article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
superconducting materials , because of their low surface resistance , are recommended for manufacturing many passive _ mw _ devices @xcite . among the various devices ,
the superconducting resonant cavity is one of the most important application in the systems requiring high selectivity in the signal frequency , such as filters for communication systems @xcite , particle accelerators @xcite , equipment for material characterization at _ mw _ frequencies @xcite . since the discovery of the high - temperature superconductors ( hts ) ,
several attempts have been done to assemble _ mw _ cavities made of bulk hts @xcite ; however , limitations in their performance were encountered .
firstly , grain boundaries in these materials are weakly coupled giving rise to reduction of the critical current and/or nonlinear effects , which worsen the device performance @xcite ; furthermore , the process necessary to obtain bulk hts in a performing textured form is very elaborated . for these reasons , in several applications , most of the superconducting cavities are still manufactured with nb , requiring liquid he as refrigerator @xcite . since the discovery of superconductivity at 39 k in mgb@xmath0 @xcite
, several authors have indicated this material as promising for technological applications @xcite .
the advantages of using mgb@xmath0 for this purpose are several .
the simple chemical composition makes the synthesis process easy , reducing the production cost in comparison with hts .
the relatively high critical temperature allows to maintain the compound in the superconducting state by using modern closed - cycle cryocoolers , not requiring liquid he as thermal bath .
a further interesting property of mgb@xmath0 bulk samples is the high connectivity of the superconducting grains .
indeed , contrary to oxide hts , bulk mgb@xmath0can be used in the polycrystalline form without a significant degradation of its critical current @xcite .
this property has been ascribed to the large coherence length , which makes the material less susceptible to structural defects like grain boundaries @xcite . due to these amazing properties ,
mgb@xmath0 has been recommended for manufacturing _ mw _ cavities @xcite , and investigation is carried out to test the potential of different mgb@xmath0 materials for this purpose . in this paper
, we report on the _ mw _ properties of the first cylindrical cavity made of bulk mgb@xmath0
the material has been produced by the reactive liquid mg infiltration technology @xcite ; it exhibits a critical temperature @xmath5 k. the frequency response of the cavity has been measured in the range @xmath1 ghz .
the quality factor of the cavity for the te@xmath2 mode has been investigated as a function of the temperature , from @xmath6 k up to @xmath7 k. from the @xmath8 values , we have determined the temperature dependence of the _ mw _ surface resistance of the mgb@xmath0 material ; these results have been compared with those independently obtained in a small plate - like sample of mgb@xmath0extracted from the same specimen used for the cavity .
as it is well known , an important valuation index to determine the performance of a resonant cavity is the quality factor @xmath8 , defined as @xmath9 in an empty resonant cavity , essentially two causes contribute to the energy losses : conductor losses , associated to the conduction currents in the cavity walls ; radiation losses , due to power that leaves the cavity through the holes for the coupling between the resonator and the external circuit .
therefore , the overall or loaded @xmath8 , denoted by @xmath10 , can be defined by @xmath11 where @xmath12 is due to the radiative losses and @xmath13 , the so - called unloaded @xmath8 , includes only the conductor losses .
+ @xmath13 is given by @xmath14 where @xmath15 is the surface resistance of the material ; @xmath16 is the geometry factor , it depends on the shape and dimensions of the cavity and the specific resonant mode . for a cavity coupled to the external circuit through two lines ,
@xmath13 can be deduced , from the measured @xmath10 , by taking into account the coupling coefficients , @xmath17 and @xmath18 , for both the coupling lines ; these coefficients can be calculated by directly measuring the reflected power at each line , as described in chap .
iv of ref .
thus , @xmath13 can be calculated as @xmath19 it is worth noting that @xmath12 can be made large by weakly coupling the excitation and detection lines . in this case ,
@xmath20 and @xmath21 .
as it is well known , resonant cylindrical cavities can support both te@xmath22 and tm@xmath22 modes ; the subscripts _
l _ , _ m _ and _ n _ refer to the number of half - wavelength variations in the standing - wave pattern in @xmath23 , @xmath24 and @xmath25 directions , respectively .
the resonant frequencies of a cylindrical cavity resonating in the te@xmath22 or tm@xmath22 modes are given by @xcite @xmath26 @xmath27 where @xmath28 and @xmath29 are the permeability and dielectric constant of the medium filling the cavity ; @xmath30 and @xmath31 the radius and length of the cavity ; @xmath32 and @xmath33 are the _ _
m__th zeros of the bessel function of order _
l _ and of its first derivative , respectively . among the various resonant modes ,
the ones more extensively used are the te@xmath34 , which give a reasonably high quality factor . in the te@xmath34 modes ,
the wall currents are purely circumferential and no currents flow between the end plates and the cylinder .
this implies that no electrical contact between plates and cylinder is required , making the cavity assembly easy .
the te@xmath34 modes are degenerate in frequency with the tm@xmath35 modes and this should be avoided to have a well defined field configuration ; however , this degeneracy can be removed introducing a small gap between cylinder and lids .
this shifts the resonant frequency of the tm@xmath35 modes downwards , leaving the te@xmath34 modes nearly unperturbed .
the unloaded quality factor of a cylindrical cavity resonating in the te@xmath34 mode is given by @xmath36 with @xmath37^{3/2}}{2(z_{01}^{\prime})^2 d^3 + 4 n^2\pi^2 a^3}\,,\;\;\ ; z_{01}^{\prime } = 3.83170\,.\ ] ]
the material from which the cavity is made has been produced by the reactive liquid mgb@xmath0 infiltration technology ( rli ) @xcite .
the rli process consists in the reaction of pure liquid mg and boron powder , sealed in a stainless steel container with a weight ratio mg / b over the stoichiometric value ( @xmath38 ) .
the reaction between liquid mg and b powder produces a more stable material , with respect to the hot pressing technique , at temperatures of a few hundred k above the mg melting point and at moderate pressure .
this procedure allows one to obtain specimens of high density and dimensions of the order of tens on centimeters , showing very high mechanical strength @xcite . as the first attempt to apply the mgb@xmath0 produced by rli to the cavity - filter technology ,
we have manufactured a simple cylindrical cavity and have investigated its microwave response in a wide range of frequencies .
all the parts of the cavity , cylinder and lids , are made of bulk mgb@xmath0 material with @xmath39 k and density @xmath40 .
the material has been obtained using commercial crystalline b powder ( starck h. c. , germany , 99.5% purity ) and pure mg ; in order to get the b powder to be used in the synthesis process , the original chunks were mechanically crushed and filtered through a 100 micron sieve . in particular , the present cylindrical cavity ( inner diameter 40 mm , outer diameter 48 mm , height 42.5 mm ) was cut by electroerosion from a thicker bulk mgb@xmath0 cylinder , previously prepared as described in sec . 4.3 of ref .
@xcite , internally polished to a surface roughness of about 300 nm . a photograph of the parts , cylinder and lids , composing the superconducting cavity is shown in fig . [ fig - cavity ] .
the holes in one of the lids have been used to insert two small loop antennas which couple the cavity with the excitation and detection lines .
the loop antenna was constructed on the end of each line , soldering the central conductor to the outer shielding of the semirigid cable . in order to remove the degeneracy between te@xmath34 and tm@xmath35 modes , we have incorporated a mode trap " in the form of circular grooves ( 1 mm thick , 2 mm wide ) inside the cylinder at the outer edges .
[ fig - cavity ] cavity .
the holes in one of the lids are used to insert the coupling loops .
after ref.@xcite.,title="fig:",scaledwidth=45.0% ]
cavity filled by liquid he in the frequency range 5.3 @xmath41 12.3 ghz . in the omitted ranges ,
no resonant curves were present.,width=275 ] the resonant cavity has been characterized by measuring its frequency response in the range @xmath1 ghz by an _
hp_-8719d network analyzer , using a two - port configuration .
the spectrum of the cavity , measured at @xmath6 k , is shown in fig .
[ fig : spettro ] , in four frequency intervals .
the arrows indicate the resonances we have certainly recognized by eqs .
( [ equ : f - te ] ) and ( [ equ : f - tm ] ) ; some peaks show a not simple structure , probably due to the superposition of different modes , which hinders to recognize the corresponding mode . among the various modes detected , two of them show the highest quality factors ; at @xmath6 k , with the cavity filled by liquid he , the resonant frequencies of these modes are 9.567 ghz and 11.291 ghz .
at room temperature , with the cavity filled by gaseous he , these resonant frequencies move to 9.79 ghz and 11.54 ghz ; according to eq .
( [ equ : f - te ] ) , they correspond to the te@xmath2 and te@xmath42 modes .
[ fig : te - curve ] shows the resonant curves , at @xmath6 k , corresponding to the te@xmath2 ( a ) and te@xmath42 ( b ) modes .
the lines are lorentzian fits of the experimental data , which give for the loaded quality factors @xmath43 and @xmath44 .
k , corresponding to the te@xmath2 ( a ) and te@xmath42 ( b ) modes .
the lines are lorentzian fits of the experimental curves from which we have extracted the values of @xmath10 and the central frequency @xmath45.,width=283 ] the quality factor of the te@xmath2 mode has been measured as a function of the temperature , in the range @xmath46 k. in order to calculate the unloaded @xmath8-values , we have determined the coupling coefficients by measuring the power reflected by each line .
the coupling coefficients , @xmath17 and @xmath18 , have been measured at different values of the temperature ; they result @xmath47 at @xmath6 k and reduce to @xmath48 when the material goes to the normal state , at @xmath49 k. fig .
[ fig : qvst - cavity ] shows the temperature dependence of the loaded and unloaded @xmath8 for the te@xmath2 mode , @xmath50 and @xmath51 . at @xmath6
k , we have obtained @xmath52 ; on increasing the temperature , the quality factor maintains values of the order of @xmath3 up to @xmath53 k and reduces by a factor of @xmath54 at @xmath55 .
values , @xmath50 and @xmath51 , for the te@xmath2 mode.,width=283 ] from the values of @xmath51 , it is possible , using eq .
( [ equ : q - te01n ] ) , to deduce the surface resistance of the mgb@xmath0 material from which the cavity is made . in order to identify possible spurious effects in the measurement of the _ mw _ response of the cavity
, we have measured the _
_ surface resistance of a small plate - like sample of mgb@xmath0 extracted from the same specimen from which the cavity has been obtained .
these measurements have been performed by the technique of hot - finger cavity perturbation @xcite , using a nb cavity resonating at @xmath56 ghz .
[ fig : rvst - cavity ] shows a comparison between the @xmath57 values deduced from @xmath58 ( full symbols ) and those obtained in the sample ( open symbols ) .
as one can see , when the cavity is in the superconducting state , the @xmath57 curves obtained by the two different techniques are consistent , while for @xmath59 they differ by @xmath60 .
this disagreement may be ascribed to the difficulty of measuring with high sensitivity the quality factor of the mgb@xmath0 cavity at @xmath59 , due to its small value . on the contrary
, the sensitivity of the measurements performed by the technique of hot - finger cavity perturbation is very high independently of the temperature ; indeed , the nb cavity is always maintained in the superconducting state and only the temperature of the sample is changed .
deduced from @xmath58 ( full symbols ) and that obtained at 9.4 ghz in a small mgb@xmath0 sample extracted from the same material of the cavity ( open symbols).,width=283 ] as one can see from fig .
[ fig : te - curve]b , the resonant curve for the te@xmath42 mode is not very well described by a lorentzian function ; this problem becomes more significant on increasing the temperature , making inaccurate the determination of the quality factor at temperatures higher that @xmath61 k. for these reason , we have not reported here the temperature dependence of @xmath62 .
however , we have verified that , also for the te@xmath63 mode , the quality factor takes on values of the order of @xmath3 up to @xmath64 k and reduced by a factor of @xmath54 when the material goes to the normal state .
it is well known that one of the most important application of superconducting materials is the implementation of _ mw _ resonant cavities , which can be used in a large variety of devices @xcite .
although several studies have been devoted to build _ mw _ cavities using hts , actually most of the superconducting cavities are still manufactured with nb , requiring liquid he as refrigerator .
the main problem encountered in using bulk hts , for this purpose , is due to the weak coupling of grain boundaries in these materials , which act as weak links . on the contrary
, it has widely been shown that in bulk mgb@xmath0 samples only a small amount of grain boundaries act as weak links @xcite , hindering the strong suppression of the critical current with magnetic field and the generation of nonlinear effects at high input power .
for these reasons , as well as for the relatively high @xmath65 , several authors have recommended mgb@xmath0 as a convenient material to build _ mw _ cavity .
recently , we have investigated the _
mw _ response of small plate - like samples of mgb@xmath0 prepared by rli process , in the linear and nonlinear regimes @xcite .
these studies have highlighted a weak nonlinear response , as well as relatively small values of the residual surface resistance .
in particular , we have found that the _ mw _ properties of such samples improve on decreasing the size of the b powder used in the synthesis process .
furthermore , bulk samples produced by rli maintain the surface staining unchanged for years , without controlled - atmosphere protection .
this useful property is most likely related to the high density , and consequently high grain connectivity , achieved with the rli process , as well as to the small and controlled amount of impurity phases @xcite .
prompted by these interesting results , we have built the _
mw _ resonant cavity using mgb@xmath0 produced by the rli process @xcite .
the results of fig .
[ fig : qvst - cavity ] show that @xmath8 takes on values of the order of @xmath3 from @xmath6 k up to @xmath66 k , a temperature easily reachable by modern closed - cycle cryocoolers . to our knowledge ,
these @xmath8 values are higher than those reported in the literature for _ mw _ cylindrical cavities manufactured with hts , both bulk and films @xcite .
so , our results show that mgb@xmath0 produced by rli is a very promising material for building _
mw _ resonant cavities .
we would remark that this is the first attempt to realize a superconducting cavity made of bulk mgb@xmath0 .
in particular , this investigation has been carried out with the aim to explore the potential of bulk mgb@xmath0 materials prepared by rli to the _ mw_-cavity technology .
the mgb@xmath0 material from which the present cavity is made has been obtained using crystalline b powder with grain mean size @xmath67 m .
our previous studies on bulk mgb@xmath0 samples , obtained by the rli method , have shown that samples prepared using microcrystalline b powder ( @xmath68 m in size ) exhibit smaller residual surface resistance ( @xmath69 ) @xcite . from fig .
[ fig : rvst - cavity ] , one can see that the residual surface resistance of the cavity is @xmath70 .
so , we infer that one could improve the quality factor by one order of magnitude manufacturing the cavity with material produced by liquid mg infiltration in micrometric b powder .
because of the shorter length of percolation of the liquid mg into very fine b powder , the production of massive mgb@xmath0 samples by rli using b powder with size @xmath68 m turns out to be more elaborate . however , work is in progress to improve the preparation process in order to manufacture large specimens using microcrystalline b powder . considering that the _ mw _ fields penetrate in a surface layer of the material of the order of the penetration depth , a more performing
_ resonant cavity can be obtained designing a composite structure ; in particular , one can use microcrystalline b powder for the inner part of the cavity and larger grain - size b powder for the outer part . in conclusion , we have shown that the rli process provides a useful method for assembling high - performance _
_ cavities , which may have large scale application .
we have obtained quality factors of the order of @xmath3 , larger than those reported in the literature for cavities made of hts , both bulk and films , in the same temperature range .
although higher quality factors have been reported for superconducting cavities made of pure nb and nb alloys @xcite , nb cavities must be kept in liquid - he bath ; on the contrary , cavity made of mgb@xmath0 can be maintained in the superconducting state by using close - cycle cryocoolers that can easily work at temperature of the order of 10 k.
the authors are very glad to thank m. bonura for his continuous interest and helpful suggestions , g. lapis and g. napoli for technical assistance . work partially supported by the university of palermo (
grant coll .
li vigni , co.ri 2005 ) .
d. c. larbalestier , l. d. cooley , m. o. rikel , a.a .
polyanskii , j. jiang , s. patnaik , x. y. cai , d. m. feldmann , a. gurevich , a. a. squitieri , m. t. naus , c. b. eom , e. e. hellstrom , r. j. cava , k. a. regan , n. rogado , m. a. hayward , t. he , j. s. slusky , p. khalifah , k. inumaru , and m. haas , nature ( london ) * 410 * ( 2001 ) 186 . | we report on the microwave properties of a resonant cylindrical cavity made of bulk mgb@xmath0 superconductor , produced by the reactive liquid mg infiltration process .
the frequency response of the cavity has been measured in the range @xmath1 ghz . among the various modes , the te@xmath2 , resonating at 9.79 ghz ,
exhibits the highest quality factor . for this mode
, we have determined the temperature dependence of the quality factor , obtaining values of the order of @xmath3 in the temperature range @xmath4 k. the values of the surface resistance deduced from the measurements of the quality factor agree quite well with those independently measured in a small sample of mgb@xmath0 extracted from the same specimen from which the cavity has been obtained . , , , , superconducting microwave cavity , mgb@xmath0 , surface resistance 85.25.am , 74.25.nf , 74.70.ad |
network science has attracted much attention in recent years due to its interdisciplinary applications @xcite .
many network results have been obtained by analyzing isolated networks , but most real - world networks do in fact interact with and depend on other networks @xcite .
thus , in analogy to the ideal gas laws that are valid only in the limiting case that molecules do not interact , so the extensive results for the case of non - interacting networks hold only when it is justified to neglect the interactions between networks .
recently several studies have addressed the resilience as well as other properties of interacting networks @xcite .
a framework based on percolation theory has been developed to analyze the cascading failures caused by interdependencies between two networks @xcite . in interdependent networks ,
when nodes in one network fail they usually cause the failure of dependent nodes in other networks , and this in turn can cause further damage to the first network and result in cascading failures , which could lead to abrupt collapse of the system .
later on , two important generalizations of the basic model @xcite have been developed .
because in real - world scenarios the initial failure of important nodes ( `` hubs '' ) may not be random but targeted , a mathematical framework for understanding the robustness of interdependent networks under an initial targeted attack on specific degree of nodes has been studied by huang et al .
@xcite and later extended by dong et al .
also in real - world scenarios , the assumption that each node in network a depends on one and only one node in network b and vice versa may not be valid . to release this assumption , a theoretical framework for understanding the robustness of interdependent networks with a random number of support and dependency relationships has been developed and studied by shao et al.@xcite .
more recently , gao et al . developed an analytical framework to study percolation of a tree - like network formed by @xmath0 interdependent networks @xcite .
gao et al .
found that while for @xmath3 the percolation transition is a second order , for any @xmath4 cascading failures occur and the network collapses as in a first order transition .
indeed cascading failures have caused blackouts in interdependent communication and power grid systems spanning several countries @xcite . to be able to design resilient infrastructures or improve existing infrastructures we need to understand how venerability is affected by such interdependencies @xcite .
here we generalize the theory of interdependent networks @xcite to regular and random regular ( rr ) network of @xmath0 interdependent networks that include loops .
figures [ fig1](a ) and [ fig1](b ) illustrate such network of networks ( netonet ) , in which each network depends on the same number @xmath1 of other networks .
we develop an exact analytical approach for percolation of a regular and a random regular netonet system composed of @xmath0 partially interdependent networks .
we show that for an rr network with degree @xmath1 of @xmath0 interdependent networks where each network has the same degree distribution , same average degree @xmath5 and the fraction of dependence nodes between a pair of interdependent networks , @xmath2 , is the same for all pairs , the number of networks is irrelevant .
we obtain analytically the fraction of survived nodes in each network after cascading failures , @xmath6 as a function of @xmath7 , @xmath1 and @xmath5 .
in our model , each node in the netonet is itself a network and each link represents a _ fully _ or _ partially _ dependent pair of networks [ see fig .
[ fig1 ] ] .
we assume that each network @xmath8 ( @xmath9 ) of the netonet consists of @xmath10 nodes linked together by connectivity links .
two networks @xmath8 and @xmath11 form a partially dependent pair if a certain fraction @xmath12 of nodes in network @xmath8 directly depend on nodes in network @xmath11 , i.e. , nodes in network @xmath8 can not function if the corresponding nodes in network @xmath11 do not function . a node in a network @xmath8 will not function if it is removed or if it does not belong to the largest connected cluster ( giant component ) in network @xmath8 .
dependent pairs may be connected by unidirectional dependency links pointing from network @xmath11 to network @xmath8 [ see fig .
[ fig1](c ) ] .
this convention indicates that nodes in network @xmath8 may get a crucial support from nodes in network @xmath11 , e.g. , electric power if network @xmath11 is a power grid .
we assume that after an attack or failure only a fraction of nodes @xmath13 in each network @xmath8 remains .
we also assume that only nodes that belong to a giant component in each network @xmath8 will remain functional .
when a cascade of failures occurs , nodes in network @xmath8 that do not belong to the giant component in network @xmath8 fail and cause nodes in other networks that depend on them to also fail .
when those nodes fail , dependent nodes and isolated nodes in the other networks also fail , and the cascade can cause further failures back in network @xmath8 . in order to determine the fraction of nodes @xmath14 in each network that remains functional ( i.e. , the fraction of nodes that constitutes the giant component ) after the cascade of failures as a function of @xmath13 and @xmath15 , we need to analyze the dynamics of the cascading failures .
we assume that all @xmath10 nodes in network @xmath8 are randomly assigned a degree @xmath16 from a probability distribution @xmath17 , they are randomly connected , and the only constraint is that a node with degree @xmath16 has exactly @xmath16 links @xcite .
we define the generating function of the degree distribution , @xmath18 where @xmath19 is an arbitrary complex variable .
the generating function of this branching process is defined as @xmath20 .
once a fraction @xmath21 of nodes is randomly removed from a network , the probability that a randomly chosen node belongs to a giant component , is given by @xcite @xmath22,\ ] ] where @xmath23 satisfies @xmath24.\ ] ] we assume that ( i ) each node @xmath25 in network @xmath8 depends with a probability @xmath26 on only one node @xmath27 in network @xmath11 , and that , ( ii ) if node @xmath25 in network @xmath8 depends on node @xmath27 in network @xmath11 and node @xmath27 in network @xmath11 depends on node @xmath28 in network @xmath8 , then node @xmath25 coincides with node @xmath28 , i.e. , we have a no - feedback situation @xcite . in section
iv we study the case of feedback condition , i.e. , node @xmath25 can be different from @xmath28 in network @xmath8 .
the no feedback condition prevents configurations from collapsing even without having their internal connectivity in each network @xcite .
next , we develop the dynamic process of cascading failures step by step . at @xmath29 , in networks @xmath8 of the netonet we randomly remove a fraction @xmath30 of nodes .
after the initial removal of nodes , the remaining fraction of nodes in network @xmath8 , is @xmath31 .
the remaining functional part of network @xmath8 therefore constituents a fraction @xmath32 of the network nodes , where @xmath33 is defined by eqs .
( [ ge2 ] ) and ( [ ge3 ] ) .
furthermore , we denote by @xmath34 the fraction of nodes in network @xmath8 that survive after the damage from all the networks connected to network @xmath8 except network @xmath11 is taken into account , so if @xmath35 , @xmath36 . when @xmath37 , all the networks receive the damages from their neighboring networks one by one . without loss of generality , we assume that network @xmath38 is the first , network @xmath39 second , ... , and network @xmath0 is last . in fig .
1(c ) , for example , since a fraction @xmath40 , @xmath41 , @xmath42 and @xmath43 of nodes of network @xmath38 depends on nodes from network @xmath39 , @xmath44 , @xmath45 and @xmath46 respectively , the remaining fraction of network @xmath38 nodes is , @xmath47,\ ] ] and @xmath48 ( @xmath49 ) satisfies @xmath50 the remaining functional part of network @xmath38 therefore contains a fraction @xmath51 of the network nodes .
similarly , we obtain the remaining fraction of network @xmath8 nodes , @xmath52\prod_{s > i } [ q_{si}y_{si , t-1}g_s(\psi'_{s , t-1})-q_{si}+1],\ ] ] and @xmath53 is @xmath54 and @xmath55 is @xmath56 following this approach we can construct the sequence , @xmath57 of the remaining fraction of nodes at each stage of the cascade of failures
. the general form is given by @xmath58}\prod_{s > i}{[q_{si}y_{si , t-1}g_s(\psi'_{s , t-1})-q_{si}+1 ] } , & \mbox { } & \\
y_{ij , t}=\frac{\psi'_{i , t}}{q_{ji}y_{ji , t-1}g_j(\psi'_{j , t})-q_{js}+1 } , & \mbox { } & \\ y_{is , t}=\frac{\psi'_{i , t}}{q_{si}y_{si , t-1}g_s(\psi'_{s , t-1})-q_{si}+1}. & \mbox { } & \\ \end{array}\ ] ] we compare the theoretical formulas of the dynamics , eqs .
( [ ge10 ] ) and simulation results in fig .
as seen the theory of the dynamics ( [ ge10 ] ) agrees well with simulations . to determine the state of the system at the end of the cascade process we look at @xmath59 at the limit of @xmath60
this limit must satisfy the equations @xmath61 since eventually the clusters stop fragmenting and the fractions of randomly removed nodes at step @xmath62 and @xmath63 are equal . denoting @xmath64 ,
we arrive for the @xmath0 networks , at the stationary state , to a system of @xmath0 equations with @xmath0 unknowns , @xmath65 where the product is taken over @xmath66 networks interlinked with network @xmath8 by partial ( or fully ) dependency links [ see fig .
[ fig1 ] ] and @xmath67 is the fraction of nodes in network @xmath8 that survive after the damage from all the networks connected to network @xmath11 except network @xmath8 itself is taken into account .
the damage from network @xmath8 itself is excluded due to the no - feedback condition .
equation ( [ e1 ] ) is valid for any type of interdependent netonet , while eqs .
( [ e2 ] ) represents the no - feedback condition . for two coupled networks , eqs .
( [ e1 ] ) and ( [ e2 ] ) are equivalent to eq .
( 13 ) of ref .
@xcite for the specific case of single dependency links . our general framework for percolation of interdependent network of networks , eqs .
( [ e1 ] ) and ( [ e2 ] ) , can be generalized in two directions : ( i ) coupling with feedback condition ( ii ) coupling with multiple - support .
\(i ) in the existence of the feedback , @xmath68 is simply @xmath69 and eqs .
( [ e1 ] ) and ( [ e2 ] ) become a single equation , @xmath70 the feedback condition leads to an extreme vulnerability of the network of interdependent networks . as we know for two fully interdependent networks with no - feedback condition @xcite if the average degree is large enough both networks exist . however , for two fully interdependent networks with feedback condition , no matter how large the average degree is , both networks collapse even after a single node is removed .
the analytical results about the feedback condition are given in section iv .
\(ii ) equation ( [ e1 ] ) can be generalized to the case of multiple dependency links studied for a pair of coupled networks in @xcite by , @xmath71\right),\ ] ] where @xmath72 represents the generating function of the degree distribution of multiple support links that network @xmath8 depends on network @xmath11 .
on one hand , the term @xmath73 reflects the topology of network @xmath8 , which can be an er network , a rr network , a scale free ( sf ) network , or even a small world ( sw ) network . on the other hand , @xmath74_{n\times n}$ ] ( @xmath0 is the number of networks ) reflects the interactions between the networks , i.e. , the topology of the netonet , which can also be any type of network
. our theoretical results eq . ( [ e1 ] ) and ( [ e2 ] ) are therefore general for any type of network of networks . by solving eqs .
( [ e1 ] ) and ( [ e2 ] ) , or eqs .
( [ e1 - 1 ] ) , or eqs .
( [ e1 - 2 ] ) , we obtain @xmath69 of each network for coupled networks with no feedback condition , feedback condition and multiple - support condition , respectively .
thus , we obtain the giant component in each network @xmath8 as @xmath75
in order to study the various forms the stationary state of the system can reach after a cascading failure , we first assume , without loss of generality , that each network depends on @xmath1 other random networks , i.e. , that we have a rr network formed of @xmath0 random networks .
we understand the rr category to also include regular non - random networks in which each network has the same number of neighbouring interdependent networks with a structure e.g. , of a lattice of er networks [ fig .
[ fig1](a ) ] . we assume , for simplicity , that the initial attack on each network is by removing randomly a fraction @xmath76 of nodes , the partial interdependency fraction is @xmath2 , and the average degree of each er network is the same @xmath77 for all networks . because of the symmetries involved , the @xmath78 equations in eqs .
( [ e1 ] ) and ( [ e2 ] ) can be reduced to two equations , @xmath79 by substituting @xmath80 , eqs .
( [ ge2 ] ) , ( [ ge3 ] ) and ( [ e1 - 3 ] ) into ( [ e11 ] ) , and eliminating @xmath81 , @xmath82 , and @xmath83 , we obtain @xmath84(1-z)}{1-h(z)},\ ] ] and @xmath85 equation ( [ ge5 ] ) can help us to understand the percolation of a rr network of any interdependent random networks where all networks have the same average degree and degree distribution . to solve eq .
( [ ge5 ] ) , we introduce an analytical function @xmath86 for @xmath87 $ ] as @xmath88 @xmath86 as a function of @xmath19 has a quite complex behaviour for various degree distributions .
we present two examples to demonstrate our general results on ( i ) rr network of er networks and ( ii ) rr network of sf networks .
* for the case of rr network of er networks we find a critical @xmath89 such that , when @xmath90 the system shows a second order phase transition and the critical threshold @xmath91 depends on @xmath2 and average degree @xmath77 .
when @xmath92 the system shows a first order phase transition , and when @xmath93 there is no phase transition because all the networks collapse even for a single node failure .
* for the case of rr network of sf networks , the phase diagram is different from the er case , because there is no pure first order phase transition .
however , there exists an effective @xmath94 , when @xmath95 , the system shows a second order phase transition and the critical threshold is @xmath96 for infinite number of nodes in each network , i.e. , the maximum degree goes to @xmath97 .
but for finite size of network , there exists an effective @xmath98 , when @xmath99 the giant component of each network is very close to 0 , and when @xmath100 finite size of giant component smoothly emerges .
it becomes more interesting .
when @xmath101 , the system shows a hybrid transition as follows .
when @xmath7 decreases from 1 to 0 , the giant component @xmath6 as function of @xmath7 shows a sharp jump at @xmath102 , which is like a first order transition to a finite small value , and then ( when @xmath7 further decreases ) goes smoothly to 0 . for @xmath93
there is no phase transition because all the networks collapse even for a single node failure . for er networks @xcite ,
the generating function @xmath103 satisfies @xcite @xmath104 , & \mbox { } &
\\ f=\exp[\bar{k } x(f-1 ) ] .
\end{array}\ ] ] substituting eqs .
( [ e13 - 0 ] ) into eqs .
( [ e11 ] ) , we get @xmath105^m(f-1)\ } } , & \mbox { } & \\ y = p[qy(1-f)-q+1]^{m-1 } , & \mbox { } & \\
p_{\infty } = -(\log f)/\bar{k}. \end{array}\ ] ] eliminating @xmath83 from eq .
( [ e13 ] ) , we obtain an equation for @xmath81 , @xmath106^{\frac{2}{m}}+(q-1)[\frac{\ln f}{\bar{k}p(f-1)}]^{\frac{1}{m}}+\frac{q}{\bar{k}}\log f = 0 .
\end{array}\ ] ] considering @xmath107^{1/m}$ ] to be a variable , eq .
( [ e14 ] ) becomes a quadratic equation that can be solved analytically having only one valid solution , @xmath108^m.\ ] ] from eq .
( [ e15 ] ) and the last equation in ( [ e13 ] ) , we determine the mutual percolation giant component for a rr network of er networks , @xmath109^m.\ ] ] figures [ fig3](a ) and [ fig3](b ) show numerical solutions of eq .
( [ e16 ] ) for several @xmath2 and @xmath1 values compared with simulations .
these solutions imply that @xmath6 as a function of @xmath7 exhibits a second ( continuous ) or a first order ( abrupt ) phase transition depending on the values of @xmath2 and @xmath1 for a given @xmath77 .
note , when q=0 or m=0 , eq .
( [ e16 ] ) is reduced to the known equation , @xmath110 , for single er networks @xcite . from eqs .
( [ ge12 - 1 ] ) and ( [ e16 ] ) , we obtain @xmath111^m}{2^m(1-z)},\ ] ] and @xmath112\sqrt{(1-q)^2 + 4q(1-z)}}. \end{split}\ ] ] next we demonstrate the behaviour of eq .
( [ ge15 ] ) as shown in fig .
[ fig4 ] . for given @xmath77 and @xmath1 , when @xmath2 is small , @xmath86 is a monotonously increasing function of @xmath19 , for example see the curve for @xmath113 .
thus , the maximum of @xmath86 is obtained when @xmath114 , which corresponds to a second order phase transition threshold @xmath115 , where @xmath116 .
when @xmath2 increases , @xmath86 as a function of @xmath19 shows a maxima at @xmath117 and @xmath118 , for example for @xmath119 in fig . [ fig4 ]
. thus , the maximum of @xmath86 is obtained when @xmath120 at the peak , which corresponds to the first order phase transition threshold @xmath121 , where @xmath122 .
the @xmath2 value in which for the first time a maxima of @xmath86 appears at @xmath117 is @xmath89 , the critical dependency which separates between the first and second order transitions .
when @xmath2 continually further increases , @xmath123 , which corresponds to a complete collapse of the netonet .
the value of @xmath2 for which @xmath124 is @xmath125 , above which the network is not stable and collapse instantaneously .
next we analyze the different behaviours of rr network of er networks in the different regimes of @xmath2 : ( i ) for @xmath90 , the percolation is a continuous second order which is characterized by a critical threshold @xmath126 .
( ii ) the range of @xmath92 is characterized by an abrupt first order phase transition with a critical threshold @xmath127 .
( iii ) for @xmath93 no transition exists due to the instant collapse of the system .
we next analyze in detail the parameters characterizing the three regimes .
( i ) for a given @xmath1 and @xmath77 , when @xmath2 is sufficiently small , there exists a critical @xmath126 such that , when @xmath7 increases above @xmath126 , @xmath6 continuously increases from zero to non - zero values . here
@xmath6 as a function of @xmath7 exhibits a second order phase transition . in order to get @xmath126
we analyze eq .
( [ ge16 ] ) . when @xmath2 is sufficiently small @xmath128 , the maximum value of @xmath86 is obtained when @xmath129 .
thus , we obtain the critical threshold for the second order phase transition , @xmath126 by substituting @xmath129 into eq .
( [ ge15 ] ) , @xmath130 \(ii ) next we obtain @xmath131 .
according to eq .
( [ ge16 ] ) , when @xmath2 increases , @xmath86 as a function @xmath19 becomes not monotonous and a maxima appears , which corresponds to the condition for first order phase transition , i.e. , when @xmath132 .
furthermore , for a given @xmath7 , the smallest of these roots gives the physically meaningful solution from which the giant component @xmath133 can be found from eq .
( [ e16 ] ) . by solving @xmath134 from @xmath135 of eq .
( [ ge16 ] ) , we obtain the critical threshold for first order phase transition @xmath127 as @xmath136^m}.\ ] ] next we study the critical coupling strength @xmath89 , i.e , the critical coupling that distinguishes between first and second order transitions .
we find that @xmath6 undergoes a second order transition as a function of @xmath7 when @xmath137 , a first order transition when @xmath92 , and no phase transition ( the system is unstable for any @xmath7 ) when @xmath93 . by definition ,
when a system changes from second order to first order at the critical point , @xmath2 , @xmath1 , and @xmath77 satisfy @xmath138 , i.e. , both conditions for the first order and second order phase transition should satisfy , @xmath139 from eqs .
( [ ge16 ] ) and ( [ ge19 ] ) , we obtain @xmath140 solving eq .
( [ e24 ] ) , we find that the physically meaningful @xmath89 is @xmath141 ( iii ) next we calculate the critical point @xmath142 , above which ( @xmath93 ) the system is unstable for any @xmath7 . from eq .
( [ ge15 ] ) , the system is unstable for any @xmath7 , when @xmath143 .
we therefore , can obtain @xmath142 by satisfying eq .
( [ ge16 ] ) and @xmath144 .
thus , we obtain @xmath142 as @xmath145 where @xmath25 satisfies @xmath146 and @xmath134 can be solved by substituting eq .
( [ e25 - 1 ] ) into eq .
( [ ge16 ] ) and set @xmath147 , @xmath135 , which is one equation with only one unknown @xmath134 .
next we obtain the numerical solution of @xmath148 as a function @xmath2 as shown in fig .
[ fig5 ] . from fig .
[ fig5 ] , we can see that for fixed @xmath1 and @xmath77 , there exist two critical values of coupling strength , @xmath89 and @xmath142 , when @xmath90 , @xmath149 which represents a second order phase transition , when @xmath92 , @xmath150 representing a first order phase transition . when @xmath93 , @xmath149 representing the netonet collapse and that there is no phase transition ( @xmath151 ) .
figure [ fig5 - 1 ] shows the phase diagram of rr network of er networks for different values of @xmath1 and @xmath77 . as @xmath1 decreases and @xmath16 increases , the region for @xmath152 increases , which shows a better robustness .
we analyze here netonets composed of sf networks with a power law degree distribution @xmath153 .
the corresponding generating function is @xmath154z^k}{(m+1)^{1-\lambda } -s^{1-\lambda}}\ ] ] where @xmath155 ( @xmath156 in this paper ) is the minimal degree cutoff and @xmath157 is the maximal degree cutoff .
sf networks approximate real networks such as the internet , airline flight patterns , and patterns of scientific collaboration @xcite . when sf networks are fully interdependent @xcite , @xmath158 , even in the case @xmath159 in contrast to a single network for which @xmath96 @xcite .
we study the percolation of a rr network composed of interdependent sf networks by substituting their degree distribution into eq .
( [ ge1 ] ) and obtaining their generating functions .
we assume , for simplicity , that all the networks in the netonet have the same @xmath160 , @xmath155 and @xmath157 , and use eq .
( [ ge5 ] ) to analyze the percolation of an rr netonet of sf networks .
the generating function of the branching process is defined as @xmath161 . substituting @xmath162 and eq .
( [ e26 ] ) into eq .
( [ ge12 - 1 ] ) , we obtain the function @xmath86 for rr of sf networks . as shown in fig .
[ fig5 - 20 ] , we find three regimes of coupling strength @xmath2 : * when @xmath2 is small ( @xmath95 ) , @xmath86 is a monotonically increasing function of @xmath19 , the system shows a second order phase transition , and the critical threshold @xmath163 is obtained when @xmath114 which corresponds to @xmath164 , i.e. , @xmath165 . * when @xmath2 is larger , @xmath101 , @xmath86 as a function of @xmath19 shows a peak which corresponds to a sharp jump to a lower value of @xmath166 at @xmath134 with a hybrid transition , because @xmath167 , which is different from the er case .
furthermore , the effective critical threshold ( sharp jump ) is @xmath168 , while for @xmath7 below this sharp jump the system undergoes a smooth second order phase transition and the critical threshold is zero , similar to ( i ) .
thus , when @xmath19 is greater than some value , @xmath86 increases with @xmath19 again and reaches @xmath169 when @xmath114 , which indicates that when @xmath7 decreases below @xmath102 , @xmath6 jumps as a first order to a finite small value and then decreases smoothly to 0 as @xmath7 approaches @xmath170 ; * when @xmath2 is above @xmath142 , @xmath86 decreases with @xmath19 first , and then increases with @xmath19 , which corresponds to the system collapse .
next we analyze the three regimes more rigorously .
\(i ) when @xmath2 is small ( @xmath171 ) , @xmath86 is a monotonically increasing function of @xmath19 , the maximum of @xmath172 is obtained when @xmath173 , which corresponds to @xmath174 , @xmath175 this is since when @xmath176 , @xmath177 , @xmath165 when @xmath95 .
\(ii ) as @xmath2 increases ( @xmath178 ) , @xmath86 as a function of @xmath19 shows a peak corresponding to @xmath179 , @xmath180 ( smaller root has the physical meaning ) , where @xmath181 corresponding to the effective critical threshold where @xmath6 as a function of @xmath7 shows an abrupt jump .
furthermore , we define @xmath182 and @xmath183 for the case of first order phase transition with a sharp jump , @xmath184 , but for the hybrid transition @xmath185 . after the sharp jump
, @xmath6 decreases smoothly to 0 until @xmath186 . for the case of two partially interdependent sf networks see zhou et .
al . @xcite .
\(iii ) as @xmath2 increases further ( @xmath187 ) , @xmath188 at @xmath189 becomes negative , thus the netonet will collapse even when a single node is initially removed .
so the maximum values of @xmath2 is obtained as @xmath190 using eqs .
( [ ge12 ] ) , ( [ ge12 - 1 ] ) and ( [ e27 - 1 ] ) , we obtain @xmath191 when @xmath192 , @xmath193 and @xmath194 , so we get @xmath195 comparison between analytical and simulation results are shown in fig . [
fig5 - 21 ] .
the above detailed analysis considers the case of no feedback condition since even for two fully interdependent networks with feedback condition ( fb ) , both networks will completely collapse even if a single node fails .
however , feedback condition can not destroy a network of partially interdependent networks when @xmath2 is sufficiently small .
for the case of feedback condition , eqs .
( [ e11 ] ) become @xmath196 substituting @xmath80 and eqs .
( [ ge1])-([ge3 ] ) into eq .
( [ ge20 ] ) and eliminating @xmath82 , we obtain @xmath197 for er networks , we obtain an equation for @xmath81 @xmath198 by substituting @xmath199 , we determine the mutual percolation giant component for a rr network of er networks with feedback condition , @xmath200 figure [ fig7 ] shows numerical solutions of eq .
( [ ge22 ] ) for several @xmath2 and @xmath1 values , which are in excellent agreement with simulations , presented as symbols .
these solutions imply that @xmath6 as a function of @xmath7 exhibits only a second order phase transition . indeed , from eq .
( [ ge22 ] ) and substituting @xmath201 ( @xmath87 $ ] ) , we obtain @xmath202 and @xmath203 next , we prove that @xmath86 is a decreasing function of @xmath19 , i.e. , @xmath204 .
it is easy to see @xmath205 and the equal condition is satisfied only when @xmath189 , so @xmath206 .
thus we obtain that @xmath86 is a monotonous decreasing function of @xmath19 , which is very different from the no feedback condition .
so the maximum of @xmath86 is obtained only when @xmath207 , which corresponds to the critical value of @xmath91 , @xmath208 which is the same as eq .
( [ e20 ] ) .
thus , the second order threshold of no feedback @xmath126 is the same as the feedback @xmath91 , which is also shown in fig .
[ fig9 ] ( a ) . however , the feedback case is still more vulnerable than the no feedback case .
[ fig9 ] ( b ) and ( c ) show @xmath6 for @xmath147 , i.e. the giant component in each network of the netonet when there is no node failures , as a function of @xmath2 .
we can see that for the no feedback case , fig .
[ fig9 ] ( b ) , the system still has very large giant component left when both @xmath1 and @xmath2 are large , but for the feedback case , there is not giant component when both @xmath1 and @xmath2 are large .
this happens because of the single connected nodes and isolated nodes in each network @xcite .
substituting @xmath209 into eq .
( [ ge25 ] ) , we obtain @xmath210 or @xmath211 , which represents the minimum @xmath77 and maximum @xmath2 for which a phase transition exists , @xmath212 and @xmath213 equations ( [ ge26 ] ) and ( [ ge27 ] ) demonstrate that the netonet collapses when @xmath2 and @xmath1 are fixed and @xmath214 and when @xmath1 and @xmath77 are fixed and @xmath215 , i.e. , there is no phase transition in these zones .
however , @xmath142 of the feedback case is smaller than that of no feedback case shown in fig .
[ fig8 ] ( a ) , which shows that the feedback case is more vulnerable than the no feedback case . in fig .
[ fig8 ] ( b ) we show that increasing @xmath77 or decreasing @xmath1 will increase @xmath142 , i.e. , increase the robustness of netonet
. next we study the feedback condition for the case of rr netonet formed of rr networks of degree @xmath16 . in this case , eq . [ ge22 ] becomes @xmath216^{\frac{1}{k}}=p\left\{1-\left[1-\frac{p_{\infty}}{p(1-q - qp_{\infty})}\right]^{\frac{k-1}{k}}\right\ } ( 1-q+qp_{\infty})^m.\ ] ] we find that the rr networks are very different from the er networks , and the system shows first order phase transition for large @xmath2 and a second order phase transition for small @xmath2 as shown in fig .
[ fig12 ] .
in summary , we develop a general framework , eqs .
( [ e1 ] ) and ( [ e2 ] ) , for studying percolation in several types of netonet of any degree distribution .
we demonstrate our approach for a rr network of er networks that can be exactly solved analytically , eqs .
( [ e16 ] ) and for rr of sf networks for which the analytical expressions can be solved numerically .
we find that @xmath142 and @xmath94 exist , where a netonet shows a second - order transition when @xmath95 , a hybrid transition when @xmath101 , and that in all other cases there is no phase transition because all nodes in the netonet spontaneously collapse .
thus the percolation theory of a single network is a limiting case of a more general case of percolation of interdependent networks .
our results show that the percolation threshold and the giant component depend solely on the average degree of the er network and the degree of the rr network , but not on the number of networks .
these findings enable us to study the percolation of different topologies of netonet .
we expect this work to provide insights leading to further analysis of real data on interdependent networks .
the benchmark models we present here can be used to study the structural , functional , and robustness properties of interdependent networks . because , in real netonets , individual networks are not randomly connected and their interdependent nodes are not selected at random , it is crucial that we understand many types of correlations existing in real - world systems and to further develop the theoretical tools studying all of them .
future studies of interdependent networks will need to focus on ( i ) an analysis of real data from many different interdependent systems and ( ii ) the development of mathematical tools for studying the vulnerability of real - world interdependent systems .
m. j. pocock , d. m. evans & j. memmott , science * 335*,973977 ( 2012 ) .
a. bashan et .
nature communications * 3 * , 702 ( 2012 ) .
k. zhao , kun & g. bianconi , arxiv preprint arxiv:1210.7498 .
* 104 * , 018701 , ( 2010 ) . c. m schneider , a .
a moreira , j. s andrade , s. havlin , h. j herrmann , proc .
sci . * 108 * , 38383841 ( 2011 ) . s. v. buldyrev et al . ,
nature * 464 * , 1025 ( 2010 ) . y. hu , b. ksherim , r. cohen , s. havlin , phys .
e * 84 * , 066116 ( 2011 ) .
s. v. buldyrev et al . , phys .
e * 83 * , 016112 ( 2011 ) .
ian dobson , benjamin a. carreras , vickie e. lynch , and david e. newman , chaos * 17 * , 026103 ( 2007 ) , i.e. , each network depends on 4 networks .
( b ) a rr network composed of 6 interdependent networks represented by 6 circles .
the degree of the netonet is @xmath217 , i.e. , each network depends on 3 networks .
the analytical results for the netonet [ eqs .
( [ e11 ] ) and ( [ ge5 ] ) ] are exact and the same for both cases ( a ) and ( b ) .
( c ) schematic representation of the dependencies of the networks .
circles represent networks in the netonet , and the arrows represent the partially interdependent pairs .
for example , a fraction of @xmath218 of nodes in network @xmath8 depends on nodes in network 3 .
pairs of networks which are not connected by dependency links do not have nodes that directly depend on each other .
, title="fig:",scaledwidth=40.0% ] , i.e. , each network depends on 4 networks .
( b ) a rr network composed of 6 interdependent networks represented by 6 circles .
the degree of the netonet is @xmath217 , i.e. , each network depends on 3 networks .
the analytical results for the netonet [ eqs .
( [ e11 ] ) and ( [ ge5 ] ) ] are exact and the same for both cases ( a ) and ( b ) .
( c ) schematic representation of the dependencies of the networks .
circles represent networks in the netonet , and the arrows represent the partially interdependent pairs . for example , a fraction of @xmath218 of nodes in network @xmath8 depends on nodes in network 3 .
pairs of networks which are not connected by dependency links do not have nodes that directly depend on each other .
, title="fig:",scaledwidth=42.0% ] , i.e. , each network depends on 4 networks .
( b ) a rr network composed of 6 interdependent networks represented by 6 circles .
the degree of the netonet is @xmath217 , i.e. , each network depends on 3 networks .
the analytical results for the netonet [ eqs .
( [ e11 ] ) and ( [ ge5 ] ) ] are exact and the same for both cases ( a ) and ( b ) .
( c ) schematic representation of the dependencies of the networks .
circles represent networks in the netonet , and the arrows represent the partially interdependent pairs .
for example , a fraction of @xmath218 of nodes in network @xmath8 depends on nodes in network 3 .
pairs of networks which are not connected by dependency links do not have nodes that directly depend on each other .
, title="fig:",scaledwidth=42.0% ] , @xmath219 , after @xmath220 cascading failures for the lattice netonet composed of 9 er networks shown in fig .
for each network in the netonet , @xmath221 , @xmath222 and @xmath223 , and @xmath224 ( predicted by eq .
( [ e25 ] ) .
the chosen value of @xmath7 is @xmath225 , and the predicted threshold is @xmath226 ( from eq .
( [ ge18 ] ) .
( b ) simulations compared to theory of the giant component , @xmath227 , for the random regular netonet composed of 6 er networks shown in fig .
1(b ) with the no feedback condition . for each network in the netonet , @xmath221 , @xmath217 , @xmath223 , @xmath228 ( predicted by eq . ( [ e25 ] ) ) , and for @xmath229 ( from eq .
( [ ge18 ] ) ) .
( c ) simulations compared to theory of the giant component , @xmath219 , for the random regular netonet composed of 6 er networks shown in fig .
1(b ) with the feedback condition . for each network in the netonet , @xmath221 , @xmath217 @xmath223 , @xmath230 ( predicted by eq . ( [ ge27 ] ) ) , and for @xmath231 ( from eq .
[ ge25 ] ) ) .
the results are averaged over 20 simulated realizations of the giant component left after @xmath220 stages of the cascading failures which is compared with the theoretical prediction of eq .
( [ ge10]).,title="fig:",scaledwidth=47.0% ] , @xmath219 , after @xmath220 cascading failures for the lattice netonet composed of 9 er networks shown in fig .
1(a ) . for each network in the netonet , @xmath221 , @xmath222 and @xmath223 , and @xmath224 ( predicted by eq .
( [ e25 ] ) .
the chosen value of @xmath7 is @xmath225 , and the predicted threshold is @xmath226 ( from eq .
( [ ge18 ] ) .
( b ) simulations compared to theory of the giant component , @xmath227 , for the random regular netonet composed of 6 er networks shown in fig .
1(b ) with the no feedback condition . for each network in the netonet , @xmath221 , @xmath217 , @xmath223 , @xmath228 ( predicted by eq . ( [ e25 ] ) ) , and for @xmath229 ( from eq .
( [ ge18 ] ) ) .
( c ) simulations compared to theory of the giant component , @xmath219 , for the random regular netonet composed of 6 er networks shown in fig .
1(b ) with the feedback condition . for each network in the netonet , @xmath221 , @xmath217 @xmath223 , @xmath230 ( predicted by eq .
( [ ge27 ] ) ) , and for @xmath231 ( from eq .
[ ge25 ] ) ) .
the results are averaged over 20 simulated realizations of the giant component left after @xmath220 stages of the cascading failures which is compared with the theoretical prediction of eq .
( [ ge10]).,title="fig:",scaledwidth=47.0% ] , @xmath219 , after @xmath220 cascading failures for the lattice netonet composed of 9 er networks shown in fig .
for each network in the netonet , @xmath221 , @xmath222 and @xmath223 , and @xmath224 ( predicted by eq .
( [ e25 ] ) .
the chosen value of @xmath7 is @xmath225 , and the predicted threshold is @xmath226 ( from eq .
( [ ge18 ] ) .
( b ) simulations compared to theory of the giant component , @xmath227 , for the random regular netonet composed of 6 er networks shown in fig .
1(b ) with the no feedback condition . for each network in the netonet , @xmath221 , @xmath217 , @xmath223 , @xmath228 ( predicted by eq .
( [ e25 ] ) ) , and for @xmath229 ( from eq .
( [ ge18 ] ) ) .
( c ) simulations compared to theory of the giant component , @xmath219 , for the random regular netonet composed of 6 er networks shown in fig .
1(b ) with the feedback condition . for each network in the netonet , @xmath221 , @xmath217 @xmath223 , @xmath230 ( predicted by eq . ( [ ge27 ] ) ) , and for @xmath231 ( from eq .
[ ge25 ] ) ) .
the results are averaged over 20 simulated realizations of the giant component left after @xmath220 stages of the cascading failures which is compared with the theoretical prediction of eq .
( [ ge10]).,title="fig:",scaledwidth=47.0% ] , as a function of @xmath7 , for er networks with average degree @xmath232 , ( a ) for two different values of @xmath1 and @xmath233 , ( b ) for two different values of @xmath2 and @xmath217 .
the curves in ( a ) and ( b ) are obtained using eq .
( [ e16 ] ) and are in excellent agreement with simulations .
the points symbols are obtained from simulations by averaging over 20 realizations for @xmath234 . in ( a )
, simulation results are shown as circles ( @xmath235 ) for @xmath236 and as diamonds ( @xmath237 ) for @xmath217 .
these simulation results support our theoretical result , eq .
( [ e16 ] ) , which is indeed independent of number of networks @xmath0 . ,
title="fig:",scaledwidth=47.0% ] , as a function of @xmath7 , for er networks with average degree @xmath232 , ( a ) for two different values of @xmath1 and @xmath233 , ( b ) for two different values of @xmath2 and @xmath217 .
the curves in ( a ) and ( b ) are obtained using eq .
( [ e16 ] ) and are in excellent agreement with simulations .
the points symbols are obtained from simulations by averaging over 20 realizations for @xmath234 . in ( a )
, simulation results are shown as circles ( @xmath235 ) for @xmath236 and as diamonds ( @xmath237 ) for @xmath217 .
these simulation results support our theoretical result , eq .
( [ e16 ] ) , which is indeed independent of number of networks @xmath0 .
, title="fig:",scaledwidth=47.0% ] as a function of @xmath19 for an rr network of er networks , for different values of @xmath2 when @xmath217 and @xmath238 .
all the lines are produced using eq .
( [ ge15 ] ) .
the symbols @xmath239 and @xmath240 show the critical thresholds @xmath126 when @xmath241 and @xmath127 when @xmath242 .
these critical thresholds coincide with the results in fig .
the dashed dotted line shows that when @xmath243 the function ( [ ge15 ] ) has no solution for @xmath147 , which corresponding to the case of complete collapse of the netonet.,scaledwidth=49.0% ] , @xmath244 , as a function of @xmath2 .
the curves are ( a ) for @xmath217 and two different values of @xmath77 , and ( b ) for @xmath238 and two different values of @xmath1 .
the curves are obtained using eqs .
( [ e16 ] ) and ( [ e20 ] ) and are in excellent agreement with simulations ( symbols ) .
panels ( a ) and ( b ) show the location of @xmath142 and @xmath89 for two values of @xmath1 . between @xmath89 and @xmath125 the transition is first order represented by @xmath150 . for @xmath90
the transition is second order since @xmath149 and for @xmath187 the netonet collapses ( @xmath149 ) and there is no phase transition ( @xmath151).,title="fig:",scaledwidth=47.0% ] , @xmath244 , as a function of @xmath2 .
the curves are ( a ) for @xmath217 and two different values of @xmath77 , and ( b ) for @xmath238 and two different values of @xmath1 .
the curves are obtained using eqs .
( [ e16 ] ) and ( [ e20 ] ) and are in excellent agreement with simulations ( symbols ) .
panels ( a ) and ( b ) show the location of @xmath142 and @xmath89 for two values of @xmath1 . between @xmath89 and @xmath125
the transition is first order represented by @xmath150 . for @xmath90 the transition is second order since @xmath149 and for @xmath187 the netonet collapses ( @xmath149 ) and there is no phase transition ( @xmath151).,title="fig:",scaledwidth=47.0% ] and @xmath238 , ( b ) for @xmath236 and @xmath232 .
the solid curves show the second order phase transition ( predicted by eq .
( [ e25 ] ) ) and the dashed - dotted curves show the first order phase phase transition , leading @xmath244 at @xmath89 from zero to non - zero values ( the rhs axis ) .
as @xmath1 decreases and @xmath245 increases , the region for @xmath152 increases , showing a better robustness .
the circle shows the tri - critical point @xmath89 , below which second order transition occurs and above which a first order transition occurs .
the square shows the critical point @xmath142 , above which the netonet completely collapse even when @xmath147.,title="fig:",scaledwidth=47.0% ] and @xmath238 , ( b ) for @xmath236 and @xmath232 .
the solid curves show the second order phase transition ( predicted by eq .
( [ e25 ] ) ) and the dashed - dotted curves show the first order phase phase transition , leading @xmath244 at @xmath89 from zero to non - zero values ( the rhs axis ) .
as @xmath1 decreases and @xmath245 increases , the region for @xmath152 increases , showing a better robustness .
the circle shows the tri - critical point @xmath89 , below which second order transition occurs and above which a first order transition occurs .
the square shows the critical point @xmath142 , above which the netonet completely collapse even when @xmath147.,title="fig:",scaledwidth=47.0% ] as a function of @xmath19 for different values of @xmath2 when @xmath217 , @xmath246 , @xmath156 and @xmath247 .
( i ) when @xmath2 is small ( @xmath248 ) , @xmath86 is a monotonically increasing function of @xmath19 , the system shows a second order phase transition .
( ii ) when @xmath2 is larger ( @xmath249 ) , @xmath86 as a function of @xmath19 shows a peak at @xmath134 which corresponds to a hybrid phase transition .
the square symbol represents the critical point of the sharp jump ( @xmath134 ) .
( iii ) when @xmath2 is large enough ( @xmath250 ) , @xmath86 decreases with @xmath19 first , and then increases with @xmath19 , which corresponds the system collapses.,scaledwidth=47.0% ] as a function of @xmath7 for different values of @xmath1 and @xmath2 for @xmath251 .
( b ) the critical threshold @xmath252 and ( c ) the corresponding giant component at the threshold @xmath253 as a function of coupling strength @xmath2 for @xmath236 and @xmath217 . the symbols in ( a ) represent simulation results , obtained by averaging over 20 realizations for @xmath254 and number of networks @xmath235 ( squares ) and @xmath255 ( circles ) .
the lines are the theoretical results obtained using eqs .
( [ ge5 ] ) and ( [ ge1])-([ge3 ] ) .
we can see in ( a ) that the system shows a hybrid phase transition for @xmath236 and @xmath256 .
when @xmath95 the system shows a second order phase transition and the critical threshold is @xmath165 .
however , in the simulation when @xmath7 is small ( but not zero ) @xmath174 .
this happens because @xmath165 is valid only when the network size @xmath257 and @xmath258 , but in simulations we have finite systems .
furthermore , when @xmath101 the system shows a hybrid transition shown in ( a ) and ( c ) , and when @xmath93 all the networks collapse even if one node fails .
we call this hybrid transition because @xmath185 , which is different from the case of er networks with first order phase transition where @xmath184.,title="fig:",scaledwidth=40.0% ] as a function of @xmath7 for different values of @xmath1 and @xmath2 for @xmath251 .
( b ) the critical threshold @xmath252 and ( c ) the corresponding giant component at the threshold @xmath253 as a function of coupling strength @xmath2 for @xmath236 and @xmath217 . the symbols in ( a ) represent simulation results , obtained by averaging over 20 realizations for @xmath254 and number of networks @xmath235 ( squares ) and @xmath255 ( circles ) .
the lines are the theoretical results obtained using eqs .
( [ ge5 ] ) and ( [ ge1])-([ge3 ] ) .
we can see in ( a ) that the system shows a hybrid phase transition for @xmath236 and @xmath256 .
when @xmath95 the system shows a second order phase transition and the critical threshold is @xmath165 .
however , in the simulation when @xmath7 is small ( but not zero ) @xmath174 . this happens because @xmath165 is valid only when the network size @xmath257 and @xmath258 , but in simulations we have finite systems .
furthermore , when @xmath101 the system shows a hybrid transition shown in ( a ) and ( c ) , and when @xmath93 all the networks collapse even if one node fails .
we call this hybrid transition because @xmath185 , which is different from the case of er networks with first order phase transition where @xmath184.,title="fig:",scaledwidth=47.0% ] as a function of @xmath7 for different values of @xmath1 and @xmath2 for @xmath251 .
( b ) the critical threshold @xmath252 and ( c ) the corresponding giant component at the threshold @xmath253 as a function of coupling strength @xmath2 for @xmath236 and @xmath217 .
the symbols in ( a ) represent simulation results , obtained by averaging over 20 realizations for @xmath254 and number of networks @xmath235 ( squares ) and @xmath255 ( circles ) .
the lines are the theoretical results obtained using eqs .
( [ ge5 ] ) and ( [ ge1])-([ge3 ] ) .
we can see in ( a ) that the system shows a hybrid phase transition for @xmath236 and @xmath256 .
when @xmath95 the system shows a second order phase transition and the critical threshold is @xmath165 . however , in the simulation when @xmath7 is small ( but not zero ) @xmath174
this happens because @xmath165 is valid only when the network size @xmath257 and @xmath258 , but in simulations we have finite systems .
furthermore , when @xmath101 the system shows a hybrid transition shown in ( a ) and ( c ) , and when @xmath93 all the networks collapse even if one node fails .
we call this hybrid transition because @xmath185 , which is different from the case of er networks with first order phase transition where @xmath184.,title="fig:",scaledwidth=47.0% ] , as a function of @xmath7 for er average degree @xmath232 , for different values of @xmath1 when @xmath233 ( a ) and for different values of @xmath2 when @xmath217 . the curves in ( a ) and ( b )
are obtained using eq .
( [ ge22 ] ) and are in excellent agreement with simulations .
the points symbols are obtained from simulations of fig .
1(b ) topology when @xmath217 and @xmath235 networks forming a circle when @xmath236 by averaging over 20 realizations for @xmath234 .
the absence of first order regime in netonet formed of er networks is due to the fact that at the initial stage nodes in each network are interdependent on isolated nodes ( or clusters ) in the other network .
however , if only nodes in the giant components of both networks are interdependent , all three regimes , second order , first order and collapse will occur , like in the case of rr netonet formed of rr networks ( see eq .
( 50 ) and fig .
title="fig:",scaledwidth=47.0% ] , as a function of @xmath7 for er average degree @xmath232 , for different values of @xmath1 when @xmath233 ( a ) and for different values of @xmath2 when @xmath217 .
the curves in ( a ) and ( b ) are obtained using eq .
( [ ge22 ] ) and are in excellent agreement with simulations .
the points symbols are obtained from simulations of fig .
1(b ) topology when @xmath217 and @xmath235 networks forming a circle when @xmath236 by averaging over 20 realizations for @xmath234 .
the absence of first order regime in netonet formed of er networks is due to the fact that at the initial stage nodes in each network are interdependent on isolated nodes ( or clusters ) in the other network .
however , if only nodes in the giant components of both networks are interdependent , all three regimes , second order , first order and collapse will occur , like in the case of rr netonet formed of rr networks ( see eq .
( 50 ) and fig .
, title="fig:",scaledwidth=47.0% ] as a function of @xmath2 for both no feedback condition and feedback condition when @xmath232 . for no feedback condition ,
the parts of curves bellow the symbols show @xmath126 and above the symbols show @xmath127 . for the feedback condition
, they only have the @xmath91 of second order , and @xmath126 for the no feedback case is equal to @xmath91 of the feedback case , but this does not mean that these two cases have equal vulnerability . @xmath259 as a function of @xmath2 for different values of @xmath1 when @xmath238 with ( b ) no feedback condition and ( c ) feedback condition .
when @xmath260 , @xmath261 for all @xmath1 and both feedback and no feedback cases . comparing ( b ) and ( c ) , we can see that the feedback case is much more vulnerable than the no feedback condition , because @xmath259 of no feedback case is much less than that of feedback case.,title="fig:",scaledwidth=49.0% ] as a function of @xmath2 for both no feedback condition and feedback condition when @xmath232 . for no feedback condition ,
the parts of curves bellow the symbols show @xmath126 and above the symbols show @xmath127 . for the feedback condition
, they only have the @xmath91 of second order , and @xmath126 for the no feedback case is equal to @xmath91 of the feedback case , but this does not mean that these two cases have equal vulnerability . @xmath259 as a function of @xmath2 for different values of @xmath1 when @xmath238 with ( b ) no feedback condition and ( c ) feedback condition .
when @xmath260 , @xmath261 for all @xmath1 and both feedback and no feedback cases . comparing ( b ) and ( c ) , we can see that the feedback case is much more vulnerable than the no feedback condition , because @xmath259 of no feedback case is much less than that of feedback case.,title="fig:",scaledwidth=46.0% ] as a function of @xmath2 for both no feedback condition and feedback condition when @xmath232 . for no feedback condition , the parts of curves bellow the symbols show @xmath126 and above the symbols show @xmath127 . for the feedback condition
, they only have the @xmath91 of second order , and @xmath126 for the no feedback case is equal to @xmath91 of the feedback case , but this does not mean that these two cases have equal vulnerability . @xmath259 as a function of @xmath2 for different values of @xmath1 when @xmath238 with ( b ) no feedback condition and ( c ) feedback condition .
when @xmath260 , @xmath261 for all @xmath1 and both feedback and no feedback cases . comparing ( b ) and ( c ) , we can see that the feedback case is much more vulnerable than the no feedback condition , because @xmath259 of no feedback case is much less than that of feedback case.,title="fig:",scaledwidth=46.0% ] as a function of @xmath77 for the case of feedback condition and no feedback condition when @xmath217 .
we can see that @xmath142 of no feedback case is larger than that of the feedback case , which indicate that the no feedback case is more robust compared to the feedback case .
( b ) the maximum value of coupling strength @xmath142 as a function of @xmath1 with the feedback condition for different values of @xmath77 , which shows that increasing @xmath77 or decreasing @xmath1 will increase @xmath142 , i.e. , increase the robustness of netonet.,title="fig:",scaledwidth=49.0% ] as a function of @xmath77 for the case of feedback condition and no feedback condition when @xmath217 .
we can see that @xmath142 of no feedback case is larger than that of the feedback case , which indicate that the no feedback case is more robust compared to the feedback case .
( b ) the maximum value of coupling strength @xmath142 as a function of @xmath1 with the feedback condition for different values of @xmath77 , which shows that increasing @xmath77 or decreasing @xmath1 will increase @xmath142 , i.e. , increase the robustness of netonet.,title="fig:",scaledwidth=49.0% ] , as a function of @xmath7 for rr of degree @xmath262 and @xmath217 , for two different values of @xmath2 .
the curves are obtained using eq .
( [ ge28 ] ) , which shows a first order phase transition when @xmath2 is large but a second order phase transition when @xmath2 is small.,scaledwidth=47.0% ] | percolation theory is an approach to study vulnerability of a system .
we develop analytical framework and analyze percolation properties of a network composed of interdependent networks ( netonet ) .
typically , percolation of a single network shows that the damage in the network due to a failure is a continuous function of the fraction of failed nodes . in sharp contrast , in netonet , due to the cascading failures , the percolation transition may be discontinuous and even a single node failure may lead to abrupt collapse of the system .
we demonstrate our general framework for a netonet composed of @xmath0 classic erds - rnyi ( er ) networks , where each network depends on the same number @xmath1 of other networks , i.e. , a random regular network of interdependent er networks .
in contrast to a _ treelike _ netonet in which the size of the largest connected cluster ( mutual component ) depends on @xmath0 , the loops in the rr netonet cause the largest connected cluster to depend only on @xmath1
. we also analyzed the extremely vulnerable feedback condition of coupling . in the case of er networks ,
the netonet only exhibits two phases , a second order phase transition and collapse , and there is no first phase transition regime unlike the no feedback condition . in the case of netonet composed of rr networks
, there exists a first order phase transition when @xmath2 is large and second order phase transition when @xmath2 is small .
our results can help in designing robust interdependent systems . |
the theory of open quantum system @xcite provides a realistic and complete description that takes into account the often uncontrollable and inevitable interaction between the system under consideration and its environment .
this particular field has attracted high attention of physicists because environment - induced effects play a vital role in a wide variety of research topics such as quantum information @xcite , quantum transport @xcite and quantum optics @xcite .
indeed , in practical quantum information processing , the inevitable interactions between the system and the environment generally lead to a deterioration of quantum information which is one of the biggest hurdle of building quantum devices or setups .
conventionally , the markov approximation was extensively used because of its simplicity and validity for systems where the system - bath coupling is weak and the memory effect of the bath is neglectable .
the markovian approximation entails that the open system dynamics is forgetful and is valid only when memory effects of the environment are negligible .
however , this approximation becomes invalid when the system - environment coupling is strong or when the environment is structured @xcite .
consequently , general non - markovian environments have to be considered in explaining new experimental advances in quantum optics @xcite , as well as in various quantum information tasks where environmental memory can be utilized to control entanglement dynamics @xcite .
therefore , it is vital to have a non - markovian description of the system s dynamics under the influence of the memory effects and the back - action of the environment without making any approximation .
however , a precise description of non - markovian open systems has long been a challenge . to this end , many theoretical approaches have been developed @xcite . among them , a stochastic schrdinger equation called the non - markovian quantum state diffusion ( qsd ) @xcite which was derived from a microscopic hamiltonian has several advantages over other exact master equations and has been proven to be a powerful tool in study of the system dynamics . while originally derived for systems embedded in bosonic bath , the qsd framework has been extended to deal with fermonic bath as well @xcite .
exact master equations were derived for many interesting systems such as dissipative multi - level atoms @xcite , multiple qubits @xcite and quantum brownian motion @xcite which was also exactly given via a path - integral approach @xcite .
recently , a generic tool for deriving non - markovian master equation has been developed using qsd @xcite which is applicable to a generic open quantum system irrespective of the system - environment coupling strength and the environment frequency distribution .
quantum continuous measurement @xcite and quantum control method @xcite employing the qsd were also studied . in quantum mechanics ,
an invariant of a quantum system remains intact during evolution of the system .
the lewis - riesenfeld dynamical invariant @xcite which was first introduced to find the solutions of time - dependent schrdinger equations has been used lately to engineer quantum states @xcite , perform quantum computation tasks @xcite as well as study shortcuts to adiabaticity @xcite .
however , it has been shown that for closed systems under hermitian hamiltonians , the system density matrix itself ( evolved by the propagator ) can be a dynamic invariant , since they share the same linear equation of motion .
the dynamical invariant has also been extended to non - hermitian hamiltonians @xcite and convolutionless master equations @xcite . in this paper , we show that for open systems whose dynamics can be described by the qsd equation , the invariants are no longer equivalent to the reduced density operator .
it is also possible to obtain an analytic solution of the qsd equation using the dynamical invariants under a bi - orthonormal basis , yielding new information on the analytical quantum trajectories . using the qsd invariants , we can also reverse - engineer a hamiltonian that is capable of driving the system to the target state . unlike unitary evolutions
, this control protocol allows the spectrum of the state to change , making it more appealing to experimental realizations .
we consider a generic quantum system embedded in a bosonic bath with the hamiltonian @xcite ( setting @xmath0 ) @xmath1 where @xmath2 is the hamiltonian of the system , @xmath3 is the lindblad operator , @xmath4 denotes the @xmath5th - mode annihilation operator of the bosonic bath with frequency @xmath6 , and @xmath7 stands for the coupling strength .
the bath state can be represented by a set of complex numbers @xmath8 which labels the bargmann coherent state of each bath mode @xmath5 .
one remarkable feature of this open system is that the influence of the bath can be fully encoded in a bath correlation function @xmath9 .
if we define a function @xmath10 that characterizes the time - dependent states of the bath and interpret @xmath11 as a gaussian random variable , then @xmath12 becomes a gaussian random process with a zero mean @xmath13=0 $ ] and the correlation function @xmath14 $ ] , where @xmath15 $ ] stands for the ensemble average . for simplicity , we first consider the case of zero - temperature bath . in this case
, the state @xmath16 , obtained by projecting the total wave function @xmath17 onto the bath state @xmath18 , corresponds to a quantum trajectory of the system and obeys a linear , time - local qsd equation @xcite @xmath19|\psi_{z^*}(t ) \rangle,\label{qsd}\ ] ] where @xmath20 is an operator ansatz defined by the functional derivative @xmath21 , and @xmath22 .
the reduced density operator @xmath23 can be obtained as @xmath24 $ ] by the ensemble average of the quantum trajectories under all possible realizations of the noise function and then the corresponding non - markovian master equations can in principle be derived .
the main challenge in the application of the qsd is to derive the functional derivative @xmath20-operator .
this @xmath20-operator can be exactly obtained for some simple models ( see e.g. @xcite ) or perturbatively derived for more general systems @xcite .
note that the qsd equation under a specific noise realization @xmath12 can be formally interpreted as a schrdinger equation with the non - hermitian effective hamiltonian @xmath25 below we use the dynamical invariants to analytically solve the qsd equation , which could give us an explicit expression for the reduced density matrix .
the lewis - riesenfeld dynamical invariant was first developed @xcite to study the magnetic - moment series for a charged particle moving nonrelativistically in an electromagnetic field and later generalized @xcite to solve the time - dependent schrdinger equation .
the invariant @xmath26 was defined so that its expectation value under any density operator @xmath27 is _ time - independent _ ,
i.e. , @xmath28\equiv \frac{\partial}{\partial t}\mathcal{i}\equiv0 $ ] .
it has been shown that the lewis - riesenfeld dynamical invariant is useful in dealing with time - dependent quantum problems , such as quantum computing in continuous time @xcite .
in fact , for closed systems whose dynamics is governed by a hermitian hamiltonian , the lewis - riesenfeld dynamical invariant obeys @xmath29 $ ] , i.e. the von neumann equation and thus shares the same dynamical behavior as the density operator . as a result , with the knowledge of the dynamical invariant , one can know the dynamics of the system under consideration .
it is shown @xcite that the dynamical invariant @xmath26 , the propagator @xmath30 and the density operator @xmath27 are mutually equivalent to each other without considering the lewis - riesenfeld phase . indeed ,
if we let @xmath31 , then for any unitary propagator @xmath30 , and density operator @xmath32 , we have @xmath33 $ ] , which is time independent .
the propagator @xmath30 can also be written as @xmath34 , where @xmath35 is the instant eigenvector of the dynamical invariant .
thus , it readily follows that if a closed system is initially prepared in one of the eigenvectors of @xmath36 , then it will necessarily evolve to the instant eigenvector of @xmath26 with the same index at a later time @xmath37 .
this property makes the dynamical invariant a valuable tool for both studying the state engineering @xcite and calculating the geometric phases .
in contrast , for an open quantum system whose dynamics is determined by the qsd equation , the problem becomes complicated .
the reduced density operator under a particular realization of noise function @xmath38 can be shown to satisfy @xmath39\label{drho},\ ] ] where @xmath40 and @xmath41 is the effective time - dependent non - hermitian hamiltonian given by the qsd in eq . .
if one directly defines the dynamical invariant as in a hermitian system by imposing @xmath42\equiv 0 $ ] , it can be seen that the invariant satisfies @xmath43 $ ] , which differs from eq . unless @xmath41 is hermitian .
thus , the dynamical invariant defined this way does not give the reduced density operator of the system under a given noise channel @xmath38 .
on the other hand , various studies @xcite have used the biorthonormal basis to study the dynamical invariants for a non - hermitian system . in such a framework , a complete biorthonormal set of eigenvectors is introduced so that the left and right eigenvectors of the hamiltonian are given , respectively , by @xmath44 and @xmath45 , where the orthonormal condition becomes @xmath46 and the completeness is @xmath47 . note that it should be carefully checked if such an eigen decomposition indeed exists for the system under consideration , because it may not be always so for any non - hermitian hamiltonian @xcite . for the criteria proposed in @xcite
, we have the system hamiltonian as the self - adjoint part , and the rest should generally be continuous and bounded for real physical scenarios .. then , the time evolution is now governed by @xmath48 and @xmath49 . as a result ,
the definition @xmath50\equiv 0 $ ] , where @xmath51 , gives @xmath52,\label{i_heff_def}\ ] ] which has the same form as in the hermitian case , albeit with a non - hermitian effective hamiltonian given by the qsd equation .
remarkably , this newly defined invariant can be used to give an analytical solution to the qsd equation . since the dynamical invariants for a non - hermitian hamiltonian
is no longer guaranteed to be hermitian , we should use the biorthonormal basis given by the instantaneous eigenvectors of @xmath26 , i.e. , @xmath53 , and @xmath54 , with @xmath55 .
we formally write the general solution to the qsd equation as @xmath56 and substitute it into the qsd equation . after some algebra ( see appendix ) , we find that under the biorthonormal basis of the invariants , the qsd equation becomes an effectively uncoupled set of differential equations of the coefficients @xmath57 , and its solution is given by @xmath58.\label{psi_solu}\ ] ] this compact , explicit solution to the qsd equation is our central result and it applies to any realizations of the noise @xmath12 .
with @xmath20 determined , we can analytically predict the quantum trajectory for each realization of the noise @xmath59 , which was previously determined numerically using the qsd method .
the reduced density operator of the system can be obtained by analytically taking the ensemble average for all realizations of noises via novikov s theorem @xcite .
under one random realization of the noise function @xmath59 for the rwa spin - boson model , where we used @xmath60 .
the solid curves correspond to the numerical solutions ; the solid , open circles and triangles correspond to the analytical results obtained via the dynamical invariant eq . .
as an illustrative example , we apply the dynamical invariants method to the quantum dynamics of a dissipative qubit under the rotating - wave approximation ( rwa ) . this model is widely used to display the decoherence effects and is exactly solvable .
the hamiltonian of the system is @xmath61 and the lindblad operator is @xmath62 , where @xmath63 is the system - bath coupling strength and @xmath64 s are pauli matrices .
the @xmath65 operator for this model @xcite takes the form of @xmath66 , where @xmath67 is a function depending on the both system parameters and the bath spectral density .
we can obtain the dynamical invariant for a given channel @xmath38 via eq . as @xmath68 du \sigma_-\nonumber\\ & \equiv\sigma_z+g(t ) \sigma_-.\end{aligned}\ ] ]
the left and right eigenvectors can be readily obtained and we finally have @xmath69\begin{bmatrix}1\\\frac{g(t)}{2}\end{bmatrix } \nonumber\\ & + \psi_2(0)\exp(it)\begin{bmatrix}0\\1\end{bmatrix},\end{aligned}\ ] ] assuming an initial state @xmath70^t$ ] , with @xmath71 denoting the transpose of a matrix . in particular , for the ornstein - uhlenbeck noise @xmath72 , @xmath67 can be explicitly given by @xmath73 .
implementation of @xmath74 $ ] gives the reduced density operator .
this provides us an analytic tool to deal with the qsd equation . for a system with known @xmath20 operator
, we can use it to directly obtain an explicit expression for the reduced density operator as a function of time neither using a master equation nor resorting to numerical calculations .
this can be very beneficial for high - dimensional systems whose numerical calculations may be very time - consuming .
another more complex example is the dissipative three - level atom with @xmath75 and @xmath76 .
the @xmath65 operator for this model @xcite explicitly depends on the noise @xmath59 and is given by @xmath77 , where @xmath67 , @xmath78 and @xmath79 are time - dependent functions that can in principle be calculated once the correlation function @xmath80 is known .
it is clear that we can assume an upper - triangular invariant of the form @xmath81 using the commutation relationship of the ladder operator for the three level system , we found from the definition eq .
the dynamic invariant @xmath26 for this model @xmath82 , \nonumber\\ & b(t)=\mathcal{r}\left[f(t)-i \omega,\sqrt{2 } \left(2a(t)p_z(t)-\left(a(t)-c(t)\right)z_t\right ) \right ] , \nonumber\\ & c(t)=\mathcal{r } \left[-2g(t)-i \omega,\sqrt{2}(z_t-2p_z(t))\right].\end{aligned}\ ] ] where @xmath83=\int_0^t \exp \left(\int_u^t g(s)ds\right)h(u)du$ ] .
we then have @xmath84|\varphi_2(t ) \rangle \nonumber\\ & + \psi_3(0)\exp\left[i \omega t-2\int_0^t f(s)ds\right]|\varphi_3(t)\rangle,\end{aligned}\ ] ] where @xmath85 it analytically reveals the quantum trajectory of the dissipative three - level system .
now we show how to use dynamical invariants to design a hamiltonian that can be used to drive an initial state to a target state by means of reverse engineering @xcite . specifically , to design the needed hamiltonian
, we first construct an invariant @xmath26 such that one of its time - dependent eigenvector @xmath86 follows the desired time - evolution path , according to eq . .
the rest eigenvectors @xmath87 , @xmath88 , are left as undetermined parameters which we will use later to make the invariant compatible with the qsd equation .
then , we take the time - derivative of this invariant to obtain its equation of motion , and compare it with eq . , where @xmath89 should be formally compatible with eq . , imposed by the qsd equation , i.e. , @xmath90 should be hermitian , both @xmath90 and @xmath3 should be _ noise - independent _ , and the @xmath65 operator is determined by @xmath90 and @xmath3 .
this is achievable by choosing an appropriate set of basis @xmath91 .
we then have the desired hamiltonian @xmath90 of the system and the corresponding @xmath3 operator . as an illustrative example
, we consider a two - level open system with the target state @xmath92 . by letting this state
be one of the eigenvector of the invariant , we first make a noise - dependent invariant of the form @xmath93 where @xmath94 is a function determined by the hamiltonian @xmath90 of the system and lindblad operator @xmath3 that we are designing .
if the coefficient @xmath57 in the general solution to eq decays to zero for the other eigenvector of the invariant , we certainly find the steady target state . unlike a closed system , the specific form of the effective hamiltonian of the qsd poses constraints on @xmath90 and @xmath3 in that @xmath90 needs to be hermitian and noise - independent .
with both @xmath90 and @xmath3 given , the @xmath65 operator is then determined . taking this into consideration and inserting eq . into eq . , we find that @xmath95 and @xmath96 . in fig .
[ fig_re ] we numerically plot the fidelity between the engineered state and the target state , as well as the purity of the engineered state for three randomly chosen mixed initial states .
it can be seen that the fidelity monotonically increases to one and become unity after some time , indicating a steady target state is reached . in sharp contrast to the closed quantum system that evolves unitarily , the spectrum of the state is free to change , and we can drive a _ mixed _ state to a target _ pure _ state by using the non - markovian dynamics of an open quantum system .
in conclusion , we studied the dynamical invariants in non - markovian open systems whose temporal evolution is governed by the non - markovian qsd equation . for systems that have an exact operator representation of the function derivative in the form of the @xmath20 operator , the dynamical invariant can be obtained analytically .
dynamical invariants of the qsd equation are discovered , and it is found that the non - hermitian dynamical invariants do not share the same equation of motion as the reduced density matrix but its eigenvectors can be used to generate an analytical expression of the solution of qsd .
this enables us to obtain temporal evolution of the open system without deriving and then solving the non - markovian master equations . using reverse engineering along with the qsd invariants , we are able to design a hamiltonian and lindblad operator that can be used to drive an initial state to a target state via non - markovian evolution .
this work is supported by the basque government ( grant no . it472 - 10 ) , the spanish micinn ( project no .
fis2012 - 36673-c03 - 03 ) , and the basque country university ufi ( project
11/55 - 01 - 2013 ) .
is supported by the hong kong grf ( project no
. 501213 ) .
j. q. you is supported by the national natural science foundation of china no .
91421102 and the national basic research program of china no .
is supported by the nsf phy-0925174 and dod / af / afosr no .
fa9550 - 12 - 1 - 0001 .
we would like to thank j. g. muga for helpful discussions . | we find dynamical invariants for open quantum systems described by the non - markovian quantum state diffusion ( qsd ) equation .
in stark contrast to closed systems where the dynamical invariant can be identical to the system density operator , these dynamical invariants no longer share the equation of motion for the density operator .
moreover , the invariants obtained with bi - orthonormal basis can be used to render an exact solution to the qsd equation and the corresponding non - markovian dynamics without using master equations or numerical simulations .
significantly we show that we can apply these dynamical invariants to reverse - engineering a hamiltonian that is capable of driving the system to the target state , providing a novel way to design control strategy for open quantum systems . |
we thank gianfranco bertone , torsten bringmann and georges meynet for useful discussions .
thanks the swedish research council ( vr ) for support .
d. n. spergel _ et al . _
[ wmap collaboration ] , astrophys .
j. suppl .
* 170 * ( 2007 ) 377 [ arxiv : astro - ph/0603449 ] .
j. angle _ et al . _ [ xenon collaboration ] , arxiv:0706.0039 [ astro - ph ] .
g. steigman , c. l. sarazin , h. quintana and j. faulkner , astron .
j. * 83 * ( 1978 ) 1050 . w. h. press and d. n. spergel , astrophys . j. * 296 * ( 1985 ) 679 .
a. gould , astrophys .
j. * 321 * ( 1987 ) 560 . k. griest and d. seckel , nucl .
b * 283 * ( 1987 ) 681 [ erratum - ibid .
b * 296 * ( 1988 ) 1034 ] .
a. gould , astrophys .
j. * 321 * ( 1987 ) 571 .
i. v. moskalenko and l. wai , arxiv : astro - ph/0608535 .
astrophys .
j. * 659 * ( 2007 ) l29 [ arxiv : astro - ph/0702654 ] .
g. bertone and m. fairbairn , arxiv:0709.1485 [ astro - ph ] .
p. salati and j. silk , astrophys .
j. * 338 * ( 1989 ) 24 .
d. n. spergel and w. h. press , astrophys .
j. * 294 * ( 1985 ) 663 .
a. gould and g. raffelt , astrophys .
j. * 352 * ( 1990 ) 654 .
astrophys .
j. * 352 * ( 1990 ) 669 .
a. bouquet and p. salati , astrophys .
j. * 346 * ( 1989 ) 284 .
d. dearborn , g. raffelt , p. salati , j. silk and a. bouquet , astrophys .
j. * 354 * ( 1990 ) 568 .
a. bottino , g. fiorentini , n. fornengo , b. ricci , s. scopel and f. l. villante , phys .
d * 66 * , 053005 ( 2002 ) [ arxiv : hep - ph/0206211 ] .
p. scott , j. edsj and m. fairbairn , in h. v. klapdor - kleingrothaus and i. v. krivosheina ( eds . ) : proc .
dark2007 , world scientific ( 2008 ) [ arxiv:0711.0991 ] .
p. scott , j. edsj and m. fairbairn , in preparation .
eggleton , mon . not .
* 151 * ( 1971 ) 351 .
mon . not .
* 156 * ( 1972 ) 361 .
o. r. pols , c. a. tout , p. p. eggleton and z. han , mon . not .
* 274 * ( 1995 ) 964 .
[ arxiv : astro - ph/9504025 ] . g. bertone and d. merritt , phys . rev .
d * 72 * ( 2005 ) 103502 [ arxiv : astro - ph/0501555 ] .
a. m. ghez _ et al .
_ , astrophys .
j. * 586 * , l127 ( 2003 ) [ arxiv : astro - ph/0302299 ] .
p. demarque and s. virani a&a * 461 * , 651 ( 2007 ) [ arxiv : astro - ph/0603326 ] | we modify a stellar structure code to estimate the effect upon the main sequence of the accretion of weakly interacting dark matter onto stars and its subsequent annihilation .
the effect upon the stars depends upon whether the energy generation rate from dark matter annihilation is large enough to shut off the nuclear burning in the star .
main sequence wimp burners look much like protostars moving on the hayashi track , although they are in principle completely stable .
we make some brief comments about where such stars could be found , how they might be observed and more detailed simulations which are currently in progress .
finally we comment on whether or not it is possible to link the paradoxically young ob stars found at the galactic centre with wimp burners .
there is growing evidence that for each gram of baryonic matter in the universe , there are around five grams of dark matter which does not couple to the electromagnetic force @xcite .
one of the more convincing candidates for this dark matter are weakly interacting massive particles ( wimps ) .
because wimps have masses and couplings close to the weak scale , they naturally give rise to a relic abundance of dark matter close to that observed after freeze out in the early universe . because wimps couple weakly to standard model particles , there is a small but non - zero wimp - nucleon cross section .
the tightest constraints for the spin - independent cross section come from the xenon experiment @xcite .
the wimp - nucleon interaction means that some wimps are gravitationally captured by stars , a process which has been well studied @xcite .
usually the focus of such investigations is the possibility of dark matter annihilating into high energy neutrinos , which then escape the star to be potentially detected by neutrino experiments like icecube .
if the accretion of dark matter were large enough however , one might expect that the stars themselves could change . the first way wimps can affect a star is by annihilating into standard model particles in its core , providing another energy source in addition to standard nuclear burning .
we refer to stars where the energy produced by wimp annihilations is greater than or comparable to that from nuclear burning as ` wimp burners ' .
recent work has focused upon white dwarfs and neutron stars @xcite .
here we instead focus on main sequence wimp burners .
previous work in this direction was carried out by salati & silk @xcite some years ago , in the context of cosmion dark matter .
these authors used an @xmath0 polytropic approximation to estimate the influence of wimp annihilation on the main sequence .
we instead employ full numerical solutions to the hydrostatic equations of stellar structure and present results from a simple code , as well as supporting preliminary results from a more advanced code .
we also update the discussion in the context of ` modern wimps ' rather than cosmions , as the nuclear scattering cross - sections and the wimp mass are more constrained than in the past .
the second way wimps can influence stellar structure is by providing an additional mechanism of heat transport in the core .
this could reduce the local temperature gradient @xcite , potentially inhibiting convection and enhancing the pulsation of horizontal branch stars @xcite .
at least for the sun however , the values currently favoured for the wimp self - annihilation cross section @xcite and the upper limit on the wimp - nucleon cross section @xcite indicate that this effect is not significant @xcite . in this work ,
we neglect heat transport by wimps , but will in a later work include this effect in more detailed , time - dependent simulations @xcite . for the current work
, we have checked that this omission does not affect our results significantly .
experiments give relatively weak constraints on the spin - dependent wimp - nucleon cross section , so it may contribute far more to wimp capture than the spin - independent one . for simplicity , we assume that the spin - dependent cross section dominates and is @xmath1@xmath2 . for a sun - like star with mass @xmath3 , a circular velocity of 220kms@xmath4 , a wimp velocity dispersion of @xmath5kms@xmath4 and a wimp mass of 100gev ,
the capture rate ( as calculated with the full capture expressions @xcite in darksusy @xcite ) is then @xmath6 we assume a hydrogen mass fraction of 75% , uniformly distributed in the star .
this is not a good approximation for the sun , but reasonable for newly born stars .
the capture rate on an arbitrary star is then approximately @xmath7 where @xmath8 is the star s mass and @xmath9 its surface escape velocity , @xmath10 is the ambient wimp density and @xmath11 is the wimp mass .
@xmath12 is the average potential at the height of the scattering nuclei ( hydrogen in this case , for which @xmath13 ) . for simplicity , we assume @xmath14 . even if this approximation is not perfect , eq .
( [ eq : caprate ] ) is only necessary for converting between the ambient wimp density @xmath10 and the capture rate @xmath15 , so it is easy to rescale if the reader wishes .
we also assume @xmath5kms@xmath4 , though for stars in very different environments to our own this must be adjusted . throughout this paper
, we use a wimp mass of @xmath16gev although to a first approximation the energy accreted by a star immersed in some fixed density of dark matter is only a function of the cross section , not the mass of the wimp .
once dark matter is captured by the star , it will form an approximately thermal ( gaussian ) internal distribution , of characteristic radius @xcite @xmath17^{1/2},\ ] ] where @xmath18 and @xmath19 are the central density and temperature .
having been concentrated in the centre of the star , the dark matter annihilates with itself , so the equation for the evolution of the number of wimps in the star over time is @xmath20 when @xmath21 , the capture and annihilation rates are in equilibrium and any dark matter which accretes onto the star is instantly converted into additional luminosity . here
we assume that all of the annihilation products interact either electromagnetically or strongly , so that they have short mean free paths in the star and the energy thermalises quickly .
in reality , some fraction of the energy will be lost to neutrinos , but this is a small effect ( typically on the order of 10% or less ) .
the timescale for the steady state to be reached will be much less than the typical evolutionary timescale of a main sequence star , so we assume that this equilibrium has been achieved .
we studied the effect of this energy input upon stars by altering the fortran code zams @xcite , which looks for stable stellar solutions assuming a constant chemical composition and a simple equation of state .
the code was modified by adding the energy generation due to wimp annihilation to the nuclear energy generation rate .
self gravitating systems have negative specific heat , so as dark matter injects energy into the core of the star , the temperature goes down .
since the nuclear reaction rates depend exponentially on temperature , they are significantly reduced for even a small drop in temperature . as the ambient density of dark matter
is increased and the capture rate goes up , the injection of energy due to wimp annihilations eventually reduces the temperature enough to shut off nuclear burning .
wimp annihilation is then the primary source of the star s luminosity .
since wimp annihilation occurs in a more centralised region than nuclear burning , the temperature gradient is much steeper in the core of the star than it would otherwise be , and the core becomes convective . outside the core , less energy is generated per unit volume than if nuclear burning were proceeding normally , so the temperature gradient is less .
the actual temperature is lower than in a normal star , but for small dark matter accretion rates it remains high enough to prevent any major increase in opacity , ensuring that energy transport in the region above the core remains radiative .
the energy from the core is easily transmitted through this radiative zone to the surface layers of the star .
the overall radius of the star remains approximately constant while the temperature decreases , so the luminosity also decreases .
as the wimp accretion rate is raised , the star continues to cool until h@xmath22 ions are able to survive at increasingly large depths below the surface .
if the star did not already have an appreciable surface convection zone , one develops . at high enough wimp capture rates , the surface convection zone merges with the inner zone and the star becomes completely convective .
the addition of more wimps and hence central luminosity beyond this point requires the star to grow in order to transport the additional energy to the surface .
the two distinct stages of this ` evolution ' can clearly be seen in figure [ hr ] .
this figure shows a black solid line corresponding to the zero age main sequence , along with ` evolutionary tracks ' plotted as the ambient density of wimps is increased .
these tracks are strongly reminiscent of the hayashi tracks which proto - stars travel along on the final stages of their evolution towards the main sequence .
the difference is that such stars are shrinking in size on the kelvin - helmholtz timescale as they radiate away gravitational potential energy . in wimp burners
, there is constant energy generation in the core ( provided the wimp capture rate remains constant ) and the stars can in principle remain at that position in the hr diagram for an arbitrarily long time .
the analysis of salati & silk @xcite led to qualitatively similar conclusions about the backing up of the main sequence along the hayashi track with increasing wimp density .
the shapes of the tracks are different in our more detailed analysis , and thus the actual positions of the resulting wimp burners in the hr diagram differ .
the effects require higher ambient wimp densities also , as we have considered modern wimps rather than cosmions ( i.e. with lower nuclear scattering cross - sections )
. in figure [ hr ] , tracks for stars of mass 0.8@xmath23 are not complete , and show increasingly large gaps for larger masses . using the modified zams code
, we could not find solutions corresponding to all temperatures between the main sequence stars and the cooler hayashi - like wimp burners .
in figure [ temp ] we plot temperature as a function of ambient wimp density .
the discontinuity occurs where the temperature of the star drops rapidly , shortly before the star becomes fully convective .
it was not initially clear whether this lack of solutions was a real effect or an artefact of the code .
we have put some effort into understanding whether some of the physical simplifications in the code ( e.g. surface boundary conditions , a simplified equation of state and thus too rigid a criterion for the onset of surface convection , or the treatment of convective heat transport itself ) are responsible for the lack of solutions .
we could find no evidence to support this idea .
nonetheless , the gaps do not appear to have a physical basis .
they seem to be a numerical artefact arising from the ` shooting ' technique used by the zams code to solve the boundary - value problem of stellar structure . in the region of the gaps , small changes in the stellar radius
cause large changes in the internal temperature .
the surface defined by the discrepancy between the inwardly and outwardly integrated partial solutions employed in this technique then becomes a highly non - trivial function of the model parameters .
finding the global minimum of this function is then nearly impossible without an initial guess extremely close to the true solution , and the algorithm fails to converge . to check that solutions do exist in this region ,
we investigated the gaps using a preliminary version of our next - generation wimp burner code darkstars .
this code has been created from generalised versions of capture routines in darksusy @xcite which use the full capture rate expressions in @xcite , and the stellar evolution package ez @xcite derived from eggleton s stars code @xcite .
the code is time - dependent , uses relaxation rather than shooting and includes wimp energy transport , more detailed treatments of the wimp distribution , capture rates , equation of state , nuclear reaction rates and opacities .
darkstars and results obtained with it will be described in full in an upcoming publication @xcite .
the results of the more detailed code ( figure [ pat ] ) , whilst consistent with the zams code close to the main sequence and in the fully convective regions on the right - hand side of the hr diagram , show that intermediate solutions do exist and that the gaps are not physical .
this cross - check also shows that we can trust the solutions of the zams code , but should nt take the actual gaps in the curves seriously .
( upper curve ) and 0.7@xmath23 as a function of wimp density .
a spin dependent wimp - nucleon cross section of @xmath24@xmath2 is assumed . _ _ ] showing that solutions can be found corresponding to all temperatures when the more sophisticated darkstars code is used . _ _ ] figure [ temp ] shows that for a spin - dependent wimp - nucleon cross section of @xmath24@xmath2 , stars only start to change their behaviour when immersed in a dark matter density of around @xmath25 or @xmath26gev@xmath27 .
this is much larger than the 0.3gev@xmath27 which is thought to be the approximate density of dark matter in the solar system .
most modern simulations of galactic dark matter halos do suggest that the density of dark matter should be much higher in the centre of the galaxy .
this density should be made more pronounced by the phenomena of adiabatic contraction and dark matter spikes created around black holes , though tempered by self - annihilation of dark matter and gravitational heating due to the motion of stars .
it is suggested @xcite that densities larger than @xmath25gev@xmath27 can be found at radii closer than @xmath28pc from the galactic centre .
observational challenges associated with constructing an hr diagram from stars in this crowded field are severe though , as the very centre of the galaxy is shrouded in dust . finally , we comment upon the ` paradox of youth ' implied by what appear to be hot , young stars close to the central black holes of m31 and our own galaxy @xcite . in our galaxy , these stars look like @xmath2915@xmath23 main sequence stars with luminosities of the order 1,000@xmath30
. such a massive main sequence star should only be around 10 myr old , so these objects would appear to have formed very recently . however , tidal forces close to the central black hole are thought to be too large to be compatible with the collapse of a gas cloud to begin star formation .
it is not easy to make normal ubv measurements of stellar temperatures at the galactic centre due to extinction , but spectral lines found in the atmosphere of such stars appear to be consistent with a temperature of 30,000 k @xcite .
if this is the case , then it would be difficult to imagine that these are normal stars which have been converted into fully convective and highly luminous wimp burners from the accretion of dark matter , as such stars would have a temperature of only a few thousand degrees .
a possible explanation for the ` paradox of youth ' is a combination of collisional stripping of red giant envelopes and merger events @xcite , owing to the high stellar densities near the centres of the two galaxies .
if this is the case , then the anomalously hot stars in fact derive from an older population .
such a population should therefore also contain less luminous , unperturbed stars , including lower mass stars still on the main sequence .
provided that this explanation is correct , and if main sequence wimp burners exist at all , then such lower mass stars should be examples of them . with upcoming observations of the galactic centre expected to probe objects as faint as 1@xmath30
, a very real possibility exists for the detection of wimp burners in the near future . |
quasi - alternating links were introduced by ozsvath and szabo in ( * ? ? ? * definition.3.9 ) as a natural generalization of alternating links .
the definition is given in a recursive way : the set @xmath0 of quasi - alternating links is the smallest set satisfying the following properties : * the unknot belongs to @xmath0 . *
if @xmath1 is link with a diagram containing a crossing @xmath2 such that 1 .
both smoothings of the diagram of @xmath1 at the crossing @xmath2 , @xmath3 and @xmath4 as in figure [ figure ] belong to @xmath0 , 2 .
@xmath5 , 3 .
@xmath6 ; then @xmath1 is in @xmath0 and in this case we say @xmath1 is quasi - alternating at the crossing @xmath2 . [ h ] at crossing @xmath2 with @xmath7 and its smoothings @xmath3 and @xmath4 respectively.,title="fig:",width=377,height=75 ] + @xmath1@xmath4 studying quasi - alternating links becomes an interesting problem in the last few years .
it has been shown in @xcite that they are homologically thin for both khovanov homology and knot floer homology . in other words ,
their khovanov homology and knot floer homology depend only on the jones and the alexander polynomials respectively and the signature of the given link .
the class of quasi - alternating links contains the class of alternating links as it has been shown in ( * ? ? ?
* lemma.3.2 ) .
champanerkar and kofman in @xcite determine an infinite family of quasi - alternating pretzel links by twisting any known quasi - alternating pretzel link .
later on , greene in @xcite characterized all quasi - alternating pretzel links . in this paper
, we generalize the technique of twisting to construct an infinite family of quasi - alternating montesinos links .
this infinite family include all classes of quasi - alternating montesinos links detected by widmar in @xcite .
moreover , we conjecture that this family contains all quasi - alternating montesinos links up to mirror image that are not alternating .
montesinos links are natural generalization of rational links . they have been classified first by the authors of @xcite and for further details and more recent reference see @xcite .
a montesinos link has a diagram as shown in figure [ figure1 ] . in this diagram
, @xmath8 is an integer that represents number of half twists and the box stands for a rational tangle with slope @xmath9 of two coprime integers @xmath10 and @xmath11 whose continued fraction is given by @xmath12 the continued fraction for any rational number @xmath13 will be denoted by @xmath14 $ ] from now on . .
( b ) represents the rational tangle with slope @xmath15 with continued fraction @xmath16$].,title="fig:",width=245,height=226 ] .
( b ) represents the rational tangle with slope @xmath15 with continued fraction @xmath16$].,title="fig:",width=170,height=132 ] + ( a ) ( b ) the montesinos links are classified by the following theorem : @xcite the montesions link @xmath17 with @xmath18 and @xmath19 , is classified by the ordered set of fractions @xmath20 up to cyclic permutations and reversal of order , together with the rational number @xmath21 .
a montesinos link @xmath17 is in standard form if @xmath22 for @xmath23 .
any montesinos link can be put into a standard form using the above theorem .
a rational tangle extends the crossing @xmath2 if it contains the crossing @xmath2 and for all @xmath24 we have @xmath25 ( ( * ? ? ?
* page.2452 ) ) .
the product of rational tangles extends the crossing @xmath2 if each rational tangle in the product extends the crossing @xmath2 . for the definition of product of two rational tangles and more details of this subject see ( *
* section.2 ) . for a given sequence of nonnegative integers
@xmath14 $ ] , the authors of @xcite define a new sequence of positive integers recursively by @xmath26 they show that @xmath27 is equal to the determinant of the rational link @xmath28 with slope @xmath13 of two coprime integers @xmath29 and @xmath30 whose continued fraction is @xmath14 $ ] . for our use , we define @xmath31 to be @xmath32 after replacing @xmath33 with @xmath34 where @xmath35 .
therefore , we obtain @xmath36 for the above sequence of nonnegative integers .
it is clear that if @xmath14 $ ] is a continued fraction of @xmath13 , then @xmath37 $ ] is a continued fraction of @xmath38 . for any positive rational number @xmath13
, there is a continued fraction @xmath14 $ ] of nonnegative integers .
let @xmath14 $ ] be a continued fraction of @xmath13 of nonnegative integers , then @xmath39 and @xmath40 .
we first show that @xmath27 and @xmath41 are relatively prime . using the recursive formula for @xmath27
, we can see that if @xmath42 and @xmath43 , then @xmath44 .
now if repeat this process we obtain that @xmath45 .
+ now , we have @xmath46 for any sequence of nonnegative integers @xmath47 $ ] with @xmath48 , we have @xmath49 @xmath50 where @xmath51 , and @xmath52 are @xmath32 , and @xmath53 for the sequence @xmath54 $ ] respectively .
the result follows since @xmath55 and @xmath56
we first prove the following lemma that will be used in proof of the main result of this paper .
[ new ] let @xmath1 be a quasi - alternating link diagram at the crossing @xmath2 , then @xmath57 where @xmath58 is the link diagram obtained from @xmath1 by replacing the crossing @xmath2 by a rational tangle that extends @xmath2 with slope @xmath59 .
we prove the case where @xmath7 and the other case will follow by taking mirror image .
let @xmath14 $ ] be the continued fraction of nonnegative integers of the rational tangle with slope @xmath59 .
we use induction on the number of denominators of the continued fraction .
it is clear that the case holds for @xmath60 by just doing simple induction on the length of @xmath61 .
we assume that the result holds for all rational tangles of number of denominators equal to @xmath62 , i.e for the any rational tangle with continued fraction @xmath63 $ ] .
now we want to show that the result holds for any rational tangle of continued fraction @xmath14 $ ] .
again we apply induction but this time on @xmath64 . in the case
that @xmath65 , then the continued fraction is given by the sequence @xmath66 $ ] and hence the result follows by the induction hypothesis .
now suppose that the result holds for @xmath67 , we want to show the result for @xmath68 .
we have to consider two cases of @xmath69 being either even or odd .
first , we assume that @xmath69 is odd .
now we smooth one of the crossings in @xmath64 and we obtain @xmath70 in the case that @xmath69 is even the result follows since the first two terms in the second equality equal to @xmath71 and the last two terms equal to @xmath72 .
now we state our main theorem that is a natural generalization of ( * ? ? ? *
theorem.2.1 ) .
[ new2 ] if @xmath1 is a quasi - alternating link diagram at some crossing @xmath2 , then @xmath73 obtained by replacing the crossing @xmath2 by a product of rational tangles that extends the crossing @xmath2 is quasi - alternating .
let @xmath1 be a quasi - alternating link diagram at the crossing @xmath2 .
we may assume that @xmath74 as shown in figure [ figure ] by taking the mirror image if it is needed .
we follow the same notation as in the proof of ( * ? ? ?
* theorem.2.1 ) . to be more precise
, we let @xmath75 denote the link diagram with @xmath76 additional crossings inserted at @xmath2 , which are vertical half - twists if @xmath77 and horizontal half - twists if @xmath78 .
to show our result , we know that @xmath79 is quasi - alternating as a result of ( * ? ? ?
* theorem.2.1 ) .
moreover , @xmath79 is quasi - alternating at any one of these two crossings @xmath80 that replaces @xmath2 .
now we replace the crossing @xmath81 in @xmath79 by any rational tangle that extends @xmath81 to obtain a new quasi - alternating link diagram @xmath82 at any crossing of this rational tangle as a result of ( * ? ? ?
* theorem.2.1 ) .
we show that @xmath82 is quasi - alternating at the crossing @xmath83 . now
if we smooth @xmath82 at the crossing @xmath83 , we obtain @xmath84 that is the link diagram of @xmath58 where @xmath58 is the link diagram obtained from @xmath1 by replacing @xmath2 by the same rational tangle that has been used before to extend the crossing @xmath81 which quasi - alternating as a result of ( * ? ? ?
* theorem.2.1 ) .
also , we obtain the link diagram @xmath85 that is a connected sum of a rational link and @xmath4 which is quasi - alternating as a result of ( * ? ? ?
* lemma.2.3 ) . finally , we check that the determinant is additive at the crossing @xmath83 .
we use lemma [ new ] , to obtain @xmath86 ( * ? ? ?
* theorem.3.2(1 ) ) 1 . if @xmath87 , then the pretzel link @xmath88 is quasi - alternating for @xmath89 .
2 . if @xmath90 , then the pretzel link @xmath91 is quasi - alternating for @xmath92 .
we show the first case and the second case follows by taking the mirror image .
we may assume that @xmath93 .
we start with the pretzel link diagram @xmath94 to be our link @xmath1 .
it is clear that if we smooth @xmath1 at the crossing of the middle tassel then we obtain that @xmath95 which is quasi - alternating by ( * ? ? ?
* lemma.2.3 ) and @xmath96 which is also quasi - alternating since it is alternating .
now we have @xmath97 finally , we apply the theorem [ new2 ] to obtain the required result .
[ main ] the montesinos link @xmath98 in standard form is quasi - alternating if one of the following conditions is satisfied 1 .
@xmath99 ; 2 .
@xmath100 ; 3 .
@xmath101 with @xmath102 for some @xmath103 ; 4 .
@xmath104 with @xmath105 for some @xmath103 .
we prove each case separate 1 .
the montesinos link with the given condition has an equivalent description @xmath106 in standard form with @xmath99 .
the associated diagram is connected and alternating , so the link is quasi - alternating .
it is enough to show that its mirror image is quasi - alternating .
the mirror image of the given link has a description @xmath107 .
now this link has an equivalent description in standard form @xmath108 .
hence the result follows from the first case since @xmath109 .
3 . in this case , we start with the montesinos link diagram @xmath58 with a description @xmath110 with @xmath111 . + we show that this diagram is quasi - alternating at the only crossing in the second integer tangle .
it is easy to see that @xmath112 is a 4-plate and by @xcite we know that it is a rational link hence it is quasi - alternating .
also , @xmath113 is a connected sum of two rational links hence it is quasi - alternating by ( * ? ? ?
* lemma.2.3 ) .
to show the additivity of the determinant at this crossing , we use the main result of @xcite to obtain the determinant of the links @xmath58 and @xmath112 @xmath114 @xmath115 + also , we know that @xmath116 + for the determinant to be additive at that crossing , we have to have @xmath117 now since @xmath58 is quasi - alternating at the given crossing , then we apply theorem [ new2 ] to obtain a quasi - alternating montesinos link with the given description .
again it is enough to obtain the result for the mirror image of the given link .
the mirror image of the given link has a description @xmath118 .
now this link has an equivalent description in standard form @xmath119 .
finally , the result follows from the third case .
the above theorem generalizes the sufficient condition of ( * ? ? ?
* theorem.1.4 ) .
all cases of ( * ? ? ?
* theorem.1.4 ) follow directly from the cases of the above theorem by putting the pretzel link into a standard form except the second case .
this case follows from the second case of the above theorem since the link has a description @xmath120 that has an equivalent description @xmath121 with @xmath122 since @xmath123 for @xmath124 .
all classes of quasi - alternating montesinos links detected by widmar in ( * ? ? ?
* theorem.1.2 and the only remark ) fall into the third case of theorem [ main ] .
we prove each case separate , but first we assume @xmath125 is any positive rational tangle of slope @xmath13 . 1 .
the link @xmath126 with @xmath127 has an equivalent description in standard form @xmath128 .
so we have @xmath129 2 .
the link @xmath130 with @xmath131 and @xmath132 has an equivalent description in standard form @xmath133 .
so we have @xmath134 3 .
the link @xmath135 with @xmath136 has an equivalent description in standard form @xmath137 .
so we have @xmath138 4 .
the link @xmath139 with @xmath140 has an equivalent description in standard form @xmath141 .
so we have @xmath142 the following proposition generalizes ( * ? ? ?
* proposition.3.1 ) : the montesinos link @xmath98 in standard form is not quasi - alternating if @xmath143 .
the montesinos link in this case has an equivalent description @xmath144 that is a non - alternating diagram obtained by summing two strongly alternating 4-end tangles .
hence , it is adequate nonalternating diagram by ( * ? ? ?
* page.532 ) and by ( * ? ? ? * proposition.5.1 ) the link is not quasi - alternating since it is homologically thick . finally , we give some examples of montesinos knots that are known to be not quasi - alternating .
the knots @xmath145 and @xmath146 are montesinos knots that are given by the descriptions in standard form @xmath147 and @xmath148 respectively .
these two knots do not satisfy the condition mentioned in the third case of theorem [ main ] and they are known to be not quasi - alternating the first knot by ( * ? ? ?
* proposition.2.2 ) since it is the pretzel knot @xmath149 and the second knot by ( * ? ? ?
* theorem.1.3 ) .
the following list of montesinos knots in standard form are candidates for homologically thin non quasi - alternating knots as stated in ( * ? ? ?
* page.4 ) .
we checked by hand that these knots do not satisfy the condition mentioned in the third case of theorem [ main ] . 1 .
@xmath150 ; 2 .
@xmath151 ; 3 .
@xmath152 ; 4 .
@xmath153 .
we end this paper with the following conjecture that is supported by the above two examples .
theorem [ main ] characterizes all quasi - alternating montesinos links in standard form . in other words ,
any quasi - alternating montesinos link has to satisfy one of the conditions of the above theorem .
the first author thanks the max planck institute for mathematics for the kind hospitality during the course of this work .
f. bonahon , _ involutions et fibr@xmath154s de seifert dans les vari@xmath154t@xmath154s de dimension 3 _ , th@xmath154se de 3e cycle , orsay 1979 . c. bankwitz and h. g. schumann , _ ober viergeflechte _ ,
hamburg 10:263 - 284 , 1934 .
m. boileau and l. siebenmann , _ a planar classification of pretzel knots and montesinos knots _ , prepublication orsay 1980 .
g. burde and h. zieschang , _ knots _ , de gruyer stud . math . * 5 * , walter de gruyter , new york , 2003 .
j. greene , _ homologically thin , non - quasi - alternating links _ , math .
lett . , 17(1):39 - 49 , 2010 . s. jablan and r. sazdanovi@xmath155 , _ quasi - alternating links and odd homology : computations and conjectures _ , arxiv:0901.0075v2 [ math.gt ] 22 jan 2010 .
m. khovanov , _ patterns in knot cohomology .
i _ , experiment . math .
, 12(3):365 - 374 , 2003 . l. kauffman and s. lambropoulou , _ unknots and molecular biology _
, milan j. math .
, 74:227 - 263 , 2006 . l. kauffman and p. lopes , _ determinants of rational knots _ , discrete math .
theoret comput .
sci . , 11(2 ) : 111 - 122 , 2009 .
w. lickorish and m. thistlethwaite , _ some links with non - trivial polynomials and their crossing - numbers _ ,
math . helv . 63(4):527 - 539 , 1988 .
c. manolescu and p. ozsv@xmath156th .
_ on the khovanov and knot floer homologies of quasi - alternating links_. in proceedings of g@xmath157kova geometry - topology conference ( 2007 ) , 60 - 81 .
g@xmath157kova geometry / topology conference ( ggt ) , g@xmath157kova , 2008 . | we construct an infinite family of quasi - alternating links from a given quasi - alternating link by replacing a crossing by a product of rational tangles each of which extends that crossing .
consequently , we determine an infinite family of quasi - alternating montesinos links .
this family contains all the classes of quasi - alternating montesinos links that have been detected by widmar in @xcite .
we conjecture that this family contains all quasi - alternating montesinos links up to mirror image that are not alternating and this will characterize all quasi - alternating montesinos links . |
ultrahigh - energy ( @xmath9 ) cosmic rays provide a formidable beam to study particle collisions at center - of - mass energies and kinematic regimes not accessible at terrestrial accelerators .
the incident cosmic radiation interacts with the atomic nuclei of air molecules and produces air showers which spread out over large areas .
if the primary cosmic ray is a baryon , hundreds to thousands of secondary particles are usually produced at the interaction vertex , many of which also have energies above the highest accelerator energies @xcite .
these secondary products are of course intrinsically hadrons . generally speaking , by extrapolating final states observed at collider experiments
, we can infer that , for @xmath10 collisions at center - of - mass energy @xmath11 , the jet of hadrons contains about 75% pions ( including 25% @xmath7 s , in accord with isospin invariance ) , 15% kaons , and 10% nucleons @xcite . during the shower evolution ,
the hadrons propagate through a medium with an increasing density as the altitude decreases and the hadron - air cross section rises slowly with energy .
therefore , the probability for interacting with air before decay increases with rising energy . moreover , the relativistic time dilation increases the decay length by a factor @xmath12 , where @xmath13 and @xmath14 are the energy and mass of the produced hadron .
when the @xmath7 s ( with a lifetime of @xmath15 ) do decay promptly to two photons , they feed the electromagnetic component of the shower . for other longer - lived mesons ,
it is instructive to estimate the critical energy at which the chances for interaction and decay are equal . for a vertical transversal of the atmosphere
, such a critical energy is found to be : @xmath16 , @xmath17 , @xmath18 , @xmath19 @xcite .
the dominant @xmath20 branching ratios are to @xmath21 , to @xmath22 , to @xmath23 , and to @xmath24 , whereas those of the @xmath25 are to @xmath26 , to @xmath27 , and for @xmath28 we have @xmath29 , @xmath30 , @xmath31 , @xmath32 @xcite . with these figures in mind , to a first approximation it seems reasonable to assume that in each generation of particles about 25% of the energy is transferred to the electromagnetic shower , and all hadrons with energy @xmath33 interact rather than decay , continuing to produce the hadronic shower .
eventually , the electromagnetic cascade dissipates around 90% of the primary particle s energy and the remaining 10% is carried by muons and neutrinos .
as the cascade process develops in the atmosphere , the number of particles in the shower increases until the energy of the secondary particles is degraded to the level where ionization losses dominate . at this point
the density of particles starts to decline .
the number of particles as a function of the amount of atmosphere penetrated by the cascade ( @xmath34 in @xmath35 ) is known as the longitudinal profile .
a well - defined peak in the longitudinal development , @xmath36 , occurs where the number of @xmath37 in the electromagnetic shower is at its maximum .
@xmath36 increases with primary energy , as more cascade generations are required to degrade the secondary particle energies
. evaluating @xmath36 is a fundamental part of many of the composition analyses done when studying air showers .
the generic shower properties can be qualitatively well understood using the superposition principle , which states that a shower initiated by a nucleus with @xmath38 nucleons and energy @xmath39 behaves to a good approximation as the superposition of @xmath38 proton showers with initial energy @xmath40
this phenomenological assumption relies on the fact that the effect of nuclear binding must be negligible at extremely high energies .
thus , for a given total energy @xmath39 , showers initiated by a heavy nucleus have smaller @xmath36 than proton induced showers .
the integrated longitudinal profile provides a calorimetric measurement of the energy of the primary cosmic ray , with a relatively small uncertainty due to the correction for energy lost to neutrinos and particles hitting the ground .
the characteristics of the cascade depend dominantly on the elasticity ( fraction of incoming energy carried by the leading secondary particle ) and the multiplicity of secondary particles in the early , high - energy interactions .
modeling the development of a cosmic - ray air shower requires extrapolation of hadronic interaction models tuned to accommodate lhc data @xcite .
not surprisingly , such extrapolation usually leads to discrepancies between measured and simulated shower properties .
the hadronic interaction models are further constrained by independent measurements of @xmath36 and the density of muons at @xmath41 from the shower core @xmath42 @xcite .
the mean @xmath36 is primarily sensitive to the cross - section , elasticity , multiplicity , and primary mass .
the mean @xmath42 is primarily sensitive to the multiplicity , the @xmath7 energy fraction ( the fraction of incident energy carried by @xmath7 s in hadronic interactions ) , and the primary mass . over the past few decades , it has been suspected that the number of registered muons at the surface of the earth is by some tens of percentage points higher than expected with extrapolations of existing hadronic interaction models @xcite .
very recently , a study from the pierre auger collaboration @xcite has strengthened this suspicion , using a novel technique to mitigate some of the measurement uncertainties of earlier methods @xcite .
the new analysis of auger data suggests that the hadronic component of showers ( with primary energy @xmath43 ) contains about @xmath44 to @xmath45 more muons than expected .
the significance of the discrepancy between data and model prediction is somewhat above @xmath46 . changing the @xmath7 energy fraction or suppressing @xmath7 decay are the only modifications which can be used to increase @xmath42 without coming into conflict with the @xmath36 observations @xcite . distribution .
since the muon lateral distribution function is a steeply falling function of the radius , a larger production angle would increase @xmath42 .
however , as shown in @xcite , the measured zenith angle dependence of the ground signal vetoes the correlated flattening of the muon lateral distribution function necessary to accommodate auger data . ]
several new physics models have been proposed exploring these two possibilities @xcite . in this paper
we adopt a purely phenomenological approach to develop an alternative scheme .
in sharp contrast to previous models , our proposal is based on the assumption that ultrahigh energy cosmic rays are heavy ( or medium mass ) nuclei .
our work builds upon some established concepts , yet contravenes others .
we conceive the production and separation of strangeness in a baryon - rich _ centauro - like _ fireball before its spontaneous _ explosive _ decay @xcite . at production
the fireball has a very high matter density and consists of gluons and two flavors of light quarks ( @xmath2 ) .
because the fireball is formed in the baryon - rich projectile fragmentation region , the high baryochemical potential slows down the creation of @xmath47 and @xmath48 pairs , resulting in gluon fragmentation dominated by @xmath49 .
the larger amount of @xmath50 and @xmath51 with respect to @xmath52 and @xmath53 gives a higher probability for @xmath54 to find @xmath50 or @xmath51 and form @xmath20 or @xmath55 than for @xmath56 to form the antiparticle counterparts .
prompt _ hard _
kaon emission then carries away all strange antiquarks and positive charge , lowering somewhat the initial temperature and entropy .
the late - stage hadronization is characterized by production of @xmath57 , @xmath58 , nucleons , pions , and strange baryons .
overall , after hadronization is complete , the relative density of strange hadrons is significantly enhanced over that resulting from a hadron gas alone , damping the @xmath7 energy fraction .
the layout of the paper is as follows .
we begin in sec .
[ s2 ] with an overview of the fireball paradigm and make a critical review of the available experimental data from colliders .
after that , in sec .
[ s3 ] we discuss the particulars of air shower evolution and present our results from monte carlo simulations .
we show that the formation of a plasma , with gluons and massive quarks , could play a key role in the hadronization process , modifying shower observables .
in particular , we demonstrate that air showers triggered by a fireball explosion tend to increase @xmath42 , and under some reasonable assumptions can accommodate auger data .
finally , we summarize our results and draw our conclusions in sec .
it has long been suspected that , for systems of high energy density , the elementary excitations can be safely approximated by an ensemble of free quarks and gluons at finite temperature and baryon number density @xcite .
this is because when the energy density is extremely high , the expected average particle separation is so small that the effective strength of interactions is weak ( asymptotic freedom ) @xcite . for many purposes ,
the order of the energy density of matter inside a heavy nucleus is immense , @xmath59 where @xmath60 is the proton mass , @xmath38 is the nuclear baryon number , and @xmath61 is the nuclear radius .
however , at typical energy densities inside of nuclei , quarks and gluons are very probably confined , on the average , inside of hadrons such as protons , neutrons , or pions . for @xmath62
all of these hadrons could be squeezed so tightly together that on the average they will all overlap and the system would become an unconfined plasma of quarks and gluons , which are free to roam the system .
the energy scale at which hadrons begin to overlap is above the energy density of matter inside the proton , @xmath63 , where we have taken the radius of the proton as measured in electron scattering @xmath64 . indeed , using phenomenological considerations
it is straightforward to estimate that the critical energy density to form a non - hadronic medium is around @xmath65 @xcite .
this result is supported by high statistics lattice - qcd calculations , which yield @xmath66 , where we have taken @xmath67 @xcite . besides the early universe , the conditions of extremely high temperature and density necessary for the appearance of unconfined quark and gluons could occur in at least two other physical phenomena : _ ( i ) _ the interiors of neutron stars @xcite and _ ( ii ) _ high - energy nucleus - nucleus collisions , whether artificially produced at accelerators or naturally occurring interactions of cosmic rays with particles in the earth s atmosphere @xcite .
we estimate the energy density in nucleus - nucleus collisions of cosmic rays following @xcite .
we assume there exists a `` central - plateau '' structure in the inclusive particle productions as function of the pseudorapidity variable .
the energy per particle should be of the order of the typical transverse momentum per particle @xmath68 .
more precisely , in the fireball frame each isotropically emitted particle has an energy given by @xmath69^{1/2 } \
, , \ ] ] where @xmath70 is the particle s mass @xcite .
the energy content is approximately @xmath71 , where @xmath72 is the charged - hadron multiplicity per @xmath10 collision , and @xmath73 the number of nucleon - nucleon interactions in the fireball during the collision .
consider two nuclei of transverse radius @xmath74 which collide in the center - of - mass frame .
the longitudinal size of the nuclei is lorentz contracted forming a transverse thin slab at mid - pseudorapidity .
the initial _ fireball _ volume is @xmath75 , where @xmath76 is the typical time scale for the formation and decay of a central fireball and hence @xmath77 is the longitudinal size , with @xmath78 the pseudorapidity width at @xmath76 .
the time scale can be estimated as the time to traverse at light speed the fireball diameter , i.e. , @xmath79 .
the energy density during the cosmic ray ( cr ) collision is then @xmath80 the average transverse - momentum of charged hadrons produced in @xmath10 collisions can be parametrized as a function of the squared center - of - mass energy , @xmath81 with @xmath56 in appropriate units of @xmath82 @xcite . on the other hand , for a cosmic - ray nitrogen nucleus ( for simplicity , we choose the beam nucleus to be that of the dominant element in air , with total @xmath83 however , the electronic binding is @xmath84 ev , whereas the nuclear binding is @xmath85 mev , so the energy scales and length scales differ by @xmath84 a million , and the dimerization is not expected to survive the cosmic - ray production process .
the nuclear binding of the single nucleus apparently does survive this process . ] ) of @xmath86 colliding with an air nucleus , we have @xmath87 , yielding @xmath88 and @xmath89 @xcite . throughout
, we take @xmath90 as fiducial in our calculations and assume that half of these number of nucleons interact per collision , producing the fireball . therefore
, taking an effective mass @xmath91 and @xmath92 , the energy density in such a scattering process is @xmath93 well above @xmath94 , complying with the requirement for the formation of a deconfined thermal fireball .
we envision the fireball as a plasma of _ massive _ quarks and gluons maintained in both kinetic and chemical equilibrium . because the total number of particles is allowed to fluctuate , we adopt the viewpoint of the grand canonical ensemble . in this representation , which allows exchange of particles among the system and the reservoir ,
the control variables are the baryochemical potential @xmath95 and the temperature @xmath96 . in the limit @xmath97 and @xmath98 the system becomes the vacuum . the momentum distribution functions @xmath99 can be approximated by fermi - dirac @xmath100 and bose - einstein @xmath101 statistics , @xmath102^{-1 } \
, , \label{fd - be}\ ] ] yielding following equilibrium number densities @xmath103 where the index @xmath104 runs over @xmath105 , @xmath106 because of the total strangeness neutrality of the initial state , @xmath107 and @xmath108 are the particle s momentum and mass , and @xmath109 is the spin - color degeneracy factor .
the plus sign is to be used for quarks and minus sign for gluons , with @xmath110 .
the density of the strange quarks is found to be ( two spins and three colors ) @xmath111 where @xmath112 is the second order modified bessel function @xcite .
in addition , there is a certain light antiquark density ( @xmath113 stands for either @xmath52 or @xmath53 ) : @xmath114 note that the baryonic chemical potential exponentially suppresses the @xmath115 pair production .
this reflects the chemical equilibrium between @xmath115 and the presence of a light quark density associated with the net baryon number .
now , since the chemical potentials satisfy @xmath116 , it follows that @xcite @xmath117 note that this result is exact and not a truncated series .
the gluon density also follows from ( [ numberd ] ) and is given by @xmath118 where @xmath119 is the riemann function .
( see appendix [ app_a ] for details . ) by comparing ( [ ns ] ) and ( [ nqbar ] ) , it is straightforward to see that there are often more @xmath54 than anti - quarks of each light flavor , @xmath120 for @xmath121 and @xmath97 , there are about as many @xmath50 and @xmath51 quarks as there are @xmath56 quarks . for @xmath122 , many of the properties of the quark - gluon plasma can be calculated in the framework of thermal perturbation theory . neglecting quark masses in first - order perturbation theory , the energy density of the _ ideal _ quark gluon plasma is found to be @xmath123 \pi^2 t^4 + \left(\frac{n_q}{4 } - \frac{6 \alpha_s}{\pi } \right ) \mu_q^2 t^2 \nonumber \\ & + & \left(\frac{n_q}{8 } - \frac{3 \alpha_s}{\pi } \right ) \frac{\mu_q^4}{\pi^2 } + b \
, , \label{eideal}\end{aligned}\ ] ] where @xmath124 and @xmath125 are the gluon and quark degrees of freedom , @xmath126 is the qcd coupling constant , and @xmath127 is the difference between the energy density of the perturbative and the nonperturbative qcd vacuum ( the bag constant ) @xcite .
one observes that ( [ eideal ] ) is essentially the equation of state of a gas of massless particles with corrections due to qcd perturbative interactions , which are always negative and thus reduce the energy density at given temperature @xmath96 .
since the @xmath50 and @xmath51 flavors are almost massless even in the fireball phase , we can get an estimate of the baryochemical potential by considering only the contribution from the nearly massless ( even in the fireball phase ) two quarks at @xmath128 . for two quark flavors , @xmath129 and
so ( [ eideal ] ) simplifies to @xmath130 substituting ( [ ecr ] ) into ( [ eidealt0 ] ) , with @xmath131 @xcite and the mit bag constant @xmath132 @xcite , we obtain @xmath133 the temperature of the plasma can be approximated by @xcite @xmath134 a point worth noting at this juncture is that the shapes of the @xmath135 spectra are expected to be determined by an interplay between two effects : the thermal motion of the particles in the fireball and a pressure - driven radial flow , induced by the fireball expansion . to disentangle the two contributions , namely thermal motion and transverse flow
, one has to rely on model calculations which seem to indicate that the observed temperature in the particle spectra is close if not exactly equal to the temperature value that would be present in the chemically equilibrated fireball @xcite . substituting ( [ mub ] ) and
( [ t ] ) into ( [ stoq ] ) , with @xmath136 , we obtain @xmath137 .
the pion - to - nucleon density ratio is found to be @xmath138/t \right\ } \
, , \label{pin}\ ] ] where the factor 3/2 comes from the number of the particle species , @xmath139 is the pion mass , @xmath140 is the average baryon mass , and @xmath141 is a normalization constant fixed by the choice of the boundary values @xcite .
it is worth commenting on an aspect of this analysis which may seem discrepant at first blush . from relativistic heavy ion experiments
one infers a temperature falling with the chemical potential , from @xmath142 . however , to accommodate the muon excess in auger data one needs @xmath143 .
details of this discrepancy and eventual accommodation are given in appendix [ app_ref ] . in our analysis
we will adopt a pragmatic approach and avoid the details of theoretical modeling of the hadronization process . which of the two points of view one may find more convincing
, it seems most conservative at this point to depend on experiment to resolve the issue .
the multiplicity ratio @xmath144 has been eyeball - fitted to reproduce the anomalous muon signal observed in auger data .
we show the success of this protocol in the next section . here
we simply note that these ratios are in partial agreement with ( [ stoq ] ) and ( [ pin ] ) for @xmath145 .
we now make contact with the shower evolution . as a first - order approximation
we adopt a basic phenomenological approach .
namely , we assume that the hadronic shower carries a fraction @xmath146 of the total energy of the primary cosmic ray @xmath39 , which scales as @xmath147 where @xmath148 is the number of generations required for most pions to have energies below @xmath149 and @xmath150 is the average fraction in electromagnetic particles per generation @xcite . in the canonical framework
hadronic interaction models transfer about 25% of the energy to the electromagnetic shower .
conspicuously , the production of light hadrons in this canonical framework is virtually local in rapidity and , therefore , since the interaction models are tuned to fit collider data , @xmath150 would remain approximately constant with energy .
we have found that for the considerations in the present work we can safely approximate that each interaction diverts about 75% of the available energy into pions , 15% into kaons , and 10% continues as nucleons . roughly speaking ,
this is consistent with a multiplicity ratio @xmath151 .
now , taking then @xmath152 and @xmath153 we conclude that @xmath154 . and that of the muon is @xmath155 , where @xmath156 is the ratio of muon to the pion mass squared .
thus , we can safely neglect the missing energy in neutrinos .
] this is in agreement with the estimates of @xcite which indicate that the number of interactions needed to reach @xmath149 is @xmath157 for @xmath158 , respectively .
a comprehensive study of the uncertainties associated with the modeling of hadronic interactions indicates that a simple reduction of @xmath148 to increase @xmath146 correlates with @xmath36 , which becomes too shallow before @xmath42 is sufficiently increased to accommodate auger data @xcite .
now , if the shower is initiated in a fireball explosion in the first generation , then we have seen that we can approximate the multiplicity ratios by @xmath159 .
moreover , this is a completely inelastic process , but differs from the usual inelastic processes in that a fireball is also produced .
we assume this fireball creates a higher multiplicity of particles and to a first approximation equally partitions energy among the secondaries ( thereby negating a large leading particle effect ) .
the fireball production thus accelerates the cooling of the cascade and could reduce the number of generations .
we denote with @xmath160 the numbers of generations for a shower initiated by a fireball .
we may then assume that @xmath161 or 8 would be enough to reach the critical shower energy @xmath149 . to include the fireball effects we rewrite ( [ fh1 ] ) as @xmath162 where @xmath163 is the fraction of electromagnetic energy emitted by the fireball .
we arrived at @xmath164 from considering the @xmath7 fraction , equal to 1/3 the total @xmath165 fraction , as before . by substituting our fiducial numbers in ( [ fh2 ] )
it is straightforward to see that the hadronic shower is increased by about 30% if @xmath166 and by about 70% if @xmath167 , in agreement with recent auger results @xcite . of course
, ( [ fh2 ] ) is a dreadful simplification of the shower evolution . as we have noted before , the shower evolution depends on primary energy , as well as the elasticity and particle multiplicity which also depend on @xmath39 .
note that all these restrictions have not been specified explicitly as separate parameters in ( [ fh2 ] ) , but rather as a combined constant @xmath160 .
moreover , heavy meson production must also be taken into consideration when modeling the shower evolution ( distinctively , @xmath168 and @xmath169 contribute about 4% to the electromagnetic cascade ) .
unfortunately , it is difficult to estimate accurate parameter values of the shower evolution in a simple analytical fashion .
for such a situation , a full - blown simulation may be the only practical approach . as a second order approximation
, we estimate the fireball spectrum and propagate the particles in the atmosphere using the algorithms of aires ( version 2.1.1 ) @xcite .
most of the large multiplicity of observable quanta emitted in the fireball is expected to come through hadronic jets produced by the quarks .
we assume that @xmath170 nucleons ( each with initial energy @xmath40 ) produce the fireball and that the remaining nucleons scatter inelastically at the collision vertex .
we further assume that for the nucleons producing the fireball , each of the valence quarks interact to give dijet final states without any leading particle .
we calculate the total energy of each jet @xmath171 in the rest frame of the fireball from the momentum fraction carried by the up and down quarks , using cteq6l parton distribution functions @xcite .
the precise nature of the fragmentation process is unknown .
we adopt the quark @xmath172 hadron fragmentation spectrum originally suggested by hill
@xmath173 \,\,(1-x)^2 \nonumber \\ & \times & \left[x \sqrt{\ln(1/x)}\right]^{-1}\ , .
\label{c}\end{aligned}\ ] ] where @xmath174 , with @xmath13 the energy of any hadron in the jet .
this is consistent with the so - called `` leading - log qcd '' behavior and seems to reproduce quite well the multiplicity growth as seen in colliders experiments @xcite .
@xmath175 provides a reasonable parametrization of ( [ c ] ) for @xmath176 .
and so we set the infrared cutoff to @xmath177 .
the main features of the jet fragmentation process derived from this simplified parameterization are listed in table [ table ] . using the multiplicity ratios derived in the previous section and the fractional equivalent energies given in table [ table ] , we construct the fireball particle spectra .
more specifically , for each jet , we start first from @xmath178 and integrate down in @xmath179 until three leading hadron particles are obtained .
the resulting interval in @xmath179 is @xmath180 , as shown in column three , with @xmath181 and @xmath182 values listed in columns two and one , respectively .
we assign one of three species - types to each hadron using the @xmath183 weights .
the energy fraction of the jet contained in these three hadrons is given in column four , and the average energy fraction per each of these three hadrons , denoted @xmath184 , is given in column five .
next , we duplicate the procedure for the remaining hadron species following the splitting of fractional energies , @xmath185 , as given in column 3 , and assigning to each hadron in the interval the corresponding @xmath184 from column 5 .
note that for each subsequent @xmath186 interval , the fractional hadron energy @xmath184 is significantly smaller ; this feature allows us to sensibly truncate the process after five intervals at our cutoff value @xmath187 .
the mean energy of each of the fifth and final batch of 30 hadrons produced by wee partons is @xmath188 of the energy of each hadron produced by large @xmath179 partons in first batch .
the average particle multiplicity per jet is the sum of column three entries , approximately 54 .
charge and strangeness conservation are separately imposed by hand .
[ table ] .properties of jet hadronization @xcite .
columns one and two define the interval @xmath189 over which @xmath190 hadrons are made .
@xmath190 is listed in the third column .
column four presents the fractional energy of the jet contained in these @xmath190 hadrons , and column five gives the mean fractional energy per each of these hadrons ( @xmath184 ) . [ cols="^,^,^,^,^",options="header " , ] all secondary particles are boosted to the laboratory frame .
the particles are tightly beamed due to their very high boost .
the boosted secondaries are then injected into aires as primaries of an air shower initiated at the collision vertex .
the vertex of the primary interaction is determined using the mean free path of a @xmath191n nucleus with @xmath192 .
we set the observation point at 1.5 km above sea level , which is the altitude of auger .
all shower particles with energies above the following thresholds were tracked : 750 kev for gammas , 900 kev for electrons and positrons , 10 mev for muons , 60 mev for mesons , and 120 mev for nucleons .
the geomagnetic field was set to reproduce that prevailing upon the auger experiment . for further details ,
see appendix [ app_b ] .
secondary particles of different types in individual showers were sorted according to their distance @xmath51 to the shower axis .
our results are encapsulated in fig .
[ fig:2 ] , where we compare the density distribution of @xmath193 at ground level for a typical nitrogen shower processed by the aires kernel with that of a nitrogen fireball explosion .
the vertical axis is given in arbitrary units to indicate the large systematic uncertainty in the normalization of the lateral distribution , which is induced by the different predictions of high - energy hadronic event generators .
importantly , the comparison of hadron and fireball fluxes is _ not _ arbitrary .
it is easily seen that the number of muons at @xmath194 is about 40% higher in the fireball - induced shower .
the third order approximation should include a precise determination of the fireball particle spectra using scaling hydrodynamic equations , which contain the probability amplitudes for @xmath5 production and annihilation both in the fireball phase and in the hadron gas phase @xcite .
it should also contain a thorough high - level competitive analysis of theoretical systematics emanating from hadronic interaction models .
this rather ambitious project is beyond the scope of this paper , but will be the topic of a future publication .
teasing out the physics of ultrahigh - energy cosmic rays has proven to be extraordinarily challenging .
the pierre auger observatory employs several detection methods to extract complementary information about the extensive air showers produced by ultrahigh - energy cosmic rays @xcite .
two types of instruments are employed : cherenkov particle detectors on the ground sample air shower fronts as they arrive at the earth s surface , whereas fluorescence telescopes measure the light produced by air - shower particles exciting atmospheric nitrogen .
these two detector systems provide complementary information , as the surface detector ( sd ) measures the lateral distribution and time structure of shower particles arriving at the ground , and the fluorescence detector ( fd ) measures the longitudinal development of the shower in the atmosphere .
a subset of _ hybrid _ showers is observed simultaneously by the sd and fd .
these are very precisely measured and provide an invaluable tool for energy calibration , minimizing systematic uncertainties and studying composition by simultaneously using sd and fd information .
very recently , the pierre auger collaboration exploited the information in individual hybrid events initiated by cosmic rays with @xmath195 to study hadronic interactions at ultrahigh energies @xcite .
the analysis indicates that the observed hadronic signal of these showers is significantly larger ( 30 to 60% ) than predicted by the leading lhc - tuned models , with a corresponding excess of muons .
the significance of the discrepancy between data ( 411 hybrid events ) and model prediction is above about @xmath46 .
a deployment of a @xmath196 scintillator on top of each sd is foreseen as a part of the augerprime upgrade of the observatory to measure the muon and electromagnetic contributions to the ground signal separately @xcite .
this will provide additional information to reduce systematic uncertainties and perhaps increase the significance of the muon excess . even though the excess is not statistically significant yet , it is interesting to entertain the possibility that it corresponds to a real signal of qcd dynamics flagging the onset of deconfinement . in this paper
, we have proposed a model that can explain the observed excess in the muon signal .
we have assumed that ultrarelativistic nuclei ( @xmath197 ) that collide in the upper atmosphere could create a deconfined thermal fireball which undergoes a sudden hadronization . at production ,
the fireball has a very high matter density and consists of gluons and two flavors of light quarks ( @xmath2 ) .
because the fireball is formed in the baryon - rich projectile fragmentation region , the high baryochemical potential damps the production of @xmath47 and @xmath4 pairs , resulting in gluon fragmentation mainly into @xmath5 .
the strange quarks then become much more abundant and upon hadronization the relative density of strange hadrons is significantly enhanced over that resulting from a hadron gas .
we have shown that the augmented production of strange hadrons by the fireball , over that resulting from a hadron gas alone , provides a mechanism to increase the muon content in atmospheric cascades by about 40% , in agreement with the data of the auger facility .
contrary to previous proposals @xcite to explain the muon excess in auger data our model relies on the assumption that ultrahigh - energy cosmic rays are heavy ( or medium mass ) nuclei . as noted elsewhere @xcite , upper limits on the cosmic diffuse neutrino flux provide a constraint on the proton fraction in ultrahigh - energy cosmic rays , and therefore can be used to set indirect constraints on the model proposed herein . in particular ,
the nearly guaranteed flux of cosmogenic neutrinos is a decay product from the generated pions in interactions of ultrahigh energy cosmic rays with the cosmic microwave background and related radiation @xcite .
the spectral shape and intensity of this flux depend on whether the cosmic - ray particles are protons or heavy nuclei . for proton primaries , the energy - squared - weighted flux peaks between @xmath198 and @xmath199 , and the intensity is around 1 in waxman - bahcall ( wb ) units @xcite .
@xcite . ] for heavy nuclei , the peak is at much lower energy ( around @xmath200 @xcite ) and the intensity is about 0.1 to 0.01 wb , depending on source evolution @xcite . the sensitivity of existing neutrino - detection facilities has a reach to 1 wb , challenging cosmic - ray models for which the highest energies are proton - dominated @xcite .
next - generation neutrino detectors will systematically probe the entire range of the parameter space of cosmogenic neutrinos @xcite .
observation of the cosmogenic neutrino flux with intensity of 0.1 to 0.01 wb could become the smoking gun for the ideas discussed in this paper .
complementary information can be obtained with accompanying cosmogenic photons @xcite . as a matter of fact
, the pierre auger observatory has begun to probe the region of the parameter space relevant for proton primaries @xcite .
a third probe of proton - dominated uhecr models is the extragalactic gamma - ray background seen by fermi - lat ( and , in the future , cta ) @xcite . for heavy nuclei
, however , the cosmogenic photon intensity is almost negligible and therefore can not be used ( for the time being ) as a harbinger signal @xcite . for @xmath201 , the mean and dispersion of @xmath36 inferred from fluorescence auger data point to a light composition ( protons and helium ) towards the low end of this energy bin and to a large light - nuclei content ( around helium ) towards the high end ( see fig . 3 in @xcite ) .
however , when the signal in the water cherenkov stations ( with sensitivity to both the electromagnetic and muonic components ) is correlated with the fluorescence data , a light composition made up of only proton and helium becomes inconsistent with observations @xcite .
the hybrid data indicate that intermediate nuclei , with @xmath202 , must contribute to the energy spectrum in this energy bin .
moreover , a potential iron contribution can not be discarded . at this stage , it is worthwhile to point out that the production of a fireball may modify the shower evolution .
for example , after emitting @xmath20 and @xmath55 , the fireball has a finite excess of @xmath56 quarks , and because of the @xmath56 quarks stabilizing effects , the fireball could form heavy multiquark droplets @xmath203 , with large strangeness @xcite .
the energy lost by the @xmath203 particles during collision with nucleons is primarily through hard scattering , and so the fractional energy loss per collision is @xmath204 @xcite .
@xmath203 production may thus slow down the shower evolution . although this effect has not been included in our simulations
, one may wonder whether the structure observed in the elongation rate above about @xmath205 @xcite could be ascribed to the onset of @xmath203 production . of course one would not expect a fireball to be created when nuclei just slide along each other .
the admixture of peripheral and fireball collisions would then produce large fluctuations in the number of muons at ground level .
however , since the critical energy for charged pions @xmath149 and kaons @xmath206 is roughly the same , the elongation rate of the muon channel would be almost unaltered and so the muon shower maximum @xmath207 would have small fluctuations .
moreover , if the fireball indeed modifies the elongation rate of the electromagnetic shower , the peripheral collisions would also tend to increase the dispersion of @xmath36 , mimicking what is expected for a light composition in the canonical framework where no fireballs are being produced in this energy range . in closing ,
we comment on the differences between our model and the proposal by farrar and allen ( fa ) @xcite , which also relies on qcd dynamics at high temperature . to better understand the differences between these models
we first note that at low energies , qcd exhibits two interesting phenomena : confinement and ( approximate ) spontaneous chiral - symmetry breaking .
these two phenomena are strong - coupling effects , invisible to perturbation theory .
the confinement force couples quarks to form hadrons and the chiral force binds the collective excitations to goldstone bosons . as a matter of course
, there is no relation _ a priori _ between these two phenomena ; in thermodynamics , the associated scales are characterized by two distinct temperatures , @xmath208 and @xmath209 . for @xmath210 , hadrons dissolve into quarks and gluons , whereas for @xmath211 , the chiral symmetry is fully restored and quarks become massless , forming an ideal quark - gluon plasma .
the @xmath212 phase diagram of hadronic matter thus contains a _ confined phase _ consisting of an interacting gas of hadrons ( a _ resonance gas _ ) and a _ deconfined phase _ comprising a ( non - ideal ) gas of quarks and gluons @xcite .
the _ phase boundary _ reflects the present uncertainties from lattice qcd extrapolated to finite baryochemical potential .
the intermediate region in - between the hadron gas and the ideal quark - gluon plasma is the domain of the thermal fireball .
the existence of this intermediate region with deconfined massive quarks and gluons is also conjectured from high - statistics lattice - qcd calculations , which indicate that , for @xmath213 , the energy density is about 85% of the stefan - boltzmann energy density for the ideal quark - gluon plasma @xcite .
note that this temperature is not inconsistent with our estimate in ( [ t ] ) . in the fa model ,
the pion suppression is a direct consequence of massless quarks living above the chiral symmetry restoration temperature , i.e. , @xmath214 . in our model , however , the pion suppression is the result of the large baryochemical potential which forbids the creation of light @xmath47 and @xmath4 pairs , allowing abundant production of massive @xmath5 via gluon fragmentation .
this process naturally occurs in the fireball boundary phase ( where @xmath215 ) , and is a consequence of the high nucleon density of lorentz - boosted nuclei . in principle
, it is possible that the muon excess observed in auger data originates in a combination of these two high - temperature qcd phenomena .
note that if ultrahigh - energy cosmic rays are heavy ( or medium mass ) nuclei and the observed muon excess is not the result of a large baryochemical potential suppressing the production of @xmath216 and @xmath4 pairs , but rather an effect of chiral symmetry restoration , then the excess should also be visible at @xmath217 , which corresponds to the center - of - mass energy in collisions of projectile cosmic ray protons with @xmath218 .
this disparity could be use to discriminate among dominance between these two pion - suppression mechanisms .
on the one hand , a model consistent with all data requiring heavy nuclei at the high - energy end of the spectrum is generally considered a bit disappointing @xcite , especially for neutrino and cosmic - ray astronomy . on the other hand ,
we have shown that even if heavy nuclei dominate at the highest energies , upon scattering in the earth s atmosphere these nuclei could become compelling probes of qcd dynamics at high temperatures , particularly in the not - so - well - understood fireball boundary phase .
should this be the case , future augerprime data would provide relevant information on the strange quark density of the nucleon , complementing measurements at heavy - ion colliders .
for the sake of completeness , in this appendix we provide all of the formulae used for computing the number densities of @xmath219 , @xmath220 , and @xmath221 .
the fermi - dirac and bose - einstein distributions ( [ fd - be ] ) can be rewritten as infinite sums of boltzmann distributions , @xmath222 following @xcite we introduce the dimensionless variables , @xmath223 and @xmath224 : latexmath:[\[\begin{aligned } z = \frac{m_i}{t } ; \quad \tau=\frac{\sqrt{p^2+m_i^2}}{t } ; \quad in terms of @xmath224 and @xmath223 , the number density for the boltzmann distribution can be written as , @xmath226 a closed form expression can be found for @xmath227 in terms of the modified bessel function @xmath228 note that @xmath229 has another representation , which follows from ( [ kn1 ] ) by partial integration , @xmath230 the modified bessel function has a recurrence relation , @xmath231 such that if the expressions for @xmath232 and @xmath233 are known , all the others can be easily obtained . from ( [ kn2 ] )
it is straightforward to obtain @xmath234 and so ( [ nb ] ) becomes @xmath235 for the fermi - dirac distribution , the number density is found to be @xmath236 for the bose - einstein distribution , the number density is given by @xmath237 in the limit @xmath238 and @xmath239 we immediately obtain @xmath240 where @xmath119 is the riemann function . finally , using ( [ fd ] ) and ( [ be ] ) , it is straightforward to obtain ( [ ns ] ) , ( [ nqbar ] ) and ( [ ng ] ) .
our best understanding of the thermodynamical properties of qcd at vanishing baryon density is rooted on high statistics lattice qcd numerical simulations . however , if the baryon density is non - zero , these simulations break down .
the only physically relevant analyses for which the obstacles of lattice qcd can be circumvented are those dealing with small baryon density or , more accurately , small @xmath95 .
namely , if we are interested in the observable @xmath242 , we can expand it in powers of @xmath95 as @xmath243 and try to determine the series coefficients @xcite .
which values of @xmath95 can be considered _ small enough _ to give reliable results with this procedure is something that can only be determined _ a posteriori _ by the convergence property of the series , limited by the accuracy in the evaluation of the expansion coefficients .
one such observable is the quark - hadron crossover line .
data from heavy - ion colliders suggest that the energy dependence of this line can be parametrized as @xmath244 with @xmath245 , @xmath246 and @xmath247 , and that the baryochemical potential can be parametrized as @xmath248 with @xmath249 and @xmath250 @xcite .
interestingly , the lattice - qcd calculation converges towards the quark - hadron crossover line as but it appears to depart from this line at large values of @xmath95 @xcite .
this a widely discussed feature @xcite which has , however , not been conclusively understood .
( several hadronization schemes have been proposed , see e.g. @xcite .
they differ in the geometry and in the flow velocity profile . )
this perplexing region may well be relevant to cosmic - ray observations .
our results in the region shown in fig
. [ fig:1 ] , suggest that this is so .
the aires simulation engine @xcite provides full space - time particle propagation in a realistic environment , taking into account the characteristics of the atmospheric density profile ( using the standard us atmosphere ) , the earth s curvature , and the geomagnetic field ( calculated for the location of auger with an uncertainty of a few percent @xcite ) .
the following particles are tracked in the monte carlo simulation : photons , electrons , positrons , muons , pions , kaons , eta mesons , lambda baryons , nucleons , antinucleons , and nuclei up to @xmath251 .
the high - energy collisions are processed invoking external hadronic event generators , whereas the low - energy ones are processed using an extension of the hillas splitting algorithm @xcite .
the aires program consists of various interacting procedures that operate on a data set with a variable number of records .
several data arrays ( or stacks ) are defined .
every record within any of these stacks is a particle entry and represents a physical particle .
the data contained in every record are related to the characteristics of the corresponding particle .
the particles can move inside a volume within the atmosphere where the shower takes place .
this volume is limited by the ground , the injection surfaces , and by vertical planes which limit the region of interest . before starting the simulation ,
all the stacks are empty .
the first action is to add the first stack entry , which corresponds to the primary particle
. then the stack processing loop begins .
the primary is initially located at the injection surface , and its downwards direction of motion defines the shower axis . after the primary s fate has been decided , the corresponding interaction begins to be processed .
the latter generally involves the creation of new particles which are stored in the empty stacks and remain waiting to be processed .
particles entries are removed when one of the following events happen : _
( i ) _ the energy of the particle is below the selected cut energy ; _ ( ii ) _ the particle reaches ground level ; _ ( iii ) _
a particle going upwards reaches the injection surface ; _ ( iv ) _ a particle with quasi - horizontal motion exists the region of interest . after having scanned all the stacks
, it is checked whether or not there are new particle entries pending further processing .
if the answer is positive , then all the stacks are scanned once more ; otherwise the simulation of the shower is complete . for the present analysis ,
we use the aires module for _ special and/or multiple primary particles_. this useful feature allows to dynamically call a user - defined module that tracks the interactions of a bundle of particles returning a handy list of secondaries , which can be conveniently controlled by the propagating engine for further processing . c. a. garcia canal , s. j. sciutto and t. tarutina , testing hadronic interaction packages at cosmic ray energies , phys .
d * 79 * , 054006 ( 2009 ) doi:10.1103/physrevd.79.054006 [ arxiv:0903.2409 [ astro-ph.he ] ] .
p. gondolo , g. ingelman and m. thunman , charm production and high - energy atmospheric muon and neutrino fluxes , astropart .
phys . * 5 * , 309 ( 1996 ) doi:10.1016/0927 - 6505(96)00033 - 3 [ hep - ph/9505417 ] . c. patrignani
_ et al . _
[ particle data group collaboration ] , review of particle physics , chin .
c * 40 * , no .
10 , 100001 ( 2016 ) .
doi:10.1088/1674 - 1137/40/10/100001 a. aab _ et al .
_ [ pierre auger collaboration ] , testing hadronic interactions at ultrahigh energies with air showers measured by the pierre auger observatory , phys . rev
. lett . * 117 * , 192001 ( 2016 ) doi:10.1103/physrevlett.117.192001 [ arxiv:1610.08509 [ hep - ex ] ] .
l. anchordoqui , m. t. dova , a. g. mariazzi , t. mccauley , t. c. paul , s. reucroft and j. swain , high energy physics in the atmosphere : phenomenology of cosmic ray air showers , annals phys . * 314 * , 145 ( 2004 ) doi:10.1016/j.aop.2004.07.003 [ hep - ph/0407020 ] .
d. denterria , r. engel , t. pierog , s. ostapchenko and k. werner , constraints from the first lhc data on hadronic event generators for ultra - high energy cosmic - ray physics , astropart .
phys . *
35 * , 98 ( 2011 ) doi:10.1016/j.astropartphys.2011.05.002 [ arxiv:1101.5596 [ astro-ph.he ] ] .
r. ulrich , r. engel and m. unger , hadronic multiparticle production at ultrahigh energies and extensive air showers , phys . rev .
d * 83 * , 054026 ( 2011 ) doi:10.1103/physrevd.83.054026 [ arxiv:1010.4310 [ hep - ph ] ] .
t. abu - zayyad _ et al . _
[ hires and mia collaborations ] , evidence for changing of cosmic ray composition between @xmath252 ev and @xmath253ev from multicomponent measurements , phys .
lett . *
84 * , 4276 ( 2000 ) doi:10.1103/physrevlett.84.4276 [ astro - ph/9911144 ] .
a. aab _ et al .
_ [ pierre auger collaboration ] , muons in air showers at the pierre auger observatory : mean number in highly inclined events , phys .
d * 91 * , no . 3 , 032003 ( 2015 ) erratum : [ phys .
d * 91 * , no . 5 , 059901 ( 2015 ) ] doi:10.1103/physrevd.91.059901 , 10.1103/physrevd.91.032003 [ arxiv:1408.1421 [ astro-ph.he ] ] .
y. a. fomin , n. n. kalmykov , i. s. karpikov , g. v. kulikov , m. y. kuznetsov , g. i. rubtsov , v. p. sulakov and s. v. troitsky , no muon excess in extensive air showers at 100 - 500 pev primary energy : eas - msu results , arxiv:1609.05764 [ astro-ph.he ] .
m. n. asprouli , a. d. panagiotou and e. gladysz - dziadus , interaction length and spontaneous decay of a cosmic ray `` centauro '' fireball , astropart .
* 2 * , 167 ( 1994 ) .
doi:10.1016/0927 - 6505(94)90039 - 6 a. l. s. angelis , e. gladysz - dziadus , y. v. kharlov , v. l. korotkikh , g. mavromanolakis , a. d. panagiotou and s. a. sadovsky , model of centauro and strangelet production in heavy ion collisions , phys . atom .
nucl . * 67 * , 396 ( 2004 ) [ yad .
fiz . * 67 * , 414 ( 2004 ) ] doi:10.1134/1.1648929 [ nucl - th/0301003 ] .
j. c. collins and m. j. perry , superdense matter : neutrons or asymptotically free quarks ?
lett . *
34 * , 1353 ( 1975 ) .
doi:10.1103/physrevlett.34.1353 f. iachello , w. d. langer and a. lande , a quark - like model of high - density matter , nucl . phys .
a * 219 * , 612 ( 1974 ) .
doi:10.1016/0375 - 9474(74)90122 - 5 n. cabibbo and g. parisi , exponential hadronic spectrum and quark liberation , phys .
* 59b * , 67 ( 1975 ) .
doi:10.1016/0370 - 2693(75)90158 - 6 d. j. gross and f. wilczek , ultraviolet behavior of nonabelian gauge theories , phys . rev .
* 30 * , 1343 ( 1973 ) .
doi:10.1103/physrevlett.30.1343 j. d. bjorken , highly relativistic nucleus - nucleus collisions : the central rapidity region , phys .
d * 27 * , 140 ( 1983 ) .
doi:10.1103/physrevd.27.140 a. bazavov
_ et al . _ ,
equation of state and qcd transition at finite temperature , phys .
d * 80 * , 014504 ( 2009 ) doi:10.1103/physrevd.80.014504 [ arxiv:0903.4379 [ hep - lat ] ] .
g. chapline and m. nauenberg , on the possible existence of quark stars , annals n. y. acad .
sci . * 302 * , 191 ( 1977 ) .
doi:10.1111/j.1749 - 6632.1977.tb37048.x c. alcock , e. farhi and a. olinto , strange stars , astrophys .
j. * 310 * , 261 ( 1986 ) .
doi:10.1086/164679 a. v. olinto , on the conversion of neutron stars into strange stars , phys .
b * 192 * , 71 ( 1987 ) . doi:10.1016/0370 - 2693(87)91144 - 0 v. khachatryan _ et al .
_ [ cms collaboration ] , transverse - momentum and pseudorapidity distributions of charged hadrons in @xmath10 collisions at @xmath254 tev , phys .
lett . *
105 * , 022002 ( 2010 ) doi:10.1103/physrevlett.105.022002 [ arxiv:1005.3299 [ hep - ex ] ] .
d. denterria , r. engel , t. pierog , s. ostapchenko and k. werner , the strong interaction at the collider and cosmic - rays frontiers , few body syst .
* 53 * , 173 ( 2012 ) doi:10.1007/s00601 - 011 - 0255 - 4 [ arxiv:1106.2453 [ hep - ph ] ] .
b. muller , physics and signatures of the quark - gluon plasma , rept .
phys . * 58 * , 611 ( 1995 ) doi:10.1088/0034 - 4885/58/6/002 [ nucl - th/9410005 ] .
m. n. asprouli and a. d. panagiotou , strangeness as probe of quark - gluon plasma : unconventional point of view , phys .
c * 51 * , 1444 ( 1995 ) .
doi:10.1103/physrevc.51.1444 e. gladysz - dziadus , are centauros exotic signals of the qgp ? , phys . part .
* 34 * , 285 ( 2003 ) [ fiz .
chast . atom .
yadra * 34 * , 565 ( 2003 ) ] [ hep - ph/0111163 ] .
o. kaczmarek and f. zantow , static quark anti - quark interactions in zero and finite temperature qcd i : heavy quark free energies , running coupling and quarkonium binding , phys .
d * 71 * , 114510 ( 2005 ) doi:10.1103/physrevd.71.114510 [ hep - lat/0503017 ] .
j. pumplin , d. r. stump , j. huston , h. l. lai , p. m. nadolsky and w. k. tung , new generation of parton distributions with uncertainties from global qcd analysis , jhep * 0207 * , 012 ( 2002 ) doi:10.1088/1126 - 6708/2002/07/012 [ hep - ph/0201195 ] .
t. matsui , b. svetitsky and l. d. mclerran , strangeness production in ultrarelativistic heavy ion collisions 1 : chemical kinetics in the quark - gluon plasma , phys .
d * 34 * , 783 ( 1986 ) erratum : [ phys .
d * 37 * , 844 ( 1988 ) ] .
doi:10.1103/physrevd.37.844 , 10.1103/physrevd.34.783 t. matsui , b. svetitsky and l. d. mclerran , strangeness production in ultrarelativistic heavy ion collisions 2 : evolution of flavor composition in scaling hydrodynamics , phys .
d * 34 * , 2047 ( 1986 ) .
doi:10.1103/physrevd.34.2047 a. aab _ et al .
_ [ pierre auger collaboration ] , evidence for a mixed mass composition at the `` ankle '' in the cosmic - ray spectrum , phys . lett .
b * 762 * , 288 ( 2016 ) doi:10.1016/j.physletb.2016.09.039 [ arxiv:1609.08567 [ astro-ph.he ] ] .
f. halzen , stable quark matter , mad / ph/216 , in proceedings of the workshop on cosmic ray and high - energy gamma ray experiments for the space station era , ( eds .
w. v. jones and j. p. wefel , louisiana state university , baton rouge , louisiana , 17 - 20 october , 1984 ) .
c. greiner , p. koch and h. stoecker , separation of strangeness from antistrangeness in the phase transition from quark to hadron matter : possible formation of strange quark matter in heavy ion collisions , phys .
lett . * 58 * , 1825 ( 1987 ) .
doi:10.1103/physrevlett.58.1825 l. a. anchordoqui , a. delgado , c. a. garcia canal and s. j. sciutto , hunting long - lived gluinos at the pierre auger observatory , phys . rev .
d * 77 * , 023009 ( 2008 ) doi:10.1103/physrevd.77.023009 [ arxiv:0710.0525 [ hep - ph ] ] .
a. aab _ et al .
_ [ pierre auger collaboration ] , depth of maximum of air - shower profiles at the pierre auger observatory .
i. measurements at energies above @xmath255 ev , phys .
d * 90 * , no .
12 , 122005 ( 2014 ) doi:10.1103/physrevd.90.122005 [ arxiv:1409.4809 [ astro-ph.he ] ] .
m. ahlers , l. a. anchordoqui and s. sarkar , neutrino diagnostics of ultrahigh energy cosmic ray protons , phys .
d * 79 * , 083009 ( 2009 ) doi:10.1103/physrevd.79.083009 [ arxiv:0902.3993 [ astro-ph.he ] ] .
l. a. anchordoqui , h. goldberg , d. hooper , s. sarkar and a. m. taylor , predictions for the cosmogenic neutrino flux in light of new data from the pierre auger observatory , phys .
d * 76 * , 123008 ( 2007 ) doi:10.1103/physrevd.76.123008 [ arxiv:0709.0734 [ astro - ph ] ] .
m. ahlers , l. a. anchordoqui , m. c. gonzalez - garcia , f. halzen and s. sarkar , gzk neutrinos after the fermi - lat diffuse photon flux measurement , astropart .
phys . *
34 * , 106 ( 2010 ) doi:10.1016/j.astropartphys.2010.06.003 [ arxiv:1005.2620 [ astro-ph.he ] ] .
k. kotera , d. allard and a. v. olinto , cosmogenic neutrinos : parameter space and detectabilty from pev to zev , jcap * 1010 * , 013 ( 2010 ) doi:10.1088/1475 - 7516/2010/10/013 [ arxiv:1009.1382 [ astro-ph.he ] ] .
k. h. kampert and m. unger , measurements of the cosmic ray composition with air shower experiments , astropart .
phys . * 35 * , 660 ( 2012 ) doi:10.1016/j.astropartphys.2012.02.004 [ arxiv:1201.0018 [ astro-ph.he ] ] .
e. waxman and j. n. bahcall , high - energy neutrinos from astrophysical sources : an upper bound , phys .
d * 59 * , 023002 ( 1999 ) doi:10.1103/physrevd.59.023002 [ hep - ph/9807282 ] .
m. ave , n. busca , a. v. olinto , a. a. watson and t. yamamoto , cosmogenic neutrinos from ultrahigh energy nuclei , astropart .
* 23 * , 19 ( 2005 ) doi:10.1016/j.astropartphys.2004.11.001 [ astro - ph/0409316 ] .
a. aab _ et al .
_ [ pierre auger collaboration ] , improved limit to the diffuse flux of ultrahigh energy neutrinos from the pierre auger observatory , phys .
d * 91 * , no .
9 , 092008 ( 2015 ) doi:10.1103/physrevd.91.092008 [ arxiv:1504.05397 [ astro-ph.he ] ] .
m. unger , g. r. farrar and l. a. anchordoqui , origin of the ankle in the ultrahigh energy cosmic ray spectrum , and of the extragalactic protons below it , phys .
d * 92 * , no .
12 , 123001 ( 2015 ) doi:10.1103/physrevd.92.123001 [ arxiv:1505.02153 [ astro-ph.he ] ] .
r. aloisio , d. boncioli , a. di matteo , a. f. grillo , s. petrera and f. salamida , cosmogenic neutrinos and ultra - high energy cosmic ray models , jcap * 1510 * , no .
10 , 006 ( 2015 ) doi:10.1088/1475 - 7516/2015/10/006 [ arxiv:1505.04020 [ astro-ph.he ] ] .
j. heinze , d. boncioli , m. bustamante and w. winter , cosmogenic neutrinos challenge the cosmic ray proton dip model , astrophys .
j. * 825 * , no . 2 , 122 ( 2016 ) doi:10.3847/0004 - 637x/825/2/122 , 10.3204/pubdb-2016 - 01612 [ arxiv:1512.05988 [ astro-ph.he ] ] .
a. d. supanitsky , implications of gamma - ray observations on proton models of ultrahigh energy cosmic rays , phys .
d * 94 * , no .
6 , 063002 ( 2016 ) doi:10.1103/physrevd.94.063002 [ arxiv:1607.00290 [ astro-ph.he ] ] .
p. allison _ et al .
_ , design and initial performance of the askaryan radio array prototype eev neutrino detector at the south pole , astropart .
phys . * 35 * , 457 ( 2012 ) doi:10.1016/j.astropartphys.2011.11.010 [ arxiv:1105.2854 [ astro-ph.im ] ] .
a. l. connolly and a. g. vieregg , radio detection of high energy neutrinos , arxiv:1607.08232 [ astro-ph.he ] .
d. hooper , a. m. taylor and s. sarkar , cosmogenic photons as a test of ultra - high energy cosmic ray composition , astropart . phys . * 34 * , 340 ( 2011 ) doi:10.1016/j.astropartphys.2010.09.002 [ arxiv:1007.1306 [ astro-ph.he ] ] .
l. a. anchordoqui , j. f. beacom , h. goldberg , s. palomares - ruiz and t. j. weiler , tev gamma - rays from photo - disintegration / de - excitation of cosmic - ray nuclei , phys .
* 98 * , 121101 ( 2007 ) doi:10.1103/physrevlett.98.121101 [ astro - ph/0611580 ] .
r. rapp and j. wambach , chiral symmetry restoration and dileptons in relativistic heavy ion collisions , adv .
phys . * 25 * , 1 ( 2000 ) doi:10.1007/0 - 306 - 47101 - 9@xmath2561 [ hep - ph/9909229 ] .
r. aloisio , v. berezinsky and a. gazizov , ultrahigh energy cosmic rays : the disappointing model , astropart .
phys . * 34 * , 620 ( 2011 ) doi:10.1016/j.astropartphys.2010.12.008 [ arxiv:0907.5194 [ astro-ph.he ] ] .
a. k. chaudhuri , a short course on relativistic heavy ion collisions , doi:10.1088/978 - 0 - 750 - 31060 - 4 arxiv:1207.7028 [ nucl - th ] .
c. bonati , m. delia , m. mariti , m. mesiti , f. negro and f. sanfilippo , across the deconfinement , arxiv:1610.03338 [ hep - lat ] .
j. cleymans , h. oeschler , k. redlich and s. wheaton , comparison of chemical freeze - out criteria in heavy - ion collisions , phys .
c * 73 * , 034905 ( 2006 ) doi:10.1103/physrevc.73.034905 [ hep - ph/0511094 ] .
f. becattini , m. bleicher , t. kollegger , t. schuster , j. steinheimer and r. stock , hadron formation in relativistic nuclear collisions and the qcd phase diagram , phys .
* 111 * , 082302 ( 2013 ) doi:10.1103/physrevlett.111.082302 [ arxiv:1212.2431 [ nucl - th ] ] .
f. becattini , j. steinheimer , r. stock and m. bleicher , hadronization conditions in relativistic nuclear collisions and the qcd pseudo - critical line , phys .
b * 764 * , 241 ( 2017 ) doi:10.1016/j.physletb.2016.11.033 [ arxiv:1605.09694 [ nucl - th ] ] .
a. tawfik , qcd phase diagram : a comparison of lattice and hadron resonance gas model calculations , phys .
d * 71 * , 054502 ( 2005 ) doi:10.1103/physrevd.71.054502 [ hep - ph/0412336 ] .
a. andronic _ et al .
_ , hadron production in ultra - relativistic nuclear collisions : quarkyonic matter and a triple point in the phase diagram of qcd , nucl .
a * 837 * , 65 ( 2010 ) doi:10.1016/j.nuclphysa.2010.02.005 [ arxiv:0911.4806 [ hep - ph ] ] .
l. mclerran , from the glasma to the qcd phase boundary , arxiv:1612.00864 [ hep - ph ] .
g. torrieri and j. rafelski , statistical hadronization probed by resonances , phys .
c * 68 * , 034912 ( 2003 ) doi:10.1103/physrevc.68.034912 [ nucl - th/0212091 ] .
j. cleymans , b. kampfer , m. kaneta , s. wheaton and n. xu , centrality dependence of thermal parameters deduced from hadron multiplicities in au + au collisions at @xmath257-gev , phys . rev .
c * 71 * , 054901 ( 2005 ) doi:10.1103/physrevc.71.054901 [ hep - ph/0409071 ] .
f. becattini , m. gazdzicki , a. keranen , j. manninen and r. stock , chemical equilibrium in nucleus nucleus collisions at relativistic energies , phys .
c * 69 * , 024905 ( 2004 ) doi:10.1103/physrevc.69.024905 [ hep - ph/0310049 ] .
f. becattini , j. manninen and m. gazdzicki , energy and system size dependence of chemical freeze - out in relativistic nuclear collisions , phys .
c * 73 * , 044905 ( 2006 ) doi:10.1103/physrevc.73.044905 [ hep - ph/0511092 ] .
j. cleymans , h. oeschler and k. redlich , influence of impact parameter on thermal description of relativistic heavy ion collisions at ( 1 - 2 ) @xmath38-gev , phys .
c * 59 * , 1663 ( 1999 ) doi:10.1103/physrevc.59.1663 [ nucl - th/9809027 ] .
f. becattini , j. cleymans , a. keranen , e. suhonen and k. redlich , features of particle multiplicities and strangeness production in central heavy ion collisions between 1.7@xmath38-gev / c and 158@xmath38-gev/@xmath258 , phys .
c * 64 * , 024901 ( 2001 ) doi:10.1103/physrevc.64.024901 [ hep - ph/0002267 ] .
r. averbeck , r. holzmann , v. metag and r. s. simon , neutral pions and eta mesons as probes of the hadronic fireball in nucleus - nucleus collisions around 1-@xmath38-gev , phys .
c * 67 * , 024903 ( 2003 ) doi:10.1103/physrevc.67.024903 [ nucl - ex/0012007 ] .
l. bravina , a. faessler , c. fuchs , z. d. lu and e. e. zabrodin , violation of energy per hadron scaling in a resonance matter , phys .
c * 66 * , 014906 ( 2002 ) doi:10.1103/physrevc.66.014906 [ nucl - th/0206014 ] .
j. adams _ et al . _ [ star collaboration ] , experimental and theoretical challenges in the search for the quark gluon plasma : the star collaboration s critical assessment of the evidence from rhic collisions , nucl .
phys . a * 757 * , 102 ( 2005 ) doi:10.1016/j.nuclphysa.2005.03.085 [ nucl - ex/0501009 ] . a. baran , w. broniowski and w. florkowski , description of the particle ratios and transverse momentum spectra for various centralities at rhic in a single freezeout model , acta phys .
b * 35 * , 779 ( 2004 ) [ nucl - th/0305075 ] .
j. knapp , d. heck , s. j. sciutto , m. t. dova and m. risse , extensive air shower simulations at the highest energies , astropart .
* 19 * , 77 ( 2003 ) doi:10.1016/s0927 - 6505(02)00187 - 1 [ astro - ph/0206414 ] . | we argue that ultrahigh energy cosmic ray collisions in the earth s atmosphere can probe the strange quark density of the nucleon .
these collisions have center - of - mass energies @xmath0 , where @xmath1 is the nuclear baryon number .
we hypothesize the formation of a deconfined thermal fireball which undergoes a sudden hadronization . at production
the fireball has a very high matter density and consists of gluons and two flavors of light quarks ( @xmath2 ) .
because the fireball is formed in the baryon - rich projectile fragmentation region , the high baryochemical potential damps the production of @xmath3 and @xmath4 pairs , resulting in gluon fragmentation mainly into @xmath5 .
the strange quarks then become much more abundant and upon hadronization the relative density of strange hadrons is significantly enhanced over that resulting from a hadron gas . assuming the momentum distribution functions can be approximated by fermi - dirac and bose - einstein statistics
, we estimate a kaon - to - pion ratio of about @xmath6 and expect a similar ( total ) baryon - to - pion ratio .
we show that , if this were the case , the excess of strange hadrons would suppress the fraction of energy which is transferred to decaying @xmath7 s by about 20% , yielding a @xmath8 enhancement of the muon content in atmospheric cascades , in agreement with recent data reported by the pierre auger collaboration .
= 1 |
since the work on complex networks by strogatz , watts , barabsi and albert ( see @xcite ) many researchers from such distinct fields as statistical mechanics @xcite , molecular biology @xcite , ecology @xcite , physical chemistry @xcite , genetics e.g. @xcite or social science @xcite have studied the emerging complex structure and the behavior of networks in their respective field of research .
a special subset of scale - free non - equilibrium networks can emerge from a construction procedure in which at each time step @xmath0 one vertex is added and connected to @xmath1 existing vertices with preferential linking .
this preference is proportional to the number of already existing connections of that particular vertex . by definition
the average number of connections remains constant @xmath2 .
the distribution of the degree of connections @xmath3 is of particular interest as it provides for the possibility to distinguish different classes .
one observes in scale - free - networks a behavior @xmath4 that was first discussed by simon @xcite and in the context of citations networks by price @xcite .
extensions to a more complex linking procedure @xcite or more general linkage properties @xcite were recently discussed in detail . to study the evolution of the distribution the continuum approximation is often used . at time @xmath0 the average number of connections @xmath5 of a vertex created at time @xmath6 is in an undirected network @xcite @xmath7 bianconi and barabsi @xcite pointed to the effects of distributions of fitness of individuals to attract new connections .
this can be already regarded as one prototypical example of incomplete information by interpreting the fitness in their model as an incomplete knowledge of all newer vertices about the _
individual _ properties or existence of the _ present _ vertices .
mossa et al.@xcite showed that the power - law behavior might be truncated due to information filtering . in their case
a newly attached vertex is only aware of a certain subset of the existing vertices .
this subset is however chosen randomly for each vertex individually .
therefore the incomplete information has no global properties but is instead a local property .
here we want however to follow another route with two distinct models to deal with the more interesting case that the incomplete information is attached to the new vertices individually and still global with respect to the whole network .
one model will mimic a generic and global effect that is present in all real citation networks while the other describes the influence of individual information unawareness .
for a newly created vertex the only relevant information is a list of existing vertices to connect to and their respective degrees .
incomplete information results in the ignorance of some of those vertices .
this effect is here mediated via an awareness function @xmath8 that makes the newly connected vertex @xmath0 aware of the existing vertex @xmath6 .
@xmath9 eq . ( [ eq::approx ] ) becomes then @xmath10 while there are many choices for @xmath8 we will propose one particular structure to resemble actual effects in citation networks .
we will further set @xmath11 as this network property has no influence on the scaling exponent @xmath12 in the models .
a newly created link might not be aware of the most recent created entities in the network .
one encounters this situation actually very often : the author of a new www - page can not be aware of other recently created pages that he would like to link to .
search engines do not provide for instantaneous listing of just created pages and therefore authors can find new pages just by chance .
this incomplete information about the vertices of the network results in a selection of older vertices for linking .
we model this by the setting @xmath13 with some constant @xmath14 and @xmath15 the heaviside - step - function .
therefore a newly created vertex is only aware of the oldest @xmath16 fraction of the existing vertices .
while this seems to be at a first glance to strong of an assumption , this is realized in exponentially growing systems like the www : suppose from the current real - time @xmath17 we can not know new pages younger than some period @xmath18 . then all existing pages that are capable of attracting a link are taken from the interval @xmath19 $ ] . as the number of pages obeys however @xmath20
this translates in the @xmath21-notion to @xmath22 $ ] .
here we have set @xmath23 .
recall that the time @xmath0 of eq .
( [ eq::approx ] ) is actually the number of vertices so that we actually draw a vertex from @xmath24 $ ] . rewriting the definition of @xmath25 we get @xmath26 and see that this function scales with respect to @xmath27 .
we can then set @xmath28 and rewrite equation ( [ eq::phi ] ) to ( with @xmath29 ) @xmath30 @xmath31 is the integral in the denominator @xcite .
the boundary condition @xmath32 reflects the fact , that upon creation a new node @xmath33 has only one connection .
the solution of this differential equation is @xmath34 as the function has to be continuous we must have @xmath35 .
we can derive the value for @xmath36 and arrive at @xmath37 the equation for @xmath38 is then @xmath39 by scaling arguments one can prove that in general the relation @xmath40 holds @xcite . using this
we conclude that here @xmath41 .
this is depicted in figure [ fig::c1 ] which compares this result with computer simulations . [ cols="^,^ " , ] we can further analyze the emerging networks by investigating the shortest - paths in these networks .
figure [ fig::sp ] shows the increase of both the average @xmath42 and the maximum @xmath43 of the shortest path lengths @xcite in independently created networks with increasing @xmath16 above @xmath44 .
a smaller @xmath45 $ ] reduces the number of available vertices in the growth process and the network gets more dense : the shortest path lengths get smaller and the probability of vertices with larger number of connections bigger - as can be seen from the previous result @xmath46 . for very small @xmath47
we see however an increase in @xmath42 and @xmath43 .
this is an artifact due to our initial setting of the first @xmath48 vertices to form a chain .
while the chain guarantees that none of the first vertices is preferred to another initial vertex , the path lengths are largely influenced now .
for the smallest @xmath49 used here we get an @xmath50 . here
the chain length dominates the path lengths . .
for every point we sampled over @xmath51 independent networks .
the first @xmath48 vertices where initialized to form a chain , so giving no preferences to any of them . ]
suppose that www - pages are the vertices in this scenario and links are the edges of the network .
this model then takes into account the time that e.g. search engines need to encounter new web - pages and make the general public aware of those sites .
the anonymity of a vertex @xmath6 is healed over time by sliding into the focus of new vertices as soon as @xmath52 . here
the incomplete information refers to the knowledge of _ every _ new vertex uniformly .
there is also the opposite scenario in which some vertices are aware of the full information ( that is the number of connections all the existing vertices possess ) and others are just ignorant and connect with equal probability to any of the existing ones . here
the incompleteness of information is restricted to a subset of individuals :
suppose that a newly added vertex is with some probability @xmath53 aware of all the connectivities of the other vertices . in this case it is attached with preferential linking described above .
with probability @xmath54 it is connected without preference .
we want to deduce the effect on the connectivity distribution from the master equation for the average number of connections of degree @xmath55 at time @xmath0 @xmath56 \nonumber\\ & & + ( 1-p)\cdot \left [ n(k , t ) + \frac{1}{t } n(k-1,t ) - \frac{1}{t}n(k , t)\right ] + \delta_{k,1 } \end{aligned}\ ] ] here @xmath57 is the average degree of each vertex .
the first term describes the preferential linking with its in- and outflow while the second term provides for the additional connections or loss thereof with equal probability .
notice that there are currently @xmath0 vertices in the network , so @xmath58 is the probability of hitting any one of those .
the third term is finally responsible for the newly added guy. by changing to continuous time we get from eq .
( [ eq::p ] ) @xmath59p(k-1,t ) - \left [ p\frac{k}{\bar{k}}+1-p\right]p(k-1,t ) + \delta_{k,1}\ ] ] where @xmath60 is the density of vertices with degree @xmath55 at time @xmath0 .
we can now solve for the stationary distribution for @xmath61 .
we arrive at the recursion @xmath62 this is further written as @xmath63 using the relationship @xmath64\ ] ] for large @xmath55 @xcite we conclude that @xmath65 for large @xmath55 with a divergence for the exponent when approaching @xmath66 as in this case we have no preferential linking at all . in this case the starting master equation ( [ eq::p ] ) leads correctly to the poisson - distribution @xmath67 .
the diverging behavior of the exponent was for instance also found by krapivsky and redner @xcite in their treatment of growing networks with redirection .
figure [ fig::np ] shows the results from computer experiments for this model .
the smaller the @xmath53 the more difficult it is to see any indication of the power - law
. of degree @xmath55 in networks of size @xmath68 averaged over @xmath69 independent runs .
the data was shifted for a better overview .
the straight line through the data of @xmath70 is the derived result of eq .
( [ eq::p2 ] ) with @xmath71 while the broken line indicates an asymptotic power - law for the cumulative number of a distribution with @xmath72 . ]
in this paper we developed two distinct models to describe the effect of 1 ) global incomplete information caused by penetration rates while constructing a citation network and 2 ) local incomplete information of individual vertices that are attached with a probability of non - knowledge. we derived the scaling behavior of the degree distribution for large degrees in both cases and compared this to computer experiments .
both models approach the analytic value @xcite of @xmath73 when reaching full information .
the incomplete information in the two models does not destroy the scale - free - behavior of the systems while mossa et al.@xcite found a cross - over from scale - free - behavior to an exponential in another model which takes information into account . by comparison
one can see the influence incomplete information may have on the global structure of growing networks .
we will work out particulars on real - world - networks and the influence of incomplete information in a forthcoming study .
kh is supported through a liebig - fellowship of the fonds der chemischen industrie .
computational resources were provided under a grant of the howard hughes medical institute and by the nsf - sponsored center for theoretical biological physics ( grant numbers phy-0216576 and 0225630 ) .
+ i thank s. redner and s. mossa for bringing ref .
@xcite and @xcite respectively to my attention . stimulating discussions with and comments from c. gros , j.a .
mccammon , and t. hwa are gratefully acknowledged . | we investigate the effect of incomplete information on the growth process of scale - free networks - a situation that occurs frequently e.g. in real existing citation networks .
two models are proposed and solved analytically for the scaling behavior of the connectivity distribution .
these models show a varying scaling exponent with respect to the model parameters but no break - down of scaling thus introducing the first models of scale - free networks in an environment of incomplete information .
we compare to results from computer simulations which show a very good agreement .
* keywords : * random graphs , networks , socio - economic networks , stochastic processes , growth processes |
in any theory of quantum gravity , classical space - time is likely to arise only as an approximate concept .
this problem is especially obvious in canonical quantum gravity . in particular , there is no notion of time in quantum gravity .
hence , it may seem like the causal structure of the classical metric is absent in the quantum theory as well .
however , we can incorporate the causal structure at the fundamental level in a theory of quantum gravity . in a path - integral formulation of quantum gravity
, we express the transition amplitude from one quantum three - geometry to another as a sum of histories .
causality can be included by limiting the sum to only those histories which have a well - defined causal structure .
this has been suggested as early as @xcite by teitelboim .
there has been further work in this direction in mini - superspace models @xcite .
similar ideas have also been explored in the work on causal evolution of spin networks @xcite and in two space - time dimensions @xcite . in this paper , we explore some of these ideas more concretely in the context of loop quantum gravity .
we shall construct a formal power series expansion of teitelboim s causal propagator in canonical ( loop ) quantum gravity .
we show that each term in the power series has a fixed causal structure which suggests that each of the terms may admit an interpretation as a quantum lorentzian space - time .
we then define a more general object a `` causal spin foam '' which incorporates the key features which allow us to do so .
causal spin foams are an extension of spin foam models which have been introduced in the quantization of euclidean gravity @xcite .
these are space - time models which have a close connection with the loop approach to canonical quantum gravity .
we shall also relate causal spin foams to the work on causal evolution of spin networks which was proposed as a model for lorentzian quantum gravity .
the rest of this paper is organized in the following manner .
we begin with a brief introduction to spin networks and canonical loop quantum gravity , spelling out the main ideas that we use later .
then , in section 3 , we give the details of the construction of the propagator . following that , we give the general definition of a causal spin foam and
show how the causal evolution model relates to it .
finally , we end the paper with a discussion of some open and interesting issues .
canonical general relativity ( gr ) can be written as a theory of a real su(2 ) connection @xmath0 over a compact three - manifold @xmath1 @xcite .
the conjugate variable to the connection is the densitized triad @xmath2 which takes values in the lie algebra ( 2 ) . using these variables , the constraints of gr are three gauss constraints which impose su(2 ) invariance , three diffeomorphism constraints which impose diffeomorphism invariance on @xmath1 , and the hamiltonian constraint , which is the generator of coordinate time evolution .
a review with the details of and references for the construction of the quantum theory can be found in @xcite . in this section
, we present a brief summary of the key results we shall use in this paper . the kinematical hilbert space of su(2 ) invariant states , @xmath3 , has a convenient orthonormal basis called the spin network basis @xcite . a spin network state is constructed as follows :
consider an oriented , closed graph @xmath4 which is embedded in the spatial manifold @xmath1 .
to each edge of @xmath4 assign a representation @xmath5 of su(2 ) and to each node of the graph , assign an intertwiner @xmath6 such that @xmath7 a spin network is then defined by the triplet @xmath8 . a spin network state is then defined as @xmath9 where @xmath10 is the holonomy of the connection @xmath11 along the edge @xmath12 in the representation @xmath5 of su(2 ) . the inner product between two spin network states @xmath13 and @xmath14 is given by @xmath15 in this basis , geometric operators such as area and volume are diagonal and have discrete spectra ( this has been discussed by a number of authors .
see for example references @xcite ) .
spin network states can thus be thought of as describing discrete three - geometries .
the group of spatial diffeomorphisms acts on the states in @xmath3 in a natural fashion .
the action of a finite diffeomorphism @xmath16 on a spin network state for example is given by @xmath17 this simply means that the action of the diffeomorphism is to send the state to one which is based on the shifted graph .
the space of solutions of the diffeomorphism constraint of the theory , @xmath18 , is defined by @xmath19-knot states which are diffeomorphism equivalence classes of spin networks .
@xmath18 is a subset of the algebraic dual of @xmath20 , so every @xmath19-knot state acts on spin networks @xmath21 where @xmath22 is the diffeomorphism equivalence class to which @xmath23 belongs .
we can use this to define an inner product on diffeomorphism invariant states : @xmath24 where @xmath25 is a representative of the diffeomorphism equivalence class defined by @xmath26 .
the hamiltonian constraint is the least understood piece of the dynamics of canonical gr .
one of the reasons for this is that it has no obvious geometrical interpretation like the gauss and diffeomorphism constraints .
the smeared hamiltonian @xmath27 = \int d^3x\ , n(x , t)hat{h}(x),\ ] ] where @xmath28 is the lapse function and @xmath29 is the hamiltonian constraint , can be thought as the generator of coordinate evolution in canonical gravity .
many versions of the hamiltonian constraint operator ( hco ) have been proposed @xcite .
we describe here only the features of the operator defined in @xcite which we shall need in the next section .
however , the results we derive there can easily be adapted to any other regularization of the hco .
the lorentzian hco can be written as a sum of two operators @xmath30=\hat{c}_e[n ] + \hat{t}[n]\ ] ] where @xmath31 $ ] is the hco for the euclidean version of the theory .
the action of the euclidean hco can be described graphically as follows the operator acts on the nodes of the spin network state . for every pair of edges at a node
, it adds a new edge between them and changes the spins of the edges at the node .
the precise location of the new edge is unimportant .
the graphical action of the second term @xmath32 $ ] is less well understood , but it can heuristically be thought of adding two non - intersecting edges at every node .
these new edges can either be added between the same pair of edges or between two different pairs of edges at the node .
figure [ hamfig ] shows the graphical actions of both the terms .
the non - symmetric operator defined in @xcite only creates new edges .
we take here the symmetrized version @xcite . ]
@xmath33 = \hat{h}_e[n ] + \hat{t}^s[n]$ ] of the operator which both adds and removes edges .
we have denoted by @xmath34 $ ] and @xmath35 $ ] , the symmetric versions of @xmath31 $ ] and @xmath32 $ ] respectively .
the matrix elements of the symmetric hco are given by @xmath36\ket{s } & = & \bra{s'}\hat{c}[n]\ket{s } + \overline{\bra{s'}\hat{c}[n]\ket{s}}\nonumber \\ & = & \bra{s'}\hat{c}_e[n]\ket{s}+\overline{\bra{s'}\hat{c}_e[n]\ket{s } } + \bra{s'}\hat{t}[n]\ket{s } + \overline{\bra{s'}\hat{t}[n]\ket{s}}.\end{aligned}\ ] ] the action of @xmath31 $ ] can be written using the notation of @xcite as follows : @xmath37\ket{s } = a^\beta_i(s ) \ket{s^\beta_i;i , n}\ ] ] where the index @xmath38 runs over the nodes of the spin network @xmath23 .
and @xmath39 runs over pairs of edges at every node and over a sign ( @xmath40 ) for each of those edges .
there is an implied summation over both @xmath38 and @xmath39 .
the object on the right hand side is defined by @xmath41 where @xmath42 is the value of the lapse function at the point @xmath43 which the position of the @xmath38th node in the spin network @xmath44 .
it is important to note that this is the only non - diffeomorphism invariant part in the definition of the hamiltonian constraint .
there is a regularization choice to be made about the positions of the new vertices @xmath45 , @xmath46 and @xmath47 which are created by the action of the operator .
we choose that all of these are distinct from the position of initial point @xmath43 .
this is a arbitrary choice , but is the most general one that can be made . in an analogous fashion , we can write the action of the second term in eq .
( [ ham ] ) as @xmath48\ket{s } = b^{\beta\beta'}_i(s ) \ket{s^{\beta\beta'}_i;i , n}\ ] ] where once again the indices @xmath39 and @xmath49 run over pairs of edges at every node and over a sign ( @xmath40 ) for each of those edges .
the coefficients @xmath11 and @xmath50 are functions only of the spins of the edges and can be calculated explicitly . an interesting point to note
is that with the above regularization , the states which result from the actions of the two pieces of the hco are in based on different graphs .
combined with the inner product defined in eq.([diffip ] ) , this implies that there do not exist states @xmath23 and @xmath26 such that both the terms of the hco have non - zero matrix elements between them .
we want to study the dynamics of quantum gravity in a sum over histories approach .
the key object in this formalism is the transition amplitude from one three - geometry to another which is given by a path integral which sums over all histories ( or four - geometries in the case of gr ) which connect the two given three - geometries .
as we have already seen , in canonical gr , the spatial or three - geometry is described by ( diffeomorphism invariant ) @xmath19-knot states . in other words ,
given an initial @xmath19-knot states @xmath26 and a final one @xmath23 we would like to calculate : @xmath51 = \sum_\mathrm{quantum\ 4-geometries\ g } a_\mathrm{g},\ ] ] where the sum is over all ( suitably defined ) quantum four - geometries ( histories ) which interpolate between the initial and final states and @xmath52 is the amplitude for each history .
since the transition amplitude is composed of many histories , we can consider the history to be a more elementary concept than the amplitude .
each history can be thought of as a quantum four - geometry , hence , it is reasonable that causality should be incorporated into each history before we calculate the transition amplitude by performing the sum over histories .
then , each history should have an interpretation as a quantum lorentzian space - time with a well defined causal structure . in the rest of the section
, we will see how some of these formal notions can be made more precise .
the finite evolution operator which propagates a state from an initial spatial hypersurface @xmath53 a time @xmath54 to the final spatial hypersurface @xmath55 at time @xmath56 is got by exponentiating the smeared hamiltonian constraint of eq .
( [ ham - evol ] ) and integrating over the lapse function , @xmath57 \exp(-i\int dt \int d^3x\ , n(x , t)h(x)),\ ] ] where @xmath58 $ ] is the ( formal ) measure over the lapse . in our calculations
, we use the normalized measure on scalar functions which was constructed in @xcite .
the key properties of this measure are that it is well defined for cylindrical functions of the lapse ( i.e. functions which depend only on the value of the lapse at a finite number of points ) and that it is invariant under the action of spatial diffeomorphisms .
the matrix elements of @xmath59 between two three - geometries define the transition amplitudes between them .
we can rewrite eq .
( [ prop ] ) in the simpler form @xmath60 \exp(-i\int d^3x\ , t(x)h(x)),\ ] ] where @xmath61 is proper time elapsed between the initial and final spatial slices and is given by @xmath62 in @xcite , the requirement that causality be incorporated into the propagator is imposed by demanding that we include in the path integral only those histories in which @xmath55 lies to the complete future of @xmath53 .
this can be incorporated into the propagator directly by restricting the range of integration over @xmath61 to only positive proper times .
we call the propagator evaluated with this restriction , the _
causal propagator_. before we present the details of the operator @xmath59 , we would like to discuss the restriction on the proper time in some more detail . recall that in the canonical decomposition of classical gr , the choice @xmath63 is not allowed since it leads to a degenerate space - time metric .
thus , assuming that the lapse is a continuous function , it must either be positive everywhere or negative everywhere .
in other words , the range of @xmath61 is divided into two disjoint classes , @xmath64 and @xmath65 .
the crucial assumption is that the transition amplitude corresponds to integration over only one of these .
this is motivated by the case of the relativistic point particle where a similar restriction leads to the feynman propagator @xcite .
the particular choice of positive proper time is a matter of convention .
since we are free to calculate the propagator in any gauge , let us now fix the gauge using the proper time gauge conditions @xcite .
in particular , these require @xmath66 . with this condition
, we have @xmath67 .
the causal propagator can then be written as @xmath68 \exp(-i\int d^3x n(x)h(x)).\ ] ] notice that in this form , the causal propagator is formally similar to the projector over physical states as defined in @xcite .
however , in that case , the range of integration over @xmath69 is unrestricted .
we would like to emphasize that while the projector can be defined purely at the level of canonical gravity , the propagator is a path - integral .
we shall explore the similarities and differences between the two operators in greater detail in a future work .
for the present , we exploit the computational techniques developed in @xcite to evaluate the causal propagator .
we wish to calculate transition amplitude which is the matrix element of @xmath59 between two @xmath19-knot states , namely @xmath70 .
let us start by limiting the range of integration of @xmath71 such that @xmath72 this can be thought of putting an infra - red regulator .
the propagator is then recovered by taking the limit @xmath73 .
we denote by @xmath74 the propagator where the upper limit of integration for @xmath69 is @xmath75 .
thus , we have @xmath76 \bra{s } \exp(-i\int d^3x n(x)h(x))\ket{s'}.\ ] ] following @xcite , we note that the above expression is formally three - diffeomorphism invariant and we can introduce an extra integration of the diffeomorphism group : @xmath77 \int_{0<n(x)<t } [ dn ] \bra{u(\phi)s } e^{-i\hat{h}[n]}\ket{s'},\ ] ] where @xmath78 is the finite diffeomorphism corresponding to @xmath16 , @xmath79 is the normalization factor introduced in the integration over diff and @xmath44 belongs to the diffeomorphism equivalence class of @xmath23 .
the integration over the diffeomorphism group is needed later to take care of the non - diffeomorphism invariant parts of the action of the hco .
we now expand the exponential as a formal power series @xmath77 \int_{0<n(x)<t } [ dn]\bra{u(\phi)s } \sum_{n=0}^\infty \frac{(-i \hat{h}[n])^n}{n ! } \ket{s'}.\ ] ] the @xmath80th power of the lorentzian hco can be expanded as @xmath81 + \hat{t}^s[n])^n = ( \hat{h}_e[n])^n + : \hat{h}_e[n])^{n-1}\hat{t}^s[n ] : + \cdots + ( \hat{t}^s[n])^n\ ] ] where : ( ) : indicates all possible orderings of the operators .
we can evaluate the matrix elements for each of the @xmath82 terms the right hand side of eq .
( [ n - expand ] ) .
each action of the one of the @xmath80 operators in the term leads to either the addition or deletion of one or two edge(s ) and a multiplication by a numerical coefficient and a factor of @xmath69 corresponding to the point where the operator acted .
thus , for the first term , we get @xmath83 \int_{0<n(x)<t } [ dn ] \bra{u(\phi)s } ( \hat{h}_e[n])^n \ket{s'}\\ & = & { \cal n}\frac{(-i)^n}{n!}\int_\mathrm{diff } [ d\phi ] a_{i_1}^{\beta_1}(s ' ) \cdots a_{i_1i_2\ldots i_n}^{\beta_1\beta_2\ldots \beta_n}({s'}_{i_1i_2 \ldots i_{n-1}}^{\beta_1\beta_2\ldots \beta_{n-1}})\braket{u(\phi)s } { { s'}_{i_1i_2\ldots i_n}^{\beta_1\beta_2\ldots \beta_n } } \\ & & \mbox { } \times \int_{n(x)>0 } [ dn ] n(x_{i_1,u(\phi)s } ) n(x_{i_2,u(\phi)s } ) \cdots n(x_{i_n , u(\phi)s } ) .\end{aligned}\ ] ] similar expressions can be written down for all the other terms as well . however , they all share the following properties : ( 1 ) for a non - zero result , the state @xmath26 must lie in the same diffeomorphism class as the state that results after @xmath80 actions of the hco on the state @xmath23 ; ( 2 ) the functional dependence of the expression on @xmath69 is now cylindrical .
it is just the product of the values of the lapse at a finite number of points .
the functional integral over @xmath71 can be performed in a simple fashion using the normalized measure on scalar functions constructed in @xcite .
it can simply be replaced by a finite number of ordinary integrals over the values of @xmath71 at the points where the operator acted . since the measure is diffeomorphism invariant , the integral over diff
can be performed trivially as well .
thus , each term in the formal power series can then be calculated and is finite .
in fact , each term in the power series has the general form @xmath84 where @xmath85 is a shorthand for the @xmath19-knot state generated after @xmath86 actions of the hco ( that is , either @xmath34 $ ] or @xmath35 $ ] ) , @xmath87 is the product of the numerical coefficients for each action and @xmath88 is the remaining integral over the lapse . to evaluate @xmath6
, we need to specify the limits of integration for the integrals .
teitelboim s original proposal is to perform the integration for all positive lapses .
however , a nave integration following @xcite ignores the causal structure induced by multiple actions of the constraint operator .
we show how this should be taken into account in evaluating the propagator .
intuitively , the idea is as follows : any time there are two or more actions of the hco , there is a possibility that a particular action takes place at a vertex which was created or affected by a previous action .
thus , viewed from a space - time point of view , it lies to the causal future of the other action .
this fact must be reflected in the propagator amplitude .
to do this more precisely , let us now look at the second order ( @xmath89 ) term of the propagator in more detail .
the matrix element for this is @xmath90 \int [ dn ] \bra{u(\phi)s}\hat{h}[n ] \hat{h}[n]\ket{s ' } \nonumber\\ & = & \frac{(-i)^2}{2 ! } { \cal n}\int [ d\phi ] \int [ dn ] \left(\bra{u(\phi)s}\hat{h}_e[n]\hat{h}_e[n]\ket{s ' } + \bra{u(\phi)s}\hat{h}_e[n]\hat{t}[n]\ket{s'}\right.\nonumber\\ & & \mbox { } + \left .
\bra{u(\phi)s}\hat{t}[n]\hat{h}_e[n]\ket{s ' } + \bra{u(\phi)s}\hat{t}[n]\hat{t}[n]\ket{s'}\right).\end{aligned}\ ] ] for the discussion , let us consider the first of the four terms on the right hand side of eq .
( [ quad ] ) , but the same considerations apply to all the four terms .
let the first action of the euclidean hco take place at a node @xmath43 of the spin network @xmath26 .
let us suppose that the action is the addition of an edge .
let @xmath45 be the position of the node after the action and let @xmath46 and @xmath47 be the positions of the nodes where the edge was added .
there are two distinct possibilities for the second action of the hamiltonian .
it can either act at one of the three nodes @xmath45 , @xmath46 and @xmath47 and @xmath47 , but this can be traced back to the choice of the volume operator in that reference . choosing a different volume operator fixes this issue . ] or it can act at some other node of the spin network .
let us study these two possibilities in more detail .
if the second action takes place at a node which is one of @xmath45 , @xmath46 or @xmath47 , then in a sense , it is to the future of the first one .
otherwise , it is causally unrelated to the first action and they commute .
thus , there is a partial order between the two actions of the hamiltonian .
we would like to identify this partial order with a discrete notion of causality .
this is motivated by the fact that in standard quantum field theory , operators at spatially separated points commute , whereas ones at causally related points do not . since , we do not have a background metric to define the causal structure , we suggest that _ the commutativity of the actions of the hco should define causality_. in other words , non - commuting actions correspond to causally related points whereas commuting actions correspond to space - like separated points
. a nave functional integration over positive lapses does not capture this feature .
if we are to follow teitelboim s basic idea of incorporating causality in each history , then we must further modify the propagator to include this discrete causality .
this is our key observation .
let us now look at one particular way of achieving this .
the scheme we propose is the following : since the points @xmath45 , @xmath46 and @xmath47 lie to the future of @xmath43 ( we denote this by @xmath91 ) , the domain of integration for the proper time should reflect this .
this can be understood as follows : suppose the second action takes place at @xmath45 .
the domain of integration of @xmath92 can be split into two parts @xmath93 and @xmath94 .
the first integral vanishes because the node at @xmath45 does not exist for @xmath95 .
thus , the only contribution is from the second integral .
that is , the domain of integration can be restricted to such that proper time at the second action is greater than that at the first action if the second action takes place at one of the nodes created by the first action . in the proper - time gauge
, this translates to a condition on the domain of integration of the lapse .
intuitively , we can think of the points @xmath45 , @xmath46 and @xmath47 as having been created by the action of the constraint operator at @xmath43 and thus being in future of this action in every history in this class .
on the other hand , if the second action takes place at some other node @xmath96 which is not a product of the first action then there is no limitation on the domain of integration over the lapses we integrate over strictly positive lapses at both @xmath43 and @xmath96 .
this can be thought of as the analogue of time - ordering in the absence of a background space - time metric .
we can now calculate the functional integrals over @xmath69 for the two cases .
when the second action is causally unrelated to the first one , then we have @xmath97\int [ dn ] \bra{u(\phi)s } \hat{h}_e[n]\hat{h}_e[n]\ket{s ' } \\ \mbox { } & = & \frac{a^{\beta_1}_{i_1}(s ' ) a^{\beta_2}_{i_2}({s'}^{\beta_1}_{i_1})}{t^2}\int_0^t dn(x ) n(x_{i_1 } ) \int_0^t dn(x_{i_2 } ) n(x_{i_2})\braket{s}{{s'}_{i_1i_2}^{\beta_1\beta_2}}\\ \mbox { } & = & \left(\frac{1}{t } \int_0^t dn\,n\right)^2 a^{\beta_1}_{i_1}(s ' ) a^{\beta_2}_{i_2}({s'}^{\beta_1}_{i_1 } ) \braket{s}{{s'}_{i_1i_2}^{\beta_1\beta_2}}\\ \mbox { } & = & \frac{t^4}{4 } a^{\beta_1}_{i_1}(s ' ) a^{\beta_2}_{i_2 } ( { s'}^{\beta_1}_{i_1})\braket{s}{{s'}_{i_1i_2}^{\beta_1\beta_2}}.\end{aligned}\ ] ] if the second action is to the future of the first one , we can repeat the above calculation and we get @xmath97 \int [ dn ] \bra{u(\phi)s } \hat{h}_e[n]\hat{h}_e[n]\ket{s'}\\ \mbox { } & = & \frac{a^{\beta_1}_{i_1}(s ' ) a^{\beta_2}_{i_2}({s'}^{\beta_1})}{t^2 } \int_0^t dn(x ) n(x_{i_1 } ) \int_{n(x_{i_1})}^t dn(x_{i_2 } ) n(x_{i_2 } ) \braket{s}{{s'}_{i_1i_2}^{\beta_1\beta_2}}\\ \mbox { } & = & \left ( \frac{1}{t^2}\int_0^t dn_1\,n_1 \int_{n_1}^t dn_2\,n_2 \right)a^{\beta_1}_{i_1}(s ' ) a^{\beta_2}_{i_2}({s'}^{\beta_1}_{i_1 } ) \braket{s}{{s'}_{i_1i_2}^{\beta_1\beta_2}}\\ \mbox { } & = & \frac{t^4}{8 } a^{\beta_1}_{i_1}(s')a^{\beta_2}_{i_2 } ( { s'}^{\beta_1}_{i_1})\braket{s}{{s'}_{i_1i_2}^{\beta_1\beta_2}}.\end{aligned}\ ] ] @xmath98 and @xmath99 are just shorthand notations for @xmath100 and @xmath101 respectively .
notice the numerical coefficient of the two terms are different .
note that we have made the choice that none of the nodes @xmath45 , @xmath46 and @xmath47 which result from the action of the hamiltonian at @xmath43 actually coincide with @xmath43 . as stated earlier ,
this is a choice in the regularization of the operator . since each term in eq .
( [ quad ] ) has the form of a numerical coefficient which is a function of the spin network state acted upon times the value of the lapse function at the vertex where the action took place .
the functional integral over the lapse then contributes a factor of @xmath102 or @xmath103 depending upon the causal relation between the two actions .
this calculation can be extended to all the terms in the power series . for any
given set of causal relations between the vertices , the functional integral over the lapse can be calculated as follows : consider a set of vertices ( @xmath104 , @xmath105 , ,
@xmath106 ) with causal relations `` @xmath107 '' between them . then , @xmath108 where the lower limits of the integrals are chosen such that if @xmath109 , then @xmath110 is the domain of integration .
the causal factor @xmath111 is a fraction smaller than @xmath112 which can be expressed as a product of local causal factors at each vertex @xmath113 where @xmath114 is the number of operator actions to the future of @xmath115 . in table 1 , we present the values of @xmath111 for the first few orders in the power series .
we have used a diagrammatic notation to indicate the causal relations : two unrelated vertices are denoted by disjoint points , whereas a causal relation is indicated by drawing an arrow from the past point to the future point .
@xmath111 & @xmath80 & causal structure & @xmath111 + 2 & & 1 & 4 & & 1 + & & & & & + 3 & & 1 & & & + & & & & & + & & & & & + & & & & & + & & & & & + & & & & & + we would like to point out one very interesting feature of the causal structure generated by multiple actions of the hco .
namely , each vertex where the action takes place can have at most one vertex to immediate past .
this means that causal diagrams such as can not appear among the terms of the power series .
this is a consequence of the fact that the action of the hco is always localized at one vertex and can be regarded as a manifestation of the ultra - local character of the hco @xcite .
we shall return to this issue briefly in the next section .
this completes our calculation of the causal propagator except for the limit @xmath116 . for any finite value of @xmath75 ,
each term in the power series is finite .
since the causal propagator is essentially a time - ordered exponential , it is clear that the limit @xmath116 should be divergent .
we should point out that this is an infra - red divergence which may be regulated by adding a small imaginary piece to the hco analogous to the case of the feynman propagator .
we shall not attempt to take this limit here , but we will discuss it some more at the end of the paper .
a very powerful way of keeping track of the terms in the power series expansion in terms of branched two - surfaces was introduced in @xcite .
we use the same basic ideas , but also incorporate the causal structure which we discussed in the previous subsection . each term in the power series for @xmath117 is a sequence of @xmath19-knots @xmath118 with local causal relations between the nodes of successive spin networks in the sequence corresponding to the action of the hamiltonian constraint .
we can represent this sequence by a branched , colored two - surface @xmath119 as follows : 1 .
construct the finite two surface generated by taking the topological product of the graph for the initial spin network with an interval .
2 . represent the first action of the hamiltonian constraint by a branching of the world - line of the node where the action took place .
3 . patch this onto a two surface generated by a finite piece of the `` world - sheet '' of the graph of the next spin network .
4 . repeat the above steps for all the remaining @xmath19-knots in the sequence .
there is a natural local causal structure built into this surface because of the partial order in the various actions of the hamiltonian constraint .
the two - surface can be sliced by `` equal - time slices '' which are simply the spin networks of the sequence .
the faces in the two - surface carry su(2 ) spin labels and the lines carry intertwiners .
the vertices or branching points of the two surface correspond to the action of the hco .
thus , the dynamics can be thought of as being localized at the vertices .
these can be thought of as being analogous to vertices in feynman diagrams .
we refer to each of these two - surfaces as a _ causal spin foam_. each causal spin foam can be assigned the amplitude @xmath120 where @xmath121 is the numerical coefficient for the action of the hco which takes @xmath122 to @xmath123 and @xmath124 is the number of operator actions in the causal future of the vertex @xmath125 .
the transition amplitude ( [ trans - amp ] ) can be then written as a sum over causal spin foams as follows : @xmath126 thus , the final form of the sum over histories can be expressed as a sum over branched , colored two - surfaces .
besides the topological information , these two - surfaces have one additional property , namely , the causal relations between their vertices . based on the formal expression ( [ formal - prop ] )
, we can identify these two - surfaces as being quantum , lorentzian four - geometries . in the next section
, we give a general definition for causal spin foams , and discuss some of their properties .
we can generalize the results of the previous section in a natural way . instead of considering two - surfaces which are generated by the action of the hamiltonian constraint , we consider more general branched , colored two - surfaces and prescribe amplitudes for all possible branchings in them .
this can be thought of as an inverse prescription of the quantum hamiltonian constraint .
for example , we can modify the ultra - local nature of the hco by allowing more general causal structures in the branched surfaces .
the requirement that the theory have the right classical limit , viz .
classical general relativity , will then choose the correct set of amplitudes .
we shall now try to define the basic structure which enables us to start addressing some of these problems .
many of the ideas in this section are closely based on @xcite . from now on
, we shall consider only piecewise - linear ( pl ) two - surfaces .
the advantage of doing so is that we can formulate the entire theory in a purely combinatorial fashion without reference to a background manifold .
we begin by giving an intuitive discussion of what properties that these surfaces should satisfy .
firstly , they should relate to spin networks in a natural way .
this means that they interpolate between an initial spin network to a final one .
also , the notion of causality which arose in the previous section should be incorporated in these two - surfaces .
this causality should also give us a prescription for `` slicing '' a two - surface and getting a spin network .
we now make these notions more precise .
consider a oriented two - dimensional cellular complex @xmath127 .
it consists of a collection of point ( 0-cells ) , edges ( 1-cells ) and faces ( 2-cells ) .
let @xmath128 denote the set of vertices of the complex , @xmath129 the set of edges and @xmath130 the set of faces . since , we have an oriented complex , for every edge @xmath131 , there exist maps @xmath19 and @xmath132 that map @xmath12 to its source and target in @xmath128 .
next , we need to include causality .
this is done by imposing be a partial order @xmath107 among the elements of @xmath128 .
that is , for all @xmath133 : @xmath134 these properties state that the order on the set is transitive and reflexive .
such sets are referred to as causal sets in the physics literature @xcite .
further , since each term we consider has a finite number of vertices , the vertex set satisfies the local finiteness requirements that are usually imposed on causal sets . in particular , the partial order @xmath107 gives us relations between any two vertices which are joined by an edge .
we shall call an edge `` causal '' if we have the condition @xmath135 otherwise we shall call it `` acausal '' or``space - like '' .
we adopt the convention that the orientation of a causal edge is always future - directed , that is , @xmath136 . only faces which have at least one causal edge are non - empty ( i.e. , belong to the foam ) .
this corresponds to the intuitive notion that the `` world - sheet '' of the edges of the spin networks should be causal surfaces .
also , every space - like edge can belong to at most two faces of @xmath127 .
this is because the dynamics defined by the action of the hco is localized at vertices .
we would like to retain this feature .
further , we assign to every face @xmath137 in @xmath127 which has at least one causal edge , a representation @xmath138 of su(2 ) and to every edge which belongs to at least two faces we shall assign an intertwiner @xmath139 which defines the map @xmath140 a face is considered to be incoming if the orientation it induces on the edge agrees with the intrinsic orientation of the edge and considered outgoing otherwise .
this in particular means that the faces which meet at space - like edges have the same spin and the intertwiner associated with such an edge is identity .
the causal spin foam @xmath141 is then defined as the object @xmath142 .
we started by requiring that a causal spin foam give us an history between some initial and some final spin network .
this translates into a requirement on the boundary of @xmath141 .
the boundary of @xmath127 consists of all edges @xmath143 which lie in only one face and the vertices that form the endpoints of these .
since the foam is supposed to be the interpolation between two spatial spin networks , we need to specify further properties of the boundary @xmath144 of @xmath127 such that this is satisfied .
it is sufficient to require that the set of vertices @xmath145 which belong to the boundary of @xmath127 consists of two disjoint sets @xmath146 and @xmath147 such that for any @xmath148 there is some @xmath149 such that @xmath150 and that there are no causal relations between any two elements of each of the sets @xmath146 and @xmath147 .
@xmath146 is then the set of vertices of the graph @xmath151 on which the initial spin network is based .
the edges @xmath143 are labeled by the same representation of the group as the face they border , while the vertices in @xmath152 are labeled by the intertwiners which labeled the edges which end on them .
hence , a causal spin foam @xmath141 can be regarded as a mapping from an initial spin network to a final one .
intermediate spin networks should correspond to spatial slices of the foam .
any acausal , connected , closed subset of the 1-skeleton of @xmath127 ( i.e. the set of all 0-cells and 1-cells which belong to @xmath127 ) can be chosen as a graph on which a spin network is based .
since all space - like edges have at most two faces incident on them , we assign the representation @xmath153 associated with either of those faces to the edge .
generically , we can also assign the intertwiner of the causal edge passing through every vertex to the vertex itself .
thus , in general , this will give us a spin network .
we can not slice through vertices where more than two causal edges meet as there is no unambiguous definition of the intertwiner for these vertices .
slicing to the past and future of these vertices gives us different spin networks .
thus these are the `` feynman vertices '' of the theory where the dynamics is effectively localized .
this completes our characterization of a causal spin foam . to summarize the key points
, we give a definition : a causal spin foam @xmath154 is a map from a spin network @xmath155 to a spin network @xmath156 such that 1 .
@xmath127 is an oriented two - dimensional cellular complex ; 2 .
@xmath107 is a partial order on the vertex set @xmath128 ; 3 .
an edge of @xmath127 is called causal if its endpoints are related by @xmath107 , otherwise it called acausal ; 4 .
every face @xmath157 has at least one causal edge ; 5 .
every face of @xmath127 is labeled by a representation @xmath138 of su(2 ) ; 6 .
every edge of @xmath127 is labeled by an intertwining tensor @xmath158 such that @xmath159 7 .
the boundary of @xmath127 , @xmath160 and for every vertex @xmath161 such that @xmath162 and @xmath163 are acausal sets ; 8 . the spins @xmath164 and intertwiners @xmath165 of @xmath123 agree with those on the faces and edges of @xmath166 which intersect the boundary .
dynamics can be considered as the assignment of amplitudes @xmath167 to each foam . in particular , this amplitude can be written as @xmath168 where the amplitude assigned to each vertex is a local quantity , that is , it depends only upon the spins and intertwiners in the neighborhood of the vertex . based on the result for the previous section
, this amplitude should also contain a factor which depends upon the causal structure of the vertex .
a specification of the amplitude @xmath169 for all possible vertices in the theory can be regarded as being equivalent to giving the definition of the hco . in the next section ,
we give an example of a specific causal spin foam model which does not arise from any hamiltonian constraint .
the causal evolution model @xcite can be viewed as causal spin foam model .
we shall start by giving a brief review of our understanding of the essential features of this model .
start with a ( abstract ) three - dimensional simplicial complex @xmath170 in which every face is labeled by a spin @xmath153 ( representation of su(2 ) ) and every tetrahedron is labeled by an intertwiner which defines the map from @xmath171 to the trivial representation .
this is supposed to correspond to a spatial slice .
the dual one - skeleton of the complex @xmath170 is a four - valent spin network ( the model restricts to spin networks based on four - valent graphs ) . now
the amplitude between an initial state ( @xmath172 and a final state ( @xmath173 ) can be written as @xmath174 where we consider all the four - dimensional simplicial complexes @xmath175 which have as their boundaries @xmath176 .
each four - simplex in @xmath175 consists of 5 tetrahedra and 10 faces .
the causal evolution model further specifies that the 4-simplex has a particular causal structure , that is , @xmath86 of the tetrahedra lie in the past of the remaining @xmath177 .
thus , the amplitude @xmath178 which corresponds to every 4-simplex is a function in the space @xmath179 which reflects this causal structure .
the contact with spin networks was made in the case of the three - simplex by looking at the dual one - skeleton of @xmath170 .
we may expect that the causal spin foam may be recovered by considering the dual two - skeleton of the four - complex .
let us consider one four - simplex in the interior of a particular @xmath175 .
the dual two - skeleton can be constructed by assigning a 0-cell to the 4-simplex itself , a 1-cell to every tetrahedron in the 4-simplex and a 2-cell to every face of the 4-simplex . by assigning the labels of every element in the simplex to its dual element
, we can generate a 2-dimensional cellular complex whose faces are labeled by spins and whose edges are labeled by intertwiners .
the only things that remains to be checked is that every face in the dual two - skeleton has a time like edge in its boundary . to verify this
, we observe that the edges in the dual 1-skeleton correspond to joining the centers of neighboring 4-simplices . since , these 4-simplices have causal relations between them , the dual edges are obviously causal .
( in the language of causal evolution , there is an order in which we placed the 4-simplices to generate @xmath175 and this is identified as a causal order ) .
there is certain amount of subtlety associated with taking the duals of the three - complexes that form the boundary .
the way to define this is to first construct the dual one - skeleton of the boundary 3-complex by associating an edge to every face in it .
next , construct the duals for these faces in the four - dimensional sense , considering only the parts of these surfaces which are inside the four - complex .
these will then be bounded by the one - skeleton of the boundary .
thus , the boundary of the dual two - skeleton is in fact the union of the graphs on which the initial and final spin networks are based .
thus , the causal evolution model satisfies all the conditions in definition 1 .
in fact , the causal evolution model can be viewed as a causal spin foam model in which only two - surfaces which are dual 2-skeletons of a 4-complex and the causal structures which are local to each 4-simplex are included in the sum over histories .
it is worth noting that the causal evolution model also gives a method for constructing ( a limited class of ) causal spin foams .
in this paper , we have shown two main results .
we have constructed the causal propagator in canonical quantum gravity as a formal power series .
this was made possible by the use of spin network states and the existence of a diffeomorphism invariant measure for the lapse function .
we found that it was necessary to modify the original proposal in @xcite to include the discrete causality that was introduced by multiple actions of the hco .
in fact , we used the non - commutativity of the hco to define a causal structure .
this was our first key result .
the other key result is that the power series for the propagator can be represented as a sum over colored , branched two - surfaces which we call causal spin foams .
we gave a general definition for a causal spin foam and showed that the causal evolution model @xcite can be regarded as a causal spin foam model .
we shall conclude this paper with a discussion of some of the open issues that remain to be understood .
these are listed below in no particular order . *
the propagator which we have considered is motivated by studying the feynman propagator for the point particle . however , the role that it plays in quantum gravity is not completely understood .
in particular , its relation to the projector onto physical states needs to be clarified .
* we have presented a particular concrete proposal to incorporate the causal structure which results from the multiple actions of the hco .
the four - diffeomorphism invariance of this proposal deserves to be better understood .
in particular , we have suggested , that the non - commutativity of the hco defines the causal structure of the quantum space - time .
this should be studied in greater detail by coupling the gravity to matter fields and verifying that our definition of causality is consistent with that in usual quantum field theory .
+ also , we have concentrated on the range on integration for the lapse .
the dependence on the measure chosen for the lapse is unclear .
we should emphasize though , that the discrete causal structure will determine the range of integration for the lapse , once the measure is chosen . *
the limit @xmath116 needs further study .
there are two different avenues which can be explored in this regard .
as we suggested , it may be possible to regulate this in a fashion analogous to the case of the feynman propagator for the relativistic point particle .
another proposal is that this limit should only be taken in physically relevant quantities where it may make sense @xcite . * the details of the calculation in section 3
depend on a particular regularization of the hco .
however , given any other regularizations of the operator , we can repeat the calculation for them .
however , we would like to point out that we gave an example of a model which can be viewed as an inverse definition of the constraint operator . the causal spin foam models and the hamiltonian constraint should be viewed as two sides of the same problem a better understanding of one will help in the study of the other . *
the appearance of causal spin foams in the path integral for lorentzian quantum gravity indicates that they may be thought of as quantum lorentzian space - times .
this aspect needs to studied in greater detail .
in particular , the relation of this work to @xcite , where the authors try to construct a spin foam model for lorentzian gravity based on the representation theory of the lorentz group , requires further investigation .
it is interesting to note that in three dimensions , a path integral model based on the group so(2,1 ) can be constructed in which each term has a consistent causal structure @xcite . *
a related issue , which arises in a natural fashion , is the choice of causal spin foams that should be included in the sum over histories
. it may be that it is sufficient to limit the sum to a subset of all causal spin foams ( for example , the ones which arise in the causal evolution model ) .
it may even turn out that this question is irrelevant in the classical limit i.e. , there may be more than one class of models for which classical gr is recovered .
further work is needed before this question can be answered in a definite way .
this work adds to the understanding of the role that causality plays in the dynamics of quantum gravity .
however , many issues are still unclear and further work needs to be done before answers to physical questions can be extracted .
i would like to thank lee smolin , fotini markopoulou , laurent freidel and michael reisenberger for numerous helpful discussions and suggestions .
this work was supported in part by nsf grant phy-9514240 to the pennsylvania state university , a miller fellowship from the physics department , and a gift from the jesse phillips foundation .
99 c. teitelboim , phys .
* 50 * 705 ( 1983 ) .
r. laflamme and j. louko , phys .
d**43 * * , 2730 ( 1991 ) . f. markopoulou and l. smolin , nucl .
phys.*b508 * , 409 ( 1997 ) .
f. markopoulou , gr - qc/9704013 .
j. ambjrn and r. loll , nucl .
* b536 * 407 ( 1998 ) .
m. reisenberger and c. rovelli , phys .
rev.d**56 * * , 3490 ( 1997 ) .
m. reisenberger , gr - qc/9711052 ; j.w .
barrett and l.crane , j. math .
* 39 * , 3296 ( 1998 ) .
baez , class .
* 15 * , 1827 ( 1998 ) .
a. ashtekar , phys . rev
. lett . , * 57 * 2244 ( 1986 ) ; phys
. rev .
d**36 * * , 1587 ( 1987 ) . f. barbero , phys .
d**49 * * , 6935 ( 1994 ) ; phys . rev
. d**51 * * , 5507 ( 1995 ) . c. rovelli , living rev
. relativ .
* 1 * , article 1 ( 1998 ) . c. rovelli and l. smolin phys .
d**53 * * , 5743 ( 1995 ) ; j.c .
baez , in `` the interface of knots and physics '' , ed .
louis kauffman ( american mathematical society , providence , 1996 ) , pp .
167 - 203 .
l. smolin , in `` quantum gravity and cosmology , the proceedings of the 1991 gift international seminar on theoretical physics '' , ( world scientific , singapore , 1992 ) ; c. rovelli and l. smolin , nucl .
phys.*b442 * , 593 ( 1995 ) ; a. ashtekar and j. lewandowski , j. geom .
phys.*17 * , 191 ( 1995 ) ; r. de pietri and c. rovelli , phys .
rev . d**54 * * , 2664 ( 1996 ) . c. rovelli and l. smolin , phys .
lett . * 72 * 446 ( 1994 ) .
r. borissov , phys .
d**55 * * 6099 ( 1997 ) .
t. thiemann , class .
grav.*15 * , 839 ( 1998 ) ; class .
grav . * 15 * , 875 ( 1998 ) . c. rovelli , phys .
d**59 * * , 104015 ( 1999 ) . c. teitelboim , phys .
d**25 * * , 3159 ( 1982 ) .
l. smolin , gr - qc/9609034 .
l. bombelli , j. lee , d. meyer , and r.d .
sorkin , phys.rev .
lett . * 59 * , 521 ( 1987 ) ; r.d . sorkin , `` spacetime and causal sets '' in proceedings of silarg vii ( 1990 ) .
barrett and l. crane , gr - qc/9904025 .
l. freidel , in preparation .
r. borissov and s. gupta , phys .
d**60 * * 024002 ( 1999 ) . | we compute teitelboim s causal propagator in the context of canonical loop quantum gravity . for the lorentzian signature , we find that the resultant power series can be expressed as a sum over branched , colored two - surfaces with an intrinsic causal structure .
this leads us to define a general structure which we call a `` causal spin foam '' .
we also demonstrate that the causal evolution models for spin networks fall in the general class of causal spin foams . |
it was observed that for blackbody radiation@xcite the mean energy @xmath0 of stationary electromagnetic waves is a function of frequency @xmath1 of these waves : at temperature @xmath2 we can write @xmath3 where @xmath4 is the boltzmann constant .
this contradicts the energy equipartition law of classical statistical mechanics in that @xmath0 is frequency independent .
the important contribution of planck came when he discovered that one could obtain the experimental result of eq .
( 1 ) by treating the energy of the stationary electromagnetic waves as a discrete spectra .
planck s law opened the way for quantum mechanics .
in such a way , einstein @xcite gave an important contribution by noticing that in the canonical ensemble the fluctuation of energy of the blackbody radiation is compatible with planck s law when it is , in terms of its variance @xmath5 , expressed in terms of the mean energy @xmath0 through the relations @xmath6 and @xmath7 where @xmath8 , @xmath9 ( @xmath10 is planck s constant ) and @xmath11 is the number of quantum particles .
einstein imagined the first relation as being exactly a type of general thermodynamic relation , while the second must have a dynamical character , being , in principle , deducible from a microscopic dynamics .
recently , carati and galgani @xcite have investigated classical systems of coupled harmonic oscillators with the procedure of jeans - landau - teller dynamic of molecular collisions @xcite .
the principal results of carati - galgani are that : i ) relaxation times to equilibrium are nonuniform and depend on the internal degrees of freedom of the considered system ; ii ) situations of nonequilibrium statistical mechanics very far from equilibrium are described , in a first order approximation , by the analogous relations found in quantum equilibrium statistical mechanics ; iii ) as conjectured by einstein , eq .
( [ ein2 ] ) is a consequence of the dynamics .
it is important to remark that planck s law considers a real equilibrium state of the quantum statistical mechanics , while within the framework of carati - galgani , a similar relation to planck s law was obtained using nonequilibrium classical statistical mechanics .
the purpose of the present work is to study classical systems of coupled harmonic oscillators along the line of carati - galgani .
we studied three different aspects of this framework .
first , we analyzed the consequence for einstein s conjecture by assuming that the exchanges of energy , in molecular collisions , follows the lvy type statistic for high enough frequencies .
the lvy statistics for the energy exchanges was recently obtained by carati et al .
@xcite , considering the dynamics over a finite time in the jeans - landau - teller model . secondly ,
admitting that there are analogous relations in the equilibrium quantum statistical mechanics of the relations found here using the nonequilibrium classical statistical mechanics approach , we will explore some possible consequences in quantum radiation theory .
third , we show that the generalization of planck s law @xcite within the framework of the recently introduced superstatistic formalism @xcite is compatible with the our analysis of einstein s conjecture .
the organization of this paper is as follows .
the carati - galgani model is briefly explained in sec .
our generalization is presented in sec .
discussions about the consequences of our results are presented in sec .
iv . concluding remarks are given in sec .
a good level of interest has been focused on the problem of estimate energy exchanges between vibrational and translational degrees of freedom in the dynamic of molecular collisions .
one of the simplest and main model used to study this problem was published by carati and galgani @xcite .
it is based on the jeans - landau - teller approach @xcite .
a possible example of a physical system that reproduce this formalism follows .
let us imagine a one - dimensional system involving two particles @xmath12 and @xmath13 in a line .
@xmath12 is attracted by a linear spring to the fixed origin through a harmonic oscillator potential . @xmath13
interacts with @xmath12 through an analytical potential . in a first approximation , the variation of energy @xmath14 in a simple collision between @xmath12 and @xmath13 for the oscillator @xmath12 of frequency @xmath1
is given by the addition of a drift term and others dependent on the fluctuation of the oscillator s initial phase .
it is easy to show @xcite that after @xmath15 collisions an oscillator has energy @xmath16 @xmath17 is the oscillator s initial energy that is considered to be very small in order that the motion of @xmath12 and @xmath13 be decoupled .
@xmath18 is the phase of the oscillator in the @xmath19th collision and @xmath20 is evaluated as a function of @xmath1 , which is known to be decreasing exponentially with @xmath1 ( for an analysis of @xmath20 see @xcite ) .
let us consider a assembly of oscillators , averaging over the phases @xmath18 , in order that @xmath21 where @xmath22 is the phase average .
considering the phases to be independent and uniformly distributed , for large @xmath15 , the average of energy and the variance are , respectively , given by @xmath23 and @xmath24 here @xmath25 and @xmath26 correspond to , respectively , the variance and mean of the exchanged energy @xmath27 .
observe that the functional form of the variance is independent of @xmath15 and @xmath20 .
assuming @xmath11 independent identical oscillators and considering the total energy @xmath28 , we immediately obtain for the variance @xmath29 and mean @xmath30 of the exchanged energy @xmath31 where @xmath32 is the initial action per oscillator @xcite .
this result represents essentially the application of the central limit theorem , that arise from a large number of independent contributions of the oscillators @xcite .
equation ( [ carat ] ) is analogous to einstein s relation obtained in the bose - einstein quantum statistical mechanics theory .
despite the fact that this relation has been obtained considering the classical statistical mechanics very far from equilibrium , the approach recovers einstein s conjecture , i.e. , proving it to be a consequence of the dynamics .
recently , carati et al .
@xcite showed that there are systems which present qualitative differences between the statistics obtained using the equilibrium distribution and that obtained through the dynamics over a finite time .
considering the dynamics over a finite time of the jeans - landau - teller model , they obtained that the exchanges of energy follow the lvy type statistics for high enough frequencies .
an immediate consequence of this result , based on the lvy flight random walk @xcite , is that @xmath33 is infinite .
we notice that in the carati - galgani model it is assumed that the oscillators have low energies and the collisions are separated from each other so that the oscillator phases are completely independent . here
, we must be able to answer what happens if the @xmath33 is not finite ? in this way , we can no longer write eq .
( [ cara0 ] ) and , as consequence , eq . ( [ carat ] ) is not valid either .
it is well known that for @xmath34 the theoretical basis is the lindeberg - lvy central limit theorem @xcite .
since we are considering the case of low energies for the oscillators , using the same steps as were used to obtain eq .
( [ cara0 ] ) , it is straightforwardly verified that @xmath35 where @xmath36 is a constant that has dimension of energy and is suitably chosen by dimensional requirements . finally , for @xmath11 independent identical oscillators
, we find @xmath37 which generalizes the dynamic proposal of einstein and will be our hypothesis to get the generalizations of einstein s conjecture and planck s law .
let us now explore the possibility of there existing analogous relations in the equilibrium quantum statistical mechanics of the relations found here using the nonequilibrium classical statistical mechanics approach .
we start the process using eq .
( [ ein1 ] ) in conformity to einstein s vision . in this case , using eq .
( [ our ] ) in eq .
( [ ein1 ] ) we have @xmath38 it is straightforward to find the energy distribution @xmath39 } \}^{\frac{1}{\alpha -1 } } \ ; , \label{plg}\ ] ] in order that we can consistently rewrite eq .
( [ plg ] ) as @xmath40 we have introduced @xmath41 , @xmath42 and @xmath43 with @xmath44 . we assumed @xmath45 that leads to the expression @xmath46 with @xmath47 a pure dimensionless number .
for @xmath48 , @xmath49 coincides with planck s constant @xmath10 , and the @xmath0 associated with bose - einstein quantum statistic is reobtained . in the same way as the planck blackbody radiation has its origin in the bose - einstein statistics , here obtained using @xmath48 , we can take other values of @xmath50 , so that we can obtain new physical systems such that eq .
( [ pld ] ) may be applied . by following along standard lines , after straightforward calculations , the density of energy per unit volume is obtained as @xmath51^{d } } \label{ome}\ ] ] where we use the standard density of states of the photons ( @xmath52 ) .
[ fig1 ] presents @xmath53 versus energy @xmath54 for typical values of @xmath50 and @xmath55 .
note that if @xmath50 increases then @xmath53 increases .
although @xmath56 has a finite value for @xmath57 , it is null for @xmath58 while it is infinity for @xmath59 ( see inset ) .
the @xmath60 is modified significantly when @xmath50 varies .
the @xmath60 are illustrated in figs .
[ fig2 ] and [ fig3 ] for , respectively , @xmath61 and @xmath62 and various values of @xmath63 . for @xmath50 fix ,
we note small changes for @xmath60 when @xmath63 varies . and @xmath55 .
( inset shows a closeup of part of the graph for @xmath57),width=415 ] and @xmath61.,width=415 ] and @xmath62.,width=415 ] in the @xmath64 regime @xmath65 which generalizes the rayleigh - jeans law .
the generalization of wien s law is obtained in the @xmath66 region @xmath67 and the stefan - boltzmann law for the total emitted power per unit surface is attained from the usual one , given by @xmath68 , where the d - dependence is present only in the prefactor @xmath69 .
in fact , one of the characteristics of eq .
( [ pld ] ) is that it reproduces the results of souza and tsallis @xcite .
observe that for @xmath70 we can obtain the particle - number distribution @xmath71 adopting the formal theory of statistical mechanics , it is straightforward to find the variance of an ensemble of @xmath72 particles as being @xmath73 which generalizes the bosonic landau s relation @xcite . ref .
@xcite obtained the eqs .
( [ nud ] ) and ( [ nsi ] ) heuristically based on the nonextensive statistical mechanics @xcite and beck - cohen superstatistics @xcite .
they were based on the discussion of the differential equation @xmath74 ( with @xmath75 ) , whose particular case @xmath76 corresponds to bose - einstein statistics .
our formalism describes the souza - tsallis generalization from the einstein s dynamical fluctuation viewpoint .
finally , we see that eq .
( [ nud ] ) can be written as @xmath77 where we obtain the energy spectrum @xmath78 and its degeneracy @xmath79 @xmath80 being the gamma function .
the spectrum is made of equidistant levels , like that of the quantum harmonic oscillator , with a constant energy difference between successive levels equal to @xmath81 .
summarizing , we introduced a generalization of einstein s conjecture considering an approach along the line of the carati - galgani model of molecular collisions .
we obtained planck s distribution compatible with the beck - cohen superstatistics .
furthermore , establishing a metaequilibrium very far from equilibrium for the system studied , and admitting that there are analogous relations in the equilibrium quantum mechanics , we generalized planck s law .
for instance , the effects in the quantum radiation theory were obtained theoretically , as a first order approximation , of a classical system very far from equilibrium @xcite .
we emphasize that we are working with classical models of molecular collisions and not quantum models for blackbody radiation .
understanding the relationship between these two points of view is an open problem .
this work was supported by cnpq ( brazilian agency ) .
m. planck , _ verhandlungen der deutschen physikalischen gessellschaft _ * 2 * , 202 and 237 ( 1900 ) [ english translation : d. ter haar , s. g. brush , _ planck s original papers in quantum physics _ ( taylor and francis , london , 1972 ) ] . a. einstein , _ annalen der physik _ * 33 * , 1275 ( 1910 ) .
a. carati and l. galgani , _ phys .
e _ * 61 * , 4791 ( 2000 ) .
a. carati and l. galgani , _ physica a _ * 280 * , 105 ( 2000 ) .
l. d. landau and e. teller , in _ collected papers of l. d. landau _ , edited by d. ter haar ( pergamon , oxford , 1965 ) .
g. benettin , a. carati , and g. gallavotti , _ nonlinearity _ * 10 * , 479 ( 1997 ) .
a. carati , l. galgani , and b. pozzi , _ phys .
lett . _ * 90 * , 010601 ( 2003 ) .
a. m. c. souza and c. tsallis , generalizing the planck distribution , in _ complexity , metastability and nonextensivity , proc .
31st workshop of the international school of solid state physics ( 20 - 26 july 2004 , erice - italy ) _ , edited by c. beck , g. benedek , a. rapisarda and c. tsallis ( world scientific , singapore , 2005 ) , p. 66
[ cond - mat/0501389 ] . c. beck and e. g. d. cohen ,
_ physica a _ * 321 * , 267 ( 2003 ) . c. tsallis and a. m. c. souza , _ phys .
e _ * 67 * , 026106 ( 2003 ) .
p. billingsley , _ probability and measure , 3rd ed . _ ( john wiley @xmath82 sons , 1995 ) . h. j. hilhorst ,
_ brazilian j. phys . _ * 39 * 371 ( 2009 ) .
g. m. viswanathan , s. v. buldyrev , s. havlin , m. g. e. da luz , e. p. raposo , h. e. stanley , _ nature _ * 401 * , 911 ( 1999 ) .
b. v. gnedenko and a. n. kolmogorov , _ limit distributions for sums of independent random variables _
( addison - wesley , 1954 ) .
l. d. landau and e. m. lifshitz , _ statistical physics .
part 1 , 3rd ed . , revised and enlarged _ ( pergamon press , oxford , 1982 ) .
c. tsallis , _ j. stat . phys . _ * 52 * , 479 ( 1988 ) .
for a regularly updated bibliography of the subject see http://tsallis.cat.cbpf.br/biblio.htm . | classical systems of coupled harmonic oscillators are studied using the carati - galgani model .
we investigate the consequences for einstein s conjecture by considering that the exchanges of energy , in molecular collisions , follows the lvy type statistics .
we develop a generalization of planck s distribution admitting that there are analogous relations in the equilibrium quantum statistical mechanics of the relations found using the nonequilibrium classical statistical mechanics approach .
the generalization of planck s law based on the nonextensive statistical mechanics formalism is compatible with the our analysis .
planck s law , nonequilibrium , superstatistics |
the formation of planets is a complex process that requires sub-@xmath0 m dust grains to grow over a dozen orders of magnitude in size . the smallest particles ,
although they are highly - coupled to gas , can slowly drift in both radial and vertical direction , and collide with each other .
that physical collisions with low relative velocities lead to creation of larger agglomerates @xcite . on the other side of the scale ,
@xmath1 m bodies are large enough to be almost completely decoupled from gas dynamics and be mostly influenced by gravitational interactions @xcite the most problematic is the intermediate step of growing dust from cm grains to km sized bodies , due to existence of several processes that counteract the possible growth or pose significant time constrains .
the latter is mainly caused by a fast radial drift of the bodies that are marginally coupled to the gas , i.e. characteristic scale of their aerodynamic drag is of the order of their orbital time @xcite .
moreover , the characteristic relative velocities for the bodies of sizes ranging from @xmath2 to @xmath3 are of the order of @xmath4 which , during collisions , result in fragmentation and bouncing rather than sticking @xcite .
one of the possible scenarios is a fast enhancement of dust density via sedimentation to the midplane due to vertical component of host star gravity .
then further fragmentation and collapse of dense dust layer is caused by a self - gravity @xcite .
though the sedimentation itself leads to the rise of kelvin - helmholz instability @xcite ( khi ) which prevents forming an infinitesimally thin layer in the midplane , recent studies show that it is not the case for massive discs with higher than solar metallicities @xcite . however , a significant flaw behind presented reasoning is a complete negligence of disc s global turbulence , which in turn is the only mechanism that would allow to explain observed accretion rates assuming @xmath5-disc model @xcite .
even though , the most plausible mechanism responsible for such turbulence , i.e. magnetorotational instability @xcite , allows for formation of local , low - level of turbulence areas due to insufficient ionization of gas , there are still other instabilities that stir the fluids @xcite . despite of this unfavorable circumstances
there is a process that commences to dominate dust evolution when ratio of concentration of dust particles to gas density approaches unity .
this mechanism was first presented by @xcite ( hereafter yg05 ) and named the _ streaming instability_. it appears that combination of dust trapping in gas pressure maxima and mutual interaction leading to dust dragging the gas , therefore enhancing the maxima even further , results in significant dust pile - up @xcite . even without the presence of self - gravity , dust concentration
may be risen up to the three orders of magnitude @xcite ( hereafter jy07 ) which could possibly lead to gravitationally bound objects @xcite .
recent numerical studies of streaming instability concentrate on various physical aspects that may influence its evolution , that includes : the influence of wide range of dust species @xcite , effects of global pressure gradients @xcite , stratification of discs @xcite . however , all known to authors works limit themselves to local disc approximation .
the main goal of this paper is to investigate the streaming instability in a more realistic circumstances of radially extended discs .
the essential points of interest are : ( 1 ) resolution studies of the streaming instability in a 2d radially extended domain , to see what is the minimum grid resolution necessary to ensure growth rates of simulated instability modes consistent with the results of linear stability analysis , ( 2 ) investigation of the streaming instability a in 3d radially extended domain , in an optimal grid resolution , deduced from resolution studies in 2d .
the plan of the paper is as follows : in section [ sec : eqs ] we set up the basic assumptions and equations relevant for the present paper . in section [ sec : lsa ]
we perform linear analysis of streaming instability in protoplanetary discs . in section [ sec : setup ]
we describe specific numerical algorithm that were used in the performed simulations along with the initial conditions and the basic simulation parameters .
section [ sec : results ] shows the obtained results and their detailed comparison with linear stability analysis .
finally in section [ sec : conclusions ] we discuss the outcome of our experiments and give short outlook of the future work .
we investigate global dynamics of two interacting fluids gas and dust in the protoplanetary disc .
we assume that neutral gas obeys the isothermal equation of state , whereas dust is treated as a pressureless fluid .
@xmath6 where @xmath7 , @xmath8 are densities of gas and dust respectively , @xmath9 , @xmath10 corresponding velocities , @xmath11 gas pressure , @xmath12 is a friction time and @xmath13 is gravitational potential .
our aim is to isolate the streaming instability from the other possible processes acting in a radially extended protoplanetary discs . in order to obtain a clear diagnostics of the excited instability we neglect the vertical component of gravity arising from the point mass source in the center of the reference frame , which would lead to dust sedimentation and eventually to the onset of khi @xcite .
we shall simulate disc starting from the relatively large radius ( 2 au ) .
we assume , following @xcite that transition to stokes regime occurs for particle radius @xmath14 where @xmath15 cm is the mean free path of the gas molecules @xcite , and @xmath16 is the radial distance to the disc centre . under these assumptions
the epstein s regime applies in the dominating part our domain , even for the largest simulated grain sizes .
therefore , to compute the friction time , we use the epstein s law @xmath17 where @xmath18 is the gas sound speed , and @xmath19 is the density of the solid material . while the term @xmath20 does not play any significant role during the evolution of si , i.e. its value never exceeds 1 per cent of @xmath21 we have decided not to neglect it in our simulations for the completeness sake . [ cols="^,^ " , ]
the initial stage of streaming instability evolution is similar for all cases of @xmath22 or @xmath23 and is governed by the dominant linear modes . detailed analysis of linear phase is covered in section [ simulation_analysis ] .
the following two sections describe different non - linear evolution of streaming instability in quasi - global setup with reference to similar case shown by jy07 . as noted by jy07 ,
streaming instability is the most prominent for @xmath26 and exhibits fast linear growth and heavy dust clumping .
figure [ fig1 ] shows temporal evolution of streaming instability for different initial dust to gas ratios @xmath22 . in all cases ( ba , bb , bc ) elongated clumps
are formed .
when the saturated state is reached , due to the high vertical speed of dust overdensities , the filaments undergo occasional fragmentation and collisions with each other , though still following `` v''-shaped trajectories that emerged during linear phase . the most prominent differences between ba , bb and bc are : ( 1 ) the characteristic length of local clumps is decreasing with the increase of initial @xmath22 and ( 2 ) their tendency to lean in radial direction : in bb structures are almost purely diagonal , whereas in ba and bc are much more elongated in the vertical direction . similarly to jy07 , in all cases ( ba , bb , bc ) the density peaks of the dust component in the non - linear regime settle at levels about 2 order of magnitude higher than initial density . since the fluid approach , contrary to the particle description , is much less susceptible to effects of poisson noise we were able to follow linear phase for run ab and abh for much longer than jy07 , though in the end we also observe sudden increase in growth rate of the instability due to spontaneous cavitation ( initial phase is caught in the middle subpanel of bottom right panel of fig . [ fig2 ] ) .
the bubbles of void start to spawn at outer radii and greatly increase the dust density and velocity on their edges ( see fig .
[ fig3 ] ) .
after just few orbital periods the accelerated instability expands to whole domain and dominate the nonlinear evolution .
afterwards the quasi - stationary state of vivid turbulence is achieved . in both cases ( ab , ac )
, the non - linear evolution leads to the enhancement of the peak densities of dust component over one order of magnitude , which is in close agreement with results of jy07 ( see fig . 8 in their work ) . .
] in the dust dominated case ( ac ) our results closely follow those of jy07 up to the point of saturation , where we achieve highly turbulent flow ( lower subpanel of upper right panel of fig .
[ fig2 ] ) and about 20 fold growth in dust density . however , we note that secular evolution of the physical system leads to formation of elongated dust overdensities and further growth up to 2 orders of magnitude as seen in runs with other parameters ( see lower subpanel of bottom right panel of fig .
[ fig2 ] ) .
in the gas dominated case ( aa ) we observe initial overdensity growth over two orders of magnitude in a characteristic grid - like pattern .
however , at time 1200 , 440 , 325 for runs aa , aah , aau respectively , in the course of nonlinear evolution the dense dust filaments are abruptly smoothed out resulting in oscillatory motion of mild overdensities that are only one order of magnitude denser than initial dust distribution ( see fig . [ fig4 ] ) over 2 orders of magnitude ( 2 ) after reaching certain level of @xmath29 , overdensities are abruptly smoothed out ( 3 ) instability reaches out saturated , non - linear phase where large , smooth clumps of dust are only 10 times denser than initial condition .
varying resolution only influences the availability of shorter and faster growing modes , i.e. shortens phase ( 1 ) . ]
the course of the evolution of our single 3d ( bb3d ) closely follows the corresponding 2d run bb ( compare fig .
[ fig5 ] and fig .
[ fig1 ] ) . during the linear phase of growth elongated clumps or rather sheets of dense dust are formed , as the deviation from axis axisymmetry is almost negligible .
upper panel shows a slice through the disc midplane , whereas lower panel is a part of slice perpendicular to disc plane zoom out to the region that corresponds to physical span of 2d simulations . ] in order to estimate whether we should expect effects of self - gravity in the nonlinear evolution of streaming instability , we have used ` yt ` @xcite clump finding facility to identify all the largest disconnected contour spanning for minimum value of 50 computational cells in bb3d run .
then we calculated their total mass and velocity dispersion @xmath30 to obtain jeans radius : @xmath31 subsequently we compared the average size of each individual clump @xmath32 and found that @xmath33 for every identified overdense clump , indicating that the clumps fulfill the gravitational binding condition ( see fig .
[ fig6 ] ) .
this indicates that in certain situations , i.e. for dust dominated runs of the semi - global disc configuration , streaming instability acting in a selfgravitating medium can lead to planetesimals formation @xcite . in order to analyse the growth of streaming instability we extract small square patches from various locations across the computational domain .
the patches have sizes of @xmath34 au@xmath35 , thus are small enough so that mean values of physical quantities do not vary significantly across them . in the case of 3d
run the patches are chosen on the _ r - z _ plane at @xmath36 .
we note that in the place of the fixed parameters describing the unperturbed equilibrium in , we use patch - averaged values of corresponding dependent variables in equations . we calculate the spatially averaged densities of gas and dust component @xmath37 , @xmath38 and their mutual ratio @xmath39 .
mean angular velocity @xmath40 is taken for the center of the patch .
dimensionless measure of sub - keplerian gas rotation is calculated by means of the formula ( see yg05 eq . ( 16 ) or jy07 eq .
( 1 ) ) @xmath41 in formula we average gas density in the vertical direction , then we calculate the mean radial derivative of @xmath42 .
the expression for mean stopping time is derived from @xmath43 for the sake of consistency gas and dust velocities @xmath44 are also approximated by their mean values @xmath45 instead of calculating them with the aid of - .
however , we note that the mean values that are naturally achieved during the simulation , do not diverge from the equilibrium solution by more than @xmath46 . to determine linear growth rates of the excited instability modes we calculate fourier transforms of density and velocity distributions of the dust component .
we analyse subsequently time variation of amplitudes of individual modes .
intermediate results are shown in fig .
[ fig7 ] .
\times[-0.15,0.15]$ ] au from simulation bb .
each line represents amplitude for a given pair of @xmath47 .
@xmath48 is time range when modes are fitted with .
@xmath49 is a reference point in time used to identify most dominant modes using criterion of maximum amplitude value .
thick line is an average of amplitudes that are grater than @xmath50 at reference time @xmath51 ] we identify time @xmath52 at which the dominating modes emerge from fluctuations and start their linear growth phase .
similarly , we identify the end of the linear growth phase @xmath53 when modes have grown by few orders of magnitude , and start to saturate their growth .
we fit exponential function : @xmath54 to the measured density amplitudes in the period @xmath48 .
that procedure allows us to determine the growth rate @xmath55 of each individual unstable mode , during the linear phase of the instability growth in the numerical experiments .
we identify the most rapidly growing modes by selecting those with the greatest amplitude at fixed point at the reference time @xmath49 before the saturation .
we compare the growth rates @xmath56 obtained for each mode with the solution @xmath57 resulting from the linear analysis for the local mean flow parameters . in fig .
[ fig8 ] we show temporal evolution of dust density perturbation amplitudes for three dominating instability modes in three experiments bb3d , bb and bbh , together with lines fitted to the phase of linear growth and lines representing growth of the amplitudes predicted by the linear analysis of the streaming instability . in the mid resolution run
bb the growth rates are @xmath58 smaller than those predicted by linear analysis , what indicates that still higher resolution is required to resolve these modes . in the high resolution run bbh the results of the numerical experiment tend to converge the results of linear stability analysis .
the final saturation amplitudes seem to be slightly smaller in high resolution runs .
one should note , however , that we have chosen the modes of the highest growth rate for each run , which are not the same modes in terms of wavenumbers .
au for the simulations : bb3d ( upper panel ) , bb ( middle panel ) and bbh ( lower panel ) . ] to extend our analysis we show in fig .
[ fig9 ] a contour plot ( following yg05 and yj07 ) of the growth rate @xmath55 , resulting from solutions of the dispersion relation .
the shape of isocontours indicates that the most unstable modes of the streaming instability , predicted by linear analysis , form a ridge extending towards large values of @xmath59 and @xmath60 .
the plots are constructed for the same set of the physical parameters and the mean state parameters derived for the domain patches at @xmath61 .
we then place @xmath62 of 9 dominating modes extracted at @xmath63 for simulations performed at different resolutions .
our procedure allows us to confirm that wavenumbers of the modes , emerging in presented numerical models , align with the ridge of linear solutions at the growth rate map .
it is apparent that dominating modes are limited by available numerical resolution . at lower numerical resolutions ( run bb3d )
the dominant modes locate below the contour labeled as
@xmath64 . at the mid resolution of run bb
some of the modes locate above the contour @xmath64 , and at the high resolution most of the modes locate above this contour .
similar tendency can be observed for the pair of runs ab and abh .
we anticipate that the reason for the absence of the very short wavenumber modes is the numerical diffusivity of the presently used relaxing tvd algorithm .
our estimations show that our code requires at least 32 computational cells per wavelength of the unstable mode , to accurately represent the linear growth rate , which is much more than is required by higher - order schemes @xcite .
we varied the resolution of our simulations to check how our numerical scheme affects the obtained solutions .
as the streaming instability in general `` prefers '' shorter wavelengths for the optimal growth , increasing resolution always leads to more , smaller overdensities emerging during the course of evolution ( see fig .
[ fig10 ] ) .
however we note that our results follow the fastest growing , linear modes ( see fig . [ fig9 ] ) and that most phenomena described in previous sections i.e. cavitation ( see fig .
[ fig3 ] ) or sudden growth dumping of the tightly coupled boulders in the gas dominated regime ( see fig .
[ fig4 ] ) are independent of the size of the smallest computational cell .
yr in two small rectangular patches for the two identical simulations parameters ( upper : bb , lower : bbh ) showing how the resolution affects simulation results at the linear and saturation phase of the instability .
it is apparent that shorter wavelength modes dominate at the higher resolution , and that dust condensation structures are sharper . ]
we performed a set of 2d simulations and 3d simulation of two mutually interacting fluids , i.e. gas and dust in protoplanetary disc , while neglecting vertical component of gravity .
for the spectrum of grain sizes ranging from @xmath65 cm we observe rapid growth of unstable modes in dust component , which are consistent with modes obtained for streaming instability within the framework of local linear analysis .
our model enhances previous work of other authors by taking into account the full dynamics of protoplanetary disc , e.g. radial migration , that lead to significant variation in physical quantities , such as gas pressure gradient , and were previously treated as constant .
moreover instead of using a general dimensionless stopping time parameter to couple both fluid components , we implement specific drag law ( eqn . )
which is best suited to the physical properties of simulated fluids . in the present quasi - global approach
we relieved the restriction of the fixed radial pressure gradient , adopted for numerical modeling of the streaming instability within the shearing box approximation .
it remains unknown how much this term affects the non - linear evolution of the system , since the additional term acts as an infinite energy reservoir driving the difference in azimuthal velocities of both fluids . in the global approach , even around fixed orbit , the parameter @xmath66 ,
measuring the radial pressure gradient , is function of time . while locality of shearing box fully justifies using a constant dimensionless stopping time to calculate mutual linear drag force , that approximation is no longer valid in the global disc .
the wave numbers of the fastest growing modes , extracted from numerical models , and their positions at the map of growth rate @xmath67 vs. @xmath59 and @xmath60 coincide with the predictions of linear analysis of streaming instability .
we reproduce reasonably well the linear growth rates , getting rapid convergence near @xmath68 cells per wavelength .
it is apparent , however , that due to diffusive nature of the relaxing tvd scheme , piernik needs high resolution to capture linear growth phase accurately .
the instability saturates when dust overdensities synchronize their velocity with the gas component .
we must note that despite the steady outward migration of the gas , dust blobs never really gain positive radial velocity and never cease to migrate inwards ( this was also noted by @xcite ) .
though in certain cases , their radial velocity is significantly slower than what is expected from the standard estimation of the migration rate .
the lack of additional force that would keep the blobs bounded also leads to high dispersion rate of the overdensities . in our simulations
we have neglected the role of stratification of the disc , self - gravity of both fluids and also mri induced turbulence . as each and every of the aforementioned physical processes plays significant and integral role in the protoplanetary disc evolution , we plan to gradually expand our setup increasing its complexity in order to create the more complete global model .
we have found that nonlinear evolution of the streaming instability forms conditions for the formation of gravitationally bound dust blobs in the examined 3d semi - global discs configuration .
this indicates , in accordance with the predictions by @xcite , the possibility for planetesimals formation in accreting circumstellar discs , and therefore we plan to incorporate selfgravity in our future work .
we note that some of our runs may be underresolved , especially the gas - dominated setups
. however the nonlinear outcomes of all setups converge to common solutions .
moreover , our initial simulation of full 3d setup , which took 0.36 mcpuh , shows that we are able to perform global simulation of streaming instability in a convincing manner . however , due to the severe cfl constraint on time - step imposed by the high azimuthal velocity in 3d simulations , we consider implementing fargo - like algorithm @xcite for our future work .
this is an important step towards complete model of planetesimals formation due to combined action of various fluid instabilities postulated by @xcite .
we would like to stress that everything that is required to reproduce results presented in this paper is publicly available : the code ( http://piernik.astri.umk.pl ) , initial conditions ( part of piernik s example problem set _ streaming global _ ) , analysis and visualization routines ( aforementioned github repository and ` yt ` package ) .
runs in the basic resolution took approx .
400 cpuh and can be conveniently reproduced on small clusters or even modern workstations .
kk is willing to assist inquisitive readers with performing said simulations along with providing access to data obtained in much more computationally demanding runs .
this work was partially supported by polish ministry of science and higher education through the grant n203 511038 .
this work is a part of powiew project supported by the european regional development fund in the innovative economy programme ( poig.02.03.00 - 00 - 018/08 ) .
this research was supported in part by pl - grid infrastructure .
all of the post - processing and procedures described in sect .
[ simulation_analysis ] was carried out using the data analysis and visualization package ` yt ` by @xcite .
99 bai x .-
n . , stone j. m. , 2010 , apj , 722 : 1437 bai x .-
n . , stone j. m. , 2010 , apj , 722 : l220 balbus s. a. , hawley j. f. 1998 , rvmp , 70 : 1 balsara , d. s. , tilley , d. a. , rettig , t. , brittain , s. d. , 2009 , mnras 397 : 24 blum j. , wurm g. , 2008 , ara&a , 46 : 21 cuzzi j. n. , dobrovolskis a. r. , champney j. m. , 1993 , icarus 106 : 102 durand , e. , 1960 , in masson et al .
_ solutions numriques des equations algbriques _ , vol .
1 , pp . 279281 kerner , i. o. , 1966 , numerische mathematik 8 : 290 - 294 .
goldreich p. , ward w. r. , 1973 , apj , 183 : 1051 hanasz m. , kowalik k. , wltaski d. , pawaszek , r. k. , kornet , k. , 2010 , _
eas publications series _ 42 , 281 hayashi , c. , 1981 , prog .
suppl . , 70 , 35 hawley j. f. , gammie c. f. , balbus s. a. , 1995 , apj , 440 , 742 kokubo e. , kominami j. , ida s. , 2006 , apj , 642 , 1131 kosiski , r. , hanasz , m. , 2006 , mnras , 368 , 759 jacquet e. , balbus s. , latter h. , 2011 , mnras , 415 , 3591 johansen a. , henning t. , klahr h. , 2006 , apj , 643 , 1219 johansen a. , oishi j. s. , mac low m .-
m . , klahr h. , henning t. , youdin a. , 2007 , nature , 448 , 1022 johansen a. , youdin a. , 2007 , apj , 662 , 627 lee a. t. , chiang e. , asay - davis x. , barranco j. , 2010 , apj , 725 , 1938 lesur g. , papaloizou j. c. b. , 2010 , aa , 513 , 60 masset f. , 2000 , a&as , 141 , 165 mignone , a. , bodo , g. , massaglia , s. , matsakos , t. , tesileanu , o. , zanni , c. , ferrari , a. , 2007 , apjs , 170 , 228 nakagawa y. , sekiya m. , hayashi c. , 1986 , icarus , 67 , 375 perucho , m. , hanasz , m. , mart , j. m. , sol , h. , 2004 , a&a , 427 , 415 shakura n. i. , sunyaev r. a. , 1973 , a&a , 24 , 337 skinner , m. a. , ostriker , e. c. , 2010 , apjss 188 , 290 takeuchi t. , muto t. , okuzumi s. , ishitsu n. , ida s. , 2012 , apj , 744 , 101 tilley d. a. , balsara d. s. , brittain s. d. , rettig t. , 2009 , mnras , 403 , 211 turk , m. j. , smith , b. d. , oishi , j. s. , skory , s. skillman , s. w. , abel , t. , norman , m. l. , 2011 , apjs , 192 , 9 weidenschilling s. j. , 1977 , mnras , 180 , 57 youdin a. , goodman , j. , 2005 , apj , 620 , 459 youdin a. , johansen a. , 2007 , apj , 662 , 613 zsom a. , ormel c. w. , guttler c. , blum j. , dullemond c. p. , 2010
, aa , 513 , 57 | we investigate streaming instability using two - fluid approximation ( neutral gas and dust ) in a quasi - global , unstratified protoplanetary disc , with the help of piernik code .
we compare amplification rate of the eigen - mode in numerical simulations , with the corresponding growth resulting from the linear stability analysis of full system of euler s equation including aerodynamic drag .
following @xcite we show that ( 1 ) rapid dust clumping occurs due to the difference in azimuthal velocities of gas and dust , coupled by the drag force , ( 2 ) initial density perturbations are amplified by several orders of magnitude .
we demonstrate that the multi - fluid extension of the simple and efficient relaxing tvd scheme , implemented in piernik , leads to results , which are compatible with those obtained with other methods .
[ firstpage ] hydrodynamics instabilities planets and sattelites : formation . |
since the seminal work of fermi , pasta and ulam @xcite , the problem of heat transfer in low dimensional systems has attracted much attention ( see , e.g. , refs .
@xcite for a review ) , in particular with regards to nanoscale systems .
while the problem is fundamentally important and interesting , understanding heat transport in nanoscale systems ( including molecular junctions ) is of considerable importance for applications , e.g. , in nano - electronic as well as thermo - electric devices where , typically , a minimal thermal conductivity is required to convert heat in electric current ; molecular junctions are also promising systems for use as thermal rectifiers @xcite .
the problem has been the object of numerous theoretical studies , which may be classified in two categories : ( i ) simple , toy - model one - dimensional systems typically a chain coupled to heat baths @xcite ; ( ii ) numerical simulations of realistic systems .
the latter category includes , among others , molecular dynamics studies of carbon nanotubes @xcite , self - assembled monolayers @xcite , molecular junctions @xcite and polyethylene chains as well @xcite , which have been motivated to a large extent by experimental studies of a number of materials , including nanowires @xcite , carbon nanotubes@xcite and diamond nano - rods @xcite . from a fundamental viewpoint ,
nanoscale heat transport is closely related to the study of the validity of fourier s law at small spatial scales . in the context of a junction of length @xmath0
subject to a temperature difference @xmath1 , the heat current @xmath2 is given by @xmath3 where @xmath4 is the thermal conductivity , which in general is a material property .
in short junctions , the physical dimensions of the systems of interest is comparable to the phonon mean free path @xmath5 , and one expects a transition from a ballistic regime , where @xmath6 when @xmath7 , to a diffusive regime when @xmath0 is large and temperature is high ; in this case , @xmath8 where @xmath9 , which may eventually lead to the divergence of the conductivity with system size for systems with vanishing concentrations of impurities .
the scaling of the thermal conductivity with length is often analysed using the landauer formula @xcite commonly used in mesoscopic physics : @xmath10 where @xmath11 is the transmission coefficient of phonons , and @xmath12 and @xmath13 are the distributions of phonons in the two heat baths ( ` left ' and ` right ' ) to which the system is connecting .
these distributions depend on the densities of vibrational states and on the thermal ( bose einstein or classical ) statistics in the limit where the temperature difference @xmath1 between the two heat baths vanishes and if the two baths have identical statistics , the landauer formula becomes consistent with the fourier law : @xmath14 with a conductivity increasing linearly with length : @xmath6 .
the dependence of the thermal conductivity on the system length has been extensively studied in nanowires and carbon nanotubes @xcite .
there has been comparatively fewer studies in molecular junctions , largely because measuring heat currents across molecular junctions is extremely difficult . the usual way to measure heat current across a 1d object
consists in connecting the object to two electrodes and measuring the heating joule effect due to a difference of voltage between the electrodes .
however , this gives access to the thermal conductance and one therefore needs a model for the nanoscale junction to extract the conductivity .
also , the thermal contacts of the junction may contribute to the joule effect , thus making the determination of the conductance inaccurate .
a promising but less explored method consists in using nanoparticles ( nps ) as thermal contacts . indeed , nps of identical or different
materials can be permanently linked with a few nanometers long molecule e.g. an alcenes or complementary dna strands .
such particle may then serve as thermal contacts , either to be inserted within a nems device for thermal transport , or manipulated , heated and thermally monitored using optical methods . in the present article , we propose a numerical protocol @xcite to measure the thermal conductivity of a molecular junction permanently linking two nps , possibly immersed in a liquid .
we examine the steady state current / temperature difference for the junction , and we compare the conductivity in the linear regime for junctions in vacuum and in a liquid .
we unveil the crucial role of the liquid solvent on the thermal conductivity of the junction with respect to its length . for a junction in vacuum ,
i.e. , `` dry '' , the thermal conductivity is found to increase linearly with the junction length @xmath0 , in agreement with the landauer formula , eq.[landauer2 ] . for a junction immersed in a liquid solvent , or `` wet '' ,
in contrast , we find a different behaviour : when the junction is short , the conductivity increases approximately linearly with @xmath0 , as for the dry case , then saturates for longer junctions ; the difference results from the `` friction '' of the solvent , which damps the low - frequency modes of the junction .
we have used molecular dynamics ( md ) simulations to model a molecular junction between two nps , embedded or not in a liquid matrix , as illustrated in fig .
[ snapshot ] .
the nps each contain @xmath15 atoms cut from an fcc lattice , interacting via the combination of a simple lennard - jones interaction and a nonlinear spring that forms the so called fene interaction @xcite .
the stiffness and maximum extension of the nonlinear spring are @xmath16 and @xmath17 , with @xmath18 and @xmath19 the diameter and energy of the lennard jones potential ( see below ) . the molecular junction , whose setup is displayed in fig .
[ setup ] , consists of @xmath20 beads ( @xmath21 ) ; each bead interacts with its neighbours via a hookean spring of stiffness @xmath22 , which we varied , and equilibrium length @xmath23 , exactly at the minimum of the lennard jones potential .
the end atoms of the junction are connected each to one atom from the two nps so that the equilibrium length of the junction is @xmath24 . however , in order to keep the junction locally perpendicular to the np , second - neighbour interactions are added in the form of soft springs of stiffness @xmath25 and equilibrium length @xmath26 . apart from the elastic springs ,
the atoms of the junction interact with their neighbours through the bending energy term @xmath27 , where @xmath28 is the angle between two successive segments , and @xmath4 is the bending constant which sets the rigidity of the chain ; in what follows , we set @xmath29 which yields a stiff molecular junction with a persistence length @xmath30 much larger than the equilibrium length of the junction , even for the longest junctions considered ( @xmath31 ) .
finally , each bead of the junction interacts with the atoms in the nps through the lennard - jones potential @xmath32 , where @xmath19 and @xmath18 fix the units of energy and length , respectively , which we set to @xmath33 , as we also do for the mass @xmath34 of all atoms . for wet junctions , the particles in the liquid interact via lennard - jones potentials with the same parameters @xmath19 and @xmath18 , as do the the atoms of both the junction and the nps with the liquid . in
what follows , all quantities are expressed in lennard - jones units ( lju ) , which are based on the elementary units @xmath19 , @xmath18 and @xmath34 .
beads ) between two nps embedded in a lennard jones liquid .
the region beyond the dashed line is thermostatted at @xmath35 . ] to prepare the wet junction in an equilibrium state , the whole system is first equilibrated at temperature @xmath36 under pressure @xmath37 using a nos - hoover thermostat .
for the dry junction , we thermalize the junction and the nps at the temperature @xmath35 and remove the total angular momentum of the system to avoid the possibility of a permanent rotational motion . to speed up equilibration in the latter case , we have applied a small friction force to all atoms in the system , running over @xmath38 time steps ( @xmath39 ) .
after this initial equilibration period , for both wet and dry junctions , one of the two nps is heated up to constant temperature @xmath40 using velocity rescaling while the temperature of the other np is maintained at @xmath35 . in presence of a liquid matrix ,
the region at distance @xmath41 from the surface of the np , schematically represented in fig .
[ snapshot ] , is thermostatted at @xmath35 ; the whole system is also maintained at constant pressure @xmath42 . under these conditions ,
the energy transferred from one np to the other across the junction corresponds to the amount of energy necessary to maintain the cold np at @xmath35 .
junction with @xmath43 as a function of time : energy supplied to the hot np ( black line ) , energy removed from the cold np ( red line ) and sum of the two ( blue line ) . ]
figure [ eng_junction_dry ] displays the energy transferred across a short ( @xmath44 ) dry junction with @xmath43 as a function of time .
the spring stiffness is chosen here such that the characteristic pulsation ( angular frequency ) , @xmath45 , matches the density of states ( dos ) of the nps , as will be discussed below .
the junction is found to conduct energy ; fig .
[ eng_junction_dry ] shows that almost immediately after heating the junction , the energy supplied to the hot np is exactly compensated by the energy removed from the cold np , and the junction enters a steady state regime : the energy transferred from one np to the other increases linearly with time , i.e. , the power @xmath46 flowing through the junction is constant . for a short ( @xmath44 ) dry junction ( @xmath47 , black line ) and wet junction ( @xmath48 , red line ) .
the lines are guides to the eye . ]
the corresponding results for the power as a function of the temperature difference @xmath1 between the hot and cold np are presented in fig .
[ power_t ] .
as expected , the power increases with @xmath1 .
a linear behaviour , is observed for temperature difference smaller than @xmath33 lju , which corresponds in real units to hundreds of kelvins .
this linear behaviour allows one to define unambiguously the conductance of the junction . with these temperature differences ,
the typical power is a few tenths of nw . for larger @xmath1 s ,
the power increases more slowly , a consequence of anharmonic contributions .
indeed , the atoms of the np are connected with non linear springs which , at high temperature , may couple strongly with the springs from the junction , making the hole system strongly anharmonic . fig . [ power_t ] shows also that the energy transfer is not affected by the presence of a liquid surrounding the system , at least for short junctions ; this suggests that the heat flow via the liquid is negligible compared to that across the junction . to verify this statement
, we have run simulations without a junction , i.e. for two nps at a surface - to - surface distance equal to the equilibrium length of the @xmath44 junction considered above , @xmath49 . in practice , this is achieved by attaching the center of mass of each np to a fixed point using a stiff spring .
the two nps are immersed in the same liquid as the np - junction system and equilibrated , after which one of the np is heated up to temperature @xmath50 while the other is thermostatted to @xmath35 ( and so is the liquid at distance @xmath41 ) . in this situation , the liquid between the two nps conducts energy due to the local temperature gradient ( which is @xmath51 k / nm ! ) .
the conductance of the liquid is obtained by measuring the energy necessary to thermalize the cold np . )
junction and for a pure liquid with a comparable interparticle separation , respectively @xmath52 ( dashed lines ) and @xmath53 ( dashed - dotted lines ) . the temperature difference between the two contacts of the junction is @xmath54 . ] fig .
[ wet_junction ] provides a comparison of the energy transferred across short wet junctions for different values of the spring constant @xmath22 to the energy transferred across the liquid alone .
the range of values of @xmath22 has been chosen so as to cover all possible situations of overlap between the dos of the nps and the vibration spectrum of the junction .
in particular , for the lowest value ( @xmath55 ) , the characteristic pulsation @xmath56 is smaller than the low frequency branch of the np dos , and one therefore expects weak energy transfer in this case .
indeed , the energy transferred across the liquid is negligibly small compared to the energy transferred across the junction . the power flowing across the @xmath33 nm thick liquid layer
is approximately @xmath33 nw . so far , we have considered a perfectly wetting ( hydrophilic ) situation . for large contact angles ( i.e. , a more hydrophobic particle ) which can be achieved by tuning the attractive part of the np - liquid interaction one expects even smaller energy transfer capability across the liquid .
indeed , at the interface between the nps and the liquid , the temperature field displays a discontinuity due to the finite conductance of the interface@xcite ; this discontinuity is known to increase with hydrophobicity , as the mismatch between the two media decreases the transmission of phonons through the interface . for our two nps and
given the length @xmath0 of the junction , the effective temperature gradient in the liquid decreases as hydrophobicity increases because the temperature jump at the np - liquid gets larger ( i.e. , the temperature of the liquid is smaller ) .
hence , the effective conductance of the liquid is expected to be even lower than in the hydrophilic case .
we now examine the variation of the conductance with the stiffness of the springs connecting the atoms within the junction .
the results are presented in fig .
[ power_stiffness ] for both the wet and the dry @xmath44 junction .
the dependence on @xmath22 is found to be non monotonous in both cases : the power first increases , reaches a maximum at the same value of @xmath22 ( @xmath57 ) for both wet and dry junctions , then decreases , the dry junction thus appearing to be a better conductor than the wet junction for a large stiffness . , for wet ( @xmath48 ) and dry ( @xmath47 ) short ( @xmath44 ) junctions ; @xmath54 . ]
this particular behaviour can be explained qualitatively in terms of the dos of the nps and the vibrational spectrum of the junction , as depicted in fig .
[ dos_np_junction ] .
the dos of the nps has been obtained by fourier transforming the velocity autocorrelation function . for the junction ,
we have only considered , for the sake of clarity , the spectrum of stretching modes , whose resonance frequencies are given by @xmath58 , @xmath59 being an integer between @xmath33 and @xmath60 .
apart for these @xmath60 modes , there are @xmath61 resonance frequencies corresponding to bending which , in the harmonic approximation , are decoupled from longitudinal modes so that the corresponding spectrum does not depend on @xmath22 .
different cases may be encountered , depending on @xmath22 , as demonstrated in fig .
[ dos_np_junction ] . for small values of @xmath22 , only a few modes of the junction overlap with the low frequency branch of the np dos , so that only the low frequency modes of the np are excited and a small conductance is expected , as indeed observed in fig .
[ power_stiffness ] . when @xmath62 , the matching between the stretching modes and the np dos is optimal , and the junction is a very good conductor , consistent with the results of fig .
[ power_stiffness ] . for large values of @xmath22 ,
the normal modes of the junction have high frequencies and there is little overlap with the np dos so that the energy transfer across the junction is expected to decline . junction ( vertical lines ) superimposed on the np dos ( full line ) for different values of the longitudinal stiffness @xmath22 : @xmath63 ( red lines ) , @xmath64 ( black lines ) and @xmath65 ( dashed line in this case , only the lowest frequency mode is shown ) . ]
it turns out , however , that the conductance of stiff junctions is larger than that of soft junctions , either dry or wet ( cf .
[ power_stiffness ] ) .
this may appear counterintuitive , as there is some overlap between the soft - junction spectrum and the np dos while there is none for the stiff junction .
a possible explanation is that , for large stiffness , the high - frequency longitudinal modes of the junction excite higher harmonics of the nps , allowing energy to be transferred between them . as noted earlier , fig .
[ power_stiffness ] also reveals that the dry junction conducts heat better than the wet junction .
the difference may be understood in terms of the friction of the liquid , which damps the low frequency modes of the nps , thus reducing the efficiency of the non - linear coupling discussed above . ) and wet ( @xmath48)junctions as a function of length , measured by the number of beads of the junction @xmath20 . here , @xmath43 and @xmath54 .
we discuss , finally , how the length of the junction affects the energy transfer between the nps .
the variation of the conductivity of dry and wet junctions with length is displayed in fig .
[ power_length ] .
the stiffness @xmath22 ( @xmath66 ) has been chosen here so as to ensure a good matching between the junction and the thermal contacts with the nps .
for the dry junction , the relation between conductivity and length is monotonous : the conductivity increases almost linearly , i.e. , the conductance is approximately constant .
this behaviour is standard for unidimensional systems , and easy to understand . upon increasing the length of the junction ,
the number of vibrational modes @xmath67 increases proportionally and the overlap with the np dos increases , provided the normal mode frequencies @xmath68 and @xmath69 are not too small .
this is demonstrated in fig .
[ dos_np_junction_dry_wet ] where we superimpose the spectra of free dry junctions ( not connected to nanoparticles ) of two different lengths ( @xmath44 and @xmath70 ) with the np dos : the overlap clearly increases with length and , as a result , the conductivity increases . for wet junctions , the scenario is different , as can be seen in fig .
[ power_length ] .
the conductivity first increases with length , following closely the results for the dry junction ; for @xmath71 , however , the current starts decreasing with increasing length , i.e. the conductance drops .
this non monotonous behaviour may be understood by analysing the vibrational spectra of wet junctions , also shown in fig .
[ dos_np_junction_dry_wet ] .
these have been calculated from simulations of a free junction ( i.e. , without nps ) immersed in a liquid . for the shortest junction ( @xmath44 ) ,
the dos of wet and dry junctions are very similar , except perhaps at low frequencies ; this is evidently consistent with the fact that the conductivities of short junction is not affected by the presence of the surrounding solvent , as observed above ( fig .
[ power_length ] ) .
[ dos_np_junction_dry_wet ] however shows that the spectra of long wet and dry junctions may be different and , in particular , the low - frequency modes of the dry junction are absent in the wet junction .
we infer from this observation that the surrounding solvent damps the low - frequency modes of the wet junction . to ascertain this
, we also show in fig .
[ dos_np_junction_dry_wet ] the dos of the lennard - jones liquid , here thermostatted at the reference temperature @xmath72 .
it is found to be significant only at low frequencies @xmath73 less than 40 or so ; in contrast , for higher frequencies , the spectrum of the liquid is vanishingly small .
hence , high - frequency modes of the junction feel the presence of the solvent as a frictionless elastic medium , and are thus not affected by the presence of the liquid .
this explains why the high - frequency part of the spectrum of dry and wet junctions are very similar , irrespective of the length of the junction .
low - frequency modes , on the other hand , feel the solvent as a viscous medium , and are therefore significantly damped by the solvent .
this is can in fact be seen in fig .
[ dos_np_junction_dry_wet ] where the low - frequency modes are almost completely absent , more precisely almost completely damped in the spectrum of the long wet junction .
( top ) and @xmath70 ( bottom ) , superimposed on the dos for the nps ( blue solid lines ) and that of the liquid ( dashed blue lines).,title="fig : " ] ( top ) and @xmath70 ( bottom ) , superimposed on the dos for the nps ( blue solid lines ) and that of the liquid ( dashed blue lines).,title="fig : " ]
we have studied the problem of heat transfer accross a molecular junction between two nanoparticles that lie either in vacuum and in a liquid environment using classical molecular dynamics simulations .
while our model remains `` generic '' , it does capture the essential physics of the problem . for the junction in vacuum ,
the well - known result of a conductance that is independent of length is recovered ; however , the conductivity depends sensitively on the overlap between the density of states of the junction itself and that of the nanoparticles to which it is connected .
the presence of a liquid is found to affect this behaviour in a significant manner : the conductance is constant to a crossover length , above which it starts to decline and the standard fourier regime is recovered .
our results can be rationalized in terms of the overlap between the vibrational spectrum of the junction and that of the immersing liquid , which is qualitatively different in short and long junctions .
we acknowledge funding from the _ agence nationale de la recherche _ ( anr ) project opthermal , as well as the natural sciences and engineering research council of canada ( nserc ) and the _ fonds qubcois de la recherche sur la nature et les technologies _ ( fqrnt ) .
part of the simulations have been done at the ple scientifique de modlisation numrique ( lyon ) using the lammps package freely available at http://lammps.sandia.gov see @xcite e. fermi , j. pasta and s. ulam , _ los alamos report _
la-1940 , 1955 .
cahill , w.k .
ford , k.e .
goodson , g.d .
mahan , a. majumdar , h.j .
maris , r. merlin , s.r .
philpot , _
_ 2003 * 93 * ( 2 ) 793 .
a. dhar , _ adv .
phys . _ * 57 * , 457 - 537 , 2008 .
chang , d. okawa , a. majumdar and a. zettl , _ science _ 2006 * 314 * 1121 .
g. casati , _ nature nanotechnology _
, 2007 * 2 * 23 .
a. dhar and d. roy , _ jour .
_ 2006 * 125 * 805 .
d. segal , a. nitzan and p. hnggi , _ j. chem .
_ 2003 * 119 * ( 13 ) , 6840 .
y. zhou and d. segal , _ j. chem .
_ 2010 * 133 * 094101 . n. mingo and d.a .
broido , _ phys .
_ 2005 * 95 * 096105 .
t. luo and j.r .
lloyd , _ int .
j. heat mass .
_ 2010 * 53 * 1
. n. mingo , _ phys .
b _ 2006 * 74 * 125402 .
a. henry and g. chen , _ phys .
_ 2008 * 101 * 235502 .
t.s tighe , j.m .
worlock , m.l .
_ 1997 * 70 * , 2687 .
k. schwab , e.a .
henriksen , j.m .
worlock and m.l .
roukes , _ nature _ 2000 * 404 * 974 . s. berber , y.k .
kwon and d. tomanek , _ phys .
_ 2000 * 84 * 4613 .
p. kim , l. shi , a. majumdar , p.l .
mceuen , _ phys .
_ 2001 * 87 * 215502 . c. yu , l. shi , z. yao , d. li , a. majumdar , _ nanolett .
_ 2005 * 5 * 1842 .
padgett , o. shenderova and d.w brenner , _ nanolett .
_ 2006 * 6 * 1827 .
r. landauer , _ philosophical magazine _ 1970 * 21 * ( 172 ) 863 .
s. merabia , p. keblinski , l. joly , l. lewis and j .- l .
barrat , _ phys .
2009 * 79 * 021404 . s. merabia , s. shenoguin , l. joly , p. keblinski and j .- l .
barrat , _ pnas _ 2009 * 106 * 15113 . k. kremer and g.s .
grest , _ j. chem .
_ 1990 * 92 * 5057 . j
barrat and f. chiaruttini , _ mol .
_ 2003 * 101 * 1605 . s. plimpton , _ j. comp . phys . _ ,
1995 * 117 * , 1 - 19 . | we investigate the problem of heat conduction across a molecular junction connecting two nanoparticles , both in vacuum and in a liquid environment , using classical molecular dynamics simulations . in vacuum ,
the well - known result of a length independent conductance is recovered ; its precise value , however , is found to depend sensitively on the overlap between the vibrational spectrum of the junction and the density of states of the nanoparticles that act as thermal contacts . in a liquid environment ,
the conductance is constant up to a crossover length , above which a standard fourier regime is recovered . |
cooperative behaviors are ubiquitous in real world , ranging from biological systems to socioeconomic systems .
however , the question of how natural selection can lead to cooperation has fascinated evolutionary biologists for several decades .
fortunately , together with classic game theory , evolutionary game theory provides a systematic and convenient framework for understanding the emergence and maintenance of cooperative behaviors among selfish individuals @xcite . especially , the prisoner s dilemma game ( pdg ) as a general metaphor for studying the evolution of cooperation has attracted considerable interests @xcite . in the original pdg
, two players simultaneously decide whether to cooperate ( c ) or to defect ( d ) .
they both receive @xmath1 upon mutual cooperation and @xmath2 upon mutual defection . a defector exploiting a c player
gets @xmath3 , and the exploited cooperator receives @xmath4 , such that @xmath5 and @xmath6 . as a result
, it is best to defect regardless of the co - player s decision .
thus , in well - mixed infinite populations , defection is the evolutionarily stable strategy ( ess ) , even though all individuals would be better off if they cooperated .
thereby this creates the social dilemma , because when everybody defects , the mean population payoff is lower than that when everybody cooperates . in
a recent review nowak suggested five rules for the evolution of cooperation ( see ref .
@xcite and references therein ) .
most noteworthy , departure from the well - mixed population scenario , the rule `` network reciprocity '' conditions the emergence of cooperation among players occupying the network vertices @xcite .
that is , the benefit - to - cost ratio must exceed the average number of neighbors per individual .
actually , the successful development of network science provides a convenient framework to describe the population structure on which the evolution of cooperation is studied .
the vertices represent players , while the edges denote links between players in terms of game dynamical interactions .
furthermore , interactions in real - world network of contacts are heterogeneous , often associated with scale - free ( power - law ) dependence on the degree distribution , @xmath7 with @xmath8 . accordingly ,
the evolution of cooperation on model networks with features such as lattices @xcite , small - world @xcite , scale - free @xcite , and community structure @xcite has been scrutinized .
interestingly , santos _
_ found that scale - free networks provide a unifying framework for the emergency of cooperation @xcite . from the best of our knowledge ,
so far much previous works of games on networks are based on crystalized ( static ) networks , i.e. the social networks on which the evolution of cooperation is studied are fixed from the outset and not affected by evolutionary dynamics on top of them . however , interaction networks in real world are continuously evolving ones , rather than static graphs .
indeed , individuals have adaptations on the number , frequency , and duration of their social ties base upon some certain feedback mechanisms . instead of investigating the evolutionary games on static networks which constitute just one snapshot of the real evolving ones ,
recently , some researchers proposed that the network structure may co - evolve with the evolutionary game dynamics @xcite .
interestingly , as pointed out in refs .
@xcite , the entangled evolution of individual strategy and network structure constitutes a key mechanism for the sustainability of cooperation in social networks .
therefore , to understand the emergence of cooperative behavior in realistic situations ( networks ) , one should combine strategy evolution with topological evolution . from this perspective
, we propose a computational model in which both the adaptation of underlying network of interactions and the evolution of behavioral strategy are taken into account simultaneously . in our model
, each agent plays a @xmath0-round prisoner s dilemma game with its immediate neighbors , after that , based upon self - interest , partial individuals may punish their defective neighbors by dismissing the social tie to the one who defects the most times , meanwhile seek for a new partner at random from the neighbors of the punished agent .
we shall show that such individual s local adaptive interactions lead to the situation where cooperators become evolutionarily competitive due to the preference of assortative mixing between cooperators .
the remainder of this paper is organized as follows . in the following section ,
the model is introduced in detail .
iii presents the simulation results and discussions .
we finally draw conclusions in sec . iv .
we consider a symmetric two - player game where @xmath9 individuals engage in the prisoner s dilemma game ( pdg ) over a network .
the total number of edges @xmath10 is fixed during the evolutionary process .
each individual @xmath11 plays with its immediate neighbors defined by the underlying network .
the neighbor set of individual @xmath11 is denoted as @xmath12 , which is allowed to evolve according to the game results .
let us denote by @xmath13 the strategy of individual @xmath11 .
player @xmath11 can follow two simple strategies : cooperation [ c , @xmath14 and defection [ d , @xmath15 in each round . following previous studies @xcite
, the payoff matrix @xmath10 has a rescaled form depending on a single parameter , @xmath16 where @xmath17 . + in each round , each agent plays the same strategy with all its neighbors , and accumulates the payoff , observing the aggregate payoff and strategy of its neighbors .
the total income of the player at the site @xmath18 can be expressed as @xmath19 where the sum runs over all the neighboring sites of @xmath18 , @xmath20 . in evolutionary games
the players are allowed to adopt the strategies of their neighbors after each round .
then , the individual @xmath18 randomly selects a neighbor @xmath21 for possibly updating its strategy .
the site @xmath18 will adopt @xmath21 s strategy with probability determined by the total payoff difference between them @xcite : @xmath22},\ ] ] where the parameter @xmath23 is an inverse temperature in statistical physics , the value of which characterizes the intensity of selection .
@xmath24 leads to neutral ( random ) drift whereas @xmath25 corresponds to the imitation dynamics where the @xmath21 s strategy replaces @xmath18 s whenever @xmath26 . for finite value of @xmath23 ,
the larger @xmath23 is , the fitter strategy is more apt to replace to the less one , thus the value of @xmath23 indicates the intensity of selection . in the present model , we assume each agent plays a @xmath0-round pdg with its neighbors ( @xmath27 ) , and then @xmath28 randomly selected individuals are allowed to adapt their social ties according to the game results ( @xmath29 ) . here
the individuals are endowed with the limited cognitive capacities
each agent records the strategies of its opponents used in the @xmath0-round game
. then they are able to decide to maintain those ties from which they benefit from , and to rewire the adverse links . for the sake of simplicity , if someone is picked for updating its neighbors , only the most disadvantageous edge is rewired .
it dismisses the link to the one , who defects the most times ( if there exist more than one individuals who defect the same maximum times , the one is chosen at random ) , and redirects the link to a random neighbor of the punished ( see fig . [ fig1 ] as an illustrative example ) .
the advantage of rewiring to neighbor s neighbor is twofold : first , individuals tend to interact with others that are close by in a social manner @xcite , i.e. friend s friend is more likely to become a friend ( partner ) ; second , every agent is seeking to attach to cooperators , thus redirecting to neighbor s neighbor will be a good choice since the neighbor also tries to establish links with cooperators . hence rewiring to a neighbor of a defector
is no doubt a good choice for individuals with local information only @xcite .
herein , the parameters @xmath0 and @xmath28 can be viewed as the corresponding time scales of strategy evolution and network structure adaptation .
as our strategy evolution uses synchronous updating , while evolution of network topology adopts asynchronous updating , in our case , the strategy updating event proceeds naturally much more frequent than evolution of network structure ( as @xmath30 ) .
nevertheless , even though network structure adaption is much lower than game dynamics , cooperation is still promoted by the efficient interplay between the two dynamics .
-round prisoner s dilemma game with immediate neighbors .
a dismisses the link to b , who defects the most times , and rewires the link to c , a random neighbor of b. [ fig1],width=226 ] let us point out the differences between our model and previous works . in refs .
@xcite , the evolution of strategy adopted the `` best - take - over '' update rule where each agent imitates the strategy of the best neighbor .
besides , individuals are divided into two types based on the payoffs : satisfied and unsatisfied . if individual s payoff is the highest among its neighbors , then it is satisfied . otherwise , it is unsatisfied .
the network adaptation dynamics is restricted to the ones who are defectors and unsatisfied .
thus the unsatisfied defector breaks the link with defective neighbor with probability @xmath31 , replaces it by randomly choosing agent uniformly from the network .
more recently , ref .
@xcite proposed another minimal model that combined strategy evolution with topological evolution .
they used asynchronous update rule both for evolution of strategy and structure by the fermi function . in their model , topological evolution is manipulated in the following way : a pair of c - d or d - d are chosen at random , the one may compete with the other to rewire the link , rewiring being attempted to a random neighbor s neighbor with certain probability determined by payoff difference between them .
whereas in our model , we argue that individuals are exclusively based on their self - interest .
even if the individual is a cooperator , it could not bear the exploitation by defectors . furthermore , in our situation , individuals have enough inspection over their opponents because they are engaged in a @xmath0-round pdg .
subsequently , each agent can punish the most defective neighbor by dismissing the link , meanwhile seeks to establish links with cooperators .
especially , in our model the agents are endowed with limited memory abilities by which they can punish the most defective one .
in addition , the way of associated time scales with respect to evolution of strategy and structure is different in our model from the previous related works . as aforementioned , after playing @xmath0-round pdg with neighbors , individuals update their local interactions according to the game results .
such timely feedback mechanism is ubiquitous in natural world .
besides , the adaption of network structure is much slower than evolution of strategies in our model .
such feature reflects the fact that individuals may not respond rapidly to the surrounding since maintaining and rewiring interactions are costly to them . in previous investigations ,
the time scale often is implemented in a stochastic manner .
although the implementation of time scales in the literature is equivalent to our model in a way , our method may be more plausible .
therefore , our model is different from the previous ones in these respects and captures the characteristics in real situation . in
what follows , we investigate under which conditions cooperation may thrive by extensive numerical simulations . and also , we show the effects of the changing parameters in our model on the evolution of cooperation .
we consider @xmath9 individuals occupying the network vertices . each interaction between two agents
is represented by an undirected edge ( a total of @xmath32 ) .
the social networks are evolving in time as individuals adapt their ties .
the average connectivity @xmath33 is conserved during the topological evolution since we do not introduce or destroy links .
this point assumes a constrained resource environment , resulting in limited possibilities of network configurations . besides
, we impose that nodes linked by a single edge can not lose this connection , thus the evolving networks are connected at all times .
we calculated the amount of heterogeneity of the networks as @xmath34 ( the variance of the network degree sequcence ) , where @xmath35 gives the number of vertices with @xmath36 edges .
additionally , in order to investigate the degree - degree correlation pattern about the emerging social networks , we adopted the assortativity coefficient @xmath37 suggested by newman @xcite , @xmath38 ^ 2}{m^{-1}\sum_i\frac{1}{2}(j_i^2+k_i^2)-[m^{-1}\sum_i\frac{1}{2}(j_i+k_i)]^2},\ ] ] here @xmath39 are the degrees of the vertices at the ends of the @xmath11th edge , with @xmath40 .
networks with assortative mixing pattern , i.e. @xmath41 , are those in which nodes with large degree tend to be connected to other nodes with many connections and vice versa . the interplay between network structure and game dynamics
is implemented as following steps : * step ( 1 ) : the evolution of strategy uses synchronous updating .
each agent plays the pdg with its immediate neighbors for consecutive @xmath0 rounds . after each round ,
each individual adapts its strategy according to eq .
( [ transp ] ) , and records down the defection times of its every neighbors . * step ( 2 ) : the update of individual s local social interactions is asynchronous .
@xmath28 agents are successively chosen at random to rewire the most adverse links ( if any ) as shown in fig .
[ fig1 ] . * step ( 3 ) : repeat the above two steps until the population converges to an absorbing state ( full cooperators or defectors ) , or stop repeating the above two steps after @xmath42 generations .
we start from a homogeneous random graph by using the method in ref .
@xcite , where all nodes have the same number of edges , randomly linked to arbitrary nodes .
initially , an equal percentage of cooperators and defectors is randomly distributed among the elements of the population .
we run 100 independent simulations for the corresponding parameters @xmath9 , @xmath43 , @xmath0 , @xmath28 , @xmath44 , and @xmath23 .
we also compute the fraction of runs that ended up with 100% cooperators .
if the evolution has not reached an absorbing state after @xmath42 generations , we take the average fraction of cooperators in the population as the final result .
moreover , we observe the time evolution of the network structure and strategy , including the degree - degree correlation coefficient , the degree of heterogeneity , the frequency of cooperators , the fraction of c - c / c - d / d - d links , etc . finally , we confirm that our results are valid for different population size @xmath9 and edge number @xmath32 .
we report a typical time evolution of the network structure as a result of the adaption of social ties in fig .
[ fig2 ] , with relevant parameters @xmath45 , @xmath46 , @xmath47 , @xmath48 , @xmath49 , and @xmath50 .
the emerging social network shows disassortative mixing pattern , indicating that large - degree nodes tend to be connected to low - degree nodes .
the degree - degree correlation coefficient @xmath37 of the network we started from is zero .
once the network structure adaption is in effect , disassortative mixing pattern will be developed .
since the rewiring process is attempted to a random neighbor s neighbor , thus the nodes with large connectivity are more possible to be attached by others . due to
such `` rich gets richer '' , inhomogeneity is induced as shown in fig .
[ fig2](b ) .
the amount of the heterogeneity ( degree variance ) @xmath51 increases in virtue of the rewiring process .
the inset in fig .
[ fig2](b ) plots the cumulative degree distribution of the stationary ( final ) network , which exhibits high heterogeneity with a power - law tail . fig .
[ fig2](c ) displays the evolution of cooperation .
we find that the frequency of cooperators decreases at first due to the temptation to defect , and then because of the adaptive interactions , the cooperation level thrives gradually and the population converges into an absorbing state of 100% cooperators .
the viability of cooperation is also in part promoted by the heterogeneity of the underlying heterogeneity . from fig .
[ fig2](d ) , we can see that local assortative interactions between cooperators is enhanced by structural updating , while assortative interactions between defectors and defectors is inhibited remarkably .
the disassortativity between cooperators and defectors is promoted in the beginning by strategy updating , however , being diminished eventually by structural updating .
clearly , it is thus indicated that the interplay between strategy and structure will facilitate the emergence of cooperation . ) , randomly linked to arbitrary nodes .
the corresponding parameters are @xmath45 , @xmath46 , @xmath47 , @xmath48 , @xmath49 , and @xmath50.[fig2 ] , width=377 ] for different values of @xmath44 .
we ran 100 simulations , starting from @xmath52 cooperators .
the values plotted correspond to the fraction of runs which ended with @xmath53 cooperators . from left to right , @xmath54 respectively .
the inset plots the critical value @xmath55 vs @xmath44 .
@xmath56 , @xmath57 , @xmath48 , and @xmath58.[fig3],width=377 ] to evolution of cooperation for fixed @xmath59 , @xmath56 , @xmath60 , @xmath47 , and @xmath58.[fig4],width=377 ] to evolution of cooperation . fraction of cooperators as a function of @xmath28 for different values @xmath61 . from left to right , @xmath62 respectively .
@xmath56 , @xmath48 , @xmath47 , and @xmath58.[fig5],width=377 ] for different values @xmath23 . from left to right , @xmath63 respectively .
@xmath56 , @xmath48 , @xmath47 , and @xmath64.[fig6],width=377 ] let us consider the effect of the amount of temptation to defect @xmath44 to the evolution of cooperation .
the relevant result is presented in fig .
[ fig3 ] . with increasing @xmath44 ,
the structural updating event must be sufficiently frequent to guarantee the survival of cooperators . in other words , the fraction of individuals chosen for updating social ties
should be accordingly increased to ensure the sustainability of cooperators . when the value of @xmath44 is enlarged , the defectors become more favorable by the nature selection .
nevertheless , with the aid of structural updating , a small fraction of surviving cooperators promotes them into hubs ( large degree nodes ) , since they are attractive to neighborhood . such co - evolution of strategy and structure leads to highly heterogeneous networks in which the cooperators become evolutionarily competitive as demonstrated in refs .
@xcite . for fixed @xmath44
, we observed a critical value @xmath55 for @xmath28 , above which the cooperator will wipe out defectors . for fixed number of rounds @xmath0
, the critical value @xmath55 monotonously increases with increasing @xmath44 , as shown at the inset of fig .
[ fig3 ] . therefore the prompt network adaption prevents cooperators from becoming extinct , further , resulting in an underlying heterogeneous social network which is the `` green house '' for cooperators to prevail under strategy dynamics .
consequently , the entangled co - evolution of strategy and structure promotes the evolution of cooperation among selfish individuals .
furthermore , we investigated the effect of number of rounds @xmath0 to the emergence of cooperation .
fix fixed @xmath28 and other parameters , there exists a critical value for @xmath0 , above which the cooperators will vanish as shown in fig .
indeed , although the structural updating promotes the cooperators to a certain extent , its role will be suppressed by the long - time strategy dynamics ( corresponding to large @xmath0 ) . in our case ,
strategy dynamics is synchronous while structural updating is asynchronous , namely , for each repetition in the simulations , strategy updating happens at a frequency of @xmath65 while the structural updating occurs at a frequency of @xmath28 .
hence the evolution of strategy is much more frequent than that of structure .
thus , with large @xmath0 , able defectors outperform cooperators through strategy dynamics , even though the heterogeneity , resulting from structural updating , is positive to evolution of cooperation .
this result illustrates that even if the evolution of network topology is less frequent than the evolution of strategy , cooperators still have chances to beat defectors under appropriate conditions .
as is well known , cooperation is promoted in the situation where individuals are constrained to interact with few others along the edges of networks with low average connectivity @xcite . to understand the cooperation in real - world interaction networks
of which the average connectivity is normally relatively high , one needs new insight into the underlying mechanism promoting cooperation . here , the role of average connectivity to evolution of cooperation is inspected . in fig .
[ fig5 ] , it is shown that for increasing @xmath61 , the individuals must be able to promptly adjust their social ties for cooperation to thrive , corresponding to increasing @xmath28 .
thus in order to explain the cooperation in communities with a high average number of social ties , the entangled co - evolution of network structure and strategy should be taken into account simultaneously .
on static networks , maximum cooperation level occurs at intermediate average degree @xcite .
moreover , when the connections are dense among individuals ( large average connectivity @xmath61 ) , cooperators die out due to mean - field behavior .
conversely , our results suggest that even in highly - connected network , on account of the proposed structural adaption , cooperators can beat back the defectors and dominate the populations .
finally , we report the influence of changing intensity of selection @xmath23 on the evolution of cooperation in fig .
it is indicated that reducing @xmath23 will demote the influence of the game dynamics , thereby increase the survivability of the less fit . clearly , the smaller the value of @xmath23 is , the smaller the critical value of @xmath28 . in fact , for small @xmath28 , cooperators survival probability increases with decreasing @xmath23 although cooperators are generally less fit . such increased survivability enhances assortative interactions between cooperators through network structure adaption . as a result
, the critical value of @xmath28 above which cooperators dominate defectors decreases with decreasing @xmath23 .
in summary , we have studied the coupled dynamics of strategy evolution and the underlying network structure adaption . we provided a computational model in which individuals are endowed with limited cognitive abilities in the @xmath0-round pdg limited memories for recording the defection times of opponents . after the @xmath0-round game , @xmath28 randomly chosen individuals are allowed to adjust their social ties based on the game results .
the values of @xmath0 and @xmath28 are corresponding to the associated time scales of strategy dynamics and structural updating respectively .
we found that for a given average connectivity of the population and the number of rounds , there is a critical value for the fraction of individuals adapting their social interactions above which cooperators wipe out defectors .
in addition , the critical value of @xmath28 above which cooperators dominate defectors decreases with decreasing intensity of selection @xmath23 .
moreover , for increasing average connectivity , the individuals must be able to swiftly adjust their social ties for cooperators to thrive . finally , the emerging social networks at steady states exhibit nontrivial heterogeneity which is the catalyst for emergence of cooperation among selfish agents . to a certain extent ,
our results shed some light on the underlying mechanism promoting cooperation among selfish individuals , and also provide an alternative insight into the properties accruing to those networked systems and organizations in natural world .
delightful discussion with dr .
wenxu wang is gratefully acknowledged .
this work was supported by nnsfc ( 60674050 and 60528007 ) , national 973 program ( 2002cb312200 ) , national 863 program ( 2006aa04z258 ) and 11 - 5 project ( a2120061303 ) . | we consider the coupled dynamics of the adaption of network structure and the evolution of strategies played by individuals occupying the network vertices .
we propose a computational model in which each agent plays a @xmath0-round prisoner s dilemma game with its immediate neighbors , after that , based upon self - interest , partial individuals may punish their defective neighbors by dismissing the social tie to the one who defects the most times , meanwhile seek for a new partner at random from the neighbors of the punished agent .
it is found that the promotion of cooperation is attributed to the entangled evolution of individual strategy and network structure .
moreover , we show that the emerging social networks exhibit high heterogeneity and disassortative mixing pattern . for a given average connectivity of the population and the number of rounds , there is a critical value for the fraction of individuals adapting their social interactions , above which cooperators wipe out defectors . besides , the effects of the average degree , the number of rounds , and the intensity of selection are investigated by extensive numerical simulations .
our results to some extent reflect the underlying mechanism promoting cooperation .
, , , social networks , network structure adaption , heterogeneity , prisoner s dilemma , cooperation 89.75.hc , 02.50.le.,89.75.fb , 87.23.ge |
one of the surprising results of the many deep _ chandra _ _ x - ray observatory _ observations of clusters of galaxies has been growing evidence for higher surface densities of point sources toward clusters of galaxies relative to blank field observations .
early evidence for this excess point source density was found with measurements of cumulative source counts relative to the expectations from deep , blank field exposures @xcite . while the natural interpretation of these sources was that they are associated with the clusters , most of these sources were sufficiently bright that they would have to be active galactic nuclei ( agn ) if they were cluster members .
this conclusion was unexpected because agn , or any emission - line galaxies , are rarely identified in clusters . in a study of elliptical galaxies
, @xcite noted that [ ] emission was only observed from isolated ellipticals or those in small groups and not from any of the 25 members of rich clusters that were observed .
@xcite significantly improved the statistical significance of this result with a survey of over 1300 galaxies .
he found that the decrease in the number of emission - line galaxies in clusters was true of lenticular and spiral galaxies , as well as ellipticals , and presented the first tentative evidence that agn may be found less commonly in rich clusters than the field .
this result was confirmed with a subsequent , more uniformly - selected spectroscopic survey of 1095 cluster galaxies and 173 field galaxies by @xcite .
these authors found that 7% of cluster galaxies have emission - line nuclei , compared to 31% of field galaxies .
agn were similarly found to be about a factor of five times more rare in clusters : only 1% of cluster galaxies were agn compared to 5% of the field sample .
the smaller agn population in rich clusters relative to the field has been ascribed to differences in the frequency of agn fueling episodes and the smaller numbers of cluster galaxies with significant reservoirs of cold gas ( e.g. * ? ? ?
one of the most popular mechanisms for fueling agn is the merger of two gas - rich galaxies as the strong gravitational torques from a major merger could provide sufficient gas inflow to the central , supermassive black hole to produce an agn ( e.g. * ? ? ?
* ) . however , the high velocity dispersions of rich clusters results in a low merger frequency because the galaxies relative velocities are too high to form bound pairs . if galaxy mergers and significant reservoirs of cold gas are the most important parameters for fueling agn , these arguments provide a natural explanation for the lower agn fraction in clusters compared to the field .
these differences in the galaxy population between rich and poor environments suggest that cluster agn are a valuable probe of agn and galaxy evolution . in @xcite
, we measured redshifts for all of the x - ray sources with bright visible - wavelength counterparts in the field of the @xmath10 cluster abell 2104 .
this work confirmed for the first time that many of the x - ray point sources were in fact cluster members and demonstrated that there may be a sufficiently large agn population in clusters of galaxies for demographic studies .
particularly , we found six x - ray sources associated with cluster members , or approximately 5% of all of the luminous cluster galaxies . since the @xcite study , several other groups have carried out x - ray studies of clusters with either new spectroscopic observations or data obtained from the literature .
these observations include the identification of two galaxies in the @xmath11 cluster ms1054 - 0321 @xcite , eight in the @xmath12 cluster abell 2255 @xcite , six galaxies in the @xmath13 cluster abell 3128 @xcite , a recent study of the coma cluster @xcite , and one in the @xmath14 cluster ms2053 - 04 @xcite .
the x - ray emission from many of these galaxies appears to be due to agn based on their x - ray luminosity and/or x - ray spectral shape .
the two sources in ms1054 - 0321 and the brightest cluster galaxy in ms2053 - 04 have @xmath15 erg s@xmath3 although only a small fraction of the lower - redshift x - ray sources have such high luminosities . for these cases it is possible that the x - ray emission is due to either low - mass x - ray binaries ( lmxbs ) or halos of hot gas .
several recent visible - wavelength @xcite and radio surveys @xcite have also identified agn in clusters of galaxies and studied the environmental dependence of agn .
new studies of _ chandra _ observations of larger samples of additional clusters have identified more examples of x - ray source overdensities @xcite , although this is not true of every study @xcite .
the largest of these studies to date is the investigation by @xcite of the fields of 51 massive clusters from the macs sample @xcite .
these authors measured significant x - ray source overdensities within 3.5 mpc of the cluster cores , and in particular a high concentration of sources within the central 0.5 mpc .
if the x - ray sources in the moderate to high - redshift clusters studied by @xcite and @xcite are associated with cluster galaxies , their high inferred luminosities makes them very strong agn candidates . at progressively lower x - ray luminosities ,
an agn classification based on x - ray luminosity alone becomes less secure because populations of lmxbs , hot gaseous halos , and massive starbursts also produce significant x - ray emission .
redshift surveys of galaxies detected in deep x - ray observations have not always detected agn spectral signatures ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
@xcite ascribed the absence of clear agn signatures from even relatively low - redshift ( @xmath16 ) cluster x - ray sources to a combination of dilution by host galaxy starlight and obscuration by gas and dust . in the absence of clear agn spectral signatures ,
multiwavelength spectral shape is the clearest indicator of the physical origin of the x - ray emission .
the x - ray and visible - wavelength properties of resolved lmxb populations have now been carefully measured with _
chandra _ for a large number of nearby , luminous elliptical galaxies @xcite and have been observed to produce broad band ( 0.38 kev ) x - ray luminosities as high as @xmath17 erg s@xmath3 .
galaxies with x - ray emission from lmxbs follow a fairly tight relation between @xmath18band and x - ray flux because the number of lmxbs approximately scales with the stellar mass of the host galaxy .
hot gaseous halos from large , early - type galaxies can also produce significant x - ray emission . while the intracluster medium may strip these halos , sources with soft x - ray luminosities of @xmath19 erg s@xmath3
have been observed in many nearby clusters with _
chandra _ ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
there is also a relation between x - ray and stellar luminosity for emission from hot gas , although with larger scatter @xcite and the larger scatter is likely due to environmental influences @xcite .
finally , the most powerful , local starburst galaxies such as arp 220 and ngc 3256 can have x - ray luminosities as high as @xmath20 erg s@xmath3 due to hot gas associated with the starburst @xcite , although these galaxies also have luminous , visible - wavelength emission lines and are expected to be rare in rich clusters .
in addition to their multiwavelength properties , resolution is another discriminant between lmxbs , hot gas , star formation , and agn .
the first three mechanisms will all produce extended x - ray emission and should be resolved in _
_ observations of low - redshift clusters . since our initial work on abell 2104
, we have completed a spectroscopic study of seven additional , low - redshift ( @xmath21 ) clusters of galaxies .
we have measured redshifts for the visible - wavelength counterparts of all of the x - ray sources in the fields of these clusters more luminous than approximately one magnitude fainter than @xmath22 and identified a total of 40 x - ray sources in eight clusters . in this paper
we describe their multiwavelength properties , conclude that most of these sources must be low - luminosity agn , and demonstrate how the x - ray sources in these clusters can account for the excess point source surface densities .
the observations and data analysis are described next in [ sec : obs ] , while their multiwavelength properties are analyzed in section
[ sec : res ] .
the results of this study , in particular the nature of the x - ray emission , calculation of the agn fraction , and implications for the x - ray source overdensities observed toward clusters are discussed in [ sec : dis ] .
we summarize this work in the last section . in future papers
we will investigate the relationship of these x - ray sources to the cluster galaxy population and perform a detailed sensitivity analysis to derive the x - ray luminosity function in clusters of galaxies .
lccccrcrrrrrc abell 3125 & 0.059 & 03:27:22 & -53:30:00 & 1.6 & 892 & acis - i & 9 & @xmath23 & 2002 sep & 2003 oct & 6 & @xmath24 + abell 3128 & 0.060 & 03:30:24 & -52:32:00 & 1.5 & 893 & acis - i & 19 & @xmath23 & 2002 sep & 2003 oct & 10 & @xmath25 + abell 644 & 0.070 & 08:17:25 & -07:30:42 & 6.4 & 2211 & acis - i & 29 & @xmath23 & 2003 mar & 2003 mar & 2 & @xmath26 + abell 2104 & 0.155 & 15:40:07 & -03:17:24 & 8.9 & 895 & acis - s & 49 & @xmath27 & 2003 mar & 2002 apr & 6 & @xmath28 + abell 1689 & 0.183 & 13:11:30 & -01:20:10 & 1.8 & 1663 & acis - i & 10 & @xmath27 & 2003 mar & 2003 mar & 2 & @xmath29 + abell 2163 & 0.203 & 16:15:46 & -06:08:55 & 12.0&1653 & acis - i & 70 & @xmath23 & 2003 mar & 2003 mar & 3 & @xmath30 + ms1008 & 0.301 & 10:10:32 & -12:39:32 & 6.7 & 926 & acis - i & 41 & @xmath27 & 2003 mar & 2003 mar & 5 & @xmath31 + ac 114 & 0.312 & 22:58:49 & -34:48:09 & 1.3 & 1562 & acis - s & 72 & @xmath27 & 2002 sep & 2002 oct & 6 & @xmath32 + this sample of eight clusters was chosen because the clusters have publicly available _ chandra _ acis data with sufficiently long exposures to detect x - ray sources as faint as @xmath33 erg s@xmath3 at the cluster redshift .
all of these clusters are also observable from las campanas observatory in chile .
the cluster sample , redshift , and basic data about the x - ray , visible - wavelength imaging , and spectroscopic observations are presented in table [ tbl : obslog ] .
the field of view of the survey in each cluster is set by the field size of the camera employed to obtain the x - ray data and is either @xmath34 ( acis - s ) or @xmath35 ( acis - i ) .
we obtained visible - wavelength imaging with the 2.5 m du pont telescope , identified likely counterparts to x - ray sources in these fields , and then obtained spectroscopy with the 6.5 m magellan clay telescope to determine which x - ray counterparts belong to the clusters .
this search identified a sample of 40 cluster members , whose coordinates , redshifts , and visible - wavelength properties are provided in table [ tbl : catalog ] .
we describe the x - ray , imaging , and spectroscopic observations in the next subsections .
lcccccccccll a3125 - 1 & j032723.5 - 532535 & 0.0642 & 15.93 ( 0.03 ) & 1.78 ( 0.06 ) & 0.65 ( 0.05 ) & ... & ... & ... & ... & lit - z & d80 - 052 + a3125 - 2 & j032754.7 - 532217 & 0.0583 & 15.10 ( 0.03 ) & 1.81 ( 0.06 ) & 0.69 ( 0.05 ) & ... & ... & ... & ... & ... & 2masxj03275473 - 5322169 + a3125 - 3 & j032752.0 - 532610 & 0.0609 & 15.26 ( 0.03 ) & 1.82 ( 0.06 ) & 0.70 ( 0.05 ) & ... & ... & ... & ... & ... & 2masxj03275206 - 5326099 + a3125 - 4 & j032724.8 - 532518 & 0.0628 & 15.26 ( 0.03 ) & 1.84 ( 0.06 ) & 0.67 ( 0.05 ) & ... & ... & ... & ... & ... & ... + a3125 - 5 & j032705.0 - 532140 & 0.0625 & 16.08 ( 0.03 ) & 1.78 ( 0.06 ) & 0.71 ( 1.05 ) & ... & -12.0 ( 0.5 ) & ... & h@xmath36,[nii],[sii ] & ... & eso155-ig031 + a3125 - 6 & j032823.6 - 533436 & 0.0630 & ... & ... & ... & ... & ... & ... & ... & lit - z , lit - pos & eso155-g037,rgc02 - 241 + a3128 - 1 & j033018.6 - 522856 & 0.0549 & 15.51 ( 0.03 ) & 1.64 ( 0.06 ) & 0.48 ( 0.05 ) & ... & ... & ... & ... & ... & csrg0262 + a3128 - 2 & j032941.4 - 522936 & 0.0586 & 17.20 ( 0.03 ) & 1.20 ( 0.06 ) & 0.36 ( 0.05 ) & ... & -2.8 ( 0.5 ) & 1.1 ( 0.2 ) & ... & ... & s03 - 6,apmuks(bj)b032816.70 - 523949.5 + a3128 - 3 & j032931.1 - 522716 & 0.0586 & ... & ... & ... & ... & ... & ... & ... & lit - z , edge & d80 - 135 + a3128 - 4 & j033051.0 - 523031 & 0.0571 & 15.04 ( 0.03 ) & 1.51 ( 0.06 ) & 0.59 ( 0.05 ) & ... & ... & ... & ... & ... & s03 - 3,2masxj03305107 - 5230315 + a3128 - 5 & j033046.0 - 522335 & 0.0588 & ... & ... & ... & ... & ... & ... & ... & lit - pos,@xmath37 & 2masxj03304589 - 5223385 + a3128 - 6 & j033017.3 - 523407 & 0.0544 & 16.96 ( 0.03 ) & 1.28 ( 0.06 ) & 0.33 ( 0.05 ) & ... & ... & ... & ... & ... & s03 - 5 + a3128 - 7 & j033013.6 - 523730 & 0.0648 & 15.15 ( 0.03 ) & 1.41 ( 0.06 ) & 0.24 ( 0.05 ) & ... & ... & ... & ... & ... & s03 - 4,2masxj03301366 - 5237304 + a3128 - 8 & j032950.6 - 523447 & 0.0643 & 15.31 ( 0.03 ) & 1.64 ( 0.06 ) & 0.48 ( 0.05 ) & ... & ... & ... & ... & @xmath37 & 2masxj03295060 - 5234471 + a3128 - 9 & j033039.3 - 523206 & 0.0622 & 16.50 ( 0.03 ) & 1.63 ( 0.06 ) & 0.46 ( 0.05 ) & ... & ... & ... & ... & ... & 2masxj03303924 - 5232066 + a3128 - 10 & j033038.4 - 523710 & 0.0599 & 14.94 ( 0.03 ) & 1.54 ( 0.06 ) & 0.59 ( 0.05 ) & ... & ... & ... & ... & @xmath37 & s03 - 2,2masxj03303848 - 5237096 + a644 - 1 & j081739.6 - 073309 & 0.0726 & 16.90 ( 0.03 ) & 0.95 ( 0.06 ) & 0.35 ( 0.05 ) & ... & ... & ... & nls1 & ... & ... + a644 - 2 & j081748.1 - 073732 & 0.0783 & 15.93 ( 0.03 ) & 1.81 ( 0.06 ) & 0.66 ( 0.05 ) & ... & ... & 2.0 ( 0.3 ) & ... & ... & ... + a2104 - 1 & j154023.6 - 031347 & 0.159 & 17.30 ( 0.03 ) & 2.43 ( 0.06 ) & 0.81 ( 0.05 ) & 0.98 ( 0.05 ) & ... & ... & s2 & ... & mkmt02 - 1 + a2104 - 2 & j154016.7 - 031507 & 0.161 & 18.65 ( 0.03 ) & 2.32 ( 0.06 ) & 0.78 ( 0.05 ) & 0.85 ( 0.05 ) & ... & ... & ... & ... & mkmt02 - 2 + a2104 - 3 & j154009.4 - 031519 & 0.155 & 17.16 ( 0.03 ) & 2.46 ( 0.06 ) & 0.79 ( 0.05 ) & 0.85 ( 0.05 ) & ... & ... & ... & ... & mkmt02 - 3 + a2104 - 4 & j154014.0 - 031704 & 0.157 & 19.60 ( 0.03 ) & 2.22 ( 0.06 ) & 0.69 ( 0.05 ) & 0.78 ( 0.05 ) & ... & ... & ... & @xmath37 & mkmt02 - 4 + a2104 - 5 & j154019.5 - 031825 & 0.162 & 19.56 ( 0.03 ) & 2.04 ( 0.06 ) & 0.67 ( 0.05 ) & 0.76 ( 0.05 ) & ... & ... & ... & ... & mkmt02 - 5 + a2104 - 6 & j154003.9 - 032039 & 0.154 & 17.77 ( 0.03 ) & 2.43 ( 0.06 ) & 0.74 ( 0.05 ) & 0.87 ( 0.05 ) & ... & ... & ... & @xmath37 & mkmt02 - 6 + a1689 - 1 & j131145.4 - 012336 & 0.187 & 17.75 ( 0.03 ) & 2.19 ( 0.06 ) & 0.77 ( 0.05 ) & 0.70 ( 0.05 ) & ... & ... & ... & ... & apmuks(bj)b130911.16 - 010741.8 + a1689 - 2 & j131135.6 - 012012 & 0.200 & 18.34 ( 0.03 ) & 1.41 ( 0.06 ) & 0.48 ( 0.05 ) & 0.50 ( 0.05 ) & ... & ... & s2 & @xmath38 & apmuks(bj)b130901.41 - 010417.1 + a2163 - 1 & j161524.4 - 060904 & 0.201 & 17.76 ( 0.03 ) & 1.89 ( 0.06 ) & 0.70 ( 0.05 ) & ... & -14.0 ( 1.0 ) & 2.5 ( 0.5 ) & s2 & ... & ... + a2163 - 2 & j161548.9 - 061512 & 0.201 & 20.04 ( 0.03 ) & 2.24 ( 0.07 ) & 0.76 ( 0.05 ) & ... & ... & ... & ... & @xmath37 & ... + a2163 - 3 & j161543.6 - 061730 & 0.200 & 18.34 ( 0.03 ) & 2.52 ( 0.06 ) & 0.86 ( 0.05 ) & ... & ... & ... & ... & ... & ... + ms1008 - 1 & j101018.7 - 123744 & 0.298 & 19.60 ( 0.03 ) & 2.12 ( 0.06 ) & 0.71 ( 0.05 ) & 0.83 ( 0.05 ) & -6.0 ( 0.5 ) & 2.9 ( 0.3 ) & [ oiii ] & ... & ppp-001519 + ms1008 - 2 & j101005.2 - 123834 & 0.305 & 21.53 ( 0.04 ) & 2.29 ( 0.10 ) & 0.77 ( 0.15 ) & 0.72 ( 0.08 ) & ... & ... & ... & @xmath37 & ... + ms1008 - 3 & j101035.3 - 124021 & 0.309 & 19.17 ( 0.03 ) & 2.65 ( 0.06 ) & 1.02 ( 0.05 ) & 0.96 ( 0.05 ) & ... & ... & ... & ... & ppp-000760 + ms1008 - 4 & j101026.5 - 123810 & 0.297 & 19.71 ( 0.03 ) & 2.27 ( 0.06 ) & 0.86 ( 0.05 ) & 0.84 ( 0.05 ) & ... & 0.7 ( 0.2 ) & ... & ... & ppp-001427 + ms1008 - 5 & j101032.3 - 123934 & 0.301 & 18.96 ( 0.03 ) & 2.46 ( 0.06 ) & 0.92 ( 0.05 ) & 0.88 ( 0.05 ) & ... & 0.8 ( 0.2 ) & ... & ... & ... + ac114 - 1 & j225852.9 - 344846 & 0.304 & 20.01 ( 0.03 ) & 1.83 ( 0.06 ) & 0.77 ( 0.05 ) & 0.47 ( 0.05 ) & ... & 8.0 ( 0.6 ) & ... & ... & cn84 - 191 + ac114 - 2 & j225851.4 - 344912 & 0.321 & 21.42 ( 0.04 ) & 2.09 ( 0.20 ) & 1.24 ( 0.20 ) & 0.74 ( 0.08 ) & ... & ... & ... & ... & ... + ac114 - 3 & j225849.3 - 344701 & 0.313 & 19.26 ( 0.03 ) & 2.01 ( 0.06 ) & 0.96 ( 0.05 ) & 0.69 ( 0.08 ) & -3.0 ( 0.8 ) & ... & h@xmath39,[oiii],h@xmath36 & ... & cn84 - 087 + ac114 - 4 & j225846.2 - 344945 & 0.318 & 20.98 ( 0.04 ) & 2.32 ( 0.15 ) & 1.16 ( 0.15 ) & 0.70 ( 0.08 ) & ... & 3.2 ( 0.1 ) & ... & ... & cbb2001 - 0552 + ac114 - 5 & j225857.4 - 345059 & 0.322 & 21.02 ( 0.05 ) & 2.30 ( 0.15 ) & 1.80 ( 0.15 ) & 1.50 ( 0.08 ) & ... & ... & ... & lit - z & cbb2001 - 0365 + ac114 - 6 & j225842.0 - 344747 & 0.310 & 18.14 ( 0.03 ) & 2.68 ( 0.06 ) & 1.27 ( 0.05 ) & 0.79 ( 0.05 ) & ... & ... & ... & lit - z & cn84 - 003 + the x - ray observations were initially processed with a standard application of the ciao pipeline and the _ wavdetect _ source identification tool , although they have since been reprocessed with a custom x - ray photometry package named xproces ( e. kim et al . 2006 , _ in prep _ ) . as this is the first application of the package we provide a brief description here , although defer a complete description to a future paper .
the first step is to produce a new level 2 event file following the standard science threads , including suitable filtering to remove flare events and cti correction .
xproces then runs the _ wavdetect _ task on the unbinned images within the ciao package ( version 3.2 ) to identify point sources .
we set the detection threshold at @xmath40 , searched scale sizes of @xmath41 pixels to identify sources , and employed an exposure map for an energy of 1.5 kev .
the _ wavdetect _ detection threshold of @xmath40 corresponds to one false detection per @xmath42 pixels , or one per unbinned acis chip .
we therefore expect between one ( acis - s ) and four ( acis - i ) false detections per cluster field .
in the same area ( 1000x1000 _ chandra _ pixels or @xmath34 ) in which we expect one false detection , there @xmath43 mag galaxies . for a generous matching tolerance of @xmath44 ,
the probability of a chance superposition between a false detection and a bright galaxy is less than 3% for the entire survey area .
this is an upper limit on the false detection probability , as we also have a strict signal - to - noise cut on the x - ray data that will eliminate many false detections .
the probability that a false detection will be cross - identified with a cluster member is lower still because many bright galaxies are not cluster members .
xproces prepares the event file for photometry with several additional steps .
the first step is detection and removal of new hot pixels and cosmic ray afterglow events , followed by the standard charge transfer inefficiency and time - dependent gain correction . at this stage flare events
are detected and removed as well .
the properties of the sources are then determined with aperture photometry , where the aperture size is set by the 95% encircled energy fraction of the psf at 1.5 kev ( the energy of maximum quantum efficiency ) at a given off - axis angle .
we measure source counts in five x - ray bands following the definitions used by @xcite .
these are the broad ( 0.3 - 8 kev ) , soft ( 0.3 - 2.5 kev ) , hard ( 2.5 - 8 kev ) , @xmath45 ( 0.3 - 0.9 kev ) , and @xmath46 ( 0.9 - 2.5 kev ) bands .
the same aperture size is used for all bands even though there are slight changes in the size of the psf with energy .
this is a small source of error .
for example , the 95% encircled energy at 1.5 kev corresponds to the 92% encircled energy at 3.5 kev . we do not use the 95% encircled energy aperture at 3.5 kev because it would increase the potential for source overlap .
the counts for each source in these five bands are presented in table [ tbl : xdata ] . only sources detected at greater than 3@xmath47 significance in the broad band
are included in this sample and all of these sources have at least seven counts in the broad band .
count values are not shown if the source counts corresponded to less than a 3@xmath47 detection in any given band .
the background flux for each source has been measured within an annulus from two to five times the source radius and this value has been subtracted from the aperture measurement . we have calculated the uncertainties and the 3@xmath47 significance with the @xcite approximation for small @xmath48 . for each source
we calculate several quantities to parametrize the x - ray spectral energy distribution .
following @xcite we define the hardness ratio @xmath49 and the colors @xmath50 and @xmath51 .
these quantities are commonly - used characterizations of spectral shape .
x - ray quantiles are a recently developed measure of spectral shape that are based on the distribution of energy values in a spectrum @xcite .
a quantile is defined such that if @xmath48% of the counts in a spectrum are below some energy @xmath52 , then the quantile is : @xmath53 where @xmath54 and @xmath55 are the lower and upper energy boundaries of the spectrum and are taken here to be 0.3 and 8.0 kev , respectively .
the virtue of x - ray quantiles is that they are less susceptible to biases due the relative energy sensitivity of an instrument .
following @xcite , we have calculated the quantities @xmath56 and @xmath57 .
the hardness ratio , x - ray colors , and quantiles for our sample are listed in table [ tbl : xdata ] .
we calculate fluxes in the broad , soft , and hard bands with an energy conversion factor calculated from the foreground galactic extinction toward each cluster @xcite and assume a @xmath58 power - law typical of agn , where @xmath59 is the slope of the power - law photon flux density @xmath60 .
the energy conversion factor is calculated for each chip at the maximum value of the exposure map and then scaled to the location of each source with the exposure map .
this approach is similar to that described in @xcite .
we calculate the x - ray luminosity of each source based on the luminosity distance ( e.g. * ? ? ?
* ) to each cluster and assume a cosmology with @xmath61 .
the fluxes are measured in the observed bandpasses and the luminosities have been calculated for the restframe bandpass assuming a @xmath58 model .
the source fluxes and luminosities are presented in table [ tbl : lum ] . for a @xmath58 power - law plus galactic absorption model
we are complete to broad - band luminosity limits of @xmath62 erg s@xmath3 ( a644 at @xmath63 ) to @xmath64 erg s@xmath3(a1689 at @xmath65 ) .
these luminosity limits are listed as the last column of table [ tbl : obslog ] and correspond to a five count detection on axis .
thirty eight of the 40 sources are unresolved , based on a comparison to the _ chandra _ psf at similar off - axis angle .
the remaining two sources ( ac114 - 1 , ac114 - 2 ) are sufficiently embedded in the diffuse , intracluster gas that it is difficult to determine if the emission due to the galaxy is unresolved .
lccccccccccc a3125 - 1 & 21.2 ( 3.7 ) & 20.4 ( 3.6 ) & ... & 5.0 ( 1.5 ) & 15.4 ( 3.0 ) & -0.5 ( 0.3 ) & ... & ... & -0.9 ( 0.1 ) & 1.7 ( 0.8 ) + a3125 - 2 & 15.4 ( 2.8 ) & 16.0 ( 3.0 ) & ... & 11.2 ( 2.4 ) & 4.9 ( 1.3 ) & 0.4 ( 0.4 ) & ... & ... & -1.1 ( 0.1 ) & 2.2 ( 0.4 ) + a3125 - 3 & 7.6 ( 1.8 ) & 8.3 ( 2.0 ) & ... & 4.8 ( 1.4 ) & 3.5 ( 1.1 ) & 0.1 ( 0.5 ) & ... & ... & -1.1 ( 0.3 ) & 1.7 ( 0.6 ) + a3125 - 4 & 8.0 ( 1.9 ) & 6.3 ( 1.7 ) & ... & ... & 3.4 ( 1.1 ) & ... & ... & ... & -0.9 ( 0.6 ) & 0.9 ( 0.6 ) + a3125 - 5 & 25.9 ( 4.1 ) & 11.9 ( 2.5 ) & 14.0 ( 2.8 ) & ... & 11.4 ( 2.5 ) & ... & -0.1 ( 0.2 ) & 0.1 ( 0.3 ) & -0.4 ( 0.1 ) & 1.5 ( 0.3 ) + a3125 - 6 & 13.7 ( 2.6 ) & 13.0 ( 2.7 ) & ... & 3.9 ( 1.2 ) & 9.1 ( 2.1 ) & -0.4 ( 0.4 ) & ... & ... & -1.0 ( 0.3 ) & 1.1 ( 0.6 ) + a3128 - 1 & 19.4 ( 3.4 ) & 17.9 ( 3.3 ) & ... & 9.6 ( 2.2 ) & 8.4 ( 2.0 ) & 0.1 ( 0.3 ) & ... & ... & -1.1 ( 0.2 ) & 1.5 ( 0.8 ) + a3128 - 2 & 176.7 ( 12.2 ) & 124.2 ( 10.1 ) & 52.5 ( 6.2 ) & 39.0 ( 5.3 ) & 85.1 ( 8.2 ) & -0.3 ( 0.1 ) & 0.2 ( 0.1 ) & -0.4 ( 0.1 ) & -0.7 ( 0.1 ) & 0.9 ( 0.1 ) + a3128 - 3 & 15.2 ( 2.5 ) & 11.2 ( 2.2 ) & 4.0 ( 1.0 ) & ... & 9.2 ( 2.0 ) & ... & 0.4 ( 0.5 ) & -0.5 ( 0.4 ) & -0.6 ( 0.3 ) & 1.3 ( 1.2 ) + a3128 - 4 & 98.2 ( 8.8 ) & 96.5 ( 8.8 ) & ... & 43.0 ( 5.6 ) & 53.5 ( 6.3 ) & -0.1 ( 0.1 ) & ... & ... & -1.0 ( 0.0 ) & 1.7 ( 0.1 ) + a3128 - 5 & 14.4 ( 2.3 ) & 10.2 ( 1.9 ) & 4.2 ( 1.0 ) & 4.8 ( 1.3 ) & 5.4 ( 1.2 ) & 0.0 ( 0.5 ) & 0.1 ( 0.6 ) & -0.4 ( 0.5 ) & -0.7 ( 0.4 ) & 0.6 ( 0.5 ) + a3128 - 6 & 216.8 ( 13.7 ) & 175.1 ( 12.2 ) & 41.8 ( 5.5 ) & 68.7 ( 7.3 ) & 106.4 ( 9.3 ) & -0.2 ( 0.1 ) & 0.4 ( 0.1 ) & -0.6 ( 0.1 ) & -0.8 ( 0.1 ) & 0.9 ( 0.1 ) + a3128 - 7 & 33.9 ( 4.6 ) & 33.7 ( 4.7 ) & ... & 20.2 ( 3.6 ) & 13.5 ( 2.6 ) & 0.2 ( 0.2 ) & ... & ... & -1.1 ( 0.1 ) & 1.7 ( 0.2 ) + a3128 - 8 & 26.8 ( 3.2 ) & 20.4 ( 2.7 ) & 6.4 ( 1.4 ) & 5.2 ( 1.2 ) & 15.2 ( 2.3 ) & -0.5 ( 0.4 ) & 0.4 ( 0.4 ) & -0.5 ( 0.3 ) & -0.8 ( 0.2 ) & 1.1 ( 0.6 ) + a3128 - 9 & 25.4 ( 4.0 ) & 19.2 ( 3.4 ) & 6.2 ( 1.6 ) & 3.4 ( 1.1 ) & 15.9 ( 3.0 ) & -0.7 ( 0.4 ) & 0.4 ( 0.3 ) & -0.5 ( 0.2 ) & -0.6 ( 0.2 ) & 1.2 ( 0.4 ) + a3128 - 10 & 38.7 ( 5.1 ) & 38.7 ( 5.2 ) & ... & 13.3 ( 2.8 ) & 25.4 ( 4.0 ) & -0.3 ( 0.2 ) & ... & ... & -1.0 ( 0.1 ) & 1.8 ( 0.2 ) + a644 - 1 & 250.4 ( 14.6 ) & 167.0 ( 11.8 ) & 83.4 ( 8.1 ) & 98.3 ( 8.9 ) & 68.7 ( 7.2 ) & 0.2 ( 0.1 ) & -0.1 ( 0.1 ) & -0.3 ( 0.1 ) & -0.7 ( 0.1 ) & 0.4 ( 0.1 ) + a644 - 2 & 28.8 ( 3.8 ) & 6.1 ( 1.3 ) & 22.6 ( 3.5 ) & ... & 3.4 ( 0.8 ) & ... & -0.8 ( 0.6 ) & 0.6 ( 0.3 ) & -0.1 ( 0.2 ) & 1.6 ( 0.3 ) + a2104 - 1 & 3778.3 ( 60.3 ) & 1797.0 ( 41.3 ) & 1981.3 ( 43.4 ) & 66.8 ( 7.0 ) & 1730.3 ( 40.5 ) & -1.4 ( 0.1 ) & -0.1 ( 0.0 ) & 0.0 ( 0.0 ) & -0.4 ( 0.0 ) & 1.2 ( 0.0 ) + a2104 - 2 & 80.9 ( 7.8 ) & 22.5 ( 3.6 ) & 58.3 ( 6.6 ) & 3.7 ( 1.1 ) & 18.9 ( 3.3 ) & -0.7 ( 0.4 ) & -0.5 ( 0.1 ) & 0.4 ( 0.1 ) & -0.2 ( 0.1 ) & 1.3 ( 0.1 ) + a2104 - 3 & 29.3 ( 4.1 ) & 29.1 ( 4.2 ) & ... & 14.3 ( 2.7 ) & 14.8 ( 2.8 ) & 0.0 ( 0.2 ) & ... & ... & -1.1 ( 0.1 ) & 1.7 ( 0.2 ) + a2104 - 4 & 25.5 ( 3.3 ) & 26.9 ( 3.6 ) & ... & 6.3 ( 1.5 ) & 20.6 ( 3.1 ) & -0.5 ( 0.3 ) & ... & ... & -0.9 ( 0.2 ) & 1.6 ( 0.3 ) + a2104 - 5 & 166.1 ( 11.7 ) & 136.5 ( 10.5 ) & 29.5 ( 4.4 ) & 36.0 ( 5.0 ) & 100.5 ( 8.9 ) & -0.4 ( 0.1 ) & 0.5 ( 0.1 ) & -0.6 ( 0.1 ) & -0.9 ( 0.1 ) & 1.2 ( 0.1 ) + a2104 - 6 & 28.6 ( 3.9 ) & 12.2 ( 2.2 ) & 16.4 ( 3.0 ) & ... & 13.0 ( 2.4 ) & ... & -0.1 ( 0.2 ) & 0.1 ( 0.3 ) & -0.3 ( 0.1 ) & 1.0 ( 0.4 ) + a1689 - 1 & 10.6 ( 2.2 ) & 10.4 ( 2.3 ) & ... & ... & 7.8 ( 1.9 ) & ... & ... & ... & -0.8 ( 0.2 ) & 1.6 ( 0.6 ) + a1689 - 2 & 16.1 ( 2.8 ) & 11.6 ( 2.3 ) & 4.5 ( 1.2 ) & ... & 11.3 ( 2.3 ) & ... & 0.4 ( 0.4 ) & -0.4 ( 0.4 ) & -0.7 ( 0.5 ) & 0.9 ( 0.3 ) + a2163 - 1 & 159.4 ( 11.0 ) & 11.3 ( 1.9 ) & 148.1 ( 10.9 ) & 6.3 ( 1.5 ) & 5.0 ( 1.1 ) & 0.1 ( 0.5 ) & -1.5 ( 0.4 ) & 0.9 ( 0.1 ) & 0.0 ( 0.0 ) & 1.9 ( 0.1 ) + a2163 - 2 & 35.8 ( 4.9 ) & 25.6 ( 4.1 ) & 10.1 ( 2.2 ) & ... & 23.0 ( 3.8 ) & ... & 0.4 ( 0.2 ) & -0.4 ( 0.2 ) & -0.6 ( 0.1 ) & 1.1 ( 0.2 ) + a2163 - 3 & 12.1 ( 2.3 ) & 12.5 ( 2.5 ) & ... & 3.3 ( 1.0 ) & 9.2 ( 2.1 ) & -0.4 ( 0.5 ) & ... & ... & -0.9 ( 0.1 ) & 1.9 ( 0.8 ) + ms1008 - 1 & 7.9 ( 1.9 ) & 4.5 ( 1.3 ) & 3.5 ( 1.1 ) & ... & ... & ... & ... & ... & -0.6 ( 0.5 ) & 0.4 ( 0.7 ) + ms1008 - 2 & 8.1 ( 2.0 ) & 5.6 ( 1.5 ) & ... & ... & 5.8 ( 1.6 ) & ... & ... & ... & -0.5 ( 0.7 ) & 0.8 ( 0.5 ) + ms1008 - 3 & 67.0 ( 6.9 ) & 52.1 ( 6.0 ) & 14.9 ( 2.9 ) & 12.9 ( 2.7 ) & 39.2 ( 5.1 ) & -0.5 ( 0.2 ) & 0.4 ( 0.2 ) & -0.6 ( 0.1 ) & -0.7 ( 0.1 ) & 1.1 ( 0.3 ) + ms1008 - 4 & 8.9 ( 1.9 ) & 7.1 ( 1.6 ) & ... & ... & 5.7 ( 1.4 ) & ... & ... & ... & -0.9 ( 0.5 ) & 1.1 ( 1.1 ) + ms1008 - 5 & 37.3 ( 3.3 ) & 29.2 ( 2.9 ) & 8.1 ( 1.4 ) & 10.3 ( 1.9 ) & 18.9 ( 2.1 ) & -0.3 ( 0.3 ) & 0.4 ( 0.4 ) & -0.6 ( 0.3 ) & -0.8 ( 0.1 ) & 1.3 ( 0.5 ) + ac114 - 1 & 933.1 ( 29.0 ) & 910.9 ( 28.7 ) & 22.2 ( 3.4 ) & 706.1 ( 25.4 ) & 204.7 ( 12.8 ) & 0.5 ( 0.0 ) & 1.0 ( 0.1 ) & -1.0 ( 0.0 ) & -1.4 ( 0.0 ) & 0.8 ( 0.1 ) + ac114 - 2 & 47.6 ( 4.3 ) & 32.1 ( 3.3 ) & 15.5 ( 2.6 ) & 7.1 ( 1.1 ) & 25.0 ( 3.0 ) & -0.6 ( 0.4 ) & 0.2 ( 0.2 ) & -0.3 ( 0.2 ) & -0.7 ( 0.2 ) & 0.7 ( 0.2 ) + ac114 - 3 & 122.0 ( 9.6 ) & 108.2 ( 9.0 ) & 13.8 ( 2.6 ) & 46.9 ( 5.7 ) & 61.3 ( 6.5 ) & -0.1 ( 0.1 ) & 0.6 ( 0.2 ) & -0.8 ( 0.1 ) & -0.9 ( 0.1 ) & 1.1 ( 0.2 ) + ac114 - 4 & 63.2 ( 6.5 ) & 23.0 ( 3.4 ) & 40.2 ( 5.3 ) & ... & 22.9 ( 3.6 ) & ... & -0.2 ( 0.1 ) & 0.3 ( 0.2 ) & -0.1 ( 0.1 ) & 1.1 ( 0.1 ) + ac114 - 5 & 7.8 ( 1.5 ) & 5.7 ( 1.2 ) & ... & 5.5 ( 1.3 ) & ... & ... & ... & ... & -1.3 ( 1.5 ) & 0.6 ( 0.7 ) + ac114 - 6 & 7.5 ( 1.3 ) & 7.2 ( 1.3 ) & ... & 3.6 ( 0.9 ) & 3.6 ( 0.8 ) & 0.0 ( 0.7 ) & ... & ... & -1.1 ( 1.3 ) & 0.7 ( 0.7 ) + we obtained ground - based @xmath23 photometry for the three @xmath66 clusters and @xmath27 photometry for the five higher - redshift clusters with the 2.5 m du pont telescope at las campanas observatory . the observations obtained in 2003 march employed the wfccd camera , which has a plate scale of @xmath67 , and the observations in 2002 september employed the tek5 ccd camera , which has a plate scale of @xmath68 .
these observations encompass nearly all of the fields - of - view of the primary _ chandra _ camera ( acis - i or acis - s ) . for two clusters ( a3125 and
a3128 ) we obtained multiple pointings with the tek5 camera to tile the _ chandra _ field of view .
these images were combined into single mosaics with the swarp package by e. bertin .
these images were all obtained under photometric conditions and calibrated with multiple observations of standard star fields from the data compiled by p.b .
stetson onto the vega magnitude system .
we calculated astrometric solutions for the images with the wcstools package @xcite , cataloged the sources with the sextractor package @xcite , and matched these sources with the _ chandra _ source catalog to create prioritized catalogs for multislit spectroscopy .
the sextractor detection parameters were approximately the default settings and the most relevant include the requirement that each source include a minimum of six pixels at least 3@xmath47 brighter than the sky level .
typically there were several x - ray sources with obvious and bright visible - wavelength counterparts in each field .
these matches were used to fine tune the registration of the x - ray catalog to match the wcs solution for the visible - wavelength catalogs .
sources were targeted for spectroscopy if they were within a generous @xmath44 of the centroid of an x - ray source .
small @xmath69band images of each x - ray source were also generated to inspect the quality of each match and the size of the x - ray psf was considered in this inspection .
the observed visible - wavelength photometry for all but three galaxies in our sample are provided in table [ tbl : catalog ] .
the photometric measurements were obtained from sextractor best magnitudes , which for these relatively crowded fields are a corrected isophotal magnitude calculated after subtraction of neighboring galaxies that may contaminate the photometry .
these three galaxies fell outside ( or at the edge ) of the field of view of our tek5 or wfccd images and are discussed further below .
visible - wavelength images of all of these sources are shown in figure [ fig : charts ] along with the 95% x - ray error circles .
these error circles were calculated by generating fake sources with the marx software and comparing their input positions to the measured positions with wavdetect .
the distribution of the differences in position were used to calculate the 95% x - ray error circle , where the size of the error circle depends on off - axis angle and the number of counts from the source .
this technique will be described in more detail in m. kim et al .
( 2006 , _ in prep _ ) . in all but two cases ( abell 3125 - 6 and abell 3128 - 5 ) these images shown in figure [ fig : charts ] are from our @xmath69band observations .
images of the two exceptions were obtained from the _ digitized sky survey_. the x - ray and visible - wavelength photometric catalogs were used to design multislit masks for spectroscopic observations with the ldss-2 spectrograph on the 6.5 m clay telescope of the magellan project .
all x - ray sources with @xmath69band counterparts in our deep images from the du pont telescope , corresponding to sources with @xmath70 mag , were given first priority .
second priority was assigned to other bright , resolved sources .
each mask included approximately 15 25 slits , where lower - redshift clusters typically had fewer sources per masks .
we observed 3 10 masks per cluster and 5 10 x - ray sources per mask .
all masks were designed with @xmath71 slit widths .
for some of the lower - redshift clusters we also designed additional masks without x - ray sources in order to obtain more data on the cluster galaxy population .
all of the x - ray sources were assigned to at least one multislit observation .
each mask was observed for between 1200 and 5400s , with longer observations for higher - redshift clusters .
flatfields and comparison lamps were observed before and/or after each mask .
we obtained multislit observations for seven clusters and present new photometry and analysis of our previously - published spectra of the six galaxies in abell 2104 @xcite .
the spectroscopic observations were processed with a series of custom python scriptskelson ] . after a bias frame
was subtracted from each raw data frame , the spatial and spectral distortions of each image were calculated from spectroscopic observations of flat field and henear comparison lamp observations , respectively , following the techniques described in @xcite .
the comparison lamps were then used to calculate a fifth - order wavelength solution for each frame .
the zeropoint of the wavelength solution was then recalculated from prominent night sky lines in the red region of the spectrum .
the flatfield observations were used to correct the frames and the flats themselves for variable slit illumination and normalized flatfields were used to removing the effects of fringing from the spectra . a model for the night sky emission was then fit to each frame following the method described in @xcite .
after this sky model was subtracted from each frame , the centroids and spatial extent of the sources were tabulated for all of the observations of a given mask .
one - dimensional spectra for each object were extracted from all of the observations of a given mask simultaneously using the b - spline algorithm described in @xcite .
this method accounted for the different distortions and dispersion solutions in each frame and generated linear output spectra with a dispersion of 5 pix@xmath3 .
the spectra extend from 4000 to 9000 , except for some that fall near an edge of the mask .
the final output spectrum for each source includes the object counts , the sky counts , and the signal - to - noise per pixel .
redshifts for all of the sources were calculated using a cross - correlation technique in pixel space , rather than fourier space , which has the advantage that each pixel may be weighted by the inverse of the variance , rather than assigned equal weight .
this technique is more fully described in @xcite .
a number of our spectroscopic targets , both with and without x - ray emission , have previously published redshifts , including abell 3125 and abell 3128 @xcite , ms1008 @xcite , and ac 114 @xcite .
these literature data were used to cross - check our redshift measurements as well as to determine the redshift range of likely cluster members . for clusters without published velocity dispersions , we conservatively estimated membership if the galaxies were within 2000 km s@xmath3of the cluster mean velocity .
only one source ( a1689 - 2 ) has a substantial offset of slightly over 4000 km s@xmath3 from the cluster mean .
this source is included in the cluster sample , although it is flagged with a @xmath38 in table [ tbl : catalog ] .
we have also flagged ( with a @xmath37 ) seven sources whose visible - wavelength centroids are only marginally within the x - ray error circles .
these sources could be either off - nuclear x - ray sources or chance superpositions of x - ray sources with a cluster member .
the probability of a chance superposition of a cluster member with a background ( or foreground ) x - ray source depends strongly on the sensitivity of the _ chandra _ dataset . from the x - ray log @xmath48
log @xmath72 relation , we expect approximately 1000 x - ray sources per square degree , or about one every three square arcminutes @xcite .
this is comparable to the average surface density of cluster members to our absolute magnitude limit and therefore chance superpositions must be extremely rare .
a second possibility is that the x - ray sources with large @xmath37 are associated with the cluster galaxies , but they are ultraluminous x - ray sources rather than nuclear emission . in this case
their high luminosity ( @xmath33 erg s@xmath3 ) makes this conclusion unlikely because known ultraluminous x - ray sources rarely have @xmath73 erg s@xmath3 @xcite .
the redshifts for the counterparts to all of the cluster x - ray sources are provided in table [ tbl : catalog ] .
the star formation properties of this sample can be studied with measurements of the [ ] and h@xmath74 equivalent widths , where we have tabulated these values for our sample in table [ tbl : catalog ] using the bandpass definitions from @xcite .
we detect [ ] emission in five galaxies and h@xmath74 absorption in eight galaxies .
detection limits for the remaining sources vary , but typically are less than 2 .
table [ tbl : catalog ] also notes any other prominent emission lines or an agn classification if the emission - line ratios suggest an agn based on the standard diagnostic diagram of @xcite .
spectra for these sources are shown in figure [ fig : spectra ] , with the exception of the six sources in abell 2104 published in @xcite .
table [ tbl : lum ] lists the absolute @xmath69band magnitude @xmath75 for each galaxy .
corrections for bandpass shifting and stellar evolution have been applied based on a simple stellar population model with solar metallicity and formation redshift of @xmath76 @xcite .
this model was also used to correct the @xmath77 colors to the rest - frame values shown in the relevant figures .
we estimated the spectroscopic completeness for these observations by calculating the fraction of spectroscopic candidates with redshifts as a function of @xmath69band magnitude .
the 95% completeness varies between @xmath78 mag and increases for higher - redshift clusters . for all of the clusters our redshift catalog is complete for all x - ray counterparts with rest - frame @xmath9 mag .
lccccccc a3125 - 1 & -21.19 & 2.6e-14 ( 7.0e-15 ) & 2.6e-14 ( 7.1e-15 ) & ... & 2.3e+41 & 2.3e+41 & ... + a3125 - 2 & -22.02 & 2.1e-14 ( 7.6e-15 ) & 2.3e-14 ( 7.6e-15 ) & ... & 1.9e+41 & 2.0e+41 & ... + a3125 - 3 & -21.86 & 9.5e-15 ( 5.1e-15 ) & 1.1e-14 ( 5.3e-15 ) & ... & 8.2e+40 & 9.6e+40 & ... + a3125 - 4 & -21.86 & 9.7e-15 ( 5.0e-15 ) & 7.9e-15 ( 4.8e-15 ) & ... & 8.4e+40 & 6.9e+40 & ... + a3125 - 5 & -21.04 & 3.4e-14 ( 8.3e-15 ) & 1.6e-14 ( 6.4e-15 ) & 9.4e-15 ( 3.3e-15 ) & 2.9e+41 & 1.4e+41 & 8.2e+40 + a3125 - 6 & ... & 2.4e-14 ( 9.3e-15 ) & 2.3e-14 ( 8.8e-15 ) & ... & 2.1e+41 & 2.0e+41 & ... + a3128 - 1 & -21.65 & 1.2e-14 ( 3.4e-15 ) & 1.1e-14 ( 3.5e-15 ) & ... & 1.0e+41 & 9.9e+40 & ... + a3128 - 2 & -19.96 & 1.1e-13 ( 9.1e-15 ) & 8.3e-14 ( 8.2e-15 ) & 1.7e-14 ( 2.7e-15 ) & 9.9e+41 & 7.5e+41 & 1.6e+41 + a3128 - 3 & ... & 1.0e-14 ( 4.1e-15 ) & 8.0e-15 ( 3.7e-15 ) & 1.4e-15 ( 1.4e-15 ) & 8.9e+40 & 7.2e+40 & 1.2e+40 + a3128 - 4 & -22.12 & 6.0e-14 ( 6.8e-15 ) & 6.2e-14 ( 7.0e-15 ) & ... & 5.4e+41 & 5.6e+41 & ... + a3128 - 5 & ... & 9.5e-15 ( 4.1e-15 ) & 7.0e-15 ( 3.6e-15 ) & 1.4e-15 ( 1.4e-15 ) & 8.5e+40 & 6.3e+40 & 1.2e+40 + a3128 - 6 & -20.20 & 1.2e-13 ( 8.4e-15 ) & 9.5e-14 ( 7.8e-15 ) & 1.1e-14 ( 2.1e-15 ) & 1.0e+42 & 8.5e+41 & 9.9e+40 + a3128 - 7 & -22.01 & 1.9e-14 ( 4.2e-15 ) & 2.0e-14 ( 4.2e-15 ) & ... & 1.7e+41 & 1.8e+41 & ... + a3128 - 8 & -21.85 & 1.5e-14 ( 4.7e-15 ) & 1.2e-14 ( 4.3e-15 ) & 1.9e-15 ( 1.3e-15 ) & 1.4e+41 & 1.0e+41 & 1.7e+40 + a3128 - 9 & -20.66 & 1.5e-14 ( 3.9e-15 ) & 1.2e-14 ( 3.5e-15 ) & 1.9e-15 ( 1.2e-15 ) & 1.4e+41 & 1.0e+41 & 1.7e+40 + a3128 - 10 & -22.22 & 2.5e-14 ( 5.0e-15 ) & 2.6e-14 ( 5.1e-15 ) & ... & 2.3e+41 & 2.3e+41 & ... + a644 - 1 & -20.61 & 1.0e-13 ( 6.9e-15 ) & 6.4e-14 ( 5.4e-15 ) & 1.7e-14 ( 2.1e-15 ) & 1.3e+42 & 8.0e+41 & 2.1e+41 + a644 - 2 & -21.58 & 1.3e-14 ( 3.4e-15 ) & 2.6e-15 ( 2.1e-15 ) & 5.2e-15 ( 1.5e-15 ) & 1.7e+41 & 3.3e+40 & 6.5e+40 + a2104 - 1 & -22.06 & 5.5e-13 ( 9.1e-15 ) & 1.1e-13 ( 2.6e-15 ) & 1.7e-13 ( 4.0e-15 ) & 4.0e+43 & 7.9e+42 & 1.2e+43 + a2104 - 2 & -20.71 & 1.1e-14 ( 1.4e-15 ) & 1.3e-15 ( 3.6e-16 ) & 4.9e-15 ( 7.4e-16 ) & 7.9e+41 & 9.4e+40 & 3.5e+41 + a2104 - 3 & -22.20 & 4.0e-15 ( 9.9e-16 ) & 1.7e-15 ( 4.0e-16 ) & ... & 2.9e+41 & 1.2e+41 & ... + a2104 - 4 & -19.76 & 4.0e-15 ( 1.2e-15 ) & 1.8e-15 ( 4.9e-16 ) & ... & 2.9e+41 & 1.3e+41 & ... + a2104 - 5 & -19.80 & 2.3e-14 ( 2.0e-15 ) & 7.9e-15 ( 7.5e-16 ) & 2.5e-15 ( 5.6e-16 ) & 1.7e+42 & 5.7e+41 & 1.8e+41 + a2104 - 6 & -21.59 & 4.0e-15 ( 1.0e-15 ) & 7.0e-16 ( 3.2e-16 ) & 1.4e-15 ( 4.6e-16 ) & 2.9e+41 & 5.1e+40 & 1.0e+41 + a1689 - 1 & -22.00 & 1.2e-14 ( 5.3e-15 ) & 1.3e-14 ( 5.8e-15 ) & ... & 1.2e+42 & 1.3e+42 & ... + a1689 - 2 & -21.41 & 1.8e-14 ( 6.2e-15 ) & 1.4e-14 ( 6.1e-15 ) & 2.5e-15 ( 2.0e-15 ) & 1.9e+42 & 1.5e+42 & 2.7e+41 + a2163 - 1 & -22.24 & 2.8e-14 ( 2.5e-15 ) & 1.8e-15 ( 9.1e-16 ) & 1.3e-14 ( 1.2e-15 ) & 3.8e+42 & 2.4e+41 & 1.7e+42 + a2163 - 2 & -19.96 & 5.6e-15 ( 1.2e-15 ) & 3.5e-15 ( 8.4e-16 ) & 8.3e-16 ( 3.7e-16 ) & 7.6e+41 & 4.8e+41 & 1.1e+41 + a2163 - 3 & -21.66 & 2.0e-15 ( 9.0e-16 ) & 1.8e-15 ( 7.1e-16 ) & ... &
2.7e+41 & 2.4e+41 & ... + ms1008 - 1 & -21.35 & 3.3e-15 ( 1.7e-15 ) & 1.7e-15 ( 1.3e-15 ) & 7.5e-16 ( 6.9e-16 ) & 1.2e+42 & 5.9e+41 & 2.6e+41 + ms1008 - 2 & -19.42 & 2.0e-15 ( 1.0e-15 ) & 1.2e-15 ( 8.1e-16 ) & ... & 7.0e+41 & 4.2e+41 & ... + ms1008 - 3 & -21.78 & 1.8e-14 ( 2.6e-15 ) & 1.2e-14 ( 2.0e-15 ) & 2.1e-15 ( 7.4e-16 ) & 6.3e+42 & 4.2e+42 & 7.3e+41 + ms1008 - 4 & -21.24 & 2.2e-15 ( 1.2e-15 ) & 1.6e-15 ( 9.6e-16 ) & ... &
7.7e+41 & 5.6e+41 & ... + ms1008 - 5 & -21.99 & 9.7e-15 ( 2.9e-15 ) & 6.6e-15 ( 2.3e-15 ) & 1.1e-15 ( 8.0e-16 ) & 3.4e+42 & 2.3e+42 & 3.8e+41 + ac114 - 1 & -21.03 & 8.1e-14 ( 2.8e-15 ) & 4.3e-14 ( 1.5e-15 ) & 1.2e-15 ( 3.6e-16 ) & 3.1e+43 & 1.7e+43 & 4.6e+41 + ac114 - 2 & -19.62 & 4.1e-15 ( 9.6e-16 ) & 1.5e-15 ( 4.6e-16 ) & 8.5e-16 ( 3.3e-16 ) & 1.6e+42 & 5.6e+41 & 3.2e+41 + ac114 - 3 & -21.78 & 1.1e-14 ( 1.1e-15 ) & 5.2e-15 ( 5.8e-16 ) & 7.8e-16 ( 2.9e-16 ) & 4.2e+42 & 1.9e+42 & 3.0e+41 + ac114 - 4 & -20.06 & 5.4e-15 ( 8.3e-16 ) & 1.1e-15 ( 3.1e-16 ) & 2.2e-15 ( 4.2e-16 ) & 2.0e+42 & 4.2e+41 & 8.4e+41 + ac114 - 5 & -20.02 & 7.7e-16 ( 5.2e-16 ) & 3.1e-16 ( 2.6e-16 ) & ... & 2.9e+41 & 1.2e+41 & ... + ac114 - 6 & -22.90 & 6.7e-16 ( 5.3e-16 ) & 3.5e-16 ( 2.7e-16 ) & ... & 2.5e+41 & 1.3e+41 & ... +
we have identified a total of 40 galaxies with x - ray emission in eight clusters , corresponding to between two and ten x - ray sources per cluster .
thirty four of these cluster members are new to this study , while six were published in our previous study of abell 2104 . of these 34 new galaxies
, we identified 29 with our spectroscopic observations and five additional sources from cross - correlating our x - ray sources with redshift data from the literature .
references to these sources are provided in the caption to table [ tbl : catalog ] .
as noted above , several of these sources were previously known to be cluster members , although only the @xcite study of a3128 previously studied the x - ray emission .
he noted that our sources a3128 - 2 and a3128 - 6 had hard x - ray spectra and may be agn , while of the three other galaxies a3128 - 7 and a3128 - 10 appeared to be normal e / sos and a3128 - 6 was an e / so with some shell structure . in the next sections below
we describe the photometric and spectroscopic properties of this sample of cluster galaxies with x - ray emission .
we then use these results to identify the likely physical mechanisms responsible for the x - ray emission and calculate the fraction of cluster galaxies that host agn .
the @xmath69band images of these galaxies ( figure [ fig : charts ] ) show that most appear to be early - type galaxies , consistent with the morphological type most commonly found at the centers of rich clusters .
this is particularly true of the lowest - redshift ( @xmath66 ) clusters , where all galaxies except for a3125 - 5 appear to be es or s0s .
six of the 40 galaxies show obvious signs of interactions .
the left panel of figure [ fig : brrmr ] presents a color - magnitude diagram of @xmath77 vs. @xmath79 and demonstrates that the x - ray counterparts with emission lines , particularly those with obvious agn spectra signatures , tend to be bluer than galaxies that do not exhibit bright emission lines .
this result is more pronounced when all of the lower - redshift ( @xmath66 ; large symbols ) galaxies are compared separately from the higher - redshift ( @xmath80 ) galaxies .
two of the four galaxies with agn spectral signatures are the bluest galaxies in the sample .
sources with other strong emission lines , such as h@xmath36 , [ ] , or [ ] also tend to be substantially bluer than the other galaxies . in the lower - redshift subsample
the emission - line galaxies are among the fainter sources , while in the higher - redshift subsample the obvious agn and other emission line galaxies tend to be in the brighter half of the distribution .
this is shown in right panel of figure [ fig : brrmr ] , which plots @xmath77 vs. @xmath75 .
this figure demonstrates that the emission - line galaxies do not tend to reside in the most luminous host galaxies , but rather in galaxies only slightly more luminous than the knee of the galaxy luminosity function @xmath81 mag @xcite .
we use the [ ] and h@xmath74 equivalent widths as estimators of the star formation history of these x - ray sources , where the [ ] equivalent width is a measure of current star formation and h@xmath74 is a measure of the remnant population of a significant starburst in the recent past .
figure [ fig : o2hd ] illustrates the location of our sample galaxies in this parameter space , along with the spectral classification scheme developed by @xcite .
only three of these galaxies are classified as e(c ) and have [ ] widths greater than 5 and moderate balmer absorption .
two galaxies have no detected [ ] , but h@xmath74 greater than 3 indicative of moderate to strong balmer absorption .
the most extreme case is ac114 - 1 , whose spectrum exhibits a strong balmer series and is a borderline poststarburst galaxy .
most of the galaxies in this sample show no strong [ ] or h@xmath74 .
x - ray colors offer a simple means of quantifying the properties of x - ray sources .
as the x - ray colors are calculated directly from count rates , the correspondence between colors and spectral properties , such as power - law shape , depend on the energy sensitivity of a given detector .
these differences are particularly acute between the front - illuminated ( fi ) acis - i ( ccds 0 through 3 ) and the back - illuminated ( bi ) acis - s ( ccds 5 and 7 ) and consequently we have calculated the conversion between x - ray colors and spectral properties for the acis - i and acis - s observations separately .
the primary cause of this difference is that the bi chips are more sensitive to very soft x - ray photons ( @xmath82 kev ) and therefore a given source will , for example , register a larger @xmath45 count rate if observed with a bi chip relative to an fi chip .
figure [ fig : xccd ] displays the x - ray colors c32 vs. c21 for the fi and bi chips , along with grids that demonstrate the location of an absorbed power - law with @xmath83 and @xmath84 @xmath85 .
the different shapes of the grids in the two panels reflects the substantial differences in the energy sensitivity of the fi and bi ccds .
the relative energy sensitivity of the chips change slightly with time and the grids shown are based on the average sensitivity of these observations .
sources are only shown if they have at least three counts in the @xmath45 , @xmath46 , and @xmath86 bands ( see table [ tbl : xdata ] ) .
nearly all of the sources shown in the figure are consistent with @xmath87 and at most modest ( @xmath88 @xmath85 ) absorption .
those that do not include the spectroscopically - identified agn ( _ filled circles _ ) and indicate that an absorbed power - law is too simplistic a model for some of these sources .
an absorbed power - law combined with emission from hot gas may be a better fit to these cases , although we lack sufficient signal - to - noise for detailed spectral modeling .
in a recent paper , @xcite argued that this traditional approach of using x - ray colors in predetermined bands is strongly biased toward particular values of @xmath59 and @xmath89 ( for an assumed power - law model ) .
these authors show that the bands used to define the colors used in figure [ fig : xccd ] can classify sources with @xmath87 and modest absorption with only a few counts , while there will be substantial uncertainties in the classification of substantially softer , harder , and/or more obscured sources .
effectively , a source with higher or lower @xmath59 must be detected with substantially higher counts in order to fall in much of the traditional x - ray color space .
the common adoption of an unobscured @xmath58 power - law for typical x - ray sources may therefore be the result of selection effects due to the band definitions , rather than a reflection of the ubiquity and appropriateness of this model for faint x - ray sources . in figure
[ fig : xq ] we plot the x - ray quantiles defined in section [ sec : xdata ] along with the same grids of @xmath59 and @xmath89 shown in figure [ fig : xccd ] ( the approximate axes of @xmath59 and @xmath89 are rotated by @xmath90 degrees ) . as in figure
[ fig : xccd ] , most of the sources are concentrated at low @xmath89 and @xmath87 , although there are many more very soft sources ( large @xmath59 ) and several more hard sources than is apparent from the x - ray colors . as in the previous figure ,
the most significant outliers include the spectroscopically - confirmed agn .
this confirms our conclusion based on figure [ fig : xccd ] that a simple , absorbed power - law model is not a good fit to even the obvious low - luminosity agn , although more counts would be required to explore a wider parameter space of models .
taken separately , the previous two sections present an apparent contradiction : from the visible - wavelength data alone , it appears that only four of these 40 x - ray sources are agn and the remainder are galaxies with modest star formation or passively - evolving galaxies .
in contrast , the x - ray data indicate that most of these sources are quite luminous in x - rays and many are consistent with the unobscured , @xmath91 power laws typical of agn .
furthermore , the spectroscopically - identified agn are those least consistent with unobscured , @xmath91 power law emission . here
we investigate the visible to x - ray properties of these sources in order to resolve this contradiction .
if the x - ray emission from these galaxies is predominantly due to agn and the x - ray to visible - wavelength flux ratio is constant for agn , then the galaxies with agn spectral signatures and bluer colors should be among the most x - ray luminous sources .
we test this hypothesis in figure [ fig : brlxb ] with a plot of @xmath77 vs. broad band x - ray luminosity .
this figure shows that only one spectroscopically identifiable agn is among the most x - ray luminous galaxies .
most of the x - ray sources with @xmath15 erg s@xmath3 do not show agn spectral signatures , although we note that all of these galaxies are above @xmath92 and such signatures are more easily diluted by host galaxy starlight .
these galaxies are generally not significantly bluer than other galaxies in these clusters , although some would be classified as butcher - oemler galaxies @xcite .
the visible x - ray flux ratio can be another indicator of the presence of an agn , as opposed to lmxbs or hot gas , because agn have higher @xmath93 compared to other galaxies . in figure
[ fig : brmrfxfr ] we plot @xmath77 and @xmath75 vs. the soft x - ray to @xmath69band flux ratio .
this figure is similar to figure 8 of @xcite , which those authors use to separate various types of x - ray emitting galaxies in the champ survey ( see also * ? ? ?
the region to the right of the dashed vertical line at @xmath94 and bluer than @xmath95 is their locus for blue quasars .
sources with redder @xmath77 colors still have excess x - ray emission over that expected for hot x - ray halos or lmxbs and may therefore be obscured agn .
the very red a2104 - 1 is an excellent example of an obscured agn with red visible - wavelength colors and excess x - ray emission .
the other extreme source in this figure ( at @xmath96 , @xmath97 ) is the poststarburst galaxy ac114 - 1 , which has an extremely soft x - ray spectrum .
less than a third of our sample is located in their quasar regime defined by @xmath94 and @xmath98 in figure [ fig : brmrfxfr ] , although this is not surprising as none of these x - ray sources have quasar luminosities ( @xmath99 erg s@xmath3 ) and therefore the @xmath69band flux could include a contribution from stars .
in fact nearly all of our spectra are dominated by host galaxy starlight .
we investigate the @xmath100 ratio as a function of @xmath75 in the lower panel of figure [ fig : brmrfxfr ] in order to determine the typical host galaxy luminosity of sources with large @xmath100 .
this figure demonstrates that most of the sources with large @xmath100 are in approximately @xmath101 or fainter galaxies and are above @xmath102 .
the absence of low - luminosity galaxies with small @xmath100 is due to the sensitivity limit of the x - ray observations . while our spectroscopic observations are complete to @xmath103 mag , the x - ray observations are only sensitive to galaxies with small @xmath100 in these clusters if they have luminous host galaxies . in contrast
, the increasing number of luminous galaxies with lower @xmath100 is significant .
for luminous galaxies this illustrates that they are more likely to host a faint x - ray source than a bright one . unlike the case for low - luminosity agn , there is a relatively tight relation between x - ray and visible - wavelength flux for galaxies with x - ray emission due to lmxbs @xcite .
this relation has been studied in detail with _
chandra _
observations of nearby elliptical galaxies ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
in a recent paper @xcite derived the relation @xmath104 erg s@xmath3@xmath105 based on broad - band x - ray measurements of 14 luminous , nearby galaxies .
figure [ fig : lxlb ] shows the broad band @xmath106 vs. @xmath107 for our sample , along with the galaxies measured by @xcite and the band that defines the @xmath108 scatter in the @xmath93 ratio for lmxbs .
this figure shows that all of these galaxies are more x - ray luminous than would be expected from lmxb emission alone , although several galaxies are marginally consistent with the lmxb relation .
x - ray emission from the hot interstellar medium in early - type galaxies may also make a significant contribution to the total x - ray luminosity .
the x - ray spectra of hot gas emission should be substantially softer than agn emission , however most of the x - ray sources lack sufficient counts to use this as a discriminant .
instead , we compare these sources to the @xmath109 relation observed for local early - type galaxies @xcite . figure [ fig : lxlb2 ] shows soft x - ray luminosity vs. @xmath110 for our sample along with the literature relation from @xcite and several measurements of early - type galaxies in abell 1367 with _ chandra _ from @xcite .
the measurements of early - type in abell 1367 galaxies are a particularly relevant comparison sample because the @xmath93 ratio is generally larger for galaxies in rich environments , provided the environments are not so rich and the galaxies have such high relative velocities that ram pressure stripping is significant @xcite .
all of our sources have at least seven counts in the broad ( 0.3 - 8 kev ) x - ray band .
seven counts correspond to broad band x - ray luminosities ranging from @xmath62 erg s@xmath3(a644 ) to @xmath64 erg s@xmath3(a1689 ) for an unobscured @xmath58 power law and galactic absorption .
when we combine the x - ray sensitivity with our spectroscopic completeness , we can conservatively state that we are sensitive to all x - ray sources more luminous than @xmath111 erg s@xmath3and host galaxies with @xmath9 mag in these eight clusters , although we are sensitive to substantially less x - ray luminous galaxies in most of the clusters .
only four x - ray sources in our sample ( a644 - 1 , a2104 - 1 , a1689 - 2 , a2163 - 1 ) are both x - ray luminous @xmath15 erg s@xmath3 and exhibit the emission - line signatures of agn .
ten additional galaxies are above @xmath111 erg s@xmath3 and therefore are most likely agn , but at progressively fainter x - ray luminosities an increasing number of physical mechanisms can produce luminous x - ray emission , particularly a population of lmxbs , thermal emission from a hot , gaseous halo , and star formation . of these three mechanisms ,
we rule out star formation for these cluster sources because the observed h@xmath36 or [ ] luminosities indicate low star formation rates that could not produce the observed x - ray luminosities ( e.g. * ? ? ?
* ; * ? ? ?
while the less luminous sources lack sufficient counts for detailed spectral fitting , most with at least moderate signal - to - noise ratio x - ray detections are consistent with a @xmath58 power law ( see figures [ fig : xccd ] and [ fig : xq ] ) .
the sources with softer x - ray spectra could be due to thermal bremsstrahlung or may be compton - thick agn .
our strongest arguments against emission from lmxbs or hot , gaseous halos are the ratios of the x - ray to b - band luminosities of the galaxies . from figures [ fig : lxlb ] and [ fig :
lxlb2 ] , all but three sources are more x - ray luminous than either the @xmath109 relation for lmxbs or hot halos ( two of the three are consistent with both relations ) , or 34 of 37 sources with visible - wavelength photometry . as noted above in section [
sec : xdata ] , all of these sources are unresolved , which strengthens the argument against an lmxb or hot halo origin for the lower - redshift clusters .
there are three additional cluster x - ray sources that are not in the figures because we do not have b - band observations .
two of the three were detected in the hard x - ray band and one with sufficient counts to determine that it is consistent with an agn .
if we conservatively assume the other two sources are not agn , along with the three mentioned previously that are statistically consistent with the lmxb or hot gas @xmath109 relation , we conclude that at least 35 of these 40 x - ray sources are agn .
we calculate the fraction @xmath4 of cluster galaxies that host agn by dividing the number of cluster agn @xmath112 by an estimate of the total number of cluster members @xmath113 to our absolute - magnitude spectroscopic limit @xmath75 within the field of the _ chandra _ observation . along with our spectroscopic survey of x - ray sources
, we also obtained spectra of many candidate cluster members to measure the cluster velocity dispersion and more generally characterize the cluster galaxy population .
we first calculate the fraction @xmath114 of all ( non x - ray ) spectroscopic targets to some magnitude limit that are cluster members
. then the total number of cluster galaxies to this magnitude limit is @xmath115 , the magnitude limit corresponds to the same @xmath75 for all clusters , and @xmath116 is the total galaxy population to this magnitude limit .
the cluster agn fraction is then @xmath117 .
we estimated the total number of galaxies in each cluster to a uniform absolute magnitude of @xmath6 mag , which was set by the spectroscopic completeness of our observations of the more distant clusters .
our success rate at identifying cluster members varied from 19 to 44% based on spectroscopic redshifts for between 12 and 54 cluster members not targeted as x - ray sources .
these calculations yield an x - ray source fraction of @xmath118% for the seven new clusters and confirms our previous result based on just one cluster @xcite .
the six cluster members fainter than @xmath6 mag in table [ tbl : lum ] are not included in this calculation .
the agn fraction is more meaningfully compared to other surveys with a constraint on agn luminosity in addition to galaxy stellar luminosity .
we use the x - ray sensitivity of our observations to constrain the agn fraction to be the fraction of agn above a broad - band x - ray luminosity of @xmath7 erg s@xmath3 .
nearly all of these x - ray observations would detect an x - ray source more luminous than this and all but four of our 40 x - ray sources are above this luminosity .
the four sources eliminated by this luminosity cut include all but one of those least likely to be agn based on x - ray to visible - wavelength flux ratio .
we find that the fraction of galaxies brighter than @xmath6 mag that host agn more luminous than @xmath7 erg s@xmath3 is @xmath119 mag@xmath120erg s@xmath3@xmath121% .
the quoted uncertainty corresponds to 90% one - sided confidence limits calculated from poisson statistics @xcite .
we may have underestimated the true agn fraction because our x - ray sensitivity is a factor of five above @xmath7 erg s@xmath3 for two clusters ( a1689 and ms1008 ) and therefore these clusters may contain additional agn below our x - ray limit and more luminous than @xmath17 erg s@xmath3 .
the advantage of this lower luminosity limit is that it maximizes the number of x - ray sources in cluster members and the statistical significance of the detection .
while there are only a small number of agn per cluster , we do find evidence for substantial variation in the agn fraction from cluster to cluster , most notably a substantially higher agn fraction in abell 3128 relative to abell 644 , even though the abell 644 x - ray observations are more sensitive .
we plan to explore this aspect of our data in a future paper ( martini et al . 2006 , _ in prep _
) that will include a more detailed completeness study .
we note uncertainties in the present completeness calculation may introduce systematic errors in the agn fraction that are larger than the statistical errors .
the agn fraction we find in clusters is approximately a factor of five times higher than the visible - wavelength spectroscopic survey of @xcite , who found a @xmath122% agn fraction ( see also * ? ? ?
. however , when we just calculate the cluster agn fraction from those identified in visible - wavelength spectroscopy , our results are completely consistent with the @xcite result .
this has some interesting implications for the field agn fraction , which these authors found was approximately five times higher than the cluster fraction .
if the properties of cluster and field agn are similar , then there are simply five times fewer agn in clusters .
a survey with similar selection criteria to this one should then identify a field agn fraction of @xmath123% .
we can also use these data to address the origin of the apparent excess of x - ray sources observed toward the fields of other clusters observed by _
chandra_. the excesses take the form of overdensities in the log n
log s relation and have been reported in the soft @xcite and hard x - ray bands ( e.g. * ? ? ?
* ) , although not for all cluster fields @xcite .
our redshift measurements allow us to directly address whether or not the x - ray sources in these eight clusters add appreciably to the x - ray surface density .
we address this question on a cluster by cluster basis by measuring the flux - limited fraction of sources that are members of a given cluster . for each of our fields
, we have counted the number of sources brighter than some limiting soft x - ray flux and calculated the fraction of these sources that are cluster members .
this effectively integrates log n log s for the field to the flux limit of the observation , where we have adopted a flux limit corresponding to five counts in the soft x - ray band at the aimpoint and adopted a model with @xmath58 and only galactic absorption .
the five count soft - band threshold was chosen because it is comparable in signal - to - noise ratio to our seven count broad - band threshold .
this threshold includes all but one of our cluster sources ( ms1008 - 1 ) .
we find that the cluster x - ray sources contribute up to 20% of the soft x - ray point source density in these fields .
for each field we use poisson statistics @xcite to calculate the one - sided confidence limits from the number of nonmembers . for three of these eight clusters ,
the point sources due to cluster members constitute a @xmath124 excess , and are slightly more significant for a fourth ( a3128 ) .
these marginally significant excesses are similar to the results toward cluster fields without membership information , particularly the @xcite study that identified an @xmath125 excess surface density from a sample of 51 clusters .
two differences between our data and their macs sample are that the later data have a shallower luminosity threshold and encompass a larger fraction of the total cluster size because the macs clusters tend to be at higher redshift .
the absence of an overdensity of x - ray sources in the @xcite study , in contrast , may be due to their preference for clusters whose intracluster medium subtends only a small fraction ( less than 10% ) of the field @xcite with the result that their sample of 29 cluster fields is dominated by high - redshift clusters .
only the most luminous cluster members could therefore be detected in these images .
in addition , as @xcite note , their cluster fields were mostly obtained with acis - i and any cluster signal may also be diluted by field contamination due to the large field of view .
the redshift range and various sensitivity limits of these eight clusters can be used to explore the importance of these quantities in detecting x - ray source excesses in other cluster fields . in the top panels of figure [ fig : logns ]
we show that there are no obvious trends with either flux limit or redshift , in spite of the approximately factor of ten variation in sensitivity
. this lack of variation may be due in part to the fact that the more distant cluster observations were chosen to be more sensitive and therefore the observations have a narrower range in luminosity sensitivity than flux sensitivity , although this could be offset by higher field contamination at fainter flux limits .
indeed three of the four clusters with marginally - significant excesses , which correspond to those with greater than 10% cluster contributions , are at @xmath66 .
we note that we are neglecting the fact that the flux limit will rise with increasing off - axis angle because both the cluster and field x - ray sources will experience the same effect , although could nevertheless be a bias toward a higher cluster fraction if the cluster sources are more centrally concentrated within the acis field of view .
the radial distribution of cluster sources is also important because the fixed angular size of the _ chandra _ observations subtends a substantially smaller fraction of a given cluster at lower - redshift .
for example , the approximately @xmath126 wide acis - i field of view corresponds to only 1.2 mpc at @xmath13 , but over 4.5 mpc at @xmath127 .
we have investigated the importance of aperture size by calculating the fraction of cluster x - ray sources within a fixed physical radius of 0.5 mpc for all eight clusters .
this physical aperture size is the largest that is contained within the _ chandra _ images of all eight clusters , particularly the two lowest - redshift clusters , and also happens to be the physical radius within which @xcite observed the most significant overdensity of x - ray sources . within this aperture size
we now measure fractions as high as @xmath128% for one cluster ( ac114 ) , although we still measure @xmath129% or less for most ( six clusters ; all but ms1008 and ac114 ) the fractional contribution of sources within 0.5 mpc as a function of limiting soft x - ray flux and redshift are shown in the bottom panels of figure [ fig : logns ] . if we now repeat the same experiment described above to determine if the cluster members make a statistically - significant contribution to the surface density within an 0.5 mpc aperture , the cluster x - ray sources in a3128 and ac114 are @xmath130 excesses over the number of other x - ray sources , while ms1008 sources make only a marginal contribution .
we note that the @xmath125 excess noted by @xcite in their sample of 51 clusters is consistent with a @xmath124 excess per cluster summed in over their sample .
while the fraction of cluster members does appear highest for the two highest - redshift clusters , the present sample of clusters is too small for us to claim the presence of redshift dependence in the cluster agn population .
we have completed a redshift survey and multiwavelength study of x - ray sources in the fields of eight clusters of galaxies and discovered luminous x - ray emission from 40 galaxies .
emission - lines , x - ray spectral properties , and x - ray to visible - wavelength flux ratios were used to determine that the vast majority of these sources are low - luminosity agn .
we conservatively estimate that at least 35 of these 40 sources are agn in the clusters and estimate that the agn fraction in clusters is @xmath8% , that is @xmath5% of all cluster galaxies more luminous than @xmath6 mag host agn with broad - band x - ray luminosities greater than @xmath7 erg s@xmath3 .
only a small fraction of these galaxies would be classified as agn from their visible - wavelength spectra . in most instances
the agn identification is due to x - ray spectral shape or x - ray to visible - wavelength flux ratios . in particular , the multiwavelength properties of these sources are used to show that they are more x - ray luminous for their observed b - band luminosity than would be expected for either a population of lmxbs or hot , gaseous halos that have survived in the intracluster medium .
the x - ray sources in these clusters do not make a statistically significant contribution to the surface density in any one of these _ chandra _ observations , although summation of a sufficient number of similar clusters would produce a statistically significant excess similar to those observed in stacked _ chandra _ observations of other cluster fields .
we therefore conclude there is a sufficiently large population of x - ray sources in clusters of galaxies to explain the observed point source excess .
this population of cluster agn is a factor of five higher than expected from previous spectroscopic surveys .
we attribute our higher success rate to the fact that optical searches for agn are strongly biased against detection of low - luminosity agn because of host galaxy dilution and obscuration .
agn with yet lower x - ray luminosities are likely present in other luminous cluster galaxies , or equivalently the agn fraction should increase with a lower @xmath106 threshold , although they may prove difficult to discriminate from emission due to lmxbs or hot gas in the absence of substantially higher signal - to - noise ratio data .
our result also has interesting implications for a higher field agn fraction .
if the agn fraction in the field is five times the cluster agn fraction , as found by dressler and collaborators for uniformly - selected samples of field and cluster galaxies , then similar selection criteria to this study will identify a field agn fraction of @xmath123% .
the substantial number of cluster galaxies hosting agn suggests that cluster galaxies can retain significant reservoirs of cold gas near their central , supermassive black holes .
these agn , and others currently dormant , may have significantly contributed to heating the intracluster medium and driving galaxy evolution via feedback processes at higher redshift .
future observations of agn and measurements of the agn fraction in higher - redshift clusters may provide valuable new insights into cluster assembly and galaxy evolution in clusters .
pm was supported by a carnegie starr fellowship and a cfa clay fellowship .
support for this work was provided by the national aeronautics and space administration through chandra award numbers 04700793 and 05700786 issued by the chandra x - ray observatory center , which is operated by the smithsonian astrophysical observatory for and on behalf of the national aeronautics space administration under contract nas8 - 03060 .
we greatly appreciate the excellent staffs of the las campanas observatory and the magellan project telescopes for their assistance with these observations . we thank tom aldcroft for help with x - ray analysis software during an earlier stage of this project and mark whittle and the referee for helpful suggestions and comments .
the canadian astronomy data centre , which is operated by the dominion astrophysical observatory for the national research council of canada s herzberg institute of astrophysics , was an extremely helpful source for our photometric calibration .
this research has made use of the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
the digitized sky surveys were produced at the space telescope science institute under u.s .
government grant nag w-2166 .
the images of these surveys are based on photographic data obtained using the oschin schmidt telescope on palomar mountain and the uk schmidt telescope .
the plates were processed into the present compressed digital form with the permission of these institutions . , m. , mazzotta , p. , elvis , m. , burke , d. j. , comastri , a. , fiore , f. , forman , w. , fruscione , a. , green , p. , harris , d. , hooper , e. j. , jones , c. , kaastra , j. s. , kellogg , e. , murray , s. , mcnamara , b. , nicastro , f. , ponman , t. j. , schlegel , e. m. , siemiginowska , a. , tananbaum , h. , vikhlinin , a. , virani , s. , & wilkes , b. 2001 , , 548 , 624 , p. j. , silverman , j. d. , cameron , r. a. , kim , d .- w . , wilkes , b. j. , barkhouse , w. a. , lacluyz ' e , a. , morris , d. , mossman , a. , ghosh , h. , grimes , j. p. , jannuzi , b. t. , tananbaum , h. , aldcroft , t. l. , baldwin , j. a. , chaffee , f. h. , dey , a. , dosaj , a. , evans , n. r. , fan , x. , foltz , c. , gaetz , t. , hooper , e. j. , kashyap , v. l. , mathur , s. , mcgarry , m. b. , romero - colmenero , e. , smith , m. g. , smith , p. s. , smith , r. c. , torres , g. , vikhlinin , a. , & wik , d. r. 2004 , , 150 , 43 , a. e. , brandt , w. n. , garmire , g. p. , schneider , d. p. , barger , a. j. , broos , p. s. , cowie , l. l. , townsley , l. k. , bautz , m. w. , burrows , d. n. , chartas , g. , feigelson , e. d. , griffiths , r. e. , lumb , d. , nousek , j. a. , ramsey , l. w. , & sargent , w. l. w. 2001 , , 554 , 742 , d .-
w . , cameron , r. a. , drake , j. j. , evans , n. r. , freeman , p. , gaetz , t. j. , ghosh , h. , green , p. j. , harnden , f. r. , karovska , m. , kashyap , v. , maksym , p. w. , ratzlaff , p. w. , schlegel , e. m. , silverman , j. d. , tananbaum , h. d. , vikhlinin , a. a. , wilkes , b. j. , & grimes , j. p. 2004
, , 150 , 19 , d .- w . ,
wilkes , b. j. , green , p. j. , cameron , r. a. , drake , j. j. , evans , n. r. , freeman , p. , gaetz , t. j. , ghosh , h. , harnden , f. r. , karovska , m. , kashyap , v. , maksym , p. w. , ratzlaff , p. w. , schlegel , e. m. , silverman , j. d. , tananbaum , h. d. , & vikhlinin , a. a. 2004 , , 600 , 59 | we have completed a spectroscopic survey of x - ray point sources in eight low - redshift clusters of galaxies ( @xmath0 ) and have identified 40 cluster members with broad - band ( 0.38 kev ) x - ray luminosities between @xmath1 and @xmath2 erg s@xmath3 .
there are between two and ten x - ray sources per cluster .
we use visible - wavelength emission lines , x - ray spectral shapes , and multiwavelength flux ratios to determine that at least 35 of these galaxies are active galactic nuclei ( agn ) . from our spectroscopic survey of other candidate cluster members
we estimate that the agn fraction @xmath4 is @xmath5% for cluster galaxies more luminous than @xmath6 mag hosting agn with broad - band x - ray luminosities above @xmath7 erg s@xmath3 , or @xmath8% . we stress that additional , lower - luminosity agn are expected to be present in the @xmath9 mag cluster members .
our data unambiguously demonstrate that cluster galaxies host agn more frequently than previously expected .
only four of these galaxies have obvious visible - wavelength agn signatures , even though their x - ray luminosities are too high for their x - ray emission to be due to populations of low - mass x - ray binaries or hot , gaseous halos .
we attribute the significant difference in visible and x - ray agn identification to dilution of low - luminosity agn spectral signatures by host galaxy starlight and/or obscuration of accretion onto the central , supermassive black hole . |
the existence of dark matter ( dm ) in the universe has motivated many experimental studies and theoretical speculations . today we know that about @xmath0 of the energy density of the universe is in form of cold dark matter but we have no idea about the origin and nature of this type of matter .
there are a lot of theoretical ideas to describe the properties of the dark matter sector , which can be as complex as the visible sector . among the very popular candidates
are the weakly interacting massive particles which appear in several extensions of the standard model ( sm ) of particle physics .
thanks to many experimental collaborations , there are relevant constraints on the properties of these candidates which play an important role in ruling out some of the theories for dark matter . for a review on different dark matter candidates and experiments
see ref .
@xcite .
we distinguish between baryonic and non - baryonic matter in the universe , and we say that the cold dark matter is non - baryonic .
this refers to the fact that it has to be different from the ordinary matter that is made of quarks ( and leptons ) . as is well known the quarks are the only particles in the context of the sm that carry baryon number , and they form protons and neutrons . in this article
we will discuss a different scenario where the dark matter carries also baryon number .
baryon number is an accidental global symmetry of the renormalizable couplings of the standard model lagrangian , but we know that it has to be broken to explain the matter
antimatter asymmetry of the universe .
recently , in ref .
@xcite , we have proposed the simplest realistic model where it is possible to have the spontaneous breaking of the baryon and lepton numbers . we will not be concerned with lepton number in this article and we will discuss only the spontaneous breaking of baryon number .
see also refs .
@xcite for earlier attempts to gauge baryon and lepton numbers . in this article
we focus on a simplified version of the model in ref .
@xcite , where there is an interesting connection between the stability of the dark matter and the spontaneous breaking of the local baryon number .
we investigate the properties of a fermionic dark matter candidate which has baryon number and refer to this type of scenario as `` baryonic dark matter . ''
we show the relic density constraints and the predictions for direct detection experiments . since this model has only four free parameters one could hope to test this idea combining the possible results from the large hadron collider and dark matter experiments .
note that we assume the dark matter to be a thermal relic in this article .
the asymmetric component of dark matter in this model has been discussed recently in ref .
@xcite .
this article is organized as follows : in section ii we discuss the theoretical framework , while in section iii we calculate the dm relic density . in sec .
iv , we show the correlation between the bounds from direct detection experiments and the relic density .
additionally , we discuss a possible test of this model by combining the efforts at the large hadron collider and dark matter experiments .
we summarize and conclude in section v.
recently , we have proposed a simple extension of the standard model where one can understand the spontaneous breaking of baryon and lepton numbers at the low scale @xcite . here
, we will discuss a simplified version of this model , only considering baryon number as a local gauge symmetry .
therefore , this model is based on the gauge group @xmath1 in order to define an anomaly - free theory using this gauge group , we need to include additional fermions that account for anomaly cancellation , @xmath2 and extend the scalar sector with a new higgs boson to allow for a spontaneous breaking of baryon number , @xmath3 here @xmath4 and @xmath5 refer to the baryon numbers of the additional fermions which are vector - like under the sm gauge group . from the conditions that ensure the cancellation of all relevant baryonic anomalies , one finds the following relation for the baryon numbers of the new fermions : @xmath6 the relevant interactions of the new fields in the theory are @xmath7 notice that one can have terms such as @xmath8 and @xmath9 only when @xmath10 . here
we will stick to the case where @xmath11 .
the yukawa interactions between the new fields and the standard model higgs boson are present as well , but they are not relevant for our main discussion .
it is important to notice that the new higgs boson must have baryon number @xmath12 in order to generate vector - like mass for the new fermions .
therefore , once @xmath13 gets a vacuum expectation value breaking local @xmath14 we never generate any operator mediating proton decay and the scale for baryon number violation can be as low as a few tev ( or even below ) . for a discussion of the bounds on the mass of a leptophobic neutral gauge boson see refs .
@xcite . in this simple theory , when the local baryon number is spontaneously broken by the vacuum expectation value @xmath15 of @xmath13 , one obtains a @xmath16 discrete symmetry which protects the stability of the dark matter candidate . under this @xmath16
the new fermionic fields transform as @xmath17 therefore , when the lightest new field with baryon number is neutral , one can have a candidate for the cold dark matter in the universe .
it is important to mention that the main idea of having a dark matter candidate with baryon number was discussed earlier in refs .
@xcite . for simplicity
, we will focus on the case when the dark matter is sm singlet - like and is the dirac fermion @xmath18 .
since the dark matter has baryon number , the relevant interactions with the new gauge boson @xmath19 related to baryon number are @xmath20 where @xmath21 and @xmath22 are the left- and right - handed projectors , and @xmath23 is the gauge coupling related to baryon number .
of course , the new gauge boson also couples to the sm quarks , which is crucial to understand the properties of the dark matter candidate .
the leptophobic gauge boson mass reads as @xmath24 while the mass of the sm singlet - like baryonic dark matter candidate is given by @xmath25 this upper limit is coming from the perturbative condition on the yukawa coupling @xmath26 , i.e. @xmath27 .
it is important to notice that this model for baryonic dark matter has only four free parameters : @xmath28 and one needs to satisfy the relic density constraints and the bounds from direct detection . here
@xmath29 is the parameter which enters in the predictions for the relevant cross sections .
one could imagine that the parameters @xmath30 and @xmath23 can be determined from the discovery of a leptophobic gauge boson at the large hadron collider .
therefore , one can say that for a given value of these two quantities we can find the values of @xmath31 and @xmath32 using the relic density and spin - independent cross section values .
we will discuss in more details the numerical predictions in the following sections in order to appreciate this connection between collider physics and dark matter experiments . in the rest of the paper we will neglect the kinetic mixing between @xmath14 and @xmath33 , as well as the mixing between the sm higgs and @xmath13 .
the dark matter particle @xmath34 can annihilate into two standard model quarks through the interaction with the leptophobic gauge boson @xmath19 .
the annihilation cross section is given by @xmath35 } \frac{c_b}{m_\chi } , \\\ ] ] with @xmath36 here @xmath37 is the mass of the quarks , @xmath38 is the square of the center - of - mass energy and @xmath39 is the decay width of the leptophobic gauge boson . in the non - relativistic limit , the above cross section reads as @xmath40 we neglected the decay width of the new gauge boson for simplicity .
note that for @xmath10 , which would allow for terms leading to majorana masses for the dm fields after symmetry breaking , @xmath41 and the annihilation cross section is velocity suppressed .
vs. the dark matter mass @xmath32 when the dm relic density is in the range @xmath42 .
here we use @xmath43 and @xmath44 tev .
[ fig : figure1 ] ] as is well known the cold dark matter relic density can be approximated by @xmath45 where @xmath46 is the freeze - out temperature and @xmath47 gev is the planck mass scale . the quantity @xmath48 can be calculated using the expression @xmath49 \nonumber \\ & - & \frac{1}{2 } \ln \left\ { \ln \left [ 0.038 \left ( \frac{g}{\sqrt{g_\ast } } \right ) m_\text{pl } m_\chi \sigma_0 \right ] \right\},\end{aligned}\ ] ] where @xmath50 is the number of internal degrees of freedom , and @xmath51 is the effective number of relativistic degrees of freedom evaluated around the freeze - out temperature @xmath48 .
the current value of the dm relic density provided by planck is @xmath52 @xcite , which we will use to understand the constraints on our model .
as we have mentioned above the relevant parameters for our study are the dark matter mass @xmath32 , the mass of the new gauge boson @xmath30 , the gauge coupling @xmath23 , and the baryon numbers of the new fermionic fields . in order to illustrate the numerical results we will choose @xmath43 for simplicity and later
discuss how one can test the model for any value of b. in fig .
[ fig : figure1 ] we show the allowed region where the dm relic density is @xmath42 in the plane spanned by the dm mass @xmath32 and the gauge coupling @xmath23 . for a range of @xmath53 , the dm mass in the range @xmath54 allows for a dm relic density around the current value .
notice that a gauge coupling @xmath23 in the interval @xmath55 $ ] and @xmath44 tev is consistent with recent collider studies @xcite .
one could use even smaller values for the gauge boson mass because the collider bounds are not so strong for this type of gauge bosons . in order to illustrate the possible values of the relic density for different values of the free parameters , fig .
[ fig : figure2 ] shows possible values of the dm relic density for dm masses @xmath56~\text{gev}$ ] for @xmath57 .
the blue dots correspond to the solutions when @xmath58 $ ] , and the red dots are for values of @xmath59 $ ] .
the region of the current relic density measured by planck @xcite is marked by a blue band . in this band , many solutions exist .
of course , our dm candidate could make up only part of the total dm relic density , and many solutions for @xmath60 smaller than the current value also exist .
as one can appreciate from these numerical results , there is no need to be on the resonance to achieve the correct relic density value and all the solutions can be in agreement with the collider constraints . in our opinion ,
the simplicity of this model is very appealing and one can make predictions for direct detection as well , which we discuss in the next section . vs. the dark matter mass @xmath32 .
the gauge boson mass was chosen to be @xmath61 .
blue dots are for values of @xmath58 $ ] , red dots are for values of @xmath59 $ ] .
the blue band shows the allowed range for the dm relic density .
[ fig : figure2 ] ]
the direct detection of the baryonic dark matter candidate is also through the baryonic force .
the elastic spin - independent nucleon dark matter cross section is given by @xmath62 notice that the numerical results will be independent of the matrix elements because the baryon number is a conserved current in the theory .
this is a nice feature of the model because we do not introduce any extra unknown parameter coming from the matrix elements , as one has in several dark matter models such as the higgs portal . in order to show our numerical results we can write the above expression as @xmath63 where @xmath64 is the reduced mass and @xmath65 . the numerical results for the spin - independent elastic dm nucleon cross section are shown in fig . [ fig : figure3 ] as a function of the dark matter mass @xmath32 for two ranges of @xmath30 .
for definiteness , we choose the same value @xmath43 as in the analysis for the relic density changing @xmath23 between 0.1 and 0.5 .
one can appreciate that the xenon100 @xcite and the lux @xcite bounds on @xmath66 significantly cut into the parameter space and allow for dm masses larger than about 400 gev .
the projected values for xenon1 t @xcite even constrain the range of dark matter masses more tightly , and allow for values larger than 700 gev .
it is important to note that the range @xmath67 allowed by current lhc results @xcite ( see blue dots in fig .
3 ) gives several scenarios consistent with the relic density constraints and the xenon100 and lux bounds .
we have to say that thanks to the experimental collaborations we can rule out a large fraction of the parameter space and with future experiments such as xenon1 t we will be able to rule out most of the solutions for the dark matter mass below 1 tev in case of no discovery .
let us discuss the possible correlation between possible discoveries at the large hadron collider and in dark matter experiments .
at the large hadron collider we could discover the new neutral gauge boson associated to the breaking of the local baryon number , the gauge boson @xmath19 .
therefore , one could know about the mass @xmath30 and the gauge coupling @xmath23 . assuming that our dark matter candidate describes all the relic density in the universe and for
a given value of the spin - independent cross section one can solve for the dark matter mass @xmath32 and the baryon number @xmath31 .
then , we could predict the values for the production cross section of a dark matter pair and an energetic jet , relevant for the monojet searches at the large hadron collider .
a possible benchmark scenario is when @xmath68 , @xmath44 tev , @xmath69 gev , and @xmath70 . in summary
, one could say that this theory provides a scenario for dark matter which could be fully tested in the future combining dark matter and collider experiments .
we have proposed a simple theory for dark matter where the cold dark matter candidate has spin one - half and carries baryon number .
we refer to this type of dark matter scenario as `` baryonic dark matter . ''
the baryon number is defined as a local gauge symmetry which is spontaneously broken at the low scale and the stability of the dark matter is a natural consequence coming from symmetry breaking .
this theory for dark matter has only four free parameters which determine the relic density and predictions for the spin - independent cross section relevant for direct detection experiments .
we have shown several numerical results in order to illustrate the possibility to have a consistent scenario for cosmology in agreement with the bounds from dark matter experiments .
in particular , a nice feature of this setup is to have a viable region of dark matter masses in the tev - range , which is therefore also testable at the lhc .
all in all , this theory provides a scenario for dark matter which could be fully tested in the future combining dark matter and collider experiments . in a future publication
we will investigate the possibility to test this theory at the large hadron collider and the predictions for indirect detection experiments .
99 m. drees and g. gerbier , `` mini - review of dark matter : 2012 , '' in [ particle data group ] , `` review of particle physics ( rpp ) , '' phys .
d * 86 * ( 2012 ) 010001 .
m. duerr , p. fileviez prez and m. b. wise , `` gauge theory for baryon and lepton numbers with leptoquarks , '' phys .
lett . * 110 * ( 2013 ) 231801 [ arxiv:1304.0576 [ hep - ph ] ] .
a. pais , `` remark on baryon conservation , '' phys .
d * 8 * ( 1973 ) 1844 .
s. rajpoot , `` gauge symmetries of electroweak interactions , '' int .
j. theor .
* 27 * ( 1988 ) 689 .
p. fileviez perez and m. b. wise , `` baryon and lepton number as local gauge symmetries , '' phys .
d * 82 * ( 2010 ) 011901 [ erratum - ibid .
d * 82 * ( 2010 ) 079901 ] [ arxiv:1002.1754 [ hep - ph ] ] .
t. r. dulaney , p. fileviez perez and m. b. wise , `` dark matter , baryon asymmetry , and spontaneous b and l breaking , '' phys .
d * 83 * ( 2011 ) 023520 [ arxiv:1005.0617 [ hep - ph ] ] .
fileviez perez and m. b. wise , `` breaking local baryon and lepton number at the tev scale , '' jhep * 08 * ( 2011 ) 068 [ arxiv:1106.0343 [ hep - ph ] ] .
fileviez perez and h. h. patel , `` baryon asymmetry , dark matter and local baryon number , '' arxiv:1311.6472 [ hep - ph ] .
b. a. dobrescu and f. yu , `` coupling mass mapping of di - jet peak searches , '' phys .
d * 88 * ( 2013 ) 035021 [ arxiv:1306.2629 [ hep - ph ] ] .
h. an , r. huo and l .- t . wang , `` searching for low mass dark portal at the lhc , '' phys .
dark univ .
* 2 * ( 2013 ) 50 [ arxiv:1212.2221 [ hep - ph ] ] .
m. t. frandsen , f. kahlhoefer , a. preston , s. sarkar and k. schmidt - hoberg , `` lhc and tevatron bounds on the dark matter direct detection cross - section for vector mediators , '' jhep * 07 * ( 2012 ) 123 [ arxiv:1204.3839 [ hep - ph ] ] .
g. r. farrar and g. zaharijas , `` dark matter and the baryon asymmetry , '' phys . rev .
* 96 * ( 2006 ) 041302 [ arxiv : hep - ph/0510079 ] .
k. agashe and g. servant , `` warped unification , proton stability and dark matter , '' phys .
* 93 * ( 2004 ) 231805 [ arxiv : hep - ph/0403143 ] .
p. a. r. ade _ et al .
_ [ planck collaboration ] , `` planck 2013 results
cosmological parameters , '' arxiv:1303.5076 [ astro-ph.co ] . the atlas collaboration , `` a search for @xmath71 resonances in the lepton plus jets final state with atlas using 14 fb@xmath72 of pp collisions at @xmath73 tev , '' atlas - conf-2013 - 052 .
e. aprile _ et al .
_ [ xenon100 collaboration ] , `` dark matter results from 225 live days of xenon100 data , '' phys . rev .
* 109 * ( 2012 ) 181301 [ arxiv:1207.5988 [ astro-ph.co ] ] .
d. s. akerib _ et al . _
[ lux collaboration ] , `` first results from the lux dark matter experiment at the sanford underground research facility , '' arxiv:1310.8214 [ astro-ph.co ] .
e. aprile [ xenon1 t collaboration ] , `` the xenon1 t dark matter search experiment , '' arxiv:1206.6288 [ astro-ph.im ] . | we investigate a simple extension of the standard model where the baryon number is a local gauge symmetry and the cold dark matter in the universe can be described by a fermionic field with baryon number .
we refer to this scenario as `` baryonic dark matter '' .
the stability of the dark matter candidate is a natural consequence of the spontaneous breaking of baryon number at the low scale and there is no need to impose an extra discrete symmetry .
the constraints from the relic density and the predictions for direct detection are discussed in detail .
we briefly discuss the testability of this model using the correlation between the large hadron collider data and possible results from dark matter experiments . |
video segments show the 8-bond transitions between ground states .
the clusters transition from the ground state pictured on the left to the ground state pictured above .
connectivity diagrams label the excited state shown in each movie .
the micrographs were divided by a background to remove static artifacts and scaled to create identical background intensities .
we created this compilation using the matplotlib library @xcite .
videos are played back at the recording rate of 3 frames per second .
this clip of 11 minutes ( 2000 frames ) of raw data shows our experimental arrangement for simultaneously observing 4 clusters of 6 spheres while they rotate , translate , and rearrange .
the clusters rearrange frequently , but rarely break apart .
playback is 10 times faster than recorded . | we study experimentally what is arguably the simplest yet non - trivial colloidal system : two - dimensional clusters of 6 spherical particles bound by depletion interactions .
these clusters have multiple , degenerate ground states whose equilibrium distribution is determined by entropic factors , principally the symmetry .
we observe the equilibrium rearrangements between ground states as well as all of the low - lying excited states .
in contrast to the ground states , the excited states have soft modes and low symmetry , and their occupation probabilities depend on the size of the configuration space reached through internal degrees of freedom , as well as a single `` sticky parameter '' encapsulating the depth and curvature of the potential . using a geometrical model that accounts for the entropy of the soft modes and the diffusion rates along them , we accurately reproduce the measured rearrangement rates .
the success of this model , which requires no fitting parameters or measurements of the potential , shows that the free - energy landscape of colloidal systems and the dynamics it governs can be understood geometrically .
colloidal clusters containing a few particles bound together by weak attractive interactions are among the simplest , non - trivial systems for investigating collective phenomena in condensed matter .
such clusters can equilibrate on experimental time scales and display complex dynamics , yet are small enough that the ground states can be enumerated theoretically , and the positions and motions of all the particles can be measured experimentally .
theoretical and experimental work on isolated three - dimensional ( 3d ) colloidal clusters of monodisperse particles has shown how the number of ground states changes with the number of particles @xmath0 @xcite and how the free energies of the rigid states are related to entropy - reducing symmetry effects and entropy - enhancing vibrational modes @xcite .
the importance of entropy in colloidal clusters stands in stark contrast to the case of atomic clusters , where potential energy effects dominate .
the entropically - favored clusters are important clues to understanding nucleation barriers in bulk colloidal fluids @xcite and the local structure of gels @xcite .
however , the excited states and structural rearrangements in such clusters have not yet been studied experimentally . in bulk materials ,
local structural rearrangements are important to a variety of dynamical phenomena , including the glass transition @xcite , aging @xcite , epitaxial growth @xcite , and the jamming transition @xcite .
a better understanding of the internal dynamics in colloidal clusters could reveal local mechanisms underpinning these bulk phenomena .
only a few experimental studies have explored internal dynamics in colloidal clusters : perry and coworkers examined transitions between two states of a 3d 6-particle cluster of spherical particles @xcite ; yunker and coworkers studied relations between the vibrational mode structure and the contact network in disordered , two - dimensional ( 2d ) clusters of polydisperse particles as a function of @xmath0 @xcite ; and chen and coworkers examined the interconversion and aggregation pathways in clusters of particles with directional attractions @xcite .
as yet , however , a quantitative understanding of the rearrangement rates and the pathways through the excited states remains challenging .
transition - state models @xcite , which relate dynamics to the heights of saddle points on the energy landscape , are not easily applied to colloids because the fluid surrounding the particles damps and hydrodynamically couples their motions , and the short - ranged interactions typical of colloidal particles are not easily measured , making the topography of the landscape difficult to accurately compute .
indeed , as we shall show , the excited state occupation probabilities and the transition rates are sensitive to fine details of the potential , which are not easily measured .
( a ) sds micelles induce a short - range depletion attraction between polystyrene microspheres and between the microspheres and the nearby glass coverslip .
( b ) time - lapse images demonstrate a transition .
( c ) the three rigid ground states and their theoretical and experimental probabilities with 95% confidence intervals @xcite ( the probabilities for the parallelogram include both chiral enantiomers ) . ] we study experimentally the excited states and rearrangement rates in perhaps the simplest type of colloidal cluster : isostatic arrangements of equal - sized , spherical colloidal particles , constrained to lie on a plane and held together by well - controlled , short - range attractions a few times the thermal energy @xmath1 in depth ( figure [ fig : diagram]a ) . because the clusters are isostatic
, all excited states have zero - frequency modes , or soft modes , in their vibrational spectra ( figure [ fig : diagram]b , @xcite ) . by tracking the particles over long times
, we quantify the equilibrium probability of each excited state and the motions of the particles within each soft mode .
surprisingly , the dynamics that emerge from this landscape can be quantitatively described by a simple geometric model involving only two parameters , a `` sticky parameter '' that characterizes both the depth and curvature of the attraction , and a diffusion coefficient , which we find to be insensitive to the mode .
both parameters can be easily measured .
therefore , no detailed knowledge of the interactions or hydrodynamics is required to reproduce the rates of rearrangement between ground states .
to make clusters , we first load an aqueous suspension of 1.3 @xmath2m - diameter sulfate polystyrene microspheres into a cell made from two plasma - cleaned glass coverslips separated by 35 @xmath2 m dupont mylar^^ a spacers @xcite .
the only additional component in the suspension is sodium dodecyl sulfate ( sds ) , a surfactant that forms negatively charged micelles in solution .
the micelles create a weak depletion interaction @xcite between the particles and a stronger depletion interaction between the particles and coverslip @xcite , as illustrated in figure [ fig : diagram]a . at 33.4 mm sds
, we observe that 2d clusters bound to a coverslip frequently transition between states but rarely split apart or merge @xcite . at this concentration
, the sodium counterions from the surfactant reduce the debye length to 2.85 nm , setting the effective hard - sphere depletion range of the micelles to 30 nm , just 2.3% of the particle diameter @xcite . as a result ,
the electrostatic and depletion interactions between the particles are short - ranged .
there is likely also a short - range van der waals attraction , which we estimate tapers off to @xmath1 when the particle surfaces are 145 nm apart @xcite . at the beginning of the experiment
, we assemble clusters at the top of the sample cell using optical tweezers .
we then turn off the tweezers and record digital micrographs for the remainder of the experiment .
the clusters , which would normally sediment , remain at the underside of the upper coverslip , confirming the depletion attraction .
we use particle tracking algorithms to locate the particles @xcite , link the locations into trajectories through time , and automatically identify the cluster configurations @xcite .
we focus on 6-particle clusters because this is the smallest system with multiple ground states . because these clusters are bound by short - range interactions , the potential energy is proportional to the number of contacts or `` bonds '' between particles .
the 6-particle clusters adopt three ground states with nine bonds each ( figure [ fig : diagram]c ) : the parallelogram ( which has two enantiomers ) , chevron , and triangle . in aggregate , the clusters occupy the parallelogram and chevron states for equal amounts of time but spend only one third as much time in the triangle state ( figure [ fig : diagram]c ) .
the measured occupation probabilities agree with the expectation for a statistical mechanics ensemble in equilibrium .
to calculate the probabilities , we assume that the translational , rotational , and vibrational degrees of freedom are independent , the vibrational modes are harmonic , and the translational contributions and potential energy differ negligibly among the 3 states @xcite .
as seen previously in 3d clusters , the differences in occupation probabilities are primarily due to symmetry , which enters into the rotational contribution @xcite .
the excited states of the system have more complex and interesting structures .
all of them have zero - frequency modes .
the modes we see at the 8-bond energy level have either hinge - like joints or diamond - square - diamond @xcite flexibility ( figure [ fig : thryvexpt ] ) .
although the 7-bond energy level has twice as many states , nearly all of the zero - frequency modes are simply combinations of these two types of motion ( figure [ fig : thryvexpt ] ) .
the exceptions are a state with a flexible ring of five spheres and a state with a single sphere detached from the cluster .
we do not include this disconnected state in our 7-bond probability calculations because it is not a true 6-sphere cluster .
the fraction of time the clusters spend in the excited energy levels depends on the surfactant concentration . at a concentration of 33.4 mm sds
, the clusters spend 95.5% of the time in states with 7 or more bonds . of this time
, 79.6% is spent in ground states , 18.0% in 8-bond excited states , and 2.4% in 7-bond excited states .
as we decrease the surfactant concentration , the distribution shifts toward the excited energy levels .
qualitatively , this shift makes sense , since decreasing surfactant concentration corresponds to decreasing depletion strength . to understand the energy level occupation probabilities quantitatively
, we must consider the entropy of the soft modes .
we return to this point later . despite the wide variety of structures in the excited states ,
few have any symmetry .
surprisingly , the few symmetric states do not occur as infrequently as we might expect , given the dominant role symmetry more specifically , permutational entropy @xcite plays in the probabilities of 6-sphere ground states in both 2d and 3d .
furthermore , the asymmetric states have a highly non - uniform distribution that is only partially explained by the increased probability of states that are pairs of chiral enantiomers ( figure [ fig : thryvexpt ] ) .
these observations suggest that the variation in probabilities arises from entropic factors other than the permutational contribution .
.structural rearrangement rates between each of the ground states : ( p)arallelogram , ( c)hevron , and ( t)riangle . in total , we observed 820 transitions in 25.6 hours of data from 44 clusters . measured values used in postfactor : @xmath3 @xmath2m@xmath4/s ( 234 @xmath2m@xmath4/hr ) ,
@xmath5 and @xmath6 = 1.3 @xmath2 m . [ cols="^,^,^,^,^,^,^,^,^,^,<,^,^,^,^,^ " , ] [ tab : zs ] [ tab : geometricalcont ] |
spatial periodicity of the lattice potential .
right : virtual four - photon process contributing to a lattice potential with @xmath1 spatial periodicity . in these simple scheme
however the usual lattice potential of @xmath0 periodicity due to second order processes dominates .
( b ) improved scheme for generation of a four - photon lattice with @xmath1 spatial periodicity , as used in our experimental work .
compared to the four - photon process of ( a ) , no second order standing wave processes occur .
the scheme can be generalized to lattice potentials with @xmath2 periodicity , where @xmath3 is an integer number . ]
spatial periodicity .
( b ) four - photon lattice with @xmath1 spatial periodicity .
due to its smaller spatial periodicity , the splitting of the clouds is a factor two above that observed in ( a ) .
( c ) asymmetric lattice realized by superimposing two- and four - photon lattice potentials .
( d ) same as in ( c ) , but with an additional phase shift of 180@xmath36 for the four - photon lattice potential . ]
( bars ) and @xmath74 ( crosses ) as a function of the phase shift of the four - photon component of the lattice potential .
each point corresponds to the average of 4 measurements .
the data has been fitted with sinusoidal functions . ] | we demonstrate a scheme for the fourier synthesis of periodic optical potentials with asymmetric unit cells for atoms . in a proof of principle
experiment , an atomic bose - einstein condensate is exposed to either symmetric or sawtooth - like asymmetric potentials by superimposing a conventional standing wave potential of @xmath0 spatial periodicity with a fourth - order lattice potential of @xmath1 periodicity . the high periodicity lattice is realized using dispersive properties of multiphoton raman transitions
. future applications of the demonstrated scheme could range from the search for novel quantum phases in unconventionally shaped lattice potentials up to dissipationless atomic quantum ratchets . the properties of solid state materials are much determined by the internal spatial structure of its constituents .
most natural solid state materials are crystals , in which the constituents are spatially ordered following the principle of lattice structures .
a large class of different unit cells is known , whose classification into crystal classes , point- and space groups is well established within the field of crystallography @xcite . recently ,
artificial crystals of atoms bound by light , so called optical lattices , have developed into model systems for solid state physics problems . in such systems ,
effects like bloch - oscillations and the mott - insulator transition were observed @xcite . on the other hand , current quantum gas experiments with optical lattices
have been limited to sinusoidal atom potentials .
recent theoretical work pointed out that in unconventional lattice structures novel quantum phases are expected @xcite . a possible way to at least in one dimension produce arbitrarily shaped potentials is the use of beams that are inclined under different angles , which may seem impractical and hard to phase stabilize especially if larger numbers of beams are required .
notably , such phase stabilization has recently been achieved for two near - resonant standing waves @xcite . for thermal atoms some nonstandard optical lattices , as subwavelength structure potentials or asymmetric lattices , have been realized with magneto - optical potentials or grey lattices @xcite .
these schemes require near - resonant optical radiation whose inherent dissipation impedes the use of quantum degenerate atomic gases . with the so far demonstrated schemes
, the class of realizable lattice potentials furthermore seems limited . here
, we demonstrate a scheme to create dissipationless optical lattice potentials with @xmath1 spatial periodicity , which is realized with a fourth - order raman process . building upon this potential , a fourier synthesis of lattice potentials
is demonstrated . by combining the @xmath1 period multiphoton potential with a conventional lattice of @xmath0 spatial periodicity with appropriate phase , an asymmetric lattice with variable asymmetry
is realized .
the scheme for the generation of the @xmath1 period lattice potential is generalizable , in principle , to optical lattice potentials with @xmath2 spatial periodicity , where @xmath3 is an integer number . by diffraction of a rubidium bose - einstein condensate at the optical light shift potential ,
absorption images of atom clouds closely connected to the reciprocal lattice of the synthesized potential structures were observed . in conventional optical lattices with two counterpropagating beams of wavelength @xmath4 ,
the resulting optical potential @xmath5 , with @xmath6 as the atomic polarizability , is proportional to @xmath7 , yielding the well - known @xmath0 spatial periodicity .
the atoms here undergo virtual two - photon processes of absorption of a photon from one laser beam and stimulated emission into the counterpropagating beam . for a fourier synthesis of arbitrarily shaped potentials ,
the required higher harmonics can , in principle , be generated by using a combination of standing waves with wavelengths of the fractional harmonics @xmath8 , where @xmath3 is integer , yielding a lattice potential of spatial period @xmath2 .
a lattice potential of the same periodicity could also be produced , in case each virtual absorption or stimulated cycle could be replaced by a simultaneous process induced by @xmath3 photons of the fundamental wavelength @xmath4 absorbed or emitted along the corresponding direction ( see fig .
1a )
. the spatial period of this higher order lattice potential can be expressed as @xmath9 , where @xmath10 denotes the effective wavelength of a @xmath3-photon field .
a direct measurement of this effective multiphoton wavelength is in general a very difficult task , as has been discussed in the context of heisenberg - limited measurements @xcite .
we here use the high frequency resolution of raman transitions between ground state sublevels to separate in frequency space the desired @xmath11-th order process from lower order contributions .
theoretical work has predicted , that a lattice potential of @xmath2 spatial periodicity is indeed expected with this approach @xcite . for early theoretical work analyzing
the use of raman transitions to create optical potentials for atoms , see @xcite . for experimental work on an improved spectral resolution with multiphoton raman transitions ,
see @xcite .
other work has demonstrated , that a near - resonant raman lattice can cool atoms to a few microkelvins @xcite . fig .
1b shows the used scheme for a four - photon lattice , which yields a periodic potential with @xmath1 spatial periodicity .
the scheme uses three - level atoms with two stable ground states @xmath12 and @xmath13 and one electronically excited level @xmath14 .
compared to the four - photon ladder scheme ( right graph of fig .
1a ) , one absorption and one stimulated emission process have been interchanged by a stimulated emission and absorption process of an oppositely directed photon respectively . to avoid second order standing wave processes , a minimum of three laser frequencies is required .
the atoms are irradiated with two driving optical fields of frequencies @xmath15 from the left and one field with frequency @xmath16 from the right .
a non - zero two - photon detuning @xmath17 is used to suppress resonant raman processes .
the adiabatic light shift potential for e.g. atoms in state @xmath12 here exhibits the desired fourth - order energy shift @xmath18 with spatial periodicity @xmath1 @xcite .
the depth of the optical potential is given by @xmath19 with @xmath20 and @xmath21 .
here , @xmath22 and @xmath23 ( @xmath24 and @xmath25 ) are the rabi frequencies for the transitions @xmath26 and @xmath27 driven by the light fields with frequency @xmath28 and @xmath16 ( @xmath16 and @xmath29 ) respectively .
further , @xmath30 denotes the optical detuning from the excited atomic level @xmath14 , which with @xmath31 is here assumed to be identical for all three optical frequencies .
it should be pointed out that the four - photon energy shift is not proportional to @xmath32 , i.e. the effect does not originate from the nonlinearity of the multiphoton transitions @xcite . in an atom optics picture ,
the obtained @xmath1 spatial periodicity can be understood by considering the momentum transfer to an atom during virtual four - photon cycles of alternating absorption and stimulated emission which is @xmath33 , being a factor 2 above the corresponding processes in a standing laser wave . by combining lattice potentials of different spatial periodicities , in principle ,
arbitrarily shaped periodic potentials can be generated . in our experimental setup
( see fig .
2 ) , light detuned some 2 nm to the red of the rubidium d2-line is generated by a high - power diode laser system .
the emitted radiation is split into two beams ( beam1 , beam2 ) , from which , after passing independent aousto - optical modulators ( aom1 , aom2 ) , the two counterpropagating beams realizing the lattice potentials are derived .
the aoms are used for switching of the optical fields and in the case of beam2 , to superimpose several optical frequency components onto a single beam path . to generate the rf drive signals for the aoms , we use a set of four phase - locked function generators .
while aom1 is driven with a single radio frequency , for aom2 , the radio frequency signals of three function generators are combined . in this way
, the three different optical frequencies required to synthesize the asymmetric lattice can be generated in beam2 .
after passing the aoms , the two beams are fed through optical fibers and focused in a counterpropagating geometry onto a bose - einstein condensate ( bec ) confined in a co@xmath34-laser dipole trap .
the typically used optical lattice beam powers are 18mw for the frequency components @xmath15 , and 24mw ( 4mw ) for the components with frequency @xmath16 in beam1 ( beam2 ) on a 40@xmath35 m beam waist .
the optical lattice beams are inclined under an angle of 41@xmath36 respectively to the horizontally orientated co@xmath34-laser trapping beam .
our @xmath37rb bec is produced all - optically by evaporative cooling in a tightly focused co@xmath34-laser beam operating near a wavelength of 10.6@xmath35 m , as described in detail in @xcite . by activating a magnetic gradient field during the evaporation phase
, a spin - polarized condensate with approximately 10,000 atoms in the @xmath38 zeeman state of the @xmath39 ground state is produced .
a magnetic bias field generates a frequency splitting @xmath40khz between neighboring zeeman ground states of the rubidium atoms . for a typical optical frequency difference @xmath41khz
we arrive at a raman detuning @xmath42khz .
the magnetic bias field forms an angle with respect to the optical beam axis , so that the atoms experience both @xmath43 , @xmath44 , as well as @xmath45 polarizations . to realize the multiphoton lattice as shown in fig .
1b , we use the @xmath39 ground state levels @xmath38 and @xmath46 as states @xmath12 and @xmath13 respectively and the @xmath47 manifold as the electronically excited state @xmath14 . since the lattice beam detuning is orders of magnitude larger than the upper state hyperfine structure splitting , raman transitions only couple neighboring ground state zeeman levels . for alkali atoms ,
the raman lattice detuning is limited to a value of order of the excited state fine structure splitting ( @xmath4815 nm for rb , 34 nm for cs ) , which for these atoms for typical parameters limits the expected coherence time of the multiphoton lattice to a value of order of a second due to photon scattering .
longer coherence times should be possible e.g. with the use of metastable p - triplett states of two - electron atoms , as e.g. the @xmath49 state in atomic ytterbium with 17s life time , an atom with which bec has recently been achieved @xcite .
after the generation of the bec , the lattice beams are activated for a period @xmath506@xmath35s . to read out the spatially varying potential of the multiphoton lattice , we study the far - field diffraction pattern of the imprinted phase grating after a free falling time of 10 ms . from the measured diffraction patterns
, we can reconstruct the optical potential . as is well known from crystal structure analysis , measurements of the intensity of a scattered or diffracted beam ,
do not in general allow one to reconstruct the lattice structure , as is known as the phase problem .
for many special cases , a reconstruction is however possible @xcite . in our experiment , assuming that the one - dimensional lattice is invariant to spatial translations of @xmath0 , the reciprocal lattice is limited to the harmonics @xmath2 and it is straightforward to show that a measurement of the atom diffraction pattern @xmath51 allows one to reconstruct the spatial lattice structure .
we have fitted the result of a numerical solution of the time - dependent , momentum - picture schrdinger equation to our experimental far - field diffraction data .
the used hamiltonian was @xmath52 , where @xmath53 . for known phase @xmath54 and pulse length @xmath55 , the potential depths @xmath56 and @xmath57 of the two- and four - photon lattice components were left as free parameters and could thus be determined from the fit .
interestingly , we found no major difference between these results and an analytical solution for the atomic diffraction pattern obtained using the simple thin grating ( raman - nath ) approximation ( roughly below 20% difference ) . in the raman - nath approximation ,
the kinetic energy in the hamiltonian is neglected , which requires ( 1 ) @xmath58 , where @xmath59 is the recoil frequency , and ( 2 ) @xmath60 , where @xmath61 is the oscillation period @xcite .
the first condition is fulfilled in our experiment , whereas the second is not .
the difference between the numerical solution and the predictions in the raman - nath approximation is due to the neglect of the center - of - mass motion in this approximation . within the raman - nath regime ,
a quite intuitive connection between the lattice structure and its far - field diffraction pattern is possible .
a bec with initially homogeneous phase here accumulates a spatially varying phase shift @xmath62 , and one finds that the far - field momentum distribution is @xmath63 , where ft denotes the fourier transform with respect to @xmath64 @xcite . for small pulse envelopes ,
in which the exponential factor can be linearized , the time of flight image is directly related to the fourier transform of the potential @xmath65 , i.e. the reciprocal lattice . within second order we find @xmath66 , which allows us to express the probability to find an atom in @xmath67 .
diffraction orders as @xmath68 $ ] with @xmath69 . by choosing variable amplitudes and phases of the lattice potentials ,
different lattices were fourier synthesized . to begin with , we investigated a conventional optical lattice .
fig .
3a shows a typical obtained absorption image along with the reconstructed potential ( which yielded a potential depth @xmath703.4@xmath35k ) . for this measurement , only two counterpropagating beams with frequency @xmath16
were required .
fig .
3b shows corresponding results obtained for the four - photon lattice potential ( @xmath712.5@xmath35k ) , as realized with the scheme shown in fig . 1b . for this measurement ,
two beams with frequency @xmath15 and one counterpropagating beam with frequency @xmath16 were activated .
the main peaks of the absorption image here are spatially separated by twice the amount shown in fig .
3a , which reflects the by a factor two smaller spatial periodicity of @xmath1 of this four - photon lattice . to demonstrate a fourier synthesis of atom potentials , two- and four - photon lattice potentials were overlapped . here
, all four light fields were used ( see fig .
2 ) , so that in addition to the optical frequency components shown in fig .
1b also a counterpropagating beam with frequency @xmath16 was active . by changing the phase of the control beam with frequency @xmath16 with respect to the other beams
, the phase of the four - photon contribution could be varied relatively to the position of the two - photon potential .
fig .
3c and fig .
3d show far - field diffraction patterns and the reconstructed unit cells for such lattices .
for the measurement shown in fig .
3d ( @xmath725.6@xmath35k , @xmath713.8@xmath35k ) , the phase of the four - photon contribution was changed by 180@xmath36 with respect to that shown in fig .
3c ( @xmath704.5@xmath35k , @xmath713.7@xmath35k ) . in both images
the diffraction pattern is clearly asymmetric , especially when considering the amplitude of the @xmath73 peaks .
this is attributed as evidence for asymmetric atom potentials , as also visible in the shown reconstructed lattice potentials .
we have also investigated diffraction patterns from lattice potentials with a more complete range of phase shifts of the four - photon potential .
as the fundamental period of the combined lattice is @xmath0 , we anticipate that the relative magnitude of the @xmath73 diffraction peaks gives an indication for the asymmetry of the lattice potential .
fig .
4 shows the fraction of atoms diffracted into the @xmath73 orders relatively to the total number of atoms as a function of the phase shift .
the data points have been fitted with sinusoidal functions which is in accordance with theory for small pulse envelopes .
one clearly observes the variation of asymmetry of the diffraction pattern as a function of the phase between the different fourier components .
the deviations of the fitted curves to the theoretical formula given above can be attributed to the fact , that were are not completely in the raman - nath regime .
to conclude , we report on the realization of atom potentials suitable for atomic bose - einstein condensates with spatial periodicity @xmath1 using multiphoton raman transitions . this has allowed us to demonstrate a scheme for the fourier synthesis of asymmetric shaped lattice potentials . in future
, it would be interesting to extend this scheme to include even smaller period lattice potentials .
besides in the search for new quantum phases , tailored high - periodicity lattices are of interest e.g. in the context of subhealing length physics , atom lithography @xcite and quantum computing @xcite .
an intriguing other perspective includes studies of quantum ratchets with atomic quantum gases .
it is expected that the investigation of the extremely weakly damped regime , in which the atomic inertia plays a large role , will allow for novel quantum dynamical phenomena @xcite .
we thank p. hnggi , d. klle , s. flach , and l. santos for discussions .
we acknowledge support by the deutsche forschungsgemeinschaft , the landesstiftung baden - wrttemberg and the european community .
25 natexlab#1#1bibnamefont # 1#1bibfnamefont # 1#1citenamefont # 1#1url # 1`#1`urlprefix[2]#2 [ 2][]#2 , _ _ ( , , ) . ,
* * , ( ) . ,
* * , ( ) . , * * , ( ) . , * * , ( ) . , * * , ( ) . , , , , , * * , ( ) . , * * , ( ) . , , , , * * , ( ) . , , , * * , ( ) . , * * , ( ) . , , , , * * , ( ) . , * * , ( ) . , , , , * * , ( ) . , , , , * * , ( ) . , * * , ( ) , , , , , * * , ( ) . , * * , ( ) . , _ _ ( , , ) . , p. ( ) , , * * , ( ) . ,
* * , ( ) . , , , , , * * , ( ) .
, , , * * , ( ) . ,
* * , ( ) . |
human behavior is hierarchically structured . even simple actions
checking email , for example consist of many steps , which span levels of description and complexity : _ moving _ fingers , arms , eyes ; _ choosing _ whether to use the mouse or keyboard ; _ searching _ for the email app among the others that are open , etc .
human behavior is also efficient : we attempt to perform each part of each action as swiftly and successfully as we can , at the least possible cost .
classical models of how humans understand the actions of others ( originating from the plan recognition literature @xcite ) seek to leverage this hierarchical structure by building in prior knowledge of others high - level actions , tasks , and goals .
behavioral evidence has shown that adults and even infants can learn simple hierarchical action structures from data , segmenting novel sequences along statistical action boundaries @xcite .
these abilities can be captured with a nonparametric action segmentation model @xcite , which learns lists of actions that occur in sequence , and potentially cause observable effects .
however , purely statistical learning from data leaves implicit the intrinsic efficiency of intentional actions . for adults and infants ,
the assumption that others actions are rational functions of their beliefs , desires , and goals is fundamental @xcite . rather than simply memorizing repeated action sequences ,
people infer goals of complexity sufficient to rationalize these actions @xcite .
a simple formalization of people s theory of rational , goal - directed action can be given in terms of approximately rational planning in markov decision processes ( mdps ) , and human goal inferences can be accurately predicted using bayesian inference over models of mdp planning ( or inverse planning ) @xcite .
this approach is closely related to recent plan recognition algorithms developed in the ai literature ( * ? ? ?
* e.g. ) . in this paper , we integrate hierarchically structured actions into the inverse planning framework using hierarchical mdps @xcite .
we consider scenarios in which agents pursue sequences of subgoals enroute to various destinations .
each destination can have multiple subgoal sequences , generated by a nonparametric bayesian model .
together , a destination and subgoal sequence induce a hierarchical mdp , and agents actions are assumed to be generated by approximately rational planning in this mdp .
representing agents subgoal - based plans allows segmentation of behavior into extended chunks , separated by sparse subgoal boundaries , and can achieve greater generalization than purely statistical models by naturally capturing the context - sensitivity of rational action sequences , rather than having to learn new sequences for each context .
further , the model predicts deviations from rationality ( with respect to a single subgoal ) at subgoal boundaries a strong cue to hierarchical structure .
this inductive bias should enable efficient learning of subgoal structure from small numbers of examples .
we present two experiments to test our model .
the first is a behavioral experiment , in which human participants inferred the subgoal structure underlying series of action sequences .
the second experiment is a simulation to show that the model is useful in an artificial user support task .
the model first learns the subgoal structure of the task from a small number of observations .
then , the model infers a user s destination and subgoals from a partial action sequence , and attempts to assist the user to achieve a subset of the remaining subgoals . for each experiment
, we compare the performance of our model with that of several natural alternatives .
[ fig : pp_str ] represents the structure of our model in terms of separate graphical models for the hierarchical mdp and nonparametric subgoal models .
these graphical models specify the structure of the joint distributions over actions , state sequences , and subgoal sequences .
our model represents the situational context in terms of a finite state space @xmath0 .
the variable @xmath1 denotes a state sequence of length @xmath2 , such that @xmath3 is the @xmath4th state in the sequence .
the variable @xmath5 represents a sequence of @xmath6 subgoals .
we denote the @xmath7th subgoal of @xmath5 as @xmath8 . for convenience , we assume that @xmath9 is the destination .
we denote the set of actions as @xmath10 , and the action executed in @xmath11 as @xmath12 ( see fig . [
fig : pp_str](a ) ) .
the remainder of this section will first define the hierarchical mdp induced by a given destination and subgoal sequence , and derive the likelihood of a subgoal sequence , given an observed state sequence .
then we describe the nonparametric model of subgoal sequences , and a markov chain monte carlo ( mcmc ) method for jointly inferring the number and composition of the subgoal sequences underlying a series of action sequences .
finally , we show how to use subgoal sequences learned from previous observations to predict the subgoals and destination of a novel partial action sequence .
ll [ fig : pp_str ] our hierarchical mdp formulation is closely related to the options framework for hierarchical reinforcement learning @xcite . for simplicity
, we assume that actions and state transitions are deterministic , that each action incurs a cost of 2 , that the discount factor @xmath13 , and that the destination yields a reward of 100 once all subgoals have been achieved .
the subgoal sequence @xmath5 is analogous to a set of options , with initiation and termination conditions that require each subgoal to be achieved in sequential order . in fig .
[ fig : pp_str](a ) , the variable @xmath14 keeps track of the current subgoal at time @xmath4 . assume the current subgoal is @xmath15 ; if the agent reaches the current subgoal , i.e. , @xmath16 , then @xmath17 should be the next subgoal , otherwise the subgoal should stay the same ( @xmath18 ) . based on this , an observed sequence @xmath1 can be divided into multiple segments corresponding to @xmath5 .
we define boundary @xmath19 to be the first timestep @xmath4 after @xmath20 is achieved : @xmath21 ; @xmath22 .
we write the boundary vector @xmath23 .
if @xmath1 achieves all subgoals in @xmath5 in order , the length of @xmath24 , @xmath25 should be @xmath26 .
otherwise , @xmath1 does not satisfy @xmath5 , so @xmath27 should be 0 .
the subgoal sequence likelihood is then : @xmath28 where @xmath29 is the marginal probability of the state transition from @xmath11 to @xmath30 , integrating over actions .
@xmath31 is a softmax policy of the local mdp state - action value function for subgoal @xmath32 , based on the assumption that the observed agent plans approximately rationally , stochastically maximizing expected reward and minimizing cost .
a similar likelihood computation was used by @xcite within an mcmc method for inferring subgoal sequences from user demonstrations . however , this approach focused on learning only one subgoal sequence from one action sequence ; in the next section , we describe a nonparametric bayesian model and mcmc methods for inferring multiple subgoal sequences , given a series of state sequences .
we now consider inference of subgoal structure by observing multiple sequences for a certain destination .
we denote the set of @xmath33 behavior sequences as @xmath34 and the @xmath35th sequence as @xmath36 .
we denote a set of @xmath37 subgoal sequences as @xmath38 , and the @xmath39th sequence as @xmath40 .
the problem of nonparametric subgoal inference is to compute @xmath41 for an unbounded number of sequences @xmath37 .
[ sec : gibbs ] we model the set of unknown subgoal sequences using a nonparametric bayesian model , which allows us to consider an unbounded number of subgoal sequences for each destination .
we use the dirichlet process ( dp ) to express the distribution over subgoal sequences , following @xcite . a graphical model of our dp model
is shown in fig .
[ fig : pp_str](b ) .
we use the chinese restaurant process ( crp ) representation to efficiently draw samples from the dp .
first , the crp selects a `` table '' for each observation @xmath36 , conditioned on all previous table assignments and the concentration parameter @xmath42 .
@xmath43 is the index of the table assigned to state sequence @xmath36 .
next , for each crp table , a subgoal sequence is sampled from the base distribution @xmath44 , and @xmath40 denotes the subgoal sequence associated with the @xmath39th table .
the state sequence @xmath36 is then generated given its associated subgoal sequence .
we use a mcmc method to compute the posterior probability over subgoal sequences , specifically , gibbs sampling .
gibbs sampling allows us to approximate the complex dp distribution by inducing a markov chain over samples from the crp .
gibbs sampling over the crp is a standard mcmc algorithm for dp inference @xcite .
algorithm [ alg : gibbs ] is an overview of our algorithm .
as an initialization step , we assign a different table for each state sequence , and draw subgoal sequences from the conditional distribution over subgoal sequences given the state sequence assigned to each table .
we then repeat the table re - assignment step ( resampling the table for each state sequence ) and the parameter re - assignment step ( drawing the subgoal sequences from the conditional distribution over subgoal sequences , given all sequences assigned to a table ) . for the table re - assignment step
, we calculate the probability @xmath45 to assign sequence @xmath36 to table @xmath39 according to standard gibbs sampling for the crp : @xmath46 where @xmath47 denotes table assignments , excluding sequence @xmath35 , and @xmath48 denotes the number of sequences assigned to table @xmath39 , excluding sequence @xmath35 .
however , for our problem , @xmath49 is the mdp likelihood of a subgoal sequence . because this is a non - conjugate distribution
, we can not integrate this equation analytically .
if the environment is small , we can enumerate all of , and compute it directly as in the previous section .
in large environments we must use an approximate method to choose a new table ; some techniques are described by @xcite . in the parameter re - assignment step , we draw the subgoal sequence from the posterior over subgoal sequences , given all sequences assigned to a table .
assume @xmath50 are the sequences assigned to table @xmath39 .
the distribution to draw a new subgoal sequence @xmath5 for table @xmath39 should be @xmath51 .
this probability for each subgoal sequence can be calculated as follows : @xmath52 at the end of each step of the loop , we count the number of subgoal sequences for each state sequence .
we represent the number of times that @xmath5 is assigned any state sequence as @xmath53 .
the normalized count corresponds to @xmath54 .
@xmath55 // initialize step ( see , eq .
[ eq : ll_sb ] ) @xmath56
// table re - assign step ( see , eq .
[ eq : gibbs ] ) @xmath57 //
parameter initialize for new table ( see , eq .
[ eq : ll_sb ] ) @xmath58 // parameter re - assign step ( see , [ eq : ll_sb_multi ] ) @xmath59 @xmath60 @xmath54
our two experiments presented a `` warehouse '' scenario involving the delivery of various items to destinations in the environment shown in fig .
[ fig : warehouse ] .
the warehouse has three delivery destinations : a , b , and c. there are nine potential items to be delivered , marked by numbers 1 - 9 .
the items are arranged into three rows , and for each delivery , one item can be delivered from each row . each job in the warehouse has a specific destination , and several possible `` item lists '' to deliver .
an item list consists of either one , two , three items that must be delivered to the destination . for each delivery ,
one of these item lists is selected by the warehouse scheduler .
in addition to items from the current item list , workers are encouraged to pick up `` add - on '' items , but only if this wo nt increase the number of steps on their path to the destination .
for example , if the item list for fig .
[ fig : warehouse](b ) is [ 2,8 ] , item 5 is a good add - on item for that delivery , because item 5 is the only additional item that does not require additional steps to obtain , given the start point , item list , and destination .
there is a direct correspondence between delivery destinations and item lists in this setting and destinations and subgoals as represented by our model .
ccc we recruited participants for this study using amazon mechanical turk .
participants were 29 adults located in the usa ( male 13 , female 14 , unknown 2 ) .
mean age was 32 years old .
first , in a training phase , subjects learned to perform an example warehouse job to learn the rules of the scenario .
in each trial of the testing phase , participants saw 8 paths for a particular job , and inferred how many item lists the job had , and which items were on each list .
we prepared 22 jobs for the testing phase , including a range of subgoal structures of varying complexity and difficulty .
first were jobs with one item list , involving 1 , 2 , or 3 subgoals .
second were jobs with two item lists , involving 1 , 2 , or 3 subgoals for both item lists , or 1 and 3 subgoals for each item list , respectively .
some jobs presented two types of action sequences . in one type
, the action sequence could reach all subgoals without deviating from the shortest path to the destination .
these cases were difficult in that they did not provide direct evidence for any subgoal . in the other type , the action had to make a detour to reach the subgoal , which provided stronger evidence for that subgoal . in our mdp representation of the environment , @xmath0 corresponds to the agent location , and @xmath10 is the set of movements ( up , left , and right ) .
we used a uniform distribution for the prior over subgoal sequences @xmath61 and set @xmath62 . for the nonparametric bayesian model
, we set the concentration parameter @xmath63 .
we used 5,000 iterations of gibbs sampling , with 1,000 steps burn - in time to reach the stationary distribution .
[ [ alternative - models ] ] alternative models + + + + + + + + + + + + + + + + + + we prepared 3 alternative models to compare with our bayesian nonparametric model : _ independent model _ , _ logical possibility model _ , and
_ copy model_. to explain each alternative model , we consider the case in which the model observes @xmath33 sequences @xmath34 .
the independent model is most similar to our approach .
the only difference is that this model does not use the crp , but instead calculates the posterior for each sequence @xmath64 independently .
the joint probability for all sequences is given by : @xmath65 .
the marginal probability for each subgoal sequence is given by : @xmath66 .
the logical possibility model is a more heuristic approach , which computes the proportion of observations in which a subgoal sequence is present within the action sequence .
formally , @xmath67 . the copy model is the simplest alternative , but perhaps the most intuitive .
it assigns probability 1 to the subgoals which have the highest likelihood for each observed sequence .
in other words , the inferred subgoal sequence is the maximal subgoal sequence that is included by an observed action sequence .
if we represent such a subgoal sequence @xmath68 , the copy model can be written : @xmath69 if @xmath70 for some @xmath35 , otherwise 0 . to compare human behavioral results with those of our models ,
we compute the proportion of participants who selected each item list for each trial .
model predictions were based on the marginal probability that each item list generated at least one path .
table [ tbl : exp1corr ] compares the correlation of the bayesian nonparametric subgoal model and our three alternative models with human inferences .
the bayesian nonparametric model correlates very strongly with human inferences .
the other models also correlate positively with human inferences , but all are substantially worse than the bayesian nonparametric model .
.pearson correlation between human inferences and computational models .
lp stands for `` logical possibility '' .
[ cols="^,^,^,^,^",options="header " , ] fig .
[ fig : exp1scatter ] plots human inferences against the bayesian nonparametric model predictions for all trials and subgoal sequences .
this shows that our model can almost perfectly predict whether humans will infer that a certain item list is included in each job .
for a small number of trials , humans are somewhat uncertain whether to include a particular item list , and the model captures this uncertainty as well .
c c finally , we compare our model predictions with human inferences for two specific jobs . in fig .
[ fig : exp1_agents](a ) , each path includes item 5 , but there are no paths which include only this item .
humans and the bayesian nonparametric model infer that [ 5 ] is the most likely subgoal sequence , based on the systematic deviation from the shortest path to the destination to reach this item .
all other models fail in revealing ways , inferring more item lists than are necessary to explain the data . in fig .
[ fig : exp1_agents](b ) , there are two item lists .
although every path is an efficient route to the destination , without inferring an item list , humans and our model can infer these two subgoal sequences correctly by assessing the probability that these items are selected intentionally rather than by coincidence .
our alternative models once again fail to capture people s judgments .
we call this the * worker - helper task*. the main environment is the same as in experiment 1 , but there are two types of agents .
one type is * worker * , whose actions are structured identically to those presented in experiment 1 .
the worker s job remains the same : to deliver an item list to a destination .
the other type is * helper * , which must support workers in achieving their goals .
the task of the helper is to learn the structure of the worker s jobs , then use this to help the worker complete a job in progress by retrieving an item from the item list of the worker , and delivering it to the destination .
helpers begin each trial on the left side of the warehouse , midway between the second row of items ( 4 , 5 and 6 ) and the third row of items ( 7 , 8 and 9 ) . fig .
[ fig : warehouse](c ) illustrates this task , with the start point of the helper marked by a green dot . on each trial ,
the worker randomly chooses a destination ( a , b , or c ) and an item list associated with that destination .
the set of item lists for each destination is fixed and known to the worker , but not the helper .
next , the worker plans a path to achieve their goals and begins to move through the warehouse .
the helper observes the worker s behavior and decides which target item to retrieve by inferring the worker s subgoal sequence ( due to the helper s starting location , the target item will always be in the third row , i.e. , item 7 , 8 , or 9 ) .
once the helper decides their target item , the helper shows this target to the worker and begins to move .
after the worker observes the helper s target item , the worker re - plans its path under the assumption that the helper will get the target item .
after the helper gets the target item , the helper moves toward the destination inferred by observing the worker s path . when the helper can not decide on a target item , they do nothing .
we assume that workers take the optimal actions , given their subgoal sequence , i.e. , @xmath71 .
the helper estimates the marginal probability that each item is included in the subgoal sequence which the worker is currently following .
the helper computes this probability based on the partial path of the worker @xmath72 , and @xmath73 previously observed paths @xmath74 , under the assumption that the worker plans approximately rationally , with softmax parameter @xmath75 .
when the probability of a target exceeds a certain value , the helper decides on this item .
the worker then re - plans by removing the target item and the helper begins to take optimal actions , as does the worker .
[ [ probability - of - target - item - and - destination ] ] probability of target item and destination + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the marginal probability of target item @xmath76 , given @xmath72 and @xmath74 is : @xmath77 , where @xmath78 is the subset of previously observed paths with destination @xmath79 .
@xmath80 is calculated using the subgoal inference method in section 2 .
@xmath81 can be computed using eq .
[ eq : ll_sb ] , but without the constraint that @xmath82 .
the marginal probability of destination @xmath79 , given @xmath72 and @xmath74 is : @xmath83 .
[ [ alternative - models-1 ] ] alternative models + + + + + + + + + + + + + + + + + + we use the same alternative models as in experiment 1 , and a common collaboration protocol ( described above ) for each subgoal inference model .
we also add two new models .
one is no helper , which means the helper has no subgoal knowledge . as a result , the helper does nothing .
the other is the ground truth model , which means the helper knows the correct subgoal .
it is a benchmark to measure the benefit of the bayesian model .
we tested three natural types of subgoal settings .
in the first , each of the three destinations has one subgoal sequence , which consists of either 7 , 8 , or 9 . in the second ,
each of the three destinations has one subgoal sequence , which consists of either item 4 , 5 , or 6 and either item 7 , 8 , or 9 . in the third ,
each of the three destinations has two subgoal sequences which consist of either item 7 , 8 , or 9 .
we generate every possible combination of items for each subgoal setting , then we learn many subgoal structures and evaluate the performance for each subgoal structure . [
[ evaluation - for - one - subgoal - structure ] ] evaluation for one subgoal structure + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we generate a number of sequences of workers from random start points for each destination .
we then compute subgoal inferences for each model using the sequences .
we then execute the collaborative task 99 times for each subgoal inference ( 11 points times 9 trials ) , and evaluate the performance of workers in achieving their destination .
we repeat these tests 5 times for each subgoal structure .
we use the following score to measure performance .
if the worker achieves their goal , they score 100 points , but each action costs 2 points .
this directly corresponds to the mdp reward and cost settings in experiment 1 .
we executed the test , varying the number of input sequences to evaluate the dependence of each model on the amount of input data .
[ fig : exp3](a ) shows the average score of each model as a function of the number of input sequences . generally , the helper using the nonparametric bayesian model is more beneficial than any alternative model .
this model achieves good performance ( almost as high as the ground truth model ) using even a small number of input sequences .
\(a ) + ( b ) further , because the alternative models depend highly on the input sequences , the scores of these models are unstable . fig .
[ fig : exp3](b ) shows the average variance of the results for each subgoal setting ( we repeated the experiment 5 times for each subgoal setting ; this score is the variance of these ) .
none of the alternative models can provide stable user support .
in contrast , the nonparametric bayesian model provides stable support using even a small number of input sequences ( over two input sequences ) .
this stability is a key factor for user support , since unstable help can confuse and frustrate users .
we presented a model of how humans infer subgoals from observations of complex action sequences .
this model used rational hierarchical planning over subgoal structures generated by a nonparametric bayesian model to capture people s intuitions about the structure of intentional actions .
we showed how bayesian inference over this generative model using a novel mcmc method predicts quantitative human subgoal inferences with high accuracy and precision .
we then showed that our model is useful in practice , enhancing performance in an application to an artificial user support task .
our modeling and experimental scenarios are extremely simplistic in comparison with real human behavior .
one limitation is our assumption that subgoal sequences are chosen according to probabilities determined by the dirichlet process .
more generally , the probability of choosing a particular subgoal sequence will depend on the efficiency of that subgoal sequence , relative to the alternatives . to enhance the expressiveness of our model , infinite pcfgs @xcite , adaptor grammars @xcite , or fragment grammars @xcite are promising extensions to the simple dirichlet process we employ .
these and other frameworks which can be applied to structured goal representation complement our work here by naturally interfacing with models of hierarchical planning analogously to the model we describe .
this work was supported by the center for brains , minds & machines ( cbmm ) , funded by nsf stc award ccf-1231216 , and by nsf grant iis-1227495 . | _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ most human behaviors consist of multiple parts , steps , or subtasks .
these structures guide our action planning and execution , but when we observe others , the latent structure of their actions is typically unobservable , and must be inferred in order to learn new skills by demonstration , or to assist others in completing their tasks .
for example , an assistant who has learned the subgoal structure of a colleague s task can more rapidly recognize and support their actions as they unfold . here
we model how humans infer subgoals from observations of complex action sequences using a nonparametric bayesian model , which assumes that observed actions are generated by approximately rational planning over unknown subgoal sequences .
we test this model with a behavioral experiment in which humans observed different series of goal - directed actions , and inferred both the number and composition of the subgoal sequences associated with each goal .
the bayesian model predicts human subgoal inferences with high accuracy , and significantly better than several alternative models and straightforward heuristics .
motivated by this result , we simulate how learning and inference of subgoals can improve performance in an artificial user assistance task .
the bayesian model learns the correct subgoals from fewer observations , and better assists users by more rapidly and accurately inferring the goal of their actions than alternative approaches . _
_ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ |
the main problem of qcd is to understand the structure of hadrons and their properties in terms of quarks and gluons .
nucleon charges defined as matrix elements of vector , axial and tensor currents between nucleon states contain complete information about quark structure of the nucleon .
these charges are connected with the leading twist unpolarized @xmath0 , the helicity @xmath1 and transversity @xmath2 parton distribution functions ( pdfs ) .
the first two pdfs have been extensively investigated theoretically and experimentally in many works ( for instance see @xcite and references therein as well as @xcite ) .
there is a big experimental problem to measure the transversity of the nucleon because of its chiral odd nature . only , recently the tensor charge @xmath2 was extracted @xcite using the data from belle @xcite , hermes @xcite and compass @xcite collaborations . this extraction is based on analysis of the measured azimuthal asymmetries in semi - inclusive scattering and those in @xmath3 processes .
since @xmath2 is a spin dependent pdf , it is interesting to investigate whether there is a `` transversity crisis '' similar to the case of `` spin crisis '' in @xmath1 .
therefore , reliable determination of nucleon tensor charge receives special attention .
theoretically , tensor charges of hadrons are studied in different frameworks such as , non
relativistic mit bag model @xcite , su(6 ) quark model @xcite , quark model with axial vector dominance @xcite , lattice qcd @xcite , external field @xcite and three point versions of qcd sum rules @xcite . in the present work ,
using the most general form of the nucleon interpolating field , we study the tensor form factors of nucleons within light cone qcd sum rules ( lcqsr ) .
the lcqsr is based on the operator product expansion ( ope ) over twist of the operators near the light cone , while in the traditional qcd sum rules , the ope is performed over dimensions of the operators . this approach has been widely applied to hadron physics ( see for example @xcite ) .
note that , the tensor form factors of nucleons up to @xmath4 ( where @xmath5 is the euclidean momentum transfer square ) within the @xmath6 chiral soliton model are studied in @xcite ( see also @xcite ) .
the anomalous tensor form factors are studied within the same framework in @xcite .
these form factors are further studied in lattice qcd ( see for instance @xcite ) .
the plan of this paper is as follows . in section 2 ,
we derive sum rules for the tensor form factors of the nucleon within lqcsr method . in section 3 ,
we numerically analyze the sum rules for the tensor form factors .
a comparison of our results on form factors with those existing in the literature is also presented in this section .
this section is devoted to derivation of lcqsr for the nucleon tensor form factors .
the matrix element of the tensor current between initial and final nucleon states is parametrized in terms of four form factors as follows @xcite : [ enucten01 ] n(p^ ) |q _ q n(p ) |u(p^ ) \ { h_t ( q^2 ) i _ - e_t(q^2 ) _
q_2 m_n + e_1t(q^2 ) _ _ - _ _ 2 m_n + _ t ( q^2 ) p_q_- p_q_2 m_n^2 } u(p ) , where @xmath7 , @xmath8 , and @xmath9 . from t
invariance it follows that @xmath10 . in order to calculate the remaining three tensor form factors within lcqsr
, we consider the correlation function , [ enucten02 ] _
( p , q ) = i d^4x e^iqx 0 t\{j^n(0 ) j _ ( x ) } n(p ) .
this correlation function describes transition of the initial nucleon to the final nucleon with the help of the tensor current .
the most general form of the nucleon interpolating field is given as , [ enucten03 ] j^n(x ) = 2 ^abc _
i=1 ^ 2 a_2^i q^c ( x ) , where @xmath11 is the charge conjugation operator , @xmath12 , @xmath13 , @xmath14 with @xmath15 being an arbitrary parameter and @xmath16 corresponds to the ioffe current and @xmath17 are the color indices . the quark flavors are @xmath18 , @xmath19 for the proton and @xmath20 , @xmath21 for the neutron .
the tensor current is chosen as , [ enucten04 ] j _
= |u _ u |d _ d , where the upper and lower signs correspond to the isosinglet and isovector cases , respectively . in order to obtain sum rules for the form factors , it is necessary to calculate the correlation function in terms of quarks and gluons on one side ( qcd side ) , and in terms of hadrons on the other side ( phenomenological side ) .
these two representations of the correlation function are then equated .
the final step in this method is to apply the borel transformation , which is needed to suppress the higher states and the continuum contributions . following this strategy
, we start to calculate the phenomenological part . saturating the correlation function with a full set of hadrons carrying the same quantum numbers as nucleon and isolating the contributions of the ground state , we get [ enucten05 ] _
( p , q ) = 0 j^n(0 ) n(p^ ) n(p^ ) j _ n(p ) m_n^2-p^2 + , where dots stands for contributions of higher states and continuum . the matrix element @xmath22 entering eq .
( [ enucten05 ] ) is defined as [ enucten06 ] 0 j^n(0 ) n(p^ ) = _
n u(p ) , where @xmath23 is the residue of the nucleon . using eqs .
( [ enucten01 ] ) , ( [ enucten05 ] ) and ( [ enucten06 ] ) , and performing summation over spins of the nucleon , we get , [ enucten07 ] _ ( /p^_ + m_n ) \
{ h_t ( q^2 ) i _ - e_t ( q^2 ) _
_ q_2 m_n + _ t ( q^2 ) p_q_- p_q_2 m_n^2 } u(p ) . from eq .
( [ enucten07 ] ) we see that there are many structures , and all of them play equal role for determination of the tensor form factors of the nucleon . in practical applications ,
it is more useful to work with @xmath24 rather than @xmath25 .
for this reason , we choose the structures @xmath26 , @xmath27 and @xmath28 for obtaining the sum rules for the form factors @xmath29 , @xmath30 and @xmath31 , respectively .
the correlation function @xmath32 is also calculated in terms of quarks and gluons in deep eucledian domain @xmath33 .
after simple calculations , we get the following expression for the correlation function for proton case : [ enucten08 ] _ _
d^4x e^iqx _
i=1 ^ 2 \ { c a_1^i _ _ 4 ^abc 0 u_^a(0 ) u_^b(x ) d_^c(0 ) n ( p ) + a_2^i _ _ 4 ^abc 0 u_^a(x ) u_^b(x ) d_^c(0 ) n ( p ) + a_2^i _ _ 4 ^abc 0 u_^a(0 ) u_^b(0 ) d_^c(x ) n ( p ) } .
obviously , the correlation function for the neutron case can easily be obtained by making the replacement @xmath34 . from eq .
( [ enucten08 ] ) it is clear that in order to calculate the correlation function from qcd side , we need to know the matrix element , [ nolabel ] 4 ^abc 0 u_^a(a_1 x ) u_^b(a_2 x ) d_^a(a_3 x ) n(p ) , where @xmath35 , @xmath36 and @xmath37 determine the fraction of the nucleon momentum carried by the corresponding quarks .
this matrix element is the main nonperturbative ingredient of the sum rules and it is defined in terms of the nucleon distribution amplitudes ( das ) .
the nucleon das are studied in detail in @xcite .
the light cone expanded expression for the light quark propagator @xmath38 is given as , [ enucten09 ] s_q(x ) = - ( 1 + ) -
i g_s _ 0 ^ 1 dv , where the mass of the light quarks are neglected , @xmath39 @xcite and @xmath40 is the gluon field strength tensor . the terms containing @xmath40 give contributions to four and five particle distribution functions .
these contributions are negligibly small ( for more detail see @xcite ) , and therefore in further analysis , we will neglect these terms .
moreover , borel transformation kills the terms proportional to the quark condensate , and as a result only the first term is relevant for our discussion . using the explicit expressions of das for the proton and light quark propagators , performing fourier transformation and then applying borel transformation with respect to the variable @xmath41 , which suppresses the contributions of continuum and higher states , and choosing the coefficients of the structures @xmath26 , @xmath27 and @xmath42 , we get the following sum rules for the tensor form factors of nucleon : + + [ enucten10 ] h_t ( q^2 ) e^m_n^2/m^2 \ { _ x_0 ^ 1 dt_2t_2 e^-s(t_2)/m^2 + _
x_0 ^ 1 dt_3t_3 e^-s(t_3)/m^2 + _
x_0 ^ 1 dt_2t_2 e^-s(t_2)/m^2 + _
x_0 ^ 1 dt_3t_3 e^-s(t_3)/m^2 + _
x_0 ^ 1 dt_2t_2 e^-s(t_2)/m^2 ( 1-t ) f_h_t^9(t_2 ) + _
x_0 ^ 1 dt_3t_3 e^-s(t_3)/m^2 ( 1+t ) f_h_t^10(t_3 ) + _
x_0 ^ 1 dt_2t_2 ^ 2 e^-s(t_2)/m^2 + e^-s_0/m^2 + _
x_0 ^ 1 dt_3t_3 ^ 2 e^-s(t_3)/m^2 + e^-s_0/m^2 + _
x_0 ^ 1 dt_2t_2 ^ 2 e^-s(t_2)/m^2 + e^-s_0/m^2 + _
x_0 ^ 1 dt_3t_3 ^ 2 e^-s(t_3)/m^2 ( 1+t ) f_h_t^17(t_3 ) + e^-s_0/m^2 ( 1+t ) f_h_t^17(x_0 ) } , where [ ? ?
] f_h_t^1(t_2 ) _ 0 ^ 1-t_2 dt_1 \ { 2 m_n^2 t_2 ( t_1,t_2,1-t_1-t_2 ) + _ 1 ( t_1,t_2,1-t_1-t_2 ) } , + f_h_t^2(t_2 ) _ 0 ^ 1-t_2 dt_1 \ { m_n^2 t_2 ( t_1,t_2,1-t_1-t_2 ) + ( t_1,t_2,1-t_1-t_2 ) } , + f_h_t^3(t_3 ) _ 0 ^ 1-t_3 dt_1 1t_3 ( t_1,1-t_1-t_3,t_3 ) , + f_h_t^4(t_3 ) _ 0 ^ 1-t_3 dt_1 ( t_1,1-t_1-t_3,t_3 ) , + f_h_t^5(t_2 ) m_n^22 _ 1^t_2 d_0 ^ 1- dt_1 ( t_1,,1-t_1- ) , + f_h_t^6(t_2 ) m_n^2 _
1^t_2 d_0 ^ 1- dt_1 ( t_1,,1-t_1- ) , + f_h_t^7(t_3 ) m_n^2 2 _ 1^t_3 d_0 ^ 1- dt_1 ( t_1,1-t_1- , ) , + f_h_t^8(t_3 ) m_n^2 2 _
1^t_3 d_0 ^ 1- dt_1 ( t_1,1-t_1- , ) , + f_h_t^9(t_2 ) _
1^t_2 d_1^ d_0 ^ 1- dt_1 1 _ 6 ( t_1,,1-t_1- ) , + f_h_t^10(t_3 ) m_n^2 _ 1^t_3 d_1^ d_0 ^ 1- dt_1 1 _ 6 ( t_1,1-t_1- , ) , + f_h_t^11(t_2 ) 2 m_n^2 _ 0 ^ 1-t_2 dt_1 q^2+m_n^2 t_2 ^ 2t_2 _ 1^m ( t_1,t_2,1-t_1-t_2 ) , + f_h_t^12(t_2 ) m_n^2 _ 0 ^ 1-t_2 dt_1 q^2+m_n^2 t_2 ^ 2t_2 ( t_1,t_2,1-t_1-t_2 ) , + f_h_t^13(t_3 ) m_n^2 _ 0 ^ 1-t_3 dt_1 q^2 t_3 ( t_1,1-t_1-t_3,t_3 ) , + f_h_t^14(t_3 ) m_n^2 _ 0 ^ 1-t_3 dt_1 2 q^2-m_n^2 t_3 ^ 2t_3 _ 1^m ( t_1,1-t_1-t_3,t_3 ) , + f_h_t^15(t_2 ) _ 1^t_2 d _ 1^ d_0 ^ 1- dt_1 1(t_1,,1-t_1- ) , + [ ? ? ] f_h_t^16(t_2 ) 2 m_n^4 ( 1 + t ) _ 1^t_2 d _ 1^ d_0 ^ 1- dt_1 ( t_1,,1-t_1- ) , + f_h_t^17(t_3 ) m_n^2 _
1^t_3 d_1^ d_0 ^ 1- dt_1 1 ( t_1,1-t_1- , ) . for the form factor
@xmath43 we obtain the following sum rule : [ enucten11 ] _ t ( q^2 ) e^m_n^2/m^2 \ { 1 m^2 _
x_0 ^ 1 dt_2t_2 ^ 2 e^-s(t_2)/m^2 ( 1-t ) f__t^1 ( t_2 ) + e^-s_0/m^2 ( 1-t ) f__t^1 ( x_0 ) + _
x_0 ^ 1 dt_2t_2 ^ 2 e^-s(t_2)/m^2 ( 1-t ) f__t^3(t_2 ) + e^-s_0/m^2 ( 1-t ) f__t^3(x_0 ) + _
x_0 ^ 1 dt_3t_3 ^ 2 e^-s(t_3)/m^2 ( 1-t ) f__t^4(t_3 ) + e^-s_0/m^2 ( 1-t ) f__t^4(x_0 ) + _
x_0 ^ 1 dt_2t_2 ^ 2 e^-s(t_2)/m^2 ( 1-t ) f__t^5 ( t_2 ) + e^-s_0/m^2 ( 1-t ) f__t^5(x_0 ) + _
x_0 ^ 1 dt_3t_3 ^ 2 e^-s(t_3)/m^2 ( 1-t ) f__t^6(t_3 ) + e^-s_0/m^2 ( 1-t ) f__t^6(x_0 ) + _
x_0 ^ 1 dt_2t_2 e^-s(t_2)/m^2 ( 1-t ) f__t^7 ( t_2 ) + _
x_0 ^ 1 dt_3t_3 e^-s(t_3)/m^2 ( 1-t ) f__t^8 ( t_3 ) } , where [ ? ?
] f__t^1(t_2 ) - 4 m_n^2 _
1^t_2 d_1^ d_0 ^ 1- dt_1 _
6(t_1,,1-t_1- ) , + f__t^3(t_2 ) - 4 m_n^2
_ 1^t_2 d_0 ^ 1- dt_1 ( t_1,,1-t_1- ) , + f__t^4(t_3 ) 4 m_n^2 _ 1^t_2 d_0 ^ 1- dt_1 ( t_1,1-t_1- , ) , + f__t^5 ( t_2 ) 8 m_n^2 _ 0 ^ 1-t_2 dt_1 _ 1^m ( t_1,t_2,1-t_1-t_2 ) , + f__t^6 ( t_3 ) 4 m_n^2 _ 0 ^ 1-t_3 dt_1 ( t_1,1-t_1-t_3,t_3 ) , + f__t^7 ( t_2 ) 8 _ 0 ^ 1-t_2 dt_1 _ 1 ( t_1,t_2,1-t_1-t_2 ) ) , + f__t^8 ( t_3 ) 4 _ 0 ^ 1-t_3 dt_1 ( t_1,1-t_1-t_3,t_3 ) .
finally for the form factor @xmath44 we get the following sum rule : [ enucten12 ] _
t(q^2 ) e^m_n^2/m^2 \ { 1m^2 _ x_0 ^ 1 dt_2t_2 ^ 2 e^-s(t_2)/m^2 ( 1-t ) f__t^1(t_2 ) +
1 q^2+x_0 ^ 2 m_n^2 e^-s_0/m^2 ( 1-t ) f__t^1(x_0 ) + _ x_0 ^ 1 dt_3t_3 ^ 2 e^-s(t_3)/m^2 ( 1-t ) f__t^2(t_3 ) e^-s_0/m^2 ( 1-t ) f__t^2(x_0 ) } , where [ ? ? ] f__t^1(t_2 ) 4 m_n _
1^t_2 d_0 ^ 1- dt_1 ( t_1,,1-t_1- ) , + f__t^2 ( t_3)4 m_n _ 1^ d_0 ^ 1- dt_1 ( t_1,1-t_1- , ) , and we use _ 2(t_i ) v_1(t_i ) - v_2(t_i ) - v_3(t_i ) , + _
2(t_i ) - a_1(t_i ) + a_2(t_i ) - a_3(t_i ) , + _
4(t_i ) - 2 a_1(t_i ) - a_3(t_i ) - a_4(t_i ) + 2 a_5(t_i ) , + _
5(t_i ) a_3(t_i ) - a_4(t_i ) , + _
6(t_i ) a_1(t_i ) - a_2(t_i ) + a_3(t_i ) + a_4(t_i ) - a_5(t_i ) + a_6(t_i ) , + _
2(t_i ) t_1(t_i ) + t_2(t_i ) - 2 t_3(t_i ) , + _
4(t_i ) t_1(t_i ) - t_2(t_i ) - 2 t_7(t_i ) , + _
5(t_i ) - t_1(t_i ) + t_5(t_i ) + 2 t_8(t_i ) , + _
6(t_i ) 2 , + _
7(t_i ) t_7(t_i ) - t_8(t_i ) , + _
2(t_i ) s_1(t_i ) - s_2(t_i ) , + _
2(t_i ) p_2(t_i ) - p_1(t_i ) , in these expressions , we also use [ nolabel ] f(x_i ) ( x_1,x_2,1-x_1-x_2 ) , + f(x_i^ ) ( x_1,1-x_1-x_3,x_3 ) , + s(x , q^2 ) ( 1-x ) m_n^2 + ( 1-x)x q^2 , where @xmath45 is the solution to the equation @xmath46 .
the residue @xmath23 is determined from two point sum rule . for the general form of the interpolating current
, it is calculated in @xcite , whose expression is given as [ enucten13 ] _
\ { e_2(x ) ( 5 + 2 t + t^2 ) - + transformations are implemented by the following subtraction rules @xcite , [ e8315 ] dx ( x ) ( q - xp)^2 & & - ( x ) e^-s(x)/m^2 , + dx ( x)(q - xp)^4 & & 1m^2 ( x ) e^-s(x)/m^2 + ( x_0)q^2+x_0 ^ 2 m_n^2e^-s_0/m^2 , + dx ( x)(q - xp)^6 & & - 1 2 m^2 ( x ) e^-s(x)/m^2 - 12 ( x_0 ) x_0 ( q^2 + x_0 ^ 2 m_n^2 ) m^2 e^-s_0/m^2 + & & + 12 x_0 ^ 2 q^2 + x_0 ^ 2 m_n^2 \
{ ddx_0 } e^-s_0/m^2 .
in this section , numerical results of the tensor form factors of nucleon are presented .
it follows from sum rules for the form factors that the main input parameters are the das of nucleon , whose explicit expressions and the values of the parameters @xmath47 , @xmath48 , @xmath49 , @xmath50 , @xmath51 , @xmath52 , and @xmath53 in the das are all given in @xcite . in the numerical analysis ,
we use two different sets of parameters : \a ) all eight nonperturbative parameters @xmath47 , @xmath48 , @xmath49 , @xmath50 , @xmath51 , @xmath54 , @xmath52 and @xmath53 are estimated from qcd sum rules ( set 1 ) .
\b ) requiring that all higher conformal spin contributions vanish , fixes five @xmath52 , @xmath53 , @xmath50 , @xmath51 , and @xmath55 , and the values of the parameters @xmath47 , @xmath48 , @xmath49 are taken from qcd sum rules .
this set is called asymptotic set or set2 .
the values of all eight nonperturbative parameters ( see for example @xcite ) are presented in table 1 , @xmath56 the next input parameter of the lcqsr for the tensor form factors is the continuum threshold @xmath57 .
this parameter is determined from the two point sum rules whose value is in the domain @xmath58 .
the sum rules contain also two extra auxiliary parameters , namely borel parameter @xmath59 and the parameter @xmath15 entering the expression of the interpolating current for nucleon .
obviously , any physical quantity should be independent of these artificial parameters .
therefore , we try to find such regions of @xmath59 and @xmath15 , where the tensor form factors are insensitive to the variation of these parameters .
firstly , we try to obtain the working region of @xmath59 , where the tensor form factors are independent of it , at fixed values of @xmath57 and @xmath15 . as an example , in figs .
( 1 ) and ( 2 ) we present the dependence of the tensor form factor @xmath60 induced by the isoscalar current on @xmath59 at different fixed values of @xmath61 and @xmath15 , and at @xmath62 and @xmath63 for sets 1 and 2 , respectively . from these figures , we see that @xmath60 is practically independent of @xmath59 at fixed values of the parameters @xmath61 , @xmath57 and @xmath15 for both sets 1 and 2 .
our calculations also show that the results are approximately the same for two sets , therefore in further discussion , we present the results only for set 1 .
we perform similar analysis also at @xmath64 and observe that the results change maximally about 5% .
the upper limit of @xmath59 is determined by requiring that the series of light cone expansion with increasing twist converges ,
i.e. , higher twist contributions should be small .
our analysis indeed confirms that the twist4 contributions to the sum rules constitute maximally about 8% of the total result when @xmath65 .
the lower bound of @xmath59 is determined by requiring that the contribution of the highest power of @xmath59 is less than , say , 30% of the higher powers of @xmath59 .
our numerical analysis shows that this condition is satisfied when @xmath66 .
hence , the working region of @xmath59 is decided to be in the interval @xmath67 .
the working region of the parameter @xmath15 is determined in such a way that the tensor form factors are also independent of it .
our numerical analysis shows that the form factors are insensitive to @xmath68 ( with @xmath69 ) when it varies in the region @xmath70 . in figs .
( 3)(5 ) we present the dependence of the form factors @xmath60 , @xmath71 and @xmath44 on @xmath61 at @xmath62 , @xmath72 and fixed values of @xmath15 , respectively , using the central values of all input parameters in set 1 for the isoscalar current . for a comparison
, we also present the predictions of self consistent chiral soliton model @xcite and lattice qcd calculations @xcite in these figures ( note that , chiral soliton model result exists only for @xmath60 ) .
we see from fig .
( 3 ) that , our results on @xmath60 are close to the lattice qcd results for @xmath73 , while the results of two models differ from each other in the region @xmath74 . our and lattice qcd results differ considerably from the predictions of the chiral soliton model .
it also follows from these figures that the form factors get positive ( negative ) at negative ( positive ) values of the parameter @xmath15 . in figs .
( 6 ) , ( 7 ) and ( 8) , we present the dependence of the form factors @xmath60 , @xmath71 and @xmath44 for the isovector current , i.e. , for the @xmath75 current .
our observations for set 1 can be summarized as follows : * the @xmath61 dependence of @xmath60 is similar to the isoscalar current case , but the values are slightly larger compared to the previous case . *
similar to the isoscalar case , the form factors @xmath60 and @xmath71 get positive ( negative ) at negative ( positive ) values of the parameter @xmath15 . * in contrary to the isoscalar current case , the values of @xmath44 are positive ( negative ) for negative ( positive ) values of @xmath15 . * our final remark is that the lcqsr results on the form factors can be improved by taking into account the @xmath76 corrections . in conclusion , using the most general form of the nucleon interpolating current , we calculate the tensor form factors of nucleon within the lcqsr .
our results on these form factors are compared with the lattice qcd and chiral soliton model predictions .
+ + after completing this work , we become aware of a very recent paper arxiv:1107.4584 [ hep - ph ] @xcite in which part of this work is studied .
we thank p. hgler for providing us the lattice qcd data .
m. el beiyad , b. pire , m. segond , l. szymanowski , s. wallon , phys .
b * 688 * , 154 ( 2010 ) .
r. enberg , b. pire , l. szymanowski , eur .
j. c * 47 * , 87 ( 2006 ) .
d. y. ivanov , b. pire , l. szymanowski , o. v. teryaev , phys .
b * 550 * , 65 ( 2002 ) .
i. i. balitsky and v. m. braun , nucl .
b * 311 * , 541 ( 1989 ) ; v. m. braun , arxiv : hep ph/9801222 ( 1998 ) ; p. colangelo , a. khodjamirian , in `` at the frontier of particle physics / handbook of qcd '' ed . by m. shifman ( worl scientific , singapore , 2001 ) , volume * 3 | using the most general form of the nucleon interpolating current , we calculate the tensor form factors of the nucleon within light cone qcd sum rules . a comparison of our results on tensor form factors with those of the chiral
soliton model and lattice qcd is given .
0^*0 5_5 o _ ^0 _ pacs number(s ) : 11.55.hx , 14.20.dh |
thus far , the precise measurements of galaxy clustering at @xmath6 that are necessary for meaningful physical interpretation have only been possible for the relatively blue , star - forming galaxies that dominate optical surveys @xcite .
however the most massive galaxies at these redshifts tend to be faint in the optical , and are more appropriately selected in the near - infrared ( e.g. * ? ? ?
in particular , galaxies meeting the @xmath7 criterion for distant red galaxies ( drgs ; * ? ? ?
* ) have been shown to dominate the high - mass end of the galaxy mass function @xcite .
measurements of drg clustering have been severely hampered by small fields and by the near - complete reliance on photometric redshifts . using the largest drg sample then available , @xcite confirmed the very strong clustering found by previous authors @xcite .
they also showed that strong clustering implies that drgs occupy dark matter halos with @xmath8 .
however , the number density of these dark matter halos is at least an order of magnitude smaller than the number density of drgs , suggesting that the drg clustering measurements are incompatible with models of dark matter clustering .
here we use a @xmath90.65@xmath10 field from the ukirt infrared deep sky survey ( ukidss ) , which is @xmath98 times larger than the area used by @xcite but with similar nir depth , to determine whether the strong observed clustering of drgs found in previous studies was an artifact due to limited field sizes or whether some other explanation must be found .
we use @xmath11 .
small changes in these parameters do not affect our basic conclusions .
magnitudes are given in the vega system , except where noted .
the ukidss project covers different areas to different depths ; here we make use of the deepest ukidss dataset , known as the ultra deep survey ( uds ) .
we use the uds data release 1 images @xcite , which reach 5 @xmath12 point - source depths of @xmath13 and @xmath14 .
we note that @xcite used the uds early data release to study the clustering of drgs down to @xmath15 , however such bright drgs lie primarily at @xmath16 and may not be directly relevant to the galaxies that are the subject of this work .
most of the uds field has coverage in the optical bands from the subaru - xmm deep survey ( sxds ) .
we use the beta - release of the @xmath17 images , which reach depths of @xmath925.327.5 ( ab magnitudes ) .
finally , we combine these data with @xmath18 and @xmath19 imaging from the _ spitzer _ wide - area infrared extragalactic survey ( swire ; * ? ? ? * ) .
the procedures used to create a multicolor @xmath20-selected catalog from these imaging data are detailed by @xcite .
accurate redshift information is fundamentally important for clustering studies .
photometric redshifts were calculated using the eazy software @xcite .
eazy fits linear combinations of galaxy templates to the observed photometry using a bayesian prior for the distribution of redshifts as a function of apparent magnitude .
the template set was carefully chosen to provide high - quality photometric redshifts for @xmath20-selected galaxies in several current deep surveys . to assess the accuracy of our photometric redshifts we compare to a sample 119 spectroscopic redshifts available in this field ( for further description ,
see * ? ? ?
the normalized median absolute deviation of @xmath21 is @xmath22 . as nearly all spectroscopic redshifts in this field
are at @xmath23 , and none are for drgs , we use public data on the well - studied chandra deep field - south ( cdf - s ) to obtain an estimate of the expected photometric redshift accuracy for the galaxies currently under consideration .
we use the photometric catalog of @xcite , and exclude the @xmath24 filters in order to match the filter set available in the uds . restricting the sample to the drgs in cdf - s with spectroscopic redshifts , we find @xmath25 , with no apparent dependence on drg redshift and negligible systematic offsets .
this suggests that the drg photometric redshifts in our study also have an accuracy of @xmath26 , although this can only be taken as a rough indication .
the overall redshift distribution of the sample is used to deproject the observed angular correlation function in order to infer the spatial correlation function .
this distribution is often estimated simply using a histogram of the photometric redshifts , without taking into account the redshift uncertainties .
because we select drgs in a specific photometric redshift range , it is expected that random errors will cause galaxies to scatter into the sample from both lower and higher redshifts .
it is also expected that galaxies within the redshift selection window will scatter out , however this will not affect the clustering so long as such galaxies are drawn randomly from the sample ( systematic errors which cause galaxies to scatter into or out of the selection window are much more problematic ) . to estimate the _ intrinsic _ redshift distribution of the galaxies in our sample , we use monte carlo simulations
in which the observed photometry is perturbed according the the photometric uncertainties ( see also * ? ? ?
this procedure broadens the distribution relative to what would be inferred from a histogram of photometric redshifts , resulting in a higher correlation length ; we return to this point in [ sec : discussion ] .
for a detailed description of the techniques used to perform correlation function measurements , see @xcite . here
we briefly describe the method , and point out differences between the method used here and in the previous paper .
the angular two - point correlation function is calculated using the @xcite estimator .
uncertainties are estimated using bootstrap resampling , which yields uncertainties that are significantly larger than those expected from poisson statistics .
we follow the method of @xcite to correct for the integral constraint , although the correction is small given the large size of the field .
a potentially serious problem for measurements of the auto - correlation function arises from variable sensitivity or calibration errors across the images .
we measured the correlation function of stars in order to test whether such variations induce artificial clustering , obtaining the expected result that stars are unclustered .
we have also measured the clustering of drgs selected from catalogs in which the @xmath27 and @xmath20 zeropoints are varied within their uncertainties independently for each of the pointings that make up the uds mosaic .
although it is difficult to rule out significant errors in the correlation function arising from such variations , it appears that they are not a dominant source of error .
the redshift distribution of galaxies selected using the drg color criterion is rather wide , extending from @xmath28 to at least @xmath29 ( e.g. * ? ? ?
* ; * ? ? ?
we therefore impose the additional criterion @xmath1 . over this redshift range
, the drg criterion effectively selects galaxies with red rest - frame optical colors @xcite .
we also limit the sample to objects with @xmath2 , ensuring that the catalog is complete and that objects have a sufficient signal - to - noise ratio to calculate accurate photometric redshifts .
we visually inspected each object in the sample , rejecting those that appeared to be artifacts in the images or were otherwise deemed to have unreliable photometry .
objects were also rejected if the templates used for photometric redshifts gave poor fits .
the final sample consists of 1528 drgs with @xmath2 .
figure [ fig : acf ] shows the angular correlation function @xmath30 of drgs .
our measurements confirm the strong angular clustering previously reported by @xcite and @xcite , but with reduced uncertainties and extending to much larger angular scales .
immediately apparent is the departure of @xmath30 from a power law on small scales . to illustrate the significance of this excess clustering signal , and to show how the results differ from what would be measured using a smaller field , we fit a power law @xmath31 over the range @xmath32 and over @xmath33 .
the former fit yields @xmath34 while the latter fit yields @xmath35 .
note that there is significant covariance between the two fitting parameters , so the uncertainties should be taken as representative only . as noted by previous authors
@xcite , the large - scale fit is preferred when comparing galaxy samples , as it better probes the clustering of dark matter halos that host the galaxies .
the bottom panel of figure [ fig : acf ] shows the residuals @xmath30 from the large - scale power law .
a small - scale excess has also been observed in large samples of optically selected galaxies at high redshift ( e.g. * ? ? ?
* ; * ? ? ?
* ) , and for various galaxy samples at @xmath36 ( e.g. * ? ? ?
. the spatial correlation length @xmath37 can be estimated from @xmath30 using the limber projection ; see e.g. @xcite for further details .
the result is @xmath38 , where we have used the large - scale fitting parameters and the uncertainty is estimated using the bootstrap simulations described above .
this uncertainty accounts for the uncertainty in @xmath39 , but not in the redshift distribution @xmath40 .
we use the method of @xcite to estimate the linear bias directly from @xmath30 , finding @xmath41 .
@xcite found that drgs significantly outnumber the dark matter halos that are clustered strongly enough to host them .
models of the halo occupation distribution ( hod ) naturally account for this type of discrepancy by allowing halos to host multiple galaxies , but @xcite suggest that the large numbers of galaxies that would be required to share halos can be ruled out by the observed small - scale clustering .
given that the drg clustering results presented in this work are largely consistent with those from previous works , it is expected that the observations are still in conflict with the models . in this section
we use an hod model to show this explicitly . in the hod framework , the galaxy correlation function is understood to be the sum of two components . on large scales the correlation function follows that of the host dark matter halos . on small scales
there is an additional contribution from galaxy pairs within individual halos .
we follow the modeling procedures described by @xcite , to which we refer the reader for details . briefly
, we calculate the number density and bias of halos using the prescriptions of @xcite .
the halo occupation number , which describes the number of galaxies per halo , is parameterized as @xmath42 for halo mass @xmath43 and @xmath44 otherwise .
thus there are three free parameters , @xmath45 , @xmath46 , and @xmath47 .
if the number density @xmath48 of galaxies is known , then one of these parameters can be fixed for assumed values of the other two parameters .
we obtain a rough estimate of @xmath49 using the effective volume probed by our sample , which is determined using the redshift selection function described in [ sec : zphots ] .
this density is in reasonable agreement with an independent estimate of @xmath50 , which is based on the luminosity functions of @xcite .
figure [ fig : hod ] shows two models chosen according to @xmath51 fits . the lower solid line shows a model that is fit over @xmath52 . while this model provides an adequate fit on smaller angular scales , it systematically under - predicts the clustering on larger scales . to better illustrate the nature of the disagreement ,
it is useful to inspect the upper solid line , which shows a model that is fit only over @xmath53 .
while this model provides an adequate fit at larger scales , the small - scale fit is unacceptable .
as already noted , the fundamental reason that no model can fit the data is that the strong clustering on large scales implies that drgs must occupy very massive halos .
but drgs outnumber these halos by a factor of @xmath920 , which is only possible if individual halos host a large number of drgs .
this would mean that each drg has a high probability of having several neighbors in the immediate vicinity , leading to a very prominent small - scale excess in @xmath30 .
, which we parameterize following @xcite .
note that this particular choice does not affect the main result of this section because , for the high @xmath54 value found here , it is generically expected that @xmath55 follows a poisson distribution for a fixed @xmath56 ( e.g. * ? ? ?
* ) ; this fully specifies the moments . ] as can be seen , the observed excess is much smaller than expected .
we have used the ukidss - uds to perform the first precise measurement of the clustering of red , @xmath20-selected galaxies at @xmath57 .
these drgs show strong angular clustering that is well - described by a power law , but with an excess at small scales .
we use photometric redshifts to deproject the angular clustering , finding the spatial correlation length @xmath38 .
this value is comparable to that measured for luminous red galaxies in the local universe @xcite , however drgs are significantly more numerous .
we show that standard models of halo occupation statistics are unable to simultaneously reproduce the observed clustering and number density , because drgs outnumber their inferred host dark matter halos by too large a margin .
the most obvious explanation is that we have used the incorrect redshift distribution in deprojecting the angular correlation function .
a narrower distribution would reduce the correlation length ( while a moderate shift in the overall distribution makes a relatively smaller difference ) . however , a narrower redshift distribution would also decrease the effective volume probed by our sample , thereby increasing @xmath48 .
we illustrate these effects in figure [ fig : r0_n ] , which shows the observed @xmath37 and @xmath48 compared to the range of values for a typical hod model .
it also shows how estimating @xmath40 directly from the unperturbed photometric redshifts which , as mentioned in
[ sec : zphots ] , represents the extreme assumption of no random photometric redshift errors affects the results
. it may still be the case that we are subject to _ systematic _ redshift errors , but we also note that a significantly narrower @xmath40 would adversely affect the reasonably good agreement between our estimate of @xmath48 and the luminosity functions derived by @xcite .
finally , we have verified that our basic results hold when using a different photometric redshift code ( hyperz * ? ? ?
* ) and with a different template set @xcite
. given the apparently high quality of our photometric redshifts , as well as the consistency with previous results for the clustering of drgs , it is worth considering alternative explanations .
one possibility is that current hod models are too simplistic , and that massive red galaxies occupy halos in unexpected ways .
the fundamental assumption underlying these models is that galaxy observables depend on halo mass , and not on the larger - scale environment or on halo properties such as structure or age . but environment may play a role , for instance via its effect on mass accretion rates @xcite . additionally , halo clustering varies with several halo properties , even at fixed mass ; this phenomenon is generically referred to as `` assembly bias '' ( e.g. * ? ? ?
to the extent that these properties affect galaxy observables , they will also affect galaxy clustering measurements .
however , we note that current estimates for the strength of the assembly bias are too small to account for the observed discrepancies . as an example , @xcite have found that , at @xmath58 and in the relevant halo mass range , halos in the upper 20@xmath59 tail of the distribution of halo spins have a @xmath920@xmath59 larger bias than the mean value .
it might then be supposed that drgs occupy less massive and more numerous halos with higher spin . but the increased number density of these low mass halos is approximately cancelled by the requirement that drgs can only occupy 20@xmath59 of them , so the discrepancy in number densities is unchanged .
the existence of massive red galaxies at @xmath60 was not predicted by models of galaxy formation , although progress has been made on this front ( e.g. * ? ? ?
* ; * ? ? ?
* but see @xcite ) .
the results shown here suggest that the conflict between models and observations extends to the relationship between galaxies and dark matter halos . however , clustering measurements are sensitive to a number of systematic effects , so our conclusions remain tentative .
the most obvious source of error comes from our use of photometric redshifts .
ongoing medium - band nir observations will significantly reduce the photometric redshift uncertainties @xcite , and in the longer term multi - object nir spectrographs will also improve the situation .
if future work confirms our results , clustering measurements such as those presented here may provide a new way to understand the detailed relationship between galaxy and halo properties .
we thank danilo marchesini , qi guo , simon white , and chuck steidel for useful discussions , as well as the anonymous referee for a constructive report .
this work is based data made public by ukidss , sxds , and swire teams .
is supported by a nova postdoctoral fellowship .
support from national science foundation grant career ast-0449678 is also gratefully acknowledged . | recent studies have shown that distant red galaxies ( drgs ) , which dominate the high - mass end of the galaxy population at @xmath0 , are more strongly clustered than the population of blue star - forming galaxies at similar redshifts . however these studies have been severely hampered by the small sizes of fields having deep near - infrared imaging .
here we use the large ukidss ultra deep survey to study the clustering of drgs . the size and depth of this survey
allows for an unprecedented measurement of the angular clustering of drgs at @xmath1 and @xmath2 .
the correlation function shows the expected power law behavior , but with an apparent upturn at @xmath3 .
we deproject the angular clustering to infer the spatial correlation length , finding @xmath4 .
we use the halo occupation distribution framework to demonstrate that the observed strong clustering of drgs is not consistent with standard models of galaxy clustering , confirming previous suggestions that were based on smaller samples .
inaccurate photometric redshifts could artificially enhance the observed clustering , however significant systematic redshift errors would be required to bring the measurements into agreement with the models .
another possibility is that the underlying assumption that galaxies interact with their large - scale environment only through halo mass is not valid , and that other factors drive the evolution of the oldest , most massive galaxies at @xmath5 . |
a single - walled carbon nanotube is a two dimensional graphene sheet rolled into a cylinder .
the diameter of the nanotube is on the order of a few nanometers , and its length can vary from hundreds of nanometers to centimeters . due to the small tube radius
, electrons are confined in the azimuthal direction , and at sufficiently low energy the quantum confinement leads to an effectively one - dimensional electronic system .
these nanotubes can be either metallic or semiconducting , and the low - energy electronic band structure can be studied using a long - wavelength expansion of the hamiltonian around each of the degenerate fermi points , labeled by k and k points .
this long - wavelength theory is given by a pair of one dimensional dirac hamiltonians .
when a nanotube is chemically functionalized or contains defects on the tube wall , localized scattering centers interrupt the free motion of its low - energy charge carriers .
generally a localized defect can backscatter a propagating low - energy electron , either by large momentum scattering between the k and k valleys or by small momentum backscattering from forward to backward moving excitations within a single valley .
superposition of right and left moving excitations produces various standing wave patterns in the electron density near such a defect .
in this paper we consider forces on the scatterers produced by their interaction .
it is easy to see that for an isolated scatterer , the backscattering - induced forces on the left and right hand side of the defect must exactly cancel , so there is no net force .
however , for pairs of defects and generally for a distribution of defects at finite density , the forces on the left and right hand sides of the scatterer do not balance and mediate a net force on each scatterer . in previous work we explored this effect within a single valley model for the nanotube and found that the scattering induced forces could be mapped to a casimir - type problem , where the propagating electron waves provide the role of the background quantum field .
importantly , the spinor character of these background fermions admits the possibility of attractive , repulsive or compensated null forces on the scatterers depending on the internal symmetry of their scattering potentials [ ] . in this work
we generalize these earlier results to study the combined effects of intravalley and intervalley backscattering .
this extension proves to be crucial for a meaningful application to the nanotube problem .
potentials that produce only intravalley scattering need to vary slowly on the scale of a lattice spacing . yet ,
any local potential with this property degenerates to a one - dimensional scalar potential that can not backscatter a massless dirac particle .
thus , for a local potential our effect ultimately requires a significant degree of spatial localization , and in this regime intervalley backscattering ultimately arises .
indeed , we find below that for local potentials there is no regime in which the force problem can be regarded as confined to a single valley , necessitating a coupled valley formulation of the scattering problem .
by contrast , non - local scattering potentials do allow the possibility of only intravalley backscattering in a controlled physically realizable limit .
this situation is realized most naturally for electrons coupled to slowly varying lattice strains on a nanotube . in this paper
we present a generalization of the formalism described in ref . [ ] suitable for application to the coupled two valley problem , and explore the forces that occur as a function of range and internal symmetry of the scattering potentials .
we provide formulae that describe the electron mediated forces in these various geometries .
table i provides a compact summary of our results .
the magnitude and sign of the interaction is dictated by the internal structure of the scatterers .
local potentials can describe atomically sharp impurities localized on a sublattice site .
we find a repulsive force between local impurities residing on equivalent sublattice sites and an attractive force between scatterers on distinct sites .
related results were recently shown for interactions between impurities in two - dimensional graphene [ ] .
we also explore interactions between impurities where only intervalley scattering is present .
interactions between defects due to large momentum backscattering were previously discussed in one - dimensional fermi liquids [ , ] . for non - local potentials
we show that scattering persists for ranges that are larger than the lattice constant leading to the single - valley scattering problem .
the results we obtain for casimir forces between non - local scatterers agree with our previous work .
we recover the universal distance dependent power law decay for the casimir force in one - dimension .
however , for local potentials , unlike for the single - valley problem , we also observe periodic spatial modulations in the force due to intervalley scattering .
the paper is organized in the following manner . in sec .
[ sec : ho ] we define the geometry and derive the low - energy electronic structure of single - walled carbon nanotubes . in sec .
[ sec : h1 ] we present scattering potentials which can describe impurities in nanotubes .
the distinction between relevant length scales is discussed in sec . [
sec : range ] . in sec .
[ sec : local ] and sec .
[ sec : non - local ] we discuss local and non - local potentials , respectively . in sec .
[ sec : force ] we outline the basic mechanism used to calculate casimir forces . in sec .
[ sec:1valley ] we review our previous work of the one - valley problem , and show how the method is generalized to the two - valley problem in sec .
[ sec:2valley ] .
our main results are presented in sec .
[ sec : results ] .
casimir forces between local and non - local potentials are shown in sec .
[ sec : flocal ] and sec .
[ sec : flocal ] , respectively . in sec .
[ sec : dis ] we discuss the relation of our findings to physical adsorbates on nanotubes .
the paper is concluded in sec .
[ sec : conclusion ] .
in this section we describe the geometric structure of a carbon nanotube and introduce the notation used in this paper .
two - dimensional graphene is a honeycomb lattice with two inequivalent sublattice sites , labeled @xmath0 and @xmath1 , as illustrated in fig .
[ fig : lattice ] .
there is one carbon atom residing on each lattice site .
the primitive lattice vectors are @xmath2 and @xmath3 , where @xmath4 is the nearest neighbor bond length .
the vectors @xmath5 s define a triad of nearest neighbor bond vectors as shown in fig .
[ fig : lattice ] .
a carbon nanotube is formed by wrapping the graphene sheet into a cylinder , such that two equivalent lattice sites are identified .
the circumferential vector @xmath6 , where @xmath7 , characterizes the nanotube .
the @xmath8-plane defines the lattice coordinate system , where bonds run parallel to the @xmath9-axis .
the tube coordinate system is defined by @xmath10 along the tube axis and @xmath11 around the circumference .
the two coordinate systems are related by the tube s chiral angle defined as the angle between @xmath12 and @xmath10 as shown in fig .
[ fig : lattice ] .
the coordinate transformation is given by @xmath13 the circumference vectors of high - symmetry achiral nanotubes that have a plane of mirror symmetry are shown in fig .
[ fig : lattice ] . in armchair ( @xmath14 ) and zigzag ( @xmath15 ) carbon nanotubes bonds run parallel to the tube s circumference and axis , respectively . fixing the origin on an @xmath0 site , the lattice translation vector @xmath16 , where @xmath17 , locates an @xmath0 sublattice site , and the vector @xmath18 locates a @xmath1 site , where @xmath19 is a vector connecting the two sublattice sites .
the lattice vectors in the nanotube coordinate system are given by @xmath20\hat{x}_{\tiny{\mbox{$\|$}}}\nonumber \\
& + \frac{a}{2}\big[-\sin\theta\big(n_1-n_2\big)+\sqrt{3}\cos\theta\big(n_1+n_2+\frac{2b}{3}\big)\big]\hat{x}_{\tiny{\mbox{$\bot$ } } } , \label{eq : lattice.vec}\end{aligned}\ ] ] where @xmath21 for @xmath22 and @xmath23 for @xmath24 .
the nearest neighbor bond vectors @xmath25 s shown in fig .
[ fig : lattice ] in the tube coordinate system are given by @xmath26 where @xmath27 , and @xmath28 .
the first brillouin zone of the honeycomb lattice is shown in fig . [
fig : star ] . in graphene the conduction and valence bands touch at the six corner points of the brillouin zone .
therefore , for undoped graphene the fermi surface lies at the @xmath29 and @xmath30 points .
the three equivalent fermi points identified by white and black circles in fig . [ fig : star ] are related by reciprocal lattice vectors @xmath31 .
@xmath29 and @xmath30 points are inequivalent since they can not be connected through a reciprocal lattice vector . in the nanotube
coordinate system the six corners of the brillouin zone are given by @xmath32 where @xmath33 for @xmath34-points , and @xmath35 . as shown in fig .
[ fig : star ] , the corner point @xmath36 is a reference defining the chiral angle @xmath37 between the lattice @xmath12-axis and the tube axis .
points along the tube axis .
the six corners of the brillouin zone are shown .
the three equivalent @xmath29 ( black circles ) and @xmath30 ( white circles ) points are related by reciprocal lattice vectors @xmath31 .
the chiral angle @xmath37 is defined as the angle between the tube axis and @xmath38 in the lattice coordinate system.,width=268 ] the energy band structure of graphene can be obtained using a tight - binding model for @xmath39 electrons .
considering nearest neighbor hopping between sites on a two - dimensional honeycomb lattice , the tight - binding hamiltonian for graphene is given by @xmath40 where @xmath41 is the nearest neighbor hopping energy , and @xmath42 creates an electron on the a(b ) sublattice .
the low - energy electronic properties are found by expanding the tight - binding hamiltonian around the two distinct @xmath34-points to linear order in momentum @xmath43 .
since the fermi points are inversely proportional to the lattice constant @xmath44 , the long - wavelength theory is valid for @xmath45 .
the energy spectrum of a carbon nanotube is obtained from the graphene hamiltonian by rotating to the tube coordinate system and quantizing the crystal momentum along the transverse direction .
single - walled carbon nanotubes are either metallic or semiconducting depending on whether the discrete lines of crystal momentum pass through the fermi points @xmath29 and @xmath30 .
it turns out that @xmath46 of all nanotubes are metallic , since mod@xmath47 is a necessary condition for the six corners of the brillouin zone to be allowed wave vectors . in our notation ,
the @xmath48 identity and pauli matrices @xmath49 span @xmath50-sublattice pseudospin space , and @xmath51 span the @xmath34-point valley isospin space , where @xmath52 . for simplicity , we introduce an operator which defines a rotation by an angle @xmath53 around @xmath54 in either @xmath55 or @xmath56 space . for example , in @xmath56 space this operator is given by @xmath57 it is convenient to define a projection operator @xmath58 which projects on a sublattice site .
likewise , @xmath59 is a projection operator in the valley space . in this paper
, we only consider the lowest energy band of metallic tubes ( gapless systems ) as will be explained in sec .
[ sec : h1 ] .
expanding @xmath60 around the brillouin zone corners and rotating to the tube coordinate system using eq .
( [ eq : euler ] ) , the long - wavelength hamiltonian for the lowest energy band of a metallic nanotube becomes @xmath61\partial_{{x_{\tiny{\mbox{$\|$}}}}}\nonumber \\ & -e\big)f_k({x_{\tiny{\mbox{$\|$ } } } } ) = 0 , \label{eq : ho}\end{aligned}\ ] ] where @xmath62 ev@xmath63 nm .
the basis states are four - component spinors defining relative amplitudes at the @xmath0 and @xmath1 sites and the @xmath29 and @xmath30 fermi points in the following order @xmath64 , where @xmath65 and @xmath66 correspond to one of the three equivalent @xmath29 and @xmath30 points , respectively , depicted in fig .
[ fig : star ] .
the eigenstates of @xmath60 , @xmath67 are right and left moving plane waves multiplied by a spinor , where @xmath68 is the momentum along the tube axis . when the chemical potential is fixed at @xmath69 the filled dirac sea has @xmath70 , and the right and left moving spinors are given by @xmath71 the eigenstates of the long - wavelength hamiltonian in eq .
( [ eq : ho ] ) are isotropic and do not depend on the crystal orientation of a nanotube .
to include lattice anisotropic potentials in the theory , we reconstruct the bloch functions from the solutions in eq .
( [ eq : ho ] ) for the effective mass theory . in the @xmath72 approximation
the electron wave function near the fermi energy is given by a bloch function at the @xmath29 point multiplied by an envelope function .
for graphene , the wave function is @xmath73 where @xmath74 s are exact bloch functions at the @xmath29 point , and @xmath75 s are slowly varying envelope functions [ ] .
bloch states are plane waves multiplying a cell periodic function .
potentials which resolve the lattice structure couple to the lattice periodic component of the bloch states . taking the fourier transform of the periodic part of the bloch function , the sublattice basis functions at any of the six corner points @xmath76
s are given by @xmath77 where @xmath78 is the fourier transform of a localized orbital function , @xmath79 s are reciprocal lattice vectors , and @xmath80 defined in eq .
( [ eq : kpoint ] ) labels the @xmath29 and @xmath30 points .
the subscript @xmath81 labels a sublattice site , such that @xmath82 and @xmath83 .
the functions in eq .
( [ eq : basis ] ) are rapidly oscillating and describe modulations on the scale of the atomic spacing . since @xmath78 decreases rapidly with momentum , in the lowest `` star '' approximation [ ] we keep terms in the sum of @xmath84 s which connect the three @xmath85 brillouin zone corners , such that @xmath86 .
this approximation is appropriate for the range of the scattering potentials we study in this paper .
the normalized basis functions at the @xmath0 and @xmath1 sites in the lowest `` star '' representation are given by @xmath87 where @xmath88 .
evaluating the matrix element of a tight - binding potential in the lowest `` star '' basis given in eq .
( [ eq : states.star ] ) and expanding to linear order in @xmath43 , one obtains the low - energy hamiltonian given in eq .
( [ eq : ho ] ) for the lowest band of a metallic nanotube .
in this paper , we study casimir interactions between two scatterers mediated by the conduction electrons of a carbon nanotube . in this section
we describe the structure of the scattering potentials used to study this problem .
we explore the dependence of the casimir interaction on the symmetry , range , strength , and orientation of the two potentials .
we discuss two types of potentials , local and non - local , which result in different scattering processes . in our previous work we studied the one - valley scattering problem valid for potentials
whose range is larger than the lattice constant , where intervalley scattering does not play a role .
the @xmath48 matrix structure of such a potential is described by its pseudospin polarization [ ] .
when the range of the potential is on the order of interatomic spacing , the two valleys are no longer decoupled [ ] . in this paper , we build upon our previous work to incorporate the effects of sharper potentials resulting in a two - valley scattering problem .
when the two valleys are coupled , the potential is described by a @xmath89 matrix and is characterized by both pseudospin and valley polarizations .
in general , the spatial variation @xmath90 of the scattering potential relevant for casimir interactions is shorter than the conduction wavelength of the envelope function @xmath91 , such that @xmath92 .
[ fig : length ] shows an illustration of a scattering process . freely propagating electrons in regions i and iii have a wavelength @xmath93 , and the scattering region ii has a width @xmath90 .
a potential can be described by delta - function as long as @xmath94 .
the important distinction between the one- and two- valley scattering problems described by the spinor structure of the hamiltonian is relevant for potentials whose range is longer and shorter , respectively , than the interatomic separation . .
a scattering potential can be represented by a delta - function when @xmath90 is much smaller than @xmath91 , the wavelength of the envelope function.,width=288 ] we study interactions between scatterers in metallic nanotubes .
since casimir interactions mediated by massive fields are exponentially suppressed at long distances [ ] , in this paper we do not address semiconducting nanotubes or scattering between bands which do not pass through the fermi energy .
the momentum transfer in the azimuthal direction between various fermi points is determined by the matrix structure of the scattering potential @xmath95 .
the free degree of freedom is the longitudinal momentum , and the scattering process is truly one - dimensional along the tube axis .
we model a delta - function scatterer by a one - dimensional square - barrier potential of the form @xmath96 where @xmath95 describes the internal structure of the potential , and @xmath97 is the barrier width [ ] . in the rest of the paper ,
long - range potentials imply a range @xmath98 longer than the lattice constant but shorter than the envelope function wavelength @xmath99 .
short - range potentials refer to atomically sharp scatterers whose range is comparable to or smaller than the lattice constant @xmath100 . a local potential can be represented as @xmath101 we are interested in the matrix structure of the scattering potential as a function of its range and position on the lattice . for example , if we consider gaussian model potential @xmath102 , then on a surface of a cylinder @xmath103 is given by @xmath104 where @xmath105 is the potential strength , @xmath106 is the center of the gaussian on the nanotube surface , @xmath107 is the radius of the tube , and @xmath98 controls the range of the potential .
the matrix elements are calculated in the lowest `` star '' basis defined in sec .
[ sec : basis ] . for example , the intravalley matrix expectation value @xmath108 of the potential given in eq .
( [ eq : local ] ) evaluated in the lowest `` star '' basis defined in eq .
( [ eq : states.star ] ) is given by @xmath109 where @xmath110 is the fourier transform of the potential .
the fourier transform of the gaussian potential in eq .
( [ eq : gaussian ] ) is normalized such that @xmath111 as @xmath112 .
therefore , @xmath113 is given by @xmath114e^{-{q_{\tiny{\mbox{$\|$}}}}^2d^2/4}e^{-i{\mathbf{q}}\cdot{\mathbf{r}}_o } , \label{eq : fourier}\ ] ] where @xmath115 is a modified bessel function of the first kind . in the large radius limit ,
the fourier transform of the gaussian potential approaches the limit of a potential on a two - dimensional flat sheet and becomes isotropic . in the @xmath116 limit eq .
( [ eq : fourier ] ) is given by @xmath117 we define the center of the gaussian by @xmath118 , where @xmath119 , such that the potential is centered on either the @xmath0 sublattice , the @xmath1 sublattice , or along any of the three bonds defined by the triad of bond vectors @xmath120 pointing away from @xmath121 .
the total impurity hamiltonian is given by @xmath122 where @xmath123 and @xmath124 are @xmath89 matrices containing @xmath125 and @xmath126 matrix elements , respectively .
initially , we focus on the intravalley part of the potential .
evaluating both the diagonal and off - diagonal matrix elements @xmath127 s of a local potential , the intravalley part of the @xmath128 becomes @xmath129 .
\label{eq : hintra.diag}\ ] ] the component of the potential that points along the electron s propagation direction does not backscatter since it simply shifts the longitudinal momentum and can be removed by a gauge transformation [ ] . applying the gauge transformation , we find that the component of the off - diagonal matrix elements which contributes backscattering is proportional to @xmath130 , where @xmath131 labels the bond where the center of the potential is positioned .
when the potential is centered in the middle of the bond @xmath132 , and the diagonal matrix elements result in a scalar potential represented by an identity matrix .
there is no backscattering by a scalar potential in metallic nanotubes due to berry s phase of the wave function under a spin rotation [ ] .
the off - diagonal intravalley matrix elements vanish when a bond - centered impurity is on a bond that is parallel to the tube circumference ( @xmath133 ) .
for example , in fig .
[ fig : lattice ] the circumferential vector @xmath134 labeling an armchair @xmath135 tube runs parallel to the bonds labeled by a vector @xmath19
. therefore , if the center of the gaussian is positioned in the middle of any @xmath19 bond , there will be no intravalley backscattering by this local impurity for an armchair tube as labeled in fig .
[ fig : lattice ] . , @xmath136 and @xmath137 defined in eq .
( [ eq : coeffa ] ) represented by dashed , dotted and solid curves , respectively , for zero potential range @xmath138 as a function of @xmath139 .
the parameter @xmath139 determines the center of the gaussian potential along a bond connecting two neighboring sublattice sites : @xmath140 indicates a potential that is @xmath0-sublattice centered , @xmath141 yields in a @xmath1-sublattice centered potential , and @xmath142 corresponds to a bond - centered potential.,width=316 ] bonds are parallel to the circumference only in armchair nanotubes , and a mirror reflection about the axis is accompanied by an exchange of an @xmath0 and @xmath1 sublattice .
therefore , a mirror reflection across the nanotube axis for armchair tubes commutes with the hamiltonian . if a potential commutes with the hamiltonian , left and right moving states will not mix , and there will be no backscattering .
therefore , perturbations that are symmetric with respect to mirror reflection about the tube axis have zero intravalley backscattering amplitudes [ ] . in the large radius @xmath116 limit when the gaussian potential becomes isotropic as shown in eq .
( [ eq:2d ] ) , the coefficients in eq .
( [ eq : hintra.diag ] ) within the lowest `` star '' approximation are given by @xmath143 \bigg\ } \nonumber \\ & v_b = v\bigg\{1+\frac{2}{3}e^{-q_o^2d^2/4}\big[2\cos\big(\frac{2\pi(\nu-1)}{3}\big)+\cos\big(\frac{2\pi(2\nu+1)}{3}\big)\big ] \bigg\ } \nonumber \\ & v_{ab}=\frac{2v}{3}e^{-q_o^2d^2/4}\bigg\{\cos\big(\frac{\pi(2\nu-1)}{3}\big)+\cos\big(\frac{2\pi(2\nu-1)}{3}\big ) \bigg\ } , \label{eq : coeffa}\end{aligned}\ ] ] where @xmath144 is the momentum transfer between equivalent fermi points depicted in fig .
[ fig : star ] in the lowest `` star '' approximation . for short - range potentials ,
the matrix structure of the scattering potential is a function of the center of the gaussian potential @xmath139 . a plot of the amplitudes in eq .
( [ eq : coeffa ] ) as a function of potential center @xmath139 for @xmath138 is shown in fig .
[ fig : potential.center ] . the dashed , dotted , and solid curves represent @xmath145 , @xmath136 , and @xmath137 , respectively .
when the gaussian potential is centered on the @xmath0 sublattice ( @xmath140 ) , there is no amplitude on the @xmath1 sublattice ( @xmath146 ) and vice versa .
the off - diagonal amplitude @xmath137 is zero for both @xmath0 ( @xmath140 ) and @xmath1 ( @xmath141 ) sublattice centered potentials and is maximum when the potential is bond - centered ( @xmath142 ) .
when the potential is centered in the middle of the bond the three amplitude are equal @xmath147 . for long - ranged @xmath148 potentials ,
the lattice structure resolution is smeared , and @xmath149 becomes a scalar potential which does not backscatter massless fermions .
the intervalley matrix elements that describe scattering between inequivalent @xmath29 and @xmath30 points are given by @xmath150 where the phases are @xmath151 , @xmath152 , @xmath153 , and @xmath154 .
the intervalley scattering coefficients @xmath155 , @xmath156 , and @xmath157 in the large radius limit and the lowest `` star '' approximation are given by @xmath158 + e^{-q_2 ^ 2d^2/4}\big[1 + 2\cos\big(\frac{4\pi\nu}{3}\big)\big]\bigg\ } \nonumber \\ & v_b^{\prime}=\frac{v}{3}\bigg\{2e^{-q_1 ^ 2d^2/4}\big[1 + 2\cos\big(\frac{2\pi(\nu-1)}{3}\big)\big ] + e^{-q_2 ^ 2d^2/4}\big[1 + 2\cos\big(\frac{2\pi(2\nu+1)}{3}\big)\big]\bigg\ } \nonumber \\ & v^{\prime}_{ab}=\frac{v}{3}\bigg\{e^{-q_1 ^ 2d^2/4}\big[-1 + 2\cos\big(\frac{\pi(2\nu-1)}{3}\big)\big]+e^{-q_2 ^ 2d^2/4}\big[1 + 2\cos\big(\frac{2\pi(2\nu-1)}{3}\big)\big ] \bigg\}. \label{eq : coeffe}\end{aligned}\ ] ] within the lowest `` star '' there are two magnitudes of momentum transfer between distinct fermi points which are given by @xmath159 intervalley amplitudes are equal to their corresponding intravalley amplitudes for atomically sharp potentials when @xmath138 . the intervalley amplitudes approach zero for long - range potentials @xmath148 , unlike the diagonal intravalley terms in eq .
( [ eq : coeffa ] ) which approach a constant .
intravalley and intervalley amplitudes given in eq .
( [ eq : coeffa ] ) and eq .
( [ eq : coeffe ] ) , respectively , are plotted as a function of potential range @xmath160 in fig .
[ fig : range ] . the curves labeled @xmath145 , @xmath136 , and @xmath155 are amplitudes of a @xmath0-sublattice centered ( @xmath140 ) potential . due to three - fold rotational symmetry of the lattice @xmath161 for a @xmath0-sublattice centered potential .
when @xmath138 the amplitudes for intravalley and intervalley scattering become equal @xmath162 , and @xmath146 .
the vice versa is true for a @xmath1-sublattice centered ( @xmath163 ) scatterer .
off - diagonal intravalley and intervalley amplitudes @xmath137 and @xmath157 vanish for a sublattice centered potential .
the remaining two curves are plots of off - diagonal amplitudes due to a potential centered in the middle of a bond ( @xmath142 ) . in general
, the intervalley amplitudes decays slower than the intravalley ones , since @xmath164 .
when the potential is anisotropic for @xmath165 , the relative magnitude of the intervalley and intravalley amplitudes is a function of the tube s chiral angle . ) and eq .
( [ eq : coeffe ] ) , respectively , due to a local gaussian potential as a function of range @xmath160 for various values of potential center @xmath139 . the plots labeled @xmath145 , @xmath136 , and @xmath155 are due to a @xmath0-sublattice centered potential ( @xmath140 ) .
in this case , @xmath156 , @xmath137 , and @xmath157 are zero . for @xmath148 , @xmath166 and @xmath167 .
the remaining curves labeled @xmath137 and @xmath157 are off - diagonal amplitudes due to a bond - centered potential ( @xmath142 ) , which decay to zero for a long - range potential.,width=316 ] to summarize , for a gaussian model potential intervalley scattering amplitudes decay as a function of @xmath160 for all values of @xmath139 and are negligible for a long - range potential .
the intravalley components of a local potential hamiltonian also do not contribute to scattering when the potential is long - ranged .
when the range of the potential is on the order of interatomic spacing @xmath148 , the potential in eq .
( [ eq : hlocal ] ) becomes a scalar and is described by an identity matrix @xmath168 , which produces no scattering for massless dirac fermions [ ] .
this holds for all values of @xmath139 , since the position of the potential is irrelevant when the potential is slowly varying on the scale of the lattice .
therefore , only atomically sharp local potentials produce backscattering , a regime where both intra- and inter- valley scattering play a role .
note , within our model one can not realize a local potential where only intravalley scattering is present .
therefore , a local potential inevitably results in a two - valley problem . in this section
we present an example of a one - body non - local potential and show that it backscatters even when the potential is long - ranged .
we model a non - local potential by @xmath169 .
\label{eq : nonlocal}\end{aligned}\ ] ] the prefactor @xmath170 depends on the average @xmath171 of the spatial coordinates , and the remaining terms depend of the difference of @xmath172 and @xmath173 .
the @xmath174-functions restrict the length scale of @xmath175 to the nearest neighbors .
the quantity @xmath175 can describe , for example , local modulation of the hopping integral between neighboring sites .
this term depends on the orientation of the @xmath176 bond in a nanotube . calculating the off - diagonal intravalley matrix element @xmath137 in the lowest `` star '' we find @xmath177 .
\label{eq : exp.nonlocal}\end{aligned}\ ] ] in order to obtain the dependence of the potential on the orientation of the lattice with respect to the tube axis , we study the first three terms in the gradient expansion of @xmath175 given by @xmath178 where @xmath179 is a scalar , @xmath180 is a vector , and @xmath181 is a tensor of rank two .
we include deviations of the hopping amplitude to zeroth order in the momentum expansion around the brillouin zone corners .
we fix the defect potential in the plane of a tube s coordinate system and obtain the dependence of the perturbation potential on the tube s chiral angle @xmath37 .
the off - diagonal intravalley matrix elements for a non - local potential have terms that are non - vanishing for zero momentum transfer .
we evaluate the @xmath182 component of the sum in eq .
( [ eq : exp.nonlocal ] ) for the first three terms in the gradient expansion of @xmath183 shown in eq .
( [ eq : expansion ] ) .
the zeroth - order scalar @xmath179 term , the average of the hopping amplitudes , has no off - diagonal contribution at the brillouin zone corners .
the first - order term proportional to @xmath180 is a vector potential that shifts the electronic spectrum around a fermi point .
vector potentials that couple to the longitudinal momentum have no effect on any physical properties and can be eliminated by a simple gauge transformation .
therefore , only the the component of the vector potential that shift the momentum in the azimuthal direction can scatter incoming states .
the second - order term in the expansion couples to @xmath181 , a tensor of rank two .
these potentials describe deformations such at strains , twists , and curvature .
some examples of such perturbations can be found in [ ] . including the first three terms in the gradient expansion , the dimensionless sum over @xmath175 that enters the @xmath182 term of eq .
( [ eq : exp.nonlocal ] ) is given by @xmath184 \big]\equiv \tilde{g}e^{-i\theta } , \label{eq : gradient}\end{aligned}\ ] ] where we have used @xmath185 , and @xmath186 .
the components of the two - dimensional vector potential @xmath187 along the tube axis and circumference are defined by @xmath188 and @xmath189 , respectively , and have dimensions of inverse length .
the vector potential does not depend on the chiral angle as seen in eq .
( [ eq : gradient ] ) .
the components of the rank two tensor @xmath181 are defined by @xmath190 with dimensions of inverse length squared .
for example , the diagonal components @xmath191 and @xmath192 can result from uniaxial strains along the axial and circumferential directions , respectively . the off - diagonal components @xmath193 and @xmath194 can represent strains such as local twists [ , ] .
the tensor potential preserves the symmetry of the honeycomb lattice since it is invariant under the transformation of the chiral angle @xmath37 by @xmath195 , which is apparent in the @xmath196 dependence in eq .
( [ eq : gradient ] ) .
gauging away the component of the potential that couples to the longitudinal momentum , the non - local defect potential due to zero - momentum transfer is given by @xmath197 .
\label{eq : hnonlocal}\ ] ] when @xmath170 is modeled by a gaussian potential , all other matrix elements of a non - local potential decay @xmath198 where @xmath199 s are defined in eq .
( [ eq : q0 ] ) and eq . ( [ eq : q12 ] ) . therefore ,
these matrix elements are parametrically smaller than the ones described in eq .
( [ eq : hnonlocal ] ) for non - zero @xmath160 and will not be considered further .
the perturbation hamiltonian due to a non - local potential given in eq .
( [ eq : hnonlocal ] ) is independent of the impurity position @xmath139 and preserves the rotational symmetry of the lattice .
the potential is non - zero for potential ranges that exceed the scale of the lattice .
the range of this potential is only limited by the envelope square barrier defined in sec .
[ sec : range ] .
therefore , for a non - local potential only intravalley scattering contributes for finite range potentials , and the problem is single - valley .
in our previous work we developed a framework for studying casimir forces between potentials relevant for the one - valley scattering in metallic carbon nanotube [ ] . in this paper
we discuss potentials where both intra- and inter- valley scattering are present . in this section
we review the one - valley force calculation , and then generalize the method to the two - valley scattering problem . in ref .
we employ the force operator approach to calculate casimir forces between one - valley scattering potentials mediated by one - dimensional massless dirac fermions .
the total hamiltonian @xmath200 for the one - valley problem is given by @xmath201 the first term in eq .
( [ eq : h1valley ] ) is the @xmath48 low - energy hamiltonian expanded around the @xmath202 point , obtained by decoupling the two valleys in eq .
( [ eq : ho ] ) . the internal structure of the scattering potential is dictated by its spinor polarization .
we study potentials with sharp walls and calculate a force as the walls becomes impenetrable .
we model a delta - function potential by a square barrier and study limits of zero width and infinite potential strength .
the potential @xmath203 is given by @xmath204 where @xmath205 is the spinor polarization of the potential , and @xmath206 is a step function .
the force operator is given by @xmath207 where @xmath208 is a position .
using the hellmann - feynman theorem the total force is the ground state expectation value of the force operator summed over all occupied states .
the force exerted on one barrier with sharp walls is given by @xmath209 , \label{eq : gen.force}\end{aligned}\ ] ] where @xmath210 is the center of the barrier , and @xmath211 is its width .
the wave functions in eq .
( [ eq : gen.force ] ) are linear combinations of right- and left- moving eigenstates of the one - valley unperturbed hamiltonian .
the relative amplitudes of the propagating states are defined by transmission and reflection coefficients .
the scattering coefficients are obtained from the transfer matrix @xmath212 relating the wave functions at the two boundaries of a barrier .
the two expectation values in eq .
( [ eq : gen.force ] ) represent the difference between pressures on the right and left sides of the barrier . for the one scatterer system the pressures on both sides of the barrier are equal , and the net force exerted on the scatterer is zero .
-point state incoming from the left .
the two barriers of width @xmath90 and height @xmath105 are separated by distance @xmath213 .
each barrier is characterized by its spinor polarization @xmath205 .
the scattering coefficients are labeled in each region of free propagation.,width=316 ] a non - zero force arises from multiple reflections of states between two or more scatterers .
a scattering process between two barriers due to a right - moving state is illustrated in fig .
[ fig : potentials.1valley ] .
the scattering potentials are labeled by their spinor polarization @xmath205 .
the reflection and transmission coefficients resulting from scattering processes within the same valley are shown in fig .
[ fig : potentials.1valley ] . for example , @xmath214 the amplitude of a right - moving @xmath29 state backscattered into a left - moving @xmath29 state .
to calculate the force between two barriers , we fix the position of the left barrier and differentiate the hamiltonian with respect to their separation @xmath213 .
the total force is given by @xmath215 , \label{eq : force.1valley}\ ] ] where the sum is over coefficients due to right and left incoming states .
the first term in eq .
( [ eq : force.1valley ] ) is an outer pressure pushing the barriers together , and the remaining terms represent an inner pressure pushing the barriers apart . in the two barrier system ,
the outer and inner pressures are not equal resulting in a non - zero force .
we obtain a force whose sign and magnitude depends on the relative spinor polarization @xmath216 of the two scatterers .
the force between two barriers separated by distance @xmath213 in the strong and weak strength @xmath217 limits is given by [ ] @xmath218 in the @xmath219 limit @xmath220 beyond which the force is periodic .
when two potentials are aligned at @xmath221 , we obtain a universal attractive force for the fermionic casimir effect in one - dimension . when @xmath222 the relative spinor polarization of the two scatterers is antiparallel resulting in a repulsive force .
the oscillatory dependence on @xmath223 persists in the weak strength limit .
note , it is convenient to express the force in the strong limit in terms of a dilogarithm function @xmath224 , @xmath225/2\pi z^2 $ ] in eq .
( [ eq : f1valley ] ) when @xmath219 . the results in eq .
( [ eq : f1valley ] ) are plotted in fig .
[ fig : force.plot ] . .
the solid and dashed lines represent the forces in the large and small potential strength limits given in eq .
( [ eq : f1valley ] ) , respectively . the magnitude of the force in the weak potential limit , the dashed curve , is rescaled to @xmath226 so the two curves can be compared.,width=316 ] in the @xmath219 limit the states between the barriers are quantized , and the number of states changes by one when @xmath223 is an odd multiple of @xmath39 resulting in the cusps seen in fig . [ fig : force.plot ] .
the weak limit does not exhibit this behavior , since the quasibound states between the scatterers are described by a continuous spectrum . and width @xmath90 .
a @xmath89 scattering matrix is obtained by relating right and left moving @xmath29 and @xmath30 states to their corresponding outgoing states.,width=316 ] in this section we generalize the method described in sec .
[ sec:1valley ] to the two - valley scattering problem , where scattering of states between different valleys as well as within the same valley is present .
therefore , the potential is described by a @xmath227 matrix characterized by sublattice and valley degrees of freedom .
the intra- and inter- valley matrix elements are obtained using the bloch basis states described in sec .
[ sec : basis ] .
the freely propagating states are eigenstates of the effective hamiltonian given in eq .
( [ eq : ho ] ) . the wave functions used to calculate the force expectation values obtained from the hellmann - feynman theorem are linear combination of right and left moving states from the two @xmath29 and @xmath30 points .
the relative amplitudes of the propagating states are defined by scattering coefficients .
a general expression for the wave function in a region of free propagation is given by @xmath228 where @xmath205 s are four component spinors given in eq .
( [ eq : spinors ] ) , and @xmath80 s and @xmath229 s are scattering coefficients . for simplicity of notation
we have dropped the @xmath65 and @xmath66 superscripts referring to one of the three equivalent corner points .
the three @xmath29 points are related by reciprocal lattice vectors , and physical quantities will not depend on the particular choice of the corner point .
the dependence on @xmath65 and @xmath66 enters only as a phase of the scattering coefficients @xmath80 s and @xmath229 s .
the full hamiltonian for the one square barrier system is given by @xmath230 where @xmath231 is the low - energy hamiltonian given in eq .
( [ eq : ho ] ) , @xmath232 is a perturbation potential , such as @xmath128 or @xmath233 described in sec .
[ sec : h1 ] , and the step functions define a square barrier . integrating eq .
( [ eq : ht ] ) across the barrier , the @xmath89 transfer matrix becomes @xmath234[k+\hat{v}]\big\},\end{aligned}\ ] ] where @xmath211 is the barrier width . from the transfer matrix , we calculate the scattering matrix .
the @xmath89 scattering matrix , obtained from incoming and outgoing states illustrated in fig . [ fig : scattering.matrix ] , is defined as @xmath235 where @xmath236 are right - moving incoming ( @xmath81 ) and outgoing ( @xmath237 ) amplitude column vectors , and @xmath229 s define left - moving states as shown in fig .
[ fig : scattering.matrix ] .
the `` primes '' in eq .
( [ eq : smatrix ] ) indicate the coefficients due to the states incoming from the right .
each coefficient in the scattering matrix in eq .
( [ eq : smatrix ] ) is a @xmath48 matrix defining both intravalley and intervalley scattering amplitudes .
for example , @xmath238 where the diagonal(off - diagonal ) terms are the intravalley(intervalley ) transmission coefficients .
for instance , @xmath239 is the forwardscattering amplitude of a right - moving @xmath29 state being transmitted into a right - moving @xmath30 state .
as in the one - valley problem , non - zero forces arise from interactions between two scatterers .
an scattering process illustration of a left - incoming @xmath29 state between two potentials @xmath240 and @xmath241 separated by distance @xmath213 along the tube axis is shown in fig .
[ fig : potentials ] . as
before , we fix the left barrier and calculate the force exerted on the right barrier using the hellmann - feynman theorem . the force is given by @xmath242 , \label{eq : force.gen}\ ] ] where the summations represent a sum over all reflection and transmission coefficients in - between the two barriers ( region ii in fig .
[ fig : potentials ] ) due to right and left incoming states , @xmath243 and @xmath244 . throughout this
paper lower - case coefficients will refer to scattering by one barrier , and upper - case ones due to scattering by a two barrier system . the first term in eq .
( [ eq : force.gen ] ) represents an outer pressure in regions iii of fig .
[ fig : potentials ] due to a continuous spectrum of states pushing the barriers together .
the second and third terms in eq .
( [ eq : force.gen ] ) result in the inner pressure pushing the barriers apart , which is obtained from the coefficients in region ii of fig .
[ fig : potentials ] .
these coefficients are given by @xmath245 when the intervalley matrix elements are zero in one of the scattering potentials @xmath232 , there is no forward- and back- scattering between inequivalent fermi points for the two - barrier system . in this case eq .
( [ eq : force.gen ] ) reduces to the one - valley force given in eq .
( [ eq : force.1valley ] ) .
using the method described in sec .
[ sec : force ] , we explore the dependence of the force between two scatterers on the matrix structure , range , and strength of the defect potentials .
we distinguish interactions between local and non - local potentials discussed in sec .
[ sec : h1 ] .
we show that the casimir force decays as @xmath246 which is a universal result in one - dimension in the far field limit .
however , we also find that in the presence of intervalley scattering there is a spatially periodic modulation of this force .
our results pertain to the limit @xmath247 where shape corrections are negligible [ ] .
a general solution of the integrals appearing in the force calculations in derived in the appendix , and a summary of our results is presented in table [ tab : forces ] . in this section
we first consider interactions between local potentials . as discussed in sec .
[ sec : local ] , backscattering from a local potential is significant for potential that vary on the scale of the lattice @xmath248 .
let us specialize eq .
( [ eq : hlocal ] ) to describe impurities that are centered at either of the two sublattice sites .
we first study the strong potential limit by fixing the area of the potential @xmath217 .
the force is independent of the magnitude of the potential in the @xmath219 limit and is relevant for the discussion of universal casimir interactions . for a sublattice centered potential in the atomically sharp limit @xmath249 intra- and inter- valley amplitudes are equal @xmath250 , as shown in fig .
[ fig : range ] .
all reflection and transmission coefficients for such scatterers approach the same value in the strong potential limit , @xmath251 @xmath252 and are independent of the sign of the potential . calculating the two - barriers scattering coefficients described in eq .
( [ eq : s2 ] ) and inserting into eq .
( [ eq : force.gen ] ) , the force between two impurities centered on equivalent sublattice sites is given by @xmath253 , \label{eq : force.aa}\ ] ] where @xmath254 is a primitive translation vector in the tangent plane separating the two impurities , and @xmath213 the component of their separation along the axial direction .
the subscripts @xmath255 and @xmath256 imply a force between impurities which are located on equivalent sites . applying eq .
( a13 ) , the solution of the force integral in eq .
( [ eq : force.aa ] ) is given by @xmath257 .
\label{eq : f.aa}\ ] ] unlike in the one - valley problem where the force decays monotonically as @xmath246 , in addition the two - valley problem results in a spatial modulation of the force , as observed in the argument of the dilogarithm function in eq .
( [ eq : f.aa ] ) .
the force oscillates with the period of the @xmath258 superlattice indicating coupling between the two valley points .
the force given by eq .
( [ eq : f.aa ] ) is plotted in fig .
[ fig : fboth ] as a function of @xmath259 for an armchair tube .
the points on the curve indicate the discrete values of the force in each period .
the force between two equivalent impurities is purely repulsive , as seen in fig .
[ fig : fboth ] , since @xmath260>0 $ ] , where @xmath261 .
next , we consider interactions between impurities residing on different sublattice sites . a force between an @xmath0-centered ( @xmath140 ) and a @xmath1-centered ( @xmath141 ) scatterer
is given by @xmath262 , \label{eq : force.ab}\ ] ] where @xmath37 is the chiral angle of a nanotube .
applying eq .
( a13 ) the force in eq . ( [ eq : force.ab ] ) becomes @xmath263 .
\label{eq : f.ab}\ ] ] for unlike impurities the force is purely attractive for all values of the chiral angle .
the argument of the dilogarithm takes three values @xmath264 which also contains @xmath265 periodicity . the force given in eq .
( [ eq : f.ab ] ) is plotted in fig .
[ fig : fboth ] on a curve labeled @xmath266 for an armchair tube as a function of position .
( [ eq : f.ab ] ) indicates that the system is invariant under the rotation of the chiral angle by @xmath39 , rather than by @xmath195 as for a defect - free lattice .
this occurs because the impurity is fixed on the lattice rather than on the tube s coordinates , and the position of the scatterer co - rotates with the lattice for various values of the chiral angle .
therefore , the three - fold symmetry in the presence of an atomically sharp impurity is broken .
the chiral angle dependence appears only in the force between unlike impurities , since the separation between the two defects is not a primitive lattice vector .
the three branches in one period of @xmath266 are plotted as a function of @xmath37 in fig .
[ fig : chiral ] .
the figure indicates that force oscillates between @xmath267 and @xmath268 for all values of @xmath269 . an attractive and repulsive interaction between defects on different and same sublattice sites , respectively , was recently shown in two - dimensional graphene [ ] . , a force between an @xmath0 and a @xmath1 sublattice centered impurities , given in eq .
( [ eq : f.ab ] ) as a function of chiral angle @xmath37 .
the force is scaled by a factor of @xmath270 and is found to be attractive for all values of @xmath37.,width=307 ] next , we study the small potential @xmath271 limit and compare results to the ones obtained in strong @xmath219 limit given by eq .
( [ eq : f.aa ] ) and eq .
( [ eq : f.ab ] ) .
we keep the first non - zero term in the expansion of small @xmath272 and take the zero width limit @xmath273 .
the next order term in the small width expansion accounting for shape corrections is @xmath274 [ ] . for simplicity , we study the case of armchair nanotubes @xmath14 and find a general expression for a force between sublattice centered defects .
the off - diagonal matrix elements @xmath137 and @xmath157 are zero for sublattice centered potentials @xmath275 .
the force between two local potentials in the @xmath271 limit is given by @xmath276 , \label{eq : flocal.small}\end{aligned}\ ] ] where @xmath277 refers to a force between potentials of the same(different ) sign of @xmath272 , and the superscripts indicate the potential describing scatterer one and two .
unlike in the large strength limit shown in eqns .
( [ eq : force.aa])-([eq : f.ab ] ) , the sign of the force is a function of the relative sign @xmath278 of the two potentials in the weak limit
. the sign of the force also depends on the relative sublattice centers of the two scatterers , as in the strong potential limit .
therefore , in the @xmath271 limit the sign of the force is controlled both by the sublattice position of the two defects and the relative sign @xmath278 of their potential strength .
the @xmath258 periodic oscillation persists in the small strength limit .
these results for specific sublattice positions of the two potentials and general chiral angle are shown in table [ tab : forces ] and are plotted as an inset in fig .
[ fig : fboth ] for an armchair tube . for long - range @xmath148 potentials the force approaches zero for all values of @xmath139 since
the sublattice intravalley matrix elements @xmath279 s and @xmath280 s become equal , and intervalley terms @xmath281 s and @xmath282 s decay to zero as shown in fig .
[ fig : range ] .
this result confirms the absence of backscattering from an scalar potential by massless dirac fermions . although a scatterer where the two valleys are decoupled can not be realized for a local potential , a case of pure intervalley scattering @xmath283 possible . for a local potential , when an impurity is centered in the middle of a bond that points along the circumference , the potential scatters
states only between inequivalent valleys as discussed in sec .
[ sec : local ] .
this holds because the intravalley part of the hamiltonian @xmath284 is a scalar potential for all values of @xmath160 , since @xmath132 for a bond - centered potential , and @xmath285 when the perturbed bond points along the circumference .
the intervalley amplitudes are equal @xmath286 for @xmath142 , when @xmath14 and @xmath287 . in this case , the intervalley transmission coefficients @xmath288 and the intravalley reflection coefficients @xmath289 vanish .
the absence of back- and forward- scattering within the same valley and between different valleys , respectively , by potentials that preserve mirror reflection symmetry about the tube axis has been also shown by ando @xmath290 [ ] . in the @xmath219 limit , the non - zero coefficients have limits @xmath291 and @xmath292 .
the phase of the reflection coefficients depends of the sign of @xmath272 .
the force between two potentials with only intervalley scattering contribution in the large potential strength limit is given by @xmath293 , \label{eq : force.inter}\ ] ] where @xmath55 is the magnitude of the transmission coefficient .
the second term in the integrand representing the inner pressure is fundamentally different from the ones seen in eq .
( [ eq : force.aa ] ) and eq .
( [ eq : force.ab ] ) .
the phase that appears in eq .
( [ eq : force.inter ] ) is associated with large momentum backscattering .
the forces shown in eq .
( [ eq : force.aa ] ) and eq .
( [ eq : force.ab ] ) involve two types of momentum transfer which appear as various terms in the equations .
when both intra- and inter- valley play a role , there is finite transmission even in the strong potential limit .
when only intervalley scattering is present , the strong potential limit results in an impenetrable wall limit since transmission coefficient approaches zero .
therefore , the inner pressure in eq .
( [ eq : force.inter ] ) results from resonant states between the boundaries .
the overall prefactor in eq .
( [ eq : force.inter ] ) is twice the magnitude than in eq .
( [ eq : force.aa ] ) and eq .
( [ eq : force.ab ] ) .
applying eq .
( a14 ) and evaluating the periodic part of the force , the solution of the integral in eq .
( [ eq : force.inter ] ) is given by @xmath294 = \frac{\pi\hbar v_f}{72z^2}\left\ { \begin{array}{rl } & \{-3,1\},~\mbox{$s=1$}\\ \\ & \{-2,6\},~~\mbox{$s=-1 $ } \end{array } \right .
\label{eq : finter}\ ] ] when only intervalley scattering amplitude is present the force oscillates between attractive and repulsive with @xmath265 period as observed in eq .
( [ eq : finter ] ) .
the magnitude of the force is determined by the relative sign @xmath278 of the two potentials .
a plot of @xmath295 as a function of @xmath259 for @xmath296 is shown in fig .
[ fig : finter ] .
the points in the plot indicate the discrete values in each period of oscillation given in eq .
( [ eq : finter ] ) . in the small strength limit
@xmath271 the force becomes @xmath297 the results of eq .
( [ eq : finter.small ] ) are shown as an inset in fig .
[ fig : finter ] .
although the prefactors of the force are different in the two limits , the oscillation between attractive and repulsive persists in both weak and strong potential limits .
similar behavior has been observed previously in one - dimensional fermi liquids where only large momentum backscattering is considered [ , ] .
refer to table [ tab : forces ] for a compact summary of the main results presented here .
[ cols="^,^,^,^,^,^ " , ] in this section we calculate casimir forces between impurities described by non - local potentials given in eq .
( [ eq : hnonlocal ] ) . when the range of a non - local potential is @xmath298 , off - diagonal intravalley matrix elements @xmath137 are dominant since all other amplitudes are parametrically smaller as noted in sec .
[ sec : non - local ] .
therefore , a non - local potential can result in a one - valley scattering problem discussed in sec .
[ sec:1valley ] .
these potentials can describe modulations to the hopping amplitudes between neighboring sites . in the absence intervalley scattering
, states are scattered only within the same @xmath29 point .
therefore , the scattering coefficients @xmath299 are zero .
likewise , the intervalley coefficients due to states incoming from the right vanish .
since the two fermi points are decoupled the perturbation matrix is described by two independent @xmath48 matrices in the sublattice @xmath56-space .
the control parameter we vary to study interactions between two non - local defects is the sign of the potential @xmath105 .
we assume that the dimensionless quantities @xmath300 s defined in eq .
( [ eq : gradient ] ) are equal for the two barriers .
in the strong potential @xmath219 limit the magnitude of the non - zero scattering coefficients approach @xmath301 and @xmath292 .
we calculate the interaction between two barriers with the same and different signs of @xmath302 . applying the one - valley force result given in eq .
( [ eq : force.1valley ] ) , the force between two non - local potentials becomes @xmath303 , \label{eq : force.intra}\ ] ] where @xmath278 is the relative sign of the two potentials .
the integrands in eq .
( [ eq : force.inter ] ) and eq .
( [ eq : force.intra ] ) are equivalent except for the phase @xmath304 appearing in eq .
( [ eq : force.inter ] ) .
this phase associated with large momentum backscattering is absent in eq .
( [ eq : force.intra ] ) since there is no intervalley scattering present by potentials given in eq .
( [ eq : hnonlocal ] ) .
the solution to the integral in eq .
( [ eq : force.intra ] ) is shown in the appendix . applying eq .
( a14 ) the force is given by @xmath305 the result in eq .
( [ eq : fintra ] ) shows that there is an attractive force between two scatterers with equal sign of @xmath272 @xmath306 and a repulsive force between defects of unequal sign of @xmath272 @xmath307 .
the relative sign of @xmath105 is analogous to the difference between the spinor polarizations @xmath223 of the two scatterers discussed in sec .
[ sec:1valley ] .
potentials of equal sign ( @xmath308 ) refer to the case of parallel scatterers @xmath309 .
two potentials of opposite sign ( @xmath310 ) , on the other hand , refer to the case of anti - parallel scatterers @xmath311 . the results in eq .
( [ eq : fintra ] ) are consistent with the force in the @xmath219 limit of eq .
( [ eq : f1valley ] ) [ ] .
likewise , @xmath312 in the @xmath271 limit agrees with eq .
( [ eq : f1valley ] ) .
the magnitude of the force is larger than the result in eq .
( [ eq : f1valley ] ) by a factor of two since we are including fermions from the two @xmath34 branches of carbon nanotubes .
these results are shown in table [ tab : forces ] .
intervalley scattering becomes important for non - local potentials when @xmath313 for @xmath248 .
a few example of such defects are a vector potential with a zero component along the tube axis ( @xmath314 ) , a tensor potential for armchair tubes and zero twist ( @xmath14 and @xmath315 ) , or a tensor potential for zigzag tube with zero uniaxial strain ( @xmath15 and @xmath316 ) .
the effect of intervalley scattering on the casimir force is discussed in sec .
[ sec : flocal ] in the context of local potentials , and the same physics apply for the case of non - local potentials .
defects or impurities on a carbon nanotube can backscatter electrons either through intravalley or intervalley scattering processes .
in general both channels are present with their relative strengths determined by the range and symmetry of the scattering potentials .
the models we present here provide a framework for understanding the backscattering - induced forces on these species .
the signature of intervalley scattering is a spatial modulation of the scattering - induced forces .
by contrast intravalley scattering mediates a force that can be either attractive or repulsive , but has a strength that decays monotonically as a function of increasing separation .
interestingly , in all cases where the interaction is described by a local potential , the scattering problem is inevitably multivalley in character , and the energy and force of the species oscillate as a function of separation .
the long - range interaction between multiple scatterers might lead to complex phase structures .
it was suggested by shytov @xmath290 [ ] that interaction between adatoms absorbed on the graphene lattice can result in defect aggregation and inhomogeneities on the lattice .
the scatterers we describe in this paper can be physically realized by various atomic and molecular species adsorbed on the tube wall .
these range from covalently bound atoms and molecules [ ] , to more weakly bound metallic species [ ] .
the range of the scattering potential is determined by the size of the absorbed species relative to the lattice constant .
the symmetry of the potential is determined by the spatial variation of on - site energies and by the modulation in the intersite hopping amplitudes produced by these species .
covalently bound species provide the most natural candidates for the strongly coupled local potential models described in section [ sec : local ] . here
, the on - site potential barrier at an adsorbed site can be as large as 5 ev enforcing an effectively hard wall boundary condition on the electronic wave functions . in this regime
the results of section [ sec : flocal ] can be used to provide a bound on the electron - induced force .
for example , the maximum attractive force between two scatterers in the impenetrable wall limit leads to an interaction energy of @xmath317 . with @xmath318 for nanotubes
this gives an energy of @xmath319 at a range @xmath320 .
note that its spatial form follows the same scaling law as the coulomb interaction between uncompensated charges , but it is reduced by a factor @xmath321 .
thus , for charge neutral dipoles @xmath322 whose electrostatic interactions scale as @xmath323 , they are dominated by the casimir interaction in the far field @xmath324 .
similarly , this one - dimensional casimir interaction completely dominates the familiar van der waals interactions between charge neutral species that are mediated by the fluctuations of the exterior three dimensional electromagnetic fields .
the weak coupling limit is relevant to the interactions of less strongly bound species , such as metal atoms or molecules bound by @xmath39 stacking interactions , e.g. benzene . here
the energy scale for the local potential is more modest , of order 1 ev which , assuming a range of order a graphite lattice constant , corresponds to a dimensionless coupling parameter @xmath325 . in this weak coupling limit @xmath326 a local potential of @xmath327 ev results in @xmath328 mev at a distance of @xmath329 nm .
though weaker , this interaction still decays slowly as a function of distance ( @xmath330 ) and will also dominate the electrostatic interaction between charge neutral dipoles in the far field . in this weak coupling regime ,
strain induced couplings , represented by non - local scatterers can be comparable in size . assuming a linear scaling of intersite hopping amplitudes with bond lengths following @xmath331 a bond length change of 0.2 and a potential range on the order of the lattice constant
, this gives a dimensionless potential strength of @xmath332 and a weak coupling interaction @xmath333 , we find @xmath334 mev .
these are of the same order as the forces produced by local potentials in the weak coupling limit . for adsorbate - induced potentials ,
it is difficult to realize a regime where the scattering is dominated by potentials with solely a nonlocal form .
thus , one concludes that intervalley scattering and a residual spatial oscillation of the force is a generic property of inter adsorbate interactions mediated by the propagating electrons . it may be possible to quench the intervalley channel by application of a magnetic field along the tube axis which would have the effect of introducing a gap at either the k or k point and isolating the effects of intravalley scattering .
we also note that strains can be engineered into these structures by application of mechanical stresses , and this might provide an avenue for realizing the predictions of the nonlocal model .
in this paper we show that interactions between scatterers in metallic carbon nanotubes results in a one - dimensional casimir problem .
we generalized our previous work which includes the one - valley problem of nanotubes , to incorporate the effects of intervalley scattering .
we show that local potentials in nanotubes produce a two - valley scattering problem .
the decoupling of the two valleys is not possible for a local potential since the range must be atomically sharp in order to produce finite backscattering .
local potentials whose spatial extent is beyond the lattice constant result in scalar potentials which do not backscatter massless dirac fermions .
non - local potentials , on the other hand , can result in a decoupled valley scattering problem .
intervalley scattering amplitudes are parametrically smaller for finite range non - local potentials .
therefore , we formulate a physically realizable potential which reduces to the one - valley scattering problem
. we study forces between two scatterers mediated by the propagating electrons of metallic carbon nanotubes . for interactions between both local and non - local potentials
we find a universal @xmath246 power law decay for a one - dimensional casimir force .
however , for local potentials , where intervalley scattering plays a role , we also observe a position dependent periodic modulation of the force .
the signs and magnitudes of the forces are not universal and are controlled by the internal symmetry of the scattering potentials .
we would like to thank j.m .
kinder and philip t. gressman for helpful discussions .
this work was supported by the department of energy under grant no .
de - fg02-er45118 .
19 natexlab#1#1bibnamefont # 1#1bibfnamefont # 1#1citenamefont # 1#1url # 1`#1`urlprefix[2]#2 [ 2][]#2 , , , * * , ( ) .
, , , * * , ( ) . , , , * * , ( ) .
, , , * * , ( ) . , * * , ( ) . , , , , , , , * * , ( ) . , * * , ( ) . , _ _ ( , , ) . ,
* * , ( ) .
, , , * * , ( ) . ,
* * , ( ) . , * * , ( ) . , * * , ( ) . , * * , ( ) . , , , * * , ( ) . , , , , , , , , , , , * * , ( ) . , , , , * * , ( ) . , , , , ,
* * , ( ) . ,
* * , ( ) .
in this appendix we provide a derivation for the integrals that appear in the calculations of casimir forces for @xmath219 .
although , a cutoff function is introduced in order to control divergences appearing in the integral , we show that the final result is cutoff independent .
the class of integrals found in this paper have a general form @xmath335 , \label{eq : integral}\ ] ] where @xmath213 is the impurity separation along the tube axis .
the integrand in eq .
( [ eq : integral ] ) can be represented in terms of a poisson kernel @xmath336 where @xmath296 , and @xmath337 . introducing an exponential cutoff function , the integral in eq .
( [ eq : integral ] ) becomes @xmath338dq .
\label{eq : cutoff}\ ] ] since the poisson kernel is @xmath339 periodic in @xmath340 , the integral can be expressed as an infinite sum times an integral over a region of @xmath341 $ ] .
rewriting eq .
( [ eq : cutoff ] ) we obtain @xmath342dq\nonumber\\ & \bigg(\sum_{n=0}^{\infty}(q+2n\pi)e^{-\mu(q+2n\pi)}\bigg ) .
\label{eq : integral.cutoff}\end{aligned}\ ] ] expressing the sum in terms of a geometric series and separating terms constant in @xmath340 , the series in eq .
( [ eq : integral.cutoff ] ) to @xmath343 is given by @xmath344 the first two terms on the rhs of eq .
( [ eq : series ] ) diverge in the limit @xmath345 , but vanish when integrated over @xmath340 since @xmath346dq=0 .
\label{eq : zero}\ ] ] to verify that the above statement is true in the case of @xmath347 we express the poisson kernel in terms of a delta function @xmath348 where @xmath349 and @xmath350 .
inserting eq .
( [ eq : lim.delta ] ) into eq .
( [ eq : zero ] ) , we find that there is either one @xmath174-function in the range of integration @xmath341 $ ] or two @xmath174-functions at the two limits of integration , each contributing half the area .
therefore , in both cases the integral over the series of @xmath174-functions yields a factor of @xmath339 , which is consistent with the result in eq .
( [ eq : zero ] ) . note , in the @xmath347 limit eq .
( [ eq : cutoff ] ) can be solved using a generalized abel - plana formula which provides a finite expression for a difference between an infinite integral and an infinite sum [ ] . combining the above results and noting that the third term in eq .
( [ eq : series ] ) is cutoff independent , eq . ( [ eq : integral.cutoff ] ) becomes @xmath351dq .
\label{eq : integral.finite}\ ] ] we use the following identity to solve the integral in eq .
( [ eq : integral.finite ] ) : @xmath352 where the `` hat '' indicates the fourier series of the original function .
the fourier series of the poisson kernel is given by @xmath353 the fourier series of the other term in eq .
( [ eq : integral.finite ] ) is given by @xmath354 using the results from eqns .
( [ eq : identity])-([eq : series2 ] ) , eq .
( [ eq : integral.finite ] ) becomes @xmath355 eq .
( [ eq : integral.final ] ) is a general result which can be applied to all the integrals encountered in this paper .
the series above can be represented in terms of dilogarithm functions .
for example , @xmath356 and , @xmath357=\sum_{n=1}^{\infty}\left\ { \begin{array}{rl } & \frac{\cos(n\varphi)}{n^2},~~~~~~~~~\mbox{$s=-1$}\\ \\ & \frac{(-1)^n \cos(n\varphi)}{n^2 } , ~~\mbox{$s=1 $ } \end{array } \right .
\label{eq : lif}\ ] ] where @xmath224 is a dilogarithm function . in sec .
[ sec : flocal ] we calculate forces between two local sublattice centered impurities .
the solution of eq .
( [ eq : force.aa ] ) for interaction between defects residing on equivalent sites is eq .
( [ eq : lir ] ) , where @xmath358 , with @xmath359 and @xmath310 . the result for the force integral in eq .
( [ eq : force.ab ] ) , applicable to interactions between impurities centered on inequivalent sites , is eq .
( [ eq : lir ] ) with @xmath360 and @xmath308 .
the integral in eq .
( [ eq : integral ] ) can also be related to integral in eq .
( [ eq : force.inter ] ) for a force between two local potentials where only intervalley scattering plays a role , and eq .
( [ eq : force.intra ] ) for interactions between non - local potentials .
the limit of zero transmission @xmath361 is equivalent to @xmath347 in eq .
( [ eq : integral ] ) , where @xmath362 . writing eq .
( [ eq : force.inter ] ) and eq .
( [ eq : force.intra ] ) in a general form in terms of @xmath363 we obtain @xmath364\nonumber \\ & = \frac{1}{\pi}\int_{0}^{\infty}k\bigg[1-\lim_{\rho\to 1}\frac{1-\rho^4}{1+\rho^4 + 2\rho s\cos(2kz+\varphi)}\bigg ] , \label{eq : fex}\end{aligned}\ ] ] where we have ignored the prefactors . the right - hand side of eq .
( [ eq : fex ] ) is equivalent to eq .
( [ eq : integral ] ) in the limit @xmath347 .
therefore , the solution of eq .
( [ eq : force.inter ] ) is given by eq .
( [ eq : lif ] ) for @xmath365 .
the solution to eq .
( [ eq : force.intra ] ) is obtained by setting @xmath358 in eq .
( [ eq : lif ] ) . | we study interactions between localized scatterers on metallic carbon nanotubes by a mapping onto a one - dimensional casimir problem .
backscattering of electrons between localized scattering potentials mediates long range forces between them .
we model spatially localized scatterers by local and non - local potentials and treat simultaneously the effects of intravalley and intervalley backscattering .
we find that the long range forces between scatterers exhibit the universal power law decay of the casimir force in one dimension , with prefactors that control the sign and strength of the interaction .
these prefactors are nonuniversal and depend on the symmetry and degree of localization of the scattering potentials .
we find that local potentials inevitably lead to a coupled valley scattering problem , though by contrast non - local potentials lead to two decoupled single - valley problems in a physically realized regime .
the casimir effect due to two - valley scattering potentials is characterized by the appearance of spatially periodic modulations of the force . |
recent cosmological observations such as the wmap results are consistent with the inflationary scenario . hence , we are prompted to seek the inflaton in a unified theory of particle physics . currently , it is widely believed that the most promising candidate for a unified theory is superstring theory .
interestingly , the superstring theory predicts the existence of the extra dimensions . in order to reconcile this prediction with our observed 4-dimensional universe , we need a mechanism to hide the extra dimensions . for a long time
, the kaluza - klein compactification was considered to be the unique choice .
however , recent developments in superstring theory suggest a braneworld picture where we are living on 4-dimensional hypersurface embedded in a higher dimensional spacetime @xcite .
this braneworld picture not only gives a way for the superstring theory to be phenomenologically viable but also suggests a new inflationary scenario , so - called brane inflation @xcite . in the brane inflation scenario , the radion
, the distance between branes , plays the role of the inflaton and inflation is terminated by the brane collision .
this is nice because the inflation is realized purely geometrical manner without introducing an ad - hoc scalar field . in this scenario , however , branes are treated as test branes . on the other hand ,
relativistic cosmologists have studied the braneworld gravity intensively @xcite . in these studies ,
the effect of the bulk geometry on the 4-dimensional braneworld cosmology have been a central concern , though inflation is usually assumed to be driven by the fundamental scalar field either on the brane or in the bulk . in this braneworld cosmology ,
the self - gravity of branes are properly treated and hence , it is clear how to calculate corrections due to the bulk effect @xcite .
taking a look at both approaches , we have come up with the idea of incorporating geometrical inflation in a simple randall - sundrum ( rs ) model @xcite . in this paper
, we would like to propose an inflationary scenario driven by the flux generated by a brane that is charged with respect to a five - form field strength .
the idea is very similar to that of brane inflation but we take into account the self - gravity of branes . our model is constructed in the rs framework @xcite .
we suppose that initially two positive tension branes are inflating as de sitter spacetimes embedded in an anti - de sitter bulk .
eventually , they collide with each other and inflation will end .
subsequently , two branes are assumed to coallesce and evolve as a single @xmath0-symmetric positive tension brane .
the gravitational theory is non - singular and the model we construct is essentially 5-dimensional way . except at the collision point
, we can use the effective action obtained by the low energy approximation @xcite .
but in the 4-dimensional einstein frame , the evolution of the universe is discontinuous at the collision point , which means that the effective 4-dimensional theory breaks down .
however at the collision point we can use 5-dimensional energy - momentum conservation to determine the dynamics @xcite .
we are thus able to analyze the spectrum of primordial scalar and tensor fluctuations produced after the collision .
it turns out that a curvaton - type mechanism is required to generate the primordial density perturbations producing the present structure of the universe .
the organization of this paper is as follows . in sec.ii ,
the basic setup is presented . in sec.iii ,
our cosmological scenario is described .
inflation driven by the flux is analyzed both in the induced metric frame and the einstein frame .
the consistency analysis gives the expansion rate after the collision determined by the expansion rate of both branes before the collision .
the cosmological history after the collision is briefly summarized . in sec.iv , the spectrum of fluctuations are calculated .
the sec.v is devoted to conclusions . in the appendix ,
the detailed derivation of the effective action is provided .
the point of brane inflation is that no fundamental scalar field is necessary and the exit from inflation is realized by the collision of branes .
what we want to do is to incorporate this idea into the codimension - one rs braneworld model .
we consider a two - brane system where one has a @xmath0 symmetry and the other does not .
hereafter , we call the former the boundary brane and the latter the bulk brane .
as we show in an appendix such a set - up can be realised as the limiting case of a three - brane system , with two boundary branes , where one of the @xmath0-symmetric branes is sent to infinity .
the model is described by the 5-dimensional action @xmath1
+ \sum_{i=1}^2 \frac{1}{\kappa^2 } \int d^4 x \sqrt{-h_i } k^{i } \nonumber \\ & & -\frac{1}{2\cdot5 ! }
\int d^5 x \sqrt{-g } f_5 ^ 2 + \sum_{i=1}^2 \mu_i \int c_4 \nonumber \\ & & + \frac{1}{4 ! } \int d^5 x \partial_a \left(\sqrt{-g } f^{abcde } c_{bcde } \right ) \nonumber \\ & & + \sum_{i=1}^2 \int d^4 x \sqrt{-h_i } \left[-\sigma_i + { \cal l}^i_{\rm matter } \right ] \label{5d : action } \end{aligned}\ ] ] where @xmath2 is the 5-dimensional gravitational coupling constant and @xmath3 is the 5-dimensional curvature .
both our branes have positive tension , @xmath4 , but opposite charges , @xmath5 , which couple to a 4-form potential field , @xmath6 .
the bulk brane separates the bulk into regions @xmath7 and @xmath8 ( see fig.1 ) .
we denote the induced metric on each brane by @xmath9 and @xmath10 denotes the trace - part of the extrinsic curvature of each brane . here
we have taken into account the gibbons - hawking boundary term instead of introducing delta - function singularities in the five - dimensional curvature .
we incorporated the 5-form field @xmath11 which can change the effective cosmological constant in the bulk @xmath12 @xcite .
the third line represents the surface term which is introduced to make the variation of the action with fixed @xmath13 consistent .
let us take the coordinate system @xmath14 the latin indices @xmath15 and the greek indices @xmath16 are used for tensors defined in the bulk and on the brane , respectively .
the 5-form equation of motion becomes @xmath17 where @xmath18 denotes the position of each brane . in each bulk ( @xmath19 ) , it is easy to solve eq .
( [ eom:5form ] ) as @xmath20 where @xmath21 are constants of integration .
because of the charge conservation , we have @xmath22 as is explained in the appendix .
thus we see @xmath13 has no local dynamics .
as @xmath13 is not dynamical , we can eliminate it from the action by simply substituting the above solution into the original action .
this can be done using the equations of motion to give @xmath23 substituting eq .
( [ eom2:5form ] ) into the original action ( [ 5d : action ] ) , we see that 4-form potentials are cancelled and the effect of 5-form field strength is indistinguishable from a cosmological constant term in the bulk .
thus , the resultant action is @xmath24 \nonumber\\ & & + \frac{1}{2\kappa^2 } \int_{i\!i } d^5 x \sqrt{-g } \left[\overset{(5)}{r } + 2\lambda \right ] \nonumber\\ & & + \sum_{i } \frac{1}{\kappa^2 } \int \sqrt{-h_i}k^{i } \nonumber \\ & & + \sum_i \int d^4 x \sqrt{-h_i } \left[-\sigma_i + { \cal l}^i_{\rm matter } \right ] \ .\end{aligned}\ ] ] the 5-form field strength @xmath13 in region @xmath7 acts like a change in the effective 5-dimensional cosmological constant in region @xmath7 , @xmath25 .
consequently , two bulk regions have a different cosmological constant .
the absolute value of the cosmological constant in region i is assumed to be small and therefore we shall see that the effective cosmological constants induced on both branes are positive .
if the expansion rate of the second brane is faster than that of first , both branes will eventually collide with each other and the opposing charges annihilate .
the resulting brane tension is assumed to become ( close to ) the randall - sundrum tuning value so that the inflation ends and the universe becomes radiation dominated .
we assume a completely inelastic collision so that the kinetic energy of the branes is transformed into radiation energy density on the brane .
the subsequent evolution is same as that of the radiation dominated universe in the rsii brane model .
thus , the original flux generated by the charged branes has caused de sitter inflation of branes and the exit from inflation is realized by the brane collision . in the following sections
, we shall look at the details of this scenario .
flux in the region i. this changes the cosmological constant in region i. after the collision , the single brane remains and the big - bang universe commences.,width=264,height=151 ]
the non - linear dynamics of de sitter branes embedded in 5-dimensional anti - de sitter space can be studied exactly without recourse to any approximation @xcite . however
the study of inhomogeneous perturbations about this background , when two de sitter branes are in relative motion , is a much harder problem .
therefore we will use a low - energy approximation @xcite in this case , valid when the hubble rate is smaller than the anti - de sitter scale . in this case the only extra degree of freedom coming from the 5-dimensional gravitational field is the radion , a scalar in the 4-dimensional effective theory , describing the distance between the two branes . before the collision , the radion field is non - minimally coupled to the 4-dimensional gravitational field on either brane and the system is described by the scalar - tensor theory .
hence , it is useful to look at the cosmological evolution both from the induced metric frame on the bulk brane and the einstein frame . after the collision the radion field vanishes and
the low - energy system is described by 4-dimensional einstein gravity .
as the collision changes the theory discontinuously , the evolution of the universe in the einstein frame looks strange .
we find a contracting universe immediately before the collision which will start to expand abruptly after the collision .
however , in the induced metric frame which is a natural frame for an observer , the universe is always expanding .
except for the collision point , the low energy approximation can be applied @xcite .
the detailed derivation of the effective action for our system can be found in the appendix .
an alternative derivation is given in ref .
the induced metric on the bulk brane is @xmath26 .
the low - energy effective action on the bulk brane is @xmath27 \ , \end{aligned}\ ] ] where @xmath28 represents the radion field , with @xmath29 when the branes are coincident .
the ads length scale @xmath30 is given in eq .
( [ eom : bulk ] ) . for simplicity , we denote @xmath31 as @xmath32 .
the effective potential for the radion is given by @xmath33 here we have defined the dimensionless parameters @xmath34 we obtain the static case of single minkowski brane at fixed distance in ads when @xmath35 , @xmath36 and @xmath37 . as we have @xmath38 , the effective cosmological constant in region @xmath7 , @xmath39 , is always smaller than that in region @xmath8 , @xmath12 , due to the flux in the region @xmath7 . from ( a2 ) , we see @xmath40 , i.e. @xmath41 . the equation of motion for the radion is @xmath42 the dynamics of the radion field thus appear non - trivial , and as the effective theory in the induced metric frame is a scalar - tensor theory , the cosmological dynamics will also depend on this non - trivial dynamics .
however , the equations of motion for the induced metric can be written as @xmath43 where @xmath44 represents the projected 5-dimensional weyl tensor on the brane @xcite and is determined by the radion field , @xmath45 this satisfies the traceless condition @xmath46 and indicates the radion field behaves as the conformally invariant matter on the brane .
hence , for the isotropic and homogeneous universe , @xmath47 the effect of the bulk gravity , @xmath44 , acts like a radiation fluid , and the friedman equation is obtained as @xmath48 where @xmath49 denotes the hubble parameter of the induced spacetime and the constant of integration @xmath50 is often called the dark radiation . as we know @xmath41 , in order to have the positive effective cosmological constant , we assume @xmath51 in practice , once the universe starts to expand , the dark radiation term @xmath52 becomes soon negligible .
thus , we see that the induced spacetime on the brane rapidly approaches de sitter at late times ( @xmath53 ) .
we can obtain enough e - foldings on the brane , if we fine tune the initial brane positions ( and hence the initial value of @xmath28 , see fig .
[ fig:2 ] ) .
thus , we have obtained an inflationary universe on the brane . .
for these parameters , hubble parameter becomes 1.2413 in unit of @xmath54 , the number of e - foldings becomes about 74.,width=264,height=188 ] when @xmath28 reaches @xmath55 , the inflation is suddenly terminated by the collision of branes in our scenario .
we have shown that we can obtain a de sitter inflationary universe in the induced metric frame due to the existence of the flux between the two branes in the bulk .
it is interesting to see this in the conformally related einstein frame @xcite .
in the einstein frame , the metric is @xmath56~h_{\mu\nu}$ ] , using the variable @xmath57 .
then the action reduces to @xmath58 \end{aligned}\ ] ] where the effective potential for the minimally - coupled radion @xmath59 is given by @xmath60 .,width=264,height=188 ] the above potential is depicted in figure [ fig:3 ] .
the unstable extremum corresponds to the static two de sitter brane solution . .
the horizontal axis represents the cosmic time in jordan frame .
, width=264,height=188 ] this extremum is located at @xmath61 determined by @xmath62 in that case , the potential energy becomes @xmath63 the effective mass - squared is negative , @xmath64 from the above , one can read off the radion effective mass @xmath65 , where @xmath66 represents the hubble parameter at @xmath61 .
this indicates the linearised instability of the static two de sitter brane system , consistent with previous analyses @xcite .
let us assume two branes are separated enough initially , which means the radion is located at near the maximum .
the radion then starts to roll down the hill and reaches the collision point @xmath67 which corresponds to @xmath68 .
in contrast to the other models discussing collisions @xcite , our model has no singularity at the collision because the bulk region never disappears in our model .
the hubble parameter in the einstein frame , @xmath69 , is related to the hubble parameter in the induced metric frame , @xmath49 , as @xmath70 the typical behavior of the hubble parameter in the einstein frame is depicted in figure [ fig:4 ] .
while @xmath28 remains almost constant the hubble rate is also nearly constant and we have almost exponential expansion in both the induced and einstein frames . due to the suppression factor @xmath71 ,
the hubble rate in the einstein frame is much smaller than that in the induced frame .
however , the e - folding number is almost the same as that in the jordan frame because the time in the einstein frame becomes longer by the conformal factor @xmath72 . shortly before the collision
, we find that the effective potential in the einstein frame becomes negative , the universe recollapses and immediately before the collision the universe is contracting in the einstein frame , though the universe is always expanding in the induced metric frame .
we need a fully 5-dimensional consideration to give a rule for evolution through the collision .
we consider the simplest case of a completely inelastic collision where the bulk brane is absorbed by the boundary brane . to describe the @xmath0-symmetric collision of the branes it is useful to consider the complete system of four branes , consisting of the incoming @xmath0 symmetric boundary brane , two copies of the bulk brane , and the outgoing @xmath0 symmetric brane ( see figure [ fig:5 ] ) .
symmetry.,width=207,height=207 ] a detailed analysis gives consistency conditions for the collision @xcite .
these are equivalent to relativistic energy - momentum conservation @xcite @xmath73 where the pseudo - lorentz factor between the colliding branes can be written as @xmath74 where @xmath75 at low energy ( small `` angles '' ) this reduces to the `` newtonian '' energy conservation law @xmath76 plus the momentum conservation law @xmath77 \
.\end{aligned}\ ] ] hence , in order to obtain an expanding universe after the collision we have to impose the constraint @xmath78 in addition to this constraint , we require that the expansion rate of the bulk brane is bigger than the @xmath0 symmetric brane , @xmath79 , in order to cause a collision .
it is easy to meet both these requirements , as we set @xmath41 . after the collision
, we assume the branes coalesce and behave as a single brane .
we further assume the resulting brane tension is given by the rs value , @xmath80 so that the effective cosmological constant on the brane vanishes after the collision .
this is the brane - world equivalent of the usual assumption that the inflaton potential is zero at its minimum . at the collision , the additional energy density ( above the rs brane tension )
is assumed to be transferred to light degrees of freedom on the brane , i.e. , radiation .
hence , the subsequent evolution will be governed by the standard friedmann equation for the hot big bang with small kaluza - klein corrections .
as the radion disappears after the collision , the difference between the induced metric frame and the einstein frame also disappears . in the einstein frame , the abrupt change of the contracting phase to the expanding phase can not be described within the 4-dimensional effective theory .
this clearly shows that our model is different from a conventional 4-dimensional inflationary model .
having constructed a homogeneous cosmological model , we can now consider the spectrum of inhomogeneous perturbations that would be expected due to small - scale quantum fluctuations . as there is no singularity at the collision
, we can unambiguously follow the evolution of fluctuations generated during inflation through the collision .
to study the behavior of fluctuations before the collision , it is convenient to work in the einstein frame in which the radion is minimally coupled to the metric .
we have the equations of motion in the einstein frame @xmath81 the homogeneous and istropic background metric is @xmath82 \ .\end{aligned}\ ] ] and we have the einstein equations become @xmath83 where a prime denotes a derivative with respect to the conformal time @xmath84 and @xmath85 . during the de sitter phase ,
the solution is @xmath86 where @xmath87 and @xmath88 now we can examine possible fluctuations separately . let us consider first the tensor perturbations @xmath89 \ , \end{aligned}\ ] ] where the tensor perturbations satisfy @xmath90 .
we can reduce the einstein equations to @xmath91 before the radion starts to roll down the hill , the background spacetime is the de sitter spacetime ( [ desb ] ) .
the positive frequency mode function is the standard one @xmath92 where @xmath93 is the hankel s function of the first kind .
this gives the standard flat spectrum for the primordial gravitational waves . during the roll down phase
, the universe will begin to contract rapidly from a numerical calculation , we see that the contracting phase is negligible in practice . during the collision process ,
as the gravitational waves are independent of gauge , we will have a flat spectrum on long wavelengths after the collision when the standard radiation dominated era begins .
to study the behavior of the radion fluctuations , we express the metric perturbation in the einstein frame as @xmath94 \ , \label{mtrc : sclr - ptb}\end{aligned}\ ] ] where @xmath95 represent the gauge - dependent scalar metric perturbations .
a convenient gauge - invariant combination is the comoving curvature perturbation , which is the intrinsic curvature perturbation on uniform - radion hypersurfaces : @xmath96 the second - order action for the curvature perturbation @xmath97 is @xmath98\ , , \label{ptb : action}\end{aligned}\ ] ] where @xmath99 the equation of motion for @xmath97 is @xmath100 therefore , on large scales , @xmath97 is constant .
equivalently we can work in terms of the radion on uniform - curvature hypersurfaces @xmath101 during the de sitter phase , the equation for @xmath102 can be written as @xmath103 where we have used eq .
hence , the positive frequency mode becomes @xmath104 where @xmath105 is the hankel s function of the first kind .
thus , we can read off the power spectrum of @xmath102 as @xmath106 . despite the exponential expansion the spectrum on large scales becomes red during the de sitter inflation because the radion has negative effective mass - squared .
these field fluctuations can be translated to the curvature perturbations on comoving hypersurfaces via eq .
( [ qr ] ) .
the coefficient @xmath107 is independent of scale and hence the comoving curvature perturbation shares the same red spectrum , @xmath108 . near the collision
, we expect the spectrum to be blue because of the rapid contraction , but this only affects small scales . finally , we need to calculate the curvature perturbation on uniform - density hypersurfaces , @xmath109 , on large scales after the collision .
this should then remain constant on large scales for adiabatic perturbations after the collision , even in the brane - world @xcite , simply as a consequence of local energy conservation @xcite . in the low - energy limit ,
energy conservation at the collision gives eq .
( [ rhoflo ] ) which implies that the collision hypersurface will be a uniform - energy hypersurface .
thus @xmath109 after the collision coincides with the curvature perturbation of this collision hypersurface .
we define the collision hypersurface in terms of the low - energy effective theory before the collision by the condition that @xmath68 .
thus the collision hypersurface is a uniform - radion hypersurface and we can identify the curvature perturbation on this hypersurface as the comoving curvature perturbation , @xmath110 ( where the negative sign comes from different historical conventions for the sign of the curvature perturbation ) . one might worry that @xmath97 was calculated in the einstein frame and the collision hypersurface corresponds to a physical hypersurface on the two branes .
in fact the comoving curvature perturbation is invariant under any conformal transformation that is function of the radion field , @xmath111 , as the conformal transformation then corresponds to a uniform rescaling on uniform-@xmath28 hypersurfaces , leaving the comoving curvature perturbation conformally invariant .
hence @xmath97 does describe the curvature perturbation on the physical collision hypersurface .
alternatively we can calculate the curvature perturbation on the collision hypersurface using the @xmath112 formalism @xcite .
this says that the curvature perturbation is given by the perturbed expansion with respect to an initial flat hypersurface , or @xmath113 where @xmath114 is the field perturbation on the initial spatially flat hypersurface at horizon exit .
this gives the usual expression @xmath115 for the comoving curvature perturbation during inflation . because the collision hypersurface is a uniform-@xmath59 hypersurface before the collision and a uniform - density hypersurface after the collision we have
@xmath116 .
however , we have shown that the comoving curvature perturbation has a steeply tilted red spectrum which means that these perturbations can not be responsible for the formation of the large scale structure in our universe . the curvaton mechanism @xcite provides a simple example of how scale - invariant perturbations could be generated in our model .
both the bulk and boundary branes experience a de sitter expansion before the collision .
the radion field acquires a red spectrum due to its negative mass - squared , but any minimally - coupled , light scalar field living on either brane would acquire a scale - invariant spectrum of perturbations before the collision . thus we add to our basic model an additional degree of freedom which , to be specific , we assume lives on the bulk brane .
the action for the curvaton @xmath117 in the induced metric frame takes the simple form @xmath118\end{aligned}\ ] ] vacuum fluctuations of a massless scalar field in de sitter spacetime have the same time - dependence as the graviton ( [ qoft ] ) , becoming frozen - in on large scales .
normalising to the bunch - davies vacuum state on small scales leads to the standard scale - invariant spectrum , @xmath119 , on super - hubble scales .
hence , the fluctuations of the curvaton have a completely flat spectrum .
this is similar the way the curvaton mechanism was originally proposed in the pre - big bang scenario @xcite where axion fields have a non - trivial coupling to the dilaton field in the einstein frame , but are minimally coupled in a conformally related frame , which can give rise to a scale - invariant spectrum if the expansion is de sitter in that conformal frame @xcite .
the curvaton mechanism requires that the curvaton has a non - zero mass after the collision , which seems entirely natural if the collision is associated with some symmetry - breaking in the degrees of freedom on the brane .
the energy density of any massive scalar field will grow relative to the radiation density once the hubble rate drops below its effective mass .
indeed the particles must decay before primordial nucleosynthesis to avoid an early matter domination in conflict with standard predictions of the abundances of the light elements .
but if the curvaton comes to contribute a significant fraction of the total energy density before it decays then perturbations in the curvaton will result in primordial density perturbations on large scales capable of seeding the large scale structure in our universe @xcite .
we have proposed a geometrical model of inflation where inflation on two branes is driven by the flux generated by the brane charge and terminated by the brane collision with charge annihilation .
we assume the collision process is completely inelastic and the kinetic energy is transformed into the thermal energy .
after the collision the two branes coalesce together and behave as a single brane universe with zero effective cosmological constant .
the four - dimensional , low - energy effective theory in the einstein frame has to change abruptly at the collision point .
therefore , our model needs to be consistently described using 5-dimensional gravity .
this can be done using consistency conditions at the collision that ensure energy - momentum conservation in the 5-dimensional theory .
as the collision process has no singularity in the 5-dimensional gravity , we can unambiguously follow the evolution of inhomogeneous vacuum fluctuations about this homogenous background during the whole history of the universe .
it turns out that the radion fluctuations produced during inflation have a steep red spectrum and can not produce the present large - scale structure of our universe .
instead the curvature perturbations observed today must be generated by the curvaton or some similar mechanism from initially isocurvature excitations of massless degrees of freedom on one or other of the branes before the collision .
the primordial gravitational waves produced are likely to be difficult to detect .
our model does suffer from a significant fine tuning problem of the initial conditions .
the radion field must start very close to an unstable extremum of its potential in order to obtain sufficient inflation .
however , the radion field has a geometrical meaning so that the initial value is determined by the initial separation between two branes .
sufficient separation of the initial brane positions does not seem so unnatural , and the fine tuning valu of the radion may not be so serious .
a quantum cosmological consideration @xcite would be needed to give a more insight on this initial condition problem .
it might be that one can find an instanton solution to describe tunnelling to this extremum .
alternatively , in a semi - classical model such as stochastic inflation , it may be that quantum fluctuations drive the field to the top of the potential before classical evolution takes over and the field rolls down , leading to the branes to collide .
we stress that our model is fundamentally 5-dimensional in nature .
it can not be constructed starting from a 4-dimensional theory .
it would of course be interesting to see if it is possible to embed our model into string theory model , but that goes beyond the scope of this paper .
_ note added : _ while this work was being written up koyama and koyama @xcite submitted a related paper deriving an equivalent effective action for anti - d - branes in a type iib string model .
we wish to thank roy maartens for useful discussions .
sk and js would like to thank the portsmouth icg for its hospitality and financial support , under pparc grant ppa / v / s/2001/00544 .
this work is supported by the grant - in - aid for the 21st century coe `` center for diversity and universality in physics '' from the ministry of education , culture , sports , science and technology ( mext ) of japan .
this work was also supported in part by grant - in - aid for scientific research fund of the ministry of education , science and culture of japan no .
155476 ( sk ) , no.14540258 and no.17340075 ( js ) .
we shall start with the three - brane system and take the two - brane limit to derive the low energy effective action for the two - brane system .
this allows us to use a simple moduli approximation method .
assuming there exists no matter on the third brane , this procedure can be justified . in the low energy limit
, the configuration should be almost static . as we have no matter in the bulk
, the bulk metric should be anti - de sitter spacetime .
hence , let us first consider the static solution of the form @xmath120 where @xmath121 and @xmath122 represents the warp factor in each region .
the bulk equation of motion imply the relation between the curvature scale in the bulk and the flux as @xmath123 we also have junction conditions for the metric which give the relation @xmath124 if we assume the inflating bulk brane @xmath125 , we have @xmath126 .
the above implies the relation @xmath127 the junction conditions for 5-form fields give @xmath128 the last relation is nothing but the charge conservation law .
the static solution is realized under the relations eqs .
( [ econservation ] ) and ( [ chgconservation ] ) . in this paper
, we set @xmath129 .
this implies @xmath130 .
let us employ the moduli approximation method @xcite which can be shown to be valid at low energy @xcite .
( an alternative derivation appears in ref .
@xcite . ) in this method , we can use the factorized metric ansatz @xmath131 the brane positions are denoted by @xmath132 and @xmath133 .
now , we calculate the bulk action , the action for each brane tensions , and the gibbons - hawking terms , separately .
let us start with the calculation of the bulk action . for the metric ( [ fmetric ] ) , using the gauss equation , it is straightforward to relate the bulk ricci scalar to the 4-dimentional ricci scalar , @xmath134 we can write the contributions from each bulk as @xmath135 \nonumber\\ & = & \frac{1}{\kappa^2 } \int d^4 x \sqrt{-g^i } \left [ -\frac{\ell_i}{2 } \left\ { b_i^2 ( \phi_2 ) -b_i^2(\phi_1 ) \right\}r(g^i)\nonumber \right .
\\ & & \left .
+ \frac{2}{\ell_i } \left\ { b_i^4 ( \phi_2 ) - b_i^4 ( \phi_1 ) \right\}\right ] \label{bulki}\end{aligned}\ ] ] where we used the bulk equation of motion ( [ eom : bulk ] ) in the first line .
the factor 2 over @xmath2 comes from the @xmath0 symmetry of this spacetime and we neglected the second order quantities . in the same way for region
ii , we have @xmath136 \nonumber\\ & & = \frac{1}{\kappa^2 } \int
d^4 x \sqrt{-g^{i\!i } } \left [ -\frac{\ell_{i\!i}}{2 } \left\ { b_{i\!i}^2 ( \phi_3 ) -b_{i\!i}^2 ( \phi_2 ) \right\}r(g^{ii})\nonumber \right .
\\ & & \left.\qquad + \frac{2}{\ell_{i\!i } } \left\ { b_{i\!i}^4 ( \phi_3 ) -b_{i\!i}^4 ( \phi_2 ) \right\}\right]\ .
\label{bulkii}\end{aligned}\ ] ] in order to calculate the action for each brane tensions , we need to know the induced metric .
the induced metric on the left boundary brane becomes @xmath137 and the similar formula holds for other branes .
then , the contributions from the brane tension terms become @xmath138 \ , \quad \label{tension1}\end{aligned}\ ] ] @xmath139 \nonumber\\ & = & \!\!\!\ ! -2\sigma_2
\int d^4 x \sqrt{-g^{i\!i } } b_{i\!i}^4 ( \phi_2 ) \left [ 1 + \frac{1}{2 b_{i\!i}^2 ( \phi_2 ) } \left ( \partial \phi_2 \right)^2 \right ] \qquad\quad \label{tension2}\end{aligned}\ ] ] and @xmath140 \ , \quad \label{tension3}\end{aligned}\ ] ] where the factor 2 in front of @xmath141 comes from the @xmath0 symmetry .
let us turn to the calculation of gibbons - hawking terms .
the extrinsic curvature in leading order is given by @xmath142 \ , \end{aligned}\ ] ] where @xmath143 is the normal vector to the brane defined by @xmath144 hence , the gibbons - hawking terms are @xmath145 \ , \
\quad \label{gh1}\end{aligned}\ ] ] @xmath146 \nonumber\\ -\frac{2}{\kappa^2 \ell_i } \int d^4 x \sqrt{-g^i } \left [ 4 b_i^4 ( \phi_2 ) + 3 b_i^2 ( \phi_2 ) \left ( \partial \phi_2 \right)^2 \right ] \ , \qquad \label{gh2}\end{aligned}\ ] ] and @xmath147 \ , \ \ \label{gh3}\end{aligned}\ ] ] where the factor 2 over @xmath2 in @xmath148 again comes from the @xmath0 symmetry substituting eqs.([bulki ] ) , ( [ bulkii ] ) , ( [ tension1 ] ) , ( [ tension2 ] ) ( [ tension3 ] ) , ( [ gh1 ] ) , ( [ gh2 ] ) and ( [ gh3 ] ) into the 5-dimensional action ( [ 5d : action ] ) , the resultant 4-dimensional effective action can be summarized by the following . the curvature part becomes @xmath149 r ( g^i ) \nonumber\\ & & \hspace{-5 mm } + \frac{\ell_{i\!i}}{2\kappa^2 } \int d^4 x \sqrt{-g^{i\!i } } \left [ b_{i\!i}^2 ( \phi_2 ) - b_{i\!i}^2 ( \phi_3 ) \right ] r ( g^{i\!i } ) \ .
\quad\end{aligned}\ ] ] the kinetic part of radions are @xmath150 \nonumber\\ & & \hspace{-2 mm } + \frac{3}{\kappa^2 \ell_{i\!i } } \int d^4 x \sqrt{-g^{i\!i } } \left [ b_{i\!i}^2 ( \phi_2 ) \left ( \partial \phi_2 \right)^2 - b_{i\!i}^2 ( \phi_3 ) \left ( \partial \phi_3 \right)^2 \right ] \ . \qquad\end{aligned}\ ] ] the potential energy will be induced as @xmath151 \nonumber\\ & & \hspace{-5 mm } -\int d^4 x \sqrt{-g^{i\!i } } \left ( \sigma_3 + \frac{6}{\kappa^2 \ell_{i\!i } } \right ) b_{i\!i}^4 ( \phi_3 ) \ .\end{aligned}\ ] ] now we can write down the effective action for the bulk brane .
the continuity condition of the metric is given by @xmath152 where @xmath26 is the induced metric of the bulk brane . defining the variables @xmath153 we obtain @xmath154 r ( h ) \nonumber \\ & & + \frac{3\ell_i}{\kappa^2 } \int d^4 x \sqrt{-h } \left [ \left(\partial \psi \right)^2 - \alpha \left ( \partial \phi \right)^2 \right ] \nonumber\\ & & + \frac{6}{\kappa^2 \ell_i } \int d^4 x \sqrt{-h } \left [ \left ( 1 - \beta_1 \right )
\psi^4 \nonumber
\right.\\ & & \left . - \left ( 1 - \frac{1}{\alpha } + \beta_2 \right ) -\left(\frac{1}{\alpha } + \beta_3 \right ) \phi^4 \right ] \ .\end{aligned}\ ] ] here we have defined the dimensionless parameters @xmath155 as we have the relation @xmath156 , the relation @xmath41 also holds . after taking the two - brane limit @xmath157 , namely , @xmath158 , we obtain the low energy effective action for the two - brane system .
@xmath159 \ .
\nonumber\end{aligned}\ ] ] the scalar variable @xmath28 which describes the distance between two positive tension branes is called as the radion .
p. horava and e. witten , nucl .
b * 460 * , 506 ( 1996 ) [ arxiv : hep - th/9510209 ] ; i. antoniadis , n. arkani - hamed , s. dimopoulos and g. r. dvali , phys .
b * 436 * , 257 ( 1998 ) [ arxiv : hep - ph/9804398 ] ; a. lukas , b. a. ovrut , k. s. stelle and d. waldram , phys .
d * 59 * , 086001 ( 1999 ) [ arxiv : hep - th/9803235 ] ; n. arkani - hamed , s. dimopoulos and g. r. dvali , phys .
b * 429 * , 263 ( 1998 ) [ arxiv : hep - ph/9803315 ] .
g. r. dvali and s. h. h. tye , phys .
b * 450 * , 72 ( 1999 ) [ arxiv : hep - ph/9812483 ] .
s. kachru , r. kallosh , a. linde and s. p. trivedi , phys .
d * 68 * , 046005 ( 2003 ) [ arxiv : hep - th/0301240 ] ; + s. kachru , r. kallosh , a. linde , j. maldacena , l. mcallister and s. p. trivedi , jcap * 0310 * , 013 ( 2003 ) [ arxiv : hep - th/0308055 ] ; r. maartens , living rev .
* 7 * , 7 ( 2004 ) [ arxiv : gr - qc/0312059 ] ; p. brax , c. van de bruck and a. c. davis , rept .
phys . * 67 * , 2183 ( 2004 ) [ arxiv : hep - th/0404011 ] ; s. kanno and j. soda , arxiv : hep - th/0407184 .
r. maartens , d. wands , b. a. bassett and i. heard , phys .
d * 62 * , 041301 ( 2000 ) [ arxiv : hep - ph/9912464 ] .
d. langlois , r. maartens and d. wands , phys .
b * 489 * , 259 ( 2000 ) [ arxiv : hep - th/0006007 ] .
s. kobayashi , k. koyama and j. soda , phys .
b * 501 * , 157 ( 2001 ) [ arxiv : hep - th/0009160 ] .
y. himemoto and m. sasaki , phys .
d * 63 * , 044015 ( 2001 ) [ arxiv : gr - qc/0010035 ] .
l. randall and r. sundrum , phys .
* 83 * , 3370 ( 1999 ) [ arxiv : hep - ph/9905221 ] ; s. kanno and j. soda , phys .
d * 66 * , 083506 ( 2002 ) [ arxiv : hep - th/0207029 ] ; s. kanno and j. soda , phys .
d * 66 * , 043526 ( 2002 ) [ arxiv : hep - th/0205188 ] ; + s. kanno and j. soda , astrophys .
space sci .
* 283 * , 481 ( 2003 ) [ arxiv : gr - qc/0209087 ] ; + s. kanno and j. soda , gen .
grav . * 36 * , 689 ( 2004 ) [ arxiv : hep - th/0303203 ] .
t. wiseman , class .
* 19 * , 3083 ( 2002 ) [ arxiv : hep - th/0201127 ] ; + t. shiromizu and k. koyama , phys . rev .
d * 67 * , 084022 ( 2003 ) [ arxiv : hep - th/0210066 ] .
d. langlois , k. maeda and d. wands , phys . rev .
lett . * 88 * , 181301 ( 2002 ) [ arxiv : gr - qc/0111013 ] .
t. shiromizu , k. koyama and t. torii , phys .
d * 68 * , 103513 ( 2003 ) [ arxiv : hep - th/0307151 ] ; k. takahashi and t. shiromizu , phys .
d * 70 * , 103507 ( 2004 ) [ arxiv : hep - th/0408043 ] ; p.
brax and n. chatillon , jhep * 0312 * , 026 ( 2003 ) [ arxiv : hep - th/0309117 ] ; p. brax and n. chatillon , phys .
d * 70 * , 106009 ( 2004 ) [ arxiv : hep - th/0405143 ] .
p. binetruy , c. deffayet and d. langlois , nucl .
b * 615 * , 219 ( 2001 ) [ arxiv : hep - th/0101234 ] .
n. kaloper , phys .
d * 60 * , 123506 ( 1999 ) [ arxiv : hep - th/9905210 ] .
p. bowcock , c. charmousis and r. gregory , class .
* 17 * , 4745 ( 2000 ) [ arxiv : hep - th/0007177 ] .
s. kanno and j. soda , phys .
b 588 , 203 - 209 ( 2004 ) [ arxiv : hep - th/0312106 ] ; + p. l. mcfadden and n. turok , phys . rev .
d * 71 * , 021901 ( 2005 ) [ arxiv : hep - th/0409122 ] .
l. cotta - ramusino , `` low energy effective theory for brane world models '' , mphil thesis , university of portsmouth ( 2004 ) .
t. shiromizu , k. i. maeda and m. sasaki , phys .
d * 62 * , 024012 ( 2000 ) [ arxiv : gr - qc/9910076 ]
. s. nojiri , o. obregon , s. d. odintsov and v. i. tkach , phys .
d * 64 * , 043505 ( 2001 ) [ arxiv : hep - th/0101003 ] .
u. gen and m. sasaki , prog .
phys . * 105 * , 591 ( 2001 ) [ arxiv : gr - qc/0011078 ] .
z. chacko and p. j. fox , phys . rev .
d * 64 * , 024015 ( 2001 ) [ arxiv : hep - th/0102023 ] . c. r. contaldi , l. kofman and m. peloso , jcap * 0408 * , 007 ( 2004 ) [ arxiv : hep - th/0403270 ] . j. khoury , b. a. ovrut , p. j. steinhardt and n. turok , phys .
d * 64 * , 123522 ( 2001 ) [ arxiv : hep - th/0103239 ] .
u. gen , a. ishibashi and t. tanaka , phys .
d * 66 * , 023519 ( 2002 ) [ arxiv : hep - th/0110286 ] .
m. bucher , arxiv : hep - th/0107148 .
s. kanno , m. sasaki and j. soda , prog .
phys . * 109 * , 357 ( 2003 ) [ arxiv : hep - th/0210250 ] .
j. garriga and t. tanaka , phys .
d * 65 * , 103506 ( 2002 ) [ arxiv : hep - th/0112028 ] .
j. j. blanco - pillado and m. bucher , phys .
d * 65 * , 083517 ( 2002 ) [ arxiv : hep - th/0111089 ] .
a. neronov , jhep * 0111 * , 007 ( 2001 ) [ arxiv : hep - th/0109090 ] .
d. langlois , r. maartens , m. sasaki and d. wands , phys .
d * 63 * , 084009 ( 2001 ) [ arxiv : hep - th/0012044 ] .
d. wands , k. a. malik , d. h. lyth and a. r. liddle , phys .
d * 62 * , 043527 ( 2000 ) [ arxiv : astro - ph/0003278 ] . m. sasaki and e. d. stewart , prog .
phys . * 95 * , 71 ( 1996 ) [ arxiv : astro - ph/9507001 ] .
k. enqvist and m. s. sloth , nucl .
b * 626 * , 395 ( 2002 ) [ arxiv : hep - ph/0109214 ] .
d. h. lyth and d. wands , phys .
b * 524 * , 5 ( 2002 ) [ arxiv : hep - ph/0110002 ] .
t. moroi and t. takahashi , phys .
b * 522 * , 215 ( 2001 ) [ erratum - ibid .
b * 539 * , 303 ( 2002 ) ] [ arxiv : hep - ph/0110096 ] .
e. j. copeland , r. easther and d. wands , phys .
d * 56 * , 874 ( 1997 ) [ arxiv : hep - th/9701082 ] .
j. garriga and m. sasaki , phys .
d * 62 * , 043523 ( 2000 ) [ arxiv : hep - th/9912118 ] . k. koyama and j. soda , phys .
b * 483 * , 432 ( 2000 ) [ arxiv : gr - qc/0001033 ] .
k. koyama and k. koyama , arxiv : hep - th/0505256 .
j. garriga , o. pujolas and t. tanaka , nucl .
b * 655 * , 127 ( 2003 ) [ arxiv : hep - th/0111277 ] ; + p. brax , c. van de bruck , a. c. davis and c. s. rhodes , phys .
d * 67 * , 023512 ( 2003 ) [ arxiv : hep - th/0209158 ] ; + g. a. palma and a. c. davis , phys .
d * 70 * , 106003 ( 2004 ) [ arxiv : hep - th/0407036 ] ; + j. l. lehners and k. s. stelle , nucl .
b * 661 * , 273 ( 2003 ) [ arxiv : hep - th/0210228 ] .
s. kanno and j. soda , phys .
d * 71 * , 044031 ( 2005 ) [ arxiv : hep - th/0410061 ] .
s. kanno , arxiv : hep - th/0504087 . | we propose a geometrical model of brane inflation where inflation is driven by the flux generated by opposing brane charges and terminated by the collision of the branes , with charge annihilation .
we assume the collision process is completely inelastic and the kinetic energy is transformed into the thermal energy after collision .
thereafter the two branes coalesce together and behave as a single brane universe with zero effective cosmological constant . in the einstein frame , the 4-dimensional effective theory changes abruptly at the collision point .
therefore , our inflationary model is necessarily 5-dimensional in nature .
as the collision process has no singularity in 5-dimensional gravity , we can follow the evolution of fluctuations during the whole history of the universe .
it turns out that the radion field fluctuations have a steeply tilted , red spectrum , while the primordial gravitational waves have a flat spectrum . instead
, primordial density perturbations could be generated by a curvaton mechanism . |
besides the well - known examples of tunneling processes that permeate the literature in molecular and nuclear physics @xcite , in recent years it has been shown that quantum tunneling also appears in as broad physical situations as magnetic moments in spin glass systems , single magnetic impurities in a crystal field , a single domain , a ferromagnetic , or antiferromagnetic , domain wall , and also in spin tunneling in molecules @xcite , emphasizing the importance of the angle / angular momentum degree of freedom in these cases . this quantum effect occuring in such mesoscopic , and even in macroscopic systems points to the necessity of a discussion on how such tunneling can be realized and controlled @xcite . in what concerns the case of spin tunneling , or the quantum tunneling of magnetization , qtm
, the theoretical approaches have pointed to some different ways of treating this problem .
in particular , the spin tunneling was first treated by van hemmen and sto @xcite , and by enz and schilling @xcite . from different perspectives ,
various approaches were introduced : by using a generalization of the usual wkb method adapted to spin systems @xcite , or by using feyman s path integral treatment of quantum mechanics @xcite , and also by using coherent states @xcite . on the other hand , from the experimental point of view ,
great advances have been made since the synthesis of the mn12-acetate molecule in 1980 @xcite . among the properties of this molecule it emerged an indication of spin tunneling that has strongly motivated great efforts in characterizing and elucidating the pure quantum contribution to that phenomenon @xcite .
the discovery of the same indications of spin tunneling in the octanuclear iron cluster , fe8 @xcite , has thus reinforced the proposed idea of pure quantum spin tunneling , and formal treatments involving phenomenological hamiltonians were then widely proposed and discussed @xcite . in any case , the angular momentum operators obeying the standard commutation relations @xmath0 = i\varepsilon _ { ijk}j_{k}\ ] ] constitute the fundamental blocks from which the starting hamiltonian describing the spin system is constructed . in general , a hamiltonian is proposed based on experimental considerations on the symmetries of the physical system , and also under the assumption of low temperature such that the pure quantum effects dominate . in this sense , a general hamiltonian encompassing the molecule symmetries plus possible paralel and tranverse magnetic fields may be written as @xmath1 being @xmath2 the molecule hamiltonian , and the constants @xmath3 measure the intensity of the additional external magnetic fields .
this description clearly separates and takes into account the molecule total angular momentum degree of freedom as the dominant one and , as far as no other degree of freedom is directly and explicitly involved , it is considered a sound description of the system . as such , all of its kinematical content can be described in terms of combinations of the set of @xmath4 states generated by all the projections of the total angular momentum operator @xmath5 , which obeys the eigenvalue equation @xmath6 it is then clear that those hamiltonians can then be diagonalized within the state space defined by @xmath7 whose dimension is given by the total angular momentum @xmath5 . at the same time , as it is well known , since the hamiltonian is written in terms of the generators of that algebra , it commutes with the casimir operator thus giving at least one more constant of the motion - besides the energy ; in fact , all the symmetries present in the hamiltonian must also play a precise role , giving rise to characteristic features in the energy spectrum . in this paper
we want to extend the content of previous publications @xcite that proposed to construct and present the main features of a potential function , described in terms of an angle variable , which can be used to picture in a more intuitive way the spin tunneling . in those works ,
we have shown how to obtain such a potential and how pure quantum tunneling shows up as a transmition through an energy barrier . in the present contribution we show how to extract a new hamiltonian which also includes a kinetic term - not only a potential energy function - also described in terms of an angle variable , from a phenomenological hamiltonian , eq.([int1 ] ) , that can be used to study the spin tunneling . to construct such a hamiltonian , as for the kinematics , we use the @xmath8 coherent states @xcite , since they constitute an overcomplete basis of states , and also because they present the desired properties for the study of the systems of interest in the semiclassical limit . at the same time
they also include quantum effects such as zero point energies associated with the corresponding angle and angular momentum variables .
we also use the generator coordinate method ( gcm ) of wide use in nuclear and molecular physics@xcite mainly because it has been shown to be a useful tool when one deals with collective behavior in many - body systems , that is the case in the spin systems of interest where the tunneling degree of freedom has a suggested collective character . from the operational point of view
, the gcm is used so that an energy surface is generated that embodies off - diagonal matrix elements that account for the quantum corrections to the diagonal contribution which , by its turn , is basically dominated by the semiclassical contribution .
the present approach proposes to take into account those off - diagonal contributions by calculating the moments of that energy distribution ; the zeroth - order moment is associated with the potential function , while the second - order moment is associated with an inertia function : together they constitute our approximate hamiltonian .
it is clear that the higher order terms can also be obtained so that the complete series gives rise to a new representation of the starting phenomenological hamiltonian .
a truncated series up to the second term is then an approximate description of the system , and its validity must necessarily be discussed .
it is precisely the use of the @xmath9 coherent states that makes this study feasible since our moment expansion of the energy surface is then akin to the @xmath10 expansion discussed in the literature @xcite , which also makes explicit use of the coherent states . in the present context
, it is the total spin of the molecule , @xmath11 , the quantum number that determines how many terms we must keep in the moment expansion series in order to get a reliable hamiltonian .
it is worth noticing that the present approach is based on quantum grounds from the beginning in such a way that it does not need to be quantized in any form .
therefore , it has a different perspective from the approaches that start from a classical hamiltonian and then quantize it .
furthermore , what is important to keep in mind in the present context is that the associated pair of quantum operators related to the rotation degree of freedom must be treated properly@xcite .
this paper is organized as follows . in section
ii we introduce the @xmath8 coherent states in the discussion of the semiclassical approach for a general phenomenological hamiltonian , and show how some preliminary results are obtained which suggest that quantum corrections may be important in some cases . the gcm is presented in section iii where we show how the energy and overlap kernels of the method are obtained for the general phenomenological hamiltonian , and the new general angle - based hamiltonian is obtained . in this section
we also carry out the calculations for the soluble lipkin quasi - spin model @xcite since it is a valuable testing ground for systems of the kind we are interested in , and also for establishing a reliable truncation criterion for the moment series in terms of the value of the spin .
the result that we obtain guarantees the application , in section iv , of the second - order approximate hamiltonian to the description of two interesting molecules studied in the literature , namely , the mn12-acetate , and the fe8 cluster , and we show that we reproduce the experimental results to a good degree of accuracy .
the spin tunneling phenomenon occuring in these molecules is qualitatively discussed in this context , and the main aspects of the potential barrier are discussed . the role of the effective inertia function in the tunneling process
is also stressed .
finally the conclusions are presented in section v.
as a starting point to get a semiclassical approach for describing spin systems we may consider the @xmath8 coherent states defined in a general form as @xcite @xmath12 or also as @xmath13 where @xmath14 is a complex variable given by @xmath15 and @xmath11 characterizes the angular momentum state multiplet . these states
are nonorthogonal and normalized , and furthermore satisfy an overcompleteness relation @xmath16 which allows us to use them as an useful state basis . moreover , these states are a very interesting starting point when semiclassical aspects of the physical system must be discussed , as has been already discussed in the literature @xcite .
let us consider now a spin hamiltonian of the type that , for simplicity , does not include transversal magnetic fields @xmath17 written in terms of the basic operators obeying the @xmath8 algebra @xmath18 = \pm j_{\pm } \ ] ] and @xmath19 = 2j_{z}.\ ] ] it is clearly seen that by properly choosing the parameters of that hamiltonian we end up with particular expressions associated with physical systems we may want to study , for instance if we choose @xmath20 we have the lipkin quasi - spin solvable model @xcite @xmath21 by choosing @xmath22 we get the mn-12 acetate model , where a paralel magnetic fields is present @xcite , namely @xmath23 and we also see that by choosing @xmath24 we have a model hamiltonian for the octanuclear iron cluster , [ fe@xmath25o@xmath26@xmath27 , or just fe8 @xcite , in the absence of external magnetic fields @xmath28 we then recall that a description based on an angle representation of these models , out of that general hamiltonian , can be obtained by directly calculating its respective normalized matrix elements with the proposed coherent states . to this end
, we collect the basic results @xmath29 @xmath30 @xmath31 and @xmath32 ^{\dagger } .\ ] ] using these results we get the energy surface associated with the hamiltonian , namely , @xmath33
+ \label{hamisemi } \\ & & 2g2j\left ( 2j-1\right ) \sin ^{2}\frac{\alpha } { 2}\cos ^{2}\frac{\alpha } { 2}% \cos 2\xi \;. \notag\end{aligned}\ ] ] at this point , we see that we can determine the value of @xmath34 that minimizes the energy surface , leading to an energy curve in the variable @xmath35 .
let us analyse this expression in a particular case .
it is evident that , if we choose the parameters so that we describe the mn12-acetate molecule in the absence of paralel and transverse magnetic fields , i.e. , @xmath36 ( the energy surface in this case is @xmath34-independent ) , the energy curve @xmath37\ ] ] presents minima at @xmath38 for @xmath39 . for the values
@xmath40 @xmath41 , where @xmath42 is the boltzmann constant , and @xmath43 , as found in the literature @xcite , we obtain the energy minima at @xmath44 @xmath45 .
this is , however , precisely the ground state energy value obtained by diagonalizing the hamiltonian associated with the corresponding model , eq.([hbasica ] ) , within the @xmath46 states basis .
this result thus indicates that such a semiclassical approach does not embody quantum content enough to describe such system because , as we see , the bottom of the energy curve coincides with the ground state energy .
this fact shows that the quantum correlations related to the rotation degree of freedom may be important and , furthermore , it is not taken into account in this approach .
let us now go one step further in order to construct an energy surface that describes the spin system and at the same time that encompasses more quantum correlations than the previous procedure does . in this sense the generator coordinate method ( gcm )
@xcite extends that approach because it also embodies off - diagonal elements , i.e. , if we start from the @xmath8 coherent states we have to calculate the so - called energy and overlap kernels , as already discussed@xcite @xmath47 @xmath48 respectively .
it must be observed that hereafter we will only consider @xmath35 as our generator coordinate .
the variable @xmath34 will be properly chosen at the beginning in each case so as to minimize the energy surface as pointed out before . in the same way as we did in the last section ,
the basic matrix elements of the operators are collected @xmath49 @xmath50 and @xmath51 @xmath52 therefore , the gcm energy and overlap kernels for the general spin hamiltonian are respectively @xmath53 @xmath54 @xmath55 and @xmath56 the contribution of the off - diagonal matrix elements is immediately recognized in these expressions , and it can be easily verified that they are less and less important as @xmath11 increases , and that the semiclassical results are thus reobtained for @xmath57 , or when @xmath58 .
thus , the off - diagonal contributions may play an essential role in the description of spin systems with small and medium values of @xmath11 since they carry additional quantum information when compared with the semiclassical approach . instead of solving the gcm integral equation with these kernels , as
usually is done@xcite , we will show in what follows how a spin hamiltonian written in terms of an angle variable can be extracted from the gcm kernels .
once we have the gcm overlap and energy kernels , we will use a new represention for describing the spin system@xcite .
following the scheme already presented before@xcite , which makes use of fourier transforms , we get the new matrix representation for the spin system @xmath59 @xmath60 @xmath61 + \ ] ] @xmath62 + \ ] ] @xmath63 where @xmath64 the diagonalization of this hamiltonian matrix clearly gives the same results as those obtained by directly diagonalizing the matrix generated by the starting hamiltonian , eq.([hbasica ] ) , using the angular momentum @xmath65 and @xmath66 .
in fact , we introduce the new variables @xmath67 and @xmath68 , and also @xmath69 and @xmath70 in order to transform the gcm energy kernel into the new energy surface . the summations over @xmath71 that appear are restricted so that they run over even / odd values if @xmath72 is even / odd .
since we already have the gcm energy kernel , eq.([hgcm ] ) , we will consider hereafter the particular case of no transverse magnetic field , @xmath73 , for simplicity , and therefore we rewrite it as @xmath74 @xmath75 \cos ^{2j}\left ( \frac{\theta } { 2}\right ) -2\left| g\right| j\left ( 2j-1\right ) \cos ^{2j-2}\left ( \frac{\theta } { 2}\right ) -\ ] ] @xmath76 where the choices for the variable @xmath34 were already carried out .
after a direct but tedious calculation we obtain the general expression , @xmath77 } { 2\pi } \sum_{l=-2j}^{2j}e^{i\frac{l}{2}u}+\frac{% \frac{b}{2}-3\left| g\right| } { 2\pi } \sum_{l=-2j}^{2j}e^{i\frac{l}{2}% u}\left ( j^{2}-\frac{l^{2}}{4}\right ) -\ ] ] @xmath78 @xmath79 so far no approximation has been made and the above expression contains exactly the same physical content as the original phenomenological hamiltonian . before carrying out the calculations in order to obtain the moments of the above shown energy surface , we must recall some aspects of the weyl - wigner transformation techniques developed for continuous degrees of freedom @xcite .
it can be readily recognized that the power series expansion in the momentum @xmath80 that appears in the defining expression of the weyl - wigner transform @xmath81 where @xmath82 , and @xmath83 is the quantum phase space representative of the hamiltonian operator , can be rewritten in such a form that the new coefficients are the moments of the energy surface with respect to the difference of the original variables , namely @xmath84 is the n - th moment of the energy surface in the variable @xmath85 .
obviously we expect that only the even moments do not vanish because otherwise the system would violate time reversal symmetry , or , in other words , terms would appear in the hamiltonian that would be proportional to odd powers of the momentum .
in fact , it is the even parity of the gcm kernels that warrants us a time reversal invariant hamiltonian . in the continuous case , by using this procedure one ends up with a series the first two terms of which are dependent on the coordinate variable only , and proportional to the squared momentum respectively . in some cases , the proportionality coefficient in the second term may result in a function of the coordinate variable or , in other words , an effective mass may come out .
therefore , these first two terms can be seen as the potential and kinetic energy terms of an associated effective hamiltonian .
when these two terms dominate the series expansion , i.e. , higher order moments can be neglected _ _
_ _ they may even be completely absent in certain cases then the approximate two - terms effective hamiltonian can be useful to describe the physical system we started with .
now , the same can be seen to be valid in the present case of spin degrees of freedom if we consider the proper angular function in respect to which the moments of the energy surface must be taken . since the angle is not the direct variable to be used here since it does not obey a standard commutation relation with the angular momentum operator @xcite in contrast with the canonical case of ordinary coordinate and momentum discussed above the moments of the energy surface must be calculated with a periodic function of the angle .
thus , we perform the calculation with respect to @xmath86 @xcite , where @xmath87 , as already defined . the zeroth order moment will give us the potential function in terms of the angle variable , as has been already presented in the past @xcite .
since the zeroth order is given by @xmath88 where @xmath89 , we end up with @xmath90 } { 2}+\frac{b-6\left| g\right| } { 2}j^{2}-\ ] ] @xmath91 @xmath92 where @xmath93 is the angular momentum state space dimension . by its turn
, the second moment will give us the expression associated with the inertia function .
in other words , @xmath94 leads to @xmath95 @xmath96 \sqrt{1-\frac{n\left ( n+1\right ) } { j\left ( j+1\right ) } } \;-\ ] ] @xmath97 .\ ] ] it is then a direct matter to see that @xmath98 where @xmath99 is the corresponding effective mass function associated with the spin system .
we now propose that these first two moments must suffice to write a hamiltonian that can , to an acceptable degree of accuracy , describe systems consisting of a suitably great number of spins . in this form , in this approximation the hamiltonian is written as @xmath100 and we straightforwardly get the eigenvalues @xmath101 as well as the eigenfunctions @xmath102 by solving the schrdinger equation @xmath103 by a fourier analysis . as such , the spin number for which the validity of this proposal is assured can be estimated by introducing a model hamiltonian and comparing the results obtained from this schrdinger equation with the exact ones , obtained from a direct diagonalization of that model hamiltonian in a @xmath104 basis , for a great range of values of the spin quantum number @xmath11 . in what follows
we will discuss the validity of that approximation using the lipkin model@xcite .
as already pointed out before , the lipkin quasi - spin model is singled out by assuming the parameters : @xmath105 , @xmath20 , and @xmath106 , where @xmath107 , so that the gcm kernels read @xmath108 @xmath109 \},\ ] ] and @xmath110 respectively , if we measure energy in unities of @xmath111 .
we obtain the potential function directly from eq.([pot ] ) , which gives @xmath112 @xmath113 in the same form we get the expression related to the inertia function @xmath114 @xmath96 \sqrt{1-\frac{n\left ( n+1\right ) } { j\left ( j+1\right ) } } -\ ] ] @xmath115 .
\label{exainer}\ ] ] with these expressions we can discuss now the approximations proposed before .
first , if the number of quasi - spins is such that @xmath116 , the summations over @xmath72 can be substituted by integrals and we obtain the expressions @xmath117 and @xmath118 @xmath119 respectively .
furthermore , the remaining summations can be seen to have well defined limits for @xmath116 , so that we get @xmath120 and @xmath121 respectively . these expressions are closely related to those obtained from an adiabatic time dependent hartree - fock ( atdhf ) approach to this model @xcite .
in fact , the expression that gives the inertia function , eq.([spininercia ] ) , almost coincides with the one obtained from that approach ( @xmath122 instead of our @xmath123 ) while the expression associated with the potential , namely @xmath124 shows that the atdhf mean field approximation does not include the term @xmath125 that appears in our approximation .
this difference is due to the initial gcm ansatz used in our approach which deals with linear superpositions of slater determinants as the starting quasi - spin wavefunctions instead of pure determinants as it is the case with atdhf .
a similar discussion using the gcm has been also presented in ref .
@xcite .
figure 1 depicts the potential functions obtained from expressions ( [ exapot ] ) and ( [ spinpot ] ) , while figure 2 shows the functions given by ( [ exainer ] ) and ( [ spininercia ] ) , with the changed signal , for @xmath126 and @xmath127 and @xmath128 , respectively . from these figures
it can be verified that the curves tend to superpose as @xmath129 increases being that they are almost indistinguishable already for @xmath130 , i.e. , @xmath131 . with these results we can now check the validity of our approach and its limiting version when @xmath116 . to compare the two approaches we first obtain the exact results by diagonalizing the energy matrix @xmath132 , eq.([exata ] ) , and , on the other hand , by solving the schrdinger equation @xmath133 \psi \left ( \phi \right ) = e\psi \left ( \phi \right ) \label{schrod}\ ] ] using the expressions for @xmath134 , eq.([spinpot ] ) , and @xmath99 , obtained from eq.([spininercia ] ) , respectively .
from the outset it is clear that the second order moment truncation proposed here will not suffice to describe the quasi - spin system for small @xmath129 since in these cases higher order terms are also important .
thus the requirement of large @xmath129 imposed so that a reliable two - terms hamiltonian can be obtained also results in the approximate expressions , eqs.([spinpot ] ) and ( [ spininercia ] ) respectively . in this form , from a direct inspection of figures 1 and 2 , we know beforehand that our approach must be acceptable for @xmath135 .
this guides us to study the spectra of systems with those quasi - spin number directly using expressions ( [ spinpot ] ) and ( [ spininercia ] ) as our starting point .
therefore , the solutions thus obtained give an way of testing the quality of our approximate expressions for the potential and inertia functions .
a comparison between the approximate and exact energies can be seen in figures 3 and 4 .
a check of our description is shown in figure 3 where the exact ground state eigenvalues of the quasi - spin systems , with @xmath136 and @xmath137,@xmath138are compared with those obtained from eq.([schrod ] ) , using eqs.([spinpot ] ) and ( [ spininercia ] ) , as a function of the strength parameter @xmath139 . in figure 4
we show the energy errors in the ground state eigenvalues as a function of the quasi - spin number @xmath129 for a fixed value of the strength parameter , namely , @xmath140 .
the results are almost the same for other values of @xmath139 .
it is clear from figure 4 that an error as low as @xmath141 is achieved already for a system with @xmath142 @xmath143 thus confirming our previous assumption , and therefore establishing a suitable working limit for the number of quasi - spins of the system when we use the present approach .
this also means that the truncation of the series of moments of the energy surface up to the second term can , already for @xmath144 , account for the dominant part of the dynamical content of the quasi - spin system .
two interesting features of the lipkin model can be directly discussed in this approach .
first , it is immediate to see , observing the expression describing the potential function , eq.([spinpot ] ) , that the strength parameter , namely @xmath139 , assumes its critical value @xmath145 the minimum at @xmath146 turns into a maximum
when @xmath147 .
this shows that only when @xmath148 we get @xmath149 as expected @xcite . clearly for values @xmath150
@xmath151 a barrier appears at the origin and tunneling can occur under certain conditions @xcite .
second , the possibility of tunneling through the barrier , when it is the case , can be blocked if the effective mass presents a divergent behaviour at or near the classical turning points , i.e. , if @xmath152 exhibits zeroes there . in this form ,
not only the potential barrier governs the tunneling process , but the inertia function also plays an essential role in this matter . due to the behaviour of the inertia some regions in the angle domain may not be acessible ; this can also be seen as a localization of the spin orientation when this singular behaviour occurs : when it turns out to be impossible for the spin to change its orientation - when the effective mass diverges - it can not tunnel . considering @xmath153 for the lipkin model
we see that its zeroes occur at @xmath154 for @xmath155 , and that they are pushed to the borders of the angle interval as @xmath139 increases to @xmath156 , i.e. , @xmath157 for @xmath158 . for @xmath159
there will be no zeroes at all .
this result shows us that whenever a barrier is present in the potential function and
this occurs for @xmath150 there will always be a nonvanishing tunneling probability . as a consequence ,
there will always be tunneling in the lipkin model when a barrier is present .
if we choose now the parameters of the phenomenological hamiltonian such that @xmath160 , @xmath161 , and @xmath22 we have the already proposed mn12-acetate molecule hamiltonian model @xcite . starting from that hamiltonian , and using the general expression , eq.([superficiegeral ] ) , we end up with the corresponding energy surface @xmath162 @xmath163 @xmath164 we can now obtain the expressions for the potential and inertia functions , @xmath165 @xmath166 and @xmath167 @xmath168 \sqrt{1-\frac{n\left ( n+1\right ) } { j\left ( j+1\right ) } } + \ ] ] @xmath169\ ] ] respectively . since in this case @xmath43 , and
all the summations can be directly calculated , we are allowed by the estimates of the previous section to directly use here the already discussed approximations such that these expressions reduce respectively to @xmath170 and @xmath171 where @xmath172 ( to within 1.4% ) , @xmath173 @xmath174 ( to within 0.1% ) , and @xmath175 ( to within 0.47% ) .
figures 5 and 6 depict @xmath176 and @xmath177 for some values of @xmath178 where we have adopted the angle domain @xmath179 for the sake of clarity . in figure 5
we see that the potential for the mn12-acetate molecule presents two minima whose energy depths depend on the applied field . at the same time
the expressions for @xmath180 shown in figure 6 present a similar behaviour . as it is shown , as the paralel field increases , one of the minima gets shallower while the other deepens as expected from phenomenological considerations @xcite .
if we solve the schrdinger equation with the obtained expressions for the potential and inertia function , we obtain @xmath181 for the ground state energy when @xmath182 with @xmath183 @xmath41 .
this result shows that the calculated ground state energy is only @xmath184 below the exact energy result ( @xmath185 ) , obtained by diagonalizing the corresponding @xmath186 matrix .
when @xmath187 the errors are of the same order of magnitude as that .
these results confirm our previous considerations , and thus allow us to further discuss the mn12-acetate molecule properties using our approach .
however , instead of extracting results from numerical calculations based on the new hamiltonian , we will discuss some basic properties of this molecule by analysing the potential and inertia function analytic expressions . observing that we can consider @xmath188 , and that we can introduce the field
at which the magnetization of the mn12-acetate molecule attains saturation @xcite , i.e. , @xmath189 we can further rewrite expressions ( [ acepot ] ) and ( [ ine ] ) as @xmath190 @xmath191 it can be immediately seen that while @xmath192 , @xmath193 , thus showing the relative dominance of the potential energy term in the approximate hamiltonian .
now , let us analyse the expressions for the potential and inertia functions .
first of all let us extract the essential features of the potential function .
it is direct to see that since the treatment is of quantum nature , it embodies the inherent quantum correlations , and as such , even in the absence of paralel magnetic fields , the potential minima are @xmath194 , instead of the semiclassical value @xmath185 obtained with @xmath195 .
in fact , in our calculation , it is the ground state energy that occurs at @xmath196 , meaning that this is the energy height of the ground state barrier in this description , and furthermore that the ground state energy does not coincide with the bottom of the potential , due to the uncertainty principle , as it must be .
this also points to the fact that the quantum correction related to the uncertainty principle amounts to about @xmath197 of the potential barrier height . in the particular case of no paralel field ,
the minima occur at @xmath198 , while the maxima occur at @xmath199 .
the introduction of a paralel magnetic field , @xmath178 , only shifts the maxima , the minima being kept unaltered , viz .
, @xmath200 in what refers to the inertia function let us study first of all its expression by looking for the extrema .
let us consider the equation @xmath201 as before , when there is no magnetic field , @xmath182 , the maxima occur at @xmath202 respectively , while the minima occur at @xmath198 . by choosing now @xmath208 and @xmath209 , with @xmath24 , we have the proposed hamiltonian for the fe8 cluster @xcite in the absence of external magnetic fields .
thus , by taking into account that , again , @xmath43 , we can use all the approximations discussed before so that we obtain @xmath210 and @xmath211 respectively . in the same way as in the mn12-acetate
, we also have @xmath212 and @xmath213 thus also showing the relative dominance of the potential energy contribution to the approximate hamiltonian . in figure
( 7 ) we depict the potential and the effective mass functions for @xmath214 and @xmath215 . the numerical solutions of the schrdinger equation written with those expressions and constants also give results that agree quite well with the ones obtained from the diagonalization of the corresponding original model hamiltonian .
the errors are less than 1% for the energy values in the lower part of the spectrum . as in the mn12-acetate molecule without paralel magnetic fields ,
the minima of the potential occur at _ _
_ _ @xmath216 _ _ , _ _ while the maxima occur at _ _
_ _ @xmath217 .
as such , the energy height defined by @xmath218 is obviously only dependent on the internal parameters of the molecule .
it is now possible to evaluate the energy barrier height associated with the lowest energy level . if we consider the energy of the ground state ( but not the botton of the potential well ) then , @xmath219 using @xmath220 , obtained from the schrdinger equation , we get @xmath221 , which is only @xmath222 higher than the experimental result , @xmath223 , presented in @xcite . in addition , it is important to observe that the top of the potential barrier lies in the negative energy domain so that there appear some energy levels all those above the barrier that seem not to be associated with tunneling processes .
in fact there are some pairs of levels that are almost degenerate although this is not due to tunneling , as is also the case in the lipkin model when @xmath150 .
this effect is due to the particular form of the fe8 hamiltonian which embodies a subtle interplay between the @xmath224 term , which tends to produce degenerate doublets , and the @xmath225 term which tends to symmetrize the energy spectrum about zero . in what concerns the expression that gives the inertia function , eq.([inefe ] ) , it is immediately seen that it presents its minima precisely at the same angles as the potential function does .
furthermore , for the specific values of the fe8 cluster parameters , the mass function does not exhibit infinities so that tunneling is always possible , even in the absence of an external transverse magnetic field .
in fact , the tunneling probability rate is strongly dependent on the behavior of the effective mass function in the barrier region
in this contribution we have shown that a hamiltonian written in terms of an angle variable can be constructed that can describe spin systems in a consistent way .
the construction scheme starts from a hamiltonian given in terms of the operators obeying the angular momentum commutation relations and that is , in general , introduced from phenomenological considerations .
the quantum hamiltonian here proposed extends a previous description , that only took into account a potential energy term , by also introducing a kinetic energy term in which an effective mass associated with the spin system is present . although this hamiltonian has been obtained in a general form , the validity of the basic hypothesis supporting the assumed dominance of the first two terms of a moment expansion of the associated matrix energy distribution has been tested . using the soluble lipkin quasi - spin model as a testing ground
, we have then shown that a criterion for the obtained two - terms hamiltonian to be a good description of the system can be established . from this study
it is clear that for systems with @xmath226 the energy spectrum - then obtained from a schrdinger equation - is a very good approximation to the one obtained from the diagonalization of the starting model hamiltonian .
it is then important to verify that some recently produced molecules are physical systems of interest that have global angular momentum values in the range that allow us to use the present approach . in this perspective , in what concerns the energy spectra and barriers , we have shown that a consistent description were obtained for the mn12-acetate as well as for the fe8 cluster molecules .
being of quantum nature , this approach extends the semiclassical estimates and introduce essential corrections that would not be present . in what concerns the possibility of spin tunneling
, we have shown that in the present description the potential energy function clearly exhibits energy barriers whose heigths are in agreement with the experimental results . at the same time
we have emphasized the essential role played by the effective mass function in the discussion of tunneling in the sense that its values in the barrier region are a measure of how easy it is for the system to change its angular orientation .
the presence of infinities in the effective mass function is an unequivocal feature indicating that tunneling is then forbidden . in the case of the mn12-acetate molecule
the divergences always occur if only a paralel magnetic field is present , as it should be .
the expression for the effective mass then shows the way through which the final effective hamiltonian embodies that well - known basic quantum mechanical result .
the same is not true for the fe8 cluster when tunneling can occur , even in the absence of a transverse magnetic field .
the phenomenological hamiltonian reflects the symmetries of the molecule , and it is precisely its particular form , with its coefficients obtained from the experiments , which gives rise to an effective mass function that is nowhere divergent . in the present paper
we have carried out a qualitative discussion pointing to the main aspects of our approach that can be used to describe the spin tunneling process . in a forthcoming contribution
we intend to present quantitative estimates of the energy levels splitting associated with the spin tunneling .
the author is grateful to prof .
novak from universidade federal do rio de janeiro , a.f.r .
de toledo piza from universidade de so paulo , and b.m .
pimentel from instituto de fsica terica- unesp for valuable suggestions .
the author was partially supported by conselho nacional de desenvolvimento cientfico e tecnolgico , cnpq , brazil .
figure 1 : comparison between the large @xmath129 potential function ( continuous line ) and the full expression , eq.(8 ) , ( dashed line ) for the lipkin model for different values of @xmath129 .
the interaction strength is @xmath227 in all cases .
figure 2 : comparison between the large @xmath129 function @xmath228 ( continuous line ) and the full expression , eq.(9 ) , ( dashed line ) for the lipkin model for different values of @xmath129 .
the interaction strength is @xmath227 in all cases .
figure 3 : relative error in the ground state energies obtained from eq.([schrod ] ) for the lipkin model when compared with the exact results , as a function of the interaction strength @xmath139 .
the continuous curve is associated with @xmath229 , the dashed curve with @xmath230 , and the dot - dashed curve with @xmath231 .
the lines only guide the eyes .
figure 4 : relative error in the ground state energies obtained from eq.([schrod ] ) for the lipkin model , when compared with the exact results , as a function of the number of spins , @xmath129 @xmath232 , for @xmath233 . for systems with @xmath234 the deviation lies below @xmath141 .
the line only guides the eyes . | starting from a phenomenological hamiltonian originally written in terms of angular momentum operators we derive a new quantum angle - based hamiltonian that allows for a discussion on the quantum spin tunneling .
the study of the applicability of the present approach , carried out in calculations with a soluble quasi - spin model , shows that we are allowed to use our method in the description of physical systems such as the mn12-acetate molecule , as well as the octanuclear iron cluster , fe8 , in a reliable way . with the present description the interpretation of the spin tunneling
is seen to be direct , the spectra and energy barriers of those systems are obtained , and it is shown that they agree with the experimental ones .
pacs : 75.45+j , 75.60jp + keywords : spin tunneling , lipkin model , mn12-acetate , fe8 cluster corresponding author d. galetti instituto de fsica terica - unesp phone : ( 011 ) 3177 9001 fax : ( 011 ) 3177 9080 e - mail : [email protected] |
the central objects of study in this paper are the block structure spaces of certain topological manifolds .
these spaces arise as follows . for a closed topological manifold @xmath0 , a
_ simple manifold structure _ on @xmath0 is a simple homotopy equivalence @xmath6 from some closed manifold @xmath7 to @xmath0 .
two simple manifold structures @xmath8 and @xmath9 are _ equivalent _ if there exists a homeomorphism @xmath10 and a homotopy from @xmath11 to @xmath12 .
the set of equivalence classes is called the _ simple structure set _ of @xmath0 and is denoted by @xmath13 .
more generally , if @xmath0 is a compact manifold with boundary , there is a simple structure set @xmath14 whose elements are represented by simple homotopy equivalences of pairs @xmath15 where the restricted map @xmath16 is a homeomorphism . for a given compact manifold @xmath0 ,
quinn constructed in @xcite a space @xmath17 , the _ block structure space _ of @xmath0 , such that @xmath18 this space is related to the automorphism spaces of @xmath0 via the homotopy fibration sequence @xmath19 here the space @xmath20 is the block version of the space @xmath21 of self homeo - morphisms of @xmath0 and @xmath22 is the space of simple self homotopy equivalences of @xmath0 .
see @xcite , @xcite .
the simple structure sets @xmath23 have been studied in many special cases . in particular
they were calculated for real projective spaces , lens spaces and complex projective spaces by work of many authors , especially wall @xcite , browder @xcite , browder - livesay @xcite and lpez de medrano @xcite .
a useful tool used by these authors is the _
join construction _ ( see @xcite ) .
it provides us with maps between simple structure sets in different dimensions .
for example in the case of real projective spaces we get maps @xmath24 .
we generalize the join construction to define maps between the block structure spaces . in principle
the homotopy type of the block structure space of any manifold is well understood as the homotopy fiber in the sullivan - wall - quinn - ranicki homotopy fibration sequence of surgery ( see ( * ? ? ?
* theorem 2.3.1 . ) ) .
unfortunately , if @xmath25 this description of the block structure space does not easily extend to an illuminating description of the maps @xmath26 induced by the join construction ( ranicki showed me how to handle the cases @xmath27 ) .
we propose to remedy this defect by using ideas from functor calculus . for this purpose
we formulate the join construction as a certain continuous functor . in the case of real projective spaces
it is a functor , which we denote @xmath28 , from the category @xmath29 of real finite dimensional vector spaces with inner product to the category @xmath30 of pointed spaces which sends the vector space @xmath4 to the block structure space of @xmath31 , the real projective space of @xmath4 .
there are variants of this for the lens spaces and complex projective spaces .
the general methods we use apply to all cases .
however , in the paper we specialize to the case of real projective spaces .
we make remarks on the other cases at the end of the introduction .
the advantage of the formulation of the join construction as a continuous functor is that it enables us to use the orthogonal calculus of functors to study the block structure spaces of real projective spaces .
the orthogonal calculus provides us with a general theory for studying continuous functors from @xmath29 ( as above ) to @xmath30 .
it was developed by weiss in @xcite .
he associates to a given continuous functor @xmath32 the tower of functors @xmath33 called the _
taylor tower_. for each @xmath2 the functor @xmath34 is a _ polynomial functor of degree _
@xmath35 in a certain sense and it is also the best approximation of @xmath32 by such a functor
. therefore it should be thought of as the @xmath2-th taylor polynomial of @xmath32 .
for example , the functor @xmath36 is defined by @xmath37 .
it is the best approximation of @xmath32 by an essentially constant functor ( all induced maps @xmath38 are homotopy equivalences ) .
thus the expansion is at _ infinity_. the functor @xmath32 is related to its taylor tower via natural transformations @xmath39 which fit together to form a commutative diagram . in favorable circumstances the maps @xmath40
are highly connected for @xmath2 large .
the idea of the calculus of functors is to obtain some information about the functor @xmath32 by investigating its taylor tower . for this purpose
we need to estimate the connectivity of the maps @xmath41 .
the best situation happens if the functor @xmath32 is polynomial of some degree @xmath35 .
then the maps @xmath41 are homotopy equivalences for all @xmath42 .
a continuous functor @xmath32 is called polynomial of degree @xmath35 if the canonical map @xmath43{cc } \mathrm{holim } \\ [ -1 mm ] \scriptstyle{0 \neq u \subseteq { \mathbb{r}}^{k+1 } } \end{array } } e({v \ ! \oplus \
! u})\ ] ] is a homotopy equivalence for all @xmath42 .
the homotopy limit is over the topological poset of all non - zero vector subspaces of @xmath44 ( in section [ chap : oc ] we give some details about homotopy limits in this setting ) .
there is a way how to verify that a given functor is polynomial of degree @xmath35 ; in the case @xmath45 we discuss it in detail in the paper .
once the convergence question is established ( that means the connectivity of maps @xmath40 is known ) , the goal is to determine the taylor tower itself .
this can be done via an important theorem in @xcite which says that the difference , @xmath46 , between the stages of the taylor tower of the functor @xmath32 is determined by a certain spectrum @xmath47 with an action of the orthogonal group @xmath48 .
this spectrum should therefore be thought of as the @xmath2-th derivative of @xmath32 at _
infinity_. the main theorem of this paper says that the functor @xmath28 defined above is essentially polynomial of degree @xmath5 .
it can be formulated as follows
. * theorem [ thm1 ] . * _ let @xmath49 be the functor defined by @xmath50
. then the canonical maps @xmath51{cc } \mathrm{holim } \\ [ -1 mm ] \scriptstyle{0 \neq u \subseteq { \mathbb{r}}^2 } \end{array } } f({v \ ! \oplus \
! u})\ ] ] are homotopy equivalences if @xmath52 .
_ this implies that the maps @xmath53 are homotopy equivalences if @xmath54 ( see corollary [ cor1 ] ) and therefore in high enough dimensions the taylor tower of @xmath28 has only two interesting stages , namely @xmath55 and @xmath56 . as mentioned above , the difference between them is measured by the first derivative of @xmath28 at infinity , which is a certain spectrum @xmath57 with an action of the orthogonal group @xmath58 .
the homotopy groups of the spectrum @xmath57 are also determined . * theorem [ thm2 ] . *
_ we have @xmath59 where the groups @xmath60 are the @xmath61-periodic @xmath62-groups from surgery theory associated to the trivial group . _
there is another spectrum @xmath63 whose homotopy groups are @xmath64 for @xmath65 ( see @xcite ) .
the relationship between @xmath57 and @xmath63 is not yet clear to us , because we do not know any map between the two spectra .
the results of theorems [ thm1 ] and [ thm2 ] can be used to provide a functorial description of spaces @xmath66 via the taylor tower of @xmath28 , if @xmath54 .
for example by comparing the taylor tower of @xmath28 evaluated at @xmath67 and at @xmath4 we obtain the homotopy fibration sequence : @xmath68 \rightarrow t_1f(0 ) \rightarrow t_1f(v),\ ] ] where @xmath69 is the unit sphere in @xmath4 with the antipodal involution , the subscript @xmath70 denotes added base point and the subscript @xmath71 denotes the homotopy orbit spectrum .
we expect to determine the homotopy type of @xmath72 and the first map of the homotopy fibration sequence ( [ alt - des ] ) jointly with weiss in @xcite .
in any case , this yields a surprising similarity of the homotopy fibration sequence ( [ alt - des ] ) with the sullivan - wall - quinn - ranicki homotopy fibration sequence of surgery theory for @xmath31 ( see ( * ? ? ?
* theorem 2.3.1 ) ) .
however , an important difference between the two homotopy fibration sequences is that ( [ alt - des ] ) is functorial in @xmath4 whereas the surgery theoretic description is not .
the taylor tower of @xmath28 can also be used to give a satisfying description of the block structure spaces of the infinite dimensional real projective space , that means the colimit @xmath73{cc } \mathrm{colim } \\ [ -1 mm ] \scriptstyle{n \in { \mathbb{n } } } \end{array } } f({\mathbb{r}}^n ) \simeq t_0f(0).\ ] ] namely , the taylor tower of @xmath28 evaluated at @xmath67 becomes the homotopy fibration sequence : @xmath74 \rightarrow t_1f(0 ) \rightarrow t_0f(0 ) \simeq f({\mathbb{r}}^\infty).\ ] ] if the promised results of @xcite are obtained , this yields the desired description of @xmath75 . the ultimate goal in the study of the block structure space of a manifold @xmath0 is to provide some information about the automorphism spaces of @xmath0 , the space of self homeomorphisms @xmath21 and the space of simple self homotopy equivalences @xmath22 .
the relation between these spaces is via the homotopy fibration sequence ( [ two ] ) , which involves the block version of the space of self homeomorphisms @xmath20 .
if @xmath76 is any of the automorphism spaces above , the join construction provides us with maps @xmath77 for @xmath78 and the automorphism space @xmath79 of the infinite dimensional real projective space is defined as the colimit of @xmath80 over @xmath81 .
the join maps commute with the maps in ( [ two ] ) and so we obtain the homotopy fibration sequence @xmath82 closely related to these spaces are also certain equivariant automorphism spaces discussed in a more detail below .
we remark that the space @xmath83 defined below is a trivial double cover of @xmath84 and in @xcite it is shown that this space is homotopy equivalent to @xmath85 , where @xmath86 and the subscript @xmath70 denotes the added base point .
therefore any information about the homotopy type of @xmath75 gives us also information about @xmath87 which is closely related to @xmath88 ( see @xcite for more information about the relation between various automorphism spaces , also note that for real projective spaces the word simple can be dropped as @xmath89 ) .
there is also a more direct way to relate the space @xmath75 to certain honest ( not block ) equivariant automorphism spaces as follows .
let @xmath90 be the regular representation of the group @xmath91 .
denote by @xmath92 the topological monoid of equivariant self homotopy equivalences of the unit sphere @xmath93 of the induced representation @xmath94 and by @xmath95 the topological group of equivariant self homeomorphisms of @xmath96 .
the join map induces inclusions @xmath97 and @xmath98 for @xmath78 and there are also inclusions @xmath99 which commute with the join map
. therefore we get the maps between quotients induced by the join construction . after passing to the colimit over @xmath100
there is the following splitting @xmath101{cc } \mathrm{colim } \\ [ -1 mm ] \scriptstyle{n \in { \mathbb{n } } } \end{array } } { \mathrm{g}}_{{\mathbb{z}}_2}(s(n \cdot w))/{\mathrm{top}}_{{\mathbb{z}}_2}(s(n \cdot w ) ) \simeq f({\mathbb{r}}^\infty)_1 \times { \mathrm{g}}/{\mathrm{top}},\ ] ] which relates the left hand side to the space @xmath75 . here
@xmath102 is the base point component . for the classical space @xmath103 see e.g. @xcite
here is a sketch of how the splitting ( [ equiv - spaces ] ) is obtained .
let @xmath104 where @xmath4 is the free part of the representation and @xmath105 is the trivial part .
there is the splitting @xmath106{cc } \mathrm{colim } \\ [ -1 mm ] \scriptstyle{n \in { \mathbb{n } } } \end{array } } { \mathrm{g}}_{{\mathbb{z}}_2}(s(n \cdot w ) ) \simeq { \begin{array}[t]{cc } \mathrm{colim } \\ [ -1 mm ] \scriptstyle{n \in { \mathbb{n } } } \end{array } } { \mathrm{g}}_{{\mathbb{z}}_2}(s(n \cdot v ) ) \times { \begin{array}[t]{cc } \mathrm{colim } \\ [ -1 mm ] \scriptstyle{n \in { \mathbb{n } } } \end{array } } { \mathrm{g}}(s(n \cdot u)).\ ] ] see ( * ? ? ?
* proposition 3.5 . ) for the statement on the level of homotopy groups .
similarly there is the splitting @xmath107{cc } \mathrm{colim } \\ [ -1 mm ] \scriptstyle{n \in { \mathbb{n } } } \end{array } } { \mathrm{top}}_{{\mathbb{z}}_2}(s(n \cdot w ) ) \simeq { \begin{array}[t]{cc } \mathrm{colim } \\ [ -1 mm ] \scriptstyle{n \in { \mathbb{n } } } \end{array } } \widetilde{{\mathrm{top}}}_{{\mathbb{z}}_2}(s(n \cdot v ) ) \times { \begin{array}[t]{cc } \mathrm{colim } \\ [ -1 mm ] \scriptstyle{n \in { \mathbb{n } } } \end{array } } { \mathrm{top}}(s(n \cdot u)).\ ] ] this is obtained by passing to the colimit over @xmath108 with the splittings @xmath109 where @xmath110 means that the equivariant self homeomorphisms of @xmath96 are fixed to be the identity on @xmath111 .
the splitting ( [ top - splitting ] ) is analogous to the splitting in ( * ? ? ?
* lemma 3.4 . ) . the second factor in ( [ top - stable - splitting ] ) is the colimit of the second factors of ( [ top - splitting ] ) .
the first factor of ( [ top - stable - splitting ] ) is obtained by relating the first factor of ( [ top - splitting ] ) to the equivariant bounded self homeomorphisms of @xmath112 using a trick of @xcite and then relating the bounded self homeomorphisms to block self homeomorphisms as in ( * ? ? ?
* proposition 1.4.2 . ) . finally , these equivariant automorphism spaces of @xmath113 map into automorphism spaces of @xmath114 by passing to quotients , and this map is a trivial double cover .
the splitting ( [ equiv - spaces ] ) then follows from ( [ two - stable ] ) because the join maps commute with the splittings ( [ htpy - splitting ] ) , ( [ top - stable - splitting ] ) and we have @xmath115 now we would like to briefly discuss the study of the block structure spaces of lens spaces and complex projective spaces . using the join construction we obtain in these cases a functor from the category of complex vector spaces with inner product to @xmath30 which sends the vector space @xmath4 to @xmath116 or @xmath117 .
here we can employ the unitary calculus which is the complex analogue of orthogonal calculus , and we can try to run the study parallel to the case of real projective spaces .
the case of lens spaces was studied by madsen and rothenberg in @xcite , where they show that the join maps induce isomorphisms on homotopy groups ( they also determine the homotopy groups , and later madsen in @xcite determines the homotopy type localized at @xmath118 and away from @xmath118 ) .
hence , from our point of view , the corresponding functor is polynomial of degree @xmath67 .
the case of complex projective spaces is not covered in @xcite , so the parallel study would be interesting .
we note that the main ingredient in the proof of theorems [ thm1 ] and [ thm2 ] is the codimension @xmath119 surgery theory , whereas in the case of complex projective spaces we have to use codimension @xmath118 surgery theory , which is considerably more complicated ( see @xcite ) . at the moment we are unable to carry over the whole proofs of theorems [ thm1 ] , [ thm2 ] .
we hope to do this in the future .
the paper is organized as follows . in section [ blockspace ]
we introduce the block structure spaces . in section [ chap : suspension ]
we construct the functor @xmath28 . in section [ chap : oc ]
we give an overview of the basic constructions from orthogonal calculus that we use in the paper .
in particular for a continuous functor @xmath120 we recall a necessary and sufficient condition for @xmath32 to be polynomial of degree @xmath5 ( theorem [ poly1 ] ) .
this involves studying a certain functor @xmath121 with some additional structure , which is called the first derivative functor of @xmath32 .
section [ chap : f1 ] contains background material about codimension @xmath119 surgery theory , which is used to give a description of the first derivative functor @xmath122 . up to now
we claim no originality
. the description of @xmath122 will then be the main tool in the proofs of theorems [ thm1 ] and [ thm2 ] which are given in section [ chap : thetaf1 ] .
the proofs are our contribution , and they are the core of the paper .
we note that section [ chap : oc ] is a short survey of orthogonal calculus , however , for complete information we refer the reader to the original source @xcite . throughout the paper
we also assume basic knowledge of surgery theory ( that means the book @xcite up to chapter 10 ) , but we provide some background about the less standard parts , such as codimension 1 surgery theory . for @xmath42 by @xmath31
we denote the real projective space of @xmath4 , by @xmath123 we denote the sphere with radius @xmath124 in @xmath4 ; and @xmath125 .
in this section we introduce the block structure space of a closed manifold @xmath0 in general and we also introduce a method for constructing other models for the block structure spaces .
the method will be used in section [ chap : suspension ] to obtain a functorial model of the block structure spaces of the real projective spaces . in the introduction
we have defined for a given closed manifold @xmath0 its simple structure set @xmath23 and more generally for a compact manifold with boundary @xmath126 the simple structure set @xmath127 .
these are the basic objects of study of surgery theory which have been calculated in many special cases ( see for example @xcite , @xcite ) . for a compact manifold @xmath0 the structure sets @xmath128 become groups when @xmath129 .
the block structure space of @xmath0 is designed to be a space such that @xmath130 it is constructed as a @xmath131-set ( see @xcite ) .
if @xmath0 is a closed manifold then a @xmath2-simplex of @xmath17 is a simple homotopy equivalence @xmath132 between @xmath133-dimensional manifold @xmath134-ads , where @xmath135 ( the notion of a manifold @xmath134-ad is discussed below ) .
the face operators are given by restrictions and the base point in dimension @xmath2 is @xmath136 .
the construction of the space @xmath17 was given by quinn in @xcite .
as indicated in the introduction the main motivation for studying this space comes from the fact that it is closely related to the automorphism spaces of @xmath0 .
we refer the reader to the survey article @xcite for more information about this application . in (
* theorem 2.3.1 . )
it is also stated that @xmath17 fits into the homotopy fibration sequence of surgery theory of sullivan - wall - quinn - ranicki , which in principle determines its homotopy type .
as already mentioned , when @xmath0 is the real projective space @xmath31 for some @xmath42 , we obtain an additional structure of maps between the block structure spaces of real projective spaces in different dimensions .
these maps , called the _ join maps _ , were described on the homotopy groups by wall @xcite for @xmath137 and by madsen and rothenberg @xcite for @xmath138 , when @xmath139 .
we generalize these maps to the level of spaces in a functorial way . for this
we need another model of the block structure space of @xmath31 , that means a space homotopy equivalent to @xmath140 .
the method for constructing such a space , which we describe in this section , is based on a careful analysis of quinn s model .
the desired model is then constructed in two steps . in this section
we construct an intermediate model @xmath141 , which also admits the join maps , as explained in @xcite . however
, these do not define a functor because they do not behave well w.r.t . the composition .
then , in the next section , we construct the functorial model @xmath142 .
[ models ] each of the models @xmath140 , @xmath143 , @xmath144 has convenient properties in different situations . as they are all homotopy equivalent ( at least in high dimensions ) and as many of the results we need are only ` up to homotopy ' we switch between these models throughout the paper according to our needs .
firstly we elaborate on the definition of a manifold @xmath134-ad .
it is a generalization of a cobordism between manifolds , namely a cobordism is a special manifold @xmath145-ad .
according to @xcite ( or @xcite ) a cobordism between two @xmath108-dimensional manifolds @xmath146 , @xmath147 is a @xmath148-tuple @xmath149 , where @xmath7 is an @xmath150-dimensional manifold with two disjoint pieces of boundary @xmath151 , @xmath152 and @xmath153 for @xmath154 are homeomorphisms of @xmath108-dimensional manifolds .
a manifold @xmath134-ad is a generalization where the manifold @xmath7 has @xmath155 pieces of boundary which fit together in a nice way .
we formalize this notion below .
more precisely we give an abstract construction of a @xmath131-set which consists of objects formally resembling manifold @xmath134-ads as an associated @xmath131-set to a certain @xmath131-groupoid .
this is done in construction [ abstract ] and then applied to the special case of the block structure spaces in construction [ blockdef ] .
we recall basic definitions for @xmath131-sets of @xcite ( see also @xcite ) .
a @xmath131-set @xmath156 is a functor @xmath157 , where @xmath131 is the category with objects finite totally ordered sets @xmath158 and morphisms injective monotone functions @xmath159 for @xmath160 .
thus a @xmath131-set @xmath156 can be described as a collection of sets @xmath161 together with face maps @xmath162 .
the standard @xmath2-simplex is a @xmath131-set in the following way : the set of @xmath163-simplices @xmath164 is the set of monotone injective functions @xmath165 and for @xmath166 the face map @xmath167 is defined by @xmath168 .
also @xmath159 defines a map of @xmath131-sets @xmath169 . more generally , a @xmath131-groupoid @xmath156 is a functor @xmath170 , that means for each @xmath2 the set of @xmath2-simplices @xmath171 is a groupoid ( a category where all morphisms are isomorphisms ) and the face maps @xmath172 are functors .
note that the standard simplex @xmath173 is also a @xmath131-groupoid in a trivial way .
a @xmath131-set @xmath156 is _ pointed _ if for every @xmath174 there is a base simplex @xmath175 such that @xmath176 for all @xmath177 .
now we recall the following well known construction for the special case of a @xmath131-groupoid .
@xcite let @xmath156 be a @xmath131-groupoid .
the _ grothendieck construction _
@xmath178 of @xmath156 is the category with objects pairs @xmath179 , where @xmath180 , @xmath181 , and morphisms @xmath182 are pairs @xmath183 , where @xmath184 is a morphism in @xmath131 and @xmath185 is a morphism in @xmath186 .
we also define the degree of the object @xmath179 to be @xmath174 .
note that the construction is functorial .
in particular for @xmath187 we have a functor @xmath188 .
in this section we construct a continuous functor @xmath253 , such that for an object @xmath42 with @xmath254 , the space @xmath255 is homotopy equivalent to the space @xmath140 and for a morphism @xmath256 the map @xmath257 is a generalization of the join construction of wall .
[ f - functor ] given an object @xmath42 , let @xmath144 be the associated pointed @xmath131-set of the pointed @xmath131-groupoid whose @xmath2-simplex is a pair @xmath237 where : * @xmath238 is an involution on @xmath258 , which , as a map , is a block map , it leaves @xmath259 invariant for each @xmath124 , its restriction to @xmath259 is free for @xmath260 , and its restriction to @xmath261 is trivial , * @xmath262 is a block map , which is equivariant w.r.t .
@xmath238 on the source and @xmath241 on the target , it leaves @xmath259 invariant for each @xmath124 , the restriction of @xmath6 to @xmath263 is an equivariant homotopy equivalence for @xmath260 , and the restriction of @xmath6 to @xmath264 is a homeomorphism .
face maps are defined by restriction , an isomorphism from @xmath237 to @xmath242 is an equivariant block homeomorphism @xmath265 such that @xmath222 , and the base point is the pair @xmath244 .
a pair @xmath237 which represents a @xmath2-simplex in @xmath144 will be shortly referred to as a _ structure _ on @xmath258 . given a morphism @xmath266 in @xmath29 ,
let @xmath267 be a map defined as follows .
the linear inclusion @xmath268 induces a direct sum decomposition @xmath269 and the product decomposition @xmath270 .
first we obtain the structure @xmath271 on @xmath272 .
the desired structure on @xmath273 is defined by @xmath274 given an object @xmath42 , let @xmath255 be the geometric realization of @xmath144 , and given a morphism @xmath256 , let @xmath257 be the geometric realization of @xmath275 .
then @xmath28 defines a functor from @xmath29 to @xmath30 .
the statement is easy to verify and we leave it for the reader . in order to show that the space @xmath144 defines another model for the block structure space of @xmath31 we use the method of the previous section to prove the following proposition .
[ spheres - vs - vector - spaces ] for each @xmath42 there exist maps @xmath276 which are homotopy inverses of each other . as already indicated , throughout the paper we also work with the models @xmath140 and @xmath143 , which is legal for the statements which are ` up to homotopy ' .
similarly , we work with the join maps on these models .
for example in section [ chap : thetaf1 ] the model for the join map is @xmath277 .
the maps @xmath278 and @xmath279 are induced by the maps between the corresponding pointed @xmath131-groupoids .
since we think of @xmath280 as of a subspace of @xmath281 , we can define @xmath282 by @xmath283 . the map @xmath284 is defined by a certain ` iterated cone construction ' which we now describe .
let @xmath237 be a structure on @xmath245 and let @xmath285 , which is a structure on @xmath258 .
we describe a general procedure for constructing a block self - map of @xmath258 , which preserves @xmath259 for each @xmath124 , from a block self - map of @xmath245 .
the maps @xmath286 , @xmath287 are constructed from @xmath238 , @xmath6 using this procedure . * the case @xmath288 . *
let @xmath0 be a topological space .
the _ open cone _ on @xmath0 is the space @xmath289 whose points are linear expressions @xmath290 , where @xmath291 .
this includes the identifications @xmath292 .
the point @xmath67 is called the _ cone point_. we have a decomposition @xmath293 , with @xmath294\}$ ] , and @xmath295 .
for @xmath124 we distinguish the subspaces @xmath296 and we identify @xmath297 .
sometimes we want to give the cone point some other name than @xmath67 , e.g. @xmath298 . in that case
the points of the subspace @xmath299 are linear combinations @xmath300 for @xmath301 and @xmath302 $ ] .
note that , if @xmath0 is @xmath69 , then @xmath303 can be identified with @xmath4 , so that @xmath304 is identified with @xmath123 for @xmath124 .
let @xmath305 be a map .
then the rule @xmath306 defines an extension @xmath307 of @xmath251 such that @xmath308 . * the cases @xmath309 .
* let @xmath310 be a block map .
we define its extension , a block map @xmath311 , such that @xmath312 .
we have the decomposition @xmath313 . on @xmath314
the extension is defined as in the case @xmath288 , i.e. if @xmath315 , then @xmath316 .
on @xmath317 we use the following ` iterated cone construction ' .
starting with @xmath1 we define the _ iterated cone _ @xmath318 by @xmath319 where @xmath320 , @xmath321 .
@xmath322 figure [ fig : c-1 m ] depicts the special case when @xmath323 and @xmath45 .
( here mutually distinct cone points should be used for different cones in sight . ) the spaces @xmath317 and @xmath318 can be identified by choosing suitable cone points in the above description .
namely , if @xmath324 , then choose the cone point of @xmath325 to be the barycenter @xmath326 .
explicitly , we obtain a homeomorphism @xmath327 as follows .
suppose that for all @xmath328 a homeomorphism @xmath329 was constructed .
then define a homeomorphism from @xmath330 to @xmath331 by @xmath332 by induction over @xmath324 we obtain the desired identification .
again , figure [ fig : c-1 m ] in the special case when @xmath323 and @xmath45 might be helpful . with this identification ,
starting with a block map @xmath333 we construct a block map @xmath334 by iterating the construction of the cone map as described in the case @xmath288 .
the resulting map clearly has the property @xmath335 ; and it is our definition of the map @xmath336 on @xmath337 .
it is easy to verify that the maps @xmath286 , @xmath287 defined via this procedure define a structure on @xmath258 as required .
the composition @xmath338 is the identity .
it remains to show that the composition @xmath339 is homotopic to the identity .
we start with a structure @xmath237 on @xmath258 , consider its restriction , the structure @xmath340 on @xmath341 and the image @xmath285 , a structure on @xmath281 .
we need to find compatible homotopies between @xmath238 and @xmath286 and @xmath6 and @xmath287 satisfying certain restrictions .
similarly as in the previous construction we give a general procedure as follows .
suppose @xmath342 is a block map such that @xmath343 .
consider the restriction block map @xmath344 and the block map @xmath345 obtained via the iterated cone construction .
we give a construction of a homotopy , say @xmath346 , between @xmath251 and @xmath347 with the property that each @xmath346 is a block map @xmath348 such that @xmath349 . applying this homotopy to the maps
@xmath238 , @xmath6 in place of @xmath251 produces the desired homotopies between @xmath238 , @xmath6 and @xmath286 , @xmath287 . *
the case @xmath288 . *
the required homotopy is obtained using the alexander trick .
we can think of the map @xmath251 as of a perturbation of @xmath350 and refer to the part of @xmath303 on which the two maps @xmath251 and @xmath350 do not agree as the domain of the perturbation .
further we consider @xmath351 and we give a different description of the homotopy on the two pieces . in words ,
the trick consists of shrinking the domain of the perturbation in @xmath299 and of pushing the domain of the perturbation towards @xmath352 in @xmath353 .
formally , the homotopy @xmath346 on @xmath299 is parametrized by @xmath354 $ ] and given by @xmath355 the homotopy @xmath346 on @xmath353 is parametrized by @xmath356 $ ] and given by @xmath357 if @xmath358 and by @xmath359 .
after reparametrizing , using a suitable homeomorphism @xmath356 \cong [ 0,1]$ ] which maps @xmath119 to @xmath119 and @xmath352 to @xmath67 , we obtain that @xmath360 and @xmath361 .
moreover , clearly @xmath362 for all @xmath363 $ ] . * the cases @xmath309 . *
again , the homotopy is described on the two pieces of the decomposition @xmath313 differently .
on @xmath364 it is given by pushing the domain of the perturbation towards @xmath352 just as in the case @xmath288 ; the formula is left for the reader . to describe the homotopy on @xmath317 , recall that in the definition of @xmath365 we iterated the cone construction .
further , in the case @xmath288 the homotopy between @xmath251 and @xmath366 was given by the alexander trick .
in analogy , in the case @xmath309 the homotopy between @xmath251 and @xmath365 is given by iterating the alexander trick .
to start , recall the identification of @xmath318 with @xmath317 and apply the alexander trick homotopy to @xmath251 considered as a block map from @xmath318 to @xmath367 .
the resulting map , say @xmath368 , coincides with @xmath369 on @xmath370 , but is possibly different on the complement , the union of @xmath371 where @xmath372 , @xmath321 ( see figure [ fig : c-1 m ] ) .
note however , that if we consider @xmath371 for each such @xmath373 as the quotient of @xmath374 $ ] , then the restriction of @xmath368 is the quotient map of the product map of the restriction of @xmath251 to @xmath375 with the identity on @xmath376 $ ] ( again see figure [ fig : c-1 m ] ) .
hence , we can perform a homotopy given as the product of the alexander trick homotopy on @xmath375 and the identity on @xmath376 $ ] .
the resulting map now agrees with @xmath365 on a larger portion of @xmath317 .
we proceed inductively until we take care of all the faces of @xmath173
. then the resulting map agrees with @xmath365 on the whole @xmath317 .
again , it is clear that the homotopy @xmath346 is such that for all @xmath377 we have @xmath378 for all @xmath124 .
before we discuss continuity of the functor @xmath28 recall that , since the maps @xmath238 , @xmath6 can vary continuously , the set of @xmath2-simplices of the @xmath131-groupoid @xmath144 ( similarly @xmath143 ) is endowed with a topology , namely the subspace topology of the product topology of the compact - open topology on the space of self - maps on @xmath258 .
so we have a topological @xmath131-groupoid , and the associated @xmath131-set @xmath144 becomes a @xmath131-space .
the geometric realization of @xmath144 as a @xmath131-space is homotopy equivalent to the geometric realization of the same object as a @xmath131-set . to see this we resolve the @xmath131-space version of @xmath144 by taking the singular simplicial set of the space of @xmath2-simplices for each @xmath174 .
we obtain a combined simplicial-@xmath131-set , which has its geometric realization homotopy equivalent to the geometric realization of the @xmath131-space . we can ` change the order ' of realization and realize first spaces @xmath379 for each @xmath174 as @xmath131-sets and then realize the resulting simplicial set .
the lemma will be proved if we show that all degeneracy and face maps between @xmath379 and @xmath380 induce homotopy equivalences .
this can be verified using the kan property of @xmath144 .
now we show that the functor @xmath28 is a continuous functor in the sense of definition [ continuous ] .
first , note that one can define a continuous functor from @xmath29 to topological @xmath131-groupoids or to @xmath131-spaces to be a functor for which for all @xmath2 the evaluation maps @xmath381 are continuous .
given such a continuous functor from @xmath382 , composing with the functor @xmath383 and taking the geometric realization of @xmath384 we obtain a continuous functor to @xmath30 .
[ f - cont ] the functor @xmath253 is a continuous functor . in view of the previous remarks it is enough to show that the functor @xmath385 is continuous .
that means we need to show that the evaluation map @xmath386 is continuous for each @xmath2 . if @xmath387 , then @xmath388 is given by @xmath389 which clearly gives continuity . for the case when @xmath390 we work locally and reduce to the above case as follows
choose @xmath391 .
we prove that the evaluation map is continuous on some neighborhood @xmath392 of @xmath268 .
there is a map @xmath393 given by @xmath394 , which is a bundle map . therefore there is a neighborhood @xmath395 of @xmath268 over which we have a section @xmath396 of this bundle .
then the evaluation map @xmath388 can be factored as @xmath397 the first map is continuous because by the assumption both of its factors are continuous and the second one is the previous case .
our method of study of the functor @xmath28 defined in the previous section is the orthogonal calculus of functors . in this section
we review some of the terms and tools of orthogonal calculus which we use in this paper .
it is intended as a short survey which describes the context in which we work . for the full account
we refer the reader to the paper of weiss @xcite in which orthogonal calculus was developed .
we recall some notation . by @xmath29
is denoted the category of finite - dimensional real vector spaces with ( positive definite ) inner product .
morphisms are linear maps preserving inner product , in particular they are linear inclusions
. the objects of @xmath29 are usually denoted by letters @xmath398 .
morphism sets @xmath399 are stiefel manifolds and so have a topology . by @xmath400 will be denoted the inclusion of @xmath4 into @xmath401 for any objects @xmath4 , @xmath105 in @xmath29 . in special cases , by @xmath402 will be denoted the inclusion of @xmath403 into @xmath404 where @xmath405 is mapped isomorphically onto @xmath406 in @xmath407 and by @xmath408 the inclusion where @xmath405 is mapped onto @xmath409 in @xmath407 . for @xmath410 , @xmath411 denotes the one - point compactification of @xmath105 .
furthermore @xmath412 and @xmath413 .
we also write @xmath414 for @xmath415 . the orthogonal calculus works for functors @xmath416 satisfying the following axiom .
@xcite [ continuous ] a functor @xmath120 is called _ continuous _ if for each @xmath417 the evaluation map @xmath418 is continuous .
we have shown in theorem [ f - cont ] that the functor @xmath28 satisfies this axiom .
there are many other functors of this type .
for example , the functor @xmath419 where @xmath420 is the classifying space of the orthogonal group @xmath421 has been studied in @xcite .
other examples come up in the study of automorphisms of manifolds , see papers @xcite , @xcite .
we have already presented the basic philosophy of orthogonal calculus in the introduction . recall that a continuous functor @xmath422 is studied by means of its taylor tower which is a tower of functors @xmath423 where the functor @xmath34 is the @xmath2-th polynomial approximation of @xmath32 .
the tower should be thought of as the expansion at infinity .
there are two aspects of the taylor tower which one needs to understand .
firstly , one needs to find out whether the taylor tower converges to @xmath32 , that means to estimate the connectivity of the canonical maps @xmath40 .
secondly , one needs to determine the differences , @xmath424 , between the stages of the tower .
both issues can be investigated using the derivatives in orthogonal calculus .
these arise as follows .
given a continuous functor @xmath422 there is defined for each @xmath425 its @xmath2-th _ derivative functor _ @xmath426 which is again a continuous functor from @xmath29 to @xmath30 with the additional structure of a coordinate free spectrum of multiplicity @xmath2 .
that means there are maps @xmath427 on objects @xmath428 is defined inductively as @xmath429 where @xmath430 is the adjoint of @xmath431 . for
the definition on morphisms , take @xmath432 and define @xmath433 by @xmath434 .
the pair @xmath435 then defines the map @xmath436 between the homotopy fibers @xmath437 . in construction [ map1 ]
we confine ourselves to a description of the maps @xmath431 for @xmath45 as we will not need the maps @xmath431 for @xmath438 .
[ map1 ] the values @xmath439 are defined as homotopy fibers of the stabilization maps @xmath440 where @xmath441 is the linear inclusion . a point in @xmath439
is described as a pair @xmath442 , where @xmath443 and @xmath444 satisfies @xmath445 and @xmath446 .
we give a description of the adjoints of the maps @xmath447 , that means the maps @xmath448 given @xmath449 we describe the image @xmath450 where @xmath451 @xmath452 such that @xmath453 , @xmath454 and @xmath455 .
each vector @xmath456 , such that @xmath457 , defines a morphism @xmath458 in @xmath29 by @xmath459 .
similarly , each vector @xmath460 , such that @xmath461 , defines a morphism @xmath462 in @xmath29 by @xmath463 .
all such unit vectors @xmath456 , @xmath460 define rays @xmath464 and @xmath465 from @xmath67 trough @xmath466 and @xmath467 , which cover @xmath105 and @xmath468 respectively .
we define @xmath469 on the closures @xmath470 , @xmath471 as the compositions @xmath472 @xmath473 here the fact that the functor @xmath32 is continuous is used .
it guarantees that the maps @xmath474 , @xmath475 are continuous .
@xmath476 figure [ fig : sigma1 ] describes the map @xmath477 .
lines inside the half - disk represent the rays @xmath465 and the half - circle represents the unit hemisphere inside @xmath468 .
the morphism @xmath478 which is used to define @xmath475 on each ray @xmath465 is determined by the point on the unit hemisphere which the ray @xmath465 intersects .
now we turn our attention to the convergence question .
we will not address this issue in full generality , that means we will not discuss a notion of an analytic functor analogous to such a notion in @xcite .
we will only consider the case when a functor @xmath32 is polynomial of some degree @xmath35 . in this case the taylor tower ` stops ' at the @xmath2-stage , that means the canonical maps @xmath41 are homotopy equivalences and thus the @xmath2-th stage @xmath34 gives us complete homotopy theoretical information about the functor @xmath32 .
polynomial functors are defined as follows .
a continuous functor @xmath120 is called _ polynomial of degree _
if for every @xmath42 the following canonical map is a homotopy equivalence @xmath479{cc } \mathrm{holim } \\ [ -1 mm ] \scriptstyle{0 \neq u \subseteq { \mathbb{r}}^{k+1 } } \end{array } } e({v \ ! \oplus \
! u}).\ ] ] the homotopy limit is taken over the topological poset of all non - zero vector subspaces of @xmath44 ( see @xcite ) . in general , for a functor @xmath480 from a small category @xmath481 ,
the homotopy limit of @xmath32 over @xmath481 can be defined as the total space of a cosimplicial space given by @xmath482 see @xcite .
if @xmath481 is a topological category , this affects the definition of the homotopy limit . in our case , when @xmath481 is @xmath483 with objects non - zero vector subspaces of @xmath44 and one morphism for every inclusion @xmath484 , the sets @xmath485 become disjoint union of stiefel manifolds .
then the disjoint union of spaces @xmath486 becomes a bundle over @xmath487 .
the space of @xmath488-cosimplices of the cosimplicial space which defines the homotopy limit then becomes the space of sections of this bundle , see @xcite for more details .
the property above is an extrapolation property .
notice that the value of the functor @xmath32 at @xmath4 is determined by the values of @xmath32 at vector spaces between @xmath4 and @xmath489 which are strictly greater than @xmath4 . for more details see @xcite .
the derivatives are related to the polynomial functors via the following proposition ( * ? ? ?
* proposition 5.3 . ) .
[ fibseq ] for any @xmath42 the value @xmath490 is homotopy equivalent to the homotopy fiber of the canonical map @xmath43{cc } \mathrm{holim } \\ [ -1 mm ] \scriptstyle{0 \neq u \subseteq { \mathbb{r}}^{k+1 } } \end{array } } e({v \ ! \oplus \
! u})\ ] ] over the base point of the target
. thus if the functor @xmath32 is polynomial of degree @xmath35 then the @xmath155-th derivative functor @xmath428 vanishes .
if the target of the map from proposition [ fibseq ] is path - connected for all @xmath4 , then the converse is also true .
if the target is not path - connected , then we have to investigate also the homotopy fibers over other points than the base point .
this leads to a necessary and sufficient condition for @xmath32 to be polynomial of degree @xmath35 which we state in theorem [ poly1 ] in the case @xmath45 .
it is due to weiss ( but did not appear in @xcite ) and we are grateful to him for showing it to us . before we state it we need to define a modification of the map @xmath477 presented in construction [ map2 ] . we also need some new notation because we will use compactifications of euclidean spaces other than the one - point compactification . for @xmath491 by @xmath492 is denoted the compactification of @xmath4 by the sphere @xmath493 at infinity . in special cases , @xmath494 is the @xmath67-sphere compactification of @xmath405 , with the points at infinity @xmath495 , @xmath496 is the @xmath119-sphere compactification of @xmath407 with points at infinity @xmath497 .
here we also distinguish @xmath498 .
by @xmath499 is denoted the @xmath119-disk compactification of @xmath500 , points at infinity are @xmath501 . [ map2 ] denote @xmath502 , where @xmath503 means the homotopy fiber taken over some point @xmath504 .
let @xmath505 denote the space of pairs of maps @xmath506 where @xmath507 @xmath508 such that @xmath509 , where @xmath510 denotes the image of the point @xmath511 under the self - map of @xmath512 induced by the reflection of @xmath405 .
furthermore we require that @xmath513 and @xmath514 where @xmath515 is as in construction [ map1 ] .
notice that in the case @xmath516 the space @xmath505 is homotopy equivalent to the loop space @xmath517 .
if @xmath518 the space @xmath505 can be thought of as a certain space of paths in @xmath519 where @xmath520 where @xmath521 .
this is depicted in figure [ fig : sigma2 ] .
@xmath522 the map @xmath523 is defined similarly as the map @xmath477 in construction [ map1 ] .
for @xmath524 we define @xmath525 as follows .
given unit vectors @xmath526 , @xmath527 we now have rays @xmath528 , @xmath529 from @xmath67 to @xmath530 or @xmath531 .
these rays cover @xmath494 , @xmath499 respectively .
define @xmath532 @xmath533 if @xmath516
then the map @xmath477 obviously coincides with the map @xmath477 from construction [ map1 ] , see figure [ fig : sigma2 ] again .
now we are ready to state the promised necessary and sufficient condition .
[ poly1 ] a continuous functor @xmath120 is polynomial of degree @xmath5 if and only if for all @xmath42 and for all @xmath534 the map @xmath535 defined in [ map2 ] is a homotopy equivalence .
we sketch the proof .
we need to show that the map @xmath536{cc } \mathrm{holim } \\ [ -1 mm ] \scriptstyle{0 \neq u \subset { \mathbb{r}}^2 } \end{array } } e({v \ ! \oplus \ !
u})\ ] ] is a homotopy equivalence if and only if for all choices of @xmath4 and @xmath511 the map @xmath477 is .
consider the map @xmath537 induced by @xmath538 and the map @xmath539 obtained by the restriction via @xmath540 . for @xmath534 consider the induced map , say @xmath541 , between the homotopy fibers over @xmath511 of both maps .
the second map is a fibration so the target homotopy fiber can be replaced by the fiber over @xmath511 . in lemma [ x_y ] below we show that the space @xmath505 is homotopy equivalent to this fiber .
the map @xmath477 is such that it is homotopic to the composition of the map @xmath541 , a suitable homotopy equivalence from the homotopy fiber to the fiber over @xmath511 of the map @xmath542 and the identification of this fiber with the space @xmath505 given in lemma [ x_y ] ( this should be clear after reading the proof of lemma [ x_y ] below ) .
now the map ( [ poly ] ) is a homotopy equivalence if and only if for all choices of @xmath511 the map @xmath541 is .
[ x_y ] the space @xmath505 is homotopy equivalent to the fiber of the map @xmath543{cc } \mathrm{holim } \\ [ -1 mm ] \scriptstyle{0 \neq u \subset { \mathbb{r}}^2 } \end{array } } e({v \ ! \oplus \ ! u } ) \rightarrow e({v \
! { \mathbb{r}}})\ ] ] over the point @xmath511 . as a topological space
the category @xmath544 is @xmath545 , with one morphism ( an inclusion , denoted by @xmath546 ) for every point of @xmath547 to @xmath548 .
we will denote by the same @xmath549 also the inclusion of @xmath401 into @xmath550 . according to the definition the homotopy limit of the functor @xmath551 over this category
is given by a pair of maps @xmath552 where @xmath553 is a bundle over the circle @xmath554 with the fiber over @xmath105 the space @xmath555 and such that @xmath556 is a section of this bundle , @xmath557 and @xmath558 for every @xmath105 , @xmath559 in @xmath554 . in a more condensed form
we can say that the source of the map @xmath560 is defined just on the cone over the circle @xmath554 .
the projection to @xmath512 is given by @xmath561 , therefore taking the fiber over @xmath534 means to require in addition that @xmath562 .
notice that we can think of @xmath553 as a subbundle of the trivial bundle over @xmath554 with the fiber @xmath563 .
the compatibility conditions between @xmath556 and @xmath560 express the fact that the two sections of the two bundles commute with the embedding of the subbundle .
the identification of the fiber over @xmath511 with the space @xmath505 is obtained via a sequence of reparametrizations of the maps @xmath556 , @xmath560 which correspond to certain isotopies of the embedding of @xmath553 into @xmath564 as follows .
( the reader might find it useful to draw some pictures of the situation similar to figure [ fig : sigma2 ] . )
firstly , recall the @xmath119-sphere compactification @xmath496 of @xmath407 with points at infinity @xmath565 and redefine the fiber over @xmath511 as the space of maps @xmath566 such that for @xmath567 with @xmath461 we have @xmath568 if @xmath569 and @xmath570 if @xmath571 , and @xmath572 if @xmath573 and @xmath574 if @xmath571 . here
@xmath465 denotes the vector subspace of @xmath548 generated by @xmath467 . by ` rotating ' the fibers of @xmath553 inside @xmath563 redefine our space again as the space of maps @xmath575 as above , but satisfying now @xmath576 if @xmath569 and @xmath577 if @xmath571 , and @xmath578 if @xmath569 and @xmath579 if @xmath571 , where @xmath580 and @xmath581 are as in construction [ map1 ] . in the next step
consider @xmath582 and redefine the fiber over @xmath511 as the space of maps @xmath583 such that for @xmath567 with @xmath461 we have @xmath584 , and @xmath585 if @xmath586 and @xmath587 .
finally choose a homeomorphism @xmath588 such that its restriction to @xmath582 is the projection onto @xmath589 and the restriction to @xmath590 is the identity .
the resulting space of maps is the desired @xmath505 .
[ pattern ] note that if we want to use theorem [ poly1 ] to prove that a certain functor @xmath32 is polynomial of degree @xmath5 there are two possibilities for each @xmath534 ; the spaces @xmath591 , @xmath505 might be either empty or not .
so the proof splits into two parts .
namely , 1 .
show that @xmath591 is empty if and only if @xmath505 is empty , and 2 .
if @xmath591 is not empty , show that @xmath592 is a homotopy equivalence .
this finishes the discussion of polynomial functors .
now we turn to the taylor tower .
we have already mentioned that for a continuous functor @xmath32 the functors @xmath34 are its polynomial approximations .
this is the content of theorem 6.3 of @xcite which in a slightly weaker form reads as : [ approx ] let @xmath120 be a continuous functor .
then for @xmath593 there exists a functor @xmath594 and a natural transformation @xmath595 such that : 1 .
@xmath596 is polynomial of degree @xmath35 , 2 .
if @xmath32 already is polynomial of degree @xmath35 then the map @xmath597 is a homotopy equivalence for all @xmath4 .
the formula for @xmath34 is given in the proof of theorem 6.3 in @xcite .
for example , if @xmath288 we have for all @xmath598 @xmath599{cc } \mathrm{hocolim } \\ [ -1.7 mm ] \scriptstyle{m \in { \mathbb{n } } } \end{array } } e({v \ ! \oplus \ !
{ \mathbb{r}}}^m).\ ] ] now we address the issue how to determine the layers of the taylor tower .
they are determined by the @xmath2-th _ derivative spectrum _
@xmath47 which is the ` ordinary ' spectrum associated to the coordinate free spectrum given by the @xmath2-th derivative functor @xmath426 .
namely @xmath600 and the structure maps are specializations : @xmath601 as we have mentioned before the tower is the expansion at infinity and therefore the spectrum @xmath47 should be thought as the @xmath2-th derivative of @xmath32 at _
we have also already indicated that there is an action of the orthogonal group @xmath48 on the @xmath2-th derivative spectrum associated to @xmath426 .
the relevant statement in @xcite is proposition 3.1 and simplified it reads as : [ action ] there exists a unique family @xmath602 of left actions @xmath603 which makes the maps @xmath604 into @xmath48-maps . here
@xmath48 acts diagonally on the domain of the map @xmath431 , it acts on @xmath605 because it acts on @xmath606 . as we have said before , the polynomial approximations @xmath34 fit together to form the _ taylor tower _ of @xmath32 which is the diagram ( [ taylor ] ) . the natural transformation @xmath607 is essentially given as the @xmath608-st polynomial approximation of @xmath596 .
the usefulness of the taylor tower follows from the fact that the differences between its stages , @xmath46 , which are called the layers of the tower , can be described using the derivatives .
the formula is given in theorem 9.1 of @xcite .
it states : [ tower ] for a continuous functor @xmath120 , any @xmath609 and any @xmath42 there is the following homotopy fibration sequence : @xmath610 \rightarrow t_k e(v ) \xrightarrow{r_k } t_{k-1 } e(v)\ ] ] where the subscript @xmath611 denotes the homotopy orbit spectrum .
the group @xmath48 acts diagonally on the smash product , it acts on @xmath47 by proposition [ action ] and it acts on @xmath612 because it acts on @xmath606 .
once we have established the convergence question and we know the layers of the taylor tower we can use the taylor tower to obtain some information about the functor @xmath32 .
one way how to do this is suggested in the introduction for our functor @xmath28 .
the input for the taylor tower of @xmath28 is given in theorems [ thm1 ] . and
[ thm2 ] . for examples how
the taylor tower can be used to give information about other functors see @xcite .
[ category ] the functor @xmath28 defined in section [ chap : suspension ] is a continuous functor @xmath613 which is obtained from the functor to the category @xmath614 ( or @xmath615 ) by the geometric realization .
although we do not generalize the machinery of orthogonal calculus to this setting we often use constructions in the category @xmath614 ( or @xmath616 ) , for example the construction of the homotopy fiber of a map or other constructions in section [ chap : thetaf1 ] in construction [ map3 ] .
after geometric realization these constructions always commute ( at least up to homotopy ) with the corresponding constructions in the category @xmath30 which justifies their use .
now we start to apply the machinery of orthogonal calculus to study the functor @xmath28 defined in section [ chap : suspension ] . the aim of this section is to identify the homotopy type of the spaces @xmath617 .
this is done in proposition [ crit - f ] where we give a criterion for the space @xmath617 to be empty and we show that when it is not empty it is homotopy equivalent to a certain space @xmath618 .
the homotopy groups of the space @xmath618 are the groups @xmath619 which are special cases of the @xmath620-groups which are obstruction groups in codimension @xmath119 surgery theory .
the groups @xmath619 are well known to be isomorphic to the @xmath62-groups of the trivial group from ` ordinary ' surgery theory .
thus we also obtain a calculation of the homotopy groups of the spaces @xmath617 .
relating the comparison between the block structure spaces of real projective spaces of different dimensions with codimension @xmath119 surgery theory is not our original idea . at the level of structure
sets it was used by browder - livesay , lpez de medrano and wall in order to calculate the structure sets of the real projective spaces ( see @xcite , @xcite , @xcite ) .
the statements at the level of block structure spaces which we review here follow from general theory of @xcite and from the treatment of codimension @xmath621 surgery theory in @xcite .
in this section we use the definition of the block structure space @xmath17 from construction [ blockdef ] and we work on the level of @xmath131-sets .
this is legal by remarks [ models ] , [ category ] .
we also adopt the convention that an embedding of manifolds will always mean locally flat embedding .
this condition is fulfilled in all cases that we treat here .
we include a brief review of the codimension @xmath621 surgery theory in which we fix the terminology , the notation and we also introduce some constructions that will be used in section [ chap : thetaf1 ] .
the reference for the general theory is @xcite .
the treatment there is slightly more general , see remarks by the editor at the beginning of the chapter .
another useful reference is @xcite .
we do assume however a basic knowledge of ` ordinary ' surgery theory , in particular , the algebraic definition of the @xmath62-groups and the construction of the surgery obstruction for a degree one normal map ( see @xcite ) .
first we need the following construction : [ split - decomposition ] let @xmath622 be an embedding of manifolds .
it induces the following decomposition of the ambient manifold @xmath0 .
the submanifold @xmath7 has a normal block bundle @xmath623 , with total space @xmath624 and with the total space of the associated sphere bundle denoted by @xmath625 .
the closure of the complement of @xmath624 in @xmath0 is denoted by @xmath626 .
thus we have @xmath627 [ def_split ] let @xmath622 be an embedding of manifolds with the normal block bundle @xmath623 .
a simple homotopy equivalence @xmath628 from a manifold @xmath629 of the same dimension as @xmath0 is said to be _ split along _
@xmath7 w.r.t @xmath623 , if @xmath6 is transverse to @xmath7 w.r.t . @xmath623 and the following restrictions are simple homotopy equivalences : * @xmath630 , * @xmath631 .
[ splitprob ] let @xmath622 be an embedding of manifolds as above .
we will refer to a simple homotopy equivalence @xmath632 as a _ splitting problem _ ( _ along _ @xmath7 ) .
given a splitting problem @xmath6 along @xmath7 the task is to change @xmath6 by a homotopy to a simple homotopy equivalence which is split along @xmath7 .
if @xmath633 the obstruction theory reduces to the ordinary surgery theory with the @xmath62-groups as obstruction groups ( ( * ? ? ?
* theorem 11.3 ) ) . in cases
@xmath634 the obstruction theory is different .
firstly , we can assume that @xmath623 is a vector bundle .
then given a splitting problem @xmath6 along @xmath7 there is an obstruction @xmath635 in a group @xmath636 .
the obstruction vanishes if and only if the problem can be solved . here
@xmath637 and @xmath638 is the following diagram of fundamental groupoids : @xmath639 \ar[r ] & \pi(c ) \ar[d ] \\
\pi(m ) \ar[r ] & \pi(x ) . } \ ] ] the groups @xmath636 depend only on @xmath638 , dimension @xmath108 , codimension @xmath621 and the orientation characters of @xmath7 and @xmath0 .
see @xcite for the details of the construction of the @xmath620-groups . in the special case
when @xmath640 is an isomorphism also the other horizontal morphism in @xmath638 is an isomorphism and thus the diagram depends only on the vertical morphism @xmath641 .
in this case the group @xmath642 is denoted @xmath643
. this will be the most important case for us .
suppose now we are given an embedding of manifolds @xmath644 which induces the diagram of fundamental groupoids @xmath638 .
following general philosophy of @xcite to ` spacify ' all obstruction groups in surgery theory we can define a space @xmath645 such that @xmath646 ( see also @xcite ) . in an outline , a @xmath2-simplex of the space
@xmath645 is a splitting problem @xmath647 of manifold @xmath134-ads along an embedding @xmath648 of manifold @xmath134-ads , which is solved on one part of the boundary and comes with a reference map @xmath649 which respects the decompositions of @xmath629 and @xmath0 given by @xmath648 and @xmath622 as described in construction [ split - decomposition ] .
we will not give a precise definition of this space in general , instead we look at the special case of an embedding @xmath650 of a manifold as the zero section of the total space of some disk bundle @xmath268 .
let @xmath651 be the induced homomorphism of fundamental groupoids .
the embedding corresponds to the @xmath652-situation , so we will talk about the space @xmath653 .
notice that we are provided with a map @xmath654 . on a @xmath2-simplex @xmath655
it is given by pulling back the disk bundle @xmath656 along @xmath6 .
let @xmath511 be a @xmath67-simplex in @xmath657 , that means @xmath511 is a simple homotopy equivalence @xmath658 which can also be seen as a splitting problem along @xmath7 .
the splitting obstruction @xmath659 lives in the group @xmath660 .
denote @xmath661 by definition ( see @xcite ) a @xmath2-simplex of @xmath662 is a simple homotopy equivalence @xmath663 which respects faces , such that @xmath664 for @xmath665 , such that @xmath666 , and @xmath667 is split along @xmath7 where @xmath668 is obtained by omitting @xmath669 in the target .
it is not hard to see that , * if @xmath670 then @xmath662 is empty , * if @xmath671 then @xmath672 . for the second statement ,
notice that a @xmath2-simplex of the space @xmath662 is a splitting problem of manifold @xmath673-ads which is solved on the part @xmath674 of the boundary .
as such it can be considered as a @xmath2-simplex of the space @xmath675 , which gives an inclusion map which is the indicated homotopy equivalence ; it induces isomorphisms on homotopy groups .
see for example @xcite for the reference .
consider now the embedding @xmath676 .
we have the decomposition @xmath677 where @xmath623 is the normal disk bundle of @xmath31 and @xmath678 , @xmath679 .
consider now the embedding @xmath680 , the induced homomorphism @xmath681 and a homotopy equivalence @xmath682 with the splitting obstruction @xmath683 .
consider @xmath684 by the above discussion we have that , * if @xmath670 then @xmath685 is empty , * if @xmath671 then @xmath686 .
now let @xmath687 be a @xmath67-simplex in @xmath688 .
consider the space @xmath689 .
a @xmath2-simplex of this space can be viewed as a splitting problem over @xmath690 along @xmath691 satisfying certain conditions on the boundary .
the homotopy type of @xmath617 is identified in the following proposition .
[ crit - f ] let @xmath692 and consider the homotopy equivalence @xmath693 as a splitting problem along @xmath31 with the splitting obstruction @xmath694
. then the following holds * if @xmath670 then @xmath617 is empty , * if @xmath671 then @xmath695 . firstly notice that there is a map @xmath696 given by ` coning off ' .
this map is a homotopy equivalence by the following argument . by transversality
there is the following homotopy cartesian square @xmath697 \ar[d ] & { \widetilde{\mathcal{s}}^s}(c , s(\nu ) ) \ar[d ] \\ { \widetilde{\mathcal{s}}^s}(e(\nu),s(\nu ) ) \ar[r ] & { \widetilde{\mathcal{s}}^s}(s(\nu ) ) . } \ ] ] by the result of ( * ? ? ?
* theorem 12.1 ) , if @xmath54 , then the forgetful map @xmath698 is a homotopy equivalence .
it is also well known that the forgetful map @xmath699 is a homotopy equivalence .
therefore also the left hand vertical map is a homotopy equivalence . the ` coning off ' map is its homotopy inverse .
abusing notation we denote by @xmath511 also the image of @xmath700 under some homotopy inverse of the coning off map .
we have the following diagram which commutes up to homotopy @xmath701 \ar[r ] & { \widetilde{\mathcal{s}}^s}({\mathbb{r}}p(v ) ) \ar@{=}[d ] \ar[r]^-{\nu^ { ! } } & { \widetilde{\mathcal{s}}^s}(e(\nu),s(\nu ) ) \ar[d ] \\
f^{(1)}(v , y ) \ar[r ] & { \widetilde{\mathcal{s}}^s}({\mathbb{r}}p(v ) ) \ar[r]^-{f(\iota ) } & { \widetilde{\mathcal{s}}^s}({\mathbb{r}}p({v \ ! \oplus \ ! { \mathbb{r } } } ) ) . } \ ] ] the left hand vertical map is induced by the right hand vertical map which is a homotopy equivalence and so the left hand map is also a homotopy equivalence . by the criterion preceding the theorem we get the desired statement .
[ 1stder ] if @xmath692 we have @xmath702 we have the homomorphism @xmath703 , where the subscript @xmath4 stands for the orientation character .
this equals @xmath704 . by
* corollary 12.9.1 . )
we have that @xmath705 where @xmath706 if the orientation character is @xmath707 and @xmath708 if the orientation character is @xmath709 .
this together with the @xmath61-periodicity of the @xmath62-groups yields the desired statement .
following @xcite we now give a concrete and explicit description of the isomorphism @xmath712 by definition a @xmath2-simplex in @xmath618 is a splitting problem @xmath713 along @xmath714 satisfying certain conditions on the boundary .
the isomorphism @xmath715 sends the homotopy class of @xmath6 to the splitting obstruction @xmath716 of @xmath6 which lives in the group @xmath711 .
simultaneously we give a description of the map @xmath717 a @xmath67-simplex of @xmath718 is a splitting problem @xmath719 along @xmath31 and the map @xmath715 sends @xmath6 to its splitting obstruction .
so we consider a splitting problem @xmath720 , where @xmath721 and we find a splitting obstruction @xmath635 in @xmath711 . if @xmath722 is odd , then @xmath711 is trivial so it is enough to consider the case when @xmath722 is even and hence also @xmath133 is even .
the even dimensional @xmath62-groups are defined algebraically in @xcite . for a group @xmath723 with a homomorphism @xmath724 and for @xmath725 an element of the group @xmath726
is represented by a _ simple @xmath727-hermitian form _ @xmath728 where @xmath729 is a stably free @xmath730$]-module and the maps @xmath731 , \quad \mu \colon g \rightarrow { \mathbb{z}}[\pi]/\{x - ( -1)^l \overline{x } \},\ ] ] where @xmath732 is an involution on @xmath730 $ ] induced by the homomorphism @xmath467 , satisfy certain conditions , see ( * ? ? ?
* theorem 5.2 ) .
a submodule @xmath733 of @xmath729 is called a _ lagrangian _ of the form
@xmath728 if the map @xmath734 induces an isomorphism @xmath735}(h,{\mathbb{z}}[\pi])$ ] , @xmath736 and @xmath737 .
the form @xmath728 admits a lagrangian if and only if it represents a zero element in the @xmath62-group in which case it is called a _
hyperbolic _ form . in @xcite
the surgery obstruction for converting an even dimensional degree one normal map into a simple homotopy equivalence is described as an element of such an @xmath62-group .
following @xcite we now describe how to associate a simple @xmath727-hermitian form which represents an element in @xmath711 to the splitting problem @xmath6 as above .
[ obstruction ] so we start with @xmath713 a homotopy equivalence satisfying appropriate conditions on the boundary . denote by @xmath738 and by @xmath739 the embedding @xmath740 and
note that @xmath739 is a degree one normal map .
the obstruction @xmath635 coincides with the obstruction to make the embedding @xmath739 into a homotopy equivalence by an ambient surgery on @xmath741 inside @xmath742 ( see @xcite or @xcite for the definition of the ambient surgery on a submanifold ) .
this is defined as follows .
let @xmath743 .
first make @xmath739 into an @xmath488-connected map by an ambient surgery .
the ordinary surgery obstruction associated to @xmath739 is a simple @xmath744-hermitian form @xmath745 over @xmath746 $ ] , where @xmath747 .
as usual the elements of @xmath733 can be represented by @xmath488-dimensional spheres immersed in @xmath741 , the form @xmath734 is given by the @xmath746$]-intersection numbers of these spheres and @xmath748 is given by the @xmath746$]-self - intersection numbers .
let us now instead look at the double covers and try to change @xmath749 by an equivariant ambient surgery . here
@xmath750 denotes the non - trivial double cover of @xmath32 .
the ordinary non - equivariant surgery obstruction associated to @xmath751 is a simple @xmath744-hermitian form @xmath752 over @xmath753 , with the same @xmath733 , only considered now as a @xmath753-module .
the maps @xmath475 , @xmath754 are now given by @xmath753-intersection , resp .
@xmath753-self - intersection numbers of the representing immersed spheres .
the form @xmath755 is hyperbolic .
more to the point , @xmath756 is a separating submanifold and we have a decomposition @xmath757 .
the @xmath753-module @xmath758 splits as @xmath759 the involution @xmath238 switches @xmath760 and @xmath761 which are both lagrangians of @xmath762 .
the equivariant ambient surgery obstruction for @xmath751 is now defined as follows .
denote by @xmath763 , @xmath764 , where @xmath377 is the nontrivial element of @xmath91 .
then @xmath765 is a simple @xmath766-hermitian form over @xmath753 where @xmath725 .
it represents the desired obstruction @xmath635 in @xmath711 .
furthermore we may assume that @xmath767 is a free @xmath753-module on generators @xmath768 whose boundaries are represented by the spheres @xmath769 disjointly embedded in @xmath770 ( because @xmath760 is a lagrangian over @xmath753 ) .
let @xmath771 be the normal disk bundle of @xmath770 in @xmath772 .
then the generators @xmath773 can be represented by the disks @xmath774 disjointly embedded in @xmath772 so that the disk @xmath774 is attached to the normal bundle @xmath771 at the sphere @xmath775 ( see @xcite ) .
similarly @xmath761 is generated by the elements @xmath776 whose boundaries are represented by the spheres @xmath777 and the elements @xmath778 are themselves represented by the disks @xmath779 attached to @xmath771 at @xmath780 .
we did not prove that @xmath715 is an isomorphism if @xmath593 , we have only given a description of the map @xmath715 .
however , from the fact that @xmath715 is an isomorphism if @xmath593 follows an important realization theorem which says that if @xmath593 then any element of @xmath711 can be represented by a splitting problem over @xmath781 along @xmath714 .
we refer to @xcite for the details .
[ equiv - model - of - f1 ] in this section we have used a model of @xmath617 such that a @xmath2-simplex was a splitting problem over @xmath690 along @xmath691 satisfying certain conditions on the boundary . in the next section it will be more convenient to use another model where a @xmath2-simplex is an equivariant splitting problem over @xmath782 along @xmath783 ( it is clear what involutions are meant ) satisfying corresponding conditions on the boundary .
in this section we prove the two main theorems of this paper , theorem [ thm1 ] . and theorem [ thm2 ] .
we start with : [ thm1 ] let @xmath253 be the functor defined by @xmath784 . if @xmath54 then the canonical maps @xmath785{cc } \mathrm{holim } \\ [ -1 mm ] \scriptstyle{0 \neq u \subset { \mathbb{r}}^2 } \end{array } } f({v \ ! \oplus \ ! u})\ ] ] are homotopy equivalences .
the definition of the functor @xmath787 for a general continuous functor @xmath120 is given in ( * ? ? ?
* theorem 6.3 ) .
it says that @xmath787 is the homotopy colimit of the direct system @xmath788 where @xmath789{cc } \mathrm{holim } \\ [ -1 mm ] \scriptstyle{0 \neq u \subset { \mathbb{r}}^2 } \end{array } } e({v \ ! \oplus \ !
u}),\ ] ] and the natural transformation @xmath790 is given by the canonical maps .
theorem [ thm1 ] says that in the case of the functor @xmath28 the maps @xmath791 are homotopy equivalences if @xmath52 .
the claim of corollary [ cor1 ] follows . the pattern of the proof is to verify the condition of theorem [ poly1 ] for the map ( [ map ] ) to be a homotopy equivalence , if @xmath54 .
this condition says that we have to verify that a certain map @xmath792 is a homotopy equivalence for all choices of @xmath793 .
the space @xmath505 and the map @xmath477 were described for a general continuous functor @xmath120 in construction [ map2 ] .
we work with the @xmath131-set models , so in construction [ map3 ] we provide a @xmath131-set model of the space @xmath505 and we also provide a @xmath131-set description of the map @xmath477 which up to homotopy commutes with the geometric realization . the homotopy type of the space @xmath617 has been described in proposition [ crit - f ] .
it says that @xmath617 is empty if and only if the splitting obstruction @xmath659 of a certain equivariant splitting problem @xmath794 over @xmath795 along @xmath69 is non - zero ( see also remark [ equiv - model - of - f1 ] ) .
if @xmath796 then the space @xmath617 is homotopy equivalent to a certain space @xmath618 whose homotopy groups are calculated in proposition [ 1stder ] .
we give a corresponding description of the homotopy type of the space @xmath505 in proposition [ crit - x ] .
it says that @xmath505 is empty if and only if the splitting obstruction @xmath797 of a certain equivariant splitting problem @xmath798 over @xmath799 and along @xmath800 is non - zero .
if @xmath801 then the space @xmath505 is homotopy equivalent to the space @xmath802 whose homotopy groups are also calculated by proposition [ 1stder ] .
as we have indicated in remark [ pattern ] the proof then splits into two parts . to prove the condition ( 1 ) we have to show that the space @xmath617 is empty if and only if the space @xmath505 is empty
this is the content of proposition [ prop1 ] .
it is proved using the criteria mentioned above .
notice that the equivariant splitting obstructions @xmath659 and @xmath797 live in the same group @xmath803 .
therefore it is enough to check that @xmath804 in @xmath803 .
the proof of this statement is the crucial part of the whole proof .
unfortunately it is rather technical .
it is given in a sequence of lemmas : lemma [ lem1 ] , [ lem2 ] , [ lem3 ] . to prove the condition ( 2 ) we have to show that if @xmath617 is not empty then the map @xmath805 is a homotopy equivalence .
this is done by showing that it induces isomorphisms on homotopy groups . using propositions [ crit - f ] , [ crit - x ] and [ 1stder ] we see that the spaces @xmath617 and @xmath505 have isomorphic homotopy groups .
thus we have to show that the map @xmath477 induces isomorphisms between abstractly isomorphic groups .
the idea here is the following .
an element of @xmath806 can be represented by a splitting obstruction @xmath635 of some equivariant splitting problem @xmath6 over @xmath807 along @xmath808 .
the image @xmath809 is a splitting problem over @xmath810 along @xmath811 with the splitting obstruction @xmath812 which represents the homotopy class of @xmath809 in @xmath813 .
the condition ( 2 ) is proved by showing that @xmath814 in @xmath711 .
this statement is the content of proposition [ prop2 ] .
its proof turns out to be just a parameterized version of the proof of the statement of condition ( 1 ) .
[ map3 ] we need a @xmath815 model of the space @xmath505 and a description of the map @xmath477 on the level of @xmath816 .
the descriptions given here and the descriptions from construction [ map2 ] commute with the geometric realization up to homotopy .
recall the @xmath815 model for @xmath617 . here
@xmath817 is an equivariant splitting problem along @xmath69 .
a @xmath2-simplex of @xmath617 is an equivariant homotopy equivalence @xmath818 w.r.t .
the antipodal involution on the target and w.r.t .
some free involution on the source , it respects faces , it is transverse to @xmath808 , for @xmath819 , @xmath820 , we have @xmath821 and @xmath822 is equivariantly split along @xmath341 .
now we give a @xmath815 model for @xmath505 .
a @xmath2-simplex of @xmath505 is an equivariant homotopy equivalence @xmath823 w.r.t .
the antipodal involution on the target and w.r.t . to some free involution on the source which respects faces and satisfies the following conditions on the boundary .
the conditions on @xmath287 are the following .
firstly @xmath287 restricted to @xmath827 is split along @xmath828 .
the restriction of @xmath287 to @xmath829 is fixed and it is given as the composition @xmath830 here @xmath831 means the unreduced suspension map and the map @xmath832 is a homeomorphism given by @xmath833 where @xmath834 is the unique rotation of @xmath550 which fixes @xmath4 and sends @xmath835 to @xmath836 .
a @xmath67-simplex of @xmath505 is depicted in figure [ fig : sigma3 ] .
the top edge corresponds to the part @xmath839 of the boundary .
the closure of the complement of @xmath840 in the boundary is @xmath841 . on each vertical line over @xmath842
the restriction of the map @xmath809 is given by @xmath843 .
compare this with figure [ fig : sigma2 ] .
the vertical lines here correspond after reparametrization to the rays from the origin in figure [ fig : sigma2 ] .
if @xmath505 is not empty then there exists a @xmath67-simplex , say @xmath6 , of @xmath505 .
since @xmath6 restricted to @xmath847 is by definition split along @xmath848 any such a @xmath67-simplex @xmath6 can be seen as a solution of the splitting problem @xmath798 and hence @xmath801 .
as we have remarked in construction [ map2 ] if the space @xmath505 is not empty it is homotopy equivalent to a certain space of paths in the space @xmath849 .
it is not difficult to see that this path space is homotopy equivalent to @xmath850 .
propositions [ crit - f ] , [ crit - x ] tell us when the spaces @xmath617 and @xmath505 are empty .
we see that both obstructions @xmath659 and @xmath797 live in the group @xmath803 . in order to prove the condition ( 1 )
it is enough to show that @xmath851 in @xmath803 .
this is done in the sequence of the following three lemmas .
note that a precomposition with a homeomorphism does not change the splitting obstruction of a splitting problem .
therefore when investigating @xmath797 it is enough to look at the splitting obstruction of the map @xmath852 and
so from now on we denote @xmath853 one would expect that in order to find the splitting obstruction @xmath797 it is necessary to first adjust the map @xmath798 to make it transverse to @xmath800 .
amazingly , the map @xmath798 as constructed by the formula ( [ mapf_y2 ] ) already is transverse to @xmath800 !
this is what we prove in this lemma , more precisely we show an equivalent statement that the map @xmath855 is transverse to @xmath856 where @xmath857 denotes the embedding @xmath858 .
we use an easier notion of transversality which is good enough for our purposes .
it goes as follows .
let @xmath622 be a codimension @xmath119 embedding .
suppose that there exist an open neighborhood @xmath32 of @xmath7 in @xmath0 and a submersion @xmath859 such that @xmath860 .
( this implies that @xmath7 has a trivial normal bundle . )
we say that a map @xmath628 is transverse to @xmath7 if @xmath861 is a submersion in a neighborhood of @xmath862 .
note that this condition can be verified locally . in our case , @xmath863 and @xmath7 is the image of @xmath864 under the embedding @xmath856 .
let @xmath865 where @xmath866 is the ` height function ' defined by the projection on the second coordinate of @xmath548 .
( this should be suitably restricted to a neighborhood of @xmath7 in @xmath0 . )
we use the following decompositions ( [ decomposition1 ] ) , ( [ decomposition2 ] ) into ( overlapping ) codimension @xmath67 ( non - compact ) submanifolds @xmath867 where @xmath868 is the total space of the open normal vector bundle @xmath869 of @xmath69 in @xmath795 , and @xmath870 , where @xmath871 with @xmath872 , and @xmath873 where @xmath874 is the total space of the open normal vector bundle @xmath875 of @xmath69 in @xmath876 and @xmath877 , where @xmath878 .
we start with the embedding @xmath880 .
this is an embedding of a codimension @xmath119 subbundle where both bundles are over @xmath881 .
the map @xmath882 restricted to @xmath883 can be described as @xmath884 where @xmath885 is the orthogonal projection in the fiber of @xmath886 onto the image of @xmath887 .
now the map @xmath794 was transverse to @xmath69 , therefore the map @xmath888 is also transverse to @xmath69 .
let @xmath889 and let @xmath890 be the open normal bundle of @xmath770 in @xmath795 and denote by @xmath891 the open normal bundle of @xmath770 in @xmath876 .
the restriction @xmath892 is a bundle map .
now it is not difficult to see that the restriction @xmath893 is transverse to @xmath894 in the sense above .
next we look at the embedding @xmath895 .
observe that the composition @xmath896 is a homeomorphism .
therefore we can consider the embedding @xmath897 as an embedding of a graph of some function @xmath898 into @xmath899 .
the restriction of @xmath882 can be expressed as @xmath900 .
again it is not difficult to see that the restriction @xmath901 is transverse to @xmath902 in the sense above .
a key observation here is that the manifold ( with boundary ) @xmath910 has a decomposition @xmath911 into codimension @xmath67 submanifolds which corresponds to the previous decomposition @xmath912 .
this can be seen as follows .
we can think of @xmath910 as a pullback of @xmath856 along @xmath913 : @xmath914
\ar[r ] & s({v \ ! \oplus \ ! { \mathbb{r } } } ) \times i \
\ar[d]^{s(\zeta ) \circ j } \\ s({v \ ! \oplus \ ! { \mathbb{r}}}^2 ) \times i \ar[r]^{\sigma ( f_y ) \times { \mathrm{id } } } & s({v \ ! \oplus \ ! { \mathbb{r}}}^2 ) \times i.}\ ] ] look first at the following restriction of this diagram : @xmath915 \ar[r ] & e(\widetilde{\nu } ) \times i \
\ar[d]^{s(\zeta ) \circ j } \\
e(\widetilde{\mu}_{{\widetilde{n } } } ) \times i \ar[r]^{\sigma ( f_y)| \times { \mathrm{id } } } & e(\widetilde{\mu } ) \times i.}\ ] ] in this diagram the lower horizontal arrow is a bundle map over the restriction @xmath916 . here
@xmath917 is a @xmath118-dimensional vector bundle over @xmath771 , @xmath883 is a @xmath118-dimensional vector bundle over @xmath918 and @xmath894 is its codimension @xmath119 subbundle .
it follows that the top row @xmath919 is a bundle map and it also follows from the construction that @xmath920 can be identified with @xmath921 .
secondly , look at the restriction of the diagram to @xmath922 .
here , because the map @xmath856 is a map over @xmath923 , we can look at the following pullback diagrams @xmath924 \ar[r ] & c_\pm \times i \
\ar[d]^{{\mathrm{pr}}_1 \circ s(\zeta ) \circ j } \\ s({v \
! \oplus \ ! { \mathbb{r}}}^2 ) \ar[r]^{\sigma ( f_y ) } & s({v \
! { \mathbb{r}}}^2).}\ ] ] where @xmath925 is the projection onto the first factor .
as noted before the composition @xmath926 is a homeomorphism and thus we obtain that @xmath927 .
this yields the decomposition ( [ decomposition3 ] ) . on the two submanifolds
@xmath929 the restriction of the map @xmath373 can be described as follows .
observe first that @xmath930 , and @xmath931 and that @xmath932 can be seen as a cone on @xmath795 .
the restriction of @xmath373 is the following composition : @xmath933 where @xmath934 denotes the canonical extension of @xmath794 to the cones suitably restricted .
the homeomorphism @xmath935 can be described as follows .
the fiber @xmath936 is embedded in @xmath937 as an open ` meridian ' through the point @xmath938 .
it is not difficult to see that the cone map @xmath934 is transverse to all such meridians except the one corresponding to @xmath939 ( details left to the reader ) .
recall that @xmath945 is a homotopy equivalence about which we can assume that it is transverse to @xmath69 and also that it is obtained from a homotopy equivalence @xmath946 by ` coning off ' ( by proposition [ crit - f ] ) . here
@xmath868 denotes the total space of the normal disk bundle of @xmath69 in @xmath795 .
the splitting obstruction @xmath659 is represented by a simple @xmath727-hermitian form @xmath943 obtained as an equivariant ambient surgery obstruction associated to the embedding @xmath947 ( see construction [ obstruction ] ) .
the homotopy equivalence @xmath948 is constructed by formula ( [ mapf_y2 ] ) .
let @xmath949 be the total space of the normal disk bundle of @xmath795 in @xmath876 and note that @xmath798 already is in the form of a homotopy equivalence obtained from another homotopy equivalence @xmath950 by ` coning off ' so there is no need to adjust it .
the splitting obstruction @xmath797 should now be read off as the equivariant ambient surgery obstruction from the embedding @xmath951 by the recipe given in construction [ obstruction ] . in previous lemma [ lem2 ]
we have made a first step by identifying the homotopy type of @xmath910 .
moreover it follows from the construction of the map @xmath798 by formula ( [ mapf_y2 ] ) that the embedding @xmath952 when restricted to @xmath953 is just the unreduced suspension map @xmath954 .
so we have a commutative diagram @xmath955^{\simeq } \ar[r]^-{\sigma \widetilde{i } } & e(\widetilde{\nu } ' ) \times \{0\ } \ \ar[d]^{\simeq } \\ { \widetilde{n } } ' \ar[r]^-{\widetilde{i } ' } & e(\widetilde{\nu } ' ) \times i}\ ] ] where all maps are embeddings .
let @xmath956 . from the diagram above it follows that @xmath952 is @xmath957-connected , and also that @xmath958 and that the corresponding splitting of @xmath733 is the same @xmath959 as before .
it only remains to find the embedded spheres representing the generators of @xmath733 and to show that their intersection numbers are preserved .
let @xmath960 be the set of generators of @xmath733 as a @xmath753-module , where generators are represented by embedded spheres @xmath961 as in construction [ obstruction ] .
denote the images under suspension map by @xmath962 .
we need to put them in a general position .
first we describe the topology of @xmath910 more .
the manifold @xmath910 is a special manifold @xmath145-ad with two distinguished parts of the boundary @xmath963 . from the decomposition ( [ decomposition3 ] )
we also see that the product @xmath771 is a separating submanifold of @xmath910 . by taking its normal disk bundle ( using collars )
we obtain the codimension @xmath67 embedding @xmath964 , which when restricted to @xmath965 coincides with the embedding of the closed normal bundle of @xmath966 .
in particular there are disks @xmath967 attached to the normal bundle @xmath771 as described in construction [ obstruction ] .
see figure [ intersections ] .
denote now by @xmath968 the maps @xmath969 . using the isotopy induced by the embedding of @xmath970 , @xmath971 into @xmath910 ( we omit the details )
we can change the embedded spheres @xmath972 to embedded spheres in a general position given by the two formulas @xmath973 and @xmath974 .
it follows that the associated intersection forms are @xmath975 , since the only intersections are the intersections in @xmath770 .
this is proved by showing that the map @xmath477 induces isomorphisms on homotopy groups .
notice that we have @xmath980 because the surgery obstruction map @xmath715 is an isomorphism we can realize any element of the group @xmath711 by a @xmath2-simplex in @xmath617 , that means by an equivariant splitting problem @xmath981 along @xmath783 satisfying certain conditions on the boundary . in order to show that the map @xmath477 induces an isomorphism on homotopy groups it is enough to show that if the splitting obstruction @xmath635 is represented by a simple @xmath727-hermitian form @xmath985 , where @xmath986 , then the splitting obstruction @xmath987 is represented by the same simple @xmath727-hermitian form . but
this is just a version of the statement that was proved in a sequence of lemmas [ lem1 ] , [ lem2 ] , [ lem3 ] parametrized over @xmath988 .
the proofs carry over to the parametrized case as well .
notice that the proof of proposition [ prop1 ] tells us in the special case @xmath516 that modulo some low - dimensional deviations the spectrum @xmath57 is an @xmath989-spectrum and we know its homotopy groups .
namely @xmath990 for all @xmath65 .
* acknowledgements . *
the present paper grew out of my thesis defended at the university of aberdeen .
i would like to thank my supervisor michael weiss for conversations and encouragement during the work on this project and my thesis referees michael crabb and andrew ranicki for useful comments .
my thanks are also to mamuka jibladze for conversations and for the pictures in the paper .
finally i would like to thank max planck institut fr mathematik in bonn for the hospitality during final stages of the preparation of this paper . | given a compact manifold @xmath0 , the set of simple manifold structures on @xmath1 relative to the boundary can be viewed as the @xmath2-th homotopy group of a space @xmath3 .
this space is called the block structure space of @xmath0 .
we study the block structure spaces of real projective spaces .
generalizing wall s join construction we show that there is a functor from the category of finite dimensional real vector spaces with inner product to the category of pointed spaces which sends the vector space @xmath4 to the block structure space of the projective space of @xmath4 .
we study this functor from the point of view of orthogonal calculus of functors ; we show that it is polynomial of degree @xmath5 in the sense of orthogonal calculus .
this result suggests an attractive description of the block structure space of the infinite dimensional real projective space via the taylor tower of orthogonal calculus .
this space is defined as a colimit of the block structure spaces of projective spaces of finite dimensional real vector spaces and is closely related to some automorphisms spaces of real projective spaces .
a @xmath2-simplex @xmath196 can be described by a list @xmath197 , where @xmath198 and @xmath199 is an isomorphism in @xmath186 for @xmath187 a morphism in @xmath131 , satisfying various compatibility conditions . here
@xmath200 .
note that if @xmath156 is a trivial @xmath131-groupoid ( all morphisms are the identity ) , then a @xmath2-simplex of @xmath189 is just a collection @xmath201 , where @xmath202 .
thus we recover the underlying @xmath131-set of @xmath156 .
if @xmath156 is a non - trivial @xmath131-groupoid a @xmath2-simplex of @xmath189 is a collection @xmath197 , where objects @xmath203 are just isomorphic to @xmath204 .
[ rem : sharp ] the assignment @xmath205 is a functor from @xmath131-groupoids to @xmath131-sets .
suppose that @xmath156 is a topological @xmath131-groupoid , that means the category @xmath171 is a topological category for each @xmath2 and the face operators are continuous functors
. then @xmath206 can also be endowed with a topology ( the subspace topology of the product topology ) and @xmath189 becomes a @xmath131-space , also in a functorial way .
the geometric realization then converts the @xmath131-set ( or the @xmath131-space ) @xmath189 into a genuine space again in a functorial way . in the notation
we do not distinguish between the underlying @xmath131-set ( or @xmath131-space ) and its geometric realization .
we note that all these constructions have pointed versions .
a @xmath2-_block _ @xmath7 is a manifold with boundary such that the boundary @xmath207 is decomposed into codimension @xmath67 submanifolds @xmath208 , for @xmath209 , whose intersections @xmath210 are codimension @xmath67 submanifolds of the boundaries of @xmath208 and @xmath211 , and so on . for @xmath187 a morphism in @xmath131 we denote @xmath212 .
a _ block map _ @xmath213 between two @xmath2-blocks is a map , such that @xmath214 for @xmath215 .
a @xmath2-block is _ special _ if @xmath216 .
an example of a special @xmath2-block is @xmath1 for @xmath0 a closed manifold .
now we are ready for the definition of the block structure space @xmath17 .
[ blockdef ] let @xmath0 be a closed @xmath108-dimensional manifold .
the block structure space @xmath17 of @xmath0 is the associated pointed @xmath131-set of the pointed @xmath131-groupoid whose @xmath2-simplices are pairs @xmath217 , where @xmath7 is a special @xmath2-block , and a block map @xmath132 is a simple homotopy equivalence between @xmath133-dimensional @xmath2-blocks .
face operators @xmath172 are obtained by taking the face @xmath218 and restricting @xmath6 to @xmath219 .
an isomorphism between @xmath217 and @xmath220 is a block homeomorphism @xmath221 such that @xmath222 .
the base point in dimension @xmath2 is @xmath136 .
a @xmath2-simplex of @xmath17 can be described by a list @xmath223 , where for @xmath187 the @xmath224 is a special @xmath163-block , the map @xmath225 is a simple homotopy equivalence of @xmath163-blocks , @xmath226 is a homeomorphism of @xmath163-blocks and the maps @xmath227 , @xmath228 satisfy various compatibility conditions .
the list @xmath229 is the correct formalization of a manifold @xmath134-ad . in the case
@xmath45 , the manifold @xmath145-ad is a cobordism in the sense of @xcite , @xcite as described earlier .
the block structure space @xmath17 as defined above is a kan @xmath131-set for any manifold @xmath0 .
for kan @xmath131-sets there is a convenient combinatorial formula for homotopy groups ( see @xcite ) and using it we immediately obtain the isomorphisms ( [ htpygroups ] ) .
the following lemma provides us with a method for constructing other models of the block structure space .
[ skelet ] let @xmath230 be a map of pointed @xmath131-groupoids such that @xmath231 and @xmath189 are kan @xmath131-sets .
if @xmath232 induces a bijection on equivalence classes of objects for every @xmath2 , then the induced map @xmath233 is a homotopy equivalence . using the combinatorial expression of @xmath138
it is easy to see that the map @xmath234 induces isomorphisms on homotopy groups .
now we specialize to the real projective spaces .
note first that any homotopy equivalence @xmath235 is simple because @xmath236 ( see @xcite ) .
therefore we drop the word simple when dealing with the real projective spaces .
here is the promised intermediate model @xmath143 for the block structure space of @xmath31 .
[ f - spheres ] [ doublecovers ] let @xmath143 be the associated pointed @xmath131-set of the pointed @xmath131-groupoid whose @xmath2-simplex is a pair @xmath237 , where : * @xmath238 is a free involution on @xmath239 , which , as a map , is a block map , * @xmath240 is a block map which is an equivariant homotopy equivalence w.r.t .
@xmath238 on the source and @xmath241 on the target , where @xmath241 is the product of to the antipodal involution on @xmath4 ( restricted to @xmath69 ) with the identity on @xmath173 .
face maps are defined by restriction and an isomorphism between @xmath237 and @xmath242 is an equivariant block homeomorphisms @xmath243 such that @xmath222 .
the base point is the pair @xmath244 .
a pair @xmath237 which represents a @xmath2-simplex in @xmath143 will be shortly referred to as a _ structure _ on @xmath245 there is a map @xmath246 defined on the underlying @xmath131-groupoids by taking quotients .
[ modelone ] the map @xmath247 is a homotopy equivalence if @xmath248 .
using lemma [ skelet ] it is enough to show that for each @xmath249 the map from the @xmath2-simplices of the underlying groupoid of @xmath143 to the @xmath2-simplices of the underlying groupoid of @xmath140 induces an isomorphism on the equivalence classes of objects .
we proceed by induction on @xmath2 . from the generalized poincar conjecture in @xmath248 we have that any manifold @xmath7 homotopy equivalent to @xmath31 has the universal cover homeomorphic to @xmath69 which proves the statement for @xmath250 . for the induction step
we apply the @xmath251-cobordism theorem which implies that the universal cover of any manifold @xmath134-ad @xmath7 mapping via a homotopy equivalence to @xmath252 is homeomorphic to @xmath245 . |
we have performed a series of experiments on the -ray spectroscopy of @xmath12 hypernuclei using a germanium detector array called hyperball @xcite .
the main purpose of these experiments is to study the spin dependence of the @xmath10 interaction via precise measurements of level spacings in @xmath13-shell hypernuclei . in this paper
, we report on results for the and hypernuclei investigated via the reaction at the brookhaven national laboratory . in principle
, interactions between baryons can be studied by baryon - baryon scattering experiments . however , such scattering experiments are extremely difficult except for the nucleon - nucleon ( @xmath15 ) case due to the short lifetimes of the other baryons . on the other hand
, @xmath12-hypernuclear data can provide information on the @xmath10 interaction .
free hyperon - nucleon ( @xmath16 ) interaction models have been theoretically constructed as extensions of @xmath15 interaction models by assuming flavor su(3 ) symmetry @xcite .
effective @xmath16 interactions have often been approximated as a g - matrix derived from the free @xmath16 interaction , either in nuclear matter as a function of density @xcite or using the pauli exclusion operator for finite nuclei @xcite .
many - body effects entering into the calculation of the effective interaction for a finite shell - model space from the g - matrix are small , except for the effects of coupling @xmath12-hypernuclear and @xmath17-hypernuclear states ( @xmath12-@xmath17 coupling ) @xcite .
this is in part due to the lack of pauli blocking for a @xmath12 in a @xmath12 hypernucleus and in part because the @xmath10 interaction is relatively weak because one - pion exchange is forbidden due to isospin conservation . on the other hand ,
the @xmath10@xmath18 interaction can be mediated by one - pion exchange and it is found to play a significant role in the energy - level spacings of s - shell hypernuclei despite the @xmath19 mev @xmath12@xmath17 mass difference @xcite .
a hypernucleus ( ) has a spin - doublet ( @xmath20 ) structure when the core nuclear ( ) level has non - zero spin ( @xmath21 ) and the @xmath12 is in the 0@xmath22 orbit . in this simple weak - coupling limit , the splitting of the doublet is due only to @xmath10 interactions that involve the @xmath12 spin ( see eq .
( [ lneff ] ) below ) together with a contribution from @xmath12@xmath17 coupling . therefore ,
such spin - doublet structures in hypernuclei provide us with information on the spin - dependence of the @xmath10 interaction . the @xmath10 effective interaction @xcite
can be written in the form @xmath23 . \label{lneff}\end{aligned}\ ] ] the terms correspond to the spin - averaged and spin - spin central , the @xmath12-spin - dependent spin - orbit ( @xmath12-spin - orbit ) , the nucleon - spin - dependent spin - orbit ( @xmath24-spin - orbit ) , and the tensor interactions , respectively . for @xmath25 configurations in @xmath13-shell hypernuclei
, there are five independent two - body matrix elements @xmath26 that can be written in terms of radial integrals associated with each of these terms .
the radial integrals are conventionally denoted by the parameters @xmath27 , @xmath28 , @xmath29 , @xmath30 and @xmath31 , respectively @xcite , with @xmath29 and @xmath30 the coefficients of @xmath32 and @xmath33 ( because @xmath34 is proportional to @xmath35 for a @xmath12 in an @xmath22 orbit ) . since @xmath27 contributes equally to all levels ,
the level spacings are given by a linear combination of the four spin - dependent parameters and corresponding core level spacings .
as noted above , only @xmath28 , @xmath29 , and @xmath31 contibute to doublet spacings in the weak - coupling limit . for the same p - shell model space , the @xmath10@xmath18 interaction can be written and parametrized in the same way as the @xmath10 effective interaction @xcite . for a fixed @xmath12@xmath17 coupling interaction ,
the four @xmath10 parameters that govern the spin dependence of the interaction can be determined phenomenologically to fit various @xmath13-shell hypernuclear level spacings .
in particular , a level spacing which is dominantly given by one parameter provides the parameter value almost independently of the others .
for example , if the core state has @xmath36 only @xmath28 contributes to the doublet spacing .
this is the case for the ground - state doublet spacing in ( @xmath37 ) due to an almost pure @xmath38 configuration for the core ( @xmath39 ) .
similarly , if the ( @xmath40 ) core state for the excited - state doublet in ( @xmath41 ) has @xmath42 and @xmath43 , only @xmath29 contributes to the doublet spacing .
then , by comparing the experimentally determined parameter values with the predicted values by theoretical models of @xmath16 interactions , we can test the validity of the models .
however , effects of @xmath16 interactions on hypernuclear levels are small .
in particular , spin - doublet spacings of @xmath13-shell and heavier hypernuclei are expected to be much smaller than 1 mev . in some cases , hypernuclei have quite small spin - doublet spacings of the order of a few tens of kev leading to the use of the term `` hypernuclear fine structure '' .
therefore , an energy resolution of the order of several kev is necessary for spectroscopic studies to resolve the level spacings .
thus , -ray spectroscopy with germanium ( ge ) detectors has exclusive access to these structures .
for this purpose , a ge detector array dedicated to the -ray spectroscopy of hypernuclei , hyperball , was built in 1998 @xcite and a project to investigate precise structure of hypernuclei started .
the energy levels of @xmath12 hypernuclei were studied by and reaction spectroscopy and by -ray spectroscopy with nai counters prior to the start of the hyperball project in 1998 .
information on the spin - spin interaction can be obtained from the observations of the @xmath44 mev spin - flip m1 transitions between the ground - state doublet ( @xmath45 ) in and with nai counters @xcite .
however , it has been shown that @xmath12-@xmath17 coupling can make a large contribution to these doublet spacings @xcite . recently
, four - body calculations have been performed @xcite for a number of the nijmegen @xmath16 potential models with the result that both the spin - spin interaction and @xmath12-@xmath17 coupling make significant ( comparable ) contributions to the splitting .
a small @xmath12-spin - orbit interaction was reported for the first time based on the fact that the splitting between the @xmath46 and @xmath47 substitutional states in is close that of the underlying hole states of the @xmath48o core @xcite .
afterwards , the -ray spectroscopy of with nai counters suggested a very small spin - orbit strength corresponding to @xmath49 mev from a limit of @xmath50 kev for the spacing of the 3-mev excited - state doublet ( @xmath41 ) based on the fact that the width of the peak containing both @xmath2 rays was comparable with the resolution of @xmath51 kev @xcite . in 1998 , high - precision -ray spectroscopy experiments for and were carried out with hyperball using the and the reactions , respectively @xcite .
the ground - state doublet spacing of 692 kev in provided a spin - spin parameter value of @xmath52 mev without the inclusion of @xmath12-@xmath17 coupling . in the @xmath13 shell
, the scale of energy shifts due to @xmath12-@xmath17 coupling is expected to be roughly a factor of 4 smaller ( with a strong state dependence ) than that for the @xmath53 hypernuclei @xcite , and provides only about 12% of the ground - state doublet spacing in .
a strength of @xmath54 mev was also established from the excitation energy of the @xmath55 state in @xcite . in the next experiment ,
the rather small spacing of the @xmath41 doublet in noted above was resolved @xcite .
the @xmath56 state has been determined to be the upper member of the doublet based on @xmath57b target data from the present experiment @xcite .
the spacing of 43(5 ) kev leads to @xmath58 mev @xcite .
the sign and magnitude of @xmath29 are consistent with the ordering and spacing of the @xmath59 and @xmath60 states in measured with a nai counter array @xcite
. the energy - level spacings discussed above are not very sensitive to the tensor interaction .
consequently , for the derivation of other three parameters ( @xmath61 and @xmath30 ) , @xmath31 was assumed to be in the range of values 0.01 0.06 mev predicted using @xmath10 interactions from the nijmegen one - boson - exchange ( obe ) models ( nsc97f , nsc89 , nd and nf ) in @xmath62-matrix calculations @xcite . with @xmath31 taken to be 0.030 mev , the remaining @xmath10 parameters that fit almost perfectly the four bound excited states of @xcite are ( parameters in mev ) @xmath63 among the four @xmath10 spin - dependent interactions ,
only direct information on the @xmath10 tensor interaction has not been obtained experimentally .
the derivation of the tensor interaction strength is important not only to complete the set of parameters but also to remove the theoretical assumption implicit in the derivation of the other three parameter values .
the strong @xmath15 tensor interaction is well understood by one - pion exchange , reduced at short distances by rho exchange .
but in the case of @xmath10 , these exchanges are forbidden due to the zero isospin of the @xmath12 . the corresponding pseudo - scalar @xmath64 and vector @xmath65 exchanges
are allowed but cancel strongly because their masses are more similar than those of the @xmath66 and the @xmath67 @xcite . as a result ,
the obe models predict small strengths for the tensor interaction .
this , and a relatively small variation for different models , can be seen from table xi of ref .
@xcite for the nijmegen hard - core and soft - core models , and from table xx of ref .
@xcite for the extended soft - core models .
the ground - state doublet spacings in @xmath46-shell hypernuclei have large contributions from the tensor interaction .
for example , in the simplified @xmath68 coupling model @xcite , the ground - state doublet spacings of and are given by @xmath69 while the @xmath56 , @xmath70 spacing in is 1.5 times larger . in contrast , the spacing of the doublet based on the @xmath47-hole state of , @xmath71 has a coefficient of @xmath31 relative to @xmath28 that is an order of magnitude smaller than for the @xmath46 doublet in eq .
( [ p12 ] ) . however , the difficulty of determining the tensor interaction strength is due not only to its small value but also to the small spacings of the spin doublets for which @xmath31 contributes strongly .
since @xmath31 is expected to have a small positive value , and @xmath28 has a relatively large positive value , the contributions of @xmath28 and @xmath31 are expected to cancel strongly in @xmath46-shell hypernuclei , as can be seen from eq .
( [ p12 ] ) .
hence the spacings are expected to be quite small and the ordering of the states in a doublet can be uncertain .
a small spacing also means that the lifetime of the upper member of a doublet may be long compared with the lifetime for weak decay ( @xmath72 ps ) .
in this situation , the only safe way to measure a doublet spacing is to measure the energies of @xmath2-rays feeding both members of the doublet from a higher level . in order to study the @xmath12n tensor interaction ,
the reaction on @xmath3o is ideal because bound states of both and can be produced .
figure [ onrelate ] shows the level schemes of and ( the isospins of 1/2 for and 0 for are omitted ) .
predicted -ray transitions are also shown . for kaon momenta less than 1 gev / c
, the reaction proceeds predominantly by non - spin - flip ( @xmath76 ) transitions .
theoretical cross sections calculated in distorted - wave impulse approximation ( dwia ) @xcite for all the natural - parity states in fig .
[ onrelate ] are shown in fig .
[ dwia ] for the incident kaon momentum used in the present experiment ( @xmath77 mev / c ) .
substitutional states , in which a neutron is replaced by a @xmath12 without changing the orbit , are strongly populated at forward angles ( @xmath78 ) via @xmath79 transitions .
the three @xmath80 states shown in fig . [ onrelate ]
are of this nature ( see below ) and were observed , along with the two @xmath6 states ( @xmath81 ) , at cern @xcite .
the excitation energies given for the @xmath80 states in fig .
[ onrelate ] come from a reanalysis of the cern data @xcite .
the @xmath6 and @xmath40 ( @xmath82 ) states are seen strongly in the reaction @xcite and can be produced in the reaction at larger angles ( see fig .
[ dwia ] ) .
the excitation energies of the @xmath40 states are taken from ref .
@xcite while the energies of the negative - parity states come from the present experiment @xcite .
the thresholds in fig .
[ onrelate ] depend on the @xmath12 binding energies ( @xmath83 values ) for the hypernuclei involved and these are not reliably determined from emulsion data @xcite ( few events with ambiguous interpretations ) .
states and @xmath80 states of ( thick lines ) are most strongly populated by this reaction .
the supscript @xmath84 shows excitation energies from the present experiment , @xmath85 from @xcite and @xmath86 from @xcite .
the particle - decay thresholds , except for + p and + n ( see text ) , are determined from emulsion data @xcite assuming @xmath87 mev for .,width=321 ] the negative - parity states of in fig .
[ onrelate ] have dominant @xmath88 configurations while the positive - parity states have configurations except for the @xmath89 state . the @xmath90 and @xmath91 states
are the @xmath13-substitutional states while the @xmath89 state has an @xmath92 configuration and is the @xmath22-substitutional state . both @xmath6 states in fig .
[ onrelate ] are particle bound . the 6.6-mev excited @xmath93 state is expected to decay to both ground - state doublet members ( @xmath94 ) by m1 transitions . the upper level of the ground - state doublet is expected to decay to the lower level by a spin - flip m1 transition .
however , for a small spacing of less than 100 kev , the detection efficiency of hyperball becomes small .
in addition , the weak decay of @xmath12 in nuclei ( @xmath95 ps ) would compete with the m1 transition for a spacing of less than 100 kev .
thus , we need to detect both rays , @xmath96 and @xmath97 and determine the doublet spacing from the energy difference between these rays .
the spin ordering of the doublet can be determined from the branching ratio for these transitions .
since the spin - flip cross section is expected to be much smaller than the non - spin - flip one , the yield of transitions from the @xmath5 state , which should decay mainly to the @xmath98 state by an @xmath99 transition , is expected to be small .
it has been pointed out @xcite that bound states of can be produced following proton emission from particle - unbound states in .
the proton - emission threshold of is expected to be at about @xmath100 7.8 mev based on systematics for the difference of the @xmath83 values for and . making the same approximation and adding the 5.9 mev between the neutron and proton thresholds in @xmath48o gives @xmath101 mev for the neutron - emission threshold in .
reaction at @xmath77 mev / c as a function of the laboratory scattering angle ( @xmath102 ) .
the cross sections are for the @xmath93 , @xmath90 , @xmath103 , and @xmath89 states shown in fig .
[ onrelate ] . for the @xmath98 , @xmath91 , and @xmath104 states ,
the first three cross sections should be multiplied by 0.61 , 1.65 , and 0.60 , respectively.,width=264 ] the full @xmath105 wave functions for the positive - parity states of contain admixtures of the following types of configurations , @xmath106 these states can decay by the @xmath22-wave and @xmath107-wave proton emission to the positive - parity states in via the @xmath108 components in their wave functions .
the lowest five - positive parity states of are shown in fig .
[ onrelate ] . for
the same reason as in , direct observation of the spin - flip m1 transition between the ground - state doublet members may not be possible .
therefore , to observe rays from , the upper doublet ( @xmath109 ) or the @xmath9 state should be produced .
the @xmath39 state at @xmath110 mev in , which is the core level for the upper doublet , decays by m1 transitions to the @xmath111 state at @xmath112 and the ground @xmath39 state with the branching rate of 100:4 .
therefore , the cascade m1 transitions @xmath113 and @xmath114 are likely to be observed in . the baryonic decay rates of the states ( @xmath115 ) have been calculated in the translationally invariant shell model ( tism ) @xcite .
the predictions are that 12% of the @xmath90 state decays to the @xmath9 state and 28% of the @xmath91 state to the @xmath116 state .
the @xmath117 states are calculated to decay mainly to the @xmath118 state .
measurement of the ground - state doublet spacings in and/or determine a relationship between the values of @xmath28 and @xmath31 from the full shell - model version ( which includes @xmath12@xmath17 coupling ) of eq .
( [ p12 ] ) .
a determination of the energies of more excited states of and tests the previously determined spin - dependent interaction strengths .
in particular , the excited - state doublet in is based on a mainly @xmath38 core state , like the @xmath119li ground state , and the doublet spacing should be large and determined mainly by @xmath28 .
the -ray spectroscopy experiment e930(01 ) on and was carried out using the d6 beam line @xcite at the brookhaven national laboratory ( bnl ) alternating gradient synchrotron ( ags ) for a period of two months in the fall of 2001 .
both and were produced via the reaction with the rays being detected by a large acceptance germanium ( ge ) detector array called hyperball .
incident and outgoing meson momenta were measured by magnetic spectrometers and -ray spectra were obtained by selecting events corresponding to the mass region . a side view of the detection apparatus is shown in fig . [ setup ] . the d6 beam line @xcite provided a high - purity and high - intensity @xmath121 beam .
the incident @xmath121 momentum was set to 0.93 gev / c .
the energy loss in the beam - line detectors and the target medium reduced the momentum to 0.91 gev / c at the reaction point .
this @xmath121 momentum was chosen for a balance between hypernuclear production yield and doppler shift effects .
the repetition rate of the beam spill was 4.6 s and the beam duration time was 1.5 s. in each spill , @xmath122 21 gev / c protons bombarded the primary platinum target and a @xmath121 beam with a typical @xmath123 ratio of 3 was delivered with a typical intensity of @xmath124 . in total ,
@xmath125 kaons were incident on the 20 g/@xmath126 water target ( hereafter `` target '' denotes the water target ) . the incident @xmath121 momentum was measured in the downstream part of the d6 beam line ( not shown in the figure ) using the transport matrix with information from a hit position in the plastic scintillation counter hodoscope before the last bending magnet and a straight track measured by three drift chambers ( id13 ) after it .
see ref .
@xcite for details .
the spectrometer for scattered particles consisted of a dipole magnet 48d48 , which has a pole size of 48@xmath12748 and a gap size of 80 cm .
it accepted scattering angles from @xmath128 to @xmath129 in the horizontal direction and from @xmath130 to @xmath131 in the vertical direction .
the scattered particles were bent vertically ( upward ) .
the momentum was measured by the 48d48 and five drift chambers located upstream ( fd13 ) and downstream ( bd12 ) of the 48d48 .
in the present experiment , the 48d48 was operated at 0.8 t , optimized to the outgoing @xmath132 momentum of about 0.8 gev / c .
the field distribution of the 48d48 used in track reconstruction analysis was calculated by the tosca code .
the reaction events were selected in the trigger level by threshold type aerogel counters ( ac ) with @xmath133 located upstream ( ic1 and ic2 ) and downstream ( fc ) of the target , and by time - of - flight in the off - line analysis .
plastic scintillation counters mt ( located upstream of the last bending magnet of the beam line and not drawn in the figure ) , it , and bt were used to measure the time - of - flight and it was used as a timing reference counter for all the detectors .
mt is a horizontally segmented hodoscope and bt is a vertically segmented hodoscope .
the typical flight lengths between corresponding tof counters were 15 m ( mt it ) for incident and 7.9 m ( it bt ) for scattered particles , respectively . a plastic scintillation counter ( fv ) , which covered a scattering angle less than 3@xmath134 was located 50 cm downstream from the target and used to reject unscattered beam particles and very - forward scattered particles in the trigger level .
events with small reaction angles were also rejected in the off - line analysis because the vertex resolution is not enough to select reaction events in the target .
the target was installed between the two ac s ( ic2 and fc ) spaced as closely as possible to minimize kaon decay events occurring in the target region .
such decay events contributed to background , but can not be eliminated in the off - line analysis because their momentum range overlaps with that of hypernuclear production .
the incident particle identification with two stages of ac s ( ic1 and ic2 ) provided an almost pure @xmath121 trigger but caused a kaon suppression of 4% .
the efficiency of the outgoing particle identification by an ac ( fc ) for pions was 98% and the misidentification of kaons as pions was 1% .
however , such kaons were rejected by fv .
hyperball consisted of fourteen sets of ge detectors with bismuth germanate ( bgo ) scintillation counters . a typical configuration and size of the ge detector and the bgo elements
are shown in fig .
[ setup ] .
each ge detector was surrounded by six bgo detectors of 19 mm radial thickness .
each ge detector had an n - type coaxial crystal of about 7 cm @xmath127 7 cm @xmath135 and a relative efficiency to a 3 @xmath127 3 nai counter of 60% .
the bgo counters were used to suppress such background events as compton scattering , @xmath136 decay , and high - energy charged particles .
the background from @xmath137 decay was particularly serious .
each ge detector was equipped with a transistor - reset type preamplifier and connected to a shaping amplifier with a gated integrator ( ortec 973u @xcite ) .
this solved the difficulty of operating the ge detectors in a high counting rate and high energy - deposit rate environment .
we also used timing filter amplifiers ( ortec 579 @xcite ) for timing information . the end cap of each ge detector facing
the beam was located at a distance of 10 cm from the beam axis .
the ge crystals covered a total solid angle of 0.25 @xmath127 @xmath138 sr from the target center .
the total photo - peak efficiency in the beam - on period was measured using -ray peaks from -cascade decays originating in the target and was determined to be ( 1.5 @xmath11 0.3)% for 2.3 mev rays and ( 4.2 @xmath11 1.0)% for 0.7 mev rays after including all electronics dead time and all analysis efficiencies .
the efficiency curve as a function of the energy was simulated by the geant code and the absolute scale was adjusted to fit the measured efficiencies .
the relative efficiency for the beam - on and the beam - off periods was measured by the monitoring system using triggerable 1 kbq @xmath139co sources , each embedded in a plastic scintillator connected to a photo - multiplier tube ( pmt ) and installed behind a ge detector as shown in fig .
[ setup ] ( pulser ) .
these monitoring data were taken during both beam - on and beam - off periods with the scintillator detecting @xmath140 from @xmath141 decay and the corresponding ge detector detecting rays in coincidence . the beam - on / beam - off efficiency ratio for each detector
was measured to be 90 @xmath11 5% .
two -ray sources , and , were used for energy calibration .
however , the -ray energies of available sources were limited to 6.13 mev , while the -ray energies from are expected to be around 6.6 mev .
therefore , we also used rays from activities ( , , and @xmath142 ) produced by the beam in the target and surrounding materials
. in particular , we used -ray peaks following @xmath3n(@xmath143 ) decay at 6129 kev and 7115 kev and their escape peaks .
the was most likely produced by the ( @xmath144 ) reaction on in the target and surrounding bgo detectors .
these rays were observed in the ge - self - trigger data taken during the beam - off period .
energy calibration was performed in the range from 0.1 to 7.1 mev .
the peak shifts between the beam - on and the beam - off periods and the long - term gain shift were also corrected . the shift between beam - on / off was typically 1 kev and the long - term shift was 2 kev at maximum .
the response function of the -ray peak shape after summing up the spectra of the fourteen ge detectors was found to be described a gaussian function up to 6.1 mev .
the beam - on energy resolution was 5.7 kev fwhm for 2-mev rays and 8.6 kev for 6.6-mev rays .
the energy calibration error was found to be 1.0 kev at 2 mev and 1.5 kev for over 5 mev .
the @xmath121 beam trigger was defined as @xmath145 .
the @xmath146 scattering trigger was defined as @xmath147 .
the data were taken according to the trigger defined as @xmath148 .
the trigger rate was typically @xmath149 per spill . to reduce the trigger rate ,
a second - level trigger was made using information on the ge detector hits . when none of the ge detectors had hits , data were not taken .
the second - level trigger rate was typically @xmath150 per spill .
adc and tdc data were collected via four fastbus crates for all the detectors except for the ge detectors , and via fera bus for ge detectors , to the corresponding vme memory modules ( umem @xcite ) on an event - by - event basis .
the accumulated data were read out by the host computer ( linux pc ) in every beam - off period and recorded on a dvd ram after being processed by the host computer .
another type of data , triggered only by the ge detectors ( ge - single - trigger data ) , was also taken in the beam - off period of every synchrotron cycle .
these data were used for energy calibration of the ge detectors .
the mass ( @xmath151 ) was reconstructed as a missing mass in the reaction .
the binding energy of a @xmath12 in the hypernucleus is defined by @xmath152 where @xmath153 and @xmath154 are the masses of the core nucleus ( ) in its ground state and the @xmath12 , respectively .
the absolute mass scale was calibrated using the @xmath155 mass reconstructed from the @xmath156 decay .
the energy losses of the @xmath121 and the @xmath146 in the target medium were corrected event by event . in the present experiment ,
the mass resolution was 15 mev ( fwhm ) .
we selected events with a reaction angle larger than @xmath157 and a reaction vertex point in the target region .
figure [ oxblam ] shows the missing - mass spectrum from the reaction plotted against the @xmath12 binding energy ( @xmath83 ) for those events accompanying rays with energies in the range from 1.5 mev to 7.0 mev . as shown in fig .
[ onrelate ] , four narrow states ( @xmath158 and @xmath91 ) and one broad state ( @xmath89 ) , corresponding to the @xmath159 substitutional state at around @xmath160 mev @xcite , are expected to be produced . however , they can not be resolved in the missing - mass spectrum due to the limited mass resolution . to observe the rays ,
the mass region corresponding to the @xmath93 state was selected .
the mass of the @xmath93 state is near @xmath161 mev from previous experiments @xcite .
consequently , the @xmath93 state region was defined by @xmath162 mev .
the 20-mev gate width corresponds to 88% of all the @xmath93 state events for the 15 mev resolution . on the other hand , to observe the rays , a mass region including the @xmath80 and @xmath40 states ( -state region ) was selected .
the @xmath90 and @xmath91 states are at @xmath163 and @xmath164 mev @xcite , as are the @xmath104 and @xmath117 states @xcite .
therefore , we defined this region to be @xmath165 mev and this gate width covers 94% of the events for the @xmath90 and @xmath91 states .
we also defined the highly unbound region as @xmath166 mev .
these defined regions are shown in fig .
[ oxblam ] . binding energy ( @xmath83 ) for those events accompanying rays with energies of 1.5 7.0 mev .
the indicated regions show the definitions of the @xmath93 state region ( @xmath167 mev ) , the states region ( @xmath168 mev ) and the highly unbound region ( @xmath166 mev ) .
[ oxblam],width=207 ] ) dependent acceptance of the scattered - particle spectrometer .
differences in acceptance for the @xmath93 ( @xmath169 mev ) and @xmath91 ( @xmath170 mev ) states of are negligibly small .
( @xmath171 mev / c ) for these states corresponding to the reaction angles are also shown .
events with reaction angles less than 2@xmath134 were rejected in the off - line analysis.,width=226 ] figure [ acceptance ] shows the acceptance , obtained by a monte - carlo simulation , of the scattered - particle spectrometer as a function of the reaction angle ( @xmath102 ) . the momentum transfer [ @xmath171 ( mev / c ) ] corresponding to @xmath102 is also shown .
we simulated the cases of producing the @xmath172 mev ( @xmath93 ) and @xmath173 mev ( @xmath91 ) states .
differences in the @xmath171-dependent acceptances for these states were found to be negligibly small .
the @xmath121 momentum at the reaction point was distributed around 0.91 gev / c with a width of 0.06 gev ( fwhm ) on account of the original beam momentum spread and energy loss in the thick target . as is evident from fig .
[ dwia ] , the cross sections for the @xmath80 , @xmath6 , and @xmath40 states formed via the reaction at @xmath174 mev / c peak at @xmath175 , @xmath176 , and @xmath177 , respectively . taking into account the angular distribution of these reactions and the spectrometer acceptance ,
the initial recoil momenta of for the @xmath93 and @xmath91 states were estimated to be in the range of @xmath178 mev / c and @xmath179 mev / c , respectively .
the recoil momentum of was also calculated to be in the range of @xmath180 mev / c assuming that the proton was emitted isotropically .
the corresponding stopping times of these hypernuclei in the target medium were calculated using the srim code @xcite to be in the range of 1.5 2.5 ps and 0 2 ps , respectively .
since the timing of a ge detector hit with respect to the beam timing ( it hit ) varies with -ray energy , the timing gate width was set as a function of the energy .
the gate width was 50 ns for 0.5 mev , 30 ns for 2 mev , and 20 ns for over 5 mev .
the gate width for the bgo counters was set to be 50 ns so as to achieve efficient background suppression without oversuppression from accidental coincidences . to observe and identify statistically weak -ray peaks ,
the background level should be minimized .
when gating the particle - unbound region , we found a lot of rays , such as and rays around the 2 mev region , originating from ( @xmath181 ) reactions in the detector and surrounding materials .
since the neutrons were assumed to reach these materials a few ns after the triggered timing , we made another tdc cut which was used only to identify these rays .
figure [ tdccut ] ( a ) shows a tdc spectrum for a typical ge detector plotted for -ray energies over 1.5 mev .
the timing cut conditions , prompt and delayed , were defined as shown in fig .
[ tdccut ] ( a ) .
figures [ tdccut ] ( b ) and ( c ) show -ray spectra plotted for the highly - unbound region in with the prompt and delayed timing cuts , respectively .
rays from the target materials are enhanced in the prompt spectrum ( b ) and rays from the detector materials are enhanced in the delayed spectrum ( c ) .
the energies of the expected rays from and are larger than 1 mev except for the spin - flip transitions within the spin doublets .
they are all m1 transitions so that their lifetimes are expected to be much shorter than , or of the same order as , the stopping time of the recoiling hypernucleus .
therefore , each -ray peak has a doppler - broadened shape determined by the lifetime of the state , the stopping time , the response function of the ray , and the kinematical conditions of the experiment .
when the lifetime of the initial state is much shorter than the stopping time , @xmath182 ps , the -ray peak shape is fully broadened by the doppler - shift effect .
if we know the velocity of the recoiling hypernucleus ( @xmath141 ) and the angle between the hypernuclear velocity and -ray emission ( @xmath183 ) in the laboratory frame , the shifted -ray energy can be corrected for by the relativistic formula , @xmath184 where @xmath185 and @xmath186 are the doppler - shift corrected energy and the measured energy , respectively .
the direction of -ray emission in the laboratory frame was calculated from the reaction vertex point and the position of the ge crystal with a hit .
the velocity ( @xmath141 ) and direction of the hypernucleus were calculated from the kaon and pion momentum vectors measured by the incident and outgoing particle spectrometers .
first , we consider the doppler - shift correction for rays . since the -ray energies of @xmath99 transitions from the @xmath93 state are larger than 6 mev
, the lifetime of the state is expected to be very short because the lifetime of the corresponding core level ( @xmath187 ) is very short ( @xmath188 fs ) .
figure [ ogam ] shows the simulated -ray peak shapes of assuming an original energy of 6.55 mev , giving ( a ) the original response function , ( b ) the fully doppler - broadened shape and ( c ) the shape after doppler - shift correction for ( b ) .
second , we consider the doppler - shift correction for rays .
the lifetimes of the upper - doublet states ( @xmath109 ) in are expected to be a few fs @xcite .
the recoil velocity and the recoil direction of are changed from those of due to the proton emission . however , since the mass of the secondary nucleus ( ) is not so different from the mass of primary nucleus ( ) and the emitted proton energy is small , the doppler - shift correction can be also applied to the the secondary nucleus .
we simulated a doppler - shift corrected -ray peak shape for using a simulated @xmath141 for ( @xmath91 ) , assumed to be produced according to the calculated angular distribution ( see fig .
[ dwia ] ) .
proton emission was assumed to be isotropic .
we found that the doppler - shift correction makes a broadened peak narrower and square - shaped .
figure [ ngamsim ] shows the simulated -ray peak shapes assuming that a 2-mev ray is emitted from following proton emission ; ( a ) fully doppler - broadened and ( b ) doppler - shift - corrected peak shapes .
in addition , we also simulated other decay processes such as + and + @xmath189 . in these cases
, it was found that the correction does not make the doppler - broadened peak shapes narrower . in these simulations
, we assumed that hypernuclei emit rays isotropically in the center - of - mass frame .
although angular correlations are expected to exist between the scattered @xmath132 and rays @xcite , the effect on the corrected energy is simulated to be negligibly small .
rays from following proton emission from the ( @xmath91 ) state produced via the reaction .
the upper figure shows ( a ) fully doppler - broadened and ( b ) doppler - shift - corrected peak shapes .
the lower figure shows the lifetime dependence of the peak shape , ( a ) @xmath190 stopping time ( no broadening , original response function ) , ( b ) @xmath191 ps and ( c ) @xmath192 ps . , width=264 ] information on the lifetime of a state can be obtained by analyzing the partly doppler - broadened peak shape .
this is called the doppler - shift attenuation method ( dsam ) . in particular ,
when the stopping time and the lifetime are of the same order of magnitude , dsam can be used to determine the lifetime .
this method was first applied for hypernuclei to determine the @xmath193 value of the @xmath55 excited state of @xcite .
dsam can also be applied to the secondary hypernuclei if the recoil momentum is known . among the states to be produced , the ( @xmath9 ) state
is expected to have a lifetime of the same order as the stopping time ( @xmath194 ps ) , an estimate for the lifetime of the state being 0.5 ps @xcite .
however , the absolute mass of is not well determined and the production rates from the @xmath90 , @xmath91 and @xmath22-substitutional states of can not be obtained because of the limited mass resolution .
these ambiguities are included in the systematic errors .
the ambiguities mainly stem from the q value of the @xmath195 + @xmath13 decay .
details are given later . in the simulation
, we assumed that decays to and a proton isotropically in the center - of - mass frame .
the simulated -ray peak shape is shown in fig .
[ ngamsim ] ( bottom ) .
the figure shows the lifetime dependent peak shape for ( a ) @xmath196 stopping time ( no broadening , original response function ) , ( b ) @xmath197 ps and ( c ) @xmath198 ps .
some results for the rays have already been reported in ref .
@xcite which was focused on the determination of the ground - state doublet spacing in . in this paper , details of the results on the rays and additional results on the rays are discussed .
figure [ oxgam ] shows the mass - gated @xmath2-ray energy spectra around 6 mev in coincidence with the reaction at @xmath199 ; ( a ) is plotted for the highly - unbound region ( @xmath200 mev ) without doppler - shift correction ( @xmath201 ) , ( b ) is for the @xmath93-state region ( @xmath202 mev ) without doppler - shift correction ( @xmath201 ) , and ( c ) is for the @xmath93 state region with doppler - shift correction ( @xmath203 ) . here ,
the doppler - shift correction was applied assuming that rays were emitted before the recoiling hypernucleus slowed down .
the highly unbound and the @xmath93 state regions are defined in fig . [ oxblam ] .
we found ( 6129 kev ) and ( 6175 kev ) -ray peaks in the spectra for both the highly - unbound region ( a ) and the @xmath93-state region ( b ) . on the other hand ,
a bump around 6.6 mev was found only in the spectrum for the @xmath93-state region ( b ) and it becomes two narrow peaks after the doppler - shift correction , as shown in ( c ) .
the peaks are attributed to @xmath99 transitions from the @xmath93 state to both ground - state doublet members ( @xmath94 ) in because the @xmath93 state is the only bound state , other than the @xmath6 member of the ground - state doublet , that can be strongly populated by the reaction and -ray peaks from other ( hyper)nuclei can not become so narrow after doppler - shift correction ( fig . [ ngamsim ] ) .
this argument is supported by the fact that the shape of the bump before doppler - shift correction in fig .
[ oxgam ] ( b ) is consistent with the fully doppler - broadened peak shape [ fig .
[ ogam ] ( b ) ] .
in addition , the structure observed at @xmath204 mev in fig .
[ oxgam ] ( c ) has a shape similar to the peaks at @xmath205 6.55 mev and corresponds to their single - escape ( se ) peaks . )
dependence of the @xmath2-ray intensities ( sum of the 6534- and 6560-kev rays ) from the @xmath93 state in measured in the reaction for the @xmath93 state region ( @xmath167 mev ) defined in fig [ oxblam ] . shown
are the -ray counts divided by solid angle taking into account the spectrometer acceptance ( fig .
[ acceptance]).,width=183 ] we made a fit to the doppler - shift corrected spectrum with the simulated peak shape [ fig .
[ ogam ] ( c ) ] assuming that the background around a full - energy peak has two components ; one is constant for the whole region below and above the -ray energy ( @xmath206 ) , and the other is caused by multiple compton scattering and exists only on the lower - energy side with the form @xmath207 for @xmath208 and 0 for @xmath209 . in a simulation taking into account the doppler - shift correction ,
the step - function was smoothed around @xmath210 and then used in the fitting .
we obtained for the lower peak an energy of @xmath211(stat ) @xmath212(syst ) kev and a yield of @xmath213 counts , and for the upper peak , @xmath214 kev and @xmath215 counts .
the systematic errors include the calibration error and the accuracy of the doppler - shift correction .
the excitation energy of the @xmath93 state becomes 6561.7 kev after applying the nuclear recoil correction to the -ray energy .
the energy difference between the two rays , @xmath216 kev , corresponds to the ground - state doublet spacing .
the relative intensities of the rays are @xmath217 where the -ray efficiencies were assumed to be the same .
we also found a gathering of events at 6760 kev in the @xmath6 state region spectrum ( c ) . to reduce the background , we set a narrower mass gate ( @xmath218 mev ) to maximize the signal to noise ratio , @xmath219 , for the 6534- and 6560-kev @xmath2-ray peaks .
then , we found a peak with a statistical significance of @xmath220 ( @xmath221 counts ) at an energy of @xmath222 kev , as shown in fig .
[ oxgam ] inset ( c ) .
the peak width was found to be consistent with the simulated doppler - shift - corrected peak shape .
applying the same mass gate , the sum of the @xmath223 -ray yields was was found to be 262 @xmath11 24 counts .
the -ray intensity ratio was @xmath224 figure [ oxangdis ] shows the reaction - angle ( @xmath102 ) dependence for the sum of the 6534- and 6560-kev rays observed from the @xmath6 state region , @xmath225 mev , defined in fig .
[ oxblam ] .
the sum of these -ray yields were obtained from a fit using the simulated peak shape with the two -ray energies and the ratio of the two -ray yields fixed to be 0.69 from eq .
( [ o16ratio ] ) .
the -ray intensity was obtained from the summed -ray yield divided by a solid angle taking into account the angular dependence of the spectrometer acceptance ( fig .
[ acceptance ] ) .
the observed angular distribution agrees well with calculated distribution in fig .
[ dwia ] for a @xmath226 transition and is thus consistent with the the state that emits the 6534- and 6560-kev rays being the @xmath93 state .
the branching ratio @xmath227 is 0.5 in the weak - coupling limit and 0.41 when the level mixing is taken into account @xcite . the measured relative intensity of the 6534- and 6560-kev rays [ eq . ( [ o16ratio ] ) ]
should therefore give information on the spin ordering of the doublet members .
first , a correction has to be made for the fact that the effective -ray efficiency of the hyperball is different for the two transitions due to angular distribution effects . for pure m1 transitions and a forward reaction angle ( @xmath228 ) @xcite , @xmath229 where @xmath230 is the angle between the @xmath146 and the @xmath2 ray .
the detectors were not arranged isotropically with respect to the @xmath146 direction and effective efficiencies were calculated from a simulation with the result @xmath231(syst . ) .
the error comes from ambiguities in the real size and the shape of the detectors . since we measured the hypernuclear production events at finite angles , the actual ratio is larger than this estimation .
taking into account the @xmath232 correlation , the expected branching ratios of 0.41 or 0.50 should be corrected to 0.33 or 0.40 .
thus , the measured ratio of @xmath233 is slightly larger than expected but still favors a @xmath8 assignment for the lower level of the ground - state doublet .
the intensity of the 6758 kev ray is much weaker than those of the transitions from the @xmath93 state [ eq .
[ ds01 ] ] . with the caveat
that @xmath40 hypernuclear states based on the 6.793-mev @xmath56 or 6.859-mev @xmath55 levels of @xmath48o could possibly be excited through small @xmath234 components in their wave functions [ cf .
( [ config ] ) ] , this -ray transition is tentatively attributed to the m1(@xmath235 ) transition from the @xmath5 member of the @xmath236 doublet , giving an excitation energy of @xmath237 kev for the @xmath5 state . if this assignment is correct , this is the first experimental data to identify directly produced spin - flip and non - spin - flip states in @xmath12 hypernuclei . binding energy ( @xmath83 ) for events accompanying a ray with @xmath238 6600 kev for ( a ) @xmath239 , ( b ) @xmath240 , and ( c ) @xmath241 .
the dotted lines show the expected background shapes plotted for events accompanying -rays with @xmath242 6600 7000 kev and scaled to fit the background region ( @xmath243 mev).,width=321 ] figure [ blam1 ] shows missing - mass spectra as a function of @xmath12 binding energy ( @xmath83 ) for events accompanying a ray with a doppler - shift corrected -ray energy @xmath238 6600 kev corresponding to the @xmath223 transitions . in fig .
[ oxblam ] ( a ) , a prominent peak is observed corresponding to the experimental value of @xmath244 mev for the @xmath93 state . by fitting this peak in with a gaussian ,
the mass resolution was determined to be 15 @xmath11 1 mev .
the mass resolutions obtained by fitting the mass spectra in fig . [ oxblam ] ( b ) and ( c ) were @xmath245 mev .
the fully broadened -ray peak shape contains information on the lifetime .
we fitted the doppler - shift uncorrected spectrum with simulated peak shapes for various lifetimes using the measured -ray energies and yield ratio .
then , we obtained an upper limit for the lifetime of the @xmath93 state in of @xmath246 ps at the 68% confidence level .
figure [ nigam ] shows the mass - gated -ray spectra measured in coincidence with the reaction ; ( a ) and ( c ) are plotted for highly unbound region ( @xmath248 mev ) , ( b ) , ( d ) and ( e ) are for the -states region ( @xmath168 mev ) .
event - by - event doppler - shift corrections are applied for ( e ) .
the highly - unbound and -states regions are defined in fig .
[ oxblam ] .
a prominent peak is found at 2268 kev in spectrum ( d ) which is gated on the -states region .
the -ray peak is taken to be hypernuclear since there are no -ray transitions of this energy in ordinary nuclei with @xmath249 16 .
the -ray peak has two components , namely a narrow part and a doppler - broadened part , which indicates that the lifetime of this transition is close to the stopping time of @xmath205 1 ps for the hypernucleus in the target .
there are four candidates for secondary hypernuclei with thresholds below the ( @xmath91 ) state , namely , , , and ( see fig .
[ onrelate ] ) .
the long lifetime of the 2268-kev ray suggests that the corresponding core transition is also slow leaving the @xmath111 , first - excited state of ( @xmath250 mev , @xmath251 fs ) as the only candidate .
moreover , the excitation energies of members of the first - excited doublets in in and are known to be 2.51 mev for ( @xmath93 ) @xcite ( 2.63 mev from kek e336 @xcite ) and 4.88 mev for ( @xmath56 ) @xcite ( 4.85 mev @xcite ) . the unknown first - excited state of must be based on the 5.17-mev @xmath6 state of and the @xmath91 state of ca nt decay to a state at this high an energy .
therefore , the 2268-kev -ray peak is attributed to .
in addition , two broad peaks at 1960 and 2440 kev can be seen in ( d ) and they become square - shaped narrower peaks in ( e ) after the doppler - shift correction .
these peak shapes are consistent with the result of the simulation for described in sect .
[ shift - correction ] [ see fig . [ ngamsim ] ( a ) and ( b ) ] while , according to the simulation , -ray peaks from and ca nt become narrower .
therefore , these two rays are also attributed to transitions in . on the other hand ,
a narrow -ray peak observed at 2215 kev can be attributed to a part of the broad -ray peak from ( @xmath181 ) because it is clearly enhanced by the `` delayed '' tdc cut [ see fig .
[ tdccut ] ] . as shown in fig .
[ nigam ] ( b ) , no prominent peak corresponding to the ground - state spin - flip m1 transition is observed in the region from 100 kev to 700 kev .
the sensitivity for such a peak and the ground - state doublet spacing will be discussed later ( sect .
[ sec : gsdoublet15 ] ) .
we made fits for the doppler - shift corrected -ray spectrum using the simulated peak shape from fig .
[ ngamsim](a ) .
we obtained energies of @xmath252 kev and 2442.0 @xmath253 @xmath111.7 kev , with corresponding yields of 190 @xmath254 @xmath11 5 and 313 @xmath11 35 @xmath11 5 counts . for the narrow peak
, we obtained the energy of 2267.6 @xmath11 0.3 @xmath11 1.5 kev and a yield of 744 @xmath11 39 @xmath11 15 counts .
the detailed fitting procedure for the 2268-kev ray is described later .
the measured -ray energies correspond to transition energies of 1960.8 , 2267.8 and 2442.3 kev .
taking into account the energy dependence of the -ray efficiency , the relative -ray intensities for the -states region were found to be @xmath255 figure [ nigam ] ( f ) shows the spectrum of rays emitted in coincidence with another ray with a doppler - corrected energy of @xmath256 kev for the mass range of @xmath257 mev .
the shaded region in ( f ) shows the expected background , which is derived from the spectrum of rays accompanied by another ray with @xmath258 kev for the same mass region , with the counts scaled to the counts in the 2442 @xmath11 25 kev gate corresponding to the shaded region in the inset ( f ) .
the mass region @xmath257 mev was set to select the @xmath91 state ( @xmath259 mev ) and higher states .
a peak with 6 counts appeared at 2268 kev , the energy of one of the other rays from .
the probability for 6 events to appear in the narrow region specified by 2268 @xmath11 12 kev from a random fluctuation of the background of 1.2 @xmath11 0.3 counts was estimated to be 0.4 % using a poisson distribution .
thus the possibility of a background fluctuation was rejected .
since the peak shapes indicate that the lifetime of the 2442-kev transition is shorter than that the 2268-kev transition , the decay chain ( 2442 kev ) @xmath195 ( 2268 kev ) was determined .
the yield of the 2442-kev ray for the @xmath257 mev region was obtained to be @xmath260 counts and the expected yield of the 2268-kev ray in coincidence is estimated to be 4.6 counts if no other transition strongly competes with the 2268-kev transition .
therefore , the 6 counts with 1.2 @xmath11 0.3 counts background is consistent with the expected coincidence yield . on the other hand ,
no excess of events was found in the - coincidence spectrum with the 1961-kev ray , consistent with the lower yield of the 1961-kev ray .
figure [ angdist ] ( a c ) shows the reaction angle ( @xmath102 ) dependence of the -ray intensities for the -states region in the reaction .
the -ray intensities were obtained from the measured -ray counts divided by the spectrometer acceptance ( fig . [ acceptance ] ) and normalized to the relative efficiency for the 6550-kev ray .
the panels of the figure show the distributions for the ( a ) 2268-kev , ( b ) 2442-kev , and ( c ) 1961-kev rays .
the spectra ( a ) and ( b ) show forward peaking angular distributions with a significant intensity remaining at large angles ( @xmath261 ) .
this indicates that the excited states de - excited by the 2268- and 2442-kev rays are produced mainly by proton emission from states excited by @xmath262 transitions with some contribution from states produced by @xmath263 transitions .
on the other hand , the excited state leading to the 1961-kev deexcitation is mainly produced from non - substitutional states excited by @xmath264 transitions .
the partly broadened shape of the 2268-kev -ray peak indicates a lifetime comparable to the stopping time of @xmath265 ps . since the @xmath266n(@xmath111 @xmath195 @xmath267 )
transition has a lifetime of 0.10 ps and an energy of 2313 kev , the 2268-kev ray is attributed to a transition corresponding to this core transition , namely , from the @xmath9 state to one of the ground - state doublet members @xmath268 . in the weak - coupling limit , the @xmath269 values for
the decay to the @xmath56 and @xmath70 states would be in the ratio of @xmath270 .
however , taking into account the level mixing , they are calculated to be in the ratio of @xmath271 @xcite .
therefore , the 2268-kev ray is taken to correspond to the @xmath272 transition .
the lowest state in which decays to the @xmath111 state ( the core level of the @xmath9 state in ) is the 3948-kev @xmath39 state and it gives rise to the @xmath118 , @xmath116 doublet .
therefore , the initial state for the 2442-kev -ray transition is most likely to be one of these doublet members .
the @xmath262 dominance shown in the angular distribution ( fig .
[ angdist ] ( b ) ) indicates that the 2442-kev ray mostly stems from the ( @xmath91 ) state or the @xmath273-hole state ( see fig .
[ onrelate ] ) .
the tism calculation @xcite suggests that the ( @xmath116 ) state , rather than the ( @xmath118 ) state , is dominantly produced from the ( @xmath91 ) state . on this basis , the 2442-kev ray is likely to be the @xmath274 transition .
all the excited states higher than the @xmath111 state in decay mainly to the @xmath111 state or the @xmath267 ground state . since no other rays from
are observed , the 1961-kev -ray transition is also likely to be a transition decaying to the @xmath9 state .
the @xmath275 dominance in the angular distribution ( fig . [ angdist ] ( c ) ) implies that the 1961-kev ray is predominantly emitted from a state populated by proton emission from the @xmath40 states of ( @xmath40 ) .
the tism calculation suggests that the @xmath40 states in decay to the @xmath118 state rather than the @xmath116 state . on this basis , the 1961-kev ray is likely to be the @xmath276 transition . as described above
, the @xmath102 distributions of the 2442-kev and 1961-kev rays imply that the @xmath70 state is the upper member and the @xmath56 state is the lower member of the excited doublet . because the 3948-kev @xmath267 state of @xmath266n is mainly @xmath277 in nature ( see sect .
[ sec : discussion ] ) , the upper - doublet spacing is dominantly due to the spin - spin interaction term ( @xmath28 ) and for @xmath278 the the spin anti - parallel state becomes the lower level , as for ( @xmath70 ) .
the contradiction between the implications of the @xmath2-ray yields as a function of pion scattering angle and the underling structure of will be discussed in sect .
[ sec : discussion ] . binding energy ( @xmath83 ) .
the solid lines are plotted for those events accompanying rays in the energy range @xmath279 8 kev for ( a ) , ( b ) , and ( c ) , @xmath280 kev for ( d ) , ( e ) , and ( f ) , and @xmath281 kev for ( g ) , ( f ) and ( i ) .
the reaction angles covered are @xmath282 for ( a ) , ( d ) , and ( g ) , @xmath283 for ( b ) , ( e ) , and ( f ) , @xmath284 for ( c ) , ( f ) , and ( i ) .
the dashed lines show the expected background shapes for the corresponding reaction angles , determined from the events accompanying rays with energies from 2500 kev to 3000 kev after scaling to fit the background region ( @xmath285 mev).,width=321 ] figure [ blam2268 ] shows missing - mass spectra , plotted as a function of @xmath12 binding energy ( @xmath83 ) , for the three @xmath2 rays attributed to in the reaction . the energy and angular ranges
are specified in the caption . by fitting the excess over the background with a gaussian , the peak widths gated on the 2268-kev ray
were found to be @xmath286 , @xmath287 , and @xmath288 mev ( fwhm ) from figs . [ blam2268 ] ( a ) , ( b ) , and ( c ) , respectively .
these widths are significantly wider than the mass resolution .
the natural widths of the states and the separations between the @xmath289 and @xmath40 states in each group of levels ( see fig . [ onrelate ] ) are much smaller than the mass resolution ( see fig .
[ onrelate ] ) @xcite , suggesting that the @xmath9 state in is not produced from a single group of states in . by comparing ( b ) with ( c ) ,
more events are clearly seen at around @xmath290 mev in ( c ) showing that the @xmath9 state of is fed by proton emission decay from several states of produced by @xmath291 transitions .
on the other hand , the widths from figs .
[ blam2268 ] ( d ) , ( e ) , and ( f ) were 17 @xmath11 2 , 17 @xmath11 3 and 17 @xmath11 5 mev ( fwhm ) , respectively .
this suggests that the state which emits 2442 kev ray is dominantly produced from the @xmath91 state and its partners in . the widths from figs .
[ blam2268 ] ( g ) , ( f ) , and ( i ) were 26 @xmath11 8 , 26 @xmath11 8 and 18 @xmath11 9 mev ( fwhm ) , respectively , and no information on the population from states of can be obtained . the lifetime of the @xmath9 state was derived from the shape of the peak by fitting with the simulated shape . since the recoil velocity is affected not by the absolute masses of and but by the q value of the proton decay , it is necessary to know the masses of the and states contributing to the decay . in the simulation , we used the excitation energies of the @xmath115 states and the proton emission threshold as shown in fig .
[ onrelate ] .
we assumed that all the excited states in are produced directly from the excited @xmath80 and @xmath40 states at 17 mev and that the 1961- and 2442-kev rays are from decays to the @xmath9 state with the intensities in eq .
( [ nigamma ] ) . here
, the lifetimes of the upper doublet states were assumed to be much shorter than that of the @xmath9 state and their effect was neglected .
this assumption is justified because the core state of the doublet [ ( @xmath292 ) ] has a short lifetime ( 7 fs ) compared to the scale set by the stopping time . the measured @xmath102 distribution from fig .
[ angdist ] ( a ) was used ( cf .
[ dwia ] ) .
in addition , the reaction angle was selected in the region from @xmath157 to @xmath129 to reduce the contributions from the non - substitutional states evident in the @xmath290 mev region of fig .
[ blam2268 ] ( c ) .
the uncertainty in the excitation energies of states in from previous experiments was found to have a negligible effect .
the systematic error on the lifetime stems mainly from lack of knowlwdge of the decay ratio of the @xmath90 and @xmath91 states to the @xmath9 state ( @xmath293 ps ) .
uncertainty in the response function for the -ray peak at this energy was found to also contribute to the error in the lifetime ( @xmath294 ps ) .
then , the lifetime obtained for the @xmath9 state is @xmath295 ( stat . )
@xmath296 ( syst . ) ps .
this lifetime is 15 times longer than that of the @xmath111 core state and longer than the theoretically predicted value of 0.5 ps @xcite .
the reason for this dramatic increase in the lifetime of the hypernuclear state is discussed in sec .
[ sec : discussion ] .
we also examined the lifetimes of the initial states of the 1961- and 2442-kev rays by fitting the spectrum without doppler - shift correction . in the simulation of the peak shape
, we assumed that the @xmath102 distributions were the same as that of the 2268-kev ray .
then the upper limits ( 68% c.l . ) on the lifetimes were found to be @xmath297 ps for the 1961-kev ray and @xmath246 ps for the 2442-kev ray . in sec .
[ sec : assign15 ] , the 2268-kev @xmath2 ray was taken to correspond to the @xmath298 transition based on the theoretical expectation that this transition should be by far the stronger of the two possible @xmath2 rays de - exciting the @xmath9 level .
the energy of the @xmath299 transition would determine the ground - state doublet spacing but no candidate is observed .
in addition , no low - energy @xmath2-ray peak corresponding to to the ground - state doublet , spin - flip transition ( @xmath300 or @xmath301 ) is observed . in these circumstances , nothing can be said about the ground - state doublet spacing if the @xmath70 state is the upper member of the doublet as the shell - model calculations predict ( see sec .
[ sec : discussion ] ) .
if , on the other hand , the @xmath302 state is the upper level of the doublet , as expected in the the @xmath68-coupling limit @xcite given that has a @xmath8 ground state , an upper limit can be put on the doublet spacing .
the b(m1 ) for the @xmath301 transition can be estimated quite accurately in the weak - coupling limit @xcite as @xmath303 where @xmath304 @xcite and @xmath305 @xcite .
then , @xmath306 ps with @xmath307 in mev . taking into account the -ray efficiency and the competition between electromagnetic and weak decay ( @xmath308 ps for most nuclei @xcite ) for the @xmath302 state ,
the sensitivity of the experiment is high enough to observe a -ray peak with @xmath309 kev even if the @xmath302 state was not produced from directly .
then we can conclude that the ground - state doublet spacing in is @xmath310 as noted previously , no prominent -ray peak corresponding to the @xmath311 transition has been observed .
an upper limit for the @xmath269 ratio between the @xmath311 and the @xmath312 transitions as a function of the ground - state doublet spacing can be obtained from the -ray spectrum .
this is shown in fig .
[ bm1r ] where the shaded region shows the allowed range of the @xmath269 ratio .
the ratios in fig .
[ bm1r ] were obtained from the upper limit of the -ray yield taking into account the relative -ray efficiencies .
the thick line shows the upper limit of the @xmath269 ratio obtained by fitting the spectrum from 2000 to 2370 kev assuming that the @xmath311 transition has the same -ray peak shape as the @xmath312 transition .
the dashed line is obtained by fitting the spectrum with the `` prompt '' tdc cut to reduce the @xmath313 ray at 2211 kev .
the thin line is obtained by fitting the spectrum from 100 kev to 270 kev , assuming that the @xmath300 ray has the original response function because the lifetime of the @xmath314 state ( half the estimate for the @xmath302 level in the previous subsection ) is much longer than the stopping time and with the estimates for the electromagnetic and weak - decay lifetimes taken into account .
ratio between the @xmath311 and @xmath312 transitions as a function of the ground - state doublet spacing , @xmath315 , and the @xmath311 transition energy .
the thick line is the result of a peak search for the @xmath311 transition obtained by fitting the spectrum from 2000 kev to 2370 kev [ fig .
[ nigam ] ( d ) ] .
the dashed line is from the same procedure but with the `` prompt '' tdc cut .
the thin line is the result of a peak search for the @xmath300 transition obtained by fitting the spectrum from 100 kev to 270 kev [ fig .
[ nigam ] ( b)].,width=321 ] the @xmath269 ratio between the @xmath311 and @xmath312 transitions is calculated to be @xmath271 with the level mixing taken into account @xcite using the @xmath10 parameters from the first analysis of the @xmath2-ray data @xcite . as shown in fig .
[ bm1r ] , either the doublet spacing is less than 5 kev or the upper limit on the b(m1 ) ratio is @xmath316 ( the calculation puts the @xmath314 level @xmath317 kev above the @xmath302 level ) .
the interpretation of this result is discussed in detail in sec .
[ sec : discussion ] . basically , small @xmath318 admixtures in the wave functions of the ground - state doublet members introduce strong destructive interferences from the strong @xmath319 m1 core transition into the @xmath2 decays from the @xmath9 level of .
figure [ oxlevel ] shows the experimentally determined level scheme of . as shown in the figure ,
the excitation energies and spin ordering of both ground and 6.7-mev excited doublets have been determined .
this is the first determination of the spin - ordering and the spacing of a @xmath46-shell hypernuclear ground - state doublet . in addition
, this is the first observation of a spin - flip state ( the @xmath5 state ) that is directly produced via the reaction .
figure [ nilevel ] shows the level scheme of together with the corresponding core levels of @xmath266n @xcite . since the spin ordering and spacing of the ground - state doublet were not experimentally determined ,
the excitation energies are given as level spacings from the @xmath302 state .
the existence of the 4710-kev excited state was determined from the coincidence of the 2442-kev ray with the 2268-kev ray . on the other hand , while the existence of the 4229-kev excited state was not unambiguously determined , the 1961-kev ray is likely a transition to the @xmath9 state .
the observed -ray transitions are summarized in table [ ngam ] .
the existence of the 4229-kev excited state was not experimentally determined but the 1961-kev ray is likely a transition to the @xmath9 state .
, width=302 ] [ cols="^,^,^,^,^,^",options="header " , ] from the shell - model calculations , the contribution from each of the @xmath10 parameters to a given eigenenergy can be calculated and by taking differences the coefficient of each parameter entering into an energy spacing can be deduced .
these are given in the first line of table [ tab : lo16 ] for each listed pair of levels in .
it can be seen that the coefficients of the parameters giving the two doublet spacings do not deviate much from those given in eqs .
( [ p12 ] ) and ( [ p32 ] ) describing the simple @xmath68-coupling limit .
this is because the @xmath10 interaction is too weak to cause large mixing between the @xmath6 weak - coupling basis states with an unperturbed core separation energy of 6.176 mev .
in fact , the purity of the dominant basis state is never less than 99.7% for the states of interest in and .
that is , the sum total of admixed @xmath12 and @xmath17 configurations is less than 0.3% .
the downward energy shifts caused by @xmath12-@xmath17 coupling are calculated to be 29 , 61 , 99 , and 7 kev for the @xmath8 , @xmath98 , @xmath93 , and @xmath5 states , respectively , and these lead to the contributions to the energy spacings listed in the fourth column of table [ tab : lo16 ] .
the sum of the contributions from the @xmath10 parameters , the core energy difference , and the @xmath12-@xmath17 coupling gives an accurate , but not perfect ( see below ) representation of the energy - level spacings . assuming that the @xmath12-@xmath17 coupling and the small value of @xmath320 from data [ see sec .
[ sec : intro - previous ] and eq .
( [ eq : param7 ] ) ] are fixed , these expressions can be used to extract values for @xmath28 , @xmath30 and @xmath31 .
since @xmath30 affects only core separations and not the doublet spacings , the two measured doublet spacings in ( see fig . [ oxlevel ] ) give a pair of simultaneous equations for @xmath28 and @xmath31 .
the solution gives @xmath321 mev and @xmath322 mev . here
, @xmath28 is derived from the excited - state doublet spacing in because the main contributor of this doublet is @xmath28 . on the other hand ,
the most important feature of the ground - state doublet splitting is the almost complete cancellation between substantial contributions from @xmath31 and @xmath28 ( aided by @xmath12-@xmath17 coupling . )
there is thus great sensitivity to the value of @xmath31 .
similarly , the difference between the centroid energies of the two doublets ( or the energy separation between the @xmath6 levels ) , with a small correction for @xmath12-@xmath17 coupling , gives @xmath323 mev ; here , @xmath33 simply augments the spin - orbit splitting of the hole states of the core nucleus @xmath48o .
a re - diagonalization of the energy matrices with new parameters leads to small shifts in the energy levels and slightly changed values for the coefficients of the parameters in the expressions that give the energy - level differences .
the parameter set ( in mev ) that fits the measured energy - level spacings in is then given by @xmath324 the second line for each pair of levels in table [ tab : lo16 ] and table [ tab : ln15 ] gives the actual breakdown of the contributions to the energy spacings for this parameter set .
as mentioned above , all parameters except for @xmath29 are derived only from the levels .
nevertheless , @xmath28 and @xmath30 show smaller but similar values to those in eq .
( [ eq : param7 ] ) obtained from the levels .
it is to be noted that a 0.15% admixture of the @xmath6 basis state in reduces the binding energy of the lower @xmath6 state by 10 kev ( @xmath325 ) and vice versa for the upper state .
this is the reason why the sum of the individual energy contributions in table [ tab : lo16 ] ( similarly in table [ tab : ln15 ] ) do not add up to precisely @xmath326 .
in addition to the effective @xmath16 interaction , the structure of is sensitive to details of the two - hole , @xmath13-shell wave functions for @xmath266n .
the tensor interaction in the @xmath13-shell hamiltonian used for the calculation was kept fixed during a fit to 90 @xmath13-shell levels and was chosen to ensure cancellation in the gamow - teller matrix element for @xmath266c(@xmath328 ) decay .
the relevant core wave functions in ls coupling are ( these are @xmath329 wave functions in a supermultiplet basis so that the phases differ from those for simple two - hole wave functions ) @xmath330 while in @xmath68 coupling the states are 85% @xmath331 , 74% @xmath332 , and 93% @xmath331 , respectively .
the gamow - teller matrix element is very closely related to the @xmath333 matrix element for the core @xmath334 m1 transition in @xmath266n , which is given by @xmath335 with the amplitudes to be taken from eq .
( [ eq:14gs ] ) .
the contributions to the energy spacings from a shell - model calculation using the parameters of eq .
( [ eq : param16 ] ) are given in table [ tab : ln15 ] .
starting with the ground state , the energy shifts due to @xmath12-@xmath17 coupling are calculated to be 59 , 15 , 116 , 75 , and 10 kev .
the coefficients that enter the ground - state doublet spacing show a significant shift away from the @xmath68-coupling limit , in which the coefficients are @xmath336 times those in eq .
( [ p12 ] ) .
specifically , the changes from 0.5 to 0.74 for @xmath28 and from @xmath337 to @xmath338 for @xmath31 mean that the higher - spin member of the doublet is predicted to be the ground state in contrast to the usual ordering for @xmath13-shell hypernuclei , including .
this spacing was measured to be @xmath339 kev and thus the calculated spacing , 83 kev , is consistent with the data .
the excitation energy of the @xmath9 state , which gets a significant contribution from @xmath12-@xmath17 coupling , is well reproduced .
note that the contribution of @xmath30 is small and would be zero in the @xmath68 limit of @xmath331 for both core states .
viewed from another perspective , the coefficient of @xmath30 for the @xmath9 state depends strongly on the matrix element connecting the @xmath340 and @xmath341 components of the core wave function in eq .
( [ eq:14gs ] ) .
these components are sensitive to the p - shell interaction used .
for example , the cohen and kurath interactions @xcite give rise to amplitudes of @xmath342 for the @xmath341 component , a smaller coefficient of @xmath30 in the hypernuclear calculation , and an energy for the @xmath9 state that is higher by more than 100 kev .
the level spacings of the members of the excited - state doublet ( @xmath109 ) from the @xmath9 state and the doublet spacing itself are calculated to be 2304 , 1851 kev and 453 kev , respectively .
this upper - doublet spacing is dominantly given by @xmath28 because the core state ( @xmath267 ) is mainly @xmath277 in nature .
this doublet spacing is not experimentally determined but the observed 2442 and 1961-kev rays are likely to be the transitions from the upper doublet members . comparing the measured energy difference , @xmath343 or @xmath344 kev , with the calculated value , 453 kev , the assignments shown in the last column of table [ tab : ln15 ] are strongly indicated . for this assumption ,
the doublet spacing is well reproduced using the parameter set in eq .
[ eq : param16 ] . on the other hand ,
the excitation energies of both members of the excited - state doublet are underestimated by about 100 kev . here
, @xmath30 provides the dominant contribution to the shift of the centroid of the doublet above the unperturbed core energy separation of 3948 kev and there is some dependence on the shell - model wave functions for the core .
for example , the coefficient of @xmath30 for the third and fourth entries in table [ tab : ln15 ] would be @xmath345 in the @xmath68-coupling limit .
the influence of the @xmath10 parameters on the spectra of and has been discussed in the preceding two subsections and some use of the spectra and transition rates from the shell - model calculations has already been made in interpreting the results of this experiment in sec .
[ sec : results ] .
the electromagnetic lifetimes and branching ratios for the observed transitions are largely controlled by the m1 matrix elements for the three core transitions in @xmath48o and @xmath266n . for the upper doublets in both hypernuclei ,
the transition rates are governed by strong m1 transitions .
consequently , the lifetimes are short ( @xmath205 fs ) and the de - excitation @xmath2-ray lines are strongly doppler broadened . for similar reasons ,
the @xmath2-ray branching ratio for the @xmath93 level of differs little from the simple weak - coupling estimate ( 2:1 in favor of the @xmath8 final state ) .
the situation is very different for the decay of the @xmath9 state of because ( 1 ) the measured lifetime of @xmath346 ps is much longer than 0.1ps for the core transition ( sec .
[ sec : lifetime15 ] ) and ( 2 ) the limit of 9% for the @xmath2-ray branch to the @xmath70 member of the ground - state doublet is much less than 33% in the weak - coupling limit ( sec .
[ sec : branch15 ] ) .
the b(m1 ) values are given by @xmath347 where @xmath348 is the reduced matrix element of the m1 operator .
the m1 core transitions of interest , the magnetic moments of @xmath266n , @xmath48n , and @xmath48o , and many other m1 properties of @xmath13-shell nuclei are well described by the present wave functions together with an effective m1 operator @xcite that specifies the six m1 matrix elements in the @xmath13 shell ( equivalently , the isoscalar and isovector @xmath349 factors for the operators @xmath350 , @xmath351 , and @xmath352 ^ 1 $ ] ) . for the @xmath353 and @xmath354 transitions in @xmath266n ,
the values of @xmath348 are calculated to be @xmath355 and 2.957 , respectively @xcite .
the matrix elements contain similar negative orbital contributions while the spin contributions are @xmath356 and large and positive , respectively , according to eq .
( [ eq : gt ] ) .
the corresponding matrix elements @xmath348 for the @xmath9 @xmath195 @xmath357 and @xmath358 transitions in are the same in the weak - coupling limit and equal to @xmath355 [ the @xmath270 branching ratio comes from the @xmath359 factor in eq .
( [ eq : m1 ] ) ] .
the shell - model calculation admixes small @xmath318 components into the final - state wave functions with positive amplitudes of 0.050 and 0.032 for the @xmath70 and @xmath56 final states .
this reduces the matrix element for the @xmath360 transition from @xmath361 to @xmath362 and the @xmath363 matrix element by a smaller amount .
the net effect is to increase the lifetime of the @xmath9 state by about a factor of five over the weak - coupling limit ( to @xmath364 ps ) while the branch to the @xmath357 state is reduced to 18% .
these effects are in the right direction , but not large enough , to explain the experimental results . with the present wave functions , the dominant contributions to the off - diagonal matrix elements between the @xmath365 and @xmath366 basis states are from @xmath30 followed by contributions from @xmath31 that are constructive for the @xmath70 state and destructive for the @xmath56 state @xcite .
the larger off - diagonal matrix element for the @xmath70 case means that the @xmath2-ray branch to the @xmath70 state is suppressed relative to the @xmath56 state ( unless the mixing is large enough to change the sign of the m1 matrix element ) .
if the amplitudes of the @xmath366 admixtures are scaled up ( by a factor of @xmath367 ) to reproduce the measured lifetime @xmath368 ps , the @xmath2-ray branch to the @xmath70 state drops to @xmath369% which is consistent with the limit derived in sec .
[ sec : branch15 ] .
these results are sensitive to the choice of @xmath13-shell wave functions .
it is also worth mentioning that the admixtures of @xmath17 configurations contribute roughly @xmath370 to the m1 matrix elements in either case @xcite and that these are significant compared to the @xmath371 for the @xmath372 transition that reproduces the measured lifetime .
tables [ tab : lo16 ] and [ tab : ln15 ] show that the parameter set in eq .
( [ eq : param16 ] ) , which was derived to reproduce the levels while using a fixed @xmath12-@xmath17 coupling , also reproduces quite well the upper - doublet spacing and the @xmath9 excitation energy in , but predicts a value for the excitation energy of the upper doublet in that is somewhat too small .
the @xmath12-spin - orbit parameter @xmath29 is fixed to be small by the @xmath2-ray data for @xcite .
the theoretical breakdown of the contributions to the measured energy spacings in tables [ tab : lo16 ] and [ tab : ln15 ] then shows that @xmath28 , @xmath30 , and @xmath31 are well determined because @xmath28 and @xmath30 dominate in certain spacings while the ground - state doublet spacings involve a delicate cancellation between significant contributions from both @xmath28 and @xmath31 .
it is this cancellation that fixes @xmath31 to be small and positive based on the measured spacing of the ground - state doublet in , in general agreement with the predictions of @xmath16 models , @xmath373 mev , as described in sec .
[ sec : intro - tensor ] .
the matrix elements @xmath28 and @xmath30 take values that are somewhat smaller than the corresponding quantities in eq .
( [ eq : param7 ] ) that are determined primarily from .
this is what would be expected if the nuclei at the end of the @xmath13 shell were significantly larger than those at the beginning of the shell .
however , it is well known that the charge radii of stable @xmath13-shell nuclei are almost constant throughout the shell @xcite , essentially because the @xmath13-shell nucleons become more deeply bound for the heavier nuclei . in fact , if one calculates the @xmath10 matrix elements from a yng interaction using woods - saxon wave functions , they tend to be slightly larger for @xmath374 than for @xmath375 @xcite .
this is what led to the choice of @xmath376 mev in the first analysis of our data @xcite .
now , the data on the upper doublets in and ( with some reservations ) imply a smaller value of @xmath377 mev . confidence in this value of @xmath28 is increased by the fact that the ground - state doublet spacing in has been established as 264 kev @xcite and is reproduced by this value of @xmath28 @xcite .
it is notable that a value @xmath378 mev was proposed @xcite to account for the non - observation of the ground - state doublet transition in above 100 kev in the first hypernuclear experiment with ge @xmath2-ray detectors @xcite ( a result confirmed in the present experiment using a @xmath57b target @xcite ) .
a @xmath2-ray spectroscopy experiment on and was performed at the bnl - ags d6 beamline employing a high quality 0.93 gev / c @xmath121 beam and the hyperball ge detector array .
the experiment is one of a series aimed at studies of spin - dependent @xmath10 interactions through the precise measurement of @xmath2-ray transitions in @xmath13-shell hypernuclei .
the bound states of both hypernuclei were produced via the reaction .
we succeeded in observing three -ray transitions in and three in . for , we determined the level scheme for the four bound negative - parity states [ two doublets , ( @xmath98 , @xmath8 ) and ( @xmath5 , @xmath93 ) ] from these rays , although another assignment for the @xmath5 state is not excluded . in particular , we determined the excitation energy of the @xmath93 state to be @xmath379 kev and found a small spacing of @xmath380 kev for the ground - state doublet ( @xmath6 , @xmath8 ) with the @xmath8 state being the ground state .
the doublet spacing determines a small but nonzero strength for the @xmath10 tensor interaction and this is the first experiment to give direct information on the @xmath10 tensor interaction .
one of the three rays in is likely to be a transition from the @xmath5 spin - flip state to the one of the ground - state doublet members ( @xmath381 ) and this constitutes the first observation of the direct production of a spin - flip state via the reaction . for , we observed a rather sharp 2268-kev @xmath2 ray and measured the corresponding lifetime via the doppler - shift attenuation method to be @xmath382 ps .
the @xmath2 ray was interpreted as a transition from the @xmath9 level of to the @xmath358 member of the ground - state doublet . because the transition to the @xmath383 member of the ground - state doublet was not observed ( a limit of @xmath384% was put on the @xmath2-ray branch ) , the spacing and the spin ordering of the ground - state doublet were not determined , but we obtained an upper limit on the spacing energy of @xmath385 kev .
we also observed the -ray transitions which can be assigned to those from the upper - doublet states @xmath386 to the @xmath9 state .
we also measured the reaction angle ( @xmath102 ) distributions of the and rays .
analysis of the distributions of rays provides information on the spins of the initial states of decaying to the excited states of via proton emission .
the level spectra obtained for and are consistently explained by the set of values for the @xmath10 interaction parameters @xmath28 , @xmath30 , and @xmath31 in eq .
( [ eq : param16 ] ) .
the determination of @xmath31 was , in fact , the main motivation for the present experiment .
the authors would like to thank the bnl - ags staff for support of the experiment .
this work is supported by the u. s. doe under contract no .
de - ac02 - 98ch10886 , by grants - in - aid nos .
11440070 and 15204014 for scientific research from the ministry of education of japan , and a grant - in - aid no .
1507122 for jsps fellows .
m. juric _ et al_. , nucl .
b * 52 * , 1 ( 1973 ) ; t. cantwell _ et al_. , nucl . phys .
a * 236 * , 445 ( 1974 ) ; p. dluzewski , k. garbowska - pniewska , j. pniewski , t. tymieniecka , p. ciok , and d. h. davis , nucl .
a * 484 * , 520 ( 1988 ) . | the bound - state level structures of the @xmath0o and @xmath1n hypernuclei were studied by @xmath2-ray spectroscopy using a germanium detector array ( hyperball ) via the @xmath3o ( @xmath4 ) reaction .
a level scheme for @xmath0o was determined from the observation of three @xmath2-ray transitions from the doublet of states ( @xmath5,@xmath6 ) at @xmath7 mev to the ground - state doublet ( @xmath6,@xmath8 ) .
the @xmath1n hypernuclei were produced via proton emission from unbound states in @xmath0o .
three @xmath2 -rays were observed and the lifetime of the @xmath9 state in @xmath1n was measured by the doppler shift attenuation method . by comparing the experimental results with shell - model calculations , the spin - dependence of the @xmath10 interaction is discussed . in particular
, the measured @xmath0o ground - state doublet spacing of 26.4 @xmath11 1.6 @xmath11 0.5 kev determines a small but nonzero strength of the @xmath10 tensor interaction . |
in the standard paradigm of cosmic structure formation , it is expected that many wide massive black hole binaries ( mbhbs ) will form following the mergers of two massive galaxies ( e.g. begelman , blandford , & rees 1980 ; volonteri , haardt , & madau 2003 ; mayer 2007 ) .
the binary will subsequently shrink due to stellar or gas dynamical processes and may ultimately coalesce by emitting a burst of gravitational waves .
it was first proposed by ebisuzaki , makino , & okumura ( 1991 ) that the heating of the surrounding stars by a decaying smbh pair would create a low - density core out of a preexisting cuspy ( e.g. @xmath6 ) stellar profile . in a purely stellar background a ` hard ' binary shrinks by capturing the stars that pass close to the holes and ejecting them at much higher velocities , a super - elastic scattering process ( ` gravitational slingshot ' ) that depletes the nuclear region .
observationally , there is evidence in early type galaxies for a systematically different distribution of surface brightness profiles , with faint ellipticals showing steep power - law profiles ( cusps ) , while bright ellipticals have much shallower stellar cores ( e.g. faber 1997 ; ravindranath 2001 ) .
dwarf ellipticals seem to elude this paradigm showing somewhat flat profiles , similarly to bright ones ( graham & guzman 2003 ) . late type spirals tend to show steep central cusps , as in the case of the milky way or andromeda .
detailed n - body simulations have confirmed the cusp - disruption effect of a hardening mbhb ( makino & ebisuzaki 1996 ; quinlan & hernquist 1997 ; milosavljevic & merritt 2001 ) , while semi - analytic modelling in the framework of hierarchical structure formation theories has shown that the cumulative damage done to a stellar cusps by decaying black hole pairs may explain the observed correlation between the ` mass deficit ' ( the mass needed to bring a flat inner density profile to a @xmath7 cusp ) and the mass of the nuclear black hole ( volonteri , madau , & haardt 2003 ) .
this is the third paper of a series aimed at the detailed study of the interaction between mbhbs and their stellar environment . in sesana , haardt , & madau ( 2006,2007a , hereafter paper i and paper ii )
, we analyzed the three - body scatterings between a mbhb and background stars _
unbound _ to the binary .
the assumption of a fixed background breaks down once the binary has ejected most of the stars on intersecting orbits , and the extraction of energy and angular momentum from the binary can continue only if new stars can diffuse into low - angular momentum orbits ( refilling the binary s phase - space `` loss cone '' ) , or via gas processes . in galaxies with inner cores or shallow
cusps , only a small fraction of the loss cone is confined within the sphere of influence of the binary , and the approximation of a background of unbound stars is reasonable .
a similar argument holds also in the case of a galaxy with a steep cusp hosting a nearly equal - mass binary ( @xmath8 ) .
the radius of influence of such a pair , @xmath9 where @xmath10 is the stellar velocity dispersion , is of the order of the binary hardening radius , @xmath11 , and only few low - angular momentum stars have orbits with semi - major axis @xmath12 . this is not true for unequal - mass binaries , where @xmath13 , and almost all interacting stars are bound to @xmath14 . in this paper
we develop a formalism for studying the dynamics of mbhbs embedded in gravitationally - bound stellar cusps . the plan is as follows . in 2 we describe our suite of three - body scattering experiments between the black hole pair and ambient bound stars .
our numerical results are used in 3 and 4 to construct a `` hybrid model '' of binary dynamics and investigate the orbital decay and shrinking of mbhbs in time - evolving stellar cusps .
the properties of the ejected hvss are discussed in 5 . the massive black hole ( sgr a@xmath4 ) in the galactic center and the stars around it offer a unique opportunity to study stellar dynamics in the extreme environment around a relativistic potential .
the scattering of stars bound to sgr a@xmath4 by an inspiraling intermediate - mass black hole ( imbh ) is treated in
finally , we present a brief summary in 7 .
consider a mbh of mass @xmath14 surrounded by a stellar cusp ( with density profile @xmath15 , @xmath16 ) , interacting with a secondary intruder " hole of mass @xmath17 .
as the binary separation decays , the effectiveness of dynamical friction slowly declines because distant stars perturb the binary s center of mass but not its semi - major axis .
the bound pair then loses orbital energy by capturing ambient stars and ejecting them at much higher velocities , a three - body scattering process known as the ` gravitational slingshot ' . for unequal - mass binaries ( mass ratio @xmath18 ) , @xmath19 , implying that the stellar mass inside @xmath20 is @xmath21 . in this case , the contribution to the potential energy given by the stellar distribution during a binary - star interaction can be ignored , and the problem can be tackled by means of three - body scattering experiments .
following the guidelines fully described in paper i , we performed numerical experiments to study the interaction of a mbhb with a star of mass @xmath22 ( @xmath23 ) bound to @xmath14 .
note that , for extreme mass ratios @xmath24 , the encounter is essentially a two - body scattering as the star and the secondary hole move in the static potential of the primary .
the integration of the three - body encounter equations is greatly simplified by setting the center of mass of the binary at rest in the origin of the coordinate system .
the binary orbits counterclockwise in the @xmath25 plane , and the apoastron of @xmath26 is located along the positive @xmath27-axis .
stars are drawn from a spherical isotropic distribution bound to @xmath14 .
the problem is completely defined by ten variables : 1 .
the binary mass ratio @xmath28 ; 2 . the binary eccentricity @xmath29 ; 3 . the stellar mass @xmath22 ; 4 . the initial distance between the primary hole and the star , @xmath30 ; 5 . the specific energy of the stellar orbit around @xmath14 , @xmath31 , or equivalently the semi - major axis of the stellar orbit , @xmath32 ; 6 . the specific angular momentum of the stellar orbit @xmath33 , where @xmath34 is the stellar velocity relative to @xmath14
four angles : @xmath35 and @xmath36 describing the latitude and longitude of the star , @xmath37 defining the orientation of @xmath38 ( the stellar velocity component normal to @xmath39 ) i.e. selecting the orbital plane of the star , and @xmath40 the initial binary phase . compared to the unbound scattering problem ( which is defined by nine variables , see paper i ) , there is here the extra variable @xmath41 , and @xmath42 and @xmath43 replace the asymptotic initial speed and impact parameter of the incoming field star .
a sketch of the experiment set - up is given in figure [ fig : incond ] .
in each numerical integration , the binary mass @xmath44 and separation @xmath45 are set equal to unity , and the stellar mass to @xmath46 .
scattering events are simulated in sets of @xmath47 trials for fixed values of @xmath48 and @xmath29 .
we sample five values of the binary eccentricity , @xmath49 , and five values of the binary mass ratio , @xmath50 , for a grand total of 25 models .
the distribution of stars bound to @xmath14 is modeled as follows . the stellar semi - major axis @xmath32 is randomly sampled from fifty logarithmic bins spanning the range @xmath51 .
the angular momentum is sampled in the interval @xmath52 $ ] according to an equal probability distribution in @xmath53 , where @xmath54 is the specific angular momentum of a circular orbit of radius @xmath32 .
a population of stars with such distribution in @xmath53 has mean eccentricity @xmath55 , corresponding to @xmath56 .
this condition defines an isotropic stellar distribution ( e.g. quinlan , hernquist , & sigurdsson 1995 ) .
the quantities @xmath42 and @xmath43 define the shape of the stellar orbit .
we sample the initial value of @xmath41 from the distribution @xmath57 , @xmath58 which is proportional to the the fraction of time spent by the star at distance between @xmath41 and @xmath59 from @xmath14 .
the probability @xmath60 is defined in the range @xmath61 , where @xmath62 the angles @xmath35 and @xmath36 are randomly generated to reproduce a uniform density distribution over a spherical surface centered on @xmath14 , while the orientation angle @xmath37 is chosen from a uniform distribution in the range @xmath63 $ ] .
we start numerical integration with @xmath26 at its apoastron ( @xmath64 ) , and have checked that the chosen initial phase of the binary does not affect our results .
the orbit of the pair is in the @xmath65 plane with the center of mass at coordinates @xmath66 .
the nine coupled , second order , differential equations of the three - body problem are integrated using the most recent version of the subroutine dopri5 , based on an explicit runge - kutta method of order 4(5 ) due to dormand & prince ( 1978 ) . a complete description of the integrator can be found in hairer , norsett , & wanner ( 1993 ) .
the integration is stopped if any of these events occurs : * the star leaves the sphere of radius @xmath67^{1/4}a_0 $ ] with positive total energy , where @xmath68 is the reduced mass of the binary and @xmath69 its initial semi - major axis . at @xmath70
the force induced by the quadrupole moment of the binary is 10 orders of magnitude smaller than the total force acting on the star at a distance @xmath69 ; * the physical integration time exceeds 1 gyr ; * the integration reaches @xmath71 time steps .
this typically corresponds to the complete integration of @xmath72 binary orbits .
examples of integrated stellar orbits are given in figures [ fig : boundej ] and [ fig : boundweak ] .
figure [ fig : boundej ] shows a strong interaction : the star has a close encounter with @xmath26 and leaves the binary sphere of influence with positive energy .
depending on the value of @xmath32 we have a variable number of numerically `` abandoned events '' , i.e. , bound triple systems that do not readily produce an ejection .
figure [ fig : boundweak ] shows one of such events : the periastron of the star is much larger than the binary orbital separation @xmath45 , and the star is not perturbed by the binary quadrupole field , except for the precession of its periastron .
we have performed a number of tests to check the sensitivity of our results on numerics .
because of the intrinsically chaotic nature of the three - body problem , the properties of the ejected stars are meaningful only for a statistically - significant sample .
the integration of the full three - body problem allows us to directly control the conservation of total energy and angular momentum .
the code adjusts the integration stepsize to keep the fractional error per step in position and velocity , @xmath73 , below @xmath74 .
this allows a total energy conservation accuracy of @xmath75 in a single orbit integration , i.e. for @xmath76 the energy of the star is conserved at level of one part in a thousand during a single orbit .
we varied @xmath73 between @xmath77 and @xmath78 and @xmath79 between @xmath80 and @xmath81 , and found no significant differences in the statistics of the ejected population .
we have also checked that the mean energy and angular momentum exchanges scale linearly with @xmath22 .
a set of longer orbit integrations was performed to test that the number of abandoned events does not depend on the finite numbers of timesteps allowed , again finding no systematic decrease in such a number .
each logarithmic bin in @xmath32 was sampled by @xmath82 stars .
we calculated the mean energy and angular momentum exchange between the mbhb and the stars , the final velocity and angular distribution of the scattered stars , the fraction @xmath83 of stars that are ejected in the interaction , and the ejection timescales .
the fraction @xmath83 is plotted versus @xmath84 in figure [ fig4.1 ] for different values of the binary parameters @xmath48 and @xmath29 .
eccentric binaries can eject stars that are initially very tightly bound to @xmath14 , i.e. with @xmath32 as small as @xmath85 . for @xmath86 ,
the expelled fraction declines significantly , regardless of binary eccentricity .
it is also evident that the ejection process is more efficient for large values of @xmath48 .
figure [ fig4.3 ] shows the mean fractional eccentricity change of the pair after each scattering , @xmath87 , as a function of @xmath84 .
this quantity is found to scale linearly with @xmath79 .
stars with @xmath88 typically tend to reduce the binary eccentricity , while stars with @xmath89 work in the opposite direction .
an eccentric binary spends most of its period near its apocenter , so in the case @xmath89 the probability of a close star - binary encounter ( and subsequent star ejection ) is maximal at binary apocenter . in the instantaneous interaction
the binary velocity decreases . as close to the apocenter the binary velocity is nearly tangential , the binary
is forced on a more radial orbit . on the contrary ,
a star with @xmath88 `` feels '' the secondary hole @xmath26 when this approaches the primary @xmath14 at a distance @xmath90 . at that point ,
the interaction with the star is unlikely to occur close to the pericenter of @xmath26 ( being the time spent by @xmath26 at the pericenter very small ) , and typically also extracts a large radial component from the velocity of the secondary black hole , hence causing circularization .
note that in this case the absolute value of @xmath91 is larger , according to the @xmath92 dependence in equation ( [ deltaebound ] ) ( see the different positive and negative @xmath93-axis scales in fig .
[ fig4.3 ] ) .
we will return on this point in section 3 , giving a simple , physically motivated , mathematical derivation of this qualitative argument .
figure [ fig4.4 ] depicts the fractional number of stars ejected in the time interval @xmath94 and @xmath95 as a function of the ejection timescale @xmath94 .
the latter is measured in units of the binary orbital period at separation @xmath96 , @xmath97 , and is defined as the time elapsed from the start of numerical integration to the moment the interacting star reaches , with positive energy , a distance @xmath98 from the binary center - of - mass . for large values of @xmath48 , it typically takes the mbhb just a few orbits to expel the star , while lowering @xmath48 makes the distribution of slingshot timescales broader and flatter .
the ejection rate remains approximately constant , or decreases slowly , for @xmath99 , and drops dramatically afterwards .
the binary eccentricity plays no role in all the tested cases .
figure [ fig6bis ] shows the mean ejection timescale ( i.e. , the integral of @xmath100 in @xmath101 ) as a function of the star eccentricity @xmath102 , for different values of @xmath48 and @xmath29 . in first approximation
, one can assume that the cusp remain isotropic during the binary orbital evolution .
we checked that our results are practically the same considering @xmath103 rather than @xmath104 as the outer boundary in the sampled @xmath32 distribution .
this is because the binary stars interaction is dominated , in terms of energy and angular momentum exchange , by those stars whose @xmath32 is @xmath105 . from our scattering experiments
, we can finally derive the bivariate distribution functions @xmath106 and @xmath107 , along the same line described in paper ii .
for a given @xmath32 , we record the number of of stars with ejection speed in the interval @xmath108 , leaving the binary with latitude ( longitude ) at infinity in the interval @xmath109 ( @xmath110 ) .
the differential distributions are normalized as follows : @xmath111 the subscript 1 " is meant to indicate that the scattering experiments are performed for a binary at separation @xmath112 .
to study the impact of the gravitational slingshot on the dynamical evolution of a mbhb , we have developed a self - consistent hybrid model , in which numerical results of scattering experiments in a fixed stellar background are coupled to an analytical formulation of loss - cone depletion .
this technique allows us to simultaneously follow the orbital decay of the pair as well as the time evolution of the stellar cusp .
the hybrid model is similar , in spirit , to the case extensively discussed in paper ii of an unbound stellar population interacting with a black hole pair .
scattering experiments are performed for a binary at fixed orbital separation @xmath45 , and the results scaled to any value of @xmath45 .
then , by specifying the pace at which @xmath45 evolves , we can solve for the binary orbital decay with time .
mathematical details , however , are different from the scheme developed in paper ii . in the unbound case , the pair interacts with a stellar population approaching its sphere of influence at a given rate , and the typical three - body interaction is a fast " process . for stars bound to @xmath14 instead , the interacting stellar population is in place from the beginning .
the temporal evolution of the system is then determined by the ejection timescale rather than by supply rate .
it is convenient to describe the stellar cusp using @xmath113 , the differential number of stars orbiting @xmath14 with semi - major axis in the range @xmath114 . for an isotropic stellar distribution @xmath115 where the stellar density profile is @xmath116 and @xmath117
is a fudge factor that depends on the cusp slope , @xmath118 for @xmath119 ( ivanov , polnarev , & saha 2005 ) .
two differential equations determine the rate of change of orbital separation and eccentricity : @xmath120 and @xmath121 we start numerical integration at time @xmath122 , orbital separation @xmath69 , and binary eccentricity @xmath123 . the terms @xmath124 , @xmath125 , and @xmath126 are measured from our scattering experiments .
the scaling of the eccentricity change with binary parameters @xmath127 and @xmath92 can be derived from ( quinlan 1996 ) @xmath128 where @xmath129 and @xmath130 are the total angular momentum and energy of the binary , respectively . from @xmath131 ( in a typical encounter , on average ,
the star gets a kick @xmath132 ) , and @xmath133 , assuming @xmath134 , we have @xmath135 this confirms the qualitative argument developed in section 2.3 and accounts for the features shown in figure 5 : the mean @xmath124 scales with the inverse of the mass ratio , and , from @xmath136 , results negative for @xmath137 ( promoting circularization ) , and positive for larger @xmath84 ( hence increasing the binary eccentricity ) . the binary energy change per scattering , @xmath125 ,
is related to the change of stellar energy @xmath138 by @xmath139 , where we have neglected the energy change of the binary center of mass ( which is a factor @xmath79 smaller than @xmath125 ) .
the quantity @xmath138 can be written as @xmath140 where @xmath141 is the circular velocity of the binary .
the numerical factor @xmath142 depends upon the ratio @xmath84 , and @xmath29 , and is derived from our numerical experiments .
the term @xmath126 quantifies the number of stars orbiting @xmath14 within @xmath32 and @xmath143 that are ejected from the system in the time interval between @xmath144 and @xmath145 .
this term depends on @xmath32 , @xmath45 , @xmath29 and @xmath144 , and is determined as follows . from scattering experiments
we derive the distribution @xmath146 describing the probability that a star at @xmath147 becomes unbound from a binary with eccentricity @xmath29 in the time interval @xmath148 .
note that the distribution of ejection timescales plotted in figure [ fig4.4 ] is simply the probability function @xmath149 averaged over @xmath150 , @xmath151 when @xmath152 ( the exact value depending on @xmath48 and @xmath29 ) most stars remain bound to @xmath14 , and @xmath153 ( see fig . [ fig4.1 ] ) . as @xmath154
all interacting stars are instead expelled .
we can see from figure [ fig4.4 ] that , in the ejection regime , the distribution @xmath149 is nearly independent on @xmath150 and has the same functional form ( as a function of @xmath94 ) of @xmath155 . from @xmath149
we can then compute @xmath126 as a function of @xmath144 if we set @xmath156 , where the term @xmath157 accounts for the change in the time units of @xmath149 as @xmath45 changes .
though not formally correct , the scheme catches the basic physics of the interaction in a time - dependent fashion .
suppose now that in a small time interval @xmath158 after the beginning of the interaction the binary remains at constant separation @xmath69 corresponding to @xmath159 .
the number of stars ( with semi - major axis in the interval @xmath160 ) ejected in such time interval is @xmath161 where for simplicity we have omitted the dependence of @xmath149 on the eccentricity .
after @xmath158 and for an interval @xmath162 , the binary settles to a new separation @xmath163 , corresponding to @xmath164 .
the number of stars ejected in the time interval @xmath165 is @xmath166 the integral at the denominator of the right - hand side renormalizes the distribution @xmath149 in the time interval @xmath167 $ ] so that the correct number of stars is involved .
iterating we have @xmath168 where @xmath169 . in differential
form : @xmath170^{3/2 } , \label{eq : d2ndiff}\ ] ] where @xmath171 . simultaneous numerical integration of the tree coupled equations ( [ eq : aev ] ) , ( [ eq : eev ] ) and ( [ eq : d2ndiff ] ) self - consistently solves for the evolution of the binary and the depletion of the stellar cusp . the integration is performed using the subroutine dopri5 . the time
step is adapted in order to keep the fractional error per step @xmath172 , a value much lower than the error associated to the employed linear interpolation of @xmath124 and @xmath125 .
the bivariate distributions @xmath106 and @xmath107 derived from our suite of scattering experiments ( see 2 ) can be convolved with the ejection rate @xmath173 to compute the final velocity distributions @xmath174 and @xmath175 .
the procedure is similar to that described in the appendix of paper ii .
as the binary shrinks to separation @xmath176 , the normalized distribution of ejection velocities for stars with semi - major axis in the interval @xmath114 is @xmath177 where the ejection speed @xmath178 was shifted by the factor @xmath179 to account for the increase in the circular velocity of the binary @xmath180 as the pair shrinks .
the prefactor @xmath179 normalizes the distribution according to equation ( [ eq : norm ] ) .
the kick velocity function of the expelled population as a whole can then be written as @xmath181 where @xmath182 is the final binary separation , @xmath183 and @xmath184 is given by equation ( [ eq : aev ] ) .
the distribution @xmath106 is evaluated for different values of the eccentricity @xmath29 .
we can account then for the evolution of binary eccentricity by interpolating the @xmath185 distribution on a grid of @xmath29-values as the orbit decays .
the above procedure returns the velocity distribution in units of @xmath186 .
the calculation of @xmath175 can be performed following the same lines . finally , integrating the rate @xmath173 over the entire evolutionary history of the binary yields the quantity @xmath187 , which can be compared to @xmath188 to study the depletion of the stellar cusp by three - body interactions .
the monte carlo technique developed to address this point is described in
the functions @xmath187 and @xmath174 can be used to check energy conservation , by simply equating the total energy gained by the stars to the increment in the binding energy of the black hole pair , i.e. @xmath189 where @xmath190 , @xmath191 , and @xmath192 .
we have checked that energy is conserved to better than @xmath193 .
it is important to remark at this point that equations ( [ eq : aev ] ) and ( [ eq : eev ] ) are independent of the absolute value of @xmath194 and @xmath195 .
indeed , in equation ( [ eq : aev ] ) , the total number of interacting stars is @xmath196 , and the energy exchange term is @xmath197 .
the scaling factor @xmath198 cancels out , so that the orbital evolution depends only on @xmath48 and @xmath29 .
the same consideration holds for equation ( [ eq : eev ] ) : as @xmath199 while the number of interacting stars @xmath196 , the dependence on @xmath194 and @xmath195 cancels out .
for comparison with our previous results on the scattering of unbound stars ( paper ii ) , it is convenient to introduce the stellar velocity dispersion @xmath10 , and model the _ outer _ stellar component as a singular isothermal sphere ( sis ) with density profile @xmath200 we assume that this profile extends inward up to the characteristic radius @xmath201 within which the total stellar mass is @xmath202 ( merritt 2004 ) . matching the inner cusp described by equation ( [ eq : density ] ) to the outer sis at @xmath201 yields @xmath203
we also assume that stellar - binary interactions start at separation @xmath69 where the enclosed mass stellar is @xmath204 , yielding @xmath205.\ ] ] this assumption is motivated by recent n - body simulations of the hardening of unequal mbhb in stellar cusps .
matsubayashi , makino , & ebisuzaki ( 2007 ) found that dynamical friction is efficient in driving orbital decay only as long as the stellar mass inside the binary semi - major axis is @xmath206 . beyond this point ,
the evolution of the pair is driven by three - body interactions with individual stars .
our results on binary orbital decay are summarized in table 1 .
a mbhb can shrink by factors ranging from 6 to 18 depending on @xmath48 , @xmath123 , and @xmath207 .
the decay factor @xmath208 grows with @xmath123 , as more eccentric binaries can expel stars that are more tightly bound ( see fig .
[ fig4.1 ] ) .
the importance of the initial eccentricity @xmath123 is modest for low values of @xmath207 , since in shallow cusps the number of tightly bound stars is small anyway .
the factor @xmath208 is also a weakly increasing function of @xmath48 , as for higher @xmath48 the radius of influence of the binary is larger ( in terms of @xmath84 , see fig .
[ fig4.1 ] ) .
the dependence of the decay factor on @xmath207 simply reflects the fact that , for steeper stellar cusps , the mean binding energy of the stars is larger . for @xmath209 and less than @xmath210 for @xmath211 ,
see table 1 of paper ii ) to binary hardening for very unequal - mass pairs . in the bound case ,
the effect of returning stars is likely to be even smaller .
consider for example the case of an isothermal stellar profile : the impact of secondary slingshots is proportional to the number of stars that can interact with the mbhb more than once , i.e. to the size of the loss - cone after the first interaction .
for bound stars this is at least a factor of @xmath212 smaller than in the unbound case : so even for @xmath209 , returning stars would cause at most a @xmath213 increase in binary hardening . ] [ tab2 ] @xmath48 & @xmath123 & @xmath208 & @xmath214 & @xmath208 & @xmath214 & @xmath208 & @xmath214 + & 0.1 & 9.63 & 0.608 & 9.89 & 0.350 & 11.38 & 0.179 + @xmath215 & 0.5 & 9.99 & 0.972 & 11.55 & 0.907 & 15.77 & 0.753 + & 0.9 & 10.06 & 0.998 & 11.80 & 0.992 & 17.73 & 0.969 + & 0.1 & 8.06 & 0.691 & 8.35 & 0.532 & 10.19 & 0.408 + @xmath216 & 0.5 & 8.26 & 0.959 & 9.35 & 0.862 & 12.35 & 0.710 + & 0.9 & 8.27 & 0.996 & 9.64 & 0.988 & 14.03 & 0.958 + & 0.1 & 6.99 & 0.755 & 7.75 & 0.650 & 9.39 & 0.542 + @xmath217 & 0.5 & 6.90 & 0.922 & 7.81 & 0.828 & 10.14 & 0.717 + & 0.9 & 6.89 & 0.996 & 7.81 & 0.974 & 11.00 & 0.937 + & 0.1 & 6.49 & 0.906 & 7.28 & 0.805 & 9.30 & 0.688 + @xmath218 & 0.5 & 6.39 & 0.971 & 7.25 & 0.914 & 9.92 & 0.818 + & 0.9 & 6.38 & 0.962 & 7.19 & 0.986 & 10.09 & 0.955 + & 0.1 & 6.12 & 0.881 & 6.91 & 0.814 & 8.94 & 0.724 + @xmath219 & 0.5 & 5.95 & 0.919 & 6.91 & 0.869 & 9.09 & 0.797 + & 0.9 & 5.94 & 0.977 & 6.92 & 0.953 & 9.41 & 0.900 + binary eccentricity also grows as a function of @xmath48 and @xmath207 . a shallow cusp increases the relative importance of stars with large @xmath32 , and
the eccentricity growth is then larger .
moreover , binaries with large @xmath48 are more effective in ejecting stars with @xmath220 . as these stars act to circularize the binary orbit , nearly equal - mass binaries decay following less eccentric orbits .
figure [ fig5.1 ] shows examples of binary evolution as a function of time .
note the different scale of the time axis for the three mass ratios : as shown in figure [ fig4.4 ] , it takes nearly equal - mass pairs a smaller number of binary orbits to unbind the stellar cusp .
results for the @xmath221 cusp can also be compared to those obtained in paper ii for the case of a binary interacting with a population of unbound stars .
( in paper ii , shrinking factors were normalized to the hardening radius " @xmath222 , where , by definition , @xmath223 for @xmath221 . ) binaries with @xmath224 and @xmath225 , for example , shrink by the factors @xmath226 according to table 1 , compared to the corresponding @xmath227 for the unbound case , i.e. , very unequal - mass binaries can only decay by extracting the cusp binding energy .
the evolution of the eccentricity in the bound and unbound cases is compared in figure [ fig5.1bis ] , for two different initial values of @xmath123 .
compared to the scattering of unbound stars , the binary eccentricity grows to larger values .
this can be understood by the following argument . in the bound case treated here
, there is a great deal of stars able to extract angular momentum from the binary right from the start .
then , for large @xmath45 , @xmath29 shows a steep increase .
when the orbit shrinks , however , stars with @xmath88 ( increasing circularization rather than eccentricity , see fig .
[ fig4.3 ] ) become more and more relevant for the binary evolution as stars with @xmath89 get progressively depleted , and , as result , @xmath29 increases at a reduced pace . in the unbound case , the binary interacts with the same distribution of stars independently upon its separation . as
, in general , significant changes of @xmath29 happen whenever a strong star - binary interaction occurs , the eccentricity grows more rapidly for small @xmath45 because stars are slower ( in terms of binary circular velocity ) , and hence more easily captured in
quasi " bound orbits before ejection . as a matter of fact ,
the eccentricity growth rate @xmath228 ( see paper i ) increases for @xmath229 . in this separation regime ,
most of the stars interacting with the binary are captured in temporarily bound orbits , and would experience a dynamical interaction similar to that of bound stars .
consider now , as in paper ii , the separation at which a mbhb can coalesce in less than 1 gyr because of the emission of gravitational waves ( gws ) : @xmath230 where @xmath231 and , to 4th order in @xmath29 , @xmath232 ( peters 1964 ) .
assuming as before that three - body interactions begin operating at separation @xmath69 , and combining equations ( [ eq : a0 ] ) and ( [ eq : agwbound ] ) , we get @xmath233 if @xmath234 , then a mbhb would coalesce in less than 1 gyr after interaction with a bound stellar cusp .
as shown in the coalescence diagram " of figure [ fig21 ] , eccentric , massive , very unequal - mass binaries embedded in steep stellar cusps are favored to reach coalescence .
the steeper the cusp , the wider the portion of the @xmath235 plane where coalescence can be reached within 1 gyr .
the orbital decay of the pair occurs following the ejection of a mass @xmath236 ( depending on cusp slope and binary eccentricity ) and results in the progressive erosion of the stellar cusp .
eccentric binaries can shrink rapidly by scattering at pericenter a fewer number of deeply bound stars . more mass
is expelled in the case of shallower cusps , because of the larger number of stars surrounding the binary just outside @xmath69 .
our hybrid approach allows us to compute the binary driven evolution of the stellar density profile , once isotropy is assumed . at this stage
, we do not perform a self consistent treatment of the evolution of the anisotropy of the stellar cusp .
consistent integration of the orbital decay yields @xmath187 , and the number of stars that remain bound to @xmath14 as a function of @xmath32 is simply @xmath237 we assume that the stellar cusp remains isotropic , i.e. @xmath238 . the @xmath32 domain is then subdivided in intervals @xmath239 , and for each interval a number of bound stars @xmath240 is generated with angular momentum distribution @xmath241 .
we then compute the probability of finding a star at distance between @xmath70 and @xmath242 from @xmath14 ( this is proportional to the time spent at such distance along its orbit ) , and reconstruct the stellar density profile @xmath243 .
figure [ fig7 ] shows the profile before and after binary shrinking .
the cusp is eroded between @xmath244 and @xmath245 , depending on @xmath123 . for @xmath246 ,
a sis is flattened to @xmath2 , while a @xmath247 cusp becomes @xmath248 .
such results are independent of the mass ratio @xmath48 .
as three - body interactions and stellar ejections tend to circularize the orbits of ambient stars , one may wonder about the validity of the assumption that the cusp remains isotropic .
we checked that even assuming that all stars still bound to the binary at the end of the process were set on circular orbits ( i.e. , maximum tangential anisotropy ) , then the cusp slope would be flattened at most by a factor @xmath249 ( for @xmath221 , and @xmath250 ) with respect to the isotropic case .
differences in the cusp slope @xmath251 are common for initially shallower cusps and/or larger mass ratios . we conclude that , depending on @xmath207 and on the final anisotropy of the bound stars , the slingshot mechanism creates a central core as flat as @xmath252 .
the distribution of stellar ejection velocities in units of the binary circular velocity at the initial separation @xmath69 , @xmath253 , is shown in figure [ fig8 ] .
the distribution can not be fit by a single or a broken power - law , as in the case of the kick velocities imparted at fixed binary given separation , as the derivative of the distribution is a monotonic decreasing function of the ejection speed . while the peak of the distribution shifts toward smaller @xmath254 values as @xmath48 decreases ( @xmath255 ) , the high velocity tail is independent of @xmath48 .
for initially eccentric binaries , the velocity of @xmath26 at periastron is @xmath256 and a significant number of hvss can then be generated .
note that the speed distribution is only weakly dependent on @xmath207 once @xmath178 is expressed in units of @xmath253 .
integration of the kick velocity distribution gives the ejected mass @xmath257 .
this quantity is plotted in figure [ fig9 ] as a function of @xmath48 for different lower velocity thresholds . as discussed above ,
the total number of ejected stars is approximately @xmath258 and is weakly dependent on @xmath123 and @xmath207 .
stellar ejection occurs in a burst lasting from few tens to several thousands binary orbital periods ( see fig .
[ fig10 ] , note the different scale of the time axis in the four panels ) , i.e. from @xmath259 to @xmath260 yr . the highest velocity stars are delayed with respect to the bulk of ejections , as large kicks require close binary separations .
the ejection rate peaks at earlier times in the case of more eccentric binaries , and is larger for steeper cusps .
it must be pointed out that the quantity @xmath257 plotted in figure [ fig9 ] accounts only for the mass ejected by energetic three body interactions occurring in the final stage of the binary evolution . during the entire binary evolution ,
stars in galaxy nuclei are also displaced by the heating associated with dynamical friction the cumulative effect of many weak encounters with distant stars . according to merritt ( 2006 ) , the total mass displaced by dynamical friction depends only weakly on binary mass ratio ( for @xmath48 in the range @xmath261 ) , and has little or no effect on the innermost stellar cusp for @xmath262 ( see fig . 5 in merritt 2006 ) .
while the total mass deficit accumulated during binary evolution appears to scale with the mass of the binary ( i.e. , to @xmath14 for small @xmath48 ) , the mass ejected after the energetic encounters discussed in this paper , and therefore the number of expected hvss , scale with the mass @xmath26 of the intruder .
the number of stars ejected above a given velocity threshold ( in units of @xmath253 ) is also a weak function of @xmath123 and @xmath207 .
the angular properties of hvss show several peculiar features , qualitatively similar to those discussed in paper i for the scattering of an unbound stellar population .
the ejected stars are flattened in the binary orbital plane , and , in the case of eccentric binaries , are grouped into a `` broad jet '' aligned to the velocity of @xmath26 at periastron .
both anisotropies are more pronounced in stars undergoing a stronger interaction and receiving larger kicks .
the anisotropy of the hvs population can be quantified by computing as a function of binary separation the angles @xmath263 and @xmath264 , where @xmath265 is the latitude of the star ( the angle between the velocity vector @xmath178 at infinity and the binary orbital plane ) , and @xmath266 is its longitude ( the angle between the projection onto the binary orbital plane of the velocity vector at infinity and the @xmath27-axis ) . for a spherically - symmetric distribution , @xmath267 , and @xmath268 .
we find @xmath269 for all hvss . as a general trend
, stars ejected above a given speed tend to become more isotropic as the binary shrinks , confirming the analytical result of levin ( 2006 ) .
this effect was already seen in three - body scattering experiments of unbound stars ( see paper i for a detailed discussion ) , and is related to the fact that , as the pair decays , its circular velocity grows and even stars experiencing relatively weak encounters can attain large kick velocities . in terms of longitude ,
hvss are ejected almost isotropically , though an azimuthal anisotropy become apparent with increasing kick velocities .
the high - velocity tail of the distribution is formed by stars expelled after a close encounter with @xmath26 near its periastron , in a broad jet with @xmath270 .
as the binary potential is non - keplerian , such broad jet will precess during binary evolution on a timescale that depends upon @xmath48 and @xmath207 ( levin 2006 ) .
most of the results presented in the previous sections are scale invariant , i.e. they are independent on the absolute value of @xmath194 and on the chosen normalization of the stellar cusp .
the only underlying assumption is that the interaction between the mbhb and ambient stars begins when the stellar mass inside the binary orbit is equal to @xmath271 .
we are interested in scaling our results to the scattering of stars bound to sgr a@xmath4 , the massive black hole in the galactic center , by an inspiraling companion of intermediate mass ( yu & tremaine 2002 ; sesana , haardt & madau 2007b ) .
let us first express equations ( [ eq : sis ] ) , ( [ eq : rbrake ] ) , and ( [ eq : a0 ] ) in physical units : @xmath272 @xmath273 @xmath274 where @xmath275 is measured in units of @xmath276 . from @xmath69 , the time unit of our experiments is then @xmath277 the stellar density profile around the galactic center can be described as a double power - law , with outer slope @xmath278 and inner slope @xmath279 ( genzel et al .
2003 ; schodel et al .
the massive black hole sgr a@xmath4 weighs @xmath280 ( schodel et al .
2002 ; ghez et al .
2005 ) . using @xmath281 , @xmath282 and @xmath283 , from equations ( [ eq : rzero ] ) and ( [ eq : rhozero ] ) we obtain @xmath284pc , and @xmath285 pc@xmath286 , in good agreement with the most recent observations ( merritt 2006 ; schodel et al .
2007 ) . for the relevant values of @xmath48 and @xmath207 , the typical timescale for orbital decay ranges between @xmath287 and @xmath288 yrs ( see eq .
[ eq : p_0 ] ) . in the following ,
we will consider two different mass ratios for the putative mbhb at the galactic center , @xmath289 and @xmath250 , corresponding to an inspiraling imbh of mass @xmath290 and @xmath291 , respectively , as well as two different slopes for the initial inner stellar cusp , @xmath283 and @xmath292 .
we will also study the impact of binary initial eccentricity .
the parameters of the different models considered are listed in table 2 . [
cols="^,^,^,^,^,^,^ " , ] the dynamical evolution of a putative imbh - sgr a@xmath4 binary is displayed in figure [ fig20 ] for @xmath224 .
the two upper panels show the time - changing semi - major axis @xmath45 and eccentricity @xmath29 .
the former shrinks to @xmath293 pc in 2 - 15 myrs , depending on @xmath48 and @xmath207 .
lighter imbhs reach smaller separations on a longer timescale .
our results for the case @xmath292 can be directly compared to the numerical simulations of baumgardt et al .
( 2006 ) and matsubayashi et al .
( 2007 ) , and are found to be in excellent agreement .
the eccentricity increases rapidly in all cases to values @xmath294 .
we have checked that , when @xmath295 , the binary eccentricity can reach values as large as @xmath296 .
this is again in agreement with the results of baumgardt et al .
( 2006 ) and matsubayashi et al .
( 2007 ) . in the figure we have marked with an horizontal dotted line the separation at which the binary can coalesce in 1 gyr because of gw emission ( eq . [ eq : agwbound ] ) . only for a cusp as steep as @xmath292 does
the pair actually reach such separation .
one should note that , while in our hybrid model the eccentricity evolves smoothly , in a realistic situation it will undergo discontinuous jumps " triggered by rare close encounters , which could induce extreme eccentricities and accelerate coalescence .
the two lower panels of figure [ fig20 ] show the resulting stellar density profiles after binary erosion .
cusps are flattened to @xmath2 in the central few@xmath297 pc .
the numerical simulations of baumgardt et al .
( 2006 ) and matsubayashi et al .
( 2007 ) produce somewhat shallower slopes , a discrepancy that may be associated with our assumption of an isotropic stellar cusp after the interaction . the heating of the cusp results in the creation of a population of hvss .
this is of particular interest since the discovery of the first hvs in the milky way ( brown et al .
2005 ) , and an imbh inspiral onto sgr a@xmath4 is regarded as a possible source of hyper velocity ejections ( though the tidal break up of close binaries by sgr a@xmath4 seems to be supported by observations and statistical studies , see e.g. perets 2007 , sesana et al .
in figure [ fig18 ] we plot the stellar ejection rates as a function of time for different models and for different velocities at the radius of influence of sgr a@xmath4 . modelling the milky way potential as the sum of a luminous component ( miyamoto & nagai 1975 ) and a dark matter halo ( widrow & dubinski 2005 )
, we find an escape velocity from the milky way of @xmath298 km / s at @xmath299 .
this velocity threshold translates , in such a gravitational potential , into about @xmath300 10 kpc away from sgr a@xmath4 .
stars with @xmath301 at @xmath299 do not leave the bulge , while stars with @xmath302 can reach @xmath303 kpc away from the galactic center .
stars with @xmath304 and @xmath305 are not bound to the milky way and , at a reference distance of @xmath306 kpc , have still velocities of @xmath307 and @xmath308 , respectively .
figure [ fig18 ] shows that the ejection occurs in a relatively short burst lasting a few myrs , with higher velocity stars being produced at increasingly later times . at peak ,
the ejection rate of hvss with @xmath304 varies between @xmath309 and @xmath310 yr@xmath311 , depending on @xmath48 and @xmath207 . the rate is larger in steeper cusps , as stars are more centrally concentrated and are scattered when the binary separation is smaller and orbital velocity larger .
the speed distribution of hvss depends on the details of the model . in the range @xmath312 ( the velocity range of the hvss observed by brown et al .
2006 ) , and at a galactocentric distance of 55 kpc ( the average distance of the observed hvss ) , the distribution can be approximated by a power - law @xmath313 , almost independent of @xmath48 , @xmath123 , and @xmath207 .
assuming @xmath314 , we find a predicted number of hvss with @xmath315 at @xmath299 @xmath316 for @xmath250 and @xmath317 .
this number roughly doubles for @xmath289 and is fairly independent on @xmath123
. our results are consistent with baumgardt et al .
( 2006 ) , who estimate @xmath318 for @xmath319 , and @xmath320 for @xmath321 .
a peak of ejection occurs after @xmath322 myr .
levin ( 2006 ) finds comparable numbers of stars expelled , and a similarly peaked ejection rate .
we have performed , for the first time , scattering experiments between a mbhb and stars drawn from a cusp bound to the primary hole .
we have studied the dynamics of the pair and its orbital decay by three - body interactions , the impact of the gravitational slingshot on the stellar density profile , the properties of the ejected stellar population , and have scaled our results to the case of sgr a@xmath4 .
our results can be quickly summarized as follows : 1 .
the extraction of the cusp binding energy causes the binary to shrink by a larger factor compared to the scattering of unbound stars .
the effect is more noticeable in the case of small mass ratios @xmath48 .
the binary orbital eccentricity increases much more rapidly compared to the unbound case .
the eccentricity growth is more pronounced in small mass - ratio binaries , and for shallower stellar cusps .
3 . the combined effects of enhanced orbital decay and eccentricity growth lead very unequal - mass binaries to the gravitational wave coalescence phase .
the detailed fate of the pair depends on the absolute value of its mass .
more massive binaries decay faster .
the stellar cusp is eroded , and the total mass removed by strong three
body encounters is 2-to-4 times the mass of the secondary hole . while the mass deficit caused by dynamical friction in a merger event scales with the binary mass ( and involves mostly distant stars ) , the mass of stars ejected from the inner cusp by highly energetic interactions scales with @xmath26 .
ejection occurs in a burst " lasting from few tenths to several thousands binary orbital periods , depending upon @xmath48 .
5 . scaled to the scattering of stars bound to sgr a@xmath4 by an inspiralling imbh , our results imply the formation of a core of 0.1 pc in 1 - 10 myrs , as well as the ejection of 500 - 2500 hvss moving with speeds sufficient to escape the gravitational field of the milky way . in sesana
( 2007b ) we have used the brown et al .
( 2007 ) sample of unbound and bound hvss together with numerical simulations of the propagation of hvss in the milky way halo to constrain this ejection mechanisms , and shown that it appears to produce a spectrum of ejection velocities that is too flat compared to the observations .
future astrometric ( as , e.g. , _ gaia _ ) and deep wide - field ( as , e.g. , _ lsst _ ) surveys should unambiguously identify the ejection mechanism of hvss , and probe the milky way potential on scales as large as @xmath323 kpc ( gnedin et al . 2005 ; yu & madau 2007 ) . | we develop a formalism for studying the dynamics of massive black hole binaries embedded in gravitationally - bound stellar cusps , and study the binary orbital decay by three - body interactions , the impact of stellar slingshots on the density profile of the inner cusp , and the properties of the ejected hypervelocity stars ( hvss ) .
we find that the scattering of bound stars shrinks the binary orbit and increases its eccentricity more effectively than that of unbound ambient stars .
binaries with initial eccentricities @xmath0 and/or
unequal - mass companions ( @xmath1 ) can decay by three - body interactions to the gravitational wave emission regime in less than a hubble time . the stellar cusp is significantly eroded , and cores as shallow as @xmath2 may develop from a pre - existing singular isothermal density profile .
a population of hvss is ejected in the host galaxy halo , with a total mass @xmath3 .
we scale our results to the scattering of stars bound to sgr a@xmath4 , the massive black hole in the galactic center , by an inspiraling companion of intermediate mass . depending on binary mass ratio , eccentricity , and initial slope of the stellar cusp , a core of radius @xmath5 pc typically forms in 1 - 10 myr . on this timescale
about 500 - 2500 hvss are expelled with speeds sufficiently large to escape the gravitational potential of the milky way . #
1to 0pt#1 |
galactic rotation curves are one of the most important evidences favoring the dark matter scenario in astrophysics .
actually dark matter was proposed by zwicky @xcite in order to accommodate observed velocities in galaxies and galaxy clusters . since then
others explanations was proposed as modifications of newtonian gravity @xcite or modifications in general relativity @xcite .
all these approaches have in common , from the modeling point of view , just one element : non - linearity .
so , is natural asks the question if non - linear effects of some sort are responsible for the discrepancy observed in galaxies and galaxy cluster dynamics .
recently , in the context of general relativity , some authors have investigated this point @xcite . here
, we will propose an alternative paradigm where newtonian gravity is maintained on its solid basements but a non - linear model of a galaxy is build .
we solve the non - linear equations in two simplified cases and calculate the resulting galaxy rotation curves , showing qualitatively that flat curves can be obtained in a given region of the parameter space .
we focus on spiral galaxies described as a fluid with mass distribution in the phase space , @xmath0 which is related to the ordinary mass distribution by@xmath1 where @xmath2 the mass for a typical star .
the dynamics is governed by the poisson equation coming from the second newton law,@xmath3 but restricted to the mass conservation,@xmath4 using a hamiltonian description , @xmath5 * * , * * @xmath6 , we obtain,@xmath7 in general , this is a set of integro - differential non - linear coupled equations .
let us restrict the model only to the disc of the galaxy in the static regime .
so , using the axial symmetry of this system , the mass conservation equation become:@xmath8 this equation can be solved using the method of characteristics . assuming a constant angular velocity , this is equivalent to a set of ordinary differential equations,@xmath9 whose uniparametric family of solutions is@xmath10 experimental data shows that the mass distribution is gaussian in the observed velocities , therefore is natural to choose the mass distribution @xmath11 to be a boltzmann distribution in the total energy:@xmath12 \right)\ ] ] the integration over the velocities can be done , in order to obtain the functional dependence of the mass distribution over the gravitational potential@xmath13 it means that the poisson equation now is a non - linear partial differential equation , given by@xmath14
let us to take an over simplified case , assuming that the field vary only with the height to the plane of the disc,@xmath15 using an integration factor @xmath16 and the boundary conditions @xmath17 this equation can be directly integrated,@xmath18 the density profile in this case is@xmath19 so far , it is a reasonable model , since the disc predicted is thin .
let us to take another over simplified model assuming variation only in the radial direction,@xmath20 multiplying both sides by @xmath21 and integrating , we find@xmath22 where @xmath23 is the radius of the galaxy core , and @xmath24 .
performing a second integration , we arrive in a volterra second order integral equation:@xmath25 to solve it , we apply the piccard s method of successive approximations:@xmath26@xmath27 therefore , the first order iterative solution is@xmath28
assuming a virial balance , @xmath29 , the galaxy rotation curve in this model is @xmath30@xmath31 in our illustrative example , @xmath32 , we have the behavior illustrated in fig .
[ f1 ] .
and @xmath33 is the distance to center in adimensional units .
@xmath34 is the parameter dictated by the boundary condition at the core .
[ f1],width=491 ] so , if @xmath35,@xmath36 which is constant .
a non - linear newtonian model for the galaxy disk was constructed by introducing a solution of the mass conservation on the phase space .
s. fay , _ astron .
astrophys . _
* 413 * ( 2004 ) 799 ; s. behar and m. carmeli , _ int . j. mod .
* 39 * ( 2000 ) 1397 ; s. capozziello _
et al . , phys .
lett . _ * a326 * ( 2004 ) 292 ; d. n. vollick , _ gen .
_ * 34 * ( 2002 ) 471 ; k. ichiki _ et al .
_ * d66 * ( 2002 ) 023514 ; r. r. cuzinatto , c. a. m. de melo , l. g. medeiros and p. j. pompeia , _ eur .
j. _ * c53 * ( 2008 ) 98 . | we develop a model for spiral galaxies based on a nonlinear realization of the newtonian dynamics starting from the momentum and mass conservations in the phase space .
the radial solution exhibits a rotation curve in qualitative accordance with the observational data . * keywords : * galactic rotation curves ; dark matter ; mathematical modeling . *
pacs numbers : * 95.35.+d , 02.90.+p |
sonoluminescence refers to the phenomenon of light emission during acoustic radiation of a liquid and is associated with cavitation bubbles present in the liquid .
the most controllable and promising experimental data were obtained for single bubble sonoluminescence ( sbsl ) : picosecond uv radiation of a single bubble pulsating in the field of the sound wave @xcite .
though the case of a single pulsating bubble is , of course , much simpler than cavitation in general , it turns out that the sbsl is a very complex phenomenon which still remains not completely understood .
there is a vast literature devoted to sbsl that was recently reviewed in ref.@xcite .
interest in this phenomenon is stimulated , on the one hand , by the fact that in relatively simple and controllable experiments extraordinary conditions ( ultra - high pressures , temperatures , and ultra - short light flashes ) are realized at the final stage of the bubble collapse ; on the other hand , it looks promising to use the sbsl to construct a source of ultraviolet ultra - short flashes that is much cheaper than lasers . the highly involved hydrodynamics of bubble collapse has been addressed in many papers that are reviewed in ref .
one of the most important problems to explain here is the very existence of stable pulsation of bubbles .
recently a theory has been developed @xcite which explains this regime in terms of the dissolved gas diffusion and chemical reactions in the gas within the bubble .
it has been shown that an accumulation of noble gas in the bubble takes place during pulsation of the bubble and that in a stable situation the gas in the bubble consists , almost entirely , of noble gas and , of course , of water vapor .
these theoretical conclusions were supported by the experimental data of ref.@xcite .
another key problem in sbsl is the mechanism of the light emission .
many mechanisms have been proposed , criticized , and reviewed to explain uv flashes radiated by the collapsing bubble @xcite .
the most popular of the suggested mechanisms are the adiabatic heating of the bubble gas @xcite , shock wave - bremsstrahlung model @xcite , and , recently proposed proton - tunneling radiation as a result of a phase transition in water at ultra - high pressure @xcite .
a general feature of these mechanisms is that extraordinary conditions are needed which can only be realized , if at all , at a very small bubble radius when the density of the bubble gas is close to the water density and both the hydrodynamics of the bubble and the properties of the bubble content and the neighboring liquid are not known from experiment but inferred from numerical simulations . in this paper
we show that the main characteristic features of sbsl can be explained even without making any assumptions about the extraordinary conditions .
we are far from saying that these extraordinary conditions are not present in experiments .
what we mean is that there is another mechanism of the sbsl which occurs in water near the bubble surface but not in the bubble gas as has been assumed in the most popular models of sbsl @xcite .
extraordinary conditions are not necessary for the operation of this mechanism even though it may well be the main cause of uv radiation .
the mechanism under consideration is based on the idea put forward in ref.@xcite that sbsl occurs because of electric breakdown in strong electric fields arising near the bubble surface as a result of flexoelectric effect , that is the effect of polarization of water because of gradients of pressure @xcite . here
we present a scenario and estimations that show that within the hypothesis the main features of sbsl can be explained using relatively moderate parameters , e.g. , temperatures ( 5@xmath010)10@xmath1k for the bubble gas @xcite , and the natural ( expected ) value of the flexoelectric coefficient of water @xcite .
let us mention that ref.@xcite has left many questions to answer .
the origin of the optical radiation of the bubble remaines unexplained .
the problem is that in the visible region pure water has very low levels of absorbtion and radiation due to interband optical transition @xcite .
moreover the breakdown scenario is far from being clear .
the mechanism of the breakdown in water ( see , e.g.,@xcite ) involves lucky electrons whose acceleration in the electric field leads to development of an avalanche . at ambient temperature
the concentration of free electrons in water is quite negligible ( the band gap is about 6.5 ev @xcite ) and it is impossible to find a
lucky electron in a small volume near the bubble surface , i.e. in the region of high electric field , during the short time of the existence of this field .
the spectacular synchronization of the emission pulses @xcite that in the case of breakdown appears , at least at first sight , to be hardly compatible with the fact that we are dealing with a probabilistic situation .
in addition , the reference to the penning effect to explain the role of the noble gases is not convincing because the ionization energy of the metastable state is more than the band gap of water as distinct from the case of breakdown in gases where the penning effect is pronounced . in this paper
we consider in more detail possible processes associated with the electric field arising from the flexoelectric effect near a bubble exhibiting sbsl .
a common difficulty with theories of sbsl is that little is known about the parameters of gas in the bubble when the bubble radius is near its minimum value .
consensus exists about one point : it is far from being a gas , because the lowest limit of the radius is governed by the van der waals repulsion and the minimum volume of the bubble is close to the van der waals hard core volume .
the equations of state used for these conditions are not reliable at present .
this is a challenging and fundamental problem but it is beyond the scope of this paper .
what is now possible is to make order - of - magnitude estimations .
that is why we shall first explain what values of relevant parameters are necessary for our scenario to be operative and then discuss whether our assumptions about these parameters are realistic .
we include the value of the flexoelectric coefficient of water among these parameters as well .
this coefficient has not been measured , unfortunately , and we use its estimated natural value but , from the other hand , the knowledge of its precise value would hardly be of decisive importance because the bubble gas parameters are not known precisely .
let us describe shortly the proposed scenario .
at certain short time interval , @xmath2ns , when the bubble radius is near its minimum , the acceleration of the radius and , therefore , the pressure gradient , assume gigantic values @xcite and the sign of the pressure gradient is such as to create a strong ( depolarizing ) electric field directed to the centre of the bubble in a thin water layer , @xmath3 , near the bubble surface , because of the flexoelctric effect .
what happens then is , in effect , screening of this depolarizing field by free electrons .
indeed , during the same time interval the temperature of the gas sharply increases up to at least several thousand k , which owing to thermal excitation of the bubble gas and water in a thin layer near the bubble surface makes the concentration of free electrons appreciable .
the free electrons are accelerated by the strong flexoelectric field up to energies sufficient to generate additional free electrons as a result of the electric breakdown of water .
the hot electrons also collide with noble gas atoms dissolved in water and transfer them to higher energy excited states .
optical transitions from these states produce light radiation with a broad spectrum whose shape is determined mainly by the energy distribution of the hot electrons .
the latter has the form of a maxwell distribution with an effective electron temperature which can be very high .
this is due to the fact that the noble gas atoms have a huge number of excited states with very high probabilities of optical transitions between them ( see , e.g. , @xcite ) .
since the radiation occurs in the region of very strong and inhomogeneous electric fields the observed radiation spectrum is featureless .
the characteristic time of the electric breakdown is much less than the characteristic time for the last stage of the collapse @xmath4 , i.e. the polarization can continue to change after the first breakdown and the electric field can reach the breakdown threshold value more than once . as a result , several breakdowns can take place during the time interval @xmath4 . during each breakdown
the noble gas atoms are excited .
an important specific feature of the noble gas atoms is the existence of long - living ( metastable ) excited states with a life time of up to several milliseconds ( see , e.g. , @xcite ) .
therefore , once excited the noble gas atom can remain in the metastable state for the entire time interval @xmath2ns , and , possibly , for many periods of the acoustic wave ( the period is about @xmath5s ) .
that means that in the stationary state the longer is the pulse the more is its intensity .
the paper is organized as follows . in sec.2
we discuss the flexoelectric effect in water in more detail than in @xcite and consider the sign of the flexoelectric field , which is crucial for the proposed scenario , at different stages of the bubble collapse .
the scenario of sbsl is presented in sec.3 where the values of relevant parameters necessary for realizing the scenario are estimated . in sec.4
we discuss the energy distribution functions of hot electrons and the breakdown conditions in strong flexoelectric fields as well as the spectrum of sbsl . in conclusions
we summarize the results of our theory , discuss their relation to the experimental data , speculate about some possible modifications of the scenario and point out some problems to solve .
we have already mentioned that the proposed mechanism of sbsl is based on the flexoelectric effect , namely the appearance of a polarization ( electric field ) due to gradient of density ( pressure)@xcite . this effect is not widely known and it was discussed in @xcite but shortly . besides , the flexoelectric coeficients are still not determined experimentally@xcite .
that is why we think it is worthwhile considering in more detail the appearance of electric field due to flexoelectric effect .
the flexoelectric effect is a particular case of a more general phenomenon : appearance of electrical polarization , @xmath6as a result of gradient of some scalar quantity , e.g. , temperature , concentration of a component , mass density , @xmath7 .
this effect should take place in any substance @xcite . as usual ,
the material equation can be written in several forms .
in particular , considering polarization as a function of dilatation , @xmath8 , and the electric field , @xmath9 , one has for an isotropic medium @xmath10 where @xmath11 is the electric susceptibility and @xmath12 is one of the flexoelectric coefficients
. we shall be interested in the case of spherical symmetry and the absence of free charges . in this case
@xmath13 then from eq.[polar ] it follows that @xmath14 since @xmath15 , where @xmath16 is the ( excess ) pressure , @xmath17 is the compressibility . to estimate the flexoelectric coefficients it is convenient to consider separately two main mechanisms of polarization : charge displacement and dipole ordering .
we shall see that in both cases the coefficient of proportionality between polarization and gradient of pressure , @xmath18 , has the same characteristic value .
the charge displacement polarization is realized , e.g. , in ionic crystals .
when a gradient of dilatation takes place there are less and more compressed regions in each unit cell .
the ions of larger radius tend to displace to the less compressed region while ions of less radius displace into the opposite direction .
since cations and anions usually have different radii their displacements produce polarization .
the coefficient @xmath12 can be estimated as follows @xcite .
the maximum ( atomic ) gradient of dilatation is equal to @xmath19 where @xmath20 is the interatomic distance.this gradient has to produce atomic
polarization @xmath21 , where @xmath22 is the electron charge , when the electric field is absent ( compensated by some charges ) , i.e. @xmath23 . then from eq.[el ] it follows that @xmath24 where we have taken into account that @xmath25 for ionic crystals .
we shall argue that the same estimation is aplicable for materials with dipole ordering .
the dipole ordering polarization in the absence of an external electric field but under a gradient of pressure arises because of geometric asymmetry of the molecular dipoles .
for example , the geometric shape of water molecule is highly asymmetric : the negatively charged end is much more compact than the positive one : the negative charge is concentrated in the oxygen ion ( o@xmath26 ) while the positive charge is shared by two hydrogen ions ( h@xmath27 ) located fairly far from each other . under a pressure gradient
a water molecule tries to orient itself in such a way that the oxygen ion would be located in the region of higher density while the hydrogens would be located in the region of lesser density .
in other words , a gradient of pressure leads to orientation of the water molecular dipoles .
it is well known that polarization due to dipole ordering is much more effective than polarization due to charge displacements .
this is reflected in the fact that dielectric constants of dipolar materials are , normally , much larger than of non - polar materials .
for the same reason it is natural to estimate @xmath28 .
then from eq.[el ] it follows that for water @xmath29 where we have taken into accout that for water @xmath30 @xmath31 . for what follows
the sign of the flexoelectric coefficients is of importance .
as it has been mentioned above the negative tip of the water molecule , the oxygen ion , tends to be located in the region of higher pressure .
this means that the polarization vector is directed opposite to the gradient of pressure , i.e. , the flexoelectric coefficient @xmath12 is negative and the coefficient @xmath18 is positive .
let us mention that we rather underestimated the flexoelectric coefficient of water than overestimated .
indeed , what we call @xmath32 is replaced , in fact , by @xmath33 defining an atomic
flexoelectric coefficient that is of the same order of magnitude for all substances , i.e. it does not take into account specific features of the substance in question .
it is natural to expect that a substance consisting of molecules whose electric asymmetry ( existence of a dipolar moment ) is accompanied by a pronounced geometric asymmetry ( as it is for water ) will exhibit a stronger flexoelectric effect than that estimated above .
however , while this coefficient is not measured ( reliable calculations seem much more problematic than measurements ) we shall assume the value given by eq.[flex ] .
on the contrary , the conclusion reached about the sign of the coefficient @xmath18 , which is crucial for the mechanism discussed in this paper , seems much more definite .
above we have neglected the conductivity of water .
this may seem questionable because the debye radius of the electric field screening in water is comparable with the characteristic size of the strong field region ( @xmath34 m ) .
however , this neglection is justified because , as a rule , the dielectric relaxation time , @xmath35 is much greater than @xmath36ns , which is the maximal characteristic time of the polarization change in our case . increasing the ionic conductivity of water by adding , for example , nacl
, one can , according to our estimations , decrease @xmath37 down to 0.1ns .
in such an electrolyte the mechanism of sbsl discussed in this paper might be less effective .
the dynamics of bubble cavitation have been studied in many papers @xcite . here
we will only discuss the short time interval when the bubble radius @xmath38 is close to its minimum value @xmath39 ( fig.1a ) .
fig.1a reflects the most essential features of fig.4 of ref.@xcite and fig.12 of ref.@xcite where experimental results were presented . within this interval the velocity of the bubble surface @xmath40 reaches its maximum and reaches zero at the point @xmath41 ( fig.1b ) and the acceleration @xmath42 reverses its sign and can achieve huge values ( fig.1c ) . during the negative acceleration period ( @xmath43 ,
1c ) the gradient of pressure @xmath44 is directed from the bubble center to its periphery , thus according to eq.[flex ] the flexoelectric depolarizing field has the same direction ( fig.1d ) .
that means that free electrons that could be generated in the gas or in water near the bubble surface can not be accelerated . during the positive acceleration period ( @xmath45 ,
1e ) the situation is opposite : the arising flexoelectric field is directed to the bubble center and the generated electrons can be accelerated by the field and produce a breakdown ( fig .
according to experimental data @xcite for a bright sbsl the characteristic value of the acceleration is about ( 10@xmath46 @xmath47 m / s@xmath48 during an interval , @xmath49 1ns , so the pressure gradient @xmath44 can reach ( 10@xmath5010@xmath51n / m@xmath52 . note that the gradient of pressure decays with distance from the bubble surface . for purpose of estimations one can consider water as incompressible liquid where @xmath53 where @xmath54 is the gradient of pressure at the bubble surface .
the same estimation for the gradient of pressure can be obtained if we take into account that the pressure is about ( 10@xmath5510@xmath56 ) n / m@xmath48 at the final stage of the bubble collapse and the extension of the high pressure region is about 1@xmath57 m @xcite .
it follows from eqs.[flex ] and [ el ] that for @xmath58=(10@xmath5910@xmath51n / m@xmath1 the value of the flexoelectric field , @xmath60 can reach ( 10@xmath61 10@xmath62v / m . according to eq.[flex ] the electric field decays just as the gradient of pressure , i.e. , the flexoelectric field in water is given by the formula @xmath63 where @xmath64 is the field at the bubble surface , i.e. the radial extension of the strong field region is about 1@xmath57 m .
the field @xmath65 can essentially exceed the threshold field of dynamic electric breakdown of water @xmath66 which is about ( 1@xmath03)@xmath6710@xmath56v / m@xcite .
it does not mean , however , that electric breakdown will occur : a lucky electron which capable of provoking an avalanche is needed .
a similar situation takes place with the laser breakdown @xcite . at the same time
the conduction electron concentration , @xmath68 in water at room temperature is practically zero ( @xmath69@xmath70 ) : water could be considered as a wide gap amorphous semiconductor with @xmath71ev @xcite .
the breakdown starts only the moment when the strong flexoelectric field is directed to the bubble center ( fig .
1f ) and conduction electrons appear because of the sharp increase of temperature near the bubble surface .
the latter takes place when the bubble radius @xmath72 which is close to @xmath73(0.5@xmath01)@xmath57 . at this moment
the flexoelectric field @xmath74 may be much more greater than @xmath66 so the coefficient of avalanche multiplication of an electron can be practically infinite and in this case just several electrons are enough to provoke the breakdown and to screen the field .
let us show that even relatively moderate temperatures near the bubble surface ( which seem to be generally acceptable for the regimes without the shock waves ) are quite sufficient to provide a sufficient number of conduction electrons to ensure the breakdown .
as in ref .
@xcite we assume that the gas has the temperature of 7000k .
that means that the water layer with a thickness of the thermal penetration length , @xmath75 , has a temperature of about 3000k . for @xmath75
one has : @xmath76 where @xmath77 , the thermal conductivity of water is about 0.4 j / m s k ; @xmath78 , the specific heat , is about 4@xmath79 j / m@xmath1 k ; @xmath7 , the density , is 10@xmath1kg / m@xmath1 . bearing in mind that the characteristic time for the last stage of the collapse is @xmath801ns @xcite
, one finds that the thermal penetration length is @xmath81 m .
following @xcite we will consider water as an amorphous semiconductor with a band gap @xmath71ev and an effective density of states @xmath82 @xmath83 for @xmath84k .
the equilibrium concentration of conduction electrons in water at @xmath85k is @xmath86@xmath70 .
taking into account that the volume of the hot surface layer is approximately @xmath87cm@xmath1 we find that the equilibrium number of conduction electrons in the layer is @xmath89 it is important also to find the time needed for the equilibrium concentration of electrons to be established .
the thermal ionization frequency ( see , e.g. , @xcite ) is @xmath90 where @xmath91 is the cross - section of the free carrier recombination , @xmath92 is the thermal velocity of conduction electrons in water . taking into account that a reasonable value of @xmath91 is about 10@xmath93@xmath94 and @xmath95cm / s we find from eq.[thermo ] that for @xmath96k the transient time to the equilibrium @xmath97s . from that
follows that during the characteristic time for the last stage of the bubble collapse @xmath801ns the number of conduction electrons in the layer , @xmath98 exceeds @xmath99 electrons , which is quite enough to provoke breakdown .
note that a similar number of free electrons can be generated by thermal ionization of the gas .
indeed , the assumed gas temperature ( 7000k ) is about two times larger and the ionization energy both of the water vapor and ar , kr , xe is about 12@xmath016ev , i.e. also about two times larger than @xmath100 .
authors of many works ( see @xcite and the references therein ) state that much higher temperatures can be reached because of shock waves forming in the bubble gas during the collapse . in principle
, one can imagine a situation where high temperatures are not reached ( and free electrons are not generated ) before the shock waves are formed . in this case
the shock wave , when it explodes , will start the electric breakdown . as a result of breakdown
the depolarizing flexoelectric field becomes screened .
the total transmitted charge in the process of the screening is @xmath101 where we have taken into account that in our case @xmath102 , i.e. , @xmath103 . assuming @xmath104 m and @xmath105 v / m we find that the maximum total number of transmitted electrons is @xmath106 .
note that for considered electric fields the time of the screening ( breakdown ) , @xmath107 , is determined by the time it takes for the conduction electrons to cross the region of the strong electric field whose size is @xmath108 .
thus @xmath109 where @xmath110 is the electron drift velocity which in strong electric fields saturates at some value @xmath111 m/s ( see sec.4 ) . assuming @xmath112
m we find @xmath113ps .
note that the value of @xmath107 is less than the observed sbsl pulse width @xcite .
the whole period of positive acceleration , @xmath4 , ( see sec.3 ) is about two orders of magnitude larger than @xmath114 therefore , after the breakdown is finished and the depolarization field is screened the polarization continues to change , because of change in acceleration .
the electric field arises once more and can exceed the breakdown threshold . as a result
a new breakdown will take place .
such a situation can be repeated several times .
effectively , it manifests itself in an increase of the pulse duration which may achieve a fraction of a nanosecond .
therefore , within our scenario the greater is the pulse duration the greater is its energy .
analogous interrelation is observed in experiment . @xcite .
we have already mentioned that the electric fields under consideration are very strong , inhomogeneous and change over time
. however , the characteristic space and time scales of the field change are correspondingly @xmath112 m and @xmath2ns and they are much more than the relaxation length @xmath115 and relaxation time @xmath116 of the hot electron energy which are respectively about @xmath117 and @xmath118s .
therefore one can consider the local electron energy distribution function using well known results for homogeneous static electric fields @xcite .
the considered electric fields are superstrong , i.e. the following condition is valid : @xmath119 where @xmath120 is the characteristic energy of local oscillations in water which is practically equal to the energy of optical phonons in ice , @xmath121mev @xcite . using @xmath122
@xmath117 one sees that the condition [ cond1 ] is satisfied for fields @xmath123v / m .
the condition [ cond1 ] means that in these fields an electron acquires , on average , an energy @xmath124 which is much more than @xmath120 . since in the process of the acceleration
an electron generates many phonons the electron energy distribution is nearly isotropic in the momentum space .
specifically , in this case the energy distribution of hot conduction electrons is approximated with high accuracy , up to energies of electrical breakdown @xmath125 , by the maxwell function @xcite : @xmath126 with an effective electron temperature @xmath127 which is determined by the balance equations for energy and momentum of the hot electrons @xmath128
@xmath129 where @xmath130 is the effective thermal velocity of electrons . since for the maxwell distribution @xmath131 from eqs.[bal1 ]
, [ bal2 ] and [ vel - temp ] it follows that @xmath132 and @xmath133 from eqs [ vel - temp ] and [ vel - dr ] as well from eq .
[ cond1 ] one sees that @xmath134 this is precisely the condition of validity for the thermalization of the hot electrons and use of the maxwell distribution .
conduction electrons in water form polarons which are conventionally called hydrated electrons @xcite .
however , in the strong electric field the polaron states decay and the current carriers are the usual conduction electrons . for purpose of estimation
we assume that their effective mass @xmath135 is close to that of free electrons @xmath136 .
then from eq.[vel - dr ] it follows that @xmath111 m/s for @xmath137mev .
it has already been mentioned that the electric fields under consideration are @xmath138v / m .
it follows from eq.[temp - el ] that for such fields @xmath139ev , i.e. , @xmath140k for @xmath141 .
one sees that the electron temperatures can be several orders higher than the gas temperature in the bubble .
recall that for our scenario ( see sec.3 ) it sufficient that the gas temperature be about 7000k .
now we will discuss the role of noble gases dissolved in water .
an important feature of an noble gas atom is the existence of metastable states .
the life time of the metastable states can reach several milliseconds when the noble gas atoms are impurities in solids ( see , e.g. , @xcite ) . since the nearest order in water is essentially the same as in solids it would be natural to expect that the life - time of the metastable states for the noble gas atoms in water is not less than @xmath142ns .
hot electrons not only generate new conduction electrons ( the breakdown avalanche ) but also excite the noble gas atoms into metastable states .
below we will estimate from the experimental data that the characteristic value of @xmath143ev .
however , a considerable number of electrons may have energies about 10ev and can excite the noble gas atoms .
note that the energies necessary for excitation to the metastable states , @xmath144 , for xe , kr , ar , ne , he are about 10ev and increase monotonically from xe to he . in other words , it is much easier to excite xe , kr and ar than ne and he .
this could be the reason for the increase in sbsl intensity in the series he - xe , which has been observed experimentally @xcite .
as we have mentioned above , the life - time of the metastable states of noble gas atoms in water is expected to be fairly long .
thus , once excited a noble gas atom can remain in a metastable state for the entire time interval of positive acceleration and multiple breakdowns , @xmath2ns .
the major part of electrons has the energies @xmath145they collide with these metastable atoms and transfer them to higher excited states .
the radiation transitions between high - energy excited states of noble gas atoms govern the sbsl spectrum .
note that the life - time of the metastable states can exceed the period of the acoustic wave . in this case
there will be an effect of accumulation of the noble gas atoms in metastable states resulting in the gradual build up of the sbsl power during several periods of the acoustic wave .
a specific feature of a noble gas is an abundance of excited states with energies higher than those of metastable states and with high probabilities of the radiation transitions between them .
that is why the optical spectra of the noble gas atoms contain many lines @xcite .
because of the stark effect in the strong electric field these lines are split . besides , in the active region the strong electric field changes at least several times and the observed radiation spectrum is a superposition of the spectra of the atoms in different strong electric fields . in other words
one can consider the density of the atomic excited states as a constant . the probability of a hot electron having an energy @xmath146 is given by eq.[distr ] and at every collision it transfers the noble gas atom in a metastable state to a state with energy @xmath147 , the reference point of energy being the energy of the metastable state ( for our estimations we consider only one metastable state ) .
the concentration of atoms excited during the time of a single electric breakdown , i.e. the screening time , @xmath148 to energies within an interval @xmath149 reads @xmath150 where @xmath151 is the number of the noble gas atoms in the metastable state , @xmath152 is the cross section of the impact excitation of an atom from the metastable state to a state with energy @xmath147 .
atoms excited to states with the energy @xmath147 generally do not go to the ground state directly but through intermediate excited states .
it reasonable to assume that with a high probability they radiate phonons with energy @xmath153 .
so the spectral density of the sbsl radiation energy per pulse for unit volume can be written as @xmath154 where @xmath155 is the probability of the spontaneous radiation transition .
taking into account that @xmath156 where @xmath157 is modulus of matrix element of the dipole moment of the transition @xcite we find @xmath158 recall that @xmath127 in eq .
[ spectrn ] is a function of the coordenates because @xmath159 ( see eq.[temp - el ] ) .
putting , as before , @xmath160 and integrating approximately eq.[spectrn ] we find the spectral power of sbsl for a single breakdown is @xmath161 where @xmath162and @xmath163 is , practically , the total number of electrons participating in the breakdown .
note that the approximative eq.[sbsl ] is not sensitive to the form of decay of the field , it is important only that @xmath164 decays more steeply than @xmath165 , i.e. it is not important that for @xmath58 we use an expression for an incompressible liquid .
the observed spectra are cut off in the shortwave region because of the absorption of water .
this can be taken into account by multiplying eq.[sbsl ] by exp(-@xmath166 ) where @xmath167 is the size of the acoustic resonator ( @xmath168 cm @xcite ) .
because of the urbach absorption tails @xcite the radiation maximum is located considerably below than @xmath169 .
it should be emphasized that the spectrum given by eq.[sbsl ] resembles the black - body spectrum but temperature is given here by the effective temperature @xmath170 of hot electrons near the bubble surface .
the value of @xmath171 may be much higher than the bubble gas temperature . experimentally observed spectra
can be fitted , in the wavelength interval ( 200@xmath172 nm , to the black - body ones with temperatures ( 2@xmath05)10@xmath173k @xcite . according to eq.[temp - el ]
such electron temperatures are reached at electric fields @xmath174 v / m for values of @xmath175ev and @xmath176 characteristic for water @xcite . as follows from eqs.[flex ] and
@xmath177 such fields are realized at accelerations @xmath178 m / s@xmath179 similar accelerations are reported in @xcite . integrating eq.[sbsl ] multiplied by exp(-@xmath166 ) we find an estimate of the radiation energy for a single breakdown @xmath180 where @xmath181 is the characteristic photon energy which is close to the energy of the maximum of the observed spectrum , which is about ( @xmath182 in the case of strong electric field of our interest .
recall that in eq.[power ] @xmath183 is the screening ( breakdown ) time and @xmath184 is the number of transmitted electrons in the screening process ( sec . 3 ) and
@xmath185 is the concentration of noble gas atoms in a metastable state near the bubble surface . as we have mentioned before it was shown in @xcite that because of dissolved gas diffusion and chemical reactions an accumulation of noble gas in the bubble takes place while the bubble pulsates and in stationary conditions the gas in the bubble consists almost entirely of noble gas .
therefore one can assume that the concentration of the noble gas atoms , @xmath186near the bubble surface is close to its saturation value .
note that this value has to be estimated for atmospheric pressure and ambient temperature rather than for the high pressures and temperatures existing at the last stage of the collapse over a very short time .
this is why we assume that @xmath187 ( 10@xmath188 ) @xmath70 .
now we will argue that the value of @xmath151 can be only one order of magnitude less than @xmath189 .
indeed , for @xmath190 ev obtained above by fitting the sbsl spectrum to the black - body spectrum , the number of electrons having an energy higher than 10ev , @xmath191 may be about 1@xmath192 of the total electron number @xmath193 i.e. , @xmath191 is about 10@xmath194 .
such a superhot
electron excites a noble gas atom to the metastable state over the time @xmath195s for @xmath196@xmath197 @xmath198 m / s and @xmath187(10@xmath199 ) @xmath200 therefore a
superhot
electron during its participation in screening excites @xmath201 noble gas atoms and the total number of the excited atoms is @xmath202 if one takes into account that @xmath203 @xmath204s .
the total number of the noble gas atoms in the region of the strong field ( @xmath2051@xmath57m@xmath206 is about @xmath207 , i.e. the percentage of the noble gas atoms excited to the metastable state during a single breakdown can be about 10@xmath208 . since the life time of the metastable state is much more than @xmath4 and during this time interval several breakdowns can take place ( sec.3 ) , one may expect that the number of the noble gas atoms excited to the metastable state during the interval @xmath209 is only one order of magnitude less than the total number of these atoms i.e. @xmath151 @xmath205 10@xmath210 @xmath70 . if the life time of the metastable states exceeds the acoustic period ( @xmath211s ) an accumulation of noble gas atoms in the metastable states can also occur during several acoustic periods .
now we can estimate the total radiation energy .
we have mentioned above that the maximum value of @xmath193 , @xmath212 1@xmath57 m , @xmath213cm@xmath48 and @xmath215 .
taking into account that the positive acceleration period @xmath2ns and the breakdown time @xmath216 is about 0.01ns one can assume that the number of breakdowns is not less than 10 .
thus one finds from eq.[power ] that the maximum photon number in the sbsl pulse can be more than 10@xmath217 .
this value corresponds to the maximum photon number observed experimentally @xcite .
note that the radiation energy is a small part of the total energy of the flexoelectric field ( see ref@xcite ) .
let us emphasize once more that within the mechanism of sbsl considered the main processes occur in water near the bubble surface unlike the most widely discussed mechanisms of sbsl @xcite . within the framework of this mechanism many experimental data about the sbsl can be explained quite naturally .
of course , since the parameters of the collapsing bubble are not reliably known when the bubble radius is close to its minimum value we can present no more than order - of - magnitude estimations .
the minimum duration of the sbsl flash is determined by the single breakdown ( screening ) time , @xmath218ps .
a larger time is possible because of the multiplicity of breakdowns .
the maximum duration is limited by @xmath2ns .
the longer is the duration the greater is the energy of the flash .
this is in agreement with experiments where it was found that the pulse duration changes from 30ps to 400ps when its energy increases @xcite .
( iii ) . within our scenario
the noble gas atoms play an important role .
they do not reduce so much the breakdown threshold as they govern the radiation process .
abundance of optical transitions in these atoms and inhomogeneous broadening because of the stark effect explain the practically continuous character of the sbsl spectrum .
( iv ) . in agreement
with the experiment @xcite the theoretical spectrum of the sbsl resembles the black - body spectrum but the temperature is given here by the effective temperature of the hot electrons which can be about several ev , what corresponds to the observed apparent radiation temperature @xcite . at the same time
the gas temperature is not directly related to the radiation temperature and can be considerably less than 1ev .
the increase of influence on sbsl intensity in the noble gas series he - xe observed in the experiment @xcite is connected with the decrease , in this series , of the energy of the lowest metastable state .
note also that within our scenario of sbsl the influence of magnetic fields on sbsl is fairly weak .
their influence becomes appreciable when they are high enough to hamper the heating of electrons@xcite .
this might be the reason for the decrease of the sbsl intensity in strong magnetic fields which has been observed experimentally@xcite .
\(i ) the strong geometric asymmetry of the water molecule , which accompanies its electric asymmetry , provides the needed sign of the flexoelectric coefficient .
this sign is such that the electric field has the correct direction
( accelerates electron ) over the same ( very short ) period in which electrons are generated because of the sharp increase of the bubble gas temperature .
the temporal coincidence of these two cicumstances is a possible reason of the effect of synchronization of the light pulses observed experimentally @xcite .
\(iii ) high solubility of noble gases in water what makes possible high concentrations of noble gas in water near the bubble filled by the noble gas accumulated there in the process of the bubble pulsations .
some experimentally found characteristics of sbsl have not been explained in this paper , in particular the dependence of the energy of the pulse upon the partial pressure of noble gases and water temperature @xcite .
these dependences may be not necessarily a manifestation of the radiation processes studied in this paper but they could be a consequence of a set of various factors
. they may be determined by the hydrodynamics of the pulsating bubble , the solubility of noble gases in water and also by the temperature dependence of the flexoelectric coefficient of water which still remains , unfortunately , unmeasured .
although the main features of sbsl seem to be explainable within our scenario , these explanations are qualitative rather than quantitative .
we are still a long way from a quantitative theory at present .
one of the main aims of the paper is to stimulate some experiments that could either support or discard the proposed mechanis of sbsl . after performing these experiments
it makes sense to develop the theory further .
in particular , it will be necessary to study in more detail the kinetics of the screening process and of the excitation of noble gas atoms while taking into account the time variation of the distribution function of the hot electrons .
this problem should be considered together with that of the determination of the spatial distribution of excited noble gas atoms near the bubble surface .
diffusion and relaxation of the excited atoms should be considered together , of course , with the processes that lead to noble gas accumulation in the bubble @xcite . in our estimations we assumed that flexoelectric coefficient of water has its natural value .
this was sufficient for the theoretical estimations to be in agreement with the experimental data .
in fact , because of the above - mentioned strong geometric asymmetry of the water molecule , the coefficient could be even larger .
if this is the case some new phenomena will take place including influence of the flexoelectric effect on hydrodynamics of the pulsating bubble and x - ray radiation induced by electrons of very high energy .
we are grateful to v.kholodnov for useful discussions .
ng and vvo thank eu esprit , the spanish csic and nato for linkage grant ref .
outrlg 970308 .
also , apl thanks nato for linkage grant htechlg 971213 . | it was proposed before that single bubble sonoluminescence ( sbsl ) may be caused by strong electric fields occurring in water near the surface of collapsing gas bubbles because of the flexoelectric effect involving polarization resulting from a gradient of pressure . here
we show that these fields can indeed provoke dynamic electric breakdown in a micron - size region near the bubble and consider the scenario of the sbsl .
the scenario is : ( i ) at the last stage of incomplete collapse of the bubble , the gradient of pressure in water near the bubble surface has such a value and a sign that the electric field arising from the flexoelectric effect exceeds the threshold field of the dynamic electrical breakdown of water and is directed to the bubble center ; ( ii ) mobile electrons are generated because of thermal ionization of water molecules near the bubble surface ; ( iii ) these electrons are accelerated in cold water by the strong electric fields ; ( iv ) these hot electrons transfer noble gas atoms dissolved in water to high - energy excited states and optical transitions between these states produce sbsl uv flashes in the trasparency window of water ; ( v ) the breakdown can be repeated several times and the power and duration of the uv flash are determined by the multiplicity of the breakdowns .
the sbsl spectrum is found to resemble a black - body spectrum where temperature is given by the effective temperature of the hot electrons .
the pulse energy and some other characteristics of the sbsl are found to be in agreement with the experimental data when realistic estimations are made . |
the fermion electric dipole moments ( edms ) offer a powerful probe for new physics beyond the standard model ( sm ) . in the sm
, the edm of the neutron is fully induced by the cp phase of the cabibbo - kobayashi - maskawa ( ckm ) matrix elements and it is predicted to be much smaller @xcite than the present experimental upper limit of @xmath0 @xcite and beyond the reach of experiment in the near future . as for the minimal supersymmetric extension of the sm ( mssm )
, there are many new sources of the cp violation that can result in larger contributions to the edm of the neutron @xcite . taking the cp phases with a natural size of @xmath1 , and the supersymmetry mass spectra at the tev range , the theoretical prediction on the neutron edm at one - loop level already exceeds the present experimental upper bound . in order to make the theoretical prediction consistent with the experimental data ,
three approaches are adopted in the literature .
one possibility is to make the cp phases sufficiently small , i.e. @xmath2 @xcite .
one can also assume a mass suppression by making the supersymmetry spectra heavy , i.e. in the several tev range @xcite , or invoke a cancellation among the different contributions to the fermion edms @xcite .
the prediction for the fermion edms at one - loop level in a supersymmetric theory has been extensively discussed in the literature . on other hand ,
the analysis on the neutron edm at two - loop level is less advanced , even though some pioneer work has been carried out , for example the two - loop barr - zee type diagrams involving the higgs bosons @xcite and the purely gluonic dimension - six weinberg operator induced by two - loop gluino - squark diagrams @xcite have been analyzed .
analyzing the edms at two - loop order can give us a better understanding of where the new physics scale may emerge and shed some light to the spectra of new physical particles around this scale .
moreover , the two - loop analysis involves some new parameters in addition to those appearing in the one - loop calculations , hence resulting in more rigorous constraints on the supersymmetry parameter space . in this work
we shall analyse the two - loop gluino corrections to the neutron edm in the mssm .
we work in the framework of the simplest model , where we will neglect all other possible sources of flavor violation except those related to the ckm matrix , and try to avoid ambiguities of the unification conditions of the soft - breaking parameters at the grand unification scale present in the msugra schemes . in the next section
, we will demonstrate how to obtain the two - loop gluino corrections to the neutron edm . in the section [ sec3 ] we will present the results of our numerical computations and we study the dependence of the neutron edm on the supersymmetry parameters .
conclusions are presented in section [ sec4 ] .
in the effective lagrangian , the fermion edm @xmath3 is defined through the dimension five operator @xmath4 where@xmath5 is a fermion field , @xmath6 is the electromagnetic field strength .
this coupling obviously signifies a cp violation .
it is not present among the fundamental interactions at tree level , but it is generated in loop level in an electroweak theory with cp violation .
moreover , because quarks also take part in strong interactions , the chromoelectric dipole moment ( cedm ) @xmath7 of quarks , where @xmath8 denote the generators of the strong @xmath9 gauge group , @xmath10 is the gluon field strength , and the pure gluon weinberg operator of dimension - six , @xmath11 , contributes to the quark edms as well . a convenient way to describe loop - induced contributions is the effective theory approach , where the heavy particles are integrated out at the matching scale .
the resulting effective lagrangian includes a full set of cp violation operators . in this work ,
we restrict ourselves to the following operators that are relevant to the neutron edm : @xmath12 where @xmath13 are the wilson coefficients evaluated at the scale @xmath14 , and the five operators of interests are @xmath15 ( 0,20)(0,0 ) ( -62,-100 ) the effective triangle diagrams responsible for the quark edms and cedms , are obtained by attaching a photon or gluon external line , for quark edm or cedm , respectively , in all possible ways to the quark self - energy diagrams .
the one - loop supersymmetric corrections to the wilson coefficients @xmath16 in eq .
( [ eq2 ] ) originate from three types of the graphs : the quark self - energy diagrams , where one has a gluino and a squark , a chargino and a squark or a neutralino and a squark as virtual particles in the loop ( fig.[fig1 ] )
. we will adopt below a terminology where , for example , the `` gluino - squark contribution '' means the sum of those triangle diagrams , which have a gluino and a squark as virtual states and where a photon or a gluon in attached in all possible ways to the external line . before moving to derive the two - loop gluino corrections , we first present , in order to familiarize our notation , the one - loop results .
those expressions can be found in the literature , and we just translate them here into our notations for later convenience .
the corrections from the gluino - squark diagrams are @xmath17 here @xmath18 , @xmath19 denotes the phase of the soft gluino mass @xmath20 , and @xmath21 are the mixing matrices of the squarks , i.e. @xmath22 where @xmath23 with @xmath24 for @xmath25 . as usual
, @xmath26 is the ratio between the vevs of the up- and down - type higgs fields , and @xmath27 is the weak mixing angle .
we also use the short - hand notations @xmath28 .
the loop functions @xmath29 and @xmath30 are defined as @xmath31^{-1 } [ 1+r+2r\ln r/(r-1)]$ ] and @xmath32^{-1}[10r-26-(2r-18)\ln r/(r-1)]$ ] . in a similar way
, the one - loop neutralino - squark contributions can be written as @xmath33 with @xmath34 here @xmath35 , @xmath36 denote the eigenvalues of neutralino mass matrix , and @xmath37 is the correspondingly mixing matrix .
( 0,80)(0,0 ) ( -50,-50 ) finally , the chargino - squark contributions are given by @xmath38 \;,\nonumber\\ & & d_{_{\chi_{_k}^\pm(1)}}^g={g_3\alpha\over4\pi s_{_{\rm w}}^2 } v_{_{qq}}^\dagger v_{_{qq } } \sum\limits_{i , k}{\bf im}\big((a_{_c}^q)_{_{k , i}}(b_{_c}^q)^\dagger _ { _ { i , k}}\big ) \nonumber\\ & & \hspace{1.2cm}\times { m_{_{\chi_{_k}^\pm}}\over m_{_{\tilde{q}_i}}^2 } b\big({m_{_{\chi_{_k}^\pm}}^2\over m_{_{\tilde{q}_i}}^2}\big)\ ; , \label{eqa3 - 4}\end{aligned}\ ] ] where @xmath39 denotes the ckm matrix , @xmath40 are the masses of charginos , and the loop function @xmath41 $ ] .
the couplings appearing in these expressions are defined as @xmath42 where @xmath43 and @xmath44 are the right- and left - handed mixing matrices of the charginos , respectively .
it is noteworthy that one - loop chargino corrections to the quark edms and cedms are proportional to the suppression factors @xmath45 contained in the couplings @xmath46 .
the two - loop gluino corrections to the wilson coefficients originate from the two - loop self - energy diagrams for quarks , depicted in fig .
the corresponding dipole moment diagrams are obtained by attaching a photon or gluon to these diagrams in all possible ways . in these diagrams
there is no new suppression factor , except a factor arising from loop integration , and the divergence caused by the sub - diagrams is subtracted in @xmath47 scheme safely .
it turns out that for some region of the parameter space the two - loop results are comparable with the one - loop contributions .
the reason for this is that the dependence of the two - loop results on the relevant cp phases differs from that of the one - loop results .
since the two - loop analysis is more subtle than the analysis at one - loop level , we present here in some detail all the processes , which contribute at two - loop level into the theoretical prediction of the quark edms . taking the same steps , which we did in our earlier works @xcite , we obtain the following expression for the relevant effective lagrangian : @xmath48 \nonumber\\ & & \hspace{1.2 cm } + ( { \cal z}_{_{\tilde q}})_{_{1,j } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{j,2}}({\cal z}_{_{\tilde q}})_{_{2,i } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{i,2}}|m_{_{\tilde g}}|e^{i\theta_{_3 } } \big[e_{_q}e{\cal n}_{_{\tilde{g}(1)}}^\gamma{\cal q}_{_2 } + g_3{\cal n}_{_{\tilde{g}(1)}}^g{\cal q}_{_4}\big ] \nonumber\\ & & \hspace{1.2 cm } + ( { \cal z}_{_{\tilde q}})_{_{2,j } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{j,2}}({\cal z}_{_{\tilde q}})_{_{2,i } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{i,1}}|m_{_{\tilde g}}|e^{-i\theta_{_3 } } \big[e_{_q}e{\cal n}_{_{\tilde{g}(2)}}^\gamma{\cal q}_{_1 } + g_3{\cal n}_{_{\tilde{g}(2)}}^g{\cal q}_{_3}\big ] \nonumber\\ & & \hspace{1.2 cm } + ( { \cal z}_{_{\tilde q}})_{_{1,j } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{j,1}}({\cal z}_{_{\tilde q}})_{_{1,i } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{i,2}}|m_{_{\tilde g}}|e^{i\theta_{_3 } } \big[e_{_q}e{\cal n}_{_{\tilde{g}(2)}}^\gamma{\cal q}_{_2 } + g_3{\cal n}_{_{\tilde{g}(2)}}^g{\cal q}_{_4}\big]\big\}+\cdots\ ; , \label{eq4}\end{aligned}\ ] ] where @xmath49 and @xmath50 is the time - space dimension .
we collect the complicated form factor @xmath51 in appendix [ ap1 ] .
the effective lagrangian of eq .
( [ eq4 ] ) originates in fact from the contributions of the effective triangle vertices induced by the `` gluino - gluino self - energy '' ( fig.2 ( a ) ) .
the terminology here is analogous to the one adopted in the one - loop case above .
the gluino - gluino self - energy is a short - hand expression for a diagram , where we attach a photon or gluon line in all possible ways to the two - loop quark self - energy diagram having , besides the squark , two gluinos are as intermediate agents .
because the sum of the triangle diagrams corresponding to each `` self - energy '' obviously respects the gauge invariance , we can calculate the contributions of all the `` self - energies '' separately .
we will use below the following identities demanded by the translational invariance of the loop - momenta : @xmath52 \nonumber\\ & & + { 2\over q_1 ^ 2-m_{_1}^2 } { q_1 ^ 2q_1\cdot(q_2-q_1)\over d } \big\}\equiv0\ ; , \nonumber\\ & & \int{d^dq_1\over(2\pi)^d}{d^dq_2\over(2\pi)^d}{1\over{\cal d}_{_0 } } \big\{-q_1\cdot(q_2-q_1 ) \nonumber\\ & & + { 2\over q_1 ^ 2-m_{_1}^2}\big[{d(q_1\cdot q_2)^2-q_1 ^ 2q_2 ^ 2\over d(d-1 ) } -{q_1 ^ 2q_1\cdot q_2\over d}\big ] \nonumber\\ & & + { 2\over q_2 ^ 2-m_{_2}^2}{q_1\cdot(q_2-q_1)q_2 ^ 2\over d } \big\}\equiv0\ ; , \label{eq5}\end{aligned}\ ] ] where @xmath53 . we find @xmath54 + { 4\over d}{q_1 ^ 2q_1\cdot(q_1-q_2)\over(q_1 ^ 2-|m_{_{\tilde g}}|^2)^2 } \nonumber\\ & & \hspace{0.0 cm } -{2+d\over d}{q_1\cdot(q_1-q_2)\over q_1 ^ 2-|m_{_{\tilde g}}|^2 } + { 2-d\over d}{q_1\cdot(q_1-q_2)\over(q_2-q_1)^2-m_{_q}^2}\big\ } \nonumber\\ = & & \hspace{.0 cm } \int{d^dq_1\over(2\pi)^d}{d^dq_2\over(2\pi)^d}{1\over{\cal d}_{_{\tilde g } } } \big\{{d-2\over d}{q_1\cdot(q_2-q_1)\over(q_2-q_1)^2-m_{_q}^2 } + { 2\over d}{q_1 ^ 2q_1\cdot(q_2-q_1)\over(q_1 ^ 2-|m_{_{\tilde g}}|^2 ) ( q_1 ^ 2-m_{_{\tilde{q}_j}}^2 ) } \nonumber\\ & & \hspace{0.0 cm } + \big[{2\over(q_1 ^ 2-|m_{_{\tilde g}}|^2 ) ( q_2 ^ 2-|m_{_{\tilde g}}|^2)}+{1\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2 ) ( q_2 ^ 2-m_{_{\tilde{q}_i}}^2)}\big]\big [ { d(q_1\cdot q_2)^2-q_1 ^ 2q_2 ^ 2\over d(d-1 ) } \nonumber\\ & & \hspace{.0 cm } -{q_1 ^ 2q_1\cdot q_2\over d}\big ] + { 2\over d}{q_2 ^ 2q_1\cdot(q_2-q_1)\over(q_2 ^ 2-m_{_{\tilde{q}_i}}^2 ) ( q_2 ^ 2-|m_{_{\tilde g}}|^2 ) } -{2\over d}{q_1\cdot(q_2-q_1)\over q_1 ^ 2-m_{_{\tilde{q}_j}}^2}\big\ } \nonumber\\ = & & \hspace{.0 cm } \int{d^dq_1\over(2\pi)^d}{d^dq_2\over(2\pi)^d}{1\over{\cal d}_{_{\tilde g } } } \big\{{d-2\over d}{q_1\cdot(q_2-q_1)\over(q_2-q_1)^2-m_{_q}^2 } -{4\over d}{q_1\cdot(q_2-q_1)q_2 ^
2\over(q_2 ^ 2-|m_{_{\tilde g}}|^2)^2 } \nonumber\\ & & \hspace{.0 cm } -{4\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)(q_2 ^ 2-|m_{_{\tilde g}}|^2)}\big [ { d(q_1\cdot q_2)^2-q_1 ^ 2q_2 ^ 2\over d(d-1 ) } -{q_1 ^ 2q_1\cdot q_2\over d}\big ] \nonumber\\ & & \hspace{.0 cm } -{4\over d}{q_1 ^ 2q_1\cdot(q_2-q_1)\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)^2 } + { q_1\cdot(q_2-q_1)\over q_2
^ 2-|m_{_{\tilde g}}|^2 } + { q_1\cdot(q_2-q_1)\over q_1 ^ 2-m_{_{\tilde{q}_j}}^2}\big\ } \nonumber\\ = & & \hspace{.0 cm } \int{d^dq_1\over(2\pi)^d}{d^dq_2\over(2\pi)^d } { { \cal n}_{_{\tilde{g}(1)}}^\gamma\over{\cal d}_{_{\tilde g}}}\;. \label{eq6}\end{aligned}\ ] ] similarly , we find @xmath55 with the help of eq.([eq6 ] ) and eq.([eq7 ] ) , the corresponding terms in eq.([eq4 ] ) are transformed into @xmath56 \nonumber\\ & & \hspace{1.2 cm } -i{4\over9}g_3 ^
5|m_{_{\tilde g}}|\int{d^dq_1\over(2\pi)^d } { d^dq_2\over(2\pi)^d}{{\cal n}_{_{\tilde{g}(1)}}^g\over{\cal d}_{_{\tilde g } } } % \nonumber\\ % & & \hspace{1.2cm}\times { \bf im}\big(({\cal z}_{_{\tilde q}})_{_{2,j}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,1 } } e^{-i\theta_{_3}}\big)\cdot\big[\bar{q}t^a\sigma^{\mu\nu}\gamma_5qg^a_{\mu\nu}\big ] \nonumber\\ & & \hspace{1.2 cm } + \cdots\;. \label{eq7a}\end{aligned}\ ] ] the two - loop scalar integrations can be reduced into the two - loop vacuum integrals defined as @xmath57 which was analyzed in detail @xcite .
however , the two - loop scalar integration @xmath58 contain the ultra - violet divergence which originates from the sub - daigrams .
after the renormalization in @xmath47 scheme , we finally obtain the corrections from the two - loop `` gluino - gluino '' diagram to the quark edms and cedms respectively as @xmath59 with @xmath60 , and the function @xmath61 are defined in appendix [ ap2 ] .
the effective lagrangian , which corresponds to the contributions of the triangle diagrams induced in the `` neutralino - gluino '' self - energy ( i.e. diagrams where loop squarks are accompanied with neutralinos and gluinos , see fig.2 ( b ) and ( c ) ) , includes the following pieces : @xmath62 \nonumber\\ & & \hspace{1.2 cm } + ( b_{_n}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,2 } } ( a_{_n}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,2}}|m_{_{\tilde g}}| e^{i\theta_{_3}}\big[e_{_q}e{\cal n}_{_{\tilde{g}(1)}}^\gamma{\cal q}_{_2 } + g_3{\cal n}_{_{\chi_{_k}^0(1)}}^{g(a)}{\cal q}_{_4}\big ] \nonumber\\ & & \hspace{1.2 cm } -(a_{_n}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,2}}(a_{_n}^q)_{_{ki } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{i,1}}m_{_{\chi_{_k}^0 } } \big[e_{_q}e{\cal n}_{_{\tilde{g}(2)}}^\gamma{\cal q}_{_1 } + g_3{\cal n}_{_{\chi_{_k}^0(2)}}^{g(a)}{\cal q}_{_3}\big ] \nonumber\\ & & \hspace{1.2 cm } -(b_{_n}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,1}}(b_{_n}^q)_{_{ki } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{i,2}}m_{_{\chi_{_k}^0 } } \big[e_{_q}e{\cal n}_{_{\tilde{g}(2)}}^\gamma{\cal q}_{_2 } + g_3{\cal n}_{_{\chi_{_k}^0(2)}}^{g(a)}{\cal q}_{_4}\big]\big\ } \nonumber\\ & & \hspace{1.2 cm } + { 2\over3}{e^2g_3 ^ 2\over s_{_{\rm w}}^2c_{_{\rm w}}^2}\int { d^dq_1\over(2\pi)^d}{d^dq_2\over(2\pi)^d}{1\over{\cal d}_{_{\chi_{_k}^0}}^{(b ) } } \nonumber\\ & & \hspace{1.2cm}\times \big\{-({\cal z}_{_{\tilde q}})_{_{2,j}}(b_{_n}^q)_{_{jk}}^\dagger ( { \cal z}_{_{\tilde q}})_{_{1,i}}(b_{_n}^q)_{_{ik}}^\dagger m_{_{\chi_{_k}^0 } } \big[e_{_q}e{\cal n}_{_{\chi_{_k}^0(1)}}^{\gamma(b)}{\cal q}_{_1 } + g_3{\cal n}_{_{\chi_{_k}^0(1)}}^{g(b)}{\cal q}_{_3}\big ] \nonumber\\ & & \hspace{1.2 cm } -({\cal z}_{_{\tilde q}})_{_{1,j}}(a_{_n}^q)_{_{jk}}^\dagger ( { \cal z}_{_{\tilde q}})_{_{2,i}}(a_{_n}^q)_{_{ik}}^\dagger m_{_{\chi_{_k}^0 } } \big[e_{_q}e{\cal n}_{_{\chi_{_k}^0(1)}}^{\gamma(b)}{\cal q}_{_2 } + g_3{\cal n}_{_{\chi_{_k}^0(1)}}^{g(b)}{\cal q}_{_4}\big ] \nonumber\\ & & \hspace{1.2 cm } + ( { \cal z}_{_{\tilde q}})_{_{2,j}}(a_{_n}^q)_{_{jk}}^\dagger ( { \cal z}_{_{\tilde q}})_{_{2,i}}(b_{_n}^q)_{_{ik}}^\dagger \big[e_{_q}e{\cal n}_{_{\chi_{_k}^0(2)}}^{\gamma(b)}{\cal q}_{_1 } + g_3{\cal n}_{_{\chi_{_k}^0(2)}}^{g(b)}{\cal q}_{_3}\big ] \nonumber\\ & & \hspace{1.2 cm } + ( { \cal z}_{_{\tilde q}})_{_{1,j}}(b_{_n}^q)_{_{jk}}^\dagger ( { \cal z}_{_{\tilde q}})_{_{1,i}}(a_{_n}^q)_{_{ik}}^\dagger \big[e_{_q}e{\cal n}_{_{\chi_{_k}^0(2)}}^{\gamma(b)}{\cal q}_{_2 } + g_3{\cal n}_{_{\chi_{_k}^0(2)}}^{g(b)}{\cal q}_{_4}\big]\big\}+\cdots \label{eq9}\end{aligned}\ ] ] with @xmath63 . the explicit expressions of the form factors @xmath64 are presented in appendix [ ap1 ] . with a help of eq .
( [ eq5 ] ) , one finds the following identities : @xmath65 substituting these identities into eq.(15 ) and removing the ultra - violet divergence in @xmath47 scheme , we may extract the two - loop `` neutralino - gluino '' corrections to the quark edms and cedms : @xmath66 \nonumber\\ & & \hspace{1.4 cm } -m_{_{\chi_{_k}^0}}f_2(x_{_q } , x_{_{\tilde{q}_j } } , x_{_{\tilde g } } , x_{_{\chi_k^0 } } , x_{_{\tilde{q}_i}})\big[{\bf i m } \big((a_{_n}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,2 } } ( a_{_n}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,1}}\big ) \nonumber\\ & & \hspace{1.4 cm } -{\bf im}\big((b_{_n}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,1 } } ( b_{_n}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,2}}\big)\big]\big\ } \;,\nonumber\\ & & d_{_{\chi_{_k}^0(2)}}^g= { 4g_3\alpha\alpha_{_s}\over3(4\pi)^2s_{_{\rm w}}^2c_{_{\rm w}}^2m_{_{\rm w}}^2 } \big\{|m_{_{\tilde g}}|f_4(x_{_q } , x_{_{\tilde{q}_j } } , x_{_{\tilde g } } , x_{_{\chi_k^0 } } , x_{_{\tilde{q}_i}})\big[{\bf im}\big((a_{_n}^q)_{_{kj } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{j,1 } } \nonumber\\ & & \hspace{1.4cm}\times ( b_{_n}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,1 } } e^{-i\theta_{_3}}\big)-{\bf im}\big((b_{_n}^q)_{_{kj } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{j,2}}(a_{_n}^q)_{_{ki } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{i,2}}e^{i\theta_{_3}}\big)\big ] \nonumber\\ & & \hspace{1.4 cm } -m_{_{\chi_{_k}^0}}f_5(x_{_q } , x_{_{\tilde{q}_j } } , x_{_{\tilde g } } , x_{_{\chi_k^0 } } , x_{_{\tilde{q}_i}})\big[{\bf i m } \big((a_{_n}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,2 } } ( a_{_n}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,1}}\big ) \nonumber\\ & & \hspace{1.4 cm } -{\bf im}\big((b_{_n}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,1 } } ( b_{_n}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,2}}\big)\big ] \big\}\;. \label{eq11}\end{aligned}\ ] ] in a similar way , we get the following results for the two - loop `` gluino - chargino '' corrections : @xmath67 \nonumber\\ & & \hspace{1.4 cm } -m_{_{\chi_{_k}^\pm}}f_7(x_{_q } , x_{_{\tilde{q}_j } } , x_{_{\tilde g } } , x_{_{\chi_k^\pm } } , x_{_{\tilde{q}_i}})\big[{\bf i m } \big((a_{_c}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,2 } } ( a_{_c}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,1}}\big ) \nonumber\\ & & \hspace{1.4 cm } -{\bf im}\big((b_{_c}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,1 } } ( b_{_c}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,2}}\big)\big]\big\ } \;,\nonumber\\ & & d_{_{\chi_{_k}^\pm(2)}}^g= { 4g_3\alpha\alpha_{_s}\over3(4\pi)^2s_{_{\rm w}}^2m_{_{\rm w}}^2}v_{_{qq}}^\dagger v_{_{qq } } \big\{|m_{_{\tilde g}}|f_4(x_{_q } , x_{_{\tilde{q}_j } } , x_{_{\tilde g } } , x_{_{\chi_k^\pm } } , x_{_{\tilde{q}_i}})\big[{\bf im}\big((a_{_c}^q)_{_{kj } } ( { \cal z}_{_{\tilde q}}^\dagger)_{_{j,1 } } \nonumber\\ & & \hspace{1.4cm}\times ( b_{_c}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,1 } } e^{-i\theta_{_3}}\big ) -{\bf im}\big((b_{_c}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,2 } } ( a_{_c}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,2 } } e^{i\theta_{_3}}\big)\big ] \nonumber\\ & & \hspace{1.4 cm } -m_{_{\chi_{_k}^\pm}}f_5(x_{_q } , x_{_{\tilde{q}_j } } , x_{_{\tilde g } } , x_{_{\chi_k^\pm } } , x_{_{\tilde{q}_i}})\big[{\bf i m } \big((a_{_c}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,2 } } ( a_{_c}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,1}}\big ) \nonumber\\ & & \hspace{1.4 cm } -{\bf im}\big((b_{_c}^q)_{_{kj}}({\cal z}_{_{\tilde q}}^\dagger)_{_{j,1 } } ( b_{_c}^q)_{_{ki}}({\cal z}_{_{\tilde q}}^\dagger)_{_{i,2}}\big)\big ] \big\}\;.\nonumber\\ \label{eq12}\end{aligned}\ ] ] notice that the last terms of the @xmath68 and @xmath69 are not proportional to the suppression factor @xmath70 .
this implies that the two - loop `` gluino - chargino '' diagrams may be dominant two - loop corrections of the chargino to the quark edms and cedms .
the wilson coefficient of the purely gluonic weinberg operator originates from the two - loop `` gluino - squark '' diagrams and is given by @xcite : @xmath71 where @xmath72 is the diagonalizing matrix for the squared mass matrix of stop , and the function @xmath73 is the same as that given in @xcite . in order to account for the resummation of logarithmic corrections
, we should evolve the quark edms , cedms and the wilson coefficients of the weinberg operator with the renormalization group equations ( rges ) from the matching scale @xmath14 down to the chirality breaking scale @xmath74 @xcite : @xmath75 where @xmath76 and @xmath77 . at a low scale ,
the quark edm consists of a weak interaction contribution , a quark cedms contribution and the contribution of the weinberg operator .
we can numerically evaluate these different contributions . according to a naive dimensional analysis one can write @xcite @xmath78 the edm of the neutron is , on the basis of the simple su(6 ) quark model , then given by @xmath79 where @xmath80 are the edms of the up- and down - type quarks respectively .
the mssm lagrangian contains several sources for cp violating phases : the phases of the @xmath81 parameter in the superpotential and the corresponding bilinear coupling of the soft breaking terms , three phases of the gaugino mass terms , and the phases of the trilinear sfermion yukawa couplings in the soft lagrangian . as we are not considering the spontaneous cp violation in this work , the cp phase of soft bilinear coupling vanishes due to the tree level neutral higgs tadpole conditions . as we remarked above , the two - loop gluino corrections to the neutron edm
are important , and may be the dominant contribution in a certain part of the parameter space .
let us clarify this point in some detail by means of numerical analysis . without losing too much generality , we will fix the following values for the supersymmetric parameters : @xmath82 , are the soft masses of @xmath83 gauginos respectively . ]
@xmath84 , and @xmath85 in our numerical analysis . for simplification
, we will also assume @xmath86 .
( 0,80)(0,0 ) ( -50,-30 ) in fig .
[ fig3 ] we present the corrections the neutron edm obtains from the one - loop `` gluino - squark '' diagrams ( see eq .
( [ eqa3 - 1 ] ) and fig . 1 ) and two - loop `` gluino - gluino '' contributions ( see eq .
( [ eq8 ] ) and fig .
2 ) as a function of the right - handed squark masses @xmath87 .
we have set in this figure @xmath88 , and @xmath89 , and what are plotted are the one - loop `` gluino - squark '' contribution ( dashed line ) and the sum of the one - loop `` gluino - squark '' and two - loop `` gluino - gluino '' contributions ( solid line ) . the dependence of the contributions on the cp violation phase @xmath19 are all proportional to the factor @xmath90 , and the two - loop correction is less than 10% of the one - loop results for our choice of the parameter values .
the neutron edm depends on the parameter @xmath91 through the squark mixing matrices , and as seen from the plots , this makes the a large difference between the corrections in a low and a high @xmath91 cases .
let us now focus on the two - loop gluino corrections of the neutron edm .
there the cp violating phases @xmath92 induce a nonzero quark edms at one - loop level ( see eq .
( [ eqa3 - 2 ] ) ) through the neutralino mixing matrix , whereas the cp phase @xmath19 contributes only through the two - loop diagrams ( see eq .
( [ eq11 ] ) ) .
we will present three sets of plots , where we set one of the phases @xmath93 at a time equal to @xmath94 while the other two are set equal to zero .
taking first @xmath95 and @xmath96 , we plot in fig .
[ fig4 ] for two values of @xmath91 the neutron edm as a function of the soft squark masses @xmath97 . in this plot
the dash - lines represent the one - loop `` neutralino - squark '' results and the solid - lines correspond to a sum of one - loop `` neutralino - squark '' and two - loop `` neutralino - gluino '' contributions . in fig .
[ fig5 ] we present the same plot for the case @xmath98 , other parameters being the same as in fig . [ fig4 ] .
we notice that the two - loop gluino corrections to the one - loop results can be as large as about @xmath99 . for the contribution of neutralino to the neutron edm
, the most interesting piece originates from the cp violating phase of the gluino mass , which does not contribute at the one - loop level .
( 0,80)(0,0 ) ( -50,-30 ) ( 0,80)(0,0 ) ( -50,-30 ) in fig .
[ fig6 ] we plot the neutron edm versus the soft squark masses @xmath100 in the case @xmath101 , the other parameters are taken as in fig .
[ fig4 ] . as the soft squark masses increase ,
the neutron edm @xmath102 varies steeply . because of the interference between the up- and down - quark edms ( see eq .
[ eq16 ] ) , there is in the case of @xmath103 a profound minimum at @xmath104 .
( 0,80)(0,0 ) ( -50,-30 ) ( 0,80)(0,0 ) ( -50,-30 ) ( 0,80)(0,0 ) ( -50,-30 ) ( 0,80)(0,0 ) ( -50,-30 ) let us move to the corrections arising from the diagrams involving virtual charginos . in this case the phase @xmath105 does not contribute to the quark edms at all , whereas @xmath106 contributes at one - loop level ( see eq .
( [ eqa3 - 4 ] ) ) and @xmath19 takes part in the game through only two - loop diagrams ( see eq .
( [ eq12 ] ) ) . at the one - loop level ,
the chargino contributions to the quark edms are proportional to the suppression factor @xmath107 . however , the last terms of the two - loop `` gluino - chargino '' corrections to the quark edms and cedms , given in eq .
( [ eq12 ] ) , do not depend on this suppression factor . taking @xmath108 , we plot in fig .
[ fig7 ] the neutron edm as a function of the soft squark masses @xmath97 for a large @xmath91 .
the dash lines represent the one - loop `` chargino - squark '' results and the solid lines correspond to the sum of the one - loop `` chargino - squark '' and two - loop `` chargino - gluino '' results . as the dominant two - loop `` gluino - chargino '' corrections are not proportional to the suppression factor given above , they play the leading role . the corresponding figure for a low @xmath91
is presented in fig .
[ fig8 ] . just as in the neutralino case ,
the cp phase @xmath19 affects also in the chargino case the quark edms only at the two - loop level . in fig .
[ fig9 ] we show the variation of the neutron edm with respect to @xmath97 in the case @xmath109 and @xmath96 .
even when the squark masses are larger than @xmath110 , the neutron edm induced by the cp phase @xmath19 can be close to @xmath111 , which is the present upper bound from experiments .
( 0,80)(0,0 ) ( -50,-30 ) ( 0,80)(0,0 ) ( -50,-30 ) by including all the contributions present in eq . ( [ eq15 ] )
, we now determine in our theoretical framework the variation of the neutron edm as a function of the cp phase @xmath19 .
taking @xmath112 and @xmath113 , we plot in fig .
[ fig10 ] the neutron edm versus the cp phase @xmath19 , where the dashed line is the sum of the one - loop results and the contributions of the weinberg operator , and the solid line represents the results including two - loop gluino corrections to the quark dipole moment operators .
[ fig11 ] gives the corresponding plot for the case @xmath103 .
it is clear that the two - loop gluino corrections can reach the @xmath114 level for our choice of the parameters . in this work
, we do not consider the effects of the cp phases associated with the trilinear soft squark yukawa coupling .
this is reflected in our numerical results as a relatively strong decrease of the neutron edm as a function increasing squark masses .
moreover , for the model we employ here , the mass of the lightest higgs boson sets a strong constraint on the parameter space of the new physics . as indicated in the literature @xcite
, the cp violation would cause changes to the neutral - higgs - quark coupling , neutral higgs - gauge - boson coupling and self - coupling of higgs boson .
the present experimental lower bound for the mass of the lightest higgs boson is relaxed to 60 gev . in our numerical analysis
we have taken this constraint for the parameter space into account .
we have investigated in this work the corrections induced by some two - loop gluino diagrams to the neutron edm . except a loop factor
, the two - loop contributions do not involve other suppression factors . since the dependence of the two - loop corrections on the relevant cp violating phases differs from that of the one - loop corrections
, there is a region in the parameter space , where the two - loop corrections are comparable with the one - loop corrections .
certainly , the present experimental upper bound on the neutron edm would set a very rigorous constraint on the possible values of cp phases if the masses of new particles are in the tev range . in order to circumvent this constraint ,
one can assume , as was done in the corresponding analysis at one - loop level , that the squarks of the first two generations are heavy , _
i.e. _ @xmath115 , or invoke a mechanism that causes an effective cancellation among different contributions to the quark edms . because the exact analytical expressions for the two - loop vacuum integrals are derivable
, one can extract , at least in principle , the wilson coefficients of the dipole moment operators from any two - loop triangle diagrams and thereby find their contributions to the fermion anomalous magnetic dipole moments and edms .
we will address our analysis to this issue in our coming works .
the work has been supported by the academy of finland under the contracts no .
104915 and 107293 , and partly also by the national natural sceince foundation of china .
@xmath116 + { d-2\over d}{q_1\cdot(q_2-q_1)\over(q_2-q_1)^2-m_{_q}^2 } \;,\nonumber\\ % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % & & { \cal n}_{_{\tilde{g}(2)}}^\gamma={2\over d } { q_2\cdot(q_2-q_1)\over q_2 ^ 2-m_{_{\tilde{q}_i}}^2}+{4\over d } { q_1 ^ 2q_2\cdot(q_2-q_1)\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)^2 } -{q_2\cdot(q_2-q_1)\over q_1 ^ 2-m_{_{\tilde{q}_j}}^2 } \nonumber\\ & & \hspace{1.4 cm } + { 4\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)(q_2 ^ 2-|m_{_{\tilde g}}|^2 ) } \big[{q_1\cdot q_2q_2 ^ 2\over d } -{d(q_1\cdot q_2)^2-q_1 ^ 2q_2 ^ 2\over d(d-1)}\big ] \nonumber\\ & & \hspace{1.4 cm } + { 4\over d}{q_2 ^
2q_2\cdot(q_2-q_1)\over(q_2 ^ 2-|m_{_{\tilde g}}|^2)^2 } -{2+d\over d}{q_2\cdot(q_2-q_1)\over q_2 ^ 2-|m_{_{\tilde g}}|^2 } + { 2-d\over d}{q_2\cdot(q_2-q_1)\over(q_2-q_1)^2-m_{_q}^2 } \;,\nonumber\\ % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % & & { \cal n}_{_{\tilde{g}(1)}}^g=-{4\over d}{q_1 ^ 2q_1\cdot ( q_2-q_1)\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)^2}+{17\over8}{q_1\cdot(q_2-q_1)\over q_1 ^ 2-m_{_{\tilde{q}_j}}^2}-{4\over d}{q_1\cdot(q_2-q_1)q_2 ^ 2\over ( q_2 ^ 2-|m_{_{\tilde g}}|^2)^2 } \nonumber\\ & & \hspace{1.4 cm } + { 4\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2 ) ( q_2 ^ 2-|m_{_{\tilde g}}|^2)}\big[{q_1 ^ 2q_1\cdot q_2\over d } -{d(q_1\cdot q_2)^2-q_1 ^ 2q_2 ^ 2\over d(d-1)}\big ] \nonumber\\ & & \hspace{1.4 cm } -{1\over8}{q_1\cdot(q_2-q_1)\over q_2 ^ 2-|m_{_{\tilde g}}|^2 } -{18 - 9d\over8d}\big[{q_2\cdot(q_2-q_1)\over q_2 ^ 2-|m_{_{\tilde g}}|^2 } + { q_2\cdot(q_2-q_1)\over q_2 ^ 2-m_{_{\tilde{q}_i}}^2}\big ] \nonumber\\ & & \hspace{1.4 cm } + { 10\over8}{2-d\over d}{q_1\cdot(q_2-q_1)\over(q_2-q_1)^2-m_{_q}^2 } \;,\nonumber\\ % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % & & { \cal n}_{_{\tilde{g}(2)}}^g={4\over d } { q_1 ^ 2q_2\cdot(q_2-q_1)\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)^2 } + { 4\over d}{q_2 ^
2q_2\cdot(q_2-q_1)\over(q_2 ^ 2-|m_{_{\tilde g}}|^2)^2 } -{q_2\cdot(q_2-q_1)\over q_1 ^ 2-m_{_{\tilde{q}_j}}^2 } \nonumber\\ & & \hspace{1.4 cm } -{34-d\over8d}{q_2\cdot(q_2-q_1)\over q_2 ^ 2-|m_{_{\tilde g}}|^2 } + { 4\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)(q_2 ^ 2-|m_{_{\tilde g}}|^2 ) } \big[{q_1\cdot q_2q_2
^ 2\over d } \nonumber\\ & & \hspace{1.4 cm } -{d(q_1\cdot q_2)^2-q_1 ^ 2q_2 ^ 2\over d(d-1)}\big ] -{1\over
4d}{q_2\cdot(q_2-q_1)\over q_2 ^ 2-m_{_{\tilde{q}_i}}^2 } + { 9\over8}{q_2\cdot(q_2-q_1)\over q_1 ^ 2-|m_{_{\tilde g}}|^2 } \nonumber\\ & & \hspace{1.4 cm } -{10\over8}{2-d\over d}{q_2\cdot(q_2-q_1)\over(q_2-q_1)^2-m_{_q}^2 } \;,\nonumber\\ % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % & & { \cal n}_{_{\chi_{_k}^0(1)}}^{g(a)}= -{4\over d}{q_1 ^ 2q_1\cdot(q_2-q_1)\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)^2 } + { q_1\cdot(q_2-q_1)\over q_1 ^ 2-m_{_{\tilde{q}_j}}^2 } -{4\over d}{q_1\cdot(q_2-q_1)q_2 ^
2\over(q_2 ^ 2-|m_{_{\tilde g}}|^2)^2 } \nonumber\\ & & \hspace{1.4 cm } -{1\over8}{q_1\cdot(q_2-q_1)\over q_2 ^ 2-|m_{_{\tilde g}}|^2 } + { 4\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2 ) ( q_2 ^ 2-|m_{_{\tilde g}}|^2)}\big[{q_1 ^ 2q_1\cdot q_2\over d } \nonumber\\ & & \hspace{1.4 cm } -{d(q_1\cdot q_2)^2-q_1 ^ 2q_2 ^ 2\over d(d-1)}\big ] + { 1\over8}{2-d\over d}{q_1\cdot(q_2-q_1)\over(q_2-q_1)^2-m_{_q}^2 } \;,\nonumber\\ % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % & & { \cal n}_{_{\chi_{_k}^0(2)}}^{g(a)}={4\over d } { q_1 ^ 2q_2\cdot(q_2-q_1)\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)^2 } + { 4\over d}{q_2 ^ 2q_2\cdot(q_2-q_1)\over(q_2 ^ 2-|m_{_{\tilde g}}|^2)^2 } -{q_2\cdot(q_2-q_1)\over q_1 ^ 2-m_{_{\tilde{q}_j}}^2 } \nonumber\\ & & \hspace{1.4 cm } -{34-d\over8d}{q_2\cdot(q_2-q_1)\over q_2 ^ 2-|m_{_{\tilde g}}|^2 } + { 4\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)(q_2 ^ 2-|m_{_{\tilde g}}|^2 ) } \big[{q_1\cdot q_2q_2 ^ 2\over d } \nonumber\\ & & \hspace{1.4 cm } -{d(q_1\cdot q_2)^2-q_1 ^ 2q_2 ^ 2\over d(d-1)}\big ] -{1\over 4d}{q_2\cdot(q_2-q_1)\over q_2 ^ 2-m_{_{\tilde{q}_i}}^2 } -{1\over8}{2-d\over d}{q_2\cdot(q_2-q_1)\over(q_2-q_1)^2-m_{_q}^2 } \;,\nonumber\\ % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % & & { \cal n}_{_{\chi_{_k}^0(1)}}^{g(b)}=-{4\over d}{q_1 ^ 2q_1\cdot ( q_2-q_1)\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)^2}+{17\over8 } { q_1\cdot(q_2-q_1)\over q_1 ^ 2-m_{_{\tilde{q}_j}}^2 } -{4\over d}{q_1\cdot(q_2-q_1)q_2
^ 2\over(q_2 ^ 2-m_{_{\chi_{_k}^0}}^2)^2 } \nonumber\\ & & \hspace{1.4 cm } + { q_1\cdot(q_2-q_1)\over q_2 ^ 2-m_{_{\chi_{_k}^0}}^2 } + { 4\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2 ) ( q_2 ^ 2-m_{_{\chi_{_k}^0}}^2)}\big[{q_1 ^ 2q_1\cdot
q_2\over d } \nonumber\\ & & \hspace{1.4 cm } -{d(q_1\cdot q_2)^2-q_1 ^ 2q_2 ^ 2\over d(d-1)}\big ] + { 1\over8}{2-d\over d}{q_1\cdot(q_2-q_1)\over(q_2-q_1)^2-m_{_q}^2 } \nonumber\\ & & \hspace{1.4 cm } -{18 - 9d\over8d}\big[{q_2\cdot(q_2-q_1)\over q_2 ^ 2-m_{_{\chi_{_k}^0}}^2 } + { q_2\cdot(q_2-q_1)\over q_2 ^ 2-m_{_{\tilde{q}_i}}^2}\big ] \;,\nonumber\\ % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % & & { \cal n}_{_{\chi_{_k}^0(2)}}^{g(b)}={4\over d } { q_1 ^ 2q_2\cdot(q_2-q_1)\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)^2 } + { 4\over d}{q_2 ^ 2q_2\cdot(q_2-q_1)\over(q_2 ^ 2-m_{_{\chi_{_k}^0}}^2)^2 } -{q_2\cdot(q_2-q_1)\over q_1 ^ 2-m_{_{\tilde{q}_j}}^2 } \nonumber\\ & & \hspace{1.4 cm } -{2+d\over d}{q_2\cdot(q_2-q_1)\over q_2 ^ 2-m_{_{\chi_{_k}^0}}^2 } + { 4\over(q_1 ^ 2-m_{_{\tilde{q}_j}}^2)(q_2 ^ 2-m_{_{\chi_{_k}^0}}^2 ) } \big[{q_1\cdot q_2q_2
^ 2\over d } \nonumber\\ & & \hspace{1.4 cm } -{d(q_1\cdot q_2)^2-q_1 ^ 2q_2 ^ 2\over d(d-1)}\big ] + { 2\over d}{q_2\cdot(q_2-q_1)\over q_2 ^ 2-m_{_{\tilde{q}_i}}^2 } + { 9\over8}{q_2\cdot(q_2-q_1)\over q_1 ^ 2-|m_{_{\tilde g}}|^2 } \nonumber\\ & & \hspace{1.4 cm } -{1\over8}{2-d\over d}{q_2\cdot(q_2-q_1)\over(q_2-q_1)^2-m_{_q}^2 } \;.
\label{aeq1}\end{aligned}\ ] ]
@xmath118\ ; , \nonumber\\ & & f_4(x_0,x_1,x_2,x_3,x_4)=\big\{\xi_{_a}+\xi_{_b}+\xi_{_c } -{1\over8}\xi_{_d}+\xi_{_e}-{1\over16}\xi_{_f}\big\}(x_0,x_1,x_2,x_3,x_4)\ ; , \nonumber\\ & & f_5(x_0,x_1,x_2,x_3,x_4)=\big\{\xi_{_a}+{15\over16}\xi_{_b}+\xi_{_c } + \xi_{_d}+\xi_{_e}-{1\over16}\xi_{_f}\big\}(x_0,x_2,x_1,x_4,x_3 ) \nonumber\\ & & \hspace{3.8 cm } + { 1\over16}\xi_{_b}(x_0,x_4,x_1,x_2,x_3)\ ; , \nonumber\\ & & f_6(x_0,x_1,x_2,x_3,x_4)=\big\{e_{_q}\big[\xi_{_a}+\xi_{_c } + \xi_{_d}+\xi_{_e}\big]+(2e_{_q}-e_{_q})\xi_{_b } \nonumber\\ & & \hspace{3.8 cm } + { e_{_q}\over2}\xi_{_f } \big\}(x_0,x_1,x_2,x_3,x_4 ) -{e_{_q}-e_{_q}\over2}\big[\xi_{_b}(x_0,x_2,x_1,x_4,x_3 ) \nonumber\\ & & \hspace{3.8 cm } + \xi_{_b}(x_0,x_4,x_1,x_2,x_3)\big]\ ; , \nonumber\\ & & f_7(x_0,x_1,x_2,x_3,x_4)=\big\{e_{_q}\big[\xi_{_a}+{3\over2 } \xi_{_b}+\xi_{_c}+\xi_{_d}+\xi_{_e}\big]+{e_{_q}\over2 } \xi_{_f}\big\}(x_0,x_2,x_1,x_4,x_3 ) \nonumber\\ & & \hspace{3.8 cm } -{e_{_q}\over2}\xi_{_b}(x_0,x_4,x_1,x_2,x_3 ) -(e_{_q}-e_{_q})\xi_{_d}(x_0,x_2,x_3,x_4,x_1)\;. \label{aeq2}\end{aligned}\ ] ] where the functions @xmath119 are formulated as @xmath120 \nonumber\\ & & \hspace{1.0cm}\times \big[{3 + 2\ln x_1\over x_1-x_3}-{2x_1(1 + 2\ln x_1)\over(x_1-x_3)^2 } + { 2(x_1 ^ 2\ln x_1-x_3 ^ 2\ln x_3)\over(x_1-x_3)^2}\big ] \nonumber\\ & & \hspace{1.0 cm } + { x_3(x_0-x_2+x_3)\over(x_1-x_3)^3(x_2-x_4)}\phi(x_0,x_1,x_2 ) -{x_3(x_0+x_3-x_4)\over(x_1-x_3)^3(x_2-x_4)}\phi(x_0,x_1,x_4 ) \nonumber\\ & & \hspace{1.0 cm } + { x_1 ^ 2-x_0x_3 - 2x_1x_3+x_2x_3\over(x_1-x_3)^2(x_2-x_4 ) } { \partial\phi\over\partial x_1}(x_0,x_1,x_2 ) \nonumber\\ & & \hspace{1.0 cm } -{x_1 ^ 2-x_0x_3 - 2x_1x_3+x_3x_4\over(x_1-x_3)^2(x_2-x_4 ) } { \partial\phi\over\partial x_1}(x_0,x_1,x_4 ) \nonumber\\ & & \hspace{1.0 cm } + { x_1(x_0+x_1-x_2)\over2(x_1-x_3)(x_2-x_4 ) } { \partial^2\phi\over\partial^2x_1}(x_0,x_1,x_2 ) \nonumber\\ & & \hspace{1.0 cm } -{x_1(x_0+x_1-x_4)\over2(x_1-x_3)(x_2-x_4)}{\partial^2\phi \over\partial^2x_1}(x_0,x_1,x_4 ) \nonumber\\ & & \hspace{1.0 cm } -{x_3(x_0-x_2+x_3)\over(x_1-x_3)^3(x_2-x_4)}\phi(x_0,x_3,x_2 ) \nonumber\\ & & \hspace{1.0 cm } + { x_3(x_0+x_3-x_4)\over(x_1-x_3)^3(x_2-x_4)}\phi(x_0,x_3,x_4 ) \big\}\ ; , \label{aeq3}\end{aligned}\ ] ] @xmath121\cdot\big[{1+\ln x_1\over x_1-x_3 } \nonumber\\ & & \hspace{1.0 cm } -{x_1\ln x_1-x_3\ln x_3\over(x_1-x_3)^2}\big ] -\big[{2\ln x_1+\ln^2x_1\over2(x_1-x_3)}-{x_1\ln^2x_1-x_3\ln^2x_3 \over2(x_1-x_3)^2}\big ] \nonumber\\ & & \hspace{1.0 cm } -{x_0-x_2+x_3\over2(x_1-x_3)^2(x_2-x_4)}\phi(x_0,x_1,x_2 ) + { x_0+x_1-x_2\over2(x_1-x_3)(x_2-x_4)}{\partial\phi\over\partial x_1 } ( x_0,x_1,x_2 ) \nonumber\\ & & \hspace{1.0 cm } + { x_0+x_3-x_4\over2(x_1-x_3)^2(x_2-x_4)}\phi(x_0,x_1,x_4 ) -{x_0+x_1-x_4\over2(x_1-x_3)(x_2-x_4)}{\partial\phi\over\partial x_1 } ( x_0,x_1,x_4 ) \nonumber\\ & & \hspace{1.0 cm } + { x_0-x_2+x_3\over2(x_1-x_3)^2(x_2-x_4)}\phi(x_0,x_3,x_2 ) -{x_0+x_3-x_4\over2(x_1-x_3)^2(x_2-x_4)}\phi(x_0,x_3,x_4)\big\}\ ; , \label{aeq4}\end{aligned}\ ] ] @xmath122 \nonumber\\ & & \hspace{1.0cm}\times \big[{3 + 2\ln x_2\over x_2-x_4}-{2x_2(1 + 2\ln x_2)\over(x_2-x_4)^2 } + { 2(x_2 ^ 2\ln x_2-x_4 ^ 2\ln x_4)\over(x_2-x_4)^3}\big ] \nonumber\\ & & \hspace{1.0 cm } -{x_4(x_0+x_1-x_4)\over(x_1-x_3)(x_2-x_4)^3}\phi(x_0,x_1,x_2 ) + { x_0x_4+x_2 ^ 2 - 2x_2x_4+x_1x_4\over(x_1-x_3)(x_2-x_4)^2}{\partial\phi \over\partial x_2}(x_0,x_1,x_2 ) \nonumber\\ & & \hspace{1.0 cm } -{x_2(x_0+x_1-x_2)\over2(x_1-x_3)(x_2-x_4)}{\partial^2\phi \over\partial^2 x_2}(x_0,x_1,x_2 ) + { x_4(x_0+x_3-x_4)\over(x_1-x_3)(x_2-x_4)^3}\phi(x_0,x_3,x_2 ) \nonumber\\ & & \hspace{1.0 cm } -{x_0x_4+x_2 ^ 2 - 2x_2x_4+x_3x_4\over(x_1-x_3)(x_2-x_4)^2}{\partial\phi \over\partial x_2}(x_0,x_3,x_2 ) + { x_2(x_0-x_2+x_3)\over2(x_1-x_3)(x_2-x_4)}{\partial^2\phi \over\partial^2 x_2}(x_0,x_3,x_2 ) \nonumber\\ & & \hspace{1.0 cm } + { x_4(x_0+x_1-x_4)\over(x_1-x_3)(x_2-x_4)^3}\phi(x_0,x_1,x_4 ) -{x_4(x_0+x_3-x_4)\over(x_1-x_3)(x_2-x_4)^3}\phi(x_0,x_3,x_4 ) \big\}\ ; , \label{aeq5}\end{aligned}\ ] ] @xmath123 \nonumber\\ & & \hspace{1.0cm}\times \big[{1+\ln
x_2\over x_2-x_4}-{x_2\ln x_2-x_4\ln x_4\over(x_2-x_4)^2}\big ] + \big[{\ln^2x_2 + 2\ln x_2\over2(x_2-x_4 ) } -{x_2\ln^2x_2-x_4\ln^2x_4\over2(x_2-x_4)^2}\big ] \nonumber\\ & & \hspace{1.0 cm } -{x_0+x_1-x_4\over2(x_1-x_3)(x_2-x_4)^2}\phi(x_0,x_1,x_2 ) + { x_0+x_1-x_2\over2(x_1-x_3)(x_2-x_4)}{\partial\phi\over \partial x_2}(x_0,x_1,x_2 ) \nonumber\\ & & \hspace{1.0 cm } + { x_0+x_1-x_4\over2(x_1-x_3)(x_2-x_4)^2}\phi(x_0,x_1,x_4 ) + { x_0+x_3-x_4\over2(x_1-x_3)(x_2-x_4)^2}\phi(x_0,x_3,x_2 ) \nonumber\\ & & \hspace{1.0 cm } -{x_0-x_2+x_3\over2(x_1-x_3)(x_2-x_4)}{\partial\phi\over \partial x_2}(x_0,x_3,x_2 ) -{x_0+x_3-x_4\over2(x_1-x_3)(x_2-x_4)^2}\phi(x_0,x_3,x_4 ) \big\}\ ; , \label{aeq6}\end{aligned}\ ] ] @xmath124 + 3\big[{x_1 + 2x_1\ln x_1\over x_1-x_3 } -{x_1 ^ 2\ln x_1-x_3 ^ 2\ln x_3\over(x_1-x_3)^2}\big]\cdot \big[{1+\ln x_2\over x_2-x_4 } \nonumber\\ & & \hspace{1.0 cm } -{x_2\ln x_2-x_4\ln x_4\over(x_2-x_4)^2}\big ] -{3\over2}\big[{x_2 + 2x_2\ln x_2\over x_2-x_4 } -{x_2 ^ 2\ln x_2-x_4 ^ 2\ln x_4\over(x_2-x_4)^2}\big ] + { \ln x_1 + 2\ln x_1\over x_1-x_3 } \nonumber\\ & & \hspace{1.0 cm } -{(x_1\ln^2x_1-x_3\ln^2x_3)\over(x_1-x_3)^2 } + { \ln x_2 + 2\ln x_2\over x_2-x_4}-{(x_2\ln^2x_2-x_4\ln^2x_4 ) \over(x_2-x_4)^2 } \nonumber\\ & & \hspace{1.0 cm } + { 2[x_1 ^ 2x_4+x_2 ^ 2x_3+(x_0 - 2x_1 - 2x_2)x_3x_4]\over(x_1-x_3)^2(x_2-x_4)^2 } \ln x_1\ln x_2 \nonumber\\ & & \hspace{1.0 cm } -{2[x_1 ^ 2-x_2x_3 - 2x_1x_3+x_0x_3]\over(x_1-x_3)^2(x_2-x_4)}\ln x_1 -{2[x_2 ^ 2-x_1x_4 - 2x_2x_4+x_0x_4]\over(x_1-x_3)(x_2-x_4)^2}\ln x_2 \nonumber\\ & & \hspace{1.0 cm } -{2(x_1+x_2-x_0)\over(x_1-x_3)(x_2-x_4 ) } -{2x_4[x_1 ^ 2+x_3(x_0 - 2x_1-x_4)]\over(x_1-x_3)^2(x_2-x_4)^2 } \ln x_1\ln x_4 \nonumber\\ & & \hspace{1.0 cm } -{2x_4(x_1+x_4-x_0)\over(x_1-x_3)(x_2-x_4)^2}\ln x_4 -{2x_3[x_2 ^ 2+x_4(x_0 - 2x_2-x_3)]\over(x_1-x_3)^2(x_2-x_4)^2 } \ln x_2\ln x_3 \nonumber\\ & & \hspace{1.0 cm } -{2x_3(x_2+x_3-x_0)\over(x_1-x_3)^2(x_2-x_4)}\ln x_3 -{2x_3x_4(x_3+x_4-x_0)\over(x_1-x_3)^2(x_2-x_4)^2}\ln x_3\ln x_4 \nonumber\\ & & \hspace{1.0 cm } + { \phi(x_0,x_1,x_2)\over2(x_1-x_3)^2(x_2-x_4)^2}\big[2x_0 ^ 2+x_1 ^ 2 -2x_2 ^ 2-x_0x_3 - 4x_0x_4 - 2x_1x_3 + 4x_2x_4+x_3x_4\big ] \nonumber\\ & & \hspace{1.0 cm } + { x_1 ^ 2 - 2x_0 ^ 2 + 2x_2 ^ 2+x_0x_1 + 4x_0x_4-x_1x_4 - 4x_2x_4\over 2(x_1-x_3)(x_2-x_4)^2}{\partial\phi\over\partial x_1}(x_0,x_1,x_2 ) \nonumber\\ & & \hspace{1.0 cm } -{2x_0 ^ 2+x_1 ^ 2 + 2x_2 ^ 2 - 4x_0x_2-x_0x_3 - 2x_1x_3+x_2x_3\over 2(x_1-x_3)^2(x_2-x_4)}{\partial\phi\over\partial x_2}(x_0,x_1,x_2 ) \nonumber\\ & & \hspace{1.0 cm } + { 2x_0 ^ 2-x_1 ^ 2 + 2x_2 ^ 2-x_0x_1 - 4x_0x_2-x_1x_2\over2(x_1-x_3)(x_2-x_4 ) } { \partial^2\phi\over\partial x_1\partial x_2}(x_0,x_1,x_2 ) \nonumber\\ & & \hspace{1.0 cm } -{2x_0 ^ 2+x_1 ^ 2 + 2x_4 ^ 2-x_0x_3 - 4x_0x_4 - 2x_1x_3+x_3x_4\over 2(x_1-x_3)^2(x_2-x_4)^2}\phi(x_0,x_1,x_4 ) \nonumber\\ & & \hspace{1.0 cm } + { 2x_0 ^ 2-x_1 ^ 2 + 2x_4 ^ 2-x_0x_1 - 4x_0x_4-x_1x_4\over2(x_1-x_3)(x_2-x_4)^2 } { \partial\phi\over\partial x_1}(x_0,x_1,x_4 ) \nonumber\\ & & \hspace{1.0 cm } -{2x_0 ^ 2 - 2x_2 ^ 2-x_3 ^ 2-x_0x_3 - 4x_0x_4 + 4x_2x_4+x_3x_4\over2(x_1-x_3)^2 ( x_2-x_4)^2}\phi(x_0,x_3,x_2 ) \nonumber\\ & & \hspace{1.0 cm } + { 2x_0 ^ 2 + 2x_2 ^ 2-x_3 ^ 2 - 4x_0x_2-x_0x_3-x_2x_3\over2(x_1-x_3)^2(x_2-x_4 ) } { \partial\phi\over\partial x_2}(x_0,x_3,x_2 ) \nonumber\\ & & \hspace{1.0 cm } + { 2x_0 ^ 2-x_3 ^ 2 + 2x_4 ^ 2-x_0x_3 - 4x_2x_4-x_3x_4\over2(x_1-x_3)^2(x_2-x_4)^2 } \phi(x_0,x_3,x_4)\big\}\ ; , \label{aeq7}\end{aligned}\ ] ] @xmath125 + { x_0+x_1-x_2\over(x_1-x_3)(x_2-x_4)}{\partial\phi\over\partial x_0 } ( x_0,x_1,x_2 ) \nonumber\\ & & \hspace{1.0 cm } -{x_0+x_1-x_4\over(x_1-x_3)(x_2-x_4)}{\partial\phi\over\partial x_0 } ( x_0,x_1,x_4 ) -{x_0+x_3-x_2\over(x_1-x_3)(x_2-x_4)}{\partial\phi\over\partial x_0 } ( x_0,x_3,x_2 ) \nonumber\\ & & \hspace{1.0 cm } + { x_0+x_3-x_4\over(x_1-x_3)(x_2-x_4)}{\partial\phi\over\partial x_0 } ( x_0,x_3,x_4)\big\}\ ; , \label{aeq8}\end{aligned}\ ] ] 99 e. p. shabalin , yad . fiz . * 28 * , 151(1978 ) ; sov .
* 28 * , 75(1978 ) .
i. s. alatrev _ et . al . _ ,
b * 276 * , 242(1992 ) ; k. f. smith _ et .
b * 234 * , 191(1990 ) .
j. ellis , s. ferrara , and d. v. nanopoulos , phys .
b. * 114 * , 231(1982 ) ; w. buchmuller and d. wyler , _ ibid .
_ * 121 * , 321(1983 ) ; j. polchinski and m. b. wise , _ ibid . _ * 125 * , 393(1983 ) .
p. nath , phys .
lett . * 66 * , 2565(1991 ) ; y. kizukuri and n. oshimo , phys .
d. * 46 * , 3025(1992 ) ; * 45 * , 1806(1992 ) .
t. ibrahim , p. nath , phys .
d. * 58 * , 111301(1998 ) ; m. brhlik , g. j. good , g. l. kane , _ ibid . _ * 59 * , 115004(1999 ) .
d. chang , w. y. keung , a. pilaftsis , phys .
, 900(1999 ) ; * 83 * , 3972(1999)(e ) ; a. pilaftsis , phys.lett .
b. * 471 * , 174(1999 ) ; nucl .
phys . b. * 644 * , 263(2002 ) ; d. chang , w. f. chang , and w. f. keung , phys .
b. * 478 * , 239(2000 ) ; t. f. feng , t. huang , x. q. li , s. m. zhao , and x. m. zhang , phys .
d. * 68 * , 016004(2003 ) .
t. ibrahim , p. nath , phys .
b. * 418 * , 98(1998 ) ; j. dai , h. dykstra , r. g. leigh , s. paban , d. a. dicus , phys .
b. * 237 * , 216(1990 ) .
tai - fu feng , phys .
d * 70 * , 096012(2004 ) ( hep - ph/0405192 ) .
a. i. davydychev and j. b. tausk , nucl .
b. * 397 * , 123(1993 ) .
j. dai , h. dykstra , r. g. leigh , s. paban , and d. a. dicus , phys .
b. * 237 * , 216(1990 ) ; d. a. dicus , phys .
d. * 41 * , 999(1990 ) .
r. arnowitt , j. lopez , and d. v. nanopoulos , phys .
d. * 42 * , 2423(1990 ) ; r. arnowitt , m. duff and k. stelle , _ ibid .
_ * 43 * , 3085(1991 ) .
a. manohar , and h. georgi , nucl .
b. * 234 * , 189(1984 ) .
a. pilaftsis , phys .
* d58 * , 096010(1998 ) ; phys . lett . *
b435 * , 88(1998 ) ; a. pilaftsis , c. e. m. wagner , nucl
b533 * , 3(1999 ) ; m. carena , j. ellis , a. pilaftsis , c. e. m. wagner , _
ibid_. * 586 * , 92(2000 ) ; _ ibid_. * 625 * , 345(2002 ) . | we analyze two - loop gluino corrections to the neutron electric dipole moment ( edm ) in the minimal supersymmetry extension of the standard model ( mssm ) .
the dependence of two - loop corrections on the relevant cp violating phases differs from that of the one - loop contributions , and there is a region in the parameter space where the two - loop contributions are comparable with the one - loop contributions .
our numerical results show that the two - loop corrections can be as large as 30% of the one - loop results . |
much prior work on resource allocation in cognitive radio networks has focused on the _ dynamic spectrum access _ model @xcite in which the secondary users seek transmission opportunities for their packets on vacant primary channels in frequency , time , or space . under this model ,
the primary users are assumed to be oblivious of the presence of the secondary users and transmit whenever they have data to send .
secondly , a collision model is assumed for the physical layer in which if a secondary user transmits on a busy primary channel , then there is a collision and both packets are lost .
we considered a similar model in our prior work @xcite where the objective was to design an opportunistic scheduling policy for the secondary users that maximizes their throughput utility while providing tight reliability guarantees on the maximum number of collisions suffered by a primary user over _ any _ given time interval .
we note that this formulation does not consider the possibility of any cooperation between the primary and secondary users .
further , it assumes that the secondary user activity does not affect the primary user channel occupancy process .
there is a growing body of work that investigates alternate models for the interaction between the primary and secondary users in a cognitive radio network . in particular ,
the idea of cooperation at the physical layer has been considered from an information - theoretic perspective in many works ( see @xcite and the references therein ) .
these are motivated by the work on the classical interference and relay channels @xcite . the main idea in these works
is that the resources of the secondary user can be utilized to improve the performance of the primary transmissions . in return , the secondary user can obtain more transmission opportunities for its own data when the primary channel is idle .
these works mainly treat the problem from a physical layer / information - theoretic perspective and do not consider upper layer issues such as queueing dynamics , higher priority for primary user , etc .
recent work that addresses some of these issues includes @xcite .
specifically , @xcite considers the scenario where the secondary user acts as a relay for those packets of the primary user that it receives successfully but which are not received by the primary destination .
it derives the stable throughput of the secondary user under this model .
@xcite use a stackelberg game framework to study spectrum leasing strategies in _ cooperative cognitive radio networks _ where the primary users lease a portion of their licensed spectrum to secondary users in return for cooperative relaying .
@xcite study and compare different physical layer strategies for relaying in such cognitive cooperative systems .
an important consequence of this interaction between the primary and secondary users is that the secondary user activity can now potentially influence the primary user channel occupancy process .
however , there has been little work in studying this scenario .
exceptions include the work in @xcite that considers a two - user setting where collisions caused by the opportunistic transmissions of the secondary user result in retransmissions by the primary user . in this paper , we study the problem of opportunistic cooperation in cognitive networks from a _ network utility maximization _ perspective , specifically taking into account the above mentioned higher - layer aspects . to motivate the problem and illustrate the design issues involved , we first consider a simple network consisting of one primary and one secondary user and their respective access points in sec .
[ section : femto_basic ] . this can model a practical scenario of recent interest , namely a cognitive femtocell @xcite , as discussed in sec .
[ section : femto_basic ] .
we assume that the secondary user can cooperatively transmit with the primary user to increase its transmission success probability . in return , the secondary user can get more opportunities for transmitting its own data when the primary user is idle .
we formulate the problem of maximizing the secondary user throughput subject to time average power constraints in sec .
[ section : femto_objective ] .
unlike most of the prior work on resource allocation in cognitive radio networks , the evolution of the system state for this problem depends on the control actions taken by the secondary user .
here , the system state refers to the channel occupancy state of the primary user . because of this dependence
, this problem becomes a constrained markov decision problem ( mdp ) and the greedy `` drift - plus - penalty '' minimization technique of lyapunov optimization @xcite that we used in @xcite is no longer optimal .
such problems are typically tackled using markov decision theory and dynamic programming @xcite .
for example , @xcite uses these tools to derive structural results on optimal channel access strategies in a similar two - user setting where collisions caused by the opportunistic transmissions of the secondary user cause the primary user to retransmit its packets .
however , this approach requires either extensive knowledge of the dynamics of the underlying network state ( such as state transition probabilities ) or learning based approaches that suffer from large convergence times . instead , in sec .
[ section : solution ] , we use the recently developed framework of maximizing the _ ratio _ of the expected total reward over the expected length of a renewal frame @xcite to design a control algorithm .
this framework extends the classical lyapunov optimization method @xcite to tackle a more general class of mdp problems where the system evolves over renewals and where the length of a renewal frame can be affected by the control decisions during that period .
the resulting solution has the following structure : rather than minimizing a `` drift - plus - penalty '' term every slot , it minimizes a `` drift - plus - penalty ratio '' over each renewal frame .
this can be achieved by solving a sequence of unconstrained _ stochastic shortest path _ ( ssp )
problems and implementing the solution over every renewal frame .
while solving such ssp problems can be simpler than the original constrained mdp , it may still require knowledge of the dynamics of the underlying network state .
learning based techniques for solving such problems by sampling from the past observations have been considered in @xcite .
however , these may suffer from large convergence times .
remarkably , in sec . [ section : solving ] , we show that for our problem , the `` drift - plus - penalty ratio '' method results in an online control algorithm that _ does not require any knowledge of the network dynamics or explicit learning _ , yet is optimal . in this respect , it is similar to the traditional greedy `` drift - plus - penalty '' minimizing algorithms of @xcite .
we then extend the basic model to incorporate multiple secondary users as well as time - varying channels in sec .
[ section : femto_extensions ] . finally , we present simulation results in sec .
[ section : femto_sim ] .
we consider a network with one primary user ( pu ) , one secondary user ( su ) and their respective base stations ( bs ) .
the primary user is the licensed owner of the channel while the secondary user tries to send its own data opportunistically when the channel is not being used by the primary user .
this model can capture a femtocell scenario where the primary user is a legacy mobile user that communicates with the macro base station over licensed spectrum ( fig .
[ fig : femto ] ) .
the secondary user is the femtocell user that does not have any licensed spectrum of its own and tries to send data opportunistically to the femtocell base station over any vacant licensed spectrum .
similar models of _ cooperative cognitive radio networks _ have been considered in @xcite .
this can also model a single server queueing system with two classes of arrivals where one class has a strictly higher priority over the other class .
we consider a time - slotted model .
we assume that the system operates over a frame - based structure .
specifically , the timeline can be divided into successive non - overlapping frames of duration @xmath0 $ ] slots where @xmath1 represents the frame number ( see fig . [
fig : renewal ] ) .
the start time of frame @xmath2 is denoted by @xmath3 with @xmath4 .
the length of frame @xmath2 is given by @xmath0 { \mbox{\raisebox{-.3ex}{$\overset{\vartriangle}{=}$}}}t_{k+1 } - t_k$ ] .
for each @xmath2 , the frame length @xmath0 $ ] is a random function of the control decisions taken during that frame .
each frame can be further divided into two periods : pu idle and pu busy .
the `` pu idle '' period corresponds to the slots when the primary user does not have any packet to send to its base station and is idle .
the `` pu busy '' period corresponds to the slots when the primary user is transmitting its packets to its base station over the licensed spectrum . as shown in fig .
[ fig : renewal ] , every frame starts with the `` pu idle '' period which is followed by the `` pu busy '' period and ends when the primary user becomes idle again . in the basic model , we assume that the primary user receives new packets every slot according to an i.i.d .
bernoulli arrival process @xmath5 with rate @xmath6 packets / slot .
this means that the length of the `` pu idle '' period of any frame is a geometric random variable with parameter @xmath6 .
however , the length of the `` pu busy '' period depends on the secondary user control decisions as discussed below . in any slot @xmath7 ,
if the primary user has a non - zero queue backlog , it transmits one packet to its base station .
we assume that the transmission of each packet takes one slot .
if the transmission is successful , the packet is removed from the primary user queue .
however , if the transmission fails , the packet is retained in the queue for future retransmissions .
the secondary user can not transmit its packets when the channel is being used by the primary user .
it can transmit its packets only during the `` pu idle '' period of the frame and must stop its transmission whenever the primary user becomes active again .
however , the secondary user can transmit cooperatively with the primary user in the `` pu busy '' period to increase its transmission success probability .
this has the effect of decreasing the expected length of the `` pu busy '' period . in order to cooperate , the secondary user must allocate its power resources to help relay the primary user packet .
this cooperation can take place in several ways depending on the cooperative protocol being used ( see @xcite for some examples ) . in this simple model , these details
are captured by the resulting probability of successful transmission .
the reason why the secondary user may want to cooperate is because this can potentially increase the number of time slots in the future in which the primary user does not have any data to send as compared to a non - cooperative strategy .
this can create more opportunities for the secondary user to transmit its own packets .
however , note that the trivial strategy of cooperating whenever possible may lead to a scenario where the secondary user does not have enough power for its own data transmission .
thus , the secondary user needs to decide whether it should cooperate or not considering these two opposing factors .
the probability of a successful primary transmission depends on the control actions such as power allocation and cooperative transmission decisions by the secondary user .
this is discussed in detail in the next section . in this model
, we assume that the network controller can not control the primary user actions . however , it can control the secondary user decisions on cooperation and the associated power allocation .
let @xmath8 represent the primary and secondary user queues respectively in slot @xmath7 .
new packets arrive at the secondary user according to an i.i.d .
process @xmath9 of rate @xmath10 packets / slot respectively .
we assume that there exists a finite constant @xmath11 such that @xmath12 for all @xmath7 .
every slot , an admission control decision determines @xmath13 , the number of new packets to admit into the secondary user queue .
further , every slot , depending on whether the primary user is busy or idle , resource allocation decisions are made as follows .
when @xmath14 , this represents the secondary user decision on cooperative transmission and the corresponding power allocation @xmath15 .
when @xmath16 , this corresponds to the secondary user decision on its own transmission and the corresponding power allocation @xmath15 .
we assume that in each slot , the secondary user can choose its power allocation @xmath15 from a set @xmath17 of possible options .
further , this power allocation is subject to a long - term average power constraint @xmath18 and an instantaneous peak power constraint @xmath19 .
for example , @xmath17 may contain only two options @xmath20 which represents `` remain idle '' and `` cooperate / transmit at full power '' . as another example ,
@xmath21 $ ] such that @xmath15 can take any value between @xmath22 and @xmath19 .
suppose the primary user is active in slot @xmath7 and the secondary user allocates power @xmath23 for cooperative transmission .
then the random success / failure outcome of the primary transmission is given by an indicator variable @xmath24 and the success probability is given by @xmath25 .
the function @xmath26 is known to the network controller and is assumed to be non - decreasing in @xmath27 .
however , the value of the random outcome @xmath24 may not be known beforehand .
note that setting @xmath28 corresponds to a non - cooperative transmission and the success probability for this case becomes @xmath29 and we denote this by @xmath30 .
likewise , we denote @xmath31 by @xmath32 .
thus , @xmath33 for all @xmath34 .
we assume that @xmath6 is such that it can be supported even when the secondary user never cooperates , i.e. , @xmath35 .
this means that the primary user queue is stable even if there is no cooperation .
further , for all @xmath2 , the frame length @xmath0 \geq 1 $ ] and there exist finite constants @xmath36 such that under all control policies , we have : @xmath37\right\ } } \leq t_{max } \end{aligned}\ ] ] specifically , @xmath38 can be chosen to be the expected frame length when the secondary user always cooperates with full power while @xmath39 can be chosen to be the expected frame length when the secondary user never cooperates . using little s theorem
, we have that : @xmath40 similarly , we have : @xmath41 using these , we have : @xmath42 finally , there exists a finite constant @xmath43 such that the expectation of the second moment of a frame size , @xmath44\right\}}$ ] , satisfies the following for all @xmath2 , regardless of the policy : @xmath45\right\ } } \leq d \label{eq : second_moment}\end{aligned}\ ] ] this follows from the assumption that the primary user queue is stable even if there is no cooperation . in appendix c ,
we exactly compute such a @xmath43 that satisfies ( [ eq : second_moment ] ) . when the primary user is idle in slot @xmath7 and the secondary user allocates power @xmath23 for its own transmission , it gets a service rate given by @xmath46 .
this can represent the success probability of a secondary transmission with a bernoulli service process .
this can also be used to model more general service processes .
we assume that there exists a finite constant @xmath47 such that @xmath48 for all @xmath49 .
given these control decisions , the primary and secondary user queues evolve as follows : @xmath50 + a_{pu}(t ) \label{eq : ch4_queues1 } \\
q_{su}(t+1 ) = \max[q_{su}(t ) - \mu_{su}(p(t ) ) , 0 ] + r_{su}(t ) \label{eq : ch4_queues2}\end{aligned}\ ] ] where @xmath51 .
consider any control algorithm that makes admission control decision @xmath13 and power allocation @xmath23 every slot subject to the constraints described in sec .
[ section : femto_decisions ] .
note that if the primary queue backlog @xmath14 , then this power is used for cooperative transmission with the primary user .
if @xmath16 , then this power is used for the secondary user s own transmission . define the following time - averages under this algorithm : @xmath52 where the expectations above are with respect to the potential randomness of the control algorithm .
assuming for the time being that these limits exist , our goal is to design a joint admission control and power allocation policy that maximizes the throughput of the secondary user subject to its average and peak power constraints and the scheduling constraints imposed by the basic model .
formally , this can be stated as a stochastic optimization problem as follows : @xmath53 it will be useful to define the primary queue backlog @xmath54 as the `` state '' for this control problem .
this is because the state of this queue ( being zero or nonzero ) affects the control options as described before .
note that the control decisions on cooperation affect the dynamics of this queue .
therefore , problem ( [ eq : cmdp ] ) is an instance of a constrained markov decision problem @xcite .
it is well known that in order to obtain an optimal control policy , it is sufficient to consider only the class of stationary , randomized policies that take control actions only as a function of the current system state ( and independent of past history ) .
a general control policy in this class is characterized by a stationary probability distribution over the control action set for each system state .
let @xmath55 denote the optimal value of the objective in ( [ eq : cmdp ] ) .
then using standard results on constrained markov decision problems @xcite , we have the following : ( optimal stationary , randomized policy ) : there exists a stationary , randomized policy _ stat _ that takes control decisions @xmath56 every slot purely as a ( possibly randomized ) function of the current state @xmath54 while satisfying the constraints @xmath57 for all @xmath7 and provides the following guarantees : @xmath58 where @xmath59 denote the time - averages under this policy . [ lem : femto_one ] we note that the conventional techniques to solve ( [ eq : cmdp ] ) that are based on dynamic programming @xcite require either extensive knowledge of the system dynamics or learning based approaches that suffer from large convergence times .
motivated by the recently developed extension to the technique of lyapunov optimization in @xcite , we take an different approach to this problem in the next section .
recall that the start of the @xmath60 frame , @xmath3 , is defined as the first slot when the primary user becomes idle after the `` pu busy '' period of the @xmath61 frame .
let @xmath62 denote the secondary user queue backlog at time @xmath3 .
also let @xmath23 be the power expenditure incurred by the secondary user in slot @xmath7 . for notational convenience , in the following we will denote @xmath46 by @xmath63 noting the dependence on @xmath23 is implicit .
then the queueing dynamics of @xmath62 satisfies the following : @xmath64 \nonumber\\ & \qquad + \sum_{t = t_k}^{t_{k+1}-1 } r_{su}(t ) \label{eq : u_su}\end{aligned}\ ] ] where @xmath13 denotes the number of new packets admitted in slot @xmath7 and @xmath65 denotes the start of the @xmath66 frame .
the above expression has an inequality because it may be possible to serve the packets admitted in the @xmath60 frame during that frame itself . in order to meet the time average power constraint
, we make use of a virtual power queue @xmath67 @xcite which evolves over frames as follows : @xmath68 p_{avg } + \sum_{t = t_k}^{t_{k+1}-1 } p(t ) , 0 ] \label{eq : x_su}\end{aligned}\ ] ] where @xmath0 = t_{k+1 } - t_k$ ] is the length of the @xmath60 frame .
recall that @xmath0 $ ] is a ( random ) function of the control decisions taken during the @xmath60 frame . in order to construct an optimal dynamic control policy
, we use the technique of @xcite where a ratio of `` drift - plus - penalty '' is maximized over every frame .
specifically , let @xmath69 denote the queueing state of the system at the start of the @xmath60 frame . as a measure of the congestion in the system
, we use a lyapunov function @xmath70 $ ] . define the drift @xmath71 as the conditional expected change in @xmath72 over the frame @xmath2 : @xmath73 then , using ( [ eq : u_su ] ) and ( [ eq : x_su ] )
, we can bound @xmath71 as follows : @xmath74 | \boldsymbol{q}(t_k)\right\ } } \nonumber\\ & \qquad - x_{su}(t_k ) { \mathbb{e}\left\{t[k ] p_{avg } - \sum_{t = t_k}^{t_{k+1}-1 } p(t ) | \boldsymbol{q}(t_k)\right\ } } \label{eq : femto_drift1}\end{aligned}\ ] ] where @xmath75 is a finite constant that satisfies the following for all @xmath2 and @xmath76 under any control algorithm : @xmath77 p_{avg } \big)^2| \boldsymbol{q}(t_k ) \bigg\ } \end{aligned}\ ] ] using the fact that @xmath78 for all @xmath7 , and using the fact ( [ eq : second_moment ] ) , it follows that choosing @xmath75 as follows satisfies the above : @xmath79}{2 } \label{eq : femto_b}\end{aligned}\ ] ] adding a penalty term @xmath80 ( where @xmath81 is a control parameter that affects a utility - delay trade - off as shown in theorem [ thm : femto_performance ] ) to both sides and rearranging yields : @xmath82 p_{avg } | \boldsymbol{q}(t_k)\right\ } } \nonumber\\ & - { \mathbb{e}\left\{\sum_{t = t_k}^{t_{k+1}-1 } \big(q_{su}(t_k ) \mu_{su}(t ) - x_{su}(t_k ) p(t ) \big ) | \boldsymbol{q}(t_k)\right\ } } \label{eq : femto_drift2}\end{aligned}\ ] ] minimizing the ratio of an upper bound on the right hand side of the above expression and the expected frame length over all control options leads to the following _ frame - based - drift - plus - penalty - algorithm_. in each frame @xmath1 , do the following : 1 .
_ admission control _ : for all @xmath83 , choose @xmath13 as follows : @xmath84 2 .
_ resource allocation _ : choose a policy that maximizes the following ratio : @xmath85 | \boldsymbol{q}(t_k)\right\ } } } \label{eq : femto_resource_alloc}\end{aligned}\ ] ] + specifically , every slot @xmath7 of the frame , the policy observes the queue values @xmath62 and @xmath67 at the beginning of the frame and selects a secondary user power @xmath23 subject to the constraint @xmath34 and the constraint on transmitting own data vs. cooperation depending on whether slot @xmath7 is in the `` pu idle '' or `` pu busy '' period of the frame .
this is done in such a way that the above frame - based ratio of expectations is maximized .
recall that the frame size @xmath0 $ ] is influenced by the policy through the success probabilities that are determined by secondary user power selections .
further recall that these success probabilities are different during the `` pu idle '' and `` pu busy '' periods of the frame .
an explicit policy that maximizes this expectation is given in the next section .
_ queue update _ : after implementing this policy , update the queues as in ( [ eq : ch4_queues2 ] ) and ( [ eq : x_su ] ) . from the above , it can be seen that the admission control part ( [ eq : femto_adm_ctrl ] ) is a simple threshold - based decision that does not require any knowledge of the arrival rates @xmath10 or @xmath6 . in the next section , we present an explicit solution to the maximizing policy for the resource allocation in ( [ eq : femto_resource_alloc ] ) and show that , remarkably , it also does not require knowledge of @xmath10 or @xmath6 and can be computed easily .
we will then analyze the performance of the _ frame - based - drift - plus - penalty - algorithm _ in sec .
[ section : femto_analysis ] .
the policy that maximizes ( [ eq : femto_resource_alloc ] ) uses only two numbers that we call @xmath86 and @xmath87 , defined as follows .
@xmath86 is given by the solution to the following optimization problem : @xmath88 let @xmath89 denote the value of the objective of ( [ eq : p_0_define ] ) under the optimal solution .
then , @xmath87 is given by the solution to the following optimization problem : @xmath90 note that both ( [ eq : p_0_define ] ) and ( [ eq : p_1_define ] ) are simple optimization problems in a single variable and can be solved efficiently . given @xmath86 and @xmath87 , on every slot @xmath7 of frame @xmath2 , the policy that maximizes ( [ eq : femto_resource_alloc ] ) chooses power @xmath23 as follows : @xmath91 that is , the secondary user uses the constant power @xmath86 for its own transmission during the `` pu idle''period of the frame , and uses constant power @xmath87 for cooperative transmission during all slots of the `` pu busy''period of the frame . note that @xmath86 and @xmath87 can be computed easily based on the weights @xmath92 associated with frame @xmath2 , and do not require knowledge of the arrival rates @xmath93 . our proof that the above decisions maximize ( [ eq : femto_resource_alloc ] ) has the following parts :
first , we show that the decisions that maximize the ratio of expectations in ( [ eq : femto_resource_alloc ] ) are the same as the optimal decisions in an equivalent infinite horizon markov decision problem ( mdp ) .
next , we show that the solution to the infinite horizon mdp uses fixed power @xmath94 for each queue state @xmath95 ( for @xmath96 ) .
then , we show that @xmath94 are the same for all @xmath97 .
finally , we show that the optimal powers @xmath86 and @xmath87 are given as above . the detailed proof is given in the next section .
recall that the _ frame - based - drift - plus - penalty - algorithm _ chooses a policy that maximizes the following ratio over every frame @xmath1 @xmath98 | \boldsymbol{q}(t_k)\right\ } } } \label{eq : resource_alloc1}\end{aligned}\ ] ] subject to the constraints described in sec .
[ section : femto_basic ] .
here we examine how to solve ( [ eq : resource_alloc1 ] ) in detail .
first , define the state @xmath99 in any slot @xmath83 as the value of the primary user queue backlog @xmath54 in that slot .
now let @xmath100 denote the class of stationary , randomized policies where every policy @xmath101 chooses a power allocation @xmath102 in each state @xmath99 according to a stationary distribution .
it can be shown that it is sufficient to only consider policies in @xmath100 to maximize ( [ eq : resource_alloc1 ] ) .
now suppose a policy @xmath101 is implemented on a _ recurrent _ system with fixed @xmath62 and @xmath67 and with the same state dynamics as our model .
note that @xmath103 for all @xmath7 when the state @xmath97 .
then , by basic renewal theory @xcite , we have that maximizing the ratio in ( [ eq : resource_alloc1 ] ) is equivalent to the following optimization problem : @xmath104 where @xmath105 is the resulting steady - state probability of being in state @xmath99 in the recurrent system under the stationary , randomized policy @xmath106 and where the expectations above are with respect to @xmath106 .
note that well - defined steady - state probabilities @xmath105 exist for all @xmath101 because we have assumed that @xmath35 so that even if no cooperation is used , the primary queue is stable and the system is recurrent .
thus , solving ( [ eq : resource_alloc1 ] ) is equivalent to solving the _ unconstrained time average maximization problem _ ( [ eq : femto_time_avg ] ) over the class of stationary , randomized policies .
note that ( [ eq : femto_time_avg ] ) is an infinite horizon markov decision problem ( mdp ) over the state space @xmath107 .
we study this problem in the following .
consider the optimal stationary , randomized policy that maximizes the objective in ( [ eq : femto_time_avg ] ) .
let @xmath108 denote the probability distribution over @xmath17 that is used by this policy to choose a power allocation @xmath94 in state @xmath99 .
let @xmath109 denote the resulting effective probability of successful primary transmission in state @xmath97 .
then we have that @xmath110 where @xmath111 denotes the probability of successful transmission in state @xmath99 when the secondary user spends power @xmath94 in cooperative transmission with the primary user .
since the system is stable and has a well - defined steady - state distribution , we can write down the detail equations for the markov chain that describes the state transitions of the system as follows ( see fig .
[ fig : bd ] ) : @xmath112 where @xmath113 denotes the steady - state probability of being in state @xmath99 under this policy .
summing over all @xmath99 yields : @xmath114 the average power incurred in cooperative transmissions under this policy is given by : @xmath115 now consider an alternate stationary policy that uses the following fixed distribution @xmath116 for choosing control action @xmath117 in all states @xmath97 : @xmath118 let @xmath119 denote the resulting effective probability of a successful primary transmission in any state @xmath97 .
note that this is same for all states by the definition ( [ eq : chi ] ) .
then , we have that : @xmath120 let @xmath121 denote the steady - state probability of being in state @xmath99 under this alternate policy .
note that the system is stable under this alternate policy as well .
thus , using the detail equations for the markov chain that describes the state transitions of the system under this policy yields @xmath122 where we used ( [ eq : eq1_new ] ) in the last step .
this implies that @xmath123 and therefore @xmath124 .
also , the average power incurred in cooperative transmissions under this alternate policy is given by : @xmath125 where we used ( [ eq : eq2_new ] ) in the second last step and @xmath123 in the last step .
thus , if we choose @xmath126 in state @xmath127 and choose @xmath128 as defined in ( [ eq : chi ] ) in all other states , it can be seen that _ the alternate policy achieves the same time average value of the objective ( [ eq : femto_time_avg ] ) as the optimal policy_. this implies that to maximize ( [ eq : femto_time_avg ] ) , it is sufficient to optimize over the class of stationary policies that use the _ same _ distribution for choosing @xmath94 for all states @xmath97 .
denote this class by @xmath129 .
then for all @xmath130 , we have that @xmath131 for all @xmath132 . using this and
the fact that @xmath133 , ( [ eq : femto_time_avg ] ) can be simplified as follows : @xmath134 \pi_0(r ) \nonumber\\ & - x_{su}(t_k ) { \mathbb{e}\left\{p_1(r)\right\ } } ( 1 - \pi_0(r ) ) \nonumber \\ \textrm{subject to : } \ ; & r \in \mathcal{r } ' \label{eq : time_avg2}\end{aligned}\ ] ] where @xmath135 is the resulting steady - state probability of being in state @xmath22 and where @xmath136 is the average power incurred in cooperative transmission in state @xmath137 ( same for all states @xmath97 ) .
next , note that the control decisions taken by the secondary user in state @xmath138 do not affect the length of the frame and therefore @xmath135 .
further , the expectations can be removed .
therefore the first term in the problem above can be maximized separately as follows : @xmath139 this is the same as ( [ eq : p_0_define ] ) .
let @xmath86 denote the optimal solution to ( [ eq : time_avg3 ] ) and let @xmath140 denote the value of the objective of ( [ eq : time_avg3 ] ) under the optimal solution .
note that we must have that @xmath141 because the value of the objective when the secondary user chooses @xmath142 ( i.e. , stays idle ) is @xmath22 .
then , ( [ eq : time_avg2 ] ) can be written as : @xmath143 the effective probability of a successful primary transmission in any state @xmath97 is given by @xmath144 . using little s theorem
, we have @xmath145 . using this and rearranging the objective in ( [ eq : time_avg4 ] ) and ignoring the constant terms
, we have the following equivalent problem : @xmath146 it can be shown that it is sufficient to consider only _ deterministic _ power allocations to solve ( [ eq : time_avg5 ] ) ( see , for example , ( * ? ? ? * section 7.3.2 ) ) .
this yields the following problem : @xmath147 this is the same as ( [ eq : p_1_define ] ) .
note that solving this problem does not require knowledge of @xmath6 or @xmath10 and can be solved easily for general power allocation options @xmath17 .
we present an example that admits a particularly simple solution to this problem .
suppose @xmath148 so that the secondary user can either cooperate with full power @xmath19 or not cooperate ( with power expenditure @xmath22 ) with the primary user .
then , the optimal solution to ( [ eq : time_avg6 ] ) can be calculated by comparing the value of its objective for @xmath149 .
this yields the following simple threshold - based rule : @xmath150 we also note that this threshold can be computed without any knowledge of the input rates @xmath151 . to summarize ,
the overall solution to ( [ eq : femto_resource_alloc ] ) is given by the pair @xmath152 where @xmath86 denotes the power allocation used by the secondary user for its own transmission when the primary user is idle and @xmath87 denotes the power used by the secondary user for cooperative transmission .
note that these values remain fixed for the entire duration of frame @xmath2 .
however , these can change from one frame to another depending on the values of the queues @xmath92 .
the computation of @xmath152 can be carried out using a two - step process as follows : 1 .
first , compute @xmath86 by solving problem ( [ eq : time_avg3 ] ) .
let @xmath153 be the value of the objective of ( [ eq : time_avg3 ] ) under the optimal solution @xmath86 .
2 . then compute @xmath87 by solving problem ( [ eq : time_avg6 ] ) .
it is interesting to note that in order to implement this algorithm , the secondary user does not require knowledge of the current queue backlog value of the primary user .
rather , it only needs to know the values of its own queues and whether the current slot is in the `` pu idle '' or `` pu busy '' part of the frame .
this is quite different from the conventional solution to the mdp ( [ eq : cmdp ] ) which is typically a different randomized policy for each value of the state ( i.e. , the primary queue backlog ) .
to analyze the performance of the _ frame - based - drift - plus - penalty - algorithm _ , we compare its lyapunov drift with that of the optimal stationary , randomized policy _ stat _ of lemma [ lem : femto_one ] .
first , note that by basic renewal theory @xcite , the performance guarantees provided by _ stat _ hold over every frame @xmath1 .
specifically , let @xmath3 be the start of the @xmath60 frame .
suppose _ stat _ is implemented over this frame .
then the following hold : @xmath154\right\ } } \upsilon^ * \label{eq : iid_1 } \\ & { \mathbb{e}\left\{\sum_{t = t_k}^{\hat{t}_{k+1}-1 } r^{stat}_{su}(t)\right\ } } \leq { \mathbb{e}\left\{\sum_{t = t_k}^{\hat{t}_{k+1}-1 } \mu^{stat}_{su}(t)\right\ } } \label{eq : iid_2 } \\ & { \mathbb{e}\left\ { \sum_{t = t_k}^{\hat{t}_{k+1}-1 } p^{stat}_{su}(t)\right\ } } \leq { \mathbb{e}\left\{\hat{t}[k]\right\ } } p_{avg } \label{eq : iid_3}\end{aligned}\ ] ] where @xmath155 and @xmath156 $ ] denote the start of the @xmath66 frame and the length of the @xmath60 frame , respectively , under the policy _
similarly , @xmath157 denote the resource allocation decisions under _
stat_. next , we define an alternate control algorithm _ alt
_ that will be useful in analyzing the performance of the _ frame - based - drift - plus - penalty - algorithm_. _ algorithm alt : _ in each frame @xmath1 , do the following : 1 .
_ admission control _ : for all @xmath83 , choose @xmath13 as follows : @xmath158 2 .
_ resource allocation _ : choose a policy that maximizes the following ratio : @xmath159 | \boldsymbol{q}(t_k)\right\ } } } \label{eq : resource_alloc_alt}\end{aligned}\ ] ] 3 .
_ queue update _ : after implementing this policy , update the queues as in ( [ eq : u_su ] ) , ( [ eq : x_su ] ) . by comparing with the _ frame - based - drift - plus - penalty - algorithm _ , it can be see that this algorithm differs only in the admission control part while the resource allocation decisions are exactly the same . specifically , under _ alt _ , the queue backlog @xmath62 at the start of the @xmath60 frame is used for making admission control decisions for the entire duration of that frame . however , under the _ frame - based - drift - plus - penalty - algorithm _ , the queue backlog @xmath160 at the start of _ each _ slot is used for making admission control decisions . note
that since the length of the frame depends only on the resource allocation decisions and they are the same under the two algorithms , it follows that implementing them with the same starting backlog @xmath76 yields the same frame lengths .
the following lemma compares the value of the second term in the lyapunov drift bound ( [ eq : femto_drift2 ] ) that corresponds to the admission control decisions under these two algorithms .
let @xmath161 and @xmath162 denote the admission control decisions made by the _ frame - based - drift - plus - penalty - algorithm _ and the _ alt _ algorithm respectively for all @xmath83 .
then we have : @xmath163 where @xmath164 is a constant that does not depend on @xmath165 .
[ lem : adm_ctrl_bound_lem ] see appendix a. we are now ready to characterize the performance of the _ frame - based - drift - plus - penalty - algorithm_. ( performance theorem ) suppose the _ frame - based - drift - plus - penalty - algorithm _ is implemented over all frames @xmath1 with initial condition @xmath166 and with a control parameter @xmath81 .
let @xmath167 denote the resource allocation decisions under this algorithm .
then , we have : 1 .
the secondary user queue backlog @xmath160 is upper bounded for all @xmath7 : @xmath168 2 .
the virtual power queue @xmath67 is mean rate stable , i.e. , @xmath169 further , we have : @xmath170\right\ } } } \leq p_{avg } \label{eq : x_max3}\end{aligned}\ ] ] 3 .
the time - average secondary user throughput ( defined over frames ) satisfies the following bound for all @xmath171 : @xmath172\right\ } } } \geq \upsilon^ * - \frac{b + c}{v t_{min } } \label{eq : femto_utility_bound}\end{aligned}\ ] ] where @xmath173}{2}$ ] and @xmath174 .
[ thm : femto_performance ] theorem [ thm : femto_performance ] shows that the time - average secondary user throughput can be pushed to within @xmath175 of the optimal value with a trade - off in the worst case queue backlog . by little s theorem
, this leads to an @xmath176 utility - delay tradeoff .
part ( 1 ) : we argue by induction .
first , note that ( [ eq : u_max ] ) holds for @xmath177 .
next , suppose @xmath178 for some @xmath179 .
we will show that @xmath180 .
we have two cases .
first , suppose @xmath181 .
then , by ( [ eq : u_su ] ) , the maximum that @xmath160 can increase is @xmath11 so that @xmath182
. next , suppose @xmath183 .
then , the admission control decision ( [ eq : femto_adm_ctrl ] ) chooses @xmath184 .
thus , by ( [ eq : u_su ] ) , we have that @xmath185 for this case as well .
combining these two cases proves the bound ( [ eq : u_max ] ) . parts ( 2 ) and ( 3 ) : see appendix b.
we consider two extensions to the basic model of sec . [ section : femto_basic ] .
consider the scenario with one primary user as before , but with @xmath186 secondary users .
the primary user channel occupancy process evolves as before where the secondary users can transmit their own data only when the primary user is idle .
however , they may cooperatively transmit with the primary user to increase its transmission success probability . in general
, multiple secondary users may cooperatively transmit with the primary in one timeslot .
however , for simplicity , here we assume that at most one secondary user can take part in a cooperative transmission per slot .
further , we also assume that at most one secondary user can transmit its data when the primary user is idle .
our formulation can be easily extended to this scenario .
let @xmath187 denote the set of power allocation options for secondary user @xmath99 .
suppose each secondary user @xmath99 is subject to average and peak power constraints @xmath188 and @xmath189 respectively .
also , let @xmath190 denote the success probability of the primary transmission when secondary user @xmath99 spends power @xmath27 in cooperative transmission .
now consider the objective of maximizing the sum total throughput of the secondary users subject to each user s average and peak power constraints and the scheduling constraints of the model . in order to apply the `` drift - plus - penalty '' ratio method
, we use the following queues : @xmath191 + \sum_{t = t_k}^{t_{k+1}-1 } r_{i}(t ) \label{eq : u_su_i } \\ & x_{i}(t_{k+1 } ) = \max[x_{i}(t_k ) - t[k ] p_{avg , i } + \sum_{t = t_k}^{t_{k+1}-1 } p_{i}(t ) , 0 ] \label{eq : x_su_i}\end{aligned}\ ] ] where @xmath192 is the queue backlog of secondary user @xmath99 at the beginning of the @xmath60 frame , @xmath193 is the service rate of secondary user @xmath99 in slot @xmath7 , @xmath194 and @xmath195 denote the number of new packets admitted and the power expenditure incurred by the secondary user @xmath99 in slot @xmath7 .
finally , @xmath65 denotes the start of the @xmath66 frame and @xmath0 = t_{k+1 } - t_k$ ] is the length of the @xmath60 frame as before .
let @xmath196 denote the queueing state of the system at the start of the @xmath60 frame . using a lyapunov function @xmath197 $ ] and following the steps in sec .
[ section : solution ] yields the following _ multi - user frame - based - drift - plus - penalty - algorithm_. in each frame @xmath1 , do the following : 1 .
_ admission control _ : for all @xmath83 , for each secondary user @xmath198 , choose @xmath194 as follows : @xmath199 where @xmath200 is the number of new arrivals to secondary user @xmath99 in slot @xmath7 .
resource allocation _ : choose a policy that maximizes the following ratio : @xmath201 | \boldsymbol{q}(t_k)\right\ } } } \label{eq : resource_alloc_i}\end{aligned}\ ] ] 3 .
_ queue update _ : after implementing this policy , update the queues as in ( [ eq : u_su_i ] ) and ( [ eq : x_su_i ] ) .
similar to the basic model , this algorithm can be implemented without any knowledge of the arrival rates @xmath202 or @xmath6 .
further , using the techniques developed in sec .
[ section : solving ] , it can be shown that the solution to ( [ eq : resource_alloc_i ] ) can be computed in two steps as follows .
first , we solve the following problem for each @xmath198 : @xmath203 let @xmath86 denote the optimal solution to ( [ eq : time_avg3_i ] ) achieved by user @xmath204 and let @xmath153 denote the optimal objective value .
this means user @xmath204 transmits on all idle slots of frame @xmath2 with power @xmath86 .
next , to determine the optimal cooperative transmission strategy , we solve the following problem for each @xmath198 : @xmath205 let @xmath87 denote the optimal solution to ( [ eq : time_avg6_i ] ) achieved by user @xmath206 .
this means user @xmath206 cooperatively transmits on all busy slots of frame @xmath2 with power @xmath87 .
next , suppose there is an additional channel fading process @xmath207 that takes values from a finite set @xmath208 in an i.i.d fashion every slot .
we assume that in every slot , @xmath209 = q_s$ ] for all @xmath210 . the success probability with cooperative transmission now is a function of both the power allocation and the fading state in that slot .
specifically , suppose the primary user is active in slot @xmath7 and the secondary user allocates power @xmath23 for cooperative transmission .
also suppose @xmath211 .
then the random success / failure outcome of the primary transmission is given by an indicator variable @xmath212 and the success probability is given by @xmath213 .
the function @xmath214 is known to the network controller for all @xmath210 and is assumed to be non - decreasing in @xmath27 for each @xmath210 . for simplicity , we assume that the secondary user transmission rate @xmath63 depends only on @xmath23 . by applying the `` drift - plus - penalty '' ratio method to this extended model , we get the following control algorithm .
the admission control remains the same as ( [ eq : femto_adm_ctrl ] ) .
the resource allocation part involves maximizing the ratio in ( [ eq : femto_resource_alloc ] ) .
using the same arguments as before in sec .
[ section : solving ] , it can be shown that maximizing this ratio is equivalent to the following optimization problem : @xmath215 where @xmath216 is the resulting steady - state probability of being in state @xmath217 in the recurrent system under the stationary , randomized policy @xmath106 and where the expectations above are with respect to @xmath106 .
we study this problem in the following .
consider the optimal stationary , randomized policy that maximizes the objective in ( [ eq : time_avg_fading ] ) .
let @xmath218 denote the probability distribution over @xmath17 that is used by this policy to choose a control action @xmath219 in state @xmath217 .
let @xmath220 denote the resulting effective probability of successful primary transmission in state @xmath217 where @xmath97 . since the system is stable under any stationary policy , @xmath221 .
thus , we get : @xmath222 where @xmath223 denotes the steady - state probability of being in state @xmath224 under this policy .
note that the system is stable and has a well - defined steady - state distribution .
the average power incurred in cooperative transmissions under this policy is given by : @xmath225 now consider an alternate stationary policy that , for each @xmath210 , uses the following fixed distribution @xmath226 for choosing control action @xmath227 in all states @xmath217 where @xmath97 : @xmath228 for each @xmath210 ,
let @xmath229 denote the resulting effective probability of a successful primary transmission in any state @xmath217 where @xmath97 under this policy .
note that this is same for all states @xmath217 where @xmath97 by the definition ( [ eq : chi_fading ] ) .
then , we have that : @xmath230 let @xmath231 denote the steady - state probability of being in state @xmath224 under this alternate policy . since the system is stable under any stationary policy , total incoming rate = total outgoing rate .
thus , we get : @xmath232 \bigg ( \sum_{k \geq 1 } \pi_{k , s } ' \bigg ) \label{eq : ipop_fading}\end{aligned}\ ] ] where we used ( [ eq : mu_s_fading ] ) in the last step . since @xmath207 is i.i.d . , for any @xmath233 ,
we have that @xmath234 similarly , we have : @xmath235 using this , for any @xmath233 , we have : @xmath236 using this in ( [ eq : ipop_fading ] ) , we have for each @xmath237 : @xmath238 \frac{\sum_{k \geq 1 } \pi_{k,\hat{s}}'}{\sum_{j \geq 1 } \pi_{j , \hat{s } } } = \lambda_{pu } \frac{\sum_{k \geq 1 } \pi_{k,\hat{s}}'}{\sum_{j \geq 1 } \pi_{j , \hat{s}}}\end{aligned}\ ] ] where we used ( [ eq : eq1_new_fading ] ) in the last step .
this implies that @xmath239 for every @xmath237 and therefore @xmath124 .
also , the average power incurred in cooperative transmissions under this alternate policy is given by : @xmath240 where we used the fact that @xmath241 for all @xmath242 .
thus , if we choose @xmath126 in state @xmath127 and choose @xmath243 as defined in ( [ eq : chi_fading ] ) in all states @xmath217 where @xmath97 , it can be seen that the alternate policy achieves the same time average value of the objective ( [ eq : time_avg_fading ] ) as the optimal policy .
this implies that to maximize ( [ eq : time_avg_fading ] ) , it is sufficient to optimize over the class of stationary policies that , for each @xmath210 , use the _ same _ distribution for choosing @xmath244 for all states @xmath217 where @xmath97 .
denote this class by @xmath129 . using this and
the fact that @xmath245 for all @xmath242 , ( [ eq : time_avg_fading ] ) can be simplified as follows : @xmath246 \pi_0(r ) \nonumber\\ & - x_{su}(t_k ) \sum_{s
\in \mathcal{s } } { \mathbb{e}\left\{p_{s}(r)\right\ } } ( 1 - \pi_0(r))q_s \nonumber \\ \textrm{subject to : } \ ; & r \in \mathcal{r } ' \label{eq : time_avg2_fading}\end{aligned}\ ] ] where @xmath135 is the resulting steady - state probability of being in state @xmath22 and where @xmath247 is the average power incurred in cooperative transmission in any state @xmath224 with @xmath97 .
using the same arguments as before , the solution to ( [ eq : time_avg2_fading ] ) can be obtained in two steps as follows .
we first compute the solution to ( [ eq : time_avg3 ] ) as before . denoting its optimal value by @xmath153 , ( [ eq : time_avg2_fading ] )
can be written as : @xmath248 using little s theorem , we have @xmath249 . using this and rearranging the objective in ( [ eq : time_avg4_fading ] ) and ignoring the constant terms
, we have the following equivalent problem : @xmath250 it can be shown that it is sufficient to consider only _ deterministic _ power allocations to solve ( [ eq : time_avg5_fading ] ) ( see , for example , ( * ? ? ?
* section 7.3.2 ) ) .
this yields the following problem : @xmath251 note that solving this problem does not require knowledge of @xmath6 or @xmath10 and can be solved efficiently for general power allocation options @xmath17 .
in this section , we evaluate the performance of the _ frame - based - drift - plus - penalty - algorithm _ using simulations .
we consider the network model as discussed in sec .
[ section : femto_basic ] with one primary and one secondary user .
the set @xmath17 consists of only two options @xmath252 .
we assume that @xmath253 and @xmath254 .
we set @xmath255 and @xmath256 . for simplicity , we assume that @xmath257 . in the first set of simulations , we fix the input rates @xmath258 packets / slot . for these parameters
, we can compute the optimal offline solution by linear programming .
this yields the maximum secondary user throughput as @xmath259 packets / slot .
we now simulate the _ frame - based - drift - plus - penalty - algorithm _ for different values of the control parameter @xmath165 over @xmath260 frames . in fig .
[ fig : femto_sim1 ] , we plot the average throughput achieved by the secondary user over this period .
it can be seen that the average throughput increases with @xmath165 and converges to the optimal value @xmath259 packets / slot , with the difference exhibiting a @xmath175 behavior as predicted by theorem [ thm : femto_performance ] . in fig .
[ fig : femto_sim2 ] , we plot the average queue backlog of the secondary user over this period
. it can be see that the average queue backlog grows linearly in @xmath165 , again as predicted by theorem [ thm : femto_performance ] .
also , for all @xmath165 , the average secondary user power consumption over this period was found not to exceed @xmath253 units / slot . for comparison
, we also simulate three alternate algorithms . in the first algorithm `` no cooperation '' , the secondary user never cooperates with the primary user and only attempts to maximize its throughput over the resulting idle periods .
the secondary user throughput under this algorithm was found to be @xmath261 packets / slot as shown in fig .
[ fig : femto_sim1 ] . note that using little s theorem , the resulting fraction of time the primary user is idle is @xmath262 .
this limits the maximum secondary user throughput under the `` no cooperation '' case to @xmath261 packets / slot . in the second algorithm
, we consider the `` always cooperate '' case where the secondary user always cooperates with the primary user .
for the example under consideration , this uses up all the secondary user power and thus , the secondary user achieves zero throughput . in the third algorithm `` counter based policy '' ,
a running average of the total secondary user power consumption so far is maintained . in each slot , the secondary user decides to transmit / cooperate only if this running average is smaller than @xmath18 .
the maximum secondary user throughput under this algorithm was found to be @xmath263 packets / slot .
this demonstrates that simply satisfying the average power constraint is not sufficient to achieve maximum throughput .
for example , it may be the case that under the `` counter based policy '' , the running average condition is usually satisfied when the primary user is busy .
this causes the secondary user to cooperate . however , by the time the primary user next becomes idle , the running average exceeds @xmath18 so that the secondary user does not transmit its own data .
in contrast , the _ frame - based - drift - plus - penalty - algorithm _ is able to find the opportune moments to cooperate / transmit optimally . in the second set of simulations , we fix the input rate @xmath264 packets / slot , @xmath265 , and
simulate the _ frame - based - drift - plus - penalty - algorithm _ over @xmath260 frames . at the start of the simulation , we set @xmath266 packets / slot .
the values of the other parameters remain the same
. however , during the course of the simulation , we change @xmath6 to @xmath267 packets
/ slot after the first @xmath268 frames and then again to @xmath269 packets / slot after the first @xmath270 frames . in figs .
[ fig : femto_sim3 ] and [ fig : femto_sim4 ] , we plot the running average ( over @xmath271 frames ) of the secondary user throughput and the average power used for cooperation .
these show that the _ frame - based - drift - plus - penalty - algorithm _ automatically adapts to the changes in @xmath6 .
further , it quickly approaches the optimal performance corresponding to the new @xmath6 by adaptively spending more or less power ( as required ) on cooperation .
for example , when @xmath6 reduces to @xmath267 packets / slot after frame number @xmath268 , the fraction of time the primary is idle even with no cooperation is @xmath272 . with @xmath273 ,
there is no need to cooperate anymore .
this is precisely what the _ frame - based - drift - plus - penalty - algorithm _ does as shown in fig .
[ fig : femto_sim4 ] .
similarly , when when @xmath6 increases to @xmath269 packets / slot after frame number @xmath270 , the _ frame - based - drift - plus - penalty - algorithm _ starts to spend more power on cooperative transmissions .
in this paper , we studied the problem of opportunistic cooperation in a cognitive femtocell network .
specifically , we considered the scenario where a secondary user can cooperatively transmit with the primary user to increase its transmission success probability . in return , the secondary user can get more opportunities for transmitting its own data when the primary user is idle .
a key feature of this problem is that here , the evolution of the system state depends on the control actions taken by the secondary user .
this dependence makes it a constrained markov decision problem traditional solutions to which require either extensive knowledge of the system dynamics or learning based approaches that suffer from large convergence times .
however , using the technique of lyaunov optimization , we designed a novel greedy and online control algorithm that overcomes these challenges and is provably optimal .
let @xmath274 denote the queue backlog value under the _ frame - based - drift - plus - penalty - algorithm _ for all @xmath275 . then
, since the admission control decision ( [ eq : femto_adm_ctrl ] ) of the _ frame - based - drift - plus - penalty - algorithm _ minimizes the term @xmath276 for all @xmath160 , we have : @xmath277 note that we are not implementing the admission control decisions of _ alt _ in the left hand side of the above .
next , we make use of the following sample path relations in ( [ eq : ineq2_new ] ) to prove ( [ eq : adm_ctrl_bound ] ) . for all @xmath275 , the following hold under any control algorithm :
@xmath278 ( [ eq : sample1 ] ) follows by noting that the maximum number of arrivals to the secondary user queue in the interval @xmath279 is at most @xmath280 .
similarly , ( [ eq : sample2 ] ) follows by noting that the maximum number of departures from the secondary user queue in the interval @xmath279 is at most @xmath281 . using ( [ eq : sample1 ] ) in the left hand side of ( [ eq : ineq2_new ] ) yields : @xmath282 using the fact that @xmath283 and @xmath284(t[k]-1)}{2}$ ] , we get : @xmath285 next , using ( [ eq : sample2 ] ) in the right hand side of ( [ eq : ineq2_new ] ) yields : @xmath286 again using the fact that @xmath287 and @xmath288 ) = \frac{t[k](t[k]-1)}{2}$ ] , we get : @xmath289 using ( [ eq : ineq1_new ] ) and ( [ eq : ineq3_new ] ) in ( [ eq : ineq2_new ] ) , we have : @xmath290
we prove parts ( 2 ) and ( 3 ) of theorem [ thm : femto_performance ] using the technique of lyapunov optimization . using ( [ eq : femto_drift2 ] ) , a bound on the lyapunov drift under the _ frame - based - drift - plus - penalty - algorithm _ is given by : @xmath291 p_{avg } | \boldsymbol{q}(t_k)\right\ } } \nonumber\\ & - { \mathbb{e}\left\{\sum_{t = t_k}^{t_{k+1}-1 } ( q_{su}(t_k ) \mu_{su}^{fab}(t ) - x_{su}(t_k ) p_{su}^{fab}(t ) ) | \boldsymbol{q}(t_k)\right\ } } \label{eq : drift3}\end{aligned}\ ] ] using lemma [ lem : adm_ctrl_bound_lem ] , we have that : @xmath292 next , note that under the _ alt _ algorithm , we have : @xmath293 | \boldsymbol{q}(t_k)\right\ } } } \\ & \leq \frac{{\mathbb{e}\left\{\sum_{t = t_k}^{\hat{t}_{k+1}-1 } ( q_{su}(t_k ) - v ) { r}_{su}^{stat}(t ) | \boldsymbol{q}(t_k)\right\}}}{{\mathbb{e}\left\{\hat{t}[k ] | \boldsymbol{q}(t_k)\right\}}}\end{aligned}\ ] ] to see this , we have two cases : 1 .
@xmath294 : then , @xmath295 for all @xmath296 , so that the left hand side above is @xmath22 while the right hand side is @xmath297 . hence , the inequality follows .
@xmath298 : then , @xmath299 for all @xmath296 , so that the left hand side becomes @xmath300 while the right hand side can not be smaller than @xmath300 . combining these , we get : @xmath301 | \boldsymbol{q}(t_k)\right\}}}{{\mathbb{e}\left\{\hat{t}[k ] | \boldsymbol{q}(t_k)\right\ } } } \ ] ] finally , since the resource allocation part of the _ frame - based - drift - plus - penalty - algorithm _ maximizes the ratio in ( [ eq : femto_resource_alloc ] ) , we have : @xmath302 | \boldsymbol{q}(t_k)\right\}}}{{\mathbb{e}\left\{\hat{t}[k ] | \boldsymbol{q}(t_k)\right\ } } } \ ] ] using these in ( [ eq : drift3 ] ) , we have : @xmath303 | \boldsymbol{q}(t_k)\right\}}}{{\mathbb{e}\left\{\hat{t}[k ] | \boldsymbol{q}(t_k)\right\ } } } \nonumber\\ & - { \mathbb{e}\left\{\sum_{t = t_k}^{\hat{t}_{k+1}-1 } ( q_{su}(t_k){\mu}_{su}^{stat}(t ) - x_{su}(t_k ) { p}_{su}^{stat}(t ) ) | \boldsymbol{q}(t_k)\right\ } } \\ & \times \frac{{\mathbb{e}\left\{t[k ] | \boldsymbol{q}(t_k)\right\}}}{{\mathbb{e}\left\{\hat{t}[k ] | \boldsymbol{q}(t_k)\right\ } } } - x_{su}(t_k ) { \mathbb{e}\left\{t[k ] p_{avg } | \boldsymbol{q}(t_k)\right\ } } \ ] ] using ( [ eq : iid_1])-([eq : iid_3 ] ) in the inequality above , we get : @xmath304 | \boldsymbol{q}(t_k)\right\ } } \label{eq : drift5}\end{aligned}\ ] ] to prove ( [ eq : x_max1 ] ) , we rearrange ( [ eq : drift5 ] ) to get : @xmath305 | \boldsymbol{q}(t_k)\right\ } } \\ & + v { \mathbb{e}\left\{\sum_{t = t_k}^{t_{k+1}-1 } r_{su}^{fab}(t ) | \boldsymbol{q}(t_k)\right\ } } \leq b + c + v t_{max } a_{max}\end{aligned}\ ] ] ( [ eq : x_max1 ] ) now follows from theorem 4.1 of @xcite .
since @xmath67 is mean rate stable , ( [ eq : x_max2 ] ) follows from theorem 2.5(b ) of @xcite . to prove ( [ eq : femto_utility_bound ] )
, we take expectations of both sides of ( [ eq : drift5 ] ) to get : @xmath306\right\}}\end{aligned}\ ] ] summing over @xmath307 , dividing by @xmath165 , and rearranging yields : @xmath308\right\ } } - \frac{(b + c)k}{v } \end{aligned}\ ] ] where we used that fact that @xmath309 and @xmath310 . from this
, we have : @xmath311\right\ } } } & \geq \upsilon^ * - \frac{(b + c)k}{v \sum_{k=1}^k { \mathbb{e}\left\{t[k]\right\ } } } \\ & \geq \upsilon^ * - \frac{b + c}{v t_{min}}\end{aligned}\ ] ] since @xmath312\right\ } } \geq k t_{min}$ ] .
this proves ( [ eq : femto_utility_bound ] ) .
here , we compute a finite @xmath43 that satisfies ( [ eq : second_moment ] ) . first , note that @xmath44\right\}}$ ] would be maximum when the secondary user never cooperates .
next , let @xmath313 $ ] and @xmath314 $ ] denote the lengths of the primary user idle and busy periods , respectively , in the @xmath60 frame .
thus , we have @xmath0 = i[k ] + b[k]$ ] . in the following , we drop
@xmath315 $ ] from the notation for convenience . using the independence of @xmath316 and @xmath75
, we have : @xmath317 we note that @xmath316 is a geometric r.v . with parameter @xmath6 .
thus , @xmath318 and @xmath319 . to calculate @xmath320
, we apply little s theorem to get : @xmath321 this yields @xmath322 . to calculate @xmath323
, we use the observation that changing the service order of packets in the primary queue to preemptive lifo does not change the length of the busy period @xmath75 .
however , with lifo scheduling , @xmath75 now equals the duration that the first packet stays in the queue . next ,
suppose there are @xmath324 packets that interrupt the service of the first packet .
let these be indexed as @xmath325 .
we can relate @xmath75 to the service time @xmath326 of the first packet and the durations for which all these other packets stay in the queue as follows : @xmath327 here , @xmath328 denotes the duration for which packet @xmath99 stays in the queue . using the memoryless property of the i.i.d .
arrival process of the primary packets as well as the i.i.d
. nature of the service times , it follows that all the r.v.s @xmath328 are i.i.d . with
the _ same _ distribution as @xmath75 .
further , they are independent of @xmath324 . squaring ( [ eq : b_k ] ) and taking expectations ,
we get : @xmath329 note that @xmath326 is a geometric r.v . with parameter @xmath30 .
thus @xmath330 and @xmath331 .
also , @xmath332 . using these in ( [ eq : b_squared ] ) , we have : @xmath333 to calculate the last term , we have : @xmath334 note that given @xmath335 , @xmath324 is a binomial r.v . with parameters
thus , we have : @xmath337 \\ & = \sum_{x \geq 1 } \big[(x \lambda_{pu})^2 + x \lambda_{pu}(1 - \lambda_{pu})\big ] ( 1-\phi_{nc})^{x-1}\phi_{nc } \\ & = \lambda_{pu}^2 \sum_{x \geq 1 } x^2 \phi_{nc } ( 1 -\phi_{nc})^{x-1 } \\ & + \lambda_{pu } ( 1 - \lambda_{pu } ) \sum_{x \geq 1 } x \phi_{nc } ( 1 -\phi_{nc})^{x-1 } \\ & = \lambda_{pu}^2 \frac{(2 - \phi_{nc})}{\phi_{nc}^2 } + \lambda_{pu } ( 1 - \lambda_{pu } ) \frac{1}{\phi_{nc}}\end{aligned}\ ] ] using this , we have : @xmath338 using this , we have : @xmath339 simplifying this yields : @xmath340 o. simeone , i. stanojev , s. savazzi , y. bar - ness , u. spagnolini , and r. pickholtz . spectrum leasing to cooperating secondary ad hoc networks . _
ieee jsac special issue on cognitive radio : theory and applications _ , 26(1):203 - 213 , jan .
2008 .
i. krikidis , j. n. laneman , j. thompson , and s. mclaughlin . protocol design and throughput analysis for multi - user cognitive cooperative systems .
_ ieee trans .
wireless commun .
_ , 8(9):4740 - 4751 , sept . 2009 . | we investigate opportunistic cooperation between unlicensed secondary users and legacy primary users in a cognitive radio network .
specifically , we consider a model of a cognitive network where a secondary user can cooperatively transmit with the primary user in order to improve the latter s effective transmission rate . in return
, the secondary user gets more opportunities for transmitting its own data when the primary user is idle .
this kind of interaction between the primary and secondary users is different from the traditional _ dynamic spectrum access _ model in which the secondary users try to avoid interfering with the primary users while seeking transmission opportunities on vacant primary channels . in our model ,
the secondary users need to balance the desire to cooperate more ( to create more transmission opportunities ) with the need for maintaining sufficient energy levels for their own transmissions .
such a model is applicable in the emerging area of cognitive femtocell networks .
we formulate the problem of maximizing the secondary user throughput subject to a time average power constraint under these settings .
this is a constrained markov decision problem and conventional solution techniques based on dynamic programming require either extensive knowledge of the system dynamics or learning based approaches that suffer from large convergence times .
however , using the technique of lyapunov optimization , we design a novel _ greedy _ and _ online _ control algorithm that overcomes these challenges and is provably optimal .
resource allocation , opportunistic cooperation , cognitive radio , femtocell networks , optimal control |
in presence of massive objects , the geometry of spacetime is warped .
this warping , commonly called the curvature of spacetime , causes clocks situated at different locations to tick at different rates .
this is known as gravitational time dilation and has been an experimentally proven phenomenon for about half of a century @xcite .
even the relatively weak gravitational field of the earth has an effect on certain technological applications such as the global positioning system and must be accounted for .
recent experiments @xcite have been able to detect time dilation near earth s surface due to height difference of 1 meter .
recently , pikovski et al . have considered for the first time the effect of gravitational time dilation on the coherence of a composite quantum system prepared in a quantum superposition of two locations corresponding to different distances from earth @xcite . for this system ,
e.g. , a molecule , time dilation induces a universal coupling between the internal degrees of freedom and the center of mass ( c.m . ) .
this leads to decoherence in the c.m .
position of the particle characterized by a time scale @xmath0 . from an open quantum systems perspective
, the internal degrees of freedom of the molecule play the role of the environment while the c.m .
degree of freedom represents an open system . for
the sake of concreteness , let us consider the case in which the internal degrees of freedom are bosonic .
a standard example is a molecule for which the vibrations of its constituent atoms are described in terms of @xmath1 normal modes represented mathematically as uncoupled quantum harmonic oscillators with frequencies @xmath2 , @xmath3 .
the coupling between system and environment arises because the vibrational frequencies become position - dependent due to time dilation , namely , @xmath4 .
this in turn induces non - dissipative decoherence , as shown in ref .
@xcite . the gravitation - induced decoherence model considered by pikovski et al
. relies on a number of quite restrictive approximations and should therefore be considered as a toy model useful to grasp the main features of a new decoherence mechanism .
more specifically , the spherically symmetric metric is considered in the newtonian limit , valid only for slowly moving particles in weak gravitational fields , therefore general relativistic effects are not taken into account .
secondly , a superposition of position eigenstates is considered as initial state , rather than a more physically significant superposition of macroscopically distinguishable classical - like states .
thirdly , the observer , i.e. , the location of the decoherence measurement device , is assumed to be asymptotically far away .
these assumptions may severely limit the generality of the conclusions made in ref .
@xcite , and it is therefore crucial to extend the proposed model towards more realistic physical scenarios . in this paper
we generalize the study of gravitationally induced decoherence in several directions .
first of all we calculate the deviations due to a non - asymptotic observer , as well as the effect of the total time dilation , in addition to the relative one . then we focus on the case in which the molecule is confined in a harmonic trap and initially prepared in a superposition of two macroscopically distinguishable coherent states of the c.m .
motion , centered at two different heights .
this allows us to describe gravitation - induced decoherence when the model open system is a quantum harmonic oscillator .
this is particularly useful in describing the quantum to classical transition and , in fact , experiments revealing the decoherence of these so called cat states in this framework have been performed in different experimental platforms @xcite .
importantly , the nonclassicality of this schrdinger cat state can be described by means of various , physically meaningful nonclassicality indicators , i.e. , the peak of the interference fringe of the wigner function @xcite , the nonclassical depth @xcite , the negativity of the wigner function @xcite , the vogel criterion @xcite and the klyshko @xcite criterion .
such quantities measure different , nonequivalent ways in which our system can be nonclassical .
therefore , a much more complete picture of the quantum to classical transition can be given by looking at the dynamics of all such indicators , since in general their characteristic decoherence timescales will be different .
this is one of the main goals of this paper . finally , we take into account realistic sources of noise arising in the experiments .
we compare the characteristic timescales of time dilation induced decoherence to the timescale of decoherence due to classical fluctuation in crucial experimental parameters .
this allows us to assess whether time dilation induced decoherence is an observable phenomenon or , alternatively , to define the necessary measurement precision for the detection of such an intriguing fundamental effect .
the paper is structured as follows . in sec .
[ sec : model ] we present the master equation and its exact solution in terms of the quantum characteristic function . in sec .
[ sec : observer ] we introduce a schwarzschild potential and a non - asymptotic observer in the relativistic sense and explore the effects of these changes on the hamiltonian as well as the decoherence time . in sec .
[ sec : decoh_dynamics ] , we study the dynamics of an initial superposition of coherent states and describe decoherence in terms of the time evolution of the interference fringe of the wigner function . in sec .
[ sec : nonclass_ind ] , we calculate , estimate and compare the values of the aforementioned nonclassicality indicators for our system . finally , in sec .
[ sec : classic_noise ] , we consider the effect of classical noise and estimate the levels of precision and noise control required to observe the time dilation induced decoherence phenomenon .
following ref .
@xcite we consider a composite system described by the subsequent hamiltonian @xmath5 where @xmath6 with @xmath7 the hamiltonian of the internal degrees of freedom described by @xmath1 quantum harmonic oscillators of frequencies @xmath2 . for weak gravitational fields and slow - moving particles , the time
dilation induced coupling between the internal degrees of freedom and the c.m . to the lowest order in @xmath8
is described by @xmath9 . in eq .
( [ eq : fullhamilhint ] ) , @xmath10 is the local gravitational acceleration and @xmath11 is the c.m . position operator . using standard open quantum systems approaches @xcite
one derives the master equation for the c.m .
dynamics under the following assumptions : ( i ) weak coupling limit , ( ii ) no initial correlations between the state of the c.m . and the state of internal degrees of freedom , ( iii ) initially thermalized state of the internal degrees of freedom , ( iv ) negligible changes in the off - diagonal elements due to the c.m .
hamiltonian on the decoherence timescale @xcite : @xmath12 \\ & - \left ( \frac{\delta e_0 g}{\hbar c^2 } \right)^2 t \left [ x , \left [ x,\rho_{\text{cm}}(t)\right ] \right ] , \end{split}\ ] ] in eq .
( [ eq : mebruk ] ) , @xmath13 is the total mass of the system while @xmath14 and @xmath15 are the expectation value and the variance of @xmath7 , respectively .
the term @xmath16 represents the c.m .
hamiltonian in a convenient picture , as defined in ref .
we now consider the case where the c.m .
motion is a quantum harmonic oscillator @xmath17 , with @xmath18 and @xmath19 creation and annihilation operators , respectively , and @xmath20 the oscillator frequency . assuming that the dominant term in the unitary dynamics is the c.m .
hamiltonian , eq .
( [ eq : mebruk ] ) becomes @xmath21 -\delta(t ) \left [ x , \left[x,\rho_{\text{cm}}(t ) \right ] \right],\ ] ] where we have used the dimensionless position operator @xmath22 , related to @xmath11 through the formula @xmath23 , with @xmath24 the width of the ground state wavefunction of the quantum harmonic oscillator . with this notation
the time - dependent coefficient @xmath25 , which is linear in time and positive , takes the form @xmath26 note that the master equation ( [ eq : mebrukshort ] ) is of the lindblad form hence the dynamics is always divisible ( markovian ) @xcite .
moreover , the master equation here considered is a special case of the well - known quantum brownian motion master equation which , in the interaction picture , reads as follows @xcite @xmath27\right ] \\ & + \pi(t ) \left [ x,\left [ p,\rho(t)\right]\right ] + \frac{i}{2 } r(t ) \left [ x^2 , \rho(t)\right ] \\ & - i \gamma(t ) \left [ x , \left\{p , \rho(t ) \right\}\right ] . \end{split}\ ] ] indeed , the master equation above reduces to eq . ( [ eq : mebrukshort ] ) for @xmath28 . using the general solution of the quantum brownian motion model ( see , e.g. , refs .
@xcite ) we obtain the following simple solution for @xmath29 , in terms of the symmetrically ordered quantum characteristic function @xmath30 @xmath31 with @xmath32 and @xmath33 we introduce here for convenience the family of @xmath34-ordered characteristic functions defined as @xmath35 e^{\frac{1}{2 } s |\xi|^2},\ ] ] where @xmath36 is the displacement operator and @xmath37 , correspond to the normally , symmetrically and antinormally ordered characteristic functions , respectively .
the quasi - probability distributions @xmath38 , defined as the fourier transform of the quantum characteristic functions , @xmath39 are the well - known wigner ( @xmath40 ) , glauber or @xmath41 ( @xmath42 ) and husimi or @xmath43 ( @xmath44 ) functions , respectively , extensively used in quantum optics . in the following we use , for simplicity , the following notation @xmath45 and @xmath46 .
the description in terms of characteristic function is completely equivalent to the density operator formalism as one sees from the equation @xmath47 the solution of the master equation ( [ eq : mebruk ] ) is therefore given by eq .
( [ rhoxi ] ) with eqs .
( [ eq : phi_with_ctdt ] ) -([nofteq ] ) . before proceeding with the study of decoherence
, we discuss in the next section the possible modifications of the master equation due to a change in the position of the observer .
in the sense of a relativistic frame of reference in schwarzschild metrics the time used in ref .
@xcite corresponds to the observer positioned asymptotically far away from the source of the gravity field .
a more realistic situation would be a finite - distance , stationary observer . in the following
we derive a more general solution for an observer at a finite distance @xmath48 from the source of gravity .
this is achieved by applying gravitational time dilation effects to the line element .
an additional effect appears from the total time dilation experienced by the system seen by a distant observer .
the effect can be introduced by expanding the newtonian potential @xmath49 beyond the non - relativistic homogeneous field approximation @xmath50 , where @xmath51 is the earth s radius , @xmath11 is the elevation from earth s surface , @xmath52 is earth s mass , and @xmath53 is the gravitational constant . in @xcite ,
the metric in question is approximated by its newtonian limit . here
we will use spherically symmetric schwarzschild metric of signature @xmath54 , where the line element has the form @xmath55 where @xmath56 is the schwarzschild radius .
the last equation holds for a stationary system @xmath57 . in the schwarzschild metric , the times of two stationary systems at distances @xmath58 and @xmath59 are related by the equation @xmath60 as the proper time @xmath61 is invariant . in @xcite , the observer is stationed at one of these systems and is located infinitely far away ( i.e. @xmath62 ) , hence setting the stationary observers time equal to proper time . using the above equation we can correct for the effect appearing from a non - asymptotic , nearby observer by modifying the asymptotic decoherence time @xmath63 with a observer - position - dependent coefficient : @xmath64 with @xmath65 .
the position of the observer is now arbitrary . on the other hand ,
when the system is at rest with respect to the gravitational source , the hamiltonian of ref @xcite can be generalized as follows @xmath66 where @xmath67 .
we see that the truncated , momentumless forms of hamiltonian components are quite similar apart from some differences in front factors : @xmath68 with @xmath69 assuming earth s mass and radius this can be written using earth surface gravitational acceleration @xmath10 : @xmath70 the master equation takes now the form @xmath71 \\ & -\left ( \frac{\delta e_0}{\hbar } \right)^2 g_r^2 t \left [ x , \left [ x,\rho_{\text{cm}}(t ) \right ] \right ] .
\end{split}\ ] ] one sees immediately that the effect on the decoherence rate is the substitution of earth s gravity acceleration rate @xmath10 factor with an observer - dependent factor @xmath72 .
thus , we finally obtain @xmath73 giving the lowest order gravitational effects to the complete quantum system - observer pair . for earth mass and radius the correction @xmath74 is only of the order of @xmath75 at surface level .
however it is more significant in proximity to heavy stellar object or near a horizon .
let us assume that the system is initially prepared in a schrdinger cat state of the form @xmath76 where @xmath77 , with @xmath78 a coherent state . here , @xmath79 is the normalization factor .
more specifically we consider the so - called even coherent state obtained for @xmath80 .
the `` size '' of the cat is given by @xmath81 , where @xmath82 is the distance between the peaks of the two gaussian functions describing the coherent state components of the superposition in phase space .
we study the time evolution of the state in terms of its wigner function defined in eq .
( [ eq : charfunctionxi ] ) with @xmath40 .
inserting eqs.([eq : phi_with_ctdt])-([nofteq ] ) into ( [ eq : charfunctionxi ] ) one can write the wigner function for the initial state here considered as follows @xcite : @xmath83 where @xmath84 @xmath85 with @xmath86 given by eq .
( [ nofteq ] ) . the expression above is particularly suited to describe environment - induced decoherence and its effect on the cat state since it singles out the interference term @xmath87 which characterizes the quantumness of the initial state .
as decoherence takes place such term decays with a characteristic time known as decoherence time .
following the standard description of environment - induced decoherence , we introduce the fringe visibility function @xcite @xmath88^{1/2}},\end{aligned}\ ] ] where @xmath89 in our case , this function takes the form @xcite @xmath90.\ ] ] inserting eq .
( [ nofteq ] ) into eq .
( [ eq : decoratemaniscalco ] ) , and in the limit @xmath91 , we obtain the following simple expression for the fringe visibility function @xmath92 where the decoherence time is given by @xmath93 the condition @xmath91 can be written as @xmath94 showing that eq .
( [ fapproxim ] ) only describes the initial decoherence behaviour since generally @xmath95 . to conclude this section we compare
the decoherence time derived with our formalism with the one defined in ref .
@xcite , where the initial state of the c.m . is a superposition of different position eigenstates separated by a height difference @xmath96 and decoherence is measured using the notion of interferometric visibility given by @xcite : @xmath97 with @xmath1 the number of internal degrees of freedom . for times
@xmath98 , with the decoherence time defined as @xmath99 the interferometric visibility is approximated by @xmath100 which has the same temporal behaviour as the wigner function fringe visibility that we have derived in eq .
( [ fapproxim ] ) . recalling from eq .
( [ nofteq ] ) the expression of @xmath101 appearing in eq .
( [ eq : decorate_fringepeak ] ) , and remembering that @xmath102 when the state of the internal degrees of freedom is thermal , one concludes straightforwardly that @xmath103 .
the definition of nonclassicality for the states of the quantum harmonic oscillator has been extensively debated in the past .
there exist indeed different quantities measuring or highlighting different ways in which this paradigmatic system departs from classical behaviour .
therefore , in order to provide a holistic view of the quantum to classical transition stemming from gravitational decoherence , in this section we explore how the most widespread nonclassicality indicators witness the loss of quantumness .
we will describe the dynamics of the following quantities : the nonclassical depth , measuring the minimum number of thermal photons required to destroy any nonclassical characteristics of the system @xcite ; the negativity of the wigner function quantifying the separation between the wigner distribution and a classical probability distribution @xcite ; the vogel criterion defined in terms of properties of the characteristic functions of the quadrature distributions having no classical counterpart @xcite ; and the klyshko criterion detecting differences between classical photon number probability distributions and the quantum ones @xcite .
let us begin by writing the quasiprobability distributions defined in eq .
( [ eq : charfunctionxi ] ) for @xmath104 as the following convolution @xmath105 where @xmath106 we now generalize to the case in which @xmath34 is a continuous parameter taking values in the interval @xmath107 $ ] .
the _ nonclassical depth _ is defined as @xcite @xmath108 \left| w(\alpha , s ) \geq 0 \right .
\right\}. \end{split}\ ] ] by following the steps of ref .
@xcite we notice that setting @xmath109 in eq .
( [ eq : genw ] ) we can express @xmath38 as a convolution of the @xmath41 function @xmath110 using eqs .
( [ eq : phi_with_ctdt])-([nofteq ] ) we obtain the following expression for the time evolution of the @xmath41 function @xmath111 where we have used the fact that the fourier transform of a product of two functions is equal to the convolution of the two corresponding fourier transforms , and where @xmath112 .
the equation above shows that our dynamics transform the initial @xmath41 function into the other characteristic functions .
since the nonclassical depth of the initial state is @xmath113 , one can prove that the time @xmath114 at which @xmath38 becomes positive corresponds to the time at which the initial @xmath41 is transformed into the @xmath43 function ( which is always positive ) @xcite , that is @xmath115 solving for @xmath114 one obtains straightforwardly @xmath116 note that @xmath117 , where @xmath118 is the decoherence time associated to the decay of the wigner function fringe visibility , given by eq .
( [ eq : decorate_fringepeak ] ) .
hence , @xmath114 does not depend on the size of the cat , contrarily to @xmath118 .
moreover , for truly macroscopic superpositions such that @xmath119 , the loss of interference in the wigner function is much faster than the loss of quantumness measured by the nonclassical depth .
we now turn our attention to the second nonclassicality indicator , namely the _ negativity of the wigner function_. we recall that this quantity is associated to nonlocal properties and , more specifically , to violation of the bell inequality @xcite
. more precisely , we are interested in identifying the time @xmath120 at which the wigner function of the initial cat state , which is negative in several zones of the phase space , becomes positive everywhere .
this time can be calculated analytically once again using eq .
( [ eqkey ] ) and corresponds to the time at which the initial @xmath41 function is turned into the wigner function @xmath121 yielding @xmath122 decoherence induced by time dilation therefore causes the negativity of the wigner function to disappear faster than the nonclassicality as measured by the nonclassical depth .
moreover , as the latter one , it does not depend on the size of the initial superposition .
the two criteria considered so far are based on properties of the quasiprobability distribution functions and , as such , are experimentally demanding since they require full tomography of the state while it evolves due to the interaction with the environment .
the next criterion examined is on the contrary experimentally easier to implement since it is defined in terms of the symmetrically ordered characteristic function which can be directly measured through balanced homodyne detection .
the _ vogel nonclassicality criterion _ is indeed simply defined as follows @xcite : a state is nonclassical at time @xmath123 iff for its normally ordered characteristic function , @xmath124 using our solution given by eq .
( [ eq : phi_with_ctdt ] ) one promptly obtains @xmath125 recalling the expression of @xmath126 for our initial schrdinger cat state we can write the vogel criterion as follows @xmath127 \le \chi_t(u,0,1 ) .
\end{split}\ ] ] the vogel criterion is inherently state - dependent and , in particular , in our case it depends on @xmath128 .
one defines the vogel nonclassicality transition time @xmath129 from the equation @xmath130 . in fig .
[ fig : criterionplot ] we plot the behaviour of @xmath129 , in units of @xmath120 , as @xmath128 increases .
as one can see from the figure , the decoherence time @xmath129 tends to saturate as the superposition becomes more and more macroscopic . in the limit @xmath131 one
easily obtains analytically that the condition @xmath130 amounts at requiring that @xmath132 this is an upper bound for the onset of classicality and , for our system , it turns out to be equivalent to the disappearence of the negativity of the wigner function . in units of @xmath120
@xmath133 represents the peak of interference fringe.,scaledwidth=48.0% ] the last criterion explored in our paper is the _ klyshko criterion _ @xcite .
similarly to the vogel criterion , also this one is experimentally friendly since it only requires the measurement of phonon number distributions @xmath134 .
the criterion states that a sufficient condition for a state to be nonclassical , or more precisely to have nonclassical phonon counting statistics , is that there exist at least one integer @xmath135 such that @xcite @xmath136 the phonon number probabilities can be written in terms of the normally ordered characteristic functions @xmath137 and the antinormally ordered characteristic function for the number states @xmath138 , where @xmath139 are the laguerre polynomials : @xmath140 finding times where @xmath141 is valid yields the plot in fig .
[ fig : criterionplot ] .
the plot shows that the klyshko nonclassicality decoherence time @xmath142 ( in the figure in units of @xmath120 ) depends on the size of the cat and , in particular , quickly decreases for increasing values of @xmath128 , i.e. , with the size of the cat state .
we also see that , in general , @xmath143 so , as in the case of @xmath118 , this type of nonclassicality quickly disappears and it is more difficult to be hidden by other sources of noise . in order to evaluate the ability to detect time
dilation induced decoherence we need to compare its characteristic time scales with those of the most significant noise sources that may affect the dynamics .
the fact that gravitational decoherence might be very hard to detect experimentally was already recognized in ref .
@xcite , where the effect of decoherence due to emission of radiation was considered for comparison .
a more thorough analysis was performed ref .
@xcite , where the effects of both collisional and thermal decoherence were analyzed . in the following section we will perform a comparison with what is perhaps the most common source of decoherence in the experiments , namely classical noise on the experimental parameters .
one of the most ubiquitous sources of decoherence in interferometric experiments is classical noise affecting the relevant experimental parameters .
the generality of this noise source makes the investigation of its effects on the new type of decoherence described in this paper a crucial step .
moreover , contrarily to the environmental effects considered before in refs . @xcite and @xcite , in the case of classical stochastic noise it is possible to perform a numerical exact analysis without invoking the born / markov approximations commonly done for quantum environments . in this way
we can extend the analysis of the observability of gravitational decoherence due to time dilation to situations where memory effects play a crucial role .
this is particularly relevant since reservoir engineering techniques nowadays allow to manipulate the properties of the environment in order to increase the coherence times , e.g. , by means of backflow of information and recoherence characterizing non - markovian dynamics @xcite .
this in turn would make it possible to prolong the coherence time associated to classical noise long enough to render time dilation induced decoherence practically observable in the experiments .
our analysis follows the results of ref .
@xcite , where a quantum harmonic oscillator is subjected to a classical stochastic field as described by the following coupling hamiltonian : @xmath144 where @xmath145 describes a gaussian stochastic process with the following properties : @xmath146 here @xmath147 is the complex conjugate of @xmath148 , the @xmath149 notation represents the average over all stochastic realizations , and @xmath150 is the kernel autocorrelation function . for the sake of concreteness
we consider a ornstein - uhlenbeck process with @xcite @xmath151 the parameter @xmath152 is the system - noise coupling constant while @xmath153 quantifies the temporal correlations of the environment , its inverse therefore measuring the so called memory time of the environment . for a gaussian stationary process
, the evolved state can be written in terms of the @xmath34-ordered characteristic functions as follows @xmath154 with @xmath155 k(s_1,s_2).\ ] ] for resonant interaction and for the ornstein - uhlenbeck process here considered , @xmath156 has a simple analytical expression @xcite @xmath157 comparing eq . ( [ eq : classno ] ) with eq .
( [ eq : phi_with_ctdt ] ) one sees immediately that the quantum characteristic function describing the time evolution in presence of classical stochastic noise has precisely the same form as the one describing time dilation induced decoherence , with @xmath158 now playing the role of @xmath159 .
we recall here the expressions of the nonclassicality indicators calculated in @xcite .
the decoherence time @xmath120 , associated to the negativity of the wigner function , is given by @xmath160 .\ ] ] as for the nonclassical depth it is sufficient to recall that , also in this case , @xmath161 . the vogel and klyshko criteria are studied according to the same lines of sec . v. we note that , following ref .
@xcite , all energy - related quantities are rescaled in units of @xmath162 . and @xmath152 for @xmath163 .
the shaded areas represent the region of parameters in which decoherence due to classical noise dominates over time dilation induced decoherence . @xmath153 and @xmath152
are given in units of @xmath20.,scaledwidth=48.0% ] to compare the effects of classical noise and time dilation induced decoherence on the quantum to classical transition we consider the ratio of the respective decoherence times as a function of @xmath153 and @xmath152 . similarly to ref .
@xcite we consider a system with superposition size of @xmath164 m and a temperature @xmath165 setting the particle number @xmath1 to @xmath166 .
regions in which the classical noise dominates are shown in fig .
[ fig : classical_nondepth_wignerneg_vogel_klyshko ] and correspond to the ratio being @xmath167 .
we notice that the behaviour of all different indicators with respect to the constants @xmath153 and @xmath152 is very similar .
this means that , independently of either the strength of the coupling or the memory time , and hence non - markovian character of the dynamics , classical noise destroys all types of quantumness much faster than time dilation induced decoherence , making this phenomenon very difficult to observe in the experiments .
one of the goals of this paper was to investigate the potential ways to expand on the results of ref .
@xcite concerning the decoherence of an open quantum system in a spatial superposition at different heights above a source of gravitational field .
this was approached from several directions .
we replaced the existing potential with a more realistic gravitational potential to act as the environment and introduced the notion of a non - asymptotic relativistic observer .
the aforementioned changes manifest themselves in simple coefficients to the total decoherence time allowing for necessary corrections .
we proved that the corrections are negligible close to earth surface but may become significant close to heavy stellar objects .
we then considered the dynamics of a schrdinger cat state and found the decoherence time measured in terms of the decay of the wigner fringe visibility function .
in addition , we explored several other measures of decoherence , called nonclassicality indicators , and presented their dependence on the parameters of the system as well as their relative magnitudes .
as some of the indicators are more appropriate for certain experimental implementations , the additional measures may prove useful for detection of the phenomenon .
finally , continuing with the theme of experimental detection , we examined the scale of precision or noise control required to detect the effect of time dilation in the presense of classical noise .
our analysis shows that the phenomenon of gravitational induced decoherence imposes very high demands on the acceptable level of classical noise in order to be observed experimentally .
this work was supported by the academy of finland ( project no .
287750 ) , finnish cultural foundation and the magnus ehrnrooth foundation .
we thank prof .
daniel braun for fruitful discussions . | we study the loss of quantumness caused by time dilation @xcite for a schrdinger cat state , considering also the effect of a non - asymptotic observer .
we give a holistic view of the quantum to classical transition by comparing the dynamics of several nonclassicality indicators , such as the wigner function interference fringe , the negativity of the wigner function , the nonclassical depth , the vogel criterion and the klyshko criterion .
our results show that only two of these indicators depend critically on the size of the cat , namely on how macroscopic the superposition is .
finally we compare the gravitation - induced decoherence times to the typical decoherence times due to classical noise originating from the unavoidable statistical fluctuations in the characteristic parameters of the system @xcite .
we show that the experimental observation of decoherence due to time dilation imposes severe limitations on the allowed levels of classical noise in the experiments . |
quantifying the cosmic fields , such as the density fields , velocity fields , gravitational lensing fields , temperature fluctuations in the cosmic microwave background , etc .
is undoubtedly crucial to study the origin and dynamics of the structure in the universe .
the structure of these fields are believed to be emerged from primordial random gaussian perturbations , as most of the inflationary models naturally predict nearly scale - invariant gaussian fluctuations @xcite .
even if the primordial perturbations are random gaussian , the gravitationally nonlinear evolution produces non - gaussianity in the cosmic fields .
quantifying the non - gaussianity is not trivial since it depends on the full hierarchy of the higher - order correlation functions in real space , or of the polyspectra in fourier space .
first several members of such hierarchy can be observationally determined , with which only partial information on non - gaussianity is quantified .
therefore , alternative statistics , such as the void probability function @xcite , the genus statistic @xcite , the minkowski functionals @xcite , etc . which contain the information on
the full hierarchy of higher - order statistics should be useful .
non - gaussianity is frequently termed `` phase correlations '' .
this term reflects the fact that the fourier phases of a random gaussian field are randomly distributed without any correlation among different modes .
therefore , phase correlations , if any , obviously characterize the non - gaussianity .
however , what kind of phase correlations arise in a given non - gaussian field have been far from obvious .
investigations along this line are quite limited in the literatures despite its importance , apparently because of the lack of theoretical guidelines . most of the earlier work @xcite only assess the nonlinear evolution of phases in individual fourier modes without statistics .
phenomenological studies of @xmath2-body simulations have revealed that the one - point phase distribution remains uniform even in non - gaussian fields @xcite , and that the phase difference between neighboring fourier modes is non - uniformly distributed @xcite . however , the meaning of the discovered phase correlations is obscure in those literatures .
since the hierarchy of the higher - order statistics contains statistically all information on the distribution @xcite , there should be some connection between phase correlations and polyspectra , which is the key to theoretically understand the phase correlations . examining a toy model
, @xcite realized the importance of the phase sums with closed wave vectors in this connection , although they have never derived the exact relations . in this letter
, the connection in the general form is discovered for the first time . as a result
, we will have much better theoretical understanding of the phase information than before .
although the real part @xmath3 and the imaginary part @xmath4 of the fourier transform @xmath5 of a random field @xmath6 are naturally the independent variables , one can also take their linear combinations @xmath7 and @xmath8 as another set of mutually independent variables . for calculational advantages
, we use the latter choice . in this letter , the reality of the random field @xmath6 is assumed since most of the cosmic fields are real , although one can readily generalize the following analysis to the complex fields .
because of the reality condition , @xmath9 , @xmath10 s are actually not independent variables , and @xmath5 s of all modes @xmath11 are taken as independent variables .
it is useful to define the normalized quantity @xmath12 , where @xmath13 is the power spectrum of the random field .
the key technique to derive the relation between phase correlations and polyspectra is given by previous work @xcite : the joint probability function @xmath14 of having particular set of @xmath15 is formally represented by @xmath16 where @xmath17 indicates the cumulants , and @xmath18 is the multivariate gaussian distribution function of variables @xmath19 . in the present case ,
@xmath20 , since the covariance matrix is @xmath21 , where the symbol @xmath22 is defined by @xmath23 for @xmath24 and @xmath25 for @xmath26 .
the periodic boundary condition with boxsize @xmath27 is assumed . since the polyspectra @xmath28 are defined from the cumulants by @xmath29 the above formula ( [ eq01 ] )
provides the relation between polyspectra and joint distribution of the fourier coefficients . expanding the exponential in equation ( [ eq01 ] ) , each term in this expansion consists of the products of polyspectra times derivatives of @xmath30 .
the derivatives of @xmath30 are given by a simple polynomial of @xmath31 s times @xmath30 .
the general term in the expansion has the form @xmath32 with appropriate coefficients , where @xmath33 are the dimensionless , normalized polyspectra of @xmath31 , and @xmath34 is a generalization of hermite polynomials and is given by polynomials of @xmath31 s and @xmath22 s .
for example , @xmath35 and so on .
thus , the joint probability function @xmath36 is represented by @xmath30 times infinite sum of products by @xmath31 s , @xmath37 s , and normalized polyspectra .
next step is to transform the complex variable @xmath31 into the modulus @xmath38 and the phase @xmath39 by @xmath40 .
this should be carefully done , since @xmath31 is considered independent on @xmath41 . at this point , we restrict the wavenumber @xmath11 in the upper half sphere ( uhs ) , @xmath42 , and the degrees of freedom in the lower half sphere is relabeled by the reality relation , @xmath43 for @xmath44 . the mode @xmath45 is excluded which ensures zero mean of the original field @xmath6 .
the term ( [ eq03 ] ) is accordingly relabeled , resulting in the sum of the products of @xmath31 s , @xmath41 s , @xmath22 s , where @xmath46 , and normalized polyspectra . with the above procedures
, one can express the ratio @xmath47 in terms of the normalized polyspectra , the modulus @xmath38 and the phase @xmath39 .
the jacobian of the transform from @xmath31 to @xmath48 is the same for the probability functions @xmath36 and @xmath49 .
therefore this meets our ends to relate the phase correlations and polyspectra , which is a completely new result .
practically , one needs to truncate the infinite series by a consistent manner .
fortunately , the non - gaussianity generated by gravitationally nonlinear evolution is known to approximately follow the hierarchical model of the higher - order correlations , in which the polyspectra @xmath50 have the order , @xmath51 $ ] ( e.g. , * ? ? ? * ) .
this means @xmath52 , where @xmath53 .
therefore one can evaluate the phase correlations in perturbative manner as long as the expansion parameter @xmath54 is small .
it is straightforward to perform the above procedure to express @xmath55 in terms of normalized polyspectra to arbitrary order in @xmath54 . in the lowest order approximation
, only the normalized bispectrum @xmath56 gives the term of order @xmath57 .
the result is @xmath58 \nonumber\\ & & \qquad\quad\times \prod_{{{\mbox{\scriptsize\boldmath $ k$}}}\in{\rm uhs } } 2|\alpha_{{{\mbox{\scriptsize\boldmath $ k$}}}}| e^{-|\alpha_{{{\mbox{\tiny\boldmath $ k$}}}}|^2 } d|\alpha_{{{\mbox{\scriptsize\boldmath $ k$}}}}|\frac{d\theta_{{{\mbox{\scriptsize\boldmath $ k$}}}}}{2\pi}. \label{eq07}\end{aligned}\ ] ] higher - order terms can be similarly calculated , although they are somehow tedious . for example , in the second - order approximation , @xmath59 , there appears the square of the first - order term , and terms like @xmath60 with appropriate normalized trispectrum or the product of normalized bispectra multiplied , and other terms which do not depend on phases .
the phases always contribute to the probability distribution by the combination of the form , @xmath61 , with closed wavevectors : @xmath1 .
this is generally true because the phase dependence in equation ( [ eq03 ] ) is the exponential of the sum of phases , and the probability is the real number so that taking real parts gives the cosine function .
the reason that phase correlations exist only among modes with closed wavevectors comes from the translational invariance . in equations ( [ eq07 ] ) and ( [ eq08 ] ) , wavenumbers are restricted to the uhs so that the modes in the lower half sphere are relabeled by @xmath62 .
the moduli @xmath38 s are easily integrated in the first - order approximation of equation ( [ eq07 ] ) , resulting in @xmath63 the practically useful relations between phase correlations and the bispectrum are obtained by further integrating some phases in equation ( [ eq09 ] ) .
one obtains @xmath64 where @xmath65 .
these are the explicit forms of the relation between phase correlations and the bispectrum in the first - order approximation .
we find that the distribution of the ` phase sum ' @xmath66 is determined only by the normalized bispectrum at the first - order level , although higher - order normalized polyspectra can contribute in general .
the higher - order calculations show that the distribution of the phase sum @xmath0 for the modes with closed wavevectors @xmath67 is determined by normalized polyspectra of order 3 to @xmath2 in the lowest - order approximation , where the identification @xmath62 is understood .
it was vaguely suggested that there is some relationship between the phase sum and polyspectra based on a particular non - gaussian model by @xcite .
we now find the explicit relationship between them in general non - gaussian fields .
if we further integrate all phases but one particular @xmath39 , the one - point probability function of a phase is uniform , @xmath68 , which is consistent with the previous @xmath2-body analysis @xcite .
this conclusion does not depend on the first - order approximation , since a single wavevector can not be closed unless @xmath24 .
similarly , the two - point probability function @xmath69 is also uniform unless @xmath70 . at first glance , this conclusion seems to contradict the reported non - uniform distribution of the phase difference of neighboring wavevectors @xmath71 in @xmath2-body data @xcite , where @xmath72 is a fixed small vector .
the same arguments are also applied to higher - order approximations , so that the phase correlations between neighboring wavenumbers should not appear even in strongly non - gaussian fields in a statistical sense . to resove this puzzle , it is useful to consider the conditional probability function given a fourier coefficient of a small wavenumber @xmath73 . in the first - order approximation ,
the joint probability of having phases @xmath74 with fixed @xmath73 is given by @xmath75 which arise the non - uniform distribution pattern of the phase difference @xmath76 .
the pattern depends on the fixed phase @xmath77 , which means the pattern varies from sample to sample , and this is exactly what is reported in the @xmath2-body analyses
. the pattern of the phase difference should be significant for red power spectrum , which is also consistent with @xmath2-body analyses .
the functional form of equation ( [ eq12 ] ) also agrees with the @xmath2-body analysis @xcite .
the statistics of phase difference is thus the manifestation of the large - scale patterns of individual realizations .
the position of the trough in the distribution of the phase difference corresponds to the phase of the mode @xmath78 , and the degree of deviations from the uniform distribution depends on the specific amplitude of the mode @xmath72 and also on the normalized bispectrum .
the equations ( [ eq10 ] ) and ( [ eq11 ] ) relate the bispectrum to the distribution of the phase sum @xmath79 . to see if this kind of phase information is practically robust , we numerically examine simple examples of non - gaussian fields . instead of examining cosmological simulations ,
the following simple example is enough to compare the numerical phase distributions and theoretical predictions .
series of non - gaussian fields are simply generated by exponential mapping of a random gaussian field : @xmath80 where @xmath81 is a random gaussian field with zero mean , unit variance , and @xmath82 is the non - gaussian parameter . we simply take a flat power spectrum for the gaussian field @xmath81 .
the field @xmath6 has zero mean and variance @xmath83 , and is called the lognormal field @xcite .
this field has quite similar statistical properties to gravitationally evolved non - gaussian fields and approximately follows the hierarchical model of higher - order correlations .
the parameter @xmath82 controls the non - gaussianity , and the random gaussian field is recovered by taking the limit @xmath84 .
the random field @xmath6 is generated on @xmath85 grids in a rectangular box with the periodic boundary condition . in fig .
[ fig1 ] , the distribution of the phase sum @xmath86 is plotted for a binned configuration of the wavevectors , @xmath87 $ ] , @xmath88 $ ] , @xmath89 $ ] , as an example , where @xmath90 is the angle between @xmath91 and @xmath92 , and the magnitudes of the wavenumber are in units of the nyquist wavenumber .
the phase sum is averaged over the wavevectors in a configuration bin .
the points represent the distributions of the phase sum in each realization .
poisson errorbars are smaller than the size of the points .
the normalized bispectra @xmath93 are numerically evaluated from each realization , which are used to draw the theoretical curves in the first - order approximation of equation ( [ eq11 ] ) .
there is not any fitting parameter at all .
the agreement is remarkable in weakly non - gaussian fields .
when the non - gaussianity becomes high , the data points deviate from the first - order approximation , and the distribution of the phase sum is sharply peaked at @xmath94 . up to @xmath95 , or @xmath96 ,
the distribution of the phase sum is accurately described by the first - order approximation , and is determined only by the normalized bispectrum . even though the non - gaussianity @xmath95 on scales of the nyquist wavenumber is beyond the perturbative regime ,
the normalized bispectrum on scales of the presently tested configuration is still within the perturbative regime , @xmath97 .
this means that the phase sum is well approximated by first - order formula of the present work even when the field is strongly nonlinear in dynamics , as long as the parameter @xmath98 on the relevant scales is small .
increasing the power on relevant scales and/or decreasing the volume drive the phase correlation large , due to the fact that the phase correlations are particularly dependent on significant features in the sample .
the structure of the phase correlations in non - gaussian fields is elucidated . the method to relate the joint distribution of phases to polyspectra
is newly found and developed .
the distribution of the phase sum of closed wavevectors is represented by the polyspectra .
we found the statistics of the phase difference reflect the particular phase of the mode within an individual sample .
the distribution of the phase sum of three or more modes carries the statistically useful information .
the understanding of the phase correlations in non - gaussian fields is now reached unprecedented level in this letter , so that many investigations to make use of the phase information will be followed , such as the analysis of the non - gaussianity of all kinds of cosmic fields , the nonlinear gravitational evolution of the density fields , the biasing and redshift - space distortion effects on the galaxy clustering , the primordial non - gaussianity from inflationary models , and so forth .
one may also hope that phase information can be useful in statistical analyses of all kinds of non - gaussian fields , from various phenomena of pattern formations to human brain mapping , etc .
bardeen , j. m. , steinhardt , p. j. , & turner , m. s. 1983 , , 28 , 679 bernardeau , f. , colombi , s. , gaztanaga , e. , & scoccimarro , r. 2002 , , 367 , 1 bertschinger , e. in lecture notes in physics , 408 , _ new insights into the universe _ , ed
. v. j. martinez , m. portilla , and d. saez .
, p65 ( springer - verlag , berlin , 1992 ) .
chiang , l. 2001 , , 325 , 405 chiang , l. , naselsky , p. , & coles , p. 2002 , arxiv arxiv : astro - ph/0208235 coles , p. & jones , b. 1991 , , 248 , 1 coles , p. & chiang , l. 2000 , , 406 , 376 gott , j. r. , dickinson , m. , & melott , a. l. 1986 , , 306 , 341 guth , a. h. & pi , s .- y .
1982 , physical review letters , 49 , 1110 hawking , s. w. 1982 , , 115 , 295 jain , b. & bertschinger , e. 1998 , , 509 , 517 matsubara , t. 1995 , , 101 , 1 matsubara , t. 2003 , , 584 , 1 mecke , k. r. , buchert , t. , & wagner , h. 1994 , , 288 , 697 minkowski , h. 1903 , math .
ann . , 57 , 447 ryden , b. s. & gramann , m. 1991 , , 383 , l33 scherrer , r. j. , melott , a. l. , & shandarin , s. f. 1991 , , 377 , 29 schmalzing , j. & buchert , t. 1997 , , 482 , l1 soda , j. & suto , y. 1992 , , 396 , 379 starobinskii , a. a. 1982 , , 117 , 175 suginohara , t. & suto , y. 1991 , , 371 , 470 watts , p. & coles , p. 2003
, , 338 , 806 watts , p. , coles , p. , & melott , a. 2003 , , in press , arxiv : astro - ph/0211408 .
white , s. d. m. 1979 , , 186 , 145 | we present the general relationship between phase correlations and the hierarchy of polyspectra in the fourier space , and the new theoretical understanding of the phase information is provided .
phase correlations are related to the polyspectra only through the non - uniform distributions of the phase sum @xmath0 with closed wave vectors , @xmath1 .
the exact relationship is given by the infinite series , which one can truncate in a consistent manner .
the method to calculate the series to arbitrary order is explained , and the explicit expression of the first - order approximation is given .
a numerical demonstration proves that the distribution of the phase sum is a robust estimator and provides an alternative statistic to search for the non - gaussianity . |
recent results from super kamiokande@xcite and earlier measurements from soudan-2@xcite and imb@xcite on the relative fluxes of electron and muon type neutrinos produced by the interaction of cosmic rays in the atmosphere show a deficit of muon - type neutrinos .
these results are consistent with the oscillation of @xmath4 into @xmath5 provided that the two mass eigenstates are fully mixed , @xmath6 , and their mass difference is of order @xmath7
. models that can accommodate this mixing have lepton flavor violation ( lfv ) and other lepton number violating processes which are likely to be observable , such as @xmath8 and @xmath9 .
it has been suggested@xcite that supersymmetric models can naturally accommodate the flavor violation .
if there is mixing of neutrinos , then one would in general expect mixing of sleptons and sneutrinos in susy .
in general , susy models can have significant flavor mixing .
the slepton mass matrix is a @xmath10 matrix involving the three flavors of left and right sleptons and has the form @xmath11 where @xmath12 label chirality and @xmath13 are generational indices . @xmath14 and @xmath15 are the supersymmetry breaking , chirality conserving , slepton masses .
the chirality breaking terms , @xmath16 , contain a flavor diagonal piece @xmath17 , where @xmath18 is the corresponding charged lepton mass , @xmath19 is the supersymmetry conserving higgs mass parameter and @xmath20 is the usual ratio of higgs vacuum expectation values .
additional supersymmetry breaking terms ( @xmath21 terms ) also appear in the chirality mixing term
. these can not be flavor diagonal if , as we require , the neutrino mass matrix is not diagonal in the flavor basis . however , these terms are expected to be proportional to charged lepton masses .
we will therefore make the simplifying assumption that these terms are relevant only for the third generation . in order to accommodate the @xmath4 into @xmath5 mixing
, there must be significant mixing in the stau , smuon sector@xcite . in the mssm ( minimal supersymmetric standard model )
there is no lepton number violation .
it can be incorporated by adding right handed neutrinos ( @xmath22 ) with couplings to the left handed lepton doublets @xmath23 of the form @xmath24 where @xmath25 is a higgs doublet .
then radiative corrections at one loop will induce lepton number violating terms in the mass matrices for the left sleptons .
there can also be similar soft supersymmetry breaking @xmath21 terms .
we therefore assume the following form for the slepton mass matrix in the @xmath26 basis @xcite : @xmath27\ ] ] where @xmath28 , @xmath29 , and @xmath30 .
minimal supersymmetric models such as sugra are recovered by neglecting @xmath31 .
the super kamiokande results suggest @xmath32 it has been suggested that this large flavor violation in the slepton sector could be observed indirectly via the decay @xmath33@xcite . by contrast , this paper considers direct signals from the observation of lepton flavor non - conservation in the susy sector at the lhc .
we will demonstrate that , in certain cases , the direct observation is more sensitive than the flavor violating tau decay .
section [ sec : model ] defines a sample point based on the minimal sugra model that has been studied in detail , and section [ sec : tau ] explains how sensitivity to @xmath34 flavor violation can be observed at the lhc for this model .
section [ sec : rare ] compares the sensitivity for this direct search with that from the rare decay @xmath35 .
section [ sec : general ] presents results indicating the range of parameters in the minimal sugra model for which such direct searches might be feasible .
finally , section [ sec : concl ] presents a summary and conclusions .
we consider a minimal sugra@xcite point with @xmath36 , @xmath37 , @xmath38 , @xmath39 , and @xmath40 .
we use the isajet@xcite implementation of the sugra model .
this point is identical to one of those studied in detail@xcite ( sugra point 5 ) except for the value of @xmath20 .
the larger @xmath20 gives a light higgs mass of @xmath41 , consistent with the current lep limit@xcite .
the masses are calculated with isajet 7.49 , which assumes @xmath42 .
a modification was made to make available the complete slepton mass matrix , giving access to all the soft mass parameters at the electroweak scale _ viz . _ : @xmath43 the mass eigenvalues for @xmath42 are @xmath44 ( @xmath45 is the lighter of the two mass eigenstates resulting from the mixing of @xmath46 and @xmath47 . ) note that in sugra type models the @xmath48 are generally lighter than the @xmath49 .
the dominant final states for @xmath50 decay are @xmath51 , @xmath52 and @xmath53 , and @xmath54 which have branching ratios of 6.9% , 12.8% ( for both @xmath55 and @xmath19 ) , and 66.1% respectively .
the @xmath56 branching ratios are 45.3% for @xmath57 and 54.4% for @xmath58 .
direct drell - yan production of sleptons at the lhc has a small cross section ( 270 fb for all sleptons in the case considered ) and is difficult to separate from susy backgrounds .
in contrast the total susy rate , made up largely of squark and gluino production is 100 times larger .
gluinos ( of mass 706 gev ) decay to squarks ( average mass 620 gev ) with an almost 100% branching ratio and left - squarks have a 30% branching ratio to @xmath50 ( of mass 212 gev ) .
hence the most copious source of observed dileptons from sleptons in this case is @xmath59 where the @xmath50 arises in squark decay
. this will always be the case provided that the decay is allowed .
if only @xmath48 is produced in this way ( in the case studied @xmath49 is heavier than @xmath50 ) , then the main effect of @xmath60 is to add a component of @xmath61 in @xmath45 so that @xmath62 and @xmath63 .
the @xmath64 mass is 116 gev .
this decay gives a @xmath65 signal and produces an asymmetry between @xmath65 and @xmath66 final states .
the slepton mass matrix was rediagonalized using the soft mass parameters given above plus the mixing term @xmath67 , inducing mixings among slepton flavors and hence flavor - violating decays such as @xmath68 and @xmath69 .
the resulting branching ratios for @xmath70 and for @xmath71 through an intermediate @xmath45 with at least one lepton - flavor violating decay are shown in figure [ c10br ] .
the latter decay is uninteresting both because it is small and because it can not be separated from the @xmath72 final state for which both taus decay to muons .
we therefore concentrate on the @xmath73 decay followed by a hadronic tau decay .
note that the mass shifts in the lightest states are very small . for @xmath74 ,
the masses of the lightest states change by @xmath75 mev ; the heavier states are more strongly affected , their masses become 221 and 243 gev ; the mass of @xmath76 is , of course , unchanged at 231 gev .
a sample of 600000 susy events was generated with isajet 7.49 , equivalent to an integrated luminosity of @xmath77 fb@xmath78 .
the detector response is parameterized by gaussian resolutions characteristic of the atlas@xcite detector without any non - gaussian tails . in the central region of rapidity
we take separate resolutions for the electromagnetic ( emcal ) and hadronic ( hcal ) calorimeters , while the forward region uses a common resolution : @xmath79 all energy and momenta are measured in gev .
a uniform segmentation @xmath80 is used with no transverse or longitudinal shower spreading . both atlas@xcite and cms@xcite
have finer segmentation over most of the rapidity range .
an oversimplified parameterization of the muon momentum resolution of the atlas detector including both the inner tracker and the muon system measurements is used , _ viz _
@xmath81 in the case of electrons we take a momentum resolution obtained by combining the electromagnetic calorimeter resolution above with a tracking resolution of the form @xmath82 this provides a slight improvement over the calorimeter alone .
missing transverse energy is calculated by taking the magnitude of the vector sum of the transverse energy deposited in the calorimeter cells .
jets are found using getjet@xcite with a simple fixed - cone algorithm .
the jet multiplicity in susy events is rather large , so we have used a cone size of @xmath83 unless otherwise stated .
jets are required to have at least @xmath84 gev ; more stringent cuts are often used .
all leptons are required to be isolated and have some minimum @xmath85 and @xmath86 .
the isolation requirement is that no more than 10 gev of additional @xmath87 be present in a cone of radius @xmath88 around the lepton to reject leptons from @xmath89-jets and @xmath90-jets .
susy production at the lhc is dominated by the production of squarks and gluinos , which decay via cascades to the lightest susy particle @xmath64 , which escapes the detector .
events were therefore selected by requiring multiple jets plus large @xmath91 : * @xmath92 jets with @xmath93 and @xmath94 ; * @xmath95 ; * @xmath96 ; the same analysis was applied to the standard model background .
isajet 7.44 was used to generate 250000 events each of @xmath97 , @xmath98 , @xmath99 , and @xmath100 and 2500000 qcd jets , divided among five @xmath85 bins covering @xmath101 .
the cut on @xmath102 was chosen to make this background fairly small but still visible .
these background samples represent much smaller integrated luminosity that the signal sample , so the resulting statistical fluctuations are large when they are scaled to the appropriate integrated luminosity . for events that pass this selection , an additional requirement that there be two isolated leptons with @xmath103 gev and @xmath104
the dilepton mass distribution for opposite sign , same flavor pairs is shown in figure [ c10mll ] and shows the characteristic endpoint for slepton mediated decay .
the end - point occurs at @xmath105 a small @xmath106 peak is also visible , primarily from the 6.9% branching ratio for @xmath107 .
the events in the opposite sign , opposite flavor sample are due to decays of the type @xmath108 and from events with leptons arising from @xmath109 decay which are not subject to the kinematic constraint .
selection of hadronic @xmath110 decays was based on the actual hadronic tau decays plus the reconstruction efficiency and jet rejection based on the full simulation analysis@xcite done for an earlier study@xcite .
pile - up effects have been ignored , so the results are applicable only at low luminosity , @xmath111 . for each hadronic @xmath110 decay with @xmath112 and @xmath113 ,
the direction was found .
a matching jet with @xmath113 was then sought with @xmath114 and @xmath115 .
if such a jet was found , it was tagged as a @xmath110 with a probability of 66% ; its charge was assigned correctly with a probability of 92% .
jets not so tagged were assigned as @xmath110 s using an approximate parameterization of the jet/@xmath110 rejection shown in figure 9 - 31 of ref . ,
@xmath116 where @xmath117 is the hadronic @xmath110 efficiency and @xmath85 is in gev .
we choose @xmath118 .
the efficiency and the results using the parameterization of@xcite are not very different .
note that we are actually using the hadronic jet resolution for all @xmath110 decays here , not the improved resolution for a subset of decays as in the earlier analysis @xcite which was focussed on the invariant mass of a pair of taus , both decaying hadronically .
since this analysis relies mainly on counting events , the resolution on the measurement of the hadronic @xmath110 decay products is not crucial .
the visible ( hadronic ) @xmath119 mass for hadronic @xmath110 decays ( hereafter denoted by @xmath120 ) is shown in figure [ c10mtautau ] . here
events were required to have two @xmath120 candidates in addition to the cuts listed above .
note that one or both of these candidates could be the same jets that were used in the initial event selection .
the background from misidentified qcd jets is approximately random in sign and so cancels in the @xmath121 combination . to the extent that @xmath122 and @xmath123 production dominates , so does the background where both @xmath110 s come from @xmath56 decay .
figure [ c10mtautau ] shows an excess of @xmath124 pairs ending approximately at the endpoint of the dilepton distribution ( from @xmath125 ) shown in figure [ c10mll ] .
the structure is less clear due to the energy carried off by the neutrinos .
even for @xmath42 there is a substantial rate for real @xmath126 pairs from @xmath110 pairs produced by the decay chain @xmath127 .
two independent chargino decays can also give @xmath128 pairs of either sign .
misidentified qcd jets give additional @xmath129 pairs ; the signs of @xmath119 pairs from such misidentified jets should be random , and this is assumed here .
the lfv mixing from @xmath60 produces additional @xmath130 pairs , 92% of which have signs that are correctly identified .
figure [ c10mltaulfv10 ] shows the os and ss backgrounds and the signal for an assumed 10% branching ratio for the direct decay @xmath131 .
this corresponds to @xmath132 ( see figure [ c10br ] ) .
since isajet assumes lepton flavor conservation , the decay of interest was simulated by finding events with two @xmath110 s consistent with @xmath133 and with at least one hadronic @xmath110 decay and then replacing the other @xmath110 with a muon with a probability equal to the assumed branching ratio .
this is an excellent approximation since , as is indicated above , the mass shift due to mixing is very small .
figure [ c10mltaudiflfv ] shows the signal and the sign - subtracted background .
the latter should cancel in the subtracted @xmath134 distribution up to statistical fluctuations ; the lfv signal occurs only in the @xmath73 channel .
the distribution from the lfv signal has a peak at larger values of invariant mass than that of the lepton flavor conserving process as the lepton in the latter must arise from the decay of a tau and is therefore softer than that the muon from @xmath135 in figure [ c10mltaulfv10 ] for @xmath136 , there are 1089 os and 707 ss events , with equal numbers of @xmath137 and @xmath73 , and there are 518 lepton flavor violating @xmath73 events , 92% of which are correctly identified as to sign .
hence in this mass range we expect @xmath138 adding the errors in quadrature , we would then measure @xmath139 giving a @xmath140 excess for @xmath141 .
the statistical @xmath142 limit for @xmath143 would be a branching ratio of @xmath144 , corresponding to @xmath145 .
the signal to background ratio is better in the sign subtracted distributions , but the statistical errors are larger . a determination of the branching ratio from the observation of an excess of @xmath73 events requires a detailed understanding of the systematic uncertainties .
the scaling of these results to the design lhc luminosity also requires a careful study of the effects of pile up of low-@xmath85 hadronic events .
it is interesting to remark that the sensitivity that we obtain from direct slepton decay is comparable to that claimed for a lepton collider operating at 500 gev center of mass energy for @xmath143@xcite .
the decay @xmath35 is sensitive to the same lepton flavor violation as the signal considered here .
the approximate formula of ref .
, @xmath146 gives for @xmath74 and @xmath147 @xmath148 however this result is sensitive to the details of the mass spectra and mixings ; cancellations can occur resulting in much smaller rates@xcite . for comparison ,
the current bound is @xmath149@xcite .
the total production of @xmath150 at the lhc is about @xmath151 events for an integrated luminosity of @xmath152 .
a study@xcite carried out for the atlas detector concluded that a 90% confidence level upper limit of @xmath153 on @xmath154 could be reached using @xmath155 events for an integrated luminosity of @xmath143 corresponding to three years of running at low luminosity .
the process is background limited from qed final state radiation in the decays @xmath156 and @xmath157 .
these contribute approximately 50 events for @xmath143 .
we have demonstrated , by using a specific example , how flavor violation in the slepton sector can be observed at the lhc . it would be difficult to repeat the analysis for a large number of other cases . in order to estimate the generality of the method we have adopted a simpler approach .
the crucial feature is the production of staus in decays of @xmath50 . for fixed values of @xmath21 , @xmath20 and @xmath19 ,
the parameter space of @xmath158 and @xmath159 was scanned . at each point , all masses and branching ratios were computed assuming that there is no slepton flavor mixing . if the gluino is lighter than the up and down squarks , then the product branching ratio @xmath160 is computed , if the gluino is heavier than the up and down squarks the combination @xmath161 is calculated .
the total supersymmetric cross section is determined at a few points in the parameter space and can be approximated by @xmath162 this formula agrees with the cross - section to within 25% for all relavant paramters ; the dependence on @xmath21 , @xmath20 and @xmath19 is small .
an approximation of the total production rate for @xmath163 from @xmath50 arising from squark and gluino production and decay can then be obtained from the product of the cross - section and the combined branching ratio .
figures [ c3rate ] through [ c30rate ] show the resulting rates , and figure [ tmass ] shows two sample sets of contours of the @xmath163 mass .
the light higgs mass is required to be larger than 103 gev in making these plots ; this accounts for the cut off at smaller values of @xmath159 for @xmath164 .
rasing the limit to 107 gev excludes the region below @xmath165 gev for @xmath40 .
the current bound on the higgs mass for such a small value of @xmath20 is approximately 112 gev @xcite . at this value @xmath159
is so large that the production rate is less than 0.1 fb .
the effect of this bound should be taken _
cum grano salis _ as uncomputed higher order corrections could modify the relationship between the model parameters and the higgs mass .
the higgs limit has no impact on the plots with @xmath166 .
the lack of rate at large values of @xmath159 is due to the large gluino and squark masses and the consequent small production rate .
the region where @xmath167 is not accessible as there the decay @xmath168 is not allowed kinematically . for comparison ,
the case studied in detail would correspond to a @xmath45 rate of @xmath169 fb according to figure [ c10rate ] .
a comparison of this value with the rates shown in figure [ c10mtautau ] shows that the combination of acceptance and tau detection efficiency is approximately 2% .
this is not expected to vary dramatically over the relavent parameter space as the kinematics of production and decay are broadly similar .
the acceptance will be less near the boundary where @xmath170 as the resulting taus will be soft .
the acceptance is greater at larger values of @xmath158 or @xmath159 where events are more central and the resulting taus more energetic . at a luminosity of @xmath171 @xmath172
sec@xmath78 a meaningful measurement should be possible if the rate is larger than 1 pb . an examination of the plots shows that this covers a sizable fraction of parameter space .
a detailed quantitative assessment of the sensitive region is not straightforward .
this is because the background arises from supersymmetry itself and not from standard model sources .
the background therefore varies across parameter space , so a more detailed sensitivity estimate would require repeating the full analysis at each point .
the identification of @xmath120 decays requires tight cuts on the shower shape and track multiplicity , so it becomes more difficult at high luminosity , @xmath173 . for high susy masses , however ,
the @xmath120 typically have higher @xmath85 and so are more easily distinguished from qcd jets . extending the analysis described here to high luminosity requires a much more detailed and detector - specific study of pileup effects than has been done here .
if it proves to be possible , then the parameter space where the rate is larger than 0.1 pb should be accessible .
this covers almost all of the physically interesting region where the decay @xmath174 is allowed .
we have demonstrated that the observation of lepton flavor violation may be possible at the lhc .
we have focussed on the case where the largest violation occurs in the @xmath175 sector . while this is motivated by the atmospheric neutrino problem this violation is harder to detect than that occuring in the @xmath176 sector since the detection of hadronically decaying taus
is involved .
if sleptons are produced in the decays of squarks and gluinos the rates are large enough that the resulting sensitivity is greater than can be reached in rare decays of the type @xmath177 .
this corresponds to the fraction of parameter space where @xmath178 . in the sugra model
the fraction of parameter space where these decays occur with a rate large enough for observation is large .
if sleptons are not produced in these decays , the sensitivity will be much less .
a study of the possibility of observing direct slepton production for the mass spectrum considered ( without flavor violation)@xcite .
stringent jet veto cuts are required to reduce the backgrounds from other susy events .
approximately 50 events from @xmath179 production were found in @xmath143 and a mass sensitivity of 20 gev obtained .
note that this sensitivity is comparable to the mass shifts induced in the @xmath180 sector for @xmath74 .
a more detailed study is needed before a definite conclusion can be made , but it appears that this process might also be more sensitive than the @xmath181 decay .
the importance of @xmath120 decays for susy extends beyond the lepton flavor violating signal considered here .
if @xmath182 , then the decays @xmath183 and @xmath184 can be the only allowed two - body decays and so have branching ratios close to unity ; then @xmath110 signatures dominate@xcite .
for any value of @xmath20 , @xmath120 signatures can provide information both on the splitting between @xmath45 and @xmath48 and on the mixing between @xmath47 and @xmath46@xcite .
it is therefore important to continue a detailed study of @xmath120 signatures for the lhc detectors .
we are grateful to georges azuelos , john ellis , fabiola gianotti , ryszard stroynowski and takeo moroi for useful discussions .
this work was supported in part by the director , office of science , office of basic energy research , division of high energy physics of the u.s .
department of energy under contracts de ac0376sf00098 and de - ac02 - 98ch10886 .
accordingly , the u.s .
government retains a nonexclusive , royalty - free license to publish or reproduce the published form of this contribution , or allow others to do so , for u.s .
government purposes .
99 t. kajita [ super - kamiokande collaboration ] , nucl .
* 77 * , 123 ( 1999 ) ; k. s. hirata _ et al . _
[ kamiokande - ii collaboration ] , phys .
* b205 * , 416 ( 1988 ) ; y. fukuda _ et al . _
[ kamiokande collaboration ] , phys .
. * b335 * , 237 ( 1994 ) . w. w. allison _ et al . _ , phys . lett . * b391 * , 491 ( 1997 ) d. casper _ et al .
_ , phys .
lett . * 66 * , 2561 ( 1991 ) ; r. becker - szendy _ et al .
_ , phys . rev . *
d46 * , 3720 ( 1992 ) .
j. ellis , m. e. gomez , g. k. leontaris , s. lola and d. v. nanopoulos , hep - ph/9911459 .
j. l. feng , y. nir and y. shadmi , phys .
rev . * d61 * , 113005 ( 2000 ) .
j. hisano , t. moroi , k. tobe and m. yamaguchi , phys .
* d53 * , 2442 ( 1996 ) [ hep - ph/9510309 ] .
l. alvarez - gaume , j. polchinski and m.b .
wise , nucl .
b221 * , 495 ( 1983);l .
ibaez , phys . lett . *
118b * , 73 ( 1982);j.ellis , d.v .
nanopolous and k. tamvakis , phys
121b * , 123 ( 1983);k . inoue _
. phys . * 68 * , 927 ( 1982);a.h .
chamseddine , r. arnowitt , and p. nath , phys .
* 49 * , 970 ( 1982 ) .
for reviews see , h.p .
nilles , phys . rep . *
111 * , 1 ( 1984);h.e .
haber and g.l .
kane , phys .
rep . * 117 * , 75 ( 1985 ) .
f. paige and s. protopopescu , in _ supercollider physics _
d. soper ( world scientific , 1986);h .
baer , f. paige , s. protopopescu and x. tata , in _ proceedings of the workshop on physics at current accelerators and supercolliders _ , ed .
j. hewett , a. white and d. zeppenfeld , ( argonne national laboratory , 1993 )
. i. hinchliffe , f. e. paige , m. d. shapiro , j. soderqvist and w. yao , phys .
* d55 * , 5520 ( 1997 ) .
l. poggioli , g. polesello , , e. richter - was , j. soderqvist , `` precision susy measurements with atlas for sugra point 5 '' , atl - phys-97 - 111 .
p. bock _ et al .
_ [ aleph , delphi , l3 and opal collaborations ] , cern - ep-2000 - 055 .
atlas collaboration , _ technical proposal _ ,
lhcc / p2 ( 1994 ) .
cms collaboration , _ technical proposal _ ,
lhcc / p1 ( 1994 ) .
atlas collaboration , `` atlas physics and detector performance technical design report , '' lhcc 99 - 14/15 .
y. coadou , atlas internal note atl - phys-98 - 126 .
i. hinchliffe and f. e. paige , phys .
rev . * d61 * , 095011 ( 2000 ) j. hisano , t. moroi , k. tobe and m. yamaguchi , phys
. lett .
* b391 * , 341 ( 1997 ) r. m. barnett _ et al .
_ , rev .
* 68 * ( 1996 ) 611 .
l. serin and r. stroynowski atl - phys-97 - 114 ( 1997).``study of lepton number violating decay tau - mu gamma in atlas '' , j. hisano , hep - ph/9906312 .
i. hinchliffe and f. paige , in preparation . | recent results from super kamiokande suggest @xmath0 mixing and hence lepton flavor violation . in supersymmetric models ,
this flavor violation may have implications for the pattern of slepton masses and mixings .
possible signals for this mixing in the decays of sleptons produced at the lhc are discussed .
the sensitivity expected is compared to that of rare decays such as @xmath1 .
* lepton flavor violation at the lhc * * i. hinchliffe@xmath2 and f.e . paige@xmath3 * @xmath2_lawrence berkeley national laboratory , berkeley , ca _
@xmath3_brookhaven national laboratory , upton , ny _ |
in field theory , a metastable false vacuum may decay either by tunneling or by thermal activation .
tunneling is described by a solution of the euclidean field equations symmetric under spacetime rotations @xcite . in flat space and at zero temperature
this is the instanton with the least action , and hence represents the dominant contribution to the decay rate @xcite . at sufficiently high temperature ,
thermal activation is more probable than tunneling , and the rate is dominated by a static solution which represents a spherical critical bubble in unstable equilibrium between expansion and collapse @xcite .
the symmetry of this solution is o(3)@xmath0 u(1 ) , where the u(1 ) factor corresponds to translations along the compactified euclidean time direction .
as shown by coleman and de luccia @xcite , gravity can be easily incorporated into the description of tunneling .
when the initial state is a false vacuum with a positive energy density , the initial geometry corresponds to a de sitter - like exponential expansion .
a bubble which materializes through quantum tunneling has zero energy , and consequently ( in the thin wall limit ) the geometry outside of the bubble remains unaffected by the nucleation event .
after nucleation , the bubble walls accelerate outward , and the volume of the new phase increases at the expense of the old one .
however , due to the presence of event horizons in de sitter , the growth of a single bubble can not engulf the whole space . if the nucleation rate per unit volume @xmath1 is large compared to @xmath2 , where @xmath3 is the inverse de sitter radius , the foam of nucleated bubbles will percolate and the phase transition will complete .
but if @xmath4 , the rate at which bubbles nucleate and grow does not catch up with the exponential expansion of the false vacuum . in this case , the volume of the false vacuum keeps increasing with time and the transition never fully completes in the whole spacetime ( leading to eternal old inflation )
. however , any observer will experience a local transition to the new vacuum phase in a finite proper time .
on the other hand , the description of thermal activation can be more involved when the self - gravity of the bubbles and of the thermal bath is considered . in the cosmological context ,
the thermal bath drives the expansion of the universe , and the temperature becomes time dependent . because of that , exact instantons can not be constructed . in low energy cosmological phase transitions ( e.g. at the electroweak scale )
it is safe to ignore the self - gravity of the bubbles , and to use the flat spacetime results for the nucleation rate .
if this rate is sufficiently large , the phase transition completes as the bubbles percolate , typically after a short period of supercooling and subsequent release of latent heat by the nucleated bubbles @xcite .
but if the rate is too small , there is a strong supercooling and the thermal bath dilutes away ; the false vacuum starts dominating before the phase transition is complete , and we are back to the situation described at the end of the previous paragraph @xcite .
note , however , that even when all matter has been diluted away , the false vacuum dominated de sitter expansion can be considered to have a non - vanishing temperature @xcite @xmath5 , which does not dilute further as the universe expands .
one may then ask whether this leads to thermal activation , and if so , at what rate does it proceed . in
what follows , we shall discuss the corresponding instanton , describing the nucleation of a pair of critical bubbles in unstable equilibrium between expansion and collapse .
it should be noted that the theorem in ref .
@xcite does not necessarily apply to de sitter space , and hence it is not clear a priori whether this new instanton has higher action than the usual coleman - de luccia one .
static self - gravitating instantons with o(3 ) symmetry have previously been considered in a variety of contexts , notably for the description of false vacuum decay in the presence of a black hole ( see e.g. @xcite and references therein ) .
the particular solution we shall consider here corresponds to pair creation of critical bubbles in de sitter , and to our knowledge it does not seem to have received much attention in the past . the following is an extended version of the discussion given by the present authors in @xcite . in section
ii we describe the solution . in section
iii we discuss the action and the nucleation rate . in section
iv we consider the limit in which the mass of the nucleated bubbles is small .
section v is devoted to the opposite limit , when the gravity of the bubbles is very important . in section
vi we compare the action for thermal activation with the action for tunneling ( through the coleman - de luccia instanton ) .
section vii compares the process of thermal activation of seeds of the new phase with a related process recently discussed by gomberoff et al .
@xcite , by which most of space would jump to the new phase except for a pair of bubbles which contain the remnant " of the old phase .
section viii is devoted to conclusions .
unlike the case of the coleman - de luccia bubble , the energy of a critical bubble is different from zero , and consequently , the metric outside of it is no longer pure de sitter but schwarzschild - de sitter ( sds ) .
the instanton is a solution of the euclidean equations of motion , with two metrics glued together at the locus of the wall , which is a surface of constant @xmath6 in the static chart of sds ( see fig .
[ estatico ] ) . for simplicity
, we shall restrict attention to the case where the vacuum energy density is positive in both the initial and the final states .
also , we shall assume that the thin wall approximation is valid @xcite .
the metric outside is given by @xmath7 where @xmath8 , and @xmath9 the metric inside is given by @xmath10 where @xmath11 here @xmath12 is newton s constant .
the parameter @xmath13 is determined by the condition that on the bubble wall ( i.e. , at @xmath14 ) the two metrics must agree , which leads to @xmath15^{1/2 } $ ] . [ ] [ r]@xmath16[][r]@xmath17[][r]@xmath18 [ ] [ r]@xmath19 [ ] [ l]@xmath20 the parameters @xmath21 and @xmath17 depend on the wall tension @xmath22 , and the hubble parameters outside and inside the bubble , @xmath23 and @xmath24 respectively .
their values are determined by the junction conditions at the bubble wall @xcite , @xmath25 = -4\pi g\sigma \gamma _ { ab } , \label{discont2}\ ] ] where @xmath26 $ ] is the difference in the extrinsic curvature @xmath27 on the two sides and @xmath28 is the world - sheet metric .
eq.([discont2 ] ) gives rise to the junction conditions , @xmath29=-4\pi g \sigma,\quad [ g']=0 , \label{juco}\ ] ] where we have introduced the new function @xmath30 . using eqs .
( [ efeo ] ) and ( [ efei ] ) , we have @xmath31 from ( [ juco ] ) , we have @xmath32 , and then @xmath33 and @xmath34 are easily obtained from eqs .
( [ cuadradosprima ] ) : @xmath35 from ( [ efeo ] ) and ( [ efei ] ) we have @xmath36 inserting ( [ ges ] ) in ( [ cuadrados ] ) we finally obtain a quadratic equation for @xmath37 .
the solution is @xmath38 ^{1/2 } , \label{x}\ ] ] where we have introduced the parameter @xmath39 representing the difference in vacuum energy difference on both sides of the bubble wall : @xmath40 .
then the parameters @xmath21 and @xmath17 are given in terms of @xmath41 by @xmath42 this concludes the construction of the instanton solution for given values of the physical parameters @xmath43 and @xmath44 .
the above equations are valid only as long as @xmath45 [ otherwise ( [ ges ] ) would yield @xmath46 , which is meaningless ] .
thus , from ( [ themass ] ) , we require @xmath47 , or @xmath48 , where @xmath49 ( the case with @xmath50 will be discussed in section v. ) the mass parameter satisfies @xmath51 for @xmath52 , the equation @xmath53 has three real solutions
. one of them , say @xmath54 , is negative and the other two are positive .
the two positive roots correspond to the black hole and cosmological horizons .
we call them respectively @xmath55 and @xmath56
. therefore we can write @xmath57 in the present case , with @xmath48 , the horizon at @xmath55 is not present , since the exterior metric is matched to an interior metric at some @xmath58 ( see fig .
[ estatico ] ) . for @xmath59
the metric is just a ball of de sitter in the static chart , and it is regular down to the center of symmetry at @xmath60 . in general , the size of the cosmological horizon is given by @xmath61 where we have introduced the angle @xmath62 in the limit @xmath63 the angle @xmath64 , and @xmath65 . according to eq ( [ ges ] ) , on the bubble wall we have @xmath66 , so the equation @xmath67 has a double zero instead of two different roots .
this means that the radius of the instanton will coincide with the radius of one of the horizons only in the special case where both horizons have the same size , @xmath68 . as we shall see in section v
, this limit corresponds to @xmath69 , for which the exterior metric is the nariai solution @xcite , with mass parameter @xmath70 given in ( [ mmax ] ) and with @xmath71 .
there , we shall also comment on the case @xmath50 , which is not covered by the present discussion .
regularity of the euclidean solution will determine the periodicity of the euclidean time coordinate ( and the thermal properties of the solution ) .
for @xmath72 , we have @xmath73 where @xmath74 in terms of the new coordinates @xmath75
the metric ( [ metrico ] ) for @xmath76 reads @xmath77 so it is clear that @xmath78 is an angle , @xmath79 , and @xmath19 varies in the range @xmath80 .
thus , the periodicity of the time coordinate is given by @xmath81 it is also of some interest to determine the physical temperature on the bubble wall worldsheet , given by the proper time periodicity @xmath82 , @xmath83 if there are field degrees of freedom living on this worldsheet , this will be the temperature that they will experience ( rather than the ambient de sitter temperatures ) . like in the case of instantons describing the production of black holes @xcite or monopoles @xcite in de sitter
, the instanton presented here describes the creation of _ pairs _ of bubbles .
as we have just seen , the euclidean time runs on a circle @xmath84 ( see fig .
the geometry at the time of nucleation is obtained by slicing the compact instanton through a smooth spacelike surface which cuts the @xmath84 factor at two places , say , @xmath85 and @xmath86 .
the resulting geometry contains two different bubbles separated by a distance comparable to the inverse expansion rate .
the nucleation rate is determined by the euclidean action , which turns out to have a rather simple expression in terms of @xmath16 .
the action is given by @xmath87 by the equations of motion , the scalar curvature is given by @xmath88 and hence the on shell action is given by @xmath89 the first integral in ( [ actionstatic ] ) is just the volume of a two - sphere of radius @xmath17 times @xmath90 .
the second integral in ( [ actionstatic ] ) splits into the contributions from the two different vacua , @xmath91 so the instanton action is @xmath92 after some algebra @xmath93 can be written in the simple form @xmath94 where @xmath95 is the area of the horizon at @xmath16 .
the exponent @xmath96 which gives the probability for brane nucleation is the difference in actions between instanton and background .
the action of the background is just @xmath97 so the difference in actions between the instanton and the background is given by @xmath98 this expression could have been anticipated from general arguments .
the action of a static euclidean solution with periodicity @xmath99 is given by @xcite @xmath100 , where @xmath101 is the _ total _ energy and @xmath102 is the entropy . for a closed solution , without a boundary , the total energy vanishes @xmath103 .
hence , the euclidean action is equal to the entropy of the solution , which is just one fourth of the sum of the areas of all horizons .
hence , eq . ( [ b ] ) can be rewritten as @xmath104 , and the transition probability is proportional to @xmath105 as both the initial and the final solutions can be said to represent macroscopic states in thermal equilibrium , their entropy can be interpreted as the logarithm of the corresponding number of microstates in the microcanonical ensemble with @xmath103 .
the transition probability is then simply proportional to the relative number of microstates .
if the bubbles are sufficiently light , @xmath106 , and provided that @xmath107 , eq .
( [ varphi ] ) gives @xmath108 , and from ( [ b ] ) we have @xmath109 where @xmath110 .
since the mass of the bubble is small , its appearance does not significantly change the temperature of the horizon . in such case
, the nucleation rate of the bubbles may be interpreted from the point of view of the observer at @xmath60 as being due to a thermal bath at the fixed gibbons - hawking temperature @xmath111 .
the corresponding probability is proportional to the boltzman factor @xmath112 eq .
( [ bm ] ) can also be understood as follows .
the energy of the bubble has been extracted from thermal reservoir . according to the first law ,
the entropy of the reservoir must decrease by @xmath113 @xcite . from the discussion at the end of the last section , @xmath114 , and hence for bubbles of small mass we obtain ( [ bm ] ) .
let us now consider a few specific limiting cases , starting with the case of low tension branes , @xmath115 , with @xmath116 . in this case
the parameter @xmath41 is large compared with @xmath23 , @xmath117 , and we have @xmath118 this is just the flat space expression for the energy of a critical bubble .
the corresponding bounce action is @xmath119 which coincides with the thermal activation rate in flat space at the temperature @xmath120
. at finite temperature , jumps to a vacuum with a higher energy density are also possible . in the absence of gravity , these jumps are frustrated because the bubbles of the new phase tend to recollapse .
when gravity is included , the expansion of the universe can keep these false vacuum " bubbles from contracting ( this is true also for the case of tunneling bubbles @xcite ) .
hence , let us consider again the case of low tension branes , @xmath121 , but now with @xmath122 .
in this case we find @xmath123 , and @xmath124\approx h_i^{-1}.\ ] ] in this limit , the bubble of the false vacuum phase is almost as big as the cosmological horizon .
we also have @xmath125 and @xmath126 , where we have used that @xmath127 to replace @xmath44 by @xmath128 in the last expression ( note that @xmath129 in the case we are considering here , so @xmath21 and @xmath96 are both positive ) .
this approximately coincides with the bounce action for the homogeneous hawking - moss instanton @xcite , representing the upward jump of a horizon sized region of de sitter space into a higher false vacuum .
finally , we may consider the case of intermediate tension @xmath130 .
this leads to @xmath131 , and @xmath132 . in this case
, the difference in pressure between inside and outside of the brane is insignificant compared with the brane tension term , which is balanced against collapse by the cosmological expansion .
the energy of the critical bubble is @xmath133 .
note , from ( [ betar ] ) , that the inverse temperature @xmath134 is different from the one experienced by a geodesic observer at the origin of coordinates @xmath60 .
this is because observers at @xmath135 are in fact accelerating . from the point of view of the observer at @xmath60 ,
the energy of the bubble is @xmath136 , because of the gravitational potential contribution .
hence , taking into account that @xmath137 the exponent in the boltzmann suppression factor can be written as @xmath138 , and we have @xmath139 .
for given @xmath128 and @xmath44 , the solution of section ii only exists provided that the tension of the bubble wall does not exceed a certain bound @xmath140 , given in eq .
( [ sigmamax ] ) .
let us now consider what happens near this bound , and beyond . as we mentioned in the discussion below eq .
( [ fofactors ] ) , the exterior metric in the limit @xmath141 corresponds to the nariai solution , with @xmath142 , and @xmath143 . replacing this value in ( [ b ] )
we find readily @xmath144 this may be compared with the action of the instanton describing the nucleation of black holes in the same de sitter universe @xcite , @xmath145 the difference @xmath146 , is just the area of the black hole horizon in the nariai solution , as expected from the general discussion of the previous section ( here , we are of course neglecting the entropy stored in the field degrees of freedom living on the bubble walls , which would show up when the determinantal prefactor in the nucleation rate is evaluated ) . the fact that @xmath147 does not mean that both horizons coincide , since the coordinates @xmath148 become inadequate in this case @xcite . near the nariai limit
the metric outside takes the form ( [ metrico ] ) , with @xmath149 } + r_+^{2}d\omega ^{2}\ ] ] and @xmath150 , plus higher orders in the parameter @xmath151 , which we defined in ( [ a ] ) . in the present limit this parameter tends to zero , @xmath152 .
now we define new coordinates @xmath153 and @xmath154 by @xmath155 so that the metric becomes @xmath156 the cosmological horizon is at @xmath157 and the black hole horizon is at @xmath158 .
now in the limit @xmath159 we just replace @xmath160 .
we must determine the position @xmath161 of the bubble wall , which is given as before by the matching conditions ( [ discont2 ] ) , where now the metric outside is ( [ nariai ] ) .
so , on the wall , we have @xmath162 the extrinsic curvature on the outside is @xmath163 , with @xmath164 and @xmath165 , i.e. , @xmath166 the curvature inside is as before @xmath167 and @xmath168 , with @xmath169 , so the israel conditions give @xmath170 these equations are easily solved and give @xmath171 so @xmath24 must be less than @xmath172 .
now regularity at the cosmological horizon @xmath173 implies that @xmath174 , so @xmath175 .
hence , @xmath176 thus , also in this case , the effective temperature of the field degrees of freedom living on the worldsheet will be of order @xmath177 ( the only exception occurs if there is some fine adjustment between @xmath128 and @xmath44 which makes the factor inside the brackets very small , in which case the temperature may be much larger . ) for @xmath178 we have @xmath179 and the construction of section ii does not apply [ since by eq .
( [ ges ] ) , @xmath180 would be negative ] .
as pointed out in @xcite , above this threshold a static solution can still be constructed by gluing the interior solution ( [ metrici ] ) to the @xmath181 portion of the exterior sds solution ( [ metrico ] ) ( rather than using the @xmath182 portion ) .
this changes the sign of @xmath183 in the exterior , and the junction conditions become @xmath184 where the curly brackets denote twice the average value on both sides of the bubble wall .
( [ ges ] ) is then replaced by @xmath185 but eqs .
( [ x ] ) through ( [ themass ] ) remain the same .
the instanton would still look pretty much as in fig . 1 , but with the cosmological horizon of radius @xmath16 replaced by a black hole horizon of radius @xmath186 .
hence , in the right pannel of fig .
1 , the horizontal maximal circles would grow from @xmath18 to @xmath17 as we move up from the center of the bubble , but then the circles would start decreasing from @xmath17 to @xmath55 as we continue from the bubble wall to the horizon .
it is straightforward to calculate the euclidean action for this solution , which is given by @xmath187 where @xmath188 is the area of the black hole horizon , with @xmath189 and where @xmath190 is given by ( [ varphi ] ) .
the corresponding bounce action @xmath191 is perfectly finite , since the instantons involved are both compact and regular .
moreover , @xmath192 , as it should be if this is to be interpreted as a process with an exponentially suppressed rate .
is this instanton suitable for describing vacuum decay in the usual sense ?
let us assume that we are in a false vacuum phase , and for simplicity , that the false vacuum decay rate per unit volume is exceedingly small compared with @xmath2 .
then we expect that after some time the metric will take the form @xmath193 over an exponentially large portion of space ( with the exception of small portions of volume carved out by bubbles of the new phase which may have nucleated ) . in the solution described in section ii
, an asymptotic region with metric ( [ dsflat ] ) can be found ( upon analytic continuation ) in the region beyond the cosmological horizon , which is asymptotically de sitter and infinite in volume in a flat slicing . in the case we are considering in this subsection
, however , the global structure of the solution is rather different .
a black hole singularity is hidden beyond @xmath55 , and the static solution does not contain any asymptotic region that looks like ( [ dsflat ] ) . [ ] [ bl]@xmath20[][r]@xmath17[][bl]@xmath18 [ ] [ r]@xmath55 the @xmath194 surface has the geometry of an einstein - rosen bridge of the old phase which connects a pair of bubbles of the new phase ( see fig . [ bridge ] ) .
each one of these bubbles is in unstable equilibrium , it can either expand or contract .
let us concentrate in one of them .
if the bubble wall expands , then its motion is perceived as expansion from both sides of the wall ( recall that in the present case the radius @xmath6 decreases as we move away from the wall in both directions ) .
conversely , contraction of the bubble wall would be perceived as contraction from both sides of the bubble wall .
if one of the bubbles expands , it eventually generates an infinitely large region of the false vacuum phase surounding the black hole , and the metric in the false vacuum region far away from the black hole has the asymptotic form ( [ dsflat ] ) .
this suggests the following interpretation for the static instanton beyond the nariai limit : it describes the thermal production of black holes of mass @xmath21 [ given by ( [ themass ] ) ] in an asymptotically de sitter region .
initially , the throat of the black hole connects with a compact baby universe , but this pinches off as the black hole singularity develops @xcite .
the baby universe contains a bubble of the new phase in unstable equilibrium ( see fig .
[ gota2 ] ) .
if the bubble of the new phase collapses , the baby universe disappears into nothing . on the contrary , if the unstable bubble expands , it ends up generating an infinite region of the new vacuum phase , separated from an infinite region of the old vacuum phase by a domain wall in constant acceleration .
at the center of the region of the old vacuum phase , there is also a black hole of mass @xmath21 . [ ] [ tl]@xmath195[][r]@xmath196 [ ] [ r]@xmath197 let us now comment on the nucleation rate . according to eq .
( [ b2 ] ) , this is given by @xmath198 here , @xmath199 is the black hole entropy . in the last step ,
we have used that for black holes of sufficiently low mass , the entropy of the cosmological horizon is smaller than the entropy of the original de sitter metric by the amount @xmath200 , as discussed in section iv . when we compare the previous result with eq .
( [ boltsu ] ) , the last factor in eq .
( [ unex ] ) strikes us as rather unexpected .
it seems to say that of all attempts at forming a black hole of mass @xmath21 in a region with effective de sitter temperature @xmath201 , only a very small fraction given by @xmath202 $ ] succeed in forming a baby universe which hosts a bubble of the new vacuum phase .
perhaps this is not unreasonable , since a baby universe which is entirely filled with the old vacuum phase would have an entropy which is higher by the amount @xmath203 , relative to the entropy of the baby universe containing the static bubble .
this suggests that most attempts should produce a baby universe of the old phase , without a bubble of the new phase .
however , we should also keep in mind that the instanton representing this alledgedly more frequent process does not exist ( the solution would contain two horizons at different temperatures , and hence the euclidean section would have a conical singularity at one of them ) .
mathematically , the factor @xmath202 $ ] arises because the instanton represented in fig .
[ bridge ] does not contain the cosmological horizon at @xmath16 .
the neighborhood of this horizon has been excised and replaced with the bubble of the new vacuum phase .
this could mean that the interpretation given above for the instanton beyond the nariai limit is not correct . in this interpretation
, we are assuming the existence of an initial region , of size larger than the cosmological horizon , where the metric takes the approximate form ( [ dsflat ] ) .
a cosmological horizon , and an asymptotic de sitter region of the form ( [ dsflat ] ) , does develop if we let one of the unstable bubbles expand , but strictly speaking it is not present in the analytic continuation of the instanton . clearly , the legitimacy of this interpretation deserves further investigation .
[ ] [ tl]@xmath204[][r]@xmath196 [ ] [ r]@xmath197 let us now compare the action of the thermal instanton with that of the tunneling process described by the coleman - de luccia ( cdl ) instanton
. the latter is given by ( see e.g. @xcite ) @xmath205,\ ] ] where @xmath206 and @xmath207 the @xmath208 s are related by @xmath209 .
using this relation and @xmath210 , @xmath211 can be written as @xmath212.\ ] ] the values of @xmath96 for the static instanton and @xmath211 are easily compared by noticing that both are of the form @xmath213 , times a function of the dimensionless parameters @xmath214 and @xmath215 . in fig .
[ cdl ] we plot the action for the cdl case , and in fig .
[ deltab2 ] we plot the difference between the two actions .
note that the static instanton action is larger than the cdl action in the whole range of parameters .
[ ] [ tl]@xmath216[][r]@xmath196 [ ] [ r]@xmath197 as we mentioned in section iv , jumps to a vacuum with higher energy density are also allowed @xcite .
note that for the case of upward jumps , @xmath217 , the actions become comparable , and in fact they are equal at the corner where @xmath218 and @xmath219 ( see figs . [ deltab2 ] and [ deltab1 ] ) .
[ ] [ tl]@xmath6[][l]@xmath196 [ ] [ r]@xmath197
[ ] [ tl]@xmath220[][l]@xmath196 [ ] [ r]@xmath197 in the interpretation which we have adopted so far , the static instanton represents the creation of pairs of critical bubbles of the new phase embedded in the false vacuum phase @xcite .
we may refer to this as the process of pair creation of seeds " of the new phase .
this process is analogous to pair creation of particles ( or even topological defects such as monopoles @xcite ) by the expanding de sitter background . as we showed in section iv , when the nucleated objects are sufficiently light , the creation rate is simply proportional to the boltzmann factor . in ref .
@xcite , gomberoff et al . suggested a rather different interpretation of the same solution .
the process they considered involves a spherical bubble wall coming in from the cosmological horizon , sweeping away the false vacuum as it moves towards smaller radii , and replacing it with the true vacuum .
the result of this process would also be a critical bubble in unstable equilibrium between expansion and collapse , but this time the bubble would be a remnant " of the old phase rather than the seed of the new phase .
we shall thus refer to this process as creation of remnants " . even if mathematically
the euclidean solution is the same as before , the interpretation and background subtractions are very different . as a consequence ,
the nucleation rate of such objects does _ not _ have the same simple boltzmann suppression form as we found in ( [ boltsu ] ) for light bubbles .
suppose for definiteness a potential with two non - degenerate vacua , labeled by 1 and 2 , with @xmath221 .
the solution representing a downward jump which leaves a remnant of vacuum 1 surrounded by vacuum 2 is the same as the instanton for an upward jump caused by a seed of vacuum 1 which has been activated from vacuum 2 .
similarly , the upward jump which leaves a remnant of vacuum 2 surrounded by vacuum 1 is related to downward jumps by activation of a seed of vacuum 2 from vacuum 1 .
hence @xmath222 here , the subindex @xmath223 stands for creation of remnants " , and @xmath224 denotes the action for the cosmological thermalon discussed by gomberoff et al . , while the subindex @xmath225 refers to creation of seeds " , and @xmath226 denotes the action given in ( [ area ] ) .
the arrows indicate whether we are considering an upward jump or a downward jump .
the bounce action is obtained by performing the relevant background subtractions @xmath227 here @xmath228 and @xmath229 are the background actions , given by eq .
( [ area ] ) , with @xmath16 replaced by the corresponding de sitter radii @xmath230 and @xmath231 respectively . before proceeding
, we should stress that since the instantons we are considering are static and compact , then according to the discussion in section iii the bounce actions are always given by and to gravity . the term which represents
the interaction of the membrane with the gauge field takes the form @xmath232 , where @xmath233 is the membrane charge , and the integral is over the membrane worldsheet .
it was argued in @xcite that @xmath234 is discontinuous accross the membrane , and a somewhat heuristic prescription was given to compute the contribution of @xmath232 to the action and to perform the background subtraction .
the result of this procedure , however , differs from eq .
( [ biarea ] ) . here
, we shall not try to elucidate the reason for this discrepancy .
we note , however , that the on - shell euclidean action for the system of a membrane coupled to @xmath234 and to gravity , and with proper inclusion of boundary terms @xcite , can be shown to be the same as the action we have taken as our starting point ( [ actionstatic ] ) with @xmath235 replaced by @xmath236 , where @xmath237 is the field strength [ see e.g. eq . ( 6.1 ) in ref .
hence , the results of the present paper , which are in principle valid for the case of vacuum decay in field theory , may as well be valid for the case of the brane coupled to the antisymmetric tensor field . ]
@xmath238 where @xmath239 is the change in the area of the horizon and @xmath240 is the change in the entropy .
it is clear from ( [ tigger ] ) that @xmath241 here , we have used the fact that @xmath242 , since the cdl instanton solution is the same for upward and for downward jumps , and all that changes is the background subtraction @xcite .
the inequalities above come from the fact that @xmath243 is always larger than @xmath211 , as shown in the previous section .
it follows that @xmath244 is also always larger than @xmath211 , and so the creation of remnants is also subdominant with respect to the tunneling process represented by the cdl instanton .
finally , we may ask which of the two processes is more important , the creation of seeds or the creation of remnants . from ( [ arrows ] )
we have @xmath245 hence , if one of the channels is dominant for upward jumps , then it means that the other process is dominant for downward jumps .
[ staticsobreterma ] shows the ratio @xmath246 .
note that if @xmath247 , corresponding to downward jumps , then the process of pair creation of seeds is much more likely than the process of pair creation of remnants . on the other hand , for @xmath248
, corresponding to upward jumps , the ratio of the bounce actions is very close to one [ although , from ( [ relat ] ) , the frequency of upward jumps through creation of remnants outweighs that of upward jumps through creation of seeds by the same factor as downward jumps through creation of seeds outweight downward jumps through creation of remnants . ] finally , let us recall that the nariai limit corresponds to @xmath249 [ see eq .
( [ sigmamax ] ) ] . here
the indices @xmath250 and @xmath251 stand for the inside and the outside of the bubble . for pair creation of seeds , outside and inside
correspond to the original vacuum and the final vacuum respectively @xmath252 and @xmath253 . in the dimensionless variables
@xmath254 and @xmath255 , the nariai curve corresponds to @xmath256 for pair creation of remnants with the same initial and final states as the seeds , we have @xmath257 and @xmath258 , and the nariai limit corresponds to @xmath259 the nariai curves @xmath260 and @xmath261 , corresponding to the circle ( [ circle ] ) and the hyperbola ( [ hyperbola ] ) are also plotted in fig .
[ staticsobreterma ] .
de sitter vacua are believed to be metastable at best ( see e.g. @xcite for a recent discussion ) .
it is well known that vacuum transitions from a metastable vacuum can proceed through quantum tunneling , which is described by the coleman - de luccia instanton .
this process can take us to lower energy vacua , but also to other de sitter vacua with a higher vacuum energy density @xcite . here
, we have investigated an alternative process , by which critical bubbles of the new phase can be pair produced .
this process is the analog of thermal activation in flat space .
the mass @xmath21 and radius @xmath17 of the `` seeds '' of the new phase are given by eqs .
( [ x ] ) and ( [ themass ] ) , in terms of the initial and final vacuum energies , @xmath262 and @xmath263 , and of the tension @xmath22 of the wall separating both phases . for @xmath264 , the geometry of the critical bubbles is the following ( see fig .
1 ) . outside the bubble , the metric is schwarzschild - de sitter , and has a cosmological horizon .
the black hole horizon is not present , since we are matching to an interior solution at some @xmath265 , where @xmath17 is the bubble radius and @xmath55 is the radius of the would be black hole horizon .
inside the bubble , the metric is pure de sitter with curvature radius @xmath266 . for @xmath267 ,
the nucleation rate is proportional to the boltzmann factor @xmath268 as would be expected from simple thermodinamical arguments . here
@xmath269 is the inverse de sitter temperature of the old vacuum phase .
for @xmath270 the metric outside of the bubble corresponds to the nariai limit of the schwarzschild - de sitter solution , for which the black hole and cosmological horizons have the same size . beyond the nariai limit ,
i.e. for @xmath271 , the asymptotic form of the solution changes quite drastically @xcite .
the static solution with a pair of critical bubbles has a black hole horizon instead of a cosmological horizon .
the interpretation of such solution is less clear than in the case @xmath272 , but we have argued that it may correspond to the creation of a baby universe containing a bubble of the new phase .
the nucleation rate is formally given by ( [ unex ] ) , and does not have the simple form ( [ conc ] ) even in the case when the mass @xmath21 is small ( here , @xmath21 is the mass of the black hole connecting the asymptotic region of the old phase with the baby universe ) .
we have compared the process of thermal activation of seeds to an alternative process recently suggested by gomberoff et al .
@xcite , by which most of space would suddenly jump to the new vacuum phase , leaving only a pair of critical bubbles as remnants of the old phase .
these could subsequently collapse into black holes , with the net result that the vacuum dark " energy , is transformed into cold dark matter " in the form of black holes .
we find that for downward jumps , this process is subdominant with respect to the thermal activation of seeds of the new vacuum . for upward jumps ,
the bounce actions are comparable , and in fact the creation of remnants may be slightly favoured with respect to the creation of seeds ( although when we are going up in energy we are not transforming dark energy into dark matter , but simply increasing both of them ! ) . also , we have compared the rate of nucleation of critical bubbles by thermal activation with the rate of bubble nucleation by quantum tunneling , described by the coleman - de luccia ( cdl ) instanton .
the cdl instanton always has a lower bounce action than the process of thermal activation of seeds or remnants .
thus , even if thermal activation is possible , it appears that jumps between neighboring vacua will be more frequent through quantum tunneling .
for the case of upward jumps , however , the corresponding actions are comparable ( see fig .
since the action for thermal activation is higher than that for tunneling , one should ask whether there are any situations where the former process may nevertheless be relevant . note that if the bubble wall carries some internal degrees of freedom , their entropy will be accounted for in the prefactor which accompanies the leading expression @xmath273 for the nucleation rate .
it is clear from eq .
( [ hite ] ) that the temperature the bubble wall can be very high if the wall tension and the vacuum energies in the two phases are suitably adjusted .
hence the entropy of the internal degrees of freedom can be very high , making up perhaps for the difference in actions .
investigation of this possibility is left for further research .
j.g . is grateful to the organizers of the 8th peyresq meeting for hospitality , and to r. emparan and m. porrati for useful discussions .
a.m. was supported by a postdoctoral fellowship of the ministerio de educacin y cultura , spain .
the work by j.g . was supported by cicyt research projects fpa2002 - 3598 , fpa2002 - 00748 , and dursi 2001-sgr-0061 .
s. r. coleman , v. glaser and a. martin , commun . math .
* 58 * ( 1978 ) 211 . a. d. linde , nucl .
b * 216 * ( 1983 ) 421 [ erratum - ibid .
b * 223 * ( 1983 ) 544 ] .
s. coleman and f. de luccia , phys . rev . *
d21 * , 3305 ( 1980 ) ; j. d. brown and c. teitelboim , phys . lett . *
b195 * , 177 , ( 1987 ) ; nucl . phys . * 297 * , 787 ( 1988 ) .
g. w. gibbons and s. w. hawking , phys .
d * 15 * , 2738 ( 1977 ) .
k. sato , m. sasaki , h. kodama and k. i. maeda , prog .
phys . * 65 * ( 1981 ) 1443 ; h. kodama , m. sasaki , k. sato and k. i. maeda , prog .
phys . *
66 * ( 1981 ) 2052 ; k. sato , h. kodama , m. sasaki and k. i. maeda , phys .
b * 108 * ( 1982 ) 103 ; h. kodama , m. sasaki and k. sato , prog .
* 68 * ( 1982 ) 1979 ; s. k. blau , e. i. guendelman and a. h. guth , phys .
d * 35 * ( 1987 ) 1747 ; e. farhi and a. h. guth , phys .
b * 183 * ( 1987 ) 149 ; e. farhi , a. h. guth and j. guven , nucl .
b * 339 * ( 1990 ) 417 .
j. garriga and a. megevand , arxiv : hep - th/0310211 .
d. ( in press ) .
a. gomberoff , m. henneaux , c. teitelboim and f. wilczek , `` thermal decay of the cosmological constant into black holes , '' arxiv : hep - th/0311011 .
m. banados , c. teitelboim and j. zanelli , phys .
* 72 * ( 1994 ) 957 [ arxiv : gr - qc/9309026 ] .
s. w. hawking and g. t. horowitz , class .
* 13 * ( 1996 ) 1487 [ arxiv : gr - qc/9501014 ] . c. teitelboim ,
`` gravitational thermodynamics of schwarzschild - de sitter space , '' arxiv : hep - th/0203258 . k. m. lee and e. j. weinberg , phys . rev .
d * 36 * ( 1987 ) 1088 ; j. garriga , phys .
d * 49 * ( 1994 ) 6327 [ arxiv : hep - ph/9308280 ] ; phys . rev .
d * 49 * ( 1994 ) 6343 ; j. garriga and a. vilenkin , phys .
d * 57 * ( 1998 ) 2230 [ arxiv : astro - ph/9707292 ] .
s. w. hawking and i. g. moss , nucl .
b * 224 * , 180 ( 1983 ) .
j. d. brown and c. teitelboim , nucl .
b * 297 * ( 1988 ) 787 .
l. susskind , the anthropic landscape of string theory " , hep - th/0302219 .
j. l. feng , j. march - russell , s. sethi and f. wilczek , nucl .
phys . * b602 * , 307 ( 2001 ) . | decay of a de sitter vacuum may proceed through a static " instanton , representing pair creation of critical bubbles separated by a distance comparable to the hubble radius a process somewhat analogous to thermal activation in flat space .
we compare this with related processes recently discussed in the literature . |
in the thirteen years that have elapsed since the term `` lyman break galaxy '' ( or lbg ) was coined , samples of galaxies at @xmath124 have increased a thousand - fold .
large surveys , selecting galaxies not only via the break at 912(due mostly to intergalactic hi absorption ) , but also via ly@xmath10 emission , a variety of colour criteria , and sub - mm emission , have given us a broad view of the galaxy population during the epoch when star - formation activity was at its peak in the history of our universe ( see pettini et al .
2008 for a brief review ) . however , typical galaxies at @xmath14 are faint , with fiducial @xmath15-band magnitude @xmath16 ( steidel et al .
1999 ; reddy et al .
2008 ) , so that , even with the most efficient optical and infrared ( ir ) telescopes and spectrographs currently available , it is only possible to record their spectra with limited resolution and signal - to - noise ratio .
until the next generation of 30+m optical / ir telescopes comes into operation , the only strategy at our disposal for studying _ in detail _ the internal properties of normal galaxies at these redshifts is to take advantage of fortuitous alignments with foreground mass concentrations ( massive galaxies or galaxy clusters ) which can provide an order - of - magnitude boost of the flux reaching the earth through gravitational lensing . for several years
, the archetypal object for this kind of study has been the lyman break galaxy ms1512-cb58 ( or cb58 for short ; yee et al .
1996 ) . the magnification by a factor of @xmath17 ( seitz et al .
1998 ) provided by an intervening galaxy cluster at @xmath18 brings this @xmath19 , @xmath20 , galaxy within reach of high resolution spectroscopy from the ground and has made it the target of extensive observations at wavelengths from the visible to the mm range ( pettini et al .
2000 , 2002 ; savaglio , panagia , & padovani 2002 ; teplitz et al .
2000 ; siana et al .
2008 ; baker et al .
2001 , 2004 ; sawicki 2001 ) .
these studies provided unprecedented clear views of the interstellar medium , young stars , star - formation history , dust , metallicity , kinematics , and many other physical properties of a ` typical ' lbg at @xmath21 . but how typical is cb58 ?
we know from the work of shapley et al .
( 2003 ) that its ultraviolet ( uv ) continuum is more reddened , and its interstellar absorption lines are stronger , than in an ` average ' lbg .
furthermore , its young age of only 1020myr ( ellingson et al . 1996 ; siana et al .
2008 ) is at the lower end of the wide range of values deduced by fitting the spectral energy distributions of galaxies at @xmath123 ( papovich , dickinson , & ferguson 2001 ; shapley et al .
2001 ; erb et al .
2006b ) , although it is likely that an older stellar population is present too .
clearly , it is important to study in detail several other similarly bright sources and establish the _ range _ of properties of galaxies at a given cosmic epoch , rather than naively assuming that any particular galaxy is representative of a whole population .
fortunately , the large area of sky surveyed by the _ sloan digital sky survey ( sdss ) _ has led to the recent identification of many other examples of highly magnified galaxies ( see , for instance , belokurov et al .
2009 ; kubo et al .
2009 , and references therein ) . these and other discoveries have spurred a number of follow - up studies ( e.g. smail et al .
2007 ; stark et al . 2008 ; finkelstein et al . 2009 ; hainline et al . 2009 ; yuan & kewley 2009 ) . with these new samples of objects it will be possible to make progress on a variety of outstanding questions which can not be satisfactorily addressed with lower resolution data . in particular : \(i )
to what extent do different metallicity indicators , based on stellar photospheric lines , wind lines from the most massive stars , interstellar absorption lines , and emission lines from hii regions , give consistent answers ? this is an important issue , not only to explore the degree of chemical homogeneity of galaxies undergoing rapid star - formation , but also to clarify the sources of the systematic offsets between different hii region metallicity calibrators , and possibly bring them into better internal agreement ( see , for example , the discussions by pettini 2006 , and kewley & ellison 2008 ) .
\(ii ) what do the profiles of the interstellar absorption lines tell us about large - scale outflows in actively star - forming galaxies and about the inflow of gas fueling star formation ?
can we see any evidence of the leakage of hydrogen ionizing photons from the sites of star formation into the intergalactic medium ( igm ) which has proved so difficult to detect directly ( e.g. shapley et al .
2006 ; iwata et al .
2009 ) and yet seems to be required by a number of indirect lines of evidence ( e.g. faucher - gigure et al .
2008 ) ?
\(iii ) can we place limits on possible variations of the stellar initial mass function ( imf ) at @xmath123 ?
claims to this effect have certainly been put forward ( e.g. wilkins et al .
2008 and references therein ) , but the evidence is still controversial ( e.g. reddy & steidel 2009 ) .
\(iv ) what are the factors affecting the wide variety of spectral morphologies of the ly@xmath10 line , from strong narrow emission to damped absorption ( e.g. mas - hesse et al .
2003 ; verhamme , schaerer , & maselli 2006 ) , and what can we learn about the relationship of the so called ` ly@xmath10 emitters ' to the more general lbg population ? in this paper
we address these and other related questions with new observations of the rest - frame uv spectrum of a recently identified highly magnified lbg , named ` the cosmic horseshoe ' by its discoverers ( belokurov et al .
the paper is organised as follows . in section 2 ,
we summarize the known properties of the cosmic horseshoe and in section 3 we provide details of our observations and data reduction . in section 4
, we consider the information provided by the stellar uv spectrum of this galaxy , concerning in particular its systemic redshift , massive star imf , and metallicity of the ob stars .
section 5 deals with the resolved profiles of the interstellar absorption lines , while sections 6 and 7 focus on the uv nebular emission lines covered by our spectrum , ly@xmath10 and ciii]@xmath22 respectively . in section 8 ,
we bring these different strands together in the light of the questions raised above . finally , we summarize our main conclusions in section 9 . throughout the paper ,
we adopt a cosmology with @xmath23 , @xmath24 , and @xmath25 km s@xmath3 mpc@xmath3 .
-band image of the cosmic horseshoe obtained with fors2 on the very large telescope of the european southern observatory ( courtesy of l. j. king ) .
north is up and east is to the left .
superposed on the image is the @xmath26arcsec entrance slit of esi used for the observations reported here . ]
the cosmic horseshoe , or j1148 + 1930 , is a nearly complete blue einstein ring of 10arcsec diameter encircling a red galaxy ( figure [ fig : horseimage ] ; see also http://www.ast.cam.ac.uk/research/cassowary/ ) .
low resolution spectroscopy by belokurov et al .
( 2007 ) revealed the lens to be a massive luminous red galaxy ( @xmath27 ) at @xmath28 and the ring to be the gravitationally lensed image of a star - forming galaxy at @xmath29 .
from the extent of the ring , belokurov et al . estimated an approximate magnification factor of @xmath30 , while the detailed lensing models constructed by dye et al .
( 2008 ) give overall magnification factors of @xmath31 . adopting the latter value , and interpolating between the total integrated magnitudes of the ring @xmath32 , @xmath33 reported by belokurov et al .
( 2007 ) , we deduce a luminosity @xmath34 for the source , relative to @xmath35 from the luminosity function of uv - selected @xmath36 galaxies by reddy et al .
( 2008 ) . the image reconstruction in the source plane with the best fitting lensing model by dye et al .
( 2008 ) shows that the major contributor to this luminosity is a compact source with half - light radius @xmath37arcsec , or @xmath38kpc at @xmath29 ; there is also evidence for a second , fainter , source @xmath39arcsec to the north .
we briefly review the information currently available on the lensed galaxy .
its rest - frame uv spectrum is essentially flat in @xmath40 , reflecting a recent star - formation episode and modest reddening [ @xmath410.2 ; belokurov et al .
2007 ] , but a more extensive characterization of its stellar population(s ) awaits longer wavelength photometry to build - up the spectral energy distribution from the uv to the near - ir ( e.g. shapley et al . 2005 ; erb et al . 2006b ) .
recently , hainline et al . (
2009 ) reported observations of several rest - frame optical emission lines formed in the hii regions of the cosmic horseshoe , from which they were able to measure a number of parameters of interest .
the h@xmath10 luminosity implies a star formation rate of @xmath42 yr@xmath3 , after correcting for reddening and a lensing magnification factor of 24 .
the emission line widths indicate a velocity dispersion @xmath43 km s@xmath3 and a corresponding virial mass @xmath44 .
the oxygen abundance deduced from the ratios of the strongest emission lines is in the range ( o / h)@xmath451.5(o / h)@xmath46 .
these parameters are typical of uv - bright galaxies at @xmath47 ( erb et al . 2006a , b , c ) , although the star formation rate is at the upper end of the distribution of values found by erb et al .
( 2006c ) .
it is of interest now to compare such estimates with those provided by the rest - frame uv spectrum .
we observed the cosmic horseshoe with the echellette spectrograph and imager ( esi ; sheinis et al .
2000 ) on the keckii telescope . the high efficiency , moderately high resolution ( @xmath48 with a 1arcsec slit , sampled with 11.4 km s@xmath3 pixels ) , and wide spectral coverage of esi from 4000 to 10000 , or 11842959 in the rest - frame of the cosmic horseshoe
make it highly suitable for an in - depth study of the uv spectrum of this galaxy .
the data were collected over two nights ( 2008 march 6 and 7 ut ) in mostly sub - arcsecond seeing ; the total exposure time was 36100s , made up of a number of 1800s or 2000s - long individual integrations .
the esi slit was oriented at sky p.a .
= 10.8 degrees east of north , and positioned so as to encompass the two brightest knots in the einstein ring , as shown in figure [ fig : horseimage ] . the northern knot , labelled ` aperture 1 ' in the figure , is knot ` a ' in the labelling of the features by belokurov et al .
( 2007 ) , while ` aperture 2 ' corresponds to their knot ` d ' .
both knots are due to the same object in the source plane , although ` aperture 1 ' may include a small contribution from the fainter component @xmath39arcsec to the north of the main source of the ring ( dye et al .
the location of the esi slit is the same as that of the nirspec observations by hainline et al .
( 2009 ) .
the data were processed with standard iraf tasks . each individual two - dimensional ( 2-d ) esi spectrum was geometrically corrected ( to correct for the spectral curvature of each echellette order ) , bias subtracted , flat - fielded , and corrected for cosmetic defects and cosmic - ray affected pixels . in the next step , all of the images thus processed were co - added into a total 2-d frame which was then background subtracted .
the background subtraction process included careful modelling of the distribution along the slit of light from the lens , which is clearly visible above the sky at the longer wavelengths of the esi range .
finally , two 1-d spectra were extracted ( using optimal extraction algorithms ) , one for each ` aperture ' .
the extraction process included merging the 10 echellette orders of esi into one continuous spectrum , with weighted addition of portions of adjacent orders which overlap in wavelength .
we compared carefully the 1-d spectra of aperture 1 and aperture 2 to look for differences which one may expect to be revealed by the gravitational magnification of any spatial structure in the source , but found none .
neither dividing one spectrum by the other nor subtracting one from the other revealed features more significant than the noise . as discussed in section [ sec : lya ] , this is also the case for the ly@xmath10 emission line which ( a ) has the highest signal - to - noise ratio ( s / n ) , and ( b ) is most sensitive to geometrical effects , being resonantly scattered .
the only difference between the two apertures is in the total flux : emission line and continuum are uniformly higher by a factor of 1.19 in aperture 2 than in aperture 1 .
thus , in order to improve the s / n , the two spectra were averaged and the resulting spectrum mapped onto 0.5 bins .
this final spectrum has a resolution ( full width at half maximum ) fwhm=75 km s@xmath3 , sampled with three wavelength bins at 6000 , as determined from the widths of narrow emission lines from the cu - ar and hg - ne - xe hollow - cathode lamps used for wavelength calibration , whose spectrum was processed in the same way as that of the horseshoe . from the rms deviation of the data from the mean
, we measured an average s / n@xmath49 per 0.5 wavelength bin between @xmath50 and @xmath51(@xmath52 and @xmath53 in the rest frame ) , with a factor of @xmath54% variation in s / n across each echellette order . at the wavelengths of the ly@xmath10 emission line
the s / n rises to a maximum s / n=44 per 0.5 bin .
we attempted to put our spectrum on an absolute flux scale by reference to those of three flux - standard stars recorded during the two nights of observation , but found a scatter of @xmath55% in the flux calibration among the three stars .
adopting the mean value of the three flux scales , we deduced a mean @xmath56erg s@xmath3 @xmath57 hz@xmath3 between 42005485 .
this wavelength interval corresponds to the fwhm of the transmission curve of the gunn @xmath58 filter through which belokurov et al .
( 2007 ) measured @xmath32 ( corresponding to @xmath59erg s@xmath3 @xmath57 hz@xmath3 ) for the entire einstein ring .
thus , if the flux calibration is correct , the esi slit captured a fraction @xmath60 of the light of the cosmic horseshoe , corresponding to an effective magnitude within the two apertures of @xmath61 .
the rest - frame uv spectra of star - forming galaxies are complex blends of interstellar absorption lines , nebular emission lines , and absorption / emission lines formed in the atmospheres of ob stars .
this wealth of spectral features , properly interpreted , is a rich source of information on the physical properties of the gas and stars in these galaxies . in this section ,
we consider photospheric absorption lines which have been shown to be useful abundance diagnostics , and p - cygni profiles formed in the expanding atmospheres of the most luminous early - type stars whose strength is also sensitive to the upper end of the initial mass function .
the first step in the analysis is the determination of the systemic redshift of the galaxy . while most photospheric features are unresolved blends of multiple lines , a few exceptions allow the redshift of the hot stars to be determined . within the wavelength range covered by our data , we identified the following unblended photospheric absorption lines : sv@xmath62 , niv@xmath63 , and ciii@xmath64 [ @xmath65all three vacuum wavelengths from the national institute of standards and technology ( nist ) database available at http://physics.nist.gov/physrefdata/asd/ ] .
the redshifts of these three lines are in good mutual agreement ( see table [ table : systemic_z ] ) , and define an average @xmath66 .
the @xmath67 error corresponds to an uncertainty of @xmath68 km s@xmath3 , or 1/5 of a wavelength bin in our final spectrum .
also included in table [ table : systemic_z ] are values of @xmath69 from the ciii]@xmath70 doublet lines which are resolved in our esi spectrum ( see section [ sec : ciii ] ) , and from h@xmath10 measured from the nirspec spectrum of hainline et al .
the mean @xmath71 } = 2.38115 \pm 0.00012 $ ] is in excellent agreement with @xmath72 ; @xmath73 differs by only 7 km s@xmath3 which may include a small systematic offset between the wavelength calibrations of the esi and nirspec spectra .
thus , we conclude that the systemic redshift of the cosmic horseshoe is @xmath74 and we adopt this value throughout the paper . in this and the following subsection
, we analyse photospheric and wind lines in the spectrum of the cosmic horseshoe by comparing our data with model spectra computed with the population synthesis code _ starburst99 _ which couples libraries of either empirical ( leitherer et al . 1999 , 2001 ) or theoretical ( rix et al . 2004 )
ultraviolet ob stellar spectra with stellar evolutionary tracks .
the input set of parameters to _ starburst99 _ can be adjusted to simulate a variety of star formation histories , ages , imf parameters , and metallicities so as to determine the combination that best fits the observations under scrutiny . in the absence of more detailed information on the previous history of star formation in the cosmic horseshoe
, we shall adopt the simplest case of continuous star formation with a salpeter slope of the upper end of the imf which has been proceeding at a steady rate ( as opposed to a declining rate , or a series of bursts ) for 100myr ( this being a safe lower limit beyond which the spectrum no longer changes with time because the stars that contribute to the uv light are born and die at the same rate ) .
while this scenario is undoubtedly an over - simplification , it may not be too different from reality when considering the spectrum of a whole galaxy , rather than individual regions of star formation .
an age of more than 100myr is typical of most @xmath12 uv - bright galaxies in the ` bx ' sample of erb et al .
( 2006b ) and seems plausible given the near - solar metallicity of the hii regions of the horseshoe ( hainline et al .
2009 ) .
leitherer et al .
( 2001 ) and rix et al .
( 2004 ) pointed out the existence of some blends of uv photospheric lines whose strengths , under the above assumptions , depend on metallicity ( see also halliday et al .
the `` 1425 '' index of rix et al .
( 2004 ) measures the equivalent width of a blend of siiii@xmath75 , ciii@xmath76 , and fev@xmath77 absorption lines spanning the wavelength interval 14151435 ; at longer wavelengths , the `` 1978 '' index measures the strengths of several feiii absorption lines from b stars between 1935 and 2020 .
both indices stabilise after @xmath54myr from the onset of star formation , and increase monotonically with metallicity . in the left - hand portion of figure [ fig : rixindices ]
, we compare the esi spectrum of the cosmic horseshoe with the model spectra of rix et al .
( 2004 ) for five values of metallicity , from 1/20 of solar to twice solar , in the regions of the 1425 and 1978 indices . for this comparison
, we smoothed our spectrum to the 2.5 resolution of the rix et al .
models , and normalized it using the psuedo - continuum windows suggested by those authors .
thus , the observed and model spectra should be directly comparable .
figure [ fig : rixindices ] shows a generally good agreement in the detailed features of the synthetic and real spectra in the two wavelength regions considered here , attesting to the sophistication reached by current hot - star atmosphere and stellar population synthesis models . referring to the 1425 region ( left - most panel in figure [ fig : rixindices ] )
, it can be seen that the strength of the photospheric lines in this region of the cosmic horseshoe spectrum is intermediate between those computed with metallicities @xmath78 and @xmath79 respectively .
using the relationship between metallicity and the 1425 index proposed by rix et al .
( 2004 ) : @xmath80 with @xmath81 and @xmath82 appropriate to the value @xmath83 we measure here , we derive @xmath84 . qualitatively , the observed
spectrum seems to be intermediate between the 0.4 solar and solar metallicity cases in the 1978 region too ( see middle panel of figure [ fig : rixindices ] ) .
however , there appears to be excess absorption between 1960 and 1980over that expected from the models for _ any _ metallicity ; the presence of these additional features does not allow us to deduce a value of @xmath85 from ew(1978 ) using the rix et al .
( 2004 ) calibration of this index .
we considered the possibilities that the excess absorption may be noise , or an intervening absorption system unrelated to the cosmic horseshoe , but could find no evidence to support either interpretation .
it also seems unlikely that the excess is due to a different mix of stellar spectral types or chemical elements , since all of the photospheric absorption in this region is thought to be due to feiii lines , and the rest of the spectral features between 1920 and 2025match the data well .
no such discrepancies were found by rix et al .
( 2004 ) in the two galaxies they considered ( ms1512-cb58 and q1307-bm1163 , the latter at @xmath86 ) , nor by halliday et al .
( 2008 ) in their composite spectrum of @xmath47 galaxies . however , the cases where these photospheric indices have been measured are still very few , and it is important to continue to look critically at the match between model and real spectra as more data of suitable quality become available .
the model spectra by rix et al .
( 2004 ) show that the spectral features most sensitive to metallicity are the p - cygni lines formed in the expanding winds of the most luminous ob stars whose mass - loss rates are thought to be lower at lower metallicities ( e.g. kudritzki & puls 2000 ) . among the p - cygni lines covered by our spectrum ,
the civ@xmath87 ( unresolved ) doublet is the strongest . referring to the rightmost panel in figure[fig : rixindices ] , we find that its strength in the cosmic horseshoe is intermediate between those of the models with metallicities @xmath88 and @xmath78 . however , as emphasised by crowther et al .
( 2006 ) , the interpretation of this feature is complicated by its blending with interstellar civ ( and to a lesser extent siii@xmath89 ) absorption , at least at the coarse resolution of most spectra of distant galaxies .
similar problems affect the generally weaker siiv@xmath90 and nv@xmath91 wind lines . at the high resolution of our esi spectrum , interstellar and stellar components
are easily separated , as can be appreciated from figure [ fig : civ ] .
the top panel compares the observed spectrum of the cosmic horseshoe in the civ region with those generated by _
starburst99 _ now using libraries of _ empirical _ stellar spectra , assembled from uv observations of stars in either the galaxy or the magellanic clouds . both models are for 100myr old continuous star formation with salpeter imf . apart from the narrow interstellar siii@xmath89 and civ@xmath92 absorption lines , which are stronger in the distant galaxy than in the spectra of the stars that make up the _ starburst99 _ libraries , the agreement between models and observations is clearly very good .
the broad blue absorption wing of the p - cygni profile reaches similar terminal velocities in the models and the data , and its depth is roughly intermediate between those of the galactic and magellanic clouds models .
the latter library of stellar spectra was built - up by leitherer et al .
( 2001 ) with _ hubble space telescope _ ( _ hst _ ) observations of stars in _ both _ the large and small magellanic clouds ( lmc / smc ) and may thus correspond to an approximate metallicity @xmath93 , given that in the lmc ( o / h)@xmath94(o / h)@xmath46 and in the smc ( o / h)@xmath95(o / h)@xmath46 ( e.g pagel 2003 ) . by
linearly interpolating between the galactic and lmc models and minimising the difference from the observed spectrum , we find @xmath96 , in good agreement with @xmath97 deduced above from consideration of the 1425 photospheric index .
it is interesting that both galactic and magellanic clouds models seem to _ underpredict _ the emission component of the p - cygni profile .
however , the narrow emission feature centred near 1551.6 may be partly of nebular origin ( leitherer , calzetti , & martins 2002 ) .
similarly , the narrow emission centred near 1533.7is probably nebular siii@xmath98 ; both features are missing from the synthetic spectra which were designed to reproduce only the stellar component of the galaxy spectrum .
the superposition of p - cygni broad emission / absorption , photospheric broad absorption , narrow interstellar absorption and narrow nebular emission attests to the complex nature of this portion of the uv spectrum of star - forming galaxies . in the lower panel of figure
[ fig : civ ] we compare the esi spectra of ms1512-cb58 ( from pettini et al .
2002 ) and the cosmic horseshoe in the wavelength region 15001570 .
there is a startling similarity in this spectral region between these two galaxies which are at different redshifts ( cb58 is at @xmath99 ) and were selected randomly , only by virtue of the fact that they are highly lensed as seen from earth ( although they are both more luminous than the ` average ' galaxy at these redshifts , even after correcting for the lensing magnifications ) . presumably , not only the chemical abundances , but also the young stellar populations of these two galaxies are remarkably similar .
only the interstellar absorption lines are different , with those in cb58 apparently stronger than in the cosmic horseshoe ( but see the discussion in section [ sec : ism ] ) .
the p - cygni component of the civ complex is due to the most massive ( and luminous ) o stars , which drive the strongest winds . on the other hand , a wider range of stellar spectral types , including all stars with masses greater than about @xmath100 ,
are thought to make up the integrated continuum light near 1550(e.g .
rix et al .
thus , the contrast of the p - cygni features relative to the continuum is sensitive not only to metallicity , but also to the slope and upper end cut - off of the imf , as well as to more subtle effects , such as differential dust extinction among stars at the upper end of the imf ( the most massive and short - lived stars may in principle be more reddened than longer lived ones
leitherer et al .
2002 ) , and the relative proportions of ob stars in star clusters and in the field ( chandar et al .
pettini et al .
( 2000 ) showed how even relatively minor changes to the imf slope ( or alternatively the upper mass cut - off ) result in noticeable alterations to the integrated spectrum of a star - forming galaxy in the civ region . on that basis
, those authors concluded that a standard salpeter imf extending to @xmath101 provides the best match to the spectrum of cb58 , without the need to invoke a different imf at high redshift , as claimed by some .
clearly , the same conclusion applies here to the massive stellar population in the cosmic horseshoe .
more generally , the similarity evident in the lower panel of figure [ fig : civ ] suggests that the minor effects discussed above ( differential dust extinction and the balance between cluster and field stars ) presumably average out in at least a subset of star - forming galaxies at @xmath102 , when their integrated spectra are considered . summarising the conclusions of this section , we find that the metallicity of the early - type stars in the cosmic horseshoe is approximately half - solar , from consideration of photospheric and wind absorption lines .
a continuous mode of star formation with a salpeter slope for stars more massive than @xmath103 gives a good representation of the uv spectrum .
the 1978 index introduced by rix et al .
( 2004 ) appears to be contaminated by additional features of uncertain origin in the spectrum of the horseshoe .
the civ region is a complex blend of stellar , interstellar and nebular features , but the stellar component in the cosmic horseshoe is surprisingly similar to that in cb58 , the only other high redshift star - forming galaxy studied in comparable detail so far .
this may of course be a coincidence , but it may also point to the fact that many of the parameters which can affect the appearance of this spectral region average out to a large extent in the integrated uv spectrum of a whole galaxy .
in the wavelength interval from 1200 to 2600 , we identified 21 interstellar absorption lines from six chemical elements in a variety of ionization stages , from oi to civ .
relevant measurements are collected in table [ tab : ism ] , while figure [ fig : ismvelocity ] shows selected transitions chosen to illustrate the range of elements and ion stages covered .
the interstellar lines in the cosmic horseshoe are broad , with absorption extending smoothly from @xmath8 to @xmath104 km s@xmath3 with respect to @xmath105 . within the limits of the noise
, we find no convincing differences between the profiles of different ions , although the maximum velocities measured depend on the strength of the transition , as is the case in local starbursts ( e.g. grimes et al .
the values of @xmath106 listed in column ( 4 ) of table [ tab : ism ] are those derived from the centroids of the lines ( that is , the mean wavelength of the line weighted by the absorption in each wavelength bin ) ; their mean value is @xmath107 which corresponds to a velocity offset @xmath108 km s@xmath3 with respect to the stars .
we also measured the redshift at the peak optical depth in the most clearly defined line profiles and found a mean @xmath109 ( @xmath110 km s@xmath3 ) .
this net blueshift of the interstellar lines is a common feature of star - forming galaxies at low ( e.g. heckman et al . 2000 ; martin 2005 ) as well as high ( e.g. pettini et al . 2001 ; shapley et al .
2003 ; steidel et al . 2009 ; vanzella et al . 2009 ) redshifts .
it is generally interpreted as empirical evidence for the existence of large - scale outflows of the interstellar medium driven by the kinetic energy deposited by supernovae and the winds of massive stars .
the values of @xmath111 we deduce here , irrespective of whether one considers the mean redshift of the line centroids or of the gas with the highest optical depth , are typical of @xmath112 galaxies for which shapley et al .
( 2003 ) derive a mean @xmath113 km s@xmath3 . similarly , the widths of the strongest lines ( @xmath114600 km s@xmath3 ) are within the range @xmath115 km s@xmath3 measured by shapley et al .
( 2003 ) from their composite spectrum of 811 lbgs . [ cols="<,<,<,<,^,^,<",options="header " , ] @xmath116 abundance relative to solar ( on a linear scale ) , using the + compilation of solar abundances by asplund et al .
+ @xmath117 as reported by hainline et al .
+ @xmath118 this work ( section [ section : r04indices ] ) .
+ @xmath119 this work ( section [ section : stellarwind ] ) .
+ it is still unclear what determines @xmath120 , the fraction of hydrogen ionizing photons which escape from star - forming galaxies into the intergalactic medium . direct detection of these lyman continuum photons has proved problematic until recently ( e.g. shapley et al .
2006 ; iwata et al .
2009 ) , and yet @xmath120 must have been large at very high redshifts for the universe to be reionized by the star - formation activity thought to have taken place at @xmath121 ( e.g. bolton & haehnelt 2007 ; ryan - weber et al .
2009 ) .
observations of gravitationally lensed galaxies may offer insights into the factors that control @xmath120 . by fully resolving the interstellar absorption lines in our esi spectrum of the cosmic horseshoe
, we reached the conclusion that the interstellar gas only covers @xmath9% of the stellar uv light , as viewed from earth .
this is a promising prerequisite for large values of @xmath120 , making the horseshoe a high priority candidate in searches for lyc emission from high-@xmath13 galaxies . on the other hand ,
if @xmath122% of the photons from the stars and surrounding hii regions really had a clear path out of the galaxy , we may have expected to see a strong and narrow ly@xmath10 emission line centred at @xmath69 , whereas no such feature is present in our spectrum .
possibly , some 40% of the stars are located behind neutral gas of too low a column density to give discernible absorption in the metal lines , and yet capable of scattering most of the ly@xmath10 photons out of the line of sight ( hansen & oh 2006 ) . whether such gas would be optically thick to lyc photons remains to be established .
another necessary condition for the escape of lyc photons is a weaker h@xmath10 emission line than expected on the basis of the uv continuum luminosity and reddening ( admittedly in the idealised case of continuous star formation at a constant rate ) .
such a disparity would arise from a ` matter - bounded nebula ' , where not all of the lyc photons emitted by the stars are absorbed and reprocessed within the hii region . as discussed in section [ sec : f_esc ] , in the cosmic horseshoe sfr(h@xmath123sfr(uv ) , although a number of different factors , apart from leakage of lyc photons , can affect this comparison . in conclusion
, it would definitely be worthwhile to search for lyc emission in the cosmic horseshoe at wavelengths below 3085 ( the redshifted value of the lyman limit at @xmath0 ) .
furthermore , it would be of interest to check for partial covering of the stars by the foreground interstellar medium in more galaxies among the newly discovered strongly lensed sources , ideally including galaxies with a range of ly@xmath10 equivalent widths . with such data in hand , we should be in a better position to understand the conditions that determine the escape fraction of lyc photons and the morphology of the ly@xmath10 line .
one of the motivations for the present study was to establish how typical are the properties of the galaxy ms1512-cb58 , the only previous case where the gravitational lensing boost was sufficient to allow a detailed look at the spectrum of a high - redshift star - forming galaxy .
while we have now doubled the ` sample ' of high-@xmath13 galaxies with good - quality esi spectra , it would clearly be premature to draw general conclusions on the basis of just two objects .
nevertheless , one can not help but being struck by how closely these two galaxies resemble each other in many of their properties .
they are very similar in their overall metallicity and probably in their detailed chemical composition , indicating that they have reached comparable stages in the conversion of their gas reservoirs into stars .
their young stellar populations are largely indistinguishable in their uv spectral features , especially the p - cygni lines which are sensitive indicators of the mode of massive star formation and of differential reddening among stars at the upper end of the imf .
most remarkable perhaps is the unexpected similarity in the properties of their outflowing interstellar media , with velocities spanning in both cases @xmath124 km s@xmath3 , from @xmath8 to @xmath104 km s@xmath3 relative to the stars , and involving comparable column densities of gas .
presumably , it is just a coincidence that the first two galaxies at @xmath123 whose uv spectra have been put under the ` microscope ' of gravitational lensing have also turned out to be similar in so many respects .
after all , we now know that the cosmic horseshoe and ms1512-cb58 are also forming stars at comparable rates of @xmath54@xmath125 yr@xmath3 , and have similar dynamical masses of @xmath126 ( hainline et al .
2009 for these measurements in the cosmic horseshoe and teplitz et al .
2000 for cb58 ) .
future studies of other gravitationally lensed objects will arguably show a wider range of properties of star - forming galaxies at these redshifts .
what is interesting then , given the underlying similarity of the horseshoe and cb58 , is to consider the aspects in which they _ differ _ : the incomplete coverage of the stellar light by the interstellar absorption lines and the strikingly different ly@xmath10 morphologies .
it is these two factors alone that would lead an observer to ` classify ' the two galaxies differently .
for example , in the rough breakdown of lbgs by shapley et al .
( 2003 ) according to their uv spectral properties , cb58 would fall in the quartile with the highest interstellar absorption line equivalent widths and reddening , while the cosmic horseshoe would be in the quartile with the second highest ly@xmath10 _ emission _ equivalent width . and
yet , from what we have uncovered in the present study , it may be the case that these differences are perhaps simply due to orientation effects viewing two intrinsically similar galaxies along differing sightlines through their interstellar media , one ( the horseshoe ) more rarefied and highly ionized than the other ( cb58 ) . with future studies of other strongly lensed galaxies
we shall hopefully learn how to distinguish intrinsic differences from those caused by second - order effects , of which the direction from which we view the central starburst may be one example . for the moment , however , we must remain cautious of overinterpreting differences between different ` classes ' of high - redshift galaxies , especially if such distinctions are based on the morphology and strength of the ly@xmath10 line , whose appearance is so sensitive to a variety of different parameters ( e.g. pentericci et al .
we have presented esi observations of the rest - frame ultraviolet spectrum of a star - forming galaxy at @xmath0 , lensed by a massive foreground galaxy into a nearly complete einstein ring dubbed the ` cosmic horseshoe ' .
the high gravitational magnification affords the rare opportunity of recording the spectrum of this member of the @xmath123 galaxy population at higher resolution and s / n than otherwise achievable , and thereby provides us with a close - up look of a galaxy during the epoch when cosmic star - formation activity was at its peak .
our main findings are as follows .
\(i ) the cosmic horseshoe shares many of its properties with the population of ` bx ' galaxies selected with the colour criteria of steidel et al .
its metallicity @xmath127 , dynamical mass @xmath4 , and reddening @xmath5 are all typical of those galaxies ( erb et al . 2006a , b , c ) .
with a star - formation rate sfr@xmath128 yr@xmath3 , the horseshoe is among the most luminous galaxies in the bx sample .
\(ii ) generally , there is good agreement between different metallicity indicators based , respectively , on ratios of strong nebular emission lines , on blends of stellar photospheric lines , and on p - cygni lines from the most luminous ob stars .
the @xmath129 index of pagel et al .
( 1979 ) seems to overpredict the oxygen abundance by a factor of @xmath130 , when calibrated with the prescription by tremonti et al .
( 2004 ) . of the stellar measures , the 1978 index of rix et al . (
2004 ) does not fit the data well due to excess absorption not present in the models and whose identity remains unclear .
\(iii ) a continuous mode of star formation with a salpeter slope for stars with masses @xmath131 gives a good representation of the uv spectrum ; we see no evidence for a departure from a salpeter imf , nor differential reddening among massive stars , in the contrast of the p - cygni lines over the underlying uv continuum .
\(iv ) the interstellar absorption lines are broad , extending over a velocity interval of @xmath124 km s@xmath3 , from @xmath8 to @xmath104 km s@xmath3 relative to the redshift of the stars and hii regions .
the gas with the highest optical depths is outflowing from the regions of star formation with a speed of @xmath132 km s@xmath3 , and only covers @xmath9% of the stellar continuum light .
\(v ) the ly@xmath10 line is in emission and shares many of the properties of the so - called ly@xmath10 emitters , although its equivalent width is a factor of @xmath133 smaller than the lower limit @xmath134commonly adopted to define this subset of galaxies .
the resolved line profile matches well those computed with models of resonantly scattered ly@xmath10 photons in an expanding medium , although such models find it difficult to account for the fact that the velocity dispersion of the gas is comparable to its bulk expansion velocity , and for the existence of gas extending over hundreds of km s@xmath3 in velocity space .
some @xmath1115% of the ly@xmath10 photons produced escape the galaxy , the remainder presumably being absorbed by dust and recycled into infrared photons , or scattered over a larger area than that covered by our narrow - slit observations .
it remains to be established what fraction of the lyman _
photons emitted by the stars leak into the igm .
\(vi ) overall , many of the physical properties of the stars and interstellar medium of the cosmic horseshoe are similar to those of the only other galaxy previously studied at comparable resolution and s / n , ms1512-cb58 at @xmath20 .
this may not be surprising , given that both galaxies are fairly typical examples of the star - forming galaxy population at these redshifts .
the fundamental similarities between the two objects highlight the fact that the aspects in which they differ , particularly the very different morphologies of their ly@xmath10lines , may be due to superficial reasons , such as differing viewing angles , rather than to more deeply rooted causes . in the near future it should be possible to make further progress on all of these topics with the increasing numbers of strongly lensed high - redshift galaxies being discovered by the _ sdss _ and other lensing surveys .
we are grateful to the staff at the w. m. keck observatory for their competent assistance with the observations , to sam rix and dan nestor who generously provided some of the software used in the analysis of the spectra , and to simon dye for clarifications concerning the lensing models .
lindsay king kindly allowed us to reproduce her eso / fors2 image of the cosmic horseshoe in figure [ fig : horseimage ] .
we also thank the anonymous referee for helpful comments which have improved the paper .
amq s research is funded by a scholarship from the marshall foundation .
aes acknowledges support from the david and lucile packard foundation and the alfred p. sloan foundation , and ccs from nsf grant ast-0606912 and the john d. and catherine t. macarthur foundation .
finally , we wish to extend thanks to those of hawaiian ancestry on whose mountain we are privileged to be guests .
asplund , m. , grevesse , n , & sauval , a. j. 2005 , in barnes t. g. iii , & bash , f. n. eds .
, asp conf .
336 , cosmic abundances as records of stellar evolution and nucleosynthesis .
, san francisco , p. 25
pagel , b. e. j. 2003 , in charbonnel , c. , schaerer , d. , & meynet , g. , eds . , cno abundances in dwarf and spiral galaxies , astronomical society of the pacific conference series , san francisco , 304 , p. 187 | taking advantage of strong gravitational lensing , we have recorded the rest - frame ultraviolet spectrum of the @xmath0 galaxy ` the cosmic horseshoe ' ( j1148 + 1930 ) at higher resolution and signal - to - noise ratio than is currently feasible for unlensed galaxies at @xmath1 . with a star - formation rate of @xmath2 yr@xmath3 ,
dynamical mass @xmath4 , half - solar metallicity , and moderate reddening @xmath5 , the cosmic horseshoe is a good example of the population of galaxies responsible for most of the star - formation activity at these redshifts . from the analysis of stellar spectral features
we conclude that a continuous mode of star formation with a salpeter slope for stars in the mass range 5@xmath6 gives a good representation of the uv spectrum , ruling out significant departures from a ` standard ' imf .
generally , we find good agreement between the values of metallicity deduced from stellar and nebular tracers .
interstellar absorption is present over a velocity range @xmath7 km s@xmath3 , from @xmath8 to + 250 km s@xmath3 relative to the stars and their hii regions , but we still lack a model relating the kinematic structure of the gas to its location within the galaxy
. there is evidence to suggest that the outflowing interstellar gas is patchy , covering only @xmath9% of the uv stellar continuum .
the ly@xmath10 line shares many of the characteristics of the so - called ly@xmath10 emitters ; its double - peaked profile can be reproduced by models of ly@xmath10photons resonantly scattered by an expanding shell of gas and dust , with @xmath1115% of the photons escaping the galaxy .
many of the physical properties of the cosmic horseshoe are similar to those of the only other galaxy at @xmath123 studied in comparable detail up to now : ms1512-cb58 . the fact that these two galaxies have drastically different ly@xmath10lines
may be due simply to orientation effects , or differences in the covering factor of outflowing gas , and cautions against classifying high-@xmath13 galaxies only on the basis of spectral features , such as ly@xmath10 , whose appearance can be affected by a variety of different parameters .
[ firstpage ] cosmology : observations galaxies : evolution galaxies : starburst galaxies : individual ( cosmic horseshoe ) |
many of the high-@xmath1 cuprate materials show signatures of static or dynamic spin and charge modulation , or _ stripes
_ @xcite .
the effect of stripes on high - temperature superconductivity is a subject of ongoing debate : some researchers consider them central to the mechanism of superconductivity , others consider them detrimental , and yet others argue that stripes are irrelevant .
the situation is further complicated by the difficulty of detecting the presence of stripes , particularly when they are only weakly pinned to the lattice , or dynamical .
the former case is typically associated with zero - frequency incommensurate - momentum response in neutron scattering @xcite , while the latter is associated with finite - frequency incommensurate - momentum response @xcite .
the difficulty of direct detection has motivated other approaches to stripe detection based on their effects , e.g. , on the phonon modes @xcite , or lattice structure @xcite .
in particular , one naturally expects that in the presence of sharp stripes there will be a rearrangement of the collective mode spectrum , with the formation of local modes split off from the bulk continuum , or _ edge modes _ @xcite .
one of the most striking effects in the striped cuprates is the suppression of the superconductivity near the @xmath0 hole - doping , where ( based on neutron scattering ) the stripes become stationary : the so - called @xmath0-anomaly " @xcite . at @xmath0 doping ,
stripes are locked in with the lattice resulting in a commensurate modulation wavenumber .
this suppresses the charge and spin fluctuations , which are considered to be important for superconductivity .
away from @xmath0 doping , on the other hand , the favorable wavenumber for the stripe structure is expected to be incommensurate and proportional to the hole concentration .
naturally , the incommensurability renders the stripes more dynamical . while this anomaly strongly suggests that the commensurate ( c ) stripes suppress superconductivity , the question of whether incommensurate ( ic ) ( or fluctuating ) stripes promote the superconductivity remains .
we believe that study of this interplay is important for developing an understanding of the high - temperature superconductivity in cuprates . as a first step , in this paper
, we discuss the c - ic transition near the commensurate doping , calculating the transition point within a weak pinning approximation .
we also calculate the energy of two charge - spin - lattice excitations , namely , the _ phason _ and the _ amplitudon_. in the last part of the paper , we provide relations between ginzburg - landau parameters ( order parameters , energy coefficients ) and physical quantities ( conductivity , dielectric function ) by considering the response to an electric field .
we model the stripe phase by a simultaneous unidirectional sinusoidal charge density wave ( cdw ) and spin density wave ( sdw ) with wavelengths @xmath2 and @xmath3 , respectively . here
@xmath4 is the lattice constant , @xmath5 is the reciprocal - lattice vector , and @xmath6 is the ordering vector .
we introduce @xmath7 and @xmath8 as the complex order parameters for the cdw and the sdw , respectively .
the real part of @xmath9 gives the deviation of the charge density from the uniform state : @xmath10 where @xmath11 is the average hole - density per site : @xmath12 for a doping @xmath13 .
the real part of @xmath14 gives the staggered spin density : @xmath15 the periodicity of the cdw , @xmath16 , is characterized by @xmath17 if the cdw and lattice are commensurate , we can write @xmath18 with relatively prime integers @xmath19 and @xmath20 . we extend @xmath18 to irrational numbers for the ic case by taking @xmath21 and @xmath22 .
we define @xmath23 as a commensurability ; thus , the commensurability for the ic case is zero .
the order parameters have the following forms : @xmath24 $ } , \label{eq : order - ch}\\ & & \mbox{$\chi(x,\phi ) = - \i m \exp[\i(\frac{q}{2 } x + \frac{\phi}{2l})]$}. \label{eq : order - spn}\end{aligned}\ ] ] we investigate the free energy near the commensurate doping @xmath25 . here
we consider only the following two cases as the groundstate candidates .
one is the ic case , where the periodicity is given by @xmath26 ( @xmath27 ) as a function of the doping @xmath13 .
the other is the simple c case with @xmath28 ( finite @xmath19 ) independent of the doping @xmath13 .
we denote @xmath6 especially in the ic case by @xmath29 .
we introduce the ginzburg - landau functional ( per site ) : @xmath30.\end{aligned}\ ] ] in general , @xmath31 can be written in the following form for a one - dimensional cdw with commensurability @xmath23 @xcite : @xmath32 & = & f_0[\psi(x),\chi(x)]+ p(x ) \re{\psi(x)^l } \nonumber\\ & & + g_0\left|[q(z)+\i\nabla]\psi(x)\right|^2 .
\label{eq : f}\end{aligned}\ ] ] here , @xmath33 is concerned with the spin and charge ordering , but not with the pinning or deformation of the cdw . the second term in expression ( [ eq : f ] )
is introduced to represent the pinning energy of the cdw . the third term , which originates from the gradient term in the free energy , gives the preference for @xmath34 ( @xmath35 ) .
we introduce this term to describe the energy which arises from the deformation of the cdw ; however , to simplify the calculation , it does not include the local effects ( e.g. , those from discommensurations @xcite ) but only gives the preference among different wavevectors .
we assume that @xmath33 is of the form @xmath36 & = & r_0 \left|\chi(x)\right|^2 + u_0
\left|\chi(x)\right|^4\nonumber\\ & & -s_0\,\left|\chi(x)^2\psi(x)\right|+v_0 \left|\psi(x)\right|^2,\end{aligned}\ ] ] with the parameters chosen in such a way that the charge order is induced by the magnetic order @xcite .
expanding @xmath37 in harmonics , @xmath38 the free energy per site can be rewritten as @xmath39 ^ 2 \rho^2 , \label{eq : f_phi}\\ f^{(0)}(\rho , m)&= & r_0 m^2+u_0m^4-s_0m^2\rho + v_0\rho^2 , \label{eq : f0}\end{aligned}\ ] ] where @xmath40 for the equilibrium state , @xmath41 is taken so that @xmath42 ( i.e. , @xmath43 is always negative , and the sign of @xmath44 determines the location of the stripe center ) .
notice that the pinning term disappears for incommensurate cases , since @xmath45 for @xmath46 .
the @xmath47 and @xmath48 terms govern the continuous transition between the stripe and the uniform states at a large doping . since we are interested in the stripe state , we take @xmath49 , @xmath50 , @xmath51 and @xmath52 .
@xmath53 vanishes at the order - disorder transition point @xmath54 .
the region between the c - ic transition points @xmath55 around the c point @xmath56 is the c region ; outside this region the system is in the ic phase ( fig .
[ fig : transition ] ) .
we will find in the following that the width of the c region rapidly shrinks with increasing @xmath19 .
although the term @xmath57 can give large negative energy for large @xmath58 , this does not cause any problem so long as we restrict @xmath59 to not too large values , so that the @xmath60 term is not dominant . in other words ,
this term makes only a minor modification of the free energy , which selects between the c and ic equilibrium states .
we make the following _ assumptions _ in this paper : 1 .
@xmath61 ( weak pinning ) .
@xmath62 ( weak coupling ) .
@xmath47 is constant around the c region ( @xmath63 ) .
4 . @xmath64 .
5 . @xmath65 ( equivalent to @xmath66 ) .
_ assumption 4 _ is consistent with the neutron scattering experiments where it is roughly estimated that @xmath67 for striped cuprates @xcite . in _ assumption 5
_ , the inequality sign stands for to be on the order of or less than " , and the _ assumption _ reflects the fact that the charge order is much weaker than the magnetic one in metallic stripes due to the strong fluctuations .
now we evaluate the order parameters for the c and ic cases so that @xmath58 and @xmath68 minimize the free energy .
first , in the ic case , it follows from @xmath69 and @xmath70 that @xmath71 the following condition is required for @xmath72 and @xmath73 : @xmath74 which is satisfied by _ assumption _ _ 2_. substituting eqs .
( [ eq : mic ] ) and ( [ eq : rhoic ] ) into eq .
( [ eq : f0 ] ) , we obtain the condensation energy @xmath75 next , in the c case , the magnetic order parameter is @xmath76 the charge order parameter for the c case satisfies the equation @xmath77 where @xmath78 since @xmath79 and @xmath80 , the following condition is required for the existence of a positive root : @xmath81 under this condition , eq . ( [ eq : rho_c ] ) has two positive roots .
the smaller one corresponds to the minimal energy in the c case , while the other corresponds to the maximal one ; therefore @xmath82 is given by the smaller positive root of eq .
( [ eq : rho_c ] ) .
@xmath83 is a function of @xmath13 and @xmath19 . for later use
, we define @xmath84 and @xmath85 independent of @xmath13 and @xmath19 near the commensurate point : @xmath86 where @xmath87 note that , near @xmath54 , @xmath85 depends on @xmath13 through @xmath47 , but not so far below @xmath54 ( from _ assumption 3 _ ) . from now on , we also use the notation @xmath88 with a subscript to represent that corresponding to @xmath13 with the same subscript through eq .
( [ eq : z ] ) .
we discuss the c - ic transition as a function of the hole concentration @xmath13 .
we will evaluate the difference of the free energy between c and ic below .
however , @xmath89 is not well - defined at @xmath90 since there is no ic state at @xmath90 . to avoid problems with this discontinuity
, we first define @xmath91 as the order parameter and the ic free energy for any @xmath88 including @xmath90 , and then we take the commensurability at the @xmath90 point into account after comparing the continuous free - energy functions if needed .
the difference between the c and ic free energies is [ from eqs .
( [ eq : mic ] ) , ( [ eq : mc ] ) , and ( [ eq : rho_c ] ) ] @xmath92 where @xmath93 . the roots of @xmath94 for @xmath95 are @xmath96 here the smaller root gives the c - ic transition , because @xmath97 should be positive so that @xmath98 at the transition point .
we require @xmath99 so that the c - ic transition exists . for @xmath95 , @xmath100 has only one minimum at @xmath101 .
we find that @xmath82 is larger than @xmath102 at @xmath103 ( @xmath104 ) : @xmath105>\rho_{\ic}. \label{eq : rho_c(df=0)}\end{aligned}\ ] ] from eq .
( [ eq : rho_c ] ) , we can write @xmath106 @xmath82 and @xmath102 vanish together at the order - disorder transition point @xmath107 .
note that we take @xmath102 as a constant for @xmath88 far from @xmath108 , but it is a function of @xmath88 near @xmath108 which vanishes at @xmath107 .
the order parameter at the c - ic phase transition , @xmath109 ( @xmath110 ) , should satisfy both eqs .
( [ eq : rho_c(df=0 ) ] ) and ( [ eq : rho_c2 ] ) . for @xmath111 , from eq .
( [ eq : rho_c(df=0 ) ] ) , @xmath112 . \nonumber\\\end{aligned}\ ] ] substituting this and eq .
( [ eq : rho_c(df=0 ) ] ) into eq .
( [ eq : rho_c2 ] ) and keeping terms up to the order in @xmath113 , @xmath114\tilde{z}_c^2\nonumber\\ \tilde{z}_c^2 & \approx&\frac{2x}{1-l(l-3)x},\end{aligned}\ ] ] where @xmath115 .
since @xmath116 from _ assumptions 1 _ and _ 2 _ , @xmath113 ( @xmath117 ) is represented by a simple function of @xmath118 : @xmath119 by virtue of _ assumption _ _ 1 _ , it is apparent that @xmath111 , which is consistent with the case we consider here ; therefore it is confirmed that @xmath120 in eq .
( [ eq : transition ] ) is the c - ic transition point for the case where _ assumptions 1 _ and _ 2 _ hold .
the width of the commensurate region is given by @xmath121 the commensurate region becomes exponentially narrow as the cdw period @xmath19 increases .
therefore , only commensurate stripes with the relatively small l may exist in practice .
note that , here , we do not consider the case where discommensurations exist @xcite .
at a commensurate point , the condensation energy is given by @xmath122
now we consider the slightly displaced cdw to study collective excitations around the equilibrium state .
first , we discuss the _ phason _ and then the _ amplitudon_. for the _ phason _ , we introduce an @xmath123-dependence into the phase of the order parameters : @xmath124 $ } , \label{eq : psi - phason}\\ & & \mbox{$\chi(x,\phi ) = -\i m \exp\[\i(\frac{q}{2 } x + \frac{\phi_0}{2l}+ \frac{\varphi(x)}{l})\]$}\\ & & \varphi(x)= \xi \cos(kx+\phi_1 ) , \label{eq : phase}\end{aligned}\ ] ] where @xmath125 with @xmath126 .
the free energy is given by @xmath127 + p(x)\re{\psi(x)^l } + g_0\left|q(z)-q+\frac{k\xi\sin(kx+\phi_1)}{l}\right|^2|\psi(x)|^2\right\}\\ & \approx & f_0(\rho)+f^{(\varphi)}_2(\rho)\xi^2,\\ f_0(\rho)&= & f^{(0)}(\rho ) -|\tilde{p}_1|\,\rho^l + \tilde{g}_0[q(z)-q]^2 \rho^2,\\ f^{(\varphi)}_2(\rho)&= & \frac{1}{2}\,|\tilde{p}_1|\,\rho^l\frac{1+\delta_{k,0}\cos(2\phi_1)}{2 } + \frac{1}{2}\frac{\tilde{g}_0\rho^2}{l^2}k^2 = \frac{1+\delta_{k,0}\cos(2\phi_1)}{2}\frac{\rho^2}{l^2 } \[\frac{l^2}{2}\,|\tilde{p}_1|\,\rho^{l-2 } + \tilde{g}_0 k^2\],\end{aligned}\ ] ] assuming that @xmath58 depends on time @xmath128 only through @xmath129 , the kinetic energy is given by @xmath130 where @xmath131 is a ginzburg - landau parameter of the kinetic energy and defined by this equation .
thus , the lagrangian corresponding to the fluctuating cdw is @xmath132 \frac{t_0 \rho^2}{l^2}$}\nonumber\\&&\mbox{$\times\ [ \(\frac{d\xi}{dt}\)^2 -\frac{1}{t_0 } \ ( \frac{l^2}{2}\,|\tilde{p}_1|\,\rho^{l-2 } + \tilde{g}_0k^2\)\xi^2 \]$}\nonumber\\ & & { } + \mathrm{const.}\end{aligned}\ ] ] solving the euler - lagrange equation of motion , @xmath133.\end{aligned}\ ] ] @xmath134 is gapless in the ic case and gapped in the c case . from eq .
( [ eq : rho_c2 ] ) , the gap at @xmath135 can be evaluated as @xmath136 \big[1-(l-2)\tilde{z}^2\big],\nonumber\\\end{aligned}\ ] ] where we have used @xmath137 in the commensurate region [ see eq .
( [ eq : transition ] ) ] . in fig .
[ fig : gap ] , the doping dependence of the phason gap is shown for the @xmath138 case . in appendix [ app : gl - estimate ] , @xmath131 is roughly estimated as @xmath139 @xmath140}$ ] using a single - band peierls - hubbard model . @xmath141},\end{aligned}\ ] ] where @xmath142 ( @xmath143 ) is the commensurability energy .
@xmath144 is roughly estimated as @xmath145 [ ev ] in appendix [ app : gl - estimate ] ( @xmath146 is the on - site coulomb interaction in the single - band peierls - hubbard model ) .
therefore , the order of the phason gap at @xmath147 is @xmath148 } \sim u^\frac{7}{2}m^4\,\mbox{[ev]}. \label{eq : phason - gap}\end{aligned}\ ] ] we expect that @xmath149 [ ev ] and @xmath150 ( _ assumption 4 _ ) , and thus find the order of @xmath151 is less than @xmath152 [ mev ] . for a smaller @xmath146
, it can be even much smaller than @xmath152 [ mev ] : e.g. for @xmath153 [ ev ] , we find @xmath154 [ mev ] .
next we consider the _ amplitudon _ mode .
for this purpose , we write the order parameters in the following form : @xmath155 \exp\[\i(q x + \frac{\phi_0}{l})\]$},\\ & & \mbox{$\chi(x,\phi ) = -\i m\[1+\mathcal{a}(x)\ ] \exp\[\i(\frac{q}{2 } x + \frac{\phi_0}{2l})\]$}\\ & & \mathcal{a}(x)=\zeta \cos(kx+\phi_1).\end{aligned}\ ] ] then , the free energy is given by @xmath156 + p(x)\re{\psi(x)^l } + g_0\[q(z)-q\]^2|\psi(x)|^2 + g_0\[k\zeta\sin(kx)\]^2\rho^2\right\}\\ & \approx&f_0(\rho)+f^{(\mathcal{a})}_2(\rho)\,\zeta^2,\\ f^{(\mathcal{a})}_2(\rho)&= & \frac{1+\delta_{k,0}\cos(2\phi_1)}{2 } \bigg\ { \frac{|r_0|^2}{u_0 } + \frac{|r_0|\,s_0}{u_0}\rho + \big(v_0 + \tilde{g}_0[q(z)-q]^2\big ) \rho^2 + \frac{l(l-1)}{2}\,|\tilde{p}_1|\,\rho^{l } + \tilde{g}_0k^2\rho^2\bigg\}.\end{aligned}\ ] ] thus , we find the frequency of the _ amplitudon _ fluctuation as @xmath157 ^ 2\nonumber\\ & & \left.\hspace{.5cm}{}+\frac{l(l-1)}{2 } + \tilde{g}_0k^2\]\\ & \equiv&\delta_{\mathcal{a}}^2 + \frac{\tilde{g}_0}{t_0}k^2\end{aligned}\ ] ] the _ amplitudon _ has a gap even in the ic case ( unlike the _ phason _ ) .
the _ amplitudon _ gap at @xmath0 doping is @xmath158}\\ & \lesssim & \mbox{$u m\times10^{16}$ [ hz]},\end{aligned}\ ] ] where @xmath146 is in the unit of ev , and @xmath159 [ ev ] ( see appendix [ app : gl - estimate ] ) .
so far , we have discussed the one - dimensional collective modes , that is the modes that are uniform in the direction along stripes .
next we consider the _ meandering _ mode of stripes , i.e. the transverse mode with the wavevector oriented along the stripe direction .
this case is a simple extension of that for the one - dimensional phason .
the only difference is that now we impose a @xmath160-dependence on @xmath161 and introduce the energy cost associated with the spatial variation in the @xmath160-direction .
then , we obtain the following lagrangian @xmath162 \frac{t_0 \rho^2}{l^2}\nonumber\\ & & \times\int \frac{dy}{v_y}\ [ \(\frac{d\xi}{dt}\)^2 -\frac{\tilde{h}_0}{t_0}\(\frac{d \xi}{dy}\)^2\right.\nonumber\\ & & \left.\hspace{1.5cm}{}-\frac{1}{t_0 } \ ( \frac{l^2}{2}\,|\tilde{p}_1|\,\rho^{l-2 } + \tilde{g}_0k_x^2\)\xi^2 \]\nonumber\\ & & { } + \mathrm{const . } , \label{eq : l - meandering}\end{aligned}\ ] ] where @xmath163 is the energy coefficient of the @xmath160-derivative term . solving the euler - lagrange equation of motion ,
we find @xmath164 ^ 2&= & \frac{1}{t_0 } \[\tilde{h}_0
k_y^2 + \tilde{g}_0k_x^2\ ] + \delta_\varphi^2.\end{aligned}\ ] ] in a similar manner , we find the two - dimensional amplitudon as @xmath165 ^ 2&= & \frac{1}{t_0}\[\tilde{h}_0k_y^2 + \tilde{g}_0 k_x^2\ ] + \delta_\mathcal{a}^2.\end{aligned}\ ] ]
to investigate collective oscillation of stripes which can be experimentally examined , we consider the _ meandering _ mode contribution to the conductivity . since we do not include impurity pinning effects in this paper ,
our treatment is equivalent to the classical particle model in the one - dimensional case . in the presence of an electric field
@xmath166 $ ] along the @xmath123 direction , the equation of motion is given by ( see appendix [ app : eq - motion ] or reference @xcite ) @xmath167,\end{aligned}\ ] ] where @xmath168 is the effective mass of the cdw ( see appendix [ app : gl - estimate ] ) .
the particular solution of this equation is @xmath169 ^ 2-\omega^2+\i\omega\gamma}.\end{aligned}\ ] ] therefore , an induced current - density is @xmath170 where @xmath171},\end{aligned}\ ] ] and @xmath172 is the lattice constant in the @xmath13-direction .
at @xmath0 doping , from the effective mass ( @xmath173 [ kg ] ) estimated in appendix [ app : gl - estimate ] , it follows @xmath174 } \sim 10^{14}\,\,\mbox{[hz]}. \label{eq : plasma}\end{aligned}\ ] ] once we know @xmath175 accurately , the charge order parameter can be determined .
the value of @xmath175 will be determined from the phason contribution to the dielectric function and the phason gap , @xmath176 ( see below ) .
the conductivity is given by @xmath177 the contribution of the _ meandering _ mode to the dielectric function is @xmath178 the contribution of the _ phason _ mode to the static dielectric constant is @xmath179 where @xmath142 is in unit of hertz . from the experimental data for @xmath180 , @xmath142 can be determined .
we can also determine @xmath175 from @xmath180 and @xmath181 , and the effective mass can be estimated from @xmath175 ( this would be close to that of a free electron ) . at @xmath0 doping , it follows from the estimated phason gap ( [ eq : phason - gap ] ) and plasma frequency of the cdw ( [ eq : plasma ] ) that @xmath182 where @xmath146 is in unit of ev .
in this paper , we investigated the competition between c and ic stripe phases by calculating the c - ic transition point in the weak - pinning approximation .
this allowed us to estimate the strength of the lock - in effect , and evaluate the phason and amplitudon fluctuation frequencies , the ac conductivity , and the dielectric function .
our calculation suggests a narrow c region around @xmath0 doping . in this region , there is a weakly - pinned _ phason _ mode ( with a charge - order - dependent gap at @xmath135 ) .
the phason gap at @xmath135 for @xmath0 doping is estimated to be of the order of @xmath183 mev .
the pinning frequency increases with @xmath184 , quadratically near @xmath185 and linearly for large @xmath184 .
this weakly - pinned behavior should be observable by electronic transport experiments . in the ic regime ( away from @xmath0 doping )
, the phason frequency is linear in @xmath184 with no gap at @xmath135 .
the _ amplitudon _ shows a similar @xmath186-dependent frequency as the _
phason_. the _ amplitudon _ gap at @xmath135 in the c phase is larger than that of the phason , and remains finite in the ic case . in the presence of an electric field across the stripes , _
phason _ or _ meandering _ fluctuations can be induced . from measurements of the static dielectric function or conductivity , it should be possible to extract the amplitude of the charge order parameter .
these and other measurements would help constrain parameters in the ginzburg - landau model considered here . also , direct real - space calculations @xcite can decrease the number of independent parameters .
a similar model can be applied to describe commensurate - incommensurate transitions and response functions in other related classes of materials , including nickelates and manganites .
this work was supported by the u.s .
here , we consider a single - band peierls - hubbard hamiltonian and estimate the electron - lattice coupling strengths .
the hamiltonian reads @xmath187c^\dagger_{i\sigma}c_{j\sigma } \nonumber\\ & & { } + u\dsum{i}{}n_{i\up}n_{i\dn } + 4\gamma\dsum{i,\sigma}{}w_ic^\dagger_{i\sigma}c_{i\sigma } \nonumber\\ & & { } + \dfrac{1}{2m}\dsum{i}{}(p_{xi}^2+p_{yi}^2 ) + \dfrac{k}{2}\dsum{i}{}(u_i^2+v_i^2),\end{aligned}\ ] ] where @xmath188 @xmath189 ( @xmath172 ) is the hole creation ( annihilation ) operator , @xmath190 is the cu - cu hopping between the nearest neighbors ( @xmath128 ) or the next - nearest neighbors ( @xmath191 ) , and @xmath192 and @xmath193 are the displacements of the @xmath123-oxygen and @xmath160-oxygen at site @xmath194 , respectively . to estimate the electron - lattice coupling strength in the single - band model , we start from the three - band ( cu - o ) model .
the cu on - site energy and the cu - cu hopping integral to the zeroth order in @xmath195 are given by @xmath196 the second - order perturbation to the cu on - site energy and the cu - cu hopping integral are given by @xmath197 where the cu - o hopping and @xmath198-@xmath199 energy difference depend linearly on lattice displacements , @xmath200 and @xmath201 refer to the oxygen ion to the right of , above , to the left of , and below the cu ion , respectively ; and @xmath202 to that between the @xmath194 and @xmath203 cu ions . by retaining up to linear terms of @xmath204 and @xmath205 , @xmath206 and @xmath207 are approximately @xmath208 combining now the zeroth and second - order results , for the effective one - band model , we obtain @xmath209 with the reduced electron - lattice coupling strengths @xmath210 and
@xmath211 : @xmath212 using the three - band parameters @xmath213 [ ev ] , @xmath214 [ ev ] , @xmath215 [ ev / ] , @xmath216 [ ev / ] and @xmath217 [ ev / @xmath218 @xcite , we obtain the single - band parameters : @xmath219 [ ev ] , @xmath220 [ ev / ] and @xmath221 [ ev / ] . here
the value of the cu - cu hopping energy is similar to that estimated in ref .
@xcite ( @xmath222 [ ev ] ) ; therefore , it is also expected that @xmath223 [ ev ] from ref .
here we represent the ginzburg - landau parameters for kinetic and pinning energies , @xmath131 and @xmath44 , by using the order parameters and other parameters of a single - band peierls - hubbard model , and estimate the commensurability energy and the effective mass of the cdw .
we start with the single - band hamiltonian in appendix [ app : single ] .
we can adopt a particular set of parameters obtained there : e.g. , @xmath219 [ ev ] , @xmath224 [ ev ] , @xmath225 [ ev / ] , @xmath226 [ ev / ] and @xmath217 [ ev / @xmath218 , etc .
however , we will not necessarily require all of these values below , because it is not our purpose to calculate the order parameters here .
moreover , the order parameters are assumed to be @xmath227 and @xmath228 in this paper ( _ assumptions 4 _ and _ 5 _ ) .
our purpose here is to estimate the commensurability energy , the kinetic energy , and the effective mass of the cdw . in the estimation of the commensurability energy
, we will find that the commensurability energy directly depends on only @xmath128 , @xmath146 and @xmath68 as long as the order parameters are small ( @xmath229 ) . in the estimation of the kinetic energy and the effective mass
, we will find that those values depend on @xmath210 , @xmath230 and @xmath58 as long as @xmath231 is small ( @xmath232 [ ] ) .
therefore , we take the parameters as @xmath219 ev , @xmath233 , and @xmath234 [ ] ; and we assume that the calculation with the resulting parameters yields order parameters which satisfy _ assumptions 4 _ and _ 5_. we also consider the terms of @xmath211 and @xmath191 to be small enough to be neglected , compared to the @xmath128 term . defining the fluctuations : @xmath235 we rewrite the hamiltonian : @xmath236 where @xmath237,\nonumber\\ \\ \mathcal{h}_{z}&= & \dsum{i}{}\big [ 4\gamma\dsum{\sigma}{}\tilde{w}_i \<c^\dagger_{i \sigma}c_{i \sigma}\ > + k\big(\<u_i\>\tilde{u}_i+\<v_i\>\tilde{v}_i\big ) \big],\nonumber\\ \\ \mathcal{h}_{int}&= & \dsum{i}{}\big[\ , \dfrac{u}{2}\dsum{\sigma } { } \tilde{n}_{i\bar{\sigma } } \tilde{n}_{i\sigma } + 4\gamma\dsum{\sigma } { } \tilde{w}_i \tilde{n}_{i\sigma } \,\big],\\ \mathcal{h}_{ph}&= & \dsum{i}{}\big[\ , \dfrac{1}{2m}\big(p_{xi}^2+p_{yi}^2\big ) + \dfrac{k}{2 } \big(\tilde{u}_i^2+\tilde{v}_i^2\big)\,\big],\\ \mathcal{h}_{c}&= & -u\dsum{i}{}\<n_{i\up}\>\<n_{i\dn}\ > + \dfrac{k}{2}\dsum{i}{}\big(\<u_i\>^2+\<v_i\>^2\big).\nonumber\\\end{aligned}\ ] ] according to the mean - field approximation , we neglect the @xmath238 and @xmath239 terms . to consider the equilibrium states , the terms linear in @xmath240 and @xmath241 are set to zero , @xmath242 .
we assume the order parameters to be described as in eqs .
( [ eq : order - ch ] ) and ( [ eq : order - spn ] ) : @xmath243,\\ \<n_{i\up}\ > - \<n_{i\dn}\ > & = & \[\chi(x_i,\phi ) + \chi^*(x_i,\phi)\](-1)^{-\frac{x_i}{a}}\nonumber\\ & = & \chi(x_i,\phi ) \e^{\i \frac{g}{2}x } + \[\chi(x_i,\phi)e^{\i \frac{g}{2}x}\]^*.\end{aligned}\ ] ] hence , @xmath244 can be expanded as @xmath245 where @xmath246 . from @xmath247 , @xmath248\\ & = & \frac{\gamma}{k } \[\psi(x_{i+1_x},\phi)-\psi(x_{i},\phi)\]+\mbox{c.c.}\\ & = & \i \frac{2\gamma}{k } \psi(x_i,\phi ) \e^{\i qa } \sin(qa ) + \mbox{c.c.}\end{aligned}\ ] ] we substitute the @xmath249-represented form of @xmath244 and @xmath250 into @xmath251 , and expand @xmath172 and @xmath189 as @xmath252 where @xmath253 runs from 0 to @xmath254 for a commensurability @xmath23 , the sum with respect to @xmath255 is taken within the reduced brillouin zone spanned by @xmath249 , and @xmath256 denotes the number of @xmath186 points in the original brillouin zone .
hence , @xmath251 can be rewritten as @xmath257,\nonumber\\\end{aligned}\ ] ] where @xmath258\nonumber\\ & & -4t'\cos(k_x a ) \cos(k_y a ) + u\frac{\rho_0}{2}\\ \tilde{u}&=&\[u-\frac{8 \gamma^2}{k}\sin^2\(qa\)\].\end{aligned}\ ] ] only the states near @xmath259 at the fermi level contribute to the condensation energy .
the energy gap around @xmath259 is approximately given by @xmath260 where @xmath261 and @xmath262 .
the approximation here is justified when @xmath263 and @xmath264 .
this is the case for @xmath138 and @xmath233 under _ assumptions 4 _ and _ 5_. therefore , the commensurability effect on the condensation energy , @xmath265 , is given by @xmath266 now we consider the energy to be calculated from the meanfield hamiltonian @xmath267 in the energy , the component proportional to @xmath268 contributes to the formation of the ordered state as well as the @xmath47 term in the ginzburg - landau free energy ( [ eq : f0 ] ) .
the condensation energy per site is given by @xmath269 where the first term is from @xmath270 , and the second from @xmath251 ; and @xmath271 depends on the model system parameters ( bandwidth , interactions , etc . ) and doping .
note that @xmath272 should have two components which are , respectively , proportional to @xmath268 and @xmath273 .
the former is the @xmath274-independent condensation energy , @xmath275 , and the latter is the commensurability energy , @xmath276 . setting @xmath277 , @xmath278 can be written by @xmath279 the commensurability energy is given by @xmath280 for an insulating cdw in the nearly - half - filled one - dimensional system , the condensation energy is roughly estimated as @xmath281 , so that we expect @xmath282 [ ev ] , @xmath283 should be on the order of @xmath146 [ ev ] . in this case , the commensurability energy per site is obtained by @xmath284}.\end{aligned}\ ] ] then , we obtain the ginzburg - landau parameter for pinning energy @xmath285 . next ,
to estimate the kinetic energy , we consider the time - dependence of the order parameter through @xmath274 @xmath286 the displacements are written as @xmath287 @xmath288 where @xmath289 .
the mean square velocity of the ion and the total kinetic energy are @xmath290 @xmath291,\end{aligned}\ ] ] where @xmath292 , @xmath293 , and @xmath168 are the ionic mass per site , the electron mass , and the effective mass of a hole in the cdw , respectively . on the other hand , the kinetic energy in the ginzburg - landau expansion is written by @xmath294 v_d^2,\end{aligned}\ ] ] where both @xmath295 and @xmath296 are independent of @xmath58 .
near @xmath0 doping , @xmath295 and @xmath296 are estimated as @xmath297}\\ \tau_1 & = & \frac{m}{2}\(\frac{2\gamma}{k}\)^2 \sin^2\(\frac{qa}{2}\)\\ & \approx & 1.7\(\frac{\gamma}{k}\)^2 \times 10^{-27}\,\,\mbox{[ev s$^2$]}\end{aligned}\ ] ] therefore , the ginzburg - landau parameter for the kinetic energy , @xmath131 , is given by ( for @xmath135 and @xmath298 ) @xmath299 \times 10^{-32}\,\,\mbox{[ev s$^2$]}\nonumber\\ & \approx & 4.2 \rho^{-2}\times10^{-32}.\end{aligned}\ ] ] where we use @xmath300 and the _ assumptions 4 _ and _ 5_. the effective mass is @xmath301/\rho_0\end{aligned}\ ] ] in the case of @xmath135 and @xmath298 , @xmath302 \times 10^{-31}\,\,\mbox{[kg]}\nonumber\\ & \approx & 9.2 \times 10^{-31}\,\,\mbox{[kg]}.\end{aligned}\ ] ] for @xmath303 , the electronic energy is dominant in the total kinetic energy .
here , we derive the equation of motion for a meandering cdw oscillating in the @xmath123-direction in an ac electric field ( discussed in sec . [ sec : electric - field ] ) .
we consider a plane electromagnetic wave with the electric field oriented in the @xmath123-@xmath160 plane , @xmath304 .
the corresponding vector potential is @xmath305 .
the vector potential couples to the electrical current @xmath306 , which can be expressed in terms of the carrier density and the velocity . the velocity of the density wave can be expressed via the time derivative of this phase , @xmath307 .
thus the form of the phase that couples to the plane - wave electric field as above is @xmath308 here we take @xmath309 .
thus , the additional contribution to the lagrangian due to the interaction with the external electromagnetic field is @xmath310 . combining it with the bare lagrangian , eq .
( [ eq : l - meandering ] ) , we obtain @xmath311 \nonumber\\ & & - \int\frac{dy}{v_y } \frac{e\rho_0 } { g } \frac{e_x}{(-\i\omega)}\dot{\xi}.\end{aligned}\ ] ] adding a dissipation term , we obtain the equation of motion : @xmath312 99 t. suzuki , t. goto , k. chiba , t. shinoda , t. fukase , h. kimura , k. yamada , m. ohashi and y. yamaguchi , phys .
b * 57 * , r3229 ( 1998 )
. j. m. tranquada , b. j. sternlieb , j. d. axe , y. nakamura , and s. uchida , nature ( london ) * 375 * , 561 ( 1995 ) .
j. m. tranquada , j. d. axe , n. ichikawa , a. r. moodenbaugh , y. nakamura , and s. uchida , phys .
lett . * 78 * , 338 ( 1997 ) .
h. a. mook , pengcheng dai , and f. doan , phys .
88 * , 097004 ( 2002 ) .
r. j. mcqueeney , y. petrov , t. egami , m. yethiraj , g. shirane , and y. endoh , phys .
lett . * 82 * , 628 ( 1999 ) .
e. s. bozin , s.j.l .
billinge , g. h. kwei , h. takagi , phys .
b * 59 * , 4445 ( 1999 ) .
a. bianconi , n. l. saini , a. lanzara , m. missori , t. rossetti , h. oyanagi , h. yamaguchi , k. oka , and t. ito , phys .
. lett . * 76 * , 3412 ( 1996 ) .
h. oyanagi , h. yamaguchi , k. oka , and t. ito z. g. yu , j. zang , j. t. gammel , and a. r. bishop , phys .
b * 57 * , r3241 ( 1998 )
. i. martin , e. kaneshita , a. r. bishop , r. j. mcqueeney , and z. g. yu , phys .
b * 70 * , 224514 ( 2004 ) .
a. r. moodenbaugh , youwen xu , m. suenaga , t. j. folkerts , and r. n. shelton , phys .
b * 38 * , 4596 ( 1988 )
. l. w. mcmillan , phys .
b * 12 * , 1187 ( 1975 ) .
l. w. mcmillan , phys .
b * 14 * , 1496 ( 1976 ) .
l. p. pryadko , s. a. kivelson , v. j. emery , y. b. bazaliy and e. a. demler , phys .
b * 60 * , 7541 ( 1999 ) .
h. kimura , k. hirota , h. matsushita , k. yamada , y. endoh , s. h. lee , c. f. majkrzak , r. erwin , g. shirane , m. greven , y. s. lee , m. a. kastner , and r. j. birgeneau , phys .
b * 59 * , 6517 ( 1999 ) .
g. grner , _ density waves in solids _
( perseus publishing , cambridge , massachusetts , 2000 ) . k. yonemitsu , a. r. bishop , and j. lorenzana , phys . rev .
b * 47 * , 12059 ( 1993 ) .
m. s. hybertsen , m. schlter , and n. e. christensen , phys .
b * 39 * , 9028 ( 1989 ) .
m. s. hybertsen , e. b. stechel , m. schlter , and d. r. jennison , phys .
b * 41 * , 11068 ( 1990 ) . | we analyze the anomalous behavior in hole - doped cuprates near @xmath0 doping in terms of the commensurate - incommensurate transition of a stripe phase . based on an effective ginzburg - landau theory appropriate for weak pinning ,
we calculate the commensurate - incommensurate transition point and the energies of the phason and amplitudon collective modes . using experimentally available parameters
, we estimate the phason gap ( pinning frequency ) , the conductivity and the contribution of the phason mode to the dielectric function . |
through its influence on the diffusion of spin - bearing particles , the physical characteristics of the medium leaves its signature on the observed nmr signal .
this makes nmr a powerful probe into the microstructure of biological specimens as well as other porous media .
nmr s sensitivity to diffusional processes can be controlled by introducing gradient waveforms into standard nmr experiments @xcite .
the effect of free diffusion on experiments with general gradient waveforms has been fully characterized @xcite .
however , the complexity of the environment typically leads to non - gaussian motion , which yields interesting features in the nmr signal decay that depend on the local structure .
for example , a time - dependence in the diffusion coefficients @xcite tend to emerge in complex environments , while the large wavenumber regime of the signal decay could exhibit unique features @xcite as well .
time dependence in diffusion coefficients was investigated extensively with emphasis on and ageing @xcite .
it has also been studied as a possible indicator of mesoscopic disorder in materials and tissues , making use of diffusion - sensitive nmr images @xcite ( see ref .
@xcite for a comprehensive review on diffusion in tissues ) .
fractional brownian motion is yet another viable model in which to view the signatures of non - gaussian motion in nmr signals , which is amenable to path integral methods @xcite similar in that regard to the approach discussed in the present article . in this paper
, we consider the case of diffusion taking place under the influence of a hookean force field .
this problem is well - studied in the field of stochastic processes @xcite .
to our knowledge , the first occurrence of it in the nmr literature is in a couple of papers published in 1960s @xcite , wherein the authors consider the case of `` diffusion near an attractive center '' and derive the signal expression for their then recent pulsed field gradient framework @xcite .
since then callaghan and pinder used the formalism on a semi - dilute solution of an entangled polystyrene to model the displacements of the resulting network , which is stable during the timescale of the nmr measurement @xcite .
le doussal and sen employed the problem as an `` artificial pore '' model to understand the effects of restricted diffusion in the presence of nonlinear magnetic field gradients @xcite . mitra and
halperin also touched upon the problem , referring to it as the `` parabolic pore '' in the context of the center - of - mass propagators @xcite . in this article
, we have two main goals : ( i ) to provide an analytical framework with which one can derive explicit relationships for the signal decay for a general time - dependent gradient waveform , and ( ii ) develop semi - analytical and numerical tools , which could be generalized with relative ease to complicated potentials . as demonstrated below , our formulation naturally leads to a new characterization of diffusion anisotropy ; similar in spirit to the diffusion tensor model @xcite .
however , while the latter tries to fit imaging data to a free diffusion model despite the underlying non - free diffusion , the present framework allows taking confinement into account , which is modeled as a harmonic force .
the advantage of taking confinement into account is found in the time dependence of the predicted signal profile , which exhibits features more similar to that of actual restricted diffusion . employing a harmonic confinement instead of a direct implementation of restricted diffusion , on the other hand
, offers a degree of tractability that encodes full anisotropy easily .
for instance , compared to a restricted diffusion model involving a capped cylinder @xcite , which has _ two _ length parameters due to its partial isotropy , the confinement tensor generally has three distinct size parameters ( related to its eigenvalues ) that can capture full three - dimensional anisotropy , much like the diffusion tensor model .
the article is organized as follows : in the next section , sec .
[ sec : pathint ] , we express the nmr signal as a path integral and evaluate it assuming a hookean potential as stated above , providing a general expression that can be used with arbitrary gradient sequences . as such , we end the section by deriving analytical expressions for the expected nmr signal for the conventional pulsed field gradient ( stejskal - tanner ) and oscillating gradient sequences . for the sake of keeping the main text to the point , technicalities involved in evaluating the path integral are discussed in great detail in appendix [ app : pathint ] . in the subsequent section , sec .
[ sec : mcf ] , we introduce the formulation of the semi - analytical multiple correlation function framework ( mcf ) @xcite , assuming a general confining potential .
its application to the special case of a hookean potential is presented in appendix [ app : mcf ] , along with relevant technical details . in sec .
[ sec : sim ] , the random walk simulations are described before proceeding with the validation of the ( semi-)analytical approaches and the deliberation of the hookean model as an alternative tool for the characterization of diffusion anisotropy in the results section , followed by concluding remarks .
here , we derive an analytical expression via a path integral for the average transverse magnetization in a specimen , stemming from spin - bearing random walkers subject to a hookean potential . in order to isolate details of the derivation from the results ,
we have placed most of the former in appendix [ app : pathint ] while restricting to the essential points in this section . in a medium subjected to a ( spatially ) non - uniform magnetic field , a diffusing spin - carrier experiences a rate of ( larmor ) precession that varies with respect to its spatial position . compared to a steady precession in the case of a uniform magnetic field
, therefore , each particle picks up a phase in its rotation , depending on its random path .
hence , from an ensemble of such random walkers in a ( time - dependent ) spatially linear magnetic field gradient @xmath0 of duration @xmath1 , an average signal of @xmath2 is obtained , which is the average transverse magnetization up to a unit . here , since the average is to be taken over all random paths @xmath3 , the expectation value has the form of a path integral , @xmath4 \ , { { \mathrm e}}^{- { { \mathrm i}}\gamma \int_0^{{{t_\mathrm{f } } } } { { \mathrm d}}t \ , \bm{g } ( t ) \cdot { \bm{r}}(t ) } \ ; , \label{eq : path.int}\end{aligned}\ ] ] where @xmath5 denotes an integral over the space of paths , and @xmath6 $ ] the weight of the path @xmath3 .
note that @xmath7 is a normalized quantity that takes the value of 1 when @xmath8 , and @xmath9 is the gyromagnetic ratio .
the probability weight @xmath6 $ ] is most easily imagined as an infinite product of stepwise probabilities of transition from each intermediate path point to the next . in this article
, we assume that the stochastic process underlying the random paths is diffusion with diffusivity @xmath10 under the effect of a _ dimensionless _
hookean potential , @xmath11 where we refer to @xmath12 , having dimension of inverse length squared , as the _ confinement tensor_. the true tensorial force constant can be obtained by multiplying @xmath13 by the boltzmann constant @xmath14 and the absolute temperature @xmath15 .
we argue that the potential defined above suffices to capture relevant features of diffusion in confined spaces .
the advantage of the assumption is that the stepwise probabilities mentioned above have a simple form ( see appendix [ app : pathint ] ) and the resulting path weight is analytically tractable .
we find that the nmr signal , or average transverse magnetization , that ensues the application of an arbitrary gradient waveform @xmath0 is expressed as @xmath16 where @xmath17 and @xmath18 is a matrix of rate constants ( inverse time ) , simply proportional to the confinement tensor @xmath12 .
the inverse of this matrix encodes the time scales involved in equilibration , namely the approach of a distribution of diffusing particles toward the boltzmann distribution @xmath19 .
essentially , it is the finite width of this final gaussian profile that is regarded as a measure of confinement in the present article , by matching it to the width of an actual confined geometry ( appendix [ sec : analytical1d ] ) . the results derived above can be employed to obtain exact expressions for the nmr signal intensity for any gradient waveform . in this section
, we shall consider two widely employed experiments illustrated in figure [ fig : waveforms ] .
the first sequence is stejskal and tanner s pulsed field gradient technique @xcite while the second one features an oscillatory ( sinusoidal ) profile @xcite . here
, we use eqs . and to predict the nmr signal that should be recovered from spin - bearing random walkers subject to a three dimensional harmonic potential in a traditional pulsed field gradient experiment introduced by stejskal and tanner @xcite . as illustrated on the top panel in figure [ fig : waveforms ] ,
the effective waveform comprises two gradient pulses in opposite directions with a delay @xmath20 between their leading edges .
each pulse has magnitude @xmath21 and duration @xmath22 . due to the tractability of our `` harmonic approximation , '' an expression for the signal
can be obtained easily , valid for all possible values of the experimental parameters .
the first thing to do is to evaluate the integral for @xmath23 , which is straightforward ( as noted before , one may wish to consider temporarily the eigenbasis ) .
one finds ( taking the transpose for convenience ) , @xmath24 { { \mathrm e}}^{\bm{\omega } t } & , & 0 < t < \delta \\
\left [ { { \mathrm e}}^{-\bm{\omega } ( \delta+\delta ) } - { { \mathrm e}}^{-\bm{\omega } \delta } \right ] { { \mathrm e}}^{\bm{\omega } t } & , & \delta < t < \delta \\ { { \mathrm e}}^{-\bm{\omega}(\delta+\delta ) } { { \mathrm e}}^{\bm{\omega } t } - \bm{1 } & , & \delta < t <
\delta + \delta \\ 0 & , & \delta + \delta < t \end{array } \right . \;,\label{eq : q.1pfg } \end{aligned}\ ] ] where @xmath25 represents the @xmath26 identity matrix . what remains is the tedious task of squaring we of course mean @xmath27 .
it is useful to note that all the matrices that will end up between @xmath28 and @xmath29 according to eqs . and commute ( since they all are eventually powers of @xmath30 ) and are symmetric , simplifying the algebra greatly . [
ft : algebra ] ] and integrating this expression to evaluate the signal .
after a bit of algebra and simplifications , we obtain the quadratic expression ( in @xmath31 ) @xmath32 where the real symmetric matrix @xmath33 \ .
\label{eq : amatrix } \end{aligned}\ ] ] the last two expressions provide a model alternative to the diffusion tensor model @xcite commonly used in biomedical applications of magnetic resonance imaging .
the main difference is the dependence of the predicted signal on the timing parameters of the sequence . in diffusion tensor imaging ( dti )
, this dependence is the same as that for free ( gaussian ) diffusion . in the hookean potential model ,
the dependence is consistent with restricted diffusion @xcite . as a check ,
one may evaluate eqs . and to order @xmath34 , to obtain @xmath35 which is the anisotropic generalization of stejskal s result @xcite for a spherically symmetric harmonic potential , with the additional @xmath36 term . in the @xmath37 ( i.e. , @xmath38 ) limit , the stejskal - tanner expression for free diffusion
is recovered @xcite as expected .
next , we shall consider the sinusoidal waveform with angular frequency @xmath39 as depicted in the bottom of figure [ fig : waveforms ] .
owing to the simplicity of diffusion inside a harmonic potential , we were able to derive an analytical expression for the average magnetization for this waveform as well .
since the gradient is applied in a fixed direction , the formulation in appendix [ sec : analytical1d ] is sufficient to obtain the exact result .
the gradient profile is of the form @xmath40 .
we consider an experiment with @xmath41 full periods , i.e. , @xmath42 vanishes outside the interval @xmath43 .
one finds , from eq . , that @xmath44 \ ; , \end{aligned}\ ] ] where @xmath45 .
the signal then follows from eq . after some rearrangement as @xmath46 with @xmath47
in this section , we shall visit the same problem using the multiple correlation function ( mcf ) formalism , which has been used in the past to characterize the effect of restricted diffusion on the nmr signal .
technical details may be found in appendix [ app : mcf ] .
we refer the reader to ref .
@xcite for a recent review of the technique consistent with the notation here , and its relations with the path integral framework discussed in the previous section .
the multiple correlation function formalism aims to compute the nmr signal @xmath48 from the time evolution of the ( appropriately - normalized ) magnetization density @xmath49 . in the presence of a potential @xmath50
, diffusion is governed by the smoluchowski equation @xcite , while the transverse magnetization carried by the random walkers need also be taken into account .
this suggests that , under a magnetic field gradient waveform @xmath0 , the evolution of @xmath49 is governed by an equation akin to the bloch - torrey equation @xcite , which we shall refer to as the bloch - torrey - smoluchowski ( bts ) equation , @xmath51 where this time the diffusion term deals also with the potential .
clearly , the above expression is reduced to the bloch - torrey equation when the potential is zero .
the solution @xmath49 of the bts equation is best considered in the abstract function space , in terms of a propagator responsible for the evolution of the magnetization in time . due to the explicit time dependence of the operator on the right hand side of eq . , the associated propagator is formally a time - ordered product of operators of the form @xmath52 \}$ ] over successive time intervals of length @xmath53 .
the abstract operators @xmath54 and @xmath55 correspond , respectively , to the two operators whose position - space representations appear on the right hand side of eq .
. we will refer to @xmath54 as the smoluchowski operator , which governs the evolution of the spin population under an arbitrary potential @xmath50 .
in appendix [ app : mcf ] , we show that the signal ends up being expressable in the eigenbasis of the smoluchowski operator as the matrix element of the propagator corresponding to the stationary eigenvalue : denoting the left and right eigenvectors with @xmath56 and @xmath57 , the signal has the form @xmath58 } \ket{u_0 } \ , \label{eq : e.mcf.2}\end{aligned}\ ] ] where the prime on the product indicates time ordering ( i.e. , the earliest factor acts first ) , and @xmath59 is to be understood as a time point representative of each time interval . with this perspective , the mcf approach relies on using the eigenbasis of the smoluchowski operator to carry out the matrix operations involved in evaluating the propagator and thereby the signal .
the two operators @xmath54 and @xmath60 are expressed in the eigenbasis of @xmath54 , which is done analytically in appendix [ app : mcf ] for the case of a hookean potential .
the remaining operations of exponentiation and matrix product are eventually carried out by software according to the waveforms we consider in the results section . note that while a general time dependence requires a continuum limit ( @xmath61 ) in eq .
, piecewise - constant waveforms such as those of pulsed - field - gradient experiments do not .
the latter situation lends itself readily to a product of constant exponential operators over the plateaus of the waveform , without any discretization error . on the other hand , the case of a general time dependence
is treated as a controlled approximation based on the size of @xmath62 .
another inevitable source of numerical error is the need to truncate the matrix representations , which are in principle infinite .
however , one simply has to include all eigenfunctions whose time scales of decay ( i.e. , inverses of their eigenvalues ) are significant compared to the time step @xmath62 .
it is illuminating to consider the bts equation as the projection of an operator equation for the vector @xmath211 onto the ( dual ) vector @xmath212 . if one defines the operators @xmath54 and @xmath55 such that @xmath213 eq .
may be rewriten , without the projection onto spatial coordinates , as @xmath214 \ket{{m}(t ) } \;. \label{eq : bt.sym}\end{aligned}\ ] ] we will refer to the operator @xmath54 defined in eq . as the smoluchowski operator . in the mcf framework
, the total time interval is sliced into smaller successive intervals @xmath215 with @xmath216 . within each of these intervals , the gradient waveform @xmath217 , hence the operator @xmath55 ,
is considered to be constant .
if the operator @xmath55 has the constant value @xmath218 in the interval @xmath219 , that is , @xmath220 then eq . implies that @xmath221 where @xmath222 is the duration of the @xmath126th time interval .
thus , when all @xmath122 intervals have elapsed , @xmath223 where the prime on the product symbol is to remind us that subsequent propagators multiply from the _ left_. finally , the signal becomes @xmath224 this is the essence of the mcf formalism .
what remains are details of implementation and evaluation , which vary according to the specifications of the problem at hand .
the expression derived above becomes increasingly accurate for an arbitrary gradient waveform @xmath225 with finer time - slicing .
or , it may be considered exact if the gradient waveform is as a matter of fact constant over finite time intervals .
the first step toward the evaluation of eq .
is to make the expression concrete by choosing a basis to project the operators and vectors onto .
the two most obvious choices are the eigenvectors of the operators @xmath54 and @xmath55 .
the eigenvectors of @xmath55 , according to eq .
, are the same set @xmath226 of vectors as the eigenvectors of the position operator . using
this basis will result in a path integral type implementation @xcite .
alternatively , one projects onto the eigenvectors of the smoluchowski operator @xmath54 , but there is a catch : the operator @xmath54 is not self - adjoint , thus a spectral decomposition may not be available .
however , it is similar to a self - adjoint operator in the sense that the operator @xmath227 satisfies @xmath228 .
is the potential operator , such that @xmath229 .
if one considers the self - adjoint wave vector ( or momentum ) operator @xmath230 such that @xmath231 , the operator @xmath54 given in eq . can be written as @xmath232 .
one can verify directly that @xmath233 satisfies @xmath234 .
] this allows @xmath54 to have a real spectrum and a complete set of bi - orthogonal eigenvectors , as we will briefly discuss below .
if the self - adjoint operator @xmath235 has a set of real eigenvalues @xmath236 and corresponding eigenvectors @xmath237 satisfying @xmath238 it can be verified via straightforward algebraic manipulations that the vector @xmath239 satisfies @xmath240 i.e. , @xmath57 is an eigenvector of @xmath54 with eigenvalue @xmath236 .
similarly , the vector @xmath241 satisfies the eigenvalue equation @xmath242 it is these left and right eigenvectors , taken in pairs , which satisfy relations of orthonormality @xmath243 and completeness @xmath244 as a result of the orthonormality and completeness of the eigenvectors @xmath237 of the self - adjoint operator @xmath235 .
when no perturbation via the gradient @xmath0 is applied , the bts equation must lead to the equilibrium magnetization density @xmath246 in the @xmath112 limit .
such stability of solutions requires the eigenvalues @xmath236 to be bounded from above : @xmath247 .
the eigenfunction corresponding to the maximal eigenvalue @xmath248 , therefore , is ( proportional to ) the equilibrium ( @xmath150 ) solution , @xmath249 with @xmath250 being a normalization factor given as @xmath251 also note that , according to eq .
, we have @xmath252 for the dual eigenfunction corresponding to the equilibrium eigenvalue @xmath253 .
we can exploit the properties mentioned above of the equilibrium eigenfunctions to simplify eq . .
first , if the initial magnetization density is at equilibrium , then @xmath254 , meaning @xmath255 .
second , we can ( ab)use eq . to the effect ,
@xmath256 these two manipulations eliminate both summations , and the integral in eq . , to yield @xmath257
the above analysis shows that the signal can be written as the top matrix element of the propagator , expressed in the eigenbasis ( and the dual basis ) of the smoluchowski operator . while it may be possible to evaluate this matrix element analytically , it is usually necessary to resort to numerical computations , both to compute the product , and also the exponentials . to this end ,
one uses the matrix elements @xmath258 and , using eq . , @xmath259 where the latter matrix element of the position operator may be computed , for instance , through @xmath260 now
, we will restrict to one spatial dimension , and the dimensionless potential @xmath261 . in this case
, the bts equation reads @xmath262 the eigenvalues and eigenfunctions of the smoluchowski operator @xmath54 can be found from @xmath263 to be @xmath264 and @xmath265 where @xmath266 is a hermite polynomial .
recall , also , from eq .
that @xmath267 .
immediately , one can evaluate the matrix element as @xmath268 for @xmath269 ( @xmath270 otherwise ) @xcite .
the matrix element , on the other hand , is simply @xmath271 one can then employ , for a given gradient waveform @xmath225 , a computer to use these matrix elements for @xmath272 the computation of the exponentials and the product in the expression for the nmr signal . in the case of a three dimensional anisotropic potential , written as @xmath273 in index notation
, the bts equation becomes @xmath274 observing the middle term on the right hand side , one sees that performing a rotation to diagonalize the tensor @xmath12 will decouple the equations over each rotated coordinate ( @xmath275 as in sec .
[ sec:3dpath ] ) .
the eigenfunctions hence factorize , @xmath276 with @xmath277 as given in eq . , while the eigenvalues of the entire equation become @xmath278 where @xmath119 are the eigenvalues of the confinement tensor ( note the vector index @xmath279 and the sum implied over @xmath165 ) .
the matrix @xmath280 is just made up of the eigenvalues above .
the orientation of the potential ( hence the rotation matrix @xmath192 ) is not known prior to the measurement of the specimen
. therefore these results are not as general as those of sec .
[ sec:3dpath ] .
the derivation of such general expressions seems intractable to us .
however , this does not render the above formulas useless .
the `` natural '' frame of reference , hence the matrix @xmath192 , established by the orientation of the specimen can be estimated from the data set , e.g. , via employing the diffusion tensor model .
the gradient waveform can be transformed into this natural reference frame in which @xmath13 is diagonal , and the above formulation can be used subsequently @xcite .
to provide a validation of the theoretical developments described above , we simulated random walks taking place under a hookean potential and computed the phase accumulated by each random walker , which was subsequently used in the estimation of the nmr signal . to this end , we performed biased bernoulli walk simulations using python .
for the purposes of the present article , we have employed these simulations only in one spatial dimension .
briefly put , each particle s motion is generated based on bernoulli trials performed at uniform time intervals @xmath63 .
the trials are biased in the sense that the probabilities @xmath64 and @xmath65 of changing the particle s coordinate by @xmath66 or @xmath67 , respectively , are not equal due to the presence of a force . the desired diffusive character corresponding to a bare diffusion constant @xmath10of a random walk generated as such is met by the condition @xmath68 on the other hand , if the drift velocity @xmath69 of the random walker is supposed to match the drift velocity @xmath70 of a brownian mover under a force @xmath71 , then the bias must be @xmath72 the conditions in and fix two of the simulation parameters @xmath73 , @xmath66 , and @xmath63 . as a final condition , one may set the resolution of the simulations by choosing @xmath63 to be smaller than the thermalization time scale @xmath74 of eq .
( [ eq : oss]a ) by a desired amount .
generating random particle trajectories according to the scheme described above , at each step we can compute the phase that would be accumulated due to a magnetic field gradient , and hence `` measure '' the nmr signal resulting from a large number of trajectories .
the simulated signal values which will be presented in the next section have been computed thusly .
here , we present and compare results obtained from ( i ) the application of the analytical expressions of sec .
[ sec : pathint ] , ( ii ) the numerical implementation of the mcf scheme of sec .
[ sec : mcf ] , and ( iii ) simulated data as described in sec .
[ sec : sim ] . to this end
, we consider the same two waveforms discussed in sec .
[ sec : explicitresults ] and illustrated in figure [ fig : waveforms ] .
the first diagram in fig .
[ fig : stejskaltanner_delta ] depicts the nmr signal as a function of pulse separation @xmath20 at fixed @xmath75 .
the pulse duration was @xmath76ms , and the value of @xmath21 was chosen so that the wavenumber is @xmath77mm@xmath78 .
the temperature and bulk diffusivity were taken to be @xmath79k , and @xmath80m@xmath81/ms , respectively .
the force constant was chosen such that @xmath82m@xmath83 ; through eq .
, this leads to an effective pore size @xmath84 m , which is roughly the size of a red blood cell .
the analytical result is due to the application of eq . , whereas the numerical implementation of the mcf signal is achieved via eqs . .
the random walk simulations were performed using a step size of @xmath85 m , and a time step of @xmath86s .
a total of 2 million trajectories were generated .
the decay of the signal with increasing diffusion time is seen to be captured faithfully by the mcf implementation and random walk simulations as well . in fig .
[ fig : stejskaltanner_delta](bottom ) , we show the absolute error of mcf and the simulations .
the mcf implementation is error - free in the pfg scheme , since the pulse sequence _ is _ actually piecewise constant .
we show the results for the oscillating waveform in figure [ fig : ogse ] .
the same values for the temperature , bulk diffusivity , and the force constant , hence the effective pore size , were used .
the phase shift of the waveform was taken to be @xmath87 , i.e. , @xmath88 with @xmath89t / m . in these simulations , the total duration of the waveform was fixed at @xmath90ms for all frequencies .
frequency was varied by changing the number of full periods ( @xmath41 ) within this timeframe .
so , @xmath91 corresponds to the lowest frequency , while the subsequent points were generated by setting @xmath92 the mcf results were obtained by setting the time interval to @xmath93s for all frequencies .
the random walk simulations were performed in the same way as for the case of the stejskal - tanner sequence .
the frequency dependence of the signal seems to be consistent in all three methods .
the errors in the random walk and mcf results are rather small as shown in the bottom panel of figure [ fig : ogse ] .
unlike before though , the mcf results do exhibit some deviation from the analytical results . at the low end of the frequency range , errors of the mcf scheme approach zero , since its staircase approximation of a slowly - varying waveform is sufficiently faithful . as the waveform explores increasingly higher frequencies ,
the approximation is seen to worsen until the period of the waveform reaches the thermalization timescale @xmath94ms . beyond this neighborhood of frequencies ,
less and less time remains for appreciable diffusion to take place in each period of the gradient waveform . since the sensitivity of the signal value to the magnetic field inhomogeneity stems from diffusion , this suppresses the influence of any misrepresentation of the precise gradient waveform , appearing as a tapering - off of the error with increasing frequency in spite of the worsening approximation of the mcf scheme : as long as the mcf time discretization scale is chosen safely below the thermalization time @xmath95 , the expected increase of error with increasing frequency due to time - discretization is tamed .
the random walk simulations offered yet another test for the validity of our result for the oscillating gradient waveforms .
for these simulations , the suppression of errors with increasing frequency has to do with statistics . at low frequencies ,
the expected value of the average magnetization is zero , implying that the spin - bearers have a wide distribution of accumulated phase angles , in order to have resulted in a lot of mutual cancellation . on the other hand ,
the expected magnetization is maximized at high frequencies , implying a sharp distribution of phase angles . in the extreme case of an infinitely sharp distribution
, one can imagine that even a single walker s accumulated phase angle will almost certainly match the expected value of the distribution .
therefore the lower frequency simulations ( with lower signal value ) require more realizations ( walkers , trajectories ) to sample the phase distribution adequately .
conversely , at a fixed number of realizations , higher frequency simulations ( with higher signal values ) have better sampling and hence better likelihood of small error . incorporating a tensorial force constant ( while keeping the diffusivity a scalar ) leads to a characterization of anisotropy alternative to the commonly employed diffusion tensor model . here , we shall focus on the explicit results for the stejskal - tanner sequence , i.e. , eqs . and .
the corresponding expression for the diffusion tensor model is @xcite @xmath96 \label{eq : e_dti } \ , \end{aligned}\ ] ] where @xmath97 is the diffusion tensor .
the diffusivity and temperature were as before while the confinement tensor was taken to be diagonal with eigenvalues chosen such that @xmath98 corresponded to a pore size of approximately @xmath99 m while @xmath100 was taken to be one - tenth of @xmath101 so that the corresponding pore size along the @xmath102 axis was roughly @xmath103 m .
we chose the eigenvalues of the diffusion tensor through the expression @xmath104 which was obtained by setting the mean squared displacements , @xmath105 , implied by the two models equal to each other ( see eq . ) .
different colors in figure [ fig : stejskaltanner_anisotropy ] represent different values for the angle between the gradient direction and the `` fiber , '' i.e. , the @xmath102-axis . in the first panel of figure [ fig : stejskaltanner_anisotropy ] , we plot the predicted nmr signal against @xmath106 .
the timing parameters were : @xmath107ms and @xmath108ms .
as expected , the @xmath106 dependence of the signal appears linear for both models on a semi - logarithmic plot .
the two models yield essentially the same results ; the slight deviation is due to the pulse duration .
the difference of the models becomes apparent when one considers the dependence of the signal on the timing parameters of the sequence . to illustrate this
, we depict the @xmath7 vs. @xmath22 curves in the second panel of figure [ fig : stejskaltanner_anisotropy ] . in these plots
, the @xmath109 values were fixed to @xmath110rad / mm and the pulse separation was again @xmath107ms . in this case , the diffusion tensor model implies a linear appearance in the semi - logarithmic plots
. however , the dependence in the confinement tensor model is more complicated with non - linearity becoming quite substantial in the most restricted directions .
parabolic potentials are useful in numerous physics problems when one is interested in modeling slight deviations away from the equilibrium state ; when the perturbation is small , any potential can be approximated by a quadratic function . in this study , however , the utility of employing a parabolic potential model was due to a different reason .
following previous work @xcite , our development can be regarded to yield an alternative model of restricted diffusion , which was the main motivation of this study .
its resemblance to restricted diffusion is due to the time - dependence of the underlying mean square displacements ( hence the diffusion coefficients ) .
the mean square displacement implied by the propagator in is @xmath111 , which approaches a plateau as @xmath112 as expected in restricted geometries . despite this resemblance ,
the signal implied by the parabolic potential model has an important difference from that for restricted diffusion models .
the phases of the random walkers turn out to be gaussian - distributed @xcite .
thus , the logarithm of the predicted signal for each isolated pore has only a quadratic term in gradient magnitude .
consequently , features that are visible only at large diffusion sensitivities ( e.g. , diffraction - like effects @xcite ) are not accomodated by the parabolic potential model .
although this seems like a limitation of the model , for most acquisitions involving microscopic pores and weak gradients , the low diffusion sensitivity behavior is the only regime of the signal that can be measured . under these conditions ,
the observed higher order terms emerge predominantly from heterogeneities within the sample rather than true compartmental effects .
heterogeneity - induced effects would be captured by the present model as well if the signal is represented as the superposition of signals associated with a distribution of confinement values , i.e. , potentials , similar to what is done in multiple diffusion tensor @xcite and diffusion tensor distribution @xcite representations . to illustrate the above points and better assess the resemblance of the hookean model to restricted diffusion , we consider molecules diffusing between two infinite parallel plates separated by @xmath113 m .
for this geometry , the potential is zero within the slab and infinite at the walls .
figure [ fig : compare2slab]a depicts the potential profile for the slab geometry ( continuous line ) as well as the parabolic potential ( dashed line ) that could be used in lieu of it . for the latter
, the confinement value was obtained through as 0.33 @xmath114m@xmath83 . in figure
[ fig : compare2slab]b , we plot the signal against the quantity @xmath115 , which is frequently referred to as the `` b - value , '' and fully characterizes the signal decay for free diffusion @xcite . here , curves in different colors represent different values of the pulse duration . for each pulse duration ,
the @xmath116-value was varied by changing the gradient magnitude .
having different signal values for the same @xmath116-value is a further manifestation of the failure of the free diffusion model in characterizing restricted diffusion . employing the parabolic potential
however , yields a behavior similar to that observed for restricted diffusion .
we observe that the deviation between the signal values for restricted and hookean models becomes more pronounced as the @xmath116-value is increased .
we show the dependence of the signal on the separation of pulses in figure [ fig : compare2slab]c . to be able to simulate shorter diffusion times , the pulse duration was fixed at @xmath117s .
here , different colors represent different values of the wavenumber @xmath109 .
as expected , the signals for both problems reach constant values as the pulse separation ( diffusion time ) is prolonged .
the curves for the two models are nearly indistinguishable at small attenuations . for stronger diffusion sensitivity ,
the transition to the long diffusion - time regime occurs slightly more quickly in the case of parabolic potential , yielding a larger eventual signal .
we reiterate that the comparisons here are based on the association in , which is only one way of assigning a pore size to a particular confinement value .
many more associations could be made ( e.g. , based on zero displacement probabilities ) that would change the discrepancies between the curves in figure [ fig : compare2slab ] though the qualitative resemblance afforded by the parabolic potential model would prevail .
m. dashed curve : parabolic potential that could be employed for the same slab geometry .
( b ) the signal for the traditional stejskal - tanner sequence plotted against the @xmath116-value for @xmath118 .
( c ) the signal plotted against the pulse separation for the stejskal - tanner sequence .
, width=317 ] the first approach we employed to tackle the problem involved posing the nmr signal as a path integral and evaluating it explicitly . as exemplified by the cases of stejskal - tanner and oscillating gradient sequences , this approach can be employed to obtain analytical results for the nmr signal featuring arbitrary gradient waveforms .
formerly , the path integral method was used to derive the expression describing the influence of free diffusion @xcite . in an earlier publication ,
the quadratic term of the signal decay due to restricted diffusion was calculated via an explicit evaluation of the path integral for the simple geometries of parallel plates , cylinders , and spheres @xcite .
employing the former two geometries in perpendicular directions , the solution for a capped cylinder geometry can be obtained @xcite , which had been used to model restricted diffusion in an anisotropic pore . however , because the capped cylinder is invariant to rotations about its axis , this model has only two size parameters its radius and length , thus limiting its applicability . on the other hand , our model has three unique confinement values ( @xmath119 , where @xmath120 ) , each of which can be related to an effective pore size along three spatial directions . in this respect ,
our model is more general than the capped cylinder geometry for modeling anisotropy .
we note that an alternative geometry involving three size parameters is the rectangular prism .
the mr signal influenced by diffusion within a rectangular prism could be obtained by employing the solution for parallel plates in three orthogonal directions . however , even when the three size parameters are set equal , the resulting geometry is not truly isotropic , making it difficult to isolate the influence of structural anisotropy .
this limitation , which is present also for the capped cylinder geometry , is overcome by our method .
setting all three principal confinement values equal to each other provides the solution for a perfectly isotropic potential akin to the case of diffusion taking place within a sphere . on the other hand , setting only two of the three eigenvalues of the confinement tensor equal to each other provides the solution for diffusion within a cylinder - like pore .
as exemplified in figure [ fig : stejskaltanner_anisotropy ] , compared to the diffusion tensor model , which is based on the assumption of free diffusion , our model has a different dependence on the timing parameters of the gradient waveforms .
this issue is likely to be important when sophisticated pulse sequences are employed . indeed
, several different pulse sequences have been studied in recent years employing pulsed @xcite and continuous @xcite waveforms . especially in the context of modeling local diffusion anisotropy , one is faced with the choice between the gaussian diffusion tensor model @xcite , and restricted capped cylinder model @xcite . although , the restricted character of the model is desirable , the capped cylinder model had two limitations , which may have prompted some studies to adopt the diffusion tensor model instead : ( i ) it has only two size parameters , so is more limited than the diffusion tensor model , ( ii ) the solutions are rather complicated .
we believe the confinement tensor model introduced in this study overcomes both of these limitations making it suitable to characterize global ( macroscopic ) as well as local ( microscopic ) diffusion in heterogeneous media .
for example , `` diffusion imaging with confinement tensor ( dict ) '' could be performed by estimating the confinement tensor via eqs . and
for each voxel of a data set comprising multi - directional diffusion - weighted images , much like eq .
is employed in diffusion tensor imaging ( dti ) studies .
although our formulation has focused for the most part on parabolic potentials and linear gradients , the mcf and random walk simulations can be generalized to more general cases .
for example , spatially nonlinear gradients can be incorporated with ease by simple manipulations of the operators employed in the mcf technique @xcite .
other potentials can also be incorporated by employing exact or approximate solutions of the corresponding helmholtz problem .
finally , the work presented in this article may be relevant in very different contexts .
for example , our approach here could be expanded to offer some utility in modeling noise induced by brownian motion in magnetic resonance force microscopy @xcite . in conclusion , we have presented an in - depth study on the effect of diffusion taking place in the presence of a parabolic potential on the nmr signal .
the scheme enables one to incorporate restriction - induced effects when general gradient waveforms are employed .
the study lends itself to a new formulation for diffusion anisotropy featuring three parameters like the diffusion tensor model .
however , by overcoming the limitations due to the assumption of free diffusion , the confinement tensor model could be considered a powerful alternative to the diffusion tensor model for characterizing anisotropy .
we thank dr .
chunlei liu for his comments on the nomenclature .
the authors acknowledge funding from tbitak - eu cofund project 114c015 , boazii university research fund ( grant number 8521 ) , bilim akademisi bagep 2014 award , nih r01mh074794 , and swedish foundation for strategic research ( no .
am13 - 0090 ) .
this appendix is reserved for the details of the derivation involved in arriving at the analytical expression for the average magnetization , or nmr signal .
we will begin with the discretization of the path integral , followed by its evaluation ( and continuum limit ) for a parabolic potential in one spatial dimension , and then generalize to higher dimensions . as is usual ,
the path integral is evaluated via slicing the time interval @xmath121 into a number @xmath122 of time steps , and before the continuum limit ( @xmath123 ) is considered , discretized versions of the integration measure and the integrand are employed . for what follows , we define the small time step @xmath124 , and @xmath59 is replaced by @xmath125 for the discretization , with @xmath126 an integer in the interval @xmath127 $ ] . also , time integrals @xmath128 become @xmath129 . following the definition of the variables @xmath130 and
@xmath131 , the discretization amounts simply to where @xmath133 is the initial probability distribution of spins , and @xmath134 is the probability of a particle ending up at the _ later _ position @xmath135 given that it was at the _ earlier _ position @xmath136 a time @xmath59 ago .. ] as a result , the time - sliced path integral becomes @xmath137 where the subscript @xmath63 reminds us that this is not yet the exact path integral , but a discretized approximation .
also , note how the integrals over later positions are nested inside those of the earlier ones .
the above discussion holds for any ( time - independent ) potential and number of dimensions .
however , evaluating the ( discretized ) path integral requires an explicit form for the _ propagator _ @xmath138 .
we focus first on the one dimensional case with a parabolic potential .
the propagator @xmath139 for diffusion under the effect of a dimensionless potential @xmath140 has the form @xcite
@xmath141 \label{eq : propagator } \ .\end{aligned}\ ] ] with @xmath10 being the bare diffusion constant , the inverse time @xmath142 is defined as note that the propagator is a normalized gaussian in its later argument @xmath148 , centered at a point that depends on its earlier argument @xmath149 . despite its gaussian appearance , this propagator has an important characteristic that is consistent with restricted diffusion . to illustrate this point
, we shall consider the long time limit ( @xmath150 ) of the propagator .
in this regime , the dependence on the initial position disappears as expected for restricted diffusion .
it is instructive to also consider the net displacements , i.e. , @xmath151 . let @xmath152 denote the equilibrium spin density .
the second moment of net displacements , defined through the expression @xmath153 diverges for free diffusion as the diffusion time is prolonged .
however , the same quantity for the propagator in eq . asymptotically approaches the constant value of @xmath154 . on the other hand ,
if one considers a restricted diffusion process taking place between two infinite plates separated by a distance @xmath155 , this quantity similarly approaches a constant value , which is @xmath156 .
setting the latter two quantities equal to each other , one can obtain an effective pore size given by @xmath157 we refer to this relationship in the results section .
the form of the propagator helps a great deal in evaluating the signal analytically .
indeed , the discretized path integral involves only familiar integrals , but each integration is affected by the result of that nested immediately inside it .
consider , for instance , the inner - most integral ( over @xmath158 ) : @xmath159 the first factor has nothing to do with the remaining integrals in eq . .
the second factor , on the other hand , combines with the exponential in the next integral ( over @xmath160 ) , shifting its `` wave number '' @xmath161 by an amount @xmath162 .
one can see that , with later ( large index ) wave numbers leaking into the phase factors of earlier ( smaller index ) integrations as such , and getting multiplied by @xmath163 at each step , it should be useful to define @xmath164 ( the index @xmath165 is not to be confused with the imaginary unit @xmath166 . ) following this recursion just described , one finds @xmath167 we assume that the distribution of spin - carriers has had enough time to reach equilibrium before the gradient sequence is applied .
hence , the relation @xmath168 can be employed ( eqs .
in the limit @xmath169 ) , with @xmath170 being some arbitrary pre - initial position
. we can thus perform the last remaining integration similarly to the previous ones to find @xmath171 in the continuum limit where @xmath172 as @xmath123 , we have @xmath173 and @xmath174 .
recalling moreover that @xmath175 really stands for @xmath176 , the @xmath177 limit of the average magnetization is easily obtained as @xmath178 the quantity @xmath179 follows from eq . similarly . recalling that @xmath180 , and using eq .
, the expression for @xmath175 can be recast as @xmath181 in the @xmath177 limit , one finds @xmath182 eqs . and yield the final result for the nmr signal for the case of molecules diffusing under the influence of a hookean restoring force in one dimension . here
, we show that our expression for the signal agrees with previous results for free diffusion nmr signal @xcite @xmath183 \label{eq : e.free}\end{aligned}\ ] ] in the limit of vanishing potential ( @xmath184 ) . in this limit , @xmath37 , and thus the exponential in eq . drops . along with the gradient echo condition , @xmath185
, we have @xmath186 and @xmath187 .
therefore , the first term in our expression for the signal matches the free diffusion nmr signal . the second term in eq .
does not seem to disappear , due to the @xmath188 prefactor . however , if one recalls that this term is simply the remaining integral in eq . , then it is recognized that the integral becomes @xmath189 when @xmath190 , which is unity by definition . in three dimensions ,
the most general harmonic potential ( assuming the attraction center is at the origin ) is of the form @xmath191 where the confinement tensor @xmath12 is real and can be assumed to be symmetric without loss of generality .
there exists , then , a rotation matrix @xmath192 such that @xmath193 , and @xmath194 where @xmath195 is diagonal , and @xmath196 . in this basis denoted by tildes
, the potential has the coordinates @xmath197 , @xmath198 , and @xmath199 decoupled , and so does the smoluchowski equation that describes the propagator @xmath200 .
therefore , the propagator simply factorizes into three instances of the propagator for each direction : @xmath201 \ , \label{eq : propagator.3}\end{aligned}\ ] ] where eqs .
become by virtue of this factorization , one sees that the evaluation of the path integral proceeds exactly the same way as before , only this time it is three - fold : @xmath203 with @xmath204 recall that the two equations above are valid in the rotated basis where the confinement tensor @xmath12 is diagonal ( hence the tildes ) . in order to revert to the lab frame ( the un - rotated basis ) , note that @xmath205 and @xmath206 are just the @xmath165th eigenvalues of the matrices @xmath207 and @xmath208 which are diagonalized by the same rotation @xmath192 as @xmath30 , and hence @xmath12see eq . .
exploiting as well the rotation properties of @xmath192 , one finds @xmath209 where @xmath210 these are eqs . and of the main text .
since one in general does not know what the principal directions are without previous knowledge about the specimen , these general expressions are more relevant than the diagonalized versions and .
however , note that in evaluating the integral in eq . with a given gradient waveform @xmath0 , one may have to make sense of the exponentiated matrix in the integrand by going back and forth between the lab frame and the eigen - basis . | we study the influence of diffusion on nmr experiments when the molecules undergo random motion under the influence of a force field , and place special emphasis on parabolic ( hookean ) potentials . to this end , the problem is studied using path integral methods .
explicit relationships are derived for commonly employed gradient waveforms involving pulsed and oscillating gradients .
the bloch - torrey equation , describing the temporal evolution of magnetization , is modified by incorporating potentials . a general solution to this equation
is obtained for the case of parabolic potential by adopting the multiple correlation function ( mcf ) formalism , which has been used in the past to quantify the effects of restricted diffusion .
both analytical and mcf results were found to be in agreement with random walk simulations .
a multi - dimensional formulation of the problem is introduced that leads to a new characterization of diffusion anisotropy .
unlike for the case of traditional methods that employ a diffusion tensor , anisotropy originates from the tensorial force constant , and bulk diffusivity is retained in the formulation .
our findings suggest that some features of the nmr signal that have traditionally been attributed to restricted diffusion are accommodated by the hookean model . under certain conditions ,
the formalism can be envisioned to provide a viable approximation to the mathematically more challenging restricted diffusion problems . |
understanding the process of high - mass star formation represents a challenge from both a theoretical and observational point of view . by now
, three main competing concepts of massive star formation have been discussed in the recent literature ( see * ? ? ? * and references therein ) : formation through coalescence of lower - mass stars in extremely dense stellar clusters @xcite ; monolithic collapse of isolated , turbulence supported cores @xcite ; and competitive accretion where the most massive protostars benefit of the gravitational well created by the lower - mass stars surrounding them , thus boosting the accretion rate ( e.g. , * ? ? ? * ) .
the available observational evidence is still inadequate to discern between these different scenarios , mostly because observations of massive young stellar objects ( mysos ) are complicated by the large distances of typically several kpc , implying small angular scales , and the fact that such objects are born still deeply enshrouded in dense dusty envelopes , opaque to optical and near - infrared wavelength radiation . a possible way to distinguish between different models is to explore the physical conditions and the dynamical processes of the gas close to the protostars .
however , present ( sub)millimeter interferometers and infrared ( ir ) telescopes attain angular resolutions @xmath3025 , corresponding to thousands of astronomical units ( au ) at the typical distances of mysos of several kpc .
this is insufficient to investigate the kinematics of the gas in the immediate surroundings of the forming star , where crucial phenomena such as accretion , rotation , and ejection are likely most prominent .
the above mentioned limitation can be bypassed by very long baseline interferometry ( vlbi ) observations of maser lines , which attain linear resolutions of a few au .
a few years ago , we started an observational project to study the high - mass star - forming process by comparing interferometric images of thermal lines of molecular tracers with multi - epoch vlbi studies in three well known maser species ( oh , and ) . at present , three sources ( iras20126@xmath44104 , and ) have been already analyzed @xcite .
the results demonstrate that the synergy between vlbi , multi - species maser observations and ( sub)mm interferometric observations of thermal molecular lines is crucial to achieve a multi - scale picture of the environment of newly formed massive ( proto)stars . in the present article we report on our vlbi maser study of the high - mass star - forming region ( hmsfr ) , and compare our results with available complementary interferometric data of continuum and thermally excited molecular line emission , as well as with sub - arcsec resolution mid - ir images obtained with the cooled mid - infrared camera and spectrometer ( comics ) on the subaru telescope .
in section 2 , we provide a review of previous observations toward this region .
section 3 describes the vlbi observations of the maser transitions , the very large array ( vla ) observations of the radio continuum emission at 1.3 cm and 3.6 cm , and the mid - ir observations performed with the subaru / comics telescope . in section 4 , we present the distributions of the three types of masers , the water maser proper motions , and the properties of the radio continuum . the results are discussed in section 5 and the conclusions are drawn in section 6 .
the hmsfr ( alias iras18411@xmath50338 ) is located at a kinematic distance of 7.4 kpc6 kpc ) and the far ( @xmath69 kpc ) kinematical distance is not resolved @xcite , we adopt the distance to the tangent point . ]
@xcite and is associated with a pc - scale clump with a mass in excess of 10@xmath7 and a bolometric luminosity of @xmath02@xmath110@xmath8 @xcite . from the observation of nh@xmath9
, the systemic velocity of the region ( @xmath10 ) with respect to the local standard of rest ( lsr ) is found to be 101.0 @xcite . on a large scale , mid - ir observations at 8 , 10.5 , and 18.1
have resolved the iras source in two objects along the north - south direction ( * ? ? ?
* ; * ? ? ?
* figure 1 ) . both sources could be associated with an midcourse space experiment ( msx ) point source @xcite and a 2.2 source with strong near - infrared excess , lying 34 to the southeast of the northern , brighter mid - ir object @xcite .
bolometer observations at 1.2 mm @xcite have detected a peak of dust continuum emission associated with the two mid - ir sources and the near - ir source .
single - dish maps in the cs(54 ) and c@xmath11o(21 ) lines have revealed an arcmin - scale molecular clump and a monopolar ( red - shifted ) outflow is seen in the @xmath12co(21 ) line @xcite .
single - dish surveys have detected emission in dense gas tracers as cs(76 ) , cs(21 ) , and cs(10 ) @xcite , and in the h110@xmath13 radio recombination line @xcite , which suggests the presence of a diffuse region .
in addition , oh maser emission has been found in the main ( 1.665 and 1.667 ghz ) hyperfine structure lines and in the 1.720 ghz satellite line , whereas the 1.612 ghz oh satellite has been observed in absorption @xcite .
weak and strong maser emission have been reported at 6.7 ghz @xcite and 22.2 ghz @xcite , respectively .
interferometric observations at millimeter wavelengths have revealed multiple sources in this region ( * ? ? ?
* figure 1 ) and show that the dust continuum peak coincides with the position of the northern , stronger mid - ir source .
the mass ( 100 @xmath14 ) from the 3 mm dust continuum emission and the rotational temperature ( 93 k ) inferred from the ch@xmath15cn lines suggest that one is dealing with a hot molecular core ( hmc ) . in this study
, we will focus on this hmc , as it contains all the tracers of high - mass star formation that we have investigated with our observations , namely the masers and the free - free and mid - ir continuum sources .
as shown by @xcite and @xcite , the hmc is seen in nh@xmath15 , hco@xmath16 , and ch@xmath15cn line emission .
the observations of @xcite were not sensitive enough to detect continuum emission at 1.3 cm , whereas our new images demonstrate the existence of a free - free continuum source inside the hmc ( see below ) . a wide - angle , massive ( @xmath3100 @xmath14 )
bipolar outflow has been detected in the @xmath17co(10 ) line , oriented perpendicular to a ch@xmath15cn velocity gradient detected in the hmc @xcite .
the absolute positions of the 1.665 ghz oh and 22.2 ghz masers , derived by @xcite using the vla with an accuracy of 05 , indicate that these maser emissions are associated with the hmc .
all these features are consistent with the presence of a deeply embedded myso(s ) inside the hmc , possibly accompanied by lower mass objects lying in the neighborhoods of the hmc itself .
we observed the hmsfr ( tracking center : r.a.(j2000 ) = @xmath18 and dec.(j2000 ) = @xmath19 ) with the vla at x ( 3.6 cm ) and k ( 1.3 cm ) bands in both the c - array ( in 2008 april ) and a - array configurations ( in 2008
november ) . at 3.6 cm in both configurations and at 1.3
cm in the c - array configuration , the continuum mode of the correlator was used , resulting in an effective bandwidth of 172 mhz . at 1.3 cm in the a - array we used mode `` 4 '' of the correlator , with a pair of intermediate frequency ( if ) bands , each with a width of 3.125 mhz ( 64 channels ) centered on the strongest maser line and a pair of 25 mhz wide if bands ( 8 channels each ) sufficiently offset from the maser lines to obtain a measurement of the continuum emission .
the two bands centered at the same frequency ( detecting both circular polarizations ) were averaged to decrease the noise . at x - band ,
3c286 ( 5.2 jy ) and 3c48 ( 3.1 jy ) were used as primary flux calibrators , while 1832@xmath5105 ( 1.4 jy ) was the phase calibrator . for the k - band observations , the primary flux calibrator was 3c286 ( 2.5 jy ) , the phase calibrator 1832@xmath5105 ( 1.0 jy ) , and the bandpass calibrator ( for the a - array only ) 1733@xmath5130 ( 3.7 jy ) .
the data were calibrated with the nrao astronomical image processing system ( aips ) software package using standard procedures . only for the a - array data at 1.3 cm ,
several cycles of self - calibration were applied to the strongest maser channel , and the resulting phase and amplitude corrections were eventually transferred to all the other line channels and to the k - band continuum data @xcite .
this procedure resulted in a significant ( at least a factor 2 ) improvement in the signal - to - noise ratio ( snr ) .
natural - weighted maps were made with task imagr of aips for both the continuum and the line data .
simultaneous observation of the line and continuum emission at k - band made it possible to obtain the relative position of the continuum image with respect to the maser spots with great precision ( @xmath001 ) .
finally , we identified the maser spots that appear in both our very long baseline array ( vlba ) and vla images and , using the accurate absolute position information obtained from the vlba observations , we re - centered the vla spots and continuum image .
the correction thus applied turns out to be @xmath2020 mas in each coordinate .
we conclude that the absolute astrometrical precision of the continuum maps presented in this paper is @xmath001 .
we conducted vlbi observations of the and masers ( at four epochs ) and of the oh maser ( at a single epoch ) toward at 22.2 ghz , 6.7 ghz , and 1.665 ghz , respectively .
we give more technical details on the observations in the following subsections . in order to determine the masers absolute positions
, we performed phase - referencing observations in fast switching mode between the maser source and the calibrator j1834@xmath50301 .
this calibrator has an angular offset from the maser source of 24 and belongs to the list of sources of the vlba calibrator survey 3 ( vcs3 ) .
its absolute position is known with a precision better than @xmath211.5 mas , and its flux measured with the vlba at s and x bands is 154 and 156 mjy beam@xmath22 , respectively .
five fringe finders ( j1642@xmath43948 ; j1751@xmath40939 ; j1800@xmath43848 ; j2101@xmath40341 ; j2253@xmath41608 ) were observed for bandpass , single - band delay , and instrumental phase - offset calibration .
the data were reduced with aips following the vlbi spectral line procedures . for a description of the general data calibration and the criteria used to identify individual masing clouds , derive their parameters ( position , intensity , flux and size ) , and measure their ( relative and absolute ) proper motions , we refer to the recent paper on vlbi observations of and masers by @xcite . for each vlbi epoch ,
maser absolute positions have been derived by employing two methods , direct phase - referencing and inverse phase - referencing , and the two procedures always gave consistent results . and
oh maser absolute positions have been derived only by using direct phase - referencing .
the derived absolute proper motions have been corrected for the apparent proper motion due to the earth s orbit around the sun ( parallax ) , the solar motion and the differential galactic rotation between our lsr and that of the maser source .
we have adopted a flat galactic rotation curve ( @xmath23 kpc , @xmath24 ) @xcite , and the solar motion of @xmath25 , @xmath26 , and @xmath27 by @xcite , who recently revised the hipparcos satellite results .
we observed the hmsfr ( tracking center : r.a.(j2000 ) = @xmath28 and dec.(j2000 ) = @xmath29 ) with the vlba in the @xmath30 transition ( rest frequency 22.235079 ghz ) .
the observations ( program code : bm244 ) consisted of 4 epochs : 2006 april 23 , 2006 june 30 , 2006 september 28 , and 2007 january 18 . during a run of about 6 h per epoch
, we recorded the dual circular polarization with 16 mhz bandwidth centered at an lsr velocity ( @xmath31 ) of 101.0 .
the data were processed with the vlba fx correlator in socorro ( new mexico ) using an averaging time of 1 s and 1024 spectral channels .
images were produced with natural weighting , cleaned and restored with an elliptical gaussian with full width at half maximum ( fwhm ) size of about 1.3 mas @xmath1 0.6 mas at a position angle ( pa ) of @xmath54 ( east of north ) , with small variation from epoch to epoch ( shown in table [ tbl : mas - abs - pos ] ) .
the interferometer instantaneous field of view was limited to @xmath027 . in each observing epoch
the on - source integration time was @xmath02.5 h resulting in an effective rms noise level of the channel maps ( @xmath32 ) in the range 0.0060.015 jy beam@xmath22 .
the spectral resolution was 0.2 .
we observed ( tracking center : r.a.(j2000 ) = @xmath33 and dec.(j2000 ) = @xmath34 ) with the european vlbi network ( evn ) in the @xmath35 transition ( rest frequency 6.668519 ghz ) .
data were taken in 4 epochs ( program code : em061 and em069 ) separated by @xmath01 year : 2006 february 28 , 2007 march 18 , 2008 march 18 , and 2009 march 17 . in the first two epochs ,
the antennae involved in the observations were cambridge , jodrell2 , effelsberg , hartebeesthoek , medicina , noto , torun , and westerbork .
since we found out that the longest baselines involving the hartebeesthoek antenna ( e.g. , the ef - hh baseline is @xmath08042 km ) heavily resolve the maser emission and do not produce fringe - fit solutions , the hartebeesthoek antenna was replaced with the onsala antenna in the third epoch . during a run of @xmath06 h , we recorded the dual circular polarization in two if bands of width of 2 mhz and 16 mhz , both centered at an lsr velocity of 101.0 .
the 16 mhz band was used to increase the snr of the weak continuum calibrator .
the data were processed with the mkiv correlator at the joint institute for vlbi in europe ( jive - dwingeloo , the netherlands ) using an averaging time of 1 s and 1024 spectral channels for each observing bandwidth .
images were produced with natural weighting , cleaned and restored with an elliptical gaussian with fwhm size of about 12 mas @xmath1 6 mas at a pa of 21 , with small variations from epoch to epoch ( shown in table [ tbl : mas - abs - pos ] ) .
the interferometer instantaneous field of view was limited to @xmath092 . using an on - source integration time of @xmath02.2
h , the effective rms noise level of the channel map ( @xmath32 ) varied in the range 0.0050.01 jy beam@xmath22 . the 2 mhz band spectral resolution was 0.09 .
we observed ( tracking center : r.a.(j2000 ) = @xmath36 and dec.(j2000 ) = @xmath34 ) with the vlba in the @xmath37 oh transition ( rest frequency 1.665401 ghz ) on 2007 april 28 ( program code : bm244o ) . during a run of @xmath06 h , we recorded the dual circular polarization with two if bands of width of 1 mhz and 4 mhz , both centered at @xmath38 .
the 4 mhz bandwidth was used to increase the snr of the weak l - band signal of the continuum calibrator .
the data were processed with the vlba fx correlator in two correlation passes using 1024 and 128 spectral channels for the 1 mhz and 4 mhz bands , respectively . in each correlator
pass , the data averaging time was 2 s. images were produced with natural weighting , cleaned and restored with an elliptical gaussian with fwhm size of 17.8 mas @xmath1 10.9 mas at a pa of 1 .
the interferometer instantaneous field of view was limited to @xmath0185 .
the on - source integration time was @xmath01.9 h , resulting in an effective rms noise level in each velocity channel of @xmath00.02 jy beam@xmath22 .
the spectral resolution was 0.2 . using the mid - infrared imaging spectrometer ( comics ; * ? ? ?
* ) at the cassegrain focus of the 8.2 m subaru telescope , we carried out imaging observations of the 24.5 emission toward on 2008 july 15 .
for this purpose , we configured the q24.5 filter , and employed chopping mode for subtracting the sky - background emission .
the camera provides a field of view of @xmath39 with a pixel size of 013 .
flux calibration was performed towards four standard sources listed in @xcite : hd146051 , hd186791 , hd198542 , and hd198542 .
we estimated the overall uncertainty in the flux calibrations to be less than 10% .
astrometric calibration was performed by smoothing the 24.5 image to the same angular resolution as the 24 mipsgal ( survey of the inner galactic plane using mips ; * ? ? ?
* ) image , i.e. 6 , and overlaying the two images until a good match was found .
note that the heavy saturation of the mipsgal image does not hinder the comparison between it and the smoothed comics image . in practice ,
the emission in mipsgal is basically pointlike and the mips psf presents a well defined circular structure even far from the ( saturated ) peak .
the match between the smoothed comics image and the mipsgal image can thus be performed successfully on such circular features .
inspection by eye indicates a positional error of less than 1 , comparable to the astrometric accuracy of mipsgal .
this section reports our main results from the observation of the radio continuum , individual maser species and infrared emission towards the hmsfr . in the following ,
the term `` spot '' is used to refer to maser emission in a single velocity channel , whereas the term `` feature '' indicates a collection of spots emitting at a similar position over contiguous channels ( i.e. an individual masing - cloud ) . in this work , the parameters of maser emission ( position , intensity , proper motion , etc .. ) are derived for maser features .
our vla measurements of the 1.3 and 3.6 cm continuum emission towards ( in both the a- and c - array configurations ) are summarized in table [ tbl : vla - cm ] , and shown in figure [ g28-vla - cm ] . at the angular resolution of the vla c - array , the emission consists of two components , both detected at 1.3 and 3.6 cm .
the northern and southern components , separated by @xmath031 , are labeled , respectively , `` hmc '' and `` a '' in table [ tbl : vla - cm ] .
the former name is due to the source being associated with the hot molecular core imaged by @xcite .
source `` a '' is partly resolved with the c - array at 1.3 cm , and is completely resolved out with the a - array at both 1.3 and 3.6 cm . in the following ,
we focus our attention on the centimeter continuum source `` hmc '' , because this appears to be spatially associated with the observed 22 ghz , 6.7 ghz , and 1.665 ghz oh masers in ( see figures [ g28-vla - cm ] and [ g28-absp - l ] ) . while at 1.3 cm the most extended vla a - array configuration partly resolves the emission of `` hmc '' ( missing about half of the flux measured with the c - array ) , at 3.6 cm the emission is essentially compact in both configurations ( yielding comparable integral fluxes ) .
note however that the vla a - array image at 3.6 cm shows a spur protruding to the south .
the peak positions of the 1.3 and 3.6 cm emissions coincide within the uncertainties .
the absolute position of the a - array 1.3 cm image has been obtained by aligning positions of persistent water maser spots between the vla and vlba observations , and should be accurate to @xmath40 . figure [ g28-hmc - sed - cm ] shows the spectral energy distribution ( sed ) at centimeter wavelengths of `` hmc '' .
a spectral index ( @xmath13 ) of 0.6 is derived from a linear fit of the vla c - array 1.3 and 3.6 cm and b - array 6 cm flux densities .
the latter flux is an upper limit obtained from the gps 6 cm epoch 2 images downloadable from the magpis web site , which result to be the most sensitive among the other gps maps and the relevant map from the cornish survey ( see * ? ? ?
we also fit the sed at centimeter wavelengths with a model of a homogeneous region , with a lyman continuum rate of @xmath41 s@xmath22 and a radius of 004 .
we caution that the shape of the sed could be significantly affected by the different sensitivity to extended structures at different wavelengths . for the same vla configuration
, the shorter the wavelength , the larger can be the fraction of emission resolved out by the interferometer .
therefore , one should consider the possibility that the slope between 3.6 cm and 1.3 cm is steeper than 0.6 .
@xcite observed the hmsfr with the vla c - array ( hpbw of 22 @xmath1 16 ) and identified 13 water maser spots , mainly grouped in two clusters separated by @xmath002 in the northeast - southwest direction .
the absolute position of the brightest spot ( with a flux density of 61.3 jy at the v@xmath42 of 106.3 ) differs by @xmath001 from that of the corresponding spot in our vlba observation .
the field of view of our vlba observations covers the whole area within which water maser emission was detected by @xcite .
figure [ g28-maser - spe ] ( upper panel ) shows the total - power spectrum of the 22.2 ghz masers toward .
this profile was obtained by averaging the total - power spectra of all vlba antennae , after weighting each spectrum with the antenna system temperature ( @xmath43 ) .
water maser absolute positions and lsr velocities are plotted in figure [ g28-absp - l ] .
imaging the range of lsr velocities from 93.2 to 114.3 , we detected 168 distinct water maser features distributed over a region of about 03 @xmath1 05 .
most ( 92% ) of them concentrate in a small area of about 03 @xmath1 02 centered on the vla 1.3 cm continuum peak . the line - of - sight ( l.o.s . )
velocity distribution is remarkably bipolar , with red - shifted features located to the west of the radio continuum and blue - shifted ones to the east .
while the blue - shifted features mainly concentrate in two small ( size @xmath2020 mas ) clusters , the distribution of red - shifted masers consists of one narrow strip elongated east - west ( size of @xmath6100 mas ) , together with two diffuse `` clouds '' of features spread to the north and to the south of the densely populated , narrow strip .
it is also interesting to note the presence of two streamlines of water features ( enclosed in dashed boxes in figure [ g28-absp - l ] upper panel ) , oriented north - south and extending over a few tenths of an arcsecond , on either side of the vla 1.3 cm continuum .
we took particular care in filtering out spurious emission owing to poor cleaning of the beam side lobes ( elongated mostly along the north - south direction ) and can confirm that these water maser streamlines are real .
the individual feature properties are presented in table [ tbl : water ] .
the position offsets are relative to feature # 1 .
the absolute positions of feature # 1 at the four observing epochs are given in table [ tbl : mas - abs - pos ] .
feature intensities range from 0.1 to 54 jy . the spread in lsr velocities ranges from 111.99 for the most red - shifted feature ( # 107 ) to 93.33 for the most blue - shifted one ( # 155 ) .
only 61 features ( 36% of the total ) persisted over at least 3 epochs , 34 of which lasting 4 epochs .
we calculated the geometric center ( hereafter `` center of motion '' , identified with label # 0 in table [ tbl : water ] , and indicated by a star in figure [ g28-rel - pro - s ] ) of features with a stable spatial and spectral structure , persisting over the 4 observing epochs , and refer our measurement of proper motions to this point . with a time baseline of 9 months , relative ( sky - projected ) velocities of water maser features
are derived with a mean uncertainty of @xmath023% .
the magnitude of relative proper motions ranges from [email protected] to [email protected] , with a mean value of 24.0 .
figure [ g28-rel - pro - s ] shows the derived water maser proper motions . while the two blue - shifted clusters move to the east and northeast , most velocities of the red - shifted water masers point either to the south or to the west
. in particular , looking at the distribution of the maser velocity vectors along the ( east - west ) narrow strip outlined by the red - shifted features , we note that maser features closer to the center of motion " move mainly to the south . getting closer to the western end of the maser strip , the orientation of the maser velocity vectors changes gradually from south to west .
the general pattern of water maser relative velocities appears to indicate expansion from a point on the sky close to the `` center of motion '' .
note also that at larger distances from the `` center of motion '' , maser velocities seem to be larger and better collimated along the ( north)east-(south)west direction . in particular ,
we identify two h@xmath44o maser clusters located at the northeast and southwest ends of the maser distribution ( named `` j@xmath45 '' and `` j@xmath46 '' , respectively ; see figure [ g28-rel - pro - s ] ) , which move with high velocities directed close to the northeast - southwest direction .
figure [ g28-abs - pro ] ( upper panel ) reports the intrinsic absolute proper motions of the water maser features , derived from the observed absolute proper motion of feature # 1 corrected for the apparent motion . the latter has been calculated adding the contribution of the parallax , the solar motion with respect to the lsr @xcite , and the differential galactic rotation , from a flat rotation curve with galactic constants ( r@xmath47 kpc , @xmath24 ) as recently determined by @xcite .
the obtained absolute proper motions differ from the relative proper motions presented in figure [ g28-rel - pro - s ] by a vector of amplitude @xmath640 pointing to the southwest , which seems to be the dominant component of all proper motions . we think that this difference could be caused by the large systematic error due to the uncertainties in the source distance and solar and galactic motion , and/or by a peculiar velocity of the maser source with respect to its lsr reference system .
our vlba water maser observations were not designed to measure the source parallax . with a positional accuracy of @xmath60.10.2 milli - arcseconds , and four epochs spanning ( only ) 9 months , the observed absolute proper motion of feature # 1 ,
not corrected for the apparent motion , looks similar to a straight line . from the small deviations from linear motion , we derive an upper limit of 0.2 mas for the amplitude of the sinusoidal parallax signature .
this corresponds to a lower limit of 5 kpc for the source distance , in agreement with the adopted kinematical distance of 7.4 kpc .
we have checked the reliability of the relative , water maser proper motions by following an alternative approach . as described in section .
[ meth_res ] , we have also derived the absolute proper motion of the strongest 6.7 ghz methanol maser feature . to date , vlbi observations of 6.7 ghz @xcite and 12 ghz @xcite methanol masers toward several sources indicate that these masers usually show small proper motions , with typical velocities of only a few km s@xmath22 . assuming that also in methanol masers move significantly slower than the water masers
, we can use the absolute proper motion of the methanol masers to correct the absolute proper motion of the water masers , and derive the water maser velocities as approximately seen in a reference system comoving with the star .
figure [ g28-abs - pro ] ( lower panel ) shows the absolute water maser proper motions after subtracting the absolute methanol maser proper motion ( evaluated in section .
[ meth_res ] ) .
note that , within the measurement errors , the new absolute proper motions are fully consistent with the relative proper motions calculated with respect to the `` center of motion '' ( figure [ g28-rel - pro - s ] ) .
this result reinforces our assertion that the `` center of motion '' can define a suitable reference system to calculate water maser velocities . in the following discussion ,
we use the relative velocities of the water masers to describe the gas kinematics close to the ( proto)star .
figure [ g28-maser - spe ] ( middle panel ) shows the total - power spectrum of the 6.7 ghz methanol masers observed on 2008 march 18 toward using the effelsberg 100-m telescope .
the 6.7 ghz methanol maser spectrum , with two narrow features at v@xmath42 of about 97 and 105 , resembles that of the water masers ( figure [ g28-maser - spe ] , upper panel ) .
note that @xcite detected only the emission at v@xmath42 of 105 with the parkes 64-m telescope .
we imaged the range of lsr velocities from 95 to 107 with the evn and detected two distinct methanol maser features , both with v@xmath42 close to 105 .
feature # 1 is detected at the first , third , and fourth epoch , whereas feature # 2 is detected at the third and fourth epoch only .
due to adverse atmospherical conditions and telescope failures , no 6.7 ghz maser signal was detected at the second epoch .
individual feature properties are presented in table [ tbl : meth ] .
positional offsets are relative to feature # 1 , and the absolute positions of feature # 1 at the three epochs of detection are given in table [ tbl : mas - abs - pos ] .
the two methanol maser features lie at the western border of the distribution of the red - shifted water masers , @xmath002 to the southwest of the vla 3.6 cm continuum peak ( see figure [ g28-absp - l ] ) . inspecting our images
, we could not find any reliable emission at the velocity of @xmath097 , in correspondence of the weakest feature detected in the total - power spectrum .
the mean rms noise on the channel maps was 20 mjy beam@xmath22 , and we could have missed that emission if falling below the 5@xmath32 detection threshold . because of the narrow line width ( @xmath481 ) and the offset in v@xmath42 from the systemic velocity ( 101.0 ) of the hmc , we tend to exclude that the weak methanol feature at v@xmath42 @xmath097 is thermal emission from the hmc resolved out by the evn observations .
after correcting for the apparent motion , the absolute proper motion for feature # 1 is : @xmath49 , @xmath50 . as discussed in sect .
[ wat_res ] , we have used this vector to correct the absolute proper motions of the water masers . the relative motion of the methanol maser feature # 1 with respect to the `` center of motion '' of the water masers is found to be : @xmath51 , @xmath52 .
this is consistent with the results from other vlbi observations which show that 6.7 ghz masers move significantly slower than water masers , and justifies employing the methanol maser absolute motion to correct the water maser absolute motions .
@xcite observed the hmsfr with the vla a - b hybrid configuration ( hpbw of 55 @xmath1 12 ) in the left circular polarization ( lcp ) band and detected eight 1.665 ghz oh maser spots .
their brightest spot had a flux density of 9.85 jy at the v@xmath42 of 102.8 .
figure [ g28-maser - spe ] ( lower panel ) shows the @xmath53-weighted mean of the vlba antenna total - power spectra of the 1.665 ghz oh maser emission in the lcp and rcp bands .
three strong , narrow features are visible in the lcp band over the lsr velocity range from 102 to 108 .
two weak features are detected in the rcp band , one of them at an lsr velocity of @xmath0108 .
these findings are similar to the results obtained by @xcite with the parkes 64-m telescope .
we imaged the range of lsr velocities from 102 to 110 , and identified 3 distinct oh maser features in the lcp band . in the rcp band , by imaging maser data after direct phase - referencing to the calibrator j1834@xmath50301 , no signal was detected above a 5@xmath32 level of 0.1 jy beam@xmath22 .
individual feature properties are presented in table [ tbl : oh ] .
the positional offsets are relative to feature # 1 , whose absolute position is given in table [ tbl : mas - abs - pos ] .
the derived absolute position is offset by @xmath009 from the reference position of @xcite .
maser intensities range from 0.7 to 1.9 jy beam@xmath22 .
the lsr velocities vary from 102.8 for the most blue - shifted feature ( # 1 ) , to 105.5 for the most red - shifted feature ( # 3 ) .
the fwhm line width of individual maser features ranges from 0.4 to 1.4 .
positions and lsr velocities of hydroxy1 maser features are plotted in figure [ g28-absp - l ] .
the three detected features lie @xmath54 to the southeast of the vla 3.6 cm continuum source and are distributed along a line extended @xmath55 .
figure [ g28-q - cm ] shows an image of the 24.5 continuum emission obtained with subaru / comics at an angular resolution of 075 .
one can see that the structure of the emitting region is quite complex . beside two components associated with the radio sources `` hmc '' and `` a '' with a total flux of @xmath080 jy , another fainter source ( @xmath03 jy ) is visible to the south , offset by @xmath04 from `` a '' .
the emission from the northern region is clearly resolved and appears to be roughly associated with the two radio objects , although one can not rule out the existence of additional , fainter unresolved sources contributing to the extended halo around the two main sources . in particular
, the 24.5 emission from `` hmc '' seems to split into two sub - components in the e - w direction , with comparable flux densities : @xmath032 jy for the eastern and @xmath043 jy for the western component .
these could trace two distinct objects or diffuse emission surrounding a single source at the origin of the `` hmc '' radio continuum .
a third , more intriguing possibility is that one is looking at an `` hour - glass '' shaped nebulosity centered on the embedded source at the origin of the free - free emission . in the latter scenario ,
the source lies at the center of a flattened core lying approximately along the line of sight .
this dusty core absorbs most of the ir photons emitted at 24 @xmath56 m towards the observer , while the ir radiation can escape along the minor axis of the core , lying close to the plane of the sky .
we have estimated an optical depth of @xmath57 for the hmc at 24 @xmath56 m , assuming the h@xmath44 column density of @xmath58 @xmath59 obtained by @xcite and the opacity adopted by @xcite see their figure 1 . such a value could be sufficiently large to explain the lack of emission towards the embedded star and determine the `` hour - glass '' shape of the ir emission .
interestingly , the e w orientation of this putative bipolar nebula is also roughly consistent with the direction of the h@xmath44o maser jet in figure [ g28-rel - pro - s ] , suggesting that one could be observing a bipolar flow piercing through a dense core , and revealed on different scales in different tracers .
however , such a scenario , albeit appealing , is not consistent with the inclination of the maser jet with respect to the plane of the sky . in fact , the blue - shifted emission lies to the e ( see figure [ g28-rel - pro - s ] ) and this implies that the eastern lobe should be pointing towards the observer and thus be brighter than the western lobe in the ir .
looking at figure [ g28-q - cm ] one can see that this is not the case , being the emission slightly stronger to the west ( see above ) . in conclusion
, we consider both the `` hour - glass '' nebula and the double source hypotheses too speculative on the basis of the observational evidence collected so far , and in the following we make the conservative assumption that the ir emission is associated with `` hmc '' , with a flux of @xmath075 jy .
the southern source `` a '' has a 24.5 flux density of @xmath05 jy .
also in this case , the association between the ir emission and the radio source is questionable , because the two are offset each other by @xmath01 .
however , this is the astrometrical error of the ir image and we thus believe that such an offset is not sufficient evidence against a common origin for the ir and radio emission toward `` a '' .
therefore , we conclude that both emissions can originate from the same object . figure [ g28-sed - ir ] shows the sed of from 3.6 to 1.1 mm . beside our comics data
, we have used data from the following surveys : galactic legacy infrared mid - plane survey extraordinaire ( glimpse ; * ? ? ?
* ) , mipsgal @xcite , msx @xcite , iras @xcite , the herschel infrared galactic plane survey ( hi - gal ; * ? ? ?
* ; * ? ? ?
* ) , the apex telescope large area survey of the galaxy ( atlasgal ; * ? ? ?
* ) , and the bolocam galactic plane survey ( bgps ; * ? ? ?
* ) . for the herschel / hi - gal data , the most recent release ( obtained in july 2011 with hipe version 7.0.0 ) was used , resulting from data reduction with the romagal software @xcite , where the most relevant image artifacts have been removed .
apart from the iras and msx flux densities , which have been taken from the corresponding point source catalogues ( sources iras18411 - 0338 and g028.8621 + 00.0657 ) , all the other values have been calculated by integrating the emission inside the dashed contour in figure [ inf - ima ] .
the latter was chosen in such a way to include the whole emitting region at all wavelengths . in this way
we overcome the problem of different angular resolutions at different wavelengths . in the calculation of the flux
the background emission was subtracted assuming that the background flux inside the contour is in all points equal to the typical flux density measured along the contour itself .
the flux densities used for the sed are given in table [ tbl : tsed ] .
figure [ inf - ima ] illustrates the pc - scale structure of the star forming region , imaged with herschel from 70 to 500 and with apex at 870 , and demonstrates that from the mid - ir to the sub - mm the continuum emission is dominated by a compact source surrounded by a weaker halo .
we have fitted the sed with the radiative transfer model developed by @xcite and @xcite ( using the sed fitting tool available on http://caravan.astro.wisc.edu/protostars/ ) , which assumes a zams star with a circumstellar disk , embedded in an infalling envelope , and allows derivation of a number of physical parameters .
while we are aware that this model can not adequately reproduce a complex situation such as that of the multiple mysos in , we believe that the fit is sufficiently good to guarantee good estimates of _ integrated _ quantities such as the luminosity and the mass of the envelope .
the best - fit values are , respectively , 2.3@xmath110@xmath2 , and 3.0@xmath110@xmath60 . note that in the fitting procedure , the iras fluxes were considered upper limits due to the limited angular resolution , whereas the mipsgal 24 flux is in fact a lower limit because the corresponding image is saturated and only the non saturated pixels have been considered in the calculation .
also , the distance was kept fixed to the nominal value of 7.4 kpc .
the best - fit value for the visual foreground extinction is 46 mag .
in this section , we focus on the cm - continuum source labeled `` hmc '' in table [ tbl : vla - cm ] , found in good correspondence with the positions of the 22 ghz , 6.7 ghz , and 1.665 ghz oh masers in ( see figure [ g28-absp - l ] ) .
the continuum spectrum ( shown in figure [ g28-hmc - sed - cm ] ) is consistent with free - free emission from an ultracompact ( uc ) region , with the turnover frequency in the range 1020 ghz , and ionized by a lyman continuum rate of 7.9@xmath110@xmath61 s@xmath22 .
this would correspond to a zams star of spectral type b1b0.5 @xcite , with a luminosity of @xmath62 @xmath63 , and a mass of @xmath64 .
it is worth mentioning that these values would be different if more recent studies than the seminal article by panagia were adopted ( e.g. , * ? ? ?
. however , the basic result that one is dealing with an early b - type star with a luminosity of @xmath65 would remain valid .
we note that the fit to the continuum spectrum requires a very small size of the uc region , of @xmath0004 or @xmath0300 au , which appears in contradiction with the fact that our a - array 1.3 cm observations partly resolve the continuum emission ( see table [ tbl : vla - cm ] ) .
since we considered a homogeneous distribution of ionized gas in our fit , one possibility is that the region is not homogeneous .
the southern spur identified in the a - array 3.6 cm map of the `` hmc '' source ( see figure [ g28-absp - l ] ) , could indeed suggest the existence of a density gradient toward the south .
one may speculate that being source `` a '' more extended and perhaps more evolved , the gas in its surroundings is less dense . in this scenario , the north - south streamlines of water masers on either sides of the southern spur , might trace shocks at the interface between the ionized gas of `` hmc '' ( expanding to the south ) and the surrounding molecular gas .
we also note that the continuum spectrum of `` hmc '' between 1.3 and 6 cm can be also fitted with a power law @xmath66 , with @xmath67 ( see figure [ g28-hmc - sed - cm ] ) .
this spectral index is typical of thermal jets @xcite and provides us with an alternative explanation for the origin of the free - free emission .
however , a caveat is in order . as discussed in section [ cm_res ] , the slope of the spectrum between 3.6 cm and 1.3 cm could be significantly affected by the inteferometer resolving out part of the emission at the shorter wavelength .
if the spectrum is steeper than estimated from our data , the slope could be still consistent with a thermal jet , but also with that of an region with a density gradient ( see e.g. , * ? ? ?
we conclude that it is impossible to decide the nature of the free - free emission on the basis of the information collected so far . notwithstanding this caveat , as a working hypothesis we prefer the jet interpretation , because this permits to interpret the continuum and maser emission in the same context , as discussed in the following section .
in this section we propose that both the continuum and the water maser emission are associated with a thermal jet impinging against the surrounding quiescent molecular gas .
this idea is based on two observational findings .
first of all , figure [ g28-rel - pro - s ] shows that , at larger distances from the center of the water maser distribution , maser ( relative ) velocities appear to be larger and better collimated along the northeast - southwest direction .
secondly , the axis of the powerful , molecular outflow mapped at arcsecond scales toward has a northeast - southwest orientation ( * ? ? ?
* figure 5 ) .
this it is plausible to assume that the two , fast moving , water maser clusters `` j@xmath45 '' and `` j@xmath46 '' , located , respectively , at the northeast and southwest edge of the maser distribution , are excited by the same jet responsible also for driving the large - scale molecular outflow .
one could object against the jet " interpretation , that the whole pattern of water maser velocities can not be explained only in terms of a collimated outflow .
in fact , at smaller distances from the center of the maser distribution , water maser velocities point to the north or south directions , and form large angles with the direction of the putative jet .
we propose that the scattered pattern of proper motions presented by the red - shifted water masers can result from the interaction of the jet with dense circumstellar material , which deflects and increase the opening angle of the flow velocities .
if the free - free continuum source `` hmc '' is a jet , the position of the ( proto)star powering it may be offset from the free - free continuum peak .
if a jet powered by the ( proto)star drives the water maser motions , and , in particular , the expansion to the south and west of the narrow strip of red - shifted water masers , the ( proto)star could be located at an intermediate position between `` hmc '' and the red - shifted maser strip .
this point is not far ( within tens of mas ) from the position of the center of motion of the water masers ( cross in figure [ g28-rel - pro - s ] ) . considering the large uncertainties in deriving the center of water maser expansion , in the following discussion we can safely take the center of motion as a reasonable estimate of the ( proto)star location . in the jet scenario , one can estimate the momentum rate of the jet from the average distance of the h@xmath44o masers from the myso and the average maser velocity . assuming a h@xmath68 pre - shock density of 10@xmath69 @xmath70 , as predicted by models @xcite , the momentum rate in the water maser jet is given by the expression @xmath71
@xmath72 yr@xmath22 km s@xmath22 . here
@xmath73 is the average maser velocity in units of 10 km s@xmath22 , @xmath74 the average distance of water masers from the yso in units of 100 au , and @xmath75 the solid angle of the jet .
this expression has been obtained by multiplying the momentum rate per unit surface transferred to the ambient gas ( @xmath76 ) , by @xmath77 , under the assumption that the jet is emitted from a source at a distance @xmath78 from the masers within a beaming angle @xmath75 . to estimate the average distance and velocity of the water masers tracing the jet in
, we use all the maser features belonging to the two fast moving clusters `` j@xmath45 '' and `` j@xmath46 '' ( see figure [ g28-rel - pro - s ] ) .
the resulting average distance is 122 mas ( or 902 au ) and the average speed is 43 km s@xmath22 .
using these values , the jet momentum rate is @xmath79 @xmath72 yr@xmath22 km s@xmath22 . for shock
induced ionization in a thermal jet , the momentum rate of the jet can be estimated from the continuum flux assuming the emission is optically thin ( see @xcite ) .
one has : @xmath80 , where @xmath81 is the measured continuum flux density in mjy , @xmath82 the jet momentum rate in @xmath72 yr@xmath22 km s@xmath22 , @xmath75 the jet solid angle in sterad , and @xmath83 the source distance in kpc . using the flux of 1.2 mjy measured at 1.3 cm with the vla c - array for source `` hmc '' , and the distance of 7.4 kpc , one derives @xmath84
@xmath72 yr@xmath22 km s@xmath22 .
note that the momentum rates estimated from the maser motion and the continuum flux have a different dependency on the jet solid angle , which can thus be estimated by equating the two momentum rate estimates .
we obtain an expression for the jet solid angle : @xmath85 .
we find @xmath86 sterad , i.e. a jet semi - opening angle @xmath87 .
correspondingly , the jet momentum rate is @xmath88 yr@xmath22 km s@xmath22 .
this value is large enough to justify the momentum rate of several @xmath89 yr@xmath22 km s@xmath22 , estimated by @xcite for the large - scale molecular outflow .
this result supports our hypothesis that the small - scale thermal and h@xmath44o maser jet is sufficiently powerful to feed the large - scale molecular outflow observed by @xcite .
we note that the derived parameters for the jet in are comparable to those of the few collimated outflows observed up to now in massive star - forming regions . for comparison , the jets in hh 8081 @xcite , in iras 16547@xmath54247 @xcite , and in iras 20126 + 4104 @xcite have momentum rates of the order of @xmath90 yr@xmath22 km s@xmath22 and opening angles of 2030 .
the two detected 6.7 ghz methanol maser features are approximately aligned along the same northeast - southwest line defined by the two maser jet clusters `` j@xmath45 '' and `` j@xmath46 '' , the peak of the vla a - array 1.3 cm , and the center of motion ( see figure [ g28-absp - l ] ) . figure [ g28-maser - spe ] shows that the total - power spectra of the 6.7 ghz methanol and 22 ghz water masers are similar , both presenting a strong red - shifted and a much weaker blue - shifted component .
the positions of the methanol maser features roughly along the jet axis , and the similarities in the spectral shape of the water and methanol maser emissions , make us speculate that also the 6.7 ghz methanol masers could originate in the ( proto)stellar jet , albeit associated with more quiescent gas than that traced by the water masers .
maser vlbi observations toward the sources iras20126@xmath44104 and @xcite have also indicated that the 6.7 ghz masers might trace entrained gas at the interface with a jet . figure
[ g28-absp - l ] clearly shows that the blue - shifted lobe of the water maser jet is sampled much more sparsely , as it presents much less features than the red - shifted lobe .
this is consistent with the blue - shifted line being significantly weaker than the red - shifted one in the total power spectrum ( see figure [ g28-maser - spe ] , upper panel ) .
this fact suggests that the circumstellar material is not isotropically distributed around the ( proto)star , but it is mostly concentrated on the side of the red - shifted lobe .
that would also account for the stronger emission line in the total - power spectrum of the masers at the red - shifted velocities .
the spectrum of the 1.665 ghz maser is dominated by three narrow lcp features ( see figure [ g28-maser - spe ] , lower panel ) , and is very similar to the spectrum of the 1.667 ghz maser transition observed by @xcite .
these authors find that most oh interstellar masers are highly polarized , either rcp or lcp , with the same predominant polarization for both the 1.665 ghz and 1.667 ghz transition .
they propose an explanation based on large - scale velocity and/or magnetic field gradients across the masing region .
figure [ g28-vla - cm ] shows that the 1.665 ghz oh maser features are found significantly offset ( @xmath54 ) from the compact vla source `` hmc '' and the center of water maser activity , and are closer to the more extended source `` a '' .
a possible interpretation is that oh masers trace more external , less dense parts of the molecular core surrounding the ( proto)star responsible for the excitation of the water and methanol masers .
observations indicate that the density of massive proto - stellar cores decreases from the center outward , with a power - law index of about 1.6 ( @xmath91 ) @xcite . in ,
the ratio of the average distance ( from the `` center of motion '' of water masers ) of the water ( @xmath92 ) and oh ( @xmath54 ) masers , is about 6 , implying a decrease in gas density by a factor of about 20
. that agrees with maser excitation models , which predict a h@xmath68 pre - shock density of 10@xmath9310@xmath94 @xmath70 for oh 1.665 ghz @xcite , and 10@xmath69 @xmath70 for 22 ghz masers @xcite , respectively .
so far , we have explored the properties of source `` hmc '' making use of the maser , radio , and ir continuum emission .
now , we want to use our findings to shed light on the nature of the myso(s ) associated with this source .
we have seen that the radio continuum emission and the water maser features suggest the presence of a jet with a large momentum rate , on the order of @xmath95 km s@xmath22 yr@xmath22 .
values that high are consistent with the jet being powered by a high - mass ( proto)stars , as indicated , e.g. , by the relationship between outflow momentum rate and myso luminosity obtained by @xcite . in particular , from their figure 4b , one can conclude that the myso must have a luminosity of at least several @xmath96 .
alternatively , if the radio emission were originating from an region , we have seen ( section [ dis_1 ] ) that the ionizing star should have a luminosity @xmath97 .
clearly , both the thermal jet and the region hypotheses imply similar luminosity for the associated myso .
the question we want to investigate here is whether such a luminosity is consistent with the results obtained from the ir data . in section
[ sed ] we have found that the total luminosity of the region is on the order of @xmath98 .
however , this includes all mysos in a radius @xmath9910 around `` hmc '' , whereas we need to estimate the contribution of the sole `` hmc '' , to achieve a consistent comparison with the luminosity estimate obtained from the water maser and/or radio continuum .
our 24.5 continuum image can be used for this purpose .
as shown in figure [ g28-q - cm ] , the `` hmc '' source is dominating the emission at this wavelength , contributing to @xmath090% of the total flux density , and a similar conclusion holds also for the glimpse images . due to the lack of sub - arcsecond imaging at longer wavelengths , it is difficult to establish whether also the flux density measured close to the peak of the sed is mostly due to source `` hmc '' . nonetheless , the hi - gal / herschel data suggest that the 70 @xmath56 m emission peaks closer to source `` hmc '' than source `` a '' .
a word of caution is in order here .
the angular resolution of the hi - gal data at 70 is estimated to be about @xmath100 ( after reduction with the roma - gal software ) , while the astrometrical error is @xmath035 ( the latter results from aligning the pacs 70 @xmath56 m images to the mipsgal 24 @xmath56 m images by matching suitable point sources detected at both wavelengths ) .
although large , these values still permit to associate the 70 emission with source `` hmc '' .
in fact , the angular separation between the 3.6 cm continuum source `` hmc '' and the 70 @xmath56 m peak is 37 , consistent with the positional error , whereas source `` a '' lies 65 from the 70 @xmath56 m peak .
this can be appreciated in figure [ fcm70 ] , where one sees that the 70 @xmath56 m emission peaks to the nw of `` hmc '' .
therefore , in the following we assume that the bolometric luminosity estimated from the sed is likely dominated by the `` hmc '' source and conclude that the myso(s ) embedded in the hmc may be as luminous as @xmath101 . notwithstanding
the large uncertainties of all our luminosity estimates , the value obtained from the ir images exceeds by an order of magnitude that derived from the maser and radio continuum data .
this inconsistency can be explained if `` hmc '' is hosting a multiple stellar system rather than a single star .
one can get an idea of how this may affect our estimate of the stellar properties ( mass , etc . ) by assuming that the luminosity of @xmath98 is emitted by a stellar cluster with a @xcite initial mass function .
one finds that the luminosity of the most massive star may be as small as @xmath102 , depending on the statistical distribution of the stellar masses in the cluster .
given the uncertainties , this value is only marginally greater than the luminosity estimate from the radio continuum and maser data .
the same method provides us with the total mass of the stellar cluster , which ranges from @xmath103 to @xmath104 , depending on the statistical distribution of the stars in the cluster .
this stellar mass can be divided by the total mass of the associated molecular clump , @xmath105 ( see section [ sed ] ) , to calculate the star formation efficiency , 17%67% .
albeit large , values like these are not implausible , as we are considering the part of the molecular cloud where star formation concentrates @xcite . in conclusion , we believe that the observational evidence collected so far for source `` hmc '' suggests that one is observing a young star of 2030 @xmath106 , still in the main accretion phase and surrounded by a rich stellar cluster .
such a conclusion holds independently of the scenario adopted to explain the free - free emission ( thermal jet or weak region ) and raises a question : why only faint free - free emission is detected from this object ?
a @xmath102 star is expected to ionize the surrounding gas emitting a flux density of @xmath02 mjy ( in the optically thin limit ) , whereas towards source `` hmc '' less than 1 mjy is measured .
a possible explanation is that the region is at least partly quenched by accretion from the parental core , as depicted , e.g. , in the model by @xcite .
although no evidence of accretion is seen in a typical infall tracer such as hco@xmath107 @xcite , this fact is inconclusive because the majority of the detected hco@xmath16(10 ) might be part of the outflowing gas , thus hindering the detection of infall .
finally , it is also worth mentioning that according to recent theoretical models @xcite , large accretion rates are bound to bloat the protostellar radius and decrease the corresponding temperature and lyman continuum flux .
this could be an alternative explanation of the lack of a ( strong ) region .
we observed the hmsfr with the vlbi phase - referencing technique in three powerful maser transitions : 22.2 ghz , 6.7 ghz , and 1.665 ghz oh . in addition , we also performed vla observations of the radio continuum emission at 1.3 and 3.6 cm with both the a- and c - array , and subaru / comics observations of the mid - infrared continuum emission .
we also made use of data from the hi - gal / herschel , atlasgal / apex , as well as other surveys of the galactic plane .
from all these observations , we draw the following conclusions : 1 .
the bipolar distribution of l.o.s . velocities and the general pattern of observed relative proper motions indicate that the water masers are tracing expansion with an average velocity of @xmath024 from a point on the sky close to the maser `` center of motion '' . while at larger distance from the `` center of motion '' , maser velocities are higher and better collimated along the northeast - southwest direction , at closer distances water masers show smaller speeds and a larger scatter in the velocity orientations .
we interpret these facts as the result of the interaction of a ( proto)stellar jet with dense circumstellar material .
two continuum sources ( named `` hmc '' and `` a '' ) , separated by @xmath031 along the north - south direction , are observed with the vla at 3.6 and 1.3 cm . the `` hmc '' source is compact and spatially associated with the observed , and oh masers . whether the free - free emission arising from this source is due to an region or a thermal ionized jet
can not be unambiguously established with the present data .
however , we believe that the jet hypothesis is more consistent with the direction of the water maser motions , which outline expansion in a bipolar flow . moreover , the orientation and momentum rate of the putative jet are compatible with those of the molecular outflow detected at arcsecond scales .
we hence argue that the jet traced by the water masers is also powering the large - scale outflow .
the combined information obtained from sub - arcsecond imaging with subaru / comics at 24.5 and lower angular resolution images from the hi - gal / herschel and atlasgal / apex surveys , permit to estimate with good accuracy the luminosity ( @xmath98 ) and gas mass ( @xmath105 ) of the star forming region of , and establish that @xmath090% of the luminosity is coming from the radio source `` hmc '' .
we conclude that this source must contain multiple stars , with the most massive being at least as luminous as @xmath108 .
the lack of an associated hii region indicates that the 2030 @xmath106 star must be still undergoing heavy accretion from the surrounding envelope .
we thank the anonymous referee for careful reading of the text and the constructive criticisms that improved the presentation and analysis of our results .
this work was supported by the chinese nsf through grants nsf 11133008 , nsf 11073054 , nsf 10733030 , and nsf 10703010 .
anglada , g. , estalella , r. , pastor , j. , rodriguez , l. f. , & haschick , a. d. 1996 , , 463 , 205 araya , e. , hofner , p. , goss , w. m. , et al .
2007 , , 170 , 152 beltrn , m. t. , cesaroni , r. , codella , c. , et al . 2006 ,
nature , 443 , 427 benjamin , r. a. , churchwell , e. , babler , b. l. , et al .
2003 , , 115 , 953 beuther , h. , schilke , p. , menten , k. m. , et al .
2002 , , 566 , 945 beuther , h. , schilke , p. , sridharan , t. k. , et al .
2002 , , 383,892 bonnell , i. a. , bate , m. r. , & zinnecker , h. 1998 , , 298 , 93 bonnell , i. a. , & bate , m. r. 2006 , , 370 , 488 bonnell , i. a. , smith , r. j. , clark , p. c. , & bate , m. r. 2011 , , 410 , 2339 bronfman , l. , nyman , l .- a . , & may , j. 1996 , , 115 , 81 brunthaler , a. , reid , m. j. , menten , k. m. , et al . 2011 , astron .
nachr . , 332 , 461 carey , s. j. , noriega - crespo , a. , mizuno , d. r. , et al .
2009 , , 121 , 76 caswell , j. l. , & haynes , r. f. 1983 , aust .
j. phys . , 36 , 417 caswell , j. l. , vaile , r. a. , ellingsen , s. p. , whiteoak , j. b. , & norris , r. p. 1995
, , 272 , 96 cesaroni , r. , & walmsley , c. m. 1991 , , 241 , 537 codella , c. , testi , l. , & cesaroni , r. 1997 , , 325 , 282 cohen , m. , walker , r. g. , carter , b. , et al . 1999 , , 117 , 1864 cragg , d. m. , sobolev , a. m. , & godfrey , p. d. 2002 , , 331 , 521 crowther , d. m. 2005 , in massive star birth : a crossroads of astrophysics , iau symposium 227 , ed .
r. cesaroni , m. felli , e. churchwell , m. walmsley ( cambridge university press ) , 389 de buizer , j. m. , radomski , j. t. , telesco , c. m. , & pia , r. k. 2005 , , 156 , 179 drosback , m. m. , aguirre , j. , bally , j. , et al .
2008 , , 212 , 9601 egan , m. p. , price , s. d. , & kraemer , k. e. 2003 , , 203 , 5708 elitzur , m. , hollenbach , d. j. , & mckee , c. f. 1989 , , 346 , 983 fandez , s. , bronfman , l. , garay , g. , et al .
2004 , , 426 , 97 forster , j. r. , & caswell , j. l. 1999 , , 137 , 43 franco , j. , kurtz , s. , hofner , p. , et al .
2000 , , 542 , l143 furuya , r. s. , cesaroni , r. , takahashi , s. , et al . 2008 , , 673 , 363 furuya , r. s. , cesaroni , r. , shinnaga , h. 2011 , , 525 , 72 genzel , r. , & downes , d. 1979 , , 72 , 234 hosokawa , t. , yorke , h. w. , omukai , k. 2010 , , 721 , 478 kataza , h. , okamoto , y. , takubo , s. , et al .
2000 , , 4008 , 1144 keto , e. 2002 , , 580 , 980 keto , e. 2003 , , 599 , 1196 klaassen , p. d. , wilson , c. d. , keto , e. r. , et al .
2011 , , 530 , 53 lada , c. j. , & lada , e. a. 2003 , , 41 , 57 little , l. t. , & cesarsky d.a . 1982 , , 112 , 49 lpez - sepulcre , a. , codella , c. , cesaroni , r. , marcelino , n. , & walmsley , c. m. 2009 , , 499 , 811 marti , j. , rodriguez , l. f. , & reipurth , b. 1998 , , 502 , 337 mckee , c. f. , & tan , j. c. 2003 , , 585 , 850 menten , k. m. , & reid , m. j. 1995 , , 445 , l157 miller , g. e. , & scalo , j. m. 1979 , , 41 , 513 molinari , s. , swinyard , b. , bally , j. , et al .
2010a , , 122 , 314 molinari , s. , swinyard , b. , bally , j. , et al .
2010b , , 518 , 100 moscadelli , l. , menten , k. m. , walmsley , c. m. , & reid , m. j. 2003 , , 583 , 776 moscadelli , l. , xu , y. , & chen , x. 2010 , , 526 , 66 moscadelli , l. , cesaroni , r. , rioja , m. j. , dodson , r. , & reid , m. j. 2011 , , 526 , 66 neugebauer , g. , habing , h. j. , van duinen , r. , et al . 1984 , , 278 , 1 panagia , n. 1973 , , 78 , 929 plume , r. , jaffe , d. t. , & evans , n. j. , ii . 1992 , , 78 , 505 purcell , c. r. , hoare , m. g. , & diamond , p. 2008 , asp conference series , 387 , 389 reynolds , s. p. 1986
, , 304 , 713 robitaille , t. p. , whitney , b. a. , indebetouw , r. , & wood , k. 2007 , , 169 , 328 rodrguez , l. f. , moran , j. m. , franco - hernndez , r. , et al .
2008 , , 135 , 2370 sanna , a. , moscadelli , l. , cesaroni , r. , et al .
2010a , , 517 , 71 sanna , a. , moscadelli , l. , cesaroni , r. , et al . 2010b , , 517 , 78 schuller , f. , menten , k. , contreras , y. , et al .
2009 , , 504 , 415 schnrich , r. , binney , j. j. , & dehnen , w. 2010 , , 403 , 1829 sewilo , m. , watson , c. , araya , e. , et al .
2004 , , 154 , 553 shirley , y. l. , evans , n. j. ii , young , k. e. , knez , c. , & jaffe , d. t. 2003 , , 149 , 375 testi , l. , felli , m. , persi , p. , & roth , m. 1994 , , 288 , 634 traficante , a. , calzoletti , l. , veneziani , m. , et al .
2011 , , 416 , 2932 whitney , b. a. , wood , k. , bjorkman , j. e. , & wolff , m. j. 2003a , , 591 , 1049 whitney , b. a. , wood , k. , bjorkman , j. e. , & wolff , m. j. 2003b , , 598 , 1079 zinnecker , h. , & yorke , h. w. 2007 , , 45 , 481 ) weighted average of the 22.2 ghz total - power spectra of the 9 vlba antennas observing on 2006 april 23 .
middle panel : effelsberg total - power spectrum of the 6.7 ghz methanol maser emission on 2008 march 18 .
lower panel : t@xmath109-weighted average of the 1.665 ghz total - power spectra of the 9 vlba antennas observing on 2007 april 28 .
continuous ( black ) and dashed ( red ) lines are used to distinguish between right ( rcp ) and left ( lcp ) circular polarizations , respectively . the dotted line crossing
the spectra denotes the systemic velocity ( v@xmath109 ) inferred from nh@xmath9 measurements .
[ g28-maser - spe ] ] + + + .
[ g28-q - cm ] ] lists the infrared telescopes used to reconstruct the sed .
we have fitted the sed with the radiative transfer model developed by @xcite ( using the sed fitting tool available on http://caravan.astro.wisc.edu/protostars/ ) .
[ g28-sed - ir ] ] cccccccccc hmc & vla@xmath5a & 1.3 & 0.122 @xmath1 0.093 & @xmath517 & 0.06 & 18 34 46.23 & @xmath503 35 29.8 & 0.62 & 0.64 + & vla@xmath5c & 1.3 & 1.053 @xmath1 0.755 & @xmath59 & 0.05 & 18 43 46.23 & @xmath503 35 29.8 & 1.03 & 1.20 + & vla@xmath5a & 3.6 & 0.33 @xmath1 0.24 & @xmath51 & 0.03 & 18 43 46.23 & @xmath503 35 29.9 & 0.35 & 0.42 + & vla@xmath5c & 3.6 & 2.62 @xmath1 2.11 & 5 & 0.02 & 18 43 46.25 & @xmath503 35 29.8 & 0.53 & 0.55 + a & vla@xmath5a & 1.3 & 0.122 @xmath1 0.093 & @xmath517 & 0.06 & & & @xmath200.18(3@xmath32 ) & + & vla@xmath5c & 1.3 & 1.053 @xmath1 0.755 & @xmath59 & 0.05 & 18 43 46.30 & @xmath503 35 32.8 & 0.39 & 0.60 + & vla@xmath5a & 3.6 & 0.33 @xmath1 0.24 & @xmath51 & 0.03 & & & @xmath200.09(3@xmath32 ) & + & vla@xmath5c & 3.6 & 2.62 @xmath1 2.11 & 5 & 0.02 & 18 43 46.29 & @xmath503 35 32.7 & 0.81 & 0.98 + lclllcll h@xmath68o & 1 & 2006 apr 23 & 18 43 46.22467 & @xmath503 35 29.8161 & 57.4 & [email protected] @ @xmath517 + & & 2006 jun 30 & 18 43 46.22465 & @xmath503 35 29.8173 & 50.3 & [email protected] @ @xmath516 + & & 2006 sep 28 & 18 43 46.22455 & @xmath503 35 29.8190 & 32.0 & [email protected] @ @xmath53 + & & 2007 jan 18 & 18 43 46.22448 &
@xmath503 35 29.8212 & 20.7 & [email protected] @ 20 + ch@xmath15oh & 1 & 2006 feb 28 & 18 43 46.22079 & @xmath503 35 29.8379 & 0.044 & [email protected] @ 5 + & & 2007 mar 18 & & & & + & & 2008 mar 18 & 18 43 46.22031 & @xmath503 35 29.8521 & 0.246 & [email protected] @ 45 + & & 2009 mar 17 & 18 43 46.22011 & @xmath503 35 29.8595 & 0.175 & [email protected] @ 13 +
oh & 1 & 2007 apr 28 & 18 43 46.26771 & @xmath503 35 30.4697 & 1.94 & [email protected] @ 1 & + [ tbl : mas - abs - pos ] clllccccc 0 & 1,2,3,4 & 103.17 & & [email protected] & [email protected] & 0 & 0 + 1 & 1,2,3,4 & 106.01 & 53.99 & 0 & 0 & @[email protected] & [email protected] + 2 & 1,2,3,4 & 105.07 & 26.07 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 3 & 1 & 104.36 & 25.38 & @[email protected] & [email protected] & & + 4 & 1,2,3 & 105.27 & 19.92 & @[email protected] & [email protected] & @[email protected] & [email protected] + 5 & 2,3,4 & 104.27 & 17.62 & @[email protected] & [email protected] & @[email protected] & [email protected] + + ccccccccc 1 & 1,3,4 & 104.88 & 0.155 & 0.190 & 0 & 0 & [email protected] & @[email protected] + 2 & 3,4 & 105.21 & 0.268 & 0.283 & @[email protected] & @[email protected] & & + ccccccc 1 & l & 102.9 & 1.94 & 5.26 & 0 & 0 + 2 & l & 106.1 & 1.44 & 3.52 & @[email protected] & [email protected] + 3 & l & 105.4 & 0.67 & 1.78 & @[email protected] & [email protected] + ccc 3.6 & 0.9 & spitzer / glimpse + 4.5 & 2.1 & spitzer / glimpse + 5.8 & 5.3 & spitzer / glimpse + 8.0 & 9.4 & spitzer / glimpse + 8.3 & 3.3 & msx + 12 & 8.6@xmath110 & iras + 12.1 & 7.1 & msx + 14.7 & 15.3 & msx + 21.3 & 34.0 & msx + 24 & @xmath11121 & spitzer / mipsgal + 24.6 & 87.4 & subaru / comics + 25 & 105@xmath110 & iras + 60 & 1720@xmath110 & iras + 70 & 1790 & herschel / hi - gal + 100 & @xmath203180 & iras + 160 & 1350 & herschel / hi - gal + 250 & 630 & herschel / hi - gal + 350 & 200 & herschel / hi - gal + 500 & 53 & herschel / hi - gal + 870 & 8.8 & apex / atlasgal + 1100 & 2.5 & cso / bgps + @xmath110 all iras fluxes are assumed to be upper limits in figure [ g28-sed - ir ]
clllccccc 0 & 1,2,3,4 & 103.17 & & [email protected] & [email protected] & 0 & 0 + 1 & 1,2,3,4 & 106.01 & 53.99 & 0 & 0 & @[email protected] & [email protected] + 2 & 1,2,3,4 & 105.07 & 26.07 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 3 & 1 & 104.36 & 25.38 & @[email protected] & [email protected] & & + 4 & 1,2,3 & 105.27 & 19.92 & @[email protected] & [email protected] & @[email protected] & [email protected] + 5 & 2,3,4 & 104.27 & 17.62 & @[email protected] & [email protected] & @[email protected] & [email protected] + 6 & 2 & 105.06 & 15.08 & [email protected] & @[email protected] & & + 7 & 2 & 105.14 & 14.36 & [email protected] & @[email protected] & & + 8 & 1,2,3,4 & 105.21 & 12.16 & [email protected] & @[email protected] & [email protected] & @[email protected] + 9 & 1,2,3,4 & 105.45 & 11.21 & @[email protected] & [email protected] & @[email protected] & [email protected] + 10 & 1,2,3,4 & 105.25 & 9.40 & [email protected] & @[email protected] & [email protected] & @[email protected] + 11 & 1 & 103.92 & 9.39 & @[email protected] & [email protected] & & + 12 & 1,2,3,4 & 104.80 & 9.30 & [email protected] & @[email protected] & [email protected] & @[email protected] + 13 & 1,2,3 & 104.00 & 9.15 & @[email protected] & [email protected] & @[email protected] & @[email protected] + 14 & 2 & 105.35 & 8.57 & @[email protected] & [email protected] & & + 15 & 1,2,3,4 & 106.34 & 8.22 & [email protected] & @[email protected] & @[email protected] & [email protected] + 16 & 1,2,3,4 & 106.73 & 7.53 & @[email protected] & [email protected] & @[email protected] & [email protected] + 17 & 2 & 105.12 & 7.33 & 3.31 @xmath210.05 & @[email protected] & & + 18 & 2,3,4 & 104.94 & 7.23 & [email protected] & @[email protected] & & + 19 & 1,2,3,4 & 105.87 & 7.22 & [email protected] & @[email protected] & [email protected] & @[email protected] + 20 & 2 & 104.70 & 7.02 & @[email protected] & [email protected] & & + 21 & 1,2,3,4 & 105.31 & 6.90 & [email protected] & @[email protected] & [email protected] & @[email protected] + 22 & 1,2 & 104.23 & 6.82 & [email protected] & @[email protected] & & + 23 & 1,2,3,4 & 105.42 & 6.57 & [email protected] & @[email protected] & [email protected] & @[email protected] + 24 & 2 & 105.11 & 6.48 & @[email protected] & @[email protected] & & + 25 & 2 & 105.10 & 6.45 & [email protected] & @[email protected] & & + 26 & 1,2,3 & 95.16 & 6.20 & [email protected] & [email protected] & [email protected] & [email protected] + 27 & 1,2,3,4 & 105.50 & 6.16 & [email protected] & [email protected] & [email protected] & @[email protected] + 28 & 2 & 104.36 & 5.79 & @[email protected] & [email protected] & & + 29 & 1,2,3,4 & 103.91 & 5.57 & @[email protected] & [email protected] & @[email protected] & @[email protected] + 30 & 2,3,4 & 103.78 & 5.30 & @[email protected] & [email protected] & [email protected] & @[email protected] + 31 & 1,2,3,4 & 105.13 & 5.28 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 32 & 1,2,4 & 104.22 & 5.23 & @[email protected] & @[email protected] & @[email protected] & @[email protected] + 33 & 1 & 104.35 & 5.23 & @[email protected] & [email protected] & & + 34 & 1,2,3,4 & 106.26 & 5.21 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 35 & 3 & 104.71 & 5.17 & [email protected] & @[email protected] & & + 36 & 1,2,3,4 & 105.46 & 4.86 & [email protected] & @[email protected] & [email protected] & @[email protected] + 37 & 1,2,3 & 105.15 & 4.78 & [email protected] & [email protected] & [email protected] & @[email protected] + 38 & 2 & 105.54 & 4.72 & [email protected] & [email protected] & & + 39 & 3 & 104.72 & 4.57 & @[email protected] & [email protected] & & + 40 & 1,2,3,4 & 108.62 & 4.46 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 41 & 2 & 104.93 & 4.37 & [email protected] & @[email protected] & & + 42 & 1,2 & 104.71 & 4.32 & @[email protected] & [email protected] & & + 43 & 2 & 105.12 & 4.17 & [email protected] & [email protected] & & + 44 & 1 & 104.23 & 4.06 & @[email protected] & [email protected] & & + 45 & 1,2,3,4 & 106.20 & 4.03 & [email protected] & @[email protected] & [email protected] & @[email protected] + 46 & 2 & 105.52 & 3.96 & [email protected] & [email protected] & & + 47 & 1,2,3,4 & 104.67 & 3.95 & @[email protected] & [email protected] & @[email protected] & [email protected] + 48 & 1,2,3 & 104.04 & 3.91 & @[email protected] & [email protected] & @[email protected] & [email protected] + 49 & 1,2,3,4 & 103.85 & 3.84 & @[email protected] & [email protected] & @[email protected] & [email protected] + 50 & 1,2,3,4 & 105.44 & 3.71 & [email protected] & @[email protected] & [email protected] & [email protected] + 51 & 1,2 & 105.06 & 3.67 & [email protected] & @[email protected] & & + 52 & 1 & 105.41 & 3.60 & @[email protected] & @[email protected] & & + 53 & 2,3,4 & 105.55 & 3.24 & [email protected] & @[email protected] & [email protected] & @[email protected] + 54 & 1,2,3,4 & 106.82 & 3.22 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 55 & 2,3 & 106.46 & 3.13 & [email protected] & @[email protected] & & + 56 & 4 & 105.35 & 3.07 & [email protected] & @[email protected] & & + 57 & 3,4 & 105.95 & 3.02 & @[email protected] & [email protected] & & + 58 & 2,3,4 & 106.37 & 2.93 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 59 & 3 & 105.39 & 2.90 & [email protected] & @[email protected] & & + 60 & 2 & 105.01 & 2.86 & [email protected] & [email protected] & & + 61 & 1 & 104.90 & 2.81 & @[email protected] & [email protected] & & + 62 & 4 & 105.34 & 2.77 & [email protected] & @[email protected] & & + 63 & 1,2,3,4 & 105.46 & 2.72 & [email protected] & [email protected] & @[email protected] & @[email protected] + 64 & 3 & 105.83 & 2.63 & [email protected] & @[email protected] & & + 65 & 1,2 & 105.30 & 2.51 & [email protected] & @[email protected] & & + 66 & 4 & 103.22 & 2.40 & @[email protected] & [email protected] & & + 67 & 1,2,3 & 105.21 & 2.34 & [email protected] & @[email protected] & @[email protected] & [email protected] + 68 & 1,2,3 & 106.13 & 2.30 & [email protected] & @[email protected] & @[email protected] & [email protected] + 69 & 1,2,3,4 & 106.85 & 2.28 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 70 & 2,3 & 105.04 & 2.26 & [email protected] & @[email protected] & & + 71 & 2 & 106.54 & 2.21 & [email protected] & @[email protected] & & + 72 & 4 & 105.71 & 2.15 & [email protected] & @[email protected] & & + 73 & 3 & 105.30 & 2.08 & [email protected] & @[email protected] & & + 74 & 1,2,3,4 & 106.74 & 1.97 & [email protected] & @[email protected] & @[email protected] & [email protected] + 75 & 2 & 106.40 & 1.95 & [email protected] & @[email protected] & & + 76 & 3 & 106.37 & 1.91 & [email protected] & @[email protected] & & + 77 & 1,2,3,4 & 104.43 & 1.87 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 78 & 3 & 106.05 & 1.87 & @[email protected] & [email protected] & & + 79 & 3 & 106.54 & 1.82 & [email protected] & [email protected] & & + 80 & 3 & 106.27 & 1.82 & [email protected] & @[email protected] & & + 81 & 3 & 105.94 & 1.81 & @[email protected] & [email protected] & & + 82 & 2,3 & 106.59 & 1.80 & [email protected] & @[email protected] & & + 83 & 1,2,3,4 & 94.90 & 1.80 & [email protected] & [email protected] & [email protected] & [email protected] + 84 & 3 & 106.57 & 1.72 & [email protected] & @[email protected] & & + 85 & 4 & 105.12 & 1.67 & [email protected] & [email protected] & & + 86 & 1 & 105.73 & 1.66 & [email protected] & @[email protected] & & + 87 & 3 & 106.06 & 1.66 & [email protected] & @[email protected] & & + 88 & 3 & 106.37 & 1.64 & [email protected] & [email protected] & & + 89 & 4 & 105.42 & 1.61 & [email protected] & @[email protected] & & + 90 & 1,2,3,4 & 95.86 & 1.60 & [email protected] & [email protected] & [email protected] & [email protected] + 91 & 1 & 103.89 & 1.58 & [email protected] & @[email protected] & & + 92 & 1,2 & 104.29 & 1.52 & [email protected] & @[email protected] & & + 93 & 4 & 105.12 & 1.49 & [email protected] & @[email protected] & & + 94 & 1,2,3 & 104.78 & 1.48 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 95 & 2,3,4 & 107.30 & 1.45 & @[email protected] & [email protected] & @[email protected] & [email protected] + 96 & 4 & 105.31 & 1.39 & [email protected] & @[email protected] & & + 97 & 1,2,3,4 & 106.59 & 1.37 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 98 & 3 & 103.95 & 1.35 & @[email protected] & @[email protected] & & + 99 & 2,3 & 106.78 & 1.30 & [email protected] & [email protected] & & + 100 & 4 & 104.94 & 1.28 & [email protected] & @[email protected] & & + 101 & 2 & 106.96 & 1.25 & [email protected] & @[email protected] & & + 102 & 1 & 103.88 & 1.25 & @[email protected] & [email protected] & & + 103 & 2,3 & 106.11 & 1.24 & @[email protected] & [email protected] & & + 104 & 3 & 104.07 & 1.12 & @[email protected] & @[email protected] & & + 105 & 4 & 105.11 & 1.11 & [email protected] & [email protected] & & + 106 & 4 & 105.50 & 1.11 & [email protected] & @[email protected] & & + 107 & 2,3 & 111.99 & 1.10 & [email protected] & [email protected] & & + 108 & 1,2,3,4 & 106.79 & 1.09 & [email protected] & @[email protected] & [email protected] & @[email protected] + 109 & 2,3 & 108.68 & 1.03 & [email protected] & [email protected] & & + 110 & 3 & 106.57 & 0.97 & [email protected] & @[email protected] & & + 111 & 1 & 106.80 & 0.94 & [email protected] & @[email protected] & & + 112 & 1,2,3 & 106.82 & 0.86 & [email protected] & @[email protected] & [email protected] & @[email protected] + 113 & 4 & 106.91 & 0.83 & @[email protected] & @[email protected] & & + 114 & 3 & 104.67 & 0.81 & [email protected] & @[email protected] & & + 115 & 4 & 103.55 & 0.79 & @[email protected] & [email protected] & & + 116 & 4 & 103.92 & 0.73 & @[email protected] & [email protected] & & + 117 & 1,2,3,4 & 103.66 & 0.72 & [email protected] & @[email protected] & [email protected] & [email protected] + 118 & 1,2 & 94.32 & 0.69 & [email protected] & [email protected] & & + 119 & 4 & 103.64 & 0.68 & [email protected] & @[email protected] & & + 120 & 3,4 & 107.47 & 0.66 & [email protected] & @[email protected] & & + 121 & 1,2,3 & 107.35 & 0.61 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 122 & 1 & 107.18 & 0.57 & @[email protected] & [email protected] & & + 123 & 1,2,3 & 94.48 & 0.57 & [email protected] & [email protected] & [email protected] & [email protected] + 124 & 3,4 & 107.11 & 0.53 & @[email protected] & [email protected] & & + 125 & 2 & 95.24 & 0.49 & [email protected] & [email protected] & & + 126 & 1 & 107.01 & 0.48 & @[email protected] & [email protected] & & + 127 & 1,2,3,4 & 94.75 & 0.41 & [email protected] & [email protected] & [email protected] & [email protected] + 128 & 4 & 108.82 & 0.38 & @[email protected] & [email protected] & & + 129 & 1,2,3,4 & 93.64 & 0.38 & [email protected] & [email protected] & [email protected] & [email protected] + 130 & 3 & 103.41 & 0.38 & @[email protected] & [email protected] & & + 131 & 2,3,4 & 107.11 & 0.37 & [email protected] & @[email protected] & @[email protected] & @[email protected] + 132 & 2 & 95.94 & 0.36 & [email protected] & [email protected] & & + 133 & 1,2,3 & 93.60 & 0.36 & [email protected] & [email protected] & [email protected] & [email protected] + 134 & 1,2 & 99.32 & 0.34 & [email protected] & [email protected] & & + 135 & 1,2,3,4 & 93.62 & 0.34 & [email protected] & [email protected] & [email protected] & [email protected] + 136 & 1,2,3,4 & 95.47 & 0.34 & [email protected] & [email protected] & [email protected] & [email protected] + 137 & 1,2,3 & 109.59 & 0.33 & [email protected] & [email protected] & @[email protected] & @[email protected] + 138 & 3,4 & 99.16 & 0.31 & [email protected] & [email protected] & & + 139 & 4 & 93.97 & 0.31 & [email protected] & [email protected] & & + 140 & 3 & 103.39 & 0.30 & [email protected] & @[email protected] & & + 141 & 3 & 107.37 & 0.28 & [email protected] & [email protected] & & + 142 & 1,2 & 95.14 & 0.27 & [email protected] & [email protected] & & + 143 & 1,2,3 & 107.23 & 0.25 & [email protected] & @[email protected] & & + 144 & 1,2,3 & 108.39 & 0.25 & [email protected] & @[email protected] & [email protected] & @[email protected] + 145 & 4 & 95.90 & 0.25 & [email protected] & [email protected] & & + 146 & 1,2,3 & 94.16 & 0.23 & [email protected] & [email protected] & [email protected] & [email protected] + 147 & 1,2,3 & 111.87 & 0.22 & [email protected] & @[email protected] & [email protected] & @[email protected] + 148 & 1,2 & 99.12 & 0.22 & [email protected] & [email protected] & & + 149 & 4 & 107.41 & 0.21 & @[email protected] & [email protected] & & + 150 & 4 & 103.23 & 0.21 & @[email protected] & @[email protected] & & + 151 & 2 & 95.20 & 0.19 & [email protected] & [email protected] & & + 152 & 4 & 108.66 & 0.19 & [email protected] & @[email protected] & & + 153 & 1,2 & 102.57 & 0.16 & [email protected] & [email protected] & & + 154 & 1,2,3 & 95.22 & 0.16 & [email protected] & [email protected] & [email protected] & [email protected] + 155 & 2,3,4 & 93.33 & 0.16 & [email protected] & [email protected] & [email protected] & @[email protected] + 156 & 1 & 99.90 & 0.14 & [email protected] & [email protected] & & + 157 & 4 & 103.02 & 0.14 & @[email protected] & @[email protected] & & + 158 & 1,2,3 & 96.62 & 0.13 & [email protected] & [email protected] & [email protected] & [email protected] + 159 & 1 & 95.29 & 0.13 & [email protected] & [email protected] & & + 160 & 2 & 107.64 & 0.13 & [email protected] & @[email protected] & & + 161 & 3 & 108.71 & 0.13 & [email protected] & @[email protected] & & + 162 & 2 & 107.42 & 0.12 & [email protected] & @[email protected] & & + 163 & 4 & 94.87 & 0.12 & [email protected] & [email protected] & & + 164 & 2 & 94.98 & 0.12 & [email protected] & [email protected] & & + 165 & 3 & 108.47 & 0.12 & [email protected] & [email protected] & & + 166 & 3 & 95.84 & 0.11 & [email protected] & [email protected] & & + 167 & 2 & 98.34 & 0.11 & [email protected] & [email protected] & & + 168 & 3 & 107.50 & 0.10 & @[email protected] & [email protected] & & + | we used the very long baseline array ( vlba ) and the european vlbi network ( evn ) to perform phase - referenced vlbi observations of the three most powerful maser transitions associated with the high - mass star - forming region : the 22.2 ghz , 6.7 ghz , and 1.665 ghz oh lines .
we also performed vla observations of the radio continuum emission at 1.3 and 3.6 cm and subaru observations of the continuum emission at 24.5 .
two centimeter continuum sources are detected and one of them ( named `` hmc '' ) is compact and placed at the center of the observed distribution of , and oh masers .
the bipolar distribution of line - of - sight ( l.o.s ) velocities and the pattern of the proper motions suggest that the water masers are driven by a ( proto)stellar jet interacting with the dense circumstellar gas .
the same jet could both excite the centimeter continuum source named `` hmc '' ( interpreted as free - free emission from shocked gas ) and power the molecular outflow observed at larger scales although one can not exclude that the free - free continuum is rather originating from a hypercompact region . at 24.5 , we identify two objects separated along the north - south direction , whose absolute positions agree with those of the two vla continuum sources .
we establish that @xmath090% of the luminosity of the region ( @xmath02@xmath110@xmath2 ) is coming from the radio source `` hmc '' , which confirms the existence of an embedded massive young stellar object ( myso ) exciting the masers and possibly still undergoing heavy accretion from the surrounding envelope . |
to obtain important information about a variety of astrophysical sources , astrophysics must move toward observations beyond the electromagnetic spectrum .
several ground based gravitational radiation ( gr ) observatories are already working at full efficiency and collecting data ( e.g. , ligo , @xcite ) .
these detectors are sensitive in a high gr frequency regime ( above a few tens of hz ) , where signals are expected to come from mergers of compact stellar mass objects like neutron stars ( nss ) and black holes ( bhs ) . however , these types of events are infrequent , their rate estimates are burdened with heavy uncertainties ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) and it is not clear whether any merger detections are likely given the current level of instrument sensitivity @xcite .
the future generation of gr detectors includes _ lisa _ , the laser interferometer space antenna ( * ? ? ?
* ; * ? ? ?
* and references therein ) , a joint esa and nasa mission .
it is a space - based all - sky interferometer which will be sensitive to lower frequencies [ hz ] @xmath4(orbital period ) [ s ] . ]
( @xmath5 hz ) than are ground - based gr detectors .
the different frequency regime will allow for the observations of sources of a different nature .
the most promising sources for _ lisa _ are mergers of supermassive and/or intermediate - mass black holes ( i.e. , * ? ? ?
* ) , extreme mass ratio inspirals ( emris ; e.g. , * ? ? ?
* ) of stellar - mass objects into supermassive black holes , and stellar mass compact remnant binaries ( see e.g. , * ? ? ?
* ; * ? ? ?
at first , it was believed that contact w uma systems @xcite would dominate the gr signal in the low frequency range , but later work ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ?
* ) has shown that close double white dwarf ( wd ) binaries are more important at low frequencies , and are a guaranteed source of gravitational waves for _ lisa_. the gravitational wave signal arising from double wds in the galaxy will put constraints on the star formation history of the milky way ( mw ) , as well as the scale height and shape of the thick disc and galactic bulge @xcite .
the galactic double wds will contribute to the _ lisa _ signal as both unresolved and resolved sources . at low frequencies ( @xmath6 mhz ; pertaining to binaries with orbital periods larger than @xmath7 minutes ) , predicted binary numbers are so high that most of the gr signals from individual systems will be unresolved , and they will form a confusion - limited ` foreground ' noise . within that frequency range ,
only gr sources that are relatively close to the detector or are strong gr emitters will stand out above the foreground noise . at higher frequencies ( @xmath8 mhz ; orbital periods @xmath9 minutes ) ,
the numbers of double white dwarfs are relatively small and thus they are expected to be resolved , offering an opportunity to study uncertain parameters associated with binary evolution and wd physics ( e.g. , possible progenitors of sne ia and r crb stars ) .
_ lisa _ will be mostly sensitive to galactic stellar - mass binaries , however a small contribution due to extragalactic systems was also predicted @xcite .
the gravitational wave radiation from coalescing double wds was investigated using smoothed particle hydrodynamics by @xcite .
it is expected that the signal arising from merging double wds would be detectable prior to the coalescence , though would not contribute significantly to the _ lisa _ gravitational wave signal overall , as the galactic double wd merger event rate is predicted to be as low as @xmath10 per century @xcite . among the objects which are expected to be important verification sources for _ lisa _ are the compact am canum venaticorum ( am cvn ) binaries ( see @xcite for gravitational wave strain amplitude estimates of five known am cvn binaries ) .
am cvn systems are a sub - class of cataclysmic variables with @xmath11 hour orbits in which a wd accretes matter via roche - lobe overflow ( rlof ) from what is believed to be a helium - rich white dwarf or a ( semi - degenerate ) helium star @xcite . these close binaries are faint and thus not readily detected electromagnetically , but are expected to be observable in gravitational waves ( 25 am cvn systems have been confirmed to date ) . besides double wd binaries
, other close compact binaries including nss and bhs have been studied by other groups ( e.g. , * ? ? ?
* ; * ? ? ?
some few tens of these systems with orbital periods within the _ lisa _ sensitivity range are known in the galaxy .
there are also @xmath12 ultracompact x - ray binaries , which consist of a wd transferring matter through rlof to a ns on @xmath13 minute orbits @xcite .
double nss are also known , however only one such system is observed with a period ( of 2.4 hrs ) within the _ lisa _ sensitivity range @xcite .
close binaries with bhs ( like bh - bh ,
bh - ns , bh - wd ) have yet to be observed , but they are predicted to populate the galaxy in smaller numbers than systems with nss and/or wds . previous work on galactic compact binary populations in context of _ lisa _ has been done with the exclusion of rlof systems @xcite or with the entire rlof @xmath14 detached populations combined @xcite .
in addition , @xcite studied am cvn systems separately using an analytical approach , finding no significant increase in the confusion - limited noise for space - based gr detectors with the addition of rlof binaries , while @xcite calculated the gr amplitude of resolved wd binaries using population synthesis methods ( both detached and am cvn systems ) , finding that _ lisa _ would resolve a total of @xmath15 galactic double wds ( @xmath16 detached , @xmath17 am cvn ) .
however , a more recent study which takes into account the local space density of am cvn stars predicts that only @xmath18 am cvn binaries will be resolved with _
this estimate is still @xmath19 orders of magnitude above the current number of am cvn systems which have so far been detected . in modeling our stellar population
, we use an updated population synthesis method with recent results on mass transfer / accretion in compact object binaries , that is significantly different and more complex than the methods used in previous calculations . in this study
, we assess the importance of detached and rlof compact binary populations in the galactic disc and bulge on the overall _ lisa _ sensitivity . in
2 we describe our binary modeling techniques and signal calculations , in 3 we show the physical properties ( masses , orbital periods ) of our double wd population , and show the most important evolutionary channels which contribute to the overall _ lisa _ signal arising from the four galactic sub - populations ( detached disc , rlof disc , detached bulge , rlof bulge ) . in
4 we show the _ lisa _ gr signals , estimate the likelihood of source resolvability , and comment on the detectability and selection effects of the potentially resolvable systems within our galaxy . in 5
we compare our results with those of previous studies , and in 6 we give a summary of our findings .
in this study , the combined galactic population of rlof and detached compact remnant binaries with orbital periods spanning the _ lisa _ band is considered .
we use the startrack population synthesis code described in detail in @xcite to evolve primordial binaries within the galaxy from the zero - age main sequence ( zams ) and calculate their properties .
the code has undergone a number of important updates since @xcite , including detailed mass transfer calculations .
recent results on mass accumulation in context of ultracompact x - ray binary formation are given in @xcite , while for double wd binaries in @xcite .
we evolve the field population of single and binary stars ( @xmath20 % binarity ) with solar - like metallicity ( @xmath21 ) .
single star evolution is followed from the zams employing modified analytic formulae and evolutionary tracks from @xcite . for the evolution of binary stars ,
the calculations also include full orbital evolution with tidal interactions , magnetic braking , supernova natal kicks and gravitational radiation emission ( see * ? ? ?
* for formulae ) .
zams masses ( @xmath22 ) span the mass range @xmath23 .
single stars and binary primaries ( @xmath24 ) are drawn from a 3-component broken power law initial mass function @xcite , and secondary masses ( @xmath25 ) are obtained from a flat mass ratio distribution @xmath26secondary / primary ( e.g. , * ? ? ?
we note as well that binary interactions leading to mass transfer events may alter the course of evolution for stars of a given initial mass .
initial orbital separations span a wide range up to @xmath27 r@xmath28 and are drawn from a distribution which is flat in the logarithm , while initial eccentricities are taken from a thermal - equilibrium eccentricity distribution ( see * ? ? ?
* ; * ? ? ?
progenitors of all binaries are initially formed on eccentric orbits .
tidal interactions between binary components circularize orbits before the first rlof in a given progenitor system occurs .
however , for a small fraction ( 10% ) of double wd progenitors that are found initially on very eccentric orbits , the first rlof is encountered when the orbit is still eccentric .
in such a case we assume an instant circularization at periastron ( @xmath29 and @xmath30 ) , and follow with the rlof calculation . at the time of double wd formation all of the orbits
are circular .
eccentric wd binaries are expected to arise from dynamical interactions in globular clusters @xcite , and though we do not consider them here could provide a unique opportunity for learning about wd structure with _
_ @xcite . for the population of galactic disk binaries
, we use a continuous star formation rate for 10 gyr , while for the bulge population we use a constant star formation rate for the first gyr , and no star formation thereafter ( entire mw age is 10 gyr ) .
this translates to a present galactic star formation rate of @xmath31 m@xmath28 yr@xmath32 ( disc mass @xmath33 m@xmath28 with a constant star formation rate for 10 gyr @xmath34 m@xmath28 yr@xmath32 ; see 2.2 ) , which is within reasonable agreement of the current galactic star formation rate estimate of @xmath35 m@xmath28 yr@xmath32 @xcite .
we note however that it has been suggested that the star formation rate of the mw has been ( mostly ) decreasing exponentially with time , only reaching @xmath36 at the current epoch ( * ? ? ?
* ; * ? ? ?
* see sect .
2.2 ) ; the integrated mass in formed stars being closer to @xmath37 m@xmath28 .
since the star formation history of the mw disc and bulge is not precisely known , we choose to model the galaxy with simple star formation histories ( 1 gyr - long ` continuous burst ' ( bulge ) , and constant for 10 gyr ( disc ) ) , which to first order is a reasonable representation of the global star formation history of the galaxy . the most uncertain phase in close binary evolution affecting the orbital separation for low- and intermediate- mass stars is the common envelope ( ce ) phase @xcite .
close binaries are expected to go through at least one common envelope event , and there are currently a few prescriptions of ce evolution in the literature @xcite . in our simulations the ce phase is treated using energy balance as discussed in @xcite , in which the orbital energy of the binary is diminished at the expense of the unbinding of the donor envelope .
the post - ce separation is governed by the parameters @xmath38 @xcite , a function of the donor s structure , and the highly uncertain @xmath39 , the efficiency with which the binary orbital energy is used to unbind the envelope .
we have chosen to use @xmath40 .
the effects of using other ce efficiencies have not been fully explored here , though lower ce efficiencies result in closer post - ce binaries , and a higher number of stellar mergers . at the current age of the galaxy ( 10 gyr )
we extract all binaries containing two compact remnants : wds , nss and bhs in the gravitational frequency range : @xmath41 hz , which encompasses the _ lisa _ sensitivity range ( orbital periods of 5.6 hr
2s ) . henceforth when we refer to `
_ lisa _ binaries ' , we are referring to binary systems in our study within this gr frequency band .
as previously stated , we consider only white dwarf binaries when calculating the _ lisa _ signal .
there are 5 types of wds considered in the evolution : helium ( he wd ) , formed by stripping the envelope off a hertzsprung gap or red giant low - mass star ; carbon - oxygen ( co wd ) , formed from progenitors with masses @xmath42 ; oxygen - neon ( one wd ) , formed from progenitors with masses @xmath43 ; hybrid ( hyb wd ) , having a co - core and a he - envelope , formed via the stripping of the envelope from a helium burning star , and hydrogen ( h wd ) , formed by stripping the envelope from a very low mass ( @xmath44 ) main sequence ( ms ) star . in our _ lisa _
gr calculations , we only consider the first four types of wds , as hydrogen white dwarfs are more representative of brown dwarf - like objects and thus we will from now on refer to them as brown dwarfs . for
the current study , the brown dwarfs ( formed through binary evolution ) do not play a large role in contributing to the _ lisa _ gr signal due to their low mass ( see 5 ) .
we note however that in older stellar populations , the fraction of binaries which contain h wds becomes more significant ( see * ? ? ?
* for the calculation of the _ lisa _ signal from mw halo double white dwarfs , which includes hydrogen white dwarfs ) . in compact binaries ,
gravitational wave emission is a strong source of angular momentum loss .
the average rate of orbital angular momentum loss due to gr for @xmath45 binaries is calculated from @xcite @xmath46 where @xmath47 is the gravitational constant and @xmath48 is the speed of light . at every time step in our calculations
the evolving system is checked for rlof . upon reaching contact ( rlof ) , we assume that mass - transfer is gr - driven and the donor ( @xmath49 ) mass transfer rate is calculated via : @xmath50 where rlof stability is determined by the parameter @xmath51 ( see i.e. , * ? ? ?
* ) @xmath52 @xmath53 is the orbital angular momentum of the binary , @xmath54 is the fraction of transferred mass accreted by the wd of mass @xmath55 ( @xmath56 for stable rlof ) , @xmath57 , and @xmath58 .
the radius mass exponent for the donor @xmath59 is obtained from stellar models in each time step ( see * ? ? ? * ) and is @xmath60 for wd donors in a phase of stable mass transfer .
stable rlof between two wds leads to an increase in orbital period where dynamically unstable rlof leads to a merger .
see @xcite for a more detailed description of treatment of mass transfer / accretion phases in startrack . in our simulations
both detached and rlof systems have been distributed about the disc and bulge separately , assuming no correlation between position and age of the system . for the current study ,
we have neglected the halo population of double wds , but these systems have been investigated in another work @xcite . the density distribution of the galactic disc is taken to have the following form : @xmath61 in cylindrical coordinates , and the density distribution of the bulge has the form : @xmath62 in spherical coordinates ( bulge scale length @xmath63 ) .
@xmath64 and @xmath65 represent the total numbers of simulated double wds in the disc and bulge , respectively .
we have calibrated our results using stellar galactic disc and bulge masses from @xcite , with masses of @xmath66 m@xmath28 and @xmath67 m@xmath28 , respectively .
we have chosen @xmath68 and @xmath69 for the disc scale length and scale height while @xmath70 with a radial cut - off of @xmath71 pc for the bulge @xcite .
while we note that even a scale height of @xmath72 pc would be a reasonable choice for disc wds @xcite , we have chosen to use @xmath73 pc so that our results are more readily comparable to those of previous studies @xcite . our original _ lisa _
double wd binaries which were born directly from startrack ( @xmath74 ) were the result of the evolution of @xmath75 zams binaries .
since evolving the whole galactic population of _ lisa _ binaries is time - intensive , an interpolation scheme was developed in order to scale our number of startrack binaries to match those of the milky way bulge and disc by stellar mass .
probability distribution functions ( pdfs ) were constructed for different evolutionary channels as a function of the primary and secondary masses and gr frequency . for rlof binaries , it was only necessary to construct pdfs as a function of the masses since the orbital period , and hence gr frequency , was then uniquely determined by the masses .
we chose to construct pdfs ( instead of simply scaling the original binaries by a factor of @xmath76 ) so that our systems would span a range of frequencies and masses pertinent to said formation channel .
the pdfs were constructed using a kernel density estimator that best replicated the characteristics of the population @xcite .
the kde package for matlab was used to generate the specific pdfs .
this technique was particularly successful for well - populated evolutionary channels , where we could be sure that the range of masses and frequencies was well covered .
for some of the least - populated channels , the full range of binary masses and frequencies was not covered sufficiently to yield a smooth extrapolation of the data .
thus , apparent clumps of binaries that are comparable in number to our scaling factors are attributable to an under - sampling of the frequency and mass range for these rare evolutionary channels .
the gravitational wave signal from the _ lisa _ data stream will be a time delay interferometry ( tdi ) variable in which the phases of the laser signals at each vertex of _ lisa _ are combined with the phases of other signals at delayed times in order to reduce the laser phase noise and accommodate the varying armlengths of the constellation of the spacecraft @xcite . at low frequencies
, the signal at any given vertex can be very well approximated by the michelson signal : @xmath77 where @xmath78 is the wave metric and @xmath79 and @xmath80 are the unit vectors pointing along the two arms that join at the vertex .
the transfer frequency is the frequency at which point the gravitational radiation wavelength becomes comparable to the armlength of _
lisa_. for high ( e.g. , above the transfer frequency ) gr frequencies , the period of the gravitational wave is then less than the time it takes for light to propagate between _ lisa _ spacecraft detectors , and the low frequency approximation breaks down .
the transfer frequency is given by @xmath81 hz , where @xmath82 is the _ lisa _ arm - length taken to be @xmath83 . at frequencies above or near @xmath84 a more accurate representation of the _ lisa _
signal is given by the _
rigid adiabatic approximation _
@xcite , but detailed analysis @xcite have shown the low - frequency approximation should be adequate for most tasks at frequencies below @xmath85 . in this work
, we have used the rigid adiabatic approximation to calculate the _ lisa _ signal for sources with frequencies above 0.003 hz .
space - borne gravitational wave observatories like _ lisa _ are constantly in motion , changing their relative speed and aspect with respect to astrophysical sources on the sky .
the sensitivity of the observatory to different gravitational wave polarizations as a function of sky position is reflected in the antenna beam patterns , which have the highest gain in the direction perpendicular to the plane of the interferometer .
as the observatory moves in its orbit , the gravitational wave signal is modulated in amplitude , frequency , and phase .
frequency ( doppler ) modulation arises from the relative motion of the observatory and the source , amplitude modulation arises from the sweep of the anisotropic antenna pattern on the sky , and phase modulation results from the detector s changing response to the gravitational wave polarization state .
the michelson signal can be calculated using eq .
( [ michelson ] ) by either holding the source fixed and putting all the motion of the detector into @xmath79 and @xmath80 @xcite , or by holding the detector fixed and putting the motion into the source @xcite .
both approaches yield the same michelson signal in the low - frequency limit : @xmath86}\ ] ] where @xmath87 is the amplitude modulation , @xmath88 ( where @xmath89 is the binary orbital period ) is the gravitational wave frequency for systems with zero eccentricity , and @xmath90 is the initial phase of the wave at @xmath91 . the polarization phase , @xmath92 , represents the phase modulation .
the doppler phase , @xmath93 , gives the frequency modulation and the amplitude modulation is described by : @xmath94 .
the sensitivity factors , @xmath95 and @xmath96 are complicated functions of the position angles and orientation of the binary , as well as the position of _ lisa _ in its orbit ( see * ? ? ? * ; * ? ? ?
* for specific forms of these functions ) .
the plus and cross polarization amplitudes are given by : @xmath97 where @xmath98 is the distance to the binary , @xmath99 is the angle of inclination , and the chirp mass is @xmath100 .
we have used the approach of @xcite to calculate the timestreams , which are then added to produce the total observatory data stream , which is then fourier transformed to produce the illustrated frequency domain representation of the double wd galactic foreground . the _ lisa _ instrument noise is simulated by assuming the power spectral density of the noise is made up of position ( or shot ) noise given by @xcite : @xmath101 and an acceleration noise ( converted to strain ) given by : @xmath102 these separate components are combined according to : @xmath103 where @xmath84 is the transfer frequency .
we roll off the acceleration below @xmath104 , so that @xmath105 . in reality , the _ lisa _ noise will probably not follow this simple power law all the way down to our choice of @xmath106 , but will begin to rise at a higher frequency below 0.1 mhz .
out of our total population of compact remnant binaries with gravitational wave frequencies within the _ lisa _ sensitivity range , we note that @xmath107% are double wds , @xmath108% are binaries involving a white dwarf and a brown dwarf ( wd - bd ) , while other types of binaries make up the rest ( i.e. , @xmath109 % of binaries are ns - wd binaries ; less than @xmath110 % are ns - ns ) .
specific predictions for double compact objects with nss and bhs are discussed in @xcite .
relevant for the current study of double white dwarfs , we find @xmath111 _ lisa _ binaries in the disc@xmath14bulge population ( he , co , one and hybrid wds ) . ] . out of this population @xmath112 ( 73 % ) and @xmath113 ( 27 % ) .
we find @xmath114 ( 25 % of total ) of disc binaries are detached and @xmath115 ( 47 % of total ) of disc binaries are rlof , while for the bulge @xmath116 ( 12 % of total ) are detached and @xmath117 ( 16 % of total ) are rlof .
the total number of detached binaries among the total disc@xmath14bulge _ lisa _ population is @xmath118 ( 37 % ; @xmath119 and @xmath120 for the disc and bulge respectively ) while for rlof , often neglected in previous studies , the number is @xmath121 ( 63 % ; @xmath115 and @xmath122 in the disc and bulge , respectively ) . within our rlof population , @xmath123 % ( @xmath124 ) are am cvn - like systems ( a wd accreting from a helium or hybrid wd ) , the majority of which are cowd - hewd binaries .
the number of am cvn systems which we find to currently exist in the mw can be compared to the estimate of @xcite .
however , we remind the reader that in our population synthesis , although we do form am cvn stars through the helium star channel , we only include _ binaries consisting of two degenerate objects _ in the final results ( see * ? ? ? * for a description of am cvn formation channels ) .
thus , we can not perform a direct comparison with the am cvn population of @xcite .
@xcite find the current number of galactic am cvn ( double wd channel only ) to range between ( @xmath125 ) @xmath126 , depending on the assumed effective tidal coupling ( see their 3.4 ) , and they find corresponding space densities of ( 0.4 - 1.7 ) @xmath127 pc@xmath128 , respectively .
thus , we find that our am cvn number estimate ( @xmath124 ) is in reasonable agreement with that of @xcite if in fact the double wd evolutionary pathway is an efficient channel for creating these systems . in the study of @xcite
, it was determined that population synthesis studies were overestimating the space densities of am cvn binaries , and that the true local space density is @xmath129 pc@xmath128 , based on spectroscopic observations of 6 sdss - i am cvn stars . adopting the disc - like density distribution as discussed in 2 , we find that the local space density of am cvn binaries from our population synthesis model is @xmath130 pc@xmath128 , which is still an order of magnitude above the most up - to - date observational estimate .
it is becoming more clear that mass transfer between two wds may not be as efficiently sustained as has been assumed in most population synthesis studies , and that a large fraction of these systems should merge upon reaching contact rather than enter the am cvn phase @xcite .
we note that in our calculations , the total population of wd binaries with gr frequencies between @xmath131 hz comprises roughly 6% of the _ total _ startrack galactic population of double wds .
while we include all rlof double wd systems that currently exist in the mw in our study , only 2.3% of galactic detached double wd systems are accounted for , given the orbital period cut - off of @xmath132 5.6 hrs . in tables 1 , 2 , 3 and 4
, we show the evolutionary history ( formation channels ) of the most common _ lisa _ double wds .
we indicate the contribution ( percentages ) of said channel to the population of _ lisa _ wd binaries from that particular galaxy component ( disc , tables 1 & 2 , or bulge , tables 3 & 4 ) .
the characters and numbers in parentheses in the central column of the tables represent the formation channel histories , and recount various stages of stellar evolution of the primary and secondary star progenitors ( see table 1 caption for description ) . in our formation channel notation , for a ` co -
he ' binary the co white dwarf is the first formed wd , which is not necessarily the more massive wd ( thus is not necessarily the primary star ) . in general ,
the detached systems have evolved from progenitors in which the stars initially have comparable masses , thus first rlof mass transfer is more often dynamically stable and the binary does not undergo a ce at this stage ( such as he - he detached double wds ) .
typically the detached double wd progenitors go through only one ce event , where as in many cases the rlof binaries undergo two ce events .
we wish to point out that in the case of several rlof binaries ( e.g. , those involving hybrid donors ) , there is a short - lived detached phase preceding the long - lived rlof phase .
for example , in the evolutionary history of a cohyb - r1 binary , the progenitor system only spends @xmath133 myr as a detached co - hyb wd after the second ce phase before driven to contact via gr .
once contact is reached however , the binary will spend a long time ( gyr ) as a stable rlof system .
this explains why the cohyb - r1 channel shows up in table 2 , but its predecessor cohyb detached channel is absent in table 1 ( cohyb - d1 binaries only account for @xmath134 of the disc formation channels for _ lisa _ double wds ) . in figure [ dndfdisdet ] ,
figure [ dndfdisrlo ] , figure [ dndfbuldet ] and figure [ dndfbulrlo ] we show in ( arbitrary ) colour the number densities of the most common formation channels for the detached and rlof populations in the disc and bulge .
we do not include the less - populated formation channels in the plot or it would be very difficult to distinguish between channels .
we do however include some formation channels involving brown dwarfs , since even though these binaries do not significantly affect the overall gr signal ( and we have ignored them in the signal calculation ) , they are relatively abundant , specifically in the bulge population .
it is expected that the disc would host binaries which have evolved from a wider variety of formation channels than the bulge , since the age of the disc binaries extends over a large range : @xmath132 few hundred myr to @xmath135 gyr .
note that high frequency systems only exist in the galactic disc .
there is a depletion of very high frequency ( few minutes orbital period ) rlof systems in the bulge since the present bulge population only contains double white dwarfs with long - lived ( @xmath136 gyr ) formation histories .
all of the heavier ( one+co , co+co , co+hybrid ) binaries originating from more massive progenitors which were born during star formation in the bulge have either long since merged , or in the case of rlof binaries , the stars has long since reached contact and are now exchanging mass in rlof on slowly expanding orbits . at which point we no longer follow the evolution . ]
once any system encounters stable rlof the binary s orbital period increases as a consequence of conservation of angular momentum under continued mass exchange , since the less - massive ( larger ) wd is losing matter to the more massive wd . for the galactic disc , some wd binaries are still relatively young , and are in an earlier phase of stable rlof and thus we ` catch ' these mass transferring systems at shorter orbital periods ( gr frequencies above @xmath137 mhz ; @xmath138 min ) .
_ typical detached evolution _ : detached double co - wds are formed through a variety of evolutionary channels .
for the most prominent co - co channel ( coco - d1 , see table 1 ) , one particular example of evolution proceeds as follows : two ms stars ( @xmath139 and @xmath140 ) start out with a @xmath89 of @xmath141 days and an eccentricity of @xmath142 when the mw is @xmath143 myr old . at @xmath144 myr ,
the more massive star ( s1 ) begins to evolve off of the ms and becomes a hertzsprung gap star .
shortly thereafter s1 fills its roche lobe and begins to transfer matter to the companion ( interaction between component stars aids in circularizing the orbit ; @xmath89 @xmath137 days ) .
rlof is dynamically stable , and since s1 quickly becomes the less - massive star during mass transfer @xmath89 increases ( to @xmath145 days ) , and rlof stops at @xmath146 myr when s1 is a red giant ( @xmath147 ; core mass @xmath148 ) and s2 ( @xmath149 ) is still on the ms .
s1 continues to evolve and shortly thereafter becomes a ( naked ) helium star ( @xmath147 ) , as it has lost the remaining part of its depleted ( by prior rlof ) envelope . at @xmath150 myr
, s2 has evolved off of the ms to become a hertzsprung gap star . by @xmath151 myr
, s2 has now evolved into an early agb star ( @xmath152 ) , and the orbital period has decreased to @xmath153 days due to tidal interactions ( expanding s2 ; s1 is still a helium star ) .
s2 then fills its roche lobe , and since now the mass ratio is rather large ( @xmath154 ) the mass transfer is dynamically unstable , leading to a ce phase . after the ce event the donor ( s2 ) has become a hertzsprung gap helium star ( @xmath155 ) and @xmath89 has decreased by more than @xmath156 orders of magnitude to @xmath157 hours .
the orbit then starts to decay slightly due to angular momentum losses associated with gr emission , and there is a brief rlof phase in which the hertzsprung gap helium star ( s2 ) loses @xmath158 , half of which is gained by s1 ( still a helium star , now @xmath159 ) . when the mw is @xmath160 myr old , s2 finally becomes a co wd ( primary wd ) .
s1 continues to burn helium and becomes a hertzsprung gap helium star at @xmath161 myr ( @xmath162 hours ) . within 20 myr
, s1 then becomes a low - mass co wd ( secondary wd ) finishing its helium star evolution . at @xmath163 gyr
( present ) the system is found as a detached double co wd with component masses of @xmath164 & @xmath159 , and @xmath165 hours ( @xmath166 mhz ) .
the two wds will be brought together in @xmath167 myr due to gr emission .
the system is expected to merge due to dynamically unstable mass transfer with a combined mass of @xmath168 ( pre - merger orbital period of @xmath169 minutes ) .
the total mass of the merged system does not exceed the canonical chandrasekhar mass and so it is an unlikely type ia supernova progenitor , though may result in the formation of a massive rapidly rotating wd @xcite .
_ typical rlof evolution _ : in figure [ rlof ] we show a representative evolution through rlof for one of the galactic double wds , a cohe - r1 channel system ( see table 2 ) .
the primordial binary consists of two low - mass ms stars ( @xmath170 and @xmath171 ) on a wide ( @xmath172 year ) and eccentric ( @xmath173 ) orbit , born when the mw is @xmath174 myr old . at @xmath175 myr the more massive star ( s1 ) begins to evolve off of the ms .
the orbit begins to shrink due to tidal interactions and by @xmath176 myr , s1 is an early agb star , the orbit has circularized ( @xmath45 ) and the period has decayed to @xmath177 years . within @xmath178 myr s1
becomes a late agb star .
magnetic braking ( convective envelope of s1 ) dominates angular momentum loss and the orbit decays slightly .
s1 fills its roche - lobe ( @xmath179 myr ) and mass transfer is unstable , leading to a ce phase and a drastic decrease in orbital period ( from @xmath180 years to @xmath181 days ) .
s1 then becomes a cowd ( primary wd , @xmath182 ) . at @xmath183 myr ,
s2 evolves off of the ms .
the orbit decays due to magnetic braking ( convective s2 ) , and at @xmath184 myr s2 initiates a ce phase resulting in another prominent period decrease ( @xmath185 days to @xmath186 mins ) and the loss of its hydrogen - rich envelope .
s2 then becomes a he wd ( @xmath187 ) .
gr causes the wds to come into contact a few myr later ( @xmath188 mins ) , and the secondary ( s2 ) fills its roche - lobe .
the mass ratio is only @xmath189 thus rlof is stable , and an am cvn system is born .
initially mass transfer is quite rapid ( see figure [ rlof ] top panel ) , but a slower mass transfer phase follows .
the binary is found at @xmath163 gyr with @xmath190 hour ( @xmath191 mhz ) and component masses of @xmath192 and @xmath193 @xmath194 for the co and he wds , respectively . many of the @xmath195 observed am cvn binaries parameters are uncertain , and detection of these systems is biased against the typical , longer - period systems with lower mass transfer rates .
however this particular simulated binary s properties are somewhat similar to the very few long - period am cvn binaries which have been detected ( e.g. , sdss j1552 & ce 315 ) , though the nature of the donors is still unknown @xcite ) .
characteristic properties of the entire detached population are shown in figure [ popd ] .
disc systems are found at all orbital periods ( top panel ) within the _ lisa _ sensitivity range , with a drop in number at shorter periods due to the fact that once systems reach contact , double wds either fall into the rlof population or otherwise merge and drop out of the _ lisa _ population altogether .
secondary ( @xmath196 , less massive wds , middle panel ) lie predominantly in four regions : @xmath158 ( he wds with co primaries ) , a prominent peak near @xmath197 ( mostly he with some hybrid wds with mostly co or he companions ) , a small clump near @xmath198 ( co secondaries with co primaries ) and a short peak at @xmath199 ( co or sometimes one secondaries with either co or one primaries ) .
the he secondaries with very low mass ( @xmath200 ) derive from the same formation channels as cataclysmic variables ( similar systems have been discussed in @xcite ) , in which a co wd accretes from a low - mass main sequence star for several gyr ( note the higher number of these systems in the galactic bulge ) .
eventually the donor becomes fairly exhausted of its hydrogen supply , and meanwhile has built up a significant helium core .
the donor eventually is depleted of so much mass that it reaches the hydrogen burning limit ( 0.08 @xmath194 ) , and thus becomes degenerate ( in this case , helium rich ; a helium wd ) .
many of the systems whose secondary mass peaks near 0.3 @xmath194 are progenitors of am cvn stars ( cohe ) , while most of the binaries involving more massive progenitors ( co , one ) will likely merge once they reach contact .
primary wds ( @xmath201 , more massive wd ; bottom panel ) display a peak at @xmath202 , though most primaries have masses @xmath203 , and a smaller number of systems are quite heavy ( @xmath204 ) .
the low mass peak is comprised of he wd primaries , with he secondaries .
the majority of the rest are co wds , with either he or co secondary companions . in both the middle and bottom panels ,
we note that there are relatively fewer heavy wds in the bulge population .
this is because all of the more massive progenitors ( e.g. , co - co ) have had a sufficiently long enough time to evolve off of the main sequence , encounter 1 or 2 ce phases , reach contact ( gr is also more efficient at bringing together two co wds than two he wds ) and have either merged out of the population altogether or in some cases , have moved into the rlof population .
a two - dimensional greyshade plot is presented in figure [ greydet ] , where we show the relation between secondary and primary masses of _ lisa _ binaries in the galactic disc , presented in terms of relative total percentages of _ lisa _ double wds .
characteristic properties of the entire rlof population are shown in figure [ popr ] .
the rlof systems with the maximum @xmath89 ( lowest gr frequencies ) appear at @xmath205 minutes ( log(@xmath206)@xmath207 ) .
the shape of the period distribution ( top panel ) is due to the fact that most systems spend the majority of their time at periods of @xmath10 hr during the dynamically stable rlof phase .
as indicated previously , 99 % of rlof systems are am cvn ( a wd accreting from a he or hybrid wd ) . for most wd binaries , at the onset of stable rlof orbital periods are on the order of only a few minutes , and mass transfer rates are initially high ( see figure [ rlof ] , top panel ) .
the donor quickly becomes exhausted of most of its mass , the mass transfer rate decreases and the period continues to increase , though at a much slower rate . in the middle panel
we show wd secondary masses , which is strongly peaked at @xmath208 , with a slight drop - off to @xmath209 .
these wds are comprised of mostly he wds , some hybrid wds and a small number of co wds .
the ` pile - up ' of wds at @xmath210 stems from the fact that once a am cvn wd - wd progenitor reaches contact , initially the mass transfer rates are high ( @xmath211 yr@xmath32 ) and the donor is rapidly exhausted of mass , thus it is rare to catch am cvn secondaries which are relatively massive ( @xmath212 ) . in the bottom panel
we show the distribution of masses among the primary wds , which are mostly co wds , with a small contribution of one wds ( masses @xmath213 ) .
we show in figure [ greyrlo ] a two - dimensional greyshade plot of rlof _ lisa _ binaries in the galactic disc , presented in terms of relative total percentages of _ lisa _ double wds .
during a one - year observation period , the width of a resolvable frequency bin is @xmath214 .
below 1 mhz , there are hundreds to thousands of binaries per resolvable frequency bin , causing the signal to be confusion - limited . in our simulation , below @xmath215 mhz the only contributing binaries to the _ lisa _ signal are the detached double white dwarfs .
the _ lisa _ spectra ( amplitude densities , or spectral amplitudes @xmath216 ) for the galactic double wds were calculated using the sophisticated _ lisa _ simulator of @xcite , and are shown in figure [ spectra ] .
close binaries which evolve to contact , such as precursors to the am cvn population or progenitors of white dwarf mergers , will spend a very short time ( @xmath217 myr ) at gr frequencies @xmath218 hz ( @xmath219 minutes , provided they reach these short periods ) , and this is reflected in our original startrack calculations .
thus , we artificially truncate the signal above 0.01 hz since we simply do not have a large enough number of startrack binaries above this frequency in order to accurately extrapolate them to the total number of galactic systems here .
however , we expect the number of _ lisa _ systems above 0.01 hz to be @xmath220 . in the bottom panel ,
we show the spectra from the four galactic sub - populations , while in the top panel we show the full spectra for the combined population ( grey ) with a running median over 1000 frequency bins ( white ) .
the white curve represents our galactic double wd foreground ( see 4.2 ) .
we also show the foreground estimates of @xcite ( dotted pink ) and @xcite ( dashed orange ) for comparison .
the foreground of @xcite was artificially truncated beyond @xmath19 mhz , above which individual binaries were expected to be resolvable in that work .
it immediately comes to attention that the bulge binaries ( green and mauve ; lower panel ) do not contribute to the gr signal above log(@xmath206 ) @xmath221 ( @xmath222 mhz ) for reasons discussed in 3.1 .
also , the rlof binaries do not contribute to the signal at low frequencies ( below ( log(@xmath206)@xmath207 , @xmath223 mhz ) since these are binaries which have been in a state of rlof for a prolonged period of time , and are on very slowly expanding orbits ( largest orbital periods @xmath205 minutes ) .
additionally , once the donor mass drops below 0.01 @xmath194 , we stop the calculation thus these very low - mass , low frequency systems do not contribute to the calculated _ lisa _ signal .
the rlof disc signal ( blue ; lower panel ) becomes comparable to the detached disc signal ( red ; lower panel ) at frequencies above @xmath132 log(@xmath206)@xmath224 ( @xmath225 mhz ) . in our calculations , the am cvn binaries spend only a short period of time ( @xmath226 myr ) as detached double white dwarfs , where as they may spend @xmath132 gyr in the rlof phase .
this is likely a consequence of the common envelope prescription that we have employed , and most am cvn systems go through a double ce ( cohe - r1 channel ) .
the @xcite ce prescription with @xmath40 produces relatively close post - ce white dwarfs , and so it does not take long for the stars to be brought into contact via gravitational radiation once the detached double white dwarf has been formed . in the upper panel , we show our combined signal ( grey ) and the median foreground ( white line ) alongside the astrophysical foreground estimates of @xcite and @xcite , both of which include detached and am cvn binaries .
note that our foreground is significantly lower than the classic @xcite estimate at low frequencies , and is more comparable to the curve of @xcite in this regime ( see below ) .
our foreground curve has been smoothed over 1000 bins , but some ` bumpiness ' still exists at higher frequencies due to low number statistics .
in contrast to some previous studies where the signal has been truncated at higher frequencies , we show the curve over a significant portion of the lisa bandwidth ; up to 0.01 hz . since our original startrack binary population was scarcely populated at high frequencies , consequentially we could not generate smooth formation channel pdfs ( 2.2 ) for high - frequency systems .
our level of foreground signal is below that of @xcite for gr frequencies @xmath227 hz .
this is in part due to the fact that at low frequencies , there is an overall lack of _ lisa _ binaries , since a large fraction of our _ lisa _ binary population are rlof systems and hence do not exist at the low end of the frequency spectrum .
higher than ours . ]
the median chirp mass of our double white dwarfs at frequencies below @xmath228 hz is @xmath229 m@xmath28 ( disc ) and 0.31 m@xmath28 ( bulge ) . despite the large number of rlof systems in part of the low frequency , confusion - limited region of the galactic gravitational wave spectrum ( figure [ dndfdisrlo ] and figure [ dndfbulrlo ] ) , the amplitude of the gravitational wave signal at low frequencies
is dominated by the detached double wds .
mass - transferring binaries at the lower frequencies have undergone a significant amount of mass transfer , which leaves them with lower chirp masses than the detached binaries .
as discussed previously , rlof systems do not occur with orbital periods greater than @xmath230 hours ( @xmath231 mhz ) .
consequently the gravitational wave amplitude of the rlof systems ( both from the disc and from the bulge ) does not contribute very strongly to the overall foreground noise in the low - frequency regime .
the gr amplitude from the bulge is smaller than that of the disc for both ( corresponding ) detached and rlof components of disc , primarily due to the fact that the stellar mass of the bulge is @xmath232 the mass of the disc , and contains fewer _ lisa _ double wds .
the ` transition frequency ' is said to correspond to the gr frequency at which the gravitational wave spectrum transitions from confusion - limited to individually resolvable sources .
the exact conditions which are necessary in order for this to occur ( one possibility would be when the average number of binaries per frequency bin drops below 1 ) are not completely clear .
the addition of rlof systems to the detached population will increase the frequency at which the average number of binaries per bin drops below 1 , simply by adding more sources to the population .
this is particularly so since rlof binaries are able maintain short orbital periods ( for longer then their detached counterparts ) once contact has been established .
furthermore , as the transition frequency increases , the spread of the signal due to the motion of _ lisa _ also increases , thus the signals will continue to overlap at higher frequencies than might be inferred from a simple analysis of the number of binaries per resolvable frequency bin . for a simple detached system in a circular orbit whose orbital period is evolving solely due to the emission of gr , seven parameters are needed to fully characterize the signal .
basic estimates from information theory describe how the limited amount of information in the _ lisa _ data stream can be mapped into an equivalent amount of information about detected binary systems .
a common rule of thumb is the `` three - bin rule '' : at least three ( otherwise unpopulated ) frequency bins of _ lisa _ data are needed to resolve a seven - parameter binary source ( see * ? ? ?
* and references therein for a discussion ) .
one can reasonably expect that a mass - transferring system will require more parameters to characterize the gr signal , and consequently more frequency bins to individually resolve the source .
therefore , the average number of binaries per frequency bin is not a particularly good indicator of how the transition frequency changes with the addition of the rlof population . for the sake of comparison with previous studies ,
we have estimated the transition frequency using the 3-bin rule method .
the point at which _
lisa _ binaries _
start _ to occupy one source per three frequency bins is found to be 1.5 mhz for the detached _ lisa _ binaries only , which is comparable to @xcite and @xcite , who found a transition frequency of @xmath233 mhz see also ( * ? ? ?
* for a discussion of the inclusion of am cvn binaries in the confusion limit calculation ) .
however we find that when we consider the entire _ lisa _ population of double wds , the transition frequency is increased to 2.4 mhz .
we would like to however point out that these frequencies do not represent the frequencies above which one expects to resolve all _ lisa _ binaries ; there are still a number of populated bins at higher frequencies . the 3-bin rule method of estimating the frequency at which binaries start to become resolved is not a robust method , since it is expected that at these higher frequencies , the frequency evolution due to mass transferring binaries can contribute a significant additional spreading of the signal .
further , this method does not take into account important criteria such as doppler , phase and amplitude modulation , and one should consider alternative ( more technical ) methods in computing the threshold frequency region ( e.g. , chapter 2 of * ? ? ?
* ) . in the following subsection
, we discuss an approach other than the 3-bin rule of thumb to identify the population of resolved sources .
instead of using a general estimator like the transition frequency , it is possible to use a predicted signal - to - noise ratio for any particular system to determine whether it is potentially resolvable or not ( see also * ? ? ? * ) . for _ lisa _ observations of length @xmath234 the sensitivity to a particular binary source can be expressed by constructing a simple estimator of the signal - to - noise ratio ( snr ) . for circularized compact binaries
, long observations will narrow the bandwidth of the source , and the binary will appear in the fourier spectrum as a narrow spectral feature with root spectral density @xmath235 , given in terms of the strain amplitude @xmath236 as @xcite @xmath237 _ lisa _ will detect binaries in the presence of competing sources of noise .
there will be two components to the noise in _ lisa _ analysis : instrumental noise in the detector itself , and irreducible astrophysical noise from the total population of galactic close binaries .
the instrumental component of the noise is given by the shape of the _ lisa _ sensitivity curve @xcite .
in this work , the galactic foreground noise is computed from the synthesized galaxy itself ( i.e. , with startrack ) .
the gravitational wave strength of every binary in the synthesized galaxy is estimated from its parameters , and a running median of the total signal is computed as a function of frequency . since the total number of individually resolved binaries is expected to be small compared to the total population , the running median should be a good estimate of the residual astrophysical noise foreground that would result from a fully realized science analysis procedure .
the foreground resulting from this procedure has been shown in figure [ spectra ] ( top panel ) .
additionally , we have calculated the median foreground signal which is expected to arise from the unresolvable sources alone ( the ` reduced galaxy ' ) , which is likely a more realistic estimate of the true galactic foreground . in figure [ curves ] , we show the galactic foreground alongside the galactic foreground signal minus the signal arising from the sources which were determined to be resolvable using our signal - to - noise estimate technique , assuming a snr ratio @xmath3 , ( see 4.3 ) . alongside our galactic foregrounds
we show for comparison the standard _
sensitivity curve as well as the average michelson noise of _ lisa_.
the latter curve ( see 2.3 ) is a more appropriate comparison of the _ lisa _ instrumental noise with our calculated gravitational wave signal from the galactic binaries , as both curves ( our signal and the michelson noise ) account for directional and frequency dependencies of the wave as measured by _ lisa _ @xcite .
the standard _
sensitivity curve @xcite is often presented in the literature , and is an approximation of the spectral amplitude h@xmath238 scaled from the barycentred sensitivity amplitude h@xmath239 from the _ online sensitivity curve generator_. the ultimate level of the confusion foreground ( and as a result , the ultimate snr of any given source ) will depend strongly on the actual data analysis algorithm used to identify and subtract sources .
to date no fully realized implementation of such a procedure has been developed .
the computed foreground shown here should be a good estimate to the output of a fully developed analysis procedure . though millions of double wds are expected to be detected with _
lisa _ , a much smaller number of these systems are predicted to be resolved , depending upon the physical properties of each binary ( see also @xcite for a discussion ) . due to the small separations of these compact binaries
, @xcite predicted that nearly @xmath240 of the resolved _ lisa _ binaries will be eclipsing in the optical regime and thus will offer an opportunity for these systems to be studied with follow - up observations , allowing for the determination of wd physical parameters ( i.e. , radii ) .
it has been noted by @xcite however that the intrinsic brightness of white dwarfs was overestimated in @xcite , thus the expected number of _ lisa _ double white dwarfs with electro - magnetic counterparts should be lower .
@xcite predict a smaller , though still relatively optimistic number of _ lisa _ systems which may have optical and/or near infrared counterparts ; possibly as many as @xmath241 sources ( see their 3.3 ) .
the total effective noise in the detector ( instrumental @xmath14 foreground ) is given by the spectral amplitude @xmath242 , and the snr is estimated as @xcite @xmath243 a binary is deemed `` resolvable '' ( distinguishable from the confusion limited galactic foreground ) if its snr is greater than some detection threshold ( typical snr @xmath244 ) . additionally , resolved sources in the population are sorted into _
monochromatic sources _ and _ chirping sources_. monochromatic sources do not evolve appreciably over the _
lisa _ observing time , @xmath234 .
a binary is considered to be chirping if its frequency changes by more than the _
frequency resolution , @xmath245 .
this criteria is a conservative one , as the expectation is _ lisa _ will be able to detect frequency changes which are smaller than the frequency resolution @xmath246 @xcite .
the detection of binaries with appreciable evolution in @xmath247 ( or ` chirp ' ) is useful , since this frequency evolution allows for the determination of @xmath248 , from which the distance to the source can be calculated ( see e.g. , * ? ? ?
the estimate of the snr in eq.([simplesnr ] ) will not be precise in the regime where orbital periods change on timescales short compared to the observation time , or where the shape of the instrumental noise changes appreciably as a function of frequency .
however , this simple estimator is a good tool to use as a first cut in a search over a large population of sources where a calculation of the snr for each binary would be computationally prohibitive . for the purposes of this paper
, this estimator should be perfectly adequate since in large part the binaries of interest are not chirping across a significant portion of the _ lisa _ band during @xmath234 .
once a set of galactic binaries has been identified using the simple estimator in eq .
( [ simplesnr ] ) , it is computationally plausible to consider a more robust estimation of the snr for each system which can then be constructed from the data analysis technique of choice .
for example , with chirping binaries a good estimate of the snr can be motivated from matched filtering techniques by integrating over a binary s spectral energy distribution , @xmath249 @xcite . using this simple snr estimator to resolve _ lisa _ binaries
, we find that a total of @xmath2 double white dwarfs will be resolvable in our galactic sample : @xmath250 chirping and @xmath251 monochromatic .
this is inferred under an optimistic assumption that all binaries with a snr @xmath252 have a chance to be resolved . within the resolvable population ,
we find @xmath253 am cvn binaries and @xmath254 possible double degenerate scenario - dds ; @xcite - sn ia progenitors ( detached coco wd binaries which will merge within a hubble time , whose total mass exceeds the chandrasekhar mass limit ) . out of these possible sn ia progenitors , the majority are formed through the coco - d4 channel , which involves the most massive primaries on the zams . as for the am cvn resolvable binaries ,
7 % are chirping and arise mostly from cohe - r1 and cohyb - r1 channels , in near - equal numbers .
both formation channels involve two ce events and thus if these systems will be resolved with _
lisa _ , they will serve as useful objects for understanding close binary evolution .
additionally , a number of progenitor am cvn binaries are resolvable ( @xmath255 ; detached channels ) , but only @xmath256 of these are chirping .
if we relax our criteria and assume that any binary with a snr @xmath257 is potentially resolvable , we find @xmath258 resolvable double wds ( @xmath259 chirping , @xmath260 monochromatic ) . additionally , we find @xmath261 _ lisa _ binaries with snr @xmath262 , @xmath167 of which are chirping sources .
the binaries with large snr ( @xmath263 ) typically all have orbital periods @xmath264 s ( @xmath265 hz ) and have either just recently encountered rlof , thus they still have relatively large chirp masses , or they are close to reaching contact . in table 5
we show a breakdown of the @xmath266 ( 2800 ) potentially resolvable _ lisa _ double wds which have snr @xmath3 ( snr @xmath262 ) , with percentages relative to the resolvable population in question .
the properties of resolved systems reflect some of the selection effects that are expected to arise from using _
lisa _ to identify double wds .
systems with short orbital periods ( and consequently high gravitational wave frequencies ) are favored both because the amplitude of the signal scales as @xmath267 and because the number of binaries per bin drops with increasing frequency .
furthermore , systems with large chirp masses are favored simply because the amplitude of their gr scales as @xmath268 .
thus , we see that the resolvable population is biased toward high mass , high frequency ( short period ) systems . realistically speaking , it is still unclear as to which method(s ) of detecting resolvable binaries with _
lisa _ will be most successful , and so we have discussed the likelihood of potentially resolving individual binaries using a few different methods . _
detached resolved sources . _ a large number of the resolved sources are of the hehe - d1 channel , where two initially low - mass ( @xmath269 ) evolved stars lose their envelopes during mass transfer phases .
hehe - d1 are not only the most numerous channel of the detached systems ( table 1 and table 3 ) , but also extend to high gr frequencies . lumping all of the coco - d formation channels together ,
coco detached binaries make up @xmath270 % of the resolved population , and potential dds progenitors make up 5% of the total resolved population .
though heco - d1 double wds are more numerous compared to cohe - d1 binaries , fewer of them are resolved .
we note that cohe - d1 progenitors involve two ce phases ( heco - d1 only one ) , and overall occupy closer orbits thus contributing to higher gr frequencies .
_ rlof resolved sources .
_ for rlof systems , the majority are of am cvn - type .
the most common rlof channel is cohe - r1 , followed by coco - r1 .
coco - r1 binaries are not common , but they lie on very close orbits , consisting of a massive ( @xmath10 m@xmath28 ) co wd accreting matter from a low - mass ( @xmath271 m@xmath28 ) co wd .
the resolved cohe - r1 am cvn systems account for most of the total number of resolved am cvn binaries . since the formation of these systems involves two ce events , future _ lisa _ observations may put some very useful constraints on this rather uncertain evolutionary phase which is crucial to the formation of close binaries . however , when comparing relative formation scenarios of am cvn binaries , we must keep in mind that the precise behaviour of the stars upon reaching contact - and the exact physical conditions of the donor at contact - are still unclear , as are the physical characteristics of the am cvn progenitors themselves . further observations and detailed
modeling will be an important step toward understanding the donors in am cvn type systems @xcite .
for example , deviations from the equilibrium evolution of am cvn systems , e.g. , brought on upon by changes in the mass transfer rate following a nova explosion , result in a measurable frequency evolution which can greatly differ from that which is expected for the long - term , equilibrium evolution case .
this was already shown by @xcite ; if short - term changes in the evolution of am cvn systems are neglected , it can lead to erroneous estimations of some of the binary parameters .
the number of binaries important in context of the _ lisa _ gr signal was estimated by @xcite and in their work they predicted @xmath272 galactic close double wds ( orbital periods @xmath273 day ) based upon a surface density star formation rate .
however that estimate seemed to over - predict the observed local density of double wds available at that time .
they concluded that even if they decreased the estimated space density by a factor of 10 ( decreasing the total number of galactic double wds ) , close wd binaries would continue to dominate the _
gr signal at low frequencies @xcite .
population synthesis studies ( e.g. , * ? ? ?
* ) which incorporate current estimates of galactic star formation rates and sne rates result in a higher number of galactic double wds ( @xmath274 ) double wds in the milky way .
] , albeit the majority of these would have orbital periods too large to be detectable with _ lisa_.
still , we note here that current estimates of the number of double wds with gr frequencies detectable with _
are a factor of @xmath275 higher than those used in 1990 . to populate the galaxy ( the disc only ) , @xcite used the density distribution @xmath276 where @xmath277 is the central density , @xmath278 and @xmath279 are galactocentric cylindrical coordinates , and @xmath280 and @xmath281 are the radial scale length and vertical scale height of the galactic disc , respectively .
@xcite used @xmath282 pc in their study , and while though a reasonable assumption for the scale height of the thin disc , likely led to their ( too ) low estimate of double wds .
more recently , it has been noted ( e.g. , * ? ? ?
* ) that the scale height for the binary population may be much higher ( @xmath283 pc ) than that used in @xcite .
the @xcite curve is a modified result based on @xcite , who in calculating the gr amplitude in 1990 , used too low of a space - density estimate for double wds . in the original study of @xcite which neglected rlof systems , galactic double wds begin to dominate the gr amplitude at @xmath284 mhz , and
continue to dominate until @xmath285 mhz , where the average number of sources per resolvable frequency bin drops below 1 ( similar results for the transition frequency were found in their study that included rlof systems ; @xcite ) .
the analytical study of @xcite was not directly comparable to more recent studies of the _ lisa _ foreground which utilized more sophisticated techniques .
@xcite performed a re - calculation of the analytical @xcite gr foreground for galactic binaries in order to compare the results directly to those of @xcite ( see * ? ? ?
* their figure 10 ) .
little was known about the true space density of double white dwarfs at the time of the @xcite study , and since @xcite was based on these results which underestimated the space density ( and hence total number ) of galactic white dwarf binaries , the @xcite study had a @xmath286% decrease in source density as compared to the study of @xcite .
this reduction in number of systems resulted in a decrease in the foreground gr amplitude . until more recent observations of double white dwarfs were available to better constrain the space density @xcite , the error went undetected for some time , as this decrease was serendipitously counteracted by an increase in amplitude imposed by the ( too ) large double wd chirp masses used in the @xcite and @xcite studies . in short ,
the studies of @xcite ( recalculated @xcite ) and @xcite have comparable foreground levels , albeit for different physical reasons .
an increased number of observations of wd binaries since the time of the @xcite study [ ] nap04,and05 have confirmed that the ( lower ) wd masses used by @xcite coincide more closely with the masses of wds in binaries , where the masses used in @xcite do not ( see also * ? ? ?
* ; * ? ? ?
* for further discussion ) . based on @xcite
, @xcite used the criterion that the transition frequency occurs when the average number of binaries per bin drops below 1 , and they found the transition at @xmath285 mhz .
they also found that @xmath287 detached double wds will be resolved with _
lisa _ , assuming all systems above the _ lisa _ sensitivity limit ( a signal - to - noise of 1 ) are detectable . in their study which included am cvn systems , @xcite found a transition frequency closer to @xmath19 mhz , and that a total of @xmath288 galactic double wds ( half detached and half rlof ) would be resolvable .
for a signal - to - noise ratio of 1 , we find that @xmath258 double white dwarfs with gr frequencies @xmath289 hz are potentially resolvable with _
lisa_. however , in order to perform a reasonable comparison with @xcite , we need to consider only those binaries whose gr frequencies are above @xmath19 mhz ( and below 0.01 hz ; see 4 ) . in this case
, we find that @xmath290 double white dwarfs may be resolvable with lisa . in previous studies
, it was assumed that the level of the galactic foreground will not contribute above the frequency at which the average number of sources per bin drops below 1 ( coined the ` transition frequency ' ) . since those studies do not calculate the gr signal beyond the transition frequency , above @xmath19 mhz their results can not be compared with ours .
we find that the frequency at which the wd foreground drops below the instrumental noise level ( @xmath291 mhz ) is above the transition frequency which was found in previous studies .
in contrast to previous work , it is expected that at higher frequencies ( @xmath8 mhz ) the rlof systems may contribute more significantly to the _ lisa _ signal than detached systems .
this has implications for the ` transition frequency ' between unresolved and resolved systems : the transition frequency is shifted to higher frequencies with the addition of rlof systems .
since we do not follow the evolution of non - degenerate stars in our _ lisa _ calculations , the binaries involving ` pre'-hydrogen wd donors ( i.e. , cvs ) are not shown on the figures , though we note they would make some contribution to the _ lisa _ data stream .
we do not include the ` brown dwarfs ' in our lisa signal populations , though we note that binaries involving these brown dwarfs will not contribute significantly to the galactic gravitational wave background as these binaries have very low chirp masses . since the formation of such objects takes on average several gyr , it is however important to include them when considering very old stellar populations such as the galactic halo @xcite , where these systems comprise over @xmath292 % of the entire remnant binary population .
in that work , it was found that the _ lisa _ gr signal from halo white dwarfs will be at least an order of magnitude below the signal arising from galactic ( disc@xmath14bulge ) white dwarf binaries .
we have used the population synthesis binary evolution code startrack and the detailed _ lisa _ simulator of @xcite to calculate the gravitational wave foreground arising from close white dwarf binaries in the milky way disc and bulge .
our population synthesis includes both detached and mass - transferring binaries , and it is found that the mass - transferring ( rlof ) binaries begin to contribute significantly to the foreground noise at high gr frequencies ( @xmath8 mhz , figure 10 bottom panel ) .
our results are in general agreement with the analytical study of @xcite and the population synthesis study of @xcite for the low frequency regime : the _ lisa _ signal is dominated by detached binaries .
however , in contrast to previous studies , it is found that , despite the fact that the ` transition frequency ' is found to be @xmath293 mhz using the 3-bin rule of thumb , at frequencies above @xmath294 mhz the presence of rlof binaries will likely make it difficult to resolve individual sources , and we find that the _ galactic foreground dominates the lisa noise up to at least @xmath295 mhz_. this is because the rlof binaries add to the total number of _ lisa _ systems at high frequencies , and at high frequencies the rlof systems chirp backward and spread out over more frequency bins , which offers a new set of challenges for _ lisa _ data analysts .
the details of the true orbital period evolution due to mass transfer and possible tidal effects ( e.g. , * ? ? ?
* ; * ? ? ?
* ) are not well understood , and this uncertainty may impact the ability of future data analysis schemes to successfully resolve these systems .
it is starting to become clear that such effects should be taken into account when studying the population of compact binaries @xcite .
one new and interesting result of our study is the possibility that rlof systems which have recently entered the mass transfer phase ( small periods , higher chirp masses ) can add to the confusion foreground at higher frequencies than were previously predicted for the double white dwarf population .
it should be repeated here that in this study , we employed one model prescription of ce evolution .
this important phase of binary evolution may have a large impact on a synthesized stellar population , and various model prescriptions of ce may lead to a different number of galactic binaries and binary physical properties , which in turn affect the shape and strength of the _ lisa _ gr signal .
the number of close ( _ lisa _ ) binaries could potentially be decreased if the ce efficiency is low ( more mergers ) or if angular momentum balance rather than energy balance is assumed during the ce phase .
it was shown in @xcite that when an angular momentum balance prescription is adopted @xcite , post - ce orbital separations are overall larger , and fewer binaries will enter a cv - like rlof phase between a white dwarf and a main sequence star , which is a crucial step for some of the older ( bulge cohe - r2 ) _ lisa _ binaries .
we have shown that there is a large population of potentially resolvable binaries in the galaxy .
the approach of estimating resolvable systems described in 4.3 ( ` snr @xmath296 5 ' ) , relies on assuming that signals arising from individual binaries are separable , i.e. , can be demodulated using some sort of matched filtering technique ( see * ? ? ?
* and references their - in ) .
if detected , once their properties are measured these systems may aid to solve several remaining problems in binary evolution theory .
especially interesting would be the detection of any double white dwarf binary with a mass close to or over the chandrasekhar mass ( potential sn ia progenitor or possible pre - ns accretion induced collapse object ) , or the measurement of the orbital periods of systems which have just emerged from a common envelope ( the phase that determines the formation of most compact object binaries ) . however , we reiterate that the detection of resolvable binaries will depend on the development of sensitive detection schemes and data analysis techniques .
we would like to thank gijs nelemans for very useful discussion on this project , which greatly improved this work , and for providing data for @xcite and @xcite .
kb , mb and sll acknowledge the hospitality of the aspen center for physics .
mb and sll were supported at the aspen center by nasa award number nng05g106 g .
mb is also supported by nasa apra grant number nng04gd52 g .
kb and ajr acknowledge support through kbn grants 1 p03d 022 28 and pbz - kbn-054/p03/2001 , and the hospitality of the center for gravitational wave astronomy ( utb ) .
sll also acknowledges support from the center for gravitational wave physics , funded by the nsf under cooperative agreement phy 01 - 14375 , and from nasa award nng05gf71 g .
ajr acknowledges the support of sigma xi and the hospitality of the nicolaus copernicus astronomical center .
the majority of ajr s calculations for this work were carried out at new mexico state university and the harvard - smithsonian center for astrophysics .
the authors also would like to thank sam finn for directing us to the kde package for matlab which was used to generate the pdfs of the various channels , and joe romano for providing a routine for generating the _ lisa _ noise .
finally , we thank the referee gijs nelemans and the anonymous referee for highly insightful questions and comments .
startrack simulations were performed at the copernicus center in warsaw , poland .
the ligo scientific collaboration : b. abbott 2007 , arxiv e - prints , 711 , arxiv:0711.3041 abt , h. a. 1983 , , 21 , 343 anderson , s. f. , et al .
2005 , , 130 , 2230 arnaud , k. a. , auger , g. , babak , s. , baker , j. g. , benacquista , m. j. , bloomer , e. , brown , d. a. , camp , j. b. , cannizzo , j. k. , christensen , n. , clark , j. , cornish , n. j. , crowder , j. , cutler , c. , finn , l. s. , halloin , h. , hayama , k. , hendry , m. , jeannin , o. , krlak , a. , larson , s. l. , mandel , i. , messenger , c. , meyer , r. , mohanty , s. , nayak , r. , numata , k. , petiteau , a. , pitkin , m. , plagnol , e. , porter , e. k. , prix , r. , roever , c. , stroeer , a .. thirumalainambi , r. , thompson , d. e. , toher , j. , umstaetter , r. , vallisneri , m. , vecchio , a. , veitch , j. , vinet , j .- y . ,
whelan , j. t. & woan , g. 2007 , classical and quantum gravity , 24 , 529 beer , m. e. , dray , l. m. , king , a. r. , & wynn , g. a. 2007 , , 375 , 1000 belczynski , k. , benacquista , m. , & bulik , t. 2008 , arxiv:0811.1602 belczynski , k. , kalogera , v. , bulik , t. 2002 , , 572 , 407 belczynski , k. , kalogera , v. , rasio , f. a. , taam , r. e. , zezas , a. , bulik , t. , maccarone , t. j. , & ivanova , n. 2008 , , 174 , 223 belczynski , k. , & taam , r. e. 2004 , apj , 616 , 1159 belczynski , k. , bulik , t. , & ruiter , a. j. 2005 , apj , 629 , 915 benacquista , m. 2001 , 20th texas symposium on relativistic astrophysics , 586 , 793 benacquista , m. j. , degoes , j. , & lunder , d. 2004 , classical and quantum gravity , 21 , 509 benacquista , m. j. , larson , s. l. , & taylor , b. e. 2007 , classical and quantum gravity , 24 , 513 benacquista , m. , & holley - bockelmann , k. 2006 , , 645 , 589 bender , p. l. , & hils , d. 1997 , classical and quantum gravity , 14 , 1439 bildsten , l. , townsley , d. m. , deloye , c. j. , & nelemans , g. 2006 , , 640 , 466 burgay , m. , et al .
2003 , , 426 , 531 cooray , a. , farmer , a. j. , & seto , n. 2004 , , 601 , l47 cooray , a. 2004 , , 354 , 25 cornish , n. j. 2002 , , 65 , 022004 cutler , c. 1998 , , 57 , 7089 cox , a. n. ( editor ) 2000 , allen s astrophysical quantities , fourth ed . , springer - verlag , new york , ny danzmann , k. 1996 , classical and quantum gravity , 13 , 247 de kool , m. 1990 , , 358 , 189 deloye , c. j. , & bildsten , l. 2003 , , 598 , 1217 deloye , c. j. , taam , r. e. , winisdoerffer , c. , & chabrier , g. 2007 , , 381 , 525 duquennoy , a. , & mayor , m. 1991 , , 248 , 485 edlund , j. a. , tinto , m. , krlak , a. , & nelemans , g. 2005 , , 71 , 122003 evans , c. r. , iben , i. j. , & smarr , l. 1987 , , 323 , 129 farmer , a. j. , & phinney , e. s. 2003 , , 346 , 1197 flanagan , .
, & hughes , s. a. 1998 , , 57 , 4535 gair , j. r 2007 , arxiv general relativity and quantum cosmology e - prints , arxiv : gr - qc/0703041 hils , d. , & bender , p. l. 2000 , , 537 , 334 hils , d. , bender , p. l. , & webbink , r. f. 1990 , , 360 , 75 hughes , s. a. 2006 , laser interferometer space antenna : 6th international _ lisa _ symposium , 873 , 13 hurley , j. r. , pols , o. r. , & tout , c. a. 2000 , , 315 , 543 iben , i. , jr . , & tutukov , a. v. 1984 , , 54 , 335 kalogera , v. , et al . 2004 , , 601 , l179 king , a. r. , & kolb , u. 1995 , , 439 , 330 kosenko , d. i. , & postnov , k. a. 1998 , , 336 , 786 klypin , a. , zhao , h. , & somerville , r. s. 2002 , , 573 , 597 kroupa , p. , tout , c. a. , & gilmore , g. 1993 , mnras , 262 , 545 larson , s. l. , hiscock , w. a. , & hellings , r. w. 2000 , , 62.062001 ligo scientific collaboration 2009 , arxiv:0901.0302 lipunov , v. m. , postnov , k. a. , & prokhorov , m. e. 1987 , , 176 , l1 lorn - aguilar , p. , guerrero , j. , isern , j. , lobo , j. a. , & garca - berro , e. 2005 , , 356 , 627 majewski , s. r. , & siegel , m. h. 2002 , , 569 , 432 marsh , t. r. , nelemans , g. , & steeghs , d. 2004 , , 350 , 113 mazeh , t. , goldberg , d. , duquennoy , a. , & mayor , m. 1992 , , 401 , 265 mironovskii , v. n. 1965 , , 42 , 977 napiwotzki , r. , et al .
2004 , spectroscopically and spatially resolving the components of the close binary stars , 318 , 402 nelemans , g. 2009 , classical and quantum gravity , 26 , 094030 nelemans , g. , & tout , c. a. 2005 , , 356 , 753 nelemans , g. , yungelson , l. r. , & portegies zwart , s. f. 2004 , , 349 , 181 nelemans , g. , 2003 , class .
quantum grav . , 20 , 81 nelemans , g. 2003 , the astrophysics of gravitational wave sources , 686 , 263 ( 2003b ) nelemans , g. , portegies zwart , s. f. , verbunt , f. , & yungelson , l. r. 2001 , , 368 , 939 ( 2001a ) nelemans , g. , yungelson , l. r. , & portegies zwart , s. f. 2001 , , 375 , 890 ( 2001b ) nelemans , g. , yungelson , l. r. , portegies zwart , s. f. , & verbunt , f. 2001 , , 365 , 491 oshaughnessy , r. , kim , c. , kalogera , v. , & belczynski , k. 2008 , , 672 , 479 peters , p. c. , 1964 , physical reviews , 136 , 1224 podsiadlowski , p. 2008 , hydrogen - deficient stars , 391 , 323 postnov , k. a. , & prokhorov , m. e. 1998 , , 494 , 674 postnov , k. a. , & yungelson , l. r. 2006 , living reviews in relativity , 9 , 6 racine , . , phinney , e. s. , & arras , p. 2007
, , 380 , 381 roelofs , g. h. a. , groot , p. j. , benedict , g. f. , mcarthur , b. e. , steeghs , d. , morales - rueda , l. , marsh , t. r. , & nelemans , g. 2007 , , 666 , 1174 roelofs , g. h. a. , nelemans , g. , & groot , p. j. 2007 , , 382 , 685 rubbo , l. j. , cornish , n. j. , & poujade , o. 2004 , , 69 , 082003 ruiter , a. j. 2009 , ph.d .
thesis ruiter , a. j. , belczynski , k. , benacquista , m. , & holley - bockelmann , k. 2009 , , 693 , 383 ruiter , a. j. , belczynski , k. , & harrison , t. e. 2006 , apjl , 640 , l167 sesana , a. , haardt , f. , madau , p. , & volonteri , m. 2005 , , 623 , 23 seto , n. 2002 , , 333 , 469 smak , j. 1967 , acta astronomica , 17 , 255 saio , h. , & nomoto , k. 2004 , , 615 , 444 stroeer , a. , & nelemans , g. 2009 , , 400 , l24 stroeer , a. , vecchio , a. , & nelemans , g. 2005 , , 633 , l33 takahashi , r. , & seto , n. 2002 , , 575 , 1030 timpano , s. e. , rubbo , l. j. , & cornish , n. j. 2006 , , 73 , 122001 van der sluys , m. v. , verbunt , f. , & pols , o. r. 2006 , , 460 , 209 vecchio , a. , & wickham , e. d. 2004 , , 70 , 082002 wang , z. , & chakrabarty , d. 2004 , , 616 , l139 warner , b. 1995 , , 225 , 249 warner , b. 1995 , cataclysmic variable stars , cambridge univ . press , cambridge webbink , r. f. 1984 , apj , 277 , 355 willems , b. , vecchio , a. , & kalogera , v. 2008 , physical review letters , 100 , 041102 williams , g. 2008 , master s thesis , university of texas at browsville yoon , s .- c . , & langer , n. 2005 , , 435 , 967 ccc hehe - d1 & mt1(3 - 1 ) ce2(10 - 3;10 - 10 ) & 8% + heco - d1 & mt1(3 - 1/2/3 ) ce2(10 - 5;10 - 8 ) & 6% + coco - d1 & mt1(2/3/4 - 1 ) ce2(4/7 - 5;7 - 8 ) mt2(7 - 8 ) & 7% +
coco - d2 & mt1(2/3/4 - 1 ) ce2(11 - 5;11 - 8 ) & 3% + coco - d3 & mt1(2/3/4 - 1/2/3 ) ce2(7 - 5;7 - 8 ) & 1% + coco - d4 & mt1(2/3 - 1 ) mt1(8/9 - 1 ) ce2(11 - 3;11 - 7 ) mt2(11 - 8 ) & 1% + coco - d5 & mt1(3/4 - 1 ) ce2(7 - 3;7 - 7 ) mt2(7 - 8 ) mt1(8 - 11 ) & @xmath297 1% +
cohe - d1 & ce1(6 - 1;11 - 1 ) ce2(11 - 3;11 - 10 ) & 5% + the rest & & 4% + total & @xmath114 & 35% + [ channelsdd ] ccc cohe - r1 & ce1(5/6 - 1;8/11 - 1 )
ce2(11 - 3;11 - 10 ) mt2(11 - 10 ) & 47% + cohyb - r1 & ce1(5/6 - 1;8/11 - 1 ) ce2(11 - 3;11 - 7/17 ) mt2(11 - 7/17 ) & 8% + heco - r1 & mt1(2/3 - 1 ) ce2(10 - 5;10 - 8 ) mt1(10 - 11 ) & 5% + the rest & & 5% + total & @xmath298 & 65% + [ channelsdr ] ccc hehe - d1 & mt1(3 - 1 ) ce2(10 - 3;10 - 10 ) & 20% + cohe - d1 & ce1(6 - 1;11 - 1 ) ce2(11 - 3;11 - 10 ) & 9% + heco - d1 & mt1(3 - 1/2/3 ) ce2(10 - 5;10 - 8 ) & 5% + coco - d & various channels & 3% + the rest & & 5% + total & @xmath120 & 42% + [ channelsbr ] lcc detached & & + hehe - d1 & 8.4 & 9.4 + heco - d1 & 6.3 & 5.5 + coco - d1 & 1.2 & 1.8 + coco - d2 & 3.6 & 3.2 + coco - d3 & 2.3 & 1.8 + coco - d4 & 9.1 & 5.3 + coco - d5 & 1.4 & 1.4 + cohe - d1 & 23.0 & 22.3 + cohe - d2 & 0 & @xmath217 + cohyb - d1 & 1.2 & 1.2 + the rest & @xmath217 & @xmath217 + total & 57.0 & 53.0 + rlof & & + cohe - r1 & 17.3 & 21.4 + coco - r1 & 10.2 & 11.0 + cohyb - r1 & 9.9 & 8.4 + heco - r1 & 2.6 & 2.6 + hehe - r1 & 1.6 & 1.9 + hybco - r1 & 1.2 & 1.5 + the rest & @xmath299 & @xmath217 + total & 43.0 & 47.0 + & & + total & @xmath300 & @xmath266 + [ res ] | double white dwarfs are expected to be a source of confusion - limited noise for the future gravitational wave observatory _ lisa_. in a specific frequency range , this ` foreground noise ' is predicted to rise above the instrumental noise and hinder the detection of other types of signals , e.g. , gravitational waves arising from stellar mass objects inspiraling into massive black holes . in many previous studies only detached populations of compact object binaries
have been considered in estimating the _ lisa _ gravitational wave foreground signal . here
, we investigate the influence of compact object detached and roche - lobe overflow galactic binaries on the shape and strength of the _ lisa _ signal .
since @xmath0% of remnant binaries which have orbital periods within the _ lisa _ sensitivity range are white dwarf binaries , we consider only these binaries when calculating the _ lisa _ signal .
we find that the contribution of rlof binaries to the foreground noise is negligible at low frequencies , but becomes significant at higher frequencies , pushing the frequency at which the foreground noise drops below the instrumental noise to @xmath1 mhz .
we find that it is important to consider the population of mass transferring binaries in order to obtain an accurate assessment of the foreground noise on the _ lisa _ data stream .
however , we estimate that there still exists a sizeable number ( @xmath2 ) of galactic double white dwarf binaries which will have a signal - to - noise ratio @xmath3 , and thus will be potentially resolvable with _ lisa_. we present the _ lisa _ gravitational wave signal from the galactic population of white dwarf binaries , show the most important formation channels contributing to the _ lisa _ disc and bulge populations and discuss the implications of these new findings . |
it has been the major goal of particle physics to discover a theoretical framework for unifying gravity with the other three known forces , _ viz .
_ , electromagnetism , and the weak and strong nuclear forces . such a theory must be compatible with quantum theory at very small scales corrsponding to very high energies .
even the possibly less ambitious goal of reconciling general relativity with quantum theory has been elusive and may require new concepts to accomplish .
there has been a particular interest in the possibility that a quantum gravity theories will lead to lorentz invariance violation ( liv ) at the planck scale , @xmath1 m. this scale corresponds to a mass ( energy ) scale of @xmath2 gev / c@xmath3 .
it is at the planck scale where quantum effects are expected to play a key role in determining the effective nature of space - time that emerges as general relativity in the classical continuum limit .
the idea that lorentz invariance ( li ) may indeed be only approximate has been explored within the context of a wide variety of suggested planck - scale physics scenarios .
these include the concepts of deformed relativity , loop quantum gravity , non - commutative geometry , spin foam models , and some string theory ( m theory ) models .
such theoretical explorations and their possible consequences , such as observable modifications in the energy - momentum dispersion relations for free particles and photons , have been discussed under the general heading of `` planck scale phenomenology '' .
there is an extensive literature on this subject .
( see @xcite for a review ; some recent references are refs . @xcite
@xcite . for a non - technical treatment of the present basic approaches to a quantum gravity theory ,
see ref .
one should keep in mind that in a context that is separate from quantum gravity considerations , it is important to test li for its own sake @xcite .
_ liv gratia liv_. the significance of such an approach is evident when one considers the unexpected discoveries of the violation of @xmath4 and @xmath5 symmetries .
in fact , it has been shown that a violation of @xmath6 would imply liv @xcite we will consider here some of the consequent searches for such effects using high energy astrophysics observations , particularly observations of high energy cosmic @xmath0-rays and ultrahigh energy cosmic rays .
we know that lorentz invariance has been well validated in particle physics ; indeed , it plays an essential role in designing machines such as the new lhc ( large hadron collider ) .
thus , any liv extant at accelerator energies ( `` low energies '' ) must be extremely small .
this consideration is reflected by adding small lorentz - violating terms in the free particle lagrangian .
such terms can be postulated to be independent of quantum gravity theory , _
e.g. _ , refs .
alternatively , it can be assumed that the terms are small because they are suppressed by one or more powers of @xmath7 ( with the usual convention that @xmath8 . ) in the latter case , in the context of effective field theory ( eft ) , such terms are assumed to approximate the effects of quantum gravity at `` low energies '' when @xmath9 . one result of such assumptions is a modification of the dispersion relation that relates the energy and momentum of a free particle or photon .
this , in turn , can lead to a maximmum attainable velocity ( mav ) of a particle different from @xmath10 or a variation of the velocity of a photon _ in vacuo _ with photon energy .
both effects are clear violations of relativity theory
. such modifications of kinematics can result in changes in threshold energies for particle interactions , suppression of particle interactions and decays , or allowance of particle interactions and decays that are kinematically forbidden by lorentz invariance @xcite .
a simple formulation for breaking li by a small first order perturbation in the electromagnetic lagrangian which leads to a renormalizable treatment has been given by coleman and glashow @xcite .
the small perturbative noninvariant terms are both rotationally and translationally invariant in a preferred reference frame which one can assume to be the frame in which the cosmic background radiation is isotropic .
these terms are also taken to be invariant under @xmath11 gauge transformations in the standard model .
using the formalism of ref .
@xcite , we denote the mav of a particle of type @xmath12 by @xmath13 , a quantity which is not necessarily equal to @xmath14 , the low energy _ in vacua _ velocity of light .
we further define the difference @xmath15 .
these definitions can be generalized and can be used to discuss the physics implications of cosmic - ray and cosmic @xmath0-ray observations @xcite .
in general then , @xmath16 .
the physical consequences of such a violation of li depend on the sign of the difference between these two mavs . defining @xmath17
one can consider the two cases of positive and negative values of @xmath18 separately @xcite .
_ case i : _ if @xmath19 ( @xmath20 ) , the decay of a photon into an electron - positron pair is kinematically allowed for photons with energies exceeding @xmath21 the decay would take place rapidly , so that photons with energies exceeding @xmath22 could not be observed either in the laboratory or as cosmic rays . from the fact that photons have been observed with energies @xmath23 50 tev from the crab nebula , one deduces for this case that @xmath24tev , or that -@xmath25 . _ case ii : _ for this possibility , where @xmath26 ( @xmath27 ) , electrons become superluminal if their energies exceed @xmath28 .
electrons traveling faster than light will emit light at all frequencies by a process of ` vacuum erenkov radiation . '
this process occurs rapidly , so that superluminal electron energies quickly approach @xmath29 .
however , because electrons have been seen in the cosmic radiation with energies up to @xmath302 tev , it follows that @xmath31 tev , which leads to an upper limit on @xmath18 for this case of @xmath32 .
note that this limit is two orders of magnitude weaker than the limit obtained for case i. however , this limit can be considerably improved by considering constraints obtained from studying the spectra of active galaxies @xcite .
a constraint on @xmath18 for @xmath33 follows from a change in the threshold energy for the pair production process @xmath34 .
this follows from the fact that the square of the four - momentum is changed to give the threshold condition @xmath35 where @xmath36 is the energy of the low energy photon and @xmath37 is the angle between the two photons . the second term on the left - hand - side comes from the fact that @xmath38 .
it follows that the condition for a significant increase in the energy threshold for pair production is @xmath39 @xmath40 @xmath41 , or equivalently , @xmath42 .
the observed @xmath0-ray spectrum of the active galaxies mkn 501 and mkn 421 while flaring @xcite exhibited the high energy absorption expected from @xmath0-ray annihilation by extragalactic pair - production interactions with extragalactic infrared photons @xcite .
this led stecker and glashow @xcite to point out that the mkn 501 spectrum presents evidence for pair - production with no indication of liv up to a photon energy of @xmath3020 tev and to thereby place a quantitative constraint on liv given by @xmath43 .
as previously mentioned , liv has been proposed to be a consequence of quantum gravity physics at the planck scale @xcite . in models involving large extra dimensions ,
the energy scale at which gravity becomes strong can occur at a quantum gravity scale , @xmath44 , even approaching a tev @xcite . in the most commonly considered case , the usual relativistic dispersion relations between energy and momentum of the photon and the electron are modified @xcite by a term of order @xmath45 . generalizing the liv parameter @xmath18 from equation ( [ delta ] ) to an energy dependent form , we find @xmath46 it follows that the threshold condition for pair production given by equation ( [ threshold ] ) implies that @xmath47 since pair production occurs for energies of at least 20 tev , we find a constraint on the quantum gravity scale @xcite @xmath48 .
this constraint contradicts the predictions of some proposed quantum gravity models involving large extra dimensions and smaller effective planck masses . in a variant model of ref .
@xcite , the photon dispersion relation is changed , but not that of the electrons . in this case
, we find the even stronger constraint @xmath49 .
one possible manifestation of lorentz invariance violation , from planck scale physics produced by quantum gravity effects , is a change in the energy - momentum dispersion relation of a free particle or a photon . if this results from the linear planck - supressed term as in equation ( [ pcube ] ) above , this results in a photon velocity retardation that is of first order in @xmath50 @xcite . in a @xmath51 cosmology , where present observational data indicate that @xmath52 and @xmath53 , the resulting difference in the propagation times of two photons having an energy difference @xmath54 from a @xmath0-ray burst ( grb ) at a redshift @xmath55 will be @xmath56 for a photon dispersion of the form @xmath57 ) , with @xmath10 being the usual low energy velocity of light @xcite . in other words , @xmath18 , as defined earlier , is given by @xmath58 .
the _ fermi _ gamma - ray space telescope , ( see figure [ glast ] ) , with its _ burst monitors ( gbm ) _ covers an energy range from 8 kev to 40 mev and its _ large area telescope ( lat ) _ covers an energy range from 20 mev to @xmath59 gev
. it can observe and study both grbs and flares from active galactic nuclei over a large range of both energy and distance .
this was the case with the grb 090510 , a short burst at a cosmological distance corresponding to a redshift of 0.9 that produced photons with energies extending from the x - ray range to a of energy @xmath60 31 gev .
this burst was therefore a perfect subject for the application of equation ( [ delay ] ) .
_ fermi _ observations of grb090510 have yielded the best constraint on any first order retardation of photon velocity with energy @xmath61 .
this result would require a value of @xmath62 @xcite in large extra dimension scenarios , one can have effective planck masses smaller than @xmath63 gev , whereas in most qg scenarios , one expects that the minimum size of space - time quanta to be @xmath64 .
this implies a value for @xmath65 in all cases . in particular ,
we note the string theory inspired model of ref .
this model invisions space - time as a gas of d - particles in a higher dimensional bulk where the observable universe is a d3 brane .
the photon is represented as an open string that interacts with the d - particles , resulting a retardation @xmath66 .
the new _ fermi _ data appear to rule out this model as well as other models that predict such a retardation .
the dispersion effect will be smaller if the dispersion relation has a quadratic dependence on @xmath50 as suggested by effective field theory considerations @xcite .
this will obviate the limits on @xmath67 given above .
these considerations also lead to the prediction of vacuum birefringence ( see next section ) .
a possible model for quantizing space - time which has been actively investigated is _ loop quantum gravity _ ( see the review given in ref . @xcite and references therein . )
a signature of this model is that the quantum nature of space - time can produce a vacuum birefringence effect .
( see also the eft treatment in ref .
this is because electromagnetic waves of opposite circular polarizations will propagate with different velocities , which leads to a rotation of linear polarization direction through the angle @xmath68t/2=\xi k^2 t/2m_{pl } \label{rotation}\ ] ] for a plane wave with wave - vector @xmath69 @xcite .
again , for simple planck - suppressed liv , we would expect that @xmath70 .
some astrophysical sources emit highly polarized radiation .
it can be seen from equation ( [ rotation ] ) that the rotation angle is reduced by the large value of the planck mass . however , the small rotations given by equation ( [ rotation ] ) can add up over astronomical or cosmological distances to erase the polarization of the source emission . therefore ,
if polarization is seen in a distant source , it puts constraints on the parameter @xmath71 .
observations of polarized radiation from distant sources can therefore be used to place an upper bound on @xmath71 .
equation ( [ rotation ] ) indicates that the higher the wave number @xmath72 , the stronger the rotation effect will be .
thus , the depolarizing effect of space - time induced birefringence will be most pronounced in the energy range .
it can also be seen that the this effect grows linearly with propoagation time .
the difference in rotation angles for wave - vectors @xmath73 and @xmath74 is @xmath75 replacing the time @xmath76 by the distance from the source to the detector , denoted by @xmath77 .
the best secure bound on this effect , @xmath78 , was obtained using the observed 10% polarized soft emission from the region of the crab nebula @xcite .
clearly , the best tests of birefringence would be to measure the polarization of from grbs .
we note that linear polarization in x - ray flares from grbs has been predicted @xcite .
most bursts have redshifts in the range 1 - 2 corresponding to distances of greater than a gpc .
should polarzation be detected from a burst at distance @xmath77 , this would place a limit on @xmath79 of @xmath80 where @xmath81 is the distance to the burst in units of 0.5 gpc @xcite .
detectors that are dedicated to polarization measurements in the x - ray and energy range and which can be flown in space to study the polarization from distant astronomical sources are now being designed @xcite .
shortly after the discovery of the 3k cosmogenic background radiation ( cbr ) , greisen @xcite and zatsepin and kuzmin @xcite predicted that pion - producing interactions of such cosmic ray protons with the cbr should produce a spectral cutoff at @xmath82 50 eev .
the flux of ultrahigh energy cosmic rays ( uhecr ) is expected to be attenuated by such photomeson producing interactions .
this effect is generally known as the `` gzk effect '' .
owing to this effect , protons with energies above @xmath60100 eev should be attenuated from distances beyond @xmath83 mpc because they interact with the cbr photons with a resonant photoproduction of pions @xcite .
let us consider the photomeson production process leading to the gzk effect .
near threshold , where single pion production dominates , @xmath84 using the normal lorentz invariant kinematics , the energy threshold for photomeson interactions of uhecr protons of initial laboratory energy @xmath85 with low energy photons of the cbr with laboratory energy @xmath86 , is determined by the relativistic invariance of the square of the total four - momentum of the proton - photon system .
this relation , together with the threshold inelasticity relation @xmath87 for single pion production , yields the threshold conditions for head on collisions in the laboratory frame @xmath88 for the proton , and @xmath89 in terms of the pion energy , where m is the rest mass of the proton and m is the rest mass of the pion @xcite . if li is broken so that @xmath90 , the threshold energy for photomeson is altered .. this process would be allowed by liv in the case where @xmath91 is negative , producing a sharp cutoff in the uhecr proton spectrum .
( for more details , see refs .
@xcite . ] because of the small liv perturbation term , the square of the four - momentum is shifted from its li form so that the threshold condition in terms of the pion energy becomes @xmath92 where @xmath93 , again in units where the low energy velocity of light is unity .
equation ( [ livpi ] ) is a quadratic equation with real roots only under the condition @xmath94 defining @xmath95 ev with @xmath96 k , equation ( [ root ] ) can be rewritten @xmath97 if liv occurs and @xmath98 , photomeson production can only take place for interactions of cbr photons with energies large enough to satisfy equation ( [ cg ] ) .
this condition , together with equation ( [ livpi ] ) , implies that while photomeson interactions leading to gzk suppression can occur for `` lower energy '' uhe protons interacting with higher energy cbr photons on the wien tail of the spectrum , other interactions involving higher energy protons and photons with smaller values of @xmath86 will be forbidden .
thus , the observed uhecr spectrum may exhibit the characteristics of gzk suppression near the normal gzk threshold , but the uhecr spectrum can `` recover '' at higher energies owing to the possibility that photomeson interactions at higher proton energies may be forbidden .
we now consider a more detailed quantitative treatment of this possibility , _ viz .
_ , gzk coexisting with liv .
the kinematical relations governing photomeson interactions are changed in the presence of even a small violation of lorentz invariance .
the modified kinematical relations containing liv have a strong effect on the amount of energy transfered from a incoming proton to the pion produced in the subsequent interaction , _
i.e. _ , the inelasticity @xcite .
the primary effect of liv on photopion production is a reduction of phase space allowed for the interaction .
this results from the limits on the allowed range of interaction angles integrated over in order to obtain the total inelasticity . for real - root solutions for interactions involving higher energy protons
, the range of kinematically allowed angles becomes severely restricted .
the modified inelasticity that results is the key in determining the effects of liv on photopion production .
the inelasticity rapidly drops for higher incident proton energies .
figure [ inelasticity ] shows the calculated proton inelasticity modified by liv for a value of @xmath99 as a function of both cbr photon energy and proton energy @xcite .
other choices for @xmath91 yield similar plots .
the principal result of changing the value of @xmath91 is to change the energy at which liv effects become significant . for a choice of @xmath100 , there is no observable effect from liv for @xmath101 less than @xmath102 eev . above this energy ,
the inelasticity precipitously drops as the liv term in the pion rest energy approaches @xmath103 . as a function of cbr photon energy and proton energy @xcite . ] with this modified inelasticity , the proton energy loss rate by photomeson production is given by @xmath104\ ] ] where we now use @xmath36 to designate the energy of the photon in the cms , @xmath105 is the photon threshold energy for the interaction in the cms , @xmath106 denotes the inelasticity , and @xmath107 is the total @xmath0-p cross section with contributions from direct pion production , multipion production , and the @xmath108 resonance .
the corresponding proton attenuation length is given by @xmath109 , where the energy loss rate @xmath110 .
this attenuation length is plotted in figure [ attn ] for various values of @xmath111 along with the unmodified pair production attenuation length from pair production interactions , @xmath112 .
( solid lines ) , shown with the attenuation length for pair production unmodified by liv ( dashed lines ) . from top to bottom , the curves are for @xmath113 ( no lorentz violation ) @xcite . ]
the effect of by a very small amount of liv on the uhecr spectrum was analytically calculated in ref .
@xcite in order to determine the resulting spectral modifications .
it can be demonstrated that there is little difference between the results of using an analytic calculation _ vs. _ a monte carlo calculation ( _ e.g. _ , see ref .
@xcite ) . in order to take account of the probable redshift evolution of uhecr production in astronomical sources
, they took account of the following considerations : + ( _ i _ ) the cbr photon number density increases as @xmath114 and the cbr photon energies increase linearly with @xmath115 .
the corresponding energy loss for protons at any redshift @xmath55 is thus given by @xmath116 .
\label{eq5}\end{aligned}\ ] ] ( _ ii _ ) they assumed that the average uhecr volume emissivity is of the energy and redshift dependent form given by @xmath117 where @xmath118 is the initial energy of the proton at the source and @xmath119 .
for the source evolution , we assume @xmath120 with @xmath121 so that @xmath122 is roughly proportional to the empirically determined @xmath55-dependence of the star formation rate . @xmath123 and @xmath124 are normalized fit the data below the gzk energy . using these assumptions
, one can calculate the effect of liv on the uhecr spectrum .
the results are actually insensitive to the assumed redshift dependence because evolution does not affect the shape of the uhecr spectrum near the gzk cutoff energy @xcite . at higher energies where the attenuation length may again become large owing to an liv effect
, the effect of evolution turns out to be less than 10% .
the curves calculated in ref .
@xcite assuming various values of @xmath91 , are shown in figure [ auger ] along with the latest _ auger _
data from ref .
they show that _ even a very small amount of liv that is consistent with both a gzk effect and with the present uhecr data can lead to a `` recovery '' of the uhecr spectrum at higher energies .
_ , taking @xmath125 ( see text ) . from top to bottom , the curves give the predicted spectra for @xmath126 ( no lorentz violation ) @xcite . ]
stecker and scully @xcite have updated compared the theoretically predicted uhecr spectra with various amounts of liv to the latest _ auger _
data from the procedings of the 2009 international cosmic ray conference @xcite , @xcite .
this update is shown in figure [ auger ] .
the amount of presently observed gzk suppression in the uhecr data is consistent with the possible existence of a small amount of liv .
the value of @xmath91 that results in the smallest @xmath127 for the modeled uhecr spectral fit using the observational data from _ auger _
@xcite above the gzk energy .
the best fit liv parameter found was in the range given by @xmath91 = @xmath128 , corresponding to an upper limit on @xmath91 of @xmath129 .
an effective field theory approximation for possible liv effects induced by planck - scale suppressed quantum gravity for @xmath130 was considered in ref .
these authors explored the case where a perturbation to the energy - momentum dispersion relation for free particles would be produced by a cpt - even dimension six operator suppressed by a term proportional to @xmath131 .
the resulting dispersion relation for a particle of type @xmath132 is @xmath133 in order to explore the implications of our constraints for quantum gravity , one can take the perturbative terms in the dispersion relations for both protons and pions , to be given by the dimension six dispersion terms in equation ( [ qg ] ) above . making this identification
, the liv constraint of @xmath134 in the fiducial energy range around @xmath135 eev indirectly implies a powerful limit on the representation of quantum gravity effects in an effective field theory formalism with planck suppressed dimension six operators . equating the perturbative terms in both the proton and pion dispersion relations , one obtains the relation @xcite @xmath136 where the pion fiducial energy is taken to to be @xmath137 , as at the @xmath108 resonance that dominates photopion production and the gzk effect @xcite .
equation ( [ dim6 ] ) , together with the constraint @xmath138 , indicates that any liv from dimension six operators is suppressed by a factor of at least @xmath139 , except in the unlikely case that @xmath140 .
these results are in agreement with those obtained independently by maccione et al . from the monte carlo
runs @xcite .
it can thus be concluded that an effective field theory representation of quantum gravity with dimension six operators that suppresses liv by only a factor of @xmath141 _ i.e. _ @xmath142 , is effectively ruled out by the uhecr observations .
as we have seen ( see figure [ auger ] ) , even a very small amount of liv that is consistent with both a gzk effect and with the present uhecr data can lead to a `` recovery '' of the primary uhecr spectrum at higher energies .
this is the clearest and the most sensitive evidence of an liv signature .
the `` recovery '' effect has also been deduced in refs .
@xcite and @xcite .. however , we have noted that the ` quasi - vacuum erenkov radiation ' of pions by protons in this case will cut off the proton spectrum and no `` recovery '' effect will occur . ] in order to find it ( if it exists ) three conditions must exist : ( _ i _ ) sensitive enough detectors need to be built , ( _ ii _ ) a primary uhecr spectrum that extends to high enough energies ( @xmath60 1000 eev ) must exist , and ( _ iii _ ) one much be able to distinguish the liv signature from other possible effects .
we now turn to examining the various techniques that can be used in the future in order to look for a signal of liv using uhecr observations .
as can be seen from the preceding discussion , observations of higher energy uhecrs with much better statistics than presently obtained are needed in order to search for the effects of miniscule lorentz invariance violation on the uhecr spectrum .
such an increased number of events may be obtained using much larger ground - based detector arrays .
the _ auger _
collaboration has proposed to build an _ auger north _ array that would be seven times larger than the present southern hemisphere auger array ( http://www.augernorth.org ) . further into the future , space - based telescopes designed to look downward at large areas of the earth s atmosphere as a sensitive detector system for giant air - showers caused by trans - gzk cosmic rays .
we look forward to these developments that may have important implications for fundamental high energy physics .
two potential spaced - based missions have been proposed to extend our knowledge of uhecrs to higher energies .
jem - euso _ ( the extreme universe space observatory ) @xcite , a one - satellite telescope mission proposed to be placed on the japanese experiment module ( jem ) on the international space station .
the other is _ owl _ ( orbiting wide - angle light collectors ) @xcite , a two satellite mission for stereo viewing , proposed for a future free - flyer mission .
such orbiting space - based telescopes with uv sensitive cameras will have wide fields - of - view ( fovs ) in order to observe and use large volumes of the earth s atmosphere as a detecting medium .
they will thus trace the atmospheric fluorescence trails of numbers of giant air showers produced by ultrahigh energy cosmic rays and neutrinos .
their large fovs will allow the detection of the rare giant air showers with energies higher than those presently observed by ground - based detectors such as _ auger_. such missions will thus potentially open up a new window on physics at the highest possible observed energies .
the _ fermi _ timing results for grb090510 rule out and string - inspired d - brane model predictions as well as other quantum gravity predictions of a retardation of photon velocity that is simply proportional to @xmath143 because they would require @xmath144 .
more indirect results from birefringence limits , the non - decay of 50 tev from the crab nebula , and the tev spectra of nearby agns also place severe limits on violations of special relativity ( liv ) .
limits on lorentz invariance violation from observations of ultrahigh energy cosmic - rays provide severe constraints for other quantum gravity models , appearing to rule out retardation that is simply proportional to @xmath145 .
various effective field theory frameworks lead to such energy dependences .
new theoretical models of planck scale physics and quantum gravity need to meet all of the present observational constraints .
one scenario that may be considered is that gravity , _ i.e. _ @xmath146 , becomes weaker at high energies .
we know that the strong , weak and electromagnetic interactions all have energy dependences , given by the running of the coupling constants .
if @xmath146 decreases , then the effective @xmath147 would decrease and the effective @xmath148 would increase . in that case , the space - time quantum scale would be less than the usual definition of @xmath64 .
such speculation is presently _ cogitare ex arcis _ , but might be plausible if a transition to a phase where the various forces are unified occurs at very high energies @xcite . at the time of the present writing , high energy astrophysics observations have led to strong constraints on liv .
currently , we have no positive evidence for liv .
this fact , in itself , should help guide theoretical research on quantum gravity , already ruling out some models .
will this lead to a new null result comparable to michelson - morley ? will a totally new concept be needed to describe physics at the planck scale ? if all of the known forces are unified at the planck scale , this would not be surprising .
one thing is clear : a consideration of all empirical data will be necessary in order to finally arrive at a true theory of physics at the planck scale . | the topic of lorentz invariance violation ( liv ) is a fundamental question in physics that has taken on particular interest in theoretical explorations of quantum gravity scenarios .
i discuss various @xmath0-ray observations that give limits on predicted potential effects of lorentz invariance violation . among these
are spectral data from ground based observations of the multi - tev @xmath0-rays from nearby agn , _ integral _ detections of polarized soft @xmath0-rays from the vicinity of the crab pulsar , _ fermi gamma ray space telescope _ studies of photon propagation timing from @xmath0-ray bursts , and _ auger _
data on the spectrum of ultrahigh energy cosmic rays .
these results can be used to seriously constrain or rule out some models involving planck scale physics .
possible implications of these limits for quantum gravity and planck scale physics will be discussed .
address = astrophysics science division , nasa goddard space flight center , greenbelt , md 20771 , usa |
the sun s 22-year cycle of activity is most clearly seen in solar surface magnetic fields ( see for example the review by for a good discussion of the solar dynamo and solar magnetic fields ) .
although the sunspot number ( ssn ) is the most readily available indicator of the state of the solar cycle , the magnetic fields over the whole surface provide a more complete measure of the dynamo process .
a variety of questions arise in using the surface magnetic fields to study the solar cycle which fields are most important , the weak general field or the strong field and associated sunspots , how do the field strengths change over various time scales from instabilities with changes in seconds or less to trends lasting many decades .
the connection between photospheric magnetic fields and magnetic fields near earth or at interplanetary spacecraft requires knowledge of the field strength at the solar surface .
is the overall strength of the magnetic field stationary or does it have any multi - decade trends ( cf : ) ?
these surface fields are also needed to calculate the strength measured by interplanetary spacecraft such as _ ulysses _ @xcite .
the topology of the field in the solar surface regions introduces additional uncertainty since the field of one polarity lower in the atmosphere may be canceled by the opposite polarity before it can emerge into the corona and heliosphere @xcite .
questions such as these involving magnetic field strength can only be addressed when we are confident in the quantitative interpretation of the measured quantity . the 150-foot solar tower telescope on mt .
wilson has been dedicated to the synoptic measurement of solar surface magnetic fields using a babcock magnetograph observing the fe i spectral line at @xmath0 .
although other spectral lines and observing techniques offer advantages over these choices , the uniformity of the data set extending back to 1967 is a strong reason to leave the observing system as nearly unaltered as possible .
the application of these data to modeling the heliospheric magnetic field then depends on our ability to understand how to interpret the magnetograms .
an early step toward this goal was carried out by and subsequently extended by and using a cross - correlation of fields observed simultaneously from two or more pairs of spectral samples .
a compelling feature of the scatter diagrams from these simultaneous measurements is the linearity between the magnetic fields obtained from the different spectral samplings .
each spectral sample measures a different altitude in the solar atmosphere and is influenced by different thermal properties of the atmosphere at that altitude . if the altitude dependent part of the relationship were also to depend on the field strength at each point on the solar surface , the scatter diagram would not be confined to a straight - line band passing through the zero - zero point . the observed straight line distributions found on the scatter diagram establish that variable filling factor is the only cause for varying field strength , all flux tubes being essentially identical .
the fact that the slope is not unity can be used to study the altitude dependence of the flux tubes . throughout this paper
we will use the term `` magnetic field strength '' and usually mean the apparent field strength which is the product of a flux tube field strength and a filling factor .
this quantity is actually the magnetic flux per unit area and is the quantity we wish to measure .
the extensive literature devoted to the determination of the flux tube fields and the filling factors is of interest to us and is referenced where appropriate below .
however , questions of flux tube physics are of secondary importance to our objectives in this paper .
where the flux tube field strength is discussed , it is made explicit .
otherwise , the term magnetic field strength should be read to mean the magnetic flux per unit area as measured in the observed pixels .
the essential step in our study of magnetic field strength is the choice of a comparison line .
our objective is to learn the relationship between measured circular polarization of a relatively large pixel ( 12 arcseconds squared or 20 arcseconds squared ) and the magnetic flux in that area .
we are less concerned with questions of the structure of magnetic flux tubes or the strength of the field within the flux tubes we only seek to learn the best way to estimate the product of the filling factor and the magnetic field in the flux tubes .
in addition , we need to be able to observe the line simultaneously with @xmath0 . for this purpose the line at @xmath2 is ideally suited due to its large width and depth and due to the good linearity of its line wings . for a related objective of determining the flux tube magnetic field strength the comparison line chosen by
was @xmath11 to take advantage of the fact that the two lines have differing @xmath12 values but very similar thermal response and line formation properties .
however in both this paper and the subsequent study by the reference to @xmath2 was retained with the assumption that magnetic fluxes from this line are obtained directly from the polarization factors and the line profile .
an interesting citation to a private communication by livingston in this last paper suggests that the magnetic fields derived from @xmath2 depend on position within this line .
we verify below that the deduced magnetic field strength for any particular area on the sun s surface that depends on the spectral sampling of @xmath2 .
our previous results based on @xmath2 yielded a correction factor of 4.5 to be applied as a multiplier for the magnetic fields obtained from @xmath0 whereas found only a factor of two .
our correction factor is larger due partly to the fact that the shifted zeeman components at @xmath0 are shifted beyond the sampling pass - band of the mwo babcock magnetograph .
we refer to the reduced apparent field due to the large line shift as the `` classical saturation effect '' since it depends only on the actual line profile , the strength of the flux tube magnetic fields and the spectral resolution of the observing system .
as a first step in establishing a more complete understanding of the scale factors relating different observing methodologies we have carried out a new series of observations utilizing simultaneous magnetic field measurements with six different spectral configurations .
the mwo system integrates all spectral channels continuously with all integrator registers being emptied sequentially at a rate of 40 hz . due to this feature of the system image motion and smearing
affects all spectral samples identically so that the spatial average of spatial inhomogenieties is also identical
. thus ratios of magnetic fields calculated from different sampling pairs yield relative magnetic fields from identical solar surface structures . starting with the concept that the magnetic field from @xmath2 indicates the true magnetic flux
, the ratio @xmath13 should be the scale factor that converts the mwo synoptic program magnetic fields to the appropriate value emerging from the photospheric layers .
for the actual observations we need to specify the spectral sampling as well as the spectral line .
the study by used @xmath14m and @xmath15m to yield the scale factor of 4.5 .
contrary to the assumptions of , and we find that the field measured by @xmath2 depends on the spectral sampling so that we must generalize the notation .
we do this by defining a new quantity @xmath16 which relates the magnetic field for the spectral line at wavelength @xmath17 having spectral sampling offset by @xmath18 to the magnetic field for the spectral line at wavelength @xmath19 having spectral sampling of @xmath20 with @xmath17 and @xmath19 in ngstroms and @xmath18 and @xmath20 in m .
thus we write : @xmath21 so that the previously defined @xmath4 is : @xmath22 the fact that the magnetic field derived from @xmath2 depends on the sampling separation is not a surprise in view of the complexity of the atmospheric flux tube structure but it does present us with the additional task of determining which portion of this line is most appropriate as an indicator of the emergent magnetic flux . during the interval 18 april 2007 to 8
may 2007 a set of five special magnetograms were obtained using the 24-channel system described by with the blue , 10-channel fiberoptic reformattor centered on @xmath2 instead of the @xmath23 line of nad .
this configuration provided us with five line pairs having separations of @xmath24m .
the observations in 1992 were made using a different spectrograph configuration wherein the spectral line sampling was at @xmath25m .
the sampling of @xmath0 was left as in the synoptic program at @xmath26m . during the scanning process
the light entering the spectrograph is unfiltered .
the dispersed light passes through narrow - band blocking filters where it falls on all spectral sampling pickups simultaneously .
the output currents from all photomultiplier tubes are summed individually and simultaneously into a set of registers .
we refer to the sequence from the entry window of a single spectral sample to the summing register as a channel .
the final digitization numbers for all channels for each pixel then originate from identical portions of the solar surface even though the sampled area is not precisely defined by the entrance aperture due to distortions of the solar image by the earth s atmosphere .
we compute magnetic fields from each spectral sampling pair and compare the derived fields in a scatter - diagram format . since we expect the interrelationships to depend on position within the solar image , we restrict the scatter diagrams to pixels having a specified range in center - to - limb angle @xmath5 .
.the correlation slopes @xmath27 and their errors as a function of @xmath20 and @xmath28 . [ cols="<,>,<,>,<,>,<,>,<",options="header " , ] to investigate the properties of the stokes @xmath3 profiles and deduced magnetic field strengths based on such profiles , 23 line scan sets were obtained over a period of several months during the declining phase of solar cycle 23 .
we use observations obtained between june and august 2007 . at this stage of the solar cycle ,
the number of active regions available for study is limited .
the observing system is that described by and utilized the 10-channel fiber - optic reformatter normally used for magnetograms in the @xmath29 line of sodium ( see for the details of the spectral sampling ) . with this system
we obtain spectral line profiles from a square selected by our entrance aperture which has sides of 12 arc - sec ( other apertures are available but were not used in this study ) .
this entrance aperture tracks solar rotation using the nominal differential rotation rate for the position observed . at the final spectrograph
focus the spectrum is sampled by one of four sets of fiber - optic image reformatters that are carried on two independently movable stages .
a small portion of the solar spectrum including the target spectral line is scanned by moving its stage alternately redward and blueward over the solar spectrum repeatedly for a total duration of 20 minutes with each scan requiring 20 seconds .
the second stage is left stationary and its output is used to remove sky transparency fluctuations .
the length of each scan is 1 and the spectral output from each fiber is sampled at an interval of 5.31m ( the spectral resolution is about 30m ) .
the temporal resolution was not used and all scans were summed to a single profile for each spectral pickup .
finally the ten profiles were shifted by the known fiber - optic separation so that all are superposed for a single final profile .
the offset between the fiber - optic inputs means that the portion of the spectrum scanned is also offset .
the overall scan length was large enough that all ten inputs covered the range of interest for comparison to the magnetogram scatter diagram results .
the circular polarization modulation was retained to permit the calculation of a single stokes @xmath3 profile . following the convention described by the blue shifted component
is left circularly polarized in a positive magnetic field region .
although we retain knowledge of the polarity of each set of line profiles , we do not consider the sign of the field to be important in this application and present the stokes @xmath3 parameter and the derived magnetic fields as if all polarities are positive .
table [ t2 ] gives the parameters for these scans including the nominal magnetic field strength based on the line bisectors for a bisector width @xmath30 of @xmath31m .
the line scan set numbers given in table [ t2 ] are not sequential because a number of line scan sets for @xmath0 were also obtained during this time frame and these observations will be discussed in a future publication .
the stokes @xmath3 profiles from the observations summarized in table [ t2 ] are shown in figure [ f5a ] .
the scale of the stokes @xmath3 plots is based on a continuum @xmath32 of unity .
the discussion in the following sections emphasizes the set of line scans from july 13 , 2007 since that region was more highly magnetized than was the case for subsequent observations .
the apparent noise for the low field strength stokes @xmath3 profiles comes from a combination of factors including photon noise and the formation of interference fringes by filters , windows and other plane - parallel optics in the system .
we have mitigated these residual fringes by tilting the blocking filter but the effect is not eliminated by this technique .
we estimate the uncertainty in the stokes @xmath3 profile over the spectral range needed for the field determination is @xmath33 .
the line slope is @xmath34(m)@xmath35 at the points @xmath31m from line center so that the zeeman shift error due to this noise is about @xmath36m for an estimated uncertainty in the field strength of 0.1 gauss as observed by the channel at @xmath37 .
we can also estimate the photon noise error in the deduced magnetic field using measurements from the normal synoptic program magnetograms .
the integration time for each magnetogram pixel is 0.127 seconds while the integration time on each pixel in the spectral scans is very similar at 0.1 seconds .
the final profile information is the result of combining scans from all 10 fiber - optic reformattor channels and from all 60 scans .
thus the relative photon noise for each line profile point compared to the synoptic program magnetograms should be reduced by a factor of @xmath38 .
the line scanning process provides four times as many spectral points compared to the magnetogram leading to a further reduction of 2 in the expected photon noise . according to the discussion of section [ scat ] ,
the gaussian width from the magnetograph distribution function is 4 gauss for the above spectral channel so that again we arrive at an estimate of 0.1 gauss for the spectral scans . for observations with the synoptic program channel at @xmath0
the equivalent error would be 0.025 gauss .
these errors are substantially lower than has been previously reported and allow us to study the magnetic fields in regions that would normally be classified as unmagnetized . for this section
, we exclusively use the profile sequence number 022 of july 13 , 2007 to illustrate different ways of interpreting the profiles .
figure [ f6 ] compares the quiet and plage profiles .
the two states of circular polarization are combined for the low magnetization profile but separated for the magnetized case .
because we are not concerned here with doppler shifts , the low magnetization line has been shifted so its bisector position coincides with the average bisector position for the two magnetized components at a bisector half - width @xmath30 of 75 m .
this parameter measures the product of the filling factor and the flux tube field strength and represents the magnetic flux emerging from the pixel .
not only do the two circularly polarized line profiles differ significantly from the quiet sun profile , they also differ from each other in such a way as to yield a magnetic flux which depends on position within the profile .
the strength of this line and its high altitude of formation at line core make it difficult to attribute this effect to typical flux tube physics where thermal effects can produce line weakening or strengthening relative to the surrounding quiet atmosphere .
clearly then this spectral line does not define a single magnetic flux and our problem becomes that of deciding how to interpret the observations in such a way as to find the best estimate of the magnetic flux emerging from the solar atmosphere . the stokes @xmath3 profile is often used as the basis for this estimate .
as an alternate way of thinking about the different parts of the spectral line it is worth considering how a babcock magnetograph might work .
if we could simultaneously track the spectral line in both states of circular polarization , the babcock magnetograph would evaluate the magnetic field on the basis of central wavelengths where the red and blue wing intensities are equal in each state of polarization . the wavelength difference between these two balance points
would then define a magnetic field from the spectrally resolved line profiles .
in general , a line bisector can be considered as either a function of the intensity relative to the continuum intensity or as a function of the wavelength separation between the equal intensity points . in view of the dependence of the line core intensity on the strength of the magnetic field ,
we find it advantageous to use bisector half - width @xmath30 as the defining parameter .
corresponding bisectors for the two states of circular polarization at a fixed but arbitrary value of @xmath30 are shown in figure [ f6 ] . the separation between the two central wavelengths at this @xmath30
is indicated as @xmath39 from which a magnetic flux per unit area @xmath40 can be deduced .
for the observation shown in figure [ f6 ] we show the derived bisector field strength as a function of @xmath30 in figure [ f7 ] .
figure [ f8 ] which gives both the stokes @xmath3 parameter and the slope of the quiet sun line profile .
the scale factor for the stokes @xmath3 has been adjusted so that the curves agree near the line center . the magnetic field can be calculated from this scaling factor if we use the linear approximation .
this gives a field strength based on the point - wise slope of the spectral line and the corresponding stokes @xmath3 value so that a magnetic field can be deduced from both wings of the spectral line .
we refer to a magnetic field deduced this way as @xmath41 . these fields along with the values of @xmath42 are shown in figure [ f9 ] .
note for this plot we use @xmath43 .
while @xmath41 can be calculated for both red and blue wings of the line , @xmath42 applies to both wings equally and has been plotted symmetrically .
although the stokes @xmath3 parameter gives independent magnetic fields in the blue and red wings of the line , a babcock magnetograph responds to the average magnetic field from the two line wings .
consequently , apart from figures [ f8 ] and [ f9 ] , the magnetic field derived from the stokes @xmath3 parameter has been made symmetric by averaging the blue and red wing field values .
the comparison between @xmath41 and @xmath42 indicates that the pointwise slope leads to a smaller deduced field strength than the bisector by 25% near the line core , a larger field in the mid part of the line wing then good agreement in the outer line wings .
we do not have a definitive interpretation of this result but point out that it could be a consequence of microturbulent broadening of the line core in the flux tube .
the influence of variable microturbulent line broadening on the interpretation of spectropolarimetry of @xmath0 has been discussed recently by . because the line wings are nearly linear over much of the line , the classical saturation effect does not influence the measurement .
the classical saturation factor derived from the stokes @xmath3 parameter is shown in figure [ f10 ] . in order to illustrate @xmath44 for a case of interest ,
a flux tube field of 1500 gauss was used for this calculation although the appropriate field strength for the line core is closer to 400 gauss according to table [ t1 ] .
even with this higher than appropriate field assumption , the value of @xmath44 deviates from unity by only 10% .
the @xmath2 has nine zeeman components . as a simple numerical test
, the unmagnetized profile was shifted and weighted by these nine amounts and a simulated profile was then produced .
this numerical test gave essentially the same result as we find by shifting the line an amount equal to the weighted mean of the zeeman factors .
we conclude from this simulation that the saturation factor is not substantially influenced by the presence of the nine sub - components .
these parts of the line may however play a role in the line transfer process in a manner that is not captured by our simple approach ; the multiple zeeman components may produce the equivalent of an excess of microturbulence .
we conclude that the dependence of the deduced field strength on position within the spectral line for @xmath2 is not consequence of the classical saturation effect even for points near the line core .
the @xmath44 parameter can also be derived for the @xmath0 line . in this case
the greater curvature of the line profile leads to @xmath45 at our working point of @xmath26m .
the factor is only 35% of the slope of the relationship indicated by figure [ f3 ] confirming that thermodynamic effects play a dominant role in the modification of the apparent magnetic field strength .
at each point on the solar surface we now have a range of estimates for the magnetic field : @xmath47 with @xmath48 and @xmath49 , @xmath42 and @xmath41 .
we should choose the estimate which gives the product of the field strength and filling factor appropriate to the pixel area contribution to the magnetic flux above the solar atmosphere .
for this purpose , the portion of the line nearest the line center would appear to be the best choice since this part of the line is highest in the atmosphere .
however , as we show in this section , the choice has to be tempered by the fact that the line core shows an excess range of variability and is vulnerable to line broadening uncertainties from microturbulence . for portions of the solar surface outside of sunspots
the apparent magnetic field strength is not greater than about 500 gauss . for zeeman shifts from fields of this strength
the linear approximation is applicable and the field deduced from a magnetograph observation depends on the spectral line slope at the point of the line where the spectrograph sampling is selected .
this means that the magnetogram field is consistent with what we derived above from the ratio of the stokes @xmath3 parameter to the local spectral line slope
i.e. the magnetogram magnetic field should be consistent with the stokes @xmath3 magnetic field interpretation of the spectral line rather than the bisector interpretation of the spectral line even though the the babcock magnetograph follows the average spectral line bisector .
thus , it is reasonable to expect that the observed magnetogram correlations shown in figures [ f1 ] to [ f3 ] and table [ t0 ] between different deduced fields should be consistent with the stokes @xmath3 deduced fields .
the line profile observations come from a relatively small number of selected regions on the solar surface that may or may not be representative of a general case .
however , since many of the slopes from different spectral sampling configurations are well correlated , these particular observed profiles must be representative of the general case since otherwise , there would be a wider scatter in the correlation diagrams .
thus the implied field values from different parts of the line profile can be normalized so that the stokes @xmath3 field strengths and bisector field strengths give scale factors that can be used in bringing the different deductions to a common value .
the 23 sets of line scans have been reduced to yield the three estimates of the field strength indicated above .
the value of @xmath50 was chosen to provide a set of field strengths roughly similar to the other two estimates .
the value of @xmath51 was interpolated in @xmath28 to be appropriate to the position of each line scan field strength function .
selected line scan results illustrating this comparison are shown in figure [ f11 ] .
the s indicate the rescaled correlation coefficients obtained from the magnetograms , the solid and dashed lines give the symmetrized fields from figure [ f9 ] .
in addition to the cases of @xmath52 from table [ t0 ] we have included the implied slope at @xmath2@xmath25m by combining the slope in figure [ f3 ] with the correlations from the observations of 1991 .
this point is shown as the symbol .
the choice of cases to present in figure [ f11 ] was made to illustrate four points : 1 ) the weaker field cases are erratic , 2 ) the bisector fields show less dependence on position in the line than is the case for the stokes @xmath3 field , 3 ) for the stronger field cases the slope dependence on spectral sampling is similar to the stokes @xmath3 dependence on @xmath53 and 4 ) there does not seem to be a dependence of the results on the center - to - limb angle . for field strengths above about 40 gauss , the patterns for the cases shown in the top four plots are typical . for the weaker fields there is a considerable range of behavior with that for the bottom left case being an extreme .
the implied line core field for this case is only 20% of that implied at @xmath54m .
other cases of weak field actually have the core field greater than that at @xmath54m .
in contrast the higher field cases show the pattern of the bisector field being less dependent on @xmath30 than the stokes @xmath3 field is on @xmath53 for portions of the line nearer the line core .
since the stokes @xmath3 field @xmath55 gives a good reproduction of the correlation coefficients from the scatter diagrams and since the core parts of the bisector field @xmath42 are less sensitive to @xmath30 than @xmath41 is to @xmath53 , we adopt a model that the most appropriate field to use in determining the magnetic flux above the solar atmosphere is @xmath42 where @xmath30 approaches zero .
we believe that correlation coefficient field as well as the stokes @xmath3 field near the line core are influenced by extra line broadening in the magnetized flux tube .
since we are unable to observe this line profile we must use the profile from a standard quiet region which has a steeper slope near the line core so that the apparent falloff in field strength is due to division by too large a number .
it is possible to estimate some properties of the spectral line profile in the magnetized flux tube by integrating the stokes @xmath3 parameter .
this gives a shape near the line core which is indeed broader compared to that of the unmagnetized line in the case of the observations of july 13 , 2007 .
the bisector field does not depend on the line slope but only on the requirement of equal intensity on the red and blue wings so that its falloff toward the line core must be a consequence of some other effect such as flux cancelation . to combine the results of the magnetogram scatter diagrams and the line profile data using our adopted model , we identify the line profile quantity @xmath56 as corresponding to @xmath50 used in the scatter diagram and we identify the line profile quantity @xmath57 as the recommended field .
the ratio : @xmath58 is then a factor which will convert @xmath59 into the final correction factor for our observed magnetograms using @xmath60 in the mwo synoptic program . we show the derived values of @xmath61 in figure [ f12 ] as a function of @xmath56 .
it is noteworthy that for areas with fields @xmath56 larger than about 30 gauss the values of @xmath61 are quite stable and do not depend on the center - to - limb angle .
the variations in @xmath61 for smaller fields are not due to error of measurement of the fields but rather are a consequence of some structural property of the solar surface . for our purpose
, we do not need to understand the cause for this behavior and can simply adopt the value of @xmath61 to be @xmath62 which is the average and error of the mean of all the cases with @xmath63gauss .
examination of figure [ f5a ] reveals that the raw @xmath3 profiles for the weaker field cases are much less regular than the stronger field cases .
the flux tube structure implies that the weaker field areas have a small number of flux tubes .
the @xmath3 irregularities may be a consequence of statistical fluctuations around a fairly stable mean structure .
theoretical studies of the @xmath2 line would be helpful in understanding the properties of this line in the magnetized and unmagnetized solar atmosphere as well as providing a separate indication of the relationship between the various observed magnetic field strength indicators and the magnetic flux emerging into the heliosphere .
based on the adoption of the @xmath57 as the recommended field value , we can use the magnetogram correlation diagrams to obtain a new scale factor for the mt .
wilson 150-foot tower telescope synoptic program which utilizes the fei line at @xmath0 . for notation
, we retain the traditional quantity @xmath4 first defined by wherein the recommended field @xmath64 is : @xmath65 the previous correlation analysis by as represented by was : @xmath66 the present analysis is shown in figure [ f13 ] and gives : @xmath67 note that this correction factor is simply to take into acount line formation processes and does not include effects related to the geometry of the field .
in particular an additional @xmath68 correction is required if one adopts a model wherein the magnetic field is radial .
our principal result is given above and provides a small correction to the procedure previously applied to the mt .
wilson observatory magnetogram observations . the previous description of this correction factor as a saturation correction is false .
the various interpretations on the polarimetry of the spectral lines used in our observations differ from one another due to processes involved with transfer of radiation through an inhomogeneous , magnetized atmosphere .
a full analysis of this problem is beyond the scope of the present investigation but is obviously needed to fully understand the changes in shape of the polarized spectral lines .
we base our recommended correction factor on the fact that the magnetic field deduced from the spectral line bisector is less variable than the field deduced from the monochromatic stokes @xmath3 parameter .
we further adopt the magnetic field implied by the properties of the spectral line near its core on the grounds that this part of the line is closest to the outer edge of the solar atmosphere .
we find that the line profile properties are most stable in regions where the field strength is significantly larger than that found in the completely quiet sun .
the line profiles for the weakly magnetized regions are well determined by our system but they show substantial deviations from symmetry . we suspect that this is a consequence of the reduced effect of adjacent flux tubes as a constraint on the flux tube structure in cases where the field is so weak that there are no adjacent flux tubes .
the unconstrained flux tubes can then adopt a wider range of structure including perhaps cases where the field lines return locally to the solar interior rather than extending out of the solar atmosphere .
the fact that the error of measurement of the fields is well below 1 gauss for our line profile system means that the measured profiles can be used to constrain models of flux tubes in regions where the field strength is very low .
our determination that the differing deduced magnetic field strength for differing spectral sampling configurations has a general potential impact on the measurement of magnetic fields with other observing systems .
all programs provide maps which give a single magnetic field strength for each observed point . had the adjustment of the mt .
wilson fields been a consequence of a saturation effect unique to this system , the potential implications would also have been limited to the mt . wilson system .
however , our determination that the effect comes from a spectral transfer effect means that other observing programs need to determine how to relate their field measurements to measurements made by other systems . even within a single observing programs ,
multiple deduced magnetic fields could occur depending on the exact specification of the algorithm bringing the spectropolarimetric data to a deduced magnetic field strength .
examples where a direct comparison has been made on a pixel by pixel basis include and .
we are also in a position to update discussion from [ profandmag ] which compares our scale factor to the deductions of .
those authors concluded from advanced stokes polarimeter data that the mdi magnetic fields are @xmath10 times the correct value . after combining the factor from to the new corrected @xmath2 scale
we find that the mdi magnetic fields are @xmath8 times the recommended value at the center - to - limb distance used by .
this agreement within errors of measurement gives support to our approach in the calibration of the @xmath0 magnetograms .
the authors thank j. harvey and j. todd hoeksema for their comments on drafts of this paper .
this work has been supported by nasa , nsf and onr through a series of grants .
most recently support has come from the project nasa / hmi 16165880 - 26967-g at stanford university and from nsf through grant nsf atm-0517729 . : 1992 , analysis of magnetic fluxtubes on the solar surface from observations at mt .
wilson of a5250 ; a5233 . in : giampapa ,
m.s . , bookbinder , j.a .
_ cool stars , stellar systems , and the sun _ , _ astronomical society of the pacific conference series _ , * 26 * , 265+ | this study based on longitudinal zeeman effect magnetograms and spectral line scans investigates the dependence of solar surface magnetic fields on the spectral line used and the way the line is sampled in order to estimate the magnetic flux emerging above the solar atmosphere and penetrating to the corona from magnetograms of the mt .
wilson 150-foot tower synoptic program ( mwo ) .
we have compared the synoptic program @xmath0 line of fe@xmath1i to the line of fe@xmath1i at @xmath2 since this latter line has a broad shape with a profile that is nearly linear over a large portion of its wings .
the present study uses five pairs of sampling points on the @xmath2 line .
line profile observations show that the determination of the field strength from the stokes @xmath3 parameter or from line bisectors in the circularly polarized line profiles lead to similar dependencies on the spectral sampling of the lines with the bisector method being the less sensitive .
we recommend adoption of the field determined with the line bisector method as the best estimate of the emergent photospheric flux and further recommend the use of a sampling point as close to the line core as is practical .
the combination of the line profile measurements and the cross - correlation of fields measured simultaneously with @xmath0 and @xmath2 yields a formula for the scale factor @xmath4 that multiplies the mwo synoptic magnetic fields . using @xmath5 as the center - to - limb angle ( cla ) , a fit to this scale factor is @xmath6 .
previously @xmath7 had been used .
the new calibration shows that magnetic fields measured by the mdi system on the soho spacecraft are equal to @xmath8 times the true value at a center - to - limb position 30@xmath9
. found this factor to be @xmath10 based on a comparison the the advanced stokes polarimeter . |
soft - collinear effective theory ( scet ) @xcite , the effective theory for processes involving energetic particles , incorporates the structure of soft and collinear interactions in qcd into an effective - theory framework .
it is based on an expansion of qcd diagrams in regions where particle momenta become soft or collinear . the underlying mathematical framework is the strategy of region technique @xcite .
while the original qcd diagrams are regularized by dimensional regularization both in the ultraviolet and in the infrared , it is well known that dimensional regularization is not always sufficient to regularize also the expanded diagrams . in such cases
it is necessary to introduce additional regulators at intermediate stages , which can only be removed after the contributions from different momentum regions are combined .
a simple example where this problem occurs is the massive sudakov form factor .
the expansion of the corresponding scalar integrals at two - loop order was performed in @xcite , and it was shown that one can use analytic regulators to make the contributions of the individual momentum regions well - defined
. in analytic regularization one typically raises some propagator denominators to a fractional power @xmath0 and chooses the regulator @xmath1 in such a way that the divergences of a given diagram are softened .
the scale @xmath2 is the analogue of the renormalization scale @xmath3 introduced in dimensional regularization .
it is clear that there is a huge amount of freedom how this regularization is performed .
one may raise one or several propagators , and for each regularized propagator one can in principle use a different regulator .
in fact , one can even introduce propagators not present in the original diagram to regularize it . when expanding individual integrals , the choice of these additional regulators is largely arbitrary . in the context of scet
, analytical regularization has been used in @xcite . however , in an effective field theory analytic regularization is problematic since the additional regulators can break the symmetries of the theory . in particular , raising propagators to fractional powers will in general destroy gauge invariance , which will then only be recovered after the contributions from the individual sectors of the theory will be added and the regulator is sent to zero .
even worse , introducing such regulators may destroy some of the properties necessary to establish factorization theorems .
a crucial element of many factorization proofs , for example , is the eikonal structure of soft emissions .
the property that such emissions rearrange themselves into wilson lines will in general be broken in the presence of analytic regulators , which makes it difficult to establish factorization properties to all orders . in this letter , we consider observables such as the spectrum of transverse momentum @xmath4 of electroweak bosons in hadron collisions , or jet broadening , an event - shape in @xmath5 collisions .
these are sensitive to small transverse momenta and suffer from the problem discussed above .
the main point of our paper is that in massless theories the additional divergences only arise in the phase - space integrations . in general
, the @xmath6-dimensional integration over the transverse momentum also regularizes the light - cone propagators which arise in the effective theory .
however , this regularization is absent when phase - space constraints restrict the transverse momentum .
this explains why the problems with unregularized light - cone singularities occur for example for jet broadening , which measures the transverse momentum relative to the thrust axis , but are absent for the event - shape variable thrust , which only depends on the longitudinal momentum . instead of regularizing individual diagrams
, it is therefore sufficient to introduce the additional regularization in the phase - space integrals .
to do so , we write the phase - space integrals as integrations over light cone components ( @xmath7=0 , @xmath8 ) @xmath9 where we choose the light - cone reference vectors in the directions of large momentum flow , i.e. along the beam direction for the @xmath4 spectrum and along the thrust axis for the jet broadening .
we then define a regularized version of the usual phase - space integral as @xmath10 the factor @xmath11 regularizes the light - cone denominators which arise in scet after expanding the qcd propagators . to see that also the @xmath12 integration is regularized by the above prescription
, we can perform the @xmath13 integration using the delta - function constraint to get @xmath14 note that the momentum component @xmath12 is regularized with @xmath15 , while the regulator appears with a @xmath16 for the plus component .
other choices for the regulator are possible . in particular
, one could use the energy @xmath17 instead of @xmath18 .
the above choice is optimal since light - cone denominators are present in the effective - theory diagrams , so that the regulator ( [ regfinal ] ) does not unnecessarily complicate higher - order computations
. we can rewrite ( [ regfinal ] ) in the form of an analytically regularized propagator @xmath19^{\alpha } } \ ; \delta(k^2 ) = \left(\frac{\nu_+}{k_+}\right)^\alpha \delta(k^2 ) \,,\ ] ] for @xmath20 , which makes it clear that at @xmath21 our regularization reduces to the prescription adopted in @xcite .
let us stress that the regularization ( [ regfinal ] ) is introduced in the qcd phase - space integrals .
since these do not require the additional regularization , it is clear that qcd is recovered in the limit @xmath22 , as long as the dimensional regulator stays in place .
the regulator becomes necessary once the qcd diagrams are expanded in the different momentum regions relevant in the effective theory . in these regions
the momentum components @xmath23 scale as follows @xmath24 where @xmath25 is the ratio of the small transverse momentum over the momentum transfer .
the effective theory may in general include other momentum modes , such as ultra - soft modes whose components scale as @xmath26 , but the regularization problems we discuss only affect the modes whose transverse components scale as @xmath27 .
power counting and dimensional analysis imply the following scaling of the integration measure in the different regions : @xmath28 since the virtual corrections do not need to be regularized analytically , the measure completely fixes the dependence of a given contribution on the analytic regulator .
the dependence is furthermore very simple , it is just the scaling of the momentum component @xmath13 in the given region . from ( [ scaling ] )
we see that divergences in the analytic regulator lead to logarithms of @xmath29 upon expanding around @xmath30 . since the regions scale differently with @xmath13 , logarithmic dependence on the momentum transfer @xmath29 remains , even after the singularities themselves cancel .
the appearance of non - analytic dependence on the large momentum scale in the low - energy diagrams was called the collinear anomaly in @xcite .
the fact that the divergences in the analytic regulators must cancel between the different regions imposes strong constraints on the contributions from the individual regions .
this has been used in @xcite to show that the @xmath29 dependence associated with the collinear anomaly exponentiates . introducing the analytic regulator in the form ( [ regfinal ] ) , on the level of the phase - space integrations
, guarantees that gauge invariance is maintained . also , since the regulator is only needed to perform phase - space integrations , the effective theory is not changed at all by the presence of the regularization .
the only property that is lost is unitarity of the individual sectors of the effective theory , since the real emissions are treated differently from the virtual corrections .
this property is , however , restored when the different sectors are combined .
furthermore , in cases where the transverse momentum is not restricted and one integrates over the region of low transverse momentum , one can take the limit @xmath31 immediately and unitarity in a single sector is recovered .
unitarity was used , for example , in @xcite to show that ultra - soft emissions do not contribute to the @xmath4 spectrum .
this argument thus still holds , since the transverse momentum of the ultra - soft emissions is not restricted at leading power .
while we are discussing regularization in the framework of effective field theory , we stress that the same matrix elements also appear in the traditional diagrammatic approach to factorization .
in particular , the well known problems arising in the naive definition of transverse parton distribution functions ( pdfs ) have exactly the same origin : at fixed transverse momentum dimensional regularization no longer regularizes the light - cone singularities which arise from the wilson lines present in the associated matrix elements .
our formalism provides a gauge invariant operator definition of transverse position @xmath32 dependent pdfs @xmath33 where the sum over final state particles is regularized according to ( [ regfinal ] ) and the object on the left - hand side depends implicitly on @xmath2 and @xmath1 .
the field @xmath34 is the usual quark field , decorated with a wilson line running from @xmath35 to infinity along the @xmath36-direction . in the same physical process ,
also a pdf for the anti - quark inside the second nucleon @xmath37 arises , in which the wilson lines run along the @xmath38 direction . however , with our prescription ( [ regfinal ] ) the regulators are the same in both pdfs , since they arise from the same cuts in the original qcd diagrams .
both the quark and the anti - quark pdf have singularities for @xmath31 due to the light - cone denominators arising from the wilson lines .
the singularities cancel in the product of the two functions , but since the plus components of the momenta scale differently in the two pdfs , the product depends on @xmath29 .
the analysis of this anomalous @xmath29-dependence to all orders was given in @xcite , where it was shown that it is a pure power , i.e. has the form @xmath39 .
with ( [ tpdf ] ) , we are now able to give an operator definition of the regularized transverse pdfs from which the anomaly function @xmath40 is obtained . a new definition of transverse pdfs was recently proposed by collins in @xcite .
it includes a carefully chosen combination of light - like and non - light - like soft wilson lines , arranged in such a way that the various singularities of the naive definition are cancelled by the singularities in these wilson lines .
the advantage of our definition ( [ tpdf ] ) is its great simplicity , which , for example , facilitates the perturbative computations to match the transverse onto standard pdfs .
our method is not sufficient for cases where also the virtual diagrams need additional regularization .
this is the case for electroweak sudakov processes and also for regge limits . for the massive case , a regulator which leaves the structure of the theory intact has been proposed in @xcite , which is not analytic and introduces an additional scale into the effective theory .
this complicates the computations and care is needed to avoid double counting .
also , no arguments were presented in @xcite that the corresponding method works beyond one loop .
a promising form of regularization was proposed in @xcite , which regularizes the wilson lines of scet analytically . by construction , this leaves the eikonal structure intact , but one will need to show that qcd is indeed recovered in the limit where the regularization is removed .
also , it appears that the regularized wilson lines have complicated behavior under gauge transformations and it is thus not clear whether gauge invariance can be maintained .
we believe that there are still open issues concerning regularization in the effective theory in these cases . in the two following , rather technical sections , we will now discuss in detail , why the prescription ( [ regfinal ] ) is sufficient to obtain well - defined expressions in the effective theory . in section [ loops ] , we first explain why an additional regularization is not needed for the virtual corrections . in section [ phasespace ] , we then demonstrate that all phase - space integrals are well - defined with our regularization prescription .
a summary and conclusions are presented in section [ conclusion ] .
the virtual corrections in scet are simply matrix elements of light - like wilson lines in qcd . for the soft function in a two - jet process , for example , they are encoded in the amplitudes @xmath41 where @xmath42 are the momenta of the final state particles and @xmath43 and @xmath44 are soft wilson lines extending along the directions of the jets .
the relevant soft lagrangian is the same as the usual qcd lagrangian and the virtual corrections are identical for any two - jet observable computed in scet .
if such matrix elements are well - defined in qcd , they are thus also well - defined in scet .
we stress that the matrix elements ( [ softvirtual ] ) do have soft , collinear and ultraviolet singularities .
the only point relevant to our discussion is that these are regularized dimensionally .
since the virtual corrections are common to all observables , any problem concerning their regularization would affect all observables in the effective theory .
the same statements are true for the virtual corrections in the collinear sectors .
they are given by matrix elements of quark and gluon fields multiplied by wilson lines in the direction associated with the large energy flow , and are again common to all observables . by
now a sizable number of two - loop computations of such quantities exist , for both jet functions @xcite , i.e. collinear matrix elements , and for soft functions @xcite .
in all these computations dimensional regularization turned out to be sufficient .
the observables which were computed at two - loop accuracy include inclusive @xmath45-decays , inclusive drell - yan production and the event - shape variable thrust . the effective theory relevant for these observables is sometimes called scet@xmath46 , and distinguished from scet@xmath47 in which the soft modes have the same virtuality as the collinear ones .
however , on - shell matrix elements such as ( [ softvirtual ] ) are independent of the virtuality , and are the same in all versions of scet . in view of the existing evidence
, most practitioners will not be worried that virtual corrections could have unregularized light - cone singularities . in the following
, we will not attempt to give a rigorous proof that all loop diagrams are indeed well - defined in the effective theory , but we find it instructive to consider a specific example to get some insight why the problem of light - cone singularities does not occur in the massless virtual diagrams .
let us examine the scalar integral associated with the left diagram in figure [ diagrams ] , @xmath48 we are interested in the two - jet kinematics where @xmath49 is a collinear and @xmath50 an anti - collinear momentum , and we will consider both the case where the momentum @xmath51 of the emitted gluon is collinear or the case when it is soft .
we assume that the external momenta correspond to real massless particles , which are on - shell and have positive light - cone components .
the usual @xmath52-prescription in the propagators is understood .
we first examine the situation where the external gluon is collinear .
the collinear momentum region of the loop integral with @xmath53 then gives a typical contribution to the jet function .
writing @xmath54 the integral becomes @xmath55 at leading power .
in contrast to the exact expression the expanded integral contains a light - cone propagator , which may induce additional singularities in the effective theory that are not present in the full theory .
we will now verify that these singularities are regularized in dimensional regularization .
to do so , it is instructive to perform the @xmath13-integration with contour methods . as both @xmath56 and @xmath57 are positive , the integral vanishes for @xmath58 where the poles end up in the same half - plane .
the same argument holds for @xmath59 and hence there is no ultraviolet divergence as @xmath60 .
the only new singularity in the effective theory may thus arise in the limit @xmath61 , where the pole @xmath62 flips into the opposite half - plane . picking up the residue of this pole gives @xmath63^{-1 } \nonumber\\ & \hspace{4cm}\times\ ; \left[\vec{k}_\perp^2+\frac{k_-}{p_-+l_- } \left(k_- ( p_+ + l_+)- 2 \vec{k}_\perp ( \vec{p}_\perp + \vec{l}_\perp ) - ( p + l ) ^2\right)\right]^{-1}.\end{aligned}\ ] ] note the factor of @xmath12 in front of the parenthesis in each propagator .
after combining the propagators with a feynman parameter and performing the standard shift in the transverse momenta , this will turn into an effective mass term that is proportional to @xmath12 .
the integration over the transverse momenta will therefore supply an overall factor @xmath64 multiplied by a remainder that is finite in the limit @xmath61 .
the considered integral thus contains a light - cone singularity when @xmath12 tends to zero , which is however regularized in dimensional regularization .
we emphasize that this is not accidental but a consequence of ( [ pole1 ] ) , which ties the scaling of @xmath65 and @xmath12 together , as stressed also in .
the regularization of the transverse momentum integration therefore carries over to the longitudinal component .
we conclude that the effective - theory integral is well - defined , and one may explicitly verify that it correctly reproduces the infrared singularities of the exact expression .
[ l]@xmath66 [ l]@xmath51 [ l]@xmath49 , title="fig:",scaledwidth=20.0% ] [ l]@xmath66 [ l]@xmath67 [ l]@xmath68 [ l]@xmath69 [ l]@xmath49 , title="fig:",scaledwidth=20.0% ] a similar argument holds for the soft integrals , which have a slightly different structure . here we assume that the external gluon and the loop momentum are soft , @xmath70 .
writing @xmath71 and @xmath54 , we obtain at leading power @xmath72 where we have made the @xmath52-prescription in the light - cone propagators explicit .
as before we perform the @xmath13-integration with contour methods .
the integral again vanishes for @xmath58 , but it is no longer bounded from above since the pole in the light - cone propagator , @xmath73 does not move into the opposite half - plane for any value of @xmath12 . in the interval @xmath74 , we again pick up the contribution from the pole ( [ pole1 ] ) and the discussion proceeds along the same lines as before . for @xmath75 , on the other hand , we take the residue of the pole ( [ pole2 ] ) , which gives @xmath76^{-1}\ ; \left[(\vec{k}_\perp-\vec{l}_\perp)^2\right]^{-1}.\end{aligned}\ ] ] the subsequent transverse integration now yields an expression that scales as @xmath77 for large values of @xmath12 . in our example
the longitudinal integration is thus finite in the limit @xmath60 .
more importantly , it again inherits the dimensional regularization .
the above reasoning also applies to a general one - loop integral . to discuss the general case in the collinear sector
, it is convenient to assign the loop momentum @xmath78 to the first gluon emitted from the anti - quark line that enters the loop ( starting from the end of the anti - quark line ) .
the subsequent collinear emissions from this gluon then induce propagators that have poles in @xmath13 which change the half - plane for values of @xmath79 . with this assignment of the loop momentum , the integral thus vanishes for negative values of @xmath12 .
if the gluon with momentum @xmath78 is the first collinear emission from the anti - quark line , we obtain a light - cone propagator @xmath12 .
other collinear emissions from the anti - quark induce propagators of the form @xmath80 , where @xmath51 represents the sum of some outgoing external momenta . these light - cone propagators are , however , infrared finite since @xmath57 is positive .
as the collinear integrals are always bounded from above , we only have to show that the integral is well - defined in the limit @xmath61 . in this limit
we can read off from ( [ residue1 ] ) how the general structure in the collinear sector will look like .
the collinear propagators will take the form of the expression in the second line , with @xmath81 replaced by the appropriate sum of collinear momenta .
our central observation from above , that the transverse momentum integral will have an effective mass term that is proportional to @xmath12 , will however not change .
the regularization of the transverse momentum integration is therefore again carried over to the longitudinal component . for the soft integrals we choose the same assignment of the loop momentum as in the collinear sector .
here the same arguments apply , except that we also have to show that the integral is well - defined in the limit @xmath60 . in this limit
the generalization of ( [ residue2 ] ) consists of a product of soft propagators that depend linearly on @xmath12 , similar to the first one in ( [ residue2 ] ) . by combining these propagators with feynman parameters and completing the square in the transverse momenta
, we obtain an effective mass term that scales with @xmath12 in the limit @xmath60 .
the integration over the light - cone component is therefore again regularized dimensionally .
the situation is more subtle for those diagrams , in which a soft propagator couples to the collinear quark line .
this yields one propagator that does not depend linearly on @xmath12 , and there is hence a single point in the feynman parameter space where our argument breaks down .
similar to the situation in ( [ residue2 ] ) , it turns out that this point corresponds to a scaleless @xmath82-integral , and therefore does not spoil the argument .
let us finally discuss three situations where the dimensional regularization of the transverse space does not carry over to the light - cone directions .
first of all , in a massive theory the relation ( [ pole1 ] ) is replaced by @xmath83 , which obviously breaks the simple scaling between the transverse and the longitudinal momenta . as a consequence
the transverse integration does not regularize the light - cone components , and the individual expressions in each sector of the effective theory are ill - defined . in the massive case
the effective theory therefore requires additional regulators beyond dimensional regularization .
the massive case is relevant for the study of electroweak sudakov corrections , which were analyzed in an effective field theory framework in @xcite .
second , even in a massless theory the loop diagrams turn out to be ill - defined when they are expanded around the regge limit @xmath84 , which corresponds to forward scattering @xcite .
let us reconsider our intermediate result ( [ residue1 ] ) to understand why the above arguments break down in this situation .
in a scattering process one of the external momenta @xmath49 and @xmath51 will be incoming , and at small angles their large components will be approximately equal . at leading power in the regge limit
we are thus left with ( [ residue1 ] ) for @xmath85 .
we see that in this case the @xmath65-term , which transports the scaling to the longitudinal components , drops out in the second propagator .
similar to the massive case , this results in an effective mass term that does not vanish in the limit @xmath86 .
scet is relevant for problems with large momentum transfer and can not directly be applied to regge problems .
an exploratory study of the regge dynamics in an effective field theory context can be found in @xcite .
the third case is the one which we address in our paper , namely real - emission processes for observables that are sensitive to small transverse momenta . in this case
the transverse momentum is fixed by some external constraint , and the transverse integration therefore can not provide a factor @xmath87 which regularizes the light - cone singularities .
in contrast to the massive case , the unregularized singularities only arise in real emissions , which allows us to apply the regularization on the level of the phase - space integrals . notice that our phase - space prescription ( [ regfinal ] ) precisely reinstalls a factor @xmath88 to make the expressions well - defined .
we will now show that all real - emission diagrams are regularized by our prescription ( [ regfinal ] ) . for concreteness
, we consider the amplitude for the decay of a massive electroweak boson with momentum @xmath89 into @xmath38 massless outgoing particles with momenta @xmath67 , , @xmath90 , so that @xmath91 .
we will integrate over the phase space of the outgoing particles , and want to show that the light - cone singularities of the phase - space integrals in the effective theory are regularized by our prescription ( [ regfinal ] ) .
the amplitude in massless qcd can have singularities only when some of the invariants @xmath92 go to zero .
in such limits , the qcd @xmath38-particle amplitude factorizes into @xmath93-point amplitudes multiplied by splitting functions , but for our purposes neither the precise form of this factorization , nor the strength of the associated singularities are important .
the light - cone singularities present in the effective theory arise when the amplitude is expanded in the different momentum regions .
if an invariant contains only momenta from a single region , the invariant remains unchanged .
however , if it contains momenta from multiple regions , it gets expanded . for example , if @xmath94 and @xmath95 are collinear , while the remaining momenta are anti - collinear , one expands @xmath96 and is thus left with light - cone denominators consisting of sums of momenta . as an explicit example , we consider the right diagram in figure [ diagrams ] , which shows successive emissions of collinear gluons from an anti - quark with anti - collinear momentum . after expanding in @xmath97 ,
the associated propagators produce light - cone denominators which contain sums of the collinear momenta . in the effective theory
they are described by emissions from a collinear wilson line .
for the example shown in the figure , one ends up with the light - cone denominators @xmath98 the crossed versions of the same diagram will give rise to sums of all other combinations of momenta .
our prescription only regularizes individual light - cone components , and we need to show that this is sufficient to also regularize the singularities which arise from sums of momenta . this is the case , since the sums only become singular , when all of the summands go to zero , and our prescription regularizes all of the individual integrations . to make this property manifest , we change variables @xmath99
the light - cone part of the phase - space integration then takes the form @xmath100 the maximal value is @xmath101 for the anti - collinear integrals , while the integration extends to infinity in the soft sector . for the collinear integrals , it is convenient to reverse the order of integrations , which leaves us with integrations over the minus components that are again bounded from above . after the change of variables we have , for example , @xmath102 it is obvious that all singularities from sums of light - cone momenta occur when @xmath103 or some of the @xmath104 s are zero
the light - cone singularities are thus regularized analytically by our prescription . for the soft integrals
, we can also have singularities when some of the integration variables tend to infinity , which are regularized as well .
this shows that in the case where all the partons are in the final state , the effective - theory phase - space integrals are regularized by our prescription ( [ regfinal ] ) .
the situation is slightly more complicated for observables at hadron colliders , since the relevant amplitudes also have two partons in the initial state , one with a collinear momentum @xmath49 , with a large component @xmath105 , and one with momentum @xmath50 in the opposite direction . from an invariant composed out of both initial and final state momenta , such as @xmath106
one might expect the occurrence of additional divergences for @xmath107 , which would not be regularized .
are of no concern since they can be rewritten as sums over final state momenta by momentum conservation . ] fortunately , the light - cone singularities from soft and collinear emissions arrange themselves into wilson lines : with an emission of an anti - collinear gluon , the collinear quark propagator simplifies to @xmath108 where we have used that only the @xmath109 component of the anti - collinear gluon field is @xmath110 . only the direction @xmath111 , but not the size of the collinear momentum is relevant , and only the anti - collinear momentum component @xmath112 picks up a light - cone denominator .
this discussion can be made general by noticing that the anti - collinear sector of scet is independent of any collinear momentum , and only knows about the presence of the other sector via the light - cone reference vector @xmath38 .
it can thus only suffer from light - cone denominators which are sums of anti - collinear momenta .
the soft sector only knows about the collinear particles via @xmath38 and @xmath36 and is completely independent of collinear momenta .
it thus has only light - cone singularities corresponding to sums of soft momenta .
we conclude that also in this case , all singularities are regularized by our prescription .
for observables sensitive to low transverse momentum , it is well known that dimensional regularization is not sufficient to make the expressions in the individual sectors of soft - collinear effective theory well - defined . in this letter , we have shown that in the massless case the unregularized singularities only arise in real - emission diagrams and that it is sufficient to regularize the associated phase - space integrals analytically with the prescription @xmath113 since the amplitudes itself do not need any additional regularization , the structure of the effective theory is not changed , and fundamental properties such as gauge invariance and the eikonal form of the soft and collinear emissions are maintained .
this is essential to establish factorization theorems to all orders .
our approach is well - suited for higher - order computations , since it does not introduce any additional scales into the problem and since the light - cone denominators which we regularize are typically already present in the effective - theory amplitudes .
we stress that it is important that we introduce the regulator in qcd itself .
since the qcd diagrams do not need additional regularization , we are guaranteed to recover the qcd result after adding the contributions from the different sectors in the effective theory and sending the regulator to zero .
our result puts the derivation of the factorization theorems for the transverse momentum spectrum in drell - yan production @xcite and for the @xmath114 event shape jet broadening @xcite on a firmer footing .
it provides operator definitions for the ingredients in the associated factorization theorems , and should also simplify the computations necessary to extend the resummation for these observables to higher logarithmic accuracy . _
acknowledgments : _ we thank thorsten feldmann , matthias neubert and lilin yang for discussions .
this work was supported in part by the swiss national science foundation ( snsf ) and `` innovations- und kooperationsprojekt c-13 '' of suk . c. w. bauer , s. fleming , d. pirjol and i. w. stewart , phys .
d * 63 * , 114020 ( 2001 ) [ arxiv : hep - ph/0011336 ] . c. w. bauer , d. pirjol and i. w. stewart , phys .
d * 65 * , 054022 ( 2002 ) [ arxiv : hep - ph/0109045 ] .
m. beneke , a. p. chapovsky , m. diehl and t. feldmann , nucl .
b * 643 * , 431 ( 2002 ) [ arxiv : hep - ph/0206152 ] . for a review , see v. a. smirnov , `` applied asymptotic expansions in momenta and masses , '' springer tracts mod .
phys . * 177 * , 1 ( 2002 ) .
v. a. smirnov , phys .
b * 404 * , 101 ( 1997 ) [ arxiv : hep - ph/9703357 ] .
m. beneke and t. feldmann , nucl .
b * 685 * , 249 ( 2004 ) [ arxiv : hep - ph/0311335 ] .
g. bell and t. feldmann , nucl .
* 164 * , 189 ( 2007 ) [ arxiv : hep - ph/0509347 ] .
j. y. chiu , f. golf , r. kelley and a. v. manohar , phys .
lett . * 100 * , 021802 ( 2008 ) [ arxiv:0709.2377 [ hep - ph ] ] .
t. becher and m. neubert , eur
. phys .
j. c * 71 * , 1665 ( 2011 ) [ arxiv:1007.4005 [ hep - ph ] ] .
j. -y .
chiu , a. jain , d. neill and i. z. rothstein , arxiv:1104.0881 [ hep - ph ] .
t. becher , g. bell and m. neubert , phys .
b * 704 * , 276 ( 2011 ) [ arxiv:1104.4108 [ hep - ph ] ] .
chiu , f. golf , r. kelley and a. v. manohar , phys .
rev . * d77 * , 053004 ( 2008 ) [ arxiv:0712.0396 [ hep - ph ] ] .
j. collins , arxiv:1107.4123 [ hep - ph ] .
j. -y .
chiu , a. fuhrer , a. h. hoang , r. kelley and a. v. manohar , phys .
d * 79 * , 053007 ( 2009 ) [ arxiv:0901.1332 [ hep - ph ] ] .
t. becher and m. neubert , phys .
lett . b * 637 * , 251 ( 2006 ) [ arxiv : hep - ph/0603140 ] .
a. jain , i. scimemi and i. w. stewart , phys .
d * 77 * , 094008 ( 2008 ) [ arxiv:0801.0743 [ hep - ph ] ] .
t. becher and g. bell , phys .
b * 695 * , 252 ( 2011 ) [ arxiv:1008.1936 [ hep - ph ] ] .
a. v. belitsky , phys .
b * 442 * , 307 ( 1998 ) [ arxiv : hep - ph/9808389 ] .
t. becher and m. neubert , phys .
b * 633 * , 739 ( 2006 ) [ arxiv : hep - ph/0512208 ] .
r. kelley , m. d. schwartz , r. m. schabinger and h. x. zhu , phys .
d * 84 * , 045022 ( 2011 ) [ arxiv:1105.3676 [ hep - ph ] ] .
f. monni , t. gehrmann and g. luisoni , jhep * 1108 * , 010 ( 2011 ) [ arxiv:1105.4560 [ hep - ph ] ] .
a. hornig , c. lee , i. w. stewart , j. r. walsh and s. zuberi , jhep * 1108 * , 054 ( 2011 ) [ arxiv:1105.4628 [ hep - ph ] ] .
y. li , s. mantry and f. petriello , arxiv:1105.5171 [ hep - ph ] .
j. y. chiu , r. kelley and a. v. manohar , phys .
d * 78 * , 073006 ( 2008 ) [ arxiv:0806.1240 [ hep - ph ] ] .
j. f. donoghue and d. wyler , phys . rev .
d * 81 * , 114023 ( 2010 ) [ arxiv:0908.4559 [ hep - ph ] ] . | in high - energy processes which are sensitive to small transverse momenta , individual contributions from collinear and soft momentum regions are not separately well - defined in dimensional regularization . a simple possibility to solve this problem is to introduce additional analytic regulators .
we point out that in massless theories the unregularized singularities only appear in real - emission diagrams and that the additional regulators can be introduced in such a way that gauge invariance and the factorized eikonal structure of soft and collinear emissions is maintained .
this simplifies factorization proofs and implies , at least in the massless case , that the structure of soft - collinear effective theory remains completely unchanged by the presence of the additional regulators .
our formalism also provides a simple operator definition of transverse parton distribution functions .
december 16 , 2011 * analytic regularization in + soft - collinear effective theory * thomas becher and guido bell + |
squeezing and displacement are the basic operations of continuous variables quantum information @xcite , and are easily performed , the former by parametric amplifiers , the latter by lasers and linear optics . squeezing in the two - mode setup
is , for example , the tool to generate entanglement in the braunstein - kimble teleportation scheme @xcite .
the combined use of real squeezing and displacement allows one to encode efficiently classical information in quantum channels using homodyne detection at the receiver @xcite . in a quantum communication scenario where a coherent signal is sent through a non - linear medium and
undergoes an amplification process , the joint estimation of displacement and squeezing provides a twofold information : the amplitude modulation of the state and a property of the communication channel itself .
this may be useful in a communication scheme designed to be robust to photon loss .
in this paper we consider the problem of jointly estimating real amplitude and real squeezing of the radiation field . in a tomographic setup ,
where a large number of equally prepared copies is available , the maximum likelihood method turns out to be very efficient in estimating the parameters that characterize the state of the quantum system @xcite . in this approach ,
first one fixes a set of single - copy measurements ( typically homodyne measurements at some random phase ) and then looks for the estimate that maximizes the probability ( density ) of producing the observed data .
however , the tomographic approach is not suitable to the case when only a small number of copies is available , and one needs to use the limited resources at disposal more efficiently .
it becomes then important to optimize not only the posterior processing of the experimental data , but also the choice of the measurement that is used to extract these data from the system .
the most natural framework to deal with this situation is quantum estimation theory @xcite , where the concept of positive operator valued measure ( povm ) provides a tool to describe at the same time both measurement and data processing .
the maximum likelihood approach in quantum estimation theory @xcite then corresponds to seek the measurement that maximizes the probability ( density ) that the estimated value of the parameters coincides with the true value .
joint estimation of squeezing and displacement is equivalent to infer an unknown transformation of a group in the present case the affine group .
this is an example of a frequent situation in quantum estimation , especially in communication problems , where a set of signal states is generated from a fixed input state by the action of a group .
consider , for example , the case of phase estimation for high - sensitivity interferometry and optimal clocks @xcite , the estimation of rotation for the optimal alignment of reference frames @xcite , and the estimation of displacement of the radiation field for the detection of a coherent signal in gaussian noise @xcite . in these cases ,
the symmetry of the problem provides a physical insight that allows one to simplify the search for efficient estimation strategies , and the concept of _ covariant _ measurement @xcite becomes crucial for optimization .
the use of maximum likelihood approach in the covariant setting has been shown to be particularly successful in obtaining explicitly the optimal measurements and in understanding the fundamental mechanism that leads to the ultimate sensitivity of quantum measurements @xcite .
group theoretical tools , such as equivalent representations @xcite and multiplicity spaces @xcite , far from being abstract technicalities , are the main ingredients to achieve the ultimate quantum limits for sensitivity @xcite . moreover , in a large number of situations , the maximization of the likelihood provides measurements that are optimal also according to a wide class of different figures of merit @xcite . from a group theoretical point of view
, the action of real squeezing and displacement on the wavefunctions provides a unitary representation of the affine group `` @xmath0 '' of dilations and translations on the real line .
the structure of the affine group underlies the theory of wavelets @xcite , and has been recently used in the characterization of @xmath1 coherent states @xcite , and in the study of oscillators in a morse potential @xcite .
the affine group is particularly interesting , since it is the paradigmatic example of a nonunimodular group , namely a group where the left - invariant haar measure is different from the right - invariant one .
this leads to orthogonality relations that differently from the usual schur lemmas
involve a positive unbounded operator , firstly introduced by duflo , moore , and carey ( dmc ) in ref . @xcite .
as we will see in this paper , the nonunimodularity of the group has some amazing consequences in the estimation problem .
for example , the so - called square - root measurements @xcite , that are commonly considered in quantum communication and cryptography , do not coincide with the maximum - likelihood measurements , the latter providing a higher probability density of correct estimation .
another bizarre feature is that for maximum likelihood measurements , the most likely value in the probability distribution can be different from the true one .
while for unimodular groups this feature never happens , for nonunimodular groups it is unavoidable , and a suitable choice of the input states is needed to reduce the discrepancy between the true value and the most likely one .
the paper is organized as follows . in order to set the optimal joint estimation of real squeezing and displacement in the general estimation method ,
first we derive in sec .
[ mlest ] the optimal measurement for the problem of estimating the unknown unitary action of a nonunimodular group . then , as a special example , the general results will be used to optimize the joint estimation of squeezing and displacement in sec .
[ jointest ] .
the efficiency of coherent states and displaced squeezed states is analyzed in detail , and the asymptotic relation between the uncertainties in the joint estimation and the uncertainties in the optimal separate measuremnts of squeezing and displacement is derived .
the conclusions are summarized in sec .
the explicit derivation of group average over the affine group is given in the appendix .
suppose that a fixed input state @xmath2 , corresponding to a density operator in the hilbert space @xmath3 , is transformed by the unitary representation @xmath4 of the group @xmath5 , so that it generates the family of signal states @xmath6 the typical quantum estimation problem is then to find the measurement that gives the best estimate for the unknown transformation @xmath7 according to some optimality criterion .
usually the criterion is given by a cost function @xmath8 , which quantifies the cost of estimating @xmath9 when the true value is @xmath7 , and enjoys the invariance property @xmath10 .
for example , the maximum likelihood criterion corresponds to the delta cost - function @xmath11 ( loosely speaking , there is an infinite gain if the estimate coincides with the true value , and no gain otherwise ) .
once a cost function is fixed , one can choose two possible approaches to optimization , namely the bayes approach and the minimax .
in the bayes approach , one assumes some prior distribution of the unknown parameters , and then minimizes the average over the true values of the expected cost @xmath12 , where @xmath13 is the conditional probability density of estimating @xmath9 when the true value is @xmath7 . in the minimax approach ,
one looks instead for the measurement that minimizes the supremum of the expected cost over all possible true values of the unknown parameters .
an important class of estimation strategies is given by the _
measurements , that are described by povms of the form @xcite @xmath14 where @xmath15 is an operator satisfying the normalization condition @xmath16 @xmath17 denoting the left - invariant haar measure on the group , namely . due to the symmetry of the set of states ( [ orbit ] ) , covariant measurements play a fundamental role in the search of the optimal estimation . for compact groups ,
the following proposition holds : [ covminimax ] for compact groups , the search for the optimal measurement in the minimax approach can be restricted without loss of generality to the class of covariant measurements .
moreover , for compact groups the optimality of covariant measurements holds also in the bayes approach @xcite , if the prior distribution is chosen to be the normalized haar measure on the group , i.e. the measure @xmath18 such that @xmath19 and @xmath20 . for non - compact groups , such as the affine group involved in the joint estimation of squeezing and displacement , the situation is more involved
of course , in the bayes approach it is no longer possible to choose the uniform haar measure as prior distribution , since it is not normalizable .
however , in the minimax the optimality of covariant measurements still holds , even though in this case the proof becomes rather technical @xcite . in this paper we will adopt the minimax approach to deal with noncompact groups , and this will allow us to restrict the optimization to covariant povms .
as we will see in section iii , the action of real squeezing and displacement on one mode of the radiation field yields a representation of the affine group `` @xmath0 '' of dilations and translations on the real line .
this group is clearly non - compact , and , moreover , it is nonunimodular , namely the left - invariant haar measure @xmath17 ( with @xmath21 ) does not coincide with the right - invariant one @xmath22 ( with @xmath23 ) .
therefore , to face the estimation problem with the affine group , we need some results about representation theory and orthogonality relations for nonunimodular groups ( for an introduction to these topics , see for example @xcite ) .
let @xmath24 be a unitary representation of a locally compact group @xmath5 .
in the following , we will make two assumptions on the representation @xmath25 that are tailored on the concrete problem of estimating real squeezing and displacement . _
first assumption : discrete clebsch - gordan series .
_ we require the representation @xmath25 to be a direct sum of irreducible representations ( _ irreps _ ) , namely that its clebsch - gordan series is discrete . in this case , there is a decomposition of the hilbert space @xmath26 as @xmath27 such that @xmath28 where @xmath29 is an irreducible representation acting on the hilbert space @xmath30 and @xmath31 is the identity in the space @xmath32 .
the hilbert spaces @xmath30 and @xmath33 are called _ representation _ and _ multiplicity _ spaces , respectively , and the index @xmath34 labels the inequivalent irreps that appear in the clebsch - gordan series @xmath35 of the representation @xmath25 . _
second assumption : square - summable irreps .
_ we require each irreducible representation @xmath36 in eq .
( [ repdecomp ] ) to be square - summable .
this means that there is at least one non - zero vector @xmath37 such that @xmath38 vectors such that the above integral converges are called _ admissible_. it is possible to show @xcite that , if a representation is square - summable , then the set of admissible vectors is dense in the hilbert space .
let us consider now the group average of an operator @xmath39 , defined as @xmath40 in general , the average may not converge for any operator ( for example , it diverges for @xmath41 ) . in analogy with admissible vectors
, we say that @xmath39 is an _ admissible operator _ if the group average in eq .
( [ defave ] ) converges in the weak operator sense .
in such a case , one can prove that @xmath42 is given by @xmath43~,\ ] ] where @xmath44 is a positive self - adjoint operator acting on the representation space @xmath30 .
the operator @xmath44 has been firstly introduced by duflo , moore , and carey ( dmc ) @xcite , and is the characteristic feature of nonunimodular groups .
in fact , if the group @xmath5 is unimodular i.e .
if the left- and right - invariant measures coincide than the dmc operator is simply a multiple of the identity , and the formula ( [ aveop ] ) for the group average is equivalent to the ordinary schur lemmas .
contrarily , if the group @xmath5 is nonunimodular , the dmc operator is a positive unbounded operator , and its presence modifies the orthogonality relations dramatically , with remarkable consequences in the estimation of an unknown group transformation .
the admissibility of vectors and operators has a simple characterization in terms of the dmc operator .
as regards vectors , the set of admissible vectors for the irrep @xmath29 can be characterized as the domain of @xmath45 . for unimodular groups , since @xmath44 is proportional to the identity , the set of admissible vectors is the whole representation space @xmath30 , while for nonunimodular groups the admissible vectors form a dense subset of @xmath30 . as regards operators , an operator @xmath39 is admissible if and only if the partial traces @xmath46 $ ] are not diverging .
for example , if @xmath37 is an admissible vector and @xmath47 is any operator acting on @xmath33 , then the operator @xmath48 is admissible , and its group average is @xmath49 . in order to find the best estimate for the signal states ( [ orbit ] ) according to the maximum likelihood criterion in the minimax approach , we consider now a covariant measurement , described by a povm @xmath50 as in eq .
( [ covpovm ] ) . the normalization condition ( [ nc ] ) can be rewritten as @xmath51 using eq .
( [ aveop ] ) for evaluating the group average @xmath52 , this condition becomes @xmath53= \openone_{\spc m_{\mu}}~.\ ] ] according to the maximum likelihood approach , we need to find the covariant measurement that maximizes the probability density that the estimated transformation coincides with the true one , i.e. @xmath54=\tr[\xi \rho]$ ] . for a pure input state @xmath55 , the maximization of the likelihood over all possible operators @xmath15 satisfying the constraints ( [ trxi ] ) follows in a simple way by a repeated use of schwartz inequality .
in fact , the input state @xmath56 can be written in the decomposition ( [ spacedecomp ] ) as @xmath57 is a bipartite state . from schwartz inequality
, we have @xmath58 at this point , we assume that each bipartite state @xmath59 is in the domain of the operator @xmath60 .
this assumption is not restrictive , since the domain of a self - adjoint operator is dense in the hilbert space . in this way , it is possible to write and to exploit the schmidt decomposition of the ( non - normalized ) vector @xmath61 .
in other words , we can write @xmath62 where @xmath63 is the schmidt rank , @xmath64 are schmidt coefficients such that @xmath65 , and latexmath:[$|\tilde \psi^{\mu}_m\ > , @xmath30 and @xmath33 , respectively
. the form ( [ bipartitemu ] ) is very convenient for optimization , in fact we can use again schwartz inequality and obtain @xmath67 finally , we have @xmath68~|\tilde\phi_m^{\mu}\>=1~,\end{aligned}\ ] ] the last equality following from the normalization constraint ( [ trxi ] ) .
therefore , the previous chain of inequalities proves the upper bound @xmath69 that holds for any covariant povm . on the other hand , it is immediate to check that the bound is achieved by the covariant povm given by @xmath70 with @xmath71 denoting the argument of a complex number , i.e. @xmath72 . the normalization of such a povm follows from eq .
( [ trxi ] ) , and one has @xmath73 namely , the povm is complete in the subspace spanned by the orbit of @xmath74 , and can be trivially completed to the whole hilbert space without affecting the probability distribution .
notice that , if the group @xmath5 is unimodular , namely @xmath75 for some positive constant @xmath76 , then we correctly retrieve the results of ref .
@xcite about maximum - likelihood measurements . _
the case of @xmath25 being a direct sum of inequivalent irreps . _ the expression for the optimal covariant povm can be further simplified in the case when all the multiplicity spaces @xmath33 are one - dimensional , i.e. when the representation @xmath25 is a direct sum of inequivalent irreps . in this case
, we can decompose the input state as @xmath77 ( as in eq .
( [ statedecomp ] ) , but without the need of introducing bipartite states ) , and now the decomposition ( [ bipartitemu ] ) becomes trivial , namely @xmath78 where @xmath79 and @xmath80 therefore , eq . ( [ opteta ] ) for the optimal povm becomes @xmath81 and the corresponding optimal likelihood is given by @xmath82 _ remark 1 : square - root measurements .
_ a possible strategy to estimate an unknown quantum state , randomly drawn from a given family , is given by the so - called square - root measurements ( srm ) , firstly introduced by hausladen and wootters @xcite .
in the case of pure states with a group symmetry as in eq .
( [ orbit ] ) , there is an important connection between srm and maximum - likelihood measurements .
for example , if the group @xmath5 is a discrete group of phase shifts , it has been proved in refs .
@xcite , that srm minimize the probability of error in estimating the unknown state , and , equivalently , they maximize the likelihood , i.e. the probability of correct estimation . more generally , the optimality of the srm in the maximum likelihood approach has been proved in ref .
@xcite for a large class of groups , including all finite groups , all compact groups , and unimodular noncompact groups , such as the weyl - heisenberg group of displacements .
however , as we will see in the following , the case of nonunimodular groups represents an exception to the fact that srm are optimal for the maximum likelihood criterion in the presence of a physical symmetry .
in fact , the srm for the estimation of a group transformation acting on a fixed state @xmath2 is given by the povm @xmath83 , where @xmath84 ( the povm @xmath85 is obviously normalized with respect to the left - invariant measure @xmath17 ) .
the comparison with the maximum - likelihood measurements of the previous section is particularly simple in the case of group representations @xmath25 that are direct sum of inequivalent irreps .
in fact , for a pure state @xmath86 with @xmath87 ) , namely @xmath88 .
notice here that the square - root measurement can be defined only if @xmath89 is in the domain of @xmath90 .
therefore , the square - root measurement is given by the covariant povm @xmath91 this covariant povm is different from the optimal one given in ( [ optetaineq ] ) , and does not achieve the optimal value ( [ optlikineq ] ) for the likelihood .
when @xmath92 is in the domain of both @xmath90 and @xmath93 , we can compare the values of the likelihood as follows .
one has @xmath94 where we used the inequality @xmath95 , from schwartz inequality applied to the vectors @xmath96 $ ] for the optimal povm achieves its maximum at the value @xmath97 .
again , this is true for unimodular groups , but fails to hold for nonunimodular groups .
let the group @xmath5 be unimodular . if the covariant povm @xmath98 maximizes the likelihood for a given input state @xmath2 , then the probability distribution @xmath13 of the estimate @xmath9 on the state @xmath99 achieves its maximum for @xmath97 .
suppose that the most likely value does not coincide with the true one .
then we can rigidly shift the whole probability distribution with a post - processing operation that brings the most likely value to the true one .
in fact , if the maximum of @xmath13 occurs at @xmath100 , we can always replace @xmath101 with a new covariant povm @xmath102 , where @xmath103 the normalization of the new povm follows from the fact that for unimodular groups the dmc operators are trivially proportional to the identity , and therefore the operator @xmath104 satisfy the normalization constraints ( [ trxi ] ) as well .
moreover , the probability distribution @xmath105 associated with @xmath106 enjoys the property @xmath107 , whence it achieves the maximum in @xmath97 . in this way
, the likelihood of @xmath106 would be higher than the likelihood of the povm @xmath101 .
but this can not happen since @xmath101 is the optimal maximum - likelihood povm
. therefore @xmath13 must be maximum in @xmath97 . for nonunimodular groups the previous argument does not apply , since the povm given by ( [ xiprime ] ) is no longer normalized .
in fact , the operator @xmath104 does not satisfy the normalization constraints ( [ trxi ] ) , since the dmc operators do not commute with the unitaries @xmath108 .
in other words , we are not allowed to bring the most likely value to coincide with the true one by rigidly shifting the whole probability distribution . as we will see in the explicit example of the estimation of real squeezing and displacement
, this situation can indeed happen . in order to reduce the discrepancy between the true value and the most likely one ,
a suitable choice of the input states is needed .
for example , in the simple case of @xmath25 being a direct sum of inequivalent irreps , if the projection of the input state onto the irreducible subspaces are eigenvectors of the dmc operators , then the most likely value coincides with the true one .
in fact , for any input state @xmath109 , using schwarz inequality , we have @xmath110 and if each @xmath111 is eigenvector of @xmath44 , then the last expression is equal to @xmath112 , then the true value is the most likely one .
in the following we will apply the general framework of section [ mlest ] to the case of joint estimation of real squeezing and displacement of a single - mode radiation field with bosonic operators @xmath113 and @xmath114 with @xmath115=1 $ ] . given the wavefunction of a pure state @xmath116 in the @xmath117-representation @xmath118 , where @xmath119 denotes the dirac - normalized eigenstate of the quadrature operator @xmath120 , the affine transformation on the real line given by @xmath121
is represented by the unitary transformation @xmath122 this transformation corresponds to the action of the unitary operator @xmath123 on the ket @xmath124 , where @xmath125\;,\nonumber \\ & & s(r)=\exp\left[\frac r2 ( a^{\dag 2}-a ^2)\right ] \;,\end{aligned}\ ] ] represent the displacement and the squeezing operator with real argument , respectively . in other words ,
the operators @xmath126 provide a unitary representation of the affine group in the hilbert space of wavefunctions .
the affine group is nonunimodular , and in the above parametrization the left- and right - invariant measures are given by @xmath127 , and @xmath128 , respectively . in order to exploit the general results of sec .
[ mlest ] , we need to know the clebsch - gordan decomposition of the representation @xmath129 , the irreducible subspaces , and the dmc operators . all these informations are given in the following , while the proof is presented in the appendix .
the clebsch - gordan series of the representation @xmath129 consists on two irreps , that we indicate with the symbols @xmath130 and @xmath131 .
accordingly , the hilbert space splits into two irreducible subspaces , i.e. @xmath132 comparing this decomposition with the general case ( [ spacedecomp ] ) , we see that the subspaces @xmath133 and @xmath134 are the representation spaces , while the multiplicity spaces @xmath135 and @xmath136 are trivially one - dimensional .
the representation spaces @xmath133 and @xmath137 can be easily characterized in terms of the quadrature @xmath138 .
in fact , writing the wavefunctions in the @xmath139-representation as @xmath140 , where @xmath141 are the dirac - normalized eigenvectors of @xmath139 , we have @xmath142 and @xmath143 .
therefore , the projection operators onto @xmath133 and @xmath137 can be written respectively as @xmath144 and @xmath145 , where @xmath146 is the customary step - function [ @xmath147 for @xmath148 , @xmath149 for @xmath150 . moreover , the dmc operators are @xmath151 with these tools we are now able to provide the optimal covariant measurement for the joint estimation of real squeezing and displacement on a given state of the radiation field .
let us denote by @xmath56 the input state that undergoes to unknown squeezing and displacement transformations .
decomposing the input state on the subspaces @xmath133 and @xmath137 as @xmath152 where @xmath153 the optimal likelihood is then @xmath154 according to the general expression of eq .
( [ optlikineq ] ) .
as already mentioned , the expression of the likelihood provides some insight about the states that are most sensitive in detecting an unknown combination of real squeezing and displacement .
essentially , one can improve the likelihood by increasing the expectation value of @xmath155 , the modulus of the quadrature @xmath139 . in addiction , the use of wavefunctions that in the @xmath139-representation are non - zero both in the positive half - line and in the negative half - line allows to exploit the interference of the components @xmath156 and @xmath157 to enhance the value of the likelihood . using eq .
( [ etajoint ] ) for the optimal povm , we can obtain the probability distribution of the estimated squeezing and displacement parameter for a given input state . in particular , for a coherent input state @xmath158 the sensitivity of the measurement can be significantly improved by increasing the imaginary part of @xmath159 , this corresponding to taking coherent states with a high expectation value of @xmath155 .
the probability distribution for the joint estimation of squeezing and displacement on the vacuum state and on a coherent state with @xmath160 has been reported in figs . 1 and 2 , respectively . comparing the two figures , a remarkable improvement in the precision of the measurement can be observed as an enhancement of the likelihood , along with a narrowing of the probability distribution .
moreover , we can observe that for the vacuum state the maximum of the probability density is not achieved by the true value ( which is given by to @xmath161 ) . the discrepancy between the most likely value and the true one , due to the fact that the affine group is nonunimodular , essentially disappears by increasing the expectation value of @xmath155 . as we can see from fig .
2 , for @xmath162 the probability distribution is approximately a gaussian centered around the true value @xmath161 . .
] now it is interesting to focus on the asymptotic behavior of the probability distribution for a coherent state @xmath163 with @xmath164 going to infinity . in the asymptotic regime , the probability distribution @xmath165 given by optimal vector @xmath166 in eq .
( [ etajoint ] ) , can be further simplified .
in fact , the wavefunction of the coherent state can be written as @xmath167 and for @xmath168 it is essentially confined in the positive part of the @xmath169-axis .
hence , in the expression ( [ etajoint ] ) we can asymptotically neglect the component in the subspace @xmath137 and drop the modulus from @xmath155 . in this way
, the probability distribution ( [ probcoh ] ) can be approximated as @xmath170 neglecting the higher order terms , we thus obtain the gaussian distribution @xmath171 notice that , asymptotically the most likely values of the unknown parameters @xmath172 and @xmath173 are the true ones @xmath161 , and , in addiction , also the mean values of @xmath172 and @xmath173 coincide with the true one , namely the estimation is unbiased . from the asymptotic expression ( [ asycoh ] ) can see that the uncertainty in the estimation of the squeezing parameter @xmath173 goes to zero with the number of photons @xmath174 , namely the r.m.s error is @xmath175 , while the uncertainty in the estimation of the displacement @xmath172 remains fixed , with the value @xmath176 .
it is interesting to compare the precision achieved by the joint estimation with the precision that could be achieved if the parameters @xmath172 and @xmath173 were measured separately .
first , it is known that the optimal estimation of real displacement is given by the observable @xmath177 , and the corresponding uncertainty in a coherent state is given by @xmath178 .
the optimal estimation of real squeezing has been recently derived in @xcite , and we report here the asymptotic distribution for an excited coherent state : @xmath179 accordingly , the r.m.s .
error for the optimal estimation of squeezing in a coherent state is @xmath180 .
it is remarkable to note the relation @xmath181 between the uncertainties in the joint measurement and the ones in the separate measurements of squeezing and displacement . in particular , the product of the uncertainties in the joint estimation is twice the product of uncertainties in the optimal separate measurements : @xmath182 surprisingly , this is exactly the same relation as the one occurring in the optimal joint measurement of two conjugated quadratures @xmath117 and @xmath139 that can be achieved by heterodyne detection @xcite . in the previous section we analyzed the optimal estimation of squeezing and displacement for an excited coherent state . in that case , while the error in the estimation of @xmath173 goes to zero with the number of photons , the error in estimating @xmath172 remains fixed .
however , it is possible to choose the input state in such a way that both variances vanish in the asymptotic limit . to this purpose ,
we consider here displaced squeezed states @xmath183 with @xmath184 .
such states have the wavefunction @xmath185 namely a gaussian centered around the mean value @xmath113 , with standard deviation @xmath186 . clearly , if the conditions @xmath168 and @xmath187 are simultaneously satisfied , such a gaussian lies almost completely in the positive half - line .
therefore , in the asymptotic limit @xmath188 , @xmath189 , the optimal probability distribution can be approximated as @xmath190 as in the case of coherent states . by calculating the expectation values and keeping the leading order terms ,
we then obtain the asymptotic distribution @xmath191 again , in the asymptotic limit the most likely values in the probability distribution coincide with the true ones , and , moreover , the estimation is unbiased .
the r.m.s .
error in the estimation of squeezing and displacement are now given by @xmath192 and @xmath193 , respectively . in order to have both errors vanishing
, one needs simultaneously @xmath194 and @xmath195 .
for example , we can have an isotropic distribution @xmath196 with the choice @xmath197 . in the isotropic case ,
we notice that only a small fraction of order @xmath198 of the total number of photons @xmath199 comes from squeezing . since one has @xmath200 , the convergence to the asymptotic regime is quite slow : the uncertainty goes to zero with order @xmath201 . as an example in the asymptotic regime , in fig .
3 we plot the exact probability distribution for @xmath202 .
( with signal photons @xmath203 , and squeezing photons @xmath198 ) . ]
also for displaced squeezed states one can compare the accuracy of the joint estimation with that of independent measurements of squeezing and displacement .
for squeezed states of the form ( [ dispsqueez ] ) the uncertainty in the measurement of the observable @xmath117 is @xmath204 . on the other hand , the optimal estimation of squeezing
is obtained asymptotically by the observable @xmath205 , and its uncertainty is @xmath206 ( according to the results of @xcite ) .
again , a factor @xmath207 relates the standard deviations of the marginals in the joint estimation with those of the optimal separate measurements .
again , we find the relation ( [ uncprod ] ) between the product of uncertainties .
we then see that the estimation of real squeezing and displacement on input states of the form ( [ dispsqueez ] ) corresponds in the asymptotic regime to the joint measurement of the observables @xmath117 and @xmath208 , whose commutator is @xmath209=i/(2 y)$ ] .
we thus have the heisenberg - robertson inequality @xmath210 where @xmath211 denotes the expectation value . from this point of view ,
the displaced squeezed states ( [ dispsqueez ] ) are characterized asymptotically as minimum uncertainty states , since they saturate the inequality , and the product of uncertainties in the maximum likelihood estimation is exactly twice the heisenberg limit . to the best of our knowledge ,
the present case is the first example of joint measurement of two harmonic - oscillator noncommuting observables whose commutator is not a @xmath212-number .
the results presented in the previous sections can be easily extended to include the estimation of the reflection on the real line that is realized by the parity operator @xmath213 . in this case
we are interested in estimating the three parameters in the transformation @xmath214 where @xmath215 can assume the values @xmath216 or @xmath217 .
the representation @xmath218 is now irreducible in @xmath3 , and the associated dmc operator is @xmath219 . for a given input state @xmath56 , the optimal povm can be written as @xmath220 , where @xmath221 according to eq .
( [ optetaineq ] ) . clearly , also in this case it is possible to enhance the sensitivity of detection by increasing the average value of the modulus @xmath155 , using coherent states or displaced squeezed states .
a fundamental mechanism leading to the optimal estimation of group transformations is the use of equivalent representations of the group , via the technique of entanglement between representation spaces and multiplicity @xcite .
such a strategy strongly improves the quality of estimation for unimodular groups @xcite , and in the case of compact groups is a way to obtain the best estimation of a unitary transformation@xcite .
equivalent representations of the affine group can be obtained by entangling the radiation with a second reference mode , which is not affected by the unknown squeezing and displacement .
this corresponds to considering the two - mode hilbert space @xmath222 and the representation @xmath223 of the affine group , namely the reference mode plays the role of an infinite dimensional multiplicity space . in this case
, the optimal estimation can be obtained with the povm specified by the general formula ( [ opteta ] ) .
a remarkable feature of the two - mode setup is that it is possible to have an _ orthogonal _
povm for the estimation , namely there exists an ordinary observable on the extended hilbert space , associated with the joint measurement of squeezing and displacement .
in fact , by defining the bipartite vectors @xmath224 we can construct the orthogonal povm @xmath225 the normalization follows straightforwardly from eq .
( [ norm ] ) .
moreover , we have the orthogonality relation @xmath226 where we used eq .
( [ dsr ] ) of the appendix and the identity @xmath227 .
the vectors in eq .
( [ pointer ] ) are not normalizable . however , similarly to the case of the heterodyne operator , where physical states can arbitrarily well approximate the unnormalizable eigenstates @xcite , here one can consider , e.g. , states of the form @xmath228 is a normalization constant , and @xmath229 is the photon number operator for the auxiliary mode @xmath230 .
the states @xmath231 approaches the optimal vectors ( [ pointer ] ) as @xmath232 , with a correspondent increasing to infinity of the average energy of the radiation field .
in this paper we presented the joint estimation of real squeezing and displacement of the radiation field from the general point of view of group parameter estimation .
the combination of squeezing and displacement provides a representation of the affine group `` @xmath233 '' , which is the paradigmatic example of a nonunimodular group . to deal with the concrete example of squeezing and displacement , we derived in the maximum likelihood approach the optimal estimation of a group transformation in the case of nonunimodular groups , providing explicitly the optimal povm for a given input state . in this analysis
, some remarkable features of estimation showed up .
firstly , while for unimodular groups the maximum likelihood measurements coincide with the usual square - root measurements , for nonunimodular groups the srm are no longer optimal , namely they do not maximize the probability density of detecting the correct value .
moreover , for nonunimodular groups one can optimize the estimation strategy in the maximum likelihood approach , but even for the optimal povm the true value is not the one which is most likely to be detected . to reduce the discrepancy between the true value and the most likely one a suitable choice of the input states is required .
both these features are in general unavoidable , and their origin tracks back to the presence of a positive unbounded operator the duflo - moore - carey operator in the orthogonality relations for nonunimodular groups . in the problem of joint estimating real squeezing and displacement , all the above effects occur .
in particular , for coherent input states and displaced squeezed states we observed how an increase in the expectation value of @xmath155 gives rise to an improvement in the quality of estimation , along with a reduction of the discrepancy between the true value and the most likely one .
in the mentioned cases the probability distributions for joint estimation become asymptotically gaussian , and the r.m.s .
errors @xmath234 and @xmath235 can be easily calculated .
remarkably , the product of uncertainties in the joint estimation is exactly twice the product of uncertainties for the optimal separate measurements of squeezing and displacement , in the same way as in the joint measurement of two conjugated quadratures .
finally , the use of entanglement with an additional mode of the radiation field allows one to perform a von neumann measurement for the joint estimation of squeezing and displacement , in terms of an ordinary observable with continuous spectrum .
the ideal input states for detecting an unknown affine transformation can then be approximated by normalizable states .
aknowledges m. hayashi for pointing out ref.@xcite , and erato project for hospitality .
m.f.s . acknowledges support from infm through the project no .
this work has been supported by ministero italiano delluniversit e della ricerca ( miur ) through firb ( bando 2001 ) .
here we derive the group theoretical structure of the representation @xmath236 of the affine group , by explicitly calculating the expression for the group average of an operator @xmath39 over the left - invariant measure @xmath237 .
the clebsch - gordan series of @xmath129 contains two irreducible irreps , @xmath130 and @xmath131 .
accordingly , the hilbert space can be decomposed as @xmath238 , and the projections onto the irreducible subspaces are @xmath239 and @xmath145 , respectively .
the dmc operators are given by @xmath240 and @xmath241 .
using twice the resolution of the identity in terms of the eigenstates @xmath242 of the quadrature operator @xmath139 , and the relation @xmath243 we can calculate the group average as follows @xmath244 - \pi \int _
{ -\infty}^0 \d\tilde r \,|\tilde r \rangle \langle \tilde r| \,\tr[a \theta ( -y)/y]\nonumber \\ & & = \theta ( y ) ~\tr [ a~ \pi \theta ( y)/|y| ] + \theta ( -y ) ~\tr [ a~ \pi \theta ( -y)/|y| ] \;.\end{aligned}\ ] ] the thesis follows by comparing the last equation with the general formula ( [ aveop ] ) for the group average .
99 s. l. braunstein and a. k. pati , _ quantum information with continuous variables _ ( kluwer academic , dordrecht , 2003 ) .
s. l. braunstein and h. j. kimble , phys .
. lett . * 80 * , 869 ( 1998 ) . c. m. caves and p. d. drummond , rev .
phys . * 66 * , 481 ( 1994 ) .
k. banaszek , g. m. dariano , m. g. a. paris , and m. f. sacchi , phys .
a * 61 * , 010304(r ) ( 2000 ) .
g. m. dariano , m. g. a. paris , and m. f. sacchi , phys .
a * 62 * , 023815 ( 2000 ) ; phys .
rev . a * 64 * , 019903(e ) ( 2001 ) . c. w. helstrom , _ quantum detection and estimation theory _
( academic press , new york , 1976 ) .
a. s. holevo , _ probabilistic and statistical aspects of quantum theory _
( north holland , amsterdam , 1982 ) .
a. s. holevo , j. multivariate anal .
* 3 * , 337 ( 1973 ) .
a. s. holevo , rep .
* 16 * , 385 ( 1979 ) .
g. m. dariano , c. macchiavello , and m. f. sacchi , phys .
a * 248 * , 103 ( 1998 ) .
v. buzek , r. derka , and s. massar , phys .
lett . * 82 * , 2207 , ( 1999 ) .
g. chiribella , g. m. dariano , p. perinotti , and m. f. sacchi , phys .
lett . * 93 * , 180503 ( 2004 ) .
h. p. yuen and m. lax , ieee trans .
it * 19 * , 740 ( 1973 ) .
g. chiribella , g. m. dariano , p. perinotti , and m. f. sacchi , phys .
a * 70 * , 062105 ( 2004 ) .
g. chiribella , g. m. dariano , p. perinotti , and m. f. sacchi , int .
( in press ) , preprint quant - ph/0507007 .
g. chiribella , g. m. dariano , and m. f. sacchi , phys .
a * 72 * , 042338 ( 2005 ) .
i. daubechies , _ ten lectures on wavelets _
( siam , philadelphia , 1992 ) .
m. hayashi and f. sakaguchi , j. phys . a * 33 * , 7793 ( 2000 ) .
j. bertrand and m. irac - astaud , j. phys .
a * 35 * , 7347 ( 2002 ) .
b. molnr , m g benedict , and j. bertrand , j. phys .
a * 34 * , 3139 ( 2001 ) .
m. duflo and c. c. moore , j. funct .
21 * , 209 ( 1976 ) ; a. l. carey , bull .
soc . * 15 * , 1 ( 1976 ) .
p. hausladen and w. k. wootters , j. mod . opt . * 41 * , 2385 ( 1994 ) .
m. ozawa , in _ research reports on information sciences , series a : mathematical sciences , n. 74 _ , department of information sciences tokyo institute of technology ( 1980 ) .
a. grossmann , j. morlet , and t. paul , j. math . phys . * 26 * , 10 ( 1985 )
. m. ban , k. kurukowa , r. momose , and o. hirota , int . j. theor
36 * , 1269 ( 1997 ) .
m. sasaki , a. carlini , and a. chefles , j. phys .
a * 34 * , 7017 ( 2001 ) .
y. c. eldar and g. d. forney , ieee trans .
it * 47 * , 858 ( 2001 ) .
g. m. dariano and m. f. sacchi , phys .
a. * 52 * , r4309 ( 1995 ) .
g. chiribella , g. m. dariano , and m. f. sacchi , preprint quant - ph/0601103 .
h. p. yuen , phys .
lett . a * 91 * , 101 ( 1982 ) . | we study the problem of joint estimation of real squeezing and amplitude of the radiation field , deriving the measurement that maximizes the probability density of detecting the true value of the unknown parameters .
more generally , we provide a solution for the problem of estimating the unknown unitary action of a nonunimodular group in the maximum likelihood approach .
remarkably , in this case the optimal measurements do not coincide with the so called square - root measurements . in the case of squeezing and displacement
we analyze in detail the sensitivity of estimation for coherent states and displaced squeezed states , deriving the asymptotic relation between the uncertainties in the joint estimation and the corresponding uncertainties in the optimal separate measurements of squeezing and displacement .
a two - mode setup is also analyzed , showing how entanglement between optical modes can be used to approximate perfect estimation . |
the search for and characterization of the binary supermassive black hole ( bbh ) population are of great importance to understanding both galaxy formation and gravitational wave physics .
since most massive galaxies are believed to harbor central smbhs , the formation of bbhs is an inevitable consequence of frequent galaxy mergers .
the dynamical evolution of the bbh system within the merged galaxy , and its interaction with the host both dynamically and via possible feedback during baryonic accretion onto one or both bhs , encode crucial information about the assembly of the bulge and the central smbh .
if the bbh eventually coalesces , it will produce a gravitational siren that could be detected with future low - frequency gravitational wave experiments ( e.g. , * ? ? ?
given their importance for understanding the co - evolution of galaxies and smbhs , and for testing fundamental physics , bbhs have been of both theoretical and observational interest since the 1970s ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
theoretical studies on bbhs in the past two decades have laid down the basic framework for the dynamical evolution of bbhs within a merged galaxy ( for a review , see , e.g. , * ? ? ?
the two bhs sink to the center of the merged host via dynamical friction , and form a bbh .
the bbh orbit continues to decay by scattering stars in the galactic nucleus until it depletes the loss cone . in the absence of other mechanisms to further extract energy from the binary , or to refill the loss cone , the bbh will stall at @xmath9 pc scale , establishing the `` final parsec barrier '' @xcite . a number of mechanisms have been suggested to circumvent this barrier , though in most cases their effectiveness is quite uncertain : non - spherical stellar potentials ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , interactions with gas ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , or hierarchical triple bh systems ( e.g. , * ? ? ?
* ) through the kozai - lidov mechanism @xcite .
once the bbh reaches orbital separations below @xmath10 pc ( for a @xmath11 bbh ) , gravitational radiation will dominate the successive decay of the binary orbit , and merge the two bhs within a hubble time ( e.g. , * ? ? ?
however there is considerable theoretical uncertainty in the bbh evolution from a few pc to the gravitational regime , and the time the bbh spends during this phase is not well constrained .
the strongest indirect evidence for bbhs perhaps comes from the observed central light deficit in massive ellipticals ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , where the scouring of the central stellar cusp is most easily explained by star interactions with the bbh ( e.g. , * ? ? ?
* ; * ? ? ?
other indirect evidence includes unusual radio morphologies such as helical / winged / x - shaped jet or double - lobed morphologies , which are commonly explained by bbh models ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) although the bbh interpretation is not unique for these observations .
directly observing bbhs during different merger stages is challenging , given the stringent resolution requirement and the difficulty of identifying active galactic nuclei ( agns ) . while close quasar pairs ( binary quasars ) on tens of kpc scales have been known for a while ( e.g. ,
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , clear examples of kpc - scale binary agns were substantially rarer until recently ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ?
* ; * ? ? ?
considerable progress has been made in the last decade on the identification of binary agns on kpc scales , mostly owing to the advent of modern , large - scale sky surveys .
below a few kpc , a growing number of confirmed or candidate binary agns have been selected from multiwavelength surveys ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
? * ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , increasing the inventory of confirmed kpc - scale binary agns already by several folds .
below @xmath12 pc , there is only one clear example , 0402 + 379 , where a pair of bhs with a projected separation of @xmath13 pc were detected in the radio ( e.g. , * ? ? ? * ) with the very long baseline array ( vlba ) ; in addition , there is one claimed binary agn with a projected separation of @xmath14 pc in the nearby seyfert galaxy ngc 3393 @xcite .
so far there is no confirmed case of a bbh with sub - pc separation .
this scale is unresolvable with current instruments except with long baseline radio interferometry .
thus the observational searches for sub - pc bbhs almost exclusively rely on spatially unresolved signatures .
the first potential indirect signature is a periodic photometric light curve , where the periodicity may indicate the presence of a bbh .
the best known candidate is oj 287 , a @xmath15 blazar with its light curve showing a quasi - period of @xmath16 yr in restframe , which is interpreted as a sub - pc bbh with a mass ratio @xmath17 in the latest models ( e.g. , * ? ? ?
other examples have been claimed in the literature but these are not as robust as oj 287 ( see * ? ? ? * and references therein )
. a second signature commonly utilized to search for sub - pc bbhs is the velocity offset between the broad emission line centroid and the systemic velocity in broad - line quasars ( e.g. , * ? ? ?
* ; * ? ? ?
? * ; * ? ? ?
* ; * ? ? ?
since the broad line region ( blr ) is gravitationally confined to the bh on sub - pc scales ( e.g. , * ? ? ?
* and see below ) , any velocity offset may reflect the orbital motion of the bh in a hypothetical binary .
two classes of broad line velocity offset have attracted attention in the literature as indicators of candidate bbhs : ( 1 ) a double - peaked broad line profile , which is interpreted to mean that both bhs are active and co - rotating along with their own distinct blrs ; ( 2 ) a single - peaked broad line profile , with the broad line centroid offset from the systemic velocity , in which case only one bh is active , or both bhs are active but their blr emission can not be separated in velocity space .
the second feature is also often used to argue for recoiling bhs after the coalescence of the bbh ( e.g. , * ? ? ?
it is straightforward to select such candidate sub - pc bbhs based on these broad line diagnostics , especially in the era of massive spectroscopic surveys .
examples are presented in either large samples ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , or individual discoveries ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . however , it is impossible to confirm these cases with single - epoch spectroscopy .
in fact , most sub - pc bbh candidates proposed this way early on have been rejected as long - term monitoring did not observe the expected radial acceleration of the broad line from binary motion ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) .
the main difficulty here involves the poorly understood blr geometry and kinematics even for single bhs , which may mimic a bbh .
for instance , the rival model for the double - peaked or offset broad lines is a disk emitter model ( e.g. , * ? ? ? * ; * ? ? ?
* ; * ? ? ?
* ) , which does not require a bbh to explain the peculiar broad line profile .
thus extraordinary caution is required when interpreting these unusual broad line profiles as bbhs ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
in addition , some theoretical arguments suggest that the double - peaked broad line profile is a very ineffective indicator for bbhs , given the requirement that both blrs are largely dynamically distinct while their relative motion must be large enough to separate the two peaks in the spectrum
such conditions are difficult to fulfill in reality ( e.g. , * ? ? ?
the most effective , and the most definitive , way to improve the power of the broad line diagnostics to identify bbhs is multi - epoch spectroscopy .
long - term spectroscopic monitoring of the broad lines can confirm or rule out the bbh hypothesis based on changes in the broad line radial velocity ( acceleration ) , given that the time baseline is long enough to see the expected binary motion and the spectral quality is adequate .
earlier spectral monitoring of samples of double - peaked disk emitters did not focus on testing the bbh scenario ( e.g. , * ? ? ?
* ; * ? ? ?
spectroscopic monitoring programs that are specifically designed to find bbhs in quasars are just starting ( e.g. , * ? ? ? * ; * ? ? ? * ) .
we are conducting systematic searches for sub - pc bbhs in broad - line quasars using multi - epoch spectroscopy of large statistical quasar samples .
in contrast to previous work ( e.g. , * ? ? ? * ; * ? ? ? * ) which only focused on objects with offset or double - peaked broad lines , we are also targeting normal quasars with no broad line velocity offset . ] .
there are two reasons for targeting normal quasars .
first , quasars with double - peaked or significantly offset broad lines are only a minority of the entire quasar population ( @xmath18 , e.g. , * ? ? ?
while they are more likely to be bbhs by selection , they do not represent the general quasar population where a bbh can be easily hidden in the spectrum ( since we only observe the line - of - sight velocity ) .
second , the acceleration in the radial velocity from the bbh is the largest at zero velocity offset ( see [ sec : prelim ] ) , therefore on average it is easier to detect acceleration for bbhs with non - offset broad lines than those with offset broad lines . in this paper we present results for the general quasar population , i.e. , without pre - selection of a subsample with double - peaked or offset broad lines , based on multi - epoch spectroscopy separated by up to several years . as shown in fig .
[ fig : hb_oiii_shift ] , the vast majority of our quasars have broad line velocity offset within @xmath19 of the systemic velocity ( based on [ oiii]@xmath205007 ) .
thus these quasars will not have been included in the samples of @xcite and @xcite .
our sample is thus substantially larger than the other two samples , allowing better statistical constraints . in the second paper of the series ( hereafter paper ii )
, we will focus on a subset of quasars selected with offset broad lines , and present our own multi - epoch spectroscopic data and analysis in the context of the bbh scenario .
the organization of the paper is as follows . in [ sec : prelim ] , we provide preliminaries on the broad line diagnostics in the bbh scenario .
the data and our approach to measure radial velocity acceleration with multi - epoch spectra are described in [ sec : data ] .
we pay particular attention to a proper error analysis , which is key to this kind of study . in
[ sec : result ] we present candidate detections of acceleration ( which implies a possible bbh ) , and the statistical constraints on the general bbh population in quasars using ( mostly ) non - detections . in
[ sec : disc ] we discuss the caveats of our approach and compare to previous studies , and we summarize our conclusions in [ sec : con ] . by default
all time separations are in the restframe of the quasar , and @xmath21 refers to the quasar bolometric luminosity unless otherwise specified .
we use `` offset '' to refer to the velocity difference between two lines in single - epoch spectra , and `` shift '' to refer to the changes in the line velocity between two epochs . for simplicity
we assume circular bbh orbits , but the general methodology can be applied to a bbh population with an eccentricity distribution as well .
velocity offset relative to the [ oiii]@xmath205007 line for different samples in this study , which are all consistent with the bulk quasar population ( e.g. , * ? ? ?
definitions of the `` good '' and `` superior '' samples can be found in [ sec : vel_shift ] and [ sec : general_pop ] , respectively .
the `` detected '' sample consists of individual quasars in which a significant velocity shift has been detected ( see [ sec : binary_candidate ] ) .
the individual detections do not seem to have a different distribution ( see [ sec : result ] ) .
, scaledwidth=48.0% ]
since essentially all quasars in our sample have single - peaked broad lines , we initially assume that the broad line emission is dominated by one active bh .
we will return to the possibility that the broad line emission comes from both bhs in [ sec : caveat ] . consider a postulated smbh binary on a circular orbit , where only bh 1 is active and powering its blr .
the orbital period , line - of - sight ( los ) velocity and acceleration ( relative to the systemic frame ) of the active bh are : @xmath22 where subscripts 1 and 2 refer to bh 1 and 2 , @xmath23 , @xmath24 is the inclination of the orbit , @xmath25 is the binary separation , and @xmath26 is the orbit phase .
we use the conventions @xmath27 and @xmath28pc ) . for zero velocity offset
relative to the systemic velocity , @xmath29 , and the acceleration is at its maximum .
the typical size of the blr ( for h@xmath1 ) around a single bh with mass @xmath30 is in the blr @xmath31 relation , and so slightly differs from directly observed relations ( e.g. , * ? ? ?
* ; * ? ? ?
it also changed slightly from @xcite to reflect the most recent measurements of the @xmath31 relation . in our analyses in [ sec : data ] and [ sec : result ] we estimate the blr size using the best - fit relation reported in @xcite instead of the approximation in eqn .
( [ eqn : r_l ] ) . ]
( shen & loeb 2010 ) @xmath32 following the observed @xmath31 relation for the reverberation mapping agn sample ( e.g. , * ? ? ?
* ; * ? ? ?
here @xmath33 and @xmath34 are constants , and @xmath35 is the luminosity eddington ratio with @xmath36 the eddington luminosity .
this observed relation is tight , with only @xmath37 scatter in the predicted blr size ( e.g. , * ? ? ?
the dynamical time of the blr is @xmath38 where @xmath39 is the broad line fwhm in units of @xmath40 , and @xmath41 is the blr size in units of 0.1 pc .
now we can derive some crude criteria for the detectability of the binary motion with two - epoch spectroscopy separated by restframe time @xmath42 .
for simplicity , we ignore the detailed shape of the line profile and consider only the width ( fwhm ) of the broad line .
suppose the sensitivity of our velocity shift measurement is a fraction of the fwhm of the broad line , @xmath43fwhm
. then we require the following condition to be met for the detection : @xmath44 .
@xmath43 is typically _ inversely _ proportional to the s /
n of the line ( e.g. , * ? ? ?
* and see [ sec : mc_sim ] ) . obviously , detection of binary motion is easier if the time separation @xmath42 is large ( so long as @xmath45 , which is generally true ) , the binary separation @xmath25 is small , and the fwhm is small
. the radius of the blr must be smaller than the separation of the bhs , so @xmath46 , where @xmath47 is less than unity .
otherwise the blr becomes circumbinary , the dynamics of the blr becomes more complicated , and there may be no coherent velocity drift in the broad line emission ( e.g. , * ? ? ?
@xmath48 is then the maximum size of the blr before it is dynamically affected by the companion dormant bh in the system .
we can define @xmath47 as the average radius of the roche lobe in a circular binary system ( e.g. , * ? ? ?
* ) : @xmath49 where @xmath50 is the binary mass ratio .
now we consider this maximum blr size , i.e. , we set @xmath51 , and derive some crude constraints on the detectability based on the properties of the active bh 1 .
since the blr around bh 1 is dominated by the gravity of bh 1 , we have @xmath52 where @xmath53 is the virial coefficient , determined by the geometry of the blr .
an empirical estimate is @xmath54 ( e.g. , * ? ? ?
substituting eqn .
( [ eqn : mvir ] ) to eqn .
( [ eqn : fom_condition ] ) we obtain @xmath55 using the observed @xmath31 relation we can recast the above equation into : @xmath56 from which we can define a figure of merit ( fom ) of the detection : @xmath57 this fom is based on two observables , line width and luminosity , and the larger it is , the more likely we will detect the velocity shift due to binary motion in a hypothetical bbh , assuming everything else is equal in eqn .
( [ eqn : fom_condition2 ] ) . if we substitute fwhm in eqn .
( [ eqn : fom_condition2 ] ) with eqn .
( [ eqn : mvir ] ) and use the @xmath31 relation ( [ eqn : r_l ] ) we obtain @xmath58 where all the other known quantities are absorbed in @xmath59 .
we can therefore define an alternative fom based on the luminosity and eddington ratio of the active bh 1 : @xmath60 from this we see that if all quasars have the same eddington ratio and other conditions are the same , including the companion mass @xmath61 , lower luminosity ( or equivalently , lower @xmath30 ) quasars are more likely to have detectable velocity shifts due to binary motion ( if instead the mass ratio @xmath62 were constant , fom@xmath63 would vary as @xmath64 at fixed eddington ratio , so the conclusion would be similar but weaker ) .
one caveat here for both figure - of - merit definitions is that @xmath43 in conditions ( [ eqn : fom_condition2 ] ) and ( [ eqn : fom_condition3 ] ) scales inversely with s / n , while generally lower luminosity objects have lower s / n at fixed redshift in a given survey , which will reduce the detectability .
note that in our postulated binary system , there is only one active bh .
this is different from the scenario studied in @xcite , where both bhs are active .
therefore we do not require the additional condition that the difference in los velocities of the two blrs must exceed the fwhm of the line to make a double - peaked profile ( e.g. , the upper limit of @xmath25 in eqn .
6 of * ? ? ?
lcccccc all & 1910 & 1347 & 688 & 521 & 193 & 163 + @xmath65 days & 757 & 613 & 235 & 207 & 0 & 0 + @xmath66 days & 834 & 674 & 316 & 268 & 193 & 163 + @xmath67 yrs & 251 & 223 & 104 & 94 & 98 & 90 + @xmath68 yrs & 117 & 106 & 53 & 49 & 53 & 49 + @xmath69 yrs & 69 & 65 & 27 & 26 & 27 & 26 + @xmath70 yrs & 6 & 6 & 4 & 4 & 4 & 4
we start from the dr7 compilation of the spectroscopic quasar catalog @xcite from sdss .
this sample contains 105,783 quasars brighter than @xmath71 that have at least one broad emission line with fwhm larger than 1000 @xmath72 or have interesting / complex absorption features .
the reduced and calibrated one - dimensional ( 1d ) spectral data used in this study are available through the sdss data archive server ( das ) .
the spectral resolution is @xmath73 , and the 1d spectra are stored in vacuum wavelength , with a pixel scale of @xmath74 in log - wavelength , which corresponds to @xmath75 .
all spectra are wavelength calibrated to the heliocentric reference , with an accuracy of better than @xmath76 . throughout the paper
, we refer to the signal - to - noise ratio per pixel as s / n .
several thousand of these dr7 quasars have multiple spectra taken at different epochs by sdss .
most of them have two epochs , and a small fraction of them have multiple epochs .
the default spectrum reported in the dr7 quasar catalog is usually the one with the highest s / n , and we always take that default spectrum as epoch 1 regardless of the relative time between the two epochs .
epoch 2 is then the other available spectra in sdss ( could be more than one ) .
we call each epoch 1/2 combination a pair of observations .
these quasars with repeated spectroscopy in sdss form the sample in which we search for velocity shift of the broad line .
we will focus on the h@xmath1 line for this systematic study , which restricts us to @xmath77 , and reduces the number of available quasars to below two thousand .
we choose the h@xmath1 line rather than other broad lines for the following reasons : ( 1 ) it has robust systemic redshift from the lines ( which are shifted out of sdss spectrum for @xmath78 ) ; ( 2 ) the presence of line in the sdss spectrum provides an empirical way to subtract the narrow h@xmath1component ; ( 3 ) the restframe time separation @xmath42 between two epochs is longer for the h@xmath1sample than for higher - redshift quasars ; ( 4 ) the broad balmer lines are better studied and better understood than or civ lines , especially in terms of reverberation mapping studies , therefore their blr size and virial bh mass estimates are the most reliable ( e.g. , * ? ? ? * and references therein ) ; ( 5 ) is difficult because of the complication of the strong emission underneath ; civ is difficult because this line is more asymmetric and may arise from a disk - wind component ; in addition , for both and civthe narrow line subtraction will be ambiguous ; ( 6 ) finally , h@xmath79 is a line complex enclosing more than one narrow line ( [ nii]@xmath806548,6584 , narrow h@xmath79 ) , and the broad line component isolation is seemingly difficult . nevertheless , if we detect any h@xmath1 velocity shift , we will investigate the and h@xmath79 lines at the two epochs , if covered in the sdss spectrum .
the basic sample statistics are summarized in table 1 . for restframe @xmath81days , there are @xmath82 pairs of observations over 674 unique quasars ; most of them have a restframe time span of less than a year .
we do not remove duplicated objects , and treat multiple pairs for the same object as independent measurements in our statistical analysis in [ sec : result ] .
component and green is the broad h@xmath1 components .
the sdss designation and the number of gaussians for the broad h@xmath1 line are marked in the top left corner of each panel .
these two examples represent different broad line widths , and strengths . , title="fig:",scaledwidth=50.0% ] component and green is the broad h@xmath1 components .
the sdss designation and the number of gaussians for the broad h@xmath1 line are marked in the top left corner of each panel .
these two examples represent different broad line widths , and strengths .
, title="fig:",scaledwidth=50.0% ] in order to characterize the properties of quasars in our sample , we measure the continuum and emission line properties using a combination of model fitting and direct measurements on the spectrum .
this process also produces the broad - line spectra ( i.e. , continuum and narrow line removed ) on which we will perform the cross - correlation to measure velocity shifts in [ sec : vel_shift ] .
the basic spectral fitting approach follows our earlier work in @xcite . in short , an template plus a power - law continuum
is fit to several line - less windows around h@xmath1 to form a pseudo - continuum ; this pseudo - continuum model is then subtracted from the spectrum , leaving the emission lines .
the narrow emission lines are modeled by gaussians ( in logarithmic wavelength ) with the same width and their relative velocity offsets fixed to those based on their restframe wavelengths . whenever necessary we fit two gaussians to the lines to account for the core and the wing components often seen in ; the narrow h@xmath1 is tied to the core component in such cases .
the broad h@xmath1 profile is fit with a mixture of gaussians ( up to 5 , but 3 are enough in most cases ) .
these gaussian functions are merely used to reproduce the line profile and we assign no physical meaning to individual gaussian components .
we visually inspect all fits to make sure they are reasonable in the sense that the line is well reproduced by the model , and we require a reduced @xmath83 for the fits . fig . [
fig : fit_examp ] shows two examples of the decomposition of different spectral components .
the [ oiii]@xmath205007/narrow h@xmath1 flux ratio is tied to be the same at both epochs .
this is important for cases where the narrow h@xmath1 contributes a substantial amount to the h@xmath1 profile , and incorrect subtraction of the narrow h@xmath1 emission will likely impact the cross - correlation analysis in
[ sec : vel_shift ] . due to aperture effects and
seeing / guiding variations between the two epochs , as well as the extended nature of the narrow line regions , the absolute narrow line flux could change , and so we do not fix the normalization of the narrow line flux at the two epochs .
visual inspections were performed to ensure there is no artificial line variation near the systemic velocity between the two epochs due to the narrow line removal .
once we have decomposed the spectrum around the h@xmath1 region , we measure a variety of broad line properties both using the model fit and using the raw ( broad - line only ) spectrum : * line peak , fwhm , line flux , and restframe equivalent width ( ew ) of the broad line from the model fit ; * line centroid or first moment of the line : @xmath84 , where @xmath85 and @xmath86 are the pixel wavelength and flux density . to compute the centroid and higher order moments we use the raw spectrum , and we only use pixels that enclose @xmath87 of the line peak .
this is to reduce the adverse effects of the noisy wings in determining the line moments .
we also compute the second moment , skewness and kurtosis of the broad line , again using all the pixels enclosing @xmath88fwhm of the line peak .
the @xmath89-th moment of the line is defined as @xmath90 , the skewness is defined as @xmath91 , and the kurtosis is defined as @xmath92 .
we compared the distributions of these broad line profile characteristics to other studies ( e.g. , * ? ? ?
* ; * ? ? ?
* ) whenever available , and found good agreement . ; ( d ) fwhm of the broad h@xmath1 .
the measurement error decreases as s / n increases , and increases as line width increases , as expected .
in addition , stronger lines ( larger ew ) have on average smaller measurement errors in line centroid , but the trend is not as strong as those with s / n and fwhm , and is likely a secondary effect of s / n .
, scaledwidth=50.0% ] we use a monte carlo approach to estimate the measurement errors in the above various quantities ( e.g. , * ? ? ?
* ; * ? ? ?
* ) . in practice
we add artificial noise ( using the reported flux errors at each pixel ) to the model spectrum to create a mock spectrum , repeat our fitting and record the results , for 50 trials .
we then use the semi - quantile of the @xmath93 range of these mock results as the nominal measurement uncertainty of each measured quantity .
[ fig : err_study_cent ] shows how the measurement error in the broad line centroid , @xmath94 , changes with various quantities .
the most significant trends are the correlations of @xmath94 with s / n and fwhm : @xmath94 decreases as s / n increases , and increases as fwhm increases .
this is expected , as the uncertainty in measuring the line centroid is proportional to line width , and is inversely proportional to s / n , i.e. , @xmath95 .
however , the trends shown in fig . [ fig : err_study_cent ] somewhat deviate from the expected dependence on fwhm and s / n , which is likely caused by the following : a ) there is a slight anti - correlation between fwhm and s / n for our objects ; b ) the line shape also changes with line width , which may increase the difficulty of measuring line centroid for large line widths .
finally , there is a weak dependence of measurement errors on the strength of the line , such that stronger lines on average have smaller errors in measuring their centroid .
this weak trend with line strength is likely a secondary effect of s /
n , i.e. , given fixed continuum level and photon noise , weaker lines have smaller s / n .
line centroid ( or line peak ) is not a robust indicator for the velocity shift we are trying to measure , because it is only one characteristic of the broad line : slight changes in the line shape may produce artificial centroid shifts or hide a real shift . in the next section , we will measure the broad line velocity shift with a cross - correlation technique , which is more robust in measuring the broad line velocity shift between the two epochs .
the measured broad h@xmath1 shape parameters will be used in our analyses in [ sec : result ] . in order to constrain the velocity shift between two epochs and hence the orbital acceleration in the hypothetical binary ,
we deploy a @xmath96 cross - correlation technique ( `` ccf '' for short ) similar to that used in @xcite .
we describe this approach in detail below .
we consider the pseudo - continuum and narrow - line subtracted spectra from [ sec : fit ] , i.e. , broad line spectra only .
this is different from the approach in @xcite , which used several broad line windows that avoid the narrow h@xmath1 line and did not subtract the underlying pseudo - continuum in the cross - correlation .
we prefer to use the broad line spectra in the cross - correlation , because possible changes in the continuum slope will potentially bias the cross - correlation result if the full spectrum is used .
we also want to use a large portion of the broad line profile to increase the statistical power in the cross - correlation , which requires removal of the central narrow h@xmath1 component . as discussed in [ sec : fit ] , we carefully controlled the systematics from narrow h@xmath1 subtraction in our fitting , and visually inspected all fits to ensure that the narrow line subtraction does not introduce artificial changes between the two epochs near the systemic velocity .
we shift the second epoch broad line spectrum by an increasing number of pixels , and compute the @xmath96 as a function of the shift : @xmath97 where @xmath98 and @xmath99 are the flux density and flux density errors of the @xmath100th pixel in the epoch 1 spectrum , and @xmath101 and @xmath102 are the corresponding quantities of the @xmath100th pixel in the shifted epoch 2 spectrum .
we only use pixels within [ 4800,4940] to compute the @xmath96 , which encloses most of the broad line flux and excludes the noisy wings . during the @xmath96 calculation
, we also scale the second - epoch spectrum by the ratio of the integrated broad line flux within [ 4800,4940] of the two epochs .
the latter step is necessary because the broad line flux could vary between the two epochs , which will reduce the statistical power of the cross - correlation .
one great advantage of using the sdss data is that both spectra were reduced using the same software and binned on the same logarithmic wavelength grid , i.e. , 1 pixel shift corresponds to @xmath103 shift .
the step size in the grid of shifts that we use to compute the cross - correlation is 1 pixel , i.e. , we do not use sub - pixel shifts and rebinning of the second - epoch spectra as in @xcite .
we chose to do so because the s / n per pixel is typically several , and so by interpolating on a sub - pixel grid and rebinning we would likely introduce systematics , which would complicate the resulting error estimation of the measured velocity shift .
once we have the @xmath96 values , we determine the minimal @xmath96 and its location as follows : we fit the 20 points enclosing the minimal @xmath96 point with a 6th - order b - spline function .
the minimal @xmath96 value is then determined from the spline , which allows us to measure sub - pixel shifts , as well as reduce the impact of noisy @xmath96 curve .
we estimate the confidence range by finding the intercepts of the b - spline at @xmath104 , which corresponds to @xmath105 confidence range ( @xmath106 ) . in this way we measure the velocity shift @xmath107 and the associated uncertainty between the two epochs .
cross - correlation .
the epoch 2 spectrum has been scaled to the epoch 1 spectrum using the integrated line flux within the cross - correlation wavelength range .
the bottom right panel shows the @xmath96 curve for h@xmath1 as a function of pixel shift ( recall that 1 pixel corresponds to 69 @xmath108 ) in open triangles , where the magenta line is the 6th - order b - spline on the 20 points centered on the point with the minimal @xmath96 .
the dashed horizontal segment indicates the @xmath109 ( 2.5@xmath110 ) range in pixel shift , also indicated in the magenta square bracket . in this example a velocity shift is detected at @xmath111 significance .
the @xmath96 cross - correlation test on the [ oiii]@xmath205007 line yields a 2.5@xmath110 velocity shift range of @xmath112 $ ] pixels , not enough to account for the velocity shift measured for h@xmath1 . , scaledwidth=48.0% ] , but for a case where no significant velocity shift is detected over a restframe time span of @xmath9 a year .
, scaledwidth=48.0% ] blr size versus redshift .
the bolometric luminosity and virial bh mass estimates were taken from the @xcite catalog .
the blr sizes were estimated from the @xmath113 continuum luminosity using the best - fit @xmath114 relation in @xcite , therefore they have explicit dependence on luminosity .
the gray points are for our parent sample , and the black points are the `` good '' subsample defined in [ sec : vel_shift ] in which we search for velocity shifts and statistical constraints on the general quasar population .
the green points are the `` superior '' sample defined in [ sec : general_pop ] , which is a subset of the `` good '' sample .
finally the red points are individual quasars in which a significant velocity shift has been detected ( see [ sec : binary_candidate ] ) .
other than removing objects with low - quality spectra at high redshift and low luminosity , the good sample is statistically indistinguishable from the parent sample .
in addition , the `` superior '' sample is statistically indistinguishable from the `` good '' sample , as confirmed by standard ks tests .
, title="fig:",scaledwidth=48.0% ] blr size versus redshift .
the bolometric luminosity and virial bh mass estimates were taken from the @xcite catalog .
the blr sizes were estimated from the @xmath113 continuum luminosity using the best - fit @xmath114 relation in @xcite , therefore they have explicit dependence on luminosity .
the gray points are for our parent sample , and the black points are the `` good '' subsample defined in [ sec : vel_shift ] in which we search for velocity shifts and statistical constraints on the general quasar population .
the green points are the `` superior '' sample defined in [ sec : general_pop ] , which is a subset of the `` good '' sample .
finally the red points are individual quasars in which a significant velocity shift has been detected ( see [ sec : binary_candidate ] ) .
other than removing objects with low - quality spectra at high redshift and low luminosity , the good sample is statistically indistinguishable from the parent sample . in addition , the `` superior '' sample is statistically indistinguishable from the `` good '' sample , as confirmed by standard ks tests .
, title="fig:",scaledwidth=48.0% ] blr size versus redshift .
the bolometric luminosity and virial bh mass estimates were taken from the @xcite catalog .
the blr sizes were estimated from the @xmath113 continuum luminosity using the best - fit @xmath114 relation in @xcite , therefore they have explicit dependence on luminosity .
the gray points are for our parent sample , and the black points are the `` good '' subsample defined in [ sec : vel_shift ] in which we search for velocity shifts and statistical constraints on the general quasar population .
the green points are the `` superior '' sample defined in [ sec : general_pop ] , which is a subset of the `` good '' sample .
finally the red points are individual quasars in which a significant velocity shift has been detected ( see [ sec : binary_candidate ] ) .
other than removing objects with low - quality spectra at high redshift and low luminosity , the good sample is statistically indistinguishable from the parent sample .
in addition , the `` superior '' sample is statistically indistinguishable from the `` good '' sample , as confirmed by standard ks tests .
, title="fig:",scaledwidth=48.0% ] between two epochs using the @xmath96 ccf method .
_ bottom _ : distributions of the measurement errors in the los acceleration measured from two epochs , i.e. , @xmath115 . the black lines are for the `` good '' sample ( [ sec : vel_shift ] ) , while the green dashed lines are for the `` superior '' sample ( [ sec : general_pop ] ) . ,
title="fig:",scaledwidth=48.0% ] between two epochs using the @xmath96 ccf method .
_ bottom _ : distributions of the measurement errors in the los acceleration measured from two epochs , i.e. , @xmath115 . the black lines are for the `` good '' sample ( [ sec : vel_shift ] ) , while the green dashed lines are for the `` superior '' sample ( [ sec : general_pop ] ) .
, title="fig:",scaledwidth=48.0% ] we also apply this cross - correlation method to the [ oiii]@xmath205007 line , using pixels within [ 4995,5020] .
this allows us to check possible issues of bad wavelength calibration associated with either epoch . in most cases
the shift of the [ oiii]@xmath205007 line is constrained to be @xmath116 , which is expected as the sdss wavelength calibration is better than @xmath76 .
we will use the uncertainty in the shift as an additional constraint to select individual detections of broad h@xmath1 velocity shift in [ sec : binary_candidate ] .
[ fig : examp_ccf ] shows an example of our @xmath96 cross - correlation procedure , where a velocity shift is detected at the 2.5@xmath110 significance level , and the restframe time separation is @xmath117 months .
[ fig : examp_ccf2 ] shows another example , where no significant velocity shift is detected over a restframe timescale of about a year . in both examples
the shape of the broad h@xmath1 remains more or less the same , as indicated by the fwhm .
many sdss spectra are of low quality and often the resulting @xmath96 curve is too noisy to measure a well - defined minimum and confidence ranges . to eliminate these , we first filter our sample using a simple s / n cut , keeping only objects where both epochs have detected broad h@xmath1 line at @xmath118 significance ( where line fluxes and errors are taken from our spectral fits in [ sec : fit ] ) .
we then visually inspect the b - spline fits for all objects , and discard cases where the @xmath96 curve is noisy , such that either a meaningful spline fit is not possible , or that the formal confidence range from the spline fit is questionable .
for such cases the two epochs can not constrain the velocity shift well using this cross - correlation method .
these objects usually have noisy spectra , or very broad lines ( fwhm@xmath119 ) . for the remaining objects ,
the @xmath96 curve shows a well - determined minimum and the b - spline provides reasonably good measurements of its minimum location and associated confidence levels .
ccf method , with mock spectra . shown here
are the measurement errors as functions of peak line s / n ( per pixel ) and fwhm .
the points and error bars are the median and standard deviations from 5000 trials in our monte carlo mock tests .
the dashed lines are not a fit to the points , but straight lines with a power - law slope of @xmath120 .
this test indicates that the measurement error decreases as s / n increases and fwhm decreases , as expected .
the two - sided arrows at the top indicate the @xmath93 quantile range of actual peak line s / n for the `` good '' sample ( black ) and the `` superior '' sample ( green ) .
our quasars have typical broad h@xmath1 fwhm@xmath121 .
thus the expected typical measurement error is @xmath6 ( 1@xmath110 ) , fully consistent with the actual values shown in fig .
[ fig : sig_v ] . , scaledwidth=48.0% ] ccf approach in the `` good '' sample defined in
[ sec : vel_shift ] . _ top : _
results for a subsample of observations with time span less than 30 days ( in restframe ) .
the dispersion of velocity shift within such short period is due to a combination of measurement errors and potential systematic errors due to data reduction and short - term variability .
the black histogram is the observed distribution of velocity shift between two epochs .
the blue dashed line is the simulated distribution using a sample of zero velocity shifts , shuffled with the measurement error distribution .
the red line is the same as the blue line , but shuffled by an additional gaussian error of @xmath122 .
the ks test probabilities of both simulated distributions being drawn from the same distribution as the observed distribution ( black ) are marked on the upper left corner .
this test suggests that any additional systematic error terms are small , and our measurement error distribution is reasonable . _
bottom : _ same as the top panel , but for observations with a time span longer than 30 days .
, scaledwidth=48.0% ] days and a simulated distribution of zero velocity shifts broadened by the measurement errors scaled by a constant . under the assumption that the intrinsic distribution for these objects should be consistent with zero
, this test suggests that our formal estimates of the 1@xmath110 measurement error in velocity shift slightly underestimate the true error by @xmath123 .
, scaledwidth=48.0% ] , for which we have measured a reliable velocity shift between the two epochs with the @xmath96 cross - correlation method .
the reduction from the parent sample to the good sample is mostly due to the quality of the spectra , and does not change the distribution of @xmath42 significantly .
the green line is the `` superior '' sample defined in [ sec : general_pop ] , which is a subset of the `` good '' sample with small measurement errors on the measured radial acceleration . as a result ,
the `` superior '' sample requires on average longer time separations than the `` good '' sample .
finally , the red points show the fraction of detections ( multiplied by @xmath124 ) in the good sample , in each time separation bin , and the error bars are poisson .
the numbers above the red points are the actual number of detections .
the detection fraction generally increases as a function of time separation .
, scaledwidth=50.0% ] line shape properties for various samples .
typical measurement errors are shown in each panel .
the dotted lines mark the median values for each sample , in their corresponding colors .
panel ( a ) shows an anti - correlation between the broad line peak offset ( relative to [ oiii]@xmath205007 ) and the skewness of the line , which is consistent with fig . 5 in @xcite .
note that the definition of skewness in the latter work has the opposite sign to ours .
panel ( b ) plots the skewness against the kurtosis .
panel ( c ) plots fwhm against broad line peak offset .
panel ( d ) plots the relative changes in fwhm against the absolute changes in skewness between the two epochs .
significant shape changes between the two epochs are not common .
these four plots indicate that objects with detected broad h@xmath1 velocity shifts between the two epochs do not stand out as a distinct population .
, scaledwidth=50.0% ] while our rejection based on visual inspection of the @xmath96 curves is not rigorous , it is necessary to reject false positives and obtain a clean sample .
as long as the remaining objects probe the same parameter space as the parent sample , we are not biasing the sample selection by this visual screening ( see below ) .
there are 688 objects kept in our final sample , which we call the `` good '' sample .
we will focus on this sample in the following analysis .
[ fig : lbol_z ] shows the statistical quasar properties of the good sample and the parent sample .
objects at high redshift and low luminosities are preferentially removed from our good sample because they have low quality spectra .
other than that , the good sample probes the same parameter space of the general sdss quasar population , thus there is no obvious bias from our sample refining .
the resulting error distributions of the measured velocity shift and acceleration of the good sample are shown in fig .
[ fig : sig_v ] . the virial bh masses probed by our sample have a median value of @xmath125 with a dispersion of @xmath126 dex .
we fully appreciate the uncertainty in these mass estimates ( for a comprehensive review of the caveats of virial bh masses , see , e.g. , * ? ? ?
the @xmath126 dex dispersion in these virial bh masses is comparable to the typical virial mass uncertainty , and so it is reasonable to assume that our sample only probes a limited dynamic range in bh mass .
the blr size of our quasars is typically below @xmath127 pc , with a median value of @xmath128 pc .
the uncertainty of these blr size estimates based on @xmath129 is @xmath37 ( e.g. , * ? ? ?
* ) , much better than that for the bh mass estimates , and so we will use these blr size estimates as true values .
these physical properties will be used in our later analyses in [ sec : result ] . to use the cross - correlation results in our statistical analyses
, we must make sure that our error estimation of the velocity shift is reliable .
we first test this using simple monte carlo experiments on mock data .
we generate mock broad line spectra with different fwhm and s / n , sampled on the same wavelength grid as the sdss spectra . for simplicity we assume a single - gaussian intrinsic profile for the broad line , and we use the peak s / n ( per pixel ) as the indicator for the spectral s / n .
we then run exactly the same @xmath96 cross - correlation procedure on the mock data , and derive the errors of the measured velocity shift between two epochs .
[ fig : mc_sigv ] summarizes the results .
the data points show the median and standard deviations from 5000 trials . for
given fwhm , the measurement uncertainty decreases with s / n roughly as @xmath130 .
this trend breaks down at both the large fwhm / low - s / n end and the small fwhm / high - s / n end . in the former regime
it is very difficult to measure the velocity shift of a noisy and broad line profile , as indicated by the large scatter in the error estimations .
the latter is likely inherent to our @xmath96 error estimation procedure , such as the finite bin size in the ccf and the b - spline , but is not a concern since it is beyond the regime of interest in terms of s / n and line width . on the other hand , at fixed s / n , the measurement uncertainty increases as fwhm increases .
this trend with fwhm is slightly slower than @xmath131 because we are using the peak s / n instead of the s / n integrated over the line profile .
overall these trends are fully consistent with the expectations of the precision in measuring the profile center with the cross - correlation method ( e.g. , * ? ? ?
* ) , suggesting that our error estimation is reasonable .
the second test is to study the distributions of measured velocity shifts for the good sample .
[ fig : test_meas_errors ] shows the distributions of measured velocity shifts for pairs separated by @xmath65 and @xmath132 days in restframe ( black histograms ) . for @xmath65 days , if we assume there is no intrinsic change in the broad line , given that this time is much shorter than the blr dynamical timescale or possible binary period , then the expected velocity shift distribution should be consistent with zero , broadened by the error distribution .
the blue dashed line in the upper panel shows the expected zero velocity shift shuffled by our measurement errors for these objects , and it is not too much different from the observed distribution .
if we add an additional error component of @xmath122 , the resulting velocity shift distribution would be the red histogram , which is inconsistent with what we observe .
therefore we are not _ underestimating _ our errors in the velocity shift by much . to quantify this statement we scale the errors up by a constant factor , and perform ks tests between the observed distribution and a simulated distribution of zero velocity shifts shuffled by the scaled - up errors .
the results are shown in fig .
[ fig : test_meas_errors_ks ] , where we plot the median and standard deviations from @xmath133 random realizations at each error scale .
we achieve the best agreement with the observed distribution if our nominal errors were scaled up by only @xmath134 .
we also verify that the measurement error in the velocity shift , @xmath135 , does not vary with the time separation between two epochs , as also indicated by the similar distributions of the blue dashed lines in the two panels of fig .
[ fig : test_meas_errors ] .
thus the observed velocity shift distribution for @xmath132 days shows excess variance compared with zeros broadened by errors .
however , the errors in the measured velocity shifts are still quite large for many objects , prohibiting tight constraints on the intrinsic velocity shift distribution .
therefore in [ sec : general_pop ] we use a subset of the good sample with small measurement errors to place more stringent constraints on the intrinsic distribution .
the tests performed in this section suggest that our error estimation of the measured velocity shift is reliable , and hence we will proceed to statistical constraints and individual detections in the next section .
for the 688 pairs of observations that define our good sample , we have derived meaningful constraints on the velocity shift between two epochs from the cross - correlation analysis . as shown in fig .
[ fig : sig_v ] , the typical uncertainty in the measured velocity shift is @xmath136 .
although our spectra are generally of lower quality , this typical uncertainty is comparable to or better than that achieved in @xcite , mainly due to the fact that we are using a larger portion of the broad line in the cross - correlation than in @xcite .
we have 30 detections ( @xmath106 , 28 unique objects ) among the 688 pairs in the good sample , 2 of which are detections with a restframe time difference of less than one month .
[ fig : hbeta_hist ] shows the detection fraction with poisson errors for objects in the good sample , as a function of time separation .
there is an apparent increase of the detection fraction as the time baseline increases .
the detection fraction is @xmath137 a few percent at @xmath138yr , but increases to @xmath139 at @xmath140yr .
since there is no dependence of the spectral quality or line width ( which determine the precision of our measurements ) on restframe time separation , this increased detection rate with time is real .
it indicates either that the broad lines are more variable on longer timescales , in agreement with the general expectation of broad line variability ( as the dynamical time of the blr is on the order of a few years , see [ sec : prelim ] ) , or that we are detecting binary bh orbital acceleration , which leads to a velocity shift that grows linearly with time .
the detection rate with a restframe time baseline longer than a few years is consistent with the results in @xcite and paper ii , but we caution that the statistics are poor and can still hide large differences . the detection rate at @xmath65 days can be used to estimate the false - positive rate due to statistics and systematics .
we only detected 2 out of @xmath141 pairs with such short time intervals , or @xmath142 this is consistent with our detection threshold of @xmath106 .
this should be considered as an upper limit on the false - positive rate as both of the detections could be real with large accelerations ( see further discussion of these two objects in [ sec : class2 ] ) .
thus the majority of detections with longer time baselines should be genuine .
[ fig : sample_line_prop ] shows the broad h@xmath1 line shape properties of the `` good '' , `` superior '' ( defined in [ sec : general_pop ] ) and the detected samples , which suggest that objects with detected broad h@xmath1 velocity shifts are not a distinct population from the parent sample .
in addition , significant profile changes ( in terms of fwhm and skewness ) are not common for the timescales probed by our sample .
below we will first derive constraints on the general population with the full distribution of observed velocity shifts ( mostly non - detections ) in [ sec : general_pop ] , and then discuss individual detections in [ sec : binary_candidate ] .
is the black histogram , with @xmath143 .
, scaledwidth=50.0% ] most of our objects do not show detectable acceleration between the two epochs .
we can use these non - detections to place constraints on a hypothetical binary population .
specifically we will compare the distribution of measured accelerations with predictions from a binary population , using the measurement error distribution in our @xmath96 ccf approach .
in doing so , we have to assume that all the observed accelerations are either due to binary motion , or due to errors in the measurements , i.e. , we neglect the complication that some apparent `` acceleration '' may be due to broad line variability .
we also stick to the simplified scenario described in [ sec : prelim ] that only one bh is active and dominates its blr dynamics .
we will come back to the caveats in our assumptions in [ sec : disc ] . throughout [ sec : general_pop ] we adopt a constant mass for bh 1 , @xmath144 , i.e. , the median virial mass estimate for our sample .
the reason for adopting a constant value rather using the full distribution of viral masses is because the latter is dominated by the large uncertainties associated with these virial bh masses ( see discussion in [ sec : vel_shift ] ) .
but we will use the full distribution of the measured blr sizes , since the uncertainties associated with these estimates are substantially smaller . .
shown here is the case where the binary fraction is @xmath145 with a fixed companion bh mass @xmath146 , the median virial mass for the active bh .
the black histogram shows the observed acceleration distribution .
the blue dashed histogram shows a simulated sample of zero acceleration broadened by the measurement error distribution , which is narrower than the observed distribution , indicating non - negligible acceleration .
we consider two simple cases : 1 ) the binary separation is twice the blr size of the active bh ; 2 ) the binary separation equals the blr size .
the simulated distributions ( with measurement errors ) of the two cases are shown in the magenta and red lines .
the case with the smaller binary separation is ruled out at a high significance . _
top : _ differential probability distributions . _
bottom : _ cumulative probability distributions .
, scaledwidth=50.0% ] , but with the measurement errors scaled up by @xmath134 .
the pure error distribution is still narrower than the observed distribution , requiring an intrinsic acceleration distribution .
, scaledwidth=50.0% ] to get the strongest constraints possible , we consider a further refinement of the `` good '' sample , with stringent limits on the precision of the measured acceleration between the two epochs , i.e. , @xmath147 .
[ fig : accel_err_cut ] shows the acceleration distributions for subsets of the `` good '' sample with different cuts on @xmath148 .
as we increase the threshold error , the observed distribution broadens . therefore to better constrain the intrinsic acceleration distribution , while still keeping a sufficient number of observations , we use the stringent @xmath149 cut in refining the `` good '' sample .
there are @xmath150 pairs that satisfy this criterion , and all pairs are separated by a restframe time span @xmath151 yr .
we refer to this sample as the `` superior '' sample . as shown in fig .
[ fig : lbol_z ] , the distribution of these objects in the quasar parameter space is statistically indistinguishable from the `` good '' sample , hence we are not biasing the sample by imposing this cut on @xmath148 .
we also verified that the error distribution in measured accelerations does not correlate with the quasar continuum luminosity ( hence the broad line size via the @xmath31 relation ) or virial bh masses , which simplifies our simulations below using the measured blr size distribution .
[ fig : test_ccf_aerr ] demonstrates how we use the observed acceleration distribution to constrain the binary population . in this example
, we fix the companion bh mass to be @xmath146 , the median virial bh mass of the active bhs in the `` good '' sample .
we also consider a @xmath145 binary fraction , i.e. , all the objects in our sample are in a binary system with only one bh active .
to obtain the expected acceleration distribution for this hypothetical binary population , we must also assume the distribution of binary separations .
we consider an ideal case where the binary separation is a fixed scaling of the blr size of the active bh , @xmath152 with @xmath153 .
the intrinsic distribution of the radial acceleration can then be derived using eqn .
( [ eqn : v_a ] ) assuming random orientations and orbital phases of the binary , and using the distribution of blr sizes measured from @xmath154 using the best - fit @xmath31 relation in @xcite .
we then convolve the intrinsic distribution with the measurement error distribution to obtain the simulated acceleration distribution expected from a binary bh population .
the black histogram in fig .
[ fig : test_ccf_aerr ] is the observed acceleration distribution , and the blue dashed histogram is a simulated distribution of zero acceleration broadened by the error distribution .
the simulated distribution of zero accelerations with errors is too narrow to account for the observed distribution , indicating non - negligible accelerations in the intrinsic distribution .
the magenta and red histograms are the simulated distributions for a binary population with @xmath155 and @xmath156 , convolved with measurement errors .
the case with @xmath155 produces consistent acceleration distribution with the observed one , but the case with @xmath156 is strongly ruled out ( again , assuming that _ all _ quasars are in equal - mass binaries with this separation ) . as we have discussed in
[ sec : mc_sim ] , it is likely that our estimated measurement errors in velocity shifts are too small by @xmath123 .
we therefore scale our formal errors up by @xmath134 and repeat the above exercise in fig .
[ fig : test_ccf_aerr2 ] .
again , this test suggests there is non - negligible intrinsic acceleration , as the simulated distribution of zero accelerations is still narrower than the observed one .
the @xmath145 binary scenario with @xmath156 is even more strongly ruled out . note .
@xmath157velocity shift and acceleration are calculated based on the broad h@xmath1 line , and the first column is the value corresponding to the minimum @xmath96 and the last two columns enclose the @xmath158 confidence range ( @xmath109 ) .
we classify the detections into three categories ( see [ sec : binary_candidate ] ) : 1 ) binary candidates ; 2 ) broad line variability ; 3 ) ambiguous cases .
now we consider the general case with different binary fractions and binary separations .
we determine the ks probability that the expected acceleration distribution is drawn from the same distribution as the observed one , for a series of values of @xmath159 , the binary fraction , and the mass of the companion bh , @xmath61 .
as discussed earlier , we adopt a constant mass @xmath144 for bh 1 .
each evaluation is performed for a mock sample of @xmath160 systems .
we choose to parameterize the ks probability in terms of @xmath159 instead of @xmath25 , because objects in our sample spread over a range of blr sizes .
so for a fixed value of @xmath25 , some objects will not satisfy the criterion @xmath161 required in our binary scenario ( see [ sec : prelim ] ) , which will then complicate the expected distribution .
if we consider the more restrictive condition that the blr size must be smaller than the roche radius , then we must have @xmath162 , where @xmath47 is given in eqn .
( [ eqn : roche ] ) .
[ fig : ks_prob ] shows the contours of the ks probability as functions of the binary fraction and separation , at several different values of @xmath61 , using the `` superior '' sample with our fiducial error estimates in the measured acceleration .
the cyan lines show the @xmath163 probability contour , and darker regions are more probable .
the dashed vertical lines in each panel mark the minimum separation of the hypothetical bbhs in units of the blr size ( i.e. , the roche limit , eqn
. [ eqn : roche ] ) , in order for the broad line diagnostics to work . in each panel , regions with larger binary separations and lower binary fractions are disfavored , because they contribute negligible acceleration and thus can not explain the observation that the observed distribution of accelerations is larger than expected from the measurement errors .
thus the observed acceleration distribution ( error broadened ) favors islands in the two - dimensional parameter space with high ks probabilities , bounded by the minimum separation ( the vertical line ) . since the mass for the active bh is @xmath125 in our sample
, these results suggest that the companion inactive bh must be several times more massive than the active bh to account for the observed acceleration distribution .
in addition , we require a high binary fraction for the companion mass range probed ( @xmath164 ) and the companion bh must be orbiting not far outside the blr .
the above conclusions are based on the assumption that all the excess variance in acceleration we see is from binary motion .
however , as argued below , apparent acceleration between two epochs can also be induced by blr variability in single bhs .
if some , or most , of the observed excess variance is caused by broad line variability , the constraints on the bbh population will be significantly weakened . in the extreme case where
the observed acceleration distribution can be entirely explained by intrinsic broad line variability and measurement errors , our method only provides upper limits to the inactive bh mass and binary fraction , and lower limits to the separation . to illustrate this , we take the observed distribution as is , and randomly add intrinsic acceleration due to a bbh population as we did before , and see when the simulated distribution becomes significantly different from the observed one .
[ fig : ks_prob2 ] shows the resulting ks contours for two values of companion bh mass , @xmath165 and @xmath166 , where @xmath144 is the mass for the active bh in our sample .
since all the variance has already been accounted for , we no longer require a bbh population to contribute to the excess variance in the observed accelerations .
as a result , there is essentially no constraint on the allowable binary parameters when the companion bh is less massive than @xmath30 ( left panel ) . on the other hand , for more massive companion masses ,
certain regions in the two - dimensional parameter space are ruled out , as the resulting accelerations will exceed the observed variance ( right panel ) .
the constraints become stronger for more massive companion bhs . with a companion mass @xmath167 ,
the scenario of a @xmath168 binary fraction with separations @xmath169 can be ruled out at @xmath170 confidence .
the real situation is probably in between the two extreme cases considered in figs .
[ fig : ks_prob ] and [ fig : ks_prob2 ] , with the excess acceleration due both to a bbh population and to broad line variability in single bhs .
future spectroscopic surveys may provide the prior information on the ( stochastic ) variance due to the latter effect , thus improving the statistical constraints on the bbh population with our method ( see discussion in [ sec : future ] ) .
we show the 30 detections ( 28 unique systems ) in figs .
[ fig : det_obj1]-[fig : det_obj3 ] , where we plot the two - epoch spectra around the broad h@xmath1 line and the [ oiii]@xmath205007 line in velocity space .
we also show the h@xmath79 or line of these objects , if it is covered in sdss spectra and the line is detected at @xmath118 .
we follow the same @xmath96 cross - correlation approach to measure the velocity shift between two epochs based on or h@xmath79 .
but we note that the inferred velocity shift from or h@xmath79 should be interpreted with caution , for the reasons discussed in
[ sec : data ] .
the wavelength range for the @xmath96 cross - correlation is @xmath171 $ ] for , and @xmath172 $ ] for h@xmath79 . for
h@xmath79 we carefully subtracted the narrow line components , fixing the relative flux ratios of all narrow lines at both epochs .
in contrast , for , we use the full line flux ( broad and narrow lines ) in the @xmath96 cross - correlation , because the narrow component can not be reliably constrained . in fig .
[ fig : mvir_d ] we show the distribution of the detected quasars in the bh 1 mass and binary orbit separation plane , along with other samples defined earlier . again
, we use the virial bh mass estimates for the active bh 1 mass and we caution that there are large uncertainties in these mass estimates .
we also use the measured blr sizes for the binary separation . as discussed in
[ sec : prelim ] and [ sec : general_pop ] , the orbit separation of the hypothetical binary must be larger than the blr size , i.e. , all points in fig . [
fig : mvir_d ] would move up if the true binary separations were known .
the cyan dashed lines indicate constant maximum acceleration , as defined in eqn .
( [ eqn : v_a ] ) , where we assume a constant binary mass ratio @xmath173 .
the magenta dotted lines indicate constant orbital decay time due to gravitational radiation in a circular binary with a mass ratio of @xmath173 @xcite @xmath174 therefore most quasars in our sample are in the regime where the orbit decays from gravitational radiation in less than a hubble time ( if the virial mass estimate is correct and the orbital separation is not much bigger than the size of the blr ) . obviously , we expect binary bhs to be rare in the lower right region of the diagram , where the orbital decay time is much less than the hubble time . .
we use the virial bh mass estimates for the active bh 1 mass .
we also use the blr size estimates for the binary separation , and the real binary separations are certainly larger , possibly by a large factor .
the cyan dashed lines indicate constant maximum acceleration as in eqn .
( [ eqn : v_a ] ) , assuming equal - mass binaries , @xmath173 .
the magenta dotted lines indicate constant orbital decay time due to gravitational radiation , @xmath175 , for circular binaries and a mass ratio @xmath173 , as defined in eqn .
( [ eqn : tgr ] ) . ,
scaledwidth=48.0% ] the biggest concern with our detections is that the broad line profile could change over multi - year timescales , which may lead to an artificial line shift with the @xmath96 cross - correlation method . while visual inspection can lend some confidence to the interpretation as a genuine shift rather than a profile change ( or vice versa ) , the correct interpretation is still ambiguous in some cases .
below we give an object - by - object discussion of individual detections , and comment on the robustness of the detection , as well as possible causes of the detected velocity shift .
to guide our discussion we divide these objects into three categories : 1 ) binary candidates ; 2 ) broad h@xmath1 line variability , and 3 ) ambiguous cases .
these categories are our best efforts to assign the `` most likely '' scenario to individual detections , and are not a rigorous classification of these systems .
a useful constraint on the expected acceleration from a bbh can be derived using the equations in [ sec : prelim ] : @xmath176 is given by eqn .
( [ eqn : roche ] ) .
the product @xmath177 is a monotonic increasing function of @xmath178 for @xmath179 .
* [email protected] * for this object consistent velocity shifts between two epochs are detected for both h@xmath1 and over 261 days , despite the apparently noisy broad line spectrum for .
the shape of the broad h@xmath1 is not significantly different between the two epochs .
the inferred acceleration is @xmath181 ( @xmath158 ) .
* [email protected] * the s / n for this object is very high for both epochs .
consistent velocity shifts are detected for both h@xmath1 and h@xmath79 over @xmath182 yrs .
there is a slight change in the broad line profile between the two epochs .
the inferred acceleration is @xmath183 ( @xmath158 ) .
* [email protected] * for this object both h@xmath1 and show consistent velocity shifts over a time period of @xmath185 yrs . the change in line shape is insignificant , although there is some hint that the red - side flux changes a little between the two epochs . in the binary scenario ,
the inferred radial acceleration based on h@xmath1 is @xmath186 ( @xmath158 ) .
* [email protected] * for this object we detected a velocity shift for h@xmath1 but not for
. however , the limits we obtained from are fully consistent with the velocity shift detected in h@xmath1 . upon visual inspection ( see fig . [
fig : det_obj2 ] ) , both h@xmath1 and show similar profile changes between the two epochs , with extra flux on the red - side in epoch 1 .
therefore we classify this object as a good bbh candidate .
the inferred radial acceleration based on h@xmath1 is @xmath186 ( @xmath158 ) .
* [email protected] * for this object consistent velocity shifts between two epochs are detected for both h@xmath1 and h@xmath79 , over @xmath13 yrs .
the broad line profile also changed dramatically between the two epochs : the line is substantially narrower , and the shape is more symmetric , in epoch 2 than in epoch 1 .
it is possible that there is only a portion of the broad line emission moving along with bh 1 in the hypothetical binary system , and the remaining broad line emission comes from a circumbinary region , which does not accelerate between the two epochs .
it is also possible that some of the broad line emission comes from bh 2 , and bh 2 is much more massive than bh 1 .
the inferred acceleration is @xmath187 ( @xmath158 ) . * [email protected] * for this object we detected the largest velocity shift in our sample , for both h@xmath1 and h@xmath79 , over @xmath188 yrs .
interestingly , the velocity shift for h@xmath1 is larger than that for h@xmath79 ( significant at the @xmath189 level ) . at face value
, this may argue against the binary hypothesis if the h@xmath1 and h@xmath79 broad line emission comes from the same blr . however , if the h@xmath1 blr is mostly confined to the active bh , while the h@xmath79 blr also contains a circumbinary component ( which does not accelerate ) , a larger velocity shift in h@xmath1 is expected .
the width of the broad h@xmath1 is substantially larger than that of the broad h@xmath79 , which is consistent with this scenario .
the inferred acceleration based on h@xmath1 is then @xmath190 ( @xmath158 ) .
* [email protected] * for this object both h@xmath1 and h@xmath79 show consistent velocity shifts over a time period of @xmath191 yrs .
the shape of the broad line profile remains more or less constant with consistent fwhm at both epochs .
the narrow h@xmath1 subtraction is not perfect for both epochs , but does not affect the detected velocity shift .
the inferred radial acceleration is in the range of @xmath192 ( @xmath158 ) .
since the last spectroscopic observations of these candidates were a few years ago , taking a third - epoch spectrum now would be very helpful to confirm ( or rule out ) the proposed binary hypothesis .
spectra for the quasar sdss [email protected] at 9 epochs : the default epoch 1 ( the black line ) , and 8 different secondary epochs as listed in the legends .
the time difference in mjd from the epoch 1 spectrum and the measured pixel shifts ( @xmath158 ) are marked on the left and right in the figure .
the three spectra taken @xmath193 yr earlier than epoch 1 have different broad line shape , and for the first two of them we have detected a velocity shift from epoch 1 ( as shown in fig .
[ fig : det_obj1 ] ) .
however , the last epoch ( the gray line ) does not show the expected continuing velocity shift from the acceleration inferred from the earlier spectra .
thus the apparent velocity shifts between the first two spectra and the reference epoch 1 are likely due to broad line variability rather than binary motion . _
bottom : _ measured velocity shifts between different epochs and the reference epoch 1 , plotted against the restframe time separation .
a straight line is expected from constant acceleration within such short time intervals .
, title="fig:",scaledwidth=48.0% ] spectra for the quasar sdss [email protected] at 9 epochs : the default epoch 1 ( the black line ) , and 8 different secondary epochs as listed in the legends .
the time difference in mjd from the epoch 1 spectrum and the measured pixel shifts ( @xmath158 ) are marked on the left and right in the figure .
the three spectra taken @xmath193 yr earlier than epoch 1 have different broad line shape , and for the first two of them we have detected a velocity shift from epoch 1 ( as shown in fig . [ fig : det_obj1 ] ) .
however , the last epoch ( the gray line ) does not show the expected continuing velocity shift from the acceleration inferred from the earlier spectra .
thus the apparent velocity shifts between the first two spectra and the reference epoch 1 are likely due to broad line variability rather than binary motion . _
bottom : _ measured velocity shifts between different epochs and the reference epoch 1 , plotted against the restframe time separation .
a straight line is expected from constant acceleration within such short time intervals . , title="fig:",scaledwidth=48.0% ] * [email protected]
* for this object we have 8 different secondary epochs to cross - correlate with the default epoch 1 ; all are in our good sample with good measurement precision . in fig .
[ fig:0306 ] ( top ) we show the broad line spectra at different epochs , along with the measured velocity shift with respect to the reference epoch 1 .
we see a narrowing of the line profile from the first three spectra taken @xmath194 months earlier than epoch 1 , then the line profile stays more or less constant . in the bottom panel
we show the measured radial velocity curve , again using epoch 1 as reference .
the time baseline is only @xmath195 yr in the quasar restframe .
the expected binary period is @xmath196 yr given the constraints of the binary separation ( from the blr size ) and the virial bh mass estimate , where @xmath50 and @xmath197 is a function of @xmath178 as defined in eqn .
( [ eqn : roche ] ) .
for @xmath198 we have @xmath199 hence @xmath200 yr . thus within such short time interval the expected bbh velocity should approximately have constant acceleration .
if this is a binary system with a los acceleration inferred from the observations up to and including epoch 1 , we should expect to see a velocity shift within @xmath201 $ ] ( pixels ) between the last epoch ( gray spectrum / point in fig .
[ fig:0306 ] ) and epoch 1 .
while this is still consistent with what we measure , the fact that the last epoch has the same line profile as epoch 1 while the first two epochs have very different profile strongly suggest that the detected velocity shift is more likely due to broad line variability .
* [email protected] * this object is an unambiguous disk emitter with double - peaked broad line profiles .
while we detected velocity shifts between the three epochs we have in the good sample , it is hard to ascribe this average velocity shift to binary motion .
it appears that the relative strength of the two peaks changed over this period ( seen in both h@xmath1 and h@xmath79 ) .
variability of this kind has been seen in long - term monitoring of disk emitters ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , which usually invokes a non - binary interpretation ( but see , e.g. , * ? ? ?
* for an alternative bbh interpretation ) .
we therefore classify this object as exhibiting broad line variability , although the bbh scenario can not be completely ruled out given only three epochs .
similar objects were reported in @xcite .
* [email protected] and [email protected] * these two objects are detections with the shortest time separation ( @xmath65 days ) between the two epochs , which implies rather large accelerations . using eqn .
( [ eqn : amax ] ) and assuming the blr sizes are accurate , it is difficult to get such large acceleration unless the bh mass @xmath30 is @xmath194 times larger than the virial bh mass estimate ( see table [ table : detection ] ) and the unseen bh is at least as massive for both objects . while this is not entirely ruled out , it is more likely that the velocity shifts detected for these two objects are due to short - term broad line variability .
in addition , for j0747 the velocity shift based on h@xmath1 is _ inconsistent _ with that based on h@xmath79 , which also favors the blr variability scenario . we classify the remaining detections as ambiguous cases .
this category includes more than half of our detections , which reflects the general ambiguity in interpreting the observed broad line velocity shifts .
they have neither a second broad line ( or h@xmath79 ) showing consistent velocity shifts , nor more than two epochs with sufficient quality to rule out the hypothetical binary acceleration .
thus , they could still be binary candidates , although the binary probability is perhaps lower than those presented in [ sec : class1 ] .
additional high - quality spectroscopic observations are needed to further investigate these systems .
the basic assumption required to utilize the broad line shift as an indicator for the bbh orbital motion is that the blr must move along with its associated bh .
this means this method is only applicable to bbhs with separations @xmath202 ( see [ sec : prelim ] and * ? ? ?
fortunately , the typical blr size for our sample is @xmath128 pc , so this method is still probing the interesting sub - pc regime where presumably the bbh has passed the final parsec barrier .
the primary uncertainty is that the blr is unlikely to be in a steady state , and changes in the blr may produce artificial broad line velocity shifts .
velocity - resolved reverberation mapping results for a handful of objects have shown diverse kinematics of the blr gas ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) involving inflows and outflows .
changes of the kinematic structure of the blr will occur on the dynamical time @xmath203 yr , i.e. , the same timescale over which we search for the bbh signature .
broad line shifts without line shape changes are perhaps better explained by the bbh scenario , but the broad line shape may vary even in a bbh system . occasionally an apparent velocity shift might be induced due to excess flux on one side of the broad line .
this may indicate a blr variability origin rather than a bbh ; however , the bbh scenario can not be completely ruled out if the blr is composed of spatially distinct components , and the `` excess flux '' is indeed associated with the motion of one bh . in a similar spirit ,
the broad line profile could also change in the bbh scenario , if both bhs are active and the broad line is composed of two spectrally unresolved components ( e.g. , * ? ? ?
invoking multiple broad lines sometimes helps confirm the detected velocity shift , but can not be used to confirm or rule out the bbh scenario given the possible differences in the blrs for different lines with various ionization potentials .
it is practically impossible to distinguish between the bbh case and the blr variability case for the observed broad line velocity shifts with only two epochs .
the best ( perhaps the only ) way to differentiate these two cases is continued spectral monitoring of the broad line .
blr variability will lead to stochastic velocity shifts , while the bbh scenario will lead to coherent velocity shifts , either keplerian ( over long timescales comparable to the binary orbit period ) or linear ( over timescales much shorter than the binary orbit period ) .
of course , one also needs high s / n to make the measurements meaningful for this purpose .
currently there is no adequate spectroscopic monitoring data to provide priors on the stochastic velocity shifts induced by blr variability , but future time - domain spectroscopic surveys may provide such information ( see [ sec : future ] ) . finally , although we have focused on circular binary orbits , the methodology developed in this work can be easily applied to a binary population with a distribution of eccentricities .
such an application will be appropriate once we have a good understanding of the velocity shifts due to broad line variability . throughout this work we have used blr sizes estimated from the blr @xmath31 relation found in reverberation mapping studies ( e.g. , * ? ? ?
this is of course an approximation , as the blr size measured in reverberation mapping represents the emissivity - weighted average radius of the blr , and the actual blr could extend beyond this typical size .
it is nevertheless reasonable to expect that the bulk of the line emission comes from within this typical blr size . if we have significantly underestimated the blr sizes , the allowable binary parameter space shown in fig .
[ fig : ks_prob ] will diminish , because then the allowable binary orbit separations ( i.e. , @xmath162 ) will be too large to account for the observed excess variance in broad line velocity shifts ; in this case the observed excess variance must be due to blr variability . an alternative way to estimate
the blr size is to use theoretical predictions based on photoionization calculations ( e.g. , * ? ? ?
* ) : @xmath204 , where @xmath205 is the gravitational radius of the active bh . in our opinion ,
such theoretical estimates are more uncertain than the blr sizes derived empirically from reverberation mapping .
in addition , such estimates have an explicit dependence on the highly uncertain bh mass estimates ( e.g. , * ? ? ? * ) .
despite these concerns we have constructed models using @xmath206 .
following our earlier discussion in [ sec : general_pop ] , we assume a constant mass of @xmath207 for bh 1 , and use the new blr size estimates ( which are then the same for all quasars in our sample ) to repeat the analysis in fig . [
fig : ks_prob ] .
the new blr size is @xmath208pc , larger than our fiducial blr size estimates ( with a median value of @xmath209pc ) by almost a factor of 3 .
therefore the allowable binary parameter space as shown in fig .
[ fig : ks_prob ] disappears , and blr variability is needed to explain the observed excess variance in broad line velocity shifts .
most theoretical work on bbhs with a gaseous disk does not make specific predictions for the broad lines ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) with few exceptions ( e.g. , * ? ? ?
in addition , their models often involve a circumbinary accretion disk , which presumably complicates the applicability of the broad line diagnostics used in this work , if this circumbinary disk also contributes to the broad line emission .
using sph simulation coupled with photoionization calculations , @xcite studied the temporal behavior of the broad h@xmath79 line from the circumbinary accretion disk .
the line profile is often double - peaked , showing the characteristics of a disk , and evolves with time , similar to the results in @xcite with toy models for the blr .
however , it is not yet feasible to make detailed comparisons between the simulation predictions and observations of quasars with double - peaked broad line profiles . @xcite
considered a bbh scenario where the blr has a stratified structure for different line species with the higher - ionization lines deeper within the gravitational potential of the bh .
tidal erosion from the companion bh then reduces the flux ratio of the lower - ionization lines to higher - ionization lines .
while this process may be at work for very close bbhs , it is difficult to differentiate this scenario from single bhs with abnormal broad line flux ratios , thus difficult to confirm the bbh nature with this feature alone .
finally , recent predictions of sub - pc bbh fractions in quasars from hierarchical cosmological merger models ( e.g. , * ? ? ?
* ; * ? ? ?
* ) still have large uncertainties , and a clear pathway connecting these predictions to direct observables is currently absent .
clearly more theoretical work is needed to probe the full potential of various broad line diagnostics , and to provide more theoretical guidance on the expected binary frequency .
there are three recent observational studies that are closely related to the current work . @xcite and
@xcite both obtained second - epoch spectra for samples ( @xmath210 and @xmath211 objects , respectively ) of @xmath212 sdss quasars selected to have peculiar broad line characteristics , such as large offsets ( @xmath213 ) between the broad h@xmath1 and narrow lines , and/or double - peaked or asymmetric line profiles .
they used the two spectra ( one from the sdss ) to measure broad h@xmath1 velocity shifts between the two epochs and reported several detections .
the detection fraction in their samples is roughly consistent with ours if we restrict to the same restframe time separation ( see fig . [
fig : hbeta_hist ] ) .
but we caution that our statistics are poor ( we only have 7 detections for the largest @xmath42 bin ) , and there are considerable differences in the sample and the quality of the spectroscopic data , in the approaches to measuring the velocity shift and in the associated error analysis among the three studies .
the most significant difference is that our sample is typical of normal quasars , while theirs were selected to be more likely to host a bbh .
therefore they could not provide constraints on the general population as we do in [ sec : general_pop ] .
a more appropriate comparison will be provided in paper ii , where we also target quasars selected to be more likely bbhs .
the third study , by @xcite , also used repeated spectroscopy from the sdss , and focused on the general quasar population , as we do here .
however , there are several key differences between the two studies : ( 1 ) their sample probes quasars at @xmath214 ( although their reported detections are mostly at @xmath215 ) while our sample is at @xmath0 and has lower quasar luminosities ; ( 2 ) they used the line instead of h@xmath1
. we did not choose the line as our primary line for the reasons discussed in
[ sec : data ] ; ( 3 ) they used the several `` detections '' to constrain bbh models , while we believe it is more informative , and more appropriate , to use the full distribution in constraining the general bbh population ; and ( 4 ) they did not utilize individual measurement errors for their sample , and used a typical ensemble value for the measurement errors instead . this could be problematic , as the several `` detections '' in their study could be due to large individual measurement errors , and their modeling is sensitive to this detected number .
all these issues may explain the different conclusions ( in particular , ju et al .
find a much lower detection fraction ) reached in the two studies , but one should also bear in mind the different assumptions that went into both analyses .
we defer a more careful comparison to future work when we perform our own analysis on the line .
we have demonstrated the potential of using the broad line diagnostics , i.e. , velocity shifts between epochs , to identify and characterize the bbh population in broad line quasars .
however , to improve the power of this technique , there is both observational and theoretical need to understand the broad line emission in single and binary bhs . from an observational point of view , we need long - term spectroscopic monitoring of large samples of quasars to construct the power spectrum of the stochastic velocity shifts due to blr variability in single bhs .
analogous to studying the photometric quasar continuum variability , we can construct a stochastic velocity shift `` structure function '' , i.e. , the random fluctuation of broad line shifts between epochs as a function of time separation .
this must be done for a large sample to capture the diversity in the quasar population , and must be done over multi - year timescales and many epochs for the same objects , and with adequate s / n , to confirm the stochastic nature of the velocity shifts .
a by - product of these programs would be potential discoveries of coherent velocity shifts , which would be the best candidates for bbhs .
there are also general benefits from improved spectral quality , increased time baseline and more epochs for our statistical constraints .
for instance , merely increasing the time baseline can shrink the measurement errors in acceleration .
parallel observational efforts are needed to better understand the blr structure with reverberation mapping . from a theoretical perspective
, we need bbh simulations with recipes for the blr gas to predict the temporal evolution of the broad line emission and its correlation with the orbital motion of the bbh , as well as the effects of the ( postulated ) circumbinary disk on the broad lines ( e.g. , how much broad line emission comes from this circumbinary disk ) , with empirical constraints from observations , such as the blr size , geometry and kinematics .
we have performed a systematic search for sub - parsec bbh candidates in optically selected , broad line quasars with multi - epoch ( mostly two - epoch ) sdss spectroscopy .
the spectra are separated by up to several years in the quasar restframe .
our sample is composed of normal quasars , which is different from the samples in other studies ( e.g. , * ? ? ?
* ; * ? ? ?
we have shown that targeting these normal quasars is not only as efficient as targeting offset or double - peaked broad line quasars in terms of finding bbh candidates , but also provides much more robust constraints on the bbh population in quasars .
our working hypothesis was that only one bh is active and that this bh dynamically dominates its own blr , and that the companion bh is orbiting at a large enough distance such that the blr is not tidally distorted or disrupted .
our sample is composed of @xmath0 quasars , with typical virial bh mass estimates @xmath125 ( for the active bh ) and blr sizes @xmath216 pc .
we searched for the expected acceleration ( velocity shift ) in the broad h@xmath1 line from the binary motion between the two epochs , using a cross - correlation technique .
our main findings and their implications are summarized below : * the broad h@xmath1 line shape ( fwhm , skewness , etc ) can change over multi - year timescales , but dramatic changes in fwhm ( @xmath217 ) are rare .
* out of @xmath218 pairs of observations for which we have successfully measured the velocity shift between two epochs , we detected 28 systems with significant acceleration at the @xmath106 confidence level .
object - by - object inspections suggest that 7 of them are the best examples of sub - parsec bbh candidates , 4 of them are best explained by broad line variability in single bhs , and the rest are ambiguous . *
the full distribution of acceleration ( mostly non - detections ) for a subset from the 688 pairs with better - quality measurements and @xmath219 yr ( restframe ) was used to constrain the bbh population among general quasars . after accounting for measurement errors , excess variance
is required to explain the observed distribution .
if all the excess variance in velocity shifts is due to a bbh population , then ( 1 ) a large binary fraction is required ; ( 2 ) the companion bh must be orbiting within a few times the blr radius ( details depending on the companion mass , see [ sec : general_pop ] ) ; ( 3 ) the companion bh must be more massive than the active one .
on the other hand , if the excess variance is largely due to stochastic blr variability in single bhs , then the requirement for a bbh population is substantially weakened , and even disfavored in the case of a very massive companion
otherwise the predicted acceleration distribution will be broader than observed ( after convolving with errors ) .
* long - term spectroscopic monitoring of the detected bbh candidates with more epochs are required to confirm or rule out the bbh hypothesis .
in addition , long - term spectroscopic monitoring of normal quasars will provide vital information on the stochastic velocity shift induced by blr variability , and thus improve statistical constraints on the bbh population in broad line quasars with the methodology developed in this paper .
we thank the anonymous referee and mike eracleous for useful comments on the manuscript .
support for the work of y.s . and
was provided by nasa through hubble fellowship grant number hst - hf-51314.01 and hst - hf-51307.01 , respectively , 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 .
this work was supported in part by nsf grant ast-0907890 and nasa grant nna09db30a ( a.l . ) , and nasa grant nnx11af29 g ( s.t . )
funding for the sdss and sdss - ii has been provided by the alfred p. sloan foundation , the participating institutions , the national science foundation , the u.s .
department of energy , the national aeronautics and space administration , the japanese monbukagakusho , the max planck society , and the higher education funding council for england .
the sdss web site is http://www.sdss.org/. | we perform a systematic search for sub - parsec binary supermassive black holes ( bhs ) in normal broad line quasars at @xmath0 , using multi - epoch sdss spectroscopy of the broad h@xmath1 line .
our working model is that ( 1 ) one and only one of the two bhs in the binary is active ; ( 2 ) the active bh dynamically dominates its own broad line region ( blr ) in the binary system , so that the mean velocity of the blr reflects the mean velocity of its host bh ; ( 3 ) the inactive companion bh is orbiting at a distance of a few @xmath2 , where @xmath3 pc is the blr size .
we search for the expected line - of - sight acceleration of the broad line velocity from binary orbital motion by cross - correlating sdss spectra from two epochs separated by up to several years in the quasar restframe . out of @xmath4
pairs of spectra for which we have good measurements of the velocity shift between two epochs ( @xmath5 error @xmath6 ) , we detect @xmath7 systems with significant velocity shifts in broad h@xmath1 , among which seven are the best candidates for the hypothesized binaries , four are most likely due to broad line variability in single bhs , and the rest are ambiguous
. continued spectroscopic observations of these candidates will easily strengthen or disprove these claims .
we use the distribution of the observed accelerations ( mostly non - detections ) to place constraints on the abundance of such binary systems among the general quasar population .
excess variance in the velocity shift is inferred for observations separated by longer than @xmath8 yr ( quasar restframe ) .
attributing all the excess to binary motion would imply that most of the quasars in this sample must be in binaries ; that the inactive bh must be on average more massive than the active one ; and that the binary separation is at most a few times the size of the blr .
however , if this excess variance is partly or largely due to long - term broad line variability , the requirement of a large population of close binaries is much weakened or even disfavored for massive companions . future time - domain spectroscopic surveys of normal quasars can provide the vital prior information on the structure function of stochastic velocity shifts induced by broad line variability in single bhs .
such surveys with improved spectral quality , increased time baseline , and more epochs , can greatly improve the statistical constraints of this method on the general binary population in broad line quasars , further shrink the allowed binary parameter space , and detect true sub - parsec binaries . |
the hubbard model is a paradigm for one of the most difficult quantum manybody systems ; strong interactions competing with the kinetic energy in a dense electron gas .
these problems must be explored either analytically by relatively poorly controlled approximate methods or by methods such as monte - carlo , dynamical mean field theory ( dmft ) or finite size calculations .
a major difficulty is the absence of a framework for such systems that competes in simplicity and rigor with conventional landau fermi - liquid theory .
a particularly intriguing scenario for non - fermi liquid behavior is `` spin - charge separation '' with a low - energy effective theory where spin and charge excitations are decoupled .
this idea has attracted widespread attention because it migh provide a route to high - temperature superconductivity in a system with strong local repulsion . in 1-d
it is known through the use of `` bosonization '' that spin - charge separatation does occur at low energy.@xcite in higher dimensions , much less is known .
the usual approach to spin charge separation is to construct electron operators as composite objects consisting of a charge part , the `` holon '' , and a spin part , the `` spinon '' . to avoid unphysical states contraints have to be included
which are then treated by approximate methods.@xcite here we shall construct electron operators as composites of charge - like and spin - like operators that do not give rise to any unphysical states and thus avoids introducing any constraints .
an exact local mapping which is valid in all dimensions will be shown to exist from a purely fermionic manybody system to an equivalent system of fermions and spin - like operators obeying bosonic commutation relations between different lattice sites .
we shall explore this with the 2-d hubbard model , for which the resulting mapping bears a resemblance to the mappings of the local hilbert space achieved by introducing spinons and holons .
let @xmath1 be the particle number at site @xmath2 , and @xmath3 and @xmath4 be respectively the number of up and down electrons : @xmath5 .
since the hubbard interaction obeys @xmath6 , by choosing the chemical potential appropriately we can replace the usual hubbard interaction with the particle - hole symmetric interaction @xmath7 .
we thus write the hubbard model in the particle - hole symmetric form @xmath8 where the sum is over all sites and nearest neighbor bonds counted once .
we now demonstrate an exact transformation of the local operator algebra of the hubbard model to new fermionic @xmath9 `` quasicharge '' @xmath10 and @xmath11 `` quasispin '' operators @xmath12 that obey , respectively , fermi and bose statistics .
the operators are given by @xmath13 where @xmath14 is @xmath15 depending on which sublattice site @xmath2 is on .
we define the @xmath16 and @xmath17 component of quasispin by @xmath18 and @xmath19 .
the factor appearing as @xmath20 can in fact be chosen as an arbitrary phase , but is fixed here to preserve symmetries specific to the hubbard model .
this nonlinear
transformation generalizes earlier work on transformations that were strictly canonical , and hence did not alter commutation relations of the fermi operators.@xcite the new operators obey the following algebra , which can be verified by straightforward manipulations : @xmath21 , @xmath22 , @xmath23 = 0 $ ] , @xmath24 = i \delta_{r{{r ' } } } \sum_{k } \epsilon_{ijk } { q^{k}_{r } } $ ] we define the quasicharge operator as @xmath29 .
we find the following relations with the usual operators expressed in terms of ordinary fermions : @xmath30 , @xmath31 and @xmath32 where @xmath33 , with @xmath34 the pauli matrices .
we also define the local pseudospin
operators @xmath35 which are the generators of the su(2 ) algebra which correponds to rotations between the empty and doubly occupied states.@xcite the formal resemblance to spin symmetry has led to the name pseudospin
. the total z - component of pseudospin can therefore be seen to be half the number of doubly occupied sites minus the number of empty sites , which is precisely the charge relative to half filling .
pseudospin is a symmetry of the particle - hole symmetric hubbard model but is broken by a chemical potential .
we note that , aside from phase factors , the quasicharge operators always create even charged states from odd and vice versa , as they must since they are fermionic .
however , there is a precise phase relationship which is not so easily interpreted that is enforced by the commutation relations and symmetries .
our hubbard model can now be rewritten _
exactly _ in terms of the quasiparticle operators as @xmath36 with @xmath37 and @xmath38 and @xmath39 . in this representation ,
the symmetry under the entire @xmath11 quasispin rotation is manifest and we note that the hubbard interaction becomes simply the quasicharge number operator .
we define total quasispin @xmath40 , spin @xmath41 and pseudospin @xmath42 respectively by @xmath43 etc .
@xmath44 and @xmath45 all obey the su(2 ) algebra @xmath46 = i \sum_{k } \epsilon_{ijk } p_k \ ; $ ] etc .
we see from the definitions above eq .
[ eq : newhub ] the relationship , @xmath47 and @xmath31 .
thus @xmath48 , i.e. quasispin can be exactly split into pseudospin and spin , with @xmath49 and @xmath50 providing the projection operator of @xmath51 into either spin or pseudospin . since @xmath51 commutes with @xmath26 and the hubbard hamiltonian @xmath52 depends on @xmath53 only through rotationally invariant terms , we conclude that @xmath54 = 0 $ ] .
we further know that the hubbard model commutes with ordinary spin whereby @xmath55 = 0 $ ] .
we can therefore conclude that @xmath56 = 0 $ ] , which confirms the well known invariance under pseudospin rotations .
the operators @xmath57 are identical to the operators defined in ref .
@xcite and used in a perturbation expansion in @xmath58 .
it is clear from their definitions that @xmath59 couples different hubbard bands defined by the sectors given by different values of @xmath60 whereas @xmath61 does not .
we observe @xmath62 = 0 $ ] .
it can be inferred that @xmath63 = 0 $ ] since @xmath64 is known to commute with ordinary spin .
we now define the usual canonical transformation @xmath65 which splits the hamiltonian into pieces that preserve total quasicharge .
we define @xmath66 $ ] and it straightforward to show that @xmath67 = ( { { t_{-1}}}+ { { t_1 } } ) $ ] , whereby it follows that @xmath68 @xmath69 @xmath70 / u - $ ] @xmath71 / ( 2 u^2 ) + ... $ ] @xmath72 splits into two terms . the first term @xmath73 is the ordinary spin interaction in the `` spin - only '' sector @xmath74 where @xmath75 and the second term @xmath76 couples quasispin and quasicharge @xmath77 the last summation runs over all three - site neighbors connected by links . terms of order @xmath78 which are already present to order @xmath79 or @xmath0 are ignored .
according to the results of ref .
@xcite , to all orders in @xmath80 the operator @xmath65 which diagonalizes total quasicharge is a polynomial of @xmath81 , and hence to all orders in perturbation theory , the transformation by @xmath82 will preserve all the symmetries @xmath83 and @xmath84 in this formulation .
under this canonical transformation the eigenstates of @xmath85 will map to the eigenstates of the hubbard model represented by the bare quasiparticle operators through the transformation @xmath86 .
we let @xmath87 be eigenstates of @xmath85 and @xmath88 .
adapting ref .
@xcite to our problem , there will be a `` zeroth hubbard band '' or `` spin - only sector '' of solutions corresponding to @xmath89 being identically zero so that @xmath85 reduces exactly to @xmath90 .
we conclude @xmath91 is identically zero and the effective hamiltonian of the zeroth subband to order @xmath92 is exactly the heisenberg model .
what can we conclude from symmetries about the properties of the zeroth hubbard band ?
it must be remembered that @xmath93 is not a `` good quantum number '' i.e. is neither invariant under the symmetries nor under @xmath94 .
we have therefore no information about @xmath95 from symmetry arguments alone .
we see that @xmath96 throughout the entire lattice in the entire zeroth hubbard subband
. we can therefore conclude that @xmath97 .
since @xmath98 commutes with @xmath65 , this assertion survives to all orders in perturbation theory and we can conclude that @xmath99 in fact quasispin and ordinary spin coincide . since @xmath100 is equal to physical charge , we see that the spin - only sector has indeed charge density corresponding to half filling . the pseudospin is given by @xmath101 if there is exactly one quasicharge , we conclude from the fact that @xmath102 that @xmath103 . we have thus identified the lowest nonzero quasicharge sector with the spin half representation of pseudospin , exactly what we need to associate it with one added positive or negative physical charge .
we can derive a set of extended charge states in the dilute limit by a mean field treatment of eq .
[ eq : pert ] , replacing @xmath104 . in two dimensions ,
these have the spectrum @xmath105 where @xmath106 and where @xmath107 , @xmath108 and @xmath109 denote nearest neighbor and longer range heisenberg correlations . using the numerical value @xmath110 @xcite
we find @xmath111 .
we find a quasicharge spectrum bounded by @xmath112 and shown in the shaded region of fig .
[ fig : fig2 ] .
the present calculation makes the `` nagaoka theorem '' natural.@xcite consider eq .
[ eq : newhub ] in the case where the quasispins are perfectly aligned along the quasispin z axis . in this case
@xmath113 , the operator @xmath114 will be identically zero and the kinetic energy @xmath115 becomes precisely @xmath116 .
these are quasicharges moving independently of the ferromagnetic quasispins and are ordinary holes in a perfect 2-d ferromagnet which the kinetic energy @xmath117 in the thermodynamic limit .
the nagaoka theorem states that as @xmath118 a single charge causes the ground state of @xmath52 to become ferromagnetic . this nagaoka ferromagnet with charge one has hubbard energy @xmath119 , whereas the antiferromagnetic solution with a single quasicharge has energy @xmath120 where @xmath121 is the number of lattice sites in the system .
we therefore find that @xmath122 when the inequality @xmath123 holds .
we see that whether or not there is nagaoka ferromagnetism depends on which order we take @xmath0 and @xmath121 to infinity . the argument can be used to estimate when the antiferromagnet becomes unstable to ferromagnetic ordering .
( see also ref .
@xcite ) . let us consider a quasicharge density of @xmath124 .
we see that if @xmath125 obeys @xmath126 the dilute ferromagnet will be energetically favored over a collection of extended quasicharge states , essentially reproducing the argument by ioffe and larkin.@xcite using the previous ideas , we shall study localized states in a 2-d antiferromagnet .
as we have seen , a quasicharge hops freely on a ferromagnetic bond where @xmath113 . in a mean field treatment of the quasispins , to linear order in @xmath127 a quasicharge can not hop an a bond with antialigned quasispins since @xmath128 .
starting with a patch of locally neel ordered quasispins oriented along @xmath129 we consider a quasispin state with nearby spins flipped relative to the neel state on diagonally adjacent sites of the spin - down sublattice .
we note that in the presence of a quasispin ferromagnet oriented along @xmath130 , quasicharge corresponds to physical charge measured from half filling
. the core of the region will thus be a ferromagnet surrounded by a closed boundary of antialigned spins . outside this patch , quasispins are allowed to relax , so that far away the system will be a heisenberg antiferromagnet .
the two smallest such configurations are shown in fig . [
fig : fig1 ] .
we identify the quasispin of the state by constructing the neel state with the same rotational symmetry around the cluster . the smallest cluster ( fig.[fig : fig1],a )
has one flipped spin thus has @xmath131 whereas the cluster with two flipped spins ( fig .
[ fig : fig1],b ) has @xmath132 .
we now add @xmath133 quasicharges to the cluster . according to the definition of pseudospin , in the case
where quasicharge is confined to a quasiferromagnetic region with @xmath134 , it can be shown that @xmath135 .
a cluster with @xmath133 units of quasicharge will thus have pseudospin @xmath136 and , from the identity @xmath137 , physical spin @xmath138 .
the clusters with no quasicharge have pseudospin zero and physical spin equal to quasispin , whereas the charge one cluster will have pseudospin @xmath139 and spin @xmath140 .
the energies of these states are approximate but the spin and pseudospin quantum numbers are exact . as quasicharge is added in the cluster , quasispin , spin and pseudospin continue to be good quantum numbers . since quasispin is conserved upon adding quasicharge each such configuration has a well defined value of quasispin independent of physical spin or charge and can be calculated in the effective theory .
quasicharge , however , is not a good quantum number , and will change upon transforming back the the physical state with @xmath141 .
we estimate the energy by keeping only the part of the kinetic energy linear in @xmath79 and ignore the spin exchange energy beyond the antiferromagnetic boundary of the cluster . for the smallest @xmath131 cluster
we see sixteen bonds that are strongly affected by the localized charge .
four bonds are ferromagnetic with zero energy and twelve bonds are antiferromagnetic with energy @xmath142 , i.e. they retain about 85% of the energy of the unaffected neel state @xmath143 .
the energy of these states is compactly written with the spectroscopic - like notation @xmath144 where @xmath83 is the quasispin , @xmath145 is the charge relative to half filling and @xmath146 is the spin .
thus @xmath147 is the energy of the the smallest uncharged cluster measured with respect to the heisenberg ground state .
we find @xmath148 .
the five kinetic energy eigenvalues of this cluster are @xmath149 .
we thus find @xmath150 .
for the @xmath132 cluster we find a kinetic energy for a single charge @xmath151 ; there are 16 antiferromagnetic bonds and 8 ferromagnetic bonds giving @xmath152 .
the energy of these states are plotted in fig .
[ fig : fig2 ] and compared to the energy of the band of extended states from eq .
[ extended ] .
we find that for large @xmath153 the clusters with localized charge are energetically favored over the extended states . as @xmath0 becomes larger , larger clusters lower the energy in accordance with the nagaoka theorem .
we have demonstrated an exact local mapping from a fermionic many - body system to system of interacting bosons and fermions that does not introduce constraints .
the mapping provides a paradigm for a non - fermi liquid behavior for correlated fermions valid in any dimension .
applied to the 2-d hubbard model the transformation makes precise earlier attempts that relied on approximations and constraints to discuss spin and charge as composite operators .
our calculations supports the idea of low energy localized states where spin is bound to charge in the limit of large @xmath153
. 8 v.j .
emery,_theory of the one - dimensional electron gas _ , in `` highly coducting one - dimensional solids '' , edited by j.t .
devreese , r.p.evrard and v.e .
van doren , 327 ( plenum , new york , 1979 ) ; j. solyom , _ adv .
_ * 28 * , 201 - 303 ( 1979 ) .
p. coleman , _ phys .
b. _ * 29 , * 3035 , ( 1983 ) ; see , e. fradkin , _ field theories of condensed matter physics _ , ( addison - wesley , 1991 ) and references therein .
similar strictly canonical transformations that do not map between fermionic and bosonic operators have been discussed by s. stlund , e.j .
mele , phys .
b. * 44 * , 12413 ( 1991 ) ; s. stlund , t.h .
hansson , a. karlhede , phys .
b*71 * , 165121 ( 2005 ) after this work was completed , we have found out that that the transformation appears in unpublished work by j.i .
chandler and j.m.f .
gunn , cond - mat/9702009 .
h. shiba , _ prog .
. phys _ * 48 * , 2171 ( 1972 ) ; c. n. yang and s. c. zhang , mod .
b 4 , 759 ( 1990 ) ; shoucheng zhang , phys .
65 , 120 ( 1990 ) ; v. j. emery , phys .
b 14 , 2989 ( 1976 ) ; a.b . eriksson , t. einarsson and s. stlund , phys . rev . * b * 52 , 3662 ( 1995 ) y. nagaoka , _ solid state comm._*3 * , 409 ( 1965 ) ; phys . rev . * 147 * , 392 ( 1966 ) ; d.j .
thouless , _ proc .
* 86 * , 893 ( 1965 ) .
a.h . macdonald , s.m .
girvin and d. yoshioka , _ phys .
b _ * 37 * , 9753 ( 1988 ) ; j .- y.p .
delannoy , m.j.p .
gingras , p.c.w .
holdsworth and a .- m.s .
tremblay , cond - mat/0412033 .
e. manousakis , rev .
* 63 * 1 , ( 1991 ) .
visscher , phys .
b * 10 * , 943 ( 1974 ) ; l.b .
ioffe and a. i. larkin , phys .
b. * 37 * , 5730 ( 1988 ) . | we demonstrate an exact local transformation which maps a purely fermionic manybody system to a system of spinfull bosons and spinless fermions , demonstrating a possible path to a non - fermi liquid state . we apply this to the half - filled hubbard model and show how the transformation maps the ordinary spin half fermionic degrees of freedom exactly and without introducing hilbert space constraints to a charge - like `` quasicharge '' fermion and a spin - like `` quasispin '' boson while preserving all the symmetries of the model .
we present approximate solutions with localized charge which emerge naturally from the hubbard model in this form . our results strongly suggest that charge tends to remain localized for large values of the hubbard @xmath0 .
# 1 |
cosmological n - body simulations have led to impressive strides in our understanding of structure formation in universes dominated by collisionless dark matter .
the impact of such simulations has been greatest in the highly non - linear regime , where analytic calculations offer little guidance .
recently , and as a result of the development of efficient algorithms and of the advent of powerful , massively parallel computer arrays , it has been possible to apply n - body studies to detailed investigations of structure on small scales .
these studies can now probe scales comparable to the luminous radii of individual galaxies , thus enabling direct comparison between theory and observation in regions where luminous dynamical tracers are abundant and easiest to observe . predicting the structure of dark matter halos on kpc and sub - kpc scales , where it can be compared directly with observations of galactic dynamics , is one of the premier goals of n - body experiments , and there has been steady progress in this area over the past few years . building upon the early work of frenk et al ( 1988 ) , quinn , salmon & zurek ( 1986 ) , dubinski & carlberg ( 1991 ) and crone , evrard & richstone ( 1993 ) , navarro , frenk & white ( 1996 , 1997 , hereafter nfw ) found that , independent of mass and of the value of the cosmological parameters , the density profiles of dark matter halos formed in various hierarchical clustering cosmogonies were strikingly similar .
this ` universal ' structure can be characterized by a spherically - averaged density profile which differs substantially from the simple power - laws , @xmath2 , predicted by early theoretical studies ( gunn & gott 1972 , fillmore & goldreich 1984 , hoffmann & shaham 1985 , white & zaritsky 1992 ) .
the profile steepens monotonically with radius , with logarithmic slopes shallower than isothermal ( i.e. @xmath3 ) near the center , but steeper than isothermal ( @xmath4 ) in the outer regions .
nfw proposed a simple formula , @xmath5 which describes the density profile of any halo with only two parameters , a characteristic density contrast .
i express the present value of hubble s constant as @xmath6 km s@xmath7 mpc@xmath7 ] , @xmath9 , and a scale radius , @xmath10 .
note that there is no well defined value for the _ central _ density of the dark matter , which can in principle climb to arbitrarily large values near the center .
this is an important point , especially because there have been a number of reports in the literature arguing that the shape of the rotation curves of many disk galaxies rules out steeply divergent dark matter density profiles ( flores & primack 1994 , moore 1994 , mcgaugh et al 1998 , de blok et al 2001 ) .
the results of nfw have been confirmed by a number of subsequent studies ( see , e.g. , cole & lacey 1996 , huss , jain & steinmetz 1999 , jing & suto 2000 ) , although there is some disagreement about the innermost value of the logarithmic slope .
moore et al .
( 1998 ) , ghigna et al .
( 2000 ) , and fukushige & makino ( 1997 , 2001 ) have argued that density profiles diverge near the center with logarithmic slopes steeper than the asymptotic value of @xmath11 in nfw s formula . since steep inner slopes
are apparently disfavored by rotation curve data it is important to establish this result conclusively ; if confirmed , it may offer a way to falsify the cdm paradigm on small scales .
unfortunately , observational constraints are strongest just where theoretical predictions are least trustworthy .
for example , the alleged disagreement between observed rotation curves and cuspy dark halo models is most evident in sub-@xmath12 galaxies on scales of @xmath13 kpc or less . for typical circular speeds of @xmath14 km
s@xmath7 , this corresponds to regions where the density contrast exceeds @xmath15 .
orbital times in these regions are of order @xmath16 of the age of the universe , implying that n - body codes must be able to follow particles accurately for several thousand orbits .
few cosmological codes have been tested in a systematic way under such circumstances .
extreme care is thus needed to separate numerical artifacts from the true predictions of the cold dark matter model . furthermore , there are _ no _ simulations ( with adequate resolution ) of dark matter halos with masses comparable to the dwarf galaxies where the disagreement between theory and observation is most evident .
so far , published work has relied upon extrapolation of results obtained for massive galaxy halos and galaxy clusters , hardly an ideal procedure for assessing the viability of the ruling cdm paradigm on small scales . in order to validate or ` rule out ' the cdm cosmogony one must be certain that model predictions on the relevant scales are accurate , robust , and free of systematic numerical uncertainties .
motivated by this state of affairs , we have undertaken a large series of numerical simulations designed to clarify the role of numerical parameters on the structure of simulated cold dark matter halos .
what regions of a simulated dark matter halo in virial equilibrium can be considered reliably resolved ?
our tests explore the influence of the gravitational softening , the time - stepping algorithm , the starting redshift , the accuracy of force computations , and the number of particles on the spherically - averaged mass profile of a galaxy - sized halo in the @xmath17cdm cosmogony .
a thorough discussion of the tests is presented in power et al ( 2001 ) ; i summarize below a list of conditions that must be satisfied in order to consider the mass profile reliably resolved at a given radius : * the timestep must be substantially shorter than the orbital timescale : @xmath18 , where @xmath19 is the total number of timesteps , @xmath20 is the circular orbit timescale , and @xmath21 is the virial radius , where the mean density contrast is @xmath22 .
this criterion applies provided that the gravitational softening spline scalelength , @xmath23 , is chosen to ensure that particle discreteness effects are negligible , @xmath24 , where @xmath25 is the number of particles within the virial radius .
smaller softenings require _ substantially smaller timesteps _ for convergence .
* accelerations do not exceed a characteristic value imprinted by the circular speed of the halo , @xmath26 , and by the softening scale : @xmath27 , with @xmath28 . * enough particles are enclosed so that the collisional relaxation timescale , @xmath29 , is comparable to the age of the universe , @xmath30 . empirically , we find that regions where @xmath31 are adequately resolved .
( @xmath32 is the enclosed number of particles and @xmath33 is the usual coulomb logarithm . )
convergence also requires that at the initial redshift , @xmath34 , the linear rms fluctuations on the smallest resolved mass scale ( the particle mass @xmath35 ) is @xmath36 , and that force calculations are highly accurate .
poor spatial , time , or force resolution leads generally to systems with artificially low central density , but may also result in the formation of artificially dense central cusps .
this feature must be monitored carefully in hierarchical clustering simulations , since cusps formed in poorly - resolved high - redshift progenitors may survive merging and compromise the inner structure of present - day systems .
the most stringent requirement for convergence is that imposed on the particle number by the collisional relaxation criterion , which implies that in order to estimate accurate circular velocities at radii as small as @xmath37 of @xmath21 , where the density contrast may reach @xmath15 , the region must enclose at least @xmath38 particles ( or more than a few times @xmath39 within the virial radius ) .
we can use these criteria to identify regions of simulated dark halos where the mass profiles are unlikely to be affected by numerical artifacts .
as discussed in 1 , there is at present no broad consensus regarding how steep the innermost slope of dark matter density profiles is , or even whether there is a well - defined value for the asymptotic innermost slope .
unfortunately , the results of the previous section make it clear that it will take extraordinary computational effort to reach a robust resolution of the controversy .
what is required is a statistically significant sample of galaxy - sized halos simulated with sub - kpc resolution ; this is an extremely onerous computational task that will stretch the capabilities of the most powerful massively parallel computers .
steps in this direction are currently being taken , but it will take some time until these efforts yield conclusive results . from the theoretical point of view , a number of plausible arguments have been advanced in order to try and explain the innermost behaviour of dark matter density profiles from stellar dynamical principles .
these efforts , however , tend to give non - unique results and have so far been unable to explain the remarkable similarity in the structure of dark matter halos of widely different mass formed in a variety of cosmogonies ( evans & collett 1997 , syer & white 1998 , nusser & sheth 1999 , lokas & hoffman 2000 ) .
we have recently explored an empirical alternative to analytic efforts addressed at estimating the innermost slope of the density profile ( taylor & navarro 2001 , hereafter tn ) .
our approach exploits the similarity between the _ phase - space density _ profiles ( defined here as the ratio of density to velocity dispersion cubed , @xmath40 , measured in spherical shells ) of cdm halos and of the self - similar solution for spherical collapse in an expanding universe found by bertschinger ( 1985 ) .
this offers an attractive scenario for understanding the shape of halo density profiles as well as a powerful tool for estimating their slope near the center .
figure 1a shows the spherically - averaged phase - space density profile of three simulated cdm halos .
two of these halos have been identified in simulations of the former `` standard '' cdm cosmogony ( scdm , @xmath41 ) and have circular velocities of @xmath42 and @xmath43 km s@xmath7 , respectively .
these two halos are part of the `` local group '' simulation reported by moore ( 1999 ) .
the third halo has a circular velocity of @xmath44 km s@xmath7 and was run in the currently popular @xmath17cdm cosmogony ( @xmath45 ) .
solid ( dashed ) lines are used for the scdm ( @xmath17cdm ) halos .
for ease of comparison , we have chosen to normalize @xmath40 so that all three curves coincide at @xmath46 .
all three systems have @xmath15 particles within @xmath21 and are amongst the highest resolution simulations available at present .
the important point illustrated by figure 1a is that , over more than two decades in radius , the phase - space density profile is very well approximated by a power law of slope @xmath47 ( solid straight line ) .
this is quite remarkable , given that both the density profiles ( shown in figure 1b ) and the velocity dispersion profiles of these halos deviate substantially from simple power laws .
also remarkable is that this power - law coincides with the self - similar solution derived by bertschinger ( 1985 ) for secondary infall onto a spherical perturbation in an unperturbed einstein - de sitter universe : bertschinger s solution is plotted with solid circles in figure 1 .
the solution corresponds to the self - similar equilibrium configuration of a @xmath48 gas formed by spherical accretion onto a point - mass perturber in an otherwise uniform einstein - de sitter universe .
the quantity shown by the solid circles in figure 1a , @xmath49 , is roughly equivalent to the phase - space density ( @xmath50 is the local pressure ) . as discussed by bertschinger
, this solution is the most appropriate to compare with our numerical results for cdm halos , given that the velocity dispersion tensor in this case is only mildly anisotropic .
radii are normalized assuming that @xmath21 equals the shock radius of the self - similar solution , which corresponds to roughly one - third of the turnaround radius .
the vertical normalization is arbitrary and has been chosen to match the n - body results .
taking @xmath51 to be a measure of the local ` entropy ' of the system , figure 1a shows that cdm halos have the same radial entropy stratification as the simple spherical collapse solution .
it is possible that this power - law stratification is a fundamental property which underlies the similarity of structure of cold dark matter halos .
density profiles consistent with the power - law phase - space density profile shown in figure 1a can be obtained by assuming hydrostatic equilibrium .
the isotropic jeans equation admits a family of solutions for the density profile under the constraint @xmath52 .
the family is controlled by a single parameter , which can be expressed in terms of the ratio of the circular velocity to the 1-d velocity dispersion at @xmath53 , the radius where the circular velocity peaks ( @xmath54 ) .
for @xmath55 , the family includes a power - law , @xmath56 , with @xmath57 , which corresponds to bertschinger s spherical infall solution . as @xmath58 increases
, the density profiles become increasingly curved , although they still approach the steep power - law divergent behaviour near the center ( see , e.g. , second curve from top in figure 1b ) . for @xmath58
greater than some ` critical ' value , @xmath59 , the density profiles become unphysical , vanishing at a finite radius near the center . the `` critical '' density profile ( @xmath60 )
corresponds to the maximum value of @xmath58 consistent with a non - vanishing density profile at the center and , as shown in figure 1b , describes the n - body results much better than a power law .
interestingly , over approximately three decades in radius , _ the shape of the nfw density profile is very similar to the ` critical ' solution _ alluded to above .
the meaning of the critical solution may be understood by considering the phase - space density distribution corresponding to the different solutions .
the phase - space density distribution function is broadest for the power - law solution ( @xmath55 ) and gets increasingly narrower as @xmath58 increases ; the critical solution ( @xmath60 ) corresponds then to the most sharply peaked phase - space density distribution compatible with a monotonic density profile . in other words , the critical solution may be interpreted as a `` maximally mixed '' configuration where the phase - space density is as uniform as possible across the system .
this leads to the following interpretation of the origin of the nfw profile .
gravitational assembly of cdm halos leads to a simple power - law radial stratification of the phase - space density .
if spherical symmetry is imposed , as in the case treated by bertschinger ( 1985 ) , the collapse of each radial mass shell generates different `` entropies '' ( phase - space densities ) as they settle into virial equilibrium , leading to steeply cusped power - law profiles with slope @xmath61 . on the other hand , when the assumption of spherical symmetry is released and the collapse proceeds through many stages of hierarchical merging , mass shells are continuously `` mixed '' and the profiles tend to the critical solution : that corresponding to the most uniform entropy distribution compatible with a monotonic ( non - hollowed ) density profile and with the power - law entropy stratification constraint .
as discussed above , the ` critical ' density profile resembles the model proposed by nfw over a large dynamic range in radius . despite this similarity
, there are important differences between the two profiles , one of the most notable being that the critical solution tends to an asymptotic central slope of @xmath62 rather than to @xmath63 as in nfw s fitting formula . a simple approximation to the radial dependence of the slope of the ` critical ' density profile
is given by @xmath64 which is accurate to @xmath65 for @xmath66 . here
@xmath67 is the radius where the logarithmic slope of the density profile equals @xmath68 .
this should be compared with @xmath69 for nfw , where @xmath70 , or with the profile proposed by moore et al ( 1998 ) , @xmath71 .
the comparison is shown in figure 2a .
the critical profile ( labeled ` tn ' in this figure ) is overall similar to nfw , although it becomes shallower inwards more gradually than nfw . for @xmath72
the critical solution is shallower than nfw , and approaches slowly the asymptotic value of @xmath73 .
these differences are reflected mainly in the inner regions , as illustrated in figure 2b . here
the density profile corresponding to the three models are compared with the simulations .
the three models agree very well at radii exceeding @xmath10 , but differ systematically within @xmath10 .
although the critical solution predicts densities larger than nfw for r just inside @xmath10 , the situation eventually reverses for radii smaller than @xmath74 , outside the range plotted in figure 2b .
interestingly , the profiles of simulated halos exhibit similar deviations from nfw fits over the same radial range : the three curves connecting solid circles , corresponding to the cdm halos alluded to in 3 , show a clear excess over the nfw fit .
such excess has been noted by moore et al ( 1998 ) , ghigna et al ( 2000 ) , and fukushige & makino ( 1997 , 2001 ) , who interpret it as a signature of very steep ( @xmath75 ) central divergent slopes .
the modification to the nfw formula proposed by moore et al ( 1998 ) , where the innermost slope is @xmath76 rather than @xmath63 , indeed appears to describe the simulation results better than the nfw formula in the @xmath77 range ( see the top curve labeled ` moore et al ' in figure 2a ) .
however , and in light of the discussion in 3 , these systematic deviations from nfw fits do not necessarily signal the onset of a very steep central slope , but rather may reflect the more gradual radial dependence of the logarithmic slope of the critical solution ( figure 2a ) .
this interpretation seems to find support in the simulations : _ the profiles of the three simulated halos are significantly shallower than the moore et al fit near the center_. one may want to dismiss this feature as an artifact of limited numerical resolution , but application of our convergence criteria ( 2 ) to their own data suggest that the deviations are genuine and point to inner slopes significantly shallower than @xmath78 , just as expected from tn s critical solution . finally , figure 3a compares the circular velocity profiles of the same cdm halos with the predictions of the three models considered in figure 2 .
the cumulative mass profiles of tn s critical solution and nfw are essentially indistinguishable ; the different radial dependence of the density profile in the two models roughly compensate , and lead to almost identical circular velocity profiles that are in very good agreement with the results of the simulations .
as expected from the discussion of figure 2a , the moore et al model significantly overestimates the dark mass near the center compared with the simulations .
this result is confirmed by a further simulation of a galaxy - sized @xmath17cdm halo , whose density profile is shown in figure 3b .
the profiles correspond to the halo analyzed by power et al ( 2001 ) for the convergence tests described in 2 .
these tests vary the softening , timestep , and number of particles and show that extremely high numerical resolution is needed in order to discriminate between the three models .
the highest resolution halo in figure 3b ( which has more than 3 million particles within @xmath21 ) has a circular velocity profile which agrees well with nfw ( and tn s critical solution ) but is much shallower near the center than expected from moore et al s fitting formula . can these results be used to place meaningful constraints on the asymptotic inner slope ? at @xmath79 kpc , the smallest radius resolved in the highest - resolution run shown in figure 3b , both the local and cumulative density profiles are robustly determined : @xmath80 , and @xmath81 .
these values can be combined with the requirement of mass conservation to place an upper limit to the inner asymptotic slope of the density profile , @xmath82 . in other words , there is not enough mass within @xmath83 to support a power - law density profile with slope steeper than @xmath84 .
we note that this conclusion depends sensitively on our ability to resolve the innermost @xmath85 kpc .
if @xmath83 were just two or three times larger the same exercise would not be able to rule out slopes as steep as @xmath86 .
in summary , a number of interesting features emerge from the analysis of four very high - resolution ( million - particle class ) simulations of galaxy - sized cdm halos . *
the spherically - averaged , phase - space density profiles of cdm halos are very well approximated by a power law , @xmath87 .
the slope of this power law is consistent with the self - similar solution for spherical secondary infall derived by bertschinger ( 1985 ) . *
the circular velocity profiles of simulated cdm halos agree well with the model proposed by nfw down to the smallest resolved scales .
* on the other hand , the shape of the density profiles of simulated cdm halos differs modestly , but significantly , from the model proposed by nfw . as a result ,
nfw fits tend to underestimate the density just inside the scale radius @xmath10 , as pointed out by moore et al ( 1998 ) , ghigna et al ( 2000 ) , and fukushige & makino ( 1997 , 2001 ) .
* there is no obvious evidence for convergence to a steep asymptotic slope over the radial range probed by the simulations .
density profiles get increasingly shallow near the center , with innermost slopes significantly shallower than @xmath76 . *
a maximally - mixed ( ` critical ' ) model where the phase - space density profile is assumed to be a power law of radius accounts for these features and predicts that the slope of the density profile should gradually approach @xmath73 near the center .
the power - law behaviour of the phase - space density thus offers a natural way to gain insight into the structure of dark matter halos at radii where simulations become increasingly difficult and expensive , but also where observational constraints are strongest .
provided that the velocity dispersion tensor remains nearly isotropic , the `` critical '' solution provides a clear prediction as to the behaviour of the logarithmic slope of the density profile : it should become progressively shallower towards the center , converging to a value of @xmath88 .
this is interesting , because slopes shallower than @xmath76 appear to be consistent with the recent reanalysis of the rotation curve dataset by van den bosch ( 2000 ) and by van den bosch & swaters ( 2001 ) .
it is important to mention , however , that a shallow central slope does not guarantee consistency with observations , which constrain the detailed radial dependence of the density profile slope as well .
our results do , however , offer a clear prediction to extrapolate the mass profiles to regions that are very difficult to probe numerically .
is the extrapolation of the power - law behaviour to very small radii warranted ?
this question ultimately will have to be answered by direct numerical simulation , although there is no obvious a priori reason why a power law scaling that is valid for over two decades in radii should break down nearer the center .
we note that these conclusions are at odds with the proposal of moore et al ( 1998 ) and ghigna ( 2000 ) that the innermost slopes of cdm halos converge to a value not shallower than about @xmath76 .
however , it should be emphasized that their conclusion was based on the simulation of a _ single _ halo simulated in a standard cdm universe ( scdm ) and on a significantly different mass regime ( galaxy clusters ) than probed here .
thus the possibility remains that this particular system may not be representative of the general population or that the density profiles of clusters are steeper than those of galaxy - sized halos .
confirming which of these possibilities holds will require a statistically significant sample of halos simulated with resolution comparable to the systems used here . finally , it is also possible that the cluster simulated by moore ( 1998 ) and ghigna ( 2000 ) differs from the galaxy - sized halos we present here in other , more subtle ways .
for example , it may be significantly more triaxial than the systems analyzed here , or perhaps its velocity dispersion tensor is very anisotropic , in conflict with the assumptions of this work .
again , a detailed reanalysis of the discrepant system , extended to a statistically meaningful sample , appears necessary in order to explain conclusively this discrepancy .
the good agreement between the ` critical ' solution and the mass profiles of simulated halos suggests that the structure of cdm halos is determined by a radial stratification of phase - space density similar to that established through collapse onto a point mass perturbation in an unperturbed expanding universe , and by the uniformization of phase - space density that occurs presumably as a result of the many merger and satellite accretion events that characterize the assembly of a cdm halo .
this identification leaves a couple of important questions unanswered : ( i ) why should the phase - space density be a power law of radius ? , and ( ii )
why is the exponent the same as in bertschinger s self - similar solution ? .
although we have no clear answer to these questions at this point , our results suggest that explaining the origin of the structural similarity of cdm halos pointed out by navarro , frenk & white may entail unraveling why the radial stratification of phase - space density in cdm halos is the same power - law generated by the simple spherical collapse model .
discovering a mechanism that achieves this may provide a simple explanation for the universal structure of cold dark matter halos .
i am grateful to my collaborators : carlos frenk , adrian jenkins , chris power , tom quinn , volker springel , joachim stadel , james taylor , and simon white , for allowing me to report our results here , and also to j.makino , t.fukushige and p.hut for organizing a very enjoyable meeting .
cool talks , fresh sushi , cold sapporo , great company : all seemed a perfect antidote for a hot july in tokyo and kyoto .
thanks as well to b.moore , s.ghigna , g.lake , and f.governato for sharing with me data from their ` local group ' simulations .
this work has been supported by various grants from nserc and cfi . | i report on recent progress in our understanding of the structure of cdm halos , and in particular of the inner mass profile of galaxy - sized systems .
numerical simulations have consistently shown that the density profiles of cdm halos steepen monotonically from the center outwards , with slopes shallower than isothermal near the center and steeper than isothermal near the virial radius .
ongoing debate centers on the precise radial dependence of the logarithmic slope , as well as on whether it approaches a well defined asymptotic central value .
the latest high - resolution simulations suggest that the circular velocity profile is well approximated by the model proposed by navarro , frenk & white ( nfw ) . on the other hand ,
the radial dependence of the slope of the density profile differs modestly , but significantly , from the model proposed by nfw . as a result ,
nfw fits tend to underestimate the density at radii just inside the scale radius . rather than implying a very steep ( @xmath0 )
inner divergent slope , i argue that the data is actually best represented by a model where the density profile becomes increasingly shallow with radius , with little sign of approach to a well - defined asymptotic value .
a model where the _ phase - space density profile _ is a power law accounts well for these results and suggests that the innermost slope may be as shallow as @xmath1 .
these conclusions are supported by a thorough numerical convergence study that elucidates the effect of numerical parameters such as the timestep , gravitational softening , and particle number , on the mass profile of simulated dark matter halos .
c ii # 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in |
recently diluted magnetic semiconductors ( dms ) are under intensive investigation due to their potential applications in spintronics . in dms materials , transition or rare - earth metal ions
are substituted onto cation sites and are coupled with free carriers to yield ferromagnetism via indirect interaction .
mn - doped inas @xcite and gaas @xcite are the best studied dms materials .
conventional iii - v semiconductors are widely used for high - speed electronic and optoelectronic devices .
the discovery of hole - mediated ferromagnetism in ( ga , mn)as opened the way to integrate spin - based phenomena into mainstream microelectronics and optoelectronics as well as taking advantage of already established fabrication processes .
electrically controlled spintronic devices based on gamnas and inmnas have been successfully designed and tested at low temperatures , _
e.g. _ , a spin - led using gamnas as a spin injector @xcite .
however the highest curie temperature ( t@xmath0 ) reported in ( ga , mn)as is @xmath1170 k @xcite , which is far below room temperature and limits its regular application . in 2000 , dietl _ et al . _
@xcite used a mean field theory to estimate the ordering temperature t@xmath0 of ferromagnetic semiconductors , and they predict that room - temperature ferromagnetism can be created by substituting mn ions in p - type wide - band gap semiconductors such as gan and zno .
et al . _
calculated the properties of n - type zno doped with 3@xmath2 tm ions ( v , cr , mn , fe , co , and ni ) @xcite .
the ferromagnetic state , with a t@xmath0 of around 2000 k , is predicted to be favourable for v , cr , fe , co , and ni in zno while mn - doped zno is predicted to be antiferromagnetic .
these predictions largely boosted intensive experimental activities on transition metal doped gan and zno .
a large number of research groups have reported the experimental observation of ferromagnetism in tm ( from sc to ni ) doped zno @xcite fabricated by various methods including ion implantation . for a comprehensive review ,
however in these reports the magnetic properties using the same dopant vary considerably . _
e.g. _ the saturation moment and curie temperature for mn doped zno ranges from 0.075@xmath3/mn , 400 k @xcite to 0.17@xmath3/mn , 30 - 45 k @xcite , respectively .
in contrast to these publications , other groups reported the observation of antiferromagnetism @xcite , spin - glass behavior @xcite , and paramagnetism @xcite in tm - doped zno .
recently it was also found that nanoscale precipitates can contribute to the ferromagnetic properties substantially @xcite .
publications claiming the intrinsic ferromagnetism in tm doped zno are often based solely on magnetization measurements using high sensitivity superconducting quantum interference device magnetometery ( squid ) and structure characterizations using lab - equipped x - ray diffraction .
the latter has been demonstrated to be not sensitive enough to detect nanoscale precipitates @xcite .
nowadays the research community puts more effort to judge if tm ions in semiconductors are homogenously distributed , and if the ferromagnetism is intrinsic @xcite .
anomalous hall effect ( ahe ) , that verifies charge carrier participation in the magnetic order , has been suggested to judge about carrier - mediated ferromagnetism arising from dms .
however a recent study presents the co - occurrence of superparamagnetism and ahe in co doped tio@xmath4 films @xcite .
another criterium is the magnetic anisotropy @xcite as a signature of the intrinsic ferromagnetism .
the controversy in the magnetic properties of zno - based dms , as stated above , might partially be due to the insufficient characterization of the samples @xcite .
particularly , a careful correlation between structure and magnetism should be established by sophisticated methods . synchrotron radiation
based x - ray diffraction ( sr - xrd ) is a powerful tool to detect small precipitates , _
e.g. _ metallic tm nanocrystals in zno @xcite . on the other hand , element selective measurements of the magnetic properties , _
e.g. _ xmcd @xcite , and mssbauer spectroscopy @xcite , address the origin of the measured magnetism directly .
the present work is devoted to a comprehensive investigation of co and ni doped zno . by sr - xrd and squid magnetometry
, we correlate the structural and magnetic properties .
we attempt to answer the following questions concerning the presence of ferromagnetic precipitates .
\(1 ) how do the crystalline precipitates orient with respect to the host matrix ?
\(2 ) which techniques are suitable to detect these precipitates ? \(3 ) are these precipitates the major contribution of the measured ferromagnetism , or is there another source which contributes ?
\(5 ) can these nanoscale precipitates exhibit magnetic properties,_e.g .
_ magnetic anisotropy , concerning the criteria above mentioned ?
\(6 ) how do these precipitates behave upon thermal annealing ?
the paper is organized as follows . in section [ experiment ]
all the experimental methods employed will be described .
then the results will be presented in two sections . in section
[ as_implanted ] we will focus on the as - implanted samples , and discuss the orientation , the superparamagnetism , the magnetic anisotropy of co and ni ncs . in section [ post_annealing ]
we will describe the structure and magnetism evolution due to high vacuum annealing .
finally in section [ discussion ] we discuss the origin of the magnetic - anisotropy for the oriented co , ni ncs system , and the possible formation of co / coo and ni / nio core / shell structures upon annealing at 923 k. this paper is concluded in section [ conclusion ] .
commercial zno single crystals grown by the hydrothermal method were implanted with co or ni ions at 623 k with a fluence ranging from @xmath5 @xmath6 up to @xmath7 @xmath6 .
the implantation energy was 180 kev , which resulted in a projected range of @xmath8 nm , and a maximum atomic concentration from @xmath11% to @xmath110% ( trim code @xcite ) .
thermal annealing was performed in a high vacuum ( @xmath910@xmath10 mbar ) furnace from 823 k to 1073 k. the lattice damage induced by implantation was evaluated by rutherford backscattering / channeling spectrometry ( rbs / c ) .
rbs / c spectra were collected with a collimated 1.7 mev he@xmath11 beam at a backscattering angle of 170@xmath12 .
the sample was mounted on a three - axis goniometer with a precision of 0.01@xmath12 . during channeling measurement ,
the sample was aligned to make the zno@xmath90001@xmath13 axis parallel to the impinging he@xmath11 beam .
@xmath14 , the channeling minimum yield in rbs / c , is the ratio of the backscattering yield at channeling condition to that for a random beam incidence @xcite .
therefore , @xmath14 labels the degree of lattice disorder upon implantation , and an amorphous sample shows a @xmath14 of 100% , while a perfect single crystal corresponds to a @xmath14 of 1 - 2% .
magnetic properties were measured with a superconducting quantum interference device ( squid , quantum design mpms ) magnetometery .
the samples were measured with the field along both the in- and out - of - plane direction .
we studied both the temperature dependence of the magnetization at a constant field and the field dependence at a constant temperature . by squid , virgin zno
is found to be purely diamagnetic with a susceptibility of -2.65@xmath1510@xmath16 emu / oe@xmath17 g .
this background has been subtracted from the magnetic data .
the temperature dependent magnetization measurement has been carried out in the following way .
the sample was cooled in zero field from above room temperature to 5 k. then a 50 oe field was applied , and the zero field cooled magnetization curve ( zfc curve ) was measured with increasing temperature from 5 to 350 k , after which the field cooled magnetization curve ( fc curve ) was measured in the same field from 350 to 5 k with decreasing temperature .
structural analysis was performed both by synchrotron radiation x - ray diffraction ( sr - xrd ) and conventional xrd .
sr - xrd was performed at the rossendorf beamline ( bm20 ) at the esrf with an x - ray wavelength of 0.154 nm .
conventional xrd was performed with a siemens d5005 equipped with a cu - target source . in xrd measurement
, we use 2@xmath18-@xmath18 scans to identify crystalline precipitates , and pole figures ( azimuthal @xmath19-scan ) for determining their crystallographical orientation . as a standard approach
, for an xrd @xmath19-scan one first tilts the sample by the angle of @xmath20 from the sample surface ( @xmath20 is the angle between the diffraction plane of interest and the sample surface ) , and fixes the bragg angle .
subsequently the spectrum is recorded during azimuthal rotation with respect to the sample normal .
the pole figure is constructed by a series of @xmath19-scans at different @xmath20 .
rbs / c is used to check the lattice damage after implantation .
figure [ fig : rbs_conizno_asimp ] shows rbs / c spectra for different fluences .
the arrow labelled zn indicates the energy for backscattering from surface zn atoms .
the implanted co or ni ions can not be detected for the very low fluence ( 0.8@xmath1510@xmath21 @xmath6 , not shown ) .
however , they are more pronounced as a hump in the random spectrum for a larger fluence of 4@xmath1510@xmath21 @xmath6 and 8@xmath1510@xmath21 @xmath6 ( not shown ) .
the humps in the channeling spectra mainly originate from the lattice disordering due to implantation .
as expected , @xmath14 increases with increasing fluence ( see table [ tab : conizno_asimp ] ) .
note that the highest co fluence induced comparable lattice damage with the middle fluence , and less than the damage created by the same fluence of ni .
the reason could be a dopant specific self - annealing process @xcite .
rbs / c measurements also reveal that the zno is a hard material with respect to irradiation .
the host material still partly remains in a crystalline state after irradiation with co and ni ions up to a fluence of 8@xmath1510@xmath21 @xmath6 ( @xmath14 of 59% and 69% , respectively ) .
sr - xrd is used to identify the precipitates in zno after co or ni implantation .
figure [ fig : xrd_conizno_asimp ] shows the xrd @xmath22-@xmath18 scans for all samples implanted with different fluences . at a low fluence ( 0.8@xmath1510@xmath21 @xmath6 ) , no crystalline co or ni ncs could be detected . at a fluence of 4@xmath1510@xmath21 @xmath6 the hcp - co(0002 ) ( or fcc - co(111 ) ) and ni(111 )
peak appear , respectively , and grow with increasing fluence .
the full width at half maximum ( fwhm ) of the co or ni peak decreased with fluence , indicating the growth of the average diameter of these ncs ( table [ tab : conizno_asimp ] ) .
the crystallite size is calculated using the scherrer formula @xcite , @xmath23 where @xmath24 is the wavelength of the x - ray , @xmath18 the bragg angle , and @xmath25 the fwhm of 2@xmath18 in radians .
note that there is only one peak for co or ni detectable , which indicates that the crystallites of co or ni are highly oriented with respect to the host matrix .
the surface orientation is hcp - co(0001)(or fcc - co(111))@xmath26zno(0001 ) and fcc - ni(111)@xmath26zno(0001 ) , respectively .
-@xmath18 scan revealing the existence of co or ni precipitates in ( a ) co implanted zno , and ( b ) ni implanted zno . ] the co hcp structure only differs in the stacking from the fcc one .
since the bragg angles ( @xmath18 ) for hcp - co(0002 ) ( @xmath18=22.38@xmath12 ) and fcc - co(111 ) ( @xmath18=22.12@xmath12 ) are rather close to each other , it is difficult to assign these peaks in figure [ fig : xrd_conizno_asimp](a ) to hcp - co or fcc - co .
a @xmath19-scan or pole figure on one of the diffraction planes not parallel with the sample surface ( i.e. tilted by an angle @xmath20 from sample surface ) helps to identify hcp or fcc - co ncs and also reveals the crystallographical orientation relationship . by this approach , we find that only hcp - co is present in the as - implanted samples .
figure [ fig : conizno_polefigure ] ( a ) and ( b ) show the pole figure of hcp - co(10@xmath271 ) and ni(200 ) , respectively .
the radial coordinate is the angle ( @xmath20 ) by which the surface is tilted out of the diffraction plane .
the azimuthal coordinate ( @xmath19 ) is the angle of rotation about the surface normal .
the pole figure shows poles of hcp - co(10@xmath271 ) at @[email protected]@xmath12 , and ni(200 ) at @[email protected]@xmath12 , respectively . both exhibit a sixfold symmetry . since zno(10@xmath272 ) and hcp - co(10@xmath271 ) have similar bragg angle , the poles of zno(10@xmath272 )
also show up at @[email protected]@xmath12 with much more intensities .
the results are consistent with the theoretical co(10@xmath271 ) pole figure viewed along [ 0001 ] , and ni(200 ) pole figure viewed along [ 111 ] direction , respectively .
therefore , we can conclude that these co and ni ncs are crystallographically oriented with respect to the zno matrix .
the in - plane orientation relationship is hcp - co[10@xmath270]@xmath26zno[10@xmath270 ] , and ni[112]@xmath26zno[10@xmath270 ] , respectively . due to the hexagonal structure of co and sixfold symmetry of ni viewed along [ 111 ] direction , it is not difficult to understand their crystallographical orientation onto hexagonal - zno . 1 )
( in square ) together with the tail of zno(10@xmath272 ) ( in circle ) ; the data points out of squares and circles are due to the background ; ( b ) ni(200 ) ( in square ) together with the tail of zno(10@xmath272 ) ( in circle ) . ] at this point , we have to remind the reader that in our former work on ni implanted zno , the xrd measurement was performed on a conventional lab - equipped diffractometry ( cxrd ) @xcite .
cxrd reveals similar results as sr - xrd ( from the ni fluence of 4@xmath1510@xmath21 @xmath6 , ni ncs start to form )
. however , cxrd fails to detect fe ncs in zno @xcite , where fe ncs are not well oriented like the case of ni in zno .
the peak intensity in xrd is proportional to the diffraction volume , _
i.e. _ the number of diffraction planes which are parallel to each other . for a 2@xmath18-@xmath18 scan , the crystallographic orientation results in much more diffraction volume than the random orientation .
therefore , the high ordered orientation of ncs make them easier to be detected .
as expected , co ncs are also detectable in the cxrd measurement ( not shown ) .
cccccccc fluence & @xmath14 & crystallite size & saturation magnetization & metallic & coercivity & t@xmath28 & crystallite size + ( @xmath6 ) & ( rbs / c ) & ( nm ) ( xrd ) & ( @xmath3/co or /ni ) & fraction & ( oe ) & ( k ) & ( nm ) + co : 0.8@xmath1510@xmath21 & 44% & - & - & - & - & - & + co : 4@xmath1510@xmath21 & 54% & 5 & 0.29 ( 5 k ) & 17% & 1400 ( 5 k ) & 45 & 4.3 + co : 8@xmath1510@xmath21 & 57% & 8 & 0.44 ( 5 k ) & 26% & 1400 ( 5 k ) & 300 & 8.1 + + ni : 0.8@xmath1510@xmath21 & 45% & - & 0.05 ( 10 k ) & 8% & 10 ( 10 k ) & @xmath29 5 & @xmath9 9.4 + ni : 4@xmath1510@xmath21 & 57% & 6 & 0.16 ( 10 k ) & 27% & 30 ( 10 k ) & 16 & 14 + ni : 8@xmath1510@xmath21 & 69% & 8 & 0.22 ( 10 k ) & 37% & 120 ( 10 k ) & 70 & 23 + the magnetic properties of co and ni implanted zno were measured by squid magnetometery with the field parallel and perpendicular to the sample surface .
co implanted samples exhibit a hard axis parallel to the sample surface , while the easy axis is perpendicular to the surface . for ni - implanted samples ,
the anisotropy directions are vice versa . in this section ,
we investigate the superparamagnetism of the implanted zno samples and their magnetic anisotropy . from the xrd results , we know that co and ni ncs have been formed in the as - implanted samples . for magnetic nanoparticles ,
the formation of domain walls is energetically unfavorable . below a certain size ( typically in the range of 15 to 30 nm depending on the material )
, the particle stays in a single - domain configuration . if the particle size is sufficiently small , above a particular temperature ( so - called blocking temperature , t@xmath30 ) thermal fluctuations dominate and no preferred magnetization direction can be defined .
such a system of superparamagnetic particles does not exhibit a hysteresis curve above t@xmath30 ; therefore the coercivity ( h@xmath0 ) and the remanence ( m@xmath31 ) are both zero . for a dc magnetization measurement in a small magnetic field by squid , t@xmath28
is given by @xmath32 where @xmath33 is the anisotropy energy density , @xmath34 the particle volume , @xmath35 the boltzmann constant @xcite .
for bulk crystals , @xmath36 is 5.0@xmath1510@xmath37 and 5.7@xmath1510@xmath38 jm@xmath39 for co and ni , respectively , at room temperature .
phenomenologically there are two characteristic features in the temperature dependent magnetization of a nanoparticle system .
one is the irreversibility of the magnetization in a small applied field ( _ e.g. _ 50 oe ) after zero field cooling and field cooling ( zfc / fc ) @xcite .
the other is the drastic drop of coercivity and remanence at a temperature close to or above t@xmath30 @xcite .
figure [ fig : cozno_mh_asimp ] shows the magnetization versus field reversal ( m - h ) measured at 5 k with the field applied perpendicular to the sample surface ( along zno[0001 ] ) . a hysteretic behavior is observed for the high - fluence implanted samples . a saturation behavior
is also observed at 300 k for the sample with the highest fluence ( figure [ fig : cozno_mh_asimp ] inset ) .
however , neither coercivity nor remanence can be observed at 300 k. this is a strong indication for the superparamagnetism of a magnetic nanoparticle system .
knowing the formation of hcp - co from xrd , it is reasonable to conclude that hcp - co ncs are responsible for the magnetic behavior . for bulk hcp - co crystals ,
the magnetic moment is 1.7 @xmath40/co at 0 k. assuming the same value for co ncs , around 17% and 26% of implanted co ions are in the metallic state for the fluence of 4@xmath15 and 8@xmath1510@xmath21 @xmath6 respectively .
similar results are observed for ni implanted zno @xcite . note that the hysteresis loop for the fluence of 4@xmath1510@xmath21 @xmath6 exhibits two reversal steps . there is a very small kink when the field decreases from -2000 to -4000 oe ( and increases from 2000 to 4000 oe ) .
this wasp - waist shape of the loop is associated with magnetic phases with different coercivities @xcite .
temperature dependent magnetization with h@xmath41 oe was measured after zfc and fc to confirm the superparamagnetism ( figure [ fig : cozno_zfcfc_asimp ] ) . for the magnetic samples , a distinct difference in zfc / fc curves is observed .
zfc curves show a gradual increase ( deblocking ) at low temperature , and reach a broad peak , while fc curves continue to increase with decreasing temperature .
the broad peaks in zfc curves are due to the size distribution of co ncs . in this paper ,
the temperature at the maximum of the zfc curve is taken as the average blocking temperature ( later referred as t@xmath30 ) .
the zfc / fc curves are general characteristics of magnetic nanoparticle systems with a broad size distribution @xcite .
t@xmath30 increases with the fluence , _
i.e. _ the size of nanoparticles .
table [ tab : conizno_asimp ] lists the average size of co ncs calculated by eq .
[ scherrer ] ( xrd ) and eq .
[ blocking_squid ] ( squid ) .
they are in a good agreement although there is also a size distribution in co ncs @xcite .
the zfc / fc magnetization was also measured for ni implanted samples @xcite .
comparing with co , ni has a much lower anisotropy energy density . for similar sizes of ni ncs ,
the blocking temperature is therefore much lower than that of co. table [ tab : conizno_asimp ] lists the average size of ni ncs calculated by eq .
[ scherrer ] ( xrd ) and eq .
[ blocking_squid ] ( squid ) .
although the trend is the same for both calculations , the values from squid are much larger than that from xrd data .
one reason could be the anisotropy energy density is underestimated by assuming the magneto - crystalline anisotropy constant .
another reason is that t@xmath30 is overestimated by taking the temperature at the maximum of the zfc curve .
according to the calculation by farle _
et al . _
@xcite , and by jacobsohn _
et al . _
@xcite , the size dispersion for a given average particle diameter broadens the peak of zfc curve and shifts it to higher temperatures .
m - h loops were also measured for selective samples with co fluence of 8@xmath1510@xmath21 @xmath6 with the field both perpendicular and parallel to the sample surface .
figure [ fig : conizno_aniso_asimp](a ) shows the comparison of the magnetization along zno[10@xmath270 ] and [ 0001 ] at 300 k. figure [ fig : conizno_aniso_asimp](c ) shows the orientation relationship between hcp - co and zno , and the measurement geometry .
obviously co[0001 ] is the easy axis , the same as a bulk hcp - co crystal .
the intersection of both curves gives an effective anisotropy field of 3000 oe . at 5
k , the magnetic anisotropy is the same ( not shown ) , and the coercivity of the easy axis is around 1400 oe .
the ratio between remanence and saturation magnetization is around 60% .
this rather low remanence ( below 100% ) is due to the size distribution of nanomagnets .
very small nanomagnets behave superparamagnetically even at low temperatures , and only have field induced magnetization .
this is a rather universal feature for nanomagnets .
for instance , epitaxial mnas nanoclusters in gaas , the remanence is also below 100% along the easy axis @xcite .
figure [ fig : conizno_aniso_asimp](b ) shows the same measurement of ni implanted zno , while ( d ) shows the orientation relationship between fcc - ni and zno , and the measurement geometry .
in contrast to bulk ni where [ 111 ] is the easy axis , the easy axis is ni[112 ] and the hard axis is ni[111 ] .
moreover , as shown in figure [ fig : conizno_aniso_asimp](b ) , another in - plane direction ni[110 ] is also an easy axis . within the applied field ,
the magnetization curve along hard axis does not intersect with that along the easy axis .
the effective anisotropy field is much larger than 1500 oe .
that means that there are other contributions to the anisotropy dominating over the crystalline magnetic anisotropy .
this will be discussed in section [ discussion ] .
( out - of - plane ) and [ 10@xmath270 ] ( in - plane ) for co / ni implanted zno with the fluence of 8@xmath1510@xmath21 @xmath6 measured at 300 k , ( a ) co implanted zno ; and ( b ) ni implanted zno .
( c ) and ( d ) show the schematic geometry for magnetization measurements . ] in the work by norton _
et al . _
@xcite , epitaxial co nanocrystals have been observed in co implanted zno single crystals .
the nanocrystal size is estimated to be @xmath13.5 nm , which is below the superparamagnetic limit at room temperature .
therefore the ferromagnetism above 300 k is very possible due to co substitution onto the zn site in the zno matrix . in our case ,
the measured superparamagnetism is well explained by the presence of co and ni nanocrystals .
the formation of metallic nanocrystals already in the as - implanted state is due to the elevated implantation temperature ( 623 k ) , which facilities the precipitation process . implantation at low temperatures ( _ e.g. _ 253 k ) prohibits precipitation , however results in non - magnetism in our case , _
e.g. _ fe implanted zno @xcite .
the post - annealing was performed in high vacuum with temperatures ranging from 823 k to 1073 k for 15 min .
the samples to be annealed were selected according to the squid measurement capability , namely the temperature range from 1.8 to 400 k. due to the higher anisotropy energy of co , t@xmath30 in zfc curve of the highest fluence is already 300 k in the as - implanted state ( see figure [ fig : cozno_zfcfc_asimp ] ) .
the annealing is expected to increase the size of co ncs , and consequently increase t@xmath30 in the zfc curve , which will exceed the temperature range of the squid magnetometery .
therefore , the sample implanted with the middle co fluence was chosen . for ni
the highest implantation fluence was chosen .
as shown in figure [ fig : rbs_conizno_asimp ] , ion implantation substantially induces lattice damage of zno crystals . here we show that this damage can be partially recovered by post - annealing .
figure [ fig : rbs_cozno_ann ] shows the rbs / c spectra for the sample implanted with a co fluence of 4@xmath1510@xmath21 @xmath6 .
the annealing temperatures are given in the figure .
with increasing annealing temperature , the channeling spectra indicate that the lattice disorder of zno progressively decreases .
after annealing to 1073 k , the channeling spectrum is almost comparable with the virgin zno .
similar rbs / c results are observed for ni - implanted zno upon annealing ( not shown ) .
10@xmath21 @xmath6 after thermal annealing at different temperatures .
the lattice damage induced by implantation is progressively reduced by increasing annealing temperature . ] -@xmath18 scans : ( a ) co implanted zno crystals with different annealing temperature .
the wide range xrd pattern for one of the samples ( inset ) reveals that no other crystalline phase ( e.g. coo ) could be detected .
the arrow points the peak shoulder coming from fcc - co(111 ) diffraction in the sample of 823 k ann .
( b ) ni implanted zno crystals with different annealing temperature .
the inset shows a comparison between as - implanted and 823 k annealed samples . ]
figure [ fig : xrd_conizno_ann](a ) shows the development of co ncs upon thermal annealing . the peak area and crystallite size calculated using the scherrer formula @xcite
are compared in table i. a broad scan ( the inset of figure [ fig : xrd_conizno_ann](a ) ) reveals only one peak from co besides the zno peaks .
an xrd @xmath19-scan has been used to distinguish between hcp- and fcc - co ( figure [ fig : phi_cozno ] ) .
we find only hcp - co in the as - implanted sample and the sample annealed at 923 k , while both fcc- and hcp - co are present in the sample annealed at 823 k. the broad peak in figure [ fig : xrd_conizno_ann](a ) ( 823 k ann . )
is a superposition of hcp - co(0002 ) and fcc - co(111 ) .
the crystallographical orientation relationship between co ncs and zno is hcp - co(0001)[10@xmath270]//zno(0001)[10@xmath270]//fcc - co(111)[112 ] . with this orientation , the 2@xmath18-@xmath18 scans for hcp - co(10@xmath271 ) and fcc - co(200 )
are expected with a skew geometry at one of the azimuthal position ( _ e.g. _ at @xmath19=0 ) as shown in figure [ fig : co_skew](a ) and ( b ) , respectively . in skew geometry
, the incident and the diffracted waves have the same angles to the surface , while the sample is titled with respect to its surface normal . by this configuration , a noncoplanar
, its surface normal does not lie in the plane defined by the incident and the diffracted waves , can be measured @xcite .
note that the peak area of co in figure [ fig : xrd_conizno_ann ] , which is an approximate measure of the amount of co ncs , increases drastically after 823 k annealing , while decreases after 923 k annealing .
it is reasonable to attribute this change to the formation and disappearance of fcc - co .
the fcc - co is probably oxidized to the amorphous coo after 923 k annealing , while finally the majority of co ncs are oxidized to amorphous state after annealing at 1073 k. -scans for hcp - co(10@xmath271 ) ( @xmath18=23.78@xmath12 and @[email protected]@xmath12 ) , fcc - co(002 ) ( @xmath18=25.76@xmath12 and @[email protected]@xmath12 ) and zno(10@xmath271 ) ( @xmath18=18.13@xmath12 and @[email protected]@xmath12 ) reveal the in - plane orient relationship for co ncs respect to zno . ] for ni implanted zno , the structure evolution upon annealing is similar as shown in figure [ fig : xrd_conizno_ann](b ) .
the peak area and the crystallite size is listed in table [ tab : conizno_ann ] .
the mild temperature annealing ( 823 k ) only slightly increases the grain size of ni .
the annealing at 923 k drastically decreases the peak area , while the grain size also decreases .
note that there is a significant shift in the ni(111 ) peak ( labelled by dash lines ) , which indicates the relexation of lattice strain .
after annealing at 1073 k , the majority of ni ncs could be oxidized to amorphous state .
-@xmath18 scans for zno(10@xmath271 ) and hcp - co(10@xmath271 ) , those small sharp peaks are from artificial noise in the measurements ; ( b ) 2@xmath18-@xmath18 scan for fcc - co(002 ) . ]
cccccccc sample & peak area & crystallite size & saturation magnetization & metallic & coercivity & t@xmath28 & crystallite size + & ( xrd ) & ( nm ) ( xrd ) & ( @xmath42@xmath30/co or /ni ) & fraction & ( oe ) & ( k ) & ( nm ) + co : as - imp .
& 380 & 5 & 0.29 ( at 10000 oe ) & 17% & 1400 & 45 & 4.3 + co : 823 k ann . & 8330 & - & 0.36 ( at 2000 oe ) & 21% & 250 & 80 & 5.2 + co : 923 k ann . & 1400 & 10 & 0.32 ( at 2000 oe ) & 19% & 450 & 330 & 8.3 + co : 1073 k ann . & 0 & - & - & - & - & - & - + + ni : as - imp . &
7312 & 8 & 0.22 ( at 2000 oe ) & 37% & 230 & 70 & @xmath9 9.4 + ni : 823 k ann .
& 6590 & 9 & 0.22
( at 2000 oe ) & 37% & 200 & 80 & 14 + ni : 923 k ann . & 750 & 7 & 0.18 ( at 2000 oe ) & 30% & 220 & - & - + ni : 1073 k ann . & 0 & - & - & - & - & - & - + the structural phase transformation of co ncs results in different magnetic properties as revealed by squid .
figure [ fig : squid_cozno_ann](a ) shows the zfc / fc magnetization curves for all samples annealed at different temperatures .
obviously , except the sample annealed at 1073 k , the zfc curves show a gradual increase ( deblocking ) at low temperature , and reach a maximum at a temperature of t@xmath30 ( shown in table [ tab : conizno_ann ] ) , while fc curves continue to increase with decreasing temperature .
no significant magnetization response is detected for the sample annealed at 1073 k. note that t@xmath30 increases drastically above 330 k after annealing at 923 k. however jacobsohn _ et al . _
@xcite reported a much lower t@xmath30 ( 250 k ) of hcp co ncs with similar grain size ( @xmath110 nm ) .
the higher t@xmath30 could be due to some other anisotropy energy , which stablized the superparamagnetism at higher temperature .
this will be discussed in section [ discussion ] .
0]@xmath26hcp - co[10@xmath270]@xmath26fcc - co[112 ] ( solid symbols ) and zno[0001]@xmath26hcp - co[0001]@xmath26fcc - co[111 ] ( open symbols ) . implantation or annealing temperature is shown . ]
0]@xmath26fcc - ni[112 ] ; ( c ) and ( d ) m - h curves measured at 5 k for all samples along zno[10@xmath270]@xmath26fcc - ni[112 ] ( solid symbols ) and zno[0001]@xmath26fcc - co[111 ] ( open symbols ) after 823 k and 923 k annealing , respectively . implantation or annealing temperature is shown . ]
hysteresis loops were measured for all samples in both parallel ( zno[10@xmath270]@xmath26hcp - co[10@xmath270]@xmath26fcc - co[112 ] ) and perpendicular ( zno[0001]@xmath26hcp - co[0001]@xmath26fcc - co[111 ] ) directions .
figures [ fig : squid_cozno_ann](b)-(d ) reveal that the anisotropy and coercivity can be tuned by different annealing procedures .
the as - implanted sample only consists of hcp co ncs , which persist the bulk - like anisotropy along [ 0001 ] direction with a high coercivity ( figure [ fig : squid_cozno_ann](b ) ) .
the sample annealed at 823 k mainly consists of fcc - co with easy axis is along the fcc - co[112 ] direction ( figure [ fig : squid_cozno_ann](c ) ) .
after annealing at 923 k , the sample shows an easy axis both along hcp - co[10@xmath270 ] and hcp - co[0001 ] , but a higher coercivity along the latter direction .
the magnetic properties of different samples are listed in table [ tab : conizno_ann ] .
figure [ fig : squid_nizno_ann ] show the magnetic properties of ni implanted zno upon annealing .
as expected , the 823 k annealing increases t@xmath30 in the zfc curve due to the increase of crystallite size .
however annealing at 923 k , the shape of zfc magnetization curve is deviated from others .
there is no real maximum , but a broad plateau . the formation of ni / nio core / shell structure could introduce the exchange coupling , which contributes another anisotropy energy
this will be discussed in the following section .
note that after annealing at 1073 k the sample still exhibits a non - neglectful response in zfc / fc magnetization measurement .
there are could be a small amount of ni nanocrystals remaining , while they are beyond the detection limit in sr - xrd measurement .
as shown before , the magnetic anisotropy of ni ncs embedded in zno are drastically different from the bulk ni . for bulk
ni , the magnetocrystalline anisotropy constant of k@xmath43 is -5.7@xmath1510@xmath38 j m@xmath39 at 300 k. the [ 111 ] direction is the easy axis . for a single magnetic nc , the magnetocrystalline , shape , and magnetoelastic anisotropy have to be considered . in principle
, a uniformly magnetized single domain spherical particle has no shape anisotropy , because the demagnetizing factors are isotropic in all directions .
however , in the case of a nonspherical sample it will be easier to magnetize along a long axis than along a short one .
the fwhm of the ni peak in xrd 2@xmath18-@xmath18 scans is a measure of the crystallite size . along the ni[111 ]
direction , the crystallite size is estimated to be around 8 nm .
for the in - plane direction , the crystallite size could be estimated by measuring a diffraction plane not parallel with sample surface .
we chosen ni(200 ) diffraction plane .
figure [ fig : xrd_ni(200)(111 ) ] shows the comparison of the normalized ni(111 ) and ( 200 ) peaks . actually , the ni(200 ) is broader than ( 111 ) , which means the crystallite size along the in - plane direction is even smaller than the [ 111 ] direction . therefore , the shape anisotropy is not the key reason to induce the easy axis along the in - plane direction .
-@xmath18 scan of ni(200 ) and ni(111 ) .
] now we consider the strain anisotropy .
this kind of anisotropy is often described by a magnetoelastic energy term @xmath44 where @xmath45 is the magnetostriction constant , @xmath46 the stress , and @xmath19 the angle between magnetization and the strain tensor axis @xcite .
the strain @xmath47 along [ 111 ] direction of our ni ncs can be calculated according to xrd measurement .
the lattice spacing is given by bragg law : @xmath48 where @xmath49 is the lattice spacing , @xmath18 the bragg angle , and @xmath24=0.154 nm the wavelength of x - ray .
@xmath18 is obtained from figure [ fig : xrd_conizno_asimp ] .
the strain is defined as following @xmath50 where @xmath2 is the theoretical lattice spacing for bulk ni or co. [ cols="^,^,^,^,^,^,^",options="header " , ] using the approach and parameters in refs .
@xcite , the magnetoelastic anisotropy constant is calculated ( see table [ tab : anisotropy ] ) . for ni ncs ,
the magnetoelastic anisotropy constant is one order higher than the magnetocrystalline anisotropy constant .
therefore the magnetoelastic anisotropy energy dominates the total anisotropy energy .
this finding demonstrates the possibility to tune the magnetic properties by embedding magnetic ncs in different host matrix .
the annealing at 823 k does not change the strain status significantly .
after annealing up to 923 k , the elastic strain is partially released .
however , the magnetoelastic anisotropy energy is still much higher than the magnetocrystalline one . note the anisotropy change in figure [ fig : squid_nizno_ann](d ) after 923 k annealing by comparing with the 823 k annealing
. there could be another anisotropy source after 923 k annealing , which will be discussed in section [ 923k ] . for hcp - co ,
the magnetocrystalline anisotropy constant is very large , and it is difficult to be dominated by other anisotropy energy contribution ( as shown in table [ tab : anisotropy ] ) .
therefore the hcp - co ncs in the as - implanted state still persist the bulk like anisotropy behavior . however , the annealing at 823 k resulted in coexistence of fcc - co and hcp - co ncs .
since fcc - co is a meta - stable state , its magnetic and mechanic data are very limited . therefore in this paper
, we will not discuss its magnetoelastic anisotropy energy .
nevertheless fcc - co[112 ] is the easy axis rather than [ 111 ] , which could be due the lattice strain , as the same reason for ni nanocrystals in zno .
after annealing at 923 k , hcp - co is the only phase , and the elastic strain is not significantly changed compared with the as - implanted state ( see figure [ fig : xrd_conizno_ann](a ) ) .
however , the anisotropy is obviously changed .
like the ni case , there could be another anisotropy source , which will be discussed in section [ 923k ] .
after annealing at 923 k , the magnetic properties , namely zfc / fc magnetization , and magnetic anisotropy , are changed significantly compared with other samples . for the co case ,
@xmath51 in the zfc curve increases drastically , while the fc curve is not mono - decreased with increasing temperature . for ni case
, there is a broad plateau in zfc curve .
one explanation for the higher @xmath51 is the increase and broadening of the co crystallite size .
_ calculated the zfc curves by varying different parameters , including the size distribution @xcite .
it is found that a slight broadening can result in a very broad and high zfc curve .
this can well explain the zfc magnetization , but not the fc magnetization .
moreover the drastic changes in coercivity , and magnetic anisotropy can not be explained in such an approach .
in addition , table [ tab : conizno_ann ] lists the comparison of the xrd peak area and the magnetization upon annealing .
note that the xrd peak area decreases drastically by almost one order after 923 k annealing , which indicates the decreasing of the amount of the co or ni ncs , however the saturation magnetization only decreases slightly .
another explanation is due to other anisotropic energy contribution .
_ have found that the exchange coupling between co ncs and their coo shells drastically increase the blocking temperature @xcite . in our xrd measurement ,
the amount of metallic co and ni decreases after annealing at 923 k , which is very probably due to the oxidation .
therefore we could assume the formation of a co / coo core / shell structure in our sample , and the exchange coupling increases the blocking temperature .
the exchange coupling is further confirmed by the vertical shift of the magnetization loop ( figure [ fig : co_mh_zfcfc ] ) .
the loop - curve is shifted along the magnetization axis after cooling from 350 k in an applied field .
the inset shows the temperature dependence of the vertical shift . ]
the vertical shift of the magnetization loop is a strong evidence for the presence of an interfacial interaction between an anitferromagnet ( afm ) and a ferromagnet ( fm ) @xcite , _ i.e. _ the hcp co ncs are surrounded by coo , which is amorphous , and therefore can not be detected by xrd .
et al . _
@xcite presented a model to discuss the exchange coupling of nanoscaled co / coo core / shell structures . below a critical size ( 12 nm ) of co cores ,
the interfacial exchange energy is larger than both the zeeman energy of fm and the anisotropy energy of the afm due to a large surface - to - volume ratio of ncs , and consequently some spins in the fm part can be frozen by the afm part , leading to a vertical shift along the magnetization axis after field cooling .
this model well explains our co / coo system with an average diameter of 10 nm for co ncs .
the vertical shift of the magnetization loop decreases with increasing temperature and disappears at a temperature between 200 and 300 k. this is consistent with the nel temperature of 290 k for coo @xcite .
such a vertical shift is also observed in the fc - loop measured along zno[10@xmath270]@xmath26hcp - co[10@xmath270 ] ( in - plane ) ( not shown ) .
for the ni implanted sample annealed at 923 k , the hysteresis loop was also measured under both zfc and fc conditions ( not shown ) . a similar shift along the magnetization axis is observed .
therefore we would assume the formation of a co / coo ( ni / nio ) core / shell structure .
the exchange coupling between co ( ni ) and coo ( nio ) contributes another anisotropy energy .
this explains the high blocking temperature and the change in magnetic - anisotropy after annealing at 923 k. note that for both of cases , the magnetization is only slightly decreased after annealing at 923 k. this is controversial with the drastic decrease of the diffraction intensity ( see figure [ fig : xrd_conizno_ann ] ) .
one possible reason is the formation of small co or ni nanocrystals due to annealing .
as shown in table [ tab : conizno_asimp ] , only around 10 - 40% of co or ni form as metallic nanocrystals and others remain as dispersed ones .
annealing at 923 k , on one hand , oxidized some metallic nanocrystals partially , on the other hand , could induce the gettering of dispersed co and ni and result in small metallic nanocrystals . below a critical size , nanocrystals are non - detectable by xrd .
another reason could be that only a thin shell of co or ni nanocrystals transforms into oxides , which results in the invisibility of oxides in xrd even they are crystalline .
obviously a detailed investigation using transmission electron microscopy or other more sensitive techniques should be performed to clarify this controversy .
a thorough characterization of the structural and magnetic properties has been presented on co and ni implanted zno single crystals .
the results by sr - xrd and squid magnetometery correlate well with each other .
the major conclusions are summarized as follows .
\(1 ) co or ni ncs have been formed in zno upon ion implantation .
their crystalline sizes , generally below 10 nm , increase with increasing fluence . from several to 37% percent of implanted co or ni is in metallic states , while the remaining could be diluted into zno matrix .
the co or ni ncs are the origin of the measured ferromagnetism .
\(2 ) the co and ni ncs are crystallographically oriented with respect to zno host matrix .
the orientation relationship is as follows : hcp - co(0001)[10@xmath270]@xmath26zno(0001)[10@xmath270 ] , and ni(111)[112]@xmath26zno(0001)[10@xmath270 ] .
this well ordered structure of ncs could result in a rather smooth interface between them and zno host , and makes the hybrid of ferromagnetic ncs and semiconductors promising for spintronics functionality .
\(3 ) magnetic anisotropy is observed for co or ni ncs in zno . especially for the ni ncs
, the anisotropy is different from the bulk crystals .
the extra anisotropy energy is attributed to the lattice strain impressed from the host matrix .
this opens a route to artificially tune the magnetic properties of nanoparticles by selecting of substrate materials .
\(4 ) the structure and magnetic properties of co or ni ncs embedded inside zno can be tuned by post - annealing . for the co case , 823 k annealing results in the co - exists of fcc - co and hcp - co .
the magnetic anisotropy is changed from out - of - plane to in - plane .
annealing at 923 k could have partially oxidized metallic co and ni , and result in co / coo ( ni / nio ) core / shell structures .
after annealing at 1073 k , no co or ni ncs can be detectable within the detection limit of sr - xrd , at the same time , the samples shows no pronounced ferromagnetism down to 5 k. | in the last decade , transition metal doped zno has been intensively investigated as a route to room temperature diluted magnetic semiconductors ( dms ) .
however the origin for the reported ferromagnetism in zno based dms remains questionable .
possible options are diluted magnetic semiconductors , spinodal decomposition or secondary phases . in order to clarify this question ,
we have performed a thorough characterization of the structural and magnetic properties of co and ni implanted zno single crystals .
our measurements reveal that co or ni nanocrystals ( ncs ) are the major contribution of the measured ferromagnetism . already in the as - implanted samples , co or ni ncs have formed , and they exhibit superparamagnetic properties .
the co or ni ncs are crystallographically oriented with respect to the zno matrix .
their magnetic properties , _ e.g. _ the anisotropy and the superparamagnetic blocking temperature can be tuned by annealing .
we discuss the magnetic anisotropy of ni ncs embedded in zno concerning the strain anisotropy . |
for cavity cooling in the weak coupling regime @xmath71 , coherences @xcite decay rapidly , and rate equations are sufficient to describe the cooling @xcite . for one - dimensional
cooling along the cavity axis @xmath43 , the rate of transitions from motional state @xmath72 to @xmath73 is @xmath74 and the rate of transitions from motional state @xmath72 to @xmath75 is @xmath76 where @xmath33 is the photon scattering rate into free space , the number @xmath10 is defined such that @xmath77 is the average recoil heating along the @xmath43 direction per free space scattering event , and @xmath78 is the heating rate along the @xmath43 direction due to environmental electric field fluctuations in quanta per second . here , @xmath79 is the lamb - dicke parameter , as determined by the ratio of recoil energy @xmath80 and trap vibration frequency @xmath2 .
note that these transition rates are only valid in the lamb - dicke regime @xmath81 , which limits the applicability of this model to @xmath82 for our experimental parameters .
the expectation value of the mean vibrational quantum number @xmath83 evolves according to @xmath84 for @xmath85 ( cooling ) , @xmath86 for @xmath32 , and @xmath87 for @xmath88 ( heating ) , where @xmath89 and @xmath90 are the initial and steady - state value of @xmath83 , respectively .
the cavity cooling rate constant @xmath91 is given by @xmath92 and the steady state average occupation number @xmath7 is given by @xmath93 \left [ 1 + \left ( \frac{\kappa } { 4 \omega } \right)^2 \right ] \ .\ ] ] for cavity cooling of the @xmath43 motional mode in our experiment , we calculate @xmath94 ( photons are scattered isotropically for a @xmath95 transition ) , and measure independently @xmath96 s@xmath69 .
thus , for our experimental parameters , the heating due to environmental field fluctuations is negligible ( @xmath97 ) and the expression for the steady - state occupation number reduces to eq . ( 1 ) . | we report a demonstration and quantitative characterization of one - dimensional cavity cooling of a single trapped @xmath0sr@xmath1 ion in the resolved sideband regime .
we measure the spectrum of cavity transitions , the rates of cavity heating and cooling , and the steady - state cooling limit .
the cavity cooling dynamics and cooling limit of 22.5(3 ) motional quanta , limited by the moderate coupling between the ion and the cavity , are consistent with a simple model [ phys .
rev .
a * 64 * , 033405 ] without any free parameters , validating the rate equation model for cavity cooling .
cooling atoms , molecules , and even macroscopic objects to the quantum mechanical ground state of their motion represents an important step towards comprehensive control of large or complex quantum mechanical systems @xcite . of particular interest are cooling methods that do not destroy the quantum mechanical coherences in other degrees of freedom .
for example , in trapped ion quantum information processing @xcite , it is necessary to cool the motion of the ions to the ground state without affecting the internal state where the quantum information is stored .
this has been accomplished by sympathetic cooling of the processing ion using a laser cooled ion of another species @xcite .
an all optical method that may allow one to cool a particle without causing decoherence of the internal states is cavity cooling .
when a particle is illuminated with monochromatic laser light , the spectrum of the scattered light is distributed around the incident frequency due to the particle s motion .
if an optical resonator is used to enhance selected portions of the emission spectrum , such that the average emission frequency is larger than that of the incident light , energy conservation implies that the particle is cooled in the scattering process ( cavity cooling ) @xcite . unlike conventional doppler cooling @xcite
, cavity cooling does not require the incident light to be matched to an atomic resonance , and can in principle be performed with non - resonant light .
as light scattered far off resonance carries no information about the atom s internal state @xcite , such a process can potentially cool an atom without destroying a quantum superposition of internal states .
cavity cooling relies on the frequency discrimination provided by the resonator , and hence cooling to the quantum mechanical ground state of the external trapping potential is possible only when the trap frequency @xmath2 exceeds the cavity linewidth @xmath3 ( cavity resolved sideband regime ) @xcite . in the resolved sideband regime cooling
is achieved by tuning the resonator an amount @xmath2 to the blue of the incident light , such that scattering events into the cavity correspond to transitions @xmath4 that lower the motional quantum number @xmath5 .
the steady - state temperature is then set by the competition between cooling by photons scattered into the cavity and recoil heating by photons scattered into free space @xcite . denoting the probability of scattering into a resonant cavity relative to free space by the cooperativity @xmath6 , the steady - state average vibrational quantum number @xmath7 in the resolved sideband regime @xmath8
is given by @xcite @xmath9 , \label{eq : steadystate}\ ] ] where @xmath10 is a dimensionless factor of order unity that depends on the cooling geometry .
thus in the strong - coupling regime @xmath11
cooling to the motional ground state is possible , while for moderate coupling @xmath12 the cooling is limited by the cooperativity to @xmath13 . cavity cooling has been demonstrated in the weak confinement regime @xmath14 with single trapped atoms without directly measuring the atomic temperature @xcite and with atomic ensembles in a different parameter regime of collective coupling @xcite . in this work , using a single trapped @xmath0sr@xmath1 ion in the resolved sideband regime , we measure for the first time cavity heating and cooling rates and the steady - state cooling limit , and observe parameter - free agreement with a rate equation model for cavity cooling @xcite . this experiment ( fig .
[ fig : experimentschematic ] ) builds on the pioneering cavity cooling work with neutral atoms in the weak confinement regime @xcite using the exquisite experimental control and strong confinement of trapped ions @xcite .
a single @xmath0sr@xmath1 ion is confined in a linear rf paul trap with motional frequencies @xmath15 mhz .
the cavity is 5 cm long and has a finesse @xmath16 , resulting in a cavity linewidth ( energy decay rate constant ) @xmath17 khz . the cavity is typically detuned by a few tens of mhz from the s@xmath18 p@xmath19 transition , which has a wavelength @xmath20 nm and a population decay rate @xmath21 mhz .
the optical cavity is oriented along the ion trap axis and the 422 nm cavity cooling laser is perpendicular to the ion trap axis .
the additional 422 nm doppler cooling laser , 1033 nm and 1092 nm repumpers , and 674 nm state preparation and temperature measurement laser are all at 45 degrees to the trap axis such that they have projections along all of the ion motional principal axes .
a 4.1 g magnetic field is applied to define the quantization axis .
the cooperativity at an antinode of the cavity for a two - level atom is @xmath22 , where @xmath23 is the wavenumber and @xmath24 @xmath25 m the radius of the cavity tem@xmath26 mode waist @xcite .
( color online ) schematic of the experimental setup .
a single @xmath0sr@xmath1 ion is confined in the center of a linear rf paul trap and coupled to an optical resonator oriented along the trap axis .
the inset shows the ion energy levels ( solid lines ) and the cavity resonance ( dashed line).,width=324 ] the effective cooperativity is reduced from the cooperativity of a two - level atom @xmath27 by several effects , including the reduced dipole matrix element for @xmath28 polarized light on the @xmath0sr@xmath1 s@xmath18 p@xmath19 transition ( 0.31 ) , the error in positioning the ion in the center of the cavity mode ( 0.89 ) , the non - zero laser linewidth ( 0.82 ) , and the ion s thermal motion ( 0.32 for our typical temperature ) . the combination of these independently measured effects results in a calculated effective cooperativity @xmath29 .
we determine @xmath30 experimentally by comparing the photon scattering rate by the ion into the resonant cavity @xmath31 ( detuning between incident laser and cavity @xmath32 ) to the scattering rate @xmath33 into free space .
the free space scatter is calibrated using a pulsed optical pumping experiment where the 422 nm and 1092 nm lasers are applied in succession and the mean number of 422 nm photons detected during the 1092 nm laser pulses gives the collection efficiency .
fig .
[ fig : cooperativity ] shows the results for several values of the cavity - ion detuning @xmath34 with the ion located at an antinode of the cavity standing wave .
the slope @xmath35 at small values of @xmath33 is the effective cooperativity @xmath36 , consistent with our calculated value . the saturation behavior of fig .
[ fig : cooperativity ] at high free - space scattering rate is due to the finite repumping rate from the metastable d@xmath37 state , which frequency - broadens the scattered light , and reduces the spectral overlap with the cavity mode .
( color online ) photon scattering rate @xmath31 into the resonant cavity ( @xmath32 ) as a function of the scattering rate into free space @xmath33 for several values of cavity - ion detuning @xmath34 .
each data point is measured by preparing the ion at fixed temperature by doppler cooling for 200 @xmath25s , then measuring the photon scattering rates for 50 @xmath25s .
the line is a fit to the form @xmath38 with fit parameters @xmath36 and @xmath39 photons / s.,width=324 ] due to the symmetries of the light - ion interaction , there are selection rules for the motional sideband transitions that depend on the location of the ion in the cavity standing wave @xcite .
the cavity scattering process is a two - photon transition where a photon is absorbed from the traveling - wave cooling laser and another photon emitted into the standing - wave mode supported by the cavity .
the momentum transfer in the scattering process enables the coupling between different ion motional states .
transitions which change the vibrational quantum number along the @xmath40 or @xmath41 directions perpendicular to the resonator ( @xmath42 , enabled by the momentum of the absorbed photon ) are strongest when the ion is located at a cavity standing - wave antinode , since this position corresponds to the strongest ion - cavity coupling . on the other hand , for vibration - changing transitions along @xmath43 that rely on the momentum of the emitted photon
, the symmetry of the standing - wave field allows @xmath44-even processes ( @xmath45 ) to occur only at an antinode , and @xmath44-odd processes ( @xmath46 ) only at a node .
consequently the relative strength of different motional transitions depends on the location of the ion in the cavity , as shown in fig .
[ fig : spectra ] .
when the laser - cavity detuning @xmath47 is varied at fixed cavity - ion detuning @xmath48 mhz , the scattering into the cavity reveals all three first - order motional sideband transitions ( @xmath49 ) , as well as some of the second - order motional sideband transitions .
the best cooling along the @xmath43-direction as investigated here is achieved via the @xmath50 transition when the atom is located at a node of the cavity field .
( color online ) photon scattering rate into the cavity as a function of laser - cavity detuning @xmath47 with the ion located at a cavity standing - wave antinode ( a ) , halfway between a node and an antinode ( b ) , and at a node ( c ) .
each data point is measured by preparing the ion at fixed temperature by doppler cooling for 200 @xmath25s , then measuring the rate of scattering photons from the cavity cooling beam into the cavity for 50 @xmath25s with @xmath48 mhz and @xmath51 photons / s .
the solid , dashed , and dotted vertical lines are at the carrier , first - order motional sideband , and second - order motional sideband transition frequencies , respectively .
the curves are lorentzian fits with linewidths consistent with the combined linewidth of the cavity and laser.,width=324 ] we investigate one - dimensional cavity cooling along the @xmath43 direction by measuring separately the recoil heating rate , as well as the cavity cooling and heating rates for pumping on the cavity red ( @xmath52 ) and blue ( @xmath53 ) motional sidebands , respectively .
to realize a situation that allows simple quantitative comparison with the theoretical model for cavity cooling @xcite , we prepare the ion in its motional ground state by standard sideband cooling on the narrow s@xmath54 d@xmath55 transition .
we then apply the cavity cooling laser for a variable time @xmath56 with detunings @xmath32 or @xmath57 and @xmath58 mhz .
finally , the mean vibrational quantum number @xmath59 is determined by measuring the rabi frequencies of the red and blue motional sidebands of the s@xmath54 d@xmath55 transition @xcite .
this cavity - ion detuning is near the optimum value for conventional doppler cooling , but the geometry of the setup dictates that the cavity cooling laser doppler cools the @xmath40 and @xmath41 motional modes to maintain them at @xmath60 but does not doppler cool the @xmath43 motional mode . the ion position is locked to a node of the cavity standing wave for this measurement by applying dc compensation voltages to the trap electrodes .
the recoil heating rate is the slope @xmath61 for @xmath32 ( fig .
[ fig : coolingdynamics ] green line ) , and the cavity cooling and heating rates are the differences of the slopes @xmath61 for @xmath62 ( fig .
[ fig : coolingdynamics ] red and blue lines ) and for the recoil heating rate .
the signature of cavity cooling is that the temperature after pumping on the cavity red sideband is smaller than the temperature after pumping on the cavity carrier , which is smaller than the temperature after pumping on the cavity blue sideband .
cavity cooling ( @xmath52 ) counteracts recoil heating by free - space scattering , and results in a finite steady - state vibrational quantum number @xmath63 .
( color online ) cavity cooling dynamics . the ion is sideband cooled to the three - dimensional motional ground state ; a cavity cooling pulse with detuning @xmath32 ( carrier ) , @xmath52 ( red axial sideband ) , or @xmath53 ( blue axial sideband ) is applied ; and the mean number of motional quanta in the @xmath43 mode is measured .
the three lines are a simultaneous fit to the model described in the supplementary information with fit parameters @xmath64 , @xmath65 photons / s , and @xmath66 .
the reduced @xmath67 of the fit is 1.7.,width=324 ] we fit the cavity cooling dynamics to a rate equation model parameterized by the initial mean occupation number @xmath68 , the free space scattering rate @xmath33 , and the effective cooperativity @xmath30 @xcite ( see the supplementary information for a description of the model ) . the data in fig .
[ fig : coolingdynamics ] fits the model with fit parameters @xmath64 , @xmath65 s@xmath69 , and @xmath66 .
these values of the fit parameters are consistent with those derived from independent direct measurements , so that the rate equation model is consistent with our data without any free parameters .
the steady - state mean occupation number due to cavity cooling is limited by the relatively small cooperativity of the cavity to @xmath70 , consistent with the value calculated from the model for our parameters . in conclusion ,
we have demonstrated one - dimensional cavity cooling of a single @xmath0sr@xmath1 ion in the resolved sideband regime .
while our small effective cooperativity prevents us from cooling to the motional ground state , we have observed clearly resolved motional sidebands in the cavity emission spectrum , and we have measured the cavity cooling and heating rates .
our results validate the rate equation model proposed by vuleti et al .
@xcite , which predicts that it is possible to cavity cool atoms or ions to the motional ground state without decohering the internal state .
this would require a large detuning of the incident laser compared to the atomic fine structure , a criterion which is easier to meet with light ions such as be@xmath1 @xcite .
resolved sideband cavity cooling might also be used to cool large molecular ions to the motional ground state @xcite . while some species of molecular ions have been previously cooled using sympathetic cooling @xcite
, large mass ratios between the atomic cooling ions and the molecular ions prevent efficient sympathetic cooling of the molecular ions at temperatures near the motional ground state .
resolved sideband cavity cooling could enable exciting new studies of large molecular ions in the quantized motional regime .
we thank s. urabe for providing the linear rf paul trap used in this work and j. simon , m. cetina , and a. grier for advice .
this work is supported in part by the japan science and technology agency , the nsf , and the nsf center for ultracold atoms . |
potentials such as the infinite square well , the harmonic oscillator , the delta function well , and the finite square well are frequently discussed in textbooks@xcite as examples that have bound states . in this paper
we consider solutions of the time independent schrdinger equation@xcite @xmath0 for the gaussian well given by @xmath1 where @xmath2 and @xmath3 . in eq .
@xmath4 is the mass of the particle and @xmath5 is the eigenfunction .
we will obtain an estimate of the ground state energy from a simple variational method and determine that there are a finite number of bound states using the wkb approximation .
we also formulate eq . as a matrix eigenvalue problem , which can be used for any hamiltonian . for simplicity ,
we choose units such that @xmath6 in the following .
consider the hamiltonian @xmath7 for a particle in one dimension subjected to a potential @xmath8 , @xmath9 to demonstrate that @xmath7 possesses a bound state , it is sufficient to construct a real normalized trial function @xmath5 such that the expectation value of @xmath7 for this trial function is negative , that is , @xmath10 this sufficiency condition becomes @xmath11dx.\ ] ] defining @xmath12 to be the width ( or , more precisely the length scale ) of @xmath5 , we can write @xmath13 in the form @xmath14 we substitute this form for @xmath13 in eq . (
[ eq : var ] ) and obtain @xmath15 where @xmath16 ( with @xmath17 ) , is a dimensionless positive number and @xmath18 . if @xmath12 is sufficiently large the second term on the right - hand side of eq .
( [ eq : condition ] ) dominates the first term , and @xmath19 ^ 2 $ ] becomes @xmath20 ^ 2 $ ] ( a positive constant ) , so that , it can be taken out of the integral .
hence , the sign of @xmath21 depends only on the sign of @xmath22 . for both @xmath23 and @xmath21 positive ,
it is not possible to obtain a positive value of @xmath12 for which @xmath24 is negative .
thus , the condition on @xmath8 to have at least one bound state is @xmath25 which depends on the shape of the potential , not on its strength .
equation is a sufficient condition for a potential to have a bound state .
it is not a necessary condition as the example of the harmonic oscillator potential shows .
note that this general criterion for the existence of the bound state follows from a simple argument involving dimensional analysis and the variational principle .
in practice , we can obtain various upper bounds of the ground state energy by calculating @xmath24 for suitably chosen trial functions .
there is a tradeoff between improving the approximate ground state energy using a complex trial wave function and the ease of calculation .
here we choose a normalized gaussian trial wave function @xmath5 with adjustable width , @xmath26 where @xmath27 is related to the width @xmath12 .
the expectation value of the hamiltonian @xmath7 becomes , @xmath28 from eq .
the condition for a minimum is @xmath29 , or @xmath30
.[tab : ground_state]comparison of the ground state energy @xmath24 calculated from the variational principle and the numerical solution ( @xmath31 .
@xmath24 is within 2% of the corresponding numerically evaluated value . [
cols="^,^,^,^,^",options="header " , ] for @xmath32 and @xmath33 finite , @xmath34 is also finite
. therefore , the number of bound states is finite for the gaussian well .
the value of @xmath34 calculated from eq . compared to the numerical solution is in table [ tab : number_state ] .
equation implies that the number of bound states is a function of the ratio @xmath35 , which is a measure of the length scale @xmath12 of the wavefunctions.@xcite because @xmath36 , @xmath8 becomes a delta function in the limit @xmath37 , for which eq .
( [ eq : bound ] ) gives @xmath38 , implying that there exists exactly one bound state for the delta function well.@xcite the reason for the half - quantum number is that the ground state has no nodes .
if @xmath39 , then @xmath40 , which is the case for an infinite square well.@xcite
the one - dimensional schrdinger equation has few analytic solutions , and most problems must be solved numerically .
we first use a taylor s series to obtain a discretization of derivatives for a function @xmath41 and write the second derivative of the function @xmath41 in the three point central difference form @xmath42 where , @xmath43 is the step size . if @xmath44 is the number of mesh points and @xmath45 , @xmath46 are the maximum and minimum value of the variable @xmath47 , the step size @xmath43 is @xmath48 the solution of eq .
is accomplished by discretizing it using eq .
( [ eq : discrete ] ) and evaluating the functions and derivatives at @xmath49 ( for @xmath50 ) , @xmath51 because the solution is expected to decay exponentially outside @xmath45 and @xmath46 , we solve eq . in the interval
@xmath52 $ ] .
these cut - offs and the step size need to be adjusted to obtain the desired accuracy .
equation ( [ eq : sch_dis ] ) is equivalent to the tight - binding approximation applied to a chain of atoms with spacing @xmath43 and one orbital per atom.@xcite we can develop a matrix representation of the schrdinger equation , with @xmath53 in the sub - diagonal and super - diagonal matrix elements and the diagonal elements are @xmath54 .
equation ( [ eq : sch_dis ] ) can then be written as a matrix equation as , @xmath55 equation is a matrix eigenvalue problem with a tridiagonal matrix of dimension @xmath56 ; thus there are @xmath57 eigenvalues . because all the matrix elements are real and the transpose of the matrix is equal to the matrix itself , it is a hermitian matrix , and hence all the eigenvalues are real .
an efficient way of diagonalizing a tridiagonal matrix is to use the standard lapack routine dstevx,@xcite which stores the symmetric tridiagonal matrix in two one - dimensional arrays , one of length @xmath57 containing the diagonal elements , and one of length @xmath58 containing the off - diagonal elements , and returns the eigenvalues along with the eigenfunctions of the matrix .
the energy eigenfunctions and the corresponding eigenvalues for @xmath59 and @xmath60 .
note the unequal spacing between different levels . ] the eigenvalues and corresponding eigenfunctions for the gaussian well with @xmath59 and @xmath60 are shown in fig .
[ fig : wavefunction ] .
the ground state is symmetric with respect to the center of the potential ( even parity ) , and the first excited state is antisymmetric ( odd parity ) .
the wave functions resemble those of the harmonic oscillator,@xcite because for @xmath61 , the dominant term in the expansion of @xmath62 is proportional to @xmath63 .
the virtue of the matrix method is that it can be applied to any potential for which the hamiltonian can be brought into a symmetric tridiagonal or bidiagonal matrix .
however , for systems with periodic boundary conditions , the hamiltonian is no longer tridiagonal and we can not use the simple matrix method .
a gaussian barrier can be constructed by changing the sign of @xmath32 in eq .
( [ eq : potential ] ) . because the condition in eq .
( [ eq : bcondi ] ) is not satisfied , there is no bound state .
the barrier formed in this way is an interesting example of tunneling .
we start with the time energy uncertainty principle @xmath64 denote the energy of the incident particle by @xmath65 .
the uncertainty in the energy is @xmath66 and for sufficiently small @xmath67 the energy of the particle @xmath68 is greater than the height of the barrier @xmath32 .
tunneling will take place if in the time @xmath67 the particle can traverse the barrier .
we take @xmath69 as the width of the barrier and write , @xmath70 from eqs . and
we find @xmath66 to satisfy the equation , @xmath71 the condition for @xmath66 to be real implies that @xmath72 which is the condition for tunneling .
the left - hand side of eq .
is the kinetic energy of the particle obtained from the position - momentum uncertainty relation , with an uncertainty of @xmath73 in the particle s position .
we see that for tunneling to occur , the kinetic energy of the particle must be greater than the difference between the height of the barrier @xmath32 and the total energy @xmath65 .
we now investigate tunneling using the wkb approximation .
consider the classically inaccessible region , @xmath74 , as @xmath75 , so that we can write the transmission coefficient as@xcite @xmath76,\ ] ] where @xmath77 .
we define the opacity of the barrier by @xmath78,\ ] ] so that eq .
( [ eq : trans ] ) becomes @xmath79 .
no attempt been made to obtain an analytical solution to the integral in eq .
( [ eq : trans ] ) .
but we can find an approximate solution of the integral that correctly predicts the transmission coefficient @xmath80 .
we introduce the dimensionless quantity @xmath81 and write the classical turning points as @xmath82 and @xmath83 .
the change in variables , @xmath84 , changes eq .
( [ eq : coeff ] ) to @xmath85}dy\right).\ ] ] dependence of the tansmission coefficient @xmath80 on @xmath86 .
as expected , @xmath80 decreases as @xmath87 increases . ]
the term @xmath88 is always less than unity . if we assume that this term is approximately equal to unity,@xcite then the binomial expansion of @xmath89^{\frac{1}{2}}$ ] to first order we have @xmath90dy\right),\ ] ] which can be further expressed as @xmath91\right),\ ] ] where erf(x ) is the error function of @xmath47 .
equation ( [ eq : theta ] ) implies that @xmath80 depends only on the ratio @xmath86 , a common feature of barrier tunneling .
the dependence of @xmath80 on @xmath87 can be seen from fig .
[ fig : transmission ] .
it is evident that the transmission coefficient decreases as @xmath87 increases .
as expected for @xmath92 , that is , @xmath93 , @xmath94 , implying no reflection from the barrier .
thus far we have discussed different aspects of the potential from a quantum theoretic approach .
the gauusian well ( or , barrier ) is not a long range potential , because it falls off faster than @xmath95 .
however , it has a crucial advantage over the widely used ` finite square well ' potential in that it is continuous throughout the entire range @xmath96 $ ] , whereas the finite square well goes to zero discontinuously . in the following
we describe two applications of this potential .
the ` mean - field ' dynamics of a single nucleon in the field of all other nucleons is the starting point of nuclear many - body theory .
the gaussian well potential is well suited for describing the interaction of a nucleon ( especially , a neutron ) with the heavy nucleus.@xcite the force between them is appreciable only over a very short distance , of the order of @xmath97 m , and in this range the forces are very large compared to forces holding atoms together . keeping in mind the three dimensionality of a nucleus
, we can change our variable from @xmath47 to the radius vector @xmath98 in spherical polar coordinates , so that the potential can be represented by a ` half ' gaussian well , @xmath99 it is monotonically increasing with distance ( i.e. attracting ) and approaches zero very quickly as @xmath44 goes to infinity , reflecting the short - range nature of the nuclear force .
the parameter @xmath33 can be adjusted to the experimental value of the nuclear interaction barrier . from the schrdinger equation in radial form ,
one has @xmath100u = eu,\ ] ] where @xmath101 is the azimuthal quantum number .
it follows immediately that this equation is identical to eq .
( [ eq : sch_tind ] ) for @xmath102 .
it is left as an exercise to the reader to solve this equation using the numerical method discussed already and to deduce the single particle nuclear energy levels for @xmath102 by taking appropriate values of @xmath32 and @xmath33.@xcite for a potential such as the one in eq .
( [ eq : gauss_half ] ) there will be no even - parity states ( because at @xmath103 there must be an antinode . ) .
the gaussian barrier can also be very useful in analyzing a scanning tunneling microscope ( stm ) .
the basic idea behind a stm is quite simple .
an atomically sharp metal tip is brought very close ( @xmath104 ) but without physical contact to a sample surface , and a small bias voltage ( 3 - 6 volts ) is applied between them . if the distance is small enough , an electron ( with an energy 3 - 6 ev ) can tunnel quantum mechanically through the potential barrier developed in between the tip and sample .
this gives rise to the tunneling current that is the result of the overlap between wave functions of the tip - atom and the surface - atom .
assume the metal to tip gap is a gaussian barrier with a height of @xmath105 ev @xmath106 in atomic units .
an electron with energy of @xmath107 ev @xmath108 in atomic units approaches the surface .
now from eq .
( [ eq : trans ] ) , the transmission coefficient is @xmath109 if the tip is @xmath110 nm @xmath111 in atomic units from the surface , we have @xmath112 .
however , it should be noted that for a large distance between the tip and surface , e.g. , if @xmath113 nm @xmath114 in atomic units , @xmath80 will become @xmath115 and practically , there will be no current to measure !
thus the magnitude of the tunneling current is extremely sensitive to the gap between the tip and sample .
the position of the tip in three dimensions is accurately controlled by piezoelectric drivers .
the tip is scanned in two lateral dimensions , while a feedback circuit constantly adjusts the tip height to keep the current constant .
as we measure the current with the tip moving across the surface , information about the atomic nature of the surface can be determined by tracing the path of the tip .
as a final example there is the alpha particle tunneling problem which also can be modeled with the gaussian barrier .
an interesting double well potential may be formed by multiplying @xmath8 in eq . by a factor of @xmath63 .
because the potential @xmath116 is symmetric , the hamiltonian can be brought into a symmetric tridiagonal form . the numerical solution for this potential
is given in fig .
[ fig : double_well1 ] .
because the barrier is finite the wave function extends into both wells .
the ground state wave function is small but nonzero inside the barrier .
assume the particle starts in the right well . from fig .
[ fig : double_well1 ] we see that initially @xmath117 ( where , @xmath118 & @xmath119 are the ground state and first excited state , respectively ) is canceled on the left . as a consequence
the particle will oscillate between the wells with the period , @xmath120 .
thus the tunneling rate depends on the energy difference , @xmath121 , and a double well with a high or wide barrier will have a smaller @xmath66 than one with a low or narrow barrier .
also , @xmath66 will become larger as the energy increases ( that is , as @xmath122 decreases ) .
an example is a quantum well laser , based on the transition that an electron makes between the ground and first excited state of a double gaussian well . by choosing @xmath33 and @xmath32
, we can tune the wavelength of the light emitted .
another not so obvious case could be a qubit .
the quantum bit , or qubit , is the simplest unit of quantum information .
measurements give only two values : ` zero ' or ` one ' .
it is described conveniently by a state in a two level quantum - mechanical system .
a pure qubit state would then be a linear superposition of those two states .
though the double gaussian well has multiple energy levels , it can be seen from fig .
( [ fig : double_well2 ] ) that the relative spacing between the ground and first excited state is very small compared to that between the first and second excited states .
therefore , the lowest two bound states can be effectively decoupled from the other states by choosing appropriate values of @xmath32 & @xmath33 .
an electron making transitions between these two states can easily be considered as a ` charge qubit ' , a much discussed way for consistent quantum data storage .
the author would like to thank sayan chakraborti , swastik bhattacharya , and shamashis sengupta for many helpful suggestions .
he is grateful to professor deepak dhar for help in understanding different aspects of quantum mechanics .
it can be thought as a potential similar to the well known woods - saxon potential .
the schrdinger equation for woods - saxon potential has been solved for @xmath102 to obtain the nuclear energy levels .
see for example , s. flgge , _ practical quantum mechanics _
( springer verlag , berlin , 1999 ) , pp . 162166 . | different features of a potential in the form of a gaussian well have been discussed extensively .
although the details of the calculation are involved , the general approach uses a variational method and wkb approximation , techniques which should be familiar to advanced undergraduates .
a numerical solution of the schrdinger equation through diagonalization has been developed in a self - contained way , and physical applications of the potential are mentioned . |
the pion , being the lightest and simplest particle in the hadronic spectrum has been studied intensively in the past . using a variety of effective and phenemenological models
, the properties of the pion have been desribed with varying succes .
however , these models make assumptions , for example , confinement is put in by hand in contrast to being the result of the underlying dynamics .
lattice qcd ( lqcd ) does nt have this drawback since it is solves qcd directly .
using lqcd , global properties of the pion such as the mass and the decay width have been calculated to satisfying accuracy .
the form factor , which directly reflects the internal structure , is clearly an important challenge .
the first lattice results were obtained by martinelli and sachrajda @xcite .
it was followed by a more detailed study by draper _
@xcite , who showed that the form factor could be described by a simple monopole form as suggested by vector meson dominance @xcite .
we extend @xcite these studies by adopting improved lattice techniques @xcite , which means that we include extra operators in order to systematically eliminate all the @xmath2 discretisation errors .
some aspects of the pion structure have been obtained @xcite using the bethe - salpeter method. we also use this approach and compare its predictions to the results of our direct calculation of the pion form factor . finally we study a chiral extrapolation to reach the physical limit .
to extract the form factor we calculate the two- and three point functions of the pion , analogously to @xcite .
the two point function is given by @xmath3 where @xmath4 is an operator creating a state with the quantum numbers of the pion . by varying the interquark distance at the sink , @xmath5
, we can improve the overlap with the physical pion and obtain information on the bethe - salpeter amplitudes. the three point function is calculated as @xmath6 with @xmath7 the fourth component of the current , inserted at time @xmath8 . since the local current @xmath9 is not conserved on the lattice , one can construct the noether current belonging to our action @xmath10 this current however , still has @xmath2 corrections for @xmath11 .
a conserved and improved current can be constructed @xcite @xmath12 with @xmath13 the bare - quark mass is defined as @xmath14 , where @xmath15 is the kappa value in the chiral limit and @xmath16 denotes the lattice spacing . for our simulation , @xmath17 @xcite .
comparison of the currents will give us information on the importance of improvement .
on a @xmath18 lattice , we generated @xmath19 quenched gluon configurations at @xmath20 .
subsequently , we calculated non - perturbatively improved ( @xmath21 @xcite ) quark propagators for 5 different values of the hopping parameter , @xmath22 .
these propagators were then combined to two- and three point functions for the pion with masses ranging from 360 to 970 mev .
is taken from @xcite . ] for the improved current , we use the parameters @xmath23 , @xmath24 and @xmath25 as determined by bhattacharya _
_ @xcite . to extract the significant parameters from our numerical data ,
we use the following parametrisations . for the two point function
, we have @xmath26 including the contribution of the ground state ( n=0 ) and a first excited one ( n=1 ) .
the @xmath27 denote the matrix elements , @xmath28 and are related to the bethe - salpeter amplitude , @xmath29 .
@xmath30 and @xmath31 are the energies of ground and excited state , respectively ; @xmath32 denotes the interquark distance .
the three - point function is parametrised as @xmath33 the production of pion pairs , wrap around effects being due to the propagation of states beyond @xmath5 and an elastic contribution from the excited state were ignored in our parametrisation after being estimated negligable .
all parameters in the two- and three - point functions - energies @xmath34 , @xmath35-factors and the form factor @xmath36 - were fit simultaneously to the data per configuration .
for the three - point function , we let the current insertion time @xmath8 vary from @xmath37 to @xmath5 .
the value for the parameters and their error in these simultaneous fits are obtained through a single elimination jackknife procedure .
as a byproduct of our simulations we also obtain pion masses for the 5 different @xmath22-values .
they agree with the literature .
we also checked the energie - momentum relation and up to the energies involved we found that a continuum relation provides the best description .
these non - trivial tests indicate that our simulations are done correctly . using the different currents
, we extract the form factor for our five @xmath22-values .
comparison of the results for the noether current and the improved current yields that the effect of improvement can be as large as 25@xmath38 for the highest momenta considered .
the improved results are shown in fig.[fig : ff_vmd ] .
for the five pion masses .
curves : monopole fits to the data._,height=302 ] the high accuracy for the datapoint at @xmath39 is due to the fact that we satisfy the ward - takahashi identity to 1 ppm . as in the previous study @xcite of the pion form factor
we compare our results to a monopole form factor @xmath40 which is suggested by the vector meson dominance ansatz . fitting our data to this model
, we extract a vector meson mass which is within 5@xmath38 of the corresponding rho mass on the lattice @xcite . from the behaviour of the form factor at low @xmath41
, we can extract the mean - square charge radius of the pion , @xmath42 where in the last step we assume eq .
[ eq : vmd ] and use the fitted parameter @xmath43 .
the results are shown in fig .
[ fig : rms ] as a function of the pion mass .
previously , the bethe - salpeter - amplitude @xmath44 has been used to obtain estimates of the charge radius , @xmath45 the results based on this procedure are also shown in fig . [
fig : rms ] . as can be seen these values
are much lower than the actual values obtained from the form factor
. moreover , the bethe - salpeter results are almost mass independent , in accordance with the observations of refs .
this is a known@xcite deficit of the approach , which fig .
[ fig : rms ] makes quantitative .
we extrapolate our results obtained with eq .
[ eq : rms_ff ] using three different parametrisations .
first , we try chiral perturbation theorie ( @xmath46pt ) . at one - loop order the prediction for the radius @xcite is @xmath47 in quenched @xmath46pt
the rms is constant to this order .
there are however indications that a mass dependence appears at higher order @xcite . for our masses
we restrict ourselves to a term linear in @xmath48 @xcite .
lastly , we have also used the vmd prescription and assume a linear dependence of @xmath43 on @xmath48 .
these three expectations are also plotted in fig .
[ fig : rms ] . _ radius of the pion as obtained from bethe - salpeter amplitudes , eq .
[ eq : rms_bsa ] , and from the form factor , eq . [ eq : rms_ff ] . also shown
are three different parametrizations of @xmath49 ( see text ) .
_ , height=302 ]
this work has been done in collaboration with justus koch and edwin laermann .
2 g. martinelli and c.t .
sachrajda , nucl . phys .
* b306 * ( 1988 ) 865 .
t. draper , r.m .
woloshyn , w. wilcox and k. liu , nucl . phys .
* b318 * ( 1989 ) 319 .
oconnell , b.c .
pearce , a.w .
thomas and a.g .
williams , prog . part .
* 39 * ( 1997 ) 201 .
j. van der heide _
et al . , _ , phys . lett .
* b566 * ( 2003 ) 131 .
chu , m. lissia and j.w .
negele , nucl . phys .
* b360 * ( 1991 ) 31 .
m.w . hecht and t.a .
degrand , phys .
* d46 * ( 1992 ) 2155 .
r. gupta , d. daniel and j. grandy , phys .
* d48 * ( 1994 ) 3330 p. lacock _
et al . _ ,
phys . rev .
* d51 * ( 1995 ) 6403 .
e. laermann and p. schmidt , eur .
j * c20 * ( 2001 ) 541 .
m. lscher _
et al . _ ,
nucl . phys .
* b491 * ( 1997 ) 323 .
b. sheikholeslami and r. wohlert , nucl .
b259 * ( 1985 ) 572 .
g. martinelli , c.t .
sachrajda and a. vladikas , nucl . phys .
* b358 * ( 1991 ) 212 .
m. lscher , s. sint , r. sommer and h. wittig , nucl . phys . *
b491 * ( 1997 ) 344 . m. guagnelli and r. sommer , nucl . phys .
proc . suppl . *
63 * ( 1998 ) 886 .
ukqcd collaboration ( k.c .
_ ) , phys . rev .
* d62 * ( 2000 ) 054506 . r.g .
edwards , u.m . heller and t.r .
klassen , nucl . phys .
* b517 * ( 1998 ) 377 .
t. bhattacharya , r. gupta , w. lee and s. sharpe , phys . rev .
* 63 * ( 2001 ) 074505 .
j. gasser and h. leutwyler , ann .
* 158 * ( 1984 ) 142 .
g. colangelo and e. pallante , nucl .
phys . * b520 * ( 1998 ) 433 . g. colangelo , private communication . | we calculate the electromagnetic form factor of the pion in quenched lattice qcd .
the non - perturbatively improved sheikoleslami - wohlert lattice action is used together with the @xmath0 improved current .
we calculate form factor for pion masses down to @xmath1 .
we compare the mean square radius for the pion extracted from our form factors to the value obtained from the bethe salpeter amplitude. using ( quenched ) chiral perturbation theory , we extrapolate our results towards the physical pion mass .
address = national institute for nuclear physics and high - energy physics ( nikhef ) , 1009 db amsterdam , the netherlands , [email protected] |
the last two decades have seen an increasing interest in phase - changing materials ( pcm ) , such as ge - sb - te ( gst ) , as promising candidates for next - generation nonvolatile memory devices @xcite .
these materials display ultrafast and reversible phase transitions between their crystalline and amorphous states at high temperature , while displaying remarkable stability at low temperatures .
their potential originates from the striking differences in electrical and optical properties exhibited by each phase , which can reach several orders of magnitude in some instances @xcite .
these two properties of gst materials reversible kinetics and high contrast in phase properties makes them ideal candidates for memory applications .
generally , a phase transformation is induced by some controlled external heat source that drives the system above the transition threshold .
however , a complete understanding of these fast phase transformations at ultra high heating rates and their remarkable properties is still lacking .
even for single component systems such as ge , the crystalline structures induced by laser - heating of amorphous thin films are very complex and depend on many factors such as the deposition history and morphology of the amorphous structure , the intensity and duration of the heating pulse , and the substrate structure , among others @xcite .
the interplay between all these processes governs the kinetic and thermodynamic evolution of pcm , which must be understood in terms of the microscopic properties of each phase .
modeling and simulation can provide physical insight into the governing phenomena , as well as a path to understanding the physical evolution of pcm under the conditions just discussed .
the goal of this paper is two - fold : ( i ) to present a novel multiscale model for phase transformations in a heat bath at constant temperature ; and ( ii ) to obtain scaling laws for ge - crystal mean radius evolution as a function of the nucleation rate during the recrystallization process .
our approach consists of a thermodynamically consistent phase field model ( tcpf ) that reproduces _ exactly _ the interface energetics and kinetics of atomistically computed crystallization fronts . as an additional feature , the interface thickness may be chosen arbitrarily large while preserving this exact atomistic - to - continuum coupling , thus delivering a highly efficient multiscale computational model .
the paper is organized as follows .
we begin in section [ sec : physicalmodel ] by discussing the physical processes captured by the model ( nucleation , growth , role of anisotropy , etc . ) and the boundary conditions employed . in section [ sec : methods ] we describe the mathematical formulation of the tcpf model , and prove that the energetics and kinetics of planar fronts at constant temperature obey universal relations when expressed in appropriate non - dimensional units .
these relations lead to a direct correspondence between the phase field parameters and physical quantities that can be measured via atomistic simulations , and thus provide with an accurate coupling strategy between the two scales .
additionally , in the same section , we present detailed information on the evaluation of continuum data from atomistic calculations , namely , the free energies for each of the faces , the interface velocities and the transformation pathway between the amorphous and crystalline phase .
next , in section [ sec : results ] we show the results of the atomistic calculations as a function of temperature and interface orientation , compute the invariance solutions of the allen - cahn equations and obtain the related phase field parameters . following these results
, we use the developed multiscale model to analyze , in section [ sec : discussions ] , the interplay between nucleation and growth during crystallization kinetics under isotropic and anisotropic growth conditions .
we conclude the paper in section [ sec : summary ] , with a summary of the results .
when an amorphous ge thin film is heated above the glass transition temperature ( @xmath0 ) but below the melting temperature ( @xmath1 ) , it spontaneously crystallizes .
the difference in free energies between the two phases constitutes the driving force for crystallization , which decreases with temperature . at the same time
, the mobilities and nucleation rates rapidly increase as the temperature rises . for most pcms
, there is an intermediate temperature at which an optimal compromise is achieved , and the crystallization rate is maximized . for ge ,
such temperature is around @xmath2 k , cf .
[ sec : results_md ] , which can be achieved , for instance , by high - intensity pulsed laser heating @xcite . under those conditions ,
crystals nucleate and grow until full impingement is achieved .
complete crystallization occurs over tens of nanoseconds , and the resulting crystals have a mean radius of tens of nanometers @xcite .
experiments suggest that nucleation occurs homogeneously in space throughout the thin film with no preferential texture , which typically indicates heterogeneous nucleation at uniformly distributed imperfections and/or irregularities in the amorphous microstructure . from a theoretical point of view
, nucleation is ill - defined despite substantial efforts @xcite , and here the nucleation rate is simply taken taken as a parameter of the model , which can be changed to understand its effect on the overall crystallization process .
once crystals have nucleated , they grow by front propagation in the surrounding amorphous material , or , if in contact with another crystal , by grain boundary motion . in this paper , we focus solely on the former , and leave the study of coarsening of the recrystallized nanograined material for future studies . the mesoscopic time and length scales involved in the crystallization process ( microns and tens of nanoseconds ) are prohibitive for fully atomistic simulations .
hence , in this paper we adopt a computational approach based on the phase field equations parameterized with atomistic data .
the details of the method are described in the following section . in the remainder of the paper
it will be assumed that the film is thin enough for critical nuclei to span the full thickness and for the problem to be treated as two - dimensional .
furthermore we will consider that the amorphous thin film is free - standing and capable of fully releasing stresses via bending .
the phase - field method is a computational approach widely used in modeling morphological and microstructural evolution in materials @xcite , including solidification @xcite , fracture @xcite , dislocation dynamics @xcite , crystal nucleation @xcite and growth @xcite , and diffusion processes @xcite , among others .
the method uses a so - called phase - field variable , @xmath3 , to distinguish among the different material phases and implicitly track interface motion .
for example , in this work we take @xmath4 to represent the crystalline phase , and @xmath5 for the amorphous region .
the interface between the two phases is then described as a smooth transition of the variable @xmath3 through a small thickness . in phase field models , the equilibrium configuration of a material point in homogeneous state is assumed to be characterized by a gibbs free energy @xmath6 that depends on the phase field variable and a set of thermodynamic state variables such as temperature and pressure .
the total free energy of the system is then described via a landau - ginzburg free energy functional of the form @xcite @xmath7 d\omega,\ ] ] where the second term accounts for the additional contribution due to the presence of the interface , which is characterized by a non - vanishing value of @xmath8 .
the variable @xmath9 in eq .
[ eq : totalenergy ] sets the interfacial energy and it is a function of the state variables and the misorientation @xmath10 , defined as the relative angle between the crystal lattice @xmath11 and the normal to the interface , i.e. @xmath12 . the existence of a total free energy of the form just described is a standard postulate in transport processes in continuum mechanics @xcite . for the system of interest in this paper , we will assume in the following that the only thermodynamic variable dependence is with the temperature , which is assumed to be controlled by thermostatting to a heat bath .
the system s time evolution is then modeled using allen - cahn s equation @xcite @xmath13 , \end{split}\ ] ] where @xmath14 is the mobility on @xmath15 .
this simple form of the kinetic relation at constant temperature guarantees a decrease of total free energy of the system @xmath16 $ ] along the solution path , in accordance to the second law of thermodynamics .
we further assume that the local free energy for intermediate values of @xmath3 can be computed from the amorphous free energy @xmath17 and the crystalline one @xmath18 as @xmath19 with @xmath20 @xmath21 is the driving force for the transformation and @xmath22 represents the energy barrier along the transformation path @xmath23 .
this functional form guarantees that the free energy has two local minima at all temperatures at @xmath24 and @xmath25 , which correspond respectively to the pure amorphous and crystalline phases . the kinetic equation ( [ eq : general_pf ] )
can then be written as @xmath26 or equivalently , as @xmath27 where @xmath28 and @xmath29 has units of length , @xmath30 has units of an inverse diffusivity , and @xmath31 is non - dimensional .
equation ( [ eq : phi_constantt ] ) determines the growth of existing nuclei .
nucleation , for its part , is treated explicitly in this work by introducing cylindrical critical nuclei in the simulation with random orientation and with a probability that is consistent with a certain nucleation rate .
it is assumed that the nuclei span the entire thickness of the film , so that their critical radius @xmath32 in 2d can be estimated via classical nucleation theory as @xmath33 where @xmath34 is the interfacial free energy .
the evolution of a flat crystalline - amorphous interface at constant temperature obeys an invariance relation that we proceed to characterize . to that end , consider a set of parameters @xmath29 and @xmath30 and the corresponding ones of a reference problem ( @xmath35 , @xmath36 ) .
the solutions characterizing the motion of the front for both sets of parameters are denoted by @xmath3 and @xmath37 respectively , and are given by eq .
( [ eq : phi_constantt ] ) @xmath38 these equations lead to a planar front moving at constant velocity in steady state , that we call @xmath39 and @xmath40 respectively . with the change of variables @xmath41 eqs .
( [ eq : phi_phi0_1 ] ) transform into the following ordinary differential equations @xmath42 where @xmath43 represents the gradient with respect to coordinates @xmath44 .
if we now define @xmath45 and denote @xmath46 , then the two equations become @xmath47 which are identical for @xmath48 . in other words ,
the non - dimensional steady state velocity ( @xmath49 ) satisfies an invariance relation of the form @xmath50 similarly , the non - dimensional interface energy @xmath51 can be shown to satisfy @xmath52 this relation can be obtained by defining the interface energy as the difference between the sum of energies of the bulk systems and the energy of the two - phase configuration separated by a sharp interface . choosing a reference frame that moves with the interface steady state velocity ,
the surface energy reads @xmath53 dx^ * \\ & \phantom{=}- \frac{h_{\ell}}{2}(g_a(t)+g_c(t ) ) , \end{split}\ ] ] where @xmath54 is the range of the coordinate orthogonal to the interface and @xmath55 represents the system length in such direction . without loss of generality , we consider that the interface is at the center of the domain under consideration .
then , at constant temperature , @xmath56 \\ & \phantom{=}+ b(t ) \bigg [ \int_{\ell}g_{dw}(\phi^*)dx^*\bigg ] \\ & \phantom{=}+ \delta g(t ) \bigg[\frac{h_{\ell}}{2 } - \int_{\ell}q(\phi^*)dx^*\bigg ]
. \end{split}\ ] ] by recourse to the changes of variable expressed by eq .
( [ eq : variablechange ] ) and the equivalence @xmath57 , this equation can be rewritten as @xmath58 \\ & \phantom{=}+ \frac{1}{\epsilon_0 } \bigg [ \int_{\ell_0}g_{dw}(\phi^*_0 ) d x_0^*\bigg ] \\ & \phantom{=}+ \frac{\delta \bar{g}(t,\theta)}{\epsilon_0 } \bigg[\frac{h_{\ell _ 0}}{2 } - \int_{\ell_0}q(\phi^*_0 ) d x_0^*\bigg ] \\ & = i_g(\delta \bar{g}(t,\theta ) ) \end{split}\ ] ] and the sought - after result is obtained . the phase field model at constant temperature described by eq . ( [ eq : phi_constantt ] ) requires that the following magnitudes be defined : @xmath59 , @xmath60 , @xmath61 and @xmath22 . in light of the invariance relations
previously derived , these parameters can be obtained directly via atomistic simulations of the following kind : 1 .
free energy calculations using thermodynamic integration of bulk crystalline and amorphous phases at zero pressure and constant temperatures . these yield @xmath18 and @xmath17 .
transition energy barrier between the amorphous and crystalline states at zero pressure and constant temperature . to obtain @xmath22
we equate the transition path to an activated process described by a general configurational nonlinear many - body reaction coordinate@xcite .
free energy calculations at zero pressure and constant temperature to compute @xmath62 . as well , molecular dynamics simulations of moving fronts in steady state to extract interface velocities @xmath63 . equations ( [ eq : v_int ] ) and ( [ eq : f_int ] ) allow to explicitly relate the atomistic data with the phase field parameters @xmath60 and @xmath61 as @xmath64 the details of such computations for the different subsystems are described in the following section . to compute the free energies of bulk phases we use thermodynamic integration .
the helmholtz free energy is defined as @xmath65 where @xmath66 and @xmath67 are the internal energy and the entropy .
the approach employed here to compute @xmath68 for solid systems involves two steps .
first , a free energy @xmath69 is obtained at a temperature @xmath70 in each case .
second , the internal energy is computed for the desired range of temperatures as @xmath71 where @xmath72 and @xmath73 are the potential and kinetic energies , and , under the usual assumption of ergodicity , @xmath74 denotes time average . the free energy at a given temperature @xmath75
can then be obtained by solving the gibbs - helmholtz integral @xmath76 where @xmath77 is made to be a smooth integrable function that expresses the temperature dependence of the internal energy . to obtain @xmath69
, we employ the @xmath78-integration technique .
the use of this technique is well documented in the literature @xcite and here we only provide a brief overview .
it is assumed that the system at @xmath70 is represented by the following energy function @xmath79 where @xmath80 is a constant , and @xmath81 and @xmath82 are , respectively , the potential energy functions of the system under study and a reference system whose free energy is known . for practical reasons ,
@xmath83 is typically taken to be the potential energy function of an analytical or a numerical system of coupled harmonic oscillators . from here
, @xmath69 is straightforwardly obtained by integrating along a @xmath78 trajectory between 0 and 1 @xmath84 d\lambda . & \end{aligned } \label{fu}\ ] ] in this work , the calculation hypotheses warranted the use of linear thermodynamic integration ( @xmath85 ) .
the numerical behavior of eq .
( [ fu ] ) under other choices of @xmath86 has been reviewed in the literarure @xcite , and we refer the reader to those works for more details . when internal transport processes derived from the existence of viscous flow , _
e.g. _ as in liquids , are important , eq . ( [ fu ] ) can no longer be used in conjunction with a reference harmonic crystal to compute @xmath69 .
in such cases , the free energy is obtained as @xcite @xmath87 , & \end{aligned } \label{liquid}\ ] ] where @xmath88 is the pressure , @xmath89 is the system s density at a temperature @xmath70 above the supercritical temperature ( here we have taken @xmath90 k ) and @xmath91 is the free energy of an ideal gas at density and temperature @xmath92 and @xmath75 @xcite .
the integrand in the second term of the r.h.s . of eq .
( [ liquid ] ) represents an isothermal expansion from @xmath89 to zero density ( _ i.e. _ infinite volume ) , where the system effectively behaves as an ideal gas .
this expansion should be reversible , which means that no first - order transition e.g .
liquid - gas should be traversed . for its part
, the equation of state @xmath93 is obtained from a set of canonical ensemble calculations of liquid ge at @xmath70 .
this is shown in fig .
[ fig : ideal ] .
the resulting data are fitted to a 3@xmath94 degree polynomial and the integral in eq .
( [ liquid ] ) is solved to yield the free energies . to establish the limits of applicability of eqs .
( [ fu ] ) and ( [ liquid ] ) , we compute the self - diffusivity of amorphous ge as a function of temperature .
results are shown in figure [ fig : diff ] .
our calculations clearly demonstrate that internal transport is important only above 900 k , at which point eq .
( [ liquid ] ) is the pertinent one for computing free energies . for ideal ( perfectly - flat , stress - free ) interfaces ,
the interfacial free energy @xmath95 can be defined as the change in free energy due to an increase in interface surface area @xmath96 @xmath97 from a practical standpoint , interfacial free energies are computed using the dupr equation @xmath98 where @xmath99 is the surface free energy of solid @xmath100 in a vacuum and @xmath101 is the so - called work of _ adhesion_. the above formula results from a two - step construction in which ( i ) slabs with exposed area @xmath96 are created in a vacuum from bulk systems @xmath100 and @xmath102 , and ( ii ) these new surfaces are brought into contact , releasing an amount of work @xmath103 . using the bulk free energy densities , @xmath104 ,
computed in section [ subsec : bulk ] by thermodynamic integration , we can express eq .
( [ dupre ] ) as @xmath105 where @xmath106 is the free energy of a system with total volume @xmath107 of a simulation cell containing a volume @xmath108 of phase @xmath100 and a volume @xmath109 of phase @xmath102 separated by the interface . as for the bulk phases ,
@xmath106 is computed using @xmath78-integration .
the calculation is simplified due to the fact that both phases correspond to monoatomic solid systems and thus the same reference thermodynamic system can be used for both . because cohesion in crystalline solids is anisotropic while in amorphous solids or liquids is isotropic , we have repeated this procedure for a number of different surface orientations to assess the orientation dependence of the interfacial free energy .
as a final point , we note that , in reality , atomistic interfaces are intrinsically rough and accurate methods for computing interfacial free energies that account for such atomic - scale roughness have been developed @xcite . to calculate interface mobilities , large - scale md simulation of crystal - amorphous mixing are performed .
for this , a bicrystal with the desired interface orientation is constructed and the system is left to evolve at a given temperature . the temperature must be kept constant on the boundaries of this box using some suitable themostat .
it is important to ensure that no temperature control is applied in the interface region , as this could result in thermodynamic artifacts .
following hoyt and song @xcite we measure the rate of change in potential energy as the front advances , @xmath110 , and obtain the velocity as @xmath111 where @xmath96 is again the interface surface area , @xmath112 is the volume per atom in the crystalline phase at the temperature of interest , and @xmath113 is the latent heat , which in this case is simply equal to the internal energy difference between the two phases . as above , the factor of two in the above expression reflects the use of periodic boundary conditions , which introduces two interfaces in the simulation volume .
next , we discuss the most salient features of the potential energy function @xmath81 employed here .
we use a stillinger - weber potential parameterized by posselt and gabriel for ge@xcite .
the potential builds on the parameterization by ding and anderson@xcite , which yields the correct values for the cohesive energy and lattice constant .
the elastic constants , melting point , as well as the formation and migration energies of point defects for diamond - structure ge are reproduced within certain limits .
the energetics of other crystalline phases and the structure of the liquid are described reasonably well .
all the simulations were run in the isobaric - isothermal ensamble @xmath114 where @xmath115 is the number of particles , @xmath88 is the pressure , and @xmath75 the abolute temperature using periodic boundary conditions in three dimensions . the pressure and temperature were controlled using a nos - hoover thermostat with a damping constant of 0.1 ps in both cases .
first , we calculate the free energy of the crystalline and amorphous phases as a function of temperature .
figure [ fig : freeenergy_temperature ] shows the results for each case , with the crystalline phase being the more stable one up to a temperature of 1350 k. this value can be considered as the melting point for the numerical force field @xmath81 employed here , about 10% larger than the experimental value of 1210 k @xcite .
figure [ fig : volume_temperature.pdf ] shows the variation of the atomic density @xmath116 and the thermal expansion coefficient @xmath117 with of temperature . as the figure shows , the amorphous structure is always approximately 10% denser than the crystalline phase , and is seen to undergo a smooth transition at @xmath118800 k. this temperature correlates well with the point at which the self diffusivity is markedly nonzero ( cf .
[ fig : diff ] ) , and can be assigned to the reversible liquid - glass transition .
the temperature at which we have recorded this transition is in excellent agreement with experimental measurements of the glass transition temperature for ge , @xmath119 k @xcite .
this glass - liquid transition manifests itself as a marked dip in the value of @xmath117 .
the glass transition is not itself a phase transition , but a laboratory phenomenon , and depends on the cooling rate and the boundary conditions .
our calculations were obtained for systems containing 13824 atoms at a cooling rate of 10 k per ps from the liquid melt with 100 ps annealings every 100 k. next , we construct a bilayer system consisting of crystalline and amorphous half crystals to measure interface properties .
we calculate interfacial free energies and interface velocities as a function of temperature . to ascertain the orientation dependence of the crystal - amorphous interface , three different crystallographic directions
have been considered , namely [ 100 ] , [ 110 ] and the [ 111 ] .
figure [ fig : interfaceenergy_temperature ] reveals the temperature evolution of the interfacial free energy , which is self - consistent in that it vanishes at the melting point . the @xmath120 $ ] orientation is the most stable one at low temperatures , followed by the @xmath121 $ ] and the @xmath122 $ ] directions . at the glass transition temperature , however , all interfacial free energies converge , likely due to a ` wetting ' effect of the crystalline surface in contact with a liquid - like system . for their part
, the interface velocities also display a behavior qualitatively similar to the self - diffusion coefficient for intermediate temperatures ( cf .
[ fig : diff ] ) .
this amounts to practically zero mobility below @xmath0 and a monotonic increase above it .
for very large temperatures though , the driving force decreases till vanishing at the melting temperature , and induces a decrease in the velocity .
the compromise between driving force and mobility leads to a peak in the interface velocity around @xmath2 k. the last parameter to be computed from atomistic simulations is the energy barrier along the crystal - amorphous transformation path , @xmath22 , introduced in section [ phase ] .
this calculation involves defining a suitable reaction coordinate between the initial amorphous and final crystalline states .
a natural choice of initial and final states is to start from a crystalline specimen and melt it to the liquid state , followed by a sensible cooling into the amorphous structure .
however , in most constrained minimization techniques used for this purpose e.g . nudged elastic band or drag methods , the reaction coordinate vector is constructed as a linear interpolation of all degrees of freedom within each end state . due to the sizable amount of mixing and thermal transport occurring during the melting and cooling phases
, however , we have seen that this construction leads to large variations in the reaction coordinate that make convergence difficult ( and the physical meaning of the procedure questionable ) . instead
, we proceed to use the bilayer geometry described above and let the interface propagate spontaneously turning the amorphous half system into a crystalline one . once the conversion is complete ,
we extract a volumetric sample from within the converted region and use it as one of the end states , while using the corresponding atoms contained within that volume in the amorphous region as the second end state .
replicas for constrained minimization are generated by linear interpolation between both end states .
this procedure establishes a direct link between both structures while minimizing the amount of internal transport , providing a more physical transition pathway .
figure [ fig : bt ] shows the relaxed minimum energy path at 0 k for a system containing 13824 atoms and 16 replicas .
the static energy barrier is approximately @xmath123 ev per atom .
finding the temperature dependence intrinsic in @xmath22 is not trivial and requires using elaborate thermodynamic integration techniques ( _ e.g. _ refs .
@xcite ) . here , for lack of a better estimate
, we assume that @xmath124
. at 0 k. the roughness in the energy path is a consequence of the multidimensional nature of the transition , with many atoms undergoing transformations at once.,title="fig:",scaledwidth=48.0% ] [ fig : bt ] the evolution equation for the phase field variable , c.f .
( [ eq : phi_constantt ] ) , is numerically solved in this work by means of a semi - implicit , backward euler discretization scheme in time @xmath125 where @xmath126 is the time step .
the resulting ordinary differential equation is then solved in space via a first order finite difference scheme followed by a discrete fourier transform ( ` fftw ` package @xcite ) .
the grid size is denoted by @xmath127 . in this section
we illustrate the convergence of the numerical scheme to known analytical solutions and demonstrate good agreement with the theoretical spatial and temporal convergence rates .
we first consider a planar front separating a two phase region at constant temperature with an energy landscape described by a perfect double - well potential , i.e. with @xmath128 .
the equilibrium solution of this system is characterized by eq .
( [ eq : phi_constantt ] ) , that is , @xmath129 this equation is analytically soluble and gives a static interface with the following profile @xmath130 where the @xmath131 axis is centered at the interface and orthogonal to it . for this solution
, @xmath29 exactly corresponds to the interface thickness , measured as the support of @xmath132 .
figure [ fig:1d_doublewell_energyerror ] shows the error with respect to the analytic solution with the following energy measure @xmath133 and @xmath134 , @xmath135 , @xmath136.,title="fig:",scaledwidth=45.0% ] we observe a quadratic convergence of the energy as it is expected from the first order finite difference scheme used in these simulations . a difference in free energy between the two phases ( @xmath137 ) will induce a planar interface to move until a constant velocity is reached , as discussed in sec .
[ sec : methods ] .
figure [ fig:1d_inclineddoublewell_frontvelocity_dt ] shows the value of the numerical steady state velocity as a function of the time step for a constant ( small ) mesh size .
as expected from a first order time scheme , the average velocity converges linearly with the time step .
however , if the mesh size used for the spacial discretization is too large , the numerical solution does not lead to a steady state velocity .
rather , the underlying mesh acts as a periodic potential inducing oscillations in the steady state numerical solution , c.f .
[ fig:1d_inclineddoublewell_frontvelocity_ne ] . additionally , the mean velocity decreases with the mesh size for a given constant time step , and can even vanish for very large mesh sizes , c.f .
[ fig:1d_inclineddoublewell_frontvelocity_multiplene ] .
this behavior , previously reported by other authors @xcite , is characteristic of the scheme used for solving the ordinary differential equation at a given constant time step .
equivalent simulations using finite elements indicate that the average steady state velocity may increase with the mesh size for sufficiently large driving forces .
, @xmath135 , @xmath138 , @xmath139,title="fig:",scaledwidth=45.0% ] the functions @xmath140 and @xmath141 described in eqs .
( [ eq : v_int ] ) and ( [ eq : f_int ] ) , can now be computed with a controlled numerical error lower than 1@xmath142 . in figures [ fig : invariancerelation_velocity ] and [ fig : invariancerelation_interfaceenergy ] we plot each invariant for different values of the phase field parameters . for small values of @xmath143 , @xmath140 and @xmath141
can be approximated by @xmath144 the linear dependence given in eq .
( [ eq : firstinvarance ] ) between the interface velocity and the free energy difference is often assumed in the literature @xcite . for its part ,
( [ eq : secondinvarance ] ) is the result of the equipartition of the interface energy at zero velocity , as shown in appendix [ sec : appendix_a ] .
equations ( [ eq : m ] ) and ( [ eq : tau ] ) provide a direct recipe for computing @xmath61 and @xmath60 from the functions @xmath140 and @xmath141 .
these are shown in figs .
[ fig : m_temperature ] and [ fig : mobility_temperature ] for different values of @xmath145 .
equivalently , the values of @xmath146 and @xmath147 , defined in eqs .
( [ eq : epsilon ] ) and ( [ eq : alpha ] ) , are shown in figs .
[ fig : epsilon_temperature ] and [ fig : alpha_temperature ] respectively . as
the figures and equations show , the phase field parameters can be adjusted to deliver the appropriate interface energy and front velocity at constant temperature for an arbitrary value of the free energy barrier @xmath22 . in view of this observation , and the computational cost associated with the computation of @xmath22 , the value obtained in sec .
[ sec : results_md ] is used .
other strategies for selecting @xmath145 based solely on computational cost , i.e. fixing the interface thickness , may also be used . in such case ,
the velocity and free energy of the interface would be exactly captured by the phase field model with a finite and potentially thick interface .
next , the multiscale model developed in the previous sections is used to examine the interplay between the nucleation and growth mechanisms in the crystallization process . for verification purposes ,
first we examine the ideal case of isotropic growth and demonstrate its consistency with the avrami relations . to that end
, we consider a temperature of 1100 k , nucleation rates that span several orders of magnitude , and growth conditions dictated by the phase field parameters previously computed for the three surface orientations . in these isotropic calculations ,
the parameter @xmath10 is simply a differentiating property for each growth condition .
we begin by rendering eq .
( [ eq : phi_constantt ] ) dimensionless with the following characteristic length and timescale @xmath148 where @xmath149 is the radius of the critical nuclei , given by eq .
( [ eq : criticalradius ] ) .
this specific choice of length and time scales has the property of delivering non - dimensional parameters @xmath29 and @xmath30 that are independent of @xmath10 . in particular , using eq .
( [ eq : m ] ) , @xmath150 consequently , the resulting non - dimensional phase field equation @xmath151 allows to examine all growth conditions with a single equation , which we proceed to couple with an explicit nucleation strategy , with non - dimensional nucleation rates ranging between @xmath152 and @xmath153 .
the numerical results attendant to this coupled nucleation and growth process are shown in fig .
[ fig : radiusevolution_adimensional ] , where the time evolution of the mean radius during the nanocrystallization process is represented .
these curves may be normalized with the time and mean radius at @xmath154 crystallization , delivering a universal crystallizatian curve , as shown in fig .
[ fig : radiusevolution_adim_normalized ] . for their part , the time required to complete @xmath154 of the crystallization , and the mean radius at that stage of crystallization , satisfy very simple relations with the nucleation rate as depicted in figs .
[ fig : time_095_j_adim ] and [ fig : radius_095_j_adim ] .
the slope of the fit to the numerical points in both curves coincide with the exponents of avrami s relations , as detailed in appendix [ sec : appendix_b ] .
next , we proceed to examine the case of anisotropic growth at @xmath155 k using the surface orientation dependent data for the phase field parameters . in particular , we use a linear interpolation scheme between the @xmath121 $ ] , @xmath120 $ ] and @xmath122 $ ] data previously computed , to define @xmath156 and @xmath157 for an arbitrary value of the misorientation angle @xmath158 , c.f .
[ fig : theta_interpolation ] .
the specific spacing between the three data points has been chosen proportional to the number of distinct @xmath159 , @xmath160 and @xmath161 orientations in a diamond cell structure ( 4 , 3 and 6 respectively ) . and
@xmath30 for a given misorientation @xmath158 will be obtained by linear interpolation of the values for orientations [ 111 ] , [ 100 ] and [ 110 ] spaced as shown in the figure.,title="fig:",scaledwidth=45.0% ] the resulting curves for the mean radius evolution as a function of time for different non - dimensional nucleation rates follow in a very similar manner to the isotropic case , c.f .
[ fig : tr_anisotropic ] . despite the strong growth anisotropy exhibited by ge , c.f .
[ sec : results_md ] , the time for crystallization and the mean radius of the final microstructure also follow fairly closely avrami s relations , c.f . fig .
[ fig : jt_anisotropic ] and [ fig : jr_anisotropic ] .
the physical reasoning behind this result lies on the fact that the final microstructure does not exhibit grains with a large aspect ratio , c.f .
[ fig : final_microstructure_j_anisotropic ] .. the homogenous nucleation creates constraints to the growth of preexisting nuclei that are , on average , equi - spaced , leading to anisotropic ratios of the final microstructure similar to the simulations of isotropic growth , c.f .
[ fig : ar_evolution ] .
the measure of anisotropy used in these calculations is @xmath162 where @xmath163 are the eigenvalues of the gyration tensor of each grain , see fig .
[ fig : ar ] for several example of ar .
the results shown in this paper indicate that the ge nanocrystalline structure resulting from highly anisotropic growth and homogenous nucleation is fairly isotropic in the range of nucleation rates examined . as a result ,
the relations between the crystallization time , final average mean radius and nucleation rate follow fairly closely avrami s relations .
these relations could be potentially used to extract nucleation rates from experimental measurements of time - to - crystallization and grain size or vice versa .
however , it must be kept in mind that the nucleation rates used here for numerical convenience may differ from those found in laboratory experiments depending on the experimental conditions @xcite , and any exercise of extrapolation for the anisotropic curves is not backed by our model . the numerical strategy chosen to obtain these results
consists on phase field simulations that exactly deliver the interface kinetics and energetics obtained via atomistic calculations .
this exact mapping between the atomistic and continuum solutions is based on two invariance relations of the allen - cahn equations at constant temperature .
more precisely , we show in this paper that the non - dimensional steady state velocity and interfacial free energy of planar crystallization fronts satisfy universal relations with the non - dimensional free energy difference between the crystalline and amorphous phase .
these two relations allow for an explicit parametrization of the phase field equations with the atomistic data calculated from molecular dynamic simulations in suitably constructed subsystems using state - of - the - art interatomic potentials .
consider a flat crystalline - amorphous interface with an energy landscape described by a perfect double - well potential , i.e. with @xmath128 .
the equilibrium solution consists of a static interface that satisfies @xmath164 for this specific case , the interface profile is also the minimizer of the interface free energy functional , c.f .
( [ eq : gint ] ) , @xmath165 + b \bigg [ \int_{\ell}g_{dw}(\phi)dx\bigg],\ ] ] where @xmath131 to be the coordinate orthogonal to the interface . then , the integrand of this functional can be seen as a lagrangian , that exclusively depends on @xmath3 and @xmath166 and does not depend explicitly on the variable @xmath131 , @xmath167 as a result , by noether s theorems , the corresponding hamiltonian ( legendre transform of the lagrangian ) is independent of the variable @xmath131 .
more precisely , @xmath168 or equivalently , @xmath169 the constant may be evaluated far away from the interface , where @xmath3 is either @xmath170 or @xmath171 , delivering a zero value . as a result , the interface energy is partitioned equally between the two terms in eq .
( [ eq : g_int_still ] ) , @xmath172 = b \bigg [ \int_{\ell}g_{dw}(\phi)dx\bigg].\ ] ] the value of either of these two integrals may be evaluated for the equilibrium solution , c.f .
( [ eq : analytic_phi ] ) , giving @xmath173 we further remark that the value of the integrals in eq .
( [ eq : equipartition ] ) are not very sensitive to the specific interface profile .
a simple piecewise linear profile with interface thickness @xmath29 , i.e. @xmath174 for @xmath175 , would deliver @xmath176
avrami s equation @xcite describes the kinetics of crystallization under the assumption of homogeneous nucleation , constant nucleation rate and constant growth rate for the radius of each nuclei independently of its size . for a two dimensional system
, it reads @xmath177 where @xmath178 is the volume fraction of the crystalline phase and @xmath179 is the growth rate .
this equation immediatly delivers the relation between the time required to acheive @xmath154 crystallization @xmath180 and the nucleation rate .
namely , it is of the form @xmath181 where @xmath182 . on its side , the mean radius at @xmath154 crystallization ( @xmath183 ) satisfies @xmath184 where @xmath72 is the total volume and @xmath185 is the number of nuclei composing the nanocrystalline structure .
the later , can be computed as @xmath186 by use of the change of variables @xmath187 , eq .
[ eq : n ] simplifies to @xmath188 with @xmath189 solely dependent on @xmath179 .
then , the mean radius at crystallization can be expressed as @xmath190 or @xmath191
this work performed under the auspices of the u.s .
department of energy by lawrence livermore national laboratory under contract de ac52 - 07na27344 . c. reina acknowledges support from the lawrence fellowship program at lawrence livermore national laboratory .
j. marian acknowledges support from doe s early career research program .
liliya nikolova , thomas lagrange , mark j. stern , jennifer m. macleod , bryan w. reed , heide ibrahim , geoffrey h. campbell , federico rosei , and bradley j. siwick .
complex crystallization dynamics in amorphous germanium observed with dynamic transmission electron microscopy . , 87:064105 , feb 2013 .
l. nikolova , t. lagrange , b.w .
reed , m. j. stern , n. d. browning , g.h .
campbell , j. c kieffer , b.j .
siwick , and f. rosei .
nanocrystallization of amorphous germanium films observed with nanosecond temporal resolution . , 97(20):2031022031023 , 2010 .
christian miehe , martina hofacker , and fabian welschinger . a phase field model for rate - independent crack propagation : robust algorithmic implementation based on operator splits .
, 199(4548):2765 2778 , 2010 . | germanium is the base element in many phase - change materials ,
_ i.e. _ systems that can undergo reversible transformations between their crystalline and amorphous phases .
they are widely used in current digital electronics and hold great promise for the next generation of non - volatile memory devices . however , the ultra fast phase transformations required for these applications can be exceedingly complex even for single component systems , and a full physical understanding of these phenomena is still lacking . in this paper
we study nucleation and growth of crystalline ge from amorphous thin films at high temperature using phase field models informed by atomistic calculations of fundamental material properties .
the atomistic calculations capture the full anisotropy of the ge crystal lattice , which results in orientation dependences for interfacial energies and mobilities .
these orientation relations are then exactly recovered by the phase field model at finite thickness via a novel parametrization strategy based on invariance solutions of the allen - cahn equations . by means of this multiscale approach , we study the interplay between nucleation and growth and find that the relation between the mean radius of the crystallized ge grains and the nucleation rate follows simple avrami - type scaling laws .
we argue that these can be used to cover a wide region of the nucleation rate space , hence facilitating comparison with experiments . |
neutron induced reactions on @xmath0ar occur in the weak component of the s - process , where the most relevant neutron energies are situated in the kev range . in order to calculate the maxwellian averaged cross section in this energy region
, the reactions under investigation have to be studied with thermal as well as with resonance neutrons . the thermal value is often used to normalise the higher energy values and besides it is needed in the calculation of the maxwellian averaged cross section as a summation of the thermal and resonance components@xcite .
samples with a well - defined mass are of great importance to perform reliable reaction cross section measurements , so a lot of effort was put in the preparation and characterisation of suited samples . a detailed description of the procedure is given in@xcite .
@xmath0ar atoms were produced via the @xmath0cl(p , n)@xmath0ar reaction by bombarding a nacl target with 30 mev protons from a cyclotron of the ucl at louvain - la - neuve ( belgium ) .
these atoms were then ionised to the 1@xmath8 state and implanted at 8 kev in a 20 @xmath9 m thick al - foil .
different samples were produced , containing @xmath10 up to @xmath11 atoms .
these numbers were determined at the irmm in geel via the detection of the 2.6 kev kx - rays ( emitted in the decay of @xmath0ar ) with a gas flow proportional counter .
the experiments with thermal neutrons were performed at the end of the 87 m curved neutron guide h22 of the high flux reactor at the ill in grenoble ( france ) .
a well - thermalised flux of about @xmath12 neutrons/@xmath13 was available at the sample position .
the @xmath0ar samples were mounted in a vacuum chamber together with suited surface barrier detectors , placed outside the neutron beam . a typical charged particle spectrum obtained during a 62 h neutron irradiation of one of the @xmath14ar samples
is shown in figure [ figure 1 ] .
the results of these measurements are summarised in table 1 .
comparison of our results with those obtained by ashgar et al.@xcite shows that our values for both the ( n@xmath15,@xmath3 ) and ( n@xmath15,p ) cross sections are about two times smaller , which indicates that the discrepancy most likely lies in the determination of the number of @xmath0ar atoms in the sample or in the neutron flux determination .
lrc + & & cross section + & & ( b ) + @xmath0ar(n@xmath15,@xmath3)@xmath16s & 4.63 & @xmath17 + @xmath0ar(n@xmath15,@xmath4)@xmath16s & 2.50 & @xmath18 + @xmath0ar(n@xmath15,@xmath19)@xmath16s & & @xmath20 + @xmath0ar(n@xmath15,p)@xmath0cl & 1.60 & @xmath21 +
the measurements with resonance neutrons were carried out at a 9 m long flight path of the linear accelerator gelina of the irmm in geel ( belgium ) , covering a neutron energy range from 10 mev up to 70 kev .
the flux determination was done via the well known @xmath22b(n,@xmath1)@xmath23li reaction .
an overview of the characteristics of the measurements is given in table 2 .
in none of the three measuring cycles the (
n , p ) , ( n,@xmath19 ) or ( n,@xmath24 ) reactions were observed , as could be expected from their small thermal values . [ cols= " < ,
< , < , < , < " , ] in the 100 hz measuring campaign a 1/v shape of the @xmath0ar(n,@xmath25 ) cross section could be established ( figure [ figure 2 ] ) .
a second measuring cycle , with the linac operating at 800 hz , provided us with cross section data for neutron energies up to 15 kev ( figure [ figure 2 ] ) .
two strong resonances were observed at 1.6 kev and at 2.5 kev , with resonance areas of @xmath26 b.kev and @xmath27 b.kev . in a third measuring cycle we used a surface barrier detector mounted in a vacuum chamber and realised good experimental conditions up to 70 kev neutron energy . here , two smaller resonances were observed at 25 kev and at 40 kev .
resonance areas are in the order of 12 b.kev and 15 b.kev respectively .
the determination of the maxwellian averaged cross section is based on a formula which calculates the maxwellian averaged cross section as a sum of the 1/v extrapolation of the thermal value and the contributions of the resonances@xcite : @xmath28 in eq .
( [ eq : curv ] ) @xmath29 is the thermal cross section value in mb , @xmath30 the stellar temperature in kev , @xmath31 the resonance energy in kev and @xmath32 the resonance area in ( mb.kev ) .
our data result in very large values for the maxwellian averaged cross section , e.g. 19 b at 2 kev which is 7 times larger than the theoretically calculated one ( figure [ figure 4 ] ) .
for the first time neutron induced reactions on @xmath0ar were performed , covering a neutron energy range from thermal energy up to 70 kev .
measurements with thermal neutrons were performed at the high flux reactor of the ill , leading to cross section values for the (
n , p ) , ( n,@xmath3 ) and ( n,@xmath4 ) reactions of @xmath5 b , @xmath33 b and @xmath34 mb respectively . for the ( n,@xmath3 ) reaction , measurements at the neutron spectrometer gelina of the irmm gave evidence for a perfect 1/v shape of the cross section in the low energy region and moreover revealed the existence of four resonances in the region up to 70 kev . the obtained resonance parameters combined with the thermal cross section value lead to very large values of the maxwellian averaged cross section . 99 ashgar m. , emsallem a. , hagberg e. , jonson b. and tidemand - petersson p. , _
z. phys _ a288 ( 1978 ) 45 .
bao z. and kappeler f. , _ atomic data and nuclear data tables _ 36 ( 1987 ) 411 .
wagemans c. , loiselet m. , bieber r. , denecke b. , reher d. and geltenbort p. , _ nucl .
instr . and meth .
_ a397 ( 1997 ) 22 . | the energy dependent @xmath0ar(n , p)@xmath0cl and @xmath0ar(n,@xmath1)@xmath2s reaction cross sections were determined for the first time in an experimental effort involving three large facilities : the cyclotron of the ucl ( louvain - la - neuve , belgium ) where implanted @xmath0ar samples were produced ; the high flux reactor of the ill ( grenoble , france ) where thermal ( n , p ) , ( n,@xmath3 ) and ( n,@xmath4 ) cross sections of @xmath5 b , @xmath6 b and @xmath7 mb respectively could be determined , and the gelina neutron time - of - flight facility of the irmm ( geel , belgium ) where strong resonances were observed in the kev region . |
oxygen vacancies were introduced in commercial single crystals of srtio@xmath0 by heating them in the temperature range of 800 to 1100 degrees celsius in a vacuum in the range of 10@xmath63 mbar in a manner reported before@xcite .
the samples typical dimensions were @xmath64 or @xmath65 .
ohmic contacts on sample corners were made by evaporating gold and heating up to 600 degrees .
compressive uniaxial stress was applied by sandwiching the sample between two metallic blocks connected with two stainless steel screws .
the magnitude of applied strain was measured using a 350@xmath66 resistive strain gage ( kyowa kfl-1 - 350-c1 - 11 ) .
a 20-atom cell ( @xmath67 ) is adopted to calculate the fermi surfaces and band structures of the tetragonal afd phase of srtio@xmath3 , while a 5-atom unit cell is used for the cubic phase . to study the effect of @xmath32 ratio , the in - plane lattice constant
is fixed to an experimental value at 50 k , i.e. , 5.507 @xcite , while @xmath32 varies from 1 to 1.0006 .
the afd tilting angle for each @xmath32 is fixed according to eq . 2 of the main article .
the calculations are carried out within the framework of density function theory ( dft ) in the local density approximation ( lda ) , using the abinit package@xcite .
the wave functions are expanded using plane - wave basis sets with kinetic energy cutoff of 45 hartree .
the self - consistent calculations are performed on an unshifted 6@xmath406@xmath404 and 12@xmath4012@xmath4012 @xmath68-point grid for the tetragonal and cubic phases , respectively .
the accurate fermi surface anisotropy ( fig .
4c of the main text ) is determined from the band structures along the pseudo - cubic [ 100 ] and [ 001 ] directions ( fig .
4b of the article ) , by finding @xmath68 for which the band intersects the chemical potential .
optimized norm conserving vanderbilt pseudopotentials are used here@xcite , in order to include the spin - orbit coupling ( soc ) , which yields a band edge splitting ( due to tetragonal distortion ) that is in good agreement with experiment @xcite .
this agreement is somewhat fortuitous , given the neglect of electron - electron interactions beyond semilocal dft and of electron - phonon interaction effects , as already observed by allen@xcite and van mechelen@xcite .
both effects will increase the effective mass , and may contribute to shift the afd splitting .
larger values of splitting were found with lapw@xcite , which should be more accurate dft as it does not pseudize the wave functions .
using the experimental values for lattice constant and afd angles removes the additional effect of dft structural relaxation . the fermi surface shape ( fig .
4a ) is obtained with explicit calculations including 200@xmath40200@xmath40200 @xmath68 points in a restricted cube of @xmath69 in each direction around @xmath70 . for the graphical representation of the fermi surface evolution as a function of the chemical potential and afd angle ,
non - relativistic extended norm conserving pseudopotentials are adopted @xcite .
the fermi surfaces of the lowest conduction band are obtained with 200@xmath40200@xmath40200 @xmath68 grid in the full brillouin zone , using maximally - localised wannier functions and the wannier90 code@xcite . without the soc
the fs are not quantitative , but qualitatively show the evolution of the shape and anisotropy . in agreement with previous theoretical studies@xcite
, we find that the cubic phase fs has six lobes at moderate to high doping . in the tetragonal state , two out of these six
are squeezed ( fig .
[ fig.s1 ] ) .
when the chemical potential ( @xmath38 ) is near the band edge , the fs shape is in agreement with the model bands given in @xcite .
10 p. zubko , g. catalan , a. buckley , p. r. l. welche , and j. f. scott .
strain - gradient - induced polarization in srtio@xmath0 single crystals .
lett . * 99 * , 167601(2007 ) .
a. koreeda , r. takano , and s. saikan .
second sound in srtio@xmath0 .
rev . lett.*99 * , 265502 ( 2007 ) . j. g. mavroides , j. a. kafalas and d. f. kolesar .
photoelectrolysis of water in cells with srtio@xmath0 anodes .
* 28 * , 241 ( 1976 ) w. d. rice _
_ persistent optically induced magnetism in oxygen - deficient strontium titanate .
nature materials * 13 * , 481 ( 2014 ) .
k. a. mller , and h. burkhard .
srtio@xmath0 : an intrinsic quantum paraelectric below 4 k. phys .
b * 19 * , 3593 ( 1979 ) .
a. spinelli , m. a. torija , c. liu , c. jan and c. leighton .
electronic transport in doped srtio@xmath0 : conduction mechanisms and potential applications .
b _ * 81 * , 155110 ( 2010 ) .
y. kozuka , t. susaki and h. y. hwang . vanishing hall coefficient in the extreme quantum limit in photocarrier - doped srtio@xmath0 .
* 101 * , 096601 ( 2008 ) .
j. son _ et al .
_ epitaxial srtio@xmath0 films with electron mobilities exceeding 30,000 @xmath71 .
nature materials * 9 * , 482 ( 2010 )
. k. behnia , on mobility of electrons in a shallow fermi sea over a rough seafloor .
j. physics : condens .
* 27 * , 375501 ( 2015 ) .
x. lin , b. fauqu and k. behnia .
scalable t@xmath72 resistivity in a small single - component fermi surface .
science 349 945 ( 2015 ) .
x. lin , z. zhu , b. fauqu , and k. behnia .
fermi surface of the most dilute superconductor .
x 3 , 021002 ( 2013 ) . l. p. gorkov .
phonon mechanism in the most dilute superconductor : n - type srtio@xmath0 .
arxiv:1508.00529 ( 2015 ) .
j. m. edge , y. kedem , u. aschauer , n. a. spaldin , a. v. balatsky .
quantum critical origin of the superconducting dome in srtio@xmath0 .
rev . lett . * 115 * , 247002 ( 2015 ) .
j. ruhman and p. a. lee .
superconductivity at very low density : the case of strontium titanate .
arxiv:1605.01737 ( 2016 ) .
s. e. rowley _
_ ferroelectric quantum criticality .
* 10 * , 367 ( 2014 )
. f. w. lytle .
x - ray diffractometry of low - temperature phase transformations in strontium titanate
* 35 * , 2212 ( 1964 ) .
h. unoki and t. sakudo .
electron spin resonance of fe3 + in srtio@xmath0 with special reference to the 110 k phase transition .
japan * 23 * , 546 ( 1967 ) .
p. a. fleury , j. f. scott , and j. m. worlock .
soft phonon modes and the 110 k phase transition in srti0@xmath0 , phys .
* 21 * , 16 ( 1968 ) .
k. a. mller , w. berlinger , and f. waldner .
characteristic structural phase transition in perovskite - type compounds .
* 21 * , 814 ( 1968 ) .
e. k. h. salje , m. c. gallardo , j. jimnez , f. j. romero , and j. del cerro .
the cubic - tetragonal phase transition in strontium titanate : excess specific heat measurements and evidence for a near - tricritical , mean field type transition mechanism .
j. phys . : condens .
matter * 10 * , 5535 ( 1998 ) .
e. f. steigmeier , field effect on the cochran modes in srtio@xmath0 and ktao@xmath0 , phys .
rev.*168 * , 523 ( 1968 ) .
d. buerle and w. rehwald . structural phase transitions in semiconducting srtio@xmath0 .
solid state commun .
* 27 * , 1343 ( 1978 ) .
s. tsunekawa , h. f . j. watanabe , and h. takei .
linear thermal expansion of srtio@xmath0 . phys .
( a ) * 83 * , 467 ( 1984 ) .
j. f. scott , e. k. h. salje , and m. a. carpenter .
domain wall damping and elastic softening in srtio@xmath0 : evidence for polar twin walls .
lett . * 109 * , 187601 ( 2012 ) .
e. k. h. salje , o. aktas , m. a. carpenter , v. v. laguta , and j. f. scott .
domains within domains and walls within walls : evidence for polar domains in cryogenic srtio@xmath0 .
. lett . * 111 * , 247603 ( 2013 ) .
a.verma , a. p. kajdos , t. a. cain , s. stemmer and d. jena . intrinsic mobility limiting mechanisms in lanthanum - doped strontium titanate .
lett . * 112 * , 216601 t. a. cain , a. p. kajdos , and s. stemmer .
la - doped srtio@xmath0 films with large cryogenic thermoelectric power factors .
. lett . * 102 * , 182101 ( 2013 ) .
h. c. montgomery .
method for measuring electrical resistivity of anisotropic materials .
. phys . * 42 * , 2971 ( 1971 ) . j. appel .
soft - mode superconductivity in srtio@xmath73 .
phys . rev .
* 180 * , 508 ( 1969 ) . c. a.
m. dos santos _ et al .
_ procedure for measuring electrical resistivity of anisotropic materials : a revision of the montgomery method .
. phys . * 110 * , 083703 ( 2011 ) .
e. fradkin , s. a. kivelson , m. j. lawler , j. p. eisenstein and a. p. mackenzie .
nematic fermi fluids in condensed matter physics
. annual review of condensed matter physics , 1 : 153 ( 2010 ) .
kuo , j .- h .
chu , j. c. palmstrom , s. a. kivelson , i. r. fisher .
ubiquitous signatures of nematic quantum criticality in optimally doped fe - based superconductors science * 352 * , 958 ( 2016 ) . b. s. de lima _ et al .
_ interplay between antiferrodistortive , ferroelectric , and superconducting instabilities in sr@xmath28ca@xmath29tio@xmath74 .
b * 91 * , 045108 ( 2015 ) .
a. tkach1 , a. almeida1 , j. a. moreira1 , t. m. correia1 , m. r. chaves , o. okhay , p. m. vilarinho , i. gregora and j. petzelt .
enhancement of tetragonality and role of strontium vacancies in heterovalent doped srtio@xmath0 .
* 98 * , 052903 ( 2011 ) c. s. koonce , m. l. cohen , j. f. schooley , w. r. hosler , and e. r. pfeiffer . superconducting transition temperatures of semiconducting srtio@xmath0 .
163 , 380 ( 1967 ) .
h. suzuki _ et al .
_ superconductivity in single - crystalline sr@xmath28la@xmath29tio@xmath0 , j. phys .
. jpn . * 65 * , 1529 ( 1996 ) .
king , s. mckeown walker , a. tamai , a. de la torre , t. eknapakul , p. buaphet , s .- k .
mo , w. meevasana , m.s .
bahramy , and f. baumberger , nature comm .
* 5 * , 3414 ( 2014 ) .
j. l. m. van mechelen _ et al .
_ electron - phonon interaction and charge carrier mass enhancement in srtio@xmath0 .
lett . * 100 * , 226403 ( 2008 ) .
d. van der marel , j. l. m. van mechelen , and i. i. mazin .
common fermi - liquid origin of t@xmath72 resistivity and superconductivity in n - type srtio@xmath0 .
b * 84 * , 205111 ( 2011 ) .
x. lin _ et al .
_ critical doping for the onset of a two - band superconducting ground state in srtio@xmath9 .
lett . * 112 * , 207002 ( 2014 ) .
s. j. allen _ et al . _ conduction - band edge and shubnikov - de - haas effect in low - electron - density srtio@xmath0 .
b * 88 * , 045114 ( 2013 ) . e. mikheev _
_ carrier density independent scattering rate in srtio@xmath0-based electron liquids , sci . rep .
* 6 * , 20865 ( 2016 ) r. joynt and l. taillefer .
the superconducting phases of upt@xmath0 .
phys . * 74 * , 235 ( 2002 ) .
a. p. mackenzie and y. maeno .
the superconductivity of sr@xmath58ruo@xmath59 and the physics of spin - triplet pairing .
* 75 * , 657 ( 2003 ) .
s. nakamae _
_ electronic ground state of heavily overdoped nonsuperconducting la@xmath75sr@xmath29cuo@xmath59 .
b * 68 * , 100502(r ) ( 2003 ) b. kalisky _
locally enhanced conductivity due to the tetragonal domain structure in laalo@xmath0/srtio@xmath0 heterointerfaces .
nature materials * 12 * , 1091 ( 2013 ) . h. k. pal , v. i. yudson , and d. l. maslov . resistivity of non - galilean - invariant fermi- and non - fermi liquids . lith . j. phys . * 52 * , 142 ( 2012 ) . w. jauch and a. palmer .
anomalous zero - point motion in srtio@xmath0 : results from -ray diffraction .
b * 60 * , 2961 ( 1999 ) .
et al . _ a brief introduction to the abinit software package .
fr kristall .
* 220 * , 558 ( 2005 ) .
abinit : first - principles approach to material and nanosystem propertiescomput .
. commun . * 180 * , 2582 ( 2009 ) .
d. r. hamann . optimized norm - conserving vanderbilt pseudopotentials .
b * 88 * , 085117 ( 2013 ) . m. teter .
additional condition for transferability in pseudopotentials .
phys . rev .
b * 48 * , 5031 ( 1993 ) .
a. a. mostofi , j. r. yates , y .- s .
lee , i. souza , d. vanderbilt and n. marzari .
wannier90 : a tool for obtaining maximally - localised wannier functions .
. commun . * 178 * , 685 ( 2008 ) . | cubic srtio@xmath0 becomes tetragonal below 105 k. the antiferrodistortive ( afd ) distortion leads to clockwise and counter - clockwise rotation of adjacent tio@xmath1 octahedra .
this insulator becomes a metal upon the introduction of extremely low concentration of n - type dopants .
however , signatures of the structural phase transition in charge conduction have remained elusive . employing the montgomery technique , we succeed in resolving the anisotropy of charge conductivity induced by the afd transition , in the presence of different types of dopants .
we find that the slight lattice distortion ( @xmath2 ) gives rise to a twenty percent anisotropy in charge conductivity , in agreement with the expectations of band calculations .
application of uniaxial strain amplifies the detectable anisotropy by disfavoring one of the three possible tetragonal domains .
in contrast with all other known anisotropic fermi liquids , the anisotropy has opposite signs for elastic and inelastic scattering . increasing the concentration of dopants leads to a drastic shift in the temperature of the afd transition either upward or downward .
the latter result puts strong constraints on any hypothetical role played by the afd soft mode in the formation of cooper pairs and the emergence of superconductivity in srtio@xmath3 .
strontium titanate , of the abo@xmath0 family of perovskites , has continuously attracted attention during the last five decades . as a band insulator
, it displays a fascinating variety of fundamental@xcite and application - oriented@xcite properties .
it is a quantum paraelectric , avoiding a ferroelectric transition thanks to quantum fluctuations@xcite . with a low - temperature electric permittivity enhanced by four orders of magnitude , it turns to a metal upon the introduction of infinitesimal concentration of dopants@xcite or even after shining light@xcite .
the carriers are remarkably mobile@xcite , as a consequence of an exceptionally long bohr radius@xcite and scatter off each other generating a resistivity which varies as @xmath4 , with a very large prefactor@xcite . in spite of being far apart , the electrons living in this peculiar environment can form cooper pairs in the most dilute superconductor currently known@xcite .
after five decades of intensive research , several questions remain unanswered .
what causes superconductivity@xcite and restricts it to a concentration range of @xmath5@xcite ?
why is the inelastic scattering so large @xcite ? does the aborted ferroelectricity@xcite play a role in superconductivity because of its quantum criticality@xcite ?
in contrast to its isovalent siblings batio@xmath0 and catio@xmath0 , srtio@xmath0 is cubic at room temperature .
however ,
it becomes tetragonal below 105 k. in the ordered state , tio@xmath1 octahedra rotate in alternating directions around the @xmath6-axis@xcite .
this antiferrodistortive phase transition has been studied by a variety of experimental techniques .
clear anomalies in specific heat@xcite , thermal conductivity@xcite , sound velocity@xcite and thermal expansion @xcite have been observed at @xmath7105 k. the transition does not break inversion symmetry , but the boundaries between tetragonal domains become polar@xcite deep inside the ordered state . in spite of being widely documented over several decades ,
the consequences of this phase transition for mobile electrons have been elusive .
a recent study@xcite concluded that the afd transition should be accompanied by en enhancement in electron scattering .
however , the experimental data does not show any visible anomaly at the transition temperature .
this is also the case of other numerous studies devoted to the transport properties of metallic srtio@xmath0 , which failed to detect any anomaly in the vicinity of 105 k in electric@xcite or thermoelectric@xcitetransport .
a ) left : semi - log plot of @xmath8 in lightly - doped srtio@xmath9 along two perpendicular axes .
the inset is a zoom near the structural transition .
no clear anomaly is visible at 105 k. right : the @xmath10 ratio shows a clear upward deviation from a constant value , close to unity , at 105 k. the inset shows the geometry of the montgomery method ( see text ) .
b ) atomic configurations in the cubic and antiferrodistortive states . in the latter adjacent octahedra rotate clockwise and anti - clockwise .
c ) representations of the three possible domains in the afd state according to the orientation of the c - axis .
blue(red ) squares represent octahedra which rotate clockwise ( anti - clockwise).,title="fig : " ] in this paper , we report on a direct measurement of the in - plane anisotropy of charge conductivity in metallic srtio@xmath0 with different dopants .
thanks to the sensitivity of the montgomery technique@xcite employed here for the first time to study srtio@xmath0 , a sharp anomaly at 105 k is clearly visible .
the very slight departure from cubicity ( @xmath11 ) gives rise to an anisotropy of conductivity several orders of magnitude larger ( @xmath12 ) .
first - principles calculations of the fermi surface in the afd state explain the sign of the anisotropy just below the transition temperature and the order of magnitude of its amplitude . remarkably , we find that the anisotropy in conductivity displays a non - monotonous temperature dependence . applying uniaxial
strain favors two out of the three possible tetragonal domains and amplifies the non - monotonous behavior .
in contrast to what has been observed in other fermi liquids , elastic and inelastic scattering have opposite sign anisotropies
. finally , we report on the evolution of @xmath13 with doping , and find that different dopants , which give the same superconducting @xmath14 , affect the afd transition in different ways .
we argue that this provides evidence against a theoretical scenario tracing the microscopic origin of superconductivity to the afd soft phonon mode@xcite .
the montgomery technique is a powerful method used to extract the anisotropy of conductivity in a solid shaped as a rectangular prism@xcite .
it is particularly well adapted to the case of srtio@xmath0 because of the availability of relatively large single crystals with precise geometry , on whose corners we evaporate gold pads . at each temperature ,
current is applied and voltage measured using two pairs of adjacent electrodes in two perpendicular configurations ( see the inset of fig .
1a ) .
the ratio of apparent resistance values ( @xmath15 = v / i ) yields a parameter @xmath16 , which is linked to the ratio of in - plane dimensions ( @xmath17 ) of an equivalent isotropic conductor ( @xmath18)@xcite . assuming that the sample is a perfect square with thickness @xmath19 , the ratio @xmath20 yields the intrinsic resistivity along each perpendicular axis@xcite : @xmath21 like a previous report@xcite , we found that the magnitude of resistivity obtained by this method is in agreement with what is obtained by a standard four - contact set - up .
the advantage of the montgomery technique is the direct resolution of the anisotropy , as shown in panels a and b of fig .
[ fig.1 ] , which the date for a lightly - doped srtio@xmath9 ( n=7.8 10@xmath22@xmath23)sample .
it is hard to see any anomaly in the vicinity of 105 k in either @xmath24 and @xmath25 , but the temperature dependence of their ratio reveals a clear abrupt jump at the onset of the afd transition . the ratio is close to unity and temperature - independent above @xmath7 105 k and suddenly jumps at @xmath13 .
the slight departure from unity above @xmath13 presumably results from imperfect squareness of the sample , the finite size of the contact size and other technical limitations . to the best of our knowledge ,
this is the first detection of any consequence of the afd transition for charge transport in strontium titanate .
the montgomery technique appears as a powerful method to detect the loss of rotational symmetry and the emergence of nematicity @xcite in other solids such as iron - based superconductors@xcite .
remarkably , multiplicity of domains does not hinder the observation of the symmetry breaking by a directional probe such as resistivity .
each of the three axes of the elementary cubic cell can become the tetragonal axis ( fig .
1c ) .
therefore , there are three possible domains in an unconstrained crystal and the @xmath26 ratio is not the intrinsic anisotropy of srtio@xmath9 .
its magnitude is set by the accidental imbalance between the population of these domains .
one would expect therefore that the magnitude and the orientation of applied strain should modify the @xmath26 ratio ; as seen in fig .
2 , this is indeed the case . cooling the same sample in the presence of strain ,
the anisotropy in the ordered state is enhanced . moreover , rotating the orientation of the applied strain by 90 degrees in the same sample yields an opposite sign for ( @xmath27 ) in the afd state .
we conclude that the samples are multi - domain and applying compressive strain favors the two domains which have their ( longer ) @xmath6-axis perpendicular to the strain . as seen in fig .
2c , strained srti@xmath28nb@xmath29o@xmath0 displays also a sizeable anisotropy in its @xmath26 ratio below the transition temperature .
the first outcome of this study is a quantification of the magnitude of @xmath26 ratio . as seen in fig .
2 , in samples subject to compressive strain , this ratio can become as large as 1.2 . at first sight , it looks surprising that an anisotropy of this amplitude in charge conductivity results from a tetragonal distortion quantified by x - ray diffraction to be as small as 1.00056@xcite
. the tetragonal distortion is linked to the angle of rotation of the tio@xmath1 octahedra@xcite : @xmath30 the measured values of @xmath31@xcite and @xmath32=1.00056 @xcite satisfy this expression . as seen in fig .
1c , the slight tetragonal distortion is accompanied by a more substantial misalignment between ti atoms .
considering that their t@xmath33 orbitals host the threefold degenerate conduction bands of metallic srtio@xmath0 , the consequences can be drastic .
a slight misalignment between adjacent tio@xmath1 octahedra can drastically change the overlap between t@xmath33 orbitals of two neighboring titanium atoms .
thus , a very small @xmath34 can cause a substantial drop in conductivity along the misalignment orientation providing a qualitative agreement for the experimental observation . as we will see below , according to first - principles band calculations , the fermi surface topology is deeply affected by the phase transition .
a second and more unexpected result is the temperature dependence of the anisotropy in the ordered state .
as seen in figures [ fig.1 ] and [ fig.2 ] , the temperature dependence of @xmath26 ratio is not monotonous . in srtio@xmath9
, it peaks around 20 k and then decreases to below unity . in srti@xmath28nb@xmath29o@xmath0 ,
the temperature dependence is even more complex , with a peak and a shallow minimum .
if the @xmath26 ratio were proportional to an order parameter , then this would suggest a second phase transition well below 105 k. the existence of additional phase transitions in srtio@xmath0 has been a longstanding controversy . however , our observation can be accounted for if carrier scattering by phonons , electrons and defects do not share the same anisotropy .
we will come back to this below .
we now turn our attention to the evolution of the afd transition temperature with doping . as seen in fig . 3 , the afd anomaly in @xmath26 ratio shifts to higher temperature with the substitution of ti with nb .
we also studied the evolution of @xmath13 with oxygen reduction and la substitution .
our results are in agreement with previous studies of the afd transition using sound velocity@xcite and ramman [email protected] emerging picture is summarized in fig . 3e .
while substituting ti with nb ( or sr with la ) leads to an upward shift of @xmath13 , introducing oxygen vacancies pulls it downward .
the extreme sensitivity of the afd transition to the presence of a tiny concentration of foreign atoms is remarkable . substituting sr with larger la or ti with larger nb both
strengthen the afd instability , presumably by weakening the stability of the cubic structure , which relies on a fragile balance of the three atomic radii in srtio@xmath0 .
intriguingly , this also happens when sr is replaced with smaller ca@xcite .
in contrast , oxygen vacancies weaken the tetragonal distortion , presumably because they destroy the antiferrodistortive coupling between adjacent octahedra .
the contrast in response of @xmath13 to oxygen reduction compared to la and nb substitution has an unnoticed implication .
all three routes for @xmath35-doping , i.e. , oxygen removal , substituting ti(by nb)or sr ( by la)@xcite lead to a dilute metal with a superconducting @xmath14 which peaks at a carrier density in the range of @xmath36 @xmath23 .
decades ago , appel@xcite suggested a link between the formation of cooper pairs and the soft phonon mode associated with the octahedra rotation and the afd transition@xcite .
the drastic difference in the dependency of @xmath13 on dopants type indicates that either the soft mode does not play a major role in superconductivity or its response to reduction and substitution is not the same as @xmath13 .
we carried out _ first - principles _ calculations to quantify the fermi surface ( fs ) anisotropy .
fig .
[ fig.4 ] shows the fs cross sections at @xmath371 mev , the electron band structure , and the anisotropy of the fs as a function of @xmath32 for chemical potentials ( @xmath38 ) from 1 to 10 mev .
the lower conduction bands in sto originate from the same orbital manifold . in the tetragonal system
they have distinct irreducible representations for the spatial degrees of freedom , which allows them to cross , only at a single point along 001 . in the plane ( or in any skew direction ) the afd rotation mixes their representations , and the bands will anticross , as is seen along 100 . in agreement with previous theoretical studies@xcite , we find that the cubic phase fs has six lobes and in the tetragonal state , two out of these six are squeezed .
according to shubnikov - de haas ( sdh ) experiments @xcite when carrier density is @xmath397.8@xmath4010@xmath22@xmath23 , the fermi energy is close to @xmath41 mev . at this level ( fig .
4a ) , we find a fs area in the ( 001 ) plane about 1.7 times larger than in the ( 010 ) plane , in fair agreement with the sdh data@xcite .
the calculated band masses are lighter than the experimental values , indicating the importance of renormalization by electron - phonon interactions and electron correlation@xcite .
the estimated renormalizations are 1.5 for the lower band ( similar to explicit calculations in ref .
@xcite ) and 3.5 for the upper band .
a quantitative description of fermi surface would requires the calculation of _ anisotropic _ mass renormalization .
the fs anisotropy , ( ratio of its radii along @xmath6 and @xmath42 ) reaches 0.65 ( fig .
4c ) for @xmath43 mev .
the fermi radius is shorter ( and @xmath44 larger ) along @xmath6 , such that the conductivity should be larger as well , in agreement with the experimental observation in the vicinity of @xmath13 . finally , according to the calculations
, anisotropy decreases with increasing @xmath38(see the supplement ) .
this matches the experimental observation of diminishing anomaly at @xmath13 with increasing doping ( fig .
[ fig.3]a - d ) .
let us now examine the puzzle of the non - monotonic temperature dependence of @xmath45 . as seen in fig .
1a , upon cooling from room temperature to liquid - helium temperature , the resistivity of metallic srtio@xmath0 drops by three orders of magnitude .
thus , inelastic scattering of electrons is by far the dominant scattering around 100 k. most of this inelastic scattering is due to phonons , but below 30 k , the electron - electron scattering leads to a quadratic temperature dependence .
fig .
5 shows the low - temperature resistivity of a strained oxygen - reduced sample as a function of @xmath46 . in agreement with previous reports@xcite ,
@xmath24 and @xmath25 both follow a behavior described by @xmath47 ( i=@xmath48,@xmath49 ) .
the anisotropy of the prefactor is set by the expected anisotropy of conductivity stemming from the topology of the fermi surface .
note that in this temperature window , below the degeneracy temperature of electrons , the relevance of the fermi surface topology is straightforward . on the other hand , when the carrier density is as low as @xmath50 , the system is non - degenerate and subject to boltzmann statistics at 105 k. but , even in this non - degenerate regime , the anisotropy of conductivity still keeps the trace of the anisotropy of the fermi velocity .
the prefactor of t - square resistivity along a - axis ( c - axis ) is @xmath51(@xmath52 ) [ the average ( @xmath53 ) is close to what was measured for a sample of comparable carrier density by a conventional four - probe technique@xcite ] .
as expected for an anisotropic fermi liquid , for example in upt@xmath0@xcite , the ratio ( @xmath54 ) is close to the ratio of the fermi velocities along @xmath6-axis and @xmath42-axis . as seen in fig . 5 , the intercept ( i.e. the residual resistivity , @xmath55 ) is lower along the direction where the prefactor @xmath56 is larger . in other words ,
elastic scattering is larger along the orientation for which inelastic scattering is smaller .
this generates a peak in @xmath26(@xmath57 ) , since the balance between the two components of the resistivity is changed at an intermediate temperature .
this feature distinguishes srtio@xmath0 from other anisotropic fermi liquids . in the latter cases , the residual resistivity and the prefactor of the @xmath46 resistivity share the same anisotropy , set by the fermi velocity anisotropy .
the same is true of upt@xmath0@xciteas well as two other well - known perovskites , the ruthenate sr@xmath58ruo@xmath59@xcite and cuprate la@xmath60sr@xmath61cuo@xmath59@xcite . in both of these layered conductors , the prefactor of the quadratic term in the resistivity and the residual resistivity share the same anisotropy .
the exceptional case of metallic srtio@xmath9 may be a result of the particular orientation of boundaries between twins in a sample subject to uniaxial strain ( see fig .
5b ) . application of strain along one direction leads to the suppression of one domain out of three ( the one with its @xmath6-axis along the strain ) .
since the fermi velocity is larger along the @xmath6-axis , the conductivity should become _ lower _ in the orientation parallel to the applied strain .
this is indeed what is seen experimentally . on the other hand , in such a configuration
the favored domain boundary would lie along the orientation of compressive strain . as a scatterer
, such a boundary would introduce additional scattering perpendicular to the direction of strain .
this could explain the opposite signs of @xmath62 at the onset of the transition ( governed by the anisotropy of the fermi velocity ) and in the zero temperature limit ( set by the orientation of twin boundaries ) .
two other features should also be included in an exhaustive quantitative treatment of the problem .
the first is the experimental evidence for enhanced conductivity along a twin boundary in srtio@xmath0@xcite , amplifying the anisotropy introduced by the presence of a twin boundary along the orientation of compressive strain .
the second is the contrast between potential wells introduced by nb dopants and oxygen vacancies .
they are expected to differ in structure and the angular distribution of the scattering they cause .
several other questions remain .
the precise shape of the fermi surface in dilute metallic srtio@xmath0 is not settled yet .
a detailed angle - dependent study on single - domain samples is necessary to map the full fs .
the quantification of the non - parabolic dispersion is important to pin down the origin of the @xmath4 resistivity in the dilute limit@xcite .
non - parabolicity leads to a distinction between momentum and velocity , and breaks galilean invariance@xcite , which is expected to play a significant role in the @xmath46 resistivity of srtio@xmath0 . in summary
, we quantify the anisotropy of charge conductivity caused by the afd transition in metallic srtio@xmath0 , using the montgomery technique .
both first - principles band calculations and experimental data find that the very slight departure from cubicity gives rise to a sizeable anisotropy in the fermi surface and conductivity .
in contrast with other anisotropic fermi liquids , elastic and inelastic resistivity do not present the same anisotropy .
the contrasting response of the afd and superconducting phase transitions to the type of dopant argues against any role played by the octahedral rotational mode in superconductivity .
this work was supported by agence national de la recherche through superfield and quantumlimit projects .
q.t . acknowledges a grant offered by china scholarship council and support by the national basic research program of china ( grant no.2012cb821404 ) .
b. x. and m. j. v. are supported by arc projects themotherm ( ga 10/15 - 03 ) and aimed ( ga 15/19 - 09 ) from the communaut franaise de belgique , and a pdr project ( ga t.1077.15 ) from the fonds national pour la recherche scientifique ( belgium ) .
computer time was provided by ceci ( fnrs ga 2.5020.11 ) , segi , and cenaero / zenobe through the walloon region ga 1117545 . |
in this work , we consider the initial value problem for the nonlinear schrdinger equation in two spatial dimensions @xmath8 this classical equation has been studied intensively for the past fifty years and many problems regarding local well - posedness , global existence , blowup and asymptotic behaviour have very complete answers ( see the monographs @xcite , @xcite , @xcite and references therein ) .
the most standard framework where one studies this initial value problem is based on the following @xmath6 local well - posedness result : for any @xmath9 , there exists a unique maximal solution @xmath10 of , which depends continuously on the initial data .
moreover , if @xmath11 , one has @xmath12 as @xmath13 .
the @xmath6 framework allows one to used certain conserved quantities ( mass , energy , variance , etc . ) to obtain precise results on the dynamical behaviour of solutions .
apart from this , few nonstandard frameworks have been developed ( see , for example , @xcite , for a theory on zhidkov spaces ) . here
, we shall introduce new functional spaces and study local well - posedness , global existence and stability ( in a way that will be clear further ahead ) .
these spaces do not lie within @xmath5 : in fact , the first one will lie in @xmath14 , but not in @xmath15 , for any @xmath16 ; the second will lie in @xmath17 , but not in @xmath15 , @xmath18 .
the most simple example of these spaces appeared for the first time in a previous work , @xcite , in the context of the hyperbolic nonlinear schrdinger equation .
we shall generalize the results of @xcite in the context of the ( nls ) for two reasons : to our knowledge , there are no results of this type for ( nls ) ; secondly , even though one may prove similar results for more general equations , that would increase needlessly the complexity of the exposition and deviate the attention of the reader from the essential ideas .
the first step is to take any @xmath19 and consider solutions of ( nls ) of the form @xmath20 ( we call these solutions _ spatial plane waves _ : the number @xmath21 is the _ speed _ and the function @xmath22 is the _ profile _ of the wave ) . introducing this ansatz ,
one arrives to @xmath23 which one may solve using the standard @xmath6 theory .
it is clear that spatial plane waves are not @xmath24 solutions of ( nls ) , for any @xmath16 , even though they lie in @xmath25 .
based on these solutions , consider the space of spatial plane waves @xmath26 then the initial value problem for ( nls ) is well - posed over @xmath27 , simply because the problem reduces to the local theory in one spatial dimension .
now consider @xmath28 notice that , due to the lack of decay of elements in @xmath27 , the sum is indeed a direct sum .
then the initial value problem for @xmath29 can be proven to be equivalent to the initial value problem for the system @xmath30 where @xmath31 is the profile of @xmath32 .
notice that the second equation is independent on @xmath33 .
this means that one may solve it and introduce the solution onto the first equation .
thus , to prove local well - posedness on @xmath34 , it suffices to show an @xmath6 local well - posedness result for the first equation .
this is achieved by making the crucial observation that the nonlinear terms lie in @xmath35 ( even though they involve @xmath36 ) .
with such a local well - posedness result , consider the following situation : take a spatial plane wave @xmath36 with initial data @xmath32 , and assume @xmath36 is globally defined . if one introduces an @xmath6 perturbation on the initial data , @xmath37 , the corresponding solution is given by @xmath38 ( since @xmath36 only depends on the plane wave part of the initial data ) .
then a natural question is the following : _ stability problem : if @xmath39 is sufficiently small , does @xmath40 stay close to @xmath36 ? _
this question corresponds to a global existence result for @xmath33 that insures that @xmath33 stays small for all times .
results of such nature are indeed valid for ( nls ) in @xmath6 ( see @xcite , @xcite , @xcite , @xcite , @xcite ) in the @xmath7-supercritical case . here , however , the question is not that clear : first , if one develops the nonlinear terms in the equation for @xmath33 , one sees that lower order terms in @xmath33 are present ; second , the plane wave component @xmath36 , which even has `` infinite mass '' and `` infinite energy '' ( due to the lack of integrability ) might act as a forcing term , pushing @xmath33 to grow indefinitely . in @xcite ,
such a result was proven for @xmath41 an even power greater or equal to 4 , in the context of the hyperbolic nonlinear schrdinger equation .
now we advance to the next level of complexity .
take a sequence of wave speeds @xmath42 , with @xmath43 , @xmath44 , and consider the space @xmath45 endowed with the induced norm ( which , as we will check , is well - defined ) . as before ,
we look for a local well - posedness result in @xmath46 which turns out to be equivalent to a local well - posedness result for the infinite system @xmath47 the equations for the profiles @xmath48 are solved using the @xmath49 local well - posedness result ( here , one must be careful to ensure that _ all _ profiles exist up to some time @xmath50 ) . once again , it remains to prove that one may solve the equation for @xmath33 in @xmath4 , which amounts to check that the nonlinear terms are in @xmath51 . under the additional hypothesis @xmath52 ,
this can be shown to be true : heuristically , if one develops the nonlinear terms , one obtains some terms with @xmath33 , which are well - behaved , and some products of diferent @xmath53 s .
remarkably , these terms lie in @xmath5 : take , for example , @xmath54 , @xmath55 . then @xmath56 however , since one has an infinite sum of such terms , one must be able to control @xmath57 the above quantity turns out to be preserved by the flow of the infinite system , since one has conservation of the @xmath7 norm of each @xmath48 , and therefore it can controlled by the initial data . hence we shall restrict ourselves to @xmath58 one finally concludes that it is possible to solve the equation for @xmath33 and arrives to the local well - posedness result over @xmath59 : [ existnumeravel ] fix @xmath52 and @xmath60 .
then there exists a unique maximal solution @xmath61 ( cf . ) of ( nls ) such that @xmath62 .
the solution depends continuously of @xmath63 .
furthermore , if @xmath64 , then @xmath65 as in the single plane wave case , the stability problem may be studied . here , we present the following result : [ estabilidadenumeravel ] set @xmath66 even . given @xmath67
, there exist @xmath68 and @xmath69 , with @xmath70 as @xmath71 , such that , if @xmath72 and @xmath73 satisfy @xmath74 @xmath75 then the solutions of ( nls ) @xmath40 and @xmath76 with initial data @xmath77 and @xmath32 , respectively , are both globally defined and satisfy @xmath78 we now arrive to the last generalization , more complex and with interesting properties .
the idea is simply to pass from a discrete sum of plane waves to an _ integration over a continuum of plane waves_. that is , we consider @xmath79 endowed with the norm @xmath80 .
one may actually define an integral transform , the _ plane wave transform _ , mapping functions in two _ speed _ variables to functions in two _ physical _ variables @xmath81 before we study the initial value problem for ( nls ) in this context , we shall study some very interesting properties related to the plane wave transform . among them , one may prove that : * @xmath82 , @xmath83 under suitable conditions on @xmath22 ( proposition [ integrabilidadelp ] and corollary [ integrabilidadel2 ] ) ; * @xmath84 if @xmath85 is positive ( corollary [ naointegrabilidadel2 ] ) ; * the convolution of two functions , the fourier transform and the laplace transform may be obtained using the plane wave transform transform ( corollary [ convolucao ] and examples [ calor ] and [ schrodinger ] ) ; * several classical linear equations , such as the heat equation , the schrdinger equation and the wave equation , may be solved by means of this transform ( section [ resolverequacoes ] ) .
one should regard all the previous definitions in analogy to the fourier series and transform : one starts with a simple periodic function with a given frequency , superposes a numerable family of functions with different frequencies to arrive to the fourier series and passes to the continuous case to obtain the fourier transform , where one also has the concepts of physical and frequency variables . evidently , our construction is not the same as the fourier one and many properties will differ .
one aspect is that , while the fourier construction makes all sense in one variable ( and its multidimensional version is simply the application of the one - dimensional case to each variable ) , the plane wave construction needs ( at least ) a two - dimensional setting . in another perspective
, the fourier transform is based on the solutions of the ode @xmath86 while the plane wave theory is based on solutions of the transport equation @xmath87 with this new transform in hand , we try to obtain some results in the spirit of the numerable case .
that is , look for local well - posedness and stability results in @xmath88 .
the main difference is that now the ( nls ) may be decoupled into @xmath89 surprisingly , when one passes to the continuous case , there exists a decoupling such that the equation for the plane wave component is linear .
we now state our main results .
[ existenciacontinuo ] fix @xmath52 and @xmath90 .
then there exists a unique maximal solution @xmath91 of ( nls ) ( cf . definition [ defisolucao ] ) such that @xmath62 .
the solution depends continuously on the initial data .
furthermore , if @xmath64 , then @xmath92 [ estabilidadecontinuo ] set @xmath93 even .
given @xmath67 , there exist @xmath68 and @xmath94 , with @xmath95 as @xmath71 , such that , if @xmath96 and @xmath73 satisfy @xmath97 @xmath98 then the solution @xmath40 of ( nls ) with initial data @xmath77 is globally defined .
moreover , if @xmath99 is the free schrdinger group in dimension two , @xmath100 as a consequence of theorem [ estabilidadecontinuo ] , we have [ grandesdados ] set @xmath93 even .
for any given @xmath101 and @xmath67 , there exists @xmath102 satisfying the conditions of theorem [ estabilidadecontinuo ] and @xmath103 consequently , for any @xmath73 such that @xmath104 the @xmath6 solution @xmath40 of ( nls ) with initial data @xmath77 is global and satisfies @xmath105 the global @xmath6 solutions given by theorem [ grandesdados ] have small @xmath106 norm , since this norm is controlled by @xmath107
. therefore the energy of these solutions will always be positive and so there is no possible contradiction with the usual virial blowup result by glassey @xcite .
finally , we present a global well - posedness result for @xmath108 .
[ gwp ] fix @xmath108 .
consider @xmath109 , @xmath73 , @xmath96 .
assume that @xmath110 then there exists a unique global solution @xmath111 of ( nls ) with initial data @xmath63 .
we now outline the structure of the paper .
section 2 will be devoted to the study of the numerable superposition of plane waves and theorems [ existnumeravel ] and [ estabilidadenumeravel ] are proved therein . in section 3
, we recall the plane wave transform , construct a suitable functional framework in which one may consider such a transform , and prove several properties . in section 4 , we apply the transform to solve some classic linear pde s . in section 5 ,
we apply the theory for the transform to obtain theorems [ existenciacontinuo ] , [ estabilidadecontinuo ] , [ grandesdados ] and [ gwp ] .
we conclude with some remarks and , in the appendix , we give an alternate proof of the injectivity of the plane wave transform .
we fix some notations .
we define the space of schwarz funtions , @xmath112 , and the space of distributions , @xmath113 .
we denote by @xmath114 the fourier transform with respect to the @xmath115 variables .
integration spaces will often be indexed with the variable that is being integrated : @xmath116 will denote the space of integrable functions in @xmath19 with values in @xmath117 , where the variable is @xmath118 .
furthermore , if no confusion can arise , we omit the domain in the notation of the integration space .
let us recall the numerable construction . fix a sequence of wave speeds @xmath42 , with @xmath43 , @xmath44 .
define the spaces @xmath119 and @xmath120 [ injectiva ] if @xmath121 is such that @xmath122 then @xmath123 .
fix @xmath124 , @xmath125 and define @xmath126 by absurd , suppose that there exists an open interval @xmath127a , b[$ ] such that @xmath128a , b [ ) } = \delta > 0.\ ] ] since @xmath121 , there exists @xmath129 such that @xmath130 on the other hand , for @xmath131 large enough , one has @xmath132a , b [ ) } < \delta/2.\ ] ] hence @xmath133a , b [ ) } \le \sum_{n\ge 1,n\neq k } \|f_n(\cdot + ( c_k - c_n)h)\|_{l^2(]a , b [ ) } < \delta,\ ] ] which is impossible .
since each element of @xmath134 and @xmath135 may be represented in a unique way , we define the norms of these spaces as the norm induced by the profile space : @xmath136 [ somadirecta ] consider @xmath51 and @xmath137 as subspaces of @xmath113 . then @xmath138 .
suppose that @xmath121 is such that the function @xmath36 defined by @xmath139 is in @xmath51 .
fix @xmath124 and define @xmath140 given @xmath141 , it is easy to check that , since @xmath142 , @xmath143 through a similar argument to that of the previous proof , one may check that , when @xmath144 , @xmath145 hence one has necessarily @xmath146 and so @xmath147 .
since @xmath124 is arbitrary , one concludes @xmath148 .
define the following subspaces of @xmath113 : @xmath149 notice that it follows from the previous lemma that these sums are indeed direct sums .
moreover , consider @xmath150 [ decomposicao ] fix @xmath52 .
take @xmath151 and @xmath152 .
writing @xmath153 define @xmath154 and @xmath155 then @xmath156 and @xmath157 . since , for any @xmath124 , @xmath158 one has @xmath159 and so @xmath156 . for
the second part of the result , recall the classical estimates @xmath160 and @xmath161 this implies that @xmath162 the first term , @xmath163 , is in @xmath164 , by sobolev s injection .
the second term is clearly in @xmath7 . finally , we prove that the third term is also in @xmath7 : @xmath165 hence @xmath166 .
_ write @xmath109 , @xmath73 , @xmath72 , @xmath167 we claim that , over @xmath59 , the initial value problem for ( nls ) is equivalent to the initial value problem @xmath168 indeed , if @xmath33 and @xmath169 are solutions of this problem , it is trivial to check that @xmath170 is a solution of ( nls ) . on the other hand ,
suppose that @xmath40 is a solution of ( nls ) with initial condition @xmath63 .
decompose @xmath40 as @xmath171 and write @xmath172 then @xmath173 since the left hand side is in @xmath51 and the right hand side is in @xmath135 , by lemma [ somadirecta ] , both sides must be equal to zero : @xmath174 furthermore , by lemma [ injectiva ] , the second equation is equivalent to the infinite system @xmath175 , which proves the claim . _ step 2 .
_ we solve the infinite system . for each @xmath176 , define @xmath177 then @xmath178 since @xmath179 , @xmath180 as @xmath181 .
now consider the initial value problem @xmath182 it follows from the @xmath183 local well - posedness results for ( nls ) that there exists a time @xmath184 and a unique maximal solution @xmath185 of the above problem .
moreover , since @xmath186 , for @xmath187 sufficiently large , @xmath188 . define the common time of existence , @xmath189 then each @xmath190 is define on @xmath191 .
if @xmath192 , then , as @xmath193 , @xmath194 setting @xmath195 it is clear that the sequence @xmath169 is the unique solution of on @xmath191 .
moreover , considering that @xmath196 if @xmath192 , one has @xmath197 finally , it follows from the conservation of the @xmath7 norm that @xmath198 setting @xmath199 this implies that @xmath200 .
_ we now solve . by lemma
[ decomposicao ] and kato s local well - posedness result in @xmath4 , has a unique maximal solution @xmath33 , defined on @xmath201 , @xmath202 , with initial condition @xmath203 .
furthermore , if @xmath204 , @xmath205 _ step 4 .
_ conclusion .
it follows from the previous steps that @xmath206 is the unique solution of ( nls ) on @xmath34 , with initial data @xmath63 . if @xmath207 , then either @xmath204 or @xmath192 . in any case , @xmath208 which implies that @xmath40 is not extendible over @xmath34 .
the continuous dependence on the initial data is a consequence of the continuous dependence given by the local well - posedness results for @xmath33 and @xmath169 .
[ decaimento ] set @xmath209 . given @xmath67
, there exists @xmath210 such that , given @xmath72 satisfying @xmath211 the solution @xmath36 of ( nls ) with initial data @xmath32 is global and satisfies @xmath212 we write @xmath213 where @xmath48 is a solution of @xmath214 using the rescaling @xmath215 , one arrives to @xmath216 _ step 1 .
_ we now collect some properties of @xmath190 .
first of all , it follows from ( * ? ? ?
* theorem 5.3.1 ) that there exists @xmath217 such that , if @xmath218 , then @xmath190 is global and @xmath219 from decay estimates , which are valid regardless of the sign of @xmath220 , one also has @xmath221 these properties imply that @xmath222 now we obtain an uniform estimate for @xmath223 .
recall , from the proof of theorem [ existnumeravel ] , that there exists @xmath50 ( independent of @xmath224 ) such that @xmath190 is defined on @xmath225 $ ] and @xmath226 , for @xmath227 .
hence @xmath228 for @xmath229 , since @xmath230 one has @xmath231 where @xmath232
thus we obtain the estimate @xmath233 _ step 2 .
_ now we write the estimates of step 1 in terms of @xmath36 .
some simple computations show that , for any @xmath234 , @xmath235 @xmath236 notice that , if @xmath237 , then @xmath238 hence @xmath239 and , in a similar fashion , @xmath240 we shall only prove the case @xmath209 , the general being completely analogous . the main idea of the proof is to obtain a `` small data global existence '' result for the @xmath6 component ( see , for example , ( * ? ? ? * theorem 6.2.1 ) ) .
_ from the previous lemma , it follows that , for @xmath241 sufficiently small , @xmath36 is global and @xmath242 fix @xmath73 and consider the corresponding solution @xmath33 of @xmath243 we recall that @xmath33 is defined on @xmath244 , where @xmath109 . since @xmath36 is global in @xmath134 , the blow - up alternative of theorem [ existnumeravel ] then implies that , if @xmath11 , @xmath245 we develop the nonlinear part as @xmath246 where each @xmath247 has exactly @xmath248 powers of @xmath33 .
define , for @xmath249 , @xmath250 and @xmath251 such that @xmath252 is an admissible pair , that is , @xmath253 .
in particular , @xmath254 .
consider , for @xmath255 , @xmath256 we write duhamel s formula , @xmath257 therefore , for any admissible pair @xmath258 , @xmath259 for the sake of simplicity , we shall omit both the temporal and spatial domains . in the next steps
, we shall estimate each term of the sum by a suitable power of @xmath260 .
all constants depending solely on @xmath261 shall be ommited .
estimate of higher - order terms in @xmath33 on .
_ here , we shall estimate @xmath262 take @xmath263 .
then it follows from step 1 that @xmath264 now we treat the case @xmath265 : @xmath266 where @xmath267 then , through the interpolation @xmath268 and the injection @xmath269 , @xmath270 _ step 3 .
estimate of the linear term in @xmath33 .
_ @xmath271 using the properties deduced in step 1 , @xmath272 _ step 4 .
estimate of the term independent on @xmath33 .
_ define @xmath273 .
then @xmath274 recalling that @xmath275 one estimates @xmath276 hence , by lemma [ decaimento ] , @xmath277 _ step 5 .
estimate of the quadratic term in @xmath33 .
_ recalling that @xmath278 are bounded in @xmath279 , one has @xmath280 _ step 6
. conclusion .
_ putting together steps 2 , 3 , 4 and 5 , there exists a constant @xmath281 , depending only on @xmath261 , such that @xmath282 for @xmath39 sufficiently small , we arrive at @xmath283 if @xmath284 is sufficiently small , then the above inequality implies @xmath285\cup [ h_1 , \infty)$ ] , for some @xmath286 .
since @xmath287 , by continuity , one has @xmath288 , for all @xmath289 .
the blow - up alternative then implies that @xmath290 .
this implies that @xmath291 now notice that this property is also valid for @xmath292 , since it is a solution of ( nls ) with @xmath293 . hence @xmath294
we recall the definition of the plane wave transform : @xmath295 now we state some simple properties of this transform , whose proof is quite straightforward . fix any @xmath85 . 1 .
translation property : @xmath296 2 .
scaling property : @xmath297 3 .
monotonicity : if @xmath298 , @xmath299 4 .
derivation : if @xmath300 , @xmath301 [ integrabilidadelp ] fix @xmath85 .
then @xmath302 and @xmath303 consequently , @xmath304 can be continuously extended in a unique way to @xmath305 , for any @xmath306 .
take @xmath307 .
then , for @xmath308 , @xmath309 this implies that @xmath310 on the other hand , given @xmath311 , for a.e .
@xmath312 , @xmath313 [ estareml4 ] as a consequence of the above result , for any @xmath85 , @xmath314 indeed , @xmath315 it is a simple exercise to compute the transform of the characteristic function of the unit square : if @xmath316 ^ 2}$ ] , then @xmath317 [ transformada ] ^2}$].,title="fig:",width=264,height=188 ] we claim that @xmath84 .
in fact , one has @xmath318 [ naointegrabilidadel2 ] if @xmath319 satisfies @xmath320 and @xmath321 , then @xmath84 . since @xmath321 and @xmath320
, there exists @xmath322 and a square @xmath323 such that @xmath324 , which implies that @xmath325 . using the translation and scaling properties
, one may write the transform of @xmath326 in terms of @xmath327 ^ 2}$ ] , which does not belong to @xmath5 . consider @xmath316 ^ 2 } - \mathds{1}_{[0,1]\times [ 1,2]}$ ] .
then , using the expression for @xmath327 ^ 2}$ ] and the translation property , it is easy to check that @xmath328 . let s compute the transform of @xmath329 : @xmath330 it is interesting that @xmath22 is the kernel of the 2d - heat kernel in fourier variables at time @xmath331 , and that its transform is the 1d - heat kernel in physical variables at time @xmath332 .
[ propformula ] given @xmath333 , one has @xmath334(x),\quad a.e.\ ( x , y)\in{\mathbb{r}}^2\ ] ] where @xmath335 denotes the fourier transform in the @xmath336 variable . consequently , one has the following inversion formula @xmath337(z , c ) , \quad a.e.\ ( z , c)\in{\mathbb{r}}^2\ ] ] first of all , since @xmath333 , one has @xmath338 . for a.e .
@xmath312 , one may then take the fourier transform of @xmath339 : @xmath340(\xi)=\int f(x - cy , c)e^{-2\pi i x\xi } dxdc = \int f(z , c)e^{-2\pi
iz\xi-2\pi i cy\xi}dzdc = ( \mathcal{f}_{z , c}f)(\xi , y\xi).\ ] ] take @xmath341 .
for each @xmath342 , define @xmath343 and let @xmath344 be the usual trace operator on @xmath345 .
moreover , consider the isomorphism @xmath346 @xmath347 and @xmath348 .
then , for a.e .
@xmath312 , @xmath340 = \pi_y\mathcal{f}_{z , c}f.\ ] ] take @xmath349 in @xmath350 , @xmath351 .
in particular , @xmath352 hence , for a.e @xmath353 , using the previous result , @xmath340 = \lim \mathcal{f}_x[tf_n(\cdot , y ) ] = \lim \pi_y\mathcal{f}_{z , c } f_n = \pi_y\mathcal{f}_{z , c } f.\ ] ] [ transform of the product of two functions ] for @xmath354 , @xmath355 the result is a simple application of proposition [ formula ] .
first of all , notice that @xmath356 hence @xmath357 [ parseval s identity for the plane wave transform][parseval ] for any @xmath358 , if @xmath359 and @xmath360 , one has @xmath361 if @xmath354 , the result follows from @xmath362 the general case is obtained by a density argument .
[ unicarepresentacao ] if @xmath363 is such that @xmath364 , then @xmath365 .
fix @xmath366 , @xmath367 .
let @xmath368 be such that @xmath369 in @xmath370 $ ] and @xmath371 in @xmath372 $ ] .
for any @xmath241 , define @xmath373 it is clear that @xmath374 has support outside the strip @xmath375 .
furthermore , @xmath376 setting @xmath377 it follows that @xmath378 and so @xmath379 . by proposition
[ formula ] , @xmath380 . using proposition [ parseval ] , @xmath381 and so @xmath365 .
[ integrabilidadel2 ] take @xmath333
. then @xmath382 consequently , if the spectrum of @xmath22 lies in @xmath383-m,-\epsilon[\cup ] \epsilon , m[)\times ] -m , m[$ ] , for some @xmath384 , @xmath328 .
this is a direct consequence of the formula for @xmath304 in terms of the fourier transform .
take @xmath333 and @xmath385
. then @xmath386 by definition , @xmath387 [ convolucao ] take @xmath388 . if @xmath389 , that is , @xmath390 , then @xmath391 more generally , for any @xmath392 , one has @xmath393 and , for @xmath394 , @xmath395 .
the result is trivial for @xmath394 . given @xmath392 , one has @xmath396 corollary [ convolucao ] and proposition [ integrabilidadelp ] give a new proof of young s inequality : @xmath397 moreover , since , for a.e .
@xmath312 , @xmath398 when @xmath399 , by dominated convergence theorem , @xmath400 which is a new way to prove convergence of mollifiers .
in this short section , we apply the plane wave transform to solve some classical equations .
consider the wave equation in @xmath0 @xmath401 if one supposes that @xmath402 , one arrives to @xmath403 which may be explicitly solved : @xmath404 notice that we introduced on purpose the speed @xmath21 as an independent variable of @xmath22 . now , taking the transform of @xmath405 , @xmath406 one obtains a solution of the 2d wave equation .
we remark that , unlike poisson s formula for classical solutions , this representation does not involve derivatives of @xmath407 .
take the linear schrdinger equation : @xmath408 introducing the plane wave ansatz @xmath402 , @xmath409 the solution is given by @xmath410 which , through the plane wave transform , gives a family of solutions to the two dimensional schrdinger equation .
@xmath411 [ calor ] we now consider a slightly different application of the plane wave transform .
take the heat equation in one dimension : @xmath412 here , instead of taking a plane wave in two spatial dimensions , we consider a plane wave in time and space .
the ansatz @xmath413 implies that @xmath414 setting @xmath415 and applying the transform @xmath304 , one arrives to the following family of solutions to the heat equation : @xmath416 actually , if one is given an initial condition @xmath417 , then one observes that @xmath418 is the inverse laplace transform of @xmath63 .
although this integral representation is certainly valid , the presence of the term @xmath419 implies a strong decay of @xmath418 at infinity .
now , _ if one replaces @xmath21 with @xmath420 in the integral representation , one still obtains a solution to the heat equation_. this procedure makes no sense when one starts with the ansatz @xmath413 ; however , the integral representation is still meaningful for complex values of @xmath21 .
hence another family of solutions is @xmath421 in this situation , one sees that @xmath422 is none other than the inverse fourier transform of @xmath63 and so @xmath423 which is precisely the solution of the heat equation given by the fourier transform .
this integral representation was studied in @xcite as a generalization of the fourier transform .
[ schrodinger ] as in the previous example , one may derive a family of solutions to the 1d - schrdinger equation using @xmath304 in both space and time : @xmath424 givan an initial condition @xmath63 , @xmath418 corresponds to the fourier transform of @xmath63 .
the substitution @xmath425 gives the family @xmath426 which is connected to the laplace transform .
we finish with a result stating that the above classical arguments to reduce the dimension of the equations are valid in a functional setting .
the result , although stated only for the linear schrdinger equation , may also be extended to more equations .
[ semigrupot ] let @xmath427 and @xmath99 be the free schrdinger groups in dimensions one and two , respectively .
given @xmath363 , one has @xmath428 suppose that @xmath429 . then , from the formula of proposition [ propformula ] , @xmath430(\xi , y ) = \mathcal{f}_{x}\left [ \mathcal{f}_{\xi,\eta}^{-1}\left(e^{-4\pi^2i(\xi^2+\eta^2)t}\mathcal{f}_{x , y}tf\right)\right](\xi , y)\\ = \ & \mathcal{f}_{\eta}^{-1}\left[e^{-4\pi^2i(\xi^2 + \eta^2)t}\mathcal{f}_{x , y}tf\right](\xi , y ) = \mathcal{f}_{\eta}^{-1}\left[e^{-4\pi^2i(\xi^2+\eta^2)t}\mathcal{f}_{y}\left((\mathcal{f}_{z , c}f)(\xi , y\xi)\right)\right](\xi , y)\\ = \ & \mathcal{f}_{\eta}^{-1}\left[e^{-4\pi^2i(\xi^2+\eta^2)t}\frac{1}{|\xi|}\left((\mathcal{f}_{z}f)(\xi,-\eta/\xi)\right)\right](\xi , y ) = \mathcal{f}_{\eta}\left[e^{-4\pi^2i(\xi^2+\eta^2)t}\frac{1}{|\xi|}\left((\mathcal{f}_{z}f)(\xi,\eta/\xi)\right)\right](\xi , y)\\ = \ & \mathcal{f}_{\eta}\left[e^{-4\pi^2i(1+\eta^2)\xi^2t}\left((\mathcal{f}_{z}f)(\xi,\eta)\right)\right](\xi , y\xi)=\mathcal{f}_{z,\eta}\left[\mathcal{f}_{\xi}^{-1}e^{-4\pi^2i(1+\eta^2)\xi^2t}\left((\mathcal{f}_{z}f)(\xi,\eta)\right)\right](\xi , y\xi)\\=\ & \mathcal{f}_{x}t(s_1((1+c^2)t)f(c))(\xi , y).\end{aligned}\ ] ] the general case follows by a density argument .
consider @xmath431 endowed with the induced norm @xmath432 notice that such a norm is well - defined , by corollary [ unicarepresentacao ] . furthermore , since @xmath433 , one has , by proposition [ integrabilidadelp ] and remark [ estareml4 ] , @xmath434 now consider the subspace of @xmath435 @xmath436 endowed with the semi - norm @xmath437 the semi - norm @xmath438 is a norm in @xmath34 .
moreover , @xmath439 is a banach space and @xmath440 .
take @xmath441 .
if @xmath206 , @xmath151 , @xmath442 , one has @xmath443 taking the infimum on the right hand side , @xmath444 hence , if @xmath445 , @xmath446 , which proves that @xmath438 is a norm .
we now prove that @xmath34 , with the norm @xmath438 , is complete .
given @xmath447 , suppose that @xmath448 for each @xmath124 , there exist @xmath449 , increasing with @xmath450 , such that @xmath451 define @xmath452 and take @xmath453 and @xmath454 , inductively , such that @xmath455 and @xmath456 this implies that @xmath457 and @xmath458 are cauchy sequences in @xmath4 and @xmath459 , respectively , and so @xmath460 hence @xmath461 and so @xmath462 in @xmath34 . since @xmath463 is cauchy in @xmath34 , this implies that @xmath464 .
let @xmath465 be the schrdinger group in dimension two . then @xmath466 given @xmath441 , write @xmath206 , @xmath151 , @xmath442 , @xmath467 .
then , by proposition [ semigrupot ] , @xmath468 hence @xmath469 and , taking the infimum on both sides , one concludes the proof . for the sake of clarity ,
we define do we mean as a solution of ( nls ) over @xmath34 .
[ defisolucao ] given @xmath90 and @xmath470 , e)$ ] , @xmath50 , we say that @xmath40 is a solution of ( nls ) with initial data @xmath63 if @xmath40 satisfies the duhamel formula @xmath471 where @xmath99 is the schrdinger group in dimension two .
[ solucaounica ] for any @xmath472 and @xmath90 , if @xmath473 , e)$ ] , @xmath50 , are two solutions of ( nls ) with initial data @xmath63 , then @xmath474 .
since @xmath475 are two solutions with the same initial data , @xmath476 let @xmath477 be such that @xmath478 . take @xmath479 such that @xmath480 then , since @xmath481 using hlder s inequality , @xmath482 define @xmath483 and @xmath484 such that @xmath485 and @xmath258 are admissible pairs , i.e. , @xmath486 for any @xmath487 $ ] , @xmath488 and for @xmath489 , it follows that @xmath490 using strichartz s estimates in , @xmath491 since @xmath492 , @xmath493 where @xmath494 does not depend on @xmath495 .
the result now follows from ( * ? ? ?
* lemma 4.2.2 ) . for any @xmath52 and @xmath441 , @xmath496 .
writing @xmath206 , @xmath497 since @xmath151 , @xmath498 and so @xmath499 . on the other hand , recalling that @xmath500 , for any @xmath501 , @xmath502 .
hence @xmath503 . write @xmath109 , @xmath73 , @xmath504 .
consider the initial value problem for @xmath505 , h^1({\mathbb{r}}^2))$ ] , @xmath506 by the previous lemma , the nonlinear term is in @xmath51 .
it follows from kato s local existence theory ( see , for example , ( * ? ? ?
* theorem 4.4.1 ) ) that there exists @xmath507 and @xmath505 , h^1({\mathbb{r}}^2))$ ] solution of , which depends continuously on @xmath203 and @xmath32 . setting @xmath508 ,
@xmath509 , it becomes clear that @xmath40 is a solution in @xmath34 of ( nls ) .
moreover , it is unique , by lemma [ solucaounica ] .
define @xmath510,e ) \mbox { solution of ( nls)},\ u(0)=u_0\}.\ ] ] if , for some @xmath511 , @xmath512 as @xmath513 , then , for each @xmath514 , one may find @xmath515 and @xmath516 such that @xmath517 for @xmath518 sufficiently close to @xmath519 , @xmath520 then one may solve with initial data @xmath521 and @xmath522 , thus obtaining a unique solution of ( nls ) defined on @xmath523 $ ] .
this implies that @xmath40 is extendible beyond @xmath519 , a contradiction .
hence there exists a unique maximal solution @xmath524 of ( nls ) and , if @xmath525 , @xmath526 the proof is very similar to that of the numerable case and again will only be done for @xmath209 .
we start with some estimates for @xmath527 : using propositions [ integrabilidadelp ] and [ semigrupot ] , @xmath528 @xmath529 the theorem will be proved if we show that the unique solution @xmath33 of @xmath530 is global and remains small in the @xmath6 norm , for all @xmath531 .
as before , we decompose the nonlinear term as @xmath532 define @xmath131 as in and apply some strichartz estimates in the duhamel s formula for @xmath33 in order to obtain inequality .
the estimates for @xmath533 are _ mutatis mutandis _ those that were derived for the numerable case , since one has control of the @xmath534 norm of @xmath527 . for the autonomous term , using the decay of @xmath535 , @xmath536 let @xmath368 be such that @xmath537 , @xmath369 in @xmath370 $ ] and @xmath371 in @xmath372 $ ] .
take @xmath538 such that @xmath539 for any @xmath241 , define @xmath540 then @xmath541 and @xmath542 finally , take @xmath429 such that @xmath543 defining @xmath544 , it is now easy to check that @xmath545 satisfies the conditions of theorem [ estabilidadecontinuo ] .
moreover , by corollary [ integrabilidadel2 ] , for @xmath241 small @xmath546 all that is left is to prove that , for @xmath241 small enough , @xmath547 .
this follows from @xmath548 suppose that @xmath549 . if @xmath550 then , for some @xmath551 , @xmath552 from proposition [ semigrupot ] and remark [ estareml4 ] , @xmath553 furthermore , from proposition [ integrabilidadelp ] , @xmath554 the estimate for @xmath555 follows by interpolation .
now take @xmath556 . by proposition
[ integrabilidadelp ] , @xmath557 using hlder s inequality , @xmath558 recall the classical estimate , valid a.e . for @xmath19 : @xmath559 hence , setting @xmath560 we have @xmath561 by a analogous reasoning to that of remark [ estareml4 ] and , @xmath562 the theorem will be proved once we show that the unique solution of @xmath563 is globally defined . in the following ,
we perform formal calculations which can be justified with suitable regularization arguments .
multiplying by @xmath564 , integrating over @xmath0 and taking the imaginary part , we obtain @xmath565 and so @xmath566 since , by the previous lemma , @xmath567 and @xmath568 are uniformly bounded , gronwall s lemma implies that @xmath569 where @xmath570 is a non decreasing continuous function .
now , multiplying by @xmath571 , integrating and taking the real part , @xmath572 recalling that @xmath573 and applying the previous lemma , @xmath574 define the energy @xmath575 from gagliardo - nirenberg s inequality , @xmath576 let @xmath519 be the maximal time of existence of the solution of .
then , for any @xmath577 , @xmath578 suppose , by contradiction , that @xmath525 . since @xmath579 as @xmath513 ,
one has , for @xmath580 sufficiently close to @xmath519 , @xmath581 however , this implies that @xmath582 is bounded on @xmath583 $ ] , which is absurd .
the plane wave transform has many interesting properties .
section 4 should be regarded as a starting point for the application of this transform to obtain new local well - posedness results for equations such as the wave equation , the heat equation and kdv equation .
furthermore , theorem [ grandesdados ] points out that this theory is not simply an appendix to the usual @xmath584 local well - posedness theory : it gives new information about @xmath584 solutions and may shine some new light on global existence criteria for these equations .
it is important to recall that the plane transform is built upon solutions of the transport equation .
other equations may give birth to other transforms : for example , consider @xmath585 solutions of this equation are of the form @xmath586 , @xmath587 , that is , functions that are periodic in the @xmath588 direction . by varying the direction of @xmath588
, we may arrive to a similar construction to that of the transform @xmath304
. it would be especially important to find a construction which can be applied to other spatial domains .
one may also try to extend the plane wave transform to more spatial variables , either by applying only to the two first variables or by iteration ( applying to the first and to the @xmath248-th variable ) .
the functional tools of section 3 should be easily applicable to either extension .
the authors were partially suported by fundao para a cincia e tecnologia , through the grant uid / mat/04561/2013 .
the first author was also partially supported by fundao para a cincia e tecnologia , through the grant sfrh / bd/96399/2013 .
the authors thank rmi carles for important discussions and comments .
in this appendix , we give another proof of @xmath589 this proof is based on arguments that do not involve the expression of @xmath304 in terms of the fourier transform . for the sake of simplicity , we shall assume that @xmath22 is good enough so that all integrations are automatically justified .
reduction to radial case .
_ since @xmath591 , @xmath592 then , given any @xmath593 , @xmath594 which means that the integration of @xmath22 over any non - vertical strip is null .
define @xmath595 then , for any @xmath596 and any strip @xmath597\epsilon-\delta,\epsilon+\delta[\times{\mathbb{r}}$ ] , @xmath598 since the rotation of a strip is still a strip .
in particular , the function @xmath599 is zero .
since @xmath600 is a radial function , @xmath601 is an even function , and so @xmath602 hence @xmath603 which implies , in polar coordinates , that @xmath604 _ step 2 .
@xmath600 must be zero .
_ we write @xmath605 , with @xmath606 and @xmath607 .
fix @xmath608 .
w.l.o.g . , let @xmath609 be a strictly increasing sequence with @xmath610 and such that @xmath611}\ge 0,\quad \tilde{f}\big|_{[a_{2k+1},a_{2k+2}]}\le 0 , k\in{\mathbb{n}}.\ ] ] there are two possibilities : either @xmath612 is bounded or not .
we consider the case where @xmath612 is not bounded , the other being quite similar .
define @xmath613 and @xmath614 since @xmath615 for @xmath616 , it is easy to check that , for @xmath617 even , @xmath618 this implies that , for all @xmath617 even , @xmath619 . on the other hand , the same reasoning shows that , for all @xmath617 odd , @xmath620 .
given @xmath621 , it follows from and from the decay of @xmath22 that @xmath622 since , for any @xmath450 , @xmath623 it follows that @xmath624 set @xmath625 . by monotone convergence theorem ,
the functions @xmath626 and @xmath627 are integrable , positive and have integral over @xmath628 equal to 0 .
hence @xmath629 .
it follows from and that @xmath630 for @xmath631 . since @xmath322 is arbitrary ,
we conclude that @xmath632 .
_ since @xmath600 is the average of @xmath22 over any circle centered at the origin and @xmath632 , it follows that @xmath633 . to prove that @xmath634 for any @xmath635 , consider @xmath636 then @xmath637 and therefore , by the previous steps , @xmath638 .
it follows directly from this proof that the integral of @xmath22 over the collection of all strips in @xmath0 determines uniquely the values of @xmath22 .
this implies that the integral of @xmath22 over all strips determines the integral of @xmath22 over all balls .
this is not trivial at all , since the value of the latter can not be obtained from the first ones using algebraic set relations .
furthermore , it is easy to check that , if one starts with a collection of strips with a finite number of possible directions , such a result is no longer valid .
simo correia + cmaf - cio and fcul + campo grande , edifcio c6 , piso 2 , 1749 - 016 lisboa ( portugal ) + ` [email protected] ` + mrio figueira + cmaf - cio and fcul + campo grande , edifcio c6 , piso 2 , 1749 - 016 lisboa ( portugal ) + ` [email protected] ` + | we consider the cauchy problem for the nonlinear schrdinger equation on @xmath0 , @xmath1 , @xmath2 , @xmath3 .
we introduce new functional spaces over which the initial value problem is well - posed .
their construction is based on _ spatial plane waves _
( cf .
@xcite ) .
these spaces contain @xmath4 and do not lie within @xmath5 .
we prove several global well - posedness and stability results over these new spaces , including a new global well - posedness result of @xmath6 solutions with indefinitely large @xmath6 and @xmath7 norms .
some of these results are proved using a new functional transform , the _ plane wave transform_. we develop a suitable theory for this transform , prove several properties and solve classical linear pde s with it , highlighting its wide range of application . * keywords * : nonlinear schrdinger equation ; local well - posedness ; stability ; spatial plane waves ; integral transform . *
ams subject classification 2010 * : 35b35 , 35c15 , 35q55 , 35a01 , 44a05 . |
we examine the results of a set of simulations of collisionless collapse of a stellar system and focus on the fit of the final quasi - equilibrium configurations by means of the so - called @xmath0 models ( see @xcite , @xcite ) .
the models , constructed on the basis of theoretical arguments suggested by statistical mechanics , are associated with realistic luminosity and kinematic properties , suitable to represent elliptical galaxies .
all the simulations have been performed using a new version @xcite of the original code of van albada ( see @xcite ; the code evolves a system of simulation particles interacting with one another via a mean field calculated from a spherical harmonic expansion of a smooth density distribution ) , starting from various initial conditions characterized by approximate spherical symmetry and by a small value of the virial parameter @xmath1 ( @xmath2 ; here @xmath3 and @xmath4 represent total kinetic and gravitational energy ) . from such initial conditions , the collisionless
gravitational plasma " evolves undergoing incomplete violent relaxation @xcite .
the models , in spite of their simplicity and of their spherical symmetry , turn out to describe well and in detail the end products of our simulations .
we should note that the present formulation neglects the presence of dark matter .
since there is convincing evidence for the presence of dark matter halos also in elliptical galaxies , this study should be extended to the case of two - component systems , before a satisfactory comparison with observed galaxies can eventually be claimed .
most simulations have been run with @xmath5 particles ; the results have been checked against varying the number of particles .
the units chosen are @xmath6 for mass , @xmath7 for length , and @xmath8 for time . in these units
the gravitational constant is @xmath9 , the mass of the model is @xmath10 , and the initial radius of the system @xmath11 .
we focus on clumpy initial conditions .
these turn out to lead to more realistic final configurations ( @xcite , @xcite ) .
they can be interpreted as a way to simulate the merging of several smaller structures to form a galaxy .
the clumps are uniformly distributed in space , within an approximate spherical symmetry ; the ellipticity at the beginning of the simulation is small , so that the corresponding initial shape would resemble that of @xmath12 galaxies .
we performed several runs varying the number of clumps and the virial ratio @xmath1 in the range from @xmath13 to @xmath14 .
usually the clumps are cold , i.e. their kinetic energy is all associated with the motion of their center of mass .
the simulations are run up to time @xmath15 , when the system has settled into a quasi - equilibrium .
the violent collapse phase takes place within @xmath16 .
the final configurations are quasi - spherical , with shapes that resemble those of @xmath12 galaxies .
the final equilibrium half - mass radius @xmath17 is basically independent of the value of @xmath1 .
the central concentration achieved is a function of the initial virial parameter @xmath1 , as can be inferred from the conservation of maximum density in phase space @xcite .
for a spherically symmetric kinetic system , extremizing the boltzmann entropy at fixed values of the total mass , of the total energy , and of an additional third quantity @xmath18 , defined as @xmath19 , leads ( @xcite ; see also @xcite ) to the distribution function @xmath20}$ ] , where @xmath21 , @xmath22 , @xmath23 , and @xmath24 are positive real constants ; here @xmath25 and @xmath26 denote single - star specific energy and angular momentum . at fixed value of @xmath24
, one may think of these constants as providing two dimensional scales ( for example , @xmath27 and @xmath18 ) and one dimensionless parameter , such as @xmath28 . in the following
we will focus on values of @xmath24 ranging from 3/8 to 1 .
the corresponding models are constructed by solving the poisson equation for the self - consistent mean potential @xmath29 generated by the density distribution associated with @xmath0 . at fixed value of @xmath24 ,
the models thus generated make a one - parameter family of equilibria , described by the concentration parameter @xmath30 , the dimensionless depth of the central potential well .
the projected density profile of the models is very well fitted by the @xmath31 law ( for a definition of the law and of the effective radius @xmath32 , see @xcite ) , with the index @xmath33 varying in the range from @xmath10 to @xmath34 , depending on the precise value of @xmath24 and on the concentration parameter @xmath35 . at worst , the residuals from the @xmath31 law are less than @xmath36 mag over a range of more than @xmath37 mag , while in general the residuals are less than @xmath38 mag in a range of more than @xmath39 mag
. the models are isotropic in the inner regions , while they are characterized by almost radial orbits in the outer parts .
the local value of pressure anisotropy can be measured by @xmath40 .
the form of the transition from isotropy ( @xmath41 ) to radial anisotropy ( @xmath42 ) is governed by the index @xmath24 : higher values of @xmath24 give rise to a sharper transition .
the anisotropy radius , that is the location where the transition takes place ( with @xmath43 ) , is close to the half - mass radius of the models @xcite .
in order to study the output of the simulations we compare the density and the anisotropy profiles , @xmath44 and @xmath45 , of the end products with the theoretical profiles of the @xmath0 family of models ; smooth simulation profiles are obtained by averaging over time , based on a total of @xmath46 snapshots taken from @xmath47 to @xmath15 . for the fitting models ,
the parameter space explored is that of an equally spaced grid in @xmath48 , with a subdivision of @xmath49 in @xmath24 , from @xmath50 to @xmath11 , and of @xmath14 in @xmath35 , from @xmath14 to @xmath51 ; the mass and the half - mass radius of the models are fixed by the scales set by the simulations .
a least @xmath52 analysis is performed , with the error bars estimated from the variance in the time average process used to obtain the smooth simulation profiles . a critical step in this fitting procedure
is the choice of the relative weights for the density and the pressure anisotropy profiles .
we adopted equal weights for the two terms , checking a posteriori that their contributions to @xmath52 are of the same order of magnitude .
as exemplified by fig . [ fig : simc2 ] , the density of all the simulations is well represented by the best fit @xmath0 profile over the entire radial range .
the fit is satisfactory not only in the outer parts , where the density falls under a threshold value that is _ nine orders of magnitude _ smaller than that of the central density , but also in the inner regions , where , in principle , there could be problems of numerical resolution .
we have performed simulations with different numbers of particles , without noticing significant changes in the relevant profiles and in the quality of the fits .
the successful comparison between models and simulations is also interesting because , depending on the initial virial parameter @xmath1 , the less concentrated end products are fitted by a density profile which , if projected along the line - of - sight , exhibits different values of @xmath33 in an @xmath31 best fit analysis . in turn , this may be interpreted in the framework of the proposed weak homology of elliptical galaxies @xcite . to some extent , the final anisotropy profiles for clumpy initial conditions are found to be sensitive to the detailed choice of initialization . in other words , runs starting from initial conditions with the same parameters , but with a different _ seed _ in the random number generator , give rise to slightly different profiles .
the agreement between the simulation and the model profiles is good ( see fig . [
fig : simc2 ] ) , except for about @xmath53 irregular " cases ; the origin of these is not clear , and we argue that they correspond to inefficient mixing in phase space during collapse ( a discussion of these issues will be provided in a separate paper , currently in preparation ) . at the level of phase space ,
we have performed two types of comparison , one involving the energy density distribution @xmath54 and the other based on @xmath55 .
the chosen normalization factors are such that : @xmath56 in fig .
[ fig : simc2 ] we plot the final energy density distribution for a simulation run called @xmath57 ( a regular case ) with respect to the predictions of the best fit model identified from the study of the density and pressure anisotropy distributions .
the agreement is very good , especially for the strongly bound particles . in some simulations ,
the distribution of less bound particles shows some deviations from the theoretical expectations , especially in the irregular " cases , when the final @xmath54 exhibits a double peak ; one peak is around @xmath58 , as expected , while the other is located at the point where there was a peak in @xmath54 at the initial time @xmath59 .
we interpret this feature as a signature of inefficient phase mixing . finally , at the deeper level of @xmath55
, simulations and models also agree very well , as illustrated in the right set of four panels of fig .
[ fig : simc2 ] . for the case
shown , the distribution contour lines essentially coincide in the range from @xmath60 to @xmath61 ; however , the theoretical model shows a peak located near the origin , corresponding to an excess of weakly bound stars on almost radial orbits .
, called @xmath57 ( starting from 20 cold clumps of radius @xmath62 ) , and the best fit @xmath0 model ( @xmath63 , @xmath64 ) .
_ left set of four panels _ : density as measured from the simulation ( crosses ) vs. the best fit profile ( top left ) .
residuals ( in magnitudes ) from the @xmath65 ( dotted line ) and from the @xmath31 ( with @xmath66 ) law for the best fit projected - density profile ( top right ) .
anisotropy profile of the simulation ( crosses ) vs. the best fit profile ( bottom left ) .
energy density distribution @xmath54 ( bottom right ) .
_ right set of four panels _ : final phase space density @xmath55 ( left column ) , compared with that of the best fitting @xmath0 model ( right column ) .
the model curve for @xmath54 and surface for @xmath55 have been computed by a monte carlo sampling of phase space .
[ fig : simc2],title="fig : " ] , called @xmath57 ( starting from 20 cold clumps of radius @xmath62 ) , and the best fit @xmath0 model ( @xmath63 , @xmath64 ) . _ left set of four panels _ : density as measured from the simulation ( crosses ) vs. the best fit profile ( top left )
. residuals ( in magnitudes ) from the @xmath65 ( dotted line ) and from the @xmath31 ( with @xmath66 ) law for the best fit projected - density profile ( top right ) .
anisotropy profile of the simulation ( crosses ) vs. the best fit profile ( bottom left ) .
energy density distribution @xmath54 ( bottom right ) .
_ right set of four panels _ : final phase space density @xmath55 ( left column ) , compared with that of the best fitting @xmath0 model ( right column ) .
the model curve for @xmath54 and surface for @xmath55 have been computed by a monte carlo sampling of phase space .
[ fig : simc2],title="fig : " ] | n - body simulations of collisionless collapse have offered important clues to the construction of realistic stellar dynamical models of elliptical galaxies .
such simulations confirm and quantify the qualitative expectation that rapid collapse of a self - gravitating collisionless system , initially cool and significantly far from equilibrium , leads to incomplete relaxation , that is to a quasi - equilibrium configuration characterized by isotropic , quasi - maxwellian distribution of stellar orbits in the inner regions and by radially biased anisotropic pressure in the outer parts . in earlier studies , as illustrated in a number of papers several years ago ( see @xcite and references therein ) , the attention was largely focused on the successful comparison between the models ( constructed under the qualitative clues offered by the n - body simulations mentioned above ) and the observations . in this paper
we revisit the problem of incomplete violent relaxation , by making a direct comparison between the detailed properties of a family of distribution functions and those of the products of collisionless collapse found in n - body simulations .
address = dipartimento di fisica , universit di milano , via celoria 16 , i-20133 milano , italy address = scuola normale superiore , piazza dei cavalieri 7 , i-56126 pisa , italy |
the standard model of particle physics and the standard model of cosmology are both rife with numerical parameters that must have values fixed by hand to explain the observed world .
the world would be a radically different place if some of these constants took a different value .
in particular , it has been argued that if any one of six ( or perhaps a few more ) numbers did not have rather particular values , then life as we know it would not be possible @xcite : atoms would not exist , or no gravitationally bound structures would form in the universe , or some other calamity would occur that would appear to make the ( alter)-universe a very dull and lifeless place .
how , then , did we get so lucky as to be here ?
this question is an interesting one because all of the possible answers to it that i have encountered or devised entail very interesting conclusions .
an essentially exhaustive list of such answers is : 1 .
we just got very lucky : all of the numbers could have been very different , in which case the universe would have been barren
but they just happened by pure chance to take values in the tiny part of parameter space that would allow life .
we owe our existence to one very , very , very lucky roll of the dice .
we were nt particularly lucky : almost any set of parameters would have been fine , because life would find a way to arise in nearly any type of universe .
this is quite interesting because it implies ( at least theoretically ) the existence of life forms radically different from our own , existing for example in universes with no atoms or with no bound structure , or overrun with black holes , etc .
the universe was specifically designed for life .
the choice of constants only happened once , but their values were determined in some way by the need for us to arise
. this might be divine agency , or some radical form of wheeler s self - creating universe " , or super - advanced beings that travel back in time to set the constants at the beginning of the universe , etc .
however the reader feels about this possibility , they must admit that it would be interesting if true .
4 . we did not have to get lucky , because there are many universes with different sets of constants
i.e. , the dice were rolled many , many times .
we are necessarily in one of the universes that allows life , just as we necessarily reside on a planet that supports life , even when most others may not .
this is interesting because it means that there are other very different universes coexisting with ours in a multiverse " .
these four answers luck , _ elan vital _ , design , and multiverse will appeal at different levels to different readers .
but i think it is hard to argue that the multiverse is necessarily less reasonable than the alternatives . moreover , as is discussed at length elsewhere in this volume , there are quite independent reasons to believe , on the basis of inflation , quantum cosmology , and string / m theory , that there might quite naturally be many regions larger than our observable universe , governed by different sets of low - energy physics .
i am not aware of any independent scientific argument for the other three possible explanations . whether they are contemplated as an answer to the why are we lucky " question , or because they are forced upon us from other considerations ,
multiverses come at a high price .
even if we have in hand a physical theory and cosmological model that lead to a multiverse , how do we test it ? if there are many sets of constants , which ones do we compare to those we observe ?
in the next section of this chapter i will outline what i think a sound prediction in a multiverse would look like .
as will become clear , this requires many ingredients , and there are some quite serious difficulties in generating some of these ingredients , even with a full theory in hand . for this reason ,
many short - cuts have been devised to try to make predictions more easily . in the third section
i will describe a number of these , and show the cost that this convenience entails . finally , in section [ sec - coinc ]
i will focus on the interesting question of whether the anthropic approach to cosmology might lead to any _ general _ conclusions about how the study of cosmology will look in coming years .
imagine that we have a candidate physical theory and set of cosmological boundary conditions ( hereafter denoted @xmath0 ) that predicts an ensemble of physically realized systems , each of which is approximately homogeneous in some coordinates and can be characterized by a set of parameters ( i.e. the constants appearing in the standard models of particle physics and cosmology ; i assume here that the laws of physics themselves retain the same form ) .
let us denote each such system a `` universe '' and the ensemble a `` multiverse '' .
given that we can observe only one of these universes , what conclusions can we draw regarding the correctness of @xmath0 , and how ?
one possibility would be if there were a parameter for which _
none _ of the universes in the ensemble had the value we observe . in this case
@xmath0 would be ruled out .
( note that any @xmath0 in which at least one parameter has a range of values that it does not take in _ any _ universe is thus rigorously falsifiable , which is a nice thing for a theory to be ) . or perhaps some parameter takes only one value in all universes , and this value matches the observed one .
this would obviously be a significant accomplishment of the theory .
both possibilities are good as far as they go , and seem completely uncontroversial .
but they do not go far enough .
what if our observed parameter values appear in some but not all of the universes ?
could we still rule out the theory if those values are incredibly rare , or gain confidence if they are extremely common ?
i find it hard to see why not .
if some theory predicts outcome a of some experiment with @xmath1 probability , and outcome b with probability @xmath2 , i think we would be reluctant to accept the theory if a single experiment were performed and showed outcome b , _ even if we did not get to repeat the experiment_. in fact , it seems consistent with all normal scientific methodology to rule out the theory at @xmath3 confidence the problem is just that without repeating our measurements we will not be able to _ increase _ this confidence .
this seems to be exactly analogous to the multiverse _ if _ we can compute , given our @xmath0 , the _ probability _ that we should observe a given value for some observable .
can we compute this probability distribution in a multiverse ? perhaps .
i will argue that to do so in a sensible way , we would need seven successive ingredients . 1 . first , of course , we require a multiverse : an ensemble of regions , each of which would be considered a universe to observers inside it ( i.e. its properties would be uniform for as far as those observers could see ) , but each of which may have different properties .
next we need to isolate the set of parameters characterizing the different universes
. this might be the set of 20-odd free parameters in the standard model of particle physics ( see , e.g. , ref .
@xcite and references therein ) , plus a dozen or so cosmological parameters @xcite
. there might be additional parameters that become important in other universes , or differences ( such as different forms of the physical laws ) that can not be characterized by differences in a finite set of parameters .
but for simplicity let us assume that some set of @xmath4 numbers @xmath5 ( where @xmath6 ) fully specify each universe .
3 . given our parameters , we need some _ measure _ with which to calculate the multi - dimensional probability distribution @xmath7 for the parameters .
we might , for example ,
count each universe equally " to obtain the probability @xmath8 , defined to be the chance that a randomly chosen universe from the ensemble would have the parameter values @xmath5 .
, the probability that @xmath5 are all within the interval @xmath9 $ ] , where @xmath10 is a cumulative probability distribution . ] this can be a bit tricky , however , because it depends on how we delineate the universes : suppose that @xmath11 universes happen to be @xmath12 times larger than @xmath13 universes .
what would then prevent us from splitting " each @xmath11 universe into 10 , or 100 , or @xmath12 universes , thus radically changing the relative probability of @xmath11 vs. @xmath13 ?
these considerations might lead us to take a different measure such as volume , e.g. to define @xmath14 , the chance that a randomly point in space would reside in a universe with parameter values @xmath5 .
but in an expanding universe volume increases , so this would depend on the time at which we choose to evaluate the volume in each universe .
we might then consider some counting " object that endures , say a baryon ( which is relatively stable ) , and define @xmath15 , the chance that a randomly chosen baryon would reside in a universe with parameter values @xmath5 .
but now we have excluded from consideration universes with no baryons .
do we want to do that ?
this will be addressed in step ( v ) . for now , note only that it is not entirely clear , even in principle , which measure we should place over our multiverse .
we can call this the measure problem . " 4 .
once we choose a measure object @xmath16 , we still need to actually compute @xmath17 , and this may be far from easy .
for example , in computing @xmath18 , some universes may have infinite volume . in this case values of @xmath5 leading to universes with finite volume will have zero probability .
how , though , do we compare two infinite volumes ?
the difficulty can be seen by considering how we would count the fraction of red vs. blue marbles in an infinite box .
we could pick one red , then one blue , and find a 50 - 50 split .
but we could also repeatedly pick 1 red , then 2 blue , or 5 red , then 1 blue .
we could do this forever and so obtain any ratio we like .
what we would like to do is just
grab a bunch of marbles at random " and count the ratio .
but in the multiverse case it is not so clear how to perform this random ordering of marbles to pick .
this difficulty , which might be termed the ordering problem " @xcite , has been discussed a number of times in the context of eternal inflation @xcite and a number of plausible prescriptions have been proposed .
but there does not seem to be any generic solution , or convincing way to prove that one method is correct .
if we have managed to calculate @xmath17 , do we have a prediction ?
we have an answer to the question : given that i am ( or can associate myself with ) a randomly chosen @xmath16-object , which sort of universe am i in ? " but this is _ not _ necessarily the same as the more general question : what sort of universe am i in ? " first of all , different @xmath16-objects will generally give different probabilities , and they can not all be the answer to the same question .
second , we may not be all that closely associated with our @xmath16-object ( which was chosen mainly to provide _ some _ way to compute probabilities ) because it does not take into account important requirements for our existence .
for example , if @xmath16 were volume , i would be asking what i should observe given that i am at a random point in space ; but we are _ not _ at a random point in space ( which would on average have a density of @xmath19g / cc ) , but rather at one of the very rare points with density @xmath20g / cc .
the reason for this improbable situation is obviously anthropic " we just do not worry about it because we can observe many other regions at the proper density ( if we could not see such regions , we might be more reluctant to accept a cosmological model with such a low average density . ) finally , it might be argued that the question we have answered through our calculation is not nearly as specific a question as we could ask , because we know a lot more about the universe than that it contains volume , or baryons .
we might , instead , ask given that i am in a universe with the properties we have already observed , what should i observe in the future ? "
+ as discussed at length in @xcite , these different specific questions can be usefully thought of as arising from different choices of conditionalization .
the probabilities @xmath17 are conditioned on as little as possible , whereas the anthropic question of given that i am a randomly chosen _ observer _
, which should i measure " specifies probabilities conditioned on the existence of an observer " , while the approach of given what i know now , what will i see " specifies probabilities conditioned on being in a universe with all of the properties that we have already observed . these are three genuinely different approaches to making predictions in a multiverse that may be termed , respectively , bottom - up " , anthropic " , and top down " .
+ let us denote by @xmath21 the conditionalization object used to specify these conditional probabilities . in bottom - up reasoning
, it would be the same as the @xmath16-object ; in the anthropic approach it would be an observer " , and in the top - down approach it could be a universe with the currently - known properties of our universe .
it can be seen that they inhabit a spectrum , from the weakest conditionalization ( bottom - up ) to the most stringent ( top - down ) . like our initial @xmath16-object , choosing a conditionalization is unavoidable and important , and there is no obviously correct choice to make .
( see refs .
@xcite for similar conditionalization schemas . ) 6 .
having decided on a conditionalization object @xmath21 , the next step is to compute the number of @xmath22 of @xmath21-objects per @xmath16-object , for each set of values of the parameters @xmath5 .
for example , if we have chosen to condition on observers , but have used baryons to define our probabilities , then we need to calculate the number of observers per baryon as a function of cosmological parameters .
we can then calculate @xmath23 , i.e. the probability that a randomly chosen @xmath21-object ( observer ) resides in a universe with parameters @xmath5 .
there are a few possible pitfalls in doing this .
first , if @xmath24 is infinite , then the procedure clearly breaks because @xmath25 then becomes undefined .
this is why the @xmath16-object should be chosen to requires as little as possible for its existence ( and hence be associated with the minimal - conditionalization bottom - up approach ) .
this difficulty will generically occur if the existence of an @xmath21-object does not necessarily entail the existence of an @xmath16-object .
for example , if the @xmath16-object were a baryon but the @xmath21-object were a bit of volume , then @xmath4 would be infinite for @xmath5 corresponding to universes with no baryons .
the problem arises because baryons require volume to be in , but volume does not require a baryon to be in it .
this seems straightforward , but gets much murkier when we consider the _ second _ difficulty when calculating @xmath4 , which is that we may not be able to precisely define what an @xmath21-object is , or what it takes to make one .
if we say that the @xmath21-object is an observer , what exactly does that mean ? a human ? a carbon - based life form
? can observers exist without water ?
without heavy elements ?
without baryons ?
without volume ?
it seems quite hard to say .
we are forced , then , to choose some proxy for an observer , e.g. a galaxy , or a star with possible planets , etc .
but our probabilities will perforce depend on the chosen proxy and this must be kept in mind .
+ it is worth noting a small bit of good news here .
if we do manage to consistently compute @xmath26 for some measure object @xmath27 , then insofar as we want to condition our probabilities on @xmath21-objects , we have solved the measure problem : if we could consistently calculate @xmath28 for a different measure object @xmath29 , then we should obtain the same result for @xmath30 , i.e. @xmath31 .
thus our choice of @xmath27 ( rather than @xmath29 ) becomes unimportant .
the final step in making predictions is to make the assumption that the probability that we will measure some set of @xmath5 is given by the probability that a randomly chosen @xmath21-object will .
this assumption really entails two others : first , that we are some how directly associated with @xmath21-objects , and second that we have not , simply by bad luck , observed highly improbable values of the parameters .
the assumption that we are _ typical _ observers has been termed the principle of mediocrity " @xcite .
one may argue about this assumption , but _ some _ assumption is necessary if we are to connect our computed probabilities to observations , and it is difficult to see what alternative assumption would be more reasonable
. the result of all this work would be the probability @xmath32 that a randomly selected @xmath21-object ( out of all of the @xmath21-objects that exist in multiverse ) would reside in a universe governed by parameters @xmath5 , along with a reason to believe that this same probability distribution should govern what we will observe .
we can then make the observations ( or consider some already - made ones ) .
if the observations are highly improbable according to our predictions , we can rule out the candidate @xmath0 at some confidence that depends on how improbable our observations were .
apart from the manifest and grave difficulties involved in actually completing the seven listed steps in a convincing way , i think the only real criticism that can be leveled at this approach is that unless @xmath33 for our observed paramters , there will always be the chance that the @xmath0 was correct and we measured an unlikely result . usually , we can rid ourselves of this problem by repeating our experiments to make @xmath34 as small as we like ( at least in principle ) , while here we do not have that option once we have used up " the measurement of all of the paramters required to describe our universe ( which appears to be rather surprisingly few , at least according to current theories ) , we are done .
although the idea of a multiverse has been around for quite a while , no one has ever really come close to making the sort of calculation outlined in the previous section . instead ,
those wishing to make predictions in a multiverse context have made strong assumptions about which parameters @xmath5 actually vary across the ensemble , about the choice of @xmath21-object , and about the quantities @xmath17 and @xmath22 that go into predicting their probabilities for measurement .
some of these shortcuts aim simply to make a calculation tractable ; others are efforts to avoid anthropic considerations , or alternatively to use anthropic considerations to avoid other difficulties .
i would not have listed any ingredients that i thought could be omitted from a really sound calculation , thus all of these shortcuts are necessarily incomplete ( some , in my opinion , disastrously so ) . but by listing and discussing them , i hope to give the reader both a flavor for what sort of anthropic ( or anthropic - esque ) arguments have been made in the literature , and where they may potentially go astray .
this assumption underlies a sort of anthropic reasoning that has earned the anthropic principle a lot of ill will .
it goes something like : `` let s assume that lots of universes governed by lots of different parameter values exist .
then since only universes with parameter values almost exactly the same as ours allow life , we must be in one of those , and we should not find it strange if our parameter values seem special . '' in the conventions i have described , this is essentially equivalent setting @xmath21-objects to be observers , then hoping that the hospitality factor " @xmath35 is very narrowly peaked around one particular set of parameters . in this case , the _ a priori _ probabilities @xmath36 are pretty much irrelevant because the shape of @xmath37 will pick out just one set of parameters .
because our observed values @xmath38 definitely allow observers , the allowed set must then be very near @xmath38 .
three problems with this type of reasoning are as follows .
first , it is rather circular : it entails picking the @xmath21-object to be an observer , but then quickly substituting a universe just like ours " for the @xmath21-object , with the reasoning that such universes will definitely support life .
thus we have arrived at : the universe we observe should be pretty much like the observed universe .
the way to avoid this silliness is to allow at least the possibility that there are life - supporting universes with @xmath39 , i.e. to discard the unproven _ assumption _ that @xmath37 has a single , dominant , narrow peak .
the second problem is that if @xmath37 were really _ so _ narrowly peaked as to render @xmath36 irrelevant , then we would be in serious trouble as theorists , because we would lose any ability to distinguish between candidates for our fundamental theory : unless our observed universe is _ impossible _ in the theory , then the anthropic factor would force the predictions of the theory to match our observations . as discussed in the next section , this is not good .
the third problem with @xmath37 being an extremely peaked function is that it does not appear to be true !
as discussed in below , it appears that for any reasonable surrogate for observers ( e.g. galaxies like ours , or stars with heavy elements , etc . ) , calculations done using our current understanding of galaxy and structure formation indicate that the region of parameter space in which there can be many of those objects may be small compared to the full parameter space , but it is much larger than the region compatible with our observations .
as mentioned in section 2 , there is a ( relatively ! ) easy thing to do with a multiverse theory @xmath0 : work out which parameters combinations can not occur in _ any _ universe .
if the combination we actually observe is one of these , then the theory is ruled out .
this is unobjectionable , but a rather weak way to test a theory because given two theories that are _ not _ ruled out , we have no way whatsoever of judging one to be better , even if the parameter values we observe are in some sense generic in one and absurdly rare in the other . , this approach which makes _ no _ assumptions about @xmath24 is equivalent to the approach just described of making the very strong assumption that @xmath24 allows only one specific set of parameter values , because in either case a theory can only be ruled out if our observed values are impossible in that theory . ]
this is not how science usually works .
for example , suppose our theory is that a certain coin - tossing process is unbiased . if our only way to test this theory was to look for experimental outcomes that are impossible , then the theory would unfalsifiable : we would have to accept it theory for _ any _ coin we are confronted with , because _ no _ sequence of tosses would be impossible in it ! even if 10,000 tosses in a row
all came up heads , we would have no grounds for doubting our theory because while getting heads 10,000 times in a row on a fair coin is absurdly improbably , it is not impossible .
nor would we have reason to prefer the ( seemingly much better ) nearly every toss comes out heads " theory . clearly this is a situation we would like to improve on , as much in universes as in coin tosses .
one possible improvement would be to assume that we will observe a
typical " set of parameters in the ensemble , i.e. that we will employ bottom - up " reasoning as described in the first section , by using the _ a priori _ ( or prior " ) probabilities @xmath36 for some choice of measure - object such as universes , and just ignore the conditionalization factor @xmath24 .
there are two possible justifications for this .
first , we might simply want to avoid any sort of anthropic issues on principle .
second , we might hope that some parameter values are much , much more common than others , to the extent that the @xmath24-factor becomes irrelevant in other words that @xmath36 ( rather than @xmath24 ) is a very strongly peaked around some particular parameters .
the problem with this approach is the measure problem " discussed above : there is an implicit choice of basing probabilities on universes ( say ) rather than on ( say ) volume elements or baryons .
each of these measures has problems for example , it seems that probabilities based on universes " depends on how the universes are delineated , which can be ambiguous .
moreover there seems to be no reason to believe that predictions made using any two measures should agree particularly well .
for example , as discussed elsewhere in this volume , in the string theory
landcape " there are many possible parameter sets , depending on which metastable minimum one chooses in a potential that depends in turn on a number of fluxes that can take a large range of discrete values .
imagine that exponentially many more minima lead to @xmath11 than lead to @xmath13 .
should we expect to observe @xmath11 ?
not necessarily , because the relative number of @xmath40-universes vs. @xmath41-universes that _ actually come into existence _ may easily differ exponentially from the relative number of @xmath40-minima vs. @xmath41-minima .
( this seems likely to me in an eternal - inflation context , where the relative number universes could depend on exponentially - suppressed tunnelings between vacua . )
worse yet , these may in turn differ exponentially ( or even by an infinite factor ) from the relative numbers of baryons , or relative volumes . in short , while we are free to use bottom - up reasoning with any choice of measure object we like , we are not free to assert that other choices would give similar predictions , or that conditionalization can be rendered irrelevant .
so we had better have a pretty good reason for the choice we make .
another way in which one might hope to circumvent anthropic issues is to condition the probabilities on some or all observations that have already been made . in this top - down " ( or perhaps pragmatic " ) approach
we ask : given everything that has been observed so far , what will we observe in some future measurement ?
it has a certain appeal , as this is often what is done in experimental science : we do not try to _ predict _ what our laboratory will look like , just what will happen given that the lab is in a particular state at a given time . in the conventions of section 2 ,
the approach could consist of choosing the @xmath21-object to be universes with parameters agreeing with the measured values .
while appealing , this approach suffers some deficiencies : * it still does not completely avoid the measure problem , because even once we have limited our consideration to universes that match our current observations , we must still choose a measure with which to calculate the probabilities for the remaining ones . * through our conditioning , we may accept theories for which our parameter values are wildly improbable , without supplying any justification as to why we observe such improbable values .
this is rather strange .
imagine that i have a theory in which the cosmological constant @xmath42 is ( with very high probability ) much higher than we observe , and the dark matter particle mass @xmath43 is almost certainly @xmath44gev .
i condition on our observed @xmath42 , simply accepting that i am in an unusual universe .
now say i measure @xmath45gev .
i would like to say my theory is ruled out .
fine , but here is where it gets odd : according to top - down reasoning , i should also have already ruled it out if i had done my calculation in 1997 , before @xmath42 was measured . and
someone who invented the very same theory next week but had not been told that i have already ruled it out would _ not _ rule it out , but instead just take the low value of @xmath43 ( along with the observed @xmath42 ) as part of the conditionalization ! * if we condition on everything we have observed , we obviously give up the possibility of _ explaining _ anything we have observed ( which at this point is quite a lot in cosmology ) through our theory .
the last two issues motivate variations on the top - down approach in which only some current observations are conditioned on .
two of which i am aware are : 1 .
we might _ start _ by conditioning on all observations , then progressively condition on less and less and try to predict " the things we have decided not to condition on ( as well , of course , as any _ new _ observations)@xcite .
the more we can predict , the better our theory is . the problem is that either ( a ) we will get to the point where we are conditioning on as little as possible ( the bottom - up approach ) , and hence the whole conditionalization process will have been a waste of time , or ( b ) we will still have to condition on some things , and admit either that these have an anthropic explanation , or that we just choose to condition on them ( leading to the funny issues discussed above ) .
2 . we might choose at the outset to condition on things that we think may be fixed anthropically ( without trying to actually generate this explanation ) , then try to predict the others @xcite .
this is nice in being relatively easy , and in providing a justification for the conditionalization .
it suffers from the problems of ( a ) guessing which parameters are anthropically important and which are not , ( b ) even if a parameter is anthropically unimportant , it may be strongly correlated in @xmath36 with one that is , and ( c ) we still have to face the measure problem , which we can not avoid by counting conditioning on observers , because we are avoiding anthropic considerations .
most of the shortcuts " discussed so far have been attempts to avoid anthropic considerations .
but we may , instead , consider how me might try to formulate an anthropic prediction ( or explanation ) for some observable , without going through the full calculation outlined in section 2 .
the way of doing this that has been employed in the literature ( largely in the efforts of vilenkin and collaborators ) is as follows .
first , one fixes all but one ( or perhaps two ) of the parameters to the observed values .
this is done for tractability and/or because one hopes that they will have non - anthropic explanations .
let us call the parameter that is allowed to vary across the ensemble @xmath46 .
next , an @xmath21-object is chosen such that _ given that only @xmath46 varies _ , it is hoped that ( a ) the number @xmath24 of these objects ( per baryon , or per comoving volume element ) in a given universe is calculable , and ( b ) this number is arguably proportional to the number of observers .
for example , if only @xmath42 varies across the ensemble , galaxies might make reasonable @xmath21-objects because a moderately different @xmath42 will probably not change the number of observers per galaxy , but _ will _ change the number of galaxies in a way that can be computed using fairly well - understood theories of galaxy and structure formation ( for examples see @xcite ) .
third , it is assumed that @xmath47 is either flat or a simple power - law , without any complicated structure .
this can be done just for simplicity , but it is often argued to be natural @xcite . the flavor of this argument is as follows . if @xmath36 is to have interesting structure over the relatively small range in which observers are abundant , there must be a parameter of order the observed @xmath46 in the expression for @xmath36 .
but precisely this absence is what motivated the anthropic approach .
for example , if the expression for @xmath48 contained the energy scale @xmath49ev corresponding to the observed @xmath42 , the origin of that energy scale would probably be more interesting than our anthropic argument , as it would provide the basis for a ( non - anthropic ) solution to the cosmological constant problem ! under these ( fairly strong ) assumptions we can then actually calculate @xmath50 and see whether or not the observed value is reasonably probable given this predicted distribution . for example
when @xmath42 alone is varied , a randomly chosen galaxy is predicted to lie in a universe with @xmath42 comparable to ( but somewhat larger than ) the value we see @xcite . with @xmath51 . for @xmath52 ,
higher values would be predicted , and with @xmath53 lower values would be favored . ]
i actually think this sort of reasoning is pretty respectable , _ given the assumptions made_. in particular , the anthropic argument in which _ only _
@xmath42 varies is a relatively clean one .
but there are a number of pitfalls when it is applied to parameters other than @xmath42 , or when one allows multiple parameters to vary simultaneously . *
assuming that the abundance of observers is strictly proportional to that of galaxies only makes sense if the number of galaxies and not their properties changes as @xmath46 varies .
however , changing nearly any cosmological parameter will change the properties of typical galaxies .
for example , increasing @xmath42 will decrease galaxy numbers , but also make galaxies smaller on average , because a high @xmath42 squelches structure formation at late times when massive galaxies form .
increasing the amplitude of primordial perturbations would similarly lead to smaller , denser but more numerous
galaxies , as would increasing ratio of dark matter to baryons . in these cases
, we must specify in more detail what properties an observer - supporting galaxy should have , and this is very difficult to do without falling into the circular - argument trap of assuming that only galaxies like ours support life .
finally , this sort of strategy seems unlikely to work if we try to change _
non_-cosmological parameters , as this could lead to radically different physics and the necessity of thinking _ very _ hard about what sort of observers there might be .
* the predicted probability distribution clearly depends on @xmath36 , and the assumption that @xmath36 is flat , or a simple power law , can break down .
this can happen even for @xmath42 @xcite but perhaps more naturally for other parameters such as the dark matter density for which particle physics models can already yield sensible values .
moreover , this breakdown is much more probable if ( as discussed below and contrary to the assumption made above ) the hospitality factor @xmath54 is significant over many orders of magnitude in @xmath46 .
* calculations of the hospitality factor @xmath54 can go awry if @xmath46 is changed more than a little .
for example , a neutrino mass slightly larger than we observe would suppress galaxy formation by erasing small - scale structure .
but neutrinos with a large ( @xmath55ev ) mass would act as dark matter and lead to strong halo formation .
whether these galaxies would be hospitable is questionable ( they would be very baryon - poor ) , but the point is that the physics becomes _ qualitatively _ different . as another example , a lower photon / baryon ratio @xmath56 would lead to earlier - forming , denser galaxies . but a _ much _ smaller value would lead to qualitatively different structure formation , as well as the primordial generation of heavy elements @xcite . as discussed at length in ref .
@xcite , these changes are very dangerous because over orders of magnitude in @xmath46 , @xmath47 will tend to change by many orders of magnitude .
thus even if these alter - universes only have a few observers in them , they may dominate @xmath25 and hence qualitatively change the predictions .
* along the same lines , but perhaps even more pernicious , when multiple parameters are varied simultaneously , the effects of some variations can offset the effect of others so that universes quite different from ours can support many of our chosen @xmath21-objects . for example , increasing @xmath42 cuts off galaxy formation at a given cosmic density , but raising the perturbation amplitude @xmath57 causes galaxies to form earlier ( thus nullifying the effect of @xmath42 ) .
this can be seen in the calculations of @xcite , and is discussed explicitly in @xcite .
many such deneneracies exist , because rasing @xmath42 , @xmath56 , or the neutrino mass all decrease the efficiency of structure formation , while raising @xmath58 or @xmath57 increase the efficiency . as an extreme case
, it was shown in @xcite that if @xmath57 and @xmath56 are allowed to vary with @xmath42 , then universes with @xmath42 of @xmath59 times our observed value could arguably support observers ! including more cosmological parameters , or non - cosmological parameters , can only make this problem worse .
these problems indicate that while anthropic arguments concerning @xmath42 in the literature are relatively clean " , it is unclear whether other parameters ( taken individually ) will work as nicely .
more importantly , a number of issues arise when several parameters are allowed to vary at once , and there does not seem to be any reason to believe that success in explaining one parameter anthropically will persist when additional parameters are allowed to vary . in some cases , it may : for example , allowing neutrino masses to vary in addition to @xmath42 does not appear to spoil the anthropic explanation of a small but nonzero cosmological constant @xcite .
on the other hand , allowing @xmath57 to vary does , unless @xmath60 is strongly peaked at small values of @xmath57 @xcite .
i suspect that allowing @xmath61 or @xmath56 to vary along with @xmath42 would have a similar effect .
for those serious about making predictions in a multiverse , i would propose that rather than working to generate additional incomplete anthropic arguments by taking shortcuts , a much better job must be done in each of the individual ingredients .
for one example , our understanding of galaxy formation is sufficiently strong that the multi - dimensional hospitality factor @xmath22 could probably be computed for @xmath5 within a few orders of magnitude of the observed values , for @xmath62objects of galaxies with properties within some range .
second , despite some nice previous work , i think the problem of how to compute @xmath36 in eternal inflation is a pretty open one . finally , the string / m theory landscape ( which is generating a lot of interest in the present topic right now ) can not hope to say much of anything about @xmath36 until its place in cosmology is understood in particular , we need both a better understanding of the statistical distribution of field values that result from evolution in a given potential , and also an understanding of how transitions between vacua with different flux values occur , and exactly what is transitioning .
the preceding sections should have suggested to the reader that it will be a huge project to compute a sound prediction of cosmological and physical parameters from a multiverse theory in which they vary .
it may be so hard that it will be a very long time before any such calculation is at all believable .
it is worth asking then : is there any way nature might give us an indication as to whether the anthropic approach is a sensible one , i.e. does the anthropic approach make any sort of _ general _ predictions even without the full calculation of @xmath25 ?
interestingly , i think the answer might be yes : i am aware of two such general ( though somewhat vague ) predictions of the anthropic approach . to understand the first , assume that only one parameter , @xmath46 , varies , and consider @xmath63 , the probability distribution in @xmath64 , given by some theory @xmath0 .
for @xmath64 near the observed value @xmath65 , @xmath66 can basically only be doing one of three things : it can rise with @xmath67 , fall with @xmath67 , or be approximately constant . in the first two cases ,
the theory @xmath0 would predict that we should see a value of @xmath46 that is , respectively , higher or lower than we actually do _ if _ no anthropic conditionalization @xmath68 is applied .
now suppose we somehow compute @xmath69 and find that it falls off quickly for values of @xmath46 much smaller or larger than we observe , i.e. that only a range @xmath70 is anthropically acceptable " .
then we have an anthropic argument explaining @xmath71 , because this falloff means that @xmath72 will only be significant near @xmath71 . but now note that _ within _ the anthropically acceptable range , @xmath73 will be peaked near @xmath74 if @xmath66 is increasing with @xmath46 , or near @xmath75 if @xmath66 is decreasing with @xmath46 .
that is , we should expect @xmath71 at one edge of the anthropically acceptable range .
this idea has been called the principle of living dangerously " @xcite .
it asserts that for a parameter that is anthropically determined , we should expect that a calculation of @xmath37 would reveal that observers would be strongly suppressed either for @xmath46 slightly larger or slightly smaller than @xmath76 , depending on whether @xmath66 is rising or falling .
now , this is not a very specific prediction : exactly where we would expect @xmath71 to lie depends both on how steep @xmath77 is , and how sharp the cutoff in @xmath37 is for @xmath46 outside of the anthropically acceptable range . and
it would not apply to anthropically - determined parameters in all possible cases .
( for example , if @xmath66 were flat near @xmath71 , but also very high at @xmath78 , anthropic effects would be required to explain why we do not observe the very high value ; but any region within the anthropically acceptable range would be equally probable , so we would not expect to be , so to speak , living on the edge . ) despite these caveats , this is a prediction of sorts , because the naive expectation would probably be for our observation to place us somewhere in the interior of the region of parameter space that is hospitable to life , rather than at the edge .
a second sort of general prediction of anthropic reasoning is connected to what might be called cosmic coincidences . " for example , many cosmologists have asked themselves ( and each other ) why the current density in vacuum energy , dark matter , baryons , and neutrinos are all within a couple of orders of magnitude of each other making the universe a much more complicated place than it might be .
conventionally , it has been assumed that these coincidences are just that , and follow directly from fundamental physics that we do not understand .
but if the anthropic approach to cosmology is really correct ( that is , if it is the real answer to the question of why these densities take the particular values they do ) , then the explanation is quite different : the densities are bound together by the necessity of observers existence , because only certain combinations will do . more explicitly ,
suppose several cosmological parameters are governed by completely unrelated physics , so that their individual prior probabilities @xmath36 simply multiply to yield the multidimensional probability distribution .
for example , we might have @xmath79 .
but even if @xmath34 factors , the hospitality factor @xmath37 will almost certainly not : if galaxies are @xmath21-objects , the number of galaxies formed at a given @xmath42 will depend on both other parameters , and only certain combinations will give a significant number of observers .
thus @xmath80 will likewise have correlations between the different parameters that lead to only particular combinations ( for example those with @xmath81 for a given @xmath57 and @xmath42 ) having high probabilities .
the cosmic coincidences would be explained in this way .
this anthropic explanation of coincidences , however , should not only apply to things that we have already observed .
if it is correct , then it should apply also to _ future _ observations ; that is , we should expect to uncover yet more bizarre coincidences between quantities that seem to follow from quite unrelated physics .
how might this actually happen ?
consider dark matter .
we know fairly precisely how much dark matter there is in the universe , and what its basic properties are .
but we have no real idea what it actually is , and there are many , many possible candidates that have been proposed in the literature .
in fact , we have no _ observational _ reason to believe that dark matter is one substance at all : in principle it could be equal parts axions , supersymmetric particles , and primordial black holes .
the reason most cosmologists do not expect this is that it would be a strange coincidence if three substances involving quite independent physics all wound up with essentially the same density in our universe .
but of course this would be just like the suprising - but - true coincidences that hold in already - observed cosmology . in the anthropic approach
, these comparable densities could be quite natural @xcite .
to see why , imagine that there are two _ completely independent _ types of dark matter permeating the ensemble : in each universe , they have some particular densities @xmath82 and @xmath83 out of a wide range of possibilities , so that the densities in a randomly chosen universe ( or around a randomly chosen baryon , etc .
) will be given probabilistically by @xmath84 .
under these assumptions there is no reason to expect that we should observe @xmath85 based just on these _ a priori _ probabilities .
now suppose , though , that @xmath24 picks out a particular narrow range of _ total _ dark matter density as anthropically acceptable .
that is , @xmath86 is narrowly peaked about some @xmath87 . in this case
, the peak of the probability distribution @xmath88 , which indicates what values a randomly chosen observer should see , will occur where @xmath84 is maximized _ subject to the condition _ that @xmath89 .
for simplicity let both prior probabilities be power laws : @xmath90 and @xmath91 .
now the coincidence : it is not hard to show that if @xmath92 and @xmath93 , then the maximum will occur when @xmath94 .
that is , the two components are likely to have similar densities unless the _ power law indices _ of their probability distributions differ by orders of magnitude .
being peaked where @xmath24 is declining , i.e. we should be living _ outside _ the anthropically comfortable range , not just dangerously but downright recklessly .
] of course , there are many ways in which this coincidence could fail to occur ( e.g. negative power - law indices , or correlated probabilities ) , but the point is that there is a quite natural set of circumstances in which the components are coincident , even though the _ fundamental _ physics is completely unrelated .
the preceding sections should have convinced the reader that there are good reasons for scientists to be very worried if we live in a multiverse : in order to test a multiverse theory in a sound manner , we must perform a fiendishly difficult calculation of @xmath32 , the probability that an @xmath21-object will reside in a universe characterized by parameters @xmath5 . and
because of the shortcomings of the shortcuts one may ( and presently must ) take in doing this , almost any particular multiverse prediction is going to be easy to criticize ; only a quite good calculation is going to be at all convincing .
much worse , we face an unavoidable and important choice in what @xmath21 should be : a possible universe , or an existing universe , or a universe matching current observations , or a bit of volume , or a baryon , or a galaxy , or an `` observer '' , etc .
i find it disturbingly plausible that observers " really are the correct conditionalization object , that their use as such is the correct answer to the measure problem , and that anthropic effects are the real explanation for the values of some parameters ( just as for the local density that we observe ) .
many cosmologiest appear to believe that taking the necessity of observers into account is shoddy thinking , and is employed only because it is the _ easy _ way out of solving problems the right " way .
but the arguments of this paper suggest that the truth may well be exactly the opposite : the anthropic approach may be the right thing to do in principle , but nearly impossible in practice .
nonetheless , we can not do away with multiverses just by wishing them away : we may in fact live in one , whatever the inconvenience to cosmologists .
the productive strategy then seems to be one of accepting multiverses as a possibility , and working toward understanding how to calculate the various ingredients necessary to make predictions in one .
whether really performing such a calculation will turn out to be possible , but it is certainly impossible if not attempted .
even if we can not calculate @xmath25 in the foreseeable future , however , cosmology in a multiverse may not be completely devoid of predictive power .
for example , of anthropic effects are at work , they should leave certain clues .
first , if we could determine the region of parameter space hospitable to observers , we should find that we are living in the outskirts of the livable region , rather than somewere in its midst .
second , if the anthropic effects are the explanation of the parameter values and coincidences between them that we see , then it ought to predict that new coincidences will be observed in future observations .
if in the next several decades dark matter is resolved into several equally important components , dark energy is found to be three independent substances , and several other `` cosmic coincidences '' are observed , even someone the most die - hard skeptics might accede that the anthropic approach may have validity
why else would be universe be so very baroque ? on the other hand ,
if we are essentially finished in defining the basic cosmological constituents , and the defining parameters are in the midst of a relatively large region of parameter space that might arguably support observers , then i think the anthropic approach would lose almost all appeal it has ; we would be forced to ask : why is nt the universe much wierder ?
99 m. rees , just six numbers .
( new york : basic books , 2000 ) .
m. tegmark , arxiv : astro - ph/0410281 .
a. d. linde and a. mezhlumian , phys .
d * 53 * , 4267 ( 1996 ) [ arxiv : gr - qc/9511058 ] .
v. vanchurin , a. vilenkin and s. winitzki , phys .
d * 61 * , 083507 ( 2000 ) [ arxiv : gr - qc/9905097 ] .
a. h. guth , phys .
rept . * 333 * , 555 ( 2000 ) [ arxiv : astro - ph/0002156 ] .
j. garriga and a. vilenkin , phys .
d * 64 * , 023507 ( 2001 ) [ arxiv : gr - qc/0102090 ] .
a. aguirre and m. tegmark , jcap , in press ; arxiv : hep - th/0409072 . n. bostrum , anthropic bias : observation selection effects in science and philosophy ( new york : routledge , 2002 ) . s. w. hawking and t. hertog , phys
. rev .
d * 66 * , 123509 ( 2002 ) [ arxiv : hep - th/0204212 ] .
a. albrecht , arxiv : astro - ph/0210527 .
m. dine , arxiv : hep - th/0410201 .
j. garriga , m. livio and a. vilenkin , phys .
d * 61 * , 023503 ( 2000 ) [ arxiv : astro - ph/9906210 ] . j. garriga and a. vilenkin , phys .
d * 61 * , 083502 ( 2000 ) [ arxiv : astro - ph/9908115 ] . j. garriga and a. vilenkin , phys .
d * 64 * , 023517 ( 2001 ) [ arxiv : hep - th/0011262 ] .
m. tegmark .
a. vilenkin , and l. pogosian , arxiv : astro - ph/0304536 .
l. pogosian , a. vilenkin and m. tegmark , arxiv : astro - ph/0404497 .
m. l. graesser , s. d. h. hsu , a. jenkins and m. b. wise , arxiv : hep - th/0407174 .
a. vilenkin , arxiv : gr - qc/9512031 .
s. weinberg , phys .
d * 61 * , 103505 ( 2000 ) l. smolin , arxiv : hep - th/0407213 . m. tegmark and m. j. rees , astrophys .
j. * 499 * , 526 ( 1998 ) s. dimopoulos and s. thomas , phys .
b * 573 * , 13 ( 2003 ) [ arxiv : hep - th/0307004 ] . | the notion that there are many `` universes '' with different properties is one answer to the question of `` why is the universe so hospitable to life ? ''
this notion also naturally follows from current ideas in eternal inflation and string / m theory .
but how do we test such a `` multiverse '' theory : which of the many universes do we compare to ours ?
this paper enumerates would would seem to be essential ingredients for making testable predictions , outlines different strategies one might take within this framework , then discusses some of the difficulties and dangers inherent in these approaches .
finally , i address the issue of whether there may be some _ general , qualitative _ predictions that multiverse theories might share . |
the advent of trigonometric parallaxes for @xmath510@xmath6 stars from the _ gaia _ mission promises to revolutionize many areas of stellar and galactic astrophysics , including exoplanet science .
for example , with eventual expected precision in the parallax of @xmath720 @xmath8as for bright exoplanet host stars , it should be possible to determine the stellar and planet radii and masses directly and empirically with accuracies of 35% ( see , e.g. , * ? ? ? * ) .
already , the _ gaia _ first data release ( dr1 ; * ? ? ? * ) provides parallaxes for @xmath52 million _
tycho-2 _ stars ( the tgas stars ) with a nominal precision of @xmath70.3 mas and with a quoted systematic uncertainty at present of 0.3 mas .
the results from dr1 are based on only 14 months of observations , and use external information in the form of earlier positions from the _ hipparcos _ @xcite and _ tycho-2 _ @xcite catalogs to help remove degeneracies ( the _ tycho - gaia _ astrometric solution ; * ? ? ?
* ) . additionally , they rely on very provisional and as yet incomplete calibrations , and as a result the astrometric products including the parallaxes are still preliminary .
nonetheless , the new parallaxes represent such an improvement in both quality and quantity that they are certain to be used by the community for a wide range of astrophysical applications , at least until future _ gaia _ releases supersede them .
it is essential , therefore , to assess the on - sky delivered performance of these parallaxes from _ gaia _ dr1 , especially the presence of any unexpected biases .
this is particularly important in light of the experience from _ hipparcos _ , which suffered a significant bias in at least the case of the pleiades cluster ( e.g. , * ? ? ?
such a check requires a set of benchmark stars whose parallaxes are determined independently and with an accuracy that is at least as good as that expected from _
gaia _ dr1 .
@xcite assembled a sample of 158 eclipsing binary stars ( ebs ) whose radii and effective temperatures are known empirically and precisely , such that their bolometric luminosities are determined to high precision ( via the stefan - boltzmann relation ) and therefore independent of assumed distance .
@xcite reported new , accurate measurements of the bolometric fluxes for these ebs which , together with the precisely known bolometric luminosities , yields a highly precise distance ( or parallax ) .
the precision of the parallaxes for this eb sample was predicted by @xcite to be @xmath7190 @xmath8as on average .
this is a factor of @xmath51.5 better than the median precision of 320 @xmath8as for _ gaia _
dr1 @xcite .
it is even somewhat superior to the expected _ gaia _ dr1 precision floor of 240 @xmath8as .
these eb parallaxes can therefore readily serve as distance benchmarks for the trigonometric parallaxes reported by _
dr1 , and in particular can be used to assess the presence of any systematics . in this letter
, we report the results of testing the _ gaia _ dr1 parallaxes against the @xcite eb benchmark sample .
section [ sec : data ] describes the eb and _ gaia _ data used .
section [ sec : results ] presents the key result of a systematic offset in the _ gaia _ parallaxes relative to the eb sample .
section [ sec : disc ] considers potential trends in the parallax offset with other parameters .
section [ sec : summary ] concludes with a summary of our conclusions .
we adopted the predicted parallaxes for the 158 ebs included in the study of @xcite .
of these , 116 had parallaxes available in the _ gaia _ first data release .
we excluded from our analysis any ebs identified as potentially problematic in @xcite .
this left 111 ebs with good parallaxes from both the eb analysis and from _
gaia_. these ebs are all relatively nearby , with parallaxes in the range @xmath9 0.330 mas .
the ebs and their relevant data are provided in table [ tab : data ] .
llccrrrrrrrrrrr uv psc & 0026 - 0577 - 1 & 5780 & 100 & 9.01 & 2.72 & 12.47 & 0.57 & 0.53 & 14.392 & 0.407 & 19.2297 & 6.8117 + xy cet & 0051 - 0832 - 1 & 7870 & 115 & 8.75 & 1.6 & 3.62 & 0.14 & 0.13 & 4.542 & 0.891 & 44.8897 & 3.5176 + v1130 tau & 0066 - 1108 - 1 & 6625 & 70 & 6.66 & 1.34 & 14.35 & 0.36 & 0.37 & 14.329 & 0.332 & 57.6748 & 1.5639 + ew ori & 0104 - 1206 - 1 & 5970 & 100 & 9.78 & 0.71 & 5.9 & 0.22 & 0.21 & 5.482 & 0.233 & 80.0381 & 2.0444 + u oph & 0400 - 1862 - 1 & 16440 & 250 & 5.9 & 1.15 & 4.35 & 0.14 & 0.17 & 3.685 & 0.775 & 259.1322 & 1.2105 +
figure [ fig : pxvspx ] shows the direct comparison of the eb parallax predictions from @xcite versus the _ gaia _ dr1 parallaxes for the study sample .
the least - squares linear best fit , weighted by the measurement uncertainties in both quantities @xcite , is @xmath10 .
while this indicates a good 1-to-1 agreement to first order , the coefficient of [email protected] could be interpreted as a modest global difference of _ scale _ in the _ gaia _ parallaxes relative to the eb parallaxes . however , considering all of the available evidence instead suggests a small _ offset _ in the _ gaia _ parallaxes as we now discuss .
figure [ fig : delta_plx_hist ] presents the overall distribution of parallax differences in the sense of @xmath11 .
the distribution appears roughly symmetric and normally distributed , with perhaps a sharper peak and more extended wings than a gaussian , and there is a clear offset relative to zero .
the mean offset is @xmath12 mas , where the quoted error is the uncertainty of the mean for 111 measurements .
( _ gaia_@xmath0eb ) .
ebs with known tertiary companions are represented in blue .
we find an offset of @xmath12 mas for the entire sample and @xmath13 mas when the triple systems are excluded .
also shown is a best - fit gaussian with @xmath14 , representing the quadrature sum of the typical random uncertainties from the _ gaia _ and eb parallaxes ; see sec .
[ sec : intro ] . ]
@xcite noted that a number of the ebs used in that study are known triple or quadruple systems .
in general these companions contribute very little to the total system light , and @xcite found no evidence for significant systematics in their predicted parallaxes .
nonetheless , the offset that we find above for the _ gaia _ parallaxes is small and could potentially have gone unnoticed in @xcite .
indeed , the effect of additional light contribution to the ebs by companions would be in the sense of making the eb stars appear brighter , therefore inferred to be closer , and in turn the _ gaia _ distances interpreted as too long ( parallax too small ) . in _ gaia _ dr1 , 12 of our ebs have known companions ( see table [ tab : data ] ; one additional eb with a known companion is already excluded by the cuts discussed in sec .
[ sec : data ] ) .
these ebs are indicated in blue in fig .
[ fig : delta_plx_hist ] , which shows the two largest outliers to be triples .
excluding all of the triples results in a parallax offset of @xmath13 mas , consistent with that determined for the full sample , though slightly smaller .
overall , from the eb sample , a systematic offset in the _ gaia _ dr1 parallaxes of @xmath15 to @xmath16 mas is indicated . for simplicity ,
we adopt the rounded value between these estimates of @xmath17 mas .
the official _ gaia _ dr1 documentation states : there are colour dependent and spatially correlated systematic errors at the level of @xmath10.2 mas . over large spatial scales ,
the parallax zero - point variations reach an amplitude of @xmath10.3 mas .... furthermore , averaging parallaxes over small regions of the sky will not reduce the uncertainty on the mean below the 0.3 mas level . "
our finding of a mean parallax offset of @xmath18 mas ( sec .
[ sec : results ] ; fig .
[ fig : delta_plx_hist ] ) corroborates this statement , and further quantifies it using an independent benchmark sample of ebs with accurately known distances @xcite . in principle
this offset could be due to systematics in one or more of the eb parameters from which the eb distances are determined .
if so , one might especially suspect the eb @xmath19 values : unlike the stellar radii , for example , which are determined from simple geometry , the @xmath19 values are determined from spectral analysis and/or spectral typing and/or color relations . the slope of the fitted relation in fig . [
fig : pxvspx ] would imply an error in the eb distance scale of @xmath53% , which in turn would require a systematic error in @xmath19 of @xmath51.5% ( because @xmath20 ) or @xmath5105 k given the typical @xmath19 of the eb sample .
the sense of the offset is that the ebs would have to be systematically too cool .
however , we do not consider this to be a likely possibility , for multiple reasons .
first , @xcite found no evidence for a systematic offset of the eb parallaxes relative to the _ hipparcos _ parallaxes which , even at the somewhat poorer precision of @xmath51 mas , should have been apparent .
@xcite compare the _ gaia _ dr1 parallaxes against @xmath587,000 _ hipparcos _ stars in common , finding a statistically significant average offset just under @xmath21 mas , smaller than , but in the same sense as the offset we find among our eb sample .
second , while systematics among various @xmath19 scales can be of order 100 k ( see , e.g. , * ? ? ?
* ; * ? ? ?
* ) , it is unlikely that they should produce a net offset of this entire magnitude in a sample of 111 ebs spanning a large range of @xmath19 , given the different methodologies and calibrations adopted by the various authors .
finally , we have directly examined the degree to which @xmath22 might correlate with @xmath19 in the eb sample ( fig .
[ fig : delta_vs_eb_plx ] , upper right ) , finding very weak evidence for a correlation : the regression relation has a coefficient of determination @xmath23 ; this parameter explains only 13% of the variance in @xmath22 .
indeed , a kendall s @xmath24 non - parametric correlation test gives a probability of 51% that @xmath22 and @xmath19 are uncorrelated .
( we checked that the parallax _ ratio _ versus @xmath19 is also not significantly correlated . ) incidentally , this also suggests little to no dependence of @xmath22 with color , since @xmath19 can be taken as a proxy for color . and @xmath25 , eb @xmath26 , and @xmath27 mag .
best fitting linear regression lines are shown for each ( see the text ) .
, title="fig : " ] and @xmath25 , eb @xmath26 , and @xmath27 mag .
best fitting linear regression lines are shown for each ( see the text ) .
, title="fig : " ] and @xmath25 , eb @xmath26 , and @xmath27 mag .
best fitting linear regression lines are shown for each ( see the text ) .
, title="fig : " ] with a current sample of 111 ebs that overlap with the _ gaia _ dr1 parallaxes , it is difficult to ascertain with certainty whether any higher - order dependencies are at work beyond a simple offset . nonetheless , it is possible that the parallax offset we observe is a function of one or more parameters .
we have considered some obvious possible parameters , including distance , and brightness .
these are represented in figure [ fig : delta_vs_eb_plx ] .
there is some evidence for a dependence of @xmath22 on distance and/or brightness , however it is difficult to gauge whether these modest distance and brightness dependencies are independent of one another . a priori ,
a possible dependence on brightness may be more likely than a dependence on distance : because of the manner in which the trigonometric parallax measurements are made ( linear offsets on the detector ) , they are more likely to depend on parameters that affect the displacement calibration ( e.g. , color or brightness ) than the displacement itself . thus there is not strong evidence for a dependence of the offset with brightness or color , at least in our sample .
the apparent correlation of @xmath22 with @xmath28 may be a consequence of the tendency for the ebs with the largest @xmath28 to be located in the ecliptic north , where we find the largest overall offset ( see below ) . in any event , among our 12 ebs with @xmath29 mas , the average offset is @xmath30 mas .
thus , it appears that the offset vanishes for very small parallaxes , @xmath31 mas .
this would be consistent with the findings of @xcite and also @xcite , who find good agreement with the _ gaia _ dr1 parallaxes in separate samples of very distant cepheids with estimates based on period - luminosity relations .
@xcite also find evidence for a parallax offset , in the same sense as we find , among the small number of very nearby , bright cepheids in their sample .
finally , we have considered the possible spatial dependence of @xmath22 .
there is evidence for a trend or difference by ecliptic latitude ( fig .
[ fig : hist_eclip ] ) .
the mean parallax offset for ebs in the northern ecliptic hemisphere is statistically significant with @xmath32 mas , whereas in the southern ecliptic hemisphere it is not significant with @xmath33 mas .
a two - sided kolmogorov - smirnov test gives a probability of 0.0001 that this difference could occur by chance .
alternatively , the trend can be represented as a linear variation with latitude , @xmath34 mas ( @xmath3 in degrees ) .
a kendall s @xmath24 test indicates that the correlation between parallax offset and ecliptic latitude is significant with 99.7% confidence .
mas ) but not in the ecliptic south ( @xmath33 mas ) .
color represents eb brightness and symbol size is proportional to @xmath28 . ]
the possibility that systematics in the _ gaia _ dr1 parallaxes are dependent on ecliptic latitude was suggested in the _ gaia _ documentation @xcite .
these authors reported that a comparison with parallaxes from the _ hipparcos _ mission indicates an overall statistically significant offset @xmath35 mas , the _ gaia _ values being smaller , but that the northern ecliptic hemisphere shows a larger offset of @xmath36 mas compared to the southern hemisphere ( @xmath37 mas ) .
here we present evidence of a small but systematic offset in the average zero - point of the parallax measurements recently released by the _
gaia _ mission of about @xmath38 mas , in the sense that the _ gaia _ values are too small
. we also find evidence to suggest that the offset is a function of ecliptic latitude .
the offset in the northern ecliptic hemisphere is @[email protected] mas and @[email protected] mas in the southern ecliptic hemisphere . alternatively
, the offset may also be represented as a linear function of the ecliptic latitude , @xmath3 ( @xmath39 ) , according to @xmath2 mas .
to apply the correction , this ( negative ) offset must be _ subtracted _ from the reported _ gaia _ dr1 parallaxes . at present
we can only confirm that the offset is statistically valid for relatively large parallaxes , @xmath40 mas .
the reference for this determination is a set of more than 100 independently inferred parallaxes from a benchmark sample of well - studied eclipsing binaries with a wide range of brightnesses and distributed over the entire sky .
this paper presents evidence of a _ difference _ between the _ gaia _ and eb parallaxes , which we have interpreted here as a systematic error in _ gaia _ after discussing the alternative .
in particular , we have considered the possibility of a systematic offset in the eb effective temperature scale as a possible , but unlikely alternative explanation .
it is expected that future releases of the _ gaia _ catalog will remove this small shift as the number of observations increases , calibrations are improved , and the astrometric solution transitions to a self - consistent global fit using only _ gaia _ data , independent of external astrometric information . indeed , these final trigonometric parallaxes may then be used to further refine the eb sample itself , such as improvements to the eb effective temperature scale . in the meantime , investigators using the parallax results from _ gaia _
dr1 are encouraged to keep the systematic error reported here in mind .
this work has made use of the filtergraph data visualization service at filtergraph.vanderbilt.edu @xcite .
k.g.s . acknowledges partial support from nsf paare grant ast-1358862 .
g.t . acknowledges partial support for this work from nsf grant ast-1509375 .
the authors are grateful to s. casertano and a. riess for sharing their results in advance of publication .
we are grateful to the referee for critiques and suggestions that improved the manuscript .
this work has made use of data from the european space agency ( esa ) mission _ gaia _
( http://www.cosmos.esa.int/gaia ) , processed by the _
data processing and analysis consortium ( dpac , http://www.cosmos.esa.int/web/gaia/dpac/consortium ) . funding for the dpac has been provided by national institutions , in particular the institutions participating in the _ gaia _ multilateral agreement . | we test the parallaxes reported in the _ gaia _ first data release using the sample of eclipsing binaries with accurate , empirical distances from @xcite . we find an average offset of @[email protected] mas in the sense of the _ gaia _ parallaxes being too small ( i.e. , the distances too long )
. the offset does not depend strongly on obvious parameters such as color or brightness .
however , we find with high confidence that the offset may depend on ecliptic latitude : the mean offset is @[email protected] mas in the ecliptic north and @[email protected] mas in the ecliptic south .
the ecliptic latitude dependence may also be represented by the linear relation , @xmath2 mas ( @xmath3 in degrees ) .
finally , there is a possible dependence of the parallax offset on distance , with the offset becoming negligible for @xmath4 mas ; we discuss whether this could be caused by a systematic error in the eclipsing binary distance scale , and reject this interpretation as unlikely . |
the majority of galaxies in the local universe reside in groups .
early group samples were defined as overdense areas on the sky from photographic plates while redshift surveys later allowed for groupings in a third dimension .
investigations have sought to determine whether these systems are bound gravitationally or simply chance projections of unrelated galaxies along a line of sight .
observations of groups enveloped in a hot , x - ray emitting halo similar to the hot medium found in rich clusters @xcite along with evidence in others for interactions between group members have shown that many groups are indeed gravitationally bound systems .
galaxy properties , like morphology and star - formation rate ( sfr ) , are observed to change with environment ; high density regions , like clusters , are characterized by a larger fraction of elliptical galaxies and a lower star - formation rate than seen in the field @xcite .
these observations are known as the morphology - density and sfr - density relations , although @xcite argue that these reflect a tighter and more fundamental morphology - radius relation .
these relations are not limited to clusters , they are seen in groups of galaxies as well @xcite which suggests that the physical mechanisms which play a role in producing these relations are also important in the group environment .
these mechanisms include some combination of tidal interactions , mergers , ram pressure stripping , and strangulation .
although , there is debate over the effectiveness of ram pressure stripping and strangulation in the group environment @xcite .
we use because it is an excellent tracer of tidal interactions and often shows evidence for these gravitational disturbances when optical observations of the same source appear undisturbed .
additionally , observations of groups are able to probe a population of gas rich galaxies which may not be detected in optical observations .
our current understanding of the evolution of groups is derived from observed correlations between the fraction of a group s galaxies that are elliptical , the presence of a diffuse x - ray emitting halo , and a high velocity dispersion @xcite . additionally , numerical simulations suggest that galaxies within a compact group eventually merge to form large ellipticals @xcite .
thus , the evolution of a group should be marked by the removal of from the individual galaxies , dumping it into the intragroup medium where it can be heated , along with primordial intergalactic gas , to x - ray emitting temperatures .
this may be initially accompanied by enhanced star formation within the individual galaxies that leads to the feedback of diffuse hot gas into the intragroup medium @xcite .
we use these data to compile an hi mass function ( himf ) for galaxy groups . in general , the least well - known component of the himf
is the slope of the low mass end .
cosmological and galaxy evolution models depend on this slope as a constraint .
a handful of blind surveys , as well as targeted hi observations of individual clusters and groups of galaxies , have attempted to accurately determine the slope of the low mass end of the mass function @xcite .
evidence from these , and other , observations suggests that the low mass slope of the himf may vary with environment @xcite . in this paper
we examine the content of a diverse sample of galaxy groups .
galaxy morphologies are taken from ned when possible and if not , hyperleda . a hubble constant of @xmath2
is used throughout this paper .
lccccccccc ngc 7582 & 62 @xmath3 49&6.41 @xmath3 5.54 & 0.9/3.8 & 13 & 0.12 & @xmath4 & @xmath5 & 1.00 & 8 + usgc u207 & 62 @xmath3 60&25.5 @xmath3 24.7 & 0.58/0.75 & 11 & 2.2 & 122 & @xmath6 & 0.86 & 7 + usgc u070 & 67 @xmath3 60&23.9 @xmath3 21.4 & 0.78/0.98 & 6.6 & 0.92 & @xmath7 & @xmath8 & 1.00 & 5 + usgc u412 & 67 @xmath3 60&24.0 @xmath3 21.5 & 0.26/0.37 & 4.7 & 0.78 & 168 & @xmath9 & 0.90 & 10 + usgc u451 & 67 @xmath3 56&29.3 @xmath3 24.2 & 0.48/0.62 & 8.7 & 1.6 & 416@xmath1035 & 42.40 & 0.31 & 74
the five groups in this paper are found in various catalogs including , but not limited to , geller and huchra ( gh ) @xcite , uzc - ssrs2 ( usgc ) group catalog @xcite , the rosat all - sky survey ( rass ) center for astrophysics ( cfa ) loose system ( rasscals ) catalog @xcite , the group evolution multi - wavelength study ( gems ) project sample @xcite , and @xcite , a spectroscopic survey of poor groups , hereafter known as zm98 .
we identify most groups in this paper by their usgc number , although alternate identifications are available in the table notes for each group .
all of these catalogs , except the gems sample and zm98 , use the standard friends - of - friends algorithm outlined in huchra and geller @xcite with parent catalogs like the center for astrophysics ( cfa ) redshift survey , the updated zwicky catalog ( uzc ) , and the southern sky redshift survey ( ssrs2 ) to identify groups .
these parent catalogs have a limiting magnitude of @xmath11 .
the gems sample uses the nasa extragalactic database ( ned ) as a parent catalog and only accepts galaxies as group members if they are within these spatial and velocity constraints : a projected group radius , @xmath12 , calculated from the x - ray temperature of the group and a velocity range of @xmath13 .
we chose the five groups discussed in this paper in order to sample group environments spanning a range of spiral fraction .
however the spiral fraction as well as the group velocity dispersion can vary , for a given group , between catalogs and may change as additional redshift information allows for the identification of new group members .
the data for usgc u412 , u451 , and u207 were taken with the vla in d configuration ( longest baseline 1030 m ) in the spring of 2003 . in order to cover the largest velocity range possible we used the correlator in 4 if mode and centered each if pair at well separated velocity overlapping the three central channels which gave a @xmath14 mhz or @xmath15 bandwidth with a velocity resolution of @xmath16 khz or @xmath17 . during processing
we combined the ifs into one continuous cube using the aips task uvglu .
we used three by three and two by two pointing mosaics resulting in spatial coverage of @xmath18 by @xmath18 ( or 45@xmath19 by 45@xmath19 ) on the sky .
pointing centers were separated by the vla half - power beamwidth of 15@xmath19 and we integrated for 5 minutes on each pointing , shifting to the phase calibrator after each @xmath20 minute mosaic sequence .
usgc u070 was observed with the vla in the d configuration during august 2004 .
the correlator was configured in the standard 2 if mode with a bandwidth of 3.1 mhz or @xmath21 and channel size of 48.4 khz or @xmath22 .
a three by three pointing mosaic was used to cover the size of the group on the sky .
we used aips to calibrate all the vla data .
the ngc 7582 group was observed using the australia telescope compact array ( atca ) near narrabri in new south wales , australia .
we used a 23 pointing mosaic to cover an area of @xmath23 on the sky .
we obtained 60 hours of data in the 0.750a configuration ( longest baseline 3750 m ) in january 2002 and 50 hours of data in the ew352 configuration ( longest baseline 4439 m ) in june 2002 .
for these observations the atca correlator collected dual - polarization data with a total bandwidth of 16 mhz or @xmath24 and a channel width of 31.2 khz or @xmath25 .
the primary calibrator used to set the absolute flux scale was the standard atca flux calibrator 1934 - 638 ( 14.9 jy at 1384 mhz ; reynolds 1994 ) and was observed for approximately 10 minutes every 24 hours .
a secondary calibrator 2323 - 407 was observed for about 5 minutes every hour for phase calibration .
miriad was used for editing , calibration , and imaging of the visibility data . with calibrated data in hand we used miriad to mosaic , subtract the continuum , and deconvolve the dirty cubes for all the groups .
the naturally weighted cubes were deeply cleaned using mossdi , the cleaning task for mosaics .
line - free channels were identified , and miriad task uvlin was used to subtract the continuum in the uv plane .
moment maps were made by smoothing the cube by twice the spatial resolution and masking that cube at the three sigma noise level .
this smoothed , masked cube was then used as a mask when taking the moment of the original data cube . additionally , calibrated atca data on the ngc 7582 group were obtained in collaboration with b. koribalski at australia telescope national facility .
these data are single pointings in the 375 ( longest baseline 5969 m ) , 750d ( longest baseline 4469 m ) , and 1.5a ( longest baseline 4469 m ) configurations , adding a total of 100 more observing hours on this group .
table [ tab : groupdata ] illustrates the basic properties of the five groups and their data cubes .
mosaicing produces an area of uniform sensitivity where the primary beams overlap surrounded by an area with increased noise .
column 4 gives the rms noise ( in mjy @xmath26 ) for the center and edge of the field in a line free channel of the deconvolved , naturally weighted cube .
column density and mass sensitivity calculations correspond to a three sigma detection at the lower rms noise level from the center of the field .
the column density sensitivity is from one line free channel while the mass sensitivity assumes a detection over three channels filling one beam .
lcccccc[ht ] eso 347-g 008 & 23 20 49 & -41 43 51 & sab(s)m & 1622 @xmath10 76 & 8.6 & 0.94 + ngc 7599 & 23 19 21 & -42 15 25 & sb(s)c & 1654 @xmath10 7 & 35 & 3.8 + ngc 7590 & 23 18 55 & -42 14 21 & s(r?)bc sy2 & 1596 @xmath10 7 & 20 & 2.1 + ngc 7582 & 23 18 24 & -42 22 14 & ( r_1)sb(s)ab sy2 & 1575 @xmath10 7 & 21 & 2.3 + lcrs b231454.2 - 423320 & 23 17 39 & -42 16 55 & - & 1609 @xmath10 80 & 0.33 &
0.036 + ngc 7552 & 23 16 11 & -42 34 59 & ( r)sb(s)ab;hiiliner & 1585 @xmath10 5 & 36 & 3.9 optical data were obtained for usgc u412 , u451 , and u207 with the wiyn 0.9 m telescope on kitt peak in arizona during march 2003 to search for optical counterparts to the detections .
groups were imaged in b and r filters using the ccd mosaic imager .
the field of view of this instrument on the wiyn 0.9 m telescope covers @xmath18 by @xmath18 and we used two pointing offsets to fill in the gaps between chips .
these images were reduced using the _ mscred _ package in iraf .
standard calibration procedures were used including overscan and cross talk corrections , bias subtraction , and flat fielding .
once we had calibrated single images , three per pointing , we stacked them to remove the chip gaps and produce a final image .
for usgc u070 and the ngc 7582 group we use digitized sky survey ( dss ) optical data to search for counterparts .
the ngc 7582 group , also known as the grus group , has eight spectroscopically confirmed members in zm98 all of which are classified as late - type galaxies . in zm98
the morphological classification of group members was done by examination of images and assignments were simply given as early ( e and s0 ) or late ( all others ) .
the group velocity is @xmath27 and the velocity dispersion is @xmath28 ( zm98 ) .
_ rosat _ failed to detect the presence of any diffuse x - ray emission @xcite as did xmm - newton @xcite with an exposure of 17.8 ksec giving an upper limit of @xmath29 .
we detect in six group galaxies : ngc 7599 , ngc 7590 , ngc 7582 , ngc 7552 , eso 347-g008 , and lcrs b231454.2 - 423320 .
we do not detect in the optically compact galaxies eso 291-g015 , which has ned morphology of sa and is well within the spatial and velocity bounds of our data , and ngc 7632 , which has a ned morphology of so and is near the edge of our field of view .
ngc 7582 and ngc 7552 are classified as active ( seyfert 2 ) galaxies @xcite and is seen in absorption against their nuclei @xcite .
table [ tab : n7582 ] lists the detections and figure [ fig:7582 ] shows the contours on an optical dss image of the group .
the compact core of the group shows multiple signatures of interactions including a large tidal extension with an integrated flux density of @xmath30 or @xmath31 which reaches from ngc 7582 towards ngg 7590 as well as southeast towards ngc 7552 .
there may also be an bridge between ngc 7599 and ngc @xmath32 but we are not able to resolve this structure .
our ngc 7582 cube covers a velocity range from @xmath33 which corresponds to distances of @xmath34 .
the region of uniform sensitivity on the sky is square with @xmath35 on a side which corresponds to @xmath36 at a @xmath37 distance , 0.5 mpc at the distance of the group , and @xmath38 at a @xmath39 distance .
thus , the volume surveyed for this group is @xmath40 .
this group has seven members : ic 0527 , ic 2434 , ugc 04799 , ngc 2759 , cgcg 180 - 036 ned01 , cgcg 180 - 048 , and cgcg 180 - 050 .
we detect in ic 0527 , ic 2434 , ugc 04799 as well as two new galaxies previously not classified as group members : fgc 0841 and cgcg 180 - 045 .
all detections are spiral galaxies except cgcg 180 - 045 whose morphology is unknown .
cgcg 180 - 036 is a galaxy pair with discrepant velocities such that cgcg 180 - 036 ned2 is a background object and not a group member .
cgcg 180 - 048 is way beyond the spatial bounds of our data .
ic 2434 has an interesting hi morphology that is reminiscent of ram pressure stripping with the classical head - tail morphology showing compressed contours opposite extended contours .
however , in this environment it is much more likely that this morphology is caused by a tidal interaction with nearby galaxy ugc 04799 .
the heliocentric group velocity is @xmath41 and the group velocity dispersion is @xmath42 @xcite . @xcite give an upper limit on the x - ray emission from this group of @xmath43 .
the dynamic range of our data for this system is limited by a bright continuum source in the field which does nt subtract cleanly from both ifs
. table [ tab : gh40 ] lists the detections in this group and figure [ fig : gh40 ] shows the total intensity hi contours on an optical image of the group . our usgc
u207 cube covers a velocity range from @xmath44 which corresponds to distances of @xmath45 .
the region of uniform sensitivity on the sky is square with @xmath46 on a side which corresponds to @xmath47 at a @xmath48 distance , @xmath49 at the distance of the group , and @xmath50 at a @xmath51 distance .
thus , the volume surveyed for this group is @xmath52 .
lccccccccc ic 0527 & 9 09 41.8 & + 37 36 05 & s ? & 6889 @xmath10 11 & 9.8 & 20 + cgcg 180 - 045 & 9 09 24.3 & + 37 41 09 & - & 7138 @xmath10 65 & 0.49 & 0.97 + ugc 04799 & 9 08 47.6 & + 37 30 06 & sb & 6873 @xmath10 27 & 0.55 & 1.1 + fgc 0841 & 9 07 38.5 & + 37 40 21 & sc & 6976 @xmath10 24 & 1.9 & 3.8 + ic 2434 & 9 07 16.0 &
+ 37 12 55 & sb ?
& 7158 @xmath10 29 & 4.6 & 9.4 usgc u070 , also know as the ngc 664 group , has five spectroscopically identified members in zm98 : ngc 664 , ic 0150 , and ucg 01204 are classified as spirals while 2masx j01431801 + 0411322 , and ngc 664:[zm98 ] 057 ( ned identification ) have unknown morphologies .
the group velocity is @xmath53 and the group velocity dispersion is reported to be @xmath7 ( zm98 ) . @xcite
give an upper limit of @xmath54 for diffuse x - ray emission from the intragroup medium using _ rosat _ pspc data with an exposure time of 7.5 ks .
we detect in the three group members previously classifed as spirals , as well as two new galaxies : ic1726 and its uncataloged neighbor ( @xmath0 , @xmath1 ) which we are designating fsw j014526.92 + 043619.1 according to iau nomenclature conventions .
additionally , we detect @xmath55 of near fsw j014526.92 + 043619.1 without an optical counterpart which may be tidal material or an cloud .
we do not detect in either of the two group members with unknown morphologies .
although ngc 664:[zm98 ] 057 and 2masx j01431801 + 0411322 both have emission lines in their spectra @xcite , ngc 664:[zm98 ] 057 is spatially beyond the coverage of our data and 2masx j01431801 + 0411322 is optically very compact compared to the other galaxies in this group .
its mass may fall below our detection limit .
table [ tab:664 ] lists the detections in this group and figure [ fig:664 ] shows the total intensity contours on an optical image of the group with an inset showing the uncataloged galaxy .
our ngc 664 cube covers a velocity range from @xmath56 which corresponds to distances of @xmath57 .
the region of uniform sensitivity on the sky is square with @xmath46 on a side which corresponds to @xmath58 at a @xmath59 distance , @xmath60 at the distance of the group , and @xmath61 at a @xmath62 distance .
thus , the volume surveyed for this group is @xmath63 .
lcccccc ic 0150 & 01 42 58 & + 04 11 58 & sb & 5551 @xmath10 38 & 2.6 & 3.4 + ugc 01204 & 01 43 12 & + 04 16 42 & sbc & 5621 @xmath10 41 & 2.8 & 3.6 + ngc 664 & 01 43 46 & + 04 13 22 & sb & 5424 @xmath10 5 & 2.9 & 3.7 + ic 1726 & 01 45 20 & + 04 37 06 & - & 5334 @xmath10 36 & 3.1 & 4.0 + fsw j014526.92 + 043619.1 & 01 45 27 & + 04 36 16 & - & 5320 @xmath10 10 & 1.0 & 1.3 + cloud & 01 45 39 & + 04 34 43 & - & 5230 @xmath10 20 & 0.34 & 0.44 this group has ten spectroscopically identified members in the uzc - ssrs2 group catalog ; nine of them are classified as spiral galaxies .
the group velocity is @xmath64 and the velocity dispersion is @xmath65 @xcite .
we detect in five of these galaxies , namely ngc 3833 , cgcg 068 - 046 , cgcg 068 - 044 , ngc 3822 , ugc 06647 .
we also detect two new group members in : ngc 3820 , which was not originally classified as a member of usgc u412 although it is a member of the compact subgroup hcg 58 , and sdss j114140.99 + 102956.4 .
cgcg 068 - 044 , cgcg 068 - 046 , and ngc 3833 have large disks extending from three to five times the optical radius of the galaxy .
ngc 3819 , the only early type group member , is not detected , while ngc 3839 and ic 0727 are undetected because they are outside the field of view of our data .
the apparent lack of an detection for ngc 3825 will be discussed in the next section with the content of hcg 58 .
table [ tab : gh89 ] lists the detections in this group and figure [ fig : gh89 ] shows the total intensity contours on an optical image of the group . our usgc
u412 cube covers a velocity range from @xmath66 which corresponds to distances of @xmath67 .
the region of uniform sensitivity on the sky is square with @xmath68 on a side which corresponds to @xmath69 at a @xmath70 distance , @xmath71 at the distance of the group , and @xmath72 at a @xmath73 distance .
thus , the volume surveyed for this group is @xmath74 .
lcccccc cgcg 068 - 046 & 11 43 41.7 & + 10 14 44 & sc & 6009 @xmath10 18 & 3.4 & 5.3 + cgcg 068 - 044 & 11 43 29.3 & + 10 20 42 & sc & 6013 @xmath10 19 & 2.1 & 3.2 + ngc 3833 & 11 43 28.9 & + 10 09 43 & sc & 6036 @xmath10 10 & 13 & 19 + ngc 3822 + extended&11 42 11.1 & + 10 16 40 & sb & 6138 @xmath10 20 & 9.0 & 15 + ngc 3820 & 11 42 04.9 & + 10 23 03 & sbc & 6102 @xmath10 33 & 0.68 & 1.0 + sdss j114140.99 + 102956.4 & 11 41 41.0 & + 10 29 56 & - & 6163 @xmath10
48 & 0.31 & 0.48 + ugc 06647 & 11 41 03.7 & + 10 13 31 & sd & 6350 @xmath10 66 & 1.7 & 2.6 most hickson compact groups ( hcgs ) are dynamically associated with larger loose groups @xcite .
this may explain the observed longevity of these systems when compared with n - body simulations which show an isolated compact group evolving into a single merger remnant in only a few orbital times @xcite .
hcg 58 @xcite appears to be at the core of the previously discussed loose group , usgc u412 .
hickson finds five members in hcg 58 , four of which are classified as spirals .
we detect in three of them , ngc 3820 , ngc 3817 , and ngc 3822 .
there is also an extension towards ngc 3825 but very little coincident with the optical galaxy .
the last group member , ngc 3819 , classified as e1 is undetected .
hcg 58 is classified as deficient by who find a total mass of @xmath75 compared to our total mass of @xmath76 which leaves little room for deficiency .
figure [ fig : hcg58 ] shows the emission ( contours on an optical image of hcg 58 ) from this group as a function of velocity .
the morphology of ngc 3822 is severely disturbed .
we see an extension towards ngc 3825 which contains a similar amount of as seen in previous single dish measurements ( @xmath77 ) @xcite .
it appears that ngc 3825 has had most of its stripped off into the intragroup medium due to recent interactions with the other group members .
the published x - ray data for this group do not agree on whether a hot intragroup medium exists .
@xcite find an x - ray luminosity , whose profile is dominated by a point source , of @xmath78 using the einstein x - ray satellite , while the rasscals study @xcite finds @xmath79 using rosat all - sky survey data .
however , using only pointed - mode rosat data @xcite do not detect x - rays in hcg 58 and put upper limits on the x - ray luminosity at @xmath80 .
nonetheless , @xcite have shown that the intragroup medium in even small groups can be relatively dense and remain undetected by the current generation of x - ray telescopes .
this group has the highest velocity dispersion of the five groups presented in this paper and is the only group with a definitive detection of x - ray emitting intragroup gas .
a redshift survey by @xcite finds 74 group members , a group velocity of @xmath81 , and a group velocity dispersion of @xmath82 .
the group has two compact subgroups , uzc - cg 156 ( including ngc 4076 ) and uzc - cg 157 ( including ngc 4098 ) which were identified by the 3d uzc catalog solely on the criterion of compactness unlike hcgs which were visually selected using multiplicity , isolation , and luminosity concordance of member galaxies .
@xcite detect bimodal x - ray emission with one peak on the galaxies ngc 4061 and ngc 4065 and the other on ngc 4066 .
the bimodal x - ray emission does not correlate with the two compact subgroups .
the gems @xcite and rasscals @xcite x - ray luminosities are in close agreement with @xmath83 and @xmath84 respectively .
we detect in vv 062 , ngc 4092 , ngc 4086 , 2masx j12063969 + 2027235 , and maps - ngp o 377 1582735 and place lower limits on the content of ngc 4098 , ngc 4090 , ngc 4076 , and kug 1203 + 206b .
ngc 4098 is a system of two interacting spiral galaxies which is itself interacting with the pair of spiral galaxies , vv062a and vv062b .
ngc 4086 and ngc 4092 are connected by an bridge . at the time of the observations
the vla did not have the bandwidth necessary to sample the entire velocity range of the group which spans from @xmath85 @xcite .
additionally , our nine pointing vla mosaic has only spatially sampled one quarter of the group size .
table [ tab : gh98 ] lists the detections in this group and figure [ fig : gh98 ] shows the total intensity contours on an optical image of the group .
galaxies whose profiles are truncated by the edge of our data cube are indicated with lower limits in table [ tab : gh98 ] .
our usgc u451 cube covers a velocity range from @xmath86 which corresponds to distances of @xmath87 .
the region of uniform sensitivity on the sky is square with @xmath68 on a side which corresponds to @xmath88 at a @xmath89 distance , @xmath90 at the distance of the group , and @xmath91 at a @xmath92 distance .
thus , the volume surveyed for this group is @xmath93 .
lcccccc 2masx j12063969 + 2027235 & 12 06 39.7 & + 20 27 24 & s ?
& 7263 @xmath10 16 & 0.79 & 1.8 + vv 062 & 12 06 18.3 & + 20 36 40 & sb & 7217 @xmath10 18 & 0.85 & 1.8 + ngc 4098 & 12 06 03.9 & + 20 36 22 & sbc & 7296 @xmath10 11 & @xmath94 & @xmath95 + ngc 4092 & 12 05 50.1 & + 20 28 37 & sa & 6719 @xmath10 8 & 1.6 & 3.0 + kug
1203 + 206b & 12 05 47.4 & + 20 19 35 & s & 6984 @xmath10 15 & @xmath96 & @xmath97 + ngc 4090 & 12 05 27.9 & + 20 18 32 & sab & 7333 @xmath10 45 & @xmath98 & @xmath99 + ngc 4086 & 12 05 29.4 & + 20 14 48 & s0 & 7093 @xmath10 24 & 2.2 & 4.7 + maps - ngp o_377_1582735 & 12 05 11.8 & + 20 31 07 & - & 7100 & 2.3 & 4.9 + ngc 4076 & 12 04 32.5 & + 20 12 18 & sa & 6214 @xmath10 6 & @xmath100 & @xmath101
the ngc 7582 , ngc 664 , and usgc u412 groups are the most spiral rich groups and they all contain varying amounts of intragroup .
this material is tidally stripped from the galaxies themselves .
these tidal interactions , which are so favorable in the group environment , are likely to play an important role in the evolution of galaxies in these systems .
as is seen here , they have the potential to remove large amounts of neutral gas from the galaxies involved .
this tidally stripped may fall back onto the galaxies or eventually be photoionized by the uv background @xcite .
the two systems with the most intragroup , ngc 7582 and hcg 58 ( a subgroup of usgc u412 ) , are both compact cores of larger systems .
these compact configurations with small impact parameters and small relative velocities between galaxies are where tidal forces will be most efficient at removing from inside galaxies into the intragroup medium .
examine the content of compact groups of galaxies and find that @xmath102 of spiral galaxies in compact systems have a perturbed morphology .
they present an evolutionary scenario that begins with the least evolved groups showing little evidence for interactions among group members .
intermediately evolved groups have tidal features and evolved groups may have a single cloud enveloping the group or show extreme deficiencies in galaxy content .
@xcite find evidence for a significant reservoir of diffuse in compact systems that is likely a result of the tidally stripped being dispersed throughout the intragroup medium .
they find a larger component of this diffuse in compact groups with deficient galaxies . within groups of galaxies ,
the formation of compact subgroups may facilitate interactions and lead to the tidal removal of significant amounts of .
the loss of neutral gas from these galaxies may then restrict their star - forming potential .
a handful of recent large surveys have attempted to accurately determine the slope of the low mass end of the mass function ( himf ) .
predictions from cold dark matter models of galaxy formation suggest that the faint end slope should be rather steep @xcite .
if these dark matter haloes contain then the corresponding himf should also have a steep faint end slope .
cosmological and galaxy evolution models use the behaviour of the mass function at the low mass end as a constraint .
we use the data in this paper , as well as additional data in our possession for the following groups : hcg 22 @xcite , ngc5846 and ngc 2563 @xcite to compile an mass function for the group environment .
these additional data were taken as mosaics with the vla and reduced in a similar manner to the data presented in this paper .
we do not have enough data to fit an independent mass function but we do check the consistency of our mass function against previous himf results .
the himf , shown in figure [ fig : himf ] , is the space density of objects in units of @xmath103 .
it is parametrized by a schechter function of the form @xmath104 where @xmath105 is the mass at which the function turns over , @xmath106 is the slope of the low mass end , and @xmath107 is a normalization .
recent large surveys have allowed for the determination of the himf with increased precision . @xcite
use 4315 galaxies from the parkes all - sky survey ( hipass ) to construct an himf and derive a low mass slope of @xmath108 . with 265 galaxies the arecibo dual - beam survey ( adbs )
@xcite finds an even steeper faint end slope of @xmath109 .
@xcite carried out a sensitive search for companions to otherwise isolated spiral galaxies .
their survey covered a total of 31.6 mpc@xmath110 around 41 different spirals , and the faint end slope of their himf is consistent with values flatter than @xmath111 .
there is some evidence for variation in the himf with environment . using only galaxies near the virgo cluster @xcite
find a faint end slope that is much more flat ( @xmath112 ) than what is seen in the full sample of galaxies .
showed that the mass functions in both the centaurus a and sculptor groups are inconsistent with a slope steeper than @xmath113 .
@xcite perform a deep survey of the canis venatici group and report a flat ( @xmath114 ) himf .
another flat himf is seen in a deep survey of the ursa major group by @xcite .
@xcite use 2771 detections from an optically selected sample of galaxies to derive an himf whose faint - end slope is @xmath115 for the full sample and see evidence for a flattening of the himf when examined in increasingly dense environments .
however , @xcite using hipass data find tentative evidence for a steepening of the faint - end slope of the himf in increasingly rich environments .
they parametrize the density around a given galaxy by calculating the distance to its n nearest hipass detected neighbors .
the flattening of the faint end slope of the himf suggests that physical processes in the group environment are responsible for removing gas from low mass galaxies .
large surveys must deal with volume corrections which account for the fact that low luminosity objects are more easily detected nearby and their contribution throughout the whole volume must be weighted accordingly .
these corrections must also account for the local overdensity that surrounds the milky way . since all galaxies in a group are at approximately the same distance no correction needs to be made for large scale structure effects throughout the depth of the survey .
we do apply a volume correction , discussed below , to account for low mass galaxies which are not detectable throughout an entire data cube .
@xcite explore the impact of distance uncertainties on the faint - end slope of mass and luminosity functions .
they use mock survey data to show that for galaxies in the virgo region the low - mass himf slope is underestimated if peculiar velocities are neglected when finding the distances to objects in the nearby universe .
objects with @xmath116 and typical peculiar velocities of a few hundred kilometers per second are well within a pure hubble flow and have errors of less than @xmath117 on their redshift distances .
we have not made any correction for the groups in our sample with @xmath118 because they are all in the anti - virgo direction where the local peculiar velocity field is not well constrained .
low mass galaxies which can be detected near the front of the cube may be undetectable at the back side of the cube .
this means that we are sensitive to low masses in a subset of the total volume surveyed . to account for this
we apply a volume correction to the lowest three mass bins @xcite . to determine this correction for a given cube we calculate the actual volume over which we could detect a galaxy with a given mass based on the noise level and size of the cube .
then we add the volumes calculated by this method together for all cubes .
the himf is then produced by dividing the total number of detections in a mass bin by the corresponding volume surveyed for masses in that bin among all groups .
our mass function for the group environment , seen in figure [ fig : himf ] , is essentially flat and inconsistent with the hipass and adbs mass functions .
our himf is similar to those compiled for other group environments .
the volume corrections for the last two bins are very large with only @xmath119-@xmath117 of the total volume being surveyed for galaxies with those masses .
of the ten galaxies in the last three bins six of them ( including all the galaxies in the last two bins ) are detections from galaxy group ngc 5846 @xcite . however , the flattening of the low mass slope is apparent even neglecting the last three bins
. a large and sensitive survey that can populate the lower mass bins and whose detections can be reliably sorted by environment will be necessary to address the variety currently seen in the himf .
we have performed mosaicked vla and atca observations of five groups of galaxies .
we find a multitude of interactions and interesting morphologies in these systems .
intragroup , not easily associated with a single group member , is present in the three most spiral rich systems .
an mass function compiled from these data , and an additional three groups , is consistent with other mass functions for the group environment in that it has a flat low mass slope .
the flattening of the faint end slope of the himf suggests that physical processes in the group environment are responsible for removing gas from low mass galaxies .
we would like to thank the referee for useful comments and gratefully acknowledge the support of nsf grant ast-0506628 . ef acknowledges the support of the wisconsin space grant consortium in the form of a graduate fellowship .
this research has made use of the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , caltech , under contract with the national aeronautics and space administration .
the national radio astronomy observatory is a facility of the national science foundation operated under cooperative agreement by associated universities , inc . , s. , yun , m. s. , & verdes - montenegro , l. 2008 , in american institute of physics conference series , vol .
1035 , the evolution of galaxies through the neutral hydrogen window , ed .
r. minchin & e. momjian , 201204 , w. k. , & richter , o .- g .
1989 , a general catalog of hi observations of galaxies . the reference catalog .
( a general catalog of hi observations of galaxies . the reference catalog , xix , 350 pp . 8 figs .. springer - verlag berlin heidelberg new york ) , m. a. w. , trentham , n. , tully , r. b. , & zwaan , m. a. 2000 , in astronomical society of the pacific conference series , vol . 218 , mapping the hidden universe : the universe behind the mily way - the universe in hi , ed
. r. c. kraan - korteweg , p. a. henning , & h. andernach , 263+ | we present the results of observations of five groups of galaxies spanning a range of velocity dispersion and spiral fraction ( brightest optical group member in parenthesis ) : ngc 7582 ( ngc 7552 ) , usgc u207 ( ngc 2759 ) , usgc u070 ( ngc 664 ) , usgc u412 ( ngc 3822 ) , usgc u451 ( ngc 4065 ) .
neutral intragroup gas is detected in three of the five groups .
we present the discovery of a previously uncataloged galaxy in the usgc u070 group at @xmath0 , @xmath1 which we are designating fsw j014526.92 + 043619.1 .
we compile an mass function for the group environment and find that the faint - end slope is consistent with being flat . |
supermassive black holes are now believed to inhabit the nuclei of all massive galaxies .
furthermore , the active galactic nucleus , or agn , phase is generally understood to be a short - term event in the life of a typical black hole , triggered either by a merger event or secular processes in the host galaxy ( cf .
the review of @xcite and references therein ) .
tight scaling relationships between the observed properties of black holes and their host galaxies point to a symbiotic relationship between the two ( e.g. , @xcite ) , in which the growth of structure and the evolution of galaxies across cosmic time is fundamentally linked to supermassive black holes . understanding this link requires an understanding of black hole demographics , not just in the local universe , but also at higher redshift where we can witness the growth of structure occurring .
black holes , as opposed to galaxies , are incredibly simple objects that can be fully characterized with only two fundamental measurements : mass and spin . in the milky way , years of astrometric monitoring of stars in the central @xmath8parsec have led to an extremely precise determination of the mass of our own supermassive black hole @xcite .
unfortunately , all other galaxies are too distant for this same technique to be employed , and different techniques must be used to understand the masses of a population of central black holes . for galaxies out to @xmath9mpc , spatially - resolved observations of the bulk motions of stars or nuclear gas disks can be combined with dynamical modeling to constrain the central black hole mass ( cf .
the reviews of @xcite ) .
reverberation mapping @xcite , on the other hand , takes advantage of agn flux variability to constrain black hole masses through time - resolved , rather than spatially - resolved , observations , thus obviating any distance limitations .
furthermore , the most widely - used technique to constrain supermassive black hole spins requires high x - ray luminosities that are only found in agns ( e.g. , @xcite and references therein ) , so the study of active black holes is an important key to unraveling the growth and evolution of cosmic structure .
unfortunately , bright agns are relatively rare in the local universe , leading to a disconnect in our current understanding of nearby black holes compared to those observed at larger look - back times .
in particular , we are lacking direct comparisons of black hole mass constraints through multiple independent techniques in the same galaxies .
there are a handful of published comparisons of reverberation masses and gas dynamical masses ( e.g. , @xcite ) , including the low - mass seyfert ngc4395 ( @xcite ) .
the agreement is generally quite good , although the number of galaxies studied is small .
stellar dynamics , on the other hand , is a good check against reverberation masses because it relies on modeling a non - collisional system , unlike gas dynamics where the agn may be expected to inject energy on resolvable spatial scales .
however , only two such comparisons currently exist for black hole masses from reverberation mapping and stellar dynamical modeling : ngc4151 @xcite and ngc3227 @xcite .
while the techniques give roughly consistent masses for these two examples , there are caveats and limitations to both reverberation mapping and dynamical modeling , and a larger comparison sample is needed to fully assess the consistency of the local and the cosmological black hole mass scales .
we have therefore undertaken a program to identify and monitor local agns where it might be possible to obtain both a reverberation and a stellar dynamical mass constraint .
both techniques are time- and resource - intensive , and there are very few broad - lined agns within @xmath10 , where the spatial resolution provided by @xmath11-m class telescopes would be likely to resolve the black hole s gravitational influence on the nuclear stellar dynamics , but we hope to increase the sample of mass comparisons by a factor of a few .
we currently have stellar dynamical modeling underway for two other local agns , and we describe here the reverberation results for an additional local agn in our sample , ugc06728 .
ugc06728 is a low - luminosity seyfert 1 located at @xmath1211:45:16.0 , @xmath1379:40:53 , @xmath14 in a late - type galaxy that is highly inclined to our line of sight .
it was monitored nightly over the course of two months in the spring of 2015 .
optical spectroscopy and photometry were obtained at apache point observatory in new mexico , with additional supporting photometry obtained at hard labor creek observatory in georgia .
we describe the details below .
spectrophotometric monitoring of ugc06728 was carried out at apache point observatory ( apo ) with the 3.5-m telescope from 2015 april 15 @xmath15 may 30 ( ut dates here and throughout ) .
our monitoring program was scheduled for the first hour of almost every night during this time period , coincident with evening twilight .
we employed the dual imaging spectrograph ( dis ) , which uses a dichroic to split the incoming beam into a red arm and a blue arm , with the low - resolution ( b400/r300 ) gratings centered at 4398 and 7493 .
the b400 and r300 gratings , when used together , cover the entire optical bandpass between the atmospheric cutoff and 1@xmath16 m , with a nominal dispersion of 1.8 / pix and 2.3 / pix respectively .
spectra were obtained through a 5 slit rotated to a position angle of 0(oriented north - south ) and centered on the agn . on each visit ,
a single spectrum with an exposure time of 600s was acquired at a typical airmass of 1.5 .
observations of the spectrophotometric standard star feige34 were also acquired with each visit .
all spectra were reduced with iraf following standard procedures .
an extraction width of 12pixels was adopted , corresponding to an angular width of 5 and 4.8 for the blue and red cameras , respectively .
the desire to minimize sampling gaps and maximize temporal coverage means that ground - based reverberation campaigns must rely on spectroscopy obtained under nonphotometric conditions . while a spectrophotometric standard star can help correct the overall shape of the spectrum for atmospheric effects , as well as those from the telescope and instrument optics , an additional technique is required to achieve absolute flux calibrations of all the spectra .
fortuitously , the narrow emission lines do not vary on short timescales of weeks to months , so they can serve as convenient `` internal '' flux calibration sources .
we utilize the @xcite spectral scaling method , which accounts for small differences in wavelength calibration , flux calibration , and resolution ( from variations in the seeing ) .
the method compares each spectrum to a reference spectrum built from the best spectra ( identified by the user ) and minimizes the differences within a specified wavelength range .
the method has been shown to result in relative spectrophotometry that is accurate to @xmath17% @xcite .
we restricted the scaling algorithm to focus on the spectral region containing the [ ] @xmath184959,5007 doublet .
additionally , we adopted an overall flux scale based on the integrated [ ] @xmath19 flux measured from the nights with the best observing conditions of @xmath20ergss@xmath5@xmath21 .
the red - side spectra showed only h@xmath22 emission smoothly blended with [ ] @xmath186548,6583 .
emission from [ ] @xmath186716,6730 and [ ] @xmath186300,6363 is extremely weak and difficult to detect above the continuum . with no suitable narrow lines available ,
we were unable to accurately intercalibrate the red - side spectra and we do not consider them further . figure [ fig : mean.rms ] displays the final mean and root mean square ( rms ) of all the calibrated spectra acquired throughout the campaign .
the rms spectrum displays the variable spectral components , of which h@xmath0 , @xmath234686 , and h@xmath24 are apparent , as is the agn continuum .
broad - band @xmath25 and @xmath26 images were obtained at apo with the imaging mode of the dis spectrograph each night directly after acquiring the spectra .
the dual - arm nature of the spectrograph allowed both images to be obtained simultaneously .
the typical exposure time was 30s , and a single image in each filter was obtained per visit .
images were reduced in iraf following standard procedures .
the dis imaging mode provides a relatively small field of view ( @xmath27@xmath287 ) , but there were a handful of convenient bright stars in all of the images ( see figure [ fig : apo.r ] ) .
we carried out aperture photometry employing circular apertures with radii of 378 in @xmath25 and 36 in @xmath26 , and sky annuli of 63@xmath15756 and 60@xmath1572 respectively .
calibrated @xmath29 and @xmath30band magnitudes for three field stars were adopted from apass ( the aavso photometric all sky survey ; @xcite ) and set the photometric zeropoints .
photometric monitoring was also carried out with the 24-inch miller telescope at hard labor creek observatory ( hlco ) , owned and operated by georgia state university in hard labor creek state park near rutledge , ga .
@xmath31band images were acquired with an apogee @xmath32 detector , spanning a field of view of @xmath33 with a pixel scale of @xmath34 . on a typical night
, three exposures were obtained at an airmass of @xmath35 , each with an exposure time of 300s .
the wide field of view of the hlco images included a large number of field stars , allowing us to derive a @xmath31band light curve for ugc06728 by employing image subtraction techniques .
we first registered all the images to a common alignment with the sexterp package @xcite .
we then carried out the image subtraction analysis with the isis package @xcite .
isis builds a reference frame from the best images ( specified by the user ) and then uses a spatially - varying kernel to convolve the reference frame to match each individual image in the dataset .
subtraction of the two results in a residual image in which all constant components have disappeared and only variable flux remains . in the case of ugc06728 ,
the host - galaxy and the average agn brightness are subtracted from all the residual images , leaving behind only the brightness of the agn relative to its mean level .
aperture photometry is then employed to measure this variable flux , which may be positive or negative , at the location of the target of interest in each residual image , providing a @xmath31band residual light curve .
lcccc[h ! ] 7127.6059 & @xmath36 & @xmath37 & 7134.7311 & @xmath38 + 7130.6023 & @xmath39 & @xmath40 & 7135.8689 & @xmath41 + 7131.6043 & @xmath42 & @xmath43 & 7136.7193 & @xmath44 + 7134.6484 & @xmath45 & @xmath46 & 7142.6382 & @xmath47 + 7134.6493 & @xmath48 & @xmath49 & 7144.6886 & @xmath50 + 7135.6058 & @xmath51 & @xmath52 & 7145.6625 & @xmath53 + 7141.6148 & @xmath54 & @xmath55 & 7146.8040 & @xmath56 + 7142.6108 & @xmath57 & @xmath58 & 7147.6872 & @xmath59 + 7143.6100 & @xmath60 & @xmath61 & 7148.5865 & @xmath62 + 7144.6099 & @xmath63 & @xmath64 & 7150.6653 & @xmath65 + 7149.6224 & @xmath66 & @xmath67 & 7151.7214 & @xmath68 + 7150.6151 & @xmath69 & @xmath70 & 7152.7171 & @xmath71 + 7151.6161 & @xmath72 & @xmath73 & 7153.6520 & @xmath74 + 7152.6159 & @xmath75 & @xmath76 & 7156.7405 & @xmath77 + 7153.6223 & @xmath78 & @xmath79 & 7158.6132 & @xmath80 + 7156.6189 & @xmath81 & @xmath82 & 7159.6255 & @xmath83 + 7159.6214 & @xmath84 & @xmath85 & 7160.6317 & @xmath86 + 7160.6215 & @xmath87 & @xmath88 & 7162.6928 & @xmath89 + 7162.6218 & @xmath90 & @xmath91 & 7164.6410 & @xmath92 + 7165.6511 & @xmath93 & @xmath94 & 7165.6504 & @xmath95 + 7166.6239 & @xmath96 & @xmath97 & 7166.7008 & @xmath98 + 7168.6555 & @xmath99 & @xmath100 & 7167.7025 & @xmath101 + 7169.6224 & @xmath102 & @xmath103 & 7172.6709 & @xmath104 + 7170.6218 & @xmath105 & @xmath106 & 7173.6752 & @xmath107 + 7171.6249 & @xmath108 & @xmath109 & & [ tab : lc.phot ]
light curves for the broad emission lines h@xmath0 , @xmath110 , and h@xmath24 were derived directly from the scaled spectra .
we fit a local , linear continuum below each emission line and then integrated the flux above this continuum to determine the total emission - line flux .
this includes the contribution from the narrow component of each emission line , which is simply a constant flux offset .
we also determined a continuum light curve from the spectra at @xmath111 , which has the merit of being completely uncontaminated by emission lines .
the strong continuum and emission - line variability over the course of the campaign allows us to determine these light curves directly from the spectra without carrying out any spectral modeling or decomposition , which has the potential to introduce artificial features into light curves . in figure
[ fig : lc.cont ] , we show the spectroscopic continuum light curve relative to the @xmath31band residual light curve and the @xmath25 and @xmath26 photometric light curves ( tabulated in table [ tab : lc.phot ] ) .
the @xmath31band residual light curve does not contain significant emission from any broad emission lines , so we combined it with the continuum light curve determined from our spectra to improve the time sampling , especially in the first half of the campaign .
we selected pairs of points from the two light curves that were contemporaneous within @xmath112days and fit for the best multiplicative and additive factors to bring the @xmath31band residual fluxes into agreement with the measured continuum flux densities .
these best - fit factors account for the differences in host - galaxy background light , average agn flux level , and bandpass .
the @xmath31band light curve was scaled according to the best - fit parameters and merged with the continuum light curve .
we then examined the @xmath29band light curve from the apo photometry and found that there was no significant time delay relative to the merged continuum+@xmath113 light curve , so we merged it as well by again finding the multiplicative and additive scale factors necessary to bring it into agreement with contemporaneous points in the continuum+@xmath113 light curve .
our final merged continuum light curve was binned to 0.5day sampling to improve the accuracy .
the overall shape of the @xmath30band light curve agrees with the other photometric light curves and the continuum light curve , but the variability level is somewhat damped by additional host - galaxy flux and there is possibly a slight delay in the light curve , so we did not merge the @xmath30band with the other light curves .
a detectable delay in @xmath26 is not unexpected , given that the filter bandpass is centered on h@xmath22 . while @xmath25 is centered on h@xmath0 , the overall contribution of h@xmath0 to the total filter bandpass is much smaller than for h@xmath22 and @xmath26 . in particular , h@xmath0 contributes only 2% of the @xmath29band flux , with the variable component of h@xmath0 accounting for only 10% of the total h@xmath0 contribution , or 0.2% of the total @xmath29band flux . on the other hand
, h@xmath22 contributes 15% of the total @xmath30band flux .
lcccc 7127.60131629 & @xmath114 & @xmath115 & @xmath116 & @xmath117 + 7130.59772753 & @xmath118 & @xmath119 & @xmath120 & @xmath121 + 7131.59975985 & @xmath122 & @xmath123 & @xmath124 & @xmath125 + 7134.64244313 & @xmath126 & @xmath127 & @xmath128 & @xmath129 + 7135.60113042 & @xmath130 & @xmath131 & @xmath132 & @xmath133 + 7141.6102368 & @xmath134 & @xmath135 & @xmath136 & @xmath137 + 7142.60623451 & @xmath138 & @xmath139 & @xmath140 & @xmath141 + 7143.6053809 & @xmath142 & @xmath143 & @xmath144 & @xmath145 + 7144.60529841 & @xmath146 & @xmath147 & @xmath148 & + 7149.61781551 & @xmath149 & @xmath150 & @xmath151 & @xmath152 + 7150.61051989 & @xmath153 & @xmath154 & @xmath155 & @xmath156 + 7151.61149262 & @xmath157 & @xmath158 & @xmath159 & @xmath160 + 7152.61133458 & @xmath161 & @xmath162 & @xmath163 & @xmath164 + 7153.61769047 & @xmath165 & @xmath166 & @xmath167 & @xmath168 + 7156.61430815 & @xmath169 & @xmath170 & @xmath171 & @xmath172 + 7159.61682615 & @xmath173 & @xmath174 & @xmath175 & @xmath176 + 7160.61690956 & @xmath177 & @xmath178 & @xmath179 & @xmath180 + 7162.61714264 & @xmath181 & @xmath182 & @xmath183 & @xmath184 + 7163.62052311 & @xmath185 & @xmath186 & @xmath187 & @xmath188 + 7165.63733328 & @xmath189 & @xmath190 & @xmath191 & @xmath192 + 7166.61939046 & @xmath193 & @xmath194 & @xmath195 & @xmath196 + 7168.65095798 & @xmath197 & @xmath198 & @xmath199 & @xmath200 + 7169.61788092 & @xmath201 & @xmath202 & @xmath203 & @xmath204 + 7170.61722538 & @xmath205 & @xmath206 & @xmath207 & @xmath208 + 7171.62035657 & @xmath209 & @xmath210 & @xmath211 & @xmath212 + 7172.62633782 & @xmath213 & @xmath214 & @xmath215 & @xmath216 [ tab : lc.spec ]
lccccccc 5100 & 26 & @xmath217 & 1.0 & @xmath218 & 0.009 & 0.090 & @xmath219 + @xmath113 & 24 & @xmath220 & 1.1 & @xmath221 & 0.050 & -2.56 & @xmath222 + @xmath25 & 25 & @xmath217 & 1.0 & @xmath223 & 0.008 & 0.072 & @xmath224 + @xmath26 & 25 & @xmath217 & 1.0 & @xmath225 & 0.007 & 0.052 & @xmath226 + h@xmath0 & 26 & @xmath217 & 1.0 & @xmath227 & 0.002 & 0.105 & @xmath228 + h@xmath24 & 26 & @xmath217 & 1.0 & @xmath229 & 0.007 & 0.135 & @xmath230 + & 25 & @xmath231 & 1.0 & @xmath232 & 0.033 & 0.500 & @xmath233 [ tab : lcstats ] figure [ fig : lc ] displays the final merged and binned continuum light curve and the broad emission - line light curves ( tabulated in table [ tab : lc.spec ] .
the variability statistics for each of the light curves are tabulated in table [ tab : lcstats ] .
column ( 1 ) lists the spectral feature and column ( 2 ) gives the number of measurements in the light curve .
columns ( 3 ) and ( 4 ) list the average and median time separation between measurements , respectively . column ( 5 ) gives the mean flux and standard deviation of the light curve , and column ( 6 ) lists the mean fractional error ( based on the comparison of observations that are closely spaced in time ) .
column ( 7 ) lists the excess variance , computed as : @xmath234 where @xmath235 is the variance of the fluxes , @xmath236 is their mean - square uncertainty , and @xmath237 is the mean flux . and column ( 8) is the ratio of the maximum to the minimum flux in the light curve , @xmath238 .
we employed the interpolated cross - correlation function ( iccf ) methodology @xcite with the modifications of @xcite to search for time delays of the emission lines relative to the continuum .
the iccf method calculates the cross - correlation function ( ccf ) twice , by interpolating first one light curve and then the other , and averages the two results together to determine the final ccf .
the ccf can be characterized by its maximum value ( @xmath239 ) , the time delay at which the maximum occurs ( @xmath240 ) and the centroid ( @xmath241 ) of the points around the peak above some value ( typically @xmath242 ) .
ccfs for each light curve relative to the continuum are displayed in figure [ fig : lc ] ( _ right panels _ ) .
for the continuum light curve , this is the autocorrelation function . to quantify the uncertainties on the time delay measurements , @xmath241 and @xmath240
, we employ the monte carlo `` flux randomization / random subset sampling '' method of @xcite .
this method is able to account for the measurement uncertainties as well as the effect of including or excluding any particular data point .
the `` random subset sampling '' is implemented such that , from the @xmath243 available data points within a light curve , @xmath243 points are selected without regard to whether a point has been previously chosen . for a point that is sampled @xmath244 times , the uncertainty on that point is scaled by a factor of @xmath245 .
the typical number of points that is not selected in any specific realization is @xmath246 .
the `` flux randomization '' component takes the newly sampled light curve and modifies the flux values by a gaussian deviation of the flux uncertainty .
these modified light curves are then cross - correlated with the iccf method described above , and the whole process is repeated many times ( @xmath247 ) . from the large set of realizations , we build distributions of @xmath241 and @xmath240 . the median of each distribution is taken to be the measurement value , and the uncertainties are set such that they mark the upper 15.87% and lower 15.87% of the realizations ( corresponding to @xmath248 for a gaussian distribution ) .
the red histograms in the figure [ fig : lc ] depict the cross - correlation centroid distribution for each emission line . to further check that combining the various photometric and spectroscopic light curves has not affected our measured time delays
, we also determined the time delay of h@xmath0 relative to each of the individual continuum , @xmath31band , and @xmath29band light curves .
each of these light curves is slightly undersampled relative to the combined continuum light curve , but the ccfs and recovered h@xmath0 time delays agree within the measurement uncertainties .
we also investigated the time delays with the javelin package @xcite .
javelin fits the continuum variations with a damped random walk model .
it then assumes a top hat model for the reprocessing function , and determines the best - fit shifting and smoothing parameters for the emission - line light curves by maximizing the likelihood of the model .
uncertainties on each of the model parameters are assessed through a bayesian markov chain monte carlo method .
we denote time delays from javelin as @xmath249 .
given the extremely short time delays , we were unable to fit a single model while including all the emission lines simultaneously , so we instead modeled each emission line separately relative to the continuum ( see figure [ fig : jav ] ) .
time delay measurements are listed in table [ tab : lags ] . while each of the measurements is an observed time delay , the rest - frame time delays , corrected for a factor of @xmath250 , are formally the same within the uncertainties .
lccc h@xmath0 & @xmath251 & @xmath252 & @xmath253 + h@xmath24 & @xmath254 & @xmath255 & @xmath256 + & @xmath257 & @xmath258 & @xmath259 [ tab : lags ]
the widths of the broad emission lines in agn spectra are interpreted as the line - of - sight velocities of the bulk motion of the gas .
the narrow emission lines , however , are known to emit from gas that is not participating in the same bulk motion .
therefore , good practice is to isolate the broad emission from the narrow emission when quantifying the line width . in the spectrum of ugc06728 , however , it is not clear what part of the h@xmath0 line is narrow emission ( cf .
figure [ fig : mean.rms ] ) .
furthermore , the narrow lines contribute almost no signal to the rms spectrum , demonstrating that our internal spectral scaling method has minimized their apparent variability from changing observing conditions throughout the monitoring campaign . as it is the variable part of the emission line ( the rms profile ) that we are most interested in , we do not attempt any narrow line subtraction for this object .
we measured the widths of the broad h@xmath0 , @xmath260 , and h@xmath24 emission lines in both the mean and the rms spectra and we report two different line width characterizations : the full width at half the maximum flux ( fwhm ) and the second moment of the line profile ( @xmath261 ) .
line widths were measured directly from the spectra , with each line profile defined as the flux above a local linear continuum .
uncertainties in the emission line widths were determined using a monte carlo random subset sampling method . from a set of @xmath243 spectra ,
a subset of @xmath243 spectra were selected without regard to whether they had been previously chosen .
the mean and rms of the subset were created , from which the fwhm and @xmath261 of an emission line were determined and recorded .
distributions of line width measurements were built up over 1000 realizations .
we take the mean and the standard deviation of each distribution as the measurement and its uncertainty , respectively .
following @xcite , we corrected the emission - line widths for the dispersion of the spectrograph .
the observed emission line width , @xmath262 , can be described as @xmath263 where @xmath264 is the intrinsic line width and @xmath265 is the broadening induced by the spectrograph .
the employment of a wide spectrograph slit for reverberation campaigns means that the spectrograph dispersion can not be determined from night sky emission lines or from arc lamp lines the unresolved agn point source , even under poor seeing conditions , will not fill the spectrograph slit .
given the relative obscurity of this particular agn , we were unable to estimate @xmath266 , and therefore constrain @xmath265 , from high - quality , high - resolution observations of the narrow emission lines in the literature .
however , we have previously monitored other agns with this same instrumental setup , and so we adopt the value of @xmath267 that we determined for ngc5273 from a spring 2014 monitoring campaign @xcite .
our final resolution - corrected line width measurements are listed in table [ tab : linewidths ] .
lcccc h@xmath0 & @xmath268 & @xmath269 & @xmath270 & @xmath271 + h@xmath24 & @xmath272 & @xmath273 & @xmath274 & @xmath275 + & @xmath276 & @xmath277 & @xmath278 & @xmath279 [ tab : linewidths ]
all of the time delays measured for ugc06728 are very short , which is to be expected given the low luminosity of the agn .
the time delays determined for h@xmath0 are the only ones that are not formally consistent with zero within the measurement uncertainties , so h@xmath0 is the only emission line we will consider for the determination of the black hole mass .
however , h@xmath0 is also the emission line for which we have the largest number of reverberation results ( cf .
@xcite for a recent summary ) , so it is also the most reliable emission line for determining @xmath280 .
the black hole mass is generally determined from reverberation - mapping measurements as : @xmath281 where @xmath282 is the time delay for a specific emission line relative to variations in the continuum , and @xmath113 is the line - of - sight velocity width of the emission line , with @xmath283 and @xmath284 being the speed of light and gravitational constants , respectively .
the emission - line time delay is interpreted as a measure of the responsivity - weighted average radius of the broad - line region for that specific emission feature ( e.g. , h@xmath0 ) .
the scaling factor @xmath285 accounts for the detailed geometry and kinematics of the broad line region gas , which is unresolvable . in practice ,
the multiplicative factor , @xmath286 , which is found to bring the @xmath7 relationship for agns with reverberation masses into agreement with the @xmath7 relationship for nearby galaxies with dynamical black hole masses ( e.g. , @xcite ) is used as a proxy for @xmath285 . in this way , the population average factor provides an overall scale for reverberation masses that should be unbiased , but the mass of any particular agn is expected to be uncertain by a factor of 2 - 3 because of object - to - object variations .
the value of @xmath287 has varied in the literature from 5.5 @xcite to 2.8 @xcite , depending on which objects are included and the specifics of the measurements .
we adopt the value determined by @xcite of @xmath288 . combining the time lag ( @xmath241 ) and line width ( @xmath261 ) measurements for h@xmath0 and
scaling by @xmath286 , we determine @xmath289m@xmath3 .
the extremely rapid response of the broad emission lines to variations in the continuum flux in ugc06728 means that our daily sampling was not fine enough to resolve time delays for all the broad optical recombination lines .
the time delay of h@xmath0 is the only one that is not formally consistent with zero delay , and it is only marginally resolved at that .
however , while we were not able to resolve the time delays for h@xmath24 and , we can examine them in light of the expected virial relationship for blr gas that is under the gravitational dominance of the black hole .
in particular , we would expect that @xmath290 .
this relationship has been shown to be a good description of observations when reverberation results from multiple emission lines have been recovered ( e.g. , @xcite ) .
figure [ fig : rv ] shows the measurements for the optical recombination lines in ugc06728 , with the expected relationship scaled to match the measurements for h@xmath0 .
there is generally good agreement with the expected relationship within the measurement uncertainties , such that we would not expect to resolve the responses of these emission lines with our current sampling .
a monitoring campaign with finer temporal resolution ( @xmath291days ) would be needed to further improve upon these constraints .
furthermore , we can examine the location of ugc06728 on the agn @xmath6 relationship to further assess the h@xmath0 time delay measurement .
for very nearby galaxies like ugc06728 , however , one complication is the large fraction of host - galaxy starlight that contributes to the continuum emission at rest - frame 5100 through the large spectroscopic slit ( @xmath292 ) employed in a reverberation mapping campaign .
the usual method to correct for this contamination is to carry out two - dimensional surface brightness modeling of a high - resolution image of the galaxy ( usually from the _ hubble space telescope _ to maximize the image quality ) , thereby isolating the host - galaxy starlight components from the unresolved agn point source . using the modeling results to create an
`` agn - free '' image allows the starlight contribution to be directly constrained @xcite .
unfortunately , there are no _ hst _ images of ugc06728 .
the highest resolution optical images available are the apo dis @xmath29band images discussed above , with a pixel scale of @xmath293/pixel . while hardly comparable to the quality afforded by _ hst _
, the dis images do allow us to place some rough constraints on the starlight contribution to the flux density at @xmath111 .
we aligned and stacked several of the @xmath29band images to increase the signal - to - noise in the combined image . using the two - dimensional surface brightness fitting program galfit @xcite
, we created a model of the point spread function ( psf ) of the stacked image by fitting multiple gaussian components to the profile of a field star in a restricted portion of the image .
we then employed this model psf while fitting the full frame , including a background sky gradient , a psf for the agn and the nearby star , an exponential profile for the disk of the galaxy , and a srsic profile for the bulge .
the bulge profile , in particular , is very compact with a half - light radius of 1.7pix ( 0.7 ) , and likely degenerate with the agn psf , so we caution that our estimate of the starlight contribution is probably more like a lower limit .
figure [ fig : galfit ] displays a 25@xmath2825 region of the stacked @xmath29band image , our best - fit model from galfit , and the residuals after subtracting the model from the image .
as described earlier , calibrated @xmath29band photometry for three field stars from apass ( the aavso photometric all sky survey ; @xcite ) was used to set the overall flux scale of the image .
we also account for a slight flux scaling factor , due to the difference in effective wavelength of the @xmath25 filter compared to @xmath111 , using synphot and a template galaxy bulge spectrum @xcite .
our estimate of the host - galaxy contribution to the spectroscopic flux density is @xmath294ergss@xmath5@xmath21@xmath5 . removing this contribution results in an agn - only continuum flux density of @xmath295ergss@xmath5@xmath21@xmath5 . assuming a luminosity distance of @xmath296mpc and correcting for galactic absorption along the line of sight @xcite
, we derive @xmath297ergss@xmath5 .
figure [ fig : rl ] displays the @xmath6 relationship for nearby agns based on reverberation mapping of h@xmath0 @xcite .
the filled circle shows the location of ugc06728 with the h@xmath0 time delay we have derived here and the luminosity after correction for the estimated starlight contribution .
the agreement between ugc06728 and its expected location based on its estimated luminosity is extremely good considering the barely - resolved nature of the time delay and the caveats in the luminosity determination .
furthermore , we can expect that the agreement is actually somewhat better than depicted , given the likelihood that the starlight correction to the luminosity is underestimated as described above .
taking our galaxy decomposition at face value , we can estimate the bulge - to - total ratio as @xmath298 , which suggests that the hubble type of the galaxy is @xmath299sb @xcite .
we also estimate the color of the galaxy as @xmath300 , which suggests @xmath301 @xcite .
the total stellar mass of the galaxy is @xmath302m@xmath3 , which also agrees with the host - galaxy being sb@xmath15sc in type . to further explore the reverberation results for ugc06728 within the context of the larger reverberation sample , we obtained supplemental observations on 2016 may 13 with the dis spectrograph on the apo 3.5-m telescope with the intent of constraining the bulge stellar velocity dispersion .
the high resolution b1200 and r1200 gratings were employed , providing nominal dispersions of 0.62 / pix and 0.58 / pix and wavelength coverages of 1240 and 1160 , respectively .
the blue grating was centered at 4900 to target the mg@xmath303 stellar absorption signature , and the red grating was centered at 8500 for the triplet absorption .
the @xmath304 slit was rotated to a position angle of 150 east of north , approximately along the minor axis of the galaxy .
given the high inclination of the galaxy , we specifically avoided the major axis of the galaxy to mitigate the effects of rotational broadening from the disk within the one - dimensional extracted spectra .
two 1200s exposures were obtained through patchy clouds and with marginal seeing at an airmass of 1.6 .
spectra of the standard star , feige34 , were also obtained to assist with the flux calibration , as well as spectra of hd125560 ( spectral type k3iii ) and hd117876 ( spectral type g8iii ) to provide velocity templates with the same wavelength coverage and dispersion as the galaxy .
all spectra were reduced with iraf following standard procedures .
an extraction width of 40pixels ( corresponding to 16 on the blue camera and 16.8 on the red camera ) was adopted to maximize the galaxy signal in the resultant spectra .
following flux calibration of the spectra , we employed the ppxf ( penalized pixel fitting ) method of @xcite to extract the stellar kinematics .
the mg@xmath303 absorption signature was not detected in the galaxy spectra , but the triplet features were detected , so we focused on fitting the red spectra only . during the fitting process , we restricted the wavelength region to 8525@xmath158850 and determined the best - fit parameters ( velocity , velocity dispersion , h3 , and h4 ) using first one velocity template star and then the other .
the best fits to the spectrum of ugc06728 are displayed in figure [ fig : veldisp ] : hd125560 ( red line ) provided a best - fit velocity dispersion of 56.5kms@xmath5 , and hd117876 ( blue line ) provided a best fit of 46.7kms@xmath5 .
we take the average of these as the bulge stellar velocity dispersion , @xmath4kms@xmath5 . with this constraint on the bulge stellar velocity dispersion in ugc06728
, we can explore its location on the agn @xmath7 relationship .
figure [ fig : msig ] displays the agn @xmath7 relationship from @xcite ( open points and line ) , with the location of ugc06728 shown by the filled circle .
the scatter at the low - mass end of the @xmath7 relationship for agns with reverberation masses seems to be much smaller than that found for megamaser host galaxies @xcite .
@xcite also found the megamaser host galaxies to have a high scatter relative to the @xmath305 and @xmath306 relationships .
each sample of direct black hole masses , whether dynamical , reverberation , or masering , has its own set of biases and assumptions that are independent of the other techniques , so further exploration into this apparent disagreement is likely to shed light on the reliability of black hole mass measurements as they are currently applied .
furthermore , we can estimate the black hole sphere of influence ( @xmath307 ) in the nucleus of ugc06728 .
generally defined as @xmath308 @xmath307 is often employed as a convenient metric for determining the probability of success for constraining @xmath280 from spatially resolved stellar dynamics .
@xcite argue that a strict reliance on resolving @xmath307 is not necessary , however , for useful constraints on black hole masses .
combining our measurements of @xmath280 and @xmath309 and again assuming a luminosity distance of @xmath310mpc , we estimate @xmath311 for ugc06728 .
while this angular size is smaller than the achievable spatial resolution of integral field spectrographs on the largest ground - based telescopes today , it is interesting to note that it is not much smaller than @xmath307 for ngc3227 . @xcite
were able to constrain the black hole mass of ngc3227 through stellar dynamical modeling , even though the reverberation mass and bulge stellar velocity dispersion predict @xmath312 .
given the very limited number of agns where it will be possible to carry out a direct comparison of reverberation - based and stellar dynamical - based black hole mass measurements with current and near - future technology , ugc06728 could potentially be a worthwhile target for dynamical modeling .
@xcite analyzed _ suzaku _ observations of ugc06728 and determined that it was a `` bare '' agn , with minimal intrinsic absorption .
fitting the x - ray spectrum with a relativistic reflection model , and assuming an accretion disk inclination of @xmath313 , they determined a dimensionless spin parameter of @xmath314 , indicating the black hole is spinning rapidly .
combined with our mass contraint of @xmath2m@xmath3 , ugc06728 is one of a small number of massive black holes that are completely characterized . a few other low - mass black holes have both mass and spin constraints , and they appear to agree with the properties derived for ugc06728 .
mcg-06 - 30 - 15 is only slightly more massive with @xmath315m@xmath3 @xcite and is spinning near maximally ( @xmath316 ; @xcite ) .
ngc4051 is another example , with @xmath317m@xmath3 @xcite and @xmath318 @xcite .
black hole evolutionary models have only recently begun to treat black hole spin in addition to mass . depending on the model , it is not clear if the properties of the black hole in ugc06728 are expected or surprising .
for example , the model of @xcite predicts that black holes with @xmath319m@xmath3 in gas - rich galaxies at @xmath320 ( including agns ) should have slowly rotating black holes with dimensionless spin parameters of @xmath321 .
this model is based on many observational constraints , including the @xmath7 relationship , with which we have shown ugc06728 to be in agreement .
one caveat to the evolutionary model of @xcite is that it does not account for black hole feeding through disk instabilities , which could be a reason for the apparent discrepancy here .
disk instability accretion events would likely be correlated and serve to spin up a black hole .
the evolutionary models of @xcite attempt to include this effect by linking the gas dynamics of the extended galaxy to the central black hole .
their models predict that local black holes with @xmath322m@xmath3 should tend to be spinning near maximally , and that _ accreting _ black holes in spiral galaxies should also tend to have near - maximal spins .
interpretation of black hole spin measurements is still somewhat debated as well .
@xcite argue that black hole spins tend to be overestimated in many cases , although they state this is likely not the case for the most maximally spinning black holes ( @xmath323 ) . furthermore , there is a very strong selection bias inherent in the sample of agns with spin measurements .
rapidly spinning black holes have significant boosts to their x - ray flux through increased radiative efficiency , and the current sample of agns with spin constraints is based on observations of the brightest x - ray sources , so the current sample will strongly favor rapidly spinning black holes @xcite . in any case , ugc06728 is an important addition to the sample . as the least massive central black hole that has been fully described
, it will help to anchor future studies , both observational and theoretical , of central black hole demographics .
we present an h@xmath0 time delay and a reveberation - based black hole mass for the nearby , low - luminosity seyfert ugc06728 . with @xmath324days and @xmath289m@xmath3 ,
ugc06728 is at the low end of observed properties within the reverberation mapping sample .
the time delay and estimated agn luminosity agree with the @xmath6 relationship for other reveberation - mapped agns , and a measurement of @xmath325kms@xmath5 from long - slit spectroscopy shows that the black hole mass agrees with the agn @xmath7 relationship . with @xmath326m@xmath3 , ugc06728 is currently the lowest - mass central black hole that is fully described by both direct mass and spin constraints .
we thank the referee for thoughtful comments that improved the presentation of this paper .
mcb gratefully acknowledges support from the nsf through career grant ast-1253702 .
this research is based on observations obtained with the apache point observatory 3.5-meter telescope , which is owned and operated by the astrophysical research consortium .
we heartily thank the staff at apo for all their help with this program .
this research has made use of the aavso photometric all - sky survey ( apass ) , funded by the robert martin ayers sciences fund .
this research has made use of the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration and the simbad database , operated at cds , strasbourg , france . | we present the results of a recent reverberation mapping campaign for ugc06728 , a nearby low - luminosity seyfert 1 in a late - type galaxy . nightly monitoring in the spring of 2015 allowed us to determine an h@xmath0 time delay of @xmath1days .
combined with the width of the variable h@xmath0 line profile , we determine a black hole mass of @xmath2m@xmath3 .
we also constrain the bulge stellar velocity dispersion from higher - resolution long slit spectroscopy along the galaxy minor axis and find @xmath4kms@xmath5 .
the measurements presented here are in good agreement with both the @xmath6 relationship and the @xmath7 relationship for agns .
combined with a previously published spin measurement , our mass determination for ugc06728 makes it the lowest - mass black hole that has been fully characterized , and thus an important object to help anchor the low - mass end of black hole evolutionary models . |
we denote @xmath1 by @xmath8 $ ] and the space of permutations of @xmath8 $ ] by @xmath9}$ ] .
we define the permuted walk on the interval @xmath8 $ ] as follows : given @xmath10}$ ] , the permuted walk @xmath2 is defined as the markov chain on @xmath8 $ ] which goes from @xmath11 to @xmath12 and @xmath13 with probability @xmath14 each , if @xmath15 .
the markov chain jumps from @xmath16 to @xmath17 or @xmath18 equally likely and similarly from @xmath19 it jumps to @xmath20 or @xmath21 equally likely .
we suppress the dependence on @xmath19 of @xmath2 for notational simplicity .
the permuted walk is obtained by composing one step of the random walk on the graph in fig.[f.loop ] with a permutation of the interval . in this paper
we look at the hitting times of the permuted random walk when the permutation is drawn uniformly from the set of all permutations . for any @xmath22
$ ] we denote by @xmath23 the hitting time of @xmath24 . for any @xmath10}$ ] , denote by @xmath25 , the expectation of @xmath23 , starting @xmath2 from @xmath11 .
we define the worst case hitting time of the walk @xmath26 as @xmath27}\mathbb e^{\sigma}_{x}(\tau_y).\ ] ] clearly @xmath28 is a function of @xmath29 when @xmath30 it is obvious that @xmath31 however in our main result we show that this typically is not the case . [ thm : mainresult]there exists a universal constant @xmath32 such that @xmath33 where @xmath34 is the uniform measure on @xmath9}$ ] .
it turns out that this problem is a special example of hitting time of the random walk on a directed expander .
random walks on undirected expanders are well understood .
we state and prove their counterparts in the directed graph regime in sections [ de ] , [ ub ] and [ lb ] . for more details on undirected expanders
see e.g. @xcite , @xcite and the references therein .
+ the structure of this article is as follows : we look at the directed graph induced by the permuted walk in section [ eg ] .
we prove expansion properties of this directed graph in section [ de ] .
we study upper bound of the hitting time in section [ ub ] and prove theorem [ thm : mainresult ] there .
we discuss lower bounds for hitting time in section [ lb ] .
in this section we look at the permuted random walk as a walk on a directed graph .
recall that a directed graph is said to be eulerian if for every vertex @xmath35 , @xmath36 given @xmath0 we define @xmath37 as a directed graph with vertex set the integers @xmath8 $ ] and edge set @xmath38 here the edge @xmath39 is the directed edge going from @xmath40 to @xmath41 . since @xmath0 is a bijection it is clear from the definition that @xmath37 is a regular eulerian digraph with @xmath42.\ ] ]
clearly @xmath2 is the simple random walk on @xmath43 .
regular eulerian . ]
we first discuss the connectivity properties of the graphs @xmath43 . a directed graph @xmath44 is said to be connected if the graph @xmath45 obtained by removing the orientations of the edges is connected .
a directed graph is said to be strongly connected if for every pair of vertices @xmath46 there is an oriented path from @xmath11 to @xmath24 .
in @xcite the authors prove that for all @xmath0 the random walk on @xmath43 is irreducible or in other words @xmath43 is strongly connected .
we include the proof for completeness .
we state a standard result about directed graphs .
[ lemma : strongconnected]let @xmath44 be a directed graph such that indegree of every vertex equals outdegree . then @xmath44 is connected implies that it is strongly connected . for the proof see the discussion following theorem @xmath47 in @xcite . using the lemma
one can show that @xmath43 is strongly connected in the following way .
[ lemma : irreducible]the random walk chain on @xmath43 denoted by @xmath2 is irreducible for all @xmath0 . by definition to prove the lemma we show that @xmath43 is strongly connected . to show this , using lemma
[ lemma : strongconnected ] it suffices to prove that @xmath43 is connected .
it is clear from the definition of @xmath43 that for all @xmath48 , both @xmath49 and @xmath50 are connected to @xmath51 and hence to each other
. thus all the numbers of the same parity are connected to each other .
also the boundary point @xmath19 is connected to @xmath52 through @xmath21 because of the self loop at @xmath19 . since @xmath19 and @xmath52
are of opposite parity we see that @xmath43 is connected . for technical reasons
we also introduce the following graph . given a directed graph @xmath44 on the vertex set @xmath53 , @xmath54 is a directed multigraph on @xmath53 .
the edge set of @xmath54 is constructed as follows : for @xmath55 we put a distinct directed edge @xmath56 for every distinct directed path of length @xmath57 from @xmath11 to @xmath24 in @xmath44 .
we note that since @xmath43 is a directed @xmath58regular graph , @xmath59 is a directed @xmath60regular graph . the arguments in lemma [ lemma : strongconnected ] also show that @xmath61 is strongly connected . in the proof of one of the subsequent results it will turn out to be more convenient to work with @xmath61 .
we quote a quadratic bound for the hitting time @xmath62 from @xcite .
for @xmath30 it is a standard result that @xmath63 where @xmath64 is the hitting time of the set @xmath65 in @xcite the authors show that for any @xmath0 , @xmath66 whether the identity permutation is the slowest , i.e. whether for all @xmath0 , the inequality @xmath67 holds appears as an open problem in @xcite .
we focus on typical permutations .
it turns out that they are considerably faster and most permuted random walks have a linear hitting time .
in this section we discuss some geometric properties of certain markov chains and the effect they have on hitting times .
see chapter @xmath68 from @xcite for details .
let @xmath69 be the transition kernel for an irreducible markov chain on state space @xmath70 with invariant measure @xmath71 .
the edge measure @xmath72 is defined by @xmath73 @xmath74 thus @xmath75 is the probability of moving from @xmath76 to @xmath77 in one step when starting from stationarity .
the bottleneck ratio of a set @xmath78 is defined to be @xmath79 where @xmath80 .
the bottleneck ratio of the whole chain is defined as @xmath81 for a directed eulerian graph @xmath82 and the simple random walk on @xmath44 , the bottleneck ratio of a set @xmath83 is @xmath84 where @xmath85 is the degree of vertex @xmath35 and @xmath86 is the number of edges going from @xmath78 to @xmath87 a family @xmath88 of digraphs is said to be a @xmath89 directed expander family , if the following conditions hold .
* @xmath90 . *
both indegree and outdegree of every vertex in @xmath91 is @xmath92 for all @xmath93 * the bottleneck ratio of the simple random walk on @xmath88 satisfies @xmath94 for all @xmath19 .
* from now on all our graphs will be digraphs*. if @xmath88 is @xmath95regular , then the stationary measure of the random walk is uniform .
hence to show that a family of regular graphs @xmath96 with vertex set @xmath97 is a directed expander family , it just suffices to show that there exists a constant @xmath98 independent of @xmath19 such that given any set @xmath99 of size at most @xmath100 , the number of edges going from @xmath101 to @xmath102 is at least @xmath103 .
first we introduce some definitions .
let @xmath82 be a digraph .
let @xmath104 and @xmath105 the set @xmath106 is called the set of neighbors and the set @xmath107 is called the boundary of the set @xmath76 in @xmath44 .
[ exbou ] for the chain @xmath2 and any @xmath108 $ ] , @xmath109 follows from @xmath110 and the following two facts : 1 .
@xmath43 is @xmath58regular .
the number of edges going from @xmath76 to @xmath111 is at least @xmath112 and at most @xmath113 [ thm : linearhitting]if @xmath88 is a @xmath89 expander family with @xmath114 , then there exists a uniform constant @xmath115 depending only on the bottleneck ratio @xmath116 , such that @xmath117 , @xmath118 , @xmath119 where @xmath120 is the hitting time of @xmath24 for the random walk started from @xmath11 . before proving the above theorem we state a standard fact about the mixing time of the random walk on expanders .
given an ergodic markov chain with state space @xmath70 , transition kernel @xmath69 and stationary measure @xmath71 , the mixing time of the chain is defined as @xmath121 where @xmath122 is the total variation norm and @xmath123 is the probability distribution on @xmath70 after running the markov chain starting from @xmath11 for @xmath124 steps .
recall that the lazy random walk on a graph is the markov chain which at every time step stays at the same place with probability @xmath14 and takes a step to a uniform random neighbor with probability @xmath14 .
+ we denote the lazy random walk on any regular directed graph @xmath44 by @xmath125 and on @xmath43 by @xmath126 we state a standard result about mixing of the lazy random walk in @xmath127 norm on expanders .
see e.g. section @xmath128 @xmath129 in @xcite .
[ thm : supmixingexpander ] if @xmath88 is a @xmath89 expander family with @xmath114 , then for all @xmath130 @xmath131 here @xmath132 is the kernel of the lazy random walk after @xmath124 steps .
a simple corollary of the above result is that the lazy random walk on an expander family satisfies @xmath133
equipped with the results in the previous section we prove theorem [ thm : linearhitting ] .
we first recall that the green s function for @xmath125 , the lazy random walk on a graph @xmath44 , is the expected amount of time the walk spends at each vertex .
we define it formally below but suppress the dependence on @xmath44 .
let @xmath134 be the number of visits to @xmath24 in the first @xmath19 steps . here
@xmath135 denotes the position of @xmath125 at time @xmath124 .
define , @xmath136 to be the expected number of visits to @xmath24 starting from @xmath11 in the first @xmath19 steps . hence , @xmath137 where @xmath138 is the kernel of @xmath139 .
+ _ proof of theorem _ [ thm : linearhitting ] .
we first prove that for all @xmath11 , @xmath24 @xmath140 , @xmath141 where the constants involved do not depend on @xmath19 .
now by theorem [ thm : supmixingexpander ] we know that @xmath142 for all @xmath124 .
the fact that @xmath143 is upper bounded by a constant follows trivially from the above upper bound , since , @xmath144 we now show a constant lower bound . we note from the above lower bound that for some @xmath145 large enough just depending on @xmath116 , and @xmath146 @xmath147 hence , @xmath148 therefore for large enough @xmath19 , @xmath149 hence , @xmath150 now we will show that there exists some constant @xmath151 depending only on @xmath116 such that for all @xmath152 $ ] we have @xmath153 where @xmath154 is the hitting time of @xmath24 for the lazy chain @xmath125 . to show this we will bound the second moment of @xmath155 and use the second moment method . expanding @xmath156 using the expression in ( [ z ] )
we get , @xmath157 hence taking expectation , @xmath158 from ( [ upbg ] ) both @xmath143 and @xmath159 are @xmath160 therefore , @xmath161 also trivially from ( [ upbg ] ) , @xmath162 we know by the second moment method that , @xmath163 now from ( [ upbg ] ) and ( [ smm ] ) we know that both the numerator and the denominator are @xmath164 and hence we get that @xmath165 where @xmath166 does not depend on the points @xmath11 , @xmath24 or @xmath19 .
this means that starting from any point @xmath167 , the lazy walk @xmath125 hits @xmath24 in the next @xmath19 steps with chance at least @xmath168 .
thus @xmath169 is dominated by a geometric random variable with success probability at least @xmath168 .
hence @xmath170 now for all @xmath171 the hitting times of the simple random walk and the lazy random walk satisfies @xmath172 hence we are done by taking @xmath32 in the statement of the theorem to be @xmath173 .
@xmath174 the next technical result roughly states that given an uniformly chosen permutation @xmath10}$ ] , with high probability the graphs @xmath37 form an expander family .
this with theorem [ thm : linearhitting ] will then directly imply theorem [ thm : mainresult ] .
[ thm : unionbound ] there exists a universal positive constant @xmath175 independent of @xmath19 , such that , @xmath176 as @xmath177 first we state the following lemma which compares the sizes of the boundary of a subset in the graphs @xmath43 and @xmath178 [ comp12 ] for any @xmath179 $ ] , @xmath180 clearly @xmath181 since @xmath43 is @xmath182 @xmath183 thus @xmath180 [ proofidea1 ] before providing formally the proof of theorem [ thm : unionbound ] we discuss the basic idea .
+ to show that @xmath43 is an expander we use the following observation : if @xmath184 $ ] has a large number of connected components in @xmath185 then it has large expansion in @xmath43 .
otherwise , if @xmath76 has a small number of components in @xmath185 we show that for a uniformly chosen @xmath186}$ ] the probability that the set of neighbors of @xmath76 in @xmath43 also has a small number of connected components in @xmath185 is small .
this then implies that the two step expansion of @xmath76 is likely to be high .
the number of sets @xmath179 $ ] with a small number of connected components in @xmath185 is rather small and hence we apply union bound to argue that with high probability the two step expansion of any subset @xmath179 $ ] is large in @xmath43 . in other words , @xmath61 is an expander .
we then use lemma [ comp12 ] to conclude that @xmath43 is an expander .
we now proceed to the proof of theorem [ thm : unionbound ] . for any set
@xmath184 $ ] , define @xmath187.\ ] ] * proof of theorem [ thm : unionbound ] .
* by lemmas [ exbou ] and [ comp12 ] it suffices to show that there exists a universal @xmath98 such that with @xmath188 probability @xmath189 for all @xmath184 $ ] with @xmath190 @xmath191 fix @xmath179 $ ] with @xmath190 . by definition @xmath192 since @xmath0 is a bijection @xmath193 with equality occurring when @xmath76 does not contain any of the points @xmath194 .
thus @xmath195 we also note that by similar reasoning as above @xmath196 let us look at the quantity @xmath197 we first show that if @xmath76 has a large number of connected components in @xmath185 then @xmath198 is large . to avoid parity issues we look at even and odd components of a set .
formally let , @xmath199 and similarly @xmath200 .
now say @xmath201_{even}}\ ] ] is the decomposition into connected components of even numbers i.e. @xmath11 and @xmath202 are considered to be in the same component .
thus @xmath203 is the number of connected components in @xmath204 and similarly @xmath205 then @xmath206_{odd}\ ] ] where @xmath207_{odd}=[a_i , b_i]_{even}\pm 1 $ ] are disjoint connected intervals of odd numbers in the interval @xmath8 $ ] . now
@xmath207_{odd}$ ] has one more element than @xmath208_{even}$ ] unless it contains one of the points @xmath194 .
note that a component can not contain both @xmath16 and @xmath19 since that implies @xmath209 thus @xmath210 similarly @xmath211 now @xmath212 is the disjoint union of @xmath213 and @xmath214 .
therefore @xmath215 @xmath216 only if @xmath217 therefore if @xmath218 , then @xmath219 ) the number of sets @xmath76 of size @xmath220 such that @xmath216 for @xmath221 is clearly at most @xmath222 the first factor comes from choosing the at most @xmath223 starting points for the components in @xmath204 and @xmath200 .
the second factor comes from dividing @xmath220 points into components ; the number of possibilities is at most the number of ways to put @xmath220 balls in @xmath223 bins .
+ let @xmath224 where @xmath225 since @xmath43 is @xmath58regular it is easy to see that @xmath226 let @xmath77 be the image of the smallest @xmath220 elements of @xmath227 under @xmath29 then the chance that @xmath228 is at most @xmath229 from ( [ number ] ) and the fact that @xmath0 is a uniformly random permutation .
now either of the following two conditions @xmath230 imply that @xmath231 this is because under the first condition @xmath232 where the first inequality follows from ( [ key ] ) and the second inequality is easy to check from the definition of @xmath43 .
the second condition by ( [ key ] ) implies that @xmath233 where the second inequality follows from the fact that @xmath234 hence by the above discussion , ( [ number ] ) , ( [ map ] ) and union bound , @xmath235 , \frac{4}{{\epsilon}}<|a|\le n,|\partial_{{g_{n,\sigma}}^2}a|<\epsilon|a| ) \\
\le \sum_{m=1}^{n}\frac{{2n+1\choose 2\epsilon m}^2 { { m(1 + 2\epsilon)}\choose{m}}^{2}}{{2n+1 \choose m}}.\end{aligned}\ ] ] by the next lemma for a small enough @xmath236 independent of @xmath19 the above sum goes to @xmath237 .
thus with @xmath188probability going to @xmath238 , all sets @xmath76 of size between @xmath239 and @xmath19 have at least @xmath240 edges going out of @xmath76 in @xmath61 .
we also know that any set of size at most @xmath239 has at least one edge going out of it since the graph is connected . choosing a suitable @xmath241 in terms of @xmath236 and using lemma [ comp12 ]
the theorem follows .
@xmath242 such that @xmath243 to approximate the binomial coefficients we use the following upper and lower bounds , @xmath244 these bounds imply that @xmath245 choose @xmath236 small enough so that @xmath246 hence from ( [ plug1 ] ) and the fact that @xmath247 , we see that @xmath248 thus @xmath249 now when @xmath250 we also have from ( [ plug1 ] ) , @xmath251 therefore @xmath252 together with ( [ lem1 ] ) , this proves the lemma .
proof of theorem [ thm : mainresult ] . using theorem [ thm : unionbound ]
we see that with @xmath188probability @xmath189 , @xmath253 has bottleneck ratio lower bounded by some fixed positive constant @xmath241 . hence using theorem [ thm : linearhitting ]
we get that there exists a universal constant @xmath76 depending on @xmath241 such that with @xmath188probability @xmath189 the expected hitting time of @xmath24 starting from @xmath11 for the random walk @xmath2 on @xmath43 is less than @xmath254 for all @xmath255 $ ] .
hence we are done .
in this section we show that for a given pair @xmath152 $ ] , with high probability , the expected hitting time from @xmath11 to @xmath24 is @xmath256 .
we however can not hope to make a stronger statement like : there exists a universal constant @xmath257 such that with high probability , for all @xmath152 $ ] @xmath258 this is because of the well known result , see @xcite , which says that if @xmath0 is a uniformly chosen random permutation then @xmath259 where @xmath260 is a poisson random variable with mean @xmath238 .
this implies that with probability lower bounded away from @xmath237 the graph @xmath43 has self loops .
let @xmath11 be a point depending on @xmath0 such that @xmath261 .
then we see that @xmath262 [ thm : typical ] @xmath263}\mathcal{p}({\mathbb e}^{\sigma}_{x}(\tau_{y})\ge \frac{n}{18})=1-o(1).\ ] ] first we look at the lazy chain @xmath264 and the case when we start from stationarity .
[ lemma : stationarity ] starting with the stationary distribution @xmath71 , for any @xmath265,$ ] @xmath266 where @xmath154 is the hitting time of @xmath24 for the lazy chain @xmath126 @xmath267 we abbreviate @xmath123 , the probability measure on @xmath43 after running @xmath264 starting from @xmath11 for @xmath124 steps by @xmath268 then we have the following corollary .
[ linearfromlog ] there exists a universal constant @xmath269 such that with @xmath188probability @xmath189 , @xmath43 satisfies : for all @xmath270 $ ] and @xmath271 , @xmath272 the corollary says that with high probability @xmath0 would be such that starting with distribution @xmath273 , the hitting time of @xmath24 for the chain @xmath264 is at least @xmath274 with chance at least @xmath275 , uniformly for all @xmath46 and @xmath276 . by theorems [ thm : supmixingexpander ] and [ thm : unionbound ]
we know that there is a universal constant @xmath269 such that with @xmath188probability @xmath189 , for all @xmath277 $ ] @xmath278 for all @xmath279 .
fix @xmath276 .
we have for all @xmath280 $ ] @xmath281 the above follows from the definition of total variation norm . since @xmath282 @xmath283}(\pi(z)-\mu_{x}^{t}(z)){\mathbb p}_{z}({\tilde{\tau}}_{y}\ge n/3)\\ & \le \sum_{\stackrel { z\in[-n , n]}{\pi(z)-\mu_{x}^{t}(z)>0}}(\pi(z)-\mu_{x}^{t}(z))=d_{tv}(\mu_{x}^{t},\pi ) .
\end{aligned}\ ] ] hence from ( [ eqn : lower ] ) , we conclude that @xmath284 now we show that starting from @xmath11 with probability lower bounded away from @xmath237 the chain does not hit @xmath24 in the first @xmath285 steps .
+ let @xmath286 be the minimum length of all directed paths from @xmath11 to @xmath24 in @xmath287 we call @xmath286 as the distance of @xmath24 from @xmath11 in @xmath287 [ notinlog]there exists a universal constant @xmath288 such that given @xmath289 with @xmath188probability @xmath189 for all @xmath152 $ ] with @xmath290 , @xmath291 by theorems [ thm : supmixingexpander ] and [ thm : unionbound ] we know that there exists an @xmath292 such that with @xmath188probability @xmath189 , @xmath293 for all @xmath280 $ ] , where @xmath294 is the kernel of @xmath295 .
now clearly @xmath296 since by hypothesis @xmath290 , we know @xmath297 for all @xmath298 .
hence by ( [ equation : geometricdecay ] ) , @xmath299 for large @xmath19 if @xmath3 is chosen a priori to be large enough just depending on @xmath236 .
the next lemma proves that given @xmath300 $ ] , the distance to @xmath24 from @xmath11 is typically large in @xmath43 .
[ far]for every positive integer @xmath3 @xmath263}\mathcal{p}(d_{\sigma}(x , y)>k)=1-o(1)\ ] ] as @xmath177 given @xmath301 $ ] , for any positive integer @xmath302 we denote by @xmath303 the set of all points @xmath304 $ ] such that there is a directed path of length at most @xmath302 from @xmath305 to @xmath167 in @xmath43 .
fix a positive integer @xmath3 . also fix @xmath152.$ ]
we show that @xmath24 does not lie in @xmath306 with @xmath188probability @xmath189 .
now @xmath0 is a uniformly chosen permutation .
we choose @xmath0 by exposing the points in @xmath307 for @xmath308 .
then we look at the chance that @xmath309 for @xmath310 .
we do this by recursion .
suppose we have exposed all points in @xmath311 and @xmath312 .
now @xmath313 is the image of the set @xmath314 under @xmath0 .
we look at the probability that the image set does not contain @xmath24 .
now clearly since the graph is @xmath58 regular @xmath315 thus there are at most @xmath316 points to be exposed and they can take values in the set @xmath8\setminus \mathbb{b}_{l}(x ) $ ] which has cardinality at least @xmath317 . so the chance that @xmath318 given @xmath309 is at least , @xmath319 clearly for a fixed @xmath320 @xmath321 hence the probability that @xmath322 is lower bounded by @xmath323 thus we are done .
_ proof of theorem [ thm : typical]_. choose @xmath324 where @xmath116 appears in lemma [ notinlog ] and @xmath325 appears in the statement of corollary [ linearfromlog ] .
first we observe that @xmath326 this follows from the fact : @xmath327 and the simple relation @xmath328 by the choice of @xmath116 and corollary [ linearfromlog ] we get that with @xmath188 probability @xmath189 @xmath329 now by lemmas [ notinlog ] and [ far ] we get that with @xmath188 probability @xmath189 @xmath330 thus @xmath331 hence with @xmath188probability @xmath189 @xmath332 which then implies that @xmath333 the proof of theorem [ thm : typical ] immediately generalizes to show that given a point @xmath48 the hitting time of @xmath6 starting from @xmath11 on @xmath43 is @xmath256 with @xmath188probability going to @xmath238 as @xmath19 goes to infinity .
we omit the details .
we thank perla sousi for careful reading of the manuscript and making helpful comments .
we also thank christopher hoffman for useful discussions .
part of this work was done while the first author was visiting the theory group at microsoft research redmond . | given a permutation @xmath0 of the integers @xmath1 we consider the markov chain @xmath2 , which jumps from @xmath3 to @xmath4 equally likely if @xmath5 .
we prove that the expected hitting time of @xmath6 starting from any point is @xmath7 with high probability when @xmath0 is a uniformly chosen permutation .
we prove this by showing that with high probability , the digraph of allowed transitions is an eulerian expander ; we then utilize general estimates of hitting times in directed eulerian expanders . |
magnetars are strongly magnetized neutron stars powered with a super - strong magnetic field ( see e.g. @xcite for a review ) . there are two classes of objects which properties could be described in terms of the magnetar model : anomalous x - ray pulsars ( axps ) and soft gamma repeaters ( sgrs ) . it is generally assumed that such sources spin down by magnetic dipole radiation .
the inferred values of the surface magnetic field strength @xmath2 g are larger than the so - called _ critical _ magnetic field
@xmath3 g where the effects of quantum electrodynamics should be taken into account .
this makes magnetars to be the unique natural laboratories to test such superstrong magnetic fields .
however , no direct measurements of magnetic fields in magnetars have been presented so far .
it was suggested in @xcite to search for proton cyclotron harmonic , which should appear in the x - ray band ( 2 - 10 kev ) due to the presence of the strong magnetic field : @xmath4 here @xmath5 denotes the _ observed _ energy of the cyclotron feature , @xmath6 is the gravitational redshift ( @xmath7 for a typical neutron star ) , @xmath8 is the magnetic field strength .
previous search revealed some features during flares and high states , which could be treated as cyclotron features , in the spectra of 7 among 19 magnetar candidates ( see table [ tab1 ] ) . summarizing the results of table [ tab1 ] , we note that the most of the absorption features were mainly observed during flaring stages and with moderate - resolution instruments such as rxte / pca .
it was also announced about the absence of the absorption features in the _ quiescent _ spectra of axp 4u 0142 + 614 , @xcite , although they were predicted in earlier theoretical works @xcite .
the aim of this paper is to provide more extended search for cyclotron features in the quiescent spectra of magnetars .
for such an analysis , a combination of good energy resolution and large effective area is necessary .
therefore , we concentrate on epic cameras onboard _ xmm - newton _ satellite , extending the earlier results of @xcite , where the same instruments were used .
in this work , we used all publically available xmm - newton / epic observations with all three cameras ( e.g. mos1 , mos2 & pn ) in the _ imaging _ mode .
it was done to ensure the better statistics and to prevent possible calibration uncertainties which may occur in _ timing _ mode .
all event lists were cleaned from soft proton solar flares using the standard routine http://www.sr.bham.ac.uk/xmm2/xmmlight_clean.csh v. 3.3 .
source and background regions were substracted manually from the image .
pile - up checking was carried out with the ` epatplot ` procedure .
if the predicted ratios of the single - to - double events were not equal to 1 within the ` epatplot ` error interval , such observations were not included in the subsequent analysis .
after taking all reductions 41 high - quality observations for 12 objects remained .
p28mmp13mmp28mmp23mmp39mmp14 mm object & energy , kev & significance , method & instrument & notes & references + 1e 2259 + 586 & 5 , 10 & & ginga / lac & during flux increase & @xcite + sgr 1806 - 20 & 5.0 , 7.5 , 11.2 , 17.5 & 3.3@xmath9 , ` gauss ` , f - test ( for a set of features ) & rxte / pca & in the harder part of a precursor & @xcite , @xcite@xmath10 + 4u 0142 + 614 & 4 , 8 , 14 & & rxte / pca & emissions , in the most energetic among a sequence of bursts & @xcite + 1e 1048 - 5937 & 14 & 3.9@xmath9 , ` gauss ` , mc & rxte / pca & emission , in a burst & @xcite + & 13 & 3.3@xmath9 , ` gauss ` , mc & rxte / pca & emission , at one part of a bursts tail & @xcite + xte j1810 - 197@xmath11 & 12.6 & 4.5@xmath9 , ` gauss ` , mc & rxte / pca & emission in a burst tail & @xcite + 1rxs j1708 - 4009 & 8.1 & 2.95@xmath9 , ` cyclabs ` , mc & _ _ beppo__sax & the longest observation ( 200ks ) , during rising phase & @xcite , @xcite , @xcite + sgr 1900 + 14 & 6.4@xmath12 & 3.7@xmath9 , ` gauss ` , f - test & rxte / pca & during precursor to the main burst & @xcite + + + + [ tab1 ] to describe the continuum , we chose ` blackbody + powerlaw ` model modified [ mandr ] by ` xspec ` photoelectric absorption model ` phabs ` .
after the fitting procedure , we moved through the spectrum , adding cyclotron absorption with cyclotron absorption model ` cyclabs ` with a step of the central energy of @xmath13 ev and looked for the @xmath14 increase when setting the depths of the harmonics to zero .
this robust analysis revealed three spectra from three different objects to have possible cyclotron absorptions . to estimate the significances of absorption features
, we followed the procedure described in @xcite , by running monte carlo simulations of the spectra with the help of ` xspec ` procedure ` fakeit ` and calculating the percentage of the spectra which are fitted without ` cyclabs ` component .
the results are summarized in the table [ tab2 ] .
the spectra of the magnetar candidate with the highest cyclotron feature significance detected are presented in figures [ fig1 ] and [ fig2 ] . using this procedure
, we found the significances of all absorption features to be below the margin @xmath15 detection limit .
[ tab2 ] .the results of our analysis . in spite of the significant improvement of the @xmath14 value
, monte carlo simulations did not reveal significant features . [
cols="<,^,^,^,^",options="header " , ]
though initial theoretical works @xcite have predicted a proton cyclotron feature of equivalent width @xmath16 , in subsequent paper @xcite it has been shown that vacuum polarization effects strongly suppress the cyclotron absorption feature , giving the equivalent width about an order of magnitude lower .
it should also be noted that the theoretical calculations of the equivalent width of the cyclotron feature were done for a local patch of the neutron star surface .
phase averaged spectra , like those we are using here , would include contributions from various magnetic field strengths , directions and effective temperatures , which would further suppress the cyclotron feature .
the results of our analysis indicate the absence of the significant cyclotron absorptions in the quiescent spectra of the magnetar candidates observed with _ xmm - newton _ , what could be interpreted due to combination of the two above - mentioned effects .
m. m. thanks to lontkovskyi d. i. and melnyk g. m. for helpful and fruitful discussions .
all the data analysis has been processed using the computing facilities of virtual roentgen and gamma - ray observatory in ukraine ( http://virgo.org.ua ) .
both authors are grateful to mcgill pulsar group for the inappreciated information placed here : http://www.physics.mcgill.ca/~pulsar/magnetar/main.html .
99 mereghetti s. the a&a rev .
, v. 15 , pp .
225 - 287 ( 2008 ) woods p. m. , thompson c. compact stellar x - ray sources , cambridge astroph .
series , no .
39 , cambridge university press , pp .
547 - 586 ( 2006 ) thompson c. , lyutikov m. , kulkarni s. r. the apj , v. 574 , pp .
332 - 355 ( 2002 ) iwasawa k. , koyama k. , halpern j. p. pasj , v. 44 , pp . 9 - 14 ( 1992 ) ibrahim a. i. , safi - harb s. , swank j. h. , et al .
the apj , v. 574 , pp .
l51-l55 ( 2002 ) ibrahim a. i. , safi - harb s. & swank j. h. the apj , v. 584 , pp .
l17-l21 ( 2003 ) gavriil f. p. , dib r. & kaspi v. m. 40 years of pulsars : millisec .
puls . , magnetars and more .
aip conference proceedings , v. 983 , pp .
234 - 238 ( 2008 ) gavriil f. p. , kaspi v. m. , woods p. m. nature , v. 419 , pp .
142 - 144 ( 2002 ) gavriil f. p. , kaspi v. m. , woods p. m. the apj , v. 641 , pp .
418 - 426 ( 2006 ) woods p. m. , kouveliotou c. , gavriil f. p. , et al .
the apj , v. 629 , pp .
985 - 997 ( 2005 ) bernardini f. , israel g. l. , dallosso s. , et al .
a&a , v. 498 , pp . 195 - 207 ( 2009 ) rea n. , israel g. l. , oosterbroek t. , et al .
astroph . & space science , v. 308 , pp . 505 - 511 ( 2007 ) rea n. , israel g. l. , stella l. nuclear phys .
b proceedings supplements , v. 132 , pp . 554 - 559 ( 2004 ) oosterbroek t. , parmar a. n. , rea n. , et al .
proceedings of the 5th integral workshop on the integral universe , munich , germany , pp . 471 - 475 ( 2004 ) strohmayer t. e. , ibrahim a. i. the apj , v. 537 , pp .
l111-l114 ( 2000 ) rea n. , nichelli e. , israel g. l. , et al .
mnras , v. 381 , pp .
293 - 300 ( 2007 ) protassov r. , van dyk d. a. , connors a. , et al .
the apj , v. 571 , pp . 545 - 559 ( 2002 ) ho w. c. g. , lai d. mnras , v. 327 , pp .
1081 - 1096 ( 2001 ) zane s. , turolla r. , stella l. , treves a. the apj , v. 560 , pp .
384 - 389 ( 2001 ) lai d. , ho w. c. g. the apj , v. 566 , pp .
373 - 377 ( 2002 ) | in this work , we perform the detailed analysis of absorption features in spectra of magnetar candidates observed by xmm - newton satellite .
no significant line - like feature has been found .
this negative result may indicate the possible presence of smoothing out the absorption features mechanisms .
@xmath0kyiv national taras shevchenko university , glushkova ave .
, 2 , 03127 , kyiv , ukraine + @xmath1bogolyubov institute for theoretical physics , 14-b , metrolohichna str . , kiev , 03680 , ukraine + [email protected] |
recently , the lowest - lying baryon spectrum , composed of up , down and strange quarks , has been studied by various lattice quantum chromodynamics ( lqcd ) collaborations @xcite . at present ,
most of the lqcd simulations are still performed in a finite hypercube and with larger than physical light quarks masses @xcite , the final results can only be obtained by extrapolating to the physical point ( chiral extrapolation ) and infinite space - time ( finite - volume corrections ) .
baryon chiral perturbation theory ( bchpt ) provides a useful framework to perform such extrapolations and to study the induced uncertainties . in the past decades , the ground - state ( g.s . )
octet baryon masses have been studied extensively in bchpt @xcite . however , up to now , a simultaneous description of all the @xmath1 lqcd data with finite - volume effects taken into account self - consistently is still missing .
such a study is necessary for a clarification of the convergence problem and for testing the consistency between different lqcd simulations @xcite .
furthermore , it helps determine / constrain the many unknown lecs of bchpt at next - to - next - to - next - to - leading order ( n@xmath2lo ) . in this work
we study the g.s .
octet baryon masses using the eoms - bchpt up to n@xmath2lo .
finite - volume corrections to the lattice data are calculated self - consistently @xcite and are found to be important in order to obtain a good fit of the lqcd data .
unlike refs .
@xcite , the virtual decuplet contributions are not explicitly included , because its effects can not be disentangled from those of virtual octet baryons due to the large number of unknown lecs .
in order to fix all the @xmath0 lecs and test the consistency of current lqcd calculations , we perform a simultaneous fit of all the publicly available @xmath1 lqcd data from the pacs - cs @xcite , lhpc @xcite , hsc @xcite , qcdsf - ukqcd @xcite and nplqcd @xcite collaborations .
the details of the calculation can be found in ref . @xcite .
here we only briefly summarize the main results .
up to n@xmath2lo , there are 19 lecs to be determined . to make sure that n@xmath2lo bchpt is suitable for the description of the lqcd data , we have chosen two groups of lqcd data according to the following two criteria . for the first group ,
we require @xmath3 gev@xmath4 and @xmath5 with @xmath6 and @xmath7 the spacial dimension of the lqcd simulation . for the second , we require @xmath8
gev@xmath4 , @xmath9 gev@xmath4 and @xmath10 . in the following ,
we refer to these two groups of lqcd data as data set - i and set - ii .
we perform a @xmath11 fit to the lqcd data and the physical octet baryon masses by varying the @xmath0 lecs .
the so - obtained values of the lecs from the best fits are listed in table 6 of ref .
@xcite . for the sake of comparison
, we have fitted set - i using the nlo and nnlo eoms - bchpt .
the values of the lecs @xmath12 , @xmath13 , @xmath14 , and @xmath15 are also tabulated in table 6 of ref .
an order - by - order improvement is clearly seen , with decreasing @xmath16 at each increasing chiral order .
apparently , only using the @xmath17 chiral expansion , one can not describe simultaneously the lqcd data from the five collaborations .
the corresponding @xmath18 is about @xmath19 . on the other hand , in the n@xmath2lo fit of lattice data set - i and experimental octet baryon masses , @xmath20 .
in addition , the values of the fitted lecs ( named fit i ) all look very natural .
especially , the baryon mass in the chiral limit @xmath21 mev seems to be consistent with the su(2)-bchpt value @xcite .
the fit to data set - ii yields a @xmath18 about @xmath22 and the fitted lecs look similar to those from fit i except @xmath23 , @xmath24 , @xmath25 and @xmath26 . the increased @xmath11/d.o.f .
indicates that data set - ii is a bit beyond the region of applicability of n@xmath2lo bchpt . + in fig .
[ nlo - nnlo - phy ] , setting the strange - quark mass to its physical value , we plot the light - quark mass evolution of @xmath27 , @xmath28 , @xmath29 and @xmath30 as functions of @xmath31 using the lecs from table 6 of ref .
we can see that the nnlo fitting results are more curved and do not describe well lattice data set - i . on the contrary
the two n@xmath2lo fits , named fit i and fit ii , both can give a good description of lattice data set i. the rather linear dependence of the lattice data on @xmath32 at large light quark masses , which are exhibited both by the lattice data @xcite and reported by other groups , is clearly seen .
we have studied the lowest - lying octet baryon masses with the eoms bchpt up to n@xmath2lo . the unknown lecs
are determined by a simultaneous fit of the latest @xmath1 lqcd simulations from the pacs - cs , lhpc , hsc , qcdsf - ukqcd and nplqcd collaborations .
finite - volume corrections are calculated self - consistently .
it is shown that the eleven lattice data sets with @xmath33 gev@xmath4 and @xmath34 can be fitted with a @xmath11/d.o.f .
= 1.0 . including more lattice data with larger pion masses or smaller volumes deteriorates the fit a bit but still yields a reasonable @xmath35 .
our studies confirm that covariant bchpt in the three flavor sector converges as expected , i.e. , relatively slowly as dictated by @xmath36 but with clear improvement order by order , at least concerning the octet baryon masses
. a successful simultaneous fit of all the latest @xmath1 lqcd simulations indicates that the lqcd results are consistent with each other , though their setups are quite different .
this work was partly supported by the national natural science foundation of china under grants no .
11005007 , no .
11035007 , and no .
11175002 , the fundamental research funds for the central universities , and the research fund for the doctoral program of higher education under grant no .
jmc acknowledges the spanish government and feder funds under contract fis2011 - 28853-c02 - 01 and the stfc [ grant number st / h004661/1 ] for support . | we report on a recent study of the ground - state octet baryon masses using the covariant baryon chiral perturbation theory with the extended - on - mass - shell renormalization scheme up to next - to - next - to - next - to - leading order . by adjusting the available @xmath0 low - energy constants , a good fit of the @xmath1 lattice quantum chromodynamics results from the pacs - cs , lhpc , hsc , qcdsf - ukqcd and nplqcd collaborations is achieved . |
one key concern in the study of early - type galaxies , is the amount of discrete x - ray sources relative to hot gas in x - ray faint ellipticals .
it is generally agreed that integrated x - ray emission from stellar sources ( such as accreting x - ray binaries ) is likely to be present at some fraction in all elliptical galaxies .
this contribution is low in x - ray luminous galaxies ( i.e. @xmath2 ) , where hot , diffuse gas dominates the total x - ray emission ( e.g. forman , jones & tucker 1985 ; canizares , fabbiano & trinchieri 1987 ; davis & white 1996 ; brown & bregman 1998 ; buote & fabian 1998 ) . the fraction of x - ray emission from stellar sources is expected to increase with decreasing @xmath3 , but the @xmath3 at which the stellar component ( @xmath4 ) dominates the observed emission is still debated . using _ einstein observatory _ , _ rosat _ , and _ asca _ data ,
researchers have derived estimates ( or upper limits ) to the stellar x - ray emission in e and s0 galaxies using various methods .
forman , jones & tucker ( 1985 ) determined an upper limit of @xmath5 ( _ einstein _ ipc band ) by assuming discrete sources make a 50% contribution to the total diffuse x - ray emission of cen a. in this limit , only the faintest galaxies ( @xmath6 in their sample ) can have their x - ray emission dominated by discrete - source emission .
irwin & sarazin ( 1998 ) concluded that the x - ray emission of faint galaxies ( @xmath7 ) is a combination of a hard stellar component and a very soft component , which is most likely stellar in origin as well .
this was inferred by comparing the x - ray
colors " of a sample of early - type galaxies to the bulge of m31 ( @xmath8 ; irwin 2000 , private communication ) .
fabbiano , gioia & trinchieri ( 1989 ) used the average @xmath9 of early - type spirals ( @xmath10 , for @xmath11 ) as the benchmark for the stellar contribution . by comparing the @xmath3 and @xmath12 of elliptical and s0 galaxies to spirals ( in the _ einstein _
0.54.5 kev ) , they concluded that x - ray emission from hot gas is not significant in systems for which @xmath13 . in the various estimates of the stellar emission fraction
, there is an order of magnitude difference , making an accurate determination even more necessary .
the primary mechanisms for the x - ray emission can be determined through the examination of x - ray spectra ( e.g. , trinchieri , kim , fabbiano , & canizares 1994 ; loewenstein & mushotzky 1997 ; buote & fabian 1998 ) . a hard spectrum ( @xmath14 10 kev ) would indicate the presence of evolved stellar sources , while soft x - rays ( @xmath14 1 kev ) would be reflective of a hot gas component .
recent work incorporating the use of multi - temperature spectral models indicate the presence of a very soft component in addition to the hard and soft components described above , the nature of which is suspected by most to be stellar in origin but which may be due in part to a warm ism ( pellegrini & fabbiano 1994 ; fabbiano , kim , & trinchieri 1994 ; kim et al . 1996 ) .
_ rosat _ , although it has better angular resolution than the _ einstein observatory _ , does not provide high enough spectral resolution to show signatures unique to a stellar population , especially in fainter systems .
another problem with spectral modeling is that acceptable fits can be obtained with both single temperature models and multi - temperature models .
even with the better resolution of _ asca _ , calibration problems at low energies may affect spectral modeling .
the use of x - ray radial surface brightness profiles offers another way to determine the relative fractions of gas and discrete sources to the observed x - ray emission .
previously , these profiles existed only for x - ray - bright galaxies for which adequate signal - to - noise data was achieved , and were used to obtain gas densities and masses , test predictions of the cooling flow model , or model the dynamical evolution of gas flows ( sarazin & white 1988 ; pellegrini & ciotti 1998 ; david , forman & jones 1991 ) . the use of surface brightness profiles to infer the contributions to the x - ray emission has been limited , and few conclusions have been reached with regard to faint galaxies ( forman , jones & tucker 1985 ; trinchieri , g. , fabbiano , g. , & canizares , c. r. 1986 ; pellegrini & fabbiano 1994 ) .
data has become available through _ rosat _ that better allow for the examination of radial surface brightness profiles of fainter galaxies .
brown & bregman ( 1998 , 2000 ) provide x - ray information on a _ rosat _ survey of early - type galaxies extending to fainter x - ray luminosities than available from _ einstein observatory _ data . in this paper ,
radial surface brightness profiles are used to determine or set upper limits to the stellar contribution in the less luminous galaxies of this survey .
the galaxies chosen for this study ( see table [ tab : sub ] ) are the x - ray faintest galaxies of the brown & bregman ( 1998 ) survey of early - type galaxies .
galaxy distances are derived from faber et al .
( 1989 ) using an @xmath15 = 50 km / s / mpc ( the mcmillan et al . 1994 distance is used for ngc 5102 ) .
values for the stellar velocity dispersion , @xmath16 , are also obtained from faber et al .
( 1989 ) , from which the dispersion temperature , @xmath17 , is calculated according to @xmath18 ( where @xmath19 is the mean molecular weight , and @xmath16 is the one - dimensional stellar velocity dispersion ) .
the galaxies have optical luminosities ranging from log@xmath20 10.211.2 ( derived from faber et al .
1989 magnitudes ) , except ngc 5102 whose distance determination indicates log@xmath20 9.0 .
x - ray - to - optical luminosity ratios fall between @xmath21 and @xmath22 .
lcclcr & dist & log @xmath16 & @xmath17 & log@xmath12 & log@xmath23 + & mpc & ( km s@xmath24 ) & kev & ( @xmath25 ) & ( @xmath26 ) + n 1344 & 28.44@xmath27 & 2.204 & 0.163 & 10.66@xmath28 & @xmath29 +
n 1549 & 24.26@xmath30 & 2.312 & 0.267 & 10.73@xmath28 & 29.31@xmath31 +
n 2768 & 30.64@xmath32 & 2.296 & 0.248 & 10.79@xmath33 & 29.62@xmath34 + n 3115 & 20.42@xmath35 & 2.425 & 0.450 & 10.83@xmath28 & @xmath36 + n 3377 & 17.14@xmath37 & 2.116 & 0.108 & 10.21@xmath33 & @xmath38 +
n 3379 & 17.14@xmath37 & 2.303 & 0.257 & 10.49@xmath28 & @xmath39 +
n 3557 & 47.98@xmath40 & 2.465 & 0.541 & 11.10@xmath28 & 29.51@xmath41 +
n 3585 & 23.54@xmath42 & 2.343 & 0.308 & 10.72@xmath28 & @xmath43 +
n 3607 & 39.82@xmath44 & 2.394 & 0.390 & 11.18@xmath33 & 29.64@xmath33 +
n 4365 & 26.66@xmath45 & 2.394 & 0.390 & 10.79@xmath28 & 29.69@xmath31 +
n 4494 & 13.90@xmath46 & 2.095 & 0.300 & 10.20@xmath28 & @xmath47 + n 4621 & 26.66@xmath45 & 2.381 & 0.367 & 10.78@xmath28 & @xmath48 +
n 4697 & 15.88@xmath49 & 2.218 & 0.173 & 10.58@xmath28 & 29.55@xmath28 +
n 5061 & 23.92@xmath50 & 2.282 & 0.233 & 10.53@xmath28 & @xmath51 +
n 5102 & 3.10@xmath52 & 1.820 & 0.500 & 8.95@xmath33 & @xmath53 + n 5322 & 33.22@xmath54 & 2.350 & 0.319 & 10.80@xmath33 & 29.31@xmath34 +
n 7507 & 35.00@xmath55 & 2.377 & 0.361 & 10.82@xmath28 & @xmath56 + columns 2 , 4 , & 5 derived from faber et al . 1989 values ( distance for ngc 5102 from mcmillan et al .
column 3 from faber et al .
column 6 from brown 1998 .
+ @xmath57 adopted value for @xmath17 .
the targeted galaxies are bounded by the stellar x - ray emission estimates of forman , jones & tucker ( 1985 ) and fabbiano , gioia & trinchieri ( 1989 ) . for comparison
, we have converted these estimates into _ rosat _ band equivalents .
the forman , jones & tucker ( 1985 ) _ rosat _ equivalent is @xmath58 obtained by assuming a raymond - smith thermal model with a galactic @xmath59 column density of @xmath60 and @xmath61 kev .
the fabbiano , gioia & trinchieri ( 1989 ) estimate is the averaged best - fit linear regression for a sample of early - type spiral galaxies . using an average distance of 18.12 mpc ( obtained from fabbiano , gioia & trinchieri s 1988 sample of early - type spiral galaxies ) , and spectral model parameters above , we obtain a _
rosat _ band equivalent of @xmath62 .
x - ray surface brightness profiles for the targeted sample were extracted out to 45 effective radii ( @xmath63 ) from processed _ rosat _ pspc or hri data .
observations from both instruments exist only for ngc 1549 and ngc 5322 .
details for the general data processing of these galaxies are given in brown & bregman ( 1998 ) .
a background , taken at large radius from the galaxy center ( typically at @xmath64 ) , was subtracted from the data .
normalized pspc and hri data were blocked into in 4@xmath65 and 2@xmath65 pixels , respectively , and subsequently binned according to the resolution of the instrument assuming azimuthal symmetry .
pspc data were binned in 25@xmath65 wide annuli ( data for ngc 3557 was binned in 3540@xmath65 widths beyond @xmath66 to improve signal - to - noise ) , and hri data were binned in annuli of widths increasing from 5@xmath65 to 40@xmath65 in the outer regions where the surface brightness was lower .
five galaxies had total surface brightness counts too low to be useful , and so were not included in further analysis .
ngc 4494 and ngc 4621 were also excluded because their detections were off - axis , which affected the shape of the profiles .
of the ten remaining galaxies , eight are classified as elliptical and two as s0 galaxies ( tully 1988 classification ) .
hot , interstellar gas and discrete stellar x - ray sources are the two primary mechanisms suggested for x - ray emission in early - type galaxies . moderate
to low signal - to - noise data prevents extensive and detailed modeling , however by separately characterizing the radial surface brightness distribution due to each component , we seek to reach an initial indication of the relative fraction of stellar x - ray sources present in x - ray faint galaxies we use a modified king model ( @xmath0 model ) to parameterize the x - ray emission attributable to hot gas .
the @xmath0 model was developed initially to describe hot gas behavior in clusters , but has been often extended to individual elliptical galaxies ( gorenstein et al .
1978 ; fabricant & gorenstein 1983 ; forman , jones & tucker 1985 ; trinchieri , fabbiano , & canizares 1986 ) .
the @xmath0 model takes on the form @xmath67^{-3\beta + \frac{1}{2}},\ ] ] where @xmath68 is the central brightness and @xmath69 is the core radius of the x - ray emission .
values of @xmath70 and @xmath71 2 kpc are typical in brighter ellipticals where @xmath72@xmath73 ( e.g. sarazin 1990 ; goudfrooij & de jong 1995 ) .
x - ray emission of bright early - type galaxies , modeled with this king function , demonstrate radial surface brightness profiles that decline more slowly than optical profiles at large radii ( e.g. , forman , jones & tucker 1985 ) .
a de vaucouleurs @xmath74 law is used to trace the integrated emission from discrete x - ray sources .
the discrete source emission is most likely dominated by low - mass x - ray binaries ( lmxbs ) in the galaxy ( irwin & sarazin 1998 ; david , forman , & jones 1991 ; trinchieri & fabbiano 1985 ) .
lmxbs come from an evolved stellar population , so their x - ray brightness distribution should follow that of stars ( pellegrini & fabbiano 1994 ) .
a de vaucouleurs @xmath74 law has been very successful in modeling the optical surface brightness of ellipticals ( fry et al .
1999 , burkert 1993 , hjorth & madsen 1991 ) , and so we adopt it here in the form @xmath75\},\ ] ] where @xmath63 is the effective radius ( the isophote radius containing half of the total luminosity ) and @xmath76 is the surface brightness at @xmath63 .
other models , including king models , have been used to parameterize the stellar profiles of early - type galaxies .
this was primarily because there is no straightforward transformation from the de vaucouleurs function to an analytic form of the density distribution .
however , since our goal is not to derive a density distribution , this is of no concern .
a fitting algorithm was developed to determine the four parameters - @xmath68 , @xmath69 , @xmath0 , and @xmath76 - that best represent the data according to the models described above .
the models are convolved with an instrumental point spread function ( a gaussian of @xmath77 fwhm for the pspc , and @xmath78 fwhm for the hri ) , which has the form @xmath79 where @xmath80 .
spherical symmetry is assumed in the convolved profiles , which take the form @xmath81 the function @xmath82 is the true profile ( gaseous or stellar ) , where @xmath83 the convolved profiles are optimized within the program by the @xmath84 test , which yields the best - fit values for the four parameters .
the optimization utilizes the downhill simplex method in multidimensions of nelder & mead ( 1965 ) .
surface brightness data for the targets were initially fitted to the @xmath0 model and then the de vaucouleurs @xmath74 law , out to 45 @xmath63 .
generally , @xmath69 was held as a fixed parameter since it often can not be resolved by the pspc .
if adequate hri data existed , then @xmath69 was determined by the @xmath0 model , the value of which was subsequently used in fits to the pspc data . if useful hri data was not available , a core radius of @xmath85 was adopted based upon examination of hri data for bright elliptical galaxies ( see table [ tab : rcore ] ) .
lccccc & & & + & @xmath86 & & @xmath69 & @xmath87 & @xmath88 + & ( @xmath65 ) & & ( @xmath65 ) & & + n 1395 & 4.10 & & 4.56 & 42 & 0.76 + n 1404 & 2.43 & & 5.68 & 11 & 2.84 + n 1549 & 4.30 & & 6.08 & 17 & 0.69 + n 4649 & 6.67 & & 7.66 & 30 & 1.49 + the core radius , @xmath69 , fitted within a @xmath0 model ( @xmath89 ) and compared to hri data using a @xmath90 test .
best - fits ( @xmath88 ) are for @xmath91 degrees of freedom ( @xmath87 )
. results of the @xmath0 model and de vaucouleurs @xmath74 law fits determined whether a particular profile would then be fit with a two - component function where @xmath92 theoretical calculations of @xmath0 models ( @xmath93 0.4 , 0.5 , 0.6 , and 0.7 ) show that where the diffuse x - ray emission can be described by @xmath93 0.5 ( common in bright ellipticals ) , it is not possible to conclusively distinguish between a @xmath0 model and an de vaucouleurs @xmath74 law in terms of goodness of fit ( a degenerate fit , see figure [ fig : models ] ) .
a unique solution can not be obtained with a two - component function to such a profile , therefore the @xmath94 profile was applied only when @xmath0 was determined to be steeper or flatter than @xmath950.5 for a dual - component fit , a best - fit parameterization to the data was first obtained .
next , upper limits to the stellar contribution ( relative to the best - fit ) were determined by incrementally increasing @xmath96 until the fit became unacceptable at the 90% and 9899% confidence levels : @xmath97 where @xmath98 = @xmath90 at 90% and 99% significance for 3 degrees of freedom .
once best - fit and upper limit values were found , the integrated stellar fraction was computed , from which log(@xmath99 ) can be derived .
lrrrrrrrrr & & & + & @xmath63 & @xmath68 & @xmath69 & @xmath0 & @xmath87 & @xmath88 & @xmath76 & @xmath87 & @xmath88 + & ( @xmath65 ) & ( @xmath100 ) & ( @xmath65 ) & & & & ( @xmath100 ) & & + n 1549(h)@xmath101 & 47.44 & 2.31 & 5.58 & 0.50 & 9 & 1.73 & 0.039 & 11 & 1.44 + n 1549(p)@xmath101 & & 6.38 & 5.58 & 0.46 & 7 & 3.36 & 0.150 & 8 & 3.12 + n 2768(p ) & 49.44 & 1.44 & 4.94 & 0.41 & 7 & 0.20 & 0.039 & 8 & 1.75 + n 3115(p)@xmath101 & 32.32 & 5.57 & 3.23 & 0.47 & 4 & 0.90 & 0.095 & 5 & 0.80 + n 3379(h ) & 35.19 & 6.86 & 3.43 & 0.64 & 5 & 0.75 & 0.035 & 7 & 4.09 + n 3557(p ) & 37.10 & 19.59 & 3.71 & 0.56 & 4 & 0.91 & 0.160 & 5 & 4.68 + n 3585(p)@xmath101 & 38.04 & 1.70 & 3.81 & 0.48 & 5 & 0.79 & 0.027 & 6 & 0.55 + n 3607(p ) & 65.49 & 4.80 & 6.55 & 0.39 & 11 & 3.03 & 0.150 & 12 & 17.10 + n 4365(p ) & 56.57 & 4.17 & 5.65 & 0.41 & 8 & 1.84 & 0.110 & 9 & 6.90 + n 4697(p ) & 73.51 & 9.20 & 7.35 & 0.40 & 12 & 3.53 & 0.260 & 13 & 22.81 + n 5322(h ) & 34.76 & 2.82 & 3.92 & 0.61 & 6 & 0.77 & 0.023 & 8 & 1.46 + n 5322(p ) & & 19.17 & 3.92 & 0.48 & 4 & 2.94 & 0.340 & 5 & 1.65 + best - fits ( @xmath102 degrees of freedom ) . in column 1 , h = hri , p = pspc .
a starred designation indicates both models fit equally well .
+ @xmath57 core radius , @xmath69 , fixed for pspc data at either hri fitted values or @xmath103 . single component @xmath0 model and de
vaucouleurs @xmath74 law model fits were performed separately to the surface brightness data for ten galaxies .
a summary of the results are given in table [ tab : sin ] .
if a galaxy had pspc and hri data available , fits to both data sets are summarized .
in addition to @xmath63 , fitted parameter values corresponding to the minimum @xmath90 for both models are tabulated along with the number of degrees of freedom ( @xmath87 ) and the reduced @xmath90 ( @xmath104 ) .
both @xmath0 and de vaucouleurs models fit the profiles of ngc 1549 , ngc 3115 , and ngc 3585 equally
well ( see , for example figure [ fig : n1549 ] ) , which is reflected in @xmath0 values around 0.5 . where @xmath0 was not @xmath105 , pure de vaucouleurs model fits were consistently worse compared to pure @xmath0 model fits , which is evidence against complete gas - depletion in these systems .
poor fits to ngc 3607 is a result of strong curvature in the brightness profile , which causes the data to be underestimated in the central region and overestimated in the outer regions .
small - scale fluctuations in the outer brightness profile of ngc 4697 also resulted in poor fits .
the results of gas - plus - stellar component models are summarized in table [ tab : dual ] for the seven galaxies whose derived @xmath0 values were different from 0.5 in the single - component fitting .
best - fit results as well as those for upper limits to the discrete source contribution ( at the 90% and 9899% confidence levels ) are tabulated .
the core radius , @xmath69 , is fixed in all but the best - fits to hri data .
also given in table [ tab : dual ] is the integrated count fraction of the stellar component ( @xmath106 ) , and the corresponding luminosity ratio log(@xmath99 ) . in the best - fit modeling , the fraction of surface brightness counts that can be attributed to discrete sources is less than 25% for each of the seven galaxies modeled .
the modeled profile of ngc 2768 ( figure [ fig : best ] ) indicates that the stellar component is enhanced in the central regions of the galaxy versus the outer regions ( relative to the gas component ) , the reverse of what is indicated for ngc 3557 and ngc 5322 .
ngc 4365 ( 8% stellar fraction ) exhibits an approximately even ratio of stars to gas throughout the data set .
the dual - component best - fits to ngc 3379 , ngc 3607 , and ngc 4697 are essentially the same as in single - component @xmath0 modeling .
lllrrlrrrr & & @xmath68 & @xmath69 & @xmath0 & @xmath76 & @xmath87 & @xmath88 & @xmath107 & log@xmath108 + & & ( @xmath100 ) & ( @xmath65 ) & & ( @xmath100 ) & & & & ( @xmath26 ) + n 2768 & best & 0.99 & 4.94 & 0.40 & 0.810e-2 & 6 & 0.23 & 0.16 & 28.83 + & 90% & 0.24e-1 & 4.94 & 0.25 & 0.426e-1 & 6 & 1.28 & 0.79 & 29.52 + & 99% & 0.19e-1 & 4.94 & 0.25 & 0.475e-1 & 6 & 2.12 & 0.84 & 29.54 + n 3379 & best & 6.87 & 3.42 & 0.64 & 0.987e-6 & 4 & 0.94 & @xmath1090.01 & 24.79 + & 90% & 6.12 & 3.42 & 0.86 & 0.280e-1 & 5 & 1.99 & 0.68 & 29.12 + & 99% & 5.86 & 3.42 & 0.97 & 0.352e-1 & 5 & 3.01 & 0.78 & 29.18 + n 3557 & best & 18.87 & 3.71 & 0.57 & 0.170e-1 & 3 & 1.20 & 0.13 & 28.63 + & 90% & 24.39 & 3.71 & 0.99 & 0.135 & 3 & 3.25 & 0.97 & 29.50 + & 99% & 17.45 & 3.71 & 0.99 & 0.154 & 3 & 4.94 & 0.98 & 29.50 + n 3607 & best & 4.79 & 6.55 & 0.39 & 0.809e-5 & 10 & 3.33 & @xmath1090.01 & 25.16 + & 90% & 2.27 & 6.55 & 0.36 & 0.583e-1 & 10 & 3.95 & 0.24 & 29.02 + & 99% & 1.58 & 6.55 & 0.34 & 0.783e-1 & 10 & 4.46 & 0.33 & 29.15 +
n 4365 & best & 3.48 & 5.65 & 0.40 & 0.130e-1 & 7 & 2.09 & 0.08 & 28.61 + & 90% & 0.43 & 5.65 & 0.31 & 0.890e-1 & 7 & 2.97 & 0.58 & 29.45 + & 99% & 0.19 & 5.65 & 0.28 & 0.103 & 7 & 3.69 & 0.66 & 29.51 + n 4697 & best & 9.22 & 7.35 & 0.40 & 0.123e-4 & 11 & 3.85 & @xmath1090.01 & 25.04 + & 90% & 4.19 & 7.35 & 0.37 & 0.103 & 11 & 4.41 & 0.27 & 28.97 + & 99% & 2.88 & 7.35 & 0.35 & 0.136 & 11 & 4.88 & 0.35 & 29.10 + n 5322 & best & 2.79 & 3.88 & 0.65 & 0.434e-2 & 5 & 0.91 & 0.22 & 28.66 + & 90% & 1.39 & 3.88 & 0.94 & 0.251e-1 & 6 & 1.65 & 0.89 & 29.26 + & 99% & 0.50e-5 & 3.88 & 0.54 & 0.307e-1 & 6 & 2.64 & 1.00 & 29.31 + first entry for each galaxy corresponds to best - fit . "
second and third entries for each galaxy corresponds to the 90% and 9899% upper limits to the discrete source contribution .
core radius , @xmath69 , is fixed for pspc data , and fitted for best - fit to hri data , and then fixed for subsequent upper limit fits .
@xmath102 degrees of freedom .
+ @xmath57 @xmath110 , where @xmath98 = @xmath90 at 90% and 99% significance for 3 degrees of freedom . in the 90% upper limit , the stellar fraction to the combined fit ranges from around 25% to near 100% .
the stellar component dominates the total emission of ngc 2768 except in the outermost regions ( at @xmath111 , see figure [ fig : upper ] ) , and @xmath0 flattens considerably . in ngc 3557 and ngc 5322 , a steep @xmath0 model curve rapidly falls at or before @xmath63 .
here , the stellar component dominates throughout the radial range , however gas enhancement is indicated in the center .
the greatest increase in stellar fraction occurs in ngc 3379 ( from @xmath1091% in the best - fit to 68% ) . in this galaxy ,
the gas component is dominant only in the very center ( radius @xmath112 ) .
ngc 3607 and ngc 4697 results indicate that the stellar component contributes equally , or possibly exceeds , in the very center , with gas dominating the total emission throughout the remainder of the data set . the results from ngc 4365 show stars dominating in the inner regions of the galaxy , with gas being more enhanced beyond radius @xmath113 .
we compare the modeled estimates of the stellar fraction to the estimates of forman , jones & tucker ( 1985 , fjt ) and fabbiano , gioia & trinchieri ( 1989 , fgt ) , and that estimated from m31 ( irwin 2000 , private communication ) .
the linear curves are superimposed upon a plot of the x - ray luminosities of the targeted galaxies against their optical blue luminosities ( figure [ fig : answer ] ) . for log(@xmath114
, the total @xmath3 of our sample spans the region between these estimates .
the seven galaxies , whose radial surface brightness profiles could be fitted with combined king and de vaucouleurs models , all have total @xmath9 values in the brighter half of the region of interest .
the best fit and 90% upper limit @xmath4 values are plotted for those galaxies as well .
best - fit modeling indicates a hot gas - dominated emission in each of the seven galaxies fit with the combined profile .
the median for the best fits is log(@xmath99 ) = 28.61 ( @xmath3 in ergs s@xmath24 and @xmath12 in @xmath25 ) , in reasonable agreement with the forman , jones & tucker ( 1985 ) upper limit of log(@xmath99 ) = 28.6 . in the 90 - 99% upper limit , however , all but two of the seven galaxies can be modeled with a combination profile indicative of a stellar - dominated emission .
the median for the 99% confidence fits is log(@xmath99 ) = 29.31 , which is a factor of @xmath951.8 and 1.6 below the limits of fabbiano , gioia & trinchieri ( 1989 ) and m31 respectively .
we further find @xmath115 upper limits of log(@xmath99 ) = 29.15 - 29.26 for the x - ray faintest galaxies not modeled , assuming that all of the observed emission is stellar .
we suggest , then , an upper limit to log(@xmath99 ) of 29.2 for the integrated x - ray emission from discrete sources in x - ray faint early - type galaxies .
this supports the irwin & sarazin ( 1998 ) suggestion that the brightest of x - ray faint galaxies ( log @xmath116 ) may retain a significant amount of gas with a temperature of 0.30.6 kev .
we explore how our results may relate to recent spectral studies of faint e and s0 galaxies .
_ asca _ studies of early - type galaxies confirm a hard spectral component attributed to the integrated emission from lmxbs ( e.g. , matsumoto et al . 1997 ) , and a very soft component ( vsc ) , initially detected in _ rosat
_ data ( kim , fabbiano , & trinchieri 1992 ; kim et al .
1996 ) . for galaxies with log @xmath117 ,
spectral signatures harden as @xmath9 decreases , until the hard component dominates the emission .
the kim et al .
( 1996 ) study of an x - ray faint s0 galaxy , ngc 4382 , reports that the hard spectral component contributes one - half to three - fourths of the total x - ray luminosity , with the remainder due to the vsc .
this is consistent with the earlier _ rosat _ results which indicate that the vcs and hard components contribute near equally to the total x - ray emission in galaxies with log @xmath118 ( @xmath3 in ergs s@xmath24 and @xmath12 in @xmath25 )
. the vsc may be due to a warm ism , arise from stellar sources , or be a combination of both ( irwin & sarazin 1998 ; fabbiano , kim , & trinchieri 1994 ) .
our results may be indicative of the vsc having an ism origin , considering that the best - fit modeling is consistent with a substantial ism component . in this initial analysis of x - ray surface brightness profiles ,
we have assumed low - mass x - ray binaries ( lmxbs ) are the primary constituents of a discrete x - ray source population .
lmxbs contribute to the bulk of x - ray emission in large spiral bulges such as m31 ( trinchieri & fabbiano 1985 ) .
because spiral bulges are very similar to elliptical galaxies in their properties and stellar populations , it is reasonable to presuppose that lmxbs are also the main contributors to the discrete source population in early - type galaxies ( david , forman , & jones 1991 ; irwin & sarazin 1998 ) .
however , it is possible that globular clusters may be a significant source of x - ray emission in x - ray faint e and s0 galaxies ( trinchieri & fabbiano 1985 ) , in which case the x - ray radial brightness distribution due to non - gaseous sources may be more appropriately described by a model other than a pure de vaucouleurs @xmath74 law .
the goal of finding the stellar contribution in x - ray faint galaxies has been hampered by low photon counts and lack of data from both _
rosat _ detectors .
long observations with chandra will be especially helpful in constraining the core radii in faint galaxies , fixing the shape of the radial surface brightness profiles , and resolving bright point sources .
it is also hoped that chandra will provide the spectral resolution needed to determine whether the very soft component " modeled by various authors ( see [ sec : intro ] ) is due to a warm ism of to discrete sources .
additionally , techniques that circumvent the problem of low counts , such as defining x - ray colors " ( irwin & sarazin 1998 ) , will be helpful in obtaining a definite answer to the long - standing question of dominant emission mechanisms in x - ray faint early - type galaxies .
we would like to thank j. irwin and j. s. arabadjis for valuable discussion , and the anonymous referee for helpful comments . also , we wish to acknowledge the use of the nasa extragalactic database ( ned ) , operated by ipac under contract with nasa .
b. brown would like to acknowledge support through a nasa graduate student researchers program grant ngt-51408 and a national academy of science research associateship nrc-9822890 .
j. bregman acknowledges nasa grant nag5 - 3247 .
brown , b. a. 1998 , ph.d .
thesis , university of michigan .
brown , b. a. , & bregman , j. n. 1998 , , * 495 * , l75 , astro - ph/9712209 .
brown , b. a. , & bregman , j. n. 2000 , , * 539 * , 592 .
buote , d. a. , & fabian , a. c. 1998 , , * 296 * , 977 , astro - ph/9707117 .
burkert , a. 1993 , , * 278 * , 23 .
canizares , c. r. , fabbiano , g. , & trinchieri , g. 1987 , , * 312 * , 503 .
david , l. p. , forman , w. & jones , c. 1991 , , 369 , 121 .
davis , d. s. , & white , r. e. 1996 , , 470 , l35 , astro - ph/9607052 .
fabbiano , g. , gioia , i. m. , & trinchieri , g. 1988 , , * 324 * , 749 .
fabbiano , g. , gioia , i. m. , & trinchieri , g. 1989 , , * 347 * , 127 .
fabbiano , g. , kim , d .- w . , & trinchieri , g. 1994 , , * 429 * , 94 .
faber , s. m. , wegner , g. , burstein , d. , davies , r. l. , dressler , a. , lynden - bell , d. , & terlevich , r. j. 1989 , , 69 , 763 .
fabricant , d. , & gorenstein , p. 1983 , , * 267 * , 535 .
forman , w. , jones , c. , & tucker , w. 1985 , , * 293 * , 102 .
fry , a. m. , morrison , h. l. , harding , p. , boroson , t. 1999 , in the third stromlo symposium : the galactic halo , eds .
gibson , b.k . ,
axelrod , t.s .
& putman , m.e . , asp conference series , * 165 * , 197 .
gorenstein , p. , fabricant , d. , topka , k. , harnden , f. r. , jr . , & tucker , w. h. 1978 , , * 224 * , 718 .
goudfrooij , p. & de jong , t. 1995 , , * 298 * , 784 , astro - ph/9504011 .
hjorth , j. , madsen , j. 1991 , , * 253 * , 703 .
irwin , j. a. , & sarazin , c. l. 1998 , , * 499 * , 650 , astro - ph/9804210 .
kim , d .- w . ,
fabbiano , g. , matsumoto , h. , koyama , k. , & trinchieri , g. 1996 , , * 468 * , 175 .
kim , d .- w . ,
fabbiano , g. , & trinchieri , g. 1992 , , * 393 * , 134 .
loewenstein , m. & mushotzky , r. f. 1997 , _ proceedings of iau symposium 187 on cosmic chemical evolution _
astro - ph/9710339 .
matsumoto , h. , koyama , k. , awaki , h. , tsuru , t. , loewenstein , m. , matsushita , k. 1997 , , * 482 * , 133 , astro - ph/9701077 .
mcmillan , r. , ciardullo , r. , & jacoby , g. h. 1994 , , 108 , 1610 .
mihalas , d. , & binney , j. 1981 , _ galactic astronomy _
( new york : w. h. freeman and company ) .
nelder , j. a. & mead , r. 1965 , _ computer journal _ , * 7 * , 308 .
pellegrini , s. & ciotti , l. 1998 , , * 333 * , 433 , astro - ph/9802035 .
pellegrini , s. & fabbiano , g. 1994 , , * 429 * , 105 , astro - ph/9312046 .
sarazin , c. l. 1990 , in the interstellar medium in galaxies , eds .
a. thronson , jr .
& j. m. shull ( dordrecht : kluwer ) , 201 .
sarazin , c. l & white , r. iii 1988 , , * 331 * , 102 .
trinchieri , g. , & fabbiano , g. 1985 , , * 296 * , 447 .
trinchieri , g. , fabbiano , g. , & canizares , c. r. 1986 , , * 310 * , 637 .
trinchieri , g. , kim , d .- w . ,
fabbiano , g. , & canizares , c. r. c. 1994 , , * 428 * , 555 .
tully , r. b. 1988 , nearby galaxies catalog ( cambridge university : cambridge ) | hot gas dominates the emission in x - ray luminous early - type galaxies , but in relatively x - ray faint systems , integrated x - ray emission from discrete stellar - like sources is thought to be considerable , although the amount of the contribution is controversial . to help resolve this issue
, we examine the radial x - ray surface brightness distribution of 17 x - ray faint galaxies observed with the rosat hri and pspc .
we assume that the stellar contribution follows a de vaucouleurs law while the hot gas component follows a king @xmath0 model .
for some galaxies , both models fit equally well , but for a number of systems , a dual component model yields the best fit , from which , upper bounds are placed on the stellar contribution .
best - fit values for the stellar contribution are inconsistent with ( lower than ) that suggested by fabbiano , gioia , & trinchieri ( 1989 ) and estimated from the bulge of m31 , but are consistent with the forman , jones , & tucker ( 1985 ) estimate of the stellar fraction in x - ray faint elliptical and s0 galaxies .
our results indicate an upper limit to discrete sources of @xmath1 . |
the study of stellar populations has been revolutionized by the techniques of crowded - field photometry applied to _ hubble space telescope
ironically , however , nearly all _ hst _ images are undersampled and are thus not optimal for such problems .
the information missing from undersampled images makes it difficult to detect faint sources , eliminate cosmic ray hits , register different exposures , accurately represent the stellar point spread function ( psf ) , and so on .
these problems are well known to practitioners of crowded - field photometry , and can be partially countered by using impressive software packages such as allframe ( stetson 1994 ) , dophot ( schechter , mateo , & saha 1993 ) , or ccdcap ( mighell 1997 ) , which are specially tuned to extract information from multi - image data sets that may be poorly represented in any single image .
the effects of undersampling on the detected flux of a stellar image , itself , may be less apparent , but are potentially important .
while one might assume that a photon will generate an electron that will land in one pixel or another , regardless of how the sampling is done , in reality the complex microstructure of a ccd or any other solid - state detector may cause its response to vary significantly over the area of a single pixel . in this paper
i present a method to calibrate and correct such _ intra - pixel _ sensitivity variations , with particular application to _ hst _ wfpc2 and nic3 images . in a well - sampled image
the pixel spacing or sampling frequency is sufficient to completely characterize its structural content on all spatial scales ; in a poorly sampled image fine - scale structure may be present that can interact with the yet higher spatial frequencies associated with the detector microstructure .
jorden , deltron , & oates ( 1994 ) used a pinhole projector to measure the intra - pixel response of a variety of ccds .
their experiments showed that the total detected flux of an undersampled psf can vary strongly ( @xmath0 ) with centering within a pixel , source color , and differences among the gate structure of the various ccds .
front - illuminated devices showed the strongest effects , but significant variations could still be seen with rear - illuminated ccds .
not surprisingly , the response varied differently as a function of the row versus column position , given the anisotropic structure of ccds .
real astronomical cameras are difficult to calibrate with such laboratory experiments , given the sensitivity to the degree of psf undersampling , but intra - pixel effects may be detected through a `` dithered '' set of images of a star field , that is images slightly offset from each other by a fraction of a pixel .
holtzman et al . (
1995 ) used such images of the @xmath1 cen globular cluster to show that the detected stellar flux varied by a few percent as a function of fractional column position in the wfpc-2 ccds .
while such a small effect may of little concern for most wfpc-2 programs , the situation is far different in nic3 images . here
the strong undersampling , coupled with the particular microstructure of the nicmos arrays ( which are not ccds ) causes the detected flux in a stellar image to vary by up to @xmath2 mag in the bluest ( and hence most poorly sampled ) colors .
clearly , this strongly limits use of nic3 for stellar photometry work .
calibration of the camera response to undersampled images can be done in a variety of ways , given a dithered image set of point sources .
my approach is to reconstruct a fully sampled `` superimage '' from the data set , which can be then used to make `` observed '' images true to the original sampling , but with any desired spatial offset ; one can then simply measure how the integrated flux of a point source varies with its fractional offset with respect to the pixel grid .
any single image can be expressed as @xmath3 where @xmath4 is the intrinsic projected appearance of the astronomical field being imaged , @xmath5 is the psf due to the telescope and camera optics , @xmath6 is a two - dimensional array of sampling points @xmath7 and @xmath8 means convolution . the critical term for the present discussion is @xmath9 the generally unknown spatial response of the pixel , itself .
this term not only includes the sensitivity response as a function location within the pixel , but also any diffusion of photons or photoelectrons within the device its extent may thus be larger than that of a single pixel . producing a dithered image set by stepping the detector a fractional amount in @xmath10 and @xmath11
can be used to produce a more finely sampled superimage .
when the dithers are done in a regular @xmath12 pattern of subpixel steps ( of relative size @xmath13 ) , @xmath14 the new superimage thus has an effective psf @xmath15 for severely undersampled images , @xmath16 may actually be more important than the core structure of @xmath5 for setting the effective resolution of @xmath17 if @xmath18 is large enough such that @xmath19 is fully sampled ( @xmath20 for wfpc-2 is sufficient ) , then @xmath19 can be interpolated to any desired location with respect to the original undersampled pixel array . drawing every @xmath18th pixel in @xmath10 and @xmath11 from the superimage generates an image as would have been observed at the given position . comparing the integrated flux in the interpolated - undersampled psf to that in @xmath19
thus allows the photometric effects of undersampling to be measured for any desired fractional location with respect to the original grid .
note that @xmath16 need not be determined itself , since it is implicitly included in @xmath21 and its effects depend critically on the structural content of @xmath5 in any case . the tricky step is generating a fully - sampled superimage . while it may be possible to step the detector position in a regular subpixel sampling grid ,
in practice this may be difficult .
sub - pixel dithers have been used in many wfpc-2 programs , for example , but were often not executed with enough precision to fall on a regular pattern . in this case , the simple interlacing of the dithered image set implied by equation ( [ interlace ] ) can not be done .
reconstruction of a well - sampled superimage from a set of undersampled images is a difficult problem if the geometric relationships among the images are complex . for
the simple case , however , where the images in the set are related by purely translational ( if arbitrary ) offsets , the sampling grid is spatially constant , and the intrinsic object , psf , and detector properties do not vary over the set , it is possible to construct a superimage in closed form through a complex linear combination of the images in fourier space ( lauer 1999 ) .
while these requirements may sound highly and perhaps impossibly idealized , in practice they can be realized in _ hst _
observations with both wfpc2 and nicmos , for example , if the dither offsets are relatively small with respect to any changes in the angular pixel sampling frequency , and the observations are obtained over a short enough time span such that any focus changes , source variability , and so on , are unimportant . in passing ,
i note that the _ drizzle _ algorithm ( fruchter & hook 1999 ) has proven to be an extremely popular and versatile tool for building superimages from _ hst _ data sets ; however , _
drizzle _ does not guarantee that the superimage is well - sampled , and since it also introduces a variable blurring function on the finest scales , i am concerned that it may not be suited to the present application .
in greater detail , the method of psf reconstruction advocated here involves stacking the set of dithered images in the fourier domain .
the fourier transform of a discretely sampled data set is periodic , repeating to @xmath22 in both @xmath10 and @xmath23 when an image is undersampled , or aliased , the higher order `` satellites '' in the fourier domain overlap with and contaminate the fundamental transform . this contamination can not be eliminated in any single image , but as the object being imaged is shifted with respect to the pixel grid , the phases within the fourier satellites vary . with a sufficient number of dithered images ,
each having different phases , the aliasing can be eliminated algebraically .
the sampling frequency in the final superimage is determined by the spatial scale at which @xmath19 no longer has significant power .
this approach is nicely summarized by bracewell ( 1978 ) for the case of one - dimensional data ; i present a tutorial on its extension to images , with particular application to _ hst _
wfpc-2 images , in lauer ( 1999 ) .
the attractive features of this method are that the superimage is well - sampled , there are no arbitrary parameters controlling its construction , and there is no blurring at the nyquist scale .
the method requires a minimum of @xmath24 images with non - degenerate dithers to construct a superimage with @xmath12 subsampling ; when more images are available the superimage is overdetermined and becomes the best - fit to the dither set .
the _ hst _ wfpc-2 and nicmos imagers provide contrasting test cases for exploring the effects of intra - pixel sensitivity variations .
intra - pixel effects are subtle in the wfc chips of wfpc-2 , but severe in the nic3 camera .
figures [ fig : wfc_psf_v ] to [ fig : nic_psf_h ] show psfs reconstructed with @xmath25 subsampling for the v ( f555w ) and i ( f814w ) filters in the wfc camera of wfpc-2 , and the j ( f110w ) and h ( f160w ) filters for nic3 .
the wfpc-2 psfs were constructed from dithered observations of @xmath1 cen ( stsci program 4819 ) actually obtained by the wfpc-2 idt for the purpose of understanding variations in the wfpc-2 psf as a function of pixel location . the image set consists of 20 images in each of two filters , f555w , and f814w .
the dither pattern consists of @xmath26 steps in the row and column directions , the 20 images mapping out a @xmath27 rectangle with a square grid .
the nic3 psfs were constructed from images obtained for the _ hubble deep field south _ program .
the image set comprised 146 exposures , of which 98 were used ( the star in the discarded images was either out of the field , or too close to its border ) .
the psfs shown are from the brightest star in the hdf - s field .
the hdf - s dither pattern consisted both of small and large angular offsets , largely dictated by the needs of the other cameras used in the hdf - s program . while this data set was fine for the present purposes , in many ways the dither pattern was far from optimal , an issue that i will discuss in further detail below .
it is simple to calibrate the effects of intra - pixel sensitivity variations and undersampling on a psf , once @xmath19 has been constructed .
since @xmath19 is well - sampled , its centroid can be shifted to any desired fractional pixel location , without loss of resolution or information
. once shifted , coarse samples can be drawn from @xmath19 to simulate a psf , @xmath28 as would be observed at that location .
@xmath29 is thus @xmath30 where the @xmath31 function refers to the spacing of the detector ( rather than subsampled ) pixels .
the photometric error , @xmath32 at the given offset is @xmath33 one can thus systematically map @xmath34 over the entire domain of fractional centroid offsets at any desired resolution ; the map essentially consists of a lookup table of photometric offsets to be applied to a reduced photometric data set .
it is critical to use a method of interpolation for @xmath19 that does not degrade the resolution ; i do this with sinc - function interpolation , which is the theoretically appropriate sampling kernel for well - sampled data .
lastly , i emphasize that no `` integration over a pixel '' is included in equation ( [ po ] ) , nor should be .
remember , @xmath19 already reflects convolution of the optical psf with the detector pixel response , thus this integration has already implicitly taken place . figures [ fig : v_error ] to [ fig : h_error ] show the error maps for the wfpc-2 and nic3 psfs as a function of the fractional pixel location of the psf centroid .
the square area of the maps corresponds to the domain @xmath35 @xmath36 in steps of 0.05 pixels ; @xmath37 is at the center of the maps .
an important caveat is that in practice finite limits of integration must be used in equation ( [ error ] ) , thus the absolute size of the errors will vary somewhat with aperture . in the present case
i measure the flux in a @xmath38 box for the wfc psfs and a @xmath39 box for nic3 .
overexposed stellar images show that significant scattered light falls outside these limits , but in the case of wfpc-2 at least , the aperture includes nearly all the pixels above the background for stellar images not saturated in the core .
the wfc error maps show that the photometric effects of undersampling are subtle , but are still significant for bright sources with sufficient signal .
the error map for the wfc v - band psf has a peak - to - peak range of 0.030 mag , and an rms dispersion of 0.008 mag .
the effects of undersampling are slightly reduced in i - band , as might be expected given its larger psf width ; the peak - to - peak error range is 0.023 mag , with a 0.006 mag dispersion .
the random color error is thus limited to 0.01 mag .
intriguingly , however , the v and i maps qualitatively resemble each other , thus errors in v and i may correlated depending on how the telescope was pointed for the two images .
both filters show that the error maps are anisotropic in the ccd row ( @xmath11 ) and column ( @xmath10 ) , with maximal flux ( @xmath40 ) actually corresponding to when the psf falls between two ccd columns ; however , the error maps can not be simply described as separable @xmath10 and @xmath11 functions .
these results are in excellent agreement with the simple measurements presented in holtzman et al .
( 1995 ) , who found little dependence of the photometry on fractional row location , but a few percent variation dependent on fractional column location , again with more light detected for stars centered between columns .
note that this implies that the intra - pixel response itself for ccds is more complex than a simple picture that might have fairly uniform pixels separated by less sensitive `` cracks . ''
jorden , deltron & oates ( 1994 ) indeed emphasized that at some wavelengths the ccd column stops corresponded to regions of enhanced sensitivity . at the same time
, they did see this pattern reverse in sign at other wavelengths the reader is cautioned that the present results are valid only for the f555w and f814w filters .
if undersampling effects are subtle in the wfpc-2 ccds , they completely dominate the photometric errors of stellar photometry done with the nic3 camera .
the peak - to - peak error range in the nic3 j - band is 0.39 mag ; the dispersion is 0.10 mag .
the h - band psf is broader , given the longer wavelength of the bandpass , thus reducing the undersampling effects still the errors remain large , with a 0.22 mag peak - to - peak range , and a 0.06 mag dispersion .
the greatest sensitivity indeed occurs for psfs centered on a pixel , unlike the case for the wfpc2 ccds ; given the architecture of the nicmos arrays , the picture of loosing light in the cracks between the pixels may be more valid for these devices .
a critical issue is understanding if the present calibrations can be used to correct for undersampling effects in wfpc2 or nic3 images .
unfortunately in the case of wfpc2 , it appears that the form of the psf core is the dominant contributor to undersampled structure .
maps made with different stars within the same image set have error maps that differ significantly from each other .
the wfc psfs shown above were taken from a bright star near the center of the w2 ccd of wfpc2 . a map generated from another star only 147 pixels away in w2
is shown in figure [ fig : v_error_st2 ] .
both maps show that the column position dominates the error term , with maximal flux detected for centroids falling between columns , and have about same dynamic range .
the detailed structure of the first map is different enough from the second map , however , such that it would provide little help for correcting the photometry of a star imaged at the second location on w2 .
the situation is somewhat better in the nic3 camera , where the pixel response appears to dominate .
figure [ fig : h_error_st2 ] shows the map for another star in the same h - band data set as the star shown in figure [ fig : nic_psf_h ] .
the peak - to - peak amplitude of the map is within a few hundredths of a magnitude of the map shown in figure [ fig : h_error ] , and its morphology is similar enough so that corrections derived from the former star would work well for the latter .
regardless of the utility of the present error maps for correcting undersampled stellar photometry directly , they do show what errors are likely to be encountered , and how well various dither patterns sample the error pattern .
the present analysis emphasizes constructing a well - sampled image from a dither set to counter undersampling present in any single image .
clearly , a simpler approach of averaging the photometry from a star observed at different positions in general may reduce the photometric scatter due to undersampling by @xmath41 where @xmath18 is the number of dither steps available .
a caveat is that neither the error maps , nor dither pattern are necessarily random , thus this simple scheme may produce less noise reduction than expected or still include biases among stars at differing positions , particularly if the number of dithers is small .
the large range of the nic3 error maps further imply that an extremely large data set may be required to obtain 1% photometry by the simple combination of random dithers . in this context
, i ve been surprised by how often dithered image sets fail to sample the full range of fractional pixel space adequately .
figure [ fig : nic_dithers ] , for example , shows the fractional dithers realized in the nic3 hdfs image set . despite the availability of nearly 100 images
, the pattern misses covering the fractional pixel space as well as a simple regular @xmath25 pattern would ; note that no or very few dithers fall within three of the corners of the figure . to be fair , the dithers in the hdfs program were optimized for the other cameras on board
_ hst _ , rather than nic3 , but clearly the assumption that such a large data set would randomly sample the full fractional space is not justified . in the present case , failure to include many dithers landing near the pixel corners in the case of nic3 clearly produces a strong bias , since these are the regions in which the flux deficit due to undersampling is most severe ; further , other stars in the same image set will have their dithers phased differently , thus biases due to incomplete dither coverage in this particular image set would be presented as large scatter among stars at different locations .
lastly , i note that these biases will not be removed by image reconstruction algorithms , such as _ drizzle , _ that simply redistribute the image flux ; in the end one is effectively still just averaging the stellar images .
the fourier reconstruction methods that i discuss in lauer ( 1999 ) , in contrast , can reconstruct an unbiased psf from even non - optimal dither patterns , such as that in figure [ fig : nic_dithers ] .
the trick is that the complete set of fourier components that describe the psf may still be represented in the dither set and isolated algebraically , even if it is not optimally encoded in the data .
while the error maps encode the photometric error as a function of the fractional location of the psf centroid , an important caveat is that measurement of the centroid itself will be affected by undersampling .
this issue is central to the concerns of anderson & king ( 1999 ) , who discuss methods to obtain high precision relative astrometry on wfpc2 images for the goal of measuring the relative proper motions of stars within globular clusters .
i will thus not dwell extensively on this issue , myself .
nevertheless , being able to obtain accurate centroids of psfs is critical to constructing the effective psf in the first place , and then evaluating the photometric error of stars at any position in the image set .
figure [ fig : h_centroids ] gives an error map of the total differences between the centroids of undersampled nic3 h - band psfs ( drawn from the effective psf presented in figure [ fig : nic_psf_h ] ) as a function of the true offset . in this case , the centroids were computed as the simple center of weight of a @xmath42 box centered on the brightest pixel of the extracted psf .
the size of the error varied smoothly over the fractional shift domain , ranging from 0.12 to 0.27 pixels ; if the computed centroids were used to look up the corresponding h - band photometric error in figure [ fig : h_error ] , clearly the implied photometric correction could be substantially in error .
now it is true that this simple method of calculating centroids perhaps would never be the algorithm of choice for undersampled data , but this is the point methods that do work well for centroiding well sampled psfs may work poorly for undersampled data , motivating the use of more sophisticated approaches .
the method that i have used to centroid undersampled psfs is to cross - correlate them with a well - sampled psf .
this , of course , is fine if one has already generated a psf from other stars in the image set , or can fold in knowledge of the pixel response with the construction of a theoretical psf ( see the next section ) . in practice , however , i ve found that the fourier reconstruction method is fairly tolerant of centroid errors , and for wfpc2 and nic3 data , an initial _ ad hoc _ effective psf can be constructed from simple centroiding algorithms as discussed in the previous paragraph .
the penalty is some blurring in the psf core , but the _ ad hoc _ effective psf can then be used to derive more accurate centroids from the cross - correlation method indeed , this can lead to an iterative loop where one is continually refining the centroids and the effective psf in successive stages .
if a pre - existing effective psf is available , however , a critical step is to normalize it as closely as possible to the expected flux of the new psf being constructed . with undersampled psfs , particularly when much of the flux in contained in a single bright pixel , positional information is lost and there can be strong covariance between intensity scaling and centroid measurement , a point emphasized by anderson & king ( 1999 ) . again in practice , however , i ve found with wfpc2 and nic3 data that one can readily construct an initial _ ad hoc _ psf with a rough initial normalization .
lastly , of course , good information on the dither steps may already be available from external information , or measurements conducted from an ensemble of other sources in the image set .
equation ( [ ps ] ) shows that the effective psf , @xmath21 is the convolution of the intrinsic optical psf with the pixel response , @xmath43 for the discussion so far , there has been no need to separate the two terms contributing to @xmath44 however , if knowledge of the optics - only psf is available , or it can be calculated by an algorithm such as tiny tim ( krist & hook 1997 ) , then it may be possible to isolate the pixel response by deconvolution . if @xmath16 is largely constant over the array , as jorden , deltron , & oates ( 1994 ) suggest is true for ccds , then is may be possible to use it in conjunction with theoretical spatially - variable psfs to construct improved subsampled psfs at any point within the field .
figure [ fig : subv ] shows an attempt to isolate @xmath16 for the f555w filter and the w2 ccd of wfpc2 by deconvolving the effective psf in figure [ fig : wfc_psf_v ] with a theoretical psf constructed with the tiny tim package .
the particular star selected has @xmath45 ( in the wfpc2 filters ) , which corresponds very closely to spectral type k0 .
the theoretical psf was constructed for the star s location on w2 and with @xmath25 subsampling .
deconvolution was done with 160 iterations of lucy - richardson deconvolution ( richardson 1972 ; lucy 1974 ) ; however , convergence occurred well in advance of this many iterations .
the f555w @xmath16 kernel is in excellent agreement with previous , but full - pixel rather than sub - pixel estimates of the wfpc2 pixel response .
holtzman et al . ( 1995 ) noted that some diffusion of light across pixel boundaries appeared to be occurring in the wfpc2 ccds .
krist & hook ( 1997 ) suggest a kernel that has 75% of its integral in a central pixel , with 5% flanking pixels in the row and column of the central pixel . the present kernel is given below with @xmath25 subsampling , with the pixel values given as percentages of the total integral .
@xmath46 pixels outside the @xmath42 kernel listed are essentially zero ; the diffusion out of the central ( full - sized ) pixel is actually limited to only a thin margin a single subpixel in width .
figure [ fig : subj ] shows a similar attempt to isolate @xmath16 for the f110w filter in the nic3 camera . as expected given the more severe undersampling effects in nic3 , the @xmath16 kernel is more compact and sharply peaked than the wf2 pixel response .
the nic3 pixel kernel is given below with @xmath25 subsampling , with the pixel values given as percentages of the total integral .
a one - sentence summary of this paper is that the precision of stellar photometry may be significantly limited when the psf is undersampled .
the common assumption that a ccd consists of an array of contiguous and uniform pixels is an excellent initial approximation , but is not correct in detail .
one may be tempted to adopt a refined picture in which the array consists of uniform pixels , but surrounded by dead `` moats ; '' however , this still is likely to be an oversimplification . in truth ,
the sensitivity pattern within a pixel is likely to be complex and highly dependent on the specifics of the detector architecture indeed , once one allows for possible diffusion of photons or photoelectrons within the detector , the total spatial response function of a single pixel may be more complex than can be described by pure sensitivity variations alone .
the import of the pixel response depends directly on the severity of the undersampling and the structural content of the astronomical source being imaged .
it should also be understood that the pixel response may be even more important than the core of the optics psf in setting the final resolution of an image .
if countering the effects of undersampling on stellar photometry is important , then i argue that the best solution is to dither the images in a regular pattern that permits easy reconstruction of a well - sampled superimage .
the information content of the superimage is as complete as can be allowed for the particular properties of the camera s detector and optics .
the optimal dither pattern is a regular @xmath12 grid of @xmath13 subpixel steps . in practice
one may want to add full integral steps to the fractional steps as a way of stepping over hot pixels , bad columns , traps , or any other compact detector defects ; however , if there are significant scale variations over the detectors field , then it is best to keep the total spatial extent of the dither pattern as compact as possible . in an ideal case ,
one would also obtain two or more exposures at each dither step , so as to eliminate cosmic rays events , or any other variable noise feature .
if large angular steps are desirable as well to counter any large scale variations in the detector response , then i suggest that the best way to proceed is to obtain the full data set as in subsets of complete compact dither sequences separated by the larger offsets .
each subset will make a well - sampled superimage ; combining the superimages into a final image is then simple .
if a regular dither pattern can be executed exactly , then construction of a superimage requires nothing fancier than simple interlacing of the individual images .
if the dither positions fall somewhat away from their optimal locations , but the fractional pixel domain still has good coverage , or the image set is over - determined , then i suggest the fourier method used in this paper as a possible reconstruction algorithm .
however , even if no formal reconstruction is attempted , a regular dither pattern will optimize the information content of the image set .
lastly , i emphasize that in general this is a fully two - dimensional problem .
while the wfcp2 pixel response , for example , is more important in the column direction , it can not be cleanly separated into separate @xmath10 and @xmath11 functions . for cameras like nic3 , two orthogonal one - dimensional patterns will fall well short of mapping the fractional pixel domain .
the choice of a pixel scale for an astronomical camera often requires a compromise between having a large a field as possible versus obtaining well sampled images . since even rather poorly sampled ccd cameras , such as the wfc channel of wfpc-2 ,
produce excellent stellar photometry for most problems , it is difficult to argue against tipping their design towards the largest field that the optics can accommodate . however , for many of the near and mid - ir cameras contemplated for space missions now in the early design phases , one must recognize that ir - arrays may be less forgiving of undersampled psfs and may limit the photometric accuracy to unacceptable levels if the undersampling is too extreme . if designing a nyquist - sampled camera causes unacceptable limitations on the field , however , then i suggest that one may want to include a dither capability directly in the camera , itself .
_ hst _ has demonstrated the value of dithering undersampled images , but dithering _ hst _ images is both awkward and prone to error or non - optimal patterns since the full spacecraft must be moved .
an in - camera dither capability , in contrast , can likely be made to be simple , highly accurate , and easy to invoke ; ideally , the dither capability would be accurate enough to allow for direct interlace reconstruction of the superimage .
the caveats are that the dithering must be conducted on timescales shorter than those on which significant variation in image structure will occur ; when readout noise or overhead becomes significant the minimum timescale between dithers may make dithering difficult .
lastly , i emphasize that even fine dithering is unlikely to lift the effects of severe undersampling . as
a camera becomes increasingly poorly sampled , not only does the size of the image set required rise as the square of the inverse dither step ( and with it the attendant difficulty of maintaining a stable image over the length of a dither sequence ) , but the precision to which small errors in the pattern can be detected and corrected for declines , making accurate image reconstruction all the more difficult . in practice ,
pushing beyond @xmath20 subsampling for either wfpc-2 or nic3 appears to be highly cumbersome , and thus may imply final upper limits to the pixel scales of other cameras in general if accurate photometry is both important and limited by undersampling .
i thank ken mighell , todd boroson , and dave monet for useful conversations .
i thank jay anderson and ivan king for an advance look at their manuscript on astrometry from undersampled psfs and many email exchanges , which helped inform this discussion .
harry ferguson kindly provided the reduced nicmos images used in the analysis .
alex storrs kindly provided the tiny tim nicmos psfs .
anderson , j. & king , i. r. ( 1999 ) , in preparation bracewell , r. n. ( 1978 ) , the fourier transform and its applications , p. 201 - 202
( mcgraw - hill ) fruchter , a. s. & hook , r. n. 1999 , submitted ; astro - ph/9808087 holtzman , j. , et al .
1995 , , 107 , 156 jorden , p. r. , deltron , j. -m .
, & oates , a. p. 1994 , proceedings of instrumentation in astronomy viii , spie , 2198 , 836 krist , j. & hook , r. 1997 , _ the tiny tim user s guide _ ( version 4.4 ) ( baltimore : stsci ) lauer , t. r. 1999 , 111 , 227 lucy , l. b. 1974 , , 79 , 745 mighell , k. j. 1997 , , 114 , 1458 richardson , w. h. 1972 , j. opt .
a. , 62 , 52 schechter , p. l. , mateo , m. , saha , a. 1993 , , 105 , 1342 stetson , p. b. 1994 , , 106 , 250 | an undersampled point spread function may interact with the microstructure of a solid - state detector such that the total flux detected can depend sensitively on where the psf center falls within a pixel .
such _
intra - pixel _ sensitivity variations will not be corrected by flat field calibration and may limit the accuracy of stellar photometry conducted with undersampled images , as are typical for _ hubble space telescope _ observations .
the total flux in a stellar image can vary by up to 0.03 mag in f555w wfc images depending on how it is sampled , for example .
for nic3 , these variations are especially strong , up to 0.39 mag , strongly limiting its use for stellar photometry .
intra - pixel sensitivity variations can be corrected for , however , by constructing a well - sampled psf from a dithered data set .
the reconstructed psf is the convolution of the optical psf with the pixel response .
it can be evaluated at any desired fractional pixel location to generate a table of photometric corrections as a function of relative psf centroid .
a caveat is that the centroid of an undersampled psf can also be affected by the pixel response function , thus sophisticated centroiding methods , such as cross - correlating the observed psf with its fully - sampled counterpart , are required to derive the proper photometric correction . |
entanglement is one of the most striking properties of quantum mechanics .
it describes a correlation between quantum mechanical systems that does not occur in classical mechanics , namely that a pure state of a composite quantum system can not be written as a product of the states of its constituents @xcite .
it has long been a subject of interest in foundations of quantum mechanics , but recently it has acquired a fundamental role in quantum computation and quantum information @xcite .
several attempts to quantify entanglement have been developed @xcite , but this is an area of research still in progress . in the case of a pure state of a bipartite system , a good and widely accepted measure of entanglement is the von neumann entropy of its reduced density matrix .
this entropy can be associated with a quantity called concurrence @xcite , which we adopt here as our measure of entanglement .
the concurrence varies from 0 to 1 , the larger it is the more entangled is the state .
spin chains are ideal systems for the study of entanglement .
they are naturally used to model quantum computers ( qc s ) : the two states of a spin-1/2 particle correspond to the two levels of a qubit and the exchange interaction corresponds to the qubit - qubit interaction .
a particle with spin up corresponds to an excitation or an excited qubit .
we consider a spin chain described by the @xmath0 model .
this is the model used in quantum computers based on electrons on helium @xcite .
the ising part of the interaction is proportional to the anisotropy coupling , as shown in the next section .
it only plays a role when two or more excitations are present in the chain .
the xy part of the interaction is responsible for hopping the excitations and can be used to create entanglement between two or more qubits @xcite .
however it can also quickly destroy bipartite entanglement , because the excitations soon spread all over the chain . a way to avoid the spreading ( or delocalization ) of the excitations
is by tuning the qubits away from resonance @xcite .
such control of the qubit level spacings enables us to entangle just some specific qubits .
this is the strategy we use in this paper . to have two chosen qubits maximally
entangled , we tune them in resonance .
their level spacings are different from all other qubits , they are the defects of the chain .
when there is only one excitation in the chain , the entanglement between the two selected sites becomes trivial . by assuming that the energy difference between the defects and the ordinary qubits is much larger than the strength of the @xmath1 interaction , the hamiltonian of the system gives two maximally entangled states .
they correspond to superpositions of the basis states where the excitation occupies one of the defects @xcite .
however , when more than one excitation is present , the entanglement between the defects can be significantly affected by the ising interaction . the main purpose of the present work is to analyze which is , in this case , the eigenstate with maximum concurrence and how the value of the maximum concurrence depends on the anisotropy .
we find numerically that for large values of the anisotropy coupling , nearest neighbor defects are strongly entangled and the value of the concurrence is almost negligibly affected by the number of excitations in the system . in the case of separated defects ,
this observation is valid only for long chains . in the opposite situation of small anisotropies ,
the concurrence between the defects is strongly dependent on the number of excitations and it can become quite small .
partial analytical justifications for these observations are provided .
the paper is organized as follows . in sec .
ii we describe the model and the adopted measure for bipartite entanglement . in sec .
iii we analyze the maximum concurrence between nearest neighbor and next nearest neighbor defects for various values of the anisotropy coupling . a discussion on how to prepare maximally entangled states and how the anisotropy may help is also presented in this section .
we consider a spin chain with nearest neighbor interaction .
the hamiltonian describing the system is given by the @xmath0 model @xmath2 , \\ & & h_{\rm{hop } } = \left(\sigma _ { n}^{+ } \sigma _ { n+1}^{- } + \sigma _ { n}^{- } \sigma _ { n+1}^{+ } \right ) \nonumber,\end{aligned}\ ] ] where @xmath3 and @xmath4 are pauli matrices .
there are @xmath5 sites and we deal with a periodic ( or closed ) chain , that is , sites @xmath6 and @xmath7 are the same . each site @xmath7 is subjected to a magnetic field in the @xmath8 direction , giving the energy splitting @xmath9 . in this description ,
the excitation energy of a qubit is the zeeman energy of a spin . in this disordered chain ,
not all the qubits have the same level spacing @xmath10 .
the energy of two qubits , called defects , differs from those of other qubits by @xmath11 .
the parameter @xmath12 is the hopping integral and @xmath13 is a dimensionless parameter related to the anisotropy coupling .
the diagonal term @xmath14 gives the ising interaction and the non - diagonal term @xmath15 is responsible for propagating the excitations .
we set @xmath16 and @xmath17 .
the anisotropy in the hamiltonian ( [ xxz ] ) is different from the one considered in some previous models @xcite .
the hopping part of the hamiltonian can be equivalently written as @xmath18 .
here @xmath19 , but in the models cited above , the degree of anisotropy comes from the difference between @xmath20 and @xmath21 .
the anisotropy in our case originates from the extra ising interaction . in terms of entanglement ,
few studies have been developed with this model @xcite .
moreover , our goal is not simply to analyze how entanglement may depend on the described anisotropy , but we aim to analyze how it is affected by the interplay between this interaction and disorder .
the disorder of the system we consider is characterized by the presence of the two defects . in principle , total control of the qubit level spacings is available , which allows the creation of defects .
the ising part of the @xmath0 model combined with defects has been used before to create maximally entangled states @xcite .
the study of entanglement with impurities , but without any @xmath22 interaction , has also been done in @xcite .
we emphasize the importance of studying the effects of this extra interaction in disordered systems , which are far from trivial .
the ising interaction is only relevant when at least two excitations are present , being therefore associated with many - body problems .
it is actually at the heart of one of the most challenging problems in condensed matter physics , namely the difficulty in localizing many - particle states @xcite .
in what follows , we count energy off the ground state energy @xmath23 , i.e. , we replace in eq .
( [ xxz ] ) @xmath24 . to address the different states of the system we use a notation that is common in the study of spin chains with the bethe ansatz @xcite .
the state corresponding to one single excitation on site @xmath7 , that is @xmath25 , or equivalently @xmath26 , is simply written as @xmath27 .
the state of two excitations , one on site @xmath7 and the other one on site @xmath28 , is @xmath29 , which is a simplified notation for @xmath30 , or equivalently @xmath31 .
basis states where each excitation is confined to a single site , such as @xmath32 , are called in quantum computing quantum registers .
these are the states where the measurements are performed . since we want to study the effects of the ising interaction , several excitations have to be considered , which limits the numerical analysis to small chains .
in the model described by eq .
( [ xxz ] ) , the @xmath8 component of the total spin @xmath33 is conserved , so states with different number of excitations are not coupled and the hamiltonian is made of uncoupled blocks .
however the blocks can still be very large . for a chain with @xmath5 sites ,
the block corresponding to @xmath34 excitations has dimension @xmath35 $ ] .
its diagonalization leads to eigenstates which correspond to linear superpositions of quantum registers @xmath36 with @xmath34 excitations .
each state @xmath37 is written as @xmath38 in order to study quantitatively the entanglement between two qubits we calculate their entanglement of formation @xmath39 . given the density matrix @xmath40 that describes our pair of qubits , @xmath39 is the average entanglement of the pure states of the decomposition of @xmath40 , minimized over all possible decompositions : @xmath41 where @xmath42 , @xmath43 , and @xmath44 . here
@xmath45 is the von neumann entropy of either of the two qubits @xcite .
et al _ @xcite have shown that , for a pair of qubits , @xmath39 is a monotonically increasing function of the concurrence , which one can prove to be an entanglement monotone . since the concurrence is mathematically simpler to deal with than @xmath39 , we adopt it here to measure the entanglement between two qubits .
it is given by @xcite : @xmath46 where @xmath47 , and @xmath48 are the square roots of the eigenvalues , in decreasing order , of the matrix @xmath49 .
the matrix @xmath50 is the time reversed matrix @xmath51 the symbol @xmath52 means complex conjugation of the matrix @xmath40 in the basis @xmath53 .
maximum entanglement corresponds to @xmath54 and no entanglement gives @xmath55 . to compute the concurrence of two qubits in a chain with several sites , we trace over the qubits we are not interested in and study the reduced density matrix of the two chosen ones .
we study the entanglement between the two defects of the chain .
the difference in energy between them and the other qubits , @xmath11 , is assumed much larger than the hopping integral , @xmath56 .
this guarantees that an excitation placed on one defect can only hop between the two defect sites . as a consequence , among all the eigenstates of the system , the ones with a single excitation shared between the defects have the largest concurrences . in the first subsection below
, we analyze the case where the defects correspond to two nearest neighbor qubits and in the second subsection they are two next nearest neighbors . since the chain has periodic boundary conditions , any pair of qubits @xmath57 translated through the chain is equivalent , so we choose , for the numerical calculations , the pair 1 and 2 as nearest neighbor defects and the pair 1 and 3 as next - nearest neighbor defects .
we study how the maximum concurrence between the defects depend on the anisotropy and which is the corresponding eigenstate .
the numerical results for the dependence of the maximum concurrence , @xmath58 , on the anisotropy of a periodic heisenberg chain with two neighbor defects are shown in fig .
[ fig1 ] . when @xmath59 , all states with one excitation on the defects have @xmath60 .
for small @xmath13 s , in general , @xmath58 decreases with the number of excitations .
the comparison of @xmath58 for chains of different sizes but with the same number of excitations indicates that , in most cases , smaller chains are more affected by the ising interaction . when there are just two excitations , the minimum value of @xmath58 happens when @xmath61 , while in the case of more excitations , this occurs for smaller values of @xmath13 .
whenever @xmath62 , the maximum concurrence stabilizes in a value close to 1 ( the larger the chain the closer to 1 it will be ) . in this case
, the dependence on the number of excitations becomes little noticeable . in what follows
, we try to find justifications , some times analytically , to these observations . .
the panels correspond to data for chains of different lengths : @xmath63 ( top ) , @xmath64 ( middle ) and @xmath65 ( bottom ) .
the maximum number of excitations considered for each chain is @xmath66 .
the curves are chosen as follows : solid ( black ) curves for 2 excitations , dotted ( red ) for 3 excitations , dashed ( blue ) for 4 excitations , long dashed ( green ) for 5 excitations , and dot - dashed ( violet ) for 6 excitations.,width=240 ] the eigenvalues and eigenstates of an anisotropic spin chain with no defects can be analytically obtained with the bethe ansatz @xcite . in the presence of defects , where @xmath67 , most commonly the chain becomes non integrable @xcite .
when @xmath68 , as in this paper , we are again capable of solving the eigenvalue problem in certain situations .
this happens because , with such large @xmath11 , the @xmath1-type interaction between defect and ordinary qubit is negligible , there is no hopping of excitations between them .
it works as if we had cut the closed chain and created two chains with free boundaries ( two open chains ) : a small one corresponding simply to the two defects and another one with @xmath69 sites . in the simple case of a single excitation in the chain ,
the spectrum is divided into two well separated bands .
one is made of @xmath69 states with energy in the range @xmath70 , where @xmath71 .
they are states with the excitation out of the defects .
the other band is made of just the following two linear superpositions @xmath72/\sqrt{2}$ ] , with energy @xmath73 , respectively .
they correspond to bell ( or epr ) states and have concurrence equal to @xmath74 . in this paper
we want to analyze how the ising interaction affects the entanglement and so we consider two or more excitations . in the case of two excitations , but with @xmath59 , we should obtain results similar to ones for one excitation , that is the states with maximum concurrence should give @xmath75 , since no ising interaction exists .
the states of maximum concurrence between the defect sites have one excitation shared between them .
they are superpositions of the quantum registers with energy @xmath76 , where @xmath77 and form an energy band that we will call defect band , or @xmath11-band for short .
the transitions between these quantum registers are schematically shown below . each quantum register ( apart from the border ones )
is coupled to three others : two corresponding to an excitation hop to an ordinary qubit and one associated to an excitation hop to the other defect .
@xmath78 the states of the @xmath11-band have the form @xmath79 [ to avoid a heavy notation , the superscript @xmath80 was suppressed from the right hand side . ] .
we use the bethe ansatz method to find the coefficients @xmath81 for these @xmath82 states .
the schrdinger equation for @xmath81 is @xmath83 .
\nonumber \\ & & \end{aligned}\ ] ] since the studied chain became equivalent to two open chains , it is natural to write @xmath84[b_{1 } \mathrm{e}^{\mathrm{i}\beta m } + b_{2 } \mathrm{e}^{-\mathrm{i}\beta m } ] .
\label{coef}\end{aligned}\ ] ] the energy @xmath85 as a function of @xmath86 and @xmath87 is obtained from ( [ e2 ] ) , ( [ coef ] ) and has the form @xmath88 using the boundary conditions for the small chain of defects @xmath89 and the boundary conditions for the chain of ordinary qubits @xmath90 we obtain @xmath91 \sin [ \beta ( m - n_0 -1)],\ ] ] where @xmath92 is a normalization constant and @xmath86 and @xmath87 are given by @xmath93 each one of the @xmath82 states have then a `` bell - type '' form @xmath94 \nonumber \\ & & \times \frac{1}{\sqrt{2}}\left [ \phi(n_0,m ) + ( -1)^{k_1 + 1 } \phi(n_0 + 1,m ) \right ] .
\label{bell}\end{aligned}\ ] ] to calculate the concurrence between the two defects , we obtain the reduced density matrix of the states above by tracing over the ordinary qubits .
only the four elements in the middle are different from zero .
we have @xmath95 therefore , when @xmath59 , all the states in the @xmath11-band have concurrence equal to @xmath74 .
these analytical results agree very well with the numerical ones .
the situation changes as we start increasing @xmath13 .
the scheme for the involved transitions from the previous sub - subsection still applies , but the quantum registers @xmath97 and @xmath98 now have on - site energy equal to @xmath99 , which is larger than the energy of the other quantum registers considered . as a consequence , the coefficients @xmath81 are not simply written as the product ( [ coef ] ) , which leads to the `` bell - type '' states given by eq.([bell ] ) .
however , eq.([coef ] ) can still be a reasonable approximation to obtain the state with the lowest energy in the @xmath11-band .
this state necessarily has small coefficients for the exceptional registers @xmath97 and @xmath98 , so the effects caused by their different energies , which move the eigenstates away from the `` bell - type '' states , are not so significant .
such `` ground '' state of the band has , as expected and also confirmed numerically , the largest concurrence .
it can be found analytically by using eqs.([coef]),([bound_a ] ) , and some new boundary conditions for the chain of ordinary qubits .
this is a relatively good approximation to the actual state and provides a good description for the decay of @xmath58 with @xmath13 .
since we are interested in the @xmath11-band ground state , we select @xmath100 , which gives the smallest @xmath101 .
the boundary conditions for the coefficients @xmath102 are now different .
when the first excitation is on site @xmath103 , the second excitation , necessarily out of the defects , has to satisfy @xmath104 moreover , in this case , the second excitation is also subjected to the condition @xmath105 = e_b b(n_0 - 1+l),\ ] ] which , when combined with the general equation , @xmath106 = e_b b(m),\ ] ] leads to @xmath107 equivalently , when the first excitation is on site @xmath108 , the boundary conditions for the second one are @xmath109 the wave function in the case of @xmath110 is then written as @xmath111 , \label{psi}\end{aligned}\ ] ] where a is a normalization constant and the function @xmath112 is given by @xmath113 - \sin[\beta x ] .
\label{s(x)}\ ] ] the @xmath69 different and non - trivial values of @xmath87 are obtained from @xmath114 = \sin[\beta ( l-1 ) ] .
\label{beta}\ ] ] we notice once again that we are only interested in the value of @xmath115 that gives the smallest @xmath116 .
the eigenvalues obtained with eqs.([energy]),([alpha ] ) , and ( [ beta ] ) are fairly similar to the ones obtained numerically , but the results for the eigenvectors ( [ psi ] ) are so not good .
the approximation used here is only acceptable for the lowest - energy state of the @xmath11-band .
its reduced density matrix becomes @xmath117 where @xmath118 the maximum concurrence obtained , @xmath119 is clearly smaller than 1 .
@xmath120 can only be equal to @xmath121 when @xmath59 and @xmath122 , that is when all quantum registers forming the @xmath11-band are in resonance .
however , @xmath123 and @xmath124 differ mostly at the borders of the chain of ordinary qubits ( where @xmath125 or @xmath126 ) and this difference increases as @xmath13 increases . in the case of 8 sites , @xmath127 and @xmath128 , for example , we find @xmath129 , while numerically we have @xmath130 . for a longer chain with 12 sites and the same parameters , @xmath131 , while numerically we have @xmath132 .
the agreement between numerics and analytics are not excellent , but the approximation gives the correct trend : the concurrence decreases as @xmath13 increases ( with @xmath110 ) , and the effects become less important in larger chains . in the case of resonance , when @xmath133 , the quantum registers @xmath134 and @xmath98 have very large energy and do not participate in the formation of the eigenstates in the defect band anymore .
however , quantum registers corresponding to the bound pairs @xmath135 and @xmath136 now play an important role , for they are in resonance with the other registers and couple with them in first order of perturbation theory in @xmath12 . as a result ,
a transition from the sites @xmath137 to @xmath138 [ equivalently from @xmath139 to @xmath140 , which _ removes _ the excitation from the defect , can happen .
this is the cause of the drop in the value of the maximum concurrence .
the scattering process is schematically shown below .
there are four states out of the core that could lead to a bell - type state .
@xmath141 notice that even though the other bound pairs far from the defects are also in resonance with the above states , the coupling with them occurs in second order in @xmath12 , so they form a separated band .
a more thorough discussion can be found in ref.@xcite , but the idea is the following .
there are two possible channels for the second order coupling between distant bound pairs and the @xmath11-band states .
a possibility is a transition from sites @xmath142 to @xmath143 , but we can equivalently hop from @xmath142 to the bound pair next to the defect @xmath138 [ two channels also exist for the excitations on sites @xmath144 . the amplitudes for these two transitions
are equal in magnitude , but opposite in sign . due to such quantum interference , or antiresonance
, we have two uncoupled bands of very close energy . the antiresonance explains why the drop in the value of the maximum concurrence when @xmath133 is not so drastic as one might have expected .
also , since there are only four states preventing the creation of a perfect bell state , the decay of @xmath58 becomes less perceptible in larger chains . when @xmath145 , the allowed transitions , shown below , now have two registers out of the core that could lead to a bell - type state .
@xmath146 this justifies why the maximum concurrence never goes back to 1 , even for very large @xmath147 .
however , the effects caused by just these two isolated registers , @xmath148 and @xmath149 , become less significant in larger chains . when more than two excitations are present , the analysis of the maximum concurrence for small @xmath147 s gets more complicated .
a more interesting and general situation emerges when @xmath145 . here
, we have again bell - type states similar to the ones given by eq.([bell ] ) .
these states have one excitation hopping between the two defects and all the others bound together in a cluster in sites far from the defects .
only clusters with the same number of excitations are coupled and the transitions between them happen in a high order of perturbation theory in @xmath12 .
these transitions involve virtual steps where a dissociation occurs .
therefore the clusters move together as a whole and very slowly .
we have found that states involving clusters at least two sites away from the defects have the largest concurrences .
they are written as @xmath150 \times \nonumber \\ & & \frac{1}{\sqrt{2}}\left [ \phi(n_0,m_1,m_1 + 1 , ... , m_1+n-2 ) + ( -1)^{k_1 + 1 } \phi(n_0 + 1,m_1,m_1 + 1, ... ,m_1+n-2 ) \right ] , \label{bell2}\end{aligned}\ ] ] where @xmath151 and @xmath152 .
notice that the states above can not be obtained when we have small chains with @xmath153 and the total number of excitations are @xmath154 respectively , as the available @xmath155 sites are not enough for the remaining @xmath156 excitations .
but , for these small chains and also for larger ones , there are other sorts of bell - type states that can be obtained making use of the anisotropy .
an example is the state @xmath157 $ ] found when @xmath64 .
we also note that , in the case of @xmath65 and @xmath158 , the states with large concurrences obtained numerically did not correspond to what was expected from ( [ bell2 ] ) , though they were also bell states . as a result
, the anisotropy coupling may be harmful when small , but very useful when sufficiently large , for we can recover bell - type states .
in the limit of very large @xmath13 we can use the formation of clusters to create maximally entangled states that will remain as such for a long time . as an illustration
, we take the case of @xmath65 , @xmath159 excitations , @xmath127 , @xmath160 and the two defects placed on sites @xmath74 and @xmath161 . the most straightforward method to study the dynamics of the system
consists of diagonalizing the @xmath162 hamiltonian ( [ xxz ] ) for the chosen parameters .
we then assume that the initial state is the quantum register @xmath163 with one excitation on site @xmath74 and the other three excitations bounded in a cluster on sites @xmath164 , and @xmath165 , which is not an eigenstate of the hamiltonian @xmath166 .
its evolution is obtained by writing it as a linear superposition of the eigenstates of @xmath166 with their respective energies .
the results are shown in fig .
[ fig2 ] .
the excitation on site @xmath74 hops between the two defects with a period @xmath167 @xcite .
the analytical reason for this is the following .
the states with one excitation on one of the defects and three excitations bounded together form their own energy band around the value @xmath168 .
these states are not coupled with the other states of the system and can be treated separately , largely reducing the hamiltonian to be considered . in first order of perturbation theory , the hamiltonian is still further reduced to a simple @xmath169 matrix written in the basis of the two registers @xmath170 and @xmath171 , which are coupled in first order .
the off diagonal elements of this matrix are @xmath172 .
trivially , the eigenstates and eigenvalues are the epr states @xmath173 $ ] and @xmath174 , respectivelly . for the initial state considered , the probability to find the excitation on site @xmath74 at time @xmath175 is given by @xmath176 \}/2 $ ] , while the probability to find it in @xmath161 is @xmath177\ } /2 $ ] . at each instant of time
@xmath178 , where @xmath179 is an odd number , a state of maximal entanglement between the defects is obtained , @xmath180 $ ] . after a very long time , the states @xmath163 and @xmath181 finally start mixing with the other six registers ( see eq .
( [ bell2 ] ) ) , where the cluster appears in a different position .
the hybridization of these states in time is shown in fig .
the top panel corresponds to a short time and the two lower panels are obtained after much longer times have passed . before allowing the system to evolve for a very long time , where the cluster delocalizes , states with concurrence very close to 1 can be created
we let the system evolve just up to one of the initial instants where a bell state emerges .
at this moment , the two defects can be quickly detuned and such maximally entangled state between the defects would be maintained for a long time until the cluster starts moving .
it is clear that if the initial state was one of the epr states @xmath180 $ ] , it would take a very long time to change , since this would require moving the heavy " cluster of three excitations together .
the ability to keep the entanglement between the defects very large for a long time is a consequence of the large anisotropy of the @xmath0 model . in this way
, this result strongly differs from the dynamics considered in ref .
@xcite . there ,
such effects of the ising interaction are not seen , because the dynamics of correlations is studied for an @xmath1 model . and @xmath182 .
we choose @xmath163 as the initial state and let the system evolve .
the three panels show the probability for finding other quantum registers in time . for the short times of
the top panel , @xmath163 ( black solid line ) mixes only with @xmath181 ( red dotted line ) .
the instants of time where bell - type states are created are shown with vertical long dashed lines .
they appear at @xmath183 , @xmath184 , and @xmath185 . in the middle panel ,
maximally entangled states are not obtained anymore , as other quantum registers start mixing with the previous two .
the probabilities for the registers @xmath186 and @xmath187 almost coincide and are shown with dashed lines ( blue and green , respectively ) .
the same happens to the registers @xmath188 and @xmath189 , which are indicated with long - dashed lines ( violet and indigo , respectively ) .
the lower panel is obtained for times even longer , where the quantum registers @xmath190 and @xmath191 barely start appearing .
they are shown in the figure with dot - dashed ( maroon ) lines , but can hardly been seen.,width=240 ] the description of the dependence of the maximum concurrence on the anisotropy coupling when the defects are next nearest neighbors is very complex , especially when several excitations are present .
just like in the end of the previous subsection , we restrict our study to the case of @xmath145 , where general remarks can be made .
the states with the largest concurrences are similar to the states given by eq.([bell2 ] ) .
they have one excitation hopping between the two defects , though they are now separated , so the transition occurs in second order in @xmath12 .
the other excitations are bound together in clusters placed at least two sites away from the defects .
compared to the case of nearest neighbor defects , these remaining @xmath156 excitations have now one site less available , there are @xmath192 sites for them . when @xmath156 is larger than @xmath192 , the states with the largest concurrence have a different form .
sometimes these states can also have large concurrences , but this is not always the case , which explains the curves with low concurrence in the top and middle panels in fig .
[ fig3 ] . .
the panels correspond to data for chains of different lengths : @xmath63 ( top ) , @xmath64 ( middle ) and @xmath65 ( bottom ) .
the maximum number of excitations considered for each chain is @xmath66 .
the curves are chosen as follows : solid ( black ) curves for 2 excitations , dotted ( red ) for 3 excitations , dashed ( blue ) for 4 excitations , long dashed ( green ) for 5 excitations , and dot - dashed ( violet ) for 6 excitations.,width=240 ] the important result of this section is the verification that bell - type states can again be obtained in the limit of large chains and @xmath193 , even though the defects are separated .
an analysis for the time evolution of the system with a particular @xmath11-band quantum register taken as the initial state could again be made .
it would be very similar to the one from the previous subsection , though a longer time is required for a maximally entangled state to appear .
we have shown how the interplay between interaction and disorder may affect the entanglement between two defects in a chain of qubits with anisotropic coupling .
when the difference in energy between the defects and the ordinary qubits @xmath11 is much larger than the hopping integral @xmath12 , several situations may be identified according to the value of the anisotropy coupling @xmath13 , the number of excitations in the chain , where the defects are placed , and the chain size .
the most general case refers to large chains and @xmath194 , where bell - type states are obtained even for defects that are not nearest neighbors .
these states have one excitation hopping between the two defects and all the others bound together in a cluster far from the defect sites .
these clusters are a consequence of the large anisotropy .
they move together and very slowly , allowing the maintenance of a large entanglement between the defects for a long time . in the case of nearest neighbor defects
, we also verify that even when the anisotropy coupling is not very large , the decrease in the value of entanglement between the defects is never very abrupt and it becomes less significant in larger chains .
analytical results were obtained for two excitations when @xmath195 and @xmath110 . for @xmath110
the state with the largest concurrence is the one with the smallest energy in its band .
we note that in this paper we have not developed any analysis in terms of quantum phase transition as done in several previous works @xcite . by selecting two very large defects we guarantee that the entanglement between them is mostly kept very large . if only the @xmath196 part of the hamiltonian ( [ xxz ] ) was present , i. e. if @xmath197 , their concurrence would always have the maximum value 1 , as seen from figs .
[ fig1 ] and [ fig3 ] .
it is the effect of the extra ising interaction that can sometimes decrease the entanglement .
the present work has its clear relevance for quantum information and quantum computing , but it should also be of interest for condensed matter physics , where one wants to understand how localization may be affected by interaction and disorder .
l. f. s. acknowledges support by the nsf through grant no .
itr-0085922 and thanks m. i. dykman for discussions .
g. r. thanks fapesp for funding this research .
we are both very grateful to c. o. escobar for helpful suggestions .
h. a. bethe , z. phys .
* 71 * , 205 ( 1931 ) ; c. n. yang and c. p.
yang , phys .
rev . * 150 * , 321 , 327 ( 1966 ) ; m. karbach and g. mller , e - print cond - mat/9809162 ; f. c. alcaraz , m. n. barber , and m. t. batchelor , ann . of phys .
* 182 * , 280 ( 1988 ) . | we use a disordered anti - ferromagnetic spin-1/2 chain with anisotropic exchange coupling to model an array of interacting qubits .
all qubits have the same level spacing , except two , which are called the defects of the chain .
the level spacings of the defects are equal and much larger than all the others .
we investigate how the entanglement between the two defects depends on the anisotropy of the system .
when the anisotropy coupling is much larger than the energy difference between a defect and an ordinary qubit , the two defects become strongly entangled .
small anisotropies , on the contrary , may decrease the entanglement , which is , in this case , also much affected by the number of excitations .
the analysis is made for nearest neighbor and next - nearest neighbor defects .
the decrease in the entanglement for nearest neighbor defects is not very significant , especially in large chains . |
since their initial discovery , methanol masers in the 6.7-ghz ( 5@xmath06@xmath1 a@xmath2 ) transition @xcite have been considered one of the best tracers of star formation .
methanol masers are divided into two classes , originally reflecting their different locations with respect to the associated young stellar object @xcite , but which are now understood to reflect their differing pumping mechanisms .
the 6.7-ghz transition is the strongest and most widespread of all the class ii ( radiatively pumped ) methanol masers and are considered to be exclusively associated with high - mass star formation regions ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
although a number of observations have targeted low - mass star formation regions ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) no sources of 6.7-ghz methanol maser emission have been detected to date , nor are they expected given the temperatures and methanol column densities required to produce class ii methanol maser emission ( e.g. * ? ? ?
* ) . while it is widely accepted that low - mass young stellar objects ( ysos ) are unable to produce class ii methanol masers , the lower mass limit on the stars that can produce them is not well constrained .
@xcite detected a low - luminosity 6.7-ghz methanol maser towards the orion b region ( ngc2024:fir 4 ) .
this is thought to be an intermediate mass protostar , although its true nature remains enigmatic ( see discussion in @xcite ) .
high - resolution images of some 6.7-ghz methanol masers show simple velocity gradients which have been interpreted as being due to the molecular gas lying within a rotating disk ( e.g. * ? ? ?
the mass inferred when the linear extent and velocity gradient of the 6.7-ghz methanol masers is fitted by keplerian rotation is typically less than 8 m@xmath3(e.g .
* ; * ? ? ?
* ) , the lower limit of what is generally considered a high - mass star .
however , this is likely the result of one or more of the many assumptions made in the fitting process ( particularly that the masers trace the full extent of the disk ) , as the bolometric luminosity of the associated protostars and/or associated molecular outflows generally suggest the presence of a high - mass protostar at the location of the 6.7-ghz methanol masers .
in contrast to the class ii transitions , class i methanol masers have been detected towards low - mass star formation regions @xcite , although they have much lower luminosities than those associated with high - mass star formation regions . widespread class i maser emission from the 36 ghz transition has also recently been detected towards the galactic centre region @xcite . from the observations outlined above we
can clearly infer that class i methanol maser transitions can be inverted in a variety of astrophysical environments , and are not exclusively associated with high - mass star formation .
class ii methanol masers are radiatively pumped and generally have a much more restricted distribution within high - mass star formation regions than the class i transitions ( e.g. * ? ? ?
recently the exclusivity of the association of class ii methanol masers with high - mass ysos has been questioned by the suggestion that these masers could also be associated with evolved stars @xcite . in both cases
the absence of any detectable dust continuum emission is integral to the questions raised about the nature of the sources .
@xcite compared complete samples of methanol masers @xcite and 870-@xmath4 m continuum emission ( atlasgal ; * ? ? ?
* ; * ? ? ?
they found 577 methanol masers had accompanying dust continuum emission and identified 43 methanol masers devoid of dust continuum emission . in the majority of cases @xcite were able to identify diffuse , or weak compact emission at the location of the 43 methanol masers and concluded that the lack of an associated 870-@xmath4 m point source @xcite was mostly due to these sources being located at much larger distances .
@xcite find that this explanation is unsatisfactory for a small number of sources , and remark that another possibility is that these masers arise in the circumstellar shells associated with evolved stars .
of the 43 methanol maser sources that @xcite find to have no accompanying dust continuum emission @xcite , 10 had been reported previously at other wavelengths .
one of these shows iras emission , two sources were detected in the bolocam galactic plane survey ( detected due to its sensitivity to low surface brightness objects being superior to atlasgal ; * ? ? ?
* ) and seven are listed in the intrinsically red source compilation of @xcite .
of the seven sources in @xcite , six are identified as ysos and a further source is listed as a possible agb star ( g328.385 + 0.131 ) .
@xcite use the point - like nature of the glimpse 8.0-@xmath4 m emission coincident with this source to assert that this object is indeed an evolved star , since they expect 8.0-@xmath4 m emission associated with high - mass star formation regions to be extended .
previously , @xcite conducted 450-@xmath4 m and 850-@xmath4 m dust continuum observations with scuba on the jcmt towards 71 methanol masers , identifying one 6.7-ghz methanol maser site that was not associated with any sub - mm continuum emission .
the methanol maser , g10.10 + 0.73 , was discovered in parkes observations towards iras 180211950 @xcite and later followed up with the atca to determine a precise position @xcite . while @xcite failed to detect any sub - mm continuum emission associated with the methanol maser , they did detect a compact source within the target field ( offset 90 " from the methanol maser ) , associated with the bipolar planetary nebula ngc 6537 .
the authors question whether there is an association between the methanol maser and the planetary nebula , suggesting that the kinematics are well matched but maintaining that a chance alignment is a possibility .
@xcite claim that this source supports their inference that methanol masers may be associated with evolved stars . here
we present some general arguments as to why evolved stars are unlikely to harbour 6.7-ghz methanol masers ( section [ sect : meth ] ) , briefly discuss the dust properties in star formation regions ( section [ sect : dust ] ) and then focus our investigation on the two 6.7-ghz methanol masers , g328.385 + 0.131 ( section [ sect:328 ] ) and g10.10 + 0.73 ( section [ sect : walsh ] ) , where an association with evolved stars has been explicitly suggested @xcite .
oh and water masers are observed in a wide range of astrophysical environments - from star forming regions ( both low and high - mass in the case of water masers ) , evolved stars , supernovae remnants ( 1720-mhz oh ) and the central regions of active galaxies .
the 6.7-ghz methanol transition has been shown to be strongly inverted over a broad range of physical conditions ( e.g. * ? ? ?
* ) , similar to the ground - state oh and 22-ghz water transitions .
therefore it is natural to ask the question : are there astrophysical environments , other than those associated with high - mass star formation regions , which can support 6.7-ghz methanol masers ?
there are nearly 1000 known 6.7-ghz methanol maser sources @xcite and in excess of 2000 late - type stars known to exhibit maser emission - primarily found from searches for 1612-mhz oh @xcite or sio ( e.g. * ? ? ?
* ) maser transitions . to date , no overlap between the 6.7-ghz methanol masers and evolved stars showing either 1612-mhz oh or sio maser emission has been identified . given the large size of both maser samples , and their typical positional accuracies of @xmath51 arc second , we can conclude that evolved stars showing 1612-mhz oh or 43-ghz sio masers do not have conditions which are suitable for 6.7-ghz methanol masers . for a molecular transition to exhibit
maser emission requires both suitable physical conditions ( temperature , density , etc ) , and also a sufficient abundance of the relevant molecule . as evolved stars host maser emission in a range of ground - state oh transitions , as well as water and sio
, it would seem likely that some regions within that environment will have the temperatures and densities within the broad range necessary to support 6.7-ghz methanol masers . the critical difference between methanol and either oh or water seems to be that methanol is less readily produced in high abundance than either oh or water . @xcite summarises the evidence as to why methanol is thought not to be produced in gas phase reactions in the interstellar environment .
the favoured mechanism for the formation of methanol is through hydrogenation of co when it freezes out on dust grains ( e.g. * ? ? ?
low grain temperatures favour the formation of formaldehyde and methanol rather than co@xmath6 @xcite , and in some very young star formation regions infrared spectroscopy has found methanol ice to be the second most abundant ice on dust grains , after water @xcite . methanol ice has been observed towards both low- and high - mass star formation regions ( e.g. * ? ? ?
* ; * ? ? ?
* ) . as the forming stars heat the surrounding medium the methanol desorbs from the dust grains and is released into the gas phase .
mechanical desorption of the dust grains by outflows is also possible ( e.g. * ? ? ?
* ) and can play an important role in increasing the methanol abundance in the gas phase in some sources .
similar to star formation regions , the environment near evolved low - mass stars contains large amounts of dust , however , it is not expected to be at temperatures of around 10 k , which favour the formation of methanol on the grain mantles ( e.g. * ? ? ?
* ) . amongst evolved - star maser sources , a small number show emission in atypical maser transitions , for example the proto - planetary nebulae k 3 - 35 possesses 1720 mhz and 6 ghz excited oh masers @xcite
some very young planetary nebulae and proto - planetary nebulae also have associated cold dust @xcite , however , the dust masses are very much less than 1 m@xmath3 and are unlikely to facilitate the formation of large amounts of methanol ice . to date there have been no detections of methanol towards any evolved stars , despite sensitive measurements towards a number of nearby sources ( e.g. * ? ? ? * ; * ? ? ?
in particular , observations towards the extreme carbon star irc+10216 ( towards which more than 50 molecular species have been observed ) , failed to detected any methanol with a 3-@xmath7 abundance upper limit of @xmath8 ( although the same observations detected formaldehyde with more than an order of magnitude greater abundance ) . in summary , current understanding of the formation process for methanol in astrophysical environments suggest that it is unlikely to be present in the vicinity of evolved stars and this is supported by existing observational studies .
@xcite made 1.2-mm dust continuum observations of a sample of 131 high - mass star formation regions , identified on the basis of either the presence of an uchii region , or a 6.7-ghz methanol maser . similar to @xcite , @xcite found a small fraction of their targets ( 20 methanol masers and 9 uchii regions ) for which they did not detect dust continuum emission . at the 3-@xmath7 sensitivity of their observations ,
@xcite were sensitive to all sources with a dust mass in excess of 600 m@xmath3 at distances of 16.3 kpc or less ( i.e. over the vast majority of the region of the galaxy where high - mass star formation is present ) .
since the majority of the dust sources they detected have masses in excess of their sensitivity limit , they suggest three possible explanations for the 29 non - detections : 1 ) these sources have characteristics dissimilar to the majority of high - mass star formation regions ; 2 ) that these sources are associated with later stages of star formation ; or 3 ) these sources are located at too great a distance and are not massive enough to be detectable .
at least two of the nine uchii regions for which @xcite find no accompanying dust continuum emission have been extensively studied ( g188.770 + 1.074 and g23.430.18 e.g. ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) and their uchii region nature is irrefutable . both of these sources are located in complexes which contain methanol masers with distances measured through trigonometric parallax observations ( g188.95 + 0.89 and g23.440.18 ; * ? ? ?
g188.770 + 1.074 and g23.430.18 are located at 2.10 and 5.88 kpc @xcite , respectively , meaning that the @xcite observations are not sensitivity limited .
this clearly demonstrates that some high - mass star formation regions have unusually low dust masses , by the evolutionary phase when uchii regions and 6.7-ghz methanol masers are present .
thus , the absence of dust emission associated with a methanol maser is not itself sufficient to suggest an origin other than a high - mass star formation region .
we have investigated the two maser sources for which an association with an evolved star has been proposed @xcite .
the results of our investigation of g328.385 + 0.131 are given in detail in section [ sect:328 ] .
the case of g10.10 + 0.73 can be simply dismissed as a spurious detection as discussed in section [ sect : walsh ] .
here we summarise the key arguments which clearly show that there is no basis to the proposition that g328.385 + 0.131 is associated with an evolved star . *
the classification of g328.385 + 0.131 by @xcite as a possible agb star is likely erroneous .
the source is most probably a luminous yso which their mid - infrared colour - magnitude criteria misclassified . *
the infrared properties of g328.385 + 0.131 are in the middle of the range seen for large samples of methanol masers .
we further find the point - like nature of the associated 8.0-@xmath4 m emission to be fairly common for sources associated with methanol masers . *
the methanol maser at g328.385 + 0.131 has a high luminosity . if such a luminous maser were associated with an evolved star we would expect to see other examples within the galaxy .
* we find a convincing @xmath53-@xmath7 870-@xmath4 m detection ( @xmath5320 mjy ) at g328.385 + 0.131 in the atlasgal data , contrary to the non - detection reported by @xcite , which surpasses the flux density possible from any agb star .
* the mass of the dust associated with g328.385 + 0.131 is @xmath5600 m@xmath3 which is consistent with other regions that support high - mass star formation .
the possible agb star designation assigned to g328.385 + 0.131 by @xcite was determined from a set of criteria they derived for the separation of ysos and agb stars . @xcite
classify sources with [ 4.5 ] @xmath9 7.8 as ` extreme ' agb star candidates , but note that this criterion will also capture a number of luminous ysos . in the case
where [ 4.5 ] @xmath10 7.8 there are two further criteria : if [ 8.0 ] [ 24.0 ] @xmath11 2.5 they classify the source as a ` standard ' agb star , and if [ 8.0 ] [ 24.0 ] @xmath12 2.5 they classify the source as a yso . examining the relevant colours and magnitudes of g328.385 + 0.131 ,
it is clear that it barely makes it into the possible agb category on the first criterion , with [ 4.5 ] = 7.68 ( in fact , this value is only 3.2-@xmath7 from the 4.8 micron magnitude threshold of 7.8 ) .
if this source had a slightly higher magnitude , it would be classified as a yso as it satisfies the second criterion ( [ 8.0 ] [ 24.0 ] = 3.5 ) .
it is therefore likely this source is one of the luminous ysos that has been wrongly classified as an agb star , as @xcite mentioned was inevitable .
@xcite used an adaptive non - circular aperture photometry technique to determine the glimpse fluxes for sources associated with the methanol masers detected in the methanol multibeam survey ( a survey for methanol maser emission in the galactic plane ; * ? ? ? * ) .
their table 1 lists 4.5-@xmath4 m magnitudes towards 444 methanol masers , a large fraction of which ( 35 per cent ) have [ 4.5 ] @xmath9 7.8 and would therefore also be classified as ` extreme ' agb stars by the @xcite criteria .
it is clearly unreasonable to expect such a large fraction of methanol masers to be associated with agb stars , when an alternative explanation exists in which these are instead the expected fraction of ysos that are inevitably captured by the simplistic colour - magnitude criteria @xcite .
this highlights the difficulties in relying upon infrared selection criteria alone to classify individual sources .
furthermore , looking more closely at the irac colours of the mid - infrared counterpart of g328.385 + 0.131 , it is evident that this source lies in the middle of the range of colour - colour and colour - magnitude plots of sources associated with methanol masers @xcite adding further support to the argument that this source is actually a standard high - mass star formation region . in their study of the glimpse properties of sources
associated with 6.7-ghz methanol masers , @xcite also found that of their 769 methanol masers , 219 ( 28 per cent ) were well characterised by a point source in all four irac bands . since this is the case , the point - like nature of the infrared source associated with g328.385 + 0.131does
not indicate that the emission is associated with an agb star , as suggested by @xcite . our ability to confidently estimate the distance of g328.385 + 0.131 is dependent on the nature of the source :
if it is a high - mass star formation region it is necessarily constrained to the spiral arms and the velocity of the methanol maser emission will follow that of galactic rotation ( 6.7-ghz methanol masers typically show their median velocity within 3 5 @xmath13of the systemic velocity of the region ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , making kinematic distance determination viable ; if , on the other hand , the source is an agb star , the range of expected velocities is not restricted to galactic rotation and kinematic distance determination is subject to much greater uncertainty .
we have inspected the co emission from the nanten galactic plane survey @xcite and find an isolated spectral feature in the co spectrum that matches the velocity of the methanol maser emission ( see fig .
[ fig : co_meth ] ) .
the spectral structure of the detected co emission is unlike typical agb co spectra ( e.g. * ? ? ? * ) which tend to be broad ( fwhm @xmath1020 @xmath13 ) , flat - topped and steep - edged ( i.e. more like a top - hat than a gaussian profile ) .
crucially , if the co was associated with an agb of an anomalous velocity we would expect to see the co emission restricted only to the immediate spatial vicinity of the agb star . @xcite measured the co properties of a representative sample of agb and post - agb stars , finding that the ( 1 - 0 ) emission was extended by no more than @xmath540 arcsec for sources at distances up to 1 kpc .
the nanten survey was observed in a 4 arcmin grid and the co emission presented in fig .
[ fig : co_meth ] extends to at least two positions adjacent to g328.400 + 0.133 , corresponding to a structure of co covering at least 8 arcmin .
such a large structure is wholly inconsistent with the co emission expected for any agb star , even one that is very close by .
conversely , such a co structure is completely consistent with the extended emission seen within the spiral arms of our galaxy , further confirmed by the velocity of the emission , which matches that of the carina - sagittarius spiral arm at this galactic longitude . given the tendency of methanol masers to trace the systemic velocities of the regions they are associated with , the well matched velocities of the methanol and co emission , together with the expected velocity of the carina - sagittarius spiral arm at this longitude and the low latitude of the source , it is most likely that the methanol maser is associated with co following galactic rotation .
thus the positive velocity of the methanol maser at 28.9 @xmath13 @xcite undoubtedly places it at the far kinematic distance , in agreement with the distance assigned by @xcite of 18.0 kpc .
we further note the presence of nearby luminous snr 328.4 + 0.2 , which is located at a similar distance ( 16.8 kpc : @xcite ; [email protected] kpc : @xcite ) , illustrating that this part of the galaxy is clearly capable of supporting high - mass star formation ; in fact , the galactic longitude versus velocity diagram presented in @xcite shows that there are several other methanol masers with similar galactic longitudes and velocities .
the detected co emission implies a h@xmath6 mass of @xmath55300 m@xmath3 ( assuming a solar - neighbourhood galactic x - factor of 1.8@xmath1510@xmath16 ; * ? ? ? * and a distance of 18 kpc ) in a single nanten position , or a total cloud mass of @xmath510@xmath17 m@xmath3 ( over the three adjacent positions where co at this velocity was detected ) .
the estimated mass of the molecular gas is also consistent with the cloud being located within the spiral arms of our galaxy .
the association of the methanol maser with the co - rotating gas within the spiral arms of our galaxy does not alone definitively suggest either an agb or high - mass star formation origin .
however , the strong likelihood of a far kinematic distance ( @xmath518 kpc ) is key to arguments made later in this section and those in section [ sect : dust ] .
the methanol maser emission associated with g328.385 + 0.131 was reported by @xcite to consist of a single spectral feature with a peak flux density of 1.6 jy , and appeared stable over the course of the initial mmb survey and followup observations . while this is a relatively modest flux density , its distance at 18 kpc places it in the upper middle of the luminosity range of all 366 mmb sources presented in fig .
7 of @xcite .
if an agb star were capable of producing such a luminous methanol maser , we would expect to detect further sources of this type within the galaxy , but we do not .
comparison between a large fraction of the full complement of 6.7-ghz methanol masers in the southern hemisphere and mid - infrared data lead @xcite to reconfirm that these masers are exclusively associated with high - mass star formation .
if there were a significant population of methanol masers associated with evolved stars , it would be expected that the @xcite study ( and other earlier studies ) would have uncovered it .
no further indications of the source s nature can be derived from the targeted 12.2-ghz methanol maser observations conducted by @xcite , which revealed no emission at 5-@xmath7 levels of 0.75 and 0.8 jy during 2008 june and december , respectively .
the 12.2-ghz methanol maser catalogues of @xcite show that in the galactic longitude range 186 to 330@xmath18 approximately 44 per cent of 6.7-ghz methanol masers have 12.2-ghz counterparts ( commonly with 6.7-ghz to 12.2-ghz peak flux density ratios greater than 2:1 ) .
we have inspected the water maser @xcite and ammonia data @xcite from the h@xmath6o southern galactic plane survey ( hops ) and find no detection of either tracer . the modest sensitivity of this survey ( 98 per cent complete at 8.4 jy for water masers and 100 per cent complete at @xmath19 = 1.0 k for ammonia ) combined with the outer galactic distance for this source preclude us from drawing any conclusions from these data .
recent observations for oh masers in the 1612- , 1665- , 1667- and 1720-mhz lines have been carried out as part of magmo ( a project to study the magnetic fields of the milky way through oh masers ; * ? ? ?
* ) towards this source and made no detections in any of the oh transitions to a 5-@xmath7 detection limit of @xmath5250 mjy . using atlasgal data , @xcite report that there is no 870-@xmath4 m emission at the location of the methanol maser to an upper limit of 320 mjy .
we have inspected the image provided by @xcite , which shows the mid - infrared environment associated with g328.385 + 0.131 overlaid with atlasgal contours .
this image shows unresolved emission at a 3-@xmath7 level ( signified by the presence of three 1-@xmath7 interval contours ) towards the location of the methanol maser .
given the coincidence of the dust emission with the methanol maser , and the absence of any further emission beyond a 1-@xmath7 level in the region surrounding the methanol maser , we believe this to be an authentic dust continuum detection .
we therefore reinterpret the @xcite upper limit of 320 mjy to be a detection at about this flux density and similarly the presented dust mass upper limit of 579.17 m@xmath3beam@xmath20 to be the estimated dust mass of the detected source .
@xcite find that 28 per cent of dust sources with methanol masers have calculated dust masses of less than 1000 m@xmath3 .
they suggest that the majority of these are more compact ( in most cases @xmath110.3 pc ) and may have higher star - formation - efficiency and be forming smaller - stellar - mass systems . from figure 15 of @xcite , it seems that all of their sources with masses greater than about 100 m@xmath3 fall in the high - mass star forming range of values satisfying the @xcite criteria
. there is therefore no evidence to suggest that g328.385 + 0.131 , with a mass of @xmath5550 m@xmath3beam@xmath20 , would be unable to support high - mass star formation .
furthermore , @xcite apply a constant temperature of 20 k to all of their sources .
while this is a reasonable average for many young star formation regions , individual sources may vary by several k from this value which significantly impacts the derived mass ( e.g. if the temperature was actually 15 k , as can often be the case ( e.g. * ? ? ?
* ) , the mass would increase by 58 per cent ) .
conversely , the likelihood that an agb star would be detectable at such a far distance in the 870-@xmath4 m atlasgal data seems unlikely . to consider
if this is possible , we have calculated the expected 870-@xmath4 m emission from an agb star located at 18 kpc .
the flux density of the dust envelope of an agb star at 870-@xmath4 m , assuming that the dust continuum emission is optically thin , is given by @xcite : @xmath21 where @xmath22 is the agb envelope mass , @xmath23 is the dust opacity coefficient , @xmath24 is the planck function at the dust temperature ( ergs s@xmath20 @xmath25 ) , and d is the distance ( pc ) .
the dust opacity at 870-@xmath4 m has an estimated value of 1.87 @xmath26 g@xmath20 for atlasgal observations @xcite , however , toward circumstellar grains it can vary by up to an order of magnitude @xcite .
@xcite conducted 870-@xmath4 m observations towards nine evolved stars and found values of dust opacity ranging from 2 to 35 @xmath26 g@xmath20 .
we therefore calculate limits on the expected flux density using these two extreme values . assuming a dust temperature of 1000 k ( estimated from observations of nearby agb stars ; * ? ? ?
* ) , together with their average dust envelope mass of @xmath52@xmath1510@xmath27 m@xmath3 , the expected flux densities for each of the two extreme dust opacity values are : @xmath28 making it is very unlikely that 870-@xmath4 m emission would be any stronger than 160 mjy and quite likely considerably less given a more reasonable value of dust opacity . since the flux density of the 3-@xmath7 870-@xmath4 m emission associated with the methanol maser is @xmath5320 mjy , it is highly probable that we can rule out that the origin of this emission is an agb star .
we have used the online sed fitting tool of @xcite to investigate the nature of the emission associated with g328.385 + 0.131 .
we allowed the fitting distance to vary between 16 and 18 kpc .
the resultant sed fit is presented in fig .
[ fig : sed ] , and shows the characteristics of a deeply embedded , high luminosity yso . using the criterion of @xcite
, we were able to confirm that there was no significant pah emission , which is not accounted for in the yso models used in the sed fitting @xcite .
@xcite investigated the seds of the centimeter through to near - infrared emission associated with 20 6.7-ghz methanol masers which comprise a complete sample of methanol masers between galactic longitudes 38.6@xmath18 and 43.1@xmath18 , and latitudes @xmath29b@xmath29 @xmath9 0.42@xmath18 .
the fitted characteristics of g328.385 + 0.131 ( derived from the best fitting model shown in fig .
[ fig : sed ] ) all fall within the range of those derived for the @xcite sample .
the only exception is the source age which is about twice that of the oldest sources given in their study .
the significance of this difference is unlikely to be high given the size of their sample , together with the range of their values ( 1.2 to 57 @xmath15 10@xmath30 years c.f .
110 @xmath15 10@xmath30 years for g328.385 + 0.131 ) .
@xcite present a sample of seds for both their standard agb and yso candidates .
although some of the ` extreme ' agb candidates presented in @xcite have seds that are more comparable to that of ysos , the standard agb candidates have very different characteristics than those of g328.385 + 0.131 shown in fig . [ fig : sed ] . in each of the 18 standard agb cases presented in @xcite ,
the sed falls significantly beyond wavelengths of 5.8-@xmath4 m ( and sometimes from even shorter wavelengths ) , a property not shared by the data presented for g328.385 + 0.131 which is well fitted by a yso model .
@xcite conducted atca observations of methanol masers detected towards iras sources @xcite in order to derive precise source positions .
the methanol maser observed by @xcite in the target field of iras source 180211950 was found to be significantly offset from the target iras source ( by 116.79 `` in ra and 38.46 '' in declination ) and was detected as a single feature of 1.2 jy at a velocity of 1.2 @xmath13 .
the location of this methanol maser was found to have no coincident dust continuum emission @xcite and , combined with its proximity to planetary nebula ngc 6537 led to the suggestion of a possible association between the two objects .
this methanol maser was not detected in the complete mmb survey @xcite , perhaps not too surprising given the intrinsically variable nature of masers and the 5-@xmath7 sensitivity of the survey observations , which at 0.7 jy is not greatly different from the flux density of the @xcite detection .
although the non - detection in the @xcite observation is not suspicious alone , the reported maser is , notably , less than a degree from the strongest methanol maser in the galaxy , g9.621 + 0.196 , with velocity of 1.3 @xmath13 and emission regularly detected with a peak flux density in excess of 5000 jy ( e.g. * ? ? ?
* ) , and this prompted us to download and inspect the data used in @xcite from the australia telescope online archive ( atoa ) .
the atca data was taken on 1994 july 30 and consisted of four cuts of two minutes each , spread over almost 8 hours .
observations of pks b1934638 were made for primary flux density and bandpass calibration , and 1808209 for gain calibration ( which had a flux density of @xmath50.25 jy at the time of the observations ) .
for these observations the correlator was set to record 1024 channels over a bandwidth of 4-mhz , yielding a channel spacing of 0.18 @xmath13 and a coverage of 180 @xmath13 @xcite . despite several attempts using miriad , employing standard data reduction techniques for atca spectral line data ,
we have been unable to reproduce the 1.2 jy detection from the same atca data .
we suggest that the reported detection is most likely an image artefact arising from the nearby bright of g9.621 + 0.196 which was more pronounced than usual because the quality of the data was inadequate to form reliable images ( in particular , the phase calibrator was too weak for the correlator configuration and system available at the time of the observations ) and possibly further due to inferior data reduction algorithms available almost 20 years ago .
we conclude that the reported methanol maser g10.10 + 0.73 detected at a velocity of 1.2 @xmath13 is not an authentic maser detection and therefore has no possibility of supporting the suggestion that methanol masers can be associated with evolved stars as suggested in both @xcite and @xcite .
we have investigated the suspected association between two 6.7-ghz methanol masers , g328.385 + 0.131 and g10.10 + 0.73 , with evolved stars @xcite .
we find no evidence to support these claims , rather finding g328.385 + 0.131 to be an average methanol - maser - associated yso , and g10.10 + 0.73 to be a spurious source caused by the strong emission from nearby methanol maser g9.621 + 0.196 . more generally , we argue that it is unlikely that the environments of evolved stars contain a high enough abundance of methanol to support maser emission .
we therefore confirm that the exclusive association of 6.7-ghz methanol masers with star formation regions remains valid .
we thank the referee , anita richards , for insightful and thorough comments that allowed us to significantly improve the quality of the paper . this research has made use of : nasa s astrophysics data system abstract service ; the nasa/ ipac infrared science archive ( which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration ) ; the simbad data base , operated at cds , strasbourg , france ; and data products from the glimpse and mipsgal surveys , which are legacy science programs of the _ spitzer space telescope _ , funded by the national aeronautics and space administration .
breen s. l. , ellingsen s. p. , caswell j. l. , green j. a. , fuller g. a. , voronkov m. a. , quinn l. j. , avison a. , 2012 , mnras , 426 , 2189 carey s. j. , feldman p. a. , redman r. o. , egan m. p. , macleod j. m. , price s. d. , 2000 , apjl , 543 , l157 deguchi s. , 2007 , in chapman j. m. , baan w. a. , eds .
iau symp . , 242 , astrophysical masers and their environments .
cambridge univ . press , cambridge ,
desmurs j .- f .
, baudry a. , sivagnanam p. , henkel c. , richards a. m. s. , bains i. , 2010 , a&a , 520 , 45 mizuno a. , fukui y. , 2004 , in clemens d. , shah r. , brainerd t. , eds , asp conf .
317 , milky way surveys : the structure and evolution of our galaxy .
, san francisco , p. 59
norris r. p. , caswell j. l. , gardner f. f. , wellington k. j. , 1987 , apj , 321 , l159 purcell c. r. , et al . , 2012 , mnras , 426 , 1972 reid m. j. , menten k. m. , zheng x. w. , brunthaler a. , moscadelli l. , xu y. , zhang b. , sato m. , honma m. , hirota t. , hachisuka k. , choi y. k. , moellenbrock g. a. , bartkiewicz a. , 2009 , apj , 700 , 137 | recently , a comparison between the locations of 6.7-ghz methanol masers and dust continuum emission has renewed speculation that these masers can be associated with evolved stars .
the implication of such a scenario would be profound , especially for the interpretation of large surveys for 6.7-ghz masers , individual studies where high - mass star formation has been inferred from the presence of 6.7-ghz methanol masers , and for the pumping mechanisms of these masers .
we have investigated the two instances where 6.7-ghz methanol masers have been explicitly suggested to be associated with evolved stars , and we find the first to be associated with a standard high - mass star formation region , and the second to be a spurious detection .
we also find no evidence to suggest that the methanol maser action can be supported in the environments of evolved stars .
we thereby confirm their exclusive association with high - mass star formation regions .
masers stars : formation ism : molecules |
for the same reason that eclipsing binary stars are important in stellar astrophysics , transiting planets play an outsized role in exoplanetary science . this can be appreciated by comparing the transiting exoplanet hd 189733b ( bouchy et al .
2005 ) , the subject of this paper , to the arbitrarily chosen non - transiting planet hd 187123b ( butler et al .
both planets are `` hot jupiters '' detected by the doppler method .
all that is known about hd 187123b is its orbital period ( @xmath0 days ) and minimum mass ( @xmath1 @xmath2 ) , despite 6 years having elapsed since its discovery .
in contrast , for hd 189733b , discovered only 1.5 yr ago , transit photometry has revealed the planet s radius ( @xmath3 @xmath4 ; bouchy et al .
2005 , bakos et al .
2006a ) and removed the @xmath5 ambiguity in the planet s mass ( @xmath6 @xmath7 ) .
infrared photometry during a secondary eclipse has led to a determination of the planet s 16 @xmath8 m brightness temperature ( @xmath9 ; deming et al .
most recently , spectroscopic observations during a transit have shown that the angle on the sky between the orbit normal and the stellar rotation axis is within a few degrees of zero ( winn et al .
2006 ) .
these measurements are essential for a complete understanding of the atmospheres and interiors of hot jupiters , as well as their formation and migration mechanisms .
an even richer set of measurements can be expected in the future , from investigators pursuing transmission spectroscopy , reflected - light observations , and other transit - related investigations ( as reviewed recently by charbonneau et al .
one goal of the transit light curve ( tlc ) project is to support these efforts by refining the estimates of the planetary , stellar , and orbital parameters , through high - accuracy , high - cadence photometry of exoplanetary transits .
we also seek to measure or bound any variations in the transit times and light - curve shapes that would be caused by the influence of additional bodies in the system ( miralda - escud 2002 ; agol et al .
2005 ; holman & murray 2005 ) . along the way
, we are exploring different techniques for photometry and parameter determination .
previous papers in this series have reported results for the exoplanets xo-1b ( holman et al .
2006a ) , ogle - tr-111b ( winn et al . 2007 ) , tres-1 ( winn , holman , & roussanova 2006 ) , and ogle - tr-10b ( holman et al .
2006b ) .
this paper presents our results for hd 189733b , along with out - of - transit photometry spanning two years .
the reason for gathering out - of - transit photometry was to attempt to measure the stellar rotation period .
the parent star is relatively active , with a chromospheric activity index @xmath10 ( wright et al .
2004 ) and @xmath11 , raising the possibility of measuring the rotation period through starspot - induced quasiperiodic flux variations .
as we will explain , the measurements of both the rotation period and the rossiter - mclaughlin effect permit the determination of the true angle between the orbit normal and stellar rotation axis ( as opposed to the angle between the projections of those vectors on the sky , which was measured by winn et al .
2006 ) .
this paper is organized as follows . in 2 , we present photometry of 8 different transits , along with nightly photometry over two consecutive observing seasons .
in particular , 2.2 presents the out - of - transit photometry and the estimation of the stellar rotation period . in 3 , we describe the parameteric model that was fitted to the data , and in 4 we present the results for the planetary , stellar , and orbital parameters , as well as the new transit ephemerides , a limit on the orbital eccentricity , and the three - dimensional spin - orbit alignment .
the last section is a summary .
our observations took place in 2005 and 2006 using telescopes at 4 different observatories .
each of the sections below presents the photometry from a given observatory . all together , we observed 8 different transits .
table 1 gives a summary of the characteristics of the data from each transit , and table 2 gives the final photometry .
a telescope at fairborn observatory ( 2.2 ) was also used to monitor the out - of - transit flux of hd 189733 over the two observing seasons .
we observed the transits of ut 2006 jul 21 and sep 10 with the 1.2 m telescope at the fred l. whipple observatory on mt .
hopkins , arizona .
we used keplercam , which has one 4096@xmath12 fairchild 486 back - illuminated ccd , with a @xmath13 field of view . for our observations we used @xmath14 binning , which gives a scale of @xmath15 per binned pixel , a readout and setup time of 11 s , and a typical readout noise of 7 e@xmath16 per binned pixel .
we observed both transits through the sdss @xmath17 band filter , in order to minimize the effect of color - dependent atmospheric extinction on the relative photometry and to minimize the effect of limb - darkening on the transit light curve .
we deliberately defocused the telescope such that the full - width at half - maximum ( fwhm ) of stellar images was about 7 binned pixels ( @xmath18 ) in order to permit a consistent exposure time of 5 s. we used automatic guiding to keep the image registration as constant as possible .
we also obtained dome flat exposures and zero - second ( bias ) exposures at the beginning and the end of each night .
on ut 2006 jul 21 , the sky conditions began partly cloudy but gradually improved as the night went on .
we observed the target star as it rose from an airmass of 1.19 to 1.01 and then descended to an airmass of 1.23 .
there was a 15-minute interruption after second contact , due to clouds .
on ut 2006 sep 10 , the sky conditions were mainly clear ( but not all - sky photometric ) .
we followed the target star from an airmass of 1.02 to 2.5 , although we discarded the data taken at airmass @xmath19 because of their much poorer quality .
there was a 15-minute interruption prior to third contact , due to a computer crash , which also caused a change in image registration and focus .
we used standard iraf procedures for the overscan correction , trimming , bias subtraction , and flat - field division .
we performed aperture photometry of hd 189733 and 14 nearby and necessarily fainter stars .
the light curve of each comparison star was normalized to have unit median , and the mean of these normalized light curves was taken to be the comparison signal .
the light curve of hd 189733 was divided by the comparison signal , and corrected for residual systematic effects by dividing out a linear function of time .
the zero point and slope of the linear function were determined as part of the model - fitting procedure , as explained in 4 .
figure 1 shows the final light curves , along with a time - averaged composite light curve created from the two data sets . 0.2 in we used the t10 0.8 m automated photometric telescope ( apt ) at fairborn observatory , in arizona , to observe four complete transits of hd 189733b and to monitor the out - of - transit stellar flux .
the t10 apt is equipped with a two - channel precision photometer employing two emi 9124qb bi - alkali photomultiplier tubes to make simultaneous measurements in the strmgren @xmath20 and @xmath21 passbands .
the apt measures the difference in brightness between a program star and a nearby constant comparison star ( or stars ) with a typical precision of 0.0015 mag for bright stars ( @xmath22 ) . for the hd 189733 transits
, we used the comparison star hd 189410 ( @xmath23 = 5.68 , @xmath24 = 0.34 , f0 ) .
the differential magnitudes were reduced with nightly extinction coefficients and transformed to the strmgren system with yearly mean transformation coefficients . to improve our photometric precision , we combined the separate @xmath20 and @xmath21 differential magnitudes into a single @xmath25 passband . for additional information on the telescopes , photometers , observing procedures , and data reduction techniques ,
see henry ( 1999 ) and eaton , henry , & fekel ( 2003 ) .
we observed the transits of ut 2005 nov 28 , 2006 may 2 , 2006 may 22 , and 2006 jun 11 . on each of those nights
, the differential magnitude of hd 189733 was recorded for 3 - 5 hours bracketing the expected time of mid - transit .
the transit light curves are shown in the top 4 panels of fig .
2 . like the flwo data
, these data have been corrected by a linear function of time that was determined as part of the fitting procedure ( see 4 ) .
in addition , we also measured the out - of - transit flux with the apt on 93 different nights spanning two observing seasons between 2005 october and 2006 july .
three comparison stars were observed on each night .
the out - of - transit flux measurements are shown in the top two panels of fig .
the flux varied erratically during the 2005 observing season and at the beginning of the 2006 season . however , a quasiperiodic signal became evident near the end of our 2006 season observations with a peak - to - peak amplitude of 1.3% and a period of about 13 days , although only 2.5 cycles were observed . this type of photometric behavior
erratic and occasionally quasiperiodic variation is common among chromospherically active stars , especially stars with low or intermediate levels of activity ( see , e.g. , henry , fekel , & hall 1995 ) .
the photometric variability in these stars arises from photospheric starspots and plages that are carried into and out of view by the stellar rotation . in the case of hd 189733 ,
the spots cover only @xmath261% of the stellar surface at any moment , and their distribution on the star changes significantly on the rotational timescale .
a periodogram of the quasiperiodic portion of the light curve ( indicated by the solid dots in the middle panel of fig .
3 ) shows a strong peak at 13.4 days , which we take to be the stellar rotation period .
we estimated the uncertainty by recalculating the light curve using each of the three different comparison stars ; the standard deviation of the results was 0.4 days .
thus , in what follows , we adopt the value @xmath27 days .
since this result is only based on 2.5 cycles , continued observations are warranted to check our estimate
. there may be additional errors because of differential rotation and the variations in spot positions and intensities on the rotational timescale .
the bottom panel of fig .
3 shows the flux as a function of rotational phase during the epoch of quasiperiodic variation .
a periodogram of the entire data set also shows a peak at 13.4 days , along with a peak at 6.7 days , presumably from a time period when star spots occurred on both sides of the star .
we observed the transit of ut 2006 aug 21 with the 2 m telescope at the multi - color active galactic nuclei monitoring ( magnum ) observatory on haleakala , hawaii ( kobayashi et al .
1998 ; yoshii 2002 ; yoshii , kobayashi , & minezaki 2003 ) .
this is the same transit that was observed spectroscopically by winn et al .
we used the multi - color imaging photometer ( mip ) , which allows for simultaneous observation with an optical 1024@xmath12 site ccd and an infrared sbrc insb 256@xmath12 detector . in this case , we used only the optical detector , because the infrared detector was saturated even in very short exposures .
we observed in the @xmath23 band . because the mip field of view is only @xmath28 , and there are no good comparison stars within this small field , we nodded repeatedly between the target star and a calibration star , with 1 s exposures .
the calibration star was hd 190449 ( @xmath23 = 8.12 , @xmath24 = 0.79 , k0 ) .
the average interval between exposures of hd 189733 was 46 s. the median fwhm of stellar images was @xmath29 .
we reduced the images with the standard mip pipeline described by minezaki et al .
we then performed aperture photometry on hd 189733 and hd 190449 , using an aperture radius of @xmath30 and a sky annulus ranging in radius from @xmath30 to @xmath31 . to produce a comparison signal ,
the time series for hd 190449 was boxcar - smoothed ( with a width of 4 points , or 3 minutes ) and then linearly interpolated onto the time stamps of the hd 189733 data .
then the hd 189733 time series was divided by this comparison signal .
a few extreme outlying points were rejected . to remove residual systematic errors in the out - of - transit flux determination ,
we divided by a linear function of time that was determined as part of the fitting procedure ( see 4 ) .
the final light curve is shown in the fifth panel of fig . 2 .
we observed the transit of 2006 sep 5 with the 1 m telescope at wise observatory , in israel .
we used a tektronix 1024@xmath12 back - illuminated ccd detector , giving a pixel scale of @xmath32 and a field of view of @xmath33 .
we observed through a johnson @xmath34 filter , the reddest optical band available on this camera .
the exposure time was 10 s and the telescope was defocused in order to avoid saturation .
autoguiding was used to keep the image registration constant throughout the night .
we also obtained sky flat exposures at sunset and zero - second ( bias ) exposures at the beginning and during the night .
we performed the data reduction and photometry using very similar procedures to those that were used on the flwo data ( 2.1 ) .
the bottom panel of fig . 2 shows the final light curve .
to estimate the planetary , stellar , and orbital parameters , and the times of transit , we fitted a parameterized model to the transit photometry . the model and
the fitting method were similar to those described in previous tlc papers ( see , e.g. , winn et al .
2007 ) , except that in this case we accounted for correlated noise , as described below .
the model is based on a keplerian orbit of a star ( with mass @xmath35 and radius @xmath36 ) and a single planet ( @xmath37 , @xmath38 ) about their center of mass . for most of the analysis we assumed that the orbital eccentricity @xmath39 is zero , because the expected timescale of tidal circularization is short in the absence of excitations from other planets ( see , e.g. , rasio et al .
1996 , trilling et al .
2000 , dobbs - dixon et al .
2004 , adams & laughlin 2006 ) .
however , in 4.3 , we discuss the empirical upper limits on the orbital eccentricity .
the orbit has a period @xmath40 and an inclination @xmath41 relative to the sky plane .
we define the coordinate system such that @xmath42 .
it is often useful to refer to the impact parameter @xmath43 ( where @xmath44 is the semimajor axis ) rather than the inclination . because one of our goals was to measure the individual transit times , we allowed each transit to have an independent value of @xmath45 , the transit midpoint , rather than forcing them to be separated by exact multiples of the orbital period . thus , the only effect of @xmath40 on the model is to determine the semimajor axis @xmath44 for a given value of the total mass .
we fixed @xmath46 days , the value determined by bouchy et al .
( 2005 ) and hbrard & lecavelier des etangs ( 2006 ) from the detection of transits in the _ hipparcos _ database .
the uncertainty of @xmath47 days was negligible for our purposes , although we were able to use the resulting values of @xmath45 to produce an independent estimate of the period , as described in 5 .
neither @xmath35 nor @xmath37 can be determined from photometry alone . as we have done in previous tlc analyses
, we fixed @xmath35 at a value that is based on an analysis of the stellar spectrum and other observable properties .
we then used the scaling relations @xmath48 and @xmath49 to estimate the systematic error associated with the uncertainty in @xmath35 . in this case
we adopted the value @xmath50
@xmath51 ( bouchy et al . 2005 ) .
the planetary mass @xmath37 hardly affects the photometric model at all , but for completeness we used the value @xmath52
@xmath7 ( winn et al . 2006 ) . to calculate the relative flux as a function of the projected separation of the planet and the star
, we employed the analytic formulas of mandel & agol ( 2002 ) to compute the integral of the intensity over the unobscured portion of the stellar disk .
we assumed the limb darkening law to be quadratic , @xmath53 where @xmath34 is the intensity , and @xmath8 is the cosine of the angle between the line of sight and the normal to the stellar surface .
we fixed the limb - darkening coefficients at the values calculated by claret ( 2000 , 2004 ) for observations of a star with the observed spectral properties .
k , @xmath54 ( cgs ) , log [ m / h]@xmath55 and @xmath56 km s@xmath57 . for the @xmath17 band , @xmath58 and @xmath59 . ]
we also investigated the effect of fitting for the limb darkening parameters , as discussed below .
in addition , the light curves exhibited gradients in the out - of - transit data , probably due to differential extinction between the target star and the comparison stars , or some other systematic error .
for this reason , each of the 8 data sets was modeled with two extra parameters : the out - of - transit flux @xmath60 and a time gradient @xmath61 .
the fitting statistic was @xmath62 ^ 2 , \label{eq : chi2}\ ] ] where @xmath63(obs ) is the flux observed at time @xmath64 , @xmath65 controls the relative weights of the data points , and @xmath63(calc ) is the calculated value . in order to derive realistic uncertainties on the parameters , it is important for @xmath65 to include not only measurement errors but also any unmodeled systematic effects .
of particular importance is the timescale of the systematic effects .
correlated noise effectively reduces the number of independent data points and correspondingly increases the uncertainties in the model parameters , an issue that pont et al .
( 2006 ) and gillon et al .
( 2006 ) have recently brought to attention in the context of transit photometry .
our approach to this problem was as follows .
first , for each of the 8 transits , we rescaled the instrumental uncertainties such that @xmath66 for the best - fitting model .
the resulting uncertainties are those that are given in table 2 .
second , we followed the procedure of gillion et al .
( 2006 ) to decompose the observed noise into `` white noise '' ( that which averages down as @xmath67 , where @xmath68 is the number of data points ) and `` red noise '' ( that which does not average down over some specified time interval ) . specifically , we calculated the standard deviation of the residuals ( @xmath69 ) and the standard deviation of the time - averaged residuals ( @xmath70 ) .
the averaging time was 1 hr ( a timescale comparable to the transit event ) , corresponding to a number @xmath68 of data points that depended upon the cadence of observations .
then we solved for the white noise @xmath71 and red noise @xmath72 from the system of equations @xmath73 finally , to account approximately for the effective reduction in the number of independent data points , we rescaled the @xmath65 in eq .
( [ eq : chi2 ] ) by the factor @xmath74 .
the results for @xmath71 , @xmath72 , and the rescaling factor for each data set are given in table 1 .
these results are not very sensitive to the choice of averaging time .
any choice between 15 min and a few hours gave similar results . among the data sets are wide disparities in the degree of red noise , ranging over a factor of 10 . by far the best data , in the sense of the smallest noise correlations , are from the flwo 1.2 m telescope and keplercam ( 2.1 ) .
for this reason , we decided to estimate the system parameters using _ only _ the flwo data , and use the other data sets only to determine transit times and as a consistency check on the flwo results .
( had we included the other data sets , their statistical weight would anyways have been much smaller . ) in short , we used the flwo data to solve for the two bodies radii ( @xmath36 and @xmath38 ) ; the orbital inclination ( @xmath41 ) ; and the mid - transit time ( @xmath45 ) , the out - of - transit flux ( @xmath60 ) , and a time gradient ( @xmath61 ) for each of the 2 transits .
we then fixed @xmath36 , @xmath38 , and @xmath41 at the best - fitting values , and fitted each of the remaining 6 data sets to find @xmath45 , @xmath60 , and @xmath61 . to solve for the model parameters and their uncertainties
, we used a markov chain monte carlo algorithm ( see , e.g. , tegmark 2004 ) .
our jump function was the addition of a gaussian random number to each parameter value .
we set the perturbation sizes such that @xmath2620% of jumps are executed .
we created 10 independent chains , each with 500,000 points , starting from random initial positions , and discarded the first 20% of the points in each chain .
the gelman & rubin ( 1992 ) @xmath75 statistic was within 0.2% of unity for each parameter , a sign of good mixing and convergence .
we merged the chains and took the median value of each parameter to be our best estimate , and the standard deviation as the 1 @xmath69 uncertainty .
the results are given in table 3 .
along with the results for the model parameters , we have provided results for some useful derived quantities such as the impact parameter @xmath20 , the radius ratio @xmath76 , and the fraction @xmath77 ( which gives the fraction of starlight reflected by the planet , for an albedo of unity ) .
we also report the calculated values of the full transit duration ( the time between first and fourth contact , @xmath78 ) , and the partial transit duration ( the time between first and second contact , or between third and fourth contact ) . in table 3 , all of the results in table 1 are based on the entire light curves , including both ingress and egress data .
our model assumes @xmath79 . ]
the result for the stellar radius is @xmath80 .
the uncertainty is dominated by the statistical error of 0.023 @xmath81 ( 3.1% ) .
the covariance with the uncertainty in the stellar mass produces an additional error of 0.009 @xmath81 ( 1.2% ) , which we have added in quadrature to the statistical error to arrive at the net uncertainty of 0.025 @xmath81 .
we find the planetary radius to be @xmath82 , where the uncertainty is again dominated by the statistical error of @xmath83 . to test the robustness of these results , we performed some additional fits .
we gauged the importance of the choice of limb - darkening law by re - fitting the data under different assumptions . when we allowed the limb - darkening coefficients @xmath84 and @xmath85 to be free parameters rather than holding them fixed , we found @xmath86 and @xmath87 , well within the 1 @xmath69 error of our original analysis .
( the optimized limb - darkening coefficients for the @xmath17 band were @xmath88 and @xmath89 , as compared to the theoretical values of @xmath90 and @xmath91 . )
likewise , the results changed by only 0.25 @xmath69 when we used a linear limb - darkening law ( @xmath92 ) , regardless of whether the linear limb - darkening coefficient was fixed or taken to be a free parameter . as another check , we fitted each of the 8 transit data sets separately , solving for the parameters @xmath93 in each case .
although the statistical power of the flwo photometry was the greatest , as noted previously , all 8 sets of results agreed within their calculated error bars . in particular , the unweighted `` ensemble averages '' of the results from the 6 non - flwo data sets were @xmath94 and @xmath95 , again in agreement with our original analysis .
how do these results compare to the previous analyses of hd 189733 by bouchy et al .
( 2005 ) and bakos et al .
( 2006a ) ?
an important difference is that those authors decided not to determine @xmath36 from the transit photometry .
, they did not trust the result .
they quite reasonably suspected that the true error bar was significantly larger than their calculations indicated , because of correlated noise in the data . ] rather , those authors used estimates of @xmath36 based on an analysis of other observable properties of the star .
bouchy et al .
( 2005 ) used measurements of the star s parallax , effective temperature , surface gravity , and metallicity , in comparison with the outputs of stellar evolution models , and concluded @xmath96 .
likewise , bakos et al . ( 2006a ) investigated four different ways of determining the stellar radius , based on broad band colors , spectral properties , and model isochrones , and found a stellar radius in the range 0.740.79 @xmath81 .
ultimately , bakos et al . ( 2006a ) adopted the value @xmath97 based on a calibration of 2mass photometry by masana et al .
our determination of @xmath36 based on the transit light curve is in agreement with those independent determinations ; the mutual agreement constitutes an important consistency check on the data and our analysis .
our value for the planetary radius agrees with the value @xmath98 found by bakos et al .
( 2006a ) .
while the bakos et al .
( 2006a ) result would appear to be more precise , the comparison is somewhat misleading because bakos et al .
( 2006a ) were not simultaneously fitting for @xmath36 , as noted above .
if we follow their procedure of fixing @xmath99 , then our result for the planetary radius becomes more precise : @xmath100 . both our result and that of bakos et al .
( 2006a ) disagree with that of bouchy et al .
( 2005 ) , who found @xmath101 based on a @xmath102-band light curve from the 1.2 m telescope at the observatoire de haute - provence ( ohp ) .
this discrepancy was traced by bakos et al .
( 2006a ) to systematic errors in the ohp photometry .
when the @xmath102 band light curve was recalculated using a greater number of comparison stars , the transit depth decreased by 20% and the inferred planetary radius shrank accordingly .
how does the planetary radius compare to theoretical expectations , given its mass ( @xmath103 ; winn et al .
2006 ) and its proximity to its parent star ( @xmath104 au ) ?
fortney , marley , & barnes ( 2006 ) have recently provided a wide range of theoretical predictions for exoplanet radii .
their calculations are for a solar - mass star , but at their suggestion we can rescale the semimajor axis to compensate for the lower luminosity of hd 189733 . assuming @xmath105 ,
then if hd 189733 were orbiting the sun at @xmath106 it would receive roughly the same incident flux as it does in its actual orbit . the resulting prediction from fig .
6 of fortney et al .
( 2006 ) is a planetary radius between @xmath1071.051.12 @xmath4 , assuming an age of @xmath264.5 gyr and depending on whether or not the planet has a massive ( @xmath108 ) core .
the core - free prediction for the radius is larger and in agreement with the observed value .
however , it seems premature to claim that a massive core is disfavored , given the uncertainties that enter into the calculations and the uncertainty in the age of the system .
it does seem safe to say that the radius of hd 189733 does not present a severe theoretical problem , unlike the cases of the apparently `` bloated '' planets hd 209458b , hat - p-1b , and wasp-1b ( for recent results on those systems , see knutson et al .
2006 , bakos et al .
2006b , collier - cameron et al .
2006 , and charbonneau et al . 2006b ) .
the mean density of hd 189733b is @xmath109 g @xmath110 , which is between the densities of saturn ( 0.6 g @xmath110 ) and jupiter ( 1.2 g @xmath110 ) .
the surface gravity of hd 189733b is @xmath111 m s@xmath112 , which is also intermediate between saturn ( 10 m s@xmath112 ) and jupiter ( 25 m s@xmath112 ) .
we note that whenever @xmath37 is measured via the spectroscopic orbit of the star , and @xmath38 is measured via transit photometry ( as is the case here ) , then the derived value of @xmath113 is immune to systematic errors in the parameters of the parent star .
this is because the fitting degeneracies are @xmath114 and @xmath48 , and hence @xmath115 is independent of @xmath35 .
here this fact is of limited interest , because the error in @xmath38 is dominated by statistical error , but it may be of importance in future transit studies .
table 4 gives the 8 transit times measured from our data .
we have used these times , along with transit times previously measured by bakos et al .
( 2006 ) , to calculate a photometric ephemeris for this system .
although bakos et .
( 2006 ) reported 15 measured times , we used only 4 of those data points in our analyis .
we did not include the 10 times that were based on only partial observations of the transit .
full transits are greatly preferable , in order to correct ( or at least assess ) systematic errors using the pre - ingress and post - egress data . in addition , we did not include the @xmath45 measurement based on the ohp @xmath102-band light curve of bouchy et al .
( 2005 ) because of the systematic errors noted previously .
what remained were 4 data points representing 4 independent measurements of the same event . in combination with our 8 data points , we fitted a linear function of transit epoch @xmath116 , @xmath117 finding @xmath118 [ hjd ] and @xmath119 days , where the numbers in parentheses indicate the @xmath120 uncertainty in the final two digits .
the fit had @xmath121 and @xmath122 .
we chose @xmath123 to correspond to the most precisely known transit time .
our derived period agrees almost exactly with the value @xmath124 days determined by bakos et al .
( 2006a ) , and it is also in agreement with the _ hipparcos_-based values of @xmath125 days ( bouchy et al .
2005 ) and @xmath126 days ( hbrard & lecavelier des estangs 2006 ) .
figure 4 is the o@xmath127c ( observed minus calculated ) diagram for the transit times , according to this new ephemeris .
the filled symbols represent data points used in the fit .
there is not yet any pattern in the residuals that would indicate the effect of a perturbing body in the system .
the unfilled square shows the ohp @xmath102-band measurement , which is indeed an outlier .
the unfilled circle is explained in the next section . 0.2 in as mentioned previously , one would expect the orbit of a hot jupiter such as hd 189733b to be very nearly circular , due to tidal effects .
previous results and our results have all shown that a circular orbit does indeed provide a satisfactory description of the available data . however , it is still interesting to make an empirical determination of the eccentricity , both in the spirit of `` what one can measure , one should measure '' and also because any additional bodies in the system could excite the orbital eccentricity .
we used a two - step procedure to determine the orbital eccentricity .
first , we used our revised ephemeris to interpret the secondary - eclipse timing of deming et al .
( 2006 ) . for a nonzero
( but small ) orbital eccentricity , the time difference between the midpoint of secondary eclipse , @xmath128 , and the time of transit , @xmath129 , may differ from half of the orbital period : @xmath130 where @xmath131 is the argument of pericenter .
deming et al .
( 2006 ) measured the midpoint of a secondary eclipse at hjd @xmath132 , which is consistent with the @xmath133 prediction of our ephemeris . the unfilled circle in fig .
4 represents the secondary eclipse measurement .
the timing offset of eq .
( [ eq : ecosw ] ) is @xmath134 minutes , corresponding to @xmath135 .
second , to determine the other component @xmath136 of the eccentricity vector , we used the radial velocities presented by winn et al .
we used only those 60 velocities measured outside of transits ( i.e. , not affected by the rossiter - mclaughlin effect ) .
we performed an mcmc analysis to solve for the keplerian orbital parameters , as well as a possible long - term velocity gradient , using the same treatment of the measurement errors that winn et al .
( 2006 ) applied to the entire data set .
we allowed both @xmath39 and @xmath131 to be free parameters , but with an a priori constraint on @xmath137 to enforce compliance with the secondary - eclipse measurement .
the result for @xmath136 was @xmath138 .
hence , both components of the eccentricity vector are consistent with zero , and @xmath137 is about 20 times more tightly bounded than @xmath136 .
values of @xmath39 as large as @xmath1070.02 are allowed , but only for @xmath131 very close to @xmath139 .
thanks to the apt data ( 2.2 ) ,
hd 189733 is the first star with a measured rotation period that also has a transiting planet .
together with the transit photometry and the observation of the rossiter - mclaughlin effect , this allows for the determination of the angle between the stellar rotation axis and the planetary orbit normal , as anticipated by queloz et al .
this angle is worth measuring because any significant misalignment may be an indication of perturbative effects during planetary migration , among other reasons ( as explained in more detail by ohta et al .
2005 , winn et al . 2005 and gaudi & winn 2007 ) .
the true ( three - dimensional ) angle @xmath140 between the stellar spin axis and the orbital axis is given by the formula @xmath141 where @xmath41 is the orbital inclination , @xmath142 is the inclination of the stellar rotation axis , and @xmath143 is the angle between the sky projections of the two axes .
( for a diagram of the coordinate system , see fig .
3 of ohta et al . 2005 . )
the transit photometry determines @xmath41 with excellent accuracy .
observations of the rossiter - mclaughlin effect have been used to determine @xmath143 , but they can not be used to determine @xmath142 independently . rather , they are sensitive to @xmath144 , the projected rotation rate of the star . given @xmath145 from the rossiter - mclaughlin observations , from an analysis of the width of photospheric absorption lines . ] , along with @xmath41 and @xmath36 from the transit photometry , and the stellar rotation period @xmath146 , one can determine @xmath147 via the formula @xmath148 hence , all of the angles in eq .
( [ eq : angles ] ) are known . using the values @xmath149 , and @xmath150 km s@xmath57 from winn et al .
( 2006 ) , we found @xmath151 . by rejecting values of @xmath152 as unphysical , and propagating the errors through eq .
( [ eq : angles ] ) , we determined an upper bound on the ( mis)alignment angle @xmath140 of @xmath153 with 95% confidence .
essentially the same result can be obtained from the approximation @xmath154 , which is valid because @xmath155 and @xmath156 are tightly constrained .
this is the first exoplanetary system for which it has been possible to measure @xmath140 .
the result is consistent with zero , but it is not as precise as the result for the projected angle @xmath143
. how could the measurement of @xmath140 be improved ?
we have already mentioned some caveats relating to the measurement of the rotation period days . ]
( 2.2 ) , but the current uncertainty in @xmath140 is dominated by the error in @xmath144 which is itself dominated by systematic errors arising from the interpretation of the transit spectra ( see winn et al .
specifically , the systematic error arose from the `` calibration '' the rossiter - mclaughlin effect using simulated spectra , which was needed because the transit spectra were observed through an i@xmath157 cell and analyzed with an algorithm that is nominally designed for measuring doppler shifts rather than spectral distortions .
further improvement might be achieved through a more sophisticated set of simulations or perhaps by re - observing the rossiter - mclaughlin effect without the i@xmath157 cell ( i.e. , using a different technique to account for instrumental variations ) .
we have presented photometry of 8 complete transits of the exoplanet hd 189733b , and modeled the light curves in order to determine the radii of the star and the planet .
our results are consistent with previous results and with theoretical expectations for close - in jovian planets .
stringent limits on the orbital eccentricity follow from the measured transit times , in conjunction with a previous detection of the secondary eclipse and with the spectroscopic orbit .
we have also presented nightly out - of - transit photometry spanning 2 yr that has revealed the stellar rotation period .
we have used this information , along with a previous analysis of the spectroscopic transit , to place an upper bound on the true angle between the stellar rotation axis and the orbital axis . with these developments ,
hd 189733b has become one of the most thoroughly characterized planets outside of the solar system .
we thank f. pont and m. gillon for helpful discussions about correlated noise .
we are grateful to g. marcy , p. butler , s. vogt , and e. turner for their help with the doppler analysis and for encouragement . a.r .
thanks the mit urop office for research funding .
2005 nov 28 & t10 apt 0.8 m & @xmath25 & 1.44 & 0.0045 & 0.00250 & 4.2 + 2006 may 2 & t10 apt 0.8 m & @xmath25 & 1.44 & 0.0024 & 0.00080 & 2.8 + 2006 may 22 & t10 apt 0.8 m & @xmath25 & 1.44 & 0.0024 & 0.00090 & 3.0 + 2006 jun 11 & t10 apt 0.8 m & @xmath25 & 1.30 & 0.0019 & 0.00140 & 5.3 + 2006 jul 21 & flwo 1.2 m & @xmath17 & 0.23 & 0.0023 & 0.00035 & 3.3 + 2006 aug 21 & magnum 2 m & @xmath23 & 1.38 & 0.0029 & 0.00100 & 2.9 + 2006 sep 5 & wise 1 m & @xmath34 & 0.73 & 0.0029 & 0.00120 & 4.6 + 2006 sep 10 & flwo 1.2 m & @xmath17 & 0.23 & 0.0022 & 0.00020 & 2.0 @xmath158 & @xmath159 & @xmath160 + @xmath161 & @xmath162 & @xmath163 + @xmath164 & @xmath159 & @xmath165 + @xmath166 & @xmath162 & @xmath167 + @xmath168 & @xmath169 & @xmath170 + @xmath77 & @xmath171 & @xmath172 + @xmath173 & @xmath174 & @xmath175 + @xmath41 [ deg ] & @xmath176 & @xmath177 + @xmath20 & @xmath178 & @xmath179 + @xmath180 [ hr ] & @xmath181 & @xmath182 + @xmath183 [ min ] & @xmath184 & @xmath185 + @xmath186 & @xmath187 & @xmath188 + @xmath189 & @xmath190 & @xmath191 + @xmath192 & @xmath193 & @xmath194 + @xmath195 & @xmath196 & @xmath197 + @xmath198 & @xmath199 & @xmath200 + @xmath201 & @xmath202 & @xmath203 + @xmath204 & @xmath205 & @xmath203 + @xmath206 & @xmath207 & @xmath208 + | we present photometry of hd 189733 during eight transits of its close - in giant planet , and out - of - transit photometry spanning two years .
using the transit photometry , we determine the stellar and planetary radii and the photometric ephemeris . outside of transits , there are quasiperiodic flux variations with a 13.4 day period that we attribute to stellar rotation . in combination with previous results ,
we derive upper limits on the orbital eccentricity , and on the true angle between the stellar rotation axis and planetary orbit ( as opposed to the angle between the projections of those axes on the sky ) . |
laser cooling and trapping of neutral atoms plays important role for many directions of modern quantum physics .
one of the directions is quantum metrology that experiences galloping progress nowadays .
it is aimed on producing of standards of various physical quantities and on carrying out precise measurements with the help of them ( e.g. , see @xcite ) . at present
the most precise measurements are possible for the physical quantities as frequency and time .
it is due to success achieved in producing etalons ( standards ) for these quantities .
contemporary time standard is based on a frequency standard , which defines its stability and accuracy to a considerable degree . at that
, frequency etalons can be used not only as a basis for time standards , but also for conducting precise measurements of other physical quantities and constants as , for instance , electrical current and voltage , magnetic field , length , rydberg and fine - structure constants .
high - accuracy experiments for versatile examination of relativistic and quantum theories have become feasible owing to modern frequency standards . among practical applications of time and frequency standards
the broadband communication networks , navigational and global positioning systems should be mentioned especially .
many laboratories in word - known scientific centers do research in the field of frequency standards .
one of the latest trends in this field is connected with the concept of intercity or even international quantum clock network that could combine time etalons from various laboratories and countries into one system @xcite .
there are two main directions of primary frequency standards development : the technology of a single ion confined in a electro - quadrupole trap and the second one based on many neutral atoms trapped in an optical lattice ( e.g. , see @xcite ) .
the latter direction is much newer than the former and it undergoes intense progress .
the idea of neutral atoms trapping in a periodic light potential is not new and it was actively studied in 1970s ( see monograph @xcite and citations in it ) . as for metrological purposes this idea has experienced the second birth in the beginning of xxi century after noticeable progress in technique and methods of laser cooling of atoms , development of the `` magic''-wavelength concept @xcite , and also experimental and theoretical success in the field of spectroscopy of forbidden atomic transitions @xcite . at present , stability of optical - lattice - based frequency standards
is on the same level with single - ion standards and in some cases even better .
the state - of - the - art prototypes reached instability and uncertainty on the relative levels of @xmath14@xmath15@xmath16 @xcite .
ones of the main candidates for producing the new - generation frequency standards are alkaline earth and alkaline - earth - like atoms : yb ( for instance , see @xcite ) , ca @xcite , sr @xcite , hg @xcite and mg @xcite .
these atoms are the most appropriate because of narrow spectroscopic lines due to forbidden optical transitions from the ground state @xmath17s@xmath18 to the lowest excited triplet state @xmath19p@xmath20 ( see fig .
[ fig1 ] ) . moreover , one more key circumstance consists in the existence of so - called `` magic '' wavelength for these transitions . under the magic - wavelength optical field
the linear ( in the intensity ) light shift is canceled .
also , one of the last tendency in this field is connected with spectroscopy of transition @xmath17s@xmath18@xmath3@xmath19p@xmath18 in even isotopes ( with zero nucleus spin ) , which is highly forbidden .
frequency of this transition is immune to many frequency - shift effects , therefore this transition can be exploited as a good `` clock '' transition . in spite of the transition is highly forbidden , it has already been observed by means of magnetic - field - induced spectroscopy @xcite in @xmath21yb @xcite , @xmath22sr @xcite and @xmath0 mg @xcite . to date
, atoms of the first four elements ( yb , ca , sr and hg ) can be effectively cooled by the laser methods down to ultralow temperatures , approaching to the recoil energy limit , what is required for effective loading an optical lattice ( @xmath23 110 @xmath13k ) . besides , sub - recoil temperatures can be obtained by using evaporative cooling technique , getting bose - einstein condensation @xcite .
unfortunately , researchers have not been able to reach the same great success with mg atoms .
in particular , neither two - photon laser cooling @xcite nor laser quenching @xcite methods have appeared to be ineffective in the case with magnesium .
the minimum temperature of a magnesium cloud that has been obtained by laser cooling method is about 500 @xmath13k , what is rather far from desirable range of values , in particular , from the recoil temperature ( 310 @xmath13k , depending on an atomic transition ) . at the same time
, magnesium atom has some advantages with respect to the other candidates for the frequency standard .
thus , black - body radiation ( bbr ) shift is one of the main limiting factors for accuracy and stability of quantum frequency standard ( e.g. , see @xcite ) .
bbr shift of the clock transition @xmath24s@xmath18@xmath3@xmath1p@xmath18 for magnesium is much smaller than for yb , ca , sr and just a little bit higher than for a mercury atom ( see tab.1 ) .
however , from an experimental viewpoint mg has some advantages against hg .
in particular , mercury atom requires noticeably smaller wavelengths for laser spectroscopy , cooling and trapping than mg .
for instance , effective trapping of cold atoms in a nondissipative optical lattice and producing lamb - dicke regime need for an optical potential depth at the level of 50300 in the recoil energy units ( e.g. , see @xcite ) .
it means that a highly intensive laser field at the magic wavelength @xmath25 should be applied .
but , as it can be seen from table 1 , @xmath25(mg)@xmath26@xmath27 nm and @xmath25(hg)@xmath26@xmath28 nm .
therefore , from an experimental viewpoint , producing deep optical potential for mercury atom is more difficult problem than for magnesium one due to the much smaller wavelength .
besides of relatively small bbr shift , magnesium atom has one more advantage with respect to ca and sr .
it consists in absence of optical pumping of atoms on the non - resonant level @xmath24d@xmath2 during the realization of laser precooling with the help of strong dipole transition @xmath24s@xmath18@xmath3@xmath24p@xmath29 ( see fig .
[ fig1 ] ) to reach the temperature down to a few millikelvins @xcite .
[ tab1 ] .data for several atomic elements relevant for new - generation frequency standards : @xmath30 is a wavelength of the clock transition @xmath24s@xmath18@xmath3@xmath1p@xmath18 and @xmath25 is its magic wavelength , bbr frequency shifts are indicated with respect to absolute frequencies of clock transition ( @xmath30 is taken from nist atomic spectra database ) . [ cols="^,^,^,^",options="header " , ] recent experiments @xcite showed some progress in cooling of @xmath0 mg atoms .
the atoms were cooled down to the record temperature equaled to 1.3 @xmath13k and confined in an optical lattice .
however , the final number of atoms was of the order of @xmath31 , what was about 0.01% from the initial number of atoms in a magneto - optical trap ( mot ) , involved the cyclic triplet dipole transition @xmath1p@xmath2@xmath3@xmath1d@xmath4 . that great loss in atomic number was due to the fact that velocity selection technique , having similarities with evaporative cooling , was used for reaching such ultralow temperature , but the laser cooling method in mot , unfortunately , showed the cloud temperature equaled to only about 1 mk .
that result noticeably yielded to successful results with the other atomic elements ( ca , sr , yb , hg ) . at the same time
, we believe that laser cooling strategy for magnesium atoms can be proper tuned for getting much better result of laser - cooling temperature as well as number of atoms trapped . therefore , we can state that problem of deep cooling of magnesium atoms by means of laser radiation is still unsolved .
moreover , increasing of ultracold atomic number has principal importance for many applications of cold atoms .
for instance , authors of the paper @xcite managed to obtain bose - einstein condensation composed of @xmath23@xmath32 strontium atoms .
besides , frequency - standard stability depends on an atomic number in an optical lattice and it increases with the number increases @xcite .
all these things considered , we can conclude that it is important to solve the problem of deep laser cooling of magnesium atoms ( down to t@xmath23110 @xmath13k ) as well as to provide much larger number of ultracold atoms in a lattice .
laser cooling of neutral atoms in a magneto - optical trap is one of the main cooling methods . at first time
the laser field composed of six beams with orthogonal circular polarizations ( @xmath33 configuration ) was suggested in @xcite as effective way for simultaneous cooling and trapping of atoms .
narrow spectral lines allow laser cooling of various atoms down to a few tens and units of microkelvin , and even lower .
in particular , narrow intercombination transition @xmath34s@xmath18@xmath3@xmath35p@xmath29 in @xmath36ca ( @xmath37@xmath26@xmath38@xmath39@xmath40hz ) provides temperatures around 46 @xmath13k @xcite just by the help of doppler cooling process .
there were good results with intercombination transitions also for the other elements : sr @xcite , yb @xcite and hg @xcite , for even as well as for odd isotopes .
in certain aspects even isotopes , having a zero nuclear spin , are more attractive for frequency standards of new generation .
however , as it has been already noted in the introduction , still there are not satisfactory results of cooling @xmath0 mg atoms by means of laser radiation in contrast to the other elements . the first our attempt to solve the problem with deep laser cooling of magnesium was undertook in the recent work @xcite , where the detailed theoretical study of magnesium kinetics in 1d mot using the dipole transition @xmath1p@xmath2@xmath3@xmath1d@xmath4 was conducted .
the theory was based on the semiclassical approach @xcite , based on the well - known assumptions : @xmath41 and @xmath42 here @xmath43@xmath44@xmath45@xmath46@xmath47 is the recoil frequency , @xmath48 is mass of an atom , @xmath49@xmath44@xmath50@xmath51@xmath52 is wave number .
the saturation parameter @xmath53 is defined as @xmath54 where @xmath37 is the spontaneous relaxation rate of excited state , @xmath55@xmath44@xmath56@xmath15@xmath57 is the detuning of laser radiation frequency @xmath56 from the transition frequency @xmath57 , and @xmath58 is the rabi frequency . condition ( [ eq1 ] ) implies that recoil frequency must be rather small in comparison with a typical rate of establishment of steady state among atomic internal degrees of freedom . in particular , in the case of an atom without any degeneracy of the ground state this rate is defined by @xmath37 .
if there is a degenerate ground state and optical pumping can occurs , this rate is defined by @xmath37 or the pumping rate @xmath37@xmath53 , depending on what is smaller . the second semiclassical requirement ( [ eq2 ] )
implies that typical width of stationary linear momentum distribution @xmath59 must be much larger than the recoil momentum from emmision / absorption of a photon .
doppler limit for temperature of laser cooling @xmath60 , which can be achieved at the frequency detuning @xmath55@xmath44@xmath61@xmath62 , can be figured out from equation for minimum kinetic energy in one - dimensional case : @xmath63 strictly speaking , this equation is valid for transition @xmath64@xmath44@xmath65@xmath3@xmath66@xmath44@xmath67 .
it was found in @xcite under @xmath7@xmath8 configuration ( also see @xcite ) .
if we use this formula for getting estimate of @xmath60 in the case of transition @xmath1p@xmath2@xmath3@xmath1d@xmath4 ( @xmath37@xmath26@xmath38@xmath68mhz ) , we immediately find @xmath60@xmath69@xmath70@xmath13k . for effective trapping of atoms with such relatively high temperature the large intensity of cw optical lattice field at the level of tens of mw/@xmath71
is required , what is hardly feasible in an experiment .
therefore , much lower temperature of atomic cloud is needed . at the same time , since the transition considered has degenerate energy levels , one can anticipate activation of so - called sub - doppler mechanism during the laser cooling in mot under the polarization - gradient field . in principal
, this process would overcome the doppler limit ( [ dopplerlimit ] ) and show much lower temperature than in the case of @xmath64@xmath44@xmath65@xmath3@xmath66@xmath44@xmath67 .
semiclassical approach is based on kinetic equation of fokker - planck type on the wigner distribution function in phase space @xmath72 .
that equation can be acquired by reducing of exact quantum kinetic equation on the density matrix in the series on small parameter @xmath73@xmath74@xmath75@xmath67 until second - order terms .
this procedure is well - known and it has been done by many authors ( e.g. , see @xcite ) .
eventually the following equation can be obtained : @xmath76\,f(z , p).\ ] ] here @xmath77 is laser - field force on an atom , @xmath78 is diffusion of an atom in the light field .
this equation must be completed with normalizing condition that in one - dimensional periodic laser field has the form : @xmath79 one - dimensional @xmath33 laser - field configuration allows significant simplifying of ( [ semiclassics ] ) , at that the dependence @xmath80 on @xmath81 vanishes ( see section [ firststage ] ) .
our semiclassical calculations @xcite have been done beyond many widely used approximations ( for instance , slow atoms and weak field approximations ) .
as it has been shown the minimum kinetic energy achievable in mot is close to 30@xmath39@xmath82 , where @xmath82@xmath44@xmath83 is the recoil energy . the effective temperature , which can be associated to this value , @xmath84@xmath26@xmath85@xmath12k .
it is approximately three times lower than the estimate of doppler limit @xmath60@xmath26@xmath86@xmath12k , but , unfortunately , it is still very far from desirable range of values and , in particular , the recoil temperature @xmath87@xmath44@xmath88@xmath12k .
let us consider a question on validity of semiclassical approach to the magnesium problem .
indeed , as it will be shown further on the basis of semiclassical treatment , the optimal parameters of cooling field for transition @xmath1p@xmath2@xmath3@xmath1d@xmath4 can be chosen as @xmath55@xmath44@xmath89@xmath39@xmath90 mhz and @xmath91@xmath44@xmath92mw/@xmath71 .
the corresponding saturation parameter @xmath53@xmath26@xmath93@xmath39@xmath94 . in spite of such low saturation ,
the first semiclassical requirement ( [ eq1 ] ) is still satisfied , because @xmath43@xmath44@xmath50@xmath39@xmath95@xmath96 .
at the same time , typical momentum distribution width @xmath97 may not satisfied the second semiclassical condition ( [ eq2 ] ) .
indeed , in general case the distribution can have complex shape .
in particular , fig .
[ fig2 ] shows two examples of momentum distributions for different values of light field intensity .
the distribution acquires two - peaked profile at low intensity @xmath91@xmath44@xmath98mw/@xmath71 : there is the high - contrast spike on top of the wide background .
this background conditionally describes `` hot '' fraction of atoms in a cloud with effective temperature @xmath84@xmath23@xmath99@xmath15@xmath100 mk , while the spike corresponds to ultracold fraction with @xmath84@xmath23@xmath101k .
similar distributions were observed earlier ( e.g. , see @xcite with semiclassical low - saturation - limited calculations of sisyphus cooling of cs atoms with the help of transition @xmath102@xmath44@xmath93@xmath3@xmath103@xmath44@xmath88 or the quantum - treatment calculations for atomic w - type scheme in @xcite ) . in our case
the narrow - spike width is about @xmath73 and the requirement @xmath97@xmath104@xmath73 is not satisfied at all . with increasing the intensity ( @xmath91@xmath44@xmath105mw/@xmath71 ) the two - peaked shape disappears .
however , the distribution as a whole is still sufficiently narrow and the second requirement ( [ eq2 ] ) is satisfied with a good margin .
also it should be noted that the condition ( [ eq2 ] ) , as a matter of fact , depends on the value of total angular momentum @xmath102 . in other words , at the same saturation parameter @xmath53 the requirement ( [ eq2 ] ) can be valid for small @xmath102 and getting not valid with its increasing . basing on the aforesaid
, we can conclude that more precise theoretical treatment is needed in the case with magnesium for adequate description of kinetics of ultracold atoms .
this treatment can be based on the density matrix formalism with full account for the recoil effect ( e.g. , see @xcite ) . moreover , as it will be seen in the next section , the quantum - treatment results noticeably differs from the semiclassical ones , based on the equation ( [ semiclassics ] ) .
that difference , in particular , gave us an idea for exploiting the second stage of sub - doppler laser cooling for getting the desirable results .
let us consider the problem of laser cooling of magnesium atoms out of semiclassical approximation limit as well as some other widely used approximations ( weak - saturation limit , secular approximation , etc . ) .
we assume the laser field to be one - dimensional , composed of two plane monochromatic counterpropagating light waves with equal frequencies and amplitudes ( the quantization axis @xmath81 is collinear to the wave vectors ) : @xmath106 where @xmath107 are the unit complex vectors of waves polarizations , while @xmath108 is the following complex vector @xmath109 nonzero components of the vectors @xmath107 in the spherical basis are @xmath110 here @xmath111 are the ellipticity parameters ( in particular , @xmath112@xmath44@xmath113 corresponds ro righ- or left - circular polarized wave , @xmath112@xmath44@xmath65 is for linear polarization ) , @xmath114 is the angle between main axes of polarization ellipses .
for instance , the case with @xmath111@xmath44@xmath65 and @xmath114@xmath44@xmath115 corresponds to _ lin_@xmath9_lin _ field configuration . here
quantum treatment of atomic kinetics under the laser field ( [ field ] ) is based on the equation on single - atom density matrix in coordinate two - point representation that has the form ( e.g. , see @xcite ) : @xmath116+\widehat{\gamma}\bigl\{\widehat{\rho}\bigr\},\ ] ] with the hamiltonian @xmath117 the first term in the hamiltonian is the operator of kinetic energy of an atom ( @xmath118 is the linear momentum operator ) , @xmath119 describes intratomic degrees of freedom , operator @xmath120 corresponds to the atom - field dipole interaction , and the linear operator functional @xmath121@xmath122 is respective for relaxation processes in an atom .
let us introduce the projection operator onto the excited atom state : @xmath123 and the wigner vector operator @xmath124 , whose spherical components are : @xmath125 with @xmath126@xmath44@xmath127 and @xmath128 the clebsch - gordan coefficients ( e.g. , see @xcite ) .
then the terms @xmath119 and @xmath120 from ( [ hamiltonian ] ) in the resonant approximation can be written as @xmath129 and @xmath130 here @xmath58@xmath44@xmath131 is the rabi frequency ( with @xmath132 the reduced matrix element of dipole operator of an atom ) , @xmath133@xmath44@xmath124@xmath134@xmath135 is the dimensionless operator of atom - field interaction , depending on the coordinate in general case , @xmath136 means hermitian - conjugate term . introduce the new coordinates : @xmath137 in which the spontaneous relaxation operator from ( [ density1 ] ) acquires the form : @xmath138 with @xmath139 note that in the absence of recoil effect , i.e. in the limit @xmath140 , we have @xmath141 .
the density matrix can be divided into four matrix blocks : @xmath142 matrix blocks @xmath143 and @xmath144 describes populations of the ground and the excited states as well as low - frequency ( zeeman ) coherences .
blocks @xmath145 and @xmath146 are responsible for optical coherences . for the new coordinates ( [ coordinates ] ) and using all introduced notations the new equations on the density matrix blocks can be easily acquired from ( [ density1 ] ) .
so , in the steady state we have @xmath147\,,\\ & & \bigl(\gamma-2i\omega_r\frac{\partial^2}{\partial q \partial z}\bigr)\widehat{\rho}^{ee}(z , q)=\nonumber\\ & & \qquad = ir\bigl [ \widehat{v}^{eg}(z+\frac{q}{2 } ) \widehat{\rho}^{ge } - \widehat{\rho}^{eg } \widehat{v}^{eg\,\dag}(z-\frac{q}{2 } ) \bigr]\,,\\ & & \bigl(\frac{\gamma}{2}+i\delta-2i\omega_r\frac{\partial^2}{\partial q \partial z}\bigr)\widehat{\rho}^{ge}(z , q)=\nonumber\\ & & \qquad = ir\bigl [ \widehat{v}^{eg\,\dag}(z+\frac{q}{2 } ) \widehat{\rho}^{ee } - \widehat{\rho}^{gg } \widehat{v}^{eg\,\dag}(z-\frac{q}{2 } ) \bigr]\,,\\ & & \bigl(\frac{\gamma}{2}-i\delta-2i\omega_r\frac{\partial^2}{\partial q \partial z}\bigr)\widehat{\rho}^{eg}(z , q)=\nonumber\\ & & \qquad = ir\bigl [ \widehat{v}^{eg}(z+\frac{q}{2 } ) \widehat{\rho}^{gg } - \widehat{\rho}^{ee } \widehat{v}^{eg}(z-\frac{q}{2 } ) \bigr]\,.\label{densityblocks4}\end{aligned}\ ] ] these equations compose a basis for further theoretical analysis .
for instance , probability density of atoms in the momentum space can be found from the formula : @xmath148 here the linear momentum of an atom evaluated in the recoil momentum units @xmath73 and @xmath149 $ ] denotes trace operation .
momentum distribution @xmath59 must be normalized : @xmath150 this means that the set of equations ( [ densityblocks])-([densityblocks4 ] ) must be supplemented with the condition : @xmath151 average kinetic energy of an atom in the recoil energy units can be evaluated , for instance , with the help of the following formula : @xmath152 as a rule , in magneto - optical trap atoms are localized in weak - magnetic - field region ( in the vicinity of trap s center ) .
therefore , magnetic field does not affect significantly on the temperature of a cloud and we omit it here from our analysis .
in other words , we consider kinetics of atoms in 1d laser field , composed of two counterpropagating beams with orthogonal circular polarizations ( @xmath7@xmath8 configuration ) .
then in equation ( [ basis ] ) we can take @xmath153@xmath44@xmath154 , @xmath155@xmath44@xmath156 , @xmath114@xmath44@xmath65 and polarization vector of total light field in spherical basis takes the form : @xmath157 this form corresponds to the laser field with linear polarization , which rotates by an angle @xmath158@xmath44@xmath159 during propagation along @xmath81-axis . at
that the dimensionless operator @xmath160 from ( [ v ] ) is @xmath161 the considered field configuration has some unique features .
first of all , the field has homogeneous intensity ( it does not depend on @xmath81-coordinate ) . and
the second , the field polarization also can be made homogeneous ( e.g. , see @xcite ) .
indeed , let us pass to the new coordinate system @xmath162 , in which @xmath163-axis coincides with @xmath81-axis in the old @xmath164-system , while the axes @xmath165 and @xmath166 rotate around @xmath81-axis by the angle @xmath158@xmath44@xmath167@xmath44@xmath168@xmath169@xmath170@xmath171@xmath172 ( see fig .
[ fig3 ] ) . in the @xmath162-system
the linearly polarized total - field vector does not rotate anymore ( without loss of generality it can be considered to be directed along @xmath165-axis ) . then in the new system
the interaction operator @xmath160 from ( [ v ] ) does not depend on the coordinate @xmath173 : @xmath174 since the relaxation operator @xmath121 from ( [ gamma ] ) also does not depend on @xmath81 , the density matrix in the new coordinate system is not the function of @xmath81 and it depends only on @xmath175-coordinate .
this circumstance significantly simplifies numerical evaluations of the density matrix equations .
the operator of rotation @xmath176 can be exploited for getting the equations on density matrix in the new basis ( e.g. , see @xcite ) . here
the unit vector @xmath177 defines a rotation axis , while @xmath158 is a rotation angle . in our case
it is natural to coincide @xmath177 with quantization axis @xmath81 and take @xmath158@xmath44@xmath167 .
then action of the rotation operator on a wave function @xmath178 reduces to a simple multiplication by @xmath179 , i.e. @xmath180 with @xmath181@xmath44@xmath182 . in the system
@xmath162 the set of equations ( [ densityblocks])-([densityblocks4 ] ) take the form : @xmath183=\widehat{\gamma}\{\widehat{\rho}(q)\}+i\delta\bigl[\widehat{p}^e,\widehat{\rho}\bigr]+\nonumber\\ & & + \,ir\,\bigl[\widehat{v}_1(q)\widehat{\rho}-\widehat{\rho}\widehat{v}_2(q)\bigr]\,.\end{aligned}\ ] ] at that the normalizing condition ( [ norm2 ] ) becomes rather simple : @xmath184 figure [ fig4]a shows average kinetic energy of an atom as the function of light field intensity , calculated on the basis of numerical solving the equation ( [ densityrotated2 ] ) .
analogical dependence is also presented , gained by applying semiclassical approach on the basis of fokker - planck equation ( [ semiclassics ] ) . as it is seen from the figure the semiclassical approach ( dashed line )
gives the minimum kinetic energy of an atom at the level of @xmath185@xmath26@xmath186@xmath82 , what is several times smaller than the doppler limit @xmath187@xmath26@xmath188@xmath82 .
at the same time the quantum approach ( solid line ) shows the result for energy just a little bit smaller than the doppler limit ( @xmath185@xmath26@xmath189@xmath82 ) .
hence the quantum treatment of the problem shows that it is hardly possible to cool magnesium atoms in mot down to desirable range of temperatures on the basis of transition @xmath1p@xmath2@xmath3@xmath1d@xmath4 .
all this agrees with the experiments of research group from the university of hannover @xcite .
effective temperature , corresponding to the minimum at the plot @xmath190 for quantum - treatment result , is about 310 @xmath13k at frequency detuning @xmath55@xmath44@xmath191@xmath26@xmath89@xmath39@xmath192mhz and light field intensity @xmath91@xmath26@xmath193mw/@xmath71 . beside the temperature of an atomic ensemble
it is also important to know a profile of momentum distribution of atoms in a cloud .
it may be found very useful , in particular , for realization of evaporative cooling stage for achieving ultralow temperatures ( @xmath1941 @xmath13k ) .
let us consider a group of atom in the momentum space with @xmath195@xmath196@xmath197@xmath198 .
tentatively speaking we call this fraction as `` ultracold '' one .
figure [ fig4]b shows number of atoms in the ultracold fraction @xmath199 as the function of light field intensity @xmath91 . as it is seen from the figure there is a maximum in vicinity of 500 mw/@xmath71 .
it should be noted that position of this optimum is not immediately the same as for the minimum of the dependence @xmath200 .
figure [ fig4]b demonstrates that about 40 % of atoms can be concentrated in the ultracold fraction . for comparison analogical dependence is presented , calculated on the basis of semiclassical approach ( dashed line ) , which lies noticeably higher than the former one .
the dependencies @xmath200 and @xmath201 lose to the semiclassical ones , because quantum treatment provides significantly different result for the momentum distribution in the vicinity of @xmath195@xmath26@xmath65 .
indeed , fig .
[ fig5 ] shows very sharp spike in the semiclassical case and a tiny peak as the result of quantum calculations .
fortunately , solution of the problem of deep laser cooling of magnesium atoms can be found by involving the second stage of sub - doppler cooling with the help of one - dimensional optical molasses .
the molasses is composed of two counterpropagating light waves with orthogonal linear polarizations ( _ lin_@xmath9_lin _ configuration ) .
in contrast to @xmath7@xmath8 configuration , in the case of _ lin_@xmath9_lin _ field the total - field polarization transforms from linear to circular ( and back ) along z - axis ( e.g. , see @xcite ) .
consequently , there is no any rotating transformation of coordinate frame @xmath164 that would make density matrix independent of @xmath81-coordinate .
therefore , we must solve the set of equations ( [ densityblocks])-([densityblocks4 ] ) on matrix @xmath202 .
we have solved the equation numerically on the basis of matrix continued fractions method .
the details of the method can be found , for example , in @xcite and we do not reproduce it here . instead of that , we just present the numerical results .
figure [ fig6]a demonstrates much lower minimum kinetic energy than in the case of @xmath7@xmath8 field ( see fig .
[ fig4]a , solid line ) .
in particular , the minimum corresponds to @xmath203@xmath26@xmath204@xmath82 at @xmath91@xmath26@xmath205 mw/@xmath71 ( @xmath84@xmath26@xmath206@xmath13k ) .
besides , as it is seen from fig .
[ fig6]b , ultracold fraction of atoms under _ lin_@xmath9_lin _
light field can be higher than in the case of @xmath7@xmath8 ( compare with fig .
[ fig4]b ) .
narrow structure in momentum profile in vicinity of @xmath195@xmath26@xmath207 becomes more visible than under @xmath7@xmath8 field ( compare fig .
[ fig7 ] and fig . [ fig5 ] ) .
therefore , the second cooling stage involving optical molasses can provide lower temperature as well as larger number of atoms in ultracold fraction ( up to 60% ) .
after the second sub - doppler cooling stage atoms may be loaded , for instance , to a dipole trap . at that the `` hot '' fraction of atoms ( wide background at fig .
[ fig7 ] ) can be evaporated by proper choice of the light potential depth , saving only the ultracold fraction in a trap ( with effective temperature @xmath208@xmath12k ) .
it should be noted that realization of the second sub - doppler stage should eventually provide much more number of ultracold atoms in a dipole trap ( or an optical lattice ) after evaporation than without this stage .
in conclusion we would like to summarize main results of the work . we have suggested using the second sub - doppler cooling stage to solve the problem of deep laser cooling of magnesium atoms .
the first stage implies using of magneto - optical trap involving dipole transition between triplet states @xmath1p@xmath2 and @xmath1d@xmath4 .
in particular , this stage was used in the experiment of researches from hannover university @xcite . in spite of
that the level @xmath1p@xmath2 is degenerate and one could anticipate activation of effective sub - doppler mechanism of cooling in polarization - gradient field @xcite , however , the conducted theoretical analysis has figured out the minimum temperature at the level of 310 @xmath13k , what is just a little bit lower than the estimate for doppler limit of cooling ( @xmath60@xmath69@xmath70@xmath13k ) . to reduce this value by several times
we have proposed using the second laser - cooling stage with the help of optical molasses composed of two counterpropagating orthogonally linearly polarized waves ( _ lin_@xmath9_lin _ field configuration ) .
in contrast to @xmath7@xmath8 field , applied in mot , the optical molasses demonstrates much lower temperature ( 80 @xmath209 ) . at the same time , the minimum achievable temperature of laser cooling in the case of @xmath0 mg still is noticeably higher than for some other atoms , where sub - doppler mechanism provides much better results .
most likely , it is due to relatively large recoil energy of magnesium atom .
for instance , for considered transition in magnesium recoil frequency @xmath43@xmath69@xmath38@xmath39@xmath210khz , while for @xmath211cs ( @xmath212s@xmath213 , f=4 @xmath3 @xmath212p@xmath214 , f=5 ) this frequency is significantly smaller ( @xmath43@xmath69@xmath38@xmath39@xmath215khz ) , what allowed cooling cesium atoms down to 2.5 @xmath13k @xcite . besides temperature ( average kinetic energy ) of ensemble of atoms
we have also paid attention to the linear momentum distributions in the steady state . in particular , we have investigated the problem of increasing concentration of atoms in ultracold fraction ( a region in momentum space in the vicinity of @xmath195@xmath44@xmath65 ) .
conducted numerical calculations have revealed the optimum parameters of laser field for maximization of the ultracold fraction ( @xmath84@xmath231 @xmath13k ) .
this fraction can be easily localized in a optical trap , while the other fraction ( `` hot '' atoms ) can be evaporated from the trap by proper choice of the optical depth . at that
, it is the second stage of sub - doppler cooling that can provide great increase of ultracold atomic number in comparison with the case with only the first stage realized ( as in the experiments @xcite ) . in our theoretical analysis quantum treatment with full account for the recoil effect
has been exploited , i.e. we have not been limited by semiclassical or secular approximations as well as weak - field limit .
it has allowed us studying kinetics of cold magnesium in a wide range of intensity and frequency detuning to determine the optimum parameters of laser field . also we have compared data provided by quantum and semiclassical approaches . as the result of that comparison
we can conclude that semiclassical approach in the case of transition @xmath1p@xmath2@xmath3@xmath1d@xmath4 in @xmath0 mg is not valid for adequate understanding the kinetics of ultracold magnesium atoms for a wide range of light - field parameters .
moreover , we can also conclude that for getting the adequate estimate of cooling parameters and understanding the problems in deep laser cooling of atoms it is quite necessary to treat the problems with the help of quantum approach . at the end
we should note that in spite of theoretical analysis has been done out of many widely used approximations , we have assumed the problem to be one - dimensional .
however , light - field configuration used in a magneto - optical trap is always three - dimensional ( three pairs of circularly polarized beams ) .
therefore , obviously , the results of such 1d analysis may differ from the real experiment with 3d field .
for example , one can refer to the papers @xcite for getting the estimate of such kind of difference . in these papers calculations
were done for 1d and 3d configurations by the example of simple transition @xmath102@xmath44@xmath65@xmath3@xmath103@xmath44@xmath67 in limits of semiclassical and slow - atoms approximations . at the same time , an optical molasses ( the second cooling stage suggested ) , which is of the most interest from the viewpoint of deep laser cooling , can be implemented in 3d as well as in 1d configuration .
three - dimensional optical molasses for various transition of the type @xmath102@xmath44@xmath216@xmath3@xmath103@xmath44@xmath216@xmath170@xmath67 was investigated in ref . @xcite under weak - saturation approximation and with the help of adiabatic reduction of density matrix equations to ground state .
unfortunately that approximation gives good results not for wide range of parameters @xmath55 and @xmath91 that can have an interest from the laser cooling view of point ( e.g. , see the work @xcite , where results of adiabatic approximation were compared with results of full quantum treatment ) .
three - dimensional quantum treatment with full account for the recoil effect and beyond the aforesaid approximations is quite difficult task that requires separate study .
the work was partially supported by the ministry of education and science of the russian federation ( gov .
2014/139 , project no .
825 ) , presidium of the siberian branch of the russian academy of sciences , and by the grants of rfbr ( nos . 15 - 02 - 06087 , 15 - 32 - 20330 , 14 - 02 - 00806 , 14 - 02 - 00712 , 14 - 02 - 00939 ) and russian presidential grants ( mk-4680.2014.2 and nsh-4096.2014.2 ) .
ludlow et al . , science * 319 * , 1805 ( 2008 ) .
p. komar , e.m .
kessler , m. bishof , l. jiang , a.s .
sorensen , j. ye , and m.d .
lukin , nature physics * 10 * , 582 ( 2014 ) .
t. akatsuka , h. ono , k. hayashida , k. araki , m. takamoto , t. takano and h. katori , jpn .
. phys . * 53 * , 032801 ( 2014 ) .
g. marra , j. kronjaeger , p .- e .
pottie et al . , in the electronic _ book of abstracts of 28th european frequency and time forum , neuchatel , switzerland , 2014 _ ( the printing house inc . ,
stoughton , 2014 ) , p. 164 .
m. takamoto , i. ushijima , m. das et al .
, c. r. physique , to be published ( 2015 ) .
h. katori , in _ proceedings of the 6th symposium on frequency standards and metrology , university of st .
andrews , fife , scotland , 2001 _ , edited by p. gill ( world scientific , singapore , 2002 ) .
h. katori , m. takamoto , v.g .
palchikov , and v.d .
ovsiannikov , phys .
lett . * 91 * , 173005 ( 2003 ) .
m. takamoto and h. katori , phys .
* 91 * , 223001 ( 2003 ) .
hoyt , z.w .
barber , c.w .
oates , t. m. fortier , s.a .
diddams , and l. hollberg , phys .
95 * , 083003 ( 2005 )
. t. hong , c. cramer , e. cook , w. nagourney , e.n .
fortson , opt .
* 30 * , 2644 ( 2005 ) .
m. petersen , r. chicireanu , s.t .
dawkins , d.v .
magalhes , c. mandache , y. le coq , a. clairon , s. bize , phys .
* 101 * , 183004 ( 2008 ) .
yudin , a.v .
taichenachev , c.w .
oates , z.w .
barber , n.d .
lemke , a.d .
ludlow , u. sterr , ch .
lisdat , f. riehle , phys .
a * 82 * , 011804 ( 2010 ) .
h. katori , nature photonics * 5 * , 203 ( 2011 ) .
s. falke , n. lemke , ch .
grebing et al . , new journal of physics * 16 * , 073023 ( 2014 ) .
b.j . bloom , t.l .
nicholson , j.r .
williams , s.l .
campbell , m. bishof , x. zhang , w. zhang , s.l . bromley and j.ye ,
nature * 506 * , 71 ( 2014 ) .
i. ushijima , m. takamoto , m. das , t. ohkubo , h. katori , nature photonics * 9 * , 185 ( 2015 ) .
barber , c.w .
hoyt , c.w .
oates , l. hollberg , a.v .
taichenachev , v.i .
yudin , phys .
lett . * 96 * , 083002 ( 2006 ) .
n. hinkley , j.a .
sherman , n.b .
phillips , m. schioppo , n.d .
lemke , k. beloy , m. pizzocaro , c.w .
oates , a.d .
ludlow , science * 341 * , 1215 ( 2013 ) .
lemke , a.d .
ludlow , z.w .
barber , t.m .
fortier , s.a .
diddams , y. jiang , s.r .
jefferts , t.p .
heavner , t.e .
parker , and c.w .
oates , phys .
* 103 * , 063001 ( 2009 ) . c. degenhardt , h. stoehr , u. sterr , f. riehle , ch .
lisdat , phys .
a * 70 * , 023414 ( 2004 ) .
t. akatsuka , m. takamoto , h. katori , phys .
a * 81 * , 023402 ( 2010 ) .
mcferran , l. yi , s. mejri , s. di manno , w. zhang , j. guena , y. le coq , s. bize , phys . rev . lett . * 108 * , 183004 ( 2012 ) .
j. friebe , m. riedmann , t. wbbena , a. pape , h. kelkar , w. ertmer , o. terra , u. sterr , s. weyers , g. grosche , h. schnatz , e.m .
rasel , new j. phys .
* 13 * , 125010 ( 2011 ) .
a.n . goncharov et al . ,
electron * 44 * , 521 ( 2014 ) .
taichenachev , v.i .
yudin , c.w .
oates , c.w .
hoyt , z.w .
barber , l. hollberg , phys .
lett . * 96 * , 083001 ( 2006 ) . t. akatsuka , m. takamoto , and h. katori , nature physics * 4 * , 954 ( 2008 ) .
t. legero , c. lisdat , j.s.r .
vellore winfred et al .
, ieee trans .
58 * , 1252 ( 2009 ) .
lisdat , j.s.r .
vellore winfred , t. middelmann , f. riehle , and u. sterr , phys .
103 * , 090801 ( 2009 ) .
n. poli , m. schioppo , s. vogt , st .
falke , u. sterr , ch .
lisdat , g.m .
tino , appl .
b * 117 * , 1107 ( 2014 ) .
rasel et al . , in the electronic _ book of abstracts of the 28th international frequency and time forum , neuchatel , switzerland , 2014 _ ( the printing house inc . , stoughton , 2014 ) , p. 138 .
mehlstubler , k. moldenhauer , m. riedmann , n. rehbein , j. friebe , e.m .
rasel , w. ertmer , phys .
a * 77 * , 021402 ( 2008 ) .
n. rehbein , t.e .
mehlstubler , j. keupp et al .
a * 76 * , 043406 ( 2007 ) .
s. rhmann , a. kulosa , d. fim , k. zipfel , w. ertmer , e. rasel , in the electronic _ book of abstracts of the 28th international frequency and time forum , neuchatel , switzerland , 2014 _ ( the printing house inc . ,
stoughton , 2014 ) , p. 213 .
porsev , a. derevianko , phys .
a * 74 * , 020502 ( 2006 ) .
h. hachisu , k. miyagishi , s.g .
porsev , a. derevianko , v.d .
ovsiannikov , v.g .
palchikov , m. takamoto , h. katori , phys .
lett . * 100 * , 053001 ( 2008 ) .
m. petersen , r. chicireanu , s.t dawkins , d.v .
magalhaes , c. mandache , y. le coq , a. clairon , s. bize , phys .
* 101 * , 183004 ( 2008 ) .
t. binnewies , g. wilpers , u. sterr , f. riehle , j. helmcke , t. e. mehlstubler , e.m .
rasel , w. ertmer , phys .
lett . * 87 * , 123002 ( 2001 ) .
curtis , c.w .
oates , l. hollberg , phys .
a * 64 * , 031403 ( 2001 ) .
h. katori , t. ido , y. isoya , m. kuwata - gonokami , phys .
lett . * 82 * , 1116 ( 1999 ) .
t. kuwamoto , k. honda , y. takahashi , t. yabuzaki , phys .
a * 60 * , 745 ( 1999 ) .
minogin , zh .
. fiz . * 79 * , 2044 ( 1980 ) .
cook , phys .
a * 22 * , 1078 ( 1980 ) .
j. dalibard and c. cohen - tannoudji , j. phys .
b * 18 * , 1661 ( 1985 ) .
j. javanainen , phys .
a * 44 * , 5857 ( 1991 ) .
choi , s.e .
park , j. chen , v.g .
minogin , phys .
a * 77 * , 015405 ( 2008 ) .
prudnikov , a.v .
taichenachev , v.i .
yudin , zh .
147 * , 678 ( 2015 ) .
y. castin , k. molmer , phys .
lett . * 74 * , 3772 ( 1995 ) . | the problem of deep laser cooling of @xmath0 mg atoms is theoretically studied .
we propose two - stage sub - doppler cooling strategy using electro - dipole transition @xmath1p@xmath2@xmath3@xmath1d@xmath4 ( @[email protected] nm ) .
the first stage implies exploiting magneto - optical trap with @xmath7 and @xmath8 light beams , while the second one uses a _
lin_@xmath9_lin _ molasses .
we focus on achieving large number of ultracold atoms ( t@xmath10@xmath1110@xmath12k ) in a cold atomic cloud .
the calculations have been done out of many widely used approximations and based on quantum treatment with taking full account of recoil effect .
steady - state average kinetic energies and linear momentum distributions of cold atoms are analyzed for various light field intensities and frequency detunings .
the results of conducted quantum analysis have revealed noticeable differences from results of semiclassical approach based on the fokker - planck equation . at certain conditions
the second cooling stage can provide sufficiently lower kinetic energies of atomic cloud as well as increased fraction of ultracold atoms than the first one .
we hope that the obtained results can assist overcoming current experimental problems in deep cooling of @xmath0 mg atoms by means of laser fields .
cold magnesium atoms , being cooled in large number down to several @xmath13k , have certain interest , for example , in quantum metrology . |
over the last decades , the energy eigenvalues and corresponding eigenfunctions between interaction systems have raised a great deal of interest in relativistic quantum mechanics as well as in non - relativistic quantum mechanics .
the exact solution of the wave equations ( relativistic or non - relativistic ) are very important since the wave function contains all the necessary information regarding the quantum system under consideration .
analytical methods such as the supersymmetry ( susy ) @xcite and the nikiforov - uvarov method ( nu ) @xcite have been used to solve the wave equations exactly or quasi - exactly for non zero angular momentum quantum number ( @xmath3 ) by means of a given potential .
the radial schrdinger equation for the hulthn potential is solved exactly by using several techniques @xcite for @xmath4 .
for the case @xmath5 , the effective hulthn potential can not be solved exactly , but a number of methods have been used to find the bound state energy eigenvalues numerically @xcite and quasi - analytically such as the variational @xcite , perturbation @xcite , shifted @xmath2 expansion @xcite , nu @xcite and susy @xcite methods .
the hulthn potential @xcite is one of the important short - range potentials in physics and it potential has been applied to a number of areas such as nuclear and particle physics , atomic physics , condensed matter and chemical physics ( see @xcite and the references therein ) .
therefore , it would be interesting and important to solve the non - relativistic radial schrdinger equation for this potential for @xmath5 , since it has been extensively used to describe the bound and the continuum states of the interaction systems . recently , an alternative method , called as the asymptotic iteration method ( aim ) , has been developed by ifti _
@xcite for solving second - order homogeneous linear differential equations and it has been applied to solve the non - relativistic radial schrdinger equation or the relativistic dirac equation . in this paper , we aim to solve the hulthn potential to obtain the energy eigenvalues and corresponding eigenfunctions for any @xmath0 states .
in the next section , aim is introduced .
then , in section [ apply1 ] , the schrdinger equation is solved by using aim for the hulthn potential for any @xmath0 states and our aim results are given in comparison with the results of the numerical integration @xcite , the variational @xcite , the shifted @xmath2 expansion @xcite and the susy @xcite methods . finally , section [ conclude ] is devoted to the summary and conclusion .
aim is briefly outlined here and the details can be found in references @xcite .
aim is proposed to solve the second - order differential equations of the form @xmath6 where @xmath7 and the prime denotes the derivative with respect to @xmath8 .
the variables , @xmath9 and @xmath10 , are sufficiently differentiable .
the differential equation ( [ diff ] ) has a general solution @xcite @xmath11 dx_{2 } \right ) dx_{1 } \right ] \ ] ] for sufficiently large @xmath12 , @xmath13 , if @xmath14 where @xmath15 note that one can also start the recurrence relations from @xmath16 with the initial conditions @xmath17 and @xmath18 @xcite . for a given potential such as the hulthn
, the radial schrdinger equation is converted to the form of equation ( [ diff ] ) .
then , s@xmath19 and @xmath10 are determined and s@xmath20 and @xmath21 parameters are calculated by the recurrence relations given by equation ( [ iter ] ) . the termination condition of the method in equation ( [ termination ] ) can be arranged as @xmath22 the energy eigenvalues are obtained from the roots of the equation ( [ quantization ] ) if the problem is exactly solvable .
if not , for a specific @xmath23 principal quantum number , we choose a suitable @xmath24 point , determined generally as the maximum value of the asymptotic wave function or the minimum value of the potential @xcite , and the approximate energy eigenvalues are obtained from the roots of this equation for sufficiently great values of @xmath12 with iteration . in this study , we seek the exact solution of the radial schrdinger equation for which the relevant second order homogenous linear differential equation takes the following general form @xcite , @xmath25 if this equation is compared to equation ( [ diff ] ) , it entails the following expressions @xmath26 @xmath27 and @xmath28 are constants and @xmath29 can be determined from the condition ( [ termination ] ) for @xmath30 and @xmath31=-1,0,1,2,3 , ... as follows : @xmath32 hence , these formule are easily generalized as , @xmath33 the exact eigenfunctions can be derived from the following generator : @xmath34 where @xmath35 , @xmath23 represents the radial quantum number and @xmath12 shows the iteration number . for exactly solvable potentials ,
the radial quantum number @xmath23 is equal to the iteration number @xmath12 and the eigenfunctions are obtained directly from equation ( [ generator1 ] ) . for nontrivial potentials that have no exact solutions ,
@xmath12 is always greater than @xmath23 in these numerical solutions and the approximate energy eigenvalues are obtained from the roots of equation ( [ quantization ] ) for sufficiently great values of @xmath12 with iteration .
it should be pointed out that @xmath36 given by equation ( [ termination ] ) is equal to zero for the ground state .
therefore , using equation ( [ termination ] ) with ( [ snln ] ) in equation ( [ generator1 ] ) , the eigenfunctions are obtained as follows , @xmath37 finally , the following general formula for the exact solutions @xmath38 is obtained as , @xmath39 it is important to note that square integrable one in @xmath40 is this total wave function which is the asymptotic form of the wave function times @xmath38 given by equation ( [ generator1 ] ) . here ,
@xmath41=@xmath42 = @xmath43 @xmath44 = @xmath45 .
the @xmath46 and the @xmath47 are known as the pochhammer symbol and the gauss hypergeometric function , respectively .
the motion of a particle with the mass @xmath48 in the spherically symmetric potential is described in the spherical coordinates by the following schrdinger equation : @xmath49+v(r ) \right)\psi_{nl m}(r,\theta,\phi)=e\psi_{nl m}(r,\theta,\phi)\label{sch}\ ] ] defining @xmath50 , we obtain the radial part of the schrdinger equation : @xmath51r_{nl}(r)=0 \label{radialr}\end{aligned}\ ] ] it is sometimes convenient to define @xmath52 and the effective potential as follows @xmath53 since @xmath54 the radial schrdinger equation @xcite given by equation ( [ radialr ] ) follows that @xmath55u_{nl}(r)=0 \label{factorschrodinger}\ ] ] the hulthn potential @xcite is given by @xmath56 where z and @xmath57 are respectively the atomic number and the screening parameter , determining the range for the hulthn potential .
the hulthn potential behaves like the coulomb potential near the origin @xmath58 , but in the asymptotic region @xmath59 , the hulthn potential decreases exponentially , so its capacity for bound states is smaller than the coulomb potential .
however , for small values of the screening parameter @xmath57 , the hulthn potential becomes the coulomb potential given by @xmath60 .
the effective hulthn potential is @xmath61 where @xmath62 is known as the centrifugal term .
this effective potential can not be solved analytically for @xmath5 because of the centrifugal term .
therefore , we must use an approximation for the centrifugal term similar to other authors @xcite . in this approximation
, @xmath63 is used for the centrifugal term . as
shown in figure [ eigenpot ] , this approximation is only valid for small @xmath64 and it breaks down in the high - screening region . for small @xmath64 , @xmath65 is very well approximated to @xmath66 and the schrdinger equation for this approximate potential is solvable analytically .
so , the effective potential becomes @xmath67 instead of solving the radial schrdinger equation for the effective hulthn potential @xmath66 given by equation ( [ veff ] ) , we now solve the radial schrdinger equation for the new effective potential @xmath65 given by equation ( [ veff1 ] ) . inserting this new effective potential into equation ( [ factorschrodinger ] ) and using the following _ ansatzs _ in order to make the differential equation more compact @xmath68 the radial schrdinger equation takes the following form : @xmath69 u_{nl}(x)=0\ ] ]
if we rewrite equation ( [ radyal ] ) by using a new variable of the form @xmath70 , we obtain @xmath71u_{nl}(z)=0\ ] ] in order to solve this equation with aim , we should transform this equation to the form of equation ( [ diff ] ) .
therefore , the reasonable physical wave function we propose is as follows @xmath72 if we insert this wave function into equation ( [ trans ] ) , we have the second - order homogeneous linear differential equations as in the following form @xmath73 \frac{df_{nl}(z)}{dz}+\left[\frac{(2\varepsilon+l+2)l+2\varepsilon-\beta^{2}+1}{z(1-z ) } \right]f_{nl}(z)\ ] ] which is now amenable to an aim solution . by comparing this equation with equation ( [ diff ] )
, we can write the @xmath74 and @xmath75 values and by means of equation ( [ iter ] ) , we may calculate @xmath76 and @xmath77 .
this gives : @xmath78 combining these results with the quantization condition given by equation ( [ quantization ] ) yields @xmath79 when the above expressions are generalized , the eigenvalues turn out as @xmath80 using equation ( [ ansatz ] ) , we obtain the energy eigenvalues e@xmath81 , @xmath82^{2 } \hspace{1cm}\ ] ] in the atomic units @xmath83 and for z = 1 , equation ( [ energyeigenvalues ] ) turns out to be @xmath84^{2 } \hspace{1cm}\ ] ] in order to test the accuracy of equation ( [ eigenvaluesatomic ] ) , we calculate the energy eigenvalues for @xmath85 , any @xmath23 and @xmath0 quantum numbers and several values of the screening parameter .
aim results are compared with the results of the numerical integration @xcite , the variational @xcite , the shifted @xmath2 expansion @xcite and the susy @xcite methods in table [ table1 ] and table [ table2 ] . as it can be seen from the results presented in these tables , the aim results are in good agrement with the results of the other methods for the small @xmath57 values . for large @xmath57 values , there are differences between our results and the results of others .
this difference is due to the @xmath65 potential , which we have used to approximate the @xmath66 potential .
as it is seen from figure [ eigenpot ] , for large @xmath64 values , the discrepancy becomes apparent between our @xmath65 and the true @xmath66 potentials .
this gives rise to the differences for the eigenvalues presented in tables [ table1 ] and [ table2 ] at large @xmath57 values .
now , as indicated in section [ aim ] , we can determine the corresponding wave functions by using equation ( [ efson ] ) .
when we compare equation ( [ compare1 ] ) and equation ( [ aimschr ] ) , we find @xmath86 , @xmath87 , @xmath88 , and @xmath89 .
therefore , we find @xmath90 and @xmath91 .
so , we can easily find the solution for @xmath92 , for the energy eigenvalue equation ( [ energyeigenvalues ] ) by using equation ( [ efson ] ) .
@xmath93 thus , we can write the total radial wave function as below , @xmath94 where @xmath31 is the normalization constant .
we have shown an alternative method to obtain the energy eigenvalues and corresponding eigenfunctions of the hulthn potential within the framework of the asymptotic iteration method for any @xmath0 states .
we have calculated the energy eigenvalues for the hulthn potential with @xmath85 and several values of the screening parameter .
the wave functions are physical and energy eigenvalues are in good agreement with the results obtained by other methods . in order to demonstrate this ,
aim results have been compared with the results of the numerical integration @xcite , the variational @xcite , the shifted @xmath2 expansion @xcite and the susy @xcite methods in tables [ table1 ] and [ table2 ] . for small @xmath57 values ,
aim results are in good agreement with the results of the other methods , but in the high screening region , the agreement is poor .
the reason is simply that when the @xmath64 increases in the high screening region , the agreement between @xmath66 and @xmath65 potentials decreases as shown in figure [ eigenpot ] . this problem could be solved by making a better approximation of the centrifugal term .
it should be pointed out that the asymptotic iteration method gives the eigenvalues directly by transforming the radial schrdinger equation into a form of @xmath95 = @xmath96 .
the wave functions are easily constructed by iterating the values of @xmath97 and @xmath98 .
the asymptotic iteration method results in exact analytical solutions if there is and provides the closed - forms for the energy eigenvalues as well as the corresponding eigenfunctions . where there is no such a solution , the energy eigenvalues are obtained by using an iterative approach @xcite . as it is presented , aim puts no constraint on the potential parameter values involved and it is easy to implement .
the results are sufficiently accurate for practical purposes .
this paper is an output of the project supported by the scientific and technical research council of turkey ( tbitak ) , under the project number tbag-106t024 and erciyes university ( fba-03 - 27 , fbt-04 - 15 , fbt-04 - 16 ) .
authors would also like to thank dr .
aye boztosun for the proofreading as well as professor r. l. hall and dr .
h. ifti for useful comments on the manuscript .
f. cooper , a. khare and u. sukhatme , phys . rep .
* 251 * ( 1995 ) 267 .
a. f. nikiforov and v. b. uvarov , ` special functions of mathematical physics ` , birkhuser , basel , ( 1988 ) . h. ciftci , r. l. hall and n. saad , j. phys .
a : math . gen .
* 36 * ( 2003 ) 11807 .
h. ciftci , r. l. hall and n. saad , j. phys .
a : math . gen .
* 38 * ( 2005 ) 1147 .
o. bayrak and i. boztosun , j. phys .
a : math . gen .
* 39 * ( 2006 ) 6955 .
l. hulthn , ark . mat .
28a * ( 1942 ) 5 .
m. akda , r. sever , j. molec .
( theochem ) * 710 * ( 2004 ) 223 .
s. flgge , ` practical quantum mechanics ` vol .
i , springer , berlin , ( 1994 ) .
varshni , phys . rev . * a 41 * 1990 4682 p. matthys , h. de meyer , phys . rev . * a 38 * ( 1988 ) 1168
tang , f.t .
chan , phys . rev . * a 35 * ( 1987 ) 911 .
s. m. ikhdair and r. sever , [ arxiv : quant - ph/0508009 ] .
b. gonul , o. ozer , y. cancelik and m. kocak , phys . lett . * a 275 * ( 2000 ) 238 .
greene , c. aldrich , phys .
rev . * a 14 * ( 1976 ) 2363 .
u. myhrman , j. phys .
a : math . gen .
* 16 * ( 1983 ) 263 .
a. bechlert , w bhring , j. phys .
b : at . mol .
* 21 * ( 1988 ) 817 .
shang - wu qian , bo - wen huang , and zhi - yu gu , new journal of physics * 4 * ( 2002 ) 13 .
o. bayrak and i. boztosun , int .
j. quantum chem .
( 2006 ) , in press .
+ i. boztosun , m. karakoc , f. yasuk and a. durmus , j. math .
* 47 * ( 2006 ) 062301 .
t. barakat , j. phys .
* 39 * ( 2006 ) 823 .
+ t. barakat , phys .
lett . * a344 * ( 2005 ) 411 .
+ t. barakat , k. abodayeh , a. mukheimer , j. phys .
* 38 * ( 2005 ) 1299 .
f. m. fernndez , j. phys .
a : math . gen . * 37 * ( 2004 ) 6173 .
ccccccccccccccc @xmath99 & & @xmath57 & & aim & & susy @xcite & & numerical int .
@xcite & & variational @xcite & & shifted @xmath100 @xcite + 2p & & 0.025 & & 0.1128125 & & 0.1127605 & & 0.1127605 & & 0.1127605 & & + & & 0.050 & & 0.1012500 & & 0.1010425 & & 0.1010425 & & 0.1010425 & & 0.1010424 + & & 0.075 & & 0.0903125 & & 0.0898478 & & 0.0898478 & & 0.0898478 & & + & & 0.100 & & 0.0800000 & & 0.0791794 & & 0.0791794 & & 0.0791794 & & 0.0791794 + & & 0.150 & & 0.0612500 & & 0.0594415 & & 0.0594415 & & 0.0594415 & & + & & 0.200 & & 0.0450000 & & 0.0418854 & & 0.0418860 & & 0.0418860 & & 0.0418857 + & & 0.250 & & 0.0312500 & & 0.0266060 & & 0.0266111 & & 0.0266108 & & + & & 0.300 & & 0.0200000 & & 0.0137596 & & 0.0137900 & & 0.0137878 & & + & & 0.350 & & 0.0112500 & & 0.0036146 & & 0.0037931 & & 0.0037734 & & + + 3p & & 0.025 & & 0.0437590 & & 0.0437068 & & 0.0437069 & & 0.0437069 & & + & & 0.050 & & 0.0333681 & & 0.0331632 & & 0.0331645 & & 0.0331645 & & 0.03316518 + & & 0.075 & & 0.0243837 & & 0.0239331 & & 0.0239397 & & 0.0239397 & & + & & 0.100 & & 0.0168056 & & 0.0160326 & & 0.0160537 & & 0.0160537 & & 0.01606772 + & & 0.150 & & 0.0058681 & & 0.0043599 & & 0.0044663 & & 0.0044660 & & + + 3d & & 0.025 & & 0.0437587 & & 0.0436030 & & 0.0436030 & & 0.0436030 & & + & & 0.050 & & 0.0333681 & & 0.0327532 & & 0.0327532 & & 0.0327532 & & 0.0327532 + & & 0.075 & & 0.0243837 & & 0.0230306 & & 0.0230307 & & 0.0230307 & & + & & 0.100 & & 0.0168055 & & 0.0144832 & & 0.0144842 & & 0.0144842 & & 0.0144842 + & & 0.150 & & 0.0058681 & & 0.0132820 & & 0.0013966 & & 0.0013894 & & + + ccccccccccccccc @xmath99 & & @xmath57 & & aim & & susy @xcite & & numerical int .
@xcite & & variational @xcite & & shifted @xmath100 @xcite + 4p & & 0.025 & & 0.0200000 & & 0.0199480 & & 0.0199489 & & 0.0199489 & & + & & 0.050 & & 0.0112500 & & 0.0110430 & & 0.0110582 & & 0.0110582 & & 0.0110725 + & & 0.075 & & 0.0050000 & & 0.0045385 & & 0.0046219 & & 0.0046219 & & + & & 0.100 & & 0.0012500 & & 0.0004434 & & 0.0007550 & & 0.0007532 & & + 4d & & 0.025 & & 0.0200000 & & 0.0198460 & & 0.0198462 & & 0.0198462 & & + & & 0.050 & & 0.0112500 & & 0.0106609 & & 0.0106674 & & 0.0106674 & & 0.0106690 + & & 0.075 & & 0.0050000 & & 0.0037916 & & 0.0038345 & & 0.0038344 & & + 4f & & 0.025 & & 0.0200000 & & 0.0196911 & & 0.0196911 & & 0.0196911 & & + & & 0.050 & & 0.0112500 & & 0.0100618 & & 0.0100620 & & 0.0100620 & & 0.0100620 + & & 0.075 & & 0.0050000 & & 0.0025468 & & 0.0025563 & & 0.0025557 & & + + 5p & & 0.025 & & 0.0094531 & & 0.0094011 & & 0.0094036 & & & & 0.0094087 + & & 0.050 & & 0.0028125 & & 0.0026056 & & 0.0026490 & & & & + 5d & & 0.025 & & 0.0094531 & & 0.0092977 & & 0.0093037 & & & & 0.0093050 + & & 0.050 & & 0.0028125 & & 0.0022044 & & 0.0023131 & & & & + 5f & & 0.025 & & 0.0094531 & & 0.0091507 & & 0.0091521 & & & & 0.0091523 + & & 0.050 & & 0.0028125 & & 0.0017421 & & 0.0017835 & & & & + 5 g & & 0.025 & & 0.0094531 & & 0.0089465 & & 0.0089465 & & & & 0.0089465 + & & 0.050 & & 0.0028125 & & 0.0010664 & & 0.0010159 & & & & + + 6p & & 0.025 & & 0.0042014 & & 0.0041493 & & 0.0041548 & & & & + 6d & & 0.025 & & 0.0042014 & & 0.0040452 & & 0.0040606 & & & & + 6f & & 0.025 & & 0.0042014 & & 0.0038901 & & 0.0039168 & & & & + 6 g & & 0.025 & & 0.0042014 & & 0.0036943 & & 0.0037201 & & & & + and the approximate effective hulthn potentials @xmath65 with respect to the @xmath64 for various values of the screening parameter .
the parameters are in atomic units @xmath101 and @xmath57 change from 0.4 to 5.2 in steps of 0.4 . ] | in this article , we present the analytical solution of the radial schrdinger equation for the hulthn potential within the framework of the asymptotic iteration method by using an approximation to the centrifugal potential for any @xmath0 states . we obtain the energy eigenvalues and the corresponding eigenfunctions for different screening parameters .
the wave functions are physical and energy eigenvalues are in good agreement with the results obtained by other methods for different @xmath1 values . in order to demonstrate this ,
the results of the asymptotic iteration method are compared with the results of the supersymmetry , the numerical integration , the variational and the shifted @xmath2 expansion methods . |
the time latitude distribution of the axisymmetric component of the large - scale solar magnetic fields is known to have a pronounced poleward branch at higher latitudes ( @xcite , @xcite , @xcite , @xcite ; @xcite , @xcite ; @xcite , @xcite ; @xcite , @xcite ) .
this branch is also present in the butterfly diagram of a number of tracers of the magnetic field such as quiescent filaments , polar faculae , or the coronal green line ( @xcite , @xcite ; @xcite , @xcite ; @xcite , @xcite ) .
several conflicting explanations exist for this polar branch . noting that the separation latitude of the two branches , 3040@xmath0 , approximately coincides with the latitude where the radial differential rotation changes sign according to helioseismology , one group of theories interprets it as the surface reflection of a high - latitude poleward propagating dynamo wave , coexisting with the low - latitude equatorward wave ( @xcite , @xcite ; @xcite , @xcite , @xcite ) .
the parker yoshimura rule of sign ( @xcite , @xcite ) then naturally leads to the correct directions of propagation if @xmath1 is negative , as expected in the lower overshooting layer where the dynamo should operate .
( in fact this is only so for certain latitudinal profiles of @xmath1 , cf .
@xcite , @xcite . )
the most modern version of these models , also incorporating some transport effects , is due to @xcite ( @xcite ) . in
what follows we will refer to these models as _
double wave models .
_ an alternative approach to the problem of the origin of the poleward drift is to interpret it in terms of magnetic transport processes in the solar photosphere and convective zone .
the common ancestor of all such models was the now classic babcock
leighton interpretation of the solar cycle ( @xcite , @xcite , @xcite , @xcite ) .
the babcock
leighton model was what one may call a _ mixed transport model in as much as the poloidal fields were brought to the surface in a concentrated form in active regions , and thereafter they were passively transported to the poles by transport processes ( diffusion and meridional circulation ) . in the more recent mixed transport models ( @xcite , @xcite ; @xcite , @xcite )
meridional circulation plays the main role in transporting the weak fields to the poles , and is also supposed to be responsible for the equatorward drift of toroidal fields in the dynamo layer .
( in the original babcock
leighton theory the equatorward branch was still supposed to be due to a dynamo wave . ) _ a third type of models may be called _ passive transport models .
these are a variation of the mixed transport models where the `` active source '' of the weak fields in the form of active regions is also substituted by a passive transport mechanism .
dikpati & choudhuri ( @xcite , @xcite ) suggested such a scenario wherein the poloidal fields are brought to the surface at low latitudes by meridional circulation . in these models
the origin of the equatorward drift is not specified ; instead , it is just given as a lower boundary condition .
the resulting butterfly diagrams for the poloidal field were in good agreement with the observations provided that the value of the turbulent magnetic diffusivity used in the calculations was artifcially reduced to about 10 km@xmath2/s , i.e. nearly 2 orders of magnitude lower than expected .
_ in the present paper we generalize the models of dikpati & choudhuri ( @xcite , @xcite ) to include all turbulent transport effects such as anisotropic turbulent diffusion , turbulent pumping , and density pumping , as well as the meridional circulation .
we construct models with realistic values of the transport coefficients in order to see whether the horizontal component of pumping can support the meridional circulation in producing a poleward migration despite the strong diffusive link between the surface and the bottom of the convective zone .
( such a possibility was suggested by @xcite , @xcite and @xcite , @xcite . )
another motivation for this study is to check to what extent is the radial pumping able to concentrate the fields to the bottom of the convective zone , thereby restricting dynamo action to that layer , and to reduce the surface field strength to the observed values , as suggested by @xcite ( @xcite ) , @xcite ( @xcite ) , and @xcite ( @xcite ) .
we calculate the mean poloidal magnetic field in the solar convective zone as a function of space and time , assuming axial symmetry and antisymmetry to the equatorial plane .
no dynamo is supposed to operate in the convective zone , i.e. the @xmath1-effect is neglected . the dynamo layer below is not included in our computational volume ; instead , the processes operating there are assumed to influence our model via the lower boundary conditions .
the equations of passive magnetic field transport with these assumptions are ( @xcite , @xcite ) : @xmath3 @xmath4 where @xmath5 @xmath6 @xmath7 @xmath8/r \nonumber \\ & & -[\beta_{rr}+\beta_{\theta\theta}\cot^2\theta + \beta_{\theta r}(1+\cot\theta ) ] /r^2 .
\end{aligned}\ ] ] @xmath9 is the mean magnetic flux density , @xmath10 is the azimuthal component of the mean vector potential , @xmath11 is the velocity of meridional circulation , @xmath12 is the anisotropic turbulent diffusivity , and @xmath13 and @xmath14 are the normal and anomalous components of the pumping ( incorporating both the density gradient and the gradient of turbulence intensity , @xcite , @xcite ) .
note that in another widespread notation @xmath13 and @xmath14 are expressed by components of the antisymmetric and symmetric parts of the @xmath1-tensor .
for @xmath11 we use the same expression as given in @xcite ( @xcite ) . for the diffusivity and pumping
standard mean - field theory expressions are used ; the formulae are collected in the appendix . for the evaluation of those expressions we use the ukx convective zone model ( @xcite , @xcite ) .
the differential rotation is assumed to be independent of depth in the computational regime . for the solution of eq .
( [ eq : main ] ) we employ the following boundary conditions . at the bottom of the computational regime @xmath15
was explicitly given , as mentioned above .
for the other boundaries we set @xmath16 the latter condition implies that the field is vertical at the surface ( @xcite , @xcite ; @xcite , @xcite ) . note that these simple boundary conditions are not necessarily unrealistic :
the field is intermittent , and if most of the flux in the photosphere is present in flux tubes exceeding @xmath17mx then eq . ( [ eq : top ] ) should apply owing to the strong buoyancy of these tubes .
if , on the other hand , the flux is mainly present in thinner tubes , then a potential field boundary condition would be better ( cf .
@xcite , @xcite for a discussion of this problem ) .
we performed some test runs with different forms of the boundary condition to find that this choice does not have a great influence on the results .
equation ( [ eq : main ] ) was solved by the alternating direction implicit ( adi ) method on a 64@xmath1864 grid .
the grid we use is generally uniform in both @xmath19 and @xmath20 .
however , immediately below the surface the scale heights of the density and of the turbulent velocity are very small . a grid uniform in @xmath20 is unable to resolve this fine structure and the corresponding potentially important pumping effects .
for this reason , in some calculations a non - uniform grid is employed , resolving these layers .
( in practice this is realized by introducing a new variable @xmath21 , rewriting eq .
( [ eq : main ] ) in @xmath22 , and solving it on a grid uniform in @xmath22 . )
it is instructive to consider the case when the lower boundary condition corresponds to the vector potential at fixed @xmath23 mm of a time - independent dipole field centered on the solar center .
the cross section of the resulting field configuration after reaching a stationary state is shown in fig . [
fig : dipcross ] .
field lines are nearly radial , as a consequence of diffusivity .
the effects of rotation and meridional circulation are negligible ( fig .
[ fig : dipbtheta ] ) .
the overall field strength varies only by a factor of 2 between the surface and the bottom of the convective zone .
the horizontal field component , however , increases by three orders of magnitude from the surface to the bottom , as a consequence of the pumping ( fig .
[ fig : horizfield ] ) .
this is the only case where the effect of rotation is not quite negligible ; besides , it is apparent that a proper resolution of the shallow layers reduces the surface value by nearly one order of magnitude .
the overall field structure is however not greatly distorted by a uniform grid , neither here , nor in any of the other cases studied .
the case when the lower boundary condition is a dipole with a dipole moment oscillating sinusoidally around zero was also considered , in order to see whether transport effects can turn this `` standing dynamo wave '' into a migrating wave .
the result was negative : owing to the overwhelming diffusivity , the surface field pattern simply reflects the pattern at the bottom with a slight time delay .
this finding is contrary to the result of @xcite ( @xcite ) who in a simplified model found that the effect of pumping , while weak , is sufficient to turn a high - latitude standing wave component into a migrating wave . while differences in the details of the models may also contribute to the different results , we believe that perhaps the most important difference is that in kichatinov s model the region of flux transport spatially coincides with the region of the dynamo ( i.e. it is a convection zone dynamo ) .
the lower boundary condition here was @xmath24 where @xmath25\right\}^{-1}\times \nonumber \\ & & \cos\left[\omega t+2\pi/\lambda_i ( \pi/2-\theta)+\delta_i\right]\end{aligned}\ ] ] with @xmath26/s the cycle frequency .
-0.8 cm the butterfly diagrams at the surface essentially reflect those at the bottom of the convective zone ( figs .
[ fig:3bfly ] and [ fig:4bfly ] ) .
a high - latitude poleward branch and a low - latitude equatorward branch can thus only be produced at the surface if such a pattern is given as input at the bottom .
figure [ fig:4cross ] shows that the field lines do not show a complicated topology inside the convective zone , although they are not exactly vertical either as the time dependence does not allow sufficient time for diffusion to smooth them . as a consequence ,
the strong downwards pumping of the horizontal field component is to some extent also `` felt '' by the vertical component , and , in contrast to the stationary dipole case , the overall field strength is also significantly concentrated to the bottom of the convective zone ( fig .
[ fig:|b|r ] ) .
in our axisymmetric model of the passive transport of the large - scale mean poloidal magnetic field in the solar convective zone we found that the latitudinal distribution of the field at the surface reflects the conditions at the bottom of the convective zone , i.e. in this regard the convective zone behaves as a `` steamy window '' .
this is due to the fact that with realistic values of the transport coefficients diffusivity is prevalent over all other effects .
passive transport theories of the origin of the poleward branch of the solar butterfly diagram are thus not viable , while double wave models are supported by these results .
owing to the large diffusivity the phase delay of the surface poloidal field relative to the bottom poloidal field is minimal . coupled with the approximately @xmath27 phase difference between the toroidal and poloidal field components for negative @xmath1 value ( @xcite , @xcite )
, this could bring the double wave models to accordance with the observed phase relationship .
a mixed transport scenario of the kind proposed by @xcite ( @xcite ) is not ruled out ( note that our figs .
[ fig : dipcross ] and [ fig:4cross ] agree with the field structure suggested in that paper ) , but the large diffusivity implies that the photospheric patterns should pervade the whole convective zone , being continuously reprocessed through it .
note that in fact the active source used in those models may also extend to deeper layers of the zone , as emerging magnetic loops are thought to shred a significant amount of flux during their rise ( @xcite , @xcite ; @xcite , @xcite ; @xcite , @xcite ) .
the vertical distribution of the field ( figs . [
fig : horizfield ] and [ fig:|b|r ] ) is strongly concentrated to the bottom of the convective zone owing to pumping effects .
this is in agreement with the earlier proposal of @xcite ( @xcite ) and the one - dimensional results of @xcite ( @xcite ) and @xcite ( @xcite ) .
the effective pumping may contribute to restricting the dynamo to the overshooting layer and it can reduce the surface field by about an order of magnitude , thereby offering a solution to the overtly high surface poloidal field values found by @xcite ( @xcite ) in their double wave model . note
that a significant part of this pumping effect occurs in a shallow subsurface layer , normally not resolved in dynamo models . in summary ,
our results support the double wave models , may be compatible with some form of the mixed transport scenario , and exclude the passive transport theory . in order to decide between the double wave and mixed transport models ,
two key issues should be solved in the future .
first , note that in the mixed transport model the direction of migration is opposite for poloidal and toroidal fields but independent of latitude .
thus , if e.g. a clear signature of migrating high - latitude toroidal fields were found , this could solve the problem in either way , depending on the direction of migration .
@xcite ( @xcite ) claims that high latitude ephemeral active regions show an equatorward migration , while @xcite ( @xcite ) claim that at least 50% of polar faculae ( well known for their poleward drift ) correspond to dipoles with a preferential east west orientation , thus forming part of the toroidal field .
it has even been claimed that the highest - latitude part of the sunspot butterfly diagram also shows a poleward drift ( @xcite , @xcite ) .
a clarification of this issue would clearly be important .
the role of causality relations between the two branches should also be clarified in both theories .
surges in low - latitude activity are followed by poleward surges of the high - latitude field , thus supporting the mixed transport scenario . on the other hand ,
the well - known fact that the level of low - latitude activity in a sunspot cycle can be predicted from high - latitude fields at the end of the previous cycle suggests a causal relationship of the opposite sense .
it is not impossible that such causal relations can be accommodated in both scenarios , but this remains to be seen .
the ukx model assumed fixed anisotropy @xmath28 for the diffusivity tensor .
we assume that the effect of rotation on the diffusivity remains the same as for quasi - isotropic turbulence ( @xcite , @xcite ) : @xmath29 @xmath30 @xmath31 @xmath32 @xmath33 where @xmath34 is the r.m.s
. radial component of the turbulent velocity in the ukx model , @xmath35 is the pressure scale height , and @xmath36 is the surface plasma differential rotation law . note that the expression of @xmath37 used is valid for non - cellular flows only ( free random walk )
. the cellular nature of the supergranular flow on the solar surface may reduce the diffusivity there ( @xcite , @xcite ) , but even this reduced value is well in excess of @xmath38km@xmath39/s and the reduction should only be confined to the photosphere , as a similar cellular flow is not expected in the deeper layers .
the general expressions of the pumping velocities for weak magnetic fields are @xmath40 @xmath41 @xmath42^{1/2 } \\ & & { } _ 1{\cal a}^3_2= 1+\left [ 1+(\phi_2/\phi_3)\sin^2\theta\right]^{-1/2}\end{aligned}\ ] ] where @xmath43 , @xmath44 , @xmath45 and @xmath46 account for ( small ) higher - order and intermittency corrections ( @xcite , @xcite , @xcite , @xcite ) ; these corrections are of minor importance .
note that the magnetic field strength does not exceed about 10 g at any point in our computational volume ( if we assume that the surface fields are about 1 g as observed ) ; so buoyancy effects were altogether neglected here . | the axisymmetric component of the large - scale solar magnetic fields has a pronounced poleward branch at higher latitudes . in order to clarify the origin of this branch we construct an axisymmetric model of the passive transport of the mean poloidal magnetic field in the convective zone , including meridional circulation , anisotropic diffusivity , turbulent pumping and density pumping . for realistic values of the transport coefficients
we find that diffusivity is prevalent , and the latitudinal distribution of the field at the surface simply reflects the conditions at the bottom of the convective zone . pumping effects concentrate the field to the bottom of the convective zone ; a significant part of this pumping occurs in a shallow subsurface layer , normally not resolved in dynamo models .
the phase delay of the surface poloidal field relative to the bottom poloidal field is found to be small .
these results support the double dynamo wave models , may be compatible with some form of a mixed transport scenario , and exclude the passive transport theory for the origin of the polar branch .
= 10000 = 10000 = 19.6 cm = 12.6 cm + k. petrovay + instituto de astrofsica de canarias + la laguna , tenerife , e-38200 spain + e - mail : [email protected] |
one of the most exciting challenges in today s nuclear structure is the physics of exotic nuclei far from the line of @xmath0-stability .
what makes this subject particularly interesting ( and difficult ) is the unique combination of weak binding and the proximity of the particle continuum , both implying the large diffuseness of the nuclear surface and extreme spatial dimensions characterizing the outermost nucleons @xcite . for the weakly bound nuclei the decay channels have to be considered explicitly .
due to the virtual scattering of nucleons from bound orbitals to unbound scattering states , the traditional shell - model technology becomes inappropriate .
the proper tool is the continuum shell model @xcite which correctly accounts for the coupling to resonances ; the single - particle basis of the continuum shell model consists of both bound and unbound states .
the explicit coupling between bound states and the continuum and the presence of low - lying low-@xmath1 scattering states invites strong interplay between various aspects of nuclear structure and reaction theory . particularly exciting are new phenomena on the neutron - rich side . because neutrons do not carry an electric charge ,
the neutron drip line is located very far from the valley of @xmath0-stability .
consequently , the neutron drip - line systems ( i.e. , those close to the neutron drip line ) are characterized by unusually large @xmath2/@xmath3 ratios . the outer zone of these nuclei are expected to constitute essentially a new form of a many - body system : low - density neutron matter ( neutron halos and skins ) .
except for the lightest nuclei , the bounds of neutron stability are not known experimentally . theoretically , because of their sensitivity to various theoretical details ( e.g. , approximations used , parameter values , interactions ) predicted drip lines are strongly model - dependent .
the placement of the one - neutron drip line , defined by the condition @xmath4 , is solely determined by the binding energy difference between two neighboring isotopes .
analogously , the vanishing two - neutron separation energy , @xmath5 , defines the position of the two - neutron drip line . since experimental masses ( binding energies ) near the neutron drip lines are unknown , in order to extrapolate far from stability , the large - scale mass calculations are usually used ( see , e.g. , @xcite ) .
however , since their techniques and parameters are optimized to reproduce known atomic masses , it is by no means obvious whether the particle number dependence obtained from global calculations at extreme values of @xmath6 is correct .
apart from strong theoretical and experimental interest in nuclear physics aspects of exotic nuclei , calculations for nuclei far from stability have strong astrophysical implications , especially in the context of the r - process mechanism @xcite . in previous work
@xcite several aspects of nuclear structure at the limits of extreme isospin were discussed by means of the macroscopic - microscopic approach . in the present study , ground - state properties of drip - line systems and the sensitivity of predictions to effective forces
are investigated by means of the self - consistent hartree - fock - bogoliubov ( hfb ) approach .
the paper is organized as follows .
section [ forces ] discusses the effective interactions employed in this study . since pairing correlations are crucial for the behavior of drip - line systems ,
particular attention is paid to the particle - particle ( p - p , pairing ) component of the interaction . after a short review of general properties of effective pairing interactions , with emphasis on the density dependence , the pairing forces investigated in our work , namely contact forces ( delta interaction , density - dependent delta interaction , and skyrme interaction ) and the finite - range gogny force ,
are described .
the basic ingredients of the hfb formalism in the coordinate representation ( single - quasiparticle orbitals , time - reversal symmetry , canonical states , and various densities ) are defined in sec .
[ sec2a ] .
in contrast to the single - quasiparticle wave functions which often contain a scattering ( outgoing ) component , canonical states ( sec .
[ sec2ab ] ) are always localized , even if they have positive average energy .
the interpretation of particle and ( especially ) pair densities in terms of single - particle and correlation probabilities is given in sec .
[ sec2ac ] .
this interpretation is essential when relating the calculated hfb densities and fields to various experimental observables .
the structure of the hfb equations is analyzed in sec .
[ sec2b ] . here
, various functions entering the equations of motion ( i.e. , mass parameters and mean field potentials ) are introduced for both p - h and p - p channels ( secs .
[ sec2ba ] and [ sec2bap ] ) .
the advantage of using the coordinate - space hfb formalism for weakly bound systems is that in this method the particle continuum is treated properly .
this important point is discussed in detail in sec .
in particular , the difference between the single - particle hartree - fock ( hf ) spectra and canonical hfb spectra ( sec .
[ sec3d ] ) , the asymptotic properties of the hfb states ( sec . [ sec2bcd ] ) and densities ( sec . [ sec3c ] ) , and the effect of the pairing coupling to positive - energy states ( sec . [ sec3e ] ) are carefully explained .
the robust predictions of the formalism for various experimental observables ( pairing gaps and pair transfer amplitudes , masses and separation energies , radii , shell gaps , and shell structure ) are reviewed in sec .
[ sec4 ] , where experimental fingerprints of the surface - peaked pairing fields and the quenching of shell effects far from stability are also given .
section [ sec6 ] contains the main conclusions of the paper .
the technical details ( i.e. , the form of a mean - field gogny hamiltonian and the discussion of the energy cut - off in the skyrme model ) are collected in the appendices .
the uniqueness of drip - line nuclei for studies of effective interactions is due to the very special role played by the pairing force .
this is seen from approximate hfb relations between the fermi level , @xmath7 , pairing gap , @xmath8 , and the particle separation energy , @xmath9 @xcite : @xmath10 since for drip - line nuclei @xmath9 is very small , @xmath11@xmath120 . consequently , the single - particle field characterized by @xmath7 ( determined by the p - h component of the effective interaction ) and the pairing field @xmath8 ( determined by the p - p part of the effective interaction ) are equally important . in other words ,
contrary to the situation encountered close to the line of beta stability , the pairing component can no longer be treated as a _
residual _ interaction ; i.e. , a small perturbation important only in the neighborhood of the fermi surface .
surprisingly , rather little is known about the basic properties of the p - p force . in most calculations ,
the pairing hamiltonian has been approximated by the state - independent seniority pairing force , or schematic multipole pairing interaction @xcite .
such oversimplified forces , usually treated by means of the bcs approximation , perform remarkably well when applied to nuclei in the neighborhood of the stability valley ( where , as pointed out above , pairing can be considered as a small correction ) . as a result ,
considerable effort was devoted in the past to optimizing the p - h part of the interaction , while leaving the p - p component aside .
up to now , the microscopic theory of the pairing interaction has only seldom been applied in realistic calculations for finite nuclei ( see ref .
@xcite for a recent example ) .
a first - principle " derivation of pairing interaction from the bare @xmath13 force using the renormalization procedure ( @xmath14-matrix technique ) , still encounters many problems such as , e.g. , treatment of core polarization @xcite .
hence , phenomenological pairing interactions are usually introduced .
two important open questions asked in this context are : ( i ) the role of finite range , and ( ii ) the importance of density dependence .
since the realistic effective interactions are believed to have a finite range , the first question seems purely academic .
however , the remarkable success of zero - range skyrme forces suggests that , in many cases , the finite - range effect can be mocked up by an explicit velocity dependence . to
what extent this is true for the pairing channel remains to be seen
. one obvious advantage of using finite - range forces is the automatic cut - off of high - momentum components ; for the zero - range forces this is solved by restricting the pair scattering to a limited energy range and by an appropriate renormalization of the pairing coupling constant ( see appendix [ appb ] ) .
the answer to the question on the density dependence is much less clear .
early calculations @xcite for nuclear matter predicted a very weak @xmath15 pairing at the saturation point ( @xmath16=1.35@xmath17 ) .
consequently , it was concluded that strong pairing correlations in finite nuclei had to be due to interactions at the nuclear surface .
this led to the surface delta interaction ( sdi ) @xcite , a highly successful residual interaction between valence nucleons .
of course , the sdi is an extreme example of surface interaction .
more realistic density - dependent pairing forces are variants of the density - dependent delta interaction ( dddi ) introduced in the migdal theory of finite fermi systems @xcite . since the effective interactions commonly used in the hf calculations are bound to be density dependent in order to reproduce the compressibility of the infinite nuclear matter @xcite ( an explicit density dependence
is also said to account for three- and higher - body components of the interaction ) , it seems natural to introduce the density dependence in the p - p channel as well @xcite .
interestingly , the presence ( absence ) of the density dependence in the pairing channel has consequences for the spatial properties of pairing densities and fields .
as early recognized @xcite , the density - independent p - p force gives rise to a pairing field that has a volume character .
for instance , the commonly used contact delta interaction , @xmath18 leads to volume pairing . by adding a density - dependent component ,
the pairing field becomes surface - peaked .
a simple modification of force ( [ didi ] ) is the dddi @xcite @xmath19^\gamma\right\},\ ] ] where @xmath20 is the isoscalar nucleonic density , and @xmath21 and @xmath22 are constants . if @xmath23 is chosen such that it is close to the saturation density , @xmath23@xmath12@xmath24 , both the resulting pair density and the pairing potential @xmath25 ( see secs .
[ sec2ac ] and [ sec2ba ] ) are small in the nuclear interior . by varying the magnitude of the density - dependent term
, the transition from volume pairing [ @xmath23@xmath26@xmath27 to surface pairing can be probed .
what are the experimental arguments in favor of surface pairing ?
probably the strongest evidence is the odd - even staggering in differential radii , explained in terms of the direct coupling between the proton density and the neutron pairing tensor @xcite .
other experimental observables which strongly reflect the spatial character of pairing are the particle widths and energies of deep - hole states @xcite , strongly influenced by the pairing - induced coupling to the particle continuum , and the pair transfer form factors , directly reflecting the shape of the pair density . because of strong surface effects , the properties of weakly bound nuclei are sensitive to the density dependence of pairing . in particular , the same type of force is used to describe the spatial extension of loosely bound light systems @xcite .
( the measurable fingerprints of surface pairing in neutron - rich systems are further discussed in sec .
[ sec4 ] . ) in this context , it is also worth mentioning that the self - consistent model with the dddi has recently been used to describe the nuclear charge radii @xcite and the moments of inertia of superdeformed nuclei @xcite . in the latter case ,
the inclusion of a density dependence in the p - p channel turned out to be crucial for the reproduction of experimental data around @xmath28hg . in a series of papers @xcite the quasiparticle lagrangian method ( qlm )
@xcite based on the single - particle green function approach in the coordinate representation @xcite has been applied to the description of nuclear superfluidity .
the resulting pairing interaction , based on the landau - migdal ansatz @xcite , has zero - range and contains two - body and three - body components , thus leading to a density - dependent contact force similar to that of eq .
( [ dddi ] ) .
( note that in the approximations of ref .
@xcite , the _ neutron _ pairing interaction is proportional to the _ proton _ density and vice versa . ) however , in practical qlm calculations @xcite , a pure density - independent delta force was used .
a better understanding of the density dependence of the nuclear pairing interaction is important for theories of superfluidity in neutron stars . as pointed out in ref .
@xcite , it is impossible at present to deduce the magnitude of the pairing gaps in neutron stars with sufficient accuracy .
indeed , calculations of @xmath15 pairing gaps in pure neutron matter , or symmetric nuclear matter , based on bare @xmath13 interactions @xcite suggest a strong dependence on the force used ; in general , the singlet-@xmath9 pairing is very small at the saturation point .
on the other hand , nuclear matter calculations with an effective finite - range interaction , namely the gogny force @xcite , yield rather large values of the pairing gap at saturation ( @[email protected] ) .
( for relativistic hfb calculations for symmetric nuclear matter , see ref .
the pairing properties of the skyrme force in nuclear matter were investigated in ref .
see also ref .
@xcite for schematic calculations of pairing properties in nuclear matter based on the green function method with a contact interaction , and ref .
@xcite for a semi - classical description of neutron superfluidity in neutron stars using the gogny force . ) in this study , several self - consistent models based upon the hfb approaches are used .
the effective interactions employed , and other model parameters , are briefly discussed below .
the spherical hfb - skyrme calculations have been carried out in spatial coordinates following the method introduced in ref .
@xcite and discussed in detail in secs .
[ sec2a]-[sec3 ] .
several effective skyrme interactions are investigated .
these are : ( i ) the skyrme parametrization skp introduced in ref .
@xcite ( skp has exactly the same form in the particle - hole ( p - h ) and pairing channels ) ; ( ii ) skyrme interaction skp@xmath30 of ref .
@xcite ( in the p - h channel , this force is the skp skyrme parametrization , while its pairing component is given by delta interaction , eq .
( [ didi ] ) ; ( iii ) the skyrme interaction skp@xmath31 of ref .
@xcite ( in the p - h channel , this force is the skp skyrme parametrization , while its pairing component is given by eq .
( [ dddi ] ) ; ( iv ) the force siii@xmath30 ( in the p - h channel , this is the siii skyrme parametrization @xcite ; its pairing component is given by the delta force of ref .
@xcite ) ; ( v ) the force skm@xmath30 ( in the p - h channel , this is the skm@xmath32 skyrme parametrization @xcite , and its pairing part is given by the delta force with the parameters of ref .
@xcite ) .
apart from other parameters , the above skyrme forces differ in their values of the effective mass for symmetric nuclear matter , @xmath33 .
namely , @xmath33 is 0.76 , 0.79 , and 1 for siii , skm@xmath32 , and skp , respectively .
all hfb - skyrme results have been obtained using the pairing phase space as determined in ref .
@xcite ( see also discussion in appendix [ appb ] ) .
a set of spherical hfb calculations has also been performed using the finite - range density - dependent gogny interaction d1s of ref .
@xcite . in this effective interaction
@xcite the central part consists of four terms parametrized with finite - range gaussians ( see appendix [ appa ] ) .
spin - orbit and density - dependent terms of zero range are also included as in the skyrme parametrizations .
the pairing field is calculated from the d1s force , i.e. , the same interaction is used for a microscopic description for both the mean field and the pairing channels .
however , by a specific choice of the exchange contribution , the pairing component of the d1s is density - independent .
it is also interesting to note that the pairing component of the d1s is repulsive at short distances and attractive at long ranges @xcite . for the d1s force ,
the effective mass for infinite nuclear matter is @xmath33=0.70 .
the parameters of the d1s interaction were chosen to reproduce certain global properties of a set of spherical nuclei and of nuclear matter @xcite .
the hfb+gogny results presented here were obtained by expanding the hfb wave functions in a harmonic oscillator basis containing up to 19 shells .
the hfb approach is a variational method which uses independent - quasiparticle states as trial wave functions .
these states are particularly convenient when used in a variational theory , because , due to the wick theorem @xcite , one can easily calculate for them the average values of an arbitrary many - body hamiltonian .
even if the exact eigenstates of such a hamiltonian can be rather remote from any one of the independent - quasiparticle states , one can argue @xcite that one may obtain in this way fair estimates of at least one - body observables .
an independent - quasiparticle state is defined as a vacuum of quasiparticle operators which are linear combinations of particle creation and annihilation operators .
this linear combination is called the bogoliubov transformation @xcite . according to the thouless theorem @xcite ,
every independent - quasiparticle state @xmath34 , which is not orthogonal to the vacuum state @xmath35 , i.e. , @xmath36@xmath370 , can be presented in the form @xmath38 where the thouless matrix @xmath3 is antisymmetric , @xmath39 , and in general complex .
the phase of state ( [ eq101 ] ) is fixed by the condition @xmath36=1 ; the norm is given by @xmath40=@xmath41 . in the following
, state @xmath34 will represent the @xmath42=0@xmath43 ground state of the even - even system .
we refer to standard textbooks @xcite for a discussion of the properties of the bogoliubov transformation . here
we start our discussion from the trial wave function ( [ eq101 ] ) which is parametrized by the matrix elements of @xmath3 .
this form of the independent - quasiparticle state is very convenient in variational applications because variations with respect to all matrix elements @xmath44=@xmath45 are independent of one another . instead of using the matrix representation corresponding to a set of single - particle creation operators
@xmath46 numbered by the discrete index @xmath47 , one may use the spatial coordinate representation .
this is particularly useful when discussing spatial properties of the variational wave functions and the coupling to the particle continuum .
therefore , in the following , we shall consider the operators creating a particle in the space point @xmath48 and having the projection of spin @xmath49=@xmath50 , @xmath51 where @xmath52 is the wave function of the @xmath47-th single - particle state . to simplify the following expressions , we consider only one type of particle .
a generalization to systems described by a product of neutron and proton wave functions is straightforward , while that involving the mixing in the isospin degree of freedom is discussed in ref .
@xcite .
the inverse relation with respect to ( [ eq103 ] ) is given by @xmath53 equations ( [ eq103 ] ) and ( [ eq104 ] ) assume that the wave functions @xmath52 form an orthonormal and complete set . in practical calculations ,
the basis has to be truncated and the completeness is realized only approximately .
the choice of the single - particle wave functions used ( size of the set and , in particular , the asymptotic behavior ) is of crucial importance to the phenomena discussed in this study . in coordinate space , the thouless state ( [ eq101 ] )
has the form @xmath54 and is defined by the antisymmetric complex function , @xmath55= @xmath56 , of space - spin coordinates .
already , at this point , we see that any variational method employing an attractive effective interaction for a _ bound finite system _ must lead to functions which are localized in space , @xmath57 recall that in the coordinate space , values of the function @xmath58 at different space - spin points are the variational parameters , and that any arbitrarily small value of this function at large distance , @xmath59 , would create at this point a non - zero probability density .
whether this would be energetically favorable depends upon the number of particles in the system and on the interaction used in the variational method .
apart from exotic phenomena such as halos , and apart from infinite matter such as in the neutron - star crust , we assume that the attractiveness of the interaction always favors compact , localized probability densities , and hence we require the localization condition ( [ eq106 ] ) for the variational parameters @xmath58 .
an expansion of the variational function @xmath58 in terms of the single - particle wave functions is a straightforward consequence of transformations ( [ eq103 ] ) and ( [ eq104 ] ) , @xmath60 the localization condition , eq .
( [ eq106 ] ) , can , therefore , be guaranteed in the most economic way by requiring that _ all _ single - particle wave functions @xmath52 vanish at large distances .
of course , this is only a matter of convenience and manageability , because any localized function can be expanded in any complete basis .
it is , however , obvious that such an expansion converges very slowly if the basis has inappropriate asymptotic properties .
for example , one can expect that a plane - wave expansion of @xmath58 would require an infinite number of basis states @xmath52 , and in practice , any reduction to a finite basis would lead to serious errors . a discussion pertaining to asymptotic properties of functions in spatial coordinates , and the choice of an appropriate single - particle basis , will be a pivotal point in our study .
the present study is entirely restricted to an analysis of ground - state phenomena , and therefore , we use only time - even variational independent - quasiparticle wave functions . the time - reversal operator can be represented as a product of the spin - flip operator and the complex conjugation ; i.e. , @xmath61=@xmath62 @xcite .
the explicit time - reversed creation operators then have the form @xmath63 a^+_{\bbox{r}\sigma}. \label{eq108b } \end{aligned}\ ] ] we now suppose that the set of basis states represented by the creation operators @xmath46 is closed with respect to time reversal , and that the state @xmath64 is actually proportional ( up to a phase factor @xmath65=@xmath66 , @xmath67=1 ) to another basis state denoted by a bar over the greek symbol , i.e. , @xmath68 in this way , the single - particle basis is assumed to be composed of pairs of time - reversed states denoted by indices @xmath47 and @xmath69 . in what follows ,
we use the convention that @xmath70@xmath71@xmath47 , and that the sums over either @xmath47 or @xmath69 are always performed over _ all _ basis states .
the phase factors @xmath72 depend on relative phases chosen for the @xmath47-th and @xmath69-th states of the basis ; it is convenient to keep them unspecified in all theoretical formulae and to make a definite suitable choice of the phase convention only in a specific final application .
a requirement of the time - reversal symmetry of the quasiparticle vacuum ( [ eq101 ] ) or ( [ eq105 ] ) , @xmath73=@xmath34 , leads to the following conditions : @xmath74 these properties allow the introduction of more suitable forms of @xmath44 and @xmath58 ; namely , @xmath75 the matrix @xmath76 and the function @xmath77 are both time - even and hermitian , @xmath78 and therefore they can be considered as usual operators in the corresponding hilbert spaces . in particular , the function @xmath79 can be diagonalized by solving the following integral eigenequation : @xmath80 where @xmath81 are real eigenvalues , @xmath81=@xmath82 .
the eigenfunctions @xmath83 form the single - particle basis , usually referred to as the _ canonical basis_. canonical states , together with the eigenvalues @xmath81 , completely define the quasiparticle vacuum @xmath34 .
( here and in the following we use the checked symbols , e.g. , @xmath84 and @xmath85 , to denote objects pertaining to the canonical basis . )
two important remarks concerning the canonical basis are now in order .
first , the localization condition ( [ eq106 ] ) directly results in the fact that _ all _ canonical - basis single - particle wave functions @xmath83 are localized in space ; i.e. , vanish at large distances , @xmath59 .
therefore , as discussed previously , a choice of the localized wave functions for the basis states @xmath86 may allow for a rapid convergence in the expansion @xmath87 second , since @xmath88 and @xmath76 are related by [ cf .
( [ eq107 ] ) ] @xmath89 a diagonalization of @xmath90 , eq .
( [ eq113 ] ) , is equivalent to a diagonalization of @xmath76 , @xmath91 therefore , in the canonical basis , the thouless state ( [ eq101 ] ) acquires the well - known separable bcs - like form @xmath92 & = & \exp\left\{\sum_{\nu>0 } z_{\nu}s_\nu \breve a^+_{\bar\nu } \breve a^+_\nu\right\ } |0\rangle \nonumber \\ & = & \prod_{\nu>0}\left(1 + z_{\nu}s_\nu \breve a^+_{\bar\nu } \breve a^+_\nu\right ) |0\rangle , \label{eq116 } \end{aligned}\ ] ] where the symbol @xmath93 denotes the sum over one - half of the basis states with only one state ( either one ) of each time - reversed pair ( @xmath94 ) included , and @xmath95 is the creation operator in the canonical basis . according to the wick theorem @xcite for the independent - quasiparticle state , eqs .
( [ eq101 ] ) or ( [ eq105 ] ) , an average value of any operator can be expressed through average values of bifermion operators , @xmath96 functions @xmath97 and @xmath98 are called the particle and pairing density matrices , respectively . for a time - reversal invariant state @xmath34 ,
both density matrices are time - even and hermitian : [ 7ab]@xmath99 therefore , the pairing density matrix @xmath98 is more convenient to use than the standard pairing tensor @xmath100 @xcite , @xmath101 which is an antisymmetric function of the space - spin arguments .
the formulae expressing @xmath97 and @xmath98 in terms of the function @xmath88 can be easily derived from those for the density matrix and the pairing tensor @xcite , and they read [ eq120]@xmath102 as a result , the density matrices obey the following relations : [ eq121]@xmath103 in the above equations , the matrix multiplications and inversions should be understood in the operator sense ; i.e. , they involve the integration over space and summation over spin variables .
for instance : @xmath104 _ local _ hfb densities , i.e. , the density matrices for equal spatial arguments , @xmath105=@xmath48 , have very well - defined physical interpretations . to see this ,
let us assume that @xmath106 is a normalized single - particle wave function ( wave packet ) concentrated in a small volume @xmath107 around the point @xmath48=@xmath108 and having the spin @xmath109=@xmath49 .
the corresponding creation operator @xmath110 together with its hermitian conjugate , define the operator @xmath111 which measures the number of particles in the vicinity of the point @xmath108 . since @xmath112 @xmath113 can be regarded as a projection operator which projects out the component of the many - body wave function that contains one spin-@xmath109 fermion in the volume @xmath107 .
therefore , its average value gives _ the probability to find a particle with spin @xmath109 in this volume _ : @xmath114 in a very similar way , the probability of finding a fermion in @xmath107 having opposite spin can be obtained by considering the time - reversed wave function @xmath115 , cf .
( [ eq108b ] ) .
this gives @xmath116 due to time - reversal symmetry , probabilities ( [ eq128 ] ) and ( [ eq129 ] ) are equal .
we may now ask the question , what is _ the probability of finding a pair of fermions with opposite spin projections _ in the volume @xmath107 , @xmath117 ? " . if one considers two _ independent _ measurements , where in the first one is found the spin-@xmath109 fermion , and in another one the spin-(@xmath118 ) fermion , @xmath117 is equal to the product of individual probabilities ; i.e. , @xmath119 .
on the other hand , if one wants to find in @xmath107 both fermions _ simultaneously _ , one should project out from @xmath34 a corresponding two - fermion component . in this case
, @xmath117 becomes the expectation value of the product of the projection operators @xmath113 and @xmath120 ; i.e. , @xmath121 . using the wick theorem , this average value is @xmath122 or in terms of the time - even spin - averaged densities : @xmath123 for [ eq320]@xmath124 since the first terms in eqs .
( [ eq132a ] ) and ( [ eq132b ] ) describe the probability of finding the two fermions in independent measurements , the second terms in these equations should be interpreted as the probability of finding _ the correlated pair _ at point @xmath108 .
the above arguments allow us to give a transparent physical interpretation to the local hfb densities .
namely , as usual , @xmath20 represents the probability density of finding a particle at the given point . on the other hand
, @xmath125 gives the correlation probability density ; i.e. , the probability of finding a pair of fermions _ in excess _ of the probability of finding two uncorrelated fermions .
it is important to note that kinematic conditions ( [ eq121 ] ) , which result from the fact that @xmath34 is an independent - quasiparticle state , eq .
( [ eq105 ] ) , _ do not _ directly constrain the local values of the particle and pairing density matrices . in particular , there is no obvious kinematic relation between the probability of finding two independent particles at a given point of space , and the probability of finding a correlated pair at the same point . in particular , the first one can be small , while the second can be large ( see discussion in secs .
[ sec2aca ] and [ sec3c ] ) .
this result means that in such a situation the experiments probing the presence of two particles will always find these two particles as correlated pairs without a `` background '' characteristic of two independent particles .
relations ( [ eq120 ] ) imply that all three functions : @xmath88 , @xmath97 , and @xmath98 are diagonal in the canonical basis , cf .
( [ eq113 ] ) . using the standard notation for the eigenvalues of @xmath126 and @xmath127 , one obtains [ eq122]@xmath128 where the real factors @xmath129 and @xmath130 are given by @xmath131 a completeness of the canonical basis leads to standard expressions for the density matrices : [ eq321]@xmath132 equation ( [ eq122a ] ) represents the traditional definition of the canonical states as the eigenstates of the hfb density matrix .
it also shows that the canonical states are the _ natural states _
@xcite for the density matrix corresponding to the independent - quasiparticle many - body state @xmath34 , eq .
( [ eq118a ] ) , and the eigenvalues @xmath133 are the corresponding natural occupation numbers .
one may now easily repeat the previous analysis of probabilities of finding a particle , or a pair of particles , in the canonical - basis single - particle state @xmath83 .
the result , analogous to eqs .
( [ eq132a ] ) and ( [ eq132b ] ) , is @xmath134=@xmath133 , and @xmath135=@xmath136+@xmath137 . in this case ,
due to the normalization condition @xmath138+@xmath133=1 , @xmath134=@xmath135 .
this result means that the particles in the canonical states with indices @xmath47 and @xmath139 are extremely correlated spatially ; i.e. , the probability of finding the canonical pair , @xmath140 , is directly dependent on the probability of finding two independent canonical fermions , @xmath137 .
however , as discussed above , a similar direct relation between @xmath141 and @xmath142 does not exist .
in particular @xmath141@xmath143@xmath142 . figures [ fig16 ] and [ fig17 ] display the particle and pairing local spherical neutron hfb densities @xmath144 and @xmath145 , eq .
( [ eq320 ] ) , as functions of the radial coordinate @xmath146=@xmath147 .
results are shown for several tin isotopes across the stability valley . for particle densities ,
the results obtained with the skp and skp@xmath148 interactions are almost indistinguishable .
therefore , fig .
[ fig16 ] ( middle panel ) shows results for the siii@xmath148 interaction . for pairing densities ,
compared in fig .
[ fig17 ] are results for skp , skp@xmath148 , and d1s effective interactions .
the particle densities obtained with these three effective interactions are qualitatively very similar .
one can see that adding neutrons results in a simultaneous increase of the central neutron density , and of the density in the surface region .
the relative magnitude of the two effects is governed by a balance between the volume and the surface asymmetry energies of effective interactions . since all three forces considered have been fitted in a similar way to bulk nuclear properties , including the isospin dependence , the resulting balance between the volume and the surface isospin effects is similar .
of course , this does not exclude some differences which are seen when a more detailed comparison is carried out .
the pairing densities shown in fig . [ fig17 ] reflect different characters of the interactions used in the p - p channel .
the contact force ( the skp@xmath148 results ) leads to the pairing densities which are , in general , largest at the origin and decrease towards the surface .
( this general trend is slightly modified by shell fluctuations resulting from contributions from orbitals near the fermi level . ) at the surface , the isospin dependence of skp@xmath148 is fairly weak .
for example , there is very little difference between the pairing densities in @xmath149sn and @xmath150sn .
these results are characteristic for the volume - type pairing correlations .
a different pattern appears for the skp results , where the density dependence renders the p - p interaction strongly peaked at the surface . in this case , the pairing densities tend to increase when going from the center of the nucleus towards its surface .
again , the shell fluctuations are superimposed on top of this general behavior . in particular ,
the central bump in the pairing density in @xmath151sn is due to a contribution from the 3@xmath152 state .
a more pronounced dependence on the neutron excess is seen in the surface region . especially near the drip line
, the pairing density develops a long tail extending towards large distances .
the results obtained for the finite - range interaction d1s exhibit intermediate features between the surface and the volume type of pairing correlations .
in particular , in the nuclear interior one observes a fairly large region of relatively constant pairing density .
the overall magnitude of the pairing densities is very similar in all three approaches . in particular
, it is interesting to see that at the nuclear surface ( @xmath146@xmath1535fm ) all three pairing densities in @xmath151sn are very close to 0.018@xmath154 .
we begin this section by presenting basic definitions and equations of the hfb approach .
the hfb theory is discussed in many textbooks and review articles ( see refs .
@xcite , for example ) , while its aspects pertaining to the coordinate representation have been presented in ref .
. an earlier discussion of the coordinate - representation hfb formalism has been given by bulgac , whose work is available only in preprint form @xcite .
recently , similar methods have also been applied to a description of light nuclei @xcite .
it is also worth mentioning that the green function approach in the coordinate representation ( the gorkov method @xcite ) , is formally equivalent to hfb cf . discussion in refs .
the only difference between the methods lies in the explicit energy dependence of the quasiparticle mass operator , an analog to the p - h single - particle hf hamiltonian ( see below ) .
the two - body effective hamiltonian of a nuclear system can be written in the coordinate representation as @xmath155 the first term represents the kinetic energy , while the second one is the two - body interaction . in the following , we assume that @xmath156 includes the exchange terms .
the average energy of the hamiltonian ( [ eq135 ] ) in a time - even hfb vacuum ( [ eq105 ] ) reads @xmath157 the last two terms are the interaction energies in the particle - hole ( p - h ) and in the particle - particle ( p - p ) channels , respectively .
equivalently , one can define the p - h and p - p single - particle hamiltonians , @xmath158 = @xmath159 + @xmath160 and @xmath161 , respectively : [ eq141]@xmath162 which gives the hfb energy in the form : @xmath163 additional terms coming from the density - dependence of the two - body interaction @xmath164 have been for simplicity omitted in eqs .
( [ eq141a ] ) , ( [ eq141b ] ) , and ( [ eq142 ] ) .
the last term in eq .
( [ eq142 ] ) , @xmath165 represents the pairing energy .
we also define the average magnitude of pairing correlations by the formula @xcite @xmath166 where @xmath167 is the number of particles ( neutrons or protons ) . the p - h and p - p mean fields ( [ eq141 ] ) have particularly simple forms for the skyrme interaction @xcite . in appendix [ appa ]
we present the form of the p - h and p - p mean - field hamiltonians in the case of a local two - body finite - range gogny interaction . in this section
we aim at comparing the self - consistent potentials obtained with the skyrme and gogny forces .
such a comparison can not be carried out directly , because the corresponding integral kernels @xmath158 and @xmath161 have different structure .
for the skyrme interaction , they are proportional to @xmath168@xmath169@xmath170 and depend also on the differential operators ( linear momenta ) @xcite , while for the gogny interaction they are sums of terms proportional to @xmath168@xmath169@xmath170 and terms which are functions of @xmath48 and @xmath105 ( appendix [ appa ] ) .
therefore , for the purpose of the present comparison we introduce operational prescriptions to calculate the local parts of the integral kernels : [ local]@xmath171 , \label{locala } \\ \tilde{u}(\bbox{r } ) & = & { \cal loc}\left [ \tilde{h}(\bbox{r}\sigma,\bbox{r'}\sigma')\right ] .
\label{localb } \end{aligned}\ ] ] these formal definitions in practice amount to : ( i ) disregarding the momentum - dependent terms of the kernels , ( ii ) considering only terms with @xmath49=@xmath172=1/2 ( which by time reversal symmetry are equal to those with @xmath49=@xmath172=@xmath1691/2 ) , and ( iii ) taking into account _ only _ the term proportional to @xmath168@xmath169@xmath170 , if such a term is present .
the expressions for @xmath173 and @xmath174 can be found in appendix a of ref .
@xcite ( skyrme interaction ) and in appendix a ( gogny interaction ) . in the skyrme calculations ,
the contribution of the coulomb interaction to @xmath174 has been neglected since it is estimated to be small . in the case of finite - range local interactions ( such as gogny or coulomb )
, the corresponding non - local pairing field @xmath175 does not contain the term proportional to @xmath168@xmath169@xmath170 ( see appendix [ appa ] ) .
consequently , the local field @xmath174 can not be extracted in a meaningful way . for instance , the diagonal ( i.e. , @xmath176=@xmath48 ) part of of the d1s pairing field is positive ; i.e. , it is dominated by the short - range repulsive component rather than the long - range attractive part @xcite . in the spherical case ,
the potentials @xmath173 and @xmath174 depend on only one radial coordinate @xmath146=@xmath147 .
this facilitates the qualitative comparison between different forces .
figure [ fig01a ] displays the self - consistent spherical local p - h potentials @xmath177 , eq .
( [ local ] ) , for several tin isotopes , calculated with skp , siii@xmath148 , and d1s interactions ( the results with skp@xmath148 are very close to those with skp ) .
the terms depending on the angular momentum , which result from a reduction to the radial coordinate , are not included .
( the general behavior of the self - consistent p - h potentials has already been discussed many times in the literature , e.g. @xcite , and we include these results only for completeness and for a comparison with the corresponding p - p potentials , for which the detailed analysis does not exist . )
qualitatively , the results for @xmath177 obtained with different effective forces are quite similar , which reflects the fact that all these interactions correctly describe global nuclear properties .
in particular , one sees that with increasing neutron excess the neutron potentials become more shallow in the interior and more wide in the outer region . interestingly , for each of these three forces there exists a pivoting point at which the potential does not depend on the neutron excess .
for the three forces presented , this occurs at @xmath146=5.9 , 4.6 , and 5.4fm , respectively .
the differences in the overall depths of the average potentials reflect the associated effective masses ( i.e. , the non - local contributions of the two - body interactions ) .
the analogous results for the p - p potentials @xmath178 calculated for the skp and skp@xmath148 interactions are shown in fig .
[ fig01b ] . on can see that the different character of pairing interactions is directly reflected in the form of the p - p potentials .
particularly noteworthy is the fact that the density - dependent pairing interaction in skp yields a very pronounced surface - peaked potential ( the behavior of @xmath178 at large distances is further discussed in sec . [ sec3e ] ) .
one can easily understand its form by recalling that this potential is equal to the product of the pairing density @xmath145 [ fig .
[ fig17 ] ] and the function which roughly resembles the behavior of dddi of eq .
( [ dddi ] ) ; i.e. , small in the interior and large in the outer region .
of course , values of @xmath145 and @xmath178 depend on each other by the fact that they both result from a self - consistent solution of the complete hfb equation in which the p - h and p - p channels are coupled together ( see sec .
[ sec2bc ] ) .
similar results were also obtained in refs .
@xcite ( in the hfb+skp model ) and @xcite ( in the qlm ) for the proton - rich rare - earth nuclei .
since the p - h channel provides the bulk part of the interaction energy , the particle densities @xmath144 closely follow the pattern of the p - h potentials ( i.e. , the density is large where the potential is deep ) .
an analogous relation is only partly true for @xmath145 and @xmath178 ; i.e. , even the dramatic surface character of the skp p - p potential ( fig .
[ fig01b ] ) does not result in the pairing density being similarly peaked at the surface .
recall that the contributions to @xmath145 come mainly from a few wave functions near the fermi surface , and that the form of these wave functions is mainly governed by the p - h channel .
since these wave functions must have significant components in the interior , the resulting pairing densities can not exactly fit into the surface - peaked p - p potentials .
nevertheless , a clear tendency towards surface localization is evident in fig .
[ fig17 ] . in the case of the pure contact interaction ( skp@xmath148 calculations )
the p - p potential is exactly proportional to the pairing density @xcite with the proportionality constant @xmath179 equal to @xmath16980mev@xmath180 @xcite .
therefore , the resulting potential is concentrated at the origin and increases towards the surface .
( early calculations of p - p potentials in the qlm with the density - independent delta interaction can be found in ref .
the general behavior of @xmath178 , denoted as @xmath181 therein , is very similar to our skp@xmath148 results . )
the variation of the hfb energy with respect to independent parameters @xmath58 leads to the hfb equation @xcite , @xmath182 where @xmath183 and @xmath184 are upper and lower components of the two - component single - quasiparticle hfb wave function , and @xmath7 is the fermi energy .
properties of the hfb equation in the spatial coordinates , eq .
( [ eq143 ] ) , have been discussed in ref .
in particular , it has been shown that the spectrum of eigenenergies @xmath185 is continuous for @xmath186@xmath187@xmath188 and discrete for @xmath186@xmath189@xmath188 . since for @xmath185@xmath1870 and @xmath7@xmath1890 the lower components @xmath184 are localized functions of @xmath48 , the density matrices , [ eq144]@xmath190 are always localized . for the case of a discretized continuum , sec .
[ sec3a ] , the integral over the energy reduces to a discrete sum @xcite but one should still carefully distinguish between contributions coming from the discrete ( @xmath191@xmath189@xmath188 ) and discretized ( @xmath191@xmath187@xmath188 ) states .
the orthogonality relation for the single - quasiparticle hfb wave functions reads @xmath192 = \delta_{n , n'}.\ ] ] it is seen from eq .
( [ orthog ] ) that the lower components are not normalized .
their norms , @xmath193 define the total number of particles @xmath194 in the hfb theory , the localization condition ( [ eq106 ] ) discussed in sec .
[ sec2a ] is automatically guaranteed for any system with negative fermi energy @xmath7 .
this allows studying nuclei which are near the particle drip lines where the fermi energy approaches zero through negative values . for the skyrme interaction , the hfb equation ( [ eq143 ] )
is a differential equation in spatial coordinates @xcite .
if the spherical symmetry is imposed , which is assumed in the following , this equation reads @xmath195 \left(\begin{array}{c } r\phi_1(e , r ) \\
r\phi_2(e , r ) \end{array}\right ) = e\left(\begin{array}{c } r\phi_1(e , r ) \\
r\phi_2(e , r ) \end{array}\right ) , \ ] ] where @xmath196 and @xmath197 are p - h and p - p mass parameters , respectively , and @xmath198 and @xmath199 are defined in sec .
[ sec2ba ] . due to the spherical symmetry ,
( [ eq152 ] ) is solved separately for each partial wave ( @xmath200 ) .
the potentials include also the centrifugal and spin - orbit terms , and the p - h mass parameter @xmath196 is expressed in terms of the effective mass @xmath201 ; i.e. , @xmath196=@xmath202 , see ref .
@xcite for details . before discussing the properties of the hfb wave functions
, we analyze the structure of the spherical hfb hamiltonian of eq .
( [ eq152 ] ) .
figure [ fig20 ] shows the behavior of @xmath203 and @xmath204 , and @xmath177 and @xmath178 ( central parts only ) obtained for neutrons in @xmath151sn in the hfb+skp model .
the p - h functions , @xmath203 and @xmath177 , are similar to those obtained in other mean - field theories .
@xmath203 has values close to @xmath205@xmath2920mev@xmath206 , which corresponds to the value of the free nucleon mass @xmath207 . in the nuclear interior ,
this function has slightly smaller values , because the effective mass @xmath201 is here slightly larger than @xmath207 .
this effect is due to the non - zero isovector effective mass of the skyrme skp interaction ; recall that for this interaction the nuclear - matter value of the isoscalar effective mass is @xmath201=@xmath207 .
the central potential @xmath177 has the standard depth of about 40mev and disappears around @xmath146=7.5fm .
the form of the p - p functions , @xmath204 and @xmath178 , characterizes the pairing properties of the system .
one may note that both these functions are essentially peaked at the nuclear surface .
in @xmath151sn they also exhibit central bumps resulting from the fact that in this nucleus the neutron 3@xmath152 orbital is located near the fermi surface .
values of @xmath208 are ( in the chosen units ) an order of magnitude smaller than those of @xmath178 .
this should be compared with the results obtained for the p - h channel , where the values of @xmath203 are only about a factor of 2 smaller than those of @xmath177 .
it means that , for the skp parametrization , the kinetic term in the p - p channel ( which simulates the finite - range effects ) is relatively less important than the kinetic energy term in the p - h channel .
this section contains the discussion of hfb wave functions @xmath209 and @xmath210 ( sec .
[ sec2bca ] ) , canonical - basis wave functions @xmath211 ( sec .
[ sec2bcb ] ) , and hf+bcs wave functions ( sec .
[ sec2bcc ] ) . in the following , the hfb equation ( [ eq152 ] )
was solved in the spherical box of the radius @xmath212=20fm for the @xmath213=1/2 and @xmath1=0 ( @xmath152 ) neutron states ; i.e. , for vanishing centrifugal , coulomb , and spin - orbit potentials .
the calculations were performed for @xmath151sn .
the neutron single - quasiparticle wave functions are presented in fig .
[ fig21 ] .
the upper components @xmath214 , and the lower components @xmath215 , are plotted in the left and right columns , respectively . because a box of a finite radius was used , the particle continuum is discretized .
the positive quasiparticle eigenenergies @xmath191 are in increasing order numbered by the index @xmath216 , and their values are tabulated in the left portion of table [ tab01 ] , together with the norms of the lower components ( [ norms ] ) , @xmath217=@xmath218 . since the lower components define the particle density matrix [ eq .
( [ eq144a ] ) ] the numbers @xmath219 + 1)@xmath217 ( i.e. , 2@xmath217 for the @xmath213=1/2 case considered ) constitute contributions of a given quasiparticle state to the total number of neutrons ( see eq .
( [ ntot ] ) ) .
wave functions in fig .
[ fig21 ] , and the entries in table [ tab01 ] , have been ordered from the bottom to the top not according to the excitation - energy index @xmath216 , but rather according to numbers of nodes of the _ large _ component .
( the large component is the lower component for hole states and the upper component for particle states see fig .
[ fig21 ] . )
the lower component of the @xmath216=8 state is large , and it has zero nodes ; therefore it is plotted at the bottom of the figure .
next comes the @xmath216=5 state , whose lower component has one node , and the @xmath216=1 state with two nodes .
lower components of these three states are larger than their upper components and they contribute almost 2 particles each to the total number of neutrons .
consequently , these quasiparticle states should be associated with the 1@xmath152 , 2@xmath152 , and 3@xmath152 , single - hole states . for all other calculated @xmath152 states the upper components are larger than the lower ones , and these states contribute small fractions to the particle number , see table [ tab01 ] .
consequently , these quasiparticle states should be associated with the @xmath152 single - particle states .
the behavior of these wave functions differs in the nuclear interior ( i.e. , for @xmath220 where @xmath221@xmath1537 fm is the nuclear radius ) and outside ( @xmath222 ) .
since the wavelength of the upper component is roughly proportional to @xmath223 , the ratio of the corresponding wavelengths behaves as @xmath224 where @xmath225 is the depth of the neutron potential well . for the @xmath152 neutron states in @xmath151sn the excitation energy , @xmath226 ,
can be found from table [ tab01 ] ( @[email protected] ) , and @xmath225@xmath153@xmath16945mev ( see fig . [ fig20 ] ) .
the upper component of the @xmath216=2 state has three nodes .
however , for @xmath222 the exterior part of the wave function corresponds to a half - wave ; i.e. , it represents the lowest - energy discretized continuum state . since @xmath226 is only 0.97mev , the wavelength in the nuclear interior is @xmath1536.5 times shorter than @xmath227 .
the next two wave functions have four and five nodes in their upper components .
compared to the @xmath216=2 state , they exhibit shorter wavelengths both outside and inside the nucleus ( the corresponding excitation energies are larger ) , and the ratio @xmath228 decreases according to eq .
( [ wave ] ) . the quasiparticle states with @xmath216=2 , 3 , and 4 should be associated with the 4@xmath152 , 5@xmath152 , and 6@xmath152 states in the particle continuum .
of course , the values of their quasiparticle energies strongly depend on the size of the box , because the wavelength of their exterior parts will increase with increasing @xmath212 ( is roughly proportional to @xmath212 ) . from the above discussion
, one can see that the structure of large components resembles very much that of the hf wave functions .
moreover , the small components are very small compared to the large ones ; in order to plot both of them in the same scale ( fig . [ fig21 ] ) they have to be multiplied by factors from 10 to 25 .
only the lowest quasiparticle state ( @xmath216=1 ) , which corresponds to the 3@xmath152 state near the fermi surface , has the two components of a similar magnitude .
it is to be noted , however , that the detailed structure of small components is decisive for a description of the pairing correlations .
indeed , both components are coupled in the hfb equations by the pairing fields @xmath229 or @xmath199 . in agreement with general asymptotic properties of the upper and lower components @xcite
, one sees in fig .
[ fig21 ] that the lower components vanish at large distances for all quasiparticle states , regardless of the excitation energy .
consequently , the resulting density matrix is localized .
it is interesting to observe ( table [ tab01 ] ) that the norms of the lower components @xmath217 do not behave monotonically with quasiparticle energy .
namely , @xmath217 is about 0.0002 for @xmath216=2 ; then it increases to 0.0019 at @xmath216=6 , and only then it decreases to about 0.0001 at @xmath216=11 .
this means that the pairing correlations couple states with very high quasiparticle excitations and short - wavelength upper components ; i.e. , located high up in the particle continuum . in the considered example , only by going to the energy region of as high as 50mev is the pairing coupling to the continuum states exhausted .
apart from the @xmath216=1 state which has the quasiparticle energy @xmath185 smaller than @xmath188 , for all other quasiparticle states the upper components oscillate at large distances ; i.e. , these states belong to the hfb continuum .
this seems natural for the 4@xmath152 , 5@xmath152 , and 6@xmath152 states discussed above , but it also holds for the deep - hole states 1@xmath152 and 2@xmath152 .
this illustrates the physical property of the deep - hole states that once such a state is excited , it is coupled to the particle continuum and acquires some particle width . of course , before such a hole is created ( e.g. , one - quasiparticle excitation in the neighboring nucleus ) the nucleus ( i.e. , quasiparticle vacuum )
is perfectly particle - bound and the contributions from the deep - hole - like quasiparticle states to the density matrix are localized in space . by solving the integral eigenequation for the density matrix ( [ eq122a ] ) , one obtains the canonical - basis wave functions @xmath211 .
actually , when the hfb equation ( [ eq152 ] ) is solved by a discretization method on a spatial mesh , as is done here , the density matrix is represented by a matrix and the integral eigenequation becomes the usual matrix eigenproblem . in the present application to @xmath151sn , the mesh of equally spaced points with @xmath230=0.25fm was used and then the canonical - basis wave functions were obtained on the same mesh of points .
these wave functions are plotted in fig .
[ fig22 ] , while other characteristics of the canonical states are listed on the right - hand side portion of table [ tab01 ] . here
the states are ordered from bottom to top according to their occupation probabilities @xmath231 .
when @xmath47 increases from 1 to 5 , the number of nodes of the canonical - basis wave functions increases from zero to four .
therefore , these states represent the 1@xmath152 to 5@xmath152 single - particle states .
the first three of them have large occupation probabilities @xmath231 , negative average values @xmath232 of the p - h hamiltonian , and positive pairing gaps @xmath233 [ see eq .
( [ eq148 ] ) ] .
these states have all the characteristics of bound single - particle states , and their wave functions strongly resemble the large components of the @xmath216=8 , 5 , and 1 quasiparticle states shown in fig . [ fig21 ] .
it is interesting to note that the two states @xmath47=4 and 5 follow exactly the same pattern of localized wave functions , despite the _ positive _ values of @xmath232 .
therefore , these two states can be understood as the representatives of the positive - energy spectrum in the ground - state of @xmath151sn .
we purposely avoid using the term `` particle continuum '' , because these orbitals represent discrete and localized eigenstates of the density matrix .
table [ tab01 ] shows that the occupation probabilities of the canonical - basis states with @xmath47=4,@xmath234,7 decrease very rapidly .
in fact only states with @xmath47=4 and 5 have tangible occupation probabilities ; one can say that the remaining orbitals are entirely empty .
this feature has to be compared with the sequence of norms of the lower hfb components , @xmath217 , which do not fall down to zero at even a nearly similar pace .
this demonstrates that even if the convergence of the hfb eigenproblem requires high quasiparticle energies , the number of physically important single - particle states is very restrained . unfortunately , as discussed below in sec .
[ sec2bb ] , one can not obtain the canonical - basis states without actually solving the hfb equations up to high energies .
for @xmath47=6 and higher , the occupation probabilities are so small that the numerical procedure used to diagonalize the density matrix returns accidental mixtures of almost degenerate eigenfunctions .
this is seen in fig .
[ fig22 ] , where the wave function with @xmath47=6 has six nodes instead of five , expected from the regular sequence .
also the energies @xmath232 are for these nearly empty states randomly scattered between 40 and 70mev , while the pairing gaps @xmath233 are scattered around zero .
the bcs quasiparticle wave functions can be obtained by enforcing the bcs approximation on the hfb equations .
this is done by setting the pairing hamiltonian @xmath235 to a constant ; i.e. , by using @xmath204=0 and @[email protected] .
this value of @xmath236 is equal to minus the hfb average neutron pairing gap , as defined in eq .
( [ eq157 ] ) . as seen in fig .
[ fig23 ] , the pattern of large components follows closely that obtained in the hfb method , while the shapes of small components are entirely different . indeed , since in the bcs approximation lower and upper components are simply proportional , small and large components have the same asymptotic properties .
this leads to serious inconsistencies , because the small lower components _ are not _ localized any more , and introduce an unphysical particle gas in the density matrix , while the small upper components _ are _ localized and the corresponding deep - hole states have no particle width . it is seen in eqs .
( [ eq136 ] ) and ( [ eq141 ] ) that the two - body interaction enters the p - h and p - p channels in a different way .
this is particularly conspicuous when the canonical basis ( [ eq122 ] ) is used ; i.e. , @xmath237 where [ eq138]@xmath238 the two - body matrix elements in the canonical basis are defined as usual : @xmath239 since we include in @xmath240 the exchange term , the matrix @xmath241 is antisymmetric in @xmath242 and in @xmath243 . due to the hermiticity and the time - reversal symmetry of the interaction , matrices @xmath244 and
@xmath245 obey the following symmetry relations , @xmath246 the matrix @xmath247 is defined by different matrix elements of the interaction than the matrix @xmath248 .
namely , the matrix element @xmath244 represents a `` diagonal '' scattering of pairs of states @xmath242@xmath249@xmath242 ( or @xmath250@xmath249@xmath250 ) .
this type of scattering concerns _ all _ pairs of states .
the resulting contributions to the energy , eq . ( [ eq137 ] ) , involve the occupation probabilities of the single - particle states constituting each pair . on the other hand
, the matrix elements @xmath245 represent a `` non - diagonal '' scattering of pairs of _ time - reversed _
states @xmath251@xmath249@xmath252 .
this scattering concerns only a very special subset of all pairs . in principle
, an effective interaction should describe both channels of interaction at the same time .
this is , for example , the case for the gogny interaction @xcite and for the skyrme skp interaction @xcite .
however , the fact that both channels of interaction play a different role in the hfb theory allows the use of different forms of interaction to model the p - h and p - p channels .
such an approach is additionally motivated by the fact that the interaction in the p - h channel , which defines , e.g. , the saturation properties , is much better known than the p - p interaction . moreover ,
the p - h channel provides a two - orders - of - magnitude larger interaction energy . since the canonical - basis wave functions @xmath253 are all localized , it is instructive to consider the hfb equations in this particular basis .
they read : [ eq145]@xmath254 where @xmath255 equation ( [ eq145a ] ) is equivalent to the variational condition that the hfb energy is minimized , while eq .
( [ eq145b ] ) defines the energy matrix @xmath256 .
( the matrix @xmath256 represents the hfb hamiltonian in the canonical basis . ) since for every pair of indices @xmath242 it holds that @xmath257+@xmath258=1 , eqs .
( [ eq145 ] ) can be written as @xmath259 the occupation probabilities @xmath129 are solely determined by the diagonal matrix elements of the p - h and p - p hamiltonians , [ eq148]@xmath260 and the result is [ eq149]@xmath261 where @xmath262 are the diagonal matrix elements of the matrix @xmath256 : @xmath263 in this representation , the average pairing gap ( [ eq157 ] ) is given by the average value of @xmath233 in the occupied states , @xmath264 equations ( [ eq149 ] ) and ( [ eq150 ] ) misleadingly resemble those of the simple bcs theory @xcite .
however , in the hfb theory , @xmath232 is not the single - particle energy ( i.e. , the eigenvalue of @xmath265 ) but the diagonal matrix element of @xmath265 in the canonical basis .
similarly , @xmath233 does not represent the pairing gap in the state @xmath266 , and @xmath262 is not the quasiparticle energy @xmath185 .
however , since these quantities define the occupation probabilities , they play a very important role in an interpretation of the hfb results , and many intuitive , quantitative , and useful features of the bcs theory can be reinterpreted in terms of the canonical representation ( cf .
[ sec3f ] ) . in particular , the average values of single - particle p - h and p - p hamiltonians fulfill the following self - consistency equations : [ eq151]@xmath267 for a given interaction @xmath244 and @xmath245 , eqs .
( [ eq151 ] ) represent a set of nonlinear equations which determine @xmath232 and @xmath233 .
equations for @xmath232 ( [ eq151a ] ) and for @xmath233 ( [ eq151b ] ) are coupled by the values of @xmath268 ( [ eq150 ] ) , which depend on both @xmath232 and @xmath233 .
however , it is clear that the interaction in the p - h channel mainly influences the values of @xmath232 , while that in the p - p channel @xmath233 .
unfortunately , eqs .
( [ eq151 ] ) can not replace the original hfb equations , because they require the knowledge of the canonical basis to determine the @xmath244 and @xmath245 matrices .
the only way to determine the canonical basis is to solve the original hfb equation ( [ eq143 ] ) , and then to diagonalize the density matrix ( [ eq144a ] ) .
moreover , solving eqs .
( [ eq151 ] ) ensures that only the @xmath47=@xmath269 subset of variational equations ( [ eq145a ] ) is met , the minimum of energy being obtained by solving the whole set ( i.e. , for all indices @xmath47 and @xmath269 ) .
the diagonalization of the energy matrix @xmath256 gives the spectrum of hfb eigenenergies , @xmath191 : @xmath270 the matrix @xmath271 represents the unitary transformation from the canonical to the quasiparticle basis @xcite .
its matrix elements provide the link between the quasiparticle energies @xmath191 and the diagonal matrix elements @xmath262 which define the occupation probabilities , i.e. , @xmath272
for weakly bound nuclei one may expect that the particle continuum influences the ground - state properties in a significant way . as discussed in sec .
[ sec2bcb ] , the phase space corresponding to positive single - particle energies should not be confused with the continuum of scattering states which asymptotically behave as plane waves , and are significant for genuine scattering phenomena .
properties of the continuum scattering states are intuitively well understood in terms of unpaired single - particle orbits . shown in fig .
[ fig12 ] are the self - consistent hf+skp neutron single - particle energies in @xmath149sn , @xmath273 , as functions of the radius @xmath212 of the spherical box in which the hf equations are solved .
it is assumed that the following boundary condition holds for all single - particle wave functions : @xmath274 for bound single - particle states , @xmath273@xmath1890 , the effect of increasing @xmath212 beyond 10fm is insignificant . as seen in fig .
[ fig12 ] , the energies of the least bound 3@xmath275 , 2@xmath276 , 1@xmath277 , and 1@xmath278 states , which form the 82@xmath279@xmath2@xmath279126 shell , are independent of @xmath212 .
the boundary condition ( [ eq153 ] ) leads to a discretization of the continuum by selecting only those states which have a node at @xmath146=@xmath212 .
when @xmath212 increases , the density of the low - energy continuum states increases as @xmath280 .
this effect is very well visible in fig .
[ fig12 ] . among those states
whose energies decrease with @xmath212 , one may easily distinguish some quasi - bound states , which have energies fairly independent of @xmath212 . in fig .
[ fig12 ] these are the high-@xmath1 states @xmath281 , @xmath282 , @xmath283 , and @xmath284 . however , at some values of @xmath212 they are crossed by , and they interact with , the real continuum states ( plane waves ) of the same quantum numbers , and their precise determination is , in practice , very difficult . a solution of the hfb equation ( [ eq152 ] ) in the spherical box amounts to using the analogous boundary conditions , @xmath285 for both components of the hfb wave function . as a result ,
the quasiparticle continuum of states with @xmath186@xmath187@xmath188 is discretized and becomes more and more dense with increasing @xmath212 .
however , as discussed in sec .
[ sec2bc ] , the density matrix depends only on the localized ( lower ) components of the quasiparticle wave functions and , therefore , is very well stable with increasing @xmath212 . by the same token , the properties of the canonical - basis states , which are the eigenstates of the density matrix , are also asymptotically stable
of course , the bigger the value of @xmath212 , the larger is the numerical effort required to solve the hfb equations .
consequently , it is important to optimize the value of @xmath212 ; i.e. , to use the smallest box sizes which reproduce all interesting physical properties of the system .
apart from ours , there are also other possible approaches to solving the hfb eigenproblem ; in particular : ( i ) the diagonalization in the large harmonic oscillator basis , and ( ii ) the two - step diagonalization . scheme ( i )
has been used , e.g. , in the hfb+gogny calculations or in the deformed hfb+skp calculations of ref .
its limitations , due to the incorrect asymptotics , are discussed in sec .
[ sec3e ] below . in method ( ii )
one first solves the hf problem and then diagonalizes the full hfb hamiltonian in the hf basis .
such a strategy has been suggested in ref .
@xcite and recently adopted in ref .
@xcite . as discussed in sec .
[ sec2bb ] , quantities which determine the p - h properties of the system are the canonical energies
@xmath232 [ eq . ( [ eq148a ] ) ] .
the neutron canonical energies in @xmath149sn are shown in fig .
[ fig10 ] as functions of the box size @xmath212 . in this figure ,
the single - particle index @xmath47 is represented by the spherical quantum numbers @xmath286 ; only the states with occupation probabilities @[email protected] are presented .
the canonical states belonging to the shell 82@xmath279@xmath2@xmath279126 have negative @xmath288 s , and they are very close to the hf single - particle energies displayed in fig . [ fig12 ] .
they do not depend on the values of @xmath212 for @xmath212@xmath18710fm . at positive values of @xmath289 , there are several orbitals which do not depend on the box size even at @xmath212@xmath18915fm .
these states correspond to the high-@xmath1 quasibound states @xmath281 , @xmath282 , @xmath283 , and @xmath284 , already identified in the hf spectrum of fig .
[ fig12 ] .
the values of @xmath288 for these states are only slightly higher than the corresponding values of @xmath290 .
however , these quasibound canonical - basis states are not accompanied by the sea of plane - wave scattering states ( cf .
the @xmath282 , and the @xmath284 states in figs .
[ fig12 ] and [ fig10 ] ) .
one can thus say that the canonical - basis states represent the quasibound states well decoupled from the scattering continuum .
many other canonical - basis states , especially those with low orbital angular momenta @xmath1 , significantly depend on the box size up to about @xmath212=18fm , and then stabilize . therefore , in all subsequent calculations we use a safe " value of @xmath212=20fm , unless stated otherwise . above 20mev
there appear states with canonical energies fluctuating with @xmath212 .
these states have very small occupation probabilities close to the limiting value of @xmath287=0.0001 , and their determination as eigenstates of the density matrix is prone to large numerical uncertainties ( see sec . [ sec2bcb ] ) .
one should note that the physical observables are calculated directly by using the hfb density matrices , and the above numerical uncertainties do not affect the results obtained within the hfb theory . as pointed out in ref .
@xcite , the canonical spectrum presented in fig .
[ fig10 ] can be used to analyze the shell effects far from stability .
in particular , the size of the @xmath2=126 gap is very small ( a 2mev gap between the 1@xmath278 and 4@xmath152 states ) , and hence it can not yield any pronounced shell effect ( seen , e.g. , in the behavior of the two - neutron separation energies , sec .
[ sec4c ] ) .
this shell - gap quenching is not a result of a too small value of the spin - orbit splitting .
indeed , a larger spin - orbit strength would push the 1@xmath278 level down in energy , without affecting the size of the @xmath2=126 shell gap ( several negative - parity states are nearby ) .
the @xmath2=126 gap , which is equal to about 4mev at @xmath212=10fm , closes up with increasing @xmath212 due to the several low-@xmath1 states whose energies steadily decrease .
this effect can be attributed to the pairing - induced coupling with the positive - energy states ( see sec . [ sec3e ] ) . in the energy window between 0 and 20mev ,
the density of single - particle canonical energies is fairly uniform and no pronounced shell effects are visible . since the fermi energy must stay at negative values , this region of the phase space can not be reached .
however , one may say that the influence of the positive - energy spectrum on the bound states ( had we analyzed it in terms of , e.g. , the strutinsky averaging ) is characterized by a rather structureless distribution of states . above 20mev ,
the occupation probabilities rapidly decrease ( cf .
table [ tab01 ] ) , and this part of the phase space can safely be disregarded , provided one stays in the canonical basis .
the eigenvalues of the hfb equation ( [ eq152 ] ) ( single - quasiparticle energies ) carry information on the elementary modes of the system .
the lowest single - quasineutron energies @xmath291 in tin isotopes between @xmath2=50 and @xmath2=126 are shown in fig .
[ fig14 ] ( top panel ) .
apart from the magic shell gaps at @xmath2=50 and @xmath2=82 , where the single - quasiparticle energies exhibit sudden jumps , they depend rather smoothly on neutron number . for a given orbital @xmath286 , the minimum of @xmath291 is attained in the isotope where the corresponding single - particle state is closest to the fermi energy .
hence , from fig . [ fig14 ] one can infer the order of single - particle energies in the beginning of the 50@xmath279@xmath2@xmath27982 shell as @xmath292 , @xmath293 , @xmath294 , @xmath295 , and @xmath296 .
similarly , the predicted order at the bottom of the next major shell is @xmath297 , @xmath298 , @xmath299 , @xmath300 , @xmath301 , and @xmath302 .
the order of spherical single - particle states does vary with @xmath2 .
for instance , according to the hfb+skp calculations of fig .
[ fig14 ] , the @xmath295 shell never becomes lowest in energy , as it should have done , had the single - particle energies been @xmath2-independent .
noteworthy is the fact that , due to the strong interaction with the low-@xmath1 continuum ( cf . sec . [ sec3d ] )
, the @xmath303 excitation becomes lowest at @xmath2@xmath187114 . above the @xmath303 state
there appear several quasiparticle states with excitation energies rapidly decreasing with @xmath2 .
these orbitals represent the low - energy continuum states .
they are very close in energy , exhibit small spin - orbit splitting , and the lowest of them are the low-@xmath1 states : @xmath304 , @xmath305 , @xmath306 , and @xmath307 .
all these features are characteristic of the continuum states @xcite .
still higher in energy , one may distinguish a similar doublet of the @xmath308 and @xmath309 states , as well as the @xmath310 state which represents a high-@xmath1 resonance .
the bottom panel of fig . [ fig14 ] shows similar results for the bcs - like canonical energies @xmath262 defined in eq .
( [ eq150 ] ) , and denoted here by @xmath311 .
a comparison between @xmath291 and @xmath311 illustrates the fact that the _ lowest _ elementary excitations of the nucleus are equally well described by both these quantities .
indeed , a general pattern and , in most cases , also the values of @xmath291 and @xmath311 are very similar .
the differences mainly concern the @xmath152 states , and also the low-@xmath1 states in the continuum , which in the canonical representation appear higher in energy ( see table [ tab01 ] for the direct comparison for @xmath152 states ) . on the other hand ,
the position of the high-@xmath1 @xmath310 resonance is almost identical in both representations .
such a similarity supports the supposition ( ref .
@xcite and sec .
[ sec3d ] ) that the canonical single - particle energies , which are the main ingredients of @xmath311 , constitute a fair representation of single - particle and single - quasiparticle properties of the system . as discussed in sec .
[ sec2ab ] , the canonical states constitute a basis in which the independent - quasiparticle state @xmath34 has the form of a product of correlated pairs [ eq .
( [ eq116 ] ) ] .
therefore , these states can be considered as fundamental building blocks describing the pairing correlations in a many - fermion system . on the other hand ,
the canonical states are determined by a solution of the hfb equation the single - quasiparticle states .
since the canonical states constitute an orthonormal ensemble , the lower and upper hfb components can be expanded as [ eqqcan]@xmath312 where @xmath313 are the associated overlaps . in order to find the relation between @xmath314 and
@xmath315 one can employ eqs .
( [ eq122 ] ) and ( [ eq144 ] ) for the hfb densities .
this gives the canonical wave functions expressed as linear combinations of the _ lower _ hfb components : [ eqqcan1]@xmath316 one should note that the expansions ( [ eqqcan1 ] ) are valid regardless of the fact that the lower components @xmath317 _ do not constitute _ an orthogonal ensemble of wave functions . by multiplying both sides of eqs .
( [ eqqcan1a ] ) and ( [ eqqcan1b ] ) with @xmath318 and taking the scalar product , one arrives at the orthogonality relations : [ eqqcan2]@xmath319 the above identities express the fact , that both @xmath315 and @xmath314 are related to the transformation matrix @xmath320 defined in eq .
( [ ediag ] ) : @xmath321 and eqs .
( [ eqqcan2 ] ) reflect the unitarity of @xmath320
. equations ( [ transl ] ) can be easily derived by inserting expansions ( [ eqqcan ] ) into the hfb equation ( [ eq143 ] ) , and then expressing the matrix @xmath256 ( [ eq145b ] ) in its eigensystem ( [ ediag ] ) .
it is instructive to express the upper hfb component in a form similar to that of eq .
( [ eqqcana ] ) : @xmath322 for @xmath323 , the upper component @xmath324 is the scattering wave function .
it can be formally expanded in the _ localized _ canonical wave functions according to eq .
( [ phib ] ) , but the main contribution comes from the particle - like states with very small values of @xmath325 .
hence , this relation is not too useful in practical applications . in order to discuss the importance of the particle continuum on the structure of canonical states ,
it is interesting to see how a given canonical state is distributed among the single - quasiparticle states . for this , it is convenient to rewrite eq .
( [ eqqcan1a ] ) in the following way : @xmath326 the spectral amplitudes @xmath327 define the distribution of the canonical states among the single - quasiparticle states .
it is important to recall at this point that the sum in eq .
( [ spectral2 ] ) represents in fact the discrete ( @xmath191@xmath189@xmath188 ) states and the discretized ( @xmath191@xmath187@xmath188 ) continuum states , i.e. , @xmath328 with the spectral amplitudes @xmath327 and
@xmath329 pertaining to the discrete and continuous hfb spectrum , respectively . the spectral amplitudes can be expressed in terms of matrices @xmath315 or @xmath271 introduced in sec .
[ canqua ] : @xmath330 we have included in @xmath327 the norms @xmath217 of the lower components , eq .
( [ norms ] ) . in this way ,
the values of spectral amplitudes measure the influence of quasiparticle states irrespective of the overall magnitude of their lower components . before discussing the properties of the spectral amplitudes ,
let us write down the two sum rules : @xmath331 @xmath332 the first two sum rules , eq . ( [ eqqcan4a ] ) , come from the the unitarity of @xmath271 . the last one , eq .
( [ eqqcan4b ] ) , expresses the condition defining the norm of the lower hfb component . in fig .
[ fig33 ] are shown the spectral amplitudes for the @xmath152 canonical states in @xmath151sn ( cf .
[ sec2bca ] and [ sec2bcb ] ) .
the phases of the single - quasiparticle wave functions have been fixed in such a way that all the amplitudes @xmath327 for @xmath47=1 are positive ( some of these amplitudes are too small to be displayed in the figure ) .
this defines the relative phases of the spectral amplitudes for @xmath47@xmath1871 .
then , the positive and negative amplitudes are in fig . [ fig33 ] shown by bars hashed in opposite directions .
results shown in this figure pertain to the same single - quasiparticle and canonical states as those shown in figs .
[ fig21 ] and [ fig22 ] , respectively , and in table [ tab01 ] .
the lowest panel in fig . [ fig33 ] shows that the @xmath333 canonical state ( @xmath47=1 ) is composed mainly of two components corresponding to the two deep - hole quasiparticles at @xmath334=31.64mev and @xmath335=17.60mev .
similarly , the @xmath47=2 and @xmath47=3 canonical states are mixtures of the @xmath335=17.60mev and @xmath336=1.54mev quasiparticles .
for all three of these canonical states , the diagonal amplitudes dominate .
another pattern appears for the positive - energy canonical states ; i.e. , for @xmath47=4 and @xmath47=5 .
these two canonical states contain large components of the hole - like quasiparticles at @xmath335=17.60mev and @xmath336=1.54mev , but in addition , they also acquire large components of the particle - type quasiparticles belonging to the continuum .
these continuum components are centered around 15 and 20mev for @xmath47=4 and @xmath47=5 , respectively .
this illustrates the fact that a correct description of the positive - energy canonical states requires solving the hfb equation to rather high energies .
the widths of the corresponding distributions are rather large , which indicates that there is not a single resonance in the particle continuum which would alone describe the high - energy @xmath152 canonical states .
this can be well understood by recalling that the @xmath1=0 resonances have usually very large widths . for the drip - line nucleus @xmath149sn ,
the spectral @xmath152 amplitudes are shown in fig .
[ fig35 ] . similarly to the case of @xmath151sn , the three lowest canonical states for @xmath47=1 , 2 , and 3 are mainly composed of the three hole - like quasiparticles at @xmath337=34.27 , @xmath338=22.12 , and @xmath339=7.24mev with dominating diagonal amplitudes . on the other hand
, the low - lying positive - energy canonical @xmath47=4 state has large and almost equal components coming from the particle - like quasiparticles at @xmath336=2.40 , @xmath340=4.84 , and @xmath341=8.93mev .
the following @xmath47=5 canonical state has dominant amplitudes from the hole - like and particle - like quasiparticles at @xmath339=7.24 and 8.93mev , respectively .
one should note that the @xmath47=4 and @xmath47=5 canonical @xmath152 states in @xmath149sn have rather large occupation factors as compared to those in @xmath151sn . both of them require including the single - quasiparticle states _ at least _ up to 10mev .
the following @xmath47=6 state ( not shown in the figure ) has the occupation probability of @xmath342=0.0003 and the spectral amplitudes extending up to 25mev .
the spectral amplitudes for the @xmath343 states in @xmath151sn and @xmath149sn are shown in figs .
[ fig34 ] and [ fig36 ] , respectively .
an interesting situation appears in @xmath151sn where two quasiparticles , one of the particle type and another one of the hole type , have rather similar single - quasiparticle energies of 17.63 and 18.97mev . as a result ,
the lowest canonical state ( @xmath47=1 ) acquires a substantial particle - type quasiparticle component , while both quasiparticles contribute almost equally to the @xmath47=3 canonical state . in @xmath149sn , the positive - energy @xmath343 canonical states
( @xmath47=3 and 4 ) have large amplitudes from the hole - like quasiparticles ( contributing almost exclusively to the structure of the negative - energy canonical states with @xmath47=1 ( 1@xmath343 ) and @xmath47=2 ( 2@xmath343 ) ) , as well as from a wide distribution of several particle - type quasiparticles extending up to 20mev .
the spectral amplitudes allow also for a determination of the asymptotic properties of canonical states .
( see ref .
@xcite for a discussion of the the asymptotic properties of natural orbits . )
the lower components @xmath317 behave asymptotically as @xmath344@xmath169@xmath345 @xcite . therefore , as seen from eq .
( [ spectral ] ) , the asymptotic properties of canonical states are governed by the lowest discrete quasiparticle , provided the corresponding spectral amplitude , @xmath346 , is not equal to zero . however ,
if such a spectral amplitude is non - zero but very small , the corresponding asymptotic behavior will be attained only at very large distances . in practice , the lowest discrete quasiparticle dominates the asymptotic behavior only if the corresponding spectral amplitude has a significantly large value . for the @xmath152 states in @xmath151sn ( fig .
[ fig33 ] ) such a situation occurs for the canonical states with @xmath47=25 on the other hand , since the value of @xmath347 is very small , the asymptotic behavior of the @xmath47=1 canonical state is dominated by the hole - like quasiparticle at @xmath335=17.60mev .
a similar situation occurs for the @xmath343 states in @xmath151sn .
namely , only for the @xmath47=2 canonical state the asymptotic behavior is determined by the lowest discrete quasiparticle .
an entirely different property can occur in drip - line nuclei , where the fermi energy is close to zero and there may be no quasiparticle excitations in the discrete spectrum between 0 and @xmath188 .
in such a situation , shown in figs . [ fig35 ] and [ fig36 ] , the canonical states are represented by superpositions of lower quasiparticle components belonging to the particle continuum .
consequently , it is the integral over the lowest continuum quasiparticle states just above the @xmath348 threshold that determines the asymptotic properties of the canonical states . in other words ,
the profile of the level density , @xmath349 , around @xmath350 becomes a crucial factor .
good examples of a very strong coupling to the particle continuum are the @xmath47=4 and 5 canonical @xmath152 and @xmath343 states in @xmath149sn , where the quasiparticle strength is distributed in a very wide energy interval ranging from 1.5 to 20mev . on the other hand , the two lowest canonical @xmath343 states in @xmath149sn can be associated with the two quasiparticle excitations well localized in energy ( see fig .
[ fig36 ] ) and their asymptotics is governed by the energy of the lowest quasiparticle .
an analysis of the spectral distribution , analogous the one presented above , has recently been performed @xcite for the natural orbits in @xmath351o determined within the green s function method using the @xmath13 interaction .
this method accounts for a much more general class of correlations as compared to the hfb correlations of the pairing type studied here .
however , the general features of the spectral distributions remain essentially the same .
namely , the low - occupation - number natural orbits are determined mostly through high - energy continuum contributions , and large box sizes ( 1520fm ) and large single - particle bases ( 20 states per @xmath352-block ) have to be used to stabilize the solutions .
this is so even if the studied nucleus ( @xmath351o ) is @xmath0-stable , well - bound , and light ; one can expect that for drip - line nuclei the aforementioned features can only be more pronounced . in the limit of weak binding , radial dimensions of atomic nuclei increase and
it becomes exceedingly important to control the radial asymptotics of many - body wave functions , not only in reaction studies but also in nuclear structure applications .
figure [ fig04 ] displays the radial dependence of the neutron density @xmath144 in @xmath149sn calculated with the values of @xmath212 between 10 and 30fm .
it is seen that , for every value of @xmath212 , @xmath144 follows its asymptotic behavior up to about @xmath212@xmath1693fm and then falls down to zero as a result of the boundary conditions ( [ eq154 ] ) .
that is , these boundary conditions affect the density only in a narrow spherical layer of the thickness equal to about 3fm , while inside this layer @xmath144 behaves independently of the value of @xmath212 .
analogous results for the pairing density @xmath145 are shown in fig .
[ fig04a ] . at very large distances
the asymptotic behavior of the particle density is governed by the square of the lower component of the single - quasiparticle wave function corresponding to the lowest quasiparticle energy @xmath353 .
similarly , the asymptotic behavior of the pairing density @xmath145 is determined by the product of the upper and the lower components of quasiparticle @xmath353 . using the asymptotic properties of the hfb wave functions derived in @xcite , one obtains : [ eq155]@xmath354 where @xmath355 in the considered example of @xmath149sn the calculated values are @[email protected] and @xmath353= 1.07mev ( a @xmath356 state ) .
consequently , @[email protected]@xmath17 and @[email protected]@xmath17 . in figs .
[ fig04 ] and [ fig04a ] the asymptotic dependencies given by eq .
( [ eq155 ] ) are shown as shaded lines .
one can see that for @xmath144 the asymptotic regime is reached only at distances as large as 25fm , which means that the contributions from other quasiparticle states , and/or from the next - to - leading - order terms in the hankel functions , still influence the particle density at rather large values of @xmath146 .
interestingly , the pairing density approaches the asymptotic limit already at @xmath146@xmath15310fm .
a rough estimate of @xmath357 and @xmath358 can be obtained by substituting the value of a typical pairing gap ( @xmath8=1mev ) for the lowest quasiparticle energy @xmath353 . for stable nuclei ( @xmath7@xmath29@xmath1698mev ) one
obtains @[email protected]@xmath17 , while for the one - neutron drip nuclei , defined by a vanishing separation energy , @xmath359@xmath29@xmath8+@xmath7@xmath290 , the result is @[email protected]@xmath17 .
this difference illustrates the increase in the spatial extension of the _ particle _ densities when going towards the neutron drip line . on the other hand , for the _ pairing _ densities
the corresponding numbers are @[email protected]@xmath17 and @xmath358=@[email protected]@xmath17 .
therefore , in stable nuclei both types of densities have rather similar asymptotic behavior , while in drip - line nuclei the pairing densities have much longer tails . in this context , it is instructive to recall the discussion from sec .
[ sec2ac ] regarding the probabilistic interpretation of the hfb densities .
the probability @xmath361 ( @xmath362 ) of finding a particle or a pair of particles at @xmath146=@xmath363 is proportional to @xmath364 or @xmath365 , respectively .
consequently , in stable nuclei @xmath362 decays much faster than @xmath361 at large distances .
this is not true for drip - line nuclei , where the asymptotics of @xmath361 and @xmath362 is the same . as discussed above
, static pairing correlations can influence dramatically the asymptotic behavior of density distributions in drip - line nuclei .
in addition , a significant modification of the density tails comes from the dynamical coupling to collective modes through the particle continuum .
such a coupling can be treated in terms of the continuum qrpa and has been shown to be very important for light systems @xcite .
an analysis of the asymptotic behavior of the particle density @xmath144 has recently been performed @xcite by finding the _ exact _
solutions for weakly bound two particles interacting through a contact force . in that study , the role of one - particle resonant states on the density asymptotics has been discussed . as illustrated in sec .
[ sec3a ] , the density of the scattering continuum states increases with @xmath212 . in the limit of very large values of @xmath212 , the set of discretized continuum states can be considered as a fair approximation of the real continuum , and the sums over the positive - energy states can correctly represent integrals over the continuous energy variable .
therefore , we may consider this limit in order to study the dynamical coupling between the bound single - particle states and the positive - energy states . in the language of pairing correlations
, one may think of this coupling in terms of a virtual scattering of pairs of fermions from the bound states to positive - energy states , and back .
such a pair scattering gives rise to the additional pairing energy to the ground - state energy . to illustrate the stability of results with increasing box size , in fig .
[ fig03 ] we show the neutron p - p potentials @xmath178 in @xmath149sn and @xmath150sn calculated in the hfb+skp model for several values of @xmath212 . in these two nuclei , the values of @xmath178 do not change when @xmath212 is larger than 20 and 22fm , respectively , but at smaller values of @xmath212 , one observes significant variations .
a rather unexpected result of this analysis is that the overall magnitude of pairing correlations , represented by the average pairing gap @xmath366 , _ decreases _ with increasing @xmath212 .
this occurs in spite of the fact that the actual density of scattering states dramatically _ increases _ with increasing @xmath212 .
this effect can be understood by noting that the pairing correlations produced by a density - dependent p - p interaction ( and hence for the skp force used here ) are concentrated at the nuclear surface ; i.e. , at a fixed location in space . for small values of @xmath212 , the boundary conditions ( [ eq154 ] ) have a tendency to push the continuum wave functions towards smaller distances , and into the surface region .
this increases the magnitude of pairing correlations . on the other hand , with increasing @xmath212 ,
the scattering states spread out uniformly outside the nucleus and effectively leave the surface region . hence @xmath366 decreases . as a consequence , with increasing @xmath212 the self - consistent attractive pairing potential @xmath178 decreases in magnitude and significantly spreads out towards large distances .
the importance of allowing the pairing interaction to couple properly to the particle continuum is illustrated in fig .
[ fig05a ] , where the neutron rms radius , the average pairing gap , and the fermi energy are shown as functions of @xmath212 .
the two upper plots confirm that a stability of results is attained beyond 20 or 22fm , while the bottom plot indicates that the pairing coupling to the positive - energy states can be a decisive factor influencing the nuclear binding . indeed , below @xmath212@xmath2920fm the nucleus @xmath150sn is unbound , and it becomes bound only when its ground state is allowed to gain an additional binding from the pairing correlations at large distances .
this indicates that , for the surface - type pairing interaction , one has to consider a rather dense particle continuum before the pairing coupling to positive - energy states is exhausted .
( for a similar discussion in a schematic model see ref .
there , it has been pointed out that because of strong coupling to the continuum , @xmath7 is significantly lowered in the case of surface pairing as compared to the case of volume pairing . ) since , for the gogny interaction , the hfb equations are solved by expansion in the harmonic oscillator basis , one can test the coupling to the positive - energy states by increasing the number @xmath367 of the oscillator shells used in the basis . in practice ,
calculations must be restricted to @xmath367@xmath27920 , which allows one to describe the wave functions up to about @xmath368@xmath29@xmath369 , where @xmath370 is the frequency of the harmonic oscillator @xcite . for @xmath367=20
this corresponds to about @xmath368=14fm .
figure [ fig06 ] compares the asymptotic behavior of the neutron particle densities in three neutron - rich tin isotopes calculated in the spatial coordinates ( skp ) or in the harmonic - oscillator basis ( d1s ) . in the former case one
obtains a clean region of the asymptotic dependence governed by eq .
( [ eq155a ] ) , which around @xmath146=18fm is perturbed by the box boundary conditions ( [ eq154 ] ) at @xmath212=20fm . in the latter case
, the region of proper asymptotic behavior becomes perturbed by the @xmath371 dependence characteristic of the harmonic - oscillator - basis wave functions .
the @xmath370 values , obtained by minimizing the total energy for the @xmath367=17 basis , are equal to 13.4 , 6.6 , and 6.3mev in @xmath372sn , @xmath149sn , and @xmath150sn , respectively . due to this , a study of the continuum influence using such a basis can be performed only up to densities of scattering states corresponding to about @xmath212=14fm in the heavier isotopes and only @xmath212=10fm in @xmath372sn , as can be seen in fig .
[ fig12 ] .
let us note , however , that the neutron densities beyond @xmath146=10fm are typically smaller than 10@xmath373@xmath154 , which explains the stability of the hfb calculations with increasing size of the basis .
this is illustrated in fig .
[ fig05b ] which is analogous to the similar study presented for the skp interaction in fig .
[ fig05a ] . here , for each value of @xmath367 and for each nucleus , the value of @xmath370 was optimized so as to minimize the total energy .
as can be seen , one obtains a nice stability of results by using @xmath367=17 .
this test corresponds to testing the coordinate - representation solutions ( fig .
[ fig05a ] ) in the range of box sizes between 12fm@xmath279@xmath212@xmath27914fm . in this rather narrow region ,
the skp results are not stable because of the dominant surface - type character of its pairing interaction . since the p - p gogny interaction is more of the volume type ( sec .
[ sec2aca ] ) it requires much smaller distances to saturate . when inspecting fig .
[ fig12 ] , it is obvious that by applying the bcs approximation to the state - independent pairing force and by allowing the bcs - type pairing correlations to develop in such a dense spectrum , the result can be disastrous .
the seniority force gives rise to the _ non - localized pairing field _
@xcite , @xmath375 i.e. , to a constant pairing gap , identical for all states .
the high density of single - particle states in the particle continuum immediately results in an unrealistic increase of bcs pairing correlations @xcite .
one may , in principle , artificially readjust the pairing strength constant to avoid such an increase , but then the predictive power of the approach is lost and , moreover , the spatial asymptotic properties of the solutions are still going to be incorrect . to illustrate the latter point , fig .
[ fig25 ] ( top panel ) shows the neutron densities in @xmath149sn calculated for several values of @xmath212 within the hf+bcs approximation . in order to avoid the increase of pairing correlations with increasing density of states ,
the calculations have been performed by fixing the values of the pairing gap . for every box size @xmath212
, the value of @xmath376 has been set equal to the average pairing gap @xmath366 obtained within the hfb method .
the corresponding @xmath366 values are quoted in fig .
[ fig03 ] .
it is not too surprising to see that the asymptotic behavior of the density calculated in the hf+bcs+@xmath366 method ( top panel ) is entirely different than that shown in fig .
[ fig04 ] . due to a nonzero occupation probability of quasibound states
, there appears an unphysical gas of neutrons surrounding the nucleus . in fig .
[ fig25 ] this gas has a constant density of @xmath126@xmath296@xmath37710@xmath378@xmath154 , independent of @xmath212 .
this result means that an external pressure would have been necessary to keep the neutrons inside the box .
namely , had the box boundary condition been released , one would have observed a stream of neutrons escaping the nucleus .
this is a completely artificial ( and unwanted ) feature of the bcs approximation , because for a negative value of the fermi energy , neutrons can not be emitted . in the above example the density of the neutron gas at @xmath212=25fm
corresponds to about 4 neutrons uniformly distributed in the sphere of @xmath221=@xmath212 .
needless to say , by increasing the box radius , the number neutrons in the gas grows at the expense of the number of neutrons constituting the nucleus in the center of the box .
since the total average number of neutrons is conserved , by changing @xmath212 one actually performs an unphysical study of _ different _ nuclei , surrounded by a neutron gas of a fixed density .
another consequence of the presence of a gas of particles is that the rms nuclear radius can not be calculated in the bcs theory , because the results strongly depend on the box size ( see discussion in refs .
@xcite ) .
it has been suggested in the literature @xcite that the above deficiencies of the bcs approximation can be cured by applying to them the state - dependent - pairing - gap version , where the pairing gap is calculated for every single - particle state using an interaction which is not of the seniority type .
( the corresponding bcs equations resemble the canonical - basis relations ( [ eq151 ] ) . ) in such an approach one hopes that the majority of continuum states would neither contribute to the pairing field ( e.g. , because of their very different spatial character ) nor result in the appearance of the unphysical gas .
this conjecture is tested in fig .
[ fig25 ] ( middle and bottom panel ) where the neutron densities obtained within the state - dependent version of the bcs approximation using the skp@xmath148 and the skp interactions are presented .
it is seen that a reduced coupling of some continuum states to the pairing field does indeed decrease the gas density , however , the asymptotic behavior of the density is still incorrect . in the above plots ,
the shaded lines represent the asymptotic behavior given by eq .
( [ eq155a ] ) assuming @xmath353=0 , i.e. , that of a single - particle state at the fermi energy
. it is seen that a surplus density above this asymptotic limit appears at large distances . however , the deficiencies of the state - dependent bcs approximation , as used for example in refs .
@xcite , are certainly less acute than those of the seniority - pairing bcs . for example , in this type of approach one may probably calculate radii of nuclei much nearer to the drip line .
it is clear that the neutron gas appears in the bcs solutions because of the nonzero occupation probabilities of scattering states .
therefore , one may think that excluding the scattering states from the pairing phase space could be a decisive solution to the problem .
however , for drip - line nuclei , where the fermi energy is by definition close to zero , the remaining phase space would then be small , and this would lead to an artificial quenching of pairing correlations .
moreover , even if the density obtained in such method would vanish asymptotically , the corresponding factor @xmath357 would not be governed by @xmath8@xmath169@xmath7@xmath292mev , as discussed in sec .
[ sec3c ] , but by the single - particle energy , @xmath379@xmath290 , of the highest - energy single - particle state considered in bcs calculations .
this again would lead to densities vanishing at much slower pace than it is required by the hfb theory .
in this section discussed are some experimental consequences of the hfb theory , particularly important for weakly bound nuclei .
pairing gaps are p - p analogs of single - particle energies .
they carry the information about the energies of non - collective excitations , level occupations , odd - even mass differences , and other observables
. the average neutron canonical pairing gaps ( [ eq148b ] ) are shown in figs .
[ fig24a ] ( @xmath151sn ) and [ fig24b ] ( @xmath149sn ) as functions of the canonical single - particle energies ( [ eq148a ] ) . as seen in the middle part of fig .
[ fig24a ] , pairing gaps obtained with the volume - type pairing interaction exhibit very weak configuration dependence . in @xmath151sn
they decrease slightly with @xmath232 but remain confined between 1.0 and 1.5mev . in general , the values of @xmath233 for the @xmath152 states are slightly larger than for other orbitals , which is again related to the volume character of volume delta interaction .
the results presented in the bottom part of fig . [ fig24a ] nicely illustrate the surface character of the skp pairing interaction . indeed , here the pairing gaps increase from 0.5mev ( deep - hole states ) to about 1.251.5mev when the single - particle energies increase towards the fermi energy , and then they decrease again to about 1.0mev for positive single - particle energies .
this is related to the fact that orbitals near the fermi level are concentrated in the surface region .
still another type of behavior is obtained for the finite range gogny interaction ( top part of fig .
[ fig24a ] ) . here , the pairing gaps decrease steadily with single - particle energy . in @xmath151sn
the values of @xmath233 decrease from about 2.5mev for deep - hole states to about 0.75mev for positive - energy states .
( a similar energy dependence of pairing gaps was obtained in the bcs calculations of ref .
@xcite with the renormalized paris potential . )
interestingly , the values obtained for the high-@xmath1 , @xmath213=@xmath1@xmath169@xmath380 orbitals ( antiparallel @xmath381@xmath169@xmath9 coupling ) are significantly larger than those for other orbitals .
the different ranges of @xmath232 values for skp and d1s in fig .
[ fig24a ] reflects the different effective masses in both models .
a rather low effective mass in d1s , @xmath33=0.70 , gives rise a reduced level density and a more bound 1@xmath152 ground state as compared with the skp model ( @xmath33=1 ) .
in fact , due to the non - local exchange contributions to the p - h mean field ( appendix [ appa ] ) , the 1@xmath152 state in the gogny model has the canonical energy lower than the bottom of the local potential well , shown in fig .
[ fig01a ] . in @xmath151sn ,
the hfb+d1s pairing gaps at the fermi energy are of the order of 1.75mev , which slightly overestimates the values corresponding to the odd - even mass staggering in this region .
however , one should bear in mind that the pairing gaps at the fermi energy are rather rough approximations to the odd - even mass difference .
a more accurate description can be obtained by performing blocked hfb calculations for odd - mass isotopes . in the vicinity of @xmath151sn
this method yields the odd - even mass staggering of 1.6mev @xcite for the d1s interaction and of 1.3mev @xcite for the skp interaction .
another contribution to the odd - even mass difference comes from the coupling to the low - lying collective modes .
therefore , the d1s parameters have been adjusted @xcite to give the pairing gap in tin to be 0.3mev larger than the experimental one . on the other hand ,
such a margin has not been taken into account for the skp and skp@xmath148 forces . clearly , a detailed comparison of the values of pairing gaps for the interactions discussed in fig .
[ fig24a ] is delicate .
much more information can actually be derived from the comparison of their dependence on the single - particle energies , which is markedly different .
the general pattern of @xmath233 remains very similar when going to the neutron - rich nucleus @xmath149sn ( fig . [ fig24b ] ) .
in particular , the magnitude of the average pairing gap in deep - hole states depends strongly on the range and density dependence of pairing interaction .
figures [ fig07 ] shows the average neutron pairing gaps ( eqs .
( [ eq157 ] ) and ( [ eq158 ] ) ) for skp , siii@xmath148 , and d1s interactions .
the large values of @xmath366 obtained in hfb+d1s can be explained by : ( i ) an overall larger magnitude of pairing correlations in tin nuclei , and ( ii ) strong pairing correlations in deep - hole states which strongly contribute to the average , eq .
( [ eq158 ] ) .
it is to be noted , however , that despite stronger pairing in d1s , the hfb+d1s pairing gaps vanish at @xmath2=126 ( near the two - neutron drip line ) , in contrast to the hfb+skp result .
this difference may be traced back to a much larger continuum phase space taken into account in our hfb+skp calculations ( sec .
[ sec3e ] ) which are performed in the coordinate representation , and to a larger @xmath2=126 shell gap ( 4.2 mev in @xmath382sn ) obtained with d1s .
( the increase of proton pairing gaps when approaching the proton drip line has been calculated previously in ref .
@xcite with the hfb+skp model and explained in a similar way . )
the disappearance of the neutron pairing at @xmath2=126 in the hfb+siii@xmath148 model is partly due to the volume character of @xmath383 ( a weaker coupling to the particle continuum ) and partly due to a larger @xmath2=126 shell gap @xcite . as discussed in sec .
[ sec2ba ] , diffused nucleonic densities and very strong , surface - peaked , pairing fields obtained with the density - dependent pairing interaction are expected to lead to very shallow single - particle potentials in drip - line nuclei . because of a very diffuse surface ( no flat bottom ) , the resulting single - particle spectrum resembles that of a harmonic oscillator with a spin - orbit term ( but with a weakened @xmath384 term ) @xcite .
schematically , this effect is illustrated in the left panel of fig .
[ fig27 ] . by comparing with the situation characteristic of stable nuclei ( right panel of fig .
[ fig27 ] ) , new shell structure emerges with a more uniform distribution of normal - parity orbits , and the unique - parity intruder orbit which reverts towards its parent shell . such a new shell structure , with no pronounced shell gaps ,
would give rise to different kinds of collective phenomena @xcite .
the effect of the weakening of shell effects in drip - line nuclei , first mentioned in the astrophysical context @xcite , was further investigated in refs .
first analyses of its consequences for the nucleosynthesis have also been performed @xcite .
microscopically , it can be explained by : ( i ) the changes in the mean field itself due to weak binding ( see above ) , and ( ii ) a strong pairing - induced coupling between bound orbitals and the low-@xmath1 continuum .
weakening of shell effects with neutron number manifests itself in the behavior of two - neutron separation energies .
this is illustrated in fig .
[ fig28 ] which displays the two - neutron separation energies for the @xmath2=80 , 82 , 84 , and 86 spherical even - even isotones .
the large @xmath2=82 magic gap , clearly seen in the nuclei close to the stability valley and to the proton - drip line , gradually closes down when approaching the neutron drip line .
a similar effect is seen in the ( @xmath385 ) map of the spherical two - neutron separation energies for the particle - bound even - even nuclei calculated in the hfb+skp model ( fig .
[ fig26 ] ) .
namely , the neutron magic gaps @xmath2=20 , 28 , 50 , 82 , and 126 , clearly seen as cliffs in the @xmath386 surface , disappear for neutron - rich systems .
the gradual disappearance of the neutron shell structure with @xmath2 is not a generic property of all effective interactions . as seen in the plot of @xmath386 and @xmath387 for the tin isotopes ( fig .
[ fig02 ] ) this effect is seen in the skp and skp@xmath31 models , and , to some degree , also in the skp@xmath30 model .
( a weak irregularity at @xmath2=126 reflects the weaker coupling to continuum for the volume pairing @xcite . ) the strong shell effect seen in the siii and skm@xmath32 results has been discussed in ref .
@xcite ; it can be attributed to the low effective mass in these forces .
the result of the d1s model , both for @xmath386 and @xmath387 , is close to that of the skp@xmath30 model .
it is interesting to point out that the qlm calculations of ref .
@xcite ( with @xmath388 ) for the sn isotopes yield very similar results to those of hfb+skp . the very neutron - rich nuclei , as those shown in fig .
[ fig02 ] , can not be reached experimentally under present laboratory conditions .
on the other hand , these systems are the building blocks of the astrophysical r - process ; their separation energies , decay rates , and cross sections are the basic quantities determining the results of nuclear reaction network calculations .
consequently , one can learn about properties of very neutron - rich systems by studying element abundances @xcite .
the recent r - process network calculations @xcite , based on several mass formulae , indicate a quenching of the shell effect at @xmath2=82 in accordance with the results of hfb+skp model .
pairing interaction between bound orbitals and particle continuum is partly responsible for the appearance of particle widths of deep - hole states and the term - repulsion phenomenon ( strong repulsion between single - particle levels ) @xcite . in the dwba and for the local pairing field
@xmath199 the particle width is given by @xmath389 here , @xmath390 is the hf wave function of the bound deep - hole state @xmath391 with the single - particle energy @xmath392 in the absence of pairing , while @xmath393 is the hf wave function of the unbound state with the energy @xmath394
. equation ( [ width ] ) is obtained by assuming that the p - p field of the hfb hamiltonian can be treated perturbatively
. a more consistent way would be to estimate @xmath395 based on self - consistent hfb solutions containing pairing correlations . the proper formulation of the nonperturbative hfb - based theory of deep hole states and one - particle transfer process still needs to be developed .
as discussed in ref .
@xcite , @xmath395 is sensitive to the type of the pairing force .
in general , the widths are larger for surface pairing than for volume pairing . however , the result for an individual state strongly depends on its angular momentum and excitation energy . experimentally , total widths of deep hole states , @xmath396 , are of the order of mev s , ( see , e.g. , refs .
that is , the partial width ( [ width ] ) , of the order of 10 - 100kev , constitutes an extremely small fraction of @xmath396 .
consequently , the experimental determination of @xmath395 alone is very unlikely .
there are many interesting aspects of physics of unstable nuclei which are related to reaction mechanism studies : weak binding , large spatial dimensions , skins ( see , e.g. , refs .
below , we discuss some consequences of surface - peaked pairing fields for pair transfer studies .
an experimental observable that may probe the character of the pairing field is the pair transfer form factor , directly related to the pairing density @xmath127 . the difference in the asymptotic behavior of single - particle density @xmath126 and pair density @xmath127 in a weakly bound system ( see secs .
[ sec2aca ] and [ sec3c ] ) can be probed by comparing the energy dependence of one - particle and pair - transfer cross sections .
such measurements , when performed for both stable and neutron - rich nuclei , can shed some light on the asymptotic properties of hfb densities ; hence on the character of pairing field .
figure [ fig32 ] displays the pair transfer form factors @xmath397 calculated in @xmath151sn , @xmath149sn , and @xmath150sn with the skp interaction .
these microscopic results are compared with the macroscopic form factors @xmath398 @xcite which are determined by using the derivative of the particle density with respect to the neutron number : @xmath399 where @xmath400 is given by eq .
( [ epair ] ) .
this expression can be motivated by the fact that only the orbitals near the fermi surface make significant contributions to the pair density . in the bcs theory , the normalization constant in @xmath401
is usually chosen @xcite as @xmath402=@xmath169@xmath403 . here , we use neither the bcs approximation nor the constant pairing strength @xmath14 .
therefore , the normalization @xmath169@xmath404 is employed .
the derivative in eq .
( [ eq305 ] ) is calculated from the finite difference between the self - consistent results for the hfb vacuum corresponding to particle numbers @xmath2 + 1 and @xmath2@xmath1691 . in these calculations , in order to explore the smooth dependence on the particle number @xmath2 , the odd - average - particle - number vacua have been calculated without using the blocking approximation .
it should be mentioned at this point that the further approximation @xcite of the derivative @xmath405 by the spatial derivative @xmath406 is not justified , because the volume - conservation condition is not valid for the neutron density distribution ( see fig .
[ fig16 ] ) .
the pair transfer form factors in fig . [ fig32 ] clearly show that this process has a predominantly surface character .
the macroscopic form factors have smaller widths and higher maxima than the microscopic ones . on the other hand , they are smaller in the interior of the nucleus as well as in the asymptotic region . in @xmath0-stable nuclei
the macroscopic approximation works fairly well , while in the drip - line nuclei the differences between the two form factors are markedly larger . in general , the corresponding differences are much larger than those obtained within the bcs and the particle - number - projected bcs approaches for the seniority interaction @xcite .
a comparison of the results obtained for different isotopes conspicuously shows a significant increase in the pair transfer form factors in the outer regions of drip - line nuclei . in @xmath151sn ,
the form factors vanish around 9fm , while in @xmath149sn and @xmath150sn they extend to much larger distances .
this effect is particularly pronounced for the microscopic pair transfer form factors .
the importance of the hfb treatment for calculations of nuclear radii has been discussed in several papers @xcite . as mentioned in sec .
[ forces ] , odd - even staggering of rms charge radii is one of the best experimental indicators of the density - dependent pairing . the proper treatment of the pairing effect on radii is especially important for weakly bound systems which exhibit halo or skin effects @xcite ( cf .
discussion in sec .
[ sec3e ] ) . apart from the information on the nuclear rms radii
, one may also gain some experimental insight into the ratios of neutron and proton densities at large distances from the center of nucleus @xcite .
this is possible due to experiments on antiproton annihilation from atomic orbits , which leads to different reaction products depending on whether the process involves a proton or a neutron .
the role of deformation in neutron drip - line nuclei still needs to be investigated .
one can anticipate that due to : ( i ) very diffused surfaces , and ( ii ) strong pairing correlations , the geometric concept of collective deformation ( defined as a deviation of nuclear surface from sphericity ) should be revisited . in this context , the symmetry - unrestricted hfb calculations in coordinate space are called for .
the advent of radioactive nuclear beams provides many exciting opportunities to create and study unstable nuclei far from the @xmath0 stability valley .
one of the unexplored areas far from stability is physics of nuclear pairing in weakly bound nuclei , especially near the neutron drip line .
contrary to the situation characteristic of stable nuclei , the coupling between the p - h field and the p - p field in nuclei with extreme @xmath6 ratios is dramatic ; i.e. , no longer can pairing be treated as a residual interaction .
the main objective of this study was to perform a detailed analysis of various facets of pairing fields in atomic nuclei .
the first part contains the comprehensive summary of the hfb formalism , with particular attention on the physical interpretation of the underlying densities and fields .
very little is known about the p - p component of the nuclear effective interaction ; its structure is of considerable importance not only for nuclear physics but also for nuclear astrophysics and cosmology .
therefore , the second part of this work focuses on the differences between various pairing interactions . in particular ,
the role of density dependence and finite range of the p - p force has been illuminated , and the importance of the coupling to the particle continuum has been emphasized . finally , the third part of our study relates the theoretical formalism to experimental observables ; i.e. , energy spectra , masses , radii , and pair transfer form factors .
it is demonstrated that these observables carry invaluable information that can pin down many basic questions regarding the effective @xmath13 force , and its pairing component in particular .
it should be stressed , however , that in order to see clearly some of the predicted effects , the excursion far from the valley of @xmath0-stability is necessary .
the analysis presented in this paper should be viewed as a useful starting point for future investigations .
one of them is the coupling between collective surface modes ( e.g. , deformation ) and pairing fields in weakly bound nuclei .
another interesting avenue of explorations is the role of dynamics ; e.g. , the importance of the particle number conservation and the coupling to pair vibrations .
a fascinating and difficult research program is the microscopic description of excited states , especially those lying above the particle emission threshold , for which the boundary conditions used in this study ( an impenetrable box ) have to be modified to account explicitly for outgoing waves .
we are only beginning to explore many unusual aspects of the nuclear many - body problem offered by systems with extreme @xmath6 ratios .
interesting discussions with h. flocard , p .- h .
heenen , and h. lenske are gratefully acknowledged .
oak ridge national laboratory is managed for the u.s .
department of energy by lockheed martin energy systems under contract no .
de - ac05 - 84or21400 .
the joint institute for heavy ion research has as member institutions the university of tennessee , vanderbilt university , and the oak ridge national laboratory ; it is supported by the members and by the department of energy through contract no .
de - fg05 - 87er40361 with the university of tennessee .
we thank the department of energy s institute for nuclear theory at the university of washington for its hospitality and partial support during the completion of this work .
this research was supported in part by the u.s .
department of energy through contract no .
de - fg05 - 93er40770 and the polish committee for scientific research under contract no . 2 p03b 034 08 .
the gogny force @xcite is composed of the central , spin - orbit , density - dependent , and coulomb interactions .
the spin - orbit and density - dependent terms have zero - range , and their contributions to the p - h and p - p mean fields are identical to those of the skyrme interaction .
the corresponding expressions can be found in several papers ; e.g. , refs . @xcite , and will not be repeated here . in the following we only consider the central finite - range and coulomb terms .
the central components read @xmath407 where @xmath408 and @xmath409 are the exchange operators for spin and isospin variables , respectively .
this interaction is local ; i.e. , it should be multiplied by @xmath410 before it is inserted in the integrals ( [ eq141 ] ) defining the mean fields .
moreover , it should also be multiplied by the antisymmetrizing operator @xmath411 , where @xmath412 is the exchange operator for space variables .
one usually calls the term involving @xmath412 the exchange term , while the term involving no space exchange is called the direct term . the space , spin , and isospin variables are denoted by @xmath48 , @xmath49=@xmath50 , and @xmath413=@xmath50 , respectively . the parameters @xmath414 , @xmath415 , @xmath416 , @xmath417 , and @xmath418 , belong to the set called d1s @xcite which has been used in this paper . since the expressions given by the @xmath213=1 and 2 components are identical , in what follows we drop the index @xmath213 to increase the legibility of the formulae
since the interaction ( [ 1a ] ) is local , the direct term gives the p - h mean field ( [ eq141a ] ) which is also local , i.e. , @xmath419 \nonumber \\ & + & \delta(\bbox{r } - \bbox{r } ' ) \delta_{\sigma -\sigma ' } \int\text{d}^{3 } \bbox{r}_1 e^{-{{(\bbox{r } - \bbox{r}_1)^2 } \over { \mu^{2 } } } } \sum_{\tau_1 } ( b - m \delta_{\tau \tau_1 } ) \rho^{\tau_1 } ( \bbox{r}_1 \sigma , \bbox{r}_1 -\sigma ) , \label{6 } \end{aligned}\ ] ] where @xmath420 is the density of nucleons ( [ eq320a ] ) of type @xmath413 .
assuming that we consider only the states which are even with respect to the time reversal , the density matrix ( [ eq118a ] ) obeys the relation ( [ 7 ] ) .
consequently , the densities @xmath421 for @xmath49=@xmath50 are equal to @xmath422 , and the densities @xmath423 vanish . therefore , the term in ( [ 6 ] ) , which is proportional to @xmath424 , vanishes , and the contribution of the direct term to the p - h mean field is the local , spin - independent potential : @xmath425 where @xmath426.\ ] ]
one should note that due to the locality of the interaction , the direct term depends only on the local densities .
due to the locality of the interaction , the contribution of the exchange term to the p - h mean field involves no integration : @xmath427 .
\label{12 } \end{aligned}\ ] ] here the time - reversal symmetry does not bring any simplification .
however , a simpler formula is obtained in cases where @xmath428 is real .
it follows from eq .
( [ 7 ] ) that the densities @xmath429 are equal to @xmath430 , which finally leads to @xmath431 \sum_{\sigma_1 } \rho^{\tau_1 } ( \bbox{r } \sigma_1 , \bbox{r } ' \sigma_1 ) , \label{14a } \\ \gamma^{\tau}_{\text{exc } } ( \bbox{r } \sigma , \bbox{r } ' -\sigma ) & = & e^{-{{(\bbox{r } - \bbox{r}')^2 } \over { \mu^{2 } } } } \sum_{\tau_1 } ( h - w \delta_{\tau \tau_1 } ) \rho^{\tau_1 } ( \bbox{r } \sigma , \bbox{r } ' -\sigma ) .
\label{14b } \end{aligned}\ ] ] derivation of the direct and exchange coulomb fields is similar to the one of the finite range term ( [ 1a ] ) with several additional simplifications . when the nuclear state is time - reversal invariant , one obtains the following contributions to the proton p - h mean field in terms of the proton densities : @xmath432 the general form of the pairing field is given by eq .
( [ eq141b ] ) . in this case
the direct and the exchange contributions are equal . for the local central force ( [ 1a ] ) , the total contributions to the p - p mean field have the form : @xmath433 , \label{27 } \\ \tilde{h}^{\tau } ( \bbox{r } \sigma , \bbox{r } ' -\sigma ) & = & e^{-{{(\bbox{r } - \bbox{r}')^2 } \over { \mu^{2 } } } } ( w + b - h - m ) \tilde\rho^{\tau } ( \bbox{r } \sigma,\bbox{r } ' -\sigma ) .
\label{28 } \end{aligned}\ ] ] again it is to be noted that due to the locality of the interaction , the corresponding p - p mean fields do not involve any integration but are proportional to the pairing density matrices . in the case considered in this study ( time - even densities ) , the contribution ( [ 28 ] ) vanishes . since the exchange parameter of the zero - range density - dependent term of the gogny d1s interaction
is fixed at @xmath434=1 , this term does not contribute to the p - p mean field .
moreover , the spin - orbit and coulomb terms usually give small contributions as compared to those of the central force ( [ 1a ] ) .
computation of the exchange p - h mean fields , eqs .
( [ 14a ] ) , ( [ 14b ] ) , and ( [ 19 ] ) , and the pairing fields , eqs . ( [ 27 ] ) and ( [ 28 ] ) , is straightforward .
it only requires the knowledge of the spatial spin - dependent non - local particle , @xmath435 , and pairing @xmath436 densities .
computation of the direct p - h mean field , eq .
( [ 10 ] ) , and the direct coulomb mean field , eq .
( [ 18 ] ) , is more complicated since it requires the evaluation of three - dimensional integrals of the form : @xmath437 in order to compute @xmath438 of eq .
( [ 30 ] ) we note that for the single - particle wave functions expanded in the harmonic oscillator basis , the local density @xmath439 is the product of a gaussian factor and of a polynomial in the spatial coordinates @xmath440 , @xmath441 , and @xmath442 @xmath443 \ ; p ( x_1 , x_2 , x_3 ) , \ ] ] where @xmath444 , @xmath445 and @xmath446 are the ho lengths of the basis states .
consequently , this integral can be evaluated exactly using the gauss - hermite quadrature .
the computation of the coulomb integral ( [ 31 ] ) is more difficult due to the infinite range of the coulomb force .
the method we have used consists of expressing the coulomb force as a sum of gaussians : @xmath447 then one obtains @xmath448 where @xmath449 is given by eq .
( [ 30 ] ) .
in order to perform the remaining one - dimensional integration , the variable @xmath47 is changed to @xmath450 where @xmath451 is the largest of the three harmonic - oscillator lengths @xmath444 , @xmath445 and @xmath446 .
this change of variable is very convenient , since then the range of integration becomes [ 0 , 1 ] .
the integral in eq .
( [ 32 ] ) can be very accurately computed using the gauss - legendre quadrature .
calculations which are based on the schematic pairing interaction , or on the contact force [ eq . ( [ didi ] ) ] require a finite space of states in the p - p channel .
for such interactions , when this space is increased , the pairing energy diverges for any fixed strength of the interaction .
this divergence is a well - known effect @xcite related to the fact that for the contact interactions the matrix elements do not ( or too slowly ) decrease with the excitation energy .
this is not a case for finite - range interactions , such as gogny , for which the pairing energy converges to a finite value .
since it is considerably easier to use the zero - range interactions than the finite - range interactions , one applies the former ones in a limited configuration space determined by a cut - off in the single - particle energy or in the single - quasiparticle energy .
this can be understood as a phenomenological introduction of the finite range @xcite .
there are two other arguments in favor of such a procedure .
first , the scattering of particles in the nuclear medium at very high energies ( or at very small distances ) is very little known , and the particular form offered by any phenomenological finite - range force is very uncertain .
second , the single - particle wave functions are primarily determined by the p - h channel of interaction , and they , in general , spread throughout distances which are much larger than the range of the p - p interaction . therefore , physical differences between the zero- and short - range p - p forces can not be expected to be very pronounced . within the bcs approximation , and assuming a constant density of the single - particle states at large energies , one can derive @xcite a prescription to renormalize the strength of the p - p interaction in such a way that the pairing gap @xmath8 does not depend on the energy cut - off .
suppose that the single - particle states with energies @xmath452@xmath279@xmath379@xmath169@xmath7@xmath279@xmath453 are used to solve the bcs equations for the force of strength @xmath454 .
then , within the specified approximations , the following relation holds : @xmath455 where @xmath456@xmath29300mev@xmath180 is a constant inversely proportional to the density of single - particle states near the fermi energy . in other words , for given values of @xmath456 and @xmath8 ,
( [ v0 ] ) gives values of @xmath457 for any other choice of the cut - off energies @xmath458 and @xmath459 .
since in the present study we use the hfb method instead of the bcs approximation , and since the density of states can hardly be considered to be independent of energy ( actually for fixed @xmath212 it increases as @xmath460 ) , the formula ( [ v0 ] ) can not be directly used
. however , the question as to what extent the pairing strengths can be renormalized for a zero - range p - p interaction can be addressed by analyzing the numerical solutions of the hfb equations .
figure [ fig29 ] ( top panel ) shows the neutron pairing energies @xmath400 [ eq . ( [ epair ] ) ] calculated for the skp@xmath148 interaction which uses the contact p - p interaction ( [ didi ] ) with @xmath454=@xmath169160mev@xmath180 .
it should be recalled at this point that for all coordinate - space hfb calculations presented in this study , the cut - off energy , @xmath461 , depends on the quantum numbers @xmath352 ( cf .
@xcite ) . in the tin nuclei
, @xmath461 decreases from about 40mev for the @xmath152 states to zero for the @xmath462 states . in figs .
[ fig29]-[fig31 ] different curves correspond to different cut - off energies , @xmath463 ( @xmath464 varied between 0 and 40mev ) . hence , @xmath464=40mev corresponds to the cut - off energy @xmath465 of 80mev for the @xmath152 states , and 40mev for the largest values of @xmath1 .
as expected , pairing energies depend significantly on the cut - off energy . comparing results for @xmath466=40mev with those for @xmath466=0 ,
one obtains differences of @xmath400 of the order of 10@xmath46720mev in the mid - shells . due to the self - consistent readjustment of the p - h and p - p energies , the corresponding differences in the total energies ( fig .
[ fig30 ] ) are much smaller , 2@xmath4674mev , but still significant . in the bottom panels of figs .
[ fig29 ] and [ fig30 ] are shown similar results for the renormalized strengths of the contact force ( [ didi ] ) .
the values of @xmath454 , quoted in the caption of fig .
[ fig29 ] , have been obtained by requiring that the average neutron pairing gap in @xmath151sn , @xmath468=1.245mev , does not depend on @xmath466 . with such renormalized interactions ,
one obtains very small changes of total energies ( fig .
[ fig30 ] , bottom panel ) .
the largest deviations do not exceed 200kev and 800kev in stable and exotic isotopes , respectively , and can be safely disregarded when compared to all other uncertainties of methods used to extrapolate to unknown nuclei , or when studying the separation energies .
figure [ fig31 ] shows the effective pairing - interaction strengths defined schematically as @xmath469=@xmath169@xmath470 .
the top panel presents the results obtained for @xmath454=@xmath169160mev@xmath180 and for different values of @xmath466 .
one can see that the dependence on the cut - off energy is very weak , and the @xmath464-dependence of @xmath471 and @xmath400 cancels out in @xmath469 .
( at @xmath2=82 the pairing gap and the pairing energy both vanish , and hence the @xmath469 values can not be calculated . )
a fixed value of @xmath454 gives , therefore , a well - defined , cut - off - independent value of the effective pairing strength for every isotope .
this result , together with the analysis of pairing gap distributions in sec .
[ sec4a ] , demonstrates that calculations employing the volume contact p - p interaction are , in many respects , similar to those with the schematic seniority - pairing force ( cf .
, however , sec .
[ sec3c ] ) .
the values of @xmath469 monotonically decrease with increasing neutron number @xmath2 .
the obtained dependence can be very well described by the simple madland - nix formula @xcite , @xmath14=11mev/(11+@xmath2 ) , while the jensen - miranda formula @xcite , @xmath14=0.18mev[[email protected]@[email protected]@xmath473 $ ] ( @xmath472=(@xmath2@xmath169@xmath3)/@xmath474 ) , gives a much faster decrease of @xmath469 with @xmath2 .
( in a recent study @xcite , based on a schematic finite - range force , the isospin dependence of @xmath14 has been discussed .
the authors found no sign of the @xmath475 term suggested in ref .
this probably explains the disagreement seen in fig .
[ fig31 ] . ) in both expressions we have normalized the multiplicative constants to obtain @xmath14=0.18mev at @xmath2=@xmath3=50 . the bottom panel of fig .
[ fig31 ] shows similar results , but for the renormalized values of @xmath454 , quoted in the caption of fig .
[ fig29 ] .
the insert shows the values of @xmath454 ( dots ) as function of @xmath464 compared with the simple fit by the formula ( [ v0 ] ) with @xmath453=@xmath476+@xmath464 , for @xmath476=40mev , @xmath456=430mev@xmath180 , and @xmath8=5.58mev ( solid line ) .
one can see that the generic dependence of the renormalized values of @xmath454 on @xmath466 is fairly well reproduced , although the numerical constants @xmath456 and @xmath8 obtained from the fit do not exactly correspond to the values inferred from the bcs theory with a constant density of single - particle states . .
results of the hfb calculations with skp force for the @xmath152 neutrons in @xmath151sn . for the @xmath216th quasiparticle state , @xmath477 and @xmath217 are the quasiparticle energy and the norm of the lower component , respectively . for the @xmath47th canonical - basis state , @xmath231 is the occupation probability , @xmath232 and @xmath233 are , respectively , the average values of the p - h and p - p mean - field hamiltonians [ eq .
( [ eq148 ] ) ] and @xmath478 is the bcs - like quasiparticle energy defined in eq .
( [ eq150 ] ) .
all energies are in mev .
[ cols="^,>,^,^ , > , < , > , > " , ] | ground - state properties of exotic even - even nuclei with extreme neutron - to - proton ratios are described in the framework of the self - consistent mean - field theory with pairing formulated in coordinate space .
this theory properly accounts for the influence of the particle continuum , which is particularly important for weakly bound systems .
the pairing properties of nuclei far from stability are studied with several interactions emphasizing different aspects , such as the range and density dependence of the effective interaction .
measurable consequences of spatially extended pairing fields are presented , and the sensitivity of the theoretical predictions to model details is discussed . |
a web crawler traditionally fulfills two purposes : discovering new pages and refreshing already discovered pages .
both of these problems have been extensively investigated over the past decade ( see the survey paper by olston and najork @xcite ) .
however , recently , the role of the web as a media source became increasingly important as more and more people start to use it as their primary source of up - to - date information .
this evolution forces crawlers of web search engines to continuously collect newly created pages as fast as possible , especially high - quality ones . surprisingly , user traffic to many of these newly created pages grows really quickly right after they appear , but lasts only for a few days .
for example , it was discussed in several papers that the popularity of news decreases exponentially with time @xcite .
this observation naturally leads to distinguishing two types of new pages appearing on the web : _ ephemeral _ and _ non - ephemeral _ pages . note that here we do not consider the ephemeral content , which might be removed before it hits the index as in @xcite ( e.g. advertisements or the `` quote of the day '' ) , but we consider persistent content that is ephemeral in terms of user interest ( e.g. news , blog and forum posts ) .
we clustered user interest patterns of some new pages discovered in one week ( see section [ contentsources ] for details ) , and figure [ fig : yabar ] shows the centroids of the obtained clusters . in section [ contentsources ] , we show that a significant fraction of new pages appearing on the web every day are ephemeral pages .
the cost of the time delay between the appearance of such ephemeral new pages and their crawl is thus very high in terms of search engine user satisfaction . moreover , if a crawler fails to find such a page during its period of peak interest , then there might be no need to crawl it at all .
it was reported in @xcite , that 1 - 2% of user queries are extremely recency sensitive , while even more are also recency sensitive to some extent .
the problem of timely finding and crawling ephemeral new pages is thus important , but , to the best of our knowledge , is not studied in the literature . indeed ,
different metrics were suggested to measure the coverage and freshness of the crawled corpus @xcite , but they do not take into account the degradation of the profit to a search engine contributed by these pages . crawling policies based on such metrics may then crawl such new pages not quickly enough , and even crawl already obsolete content .
thus , we need a new quality metric , well thought out for this task , and a crawling algorithm optimized to maximize this metric over time , that takes into account this degradation of pages utility .
our daily experience of using the web also suggests that such ephemeral new pages can be found from a relatively small set of `` hubs '' or _ content sources_. we investigate this intuition and show that it is possible and practical to find such sources at scale .
examples of content sources are main pages of blogs , news sites , category pages of such news sites ( e.g. politics , economy ) , rss feeds , sitemaps @xcite , etc . , and one needs to periodically recrawl such sources in order to find and crawl ephemeral new pages way before their peak of user interest . however
, frequent recrawling of all these sources and all new pages found on them requires a huge amount of resources and is quite inefficient . in order to solve this problem efficiently ,
we analyze the problem of dividing limited resources between different tasks ( coined as _ holistic crawl ordering _ by oslton and najork in @xcite ) , i.e. , here between the task of crawling ephemeral new pages and the task of recrawling content sources in order to discover those new pages . a possible solution for this problem
is to give a fixed quota to each policy ( see , e.g. , @xcite ) , but we will show that such solutions based on fixed quotas are far from being optimal . in this paper , we propose a new algorithm that dynamically estimates , for each content source , the rate of new links appearance in order to find and crawl newly created pages as they appear . as a matter of fact , it is next to impossible to crawl all these new pages immediately due to resource constraints , therefore , a reasonable crawling policy has to crawl the highest quality pages in priority .
the quality of a page can be measured in different ways , and it can , for example , be based on the link structure of the web graph ( e.g. , in - degree @xcite or pagerank @xcite ) , or on some external signals ( e.g. , query log @xcite or the number of times a page was shown in the results of a search engine @xcite ) . in this paper , we propose to use the number of clicks in order to estimate the quality of pages , and predict the quality of newly created pages by using the quality of pages previously linked from each content source . by the number of clicks , we mean the number of times a user clicked on a link to this page on a search engine results page ( serp ) , which most reliably indicates a certain level of user interest in the page s content . in this way , we are able , in fact , to incorporate user feedback into the process of crawling for our algorithm to find and crawl the best new pages . to sum up , this paper makes the following contributions : * we formalize the problem of timely crawling of high - quality ephemeral new web content by suggesting to optimize a new quality metric , which measures the ability of a crawing algorithm to solve this specific problem ( section [ formalization ] ) .
* we show that most of such ephemeral new content can be found at a small set of content sources , and we propose a method to find such sources ( section [ contentsources ] ) . *
we propose a practical algorithm , which periodically recrawls content sources and crawls newly created pages linked from them , as a solution of this problem .
this algorithm uses user feedback to estimate the quality of content sources ( section [ algo ] ) .
* we validate our algorithm by comparing it to other crawling strategies on real - world data ( section [ exp ] ) . besides , in section [ relatedwork ] , we review related work , while in section [ conclusion ] , we conclude the paper and discuss possible directions for future research .
in this section , we formalize the problem under consideration by introducing an appropriate quality metric , which measures the ability of a crawling algorithm to solve this problem . as we discussed in the introduction
, we deal with pages for which user interest grows within hours after they appear , but lasts only for several days .
the profit of crawling such ephemeral new pages thus decreases dramatically with time .
assume that for each page @xmath0 , we know a decreasing function @xmath1 , which is the profit of crawling this page with delay @xmath2 seconds after its creation time @xmath3 ( by profit , one can mean the expected number of clicks or shows on serp ) .
if , finally , each page @xmath0 was crawled with a delay @xmath4 , we can define the _ dynamic quality _ of a crawler as : @xmath5 } p_i ( \delta t_i).\ ] ] in other words , the dynamic quality is the average profit gained by a crawler per second in a time window of size @xmath6 . the dynamic quality defined above
can be useful to understand the influence of daily and weekly trends on the performance of a crawler .
let us now define the _ overall quality _ of a crawler , which allows to easily compare different algorithms over larger time windows .
it is natural to expect that if @xmath6 is large enough then the influence of season and weekly trends of user interest will be reduced . in other words
, the function @xmath7 tends to a constant while @xmath6 increases .
thus , we can consider the _ overall quality _ : @xmath8 } p_i ( \vartriangle t_i),\ ] ] which does not depend on @xmath9 and @xmath6 . in this paper , by @xmath1 profit of crawling a page @xmath0 at time @xmath10 , we mean the total number of clicks this page will get on a serp after this time ( ignoring any indexing delay ) . in this way
, we can approximate the relevance of a page to current interests of users . from a crawler s perspective
, it is thus an appropriate measure of the contribution of new pages to a search engine performance ( given a specific ranking method ) . alternatively , instead of the number of clicks , we could use the number of shows , i.e. , the number of times a page was shown in the top @xmath11 results of a search engine .
this value also reflects a crawler s performance in the sense that we only want to crawl ephemeral new pages , which are going to be shown to users .
but , as we discuss further in this section , the number of clicks and the number shows behave similarly and we thus use clicks since they reflect the actual preference of users . at this point
, we have defined a new metric to measure the quality of crawling ephemeral new pages .
we are going to use this metric to validate our algorithm in section [ exp ] .
note , that this metric can only be evaluated with some delay because we need to wait when crawled pages are not shown anymore to take their profit , i.e. , their number of clicks , into account .
0.5 0.5 however , for our crawling algorithm , we do not want to wait for such a long period of time to be able to use the profit of freshly crawled pages in order to quickly adapt to changes of content sources properties .
of course , we do not know the function @xmath1 for pages that just appeared , but we can try to predict it .
it is natural to expect that pages of similar nature exhibit similar distributions of user interest . in order to demonstrate this , on figure [ fig : avg_shows ] and figure [ fig : avg_clicks ] we plot respectively the average number of cumulative shows and clicks depending on the page s age for all pages published on a news site and a blog ( both being randomly chosen ) over @xmath12 week .
we can see that almost all clicks and shows appear in the first week of a page s life , and that the dependency of the cumulative number of clicks ( shows ) gathered by a page on the page s age is pretty well described by the function : @xmath13 , where @xmath14 is the total number of clicks ( shows ) a page gathers during its life .
we thus propose the following approximation of the profit @xmath1 ( i.e. , the number of future clicks ) : @xmath15 where the _ rate of decay _ @xmath16 and the _ profit _
@xmath17 are content - source - specific and should be estimated using historical data ( see section [ p&mu ] for details ) .
we use this approximation in section [ algo ] in order to analyze the problem under consideration theoretically .
[ howtofind ] in this section , we show that most ephemeral new content can indeed be found at a small set of content sources and then describe a simple procedure for finding such a set , that fits our use case .
our hypothesis is that one can find the most of ephemeral new pages appearing on the web at a small set of content sources , but links from these sources to new pages are short living so a crawler needs to frequently recrawl these sources to avoid missing links to new pages , especially to high - quality pages . in order to validate this hypothesis about content sources ,
we need to follow the evolution over time of the link structure of the web , to understand which content sources refer which new pages as they appear . our web crawler logs could be used for this , but there are two main issues with this approach : 1 ) keeping the full history of new pages linked from each content source , even for some small time period , is impractical due to resource constraints ; and 2 ) existing crawlers do not revisit each content source often enough to provide more than a really partial view of the evolution of the link structure of the web .
so , instead , we used the toolbar logs continuously collected at yandex ( russia s most popular search engine ) to monitor user visits to web pages . in this way
, we can easily track the appearance of new pages that are of interest to at least one user , know which content sources referred them , and also follow the evolution of user interest over time for each of these new pages .
this data is a representative sample of the web as this toolbar is used by millions of people across different countries .
but we can not use this data in the algorithm itself since it is not available in all countries , and thus we use it only in order to validate our hypothesis about the existence of relatively small set of content sources . using this toolbar data , we randomly sampled 50k pages from the set of new pages that appeared over a period of one week and were visited by at least one user .
these pages were distributed over @xmath18k different hosts . for each page
, we computed its number of daily visits for a period of two weeks after it appeared .
then , using this 14-dimensional ( one per day ) feature vector ( scaled to have its maximum value equal to 1 ) , we clustered these pages into 6 clusters by applying the k - means method .
let us note that when we tried less clusters , non - ephemeral pages were not assigned to one cluster .
finally , we obtained only @xmath19% non - ephemeral pages .
the percentage of new pages that are ephemeral ( and were visited at least once ) for this week is thus 96% , which is really significant .
centroids of these clusters are plotted on figure [ fig : centroids-6 ] ( in section [ intro ] we showed only two of them ) . our toolbar logs also contain , in most cases , the referrer page for each recorded visit to a page , i.e. , the source page from which the user came to visit this target page .
we extracted all these _ links _ ( users transitions between pages ) found in the logs pointing to one of the ephemeral new pages in our sample over the same period of one week plus several days ( for the most recent pages to become obsolete ) , and obtained @xmath20k links . using these links , we studied the percentage of ephemeral new pages reached depending on the number of content sources .
we want to find the smallest set of content sources that allows to reach the most of new pages and thus proceed as follows ( in a greedy manner ) .
we first take the source page , which allows to reach the most of pages , then remove it and all covered pages , then select the second one in the same way , and so on .
we see on figure [ fig : hubs ] , that only 3k content sources are required to cover 80% of new content , which validates our hypothesis about content sources .
interestingly , 42% of these 3k content sources are main pages of web sites , while 44% are category pages , which can be accessed from the main page .
so , overall , 86% of them are at most 1 hop away from the main page .
now , we need to understand how to effectively find these content sources at scale without relying on toolbar logs , which , as said , are not available world - wide .
our crawling algorithm ( described in the next section ) focuses on high - quality ephemeral new content .
therefore , even if content sources that produce low quality content or almost no new content at all are given to this algorithm , they will almost never be crawled or crawled just in case much later when some spare resources are available ( see section [ algo ] ) .
we are thus , to some extent , only interested in recall when finding such sources here , i.e. , to get most of them , which makes this task much easier to solve in practice .
analyzing the set of content sources discovered using toolbar data , we noticed that 86% of content sources that we found are actually at most 1 hop away from the main page of their host , as said .
the following procedure will thus yield a relatively small set of content sources that generate most of the ephemeral new content on a given set of hosts . 1 .
crawl the main page of each host ( once ) and keep pages linked from it ( all pages 1-hop away from the main page ) ; 2 . select the main page , and all found pages older than few days , by using historical data , as content sources ( as new pages are almost never content sources ) .
let us note that this procedure is easy to run periodically to refresh the set of content sources , and to find new content sources .
it is possible to use url patterns as described in @xcite in order to get a better precision , but this optimization is not required here , because our crawling algorithm optimizes precision itself by avoiding to crawl low - quality content sources ( see section [ algo ] ) . the input list of hosts for this procedure can be obtained during a standard web crawling routine by ranking found hosts by their tendency to generate new content .
this simple method also fits our usage scenario considering that , as said , only recall is important for us when finding content sources as input for our algorithm .
in this section , we assume that we are given a relatively small set of content sources , which regularly generate new content ( see section [ howtofind ] for the procedure to select such a set ) .
our current aims are to ( 1 ) find an optimal schedule to recrawl content sources in order to quickly discover high - quality ephemeral new pages , and ( 2 ) understand how to spread resources between crawling new pages and recrawling content sources .
first , we analyze this problem theoretically and find an optimal solution .
then , we describe an algorithm , which is based on this solution .
assume that we are given a set of content sources @xmath21 .
note that the rate of new content appearance may differ from source to source .
for example , usually there are much more news about politics than about art , and therefore , different categories of a news site generate new content with different rates .
let @xmath22 be _ the rate of new links appearance _ on the source @xmath23 , i.e. , the average number of links to new pages , which appear in one second .
let us consider an algorithm , which recrawls each source @xmath23 every @xmath24 seconds , discovers links to new pages , and also crawls all new pages found .
we want to find a schedule for recrawling content sources , which maximizes the overall quality @xmath25 ( see equation ( [ q ] ) ) , i.e. , our aim is to find optimal values of @xmath24 .
suppose that our infrastructure allows us to crawl @xmath26 pages per second ( @xmath26 can be non - integer ) .
due to these resource constraints , we have the following restriction : @xmath27 on average , the number of new links linked from a source @xmath23 is equal to @xmath28 , therefore every @xmath24 seconds we have to crawl @xmath29 pages ( the source itself and all new pages found ) .
obviously , the optimal solution requires to spend all the resources : @xmath30 and we want to maximize the overall quality ( see equation ( [ q ] ) ) , i.e. , @xmath31 } p_j(\vartriangle t_j ) \rightarrow \max.\ ] ] note that this expression is exactly the average profit per second .
content sources may not be equal in quality , i.e. , some content sources may provide users with better content than others .
we now assume that , on average , the pages from one content source exhibit the same behavior of the profit decay function and hence substitute @xmath32 by the approximation @xmath33 discussed in section [ formalization ] .
we treat the profit @xmath17 and the rate of profit decay @xmath16 as the parameters of each content source @xmath23 .
thus , we obtain : @xmath34 @xmath35 here @xmath36 and @xmath37 . without loss of generality
, we can assume that @xmath38 .
we now want to maximize @xmath39 subject to ( [ restriction ] ) .
we use the method of lagrange multipliers : @xmath40 where @xmath41 is a lagrange multiplier .
note that @xmath42 , so we get : @xmath43 the function @xmath44 increases monotonically for @xmath45 with @xmath46 and @xmath47 .
if we are given @xmath48 , then we can find the unique value @xmath49 as shown in figure [ fig : omega ] .
one can easily compute @xmath50 using a binary search algorithm .
note that bigger values of @xmath51 lead to bigger values of @xmath52 .
that is why @xmath53 is a monotonic function of @xmath51 and we can , here also , apply a binary search algorithm ( see algorithm [ algo_disjdecomp ] ) to achieve the condition @xmath54 .
@xmath55 let @xmath56 and @xmath57 be , respectively , the lower and upper bounds for @xmath51 .
at the first step , we can put @xmath58 and @xmath59 .
indeed , @xmath60 is the obvious upper bound for @xmath51 , since in this case we do not crawl any content source . at each step of the algorithm
, we consider @xmath61 .
for this value of @xmath51 , we recompute intervals @xmath62 .
note that if we get @xmath63 for some @xmath64 , then @xmath65 and we never recrawl this content source .
after that , if @xmath66 , then we can put @xmath67 , since it is an upper bound .
if not , we put @xmath68 .
we proceed in this way until we reach the required precision @xmath69 .
the value of @xmath51 may be interpreted as the threshold we apply to content sources utility .
actually , we can find the minimal crawl rate required for the optimal crawling policy not to completely refuse to crawl content sources with the least utility .
we completely solved the optimization problem for the metric suggested in section [ formalization ] , the solution of ( [ solution2 ] ) is theoretically optimal ( we use the name _ echo - based crawler _ for the obtained algorithm , where echo is an abbreviation for ephemeral content holistic ordering ) . however , some further efforts are required in order to make this algorithm practically useful .
there are parameters , which we need to estimate for each source : the profit @xmath17 , the rate of profit decay @xmath16 , and the rate of new links appearance @xmath22 . in the following section
, we describe a practical crawling algorithm , which is based on our theoretical results .
let us describe a concrete algorithm based on the results of section [ theory ] .
first , we use the results from section [ howtofind ] to obtain an input set of content sources .
then , in order to apply algorithm [ algo_disjdecomp ] for finding an optimal recrawl schedule for these content sources , we need to know for each source its profit @xmath17 , the rate of profit decay @xmath16 , and the rate of new links appearance @xmath22 .
we propose to estimate all these values dynamically using the crawling history and search engine logs .
since these parameters are constantly changing , we need to periodically re - estimate time intervals @xmath24 ( see algorithm [ algo_disjdecomp ] ) , i.e. , to update the crawling schedule . obviously , the more often we re - estimate @xmath24 , the better results we will obtain , and the choice of this period depends on the computational resources available . thus , we first discuss how to estimate these sources characteristics ( sections [ p&mu ] and [ lambda ] ) and then how to deal with deviations of content sources behavior from our idealistic assumptions to make a practical scheduling algorithm ( section [ scheduling ] ) .
for this part , we need search engine logs to analyze the history of clicks on new pages . we want to approximate the average cumulative number of clicks depending on the page s age by an exponential function .
this approximation for two chosen content sources is shown on figure [ fig : avg_clicks ] .
let us consider a cumulative histogram of all clicks for all new pages linked from a content source , with the histogram bin size equals to @xmath70 minutes .
let @xmath71 be the number of times all @xmath26 new pages linked from this content source were clicked during the first @xmath72 minutes after they appeared .
so , @xmath73 is the average number of times a new page was clicked during the first @xmath72 minutes .
we can now use the least squares method , i.e. , we need to find : @xmath74 in other words , we want to find the values of @xmath75 and @xmath14 , that minimize the sum of the squares of the differences between the average cumulative number of clicks and its approximation @xmath76 .
it is hard to find an analytical solution of ( [ leastsquares ] ) , but we can use the gradient descent method to solve it : [ gradientdescent ] @xmath77 ; from the production point of view , it is very important to decide , how often to push data from search engine logs to re - estimate the values of @xmath16 and @xmath17 as it is quite an expensive operation .
we denote this _ logs push period _ by @xmath78 . in section
[ exp ] , we analyze how the choice of @xmath78 affects the performance of the algorithm .
the rate of new links appearance @xmath79 may change during the day or during the week .
we thus dynamically estimate this rate for each content source . in order to do this ,
we use historical data : we consider the number of new links found at each content source during the last @xmath6 crawls .
we analyze how different values for @xmath6 affect the performance of the algorithm in section [ exp ] .
finally , in order to apply our algorithm , we should solve the following problem : in reality the number of new links that appear on a content source during a fixed time period is random and we can not guarantee that we find exactly @xmath28 new links after each crawl .
we can find more links than expected after some recrawl and if we crawl all of them , then we will deviate from the schedule .
therefore , we can not both stick to the schedule for the content sources and crawl all new pages .
so we propose the two following variants to deal with these new pages , that we can not crawl without deviating from the schedule .
* echo - newpages .
* in order to avoid missing clicks , we always crawl newly discovered pages right after finding them .
if there are no any new pages in the crawl frontier , we try to come back to the schedule .
we crawl the content source , which is most behind the schedule , i.e. , with the highest value of @xmath80 , where @xmath81 is time passed after the last crawl of the @xmath0-th content source . *
* echo - schedule.**we always crawl content sources with intervals @xmath24 and when we have some resources to crawl new pages , we crawl them ( most recently discovered first ) .
we compare these two variants experimentally in the next section .
we finish this section by presenting a possible production architecture for our algorithm ( see figure [ fig : architecture ] ) to emphasize that it is highly practical to implement it in a production system .
initially , a set of content sources with good recall of new pages linked from these sources is created using the procedure described in section [ contentsources ] .
this procedure must be run periodically to refresh this set and include new content sources .
then , _ scheduler _ finds the optimal crawling schedule for content sources , while _ fetcher _ crawls these sources according to this schedule .
_ scheduler _ dynamically estimates the rate of new links appearance for content sources and also estimates the profit decay function using the number of clicks from the search engine logs .
given a fixed ranking method , the number of clicks measures the direct effect of the crawling algorithm on the search engine s performance .
_ scheduler _ dynamically uses this feedback in order to improve the crawling policy .
in this section , we compare our algorithm with some other crawling algorithms on real - world data . since it is impractical and unnecessary to conduct research experiments at a production scale , we selected some sites that provide a representative sample of the web , on which we performed our experiments .
we selected the top 100 most visited russian news sites and the top 50 most visited russian blogs using publicly available data from trusted sources .
we consider these web - sites to be a representative sample of the web for our task as they produce 5 - 6% out of the @xmath82k new pages ( visited by at least one user ) that appear in this country daily ( we estimated this second value using toolbar logs ) . for each such site
, we applied the procedure described in section [ contentsources ] and obtained about 3k content sources .
then , we crawled each of these content sources every 10 minutes for a period of 3 weeks ( which is frequent enough to be able to collect all new content appearing on them before it disappears ) .
the discovery time of new pages we observed is thus at most delayed by these 10 minutes .
we considered all pages found at the first crawl of each source ( each content source was crawled @xmath83 times ) to be old and discovered @xmath84k new pages during these @xmath85 weeks . keeping track of
when links to new pages were added and deleted from the content sources , we created a dynamic graph that we use in the following experiments .
this graph contains @xmath86 m unique links . additionally , we used search engine logs of a major search engine to collect user clicks for each of the newly discovered pages in our dataset for the same period of @xmath85 weeks plus @xmath12 week for the most recent pages to become obsolete .
we observed that @xmath87% of the pages were clicked at least once during this 4 weeks period .
we compare the algorithm suggested in section [ algo ] with several other algorithms .
there are no state - of - the - art algorithms for the specific task we discuss in this paper , but one can think of several natural ones : * * breadth - first search ( bfs ) * we crawl content sources sequentially in some fixed random order . after crawling each source ,
we crawl all new pages linked from this source , which have not been crawled yet .
we also compare our algorithm with the following simplifications to understand the importance of 1 ) the holistic crawl ordering and 2 ) the usage of clicks from search engine logs . *
* fixed - quota * this algorithm is similar to _
echo- + schedule _ , but we use a fixed quota of @xmath88 for recrawling content sources and for crawling new pages that have not been crawled before . * * frequency * this algorithm is also similar to _ echo - schedule _ , but we do not use clicks from search engine logs , i.e. , all content sources have the same quality and content sources are ordered only by their frequency of new pages appearance .
we also propose a simplification of our algorithm , based on section [ algo ] , which could be much easier to implement in a production system . * * echo - greedy * we crawl the content source with the highest expected profit , i.e. , with the highest value of @xmath89 , where @xmath90 is the time passed since the last crawl of the content source , @xmath22 is its rate of new links appearance , and @xmath17 is the average profit of new pages linked from the content source . then , we crawl all new pages linked from this source , which have not been crawled yet , and repeat this process . in this section ,
we experimentally investigate the influence of parameters on our algorithm s performance and compare the algorithm with the approaches from section [ benchmarks ] on real - world data .
we simulated , for each algorithm , the crawl of the dynamic graph described in section [ data ] , using the content sources as seed pages .
each algorithm can thus , at each step , decide to either crawl a newly discovered page or to recrawl a content source in order to find new pages . in the following experiments analyzing parameters influence , we used the crawl rate per second @xmath91 .
this crawl rate is enough to crawl a significant fraction of the new pages as shown on figure [ fig : lambda ] , but is not too high to let bfs algorithm crawl all new pages ( which is highly unrealistic in a production context ) .
we then also use two other crawl rates @xmath92 and @xmath93 per second to investigate the influence of this value .
we apply algorithm [ algo_disjdecomp ] to re - estimate @xmath24 values every 30 minutes , which is frequent enough so that smaller intervals have almost no influence on its performance , and which is also realistic in a production context .
we also set the bin size @xmath70 used in algorithm 2 to 20 minutes , which is good enough to have robust estimations of @xmath17 and @xmath16 , as , typically , the profit decay function @xmath94 does not change significantly in such a small time period .
we do not study in details , here , the influence of these two parameters due to space constraints as , according to the experiments we performed , it is negligible , for values below these realistic choices , in comparison with other parameters . besides that
, we need default values for profits @xmath17 as we start crawling without knowing anything about the quality of each content source .
please , note that we need to use _
pessimistic _ default values because we want to avoid crawling low quality sources too frequently , while we do not have enough feedback to have precise estimations .
we can not use @xmath95 as according to algorithm [ algo_disjdecomp ] , we do not crawl content sources with zero profit , so , we used some small non - zero value @xmath96 .
we compared the two variants of echo - based crawler from section [ scheduling ] with different values for : 1 ) the crawl history size used to estimate the rate of new links appearance @xmath79 discussed in section [ lambda ] ( from 3 to 10 crawls ) , and 2 ) the logs push period @xmath78 , which was described in section [ p&mu ] ( we considered 1h , 12h , 24h , and 1 week ) . interestingly , for both variants we noticed no difference , and we therefore conclude that these parameters do not affect the final quality of the algorithm in our setup . on the other hand ,
the logs push period has a really big influence during the warm - up period and the smaller the logs push period , the better the results ( see figure [ fig : profit_7 ] ) .
there is nothing interesting to observe for the crawl history size .
let us also note that the optimal schedule of echo - based algorithms almost does not recrawl 70% of content sources , which means that it does not spend much resources on low quality content sources .
.average dynamic profit for a 1-week window . [ cols="<,^,^,^,^",options="header " , ] then , we took the crawl history of size 7 and the logs push period of 1 hour ( randomly , following the discussion in section [ influence ] ) , and compared echo - based crawlers with other algorithms on three different crawl rates . in order to compare our algorithms during the last week of our observations ( after the warm - up period ) we measured the dynamic profit every two minutes using a time window of one week ( enough to compensate daily trends ) .
table [ table1 ] shows average values and their standard deviations .
note that we also include the upper bound of algorithms performance that we computed using bfs algorithm with an unbounded amount of resources , which allows to crawl all new pages right after they appear .
this upper bound therefore does not depend on the crawl rate and equals @xmath97 of profit per second .
echo - newpages shows the best results , which are really close to the upper bound , although the crawl rate used is much smaller than the rate of new links appearance .
this means that our algorithm effectively spends its resources and crawls highest quality pages first .
note that the smallest crawl rate that allows bfs to reach 99% of the upper bound is 1 per second ( this value is measured , but not present in the table ) , as bfs wastes lots of resources recrawling content sources to find new pages , while echo - newpage and echo - schedule reach this bound with crawl rate 0.2 per second ( see the last column of the table ) . note that the profit of echo - greedy is also high .
this fact can be a good motivation for using it in a production system , where ease of implementation is a strong requirement ( as it is much easier to implement ) .
first , it only requires a priority queue of content sources rather than a recrawl schedule updated using the binary search method from algorithm [ algo_disjdecomp ] .
second , it does not use @xmath16 , so @xmath17 is thus simply the average number of clicks on pages linked from the @xmath0-th content source , and can therefore be computed easier than by using the gradient descent method from algorithm 2 .
let us show a representative example ( at @xmath91 ) demonstrating the advantage of echo - based algorithms over the baselines ( see figure [ fig : dynamic ] ) .
one can observe that echo - based algorithms perform the best most of the time .
it is interesting to note though that during the night bfs shows better results .
it happens as bfs is `` catching up '' by crawling pages , which were crawled by other algorithms earlier .
this follows how the dynamic profit is defined : we take into account the profit of the pages , which were crawled during the last 5 hours .
we also see that the algorithm with fixed quota for crawl and recrawl perform well during the weekend because less new pages appear during this period and the crawl rate we use is thus enough to crawl practically all good content without additional optimizations .
most papers on crawling are devoted to discovering new pages or refreshing already discovered pages . both of these directions are , to some extent , related to the problem we deal with , though can not serve as solutions to it .
[ [ refresh - policies ] ] refresh policies + + + + + + + + + + + + + + + + the purpose of refresh policies is to recrawl known pages that have changed in order to keep a search engine s index fresh . usually , such policies are based on some model , which predicts changes on web pages . in pioneering works
@xcite , the analysis of pages changes was made in the assumption of a time - homogeneous poisson process , i.e. , it was assumed that the pages change rate does not depend on time .
however , in @xcite , it was noted that there are daily and weekly trends in the pages change rate .
then , a history - based estimator , which takes such trends into account , was proposed in @xcite . a more sophisticated approach based on machine learning
is used in @xcite , where the page s content , the degree of observed changes and other features are taken into account . for our specific task ,
refresh policies can be used to find links to ephemeral new pages , that appeared on already known pages ( content sources ) .
so , pages changes are relevant for us only if new links to such new pages can be found .
interestingly , this simplifies the estimation of pages change rate as one can easily understand , given two successive snapshots of a page , that two new links appeared , while it is much harder to know if the page s text changed once or twice .
this fact allows us to use a simple estimator for the rate of new links appearance , which reflects timely trends .
of course , more sophisticated methods ( e.g. , using machine learning @xcite ) , can be applied here , but it was out of focus of the current paper .
moreover , our method actually monitors content sources updates to avoid missing new links . in this way , our work is more related to the problem of creating an efficient rss feeds reader , which needs to monitor rss feeds updates to avoid missing new postings @xcite .
the rss reader described in these papers learns the general posting pattern of each rss feed to create an efficient scheduling algorithm that optimizes the retrieval of rss feeds in order to provide timely content to users .
this rss reader uses a short description from the rss feed , when presenting an article to users , and there is thus no need for it to crawl these articles .
it is not exactly our case as we need to crawl and index newly discovered pages to allow users to access them via the search engine .
thus , although rss monitoring policies are somehow similar to the problem of finding ephemeral new pages on content sources , one can not use such policies out of box for our task .
the main reason is that we need to spend significant amount of resources for crawling new pages and , if we want to do this in an efficient way , then we need to change the recrawl schedule to take this fact into account
. however , rss feeds themselves can be used in our approach as content sources .
[ [ discovery - policies ] ] discovery policies + + + + + + + + + + + + + + + + + + the main idea behind discovery policies is to focus breadth - first search on high quality content , i.e. , to prioritize discovered , but not yet crawled pages ( the crawl frontier ) according to some quality measure .
some approaches are based on the link structure of the web graph , e.g. , in @xcite pages with the largest number of incoming links are crawled first , while pages with largest pagerank are prioritized in @xcite . in @xcite ,
such approaches were compared in their impact on web search effectiveness . however , as pandey and olston discussed in @xcite , the correlation between link - based importance measures and user interest is weak , and hence they proposed to use search engine logs of user queries to drive the crawler towards pages with higher potential to be interesting for users . in turn , we follow recent trends and do not rely on the link structure of the web , but use clicks from search engine logs .
our crawler discovers and crawls new pages , but there is a principal difference between our approach and previous ones .
previous approaches are based on the assumption that the web is relatively deep and therefore , starting from some seed pages , a crawler needs to go deeper , in direction of high - quality pages if possible , to find new pages .
we instead argue that one can find most ephemeral new pages that are appearing on the web at a relatively small set of content sources , but that a crawler needs to frequently recrawl these sources to avoid missing short living links to new pages .
this observation , on one hand , simplifies the problem but , one the other hand , introduces new challenges to find the right balance between crawling new pages and recrawling content sources .
[ [ holistic - crawl - ordering ] ] holistic crawl ordering + + + + + + + + + + + + + + + + + + + + + + + usually , papers about crawling focus either on discovery of new pages or on refreshing already known pages , but the important question of how to divide limited resources between refreshing and discovery policies is usually underestimated .
some authors proposed to give a fixed quota to each policy @xcite .
however , as it follows , e.g. , from our analysis ( see section [ theory ] ) , such fixed quotas can be far from optimal .
in contrast , our optimization framework simultaneously deals with refreshing and discovery and can thus find an optimal way to share resources .
moreover , the problem of making a holistic crawl ordering , i.e. , to unify different policies into a unified strategy , was proposed by oslton and najork as a future direction in their extensive survey on web crawling @xcite and we tried to make a step forward in this direction .
to the best of our knowledge , the problem of timely and holistic crawling of ephemeral new content is novel . in this paper , we introduce the notion of ephemeral new pages , i.e. , pages that exhibit the user interest pattern shown on figure [ fig : yabar ] , and emphasize the importance of this problem by showing that a significant fraction of the new pages that are appearing on the web are ephemeral .
we formalized this problem by proposing to optimize a new quality metric , which measures the ability of an algorithm to solve this specific problem .
we showed that most of the ephemeral new content can be found at a relatively small set of content sources and suggested an algorithm for finding such a set .
then , we proposed a practical algorithm , which periodically recrawls content sources and crawls newly created pages linked from them as a solution of this problem .
this algorithm estimates the quality of content sources using user feedback . finally , we compared this algorithm with other crawling strategies on real - world data and demonstrated that the suggested algorithm shows the best results according to our metric .
our theoretical and experimental analysis aims at giving a better insight into the current challenges in crawling the web . in this paper
, we predict the expected profit of a new page using two features : the time when this page was discovered by a crawler , and the content source where a link to this page was found .
the natural next step , which we leave for future work , is to predict this profit using more features , e.g. , give a higher priority to pages having an anchor text related to the current trends in user queries like in @xcite , or to the pages with more incoming links .
also , url tokens and its hyperlink context ( anchor text , surrounding text , etc . ) may be useful for such prediction .
this will help to prioritize new pages with seemingly higher quality found on the same content source at the same time .
brewington , b.e . ,
cybenko , g. : how dynamic is the web ?
computer networks , vol .
33(16 ) , 257276 , 2000 .
brewington , b.e . ,
cybenko , g. : keeping up with the changing web .
computer , vol .
33(5 ) , 5258 , 2000 . cho , j. , garcia - molina , h. : effective page refresh policies for web crawlers .
acm transactions on database systems , vol .
28(4 ) , 2003 .
kumar , r. , lang , k. , marlow , c. , tomkins , a. : efficient discovery of authoritative resources . in data engineering , 2008 .
liu , m. , cai , r. , zhang , m. , zhang , l. : user browsing behavior - driven web crawling . in proc .
cikm conference , 2011 . | nowadays , more and more people use the web as their primary source of up - to - date information . in this context , fast crawling and indexing of newly created web pages has become crucial for search engines , especially because user traffic to a significant fraction of these new pages ( like news , blog and forum posts ) grows really quickly right after they appear , but lasts only for several days . in this paper , we study the problem of timely finding and crawling of such _ ephemeral _ new pages ( in terms of user interest ) .
traditional crawling policies do not give any particular priority to such pages and may thus crawl them not quickly enough , and even crawl already obsolete content .
we thus propose a new metric , well thought out for this task , which takes into account the decrease of user interest for ephemeral pages over time .
we show that most ephemeral new pages can be found at a relatively small set of content sources and present a procedure for finding such a set .
our idea is to periodically recrawl content sources and crawl newly created pages linked from them , focusing on high - quality ( in terms of user interest ) content .
one of the main difficulties here is to divide resources between these two activities in an efficient way .
we find the adaptive balance between crawls and recrawls by maximizing the proposed metric .
further , we incorporate search engine click logs to give our crawler an insight about the current user demands . efficiency of our approach is finally demonstrated experimentally on real - world data . |
since the discovery of mssbauer effect spectroscopy more than half a century ago @xcite , superconductivity has been one of the states that has been investigated by this technique @xcite .
although mssbauer spectroscopy is widely accepted as one of the most sensitive techniques in terms of energy resolution , it has not contributed significant insight to studies of conventional superconductors . after the discovery of cuprate high temperature superconductors ( htsc ) , mssbauer spectroscopy was widely used for studies of these materials and reports of observation of some anomalies in the spectral parameters in the vicinity of the superconducting critical temperature ( t@xmath4 )
@xcite were published . due to the absence of commonly available mssbauer nuclides in the cuprates ,
most studies were accomplished either by partial substitution of copper atoms by @xmath0fe and/or @xmath5sn , or by using resonant isotopes of the rare earth metals , like @xmath6eu , which increases the degree of difficulty of the measurements and reduces the clarity of the results @xcite .
recently , the discovery of iron - based superconductors , that naturally contain the common mssbauer nuclide , @xmath0fe , has triggered intense mssbauer studies of these superconductors @xcite .
superconductivity in iron - pnictides is usually achieved by doping a magnetic parent compound with electrons or holes , or by application of chemical or physical pressure , and thereby , suppressing the magnetic order , suggesting that superconductivity and magnetism are closely related in this system . although there are some studies on iron - based superconductors , which state that mssbauer spectral parameters show anomalies near t@xmath4 , in these materials it is hard to attribute the variation of the hyperfine parameters observed by mssbauer spectroscopy to purely magnetic or purely superconducting origins @xcite . in order to study possible variations of hyperfine parameters caused by the transition from the normal to the superconducting state in a conventional superconductor
, we revisited lutetium - iron - silicide , lu@xmath1fe@xmath2si@xmath3 .
lu@xmath1fe@xmath2si@xmath3 is a stoichiometric , fe - containing , superconductor with relatively high t@xmath7 6 k @xcite . in a previous mssbauer study @xcite , the non - magnetic nature of fe
was already confirmed .
hence , lu@xmath1fe@xmath2si@xmath3 can be considered as an ideal compound to investigate the variation of hyperfine parameters caused only by the superconducting transition without any complications associated by the absence of mssbauer nucleus and/or the presence of magnetism . to the best of our knowledge no detailed , temperature dependent , @xmath0fe mssbauer spectroscopy measurements were performed on this material so far , and our goal is to shed some light on the applicability of mssbauer spectroscopy for studies of conventional , albeit multigap superconductors @xcite .
polycrystalline samples of lu@xmath1fe@xmath2si@xmath3 were prepared by arc melting constituent elements with the nominal composition of lu@xmath8fe@xmath9si@xmath10 ( corresponding to lu@xmath1fe@xmath2si@xmath3 + fe@xmath11si@xmath12 ) in zr - gettered ar atmosphere .
extra iron was added to suppress the formation of a lu@xmath1fesi@xmath13 second phase and extra silicon was added to compensate apparent loss during the arc melting . to ensure the homogeneity of the sample
, the arc melting was repeated iteratively after flipping of the melted and resolidified ingot , for more than ten times .
the weight loss was about 0.26% .
the arc - melted ingot was then sealed in an amorphous silica tube , under a partial pressure of argon , and annealed at 1050 for 12 days .
powder x - ray diffraction ( xrd ) was performed using a rigaku miniflex diffractometer with cu k@xmath14 radiation at room temperature ( rt ) .
the powder x - ray spectra of the samples were refined by rietveld analysis using the expgui software @xcite .
magnetic measurements were performed using a quantum design magnetic property measurement system squid magnetometer , specific heat capacity was measured in a quantum design physical property measurement system .
mssbauer spectroscopy measurements were performed using a see co. conventional constant acceleration type spectrometer in transmission geometry with an @xmath0co(rh ) source , which had an initial intensity 50 mci , kept at rt .
the absorber was prepared in a powder form ( 10 mg of natural fe/@xmath15 ) by grinding of approximately 175 mg piece of the arc - melted and annealed button .
the absorber holder comprised two nested white delrin cups .
the powder was placed uniformly on the bottom of the larger cup and was held in place by a smaller cup .
the absorber holder was locked in a thermal contact with a copper block with a temperature sensor and a heater , and aligned with the @xmath16 - source and detector .
the absorber was cooled to a desired temperature using janis model shi-850 - 5 closed cycle refrigerator ( with vibrations damping ) that has long - term temperature stability better than 0.1 k at low temperature .
the driver velocity was calibrated by @xmath14-fe foil and all isomer shifts ( is ) are quoted relative to the @xmath14-fe foil at rt . at first , spectra with maximum velocity 6 mm / s and 3 mm / s were both measured at rt to check that no iron - containing impurity can be seen in the mssbauer spectra . then , three rounds of measurements , progressively focusing in on temperatures near t@xmath17 k , were carried out : collecting 24 h with maximum velocity 2 mm / s from 4.3 k to 293.8 k ( s1 ) ; collecting 48 h with maximum velocity 3 mm / s from 4.4 k to 10 k ( s2 ) ; collecting 48 h with maximum velocity 3 mm / s from 4.7 k to 6.4 k ( s3 ) .
all the mssbauer spectra were fitted by the commercial software package mosswinn @xcite , in which the standard error of parameters can be estimated either by calculating and inverting the curvature matrix of the @xmath18 with respect to the fit parameters , or by the monte carlo method .
the standard error of is , quadroupole splitting ( qs ) and line width ( @xmath19 ) in this work were obtained from the curvature matrix , while the error the area under the spectra here were obtained by monte carlo method by iterating 100 times .
( color online ) rietveld refinement of the powder xrd spectrum of lu@xmath1fe@xmath2si@xmath3 .
measured ( black cross ) , calculated intensities ( red line ) and background ( green line ) and difference curve ( blue line ) are shown .
vertical bars at the bottom indicate the positions of the bragg reflections .
the peaks of unknown phase are marked by the blue arrows.,scaledwidth=50.0% ] ( color online ) expanded view of the background part of the rt lu@xmath1fe@xmath2si@xmath3 mssbauer spectrum in large velocity scale .
red arrows show the expected peaks positions for @xmath14-fe.the inset is the full view of the spectrum.,scaledwidth=50.0% ]
( color online ) ( a ) temperature dependent of zero field cooled magnetic susceptibility of lu@xmath1fe@xmath2si@xmath3 ( h = 10 oe ) ; ( b ) the temperature dependent low temperature specific heat capacity .
vertical dashed line marks t@xmath17 k.,scaledwidth=48.0% ] ( color online ) the specific heat capacity divided by temperature , c@xmath20/t , as a function of t@xmath21 for lu@xmath1fe@xmath2si@xmath3 .
the dashed line represents the linear fit to the data in the normal state as described in the text.,scaledwidth=50.0% ] the xrd pattern of the lu@xmath1fe@xmath2si@xmath3 polycrystalline sample is presented in fig .
1 . the majority of the peaks match to the tetragonal structure with the @xmath22 space group .
the rietveld refinement results in an estimate of @xmath23 wt.% @xmath14-fe impurity and a trace amount of unknown phase .
a 1.8 wt.% of @xmath14-fe corresponds to 6.7 % of fe atoms in the @xmath14-fe form in the ground sample , which , for un - enriched fe , is below the resolution of @xmath0fe mssbauer spectroscopy to detect @xmath14-fe . as can be seen in fig .
2 , there are no peaks associated with the @xmath14-fe or with the other impurity in the mssbauer spectrum , which means the unknown phase is either iron - free or , if contains iron , is below the resolution limit .
consequently , the mssbauer spectra can be analyzed as a single - phase ( lu@xmath1fe@xmath2si@xmath3 ) spectra .
the superconductivity of the sample is confirmed by the dc susceptibility measurement in a magnetic field of 10 oe . as shown in fig .
3 ( a ) , the susceptibility data shows diamagnetic signal below @xmath24 k. the transition is sharp with a width of less than 0.4 k. fig .
3 ( b ) shows the temperature dependent specific heat capacity ( c@xmath20 ) . a sudden jump caused by superconducting transition
can be observed below 6.3 k on cooling , the transition temperature is close to that obtained from the susceptibility data .
based on these two thermodynamic measurements we take t@xmath4 = [email protected] k. c@xmath20/t is also plotted as a function of t@xmath21 in fig .
a linear fit above the superconducting transition yields the values of @xmath26 ( @xmath26 t is the electronic contribution to specific heat capacity ) and @xmath27 ( @xmath27@xmath28 represents the phonon contribution ) of @xmath26 = 24.6(2 ) mj / mol k@xmath21 , @xmath27 = 0.287(2 ) mj / mol k@xmath29 , which are very close to the previously reported values @xcite . from the @xmath30 value
, we can estimate the value of the debye temperature ( @xmath31@xmath32 ) using the relation : @xmath31@xmath32 = ( 12@xmath33@xmath29@xmath34@xmath35k@xmath36/5@xmath30)@xmath37 , where @xmath34 is avogadro s number , @xmath35 is the number of atoms per formula unit , and k@xmath36 is the boltzmann s constant .
we further obtained @xmath31@xmath32 is 408 k. the value of the normalized specific - heat jump at t@xmath4 , @xmath38c/@xmath26t@xmath4 , is @xmath39 , a value that is smaller than the bcs value of 1.43 , but consistent with the previously reported value 1.05 @xcite . from the above characterizations
, we can conclude that our sample is a bulk superconductor and can be considered as single phase for mssbauer measurements and analysis . in the lu@xmath1fe@xmath2si@xmath3 crystal structure , there are two , nonequivalent , fe positions , fe@xmath40 and fe@xmath41 of 1 : 2 occupation .
fe@xmath40 atoms are located at the 4d sites , which form 1d chains along the @xmath42 axis .
fe@xmath41 atoms are located at 8h sites , which form squares with planes perpendicular to the @xmath42 axis . in each fe position , the fe atom is located in a polyhedron formed by si atoms .
fe@xmath40 has four si atoms at a distance of 2.31 which form an irregular tetrahedron and two si atoms at a distance of 2.54 with the nearest fe - fe distance is 2.67 .
the fe@xmath41 has four si atoms at a distance of 2.34 which are in the same face and form a quadrangle . on each side of the face
, there are two si atoms with the 2.35 fe - si distances .
the nearest fe - fe distance for fe@xmath41 is 2.71 .
the anisotropic environments of the fe atoms ensure nonzero electric field gradient ( efg ) tensor at both sites .
hence , in analyzing the data , two doublets are expected .
all the @xmath0fe mssbauer spectra of lu@xmath1fe@xmath2si@xmath3 over the whole observed temperature range share similar spectral shapes with a clear quadrupole splitting as shown in fig . 5 ( a ) .
a very small asymmetry and small shoulders were observed , which suggest that at least two subspectra are needed to resolve the spectra .
however , due to the small splitting and consequently poor resolution of the spectra , the data can be analyzed with more than one set of parameters . in the previous mssbauer studies of r@xmath1fe@xmath2si@xmath3 ( r = rare earth ) , both ( i ) one doublet , and ( ii )
two doublets with fixed area ratio were employed to fit the spectra @xcite . in our approach to the fitting , two doublets without any restriction are used . in model 1 ,
the two subspectra have close values of qs but obviously different values of is , which is similar to the reported sc@xmath1fe@xmath2si@xmath3 fit result @xcite . in model 2 ,
the two subspectra have similar values of is , but distinct qs values
. two sets of parameters yielding fits of acceptable quality can be obtained , the corresponding parameters of the rt spectrum fits are listed in the table 1 .
the rt spectrum fitted using our two models is shown in fig . 5 ( b ) and ( c ) .
the relative area of two subspectra in model 2 are closer to the theoretical value 1 : 2 and @xmath18 values are closer to 1 , so we have chosen the model 2 to fit all the collected spectra . as an example , the fit using model 2 of the spectrum measured at 4.4 k from the s2 set is shown in fig . 5 ( d ) and the corresponding parameters are listed in table 1 .
( color online ) @xmath0fe mssbauer spectra of lu@xmath1fe@xmath2si@xmath3 .
( a)the spectrum at rt ; the results of the fits using model 1 and model 2 are shown in ( b ) and ( c ) , respectively ; ( d ) is the spectra collected at 4.4 k and fitted using the model 2 .
the model 1 and model 2 are described in detail in the text.,scaledwidth=48.0% ] .[tab : table4 ] hyperfine parameters obtained by fitting using different models as discussed in the text .
is is the isomer shift , qs is the quadrupole splitting and i is the relative area of the two subspectra . [ cols="^,^,^,^,^,^,^ " , ] fig .
6 summarizes the variation of hyperfine parameters of the spectra with temperature .
the plots in the right column present an expanded view of the low temperature range .
the is of lu@xmath1fe@xmath2si@xmath3 plotted as a function of temperature is shown in fig .
6 ( a ) and ( b ) .
the is values of the 4d and 8h sites at 294 k are 0.225(2 ) mm / s and 0.227(1 ) mm / s , respectively , which are slightly larger than the typical values for iron - silicon compounds in which fe carries no moment @xcite , but are about 0.2 mm / s smaller than that in the iron - pnictide compounds @xcite .
the temperature dependencies of the is corresponding to the two sites are very similar , and no anomalies can be observed around t@xmath4 ( fig .
6(b ) ) . the is values obtained from the fits includes contributions from both the chemical shift and the second - order doppler shift , which is known to increase convexly upon decreasing temperature , due to gradual depopulation of the excited phonon states .
however , it should be constant at low temperature , because of the quantum mechanical zero - point motion .
the chemical shift should not depend on temperature .
the main contribution to this variation is from the second - order doppler shift , which is usually described by debye model : @xmath43 where c is the velocity of light , m is the mass of the @xmath0fe nucleus , and @xmath44 is is the temperature - independent part , i.e. the chemical shift .
a fit with eq .
( 1 ) to the data of s1 shown in fig . 6 ( a ) yields @xmath31@xmath32 = 517(18 )
k and 545(16 ) k for 4d and 8h sites , respectively .
( color online ) ( a ) temperature dependence of the spectral area for s1 set of measurements on a semi - log scale .
the solid line is a fit to eq.(2 ) , as explained in the text .
( b ) normalized to the values at @xmath45 k , temperature dependent spectral area data for all three sets of measurements .
inset : enlarged low temperature part.,scaledwidth=50.0% ] the quadrupole splittings in lu@xmath1fe@xmath2si@xmath3 at the 4d and 8h sites are 0.50(1 ) mm / s and 0.19(1 ) mm / s at 294 k , respectively .
the magnitude of the qs is proportional to the @xmath46 component v@xmath47 of the efg tensor , which is composed of two contributions : ( v@xmath47)@xmath48 , from the ligand charges around the mssbauer nucleus , and ( v@xmath47)@xmath49 , from the valence electrons of mssbauer nucleus .
usually , ( v@xmath47)@xmath48 is small and weakly dependent on temperature , whereas ( v@xmath47)@xmath49 is strongly temperature dependent . as can be seen from fig .
the qs of two sites are both almost temperature independent , which indicates the qs is mainly determined by the contribution from the ligand charge distribution around the @xmath0fe sites . in order to get a better understanding of the electronic origin of efg at the fe sites in lu@xmath1fe@xmath2si@xmath3 , a first - principles calculation was performed using the full - potential linearized augmented plane wave method as embodied in the wien2k @xcite .
the generalized gradient approximation ( gga ) suggested by perdew , burke , and ernzerhof ( pbe gga ) @xcite was employed for the exchange - correlation effects .
once the electron densities are calculated self - consistently and with high accuracy , the efg tensor can be obtained from an integral over the non - spherical charge density .
the principal component v@xmath47 for the 4d site is 2.25@xmath5010@xmath51v / m@xmath21 and asymmetry parameter @xmath52 = 0.457 ; the v@xmath47 for the 8h site is -1.16@xmath5010@xmath51v / m@xmath21 and @xmath53 = 0.307 , which qualitatively agree with experimental results .
it is found that the @xmath54-@xmath54 and @xmath55-@xmath55 interactions mainly contribute to the efg of lu@xmath1fe@xmath2si@xmath3 and @xmath54- electrons is play a significant role for states far from the fermi energy whereas the @xmath55-@xmath55 interaction dominates around the fermi energy , which is similar to what is found for the iron - pnictides superconductors @xcite .
the relative areas of the subspectra are determined by the proportion of the fe atoms on different lattice sites . as mentioned above , for lu@xmath1fe@xmath2si@xmath3
, the theoretical relative area of the two subspectra representing iron atoms at the 4d and 8h sites should be 33.3% and 66.7% . at rt
the relative areas are close to expected value . however , as shown in fig .
6 ( e ) as temperature decreases , the relative area values deviate from theoretical values gradually , with the relative area for the 4d site becoming even larger than that for 8h site at low temperature .
this phenomenon may be related to the variation of the linewidth , @xmath19 , for the doublets corresponding to the two sites with reducing temperature . as shown in fig . 6 ( g ) ,
the @xmath19 of 4d site increases slightly at lower temperature , but the @xmath19 of 8h site is almost constant during the whole temperature range .
the slightly increase of @xmath19 of 4d site suggests the existence of instability at this lattice site at low temperature . in mssbauer studies of superconductivity ,
the variation of the total spectral area has also been the focus of discussion .
there are number of reports showing a decrease of the spectral area near t@xmath4 due to the softening of lattice with the opening of superconducting gap in cuprates @xcite but to the best of our knowledge only one report on iron - pnictide superconductors @xcite .
there are two kinds of behavior reported : a rapid decrease near t@xmath4 @xcite ; and a pit - like decrease either around t@xmath4 or at higher temperature , serving as a precursor to t@xmath4 @xcite . at the same time
, some publications noted a poor reproducibility of those observations in the cuprates @xcite . in our measurements , as can be seen in the fig .
7 , for the s1 , we did not observe any abnormal variation around 6.1 k. to make sure we did nt miss any minor variation around t@xmath4 , we remeasured the mssbauer spectra of the same sample between 4.4 k to 10 k with higher density of the data points around 6.1 k for 48 h. surprisingly , for data set s2 , there is a sharp , 6% , decrease around 5.5 k. nevertheless , when we repeated the same measurement again , data set s3 , this phenomenon disappeared .
the sharp spectral area change seen in data set s2 is , in our opinion , most likely an artifact .
the feature in data set s2 occurs resolvably below t@xmath4 = 6.1 k with the very sharp and rather large jump occuring between spectra taken at 5.3 k and 5.7 k ( with spectra at 5.7 k , 6.0 k , and 6.3 k all having normalized areas near 1.0 ) .
if this feature were associated with t@xmath4 , it should have occured either between 5.7 k and 6.0 k or between 6.0 k and 6.3 k. in addition , there are no corresponding anomalies in data set s2 s isomer shift , quadrupole splitting or relative areas ( as shown in fig . 6 ) . a simple explanation for such behavior could be , among others , some mechanical shift / rearrangement of the powder composing the absorber .
this observation gives a warning that even in the measurement of iron - containing stoichiometric material , irreproducibilities / artifacts might exist and should be addressed appropriately .
finally , we also fitted the temperature dependence of area under the two doublets line of s1 measurements with debye model : @xmath56,\ ] ] where @xmath57 is the recoilless fraction , which is proportional to the area for thin sample and e@xmath58 is the @xmath16-ray energy .
this expression also allows to estimate the value of @xmath59 .
we obtained the @xmath59 = 385(8 ) k which is very close to the 408 k obtained from the analysis of the low temperature specific heat capacity data and about 140 k less than the value estimated by temperature dependence of is .
a similar difference was found earlier in studies of e.g. fese@xmath60te@xmath60 and @xmath0fe- doped yba@xmath1cu@xmath2o@xmath61 compounds @xcite .
this discrepancy may be explained by the fact the area reflects the average mean - square displacements , while is related to the mean - square velocity of the mssbauer atom .
both quantities may respond in different ways to the lattice anharmonicities .
in summary , we performed detailed @xmath0fe mssbauer measurements on lu@xmath1fe@xmath2si@xmath3 in the temperature range of 4.4 k to rt .
the contributions from two fe crystallographic sites can be well distinguished by mssbauer spectra .
the main contribution of efg in this compound comes from the lattice anisotropy and the first principles calculations yield the values of efg that qualitatively agree with the experiment .
the debye temperature was estimated by the temperature dependence of specific heat capacity , spectral area and is .
the @xmath59 obtained from temperature dependence of spectral area and heat capacity are very similar , but about 140 k smaller than the value estimated by the is variation with temperature .
additionally , we did nt observe any obvious , abnormal variation of hyperfine parameters around t@xmath4 .
two possibilities could lead to this result : the opening of the superconducting gap does nt bring variation of the environment at fe site at all ; the t@xmath4 of this system is too low , and mssbauer spectroscopy is not sensitive enough to detect the minute change . x. m. was supported in part by the china scholarship council .
the authors ( h. p. and f. l. ) gratefully acknowledge the financial support from the national natural science foundation of china under grant no .
work at the ames laboratory ( x. m. , s. r. , p. c. c. and s. l. b. ) was supported by the us department of energy , basic energy sciences , division of materials sciences and engineering under contract no .
de - ac02 - 07ch11358 .
34ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @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 * * , ( ) | with the advent of fe - as based superconductivity it has become important to study how superconductivity manifests itself in details of @xmath0fe mssbauer spectroscopy of conventional , fe - bearing superconductors . to this end
, the iron - based superconductor lu@xmath1fe@xmath2si@xmath3 has been studied by @xmath0fe mssbauer spectroscopy over the temperature range from 4.4 k to room temperature with particular attention to the region close to the superconducting transition temperature ( t@xmath4 = 6.1 k ) .
consistent with the two crystallographic sites for fe in this structure , the observed spectra appear to have a pattern consisting of two doublets over the whole temperature range .
the value of debye temperature was estimated from temperature dependence of the isomer shift and the total spectral area and compared with the specific heat capacity data .
neither abnormal behavior of the hyperfine parameters at or near t@xmath4 , nor phonon softening were observed .
pacs numbers : : 74.70.dd , 76.80.+y |
because the dynamics of financial markets are of great importance in economics and econophysics @xcite , the dynamics of both stock price and trading volume have been studied for decades as a prerequisite to developing effective investment strategies .
econophysics research has found that the distribution of stock price returns exhibits power - law tails and that the price volatility time series has long - term power - law correlations @xcite . to better understand these scaling features and correlations , yamasaki _ et al .
_ @xcite and wang _ et al . _
@xcite studied the behavior of price return intervals @xmath0 between volatilities occurring above a given threshold @xmath1 . for both daily and intraday financial records
, they found that ( i ) the distribution of the scaled price interval @xmath8 can be approximated by a stretched exponential function , and ( ii ) the sequence of the price return intervals has a long term memory related to the original volatility sequence .
the scaling and memory properties of financial records are similar to those found in climate and earthquake data @xcite .
a feature of the recent history of the stock market has been large price movements associated with high volume . in the black monday stock market crash of 1987 ,
the dow jones industrials average ( djia ) plummeted 508 points , losing 22.6 percent of its value in one day , which led to the pathological situation in which the bid price for a stock actually exceeded the ask price . in this financial crash
approximately @xmath9 shares traded , a one - day trading volume three times that of the entire week previous .
understanding the precise relationship between price and volume fluctuations has thus been a topic of great interest in recent research @xcite .
trading volume data in itself contains much information about market dynamics , e.g. , the distribution of the daily traded volume displays power - law tails with an exponent within the lvy stable domain @xcite .
recently , ren and zhou @xcite studied the intraday database of two composite indices and 20 individual indices in the chinese stock markets .
they found that the intraday volume recurrence intervals show a power - law scaling , short - term correlations and long - term correlations in each stock index . in this study
we analyze u.s .
stock market data over a range broad enough to allow us to identify how several financial factors significantly affect scaling properties .
we study the daily trading volume volatility return intervals @xmath0 between two successive volume volatilities above a certain threshold @xmath1 , and find a range of power - law distributions broader than that found earlier in price volatility return intervals @xcite .
we find a unique scaling of the probability density function ( pdf ) @xmath2 for different thresholds @xmath1 .
we also perform a detailed analysis of the relation between volume volatility return intervals and four financial stock factors : ( i ) stock lifetime , ( ii ) market capitalization , ( iii ) average trading volume , and ( iv ) average trading value .
we find systematically different power - law exponents for @xmath2 when binning stocks according to these four financial factors .
similar to that found for the chinese market @xcite , we find that in the u.s .
stock market the conditional probability distribution , @xmath6 for @xmath0 following a certain interval @xmath7 , demonstrates that volume return intervals are short - term correlated .
we also find that the daily volume volatility shows a stronger long - term correlation for sequences of longer lifetime but no clear changes in long - term correlations for different stock size factors such as capitalization , volume , and trading value .
in order to obtain a sufficiently long time series , we analyze the daily trading volume volatility of 17,197 stocks listed in the u.s .
stock market for at least 350 days .
we obtain our data from the center for research in security prices ( crsp ) us stock database , which lists the daily prices of all listed nyse , amex , and nasdaq common stocks , along with basic market indices .
the period we study extends from 1 january 1989 to 31 december 2008 , a total of 5042 trading days .
for a stock trading volume time series , in a manner similar to stock price analysis @xcite , we define two basic measures : volume return @xmath10 and volume volatility @xmath11 .
the volume return @xmath10 is defined as the logarithmic change in the successive daily trading volume for each stock , @xmath12 where @xmath13 is the daily trading volume at time @xmath14 .
we define volume volatility to be the absolute value of the volume return . in order to compare different stocks , we determine the volume volatility @xmath15 by dividing the absolute returns @xmath16 by their standard deviation , @xmath17 where @xmath18 is the time average for each stock .
the threshold @xmath1 is thus measured in units of standard deviation of absolute volume return @xmath16 . for a volume volatility time series
, we collect the time intervals @xmath0 between consecutive volatilities
@xmath15 above a chosen threshold @xmath1 and construct a new time series of volume return intervals @xmath19 . fig . [ fig1](a ) shows the dependence of @xmath2 on @xmath1 , where @xmath2 is the pdf of the volume volatility return interval time series @xmath19 .
obviously , @xmath2 decays more slowly for large @xmath1 than for small @xmath1 . for large @xmath1
, @xmath2 has a higher probability of having large interval values because extreme events are rare in a high threshold series .
we next determine whether there is there any scaling in the distribution by plotting the pdfs of the volume return intervals @xmath2 , scaled with the mean volume return interval @xmath20 , for different thresholds in fig .
[ fig1](b ) .
we can see that all five threshold values @xmath1 curves ( full symbols ) callapse onto a single curve , suggesting the existence of a scaling relation , @xmath21 as the threshold @xmath1 increases , the curve ( rare events ) tends to be truncated due to the limited size of the dataset .
the tails of the scaling function can be approximated by a power - law function as shown by the dashed line in fig .
[ fig1](b ) , @xmath22 where the tail exponent is @xmath23 .
the exponent of the scaled pdfs for @xmath24 is @xmath25 by the least square method , which is the same as the unscaled pdf exponent @xmath25 as shown in fig .
[ fig1](a ) .
the power - law exponents for intraday volume recurrence intervals of several chinese stock indices are from @xmath26 to @xmath27 @xcite .
our exponents @xmath23 are larger than those in the chinese stock markets
. this might be due to differing definitions of volume volatility . in ref .
@xcite , the volume volatility is defined as intraday volume divided by the average volume at one specific minute of the trading day averaged over all trading days . here
we define the volume volatility to be the logarithmic change in the successive daily volumes [ eqs .
[ return.eq ] and [ volatility.eq ] ] . for comparison , and using the same approach , fig .
[ fig1](c ) and fig .
[ fig1](d ) show the analogous results for price volatilities ( see also the studies in refs .
note that it is not easy to distinguish between a stretched exponential and a power - law when studying price volatilities @xcite , i.e. , the power - law range is small and a stretched exponential could also provide a good fit .
in contrast , the pdfs of the volume volatility return intervals display a wide range of power - law tails , which differs from the stretched exponential tail apparent in the price return intervals @xcite . our results for volume volatility may suggest that @xmath2 for price volatility is also a power - law , but this could not be verified because the range of the observed power - law regime [ see figs .
[ fig1](c ) and [ fig1](d ) ] is more limited than the broad range of scales seen in the volume volatility [ figs .
[ fig1](a ) and [ fig1](b ) ] .
the difference between the power - law and stretched exponential behavior of @xmath2 may be related to the existence or non - existence respectively of non - linearity represented in the multifractality of the time series . when non - linear correlations appear in a time record , bugachev _ et al . _
@xcite showed that @xmath2 is a power - law . on the other hand , when non - linear correlations do not exist and
only linear correlation exists , bunde _
_ @xcite found stretched exponential behavior .
a comparison with the shuffled records allows us to see how the empirical records differ from randomized records .
we shuffle the volume volatility time series to make a new uncorrelated sequence of volatility , and then collect the time intervals above a given threshold @xmath1 to obtain synthetic random control records .
the curve that fits the shuffled records [ the open symbols in fig .
[ fig1](b ) ] is an exponential function , @xmath28 , and forms a possion distribution .
a poisson distribution indicates no correlation in shuffled volatility data , but the empirical records suggest strong correlations in the volatility .
we study the relation between the scaled pdfs @xmath29 as a function of @xmath8 for four financial factors : ( a ) stock lifetime , ( b ) market capitalization , ( c ) mean volume , and ( d ) mean trading value for threshold @xmath30 . for higher @xmath1 values , we do not have sufficient data for conclusive results @xcite . in fig .
[ fig2 ] , we plot the scaled pdfs for these four factors .
the volume return intervals characterize the distribution of large volume movements . a high probability of having a large volume return interval @xmath0 suggests a correlation in volume volatility ,
because small volatilities are followed by small volatilities and the time interval between the two large volatilities becomes relatively longer than those of random records . in order to charaterize how these four factors affect the distribution of volume return intervals
, we divide all stocks into four bins for each factor . in fig .
[ fig2](a ) , the probability that @xmath0 will be large is greater in the bin with 15@xmath3120 year old stocks ( triangles ) than in the bins of younger stock .
this indicates that small volatilities ( below the threshold ) tend to follow small volatilities and that the time intervals between large volatilities in the bin of 15@xmath3120 year - old stocks are larger than the time intervals in the bin of 5 years old stocks ( dots ) .
this also suggests that the volume volatility time records of older stocks are more auto - correlated than those of younger stocks .
the decaying parameters represented by the power - law exponents are quite different : @xmath32 for the shortest lifetime bin and @xmath33 for the longest lifetime bin
. this significant difference might be caused by differences in autocorrelation in these series . in figs .
[ fig2](b ) , [ fig2](c ) , and [ fig2](d ) , we show a similar tendencies for stock bins with different capitalizations , mean volumes , and mean trading values . trading value is defined as stock price multiplied by transaction volume . for each stock ,
we designate the lifetime average of capitalization , volume , and trading value as performance indices .
for example , the power - law exponents of the pdfs , @xmath29 , increase as the capitalization becomes larger [ see fig . [ fig2](b ) ] . to clarify the picture
, we divide all stocks into different subsets and study the behavior of the power - law exponent @xmath23 with regard to these four factors . in fig .
[ fig3](a ) , stocks are sorted into 10 subsets , from 508 days ( 2 years ) to 5080 days ( 10 years ) .
we fit the power - law tails of the volume return intervals for each subset and plot the exponent @xmath23 versus the lifetime of the stocks . in fig .
[ fig3](a ) , we can observe a systematic trend with stock lifetime .
it is seen that a smaller exponent @xmath23 which indicates a stronger correlation in older stock subsets .
similarly , we sort the stocks by capitalization , mean volume , and mean trading value , as shown in figs .
[ fig3](b ) , [ fig3](c ) , and [ fig3](d ) .
it is seen that @xmath23 decreases with increasing of all these three factors but seem to become constant for large values of capitalizations , mean volumes and mean trading values .
since all factors similarly affect the scaling of the pdf , @xmath29 , we now determine how much these factors are correlated .
to study the relations between different stock bins , we plot the relation between trading value versus capitalization , mean volume versus capitalization , and mean trading value versus mean volume for all the stocks shown in fig .
we see that larger capitalization stocks tend to have a larger trading volume and a larger trading value , which is consistent with figs . [ fig1](b ) , [ fig1](c ) , and [ fig1](d ) .
the correlation coefficients between trading value and capitalization , mean volume and capitalization , and trading value and volume are 0.62 , and 0.55 , and 0.78 , respectively .
the correlation coefficients are high because these capitalization , volume , and trading value factors are all affected by firm size .
our analyses do not , however , show a significant relationship between stock lifetime and its trading value , capitalization , and mean volume , and the correlation coefficients are all @xmath34 .
we characterize a sequence of volume return intervals in terms of the autocorrelations in the time series .
if the volume return intervals series are uncorrelated and independent of each other , their sequences can be determined only by the probability distribution . on the other hand ,
if the series is auto - correlated , the preceding value will have a memory effect on the values following in the sequence of volume volatility return intervals . in order to investigate whether short - term memory is present , we study the conditional pdf , @xmath6 , which is the probability of finding a volume return interval @xmath0 immediately after an interval of size @xmath7 . in records without memory ,
@xmath6 should be identical to @xmath2 and independent of @xmath7 .
otherwise , @xmath6 should depend on @xmath7 . because the statistics for @xmath7 of a single stock are of poor quality
, we study @xmath6 for a range of @xmath35 .
the entire dataset is partitioned into eight equal - sized subsets , @xmath36 , with intervals of increasing size @xmath37 .
figure [ fig5 ] shows the pdfs @xmath6 for @xmath38 , i.e. , small interval size @xmath39 and @xmath40 large interval size @xmath41 for different @xmath1 .
the probability of finding large @xmath8 is larger in @xmath40 ( open symbols ) than in @xmath38 ( full symbols ) , while the probability of finding small @xmath8 is larger in @xmath38 than that in @xmath40 .
thus large @xmath7 tends to be followed by large @xmath0 , and vice versa , which indicates short - term memory in the volume return intervals sequence .
moreover , note that @xmath6 in the same subset for different thresholds @xmath1 fall onto a single curve , which indicates the existence of a unique scaling for the conditional pdfs as well .
similar results were found for the volume volatility of the chinese markets @xcite and for price volatilities @xcite .
in previous studies , the price volatility series was shown to have long - term correlations . using a similar approach
, we test whether the volume volatility sequence also possesses long - term correlations . to answer this question
, we employ the detrended fluctuation analysis ( dfa ) method @xcite to further reveal memory effects in the volume volatility series . using the dfa method ,
we divide an integrated time series into boxes of equal length @xmath42 and fit a least squares line in each box .
next we compute the root - mean - square fluctuation @xmath43 of the detrended time series within a window of @xmath42 points and determine the correlation exponent @xmath44 from the scaling function @xmath45 , where @xmath46 $ ] .
the correlation exponent @xmath44 characterizes the autocorrelation in the sequence .
the time series has a long - term memory and a positive correlation if the exponent factor @xmath47 , indicating that large values tend to follow large values and small values tend to follow small values .
the time series is uncorrelated if @xmath48 and anti - correlated if @xmath49 . using the dfa method
, we analyze the price volatility and volume volatility time series by plotting in bins the relation between correlation exponent @xmath44 and the four financial factors , including stock lifetime , market capitalization , mean trading volume , and mean trading value . all the price volatility and volume volatility correlation exponents are significantly larger than 0.5 , suggesting the presence of long - term memory in both price volatility sequences and volume volatility sequences . in all of the plots ,
the price volatility series shows a stronger long - term correlation than the volume volatility series .
moreover , as shown in fig .
[ fig6](a ) , @xmath44 on average increases for the stocks with a lifetime ranging from 350 days to 3800 days ( about 15 years ) , and then shows a slight decrease , suggesting that long - lasting stocks tend to have a persistent price and volume movement on large scales .
the increasing exponent @xmath44 indicates that the volume volatility of older stocks is more correlated than that of younger stocks .
this is consistent with the indication in fig .
[ fig2](a ) that the volume volatility of older stocks are more auto - correlated .
figures [ fig6](b ) , [ fig6](c ) , and [ fig6](d ) show that there is no systematic tendency relation between @xmath44 and market capitalization , trading volume , and trading value .
we have shown the scaling properties and memory effect of volume volatility return intervals in large stock records of the u.s .
the scaled distribution of volume volatility return intervals displays unique power - law tails for different thresholds @xmath1 .
we also find different power - law exponents @xmath23 of @xmath2 for the four essential stock factors : stock lifetime , market capitalization , average trading volume , and average trading value .
these different exponents may be related to long - term correlations in the interval series .
significantly , the daily volume volatility exhibits long - term correlations , similar to that found for price volatility .
the conditional probability , @xmath6 for @xmath0 following a certain interval @xmath7 , indicates that volume return intervals are short - term correlated .
g. w. schwert , j. financ .
* 44 * , 1115 ( 1989 ) ; k. chan , k. c. chan , and g. a. karolyi , rev .
financ . stud . * 4 * , 657 ( 1991 ) ; t. bollerslev , r. y. chou , and k. f. kroner , j. econometr . * 52 * , 5 ( 1992 ) ; a. r. gallant , p. e. rossi , and g. tauchen , rev .
financ . stud . * 5 * , 199 ( 1992 ) ; b. le baron , j. business * 65 * , 199 ( 1992 ) . | we study the daily trading volume volatility of 17,197 stocks in the u.s .
stock markets during the period 19892008 and analyze the time return intervals @xmath0 between volume volatilities above a given threshold @xmath1 . for different thresholds @xmath1 ,
the probability density function @xmath2 scales with mean interval @xmath3 as @xmath4 and the tails of the scaling function can be well approximated by a power - law @xmath5 .
we also study the relation between the form of the distribution function @xmath2 and several financial factors : stock lifetime , market capitalization , volume , and trading value . we find a systematic tendency of @xmath2 associated with these factors , suggesting a multi - scaling feature in the volume return intervals .
we analyze the conditional probability @xmath6 for @xmath0 following a certain interval @xmath7 , and find that @xmath6 depends on @xmath7 such that immediately following a short / long return interval a second short / long return interval tends to occur .
we also find indications that there is a long - term correlation in the daily volume volatility .
we compare our results to those found earlier for price volatility . |
recent development of quantum groups @xcite has resulted in the construction of @xmath0-analogs of quite a few quantum systems including a very popular system that is the harmonic oscillator .
a @xmath0-analog realization of harmonic oscillator has been given by macfarlane@xcite and biedenharn@xcite . in ref .
@xcite coherent analogs of glauber states have also been proposed .
it is of great interest to use these quantum analog formulations to study physical systems and there are situations where a description in terms of @xmath0- analog structures is apparently more natural .
the case of an atom being cooled through interaction with the radiation offers an interesting case study .
it has been shown @xcite recently that the atom can be cooled down to energies very close to it s zero point vibrational energy . at these energies
the quantum effects due to center of mass motion of the atom are expected to be very important@xcite .
blockley et .
al.@xcite have studied the collapses and revivals of population inversion in a single two - level atom interacting with a classical single mode travelling light field while constrained to move in a one - dimensional harmonic oscillator trap .
they also discuss the possibility of observing the collapses and revivals experimentally .
the interaction of the atom with the trap potential results in energy exchange between the internal degrees of freedom of the atom and the center of mass motion . as the atom slows down , the amount of energy exchanged in each step of the cooling process is not expected to be constant but a variable dependent on the initial energy state of the atomic center of mass .
presently we study the time evolution of a two - level atom in a @xmath0-deformed harmonic oscillator trap , in interaction with a single mode travelling light field . in a @xmath0-deformed harmonic oscillator trap ,
the energy spacing between the trap states , as seen by the atom , is a function of the initial vibrational state of the atom . as such , when there is a lot of energy associated with center of mass motion energy loss and gain occurs in large energy quanta .
however as the atom cools down , the energy exchange takes place in smaller units of energy . besides that the relative energy separation between successive states is determined by the value of the deformation parameter .
blockley et al@xcite have shown that in the lamb - dicke regime , when only the interaction between nearest neighbours is significant the model is similar to the jaynes - cummings model(jcm)@xcite with trap quanta playing a role similar to that of the light quanta in jcm . an atom with its center of mass in a coherent trap state initially , shows collapses and revivals of its population inversion .
our object is to investigate the system response for a trap potential expected to be closer to experimental situation in comparoson with a harmonic oscillator trap .
the energy spacings between energy eigen states of the @xmath0-analog harmonic oscillator are not constant but are a function of @xmath0-deformation . as such for a given system a suitable choice of the deformation parameter can take us from the classical limit where the trap states are closely spaced to the lamb- dicke regime with well spaced trap states . in our earlier attempts
@xcite to understand the physical nature of deformation in the context of pairing of nucleons , it has been found that the deformation amounts to simulating the nonlinearities of the problem or part of the interaction not included in the hamiltionian . with these results in mind
we expect the -atom in @xmath0 deformed oscillator trap model- to be a better description of the physical situation involved .
consider a single two level atom having atomic transition frequency@xmath3 in a quantized @xmath0-analog quantum harmonic oscillator trap(@xmath0-deformed harmonic oscillator trap ) interacting with a single mode travelling light field .
the creation and annihilation operators for the trap quanta satisfy the following quocommutation relations , @xmath4 here n is the number operator .
the operators @xmath5 and @xmath6 act in a hilbert space with basis vectors @xmath7 , @xmath8 given by , @xmath9_q!)^{\frac 12}}% \left| 0\right\rangle\ ] ] such that @xmath10 the vacuum state is @xmath11 we define here @xmath12_q$ ] as @xmath13_q=\frac{q^x - q^{-x}}{q - q^{-1}}\ ] ] and the @xmath0-analog factorial @xmath14_q!$ ] is recursively defined by [ 0]@xmath15_q!=1 $ ] and @xmath14_q!=[n]_q[n-1]_q!.$ ] it is easily verified that @xmath16_q^{\frac 12}\left| n+1\right\rangle \medskip\ ; \medskip\
a\left| n\right\rangle = [ n]_q^{\frac 12}\left| n-1\right\rangle\ ] ] and @xmath17 is not equal to @xmath18 analogous to the harmonic oscillator one may define the @xmath0-momentum and @xmath0-position coordinate @xmath19 the @xmath0-analog harmonic oscillator hamiltonian is given by @xmath20 with eigenvalues @xmath21_q+[n]_q).\ ] ] for @xmath22 we have for the energy spacing @xmath23 as the atomic state changes by absorption or emission of a light quantum there is a change in it s vibrational state in the trap as well .
we note that the trap states are not evenly spaced , the energy spacing being a function of deformation . besides that as we move up in the number of vibrational quanta in the states the spacing between successive states increases . for a given region of vibrational states
a suitable choice of deformation parameter should correspond to the lamb dicke regime .
the hamiltonian for the system that consists the atom vibrating in the trap and interacting with a classical single - mode light field of frequency @xmath24 is given by@xcite @xmath25 where @xmath26 , is the detuning parameter and @xmath27 is the rabi frequency of the system .
the operator @xmath28 stands for @xmath29.$ ] the parameter @xmath30 is a function of the ratio of the recoil energy of the atom @xmath31and the characteristic trap quantum energy @xmath32 in the limit @xmath33 here @xmath34 is the wave vector of the light field .
the second term in the hamiltonian refers to the energy associated with internal degrees of freedom of the atom , whereas the third term is the interaction of the atom with the light field . for nonzero values of deformation parameter @xmath0 , the kinetic energy and potential energy of the atom in the trap
are a function of deformation .
in addition the interaction of the atom with the light field is also deformation dependent .
the state of the system at a time t , @xmath35 is a solution of the time dependent schrodinger equation @xmath36 the vector , @xmath37 , corresponds to the atom being in its ground state with its center of mass in m - th trap state
. the label e stands for the excited state of the atom .
the probability amplitudes @xmath38and @xmath39 satisfy the following set of coupled equations @xmath40_q+[m]_q)-\delta \right ) + \frac 12\omega \sum\limits_ne_n(t)\left\langle g , m\right| \sigma ^{-}f^{*}\left| e , n\right\rangle\ ] ] @xmath41_q+[m]_q)+\delta \right ) + \frac 12\omega \sum\limits_ng_n(t)\left\langle e , m\right| f\sigma ^{+}\left| g , n\right\rangle\ ] ] in order to evaluate the matrix elements @xmath42 firstly we make use of a special case of baker - hausdorff theorem to rewrite the operator f as a product of operators i.e we use the equality @xmath43=e^{-(\left| \epsilon \right| ^2[a^{+},a])}e^{i\epsilon a^{+}}e^{i\epsilon a}.$ ] in doing so we have substituted for the commutation relation of the operators @xmath5 and @xmath6 the value of the commutator given in eq.([comm ] ) in the limit @xmath33 this approximation considerably simplifies the further evaluation of matrix elements of @xmath28 using the defining eqs.([comm],[vec ] ) for the operators a , a@xmath44 and the vectors @xmath45 the final expression for @xmath46 is given as , @xmath47_q^{\frac 12}!}{[n]_q^{\frac 12}!}% \sum\limits_{k=0}^m\frac{(\epsilon ) ^{2k}(-1)^k[n]_q!}{k!(n - m+k)![m - k]_q!}\ ] ] depending on the difference @xmath48 the matrix element @xmath49 can be real , imaginary , positive or negative . for the special case of @xmath50 and to first order in @xmath51 the matrix element reduces to @xmath52_{q}}.$ ] in the lamb dicke regime the model is analogous to @xmath0-analog of jaynes - cumming model@xcite with center of mass motion quanta playing a role similar to that of the quantized radiation field .
as pointed out in ref.@xcite , the effective rabi frequency for trapping model can be easily calculated in the lowest approximation . in a similar way , considering only those transitions involving single trap quantum exchange we can easily calculate the analog of the effective rabi frequency in the deformed harmonic oscillator trap . in case the driving field is tuned to the first vibrational sideband , @xmath53 all other transitions can be neglected if these are oscillating at sufficiently high frequencies . for @xmath54 ,
the rotating wave approximation can be used and the dynamical equations solved to give the eigenvalues and the effective rabi frequency . in ref.@xcite
the effective rabi frequency obtained is @xmath55 for the @xmath0-deformed oscillator trap the effective frequency , to first order in @xmath51 , is a function of deformation given by @xmath56 + \delta]^2 + \omega ^2\epsilon ^2[m+1]_q}.$ ]
the initial state center of mass motion of the atom is represented by @xmath0-analog glauber coherent state(gcs ) , while the atom is in the ground state .
the @xmath0 -analog of gcs is written as @xmath57_q!}}% \left| n\right\rangle\ ] ] with the @xmath0-exponential defined through @xmath58_q!}\ ] ] the complex parameter @xmath59 determines the average number of trap quanta associated with the state @xmath60 .
in figures ( 1 ) and ( 2 ) , we plot the population inversion as a function of redefined time parameter @xmath61 for @xmath62 with @xmath63 taking real values 0.0 , 0.002 , 0.003 , 0.004 , 0.006 , 0.01 and 0.1 respectively . besides that we also use the parameter values @xmath64 and @xmath65 .
for @xmath66 the results obtained by blockley et .
al.@xcite are reproduced , when a harmonic oscillator trap is used and mean number of trap quanta is @xmath67 . in the numerical calculation the maximum value of m in eq.([psi ] ) is restricted to @xmath68 . in fig.(1 ) for @xmath69 , the first revival is seen to peak earlier than the one for @xmath66 and a well defined collapse appears between the first and the second revival . for a very small value of @xmath70
the collapse and revival pattern is seen to become much more pronounced(well defined ) as compared to the undeformed case .
however as the deformation is increased to @xmath71 and @xmath72 , the pattern again becomes diffuse after the first revival .
the time interval after which the first revival occurs is seen to become shorter as @xmath73 is increased as seen in fig.(2 ) .
another interesting feature of the collapses and revivals is that as the deformation increases the collapse period of the system is seen to exhibit increasing level of coherent population trapping . for a large deformation ,
that in the context of this system is something like @xmath74 , the time dependence of popuation inversion is considerably washed away . for still larger values of @xmath73 for example @xmath75 in fig.(2 ) no time dependence of the system is seen anymore .
similar results are obtained when imaginary values of the parameter @xmath63 are used .
it is expected because for very small deformations the values of @xmath12_q$ ] for @xmath76 with @xmath63 real are very close to the values of @xmath12_q$ ] for @xmath77 imaginary so long as the modulus of @xmath63 is the same . to obtain further insight into the energy exchange between the atomic center of mass motion and the internal degrees of freedom of the atom
, we calculate the quasi - probability distributions in the @xmath78 plane .
the quasi - probability or q - function is defined as @xmath79 , @xmath80 being the density matrix reduced for degrees of freedom of center of mass motion .
figure ( 3 ) shows in the upperv part the time evolution of q - function for zero deformation at @xmath81 and @xmath82 respectively .
starting with the center of mass initial state as a coherent state @xmath83 , the initially single peaked q - function splits into two peaked function counterrotating in the complex plane . at @xmath84 ,
when two peaks collide a revival of inversion oscillation occurs . a similar plot for @xmath85
is shown for @xmath86 and @xmath87 respectively .
an interesting feature of the q - function for @xmath85 is the presence of as many as five peaks at @xmath88 .
a sequenece of q - function distributions , for @xmath89 and @xmath90 shows the spreading out and breaking up of the quasi - probability into as many as eight distinct peaks(for @xmath91 .
the revival of rabi- oscillations occurs when all peaks collide together into a single peak at @xmath92 .
we also plot in fig.(3 ) the q - function for @xmath93 at @xmath94 again showing as many as seven peaks .
it is apparent that when trap potential is q - deformed the quasi - probability distribution is no longer symmetrically distributed about @xmath95 line.this asymmetry is a manifestation of unequal energy spacings between different trap states .
this calculation shows that in a q - deformed oscillator trap for a suitable deformation the collapse and revival patterns become well defined .
as proposed by blockley at el.@xcite the theoretical predictions in this case can be verified experimentally . from our earlier work@xcite with physical systems
we may conjecture that the deformation corresponds to the presence of some nonlinearitis in the trapping potential . in this case small scale nonlinearities
are expected to play a beneficial role facilitating experimental observation . but large nonlinearities can wash out the collapse and revival pattern completely .
this result is very important for constructing -an atom in a trap system- with a high probability for observing collapses and revivals of population inversion . in the study of interaction of a three level atom with radiation in ref
@xcite it is shown that by preparing the atom in a special way the time dependent collapses and revivals are either greatly diminished or vanish all together .
the initial preparation forces the atomic population to remain coherently trapped in this configuration .
apparently @xmath0-deformed initial state is similar to a dressed atomic states .
we may conclude therefore that the nonlinearities of the trapping potential lead to effects similar to those obtained by setting initial atomic state amplitudes to selected values . in ref.@xcite collapses and revivals
have been studied for jcm with an intensity dependent coupling constant by using @xmath0-analog of harmonic oscillator to represent the radiation quanta . in their calculation the collapses and revivals are seen to become more diffuse as the deformation is increased .
our result is different from theirs in the sense that the collapses and revivals become well defined .
the major difference in their calculation and ours is that they use the zero order hamiltonian to be a harmonic oscillator hamiltonian whereas we have explicitely included the zero order spectrum of @xmath0-analog oscillator .
sharma would like to thank the department of physics and astronomy at rutgers university for its hospitality and to acknowledge financial support from cnpq , brazil .
n. k. sharma would like to thank the department of mathematics at rutgers university for its hospitality .
m. jimbo , lett .
* 10 * ( 1985)63 ; * 11 * ( 1986)247 .
s. woronowicz , publ .
rims(kyoto university ) , * 23 * ( 1987)117 ; commun .
* 111 * ( 1987)613 ; invent . math .
* 93 * ( 1988)35 .
v. pasquier , nucl . phys .
* b 295 * , ( 1988)491 ; commun . math .
* 118 * ( 1988)355 .
macfarlane a. j. , j. phys .
a*22 * ( 1989)4581 ;
wineland d. j. , itano w. m. , bergquist j. c. and hulet r. g. , phys . rev .
a , * 36 * ( 1987 ) 2220 .
biedenharn l. c. , j. phys .
a*22 * ( 1989)l873 .
diedrich f. , bergquist j. c. , itano w. m. and wineland d. j. , phys .
* 62 * ( 1989)403 . javanainen j. and stenholm s. , appl .
phys . , * 24 * ( 1981)151 .
blockley c. a. , walls d. f. and risken h. , europhys .
lett . , * 17 * ( 1992)509 . jaynes e. t. and cummings f. w. , proc
eng . , * 51 * ( 1963 ) 89
. s. shelly sharma , phys .
rev . c*46 * ( 1992)904 .
s. shelly sharma and n. k. sharma , phys .
c*50 * ( 1994)2323 .
d. a. cardimona , m. p. sharma and m. a. ortega , j. phys .
b : at . mol .
* 22 * ( 1989)4029 .
chaichian m. , ellinas d. and kullish p. , phys .
* 65 * ( 1990)980 . | we study the population inversion and q - function of a two - level atom , interacting with single - mode laser light field , in a @xmath0-analog harmonic oscillator trap for increasing @xmath0 . for @xmath1
the collapses and revivals of population inversion become well defined facilitating experimental observation but for large @xmath2 the time dependence of population inversion is completely wiped out . |
_ multiple hard parton interactions _ ( mpi ) started to play an important role in the description of the inelastic @xmath16 collisions at the collider energies . hence , although the studies of mpi began in eighties @xcite , they attracted a lot of theoretical and experimental attention only recently .
extensive theoretical studies were carried out in the last decade , both for pp collisions @xcite , and for @xmath3 collisions @xcite .
attempts have been made to incorporate multi - parton collisions in the monte carlo event generators @xcite .
mpis can serve as a probe for _ non - perturbative correlations _ between partons in the nucleon wave function and are crucial for determining the structure of the underlying event at the lhc energies .
they constitute an important background for the new physics searches at the lhc .
a number of experimental studies were performed at the tevatron @xcite .
new experimental studies are underway at the lhc @xcite .
the analysis of the experimental data indicates @xcite that the rate of such collisions exceeds significantly a naive expectation based on the picture of the binary collisions of the uncorrelated partons of the nucleons ( provided one uses information from hera on the transverse distribution of gluons in nucleons ) . in the parton model
inspired picture mpi occur via collisions of the pairs of partons : the 2 mechanism ( collision of two pairs of partons ) . in pqcd
the picture is more complicated since the qcd evolution generates short - range correlations between the partons ( splitting of one parton into two , ... ) the 2 mechanism@xcite .
it was demonstrated that account of these pqcd correlations enhances the rate of mpi as compared to the parton model by a factor of up to two and may explain discrepancy of the data @xcite with the parton model .
( a much larger enhancement recently reported in the double @xmath17 production @xcite can hardly be explained by this mechanism ) .
presence of two mechanisms and limited knowledge of the nucleon multiparton structure makes a unique interpretation of the data rather difficult .
hence here we propose to study the mpi process of @xmath18 interaction with production of four jets in the kinematics where two jets carry most of the light cone fraction of the photon four momentum - direct photon mechanism . in this process
the 2 mechanism is absent and the only process which contributes is an analog of the 2 process.since in the proposed kinematics the contribution of the resolved photon is strongly suppressed the cross section in the leading log approximation ( lla ) i.e. summing leading collinear singularities is expressed through the integral over two particle gpds , @xmath19 , introduced in @xcite .
this is in difference from the case of @xmath20 scattering where 2 contribution is proportional to a more complicated integral with the integrant proportional to the product of two double parton gpds .
the main goal of the present paper is to show that processes with direct photon in photon proton collisions provide a golden opportunity for the model independent determination of the double parton distributions @xmath21 , free of the ambiguities inherent in @xmath22 scattering @xcite
. we will consider the process of the interaction of the real / quasireal photon with proton with production of two pairs of hard jets in the back to back kinematics with each dijet consisting of a heavy ( charm ) quark and gluon jets ( see figs . 1 and 2 ) .
we focus on the production of charm to suppress the contribution of the resolved photons . in the discussed process
a @xmath23 pair is produced in the photon fragmentation region , while two gluon jets are created predominantly in the target region , so that there is a large rapidity gap between the gluon and quark jets .
the gluon and c - quark jets are approximately balanced pair vice .
the cross section of the analogous process in @xmath16 collisions is influenced by parton correlations in both nucleons participating in the process , while in the case of the photon the cross section depends only on the integral over one wave function .
the reason is that the process involves only one gpd from the nucleon , while the upper part of the diagram 1 is determined by the hard physics of the photon splitting to @xmath24 pair in an unambiguous way .
it does not involve the scale @xmath25 that separates perturbative and nonperturbative correlations in a nucleon .
thus the cross section of such a process is directly expressed through the nucleon double gpd .
hence the measurement of the discussed cross section would allow to perform a nearly model independent analysis of dpi in @xmath16 scattering .
we will demonstrate below that it would be possible also to study these processes at the future electron - proton / nucleus colliders .
it maybe possible also to investigate these processes in @xmath2 and @xmath3 ultraperipheral collisions at the lhc . here
we will consider the mpi rates for all three types of processes mentioned above , @xmath26 .
we will restrict ourselves to the kinematics @xmath0 , thus guaranteeing the dominance of the direct photon contribution ( for this cutoff the direct photons contribute 60% of the dijet cross section ) .
for a lower @xmath27 cutoffs the relative contribution of the direct photon mechanism rapidly decreases for transverse momenta under consideration .
we will demonstrate that for the lhec collider energies @xmath28 gev the rate of the discussed reaction will be very high : @xmath5 events per @xmath8 s for the luminosity @xmath6 @xmath29s@xmath30 and @xmath9 gev .
the relative rate of mpi to to dijets is found to be @xmath31 . in principle
a large number of events in the discussed kinematics was produced at hera : @xmath32 for the total luminosity @xmath33 fb@xmath30 . however
the efficiency of the detection of @xmath34 was pretty low @xcite so it appears that it would be very difficult to study the discussed process using the hera data .
another way to observe the discussed process in the near future maybe possible - study of mpi in the ultraperipheral @xmath35 processes at the full lhc energy . for example , for @xmath9 gev , we have @xmath12 events for @xmath2 , and @xmath36 events for @xmath3 scattering where we used luminosities @xmath37 ( @xmath2 ) , @xmath38 @xmath29s@xmath30 ( @xmath3 ) and running time of @xmath8 s. in the discussed kinematics mpi events constitute @xmath39 ( @xmath40 , @xmath41 ) of the dijet events for @xmath42 collisions respectively for the same jet cutoff .
these fractions decrease rather rapidly with @xmath43 increase .
of course , the mpi processes are contaminated by the leading twist 4 jet production the so called 2 to 4 processes
. however it is possible to argue that in the back to back kinematics the contribution of these processes(see fig.3 ) are parametrically small in a wide region of the phase space @xcite .
( moreover for the @xmath2 collisions there is an additional combinatorial @xmath44 enhancement over parasitical 2 to 4 contributions @xcite . )
indeed , a detailed mc simulation analysis was done using pythia and madgraph for @xmath16 collisions @xcite .
these authors have demonstrated that it is possible to introduce observables that are dominated by mpi in the back to back kinematics , thus allowing to measure mpi cross sections , as distinct from 2 to 4 processes .
of course , further work is required in this direction , especially including nlo effects .
note that mpi in the photon proton collisions were also studied in @xcite .
these authors considered resolved photon kinematics , which is very different , from the one that is considered here .
so there is no overlap with the present study .
the paper is organized as following . in section 2
we calculate the mpi contribution to @xmath45 process in the back to back kinematics . in section 3
we calculate the rates of the discussed process for @xmath46 collisions at lhec and hera , and for ultraperipheral @xmath3 and @xmath2 collisions at the lhc . in section 4
we carry the numerical simulations for realistic parameters corresponding to lhc and hera runs .
the results are summarized in section 5 .
first we consider the process of production of two dijets with single charm in each pair ( fig .
4a ) in the parton model . in this case
the process is essentially the same as the one already considered in ref .
the only difference is that the parton created in the split vertex is a charmed quark
antiquark pair .
the corresponding kinematics is depicted in fig .
4a , and is analogous to the 2 transition in @xmath16 interactions .
let us parameterize the momenta of quarks and gluons using sudakov variables ( @xmath47 are momenta of virtual charm quarks and antiquark of the @xmath48 pair and @xmath49 are the gluon momenta ) .
let us analyze the lowest order amplitude shown in fig .
4a for the double hard collision which involves parton splitting .
we decompose parton momenta @xmath50 in terms of the so called sudakov variables using the light - like vectors @xmath51 and @xmath52 along the incident photon and proton momenta : @xmath53 here @xmath54 is the momentum of the quasireal photon .
we can neglect the charm quark masses except while dealing with infrared singularities .
the light - cone fractions @xmath27 , ( i=1, .. 4 ) , are determined by the jet kinematics ( invariant masses and rapidities of the jet pairs ) .
the fraction @xmath55 that measures the _ difference _ of the longitudinal momenta of the two partons coming from the hadron , is arbitrary .
the fixed values of the parton momentum fractions @xmath56 and @xmath57 correspond to the plane wave description of the scattering process in which the longitudinal distance between the two scatterings is arbitrary .
this description does not correspond to the physical picture of the process we are discussing , where two partons originate from the same bound state . in order to ensure that partons @xmath58 and @xmath59 originate from the _ same hadron _ of a finite size , we have to introduce integration over @xmath55 in the _ amplitude _ , in the region @xmath60 , as was explained in detail in @xcite .
the feynman amplitude contains the product of two virtual propagators .
the virtualities @xmath61 and @xmath62 in the denominators of the propagators can be written in terms of the sudakov variables as k_1 ^ 2 = x_1s -k_^2 , k_2 ^ 2 = -x_2s -k_^2 , where @xmath63 and @xmath64 the square of the two - dimensional transverse momentum vector .
the singular contribution we are looking for originates from the region @xmath65 .
hence the precise form of the longitudinal smearing does not play role and the integral over @xmath55 yields the amplitude @xmath66 a~ = .
the numerator of the full amplitude is proportional to the _ first power of the transverse momentum _ @xmath67 . as a result , the squared amplitude ( and thus the differential cross section ) acquires the necessary factor @xmath68 that enhances the back - to - back jet production .
the integration over @xmath69 gives a single log contribution to the cross section @xmath70 where @xmath71 is the characteristic transverse scale of the hard processes .
note , that strictly speaking the answer is proportional to @xmath72 .
the parton model answer is only single collinearly enhanced , while we are looking for the double collinear enhanced contributions @xcite .
it is well known that these contributions originate from the gluon dressing of the parton model vertex , with the @xmath73 function becoming a new pole . a typical lowest order qcd diagram which accounts for the gluon emission
is presented in fig .
the compensating gluon relaxes the transverse momentum @xmath74 function .
note however that the gluon can not be emitted from a photon , while in the @xmath16 case such emissions contribute , since the splitting parton carries color .
this eliminates the so called short split contribution , which is present in the case of hadron - hadron scattering .
the rest of the calculation is completely analogous to the `` long split '' calculation in the @xmath75 case . thus using eqs .
25,26 in @xcite we can write right away the differential cross section as [ eq : dd31 t ] ^2 & = & \ { _ [ 1]d_a^1,2(x_1,x_2;_13 ^ 2 , _
24 ^ 2 ) _ [ 2]d_b^3,4(x_3,x_4;_13 ^ 2 , _ 24 ^ 2 ) + & & s_1(q^2,_13 ^ 2 ) s_3(q^2,_13 ^ 2 ) s_2(q^2,_24 ^ 2 ) s_4(q^2,_24 ^ 2 ) } , where @xmath76 are the quark ( @xmath77 ) and gluon ( @xmath78 ) sudakov form factors @xcite : @xmath79 \right\}.\end{aligned}\ ] ] here @xmath80 are the non - regularized one - loop dglap splitting functions ( without the `` + '' prescription ) : @xmath81 p_g^q(z ) & = & t_r\big [ z^2+\ ! ( 1\!-\!z)^2\big ] , \quad\ , p_g^g(z ) = c_a\frac{1+\!z^4 + \ ! ( 1\!-\!z)^4}{z(1-z)}.\nonumber\\\end{aligned}\ ] ] the upper limit of the integration over @xmath82 properly regularizes the soft gluon singularity , @xmath83 ( in physical terms , it can be viewed as a condition that the energy of the gluon should be larger than its transverse momentum , @xcite ) .
the function @xmath84 now corresponds to the photon split into the charm anticharm pair .
moreover , since we are looking for the production of the @xmath23 pair in the photon fragmentation region , we can neglect all processes except a possible emission of the compensating gluon by the @xmath85- quark .
hence we obtain [ eq : termpt ] _ [ 1 ] d(x_1,x_2 ; q_1 ^ 2,q_2 ^ 2 ; ) = _ ^2^ r(z ) g^q_q(;q_1 ^ 2,k^2 ) g^q_q(;q_2 ^ 2,k^2 ) .
the function @xmath86 is the @xmath87 vertex@xcite
. r(z)=z^2+(1-z)^2 . [ 2 ] the @xmath88-dependence of @xmath89 } d$ ] is very mild as it emerges solely from the lower limit of the logarithmic transverse momentum integration @xmath90 . here
@xmath91 is a quark - quark evolution kernel . in the lla for hard scale
@xmath92 we can use the kernel for massless quarks . above we have calculated the differential mpi distributions .
we now can integrate the cross section obtaining = _ [ 1]d_a(x_1,x_2;q^2,q^2 ) _ [ 2]d_b(x_3,x_4;q_1 ^ 2,q_2 ^ 2;^2 ) .
note that we write here the dijet differential cross sections @xmath93 without including the corresponding pdf factors .
we see that the cross section is unambiguously determined by the integral of @xmath94gpd over @xmath95 .
the factor @xmath84 is given by eq .
[ eq : dd31 t ] ( with @xmath96 ) and does not pose any infrared problem , in difference from the @xmath16 case .
there are three possible applications of our formalism collisions at hera and future @xmath97 colliders and ultraperipheral @xmath2 and @xmath3 collisions at lhc . in order to estimate whether it is feasible to observe the mpi events discussed in the previous section
, we have to calculate the double differential cross section and then to convolute it with the photon flux .
for the case of the proton target we have = d(x_1,x_2,p_1t^2,p_2t^2)g(p_1t^2,x_3)g(p_2t^2,x_4 ) u ( ) . here we carried the integration over the momenta @xmath88 conjugated to the distance between partons , obtaining the last multipliers in the equations above .
this integral measures the parton wave function at zero transverse separation between the partons and hence it is sensitive to short - range parton - parton correlations . for @xmath1 case the factor @xmath98 , in the approximation when two gluons are not correlated , is equal to a product of two gluon form factors of the proton : u(,x_3,x_4)=f_2g(,x_3 ) f_2g(,x_4 ) .
[ u ] for the numerical estimates we use the following approximation for @xmath99 of the nucleon : _
2d(x_3,x_4,p_1t^2,p_2t^2,)=g(x_3,p_1t)g(x_4,p_2t)f_2g(,x_3)f_2g(,x_4 ) where the two gluon form factor f_2g()= and the parameter m_g^2=8/ , where = max(0.28fm^2 , 0.31fm^2 + 0.014fm^2(0.1/x ) ) , and was determined from the analysis of the exclusive @xmath100 diffractive photoproduction@xcite . the functions @xmath101 are the gluon pdf of the proton , which we parameterize using @xcite . then u()=(1/m^2_g(x_3)-1/m^2_g(x_4)+ . in the limit @xmath102 we recover u()=. the general expressions for a nuclear target is = d(x_1,x_2,p_1t^2,p_2t^2)g(p_1t^2,x_3)g(p_2t^2,x_4)d^2f_a(,- ) where f_a(,-)=f_a(,-)+au ( ) .
[ tup ] here @xmath104 is the nucleus body form factor , and the form factor @xmath105 was defined in eq .
[ u ] . the first term in eq .
[ tup ] corresponds to the processes when two gluons originate from the different nucleons in the nucleus while the second term in eq.[tup ] corresponds to the case when they originate from the same nucleon .
the first term is expected to dominate for heavy nuclei as it scales as @xmath106 @xcite . for the nuclear target we have f_a(,-)=f^2 ( ) , f()=d^2b ( i)t(b ) , and t(b)=dz _ a(b , z)dz is the nucleus profile function .
the nuclear form factor integral is expressed through the profile function as f(,-)=t^2(b)d^2b = t^2(b)db^2 , where t(b ) is calculated using the conventional mean field nuclear density @xcite _ a(b , z)=. the factor @xmath107 is a normalization constant d^2bdz _ a(b , z)=1 . here
the distance scales are given in fm .
there can be also the ladder splitting from the proton side .
however such process corresponds to 2 to 4 process in notations of @xcite and thus does not contribute to mpi in the lla we consider here .
such processes constitute @xmath108 corrections to conventional 2 to 4 four jet production , and it is expected they give a small contribution in the back to back kinematics .
this is consistent with the results of modeling a tree level processes 2 to 4 in @xmath109 in tevatron carried out by d0 and cms and atlas at lhc - see ref .
@xcite . still this issue definitely deserves a further study .
the relative rate of mpi and 2 to 4 processes plays an important role in accessing feasibility of observing mpi .
the @xmath3 collisions are dominated by the @xmath110 process where a much larger flux factor is generated by projectile nuclei leading to dominance of the ultra peripheral collisions of photons with protons .
hence in such process one predominantly measures a double gpd of a proton . at the same time in the ultraperipheral @xmath2 process
the dominant contribution originates from the interaction of charmed pair with two gluons coming from different protons @xcite . the ratio of cross section of such dpi process in @xmath2 scattering to the cross section of dpi cross section in @xmath3 scattering , in which both gluons belong to the same nucleon is ( since the photon flux from nuclei is the same ) ~2,[simon1 ] where we take @xmath111 .
thus the ratio of the total number of the mpi events in @xmath2 to the rate calculated in the impulse approximation is @xmath112.this is consistent with a numerical analysis that shows that for the same c.m .
energies the ratio of number of mpi events to dijet rate in @xmath2 collisions is 2.53 times larger than the same ratio for @xmath113 process . note that this result is purely geometrical , we find a similar ratio [ simon1 ] for @xmath1 and @xmath103 collisions at the lhec .
the cross sections @xmath114 are usual dijet cross sections calculated with @xmath115 , where @xmath116 is the invariant energy of the @xmath117 . .
we have d / dt= , where @xmath118 is the dijet transverse scale . here
the matrix element @xmath119 of the @xmath120-quark - gluon scattering is given by m^2=(4_s(q^2))^2(-(+)+ ) , where s = x_1x_3s , t =- s(1-z ) , u =- s ( 1+z ) , z==(y_1-y_3)/2=. the angle @xmath121 is the scattering angle in the c.m .
frame of the dijet .
the region of integration is given by x_1x_3s-4p_t^2>0,x_1>0.2 the integration over the second dijet event goes in the same way , with @xmath122 .
for two dijet event , in order to find the event rate , the cross section calculated above must be convoluted with photon flux determined using weiczsacker - williams approximation : n_2=dp_1t^2dp_2t^2 dudx_1dx_2dx_3dx_4 , where the limits of integration are determined by @xmath123 , @xmath124 , @xmath0 . in the same way we calculate the rate for production of one pair of jets : n_1=dp_1t^2 dudx_1dx_3 .
the limits of integration are determined by @xmath123 .
for @xmath46 collisions we use the standard variable @xmath125 s_p / s = y the photon flux is dn / dy=()(q^2_max / q^2_min)-2m^2_ey(1/q^2_min-1/q^2_max ) , where @xmath126 gev@xmath127 , and @xmath128 . in ultraperipheral collisions two nuclei ( proton and nucleus )
scatter at large impact parameters with one of the colliding particles emitting a weizsacker williams photon which interacts with the second particle producing two jets in the @xmath129 jets@xmath130 reaction ( for a detailed review see @xcite ) .
the corresponding total cross section is calculated by convoluting the elementary @xmath131 cross section with the flux factor = ( wk_0(w)k_1(w)-(k_1 ^ 2(w)-k_0 ^ 2(w))),[tup21 ] where @xmath132 , @xmath133,@xmath134 . for the proton - nucleus reactions
the flux is described by the same eq .
[ tup21 ] , the only difference is that in definition of @xmath135 we substitute @xmath136 where @xmath137 is the proton radius . in the second process
the dominant contribution is the interaction of the a photon radiated by a heavy nucleus with the proton .
the factor in the square brackets accounts for the full absorption at impact parameters @xmath138 ( @xmath139 for @xmath3 scattering ) .
the bjorken fractions in the previous section where calculated relative to @xmath140 . in order to calculate the total inclusive cross section we must integrate over @xmath141 from @xmath142 corresponding to minimal @xmath141 necessary to produce four jets in the discussed kinematics up to @xmath143 , @xmath144 , and @xmath145 .
we must fix @xmath146 .
then we have to calculate the cross section at @xmath147 where @xmath148 .
thus to determine the total inclusive cross section for given @xmath146 , we have to integrate over @xmath141 , substituting in the formulae of the previous section , @xmath149 instead of @xmath150 .
the integration region is @xmath151 .
since there is no corresponding 4 to 4 process , it makes no sense to define @xmath152 for these collisions as it is usually done in the studies of mpi in @xmath16 scattering .
instead we will calculate the number of mpi events as a function of jet cutoff - starting from 5 gev , as well as the ratio of mpi events to a total number of dijet events with the same cutoffs for @xmath46 , @xmath110 and @xmath2 collisions . in all cases
we observe a rather rapid decrease of mpi rate as a function of @xmath43 . in order to calculate the rates we use the grv structure functions for proton @xcite and the grv structure function for photon @xcite . to estimate the mpi event rate at lhec at @xmath4 tev we used luminosity @xmath6 @xmath153 .
the number of events and their ratio to the total number of dijet event are presented in figs .
5,6 . for cutoff @xmath9
gev we get @xmath5 events for the running time of @xmath8 s. the ratio to a number of dijet events with the same cutoffs on @xmath154 and @xmath43 is @xmath31 .
we also considered mpi event rate in the similar kinematics at hera . to estimate the mpi event rate at hera we use the total integrated luminosity accumulated at hera of @xmath33 fb@xmath30 , at the @xmath155=300 gev . for cut off @xmath9
gev we get @xmath156 events
. the ratio to the number of dijet events with the same cutoffs on @xmath154 and @xmath43 is @xmath157 .
however , it seems difficult to connect these results with the available data on multiparton processes studied at hera @xcite .
the reason is a rather low efficiency of selecting events with charm production .
for example , the total number of events with charm identified in the zeus study was 11000 @xmath34 events corresponding to a handful of mpi events of the type we discuss @xcite . for @xmath2 collisions we use : ( i ) luminosity @xmath158 , ( ii ) running time @xmath8 s , and @xmath159 tev , @xmath160 the radius of the lead nucleus is 6.5 fm . the exponentially decreasing macdonald function cuts off the contribution of high photon energy .
the total number of the events for the @xmath43 cut of @xmath161 gev is @xmath162 , while the ratio of mpi events to the total number of dijet events is relatively high@xmath163 ( cf .
discussion in sec .
the number of the mpi events decreases as a function of cutoff slightly faster than as @xmath164 , while the ratio of mpi to total number of dijet events scales approximately as @xmath165 .
the number of events and the ratio of a number of mpi and dijet events are presented in figs .
recall that @xmath166 in figs .
7,8 is 0.4 .
the dependence of @xmath167 on @xmath168 is shown in fig .
we see that for large @xmath168 the number of events rapidly decreases , while for @xmath169 the decrease becomes much less rapid .
if we increase @xmath168 to 0.8 the number of @xmath2 mpi events decreases by a factor of four only .
so it may be possible to focus on the higher @xmath170 regions where contribution of resolved photon is very small , while loosing relatively small fraction of events .
finally note that this ratio rapidly increases with energy . if we for example take @xmath2 energies equal to those of @xmath3 scattering ( i.e. a factor of 2.5 increase of @xmath116 ) the ratio increases by @xmath171 . for @xmath3 collisions we use luminosity @xmath172 , running time @xmath8 s , and @xmath173 tev .
the number of events for cut @xmath161 gev is of the order @xmath174 , and ratio is of order @xmath175 , rapidly decreasing as in the @xmath2 case with increase of @xmath43 .
the corresponding numbers are shown in figs .
10,11 as a function of the jet cutoff . from the comparison of figs . 9 and
11 one can see that there is a factor of @xmath176 enhancement of the ratio dpi to dijet events in @xmath2 scattering , relative to @xmath3 case which is due to a combination of the geometrical enhancements we discussed above and suppression due to the smaller energy per nucleon in @xmath2 case .
we derived general equations for mpi processes with production of charm in direct photon hadron ( nucleon , nuclei ) collisions and used them to calculate the corresponding rates and compare them with dijet rates .
we demonstrated that the discussed processes directly measure nucleon and nucleus @xmath99s .
we found a significant enhancement of the mpi / dijet cross section ration in the @xmath177 scattering as compared to @xmath1 scattering due to the scattering off two nucleons along the photon impact parameter .
the analysis was done for jet photoproduction in the realistic kinematics , of production of two pairs charm - gluon dijets with @xmath178 gev , and cut of @xmath0 , ensuring they are created mainly due to the direct photon mechanism .
we considered these mpi processes for @xmath46 collisions at the lhec and hera and for @xmath2 and @xmath3 collisions at lhc and @xmath46 collisions at hera .
we conclude that the studies would definitely be feasible at the lhec . in the case of the lhc
the rates appear reasonable , and the key question is the efficiency of the lhc detectors .
further studies of the feasibility of the measuring discussed processes at the lhc is highly desirable .
here we just notice that since a larger fraction of charm in the discussed processes is produced at the central rapidities we expect the efficiency of the detection of the discussed process would be pretty high for atlas and cms .
we thank the cern theory division for hospitality during the time the initial part of this work was done .
m.s.s research was supported by the us department of energy office of science , office of nuclear physics under award no .
de - fg02 - 93er40771 .
m. mekhfi , phys .
rev . d*32 * , 2371 ( 1985 ) .
r. kirschner , phys .
* b84 * , 266 ( 1979 ) ; + v.p .
shelest , a.m. snigirev and g.m .
zinovjev , phys .
* b113 * , 325 ( 1982 ) .
a. del fabbro and d. treleani , phys .
d * 61 * , 077502 ( 2000 ) [ arxiv : hep - ph/9911358 ] ; phys .
d * 63 * , 057901 ( 2001 ) [ arxiv : hep - ph/0005273 ] ; a. accardi and d. treleani , phys .
d * 63 * , 116002 ( 2001 ) [ arxiv : hep - ph/0009234 ] .
b. blok , yu . dokshitzer , l. frankfurt and m. strikman , phys .
d * 83 * , 071501 ( 2011 ) [ arxiv:1009.2714 [ hep - ph ] ] .
b. blok , yu . dokshitser , l. frankfurt and m. strikman , eur .
j. c * 72 * , 1963 ( 2012 ) [ arxiv:1106.5533 [ hep - ph ] ] . b. blok , yu . dokshitser , l. frankfurt and m. strikman , arxiv:1206.5594v1 [ hep - ph ] ( unpublished ) .
b. blok , y. dokshitzer , l. frankfurt and m. strikman , eur .
j. c * 74 * ( 2014 ) 2926 [ arxiv:1306.3763 [ hep - ph ] ] .
j. r. gaunt , r. maciula and a. szczurek , arxiv:1407.5821 [ hep - ph ]
gaunt , jhep * 1301 * , 042 ( 2013 ) [ arxiv:1207.0480 [ hep - ph ] ] m. bhr , m. myska , m.h .
seymour and a. sidmok , arxiv : 1302.4325 [ hep - ph ] m. strikman and d. treleani , phys .
* 88 * , 031801 ( 2002 ) [ hep - ph/0111468 ] .
b. blok , m. strikman and u. a. wiedemann , eur .
j. c * 73 * ( 2013 ) 2433 [ arxiv:1210.1477 [ hep - ph ] ] .
d. denterria and a. m. snigirev , phys .
b * 718 * ( 2013 ) 1395 [ arxiv:1211.0197 [ hep - ph ] ] .
for pythia summary see t. sjostrand and p.z .
skands , c. flensburg , g. gustafson , l. lonnblad and a. ster , arxiv:1103.4320 [ hep - ph ]
. j. m. butterworth , j. r. forshaw and m. h. seymour , z. phys .
c * 72 * ( 1996 ) 637 [ hep - ph/9601371 ] f. abe _ et al .
_ [ cdf collaboration ] , phys . rev .
d * 56 * , 3811 ( 1997 ) .
abazov _ et al . _
[ d0 collaboration ] , phys .
d * 81 * , 052012 ( 2010 ) . v.m . abazov _ et al . _
[ d0 collaboration ] , phys .
d * 83 * , 052008 ( 2011 ) .
g. aad _ et al . _ [ atlas collaboration ] , new j. phys .
* 15 * ( 2013 ) 033038 [ arxiv:1301.6872 [ hep - ex ] ] .
s. chatrchyan _ et al . _
[ cms collaboration ] , jhep * 1403 * ( 2014 ) 032 [ arxiv:1312.5729 [ hep - ex ] ] .
s. chatrchyan _ et al . _
[ cms collaboration ] , jhep * 1312 * ( 2013 ) 030 [ arxiv:1310.7291 [ hep - ex ] ] .
cms collaboration [ cms collaboration ] , `` measurement of the double - differential inclusive jet cross section at sqrt(s ) = 8 tev with the cms detector , '' cms - pas - smp-12 - 012 .
v. m. abazov _ et al .
_ [ d0 collaboration ] , arxiv:1406.2380 [ hep - ex ] .
h. abramowicz _ et al .
_ [ zeus collaboration ] , arxiv:1405.5068 [ hep - ex ] .
dokshitzer sov .
jetp * 46 * , 641 ( 1977 ) ; + yu.l .
dokshitzer , v.a .
khoze , a.h .
mueller and s.i .
troian , `` basics of perturbative qcd '' , ed .
j. tran thanh van , editions frontires , gif - sur - yvette , 1991 yu.l .
dokshitzer , d.i .
diakonov and s.i .
troian , phys .
rept . * 58 * , 269 ( 1980 ) .
m. peskin , introduction to quantum field theory , addison - wesley publishing company , reading , massachusets,1996 .
m. gluck , e. reya and a. vogt , z. phys .
c * 53 * , 127 ( 1992 ) .
a. bohr and b.r .
mottelson , nuclear structure , v.1 , w. a. benjamin , new york , 1969 .
a. j. baltz , g. baur , d. denterria , l. frankfurt , f. gelis , v. guzey , k. hencken and y. kharlov _ et al .
_ , phys .
* 458 * , 1 ( 2008 ) [ arxiv:0706.3356 [ nucl - ex ] ] .
m. gluck , e. reya and a. vogt , phys .
d * 45 * ( 1992 ) 3986 .
m. gluck , e. reya and a. vogt , phys .
d * 46 * ( 1992 ) 1973 .
m. gluck , k. grassie and e. reya , phys .
d * 30 * , 1447 ( 1984 ) . | we derive expressions for the differential distributions and the total cross section of double- parton interaction in direct photon interaction with proton and nuclei .
we demonstrate that in this case the cross section is more directly related to the nucleon generalized parton distribution than in the case of double parton interactions in the proton - proton collisions .
we focus on the production of two dijets each containing charm ( anticharm ) quarks and carrying @xmath0 fractions of the photon momentum .
numerical results are presented for the case of @xmath1 collisions at lhec , hera and in the ultraperipheral @xmath2 and @xmath3 collisions at the lhc .
we find that the events of this kind would be abundantly produced at the lhec .
for @xmath4 tev the expected rate is @xmath5 events for the luminosity @xmath6 @xmath7 , the running time of @xmath8 s and the transverse cutoff of @xmath9 gev .
this would make it feasible to use these processes for the model independent determination of two parton gpds in nucleon and in nuclei . for hera
the total accumulated number of the events is also high , but efficiency of the detection of charm seems too low to study the process .
we also find that a significant number of such double parton interactions should be produced in @xmath10 and @xmath11 collisions at the lhc : @xmath12 for @xmath13 , and @xmath14 for @xmath15 collisions for the same transverse momentum cutoff . |
galaxies have a wide range of luminosities , colors , masses , sizes , surface brightnesses , morphologies , star formation histories and environments .
this heterogeneity is not surprising , given the variety of physical processes which likely influence their formation and evolution , including gravitational collapse , hydrodynamics , turbulence , magnetic fields , black - hole formation and accretion , nuclear activity , tidal and merger interactions , and evolving and inhomogeneous cosmic radiation fields .
what _ is _ surprising is that populations of galaxies show several very precise relationships among their measured properties .
the properties we use to describe galaxies span a large `` configuration space '' , but galaxies do not fill it . galaxy spectral energy distributions , when scaled to a fixed broad - band luminosity , appear to occupy a thin , one - dimensional locus in color space or spectrum space ( e.g. , connolly & szalay 1999 ) .
spiral galaxies show a good correlation between rotation velocity and luminosity ( e.g. , tully & fisher 1977 ; giovanelli et al .
galaxy morphology is strongly correlated with broad - band colors , strengths of spectral features , and inferred star - formation histories ( e.g. , roberts & haynes 1994 ) . among all galaxy families , early - type (
elliptical and s0 ) galaxies show the most precise regularities ( djorgovski & davis 1987 ; burstein , bender , faber & nolthenius 1997 ) .
early - type galaxy surface - brightness distributions follow a very simple , universal `` de vaucouleurs '' profile ( de vaucouleurs 1948 ) .
their spectral energy distributions appear to be virtually universal , showing very little variation with mass , environment , or cosmic time ( e.g. , van dokkum & franx 1996 ; pahre 1998 ) .
what variations they do show are measurable and precise .
early - type galaxy colors , luminosities , half - light radii , velocity dispersions , and surface brightnesses are all correlated ( baum 1959 ; fish 1964 ; faber & jackson 1976 ; kormendy 1977 ; bingelli , sandage & tarenghi 1984 ) ; they can be combined into a two - dimensional `` fundamental plane '' with very little scatter ( e.g. , dressler et al .
1987 ; djorgovski & davis 1987 ; faber et al .
1987 ) .
the homogeneity of the early - type galaxy population is difficult to understand if early - type galaxies are assembled at late times by stochastic mergers of less - massive galaxies of , presumably , different ages , star formation histories , and gas contents , as many models postulate ( e.g. , larson 1975 ; white & rees 1978 ; van albada 1982 ; kauffmann 1996 ; kauffmann & charlot 1998 ) .
it is possible that the homogeneity of early - type galaxies points to early formation ( e.g. , worthey 1994 ; bressan et al .
1994 ; vazdekis et al .
1996 ; tantalo , chiosi & bressan 1998 ) ; certainly their stellar populations appear old ( e.g. , bernardi et al .
1998 ; colless et al .
1999 ; trager et al .
2000a , b ; kuntschner et al . 2001 ) . alternatively ,
the observable properties of the stellar content of early - type galaxies are fixed entirely by the properties of the collisionless , self - gravitating , dark - matter haloes in which we believe such galaxies lie ( e.g. , hernquist 1990 ) .
these halos , almost by definition , are not subject to the vagaries of gas dynamics , star formation , and magnetic fields ; they are influenced only by gravity .
it is essentially a stated goal of the sloan digital sky survey ( sdss ; york et al .
2000 ; stoughton et al . 2002 ) to revolutionize the study of galaxies .
the sdss is imaging @xmath1 steradians of the sky ( northern galactic cap ) in five bands and taking spectra of @xmath2 galaxies and @xmath3 qsos . among the @xmath4 sdss spectra
there will be roughly @xmath5 spectra taken of early - type galaxies ; in fact @xmath6 of the spectroscopic fibers are being used to assemble a sample of luminous early - type galaxies with a larger mean redshift than the main sdss sample ( eisenstein et al .
the high quality of the sdss 5-band ccd imaging ( gunn et al . 1998 ; lupton et al . 2001 ) allows secure identification of early - type galaxies and precise measurements of their photometric properties ; most spectroscopic targets in the sdss are detected in the imaging at signal - to - noise ratios @xmath7 .
early - type galaxy studies in the past , for technical reasons , have concentrated on galaxies in clusters at low ( e.g. , jrgensen , franx & kjrgaard 1996 ; ellis et al . 1997
; pahre , djorgovski & de carvalho 1998a , b ; scodeggio et al .
1998 ; colless et al .
2001 ; saglia et al . 2001 ; kuntschner et al . 2001
; bernardi et al . 2002a ,
b ) and intermediate redshifts ( e.g. , van dokkum et al . 1998 , 2001 ; kelson et al . 2000 ; ziegler et al . 2001 ) . only the large area ` seven samurai ' ( e.g. , faber et al .
1989 ) and enear surveys ( e.g. , da costa et al .
2000 ) of nearby early - types , recent work with galaxies in the sbf survey ( blakeslee et al .
2001 ) , and some studies at intermediate redshifts by schade et al .
( 1999 ) , treu et al .
( 1999 , 2001a , b ) and van dokkum et al .
( 2001 ) , are not restricted to cluster environments . in constrast ,
the sdss is surveying a huge volume of the local universe , so the sample includes early - type galaxies in every environment from voids to groups to rich clusters . as of writing , when only a small fraction of the planned sdss imaging and spectroscopy has been taken , the number ( @xmath8 ) of early - type galaxies with well - measured velocity dispersions and surface - brightness profiles in the sdss greatly exceeds the total number in the entire astronomical literature to date .
this is the first of four papers in which we use the sdss sample to measure the fundamental plane and other early - type galaxy correlations in multiple bands .
this first paper describes how the sample was selected and presents the data .
section [ sample ] describes the main properties of the sdss database .
the criteria used to select the early - type galaxy sample are described in section [ finalsample ] .
this section also presents a selection of images from a range of redshifts , shows de vaucouleur profile fits to the observed surface brightness profiles , and argues that the sdss photometric pipeline treatment of seeing and sky - subtraction have not strongly compromised the estimates of the best - fitting model parameters .
it also displays examples of the spectra from which we estimate velocity dispersions .
section [ catalog ] presents the final catalog of photometric and spectroscopic parameters which we use for our subsequent analyses of early - type galaxy properties , and discusses some properties of the sample .
many details are relegated to appendices .
appendix [ kcorrs ] contains a discussion of the various k - corrections we have tried . the way we estimate velocity dispersions is presented in appendix [ vmethods ] .
a novel method for estimating aperture corrections to the velocity dispersions is discussed in appendix [ vr ] .
finally , the covariance matrix of the errors is discussed in appendix [ errors ] .
paper ii ( bernardi et al .
2003a ) studies correlations between various pairs of observables , such as the faber - jackson and kormendy relations .
it also presents the luminosity function and its evolution .
the fundamental plane and its dependence on waveband , color , redshift , and environment is studied in paper iii ( bernardi et al .
the co - added spectra of these galaxies are studied in paper iv ( bernardi et al .
one of the results of that paper is a library of co - added spectra which contains spectra that represent a wide range of early - type galaxies .
this library is available electronically .
these spectra indicate that the chemical composition of the early - type galaxy population evolves with redshift .
the chemical abundances and evolution are then combined with stellar population models to estimate the ages and metallicities of the galaxies in our sample .
this paper also analyzes correlations with color ( e.g. , color magnitude and color@xmath9 relations ) and discusses the effects of color gradients on measurements of the strength of the correlation between color and magnitude . except where stated otherwise , we write the hubble constant as @xmath10 , and we perform our analysis in a cosmological world model with @xmath11 , where @xmath12 and @xmath13 are the present - day scaled densities of matter and cosmological constant . in such a model
, the age of the universe at the present time is @xmath14 gyr . for comparison ,
an einstein - de sitter model has @xmath15 and @xmath16 gyr .
we frequently use the notation @xmath17 as a reminder that we have set @xmath18 .
also , we will frequently be interested in the logarithms of physical quantities .
our convention is to set @xmath19 and @xmath20 , where @xmath21 and @xmath9 are effective radii in @xmath22 kpc and velocity dispersions in km s@xmath23 , respectively .
the sdss project is described in stoughton et al . ( 2002 ) .
the data we analyze in this paper were selected from the sdss database in summer 2001 . at that time
the sdss had imaged @xmath24 square degrees ; @xmath25 galaxies and @xmath26 qsos had both photometric and spectroscopic information . the photometric and spectroscopic data were taken with the 2.5-m sdss telescope at the apache point observatory ( new mexico ) between 1999 march and 2000 october .
details of the photometric and spectroscopic observations and data reduction procedure will be presented elsewhere . here
we briefly summarize .
images are obtained by drift scanning with a mosaic ccd camera ( gunn et al .
1998 ) which gives a field of view of @xmath27 , with a spatial scale of @xmath28 in five bandpasses ( @xmath29 , @xmath30 , @xmath31 , @xmath32 , @xmath33 ) with central wavelengths ( 3560 , 4680 , 6180 , 7500 , 8870 ) ( fukugita et al .
the errors in @xmath29 band measurements are larger than the others , so we will only present results in the other four bands .
in addition , the photometric solutions we use in this paper are preliminary ( for details , see discussion of the early data release in stoughton et al .
2002 ) ; we use @xmath34 rather than @xmath31 , and similarly for the other bands , to denote this .
the effective integration time is 54 sec .
the raw ccd images are bias - subtracted , flat - fielded and background - subtracted .
pixels contaminated by the light of cosmic rays and bad columns are masked .
astronomical sources are detected and overlapping sources are de - blended .
the data are flux - calibrated by comparison with a set of overlapping standard - star fields calibrated with a 0.5-m `` photometric telescope '' ( hogg et al .
2001 ; smith et al .
the photometric telescope is also used for measuring the atmospheric extinction coefficients in the five bands .
the median effective seeing ( the median fwhm of the stellar profiles ) for the observations used here is @xmath35 .
all of this image processing is performed with software specially designed for reducing sdss data ( lupton et al .
the uncertainty in the @xmath34-band zero - point calibration is @xmath36 mag ; the uncertainty in the sky background subtraction is less than about 1% .
surface photometry measurements are obtained by fitting a set of two - dimensional models to the images .
the model fits account for the effects of seeing , atmospheric extinction , and galactic extinction ( this last uses the results of schlegel , finkbeiner , & davis 1998 ) . the sdss model for seeing
is described in stoughton et al .
briefly , the psf is expected to have a fourier transform @xmath37 $ ] characteristic of kolmogorov turbulence ( e.g. , saglia et al . 1993 ) .
this psf is approximated as the sum of gaussians ( this parametrization allows for a substantial reduction in the processing time required to deconvolve the effects of seeing ) .
the sdss photometric pipeline fits two models to the two - dimensional image of each object in each band : a pure de vaucouleurs profile and a pure exponential profile ( e.g. stoughton et al . 2002 ) .
briefly , the photon counts are binned into a number of radial and twelve angular bins .
then , two - dimensional de vaucouleurs and exponential models , convolved with the seeing , are fitted to the cumulative binned counts .
the pipeline does not include more complicated models ( e.g. , bulge plus disk ) because the computational expense to compute them is not justified for the majority of the detected objects .
algorithms which fit bulge and disk components ( which are desirable to fit on large galaxies ) to the surface brightness profiles are currently being developed within the collaboration , but are not yet available .
we present examples of @xmath34-band images and mean surface brightness profiles in section [ checkphoto ] .
the sdss image processing software provides several global photometric parameters , for each object , which are obtained independently in each of the five bands . because we are interested in early - type galaxies , we use
primarily the following : * the ratio @xmath38 of the lengths of the minor and major axes of the observed surface brightness profile . * the effective radius ( or half - light radius ) @xmath39 along the major axis and * the total magnitude @xmath40 ; these are computed by fitting a two - dimensional version of de vaucouleurs ( 1948 ) @xmath41 model to the observed surface brightness profile .
( the fitting procedure accounts for the effects of seeing we test the accuracy of this procedure below . )
* likelihood parameters @xmath42 and @xmath43 that indicate how well the de vaucouleurs or exponential models , when convolved with a model for the seeing , fit the observed light profile .
* the _ model _ magnitude @xmath44 ; this is the total magnitude calculated by using the ( de vaucouleurs or exponential ) model which fits the galaxy profile best in the @xmath34-band .
the _ model _
magnitudes in the other four bands are computed using that @xmath34 fit as filter ; in effect , this measures the colors of a galaxy through the same aperture . * the petrosian magnitude @xmath45 is also computed ; this is the flux within @xmath46 , where @xmath47 is defined as the angular radius at which the ratio of the local surface brightness at @xmath31 to the mean surface brightness within a radius @xmath31 is 0.2 ( petrosian 1976 ) . * the petrosian radii @xmath48 and @xmath49 ; these are the angular radii containing 50% and 90% of the petrosian light , respectively .
although our analyses on early - type galaxy properties presented here and in the companion papers have been performed with both the de vaucouleurs fit parameters and the petrosian quantities , in most of our analyses only the results of the de vaucouleurs fits are presented .
this is because the de vaucouleurs model appears to be a very good fit to the early - type galaxy surface - brightness profiles in the sdss sample and because it is conventional , in the literature on early - type galaxies , to use these quantities . on the other hand , for reasons given in stoughton et al .
( 2002 ) , unless stated otherwise , galaxy colors are always computed using _
model _ magnitudes .
the sdss takes spectra only for a target subsample of objects .
target selection criteria are described in stoughton et al .
( 2002 ) and strauss et al .
spectra are obtained using a multi - object spectrograph which observes 640 objects at once .
each spectroscopic plug plate , 1.5 degrees in radius , has 640 fibers , each @xmath50 in diameter .
two fibers can not be closer than @xmath51 due to the physical size of the fiber plug .
typically @xmath52 fibers per plate are used for galaxies , @xmath53 for qsos , and the remaining for sky spectra and spectrophotometric standard stars .
each plate typically has three to five spectroscopic exposures of fifteen minutes , depending on the observing conditions ( weather , moon ) ; a minimum of three exposures is taken to ensure adequate cosmic ray rejection . for galaxies at @xmath54
the median spectrum @xmath55 per pixel is 16 ( see figure [ fig : vmeth ] in appendix [ vmethods ] ) .
the wavelength range of each spectrum is @xmath56 .
the instrumental dispersion is @xmath57dex / pixel which corresponds to 69 km s@xmath23 per pixel .
( there is actually some variation in this instrumental dispersion with wavelength , which we account for ; see figure [ fig : resol ] and associated discussion in appendix [ vmethods ] . )
the instrumental resolution of galaxy spectra , measured from the autocorrelation of stellar template spectra , ranges from 85 to 105 km s@xmath23 , with a median value of 92 km s@xmath23 .
a highly automated software package has been designed for reducing sdss spectral data .
the raw data are bias - subtracted , flat - fielded , wavelength calibrated , sky - lines removed , co - added , cleaned from residual glitches ( cosmic rays , bad pixels ) , and flux calibrated . the spectro - software classifies objects by spectral type and determines emission and absorption redshifts .
( redshifts are corrected to the heliocentric reference frame . ) the redshift success rate for objects targeted as galaxies is @xmath58% and errors in the measured redshift are less than about @xmath59 . once the redshift as been determined the following quantites are computed : absorption - line strengths ( brodie & hance 1986 ; diaz , terlevich & terlevich 1989 ; trager et al .
1998 ) , equivalent widths of the emission lines , and eigen - coefficients and classification numbers of a pca analysis ( connolly & szalay 1999 ) .
some information about the reliability of the redshift and the quality of the spectrum is also provided .
we present examples of the spectra in section [ exspec ] .
the sdss pipeline does not provide an estimate of the line - of - sight velocity dispersion , @xmath9 , within a galaxy , so we compute it separately for the early - type galaxy sample ( see section [ vdisp ] ) .
the main goal of this series of papers is to study the properties of early - type galaxies using the main galaxy sample of the sdss database . therefore , one of the crucial steps in our study is the separation of galaxies into early and late types .
we want to select objects whose spectra are good enough to compute the central velocity dispersion .
in addition , because we wish to study the colors of the galaxies in our sample , we must not use color information to select the sample . to reach our goal ,
we have selected galaxies which satisfy the following criteria : * concentration index @xmath60 in @xmath61 ; * the likelihood of the de vaucouleurs model is at least 1.03 times the likelihood of the exponential model ; * spectra with pca classification numbers @xmath62 , typical of early - type galaxy spectra ( connolly & szalay 1999 ) ; * spectra without masked regions ( the sdss spectroscopic pipeline outputs a warning flag for spectra of low quality ; we only chose spectra for which this flag was set to zero ) ; * @xmath63 ; * redshift @xmath64 .
section [ vdisp ] describes how we estimate the velocity dispersion @xmath9 for galaxies which satisfy the above criteria .
we consider velocity dispersion estimates smaller than about 70 km s@xmath23 to be unreliable ( see appendix [ vmethods ] ) .
therefore , in the final sample presented in section [ catalog ] , galaxies with @xmath65 km s@xmath23 have been excluded .
the results of this and the companion papers are not significantly different if we change the cut - off on velocity dispersions to 100 km s@xmath23 .
about 9,000 objects satisfied all the above criteria . as stated earlier , the sdss pipeline does not output disk - to - bulge ratios from fits to the light profiles .
the first two requirements above attempt to select profile shapes which are likely to be those of spheroidal systems .
the @xmath61 band measurements tend to be less noisy than @xmath66 or @xmath67 , so we chose to use the @xmath61 band estimate of how centrally concentrated the light is .
the second requirement reduces approximately to requiring that @xmath68 ; that is , the surface brightness profile should be better fit by the de vaucouleurs model than by an exponential .
the spectra of late - type galaxies show emission lines , so examining the spectra ( e.g. , using the pca classification ) is a simple way of removing such objects from the sample . because the aperture of an sdss spectroscopic fiber ( @xmath50 ) samples only the inner parts of nearby galaxies , and because the spectrum of the bulge of a nearby late - type galaxy can resemble that of an early - type galaxy , it is possible that some nearby late - type galaxies could be mistakenly included in the sample ( e.g. , kochanek , pahre & falco 2000 ) .
most of these will have been excluded by the first two cuts on the shape of the light profile . to check this , we visually inspected all galaxies with @xmath69 . about @xmath70 of 225 ( i.e. , about 20% ) looked like late - types , and so we removed them .
note that weak emission lines , such as h@xmath71 and/or o ii , are still present in the early - type galaxy spectra in our sample .
for example , the median h@xmath71 equivalent width for galaxies in our sample is @xmath72 , but @xmath73% of the galaxies have equivalent widths larger than @xmath74 and @xmath75% of these ( i.e. 0.5% of the sample ) have equivalent widths in the range @xmath76 .
similarly , although the median equivalent width of o ii is @xmath77 , @xmath78 of the galaxies show o ii equivalent widths larger than @xmath79 and @xmath80 of these ( i.e. , 1.5% of the sample ) have equivalent widths ranging from @xmath81 to @xmath82 . the reason for our redshift cut is as follows .
the sdss main galaxy sample from which we select our early - type galaxies is apparent magnitude limited in @xmath34 , where the limits are defined using petrosian magnitudes @xmath45 .
this means that the main galaxy sample is not magnitude limited in the other bands .
blanton et al . ( 2001 ) describe the cuts in @xmath45 one must apply in the other bands to obtain complete magnitude limited samples .
they also note that @xmath83 .
since we will almost always be working with @xmath40 rather than @xmath45 , our cuts are slightly different from theirs ; the cuts we use to define magnitude limited samples in the different bands are summarized in table [ tab : photerr ] ( also see figure [ fig : appmz ] ) . at @xmath84 , however , most early - type galaxies in the sdss database were targeted using different selection criteria than were used for the main sdss galaxy sample ( see strauss et al . 2002 ; they make - up the luminous red galaxy sample described by eisenstein et al .
2001 ) . in the interest of keeping our sample as close to being magnitude limited as possible , we restricted our sample to @xmath85 .
in addition , one might expect an increasing fraction of the early - type population at higher redshifts to have emission lines : if so , then our removal of emission line objects amounts to a small but redshift dependent selection effect .
since our sample is restricted to @xmath85 , this bias should be small .
= before moving on , it is worth pointing out that there is a morphologically based selection cut which we could have made but did nt .
elliptical galaxies are expected to have axis ratios greater than about @xmath86 ( e.g. , binney & tremaine 1987 ) .
since we have axis ratio measurements of all the objects in our sample , we could have included a cut on @xmath38 .
the bottom right panel of figure [ abscatter ] shows the distribution axis ratios @xmath38 in our @xmath34 sample : about 20% of the objects in it have @xmath87 .
( the spike at @xmath88 is artificial
the pipeline reports large uncertainties in the fitted value of @xmath38 for these objects . )
our combination of cuts on the shapes of the light profiles and spectral features mean that these objects are unlikely to be late - type galaxies .
indeed , a visual inspection of a random sample of the objects with axis ratios smaller than 0.6 shows that they look like s0s .
the bottom left panel of figure [ abscatter ] shows that @xmath38 does not correlate with color : in particular , the colors of the most flattened objects are not bluer than in the rest of the sample .
also , recall that all objects with @xmath89 arcsec were visually inspected and these , despite having @xmath87 ( top left panel ) , did not appear peculiar .
in addition , @xmath38 does not correlate with surface brightness or apparent magnitude .
galaxies with small angular sizes @xmath39 are assigned large values of @xmath38 only slightly more often than average ( top left panel ; the median @xmath38 is 0.79 , 0.78 , 0.76 , and 0.7 for @xmath39 in the range 12 , 23 , 34 and greater than 4 arcsec ) .
however , there is a weak trend for the objects at higher @xmath33 to be rounder ( middle right panel ) .
this trend with @xmath33 may be related to the magnitude limit of our sample rather than reflecting problems associated with the fits to the observed light profiles .
( the two bands at @xmath90 and @xmath91 show that the effects of large scale clustering in our sample are significant . )
section [ landr ] describes how we convert from the observed half - light radius @xmath39 to an estimate of the physical half - light radius @xmath21 , and section [ vdisp ] discusses how we estimate velocity dispersions .
objects with smaller values of @xmath21 tend to be more flattened ( top right panel ) , and to have slightly smaller velocity dispersions ( middle left ) ; because of the magnitude limit , these objects drop out of our sample at higher redshifts . requiring
that @xmath92 would remove such objects from our sample completely . in paper
iii we study the fundamental plane populated by the galaxies in this sample .
excluding all objects with @xmath93 has no effect on the shape of this plane .
so , in the interests of keeping our sample as close to being magnitude limited as possible , we chose not to make an additional selection cut on @xmath38 .
= = figure [ fig : appmz ] shows the distribution of the @xmath94 apparent magnitudes in our sample as a function of redshift .
the dashed lines in each panel show the magnitude limits summarized in table [ tab : photerr ] ; the magnitude limited samples span the redshift range @xmath95 .
to convert the apparent magnitude @xmath96 to an absolute magnitude @xmath97 we must assume a particular cosmology and account for the fact that at different redshifts an observed bandpass corresponds to different restframe bands ( the k - correction ) .
we write the hubble constant today as @xmath98 km s@xmath23mpc@xmath23 and use @xmath99 .
most of our sample is at @xmath100 km s@xmath23 ; since line - of - sight peculiar velocities are not expected to exceed more than a few thousand km s@xmath23 , we feel that it is reasonable to assume that all of a galaxy s redshift is due to the hubble recession velocity .
this means that we can compute the absolute magnitude in a given band by @xmath101 - 25 - k(z)$ ] , where @xmath96 is the apparent magnitude , @xmath102 is the luminosity distance in mpc ( from , e.g. , weinberg 1972 ; hogg 1999 ) , and @xmath103 is the k - correction for the band . because we have five colors and a spectrum for each galaxy , we could , in principle , compute an empirical k - correction for each galaxy .
this requires a good understanding of the accuracy of the sdss photometry and spectroscopy , and should be possible when the survey is closer to completion . rather than follow the procedure adopted by the 2dfgrs ( madgwick et al . 2002 ) , or a procedure based on finding the closest template spectrum to each galaxy and using it to compute the k - correction ( e.g. , lin et al .
1999 for the cnoc2 survey ) , we use a single redshift dependent template spectrum to estimate the k - correction . in effect , although this allows galaxies at different redshift to be different , it ignores the fact that not all galaxies at the same redshift are alike . as a result ,
the absolute luminosities we compute are not as accurate as they could be , and this can introduce scatter in the various correlations we study below . of course , using a realistic k - correction is important , because inaccuracies in @xmath103 can masquerade as evolutionary trends . for the reasons discussed more fully in appendix [ kcorrs ] ,
our k - corrections are based on a combination of bruzual & charlot ( 2003 ) and coleman , wu & weedman ( 1980 ) prescriptions .
specifically , the k - corrections we apply were obtained by taking a bruzual & charlot model for a @xmath104 object which formed its stars with an imf given by kroupa ( 2001 ) in a single solar metallicity and abundance ratio burst 9 gyr ago , computing the difference between the k - correction when evolution is allowed and ignored , and adding this difference to the k - corrections associated with coleman , wu & weedman early type galaxy template .
the results which follow are qualitatively similar for a number of other k - correction schemes ( see appendix [ kcorrs ] for details ) .
in addition to correcting the observed apparent magnitudes to absolute magnitudes , we must also apply two corrections to convert the ( seeing corrected ) effective angular radii , @xmath39 , output by the sdss pipeline to physical radii .
first , we define the equivalent circular effective radius @xmath105 .
( although the convention is to use @xmath106 to denote the effective radius , we feel that the notation @xmath107 is better , since it emphasizes that the radius is an effective circular , rather than elliptical aperture . )
the reason we must make a second correction is shown in figure [ angre ] .
the different panels show the distribution of ( seeing corrected ) effective angular sizes @xmath39 of the galaxies in the different bands .
notice that @xmath39 for most of the objects is larger than the typical seeing scale of 1.5 arcsec : we discuss this further in section [ checkphoto ] .
comparison of the mean sizes in the different panels ( the text in the top right corner of each panel ) shows that galaxies appear slightly larger in the bluer bands .
because our sample covers a reasonably large range in redshift , this trend means we must correct the effective sizes to a fixed restframe wavelength . therefore , when converting from effective angular size @xmath107 to effective physical size @xmath21 we correct @xmath107 ( and the petrosian radii @xmath48 and @xmath49 ) in each band by linearly interpolating from the observed bandpasses to the central rest wavelength of each filter .
the typical correction is of the order of 4% , although it is sometimes as large as 10% . in this respect , this correction is analogous to the k - correction we would ideally have applied to the magnitude and surface brightness of each galaxy .
note that the dependence of size on wavelength is also important for analyses of color gradients and the color
magnitude relation presented in paper iv .
our study will also require the effective surface brightness @xmath108 , where @xmath109 is the mean surface brightness within the effective radius @xmath21 ( as opposed to the surface brightness at @xmath21 ) .
in particular , we set @xmath110 .
note that this quantity is k - corrected , and also corrected for the cosmological @xmath111 dimming .
our earlier remarks about the k - correction are also relevant here .
= there is some discussion in the literature about the effects of seeing on estimates of the effective radius , and hence the effective surface brightness .
these effects are unimportant if the seeing scale , @xmath112 ( the fwhm in arcsec ) , is an order of magnitude smaller than the effective radius ( e.g. , saglia et al .
1993 , 1997 ) .
figure [ fig : seeinghist ] shows histograms of @xmath113 in the @xmath66 , @xmath34 , @xmath61 and @xmath67 bands : the effective radius is typically only a factor of two larger than the seeing .
therefore , the accuracy of our estimates of the effective radii depend crucially on the correction for seeing being accurate .
= to illustrate that the de vaucouleurs model fits output by the sdss photometric pipeline are reasonable , figure [ fig : seeing ] shows the ratio of @xmath114 $ ] times the half - light radius to @xmath112 in the @xmath66 and @xmath34 bands ( top and bottom panels ) as a function of redshift , in a few volume limited catalogs .
the catalog limits were chosen to be 0.5 mags wide , so the galaxies in any given catalog should all be quite similar to each other .
( the factor @xmath115 removes the expected scaling of angular diameter with redshift . ) it is important to use volume limited samples to make the measurement , because size is known to correlate with luminosity ( more luminous galaxies are larger on average ) , and in a magnitude limited sample such as ours , only the more luminous objects are present at higher redshifts . in the panels on the left
, the half - light radius is estimated by @xmath39 , which has been corrected for seeing , whereas the panels on the right use @xmath48 , which is not seeing - corrected .
seeing tends to increase the half - light radius , and indeed , in the panel on the right , there is a clear trend for the half - light radii to increase with redshift .
this trend is absent in the panels on the left , suggesting that the fitting procedure accounts quite well for the effects of seeing .
the panels on the left even show a slight decrease with redshift we argue later ( see discussion of figure [ fig : xzr ] in section [ catalog ] ) that this may be a consequence of luminosity evolution .
= we have argued that our estimates of @xmath39 are probably not strongly compromised by seeing . in our dataset , @xmath39 is about a factor of two smaller than the radius @xmath116 at which the fitted surface brightness is 10% of the sky level .
therefore , errors in the sky subtraction may also compromise our estimates of @xmath39 .
if the same object is moved to higher redshift but the sky level is kept fixed , then one expects @xmath117 because cosmological surface brightness dimming scales as @xmath111 .
therefore , a plot of @xmath118 versus @xmath119 , measured from volume limited catalog , should have a slope of @xmath120 .
however , because each volume limited catalog spans a large range in redshift , we must also account for the surface brightness dimming associated with the fact that redshift dependent @xmath121-corrections must be made to estimate fixed restframe luminosities .
therefore , we expect the scaling with redshift to be slightly stronger : @xmath122
. figure [ fig : re10sky ] shows how @xmath123 depends on @xmath119 , in the same volume limited catalogs as before ; solid lines show the mean trend in the faintest ( top ) to the most luminous ( bottom ) catalogs .
comparison with the dashed line , which has a slope of @xmath124 , shows that the scaling with redshift is close to that expected .
( note that it is important to use volume limited catalogs to make the measurement : the trend in any given volume limited subsample is shallower than the trend one would infer by first combining together all the subsamples and then making the measurement . to see why this happens ,
suppose that the mean size at fixed luminosity scales as @xmath125 .
then the magnitude limit induces a correlation which should scale approximately as @xmath126 .
if @xmath127 then @xmath39 increases with @xmath33 , so @xmath128 decreases with @xmath33 simply because of the magnitude limit . )
some of the small differences in slope between the dashed and solid lines can be attributed to the effects of luminosity evolution . by fixing @xmath129 , we select objects which are slightly smaller at higher @xmath33 ( c.f .
figure [ fig : xzr ] ) , hence have slightly smaller values of @xmath39 and so larger values of @xmath128 than expected ; so they appear to decline less rapidly than const@xmath130 .
luminosity evolution is expected to be stronger in the bluer bands , and the @xmath66 band relation does appear to be slightly shallower than the other bands .
thus , as was the case for @xmath113 , we see redshift dependent trends in @xmath128 which we believe are physically reasonable .
the sky - subtraction procedure does not appear to have introduced any obvious artifacts into our estimates of @xmath39 .
finally , we turn to an inspection of how well the de vaucouleurs model actually fits the observed surface brightness profiles .
figure [ fig : devfits ] shows angular averaged surface brightness profiles in the @xmath66 , @xmath34 , @xmath61 and @xmath67 bands ( bottom to top in each of the upper panels ) , for a random selection of galaxies in our sample , ordered by redshift and size .
the text in each panel shows the plate number , mjd , and fiber i d number of the observation , the redshift @xmath33 , the fitted apparent @xmath34 magnitude @xmath131 , the fitted axis ratio @xmath38 , and the effective physical radius @xmath21 , computed from the fitted angular radius @xmath39 , the fitted axis ratio and the redshift , as described previously .
the solid lines show de vaucouleurs profiles which , when convolved with the pipeline s model for the seeing , provide the best fit to the symbols ( because the fits are actually not to these azimuthally averaged one - dimensional profiles , but to the two dimensional profile , they serve mainly as guides to the eye . )
the horizontal lines show the surface brightness at 1% of the sky in the different bands .
long , medium and short tick marks on these lines show @xmath112 , @xmath39 and @xmath132 , the radius at which the fitted surface brightness equals 10% of the sky .
the bottom panels show the residuals ( observed minus fit ) in the different bands ( crosses , stars , diamonds and triangles show the @xmath66 , @xmath34 , @xmath61 and @xmath67 band results , respectively ) .
the residuals are systematically low on small scales , but comparison with the large tick marks , which indicate the size of the seeing disk , show that this discrepancy is because we are comparing fits which have been corrected for seeing with the observed surface brightnesses which are not seeing - corrected . on larger scales ( typically larger than twice the half - light radii ) , the residuals are also large , indicating that there may be problems with the sky subtraction . in linear flux units ,
the residuals tend to a constant value which is typically smaller than 0.3% of the sky level ( recall that the horizontal lines in the panels show the surface brightness at 1% of sky ) .
although this is not an entirely fair comparison , since the fits are not to performed on these azimuthally averaged one - dimensional profiles , but to the two dimensional profile , this test suggests that the fit is reliable , but that the zero - point level of the sky may be systematically biased by @xmath133% . note that the residuals are larger in the @xmath66 and @xmath67 bands ; but this is consistent with the larger errors we quote on the photometry in these bands .
figure [ fig : images ] shows the @xmath34 images of the same objects whose profiles are shown in figure [ fig : devfits ] .
the images are each @xmath134 arcsec@xmath135 ( i.e. , 30 pixels on a side ) .
one might expect some of the worst fits are due to the presence of a disk component which our de vaucouleur model does not account for .
however , comparison of the images and the fits shows no obvious correlation between goodness - of - fit and , e.g. , @xmath38 .
= 0.9 = = fig .
[ fig : devfits ] . continued .
redshift range @xmath136 .
[ fig : images ] .
continued .
redshift range @xmath136 .
[ fig : devfits ] . continued .
redshift range @xmath137 .
[ fig : images ] . continued .
redshift range @xmath137 .
[ fig : devfits ] . continued .
redshift range @xmath138 .
[ fig : images ] . continued .
redshift range @xmath138
. = fig .
[ fig : devfits ] . continued .
redshift range @xmath139 .
[ fig : images ] . continued .
redshift range @xmath139 .
[ fig : devfits ] . continued .
redshift range @xmath140 .
[ fig : images ] . continued .
redshift range @xmath140 . in summary
, we have performed a number of tests of the sdss photometric pipeline reductions and have found little evidence that they are significantly biased .
therefore , in what follows , the apparent magnitudes we use are those output by the sdss photometric pipeline .
to convert to luminosities , we k - correct ( and use the inverse square law ) .
the angular sizes are output from the sdss photometric pipeline .
we correct these to a fixed restframe size , and we correct for @xmath141 . both these corrections are standard . thereafter , surface brightness is defined from the restframe luminosity and size with no additional corrections . the uncertainties on the photometric parameters we use in this paper , and the magnitude limits in each band , are summarized in table [ tab : photerr ] .
there is some subtlety in obtaining them from the format available in the sdss database ; this is discussed in appendix [ errors ] .
.photometric parameters and median errors of the objects in our sample . + [ cols="^,^,^,^,^,^,^,^,^,^ " , ] [ tab : photerr ] the sdss pipeline does not provide an estimate of the line - of - sight velocity dispersion , @xmath9 , within a galaxy , so we must compute it separately for the early - type galaxy sample .
the observed velocity dispersion @xmath9 is the result of the superposition of many individual stellar spectra , each of which has been doppler shifted because of the star s motion within the galaxy .
therefore , it can be determined by analyzing the integrated spectrum of the whole galaxy .
a number of objective and accurate methods for making velocity dispersion measurements have been developed ( sargent et al . 1977 ; tonry & davis 1979 ; franx , illingworth & heckman 1989 ; bender 1990 ; rix & white 1992 ) .
each of these methods has its own strengths , weaknesses , and biases .
appendix [ vmethods ] describes how we combined these different techniques to estimate @xmath9 for the galaxies in our sample .
the velocity dispersion estimates we use in what follows are obtained by fitting the restframe wavelength range @xmath142 , and then averaging the estimates provided by the _ fourier - fitting _ and _ direct - fitting _ methods to define what we call @xmath143 .
( we do not use the cross - correlation estimate because of its behavior at low @xmath55 as discussed in appendix [ vmethods ] . note that the @xmath55 of the sdss spectra depend on wavelength . ) the error on @xmath143 is determined by adding in quadrature the errors on the two estimates ( i.e. , the fourier - fitting and direct - fitting ) which we averaged .
the resulting error is between @xmath144 dex and 0.06 dex , depending on the signal - to - noise of the spectra , with a median value of @xmath145 dex .
a few galaxies in our sample have been observed more than once .
the scatter between different measurements is @xmath146 dex , consistent with the amplitude of the errors on the measurements ( see figure [ fig : vrptd ] ) .
based on the typical @xmath55 of the sdss spectra and the instrumental resolution , we chose 70 km s@xmath23 as a lower limit on the velocity dispersions we use in this paper . following jrgensen et al .
( 1995 ) and wegner et al .
( 1999 ) , we correct @xmath143 to a standard relative circular aperture defined to be one - eighth of the effective radius : @xmath147 where @xmath148 arcsec and @xmath107 is the effective radius of the galaxy measured in arcseconds . in principle , we should also account for the effects of seeing on @xmath143 , just as we do for @xmath107 .
however , because the aperture correction depends so weakly on @xmath107 ( as the 0.04 power ) , this is not likely to be a significant effect . in any case , most galaxies in our sample have @xmath149 arcsecs ( see figure [ angre ] ) . note
that this correction assumes that the velocity dispersion profiles of early - type galaxies having different @xmath107 are similar . at the present time , we do not have measurements of the profiles of any of the galaxies in our sample , so we can not test this assumption . in paper ii of this series
we will argue that the galaxies in our sample evolve very little ; this means that if we select galaxies of the same luminosity and effective radius , then a plot of velocity dispersion versus redshift of these objects should allow us to determine if the aperture correction above is accurate .
the results of this exercise are presented in appendix [ vr ] . to illustrate the typical quality of the spectra in our sample , figure [ fig : spectra ] shows spectra of the galaxies which lie along the diagonals in figure [ fig : devfits ] ( panels from top to bottom here are for objects which are top left to bottom right in figure [ fig : devfits ] ) .
all the spectra in our sample have pca classification numbers @xmath62 , typical of early - type galaxy spectra ( connolly & szalay 1999 ) , although a few of the spectra do show weak emission lines .
= = fig .
[ fig : spectra ] . continued .
redshift range @xmath136 .
[ fig : spectra ] . continued .
redshift range @xmath137 .
[ fig : spectra ] . continued .
redshift range @xmath138 .
[ fig : spectra ] . continued .
redshift range @xmath139 .
[ fig : spectra ] . continued .
redshift range @xmath140 . the typical signal - to - noise ratio of the spectra in our sample is 16 , although @xmath55 does vary with wavelength .
therefore , individual spectra can not be used to make reliable estimates of , e.g. , the lick indices . however , by co - adding spectra of similar objects ( i.e. , galaxies with similar luminosities , sizes , velocity dispersions , and redshifts ) it is possible to construct composite spectra which have considerably better @xmath55 .
this is the subject of paper iv . in subsequent papers
, we will be interested in how the properties of the galaxies in our sample depend on their local environment .
to do so , we must come up with a working definition of environment .
the set of galaxies in the sdss photometric database is much larger than those for which the survey actually measures redshifts .
some of these galaxies may well lie close to galaxies in our sample , in which case they will contribute to the local density .
we would like to find some way of accounting for such objects when we estimate the local density . for a subset of the galaxies in our sample ,
the colors expected of a passively evolving early - type were used to select a region in @xmath66@xmath34 versus @xmath34@xmath61 color space at the redshift of the galaxy of interest .
all galaxies within 0.1 magnitudes in color of this point were included if they were : a ) within @xmath150 mpc of the main galaxy , and b ) brighter than @xmath151 in @xmath152 .
( the box in color space is sufficiently large that the difference between this techinique , and using the observed colors themselves to define the selection box is not important . )
these two cuts are made assuming every galaxy in the color - color range is at the redshift of the galaxy of interest .
the end result of this is that each galaxy in the subsample is assigned a number of neighbors .
note that , because of the selection on color , our estimate of the local density is actually an estimate of the number of neighbors which have the same colors as early - type galaxies . in the papers which follow
, we will often present results for different redshift bins .
when we do , it is important to bear in mind that this procedure for assigning neighbours is least secure in the lowest redshift bin ( typically @xmath153 ) .
figure [ lrvden ] shows how the luminosities , surface brightnesses , sizes , velocity dispersions and ( a combination of the ) axis ratios depend on environment .
the different symbols for each bin in density show averages over galaxies in different redshift bins : circles , diamonds , triangles , and squares are for galaxies with redshifts in the range @xmath154 , @xmath155 , @xmath156 , and @xmath157 .
error bars show the error in the mean value for each bin .
symbols for the higher redshift bins have been offset slightly to the right .
= 1.2 = for any given set of symbols , the bottom right panel shows that the mean redshift in each bin in density is not very different from the mean redshift averaged over all bins .
this suggests that our procedure for estimating the local densities is not biased .
the other panels show corresponding plots for the other observed parameters . when the number of near neighbours is small ,
the luminosities , sizes and velocity dispersions all increase slightly as the local density increases , whereas the surface brightnesses decrease slightly .
all these trends are very weak .
the bottom right panel shows @xmath158 , where @xmath38 is the axis ratio .
[ some authors ( e.g. de zeeuw & franx 1991 ) suggest that this is a good approximation of the ratio of rotational to random motions within the galaxy , ( @xmath159)@xmath160 , for oblate , rotationally flattened galaxies .
bright ellipticals have been shown to have a mean ( @xmath159)@xmath161(@xmath159)@xmath162 . ] there are no obvious trends with environment .
it is difficult to say with certainty that the trends with environment in the top four panels of figure [ lrvden ] are significant . a more efficient way of seeing if the properties of galaxies depend on environment is to show the residuals from the fundamental plane .
this is done in paper iii .
figure [ fig : pie ] shows a redshift - space pie - diagram distribution of our sample .
most of the sample is at low declination ( @xmath163 ) ; in addition , there are three wedges from three different disconnected regions on the sky .
red and blue symbols denote galaxies which were classified as being in dense and underdense regions ( as described in section [ environ ] below ) , whereas black symbols show galaxies in groups of intermediate richness , or for which the local density was not determined .
the observed parameters of each galaxy in our early - type galaxy sample are given in table [ tab : obs ] .
the different columns show the ra , dec , redshift @xmath33 and its error estimate , the signal - to - noise ratio @xmath55 of the spectrum , the apparent half light radius @xmath39 and error estimate , the fitted apparent magnitude @xmath94 and error estimate , the model magnitude @xmath44 and error estimate , and the axis ratio @xmath38 and error estimate .
the photometric parameters in this table are those measured in the @xmath34 band . the complete version of this table , and similar tables of the catalog parameters in the @xmath66 , @xmath61 and @xmath67 bands are available in the electronic edition of the journal .
because we will be interested in the distribution of colors for a galaxy sample defined in , e.g. , the @xmath34 band , the tables present the parameters of the same galaxies in all the bands . to select the magnitude limited samples we use in subsequent papers refer to table [ tab : photerr ] .
the errors @xmath164 and @xmath165 are the error estimates output by the sdss photometric pipeline . as discussed in appendix [ errors ] , they are _ not _ appropriate estimates of the errors in @xmath40 and @xmath166 .
our procedure for making more appropriate error estimates is described in appendix [ errors ] .
ccccccccccccc [ tab : obs ] 113.810738 & 36.307522 & 0.25100 & 0.00009 & 12 & 3.45 & 0.15 & 17.234 & 0.009 & 17.234 & 0.009 & 0.82 & 0.03 + 115.321098 & 37.511093 & 0.08513 & 0.00008 & 19 & 2.55 & 0.04 & 16.400 & 0.009 & 16.400 & 0.009 & 0.99 & 0.01 + 114.201271 & 36.971313 & 0.07832 & 0.00008 & 15 & 2.16 & 0.05 & 17.202 & 0.009 & 17.202 & 0.009 & 0.71 & 0.02 + 114.306282 & 37.111927 & 0.13049 & 0.00008 & 20 & 2.48 & 0.05 & 16.828 & 0.008 & 16.828 & 0.008 & 0.68 & 0.01 + 113.294510 & 36.370544 & 0.21756 & 0.00007 & 13 & 2.18 & 0.08 & 17.451 & 0.008 & 17.451 & 0.008 & 0.80 & 0.03 + 113.376007 & 36.478168 & 0.13365 & 0.00008 & 14 & 2.45 & 0.08 & 17.304 & 0.008 & 17.304 & 0.008 & 0.63 & 0.02 + 113.335358 & 36.648064 & 0.14016 & 0.00007 & 21 & 2.01 & 0.04 & 16.871 & 0.006 & 16.871 & 0.006 & 0.68 & 0.01 + 113.902443 & 37.023502 & 0.20225 & 0.00008 & 12 & 4.17 & 0.17 & 16.998 & 0.011 & 16.998 & 0.011 & 0.96 & 0.03 + 114.022346 & 37.099522 & 0.17845 & 0.00008 & 14 & 2.07 & 0.08 & 17.453 & 0.011 & 17.453 & 0.011 & 0.76 & 0.03 + 113.925407 & 37.113167 & 0.18365 & 0.00009 & 15 & 2.71 & 0.08 & 17.053 & 0.010 & 17.053 & 0.010 & 0.80 & 0.02 + 114.060677 & 37.211426 & 0.09610 & 0.00008 & 13 & 2.11 & 0.07 & 17.279 & 0.010 & 17.279 & 0.010 & 0.85 & 0.03 + 113.827904 & 37.380821 & 0.18387 & 0.00008 & 12 & 4.82 & 0.18 & 16.898 & 0.009 & 16.898 & 0.009 & 0.97 & 0.03 + 113.945900 & 37.708260 & 0.07476 & 0.00006 & 19 & 2.98 & 0.05 & 16.882 & 0.009 & 16.882 & 0.009 & 0.34 & 0.01 cccccccccccccc
[ tab : phys ] 113.810738 & 36.307522 & 2.501 & 0.034 & 0.979 & 0.020 & 19.873 & 0.060 & -23.587 & 0.033 & -23.587 & 0.033 & 0.309 & 4 + 115.321098 & 37.511093 & 2.133 & 0.029 & 0.600 & 0.007 & 19.918 & 0.020 & -21.648 & 0.018 & -21.648 & 0.018 & 0.104 & 4 + 114.201271 & 36.971313 & 2.055 & 0.045 & 0.414 & 0.012 & 19.993 & 0.034 & -20.643 & 0.021 & -20.643 & 0.021 & 0.093 & 3 + 114.306282 & 37.111927 & 2.363 & 0.022 & 0.664 & 0.010 & 19.620 & 0.030 & -22.268 & 0.019 & -22.268 & 0.019 & 0.159 & 2 + 113.294510 & 36.370544 & 2.307 & 0.040 & 0.871 & 0.017 & 19.944 & 0.052 & -22.978 & 0.029 & -22.978 & 0.029 & 0.268 & 4 + 113.376007 & 36.478168 & 2.319 & 0.026 & 0.619 & 0.016 & 19.810 & 0.046 & -21.851 & 0.026 & -21.851 & 0.026 & 0.163 & 2 + 113.335358 & 36.648064 & 2.266 & 0.024 & 0.541 & 0.010 & 18.866 & 0.028 & -22.407 & 0.017 & -22.407 & 0.017 & 0.173 & 5 + 113.902443 & 37.023502 & 2.352 & 0.045 & 1.110 & 0.019 & 20.886 & 0.059 & -23.229 & 0.034 & -23.229 & 0.034 & 0.244 & 999 + 114.022346 & 37.099522 & 2.388 & 0.038 & 0.749 & 0.018 & 19.872 & 0.054 & -22.440 & 0.031 & -22.440 & 0.031 & 0.212 & 7 + 113.925407 & 37.113167 & 2.433 & 0.029 & 0.856 & 0.014 & 19.934 & 0.041 & -22.914 & 0.025 & -22.914 & 0.025 & 0.217 & 6 + 114.060677 & 37.211426 & 2.173 & 0.034 & 0.542 & 0.015 & 20.219 & 0.044 & -21.058 & 0.027 & -21.058 & 0.027 & 0.114 & 1 + 113.827904 & 37.380821 & 2.347 & 0.040 & 1.094 & 0.018 & 20.965 & 0.055 & -23.071 & 0.030 & -23.072 & 0.030 &
0.217 & 4 + 113.945900 & 37.708260 & 2.090 & 0.028 & 0.394 & 0.009 & 19.687 & 0.024 & -20.851 & 0.019 & -20.851 & 0.019 & 0.088 & 3 table [ tab : phys ] shows restframe quantities we computed from the quantities in table [ tab : obs ] .
the different columns show the ra , dec , aperture corrected velocity dispersion @xmath167 and its error estimate , the physical size @xmath168 and error estimate , the surface brightness @xmath169 and error estimate , the absolute magnitude @xmath170 and error estimate , the absolute model magnitude @xmath171 and error estimate , the k - correction , and the number of neighbours which we use as an estimate of the local density . as for table [ tab : obs ] , a complete version of this table is in the electronic edition of the journal , as are similar tables showing parameters in the @xmath66 , @xmath61 and @xmath67 bands .
( the errors in @xmath170 and @xmath172 listed in this table were derived from the estimates @xmath173 of the previous table . as discussed in appendix [ errors ] , they are also the appropriate estimates for the error on @xmath40 and @xmath174 . )
= some properties of our @xmath34 sample as a function of redshift @xmath33 are shown in figure [ fig : xzr ] .
the panels show the k - corrected absolute magnitude @xmath97 , the k - corrected effective surface brightness @xmath169 , also corrected for cosmological surface brightness dimming , the effective circular radius @xmath21 in @xmath175 kpc , corrected to a standard restframe wavelength , the aperture corrected velocity dispersion @xmath9 in km s@xmath23 , and two quantities which are related to an effective mass and density , all plotted as a function of redshift .
the bold lines in the top left panel show the effect of the apparent magnitude cuts .
there is , in addition , a cut at small velocity dispersion ( @xmath176 km s@xmath23 ) which , for our purposes here , is mostly irrelevant .
the apparent magnitude cuts imply complex @xmath33-dependent cuts on the other parameters we observe . in the papers which follow ,
we attempt to account correctly for the selection effects that the magnitude cuts imply . for small intervals in luminosity , our sample is complete over a reasonably large range in redshifts .
to illustrate , the thin boxes in the top - left panel show bins in absolute magnitude of width 0.5 mags over which the sample is complete .
the solid lines in the other panels show how the median surface - brightnesses , sizes , velocity dispersions , masses and densities of galaxies in each @xmath34 luminosity bin change as a function of redshift .
although all these quantities depend on luminosity , the figure shows that , at fixed luminosity , there is some evidence for evolution : at fixed luminosity the average surface brightness is brightening .
the size at fixed luminosity decreases at a rate which is about five times smaller than the rate of change of @xmath169 .
this suggests that it is the luminosities which are changing and not the sizes .
( to see why , suppose that the average size at fixed absolute magnitude is @xmath177\rangle = s[m - m_*(z)]$ ] , where @xmath178 and @xmath179 are characteristic values , and @xmath180 is the slope of this mean relation .
if the characteristic luminosity increases with @xmath33 , but the characteristic size remains constant , @xmath181 , and the slope of the relation also does not change , then the mean size at fixed @xmath97 decreases with @xmath33 .
the surface brightness is @xmath182 , so the mean @xmath169 at fixed @xmath97 changes five times faster than the mean @xmath21 , at fixed luminosity .
this argument is also relevant to figures [ fig : seeing ] and [ fig : re10sky ] , since those plots use @xmath39 which differs from @xmath21 by two small correction factors [ c.f .
section [ landr ] ] . ) in paper ii we argue that this trend is qualitatively consistent with that expected of a passively evolving population .
this is the first of four papers which study the properties of early - type galaxies at relatively low redshifts ( @xmath85 ) .
this paper describes how we selected the sample from the sdss database using objective criteria and present the measured photometric and spectroscopic parameters for @xmath8 early - type galaxies ( table [ tab : obs ] ) .
the database contains the redshift of these galaxies as well as reliable measurements of a number of photometric properties ( luminosities , sizes , surface brightnesses , colors ) in the @xmath66 , @xmath34 , @xmath61 and @xmath67 bands .
( data in the @xmath183 band is also available , but , because it is noisier , we did not use it here . ) the galaxies in our sample span the redshift range @xmath95 .
therefore , differences between observed and restframe wavelengths are not negligible . we discussed a number of methods for estimating the appropriate k - correction which must be applied to obtain luminosities which sample the same restframe wavelength range ( appendix [ kcorrs ] ) .
since the half - light radii also depend on wavelength ( figure [ angre ] ) , we also corrected these to sizes at fixed restframe wavelengths .
finally , we estimated the internal velocity dispersions of these galaxies ( appendix [ vmethods ] ) , which , following common practice , we corrected to an aperture of @xmath184 ( c.f .
equation [ appcorr ] ) .
the resulting catalog listing the restframe photometric and spectroscopic parameters of @xmath8 early - type galaxies ( table [ tab : phys ] ) is useful for measuring correlations between these various observables , such as the faber - jackson @xmath185 relation , the kormendy @xmath186 relation , the @xmath187 relation , the fundamental plane @xmath188 relation , and the color magnitude relation .
the sample is essentially magnitude limited ( table [ tab : photerr ] , figure [ fig : appmz ] ) , and the galaxies in it populate a range of environments ( figure [ fig : pie ] ) . therefore , the sample can be used for studying how the correlations discussed above depend on environment and redshift .
for instance , figure [ fig : xzr ] shows evidence for weak evolution in the luminosities , and figure [ lrvden ] shows evidence for some weak environmental dependences .
these will be quantified in papers ii and iii .
neither the k - corrections nor the aperture corrections we apply to the velocity dispersions are ideal . by the time the sloan digital sky survey is complete , the uncertainties in the k - corrections , which prevent us at the present time from making precise quantitative statements about the evolution of the luminosities and colors ,
will be better understood .
in addition , the size of the sample will have increased by more than an order of magnitude .
this larger sample will allow us to use the data directly to estimate the correct aperture correction which should be applied to the velocity dispersions .
this is discussed more fully in appendix [ vr ] .
we would like to thank s. charlot for making his stellar population synthesis predictions for the sdss filters available to the collaboration and n. benitez for making his package available .
we thank m. strauss for helpful discussions .
funding for the creation and distribution of the sdss archive has been provided by the alfred p. sloan foundation , the participating institutions , the national aeronautics and space administration , the national science foundation , the u.s .
department of energy , the japanese monbukagakusho , and the max planck society .
the sdss web site is http://www.sdss.org/. the sdss is managed by the astrophysical research consortium ( arc ) for the participating institutions .
the participating institutions are the university of chicago , fermilab , the institute for advanced study , the japan participation group , the johns hopkins university , los alamos national laboratory , the max - planck - institute for astronomy ( mpia ) , the max - planck - institute for astrophysics ( mpa ) , new mexico state university , university of pittsburgh , princeton university , the united states naval observatory , and the university of washington .
baum , w. a. 1959 , pasp , 71 , 106 bender , r. 1990 , a&a , 229 , 441 bender , r. , saglia , r. p. , & gerhard , o. e. 1994 , mnras , 269,785 benitez , n. 2000 , apj , 536 , 571 bernardi , m. , renzini , a. , da costa , l. n. , wegner , g. , alonso , m. v. , pellegrini , p. s. , rit , c. & willmer , c. n. a. 1998 , apjl , 508 , 143 bernardi , m. , alonso , m. v. , da costa , l. n. , willmer , c. n. a. , wegner , g. , pellegrini , p. s. , rit , c. & m. a. g. maia 2002a , aj , 123 , 2159 bernardi , m. , alonso , m. v. , da costa , l. n. , willmer , c. n. a. , wegner , g. , pellegrini , p. s. , rit , c. & m. a. g. maia 2002b , aj , 123 , 2990 bernardi , m. , sheth , r. k. , annis , j. et al .
2003a , aj , _ in pree _
bernardi , m. , sheth , r. k. , annis , j. et al .
2003b , aj , _ in press _
bernardi , m. , sheth , r. k. , annis , j. et al .
2003c , aj , _ in press _
bingelli , b. , sandage , a. , tarenghi , m. 1984 , aj , 89 , 64 binney , j. , & tremaine , s. 1987 , galactic dynamics ( princeton : princeton univ . press ) blakeslee , j. p. , lucey , j. r. , barris , b. j. , hudson , m. j. , & tonry , j. l. 2001 , mnras , 327 , 1004 blanton , m. r. , dalcanton , j. , eisenstein , d. et al .
, 2001 , aj , 121 , 2358 bressan , a. , chiosi , c. , & fagotto , f. 1994 , apjs , 94 , 63 brodie , j. p. , & hanes , d. a. 1986 , apj , 300 , 258 bruzual , g. , & charlot , s. 2003 , in preparation burbidge , e. m. , burbidge , g. r. , & fish , r. a. 1961 , apj , 134,251 burstein , d. , bender , r. , faber , s. m. , & nolthenius , r. 1997 , aj , 114 , 1365 coleman , g. d. , wu , c .- c . , &
weedman , d. w. 1980 , apjs , 43 , 393 colless , m. , saglia , r. p. , burstein , d. , davies , r. l. , mcmahan , r. k. , & wegner , g. 2001 , mnras , 321 , 277 connolly , a. j. , & szalay , a. s. 1999 , aj , 117 , 2052 da costa , l. n. , bernardi , m. , alonso , m. v. , wegner , g. , willmer , c. n. a. , pellegrini , p. s. , rit , c. , & maia , m. a. g. 2000 , aj , 120 , 95 de vaucouleurs , g. 1948 , annales dastrophysique , 11 , 247 de zeeuw , t. & franx , m. 1991 , in annual review of astronomy and astrophysics .
29 , p. 239 - 274 .
palo alto , ca , annual reviews , inc . diaz , a. i. , terlevich e. , & terlevich , r. 1989 , mnras , 239 , 325 djorgovski , s. & davis , m. 1987 , , 313 , 59 dressler , a. 1984 , apj , 286 , 97 dressler , a. , lynden - bell , d. , burstein , d. , davies , r. l. , faber , s. m. , terlevich , r. j. , & wegner , g. 1987 , apj , 312 , 42 eisenstein , d.j . ,
annis , j. , gunn , j.e .
, et al . 2001 , aj , 122 , 2267 ellis , r. s. , smail , i. , dressler , a. , couch , w. j. , oemler , a. , butcher , h. , & sharples , r. m. 1997 , apj , 483 , 582 faber , s. m. , & jackson , r. 1976 , apj , 204 , 668 faber , s. m. , dressler , a. , davies , r. l. , burstein , d. , lynden - bell , d. , terlevich , r. j. , & wegner , g. 1987 , in nearly normal galaxies , from the planck time to the present , ed .
s. m. faber ( ny : springer ) , 175 faber , s. m. , wegner , g. , burstein , d. , davies , r. l. , dressler , a. , lynden - bell , d. , & terlevich , r. j. 1989 , apjs , 69 , 763 fish , r. a. 1964 , apj , 139 , 284 franx , m. , illingworth , g. d. , & heckman , t. 1989 , apj , 344 , 613 fukugita , m. , shimasaku , k. , & ichikawa , t. 1995 , pasp , 107 , 945 fukugita , m. , ichikawa , t. , gunn , j. e. , doi , m. , shimasaku , k. , & schneider , d. p. 1996 , aj , 111 , 1748 giovanelli , r. , haynes , m. p. , herter , t. , vogt , n.p .
, da costa , l.n . , freudling , w. , salzer , j.j . and wegner , g. 1997 , aj , 113 , 53 gunn , j.e . ,
carr , m.a .
, rockosi , c.m . , sekiguchi , m. , et al .
1998 , aj , 116 , 3040 hernquist , l. 1990 , apj , 356 , 359 hogg , d. 1999 , ( astro - ph/9905116 ) hogg , d.w .
, schlegel , d.j . ,
finkbeiner , d.p . , & gunn , j.e .
2001 , aj , 122 , 2129 jrgensen , i. , franx , m. & kjrgaard , p. 1995
, mnras , 276 , 1341 jrgensen , i. , franx , m. , & kjrgaard , p. 1996 , mnras , 280 , 167 kauffmann , g. 1996 , mnras , 281 , 487 kauffmann , g. & charlot , s. 1998 , mnras , 294 , 705 kelson , d. d. , illingworth , g. d. , van dokkum , p. g. , & franx , m. 2000 , apj , 531 , 184 kochanek , c. s. , pahre , m. a. , & falco , e. e. 2000 , ( astro - ph/0011458 ) kormendy , j. 1977 , apj , 218 , 333 kormendy , j. 1982 , saas - fee lectures 12 , p. 115
kroupa , p. 2001 , in asp conf .
228 , dynamics of star clusters and the milky way , ed .
s. deiters , b. fuchs , r. spurzem , a. just , and r. wielen ( san francisco : astr . spc . pac . ) , p. 187
kuntschner , h. 2000 , mnras , 2000 , 315 , 184 kuntschner , h. , lucey , j. r. , smith , r. j. , hudson , m. j. , & davies , r. l. 2001 , mnras , 323 , 615 larson , r. b. 1975 , mnras , 173 , 671 lin , h. , yee , h. k. c. , carlberg , r. g. , morris , s. l. , sawicki , m. , patton , d. r. , wirth , g. , & shepherd , c. w. 1999 , apj , 518 , 533 lupton , r. , gunn , j. e. , ivezi , z. , knapp , g. r. , & kent , s. 2001 , in asp conf .
238 , astronomical data analysis software and systems x , ed
. f. r. harnden , jr . , f. a. primini , and h. e. payne ( san
francisco : astr .
spc . pac . ) , p. 269
( astro - ph/0101420 ) madgwick , d. s. , lahav , o. , baldry , i. k. et al .
2002 , mnras , 333 , 133 pahre , m. a. , 1998 , pasp , 110 , 1249 pahre , m. a. , djorgovski , s. g. , & de carvalho , r. r. 1998a , aj , 116 , 1591 pahre , m. a. , de carvalho , r. r. , & djorgovski , s. g. 1998b , aj , 116 , 1606 petrosian , v. 1976 , apjl , 209 , 1 rix , h .- w .
, & white , s. d. m. 1992 , mnras , 254 , 389 roberts m. s. & haynes m. p. , 1994
, ara&a , 32 , 115 saglia , r. p. , bertschinger , e. , baggley , g. , burstein , d. , colless , m. , davies , r. l. , mcmahan , jr . , r. k. & wegner , g. 1993 , mnras , 264 , 961 saglia , r. p. , bertschinger , e. , baggley , g. , burstein , d. , colless , m. , davies , r. l. , mcmahan , jr . , r. k. & wegner , g. 1997 , apjs , 109 , 79 saglia , r. p. , colless , m. , burstein , d. , davies , r. l. , mcmahan , r. k. , & wegner , g. 2001 , mnras , 324 , 389 sargent , w. l. w. , schechter , p. l. , boksenberg , a. , shortridge , k. 1977 , apj , 212 , 326 schade , d. , lilly , s. j. , crampton , d. , ellis , r. s. , le fevre , o. , hammer , f. , brinchmann , j. , abraham , r. , colless , m. , glazebrook , k. , tresse , l. , & broadhurst , t. 1999 , apj , 525 , 31 schlegel , d. j. , finkbeiner , d. p. , & davis , m. 1998 , apj , 500 , 525 scodeggio , m. , gavazzi , g. , belsole , e. , pierini , d. , & boselli , a. 1998 , mnras , 301 , 1001 smith , j. a. , tucker , d.l . ,
kent , s.m . ,
2002 , aj , 123 , 2121 stoughton , c. , lupton , r.h . ,
bernardi , m. , et al .
2002 , aj , 123 , 485 ( early data release ) strauss , m.a . ,
weinberg , d.h . ,
lupton , r.h .
2002 , aj , 124 , 1810 tantalo , r. , chiosi , c. , & bressan , a. 1998 , a & a , 333 , 490 tinsley , b. m. , & gunn , j. e. 1976 , apj , 206 , 525 tonry , j. , & davis , m. 1979 , aj , 84 , 1511 trager , s. c. , worthey , g. , faber , s. m. , burstein , d. , & gonzlez , j. j. 1998 , apjs , 116 , 1 trager , s. c. , faber , s. m. , worthey , g. , & gonzlez , j. j. 2000a , aj , 119 , 1645 trager , s. c. , faber , s. m. , worthey , g. , & gonzlez , j. j. 2000b , aj , 120 , 165 treu , t. , stiavelli , m. , casertano , s. , mller , p. , & bertin , g. 1999 , mnras , 308 , 1037 treu , t. , stiavelli , m. , mller , p. , casertano , s. , & bertin , g. 2001a , mnras , 326 , 221 treu , t. , stiavelli , m. , bertin , g. , casertano , s. , & mller , p. 2001b
, mnras , 326 , 237 tully , r. b. & fisher , j. r. , 1977 , a&a , 54 , 661 van albada , t. s. 1982 , mnras , 201 , 939 van der marel , r. p. , & franx , m. 1993 , apj , 407 , 525 van dokkum , p. g. & franx m. , 1996 , , 281 , 985 van dokkum , p. g. , franx m. , kelson , d. d. , illingworth , g. d. 1998 , apjl , 504 , 17 van dokkum , p. g. , franx , m. , kelson , d. d. , & illingworth , g. d. 2001 , apjl , 553 , 39 vazdekis , a. , casuso , e. , peletier , r. f. , & beckman , j. e. 1996 , apjs , 106 , 307 wegner , g. , colless , m. , saglia , r. p. , mcmahan , r. k. , davies , r. l. , burstein , d. , baggley , g. , 1999 , mnras , 305 , 259 weinberg , s. , 1972 , gravitation , and cosmology : principles and applications of the general theory of relativity , john wiley & sons , new york white , s. d. m. , & rees , m. j. 1978 , mnras , 183 , 341 worthey , g. 1994 , apjs , 95 , 107 york , d.g . , adelman , j. , anderson , j.e . , et al . 2000 , aj , 120 , 1579 ziegler , b. l. , bower , r. g. , smail , i. , davies , r. l. , & lee , d. 2001 , mnras , 325 , 1571
when converting the observed apparent magnitude to the rest - frame absolute magnitude of an object , we must account for the fact that the sdss filters measure the light from a fixed spectral range in the observers rest - frame ; therefore , they measure different parts of the rest - frame spectrum of galaxies at different redshifts . correcting for this
is known as the k - correction .
one way to make this correction is to assume that all the galaxies at a given redshift are similar , and to use an empirically determined template spectrum , measured from a few accurately measured spectra , to estimate the k - correction .
using a mean color to estimate the k - correction is not ideal . when the survey is closer to completion it should become possible to make this correction on an object - by - object basis .
empirically determined template spectra for early - type galaxies at low redshifts exist ( e.g. coleman , wu & weedeman 1980 ; fukugita , shimasaku & ichikawa 1995 ) .
( we used n. benitez s bayesian photometric redshift package ( benitez 2000 ) to derive k - corrections for the coleman , wu & weedman early - type galaxy template in the sdss passbands . ) if we were certain that evolution effects were not important , then these empirically determined k - corrections would allow us to work out the k - corrections we should apply to the high redshift population .
however , if the stars in early - type galaxies formed at approximately the same time , and if the mass in the galaxies has remained constant , so the evolution is entirely due to the passive aging of the stellar population , then , the mass to light ratio of early - type galaxies is expected to vary approximately as @xmath189 ( e.g. , tinsley & gunn 1976 ) , with the precise scaling being different in different bands .
because our sample spans a relatively large range in redshift , we may be sensitive to the effects of this passive evolution .
in addition , because the sample is large , we may be able to measure , and hence be sensitive to , even a relatively small amount of evolution .
for this reason , it would be nice to have a prescription for making k - corrections which accounts for evolution .
absent empirically determined templates for this evolution , we must use stellar population systhesis models to estimate this evolution , and so determine k - corrections for different bands .
= as a first example , we chose a bruzual & charlot ( 2003 ) model for a @xmath104 object which formed its stars with the universal imf given by kroupa ( 2001 ) in a single solar metallicity and abundance ratio burst 9 gyr ago .
we then recorded how its colors , as observed through the sdss filters , changed as it was moved through redshift without altering its age .
this provides what we will call the no - evolution k - correction .
figure [ kcww - kbc ] shows a comparison of this with the empirical coleman , wu & weedman ( 1980 ) nonevolving k - corrections .
the two estimates are in good agreement in @xmath66 and @xmath61 out to @xmath190 .
they differ substantially at higher redshifts , but this is not a concern because none of the galaxies in our sample are so distant . in @xmath34 and @xmath67 the two estimates agree only at @xmath191 .
therefore , quantitative estimates of evolution in luminosity and/or color will depend on which k - correction we use .
= figure [ fig : kcorr ] compares both sets of nonevolving templates with the observed colors of the galaxies in our sample .
the upper set of curves in each panel show the colors associated with the nonevolving coleman , wu & weedman ( 1980 ) template ( dotted ) and the bruzual & charlot ( 2003 ) no - evolution model ( dashed ) .
the figure shows that both predictions for @xmath192 are similar , but that they are different for @xmath193 and @xmath194 , with the differences increasing with redshift .
[ in both cases , we have shifted the predicted @xmath192 blueward by 0.08 mag at all @xmath33 .
such an offset appears to be required for the sdss photometric calibrations in stoughton et al .
( 2002 ) which we use here ( also see eisenstein et al . 2001 and strauss et al . 2002 ) , although the reason for it is not understood . ]
= we argue in papers ii , iii and iv that the luminosities , colors and chemical abundance ratios of these objects show evidence for passive evolution : the higher redshift population appears to be slightly younger .
therefore , our next step is to include the effects of evolution . because the predicted observed colors at redshift zero are in good agreement with our data , we took the same bruzual & charlot ( 2003 ) model , but this time we recorded how its rest - frame colors evolve with redshift , and then computed what these evolved ( i.e. , @xmath33-dependent ) colors look like when observed in the sdss filters .
these evolving colors are shown as the dot - dashed lines in figure [ fig : kcorr ] . in an attempt to include evolution in the cww templates , we set @xmath195 .
the lower solid lines in each panel of figure [ fig : kcorr ] show the observed @xmath192 , @xmath193 and @xmath194 colors associated with these evolving models ( and again , the predicted @xmath192 curves have been shifted blueward ( downward ) by 0.08 mag at all @xmath33 ) . comparing the evolving bruzual
charlot and the coleman et al .
colors with the upper set of no - evolution curves shows the evolution towards the blue at high redshift . although the data appear to be very well fit by the no - evolution curves , this agreement is slightly misleading .
more luminous galaxies tend to be redder . as a consequence ,
a magnitude limited catalog contains only the redder objects of the higher redshift population .
a curve which describes the colors of the population as a whole will therefore appear to be biased blue .
figure [ kcorrsim ] shows this explicitly .
the two panels were constructed by making mock catalogs of a passively evolving population ( i.e. , the higher redshift population is brighter and bluer ) in which our estimates of the correlation between colors , luminosities , velocity dispersions and color and luminosity evolution were included ( see appendix a in paper ii ) .
the top panel shows the distribution of observed colors if there were no magnitude limit , and the bottom panel shows the observed colors of a magnitude limited sample .
the solid curve , the same in both panels , is the predicted trend of color with redshift which we use to make our k - corrections ; i.e. , @xmath196 .
notice that although it describes the complete simulations well , it is bluer than the higher redshift galaxies in the magnitude limited sample .
comparison with the previous figure shows that the difference here is similar to that seen in the real data , suggesting that our k - corrections and evolution estimates of the mean of the population are self - consistent . of course
, if we do not observe the mean of the high redshift population , but only the redder fraction , then we must decide whether it is realistic to use a k - correction which has been constructed to fit the truely typical galaxy at each redshift .
for example , if color is an indicator of age and/or metallicity , then the results above suggest that our sample contains the oldest and/or most metal rich part of the high redshift population .
if the objects which satisfy our apparent magnitude limit are in fact , older than the typical high redshift galaxy , then it may be that those objects are similar in age to the average object at lower redshifts in our sample .
if so , then we are better - off using a nonevolving k - correction even though the higher redshift sample as a whole is younger . none of the results presented in the main text change drastically if we use non - evolving rather than evolving k - corrections
. to decide which k - corrections to use , we computed the color - magnitude and color-@xmath9 relations discussed in paper iv using both the coleman et al .
( 1980 ) template and the bruzual & charlot ( 2003 ) models .
the slopes of the mean relations , and the scatter around the mean , remained approximately the same for both k - corrections , so we have chosen to not show them here .
this suggests that our ignorance of the true k - correction does not strongly compromise our conclusions about how color correlates with magnitude and velocity dispersion .
conclusions about evolution , however , do depend on the k - correction . the color@xmath9 relations constructed using either k@xmath197 or k@xmath198 show evidence for evolution
. however , k@xmath197 yields evolution in @xmath192 of 0.04 mags , and in @xmath193 of 0.07 mags , whereas k@xmath198 has changes of 0.07 and 0.03 , respectively .
thus , k@xmath197 suggests that the evolution in @xmath193 is larger than in @xmath192 .
this is not the expected trend ; the @xmath192 and @xmath193 wavelength baselines are about the same , so one expects more of the evolution to come in at the bluer color .
using k@xmath198 instead suggests that most of the evolution is in @xmath192 , which is more in line with expectations .
we also tried k - corrections from fukugita et al .
( 1995 ) . at low redshifts ,
the predicted early - type colors are redder than those in our sample , the predicted s0 colors are bluer , and the differences depend on redshift .
a straight average of the two is an improvement , although the resulting low redshift @xmath192 is red by 0.05 mags .
if we shift by this amount to improve the agreement at low redshifts , then the observed @xmath192 colors at @xmath199 are redder than the predicted no evolution curves by about 0.2 mags .
this is larger than the offset we expect for the selection effect introduced by the magnitude limit , so we decided against presenting further results from these k - corrections .
this appendix describes how we estimated the line - of - sight velocity dispersions @xmath9 for the sample of galaxies selected for this paper .
estimates of @xmath9 are limited by the instrumental dispersion and resolution .
recall that the instrumental dispersion of the sdss spectrograph is 69 km s@xmath23 per pixel , and the resolution is @xmath53 km s@xmath23 .
in addition , the instrumental dispersion may vary from pixel to pixel , and this can affect measurements of @xmath9 .
these variations are estimated for each fiber by using arc lamp spectra ( upto 16 lines in the range 3800 - 6170 and 39 lines between 5780 - 9230 ) .
an example of the variation in instrumental dispersion for a single fiber is shown in figure [ fig : resol ] .
the figure shows that a simple linear fit provides a good description of this variation .
this is true for almost all fibers , and allows us to remove the bias such variations may introduce when estimating galaxy velocity dispersions . = a number of methods for making accurate and objective velocity dispersion measurements as have been developed ( sargent et al . 1977 ; tonry & davis 1979 ; franx , illingworth & heckman 1989 ; bender 1990 ; rix & white 1992 ) .
these methods are all based on a comparison between the spectrum of the galaxy whose velocity dispersion is to be determined , and a fiducial spectral template .
this can either be the spectrum of an appropriate star , with spectral lines unresolved at the spectra resolution being used , or a combination of different stellar types , or a high @xmath55 spectrum of a galaxy with known velocity dispersion . in this work ,
we use sdss spectra of 32 k and g giant stars in m67 as stellar templates .
since different methods can give significantly different results , thereby introducing systematic biases especially for low @xmath55 spectra , we decided to use three different techniques for measuring the velocity dispersion .
these are 1 ) the _ cross - correlation _ method ( tonry & davis 1979 ) ; 2 ) the _ fourier - fitting _ method ( tonry & davis 1979 ; franx , illingworth & heckman 1989 ; van der marel & franx 1993 ) ; and 3 ) a modified version of the _ direct - fitting _ method ( burbidge , burbidge & fish 1961 ; rix & white 1992 ) . because a galaxy s spectrum is that of a mix of stars convolved with the distribution of velocities within the galaxy , fourier space is the natural choice to estimate the velocity dispersions
the first two methods make use of this .
however , there are several advantages to treating the problem entirely in pixel space .
in particular , the effects of noise are much more easily incorporated in the pixel - space based _ direct - fitting _ method . because the @xmath55 of the sdss spectra are relatively low , we assume that the observed absorption line profiles in early - type galaxies are gaussian ( see rix & white 1992 and bender , saglia & gerhard 1994 for a discussion of how to analyze the line profiles of high @xmath55 spectra in the case of asymmetric profiles ) .
it is well known that all three methods have their own particular biases , so that numerical simulations must be used to calibrate these biases . in our simulations , we chose a template stellar spectrum measured at high @xmath55 , broadened it using a gaussian with rms @xmath200 , added gaussian noise , and compared the input velocity dispersion with the measured output value .
the first broadening allows us to test how well the methods work as a function of velocity dispersion , and the addition of noise allows us to test how well the methods work as a function of @xmath55 .
= the best - case scenario is one in which there is no ` template mismatch ' : the spectrum of the template star is exactly like that of the galaxy whose velocity dispersion one wishes to measure .
figure [ fig : vsim ] shows the fraction of systematic bias associated with each of the different methods in this best - case scenario .
slightly more realistic simulations , using a combination of stellar spectra as templates , were also done .
the results are similar to those shown in figure [ fig : vsim ] . with the exception of the _ cross - correlation _ method at low ( @xmath201 km s@xmath23 ) velocity dispersion , the systematic errors on the velocity dispersion measurements appear to be smaller than @xmath202 .
= = although the systematics are small , note that the measured velocity dispersion is more biased at low velocity dispersions ( @xmath201 km s@xmath23 ) .
for any given @xmath55 and resolution , there is a lower limit on the velocity dispersion measurable without introducing significant bias . since the @xmath55 of the sdss spectra is not very high ( see , e.g. , figure [ fig : vmeth ] ) , and because the instrumental resolution is @xmath53 km s@xmath23 , we chose 70 km s@xmath23 as a lower limit .
figure [ fig : vmeth ] shows a comparison of the velocity dispersion estimates obtained from the three different methods for the galaxies in our sample .
the median offsets are not statistically significant and the rms scatter is @xmath203 . on the other hand , the top panels suggest that the cross - correlation method sometimes underestimates the velocity dispersion , particularly at low @xmath55 .
we evaluate the dependence of the velocity dispersion on the wavelength range by fitting the spectra in different intervals : @xmath204 which is the usual wavelength range used in the literature ; @xmath205 to test the effect of including the ca h and k absorption lines ( e.g. , kormendy 1982 ) ; @xmath206 to test the effect of including the nad line ( e.g. , dressler 1984 ) ; and @xmath207 and @xmath208 to test the effect of including longer wavelengths
. the velocity dispersion obtained with the ca h and k is @xmath209 larger than that obtained using the standard wavelength region @xmath204 , and the rms difference between the three different methods increases to @xmath210 . including the nad line increases the velocity dispersion by @xmath202 but does not increase the scatter between the different methods . using the wavelength range @xmath207 only provides velocity dispersions which are @xmath202 larger than the values obtained if only @xmath204 range is used . on the other hand , in this wavelength region , the different methods ( and measurements from repeated observations ) are in better agreement ;
the scatter is @xmath211 smaller than in the @xmath204 region . in the range
@xmath208 , the velocity dispersion estimates increase by @xmath210 .
this last effect is probably due to the presence of molecular bands in the spectra of early - type galaxies at long wavelengths ( i.e. , to the presence of cool stars ) .
furthermore , the scatter in this wavelength region increases dramatically ( @xmath78 ) .
presumably this is due to the presence of higher sky - line residuals and lower @xmath55 .
the estimated velocity dispersion we use in the main text are obtained by fitting the wavelength range @xmath142 and then using the average of the estimates provided by the _ fourier - fitting _ and _ direct - fitting _ methods to define what we call @xmath143 .
we do not use the cross - correlation estimate because of its behavior at low @xmath55 as discussed earlier .
the top panel of figure [ fig : vrptd ] shows the distribution of the errors on the velocity dispersion as a function of the @xmath55 of the spectra .
the errors for each method were computed by adding in quadrature the statistical error due to the noise properties of the spectrum , and the systematic error associated with the template and galaxy mismatches . the final error on @xmath143
is got by adding in quadrature the errors on the two estimates ( i.e. , the fourier - fitting and direct - fitting ) which we average .
the resulting errors range from @xmath212 dex , depending on the @xmath55 of the spectra , with a median value of 0.03 dex . a few galaxies in our sample
have been observed more than once .
the bottom panel shows a comparison of the velocity dispersion estimates from multiple observations .
the scatter between different measurements is @xmath146 dex , consistent with the amplitude of the errors on the measurements .
= = the sdss spectra measure the light within a fixed aperture of radius 1.5 arcsec .
therefore , the estimated velocity dispersions of more distant galaxies are affected by the motions of stars at larger physical radii than for similar galaxies which are nearby .
if the velocity dispersions of early - type galaxies decrease with radius , then the estimated velocity dispersions ( using a fixed aperture ) of more distant galaxies will be systematically smaller than those of similar galaxies nearby .
we have not measured the velocity dispersion profiles @xmath213 of any of the galaxies in our sample , so we can not correctly account for this effect .
if we assume that the galaxies in our sample are similar to those for which velocity dispersion profiles have been measured , then we can use the published @xmath213 curves to correct for this effect .
this is what equation ( [ appcorr ] ) in section [ vdisp ] does .
an alternative procedure can be followed if evolution effects are not important for the velocity dispersions in our sample . to illustrate the procedure
, the galaxies in each of the volume - limited subsamples shown in figure [ fig : xzr ] were further classified into small bins in effective physical radius ( i.e. , @xmath21 in kpc/@xmath214 , not @xmath107 in arcsec ) . figures [ apcorr1][apcorr3 ] show the result of plotting the velocity dispersions of these galaxies versus their redshifts .
since the galaxies at low and high redshift are supposed to be similar , any trend with redshift can be used to infer and average velocity dispersion profile , and how the shape of this profile depends on luminosity and effective radius . in this way , the sdss data themselves can , in principle , be used to correct for the effects of the fixed aperture of the sdss spectrograph . in practice , because there is substantial scatter in the velocity dispersions at fixed luminosity and size ( paper ii shows that this scatter is about 14% ) , the trends in the present data set are relatively noisy .
when the dataset is larger , it will be worth returning to this issue .
for now , because the corrections are small anyway , we have chosen to use equation ( [ appcorr ] ) to correct the velocity dispersions .
nevertheless , curves like those presented above provide a novel way to study the velocity dispersion profiles of early - type galaxies .
let @xmath215 denote the covariance matrix of the errors in our estimates of the the absolute magnitude @xmath97 , the half light radius @xmath216 , and the velocity dispersion @xmath20 : @xmath217 the elements of the error matrix @xmath215 are obtained as follows . the photometric pipeline estimates the size @xmath39 and the apparent magnitude @xmath40 from the same fitting procedure . as a result , errors in these two quantities are correlated .
let @xmath218 denote the error in @xmath219 , and @xmath220 the error in @xmath40 .
the correlation means that we need three numbers to describe the errors associated with the fitting procedure , @xmath221 , @xmath222 , and @xmath223 , but the pipeline only provides two . the error output by the pipeline in @xmath39 , is correctly marginalized over the uncertainty in @xmath40 , so it is essentially @xmath221 .
on the other hand , the quoted error in @xmath40 , say @xmath224 is really @xmath225 . to estimate the values of @xmath223 and @xmath222 which we need
, we must make an assumption about the correlation between the errors .
fortunately , this can be derived from the fact that , for a wide variety of galaxy profile shapes , the quantity @xmath226 , with @xmath227 , has a very small scatter ( e.g. saglia et al .
1997 ) . here
@xmath228 is the surface brightness , and @xmath229 is the error in the surface brightness . as a result , @xmath230 ( saglia et al . 1997 ) .
this means that @xmath231 , so that @xmath232,\end{aligned}\ ] ] and @xmath233 the final equality shows that the error output from the pipeline provides an estimate of @xmath234 which we can insert into our expressions for @xmath222 , @xmath223 , and @xmath235 .
( notice that if @xmath236 , then it would be a good approximation to set @xmath237 .
since this is not always the case for our dataset , we must solve the quadratic . )
once this has been done , we set @xmath238 that is , we compute the error in the absolute magnitude by assuming that there are no errors in the determination of the redshift ( and @xmath121 correction ! ) which would otherwise propagate through .
subsequent papers will focus almost exclusively on the circularly averaged radius @xmath21 defined in section [ landr ] .
the errors on it are given by adding the errors in the size @xmath39 to those which come from the error on the shape @xmath38 .
we assume that the errors in @xmath38 are neither correlated with those in @xmath219 nor with those in the absolute magnitude .
finally , we assume that errors in magnitudes are not correlated with those in velocity dispersion , so @xmath239 is set to zero , and that errors in size and velocity dispersion are only weakly correlated because of the aperture correction we apply . here
@xmath240 is the error in what was called @xmath241 in the main text . | a sample of nearly 9000 early - type galaxies , in the redshift range @xmath0 , was selected from the sloan digital sky survey using morphological and spectral criteria .
this paper describes how the sample was selected , presents examples of images and seeing corrected fits to the observed surface brightness profiles , describes our method for estimating k - corrections , and shows that the sdss spectra are of sufficiently high quality to measure velocity dispersions accurately .
it also provides catalogs of the measured photometric and spectroscopic parameters . in related papers ,
these data are used to study how early - type galaxy observables , including luminosity , effective radius , surface brightness , color , and velocity dispersion , are correlated with one another . |
the dispersion of the magnetic excitations is an important source of information on experimental low - dimensional spin systems .
knowledge of the dispersion relation @xmath5 helps essentially to identify the model appropriate to describe the compound under study .
the dispersion relation provides also important insight in the nature of the ground state .
very common in low dimensional systems is the scenario of a singlet @xmath6 ground state _ without _ magnetic long range order ( a `` spin liquid '' ) of which the elementary excitations are triplets @xmath7 .
these systems are generically gapped .
examples are isolated or weakly coupled dimerized spin chains and spin ladders such as ( vo)@xmath8p@xmath8o@xmath9,@xcite the spin - peierls phase of cugeo@xmath0,@xcite and srcu@xmath8o@xmath0 .
@xcite a true 2d example is srcu@xmath8(bo)@xmath8 which is characterized by frustrated dimers.@xcite in these gapped @xmath10 systems where the gap is related to some `` strong '' bond ( which can be also the rung of a 2-leg ladder ) the elementary triplet excitations are in principle accessible by a perturbative expansion about the limit of isolated dimers .
this approach , however , becomes tedious for the description of realistic materials since the expansion parameter is often not really small .
thus one has to compute high orders to achieve quantitative agreement .
for this reason various automated approaches have been conceived which leave the tedious part to computers @xcite . in the present article , we will apply the previously introduced perturbation by flow equation @xcite to the two - dimensional , though anisotropic , system of cugeo@xmath0 in its dimerized low - temperature phase @xcite . thereby we extend the previous analysis ( ref .
@xcite , henceforth cited as [ i ] ) considerably .
our starting point remains the same as before .
the strongest coupling is given by @xmath11 ; the other couplings are given relative to @xmath11 as indicated in fig .
[ koppl1 ] ( for details see fig . 1 in [ i ] ) .
the couplings are denoted relative to the strongest coupling @xmath11 which is set to unity in the figure .
[ koppl1],width=302 ]
the problem to be solved reads @xmath12 as in the chain in ref .
@xcite the isolated dimer limit ( @xmath13 at finite @xmath14 ) has an equidistant energy spectrum and the perturbation can alter the number of energy quanta ( here : triplets on the dimers ) by 2 at maximum .
hence @xmath15 can be represented as @xmath16 where @xmath17 stands for the perturbing part changing the number of elementary triplets by @xmath18 .
the same formalism as in ref .
@xcite can be used .
this formalism maps the perturbed hamiltonian by a continuous unitary transformation , the so - called flow equation method @xcite , to an effective hamiltonian @xmath19 which _ conserves _ the number of energy quanta , i.e. @xmath20 $ ] . the effective hamiltonian has the form @xmath21 where @xmath22 is a vector of dimension @xmath23 of which the components are in @xmath24 ; @xmath25 signifies that the sum of the components vanishes which reflects the conservation of the number of energy quanta ( triplets ) . the coefficients @xmath26 are generally valid fractions computed in ref .
@xcite where also further details on the flow equation method can be found . since @xmath19 conserves the triplet number the one - triplet sector is particularly easy to solve .
acting on one triplet the action of @xmath19 may only consist in shifting the triplet .
this means that the triplet hops on an effective lattice where one site stands for one dimer on the original lattice , see fig . 2 in [ i ] or fig . [ eff - latt ] .
for all hopping processes starting on @xmath27 and ending on one of the depicted dimers ( circles ) .
for all circles that are accessible by an arbitrary hopping of length six or less the amplitudes have been calculated within 6@xmath1 order .
( the length of a hop is the minimum number of bonds ( solid or dashed ) required to link start and end point . )
the amplitudes for light gray , dark gray and black circles have been extended in 8@xmath1 order , provided that these sites can be reached by a hopping @xmath28 of length 8 .
analogously , the amplitudes for dark gray and black circles were extended by calculating hopping processes @xmath29 within 9th order .
finally , the amplitudes for all black circles were extended by processes @xmath30 within 10th order .
the arrows indicate axes with respect to which reflection symmetry holds.,width=332 ] the full dispersion @xmath5 is obtained by fourier transform @xmath31 ( 5.2,5)(1.5,0.2 ) the hopping amplitudes @xmath32 can be calculated on finite clusters of the ( in principle infinite ) effective lattice : from the linked cluster theorem we know that the finite order contribution of a short - ranged perturbation does not depend on the cluster size for sufficient large clusters .
carrying out our perturbation within order @xmath33 implies that one allows dimer to dimer hopping processes of length @xmath33 @xcite .
the minimum cluster for a given amplitude @xmath32 in a given order @xmath33 contains all dimers and links that are involved in a hopping of length @xmath34 starting at dimer @xmath27 and ending at @xmath35 .
the minimum cluster is determined by considering all paths from @xmath27 to @xmath35 .
all dimers and links covered by one of these paths are part of the minimum cluster . in fig .
[ cluster ] , the computer generated minimum cluster for calculating @xmath36 in order 8 is shown
. due to the strong anisotropy of the quasi-1d system cugeo@xmath0 it is reasonable to use higher order terms only along the chains .
this simplifies the computational task considerably since the calculation of a hopping process along the chain is much simpler .
the cluster to be considered can be chosen smaller .
the same is true for hopping processes _ close _ to the chain direction .
here we restrict the hopping processes to be at maximum quadratic in the interchain hopping @xmath37 , which reduces the cluster sizes significantly so that the perturbation order can be enlarged .
the results for the @xmath32 are too lengthy to be published in written form
. we will provide them in electronic form on our home pages on appearance of this article . in [ i ]
the @xmath32 in third order in @xmath38 and @xmath37 were presented .
a few of these are erroneous .
they are corrected herewith @xcite .
the corrections , however , have no influence on the conclusions in [ i ] ( see also discussion below ) .
once all amplitudes @xmath32 are calculated the dispersion relation is given by eq .
( [ disp ] ) . after rewriting eq .
( [ disp ] ) in terms of @xmath39 and @xmath40 ( the reciprocal basis to @xmath41 and @xmath42 ) we add the term @xmath43 with @xmath44 mev to account for the dispersion in a - direction ( cf .
to fix the parameters @xmath11 , @xmath45 , @xmath46 , @xmath37 and @xmath38 ( cf .
[ koppl1 ] ) we use the one - magnon dispersion data for cugeo@xmath0 experimentally determined by inelastic neutron scattering @xcite .
note that the hopping amplitudes are computed as polynomials over @xmath47 in the parameters .
as noticed in [ i ] the parameter @xmath46 has almost no influence on the shape of the dispersion .
hence we refrain from determining @xmath46 from the dispersion but set it beforehand to some reasonable values in the interval @xmath48 $ ] .
this choice is motivated by comparing the microscopic _ direct _ super - exchange path @xmath37 and the _ shifted _ super - exchange path @xmath49 shown schematically in fig .
[ boris ] ( cf .
[ koppl1 ] ) .
there is only one path per cu@xmath50-site for the shifted coupling whereas there are two paths for the direct coupling .
thus we expect @xmath51 , i.e. @xmath52 .
there are also results from ab - initio calculation for the interchain couplings which indicate the existence of interchain frustration @xcite .
further evidence is provided below by the analysis of the susceptibility .
furthermore , we find that for @xmath53 fits to the dispersion data become worse . .
the three - dimensional situation is depicted in ref .
@xcite.,width=226 ] to determine the remaining parameters we equate four different experimental points with the corresponding parameter dependent dispersion values given by eq .
( [ disp ] ) .
the parameters are fixed by solving the resulting system of equations . for @xmath54 and @xmath55 figs .
[ disp_big_c ] , [ disp_big_b ] show the resulting dispersion curves in c@xmath56- and in b@xmath56-direction , respectively , using all @xmath32 calculated . in c@xmath56-direction .
the arrows indicate the experimental points used to fix the parameters .
@xmath57 order fits based on @xmath58 = ( 0.3 , 0.836 , 0.225 , 0.266 , 13.1mev ) and ( -0.3 , 0.846 , 0.209 , 0.081 , 12.3mev ) , respectively.,width=332 ] in b@xmath56-direction .
otherwise as in fig . [ disp_big_c],width=332 ] as can be seen from figs .
[ disp_big_c ] and [ disp_big_b ] the plain series up to @xmath59 order provides excellent fits .
yet one realizes that the parameter values still change on passing from order to order .
so it appears that even at @xmath59 order the results are not quantitative . in order to obtain quantitative reliable results
we adopt a systematic extrapolation in the order . in each order @xmath60
we determine the optimum fit parameters . for illustration ,
[ trend_1 ] shows results for @xmath61 . at @xmath61 .
the quasi - constant behavior of @xmath37 is found for all @xmath46-values checked .
the lines are fits to the data according to eq .
( [ extrapol ] ) .
the solid lines consider all points , the dashed ones only the last four points.,width=332 ] assuming exponential convergence we use @xmath62 where @xmath63 is the asymptotic value of the parameter considered and @xmath64 and @xmath65 are constants .
the choice ( [ extrapol ] ) is motivated on one hand by its obvious applicability ( see fig . [ trend_1 ] ) . on the other
it stems from the fact that cugeo@xmath0 is a quasi one - dimensional gapped spin system .
so one expects the magnetic correlations to drop exponentially with distance .
furthermore the order @xmath33 determines the maximum distance over which correlations occur ( cf . ref @xcite , namely @xmath33 counted in dimers or @xmath66 counted in spin sites .
hence the constant @xmath65 in eq .
( [ extrapol ] ) can be understood as the inverse of a correlation length @xmath67 . with the usual relation @xmath68 for one dimensional systems we obtain @xmath69 based on the rough estimates @xmath70 @xcite and @xmath71
this is indeed what is found ( cf
i , ii ) so that we judge our extrapolations as being well justified . in fig .
[ trend_1 ] the extrapolations are depicted by lines .
the solid lines were obtained by using all calculated parameter values .
the dashed lines are obtained from on the last four points , i.e. the results in order 7 , 8 , 9 , and 10 .
the deviation between these two extrapolations are used as a measure for the extrapolation error .
this procedure is carried out for @xmath72 , and @xmath11 .
there is no systematic dependence of @xmath37 on the order @xmath33 .
the parameter @xmath37 oscillates between the two thin horizontal lines in fig .
[ trend_1 ] .
so we take the average of these two bounds as our estimate for @xmath37 and their difference as the error in the determination of @xmath37 .
i and ii summarize the results of the fits for the parameters @xmath45 , @xmath38 and @xmath11 for four preset values of @xmath46 . the values for @xmath37
are listed in tab .
table i. table ii . a closer inspection of figs .
[ disp_big_c ] and [ disp_big_b ] reveals that we are confronted with a certain arbitrariness of which fit we should favor .
the experimental errors enhance this problem .
the filled circles in the range of small wave vectors in fig .
[ disp_big_c ] represent experimental points which have been measured with a high degree of precision .
thus it is reasonable to fit the theoretical curve as well as possible to these points .
[ disp_small_c ] shows an enlargement of this region .
the solid line is the 10@xmath1 order fit result for @xmath61 as the solid line in fig .
[ disp_big_c ] .
the depicted arrow indicates the experimental point ( @xmath73mev ) used to obtain tab .
i. for small wave vectors at @xmath54 .
the solid curve is the same as in fig .
[ disp_big_c ] , leading to tab .
i. the dashed line corresponds to a 10@xmath1 order fit where the point ( @xmath73mev ) indicated by the arrow is replaced by ( @xmath74mev ) , leading to tab .
the filled circles correspond to highly accurate experimental points for which the error bars are of the size of the symbols.,width=332 ] a likewise well suited curve , however , is produced if one uses the point ( @xmath74mev ) for the fit keeping the other points ( tab .
it is not possible to prefer one of the two lines in fig .
[ disp_small_c ] to the other on the basis of their agreement to the experimental data .
hence we choose these two fits as the bounds within which all fits are acceptable .
the corresponding parameter values @xmath75 ( fit 1 ) and @xmath76 ( fit 2 ) provide an interval [ @xmath77 which we expect to contain the true model parameter @xmath78 .
hence the latter is estimated by @xmath79 , with @xmath80 , @xmath81 and @xmath82 .
the results are summarized in tab .
+ @xmath45&0.27(4)&@xmath3&0.25(3 ) + @xmath38&0.86(2)&@xmath83&0.08(1 ) + @xmath37&0.27(1)&@xmath84&0.29(1 ) + @xmath11/mev&13.8(5)&@xmath85/mev&12.8(6 ) + + @xmath45&0.27(4)&@xmath3&0.25(3 ) + @xmath38&0.86(2)&@xmath83&0.08(1 ) + @xmath37&0.21(1)&@xmath84&0.23(1 ) + @xmath11/mev&13.4(4)&@xmath85/mev&12.5(5 ) + + @xmath45&0.27(5)&@xmath3&0.25(5 ) + @xmath38&0.88(3)&@xmath83&0.07(2 ) + @xmath37&0.13(1)&@xmath84&0.14(1 ) + @xmath11/mev&13.2(6)&@xmath85/mev&12.4(7 ) + + @xmath45&0.27(5)&@xmath3&0.25(5 ) + @xmath38&0.88(3)&@xmath83&0.06(2 ) + @xmath37&0.08(2)&@xmath84&0.09(1 ) + @xmath11/mev&13.1(5)&@xmath85/mev&12.3(7 ) + table iii . for the readers convenience tab .
iii also gives the results in the more commonly used parameters @xmath83 , @xmath3 , @xmath84 , @xmath85 and @xmath46 .
this notation is connected to the one used so far in this article by @xmath86 it corresponds to the hamiltonian depicted in fig .
[ koppl2 ] . .
the couplings are denoted relative to the average nearest - neighbor coupling @xmath85 in the chains .
[ koppl2],width=302 ]
the temperature dependence of the homogeneous susceptibility @xmath87 is often used to determine the parameters of cugeo@xmath0 @xcite .
already the curie - weiss temperature @xmath88 provides valuable information on the sum of the coupling constants .
this is particularly useful to detect frustration .
the dispersions are governed by the difference of the direct and the frustrating coupling whereas @xmath87 at larger temperatures is more influenced by the sum of direct and frustrating coupling .
the analysis of the curie - weiss temperature alone bears some risks .
it is easy to calculate but difficult to determine experimentally since it has to be deduced from values at high temperatures where @xmath87 is fairly small and hence strongly influenced by background effects ( van vleck , diamagnetism ) or by slight structural changes .
a convincing fit for temperatures above 50k is given by fabricius _
_ in ref .
@xcite on the basis of frustrated chains .
the inclusion of interchain couplings , however , would spoil the excellent agreement and a re - determination of the constant would be necessary .
a description of @xmath87 on the basis of a two - dimensional model has not been done except for a consideration of the two leading coefficients in an expansion in @xmath89 in [ i ] . in fig . [
fig : susz ] we show the same high quality experimental data as in ref .
@xcite and compare it to theoretical curves at four values of @xmath46 .
the theoretical curves are obtained by computing a [ 4,5 ] dlog pad approximant @xmath90 based on the high temperature series provided in ref .
@xcite for the frustrated chains .
this procedure provides excellent results down to @xmath91 @xcite .
the asymptotic behavior of the approximant is chosen such that @xmath90 vanishes linearly on @xmath92 as is to be expected for a two - dimensional massless antiferromagnet .
besides this feature the two - dimensionality is incorporated on a chain - mean - field level @xmath93 this relation is exact in linear order in @xmath37 .
estimates of corrections quadratic in @xmath37 indicate that they are negligible for the values of @xmath37 and @xmath46 in which we are interested .
value used is @xmath94 @xcite .
[ fig : susz],width=332 ] from the results in fig .
[ fig : susz ] it is evident that the interchain frustration can not be neglected .
only for a finite value of about @xmath95 a very good agreement can be obtained .
the agreement to the frustrated chain model @xcite is still better since the position of the maximum is also reproduced .
but on the basis of the neutron scattering results @xcite it is undoubtful that cugeo@xmath0 is a two - dimensional substance .
furthermore , it must be considered that the previous fit @xcite was a two - parameter ( @xmath11 and @xmath3 ) fit whereas only one parameter ( @xmath46 ) is fitted to obtain fig .
[ fig : susz ] . the other parameters ( @xmath96 )
were determined from an entirely different experiment .
hence the agreement for @xmath87 corroborates also the validity of the parameters determined in the preceding section .
we will first discuss our results and propose a set of parameters .
then we will put these results into the context of other results in the literature .
let us consider the remaining difference between experiment and theory concerning @xmath87 in fig .
[ fig : susz ] .
there are four conceivable sources for it .
the first are experimental inaccurcies .
we are not in the position to judge this aspect . we just like to comment that from the results and error bars in fig .
[ disp_small_c ] it is obvious that the experimental data is not completely consistent so that this explanation is possible .
second , it is conceivable that the couplings change across the transition , i.e. the intrachain frustration at @xmath97 ( where the dispersions are measured ) is different from the one above @xmath98 ( where @xmath87 is measured ) . since we are considering a spin - peierls transition
there is definitely a structural change .
so far , however , the assumption that only the dimerization changes worked well . the structural changes in the transition are very small @xcite whereas the changes needed to explain the discrepancy are of the order of 20 to 30 % ( assuming a change in the intrachain frustration ) .
estimates point into the direction that the changes on the couplings are unimportant @xcite .
yet the estimates concern in the first place the nearest - neighbor coupling only .
quantitative ab initio calculations of the frustration are very difficult @xcite , even more so for changes of the frustration .
so , again , this explanation is perhaps not the most promising but can not be excluded either .
third , the influence of the phonon dynamics is to be considered .
it is shown that spin - peierls systems can be unitarily transformed in such a way that an effective spin model remains at low energies uncoupled to the phonons @xcite .
the effective couplings in a single chain model become temperature dependent so that this may account for the discrepancy .
but it turns out for non - resonant phonons ( @xmath99 ) that these effects leave the susceptibility fairly unchanged .
this is so since these effects become significant for relatively large temperatures where @xmath87 depends only on the sum of all couplings which is unchanged by the unitary transformation ( this is observed in ( vo)@xmath8p@xmath8o@xmath9 @xcite ) .
so this reason appears rather unlikely even though it looked plausible at first sight .
fourth , one has to think about any kind of precursors of the spin - peierls transition . by this we mean on one side the critical fluctuations which appear in a narrow region ( @xmath100 ) around the spin - peierls temperature @xcite . on the other side , we mean any precursor which goes beyond a purely static spin model .
experimentally , a finite lattice correlation length can be detected already at @xmath101k far above the actual transition @xcite .
from there on deviations from the behavior of a static spin model should be observable . in the adiabatic limit , for instance
, the fluctuation yield already a reduction of the susceptibility @xcite .
what happens in the antiadiabatic limit has not yet been investigated for a two or higher dimensional model .
the mapping in ref .
@xcite leads to four - point interchain couplings the influence of which is unclear so far . in view of the above mentioned possible pitfalls of the static model
the agreement in fig .
[ fig : susz ] is already very convincing . summarizing our results we propose the parameters given for @xmath102 in tab .
iii to be the ones deduced from the dispersion data . assessing the reliability of our estimates , we redo the analysis of the susceptibility for @xmath103 ( the upper bound of our estimate for @xmath3 ) with the corresponding value of @xmath104mev . then the optimum @xmath87 is obtained for @xmath105 ; the @xmath87 curve is almost identical to the one shown in fig .
[ fig : susz ] .
so the value of @xmath46 can not be determined very precisely , but it should be in the range @xmath106 .
a certain dicrepancy between the optimum parameters for the @xmath107 dispersion data and for the @xmath108 data remains .
we split the comparison of our findings to previous works into three groups .
the first comprises the analyses on the basis of a one - dimensional model @xcite . the most striking difference to the results for static spin models
@xcite is that the dimerization @xmath83 is not of the order of 1% but significantly larger .
this is not astounding since it has been noted already in [ i ] that the gap is lowered by the interchain hopping .
hence the neglect of the latter requires to lower the gap otherwise , i.e. by assuming a lower dimerization .
our intrachain frustration is slightly larger than the one of castilla _
et al_. ( @xmath109 ) , but significantly smaller than the value of riera and dobry ( @xmath110 ) or the value of fabricius _
et al_. ( @xmath111 ) .
et al_.showed that the value @xmath112 is too small for a single chain model .
the difference between the larger frustration value in the single chain model to our value results directly from the interchain coupling . as can be nicely seen in eq .
( [ eq : cmf ] ) the interchain coupling lowers the susceptibility without changing ( in the chain mean - field approximation ) the position of its maximum .
the one dimensional models favor a larger intrachain frustration and a concomitant larger coupling @xmath11 in order to reduce the magnitude of the susceptibility . the claim by wellein and coworkers that the dimerization experimentally found to be larger than would fit to a static 1d model @xcite is due to the phonon dynamics is not compelling .
they use a root - mean - square definition of the dimerization which naturally provides larger values for the dimerization since it includes all the fluctuations .
the dispersion perpendicular to the chains , however , is an unambiguous experimental fact .
furthermore , trebst and coworkers @xcite do not find a substantial gap renormalization for parameters relevant for cugeo@xmath0 even though one should take care of different schemes to couple the phonons .
let us turn to the second group of papers considering the essentially two - dimensional character of cugeo@xmath0 .
the first work used a bond - operator technique @xcite .
no frustration was considered , hence rather small values of @xmath113mev and a rather small interchain hopping @xmath114 resulted .
the same technique was also applied later again by brenig @xcite including frustration .
it turned out , however , that only @xmath115 and @xmath116 matter on the free - boson level .
hence an independent determination of the frustration is not possible .
using additional input ( @xmath117 ) the values @xmath118 and @xmath119 were obtained . in view of the extensive comparisons to numerical results made in ref .
@xcite it appears that the bond - operator method overestimates the influence of additional couplings such as dimerization or frustration . generally , the values for dimerization or frustration tend to be too low .
this is confirmed by our findings in the present work .
compared to [ i ] ( @xmath120mev , @xmath121 ) the extended series on which our present analysis is based gives a much better handle on intrachain frustration , see fig .
[ trend_1 ] . only in the present work ,
we are able to assess its value reliably . with respect to the interchain frustration ,
the present results agree qualitatively with those in [ i ] where such a frustration was proposed for cugeo@xmath0 first .
the use of susceptibility information has been improved in the present work since the whole @xmath87 curve is used , not only the leading coefficients .
bouzerar _ et al .
_ have carried out an estimate leading to results not too far from ours : @xmath122mev , @xmath123 .
they used just linear order in the interchain hopping without interchain frustration and some square - root averaging with numerical results for chains to describe the dispersion . the intrachain frustration ( @xmath124 )
could only be taken from the curie - weiss constant .
the resulting @xmath87 has similarities with the experimental one .
the third group comprises ab - initio calculations of the exchange couplings and of the spin - phonon couplings .
microscopic calculations @xcite find relatively large values of the dimerization between @xmath125 and @xmath126 in agreement with our findings .
( even though there is also a different result @xcite ) very important for our work are recent results by drechsler and coworkers supporting the existence of interchain frustration @xcite .
werner and coworkers @xcite estimate a large dimerization from the spin - phonon couplings and the shifts of the ions ( @xmath127 ) . from the balance of elastic and magnetic energy in the d phase they obtain without frustration a lower bound of @xmath128 . assuming critical frustration @xmath129 they find even @xmath130 which fits very nicely to our findings . in summary
, we provide by the present work a determination in great detail of the coupling parameters ( @xmath106 and right column in tab .
iii under @xmath102 ) of cugeo@xmath0 based on a static dimerized spin model at @xmath107 .
the experimental input comes from inelastic neutron scattering .
the implications of the parameters found for the susceptibility are also studied .
very good agreement could be obtained fitting the interchain frustration appropriately .
a small discrepancy at low temperatures around @xmath131k indicates that the static spin model is probably insufficient to describe cugeo@xmath0 completely . by this work
, we proved the outstanding possibilities of high - order series expansions ( around the dimer limit or around the limit @xmath132 ) in the analysis of experimental data .
the authors like to thank b. bchner , a. bhler , u. lw , b. mari , e. mller - hartmann , and f. schnfeld for fruitful discussions .
the provision of the experimental data by b. bchner , t. lorenz and by l. p. regnault is gratefully acknowledged .
furthermore , we thank the regional computing center of the university of cologne for its kind and efficient support .
this work was supported by the deutsche forschungsgemeinschaft in the schwerpunkt 1073 and in the sonderforschungsbereich 341 . some mixed third order hopping amplitudes in ref .
@xcite , eq . ( 2 ) ) are incorrect : @xmath133 : @xmath134 should be @xmath135 @xmath136 should be @xmath137 @xmath138 should be @xmath139 .
@xmath140 : @xmath141 should be @xmath142 @xmath143 should be @xmath144 .
@xmath145 : @xmath146 should be @xmath147 @xmath148 should be @xmath149@xmath150 should be @xmath151 .
@xmath152 : @xmath153 should be @xmath154 @xmath155 should be @xmath156 . | we reconsider the 2d model for cugeo@xmath0 introduced previously ( phys .
rev .
lett . 79 , 163 ( 1997 ) ) . using a computer aided perturbation method based on flow equations we expand the 1-triplet dispersion up to 10@xmath1 order .
the expansion is provided as a polynom in the model parameters .
the latter are fixed by fitting the theoretical result to experimental data obtained by ins . for a dimerization @xmath2
we find an excellent agreement with experiment .
this value is at least 2 to 3 times higher than values deduced previously from 1d chain approaches . for the intrachain frustration @xmath3
we find a smaller value of @xmath4 .
the existence of interchain frustration conjectured previously is confirmed by the analysis of temperature dependent susceptibility . |
systems undergoing a phase separation from a disordered phase to a more ordered state often form interesting patterns as they select and approach the final state@xcite .
the process of phase ordering , or coarsening dynamics , can be classified into different universality classes depending on the physical dimension , the dimension of the order parameter , and conservation laws@xcite .
typically the associated scaling properties and dynamical universalities can be understood using a coarse - grained free energy functional .
a time - dependent ginzburg - landau approach then allows study of the evolving patterns .
ferro - fluids@xcite , which are colloidal suspensions of magnetic particles in solution , exhibit complex labyrinthine pattern formation when trapped between closely spaced glass plates and subjected to a transverse magnetic field@xcite .
the complex patterns result from competition between the surface tension and long - range forces , and have much in common with patterns observed in amphiphilic systems and langmuir - blodget films . in this paper
we report on a similar effect in colloidal suspensions in which phase separation and a concomitant pattern formation are driven by an external electric field .
our results are based upon a macroscopic thermodynamic model of a dielectric binary liquid .
this work was motivated by recent experiments demonstrating such behavior@xcite , and shares some features in common with the experiments .
we reported some preliminary results in ref . .
we consider colloids consisting of nanometer scale ( typical diameter @xmath0 nm ) dielectric particles suspended in a dielectric solution . in the present work we are interested in patterns on length scales much larger than the particle size and the typical inter - particle spacing .
consequently we use a coarse - grained description similar to the model b used to study phase separation in binary liquids@xcite . for an order parameter
we use the local particle concentration or volume fraction @xmath1 ( @xmath2 ) .
we model the particle interaction using a local dielectric constant @xmath3 which depends nonlinearly on the local concentration .
the use of a phenomenological dielectric constant with a positive curvature is an important ingredient of the model .
it turns out that beyond a critical value of the electric field the electrostatic energy dominates and drives phase separation and pattern formation .
the approach developed in this paper is thermodynamic , based on a macroscopic free energy . analyzing equilibrium properties yields a phase diagram which exhibits instabilities that can lead to pattern formation .
some understanding of the dynamics of pattern formation then comes from investigating the corresponding time dependent ginzburg - landau equation .
we begin by describing the model free energy and its equilibrium properties , and then turn to dynamics .
consider a binary dielectric liquid as described above placed between parallel capacitor plates separated by @xmath4 .
the liquid has capacitance @xmath5 when the internal energy is dominated by the electrostatic contribution , the helmholtz free energy is @xmath6 where @xmath7 is the voltage between the plates , @xmath8 the temperature , and @xmath9 the entropy .
the equilibrium state is obtained by minimizing @xmath10 at constant @xmath11 . for our purposes
it is more convenient to make a legendre transformation to a different free energy : @xmath12 equilibrium is found by minimizing @xmath13 at constant @xmath14 ( note that @xmath7 is voltage and not volume ) .
we will refer to @xmath13 as `` the '' free energy . for the dielectric constant @xmath15 we use a phenomenological model due to lichteneker@xcite : @xmath16 this interpolates between the dielectric constant of pure solvent ( @xmath17 ) and that of pure particle ( @xmath18 ) as the concentration varies from @xmath19 to @xmath20 . in this paper
we use a physically reasonable value @xmath21@xcite . the specific choice of @xmath22 in eq .
( [ epsilon ] ) is not essential .
the results we report turn out to be largely independent of the form of @xmath22 , as long as it has positive curvature ( a point we discuss later ) .
we use an approximate entropy functional appropriate for interacting particles at low concentration @xcite : @xmath23 . \label{s}\ ] ] this expression can be obtained by a simple counting argument . in the absence of interactions
@xmath24 is the particle volume and eq .
( [ s ] ) is the entropy of free particles @xcite .
here however @xmath24 is instead the so - called ` correlation volume , ' which accounts phenomenologically for the particle interactions . in a typical experiment
the plate separation @xmath4 is so small that the concentration @xmath1 can be treated as effectively two - dimensional
. then we can let the coordinate @xmath25 denote lateral position within the plane . in this limit
the electric field @xmath26 becomes uniform ( @xmath27 ) , and the free energy becomes [ omega ] @xmath28.\end{aligned}\ ] ] for @xmath22 given in eq .
( [ epsilon ] ) , the temperature and electric field naturally scale to the dimensionless form @xmath29
we are interested in potential phase separations in the system
. these can occur if @xmath30 possesses inflection points@xcite .
suppose that in equilibrium the system phase separates into volume fractions @xmath31 with concentrations @xmath32 .
if the average concentration is @xmath33 , then these must satisfy @xmath34 and so @xmath35 to find the conditions under which the equilibrium state is phase - separated in this manner , we minimize the free energy @xmath36 with respect to @xmath37 , subject to the constraints eq .
( [ constraint ] ) . using lagrange multipliers , this means minimizing @xmath38 . \label{tomega}\ ] ] except at the endpoints ( @xmath39 or @xmath40 equal to 0 or 1 ) this minimization yields [ f ] @xmath41 if @xmath42 and @xmath43 both satisfy eq .
( [ fa ] ) , then @xmath44 this result is summarized by a simple geometrical construction illustrated in fig .
[ inflection ] .
if the straight line joining the points @xmath45 and @xmath46 is tangent to @xmath30 at both points , then phase separation will occur for @xmath47 . this can occur only if @xmath30 has two or more inflection points ( see fig . [ inflection ] ) . for our model , the endpoints @xmath48 never correspond to a minimum free - energy state .
consider now the free energy density @xmath30 for our dielectric fluid , eq .
( [ fc ] ) . the entropic term @xmath49 has positive curvature .
at high temperatures this term dominates , @xmath30 has positive curvature everywhere , and the equilibrium state is homogeneous .
this is illustrated by the highest - temperature trace ( @xmath50 ) in fig .
[ fplot ] , which uses the model dielectric function eq .
( [ epsilon ] ) .
as the temperature is lowered the electrostatic energy term @xmath51 plays an increasingly important role .
when @xmath22 has positive curvature , the electrostatic term has negative curvature , and at low temperatures this term leads to inflection points and phase separation .
this is illustrated by the two lower - temperature traces in fig .
[ fplot ] . the phase diagram for the free energy eq .
( [ fc ] ) , developed using the above procedure , is shown in fig .
[ phase ] for @xmath21 .
below the coexistence curve ( the solid line ) the homogeneous phase becomes metastable and in equilibrium the system phase - separates .
this curve is the locus of points @xmath32 obtained from solving eq .
( [ f ] ) at all temperatures .
although there is no exact expression for the coexistence temperature , for the physically important case of low concentrations we find approximately @xmath52 for @xmath21 this is accurate for concentrations @xmath53 .
lying below the dashed phase boundary in fig .
[ phase ] is the classical spinodal region , where the homogeneous state becomes unstable [ @xmath54 .
this phase boundary is given by the condition @xmath55 , which yields @xmath56 the coexistence and spinodal curves touch at a critical point ( marked by a diamond in fig .
[ phase ] ) .
this occurs at the critical concentration @xmath57 and critical temperature @xmath58 where the two inflection points present at lower temperatures coincide [ and so @xmath59 . for the model dielectric function eq .
( [ epsilon ] ) the critical values are : [ crit ] @xmath60 the instabilities give the possibility of pattern formation , which we now discuss .
pattern formation after quenches into the metastable or spinodal regions provides important insight into the underlying mechanisms for phase transformation in the system . for a system like the one modeled here
it is experimentally easier to increase the voltage suddenly while holding the temperature fixed ; by eq .
( [ t ] ) this is equivalent to a quench . here
we investigate pattern formation under quenches using a phenomenological diffusion equation developed from the free energy .
notice first that when inhomogeneities arise the free energy needs an additional @xmath61 term .
this can be viewed as the surface energy between domains of different concentration ; such terms arise generally in any interacting system . consequently in this section
we write the free energy @xmath62 , \label{omegak}\ ] ] where @xmath63 is a phenomenological parameter .
fick s law relates the concentration current density to the local chemical potential @xmath64 : @xmath65 where @xmath66 is the mobility ( taken to be constant ) and @xmath67 is the volume of the colloidal particle .
combining this with the continuity equation @xmath68 yields a diffusion equation @xmath69 the local chemical potential is obtained from the free energy : @xmath70 . \label{mu}\ ] ] inserting this into eq .
( [ de ] ) gives @xmath71 .
\label{tdgl}\ ] ] the free energy density @xmath30 is in eqs .
( [ fc],[epsilon ] ) .
a convenient dimensionless form comes from writing time in units of @xmath72 and length in units of @xmath73 , where [ units ] @xmath74 [ diffusion ] then the diffusion equation becomes @xmath75 , \label{gleqn}\ ] ] where @xmath76.\ ] ] eq .
( [ gleqn ] ) is nothing but the time - dependent ginzburg - landau equation for a scalar conserved order parameter ; however , the free energy in this particular case is constructed with the parameters directly relevant to experiments on dielectric colloidal fluids@xcite .
we study pattern formation after quenches by solving this nonlinear diffusion equation , starting from a uniform distribution ( plus a small random component ) .
one can get some intuition about dynamics by linearizing the diffusion equation@xcite .
the result is accurate for early times , before the nonlinear terms become significant .
write @xmath77 , where @xmath33 is the average concentration . for
small @xmath78 expand @xmath79 where @xmath80\\ \label{b } b & = & -\tilde{f}''(c ) = \frac{\tilde{t}_s-\tilde{t}}{c(1-c ) } .\end{aligned}\ ] ] @xmath81 is given in eq .
( [ ts ] ) . the linear and constant terms in eq .
( [ fseries ] ) play no role in the dynamics . keeping only the second order term from eq .
( [ fseries ] ) , the free energy eq .
( [ omegak ] ) becomes @xmath82 .
\label{omegaapprox}\ ] ] this expansion connects our model to the coarsening dynamics of a binary alloy described by a conserved scalar order parameter @xmath83 .
the free energy for such a system is generally assumed to have the ginzburg - landau form @xmath84 \label{bf}\ ] ] where @xmath63 , and @xmath85 are assumed to be positive . below the critical temperature @xmath86 becomes positive , yielding a broken symmetry .
higher - order terms beyond @xmath87 are irrelevant for the dynamic universality class .
in fact , the early - time dynamics are dominated by @xmath88 terms . from the correspondence between eqs .
( [ omegaapprox ] ) , ( [ bf ] ) we expect our model to have early - time dynamics very similar to those of model b. using the expansion eq .
( [ omegaapprox ] ) linearizes the diffusion equation eq .
( [ diffusion ] ) : @xmath89.\ ] ] this is solved by expanding the fluctuation @xmath78 in fourier components : @xmath90 we find @xmath91 outside of the spinodal region @xmath92 and so by eq .
( [ b ] ) @xmath93 .
consequently all components @xmath94 decay .
this describes the homogeneous equilibrium state : fluctuations away from a uniform concentration decay .
however , within the classical spinodal region @xmath95 .
thus long wavelength modes , those with wave vector @xmath96 , grow exponentially .
the most rapid growth occurs for @xmath97 , where @xmath98 reaches its maximum value @xmath99 .
thus in the classical spinodal regime , where the homogeneous state is unstable , early - time exponential growth leads to structures a typical ( dimensionless ) length scale @xmath100 .
these structures grow exponentially in time with a ( dimensionless ) time scale @xmath101 .
for example , consider the structure factor @xmath102 here angle brackets represent a statistical average .
suppose that the initial fluctuations @xmath103 are small and uncorrelated : @xmath104 then for early times @xmath105 this grows ever more peaked at @xmath106 as time increases .
fourier transforming gives the real - space correlation function @xmath107 which exhibits the length and time scales described above .
including dimensions , the structures developing in early times have a length scale @xmath108 and grow with a time scale @xmath109 ^ 2 = \frac{\kappa}{mv_0 ^ 2 } \left [ \frac{2c(1-c)k_b / v}{t - t_s } \right]^2 .\ ] ] deep quenches into the classical spinodal region should lead to dynamically - developing structures on a length scale @xmath110 given by eq .
( [ l ] ) .
these structures result from a long - wavelength instability .
they represent not the equilibrium state , but the early - time evolution from the highly non - equilibrium homogeneous state toward the ( eventual ) phase - separated equilibrium .
the linearized description is accurate for very small initial times .
soon the fluctuations grow large enough for the nonlinear terms to play a role , and we turn to numerical means for the solution of eq .
( [ diffusion ] ) .
we numerically integrate the lattice discretized version of eqs .
( [ diffusion ] ) using a first - order euler scheme , choosing step sizes to avoid unphysical instabilities@xcite .
this algorithm is adequate for times during which the system begins to organize into patterns . to carry the simulation
all the way to phase separation requires a more sophisticated approach@xcite .
however the model we use also omits hydrodynamic effects which may be important for the late time coarsening process . consequently here we present results during the era of pattern formation only .
the initial configuration consists of a uniform concentration @xmath33 plus a small random component . starting from this initial state , eq .
( [ diffusion ] ) is integrated for various values of @xmath111 and @xmath112 . in agreement with the results of the previous sections we find that phase separation occurs below the coexistence temperature @xmath113 ( or , equivalently , above a threshold electric field ) . in the phase - separation region
the concentration develops labyrinthine patterns consisting of stripes of higher and lower concentration .
typical results for a quench deep into the phase separation region are shown in figs .
[ contrast ] and [ snapshot ] .
the latter clearly shows that a labyrinthine pattern has developed .
we monitor two quantities during the simulation . at each time step
we search for the maximum and minimum concentrations , and define a contrast parameter @xmath114 .
this parameter indicates the visibility of the developing pattern .
another quantity of interest is the pair correlation function @xmath115 the average domain width @xmath116 can be calculated from the first zero of the normalized pair correlation function @xmath117 .
the upper panel of fig .
[ contrast ] shows the early time variation of the contrast parameter and the domain width @xmath116 as a function of the scaled time .
the average domain width after a rapid initial increase saturates at this temperature to a value @xmath118 .
we have studied the electric - field - induced phase separation of a dielectric binary fluid using a coarse - grained free energy functional .
the free energy investigated is guided by experiments on field - driven instabilities in ferro - colloids , and shares features of a ginzburg - landau free energy that describes the phase separation of a binary mixture .
certain other features of ferro - particles which would require a more complicated order parameter have been omitted , _
e.g. _ the orientational ordering arising from the magnetic dipolar interaction . the first order co - existence curve of the model clearly shows how a spinodal decomposition can be driven by an electric field in colloidal systems .
analytical estimates of the critical field are consistent with the numerical results .
deep quenches show a typical labyrinthine pattern formation with characteristic length and time scales set by the free energy .
the phenomenology used here is most accurate at low particle concentrations . in particular ,
the entropy is essentially that of free particles , modified to take interactions into account _ via _ the phenomenological correlation volume @xmath24 .
the resulting entropy is at least qualitatively correct at low concentrations but not high .
experiments with low overall concentrations should be described with reasonably accuracy by this model .
we thank geoff canright and john evans for helpful conversations .
we acknowledge support from the nsf through grants dmr99 - 72683 ( mdj ) , dmr00 - 72901 ( wl ) , and nsf - nirt ( eng / ecs and cise / eia ) grant 0103587 ( ab and wl ) .
brett riley , aniket bhattacharya , m.d .
johnson , x. duan , and w. luo , in _ electrorheological fluids and magnetorheological suspensions : proceedings of the eighth international conference , nice , france 9 - 13 july 2001 _ , ed . by g. bossis ( world scientific , 2002 ) ; int .
b * 16 * , 2041 ( 2002 ) ; _
vs_. @xmath33 for the three scaled temperatures @xmath120 , using eqs .
( [ fc],[epsilon ] ) with @xmath21 . at the two lower temperatures
two inflection points are present .
these are more easily seen in ( b ) @xmath121 .
the inflection points are indicated by filled circles . ] and domain size ( inset ) as a function of time . here
time and space are in the units of eq .
( [ units ] ) .
the simulation was done on a 128@xmath122128 lattice with an initial random configuration @xmath123 around @xmath124 for @xmath125 and @xmath126 . ] | an electric - field - induced phase transition and pattern formation in a binary dielectric fluid layer are studied using a coarse - grained free energy functional .
the electrostatic part of the free energy is a nonlinear functional of the dielectric function , which depends in turn on the local colloidal concentration .
we determine the phase co - existence curve and find that beyond a critical electric field the system phase separates . accompanying the phase separation are patterns similar to those observed in a spinodal decomposition of an ordinary binary fluid .
the temporal evolution of the phase separating patterns are discussed both analytically and numerically by integrating a cahn - hilliard type of equation . |
the current census of white dwarfs in the solar neighborhood is believed to be complete only to about 13 pc , and is measurably incomplete within 20 pc @xcite .
the local sample of white dwarfs appears to be an old population consisting of mostly cool white dwarfs .
the average absolute magnitude for the local census of white dwarfs is @xmath2 which corresponds to a temperature of @xmath3 k. a significant fraction of white dwarfs within 13 pc ( * ? ? ? *
@xmath4 ) show the presence of a magnetic field .
@xcite determined that over a quarter of the white dwarfs within 20 pc are in binary systems .
based on recent literature , we updated the spectral classifications of the sample of white dwarfs within 20 pc @xcite and found that about 70% of the stars are hydrogen - rich .
the remaining 30% are helium - rich , and out of these helium rich white dwarfs about half are dc white dwarfs , about a third are cool dq white dwarfs , while the remainder are dz white dwarfs .
the new luyten two - tenths ( nltt ) catalog has been used to search for nearby cool dwarfs and subdwarfs @xcite and for white dwarfs @xcite .
however , a large number of objects still require a formal spectroscopic confirmation . recently
, @xcite have revised the coordinates and proper motions of most stars in the nltt catalog by cross - correlating the catalog with the two micron all sky survey ( 2mass ) and the usno - a catalogs .
one problem with the original catalog was that there was no obvious separation between different populations because the r and b photographic bands do not provide a broad enough baseline to distinguish between main - sequence , subdwarf and white dwarf populations .
@xcite showed that using an optical / infrared ( @xmath5 ) reduced proper motion diagram they could distinguish between these groups of stars and from this diagram they listed 23 white dwarf candidates within 20 pc from the sun . @xcite and @xcite found that only a third of these nearby candidates are white dwarfs , with the remainder being cool red dwarfs and subdwarfs .
recently , several projects have been initiated to search for high - proper motion stars , such as the supercosmos - recons survey @xcite and lspm catalog @xcite which utilizes the supercosmos sky survey @xcite and superblink @xcite , respectively .
these surveys have detected many high - proper motion stars which are either nearby or have been labeled as halo candidates .
one such object is the cool white dwarf wd 0346@xmath6246 @xcite , which has definite halo kinematics and is very cool ( @xmath7 k ) and hence old enough to belong to the halo .
a total of 299 new stars with @xmath8 yr@xmath1 have been discovered using the supercosmos - recons survey @xcite , out of which 148 have proper motions greater than @xmath9yr@xmath1 and 43 are within 25 pc from the sun .
@xcite reported the discovery of 198 new stars with proper motions greater than @xmath9yr@xmath1 using superblink .
their survey also included the nearby halo white dwarf pm j13420@xmath103415 @xcite .
the nltt catalog appears to be complete at galactic latitudes @xmath11 , however at low galactic latitudes the catalog appears to be significantly incomplete . on the other hand , @xcite notes
that the completeness for white dwarfs is very high , i.e. , the coverage of white dwarfs appears to be completely uniform .
the reason for this may be that luyten concentrated on blue objects , which are no more common in the plane than elsewhere , and hence luyten was able to detect most high - proper motion white dwarfs above @xmath12 . another survey to search for cool white dwarfs with high - proper motion using the guide star catalog ii was initiated by @xcite with the aim of placing constraints on the halo white dwarf space density .
their search resulted in the discovery of 24 new white dwarfs .
many earlier proper motion surveys , in particular the luyten proper motion surveys relied on plate pairs taken a decade or more apart , and very high - proper motion stars were most likely lost in the background .
a search for ultra - high proper motion stars was performed by @xcite using the skymorph database of the near earth asteroid tracking ( neat ) project @xcite and found a main - sequence star with spectral type m6.5 and with a proper motion of @xmath13 at a distance of @xmath14 pc which they obtained from the trigonometric parallax .
since most of the stars in the nltt catalog are brighter than 19th magnitude , we do not expect to find many halo white dwarf candidates .
since the halo star formation history is assumed to be a burst occurring 12 gyrs ago , then most of the halo white dwarfs would be required to have cooling ages greater than 10 gyrs ( and hence @xmath15 ) .
given that the percentage of halo white dwarfs in the local population is @xmath16 @xcite , then most halo candidates are likely to be at large distances and hence too faint to be included in the nltt catalog . in this paper , we pursue our study of white dwarf candidates @xcite from the revised nltt ( rnltt ) catalog of @xcite .
some preliminary results were presented in @xcite .
we present the list of white dwarfs observed in 2 and their spectroscopic observations in 2.1 with complementary optical and infrared photometric data presented in 2.2 .
the model atmospheres and spectra we used to analyze our data are described in 3 and in 4 we present our analysis of the new white dwarf stars .
we discuss our findings in 5 and summarize in 6 .
table [ tbl1 ] lists the white dwarf candidates for which we have obtained a spectroscopic identification .
many of the white dwarfs were listed in the luyten white dwarf catalogs ( lwdc ) @xcite , and required spectroscopic confirmation of their classification .
lllccc 287 & lp 464 - 57 & 0.424 , 124.0 & @xmath17 & @xmath18 & @xmath19 + 529 & gd 5 , lp 192 - 41 & 0.237 , 193.4 & @xmath20 & @xmath21 & @xmath22 + 3022 & g 240 - 93 , phl 3101 & 0.210 , 268.9 & @xmath23 & @xmath24 & @xmath25 + 3915 & lp 294 - 61 & 0.227 , 219.3 & @xmath26 & @xmath27 & @xmath28 + 6275 & g 72 - 40 , g94 - 21 & 0.213 , 76.6 & @xmath29 & @xmath30 & @xmath31 + 8435 & lhs 1421 & 0.611 , 190.0 & @xmath32 & @xmath33 & @xmath34 + 8581 & g 36 - 29 & 0.340 , 117.0 & @xmath35 & @xmath36 & @xmath37 + 8651 & gd 32 & 0.172 , 235.7 & @xmath38 & @xmath39 & @xmath40 + 9933 & lp 299 - 22 & 0.190 , 119.6 & @xmath41 & @xmath42 & @xmath43 + 10398 & gd 44 & 0.176 , 153.7 & @xmath44 & @xmath45 & @xmath46 + 14307 & g 84 - 26 & 0.293 , 142.9 & @xmath47 & @xmath48 & @xmath49 + 14352 & lp 656 - 32 & 0.239 , 157.6 & @xmath50 & @xmath51 & @xmath52 + 17285 & lp 661 - 15 & 0.211 , 312.2 & @xmath53 & @xmath54 & @xmath55 + 18555 & lp 207 - 50 & 0.420 , 165.8 & @xmath56 & @xmath57 & @xmath58 + 18642 & lp 207 - 55 , kuv 07531 + 4148 & 0.322 , 181.4 & @xmath59 & @xmath60 & @xmath61 + 19019 & lp 311 - 2 , gd 262 & 0.182 , 113.3 & @xmath62 & @xmath63 & @xmath64 + 19138 & g 111 - 64 & 0.258 , 166.5 & @xmath44 & @xmath65 & @xmath66 + 20165 & kuv
08422 + 3813 & 0.217 , 152.4 & @xmath67 & @xmath68 & @xmath69 + 21241 & lp 210 - 58 & 0.200 , 258.6 & @xmath70 & @xmath71 & @xmath72 + 27901 & gd 311 , lp 94 - 285 , lb 2052 & 0.232 , 238.7 & @xmath73 & @xmath74 & @xmath75 + 28772 & lp 129 - 587 , g197 - 35 & 0.300 , 266.8 & @xmath23 & @xmath76 & @xmath77 + 28920 & lp 19 - 411 & 0.282 , 209.3 & @xmath78 & @xmath79 & @xmath80 + 29233 & g122 - 61 , cbs 451 & 0.300 , 262.8 & @xmath81 & @xmath82 & @xmath83 + 30738 & lp 435 - 109 & 0.162 , 260.8 & @xmath84 & @xmath85 & @xmath86 + 31347 & lp217 - 47 , lhs 5222 & 0.506 , 315.0 & @xmath87 & @xmath88 & @xmath89 + 31748 & pg 1242 - 106 & 0.349 , 257.1 & @xmath90 & @xmath91 & @xmath92 + 32390 & lp 96 - 9 & 0.233 , 243.3 & @xmath93 & @xmath94 & @xmath95 + 32695 & lp 616 - 70 & 0.190 , 271.9 & @xmath96 & @xmath97 & @xmath98 + 34264 & lp 617 - 70 & 0.230 , 175.5 & @xmath99 & @xmath100 & @xmath101 + 34564 & lp 678 - 8 & 0.245 , 280.3 & @xmath102 & @xmath103 & @xmath104 + 35880 & lp 97 - 430 & 0.309 , 281.0 & @xmath105 & @xmath106 & @xmath107 + 38499 & lp 801 - 14 , ec 14473 - 1901 & 0.269 , 288.0 & @xmath108 & @xmath109 & @xmath110 + 40489 & lp 503 - 7 , gd 184 & 0.179 , 177.2 & @xmath59 & @xmath111 & @xmath112 + 40607 & g 137 - 24 & 0.248 , 221.6 & @xmath113 & @xmath114 & @xmath115 + 40881 & lp 273 - 64 , ltt 14655 , gd 187 & 0.186 , 295.4 & @xmath116 & @xmath117 & @xmath118 + 40992 & lp 384 - 38 & 0.417 , 173.2 & @xmath119 & @xmath120 & @xmath121 + 41800 & g 202 - 26 & 0.318 , 330.3 & @xmath122 & @xmath123 & @xmath124 + 42050 & lp 274 - 53 & 0.255 , 172.7 & @xmath119 & @xmath125 & @xmath126 + 42153 & lp 861 - 31 & 0.194 , 312.4 & @xmath127 & @xmath128 & @xmath129 + 43806 & lp 276 - 33 & 0.329 , 177.6 & @xmath130 & @xmath131 & @xmath132 + 43827 & lp 387 - 21 & 0.208 , 164.7 & @xmath133 & @xmath134 & @xmath135 + 43985 & lp 331 - 27 & 0.343 , 224.8 & @xmath136 & @xmath137 & @xmath138 + 44000 & g 203 - 39 & 0.385 , 228.9 & @xmath139 & @xmath140 & @xmath141 + 44149 & lp 276 - 48 & 0.250 , 135.3 & @xmath142 & @xmath143 & @xmath144 + 44447 & lp 226 - 48 & 0.195 , 355.4 & @xmath145 & @xmath146 & @xmath147 + 44986 & eg 545 , gd 362 , pg 1729 + 371 & 0.224 , 173.1 & @xmath38 & @xmath148 & @xmath149 + 45344 & lp 227 - 31 & 0.217 , 189.1 & @xmath150 & @xmath151 & @xmath152 + 45723 & lp 228 - 12 & 0.334 , 2.9 & @xmath153 & @xmath154 & @xmath155 + 49985 & lp 872 - 20 & 0.278 , 130.0 & @xmath156 & @xmath157 & @xmath158 + 50029 & lp 872 - 48 , bps cs 22880 - 0126 & 0.136 , 3.1 & @xmath159 & @xmath160 & @xmath161 + 51103 & lp 637 - 32 & 0.291 , 209.2 & @xmath162 & @xmath163 & @xmath164 + 52404 & lp 930 - 61 & 0.234 , 44.0 & @xmath165 & @xmath166 & @xmath167 + 53177 & lp 759 - 50 , he 2209 - 1444 & 0.267 , 70.4 & @xmath168 & @xmath169 & @xmath170 + 53447 & lp 287 - 39 & 0.463 , 79.3 & @xmath171 & @xmath172 & @xmath173 + 53468 & lp 619 - 64 , phl 218 & 0.249 , 72.0 & @xmath174 & @xmath175 & @xmath176 + 53996 & lp 400 - 6 & 0.320 , 134.9 & @xmath177 & @xmath178 & @xmath179 + 54047 & lhs 3821 , lp 520 - 28 & 0.538 , 68.4 & @xmath180 & @xmath181 & @xmath182 + 55932 & g 275 - 8 , lp 877 - 69 , ltt 9373 & 0.339 , 106.0 & @xmath183 & @xmath184 & @xmath185 + 56122 & lp 522 - 17 & 0.294 , 206.3 & @xmath186 & @xmath187 & @xmath188 + 56805 & lp 522 - 46 & 0.352 , 71.0 & @xmath189 & @xmath190 & @xmath191 + 58283 & lp 463 - 68 & 0.225 , 148.3 & @xmath192 & @xmath193 & @xmath194 + the rnltt white dwarf candidates were observed using the dual imaging spectrogram ( dis ) attached to the 3.5 m telescope at the apache point observatory ( apo ) on 2004 may 31 , june 1 , 2005 may 12 , 16 , june 2 , july 2 , 15 , august 9 and november 12 , 14 .
we used 300 lines / mm gratings to obtain a spectral range of 3700 to 5600 with a dispersion of 2.4 per pixel in the blue , and a spectral range of 5300 to 9800 with a dispersion of 2.3 in the red .
we also obtained spectroscopy by cross - correlating the list of rnltt white dwarf candidates with the sloan digital sky survey ( sdss ) .
we used the archival spectra from the 4th data release of sdss in our analyses of these objects .
figure [ fig1 ] shows the spectra of all the hydrogen - rich white dwarfs , and figure [ fig2 ] shows the spectra of the helium - rich white dwarfs obtained at apo .
table [ tbl1 ] lists the @xmath195 magnitudes and @xmath196 photometry for the rnltt white dwarf candidates for which we have obtained spectroscopy .
the table also lists the stars alternate names and their proper motion taken from rnltt catalog of @xcite .
the @xmath195 magnitudes are taken from the rnltt catalog of @xcite , except for stars where more accurate photometry from another source was available .
the @xmath196 photometry was obtained from the 2mass database available at the centre de donnes astronomique de strasbourg .
the data were converted from 2mass system to the cit system using color transformations provided by @xcite .
we cross - correlated the list of rnltt white dwarf candidates with the sdss to obtain _ ugriz _ photometry ( table [ tbl3 ] ) .
llccc 3022 & j005438.84 - 095219.8 & @xmath197 & @xmath198 & @xmath199 + 18555 & j075313.27 + 423001.5 & @xmath200 & @xmath201 & @xmath202 + 18642 & j075631.11 + 413950.9 & @xmath203 & @xmath204 & @xmath205 + 19019 & j080946.19 + 292032.1 & @xmath206 & @xmath207 & @xmath208 + 19138 & j081411.16 + 484529.6 & @xmath209 & @xmath210 & @xmath211 + 20165 & j084524.62 + 380156.0 & @xmath212 & @xmath213 & @xmath214 + 21241 & j091356.83 + 404734.6 & @xmath215 & @xmath216 & @xmath217 + 27901 & j113534.61 + 572451.7 & @xmath218 & @xmath219 & @xmath220 + 28772 & j115123.93 + 541147.7 & @xmath221 & @xmath222 & @xmath223 + 29233 & j120003.28 + 433541.5 & @xmath224 & @xmath225 & @xmath226 + 31347 & j123752.12 + 415625.8 & @xmath227 & @xmath228 & @xmath229 + 32390 & j125629.36 + 610200.9 & @xmath230 & @xmath231 & @xmath232 + 32695 & j130247.97 - 005002.7 & @xmath233 & @xmath234 & @xmath235 + 34264 & j132937.15 - 013430.5 & @xmath236 & @xmath237 & @xmath238 + 35880 & j135758.43 + 602855.2 & @xmath239 & @xmath240 & @xmath241 + 40881 & j154033.31 + 330852.8 & @xmath242 & @xmath243 & @xmath244 + 40992 & j154234.65 + 232939.8 & @xmath245 & @xmath246 & @xmath247 + 41800 & j160112.70 + 531700.0 & @xmath248 & @xmath249 & @xmath250 + 42050 & j160714.18 + 342345.7 & @xmath251 & @xmath252 & @xmath253 + 43806 & j165445.69 + 382936.5 & @xmath254 & @xmath255 & @xmath256 + 43827 & j165538.92 + 253345.9 & @xmath257 & @xmath258 & @xmath259 + 43985 & j165939.99 + 320320.0 & @xmath260 & @xmath261 & @xmath262 + 44149 & j170447.70 + 360847.4 & @xmath263 & @xmath264 & @xmath265 + 56122 & j231206.08 + 131057.6 & @xmath266 & @xmath267 & @xmath268 + 56805 & j232519.88 + 140339.7 & @xmath269 & @xmath270 & @xmath271 +
the stars presented in this paper have been analyzed for their @xmath272 , @xmath273 using a grid of computed pure hydrogen lte plane - parallel models . the grid of models extend from @xmath274 to 6500 k ( in steps of 500 k ) at @xmath275 and @xmath276 , from @xmath277 to 16000 k ( in steps of 1000 k ) , from 18000 to 32000 k ( in steps of 2000 k ) and from 36000 to 84000 k ( in steps of 4000 k ) at @xmath278 to @xmath279 ( in steps of 0.25 dex ) .
convective energy transport in cooler atmospheres is included by applying the schwartzschild stability criterion and by using the mixing length formalism described by @xcite .
we have assumed the ml2 parameterization of the convective flux @xcite and adopting @xmath280 @xcite , where @xmath281 is the ratio of the mixing length to the pressure scale height .
the equation of convective energy transfer was fully linearized within the feautrier solution scheme and subjected to the constraint that @xmath282 , where @xmath283 is the convective flux and @xmath284 is the radiative flux .
the dissolution of the hydrogen energy levels in the high - density atmospheres of white dwarfs was calculated using the formalism of ( * ? ? ?
* hereafter hm ) and following the treatment of ( * ? ? ?
* hereafter hhl ) ) . for the sake of reproducibility we now provide details of the calculations as included in our fortran code .
that is we calculated the occupation probability from charged particles of level @xmath285 using the form ( hm eq . 4.26 and hhl eq .
a.2 ) : @xmath286 @xmath287 is the holtzmark distribution function .
@xmath288 can be calculated using : @xmath289 where @xmath290 is the ionic charge , @xmath291 is the electron number density @xmath292 is the ion number density and @xmath293 is bohr radius ( @xmath294 cm ) . and
assuming that @xmath295 we can simplify @xmath288 to : @xmath296 @xmath297 is defined by : @xmath298 we can approximate @xmath299 by the rational expression presented by hhl : @xmath300 where @xmath301 is given by : @xmath302 and @xmath303 and @xmath304 is the correlation parameter which is defined as @xmath305 here , @xmath306 is the temperature and @xmath307 is the number density of neutral hydrogen .
we introduced the scaling factor @xmath308 to the correlation parameter @xmath304 which can have values between 0 and 1 . in calculating our model spectra
we set @xmath309 . lowering the value for @xmath308 will generate lower values in surface gravity measurements .
hm also considered the contribution to the occupation probability from neutral particles , which is necessary in the case of cool white dwarfs where neutral hydrogen becomes dominant .
therefore , the neutral particle contribution to the occupation probability is given by : @xmath310 where we can assume that the effective interaction radius is a fraction @xmath301 of the hydrogen atomic radius which is associated with level @xmath285 i.e. , @xmath311 . for our calculations
we have assumed @xmath312 ( see * ? ? ?
* ) , and therefore @xmath313 and finally , the combined occupation probability is the product of the occupation probabilities from neutral and charged particles , i.e. , @xmath314 since the occupation probability depends on the number density of neutral hydrogen @xmath307 , we first obtain an estimate of the populations for all levels by including the contribution from only neutral particles , i.e. , @xmath315 .
we then repeat the calculations by including the contribution from both neutral and charged particles , until the @xmath316 remains constant for all @xmath285 .
the calculated level occupation probabilities are then explicitly included in the calculation of the line and continuum opacities . the balmer line profiles
are calculated using the tables of stark - broadened line profiles of @xcite convolved with normalized resonance line profiles .
we have adopted the stark - broadened line profiles of @xcite over those of @xcite because the tables of @xcite are complete at higher electron densities ( @xmath317 @xmath318 ) .
the tables of @xcite are incomplete at higher electron densities for many lines .
we modeled the heavy element lines using voigt profiles @xcite including stark and van der waals broadening parameters . in cool da white dwarfs which display metal lines van der waals broadening is dominant and the pertubers are neutral hydrogen atoms . for atmospheres dominated by hydrogen , we have used an approximation of the damping constant , @xmath319 as defined by @xcite : @xmath320 where @xmath321 is the pressure , t is the temperature and @xmath322 can be calculated using : @xmath323 here , @xmath324 is the ionization potential , @xmath325 is the excitation potential of the lower level for the heavy metal atom .
and @xmath326 is the energy of a photon in the line , that is @xmath327 , where @xmath328 is in .
the balmer lines of the hydrogen - rich white dwarfs were analyzed using a @xmath329 minimization technique and our cool hydrogen model atmospheres .
the quoted uncertainties are only statistical ( @xmath330 ) and do not take into account possible systematic effects in model calculations or data acquisition and reduction procedures . using the computed spectral grid , we have computed sdss synthetic colors . for the white dwarf candidates for which we obtained spectra and for which sdss photometry is available , we have compared the observed colors to the theoretical colors as a check of the temperatures obtained from the balmer line profile fits .
we have also calculated sdss colors for a black - body spectrum for temperatures ranging from 4000 to 36000 k ( in 1000 k intervals ) , and used these colors to obtain temperature estimates of dc white dwarfs .
the observed colors for a number of white dwarfs are compared to the synthetic @xmath331 versus @xmath332 colors in figure [ fig_gmr_umg_ext ] and @xmath333 versus @xmath332 colors in figure [ fig_gmr_rmi_ext ] .
the @xmath331/@xmath332 diagram shows that the observed @xmath331 is larger than the @xmath331 calculated from the model atmospheres for h - rich white dwarfs with temperatures less than @xmath277 k ( also noted by * ? ? ?
* ) , hence implying some missing opacity in the @xmath334 band .
@xcite suggests that a similar effect in the @xmath335 band may be due to the bound - free opacity associated with the dissolved atomic levels of hydrogen , i.e. , that following a bound - bound transition there is a probability that the upper level may be sufficiently perturbed by the surrounding particles that the electron will no longer be bound to the nucleus . for the stars with sdss @xmath336 photometry ,
the temperatures we obtained from spectroscopic fits were compared to the temperatures obtained using the @xmath337 and @xmath338 colors ( fig . [ fig_temp_comp ] ) .
for da white dwarfs , the label `` spectroscopic fits '' refers to temperatures obtained by fitting the balmer line profiles with model spectra , and for dc white dwarfs the same label refers to temperatures obtained by comparing a black - body to the observed spectrum .
the only point that appears to be in disagreement ( i.e. , the photometric temperature is lower than the spectroscopic temperature ) , is the ultramassive white dwarf nltt 43827 , which has a spectroscopic temperature of 11690 k. at this temperature the effect of gravity on the @xmath337 colors is the strongest and since we assumed @xmath339 in our photometric temperature estimates , it is likely to effect the final temperature estimate for white dwarfs with significantly different surface gravities in this region of the color diagram . in figure [ fig_temp_comp ] the temperatures determined from the photometry
appear to be slightly higher than those determined from spectroscopy for @xmath340 k and vice versa for @xmath341 k. this may be due to measurement errors or possibly to systematic errors in either spectroscopic ( i.e. , via the fitting of balmer lines with model spectra ) or photometric calibration . table [ tbl2 ] summarizes the results of the analyses of the white dwarfs .
the table also includes white dwarfs from @xcite and @xcite .
the temperatures and surface gravities of the da white dwarfs from these papers were remeasured with the improved synthetic spectra .
for the cool da white dwarfs the gravities went down significantly , in particular the gravity ( @xmath342 ) of nltt 49985 went down to @xmath343 and the high gravity should be discarded .
also the surface gravity of nltt 31748 went down from @xmath344 to @xmath345 .
lllccccccr 287 & apo & 0004@xmath6122 & dc & @xmath346 & ( 8.0 ) & ( 0.57 ) & 13.91 & 26 & -14 , -33 , -19 + 529 & apo & 0008@xmath6423 & da & @xmath347 & @xmath348 & @xmath349 & 13.96 & 18 & 20 , 3 , -10 + 3022 & sdss & 0052@xmath10101 & da & @xmath350 & @xmath351 & @xmath352 & 12.86 & 58 & 59 , 36 , 6 + 3915 & apo & 0108@xmath6277 & daz & @xmath353 & @xmath354 & @xmath355 & 15.41 & 14 & 21 , 5 , -3 , + 6275 & apo & 0150@xmath6256 & da & @xmath356 & @xmath357 & @xmath358 & 13.56 & 28 & -12 , -8 , 19 + 8435 & mso / sso & 0233@xmath10242 & dc & @xmath359 & ( 8.0 ) & ( 0.58 ) & 14.65 & 17 & 49 , -24 , -1 + 8581 & mso & 0236@xmath6259 & da & @xmath360 & ( 8.0 ) & ( 0.58 ) & 14.67 & 21 & -7 , -24 , 8 + 8651 & apo & 0237@xmath6315 & da & @xmath361 & @xmath362 & @xmath363 & 14.12 & 26 & 22 , 10 , -10 + 9933 & apo & 0304@xmath6317 & da & @xmath364 & @xmath365 & @xmath366 & 11.30 & 81 & -23 , -60 , 10 + 10398 & apo & 0313@xmath6393 & dc & @xmath367 & ( 8.0 ) & ( 0.58 ) & 12.02 & 56 & -4 , -33 , -16 + 14307 & apo & 0457@xmath10004 & da & @xmath368 & @xmath369 & @xmath370 & 14.09 & 17 & 17 , -17 , 10 + 14352 & apo & 0458@xmath10064 & da & @xmath371 & @xmath372 & @xmath373 & 14.19 & 33 & 28 , -28 , 4 + 17285 & apo & 0658@xmath10075 & dc & @xmath374 & ( 8.0 ) & ( 0.57 ) & 14.20 & 32 & -9 , 28 , -4 + 18555 & apo & 0749@xmath6426 & dc & @xmath375 & ( 8.0 ) & ( 0.57 ) & 16.20 & 18 & 9 , -31 , 8 + 18642 & apo & 0753@xmath6417 & da & @xmath376 & @xmath377 & @xmath378 & 14.68 & 24 & 5 , -30 , 0 + 19019 & sdss & 0806@xmath6294 & da & @xmath379 & @xmath380 & @xmath381 & 13.78 & 29 & 23 , -8 , 24 + 19138 & apo & 0810@xmath6489 & dc & @xmath382 & ( 8.0 ) & ( 0.57 ) & 13.31 & 23 & 8 , -23 , 11 + 20165 & sdss & 0842@xmath6382 & da & @xmath383 & @xmath384 & @xmath385 & 13.31 & 35 & 20 , -28 , 17 + 21241 & sdss & 0910@xmath6410 & da & @xmath386 & @xmath387 & @xmath388 & 13.72 & 51 & -23 , -4 , -27 + 27901 & sdss & 1132@xmath6574 & dc & @xmath389 & ( 8.0 ) & ( 0.58 ) & 12.73 & 61 & -32 , -47 , 7 + 28772 & sdss & 1148@xmath6544 & da & @xmath390 & @xmath391 & @xmath392 & 12.51 & 68 & -73 , -40 , -13 + 28920 & apo & 1151@xmath6795 & da & @xmath393 & @xmath394 & @xmath395 & 12.30 & 52 & -45 , -33 , -12 + 29233 & apo & 1157@xmath6438 & da & @xmath396 & @xmath397 & @xmath398 & 13.40 & 33 & -28 , -21 , -1 + 30738 & apo & 1223@xmath6188 & dc & @xmath399 & ( 8.0 ) & ( 0.57 ) & 14.12 & 26 & -5 , -7 , 5 + 31347 & apo & 1235@xmath6422 & dq & @xmath400 & ( 8.0 ) & ( 0.57 ) & @xmath401 & 27 & -52 , 19 , -7 + 31748 & sso & 1242@xmath10105 & da & @xmath402 & @xmath403 & @xmath404 & 12.72 & 22 & -19 , -20 , 2 + 32390 & sdss & 1254@xmath6613 & da & @xmath405 & @xmath406 & @xmath407 & 12.75 & 92 & -44 , -76 , 35 + 32695 & sdss & 1300@xmath10005 & da & @xmath408 & @xmath409 & @xmath410 & 12.32 & 80 & -51 , -33 , 11 + 34264 & sdss & 1327@xmath10013 & da & @xmath411 & @xmath412 & @xmath413 & 13.62 & 37 & 31 , -23 , -13 + 34564 & apo & 1333@xmath10054 & da & @xmath414 & @xmath415 & @xmath416 & 14.04 & 30 & -19 , -11 , 16 + 35880 & apo & 1356@xmath6607 & dc & @xmath417 & ( 8.0 ) & ( 0.57 ) & 13.77 & 59 & -63 , -39 , 20 + 38499 & apo & 1447@xmath10190 & da & @xmath418 & @xmath419 & @xmath420 & 12.97 & 37 & -21 , -15 , 37 + 40489 & apo & 1529@xmath6141 & da & @xmath421 & @xmath422 & @xmath423 & 14.89 & 22 & 23 , -7 , 1 + 40607 & mdm / sso & 1532@xmath6129 & dz & @xmath424 & ( 8.0 ) & ( 0.57 ) & 13.20 & 24 & 14 , -22 , 12 + 40881 & apo & 1538@xmath6333 & da & @xmath425 & @xmath426 & @xmath427 & 13.35 & 22 & -5 , -2 , 18 + 40992 & apo & 1540@xmath6236 & da & @xmath428 & @xmath429 & @xmath430 & 14.77 & 27 & 52 , -25 , -5 + 41800 & sdss & 1559@xmath6534 & da & @xmath431 & @xmath432 & @xmath433 & 14.04 & 31 & -36 , 3 , 14 + 42050 & apo & 1605@xmath6345 & da & @xmath434 & @xmath435 & ( 0.18 ) & 13.73 & 44 & 57 , -18 , 0 + 42153 & apo & 1607@xmath10250 & da & @xmath436 & @xmath437 & @xmath438 & 12.41 & 41 & -2 , 6 , 43 + 43806 & apo & 1653@xmath6385 & daz & @xmath439 & @xmath440 & @xmath441 & 14.91 & 15 & 31 , -4 , 4 + 43827 & apo & 1653@xmath6256 & da & @xmath442 & @xmath443 & @xmath444 & 14.29 & 33 & 37 , -7 , -7 + 43985 & apo & 1657@xmath6321 & da & @xmath445 & @xmath446 & @xmath447 & 15.20 & 22 & 28 , -22 , 23 + 44000 & apo & 1658@xmath6445 & da & @xmath448 & @xmath449 & @xmath450 & 14.30 & 31 & 38 , -32 , 39 + 44149 & apo & 1703@xmath6362 & dc & @xmath451 & ( 8.0 ) & ( 0.57 ) & 15.83 & 27 & 34 , 9 , -14 + 44447 & apo & 1713@xmath6393 & dap & @xmath452 & magnetic & ( 0.59 ) & 13.60 & 32 & -18 , 14 , 13 + 44986 & apo & 1729@xmath6371 & daz & @xmath453 & @xmath454 & @xmath455 & 14.54 & 22 & 31 , -2 , 0 + 45344 & apo & 1741@xmath6436 & da & @xmath456 & @xmath457 & @xmath458 & 14.59 & 34 & 43 , -6 , 6 + 45723 & apo & 1755@xmath6408 & dc & @xmath459 & ( 8.0 ) & ( 0.57 ) & 14.65 & 33 & -39 , 21 , 16 + 49985 & sso & 2048@xmath10250 & da & @xmath460 & @xmath461 & @xmath462 & 13.72 & 22 & -1 , -11 , -14 + 50029 & apo & 2049@xmath10222 & dc & @xmath463 & ( 8.0 ) & ( 0.58 ) & 12.49 & 28 & 5 , 22 , 11 + 51103 & apo & 2119@xmath10017 & da & @xmath464 & @xmath465 & @xmath466 & 14.09 & 32 & 43 , -24 , 5 + 52404 & sso & 2152@xmath10280 & dc & @xmath467 & ( 8.0 ) & ( 0.57 ) & 13.77 & 32 & -15,27,-5 + 53177 & mso & 2209@xmath10147 & da & @xmath468 & @xmath469 & @xmath470 & 12.89 & 28 & -21 , 12 , -7 + 53447 & apo & 2215@xmath6368 & dc & @xmath471 & ( 8.0 ) & ( 0.57 ) & 15.44 & 20 & -30 , -1 , -10 + 53468 & sso & 2215@xmath10204 & da & @xmath472 & @xmath473 & @xmath474 & 10.89 & 76 & -69,15,-35 + 53996 & apo & 2227@xmath6232 & da & @xmath475 & @xmath435 & ( 0.18 ) & 13.92 & 64 & -9 , -40 , -77 + 54047 & apo & 2228@xmath6151 & da & @xmath476 & @xmath477 & @xmath478 & 15.70 & 27 & -57 , 5 , -9 + 55932 & sso & 2306@xmath10220 & da & @xmath479 & @xmath480 & @xmath481 & 11.08 & 33 & -28 , -25 , -14 + 56122 & apo & 2309@xmath6129 & da & @xmath482 & @xmath483 & @xmath484 & 14.45 & 37 & 53 , -15 , -14 + 56805 & mso & 2322@xmath6137 & da & @xmath485 & @xmath435 & ( 0.18 ) & 14.27 & 20 & -23 0 , 4 + 58283 & apo & 2350@xmath6205 & da & @xmath486 & @xmath487 & @xmath488 & 13.31 & 45 & 6 , -27 , -28 + [ [ nltt-529 ] ] nltt 529 + + + + + + + + was observed by @xcite and they classified this object as a cool da white dwarf .
we have reobserved this star and fitted the balmer line profiles ( h@xmath281 - h9 ) to obtain an effective temperature of @xmath347 k and @xmath489 .
[ [ nltt-3022 ] ] nltt 3022 + + + + + + + + + is also known as g 240 - 93 and phl 3101 , and was observed as part of the sdss as sdss j005438.3 - 095219.8 .
@xcite obtained @xmath490 k and @xmath491 .
we fitted the balmer lines with model spectra to obtain @xmath492 and @xmath493 ( see figure [ fig_sdss_balmer ] ) .
[ [ nltt-3915 ] ] nltt 3915 + +
+ + + + + + + is also known as lp 294 - 61 and its spectrum in figure [ fig1 ] shows that this is a cool daz white dwarf . we first fitted the balmer lines ( h@xmath281 and h@xmath288 ) with model spectra to obtain an effective temperature of @xmath494 and a surface gravity of @xmath495 .
we then calculated a series of spectra with a varying abundance of sodium , for @xmath496 k and @xmath497 .
we compared these synthetic spectra to the observed spectrum to find that @xmath498 .
[ [ nltt-8651 ] ] nltt 8651 + + + + + + + + + is also known as gd 32 and was listed as a white dwarf suspect by @xcite , however it was not listed in the lwdc .
figure [ fig1 ] shows that this is a cool da white dwarf .
we fitted the balmer lines with model spectra to obtain an effective temperature of @xmath499 k and surface gravity of @xmath500 .
[ [ nltt-10398 ] ] nltt 10398 + + + + + + + + + + is also known as gd 44 and was listed as a white dwarf suspect by @xcite .
figure [ fig2 ] shows this object to be a dc white dwarf . to obtain a temperature estimate we compared the optical / infrared colors ( @xmath5/@xmath501 ) to blackbody colors and we only obtain a very uncertain termperature @xmath502 k , but comparing the spectrum to a blackbody we obtain a temperature estimate of @xmath367 k. [ [ nltt-14307 ] ] nltt 14307 + + + + + + + + + + is also known as g 84 - 26 .
@xcite obtained @xmath503 photometry ( @xmath504 , @xmath505 , @xmath506 ) and classified it as a white dwarf based on these photometric colors .
figure [ fig1 ] shows that this object is a da white dwarf . we fitted the balmer lines with model spectra and obtained an effective temperature of @xmath507 k and surface gravity of @xmath508 .
[ [ nltt-17285 ] ] nltt 17285 + + + + + + + + + + is also known as lp 661 - 15 and it was not listed in the lwdc .
figure [ fig2 ] shows this object to be a dc white dwarf .
we obtained a temperature estimate of @xmath509 by comparing the optical / infrared colors ( @xmath5/@xmath501 ) to calculated blackbody colors .
a comparison of the spectrum with a blackbody spectrum suggested a temperature of @xmath510 k. [ [ nltt-18555 ] ] nltt 18555 + + + + + + + + + + is also known as lp 207 - 50 .
figure [ fig2 ] shows this object to be a dc white dwarf .
sdss photometry is available for this star ( sdss [email protected] ) and comparing the @xmath331/@xmath332 and @xmath333/@xmath332 to blackbody colors provides an estimate of the temperature of @xmath511 k. a comparison of the spectrum to a blackbody spectrum suggests a temperature of @xmath512 k. given that the infrared photometry has large uncertainty in the @xmath513 band , we will adopt the temperature obtained using sdss colors .
[ [ nltt-18642 ] ] nltt 18642 + + + + + + + + + + is also known as lp 207 - 55 and was also included in the kiso survey for ultraviolet objects ( kuv07531@xmath64148 ) .
@xcite classified this object as nhb ( i.e. , stars with a hi dominated spectrum that could either be normal main - sequence or a horizontal - branch star ) . figure [ fig1 ] shows that this object is a cool da white dwarf .
we fitted the balmer lines with model spectra to obtain an effective temperature of @xmath514 k and surface gravity of @xmath515 .
the available sdss colors ( @xmath331/@xmath332 and @xmath333/@xmath332 ) implies a temperature of @xmath516 k and confirms the temperature obtained from the spectral fit .
[ [ nltt-19019 ] ] nltt 19019 + + + + + + + + + + is also known as lp 311 - 2 and gd 262 , and was observed as part of the sdss as sdss [email protected] .
it was listed as a white dwarf suspect @xcite .
we fitted the balmer lines with model spectra to obtain @xmath517 and @xmath518 ( see figure [ fig_sdss_balmer ] ) .
[ [ nltt-19138 ] ] nltt 19138 + + + + + + + + + + was observed by @xcite who classified this object as a dc white dwarf .
we have reobserved this star and confirm this classification .
the broad absorption features observed by @xcite were not detected in the present observations .
a comparison of the spectrum to a blackbody resulted in a temperature estimate of 7500 k. using the available sdss photometry for this object , we compared the observed colors ( @xmath337 and @xmath338 ) to calculated colors for a black - body to obtained a temperature estimate of @xmath519 k , which is the temperature we adopt .
[ [ nltt-20165 ] ] nltt 20165 + + + + + + + + + + was observed as part of the sdss as sdss [email protected] .
it was also observed spectroscopically as follow up of the kiso survey as kuv 08422@xmath63813 by @xcite , who observed a hydrogen dominated spectrum and suggested that it may be normal main - sequence or horizontal branch star .
this star was not listed in the lwdc .
the sdss spectrum shows this to be a da white dwarf .
we fitted the balmer lines with model spectra to obtain an effective temperature of @xmath520 k and a surface gravity of @xmath521 ( see figure [ fig_sdss_balmer ] ) .
[ [ nltt-21241 ] ] nltt 21241 + + + + + + + + + + is also known as lp 210 - 58 and was observed as part of the sdss as sdss [email protected] .
the sdss spectrum shows this object is a da white dwarf .
we fitted balmer line profiles with model spectra to obtain an effective temperature of @xmath522k and a surface gravity of @xmath523 ( see figure [ fig_sdss_balmer ] ) .
[ [ nltt-27901 ] ] nltt 27901 + + + + + + + + + + is also known as gd 311 , lp 94 - 285 and lb 2052 and was observed as part of the sdss as sdss j113534.61 + 572451.7 .
it was included in the list of white dwarf suspects by @xcite .
the sdss spectrum shows this star to be a dc white dwarf . comparing this spectrum to a blackbody
, we obtained a temperature estimate of 9000 k. using @xmath329 minimization we compared the sdss colors ( @xmath337 and @xmath338 ) to blackbody colors to obtain a temperature estimate of @xmath524 k. [ [ nltt-28772 ] ] nltt 28772 + +
+ + + + + + + + is also known as lp 129 - 587 , g197 - 35 and it was observed spectroscopically by @xcite who classified it a da white dwarf .
@xcite using multichannel photometry , obtained an estimate of the temperature and surface gravity , @xmath525 k and @xmath526 .
this white dwarf is the common proper motion companion to lp 129 - 586 and was observed as part of the sdss as sdss [email protected] . we fitted the balmer lines of the sdss spectrum with model spectra to obtain an effective temperature of @xmath527 k and a surface gravity of @xmath528 ( see figure [ fig_sdss_balmer ] ) .
[ [ nltt-29233 ] ] nltt 29233 + +
+ + + + + + + + is also known as g122 - 61 .
@xcite obtained @xmath503 photometry ( @xmath529 , @xmath530 , @xmath531 ) and classified it as a white dwarf star .
a low - dispersion spectrum in the blue was obtained as part of the case low - dispersion northern sky survey as cbs 451 @xcite .
it was classified as a featureless spectrum , figure [ fig1 ] shows that nltt 29233 is a cool da white dwarf .
we fitted balmer line profiles with model spectra to obtain an effective temperature of @xmath532 k and surface gravity of @xmath533 .
[ [ nltt-30738 ] ] nltt 30738 + + + + + + + + + + is also known as lp 435 - 109 and was not listed in the luyten white dwarf catalog .
our spectroscopic observations revealed this object to be a dc white dwarf .
we also observed weak h@xmath281 emission , suggesting that this may be close binary system .
comparison of optical - infrared photometry ( @xmath534 ) to blackbody colors we obtain an effective temperature of 6400 k. however , if a companion is present , then this temperature estimate is most likely to be inaccurate .
[ [ nltt-31347 ] ] nltt 31347 + +
+ + + + + + + + is the high - proper motion object lhs 5222 ( lp 217 - 47 ) which in figure [ fig1 ] is shown to be a dq white dwarf .
nltt 31347 was also observed as part of the sdss as sdss j123752.12 + 415625.8 .
nltt 31347 appears very similar to the white dwarf gsc2u j131147.2@xmath6292348 @xcite .
two temperature estimates are available for gsc2u j131147.2@xmath6292348 , @xmath535 k @xcite and @xmath536 k @xcite .
we compared the sdss colors ( @xmath331/@xmath332 ) to the theoretical colors for dq white dwarfs at @xmath339 presented in @xcite to obtain a temperature estimate of @xmath537 k with @xmath538 .
therefore , nltt 31347 is one of the coolest dq white dwarfs .
[ [ nltt-32695 ] ] nltt 32695 + + + + + + + + + + is also known as lp 616 - 70 and was observed as sdss j130247.9 - 005002.7 in the sdss .
the sdss spectrum shows this object to a da white dwarf . we fitted the balmer line profiles with model spectra to obtain @xmath539 k and @xmath540 .
@xcite obtained @xmath541 k and @xmath542 ( see figure [ fig_sdss_balmer ] ) and @xcite observed this star for variability , they did not find this object to vary .
[ [ nltt-35880 ] ] nltt 35880 + + + + + + + + + + is also known as lp97 - 430 .
figure [ fig1 ] shows that this is dc white dwarf .
this star was observed in the sdss and _ ugriz _ photometry is available for this object . comparing the photometric colors ( @xmath337 ) to blackbody colors
, we obtain an effective temperature of @xmath543 k. [ [ nltt-38499 ] ] nltt 38499 + + + + + + + + + + is also known as lp 801 - 14 and was observed as part of the edinburgh - cape survey ( ec 14473@xmath101901 ) and was classified as a sdb @xcite .
figure [ fig1 ] shows this object to be cool da white dwarf rather than a sdb .
we fitted the balmer line profiles with model spectra to obtain @xmath544 k and @xmath545 . @xmath503
photometry from the ec survey places this object on the white dwarf sequence with a temperature of @xmath424 k. [ [ nltt-40489 ] ] nltt 40489 + + + + + + + + + + is also known as lp 503 - 7 and was listed as gd 184 in the list of suspected white dwarfs by @xcite .
figure [ fig1 ] shows this to be a cool da white dwarf , fitting balmer line profiles with model spectra resulted in an effective temperature of @xmath421 k and surface gravity of @xmath546 . due to the white dwarf being very cool , only the lines of h@xmath281 and h@xmath288 can be seen and hence fitted .
[ [ nltt-40881 ] ] nltt 40881 + + + + + + + + + + is also known as lp 273 - 64 , ltt 14655 and gd 187 and is listed as a white dwarf suspect by @xcite but was not listed in the lwdc .
figure [ fig1 ] shows that this is a da white dwarf . we fitted the balmer line profiles with model spectra to obtain @xmath547 k and @xmath548 .
[ [ nltt-40992 ] ] nltt 40992 + + + + + + + + + + is also known as lp 384 - 38 .
figure [ fig1 ] shows that this is a cool da white dwarf . fitting the balmer line profiles with synthetic spectra we obtain @xmath549 k and @xmath550 .
again , due the weak balmer lines in cool white dwarfs , only h@xmath281 and h@xmath288 were used in the spectral fit .
[ [ nltt-41800 ] ] nltt 41800 + + + + + + + + + + is also known as g 202 - 26 and was not included in the lwdc .
it was observed as sdss j160112.70 + 531700.0 in the sdss .
this object was included in the sdss 1st data release of white dwarfs @xcite , however they quoted an estimate of the temperature ( @xmath551 k ) with a note that the fit was unsatisfactory . using the sdss spectrum
, we fitted the balmer line profiles with model spectra to obtain @xmath552 k and @xmath553 ( see figure [ fig_sdss_balmer ] ) .
[ [ nltt-42050 ] ] nltt 42050 + +
+ + + + + + + + is also known as lp 274 - 53 .
figure [ fig1 ] shows that this is a cool da white dwarf . due to the limited signal to noise ratio , we only fitted the h@xmath281 and @xmath554 line profiles to obtain an effective temperature of @xmath555 and a surface gravity of @xmath556 . to check these values
we used the available sdss photometry to obtain a temperature estimate of @xmath557 k. we then compared synthetic spectra at different gravities ( @xmath278 , 8.0 and 9.0 ) , a visual inspection shows that a low surface gravity is required to match the h@xmath281 profile .
we also performed a balmer line fit assuming @xmath558 for which we obtained an effective temperature of 5400 k. [ [ nltt-42153 ] ] nltt 42153 + + + + + + + + + + is also known as lp 861 - 31 and was recovered in the yale / san juan southern proper - motion ( spm ) program and is listed in the spm catalog 2.0 ( @xmath559 , @xmath560 ) .
figure [ fig1 ] shows this object to be a da white dwarf .
we fitted the balmer line profiles with model spectra to obtain @xmath561 k and @xmath562 .
[ [ nltt-43806 ] ] nltt 43806 + + + + + + + + + + is also known as lp 276 - 33 .
figure [ fig1 ] shows this object to be a cool daz white dwarf .
table [ tbl_line_nltt43806 ] presents the line identification of the most prominent heavy element lines . we first fitted the balmer line profiles ( h@xmath281 and h@xmath288 ) with pure hydrogen model spectra to obtain @xmath563 k and @xmath564 .
we then calculated a series of spectra with varying abundance of sodium , calcium and magnesium , for @xmath565 k and @xmath566 .
we compared these synthetic spectra to the observed spectrum to find that @xmath567 , @xmath568 , and @xmath569 .
the large uncertainty in the surface gravity does not have significant effect on the abundance measurements .
the uncertainty in the abundance measurements is 0.1 dex for all three heavy elements .
the new results supersede our previous results published in @xcite .
figure [ fig_nltt44986 ] shows the observed spectrum compared to a synthetic spectrum with the computed abundances . at 5700k , nltt 43806 is possibly one of the coolest daz ever observed .
the abundance ratio @xmath570 is comparable to other measurements @xcite , but the the measured calcium abundance in nltt 43806 is above the trend portrayed by @xcite .
lcc 3835.30 & mgi + 3933.66 & caii + 3968.47 & caii + 4226.73 & cai + 5178.14 & mgi + 5892.94 &
nai + [ [ nltt-43827 ] ] nltt 43827 + + + + + + + + + + is also known as lp 387 - 21 .
figure [ fig1 ] shows this object to be a da white dwarf .
we fitted the balmer line profiles with model spectra to obtain @xmath571 k and @xmath572 .
the temperature and surface gravity of nltt 43827 places it outside the zz ceti instability strip @xcite .
however , due to its very high mass it remains interesting in defining the strip at high - gravity .
[ [ nltt-43985 ] ] nltt 43985 + + + + + + + + + + is also known as lp 331 - 27 .
figure [ fig1 ] shows that this is a cool da white dwarf . fitting the balmer line profiles with synthetic spectra we obtain @xmath573 k and @xmath574 .
sdss photometry ( @xmath337 and @xmath333/@xmath332 ) results in a temperature of @xmath575 k which is in agreement with the spectroscopic determination .
[ [ nltt-44000 ] ] nltt 44000 + + + + + + + + + + is also known as g 203 - 39 and it is not included in the lwdc .
figure [ fig1 ] shows this to be a cool da white dwarf .
fitting balmer line profiles ( h@xmath281 to h@xmath319 ) to model spectra we obtained @xmath576 k and @xmath577 .
[ [ nltt-44149 ] ] nltt 44149 + + + + + + + + + + is also known as lp 276 - 48 .
figure [ fig2 ] shows that this is a dc white dwarf .
fitting sdss colors ( @xmath337 and @xmath338 ) to blackbody colors results in an estimate of the temperature of @xmath578 k. a blackbody fit to the spectrum also results in a temperature of 4500 k. at this temperature the balmer lines are very weak and a higher signal - to - noise ratio is required to conclude whether this is hydrogen- or helium - rich white dwarf .
a comparison of the sdss colors to synthetic h - rich colors also results in a temperature of 4500 k , however the lower limit of our synthetic colors grid is 4500 k. [ [ nltt-44447 ] ] nltt 44447 + + + + + + + + + + is also known as lp 226 - 48 .
figure [ fig1 ] shows this object to be a cool magnetic da white dwarf . due to the zeeman
split balmer lines , profile fitting would result in an inaccurate temperature and gravity . therefore assuming @xmath558 we compared @xmath579 to hydrogen - rich synthetic colors to obtain a temperature estimate of 7000 k. the zeeman splitting corresponds to a magnetic field of 1.3 mg .
[ [ nltt-44986 ] ] nltt 44986 + + + + + + + + + + is also known as eg 545 or gd 362 and is listed as a white dwarf suspect by @xcite , however it was not included in the lwdc .
@xcite observed this star and classified it as a da with a possible dk companion based on multichannel spectrophotometry .
this object was also observed as part of the palomar - green survey ( pg 1729 + 371 : * ? ? ?
* ) and classified `` sd '' .
figure [ fig1 ] shows this object to be a cool daz white dwarf with no evidence for a companion .
table [ tbl_line_nltt44986 ] presents the line identifications of the most prominent heavy element lines . we first fitted the balmer line profiles with pure hydrogen model spectra to obtain @xmath580 k and @xmath581 .
we then calculated spectra with varying abundances of calcium , magnesium and iron at @xmath582 k and @xmath583 .
we compared these spectra to the observed spectrum to find that @xmath584 , @xmath585 , and @xmath586 .
figure [ fig_nltt44986 ] shows the observed spectrum compared to a synthetic spectrum with the computed abundances .
this star was also recently observed by @xcite and determined @xmath587 k and @xmath588 with abundances @xmath584 , @xmath589 , and @xmath590 . which is in agreement with our determinations except for the fe abundance , for which we obtained a value a factor of 10 lower .
infrared observations of nltt 44986 showed that it has significant excess from the k - band to the n - band and the presence of a debris disk around the ultra - massive white dwarf @xcite .
lcc 3835.30 & mgi + 3933.66 & caii + 3968.47 & caii + 4045.81 & fei + 4226.73 & cai + 4271.76 & fei + 4383.54 & fei + 5178.14 & mgi + [ [ nltt-45344 ] ] nltt 45344 + + + + + + + + + + is also known lp 227 - 31 .
figure [ fig1 ] shows this object to be a cool da white dwarf . fitting balmer lines ( h@xmath281 to h@xmath319 ) to model spectra we obtained an effective temperature of @xmath591 and surface gravity of @xmath592 .
[ [ nltt-45723 ] ] nltt 45723 + + + + + + + + + + is also known lp 228 - 12 .
figure [ fig2 ] shows this object to be a dc white dwarf . using the optical / infrared colors ( @xmath534 ) we estimate a temperature of @xmath459 k. fitting a black - body to the spectrum we obtained a temperature of @xmath593 k. combining the two results we will adopt a temperature of @xmath594 k. [ [ nltt-50029 ] ] nltt 50029 + + + + + + + + + + is also known as lp 872 - 48 .
@xcite obtained a spectrum ( bps cs 22880 - 0126 ) as part of the spectroscopic follow - up of a selected sample of objects from the hk - survey .
based on ubv photometry ( @xmath595 , @xmath596 , @xmath597 ) they estimated a temperature of 12600 k. they classified the object as a dc white dwarf .
figure [ fig2 ] confirms this classification . using optical / infrared colors we estimate a temperature of 9600 k. fitting a black body to the spectrum we estimate a temperature of 9500 k. [ [ nltt-53177 ] ] nltt 53177 + + + + + + + + + + was observed by @xcite and classified as a da .
we refitted the star to obtain @xmath598 k and @xmath599 .
however , @xcite found nltt 53177 ( also he 2209 - 1444 ) to be a short - period ( @xmath600 day ) double degenerate system comprised of two white dwarfs with similar similar temperatures ( @xmath601 , 7140 k ) and equal mass ( @xmath602 ) .
the spectrum reported in @xcite was of low - resolution and hence would not reveal the line core splitting ; high resolution spectroscopy ( i.e. , echelle ) such as that carried out by @xcite is required .
therefore there may be more double degenerates in our sample .
also note that our temperature determination appears close to the average temperature of the two components @xcite .
[ [ nltt-53447 ] ] nltt 53447 + + + + + + + + + + is also known as lp 287 - 39 .
figure [ fig2 ] shows this to be a dc white dwarf , fitting a black - body to this spectrum we obtain a temperature estimate of @xmath603 k. using the @xmath534 diagram we obtain a temperature of 5000 k , therefore we will adopt a temperature of 4750 k. [ [ nltt-53996 ] ] nltt 53996 + + + + + + + + + + is also known as lp 400 - 6 .
figure [ fig2 ] shows this to be a cool da white dwarf . due to the limited signal - to - noise ratio we obtained an effective temperature by comparing the @xmath534 colors to synthetic colors for h - rich white dwarfs to obtain @xmath604 k. we then compared synthetic spectra at different gravities ( @xmath278 , 8.0 and 9.0 ) , a visual inspection shows that a low surface gravity ( @xmath605 ) is required to match the h@xmath281 profile .
we also repeated this procedure for @xmath274 and 5500 k and we conclude that the best fit temperature is 5000 k and therefore we will adopt @xmath604 k and @xmath278 . [ [ nltt-54047 ] ] nltt 54047 + +
+ + + + + + + + is the high - proper motion star lhs 3821 ( lp 520 - 28 ) .
recently , @xcite obtained spectroscopy of this object and classified it a dc white dwarf . the spectrum shown in figure [ fig1 ] shows this to be a cool da white dwarf . fitting the balmer lines to model spectra we obtained @xmath606 k and @xmath607 .
[ [ nltt-56122 ] ] nltt 56122 + + + + + + + + + + is also known as lp 522 - 17 .
figure [ fig1 ] shows this to be a cool da white dwarf .
we fitted the balmer lines to model spectra to obtain @xmath608 k and @xmath609 .
comparison of sdss colors ( @xmath337 and @xmath338 ) to synthetic h - rich colors results in a temperature of @xmath610 k. our sample of da white dwarfs comprises nine stars with @xmath611 k , and of those nine , six have surface gravities below @xmath558 .
there is probably a selection effect in finding more low - gravity cool da white dwarfs , because h@xmath281 is predicted to be weaker ( i.e. , shallower lines ) at higher gravities , and we are likely to classify these stars as dc white dwarfs .
therefore , higher signal - to - noise spectroscopy is required for such objects . also
our analyses of cool objects may suffer from uncertainties because our models currently exclude the effect of molecular hydrogen @xcite .
the sample of rnltt white dwarfs are old white dwarfs with ages of the order of @xmath612 to @xmath613 years .
figure [ fig_age_mass ] shows the effective temperature versus surface gravity of the da white dwarfs compared to the mass - radius relations of @xcite , as well as the masses and ages for all the white dwarfs .
no independent gravity measurements are available for non - da white dwarfs and for the ultra - cool da white dwarf nltt 8581 , and we simply assumed @xmath614 .
three objects ( nltt 14307 , nltt 43827 , and nltt 44986 ) have masses in excess of @xmath615 and belong to a sequence of high - mass white dwarfs also identified in euv surveys @xcite and in common - proper motion surveys @xcite .
note also that three cool white dwarfs ( @xmath616k ) also seem to be characterized by a low surface gravity ( @xmath617 ) ; the low surface gravity was diagnosed by their relatively strong h@xmath281 line strengths for their assigned effective temperatures . for the remaining 37 da white dwarfs , the mass average and dispersion are @xmath618 and @xmath619 which is slightly higher than the mass average ( @xmath620 ) of the h - rich subsample of @xcite .
the distribution among various spectral types follows established trends .
first , 45 objects ( 74% ) from the present selection of 61 white dwarfs are hydrogen - rich , three are daz white dwarfs ( nltt 3915 , nltt 43806 and nltt 44986 ) . for at least three of the new da
white dwarfs the classification is primarily secured by the presence of h@xmath281 , which underlines the importance of obtaining red spectroscopy for a proper classification of cool white dwarfs .
we suspect that a few dc white dwarfs in the @xcite catalog of spectroscopically identified white dwarfs may turn out to be da white dwarfs .
some 14 objects are classified as dc white dwarfs ( 23% ) and the two remaining objects consist of a new dq white dwarf ( nltt 31347 ) and the dz white dwarf nltt 40607 @xcite . for comparison , we determined the composition of 200 well - studied rnltt white dwarfs using the @xcite catalog : some 134 objects ( 67% ) received a primary label `` da '' , 14 received the label `` db '' ( 7% ) , 15 are dq ( 8% ) , 11 are `` dz '' ( 5% ) , and in last resort 26 received the label `` dc '' ( 13% ) . combining the two data sets ( @xmath621 ) , 69% of rnltt white dwarfs are spectral type da , and 31% non - da . the da : non - da number ratio varies as a function of temperatures . by dividing the survey in the temperature bins @xmath622 , @xmath623 , @xmath624 , and @xmath625k , we observe a varying ratio 1.5:1 , 3.6:1 , 3.5:1 , and 5:1 , respectively
. the ratio increases sharply with effective temperatures .
the result of this proper motion survey are to be contrasted with the result of the parallax survey of @xcite who estimated da : non - da ratios varying from 1.2:1 in the @xmath622 bin , to 2:1 in a sparsely populated @xmath624 bin .
the number of dc white dwarfs are markedly lower in the present proper motion survey .
the da : non - da ratio in @xmath622 bin is only a lower limit , since for many white dwarfs in this bin it is very difficult to distinguish between h - or he - rich atmospheres without infrared photometry . for many of the objects
accurate infrared photometry is required to ascertain the classification .
note that in this proper motion survey we selected the brighter candidates for spectroscopic observations , and therefore there may exist a slight bias against dc white dwarfs which tend to be fainter than da white dwarfs .
over 150 white dwarf candidates from our original selection @xcite remain to be spectroscopically confirmed .
eight white dwarfs ( nltt 529 , nltt 3915 , nltt 8435 , nltt 14307 , nltt 18555 , nltt 43806 , nltt 53447 and nltt 56805 ) in the present compilation are found at distances closer than 20 pc from the sun and therefore contribute to the local census .
other nearby white dwarfs are potentially among these remaining @xmath626 white dwarf candidates .
we evaluate that @xmath627 of these are likely to lie within 20pc of the sun .
we assumed that all of these candidates are white dwarfs for which we calculated their absolute magnitude using @xmath628 photometry , and then determined their distance .
most white dwarfs in the neighborhood of the sun belong to the thin disk population , but a significant number of thick disk white dwarfs are expected .
we have calculated the velocity components @xmath629,@xmath195 and @xmath630 using @xcite and assuming @xmath631 .
we have calculated absolute magnitudes using the temperatures and surface gravities as described in the previous sections and listed in table [ tbl2 ] .
we obtained the distances using the apparent magnitudes and the absolute magnitudes listed in table [ tbl1 ] and table [ tbl3 ] .
figure [ fig_kinematic ] compares measured @xmath629 and @xmath195 to the @xmath632 velocity ellipse of the thin disk population , the @xmath632 ellipse of the thick disk population and the @xmath330 ellipse of the halo population @xcite .
the following velocities are all expressed in km s@xmath1 .
all three distributions , @xmath633 , were fitted with gaussian functions , where @xmath634 , @xmath635 , @xmath636 and @xmath637 , @xmath638 , @xmath639 .
our velocity dispersions are in good agreement with those of a thin disk population ( @xmath634 , @xmath640 , @xmath641 * ? ? ?
* ) , however our @xmath638 is only slightly more negative then their quoted value of @xmath642 for normal stars in the thin disk .
hence , the measured velocity distributions imply that the majority of the objects in this study belong to the thin disk population .
the local sample of white dwarfs should consist of a significant number of white dwarfs that belong to the thick disk population .
several estimates of the fraction of thick disk white dwarfs in the solar neighborhood have been made , from @xmath643 @xcite up to 25% @xcite .
the main reason for the difference between the two values is that the first assumes a single power - law for the initial mass function , which produces more low - mass main - sequence stars and hence a small white dwarf mass - fraction in the disk ( @xmath644 ) .
@xcite used a two - component power law , which produces fewer low - mass main sequence stars and hence the white dwarf mass fraction for the thick disk will be higher .
therefore , the thick disk could contribute as much as 25% of the local white dwarfs .
it is difficult to disentangle thin versus thick disk populations solely based on the kinematics .
a study of the distribution of white dwarfs in the @xmath645 plane led @xcite to conclude that 5% of their sample of @xmath646 white dwarf candidates extracted from the rnltt catalog of @xcite lay in a high - velocity tail and , therefore , may belong to the thick disk population .
@xcite used the @xmath195-velocity distribution which is most sensitive to the presence of thick disk white dwarfs which are characterized by a velocity dispersion twice the dispersion of thin disk white dwarfs ( @xmath647 versus @xmath648 ) .
therefore , the fraction of thick disk white dwarfs estimated by @xcite is a lower limit , considering that several objects assimilated with the thin disk population may in fact be part of the bulk of the thick disk population only betrayed by its high - velocity tail . similarly in this study of 61 objects , four objects in figure [ fig_kinematic ] are clearly lying outside the @xmath632 velocity ellipse of the thin - disk population .
if these four objects were assumed to belong to the thick - disk , then the percentage of thick - disk white dwarfs in our sample would be @xmath649 which is consistent with @xcite .
note that this is a lower limit .
we have spectroscopically identified 49 new white dwarfs from the rnltt survey of @xcite in addition to the 12 white dwarfs from @xcite and @xcite .
their proper motions range from 0.136 to 0.611@xmath0yr@xmath1 which is to be contrasted with the proper motions of suspected halo white dwarf candidates wd 0346@xmath6246 ( @xmath650yr@xmath1 ) and pm j13420@xmath103415 ( @xmath651yr@xmath1 ) .
some 45 objects from this sample are hydrogen rich white dwarfs for which we have determined effective temperatures and surface gravities fitting the balmer lines to synthetic spectra . for 14 of the objects we provided a dc classification and estimated their temperature by comparing their spectra and photometric colors to a black body . of the 45 hydrogen - rich white dwarfs , three display a high abundance of heavy elements ( nltt 3915 , nltt 43806 and nltt 44986 ) .
one of these ( nltt 44986 ) is also an ultramassive white dwarf with a mass of @xmath652 , and another one is a magnetic white dwarf ( nltt 44447 ) with an estimate surface magnetic field of 1.3 mg . in our sample
there is also one cool dq white dwarf ( nltt 31347 ) with an estimated temperature of 5200 k and the cool dz nltt 40607 presented by @xcite .
eight of the white dwarfs have distances that place them within 20 pc of the sun , hence contributing toward the local census .
parallax measurements of the new nearby white dwarfs should be obtained in order to provide independent estimate of their radius and mass .
we determined u , v and w space velocities for all 61 objects .
the means and dispersions of @xmath629 and @xmath195 suggest that these objects belong to the thin disk .
however , due to the overlap of the thick disk population distribution with the thin disk population distribution in the @xmath645 plane , some of the stars in our sample may belong to the thick disk . to help distinguish between thick and thin disk white dwarf populations , the z - component of the angular momentum @xmath653 , the eccentricity of the orbit @xmath654 and the galactic orbit
should be calculated @xcite .
therefore , to obtain full information about the kinematics of the white dwarfs , accurate radial velocity measurements are necessary .
this research received financial support from the college of science of the florida institute of technology .
a. kawka is supported by ga r 205/05/p186 .
we thank pierre chayer for sharing apo observing time .
based on observations obtained with the apache point observatory 3.5-meter telescope , which is owned and operated by the astrophysical research consortium ( arc ) .
we thank t. oswalt , m. wood , j. kubat and and the referee for interesting comments .
funding for the creation and distribution of the sdss archive has been provided by the alfred p. sloan foundation , the participating institutions , nasa , the nsf , the u.s .
department of energy , the japanese monbukagakusho , and the max planck society .
the sdss web site is http://www.sdss.org/. the sdss is managed by the arc for the participating institutions : the university of chicago , fermilab , the institute for advanced study , the japan participation group , the johns hopkins university , the korean scientist group , los alamos national laboratory , the max - planck - institute for astronomy , the max - planck - institute for astrophysics , new mexico state university , university of pittsburgh , princeton university , the united states naval observatory , and the university of washington .
becklin , e.e . ,
farihi , j. , jura , m. , song , i. , weinberger , a.j . , &
zuckerman , b. 2005 , , 632 , l119 beers , t.c . ,
preston , g.w . , shectman , s.a . ,
doinidis s.p . , & griffin , k.e .
1992 , , 103 , 276 bergeron , p. 2001
, , 558 , 369 bergeron , p. , leggett , s. k. , & ruiz , m. t. 2001 , , 133 , 413 bergeron , p. , wesemael , f. , & fontaine , g. 1991 , , 367 , 253 bergeron , p. , wesemael , f. , & fontaine , g. 1992 , , 387 , 288 benvenuto , o.g . , & althaus , l.g . 1999 ,
mnras , 303 , 30 binney , j. , & merrifield , m. 1998 , galactic astronomy ( princeton : princeton univ . press ) carollo , d. , bucciarelli , b. , hodgkin , s.t . ,
lattanzi , m.g . ,
mclean , b.j . ,
morbidelli , r. , smart , r.l . ,
spagna , a. , & terranegra , l. 2005 , , submitted ( astro - ph/0510638 ) carollo , d. , hodgkin , s.t . ,
spagna , a. , smart , r.l .
, lattanzi , m.g.,mclean , b.j . , & pinfield , d.j .
2002 , , 393 , l45 carollo , d. , koester , d. , spagna , a. , lattanzi , m.g . , &
hodgkin , s.t .
2003 , , 400 , l13 chiba , m. , & beers , t.c .
2000 , , 119 , 2843 cutri , r.m . ,
2003 , explanatory supplement to the 2mass all sky data release ( pasadena : caltech ) dufour , p. , bergeron , p. , & fontaine , g. 2005 , , 627 , 404 eggen , o.j .
1968 , , 16 , 97 fontaine , g. , villeneuve , b. , & wilson , j. 1981 , , 243 , 550 gizis , j.e . , & reid , i.n .
1997 , , 109 , 849 green , r.f . ,
schmidt , m. , & liebert , j. 1986 , , 61 , 305 greenstein , j.l .
1980 , , 242 , 738 gianninas , a. , bergeron , p. , & fontaine , g. 2005 , , 631 , 1100 gianninas , a. , dufour , p. , & bergeron , p. 2004
, , 617 , l57 giclas , h.l .
, burnham , r. , & thomas , n.g .
1965 , lowell observatory bulletin , 6 , 155 giclas , h.l .
, burnham , r. , & thomas , n.g .
1967 , lowell observatory bulletin , 7 , 49 gray , d.f .
1992 , the observation and analysis of stellar photospheres , 2nd ed .
( cambridge : cambridge univ . press ) hambly , n.c . , et al .
2001 , , 326 , 1279 hambly , n.c . ,
henry , t.j . ,
subasavage , j.p . ,
brown , m.a .
, & jao , w .- c . 2004 , , 128 , 437 hambly , n.c . ,
smartt , s.j . , & hodgkin , s.t .
1997 , , 489 , l157 hansen , b.m.s .
1998 , , 520 , 680 hansen , b.m.s . , & liebert , j. 2003 , , 41 , 465 hintzen , p. 1986
, , 92 , 431 holberg , j.b . , oswalt , t.d . , & sion , e.m .
2002 , , 571 , 512 hubeny , i. , hummer , d.g . , & lanz , t. 1994 , , 282 , 151 hummer , d.g . , & mihalas , d. 1988 , apj , 331 , 794 johnson , d.r.h .
, & soderblom , d.r .
1987 , , 93 , 864 karl , c.a . , napiwotzki , r. , nelemans , g. , christlieb , n. , koester , d. , heber , u. , & reimers , d. 2003 , , 410 , 663 kawka , a. , vennes , s. , & thorstensen , j.r .
2004 , , 127 , 1711 kawka , a. , & vennes , s. 2005 , in 14th european workshop on white dwarfs , asp conf .
ser . vol .
334 , eds .
d. koester , & s. moehler , 101 kawka , a. , vennes , s. , wickramasinghe , d.t . ,
schmidt , g.d . , & koch , r. 2003 , in white dwarfs , nato science series ii : mathematics , physics and chemistry ( kluwer ) , vol .
105 , eds .
de martino , d. , silvotti , r. , solheim , j .- e .
, & kalytis , r. , 179 kilic , m. , munn , j.a . , harris , h.c . ,
liebert , j. , von hippel , t. , williams , k.a . , metcalfe , t.s .
, winget , d.e . , & levine , s.e .
2006 , , 131 , 582 kilic , m. , von hippel , t. , leggett , s.k . , & winget , d.e .
2005 , , 632 , l115 kilkenny , d. , odonoghue , d. , koen , c. , stobie , r.s .
, & chen , a. 1997 , , 867 kleinman , s.j . , et al .
2004 , , 607 , 426 lemke , m. 1997 , a&as , 122 , 285 lepine , s. , & shara , m.m .
2005 , , 129 , 1483 lepine , s. , shara , m.m . , & rich , r.m .
2002 , , 124 , 1190 lepine , s. , shara , m.m . , & rich , r.m .
2003 , , 126 , 921 lepine , s. , rich , r.m . , & shara , m.m .
2005 , , 633 , l121 liebert , j. & strittmatter , p.a .
1977 , , 217 , l59 luyten , w.j .
1970 , white dwarfs ( minneapolis : univ .
minnesota press ) luyten , w.j .
1977 , white dwarfs ii ( minneapolis : univ .
minnesota press ) mccook , g. p. , & sion , e. m. 1999 , , 121 , 1 mihalas , d. 1978 , stellar atmospheres , 2nd ed .
( san francisco : freeman ) mukadam , a.s . , et al . 2004 , , 607 , 982 pauli , e .- m . , napiwotzki , r. , heber , u. , altmann , m. , & odenkirchen , m. 2005 , , in press ( astro - ph/0510494 ) pesch , p. , stephenson , c.b .
, & macconnell , d.j .
1995 , , 98 , 41 pravdo , s.h . ,
et al . 1999 , , 117 , 1616 reid , i.n .
2003 , , 126 , 3007 reid , i.n .
2005 , , 43 , 247 reid , i.n . , & gizis , j.e .
2005 , , 117 , 676 saumon , d. , & jacobson , s.b . 1999 , , 511 , l107 salim , s. , & gould , a. 2002 , , 575 , l83 salim , s. & gould , a. 2003 , , 582 , 1011 schning , t. 1994 , , 282 , 994 schrder , k .-
p . , pauli , e .- m . , & napiwotzki , r. 2004 , , 354 , 727 silvestri , n. m. , oswalt , t. d. , wood , m. a. , smith , j. a. , reid , i. n. , & sion , e. m.2001 , , 121 , 503 subasavage , j.p . ,
henry , t.j . , hambly , n.c . ,
brown , m.a .
, jao , w .- c . , & finch , c.t .
2005 , , 130 , 1658 teegarden , b.j .
, et al . 2003 , , 589 , l51 vennes , s. , thejll , p. , genova - galvan , r. , & dupuis , j. 1997 , , 480 , 714 vennes , s. , & kawka , a. 2003 , , 586 , l95 wegner , g. , & mcmahan , r.k .
1988 , , 96 , 1933 wegner , g. , & swanson , s.r .
1990 , , 99 , 330 weidemann , v. , & koester , d. 1984 , , 132 , 195 , yong , d. , & lambert , d.l .
2003 , , 115 , 22 zuckerman , b. , koester , d. , reid , i. n. , hnsch , m. 2003 , , 596 , 477 | the new luyten two - tenths catalog contains a large number of high - proper motion white dwarf candidates that remain to be spectroscopically confirmed .
we present new spectroscopic observations as well as sdss archival spectra of 49 white dwarf candidates which have been selected from the revised nltt catalog of @xcite . out of these , 34 are cool da white dwarfs with temperatures ranging from approximately 5000 k up to 11690 k , and 11 are dc white dwarfs with temperatures ranging from 4300 k ( nltt 18555 ) up to 11000 k. three of the da white dwarfs also display abundances of heavy elements ( nltt 3915 , nltt 44986 and nltt 43806 ) and one is a cool magnetic white dwarf ( nltt 44447 ) with an estimated magnetic field strength of 1.3 mg . we also present a new cool dq white dwarf ( nltt 31347 ) with an estimated temperature of 6250 k. we supplement our sample with sdss _ ugriz _ photometry for a fraction of the newly identified white dwarfs .
a kinematical study of this sample of white dwarfs , characterized by proper motions ranging from 0.136 to 0.611@xmath0yr@xmath1 suggest that they belong to the thin disk population . |
one of the central issues in nuclear physics is to understand the nuclear structure from the nucleon - nucleon ( @xmath4 ) interaction .
the @xmath4 interaction has a strong tensor force at long and intermediate distances and a strong repulsion at short distance @xcite .
it is important to investigate the nuclear structure considering the above characteristics of the @xmath4 interaction .
the origin of the tensor force is the one - pion exchange interaction , which brings the high - momentum components of nucleon motion in nuclei .
it is necessary to treat the high - momentum components induced by the tensor force in the nuclear wave function .
the tensor force also produces the characteristic @xmath5-wave state of a nucleon pair in nuclei , which comes from the strong @xmath6-@xmath5 coupling of the tensor force .
this @xmath5-wave state is spatially compact as compared with the @xmath6-wave state due to the high - momentum component of the tensor correlation @xcite .
the high - momentum component in nuclei coming from the tensor correlation has been investigated experimentally with the ( @xmath7,@xmath8 ) reaction @xcite .
so far , we have described the tensor correlation with high - momentum components on the shell model basis , which we name `` tensor - optimized shell model '' ( tosm ) @xcite . in tosm , we fully optimize the two - particle two - hole ( 2p2h ) states in the wave function . there is no truncation for the particle states in tosm .
in particular , the spatial shrinkage of the particle states is essential to achieve convergence of the contributions of tensor force .
this property is related to the inclusion of the spatially compact @xmath5-wave state with high momentum in the wave function .
the clustering of nucleons is one of the important aspects in the nuclear structure , such as the two-@xmath9 state in @xmath10be and the hoyle state in @xmath11c as the triple-@xmath9 state @xcite .
those clustering states can coexist with shell model - like states in a nucleus such as @xmath11c , the ground state of which is considered to be the shell model - like state . theoretically , it is generally difficult to describe the clustering states in the shell model type approach , while the shell model - like states are fairly described @xcite .
it is also known that the @xmath9 cluster itself contains the large contribution of the tensor force @xcite .
the relation between the @xmath4 interaction and the coexistence of the clustering states and the shell model - like states is unclear .
it is important to understand the nuclear clustering phenomena from the viewpoint of the @xmath4 interaction and the tensor force .
one of the theoretical approaches to describe the nuclear clustering is the antisymmetrized molecular dynamics ( amd ) @xcite .
the amd wave function consists of the gaussian wave packet for each nucleon , which is suitable to express the formation of cluster with spatial localization of some of nucleons in a nucleus .
so far , amd has shown the successful results in the description of various clustering states in finite nuclei from light mass to medium mass region @xcite . however
, this model can not treat the tensor force and/or short - range repulsion , and it is necessary for the amd analysis to rely on the effective interaction of mild central force and @xmath12 force without the tensor force . for
the clustering description of nuclei based on the @xmath4 interaction , the unitary correlation operator method ( ucom ) has been developed to treat the short - range and tensor correlations @xcite . using the fermionic molecular dynamics ( fmd ) with ucom
, they have discussed the clustering phenomena @xcite . in ucom
, the unitary - transformed hamiltonian is truncated up to the two - body operator , while the exact transformation produces many - body operators .
this truncation seems reasonable for short - range repulsion because of the short - range character , but tensor force has a long - range character and many - body operators should be important for the tensor correlation to work correctly .
the many - body operators are also important for the consistent treatment of the variational principle starting from the @xmath4 interaction .
in our study of tosm , only the short - range part of ucom is adopted to describe the short - range correlation in the shell - model type basis states , while the tensor correlation is explicitly treated using the full 2p2h excitation in the wave function .
the method of tosm+ucom nicely works to describe the shell model - like states with the correct order of the energy level in the @xmath7-shell nuclei , while the @xmath9 clustering states such as those in @xmath10be and @xmath11c are difficult to describe quantitatively @xcite . toward the nuclear clustering description from the @xmath4 interaction , we have proposed a new variational theory @xcite .
we employ the antisymmetrized molecular dynamics ( amd ) @xcite as the basis state .
we introduce two - kinds of correlation functions of the tensor - operator type for the tensor force and the central - operator type for the short - range repulsion .
this physical concept is similar to ucom @xcite .
the correlation functions are multiplied to the amd wave function as the correlated basis states and superposed with the amd wave function .
we name this framework `` tensor - optimized antisymmetrized molecular dynamics '' ( toamd ) @xcite . in toamd ,
the products of the hamiltonian and correlation functions become the series of the many - body operators , which are exactly treated using the cluster expansion .
we take all the necessary many - body operators without any truncation , which enable us to determine the correlation functions variationally .
the formulation of toamd is common for all nuclei with various mass numbers .
the scheme of toamd is extendable by taking the series of the multiple product of correlation functions as the power expansion .
this is done systematically and successively in toamd and necessary formula are published @xcite . in this paper , we take up to the double products of correlation functions of tensor and short - range types , and investigate the convergence of the solutions with respect to the multiples of correlation functions and discuss the role of each term . to demonstrate the new successive variational method , we take the @xmath0-shell nuclei , @xmath1h and @xmath2he , using the av8@xmath3 @xmath4 interaction .
we explain the basic formulation of toamd , while all the details are given in ref .
we start from the amd wave function , which is expressed by using the slater determinant of the gaussian wave packets of nucleons with mass number @xmath13 . the amd wave function @xmath14 is explicitly given as : @xmath15 the single - nucleon wave function @xmath16 consists of a gaussian wave packet with a range parameter @xmath17 and a centroid position @xmath18 , the spin part @xmath19 and isospin part @xmath20 . in this study of @xmath0-shell nuclei ,
@xmath19 is fixed as up or down component and @xmath20 is proton or neutron component .
the range @xmath17 is common for all nucleons and this condition factorizes the center - of - mass wave function from @xmath14 .
the range @xmath17 also contributes to the spatial size of @xmath14 . in toamd
we include two - kinds of correlations induced by the tensor force and short - range repulsion , which are difficult to treat in the amd wave function @xmath14 .
following the concept given in ref .
@xcite , we introduce the pair - type correlation functions @xmath21 for tensor force and @xmath22 for short - range repulsion and multiply them to the amd wave function .
this choice of the toamd wave function is motivated by the success of tosm @xcite .
we superpose these components with the original amd wave function . here
we define the basic toamd wave function as : @xmath23 with relative coordinate @xmath24 , @xmath25 and @xmath26 . here
@xmath27 and @xmath0 represent the isospin and spin channel of a pair , respectively .
the correlation functions @xmath21 and @xmath22 affect only the relative motion of nucleon pairs in @xmath14 , and do not excite the center - of - mass motion .
the center - of - mass motion is completely removed in toamd .
the function @xmath21 induces the relative @xmath5-wave transition via the tensor operator @xmath28 : @xmath29 the functions @xmath21 and @xmath22 are scalar operator and do not change the total angular - momentum state of @xmath14 . in general , two functions @xmath21 and @xmath22 are not commutable .
physically , the functions @xmath21 and @xmath22 can excite two nucleons in the amd state to the high - momentum region corresponding to the 2p2h excitation in the shell model .
this formulation of toamd is independent of the mass number @xmath13 and commonly used for all nuclei .
we state here the essential difference of toamd from the green s function monte - carlo ( gfmc ) method @xcite . in the gfmc method ,
the standard concept of correlation function is used , where it is expressed by a product : f_s^gfmc&=&_i < j^a(1+_t=0 ^ 1_s=0 ^ 1 f^t , s_s(r_ij)o^t_ijo^s_ij ) .
[ eq : product ] for the short - range correlation as an example .
this makes the calculation of matrix elements complicated , since full @xmath13-body operators should be calculated with the correlations of every nucleon pair . on the other hand , the basic toamd wave function given in eq .
( [ eq : toamd ] ) uses the lowest order term of the standard correlation function .
this truncation of the correlation operators makes the calculation easier and more systematical . in toamd
, we can increase the necessary terms of explicit calculations with the multiple number of correlation functions .
for the short - range correlation , the toamd wave function can be extended as : _
toamd^short = ( 1+f_s+f_s^2 + ) _ amd .
[ eq : expansion ] we can examine the convergence of the solutions with the power of the correlation functions step by step .
it is noted that the term @xmath30 in eq .
( [ eq : expansion ] ) has the component of @xmath31 which does not appear in eq .
( [ eq : product ] ) . as for the tensor correlation @xmath21 , the same discussion holds . in the present study of toamd
, we include up to the double products of the correlation functions consisting of @xmath22 and @xmath21 .
the effort of calculating the @xmath32 term in eq.([eq : toamd ] ) is the same as calculating further the @xmath33 , and @xmath34 terms .
this consideration brought us to an idea to use the toamd as a successive variational method with respect to the multiple of correlation functions .
writing all the possible double products of @xmath21 and @xmath22 , we come up with the next order of the toamd wave function as : @xmath35 it is noted that the correlation functions in each term in eq.([eq : toamd2 ] ) are determined independently .
this means that each correlation function can be different in toamd .
all the matrix elements are summed up for the final results .
hereafter , we call @xmath21 and @xmath22 single correlation functions and their double products are double correlation functions .
the total energy in toamd is given as : @xmath36 we calculate the matrix elements of the correlated hamiltonian @xmath37 and the correlated norm @xmath38 with the amd wave function , where @xmath37 and @xmath38 include the products of correlation functions , such as @xmath39 and @xmath40 .
these correlated operators become the series of many - body operators according to the particle index of each operator . in the case of the two - body interaction @xmath41
, @xmath42 is expanded from two - body to six - body operators with various combinations of particle index .
similarly , @xmath43 gives ten - body operators at maximum .
we classify these many - body operators fully in terms of the cluster expansion method , the detailed procedure of which is given in ref .
we take the matrix elements of all the resulting many - body operators with the amd wave function without any truncation .
this treatment is important to retain the variational principle .
the procedure is performed systematically for all the correlated operators with multiple correlation functions in toamd . in general , many - body operators produce larger number of terms of the cluster expansion for larger mass systems and the calculation of their matrix elements becomes much more demanding .
the toamd wave function has three - kinds of variational functions , two - kinds of correlation functions @xmath21 , @xmath22 and the amd wave function @xmath14 .
we determine them using the ritz variational principle with respect to the toamd energy @xmath44 .
the radial forms of @xmath21 and @xmath22 are optimized in each spin - isospin channel to minimize the total energy @xmath45 in eq .
( [ eq : e_toamd ] ) .
we use the gaussian expansion method to express the relative motion of the pair functions @xmath46 in eq .
( [ eq : fd ] ) and @xmath47 in eq .
( [ eq : fs ] ) , respectively , which are given as : @xmath48 here , @xmath49 , @xmath50 , @xmath51 and @xmath52 are variational parameters for gaussian basis functions .
we employ the common value for the gaussian basis number @xmath53 , which is taken seven at most until the solutions are converged .
for the ranges @xmath49 , @xmath50 , we search for the optimized values from short to long ranges to express the spatial correlation adequately , which gain the total energy .
the coefficients @xmath51 and @xmath52 are linear parameters in the toamd wave function and are determined variationally by diagonalization of the hamiltonian matrix elements . in the double correlation functions in eq .
( [ eq : toamd2 ] ) , the products of two gaussian functions in eqs .
( [ eq : cr_d ] ) and ( [ eq : cr_s ] ) are treated as the basis functions and the products of @xmath51 and @xmath52 are variational parameters for these basis functions .
the centroid positions of the gaussian wave packets , @xmath54 ( @xmath55 ) in eq .
( [ eq : gauss ] ) for @xmath14 are determined variationally by using the cooling method @xcite . in the calculation of matrix elements of many - body operators ,
we express also the @xmath4 interaction as a sum of gaussian functions .
technically , we adopt the fourier transformation of the gaussian form of the correlation functions and the @xmath4 interaction into the momentum space @xcite .
this transformation decomposes the many - body operators expressed with the various relative coordinates in the exponent into the separable form with respect to the single particle coordinates . in the momentum space ,
the matrix elements of the many - body operators in @xmath37 and @xmath38 result in the products of single - particle matrix elements with the amd basis functions , which is easily calculated even for various combinations of particle index in the operators . in this study
, we use the realistic @xmath4 interaction , av8@xmath3 @xcite consisting of central , tensor and @xmath12 terms .
to demonstrate the toamd as a successive variational method in many - body systems with strong interaction , we take the @xmath0-shell nuclei , @xmath1h and @xmath2he , where a large number of theoretical results in various few body methods are available @xcite .
we discuss first the toamd results for @xmath1h and @xmath2he with the single correlation function as @xmath56 . in the variation of the amd wave function
, it is found that the centroid positions of nucleons @xmath18 are zero for all nucleons in both nuclei .
this result indicates that even with the realistic @xmath4 interaction , the @xmath0-wave configurations are favored in the amd wave function , which is equivalent to the shell - model states of @xmath57 for @xmath1h and @xmath58 for @xmath2he , respectively .
the correlation functions @xmath21 and @xmath22 are optimized for each nuclei .
the energies are @xmath59 mev for @xmath1h and @xmath60 mev for @xmath2he , which are underbound in comparison with the gfmc results ; @xmath61 mev for @xmath1h and @xmath62 mev for @xmath2he . here , the range parameters in @xmath14 are @xmath63 @xmath64 for @xmath1h and @xmath65 @xmath64 for @xmath2he , which are a little bit different from the values of toamd optimized with double correlation functions to be discussed later . it is noted that both of the tensor and short - range correlations are necessary to make nuclei bound like in the case of the deuteron @xcite .
h with av8@xmath3 as function of the range parameter @xmath17 in toamd with single correlation function ( solid line ) .
dotted line represents the results with up to the two - body operators in the correlated hamiltonian @xmath37 .
dashed line represents the gfmc result.,width=321 ] in the variational calculation , the many - body operators with up to three - body for @xmath1h and up to four - body for @xmath2he are to be included in the correlated hamiltonian @xmath37 and norm @xmath38 .
if we perform a limited calculation with up to the two - body operators in the correlated hamiltonian @xmath37 , which is the similar treatment as ucom @xcite , the energies of two nuclei continue to decrease with respect to large @xmath17 and we do not obtain the physical energy minimum . in this limited calculation in which higher - body operators than two - body are omitted , the variational principle is not satisfied . in fig .
[ fig : ene_3h_nu ] , we confirm this fact by showing the energy surface of @xmath1h as function of the range parameter @xmath17 in toamd with single correlation function .
the full treatment of many - body operators in @xmath37 provides the energy minimum properly as shown by solid line . on the other hand ,
when we omit the three - body operators of @xmath37 , there is no energy minimum in @xmath1h as shown by dotted line .
this result indicates the inevitable role of the many - body operators in the correlated hamiltonian to obtain the energy minimum properly .
the similar result is obtained in the brueckner - bethe - goldstone approach for nuclear matter @xcite , in which the three - body correlation terms induced by the @xmath66-matrix are shown to be necessary to obtain the proper saturation point for density of nuclear matter . as seen in fig .
[ fig : ene_3h_nu ] , our toamd approach is very powerful for studying the important roles of many - body operators in finite nuclei because it is the variational calculation .
the detailed analysis of many - body operators will be published in the forthcoming paper .
next we include the double correlation functions in toamd given in eq.([eq : toamd2 ] ) .
we keep the nucleon positions @xmath67 for all nucleons . in fig.[fig : ene_3h_ff ] , we show the energy of @xmath1h successively obtained by adding the single and double correlation functions one by one , where @xmath68 @xmath64 determined variationally in the full calculation . in each calculation , this value of @xmath17 is common but the correlation functions are optimized independently . in the figure , the labels d and s indicate @xmath21 and @xmath22 , respectively , and `` + s '' is the result obtained with the wave function of @xmath69 .
`` + ss '' is the result by adding @xmath33 component as @xmath70 in total .
the components of @xmath71 and @xmath32 are found to give almost identical effect on the solutions , and the combined energy is provided for `` + sd+ds '' . ``
+ dd '' is the full calculation with the double correlation functions in toamd .
the final energy up to @xmath72 ( + dd ) is @xmath73 mev , which almost reproduces the gfmc results within 80 kev .
h with av8@xmath3 by adding each term of toamd successively . dashed line represents the gfmc result.,width=321 ] h by adding each term of toamd successively . dashed lines represent the gfmc results for each component . for the kinetic energy , a half value is shown with the symbol `` k/2 ''
. the symbols c , t and ls indicate the central , tensor and @xmath12 forces , respectively.,width=321 ] figure [ fig : ene_3h_ff ] shows clearly the effect of each term on the energy . at the level of pure amd the nucleus
is not bound , while the addition of @xmath22 decreases the energy and further addition of @xmath21 finally brings the nucleus bound .
adding further the double correlation functions step by step , the energy curve shows converging behavior and eventually converges to the gfmc value . in this sense
we are able to say that the toamd is a successive variational method in many - body systems with strong interaction .
he with av8@xmath3 by adding each term of toamd successively . dashed line represents the gfmc result.,width=321 ] he by adding each term of toamd successively .
dashed lines represent the gfmc results for each component . for the kinetic energy , half value
is shown with the symbol `` k/2 '' .
the symbols c , t , ls indicate the central , tensor and @xmath12 forces , respectively.,width=321 ] we provide the contributions of each term in the hamiltonian of @xmath1h : the kinetic energy ( k ) , central ( c ) , tensor ( t ) and @xmath12 ( ls ) forces in fig.[fig : ham_3h_ff ] .
we see how each term in the hamiltonian changes as the correlation functions are added successively .
the contributions of the tensor and @xmath12 forces are found to have the values after adding the tensor correlation ( + d ) .
again each term converges to the corresponding results of the gfmc calculation .
the matter radius is obtained as 1.746 fm . from these results
, we can learn the accuracy and power of toamd and the efficiency of correlation functions @xmath21 and @xmath22 to treat the @xmath4 interaction .
good reproduction of hamiltonian components of @xmath1h including the @xmath12 force indicates that the tensor and short - range correlations are essential and sufficient in the description of the @xmath1h wave function .
this is similar to the case of the deuteron , where the @xmath5-wave component brought by the tensor force provides the @xmath12 energy . in fig .
[ fig : ham_3h_ff ] , it is found that the amd+s calculation does not provide the enhancement of the kinetic energy from the amd result in spite of the inclusion of the short - range correlation .
this is because @xmath22 can represent not only the short - range character but also the intermediate- and long - range ones .
we discuss now the case of @xmath2he .
figure [ fig : ene_4he_ff ] shows the total energy obtained by adding successively the correlation functions in toamd wave function in eq .
( [ eq : toamd2 ] ) , where the range parameter of the amd wave function is @xmath74 @xmath64 .
we see quite a similar behavior of the total energy as that of @xmath1h .
we see a good convergence pattern with the correlation functions . with full components up to @xmath34 , the energy is obtained as @xmath75 mev in toamd .
the energy difference between toamd and gfmc is about 1.2 mev .
figure [ fig : ham_4he_ff ] shows the hamiltonian components . each component as kinetic energy ( k ) , central ( c ) , tensor ( t ) and @xmath12 ( ls ) forces show gradual convergence with successive addition of the multiple of the correlation functions .
these energies deviate slightly from the gfmc results .
the matter radius is obtained as 1.497 fm , which is very close to the gfmc value of 1.490 fm .
we compare the present results of @xmath2he with tosm using short - range ucom @xcite , which gives the energy of @xmath76 mev .
regarding the short - range ucom as @xmath77 and the 2p2h excitations in tosm as the role of @xmath22 or @xmath21 in toamd , tosm corresponds to toamd without the @xmath34 term , which gives the energy of @xmath2he as @xmath78 mev .
this value is close to the tosm result . in this sense
, toamd includes the tensor correlation more than that of tosm , owing to the @xmath34 term . in toamd , the correlation functions @xmath22 and @xmath21 are optimized independently in each term of eq . ( [ eq : toamd2 ] ) .
it is interesting to see this effect on the energy , and we perform the following calculation .
first , @xmath22 and @xmath21 are determined in the single correlation function of toamd as @xmath79 .
second , keeping the functional form of @xmath22 and @xmath21 with gaussian expansion , we perform the calculation including double correlation functions , where the weights of the double correlation functions are variational parameters .
this calculation provides the energies of @xmath1h and @xmath2he as @xmath80 mev and @xmath81 mev , respectively .
the energy loss from the full calculation is 1.44 mev for @xmath1h and 2.34 mev for @xmath2he .
these amounts indicate the importance of the independent optimization of the correlation functions in each term of toamd , contributing to the rapid energy convergence . from the numerical results of @xmath1h and @xmath2he shown in figs.[fig :
ene_3h_ff ] and [ fig : ene_4he_ff ] , it is found that the convergence of energies with respect to the correlation functions is rapid in toamd .
this fact indicates the validity of the present expansion of the wave function in power of the tensor and short - range correlations . considering the convergence of the solutions in the present analysis
, we can expect that we reach the precise energy further by increasing the multiples of correlation functions to the next order .
this extension of more correlation functions is handled systematically in toamd by taking all kinds of many - body operators emerging from the correlated hamiltonian and norm .
the next order is the triple products of the correlation functions consisting of @xmath21 and @xmath22 .
we can put a priority on the basis functions involving the tensor and short - correlations at the same time , which leads to the @xmath82 and @xmath83 products in toamd .
increasing the power of correlation functions , the analysis of the results at each power is physically meaningful to get a knowledge of the role of correlation functions .
this successive variational method in many - body systems with strong interaction is important when we calculate heavier nuclei , since we are able to see how we approach the convergence .
one of the advantages of toamd is the clustering description based on the amd basis states .
it is interesting to consider the system consisting of several clusters such as @xmath10be with two @xmath2he .
it is found that each @xmath2he nucleus needs the double products of the correlation functions to get the sufficient binding energy .
this fact naively suggests that the spatially separated two-@xmath2he state will need fourth power of the correlation functions totally in toamd .
it is an interesting problem how the cluster states are described in toamd with the increase of the power of correlation functions .
we have developed a new variational theory `` tensor - optimized antisymmetrized molecular dynamics '' ( toamd ) to describe the nuclear structure using the nucleon - nucleon ( @xmath4 ) interaction , in particular , toward the nuclear clustering description . in toamd , the tensor- and central - type correlation functions
are introduced considering the characteristics of the @xmath4 interaction .
these correlation functions are multiplied to the amd wave function to express the effects of tensor force and short - range repulsion explicitly and the correlated basis states are superposed with the amd wave function .
this scheme of toamd is independent of the mass number and extendable by increasing the power of the multiple products of the correlation functions successively . in the calculation of matrix elements , the products of the hamiltonian and the correlation functions produce the many - body operators in principle , which are exactly treated without any truncation using the cluster expansion
this is important to keep the variational principle starting from the @xmath4 interaction .
the genuine three - nucleon interaction such as fujita - miyazawa type can be tractable in toamd in the same manner as treating many - body operators . in this study , we take up to the double products of correlation functions in toamd and show the effect of each term step by step successively . using the av8@xmath3 @xmath4 interaction
, we show the efficiency of toamd in the description of @xmath0-shell nuclei .
the toamd results reproduce the @xmath1h energy and provide a good binding energy for @xmath2he in the scheme of the double correlation functions .
the radius is also reproduced for two nuclei .
these results indicate that the essential correlations induced by the @xmath4 interaction is sufficiently included in the present order of toamd for @xmath0-shell nuclei , which is physically meaningful and also useful when we treat the system of heavier mass nuclei .
in addition , the convergence of the solutions with respect to the multiples of correlation functions is rapid .
this indicates the validity of the present expansion approach of toamd .
we believe that it is important to obtain the precise energy for @xmath2he , since it is a building block of light nuclei . in order to reduce the energy difference from the gfmc value , we shall increase the multiples of correlation functions with the triple products such as @xmath82 as the next order .
based on the success for the @xmath0-shell nuclei , we shall apply toamd to the @xmath7-shell nuclei with the three - nucleon interaction .
this work was supported by jsps kakenhi grant numbers jp15k05091 , jp15k17662 , jp16k05351 .
# 1#2#3#4 # 1 * # 2 * , # 4 ( # 3 ) 00 s. c. pieper and r. b. wiringa , .
s. c. pieper , v. r. pandharipande , r. b. wiringa and j. carlson , .
k. ikeda , t. myo , k. kat and h. toki , _ clusters in nuclei _
, lecture notes in physics 818 vol.1 , ( springer , berlin , 2010 ) , pp.165 - 221 .
h. j. ong et al . , .
t. myo , h. toki and k. ikeda , . t. myo , a. umeya , h. toki and k. ikeda , . k. ikeda , h. horiuchi , and s. saito , .
h. horiuchi , k. ikeda , and k. kat , .
b. r. barrett , p. navrtil , j. p. vary , .
t. myo , a. umeya , k. horii , h. toki and k. ikeda , . h. kamada et al . , , and the references therein .
y. kanada - enyo , m. kimura and h. horiuchi , .
y. kanada - enyo , m. kimura , and a. ono , .
h. feldmeier , t. neff , r. roth and j. schnack , .
t. neff , and h. feldmeier , .
t. neff , .
t. myo , a. umeya , h. toki and k. ikeda , .
t. myo , h. toki , k. ikeda , h. horiuchi , t. suhara , .
a. sugie , p. e. hodgson and h. h. robertson , .
s. nagata , t. sasakawa , t. sawada and r. tamagaki , .
y. goto and h. horiuchi , .
k. fukukawa , m. baldo , g. f. burgio , l. lo monaco , and h .- j .
schulze , . | we study the tensor - optimized antisymmetrized molecular dynamics ( toamd ) as a successive variational method in many - body systems with strong interaction for nuclei . in toamd ,
the correlation functions for the tensor force and the short - range repulsion and their multiples are operated to the amd state as the variational wave function .
the total wave function is expressed as the sum of all the components and the variational space can be increased successively with the multiple correlation functions to achieve convergence .
all the necessary matrix elements of many - body operators , consisting of the multiple correlation functions and the hamiltonian , are expressed analytically using the gaussian integral formula . in this paper
we show the results of toamd with up to the double products of the correlation functions for the @xmath0-shell nuclei , @xmath1h and @xmath2he , using the nucleon - nucleon interaction av8@xmath3 .
it is found that the energies and hamiltonian components of two nuclei converge rapidly with respect to the multiple of correlation functions .
this result indicates the efficiency of toamd for the power series expansion in terms of the tensor and short - range correlation functions . |
this paper reports structural predictions for the decagonal quasicrystal @xmath0(alconi ) in the cobalt - rich ( ` basic co ' ) , of approximate composition al@xmath1co@xmath2ni@xmath3 and atomic density 0.068 atoms / @xmath4 , using simulations to minimize energy , using the same methods ( and codes ) as previous papers by mihalkovic , widom , and henley @xcite on the ` ni - rich ' ; this constitutes a first test of the transferability of that approach to other compositions that are described by other tiling geometries .
the aluminum - nickel - cobalt alloy @xmath0(alconi ) is the best studied and perhaps the highest quality of equilibrium decagonal quasicrystals .
decagonal @xmath0(alconi ) has about eight modifications ( which we call ` ' here ) , each existing in a tiny domain of the phase diagram.@xcite .
the compositions ` basic ni ' ( around @xmath5 ) , and ` basic co ' are the best - studied , having the simplest diffraction patterns .
( the other show superstructure peaks , indicating modulations . ) several crystal approximants related to ` basic co ' are known @xcite , one of which has a solved structure @xcite .
we find that ( in contrast to the ` basic ni ' ) , the framework of ` basic co ' at @xmath6 is a network of edge - sharing @xmath7 diameter decagons ( placed like the large atoms in a `` binary tiling '' quasicrystal ) . a strong ( but not always simple ) co / ni ordering is found .
our model reproduces most , but not all , of the atomic positions in w(alconi ) .
( a complementary brief account of our results is in ref .
@xcite , and a more complete account is in preparation @xcite . )
our main assumption is that the @xmath0(alconi ) structure is well approximated by a stacking of equally spaced two - dimensional tilings built from the penrose rhombi under periodic boundary conditions .
the chief experimentally determined inputs were the values @xmath84.08 for the stacking period and @xmath92.45 for quasilattice constant ( edge length of small rhombi ) .
this framework of decorated tilings takes advantage of the fact that all the known atomic structures of decagonals are penrose - tiling - like .
further , even in random tilings penrose rhombi admit `` inflation '' constructions that can relate a tiling to another one using rhombi with edges enlarged by @xmath10 ( the golden ratio ) , which gives a convenient way to build a chain of connections from the atomic level to large tiles so as to describe @xmath0(alnico ) on large scales .
[ fig : tiling ] includes two ways of subdividing a unit cell into a tiling of rhombi ( with edges @xmath11 and @xmath12 , respectively ) .
we also assume atomic pair potentials between al , co , and ni as derived ab - initio using generalised pseudopotential theory @xcite ( gpt ) , but modified using results from _ ab initio _ calculations to approximate the sums of the omitted many - body interactions when transition metals ( tm ) are nearest neighbours .
[ fig . 1 of ref .
@xcite plots these potentials . ]
the potentials are cut off by a smooth truncation past 7.0 . we know @xcite that the strongest interaction is the al - tm first potential well .
the number of al - tm neighbours is maximised and , as a corollary , that of tm - tm nearest neighbours is minimised ( but tm - tm second neighbours , at @xmath13 separation , are common ) .
however , this scarcely constrains the structure since the weak al - al potential allows enormous freedom in placing al . in practice
, one finds in al - tm quasicrystals that a rather rigid tm - tm network forms , with separations near to the strong second minimum of the tm - tm pair potential ; the al atoms fill the tm interstices in a more variable fashion .
subtle details of the density and composition , as well as the small differences between co and ni in the pair potentials , decide which structure optimises the energy .
we understand only a few of these factors in detail for the ` basic co ' phase , so in this paper we will only describe the atomic structures we found and do not attempt a microscopic rationalization of them .
our program uses metropolis monte carlo ( mmc ) simulation to perform atom - atom and atom - vacancy swaps within a site list placed on the penrose tiles .
the first stage simulations include an additional degree of freedom in the form of ` tile flips . '
these ` tile flips ' obey the mmc algorithm probabilities and conserve tile number , atomic species number , and the collective outline of tiles rearranged .
our procedure is to anneal from a high temperature @xmath14 - 10 ev@xmath15 to a low temperature at which most degrees of freedom are frozen out , @xmath16 ev@xmath15 , using a step size @xmath17 - 1 ev@xmath15 .
( here @xmath18 . )
at each temperature step , approximately 2000 atomic swaps per site are attempted .
23@xmath194 unit cell .
the left half is subdivided into @xmath11 scale rhombi ( in two ways in part of the cell [ dotted lines ] ) .
the right half of the cell is subdivided into @xmath12 scale rhombi .
insets at top left and top right display the candidate sites on the @xmath11 and @xmath12 scale tiles , respectively . filled large circles depict atoms in one layer and filled small circles are those in the other layer .
black denotes nickel , dark gray cobalt , and light gray aluminium .
unoccupied sites of the @xmath12 scale tiling are shown by very small empty circles ; those of the @xmath11 scale tiling are not shown .
, width=297 ] the strategy of our series of simulations loosely copies that reported previously @xcite on the ni - rich of @xmath0(alnico ) , and can be thought of as a sort of multiscale modelling .
it proceeds in the following stages : * begin with monte carlo simulations allowing atom swaps and ` tile - flips ' with a tiling of relatively small penrose rhombi ; note prominent features and formations that recur in all the low energy structures from a sufficient number of independent runs . *
take the observations from the small - scale tiling and promote them to rules : i.e. make the formations inferred in the first stage into fundamental objects of a larger scale tiling .
unused ( or underused ) degrees of freedom are removed . *
perform new monte carlo simulations on the larger - scale tiling to search for new low energy structures .
this larger- scale tiling is generally more efficient at finding lower energies because of the reduced degrees of freedom . *
verify that the degrees of freedom removed in stage ( 2 ) were indeed unnecessary by restoring some of them , judiciously .
* repeat for larger and larger scale tilings .
an eventual goal is to extract a ` tile hamiltonian ' @xcite which is the bridge to modelling long - range order and diffuse scattering .
that means ascribing the atom - atom energies to an effective interaction among neighbouring tiles , so that tiles are the only remaining degree of freedom .
it is also necessary to perform relaxations / molecular dynamics to escape the biases which may be introduced by the discrete site list used up through stage ( 4 ) .
the first stage of our tiling simulations uses small penrose rhombi that have @xmath20 edges .
the unit cell simulations had dimensions @xmath21 @xmath22@xmath23 , which we shall call the 32@xmath1923@xmath194 cell , shown in fig . [
fig : tiling ] .
the structure is taken to repeat after two layers of the penrose tiling . at 0.068 atoms / @xmath4 , there are about 200 atoms in the unit cell .
our low energy configuration generated at this level showed distinct rings of 5 tm and 5 al atoms surrounding an al atom and surrounded by an 10 al atom decagon with an edge length of @xmath11 .
these 21-atom motifs were evidently energy favourable .
linear regressions of the count of these 21-atom clusters versus energy revealed a noisy , but consistent correlation of @xmath24 ev per 21-atom cluster . however , the ideal structure did not simply maximsze the number of 21-atom clusters : though they may be placed adjacent to pack six clusters per cell , that was typically not observed in the low - energy configurations .
instead , further mmc annealing ( starting from a low - energy configuration ) found that a larger cluster emerged as the characteristic motif at this composition : a decagon with an edge length of @xmath25 ( hence a diameter of @xmath26 ) , which we will call ` 13 decagon ' ( 13d ) .
this object contains the 21-atom motif at its centre , ten tm atoms at its vertices , and al atoms irregularly interspersed along and within its edges .
we use the following nomenclature for the components of 13d , from its centre outwards .
the centre of the 13d is an al atom .
the 5 tm 5 al ring is known as the ` first ring . '
the tm atoms in the first ring are in the same layer as the central al atom .
the 10 al ring is known as the ` second ring . '
the 10 tm decagon and the al atoms along and within the edges are collectively known as the ` third ring . '
miscellaneous al sites occur between the second and third rings , which we designate collectively as the ` 2.5 ring . ' in the 13d , the first ring tm sites are mainly occupied by co and the third ring ( decagon vertex ) tm sites also tend to co occupation , but there is a very strong dependence on the local environment that distinctly favour ni in certain sites .
[ this became clear at the second stage of simulation , subsec .
[ sec : inflated ] . ]
the al positions of third - ring and 2.5 ring al atoms are closely correlated and the rules were not resolved from the 2.45 tiling simulation .
a second type of motif fills the interstices between the 13ds : it consists of a 10-atom ring and an al at the centre , much like ring 1 in the 13d cluster , except this kind of motif has mixed al - tm occupations for the 5 candidate tm sites . in general , about 3 sites out of the 5 candidates are filled with tm atoms , preferably ni .
we name these 11-atom clusters `` star clusters '' ; a few of these are contained in fig .
[ fig : tiling ] . in keeping with our ` multiscale ' methodology , we reassessed the degrees of freedom necessary for the next level of simulation .
the dominant geometric object is a decagon with edge @xmath28 , but we found it economical to represent its decoration by candidate sites using a thin 4 rhombus and the second fat rhombus shown in the right inset of fig .
[ fig : tiling ] ; similarly , the star cluster decoration is represented using the first fat rhombus .
we assumed that the minimum - energy structure has a maximum density of 13d clusters .
a monte carlo simulation was carried out ( admitting only tile flips ) on the ensemble of @xmath12 rhombus tilings using an artificial tile hamiltonian favouring the creation of nonoverlapping ` star decagons ' ( of edge @xmath12 ) containing a fivefold star of fat rhombi .
it was found that the ground state configuration was always a ` binary tiling ' @xcite of rhombi with edge length @xmath2910.4 with the ` large atom ' and ` small atom ' vertices replaced by 13ds and star clusters , respectively .
we * designate * a fixed site list on the @xmath12 rhombi , as shown in the right half of fig .
[ fig : tiling ] , designed to match the most frequently occupied sites observed in the low - energy states from out @xmath11 scale simulations .
this tiling , unlike the @xmath11 tiling , consists of only _
one _ layer of penrose tiles , each of which has site decorations in _ two _ layers spaced 2.04 in the @xmath30 direction .
note how the two versions of the fat penrose rhombus have different site lists .
we use the 32@xmath1923@xmath194 cell , as before so we may compare the energies for systems with exactly the same density and composition . at this second stage ,
our mmc runs had purely lattice gas moves on a fixed tiling ( no tile flips ) .
these were able to find lower energy configurations at a much faster rate than the first - stage ( @xmath11 scale ) simulations at the same composition and atomic density , due to the much decreased site list .
hence , the occupations of many types of sites were discerned with greater accuracy , such as the tm sites in the 13 decagon ( already described in subsec .
[ sec : decagon ] ) . furthermore the tendencies of more variable sites ( al in rings 2.5 and 3 , and tm in the star cluster ) , which will be described in a longer paper @xcite , started to come into focus .
finally , despite having _ fewer _ allowed configurations , the @xmath12 tiling simulation consistently found _ lower _ energies , which serves as our post hoc justification for assuming 13d clusters in the binary tiling geometry and the reduced site list .
tiles , and it was verified this had a negligible effect . ] in the prior case of the ` basic ni ' , the decoration model @xcite was implemented as a deterministic tiling .
a marked contrast in the present ` basic co ' case is that we can not impose a simple rule that fixes the chemical occupation of each site . on certain tm sites ,
it is difficult to resolve the occupation ( co / ni ) ; certain al sites are also variably occupied ( al / vacant ) depending on the environment . presumably , with a sufficiently thorough understanding of our model
, we could formulate a deterministic decoration rule on the binary tiling .
it is possible , however , that the energy differences among some competing structures are too small to be visible in a reasonable simulation , or to influence the real properties at any accessible temperature .
the physical accessible candidate sites in the @xmath12 decagons ( with our decoration ) in fact have a @xmath31 point group symmetry , which is not broken by the binary tiling geometry .
the fivefold symmetric 13d cluster breaks this symmetry , by the layer on which the central al sits and the al / tm alternation on the first ring .
thus , a major obstacle to writing a deterministic decoration is that the relative orientations of 13ds must be specified , which depends on subtle interaction energies between them .
experimentally , @xmath0(alconi ) with our co - rich composition was observed to order in a fivefold symmetry , wherein all the decagons are oriented the same way @xcite . as a test
, we compared the lowest - energy configurations resulting from mmc simulations on identical @xmath12 tilings in which neighbouring 13ds were forced either to have always identical or always opposite orientations .
[ this is controlled by the orientation of the decagon of @xmath12 rhombi in the tiling that gets decorated .
] we evaluated the minimum - energy configurations from @xmath32 independent runs of each type .
we used the 32@xmath1923@xmath194 unit cell , using a fixed composition al@xmath1co@xmath2ni@xmath3 , but repeating the tests for a a series of atom number densities @xmath33 .
the energy difference between these two orientations was density dependent . at @xmath34
atoms/ @xmath4 the two orientational patterns are practically degenerate ( though with visibly distinct atom configurations . ) from @xmath350.068 @xmath36 to 0.074 @xmath36 a range which includes the most realistic compositions the pattern with uniform orientations has the lower energy , by up to @xmath37ev / atom , but this difference disappears again at @xmath380.074 @xmath36 .
we find this energy difference arises from different ways of filling the possible tm sites in the respective orientation schemes . in the uniform - orientation scheme ,
as the tm density is raised ( as part of the total density ) , tm atoms fill the ring 2.5 in the 13 decagon before they exceed @xmath39 filling in the star cluster tm sites ; whereas in the alternating - orientation case , the ring 2.5 is filled to a lesser degree with tm while the star cluster tm sites become overfilled .
[ in ref .
@xcite , we gave an explanation how these effects could favour _ alternating _ orientations .
] but the results just presented are valid only for the ( unphysical ! ) case of atoms confined to hop on fixed ideal sites .
when displacements are allowed , so that many atoms `` pucker '' away from the flat layers ( see sec . [ sec : rmr ] ) , then the uniform arrangement is preferred more robustly .
up to this point , we reported simulations using fixed atom site positions .
we can make our configurations more realistic by subjecting them to relaxation and molecular dynamics ( md ) .
the removal of the tiling and discrete site list leads to effects such as out - of - layer `` puckering '' and so - called ` period doubling , ' whereby certain atoms relax into positions that violate the @xmath30 period of a single unit cell , but instead are periodic with respect to a unit cell with @xmath40 8.16 . exactly such distortions are familiar in the structures of decagonal phases and approximants .
our standard cycle for such ` off - site ' studies is a relaxation to a local energy minimum ( i.e. to 0 k ) , followed by an md cycle beginning at @xmath41600 k and ending at @xmath4150k in increments of 50k .
the md results are then relaxed again to 0k .
this protocol will be denoted relaxation - md - relaxation ( rmr ) . upon relaxation to @xmath42 , we find that the tm atoms are quite immobile and move only slightly .
the al atoms , however are subject to displacements as large as @xmath43 . after molecular dynamics and re - relaxation to @xmath42 , we find a few al atoms be further displaced , but to similar sites so that no systematic difference is apparent in the overall pattern .
major diffraction - based structure models were available for ` basic ni ' @xmath0(alnico ) , to which the simulation predictions could be compared , but no such model exists for the ` basic co ' @xmath0(alconi ) .
however , the structure of the ` w phase ' crystal approximant of ` basic co ' has been determined by sugiyama _
et al _ @xcite , and we may apply what we have learned so far in particular the period doubling and puckering to predict its structure .
[ this study has inspired a more detailed modelling of w(alconi ) , based on ab - initio energies rather than pair potentials @xcite .
] using exactly the same tiles and interlayer spacing as in our other simulations , our unit cell is @xmath44@xmath45@xmath46 , which differs by less than 1% from the experimentally determined lattice constants .
the approximate atom content implied by sugiyama s structure solution and co : ni ratio is al@xmath47co@xmath48ni@xmath49 , with uncertainties arising from the partial and mixed occupations . adopting this composition as our ideal
, we have al@xmath50co@xmath51ni@xmath52 at a density 0.0714 @xmath36 .
we applied our mock hamiltonian from sec .
[ sec : inflated ] to generate a tiling consistent with the @xmath53 binary rhombus tiling .
the optimum configuration is unique ( modulo symmetries ) and has four 13ds , in an arrangement which turns out to be the same as observed in in w(alconi ) .
cluster orientation comparisons were like performed like those of subsec .
[ sec : orientation ] , but using the relaxed energies ( without md ) .
we found that , between the configurations with alternating and uniform cluster orientations , the latter had a lower relaxed energy , in accordance with the experimental w - alconi structure .
the atomic locations generated on this unit cell using the 4 rhombus decoration site list is approximately consistent with the experimentally determined sites .
our mmc simulations are able to capture the gross features of w - alconi with excellent accuracy . since this approximant is not far from the ` basic co ' decagonal composition
, we are not surprised to find 13ds and star clusters as the major motifs , arranged in a manner consistent with our binary and @xmath12 tilings , consistent with the observed w - phase . the differences in the exact atomic locations lie mainly within the highly context dependent 2.5th and 3rd rings .
we find that the increase in density from our original simulations results in a more dense occupation of these rings .
atomic configurations found after rmr are in even better agreement . fig .
[ fig : wa ] compares one layer from experimental data to our rmr results ; the match in the other layers are equally good .
-axis dimension.,width=297 ] however , a significant number of al atoms in our rmr picture are in disagreement with the experimental refinement . because certain puckered atoms are not present in our simulations .
the top right of fig .
[ fig : wa ] contains a pentagon of displaced al atoms centred upon the missing al atom .
these extremely puckered al atoms are perhaps located in energy minima too far away for our md program to traverse .
the existence of such atoms is one of the greatest flaws for beginning with a discrete site simulation .
we find that in the approximate composition ( al@xmath1ni@xmath3co@xmath2 , the structure of @xmath0(alnico ) is dominated by the formation of the 13d clusters and the star clusters which complement them .
we find that the gross features are robust under variations of the composition and densities of @xmath13% ; these will strongly affect a few details of the sites and we have not established the correct answers for those atoms .
our method of tile decorations , and successive reductions of the degrees of freedom , affords an enormous speedup compared to brute - force molecular dynamics , which would get stuck in glassy configurations , at the price of a few plausible assumptions . by applying our results to the approximant w - alconi
, we demonstrated both the validity and the pitfalls of our approach .
the pitfall lies in that fact that our explorations are conducted using a fixed site list in two atomic layers , whereas the real structure has four layers and some atoms are strongly puckered out of them .
even though most puckered atoms may be represented as relaxations from ideal sites , the optimal puckered structure might be derived from a fixed - site structure of comparatively high energy that our simulations would pass over .
reliable answers can not be obtained by pure numerical exploration , but seem to demand some physical understanding of the puckering ( and other displacements ) , as we have begun to do in ref .
the context dependences evident in the ` basic co ' show that beyond a point , the errors in defining the pair potentials will surely exceed the energy differences due to some swaps of species , which is a second pitfall that may be alleviated only by checking against experimental phase diagrams .
this work is supported by doe grant de - fg02 - 89er45405 ; computer facilities were provided by the cornell center for materials research under nsf grant dmr-0079992 .
mm was also supported by grant vega-2/5096/25 of the slovak academy of sciences .
we thank m. widom for discussions .
99 c. l. henley , m. mihalkovi , and m. widom , j. all .
342 ( 1 - 2 ) : 221 ( 2002 ) .
m. mihalkovi , c. l. henley , and m. widom , j. non - cyst .
sol . 334 : 177 ( 2004 )
. s. ritsch , c. beeli , h. u. nissen , t. gdecke , m. scheffer , and r. lck , phil . mag . lett . , 74 , ( 1996 ) 99 - 106 k. edagawa , h. tamaru , s. yamaguchi , k. suzuki , and s. takeuchi , phys .
b 50,12413 ( 1996 ) .
b. grushko , d. hollard - moritz , and k. bickmann , j. all .
236 , 243 ( 1996 ) a. p. tsai , a. fujiwara , a. inoue , and t. masumoto , phil . mag . lett .
74 , 233 ( 1996 ) .
s. ritsch , c. beeli , h. u. nissen , t. gdecke , m. scheffer , and r. lck , phil . mag . lett .
78 , 67 - 76 ( 1998 ) .
philos . mag .
78 , 67 ( 1998 ) n. gu , m. mihalkovi , and c. l. henley , preprint ( www.arxiv.org : cond - mat/0602095 ) j. a. moriarty , and m. widom , phys .
b 56 ( 1997 ) 7905 - 17 .
m. mihalkovi , w .- j .
zhu , c. l. henley , and m. oxborrow , phys .
b 53 , 9002 ( 1996 ) .
s. ritsch , c. beeli , and h .- u .
nissen , phil . mag . lett .
74 , 203 ( 1996 ) .
x. z. li , r. c. yu , k. h. kuo , and k. hiraga , phil . mag . lett .
73 , 255 ( 1996 ) .
m. mihalkovi and m. widom , phil . mag .
86 , 557 ( 2006 ) . | we use energy minimization principles to predict the structure of a decagonal quasicrystal - @xmath0(alconi ) - in the cobalt - rich phase .
monte carlo methods are then used to explore configurations while relaxation and molecular dynamics are used to obtain a more realistic structure once a low energy configuration has been found .
we find five - fold symmetric decagons 12.8 in diameter as the characteristic formation of this composition , along with smaller pseudo - five - fold symmetric clusters filling the spaces between the decagons .
we use our method to make comparisons with a recent experimental approximant structure model from sugiyama et al ( 2002 ) . # 1 # 1 # 1 * clh : # 1 * # 1 * nan : # 1 * # 1 * mm : # 1 * # 1 _ # 1 _ # 1 |
the objective of this paper is to set the foundation for a research program on quantum knots[multiblock footnote omitted ] . _ for simplicity of exposition , we will throughout this paper frequently use the term `` knot '' to mean either a knot or a link . _
in part 1 of this paper , we create a formal system @xmath0 consisting of * a graded set @xmath1 of symbol strings , called * knot mosaics * , and * a graded subgroup @xmath2 , called the * knot mosaic ambient group * , of the group of all permutations of the set of knot mosaics @xmath1 .
we conjecture that the formal system @xmath3 fully captures the entire structure of tame knot theory .
three examples of knot mosaics are given below : @xmath4{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } $ ] , @xmath4{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } $ ] , and @xmath4{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } $ ] each of these knot mosaics is a string made up of the following 11 symbols @xmath5 @xmath6{ccccccccccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}\end{array } , \ ] ] called * mosaic tiles*. an example of an element in the mosaic ambient group @xmath2 is the * mosaic reidemeister 1 move * illustrated below : @xmath7{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}\end{array } \overset{\left ( 0,1\right ) } { \longleftrightarrow}\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps}}\end{array}\ ] ] this mosaic reidemeister 1 move @xmath8 is a permutation which is the product of disjoint transpositions , as illustrated by observing that the reidemeister 1 move @xmath8 interchanges the the following two knot mosaics : @xmath9{c}$\left ( n\overset{\left ( 0,1\right ) } { \longleftrightarrow}n^{\prime } \right ) \left ( \begin{array } [ c]{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \right ) = \begin{array } [ c]{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } $ \\ \\ $ \left ( n\overset{\left ( 0,1\right ) } { \longleftrightarrow}n^{\prime } \right ) \left ( \begin{array } [ c]{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \right ) $ = $ \begin{array } [ c]{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } $ \\ \qquad\qquad\qquad\textbf{knot mosaics are interchanged by } $ n\overset{\left ( 0,1\right ) } { \longleftrightarrow}n^{\prime}$\end{tabular } \ \ \ \ , \ ] ]
while it leaves the following mosaic unchanged : @xmath9{c}$\left ( n\overset{\left ( 0,1\right ) } { \longleftrightarrow}n^{\prime } \right ) \left ( \begin{array } [ c]{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \right ) $ = $ \begin{array } [ c]{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } $ \\
\qquad\qquad\qquad\textbf{the knot mosaic is left fixed by } $ n\overset{\left ( 0,1\right ) } { \longleftrightarrow}n^{\prime}$\end{tabular } \ \ \ ] ] in part 2 , the formal system @xmath0 is used to define a * quantum knot system * @xmath10 , which is a nested sequence of quantum systems consisting of * a graded hilbert space @xmath11 , called the * quantum knot state space * , defined by an orthonormal basis labelled by and in one - to - one correspondence with the set of knot mosaics @xmath1 , and * an * * associated graded control group , also called the * ambient group * , and also denoted by @xmath2 .
the ambient group @xmath2 is a discrete subgroup of the group @xmath12 of all unitary transformations on @xmath11 .
a * quantum knot * is simply a state of the quantum knot system , i.e. , an element of the quantum knot state space @xmath11 . *
quantum knot type * is defined as the orbit of the quantum knot under the action of the ambient group @xmath2 .
once having defined what is meant by a quantum knot , we then proceed to find the hamiltonians associated with the generators of the ambient group @xmath2 , and to study the quantum dynamics induced by schroedinger s equation .
we move on to discuss other hamiltonians , such as for example those associated with overcrossings quantum tunnelling into undercrossings .
we also study a class of quantum observables which are quantum knot invariants . we should mention that , if one selects a fixed upper bound @xmath13 on knot complexity ( i.e. , a fixed upper bound on the edge length @xmath13 of the knot @xmath13-mosaics under consideration ) , then a quantum knot system @xmath14 is a blueprint for the construction of an actual physical quantum system .
quantum knots could possibly be used to simulate and to predict the behavior of quantum vortices that appear both in liquid helium ii and in the bose - einstein condensate they might also possibly be a mathematical model for gaining some insight into the charge quantification that is manifest in the fractional quantum hall effect . in the conclusion , we list a number of open questions and possible future research directions . a complete table of all knot @xmath15-mosaics is given in appendix a. finally , in appendix b , we briefly outline the theory of and the construction of * oriented knot mosaics * and * oriented quantum knots*. * the motivating intuition for the above mathematical construct * @xmath10 * is as follows : * a quantum knot system is intended to represent the `` quantum embodiment '' of a closed knotted physical piece of rope .
a quantum knot is meant to represent the state of the knotted rope , i.e. , the particular spatial configuration of the knot tied in the rope .
the elements of the the ambient unitary group are intended to represent all possible ways of moving the rope around ( without cutting the rope , and without letting it pass through itself . )
the quantum system is necessarily a nested set of quantum systems because one must use longer and longer pieces of rope to tie knots of greater and greater complexity . of course ,
unlike classical knotted pieces of rope , quantum knots can also represent the quantum superpositions ( and also the quantum entanglements ) of a number of knotted pieces of rope .
this raises an interesting question about the relation between topological entanglement and quantum entanglement .
let @xmath16 denote the set of the following 11 symbols @xmath17{ut00.ps}}\qquad\ \ \ \ { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\ { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\ { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps}}\ { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\qquad\ \ \ \ { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\ { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\qquad\ \ \ \ { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps}}\ { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\qquad\ \ \ \ { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}\ { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}\ ] ] called ( * unoriented * ) * tiles*. we often will also denote these tiles respectively by the following symbols@xmath6{ccccccccccccccc}t_{0}^{(u ) } & \quad & t_{1}^{(u ) } & t_{2}^{(u ) } & t_{3}^{(u ) } & t_{4}^{(u ) } & \quad & t_{5}^{(u ) } & t_{6}^{(u ) } & \quad & t_{7}^{(u ) } & t_{8}^{(u ) } & \quad & t_{9}^{(u ) } & t_{10}^{(u)}\end{array } \text { .}\ ] ] moreover , we will frequently omit the superscript ` @xmath18 ' ( standing for ` unoriented ' ) when it can be understood from context .
please note that up to rotation there are exactly 5 distinct unoriented tiles .
the above unoriented tiles are grouped according to rotational equivalence .
let @xmath13 be a positive integer .
we define an * ( unoriented ) * @xmath19**-mosaic * * as an @xmath20 matrix @xmath21 of ( unoriented ) tiles * with rows and columns indexed from * @xmath22 * to * @xmath23 .
we denote the * set of * @xmath19**-mosaics * * by @xmath24 .
two examples of unoriented 4-mosaics are shown below:@xmath6{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \qquad\qquad\qquad\begin{array } [ c]{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array}\ ] ] we now proceed to define what is meant by a knot mosaic : a * connection point * of a tile is defined as the midpoint of a tile edge which is also the endpoint of a curve drawn on the tile .
examples of tile connection points are illustrated in figure 1 below : connection-points.ps we say that two tiles in a mosaic are * contiguous * if they lie immediately next to each other in either the same row or the same column .
an unoriented tile within a mosaic is said to be * suitably connected * if each of its connection points touches a connection point of a contiguous tile .
an * ( unoriented ) * * knot * @xmath19-*mosaic * is a mosaic in which all tiles are suitably connected .
we let @xmath25 denote the subset of @xmath24 of all knot @xmath19-mosaics .
the previous two @xmath26-mosaics shown above are examples respectively of a non - knot @xmath26-mosaic and a knot @xmath26-mosaic .
other examples of knot ( or links ) mosaics are the hopf link @xmath26-mosaic , the figure eight knot @xmath27-mosaic , and the borromean rings @xmath28-mosaic , respectively illustrated below : @xmath6{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\end{array } \qquad\qquad\begin{array } [ c]{ccccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array}\ ] ] @xmath6{cccccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array}\ ] ] we now continue with our program of using mosaics to create a formal model of ( tame ) knot theory .
let @xmath29 and @xmath13 be positive integers such that @xmath30 .
a @xmath29-mosaic @xmath31 is said to be a @xmath29**-submosaic * * of an @xmath13-mosaic @xmath32 if it is a @xmath33 submatrix of @xmath32 .
the @xmath29-submosaic @xmath31 is said to be at * location * @xmath34 in the @xmath13-mosaic @xmath32 if the top left entry of @xmath31 lies in row @xmath35 and column @xmath36 of @xmath32 .
obviously , the set of possible locations for a @xmath29-submosaic of an @xmath13-mosaic is @xmath37 .
moreover , there are exactly @xmath38 different locations .
let @xmath39 denote the @xmath29**-submosaic of * * @xmath40 * at location * @xmath41 .
for example , the 3-mosaic @xmath6{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\end{array}\ ] ] is the submosaic @xmath42 of the 4-mosaic @xmath43{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \text { , } \ ] ] and the 2-mosaic @xmath6{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\end{array}\ ] ] is the submosaic @xmath44 of the 4-mosaic @xmath43{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \text { .}\ ] ] let @xmath29 and @xmath13 be positive integers such that @xmath30 .
for any two @xmath29-mosaics @xmath31 and @xmath45 , we define a @xmath46**-move at location * * @xmath47 on the set of @xmath13-mosaics @xmath24 , denoted by @xmath48 as the map from @xmath24 to @xmath24 defined by@xmath49{ll}m\text { with } m^{\left ( k : i , j\right ) } \text { replaced by } n^{\prime } & \text{if } m^{\left ( k : i , j\right ) } = n\\ m\text { with } m^{\left ( k : i , j\right ) } \text { replaced by } n & \text{if } m^{\left ( k : i , j\right ) } = n^{\prime}\\ m & \text{otherwise}\end{array } \right.\ ] ] as an example , consider the @xmath50-move @xmath51 defined by@xmath6{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\end{array } \overset{\left ( 0,1\right ) } { \longleftrightarrow}\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\end{array } \text { .}\ ] ] then , @xmath52{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\end{array } \right ) & = \begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\end{array } \text { \ ( mosaics switched)}\\ & \\ \left ( n\overset{\left ( 0,1\right ) } { \longleftrightarrow}n^{\prime}\right ) \left ( \begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\end{array } \right ) & = \begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\end{array } \text { \ ( mosaics switched)}\\ & \\ \left ( n\overset{\left ( 0,1\right ) } { \longleftrightarrow}n^{\prime}\right ) \left ( \begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\end{array } \right ) & = \begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\end{array } \text { \ ( mosaic unchanged)}\ ] ] the following proposition is an almost immediate consequence of the definition of a @xmath29-move : each @xmath29-move @xmath53 is a permutation of @xmath24 .
in fact , it is a permutation which is the product of disjoint transpositions . for the purpose of achieving clarity of exposition and of simplifying the exposition as much as possible , we adopt the following three nondeterministic notational conventions which will eliminate a great deal of combinatorial clutter : * notational convention 1 .
* _ we will use each of the following tiles _
@xmath6{cccccccccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut01.ps } } , & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut02.ps } } , & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps } } , & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut04.ps } } , & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut05.ps } } , & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps } } , & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps } } , & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } , & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut09.ps } } , & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut10.ps}}\end{array } \text { , } \ ] ] _ called * nondeterministic tiles * , to denote either both or any one ( depending on context ) of two possible tiles . _ for example , the nondeterministic tile @xmath54{iut07.ps}}$ ] denotes either both or any one of the two tiles @xmath54{ut01.ps}}$ ] and@xmath55{ut07.ps}}$ ] . *
notational convention 2 .
* _ it is to be understood that each mosaic move _
@xmath56 _ denotes either all or any one ( depending on context ) of the moves obtained by simultaneously rotating _
@xmath31 _ and _ @xmath45 _ about their respective centers by _ @xmath22 _ _ , _ _ @xmath57 _ _ , _ _ @xmath58 _ _ , or _ _ @xmath59 _ degrees . _ for example
, @xmath6{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \overset{\left ( 0,1\right ) } { \longleftrightarrow}\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\end{array}\ ] ] represents either all or any one ( depending on context ) of the following four 2-moves : @xmath60{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \overset{\left ( 0,1\right ) } { \longleftrightarrow}\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\end{array } \qquad\qquad\qquad\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps}}\end{array } \overset{\left ( 0,1\right ) } { \longleftrightarrow}\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\end{array } \\ & \\ & \begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\end{array } \overset{\left ( 0,1\right ) } { \longleftrightarrow}\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \qquad\qquad\qquad\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\end{array } \overset{\left ( 0,1\right ) } { \longleftrightarrow}\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\end{array}\end{aligned}\ ] ] as our final notational convention , we have : * notational convention 3 .
* _ finally , we omit the location superscript _ @xmath34 , and write @xmath61 to denote either all or any one ( depending on context ) of the possible locations .
* caveat : * _ we caution the reader that throughout the remainder of this paper , we will be using all of the above nondeterministic notational conventions . _ as an analog to the planar isotopy moves for standard knot diagrams , we define for mosaics the 11 * mosaic planar isotopy moves * given below:@xmath62{c}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut02.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } $ \end{tabular } \ \ \ \ } \ ] ] @xmath63{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut02.ps}}\end{array } $ \end{tabular } \ \ \ } \qquad\underset{p_{3}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\end{array } $ \end{tabular } \ \ \ } \ ] ] @xmath64{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut02.ps}}\end{array } $ \end{tabular } \ \ \ } \qquad\underset{p_{5}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps}}\end{array } $ \end{tabular } \ \ \ } \ ] ] @xmath65{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut02.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } $ \end{tabular } \ \ \ } \qquad\underset{p_{7}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut02.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } $ \end{tabular } \ \ \ } \ ] ] @xmath66{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } $ \end{tabular } \ \ \ } \qquad\underset{p_{9}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } $ \end{tabular } \ \ \ } \ ] ] @xmath67{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } $ \end{tabular } \ \ \ } \qquad\underset{p_{11}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } $ \end{tabular } \ \ \ } \ ] ] the above set of 11 planar isotopy moves was found by an exhaustive enumeration of all 2-mosaic moves corresponding to topological planar isotopy moves . the completeness of this set of moves , i.e. , that every planar isotopy moves for mosaics is a composition of a finite sequence of the above planar isotopy moves , is addressed in section 2.7 of this paper . as an analog to the reidemeister moves for standard knot diagrams , we create for mosaics the * mosaic reidemeister moves*. the * mosaic reidemeister 1 moves * are the following : @xmath68{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } $ \end{tabular } \ \ \ } \qquad\underset{r_{1}^{\prime}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } $ \end{tabular } \ \ \ } \ ] ] and the * mosaic reidemeister 2 moves * are given below : @xmath69{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } $ \end{tabular } \ \ \ } \qquad\underset{r_{2}^{\prime}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut06.ps}}\end{array } $ \end{tabular } \ \ \ } \\ & \\ & \underset{r_{2}^{\prime\prime}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps}}\end{array } $ \end{tabular } \ \ \ } \qquad\underset{r_{2}^{\prime\prime\prime}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps}}\end{array } $ \end{tabular } \ \ \ } \ ] ] for describing the mosaic reidemeister 3 moves , we will use for simplicity of exposition the following two additional notational conventions : * notational convention 4 . *
_ _ we will make use of each of the following tiles__@xmath17{ul.ps}}\quad{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ur.ps}}\quad{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ll.ps}}\quad{\includegraphics [ height=0.3269 in , width=0.3269 in ] { lr.ps}}\text { , } \ ] ] _ also called * nondeterministic tiles * , to denote either one of two possible tiles . _ for example
, the nondeterministic tile @xmath54{ul.ps}}$ ] denotes either of the following two tiles@xmath70{ul.ps}}=\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\text { \ or \ \ } \raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\text { .}\ ] ] * notational convention 5 . * _ nondeterministic tiles labeled by the same letter are synchronized as follows : _ @xmath71{c}\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ula.ps}}=\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\longleftrightarrow\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ura.ps}}=\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\\ \\\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ula.ps}}=\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\longleftrightarrow\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ura.ps}}=\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps}}\end{array } \right
. \text { \ \ and \ \ } \left\ { \begin{array } [ c]{c}\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { llb.ps}}=\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\longleftrightarrow\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { lrb.ps}}=\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps}}\\ \\\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { llb.ps}}=\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\longleftrightarrow\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { lrb.ps}}=\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}\end{array } \right.\ ] ] with these two additional notational conventions , the * mosaic reidemeister 3 moves * are given below : @xmath72{l}$\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ula.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { llb.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ura.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { lrb.ps}}\end{array } $ \end{tabular } \ } \qquad\underset{r_{3}^{\prime}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ula.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { llb.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ura.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { lrb.ps}}\end{array } $ \end{tabular } \ } $ ] @xmath73{l}$\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ula.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { llb.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ura.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { lrb.ps}}\end{array } $ \end{tabular } \ } \qquad\underset{r_{3}^{\prime\prime\prime}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ula.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { llb.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ura.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { lrb.ps}}\end{array } $ \end{tabular } \ } $ ] @xmath74{l}$\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ula.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { llb.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ura.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { lrb.ps}}\end{array } $ \end{tabular } \
\qquad}\underset{r_{3}^{(v)}}{\begin{tabular } [ c]{l}$\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ula.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut03.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { llb.ps}}\end{array } \longleftrightarrow\begin{array } [ c]{ccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ura.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { iut07.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { iut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { lrb.ps}}\end{array } $ \end{tabular } } $ ] as noted in a previous section , all mosaic moves are permutations on the set of mosaics @xmath24 .
in particular , the planar isotopy moves and the reidemeister moves lie in the permutation group of the set of mosaics .
it easily follows that the planar isotopy moves and the reidemeister moves also lie in the group of all permutations of the set of knot mosaics @xmath25 .
hence , we can make the following definition : we define the ( * knot mosaic * ) * ambient group * @xmath75 as the group of all permutations of the set of knot @xmath13-mosaics @xmath25 generated by the mosaic planar isotopy and the mosaic reidemeister moves .
it follows from a previous proposition that the mosaic planar isotopy moves and reidemeister moves , as permutations , are each the product of disjoint transpositions .
the completeness of the set of planar isotopy and reidemeister moves is addressed in section 2.7 of this paper .
we now are prepared to define the analog of knot type for mosaics .
we define the * mosaic injection * @xmath6{rrr}\iota:\mathbb{m}^{(n ) } & \longrightarrow & \mathbb{m}^{(n+1)}\\ m^{(n ) } & \longmapsto & m^{(n+1)}\end{array}\ ] ] as@xmath76{cl}m_{ij}^{(n ) } & \text{if } 0\leq i , j < n\\ & \\\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & \text{otherwise}\end{array } \right.\ ] ] thus , @xmath77{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \overset{\iota}{\longrightarrow}m^{(n+1)}=\begin{array } [ c]{ccccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array}\ ] ] we now can explicitly define the * graded system * @xmath78 that was mentioned in the introduction . the symbol @xmath1 denotes the directed system of sets @xmath79 and @xmath2 denotes the directed system of permutation groups @xmath80 .
thus , @xmath81 two @xmath13-mosaics @xmath32 and @xmath31 are said to be of the * same knot * @xmath13**-type * * , written@xmath82 provided there is an element of the ambient isotopy group @xmath75 which transforms @xmath32 into @xmath31 .
an @xmath83-mosaic @xmath32 and an @xmath13-mosaic @xmath31 are said to be of the * same knot mosaic type * , written@xmath84 provided there exists a non - negative integer @xmath85 such that , if @xmath86 , then @xmath87 or if @xmath88 , then @xmath89 where , for each non - negative integer @xmath90 , @xmath91 denotes the @xmath90-fold composition @xmath92 . in the introduction of this paper , we conjecture that the formal ( re - writing ) system @xmath0 of knot mosaics fully captures the entire structure of tame knot theory .
we now explain in greater detail what is meant by this conjecture .
let @xmath93 denote the set of integers , and @xmath94 the two dimensional euclidean plane .
let @xmath95 denote the * square tiling * of @xmath94 induced by the sublattice @xmath96 of @xmath94 , and for each @xmath35 , @xmath36 in @xmath93 , let @xmath97 denote the subregion of @xmath94 defined by@xmath98 let @xmath29 be an arbitrary tame knot in @xmath15-space @xmath99 . a knot diagram of @xmath29 , i.e.
, a regular projection@xmath100 is said to be a * mosaic knot diagram * if * the image under @xmath101 of @xmath29 lies in the first quadrant of @xmath94 , and * for all @xmath35 , @xmath36 in @xmath93 , the pair @xmath102 is identical with the cell pair on one of the faces of the 11 tiles @xmath103 , @xmath104 , @xmath105 , @xmath106 . clearly , using standard arguments in knot theory
, one can prove that every tame knot ( or link ) has a mosaic knot diagram .
each mosaic knot diagram @xmath107 of a knot @xmath29 can naturally be identified with a knot @xmath13-mosaic @xmath108 , where @xmath13 is the smallest positive integer such that @xmath109 lies in the region@xmath110 moreover , every knot @xmath13-mosaic can naturally be identified with the diagram of a knot @xmath29 .
we call this associated knot mosaic @xmath108 a * ( knot * ) * mosaic representative * of the original knot @xmath29 .
this leads us to the following conjecture : let @xmath111 and @xmath112 be two tame knots ( or links ) , and let @xmath113 and @xmath114 be two arbitrary chosen mosaic representatives of @xmath111 and @xmath112 , respectively .
then @xmath111 and @xmath112 are of the same knot type if and only if the representative mosaics @xmath113 and @xmath114 are of the same knot mosaic type .
in other words , knot mosaic type is a complete invariant of tame knots .
our sole purpose in creating the formal system @xmath0 of knot mosaics was to create a framework within which we can explicitly define what is meant by a quantum knot .
we are finally in a position to do so .
we begin by assigning a left - to - right * linear ordering , * denoted by ` @xmath115 ' , to the 11 mosaic tiles as indicated below @xmath116{ut00.ps}}}\begin{array } [ c]{c}<\\ \mathstrut \end{array } \underset{t_{1}}{{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}}\begin{array } [ c]{c}<\\ \mathstrut \end{array } \underset{t_{2}}{{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps}}}\begin{array } [ c]{c}<\\ \mathstrut \end{array } \underset{t_{3}}{{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps}}}\begin{array } [ c]{c}<\\ \mathstrut \end{array } \underset{t_{4}}{{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}}\begin{array } [ c]{c}<\\ \mathstrut \end{array } \underset{t_{5}}{{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut05.ps}}}\begin{array } [ c]{c}<\\ \mathstrut \end{array } \underset{t_{6}}{{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut06.ps}}}\begin{array } [ c]{c}<\\ \mathstrut \end{array } \underset{t_{7}}{{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut07.ps}}}\begin{array } [ c]{c}<\\ \mathstrut \end{array } \underset{t_{8}}{{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}}\begin{array } [ c]{c}<\\ \mathstrut \end{array } \underset{t_{9}}{{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut09.ps}}}\begin{array } [ c]{c}<\\ \mathstrut \end{array } \underset{t_{10}}{{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}}$ ] we let @xmath117 be the 11 dimensional hilbert space with orthonormal basis @xmath118 labeled by the above mosaic tiles , and we define the * hilbert space * @xmath119 * of * @xmath19**-mosaics * * as the tensor product @xmath120 thus , the induced orthonormal basis of @xmath119 consists of all possible @xmath121-fold tensor products of the above 11 mosaic tiles , i.e. , the induced basis is @xmath122 we then use the above defined linear ordering on the set @xmath123 of mosaic tiles to * lexicographically ( lex ) order * all the basis elements of @xmath119 .
we also denote this linear ordering by ` @xmath115 ' . finally , using * row major order *
, we identify each basis element @xmath124 with the ket @xmath125 labeled by the @xmath13-mosaic @xmath126 .
in other words , we have used row major order to set up a one - to - one correspondence between basis elements of @xmath119 and the set of @xmath13-mosaics @xmath24 .
for example , for @xmath127 the basis element@xmath128 is identified with the @xmath13-mosaic labeled ket@xmath129{ccc}t_{2 } & t_{5 } & t_{4}\\ t_{9 } & t_{2 } & t_{1}\\ t_{5 } & t_{8 } & t_{3}\end{array } \right\rangle \text { .}\ ] ] we can now define the * hilbert space of knot * @xmath19**-mosaics * * @xmath130 as the sub - hilbert space @xmath130 of @xmath119 spanned by all orthonormal basis elements labeled by knot @xmath13-mosaics . since @xmath131{ut00.ps}}\right\ } \text { \ \ and \ \ } \mathbb{k}^{(2)}=\left\ { \ \begin{array } [ c]{cc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } , \quad\begin{array } [ c]{cc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } \ \right\ } \text { , } \ ] ] the first two hilbert spaces @xmath132 and @xmath133 are of dimensions @xmath134 and @xmath50 , respectively .
the third hilbert space @xmath135 is of dimension @xmath136 , as is demonstrated by the complete list of all possible knot @xmath15-mosaics given appendix a. an example of an element of the hilbert space @xmath137 is @xmath138{cccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle + \left\vert \begin{array } [ c]{cccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle } { \sqrt{2}}\ ] ] our next step is to identify each element @xmath139 of the ambient group @xmath75 with the linear transformation@xmath6{ccc}\mathcal{k}^{(n ) } & \overset{g}{\longrightarrow } & \mathcal{k}^{(n)}\\ \left\vert k\right\rangle & \longmapsto & \left\vert gk\right\rangle \end{array } \text { .}\ ] ] this is a unitary transformation , since the element @xmath139 simply permutes the basis elements of @xmath130 . in this way
, the ambient group @xmath75 is identified with a discrete unitary subgroup ( also denoted by @xmath75 ) of the group @xmath140 , where @xmath140 denotes the * group of all unitary transformations on the hilbert space * @xmath130 .
the unitary subgroup @xmath75 will also be called the * ambient group*. finally , everything comes together with the following definition : let @xmath13 be a positive integer . a * quantum knot system * @xmath14 * of order * @xmath13 is a quantum system with the hilbert space @xmath130 of knot @xmath13-mosaics as its state space , and having the ambient group @xmath75 as an accessible unitary control group .
the states of the quantum system @xmath14 are called * quantum knots of order * @xmath13 , and the elements of the ambient group @xmath75 are called * unitary knot moves*. moreover , the quantum knot system @xmath14 of * * order * * @xmath13 is a subsystem of the quantum knot system @xmath141 of order * * @xmath142 .
thus , the quantum knot systems @xmath143 collectively become a * nested * * sequence of quantum knot systems * which we will denote simply by @xmath10 . in other words ,
@xmath144 the nested quantum knot system @xmath10 is probably not a physically realizable system .
however , each quantum knot system @xmath143 of order @xmath13 is physically realizable is physically realizable , it may or may not be implentable within today s existing technology . ] . as an example , in the quantum system @xmath145 , we see that the action of the unitary reidemeister 2 move @xmath6{cc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps}}\end{array } \overset{(2,1)}{\longleftrightarrow}\begin{array } [ c]{cc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps}}\end{array } \in\mathbb{a}(5)\ ] ] on the quantum knot @xmath138{ccccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle + \left\vert \begin{array } [ c]{ccccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle } { \sqrt{2}}\ ] ] produces the quantum knot @xmath138{ccccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle + \left\vert \begin{array } [ c]{ccccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle } { \sqrt{2}}\text { .}\ ] ] when are two quantum knots the same ?
let @xmath146 and @xmath147 be two quantum knots of a quantum system @xmath143 of order @xmath13 .
then @xmath148 and @xmath149 are said to be of the * same quantum knot * @xmath19**-type * * , written@xmath150 provided there exists a unitary transformation @xmath139 in the ambient group @xmath75 which transforms @xmath146 into @xmath149 , i.e. , such that@xmath151 they are said to be of the * same quantum knot type * , written @xmath152 provided that for some non - negative integer @xmath85 , @xmath153 where @xmath154 and @xmath155 are states of the quantum system @xmath156 , where @xmath157 is the monomorphism induced by the previously defined injection @xmath158 .
thus , the two quantum knots found in the last example of the previous section are of the same quantum knot @xmath27-type , and also of the same quantum knot type .
surprisingly , the following two quantum knots @xmath148 and @xmath149 are neither of the same quantum knot @xmath15-type nor knot type : @xmath159{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } \right\rangle \text { \ \ and \ \ } \left\vert \psi_{2}\right\rangle = \frac { 1}{\sqrt{2}}\left ( \left\vert \begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle + \left\vert \begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } \right\rangle \right)\ ] ] this follows from the fact that the ambient group @xmath75 is generated by a finite set of involutions . in this section , we show how to find the hamiltonians associated with the generators of the ambient group @xmath160 , i.e. , the planar isotopy and reidemeister moves for quantum knots .
let @xmath139 be an arbitrary planar isotopy move or reidemeister move in the ambient group @xmath75 . from proposition 1
, we know that @xmath139 , as a permutation , is the product of disjoint transpositions of knot @xmath13-mosaics , i.e. , of the form@xmath161 without loss of generality , we can assume that @xmath162 for @xmath163 , and that @xmath164 for @xmath165 , where ` @xmath115 ' denotes the previously defined lexicographic ( lex ) order on the set of @xmath13-mosaics @xmath24 . for each permutation @xmath166 of @xmath24 , let ` @xmath167 '
denote the new linear ordering created by the application of the permutation @xmath166 .
choose a permutation @xmath166 such that @xmath168 with @xmath169 @xmath167 all other @xmath13-mosaics , and let @xmath170 and @xmath171 denote respectively the identity matrix and the first pauli spin matrix given below@xmath172{cc}1 & 0\\ 0 & 1 \end{array } \right ) \text { \ \ and \ \ } \sigma_{1}=\left ( \begin{array } [ c]{cc}0 & 1\\ 1 & 0 \end{array } \right ) \text { .}\ ] ] then in the @xmath166-reordered basis of the hilbert space @xmath130 , the element @xmath139 , as a unitary transformation , is of the form@xmath173{ccccc}\sigma_{1 } & o & \ldots & o & o\\ o & \sigma_{1 } & \ldots & o & o\\ \vdots & \vdots & \ddots & & \vdots\\ o & o & \ldots & \sigma_{1 } & o\\ o & o & \ldots & o & i_{n-2\ell}\end{array }
\right ) = \left ( i_{\ell}\otimes\sigma_{1}\right ) \oplus i_{n-2\ell}\text { , } \ ] ] where ` @xmath174 ' denotes an all zero matrix of appropriate size , where @xmath175 denotes the @xmath176 identity matrix , and where ` @xmath177 ' denotes the direct sum of matrices , i.e. , @xmath178{cc}a & o\\ o & b \end{array } \right ) $ ] . the natural log of @xmath171 is@xmath179 where @xmath180 denotes an arbitrary integer . hence , the natural log , @xmath181 , of the unitary transformation @xmath182 is @xmath183{cc}\left ( \begin{array } [ c]{cccc}\left ( 2s_{1}+1\right ) \left ( \sigma_{0}-\sigma_{1}\right ) & o & \ldots & o\\ o & \left ( 2s_{2}+1\right ) \left ( \sigma_{0}-\sigma_{1}\right ) & \ldots & o\\ \vdots & \vdots & \ddots & \vdots\\ o & o & \ldots & \left ( 2s_{\ell}+1\right ) \left ( \sigma_{0}-\sigma _ { 1}\right ) \end{array } \right ) & o\\ o & o_{\left ( n-2\ell\right ) \times\left ( n-2\ell\right ) } \end{array } \right ) $ ] where @xmath184 are arbitrary integers .
be an arbitrary finite @xmath185 unitary matrix , and let @xmath186 be a unitary matrix that diagonalizes @xmath187 , i.e. , a unitary matrix @xmath186 such that @xmath188 .
then the natural log of @xmath189 is @xmath190 . ]
since we are interested only in the simplest hamiltonian , we choose the * principal branch * @xmath191 of the natural log , i.e. , the branch for which @xmath192 , and obtain for our hamiltonian@xmath193 \eta ^{-1}=\frac{\pi}{2}\eta\left ( \begin{array } [ c]{cc}i_{\ell}\otimes\left ( \sigma_{0}-\sigma_{1}\right ) & o\\ o & o_{\left ( n-2\ell\right ) \times\left ( n-2\ell\right ) } \end{array } \right ) \eta^{-1}\ ] ] thus , if the initial quantum knot is@xmath129{ccccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle \text { , } \ ] ] and if we use the hamiltonian @xmath194 for for the reidemeister 2 move @xmath195{cc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps}}\end{array } \overset{(2,1)}{\longleftrightarrow}\begin{array } [ c]{cc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps}}\end{array } \text { , } \ ] ] then the solution to schroedinger s equation is @xmath196{ccccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle -i\sin\left ( \frac{\pi t}{2\hslash}\right ) \left\vert \begin{array } [ c]{ccccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle \right)\ ] ] where @xmath197 denotes time , and where @xmath198 denotes planck s constant divided by @xmath199 . we should also mention a number of other miscellaneous unitary transformations that do not lie in the ambient group @xmath160 , but are nonetheless of interest . there is the unitary transformation @xmath97 , called the * tunnelling transformation * , given by @xmath200{ut09.ps}}\overset{(i , j)}{\longleftrightarrow}\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}\text { , } \ ] ] which enables a quantum knot overcrossing to tunnel into an undercrossing , or vice versa . from this , we can construct the * mirror image transformation * given by@xmath201{ut09.ps}}\overset{(i , j)}{\longleftrightarrow}\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut10.ps}}\right)\ ] ] that transforms a quantum knot into its mirror image .
the following two unitary transformations can be used to create four dimensional quantum knots .
the first is the * hyperbolic transformation * @xmath202 given by@xmath203{ut07.ps}}\overset{(i , j)}{\longleftrightarrow}\raisebox{-0.1003in}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut08.ps}}\text { , } \ ] ] and the second is the * elliptic transformation * @xmath204 given by@xmath205{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut01.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut04.ps}}\end{array } \overset{(i , j)}{\longleftrightarrow}\begin{array } [ c]{cc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \text { .}\ ] ] more will be said about these transformations in future papers .
we now consider the following question : * question . *
_ what is a quantum knot invariant ?
how do we define it ? _ the objective of the first half of this section is to give a discursive argument that justifies a definition which will be found to be equivalent to the following : _ a _ * quantum knot * @xmath13**-invariant * * _ for a quantum system _
@xmath143 _ is an observable _ @xmath206 _ on the hilbert space _ @xmath130 _ of quantum knots which is invariant under the action of the ambient group _ @xmath75 _ _
, i.e. , such that _ _
@xmath207 _ for all _ @xmath187 in @xmath75__. _ _ * caveat : * _ we emphasize to the reader that the above definition of quantum knot invariants is not the one currently used in quantum topology .
quantum topology uses analogies with quantum mechanics to create significant mathematical structures that do not necessarily correspond directly to quantum mechanical observables .
the invariants of quantum topology have been investigated for their relevance to quantum computing and they can be regarded , in our context , as possible secondary calculations made on the basis of an observable . here
we are concerned with observables that are themselves topological invariants .
_ to justify our new use of the term ` quantum knot invariant , ' we will use the following yardstick : yardstick : _ quantum knot invariants are to be physically meaningful invariants of quantum knot type . by `` physically meaningful , ''
we mean that the quantum knot invariants can be directly obtained from experimental data produced by an implementable physical experiment .
is physically implementable , it may or may not be implentable within today s existing technology . ] _ let @xmath143 be a quantum knot system , where @xmath130 is the hilbert space of quantum knots , and where @xmath75 is the underlying ambient group on @xmath130
. moreover , let @xmath208 denote some yet - to - be - chosen mathematical domain . by an @xmath13**-invariant * * @xmath209 of quantum knots , we mean a map@xmath210 such that , when two quantum knots @xmath146 and @xmath149 are of the same knot @xmath13-type , i.e. , when@xmath211 then their respective invariants must be equal , i.e. , @xmath212 in other words , @xmath213 is a map which is invariant under the action of the ambient group @xmath75 , i.e.,@xmath214 for all elements @xmath139 in @xmath75 . *
question : * _ but which such invariants are physically meaningful ? _ we begin to try to answer this question by noting that the only way to extract information from a quantum system is through quantum measurement .
thus , if we wish to extract information about quantum knot type from a quantum knot system @xmath143 _ _ , _ _ we of necessity must make a measurement with respect to some observable .
but what kind of observable ? with this in mind , we will now describe quantum measurement from a different , but nonetheless equivalent perspective , than that which is usually given in standard texts on quantum mechanics . for knot theorists who might not be familiar with standard quantum measurement ,
we have included in the figure below a brief summary of quantum measurement .
+ * von neumann measurement .
* let @xmath206 be an observable for a quantum system @xmath143 , i.e. , a hermitian ( self - adjoint ) linear operator on the hilbert space @xmath130 .
moreover , let @xmath215 be the spectral decomposition of the observable @xmath206 , where @xmath216 is the @xmath36-th eigenvalue of @xmath206 , and where @xmath217 is the corresponding projection operator for the associated eigenspace @xmath218 .
let @xmath219 denote the * set of all probability distributions on the spectrum * @xmath220 of @xmath206 , i.e.,@xmath221 : \sum _ { j=1}^{m}p\left ( \lambda_{j}\right ) = 1\right\}\ ] ] we will call the probability distributions @xmath90 of @xmath219 * stochastic sources*. then each observable @xmath206 uniquely determines a map@xmath6{cccc}\widetilde{\omega } : & \mathcal{k}^{(n ) } & \longrightarrow & \mathcal{p}_{\omega}^{(n)}\\ & \left\vert \psi\right\rangle & \longmapsto & p \end{array}\ ] ] from quantum knots to stochastic sources on the spectrum of @xmath206 given by@xmath222 thus , what is seen , when a quantum system @xmath223 in state @xmath224 is measured with respect to an observable @xmath206 , is a random sample from the stochastic source @xmath225 .
but under what circumstances is such a random sample a quantum knot invariant ?
our answer to this question is that quantum knots @xmath148 and @xmath149 of the same knot @xmath13-type must produce random samples from the same stochastic source when measured with respect to the observable @xmath206 .
this answer is captured by the following definition : let @xmath143 be a quantum knot system , and let @xmath206 be an observable on @xmath130 with spectral decomposition @xmath226 then the observable @xmath206 is said to be a * quantum knot * @xmath13**-invariant * * provided@xmath227 for all quantum knots @xmath228 , for all @xmath229 , and for all projectors @xmath217 .
let @xmath143 and @xmath206 be as given in the above definition .
then the following statements are equivalent : * the observable @xmath206 is a quantum knot @xmath13-invariant * @xmath230 = 0 $ ] for all @xmath229 and for all @xmath217 .
* @xmath231 = 0 $ ] for all @xmath229 , where @xmath232 $ ] denotes the commutator @xmath233 of operators @xmath189 and @xmath234 .
the remaining half of this section is devoted to finding an answer to the following question : * question : * _ how do we find observables which are quantum knot invariants ? _
one answer to this question is the following theorem , which is an almost immediate consequence of the definition of a minimum invariant subspace of @xmath130 : let @xmath143 be a quantum knot system , and let @xmath235 be a decomposition of the representation@xmath236 into irreducible representations of the ambient group @xmath75 .
then , for each @xmath85 , the projection operator @xmath237 for the subspace @xmath238 is an observable which is a quantum knot @xmath13-invariant
. here is yet another way of finding quantum knot invariants : let @xmath143 be a quantum knot system , and let @xmath206 be an observable on the hilbert space @xmath130 .
let @xmath239 be the stabilizer subgroup for @xmath206 , i.e. , @xmath240 then the observable@xmath241 is a quantum knot @xmath13-invariant , where @xmath242 denotes a sum over a complete set of coset representatives for the stabilizer subgroup @xmath243 of the ambient group @xmath75 .
the observable @xmath244is obviously an quantum knot @xmath13-invariant , since @xmath245 for all @xmath246 .
if we let @xmath247 denote the order of @xmath247 , and if we let @xmath248 denote a complete set of coset representatives of the stabilizer subgroup @xmath249 , then @xmath250 is also a quantum knot invariant .
we end this section with an example of a quantum knot invariant : the following observable @xmath206 is an example of a quantum knot @xmath26-invariant : @xmath251{cccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle \left\langle \begin{array } [ c]{cccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\vert + \left\vert \begin{array } [ c]{cccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle \left\langle \begin{array } [ c]{cccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\vert $ ] for yet another approach to quantum knot measurement , we refer the reader to the brief discussion on quantum knot tomography found in item 11 ) in the conclusion of this paper .
there are many possible open questions and future directions for research .
we mention only a few .
* what is the exact structure of the ambient group @xmath75 and its direct limit @xmath252 can one write down an explicit presentation for @xmath75 ?
for @xmath2 ?
the fact that the ambient group @xmath75 is generated by involutions suggests that it may be a coxeter group .
is it a coxeter group ? * unlike classical knots , quantum knots can exhibit the non - classical behavior of quantum superposition and quantum entanglement .
are topological entanglement and quantum entanglement related to one another ?
if so , how ?
* what other ways are there to distinguish quantum knots from classical knots ? * how does one find a quantum observable for the jones polynomial ? this would be a family of observables parameterized by points on the unit circle in the complex plane . does this approach lead to an algorithmic improvement to the quantum algorithm given by aharonov , jones , and landau in @xcite ?
( see also @xcite , @xcite . )
* how does one create quantum knot observables that represent other knot invariants such as , for example , the vassiliev invariants ? * what is gained by extending the definition of quantum knot observables to povms ? * what is gained by extending the definition of quantum knots to mixed ensembles ? * define the * mosaic number * of a knot @xmath29 as the smallest integer @xmath13 for which @xmath29 is representable as a knot @xmath13-mosaic .
for example , the mosaic number of the trefoil is @xmath26 , as is illustrated by the following knot @xmath13-mosaic:@xmath6{cccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array}\ ] ] in general , how does one compute the mosaic number ? is the mosaic number related to the crossing number of a knot ? how does one find an observable for the mosaic number ?
* let @xmath253 denote the dimension of the hilbert space @xmath130 of quantum knot @xmath13-mosaics .
we have shown that @xmath254 , @xmath255 , and @xmath256 .
find @xmath253 for other values of @xmath13 .
a very loose upper bound for @xmath253 is obviously @xmath257 . *
consider the following alternate stronger definitions of quantum knot @xmath13-type and quantum knot type : let @xmath14 be a quantum knot system , and let @xmath258 denote the lie group of all unitary transformations on the hilbert space @xmath130 .
define the * continuous ambient group * @xmath259 as the smallest connected lie subgroup of @xmath258 containing the discrete ambient group @xmath75 .
+ let @xmath260 denote the set of planar isotopy and reidemeister generators of the discrete ambient group @xmath75 , and let @xmath261 be the lie algebra generated by the elements of the set@xmath262 where @xmath191 denotes the principal branch of the natural log on @xmath258 .
then the continuous ambient group is given by@xmath263 + we define two quantum @xmath13-knots @xmath146 and @xmath149 to be of the * same continuous knot * @xmath13**-type * * , written @xmath264 provided there exists an element @xmath139 of the continuous ambient group @xmath259 which transforms @xmath265 into @xmath149 , i.e. , such that @xmath266 . they are of the * same continuous knot type * , written @xmath267 , if there exists an integer @xmath85 such that @xmath268
. + let @xmath113 and @xmath114 denote two knot @xmath13-mosaics , and let @xmath269 and @xmath270 denote the corresponding quantum knots .
then @xmath271 + thus , if this conjectures 1 and 2 are true , these two stronger definitions of quantum knot @xmath13-type and quantum knot type fully capture all of classical tame knot theory .
moreover , these two stronger definitions have a number of advantages over the weaker definitions , two of which are the following : * under the hamiltonians associated with the generators @xmath260 , the schroedinger equation determines a connected continuous path in @xmath130 consisting of quantum @xmath13-knots all of the same quantum continuous knot @xmath13-type . *
although the following two quantum knots are not of the same discrete knot @xmath13-type , they are however of the same continuous knot @xmath13-type@xmath159{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } \right\rangle \text { \ \ and \ \ } \left\vert \psi_{2}\right\rangle = \frac { 1}{\sqrt{2}}\left ( \left\vert \begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle + \left\vert \begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } \right\rangle \right)\ ] ] is the above conjecture true ?
if so , what is the structure of the continuous ambient group @xmath259 ?
what are its irreducible representations ?
* quantum knot tomography : given repeated copies of a quantum knot @xmath272 , how does one employ the method of quantum state tomography@xcite to determine @xmath224 ?
most importantly , how can this be done with the greatest efficiency ? for example , given repeated copies of the unknown quantum knot basis state @xmath273{cccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } \right\rangle \text { \ , } \ ] ] one could make repeated measurements of this state with respect to the @xmath274 ( for @xmath275 ) tile observables @xmath276 to determine the state with a probability of error less than a chosen positive threshold @xmath277 .
this is obviously not the most efficient set of observables for the task , nor is it a universal set of observables for quantum knot tomography .
given many copies of an arbitrary quantum knot , how does find a universal set observables that is best in the sense of greatest efficiency for a given threshold @xmath277 ? * quantum braids : one can also use mosaics to define quantum braids .
how is this related to the work found in @xcite , @xcite , @xcite ?
* can quantum knot systems be used to model and to predict the behavior of * * quantum vortices in liquid helium ii ?
( see @xcite . ) * * quantum vortices in the bose - einstein condensate ? *
* fractional charge quantification that is manifest in the fractional quantum hall effect ?
( see @xcite and @xcite . )
we should mention that we have throughout this paper used a version of knot diagrams ( i.e. , two dimensional knot mosaics ) , that is susceptible to quantization . within this context
, the familiar reidemeister moves have become unitary transformations on an appropriate hilbert space .
knots as we know them in topology and geometry occur as embeddings in three dimensional space , and are projected to knot diagrams for combinatorial and topological purposes .
thus , we have chosen to model quantum knots through the extra structure of knot diagrams .
however , it is also possible to investigate quantum knots in full three dimensional space by using three dimensional knot mosaics
. this will be the subject of a forthcoming paper . in closing this section
, we should finally also say that , in the open literature , the phrase `` quantum knot '' has many different meanings , and is sometimes a phrase that is used loosely .
we mention only two examples . in @xcite , a quantum knot
is essentially defined as an element of the hilbert space with orthonormal basis in one - one - correspondence with knot types , rather than knot representatives . within the context of the mosaic construction , a quantum knot in @xcite corresponds to an element of the orbit hilbert space @xmath278 .
in @xcite and in @xcite the phrase `` quantum knot '' refers not to knots , but to the use of representations of the braid group to model the dynamic behavior of certain quantum systems . in this contexts
, braids are used as a tool to model topological obstructions to quantum decoherence that are conjectured to exist within certain quantum systems .
the first author would like to thank the institute for scientific interchange ( isi ) in torino , italy , and the mathematical sciences research institute in berkeley , california for providing a research climate where many of the ideas found within this paper could germinate .
this effort was partially supported by the defense advanced research projects agency ( darpa ) and air force research laboratory , air force materiel command , usaf , under agreement number f30602 - 01 - 2 - 0522 , and by l - oo - p fund grant beca2002 .
the second author would like to thank the national science foundation for support under nsf grant dms-0245588 .
99 dorit aharonov , vaughan jones , and zeph landau , * on the quantum algorithm for approximating the jones polynomial * , http://arxiv.org/abs/quant-ph/0511096 chen , goong , louis kauffman , and samuel j. lomonaco , ( eds . ) , * `` the mathematics of quantum computation and quantum topology , '' * chapman & hall / crc , ( 2007 ) collins , graham p. , * computing with quantum knots , * sci .
amer . , april , 2006 , pp56 - 63 .
crowell , richard h. , and ralph h. fox , * `` introduction to knot theory , '' * springer - verlag , ( 1977 ) .
helstrom , carl w. , * `` quantum detection and estimation theory , '' * academic press , ( 1976 ) .
jacak , lucjan , piotr sitko , konrad wieczorek , and arkadiusz wjs , * `` quantum hall systems : braid groups , composite fermions , and fractional charge , '' * oxford university press , ( 2003 ) .
kauffman , louis h. , and samuel j. lomonaco , * quantum knots * , quantum knots , proc .
spie , ( 2004 ) .
http://arxiv.org/abs/quant-ph/0403228 .
kauffman , louis h. and samuel j. lomonaco , jr .
, * q - deformed spin networks , knot polynomials , and anyonic topological quantum computation * , journal of knot theory , vol . 16 , no .
3 , ( 2007 ) , pp 267 - 332 .
http://xxx.lanl.gov/abs/quant-ph/0606114 kauffman , louis h. , and samuel j. lomonaco , jr .
, * spin networks and anyonic topological computing * , proc .
spie , vol .
6244 , ( 2006 ) , http://xxx.lanl.gov/abs/quant-ph/0603131 kauffman , louis h. , * `` knots and physics , '' * ( 3rd edition ) , world scientific , ( 2001 ) .
kitaev , alexei yu , * fault - tolerant quantum computation , by anyons * , http://arxiv.org/abs/quant-ph/9707021 .
leonhardt , u. , * `` measuring the quantum state of light , '' * cambridge university press , ( 1997 ) .
lickorish , w.b .
raymond , * `` an introduction to knot theory , '' * springer , ( 1997 ) .
lomonaco , samuel j. , * the modern legacies of thomson s atomic vortex theory in classical electrodynamics * , ams psapm/51 , providence , ri ( 1996 ) , 145 - 166 .
lomonaco , samuel j. , jr . , and louis h. kauffman , * topological quantum computing and the jones polynomial * , proc .
spie , vol .
6244 , ( 2006 ) , http://xxx.lanl.gov/abs/quant-ph/0605004 lomonaco , samuel j. , jr . , * a rosetta stone for quantum mechanics with an introduction to quantum computation * , psapm , vol .
58 , ams , providence , ri , ( 2002 ) , 3 - 65 .
lomonaco , samuel j. , jr . , ( ed . ) , * `` quantum computation , '' * psapm , vol .
58 , american mathematical society , providence , rhode island , ( 2002 ) .
lomonaco , samuel j. , jr .
, * five dimensional knot theory * , ams conm , vol .
20 , ( 1984 ) , pp 249 - 270 .
( http://www.cs.umbc.edu/lomonaco/5knots/5knots.pdf ) lomonaco , samuel j. , jr . , * the homotopy groups of knots i. how to compute the algebraic 3-type , * pacific journal of mathematics , vol .
95 , no . 2 ( 1981 ) , pp 349 - 390 .
murasugi , kunio , * `` knot theory and its applications , '' * birkhauser , ( 1996 ) .
nielsen , michael a. , and isaac l. chuang , * `` quantum computation and quantum information , '' * , cambridge university press , ( 2000 ) . peres , asher , * `` quantum theory : concepts and methods , '' * kluwer , ( ( 1993 ) .
rasetti , mario , and tullio regge , * vortices in he ii , current algebras and quantum knots * , physica 80 a , north - holland , ( 1975 ) , 217 - 2333 .
sakuri , j. j. , * `` modern quantum mechanics , '' * ( revised edition ) , addison - wesley , ( 1994 ) .
shankar , r. , * `` principles of quantum mechanics , '' * ( 2nd edition ) , plenum , ( 1994 ) .
sarma , sankar das , michael freedman and chetan nayak , * topologically protected qubits from a possible non - abelian fractional quantum hall state , * phys .
letters , vol .
94 , ( 2005 ) , pp 166802 - 1168802 - 4 .
shor , peter w. , and stephen p. jordan , * estimating jones polynomials is a complete problem for one clean qubit , * http://arxiv.org/abs/0707.2831 .
wen , xiao - gang , * `` quantum field theory of many - body systems , '' * oxford press , ( 2004 ) .
zee , a. , * `` quantum field theory in a nutshell , '' * princeton university press , ( 2003 ) .
a complete list of all knot 3-mosaics , listed in lexicographic ( lex ) order , is given below : @xmath279{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } $ } } \quad\underset{k_{1}=\text { 000 - 021 - 034}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } $ } } \quad\underset{k_{2}=\text { 000 - 210 - 349}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } $ } } \quad\underset{k_{3}=\text { 000 - 251 - 354}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } $ } } \quad\underset{k_{4}=\text { 021 - 034 - 000}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } $ } } $ ] @xmath280 @xmath281{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } $ } } \quad\underset{k_{6}=\text { 021 - 246 - 354}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } $ } } \quad\underset{k_{7}=\text { 021 - 274 - 340}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } $ } } \quad\underset{k_{8}=\text { 021 - 284 - 340}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut08.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } $ } } \quad\underset{k_{9}=\text { 021 - 294 - 340}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } $ } } $ ] @xmath282{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } $ } } \quad\underset{k_{11}=\text { 210 - 340 - 000}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } $ } } \quad\underset{k_{12}=\text { 210 - 371 - 034}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut07.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } $ } } \quad\underset{k_{13}=\text { 210 - 381 - 034}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut08.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } $ } } \quad\underset{k_{14}=\text { 210 - 391 - 034}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut09.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } $ } } $ ] @xmath283{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut10.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } $ } } \quad\underset{k_{16}=\text { 210 - 631 - 354}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in ,
width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } $ } } \quad\underset{k_{16}=\text { 210 - 660 - 340}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in ,
width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } $ } } \quad\underset{k_{18}=\text { 251 - 316 - 034}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } $ } } \quad\underset{k_{19}=\text { 251 - 354 - 000}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in ,
width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } $ } } $ ] @xmath284{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\end{array } $ } } \quad\underset{k_{21}=\text { 251 - 624 - 340}}{\fbox{$\begin{array } [ c]{ccc}{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut05.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut01.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut06.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut02.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps}}\\{\includegraphics [ height=0.1773 in , width=0.1773 in ] { ut03.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut04.ps } } & { \includegraphics [ height=0.1773 in , width=0.1773 in ] { ut00.ps}}\end{array } $ } } $ ]
so far , we have only discussed unoriented objects in this paper , e.g. , unoriented mosaics , unoriented knot mosaics , unoriented quantum knots , and so forth . in this appendix
, we briefly discuss how all these unoriented objects can be transformed into oriented ones . called ( * oriented * ) * tiles*. we often will also denote these tiles respectively by the symbols @xmath286 moreover , we will frequently omit the superscript ` @xmath287 ' ( standing for ` oriented ' ) when it can be understood from context .
let @xmath13 be a positive integer .
we define an * ( oriented ) * @xmath19**-mosaic * * as an @xmath20 matrix @xmath21 of ( oriented ) tiles with rows and columns indexed from @xmath22 to @xmath23 .
we let @xmath24 also denote the * set of oriented * @xmath19**-mosaics**. two examples of oriented @xmath26-mosaics are shown are shown below:@xmath6{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot01.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ot12.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot22.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot09.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ot27.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot11.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot15.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot06.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ot07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot04.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot12.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array } \qquad\qquad\qquad\begin{array } [ c]{cccc}{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot05.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ot06.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot21.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot24.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot05.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ot02.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot23.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot08.ps}}\\{\includegraphics [ height=0.3269 in , width=0.3269 in ] { ot07.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot03.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ot08.ps } } & { \includegraphics [ height=0.3269 in , width=0.3269 in ] { ut00.ps}}\end{array}\ ] ] a * connection point * of an oriented tile is defined as the midpoint of a tile edge which is either the beginning or ending point of an oriented curve drawn on the tile . we define the * sign * of a connection point as * minus * @xmath288 or * plus * @xmath289 accordingly as it is either the beginning point or the ending point of oriented tile curve .
we say that two tiles in an oriented mosaic are * contiguous * if they lie immediately next to each other in either the same row or the same column .
an oriented tile within a oriented mosaic is said to be * suitably connected * if each its connection points touches a connection point of opposite sign of a contiguous tile .
an * oriented knot * @xmath19-*mosaic * is an oriented mosaic in which all tile connection points are suitably connected .
we also let @xmath25 denote the subset of @xmath24 of all oriented @xmath19-mosaic knots . | in this paper , we give a precise and workable definition of a * quantum knot system * , the states of which are called * quantum knots*. this definition can be viewed as a blueprint for the construction of an actual physical quantum system .
moreover , this definition of a quantum knot system is intended to represent the quantum embodiment " of a closed knotted physical piece of rope .
a quantum knot , as a state of this system , represents the state of such a knotted closed piece of rope , i.e. , the particular spatial configuration of the knot tied in the rope .
associated with a quantum knot system is a group of unitary transformations , called the * ambient group * , which represents all possible ways of moving the rope around ( without cutting the rope , and without letting the rope pass through itself . )
of course , unlike a classical closed piece of rope , a quantum knot can exhibit non - classical behavior , such as quantum superposition and quantum entanglement .
this raises some interesting and puzzling questions about the relation between topological and quantum entanglement .
the * knot type * of a quantum knot is simply the orbit of the quantum knot under the action of the ambient group .
we investigate quantum observables which are invariants of quantum knot type .
we also study the hamiltonians associated with the generators of the ambient group , and briefly look at the quantum tunneling of overcrossings into undercrossings .
a basic building block in this paper is a * mosaic system * which is a formal ( rewriting ) system of symbol strings .
we conjecture that this formal system fully captures in an axiomatic way all of the properties of tame knot theory . |
relativistic jets are known to be a ubiquitous feature of environments where accretion onto a compact object leads to outflowing material : they are observed directly in galactic sources , e.g. , @xcite , and in active galaxies , e.g. , @xcite ; and their presence is inferred for gamma - ray bursts @xcite .
while extraordinarily high temporal resolution of activity in galactic sources has been achieved using observatories such as rxte , e.g. , @xcite , the parsec - scale jets of agns remain crucial for the exploration of evolving , spatially - resolved structures @xcite .
furthermore , linear polarization observations , e.g. , @xcite , of such sources provide valuable information about flow magnetic field structures , and thus indirectly about the flow dynamics .
the interpretation of these data is complicated , because allowance must be made for line - of - sight integration through a partially opaque source and relativistic effects ; furthermore , it is not yet clear whether magnetic relaxation , shear or oblique shocks are primarily responsible for the observed magnetic field order , and observational signatures capable of distinguishing these processes are yet to be thoroughly quantified @xcite .
the low ( few percent ) degree of polarization exhibited by compact extragalactic radio sources when in a quiescent state , and the modest degree of integrated polarization ( rarely more than ten percent ) during outbursts , has been widely interpreted as due to ` root - n ' depolarization in a synchrotron source with many randomly oriented magnetic cells within the telescope beam .
in particular , @xcite , working with data between 1.4 and 90 ghz , at the upper end of which range opacity and faraday effects should be negligible , were able to demonstrate using monte carlo simulation that observed ` rotator events ' arise naturally as a consequence of a random walk as turbulent field is advected across observable part of the flow channel . in developments of this picture ,
@xcite , and @xcite , and @xcite have successfully modeled the temporal , spatial , and spectral attributes of outbursts in a number of individual sources , with a scenario in which the shock compression of a flow provides an effective order to the otherwise random magnetic field , and thus increases not only the total and polarized fluxes , but also the percentage polarization . for the best - studied source , bl lac
, the model is particularly convincing , because the same flow speed , orientation and compression are needed to independently reproduce the outburst strength , temporal profile , and percentage polarization .
while it has proven more difficult to model other outbursts [ even more recent outbursts in the same source @xcite ; it appears that the earliest work in this field was fortuitously modeling transverse shock events , while many perhaps the majority of events involve oblique shocks @xcite ] , accounting for the shock orientation has yielded promising results : as shown by @xcite , even highly oblique shock structures may yield the observed level of polarized emission due to shock compression , when allowance is made for relativistic aberration .
this picture has been strengthened by the work of @xcite , who argued that the absence of large circular polarization severely limits the distribution of low energy particles ; thus even if the jet is an electron - ion rather than pair plasma , there is a severe limit to the ability of faraday effects to produce low levels of linear polarization , which is more likely to result from ` root - n ' depolarization .
more recent work allows for some of the magnetic energy to be in an ordered component of magnetic field @xcite , but still requires a significant random component .
the first results from the mojave project @xcite find enhanced magnetic field ordering at flow bends , where shear would be expected to be most influential in modifying an otherwise disordered magnetic field . and ,
unless these sources are substantially beamed , chandra and asca x - ray observations of kiloparsec scale jet - knots imply magnetic fields as much as four orders of magnitude in energy density below equipartition value @xcite .
this suggests a dynamically unimportant magnetic field , subject to the dynamical influence of entrainment and the onset of turbulence .
it would be surprising if the character of the flow on a scale of 10 - 100s parsecs is radically different .
contrary to recent claims @xcite made in support of evidence for ordered parsec - scale jet fields , a continuous sequence of shocks is not needed to yield non - negligible interknot polarization .
indeed , hydrodynamic simulations , with the most minimal disturbance to the jet , reveal a complex web of weak oblique structures that can impose a degree of order consistent with the observed polarization : little compression is required in order to put a fraction of the field energy in an effectively ordered form , and a significant fraction of the field energy may be in a turbulent component , while still admitting significant net polarization .
the macroscopic propagating knots are then distinct features imposed on this underlying pattern , and their presence implies a spectrum of disturbances to the flow that will simultaneously generate a degree of ` background ' order .
thus , while it is somewhat counterintuitive that in a flow subject to shear and repeated shocking , magnetic field structure would remain highly disordered three orders of magnitude ( or more ) in length scale beyond the ` central engine ' , multiple lines of evidence , accumulating since the early 1980s , point to the fact that parsec - scale jet fields are highly disordered , and that the discrete structures observed , form in such flows . this observational constraint has motivated the methodology applied in the work discussed below , which assumes that the flow dynamics is not dominated by the magnetic field , and that the magnetic field can therefore be evolved independently of the hydrodynamics , in a ` post - processing ' step , once the evolution of the hydrodynamic state variables has been established . in principle , three dimensional relativistic magnetohydrodynamic ( rmhd ) simulations of jets can make a major contribution to our understanding of their evolution and propagation , and provide a basis for interpretation of the data .
however , rigorous quantification of the emergent radiation in all stokes parameters requires not only a detailed knowledge of the mhd variables , but also of the radiating particle distribution ; the only simulations @xcite to follow the nonthermal particles are not relativistic .
furthermore , while multidimensional rmhd simulations are being performed , the emphasis is on understanding jet production by ergospheres and disks , and considers only ordered , not turbulent magnetic field @xcite .
apart from the technical challenges of numerical rmhd , in particular simultaneously preserving the divergence - free constraint on the field and conserving its flux , there are two purely astrophysical issues , which are particularly problematic for turbulent magnetic fields .
first , interesting structures ( e.g. , oblique shocks ) may form in the simulated flow , but the response of a turbulent magnetic field to the formation of such structures is difficult to judge using conventional numerical rmhd methods , because an initially turbulent field injected with jet material may have been subjected to other compressions and shear ( providing a significant degree of order , which in a real flow might have been offset by entrainment or the spontaneous generation of turbulence in a high reynolds number medium ) , before the structure of interest forms .
models for this may certainly be constructed , but the details of the turbulence are poorly known , and deconvolving the effects of a history of compressions and shear is very difficult .
second , unknown microphysics will determine the resistivity of the flow , and thus the rate of field line reconnection and relaxation . over or under - estimating the resistivity for a numerical simulation will lead to potentially spurious field morphology far from the point of injection , also providing a significant degree of order prior to formation of interesting structures .
the point of this study is to investigate the polarization structures that result from the formation of particular flow structures ( e.g. , oblique shocks ) , in the context of an initially random magnetic field . to avoid the aforementioned problems ,
epochs have been identified in the jet simulations , that span the formation and propagation of distinct structural features ( such as oblique shocks ) , made spatial and temporal cuts to isolate such evolving features , and used the corresponding velocity information to passively evolve initially turbulent field distributions .
this allows us to model the magnetic structures that result from the response of an initially turbulent field to the development of local hydrodynamic features the scenario strongly supported by observations , as discussed above .
the results presented below were obtained using the fully 3d , relativistic , hydrodynamic code for which the numerical method ( rhlle ) , tests , and first results were discussed in detail by @xcite .
the only two modifications for the current application are as follows .
first , the code has been adapted to run on a linux cluster using cactuscode ( http://www.cactuscode.org/ ) ; this has required minimal code modification ( the addition of _ includes _ and _ declarations _ to the fortran 90 source , the declaration of global variables as the appropriate cactus types , and the construction of the appropriate _ interface _ , _ parameter _ and _ schedule _ units ) .
shock - tube tests have been used to verify the integrity of the solver in this new environment .
the run described here was performed on @xmath0 processors of a cluster .
second , in anticipation of modeling jets propagating into cluster atmospheres , ` pseudo - gravity ' terms have been added to the solver , in order to maintain an initial pressure gradient in the ambient medium .
i assume an inviscid and compressible gas , and an ideal equation of state with constant adiabatic index , @xmath1 , and evolve mass density @xmath2 , the three components of the momentum density @xmath3 , @xmath4 and @xmath5 , and the total energy density @xmath6 relative to the laboratory frame . defining the vector ( in terms of its transpose for compactness ) @xmath7 and the three flux vectors @xmath8 @xmath9 @xmath10 where the three components of velocity are @xmath11 , @xmath12 and @xmath13 , the conservative form of the relativistic euler equation is @xmath14 ( the pressure , @xmath15 , is given by the ideal gas equation of state in terms of the internal energy , @xmath16 , and mass density , @xmath17 , by @xmath18 and @xmath19 ) to model a gravitational field that maintains an arbitrarily imposed pressure gradient , i add a source term of form @xmath20 to the @xmath5 update , and a source term of form @xmath21 where @xmath22 , to the @xmath6 update .
the effectiveness of this approach was easily tested , by noting the absence of evolution of an initial pressure gradient , in the absence of jet inflow , for a number of computational cycles comparable to that used in the final simulation .
the simulation discussed here is a low resolution trial set up to explore the energization of cooling cores by agn jets , and spans 4480 computational cycles on a grid of extent @xmath23 cells ; the lateral extent of the domain is @xmath24 jet - radii , and the longitudinal extent is @xmath25 jet - radii . both jet and ambient medium are characterized by an adiabatic index @xmath26 , the inflowing jet has ( modest ) lorentz factor , @xmath27 and relativistic mach number @xmath28 , and an opening angle of @xmath29 .
a precessional perturbation of lateral velocity @xmath30 and frequency @xmath31 ( in units set by inflow speed and jet radius ) is applied to stimulate the development of normal modes of the cyclindrical flow .
the ambient medium has a density at the inflow plane @xmath32 that of the jet , and declines with ambient pressure , corresponding to an isothermal atmosphere .
pressure in the ambient medium is initially equal to that of the inflowing jet material , and declines along the jet direction as @xmath33 where @xmath34 is the jet radius , @xmath35 is a scale height , taken here to be @xmath36 , and the power law index was taken to be @xmath37 ( eilek , private communication ) .
figure [ fig1 ] shows the formation and evolution of an oblique shock in the body of the jet , from computational cycle 17520 to 22000 .
the physical time spanned by this stage in the flow s evolution is @xmath38 jet - diameter light crossing times , and the pressure and density jumps along a normal to the shock plane ( evident in the figure ) at its central point are respectively @xmath39 and @xmath40 . following @xcite , it is assumed that the radiating particle density varies as thermal pressure , and assuming a jump in the magnetic field corresponding to the density jump , the shocked flow would have emissivity enhanced by @xmath41 ( assuming an optically thin spectral index @xmath42 ( @xmath43 ) , and ignoring small relativistic effects ) , making this a good candidate event to model a parsec - scale jet component .
( this particular compression feature moves marginally subluminally
@xmath44 but its speed has no impact on the results and conclusions presented below . ) it is , therefore , interesting to explore how a magnetic field , that is random at the start of the period studied , would respond to the formation and evolution of this structure .
the rectangles in figure [ fig1 ] outline the region selected for exploration of magnetic field evolution . as the hydrodynamic data are extracted for use in a post - processing step , and it is necessary to retain sufficient time resolution , to keep the computation tractable , the velocity data are averaged over cubes of size @xmath45 cells , and saved @xmath46 time slices , each of size @xmath47 cells .
an initially random magnetic field was generated by selecting random phases , and random amplitudes from a rayleigh distribution , for the fourier transform of the vector potential , @xmath48 .
the fourier transform of the magnetic field is then @xmath49 and an inverse fourier transform of this yields a magnetic field that is guaranteed to satisfy the divergence - free constraint .
( see , e.g. , @xcite for a detailed exposition of this approach . ) by choosing a gaussian form for the variance , @xmath50 in the rayleigh distribution , the correlation scale can be controlled , and it was chosen to be @xmath51 computational cells
10% of the computational volume height ; ` root - n ' depolarization will then lead to a net polarization for the total emission from the volume of order a few percent , typical of the quiescent state of radio - loud agn @xcite . over the @xmath52 light crossing times covered by the evolution followed here ,
magnetic field is advected primarily along the jet axis
i.e. , in the @xmath53 direction .
thus , to provide random field to be advected into the ` active ' computational volume from below , a @xmath54 field distribution ( cut from a @xmath55 data cube to ensure that asymmetry did not induce power preferentially along any one axis ) was generated ; velocity data from the hydrodynamic simulation were associated with the uppermost @xmath56 planes of cells . at each time step , the lowest-@xmath53 velocity data were copied into what is effectively an extended boundary of @xmath57 planes of cells below the inflow plane of the hydrodynamic simulation ; velocity and magnetic field components were copied into a single - cell - wide ghostzone along each of the other faces of the volume .
the magnetic field is advected according to the induction equation , @xmath58 where @xmath59 is the resistivity ( the dissipative term being added for computational expediency see below by analogy with the nonrelativistic induction equation ) , and the field @xmath60 relates to the spatial components of the magnetic four vector , @xmath61 , through @xmath62 for lorentz factor @xmath63 for numerical purposes the induction equation is recast in integral form : @xmath64 discretization then yields algebraic equations for the updates to the average of the magnetic field over a surface , proportional to the discretization scales as @xmath65 the velocities are regarded as stored at cell centers , and the magnetic field components on cell faces the surface for integration ( the @xmath66-component on the higher @xmath66 face of a voxel , etc . ) .
the line integral thus becomes the sum of terms around face edges , and the velocity and magnetic field component values at the centers of these edges are derived by interpolation as needed .
upwinding is not relevant here , as the velocity data are already known ; fluxes are not computed from the characteristic speeds .
( see , e.g. , @xcite for a detailed exposition of this approach . )
this technique preserves the divergence - free constraint to machine accuracy , but unfortunately does not conserve flux .
it has been established by advecting a random field with a _ uniform _ velocity distribution , that by adding a small dissipation , @xmath67 the global magnetic energy density does not grow , while the field remains random ( as judged by the variance of the components ) to within 15% at the mid - time in evolution , when structure of interest develops .
figure [ fig2 ] shows the fluid frame magnetic field components in a @xmath68 plane ( the same midplane as that shown in figure [ fig1 ] ) at @xmath69 epochs during evolution by the pre - computed velocity field . in each panel
the length of the lines is a measure of the average magnetic field in a @xmath70 cube centered on the cut plane , and their maximum length is the same in each panel , so that display of the evolution of field amplitude is suppressed to emphasize changes in the field direction .
the orientation of the lines shows the overall sense of the field within each local @xmath70 subvolume ; this is computed from @xmath71 where @xmath72 and @xmath73 the angles @xmath74 being the local field orientation to the @xmath75-axis .
this figure does not therefore show the degree of _ order _ of the magnetic field ( a field that is almost random locally will , in general , display some net sense , and can be of arbitrary strength ) , which will determine the degree of polarization of emergent radiation .
the degree of polarization measured by an observer is not a local property of the field ; within each @xmath70 subvolume , the adopted correlation imposes a degree of order , and the polarization of the emergent radiation will depend upon integration along some chosen line - of - sight through a number of planes such as shown here .
a measure of the field order has , however , been computed by looking at the rms values of the field components parallel and perpendicular to the preferred sense of the field in each @xmath70 subvolume , and it is found that the field order is a complex function of position .
for example , shear imposes a significant degree of order on the field even in the outlying regions of much lower magnetic field strength , implying that a measure of field order alone can not be used to estimate the regions from which a high flux of polarized radiation results , since the intrinsic emissivity of these weaker field regions will be low . our measure of field order does confirm that the field is ordered in the downstream flow of the oblique shock , by an amount that would be expected from the density jump , which is the main feature of interest here .
the appearance of these structures will depend on their relative doppler boosting ; however , the interest here is to model knots that display measurable motion , in which case we will not be seeing the flow far from its critical cone and statistically be more likely to be outside of that cone
so boosting may be somewhat limited .
thus , although a full radiative transfer calculation must be performed to understand in detail how these simulations relate to vlbp maps , the extended regions of well - defined field structure seen in the later panels of figure [ fig2 ] can be expected to correspond to knots of high intensity and polarization .
the transfer of polarized radiation through a diffuse plasma , allowing for emission , absorption , the birefringence effects of faraday rotation and mode conversion ( which can produce modest levels of circular polarization ) , and relativistic aberration and boost , has been described in detail by a number of authors .
i follow the analysis of @xcite , which is compactly summarized by @xcite , and which has been used previously in the study of parsec - scale jets ; see , for example , @xcite and @xcite .
an algorithm for radiative transfer has been constructed as follows : the observer may be placed at arbitrary polar angles ( @xmath76,@xmath77 ) , defined in the conventional sense with respect to the cartesian system used to describe the hydrodynamics ; thus an observer viewing the @xmath68 plane of figure [ fig2 ] will be at polar coordinate ( @xmath78,@xmath79 ) .
an array of lines - of - sight is established , using a preset density of lines along the longest axis of the projection of the computational volume on the plane of the sky , with a commensurate number orthogonal to that , to ensure equal resolution in the two directions . for the results presented below
, a resolution of @xmath80 pixels along the longest axis was used . for each line - of - sight
, the algorithm finds the most distant cell , and radiative transfer is performed , cell to cell , until the ` near side ' of the volume is exited . within each cell ,
an aberrated magnetic field direction is computed from the rest frame field , velocity , and observer location , and is used with transfer coefficients modified by the relevant doppler boost , as described by @xcite .
a full radiative transfer requires some knowledge of the particle distribution function : the low energy cutoff lorentz factor , the lorentz factor of particles radiating at the observer frequency , and the spectral slope of the distribution ; these are arbitrarily fixed at @xmath81 , @xmath82 , and @xmath83 ( see 2 ) . in principle
these might change from place to place , but as no model of this is available , these are regarded as fixed parameters , and when exploring optically and faraday thick flows , the predicted behavior as illustrative only .
the transfer coefficients are computed by assuming that the radiating particle density is proportional to the thermal pressure of the hydrodynamic simulation ; however , one can not a priori know the optical depth through some arbitrary line - of - sight .
the magnetic field strength and hydrodynamic pressure are scaled by their rms value and mean value respectively ( within the computational volume at the initial time ) , and the optical depth is adjusted by scaling to yield some target optical depth for a line - of - sight with the rms field and mean pressure through the longest extent of the volume .
for the results presented below , values were chosen to yield an optically thin flow .
observations strongly suggest that the particle distribution and magnetic field , jointly responsible for the synchrotron radiation , are confined to the inflowing jet , and do not permeate the ambient medium . to model
this , a filter is applied , so that no emission arises from cells with velocity @xmath84 ( @xmath85 ) .
the intensity pattern on the plane of the sky has been convolved with a @xmath86-pixel gaussian beam corresponding to four beam widths across the shock structure of interest , and typical of the best vlbi resolutions now being attained for parsec scale jets , where in some sources it is now possible to detect limb brightening , and rotation measure gradients , across the flow and all four stokes parameters ( @xmath87 , @xmath88 , @xmath89 and @xmath90 ) stored in fits files , for subsequent display using dan homan s pgperl package fitsplot ( http://personal.denison.edu/@xmath91homand/#useful ) .
in principle , the radiative transfer should use retarded times ; that has not been done in the results shown below , but this will have minimal impact on the validity of the results , as individual features move rather slowly .
rigorous , retarded time calculations , will be performed in a subsequent study .
figure [ fig3 ] shows , from upper - left to lower - right , ` fake ' maps of polarized intensity , corresponding to the @xmath92th through @xmath93th epochs shown in figure [ fig2 ] , computed for an observer orientation of ( @xmath78,@xmath94 ) , with other parameters as described above ; this corresponds approximately to the view through the midplane of the computational volume shown in figure [ fig2 ] .
care must be taken in comparing figure [ fig2 ] with figure [ fig3 ] : the former is a slice through a computational volume , whereas the latter shows an integration along lines of sight ; the ` fake ' maps are made for a viewer not precisely along a coordinate direction , slightly changing the aspect ratio of the panels ; in general , retarded time calculations are needed , and there is no simple connection between the sequence of frames in the two figures . as retarded time calculations have not been employed here , the corresponding frames in the two sequences have been labeled accordingly .
the scaling has been chosen so that the maps have comparable dynamic range .
corresponding maps of total intensity have the same general morphology , and i choose to discuss the @xmath95 map because the components are slightly more distinct .
corresponding polarization maps for a flow exhibiting significant opacity have quite different structure , but in that case the percentage polarization is negligible ; the structure results from lines - of - sight with , by chance , some residual polarization , and the pattern bears no simple relation to the underlying mhd .
( this may provide a useful model for the exploration of low polarization cores , but here we are interested in exploring the evolving structures evident on maps in regions of low optical depth . ) maps made with different observer orientation can look radically different , because a few spatially distinct peaks of emission within the 3d volume are seen in quite a different projection .
that varied and complex behavior can result from simple flows is one of the messages of this work ; i opt to present the map sequence that can be most easily related to the underlying flow features .
the localized peak in intensity in the first panel of figure [ fig3 ] is offset from the regions of most - ordered field at that epoch associated with the shock above , and shear at right seen in the central panel of figure [ fig2 ] , because that line - of - sight intersects regions of ordered field and high pressure not revealed by the simple center - plane cut of figure [ fig2 ] .
careful exploration of the flow structure shows that what appears to be a simple , oblique shock , does indeed comprise thin , approximately planar features oblique to the flow direction , but that there are two nested discontinuities , which curve slightly to follow the local jet boundary , extending for about one - third of the jet circumference , and converging at their base on the flow edge .
the devil is in the detail , and in this case , the devil is in the knot of emission . even the simplest of vlbi structures may be hiding a complex flow pattern .
subsequent evolution , shown in the upper - right , lower - left , and lower - right panels of figure [ fig3 ] , leads to a weak feature downstream of the initial bright knot , reflecting the increasing influence of emission from both the downstream flow of the shock ` complex ' , and the shear at the jet edge above this , and , finally , another distinct knot associated with the region of strong shear
evident in the upper left of the last panels of figure [ fig2 ] that arises as the jet direction changes .
that effect propagates in the flow s upstream direction , elongating this knot .
this simulation is applicable to the interpretation of sources such as bl lac objects , where components are short - lived , and where , within a few beam widths which barely start to resolve these components , leading to maps with structure smoother than one might anticipate
complex dynamical behavior can be seen , with variable apparent speed and/or nonlinear trajectories @xcite .
no claim is made that this simulation explains the behavior of any particular vlbi source , but it is interesting to see that an asymmetric shear layer is easily produced , which might have relevance for the interpretation of vlbi features suggestive of the presence of a flow boundary @xcite , and it clearly warns against the naive interpretation of observations that reveal knots that seemingly abruptly change direction and decelerate .
fortunately , the sense of the polarized emission would provide some help in disentangling the behavior displayed here , if these were actual data : the electric vector of the knot of emission in the first panel of figure [ fig3 ] is highly inclined to the average flow direction , being orthogonal to the oblique structure shown in the corresponding schlieren render of figure [ fig1 ] , whereas the electric vector for the emission evident in the last panel is orthogonal to the flow boundary .
this gives strong clues as to the magnetic field morphology ( at least for an optically thin , weakly relativistic , flow ) and thus to the underlying hydrodynamics .
the structure and dynamical significance of extragalactic jet magnetic fields has received much attention over the last years , with considerable emphasis on the case of fields with large scale order , and a tendency to dismiss the possibility of a significant random component .
lessons from studying the magnetic fields of other astrophysical bodies suggest that such an ` either - or ' view is misplaced , and that it is most likely that there is magnetic energy on a range of scales : both ordered and random components may play a role in determining the emission properties of many sources , or one may dominate under certain conditions .
strong evidence exists for the prescence of a random component , as discussed in the introduction , and we have explored some of the observational consequences of that picture .
the simulation discussed above was chosen as a reasonable data set with which to develop the technology for field evolution and radiative transfer .
results should be regarded as illustrative , and not over - interpreted .
the flow is of low lorentz factor and low mach number , the shock that forms is weak , and the resultant field compression produces very modest levels of polarization .
however , it does clearly expose the consequences of having spatially distinct , varying centers of activity , yielding a complex behavior , the character of which is very sensitive to the observer s orientation
. the hydrodynamic structure will be similar to that seen here , for lorentz factors up to about 5 @xcite , while faster flows will in general be rather featureless . at speeds higher than that addressed here
, a major role will be played by relative doppler boosting , which can be expected to lead to a detailed pattern of intensity that is very dependent on observer orientation ; this will probably lead to an even more complex interplay between regions of compression and shear , but only detailed modeling for a range of parameters can quantify this .
examination of figure [ fig2 ] shows that a region of ordered , obliquely oriented , magnetic field arises in the downstream flow of the oblique shock which is well - defined by the mid - time in evolution .
this structure propagates subluminally along the flow direction , weakening as it evolves , and , most strikingly , merges with the more oblique , wedge - shaped region dominated by shear , most obvious above center - right in the last panel .
radiative transfer reveals ( as shown in figure [ fig3 ] ) an evolving pattern of emission ( both total and polarized ) that has somewhat different morphology from that which might be inferred from the sampling of the magnetic field structure , because lines - of - sight probe flow structure hidden by simple data slicing , and incorporate the effects of radiating particle density and line - of - sight depth .
however , whether based on mhd structures alone , or accounting for a full radiative transfer , the conclusion is that the interrelated evolution of internal jet structures and shear layers can lead to apparently complex behavior in which the changing relative importance of these facets of the flow mimic the evolution of a single component with complex dynamical behavior .
it is important to realize that the domains dominated by shear at the jet edge are not static , and that these regions of shear - influenced field evolve as the jet shifts laterally .
if the flow cross - section were barely resolved by observation , it would appear that a region of more - nearly transverse field evolved into a region of longitudinal field ; higher resolution might suggest that a component within the flow propagates along a somewhat curved trajectory , with a rotation of the plane of polarized emission towards the longitudinal sense .
simulations such as presented here caution against the naive interpretation of multi - epoch vlbi / p maps in particular showing that non - linear trajectories and a rotation of the plane of polarization as a component evolves may be manifestations of evolution _ within _ a channel broader than that suggested by the data and the simulations provide a framework within which to interpret them .
the global flow pattern may be more complicated than the trajectory of one knot suggests . the simulations also serve to clarify that evolving , complex , internal structure , that may be readily identified on a time series of vlbp maps , does not necessarily result from the interaction of the jet with ambient density or pressure structures , but , rather , may form and evolve due to the response of the jet to infinitesimal perturbations . defining the dynamics of such flows through vlbi observations clearly demands a sequence of maps with high time resolution , and of sufficient duration to probe the global structure of the flow .
isolated epochs may reveal transient peculiarities , not representative of the overall flow morphology .
this work was supported in part by nsf grant ast 0205105 .
the work was motivated in part by results from the umrao monitoring program supported in part by a series of grants from the nsf and by funding from the department of astronomy , and the author thanks hugh and margo aller for illuminating discussions on the observational implications of the study .
comer duncan and mark miller have made major contributions to the hydrodynamic simulations .
a referee made numerous comments that helped to significantly improve the manuscript .
aller , h. d. , aller , m. f. , & hughes , p. a. 2002 , proceedings of the 6th european vlbi network symposium , ed .
e. ros , r. w. porcas , & j. a. zensus , ( bonn , germany : mpifr ) , 111 aller , h. d. , hughes , p. a. , freedman , i. , & aller , m. f. 1999 , bl lac phenomenon , ed . l. o. takalo , & a. sillanp ( asp conference series 159 ) , 45 attridge , j. m. , roberts , d. h. , & wardle , j. f. c. 1999 , , 518 , l87 beckert , t. , & falcke , h. 2002 , , 388 , 1106 belloni , t. , et al . 2000 , , 355 , 271 dar , a. 1998 , , 500 , l93 denn , g. r. , mutel , r. l. , & marscher , a. p. 2000 , , 129 , 61 duncan , g. c. , & hughes , p. a. 1994 , , 436 , l119 fender , r. , et al .
2004 , nature , 427 , 422 gmez , j. l. , alberdi , a. , & marcaide , j. m. 1993 , , 274 , 55 gmez , j. l. , alberdi , a. , & marcaide , j. m. 1994 , , 284 , 51 gomez , j .- l . , alberdi , a. , marcaide , j. m. , marscher , a. p. , & travis , j. p. 1994
, , 292 , 33 homan , d. c. , & wardle , j. f. c. 2004 , , 602 , l13 hughes , p. a. , aller , h. d. , & aller , m. f. 1989a , , 341 , 54 hughes , p. a. , aller , h. d. , & aller , m. f. 1989b , , 341 , 68 hughes , p. a. , aller , h. d. , & aller , m. f. 1991 , , 374 , 57 hughes , p. a. , miller , m. a. , & duncan , g. c. 2002 , , 572 , 713 jones , t. w. 1988 , , 332 , 678 jones , t. w. , & odell , s. l. 1977 , , 214 , 522 jones , t. w. , et al . 1985 , , 290 , 627 kataoka , j. , & stawarz , l. 2004 , astro - ph/0411042 kellermann , k. i. , et al .
2004 , astro - ph/0403320 koide , s. , meier , d. l. , shibata , k. , & kudoh , t. 2000 , , 536 , 668 komissarov , s. s. 1999a , , 303 , 343 komissarov , s. s. 1999b , , 308 , 1069 lister , m. l. 2003 , astro - ph/0309413 lyutikov , m. , pariev , v. i. , & gabuzda , d. c. 2004 , astro - ph/0406144 mioduszewski , a. j. , hughes , p. a. , & duncan , g. c. 1997 , , 476 , 649 ojha , r. , et al .
2004 , , 150 , 187 rosen , a. , hughes , p. a. , hardee , p. e. , & duncan , g. c. 1999 , , 516 , 729 tregillis , i. l. , jones , t. w. , & ryu , d. 2001 , , 557 , 475 tribble , p. c. 1991 , , 253 , 147 van putten , m. h. p. m. 1996 , , 467 , l57 zensus , j. a. 1997 , , 35 , 607 | evidence that the magnetic fields of extragalactic jets have a significant fraction of their energy in a random component is briefly summarized , and a detailed model of evolving , jet polarization structures is constructed , based on this picture .
the evolving magnetic field structure of an oblique shock complex that forms in a relativistic jet simulation is explored by using velocity data from the hydrodynamical simulation to advect an initially random magnetic field distribution .
radiative transfer calculations reveal that emission from a propagating region of magnetic field , ` ordered ' by the shock , and lying approximately transverse to the flow direction , merges with that from an evolving sheared region at the flow periphery .
if such a flow were barely resolved , observation would suggest evolution from a somewhat oblique , to a more longitudinal , magnetic field structure with respect to the flow axis , while higher resolution observations would infer a component following a non - linear trajectory , and with a magnetic field orientation that rotates during evolution .
this result highlights the ambiguity in interpreting vlbp data , and illustrates the importance of simulations in providing a framework for proper interpretation of such data . |
the study of superfluidity in infinitely extended nuclear systems has a long history [ 182 ] , predating the 1967 discovery of pulsars @xcite , which were soon identified as rapidly rotating magnetic neutron stars @xcite .
interest in nucleonic pairing has intensified in recent years , owing primarily to experimental developments on two different fronts . in the field of astrophysics ,
a series of @xmath4-ray satellites ( including einstein , exosat , rosat , and asca ) has brought a flow of data on thermal emission from neutron stars , comprising both upper limits and actual flux measurements .
the recent launching of the chandra @xmath4-ray observatory provides further impetus for more incisive theoretical investigations . realistic _
ab initio _ prediction of the microscopic physics of nucleonic superfluid components in the interiors of neutron stars is crucial to a quantitative understanding of neutrino cooling mechanisms [ 8595 ] that operate immediately after their birth in supernova events , as well as the magnetic properties , vortex structure , rotational dynamics , and pulse timing irregularities [ 9699 ] of these superdense stellar objects .
in particular , when nucleonic species enter a superfluid state in one or another region of the star , suppression factors of the form @xmath5 are introduced into the expression for the emissivity , @xmath6 being an appropriate average measure of the energy gap at the fermi surface . on the terrestrial front ,
the expanding capabilities of radioactive - beam and heavy - ion facilities have stimulated a concerted exploration of exotic nuclei and nuclei far from stability , with special focus on neutron - rich species @xcite .
pairing plays a prominent role in modeling the structure and behavior of these new nuclei . in this article
we will be principally concerned with nucleonic pairing in the astrophysical setting .
to a first approximation , a neutron star is described as a neutral system of nucleons ( and possibly heavier baryons ) and electrons ( and possibly muons ) in beta equilibrium at zero temperature , with a central density several times the saturation density @xmath7 of symmetrical nuclear matter @xcite .
the gross structure of the star ( mass , radius , pressure and density profiles ) is determined by the tolman - oppenheimer - volkov general relativistic equation of hydrostatic equilibrium , consistently with the continuity equation and the equation of state ( which embodies the microscopic physics of the system ) .
the star contains ( i ) an _ outer crust _ made up of bare nuclei arranged in a lattice interpenetrated by relativistic electrons , ( ii ) an _ inner crust _ where a similar coulomb lattice of neutron - rich nuclei is embedded in fermi seas of relativistic electrons and neutrons , ( iii ) a _ quantum fluid interior _ of coexisting neutron , proton , and electron fluids , and finally ( iv ) a _ core region _ of uncertain constitution and phase ( but possibly containing hyperons , a pion or kaon condensate and/or quark matter ) .
it is generally accepted @xcite that the neutrons in the background fluid of the inner crust of a neutron star will pair - condense in the @xmath0s@xmath1 state .
qualitatively , this phenomenon can be understood as follows . at the relatively large average particle spacing at the `` low '' densities involved in this region ( @xmath8 ) , the delocalized neutrons experience mainly the attractive component the @xmath0s@xmath1 interaction .
however , the pairing effect is quenched at higher densities , @xmath9 and beyond , due to the strong short - range component of this interaction . by similar reasoning ,
one expects @xmath0s@xmath1 proton pairing to occur in the quantum fluid interior , in a density regime where the proton contaminant ( necessary for charge balance and chemical equilibrium ) reaches a partial density @xmath10 .
the energy dependence of the nucleon - nucleon ( @xmath11 ) phase shifts in different partial waves offers some guidance in judging what nucleonic pair - condensed states are possible or likely in different regions of a neutron star .
a rough correspondence between baryon density and @xmath11 bombardment energies can be established through the fermi momenta assigned to the nucleonic components of neutron - star matter @xcite , which provide a crude measure of the maximum relative momenta attained in the nuclear medium . in ( pure ) neutron matter ,
only @xmath12 partial waves are allowed by the pauli principle . moreover , one need only consider partial waves with @xmath13 in the range of baryon density optimistically , @xmath14 where a nucleonic model of neutron - star material is tenable .
the @xmath0s@xmath1 phase shift is positive at low energy ( indicating an attractive in - medium force ) but turns negative ( repulsive ) at around 250 mev lab energy .
thus , unless the in - medium pairing force is dramatically different from its vacuum counterpart , the situation already suggested above should prevail : s - wave pairs should form at low densities but should be inhibited from forming when the density approaches that of ordinary nuclear matter .
the next lowest @xmath12 partial waves are the three triplet p waves @xmath2p@xmath15 , with @xmath16 , which are actually coupled by the tensor force to the triplet f waves with the same total angular momentum @xmath17 . of these channels ,
the first two are not good candidates for pairing . in the @xmath2p@xmath1 state ,
the phase shift is only mildly attractive at low energy , turning repulsive at a lab energy of 200 mev , while the @xmath2p@xmath18 phase shift is repulsive at all energies . on the other hand ,
the @xmath2p@xmath3 phase shift is negative at all energies , and indeed is the most attractive @xmath12 phase shift at energies above about 160 mev .
this exceptional attractive strength in the @xmath2p@xmath3 state is due mainly to the spin - orbit interaction and in part to the coupling to @xmath2f@xmath3 wave via the tensor component of the @xmath11 force .
a substantial pairing effect in the @xmath2p@xmath3@xmath2f@xmath3 channel may be expected at densities somewhat in excess of @xmath7 , again assuming that the relevant in - vacuum interaction is not greatly altered within the medium .
the remaining @xmath19 partial waves with @xmath13 are both singlets : @xmath0d@xmath3 and @xmath0g@xmath20 . however , the @xmath0d@xmath3 phase shift , though positive over the energy domain of interest , is clearly dominated by the @xmath2p@xmath3 phase shift , while the @xmath0g@xmath20 phase shift , again positive , is even less important . therefore , in the conventional picture of neutron - star matter ( see , however , ref .
@xcite ) , singlet and triplet pairing are operationally equivalent to @xmath0s@xmath1 and @xmath2p@xmath3@xmath2f@xmath3 pairing , respectively .
there have been many illuminating studies of the former problem , notably @xcite .
here we shall concentrate on triplet pairing in the pure neutron system , ignoring the relatively minor modifications that arise in beta - stable neutral matter with its dilute admixture of protons .
bcs theory @xcite furnishes a general framework for microscopic treatments of nucleonic pairing . as is the case for singlet pairing @xcite , two rather different problems
must be overcome to realize a quantitative description of triplet pairing within this framework and make realistic predictions for the associated energy gap .
the first problem to be solved is the construction of an accurate in - medium ( or `` effective '' ) particle - particle interaction that provides the pairing force , together with a consistent mass operator that provides the single - particle input for the bcs equation .
the second problem is to solve the bcs gap equation ( actually a system of gap equations when one deals with @xmath21 ) , once the inputs for the pairing matrix and normal - state single - particle energies have been determined .
this first task places great demands on even the most advanced tools of many - body theory . for @xmath0s@xmath1 pairing ,
there exists a `` first generation '' of results on this facet of the problem , obtained by several different approaches ( landau fermi - liquid theory @xcite , the method of correlated basis functions @xcite , the polarization - potential approach @xcite and diagrammatic perturbation theory @xcite ) .
however , little progress has been made on this aspect of @xmath2p@xmath3@xmath2f@xmath3 pairing , and the issue of modifications of the associated input pairing interaction and single - particle energies due to short- and long - range correlations inside the nuclear medium remains open .
we shall not attempt to address this issue here .
rather , we shall be concerned with the task of actually solving the triplet bcs system , assuming the particle - particle interaction is identical with the bare or vacuum @xmath11 interaction , and that the input single - particle energies needed to complete the bcs equation(s ) have the same form as for free particles ( perhaps with an effective mass ) .
this problem also proves to be nontrivial , since we face essentially the same difficulties as in singlet s - wave pairing , and more : 1 .
again , the kernel of the bcs gap equation(s ) has a pole at zero gap amplitude , implying an incipient logarithmic divergence .
2 . again
, the nonlocality of the particle - particle interaction in momentum space implies slow convergence of the integral over momentum modulus @xmath22 in the bcs equation(s ) precluding application of the standard weak - coupling estimates of bcs theory @xcite .
this feature of nucleonic pairing implies that one must integrate out to large values of the momentum variable .
as already intimated , an additional complication of pairing in higher angular momentum states with @xmath21 arises from the ( possible ) angle - dependence of the gap , which can be parametrized in terms of dependences on the quantum numbers @xmath23 of the total and relative orbital angular momenta and on the magnetic quantum number @xmath24 associated with @xmath17 .
one must deal with a set of coupled gap equations for the gap components @xmath25 in an expansion of the ( generally anisotropic ) gap in the spin - angle functions @xmath26 .
the combination of difficulties ( a ) and ( b ) can make numerical solution of the gap equation(s ) a problematic exercise .
to overcome these obstacles , a special procedure called the separation approach has been introduced in ref .
@xcite and applied to @xmath0s@xmath1 pairing . in this approach , the generic pairing matrix @xmath27 is decomposed , identically , into a separable part , plus a remainder whose every element vanishes when either momentum variable lies on the fermi surface .
( as is characteristic of bcs theory , we assume that the particle - particle interaction is independent of frequency . ) applied to @xmath0s@xmath1 pairing , this separation procedure reduces the bcs gap equation to two equivalent coupled equations : a nonsingular , quasilinear integral equation for the shape of the gap function and a nonlinear equation determining its amplitude at the fermi surface .
we hasten to point out that this strategy is by no means limited to the case where the pairing matrix elements are supplied by the bare @xmath11 interaction it will retain its practical and analytical value when a realistic in - medium particle - particle interaction is employed .
the primary aim of this article is to generalize the separation approach to pairing in an arbitrary angular momentum channel and apply it to @xmath2p@xmath3 pairing in neutron matter . since
, according to consideration ( c ) above , we are now confronted in general with a set of coupled gap equations rather than a single one ,
the situation is much more complicated than for @xmath0s@xmath1 pairing .
however , in this case the power of the separation method becomes even more apparent .
it enables a decomposition of the problem into ( i ) the determination of the shape of the gap function in @xmath22-space through an appropriate nonsingular quasilinear integral equation whose kernel is practically independent of the energy gap , and , once this is done , ( ii ) the determination of a set of numerical coefficients specifying the angular dependence of the gap through a system of nonlinear ` algebraic ' equations . in both singlet and triplet cases ,
the effectiveness of the separation scheme is greatly enhanced by the existence of a small parameter , given by the ratio of a suitable measure of the gap amplitude to the fermi energy .
small - parameter expansions work well in the singlet problem , and even better in the triplet case .
corrections to the leading order in a small - parameter expansion of the @xmath2p@xmath3@xmath2f@xmath3 energy gap are down by a factor @xmath28 or @xmath29 .
this property leads to considerable simplifications in the analytical and numerical treatment of the triplet problem , greatly ameliorating complication ( c ) .
the separation approach is not merely a tool for numerical solution of gap equations .
following the lead of refs .
@xcite , we shall demonstrate its very important role in facilitating incisive analysis of the nature of pairing solutions in different angular momentum channels .
in particular , the separation formalism provides a basis for establishing certain universal properties of @xmath2p@xmath3 pairing , properties that are independent of density and independent of the specifics of the pairing interaction itself , whether furnished by the in - vacuum or in - medium particle - particle interaction .
the analytical results obtained here and in refs .
@xcite remain valid outside the context of neutron matter and carry implications for the still more complex problem of pairing in liquid @xmath2he @xcite . in sect .
[ sec : gapeqn ] , we display the bcs gap equations that describe pairing in an arbitrary two - body channel of given total spin @xmath30 and isospin @xmath31 .
[ sec : sepformalism ] extends the separation method of ref .
@xcite to this general problem , with special attention to @xmath2p@xmath3@xmath2f@xmath3 pairing .
[ sec : nocoupling ] is devoted to an analysis of @xmath2p@xmath3 pairing in the absence of tensor coupling to the @xmath2f@xmath3 state ( `` pure '' @xmath2p@xmath3 pairing ) .
we offer a straightforward criterion for determination of the upper critical density at which a solution of the pairing problem ceases to exist in the given two - body channel , in terms of vanishing of a characteristic determinant associated associated with the separation formulation . in sect .
[ sec : aaa ] we introduce the small - parameter expansion of the pairing problem and use it to clarify the nature of the angle - average approximation proposed by baldo et al .
@xcite . in turn
, the latter approximation is adopted to formulate an efficient procedure that allows reasonably accurate numerical solution of the coupled - channel , @xmath2p@xmath3@xmath2f@xmath3 pairing system when the goal is to find the magnitude of the pairing effect and the detailed structure of the pairing solutions is a lesser concern .
we go on to recount how the fundamental character of the diverse solutions of the nonlinear pairing problem may be elucidated within the separation method .
the derivation of universal features of @xmath2p@xmath3 pairing is exemplified and discussed .
results from numerical application of the separation approach to triplet pairing in neutron matter are reported in sect .
[ sec : tripletnumapp ] .
three representative models of the free - space @xmath11 interaction are used to construct pairing matrix elements , the argonne @xmath32 potential @xcite being chosen as an example from the current generation of models yielding high - precision fits of the @xmath11 scattering data @xcite .
the results presented explicitly include the critical upper density for gap closure , the momentum dependence of gap components , and the conventional average measure of the gap amplitude at the fermi surface , variously calculated for `` pure '' and tensor - coupled @xmath2p@xmath3 pairing , with and without use of the angle - average approximation .
these results are compared with previous findings of other authors . in sect .
[ sec : concl ] , we indicate some promising future directions for research on pairing in infinitely extended nucleonic systems .
an appendix reviews the separation method as developed for @xmath0s@xmath1 pairing .
small - parameter expansion is employed to devise an approximation for the gap amplitude that may be taken as a generalization of the bcs weak - coupling formula .
additionally , the technique of small - parameter expansion is extended to the triplet pairing problem .
to describe pairing in an arbitrary two - body state with total spin @xmath30 , angular momentum @xmath17 , and isospin @xmath31 , it is useful to introduce a @xmath33 gap matrix @xmath34 which has the expansion ( cf .
@xcite ) @xmath35 in the spin - angle matrices @xmath36 with @xmath37 .
the total spin and isospin quantum numbers @xmath38 ( @xmath39 for triplet states and neutron matter ) are fixed throughout and generally suppressed . setting @xmath40 $ ]
, the spin - angle matrices ( [ spinm ] ) may be shown to obey the orthogonality condition @xmath41 the particle - particle interaction has the corresponding expansion @xmath42 with @xmath43 in the case of tensor forces .
the coupled set of generalized gap equations for the components @xmath44 then reads @xmath45 with @xmath46 and @xmath47 .
the energy denominator @xmath48^{1/2}$ ] involves the square of the single - particle excitation energy @xmath49 of the normal system , measured relative the fermi energy @xmath50 , and a term @xmath51 giving rise to an energy gap in the single - particle excitation spectrum of the pair - condensed system .
examination of the the explicit formula for the angular factor @xmath52 reveals that a summation over total momentum @xmath53 must be present in ( [ gengap ] ) and ( [ squd ] ) .
( this fact is well known in the theory of superfluid @xmath2he @xcite : the anisotropic abm phase is built of states with total momentum @xmath54 _ and _ 2 . )
the sum over orbital momenta @xmath55 on the l.h.s . of ( [ gengap ] ) appears only when tensor forces are present and is then restricted by @xmath43 and parity conservation . on the other hand ,
the sum over @xmath56 resulting from the angular dependence of the gap contains , in principle , an infinite number of terms .
we identify the energy gap as @xmath57 and note that it is in general angle dependent and in general has nodes .
it is convenient to define an angle - averaged gap function @xmath58 through the positive root of @xmath59 ( the last equality follows from the orthogonality property ( 3 ) . )
analogously to the @xmath0s@xmath1 pairing problem , the quantity ( 7 ) , evaluated at the fermi momentum @xmath60 and denoted @xmath61 furnishes an overall measure of the strength of pairing correction to the ground - state energy in the preferred state . the notation @xmath62 is included here since it has been used earlier in the literature @xcite .
we take this opportunity to caution the reader that the literature is inconsistent and sometimes vague in the choice of energy gap measure .
care must be taken in interpreting statements about the magnitude of `` the gap . ''
moreover , the choice made in normalizing or defining the pairing matrix elements is not uniform .
we have attempted to be precise , and for the most part we follow the conventions established by takatsuka and tamagaki @xcite .
generalizing the separation technique developed in ref .
@xcite for the @xmath0s@xmath1 problem , we again split the pairing matrix elements into a separable part and a remainder @xmath63 that vanishes when either of the momentum variables @xmath22 , @xmath64 lies on the fermi surface .
thus @xmath65 where @xmath66 and @xmath67 .
substitution of this identity into the generic - channel gap equation ( [ gengap ] ) yields @xmath68 where the coefficients @xmath69 ( with @xmath70 are just numbers . as in ref .
@xcite ( see also the appendix ) , we have introduced a typical scale @xmath71 in place of the gap in single - particle spectrum , in the term of ( [ delsf ] ) involving the portion @xmath72 of the pairing matrix that vanishes on the fermi surface .
such terms are quite insensitive to the choice of this scale , since the vanishing of @xmath72 at @xmath73 ensures that the integral over @xmath64 receives its overwhelming contributions some distance from the fermi surface where @xmath74 is negligible compared to @xmath75 . in the present context where the gap functions are angle - dependent , it is important to note that the @xmath71 quantity may be regarded as angle - independent .
numerical calculations show that the final result is practically independent of the choice of @xmath71 and we shall normally take it to be zero . employing the orthogonality condition ( [ ort1 ] )
, one finds that only diagonal terms ( meaning @xmath76 , @xmath77 , @xmath78 ) contribute to the sum on the l.h.s . of eq .
( [ delsf ] ) , which becomes @xmath79 henceforth we shall frequently suppress the fixed index @xmath17 .
the functions @xmath80 superposed on the r.h.s . of ( [ base1 ] ) do not depend on @xmath24 .
accordingly , the gap components @xmath81 can be constructed as linear combinations @xmath82 of @xmath83independent functions @xmath84 multiplied by coefficients @xmath85 determined by eq .
( [ defdf ] ) .
it is helpful to regard the quantities @xmath86 as forming a vector @xmath87 with components indexed by @xmath88 , and similarly the functions @xmath89 as composing a vector @xmath90 .
the equations @xmath91 for the shape factors @xmath92 are readily obtained from eq .
( [ base1 ] ) . from this result
we see that @xmath93 for any @xmath94 , since the potentials @xmath95 vanish for @xmath22 on the fermi surface while @xmath96 for any @xmath97 .
it is important to observe that the equations ( [ chiviaphi ] ) are free from any log singularities and form a coupled system of _ linear _ integral equations .
we further remark that these equations can be solved _ without regard _ to the values of the coefficients @xmath98 given by the set of equations ( [ defdf ] ) with @xmath99 expanded as in ( [ gen1 ] ) .
it is also important to take note of an essential difference in the reformulated equations ( [ base1])([chiviaphi ] ) as compared to the original gap equations ( [ gengap ] ) .
the sum over orbital angular momenta @xmath56 in ( [ gengap ] ) is in principle infinite , while in ( [ base1 ] ) only the states with @xmath100 and @xmath101 can contribute , due to the conservation of the total angular momentum @xmath17 and parity . therefore the number of components of the vectors @xmath102 and @xmath90 does not exceed 3 , since @xmath43 and in eqs .
( [ base1])([chiviaphi ] ) the total angular momentum @xmath17 has a fixed value .
we may infer that , in the case of a @xmath2p@xmath3@xmath2f@xmath3 mixture where @xmath103 , orbital angular momentum quantum numbers beyond 5 contribute only to the numerical factors @xmath85 .
it is instructive to write out eqs .
( [ base1 ] ) and ( [ chiviaphi ] ) in extended form for the @xmath2p@xmath3@xmath2f@xmath3 problem . using a symbolic notation in terms of the operator @xmath104^{1/2}$ ] , the system ( 11 ) reads @xmath105 suppose all the coefficients @xmath106 are zero except @xmath107 ; then the solution @xmath87 of ( [ ds ] ) , a two - component vector , reduces to @xmath108 where the components @xmath109 and @xmath110 of @xmath111 obey the pair of equations @xmath112 the vector @xmath113 is given by a second pair of equations , @xmath114 derived by supposing that @xmath115 is present but all other @xmath116 are zero .
the remaining two pairs of equations for the vector functions @xmath117 and @xmath118 , namely @xmath119 and @xmath120 are constructed in the same fashion .
the general solution of the @xmath2p@xmath3@xmath2f@xmath3 pairing problem is given by eq .
( [ gen1 ] ) with the indices @xmath56 and @xmath94 taking on the values 1 or 3 .
it is worth noting that the same system of equations arises when one deals with @xmath121 pairing in the deuteron channel , except that the indices @xmath56 and @xmath94 then take on the values 0 and 2 .
let us return now to the problem of pairing in an arbitrary two - body channel .
having learned that the shape factors @xmath122 in the representation ( [ gen1 ] ) may be found by solving a set of nonsingular linear integral equations independently of the coefficients @xmath123 , we must next consider the determination of these coefficients , assuming knowledge of the @xmath124 functions . substitution of ( [ gen1 ] ) into ( [ defdf ] ) provides a set of nonlinear ` algebraic ' equations to be solved for the desired set of numbers : @xmath125 in the vicinity of the ( upper ) critical value @xmath126 of the density @xmath127 where the gap value @xmath6 vanishes , the term @xmath128 in the denominator of ( [ deq1 ] ) can be omitted , and in this case we arrive at a system of _ linear _
equations for the coefficients @xmath129 .
the determinant of the linearized system should vanish at @xmath130 , since by definition there is no solution at that point .
thus we obtain a new condition for gap closure .
for example , in the case of @xmath2p@xmath3@xmath2f@xmath3 pairing ( where channel coupling is included ) , the zero of the determinant corresponding to the dominant @xmath131 matrix elements of the argonne @xmath32 potential @xcite gives a value @xmath132 @xmath133 for the critical fermi momentum when the input function @xmath49 in eqs .
( [ deq1 ] ) is constructed from free single - particle energies @xmath134 . in the standard numerical approach to bcs pairing problems , the original system ( [ gengap ] ) for the gap components @xmath44
is solved by iteration .
one starts with suitable `` guesses '' for these functions ( ten in number , for @xmath2p@xmath3@xmath2f@xmath3 pairing ) and inserts them into the r.h.s . of ( [ gengap ] ) , generating a new set of component functions ; and so on until convergence is hopefully achieved .
such a procedure encounters obstacles even in the case of @xmath0s@xmath1 pairing @xcite .
a conventional iterative approach is even more problematic for pairing in higher angular momentum states , where the dimensionality of the nonlinear problem is much larger . from this point of view ,
( [ deq1 ] ) are far more appropriate as basis for numerical computation , since shape factors @xmath122 are determined separately and iterations are needed only to find the set of coefficients @xmath135 entering the representation ( [ gen1 ] ) . in principle , instead of trying to solve the system ( [ deq1 ] ) directly , we could use the variables @xmath136 ( twenty in number , for @xmath2p@xmath3@xmath2f@xmath3 pairing ) to parametrize the condensation energy and apply the variational principle to determine an appropriate set of these factors and a corresponding set of gap components @xmath25 .
however , one must remember that the performance of optimization algorithms , as well as the convergence of iterative procedures , is very sensitive to the degeneracies of the functional under consideration .
accordingly , any attempt at implementing such an approach should be prefaced by a thorough study of the symmetries of the gap problem and the degeneracies of its solutions .
such a study is of considerable fundamental interest in any case @xcite .
to gain a better understanding of the triplet pairing problem in neutron matter , we revisit the pure ( i.e. , uncoupled ) @xmath2p@xmath3 case analyzed in some detail in @xcite .
if tensor coupling to the @xmath2f@xmath3 state is neglected , the system of gap equations ( [ gengap ] ) reduces to a set of five equations , corresponding to @xmath137 . taking into account the property of time - reversal invariance , which implies the relations @xmath138
it is seen that only three of the generally complex components @xmath139 are independent .
the independent components have customarily been written as @xcite @xmath140 where the @xmath141 and @xmath142 are real functions of @xmath22 .
takatsuka and tamagaki @xcite , as well as amundsen and stgaard @xcite , have carried out numerical studies of the pure @xmath2p@xmath3 problem that demonstrate an interesting feature of triplet pairing in neutron matter ( and in fact of triplet - p pairing more generally ) .
solutions of different types were found to be nearly degenerate , i.e. , they yield very nearly the same condensation energy . the spread of condensation energies reported by takatsuka and tamagaki at a typical density is only 3% , and stgaard and amundsen obtained very similar results .
the calculations of these authors were based on an older nucleon - nucleon potential called opeg ( see sec .
[ sec : tripletnumapp ] for details ) .
although this potential model is no longer competitive , its essential aspects are semi - quantitatively correct . at any rate , it will be argued that the nearly degenerate character of the level structure in the triplet pairing problem and the energy ordering of different solution types are very insensitive to the interaction assumed .
important insights into the uncoupled problem can be achieved within the separation method @xcite . for pure @xmath2p@xmath3 pairing , eq .
( [ sol ] ) becomes simply @xmath143 while the shape factor @xmath144 obeys the uncoupled equation @xmath145 we now make the decomposition @xmath146 , where @xmath147 serves as a scale factor .
relations between the parameters @xmath148 and @xmath149 and the quantities @xmath150 and @xmath151 ( @xmath152 ) stem from the definitions contained in eqs .
( [ deltaeta ] ) . evaluating
the spin sums @xmath153 and separating real and imaginary parts in eq .
( [ deq1 ] ) , we extract the following set of equations determining the coefficients @xmath154 ( cf . refs .
@xcite ) : @xmath155\ , , \nonumber \\
\kappa_2 & = & -v'[\kappa_2(j_0+j_5)-\kappa_1 j_1 + \lambda_1j_2+j_4 ] \ , , \nonumber\\ \lambda_1 & = & -v'[\lambda_1j_6-(\lambda_2 + 1)j_1+\kappa_2j_2-\kappa_1j_4/2]\ , , \nonumber\\ \kappa_1 & = & -v'[\kappa_1j_7-\kappa_2j_1-(\lambda_2 - 1)j_2-\lambda_1j_4/2]\ , , \nonumber\\ 1 & = & -v'[-(\lambda_1 j_1-\kappa_1
j_2+\lambda_2 j_3-\kappa_2 j_4)/3+j_5 ] \ , , \label{base2}\end{aligned}\ ] ] where @xmath156 .
the integrals @xmath157 are given by @xcite @xmath158 with @xmath159 , @xmath160 , @xmath161 , @xmath162 , @xmath163 , and @xmath164 , where @xmath165 , @xmath166 , and @xmath167 .
of these integrals , only @xmath168 contains a principal term behaving like @xmath169 , since only @xmath170 among the @xmath171 does not vanish under the angular integration when multiplied by an angle - independent factor .
the normal - state single - particle energy @xmath172 present in the energy denominator @xmath48^{1/2}$ ] is conventionally assumed to have finite slope @xmath173 , while the gap term takes the explicit form @xmath174 [ \chi_1^{11}(k)]^2 \ , .
\label{dsquare } \end{aligned}\ ] ] in their angular content , the system ( [ base2 ] ) and the trigonometric decomposition ( [ dsquare ] ) are consistent with the formulations of the uncoupled @xmath2p@xmath3 problem given in refs .
@xcite , and specifically with eqs .
( 4.7)(4.8 ) of amundsen and stgaard .
however , our separation formulation , as expressed in eqs .
( [ dchi])([dsquare ] ) , has distinct advantages as a basis for numerical solution of this problem . to begin with
, there is only _
one _ shape function @xmath175 , which satisfies a linear integral equation of exactly the same form as for singlet - s pairing ( cf .
( [ chi ] ) .
this property of shape invariance follows naturally from the formal development presented in sect .
[ sec : sepformalism ] . given the key physical fact that the angle dependence of the problem only comes into play significantly in the close vicinity of the fermi surface , we observe that this part of momentum space is almost entirely eliminated from the @xmath64 integration in eq .
( [ samechi ] ) by the vanishing of @xmath176 at @xmath73 .
thus , to an excellent approximation , all solutions @xmath139 of the uncoupled triplet problem have identical shapes .
in particular , the positions of their zeros and maxima coincide , as is clearly seen in the figures of previous works on triplet pairing in neutron matter @xcite . to establish the angular dependences of the allowed solutions of the @xmath2p@xmath3 problem ,
the shape equation ( [ samechi ] ) must of course be supplemented by a set of equations that determine the coefficients or amplitudes corresponding to the different @xmath24 values .
an important benefit of the separation strategy is that the nonlinear aspects of the problem are now concentrated entirely in the latter system of equations , given by ( [ base2 ] )
equations for a set of _ numbers _ @xmath177 rather than for a set of _
functions_. by contrast , earlier authors @xcite have worked directly with a set of singular linear integral equations for the five functions @xmath178 , @xmath179 , @xmath180 , @xmath181 , and @xmath182 defined in eq .
( [ deltaeta ] ) . in this approach , the dependences of gap quantities on the modulus of the momentum and on angles ( or magnetic quantum numbers ) are confronted simultaneously , with attendant opportunities for numerical difficulties .
obviously , the process of numerical solution is facilitated by the separation of `` shape '' and `` angular '' aspects of the problem that has been accomplished in eqs .
( [ dchi])([dsquare ] ) .
less obviously , this transformation opens the way to an incisive analysis of the nature of these solutions .
such an analysis has been carried out in refs .
@xcite , extending the formalism to finite temperature @xmath31 via introduction of the factor @xmath183 $ ] in the integrands of the basic gap equations eq .
( [ gengap ] ) and the @xmath157 integrals ( [ int10 ] ) .
appealing to the orthogonality of the spherical functions for different @xmath24 , it is possible to find special solutions of the system ( [ base2 ] ) ( or of the system ( 4.7 ) of ref .
@xcite ) that satisfy most or some of its equations identically .
the simplest types of consistent solutions are characterized as follows @xcite : 1 .
only @xmath184 components are present : @xmath185 .
only @xmath186 is present : @xmath187 .
3 . only @xmath188 are present : @xmath189 .
4 . coupled case containing @xmath190 : @xmath191 .
in addition , there is naturally the general case involving all five @xmath24 values . for purposes of illustration ,
let us examine the situation for solution types 1 and 2 in some detail . _
type 1_. under the assumption that only the component with @xmath192 is present ( its @xmath193 partner being included trivially ) , we need only deal with @xmath194 , noting that a free choice of phase allows us to set @xmath195 . in this case
only the first of eqs .
( [ base2 ] ) is relevant and it becomes @xmath196 where @xmath197 .
all the other equations of the set ( [ base2 ] ) are met identically .
( we remind the reader that @xmath198 is the value , on the fermi surface , of the angle average defined in eq .
( [ aagap ] ) . )
apart from the angular dependences and angular integration , eq .
( [ mtwo ] ) has the same form as the equation ( [ deltaf ] ) for the gap amplitude in the @xmath0s@xmath1 problem ( see the appendix ) .
the angular integration can actually be performed analytically , with the result @xmath199 where @xmath200 \ , , \nonumber \\ ( \beta_2(k))^2 & = & { { 3\over 2}\delta^2_f\chi^2(k)\over \xi^2(k ) + { 3\over 2}\delta_f^2\chi^2(k ) } \ , .\end{aligned}\ ] ] the root of eq .
( [ eqtype1 ] ) can be found numerically without difficulty , after finding the universal shape function @xmath201 by matrix inversion of the discretized form of eq .
( [ samechi ] ) .
this has been done for pairing matrix elements calculated from the opeg potential employed by takatsuka and tamagaki @xcite and amundsen and stgaard @xcite ; some results for this case are reported in sect .
[ sec : tripletnumapp ] .
_ type 2_. here there is the single component @xmath202 .
the dispersion equation determining @xmath6 is given by the last of eqs .
( [ base2 ] ) , which becomes @xmath203 analytic integration over angles is again possible , yielding @xmath204 where @xmath205 , \nonumber \\
( \beta_0(k))^2 & = & { { 3\over 4}\delta_f^2\chi^2(k)\over \xi^2(k ) + { 1\over 4}\delta_f^2\chi^2(k ) } .\end{aligned}\ ] ] some numerical results for this case are also reported in sect . [
sec : tripletnumapp ] , again for the opeg potential .
quantitative description of pairing in nuclear systems is beset with special difficulties because the strong inner repulsion ( or strong momentum dependence ) present in some partial waves ( notably s states ) implies that the momentum integral entering the gap equation must be extended to large momenta @xcite .
this behavior precludes the application of simple estimates such as the bcs weak - coupling formula , which assumes sharp localization of the pairing interaction around the fermi surface .
however , important simplifications are permitted by the existence of a small dimensionless parameter @xmath206 , given essentially by the scale of the energy gap at the fermi surface , relative to the value of the fermi energy @xmath207 . in the case of @xmath0s@xmath1 pairing ,
the small parameter is just @xmath208 , where @xmath209 . framing this problem within the separation method ,
taylor - series expansion of the energy denominator of the equation for the gap amplitude @xmath6 has been employed in ref .
@xcite to derive a sequence of approximants for this quantity expressed as series of terms in @xmath210 and @xmath211 with integral @xmath212 .
the derivations involved are outlined in appendix a. even the leading ( zeroth ) approximant yields reasonable results : the exact solution for @xmath6 within the relevant density range is reproduced within 5% , based on pairing matrix elements formed from the @xmath0s@xmath1 component of the argonne @xmath32 interaction . in the triplet case considered here ,
the small parameter is appropriately defined as @xmath213^{1/2}/ \varepsilon_f \equiv \delta_f/\varepsilon_f$ ] .
a characteristic value of this quantity in the density region of interest is 0.5 mev/80 mev @xmath214 , which is at least an order of magnitude smaller than typical values for singlet - s pairing . on this basis
, one may expect to achieve good accuracy through an approximation that retains only the principal logarithmic terms @xmath215 together with those independent of @xmath128 , while omitting corrections @xmath216 .
small - parameter approximations are examined in more detail in the appendix . here
we restrict ourselves only to brief remarks .
consider the integral on the r.h.s . of eqs .
( [ deq1 ] ) for the coefficients @xmath136 .
we may carry out an integration by parts with respect to the variable @xmath217 , to derive a principal term ( within the angular integration ) that goes like @xmath218 and a remainder that involves @xmath128 only in the factor @xmath219^{1/2}\right ) \,.\ ] ] since latter quantity is quite smooth in the vicinity of the fermi surface , its replacement by @xmath220 is valid within corrections of order of @xmath221 that belong to the next stage of approximation .
writing @xmath222 , we extract the scale @xmath198 and express the angular dependence by @xmath223 .
the dominant log term then decomposes as @xmath224 . by virtue of the orthogonality condition ( [ ort1 ] )
, the angle - independent portion @xmath225 contributes only to terms on the r.h.s.of eq .
( [ deq1 ] ) that are diagonal in the magnetic quantum number , i.e. , only for @xmath226 .
since @xmath6 is a small quantity , the contribution from the scale term @xmath225 greatly exceeds that from the angle - dependent part @xmath227 , which remains of order of 1 even in the limit @xmath228 .
it is worth noting that the @xmath229-independent contributions coming from terms in @xmath220 are also of order of 1 .
however , these terms are angle - independent and , like the scale term , contribute only to the diagonal part of the r.h.s . of ( [ deq1 ] ) .
in fact , to within corrections of order @xmath216 , the result for the @xmath230 coefficients calculated from the angle - independent terms alone ( i.e. , dropping @xmath227 ) , is the _ same _ as the result we obtain if we simply replace the quantity @xmath128 in the energy denominator of the original equation by its angle average ( [ aagap ] ) , @xmath231 we reiterate that justification of this replacement , which corresponds to the _ angle - average approximation _ first introduced by baldo et al .
@xcite , rests on ( i ) the unimportance of the angle - dependent quantity @xmath232 relative to @xmath233 and ( ii ) the unimportance of correction terms of order @xmath234 or higher in the small parameter expansion .
the angle - average approximation generally overestimates the pairing gap and the magnitude of the pairing energy ( i.e. the pair condensation energy correction to the energy of the normal ground state ) .
however , it allows a considerable simplification of the triplet pairing problem with very modest sacrifice in accuracy , if our primary interest lies in the magnitude of the pairing effect rather than the detailed structure of pairing solutions . making the replacement ( [ avgap ] ) in the general coupled - channel equations ( [ deq1 ] ) for the gap amplitudes , we arrive at the quasi - uniform system
@xmath235 since the index @xmath24 appears explicitly only on the quantities being sought , there exists a degeneracy with respect to this quantum number : all sets of @xmath136 yielding the same result for the gap quantity @xmath236 of eq .
( [ avgap ] ) are equivalent .
we first apply the angle - averaging prescription to the pure @xmath2p@xmath3 problem . in this case
( [ deq2 ] ) becomes simply @xmath237 with @xmath238 ^ 2 = \delta^2_f\chi^2(k)$ ] and @xmath197 .
this equation , having no explicit reference to magnetic substates , is identical in form to the singlet - channel equation ( [ deltaf ] ) and is solved by the same method .
numerical results from solution of the uncoupled gap problem in this approximation are presented and discussed in sect .
[ sec : tripletnumapp ] .
if we do not care about the detailed angular behavior of solutions , but only about the size of the gap ( as measured by @xmath6 ) and the condensation energy , the angular - averaging prescription gives very good predictions .
this statement is justified by the close agreement with the results shown in fig .
[ fig : aaaandm ] for solution types 1 and 2 and by the near degeneracy of the @xmath6 values and condensation energies found numerically @xcite and analytically @xcite for the different solution types . turning to the general coupled - channel case ,
observe that the terms with @xmath239 on the r.h.s
. of ( [ deq2 ] ) contain integrals @xmath240 which are almost independent of the gap value , since the property @xmath241 implies that the numerators of the integrands vanish at the fermi surface . in the diagonal case ,
@xmath76 , we introduce an unknown positive number @xmath242 ^ 2 } { 2\sqrt{\xi^2(k ) + { \overline{d^2}}(k ) } } { \rm d } \tau \label{zdef}\ ] ] and write @xmath243 where @xmath244 ^ 2 } { 2\sqrt{\xi^2(k ) + { \overline{d^2}}(k ) } } { \rm d } \tau \approx \int \frac { \phi^{ll'}(k)\chi^{l'l_2}_{l'}(k ) -[\chi^{11}_1(k)]^2 } { 2|\xi(k)| } { \rm
d } \tau \ , .
\label{bform}\ ] ] the simplification on the right in eq .
( [ bform ] ) is permitted since @xmath245 and the numerator in the integrands again vanishes at the fermi surface .
the integrals @xmath246 are effectively already known : with the indicated simplifications , they only depend on the @xmath124 functions , which may be found from eqs .
( [ chiviaphi ] ) without reference to the gap quantities @xmath116 being sought .
we now specialize to coupled - channel , @xmath2p@xmath3@xmath2f@xmath3 pairing where the system ( [ deq2 ] ) for the coefficients @xmath116 contains four equations .
for a solution of this set of equations to exist , the determinant @xmath247 which depends on @xmath6 ( or the @xmath116 quantities ) only through the parameter @xmath248 , must vanish .
evaluation of this determinant may be simplified by subtracting the third row from the second row and the second row from the fourth row , and then subtracting the second column from the first column and the third column from the fourth column .
this leads to a _
equation for the determination of the unknown parameter @xmath248 : @xmath249 where @xmath250 @xmath251 & c_{33 } - c_{13 } - [ c_{32 } - c_{12}]\\ c_{00 } - c_{20 } - [ c_{01 } - c_{21 } ] & c_{03 } - c_{23 } - [ c_{02 } - c_{22 } ] \end{array } \right| \ , , \ ] ] @xmath252 equation ( [ viet ] ) can have two roots @xmath253 , and only positive solutions are of interest . having found the appropriate root of ( [ viet ] ) , one may solve any three of the four equations of the system ( [ deq2 ] ) for the ratios @xmath254 .
these ratios may then be used to obtain an equation for @xmath255 that is analogous to the algebraic equation for the gap amplitude @xmath6 in the @xmath0s@xmath1 problem , namely @xmath256}^2 } { 2\sqrt{\xi^2(k ) + { \overline{d^2}}(k ) } } { \rm d}\tau\ , .\ ] ] solution of this equation yields the desired gap function in angle - average approximation .
the angle - average approximation serves well if the goal is to calculate the gap amplitude @xmath257^{1/2}$ ] .
however , if information is required on the spectrum of pairing solutions and the energy splittings between these solutions , it is essential to include the angle - dependent effects from the terms in eq .
( [ deq1 ] ) going like @xmath258 . by so doing
, accurate results and valuable insights can be gained even if we stop at leading order in the small - parameter approximation .
the situation is made more tangible by considering two simple solutions involving only one @xmath24 value ( and its negative ) , specifically @xmath192 or @xmath259 ( corresponding to types 1 and 2 , respectively ) .
working at lowest order in the small - parameter expansion , partial integration of the definition ( [ int10 ] ) of the integrals @xmath157 over the variable @xmath217 leads to the simplified form @xmath260 having made use of the equalities @xmath261 . with this result ,
( [ mtwo ] ) and ( [ mzero ] ) are recast , respectively , as @xmath262 \ , .\ ] ] the splitting between the states in question is determined by subtracting one equation from the other .
straightforward manipulations give @xmath263 when the angle - average approximation is applied , eq .
( [ mtwo ] ) or ( [ mzero ] ) is replaced by @xmath264 which is to be solved for @xmath265 .
( here , @xmath266 and @xmath267 after a change of integration variable from @xmath22 to @xmath268 . )
evaluating the ratios of @xmath269 and @xmath270 to @xmath271 along the lines of the above derivation of @xmath272 , the ordering of the three gap amplitudes is seen to be @xmath273 .
this order agrees with the numerical results appearing in fig .
[ fig : aaaandm ] and with the findings of other authors @xcite .
it is interesting and highly significant that the result ( [ ratio ] ) for the ratio of pairing energies for the type 1 ( @xmath274 ) and type 2 ( @xmath259 ) solutions obtained to leading order in the small parameter of the theory yet very accurate is completely independent of the parameters of the actual interaction and independent of density .
in other words , it is _
universal_. proceeding to the next layer of complexity , let us consider the type 4 solution , again studied in leading order .
in this case , @xmath259 and @xmath275 substates coexist , and the system of equations that determines the coefficients @xmath154 becomes @xmath276 \ , , \label{20lam}\ ] ] @xmath277 \ , , \label{scale}\ ] ] where we have introduced the more compact notation @xmath278 and the @xmath157 are given by ( [ int10 ] ) .
the first of these equations can be conveniently rewritten as @xmath279 by subtracting @xmath280 times eq .
( [ scale ] ) from eq .
( [ 20lam ] ) .
the alternative form ( [ dif ] ) is distinctly advantageous : unlike @xmath168 , the integrals @xmath281 and @xmath282 do not contain a term proportional to @xmath283 and hence are independent of any details of interaction .
consequently , eq . ( [ dif ] ) provides a closed equation for the parameter @xmath280 whose solution is universal
i.e. , the same for all pairing interactions .
once the universal parameter @xmath280 is determined , the scale parameter @xmath284 is calculated from ( [ scale ] ) ; its value does , of course , depend on the interaction .
the formulas for @xmath281 and @xmath282 can be further simplified by partial integration of @xmath281 over @xmath285 and @xmath282 over @xmath286 .
the remaining integration over @xmath286 is performed analytically .
after simple operations , we arrive at @xmath287 where @xmath288^{1/2}$ ] . in general , these integrals are expressible in elliptic functions , but in exceptional cases they reduce to elementary integrals .
this is exactly what happens when we seek roots of the system ( [ 20lam]([dif ] ) .
in addition to the root @xmath289 corresponding to the @xmath259 state ( the type 2 solution ) , we find the roots @xmath290 and @xmath291 .
it should not be surprising that the roots are degenerate with respect to signs , since eq .
( [ dif ] ) is invariant under reversal of the sign of @xmath280 .
what is most surprising is that these roots are rational numbers .
rationality of the roots @xmath280 is a consequence of a hidden symmetry of the problem that transcends the approximation we have employed . to substantiate this assertion ,
consider first the particular solution @xmath292 , for which the squared gap function ( [ dsquare ] ) becomes @xmath293 ^ 2[4 - 3(x^2+z^2)]$ ] .
the symmetry of this function with respect to @xmath294 and @xmath285 implies the relation @xmath295 , since the combination @xmath296 changes sign on interchange of @xmath294 and @xmath285 whereas @xmath297 and other factors within the integrand of ( [ int10 ] ) are left unchanged .
the above relation between @xmath281 and @xmath282 satisfies eq .
( [ dif ] ) .
the essential point is that this relation derives from a simple symmetry of the problem ; it is _ not _ predicated on any approximate scheme for the evaluation of the @xmath157 . now consider the root @xmath298 , for which @xmath299 ^ 2(1-x^2)$ ] .
since @xmath300 is independent of @xmath301 and @xmath285 , one may trivially integrate ( [ int10 ] ) over these variables .
simple algebra then gives @xmath302 , and this relation also satisfies eq .
( [ dif ] ) .
we may therefore conclude that the ratio of @xmath192 and @xmath259 components in the type 4 solution is independent of any input parameters including density , temperature , and the parameters specifying the pairing interaction .
a general analysis along similar lines , again exploiting the simplifications produced by the separation method , reveals the fundamental structure of the full set of solutions of the bcs @xmath2p@xmath3 pairing problem in neutron matter @xcite .
considering general multicomponent solutions , the allowed ratios of coefficients of different spin - angle matrices @xmath303 in the expansion ( [ matgap ] ) of the gap matrix turn out to be universal .
as in the above derivation , any reference to interaction , density , or temperature cancels out to very high accuracy in forming the ratios .
the complete set of solutions for these ratios is obtained explicitly by solving a set of five coupled ` algebraic ' equations , one of which serves to determine the energetic scale of the pairing effect .
the solution set and the spectrum of pairing energies are found to be highly degenerate in character .
the small splittings between different solutions can be found by manipulations similar to those performed above for the type 1 and type 2 solutions . in principle , analogous analytic arguments may be used to establish universalities of pairing in the much more difficult problem of liquid @xmath2he , where states with @xmath304 and @xmath16 contribute on an equal footing and the number of equations to be solved rises from five to nine .
the separation approach to bcs - type pairing problems which has been further developed and studied in this paper leads to considerable simplifications of the process of solving the gap equations in a generic pairing channel . aided by the existence of a small parameter , this approach allows one to find the _ interval of existence _ of solutions without having to perform any iterations , and it allows one actually to find _ solutions _ without having to iterate a set of integral equations .
some iterations may be required to construct solutions , but they need be performed only on a set of algebraic equations for the amplitudes of the components of the gap matrix ( quantities that are momentum - independent but dependent on the magnetic projection @xmath24 ) .
the `` division of labor '' that underlies the separation method thus facilitates the process of finding solutions ; moreover , because of the reduced number of numerical operations , it permits higher accuracy to be achieved .
in addition to these numerical advantages , the new method offers an efficient framework for fundamental analytic investigations of pairing , as has been demonstrated in sect .
[ sec : nocoupling ] and more systematically in refs .
. a necessary step in testing our procedures for numerical solution of the triplet pairing problem in neutron matter is comparison with results obtained by other authors . as indicated in sect .
[ sec : nocoupling ] , takatsuka and tamagaki @xcite ( see also their review @xcite ) have studied the pure @xmath2p@xmath3 pairing problem at some depth , classifying the types of solutions and generating numerical solutions by iteration .
subsequently , takatsuka @xcite ( see also @xcite ) implemented an iterative procedure for solution of the gap equations in the @xmath2p@xmath3@xmath2f@xmath3 coupled - channel case with fixed @xmath186 .
this early numerical work was based on several older potential models @xcite with soft ( gaussian ) cores .
the takatsuka - tamagaki calculations were repeated and supplemented by amundsen and stgaard @xcite .
baldo et al .
@xcite introduced the idea of angular averaging in considering @xmath259 and @xmath275 solutions ( i.e. , type 2 and type 1 solutions ) of the pure @xmath2p@xmath3 case and reported preliminary numerical results for a separable approximation to the argonne @xmath305 potential @xcite .
more recently , elgary et al .
@xcite have performed uncoupled @xmath2p@xmath3 and coupled @xmath2p@xmath3@xmath2f@xmath3 calculations with fixed @xmath306 ( 0 or 2 ) in @xmath307stable matter for the bonn potential models designated a , b , and c @xcite . in the most recent microscopic study of triplet pairing , baldo et al .
@xcite obtained results for several realistic models of the nucleon - nucleon interaction , including four that may legitimately be regarded as phase - shift equivalent . the latter potentials
argonne @xmath32 @xcite , cd - bonn @xcite , and nijmegen i and ii @xcite belong to a new generation of models that yield high - precision fits of the sanitized world supply of @xmath308 and @xmath309 scattering data up to a bombarding energy of 350 mev .
all four produce fits with @xmath310 per data point close to unity .
in addition , gap calculations were performed for three `` older '' potentials
argonne @xmath305 , paris @xcite , and bonn b. the authors of this extensive study conclude that reliable predictions of the @xmath2p@xmath3@xmath2f@xmath3 pairing gap at densities above about 1.7 times the saturation density of symmetrical nuclear matter will require potential models that fit the nucleon - nucleon phase shifts up to 1 gev . for the purpose of comparison with the pre-1990 results on the triplet pairing problem
, we have chosen the opeg ( `` one - pion - exchange gaussian '' ) potential @xcite used in refs .
@xcite , and specifically the version denoted opeg @xmath311 . this potential has been fitted to scattering data collected before 1970 , and it can not be regarded as a realistic model of the nucleon - nucleon interaction .
however , it possesses all of the essential qualitative features of the interaction and describes them at a semi - quantitative level . of local character
, it contains central , tensor , spin - orbit ( @xmath312 ) , and quadratic spin - orbit ( e.g. @xmath313 ) components , as written out in detail by amundsen and stgaard @xcite . to compare with the latest calculations @xcite ,
we have generated results for the argonne @xmath305 and argonne @xmath32 models .
the diagonal pairing matrix elements of these two potentials are plotted in fig .
[ fig : diagme ] . in presenting the numerical results , we employ two measures of the gap matrix , namely ( i ) the average quantity @xmath314^{1/2}$ ] defined via eq .
( [ aagap ] ) or eq .
( [ avgap ] ) and ( ii ) the gap component @xmath25 , either as a function of the momentum modulus @xmath22 or evaluated at the fermi momentum @xmath60 .
we note that when the pairing problem is solved in the angle - average approximation specified in sect .
[ sec : aaa ] , the @xmath24 dependence of the latter measure becomes moot .
thus the index @xmath24 will often be suppressed , along with the fixed index @xmath103 .
the zeros of the characteristic determinant of the linearized version of eq .
( [ deq1 ] ) provide us with candidate values for the upper critical fermi momentum @xmath315 .
this is true whether we restrict ourselves to pure @xmath2p@xmath3 pairing or address the coupled - channel @xmath2p@xmath3@xmath2f@xmath3 case .
examining first the uncoupled case , in fig .
[ fig:3in1 ] we superimpose plots of the diagonal matrix elements @xmath316 of the argonne @xmath32 potential , the associated characteristic determinant , and the gap component @xmath317 obtained in the angle - average approximation , all considered as functions of the fermi momentum @xmath60 . from the preceding formal developments it is evident that a zero of the potential implies a pole of @xmath318 and hence a simple pole of the determinant .
such a pole is very prominent in fig .
[ fig:3in1 ] and is naturally followed closely by a zero of the determinant . the position of this zero gives a value @xmath319 @xmath133 for the upper critical @xmath60 value at which the gap closes , the corresponding upper critical density being @xmath320 . on the plot ,
the gap amplitude is clearly negligible at this point . because of the exponential falloff
, it appears to be zero even below this point , but is in fact still finite over an interval in @xmath60 where @xmath321 has positive values .
[ fig : tripdet ] displays the characteristic determinant in the coupled - channel case , again with the argonne @xmath32 potential as input for the pairing force .
the determinant is evaluated for three different effective - mass ratios @xmath322 , the free normal - state single - particle spectrum @xmath323 being replaced by @xmath324 to simulate dispersive effects of the medium . for any of the chosen values of @xmath325 ,
the determinant has a pole near @xmath326 @xmath133 followed closely by a zero near @xmath327 @xmath133 , and a second pole near @xmath328 @xmath133 followed closely by a zero near @xmath329 @xmath133 .
it is the second zero that should be identified with the upper critical density , since the adjacent pole corresponds to a zero of @xmath330 .
the point is that we are seeking to describe @xmath12 triplet - p - wave pairing as modified by the tensor coupling to the @xmath12 triplet - f channel .
the first pole of the determinant corresponds to the zero of @xmath331 .
thus the accompanying determinantal zero is not relevant to the task at hand , which is the identification of the upper critical point for triplet - p rather than triplet - f pairing .
generally , a _ lower bound _ on @xmath315 for pairing predominantly in a state with orbital angular momentum quantum number @xmath88 is given by the location of the pole that is produced by the vanishing of @xmath332 ( we assume that this occurs only at one density ) .
conventional wisdom gives @xmath333 as a sufficient ( but not necessary ! ) condition for a pairing instability in partial wave @xmath88 .
the indicated lower bound property is of course consistent with this statement , but we are able to provide a more incisive condition for the extent of the pairing instability in terms of the behavior of the characteristic determinant associated with the integral equation ( [ deq1 ] ) ( or alternatively , with the characteristic determinant of the integral equation ( [ chiviaphi ] ) for the shape function @xmath334 @xcite ) . concerning the dependence of @xmath315 on effective mass , it is seen in fig . [
fig : tripdet ] that a smaller @xmath325 leads to a lower value of the upper critical fermi momentum ( cf . further discussion below ) .
numerical results for the opeg interaction that may be compared with those of takatsuka and tamagaki and amundsen and stgaard ( and with general findings of other authors ) are displayed in fig .
[ fig : aaaandm ] and fig .
[ fig : opegcoupled ] . fig .
[ fig : aaaandm ] refers to the uncoupled @xmath2p@xmath3 case with @xmath335 and shows results for the @xmath60 dependence of values of the energy - gap parameter @xmath6 corresponding , respectively , to the angle - average approximation ( uppermost curve ) , the pure @xmath259 solution ( middle curve ) , and the @xmath275 solution ( lowest curve ) .
the latter two curves are in reasonable agreement with results reported in refs .
@xcite , having maxima near 2 mev in the vicinity of @xmath336 @xmath133 .
it is worth mentioning again that our calculations are in accord with the energy ordering of the @xmath259 and @xmath275 solutions found numerically in the early work @xcite ; moreover , they confirm the generic results that have been derived in sect .
[ sec : aaa ] using the separation method and the small parameter expansion . at @xmath327 @xmath133 the ratio predicted by eq .
( [ ratio ] ) is reproduced with better than 1% accuracy , and the energy order @xmath337 implied by eqs .
( [ ratio ] ) and ( [ aaa0 ] ) is preserved . considering the near - degeneracies in the condensation energies for the different solutions , documented originally by takatsuka and tamagaki @xcite
, these favorable comparisons warrant some confidence in the application of the angle - average approximation to the coupled - channel case . introducing the coupling to the @xmath2f@xmath3 channel
, we have solved the triplet pairing problem for the opeg and argonne potentials in the angle - average approximation , using the procedure outlined in sect .
[ sec : aaa ] .
[ fig : opegcoupled ] shows the calculated momentum dependence of the gap components @xmath338 ( solid curve ) and @xmath339 ( dashed curve ) for the case of the opeg potential and @xmath335 , at a fermi momentum @xmath327 @xmath133 near the value yielding the maximum strength of triplet pairing .
the locations of the maxima , minima , and nodes of these functions are in close agreement with results given in refs .
@xcite , and the magnitudes are in general agreement .
( it is to be noted that in his coupled - channel calculation , takatsuka included only the @xmath259 magnetic substate . )
[ fig : v18coupled ] , [ fig : deltav18 ] , and [ fig : opegandv18coupled ] depict the principal results of our coupled - channel @xmath2p@xmath3@xmath2f@xmath3 gap calculations for the more realistic potential models , @xmath32 and @xmath305 . fig .
[ fig : v18coupled ] illustrates the momentum dependence of the @xmath340 and @xmath341 components @xmath342 for the argonne @xmath32 interaction ( abbreviated as @xmath343 and @xmath344 ) .
as expected , the shapes of these gap functions are rather similar to those found for the opeg interaction at the same density ( cf .
[ fig : opegcoupled ] ) .
however , the overall scale of the pairing effect in the opeg case is almost 50% larger than for the @xmath32 model . in fig .
[ fig : deltav18 ] , we view the dependence of the gap component values @xmath345 and @xmath346 on the fermi momentum @xmath60 , as calculated in angle - average approximation for the @xmath32 interaction .
contrasting the results for @xmath347 with and without coupling to the @xmath348 wave , it is seen that the tensor force acts to substantially increase the pairing effect , as has been demonstrated in greater or lesser measure by earlier investigators @xcite .
[ fig : opegandv18coupled ] compares the @xmath60 dependence of the @xmath2p@xmath3@xmath2f@xmath3 energy - gap parameter @xmath6 for the three potentials investigated , the angle - average approximation being invoked in all cases . in all three cases , the gap measure @xmath6 peaks in the range between @xmath349 and 2.5 @xmath133 . with the free single - particle spectrum , i.e. , @xmath335
, the maximum value reached by this quantity for the @xmath305 interaction is twice that attained for the more refined @xmath32 model .
as is clear from its definition , @xmath6 contains direct contributions both from p - wave and f - wave pairing . inspecting fig .
[ fig : diagme ] , we observe that the greater pairing effect for the @xmath305 potential relative to @xmath32 can be traced to the greater attraction present for @xmath305 in both p- and f - wave @xmath103 channels over the relevant range of the momentum variable @xmath22 .
( the coupling matrix elements @xmath350 for the two potentials are essentially equivalent over the latter range . ) fig . [ fig : opegandv18coupled ] also demonstrates the greater pairing effect of the opeg potential relative to @xmath32 .
such discrepancies between the predictions obtained for different interaction models are not unexpected : due to the log singularity intrinsic to the bcs pairing phenomenon , the overall strength @xmath6 of the effect in the @xmath2p@xmath3@xmath2f@xmath3 channel is extremely sensitive to the behavior of the input pairing matrix elements at momenta in the vicinity of the fermi surface . in turn , since the pairing matrix elements are constructed directly from the given free - space @xmath11 potential model and since @xmath351 @xmath133 , one must expect a sensitive dependence on properties of the bare interaction that are not determined , or only poorly determined , by fits of the @xmath11 scattering data up to 350 mev . this point has been developed thoroughly and convincingly in the recent work of baldo et al .
@xcite . in table 1
, we report some explicit numerical values of @xmath6 obtained in our calculations based on the two `` realistic '' potential models , argonne @xmath305 and @xmath32 . a comparison of our numerical results for these potentials with the corresponding results of baldo et al .
@xcite is in order , restricting attention perforce to the case of free single - particle energies @xmath134 .
first , we point out that the pairing matrix elements entering our formulas are larger than those appearing in ref . @xcite by a numerical factor @xmath352 , due to different choices of normalization for the spherical bessel functions . upon accounting for this ( immaterial )
factor , the diagonal matrix elements plotted for the @xmath32 interaction in fig . 3 of ref .
@xcite are apparently in excellent agreement with their counterparts in fig .
[ fig : diagme ] .
second , it must be recognized that the quantity @xmath353 defined in eq .
( 6 ) of ref .
@xcite is actually twice the function @xmath353 appearing in our treatment ( see our eqs .
( [ squd ] ) and ( [ avgap ] ) ) .
thus the gap measure @xmath354^{1/2}$ ] we have calculated is smaller than the @xmath6 quantity of baldo et al . by a factor @xmath355 .
the definition we have adopted is motivated by the desideratum of consistency with the earlier literature @xcite . in any case
, the two calculations should agree once this scale factor is applied , since both calculations employ the angle - average approximation .
the results are generally in good agreement ; however , there are some modest but significant numerical discrepancies . for the @xmath305 interaction ,
the current results for @xmath6 versus @xmath60 are somewhat higher than those of ref .
@xcite below the peak , and somewhat lower above .
the position of the peak , occurring near @xmath356 @xmath133 in the calculation of baldo et al .
, is shifted to a slightly smaller @xmath60 value , near 2.2 @xmath133 .
the value of @xmath6 at the peak is higher in our calculation by roughly 5% . in the case of the @xmath32 interaction ,
the numerical agreement is excellent below the peak , but beyond the maximum the current results lie below those of ref .
@xcite by an amount that can reach about 0.1 mev ( in the @xmath6 measure used by baldo et al . ) .
the origin of these relatively minor numerical disparities is not clear .
we would , however , like to reiterate the concern expressed in refs .
@xcite about the sensitivity of the numerical integration over @xmath22 in the triplet gap equations to the choice of mesh points near the fermi momentum .
the difficulty arises because the functions @xmath357 and @xmath358 are very strongly peaked at @xmath60 . to handle this problem within the present treatment , we have integrated out the troublesome log - like part by hand and performed a numerical integration only for the remainder , which is much smoother and well - behaved . for the latter integration
, we employ 288 points , dividing the @xmath22 domain into three intervals : @xmath359 $ ] , @xmath360 $ ] , and @xmath361 $ ] , with @xmath362 @xmath133 .
each interval contains 96 mesh points distributed in gaussian fashion very dense around the ends of the interval and sparse in the middle . in fig .
[ fig : opegandv18coupled ] and table 1 , we have included @xmath32 results not only for the baseline choice @xmath335 , but also for a value @xmath363 of the effective - mass parameter conventionally thought @xcite to provide a sensible representation of the in - medium single - particle energy .
this brings us to the last point that we would like to address in the current treatment of triplet pairing in neutron matter .
relative to the bare - mass case , the propagator renormalization implied by @xmath363 yields a substantially smaller gap as well as a lower value for the upper critical density ( see fig . [
fig : tripdet ] ) .
these effects can be readily understood in terms of a reduction of the density of states @xmath364 near the fermi surface , which is related to the effective mass by the familiar expression @xmath365 . in the singlet - s - wave problem ,
the approximate formula ( [ df0 ] ) derived in the appendix shows clearly the inverse exponential dependence of the leading component of the gap on the density of states in the case of a negative pairing interaction on the fermi surface , as does the bcs weak - coupling formula in a more limited context . the same physical reasoning and similar mathematical connections apply in the triplet case .
it is a common belief that at the higher densities where triplet pairing becomes important , the interactions with particles in the background neutron sea produce a considerably greater reduction from the bare mass than is the case at inner - crust densities where singlet pairing dominates @xcite .
this conclusion rests on the results of the brueckner - hartree - fock and correlated - basis calculations .
however , in these calculations , the contribution of pion - exchange effects to the renormalization of the nucleon effective mass is ignored .
the role of the pion degrees of freedom grows rapidly with increasing density , eventually giving rise to a rearrangement of the ground state ( associated , in particular , with pion condensation @xcite ) .
the latest variational prediction @xcite of the density @xmath126 at which this phase transition sets in is based on a nuclear hamiltonian containing the argonne @xmath32 two - nucleon interaction ( with boost corrections to account for the leading relativistic effects ) and phenomenological three - nucleon interactions .
the result @xmath366 @xmath367 actually lies below the density at which triplet pairing is expected to reach its maximum strength .
straightforward considerations @xcite demonstrate that in the vicinity of this transition point the contribution of the pion - exchange diagram to the effective mass @xmath368 diverges and @xmath325 goes to infinity .
therefore it is not unlikely that the operative value of the effective mass in the domain of triplet pairing considerably exceeds the bare mass . from this standpoint ,
conventional treatments that assume an @xmath325 value somewhat below unity may lose their relevance .
exploiting the separation approach and the existence of a small parameter , we have been able to clarify important general properties of triplet pairing in neuron matter and to establish a framework and explicit procedures for reliable numerical solution of the relevant bcs gap equations .
this effort complements a recent exhaustive analysis of the full set of generic solutions of the pure ( uncoupled ) @xmath2p@xmath3 problem , also performed with the aid of the separation strategy @xcite . in view of the corresponding advances that have been made in the @xmath0s@xmath1 pairing problem @xcite ,
it can be reasonably claimed that the _ mathematical and analytical _ aspects of bcs pairing of identical fermions in @xmath0s@xmath1 and @xmath2p@xmath3 states are now well understood , both within the relevant nucleonic contexts and in a more universal , context - independent sense .
the same can not be said of the specific _ physical and mechanistic _ aspects of triplet or singlet pairing of strongly - interacting fermions , which are associated with the realistic in - medium particle - particle interaction and renormalized single - particle energies that should provide the raw material for the gap equations .
a complete and realistic treatment of pairing in a given strongly - coupled fermi system demands _ ab initio _ calculation of these essential inputs .
the success of the method of correlated - basis ( or cbf ) theory across a broad range of many - body systems and phenomena @xcite suggests the following approach to the quantitative physics of pairing in extended nucleonic systems : 1 .
dressing of the pairing interaction by jastrow correlations within cbf theory @xcite 2 .
dressing of the pairing interaction by dynamical collective effects within cbf theory @xcite ( including polarization effects arising from exchange of density and spin - density fluctuations , etc . ) 3 .
consistent renormalization of single - particle energies by short- and long - range correlations within cbf theory ( cf . ref .
@xcite ) this approach has already been explored in the @xmath0s@xmath1 neutron pairing problem @xcite , although the assumed jastrow correlations have not been optimized and only a second - order cbf perturbation treatment is available for step ( b ) .
application of this scheme to @xmath2p@xmath3@xmath2f@xmath3 pairing in neutron - star matter is a challenging but potentially rewarding prospect
. the inclusion of other physical effects will also be important in achieving a realistic description of pairing in the neutron - star medium .
modifications due to three - nucleon interactions should be considered , the calculations should be performed in @xmath369-stable matter ( cf .
@xcite ) , and relativistic effects should be examined ( see , for example , ref .
@xcite ) . as pointed out in sect .
[ sec : tripletnumapp ] , the impact on pairing of mesonic degrees of freedom , especially through an enhancement of the nucleon effective mass in the vicinity of pion condensation , deserves close investigation . in conclusion , it is important to emphasize that although the specific calculations performed in the present work were based on pairing matrix elements computed directly from bare , in - vacuum @xmath11 interactions and on quadratic single - particle energies , many of our findings are unaffected by these simplifications .
indeed , the analyses that have been performed ( leading to the universal properties discussed here and in ref .
@xcite ) , as well as the calculational procedures that have been developed , will retain their applicability and essential validity when realistic in - medium inputs are employed in the gap equations . within this in mind , further studies in the same spirit aimed at developing accurate computational procedures and uncovering universal properties are of considerable interest . as a continuation of the work of sect .
[ sec : tripletnumapp ] , the separation - transformed bcs gap equations for pure neutron matter may be solved in the @xmath2p@xmath3@xmath2f@xmath3 coupled channel without resorting to the angle - average approximation , but still within the setting of the rapidly convergent small - parameter expansion , and with bare pairing matrix elements and unrenormalized single - particle energies .
another project that awaits conclusion is a perturbative treatment of the lifting of the universalities and degeneracies that were established for the uncoupled @xmath2p@xmath3 pairing by the analysis carried out in ref .
also , the separation analysis performed in that paper may be extended to explore the superfluid phase diagram of liquid @xmath2he although this task is considerably more demanding than in the case of @xmath2p@xmath3 pairing in neutron matter .
finally , application of the separation approach to studies of @xmath2s@xmath18@xmath2d@xmath18 pairing in symmetrical nuclear matter and in the expanding phase of heavy - ion collisions may produce new insights into these intriguing problems .
this research was supported in part by the us national science foundation under grant nos .
phy-9602127 and phy-9900713 and by the mcdonnell center for the space sciences .
we thank m. hjorth - jensen , .
elgary , and r. b. wiringa for stimulating discussions and helpful communications .
in this appendix we review the application of the separation method to @xmath0s@xmath1 pairing @xcite and display the first two approximants to the gap amplitude @xmath370 within a small - parameter expansion of the separated problem @xcite .
having set the pattern , we then sketch the extension of the small - parameter expansion to @xmath2p@xmath3 pairing . to provide a baseline for the separation method
, it is useful to revisit the problem of a separable pairing matrix @xmath371 with @xmath372 and @xmath373 .
the standard gap equation in the @xmath0s@xmath1 channel , @xmath374 with @xmath375 , becomes @xmath376 ( one may note that the differential volume element @xmath377 entering eqs .
( [ app1])([app2 ] ) and other formulas below differs from the volume element @xmath378 occurring in integrals elsewhere in this paper .
the difference arises from a different choice of normalization in forming the pairing matrix elements . in this appendix ,
the s - wave pairing matrix elements are defined as in refs .
@xcite , whereas the convention employed by takatsuka and tamagaki @xcite for general pairing matrix elements is adopted in the rest of the paper . ) since the integral on the r.h.s . of eq .
( [ app2 ] ) is just a number , we see that @xmath379 determines the shape of the gap function , while the integral determines its scale .
introducing a dimensionless shape function @xmath334 such that @xmath380 , the gap equation for this problem is equivalent to @xmath381 thus , one has only to solve a nonlinear equation for the gap amplitude rather a nonlinear singular integral equation for the gap function .
we note that a solution of this equation exists only when @xmath382 is negative , in which case the monotonicity of the r.h.s .
implies that there exists one and only one solution .
now we turn to the general situation where @xmath27 need not be separable . to gain some benefits of the separable case , let us split the pairing interaction into a separable part and a remainder that vanishes when either momentum variable is on the fermi surface : @xmath383 where @xmath384 and , as before , @xmath385 . again writing @xmath386 , the gap equation ( [ app1 ] )
is readily seen to be equivalent to the integral equation for the shape function @xmath334 @xmath387 together with the ` algebraic ' equation @xmath388 for the gap amplitude @xmath6 ( assumed nonzero ) .
it is important to note that since @xmath389 is zero by construction , the integral equation ( [ chi ] ) has a nonsingular kernel , the log - singularity of the original bcs formulation having been isolated in the amplitude equation ( [ deltaf ] ) .
an iterative solution of this set of equations converges very rapidly .
indeed , for neutron matter at densities corresponding to the inner - crust region of a neutron star , a good approximation to the gap is already obtained by replacing @xmath390 in the denominator of ( [ chi ] ) by a small constant scale factor @xmath71 ( e.g. @xmath391 ) solving ( [ chi ] ) for @xmath334 by matrix inversion , and substituting the result into ( [ deltaf ] ) to obtain the scale factor @xmath6 by ( say ) newton s method .
the existence of a small parameter @xmath392 for the problem plays a key role in the success of the separation method based on eqs .
( [ chi])([deltaf ] ) .
consider the equation ( [ chi ] ) for the shape function @xmath334 .
since the taylor expansion of the remainder function @xmath176 around @xmath73 starts with the first - order term @xmath393 , the integral over @xmath64 in the symmetric vicinity @xmath394 $ ] of this point vanishes for sufficiently smooth functions @xmath176 and @xmath334 , @xmath395 also , if beyond this interval we have @xmath396 , or in other words if @xmath397 , it is a safe approximation to replace the @xmath6 term in the denominator of ( [ chi ] ) by a rather arbitrary scale factor @xmath71 .
we thereby obtain a practically equivalent _ linear _ integral equation @xmath398 which is always soluble for well - behaved interactions .
for most purposes , it is permissible simply to set @xmath71 equal to zero .
the nature of this `` @xmath71 approximation , '' and more generally of expansions in the small parameter @xmath206 , is developed more fully below .
for realistic nuclear potentials , the accuracy of this approximation in the singlet - s problem is in fact _ much _ higher than indicated by the simple error estimate @xmath399 that accompanies the above preceding .
the reason for this heightened accuracy is that the main contribution to the integral in ( [ chi ] ) comes from a region well separated from the fermi surface .
the @xmath0s@xmath1 pairing matrix elements of realistic nuclear potentials show maxima at @xmath400 @xmath133 .
hence the contribution from the fermi surface region , already small because of the structure of @xmath176 , becomes quite negligible in comparison .
based on taylor - series expansion of the energy denominator of eq .
( [ deltaf ] ) , the gap amplitude @xmath6 may be systematically expressed as a combination of series of terms in @xmath401 and @xmath402 with integral @xmath212 . introducing the dimensionless variable @xmath403 , eq .
( [ deltaf ] ) is first recast as @xmath404 the integration over @xmath405 has been divided two intervals , namely ( i ) the vicinity of fermi surface @xmath406 $ ] , where terms of both types indicated above will appear , and ( ii ) the tail @xmath407 $ ] , where a simple series in @xmath408 is generated , @xmath409 extracting the logarithmic contribution from the second term of equation ( [ split ] ) , we have @xmath410 the leading approximation to @xmath206 , @xmath411 } \ , , \label{df0}\ ] ]
is obtained by setting @xmath412 everywhere except within the first logarithmic term of eq .
( [ df ] ) .
the first term inside the square brackets is the usual bcs contribution and is clearly dominant if the interaction is weak and localized in momentum space around the fermi surface .
if as is the case for nuclear forces the localization requirement is not fulfilled , the contribution from the second term of eq .
( [ df0 ] ) must also be taken into account .
one can then derive meaningful results even for potentials with positive @xmath413 .
the approximant ( [ df0 ] ) is offered as a useful generalization of the bcs weak - coupling formula to a broad class of pairing interactions . the first correction and higher corrections to the formula ( [ df0 ] )
may be developed as follows . near the fermi surface , the denominators appearing in eq .
( [ df ] ) are simplified by noting that @xmath414\ ] ] can be expanded in powers of @xmath415/(\epsilon^2 + d_f^2 ) $ ] .
this is a small quantity : for @xmath416 it is of order @xmath417 , while for @xmath418 we can make the expansion @xmath419 where the subscript 0 stands for @xmath420 , and use @xmath421 ( the prime indicates the derivative with respect to @xmath405 , while the subscript 0 stands for @xmath420 . )
accordingly , we invoke the expansion @xmath422 \ , .
\label{chiexpansion}\ ] ] at this point , any smooth functions in the numerators of the integral terms in eq .
( [ df ] ) may be expanded in taylor series . since ( i )
the integrations in the fourth and sixth addends on the l.h.s.of ( [ df ] ) are performed on a symmetric interval and ( ii ) the denominators contain only even powers of @xmath405 , we need only consider terms with even powers of @xmath405 in the corresponding numerators .
for example , @xmath423 @xmath424 \left[-2 + 2 \ln\frac2{d_f}\right ] + \frac38 d^4_f 4 ( \chi'_0)^2 \left[\frac23\frac1{d^2_f } - 1\right ] \cdots \ , \nonumber .\end{aligned}\ ] ] this step reveals a slight complication in the analysis .
the integral of the first term of the taylor expansion of @xmath425 vanishes , and consequently the integral of the second term of this taylor expansion is of the same order as the integral of the term of the expansion ( [ chiexpansion ] ) containing @xmath426 are of the same order in @xmath206 . in evaluating corrections to the result ( [ df0 ] ) , it is convenient to define @xmath427 and @xmath428 the first correction to @xmath429 is then determined by applying recursion relations involving the @xmath430 ( see ref .
( [ split ] ) is written as @xmath431 where @xmath432 { \rm d}\epsilon \nonumber \\ & \quad & -\frac14 a_1 + \frac12 a_3 + \frac14 a_5 + 1 \ , , \end{aligned}\ ] ] and the improved estimate is @xmath433\ ] ] follows .
a calculation for the argonne @xmath305 interaction illustrates the utility and accuracy of the leading approximation ( [ df0 ] ) to the energy gap @xmath434 within the small - parameter expansion . in fig .
( [ fig : spapprox ] ) , the @xmath60 dependence of @xmath6 as given by the zeroth - order approximation in @xmath206 is compared with the essentially exact result obtained by iteration of eqs .
( [ chi ] ) and ( [ deltaf ] ) .
free single - particle energies ( @xmath335 ) are assumed .
the estimate ( [ df0 ] ) produces excellent results , especially at densities beyond the peak of the @xmath6 curve .
the same strategy of constructing approximants to the gap problem through small - parameter expansion is applicable to pairing in higher angular momentum states and in particular to @xmath2p@xmath3 pairing . by way of illustration , consider the type 1 ( @xmath435 ) solution of the uncoupled - channel problem .
the transformed gap equation ( [ mtwo ] ) associated with this case has a form similar to that for @xmath0s@xmath1 pairing , with differences arising from the anisotropy of the p - wave problem .
a weight factor @xmath436 appears in the integration over angles in eq .
( [ mtwo ] ) , and @xmath437 appears in place of @xmath198 inside the square - root energy denominator .
doing the @xmath22 integral first , we can directly exploit results and relations of sects .
[ sec : nocoupling ] and [ sec : aaa ] ( specifically , the first of eqs .
( [ base2 ] ) along with eqs .
( [ int10 ] ) and ( [ sss ] ) ) .
ignoring corrections proportional to @xmath438 , the only additional term that needs to be considered , relative to the angle - average approximation , is @xmath439\ , .\ ] ] for the type 2 solution ( with @xmath259 ) , the additional term has the slightly different expression ( see the last of eqs.([base2 ] ) ) @xmath440 \,.\ ] ] the most salient points are that both these contributions are of order unity relative to the dominant logarithm in the gap and both are negative .
we may thus conclude that the true gap will be smaller than that given by the angle - average approximation and that the correction to the gap value calculated in this approximation will generally be small .
g. baym and c. j. pethick , ann .
sci . * 25 * ( 1975 ) 27 ; + g. baym and c. j. pethick , ann .
astrophys .
* 17 * ( 1979 ) 415 ; + c. j. pethick and d. g. ravenhall , ann . rev . nucl .
* 45 * ( 1995 ) 429 .energy gap @xmath6 for @xmath2p@xmath3@xmath2f@xmath3 pairing in neutron matter at several values of the fermi momentum , for the argonne @xmath32 and @xmath305 interactions and the opeg potential .
free single - particle energies ( @xmath335 ) are used , except for the last column , where @xmath363 . [ cols="<,<,<,<,<",options="header " , ] | the separation method developed earlier by us [ nucl . phys .
* a598 * 390 ( 1996 ) ] to calculate and analyze solutions of the bcs gap equation for @xmath0s@xmath1 pairing is extended and applied to @xmath2p@xmath3@xmath2f@xmath3 pairing in pure neutron matter .
the pairing matrix elements are written as a separable part plus a remainder that vanishes when either momentum variable is on the fermi surface .
this decomposition effects a separation of the problem of determining the dependence of the gap components in a spin - angle representation on the magnitude of the momentum ( described by a set of functions independent of magnetic quantum number ) from the problem of determining the dependence of the gap on angle or magnetic projection .
the former problem is solved through a set of nonsingular , quasilinear integral equations , providing inputs for solution of the latter problem through a coupled system of algebraic equations for a set of numerical coefficients .
an incisive criterion is given for finding the upper critical density for closure of the triplet gap .
the separation method and its development for triplet pairing exploit the existence of a small parameter , given by a gap - amplitude measure divided by the fermi energy .
the revised bcs equations admit analysis revealing universal properties of the full set of solutions for @xmath2p@xmath3 pairing in the absence of tensor coupling , referring especially to the energy degeneracy and energetic order of these solutions .
the angle - average approximation introduced by baldo et al . is illuminated in terms of the separation - transformed bcs problem and the small parameter expansion .
numerical calculations of @xmath2p@xmath3 pairing parameters and gap functions , with and without coupling to the @xmath2f@xmath3 state , are carried out for pairing matrix elements supplied by ( vacuum ) two - neutron interactions that fit nucleon - nucleon scattering data .
it is emphasized that _
ab initio _ evaluation of the in - medium particle - particle interaction and associated single - particle energies will be required if a reliable quantitative description of nucleonic superfluids is to be achieved .
psfig.sty , , and |
the nuclear regions of active galaxies can harbor different energetic phenomena such as starbursts ( sbs ) and active galactic nuclei ( agn ) which are responsible for producing the bright emission stemming from their central regions .
the heating of the interstellar medium ( ism ) in nuclei with different type of activity is expected to be dominated by a variety of mechanisms , such as uv , x - rays , and/or shocks , which will ultimately drive their observed chemical richness ( * ? ? ? * and references therein ) . disentangling which of this mechanisms is the main power source of galactic nuclei
is complicated by the high obscuration affecting these objects .
this fact becomes critical in the case of ultra luminous galaxies ( ulirgs ) and high - z sources .
therefore the observation of the dense molecular material has become an essential tool to get an insight in the evolution and classification of the nuclear activity in galaxies .
finding appropriate molecular species to accurately trace each of these heating mechanisms is crucial to stablish the nature of the central engine .
molecular species like hcn and hco@xmath4 , and in particular the ratios hcn / co and hcn / hco@xmath4 , have been claimed to be appropriate to differentiate between the sb and the agn contribution in galactic nuclei @xcite .
similarly , species such as hnc and cn are thought to show enhanced abundances in extremely irradiated environments @xcite .
a dichotomy still remains concerning the different evolutionary state of nuclear starburst .
while at an early stage , the heating of the ism is thought to be dominated by shocks affecting the molecular clouds fueling the starburst , the late stages of starburst are vastly dominated by the uv radiation originated in the newly formed massive stars .
this scenario has been inferred from the extensive observation of the prototype sources ngc253 and m82 in a number of key molecular tracers such as so@xmath5 , h@xmath5s , ocs @xcite , hoc@xmath4 , and co@xmath4 @xcite .
unfortunately , these tracers are generally too faint to be detected , but for the brightest prototypical nearby galaxies . in a recent observational study carried out towards a sample of molecular clouds dominated by different heating mechanisms within the galactic center region @xcite
, we have found that the cs / hnco abundance ratio is highly contrasted between molecular clouds illuminated by the uv radiation from massive star clusters and the giant molecular cloud complexes shielded from photodissociation and mostly heated by shocks .
the origin of the large changes in the cs / hnco ratio is due to the enhancement of cs in photon - dominated regions ( pdrs ) through reactions involving s@xmath4 @xcite as opposed to the highly photodissociable hnco , which is destroyed by the uv photons .
both observations of photodissociation regions within our galaxy and photochemical models show the suitability of cs as a pdr tracer @xcite .
in addition , hnco is enhanced in the presence of shocks due to its injection in the gas phase from the grain mantles @xcite .
this paper presents a follow up study of the cs / hnco ratio towards a sample of prototype galaxies with different nuclear activity .
although cs has been extensively observed in external galaxies @xcite , hnco had only been detected in four galaxies prior to this work @xcite .
observations were carried out with the iram 30 m telescope on pico veleta , spain , during three observing periods from summer 2005 through summer 2007 .
table [ tab : sourceandlines ] shows the sample of galaxies with their coordinates and the rest frequencies of the observed molecular transitions .
observations were performed in symmetrical wobbler switched mode with a frequency of 0.5hz and a beam throw of @xmath6 in azimuth . as spectrometers we used the @xmath7mhz filter banks for the 3 mm transitions and the @xmath8mhz filter banks for those at 2 and 1.3 mm .
pointing accuracy was estimated to be of @xmath9 from frequent continuum cross scans on nearby pointing sources .
data were calibrated using the standard dual load system and main beam temperatures were obtained as @xmath10 , where @xmath11 is tabulated in table [ tab : sourceandlines ] and @xmath12 are 0.95 , 0.93 , and 0.91 for 3 , 2 , and 1.2 mm , respectively .
calibration uncertainties are estimated to be @xmath13 .
the observed spectra for each source are shown in fig .
[ fig.spectra ] .
the hnco @xmath14 and @xmath15 lines were observed simultaneously with the c@xmath1o @xmath16 and @xmath17 lines , respectively .
this allows a reliable determination of the hnco relative abundance , as no uncertainties due to pointing drifts and/or calibration are involved .
for some of the sources with the larger observed linewidths , the hnco and c@xmath1o lines appear moderately blended . such is the case of ngc7469 , ngc1068 , and arp220 .
gaussian shapes have been fitted to the observed profiles .
table [ tab.gaussfit ] compiles the derived fitted parameters . for the non - detected transitions ,
upper limits are tabulated where it was considered a @xmath18 level and an approximate linewidth similar to those of the detected cs and hnco transitions , or that of c@xmath1o in the case of ngc5194 .
rotational temperatures and molecular column densities have been derived assuming local thermodynamic equilibrium ( lte ) and optically thin emission .
as derived for cs @xmath19 in ngc253 @xcite we only might expect the emission of cs @xmath20 to be significantly optically thick . however , our observation of c34s will be less affected by the opacity effects .
we have accounted for the dilution of the source in the telescope beam by assuming an averaged emission extent of @xmath21 for all sources .
this simplification might affect the determination of the column densities for sources much smaller than @xmath21 ( see * ? ? ?
* ) , but the abundance ratios between species dealt with in this paper ( sect .
[ sec.csvshnco ] ) will not be significantly affected .
thus , the main source of uncertainty will be due to possible different emission extents between molecules .
table [ tab.coldens ] shows the column density and rotational temperature derived for every source and species . from the lte analysis , we derive c@xmath1o rotational temperatures ( @xmath22 ) of @xmath23k for most of the galaxies except for m82 where we determine @xmath22=15k .
the temperatures derived from the hnco line emission of @xmath24k are a factor of @xmath25 higher than those derived from c@xmath1o .
whenever only one transition was observed , the average rotational temperature of 10k was assumed .
we have also included the available observations towards ngc253 to derive its physical parameters shown in table [ tab.coldens ] @xcite .
note that the c@xmath1o transitions in ngc253 were observed @xmath26 north of the position shown in table [ tab : sourceandlines ] .
for ngc4945 we used the observations by @xcite .
not all sources in the sample were observed in the cs @xmath20 transition . the upper limit to the emission to this transition in ngc6946 @xcite allows to constrain the cs column density to @xmath27 .
we have thus used the cs @xmath19 transitions available on ngc6946 and ngc5194 @xcite to estimate the column densities presented in table [ tab.coldens ] . note that for ngc6946 we used the observed offset ( @xmath28 ) position in @xcite , which is located just @xmath29 away from our nominal position .
similarly , their ngc5194 ( @xmath30 ) position is @xmath31 away from our observed position .
the critical densities for cs and hnco transitions observed in this works are expected to be very similar since the einstein coefficients and collisional cross sections are estimated to be similar @xcite .
thus , the hnco / cs ratio should reflect a reliable abundances ratio . in any case
, the possible uncertainties introduced by slightly different critical densities will never account for the large difference in the abundance ratios measured in different galaxies .
the differences in the derived molecular abundances might also be affected by molecular excitation differences of each individual source .
however the detailed multitransition study and lvg modeling of dense gas tracers such as hcn and hco+ in a sample of galaxies by @xcite , shows how for several molecular transitions with similar critical densities , the derived abundances from the line intensities are really determined by real differences in abundances and not by excitation effects .
the low excitation temperatures derived from c@xmath1o indicates that a substantial fraction of the column density will be dominated by relatively low density gas . since in our diagnostic diagrams in fig .
[ fig.cshnco ] are normalized to the total column densities derived from c@xmath1o , changes in the h@xmath32 total column densities will not change the main conclusions of the paper .
the sample of sources observed in this paper has been selected among the brightest prototypes for the different types of nuclear activity .
their distances and corresponding linear scales are shown table [ tab : sourceandlines ] .
the sources can be grouped as follows . _
m82 _ and _ ngc253 _ are the strongest and richest extragalactic molecular sources .
these are the archetypes of starburst galaxies housing the two brightest extragalactic iras sources @xcite .
m82 , as opposed to the young starburst in ngc253 , is claimed to be the prototype of evolved starburst .
this is supported by its observed chemistry characterized by low abundance of complex molecules such as sio , ch@xmath33oh @xcite and large abundance of molecular ions like co@xmath4 , hoc@xmath4 , and h@xmath33o@xmath4 which are expected to be enhanced in pdrs @xcite . in this work
we did not observe the central position towards m82 , but a position in the north - east molecular complex ( hereafter m82@xmath34 ) where most of the photodissociation regions are located as observed in the hco emission high - resolution maps @xcite .
_ maffei2 _ , is another well studied starburst spiral galaxy with a high molecular gas concentration in its nuclear region and showing traces of a tidal interaction with a dwarf companion galaxy @xcite .
_ m83 _ is a nearby face on barred galaxy undergoing strong nuclear starburst likely due to gas inflow along its bar @xcite .
_ ngc6946 _ shows a moderate starburst in its nucleus also claimed to be related to the presence of a bar @xcite . _
ngc1068 _ is a nearby luminous infrared galaxy ( lirg ) prototype of a seyfert 2 nucleus @xcite .
the agn is enclosed by a circumnuclear starburst ring with @xmath35 ( @xmath36kpc ) radius @xcite .
thus , we have observed two positions in ngc1068 .
one towards the central agn and an offset position @xmath37 towards a peak of emission within the circumnuclear ring , hereafter ngc1068@xmath34 , mostly tracing the starburst .
_ ngc4945 _ , being one of the three brightest iras point sources , has been the target of some a detailed molecular study by @xcite .
this nearly edge - on spiral galaxy harbor a heavily obscured seyfert 2 nucleus @xcite also surrounded by a starburst ring @xcite .
_ ngc5194 _ ( m51 ) is a nearby grand - design spiral galaxy with a seyfert 2 nucleus .
the proximity of this galaxy as well as its nearly face - on orientation has made it the target of numerous large scale multi - wavelength studies @xcite .
its interaction with the companion galaxy ngc5195 seems to have been the origin of a past period of intense star formation @xcite .
_ ngc7469 _ , the only example of seyfert 1 nucleus in our sample , also shows a circumnuclear star - forming ring of @xmath38 in diameter @xcite . _
arp220 _ is one of the best studied nearby ulirgs @xcite with more than a 95@xmath39 of its luminosity radiated at ir / submm wavelengths @xcite . at a distance of d@xmath40mpc ( @xmath41 ) and star formation rates of 300@xmath42yr@xmath43 , arp220 is often referred as a nearby template of the luminous star - forming galaxies at high - z .
arp220 is the result of an advanced merger system with two nuclei separated by @xmath29 ( @xmath44pc ) as seen from near - ir and radio images @xcite with an enormous concentration of molecular gas within its central region ( a few @xmath45 , * ? ? ?
* ) . we refer to _ ic342 _ is as a `` normal '' galaxy as being the closest spiral galaxies resembling our own galaxy .
it shows a minispiral structure meeting at the inner molecular ring surrounding the central stellar cluster @xcite .
chemical differences have been observed between the gas in the nuclear region , affected by the intense star formation through the radiation from the central cluster , and the molecular material in the spiral structure , mainly affected by shocks @xcite . in this work
we have observed two positions , one at the central region and the other at the offset position @xmath46 , hereafter _ ic342@xmath34 _
, located just in one of the arms of the minispiral .
we have derived the abundances of cs , c@xmath0s , and hnco relative to h@xmath5 . in order to estimate the h@xmath5 column density we have used the observed c@xmath1o emission , assuming a @xmath49o/@xmath1o abundance ratio of 150 , as derived in ngc253 @xcite and the standard co / h@xmath5 ratio of @xmath50 .
the cs fractional abundance shows a narrow range of values within @xmath51 .
similarly the c@xmath0s abundances are concentrated in the range of @xmath52 .
there are only three exceptions to this trend , namely ic342@xmath34 , arp220 , and ngc5194 .
the abundances of cs observed towards ic342@xmath34 is a factor of 6 below the average in the sample , while the c@xmath0s abundance is close to the average value .
this is explained by the highly concentrated cs emission towards the nuclear region @xcite and the almost twice smaller beam size at the cs@xmath14 transition frequency compared to that at the c@xmath0s@xmath53 line ( see table [ tab : sourceandlines ] ) .
we did not detect neither cs nor c@xmath0s emission towards ngc5194 . while the c@xmath0s upper limit is still within
the average value obtained for the other galaxies , the limit we derive in the main isotopologue is a factor of 3 below the sample average . on the other hand ,
the galaxy arp220 is observed to have an abundance a factor of @xmath54 over the average in both isotopologues .
this points out a real overabundance of cs towards this ulirg .
[ fig.csc34s ] shows the c@xmath2s / c@xmath0s abundance ratio versus the fractional abundance of cs .
this ratio is calculated with the 1 mm cs @xmath14 and the 2 mm c@xmath0 @xmath53 transitions which makes it dependent on the assumed rotational temperature to derive the column densities , and to a lesser extent ( up to a 30% for the smallest sources ) on the considered source size .
however , the ratio derived for ngc4945 and ngc253 agree within a factor of 1.5 with the values derived from the complete study of cs by @xcite and @xcite , respectively .
although the effects of line opacity have not been taken into account , we can consider the derived c@xmath2s / c@xmath0s abundance ratio as a rough representation of the @xmath55s isotopic ratio .
it appears that almost all the sources in the sample have ratios around @xmath3 , in agreement with those previously derived for ngc4945 and ngc253 , and well below the value of @xmath56 measured in the galactic disk @xcite .
this result supports the idea of an @xmath0s overproduction in the nuclei of galaxies by massive stars .
furthermore , we observe this ratio to be independent of the type of nuclear activity , both in agn and sb dominated galaxies . in fig .
[ fig.cshnco ] , we represent the fractional abundances of cs and c@xmath0s versus those of hnco .
we observe that most of the galaxies in the sample are observed with hnco fractional abundances ranging from 10@xmath57 to 10@xmath58 , but for two sources , namely ngc5194 and m82 , showing remarkable low hnco abundances .
for these sources we derived fractional abundance below the average derived from the rest of galaxies in our sample by a factor of @xmath59 and @xmath60 , respectively .
we do not observe major differences in the derived abundances among the agns and sbgs in our sample .
this homogeneity suggests that the type of nuclear activity does not play a major role at large scales in the production / destruction of these species in the bulk of the molecular gas .
this effect is illustrated by the two positions observed in ngc1068 , where the hnco and @xmath0cs abundances decrease by less than a factor of two from the nuclear agn to the sb ring .
of course we have to bear in mind that the two regions are not fully resolved by the beam of the iram 30 m telescope at these frequencies .
if , as suggested by the observations in the galactic center region @xcite , the abundance of these species is mostly driven by the presence of shocks and strong uv radiation fields in the ism , then the observed differences are likely linked to star formation activity in these galaxies .
significant differences are observed within the galaxies ic342 and maffei2 . in ic342 , the hnco / c@xmath1o ratio increases towards the position in the minispiral and decreases in the central region where ism is illuminated by the uv radiation from the central clusters .
these chemical differences , smoothed to the @xmath61 single - dish beam size , are much more contrasted in the high - resolution interferometric maps by @xcite . although not spatially , we resolve the nuclear region of maffei2 by the two velocity components . as seen from the high - resolution maps of @xmath62co @xcite the lower velocity component ( @xmath63kms@xmath43 )
is associated to the nuclear and northeastern molecular arm , while the higher velocity component ( 0kms@xmath43 ) will be mostly associated to the nuclear and southwest molecular arm .
we clearly see a difference in the chemistry of both components with an increase of the hnco abundance towards the southwest molecular arm of a factor of @xmath64 as well as a decrease of the cs abundance of a factor of @xmath65 .
overall , the hnco / cs ratio increases by almost an order of magnitude
. we can compare the ratios measured in galaxies with the abundances derived in the central region of the milky way @xcite .
the vast majority of galaxies show hnco abundances intermediate between those found in the shock dominated giant molecular clouds ( gmcs ) and the galactic pdr prototypes ( see dashed lines in fig [ fig.cshnco ] ) .
two sources in our sample , namely m83 and ic342@xmath34 clearly fall in the range of abundances found in gmcs . on the contrary
, m82 represents the other extreme where hnco emission is utterly vanished . the upper limit to the detection derived after a deep 5.5 hours integration on the hnco @xmath14 is an order of magnitude lower than the previous limit to the detection in the central position @xcite , but still hnco remains undetected .
interferometric maps with the bima array shows no emission anywhere in the nuclear region of m82 @xcite .
this non - detection of hnco supports the idea that m82 is highly pervaded by a strong uv photodissociating radiation and depleted of dense molecular as also suggested by the low abundance of ch@xmath33oh and nh@xmath33 @xcite .
we would expect to observe a higher abundance of cs in this ne position where photodissociation is claimed to be dominant @xcite .
however , we observe just a slightly higher cs abundance than in the whole sample , and not even that if we consider the isotopologue c@xmath0s . altogether with m82 ,
the upper limit to the hnco abundance derived for ngc5194 are similar to those derived in the galactic center pdrs .
most of the molecular studies of ngc5194 so far are restricted to the emission of co and hcn ( * ? ? ? * and references therein ) .
the results from this work point out the possible similarities between the chemistries of ngc5194 and m82 , both with very low abundances of hnco as compared with the rest of galaxies in this sample .
this result agrees with the idea of ngc5194 being in a post starburst stage after the star formation event triggered by its interaction with ngc5195 @xcite .
additionally , the hnco and cs abundances in ngc253 and ngc1068@xmath34 , a factor of 3 below the average , as well as the significantly high c@xmath0s abundance , could be explained by photodissociation .
this would suggest that the uv radiation from the evolved sb plays a significant role in the chemistry of the nuclear ism in these galaxies .
therefore ngc253 would contain a significant pdr molecular component , similar to what is observed in m82 . however , the amounts of dense molecular gas shielded from the uv radiation is still one order of magnitude larger than that observed in m82 .
our observations of hnco confirm that this molecule is one of the best tracers to study the evolutionary state of nuclear starbursts in galaxies because of the high contrast shown between the uv pervaded evolved starbursts and the dense and well shielded molecular material in the early stages of star formation .
this molecule shows a higher abundance variation in galaxies than other dense molecular gas tracer such as methanol @xcite . on the other hand , cs
does not show a significant variation among the observed sources .
this might be due to the fact of cs is not only enhanced in regions with strong uv radiation , but its emission also increases towards the densest molecular cores due to its high critical density .
it is interesting to note that the similar fuv field derived by @xcite towards the nucleus of m83 and ngc5194 might explain the different abundances observed in our sample mostly as a consequence of the dense molecular fuel depletion and the molecular cloud structure in their nuclei , and not so much by the differences in the photodissociation fields affecting the nuclear ism .
however , the cs variations in the galactic center clouds are more moderate ( a factor of @xmath66 , * ? ? ?
* ) than those of hnco , and therefore it is likely to be averaged out over the central several hundred parsecs covered by the single dish beam .
the highly contrasted variation of hnco abundances over the galaxy sample in this work allows us to trace the dense molecular gas available for fueling the nuclear starburst of these galaxies .
this is equivalent to defining a chemical timeline in the evolution of the starburst event that could be read from right to left hand side in fig .
[ fig.cshnco ] .
it is important to note that the time scales for the chemical processess due to shocks and photodissociation by uv radiation ( @xmath67yr , * ? ? ?
* ) are extremely short as compared with the typical time scale of the starburst phenomenon ( @xmath68yr , * ? ? ?
* ) . in this scenario , at the onset of the starburst ,
large amounts of dense molecular material are driven and compressed into the nuclear region of the galaxy unleashing the star formation .
high abundances of hnco are injected into the gas - phase within these clouds due to shocks , likely as a consequence of the cloud - cloud collisions , as observed in the nuclear spiral arms of ic342 @xcite . from our galaxy sample
, m83 is shown to have the highest abundances of hnco , and similar to those observed in the galactic center gmcs .
subsequently , as soon as the first newly massive star clusters are formed , uv illumination begins to come on stage , photodissociating the material surrounding them .
the hnco abundance in these regions will be quickly dissociated where the cs molecule will be formed .
a sort of average equilibrium will be reached between the pdr and the dense gas contribution as long as the infall of molecular fuel keeps feeding the starburst .
this is the status in which most of the starburst galaxies in our sample are observed .
the agn dominated galaxies we have observed also show this average cs / hnco abundances similar to the intermediate state starbursts , suggesting the presence of a significant uv radiation field .
once the galaxy gets closer to the end the short few 10myr starburst period , the fuel supply is cut off or stopped by superwinds . from our comparative study , this is apparently the point at which ngc253 stands , where photodissociation becomes more important and the fuel reservoir is being consumed by the last gasp of star formation .
finally , as we observe in m82 and ngc5194 , most of the molecular gas is pervaded by the uv radiation field with barely no dense uv shielded molecular material left .
the galaxy remains then waiting for a likely next burst of star formation @xcite .
indeed , chemical enrichment in m82 points out to recursive starburst periods @xcite .
from our study of the cs and hnco abundances in a selected sample of extragalactic sources , we have found differences of nearly two orders of magnitude in the hnco / cs abundance ratios among the starburst galaxies .
these differences are in agreement with the results from the observations of galactic center clouds dominated by various heating mechanisms suggesting that the observed hnco abundances could be related to the evolutionary stage of their nuclear starburst .
shocks affecting the massive dense molecular clouds in the early stages and strong uv fields in the evolved stages dominate the hnco abundance in galaxies .
we do not find significant differences in the molecular abundances of hnco in galaxies with different type of nuclear activity ( i.e. agn or sb ) which implies this species are not affected differently by these processes at the scales observed by single dish beam telescopes .
as derived from the hnco abundances we can point out prototypes of these nuclear starburst evolutionary stages .
m83 and the nuclear spiral arms in ic342 would be representative of the early stage , while m82 and ngc5194 nuclei would be in the later stages of evolution . ngc253 , claimed to be a prototype of early starburst , shows abundances below the average suggesting an important influence of photodissociation in the heating of its ism , and therefore chemically closer to m82 than to m83 .
the @xmath2s/@xmath0s isotopic ratio , derived from the c@xmath2s / c@xmath0s abundance ratio , is found to be @xmath3 , and thus similar for all the observed sources .
this ratio , found to be well below the value of @xmath69 in the galactic disk , supports the idea of the @xmath0s isotope being enriched due to massive star formation events in the nuclear region of galaxies .
the cs emission is suggested to be a good tracer of the molecular gas affected by photodissociation in the galactic center @xcite .
we do not observe a significant contrast in our sample .
though its abundance is enhanced in uv radiated regions , its emission is also favored in high density molecular clouds .
thus the overall emission over the central few hundred parsecs of galactic nuclei is averaged out , and no significant trends in the cs abundances is found in the observed galaxies .
aalto , s. , polatidis , a. g. , httemeister , s. , & curran , s. j. 2002 , , 381 , 783 aalto , s. , spaans , m. , wiedner , m.c . , & httemeister , s. 2007 , , 464 , 193 antonucci , r. r. j. , & miller , j. s. 1985 , , 297 , 621 baan , w. a. , & haschick , a. d. 1995 , , 454 , 745 baan , w. a. , henkel , c. , loenen , a. f. , baudry , a. , & wiklind , t. 2008 , , 477 , 747 bergin , e. a. , neufeld , d. a. , & melnick , g. j. 1998 , , 499 , 777 boker , t. , forster - schreiber , n. m. , & genzel , r. 1997 , , 114 , 1883 braatz , j. a. , wilson , a. s. , & henkel , c. 1997 , , 110 , 321 chin , y .-
n . , henkel , c. , whiteoak , j.b . ,
langer , n. , & churchwell , e.b .
1996 , , 305 , 960 coziol , r. 1996 , , 309 , 345 dopita , m. a. , et al .
2005 , , 619 , 755 fuente , a. , garca - burillo , s. , gerin , m. , teyssier , d. , usero , a. , rizzo , j. r. , & de vicente , p. 2005
, , 619 , l155 fuente , a. , garca - burillo , s. , gerin , m. , rizzo , j. r. , usero , a. , teyssier , d. , roueff , e. , & le bourlot , j. 2006 , , 641 , l105 garca - burillo , s. , martn - pintado , j. , fuente , a. , usero , a. , & neri , r. 2002 , , 575 , l55 goicoechea , j. r. ; pety , j. ; gerin , m. et al .
2006 , a&a , 456 , 565 greenawalt , b. , walterbos , r. a. m. , thilker , d. , & hoopes , c. g. 1998 , , 506 , 135 harrison , a. , henkel , c. , & russell , a. 1999 , , 303 , 157 helfer , t. t. , thornley , m. d. , regan , m. w. , wong , t. , sheth , k. , vogel , s. n. , blitz , l. , & bock , d. c .- j .
2003 , , 145 , 259 hurt , r. l. , turner , j. l. , & ho , p. t. p. 1996
, , 466 , 135 httemeister , s. , mauersberger , r. , & henkel , c. 1997 , , 326 , 59 ishizuki , s. , kawabe , r. , ishiguro , m. , okumura , s. k. , & morita , k .- i . 1990 , , 344 , 224 kramer , c. , mookerjea , b. , bayet , e. , garcia - burillo , s. , gerin , m. , israel , f. p. , stutzki , j. , & wouterloot , j. g. a. 2005 , , 441 , 961 kohno , k. , kawabe , r. , tosaki , t. , & okumura , s. k. 1996 , , 461 , l29 kohno , k. , matsushita , s. , vila - vilar , b. , okumura , s. k. , shibatsuka , t. , okiura , m. , ishizuki , s. , & kawabe , r. 2001 , asp conference proceedings vol . 249 : the central kiloparsec of starbursts and agn : the la palma connection `` , 672 kohno , k. 2005 , aip conference proceedings 783 ' ' the evolution of starburst : the 331st wilhelm and else heraeus seminar " , 203 krips , m. , neri , r. , garca - burillo , s. , martn , s. , combes , f. , graci - carpio , j. , & eckart , a. 2008 , , 677 , 262 maiolino , r. , risaliti , g. , & salvati , m. 1999 , , 341 , l35 marconi , a. , oliva , e. , van der werf , p. p. , maiolino , r. , schreier , e. j. , macchetto , f. , & moorwood , a. f. m. 2000 , , 357 , 24 martn , s. , mauersberger , r. , martn - pintado , j. , garca - burillo , s. , & henkel , c. 2003 , , 411 , l465 martn , s. , martn - pintado , j. , mauersberger , r. , henkel , c. , & garca - burillo , s. 2005 , , 620 , 210 martn , s. , martn - pintado , j. , mauersberger , r. 2006a , , 450 l13 martn , s. , mauersberger , r. , martn - pintado , j. , & henkel , c. , garca - burillo , s. 2006b , , 164 , 450 martn , s. , requena - torres , m. a. , martn - pintado , j. , & mauersberger , r. 2008 , , 678 , 245 mason , a. m. , & wilson , c. d. 2004 , , 612 , 860 mauersberger , r. , henkel , c. , wilson , t.l . , & harju , j. 1989 , a&a , 226 , l5 mauersberger , r. & henkel , c. 1993 , rev .
modern astron . 6 , 69 mauersberger , r. , henkel , c. , wei , a. , peck , a.b . , hagiwara , y. 2003 , , 403 , 561 meier , d. s. , & turner , j. l. 2005 , , 618 , 259 meier , d. s. , turner , j. l. , & hurt , r. l. 2008 , , 675 , 281 myers , s. t. , & scoville , n. z. 1987 , , 312 , l39 nguyen - q - rieu , henkel , c. , jackson , j. m. , & mauersberger , r. 1991 , , 241 , l33 origlia , l. , ranalli , p. , comastri , a. , & maiolino , r. 2004 , , 606 , 862 petitpas , g. r. , & wilson , c. d. 1998 , , 503 , 219 rekola , r. , richer , m. g. , mccall , m. l. , valtonen , m. j. , kotilainen , j. k. , & flynn , c. 2005 , , 361 , 330 sakamoto , k. , scoville , n. z. , yun , m. s. , crosas , m. , genzel , r. , & tacconi , l. j. 1999 , , 514 , 68 sanders , d. b. , mazzarella , j. m. , kim , d .-
, surace , j. a. , & soifer , b. t. 2003 , , 126 , 1607 schinnerer , e. , eckart , a. , tacconi , l. j. , genzel , r. , & downes , d. 2000 , , 533 , 850 schinnerer , e. , bker , t. , emsellem , e. , & downes , d. 2007 , , 462 , l27 schier , f. l. , van der tak , f. f. s. , van dishoeck , e. f. , & black , j. h. 2005 , , 432 , 369 schuster , k. f. , kramer , c. , hitschfeld , m. , garcia - burillo , s. , & mookerjea , b. 2007 , , 461 , 143 scoville , n. , & young , j. s. 1983 , , 265 , 148 scoville n. z. et al . ,
2000 , , 119 , 991 scoville , n. z. , polletta , m. , ewald , s. , stolovy , s. r. , thompson , r. , & rieke , m. 2001 , , 122 , 3017 soifer , b.t . ,
boehmer , l. , neugebauer , g. , & sanders , d.b .
1989 , aj , 98 , 766 sternberg , a. , & dalgarno , a. 1995 , apjs , 99 , 565 talbot , r. j. , jr . , jensen , e. b. , & dufour , r. j. 1979 , , 229 , 91 takano , s. , nakai , n. , & kawaguchi 2002 , pasj , 54 , 195 thronson , h. a. , jr . , rubin , h. , & ksir , a. 1991 , , 252 , 550 turner , j. l. , & ho , p. t. p. 1983
, , 268 , l79 turner , j. l. , & meier , d. s. 2008 , , 313 , 267 usero , a. , garca - burillo , s. , martn - pintado , j. , fuente , a. , & neri , r. 2006 , , 448 , 457 van der tak , f. f. s. , aalto , s. , & meijerink , r. 2008 , , 477 , l5 wang , m. , henkel , c. , chin , y .-
n . , whiteoak , j. b. , hunt cunningham , m. , mauersberger , r. , & muders , d. 2004 , , 422 , 883 wilson , a. s. , helfer , t. t. , haniff , c. a. , & ward , m. j. 1991 , , 381 , 79 zinchenko , i. , henkel , c. , & mao , r.q .
2000 , , 361 , 1079 o @xmath16 and c@xmath0s @xmath53 spectra have been scaled up for the sake of clarity in the profile comparison .
spectra with line intensities below 10mk have been resampled to a velocity resolution of 20kms@xmath43 , otherwise original velocity resolution is shown .
c@xmath0s@xmath70 in ngc1068 and c@xmath0s@xmath71 in arp220 are shown with a 30kms@xmath43velocity resolution .
gaussian fits are shown for all detected transitions .
dotted fits indicate individual velocity components . for arp220 , where c@xmath1o and hnco transitions appear significantly blended
, we show both fits in the same spectrum where the overall fit is shown with a continuous line and the individual transition with dotted lines .
temperature scale is @xmath72.,title="fig : " ] o @xmath16 and c@xmath0s @xmath53 spectra have been scaled up for the sake of clarity in the profile comparison .
spectra with line intensities below 10mk have been resampled to a velocity resolution of 20kms@xmath43 , otherwise original velocity resolution is shown .
c@xmath0s@xmath70 in ngc1068 and c@xmath0s@xmath71 in arp220 are shown with a 30kms@xmath43velocity resolution .
gaussian fits are shown for all detected transitions .
dotted fits indicate individual velocity components . for arp220 , where c@xmath1o and hnco transitions appear significantly blended , we show both fits in the same spectrum where the overall fit is shown with a continuous line and the individual transition with dotted lines .
temperature scale is @xmath72.,title="fig : " ] o @xmath16 and c@xmath0s @xmath53 spectra have been scaled up for the sake of clarity in the profile comparison .
spectra with line intensities below 10mk have been resampled to a velocity resolution of 20kms@xmath43 , otherwise original velocity resolution is shown .
c@xmath0s@xmath70 in ngc1068 and c@xmath0s@xmath71 in arp220 are shown with a 30kms@xmath43velocity resolution .
gaussian fits are shown for all detected transitions .
dotted fits indicate individual velocity components .
for arp220 , where c@xmath1o and hnco transitions appear significantly blended , we show both fits in the same spectrum where the overall fit is shown with a continuous line and the individual transition with dotted lines .
temperature scale is @xmath72.,title="fig : " ] o @xmath16 and c@xmath0s @xmath53 spectra have been scaled up for the sake of clarity in the profile comparison .
spectra with line intensities below 10mk have been resampled to a velocity resolution of 20kms@xmath43 , otherwise original velocity resolution is shown .
c@xmath0s@xmath70 in ngc1068 and c@xmath0s@xmath71 in arp220 are shown with a 30kms@xmath43velocity resolution .
gaussian fits are shown for all detected transitions .
dotted fits indicate individual velocity components . for arp220 , where c@xmath1o and hnco transitions appear significantly blended , we show both fits in the same spectrum where the overall fit is shown with a continuous line and the individual transition with dotted lines .
temperature scale is @xmath72.,title="fig : " ] o @xmath16 and c@xmath0s @xmath53 spectra have been scaled up for the sake of clarity in the profile comparison .
spectra with line intensities below 10mk have been resampled to a velocity resolution of 20kms@xmath43 , otherwise original velocity resolution is shown .
c@xmath0s@xmath70 in ngc1068 and c@xmath0s@xmath71 in arp220 are shown with a 30kms@xmath43velocity resolution .
gaussian fits are shown for all detected transitions .
dotted fits indicate individual velocity components . for arp220 , where c@xmath1o and hnco transitions appear significantly blended , we show both fits in the same spectrum where the overall fit is shown with a continuous line and the individual transition with dotted lines .
temperature scale is @xmath72.,title="fig : " ] o @xmath16 and c@xmath0s @xmath53 spectra have been scaled up for the sake of clarity in the profile comparison .
spectra with line intensities below 10mk have been resampled to a velocity resolution of 20kms@xmath43 , otherwise original velocity resolution is shown .
c@xmath0s@xmath70 in ngc1068 and c@xmath0s@xmath71 in arp220 are shown with a 30kms@xmath43velocity resolution .
gaussian fits are shown for all detected transitions .
dotted fits indicate individual velocity components . for arp220 , where c@xmath1o and hnco transitions appear significantly blended , we show both fits in the same spectrum where the overall fit is shown with a continuous line and the individual transition with dotted lines .
temperature scale is @xmath72.,title="fig : " ] o @xmath16 and c@xmath0s @xmath53 spectra have been scaled up for the sake of clarity in the profile comparison .
spectra with line intensities below 10mk have been resampled to a velocity resolution of 20kms@xmath43 , otherwise original velocity resolution is shown .
c@xmath0s@xmath70 in ngc1068 and c@xmath0s@xmath71 in arp220 are shown with a 30kms@xmath43velocity resolution .
gaussian fits are shown for all detected transitions .
dotted fits indicate individual velocity components . for arp220 , where c@xmath1o and hnco transitions appear significantly blended , we show both fits in the same spectrum where the overall fit is shown with a continuous line and the individual transition with dotted lines .
temperature scale is @xmath72.,title="fig : " ] o @xmath16 and c@xmath0s @xmath53 spectra have been scaled up for the sake of clarity in the profile comparison .
spectra with line intensities below 10mk have been resampled to a velocity resolution of 20kms@xmath43 , otherwise original velocity resolution is shown .
c@xmath0s@xmath70 in ngc1068 and c@xmath0s@xmath71 in arp220 are shown with a 30kms@xmath43velocity resolution .
gaussian fits are shown for all detected transitions .
dotted fits indicate individual velocity components . for arp220 , where c@xmath1o and hnco transitions appear significantly blended , we show both fits in the same spectrum where the overall fit is shown with a continuous line and the individual transition with dotted lines .
temperature scale is @xmath72.,title="fig : " ] o @xmath16 and c@xmath0s @xmath53 spectra have been scaled up for the sake of clarity in the profile comparison .
spectra with line intensities below 10mk have been resampled to a velocity resolution of 20kms@xmath43 , otherwise original velocity resolution is shown .
c@xmath0s@xmath70 in ngc1068 and c@xmath0s@xmath71 in arp220 are shown with a 30kms@xmath43velocity resolution .
gaussian fits are shown for all detected transitions .
dotted fits indicate individual velocity components . for arp220 , where c@xmath1o and hnco transitions appear significantly blended , we show both fits in the same spectrum where the overall fit is shown with a continuous line and the individual transition with dotted lines .
temperature scale is @xmath72.,title="fig : " ] o @xmath16 and c@xmath0s @xmath53 spectra have been scaled up for the sake of clarity in the profile comparison .
spectra with line intensities below 10mk have been resampled to a velocity resolution of 20kms@xmath43 , otherwise original velocity resolution is shown .
c@xmath0s@xmath70 in ngc1068 and c@xmath0s@xmath71 in arp220 are shown with a 30kms@xmath43velocity resolution .
gaussian fits are shown for all detected transitions .
dotted fits indicate individual velocity components . for arp220 , where c@xmath1o and hnco transitions appear significantly blended , we show both fits in the same spectrum where the overall fit is shown with a continuous line and the individual transition with dotted lines .
temperature scale is @xmath72.,title="fig : " ] o @xmath16 and c@xmath0s @xmath53 spectra have been scaled up for the sake of clarity in the profile comparison .
spectra with line intensities below 10mk have been resampled to a velocity resolution of 20kms@xmath43 , otherwise original velocity resolution is shown .
c@xmath0s@xmath70 in ngc1068 and c@xmath0s@xmath71 in arp220 are shown with a 30kms@xmath43velocity resolution .
gaussian fits are shown for all detected transitions .
dotted fits indicate individual velocity components . for arp220 , where c@xmath1o and hnco transitions appear significantly blended , we show both fits in the same spectrum where the overall fit is shown with a continuous line and the individual transition with dotted lines .
temperature scale is @xmath72.,title="fig : " ] s / c@xmath0s abundance ratio for the detected galaxies as a function of the cs fractional abundance .
sulfur isotopic ratios from the literature for the galactic disk , and the galaxies ngc4945 and ngc253 are represented with continuous , dotted and dashed lines , respectively , and their corresponding error shown as a grey shade .
labeling of sources is the same as in fig .
[ fig.cshnco ] . ] s ( _ bottom _ ) vs the fractional abundance of hnco with respect to h@xmath5 .
upper limits to the abundances are represented by arrows . as in the text ,
offset observed position are labeled with an * following the galaxy name .
the two velocity components fitted in maffei2 are labeled as maffei2a and maffei2b .
dashed lines represent the limits of the regions used to discriminate the uv dominated ( left side ) from the shock dominated ( right side ) molecular gas , as derived from galactic center observations ( see * ? ? ?
* for details ) .
] l c c c + source & @xmath73 & @xmath74 & @xmath75(mpc@xmath76pc/@xmath77 ) + & 00:47:33.3 & -25:17:23 & @xmath78 + maffei2 & 02:41:55.2 & 59:36:11 & @xmath79 + ngc1068 & 02:42:40.9 & -00:00:46 & @xmath80 + ic342 & 03:46:48.6 & 68:05:46 & @xmath81 + m82 & 09:55:51.9 & 69:40:47 & @xmath82 + _ ngc4945 _ & 13:05:27.2 & -49:28:05 & @xmath83 + ngc5194 ( m51 ) & 13:29:52.7 & 47:11:43 & @xmath84 + m83 & 13:37:00.9 & -29:51:56 & @xmath81 + arp220 & 15:34:57.1 & 23:30:12 & @xmath85 + ngc6946 & 20:34:52.4 & 60:09:14 & @xmath86 + ngc7469 & 23:03:15.6 & 08:52:26 & @xmath87 + + molecule & frequency ( ghz ) & @xmath88 & @xmath89 + c@xmath1o @xmath90 & 109.782 & 22 & 0.75 + c@xmath1o @xmath91 & 219.560 & 11 & 0.55 + cs @xmath20 & 244.935 & 10 & 0.49 + c@xmath0s @xmath91 & 96.412 & 25 & 0.76 + c@xmath0s @xmath19 & 144.617 & 17 & 0.69 + hnco @xmath92 & 109.905 & 22 & 0.75 + hnco @xmath93 & 131.885 & 19 & 0.71 + hnco @xmath94 & 219.798 & 11 & 0.55
+ l c c c c c transition & @xmath95 & @xmath96 & @xmath97 & @xmath98 & @xmath99 + & ( kkms@xmath43 ) & ( kms@xmath43 ) & ( kms@xmath43 ) & ( mk ) & ( mk ) + + c@xmath1o @xmath16 & @xmath100 & @xmath101 & @xmath102 & 18.2 & 1.9 + & @xmath103 & @xmath104 & @xmath105 & 29.0 & + c@xmath1o @xmath17 & @xmath106 & @xmath107 & @xmath108 & 28.0 & 1.0 + & @xmath109 & @xmath110 & @xmath111 & 49.8 & + cs @xmath14 & @xmath112 & @xmath113 & @xmath114 & 16.3 & 1.4 + & @xmath115 & @xmath116 & @xmath117 & 14.3 & + c@xmath0s @xmath53 & @xmath118 & @xmath63 & 65 & 6.0 & 1.4 + & @xmath119 & 0 & 110 & 3.2 & + hnco @xmath14 & @xmath120 & @xmath63 & 62 & 5.8 & 2.1 + & @xmath121 & @xmath122 & @xmath123 & 20.1 & + hnco @xmath124 & @xmath125 & & 65 & @xmath126 & 2.5 + & @xmath127 & @xmath128 & @xmath129 & 23.9 & + hnco @xmath15 & @xmath130 & @xmath63 & 65 & 1.8 & 1.0 + & @xmath131 & @xmath132 & @xmath133 & 5.9 & + + c@xmath1o @xmath16 & @xmath134 & @xmath135 & @xmath136 & 15.8 & 0.8 + c@xmath1o @xmath17 & @xmath137 & @xmath138 & @xmath139 & 9.0 & 1.9 + cs @xmath14 & @xmath140 & @xmath141 & @xmath142 & 17.6 & 3.0 + c@xmath0s @xmath53 & @xmath143 & @xmath144 & @xmath145 & 2.2 & 1.0 + hnco @xmath14 & @xmath146 & @xmath147 & @xmath148 & 3.9 & 0.8 + hnco @xmath124 & @xmath149 & @xmath150 & @xmath139 & 2.5 & 0.6 + hnco @xmath15 & @xmath151 & & 210 & @xmath152 & 1.9 + + c@xmath1o @xmath16 & @xmath153 & @xmath154 & @xmath155 & 15.0 & 1.1 + c@xmath0s @xmath53 & @xmath156 & & 170 & @xmath157 & 1.2 + hnco @xmath14 & @xmath158 & @xmath159 & @xmath160 & 2.3 & 1.1 + hnco @xmath15 & @xmath156 & & 170 & @xmath161 & 0.8 + + c@xmath1o @xmath16 & @xmath162 & @xmath163 & @xmath164 & 89.9 & 1.7 + c@xmath1o @xmath17 & @xmath165 & @xmath166 & @xmath167 & 164.1 & 1.1 + cs @xmath14 & @xmath168 & @xmath169 & @xmath170 & 40.3 & 1.8 + c@xmath0s @xmath53 & @xmath171 & @xmath172 & @xmath173 & 8.4 & 1.5 + hnco @xmath14 & @xmath174 & @xmath175 & @xmath176 & 33.0 & 1.7 + hnco @xmath124 & @xmath177 & @xmath178 & @xmath179 & 33.0 & 1.9 + hnco @xmath15 & @xmath180 & @xmath181 & @xmath182 & 8.5 & 1.1 + + c@xmath1o @xmath16 & @xmath183 & @xmath184 & @xmath185 & 78.8 & 2.0 + cs @xmath14 & @xmath186 & @xmath187 & @xmath188 & 11.0 & 2.6 + c@xmath0s @xmath53 & @xmath189 & @xmath190 & @xmath191 & 8.3 & 1.0 + hnco @xmath14 & @xmath192 & @xmath193 & @xmath194 & 36.2 & 2.0 + hnco @xmath124 & @xmath195 & @xmath196 & @xmath197 & 22.7 & 2.0 + + c@xmath1o @xmath16 & @xmath198 & @xmath199 & @xmath200 & 16.8 & 0.7 + & @xmath201 & 310 & 80 & 51.9 & + c@xmath1o @xmath17 & @xmath202 & @xmath203 & @xmath204 & 261.5 & 2.0 + cs @xmath14 & @xmath205 & @xmath206 & @xmath207 & 82.6 & 2.5 + c@xmath0s @xmath53 & @xmath208 & @xmath209 & @xmath210 & 6.0 & 1.0 + hnco @xmath14 & @xmath211 & & 80 & @xmath212 & 0.7 + hnco @xmath124 & @xmath213 & & 80 & @xmath214 & 1.3 + hnco @xmath15 & @xmath125 & & 80 & @xmath215 & 2.0 + + c@xmath1o @xmath16 & @xmath216 & @xmath217 & @xmath218 & 19.1 & 0.6 + & @xmath146 & @xmath219 & @xmath220 & 10.1 & + c@xmath0s @xmath53 & @xmath156 & & 100 & @xmath221 & 1.4 + hnco @xmath14 & @xmath222 & & 100 & @xmath161 & 0.6 + + c@xmath1o @xmath16 & @xmath223 & @xmath224 & @xmath225 & 30.9 & 1.9 + cs @xmath14 & @xmath226 & @xmath227 & @xmath228 & 26.5 & 0.5 + c@xmath0s @xmath53 & @xmath229 & @xmath230 & @xmath231 & 11.1 & 2.7 + hnco @xmath14 & @xmath232 & @xmath233 & @xmath234 & 7.7 & 1.9 + hnco @xmath124 & @xmath235 & @xmath236 & @xmath237 & 5.4 & 1.2 + + c@xmath1o @xmath16 & @xmath238 & @xmath239 & @xmath240 & 5.7 & 0.9 + cs @xmath14 & @xmath241 & @xmath242 & @xmath243 & 22.0 & 2.7 + c@xmath0s @xmath17 & @xmath244 & @xmath245 & @xmath246 & 3.4 & 1.4 + hnco @xmath14 & @xmath112 & @xmath239 & @xmath247 & 2.3 & 0.9 + + c@xmath1o @xmath16 & @xmath248 & @xmath249 & @xmath250 & 23.8 & 0.7 + & @xmath251 & @xmath252 & @xmath253 & 31.2 & + c@xmath0s @xmath53 & @xmath254 & @xmath255 & @xmath256 & 3.8 & 0.7 + hnco @xmath14 & @xmath257 & @xmath258 & @xmath259 & 5.9 & 0.7 + hnco @xmath124 & @xmath260 & @xmath261 & @xmath262 & 4.1 & 1.0 + + c@xmath1o @xmath16 & @xmath263 & @xmath264 & @xmath265 & 2.3 & 0.4 + c@xmath0s @xmath53 & @xmath266 & & 150 & @xmath267 & 0.6 + hnco @xmath14 & @xmath268 & @xmath269 & @xmath270 & 1.0 & 0.4 + lccccccc source & & cs & c@xmath0s & + & ( @xmath271 ) & ( k ) & ( @xmath272 ) & ( @xmath273 ) & ( @xmath272 ) & ( k ) + ngc253 & @xmath274 & @xmath275 & @xmath276 & @xmath277 & @xmath278 & @xmath279 + maffei2 & @xmath280 & @xmath281 & @xmath282 & @xmath283 & @xmath284 & @xmath285 + & @xmath286 & @xmath287 & @xmath288 & @xmath289 & @xmath290 & @xmath291 + ngc1068 & @xmath292 & @xmath293 & @xmath294 & @xmath295 & @xmath296 & @xmath297 + ngc1068 & @xmath298 & ... & ... & @xmath299 & @xmath300 & ... + ic342 & @xmath301 & @xmath302 & @xmath303 & @xmath304 & @xmath305 & @xmath306 + ic342 & @xmath298 & ... & @xmath244 & @xmath307 & @xmath308 & @xmath309 + m82 & @xmath310 & @xmath311 & @xmath312 & @xmath313 & @xmath314 & ... + ngc4945 & @xmath315 & @xmath316 & @xmath317 & @xmath318 & @xmath319 & ... + ngc5194 & @xmath320 & ... & @xmath321 & @xmath322 & @xmath314 & ... + m83 & @xmath323 & ... & @xmath324 & @xmath325 & @xmath296 & @xmath326 + arp220 & @xmath327 & ... & @xmath328 & @xmath329 & @xmath330 & ... + ngc6946 & @xmath331 & ... & @xmath332 & @xmath289 & @xmath333 & @xmath334 + ngc7469 & @xmath335 & ... & ... & @xmath336 & @xmath337 & ... + | the chemistry in the central regions of galaxies is expected to be strongly influenced by their nuclear activity . to find the best tracers of nuclear activity is of key importance to understand the processes taking place in the most obscured regions of galactic nuclei . in this work we present multi - line observations of cs , c@xmath0s , hnco and c@xmath1o in a sample of 11 bright galaxies prototypical for different types of activity .
the @xmath2s/@xmath0s isotopic ratio is @xmath3 , supporting the idea of an isotopical @xmath0s enrichment due to massive star formation in the nuclear regions of galaxies .
although c@xmath2s and c@xmath0s do not seem to be significantly affected by the activity type , the hnco abundance appears highly contrasted among starburst .
we observed hnco abundance variations of nearly two orders of magnitude .
the hnco molecule is shown to be a good tracer of the amount of molecular material fueling the starburst and therefore can be used as a diagnostics of the evolutionary state of a nuclear starburst . |
in 1953 watson and crick @xcite proposed the structure of dna to be a stable double - stranded helix .
the stability comes through the staking interaction and the hydrogen bonding between the base pairs in the opposite strands .
but actually this picture represents the equilibrium structure of dna under physiological conditions . as the interaction energy between these base pairs is only few @xmath0 , even at room temperature due to thermal fluctuations
locally dna strand opens up creating what is called bubbles " .
these bubbles have different sizes and lifetimes .
the creation and annihilation kinetics of these bubbles is termed as the breathing dynamics .
on increasing the temperature or changing the ph these bubbles add up to form larger bubbles and eventually the double stranded dna denatures .
in recent past single molecule fluorescence correlation spectroscopy ( fcs ) experiment by altan - bonnet et .
al @xcite has provided the first quantitative insight to the relaxation kinetics of the breathing mode of the double - stranded dna .
this breathing mode refers to local denaturation and reclosing of the double - stranded structure . in their experiment ,
two bases of the double - stranded dna , corresponding to opposite strands are tagged with a fluorophore and a quencher respectively .
so when the dna structure is closed , fluorophore and the quencher are in close proximity and the fluorescence is quenched .
but due to thermal fluctuation when the dna structure opens up creating a bubble , the fluorescence is restored .
hence the base pair fluctuation leads to a fluctuation in fluorescence intensity .
this fluctuation in fluorescence intensity is monitored by fcs which determines the characteristic dynamics of the relaxation of dynamic correlations in the fluctuation of base pairs .
they introduced a correlation function @xmath1 where @xmath2 is the fluorescent intensity at time @xmath3 .
@xmath4 was found to be multiexponential .
interestingly the correlation function for all the dna constructs , at all temperatures follow the same universal temporal behavior and when presented as a function of rescaled time they all collapse into a single universal curve @xmath5 , where @xmath6 is such that @xmath7 . to explain the experimental data they proposed a simple kinetic model in which @xmath5 becomes @xmath8 where , @xmath9 and @xmath10 is a parameter which is adjusted to @xmath11 to ensure that @xmath12 . since this experimental work by altan - bonnet et .
al , the topic bubble dynamics " as it is commonly refereed to has received a great deal of theoretical attention .
just few years after this seminal experimental study of the transient time - dependent rupture and re - healing of double stranded dna , bicout and kats @xcite proposed a kinetic scheme based on two state model ( closed or open ) of the double stranded dna .
their formulation gave an analytical expression for the survival probability , correlation function and life time for the bubble relaxation dynamics but formulation did not consider the structure of the double stranded dna at any level .
very recently srivastava and singh @xcite proposed a theoretical model where the interaction between the base pairs of the opposite dna strands is described by peyard - bishop - dauxois ( pbd ) @xcite potential and the separation between the base pairs ( y " in their notation ) follows a fokker - planck equation . thus only making the interaction realistic but still not taking into account of the dynamics of the chain and restricting to a single relevant dynamical variable ( separation y " ) description . in a paper by jeon , sung and ree @xcite double stranded dna was modeled as a duplex of semiflexible chains mutually bonded by weak interactions in other words they used an extended worm - like chain model and a langevin dynamics simulation was performed to examine the size distribution and dynamics of the bubble . shortly after this fogedby and metzler @xcite came up with a theoretical model for the bubble dynamics .
they used the poland - scheraga free energy @xcite where the free energy is a function of bubble size @xmath13 .
the dynamics of @xmath13 follows a langevin equation and the corresponding fokker - planck equation is analogous to the imaginary time schrodinger equation for a particle moving in a coulomb potential subject to a centrifugal barrier .
this mapping enabled them to calculate the correlation function . but the best fit to the long time behavior was obtained only when they considered the dynamics of @xmath13 in a linear potential . thus to reproduce the long time data
one can merely start with a linear potential but unfortunately it may not fit to the experimental data in the intermediate time range .
interestingly the shortcomings of the fogedly - metzler model was pointed out by chatterjee et .
al @xcite .
they assumed that the distance between fluorophore and the quencher follows an overdamped langevin equation in a harmonic potential .
although their theory predicts the experimental data reasonably well it does not take into account of the fluctuation of different modes of the dna strand contributing to the bubble dynamics .
a better theory should account for those and actually our model does . in our formulation
the dna - strand is described by a rouse chain @xcite confined at both the ends .
transverse displacement of the string accounts for the bubble formation .
naturally our model takes into account of the contribution of different modes of the string to the breathing dynamics .
we describe the bubble by a confined rouse chain @xcite in a harmonic potential , @xmath14 , where @xmath15 denotes the position of the @xmath16th segment of the confined rouse chain in space and @xmath3 is the time . in other words
the chain is described by a field @xmath15 . because of the thermal fluctuations the bubble undergoes brownian motion and its time development is described by the equation @xmath17 in the above
, @xmath18 is the friction coefficient for the @xmath16th segment and @xmath19 is the force constant for the confining potential which accounts for the staking interaction between the opposite strands of the dna molecule .
the fluorescent intensity at time @xmath3 , denoted by @xmath2 can be written as @xmath20 where @xmath21 denotes the position of the @xmath22th segment in space . later on
we will choose @xmath22 such that it corresponds to the center of the bubble .
as mentioned earlier in the experiment one measures the following correlation function .
it is worth mentioning that this is the the relevant dynamical coordinate / variable in our model .
obviously it is not a one dimensional phenomenological reaction coordinate / dynamical variable like the separation between the donor and the acceptor as considered by chatterjee et al .
@xcite but actually a collective dynamical variable in the sense that it has the contribution from all the normal modes of the chain as is shown later in eq.(6 ) .
@xmath1 so in order to evaluate @xmath4 we should first evaluate @xmath23 and @xmath24 . in our model the correlation function , @xmath24 would be @xmath25 then we use the fact that the step function can be expressed as an integral over delta function and in the next step we use the fourier integral representation of the delta function .
@xmath26 now @xmath15 follows rouse dynamics and it has a boundary conditions @xmath27 and @xmath28 , where @xmath29 is the chain length .
keeping these boundary conditions in mind one can express @xmath15 in terms of fourier modes @xmath30 @xmath31 should satisfy the boundary conditions @xmath32 and @xmath33 .
@xmath34 is the @xmath35th normal mode of the chain . substituting @xmath15 from eq .
( [ rnt ] ) into the expression for @xmath24 to get @xmath36 the above average is evaluated as follows first we introduce @xmath37 then the quantity in the angular bracket can be written as @xmath38 then we use the definition of the characteristic functional @xcite and write the above quantity as @xmath39 where @xmath40 with this the integral in the exponent becomes @xmath41 where @xmath42 where we have used the fact @xmath43
then @xmath44 and subsequently @xmath24 can be evaluated by integrating over @xmath45 and @xmath46 .
@xmath47\ ] ] now the above expression for @xmath24 does not include the contribution coming from the displacement of the bubble in the opposite direction , i.e. for negative values of @xmath48 . to include that one has to multiply the above expression by @xmath49 to get the final correct expression for @xmath24 .
@xmath50.\ ] ] next we evaluate the correlation function @xmath23 . here
also we follow the same technique and write the correlation function as @xmath51 next introducing @xmath52 one writes the average as @xmath53 similarly the above average can be written as @xmath54 then one gets @xmath55 where @xmath56 to get the final result one has to perform integrations over @xmath46 , @xmath57 , @xmath45 and @xmath58 .
unfortunately one of the integrals over @xmath48 ( say @xmath58 ) can not be performed analytically and has to be carried out numerically .
now as done earlier in the calculation of @xmath24 here also one should consider the contribution due to the displacement of the string in the opposite direction which corresponds to negative @xmath48 .
this means not only the above expression should be multiplied by a factor of @xmath49 but also there would be two cross terms @xmath59 and @xmath60 .
these cross terms would contribute equally to the correlation function . including all these contributions ,
the final correct expression for the correlation function becomes @xmath61 where @xmath62\ ] ] and @xmath63\ ] ] now in principle one can use eq .
( [ itfinal ] ) and eq .
( [ iti0final ] ) to calculate @xmath4 defined in eq .
( [ gt ] ) . but to do that one has to evaluate @xmath64 first .
now as mentioned earlier @xmath65 introduced in eq .
( [ rnt ] ) has to vanish at the boundaries which is satisfied by choosing @xmath66 .
one can show that the time dependent coefficients @xmath67 obeys the following equation .
@xmath68 @xmath69 and @xmath70 and the @xmath71 s are the random forces which satisfy @xmath72 @xmath73 from eq .
( [ xpeq ] ) and using the statistical properties of the random forces one can derive @xmath74 where @xmath75 next we will choose @xmath76 which corresponds to the mid point of the bubble .
this further reduces the expression for @xmath64 to @xmath77 then using eq .
( [ xpcorr ] ) and replacing the sum in eq .
( [ ctsum2 ] ) by an integral one can write @xmath64 as @xmath78 the factor @xmath79 comes because only odd modes contribute to the sum .
fortunately the above integral is analytical . after carrying out the integration one gets @xmath80.\ ] ] using this expression for @xmath64 and eq .
( [ itfinal ] ) and eq .
( [ iti0final ] ) one can evaluate @xmath4 defined in eq .
( [ gt ] ) .
here we make a comparison of our model with the experimental data obtained by altan - bonett et al @xcite.the fitting function they used has already been mentioned at the beginning ( eq . ( [ gexp ] ) ) .
figure 1 is a comparison of our model with altan - bonett fitting function ( eq .
( [ gexp ] ) ) . the comparison
are made for a fixed set of parameters ( @xmath81 ) while changing the force constant of the confined harmonic well ( @xmath19 ) which is also a measure of the strength of h - bonding and staking interaction between the base pairs .
changing @xmath19 by keeping other parameters unchanged physically means changing the relaxation time of the normal modes of the chain as @xmath82 .
thus a smaller value of @xmath19 results slow relaxation of the normal modes and a larger @xmath19 results faster relaxation of normal modes .
another interesting observation is that the relaxation time for the higher normal modes ( larger @xmath35 ) have very weak dependence on @xmath19 , while the relaxation times for the lower normal modes ( smaller @xmath35 ) have stronger @xmath19 dependence .
one can see from fig .
1 that there exist at least one value of @xmath19 ( when all the other parameters are kept fixed ) for which a very good comparison with altan - bonett results can be made . moreover a small change in the value of @xmath19 results poor comparison with the experimental result as shown in fig .
1 ( dashed , blue ) . a smaller value of @xmath19 makes the dynamics slower as expected and also a higher value results faster decay of the autocorrelation function ( green , dashed - dot ) .
this strong @xmath19 dependence also suggests that the lower normal modes of the strand mostly contribute to bubble dynamics .
here we would also like to mention that the choice of a particular set of parameters is not unique as more than one set of values can also be used to make a reasonably well comparison as is also found with the model of chatterjee et al @xcite . in this section
we explore the short and long time behavior of @xmath4 .
let us first consider the correlation function used to fit the experimental data by altan - bonett eq .
( [ gexp ] ) . at short time
one can approximate , @xmath83\simeq 1-\frac{2 z}{\sqrt{\pi}}$ ] as ( @xmath84 ) and the correlation function becomes @xmath85 .
thus it behaves as a power law .
to explore the long time limit we use the following approximation , @xmath86\simeq \frac{e^{-z^2}}{z\sqrt{\pi } } $ ] as ( @xmath87 ) , which gives @xmath88 .
now it would be interesting to see what happens to @xmath4 ( or @xmath89 , @xmath90 ) in our formulation .
all time dependence in @xmath4 comes through @xmath23 or in other words through @xmath91 and @xmath92 defined earlier . to analyze the the short time behavior of @xmath4 we first rewrite short time @xmath64 as @xmath93 where , @xmath94 and @xmath95 .
using the above short time expression for @xmath64 and keeping the leading order in @xmath3 the short time expression of @xmath91 simplifies to @xmath96 with @xmath97 and @xmath98 similarly @xmath99 where @xmath100 , @xmath101 , @xmath102 , @xmath103 , @xmath104 , @xmath105 . then combining eq .
( [ a1short ] ) , eq .
( [ a2short ] ) one finally gets the short time limit of @xmath4 .
@xmath106 where , @xmath107 , @xmath108 ( as @xmath4 is normalized ) , @xmath109)^2 $ ] , @xmath110 . hence at short time it has the similar time dependence as the one used by altan - bonett to fit the experimental data .
next we analyze the long time behavior of @xmath4 .
as mentioned earlier all the time dependence of @xmath4 comes from @xmath64 which is embedded in @xmath91 and @xmath92 . as in the long time
, @xmath111 , one can further simplifies the expressions for @xmath91 and @xmath92 . to do this consider the complimentary error function sitting inside the integral in eq .
( [ a1 ] ) .
@xmath112=erfc[\frac{2c(0)\alpha -c(t)\alpha _ 2}{2\sqrt{2c(0)(4c(0)^2)}}].\ ] ] now the complimentary error function is in the form @xmath113 $ ] with @xmath114 , @xmath115.as @xmath64 approaches zero in the long time we can make a series expansion of the above complimentary error function and keep only the first two terms to get @xmath116=erfc[p_1]+\frac{2 e^{-p_1 ^ 2 } q_1 \sqrt{t}}{\sqrt{\pi}}.\ ] ] now one can analytically perform the integration over @xmath58 to get the long time expressions for @xmath91 .
similarly we get the long time limit of @xmath92 by following the above steps .
it shows that the autocorrelation function @xmath4 approaches zero the same way @xmath64 approaches zero in the long time limit .
thus in the long time limit @xmath4 approaches zero as @xmath117 as is also predicted by altan - bonett fitting model .
the stochastic dynamics of dna bubble formed due to rupture and reformation of hydrogen bonds is modeled based on a rouse chain description of the dna strand . although the model is very simple it produces the experimental results of altan - bonett reasonably well .
unlike other well known models @xcite our model goes beyond a single reaction coordinate or order parameter description by taking into account of the collective nature of the dynamics through different modes of the chain as is done in the rouse description in the simplest possible way .
the dynamics seems to be very sensitive to the relaxation times of different normal modes of the chain which is physically understandable as the bubble dynamics " should have the contribution from all the possible normal mode of vibration of the chain .
however the current model probably can not account for the bubble size distribution .
we are in the process of developing a more realistic model which can account for the issue like bubble size distribution .
the author thanks k. l. sebastian for encouragements . | we propose a model for the fluctuation dynamics of the local denaturation zones ( bubbles ) in double - stranded dna . in our formulation ,
the dna strand is model as a one dimensional rouse chain confined at both the ends .
the bubble is formed when the transverse displacement of the chain attains a critical value .
this simple model effectively reproduces the autocorrelation function for the tagged base pair in the dna strand as measured in the seminal single molecule experiment by altan - bonnet et .
al ( phys
. rev .
lett . 90 , 138101 ( 2003 ) ) .
although our model is mathematically similar to the one proposed by chatterjee et al .
( j. chem .
phys .
127 , 155104 ( 2007 ) ) it goes beyond a single reaction coordinate description by incorporating the chain dynamics through a confined rouse chain and thus considers the collective nature of the dynamics .
our model also shows that the autocorrelation function is very sensitive to the relaxation times of the normal modes of the chain , which is obvious since the fluctuation dynamics of the bubble has the contribution from the different normal modes of the chain . |
one - dimensional totally asymmetric simple exclusion process ( tasep ) has been studied for many years @xcite .
it is one special type of aseps which represent one of the basic models studying non - equilibrium behavior of particle transport along one - dimensional lattices .
asep was first introduced in @xcite for the explanation of ribosome inside the cell of creature .
now , asep has also been used to simulate a lot of physical processes including surface diffusion @xcite , traffic model @xcite , and molecular motors @xcite,@xcite , etc .
tasep , as a special case of asep , has two identical characters .
one is that the model is discrete , which means that it is studied in finite time intervals .
the other is that the particles in the lattices can only move in one direction .
the analytical solutions under open boundary conditions had been obtained in @xcite .
under open boundary conditions , the solutions yield phase diagrams with three phases @xcite . at small values of injection rates @xmath0 and @xmath1 , the system is found in a low - density entry - limited phase where @xmath2 where @xmath3 and @xmath4 are the densities at the entrance , exit and the bulk of the lattice far away from the boundaries , respectively .
@xmath5 denotes the flux . at small values of extraction rates
@xmath6 and @xmath7 , the system is in a high - density exit - limited phase with + @xmath8 at large values of the injection ( @xmath9 ) and extraction ( @xmath10 ) rates the system is in a maximal current phase with + @xmath11 a large number of varieties of extensions of taseps have been investigated , such as tasep with hierarchical long - range connections @xcite , two lane situations @xcite , two speed tasep@xcite , the effect of defect locations@xcite , particle - dependent hopping rates @xcite , and so forth .
recently , j. brankov @xcite and e. pronina @xcite studied an asep with two chains in the middle of the filament .
they supposed that a particle chooses to move into these two chains with equal probability , 0.5 and these two chains have the same length .
then in 2007 , yao - ming yuan and rui jiang @xcite investigated a tasep with a shortcut in the middle .
they set a possibility @xmath12 for a particle to jump through the shortcut when it faces the bifurcation , and they set the length of the shortcut to be zero .
unfortunately , they did not state the situation clearly . in this paper , we will investigate tasep with one or two shortcuts respectively .
the length of the shortcut is also assumed to be zero .
for example , the filament on which the motor moves may be twisted as figure 1(a ) , a motor may have a chance to jump directly from site @xmath13 to @xmath14 , as shown in figure 1(b ) .
to the basic model , i.e. there is only one shortcut along the filament .
it is natural for us to divide the whole filament into three segments . as shown in figure 1(b ) , a molecular motor will face a choice of whether to jump through the shortcut or to move ordinarily through segment 2 , when it reaches site @xmath13 .
an important problem is that the particle at site @xmath13 may have to wait to go through the shortcut if there is also another particle occupying site @xmath15 . which of them to go first should be determined clearly .
after the basic model investigated , we would like to do some research work on our advanced model 1 .
in advanced model 1 , there are two shortcuts which begin and end at different sites .
these two shortcuts stay respectively along the filament , that is to say , shortcut 2 begins after shortcut 1 ends , as shown in figure 2 . also , considering the difference of places where the shortcuts stay , we have advanced model 2 , as shown in figure 3 .
in advanced model 2 there are also two shortcuts .
but they begin at the same site and end at different , which means when a motor moves into site @xmath13 , it will choose whether to pass through shortcut 1 or shortcut 2 or to go ordinarily through segment 2 . in the following ,
we firstly give the detailed description of the different models , then we will discuss their phase situation of the corresponding segments theoretically , which is followed by the numerical simulations .
[ fig : side : a ] [ fig : side : b ] [ fig : side : a ] [ fig : side : b ] [ fig : side : a ] [ fig : side : b ]
first , we divide the lattices into three segments , as shown in figure 1(b ) .
segment 1,2,3 start from site @xmath16 and end in site @xmath17 , respectively . in this model ,
all the particles traveling in the lattices are identical to each other . at time @xmath18
, we may suppose that the lattices are empty .
as time goes , there will be a particle in site 1 .
this particle will pass continuously through the whole filament .
given a site @xmath19 , in an infinitesimal time interval @xmath20 , if @xmath21 , a particle is inserted with probability @xmath22 , provided the site is empty , if @xmath23 and there is a particle in it , then the particle in site l is extracted with probability @xmath24 , and if @xmath25 and site @xmath26 is occupied , the particle in site @xmath26 will move into site @xmath27 with probability @xmath20 , providing site @xmath27 is empty .
however , if a particle is in site @xmath13 or @xmath15 , we have to clarify the way it moves .
although yao - ming yuan , rui jiang @xcite have discussed the method of choice at site @xmath13 , the situation at site
@xmath15 has not been studied clearly .
it is possible that when both site @xmath13 and site @xmath15 have particles , it may produce a collision in site @xmath14 .
if site @xmath13 has a particle , then the particle will choose the way as following : @xmath28 if site @xmath29 and site @xmath14 are both occupied , then the particle in site @xmath13 does not move .
@xmath28 if site @xmath14 is occupied but site @xmath29 is empty , then the particle in site @xmath13 will move into site @xmath29 with probability @xmath20 .
@xmath28 if site @xmath29 is occupied but site @xmath14 is empty then : @xmath30 if site @xmath15 is empty , the particle in site @xmath13 will move into site @xmath14 with probability @xmath31 .
@xmath30 if site @xmath15 is occupied , the particle in site @xmath13 will move into site @xmath14 with probability @xmath32 .
@xmath28 if both site @xmath29 and site @xmath14 are empty , then the particle in site @xmath13 will move into site @xmath29 with probability @xmath33 , and : @xmath30 if site @xmath15 is empty , the particle in site @xmath13 will move into site @xmath14 with probability @xmath31 .
@xmath30 if site @xmath15 is occupied , the particle in site @xmath13 will move into site @xmath14 with probability @xmath32 .
if site @xmath15 has a particle , and site @xmath14 is empty , then the particle will choose the way as following : @xmath28 if site @xmath13 is empty , then the particle in site @xmath15 will move into site @xmath14 with probability @xmath20 .
+ @xmath28 if site @xmath13 is occupied , then the particle in site @xmath15 will move into site @xmath14 with probability @xmath34 .
+ the character of this model is that the particle in site @xmath13 first choose whether to move through the shortcut and once it has decided to jump through the shortcut , it will face the problem of whether site @xmath15 is occupied .
if site @xmath15 is occupied , then the particle which has chosen the shortcut can only move into site @xmath14 with probability @xmath35 in order to prevent collision .
this means that the particle takes a risk to jump the shortcut because once it has chosen the shortcut it can not go back into segment 2 , even if site @xmath29 is empty .
if site @xmath13 has a particle , then the particle will choose the way as following : @xmath28 if site @xmath29 and site @xmath14 are both occupied , then the particle in site @xmath13 does not move .
@xmath28 if site @xmath14 is occupied but site @xmath29 is empty , then the particle in site @xmath13 will move into site @xmath29 with probability @xmath20 .
+ @xmath28 if site @xmath29 is occupied but site @xmath14 is empty then : @xmath30 if site @xmath15 is empty , the particle in site @xmath13 will move into site @xmath14 with probability @xmath31 .
@xmath30 if site @xmath15 is occupied , the particle in site @xmath13 will move into site @xmath14 with probability @xmath32 .
@xmath28 if both site @xmath29 and site @xmath14 are empty , then : @xmath30 if site @xmath15 is empty , the particle in site @xmath13 will move into site @xmath14 with probability @xmath31 , and it will move into site @xmath29 with probability @xmath33 .
@xmath30 if site @xmath15 is occupied , the particle in site @xmath13 will move into site @xmath14 with probability @xmath32 , and it will move into site @xmath29 with probability @xmath36 . if site @xmath15 has a particle , and site @xmath14 is empty , then the particle will choose the way as following : @xmath28 if site @xmath13 is empty , then the particle in site @xmath15 will move into site @xmath14 with probability @xmath20 .
@xmath28 if site @xmath13 is occupied , then the particle in site @xmath15 will move into site @xmath14 with probability @xmath34 .
the character of this model is that once the particle has chosen the shortcut but see that the particle in site @xmath15 has priority to move into site @xmath14 , it can change its mind and return to site @xmath29 , providing site @xmath29 is empty .
this model is closer to reality because actually bus drivers can change their mind in time when they are told that the road ahead is to some extent blocked .
it is not reasonable for drivers to take a shortcut which is actually difficult to go through .
now we divide the lattices into five segments , as shown in figure 2(b),segment 1,2,3,4,5 starts in site @xmath37 and ends in site @xmath38,respectively . in this model ,
all the particles traveling in the lattices are identical to each other . at time @xmath18
, we may suppose that the lattices are empty .
as time goes , there will be a particle goes into site 1 .
this particle will pass continuously through the whole filament . given site @xmath19 , in an infinitesimal time interval @xmath20 , if @xmath21 , a particle is inserted with probability @xmath22 , provided the site is empty , if @xmath23 and there is a particle in it , then the particle in site l is extracted with probability @xmath24 , and if @xmath39 and site @xmath26 is occupied , the particle in site @xmath26 will move into site @xmath27 with probability @xmath20 , providing site @xmath27 is empty .
however , if a particle is in site @xmath13 or @xmath15 or @xmath40 or @xmath41 , we have to clarify the way it moves .
we will construct advanced model 1 using similar method according to basic model(situation 1 ) .
if site @xmath13 has a particle , then the particle will choose the way as following : @xmath28 if site @xmath29 and site @xmath14 are both occupied , then the particle in site @xmath13 does not move .
@xmath28 if site @xmath14 is occupied but site @xmath29 is empty , then the particle in site @xmath13 will move into site @xmath29 with probability @xmath20 .
@xmath28 if site @xmath29 is occupied but site @xmath14 is empty then : @xmath30 if site @xmath15 is empty , the particle in site @xmath13 will move into site @xmath14 with probability @xmath42 .
@xmath30 if site @xmath15 is occupied , the particle in site @xmath13 will move into site @xmath14 with probability @xmath43 .
@xmath28 if both site @xmath29 and site @xmath14 are empty , then the particle in site @xmath13 will move into site @xmath29 with probability @xmath44 , and : @xmath30 if site @xmath15 is empty , the particle in site @xmath13 will move into site @xmath14 with probability @xmath42 .
@xmath30 if site @xmath15 is occupied , the particle in site @xmath13 will move into site @xmath14 with probability @xmath43 .
if site @xmath15 has a particle , and site @xmath14 is empty , then the particle will choose the way as following : @xmath28 if site @xmath13 is empty , then the particle in site @xmath15 will move into site @xmath14 with probability @xmath20 .
if site @xmath13 is occupied , then the particle in site @xmath15 will move into site @xmath14 with probability @xmath45 .
if site @xmath40 has a particle , then the particle will choose the way as following : @xmath28 if site @xmath46 and site @xmath47 are both occupied , then the particle in site @xmath40 does not move .
@xmath28 if site @xmath47 is occupied but site @xmath46 is empty , then the particle in site @xmath40 will move into site @xmath46 with probability @xmath20 .
@xmath28 if site @xmath46 is occupied but site @xmath47 is empty then : @xmath30 if site @xmath41 is empty , the particle in site @xmath40 will move into site @xmath47 with probability @xmath48 .
@xmath30 if site @xmath41 is occupied , the particle in site @xmath40 will move into site @xmath47 with probability @xmath49 .
@xmath28 if both site @xmath46 and site @xmath47 are empty , then the particle in site @xmath40 will move into site @xmath46 with probability @xmath50 , and : @xmath30 if site @xmath41 is empty , the particle in site @xmath40 will move into site @xmath47 with probability @xmath48 .
if site @xmath41 is occupied , the particle in site @xmath40 will move into site @xmath47 with probability @xmath49 .
if site @xmath41 has a particle , and site @xmath47 is empty , then the particle will choose the way as following : @xmath28 if site @xmath40 is empty , then the particle in site @xmath41 will move into site @xmath47 with probability @xmath20 .
@xmath28 if site @xmath40 is occupied , then the particle in site @xmath41 will move into site @xmath47 with probability @xmath51 .
also we can divide the lattices into four segments , as shown in figure 3(b ) , segment 1,2,3,4 starts in site @xmath52 and ends in site @xmath53,respectively . in this model ,
all the particles traveling in the lattices are identical to each other . at time @xmath18
, we may suppose that the lattices are empty .
as time goes , there will be particles goes into site 1 .
these particles will pass continuously through the whole filament . given site
@xmath19 , in an infinitesimal time interval @xmath20 , if @xmath21 , a particle is inserted with probability @xmath22 , provided the site is empty , if @xmath23 and there is a particle in it , then the particle in site l is extracted with probability @xmath24 , and if @xmath54 and site @xmath26 is occupied , the particle in site @xmath26 will move into site @xmath27 with probability @xmath20 , providing site @xmath27 is empty . however , if a particle is in site @xmath13 or @xmath15 or @xmath55 , we have to clarify the way it moves .
we will construct advanced model 2 using similar method according to basic model(situation 1 ) .
if site @xmath13 has a particle , then the particle will choose the way as following : + @xmath28 if sites @xmath29,@xmath14 and @xmath40 are all occupied , the the particle in site @xmath13 does not move .
@xmath28 if site @xmath29 is empty but sites @xmath14 and @xmath40 are occupied , the particle in site @xmath13 will move into site @xmath29 with probability @xmath20 .
@xmath28 if site @xmath14 is empty but sites @xmath29 and @xmath40 are occupied , then : @xmath30 if site @xmath15 is empty , the particle in site @xmath13 will move into site @xmath14 with probability @xmath42 .
@xmath30 if site @xmath15 is occupied , the particle in site @xmath13 will move into site @xmath14 with probability @xmath43 .
@xmath28 if site @xmath40 is empty but sites @xmath29 and @xmath14 are occupied , then : @xmath30 if site @xmath55 is empty , the particle in site @xmath13 will move into site @xmath40 with probability @xmath48 .
@xmath30 if site @xmath55 is occupied , the particle in site @xmath13 will move into site @xmath40 with probability @xmath49 .
@xmath28 if site @xmath29 and site @xmath14 are empty but site @xmath40 is occupied , then the particle will move into site @xmath29 with probability @xmath44 , and : @xmath30 if site @xmath15 is empty , the particle in site @xmath13 will move into site @xmath14 with probability @xmath42 .
@xmath30 if site @xmath15 is occupied , the particle in site @xmath13 will move into site @xmath14 with probability @xmath43 .
@xmath28 if site @xmath29 and site @xmath40 are empty but site @xmath14 is occupied , then the particle will move into site @xmath29 with probability @xmath50 , and : @xmath30 if site @xmath55 is empty , the particle in site @xmath13 will move into site @xmath40 with probability @xmath48 .
@xmath30 if site @xmath55 is occupied , the particle in site @xmath13 will move into site @xmath40 with probability @xmath49 .
@xmath28 if site @xmath14 and site @xmath40 are empty but site @xmath29 is occupied , then : + @xmath30 if site @xmath15 is empty , the particle in site @xmath13 will move into site @xmath14 with probability @xmath42 .
@xmath30 if site @xmath15 is occupied , the particle in site @xmath13 will move into site @xmath14 with probability @xmath43 .
@xmath30 if site @xmath55 is empty , the particle in site @xmath13 will move into site @xmath40 with probability @xmath48 .
@xmath30 if site @xmath55 is occupied , the particle in site @xmath13 will move into site @xmath40 with probability @xmath49 .
@xmath28 if sites @xmath29 , @xmath14 and @xmath40 are all empty , then the particle in site @xmath13 will move into site @xmath29 with probability @xmath56 , and : @xmath30 if site @xmath15 is empty , the particle in site @xmath13 will move into site @xmath14 with probability @xmath42 .
@xmath30 if site @xmath15 is occupied , the particle in site @xmath13 will move into site @xmath14 with probability @xmath43 .
@xmath30 if site @xmath55 is empty , the particle in site @xmath13 will move into site @xmath40 with probability @xmath48 .
@xmath30 if site @xmath55 is occupied , the particle in site @xmath13 will move into site @xmath40 with probability @xmath49 .
if site @xmath15 has a particle , and site @xmath14 is empty , then the particle will choose the way as following : @xmath28 if site @xmath13 is empty , then the particle in site @xmath15 will move into site @xmath14 with probability @xmath20 .
@xmath28 if site @xmath13 is occupied , then the particle in site @xmath15 will move into site @xmath14 with probability @xmath45 .
if site @xmath55 has a particle , and site @xmath40 is empty , then the particle will choose the way as following : @xmath28 if site @xmath13 is empty , then the particle in site @xmath55 will move into site @xmath40 with probability @xmath20 .
@xmath28 if site @xmath13 is occupied , then the particle in site @xmath55 will move into site @xmath40 with probability @xmath51 .
in this section , we will give a theoretical analysis of the phase situation of three segments of our basic model .
we choose situation 1 of our basic model described above as the base of our analysis .
according to figure 1(b ) , we are able to express the insertion and extraction rates using densities of certain sites .
these rates are the following : @xmath57\\ \alpha_{eff2}=\rho_{k_{1}}[\rho_{k_{2}}+(1-\rho_{k_{2}})(1-q)]\\ \beta_{eff2}=(1-\rho_{k_{2}})[(1-\rho_{k_{1}})+q(1-p)\rho_{k_{1}}+(1-q)\rho_{k_{1}}]\\ \alpha_{eff3}=\rho_{k_{1}}(1-\rho_{k_{2}-1})q+\rho_{k_{2}-1}(1-\rho_{k_{1}})+\rho_{k_{1}}\rho_{k_{2}-1}\\ \beta_{eff3}=\beta\\ j_{sc}=q\rho_{k_{1}}(1-\rho_{k_{2}})[1-\rho_{k_{2}-1}+p\rho_{k_{2}-1 } ] \end{array}\ ] ] certainly , the flux should be conserved : @xmath58
firstly , we are going to prove that segment 1 and segment 3 should be in the same phase .
if segment 1 is in maximum - current phase , then we have @xmath59 according to ( 3 ) . according to ( 5 ) , we have @xmath60 .
so segment 3 is also in maximum - current phase .
secondly , we claim that the following four cases are impossible @xmath61 to prove that , we firstly focus on the first two situations . if the phase condition is @xmath62 , due to ( 1 ) , we have : @xmath63 because @xmath64 , we obtain : @xmath65 when @xmath66 , from ( 7 ) we have @xmath67 , which contradicts @xmath68 . thus the phase situations @xmath62 are impossible .
if the phase condition is @xmath69 , it is difficult for us to analysis the original model with parameter @xmath70 .
so we set @xmath71 , which means that the particle in site @xmath13 has priority to move into site @xmath14 when site @xmath15 is occupied .
due to ( 2 ) , we have : @xmath72 from which we can obtain : @xmath73 let @xmath74 , then ( 9 ) turns into : @xmath75 due to ( 1 ) , we have : @xmath76\ ] ] the equation above can be turned into : @xmath77 from ( 10 ) and ( 12 ) , we reach : @xmath78 if @xmath79 , then @xmath80 does not agree with reality .
so @xmath81 could not be 0 .
thus @xmath82 .
for all @xmath83 $ ] , @xmath84 on the other hand , due to ( 5 ) , we have : @xmath85 because @xmath86 , we know that @xmath87 . together with ( 15 ) , we have @xmath88 . together with ( 14 ) , we get @xmath79 .
it is impossible .
thus the phase situation ca nt be @xmath69 .
if the phase condition is @xmath89 , also we can set @xmath71 for our convenience to prove .
because segment 2 is in high - density phase , we have @xmath90 . because segment 3 is in low - density phase
, we have @xmath91 . according to ( 4 ) , we have the cases : @xmath92 let @xmath93 , then the cases become : @xmath94 from ( 17 ) we can obtain : @xmath95 because @xmath96 , we know that @xmath97 , together with ( 18 ) , we will have @xmath98 , which leads to @xmath79 , which does not agree with reality . thus @xmath89 has been excluded .
now we have proved that segment 1 and segment 3 should be in the same phase . in the following , we will discuss which phase may segment 2 be in . if the three segments are in @xmath99 , from the proof which has demonstrated that @xmath89 is impossible , we know that @xmath100 can not be true either .
thus we can definitely say that for @xmath99 , only @xmath101 can be right .
if the three segments are in @xmath102 , let us see whether @xmath103 is true .
if the three segments are in @xmath103 , then we have : @xmath104=\rho_{k_{1}+1 } \end{cases}\ ] ] let @xmath74 , then from ( 19 ) we can get @xmath105 , then @xmath82 , then @xmath106 .
because of ( 5 ) , we have : @xmath107
if we put @xmath106 into ( 20 ) , we will get : @xmath108 because @xmath109 , we have @xmath110 . from @xmath111 , we know that @xmath112 , then @xmath113 , which does not agree to reality .
so @xmath103 is impossible .
thus we have excluded @xmath103 .
we now know that for @xmath102 , only @xmath114 is possible . as for @xmath115 , it is a tough task for us to analyze the situation of segment 2 using equations .
but under assistance of numerical simulations , it will be possible for us to find which phase is true for segment 2 . for situation 2 of model
a , we can also write the insertion rates and extraction rates at special sites : @xmath116\\ \alpha_{eff2}=\rho_{k_{1}}[\rho_{k_{2}}+(1-\rho_{k_{2}})(1-q((1-\rho_{k_{2}-1})+p\rho_{k_{2}-1}))]\\ \beta_{eff2}=(1-\rho_{k_{2}})[(1-\rho_{k_{1}})+q(1-p)\rho_{k_{1}}+(1-q)\rho_{k_{1}}]\\ \alpha_{eff3}=\rho_{k_{1}}(1-\rho_{k_{2}-1})q+\rho_{k_{2}-1}(1-\rho_{k_{1}})+\rho_{k_{1}}\rho_{k_{2}-1}\\ \beta_{eff3}=\beta\\ \end{array}\ ] ] the analysis of situation 2 of basic model is the same of the analysis of situation 1 , providing @xmath71 , so we do not state it repeatedly . from the analysis above
, we can conclude that for the situation @xmath71 , the three segments should be among the following situations : @xmath117 , @xmath115 .
we will do further investigation using numerical simulations in the following sections . in this section
, we will give the phase situation of the five segments of advanced model 1 . according to figure 2(b ) , we are able to express the insertion and extraction rates using densities of certain sites .
these rates are the following : @xmath118\\ \alpha_{eff2}=\rho_{k_{1}}[\rho_{k_{2}}+(1-\rho_{k_{2}})(1-q_{1})]\\ \beta_{eff2}=(1-\rho_{k_{2}})[(1-\rho_{k_{1}})+q_{1}(1-p_{1})\rho_{k_{1}}+(1-q_{1})\rho_{k_{1}}]\\ \alpha_{eff3}=\rho_{k_{1}}(1-\rho_{k_{2}-1})q_{1}+\rho_{k_{2}-1}(1-\rho_{k_{1}})+\rho_{k_{1}}\rho_{k_{2}-1}\\ \beta_{eff3}=(1-\rho_{k_{3}+1})\rho_{k_{4}}+(1-\rho_{k_{4}})(1-\rho_{k_{3}+1})(1-q_{2})+q_{2}(1-\rho_{k_{4}})[(1-\rho_{k_{4}-1})+p_{2}\rho_{k_{4}-1}]\\ \alpha_{eff4}=\rho_{k_{3}}[\rho_{k_{4}}+(1-\rho_{k_{4}})(1-q_{2})]\\ \beta_{eff4}=(1-\rho_{k_{4}})[(1-\rho_{k_{4}})+q_{2}(1-p_{2})\rho_{k_{3}}+(1-q_{2})\rho_{k_{3}}]\\ \alpha_{eff5}=\rho_{k_{3}}(1-\rho_{k_{4}-1})q_{2}+\rho_{k_{4}-1}(1-\rho_{k_{3}})+\rho_{k_{3}}\rho_{k_{4}-1}\\ \beta_{eff5}=\beta\\ j_{sc_{1}}=q_{1}\rho_{k_{1}}(1-\rho_{k_{2}})[1-\rho_{k_{2}-1}+p_{1}\rho_{k_{2}-1}]\\ j_{sc_{2}}=q_{2}\rho_{k_{3}}(1-\rho_{k_{4}})[1-\rho_{k_{4}-1}+p_{2}\rho_{k_{4}-1}]\\ \end{array}\ ] ] also , the flux should satisfies : @xmath119 in the following , we will prove that the segment 1,3 and 5 are in the same phase .
conclusions obtained in the proof of basic model will be used in the following steps . for the sake of simplicity , we always set @xmath120
. if one of the segments 1 , 3 or 5 is in maximum - current phase , then the other two will also be in maximum - current phase , since @xmath121 .
due to ( 6 ) and ( 7 ) , we also can find that the situation is @xmath122 is impossible . in the following ,
we discuss the situation @xmath123 .
if segment 3 is in low - density phase , then segment 1,2 and 3 will be in @xmath124 . according to the proof of the basic model , we know that this situation is impossible . because we can regard segment 1,2 and 3 as basic model , only by changing @xmath125 to be @xmath126 .
this change does not influence the proof since the rate expression of @xmath126 is independent of the densities of sites in the first three segments .
similarly , segment 3 can not be in high - density phase since segment 3,4 and 5 will be in @xmath124 .
so now we have excluded @xmath122 and @xmath123 , . on the other hand , if the phase situation is @xmath127 , then we know immediately that segment 3 can not be in high - density phase .
so segment 1,3 and 5 should all be in low - density phase .
if the situation is @xmath128 , then we know immediately that segment 3 can not be in low - density phase .
so segment 1,3 and 5 should all be in high - density phase .
since we have proved that @xmath100 and @xmath103 does not exist in basic model , so we can conclude that all of the five segments should all be in low - density phase , providing segment 1 is in low - density phase , and all be in high - density phase , when segment 1 is in high - density phase .
as for the situation @xmath129 , it is difficult for us to analyze it theoretically .
so we will use numerical simulations to see whether segment 2 and 4 are in high or low - density phase .
the simulations will be presented in the following section . in this section
, we are going to demonstrate the phase situation of four segments of advanced model 2 . according to figure 3(b ) , we are able to express the insertion and extraction rates using densities of certain sites .
these rates are stated as follows : @xmath130\\ \indent\indent + ( 1-\rho_{k_{3}})\rho_{k_{1}+1}\rho_{k_{2}}q_{2}[(1-\rho_{k_{3}-1})+p_{2}\rho_{k_{3}-1}]\\ \indent\indent+(1-\rho_{k_{1}+1})(1-\rho_{k_{2}})\rho_{k_{3}}[1-q_{1}+q_{1}[(1-\rho_{k_{2}-1})+p_{1}\rho_{k_{2}-1}]]\\ \indent\indent+(1-\rho_{k_{1}+1})(1-\rho_{k_{3}})\rho_{k_{2}}[1-q_{2}+q_{2}[(1-\rho_{k_{3}-1})+p_{2}\rho_{k_{3}-1}]]\\ \indent\indent+\rho_{k_{1}+1}(1-\rho_{k_{2}})(1-\rho_{k_{3}})[q_{1}[(1-\rho_{k_{2}-1})+p_{1}\rho_{k_{2}-1}]+q_{2}[(1-\rho_{k_{3}-1})+p_{2}\rho_{k_{3}-1}]]\\ \indent\indent+(1-\rho_{k_{1}+1})(1-\rho_{k_{2}})(1-\rho_{k_{3}})[1-q_{1}-q_{2}+q_{1}[(1-\rho_{k_{2}-1})+p_{1}\rho_{k_{2}-1}]\\ \indent\indent+q_{2}[(1-\rho_{k_{3}-1})+p_{2}\rho_{k_{3}-1}]]\\ \alpha_{eff2}=\rho_{k_{1}}[\rho_{k_{2}}\rho_{k_{3}}+(1-\rho_{k_{2}})\rho_{k_{3}}(1-q_{1})+(1-\rho_{k_{3}})\rho_{k_{2}}(1-q_{2})\\ \indent\indent+(1-\rho_{k_{2}})(1-\rho_{k_{3}})(1-q_{1}-q_{2})]\\ \beta_{eff2}=(1-\rho_{k_{2}})[(1-\rho_{k_{1}})+q_{1}(1-p_{1})\rho_{k_{1}}+(1-q_{1})\rho_{k_{1}}]\\ \alpha_{eff3}=\rho_{k_{1}}(1-\rho_{k_{2}-1})q_{1}+\rho_{k_{2}-1}(1-\rho_{k_{1}})+\rho_{k_{1}}\rho_{k_{2}-1}\\ \beta_{eff3}=(1-\rho_{k_{3}})[(1-\rho_{k_{1}})+q_{2}(1-p_{2})\rho_{k_{1}}+(1-q_{2})\rho_{k_{1}}]\\ \alpha_{eff4}=\rho_{k_{1}}(1-\rho_{k_{3}-1})q_{2}+\rho_{k_{3}-1}(1-\rho_{k_{1}})+\rho_{k_{1}}\rho_{k_{3}-1}\\ \beta_{eff4}=\beta\\ j_{sc_{1}}=q_{1}\rho_{k_{1}}(1-\rho_{k_{2}})[1-\rho_{k_{2}-1}+p_{1}\rho_{k_{2}-1}]\\ j_{sc_{2}}=q_{2}\rho_{k_{1}}(1-\rho_{k_{3}})[1-\rho_{k_{3}-1}+p_{2}\rho_{k_{3}-1}]\\ \end{array}\ ] ] also we have the equation of flux : @xmath131
our conclusion is that the segments 1 and 4 are in the same phase . in order to simplify our proof , we always set @xmath120 .
firstly , if segment 1 or segment 4 is in maximum - current phase , then the other should also be in maximum - current phase due to ( 26 ) . if the phase condition is @xmath132 , then due to @xmath133 , we know that when @xmath66 we have @xmath67 , which is contradictory to @xmath0 and @xmath134
. now we discuss the situation @xmath135 .
if the segment 3 is in low - density phase , then we have : @xmath136 and @xmath137 if segment 2 is in low - density phase , then @xmath138 let @xmath139 , we have : @xmath75 since segment 1 is in high - density phase , we have : @xmath140 so we have : @xmath77 from ( 30 ) and ( 32 ) , we know that : @xmath78 from ( 33 ) we have @xmath82 because @xmath81 can not be 0 .
so @xmath141 for all @xmath142 $ ] , which implies @xmath143 .
so segment 2 can not be in low - density phase .
if segment 2 is in high - density phase , then we have : @xmath144 from ( 34 ) we know that @xmath79 or @xmath145 , which will lead to @xmath146 or @xmath147 that does not make sense when @xmath148 .
so we have excluded @xmath149 .
if the segment 3 is in high - density phase , then @xmath150 from ( 35 ) we know that @xmath79 or @xmath151 , which will lead to @xmath146 or @xmath143 that does not make sense .
@xmath152 is also impossible .
now we have proved that segment 1 and segment 4 are in the same phase .
we now discuss the situation @xmath153 . from the proof above we know that segment 3 can not be in high - density phase , either can not segment 2 .
thus the phases can only be @xmath154 . while considering @xmath155 , we should first assume that segment 3 be in low - density phase .
from the proof above we will know that segment 2 can not be either in low or in high - density phase . and if we assume segment 3 to be in high - density phase , we know that segment 2 can not be in low - density phase .
thus the phases can only be @xmath156 .
to the situation @xmath157 , it is rather difficult for us to analyze it theoretically .
so we also use numerical simulations to get the results .
in this section , we use numerical methods to simulate basic models . first we demonstrate the results of situation 1 and situation 2 . according to analysis in last section
, we set insertion rate @xmath158 and extraction rate @xmath125 to be 0.3 and 0.8 for @xmath101 , 0.8 and 0.3 for @xmath114 , 0.8 and 0.8 for @xmath115 .
the results of the simulations are plotted in figure 4 . ;
( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : a ] ; ( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : b ] ; ( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : c ] ; ( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : d ] from figure 4 , we can see that the results of numerical solutions are in agreement with the results of equation analysis .
when @xmath161 and @xmath162 , all the three segments are in low - density phase .
when @xmath163 and @xmath164 , all the three segments are in high - density phase . when @xmath160 , segment 1 and segment 3 are in maximum - current phase .
from figure 4(a ) , we can find that the density in segment 2 decreases when @xmath12 increases , providing @xmath70 is constant .
for the same @xmath12 , when @xmath70 increases , the density in segment 2 decreases . from figure 4(b ) , we see that the density in segment 2 increases when @xmath12 increases , providing @xmath70 is constant . for the same @xmath12 , when @xmath70 increases , the density in segment 2 increases .
this is because when @xmath70 is relatively large , more particles are able to jump through the shortcut to reach site @xmath14 , thus leaving particles in segment 2 in a serious traffic jam . from figure 4(c ) and ( d ) , we see that when segment 1 and segment 3 are both in maximum - current phase , segment 2 can be either in high - density phase or low - density phase , depending on the initial density rate of each lattice
. if the average initial density rate of the lattices is high , then segment 2 will be in high - density phase when the flux becomes stable .
if the average initial density rate of the lattices is low , then segment 2 will be in low - density phase when the flux becomes stable .
now we use numerical methods to simulate situation 2 of basic model . the insertion and extraction rates
are set as the same as in the simulation of situation 1 . ;
( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : a ] ; ( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : b ] ; ( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : c ] ; ( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : d ] from figure 5 we are able to see that the results of situation 2 of basic model seem to be similar to the results of situation 1 . but after careful inspection , we may find that there are several differences between the results of the two situations , which manifest the distinct characters of these two situations .
in order to do further investigation , we use figures which can show the differences between two situations .
we calculate the values which are obtained by subtracting the density of each lattice of situation 2 from the density of each lattice of situation 1 .
when we do these subtractions , we keep parameters @xmath12 and @xmath70 in constant . the results are shown in figure 6 . and extraction rate @xmath162 : ( a ) when @xmath165 and @xmath71 ; ( b ) when @xmath166 and @xmath71 ; ( c ) when @xmath167 and @xmath168 ; ( d ) when @xmath166 and @xmath168.,title="fig:",width=240 ] [ fig : side : a ] and extraction rate @xmath162 : ( a ) when @xmath165 and @xmath71 ; ( b ) when @xmath166 and @xmath71 ; ( c ) when @xmath167 and @xmath168 ; ( d ) when @xmath166 and @xmath168.,title="fig:",width=240 ] [ fig : side : a ] and extraction rate @xmath162 : ( a ) when @xmath165 and @xmath71 ; ( b ) when @xmath166 and @xmath71 ; ( c ) when @xmath167 and @xmath168 ; ( d ) when @xmath166 and @xmath168.,title="fig:",width=240 ] [ fig : side : b ] and extraction rate @xmath162 : ( a ) when @xmath165 and @xmath71 ; ( b ) when @xmath166 and @xmath71 ; ( c ) when @xmath167 and @xmath168 ; ( d ) when @xmath166 and @xmath168.,title="fig:",width=240 ] [ fig : side : c ] in figure 6(a ) and 6(b ) , we can see that the differences between the density of two situations are within @xmath169 , which means that when @xmath71 , there is nearly no difference between the two situations .
that is because when @xmath71 , the particle at site @xmath170 in the figure ) has priority to get to site @xmath171 in the figure ) .
so whatever @xmath12 may be , the particle can go smoothly once it chooses the shortcut .
there is no difference whether it returns back to segment 2 or not . in figure 6(c ) and ( d ) , when @xmath168 , the differences seem to be obvious .
when we contrast these two figures , we are able to find that the extent of difference is decided by parameter @xmath12 . while @xmath12 changes from 0.1 to 0.9 , the error changes from within @xmath172 to 0.015 .
this is simply because the smaller @xmath12 means less particles to choose the shortcut .
but if we compare figure 6(b ) with 6(d ) , it can apparently be found that the error changes greatly when @xmath70 is switched from 1 to 0 , providing @xmath12 is comparatively large .
so we can say that it really matters whether a particle can return back or not when many particles choose the shortcut .
if the particle can not return , a great traffic jam will happen at the certain site , reflected in figure 6(d ) as a high - positioned blue star on the top of the figure .
now we use numerical methods to simulate advanced model 1 . in order to find how the phases of five segments change when parameters @xmath173 and @xmath174 change , we set @xmath120 , thus corresponding to our equation analysis in section 3 .
we will see whether the numerical results correspond to our analytical results . ;
( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : a ] ; ( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : b ] ; ( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : c ] ; ( b)@xmath159 ; ( c)@xmath160 , initial densities of all sites are high ; ( d)@xmath160 , initial densities of all sites are low.,title="fig:",width=240 ] [ fig : side : d ] from figure 7(a ) , we can see that all of the five segments are in low - density phase .
the situation in segment 2 has nothing to do with the situation in segment 4 .
we can treat them as a connection of two basic models .
the density rate in segment 2 decreases when @xmath173 increases .
the density rate in segment 4 decreases when @xmath174 increases . from figure 7(b ) , we can see that all of the five segments are in high - density phase .
the situation in segment 2 has nothing to do with the situation in segment 4 .
we can treat them as a connection of two basic models .
the density rate in segment 2 increases when @xmath173 increases .
the density rate in segment 4 increases when @xmath174 increases . from figure 7(c ) and 7(d ) , we can see that the situation in segment 2 and 4 are relevant to the initial density of each site .
if the average initial density of all sites is high , then segment 2 and 4 will be both in high - density phase .
if the average initial density of all sites is low , then segment 2 and 4 will be both in low - density phase .
now we use numerical methods to simulate advanced model 2 . in order to find how the phases of four segments change when parameters @xmath173 and @xmath174 change , we set @xmath120 , thus corresponding to our equation analysis in chapter 3 .
we will see whether the numerical results correspond to our analytical results . also , we are interested in whether there are interactions between the two parameters @xmath173 and @xmath174 . and @xmath162 : ( a)@xmath174 changes when @xmath173 remains constant ; ( b ) @xmath173 changes when @xmath174 remains constant.,title="fig:",width=240 ] [ fig : side : a ] and @xmath162 : ( a)@xmath174 changes when @xmath173 remains constant ; ( b ) @xmath173 changes when @xmath174 remains constant.,title="fig:",width=240 ] [ fig : side : b ] and @xmath164 : ( a)@xmath174 changes when @xmath173 remains constant ; ( b ) @xmath173 changes when @xmath174 remains constant.,title="fig:",width=240 ] [ fig : side : a ] and @xmath164 : ( a)@xmath174 changes when @xmath173 remains constant ; ( b ) @xmath173 changes when @xmath174 remains constant.,title="fig:",width=240 ] [ fig : side : b ] and @xmath162 with high initial density for each lattice : ( a)@xmath174 changes when @xmath173 remains constant ; ( b ) @xmath173 changes when @xmath174 remains constant.,title="fig:",width=240 ] [ fig : side : a ] and @xmath162 with high initial density for each lattice : ( a)@xmath174 changes when @xmath173 remains constant ; ( b ) @xmath173 changes when @xmath174 remains constant.,title="fig:",width=240 ] [ fig : side : b ] and @xmath162 with low initial density for each lattice : ( a)@xmath174 changes when @xmath173 remains constant ; ( b ) @xmath173 changes when @xmath174 remains constant.,title="fig:",width=240 ] [ fig : side : a ] and @xmath162 with low initial density for each lattice : ( a)@xmath174 changes when @xmath173 remains constant ; ( b ) @xmath173 changes when @xmath174 remains constant.,title="fig:",width=240 ] [ fig : side : b ] first , from figure 8 to figure 11 , we see that the numerical simulation results are corresponding to our analytical results
. the four segments should be all in low - density phase or high - density phase , and when segment 1 and 4 are in maximum current phase , the phase of segment 2 and 3 will be based upon the initial density of each lattice .
if we focus on figure 8 , we will find some interesting things by comparing figure 8(a ) and figure 8(b ) . in figure 8(a ) , we let @xmath174 change but @xmath173 remain constant . we see that both the densities in segment 2 and segment 3 decrease in the same pace when @xmath174 increases . in figure 8 ( b ) , we let @xmath173 change when @xmath174 remains constant .
the result is that although the density in segment 2 decreases with increasing of @xmath173 , the density in segment 3 does not have apparent change .
this phenomenon means that the change of @xmath173 can not influence the density in segment 3 greatly , but the change of @xmath174 can really influence the density in segment 2 greatly . from this
we know that the parameter @xmath174 has priority to control the density of segment 2 and 3 . from figure 9 to 11 , if we compare ( a ) and ( b ) of each figure carefully , it is easy for us to find that for each figure , there exists the same phenomenon as in figure 8 . in conclusion : in advanced model 2 ,
the density of segment 2 and 3 depends on parameter @xmath174 greatly .
the change of @xmath174 will bring about the change of density of both of the segments . once @xmath174 is set constant
, parameter @xmath173 will also has effects on the density in segment 2 , but the density in segment 3 will remain in a relatively stable condition .
in this paper , tasep with one or two shortcuts have been analyzed in details .
we have found that when a particle arrives at the beginning of the shortcut , it faces the choice of whether to jump through the shortcut or not .
if it chooses the shortcut , then a problem exists that whether it can return back to the ordinary road when it finds the road ahead blocked , according to which we have two different situations . after the study
we know that if the particle can not return back , the beginning of the shortcut will most probably be filled with particles , producing a heavy traffic jam .
this research offers us an idea that when we construct a road with a shortcut , the entrance of the shortcut , that is to say the place where the main road bifurcates , ought to be built widely .
once a driver chooses to jump the shortcut but find the signal showing `` blocked ahead '' , he can immediately turn back to the ordinary road to prevent time - wasting . from the simulation of basic model , we find that no matter which situation is , the three segments should be in the following four phases : @xmath175 whether they are in @xmath176 or @xmath177 is based on the initial density of the lattices . for advanced model 1 we have found that the five segments of the model should be in the following four phases : @xmath178 for advanced model 2 we have found that the four segments of the model should be in the following four phases : @xmath179 from the simulation of advanced model 1 , we have found that this model can also be regarded as two basic models , the segment 3 of the first being the segment 1 of the second .
the probabilities of the choices of a particle facing the two shortcuts do not interfere with each other , which means that the probability of the choice of the particle facing one of the shortcuts does not influence the density in the other segment .
this phenomenon leads us to suppose that if there are many shortcuts at different places of the road , we can view them as a connection of basic models . from the simulation of advanced model 2
, we have found that the second shortcut can decide the density of both of the two segments in the middle while the first shortcut can only decide the density in segment 2 .
it offers us an idea that if we induce more drivers to go directly through the second shortcut , the density of the ordinary road will essentially decrease , providing the whole road is not so crowded
. however , in this paper all of the proofs are under the condition @xmath71 .
future work can be focused on how to proof the phase situations for all @xmath180 $ ] .
also , two shortcuts overlapping each other will be studied in the future .
99 g. m. schtz in _ phase transitions and critical phenomena _
vol 19 , eds . c. domb and j. lebowitz ( academic , london , 2001 ) .
y. zhang _ china journal of physics _
( to appear ) 2009 j. t. macdonald , j. h. gibbsand and a. c. pipkin 1968 _ biopolymer _ 6 1 g. dor , b. liedke , and k .- h .
_ physical review e _ 79 , 021125 ( 2009 ) s. cheybani , j. kertesz , m. schreckenberg _ physical review e _ 63 016108 ( 2000 ) y. zhang _ journal of statistical physics _ 134 ( 2009 ) 669 - 679 y. zhang _ biophysical chemistry _ 136 ( 2008 ) 19 - 22 b. derrida , m. r. evans , v. hakim and v. pasquier 1993 _ j. phys . a : math .
_ 26 1493 g. m. schtz and e. domany 1993 _ j. stat .
_ 72 277 j. otwinowski , s. boettcher _ tasep with hierarchical long - range connections _ arxiv:0902.2262 , 2009 t. mitsudo , h. hayakawa 2005 _ journal of physics a mathematical and general _ , 38 3087 a. borodin , p. l. ferrari , t. sasamoto _ two speed tasep _
eprint arxiv : 0904.4655 , 2009 j. j. dong , b. schmittmann , r. k. p. zia 2007 _ physical review e _ 76 051113 a. rkos , g. m. schutz _ cond - mat_/0506525 , 2005 j. brankov , n. pesheva and n. bunzarova 2004 _ phys .
_ 69 066128 e. pronina and a. b. kolomeisky 2005 _ j. stat .
p07010 y. m. yuan , r. jiang , r. wang , m. b. hu , q. s. wu 2007 _ journal of physics a - mathematical and theoretical _ 40 12351 | in this paper , the operation of totally asymmetric simple exclusion process with one or two shortcuts under open boundary conditions is discussed . using both mathematical analysis and numerical simulations ,
we have found that , according to the method chosen by the particle at the bifurcation , the model can be separated into two different situations which lead to different results .
the results obtained in this paper would be very useful in the road building , especially at the bifurcation of the road . |
shock reflection problem , which has captured the interest of researchers over the years , is one of the most important problems for the mathematical theory of multidimensional conservation laws that is still largely incomplete .
the experimental , computational and asymptotic analyses show that various patterns of reflected shocks may occur including the regular and mach reflections ( see , courant and friedrichs @xcite , glimm and majda @xcite , glass @xcite , zheng @xcite , ben - dor @xcite , chen @xcite , and chang & hsiao @xcite ) .
when a weak plane shock hits a wedge head on , two processes take place simultaneously .
the incident shock wave is reflected by the wedge surface and at the same time the flow behind it is deflected by the wedge corner , producing a nearly circular diffracted wave expanding from the vertex ; the circular wave emanating from near the vertex moves at sonic speed of the incident flow . here
, we consider the case in which only regular reflection is expected to occur ; for small disturbance approximation in weak shock reflection , this corresponds to relatively large wedge angles . the reflected shock , which is weak enough , travels backward to join the diffracted wave smoothly .
the flow in the diffracted wave region was calculated by keller and blank @xcite using the linearized theory ; this solution was modified later by hunter and keller @xcite using the theory of weakly nonlinear geometrical acoustics @xcite .
they showed that the diffracted wavefronts of linear acoustics are actually shocks .
an approximate analytical solution to the shock - wedge diffraction problem was proposed by harabetian @xcite using an alternate method that invokes perturbation expansions with multiple scales .
much effort has been devoted to the study of this problem through simplified models that capture various features of euler system within the context of an ideal gas ( see , morawetz @xcite , zheng @xcite , rosales and tabak @xcite , brio and hunter @xcite , canic et al .
@xcite , and hunter & tesdall @xcite ) .
it is well known that at high pressure or low temperature , the behavior of gases deviates from the ideal gas - law and follows van der waals type gas that deals with the possible real gas effects ( without phase - transition ) ; examples cover a family of shock wave problems with complicated interface patterns and a hydrodynamic model of sonoluminescene ( an acoustic - induced light emission phenomenon ) ( see,@xcite ) . in the present paper ,
we study the regular reflection configuration and the detachment criterion when the real gas effects are taken into account . in the limit of vanishing van der waals excluded volume
, we recover the result obtained by chang and chen @xcite , which is a refinement of von neumann s criterion .
one of the main objectives of the present paper is to study how the real gas effects influence the behavior of the local structure of the self - similar solutions of the compressible euler equations near a singular point ; the motivation stems from the work carried out in @xcite .
the real gas effects , presented here , are characterized by a van der waals type equation of state .
the analysis presented here predicts several inviscid properties of real gases undergoing shock reflection - diffraction phenomenon ; a summary of the results is presented in the last section .
+ the set - up of the reflection consists of a straight shock hitting a wedge at the origin at time @xmath0 .
the shock is assumed to be weak and moving , parallel to the y - axis , towards the wedge which is placed symmetrically about the flow direction , namely the x - axis , ( figure @xmath1 ) .
the gas ahead of the shock is at rest . since the problem is symmetric with respect to the x - axis , it suffices to consider the problem in the upper half plane outside the half wedge .
+ the basic equations of this study are the euler equations @xmath2 where @xmath3 , ( @xmath4 , @xmath5 ) , @xmath6 , @xmath7 , and @xmath8 denote respectively the density , velocity components , pressure , internal energy , and specific enthalpy , while @xmath7 and @xmath8 are given functions of @xmath3 and @xmath6 which satisfy the thermodynamical constraints @xmath9 with @xmath10 being the temperature , @xmath11 the specific volume and @xmath12 the specific entropy .
we consider the situation when the gas obeys a van der waals equations of state of the form @xmath13 where r is the gas constant , @xmath14 the ratio of specific heats , and @xmath15 the van der waals excluded volume . consider a weak shock hitting a wedge with half angle @xmath16 .
the state ahead of the shock is @xmath17 for some @xmath18 .
the state behind the shock is @xmath19 with @xmath20 .
so we seek a solution of the system ( [ equ1 ] ) with initial data @xmath21 and the slip boundary condition along the wedge @xmath22
since the coefficients in equations governing the physical process do not depend on variables , equations ( [ equ1 ] ) together with initial and boundary conditions ( [ equ2 ] ) and ( [ equ3 ] ) are invariant under the dilation @xmath23 , @xmath24 , @xmath25 , where @xmath26 is an arbitrary constant , and so we look for the solution with the property + i.e. , @xmath27 thus , by introducing the variables @xmath28 and @xmath29 , we express equations ( [ equ1 ] ) in self - similar polar form @xmath30 with initial and boundary conditions : @xmath31 and @xmath32 where @xmath8 is given by @xmath33 , and @xmath34 if @xmath3 and @xmath35 are chosen as independent variables , many calculations for euler system ( [ equ1 ] ) can be simplified ; for instance the speed of sound is @xmath36 and the energy equation @xmath37 , for smooth solutions , may be written as @xmath38 it may be noticed that the unsteady euler system ( [ equ1 ] ) , governing the gas flow , is hyperbolic ; however , the corresponding pseudo - stationary flow in self - similar coordinates is governed by mixed type equations . indeed , the system ( [ equ5 ] ) changes its type from elliptic to hyperbolic when the point @xmath39 runs from the origin to infinity . since the problem is symmetric with respect to the line @xmath40 , it suffices to consider the problem in the upper half plane , @xmath41 , outside the half wedge .
as the shock front hits the wedge head on and propagates further along the wedge , it is reflected by the wedge surface at @xmath42 ( see figure @xmath43 ) , whereas the induced flow behind the incident shock wave is diffracted by the wedge corner at sonic speed .
assuming that the shock reflection is a regular one , the location of the incident shock after it has moved beyond the domain of the influence of the origin ( vertex of the wedge ) is given by @xmath44 where @xmath45 ; the straight line segment @xmath46 of the reflected shock is given by @xmath47 the unknown curved portion @xmath48 , due to the influence of the origin , joins the diffracted wavefront smoothly and gives rise the overall shock reflection - diffraction phenomenon . here
@xmath49 is the fixed boundary referred to as the sonic arc , @xmath50 , across which there is a continuous transition from the supersonic region @xmath51 to the subsonic region @xmath52 , whereas @xmath48 is the free boundary , called diffraction of the planar shock , across which the transition undergoes a jump from the supersonic region @xmath53 to the subsonic region @xmath52 near the origin .
these wavefronts , referred to as boundaries , separate the upper half @xmath39-plane into four regions . @xmath54 from equation ( [ equ7 * ] ) , it may be noticed that @xmath55 , where @xmath56 ; this implies that an increase in @xmath57 causes the domain @xmath52 of linearized solution to become larger as compared to the ideal gas case .
+ hence in order to determine the entire flow field and the wave structure , one needs to solve the free boundary value problem for a degenerate elliptic equation .
the system ( [ equ5 ] ) has four eigenvalues @xmath58 @xmath59 with @xmath60 .
equations ( [ equ10 ] ) show that the system ( [ equ5 ] ) is hyperbolic with four eigenvalues and the flow is supersonic ; when @xmath61 , the system is mixed type as two equations in ( [ equ5 ] ) are hyperbolic and the other two are elliptic .
however , @xmath62 represents a sonic curve in @xmath39 plane . in general
, the system ( [ equ5 ] ) is mixed type and the flow is transonic .
in order to find the states behind the incident and reflected shocks , denoted by subscripts-1 and -2 respectively , we need the rankine - hugoniot ( r - h ) conditions in 2d .
let @xmath63 be a shock curve with slope @xmath64 .
then it follows from ( [ equ5 ] ) that @xmath65{\zeta}d{\theta}=\left[\rho{\mbox{\scriptsize{v}}}\right]d{\zeta},\\ & [ \rho(\mbox{\scriptsize{u}}-\zeta)^2+p]{\zeta}d{\theta}=[{\rho}(\mbox{\scriptsize{u}}-\zeta)\mbox{\scriptsize{v}}]d{\zeta},\\ & [ { \rho}(\mbox{\scriptsize{u}}-\zeta)\mbox{\scriptsize{v}}]{\zeta}d{\theta}=[\rho{\mbox{\scriptsize{v}}^2}+p]d{\zeta},\\ & [ \rho(\mbox{\scriptsize{u}}-\zeta)(h+\frac{\mbox{\scriptsize{u}}^2+\mbox{\scriptsize{v}}^2}{2})+{\zeta}p]{\zeta}d{\theta}=[\rho{\mbox{\scriptsize{v}}}(h+\frac{\mbox{\scriptsize{u}}^2+\mbox{\scriptsize{v}}^2}{2})]d{\zeta } , \end{split}\end{aligned}\ ] ] where square brackets , [ .
] , denote jumps across the shock .
let @xmath66 , and @xmath67 ( with @xmath68 ) be the components of pseudo - velocity vector @xmath69 along the tangent and normal to the shock curve @xmath63
. then equations ( [ equ11 ] ) may be written as @xmath70=0,~[{\rho}{q_t}{q_n}]=0,~[\rho{q_n}^2+p]=0,~\left[\rho{q_n}\left(h+\frac{{q_t}^2+{q_n}^2}{2}\right)\right]=0.\ ] ] when @xmath71 , it follows from @xmath72 that @xmath73 , showing thereby that the pseudo - flow is no longer tangential to the shock curve ; indeed , it moves from state-0 to state-1 satisfying the entropy condition @xmath74 .
let @xmath75 ( respectively , @xmath76 ) be the up - stream ( respectively , down - stream ) shock mach number relative to the up - stream ( respectively , down - stream ) flow .
( -4,-4)(4,-4 ) ; ( -1,-4)(-1,3 ) ; ( -1,-4)(7,-4 ) ; ( 1.8,3.5)(1.8,0 ) ; ( 1.8,0)(-1.5,-1.5 ) ; ( .5,2.5)(3,2.5 ) ; ( 2.7,3.7)(1.8,2.5 ) ; ( 1.8,2.5)(.9,1.5 ) ; ( 1.8,2.5)(1.2,1.4 ) ; ( 1.3,0)(-1,-1.5 ) ; ( -.9,-3.9)(-.8,-3.8 ) ; ( .7,.5)(.1,-.8 ) ; ( .1,-.8)(-.5,-2.1 ) ; ( .1,-.8)(-.5,-2.1 ) ; ( 2.1,2.5 ) arc ( 0:45:.35 ) ; ( 1.1,1.7 ) arc ( -10:70:.2 ) ; ( .2,-.6 ) arc ( 0:100:.25 ) ; ( -.5,-1.6 ) arc ( 0:80:.2 ) ; ( -.5,-4 ) arc ( 0:80:.35 ) ; ( 4.1,-4.1 ) node @xmath77 ; ( -1.1,3.2 ) node @xmath78 ; ( 2.1,-.1 ) node @xmath42 ; ( -1,-4.3 ) node @xmath79 ; ( -1.3,-1.1 ) node@xmath80 ; ( 1.6,3.6 ) node @xmath81 ; ( 3.1,3 ) node @xmath82 ; ( -.2,1.5 ) node @xmath83 ; ( -.6,-2.8 ) node @xmath84 ; ( 2.1,3.3 ) node @xmath85 ; ( 1.5,1.3 ) node @xmath86 ; ( .9,.2 ) node @xmath86 ; ( -.8,-2.2 ) node @xmath87 ; ( 2.3,2.7 ) node @xmath88 ; ( 1.1,1.4 ) node @xmath89 ; ( .1,-.1 ) node @xmath90 ; ( -.1,-2 ) node @xmath91 ; ( -.3,-3.8 ) node @xmath92 ; then , following @xcite , it can be shown that in a van der waals gas , the states @xmath93 and @xmath94 on the two sides of the incident shock are related as @xmath95 @xmath96 @xmath97 where @xmath98 , @xmath88 is the angle between the shock velocity vector @xmath99 and the shock normal with @xmath100 , and @xmath89 is the angle between @xmath101 and @xmath102 .
+ further , since @xmath103 and @xmath104 , equation ( [ equm ] ) implies that @xmath105 similarly , the r - h conditions for the reflected shock can be written as @xmath106 @xmath107 @xmath108 where @xmath109 , @xmath90 is the angle between the shock velocity vector @xmath110 and the shock normal with @xmath111 , and @xmath91 is the angle between @xmath102 and @xmath112 .
+ the boundary condition ( [ equ7 ] ) requires that the state @xmath113 be such that @xmath114 ; this condition together with r - h conditions determine the state-2 .
let @xmath115 and @xmath116 , then the requirement that the flow in state-2 is parallel to the wedge implies that @xmath117
we now consider @xmath118 and @xmath119 as independent variables in their respective domains , and derive the condition which ensures that the regular reflection takes place in the neighborhood of the reflection point @xmath42 .
on eliminating @xmath120 from ( [ equ23 ] ) and @xmath121 , we get @xmath122 substituting @xmath123 into @xmath124 , we get @xmath125}{\beta_i(\gamma+1)(1+{\beta_i}{\tan^2{\phi_i}})\sec^2{\phi_r}-2(1-\tilde{b}\beta_i)({\beta^2_i}{\tan^2{\phi_i}}-\tan^2{\phi_r})}.\ ] ] using @xmath126 and ( [ equ31 ] ) in ( [ equ29 ] ) , we obtain @xmath127=0 , \end{split}\end{aligned}\ ] ] which implies that either @xmath128 or @xmath129 with an appropriate branch , where @xmath130 is given by @xmath131 now using ( [ equ32 ] ) in ( [ equ30 ] ) , we get @xmath132 which violates entropy condition @xmath133 , so it needs to be discarded .
thus the required condition follows from ( [ equ33 ] ) if , and only if , @xmath134 is nonnegative , i.e. , @xmath135 assuming this to be true , equation ( [ equ30 ] ) when used in ( [ equ28 ] ) yields , @xmath136 the first inequality in ( [ equ35 ] ) , implies that @xmath137 which , in view of ( [ equ33 ] ) , yields @xmath138 it may be noticed that the inequality ( [ equ36 * ] ) is always true ; this , indeed , implies that ( [ equ28 ] ) follows from ( [ equ34 ] ) .
further it follows from ( [ equ33 ] ) that the plus branch for @xmath139 ( with @xmath140 ) yields @xmath141 , whereas the minus branch yields @xmath142 , implying thereby that the branch with plus sign needs to be discarded since it is irrelevant here ; for a valid solution , we use the minus branch for @xmath139 . in order to solve the inequality ( [ equ34 ] ) ,
we write @xmath130 in the following form @xmath143 where @xmath144 @xmath145 , @xmath146 , + @xmath147 , and + @xmath148 .
+ in view of inequality ( [ equ26 * ] ) and the fact that @xmath149 , it follows that @xmath150 , @xmath151 and @xmath152 are negative , whereas @xmath153 is positive . from ( [ equ37 ] )
, we have @xmath154 implying thereby that the interval @xmath155 must contain a zero of @xmath156 .
further , as the critical points of @xmath156 , given by @xmath157 are such that @xmath158 at @xmath159 and @xmath160 at @xmath161 , it follows that the only positive zero of @xmath156 is given by @xmath162 with @xmath163 and @xmath164 since @xmath165 = @xmath166 and @xmath167 = @xmath168 , the only positive zero of @xmath134 is given by @xmath169 .
in other words , the necessary condition for regular reflection to take place in the neighborhood of the reflection point @xmath42 follows from ( [ equ37 * ] ) , namely , @xmath170 leading us to conclude that there exists a critical @xmath171 of @xmath88 , in the interval @xmath172 , depending on @xmath118 , @xmath57 and @xmath173 , given by @xmath174 such that there exists a unique state @xmath113 for each @xmath175 with @xmath90 given by ( [ equ33 ] ) and satisfying the inequality ( [ equ28 ] ) . in the absence of real gas effects
@xmath176 , we recover the result obtained by chang and chen @xcite , who studied the problem of shock diffraction along a compressive corner .
+ the following table shows the effects of density ratio @xmath118 and @xmath57 on ` @xmath177 ' ; this , indeed , shows that an increase in @xmath118 or @xmath57 causes the critical value ` @xmath177 ' to increase , implying thereby that for a regular reflection to take place , an increase either in the shock strength @xmath118 or in the van der waals excluded volume requires the incident angle @xmath88 to be larger , relative to what it would have been in the absence of real gas effects .
+ .values of @xmath177 influenced by @xmath178 and @xmath57 . [ cols= " <
, < , < , < , < , < , < , < , < , < " , ]
as pointed out earlier that in a weak shock regular reflection , it is experimentally observed that the reflected shock is no longer rectilinear , it joins the diffracted wave front @xmath48 at a point @xmath179 that arises due to the influence of the compressive corner at @xmath79 . behind the reflected shock @xmath46
the flow is a uniform supersonic flow ; further downstream near the origin the flow is subsonic .
therefore , the state behind the reflected shock is not uniform and the system of governing equations , in self - similar coordinates , becomes degenerate on the boundary @xmath49 ( see figure @xmath43 ) ; so we look for an asymptotic solution to this problem , which is uniformly valid throughout the flow field .
the boundary @xmath49 is indeed a characteristic across which solution is continuous but discontinuities in its derivatives are permitted ; the r - h conditions for the reflected shock , derived in section-3 , provide the boundary conditions for the problem .
we consider the incident shock with states @xmath93 and @xmath94 on the two sides of it .
let @xmath180 be a dimensionless parameter measuring the shock strength , i.e. , @xmath181 then using ( [ equ7 * * ] ) , ( [ equ2 ] ) and ( [ equ5 * ] ) in ( [ equ11 ] ) , the r - h conditions for the incident shock are given by the following relations @xmath182 in view of ( [ equ41 ] ) and ( [ equ5 * ] ) , equations ( [ equ8 ] ) yield the following asymptotic expansions of the state-1 variables as @xmath183 : @xmath184 where @xmath185 , @xmath186 , @xmath187 , @xmath188 , @xmath189 , @xmath190 , @xmath191 , @xmath192 , @xmath193 , @xmath194 , @xmath195 and @xmath196 .
+ we look for the asymptotic expansions of the state-2 variables in the following form @xmath197 substituting ( [ equ7 * ] ) , ( [ equ42 ] ) , and ( [ equ44 ] ) into ( [ equ11 ] ) , we obtain the perturbed quantities as follows : @xmath198 thus , in view of ( [ equ45 ] ) , equations ( [ equ44 ] ) become @xmath199 it follows from ( [ equ6 ] ) , @xmath201 and @xmath202 that the solution , to the first order approximation , in regions @xmath200 and @xmath51 , is piecewise constant , i.e. , @xmath203 the solution ( [ equ50 ] ) shows that the point @xmath179 , where the reflected wave merges with the diffracted wave smoothly , divides the diffracted wavefront @xmath50 into two parts such that @xmath204 on @xmath49 and @xmath205 on @xmath48 . in order to seek a uniformly valid asymptotic solution to the problem under consideration , we look for asymptotic expansions of the form @xmath207 introducing the non - dimensional variable @xmath208 and inserting the asymptotic expansions ( [ equ47 ] ) into @xmath209 and ( [ equs ] ) , we get the following system of equations for the first order perturbation variables @xmath210 eliminating @xmath211 , @xmath212 and @xmath213 , equations([equ48 * ] ) yield the following equation for the unknown @xmath214 @xmath215 the solution of which satisfying the boundary condition ( [ equ50 ] ) at @xmath216 and the boundary condition ( [ equ7 ] ) along the wedge surface in terms of the first order variables , i.e. , @xmath217 , takes the form @xmath218 where @xmath219 , @xmath220 , @xmath221 $ ] , and @xmath222 , with @xmath223 on the wedge , indeed , in the limit of vanishing van der waals excluded volume @xmath224 , solution ( [ equ51 ] ) reduces exactly to the one obtained in @xcite .
the point @xmath179 , where the reflected wave @xmath46 merges into the diffracted wave tangentially , separates the front @xmath216 into two parts @xmath48 and @xmath49 , which are indeed the thin regions about the diffracted wave .
it follows from ( [ equ51 ] ) that the derivative of the linearized solution normal to @xmath48 is unbounded , whereas both the normal and tangential derivatives of the linearized solution are unbounded in the neighborhood of @xmath179 .
therefore , we need an asymptotic expansion near the wavefront @xmath216 in which nonlinear effects are significant ; note that ( [ equ49 ] ) becomes degenerate at @xmath216 .
the asymptotic behavior of ( [ equ51 ] ) near @xmath216 is given by @xmath225 which ceases to be valid at @xmath226 because @xmath227 ; we , therefore , need a different asymptotic expansion using the method of matched expansion .
we will discuss the asymptotic expansion for @xmath228 in section ( 5.6 ) .
+ in view of ( [ equ52 ] ) , the linear approximation of the solution near @xmath229 for @xmath230 , @xmath231 yields @xmath232 which , using the polar form of @xmath233 , can be written as @xmath234 equations @xmath209 and ( [ equs ] ) can be written in polar coordinates @xmath235 , assuming the fact that the wedge is symmetric about the line @xmath40 i.e. , @xmath236 where @xmath237 and @xmath238 are radial and rotational velocities defined in ( [ equ5 * ] ) .
it may be remarked that in order to account for the nonlinear effects near the diffracted wavefront and sonic curve @xmath216 , where the singularity occurs , we need to construct a new expansion when @xmath233 is close to @xmath239 .
+ we consider a uniform state @xmath240 , @xmath241 , @xmath242 , @xmath243 with @xmath244 as in ( [ equ42 ] ) or ( [ equ46 ] ) , into which a small amplitude wave is propagating and , following the ideas of weakly nonlinear geometrical acoustics @xcite , look for an asymptotic expansion for @xmath245 of the form : @xmath246 where @xmath247 is the ` fast ' variable with @xmath248 as a measure of the wave amplitude . equations ( [ equ60 * ] ) , in view of ( [ equ62 ] ) , yield at @xmath249 the following relations @xmath250 which , on using vector - matrix notation , can be written as @xmath251 where @xmath252 and @xmath253 is a @xmath254 matrix with @xmath255 , @xmath256 , @xmath257 , @xmath258 , @xmath259 , and the remaining entries being zero .
+ for a nontrivial solution of ( [ equ64 ] ) , we should have @xmath260 ; this implies that the phase function @xmath261 satisfies the following eikonal equation @xmath262 which can be solved using the method of characteristics , showing thereby that the characteristics or rays are straight lines in the @xmath263 plane , and @xmath264 is constant along each ray .
we shall label each ray with a parameter @xmath265 , which is constant along each ray @xmath266 .
thus , system ( [ equ64 ] ) yields @xmath267 where @xmath268 is the right null vector of @xmath42 and @xmath269 is an arbitrary scalar function , in which @xmath265 occurs as a parameter
. + similarly , equations ( [ equ60 * ] ) , to the order @xmath270 , yield @xmath271 where @xmath272 = @xmath273 , @xmath274 and @xmath275 are @xmath276 and @xmath254 matrices with @xmath277 , @xmath278 , @xmath279 , @xmath280 , @xmath281 , @xmath282 , @xmath283 , @xmath284 , @xmath285 , @xmath286 , @xmath287 , @xmath288 , and @xmath289 . contracting ( [ equ66 ] ) by the left null vector @xmath290 of @xmath42 , and using ( [ equ65 ] ) , and taking diffracted wavefront @xmath291
, we get @xmath292 it may be remarked that the above equation , in the absence of real gas effects , reduces to the cylindrical inviscid equation for an ideal gas reported in @xcite ; this , indeed , implies that @xmath293 along the characteristics given by @xmath294 where @xmath295 is an arbitrary function describing the wave profile through its dependence on @xmath265 , which is constant along each ray , and @xmath296 is a fast variable that parametrizes the characteristic curves given by ( [ equ68 ] ) . using ( [ equ69 ] ) in ( [ equ68 ] ) and keeping in mind that @xmath296 is constant on the solution curves of ( [ equ68 ] ) , we obtain @xmath297 where @xmath298 is an arbitrary function , which for convenience , can be replaced by @xmath299 with @xmath300 as a more convenient parameter .
subsequently , ( [ equ69 * ] ) can be expressed as @xmath301 which , for @xmath302 , gives @xmath300 implicitly , a solution for which can be multivalued ; this is , indeed , the nonlinearization technique introduced by landau @xcite and whitham @xcite , which accounts for the nonlinear effects by changing the phase function in the linear solution .
equation @xmath303 , in view of equations ( [ equ42 ] ) , ( [ equ46 ] ) , ( [ equ69 ] ) and ( [ equ65 ] ) , becomes @xmath304 now , for a uniformly valid solution , the linearized solution ( [ equ54 ] ) must match with nonlinear solution ( [ equ71 ] ) near the diffracted boundary @xmath305 , implying thereby @xmath306 where @xmath307 .
+ now , in view of @xmath308 , @xmath309 and @xmath310 , the solution near nonlinear diffracted wavefront @xmath311 is given by @xmath312 using @xmath313 into ( [ equ70 ] ) , we obtain @xmath314 which on differentiating with respect to @xmath300 , yields @xmath315 equations ( [ equ73 ] ) and ( [ equ73 * ] ) imply that the diffracted wavefront is either a shock or a rarefaction depending on the sign of @xmath316 . @xmath317 * case 1 .
when diffracted wave is rarefaction ( @xmath318 ) * : + in this case , equation ( [ equ73 ] ) for @xmath319 can be solved to yield a positive root @xmath320 which on using in ( [ equ72 ] ) gives @xmath321 where @xmath322 .
+ however for @xmath323 , in view of ( [ equ46 ] ) , we get @xmath324 , @xmath325 , @xmath326 , and @xmath327 .
indeed , when the diffracted wavefront @xmath311 is rarefaction , all the flow variables are continuous across it but the discontinuity occurs in their derivatives ; in view of ( [ equ75 ] ) , the jump in the density gradient @xmath328 , in radial direction , is given by @xmath329=\frac{(1-\tilde{b})\rho_0}{(\gamma+1)r},\ ] ] which shows that an increase in @xmath57 causes the density gradient to decrease , implying thereby that the rarefaction wave becomes weaker and decays slowly as compared to the corresponding ideal gas ( @xmath330 ) case ( see figure @xmath331 ) .
+ + * case 2 .
when diffracted wave is a shock ( @xmath332 ) * : + in this case , it is clear from ( [ equ73 * ] ) that the diffracted wavefront is compressive ; the neighboring characteristics intersect and the envelop of the multivalued region is given by ( [ equ73 ] ) and ( [ equ73 * ] ) .
the multivalued region is replaced by a shock , the position of which may be determined by the equal area rule @xcite .
let @xmath333 be the shock location with @xmath334 and @xmath335 as the characteristics on the two sides of the shock . then using the equal area rule , the equation for the shock location
may be written as @xmath336 which shows that the van der waals excluded volume @xmath57 affects the diffracted shock location as well as the shock velocity ; indeed , an increase in @xmath57 causes the velocity of diffracted shock to increase .
( see figure @xmath337 ) .
+ now , using ( [ equ77 ] ) and ( [ equ74 ] ) in @xmath338 , and matching with the boundary condition ( [ equ41 ] ) as @xmath339 , the density immediately behind the diffracted shock is obtained as @xmath340 it may be noticed that the density ahead of the shock is given by @xmath341 .
thus , the shock strength across the diffracted shock is given by @xmath342=\frac{{\epsilon^2}{c^2(\beta)}(\gamma+1)}{2(1-\tilde{b})}+o(\epsilon^3).\ ] ] from ( [ equ79 ] ) , it may be noticed that an increase in the van der waals excluded volume @xmath57 causes the strength of diffracted shock to increase relative to what it would have been for the ideal gas case @xmath176 ( see figure @xmath343 ) .
+ as noticed in section ( 5.4 ) that the asymptotic behavior of the diffracted wave , given by ( [ equ52 ] ) , breaks down in the neighborhood of @xmath179 , we construct asymptotic expansion valid in the neighborhood of @xmath179 by stretching the variables @xmath233 and @xmath78 ; to this end we introduce new variables @xmath344 and @xmath345 , where @xmath346 is the gauge function with @xmath347 to be determined . in terms of these new variables , the dominant part of ( [ equ49 ] ) , after simplification ,
results into @xmath348 at this point , based on the principle that the leading order equation should be kept as rich as possible so that the solution contains the maximum possible information , the only choice for @xmath349 , which gives a non - degenerate reduced problem and allows all the terms in ( [ equ52 * ] ) to be retained is @xmath350 . in terms of these new variables ( [ equ51 ] )
yields @xmath351 where @xmath352 .
+ accordingly , in the neighborhood of @xmath179 , we seek asymptotic expansions of the form : @xmath353 substituting ( [ equ80 ] ) into ( [ equ41 ] ) , and collecting respectively , @xmath249 , @xmath354 and @xmath355 terms , we get @xmath356 where @xmath357 is a vector of flow variables @xmath358 , + @xmath359 , @xmath360 , + and @xmath361 is a column vector with the components @xmath362 , @xmath363 , @xmath364 and @xmath365 .
let @xmath366 and @xmath367 be the right and left null vectors of @xmath368 , respectively ; then equation @xmath369 yields @xmath370 where @xmath371^t$ ] and @xmath372 is an arbitrary scalar valued function .
+ in view of ( [ equ82 ] ) , equation @xmath373 implies that @xmath374 where @xmath375 is an arbitrary scalar valued function satisfying the following relation @xmath376 now , using ( [ equ82 ] ) and ( [ equ84 ] ) in @xmath377 , we obtain @xmath378 it may be remarked that the pdes ( [ equ85 ] ) and ( [ equ86 ] ) bear a close structural resemblance with the self - similar ( utsd ) equations analyzed in @xcite .
it may be recalled that the system ( [ equ85])-([equ86 ] ) is the first approximation to the flow near the point @xmath179 ; in order to obtain a uniform solution valid throughout the flow field , boundary conditions for the system ( [ equ85])-([equ86 ] ) must be specified in conformity with ( [ equ42 ] ) , ( [ equ44 ] ) , and ( [ equ47 ] ) . on using @xmath379 and @xmath380 in ( [ equ7 * ] ) , we get the following approximations for the reflected shock in @xmath381 plane @xmath382 showing thereby that , in the neighborhood of @xmath179 , the straight reflected shock becomes parabolic in the limit @xmath383 ; we notice that an increase in the van der waals excluded volume @xmath57 causes an increase in its latus - rectum , indicating thereby that an increase in @xmath57 causes the real gas boundaries to become larger than the corresponding ideal gas case .
+ in view of ( [ equ42 ] ) , ( [ equ44 ] ) , ( [ equ53 ] ) , and ( [ equ88 ] ) , the boundary conditions for ( [ equ85])-([equ86 ] ) can be specified as @xmath384 let @xmath385 be the location of the reflected shock , which is a weak solution of the conservative system ( [ equ85])-([equ86 ] ) .
then the jump conditions across the reflected shock may be written as @xmath386+(d{s_{r}}/{d\theta^{'}})[u]=0.\ ] ] @xmath387+(d{s_{r}}/{d\theta^{'}})[v]-2\kappa_0 s_{r}[u]=0.\ ] ] on using ( [ equ90 ] ) in ( [ equ91 ] ) , we get @xmath388 where @xmath389 denotes the average value of @xmath390 on either side of the shock . the solution of the above equation , in view of the fact that the values of @xmath390 ahead and behind of @xmath80 are 1 and 2 , respectively , can be written as @xmath391 where @xmath392 is an arbitrary constant .
thus , a weak solution for the system ( ( [ equ85])-([equ86 ] ) ) satisfying the boundary conditons @xmath393 can be written as @xmath394 in a similar manner , the equation of the diffracted shock @xmath395 is obtained in the following form @xmath396 and the diffracted wave solution of the system ( ( [ equ85])-([equ86 ] ) ) satisfying the boundary conditons @xmath397 can be written as @xmath398 for smooth solutions , we can eliminate @xmath11 from ( [ equ85 ] ) and ( [ equ86 ] ) to obtain @xmath399 where @xmath400 .
+ it may be noticed that equation ( [ equ87 ] ) is of mixed type , namely , it is hyperbolic when @xmath401 , and elliptic when @xmath402 , however , when @xmath403 , it corresponds to two sonic lines , @xmath404 and @xmath405 .
indeed , at @xmath406 , the reflected shock starts bending and merges asymptotically into the diffracted shock @xmath395 ; however the sonic line @xmath407 is asymptotic to the diffracted shock @xmath395 in the neighborhood of the point @xmath408 ( see figure @xmath409 ) .
+ it may be observed that the streching transformation @xmath410 , @xmath411 , @xmath412 , for every parameter @xmath413 , leaves the equation ( [ equ87 ] ) invariant and , therefore , it admits a similarity solution of the form @xmath414 such that @xmath415 with @xmath416 ; further , as the homogeneous equation ( [ equ96 ] ) admits a solution of the form @xmath417 , an expansion wave solution of ( [ equ87 ] ) in the region @xmath418 between sonic lines @xmath407 and @xmath406 and satisfying the boundary conditions @xmath419 can be written as @xmath420 it may be noticed that an increase in @xmath57 not only causes the sonic lines @xmath421 and @xmath422 to shift along the positive @xmath423 direction , but it also increases the breadth between them ( see figure @xmath409 ) .
further , equation ( [ equ53 ] ) shows that an increase in @xmath57 serves to reduce the jump in the derivatives of flow variables across the expansion wave @xmath418 ; also , a change in the parabolic configuration in figure 4 leads us to reinforce our conclusion that the domain of the elliptic region in the neighborhood of the singular point @xmath408 exhibits an increase with an increase in the van der waals parameter @xmath57 .
in this article , we explore how the real gas effects influence the self - similar solutions of the compressible euler equations .
the regular reflection configuration and the detachment criterion , influenced by the real gas effects , are studied in detail .
a necessary condition is derived for the existence of regular reflection ; the manner in which it is influenced by the shock strength and the van der waals excluded volume , is clearly brought out . in the limit of vanishing van der waals excluded volume ,
the ideal gas case presented in the work of chang and chen @xcite , who studied the problem of shock diffraction along a compressive corner , is recovered .
it is shown that for a regular reflection to take place , there exists a critical value of the angle that the shock velocity vector makes with the shock normal ; it is found that an increase in the shock strength or in the van der waals excluded volume induces an increase in the critical value , implying thereby that an increase either in the shock strength or in the van der waals excluded volume requires the incident angle to be larger , relative to what it would have been in the absence of real gas effects .
we find that the reflected and diffracted regions as well as their boundaries , referred to as wavefronts , are significantly influenced by the real gas effects in the sense that an increase in the van der waals excluded volume fosters an expansion of the linearized solution domain .
as the state behind the reflected shock is not uniform , and the system of governing equations becomes degenerate on the boundary ( referred to as the sonic arc ) , across which there is a continuous transition from the supersonic region to the subsonic region , we look for a uniformly valid asymptotic approximations in the flow field .
following the ideas of weakly nonlinear geometrical acoustics @xcite , we construct weakly nonlinear solutions in these regions and match them with the linearized solution .
it is shown that if the diffracted wave is a rarefaction wave , it gets weakened by the real gas effects and decays slowly as compared to the corresponding ideal gas case . however ,
when the diffracted wave is a shock , we obtain an equation for its asymptotic location , showing thereby that the real gas effects serve to enhance the speed and strength of the diffracted shock wave .
asymptotic expansions are constructed near the singular point , where the @xmath355 approximation ceases to be valid ; these expansions lead to a pair of pdes , which bear a close structural resemblance with the self - similar ( utsd ) equations , analyzed in @xcite .
we obtain an equation for the asymptotic position of a reflected shock in the neighborhood of the singular point , reinforcing our conclusion that the real gas effects engender the real gas boundaries to inflate .
positions of the sonic lines , at which the reflected shock starts bending and the equations change type , are determined . it is shown that the asymptotic system of coupled equations , that hold near the singular point , admits an exact similarity solution ; a rarefaction wave solution , satisfying the specific boundary conditions , is obtained .
it is concluded that the real gas effects serve to weaken the rarefaction wave and to enlarge the diffracted wave region supporting our earlier viewpoint .
99 courant , r. , friedrichs , k. o. ( 1976 ) .
_ supersonic flow and shock waves , _ springer - verlag , newyork .
glimm , j. , majda , a. ( 1991 ) .
multidimensional hyperbolic problems and computations .
_ i m a volumes in mathematics and its applications _ , * 29 * , springer - verlag , newyork .
glass , i. i. , sislian , j. p. ( 1994 ) .
_ nonstationary flows and shock waves , _ oxford university press .
zheng , y. ( 2001 ) .
_ system of conservation laws : two - dimensional riemann problems , _ birkhauser , boston .
zheng , y. ( 2006 ) .
two - dimensional regular shock reflection for the pressure gradient system of conservation laws .
_ acta math .
_ , * 22 * , 177 - 210 .
ben - dor , g. ( 2007 ) .
_ shock wave reflection phenomena , _
2nd edition , springer - verlag , newyork .
chen , g. q. ( 2011 ) .
multidimensional conservation laws : overview , problems , and perspective .
nonlinear conservation laws and applications .
_ i m a vol .
_ , * 153 * , 23 - 72 .
chang , t. , hsiao , l. ( 1989 ) .
_ the riemann problem and interaction of waves in gas dynamics , _ john wiley and sons , new york .
keller , j. b. , blank , a. a. ( 1951 ) .
diffraction and reflection of pulses by wedges and corners .
pure appl .
_ , * 4 * , 75 - 94 .
keller , j. b. , hunter , j. ( 1984 ) .
weak shock diffraction .
_ wave motion _ , * 6 * , 79 - 89 .
hunter , j. , keller , j. b. ( 1983 ) .
weakly nonlinear high frequency waves .
pure appl .
_ , * 36 * , 547 - 569
. hunter , j. k. , majda , a. , rosales , r. ( 1986 ) . resonantly interacting , weakly nonlinear hyperbolic waves .
ii - several space variables .
_ std . in appl .
_ , * 75 * , 187 - 226 .
harabetian , e. ( 1987 ) .
diffraction of a weak shock by a wedge .
pure appl .
_ , * xl * , 849 - 863 .
morawetz , c. s. ( 1994 ) .
potential theory for regular and mach reflection of a shock at a wedge .
pure appl .
_ , * 47 * , 593 - 624 .
zheng , y. ( 1997 ) .
existence of solutions to the transonic pressure gradient equations of the compressible euler equations in elliptic regions .
pure appl .
_ , * 22 * , 1849 - 1868 .
rosales , r. r. , tabak , e. g. ( 1998 ) .
caustics of weak shock waves .
, * 10 * , 206 - 222 .
hunter , j. , brio , m. ( 2000 ) .
weak shock reflection .
_ j. fluid mech .
_ , * 410 * , 235 - 261
. canic , s. , keyfitz , b. l. , kim , e. h. ( 2002 ) .
a free boundary problem for a quasilinear degenerate elliptic equation : regular reflection of a weak shocks .
pure appl . math .
_ , * 65 * , 71 - 92 .
tesdall , a. m. , hunter , j. k. ( 2002 ) .
self - similar solutions for weak shock reflection .
_ siam j. appl .
_ , * 63 * , 42 - 61 .
cramer , m. s. , sen , r. ( 1987 ) .
exact solutions for sonic shocks in van der waals gases .
fluids _ , * 30 * , 370 - 385 .
kluwick , a. ( 1991 ) . _ nonlinear waves in real fluids , _ springer - verlag , berlin .
wu , c. c. , roberts , p. h. ( 1996 ) .
structure and stability of a spherical shock waves in a van der waals gas . _ quart j mech appl math _ ,
* 49 * , 501 - 543 .
arora , r. , sharma , v. d. ( 2006 ) . convergence of strong shock in a van der waals gas .
_ siam j. appl .
_ , * 66 * , 1825 - 1837 .
pandey , m. , sharma , v. d. ( 2007 ) .
interaction of a characteristic shock with a weak discontinuity in a non - ideal gas . _ wave motion _ , * 44 * , 346 - 354 .
pandey , m. , sharma , v. d. ( 2009 ) .
kinematics of a shock wave of arbitrary strength in a non - ideal gas .
_ , * 67 * , 401 - 418 .
chang , t. , chen , g. q. ( 1986 ) .
diffraction of planar shock along a compressive corner .
_ , * 6 * , 241 - 257
. hunter , j. k. , tesdall , a. m. ( 2004 ) .
_ weak shock reflection in ` a celebration of mathematical modeling ' , _ kluwer academic press , newyork .
landau , l. d. ( 1945 ) . on shock waves at large distances from their place of origin .
_ soviet j. physics _ , * 9 * , 496 - 500 .
whitham , g. b. ( 1973 ) . _
linear and nonlinear waves , _ john wiley and sons , new york .
hunter , j. k. , tesdall , a. m. ( 2012 ) . on the self - similar diffraction of a weak shock into an expansion wavefront .
_ siam j. appl .
_ , * 72 * , 124 - 143 . | the problem of a weak shock , reflected and diffracted by a wedge , is studied for the two - dimensional compressible euler system .
some recent developments are overviewed and a perspective is presented within the context of a real gas , modeled by the van der waals equation of state .
the regular reflection configuration and the detachment criterion are studied in the light of real gas effects .
some basic features of the phenomenon and the nature of the self - similar flow pattern are explored using asymptotic expansions .
the analysis presented here predicts several inviscid flow properties of the real gases undergoing shock reflection - diffraction phenomenon .
* keywords * : _ weak shock reflection , asymptotic expansion , nonlinear geometrical acoustics , van der waals excluded volume , r - h relations . _ |
high energy neutrino astrophysics is currently very active , in particular experimentally @xcite .
neutrino telescopes are reaching sensitivities comparable to theoretical expectations for neutrino fluxes based on their connection to primary cosmic rays and secondary @xmath0rays @xcite .
this is in particular the case for upper limits from amanda ii @xcite in an energy range between @xmath1ev and @xmath2ev , and from the rice experiment @xcite above @xmath3ev .
the former aims to detect neutrinos by looking for showers and/or tracks from charged leptons produced by charged current reactions of neutrinos in ice , whereas the latter is searching for radio pulses emitted by neutrino induced showers in south polar ice .
in addition , based on the non - observation of radio pulses from the earth s surface expected from neutrinos above @xmath4ev , the forte satellite has established upper limits on their fluxes in this hitherto unexplored territory @xcite . in the near future sensitivities will further improve by next generation versions of these techniques . for amanda this will be icecube @xcite at the south pole and possibly a comparable kilometer scale neutrino telescope in the mediterranean @xcite , based on antares @xcite , nemo @xcite , and nestor @xcite .
improved limits from the radio technique may come from the antarctic impulsive transient antenna ( anita ) which is a planned long duration balloon mission to detect radio waves from showers induced by neutrinos in the antarctic ice @xcite .
next generation experiments for ultra - high energy cosmic rays ( uhecr ) above @xmath5ev will also have considerable sensitivity to neutrinos , typically from the near - horizontal air - showers that are produced by them @xcite .
these projects include the southern site of the pierre auger observatory @xcite , a combination of a charged particle detector array with fluorescence telescopes for air showers produced by cosmic rays above @xmath5ev , and the telescope array @xcite , which may serve as the optical component of the northern pierre auger site .
there are also plans for space based observatories such as euso @xcite and owl @xcite of even bigger acceptance . finally , there are plans to construct telescopes to detect fluorescence and erenkov light from near - horizontal showers produced in mountain targets by neutrinos in the intermediate window of energies between @xmath6ev and @xmath5ev @xcite
. acoustic detection of neutrino induced interactions is also being considered @xcite . in an earlier paper @xcite we reviewed fluxes in various scenarios in the context of constraints from current cosmic ray data and upper limits on @xmath0ray and neutrino fluxes . besides the improved neutrino flux limits from amanda ii , rice , and forte , a possibly lower extragalactic contribution to the diffuse gev @xmath0ray background observed by the egret instrument on board
the compton @xmath0ray observatory @xcite has been pointed out recently .
an upper limit on the extragalactic diffuse @xmath0ray flux constrains the total amount of electromagnetic ( em ) energy injected above @xmath6ev which cascades down to below the pair production threshold for photons on the cosmic microwave background ( cmb ) @xcite . since in any scenario involving pion production
the em energy fluence is comparable to the neutrino energy fluence , a change in the constraint on em energy injection can also influence allowed neutrino fluxes . furthermore , future @xmath0ray detectors such as glast @xcite will test whether the diffuse extragalactic gev @xmath0ray background is truly diffuse or partly consists of discrete sources that could not be resolved by egret .
this could further improve the cascade limit . motivated by these improved constraints and prospects , and by more detailed information available on neutrino sensitivities of future experiments , in the present paper we reconsider flux predictions in scenarios currently often discussed in the literature . as in ref .
@xcite , we apply our recently combined propagation codes @xcite .
ii summarizes the numerical technique used in this paper . in sect .
iii we discuss the cosmogenic neutrino flux , i.e. the flux of neutrinos produced as secondaries of extragalactic cosmic rays during propagation , and its dependence on various uhecr source characteristics . in sect .
iv we review neutrino flux predictions in extragalactic top - down scenarios where uhecrs are produced in decays of super - massive particles continuously released from topological defect relics from the early universe . if uhecrs are new hadrons , they have to be produced as secondaries of accelerated protons which also gives rise to neutrinos .
their fluxes in these scenarios are reviewed in sect .
v. sect .
vi discusses primary neutrino fluxes required in scenarios where the cosmic rays observed at the highest energies are produced as secondaries from interactions with the relic cosmological neutrino background , often called z - burst scenario . in sect .
vii we discuss neutrino fluxes from active galactic nuclei ( agn ) models . finally , in sect .
viii we conclude .
our simulations use an implicit transport code that evolve the spectra of nucleons , @xmath0rays , electrons , electron- , muon- , and tau - neutrinos , and their anti - particles along straight lines .
arbitrary injection spectra and redshift distributions can be specified for the sources and all relevant strong , electromagnetic , and weak interactions have been implemented .
for details see refs .
our results apply to average large scale extragalactic magnetic fields of the order @xmath7 g and a universal radio background between the minimal values consistent with observations @xcite and moderately low theoretical estimates from radio source counts @xcite . for the neutrinos
we assume for simplicity that all three flavors are maximally mixed which for our purposes is an excellent approximation @xcite and thus have equal fluxes . for each flavor
we sum fluxes of particles and anti - particles . in the present investigation we parameterize power law injection spectra of either protons ( for uhecr sources ) or neutrinos ( for z - burst models ) per co - moving volume in the following way : @xmath8 where @xmath9 is the normalization that has to be fitted to the data .
the free parameters are the spectral index @xmath10 , the maximal energy @xmath11 , the minimal and maximal redshifts @xmath12 , @xmath13 , and the redshift evolution index @xmath14 .
the resulting neutrino spectra depend insignificantly on @xmath12 in the range @xmath15 where local effects could play a role , and thus we will set @xmath16 in the following . to obtain the maximal neutrino fluxes for a given set of values for all these parameters , we determine the maximal normalization @xmath9 in eq .
( [ para_inj ] ) by demanding that both the accompanying nucleon and @xmath0ray fluxes are below the observed cosmic ray spectrum and the diffuse @xmath0ray background observed by egret , respectively .
the flux of `` cosmogenic '' neutrinos is created by decaying charged pions produced in interactions of primary nucleons of energy above @xmath17ev with cmb photons , the greisen - zatsepin - kuzmin ( gzk ) effect @xcite .
this flux depends on the production rate of the primary nucleons which we parameterize according to eq .
( [ para_inj ] ) . '' ) on the contribution of the accompanying photon flux ( labeled `` @xmath18 '' ) to the old ( upper error bars on the left ) and new ( lower error bars on the left ) diffuse egret flux estimate .
the proton primary parameters in eq .
( [ para_inj ] ) have been fixed to @xmath19 , @xmath20 , and @xmath21 , whereas @xmath11 was varied .
the uhecr proton flux ( labeled `` p '' ) is normalized to the data of the agasa @xcite and hires @xcite experiments .
the numbers indicate the fraction of the new egret estimate of the extragalactic diffuse @xmath0ray flux contributed by the respective scenario , where the unlabeled thick curves correspond to 100%.,scaledwidth=48.0% ] we first consider the dependence of the cosmogenic neutrino flux on the contribution of the accompanying photons to the diffuse @xmath0ray flux in the 100mev - 100 gev region . as an example , we do this by fixing @xmath19 , @xmath20 , and @xmath21 in eq .
( [ para_inj ] ) , while varying the maximal energy @xmath11 .
the relatively hard proton spectrum in this scenario could be produced , for example , by acceleration in potential drops or reconnection @xcite .
we normalize the resulting uhecr flux to the observations above @xmath22ev and note that the discrepancy between the agasa @xcite and hires @xcite fluxes above @xmath23ev has a negligible influence on the predicted cosmogenic neutrino flux . as fig .
[ f1 ] shows , a decrease of the diffuse photon flux results from a decrease of @xmath11 in this case .
the old egret flux estimate @xcite corresponds to @xmath24ev , whereas the @xmath25% smaller new egret flux @xcite corresponds to @xmath26ev .
most likely only a fraction of the measured diffuse photon background is connected to gzk neutrinos . in the present scenario 0.5 and 0.2 of the flux measured by egret
corresponds to @xmath27ev and @xmath28ev , respectively .
note that the uhecr proton flux is the same in all cases , except for the highest not yet observed energies where the flux can be affected by the distance to the nearest sources .
the cosmogenic neutrino flux has recently been re - evaluated also in ref .
@xcite where the egret constraint has not been taken into account .
the latter , however , eliminates a considerable part of the higher fluxes considered there . ) to @xmath20 , @xmath29ev , @xmath19 , @xmath21 . also shown
are predicted and observed cosmic ray and @xmath0ray fluxes , the atmospheric neutrino flux @xcite , as well as existing upper limits on the diffuse neutrino fluxes from macro @xcite , amanda ii @xcite , baikal @xcite , agasa @xcite , the fly s eye @xcite and rice @xcite experiments , and the limits obtained with the goldstone radio telescope ( glue ) @xcite and the forte satellite @xcite , as indicated .
the cosmic ray data are as in fig .
[ f1 ] , whereas only the new egret flux is shown to the left.,scaledwidth=48.0% ] in comparison with expected sensitivities of the currently being constructed pierre auger project to tau - neutrinos @xcite , the planned projects telescope array ( ta ) @xcite , the fluorescence / erenkov detector nutel @xcite , the space based euso @xcite , the water - based baikal @xcite and antares @xcite ( the nestor sensitivity for 1 tower would be similar to amanda - ii and for 7 towers similar to antares @xcite ) , the ice - based amanda - ii @xcite and icecube @xcite ( similar to the intended mediterranean km@xmath30 project @xcite ) , and the radio detectors rice @xcite and anita @xcite , as indicated .
all sensitivities except for anita and rice refer to one year running time . for comparison ,
the @xmath0ray bound derived from the egret gev @xmath0ray flux @xcite is also shown.,scaledwidth=48.0% ] in comparison with differential sensitivities expected for 2006 for the pierre auger project @xcite ( for two years of data taking , assuming strong ( upper curve ) and no ( lower curve ) deep inelastic tau lepton scattering ) and the anita project @xcite for 10 ( upper curve ) and 30 ( lower curve ) days of observation .
the corresponding number of events expected are listed in tab .
[ tab].,scaledwidth=48.0% ] the two major categories of experiments are based on detection in water , ice or underground , typically sensitive below @xmath31ev , and on air shower detection , usually sensitive at higher energies .
existing neutrino flux upper limits come from the underground macro experiment @xcite at gran sasso , amanda ii @xcite in the south pole ice , and the lake baikal neutrino telescope @xcite in the first category , and the agasa ground array @xcite , the former fluorescence experiment fly s eye @xcite , the radio ice erenkov experiment rice @xcite ( there is also a limit from the hires experiment @xcite which is between the rice and agasa limits ) , the goldstone lunar ultra - high energy neutrino experiment glue @xcite , and the fast on - orbit recording of transient events ( forte ) satellite @xcite in the second category . as an example
, an optimistic cosmogenic neutrino flux is compared with current neutrino flux upper limits in fig .
future experiments in the first category include nt200 + at lake baikal @xcite , antares @xcite , nestor in greece @xcite , as well as a possible common km@xmath30 scale detector in the mediterranean @xcite , and icecube @xcite , the next - generation version of amanda at the south pole .
the air shower based category includes the pierre auger project @xcite , the telescope array @xcite , the fluorescence / erenkov detector nutel @xcite , and the space based euso @xcite and owl @xcite experiments .
the euso sensitivity estimate used here is based on deeply penetrating atmospheric showers induced by electron or muon - neutrinos only @xcite and may thus be considerably better if tau neutrinos , erenkov events , and earth skimming events are taken into account @xcite , for which there are no final estimates available yet .
the same applies to the owl project @xcite .
the cosmogenic neutrino flux models shown in fig
. [ f1 ] are compared with future sensitivities in fig .
[ f3 ] .
the fluxes shown in figs . [ f3 ] and [ f4 ] are considerably higher than the ones discussed in refs .
@xcite , and should be easily detectable by at least some of these future instruments , as demonstrated by the expected number of events listed in tab .
[ tab ] .
.[tab ] the number of tau neutrino ( pierre auger ) and electron neutrino ( anita ) events expected to be measured by 2006 for the four neutrino fluxes shown in figs . [ f3 ] and [ f4 ] . the pierre auger differential sensitivity shown assumes two years of data taking until 2006 . the minimal ( upper line in fig .
[ f4 ] ) and maximal ( lower line in fig . [ f4 ] ) sensitivity depends on the assumptions made for the strength of deep inelastic scattering of tau leptons .
for anita the case for balloon flights of 10 and 30 days are shown . [ cols="^,^,^,^,^",options="header " , ] we note that the non - observation of gzk neutrinos in 2006 will significantly restrict the accompanying photon contribution to the egret diffuse flux .
historically , top - down ( td ) scenarios were proposed as an alternative to acceleration scenarios to explain the huge energies up to @xmath32ev observed in the cosmic ray spectrum @xcite . in these top - down scenarios
uhecrs are the decay products of some super - massive `` x '' particles of mass @xmath33ev close to the grand unified scale , and have energies all the way up to @xmath34 .
thus , the massive x particles could be meta - stable relics of the early universe with lifetimes of the order the current age of the universe or could be released from topological defects that were produced in the early universe during symmetry - breaking phase transitions predicted by in grand unified theories ( guts ) .
the x particles typically decay into leptons and quarks .
the quarks hadronize , producing jets of hadrons which , together with the decay products of the unstable leptons , result in a large cascade of energetic photons , neutrinos and light leptons with a small fraction of protons and neutrons , some of which contribute to the observed uhecr flux .
the resulting injection spectra have been calculated from qcd in various approximations , see ref .
@xcite for a review and ref .
@xcite for more recent work . in the present work
we will use the qcd spectra discussed in ref .
@xcite and shown in fig .
. @xcite . for the purposes of the current work
this is not expected to make a significant difference as compared to the more accurate fragmentation spectra @xcite . ,
@xmath35gev . the contribution of photons and protons to the uhecr flux decreases with decreasing fractional contribution to the diffuse photon flux at egret energies which is denoted in numbers .
even a td contribution to the present estimate of the diffuse egret flux as high as 100% ( unlabeled thick curves ) is only marginally consistent with the agasa uhecr excess .
the line key is the same as in fig .
[ f5 ] ] for dimensional reasons the spatially averaged x particle injection rate can only depend on the mass scale @xmath36 and on cosmic time @xmath37 in the combination @xmath38 where @xmath39 and @xmath40 are dimensionless constants whose value depend on the specific top - down scenario @xcite .
extragalactic topological defect sources usually predict @xmath41 , whereas decaying super - heavy dark matter ( shdm ) @xcite of lifetime much larger than the age of the universe corresponds to @xmath42 @xcite .
it has also been suggested that annihilation of shdm particles instead of their decay might contribute to the observed uhecrs @xcite . in the shdm scenario
the observable uhecr flux will be dominated by the decay or annihilation products of shdm in the galactic halo and thus by sources at distances smaller than all relevant interaction lengths .
composition and spectra will thus be directly given by the injection spectra which are dominated by photons .
this is most likely inconsistent with upper limits on the ultra - high energy ( uhe ) photon fraction above @xmath22ev @xcite .
however , requiring that this scenario explains only uhecrs above @xmath43ev allows to avoid this problem @xcite .
a more severe problem of the shdm scenario is the spatial anisotropy of the expected signal predicted due to the non - central position of the sun in our galaxy @xcite . in a recent paper
@xcite it was shown that the non - observation of anisotropy in the data of the sugar experiment excludes any shdm scenario at the @xmath44 level assuming that all events above @xmath43ev are from shdm sources .
for the extreme case where shdm is responsible only for uhecrs above @xmath45ev , the annihilation scenario is disfavored at least at 99% cl by the sugar data , while decaying shdm still has a probability of @xmath46 to be consistent with the sugar data .
the shdm scenario is therefore disfavored by present experimental data and will be finally tested by the pierre auger experiment in the near future .
we will therefore here focus on topological defect models with @xmath41 . compared to future experimental sensitivities .
the line key is as in fig .
[ f6 ] ] fig .
[ f5 ] shows the results for @xmath35gev , with @xmath47 g , and the moderately low theoretical estimate from ref .
these parameters lead to optimistic neutrino fluxes for the maximal normalization consistent with all data . for detailed earlier discussions of extragalactic top - down fluxes see refs .
@xcite .
[ f5 ] shows that already the improved upper limit on the true diffuse photon background by egret implies too small a uhecr flux compared to the agasa excess at energies @xmath48ev .
the parameters used in the figure represent the `` best fit point '' for this model ; in particular for all other masses @xmath36 the disagreement is more severe . the new egret upper limit thus strongly disfavors extragalactic top - down scenarios . in addition , independently of this problem of overproduction of gev @xmath0rays , a non - observation of td neutrinos by 2006 will rule out the possibility that extragalactic top - down mechanisms significantly contribute to the uhecr flux , as can be seen from comparing fig . [ f6 ] with fig . [ f5 ] .
, @xmath20 , compared to expected experimental sensitivities . the solid line is for a flux of primary protons peaked at @xmath49ev and the dashed line is for a primary proton flux @xmath50 up to @xmath51 ev . the line key is as in fig .
[ f7 ] ] supersymmetric ( susy ) models with a strongly interacting particle as lightest supersymmetric particle ( lsp ) or next - to - lightest susy particle ( nlsp ) are very interesting for explaining the agasa excess above the gzk cutoff .
hadrons containing a gluino were first suggested by farrar as uhecr primaries @xcite .
this model with a light gluino together with a light photino as cold dark matter candidate is meanwhile excluded @xcite
. however , more general models with a light gluino or a light sbottom quark are still viable . in a recent paper
@xcite a model - independent , purely phenomenological approach was developed .
since the observed extensive air showers ( eas ) are consistent with simulated eas initiated by protons , any new primary proposed to solve the gzk puzzle has to produce eas similar to those of protons . experimentally still allowed
are photons as uhecr primaries : at 90% cl , @xmath52 of uhecrs above @xmath5ev can be photons @xcite . however , the simplest possibility consistent with eas observations is to require that the new primary is strongly interacting .
the requirements of efficient production in astrophysical accelerators as well as proton - like eas in the atmosphere ask for a light hadron , @xmath53gev , while shifting the gzk cutoff to higher energies results in a lower bound for its mass , @xmath54gev @xcite . from these requirements
general conditions on the interactions of new uhe primaries were derived .
the production of new hadrons in astrophysical objects was investigated in ref .
it was found that proton - proton collisions in astrophysical accelerators can not produce sufficiently high fluxes of new primaries without contradicting existing measurements of photon @xcite and neutrino fluxes @xcite .
in contrast , for a light shadron with mass @xmath53gev and the astrophysically more realistic case of uhe proton collisions on optical / infrared background photons there is no contradiction with existing limits .
also , the required initial proton energy is not too extreme , @xmath55ev , which may be achieved by astrophysical acceleration mechanisms .
the only essential condition for the sources is that they should be optically thick for protons in order to produce these new hadrons .
this condition applies to all models with new particles produced by protons .
one of the important features of scenarios with new hadrons , and of any model in which the production cross section @xmath56 of a new particle @xmath57 is much smaller than the total proton - photon cross section @xmath58 , is the high flux predicted for secondary high - energy neutrinos .
this neutrino flux is approximately @xmath59 in terms of the maximal contribution of @xmath57 particles to the observed cosmic ray flux , @xmath60 ev/(@xmath61 s sr ) .
[ f7 ] shows that for a primary proton flux @xmath50 , this model can be restricted already with 3 years of amanda - ii data .
a non - observation of neutrinos by 2006 will make it impossible to render the production cross section of the new hadrons consistent with existing limits in these scenarios .
in the z - burst scenario uhecrs are produced by z - bosons decaying within the distance relevant for the gzk effect .
these z - bosons are in turn produced by uhe neutrinos interacting with the relic neutrino background @xcite .
if the relic neutrinos have a mass @xmath62 , z - bosons can be resonantly produced by uhe neutrinos of energy @xmath63 .
the required neutrino beams could be produced as secondaries of protons accelerated in high - redshift sources .
the fluxes predicted in these scenarios have recently been discussed in detail in refs .
@xcite . in fig .
[ f8 ] we show an optimistic example taken from ref .
@xcite . as in refs .
@xcite no local neutrino over - density was assumed .
the sources are assumed to not emit any @xmath0rays , otherwise the z - burst model with acceleration sources overproduces the diffuse gev @xmath0ray background @xcite .
we note that no known astrophysical accelerator exists that meets the requirements of the z - burst model @xcite . ,
@xmath64 , @xmath21 , @xmath65 , @xmath66ev in eq . ( [ para_inj ] ) for neutrino primaries .
the sources are assumed to be exclusive neutrino emitters .
all neutrino masses were assumed equal with @xmath67 ev and we again assumed maximal mixing between all flavors .
the line key is as in fig .
[ f8 ] ] however , a combination of new constraints allows to rule out the z - burst mechanism even for pure neutrino emitting sources : a combination of cosmological data including the wmap experiment limit the sum of the masses of active neutrinos to @xmath68ev @xcite .
solar and atmospheric neutrino oscillations indicate that individual neutrino masses are nearly degenerate on this scale @xcite , and thus the neutrino mass per flavor must satisfy @xmath69ev . however , for such masses phase space constraints limit the possible over - density of neutrinos in our local group of galaxies to @xmath70 on a length scale of @xmath71mpc @xcite .
since this is considerably smaller than the relevant uhecr loss lengths , neutrino clustering will not significantly reduce the necessary uhe neutrino flux compared to the case of no clustering . for the maximal possible value of the neutrino mass @xmath72ev the neutrino flux required for
the z - burst model is only in marginal conflict with the forte upper limit @xcite , as shown in fig .
[ f8 ] . for all other cases
the conflict is considerably more severe .
also note that this argument does not depend on the shape of the low energy tail of the primary neutrino spectrum which could thus be even monoenergetic , as could occur in exlusive tree level decays of superheavy particles into neutrinos @xcite .
however , in addition this possibility has been ruled out by overproduction of gev @xmath0rays due to loop effects in these particle decays @xcite . as was discussed in ref .
@xcite , the z burst scenario involving normal astrophysical sources producing neutrinos and photons by pion production within the source were already ruled out by the former egret limit .
ray flux to the new diffuse egret flux estimate .
the line key is as in fig .
[ f9 ] ] active galactic nuclei ( agn ) are very promising sites of particle acceleration .
though there is no direct evidence of proton acceleration in these objects , according to the hillas condition @xmath73 for the maximal energy , where @xmath74 is the charge , @xmath75 is the magnetic field and @xmath76 is the linear size of the acceleration region , agn cores , jets or hot spots can be sites for acceleration of uhecrs up to the highest energies @xmath77 ev . once accelerated , protons can escape the agn freely or can lose part of their energy in interactions with background protons and photons .
high energy neutrinos can be produced in agn via pion production by accelerated protons .
the total power of a given object in neutrinos can be related to the total power in @xmath0rays of mev - gev energies .
however , such a connection is not straightforward because the observed high energy @xmath0rays can be produced by several mechanisms not involving neutrino production , for example proton synchrotron radiation in magnetic fields and inverse compton scattering of mev - gev electrons .
moreover , the spectrum of neutrinos from agn are even more difficult to predict than the spectrum of cosmogenic neutrinos .
the reason is that besides the unknown model - dependent spectrum of primary protons , the spectrum of background photons is also not known and in general is model dependent .
also , agn are divided into subclasses with different properties of the observed photon spectrum .
many of those properties do not directly relate to the possible neutrino spectrum .
this means that it is very difficult to predict the space distribution of those agn which contribute to the neutrino flux from the distribution of agn subclasses .
at least these distributions can be very model dependent . due to the above complications we suggest here a phenomenological approach to the prediction of neutrino fluxes . within this approach the neutrino flux in most of agn models
can be approximately characterized by three parameters , namely the amplitude , width and position of the peak ( plateau ) in the differential spectrum .
the position of the peak ( plateau ) is related to the spectrum of background photons .
the combination of amplitude and width defines the total power in neutrinos which can be related to the total power in mev - gev photons produced in the same pion production reactions .
experimental bounds on neutrino fluxes can be converted to constraints on model parameters .
a similar approach can be used for most existing models which predict neutrino fluxes from agn , see for example ref .
the only difference would be the connection of the phenomenological parameters of the observed neutrino spectrum to the physical agn parameters in the given model .
as an example , we will use the model of @xmath0ray powered jets of ref .
@xcite . in this model
the high energy @xmath18-rays are produced by accelerated protons interacting with the ambient photon fields ( supplied , for example , by the accretion disk around the massive black hole ) through photo - meson processes . at the same time
those protons produce neutrinos which are emitted in the direction of the jet .
therefore , this model predicts a high neutrino flux comparable in power with the @xmath18-ray flux .
the detailed numerical simulations of proton acceleration in the central engine of the agn @xcite show that the collimated jet of almost mono - energetic high energy protons ( linear accelerator ) can be created in the electro - magnetic field around the black hole and the energy of those protons can be converted into photons and neutrinos , while protons can be captured inside the source .
the nucleon flux leaving the agn is well below the observed cosmic ray flux in this scenario .
furthermore , since all nucleons leaving the source are well below the gzk cutoff , there is no cosmogenic contribution to the neutrino flux from these sources .
[ f9 ] shows a typical prediction for the diffuse neutrino flux in this model .
the neutrino flux is maximized in such a way that the accompanying photon flux saturates the new egret bound . as seen from fig .
[ f9 ] , already now amanda - ii and rice data start to restrict agn models .
three years of amanda - ii data will significantly restrict the parameter space of those models . this will also restrict the contribution of @xmath78 production in agn to the extragalactic diffuse @xmath0ray flux at egret energies . in the agn model discussed above ,
blazars would be seen by neutrino telescopes as point - like sources with neutrino fluxes which are smaller or of the same order as the photon flux emitted by these same sources and which are detectable by @xmath18-ray telescopes .
the most probable sources were discussed in ref .
based on our transport code we reconsidered neutrino flux predictions and especially their maxima consistent with all current data on cosmic rays and updated upper limits on neutrino fluxes and the diffuse extragalactic gev @xmath0ray background .
we discussed predictions for fluxes of cosmogenic neutrinos produced through pion production of uhecrs during propagation , and for fluxes produced by agn .
we showed that extragalactic top - down scenarios can not contribute significantly to the observed ultra - high energy cosmic ray flux if standard evolution histories and injection spectra are used .
the z - burst mechanism where ultra - high energy neutrinos produce cosmic rays via interactions with relic neutrinos is ruled out except if cosmological neutrino mass limits are invalid and/or if relic neutrinos cluster more strongly than expected based on standard phase space principles .
only the case of maximal neutrino mass @xmath79ev consistent with large scale structure and cmb observations is only moderately excluded .
the fact that a good part of the speculative scenarios of uhecr origin are now ruled out makes the enigma of the highest energy particles an even more exciting subject of study in our opinion .
we acknowledge sergey troitsky and michael kachelrie for useful comments on a draft version of this paper .
we would like to thank cyrille barbot and pasquale blasi for fruitful discussions .
we also thank douglas mckay , teresa montaruli and cristian spiering for correspondence on the rice , antares and amanda - ii experiments , respectively .
this work was supported in part by the aci `` jeunes chercheurs '' entitled `` la cosmologie : un laboratoire pour la physique des hautes nergies '' awarded by the french research ministry .
the work of d.s . was supported by deutsche forschungsgemeinschaft ( dfg ) within the emmy noether program .
g.s . acknowledges hospitality and financial research during his stay at the max - planck institut fr physik in munich . c. spiering , j. phys .
g * 29 * , 843 ( 2003 ) [ arxiv : astro - ph/0303068 ] .
for a recent review see , e.g. , f. halzen and d. hooper , rept .
phys . * 65 * , 1025 ( 2002 ) [ arxiv : astro - ph/0204527 ] .
p. niessen [ the amanda collaboration ] , arxiv : astro - ph/0306209 ; for the energy range @xmath80ev we rescaled the amanda - b10 limit from http://www-rccn.icrr.u-tokyo.ac.jp/icrc2003/proceedings/pdf/326.pdf to amanda - ii exposure .
i. kravchenko _ et al .
_ , arxiv : astro - ph/0206371 . i. kravchenko , arxiv : astro - ph/0306408 . n. g. lehtinen , p. w. gorham , a. r. jacobson and r. a. roussel - dupre , arxiv : astro - ph/0309656 . for general information see http://icecube.wisc.edu/ ; see also f. halzen : am . astron . soc . meeting 192 , # 62 28 ( 1998 ) ; amanda collaboration : astro - ph/9906205 , proc .
_ 8@xmath81 international workshop on neutrino telescopes _ , venice , feb .
see , e.g. , talk by u. katz on hena 2003 workshop , http://antares.in2p3.fr/antares/stolar/km3/program.htm . j. j. blanco - pillado , r. a. vazquez and e. zas , phys .
* 78 * , 3614 ( 1997 ) [ astro - ph/9612010 ] ; k. s. capelle , j. w. cronin , g. parente and e. zas , astropart .
* 8 * , 321 ( 1998 ) [ astro - ph/9801313 ] ; a. letessier - selvon , nucl .
suppl . * 91 * , 473 ( 2000 ) [ astro - ph/0009416 ] ; x. bertou , p. billoir , o. deligny , c. lachaud and a. letessier - selvon , astropart .
* 17 * , 183 ( 2002 ) [ astro - ph/0104452 ] .
j. w. cronin , nucl .
* 28b * ( 1992 ) 213 ; the pierre auger observatory design report ( 2@xmath82 edition ) , march 1997 ; see also http://www.auger.org/ and http://www-lpnhep.in2p3.fr/auger/welcome.html proc .
25th _ international cosmic ray conference _ , eds . :
m. s. potgieter et al .
( durban , 1997 ) .
proc . of international symposium on
_ extremely high energy cosmic rays : astrophysics and future observatories _ , ed
. m. nagano ( institute for cosmic ray research , tokyo , 1996 ) .
j. f. ormes et al . , in @xcite , vol . 5 , 273 ; y. takahashi et al . , in @xcite , p. 310 ; see also http://lheawww.gsfc.nasa.gov / docs / gamcosray / hecr / owl/. see , e.g. , d. fargion , m. khlopov , r. konoplich , p. g. de sanctis lucentini , m. de santis and b. mele , arxiv : astro - ph/0303233 . g. w. hou and m. a. huang , arxiv : astro - ph/0204145 .
see , e.g. , l. g. dedenko , i. m. zheleznykh , s. k. karaevsky , a. a. mironovich , v. d. svet and a. v. furduev , bull .
* 61 * ( 1997 ) 469 [ izv . ross . akad .
* 61 * ( 1997 ) 593 ] .
o. e. kalashev , v. a. kuzmin , d. v. semikoz and g. sigl , phys .
d * 66 * , 063004 ( 2002 ) [ arxiv : hep - ph/0205050 ] .
p. sreekumar _ et al .
_ , astrophys .
j. * 494 * , 523 ( 1998 ) [ astro - ph/9709257 ] .
a. w. strong , i. v. moskalenko and o. reimer , arxiv : astro - ph/0306345 .
u. keshet , e. waxman and a. loeb , arxiv : astro - ph/0306442 .
v. s. berezinsky and a. yu . smirnov , ap .
sci . * 32 * , 461 ( 1975 ) .
o. e. kalashev , v. a. kuzmin and d. v. semikoz , astro - ph/9911035 ; mod .
a * 16 * , 2505 ( 2001 ) [ astro - ph/0006349 ] .
o. e. kalashev , v. a. kuzmin , d. v. semikoz , i. i. tkachev , astro - ph/0107130 . o. e. kalashev , v. a. kuzmin , d. v. semikoz and g. sigl , phys . rev .
d * 65 * , 103003 ( 2002 ) [ arxiv : hep - ph/0112351 ] .
t. a. clark , l. w. brown , and j. k. alexander , nature * 228 * , 847 ( 1970 ) .
r. j. protheroe and p. l. biermann , astropart .
* 6 * , 45 ( 1996 ) [ erratum - ibid . *
7 * , 181 ( 1996 ) ] [ astro - ph/9605119 ] .
y. fukuda _ et al .
_ [ super - kamiokande collaboration ] , phys . rev .
81 ( 1998 ) 1562 ; q. r. ahmad _ et al .
_ [ sno collaboration ] , phys . rev .
lett . 87 ( 2001 ) 071301 .
see , e.g. , m. maltoni , t. schwetz , m. a. tortola and j. w. valle , arxiv : hep - ph/0309130 . k. greisen , `` end to the cosmic ray spectrum ?
, '' phys .
* 16 * , 748 ( 1966 ) .
g. t. zatsepin and v. a. kuzmin , jetp lett .
* 4 * , 78 ( 1966 ) [ pisma zh .
fiz . * 4 * , 114 ( 1966 ) ] .
m. takeda _ et al .
_ , phys .
lett . * 81 * , 1163 ( 1998 ) , [ astro - ph/9807193 ] ; see n. hayashida _ et al . _ , astro - ph/0008102 , for an update ; see also http : //www - akeno.icrr.u - tokyo.ac.jp / agasa/. d. kieda et al . , proc . of the 26th icrc , salt lake , 1999 ; http : //www.physics.utah.edu / resrch.html .
see , e.g. , s. a. colgate , phys .
scripta * t52 * , 96 ( 1994 ) .
z. fodor , s. d. katz , a. ringwald and h. tu , arxiv : hep - ph/0309171 .
see , e.g. , p. lipari , astropart .
* 1 * , 195 ( 1993 ) . for general information
see http://wsgs02.lngs.infn.it:8000/macro/ ; see also m. ambrosio _ et al .
_ [ macro collaboration ] , astro - ph/0203181 .
v. balkanov _ et al . _ [ baikal collaboration ] , astro - ph/0112446 ; for general information see http://www-zeuthen.desy.de/baikal/baikalhome.html s. yoshida for the agasa collaboration , proc .
of 27th icrc ( hamburg ) * 3 * , 1142 ( 2001 ) .
r. m. baltrusaitis _ et al .
_ , astrophys .
j. * 281 * , l9 ( 1984 ) ; phys .
d * 31 * , 2192 ( 1985 ) .
proceedings of _ first international workshop on radio detection of high - energy particles _ , amer .
inst . of phys . ,
vol . 579 ( 2001 ) , and at http://www.physics.ucla.edu/ moonemp / radhep / workshop.html .
for general information see http://antares.in2p3.fr ; see also antares collaboration , e - print astro - ph/9907432 ; t. montaruli _ et al .
_ [ antares collaboration ] , arxiv : physics/0306057 .
i. kravchenko et al .
( rice collaboration ) , e - print astro - ph/0206371 ; for general information on rice see http://kuhep4.phsx.ukans.edu/ iceman / index.html .
see http://www.fcp01.vanderbilt.edu/schedules/upload/ john_krizmanic - owl - vandy.pdf ; see also d. b. cline and f. w. stecker , owl / airwatch science white paper , astro - ph/0003459 .
a. kusenko and t. j. weiler , phys .
lett . *
88 * , 161101 ( 2002 ) [ hep - ph/0106071 ] ; j. l. feng , p. fisher , f. wilczek and t. m. yu , phys .
lett . * 88 * , 161102 ( 2002 ) [ hep - ph/0105067 ] .
v. s. beresinsky and g. t. zatsepin , phys .
b * 28 * , 423 ( 1969 ) . v. s. berezinsky and g. t. zatsepin , sov .
j. nucl . phys .
11 ( 1970 )
111 [ yad . fiz . 11 ( 1970 )
f. w. stecker , astrophys .
j. * 228 * , 919 ( 1979 ) .
r. j. protheroe and p. a. johnson , astropart .
phys . * 4 * , 253 ( 1996 ) [ astro - ph/9506119 ] .
s. yoshida , h. y. dai , c. c. jui and p. sommers , astrophys .
j. * 479 * , 547 ( 1997 ) [ astro - ph/9608186 ] .
r. engel , d. seckel and t. stanev , phys .
d * 64 * , 093010 ( 2001 ) [ astro - ph/0101216 ] .
p. bhattacharjee , c. t. hill and d. n. schramm , phys .
lett . * 69 * , 567 ( 1992 ) .
n. rubin , m. phil .
thesis , cavendish laboratory , university of cambridge , 1999 , http://www.stanford.edu/ nrubin / thesis.ps z. fodor and s. d. katz , phys .
* 86 * , 3224 ( 2001 ) [ arxiv : hep - ph/0008204 ] .
s. sarkar and r. toldra , nucl .
b * 621 * , 495 ( 2002 ) [ hep - ph/0108098 ] . c. barbot and m. drees , phys .
b * 533 * , 107 ( 2002 ) [ arxiv : hep - ph/0202072 ] ; r. aloisio , v. berezinsky and m. kachelriess , arxiv : hep - ph/0307279 ; c. barbot and m. drees , arxiv : hep - ph/0211406 .
l. dokshitzer , v. a. khoze , a. h. mueller , and s. i. troyan , _ basics of perturbative qcd _ ( editions frontiers , saclay , 1991 ) ; r. k. ellis , w. j. stirling , and b. r. webber , _ qcd and collider physics _
( cambridge univ . press , cambridge , england , 1996 ) ; v. a. khoze and w. ochs , int .
a12 ( 1997 ) 2949 .
v. berezinsky , m. kachelrie and a. vilenkin , phys .
lett . * 79 * , 4302 ( 1997 ) [ astro - ph/9708217 ] .
v. a. kuzmin and v. a. rubakov , phys .
nucl . * 61 * , 1028 ( 1998 ) [ yad .
fiz . * 61 * , 1122 ( 1998 ) ] [ astro - ph/9709187 ] .
p. blasi , r. dick and e. w. kolb , astropart .
phys . * 18 * , 57 ( 2002 ) [ astro - ph/0105232 ] .
m. ave , j. a. hinton , r. a. vazquez , a. a. watson and e. zas , phys .
* 85 * , 2244 ( 2000 ) [ arxiv : astro - ph/0007386 ] ; k. shinozaki _ et al .
_ , astrophys .
j. * 571 * ( 2002 ) l117 .
m. kachelriess and d. v. semikoz , arxiv : astro - ph/0306282 ; h. b. kim and p. tinyakov , arxiv : astro - ph/0306413 . s. l. dubovsky and p. g. tinyakov , jetp lett .
* 68 * , 107 ( 1998 ) [ arxiv : hep - ph/9802382 ] . g. sigl , s. lee , d. n. schramm and p. coppi , phys
. lett .
b * 392 * , 129 ( 1997 ) [ astro - ph/9610221 ] .
g. sigl , s. lee , p. bhattacharjee and s. yoshida , phys .
d * 59 * , 043504 ( 1999 ) [ hep - ph/9809242 ] .
farrar , phys .
* 76 * , 4111 ( 1996 ) .
[ arxive : hep - ph/9603271 ] .
d. j. chung , g. r. farrar and e. w. kolb , phys .
d * 57 * , 4606 ( 1998 ) [ arxiv : astro - ph/9707036 ] .
see also s. balberg , g. r. farrar and t. piran , astrophys . j. * 548 * , l179 ( 2001 ) [ arxiv : astro - ph/0010112 ] .
j. adams _ et al . _
[ ktev collaboration ] , phys .
lett . * 79 * , 4083 ( 1997 ) [ arxiv : hep - ex/9709028 ] ; v. fanti _ et al . _
[ na48 collaboration ] , phys .
* b446 * , 117 ( 1999 ) ; a. alavi - harati _ et al . _
[ ktev collaboration ] , phys .
* 83 * , 2128 ( 1999 ) [ arxiv : hep - ex/9903048 ] .
albuquerque _ et al . _
[ e761 collaboration ] , phys .
lett . * 78 * , 3252 ( 1997 ) [ arxiv : hep - ex/9604002 ] .
m. kachelriess , d. v. semikoz and m. a. tortola , phys .
d * 68 * , 043005 ( 2003 ) [ arxiv : hep - ph/0302161 ] .
v. berezinsky , m. kachelrie , and s. ostapchenko , phys .
d * 65 * , 083004 ( 2002 ) [ arxiv : astro - ph/0109026 ] .
t. j. weiler , phys .
lett . *
49 * , 234 ( 1982 ) .
astrophys .
j. * 285 * , 495 ( 1984 ) .
astropart .
phys . * 11 * , 303 ( 1999 ) [ hep - ph/9710431 ] .
d. fargion , b. mele and a. salis , rays , astrophys . j. * 517 * , 725 ( 1999 ) [ astro - ph/9710029 ] .
s. yoshida , g. sigl and s. j. lee , phys .
lett . * 81 * , 5505 ( 1998 ) [ hep - ph/9808324 ] .
z. fodor , s. d. katz and a. ringwald , phys .
* 88 * , 171101 ( 2002 ) [ hep - ph/0105064 ] ; hep - ph/0105336 ; hep - ph/0203198 ; a. ringwald , hep - ph/0111112 .
d. s. gorbunov , p. g. tinyakov and s. v. troitsky , astropart .
phys . * 18 * , 463 ( 2003 ) [ arxiv : astro - ph/0206385 ] .
s. hannestad , jcap * 0305 * , 004 ( 2003 ) [ arxiv : astro - ph/0303076 ] .
s. singh and c. p. ma , phys .
d * 67 * , 023506 ( 2003 ) [ arxiv : astro - ph/0208419 ] .
g. gelmini and a. kusenko , phys .
* 84 * , 1378 ( 2000 ) [ arxiv : hep - ph/9908276 ] . v. berezinsky , m. kachelriess and s. ostapchenko , phys . rev
. lett . * 89 * , 171802 ( 2002 ) [ arxiv : hep - ph/0205218 ] .
a. neronov , d. semikoz , f. aharonian and o. kalashev , astro - ph/0201410 . f. w. stecker and m. h. salamon , space sci . rev . * 75 * , 341 ( 1996 ) [ arxiv : astro - ph/9501064 ] .
a. y. neronov and d. v. semikoz , new astronomy reviews , * 47 * 693 ( 2003 ) .
a. y. neronov and d. v. semikoz , phys .
d * 66 * , 123003 ( 2002 ) [ arxiv : hep - ph/0208248 ] . | we apply new upper limits on neutrino fluxes and the diffuse extragalactic component of the gev @xmath0ray flux to various scenarios for ultra high energy cosmic rays and neutrinos . as a result
we find that extra - galactic top - down sources can not contribute significantly to the observed flux of highest energy cosmic rays .
the z - burst mechanism where ultra - high energy neutrinos produce cosmic rays via interactions with relic neutrinos is practically ruled out if cosmological limits on neutrino mass and clustering apply . |
it is well known that the disks of late - type galaxies are not completely flat .
warping of the outer gas disk into an `` integral sign '' shape is a nearly ubiquitous feature of spiral galaxies , even those that appear to be quite isolated ( garca - ruiz et al . 2002 ) .
a significant fraction of galaxies also have warped stellar disks ( snchez - saavedra et al .
1990 ; reshetnikov & combes 1998 ) . for several decades
it has been recognized that the disk and other components of the milky way exhibit an additional type of vertical structure in the form of systematic displacements from the mean plane known as `` corrugations '' ( e.g. , gum et al .
1960 ; quiroga 1974,1977 ; lockman 1977 ; sanders et al . 1984 ) .
these corrugations are present along both radial and azimuthal directions and have scales ranging from @xmath250 to 350 pc in amplitude and from one to several kiloparsecs in wavelength ( spicker & feitzinger 1986 ) . despite predictions that corrugations may be a common phenomenon in disk galaxies , systematic searches for these features
have been limited , and consequently , little has been known about their true frequency of occurrence in external galaxies . to date ,
evidence of corrugations has been reported for only a handful of cases outside the milky way : in the emission nebulae of m31 ( arp 1964 ) , in the stellar ( optical ) light of three edge - on , late - type galaxies ( florido et al .
1991,1992 ) , and in the velocity structure of the face - on spiral ngc 5427 ( alfaro et al .
2001 ) .
the origin of disk corrugations is a matter of longstanding debate given the small number of galaxies in which they have so far been detected ( see the reviews by alfaro & efremov 1996 and alfaro 2003 ) .
suggestions have included : gravitational instabilities ( e.g. , nelson 1976 ) ; tidal interactions ( edelsohn & elmegreen 1997 ) , collisions of high - velocity clouds with the disk ( franco et al .
1988 ; santilln et al .
1999 ) ; interaction of spiral waves with the gaseous disk ( alfaro et al .
2001 ) ; and the undular mode of the parker instability ( franco et al . 2002 ) .
depending on their origin , the presence or absence of corrugations can provide important clues into the role of magnetic fields on galaxy disk structure ( e.g. , spicker & feitzinger 1986 ; santilln et al .
2000 ; franco et al . 2002 ) , the formation of molecular clouds ( e.g. , nelson & matsuda 1980 ) , the degree of self - gravity of galaxy disks ( e.g. , revaz & pfenniger 2004 ) , or the timescales of recent interactions ( e.g. , edelsohn & elmegreen 1997 ) .
moreover , current evidence suggests that whatever the physical process(es ) responsible for corrugations , they are closely linked to the mechanisms responsible for the formation of the dense gas condensations and therefore with the regulation of star formation ( e.g. , nelson & matsuda 1980 ; alfaro et al . 1992 ; alfaro & efremov 1996 ; alfaro 2003 ) .
in a recent paper ( matthews & uson 2008 ; hereafter mu08 ) , we presented optical and imaging observations of the nearby , edge - on spiral galaxy ic 2233 .
there we reported the discovery of a corrugated pattern in the disk of this galaxy i.e . , we found that the layer of ic 2233 exhibits a remarkably regular pattern of positive and negative vertical displacements from the kinematically - defined midplane .
this pattern has a wavelength of @xmath3 ( 7 kpc ) and an amplitude that increases with distance from the center of the galaxy , reaching a maximum of @xmath4 ( @xmath2150 pc ) .
[ for comparison , the fwhm thickness of the layer ranges from @xmath2@xmath5 near the disk center to @xmath2@xmath6 toward the outer disk ( mu08 ) ] .
subsequently , we have explored the vertical structure of ic 2233 using a variety of additional tracers of the stellar and interstellar components of its disk .
we have also examined the relationship between the peaks of the vertical excursions with both density enhancements along the disk and with velocity perturbations in the disk rotation curve . here
we report the discovery of undular patterns in several additional tracers in ic 2233 .
in contrast , we find the vertical displacements are minimal in the old stars .
we also report the discovery of a link between the locations of the peak excursions from the midplane in with features in the rotation curve . in [ discussion ]
we discuss the implications of these findings for understanding the origin of the disk corrugations seen in ic 2233 and other galaxies .
ic 2233 is a bulgeless sd galaxy located at a distance of @xmath210 mpc .
it has a moderately low surface brightness , `` superthin '' disk structure ( goad & roberts 1981 ) and a low current star formation rate ( @xmath7 yr@xmath8 ; mu08 ) .
the transparency and structural simplicity of ic 2233 make it well - suited to a detailed vertical structure analysis .
ic 2233 also appears to be a very isolated system ( mu08 ) , implying that its disk structure should be minimally affected by external perturbations . to investigate the vertical structure of the disk of ic 2233 , mu08 extracted a series of intensity profiles perpendicular to the disk , spaced at regular intervals .
these profiles were derived from an total intensity image with a spatial resolution of @xmath9 . the position angle for the galaxy
was taken to be 172@xmath10 , and the adopted center of the galaxy , ( @xmath11)=(0,0 ) , was the kinematic center established by mu08 : @xmath12 , @xmath13 @xmath14 .
( here @xmath15 is the distance along the galaxy major axis , with values increasing toward the north , and @xmath16 is taken as the distance above the midplane , with positive values to the east . )
mu08 fitted a single gaussian to each intensity slice and plotted the displacement of the computed centroid from the midplane ( @xmath16=0 ) as a function of distance along the major axis . the resulting `` corrugation curve '' for the data
is reproduced in the upper left panel of figure [ fig : corplots ] .
the arrows on the upper left panel of figure [ fig : corplots ] indicate the extent of the stellar disk of ic 2233 , as determined from the 25 magnitude arcsec@xmath17 @xmath18-band isophote . it can be seen that near the edge of the stellar disk , the layer twists to form an `` integral sign '' warp . on the side of the disk where the layer is more extended
, the layer bends back toward the midplane at larger radii .
however , at small and intermediate projected radii , regular undulations are clearly visible across the disk of the galaxy .
furthermore , it can be seen that a portion of the galaxy near @xmath19 is displaced from the kinematically - defined center - of - mass .
one possible explanation is that this is linked with the presence of an inner bar . to insure that our derived corrugation curve is not significantly impacted by optical depth effects ,
we have compared intensity profiles extracted from channel images with 16@xmath20 resolution ( presented in mu08 ) with the corresponding ones made with more uniform weighting of the raw visibilities ( i.e. , with a robustness parameter @xmath21 = @xmath22 ) , which leads to images with @xmath23 resolution , albeit with rms noise about 40% higher .
we find that the respective profiles and centroids agree to within the errors , although the higher resolution images resolve out some of the more diffuse component .
the peak brightness temperatures are @xmath24 k for the @xmath23 images and @xmath25 k for the @xmath26 images , respectively , near the position of the kinematic center of ic 2233 and at the systemic velocity .
examination of the profiles shows that the is optically thin with @xmath27 .
this is consistent with the generally excellent agreement between the and h@xmath28 rotation curves ( see figure 13 of mu08 ) . to better constrain the nature and origin of the corrugations in ic 2233
, we have now extended our analysis of its vertical disk structure using a number of additional data sets , spanning far - ultraviolet ( fuv ) to far - infrared ( fir ) wavelengths ( table 1 ) . for each wavelength ,
we have computed a corrugation curve in a manner analogous to that used for the data .
the results are presented in electronic form in table 2 .
we have compared the positions of background radio sources and foreground stars with catalogued values from the nrao vla sky survey ( condon et al .
1998 ) and the tycho reference catalogue ( h@xmath29 g et al .
2000 ) , respectively . with the exception of the 70@xmath30 m and 160@xmath30 m images ( where no foreground or background sources useful for registration were detected )
, we find that astrometric registration between images at different wavelengths is accurate to within a small fraction of a resolution element .
therefore , we assume that astrometric uncertainties do not impact our results unless explicitly noted .
because of the low surface brightness nature of ic 2233 , we have also assumed that optical depth effects have a minimal impact on our analysis at all available wavelengths .
above we noted that the optical depth of the galaxy is extremely low . at visible wavelengths
the optical depth of ic 2233 also appears to be minimal .
mu08 estimated the @xmath18-band extinction toward the center of the galaxy to be @xmath20.7 magnitudes , but this is likely to be an upper limit , as fits to the radial surface brightness profiles of ic 2233 show that the light distribution is more strongly peaked than predicted by a simple exponential disk model , and both the wiyn images published by mu08 and the _ hubble space telescope _
images published by seth et al .
( 2005 ) reveal that there are very few optically thick dust clumps in this galaxy and no true dust lane .
because of this , optical depth effects are expected to be negligible in all of the _ spitzer _ bands .
the optical depth of ic 2233 at uv wavelengths is more uncertain .
however , better characterization of this will require detailed radiative transfer modelling ( see e.g. , matthews & wood 2001 ) and is beyond the scope of the present paper .
lccccl nuv & 1350 - 1750 & _ galex _ & 4.6 & 1 & young - to - intermediate age stars ; b - star photospheres + fuv & 1750 - 3000 & _ galex _ & 4.6 & 1 & young - to - intermediate age stars ( @xmath31 yr ) ; b - star photospheres + @xmath32 & 5631 - 7156 & wiyn & 0.6 & 2 & young - to - intermediate age stars ( up to @xmath33 yr ) + h@xmath28@xmath34 & 6534 - 6606 & wiyn & 0.7 & 2 & ionized gas associated with young ( @xmath35 yr ) , massive stars + 3.6@xmath30 m & 3.18 - 3.94@xmath30 m & _ spitzer _ ( irac ) & 1.7 & 3 & old stellar populations ; m - star photospheres + 4.5@xmath30 m & 4.02 - 5.02@xmath30 m & _ spitzer _ ( irac ) & 1.7 & 3 & old stellar populations ; m - star photospheres + 8.0@xmath30m@xmath36 & 6.44 - 9.38@xmath30 m & _ spitzer _ ( irac ) & 2.0 & 3 & pahs ( linked with spiral arms , molecular clouds ) + 24@xmath30 m & 18.0 - 32.2@xmath30 m & _ spitzer _ ( mips ) & 6 & 3 & hot dust near sites of high - mass star formation + 70@xmath30 m & 50.0 - 111.0@xmath30 m & _ spitzer _ ( mips ) & 18 & 3 & warm dust linked with star formation + 160@xmath30 m & 100.1 - 199.9@xmath30 m & _ spitzer _ ( mips ) & 40 & 3 & cool , extended dust ; infrared cirrus + & 21 cm & vla & 16 & 2 & neutral , atomic gas + figure [ fig : corplots ] shows corrugation plots derived from the datasets summarized in table 1 . alongside each curve , a representation of ic 2233 at the corresponding wavelength is shown . for reference , each of the panels shows overplotted a polynomial fit to the data points lying within the boundaries of the stellar disk ( i.e. , excluding the outer disk regions where warping dominates the vertical structure ) .
gaps in the corrugation plots near @xmath37 at some wavelengths are due to contamination from a foreground star .
figure [ fig : corplots ] illustrates that excursions from the midplane of up to @xmath38 ( @xmath1250 pc ) are present in a variety of tracers in ic 2233 . in the case of the mid - infrared ( 3.6@xmath30 m and 4.5@xmath30 m ) continuum ,
these highest - amplitude excursions occur only at the edges of the disk and can be attributed to the disk warp ; however , at all other wavelengths shown in figure [ fig : corplots ] , undulations with amplitudes of up to @xmath38 are present at low and intermediate projected radii along the disk .
none of these vertical undulations are as regular as those observed in ; nonetheless , some interesting relationships emerge between the patterns seen in various tracers . in the southern portion of the galaxy ( i.e. , along negative @xmath15 values ) , we see that tracers of young to intermediate - age stellar populations ( nuv , fuv , h@xmath28 , and @xmath32-band ) bend away from the plane with the same sign and comparable amplitude to that observed in ( @xmath39 ) . for the nuv and fuv light ,
there is some hint of a slight phase lag with respect to the curve .
this correlation is largely maintained in the northern portion of the galaxy , to roughly @xmath40 .
in contrast , further north ( @xmath41 ) , the vertical structure of these tracers diverges sharply from what is seen in ; here , excursions are observed toward positive @xmath16 , while the layer bends toward negative @xmath16 values . only beyond @xmath42 do the vertical centroids again converge to values similar to those observed in .
visual inspection of the uv , @xmath32-band , and h@xmath28 images suggests that higher frequency undulations may also be present in these bands .
these are difficult to quantitatively characterize , but may account for some of the scatter seen in these tracers .
turning toward tracers of the dust component in ic 2233 , vertical undulations are also clearly visible .
the corrugation curve derived from the 8@xmath30 m light ( believed to be dominated by pah emission ) exhibits a pattern comparable in shape to the data , but with larger dispersion .
both the 24@xmath30 m and the 70@xmath30 m corrugation curves also show close agreement with the data over much of the disk . the exception is the region between @xmath43 where the 24@xmath30 m and 70@xmath30 m emission diverge sharply from the .
in other galaxies both the 8@xmath30 m and 70@xmath30 m emission have been found to have distributions similar to the gas ( e.g. , helou et al .
2001 ; walter et al .
2007 ) , although the pah emission tends to correlate more closely with the molecular gas and cold dust , while the 70@xmath30 m tends to be most prominent in the highest column density regions .
the vertical structure of the cool dust traced by the 160@xmath30 m light shows relatively little correlation with any other wavelength .
our sensitivity is too limited to detect the outer @xmath220% of the galaxy at this wavelength , and in addition , the low resolution of these data ( @xmath44 ) makes the uncertainty in the location of the zero point comparable to the amplitude of the vertical excursions .
nonetheless , the data are clearly inconsistent with a flat , planar distribution for the cool dust . on the northern side of the galaxy
, the 160@xmath30 m emission appears qualitatively similar to the 70@xmath30 m emission , although the amplitude of the vertical excursions appears larger . on the southern side of the disk , the 160@xmath30 m emission appears to be anticorrelated with the ( and with all the other tracers ) . convolving and resampling the 70@xmath30 m image to match the resolution and pixel size of the 160@xmath30 m data
, we were unable to reproduce either of these trends from resolution effects alone .
the divergent behavior between several of the tracers on the northern side of ic 2233 complicates the description of the galaxy as a simple undulating plane .
while projection effects due to warping or to the different radial extents of various tracers may make some contribution to these differences , there is no _ a priori _ reason to expect that these effects should be more severe on the northern side of the galaxy .
therefore , the observed differences between the vertical displacements of different disk components should offer some clues on the origin of the corrugation patterns and their possible link to the regulation of star formation in the galaxy .
for this reason , we now examine some of these differences in greater detail . visual inspection of the uv , @xmath32 , and h@xmath28 images between @xmath45-@xmath46 reveals that the vertical intensity profiles in these bands are dominated by the light associated with a smattering of compact regions located several arcseconds above the plane .
the brightest of these ( near @xmath47 ) has counterparts at 24@xmath30 m and 70@xmath30 m .
interestingly , this position also corresponds to the location of the only major disparity between the and h@xmath28 rotation curves for the ic 2233 ; here , the h@xmath28 curve dips toward the systemic velocity by @xmath2 35 ( see figure 13 of mu08 ) .
this difference between the and h@xmath28 rotation curves suggests that either there is an absence of ionized gas at the tangent point along this line - of - sight , or else that the ionized material along this direction has been kinematically perturbed .
the former scenario seems more likely given that we detect significant emission near @xmath48 at the same velocity as the h@xmath28 emission ( see figure 11 of mu08 ) .
however , the second possibility can not be ruled out ( see also [ tides ] ) . in the total intensity image of ic 2233
, we see a distinct clump near @xmath49 that is visibly displaced toward negative @xmath16 values ( figure 1 ) .
tracers of hot young stars ( fuv , nuv , h@xmath28 , @xmath32-band ) all show a depressed surface brightness at the corresponding location , although we do detect counterparts to the clump at 8.0@xmath30 m , 24@xmath30 m , 70@xmath30 m , and 100@xmath30 m .
this dearth of starlight coupled with the presence of fir emission suggests that this sight - line may intersect a young star - forming region where the gas is still predominantly cold and dense .
recent studies show that pah emission tends to be a good proxy for dense molecular gas in quiescent galaxies ( e.g. , haas et al .
2002 ; paladino et al .
2008 ) , while emission at longer ir wavelengths can be a hallmark of hot young stellar objects that are still highly embedded .
the disparity between the vertical structure of different tracers near this location thus could be partly a consequence of this region being at an earlier evolutionary state compared with some of the other optically prominent star - forming regions in the galaxy .
models for the origin of structural corrugations in disk galaxies also predict corrugations in the velocity field ( see alfaro et al . 2001 and references therein ) .
the amplitude of the velocity corrugations and the relationship between velocity and structural corrugations can in principle be used to distinguish between various models for their origin .
unfortunately , with the exception of the milky way , it is difficult to measure both types of corrugations in the same galaxy . to explore whether any systematic patterns are discernible in the velocity field of ic 2233 ,
we have re - examined the observed rotation curve from mu08 .
we have parameterized the approaching and receding sides of the curve with a brandt ( 1960 ) function of the form : @xmath50 \left[\frac{1}{3 } + \frac{2}{3}\left(\frac{x}{x_{\rm max}}\right)^n\right]^{\frac{-3~}{2n}}\ ] ] where @xmath51 , @xmath5285.0 , @xmath53=@xmath54 , and @xmath55=1.09 . a comparison with this type of smooth , idealized curve can help to highlight subtle , systematic excursions from the expected circular velocity at a given radius . in figure
[ fig : velres ] we plot the residual velocity difference between the observed rotation curve and the brandt model as a function of distance along the major axis .
the dashed sections of the curve designate the portions of the disk where warping dominates the vertical structure ( see above ) .
the residual curve in figure [ fig : velres ] reveals that the observed radial velocities along the ic 2233 disk exhibit systematic positive and negative excursions from the brandt model , with amplitudes of @xmath565 .
this type of systematic `` waviness '' has been seen previously in the rotation curves of other disk galaxies , particularly those with spiral arms ( e.g. , de vaucouleurs & de vaucouleurs 1963 ; pimi 1965 ; fridman et al .
1998 ; alfaro et al 2001 ) .
however , what is particularly striking in the case of ic 2233 is that the locations of the local minima and maxima in the residual velocity curve correspond closely to the locations of the peak vertical structural excursions observed in the gas .
as noted above , a number of mechanisms have been proposed to account for the observed corrugations in disk galaxies .
we now discuss the suitability of several of these mechanisms for explaining the observed corrugations in ic 2233 .
numerical modelling by edelsohn & elmegreen ( 1997 ) demonstrated that tidal interactions with a low - mass companion are one means of producing a corrugated disk structure . in this model ,
the companion generates a spiral - shaped wave that propagates inwards from the edge of the disk , resulting in perpendicular motions . although the simulations of edelsohn & elmegreen were performed for a milky way - like galaxy and a perturber roughly 8% as massive , the results exhibit qualitative similarities to our observations of ic 2233 .
however , mu08 found that ic 2233 appears to be an extremely isolated galaxy , exhibiting no obvious signs of a recent interaction or minor merger .
this casts significant doubt upon tidal effects as being the primary driver of the vertical disk structure of ic 2233 .
in addition , since tidally induced corrugations are excited from the outer regions of the disk ( see edelsohn & elmegreen 1997 ) , this does not provide a natural explanation for the `` null '' seen at the kinematic center of ic 2233 ( see [ data ] and figure [ fig : corplots ] ) .
interestingly , at least one of the galaxies found by florido et al .
( 1991 ) to have disk corrugations ( ngc 4244 ) , also appears to be quite isolated ( dahlem et al . 2005 ) .
a search for corrugations in additional isolated systems , coupled with deep searches for low - mass companions , could provide better statistical constraints on the likely role of companions in exciting corrugated patterns .
while an intrusion from a companion galaxy seems unlikely in the case of ic 2233 , the influence of impacts from very low - mass perturbers ( @xmath57 ) , such as high - velocity clouds , can not be entirely ruled out . based on their energetics ,
high - velocity cloud impacts were the explanation favored by alfaro et al .
( 1991 ) to account for vertical displacements of the midplane of the milky way and their associated star formation activity ( see also franco et al .
moreover , ic 2233 is known to have a vertically extended component that is kinematically distinct from the main disk component ( mu08 ) , and whose presence suggests that the circulation of material between the plane and higher galactocentric latitudes is actively occurring ( e.g. , fraternali & binney 2008 ) . while it would be difficult to explain how a _
corrugation structure as regular as that seen in the disk of ic 2233 could be explained by a series of cloud impacts , such events may help to account for _ local _ vertical displacements and offsets between certain disk components .
for example , such an event offers one possible explanation for the differences in the vertical structure of various disk tracers that is observed over a portion of the northern half of ic 2233 as well as the perturbation in the h@xmath28rotation curve observed near this position ( see [ results ] ) .
the results presented in figure [ fig : velres ] seem to exclude another class of model for explaining the vertical meanderings of the ic 2233 disk .
in particular , our findings appear to be inconsistent with a scenario where corrugations result from gas being hydraulically `` lifted '' off the plane by a spiral arm as a result of interaction of the spiral density wave with a thick , magnetized , gas disk ( martos & cox 1998 ) . in this so - called `` galactic bore '' model
, the lifting effect is expected to show an even symmetry about midplane ( see alfaro et al .
2001 ) , resulting in some thickening of the plane and no net velocity excursions in an edge - on galaxy .
in contrast to the galactic bore picture , the undular mode of the parker ( 1966 ) instability in a thick , magnetized gas disk predicts a velocity pattern similar to that seen in figure [ fig : velres ] ( see franco et al .
2002 ) . however , an additional prediction of this model is that density enhancements should be present at the locations of maximum deviation from the mean plane ( e.g. , hanawa 1995 ) .
indeed , such a trend has been observed in the corrugated carina - sagittarius arm of the galaxy ( franco et al .
2002 ) . while ic 2233 does show a series of density enhancements along its disk ( mu08 ) , we find no clear correlation between the locations of these clumps and the maxima in the corrugated pattern in the disk .
because corrugations resulting from the undular mode of the parker instability occur along the azimuthal direction , line - of - sight integration effects may make it more difficult to match the location of density enhancements and vertical displacements in an edge - on galaxy .
however , for the case of ic 2233 , the parker instability model encounters an additional challenge in that magnetic fields in small spiral galaxies with rotational velocities comparable to ic 2233 are typically quite weak ( e.g. , chy .
2007 ) .
the observation that the corrugated structure of ic 2233 is more pronounced in the younger stellar components than in the old stars would seem to argue against a purely gravitational origin for the corrugations , since both gas and stars will be influenced by gravity . however , gas is in general more dynamically responsive than stars owing to its dissipative nature and its lower velocity dispersion ( e.g. , kim & ostriker 2007 ) .
in addition , gaseous distortions may be amplified by self - gravity and/or magnetic effects ( e.g , edelsohn & elmegreen 1997 ) . for these reasons ,
even if the corrugations have a purely gravitational origin , it is not entirely unexpected to find the largest amplitude corrugations in the gaseous components of the galaxy . in this picture , the youngest stars ( which still reside near their birth locations ) would be expected to exhibit a corrugated structure , while dynamical evolution and heating would slowly erode these signatures in the older stars . to determine whether an origin for the corrugations in ic 2233 via gaseous instabilities is feasible , it is valuable to compare the ages of the stellar populations participating in the corrugations with relevant dynamical timescales in the disk .
for example , using equation 6 - 85 from binney and tremaine ( 1987 ) , we can estimate the oscillation period , @xmath58 , for stars in the disk of ic 2233 . assuming a mass equal to the dynamical mass of the galaxy ( @xmath59 ) , a disk radius of 9.2 kpc , and a maximum rotational velocity @xmath60 ( mu08 ) , then at intermediate galactocentric radii ( near @xmath61=4.5 kpc ) , we find @xmath62 years .
this is comparable to the lifetime of an region , implying that the hot young stars giving rise to these ionized regions must be born near their present off - planar locations .
if the corrugations arise initially in the gas , then the presence of corrugations in the nuv and fuv light ( corresponding to stars with ages @xmath31 years ) implies that the midplane deviations need to persist for only a few oscillation periods to be visible in these bands . the mix of stars dominating the @xmath32-band light in ic 2233 is more uncertain , although the very blue color of this galaxy ( @xmath63=0.67 ; mu08 ) suggests that this band may too be weighted toward moderately young stars , with ages no more than a few oscillation periods .
as discussed by fridman & polyachenko ( 1984 ) , vertical displacements in a galactic disk may arise from two types of internal gravitational mechanisms : spiral density waves and vertical bending oscillations ( see also fridman et al .
variations of both types of mechanisms could be linked with disk corrugations , even in galaxies with negligible magnetic fields .
for example , nelson ( 1976 ) suggested that corrugations could arise from the perturbations of a spiral potential on a differentially rotating gas layer ( see also nelson & matsuda 1980 ) , while masset & tagger ( 1997 ) postulated that their origin could be the result of a non - linear coupling between a spiral density wave and two so - called `` warp waves '' . in the model of masset & tagger , a transmitted wave results in an integral sign warp , while a reflected wave produces corrugations .
several arguments suggest that ic 2233 is likely to exhibit at least rudimentary spiral structure ( see mu08 ) , but unfortunately its edge - on orientation makes it impossible to discern the details of this spiral pattern and thus to examine any possible relationship to the observed corrugations .
vertical bending oscillations in galaxy disks have been explored by various authors as a possible origin for galaxy warps ( e.g. , sellwood 1996 ; revaz & pfenniger 2004 ) , but to our knowledge , relatively little attention has been paid to these oscillations as a possible origin of disk corrugations ( although see sparke 1995 ; fridman et al .
1998 ; griv & chiueh 1998 ) . while the sophisticated numerical simulations needed to explore this phenomenon are beyond the scope of the present paper , we point to the results of some existing studies that suggest that such an exploration might be fruitful .
for example , in certain galaxy disk models , sellwood ( 1996 ) found higher order bending modes whose amplitude and wavelength are reminiscent of the corrugation pattern of ic 2233 ( e.g. , his figure 1d ) , although it remains to be explored how these results would translate to a disk containing both gas and stars .
more recently , revaz & pfenniger ( 2004 ) used multi - component disk models to show that bending instabilities can be long - lived ( i.e. , persisting through several rotations ) provided that disks have sufficient self - gravity . in their models
it is assumed that some portion of the galaxy s dark matter resides in the disk rather than in a halo .
revaz & pfenniger explored only the low - order ( @xmath64 and @xmath65 ) bending modes relevant to producing galaxy warps ; however , a search for higher - order modes , as well as an extension of their models to include the dissipational behavior of the gas would be of interest .
we have presented a multi - wavelength analysis of the vertical structure of the late - type , edge - on galaxy ic 2233 .
systematic excursions from the mean plane with amplitude @xmath0 ( @xmath2250 pc ) are observed in a variety of tracers in this galaxy , including the neutral hydrogen gas , the young stars , the regions , and dust components of various temperatures .
this corrugated pattern appears most regular in the gas , and is absent from the old stars traced by the mid - infrared light . over a significant fraction of ic 2233 ,
the vertical deviations observed in a variety of tracers show similar signs and amplitudes .
a notable exception occurs along the disk region between @xmath66-@xmath46 , thus complicating the interpretation of the corrugation pattern in terms of a simple , global pattern .
one contribution to the differences seen along this region could be variations in the filling factors for various disk tracers .
star - forming sites may also have different spectral energy distributions depending on their ages , with the youngest being dim at uv and optical wavelengths .
a third possibility is that perturbations ( such as impacts from infalling gas clouds ) may locally influence the vertical structure of the disk . despite these complexities
, we find that the locations of maximum deviation from the plane in the gas correspond to perturbations in the rotation curve with amplitude @xmath67 .
a number of different scenarios have been suggested in the past literature to account for the corrugated structure in disk galaxies , and we have evaluated their applicability to the case of ic 2233 .
models invoking tidal interactions with a companion , parker instabilities , or a galactic bore all encounter significant challenges in explaining the radial , galaxy - wide structural and velocity corrugations of ic 2233 .
models involving gravitational instabilities of the gas layer appear to be more promising .
however , sophisticated numerical modelling will be needed to evaluate such models further .
the presence of a corrugated structure in this rather low surface brightness galaxy implies that vigorous star formation is not a necessary condition for the appearance of complex vertical structure in the disks of galaxies .
indeed , the quiescent nature of ic 2233 coupled with its low star formation density is undoubtedly an important factor in allowing us to observe the correspondence between vertical structures at a variety of wavelengths despite our edge - on viewing angle .
we thank the referee for thoughtful comments that helped to improve this manuscript .
a portion of this worked was based on observations from the _ spitzer space telescope _ , operated by the jet propulsion laboratory ( jpl ) , california institute of technology , under contract 1407 with the national aeronautics and space administration ( nasa ) .
financial support for this work was provided to ldm through jpl contract 1279242 .
the national radio astronomy observatory ( nrao ) is a facility of the national science foundation , operated under cooperative agreement by associated universities , inc .
the data presented here were obtained as part of the nrao observing program am649 .
franco , j. , kim , j. , alfaro , e. , & hong , s. s. 2002 , apj , 570 , 647 franco , j. , tenorio - tagle , g. , bodenheimer , p. , r .
zyczka , m. , & mirabel , i. f. 1988 , apj , 333 , 826 fraternali , f. & binney , j. j. 208 , mnras , 386 , 935 across and are centered at @xmath68 . in the upper left panel ,
arrows denote the edges of the stellar disk defined by the 25 mag arcsec@xmath17 @xmath18-band isophote . in all other panels ,
the solid curve shows a polynomial fit to the data points interior to the arrows in the top left panel . at the distance of ic 2233 , @xmath6950 pc.,scaledwidth=70.0% ] nuv & -106.72 & 4.86 & 0.04 + & -96.52 & 2.07 & 0.05 + & -86.32 & 0.40 & 0.04 + & -76.12 & 1.49 & 0.08 + & -65.92 & 1.43 & 0.08 + & -55.72 & -0.17 & 0.06 + & -45.52 & -1.54 & 0.06 + & -35.32 & -0.67 & 0.07 + & -25.12 & -0.38 & 0.07 + & -14.92 & -1.02 & 0.06 + & -4.72 & -0.80 & 0.04 + & 5.48 & 0.11 & 0.06 + & 15.68 & 0.07 & 0.07 + & 25.88 & 1.06 & 0.03 + & 36.08 & 1.99 & 0.03 + & 46.28 & 0.60 & 0.06 + & 56.48 & 0.74 & 0.09 + & 66.68 & 1.41 & 0.08 + & 76.88 & 0.00 & 0.09 + & 87.08 & 3.21 & 0.03 + & 97.28 & 2.46 & 0.05 + & 107.48 & 1.00 & 0.09 + & 117.68 & 3.88 & 0.10 + & 127.88 & 2.41 & 0.06 + & 138.08 & 1.36 & 0.06 + & 148.28 & -1.26 & 0.08 + & 158.48 & -0.09 & 0.11 + & 168.68 & 0.27 & 0.11 + | we recently reported the discovery of a regular corrugation pattern in the disk of the isolated , edge - on spiral galaxy ic 2233 . here
we present measurements of the vertical structure of this galaxy at several additional wavelengths , ranging from the far ultraviolet to the far infrared .
we find that undular patterns with amplitude @xmath0 ( @xmath1250 pc ) are visible in a variety of tracers in ic 2233 , including the young - to - intermediate age stars , the regions , and the dust .
however , the vertical excursions become less pronounced in the older stellar populations traced by the mid - infrared light .
this suggests that the process leading to the vertical displacements may be linked with the regulation of star formation in the galaxy .
we have also identified a relationship between the locations of the density corrugations and small - amplitude ( @xmath15 ) velocity undulations in the rotation curve .
we are able to exclude several possible mechanisms for the origin of the observed corrugations , including tidal interaction from a companion , parker instabilities , or a galactic bore .
global gravitational instabilities appear to be the most likely explanation , although local perturbations may also be important . |
exact solutions of higher dimensional gravity and supergravity theories play a key role in the development of string theory . recently , a gdel - like exact solution of five - dimensional minimal supergravity having the maximum number of supersymmetries has been constructed @xcite . as its four dimensional predecessor , discovered by gdel in 1949 @xcite , this solution possesses a large number of isometries .
it can be lifted to higher dimensions and has recently been extensively studied as a background for string and m - theory , see e.g. @xcite .
the gdel - like five - dimensional solution found in @xcite is supported by an abelian gauge field .
this gauge field has an additional chern - simons interaction and produces the stress - energy tensor of a pressureless perfect fluid . since a chern - simons term can also be added in three dimensions , it is a natural question to ask whether a gdel like solution exists in three - dimensional gravity coupled to a maxwell - chern - simons field .
actually , there is a stronger motivation to look for this kind of solutions of three - dimensional gravity .
the reason is that the original four - dimensional gdel spacetime is already effectively three dimensional , see e.g. @xcite .
in fact , the metric has as direct product structure @xmath0 where @xmath1 satisfies a purely three - dimensional einstein equation .
the goal of this paper is twofold .
on the one hand we will show that the three - dimensional factor @xmath1 of the gdel spacetime and its generalizations @xcite are exact solutions of the three - dimensional einstein - maxwell - chern - simon theory described by the action , @xmath2 . \label{action}\end{aligned}\ ] ] the stress - energy tensor of the perfect fluid will be fully generated by the gauge field @xmath3 , in complete analogy with the five - dimensional results reported in @xcite . our second goal deals with gdel particles and black holes .
within the five dimensional supergravity theory , rotating black hole solutions on the gdel background have been investigated in .
it is then natural to ask whether the three - dimensional gdel spacetime @xmath1 can be generalized to include horizons .
this is indeed the case and a general solution will be displayed .
let us now briefly discuss the general structure of the stress - energy tensor of maxwell - chern - simons theory .
the original gdel geometry is a solution of the einstein equations in the presence of a pressureless fluid with energy density @xmath4 and a negative cosmological constant @xmath5 such that @xmath6 .
equivalently , it can be viewed as a homogeneous spacetime filled with a stiff fluid , that is , @xmath7 and vanishing cosmological constant . in ( 2 + 1)-spacetime dimensions
, an electromagnetic field can be the source of such a fluid .
to see this it is convenient to write the stress - energy tensor in terms of the dual field @xmath8 , @xmath9 in any region where the field @xmath8 is timelike , the electromagnetic field behaves as a stiff fluid with @xmath10 if gdel s geometry is going to be a solution of the einstein - maxwell system , then @xmath11 must be a constant .
moreover in comoving coordinates , in which @xmath12 , @xmath8 must be a constant vector pointing along the time coordinate .
one can easily see that such a solution does not exist .
in fact , the maxwell equations for this solution , @xmath13 imply in these coordinates that @xmath14}=0 $ ] which can not be achieved for gdel .
if the electromagnetic field acquires a topological mass @xmath15 , however , maxwell s equations ( [ max ] ) will be modified by the addition of the term @xmath16 . in that case , the timelike , constant , electromagnetic field is , as we will see below , a solution of the coupled einstein - maxwell - chern - simons system , and the geometry is precisely that of gdel . finally , we compute the conserved charges - mass , angular momentum and electric charge - for these solutions and derive the first laws for the three dimensional black holes , adapted to an observer at rest with respect to the electromagnetic fluid .
we then show how to adapt this first law in order to compare with the one for ads black holes in the absence of the electromagnetic fluid .
we start by reviewing the main properties , relevant to our discussion , of the four - dimensional gdel spacetimes @xcite .
these metrics have a direct product structure @xmath17 with three - dimensional factor given by @xmath18 the original solution discovered by gdel corresponds to @xmath19 .
furthermore , it was pointed out in @xcite that the property of homogeneity and the causal structure of the gdel solution also hold for @xmath20 and @xmath21 independent , provided that @xmath22 , the limiting case @xmath23 corresponding to anti - de sitter space .
the three - dimensional metric ( [ godel ] ) has 4 independent killing vectors , two obvious ones , @xmath24 and @xmath25 , and two additional ones , @xmath26 which span the algebra @xmath27 .
finally , the metric ( [ godel ] ) satisfies the three dimensional einstein equations , @xmath28 for all values of @xmath29 , and we see that @xmath20 plays the role of a negative cosmological constant . note that a solution @xmath30 to einstein s equations in @xmath31 dimensions can be lifted to a solution in @xmath32 dimensions through the addition of a flat direction @xmath33 if the additional components of the stress - energy tensor are chosen as @xmath34 and @xmath35 . for the solutions , @xmath36 and vanishes , as it should , for the original gdel solution .
our first goal is to prove that ( [ godel ] ) can be regarded as an exact solution to the equations of motion following from ( [ action ] ) . to this end
, we need to supplement ( [ godel ] ) with a suitable gauge field which will provide the stress - energy tensor ( right hand side of ( [ ee ] ) ) .
consider a spherically symmetric gauge field in the gauge @xmath37 , @xmath38 inserting this ansatz for the gauge field into the equations of motion associated to the action ( [ action ] ) , and assuming that the metric takes the form ( [ godel ] ) , one indeed finds a solution for @xmath39 and @xmath40 .
moreover , the two parameters @xmath41 entering in ( [ godel ] ) become related to the coupling constants @xmath15 and @xmath42 as @xmath43 with this parameterization , the gdel sector is determined by @xmath44 , with @xmath45 corresponding to anti - de sitter space . for future convenience , we shall write the solution in terms of a new radial coordinate @xmath46 defined by @xmath47 explicitly , the metric and gauge field are given by , @xmath48d\varphi^2 \nonumber \\ & & + \left(2r+(\alpha^2l^2 + 1){2r^2\over l^2 } \right)^{-1 } dr^2\ , , \label{godel2}\\ a & = & \sqrt{\alpha^2l^2 -1}\ , { 2r \over l}\ , d\varphi\ ,
. \label{gauge2}\end{aligned}\ ] ] from now on , we always write @xmath20 and @xmath21 in terms of @xmath15 and @xmath49 using ( [ trans ] ) .
the general solution for @xmath50 involves the addition of arbitrary constant terms along @xmath51 and @xmath52 in ( [ gauge2 ] ) . at this stage , we choose the constant in @xmath39 to be zero .
we will come back to this issue when we discuss black hole solutions below .
a constant term in @xmath40 is not allowed , however , if one requires @xmath53 to be regular everywhere .
indeed , near @xmath54 , the spacelike surfaces of ( [ godel2 ] ) are @xmath55 in polar coordinates , the radial coordinate @xmath46 in ( [ godel2 ] ) being the square root of a standard radial coordinate over @xmath55 , and thus @xmath40 must vanish at @xmath54 because the 1-form @xmath52 is not well defined there .
the gauge field ( [ gauge2 ] ) is also invariant under the isometries of ( [ godel ] ) , up to suitable gauge transformations : for each killing vector @xmath56 there exists a function @xmath57 such that @xmath58 in this sense , the killing vectors @xmath56 of ( [ godel ] ) are lifted to gauge parameters @xmath59 that leave the full gravity plus gauge field solution invariant .
the generalized gdel metric ( [ godel2 ] ) together with the gauge field ( [ gauge2 ] ) define a background for the action ( [ action ] ) with 4 linearly independent symmetries of this type .
we shall now use these symmetries in order to find new solutions describing particles and black holes ( see also @xcite ) .
we have proven in the previous section that the gdel metric can be regarded as an exact solution to action ( [ action ] ) . the associated gauge field ( [ gauge2 ] ) is however real only in the range @xmath61 .
we consider in this section the case @xmath62 and introduce particle - like objects on the background ( [ godel2 ] ) by means of spacetime identifications .
identifications in three - dimensional gravity were first introduced by deser , jackiw and thooft @xcite and the resulting objects called cosmons " . in the presence of a topologically massive electromagnetic field , cosmons living in a gdel background
may also be constructed along these lines .
take the metric ( [ godel2 ] ) and make the following identification along the killing vectors @xmath63 and @xmath64 @xmath65 where @xmath66 are real constants .
if @xmath67 this procedure will turn the spatial plane into a cone .
the cosmon lives in the tip of this cone , and its mass is related to @xmath68 and @xmath69 ( see below ) .
the time - helical structure given by @xmath69 will provide angular momentum . to analyze the resulting geometry
it is convenient to pass to a different set of coordinates , @xmath70 where the above identification amounts to @xmath71 also , the new time @xmath72 flows ahead smoothly , that is , it does not jump after encircling the particle . inserting these coordinates into ( [ godel2 ] ) , and erasing the primes , we find the new metric @xmath73 d\varphi^2 \nonumber \\ & & + \left ( ( \alpha^2l^2 + 1){2r^2\over l^2 } + 8g\nu r - { { 4gj } \over \alpha}\right)^{-1 } dr^2 .
\label{particles}\end{aligned}\ ] ] for fixed @xmath68 and @xmath74 , the new constants @xmath75 and @xmath76 are given by @xmath77 these constants will be shown to be related to the mass and angular momentum respectively .
since under ( [ nc ] ) @xmath78 scales with @xmath68 while @xmath46 with @xmath79 we see that the @xmath46-dependent part of gauge field ( [ gauge2 ] ) is invariant under ( [ nc ] ) .
however , the manifold now has a non - trivial cycle , and it is not regular at the point @xmath80 invariant under the action of the killing vector whose orbits are used for identifications .
explicitly , @xmath81 which corresponds to @xmath54 before the shift of @xmath46 in .
this means that one can now add a constant piece to @xmath40 .
the new gauge field becomes @xmath82 the constant @xmath83 will be identified below as the electric charge of the particle sitting at @xmath54 .
the metrics only admit the 2 killing vectors @xmath64 and @xmath84 .
indeed , the other candidates @xmath85 and @xmath86 do not survive as they do not commute with the killing vector along which the identifications are made @xcite .
so far we have only used the killing vectors @xmath87 and @xmath88 of ( [ godel ] ) to make identifications . besides these killing vectors , the metric ( [ godel ] ) has two other isometries defined by the vectors ( [ xi3 ] ) and ( [ xi4 ] ) , and one may consider identifications along them . we shall not explore this possibility in this paper .
distinguished places of the geometry ( [ ds1 ] ) may appear on those points where either @xmath89 or @xmath90 vanishes .
the vanishing of @xmath89 indicates that @xmath89 changes sign and hence closed timelike curves ( ctc ) appear . on the other hand ,
the vanishing of @xmath90 indicates the presence of horizons , as can readily be seen by writting ( [ particles ] ) in adm form .
the function @xmath89 in ( [ particles ] ) is an inverted parabola , and , it will have two zeros , say @xmath91 and @xmath92 whenever @xmath93 we must require this condition to be fulfilled in order to have a normal " region where @xmath84 is spacelike .
the boundary of the normal region are two spacelike surfaces , the velocity of light surfaces ( vls ) at @xmath94 and @xmath95 ( assume @xmath96 ) .
these surfaces are perfectly regular as long as @xmath97 there , which is indeed the case for the metric ( [ particles ] ) , when @xmath98 . on the other hand , it is direct to see from ( [ particles ] ) that @xmath99 since @xmath89 is positive in the normal region , there are no horizons there and @xmath90 is positive in that region .
this together with the fact that @xmath90 is the parabola of fig . [ pneg ] , means that , if any , both zeros of @xmath90 are on the same side of the normal region . the sides in which no zero of @xmath90 are present are analog to the gdel time machine , an unbounded region , free of singularities , where @xmath84 is timelike .
if @xmath100 , the roots of @xmath90 are smaller than the roots of @xmath89 . without loss of generality , we can restrict ourselves to this case because the solutions parametrized by @xmath101 are related to those with @xmath102 by the change of coordinates @xmath103 , @xmath104 .
the condition for `` would be horizons '' is @xmath105 as depicted in fig .
[ pneg ] , once one reaches the largest root @xmath106 of @xmath90 , the manifold comes to an end .
indeed , the signature of the metric changes as one passes @xmath107 .
this can be seen by putting the metric in adm form ( see below ) .
note that in this case , given @xmath108 , there is a unique @xmath109 satisfying - .
using then @xmath110 , with @xmath111 the smallest root of @xmath90 and @xmath112 , one finds near @xmath113 , @xmath114 this means that the spacetime has a naked singularity at @xmath113 , which is the analog of the one found in the spinning cosmon of @xcite .
alternatively , as proposed originally in @xcite for the case where the would be horizon is inside the time machine , one can periodically identify time @xmath115 with period @xmath116 .
this leads to having ctc s lying everywhere , including the normal region .
gdel cosmons , width=302 ]
we have shown in sec .
[ godel ] that the metric ( [ godel ] ) can be embedded as an exact solution to the equations of motion derived from ( [ action ] ) .
the necessary gauge field , given in ( [ gauge2 ] ) is , however , real only in the range @xmath118 . as we mentioned in sec .
[ godel ] , the gauge field ( [ gauge2 ] ) represents the most general static spherically symmetric solution , given the metric ( [ godel ] ) ( or , in the new radial coordinate , ( [ godel2 ] ) ) .
this means that if we want to find a real gauge field in the range @xmath119 we need to start with a different metric .
the goal of this section is to explore the other sector , @xmath120 , where black holes will be constructed .
starting from the metric ( [ godel2 ] ) and gauge field ( [ gauge2 ] ) it is easy to construct a new exact solution which will be real in the range @xmath121 . consider the following ( complex ) coordinate changes ) : @xmath122 , @xmath123 , @xmath124 , and the same for @xmath78 .
] acting on ( [ godel2 ] ) and ( [ gauge2 ] ) : @xmath125 , @xmath126 , and @xmath127 .
the new metric and gauge field read , @xmath128 d\varphi^2 \nonumber \\ & & + \left ( ( \alpha^2 l^2 + 1){2r^2\over l^2 } - 2r \right)^{-1 } dr^2 \label{godel3}\\ a & = & \sqrt{1-\alpha^2l^2}\ , { 2r \over l}d\varphi . \label{gauge3}\end{aligned}\ ] ] several comments are in order here .
first of all , the intermediate step of making some coordinates complex is only a way to find a new solution . from now on , all coordinates @xmath129 are defined real , and , in that sense , the fields ( [ godel3 ] ) and ( [ gauge3 ] ) provide a new exact solution to the action ( [ action ] ) which is real in the range @xmath130 .
second , in the original metric ( [ godel2 ] ) , the coordinate @xmath78 was constrained by the geometry to have the range @xmath131 .
this is no longer the case in the metric ( [ godel3 ] ) .
the 2-dimensional sub - manifold described by the coordinates @xmath132 does not have the geometry of @xmath55 near @xmath133 anymore ; the coordinate @xmath78 is thus not constrained to be compact , and in principle it should have the full range @xmath134 the reason that @xmath78 in ( [ godel3 ] ) is not constrained by the geometry is that the @xmath90 component of the metric ( [ godel3 ] ) changes sign as we approach @xmath54 .
this is an indication of the presence of a horizon , although this surface is not yet compact . finally , it is worth mentioning that the metrics ( [ godel3 ] ) and ( [ godel2 ] ) are real and are related by a coordinate transformation , so that all local invariants involving the metric alone have the same values .
however , as solutions to the einstein - maxwell equations , they are inequivalent . indeed , the diffeomorphism and gauge invariant quantity @xmath135 changes sign when going from ( [ godel2])-([gauge2 ] ) to ( [ godel3])-([gauge3 ] ) .
this is different from the pure anti - de sitter case where particles and black holes are obtained by identifications performed on the same background .
let us go back to ( [ godel3 ] ) and note that the function @xmath90 vanishes at @xmath136 . in order to make the @xmath137 surface a regular , finite area , horizon
we shall use the killing vector @xmath84 of ( [ godel3 ] ) to identify points along the @xmath78 coordinate . in this case
, @xmath63 has a non - compact orbit and identifications along it does not produce a conical singularity , but a cylinder " .
more generically , we may proceed in analogy with the cosmon case and identify along a combination of both @xmath84 and @xmath138 so that @xmath65 so that the resulting geometry will also carry angular momentum .
we again pass to a different set of coordinates , @xmath139 so that the new angular coordinate @xmath140 is identified in @xmath141 , and the time @xmath72 flows ahead smoothly .
the new metric reads ( after erasing the primes ) , @xmath142 as for the particles analyzed in the previous section , for given @xmath109 , we define new constants @xmath143 and
@xmath76 according to @xmath144 again , these constants will be related below to the mass and the angular momentum and without loss of generality , we can limit ourselves to the case @xmath100 . in the new coordinates ,
the electromagnetic potential takes the form @xmath145 , where @xmath146 the constant @xmath83 is arbitrary because , once again , the nontrivial topology allows the addition of an arbitrary constant in @xmath40 .
it is worth stressing that if @xmath78 was not compact , then @xmath68 and @xmath83 would be trivial constants .
it also follows that the killing vectors of ( [ godel3 ] ) have the same form as those of ( [ godel2 ] ) , but with the trigonometric functions @xmath147 and @xmath148 replaced by hyperbolic ones .
again , these vectors do not survive after the identifications .
we now proceed to analyze the metric in the same way we did in the preceding section .
again we have a condition for having a normal region , which , in this case reads @xmath149 the functions @xmath90 and @xmath89 now behave as in fig .
[ posp ] .
gdel black holes , width=340 ] note that @xmath150 and therefore horizons may only exist in the normal region of positive @xmath89 .
note , however , that for horizons to exist we must require @xmath151 if this requirement is fulfilled , we get two horizons inside the normal region , @xmath152 and @xmath113 , which coincide in the extremal case .
the whole normal region is in fact an ergoregion because @xmath153 is spacelike everywhere .
again , for given @xmath108 , one can then find a unique solution @xmath109 satisfying - .
following carter @xcite , the metric and the gauge field can be made regular at both horizons by a combined coordinate and gauge transformation .
indeed , if @xmath154 the black hole metric can be written as @xmath155 the analog of ingoing eddington - finkelstein coordinates are the angle @xmath156 and the time @xmath157 defined by @xmath158 , @xmath159 , giving the regular metric @xmath160 with @xmath161 given by , the @xmath46 dependent gauge transformation @xmath162 , where @xmath163 gives the regular potential @xmath164 whose norm @xmath165 is zero .
outgoing eddington - finkelstein coordinates are defined by @xmath166 , @xmath167 .
the metric then takes also the form with @xmath168 and @xmath156 replaced by @xmath169 and @xmath170 and the potential can be regularized by @xmath171 .
the null generators of the horizons are @xmath172 .
the associated ignorable coordinates which are constant on these null generators are then given by @xmath173 kruskal type coordinates @xmath174 are obtained by defining @xmath175 where @xmath176 in these coordinates , the metric is manifestly regular at the bifurcation surfaces , @xmath177 ^ 2\nonumber\\ & + & \frac{2k_\pm ( r - r_\mp)^2}{r_\pm - r_\mp}du^\pm dv^\pm,\end{aligned}\ ] ] with @xmath46 given implicitly by @xmath178 in kruskal coordinates , the gauge field becomes @xmath179 the potential can be regularized at @xmath180 by the transformations @xmath181\nonumber\\ & = & \frac{k_\pm\sqrt{1-\alpha^2l^2}}{2l}(r - r_\mp)(u^\pm dv^{\pm}-v^\pm du^{\pm}).\end{aligned}\ ] ] in the original coordinates , however , the parameters of these transformations explicitly involve the angle @xmath78 , @xmath182 $ ] and , as explicitly shown below , they change the electric charge . in order to avoid this
, one can add a constant piece proportional to @xmath183 , so that @xmath184 in the original coordinates , the gauge parameter is now a linear function of @xmath115 alone , @xmath185 according to the definition given below , such a transformation does not change the charges . the causal structure of the gdel black hole is displayed in the carter - penrose diagram fig . [ cp ] , where each point represents a circle .
carter - penrose diagram of gdel black hole , width=188,height=340 ]
in the case @xmath45 the gauge field vanishes and the gdel metric ( [ godel2 ] ) reduces to the three - dimensional anti - de sitter space ( to see this , do the coordinate transformations @xmath186 and @xmath187 ) .
this means that the identifications in this case yield the usual three - dimensional black holes , and conical singularities .
we have seen in previous sections that the gdel metrics ( [ godel ] ) and ( [ godel3 ] ) , as well as the corresponding quotient spaces describing particles and black holes , can be regarded as exact solutions to the action ( [ action ] ) .
we have distinguished three cases according to the values of the dimensionless quantity @xmath188 .
our purpose in this section is to write a general solution which will be valid for all values of @xmath189 .
we shall now construct the solution by looking directly at the equations of motion .
it is useful to write a general spherically symmetric static ansatz in the form @xcite @xmath190 where @xmath191 are functions of @xmath46 only .
this ansatz can also be written in the adm form " , @xmath192 this confirms that the function @xmath90 @xmath193 controls the existence of horizons .
note that for all @xmath191 , the determinant of this metric is @xmath194 for the gauge field , we use the radial gauge @xmath37 , and assume that @xmath39 and @xmath40 depend only on the radial coordinate , @xmath195 in this parametrization , einstein s equations take the remarkably simple form , @xmath196 where primes denote radial derivatives .
maxwell s equations reduce to @xmath197 before we write the solution to these equations , we make some general remarks on the structure of the stress - energy tensor associated to topologically massive electrodynamics . as we pointed out in the introduction
, we will seek for solutions with a constant electromagnetic field @xmath198 .
hence , we will only consider potentials @xmath50 which are linear in @xmath46 . in this case ,
( [ 5 ] ) are @xmath199 we now multiply the first by @xmath200 and the second by @xmath201 , then we subtract them to obtain @xmath202 in the last step we have used eq .
( [ dd ] ) .
this implies that , if @xmath203 then @xmath204 . by properly manipulating eqs .
( [ dd ] ) we see that this result is also valid if @xmath205 but @xmath206 , and therefore is it true as long as the electromagnetic field does not vanish .
now we insert this in the last equation in ( [ 3 ] ) , and obtain , @xmath207 this equation tells us that when the topological mass @xmath208 is greater ( smaller ) than the negative cosmological constant @xmath209 , the theory only supports timelike ( spacelike ) constants fields . therefore , for the generalized gdel spacetimes ( [ godel ] ) , we will need a topological mass @xmath210 . in the other region ,
the constant electromagnetic field will describe a tachyonic perfect fluid . anyway , as we will see below , it is this region in which black hole solutions are going to exist . by direct computation
one can check that equations ( [ 3])-([5 ] ) are satisfied by the field @xmath211 where @xmath212 the parameters @xmath143 , @xmath76 and @xmath83 are integration constants with a physical interpretation as they will be identified with mass , angular momentum and electric charge below .
the arbitrary constant @xmath213 on the other hand will be shown to be pure gauge .
for later convenience , it is however useful to keep it along and not restrict ourselves to a particular gauge at this stage .
this will be discussed in details sec .
[ conschar ] . in the sector @xmath214 ,
the solution is real only for @xmath143 negative .
these are the gdel particles , i.e. , the conical singularities , discussed in sec . [ particles ] .
the metric is recovered when @xmath215 and the change of variables @xmath216 , @xmath217 is performed .
for the special values @xmath218 and @xmath219 , which correspond to the trivial identification @xmath220 , @xmath221 in sec .
[ particles ] , the conical singularities disappear and we are left with the gdel universes , used for the identifications producing the cosmons . when @xmath222 , @xmath143 has to be positive .
the black hole metrics of sec .
[ bh ] are recovered when @xmath223 and @xmath224 , @xmath225 . for @xmath226 and @xmath219 , they reduce to the solution from which the black holes have been obtained from non - trivial identifications . by construction , the electromagnetic stress - energy tensor for the solutions takes the form @xmath227 where the unit tangent vector of the fluid is @xmath228 . for @xmath229 , the effect of the electromagnetic field
can be taken into account by replacing the original cosmological constant @xmath230 by the effective cosmological constant @xmath231 and introducing a pressure - free perfect , ordinary or tachyonic , fluid with energy density @xmath232 .
from this point of view , _ the chern - simons coupling transmutes into a cosmological constant_. for @xmath233 , the fluid flows along timelike curves while for @xmath234 , the fluid is tachyonic .
when @xmath45 , the fluid disappears , the stress - energy tensor vanishes and the solution is real for @xmath235 .
the metric reduces to the btz metric @xcite , as can be explicitly seen by transforming to the standard frame that is non - rotating at infinity with respect to anti - de sitter space , @xmath236 as will be explained in more details below , in the rotating frame that we have used , the energy and angular momentum are @xmath143 and @xmath76 respectively , while they become @xmath237 and @xmath76 in the standard non - rotating frame .
regular black holes have the range ( see fig . [ diag2 ] ) @xmath238 note that the solution still possess a topological charge @xmath83 .
it has been discussed in more details in @xcite .
when @xmath239 , the limit @xmath240 can be taken smoothly in the coordinates @xmath241 , @xmath242 in which the solution becomes @xmath243 where @xmath244 .
sectors of the general solution.,width=226 ] sectors of the @xmath245 solution .
the btz mass axis @xmath246 and the extremal solutions are explicitly indicated.,width=226 ]
the charge differences between a given solution @xmath247 and an infinitesimally close one @xmath248 are controlled by the linearized theory around @xmath249 .
the equivalence classes of conserved @xmath250-forms ( here 1-forms ) of the linearized theory can be shown @xcite to be in one - to - one correspondence with gauge parameters @xmath251 satisfying @xmath252 the associated on - shell closed 1-forms can be written as @xcite ( see also @xcite for the gravitational part ) : @xmath253 = k^{grav}_{\xi } + k^{em } _ { \xi,\epsilon } + k^{cs}_{\xi,\epsilon},\end{aligned}\ ] ] where @xmath254 where @xmath255 is the komar 1-form and @xmath256 the electromagnetic contribution is @xmath257 where @xmath258 the chern - simons term contributes as @xmath259 finite charge differences are computed by choosing a path @xmath260 in parameter space joining the solution @xmath261 to a background solution @xmath262 as @xmath263,\end{aligned}\ ] ] where @xmath264 is a closed 1-dimensional submanifold and @xmath265 denotes the directional derivative of the fields along @xmath260 in the space of parameters .
these charges only depend on the homology class of @xmath264 .
they also do not depend on the path , provided the integrability conditions @xcite @xmath266=0 $ ] are satisfied . for generic metrics and gauge fields of the form ( [ sol2 ] ) ,
the general solution @xmath267 of ( [ eq:2 ] ) is a linear combination of @xmath268 , @xmath269 and @xmath270 .
these basis elements are associated to infinitesimal charges as follows , @xmath271 where the contribution proportional to @xmath272 in @xmath273 and @xmath274 originate from the chern - simons term through .
the conserved charges associated with @xmath268 , @xmath269 are thus manifestly integrable .
we choose to associate the angular momentum to @xmath275 so that its value be algebraically independent of @xmath83 .
if one takes as basis element @xmath276 instead of @xmath270 , one gets a third integrable conserved charge equal to @xmath277 .
the integrated charges computed with respect to the background @xmath278 and associated to @xmath276 , @xmath279 and @xmath268 are the mass , the angular momentum and the total electric charge respectively , @xmath280 note that even though the metric and gauge fields in become singular at the background @xmath278 , we can see from the form that this is just a coordinate singularity .
the parameter @xmath213 is pure gauge because the variation @xmath281 is not present in the infinitesimal charges .
note however that @xmath213 appears explicitly in the definition of the mass by associating it with the basis element @xmath282 .
it is only in the gauge @xmath283 , that the mass is associated with the time - like killing vector @xmath270 .
this definition ensures in particular that the mass of the black hole does not depend on the gauge transformations ( [ time_gauge ] ) needed to regularize the potential on the bifurcation surfaces . in order to compare with standard ads black holes ,
one has to compute the mass in the frame instead of using the rest frame for the fluid . the conserved charge @xmath284 associated with @xmath285 is now given by @xmath286 which coincides with the conventional definition of the mass for the btz black holes .
when it exists , the outer horizon @xmath287 is located at @xmath113 , the largest positive root of @xmath288 . in the following
, a subscript @xmath289 on a function means that it is evaluated at @xmath113 .
the generator of the horizon is given by @xmath290 , where the angular velocity @xmath20 of the horizon has the value @xmath291 where @xmath292 denotes the sign of @xmath293 .
the first law can be derived by starting from @xmath294 the first term on the right - hand side is computed using standard arguments ( see for example @xcite ) to give @xmath295 where the total electric potential is given by @xmath296 the surface gravity is given by @xmath297 and the proper area by @xmath298 note that the choice of signs in the definition of electric charge and angular momentum were made so that the first laws appear in the conventional form .
we have @xmath299 so that @xmath300\label{rplus } \label{eq:3}\end{aligned}\ ] ] in order to explicitly verify the first law , we start by showing that @xmath301 .
we need to verify that @xmath302 using the explicit expressions for the components of @xmath50 , this equation reduces to @xmath303 taking into account @xmath304 together with @xmath305 , this equality can then easily be checked using @xmath306 , @xmath307 , implying @xmath308 . since @xmath309
, the first law reduces to @xmath310[\frac{2\delta r_+}{r_+}-\frac{\delta\mu}{\mu}],\end{aligned}\ ] ] which can be explicitly checked using .
in particular , the first law can be evaluated in the gauge where the potential is regular on the horizon @xmath113 .
because the two forms ( [ blackhole ] ) and ( [ sol2 ] ) of the black hole solution are related by the change of coordinates @xmath311 , @xmath312 , the gauge now corresponds to @xmath313 this amounts to the choice @xmath314 in .
it follows that @xmath315 and that the vector associated to @xmath50 is proportional to @xmath316 on the horizon .
the first law adapted to the energy @xmath317 is obtained by changing @xmath20 to @xmath318 in .
this form of the first law reduces to the standard form for 3 dimensional ads black holes ( with or without topological charge ) when @xmath319 .
finally , we note that the first law applies both to the outer event horizon of a black hole in the normal region and to the horizon at @xmath320 of a cosmon , when time is identified with real period @xmath321 .
m. b. wants to thank m. henneaux for useful discussions .
his work is supported by proyectos fondecyt 1020832 and 7020832 .
g. b. and g. c. are associated with the national fund for scientific research , belgium .
their work is supported in part by a `` ple dattraction interuniversitaire '' ( belgium ) , by iisn - belgium , convention 4.4505.86 , by proyectos fondecyt 1970151 and 7960001 ( chile ) and by the european commission program mrtn - ct-2004 - 005104 , in which these authors are associated to v.u . brussels .
the work of a.g . is supported in part by fondecyt grants 1051084,1051064 and by the research grant 26 - 05/r of universidad andrs bello . | we construct gdel - type black hole and particle solutions to einstein - maxwell theory in 2 + 1 dimensions with a negative cosmological constant and a chern - simons term . on - shell
, the electromagnetic stress - energy tensor effectively replaces the cosmological constant by minus the square of the topological mass and produces the stress - energy of a pressure - free perfect fluid .
we show how a particular solution is related to the original gdel universe and analyze the solutions from the point of view of identifications .
finally , we compute the conserved charges and work out the thermodynamics . |
being planned to start around 2020/2022 , the deep inelastic electron - nucleon scattering at the lhc ( lhec ) machine is a possible extension of the current lhc at cern , an electron - proton collider @xcite .
it is a convenient way to go beyond the lhc capabilities , exploiting the 7 tev proton beams which will be produced at the lhc , to drive research on @xmath3 and @xmath4 physics at some stage during the lhc time .
this lhc extension will open a new kinematic window - the @xmath5 cm energy can reach up to tev scale , far beyond the @xmath6 200 gev at hera , a very proficuous region for small-@xmath7 physics and many other physics studies .
in particular , the energy of the incoming proton is delivered by the lhc beam , and a list of possible scenarios is considered for the energy of the incoming electron as @xmath8 tev and @xmath9 gev , corresponding to the center of mass energies of @xmath10 tev @xcite .
the anticipated integrated luminosity is of order @xmath11 fb@xmath12 that depends on the energy of the electron beam and also the machine design . despite of great successes of the standard model ( sm ) and difficulties in finding new physics such as supersymmetry , which is the most popular scenario ,
the non - abelian self - couplings of @xmath13 , @xmath14 , and the photon remain poorly measured up to now . in this context
, the investigation of three gauge boson couplings plays an important role to manifest the non - abelian gauge symmetry in standard electroweak theory .
their precision measurement will be the crucial test of the structure of the sm .
the inclusive and exclusive production of @xmath13 and @xmath14 at the lhc already provides important tests of the sm and beyond .
however , the photoproduction channel has the advantage of being much cleaner than the @xmath15 collision channels .
the physics program of the lhec will explore the high - energy domain complementing the lhc and its discovery potential for physics beyond the sm with the great precision dis measurements at high luminosities .
the design report already contains @xcite some estimates for a variety of electroweak interaction processes such as leptoquarks / leptogluons , new heavy leptons , new physics in boson - quark interactions , and sensitivity to a higgs boson . in this work
we investigate the photoproduction of massive gauge bosons at the tev scale and also examine the potential of the lhec collider to probe anomalous ww@xmath16 coupling . along these lines , we propose some observables that are sensitive to deviations from sm physics .
previous theoretical studies on such a subject are quite compelling , and the @xmath2 vertex in @xmath3 colliders was addressed in refs .
@xcite long ago .
the aim of this work is twofold - first , we show predictions for the photoproduction of massive gauge bosons at future lhec energies within sm physics .
the photoproduction cross section including the resolved and direct processes are obtained , as well as their number of events in the most promising final states decays .
we then go beyond it , with the photoproduction of @xmath13 bosons , analyzing the production rates of @xmath13 bosons , as they move away from the sm .
the sensitivity of the lhc for deviations from the sm is investigated , and some additional observables are proposed .
this article is organized as follows .
the basic formulas to calculate the photoproduction of @xmath0 , @xmath1 and virtual photons are presented in the next section , including the expressions for the anomalous coupling in the @xmath13-production case .
our numerical results for the photoproduction cross section and event rates within the sm and beyond are presented in section [ wbeyond ] , followed by the corresponding discussion .
the summary and conclusions are presented in section [ conc ] .
let us start by considering the c and p parity conserving effective lagrangian for two charged w - boson and one photon interactions @xcite .
the motivation is to use the @xmath13 photoproduction cross section as a test of the @xmath2 vertex .
in such a case , it is introduced by two dimensionless parameters , @xmath17 and @xmath18 , which are related to the magnetic dipole and electric quadrupole moments , namely , @xmath19 and @xmath20 . in the case of values @xmath21 and @xmath22 ,
the sm is recovered at tree level .
we are left with three diagrams for the subprocess @xmath23 and only t - channel w exchange graph contributes to the @xmath2 vertex .
the unpolarized differential cross section for the subprocess @xmath24 can be obtained using helicity amplitudes from summing over the helicities . for the signal
, we are considering a quark jet and an on - shell w with leptonic decay mode @xmath25 , where @xmath26 . in the current mode , the charged lepton and the quark jet
are nicely separated and the signal is prevented from being in the background of the sm .
the cross section for the subprocess @xmath27 is composed of the direct and resolved - photon production , @xmath28 .
the direct - photon contribution is given by @xcite @xmath29 \log{\hat{z}}+ [ ( 2\kappa+{{(1-\kappa)^{2}}\over{16 } } ) { 1\over \hat{z } } \nonumber \\ & + & ( { 1\over 2 } + { { 3(1+|e_{q}|^{2})}\over{2}})\hat{z } + ( 1+\kappa)|e_{q}|-{{(1-\kappa)^{2}}\over{16 } } \nonumber \\ & + & { |e_{q}|^{2}\over 2}](1-\hat{z } ) -{{\lambda^{2}}\over{4\hat{z}^{2}}}(\hat{z}^{2}-2\hat{z } \log{\hat{z}}-1)\nonumber \\ & + & { { \lambda}\over{16\hat{z } } } ( 2\kappa+\lambda-2)[(\hat{z}-1)(\hat{z}-9 ) + 4(\hat{z}+1)\log{\hat{z}}]\ } , \nonumber \\ \label{sigmadir}\end{aligned}\ ] ] where @xmath30 , @xmath31 and @xmath32 is the cutoff scale in order to regularize the @xmath33-pole of the collinear singularity for massless quarks .
in addition , @xmath34 is the scale that determines the running of photon structure functions in the resolved part .
the quantity @xmath35 is the cabibbo - kobayashi - maskawa ( ckm ) matrix and @xmath36 is the quark charge .
the direct part of the cross section then reads @xmath37 where @xmath38 are the parton distributions functions in the proton , @xmath39 and @xmath40 .
the resolved - photon part of the cross section can be calculated using the usual electroweak formula for the @xmath41 fusion process , @xmath42 . for the photoproduction cross sections one needs parton distribution functions inside the photon and proton .
the photon structure function @xmath43 consists of perturbative pointlike parts and hadronlike parts .
putting it all together , the resolved - photon part reads @xmath44 , \nonumber \\\end{aligned}\ ] ] where in order to avoid double counting on the leading logarithmic level , one subtracts the pointlike part of the photon structure function ( photon splitting at large @xmath7 ) , @xmath45\log ( q_{\gamma}^{2}/\lambda^{2})$ ] . in addition , here @xmath46
. similar calculation can be done for the @xmath14 boson photoproduction . here , we focus on the sm prediction . once again
, the cross section for the subprocess @xmath47 is composed of the direct and resolved - photon production , @xmath28 .
the direct - photon contribution is given by @xmath48,\end{aligned}\ ] ] where now @xmath49 and @xmath50 , with @xmath51 .
the direct part of the @xmath14-photoproduction cross section then reads @xmath52 where @xmath38 are the parton distributions functions in the proton , @xmath53 . the resolved - photon part of the cross section stands for the subprocess @xmath54 , and it is written as @xmath55 .
\nonumber \\\end{aligned}\ ] ] in the next section we compute the numerical results for the @xmath0 and @xmath1 photoproduction cross section in the lhec regime of energy / luminosity .
we also investigate the sensitivity to anomalous @xmath2 couplings associated to beyond sm physics .
let us now perform a preliminary study for the lhec machine @xcite . using the design with an electron beam having laboratory energy of @xmath56 gev , the center of mass energy will reach @xmath57 tev and a nominal luminosity of order @xmath58 @xmath59s@xmath12 .
our estimates for the massive boson photoproduction cross sections in the sm are the following .
one gets @xmath60 pb and @xmath61 pb .
these are roughly estimates , since we have not introduced the @xmath62-factors associated with next - to - leading - order ( nlo ) corrections to the processes .
we have summed the resolved and direct contributions .
the energy behavior for the cross section is presented in fig .
[ fig:1 ] .
the dependence is quantitatively given by @xmath63 , with @xmath64 .
it is seen that the cross sections are at least one order of magnitude larger than for desy - hera machine , @xmath65 gev . and
@xmath1 gauge bosons as a function of the cm energy . ] in table [ tab1 ] the photon - proton total cross sections times branching ratio of @xmath66 and the corresponding number of events are shown for sm parameters for w ( @xmath21 and @xmath22 ) and also for the @xmath1 boson with a corresponding branching ratio of @xmath67 .
the number of events has been computed using @xmath68 . at this point
we consider the acceptance in the leptonic channel as 100% .
the photoproduction cross section is calculated by convoluting the weizscker - williams spectrum @xmath69,\end{aligned}\ ] ] with the differential hadronic cross section . here ,
@xmath70 and we impose a cut of @xmath71 . through the calculations , proton structure functions of cteq @xcite and photon structure functions of grv @xcite have been used with @xmath72 .
the usual electroweak parameters are taken from ref .
we have assumed an integrated luminosity @xmath73 at 10 fb@xmath12 @xcite in order to compute the number of events , @xmath74 .
the number of events is large enough to put forward further analysis , as we have units of events per second for @xmath0 .
.the photon - proton cross sections times branching ratios @xmath75 and @xmath76 in units of pb . the number of events @xmath74 is also presented at an integrated luminosity 10 fb@xmath12 .
[ tab1 ] [ cols="<,^,^",options="header " , ] as we have already noticed , the cross sections for the @xmath13 and @xmath14 production may have contributions due to higher - order terms that where not included in our calculation .
cross sections in nlo considering the default standard model vertices were already calculated in @xcite . in order to get rid of normalization uncertainties , one can take the ratios @xmath77 to test the @xmath2 vertex and the @xmath17 and @xmath18 parameters . to do this
we propose the study of the following observable : @xmath78 which can be constructed from equations given in the previous section .
such an observable was already proposed some time ago , in refs . @xcite . in fig .
[ ratiowz ] we show our results for the @xmath79 ratio , for hera and lhec energies . in the left plot , the dependence of the ratio is shown as a function of the @xmath17 parameter for fixed @xmath22 .
in addition , in the right plot the ratio is presented as a function of the @xmath18 parameter for fixed @xmath80 .
the results are presented for both desy - hera and lhec energies .
we can study the sensitivity with @xmath17 and @xmath18 parameters , regarding the sm and possible new physics .
the results show that the ratio has much more sensitivity to the @xmath17 and @xmath18 parameters for lhec energies . regarding the @xmath18 parameter
, it is unimportant at hera energies . thus , the lhec collider would be able to pin down the correct values for these parameters and then determine the magnet dipole and electric quadrupole of the @xmath13 .
another observable that could be studied and tested at the lhec is the @xmath81 asymmetry , defined by @xmath82 the results for this asymmetry are shown in fig .
[ asy ] , where we show the sensitivity with the @xmath17 and @xmath18 parameters .
as we can see , for lhec energies the @xmath81 asymmetry depends strongly on the @xmath17 and @xmath18 parameters and is therefore a useful observable to help determine the best scenarios . in the left plot
we show the dependence of asymmetry as a function of the @xmath17 parameter for fixed @xmath22 ( its sm value ) only for the lhec energy .
moreover , in the right plot we present the ratio as a function of the @xmath18 parameter for fixed @xmath80 , where the corresponding result for desy - hera energy is also shown . as a general conclusion about the anomalous coupling
, we see that with the lhec collider the parameters @xmath17 and @xmath18 have better sensitivity than desy - hera @xmath3 collider , and it would give complementary information to the lhc collider .
finally , let us compare the present calculation to previous studies . in ref .
@xcite the massive boson photoproduction is considered for an energy of @xmath83 tev and integrated luminosity of 1 fb@xmath12 .
for the sm values of @xmath17 and @xmath18 parameters , the photoproduction cross section was obtained as @xmath84 pb , @xmath85 pb and @xmath86 pb for @xmath87 , @xmath88 and @xmath1 , respectively . the number of events for @xmath1 was estimated to be 360 .
these results are completely consistent with ours when considering the integrated luminosity of 10 fb@xmath12 .
concerning the sensitivity to the parameters , we found the general trend is similar ; however , the cross sections for higher values of parameters are relatively larger than ours . we have checked that our ratio @xmath89 is 50 % smaller than in ref .
@xcite for several values of parameters @xmath90 .
this probably is due the different energy and the theoretical uncertainties coming from the pdfs considered . in ref .
@xcite an analysis quite similar to ours was performed focusing on the spectrum of the backscattered laser photon ( energy of @xmath91 tev and integrated luminosity of 200 pb@xmath12 ) .
for a weizscker - williams spectrum , we have checked that the numbers of events in process @xmath92 is quite consistent with ours when considering the same integrated luminosity .
their original values are , for instance , 288 and 1151 events for sets ( @xmath93 ) and ( @xmath94 ) , respectively .
the compatibility is good , as we are using a luminosity 100 times larger .
the photoproduction of the @xmath13 boson at hera has been addressed at the nlo level of qcd corrections in ref .
the prediction is @xmath95 pb and @xmath96 pb at @xmath97 gev , imposing a cut @xmath98 gev .
the main conclusion is that the qcd corrections reduce the factorization scale dependence significantly and modify the leading - order prediction by a factor 10% .
this can be used as a good argument for our lo calculations here .
in addition , we can rescale their prediction for the lhec case .
a rough estimate would give @xmath99 pb and @xmath100 pb at the lhec .
this is smaller than our results in table 1 , where one has @xmath101 pb , which is is associated with the cut on boson transverse momentum and distinct kinematic cuts in the integration of the weizscker - williams spectrum .
we have examined the prospects for massive gauge bosons detection at the proposed deep inelastic electron - nucleon scattering at the lhc ( lhec ) machine .
the photon - proton cross sections have been computed for @xmath0 and @xmath1 inclusive production and are of the order of dozens of picobarns .
the number of events is evaluated for the photoproduction cross section , assuming an integrated luminosity of 10 fb@xmath12 and , they are large enough to make the measurements feasible . we have also investigated the anomalous @xmath2 coupling using the machine design .
we found that the likelihood of that kinematic limit to be available at the lhec is somewhat increased relative to the previous desy - hera machine .
we have tested some sample scenarios beyond sm physics by scanning the values of parameters @xmath17 and @xmath18 considering anomalous @xmath2 coupling . in the case of anomalous coupling
, the photoproduction process at the lhec proves to be a powerful tool .
finally , we consider different observables that together could contribute to pinning down the correct @xmath2 vertex .
we introduced the ratio @xmath102 , which is less sensitive to the nlo qcd corrections and the @xmath13-asymmetry observable @xmath103 that scanns asymmetries in the @xmath13-photoproduction .
j. l. abelleira fernandez _
[ lhec study group collaboration ] , `` a large hadron electron collider at cern : report on the physics and design concepts for machine and detector , '' arxiv:1206.2913 [ physics.acc-ph ] . | in this work we investigate the photoproduction of massive gauge bosons , @xmath0 and @xmath1 , as part of relevant physics topics to be studied in the proposed electron - proton collider , the lhec .
the estimates for production cross sections and the number of events are presented .
in addition , motivated by the intensive studies to test the deviations from the standard model at present and future colliders , we discuss the @xmath0 asymmetries and perform an analysis on the role played by anomalous @xmath2 coupling . |
chiral type models such as the njl model have been successful in explaining the low energy physics of mesons and nucleons in vacuum utilizing the concept of the spontaneous chiral symmetry breaking @xcite .
application of the model to low density , however , has posed serious problems like the absence of saturation @xcite .
this is in contrast to non - chiral models like the walecka model @xcite which has been phenomenologically successful in describing the nuclear matter ground state in the mean field approximation . in chiral models with a mexican hat potential , the nuclear medium moves the minimum of the vacuum effective potential towards smaller values .
this implies a smaller curvature , i.e. the scalar sigma mass decreases .
consequently , the attraction between nucleons due to sigma meson exchange increases .
this effect may destroy the stability of nuclear matter . on the contrary , in non - chiral models
e.g. the walecka type model , where the potential is simpler , a stable nuclear matter can be found . at the nuclear matter
saturation point the fermi momentum @xmath0 and the pion mass @xmath1 are of comparable scale ( @xmath2 ) .
therefore , chiral dynamics may play an important role in a nuclear matter saturation mechanism .
chiral symmetry is an approximate symmetry for quarks which is not necessarily translated into a symmetry for nucleon degrees of freedom .
furthermore , the delta - nucleon mass splitting @xmath3 mev is also comparable to the fermi momentum @xmath0 at nuclear matter saturation .
it has been shown that the main mechanism behind the delta - nucleon mass splitting stems from the fact that the delta and the nucleon have different internal quark structure @xcite .
therefore one may naturally be led to ask if the internal structure of nucleon is important in order to describe the empirical nuclear matter ground state .
quark degrees of freedom have also been important in order to describe deep - inelastic scattering at momentum transfers of several gev and the emc effect @xcite .
moreover , at the moment a direct calculation of the many nucleons system based on qcd itself is not feasible .
therefore , effective quark theories are the best tools to develop a microscopic understanding of a single nucleon together with nuclear and quark matter and their phase structure in a unified framework . it has been shown that a non - local covariant extension of the njl model inspired by the instanton liquid model @xcite can lead to quark confinement for acceptable values of the parameters @xcite .
there are other advantages of the non - local version of the model over the local njl model , for example : the constituent quark mass is momentum - dependent , as also found in lattice calculation @xcite .
the regulator makes the theory finite to all orders in the loop expansion and leads to small next - to - leading order corrections @xcite .
this model has been phenomenologically very successful to describe mesons @xcite and baryons @xcite in vacuum .
more applications can be found in refs .
@xcite . here
, we employ this model to investigate the role of the internal dynamics of nucleon in a nuclear matter environment .
we explicitly construct nucleon degrees of freedom by binding diquarks and quarks with relativistic faddeev equations to describe nuclear matter . the relativistic faddeev approach to describe the nucleon as a composite object has been very successful @xcite . very recently
bentz and thomas @xcite have shown that local njl models may lead to saturating nuclear matter when the quark confinement effect is incorporated to avoid unphysical thresholds .
although the idea is interesting , there may be some short - coming in their approach .
quark confinement in their prescription is incorporated by introducing an infrared cutoff in the model .
this way quark confinement is permanent and there is no confinement - deconfinement transition in their model . in their hybrid approach
, the nucleon is described as a relativistic bound state in the static approximation of a scalar diquark and quark .
it has been already noticed that the attractive axial - vector diquark plays an important role in description of a single nucleon @xcite .
therefore , in order to treat the internal structure of the nucleon and its modification in the medium adequately , we will take into account both the scalar and the axial - vector diquark channels and solve the faddeev equation without invoking the static approximation .
we start by investigating the role of diquarks in the bound nucleon . the in - medium modification of the nucleon mass and the nucleon size will be calculated .
we obtain the nuclear matter equation of state when nucleon is taken as a composite object made of diquark and quark .
this paper is organised as follows : in sec .
ii we introduce the model , discuss its pionic sector and fix the parameters . in sec .
iii we set up the description of the nucleon based on a diquark - quark picture in the relativistic faddeev approach . in sec .
iv we consider nuclear matter with composite nucleons and present our numerical results .
finally , a summary and outlook is given in sec .
we consider a non - local njl model lagrangian in terms of quark degrees of freedom with @xmath4 symmetry @xmath5 where @xmath6 is the current quark mass of the @xmath7 and @xmath8 quarks .
first , we restrict the interaction terms to four - quark interaction vertices in the quark - antiquark channels
. we will add in section iii the diquark channels in order to model the quark sub - structure of the nucleon in the medium .
the current @xmath9 has the scalar ( @xmath10 ) and pseudoscalar ( @xmath11 ) components @xmath12 with @xmath13 .
the @xmath14 s are flavour @xmath15 matrices with @xmath16 .
notice that there are different approaches @xcite to introduce non - locality into the interactions .
we use the four - way separability of the non - local interaction which is also present in the instanton liquid model @xcite .
this assumption considerably simplifies the calculation .
non - locality also emerges naturally in the presence of other gluonic field configurations within the qcd vacuum @xcite and in models based on schwinger - dyson resummation techniques @xcite . in the medium the repulsive vector mesons are important .
for simplicity we add a local vector field interaction @xmath17 to the lagrangian .
locality of this term is permissible as long as the interactions in lagrangian are not fixed by some underlying theory through a fierz transformation . upon standard bosonization we obtain @xmath18\psi(x_{3})-\frac{1}{2g_{\pi}}[\sigma^{2}(x)+\pi^{2}(x)]+\frac{1}{4g_{v}}\omega_{\mu}(x)\omega^{\mu}(x).\label{e2}\\end{aligned}\ ] ] lagrangians ( [ e1 ] ) and ( [ e2 ] ) are equivalent in the classical limit .
the equations of motion for bosonic fields lead to @xmath19 the lagrangian ( [ e1 ] ) can be immediately recovered from the bosonized version eq .
( [ e2 ] ) by substituting eq .
( [ e3 ] ) into eq .
( [ e2 ] ) .
we define the fourier transform of the form factor by @xmath20 the function @xmath21 influences the momentum distribution of quarks in the non - perturbative vacuum and will be defined later .
next , we perform a mean field approximation by expanding the meson fields around their expectation values @xmath22 , @xmath23 neglecting meson fluctuations .
the mean values of the pion field and space components of the vector field vanish in vacuum and for a baryonic matter at rest due to symmetry .
the mean field lagrangian in momentum space can then be written as @xmath24 where the momentum - dependent quark mass and quark propagator are given by @xmath25 we treat the quark fields at one - loop level and the meson fields at tree level .
this approximation is consistent with the leading - order behaviour of the model in a @xmath26 expansion . within this approximation
the effective potential can be calculated : @xmath27 where @xmath28 is the degeneracy factor for quarks .
a gap equation is obtained by requiring @xmath29 : @xmath30 where the dynamical quark mass is given in eq .
( [ mass ] ) .
our model contains four parameters : the current quark mass @xmath6 , the cutoff ( @xmath31 ) hidden in the form factor @xmath21 and the coupling constants @xmath32 , @xmath33 .
the vector field coupling @xmath33 is treated as a free parameter and will be adjusted in the medium .
we first fix @xmath32 and @xmath6 for arbitrary values of @xmath34 by fitting the pion mass @xmath1 and the pion decay constant @xmath35 to their empirical values at zero baryon density . in this way
, we can consider the entire parameter space of the model .
the pion mass is given by the zero of the 1pi two - point function @xmath36 for the pion which in random phase approximation has the following form @xmath37 where @xmath38 denotes the total momentum of the quark - antiquark pair denotes a trace over flavour , colour and dirac indices . ] .
the pion decay constant @xmath35 is obtained from the coupling of the pion to the axial - vector current @xcite @xmath39f(p_{-})f(p_{+})\nonumber\\ & -&\frac{g_{\pi}g_{\pi \bar{q}q}}{2m^{2}_{\pi}}\int \frac{d^{4}k}{(2\pi)^{4}}\operatorname{tr}[s(k)]\int \frac{d^{4}p}{(2\pi)^{4}}\operatorname{tr}[\tau^{a}\gamma_{5}f(p_{-})f(p_{+})s(p_{-})\gamma_{5}\tau_{a}s(p_{+})]\nonumber\\ & & \times [ f^{2}(k)\left(f^{2}(p_{+})+f^{2}(p_{-})\right)-f(p_{+})f(p_{-})f(k)\left(f(k+q)+f(k - q)\right ) ] , \label{fpi}\\end{aligned}\ ] ] where @xmath40 is the pion - quark - antiquark coupling constant and related to the corresponding loop integral eq .
( [ lo1 ] ) by @xmath41 .
we define @xmath42 .
notice that due to non - locality the one - pion - to - vacuum matrix element gets an additional contribution .
this extra term is very important to maintain the gell - mann - oakes - renner relation @xcite and makes a significant numerical contribution . in our numerical treatment of the model ,
we evaluate loop - integrals like that in eqs .
( [ sel1 - 0],[lo1],[fpi ] ) in euclidean space ) ) a standard transcription rules from minkowski to euclidean momentum ( @xmath43 ) and reverse has been assumed .
so far this assumption has been phenomenologically correct @xcite . ] .
we choose for the form factor in eq .
( [ forfa ] ) a gaussian function in euclidean space , @xmath44 where @xmath31 is the cutoff of the theory . this choice respects poincar invariance and for certain values of the parameters it leads to the quark confinement in the sense that the dressed quark propagator has no poles at positive @xmath45 in minkowski space and consequently , quarks do not appear as asymptotic states .
confinement can occur if the following relation is satisfied between the parameters of the model @xmath46 notice that the quark propagator has many pairs of complex poles , both for confining and non - confining parameter sets which can be conceived as a remnant of the underlying confinement .
confinement in this model has direct consequence for the pion , but condition eq .
( [ n4 ] ) is not derived by using any assumption about pions .
mev and @xmath47 mev when @xmath48 .
, width=14 ] as a function of momentum @xmath49 for the three parameter sets given in table 1 .
, width=7 ] .the first two sets of poles of the quark propagator @xmath50 in minkowski space with @xmath51 .
[ tab : quarkpoles ] [ cols="<,<,<,<",options="header " , ] the emergence of complex poles for quark propagators was also noticed in schwinger - dyson equation studies in qed and qcd @xcite .
it has been recently shown that baryons become more compact in the presence of imaginary poles in the quark propagator @xcite . in another study
it has been noticed that the mass - like singularities are located on the time axis if one removes the confining potential in qed in 2 + 1d , and the mass - like singularities move from the time axis to complex momenta if there is a confining potential @xcite .
we refer to appendix for details of numerical evaluation of loop integrals as of eq .
( [ lo1 ] ) .
solutions of the gap equation ( [ sel1 - 0 ] ) are shown in fig . 1 .
in the left panel the constituent mass at zero momentum @xmath52 is shown as a function of the cutoff .
it is obvious that for very small cutoff , there is no solution for the gap equation . on the right panel of fig . 1
, we show the corresponding values of the quark condensate .
these values are within the limits extracted from qcd sum rules @xcite and also lattice calculation @xcite .
notice that the trend of the model parameters in fig . 1 is slightly different to that obtained in the local njl model with a sharp cutoff @xcite .
in contrast to the local njl model , the dynamical quark mass eq .
( [ mass ] ) is momentum dependent ( see fig
. 2 ) and follows a very similar trend to lattice simulations @xcite .
2 shows that at low virtualities the quark mass is close to the constituent mass while at large virtualities it approaches the current mass .
we analyse three sets of parameters , as indicated in table [ tab : modpar ] .
set @xmath53 is called the non - confining parameter set since it does not contain imaginary poles of the quark propagator .
sets @xmath54 and @xmath55 lead to the quark confinement in our convention with complex poles .
we define the pole mass to be given through the equation @xmath56 where pole solutions are denoted @xmath57 . equation ( [ on - shell ] )
is a non - linear equation and can be solved numerically .
the position of the quark poles are given in table [ tab : quarkpoles ] where it is seen that for the confining sets the quark poles lie in the complex plane .
in contrast to the local njl model , a non - zero current quark mass of order @xmath58 mev leads to an increase of about @xmath59 mev in the zero - momentum dynamical quark mass ( see table 1 set b ) .
it is also noted that the real part of the pole mass is bigger than the zero - momentum constituent quark mass @xmath52 for both the confining and the non - confining sets , see tables 1,2 .
in order to have a physically realistic description of matter at low density , one has to construct nucleons and nucleon matter from the quark degrees of freedom since quarks are confined at very low density . in this section ,
we take into account the quark structure of the nucleon in a diquark - quark picture .
we first build up a single nucleon out of the quark degrees of freedom by solving the relativistic faddeev equation , then ( in the next section ) we construct nuclear matter in a mean field approximation by means of these individual nucleons . in this way
, any reference to the quarks will be naturally hidden in the nucleons . in the vacuum baryons and diquarks
have already been solved in this model by one of the authors @xcite . here , we shortly recapitulate the procedure to construct the nucleon as a diquark - quark bound state , more details can be found in @xcite . in order to describe the nucleon as a bound state of a diquark and a quark
, we firstly introduce diquarks in the model .
we truncate the quark - quark interaction to the scalar @xmath60 and the axial vector @xmath61 colour antitriplet , @xmath62 where the currents are defined @xmath63\overline{\psi}^{t}(x_{3}),\nonumber\\ j_{s}(x)&=&\int d^{4}x_{2}d^{4}x_{4}f(x - x_{4})f(x_{2}-x)\psi^{t}(x_{2})\big[c^{-1}\gamma_{5}\tau_{2}\beta^{a}\big]\psi(x_{4}),\nonumber\\ \overline{j}_{a}(x)&=&\int d^{4}x_{1}d^{4}x_{3}f(x - x_{3})f(x_{1}-x)\overline{\psi}(x_{1})\big[\gamma_{\mu}c\tau_{i}\tau_{2}\beta^{a}\big]\overline{\psi}^{t}(x_{3}),\nonumber\\ j_{a}(x)&=&\int d^{4}x_{2}d^{4}x_{4}f(x - x_{4})f(x_{2}-x)\psi^{t}(x_{2})\big[c^{-1}\gamma^{\mu}\tau_{2}\tau_{i}\beta^{a}\big]\psi(x_{4}).\\end{aligned}\ ] ] the object @xmath64 is the charge conjugation matrix .
the matrices @xmath65 project onto the colour @xmath66 channel with normalisation @xmath67 .
the couplings @xmath68 and @xmath69 specify the strength in the scalar and axial - vector diquark channels , respectively .
we fix them by the empirical nucleon mass .
the form factor in the diquark sector is assumed to be the same as in the meson sector .
@xmath70-matrix in rpa approximation .
the solid lines denote the dressed quark propagators eq .
( [ pro ] ) and double lines denote diquark propagators @xmath71 .
the filled circles denote the diquark vertices @xmath72 .
down panel : the coupled effective bethe - salpeter equation for the nucleon vertex function @xmath73 is defined by amputating the external quark and diquark propagator from the wave function @xmath74.,width=16 ] in the ladder approximation the quark - quark @xmath70-matrix near the pole can be parametrised as @xcite @xmath75 where @xmath76 is the total momentum of the quark - quark pair and @xmath77 are the relative momenta within the diquarks .
the momentum - dependent scalar @xmath78 , the axial @xmath79 diquark vertex functions , their adjoint functions @xmath80 and @xmath81 and the corresponding propagators @xmath82 and @xmath83 can be obtained from solving the bethe - salpeter equation @xcite for @xmath84 scattering matrix shown in fig .
the scalar and the axial diquark masses @xmath85 and @xmath86 are obtained as the position of the pole of the corresponding @xmath70-matrix in the scalar and axial - vector diquark channels , respectively . with the diquark
, the baryon can be constructed as a bound state of diquark - quark .
we ignore the three - body irreducible graphs . due to the separability of the two - body interaction in momentum - space ,
the relativistic faddeev equation can be recast into an effective two - body bethe - salpeter type equation ( pictorially shown also in fig .
3 ) @xmath87 where the @xmath88 denotes the inverse of the full quark - diquark @xmath89-point function and contains the sum of the disconnected part and the one - quark exchange interaction kernel . after projecting the effective diquark - quark bethe - salpeter equation in the colour singlet and isospin @xmath90 channel
, one finds @xmath91 where @xmath78 and @xmath92 stand for the dirac structures of the diquark vertices introduced in eq .
( [ new - ve ] ) .
the dressed quark propagator @xmath93 is defined in eq .
( [ pro ] ) .
the numerical coefficient in kernel of the faddeev equation ( [ n9 ] ) comes from projecting the kernel to the physical baryon states ( colour singlet and isospin half ) .
we define the spectator quark momentum @xmath94 and the diquark momentum @xmath95 , where @xmath96 is the total momenta in the diquark - quark pair and @xmath97 $ ] is the mandelstam parameter which parametrises the relative momenta within the quark - diquark system @xcite . ] .
the relative momentum of quarks in the diquarks vertices are defined as @xmath98 and @xmath99 , and the momentum of the exchanged quark is @xmath100 .
we refer to appendix for details of numerical methods involved in solving the faddeev equation ( [ n9 ] ) .
the parameters of the models @xmath101 and @xmath31 are fixed by meson properties i.e. the pion mass and decay in vacuum as described in section ii .
we select parameter set c given in table 1 . now
the only unknown parameters are diquark couplings @xmath68 and @xmath69 and vector meson coupling @xmath33 . the scalar and axial - vector diquark couplings are taken as independent parameters .
one can find a line in parameter space of diquark couplings in which a reasonable description of nucleon does exist , ( see fig .
11 in ref .
@xcite ) . in other words
, the interaction is shared between the scalar and the axial - vector diquark and for small scalar diquark coupling @xmath68 one needs a dominant axial - vector diquark @xmath102 and inverse .
we fix these diquark couplings in vacuum to obtain a nucleon mass of @xmath103 mev .
we choose @xmath104 and @xmath105 ( the value of @xmath32 for set @xmath55 is given in table 1 ) and corresponds to a scalar diquark with a mass @xmath106 mev and an axial diquark with a mass @xmath107 mev in vacuum @xcite .
note also that it has been shown that the axial - vector channel is much more important for the confining than non - confining parameter sets of our model @xcite .
all the results in the next section are given by the above - mentioned parameter set .
the vector coupling @xmath33 is the only free parameter which is left for the finite density calculation .
in general , using a functional integration technique ( similar to bosonization scheme presented in section ii ) one can recast the lagrangian in terms of hadron degrees of freedom and some auxiliary fields such as diquarks .
the quark and the auxiliary fields can be integrated out and a chemical potential for the nucleons should be introduced .
having done that one can then directly apply conventional many - body techniques to the effective lagrangian . following bentz and thomas , we construct baryons out of the diquark - quark loop by solving the faddeev equations and solve baryonic matter in a mean - field approach .
since we already have ignored meson loops , we do not keep diquark loops as well .
the vacuum contribution of quark fields is kept at one - loop level .
therefore , the effective potential @xmath108 regularised by subtracting its corresponding value at zero density can be written as @xmath109 the contribution due to vacuum part @xmath110 is @xmath111 the medium part contains the nucleons and is defined @xmath112 where @xmath113 is the spin - isospin degeneracy factor for the nucleon .
the energy spectrum of a single nucleon has a simple form @xmath114 ( for @xmath115 ) @xcite where @xmath116 is the effective nucleon mass in medium obtained by solving the bound state faddeev equation ( [ fad ] ) .
the nucleon mass now is a complicated function of scalar field @xmath117 .
we define the baryon density @xmath118 by @xmath119 where the fermi distribution function @xmath120 is at zero temperature the step function @xmath121 . imposing the self - consistence condition
@xmath29 leads to the following finite - density gap equation @xmath122 this is a highly non - linear equation which is now coupled with the solutions of the faddeev equations .
therefore , in order to solve the gap equation at a given density one needs to solve the faddeev equation at the same time . in this way , the non - perturbative feature of the nucleon as a composite object is taken into account .
this equation resembles the gap equation in the vacuum eq .
( [ sel1 - 0 ] ) . in order to obtain the physical value of @xmath123
, we require that @xmath124 , for nuclear matter at rest this yields @xmath125 notice that the general form of the faddeev equation ( [ fad ] ) for the nucleon and the bethe - salpeter equation for the diquark in nuclear medium remains the same as in the vacuum .
the only new input is the in - medium modified scalar mean field eq .
( [ sel1-n ] ) which changes quark and diquark propagators appearing in eq .
( [ n9 ] ) . at low density
we do not have quark matter background , the pauli - blocking is taken into account in constructing the composite nucleon and in the nuclear matter @xcite . and the ( real , imaginary ) part of the first pole of the dressed quark propagator ( @xmath126 , @xmath127 ) are displayed as a function of density @xmath118 in the unit of the nuclear matter density @xmath128 .
we also show the effective nucleon @xmath129 and the diquark masses @xmath130 ( for @xmath131 ) at finite density . on the right
we show the fictitious scalar , axial diquark - quark threshold @xmath132 and @xmath133 , respectively as a function of density @xmath134 ( the nucleon mass @xmath129 is also shown ) . ] in fig .
4 ( left panel ) we show the effective quark mass obtained as the solution of the finite - density gap equation ( [ sel1-n ] ) , the scalar diquark mass @xmath85 and the axial - diquark mass @xmath86 at finite density .
the density dependent nucleon mass obtained from faddeev eq .
( [ fad ] ) is also shown .
notice that the nucleon mass does not follow from the addition of the diquark and constituent quark masses because of the relativistic faddeev equation .
the effective constituent quark mass in the nuclear matter decreases with a different slope compared to that in the quark matter background @xcite .
this indicates that the many - body effect felt by a constituent quark in the nucleon is different from the quark matter . in fig . 4 ( left panel ) we also show the real @xmath135 and imaginary @xmath136 part of the first pole of the dressed quark propagator as a function of nuclear matter density . the deconfinement occurs where @xmath137 , this corresponds to @xmath138 .
the real part of the first pole of the dressed quark propagator @xmath126 increases with density up to the deconfinement point and then it decreases and approaches the zero - momentum constituent quark mass @xmath52 near the nuclear matter density @xmath128 . in our model , we do not have a well - defined quark - diquark threshold .
however , one may define a fictitious quark - diquark threshold as @xmath132 ( or @xmath139 ) , where @xmath126 is the real part of the first pole of the dressed quark propagator .
it is noted from fig .4 ( right panel ) that the nucleon mass remains below the fictitious scalar diquark - quark threshold for all densities .
however , it moves slightly above the fictitious axial diquark - quark threshold near the nuclear matter density @xmath128 .
note that we show in fig .
4 only the real part of nucleon mass . above the diquark - quark threshold nucleon pole position is moved in the complex plane where the real part is taken as physical mass .
our numerical procedure ( see appendix ) is valid when nucleon solution is not far above the fictitious threshold .
it is natural to assume that at some point as we increase the density , nucleons decay to quarks and diquarks due to deconfinement . at the moment
, we do not know yet the order of a possible deconfinement transition and its critical density . based on our model approach
, one can not address such questions quantitatively .
the scalar and the axial - diquark masses in medium decrease in a very similar way .
therefore , their relative importance in the nucleon description remains almost the same at different densities .
the nucleon mass compared with the diquarks and the quarks has a bigger decreasing slope at very low density .
however , it tends to saturate near the nuclear matter density .
this effect is also observed for the diquarks , however , less pronounced .
the stabilisation of the nucleon mass @xmath129 at about nuclear matter density and the dynamical partial restoration of the chiral symmetry may closely be interconnected .
this can be realized , since after chiral restoration of the constituent quark mass , the nucleon can not feel the presence of the nuclear medium any more .
the medium is incorporated via the quark propagator eq .
( [ pro ] ) through the finite - density gap equation ( [ sel1-n ] ) into the diquark and consequently in the diquark - quark bethe - salpeter equations .
the pauli - blocking effect in the medium is not present for the quarks since in the medium we do not have a quark matter background .
we employ the quarks only to construct the nucleons and assume that they are not resolved by the medium .
the stabilisation of the nucleon mass in nuclear matter is the same as before . in the local njl model
where quark confinement is simulated by introducing an infrared cutoff the stability of the nucleon mass is reached at higher density @xcite . in the context of the derivative scalar coupling model
this phenomenon is related to dynamical screening of the effective coupling @xcite . in our model , it is derived from the non - locality of the underlying theory . with respect to relative momentum between diquark and quark .
all given components have spin @xmath140 but different orbital angular momenta @xmath141 as indicated in the plot . ] in order to find a better understanding of the internal structure of the nucleon in medium , we construct the spectator quark momentum distribution function within nucleon @xcite as a function of nuclear matter density in terms of various diquark components in the nucleon rest frame , @xmath142 where @xmath143 and @xmath144 are the faddeev vertex functions obtained from eq .
( [ fad ] ) and @xmath145 stands for the space component of relative momentum @xmath49 . @xmath146 and @xmath147 denote the diquark propagators .
although this definition is not unique , it provides some useful information about the nature of compositeness of the nucleon and their response to the medium .
note that the nucleon vertex function contains information about the compositeness of the diquark through the diquark vertex function @xmath148 , see eqs .
( [ new - ve],[n9 ] )
. in the rest frame the mass of the nucleon and its total angular momentum are good quantum numbers . in the rest frame
it is also possible to decompose the nucleon vertex function in terms of tri - spinor in different diquark channels each possessing definite orbital angular momentum and spin which allows a direct interpretation of different components @xcite .
the results are shown in fig .
it is observed that at finite density the s - wave in the scalar diquark channel is the dominant contribution to the nucleon ground state . as the nuclear matter density increases the scalar and the axial diquark masses decrease , consequently the diquark interaction couplings @xmath68 and @xmath69 grow .
this effect can be realized from fig .
5 , since the maximum strength of the density function increases as we increase the baryonic density .
it is interesting to notice that from fig .
5 one can observe that the relative importance of the scalar and axial diquarks remains intact in medium ( at least at low density ) . in order to estimate the change of the nucleon size in response to nuclear matter environment
, we compute @xmath149 at finite density .
this leads to @xmath150 for @xmath151 and @xmath152 , respectively .
therefore , as baryonic density increases the nucleon size grows .
an increase about @xmath153 is found for density @xmath154 .
6 shows the binding energy per nucleon as a function of density .
the equation of state when the internal structure of nucleon are taken into account in a presence of a vector field indeed saturates .
we adjust the vector mean field coupling @xmath33 in order to have right binding energy per particle @xmath155 mev .
this corresponds to @xmath156 when @xmath157 .
the fact that the equation of state curve can not pass exactly through the empirical saturation point implies that our hybrid approach based on a mean field approximation in the medium is still a very crude approximation in medium and it is not straightforward to fix two values via it .
a similar study by bentz and thomas in the context of the local njl model has shown that the binding energy saturates at too high density @xcite . ] .
it is well known that in the chiral models the stability of the nuclear matter depends very much on the dynamical chiral restoration @xcite .
the necessary condition for saturation of nuclear matter in any relativistic mean field theory is that nucleon attraction mediated from the @xmath158-meson exchange should decrease at high density .
therefore , if @xmath158-meson mass decreases too rapidly with density due to chiral restoration , it may work against the stabilization of the system @xcite .
however , as we have numerically proven , it is in principle possible to obtain a saturating nuclear matter equation of state , if the scalar field couples with the quarks instead of directly with the nucleons ( see also @xcite ) . to this end
, the nucleon is taken as a diquark - quark state which moves in self consistent scalar and vector fields coupling to the quarks .
( @xmath128 is nuclear matter density ) , for parameter set c , when @xmath157 . ]
in this paper we have investigated the nuclear matter in a non - local njl model .
we have worked out the equation of state of the nuclear matter in a hybrid approximation .
first , we constructed the nucleon as a bound state of a diquark and a quark in the relativistic faddeev approach and then built up nuclear matter by means of these composite nucleons . in this way
, we incorporated the internal quark structure of the nucleon in a very simple framework .
we showed that in this approximation the binding energy saturates .
this is in contrast to the local njl model , where in a similar framework , a stable normal nuclear matter does not exist unless one introduces a new extra parameter into the model either by including an @xmath159-fermi interaction term @xcite to induce the original coupling of the model density dependent or an infrared cutoff to simulate the confinement effect @xcite .
therefore , the long standing problem of matter stability in the njl model can be resolved by introducing non - locality without invoking any adhoc new parameters .
note that the form factor eq .
( [ f - f ] ) does not introduce any new parameter except the cutoff @xmath31 which is also present in the local njl model .
we have also studied the nucleon properties such as the modification of the nucleon mass and size in a nuclear matter medium .
the nucleon mass in the medium decreases very rapidly but saturates near the nuclear matter density @xmath128 .
we obtained the nucleon wave function from the faddeev equation and showed that the nucleon size significantly increases in the medium .
this implies a swollen nucleon in nuclear matter which has many attendant consequences @xcite .
our estimation of in - medium nucleon size should be taken more qualitatively since the quark density function given in eq .
( [ density - f ] ) has not a unique definition .
we found that the swelling of the nucleon is about @xmath153 at about half nucleon matter density .
one should note that in our model at nuclear matter saturation density @xmath160 the `` confinement '' mechanism is no longer at work , see fig .
4 ( left panel ) and the nucleon becomes close to the fictitious diquark - quark threshold , see fig .
4 ( right panel ) . moreover , since the nucleon mass saturates at high density , the nucleon will not expand forever .
other studies have predicted different values . for an example , it has been shown within the friedberg - lee nontopological - soliton bag model that the swelling of the nucleon is about @xmath161 at normal nuclear matter density @xcite . in the skyrmion picture nucleon
swelling about @xmath162 has been reported @xcite .
it has been also shown that in the quark - meson coupling model nucleon swelling about @xmath163 in saturated nuclear matter can explain the observed depletion of the structure function in the medium bjorken @xmath164 region @xcite .
we have also investigated the role of diquarks within the nucleon in the nuclear matter medium .
the scalar and axial diquark masses decrease with a very similar slope in the medium . despite the fact that the nucleon mass decreases ( and its size increases ) significantly
, the role of the scalar and the axial diquark in the nucleon description remains almost the same in the nuclear medium . in this paper , for simplicity we assumed that the auxially vector field to be local .
the role of the vector meson in medium is still an open question and it deserves further investigation in our model as well .
ahr acknowledges the financial support from the alexander von humboldt foundation .
ahr would like to thank m. beyer and d. blaschke for useful discussions .
there are some subtleties involved in evaluating loop integrals as of eq .
( [ lo1 ] ) in euclidean space . for simplicity
this integral is evaluated at the timelike momentum @xmath165 in euclidean space . for the confining parameter set ,
each quark propagator has a pair of complex conjugate poles .
as we increase @xmath166 , these poles in @xmath167 ( where @xmath168 ) have a chance to cross the real axis and may produce an imaginary part in the meson ( or diquark ) propagator when a threshold for decay of a meson ( or diquark ) into @xmath84- unphysical states opens - loops ( or @xmath169-loops ) does not change the structure of the quark propagator poles , however it changes their positions . in other words , if the relation eq .
( [ n4 ] ) is satisfied with the parameters of the model , poles will always remain in form of quartets @xcite .
] , see fig .
therefore , the standard wick rotation of the integration contour can not be applied .
we use the prescription proposed by cutkosky _
_ @xcite and successfully applied to such models in refs .
this amounts to a deformation of the integration contour ( as indicated in fig .
10 ) to ensure that the meson or diquark propagator does not develop an imaginary part , when the first pair of complex poles of quark propagator cross the real axis . at the pinch point , both the naive integral over euclidean four - momentum and the residue contribution diverge , although these divergences cancel to leave a finite result @xcite . plane . on the left panel ,
the pole positions at @xmath170 are shown .
the open boxes and filled circles denote the poles of the @xmath171 and @xmath172 , respectively . on the right panel
, we show the deformed integration contour beyond the pinch point . ]
the diquark - quark loops in the faddeev equation ( [ fad ] ) are computed in the rest frame of the nucleon @xmath173 .
first , we identify the singularities involved in the kernel of faddeev equation ( [ n9 ] ) with respect to the integration momentum @xmath174 .
the singularities from quark propagator will occur at : @xmath175 where @xmath176 .
the singularities from diquark propagator occurs : @xmath177 where we define @xmath178 and @xmath179 is the the lower - energy pole in the diquark @xmath70-matrix ( the mass of the lighter diquark ) .
the exchanged quark propagator is also singular at : @xmath180 with @xmath181 .
note that all masses in the above equations are density dependent .
we define @xmath182.\ ] ] it is easy to verify that in the complex @xmath174 , the poles @xmath183 and @xmath89 lie always left of @xmath184 and poles @xmath185 and @xmath186 lie always right of @xmath187 for @xmath188 . to avoid the singularities of the kernel
, we perform a wick rotation at a given density on the energy variables around @xmath184 , i. e. @xmath189 ( @xmath190 ) .
although @xmath191 changes with density , it is in principle independent of the mandelstam parameter @xmath192 .
this provides an internal check on numerical procedure . in refs .
@xcite , the parameter @xmath192 is taken @xmath90 from the outset .
the location of the quark exchanged poles @xmath193 and @xmath194 depends on the @xmath195 .
one needs to bend the integration contour in such a way that the pole @xmath193 and @xmath194 to lie right and left of the path , respectively . to this end
, we continue the momentum @xmath195 to complex plane with @xmath196 then the poles @xmath193 and @xmath194 lie always right and left of the integration path @xmath197 . above the diquark - quark threshold the solution of the effective faddeev equation should be searched in complex @xmath198-plane . in this case
the complex pole coordinates @xmath199 are interpreted as a mass , @xmath200 , and @xmath201 is related to a corresponding width @xcite .
our numerical method is valid when the location of the complex pole is close to the real axis and we concentrate only on computation of the mass . in practice
, one may increase the stability of the numerical procedure by taking advantage of @xmath192 ( in order to escape the above - mentioned poles far enough away ) .
this implies that one can find a plateau within the range @xmath202 where the results do not depend on the parameter @xmath192 @xcite .
the resulting complex coupled integral equations are solved by using the algorithm introduced by oettel _
@xcite ( see also ref.@xcite ) . 20 for a review see : s. p. klevansky , rev .
* 64 * , 649 ( 1992 ) . v. koch , t. s. biro , j. kunz and u. mosel , phys . lett . * b185 * , 1 ( 1987 ) .
j. da providencia , m. c. ruivo and c. a. de sousa , phys . rev . * d36 * , 1882 ( 1987 ) . m. buballa , nucl . phys . *
a611 * , 393 ( 1996 ) .
w. bentz and a. w. thomas , nucl . phys . *
a696 * , 138 ( 2001 ) .
b. d. serot and j. d. walecka , int
* e6*,515 ( 1997 ) ; adv .
. phys . * 16*,1 ( 1986 ) . for an example : t. schafer , e. v. shuryak , j. j. m. verbaarschot , nucl .
phys . * b412*,143 ( 1994 ) . n. ishii , w. bentz , k. yazaki , nucl .
phys . * a587*,617 ( 1995 ) .
o. nachtmann and h. j. pirner , z. phys .
* c21 * , 277 ( 1984 ) .
j. r. smith , g. a. miller , phys
. rev . * c65 * , 055206 ( 2002 ) ; phys .
lett . * 91 * , 212301 ( 2003 ) .
cloet , w. bentz and a. w. thomas , phys .
lett . * 95 * , 052302 ( 2005 ) .
d. diakonov , v. yu .
petrov , sov .
jetp * 62 * , 204 ( 1985 ) ; nucl .
* b245 * , 259 ( 1984 ) ; d. diakonov , v. yu .
petrov , p. v. pobylitsa , nucl . phys . *
b306 * , 809 ( 1988 ) ; i. v. anikin , a. e. dorokhov and l. tomio , phys .
* 31 * , 509 ( 2000 ) .
r. d. bowler and m. c. birse , nucl .
a582 * , 655 ( 1995 ) ; r. s. plant and m. c. birse , nucl .
a628 * , 607 ( 1998 ) .
j. skullerud , d. b. leinweber and a. g. williams , phys
* d64 * , 074508 ( 2001 ) .
r. s. plant and m. c. birse , nucl .
a703 * , 717 ( 2002 ) .
a. scarpettini , d. g. dumm and n. n. scoccola , phys . rev . *
d69 * , 114018 ( 2004 ) .
w. broniowski , b. golli and g. ripka , nucl . phys . *
a703 * , 667 ( 2002 ) ; b. golli , w. broniowski and g. ripka , phys . lett . * b437 * , 24 ( 1998 ) . a. h. rezaeian , n. r. walet and m. c. birse , phys . rev . *
c70 * , 065203 ( 2004 ) .
d. g. dumm and n. n. scoccola , phys . rev . *
d65 * , 074021 ( 2002 ) .
i. general , d. g. dumm and n. n. scoccola , phys .
b506 * , 267 ( 2001 ) ; d. g. dumm and n. n. scoccola , phys . rev . *
c72 * , 014909 ( 2005 ) ; r. s. duhau , a. g. grunfeld and n. n. scoccola , phys .
rev . * d70 * , 074026 ( 2004 ) .
a. buck , r. alkofer and h. reinhardt , phys .
b286 * , 29 ( 1992 ) ; s. huang and j. tjon , phys . rev . *
c49 * , 1702 ( 1994 ) ; h. asami , n. ishii , w. bentz and k. yazaki , phys . rev . * c51 * , 3388 ( 1995 ) ; c. hanhart and s. krewald , phys . lett . * b344 * , 55 ( 1995 ) .
m. oettel , g. hellstern , r. alkofer and h. reinhardt , phys .
rev . * c58 * , 2459 , ( 1998 ) .
g. hellstern , r. alkofer , m. oettel and h. reinhardt , nucl . phys . *
a627 * , 679 ( 1997 ) ; m. oettel , r. alkofer and l. von smekal , eur .
j. * a8 * , 553 ( 2000 ) ; s. ahlig , r. alkofer , c.s .
fischer , m. oettel , h. reinhardt and h. weigel , phys . rev . *
d64 * , 014004 ( 2001 ) ; m. oettel , l. von smekal , r. alkofer , comput .
. commun . *
144 * , 63 ( 2002 ) .
a. h. rezaeain , hep - ph/0507304 . c. d. roberts and a. g. williams , prog .
. phys . * 33 * , 477 ( 1994 ) ; c. d. roberts and s. m. schmidt , prog .
* 45 * , s1 ( 2000 ) .
h. ito , w. w. buck and f. gross , phys .
b248 * , 28 ( 1990 ) ; h. ito , w. buck and f. gross , phys .
rev . * c43 * , 2483 ( 1991 ) . t. schfer and e. v. shuryak , rev .
phys . * 70 * , 323 ( 1998 ) ; d. diakonov , prog .
. phys . * 36 * , 1 ( 1996 ) .
g. v. efimov and s. n. nedelko , phys . rev . *
d51 * , 176 ( 1995 ) ; y. v. burdanov , g. v. efimov , s. n. nedelko and s. a. solunin , phys . rev . * d54 * , 4483 ( 1996 ) .
d. atkinson and d. w. e. blatt , nucl .
* b151 * , 342 ( 1979 ) .
maris and h. a. holties , int .
a7 * , 5369 ( 1992 ) ; s. j. stainsby and r. t. cahill , int
phys . * a7 * , 7541 ( 1992 ) ; p. maris , phys . rev .
* d50 * , 4189 ( 1994 ) .
p. maris , phys . rev . * d52 * , 6087 ( 1995 ) . h.g .
dosch and s. narison , phys .
lett.*b417*,173 ( 1998 ) .
l. giusti , f. rapuano , m. talevi and a. vladikas , nucl . phys . *
b538 * , 249 ( 1999 ) .
m. buballa , phys . rept . * 407*,205 ( 2005 ) . m. lutz , s. klimt and w. weise , nucl . phys . *
a542 * , 521 ( 1992 ) .
j. meyer , g. papp , h. j. pirner , t. kunihiro , phys . rev . * c61 * , 035202 ( 2000 ) .
j. zimanyi and s. a. moszkowski , phys . rev . * c42 * , 1416 ( 1990 ) ; r. aguirre and a. l. de paoli , eur .
j. * a13 * , 501 ( 2002 ) . for examples : j. v. noble , phys .
lett . * 46 * , 412 ( 1981 ) ; f. e. close , r. g. roberts , g. g. ross , phys .
b129 * , 346 ( 1983 ) ; m. jandel , g .
peters , phys . rev . *
d30 * , 1117 ( 1984 ) ; r. l. jaffe , f. e. close , r. g. roberts , g. g. ross , phys . lett . *
b134 * , 449 ( 1984 ) ; l. s. celenza , a. rosenthal and c. m. shakin , phys . rev . *
c31 * , 232 ( 1985 ) ; g. chanfray , h. j. pirner , phys . rev .
* c35 * , 760 ( 1987 ) . m. jandel , g. peters , phys . rev .
* d30 * , 1117 ( 1984 ) .
m. rho , phys .
* 54 * , 767 ( 1985 ) .
x. jin , b. k. jennings , phys . rev . *
c54 * , 1427 ( 1996 ) ; phys .
rev.*c55 * , 1567 ( 1997 ) .
r. e. cutkosky , p. v. landshoff , d. i. olive and j. c. polkinghorne , nucl . phys . *
b12 * , 281 ( 1969 ) .
m. s. bhagwat , m. a. pichowsky and p.c .
tandy , phys . rev . *
d67 * , 054019 ( 2003 ) .
s. pepin , m. c. birse , j. a. mcgovern and n. r. walet , phys . rev . * c61 * , 055209 ( 2000 ) .
a. h. rezaeian , nucl - th/0512027 .
i. c. cloet , w. bentz and a. w. thomas , phys .
lett . * 95 * , 052302 ( 2005 ) .
v. bernard , a. h. blin , b. hiller , y. p. ivanov , a. a. osipov and u. g. meissner , phys . lett . * b409 * , 483 ( 1997 ) . | we study the stability of the nuclear matter in a non - local nambu - jona - lasinio model .
we work out the equation of state in a relativistic faddeev approach where we take into account the internal structure of nucleon .
we show that the binding energy saturates when the nucleon as a composite particle made of quarks is incorporated .
after truncation of the two - body channels to the scalar and axial - vector diquarks , a relativistic faddeev equation for nucleon bound states in medium is solved in the covariant diquark - quark picture .
we investigate the nucleon properties in the nuclear medium such as the role of diquarks within the nucleon and the in - medium modification of the nucleon mass and size . |
conjugated organic polymers have been a focus of intensive research over the last twentyfive years after the discovery of their remarkable conducting properties tuned by the dopant concentration @xcite .
trans_-polyacetylene sustains neutral solitons in the dimerized chain as a consequence of the peierls instability @xcite .
charge injection in polymers induces a lattice distortion with the associated formation of localized excitations , polarons and/or charged solitons . while the latter can exist only in _
trans_-polyacetylene whose ground state is twofold degenerate , the former solutions are more general since they do not require such degeneracy .
the su - schrieffer - heeger ( ssh ) model hamiltonian @xcite is the fundamental tool in the analysis of quasi one dimensional polymers . in the weak coupling regime
a continuum version @xcite of the ssh model has been developed and polaronic solutions have been obtained analytically @xcite .
although the 1d properties of the ssh model have been mainly investigated so far @xcite , extensions to two dimensions were considered in the late eighties in connection with the fermi surface nesting effect on quasi 2d high @xmath2 superconductors @xcite .
recent numerical analysis @xcite have revealed the rich physics of 2d - ssh polarons whose mass seems to be larger than in the 1d case , at least in the intermediate regime of the adiabatic parameter @xcite .
one of the main issues in the study of _ e - ph _ models regards the possibility to monitor the conditions for polaron formation as a function of the coupling strength .
the feynman path integral method @xcite is a landmark in this respect and several studies @xcite have applied the path integral formalism both to the acoustical polaron of the frhlich hamiltonian and to the holstein model @xcite . as a main property of the ssh model , the _ e - ph _
interaction modifies the bare electron hopping integral of the tight binding approximation thus leading to an hamiltonian linearly dependent on the phonon displacement field .
this feature will allow us to attack the ssh model by path integrals techniques after introducing a generalized version of the semiclassical model which treats only the electrons quantum mechanically .
the general formalism is described in section ii together with the method to compute the full partition function of the interacting system . in section iii , we focus on the peculiarities of the free energy and heat capacity as a function of the model parameters .
section iv contains the final remarks .
the 1d ssh interacting hamiltonian reads : @xmath3 \label{1}\end{aligned}\ ] ] where @xmath4 is the nearest neighbors hopping integral for an undistorted chain , @xmath5 is the electron - phonon coupling , @xmath6 is the dimerization coordinate which specifies the displacement of the monomer group on the @xmath7 lattice site along the molecular axis , @xmath8 and @xmath9 create and destroy electrons ( i.e. , @xmath10 band electrons in polyacetylene ) on the @xmath7 group .
our non - interacting hamiltonian is given by a set of independent oscillators which will be treated classically in the following .
the real space hamiltonian in eq.(1 ) can be transformed in a time dependent hamiltonian by introducing @xmath11 and @xmath12 as the electron coordinates at the @xmath13 and @xmath14 lattice sites , respectively .
the spatial _ e - ph _ correlations contained in eq.(1 ) are then mapped onto the time axis by changing : @xmath15 and @xmath16 . accordingly , the time dependent hamiltonian is : @xmath17 \label{2}\end{aligned}\ ] ] the twofold degenerate ground state of the ssh hamiltonian undergoes a peierls instability which opens a gap at the fermi energy . in real space
the soliton connects the two degenerate phases with different senses of dimerizations and a localized electronic state is associated with each soliton .
both electron hopping between solitons and electron hopping to band states ( thermal excitation ) are in principle possible within the model .
mapping the hamiltonian onto the time scale we maintain these properties while thermally activated electron hops become time dependent .
@xmath18 is therefore more general than the real space ssh hamiltonian since hopping processes are not constrained to first neighbors sites along the chain .
we emphasize that a variable range hopping introduces some local disorder in the system .
eq.(2 ) displays the semiclassical nature of the model in which quantum mechanical degrees of freedom interact with the classical variables @xmath19 . setting @xmath20 , @xmath21 and averaging the electron operators over the ground state we obtain the time dependent semiclassical energy which is linear in the displacements : @xmath22 + g[-x(\tau ) , -\tau ] \bigr)\ , \nonumber \\
g[x(\tau ) , \tau]= & & \,{l \over { \beta
\hbar}}\int_{-\pi / a}^{\pi / a } { { dk}\over { 2\pi } } exp[ik x(\tau)]\sum_n { { exp(-i\nu_n \tau ) } \over { i\nu_n - \epsilon_k /\hbar } } \ , \nonumber \\ \label{3}\end{aligned}\ ] ] where , @xmath23 is the chain length , @xmath24 is the lattice constant , @xmath25 is the inverse temperature , @xmath26 are the fermionic matsubara frequencies and @xmath27 is the electron dispersion relation . the chemical potential has been pinned to the zero energy level . after summing over the frequencies in eq.(3 ) we observe that the average energy per site can be written as @xmath28 \cosh(\epsilon_k \tau /\hbar )
n_f(\epsilon_k ) \ , \nonumber \\ & & j(\tau)= -\alpha { a \over \pi } \int_{0}^{\pi / a}{dk } \cos[k x(\tau ) ] \cosh(\epsilon_k \tau /\hbar ) n_f(\epsilon_k ) \ , \nonumber
\\ \label{4}\end{aligned}\ ] ] with @xmath29 being the number of lattice sites and @xmath30 is the fermi function .
@xmath31 is an effective term accounting for the @xmath32 dependent electronic hopping while @xmath33 is the external source @xcite current for the oscillator field @xmath19 .
averaging the electrons over the ground state we neglect the fermion - fermion correlations @xcite which lead to effective polaron - polaron interactions in non perturbative analysis of the model @xcite .
this approximation however is not expected to affect substantially the following calculations . taking a large number of oscillators ( @xmath34 ) as the _ bath _ for the quantum mechanical particle
whose coordinate is @xmath11 , one can write the general path integral at any temperature as : @xmath35 \cdot \ , \nonumber \\ & & exp\biggl[-{1 \over \hbar } \int_0^{\hbar \beta } d\tau \biggl({m \over 2 } \dot{x}^2(\tau ) + v\bigl(x(\tau)\bigr ) - \sum_i\gamma_iu_i(\tau)j(\tau)\biggr ) \biggr ] \
, \nonumber \\ \label{5}\end{aligned}\ ] ] where , @xmath36 is the electron mass , @xmath37 is the mass of the @xmath38 ionic group , @xmath39 is the oscillator frequency , @xmath40 is the coupling constant between external source term @xmath33 and @xmath38 oscillator @xmath41 . in our model , @xmath42 and @xmath43 hence , integrating out the oscillators coordinates over the paths @xmath44 , imposing a closure condition on the paths and replacing @xmath45 , we get @xmath46 , \ , \nonumber \\ & & a(x(\tau))=\ , -{{\hbar^2 } \over { 4m}}\sum_{i=1}^{\bar n } { 1 \over { \hbar \omega_i \sinh(\hbar\omega_i\beta/2)}}\cdot \ , \nonumber \\ & & \int_0^{\beta } d\tau j(\tau ) \int_0^{\beta}d{\tau '' } \cosh\bigl(\omega_i \bigl ( |\tau - { \tau''}| - \beta/2 \bigr ) \bigr ) j({\tau '' } ) , \ , \nonumber \\ & & z_i=\ , { 1 \over { 2\sinh(\hbar\omega_i\beta/2 ) } } \label{6}\end{aligned}\ ] ]
the quadratic source term @xmath47 is the non local ( in time ) action for the particle having memory of the interactions with the oscillators field . assuming periodic conditions @xmath48 , the particle paths can be expanded in fourier components @xmath49 and , taking the following measure of integration @xmath50 \label{8}\ ] ] a direct integration of eq.(6 )
can be carried out in order to derive the full partition function of the system versus temperature .
let s point out that , by mapping the electronic hopping motion onto the time scale , a continuum version of the interacting hamiltonian ( eq.(2 ) ) has been de facto introduced . unlike previous @xcite approaches however , our path integral method is not constrained to the weak _ e - ph _ coupling regime and it can be applied to any range of physical parameters .
as a preliminar step we determine , for a given path and at a given temperature : i ) the minimum number ( @xmath51 ) of @xmath52points in the brillouin zone to accurately estimate the average interacting energy per lattice site and , ii ) the minimum number ( @xmath53 ) of points in the double time integration to get a numerically stable source action in eq.(6 ) .
we find : @xmath54 and @xmath55 at @xmath56 .
computation of eqs.(6)-(8 ) requires fixing two sets of input parameters .
the first set contains the physical quantities characterizing the system : the bare hopping integral @xmath4 , the oscillator frequencies @xmath39 and the effective coupling @xmath57 ( in units @xmath58 ) .
the second set defines the paths for the particle motion which mainly contribute to the partition function through : the number of pairs ( @xmath59 ) in the fourier expansion of eq.(7 ) , the cutoff ( @xmath60 ) on the integration range of the expansion coefficients in eq.(8 ) and the related number of points ( @xmath61 ) in the measure of integration which ensures numerical convergence . after introducing a dimensionless path @xmath62 with : @xmath63 and @xmath64 , the functional measure of eq.(8 )
can be rewritten for computational purposes as : @xmath65 we take the lattice constant @xmath66 . as a criterion to set the cutoff @xmath60 on the integration range , we notice that the functional measure normalizes the kinetic term in eq.(6 ) : @xmath67\equiv 1 , \label{11}\ ] ] and this condition holds for any number of pairs @xmath68 truncating the fourier expansion in eq.(9 ) .
then , taking @xmath69 and using eq.(10 ) , the left hand side transforms as : @xmath70 \simeq { 4 \over \pi } \bigl[\int_{0}^u dy exp(-y^2)\bigr]^2 \ , \nonumber \\ & & u \equiv \sqrt{2 \pi^3}{\lambda \over \lambda } \ , \nonumber \\ & & \lambda=\ , \sqrt{{2 \pi \hbar^2 } \over { m k_bt}}. \label{12}\end{aligned}\ ] ] using the series representation @xcite @xmath71 and taking @xmath72 ( the series converges with @xmath73 ) eq.(13 ) yields 0.886207 which well approximates the poisson integral value @xmath74 .
thus , the cutoff @xmath60 can be expressed in terms of the thermal wavelength @xmath75 as @xmath76 hence , it scales versus temperature as @xmath77 .
this means physically that , at low temperatures , @xmath60 is large since many paths are required to yield the correct normalization . for example , at @xmath0 , we get @xmath78 . numerical investigation of eq.(6 ) shows however that a much shorter cutoff suffices to guarantee convergence in the path integral , while the cutoff temperature dependence implied by eq.(12 ) holds also in the computation of the interacting partition function .
the thermodynamical results hereafter presented have been obtained by taking @xmath79 .
summing over @xmath80 points for each integration range and taking @xmath81 , we are then evaluating the contribution of @xmath82 paths ( the integer part of @xmath61 is obviously selected at any temperature ) .
thus , at @xmath0 we are considering @xmath83 different paths for the particle motion while , at @xmath84 the number of paths drops to 243 .
low temperature calculations prove therefore to be extremely time consuming .
note that larger @xmath68 in the path fourier expansion would further increase the computing time without introducing any substantial improvement in the thermodynamical output of our calculation .
although the history of the ssh model is mainly related to wide band polymers , we take here a narrow band system ( @xmath85 ) to be consistent with previous investigations @xcite and with the caveat that electron - electron correlations may become relevant in narrow bands .
free energy and heat capacity have been first ( figures 1 - 3 ) computed up to room temperature assuming a bath of @xmath86 low energy oscillators separated by @xmath87 : @xmath88 .
the lowest energy oscillator yields the largest contribution to the phonon partition function mainly in the low temperature regime while the @xmath89 oscillator essentially sets the phonon energy scale which determines the size of the _ e - ph _ coupling .
a larger number @xmath90 of oscillators in the aforegiven range would not significantly modify the calculation . in the discrete ssh model , the value @xmath91 , marks the crossover between weak and strong _ e - ph _ coupling , with @xmath92 being the effective spring constant . in our continuum and semiclassical model
the effective coupling is the above defined @xmath93 .
although in principle , discrete and continuum models may feature non coincident crossover parameters , we assume that the relation between @xmath5 and @xmath4 obtained by the discrete model crossover condition still holds in our model .
hence , at the crossover we get : @xmath94 .
this means that , in figures 1 - 3 , the crossover value is set at @xmath95 . in fig.1 ,
the phonon free energy ( @xmath96 ) is plotted versus temperature together with the free energy due to the total action in eq.(6 ) .
@xmath97 results from the competition between the free path action ( kinetic term plus hopping potential in the exponential integrand ) and the source action depending on the _ e - ph _ coupling .
while the former enhances the free energy the latter becomes dominant at increasing temperatures thus reducing the total free energy . as a main feature one notes that , by _ increasing _
@xmath93 , @xmath97 gets a negative derivative at _ decreasing _ temperatures . in the very weak coupling case ( @xmath98 ) @xmath97 never intersects @xmath96 whereas an intersection point shows up both at moderately weak ( @xmath99 ) and at moderately strong ( @xmath100 ) couplings .
the intersection temperature decreases as expected by increasing the strength of @xmath93 but , at low @xmath101 ( @xmath102 ) , the contribution of the free electron path action prevails : this feature is reflected on the low @xmath101 behavior of the heat capacity linear coefficient hereafter discussed .
fig.2 shows the source term contributions to the heat capacity : the previously described summation over a large number of paths turns out to be essential to recover the correct thermodynamical behavior in the zero temperature limit . the small phonon contribution to the heat capacity
is plotted in fig.3 to point out that the dulong petit value is achieved at @xmath103 .
looking at the _ total heat capacity over temperature _ ratio , we find a peculiar low temperature upturn ( also in the weak @xmath93 regime ) which can be mainly ascribed to the sizable effective hopping integral term @xmath31 .
the _ e - ph _ coupling however determines the shape of the low @xmath101 anomaly .
the oscillators bath affects the electronic correlations both on the space and time scales .
analysis of the thermal correlation function shows infact that high energy phonons substantially reduce the electronic correlation length at low temperatures while , in presence of low energy phonons , the electronic paths are correlated over the @xmath32 scale . by increasing @xmath101 ,
the role of the phonons on the correlation function is less pronounced .
the effect of the oscillators bath on the thermodynamical properties is discussed in figures 4 - 6 where the ten phonon energies are : @xmath104 . accordingly
the crossover is set at @xmath105 and three plots out of five lie in the strong _ e - ph _ coupling regime .
as shown in fig.4 , much larger @xmath93 values ( with respect to fig.1 ) are required to get strongly decreasing free energies versus temperature while the @xmath100 curve now hardly intersects the phonon free energy at room temperature .
fig.5 shows the rapid growth of the source heat capacity versus temperature at strong couplings whereas the presence of the low @xmath101 upturn in the _ total heat capacity over t _ ratio is confirmed in fig.6 .
note that , due to the enhanced oscillators energies , the phonon heat capacity saturates at @xmath106 . according to our integration method ( eq.(10 ) , at any temperature ,
a specific set of fourier coefficients defines the ensamble of relevant particle paths over which the hopping potential @xmath31 is evaluated .
this ensamble is therefore @xmath101 dependent .
however , given a single set of path parameters one can monitor the @xmath31 behavior versus @xmath101 .
it turns out that the hopping decreases ( as expected ) by lowering @xmath101 but its value remains appreciable also at low temperatures ( @xmath107 ) . since the @xmath108 integration range is larger at lower temperatures , the overall hopping potential contribution to the total action is relevant also at low @xmath101 .
it is precisely this property which is responsible for the anomalous upturn in the heat capacity linear coefficient .
further investigation also reveals that the upturn persists both in the extremely narrow ( @xmath109 ) and in the wide band ( @xmath110 ) regimes .
our computation method accounts for a variable range hopping on the @xmath32 scale which corresponds physically to introduce some degree of disorder along the linear chain .
this feature makes our model more general than the standard ssh hamiltonian ( eq.(1 ) ) with only real space nearest neighbor hops . while hopping type mechanisms
have been suggested @xcite to explain the striking conducting properties of doped polyacetylene at low temperatures we are not aware of any direct computation of specific heat in the ssh model .
since the latter quantity directly probes the density of states and integrating over @xmath101 the _ specific heat over t _ ratio one can have access to the experimental entropy , our method may provide a new approach to analyse the transition to a disordered state which indeed exists in polymers @xcite .
in this connection it is also worth noting that the low @xmath101 upturn in the specific heat over @xmath101 ratio is a peculiar property of glasses @xcite in which tunneling states for atoms ( or group of atoms ) provide a non magnetic internal degree of freedom in the potential structure @xcite .
mapping the real space su - schrieffer - heeger model onto the time space we have developed a semiclassical version of the interacting model hamiltonian which can be attacked by path integrals methods .
the acoustical phonons of the standard ssh model have been replaced by a set of oscillators providing a bath for the electron interacting with the displacements field .
time retarded interactions are naturally introduced in the formalism through the source action @xmath47 which depends quadratically on the bare _ e - ph _ coupling strength @xmath5 . via calculation of the electronic motion path integral , the partition function can be derived in principle for any value of @xmath5 thus avoiding those limitations on the _ e - ph _ coupling range which burden any perturbative method .
particular attention has been paid to establish a reliable and general procedure which allows one to determine those input parameters intrinsic to the path integral formalism .
it turns out that a large number of paths is required to carry out low temperature calculations which therefore become highly time consuming .
the physical parameters have been specified to a narrow band system and the behavior of some thermodynamical properties , free energy and heat capacity , has been analysed for some values of the effective coupling strength lying both in the weak and in the strong coupling regime .
we find a peculiar upturn in the low temperature plots of the heat capacity over temperature ratio indicating that a glassy like behavior can arise in the linear chain as a consequence of a time dependent electronic hopping with variable range .
c.k.chiang , c.r.finker jr . ,
y.w.park , a.j.heeger , h.shirakawa , e.j.louis , s.c.gau , a.g.macdiarmid , phys . rev
. lett . * 39 * , 1098 ( 1977 ) .
a.j . heeger , s.kivelson , j.r.schrieffer , w .- p.su , rev .
* 60 * , 781 ( 1988 ) .
w.p.su , j.r.schrieffer , a.j.heeger , phys .
42 * , 1698 ( 1979 ) .
h.takayama , y.r.lin-liu , k.maki , phys .
b * 21 * , ( 1980 ) 2388 .
d.k.campbell , a.r.bishop , phys .
b * 24 * , 4859 ( 1981 ) .
h.j.schulz , phys .
b * 18 * , 5756 ( 1978 ) .
j.e.hirsch , e.fradkin , phys .
* 49 * , 402 ( 1982 ) .
s.stafstrm , k.a.chao , phys .
b * 30 * , 2098 ( 1984 ) .
k.michielsen , h.de raedt , z.phys.b * 103 * , 391 ( 1997 ) .
h.zheng , phys .
b * 56 * , 14414 ( 1997 ) .
m.zoli , phys .
b * 66 * , 012303 ( 2002 ) .
s.tang , j.e.hirsch , phys .
b * 37 * , 9546 ( 1988 ) .
y.ono , t.hamano , j.phys.soc.jpn . * 69 * , 1769 ( 2000 ) ; n.miyasaka,y.ono , j.phys.soc.jpn . *
70 * , 2968 ( 2001 ) .
m.zoli , physica c * 384 * , 274 ( 2003 ) .
r.p.feynman , phys . rev . * 97 * , 660 ( 1955 ) .
j.t.devreese , _ polarons _ in encyclopedia of applied physics ( vch publishers , ny ) * 14 * , 383 ( 1996 ) .
g.a.farias , w.b.da costa , f.m.peeters , phys .
b * 54 * , 12835 ( 1996 ) .
g.ganbold , g.v.efimov , in _ proceedings of the 6th international conference on : path integrals from pev to tev - 50 years after feynman s paper _ , world scientific publishing ( 1999 ) p.387 h.de raedt , a.lagendijk , phys .
b * 27 * , 6097 ( 1983 ) ; ibid .
, * 30 * , 1671 ( 1984 ) .
h.kleinert , _ path integrals in quantum mechanics , statistics and polymer physycs _ world scientific publishing , singapore ( 1995 ) .
j.e.hirsch , phys . rev
. lett . * 51 * , 296 ( 1983 ) .
m.cococcioni , m.acquarone , int .
b * 14 * , 2956 ( 2000 ) .
yu lu , _ solitons and polarons in conducting polymers _ world scientific , singapore ( 1988 ) .
i.s.gradshteyn , i.m.ryzhik , _ tables of integrals , series and products _ academic press ny ( 1965 ) .
s.kivelson , phys .
* 46 * , 1344 ( 1981 ) .
th.m.nieuwenhuizen , arxiv : cond - mat/9701044 .
r.c.zeller , r.o.pohl , phys.rev.b * 4 * , 2029 ( 1971 ) .
p.w.anderson , b.i.halperin , c.h.varma , philos.mag .
* 25 * , 1 ( 1971 ) .
m.zoli , phys . rev .
b * 44 * , 7163 ( 1991 ) . | the electron motion along a chain is described by a continuum version of the su - schrieffer - heeger hamiltonian in which phonon fields and electronic coordinates are mapped onto the time scale .
the path integral formalism allows us to derive the non local source action for the particle interacting with the oscillators bath .
the method can be applied for any value of the _ e - ph _ coupling .
the path integral dependence on the model parameters has been analysed by computing the partition function and some thermodynamical properties from @xmath0 up to room temperature . a peculiar upturn in the low temperature _ heat capacity over temperature _ ratio ( pointing to a glassy like behavior ) has been ascribed to the time dependent electronic hopping along the chain . |
since the discovery of the higgs boson at the lhc by the atlas and cms collaborations @xcite , experimental measurements have been focusing on the determination of its consistency with standard model ( sm ) expectation . within the current experimental and theoretical uncertainties ,
the discovered higgs boson is in agreement with the predictions for a sm higgs boson considering its spin zero property @xcite .
the measurement of higgs coupling to fermions through @xmath11 @xcite , @xmath12 @xcite , @xmath4 @xcite and to gauge bosons through @xmath13 @xcite , @xmath14 @xcite have also been performed with current lhc data and will be measured more precisely in the future run . the measurement of higgs self - coupling is also necessary to reconstruct the scalar potential of the higgs doublet field , while due to the smallness of the signal and large qcd backgrounds the probing of the higgs self - coupling has to await for a high - luminosity lhc ( hl - lhc ) , 33 tev , or 100 tev @xmath3 collider @xcite .
it is important to stress that so far all experimental determinations of the higgs @xmath2 properties have been obtained from the higgs to vector boson couplings with lepton final states . the hypothesis that higgs boson being a pure pseudoscalar state has been excluded by the present data since the @xmath15 and @xmath16 signals from @xmath17
@xcite and @xmath14 @xcite decays have been observed . in particular ,
the angular distributions of the lepton pairs in the @xmath18 channel are sensitive to the spin - parity of the higgs boson .
although the present lhc data strongly prefer that @xmath0 is a @xmath19 state , it is not yet excluded that the higgs boson has a pseudoscalar component @xcite . using the present measured signal strengths in various higgs search channels , theoretical studies on the higgs @xmath2 properties
have been performed , and constraints on the mixing angle of the @xmath2-even and @xmath2-odd component are given model dependently @xcite .
there are also investigations on how and to what extent to pin down the higgs @xmath2 mixing angle via other higgs decay channels at the lhc and future linear collider @xcite . among all the higgs production channels , the investigation of the @xmath4 process plays a complementary role in higgs characterization in the sense that it is very sensitive to the relative magnitudes of the @xmath2-even and @xmath2-odd top - higgs yukawa coupling coefficients .
although the @xmath4 production cross section is only around 1/200 of the inclusive higgs production for @xmath20 with a mass of @xmath21 gev at run i lhc and suffers from large multiplicity of the objects in the final state from top pair decays , the hl - lhc and proposed future hadron colliders with higher collision energy , the high energy lhc ( he - lhc ) , sppc , etc . ,
are capable of enhancing the @xmath4 event rates . in this paper
, we will revisit certain operators defined from the momenta of the @xmath4 process , which are sensitive to the coefficients of different @xmath2-mixed states .
we choose two particular higgs decay modes @xmath22 and @xmath11 for clearness of signal and the wealth of event rates respectively , and study the behaviour of weighted moments at the lhc and the future hadron colliders .
we find that these operators for the corresponding @xmath10 and @xmath7 processes have a good discrimination power for a @xmath2-even higgs from a @xmath2-mixed state .
to accommodate possible deviations of higgs to top quark @xmath4 coupling in both strength and @xmath2 properties , we parametrize the @xmath4 coupling in the following form @xmath23 where @xmath24 is the vacuum expectation value ( vev ) of higgs field , and @xmath25 and @xmath26 corresponds to the @xmath2-even and @xmath2-odd coupling coefficient respectively . in particular , the sm higgs boson has @xmath27 and @xmath28 .
the angle @xmath29 is an indication of the degree of mixture @xmath2 in the physical higgs @xmath0 .
this form of higgs fermion interaction will be assumed also for other fermions in our discussions .
the dominant higgs couplings to vector gauge bosons , @xmath30 and @xmath31 will come from the @xmath2-even component of @xmath0 in the following form@xmath32 the main production channel of higgs boson at the lhc is via gluon fusion process .
the effective vertices for a @xmath2-mixed higgs state interacting with gluons at one loop order are given by @xcite @xmath33h,\end{aligned}\ ] ] with @xmath34 defined as the scalar and pseudoscalar form factors for @xmath35 which retain the dominant contributions from top and bottom quarks , @xmath36 .
@xmath37 is the gluon field strength and @xmath38 is the dual of @xmath37 given by @xmath39 .
the two form factors @xmath40 and @xmath41 can be expressed in terms of scaling function @xmath42 as @xmath43,\quad f_b(\tau)=\tau^{-1}f(\tau),\end{aligned}\ ] ] where @xmath44 ^ 2 , & \text{if}\quad \tau\leq 1,\\ -\frac{1}{4}\left[\ln(\frac{\sqrt{\tau}+\sqrt{\tau-1}}{\sqrt{\tau}-\sqrt{\tau-1}})-i\pi\right]^2 , & \text{if}\quad\tau>1 .
\end{cases}\end{aligned}\ ] ] the higgs to diphoton effective coupling is formulated as @xcite @xmath45h,\end{aligned}\ ] ] where @xmath46 and @xmath47 are the photon field strength and its dual , respectively .
the form factors @xmath48 are given by @xmath49 $ ] , @xmath50 with @xmath51 the colour factor for quarks and @xmath52 the electric charge of the corresponding particle .
@xmath53 is from @xmath30 boson loop contribution with @xmath54
deviations of higgs boson coupling to top quark from sm one can show up in different ways .
we study how to identify such deviations using @xmath55 with the higgs boson identified by @xmath56 and @xmath57 since the former has the largest branching ratio and the latter has clearer signals compared with other decay modes .
both atlas @xcite and cms @xcite have performed experimental searches on the @xmath4 production process , with @xmath0 decays into @xmath22 and @xmath11 separately . for the @xmath7 process , collecting the @xmath58 tev data final state are categorized according to their jet and @xmath59-tagged jet multiplicities .
the observed cross section upper bound is @xmath60 @xcite , and @xmath61 @xcite . for the @xmath10 process ,
the observed upper limit on cross section is @xmath62 @xcite .
we still need to wait for more accurate data to draw conclusions about the production cross section .
the interaction in eq.([htt ] ) can cause deviation in @xmath4 cross section from sm prediction and in the higgs decays .
we have evaluated the @xmath4 cross sections using madgraph5 particularly heft model where the higgs effective theory is included @xcite with a mild rapidity cuts on the top and antitop quarks @xmath63 .
the results with the parameter @xmath27 and range of @xmath29 from 0 to @xmath64 and collision energy @xmath65 13 tev , 14 tev , 33 tev and 100 tev with @xmath66 gev are shown in fig .
[ fig : strength ] .
here we would like to point out that one should consistently take the high - order correction to the cross section into account . for @xmath4 process
, the nlo prediction for a higgs boson mass of 125 gev has cross section around 86 fb , 130 fb , 611 fb and 33700 fb @xcite at @xmath67 tev , 8 tev , 14 tev and 100 tev . estimated using cteq6.6 , the nlo corrections are positive .
this is higher than that obtained with the cross section of about 72 fb , 107 fb , 476 fb and 24010 fb using the lo results .
a @xmath68-factor of 1.19 - 1.40 @xcite should be applied when we use the values shown in fig .
[ fig : strength ] . [ cols="^ " , ] if we further include the @xmath69 pair decay , a total factor of the change on the cross section @xmath8 and discrimination power @xmath70 can be estimated by the corresponding decay branching ratios .
we choose top quark decays semileptonically and antitop quark decays hadronically in order for the identification of @xmath1 and @xmath71 and also the reconstruction of the transverse momentum of @xmath1 and @xmath71 .
therefore , the decay products from top pair will be @xmath72-jet @xmath73 , the reduced signal cross section can be estimated by @xmath74 with @xmath75 for @xmath10 process .
while for @xmath7 cross section , an additional suppression factor from @xmath59-tagging efficiency ( @xmath76 ) should be included leading to @xmath77 .
consequently , by including top pair decay the value of @xmath70 will be affected by a total suppression factor of @xmath78 0.42 ( 0.25 ) for @xmath10 ( @xmath7 ) from eq.([eq : d ] ) .
given @xmath79 , @xmath80 , @xmath81 , @xmath82 and background event number @xmath83 , it is straightforward to compute the required signal event number @xmath84 for a sm - like higgs to achieve certain value of @xmath70 .
for example , if one wants to discriminate @xmath4 coupling of @xmath85 from @xmath86 at the @xmath87 statistical level , the prediction from @xmath10 and @xmath7 are different .
the @xmath10 background event number is @xmath88 ( 2280 ) at @xmath89 ( 100 ) tev , estimated from the operator with the largest discrimination power @xmath90 we find that @xmath91 requires signal event number to be around @xmath92 ( 1950 ) .
the simulated value for lo signal rates are about 160 ( 9870 ) .
therefore , using @xmath10 channel it is difficult to discriminate the @xmath4 of @xmath85 coupling from a @xmath86 mixed coupling at @xmath87 statistical level at 13 tev .
however , a future @xmath3 collider at 100 tev can produce enough signal events to achieve the discrimination at larger than @xmath87 level .
the situation is better in @xmath7 channel .
the sm @xmath7 background has a event number of around @xmath93 ( 7.8@xmath94 ) at @xmath89 ( 100 ) tev , estimated from the operator with largest discrimination power @xmath95 we find that @xmath91 requests signal events number to be @xmath96 ( @xmath97 ) .
the real lo signal event number is around @xmath98 ( @xmath99 ) , which is significantly lager than required .
consequently , from the @xmath7 process , one is able to discriminate the @xmath4 coupling with @xmath85 from a @xmath86 @xmath2-mixed state at more than @xmath87 statistical level with any collision energy larger than 13 tev at @xmath3 collider .
the figures [ fig : d_ttrr ] and [ fig : d_ttbb ] show the corresponding @xmath70 values for @xmath10 and @xmath7 final state after including the above decay and @xmath59-tagging efficiencies with the @xmath2 mixing angle running from @xmath100 to @xmath64 with collision energy at 13 tev , 14 tev , 33 tev and 100 tev . for both processes ,
the number of events are obtained with integrated luminosity assumed to be @xmath101 and an enhancing factor of @xmath102 on @xmath70 will appear if we consider the hl - lhc with @xmath1033 ab@xmath104 .
comparison between each two operators shows that @xmath105 , and @xmath95 are most useful balancing the sensitivity and errors , among which @xmath90 probably performs the best .
@xmath90 is also easy to be constructed which is irrelevant to the @xmath106-component of top and antitop momentum @xmath107 .
the results shown in fig . [ fig : d_ttrr ] tell that for @xmath10 process at @xmath108= 13 tev and 14 tev the discrimination powers for operators @xmath109-@xmath95 are generally less than three in the allowed @xmath2 mixing region . when the integrated luminosity is increased to be 3 ab@xmath104 at hl - lhc
, @xmath110 can be reached in the region @xmath111 for all operators .
it is much more optimistic at @xmath112 tev and @xmath113 tev , which shows that all the operators can give large enough statistical significance in the region @xmath114 .
the operator @xmath90 gives the best discrimination power irrelevant of the collision energy and @xmath2 mixing angle .
all the operators reach maximal discrimination power at @xmath115 . comparing the @xmath7 process with @xmath10 , we want to emphasize , as can be seen from fig .
[ fig : d_ttbb ] , that @xmath56 decay modes shows better discrimination powers due to the large event rate of the higgs to @xmath11 decay branching ratio . for @xmath7 process
, the operator @xmath90 and @xmath95 give the best discrimination powers .
note that if we make full analysis considering final state jets and leptons from top quark decay , and impose further cuts on the jet multiplicities , the value of discrimination power @xmath70 will appear even better than present numbers .
the higgs boson @xmath0 has the largest coupling to the top quark @xmath1 .
the @xmath4 interaction can be sensitive to the investigation of new physics beyond sm . in this work ,
we have studied the potential of determining higgs boson @xmath2 properties at the lhc and future 33 tev and 100 tev @xmath3 coliders by analysing various operators formed from final states variables in @xmath4 production .
a @xmath2 violating higgs coupling to top quark can cause significant deviations from sm predictions for @xmath116 .
we have evaluated the @xmath4 cross sections using madgraph5 particularly heft model .
we find that the cross sections become considerably smaller than those of the sm predictions when cp violating component gets bigger .
this fact may be used to distinguish the sm @xmath4 coupling with that from a beyond sm . but this may be caused by an overall scaling factor to the coupling . to further identify the relative magnitudes of the @xmath2-even and @xmath2-odd interactions in @xmath117 , certain sensitive weighted moments in @xmath118 process can help to achieve this .
we obtained results for discrimination power for several operators with the higgs boson identified by @xmath56 and @xmath57 .
for an integrated luminosity is 300 @xmath9 , we find that , with @xmath119 , the discrimination power will be below 3@xmath8 at the lhc , while for 33 tev and 100 tev colliders , more than 3@xmath8 sensitivity can be reached .
on the other hand , @xmath120 process can provide more than 3@xmath8 discrimination power in a wide range of allowed higgs to top couplings for the lhc , the 33 tev and 100 tev colliders .
the work was supported in part by moe academic excellent program ( grant no .
102r891505 ) and most of roc , and in part by nnsf(grant no.11175115 ) and shanghai science and technology commission ( grant no.11dz2260700 ) of prc .
s. chatrchyan _ et al .
_ [ cms collaboration ] , phys .
b * 716 * , 30 ( 2012 ) [ arxiv:1207.7235 [ hep - ex ] ] .
g. aad _ et al . _ [ atlas collaboration ] , phys .
b * 716 * , 1 ( 2012 ) [ arxiv:1207.7214 [ hep - ex ] ] .
g. aad _ et al . _ [ atlas collaboration ] , phys .
b * 726 * , 120 ( 2013 ) [ arxiv:1307.1432 [ hep - ex ] ] .
v. khachatryan _ et al . _
[ cms collaboration ] , arxiv:1411.3441 [ hep - ex ] .
[ atlas collaboration ] , atlas - conf-2013 - 079 .
s. chatrchyan _ et al . _
[ cms collaboration ] , phys .
d * 89 * , 092007 ( 2014 ) [ arxiv:1312.5353 [ hep - ex ] ] .
m. j. dolan , c. englert and m. spannowsky , jhep * 1210 * , 112 ( 2012 ) [ arxiv:1206.5001 [ hep - ph ] ] .
j. baglio , a. djouadi , r. grber , m. m. mhlleitner , j. quevillon and m. spira , jhep * 1304 * , 151 ( 2013 ) [ arxiv:1212.5581 [ hep - ph ] ] . s. chatrchyan _ et al . _
[ cms collaboration ] , phys .
lett . * 110 * , 081803 ( 2013 ) [ arxiv:1212.6639 [ hep - ex ] ] .
a. freitas and p. schwaller , phys .
d * 87 * , no . 5 , 055014 ( 2013 ) [ arxiv:1211.1980 [ hep - ph ] ] . g. bhattacharyya , d. das and p. b. pal , phys .
d * 87 * , 011702 ( 2013 ) [ arxiv:1212.4651 [ hep - ph ] ] .
k. cheung , j. s. lee and p. y. tseng , jhep * 1305 * , 134 ( 2013 ) ; phys .
d * 90 * , 095009 ( 2014 ) .
j. shu and y. zhang , phys .
lett . * 111 * , no . 9 , 091801 ( 2013 ) [ arxiv:1304.0773 [ hep - ph ] ] .
s. inoue , m. j. ramsey - musolf and y. zhang , phys .
d * 89 * , 115023 ( 2014 ) [ arxiv:1403.4257 [ hep - ph ] ] .
s. bolognesi , y. gao , a. v. gritsan , k. melnikov , m. schulze , n. v. tran and a. whitbeck , phys .
d * 86 * , 095031 ( 2012 ) [ arxiv:1208.4018 [ hep - ph ] ] . c. englert , d. goncalves - netto , k. mawatari and t. plehn , jhep * 1301 * , 148 ( 2013 ) [ arxiv:1212.0843 [ hep - ph ] ] .
j. f. gunion , b. grzadkowski and x. g. he , phys .
* 77 * , 5172 ( 1996 ) [ hep - ph/9605326 ] .
p. s. bhupal dev , a. djouadi , r. m. godbole , m. m. muhlleitner and s. d. rindani , phys .
lett . * 100 * , 051801 ( 2008 ) [ arxiv:0707.2878 [ hep - ph ] ] . s. berge , w. bernreuther and j. ziethe , phys . rev .
lett . * 100 * , 171605 ( 2008 ) [ arxiv:0801.2297 [ hep - ph ] ] .
x. g. he , y. tang and g. valencia , phys .
d * 88 * , 033005 ( 2013 ) [ arxiv:1305.5420 [ hep - ph ] ] .
a. pilaftsis and c. e. m. wagner , nucl .
b * 553 * , 3 ( 1999 ) [ hep - ph/9902371 ] .
x. g. he , g. valencia and h. yokoya , jhep * 1112 * , 030 ( 2011 ) [ arxiv:1110.2588 [ hep - ph ] ] .
j. s. lee , a. pilaftsis , m. s. carena , s. y. choi , m. drees , j. r. ellis and c. e. m. wagner , comput .
phys . commun .
* 156 * , 283 ( 2004 ) [ hep - ph/0307377 ] .
j. f. gunion , h. e. haber , g. l. kane and s. dawson , front .
* 80 * , 1 ( 2000 ) .
m. spira , a. djouadi , d. graudenz and p. m. zerwas , nucl .
b * 453 * , 17 ( 1995 ) [ hep - ph/9504378 ] .
[ atlas collaboration ] , atlas - conf-2014 - 011 .
s. dittmaier _ et al .
_ [ lhc higgs cross section working group collaboration ] , arxiv:1101.0593 [ hep - ph ] .
w. beenakker , s. dittmaier , m. kramer , b. plumper , m. spira and p. m. zerwas , phys .
lett . * 87 * , 201805 ( 2001 ) [ hep - ph/0107081 ] ; nucl .
b * 653 * , 151 ( 2003 ) [ hep - ph/0211352 ] .
s. dawson , l. h. orr , l. reina and d. wackeroth , phys .
d * 67 * , 071503 ( 2003 ) [ hep - ph/0211438 ] .
s. frixione , v. hirschi , d. pagani , h. s. shao and m. zaro , jhep * 1409 * , 065 ( 2014 ) [ arxiv:1407.0823 [ hep - ph ] ] .
f. demartin , f. maltoni , k. mawatari , b. page and m. zaro , eur .
j. c * 74 * , no .
9 , 3065 ( 2014 ) [ arxiv:1407.5089 [ hep - ph ] ] . | the higgs boson @xmath0 has the largest coupling to the top quark @xmath1 among the standard model ( sm ) fermions .
this is one of the ideal places to investigate new physics beyond sm . in this work
, we study the potential of determining higgs boson @xmath2 properties at the lhc and future 33 tev and 100 tev @xmath3 colliders by analysing various operators formed from final states variables in @xmath4 production .
the discrimination power from sm coupling is obtained with higgs boson reconstructed from @xmath5 and @xmath6 .
we find that @xmath7 process can provide more than 3@xmath8 discrimination power with 300 @xmath9 integrated luminosity in a wide range of allowed higgs to top couplings for the lhc , the 33 tev and 100 tev colliders .
for @xmath10 the discrimination power will be below 3@xmath8 at the lhc , while for 33 tev and 100 tev colliders , more than 3@xmath8 sensitivity can be reached . |
quantum memory is , in a wide sense , a device that stores or freezes a quantum state both spatially and in time .
a highly successful example is that a light pulse is frozen in a cloud of atoms @xcite .
also quantum memory is of technological importance particularly in quantum information science , such as the quantum repeater for quantum communication @xcite . because of these scientific / technological importance
, the field of quantum memory has experienced significant progress in both theory and experiment .
we refer to for instance @xcite for reviewing the current situation of this very active research area .
basic schematic of an ideal quantum memory .
@xmath0 denotes the mode of the memory subsystem used for the storage , while @xmath1 is the mode of the buffer subsystem , which transports the input state to ( from ) the memory subsystem from ( to ) the optical field with mode @xmath2 ( @xmath3 ) .
the system structure can be switched from the stage ( a ) to ( b ) , or from ( b ) to ( c ) , by tuning some controllable parameters . in the stage ( b ) , @xmath0 is decoupled from @xmath1 and thus the optical field , implying that @xmath0 is decoherence free . in the stages ( a ) and ( c ) , on the other hand , @xmath0 couples to the field for state transportation . ] now let us see a basic and general schematic of an ideal quantum memory in an abstract way .
first of all , we assume that the system contains a subsystem that can store any quantum state without loss in a long period of time ; we call this specific component the _ memory subsystem_. ideally , the memory subsystem should be completely decoupled from the other system components and surrounding environment during it stores the state ; i.e. , it is _ decoherence free ( df ) _ @xcite . note that , in this storage stage , the memory subsystem is decoupled even from the channel used for transferring an input state or retrieving the stored state .
hence , the second assumption is that , during the writing / reading process , the system can be tuned so that the memory subsystem couples to that transportation channel .
that is , the system should be the one that contains a tunable port switching the opening / closing of the memory subsystem .
indeed this basic schematic is employed in each specific memory device . in the case of atoms based on the electromagnetic induced transparency ( eit ) effect ,
an isolated memory subsystem is served by a set of metastable collective atomic states , and an external control field ( with rabi frequency @xmath4 ) can switch on / off of the coupling between the metastable states and the optical field for state transportation @xcite .
we also find successful demonstrations in optical cavity or optomechanical oscillator arrays @xcite , where the switching mechanism is served by adiabatic frequency detuning of the memory subsystem .
further , a similar switching procedure is employed in the photon echo quantum memory @xcite .
note that , if the system does not contain a switching mechanism , it is generally impossible to perfectly transfer an unknown state to a memory subsystem ( i.e. df subsystem ) @xcite .
the above - mentioned basic schematic for quantum memory is illustrated in fig .
[ df memory schematic ] . in the writing stage ( a )
, an input state is sent to the system over an input channel with mode @xmath5 .
let us here assume that , by devising a certain nice procedure " , all the input state is transferred to the memory subsystem with mode @xmath0 .
we then close the port , and @xmath0 becomes decoherence free . in this storage stage ( b ) , ideally , the memory subsystem can store the state for a long time . finally , in the reading stage ( c ) , by opening again the port we can retrieve the state at any later time , which is sent over the output channel with mode @xmath6 .
so what is a certain nice procedure " to achieve the best or hopefully perfect state transfer ?
for some typical quantum memories such as eit and off - resonant raman memories , we can explicitly formulate this problem ; that is , the question is what is the optimal temporal shape of the pulsed light field carrying the input state .
this optimization problem is very important and actually has been deeply investigated in several papers ; for instance , towards the most efficient atomic memory with eit , in @xcite the input wave packet as well as the controllable rabi frequency @xmath4 ( see the second paragraph ) are carefully designed , although the method is a heuristic one based on iterative tuning of the parameters .
it is also notable that the so - called _ rising exponential _
type function is found as an effective pulse shape @xcite .
this is a pulse with exponential increase of e.g. the probability density of photon counting or the amplitude of a coherent field , which can be physically implemented @xcite .
motivated by these results and further the fact that a large - scale quantum memory is required for constructing practical quantum communication architecture , the following questions naturally arise .
is there a pulse shape for achieving perfect state transfer for general and large - scale quantum memory networks ?
is there a general yet simple guideline for synthesizing such desirable pulse shape ?
what is the fundamental origin of the rising exponential function as a desirable pulse shape ? is the rising exponential pulse optimal in a certain sense ?
also , is it effective even for a large - scale network ? solving these problems should accelerate the progress of the quantum memory research in its deeper understanding and practical implementation . in this paper
, we consider a general _ passive linear system _ , which models a wide variety of systems such as optical systems @xcite , nano - mechanical oscillators @xcite , vibration mode of a trapped particle @xcite , and atomic ensembles @xcite . as mentioned before , the system is assumed to contain a tunable memory subsystem ; that is , by tuning a certain parameter , we can switch opening / closing of the memory subsystem , which is df during the storage period . note that in our case this df subsystem corresponds to a system having the so - called _ dark mode _ @xcite .
another assumption is that an unknown input state to be transferred is encoded in a continuos - mode single - photon field or a coherent field . with these setups ,
we give answers to all the questions posed in the above paragraph .
the essential idea is the use of _ zero - dynamics principle_. this concept originates from the classical notion of zero " of a _ transfer function _ , which is a fundamental tool used in systems and control theory @xcite .
more precisely , for a general linear classical system @xmath7 where @xmath8 is the input , @xmath9 is the output , and @xmath10 are matrices , its input - output relation is simply characterized by a transfer function @xmath11=d+c(si - a)^{-1}b$ ] as @xmath12=h[s]u[s]$ ] ( see section 4.3 for detailed description ) ; then , if the input is given by @xmath13 , where @xmath14 is a _ zero _ of the transfer function ( meaning @xmath15=0 $ ] ) , we have @xmath16e^{zt } = 0 $ ] under some additional conditions .
that is , the _ zero - dynamics _ is a system whose output is always zero .
in fact , the concept of zero - dynamics is very important in analysis and synthesis for even general nonlinear systems @xcite .
actually , the zero - dynamics principle described above can be directly applied to general quantum memory problem ; if the state transfer is perfectly carried out , the input field must be completely absorbed in the system , and the output field must not contain any small pieces of the input field .
that is , as a principle , the output should be zero " during the writing and storage stages .
surprisingly , this simple zero - dynamics principle leads us to prove , very easily , that a rising exponential function is a unique pulse shape achieving the perfect state transfer for general ( and thus large - scale ) linear passive networks .
moreover , based on this first main result , we give an explicit , simple , and general procedure for designing the wave packet carrying an unknown state that is as a result perfectly absorbed in the memory subsystem .
this paper is organized as follows .
section 2 provides some preliminaries , describing general passive linear systems , optical field states , and linear df subsystems .
also the problem is explicitly formulated in section 2.4 . in section 3
, we study a simple single - mode oscillator as a memory system , to show the fact that a rising exponential function appears as a unique pulse shape achieving the perfect state transfer ; based on this result , the idea of zero - dynamics principle is discussed .
section 4 provides our first main result ; for general passive linear systems , we derive the rising exponential pulse from the zero - dynamics principle .
also we give a useful interpretation of this fact in terms of the transfer function . in section 5 ,
we present our second main result , showing the concrete procedure for perfect writing , storage , and reading ; this explicitly shows the pulse shape to be synthesized for perfect state transfer from the optical field to the memory subsystem .
section 6 is devoted to derive the time evolution equations of some statistical quantities of the dynamics , which is useful for numerical simulation . in section 7
, we study a linear network composed of atomic ensembles trapped in a cavity , which contains a tunable memory subsystem ; this example shows how the designed pulse actually looks like and how the state transfer from the optical field to the atomic ensembles evolves in time .
section 8 summarize the paper and discusses some future works . in appendix
a , we briefly examine the case of an _ active _ linear memory system .
appendix b provides a case study comparing the zero - dynamics principle and the so - called _ dark state principle_. * notation : * we use the following notations : for a matrix @xmath17 , the symbols @xmath18 , @xmath19 , and @xmath20 represent its hermitian conjugate , transpose , and complex conjugation in elements of @xmath21 , respectively ; i.e. , @xmath22 , @xmath23 , and @xmath24 . for a matrix of operators we use the same notation , in which case @xmath25
denotes the adjoint to @xmath26 . for a time - dependent variable @xmath27
, we denote @xmath28 .
in this paper , we study a general linear open system composed of @xmath29 oscillators with mode @xmath30^\top$ ] that couples to an optical field with continuous mode @xmath5 ; hence they satisfy @xmath31=\delta_{ij}$ ] and @xmath32=\delta(s - t)$ ] .
the system is driven by a quadratic hamiltonian @xmath33 with @xmath34 an @xmath29-dimensional hermitian matrix .
the system and the field instantaneously couple through the interaction hamiltonian @xmath35 $ ] with @xmath36 an @xmath29-dimensional complex _ row _ vector .
then the unitary operator @xmath37\ ] ] ( @xmath38 is the initial time ) produces the heisenberg equations of @xmath39 and @xmath40 as follows ; @xmath41 where @xmath42^\top$ ] and @xmath43 .
the second term in @xmath21 stems from the ito - correction .
note also that , due to the ideal markov property , the optical field @xmath5 does not have its own dynamical time evolution ; rather @xmath6 represents the output of the system .
the above open system with input @xmath5 and output @xmath6 is called the _ passive linear system _ in the sense that it does not contain any active component such as an optical parametric amplifier in optics case .
as mentioned in section 1 , a passive linear system can model a wide variety of systems .
the main reason why we focus on this class of general systems is that it preserves the total energy during the interaction between the system and the field .
this implies that the perfect state transfer is equivalent to the perfect energy transfer , which as a result allows us to use the zero - dynamics principle to characterize the perfect memory ; this will be discussed in detail in section 4 .
also in appendix a , we show a brief case study where the system contains an active component . for a general treatment of passive linear systems , see @xcite ; the notation used in this paper follow these references , where in general @xmath36 is an @xmath44 complex matrix representing @xmath45 input optical fields . in this paper
, we consider the case where the input is given by a single photon state or a coherent state , which is carried by an optical pulse field with continuous - mode @xmath5 .
they are described as follows .
* single photon field state : * the single photon state in a single mode system is , as is well known , produced by acting a creation operator @xmath46 to the ground state @xmath47 ; i.e. @xmath48 to describe the continuous - mode single photon field state , we define the annihilation and creation process operators @xmath49 @xmath50 is an associated function in @xmath51 , representing the shape of the optical pulse field .
also @xmath50 satisfies the normalization condition @xmath52 . due to this ,
@xmath53 and @xmath54 satisfy the usual ccr ; @xmath55=1 $ ] .
the single photon field state is , in a similar way as above , produced by acting the creation process operator on the vacuum field @xmath56 as follows @xcite : @xmath57 due to the normalization condition of @xmath50 , we find that @xmath58
. also note the relation @xmath59 ; thus , @xmath50 has the meaning of the wave function with @xmath60 the probability of photo detection per unit time .
let us now assume that the pulse shape @xmath50 can be expanded as @xmath61 the coefficient @xmath62 represents ( unknown ) classical information encoded in the optical field , which satisfies @xmath63 , and @xmath64 is a set of orthonormal functions satisfying @xmath65 note that @xmath29 is the number of modes of the system .
then the single photon field state can be written as @xmath66 @xmath67 is called the single photon code state with pulse shape @xmath68 @xcite ; from the condition , they are orthonormal , i.e. @xmath69 . also the field operator @xmath70 satisfies the ccr @xmath71=\delta_{jk}$ ] .
* coherent field state : * another important state is a coherent state .
a coherent state in a single mode system is generated by acting a displacement operator @xmath72 on the ground state as follows ; @xmath73 where @xmath74 denotes the amplitude of @xmath75 .
likewise , a coherent field state is defined in terms of the creation and annihilation process operators as follows ; @xmath76 { |{0}\rangle}_f,\ ] ] where @xmath77 is a complex - valued function , representing the amplitude of the state ; that is , this is a coherent pulse field modulated with envelope function @xmath77 .
note that @xmath77 is not necessarily normalized , but its power @xmath78 is finite .
now we assume that @xmath77 is given , in terms of the orthonormal functions @xmath64 , by @xmath79 where @xmath80 represents ( unknown ) classical information to be stored .
the power of @xmath81 ,
i.e. the mean photon number in unit time , is then given by @xmath82 .
the coherent field state is as a result described by @xmath83 { |{0}\rangle}_f.\ ] ] note that this is not a superposition of the coherent field states @xmath84 , unlike the single photon field state .
the general theory of df subsystems states that , if a df subsystem exists , then the system hilbert space @xmath89 is decomposed to @xmath90 , where any observable in @xmath91 evolves unitarily . in our case ,
the decomposition is of the form @xmath92 ; thus the system variables are , more precisely , represented by @xmath93 and @xmath94 .
this special class of continuous - variable df subsystems appears in several situation @xcite . also for a general theory of the linear df subsystem , including a necessary and sufficient condition for a given linear system to have a df mode ,
see @xcite .
here we describe the problem discussed throughout the paper .
our system is the general passive linear system , and it is assumed to be tunable ; that is , by appropriate tuning of its parameter(s ) , a part of the system , the memory subsystem , couples or decouples to the optical field carrying the information .
hence the memory subsystem can be switched to be a df or a non - df subsystem . as mentioned in section 2.3 , the memory subsystem stores the state during it is in the df mode , while it should be in the non - df mode when we transfer the input state or retrieve the stored state . in particular , we assume that the matrix @xmath95 in the writing / reading stages is _ hurwitz _ , i.e. , the real parts of all the eigenvalues of @xmath21 are negative ; as will be shown later , this condition is necessary for perfect state transfer . on the other hand , in the storage stage
, the dynamical equation takes the form , thus @xmath21 is not hurwitz . the field s initial state is given by @xmath96 in the case of single photon input field or @xmath81 in the case of coherent input field . the system s initial state @xmath97 is assumed to be separable , hence it is given by @xmath98 .
in particular , we will set it to be the ground state @xmath99 satisfying @xmath100 . at time @xmath38 , the system and the field start to interact , via the unitary operator @xmath101 given in eq . .
the composite state at time @xmath102 is then given by @xmath103 or @xmath104 . in this writing stage ,
the memory subsystem is in the non - df mode , so it couples to the field .
but once the state transfer has been completed , then we switch the system parameters so that the memory subsystem becomes decoherence free , and its state is preserved during the storage stage .
hence , it would be desirable if the state @xmath105 is of the separable form @xmath106 and the memory subsystem s state @xmath107 contains the full information of the input field state @xmath96 or @xmath81 .
therefore , our goal is to appropriately synthesize the pulse shape @xmath50 or @xmath77 , or more precisely their basis functions @xmath64 , so that the above desirable transition from the field s initial state to @xmath107 occurs .
lastly we remark on the switching configuration . in general ,
the system matrices @xmath34 and @xmath36 ( and thus @xmath21 ) can change in time ( i.e. time varying ) in order to realize high quality quantum memory .
for instance in @xcite , the authors consider the time varying system matrices depending on the control field with frequency @xmath4 , which is optimized via a heuristic method . on the other hand , in this paper , we assume that @xmath34 and @xmath36 are time varying , but they are constant in each stage of the memory procedure ; that is , they are _ piecewise constant_. in particular , we will take the same system matrices in the writing and reading stages .
in this section , we examine a simple case where the memory system is given by a single - mode passive linear system . as will be shown later
, this system does not contain a tunable df component , so it does not work as a perfect storage device .
rather the purpose here is that , by focusing only on the writing stage , requiring perfect state transfer uniquely determines the pulse shape of the input optical field .
based on this result , we then derive the explicit form of the output field , and show the notion of zero - dynamics in this case . here
we study only the single - photon input case , but it is straightforward to obtain a similar result in the case of coherent field state .
let us consider the following single - mode ( i.e. @xmath108 ) linear system interacting with an optical field , which is obtained by setting @xmath109 and @xmath110 in eq .
: @xmath111 where @xmath112 is the interaction strength ; in optics case , this system is typically given by an optical cavity with @xmath112 proportional to the transmissivity of the coupling mirror .
the goal is to send a single photon state over the input pulse field and write it perfectly down to the system .
note that , however , clearly this system does not contain a tunable df component , hence our interest here is only in the state transfer .
the input state is given by eq .
, which is now essentially @xmath113 .
thus in this case let us take a superposition of the vacuum and the single photon field state @xmath114 where @xmath115 are the encoded ( unknown ) classical information .
recall that the system s initial state is set to the ground state @xmath116 .
the dynamical equation has the following solution : @xmath117 where @xmath118 is the time we stop the interaction .
this can be rewritten as @xmath119 where @xmath120 note that @xmath121 .
equation can be further represented as @xmath122 @xmath123 is the field creation operator with pulse shape @xmath124 , which is defined in eq . ,
and @xmath125 is the unitary operator given in eq . .
from the above equation we find that , in the limit @xmath126 , the field creation operator @xmath123 is completely mapped to the system creation operator @xmath127 .
this means that the perfect state transfer from the optical pulse field to the system mode can be carried out as shown below . in the schrdinger picture , the whole state at time @xmath118 is given by @xmath128 { |{0}\rangle}_s{|{0}\rangle}_f \nonumber \\ & & \hspace*{-1.5em } = u(t_0 , t_1 ) \big [ \alpha i_s\otimes i_f + \beta i_s\otimes b^*(\xi)\big ] u^*(t_0 ,
t_1 ) u(t_0 , t_1 ) { |{0}\rangle}_s{|{0}\rangle}_f \nonumber \\ & & \hspace*{-1.5em } = \big [ \alpha i_s\otimes i_f + \beta u(t_0 , t_1)(i_s\otimes b^*(\xi))u^*(t_0 , t_1 ) \big ] { |{0}\rangle}_s{|{0}\rangle}_f , \nonumber\end{aligned}\ ] ] where in the last equality @xmath129 is used .
let us now set the input pulse shape to be @xmath130 .
then , from eq .
, we have @xmath131 { |{0}\rangle}_s{|{0}\rangle}_f \nonumber \\ & & \hspace*{-1.6em } = \big [ \alpha { |{0}\rangle}_s + \frac{\beta}{\sqrt{1-e^{-\kappa(t_1-t_0 ) } } } { |{1}\rangle}_s \big ] \otimes { |{0}\rangle}_f \nonumber \\ & & \hspace*{2em } \mbox { } - \frac{\beta e^{-\kappa(t_1-t_0)/2}}{\sqrt{1-e^{-\kappa(t_1-t_0 ) } } } u(t_0 , t_1 ) a^*(t_0 ) u^*(t_0 , t_1 ) { |{0}\rangle}_s{|{0}\rangle}_f . \nonumber\end{aligned}\ ] ] therefore , in the limit @xmath126 , we have @xmath132 \otimes { |{0}\rangle}_f,\ ] ] which means that the input field state is completely transferred to the system state . in particular , in the case @xmath126 and @xmath133 , the input pulse shape is given by @xmath134 this is called the _ rising exponential pulse _ @xcite .
it is clear from the above discussion that the rising exponential is the unique pulse shape for perfect state transfer from the single photon field to the system .
let us further study the final state @xmath135 , where the input pulse shape is now set to @xmath136 it is possible to obtain the explicit solution : @xmath137 where @xmath138 with @xmath139 first , at @xmath133 , we have @xmath140 + \frac{2\kappa\beta}{\kappa+\gamma}\sqrt{\frac{\gamma}{\kappa } } { |{1}\rangle}_s{|{0}\rangle}_f,\ ] ] which becomes @xmath141 only when @xmath142 .
that is , the frequency bandwidth of the input pulse shape has to be exactly equal to that of the memory system to attain the perfect state transfer .
this is a form of the so - called _ impedance matching _ for efficient energy transfer ; in section 7 , we will discuss the matching condition in a more practical setup where the system is composed of a cavity and atomic ensembles .
next , in the limit @xmath143 , the whole state again becomes separable : @xmath144 where @xmath145 this @xmath146 represents the pulse shape of the output optical field over the whole period .
hence , if @xmath142 , the output field is vacuum in the writing stage @xmath147 ; this means that the single photon input field is completely absorbed into the system , and the output field does not contain any pieces of the input state . in the optics case where the system is given by a cavity , a physical meaning of
this fact is that it happens destructive interference between the light field reflected at the coupling mirror and the transmissive light field leaking from the cavity ; as a result , the output field of the cavity is always in vacuum , i.e. zero " , while the system s state still dynamically changes in time . in general , the dynamics of a system whose output is always zero is called the zero dynamics @xcite . hence in this case
the cavity dynamics during the writing process is exactly a zero dynamics .
in this section , based on the so - called energy - balanced identity , we show the notion of zero - dynamics principle as a guideline for perfect state transfer in general passive linear systems .
then , we prove that the zero - dynamics principle readily leads to the rising exponential function as a unique pulse shape .
further , a useful view of the zero - dynamics principle in terms of the transfer function is provided .
note that the zero - dynamics principle is essentially equivalent to the so - called _ dark state principle _ ; this idea was first employed in @xcite for the application to a lossless node - to - node state transfer in a cavity qed network , and later several applications have been developed , e.g. lossless gate operation @xcite .
appendix b provides a detailed case study comparing the zero - dynamics principle and the dark state principle and then discusses their difference .
first we remark that the pulse shape of the single - photon input field , @xmath50 , and that of the output field , say @xmath146 , can be connected through a dynamical equation having the same form as eq . .
actually , by multiplying @xmath148 by eq .
from the right hand side and using the relation @xmath149 , we find that the mean photon number of the output field is given by @xmath150 by definition this should be written as @xmath151 , hence @xmath50 and @xmath146 are related through the following dynamics : @xmath152 where @xmath153 is a @xmath29-dimensional c - number vector .
note that @xmath153 does not have a particular physical meaning , unlike the vector @xmath154 appearing just below .
the same classical dynamical equation holds for the case of coherent input field ; noting that @xmath155 , we readily see that the vector of mean amplitudes , @xmath156^\top$ ] with @xmath157 , follows @xmath158 where @xmath159 is the amplitude of the output field @xmath6 .
this equation has the same form as eq . ,
hence in what follows we use eq . when discussing the input - output relation of the corresponding wave packets .
to pose the zero - dynamics principle , it is important to first remember that , for the general passive linear system , the following _ energy balance identity _
@xcite holds : @xmath160 this indicates that the total energy contained in the system and the field is preserved for all time . indeed , from eq .
we immediately have @xmath161 where the mean is taken for the state @xmath162 .
now we assume @xmath163 . then , for the energy of the input pulse field to be completely transferred to the system , we need @xmath164 for @xmath165 $ ] with @xmath102 the stopping time of the writing process .
this is a rigorous description , in the case of passive linear systems , of the zero - dynamics principle ; that is , for the general quantum memory problem with a passive system , the output field must be vacuum ( i.e. zero " ) for perfect state transfer .
surprisingly , this simple condition uniquely determines the form of the input pulse shape @xmath50 as shown below .
first , from the requirement @xmath166 , we have @xmath167 , which further leads to @xmath168 this has the solution @xmath169 , with @xmath170 a fixed vector .
thus , again from the condition @xmath171 , we have @xmath172 note that the input is sent during the writing stage @xmath173 , and @xmath174 in the storage and reading stages in @xmath175 . taking this into account
, we end up with the expression @xmath176 where @xmath177 is the heaviside step function taking @xmath178 for @xmath179 and @xmath180 for @xmath181 . since @xmath21 is hurwitz , as assumed in section 2.4 , the real parts of all the eigenvalues of @xmath182 are strictly positive .
. is a generalization of the rising exponential function .
in fact , this immediately recovers the result in the example , where @xmath183 , @xmath110 , and particularly @xmath133 .
lastly we remark that the zero dynamics is given by eq . , which is defined up to time @xmath102 .
first , to see the idea let us return to the example studied in section 3 .
the input pulse shape is given by eq .
, whose laplace transformation is @xmath188=\sqrt{\kappa}/(s-\kappa/2)$ ] .
also in this case the transfer function is given by @xmath189 = 1 - \frac{\kappa}{s+\kappa/2 } = \frac{s-\kappa/2}{s+\kappa/2}.\ ] ] hence , the output is computed as @xmath190 = \frac{s-\kappa/2}{s+\kappa/2}\cdot \frac{\sqrt{\kappa}}{s-\kappa/2 } = \frac{\sqrt{\kappa}}{s+\kappa/2},\ ] ] and its inverse laplace transform then yields @xmath191 thus , we again see that the input field is completely absorbed in the system during @xmath192 ; i.e. the perfect state transfer has been carried out .
the most notable point is clearly that the _ zero _ of @xmath187 $ ] is erased ( in general , if for a transfer function @xmath15 $ ] there exists a @xmath14 satisfying @xmath15=0 $ ] , then @xmath14 is called a zero ) . now we can generalize the above fact ; for simplicity , we set @xmath133
. the transfer function of the general passive linear system is given by eq . .
also the laplace transformation of the rising exponential function is given by @xmath193 = \int_{-\infty}^0 -\eta_1^\top e^{-a^\sharp t } c^\top e^{-st } dt = c(si+a^\dagger)^{-1}\eta_1.\ ] ] therefore , the output is computed as @xmath194 = g[s]\xi[s ] = \big [ 1-c(si - a)^{-1}c^\dagger \big]c(si+a^\dagger)^{-1}\eta_1 \nonumber \\ & & \hspace*{-2.5em } = c(si+a^\dagger)^{-1}\eta_1 - c(si - a)^{-1 } \big [ ( si - a ) - ( si+a^\dagger ) \big](si+a^\dagger)^{-1}\eta_1 \nonumber \\ & & \hspace*{-2.5em } = c(si - a)^{-1}\eta_1 .
\nonumber\end{aligned}\ ] ] since @xmath21 is hurwitz , the output @xmath195 $ ] does not contain a zero , implying @xmath196 . in general , if a transfer function contains a ( transmission ) zero , then there alway exists an input signal such that the corresponding output takes zero @xcite .
therefore , we have the following interpretation of the zero - dynamics principle for quantum memory in terms of transfer function ; in general , a linear memory system needs to have a zero for perfect state transfer , and the input state is sent over an optical field whose pulse shape is characterized by that zero .
this view would be useful particularly in the case of multi input channels .
in this section , we provide a detailed procedure to achieve the perfect memory , which is composed of the following three stages ; the perfect state transfer from the input field to the memory subsystem ( writing ) , decoherence - free preservation of the transferred state ( storage ) , and the appropriate retrieving of the system state into the output field ( reading ) .
the setup was described in section 2.4 ; note again that the system matrices @xmath34 and @xmath36 ( thus @xmath21 ) change depending on the memory stage , but they are piecewise constant . also , as motivated by the result obtained in section 3 , we will take @xmath126 , while keeping general @xmath102 .
one of the main questions is as follows ; although we have derived the rising exponential function from the zero - dynamics principle , it still contains some parameters that should be chosen appropriately ; more precisely , it is given by @xmath197 , and we need to determine @xmath170 so that the input field state is completely transferred to the memory subsystem . in this section , we will see that this synthesis problem is clearly solved .
the solution of the general linear equation is explicitly given by @xmath198 since @xmath21 is hurwitz , we can take the limit @xmath126 , which yields @xmath199 where again @xmath102 is the stopping time of the writing process .
let us now define the following vector of rising exponential functions : @xmath200 then the above solution of @xmath201 can be expressed as @xmath202^\top.\ ] ] this is a vector of creation operators @xmath203 , implying that it has to satisfy the canonical commutation relation @xmath204 ; actually , we have @xmath205 \nu(\tau)^\top ds d\tau = \int_{-\infty}^{\infty } \nu(s)^\sharp \nu(s)^\top ds
\nonumber \\ & & \hspace*{-3em } = e^{a t_1 } \big [ \int_{-\infty}^{t_1 } e^{-as}c{^{\dagger}}c e^{-a{^{\dagger}}s}ds \big]e^{a{^{\dagger}}{t_1 } } = e^{a t_1 } \big [ \int_{-\infty}^{t_1 } \frac{d}{ds}\big ( e^{-as}e^{-a{^{\dagger}}s}\big)ds \big]e^{a{^{\dagger}}{t_1 } } = i. \nonumber\end{aligned}\ ] ] this relation shows that @xmath206 are orthonormal ; @xmath207 .
* case i : single photon state . *
we consider the case where the input is the single photon field state . also the system is assumed to be in the separable ground state at the initial time @xmath38 .
then , through the interaction ( see fig . 2 ( a ) ) the whole state changes to @xmath208 { |{0,\ldots,0}\rangle}_s{|{0}\rangle}_f \nonumber \\ & & \hspace*{-0.05em } = \sum_k s_k u(-\infty , t_1)\big [ i_s\otimes b^*(\gamma_k ) \big ] u^*(-\infty , t_1 ) { |{0,\ldots,0}\rangle}_s{|{0}\rangle}_f , \nonumber\end{aligned}\ ] ] where @xmath125 denotes the unitary time evolution .
we here set the basis functions @xmath68 to the rising exponential functions @xmath209 given in eq .
, meaning that the input pulse shape is chosen as @xmath210 then , noting that @xmath211 , we obtain @xmath212 \otimes { |{0}\rangle}_f,\ ] ] where @xmath213 with @xmath178 appearing only in the @xmath214th entry .
therefore , through the interaction , at time @xmath118 the system completely acquires the input code state with coefficient @xmath215 ; the resulting system state is highly entangled among the nodes ( see fig . 2 ( b ) ) .
the optimal input pulse shape is given by the rising exponential function of the form , as expected in section 4.2 .
but the point here is that we now know that the parameter vector @xmath170 in eq
. exactly corresponds to the superposition coefficients @xmath215 .
together with the structure of the memory subsystem , this fact tells us how we should design the input pulse shape @xmath50 ; this will be more precisely discussed in the next subsection .
* case ii : coherent state .
* next , let us consider the case where the input is the coherent field state .
again the system is in the ground state at @xmath126 .
then , through the interaction the whole state becomes @xmath216 therefore , by setting the basis functions @xmath68 to the rising exponential i.e. @xmath217 and again noting that @xmath218 , we obtain @xmath219 that is , the system state is changed to the product of coherent states @xmath220 .
hence , similar to the single photon case , the perfect state transfer is possible by sending the information @xmath221 over the rising exponential pulse field .
the idea is the same for the coherent input case .
that is , the state @xmath233 can be perfectly transferred and stored in the memory subsystem . the perfect memory procedure for the single photon input state in a 5-nodes passive linear network .
the system can be tuned so that the ( 3 , 4 , 5 ) nodes become decoherence free ; hence these nodes constitute the memory subsystem .
the ( 1 , 2 ) nodes does the buffer subsystem .
( a ) the single photon code state with @xmath227 is sent through the input optical field with pulse shape @xmath124 .
( b ) at time @xmath118 , the perfect state transfer has been completed .
the system is then modulated and the memory subsystem is decoupled .
( c ) the transferred state is preserved during the period @xmath234 $ ] .
( d ) at @xmath235 the memory subsystem is again coupled to the buffer subsystem and thus the optical field .
( e ) the perfect copy appears in the output optical field with pulse shape @xmath236 . ]
suppose that the state has been perfectly stored during the period @xmath234 $ ] ; hence the reading stage starts at time @xmath235 with the initial state @xmath237 for the single photon input case or @xmath238 for the coherent input case ; see fig . 2 ( d ) .
note that , as described in section 5.2 , only some elements of @xmath215 or @xmath221 , which represents the classical information of the stored state , are not zero . to retrieve this initial state , we switch the system matrices so that the memory subsystem again couples to the buffer subsystem and thus the optical field ; in particular , we take the same system matrices @xmath34 and @xmath36 ( and thus @xmath21 ) as in the writing stage . thus note that @xmath21 is hurwitz . to describe the reading stage , first , we particularly focus on the following quantity : @xmath239dt \nonumber \\ & & \hspace*{-3em } = \big [ \int_{t_2}^\infty e^{a^\dagger ( t - t_2 ) } c^\dagger c e^{a ( t - t_2 ) } dt\big ] a(t_2 ) - \int_{t_2}^\infty e^{a^\dagger ( t - t_2 ) } c^\dagger c e^{a t } \big [ \int_{t_2}^t e^{-a s } c^\dagger b(s)ds \big ] dt \nonumber \\ & & \hspace*{1em } + \int_{t_2}^\infty e^{a^\dagger ( t - t_2 ) } c^\dagger b(t ) dt .
\nonumber\end{aligned}\ ] ] the first term is @xmath240 . for the second and the third terms , by defining @xmath241 , which leads to @xmath242 , we find that they become @xmath243 dt = -e^{at_2}k(t_2)=0 . \nonumber\end{aligned}\ ] ] as a result , we have @xmath244 where @xmath245 as in the previous case , @xmath246 are orthonormal ; @xmath247 . note that @xmath236 is a generalization of a _ decaying _ exponential function .
moreover , eq . leads to @xmath248^\top .
\nonumber\end{aligned}\ ] ] * case i : single photon state . *
the initial state is now @xmath237 .
then , through the interaction , this state changes to : @xmath249 \otimes { |{0}\rangle}_f = u(t_2 , \infty ) \big [ \sum_k s_k a_k^*(t_2 ) \big ] { |{0,\ldots,0}\rangle}_s{|{0}\rangle}_f \nonumber \\ & & \hspace*{-0.8em } = u(t_2 , \infty ) \big [ \sum_k s_k a_k^*(t_2 ) \big ] u^*(t_2 , \infty ) { |{0,\ldots,0}\rangle}_s{|{0}\rangle}_f \nonumber \\ & & \hspace*{-0.8em } = \sum_k s_k \big [ i_s\otimes b^*(\tilde{\nu}_k ) \big ] { |{0,\ldots,0}\rangle}_s{|{0}\rangle}_f = { |{0,\ldots,0}\rangle}_s \otimes \sum_k s_k { |{1_{\tilde{\nu}_k}}\rangle}_f .
\nonumber\end{aligned}\ ] ] therefore , certainly the field state recovers the input state , which is now carried by the output field with pulse shape .
the system state returns to the ground state ; see fig . 2 ( d , e ) .
* case ii : coherent state .
* the initial state is @xmath238 .
then , through the interaction , this state changes to : @xmath250 where @xmath251 is a coherent field state with pulse shape @xmath252 thus , similar to the single photon input case , the stored coherent states @xmath253 leaks into the output field with pulse shape , and we can retrieve the full information about @xmath221 contained in the coherent field state @xmath251 .
here we derive the time evolution equations of the statistics in the writing stage .
these equations are useful for numerical simulation , as demonstrated in the next section . * case i : single photon state . * in the case of single photon state , we evaluate the following matrix of operators : @xmath254 [ a_1 , \ldots , a_n].\ ] ] the photon is distributed in the system according to the statistics represented by the correlation matrix @xmath255 .
the time - evolution equation of @xmath256 is , together with the vector @xmath257^\top$ ] , given by @xmath258 the solution of eq .
is readily obtained as @xmath259 where @xmath260 is used . in general ,
the lyapunov differential equation @xmath261 , with @xmath262 time varying , has the solution of the form @xmath263 if @xmath21 is hurwitz , in the limit of @xmath126 , this becomes @xmath264 using this result and the expression , we have @xmath265 if we send the single photon code state over the input field with rising exponential pulse shape @xmath266 , then we have @xmath267 due to @xmath268 .
this means that the input single photon state is distributed among the network so that the @xmath214th node has the mean photon number @xmath269 at time @xmath102 .
* case ii : coherent state . * in this case the statistics is more convenient , because a coherent state is completely charactered only by its mean and variance .
in particular , the dynamics of the mean , @xmath270 , was already obtained in eq . , with @xmath77 particularly given by @xmath271 in eq . .
hence it is immediate to obtain the solution in @xmath173 : @xmath272 hence , by taking the limit @xmath126 , we have @xmath273 ; i.e. @xmath274 .
also , we evaluate the covariance matrix @xmath275 with @xmath276 , which takes zero if and only if the state is a coherent state .
similar to the single photon case , we find that @xmath277 obeys @xmath278 , which readily yields @xmath279 as @xmath126 . as a result
, the @xmath214th node becomes the coherent state @xmath220 at time @xmath118 .
we note that the mean of the output field is @xmath280 for all @xmath173 ; thus the zero - dynamics principle is certainly satisfied .
this section is devoted to study a passive linear network composed of atomic ensembles , which contains a tunable df component .
a numerical simulation will demonstrate how the input field state is transferred to the memory subsystem and how the input pulse shape to be engineered for perfect memory looks like . the passive linear network composed of three large atomic ensembles and a ring cavity .
@xmath281 denotes the cavity mode , and @xmath282 is the annihilation operator approximating the collective lowering operator of the @xmath214th atomic ensemble . ]
we can prove that , when @xmath300 , the matrix @xmath21 is hurwitz ; i.e. the real part of all the eigenvalues of @xmath21 is negative . a convenient way to see
this fact is to use the property that the controllability matrix @xmath301 $ ] is of full rank iff @xmath21 is hurwitz @xcite .
thus the system does not contain a df component when @xmath300 . on the other hand ,
if we turn off the magnetic field and set @xmath302 , then a df subsystem appears , as shown below .
let us take the following unitary matrix : @xmath303.\ ] ] this transforms the system equation to @xmath304 where @xmath305 , \nonumber \\ & & \hspace*{0em } a'=u{^{\dagger}}au = \left [ \begin{array}{cccc } -\kappa/2 & \sqrt{3}g & 0 & 0 \\
-\sqrt{3}g & 0 & -\sqrt{2}i\delta/2 & \sqrt{6}i\delta/6 \\ 0 & -\sqrt{2}i\delta/2 & -i\delta/2 & -\sqrt{3}i\delta/6 \\ 0 & \sqrt{6}i\delta/6 & -\sqrt{3}i\delta/6 & i\delta/2 \\ \end{array } \right ] , \nonumber \\ & & \hspace*{0em } c ' = cu = [ \sqrt{\kappa},~0,~0,~0].\end{aligned}\ ] ] therefore , when @xmath302 , the system takes a form of eq . .
that is , @xmath306^\top$ ] is not affected by the incoming field @xmath5 and it does not appear in the output field @xmath6 ; hence @xmath306^\top$ ] is the memory subsystem that can be switched to a df or non - df subsystem , just by controlling the external magnetic field .
this two - mode subsystem works as a perfect memory that preserves any state of the form @xmath307 in the case of single photon state or @xmath308 in the case of coherent state .
note that @xmath309 and @xmath310 depend on the atomic modes @xmath311 and not on the cavity mode @xmath281 , implying that the state is indeed stored in the atomic ensembles .
also it should be remarked that @xmath309 and @xmath310 take the form of continuous - variable _ syndromes _ used for quantum error correction @xcite .
here we describe the concrete procedure of the writing , storage , and reading processes , in the case of single photon input ; see fig .
[ fig example ] . *
a single photon field state is prepared in the form @xmath312 , where @xmath313 and @xmath314 are the third and fourth elements of the vector of rising exponential functions @xmath315 with @xmath316 and @xmath317 given in eq . .
note in this stage the magnetic field is on ; @xmath300 . * the field couples to the system until @xmath173 .
the perfect state transfer is achieved in the end , at @xmath118 , by sending the input state over the optical field with pulse shape @xmath318 described above . the whole state changes to @xmath319 .
* we turn off the magnetic field and set @xmath302 ; then the memory subsystem with modes @xmath320 becomes decoherence free and its state @xmath321 is preserved during an arbitrary time interval @xmath234 $ ] .
* at a later time @xmath322 , we turn on the magnetic field ( i.e. set @xmath300 ) to retrieve the stored state
. then the memory subsystem again couples to the optical field , and the perfect copy @xmath323 appears in the output field with the pulse shape specified by @xmath324 . the memory procedure for the passive linear network composed of three atomic ensembles trapped in a single - mode cavity .
the number @xmath325 indicates the subsystem with mode @xmath326 .
( a ) the single photon state is sent through the input optical field with pulse shape @xmath318 , where in this stage the magnetic field is turned on ( @xmath300 ) .
( b ) at time @xmath118 the system acquires the state @xmath307 ; that is , the input state is perfectly transferred into the 3rd and 4th nodes .
( c ) then the magnetic field is turned off ( @xmath302 ) so that the memory subsystem with modes @xmath327 is decoupled from the buffer subsystem with modes @xmath328 and the input - output optical field ; hence it becomes decoherence free and the transferred state is perfectly preserved .
( d ) at @xmath235 we again set @xmath300 .
the memory subsystem again couples to the buffer subsystem and the optical field .
( e ) the perfect copy appears in the output field with the pulse shape @xmath329 . ]
recall that the optimal input pulse shape is determined by the properties ( zeros ) of the memory system and this corresponds to the impedance matching mentioned in section 3.2 .
now we should note that an additional matching condition is not imposed on the interaction between the cavity mode and the atomic ensembles , although perfect state transfer from the former to the latter is certainly achieved .
this result seems to be inconsistent with the fact obtained in @xcite , showing that perfect state transfer from a cavity to an inhomogeneously broadened ( ib ) atomic ensemble requires a strict impedance matching between them .
but there is a clear difference between our case and those studies ; in the case dealing with the ib ensemble , due to the matching condition , an input field state with _ arbitrary _ ( yet within a finite band - width ) temporal shape is allowed to be completely absorbed into the ensemble ( see e.g. @xcite for the recent experimental results ) , while in our case the optimal pulse shape has to be strictly specified .
exploring a combined schematic of these two memory procedures , which would allow weaker pulse shaping and weaker impedance matching , should be an interesting future work .
( a ) time evolutions of the absolute value of @xmath313 ( red ) and @xmath314 ( green ) .
( b ) time evolution of the mean photon number at the @xmath330th nodes , @xmath331 .
the blue , black , red , and green lines represent the time evolutions of @xmath332 , and @xmath333 , respectively . ] here we demonstrate a numerical simulation of the writing stage of the above memory procedure .
the parameters are set to @xmath334 , @xmath335 , and @xmath336 ; note again in this stage @xmath300 and there is no df subsystem .
the input is a single photon field state with coefficients @xmath337 , which is carried by the optical field with pulse shape @xmath313 and @xmath338 as mentioned above .
the initial time is @xmath339 and the stopping time is @xmath133 .
first , fig .
[ network simulation ] ( a ) shows the absolute value of @xmath313 and @xmath338 .
these are the pulse shapes we need to correctly engineer for the desirable perfect state transfer .
a notable point is that they are not anymore of a rising exponential shape such as eq . ; particularly they take the value zero at the stopping time @xmath133 .
a similar non - rising exponential pulse function was also found in @xcite , achieving perfect state transfer in an integrated quantum memory system .
it looks that we can realize this kind of pulse shape by combining some gaussian wave packets , which might be a desirable feature from the engineering viewpoint .
next , fig .
[ network simulation ] ( b ) shows the time - evolutions of the mean photon number of each node , i.e. @xmath340 , which can be computed by numerically solving eqs . and .
as expected from the theory , the memory subsystem with modes @xmath327 perfectly acquires the photon with mean photon number @xmath341 at @xmath133 .
we should note that the transportation of the photon from the input field to the memory subsystem occurs rapidly only in the last few period ; in fact , almost all the energy contained in the input pulses @xmath313 and @xmath314 is confined in this short period .
hence , we need to be very careful to stop the writing process at the accurate time @xmath133 , because the desired state @xmath342 is fragile in the following sense . for instance
if we turn off the magnetic field a bit earlier than @xmath133 , say @xmath343 , then the whole system s state generated is roughly @xmath344 ( unnormalized ) ; thus the state of the memory subsystem becomes a mixed state ( unnormalized ) @xmath345 due to the decoherence added to the buffer subsystem with modes @xmath328 during the storage period .
hence , an important future work is to find a suitable set of parameters @xmath346 so that the time - evolutions of the mean photon numbers of the memory subsystem become as flat as possible at the stopping time @xmath102 .
in this paper , for a general passive linear system , we have provided a designing method of input pulse shape that perfectly transports a single photon or coherent field state to a memory subsystem , which can be switched to a df subsystem .
the method is general and simple , so it can be directly applied to a large - scale network ; in fact , in the example studied in section 7 , we found that the explicit form of @xmath313 and @xmath314 are readily obtained .
the results are based on the zero - dynamics principle .
although in this paper this principle was used only for synthesizing the input pulse shape , it is indeed a wide concept that works in a more general situation .
for example , the zero - dynamics principle can be applied to the case where , instead of pulse shaping of the input field , some time - varying controllable parameters of the system should be engineered due to practical limitation ; also the system can be nonlinear ; further , we could deal with an inhomogeneously broadened atomic ensemble memory that allows an arbitrary temporal shape for perfect state transfer , which was discussed in section 7.2 . in any case , following the zero - dynamics principle , we should design the system so that the output is zero or more generally the output is minimized .
moreover , the zero - dynamics corresponds to the time - evolution of a state free from any energy loss , thus it represents a coherent , yet non - unitary , gate operation on the system state for quantum information processing ; that is , designing a desired manipulation of a state in an open system is no more than designing a desired zero - dynamics .
all these problems will be addressed in future works .
the authors acknowledges helpful discussions with m. r. hush and a. r. r. carvalho .
this research was supported by the australian research council centre of excellence for quantum computation and communication technology ( project number ce110001027 ) , and afosr grant fa2386 - 12 - 1 - 4075 .
also ny s work was supported by jsps grant - in - aid number 40513289 .
in this paper , we thoroughly study a passive linear system , but there are many systems containing an _ active _ component . in general , for such an active system the energy balance identity does not hold , hence the zero - dynamics principle does not anymore mean the perfect energy transfer . hence , it should be worth doing a case study to see if an active system could allow perfect state transfer .
let us consider the following active system : @xmath347 = -\frac{1}{2}\left [ \begin{array}{cc } \kappa & -\epsilon \\
-\epsilon & \kappa \\ \end{array } \right ] \left [ \begin{array}{c } a \\
a^ * \\
\end{array } \right ] -\sqrt{\kappa } \left [ \begin{array}{c } b \\
b^ * \\
\end{array } \right].\ ] ] in optics , this represents the dynamics of an optical parametric oscillator , where @xmath348 denotes the squeezing strength @xcite . note that the system becomes passive if @xmath349 .
the above equation can be explicitly solved : @xmath350 \nonumber \\ & & \hspace*{-3em } \mbox { } - \sqrt{\kappa } \int_{t_0}^{t_1 } e^{-\kappa(t_1-s)/2 } \big [ \sinh(\epsilon(t_1-s)/2 ) b(s)ds + \cosh(\epsilon(t_1-t_0)/2 ) b^*(s)ds \big ] .
\nonumber\end{aligned}\ ] ] unlike the passive case , the field annihilation operator @xmath351 appears in the equation . then under the same setting taken in section 3 where the input field state is given by a superposition of the vacuum and @xmath352 with the pulse shape function @xmath353 given below ,
we obtain ( @xmath126 and @xmath133 ) @xmath354 \otimes { |{0}\rangle}_f - \frac{\beta \epsilon}{\sqrt{2\kappa^2-\epsilon^2 } } u(t_0 , t_1 ) b(\xi_2 ) u^*(t_0 , t_1 ) { |{0}\rangle}_s{|{0}\rangle}_f , \nonumber\end{aligned}\ ] ] where @xmath355 this equation implies that , when @xmath356 , the perfect state transfer is impossible due to the third term , which clearly stems from the active element of the system . to carry out efficient state transfer ,
we need some approximation ; in the above case , if @xmath112 is much bigger than @xmath348 , then the system state becomes approximately the desired one to be stored .
another example is found in @xcite , where the system is an atomic ensemble containing an active component , but by introducing a fast oscillating magnetic field it is approximated by a passive one , which was further shown to be a perfect memory .
the basic idea of dark state principle is as follows . for a system coupled to a probe field
, we continuously monitor the system by a photo detector measuring the output field ; then if the detector counts no photon , this means that the system is in a _ dark state _ and has a time evolution without loss of energy .
here we apply this dark state principle to the writing problem discussed in section 3 and derive the same result ; that is , in this sense , the zero - dynamics principle and the dark state principle are equivalent , though there is a big difference in practice as shown below .
first let us consider the case where we want to send a coherent field state to the system .
in general , if we use a photon counter to estimate the system observables , our state ( knowledge ) conditioned on the measurement results is updated by the following stochastic master equation @xcite ( the scattering operator is now set to be the identity ) : @xmath357\alpha + [ l , \rho]\alpha^*\big)dt + \big [ \frac{1}{\cal n } ( l\rho l^ * + \alpha^*l\rho + \alpha \rho l^ * + |\alpha|^2\rho ) - \rho \big ] ( dy - { \cal n}dt),\ ] ] where @xmath358 + l\rho l^ * - l^ * l\rho/2 - \rho l^ * l/2,~~ { \cal n } = \tr\big [ \rho\big(l^*l + \alpha^*l + \alpha l^ * + |\alpha|^2\big ) \big].\ ] ] @xmath359 is the pulse shape of the input coherent light field and @xmath360 is the measurement result ( 0 or 1 ) obtained during the small time interval @xmath361 .
also @xmath362 and @xmath363 are the system operators .
now since the ensemble averaging over the measurement results leads to a standard master equation , we have @xmath364 .
then , the counting probability of the measurement result 1 " during @xmath361 is given by @xmath365 . hence if @xmath366 , the system is in a dark state and loses no energy into the output field ; the state satisfying this condition is called the dark state . in our case where the system is the single - mode passive linear system with @xmath367 and @xmath368 , the dark state can be specified to a coherent state @xmath369 because we now know that the system s state is always a coherent state .
the condition @xmath370 then becomes @xmath371 , which yields @xmath372 .
now , under the dark state condition the time evolution of the conditional state is identical to that of the averaged one , which consequently leads to @xmath373 .
these two equations yield @xmath374 , thus the input pulse shape must be a rising exponential function @xmath375 .
next let us consider the case where the input is a single photon field state .
as in the above case , the dark state principle is represented in terms of the conditional state subjected to the single - photon stochastic master equation ; see eq .
( 43 ) in @xcite . in this case
the probability to obtain the measurement result 1 " during @xmath361 is given by @xmath376 where @xmath50 is the temporal pulse shape of the single photon field .
@xmath377 are the operators characterizing the conditional state . under the dark state condition @xmath370 , they obey @xmath378\xi + [ l , \rho^{10}]\xi^*,~~ \dot{\rho}^{10 } = { \cal l}\rho^{10 } + [ \rho^{00 } , l^*]\xi,~~ \dot{\rho}^{00 } = { \cal l}\rho^{00},\ ] ] and @xmath379 , which are identical to the single - photon master equation @xcite . the initial conditions are @xmath380 and @xmath381 . in our case @xmath367 and
@xmath368 , these differential equations can be explicitly solved , yielding @xmath382 , @xmath383 , and @xmath384 , where @xmath27 and @xmath385 satisfy @xmath386 and @xmath387 . substituting these solutions @xmath377 for the dark state condition @xmath370
, we have @xmath388 . combining these three equations
, we end up with the relation @xmath389 and thus see that the input pulse shape has to be a rising exponential function @xmath390 . summarizing
, we have recovered the same result obtained in section 3 , showing that the dark state principle is equivalent to the zero - dynamics principle .
both principles require no energy leaking from the system into the output field , but their approaches are different ; the dark state principle is represented in the schrdinger picture , while the heisenberg picture is used to describe the zero - dynamics principle . as a result ,
in the former case we need to solve the master equation , which is sometimes a hard task as demonstrated above especially in the single photon field case .
on the other hand , we have seen in section 4.2 that the zero - dynamics principle allows us to derive the rising exponential function very easily , even in the general setup ; also the transfer - function - based treatment of the principle is notable and possibly very useful from the viewpoint of the applicability of the linear response theory to quantum memory .
of course these special advantages appear particularly in the linear case , and for more general nonlinear memory systems we should be careful in choosing the approach .
10 t. chaneliere , d. n. matsukevich , s. d. jenkins , s .- y .
lan , t. a. b. kennedy , and a. kuzmich , storage and retrieval of single photons transmitted between remote quantum memories , nature * 438 * , 833/836 , 2005 .
d. a. lidar and k. b. whaley , decoherence - free subspaces and subsystems , irreversible quantum dynamics , edited by f. benatti and r. floreanini , lecture notes in physics * 622 * , 83 , springer berlin / heidelberg , 2003 .
a. v. gorshkov , a. andre , m. d. lukin , and a. s. sorensen , photon storage in lambda - type optically dense atomic media , i. cavity model , phys .
a * 76 * , 033804 , 2007 ; ii .
free - space model , phys .
a * 76 * , 033805 , 2007 ; iii .
effects of inhomogeneous broadening , phys .
a * 76 * , 033806 , 2007 . | in this paper , we study a general linear networked system that contains a tunable memory subsystem ; that is , it is decoupled from an optical field for state transportation during the storage process , while it couples to the field during the writing or reading process .
the input is given by a single photon state or a coherent state in a pulsed light field .
we then completely and explicitly characterize the condition required on the pulse shape achieving the perfect state transfer from the light field to the memory subsystem .
the key idea to obtain this result is the use of zero - dynamics principle , which in our case means that , for perfect state transfer , the output field during the writing process must be a vacuum .
a useful interpretation of the result in terms of the transfer function is also given .
moreover , a four - nodes network composed of atomic ensembles is studied as an example , demonstrating how the input field state is transferred to the memory subsystem and how the input pulse shape to be engineered for perfect memory looks like . |
g2 , a recently discovered spatially - extended red source , is on a nearly radial orbit headed towards the @xmath1 ( @xmath2 ) , supermassive black hole at the galactic center , sgr a * @xcite . as g2 plunges towards sgr a * at a supersonic speed
, it will drive a bow shock into the hot interstellar medium ( ism ) .
as electrons cross the forward shock , they will be accelerated and emit synchrotron radiation . the radio synchrotron emission from the forward shock
has been predicted to be observable by @xcite .
@xcite predicted that the radio emission should peak 7 to 9 months before the center of mass crosses periapsis in spring of 2014 , reaching a distance of a mere @xmath3 away from sgr a * @xcite .
this close encounter may give a unique opportunity to probe the accretion disk of sgr a * and the ism near periapsis by measuring the radio flux of the forward bow shock . the radio flux from the forward shock has yet to be observed ; depending on the orbital parameters of g2
, it may not be expected to peak until late summer or early autumn of 2013 , although perhaps @xmath4 of g2 has already passed periapsis @xcite . in this letter , we show that the expected radio flux from the forward shock is model dependent , suggesting an intriguing possibility : the radio flux from the forward shock may reveal the nature of g2 .
the nature of g2 is undetermined .
when first discovered , @xcite hypothesized that g2 was a pressure - confined , non - self - gravitating gas cloud , due to the fact that the brackett - gamma ( br-@xmath5 ) luminosity of g2 is not changing with time , @xmath6 , and the br-@xmath5 velocity dispersion is increasing in a manner that is well fit by a gas cloud with a radius of @xmath7 being tidally sheared by sgr a*. the compactness of g2 necessitates that this gas cloud must have formed shortly before g2 was first discovered in 2002 .
however , there is not a clear source of a gas cloud in the region as the ism at a distance @xmath8 from sgr a * is not susceptible to thermal instabilities .
one possible source of a gas cloud is colliding stellar winds @xcite .
alternatively , there is another class of models where g2 contains a very faint stellar core that emits gas as it falls towards sgr a * @xcite .
the ionized gas is then tidally sheared and is the source of the br-@xmath5 radiation seen as g2 . in @xcite , the ionized gas that is the source of the br-@xmath5 radiation is located in the cold dense inner shock of a momentum - supported bow shock between a stellar wind from a hidden , young star and the hot ism .
@xcite performed hydrodynamical simulations in which they evolved the stellar wind shock in the gravitational potential of sgr a * , and they found that the @xmath9 dispersion observations are well - matched by a star with a mass - loss rate of @xmath10 and with a wind speed of @xmath11 .
the ionization source of the inner shock in this model is not entirely clear , as the predicted br-@xmath5 radiation does not perfectly match the observations .
@xcite proposed that hydrogen is collisionally ionized in the inner shock of the stellar wind , although they overestimated the ionization ability of the wind and underestimated the ionizing background from the o stars in the galactic center .
@xcite just assumed the entire inner shock was ionized , which leads to the br-@xmath5 flux increasing as g2 approaches pericenter , which is not what has been observed .
future work will need to be done to understand the radiation mechanism in the stellar wind model of g2 .
if g2 is a pressure - confined diffuse cloud , then it will likely be destroyed during periapsis passage . if g2 is the inner bow shock of a star it will likely survive @xcite . additionally , if g2 is a cloud , how its size changes as it approaches sgr a * is not clear while the stellar wind model of g2 makes a strong prediction about how the size of the forward bow shock evolves as g2 passes through periapsis . in the cloud model , as g2 heads closer to sgr a * it will be stretched in the longitudinal direction , compressed in the transverse direction by tidal forces , and compressed further by the increasing pressure of the ism . in the 3-d simulation by @xcite , for an isothermal cloud that formed in 1995.5 with a radius of @xmath12 , they find that the cross sectional area of g2 one year before periapsis shrunk by a factor of 4 to @xmath13 . it appears to be the same size at periapsis .
the cloud is in the process of being completely disrupted , so it is difficult to tell . in this letter
, we use the term `` cloud model '' to mean a model where the cross section of g2 is approximately constant as it passes through periapsis as in @xcite . in the stellar wind model ,
far away from sgr a * the size of the shock is approximately the same as in the cloud model , but when g2 gets close to periapsis , its size shrinks significantly , inversely proportional to the increase of the pressure of the ism . in other words ,
the area is proportional to the distance from sgr a * squared ( see eq [ eq : areaest ] ) .
the hydrodynamic simulations in @xcite show a similar decrease to the shocked wind area .
the decrease in area at periapsis drastically reduces the expected radio flux .
g2 appears to be an extended object with the compact head structure called g2 , and a less dense and larger tail - like structure , g2 t @xcite .
this paper focuses solely on the over - dense head , g2 , as the origin of g2 t is not well known .
if g2 is a diffuse cloud then g2 t is either material that was stripped from g2 as it fell towards sgr a * , or it was somehow formed through a similar process as g2 . if g2 has a stellar core emitting a wind , the tail may have been from the outflow of the star when it was closer to apocenter . in this letter
, we extend the work of @xcite to the stellar wind model of g2 .
we apply @xcite analytic solution of the momentum - supported bow shock model proposed by @xcite to estimate the size of the bow shock of g2 .
analytical calculations are better suited to estimate the forward shock size than a hydrodynamic simulation such as @xcite because of stability issues in the simulation .
the ism surrounding sgr a * is convectively unstable .
previous simulations ( with the exception of * ? ? ?
* ) dealt with this issue by evolving a passive tracer field along with g2 and resetting the ism to its equilibrium value whenever a cell did not have a sufficient ratio of tracer particles to ism particles @xcite . as noted in @xcite ,
this stabilization technique suppresses the forward shock in the ism , making these simulations incapable of properly resolving the forward shock .
analytical calculations also allow us to predict how the radio flux depends on the undetermined parameters of the stellar wind .
we use the geometry of the bow shock to calculate the expected synchrotron flux in the radio band of both the outer forward shock and the inner termination shock for the stellar wind model of g2 .
the predicted flux of the forward shock is roughly two orders of magnitude less than the flux predicted by @xcite , and lies an order of magnitude below the quiescent radio emission of sgr a * at 2 ghz . if the stellar wind model is correct , g2 will likely not be observable in radio frequencies .
therefore , a radio detection of g2 will shed light on the nature of g2 well before periapsis passage . in section
[ sec : methods ] of this letter , we briefly describe the model we used for the ism at the galactic center as well as the geometry of the bow shock . in section [ sec : results ] we show how the expected peak synchrotron flux at 1.4 ghz and spectrum depend on the nature of g2 . in section [ sec : s2 ] we extend our results to the star s2 .
finally , in section [ sec : conclusions ] we briefly discuss our findings .
for the hot ism environment surrounding sgr a * , we adopt the same dependence of density and temperature on distance @xmath14 from sgr a * as those used by @xcite , who used the model of @xcite . however , the most recent _ chandra _ x - ray observations of the galactic centre suggest that the radial density profile may be flatter than the one used by schartmann et al .
@xcite @xmath15 @xmath16 throughout this paper the convention @xmath17 is used , and unless otherwise noted all units are cgs . to calculate the distance and velocity of g2 , we use the orbital parameters derived from the br-@xmath5 observations given in @xcite .
however , for the purposes of illustration g2 s velocity is well matched by a free fall approximation , _
i.e. _ @xmath18 .
the mach number is approximately @xmath19 . because of the approximation we made to the velocity
, the mach number has no time dependence , but it will change with time when using the proper elliptical motion of the orbit . to calculate the radio flux from the forward bow shock of g2 , we need to determine the area of its cylindrical cross section .
we use the same parameters for the isothermal stellar wind as in @xcite : a star with a mass - loss rate of @xmath20 , a constant wind speed of @xmath21 , and a temperature of @xmath22 .
we include the dependence of our results on the stellar wind parameters as they are not precisely constrained by observations .
the wind is supersonic , with a mach number of @xmath23 . to calculate the properties of a stellar wind bow shock
, we used the equations given in @xcite instead of @xcite because @xcite correctly accounts for the centrifugal force due to the fluid moving in a curved path inside of the shock .
the effect of the centrifugal force is to produce a larger bow shock , _ e.g. _
@xcite , than the solution given in @xcite ( see figure [ fig : area ] for a comparison ) .
the stagnation radius , @xmath24 , is the same in both solutions and it occurs where the two ram pressures equal one another , _
i.e. _ @xmath25 . for the adopted parameters of @xmath26 and @xmath27 ,
@xmath28 the shock is axisymmetric about the stellar velocity vector , and the distance of the bow - shock surface from the star depends on the azimuthal angle measured from the stagnation point , @xmath29 , and can be written as @xmath30 where @xmath31 is given as an analytical function in @xcite , @xmath32 the velocity of the gas inside of the bow shock is tangential to the shock surface and equal to @xmath33 the strength of the outer shock depends on the angle @xmath29 .
the shock will terminate at an angle , @xmath34 , where the normal component of the velocity of the ism with respect to the star , @xmath35 , is equal to the sound speed of the ism .
as in @xcite , it is useful to define the following angles : @xmath36 , and @xmath37 , so that @xmath38
. then @xmath39 is found by solving @xmath40 . when @xmath41 , @xmath42 and @xmath43 .
the cylindrical cross - section area of the forward shock is @xmath44 @xmath45 has some degree of uncertainty from modeling the galactic center , and the fact that the inclination of g2 s trajectory with respect to a possible accretion disk of sgr a * is unknown .
the change in the area can be estimated using @xmath46 , when @xmath47 .
the cross - sectional area versus time is plotted in figure [ fig : area ] .
the area in the wind models decreases by @xmath48 orders of magnitude as g2 approaches periapsis .
synchrotron flux is expected to peak at periapsis where the magnetic field is strongest .
the decrease in the area of the forward shock as g2 approaches periapsis is responsible for the large difference in flux for the pressure - confined cloud and stellar wind models . , the cross - sectional area of the inner shock of g2 , which agrees very well with the simulation by @xcite .
the thick red line is the area used in @xcite , @xmath49 cm . ]
to calculate the expected synchrotron emission of the forward shock of g2 we extend the methodology of @xcite to the stellar wind model of g2 , who used particle - in - cell simulations of low mach number shocks to estimate the electron population accelerated by g2 . as in @xcite , we assume that a fraction @xmath50 of the electrons swept into the forward shock are accelerated to a high - energy power law spectrum with a slope of @xmath51 and that has a minimum lorentz factor @xmath52 , where @xmath53 and @xmath54 is the preshock temperature . @xmath55 and @xmath56 were empirically determined by @xcite with simulations .
they found @xmath55 and @xmath56 to be nearly independent of @xmath54 and @xmath45 .
it may be counter - intuitive that the injection energy of the power law spectrum depends on the preshocked rather than shocked temperature , but because of the very narrow range of @xmath45 applicable to g2 and considered by @xcite , _
i.e. _ @xmath57 , it makes little difference which temperature one chooses ( as long as @xmath56 is adjusted accordingly ) .
the energy distribution of the rate at which electrons enter the shock is given by @xcite , @xmath58 for @xmath59 , and where @xmath60 , @xmath61 , @xmath62 , and @xmath54 are all functions of time .
to calculate the expected radio flux , we assume that @xmath63 , where @xmath64 is the pressure of the unshocked ism to calculate the unshocked magnetic field .
the shocked magnetic field is calculated assuming only shock compression with no additional amplification of the magnetic field , as expected for small mach number shocks .
we use the rankine - hugoniot jump conditions of an adiabatic shock to find the shocked magnetic field from @xcite , @xmath65 when @xmath66 , @xmath67 . as an upper limit on @xmath68 ,
we make sure that the shocked magnetic pressure does not exceed the ram pressure of the ism , this is true for all mach numbers if @xmath69 .
we take @xmath70 , which corresponds to the trajectory with largest flux at 1.4 ghz in @xcite .
the synchrotron specific luminosity at the frequency @xmath71 is @xmath72 where @xmath73 is the number of electrons with @xmath74 at time @xmath75 .
the observed flux is calculated using a distance to the galactic center of @xmath76 : @xmath77 to calculate the flux , we extend the two scenarios defined in @xcite to a bow shock with changing area which starts accelerating electrons at sometime @xmath78 and the forward shock reaches periapsis at @xmath79 .
the two models are ( 1 ) the plow model , where every electron that enters the shock stays in the shocked area and radiates in the shocked magnetic field , and ( 2 ) the local model , where the electrons are energized by the shock but then they quickly exit the shock and radiate _ in situ _ in the unshocked local magnetic field . in the plow model , the flux at 1.4 ghz is calculated from eq ( [ eq : flux ] ) using the area from eq ( [ eq : area ] ) and the shocked magnetic field , eq ( [ eq : bshocked ] ) . when @xmath80 , @xmath81 is @xmath82 where @xmath60 is the instantaneous bow shock area at time @xmath75 .
when @xmath83 , @xmath81 is @xmath84 in the local model , the flux is calculated with the unshocked magnetic field @xmath85 the light curves of the flux at 1.4 ghz in the stellar wind model of g2 are shown near pericenter passage of the forward shock in the inset of figure [ fig : lightcurves ] . the light curves in the stellar wind model peak shortly after @xmath79 as in the cloud model of @xcite . to compare the expected fluxes from the stellar wind and cloud models , we also calculate the expected flux using a constant area of @xmath86 in equations ( [ eq : cbefore ] ) and ( [ eq : goodlocal ] ) .
the light curves close to pericenter of both the cloud and stellar wind models are plotted in figure [ fig : lightcurves ] . to see how our results depend on our parameter choices , we estimate @xmath73 in both the plow and local models by multiplying @xmath87 by the dynamical time ( @xmath88 ) and using the area from eq ( [ eq : areaest ] ) . @xmath89
then the flux at 1.4 ghz when @xmath80 is estimated using eq ( [ eq : flux ] ) with magnetic fields from eq ( [ eq : b ] ) and ( [ eq : bshocked ] ) for the local and plow model respectively @xmath90 the analytical results given by equations ( [ eq : localest ] ) and ( [ eq : plowest ] ) are found to be within a factor of 2 from the numerical results in figure [ fig : lightcurves ] . , for different models of g2 .
solid lines correspond to the plow model , where all of the accelerated electrons stay inside of the shock .
the dashed lines correspond to the local model , where the electrons quickly leave the bow shock region after being accelerated .
thick red lines correspond to when g2 is a cloud with an area of @xmath91 as in @xcite .
the lower blue lines represent when g2 is a stellar wind with an area calculated using eq ( [ eq : area ] ) .
the dotted blue line is the flux from electrons residing in the inner shock of the stellar wind assuming a steady state solution .
_ inset _ : a plot showing the flux at 1.4 ghz in jy for a stellar wind model of g2 in a linear scale . ] in the stellar wind model of g2 , electrons will be accelerated at both the outer , forward shock traveling into the ism as well as at the inner , termination shock of the stellar wind . for the inner shock to radiate at 1.4 ghz
, it must be able to accelerate electrons from a upstream temperature of @xmath92 ev to @xmath93 ev .
it is not clear whether a non - relativistic shock with @xmath94 will be capable of doing this efficiently .
therefore , our estimation of the radio flux from the inner shock in this section should be taken as an upper limit .
we find the flux at 1.4 ghz lies four orders of magnitude below the fluxes calculated in the previous section .
therefore the inner shock will not contribute to the radio synchrotron flux of g2 .
the stellar wind at the bow shock is much more dense than the ism , its density equal to @xmath95 the shocked gas will radiate efficiently and cool quickly , and therefore the shock can be taken to be isothermal .
so @xmath96 will be boosted by a factor @xmath97 : @xmath98 assuming that in the shocked wind @xmath99 , @xmath100 , the magnetic field inside of the inner shock is @xmath101 we calculate the volume of the inner shock by assuming that it is in steady state . in a steady state ,
the thickness of the shock , @xmath102 , is found by numerically solving the continuity equation for @xmath102 as a function of @xmath29 , @xmath103 @xmath102 is estimated as being constant in @xmath29 and equal to @xmath104 : @xmath105 we calculate the volume of the inner shock , @xmath106 , accounting for the geometry of the bow shock and solving eq ( [ eq : continuityeq ] ) for @xmath102 . approximately , @xmath107 .
using @xmath106 we calculate @xmath108 , the number of electrons in the inner shock with @xmath74 : @xmath109 the flux is estimated using eq ( [ eq : flux ] ) : @xmath110 using eq ( [ eq : c_inner ] ) , ( [ eq : bboosted ] ) , ( [ eq : power ] ) and ( [ eq : flux ] ) , we calculate the expected radio flux for the inner shock at @xmath111 , and plot it in figure [ fig : lightcurves ] .
our estimate underpredicts the more accurate calculation of the inner shock radio flux at periapsis by a factor @xmath112 .
the radio flux from the inner shock lies well below the forward shock and can be ignored . , where @xmath79 is the periapsis crossing time of the forward shock .
the color scheme is the same as figure [ fig : lightcurves ] , where solid lines correspond to the plow model , and dashed lines to the local model .
thick red lines represent the spectra if g2 is a cloud , while the lower blue lines are if g2 is a stellar wind .
the data points are radio fluxes measured during periods of inactivity of sgr a * by @xcite . ] in figure [ fig : spectra ] we show the spectra expected from forward shock at 0.05 years after periapsis of the forward shock .
to calculate the synchrotron self absorption frequency in the plowing models , we simply use the surface area of the forward shock at time @xmath113 years , accounting for the geometry of the bow shock in the stellar wind model . in the local models
, we use the same approach as @xcite , where synchrotron self absorption frequency and @xmath114 are calculated separately for each time step @xmath115 up until @xmath113 yr , and then the fluxes are summed .
the surface area of the radiation in the local model is @xmath116 , where @xmath117 cm in the cloud model and is @xmath118 in the stellar wind model .
the radio flux in the stellar wind model lies below the quiescent flux of sgr a * at all frequencies and therefore will likely not be observable .
there is a cluster of young stars at the galactic center , and one of the brightest of these stars s2 , a b type star that has an orbital period of 15.5 years , reaches nearly twice as close to sgr a * as g2 ( a distance of @xmath119 ) @xcite .
s2 is expected to have a strong wind which will form a momentum - supported bow shock similar to g2 in the stellar wind model .
the x - ray radiation from the inner shock was investigated by @xcite , who estimated the s2 s wind has a mass - loss rate @xmath120 and a velocity @xmath121 km / s . using these parameters , we estimate peak radio flux at periapsis from the electrons accelerated by the outer shock of s2 using eq ( [ eq : plowest ] ) .
we estimate that the peak flux from the forward shock of s2 at 1.4 ghz is @xmath122 jy and @xmath123 jy at 14 ghz ( assuming @xmath51 ) .
the last pericenter crossing of s2 was in 2002.3 and there is good data of sgr a * around this time .
the flux at 15 ghz varies from @xmath1240.81.1 jy and the flux at 23 ghz varies from @xmath1240.81.2 jy @xcite .
while there is variability in the radio flux before and after the pericenter crossing of s2 , there is not a flare that lasts for a long time or is spectrally consistent with synchrotron radiation from s2 . throughout the pericenter crossing time the flux at 23
ghz is equal to or larger than the flux at 15 ghz , while the opposite would be expected from the synchrotron radiation of an electron power - law distribution with @xmath51 .
thus it is likely this variation is due to intrinsic variation of sgr a * and not due to s2 .
the lack of detection of the forward shock of s2 in 2002.3 gives us additional confidence in our prediction that the forward - shock of g2 will not be observable in the radio band .
in this paper we calculated the expected radio flux from the forward shock and reverse shock of g2 assuming that g2 is the inner bow shock of a stellar wind as suggested by @xcite .
we used an analytic solution of a momentum - supported bow shock to calculate the geometry of the forward shock and provide analytical estimates of the flux at 1.4 ghz at pericenter passage of the forward and reverse shock to show how the flux depends on the undetermined parameters of the stellar wind .
if g2 is the inner shock of a stellar wind , its forward shock cylindrical cross section will decrease as g2 approaches sgr a * and the ram pressure increases , driving the stagnation point to a smaller distance from the star .
the decrease in area by over two orders of magnitude of results in a similar decrease in the radio flux expected at periapsis , falling well below the previous estimates of @xmath1241 - 20 jy predicted by @xcite , where they assumed that g2 was a gas cloud that formed in pressure equilibrium and that the cross section area stayed constant .
if g2 is the inner bow shock from a stellar wind , the radio flux from the forward shock lies below the quiescent radio emission of sgr a * at all frequencies , so it will be difficult to detect .
if g2 is a pressure - supported gas cloud , it will likely be destroyed during pericenter passage ; on the other hand g2 will survive if it is the inner bow shock of a stellar wind . therefore whether or not g2 survives will be important in determining the makeup of g2 .
however , the radio emission will peak when the forward shock crosses periapsis , which in the cloud model is predicted to happen 7 to 9 months before g2 passes through periapsis .
we have shown that the stellar wind model makes a very different prediction of the radio synchrotron flux of g2 , so the radio flux will be an important early clue about the nature of g2 . | the radio flux from the synchrotron emission of electrons accelerated in the forward bow shock of g2 is expected to peak when the forward shock passes pericenter , possibly 7 to 9 months before the center of mass of g2 reaches pericenter @xmath0 cm from the galactic center @xcite . in this letter
, we calculate the radio emission from the forward and reverse shock if g2 is a momentum - supported bow shock of a faint star with a high mass - loss rate as suggested by @xcite .
we show that the radio flux lies well below the quiescent radio flux of sgr a * and will be difficult to detect . by contrast , in the cloud model of g2 , the radio flux of the forward shock is predicted to be much larger than the quiescent radio flux and therefore should have already been detected @xcite .
therefore , radio measurements can reveal the nature of g2 well before g2 completes its periapsis passage .
galactic center , black hole physics , radiation mechanism , non - thermal |
bootstrap resampling , or sampling with replacement from the given data , is used to mimic the sampling process that produced the original sample in the first place . by randomly drawing items from the original dataset with equal probability
, one is effectively drawing a fresh sample from the `` empirical distribution '' , a multinomial with equal probability assigned to each item in the original sample .
this can be seen as a best guess at the true population distribution .
it has been over 30 years since @xcite introduced the bootstrap to statistics , and now there exist many different applications and related resampling techniques are found in statistics and computer science .
of particular interest to researchers in pattern recognition and machine learning are the bagging ensemble method of @xcite and the .632 + validation scheme of @xcite for supervised algorithms . in these settings , and others involving the construction of prediction rules on bootstrap samples , an important quantity of interest is the number of unique items from the original sample @xcite . in many cases , this can be seen as a limit on the amount of information carried down from the original sample . for learning algorithms trained on bootstrap samples ,
the reduction in the number of unique items expected can be viewed as an effective reduction in training sample size , particularly in high dimensional problems where the prediction model is unstable with respect to changes in the training data .
while generally it is a desirable property of learning algorithms that they are flexible and responsive to new information , there also exist specific cases where the unique items in a training sample are the sole determinant of the prediction rule . in ( hard - margin )
support vector machine classification where the number of dimensions exceeds the number of items observed , it is always possible to divide the two classes in the training data .
the prediction rule is determined only by the points that lie on the maximum margin separating hyperplane , whose duplication has no effect on its location .
another simple example is found in nearest - neighbour classification and regression in continuous feature spaces where the probability of obtaining any truly identical training items is zero . without competitions between multiple prediction values due to items at exactly the same location , the resulting prediction rule is unaffected by duplications . as the number of unique items in a bootstrap sample is an important determinant of the behaviours of prediction rules learned on it , the distribution of this quantity should be of interest to researchers working on their development and validation .
while related distributions have long been studied in a purer mathematical context @xcite , and this distribution has been identified before in this setting @xcite , nowhere were we able to find a concise and accessible summary of the relevant information for the benefit of researchers in machine learning .
our aim here is to fill this gap by presenting this distribution along with its key properties , and to make it easier for others who to understand or modify resampling techniques in a machine learning context . in section [ background ]
, we discuss the relevance of the number of unique items to the bagging ensemble method and the .632 + validation scheme . in section [ present_dist ]
we give a closed form of this distribution together with its notable properties .
we give an empirically derived rule to decide whether normal approximation is permissible , and describe how we produced it . in section [ multi ] , we consider the case where the items in the original sample belong to one of several categories , as is the case in classification problems . here , where the outcome is the vector of the number of unique original items from each category , we show that the limit distribution is multivariate normal , and consider limits to produce a heuristic for the normal approximation of the number of unique items from a single category .
bootstrap aggregation , or bagging , uses the perturbed samples generated by sampling with replacement to produce a diverse set of models which can then be combined to promote stability @xcite . though a given model may be over - fit to its sample , the combination should less reflect those unstable aspects particular to an individual sample @xcite .
perhaps its most important use in pattern recognition has been in random forest classification and regression @xcite , where it is key to allowing the use of flexible decision trees without over - fitting .
while bagging is still most commonly performed by drawing @xmath0 items with replacement , the way in which the samples are drawn is not necessarily fixed .
the number of unique items present in a bootstrap sample has been identified as an important predictor of algorithm performance , and the sampling method can be purposefully modified to control its range and variability @xcite .
an added bonus of bagging is the possibility of producing a performance estimate for the ensemble without doing additional cross validation @xcite .
predictions are made for items in the training sample using only those models which did not use them in training . in this way , it is possible to get a performance estimate for an ensemble without the bias of over - fitting . a pessimistic bias has been observed in high dimensional classification problems that was ameliorated using different sampling techniques @xcite , suggesting the importance of unique items in this phenomenon .
there is no one dominant scheme for the statistical comparison of learning algorithms on small to moderately sized datasets , and researchers must choose between different forms of hold out tests and cross validation .
one option available to them is the .632 + bootstrap of @xcite .
this , and the earlier .632 scheme @xcite , extend the bootstrap to the validation of learned prediction rules . both of these are functions of a quantity called the leave - one - out bootstrap error , which is the average error of the prediction models constructed on many bootstrap samples . to avoid the optimistic bias of testing on the training set ,
the models are tested only on those points not in their respective bootstrap samples . as in bagging
, the prediction rules created can have an effectively reduced training sample size in line with the lower number of unique original items in their samples .
the expectation of the leave - one - out error then becomes a weighted average of the expected algorithm performance at the different sizes given by the distribution of that quantity . unlike conventional cross validation schemes where the reduced sample size is obvious , users of this technique
may not have the effective reduction in mind . while both the .632 and .632 + methods attempt to correct for the pessimistic bias that may result , the true corrective mapping of this value will be problem specific and can not be known ahead of time .
this is almost certainly the cause of the bias observed by @xcite .
knowledge of the distribution of the effective sample sizes in these schemes should be useful to those interpreting the results of these schemes , particularly when there is concern about bias .
in this section , we first provide a closed form of the distribution of the number of unique items and describe its connection to the family of occupancy distributions . with this information ,
we are then able to derive the integer moments and pormal limit .
finally , we describe the construction of a heuristic to decide when normal approximation is appropriate .
some illustrations of the distribution are given in fig .
[ onedim ] . where @xmath1 items are taken with replacement from a sample of size @xmath0 , the probability of obtaining @xmath2 unique original items is @xmath3 where @xmath4 , @xmath5 is the falling factorial , and @xmath6 is a stirling number of the second kind .
equivalently , this is the probability of obtaining @xmath2 unique outcomes after @xmath1 categorical trials in a multinomial problem with @xmath0 equally likely outcomes . because each of the @xmath0 items is equally likely to be drawn at each of the @xmath1 sampling events , there are @xmath7 equally likely bootstrap samples that may be drawn if order is accounted for .
the number of ways to make an ordered selection of @xmath2 unique items from the original @xmath0 to be present in the bootstrap sample is @xmath5 , the falling factorial .
this is defined as follows : @xmath8 having already considered the ordering of the @xmath2 unique items to be included means that we may consider them unlabelled in the final step ; here , we consider the number of ways to take @xmath1 distinguishable draws from @xmath2 unlabelled unique items such that at least one draw is taken from each . in more generic terms
, this is the number of ways to place @xmath1 labelled items into @xmath2 subsets such that none of them are empty .
this quantity is written as @xmath6 , and is called a stirling number of the second kind ( * ? ? ?
* chapter 6 ) . combining these three results
gives us eq . .
the number of unique items present may be viewed as the sum of the set of @xmath0 indicator functions @xmath9 .
the indicator @xmath10 corresponds to original item @xmath11 , taking the value of one if this is present and zero if not .
a simple consideration of the sampling process allows one to find the mean and covariance of the indicator functions .
as all points are equally likely to be selected , the chance of an item not being selected at a given sampling trial is @xmath12 . in order for @xmath10 to be zero ,
its corresponding point must not be selected at all @xmath1 events .
this gives us @xmath13 .
as the indicators are binary variables , this is sufficient to determine their mean and variance : @xmath14 & = { \text{p}}(d_i = 1 ) = 1 - { \text{p}}(d_i = 0 ) \\ & = 1 - \big ( 1 - \frac{1}{n } \big)^a , \\ \text{var}[d_i ] & = { \text{p}}(d_i = 0 ) \cdot { \text{p}}(d_i = 1 ) \\ & = \big ( 1 - \frac{1}{n } \big)^a - \big ( 1 - \frac{1}{n } \big)^{2a}. \end{split}\ ] ] a similar consideration gives us the probability of two items , @xmath11 and @xmath15 ( where @xmath16 ) both being excluded from a bootstrap sample .
this is given @xmath17 , and is enough to give us the covariance of two indicator functions , @xmath18 & = { \text{p } } ( d_i = 1 , d_j = 1 ) - \big({\text{p}}(d_i = 1)\big)^2 \\ & = \big(1-\frac{2}{n}\big)^a - \big(1-\frac{1}{n}\big)^{2a}. \end{split}\ ] ] now that we know the mean , variance and covariance of the indicators , we have enough information to determine the same quantities for any linear combination of them , including their sum , @xmath2 .
of particular interest to us is the case where the number of samples drawn is proportional to the number of original items .
for this reason , along with the general formulae , we also provide limits for the mean and variance as @xmath0 and @xmath1 jointly approach infinity : @xmath19 & = n(1 - \big(1 - \frac{1}{n}\big)^a ) = \frac { n^a - ( n-1)^a } { n^{a-1 } } \\ & \to n(1 - e^{-\alpha } ) , \\ \text{var}[k ] & = n(n-1)\big(1-\frac{2}{n}\big)^a + n\big(1 - \frac{1}{n}\big)^a - n^2\big(1-\frac{1}{n}\big)^{2a } \\ & \to n(e^{-\alpha } - ( 1+\alpha)e^{-2\alpha } ) , \end{split}\ ] ] where the limits refer to the case @xmath20 and @xmath21 .
, scaledwidth=55.0% ] while we came to the problem of item inclusion through machine learning , the study of this problem far predates our field of research @xcite .
if , instead of the number of items included in the bootstrap sample , we were to consider the number of items _ excluded _ , we would be studying one of the family of occupancy distributions @xcite .
these arise in the study of urn problems when @xmath1 balls are independently and randomly placed into @xmath0 urns in such a way that each ball is equally likely to be placed into any one of the urns .
the @xmath22 occupancy distribution then details the probabilities of obtaining different values of @xmath23 , the number of urns containing exactly @xmath11 balls . in our problem ,
the @xmath1 drawing events are the balls of the urn model , and the @xmath0 original items that may be selected are the urns with which the balls / draws can be associated .
the number of empty urns is @xmath24 , and it has the same distribution as the number of excluded items in our sampling problem . as @xmath2 is just @xmath25 , it is easy to drawn upon useful results . while there exist a great many limit theorems for occupancy distributions ( *
* chapter 6 ) , in our case we need only the earliest .
@xcite proved that the limit distribution of @xmath24 was normal in the case that @xmath20 , and @xmath21 .
as @xmath2 is distributed as @xmath25 , it too must be normally distributed .
the @xmath26 raw moment of the distribution of @xmath2 is @xmath27 a proof for this is provided in appendix [ rawproof ] .
@xcite provides a formula for the central moments of the @xmath28 occupancy distribution , which for us corresponds to the number of excluded items . by simplifying this and adapting
the sign to the distribution of the number of items _ included _ , we can give the @xmath26 central moment as @xmath29 in both of these formulae the sum over @xmath30 and @xmath31 comprises @xmath32 unique locations .
reassuringly , they provide the identity @xmath33 , as well as results for the mean and variance that match up to those derived by the considerations of [ indicators ] .
the stirling numbers quickly overflow conventional precision , making the form of the distribution in eq . hard to work with at reasonable values of @xmath0 and @xmath1 .
while it is known that the distribution does converge to normal , this is not practically useful unless one knows when a normal approximation is appropriate . to allow others to use a normal approximation with confidence
, we decided to build a heuristic based on existing rules of thumb for normal approximation of the binomial distribution . where the number of bernoulli trials is @xmath34 and the probability of success is @xmath35 at any one , a normal approximation works best when the @xmath34 is high and @xmath35 is not too close to zero or one . while there is a degree of arbitrariness in deciding when two distributions are `` close enough '' , there have long existed conventions for doing so .
two of the most common such rules from applied statistics @xcite are @xmath36 in order to build our own rule for approximation of @xmath37 , we decided first to measure the convergence of the true distribution and its normal approximation to measure the quality of the approximation of the binomial case under the rules of .
we could then experiment with building rules to permit approximation of the distribution of @xmath2 such that convergence was always better than the worst case seen for the binomial approximation . to measure the quality of approximation , we chose a metric closely linked to the concept of convergence in distribution : the maximum absolute difference in cumulative distribution ( madcd ) between the true distribution and its continuity - corrected normal approximation on the set of all possible outcome values .
the continuity correction was performed using a shift of @xmath38 .
computation of the exact value of the true distribution was performed using the symbolic mathematics toolkit of the matlab software package . to find the worst permissible binomial approximation , we measured the madcd between the binomial distribution and its continuity - corrected normal approximation across the grid of parameters specified by @xmath39 and @xmath40 .
we combined the two rules of eq . to produce a third , stricter rule , and found a maximum madcd of 0.0205 within the resulting acceptance region .
as this value occurred far from the boundary of the allowed set , we think it unlikely that expanding the grid would produce higher values .
we then mapped the madcd for @xmath37 and its continuity - corrected normal approximation across the grid of parameters @xmath41 and @xmath42 , so that we could begin constructing a rule .
the results can be seen in fig .
[ madcd_grid ] . after inspecting this map ,
we decided to construct rules for the minimum permissible @xmath1 at a given @xmath0 ( and vice versa ) , of the form @xmath43 .
while there are areas near @xmath44 and @xmath45 where the madcd value is low , this did not represent convergence to normal , rather , most of the probability mass has collapsed to a single value of @xmath2 .
for this reason , we chose to focus on the central valley of good approximation near the axis @xmath46 . , scaledwidth=55.0% ] we were mainly concerned with producing a reliable rule that would not fail at parameter setting outside the mapped grid ; while we experimented with different fitting methods , we chose to select the rule parameters by hand so as to have greater control over features likely to make it more conservative .
this also made it easier to restrict the rule parameters to those fully specified by two digits of decimal precision .
we produced a rule such that * all madcd values in the acceptance region were below the specified level of 0.0205 , * the boundaries of the acceptance region did not cross the ridges of the distribution as @xmath0 and @xmath1 increase , unless it was to move towards the central valley , and * the values on the boundary had to appear to be decreasing with increasing @xmath0 and @xmath1 .
we excluded the smallest values of @xmath1 and @xmath0 in order to better fit of the general pattern .
the resulting rule is that a normal approximation is permissible when @xmath47 the acceptance region specified and the values of madcd at its boundaries can be seen in fig .
[ madcd_grid ] and [ madcd_grid_bound ] respectively .
in addition to the madcd , we also computed the jensen - shannon divergence between the true probability distribution and the normal approximation .
as this is function of a probability distribution itself , rather than a cumulative distribution , it was necessary to discretise the normal approximation .
this was done by defining the probability mass at a location as the difference in the continuity - corrected cumulative distribution between that location and the next .
we found a maximum value of 0.0444 within the allowed region for the binomial distribution , and 0.0631 for the allowed region for the distribution of @xmath2 .
all the highest divergence values for @xmath37 and its approximation occurred on the boundaries near the very lowest allowed values of @xmath1 and @xmath0 .
, scaledwidth=55.0% ]
in some machine learning tasks , notably in supervised classification , items in the sample belong to one of several categories .
it is then of interest to those studying bootstrap techniques to know the distribution of the balance of the unique items drawn from each category .
we consider the case where there are @xmath48 categories present , the @xmath49 of which contains @xmath50 items in the original sample of @xmath0 . when @xmath1 samples are drawn without replacement , the distribution can be found by marginalising over the number of draws taken from each category , the various @xmath51 .
the probability of obtaining the vector @xmath52 detailing the number of items @xmath53 from each of category is @xmath54 an example of this distribution in the case of two categories is illustrated in fig .
[ twodim ] . ,
scaledwidth=55.0% ] from consideration of the mean and covariance of the individual items participations as in section [ indicators ] , it is straightforward to derive the mean , variance and covariance ; these are the following : @xmath55 & = n_i ( 1 - \big(1 - \frac{1}{n}\big)^a ) , \\ \text{var}[k_i ] & = n_i(n_i-1)\big(1-\frac{2}{n}\big)^a + n_i\big(1 - \frac{1}{n}\big)^a - n_i^2\big(1-\frac{1}{n}\big)^{2a } , \\ \text{cov}[k_i , k_j ] & = n_in_j\big[\big(1-\frac{2}{n}\big)^a - \big(1-\frac{1}{n}\big)^{2a } \big ] , \text { for $ i \neq j$. } \end{split}\ ] ] not only is the total number of unique items from the original sample asymptotically normally distributed under appropriate conditions , but so is the number from the subset belonging to each category . to show this , we must turn to limit theorems for dependent variables rather than the study of urn problems .
the covariance between the presence indicators of section [ indicators ] is always negative ( see eq . ) .
as these are binary variables , this means that the joint distribution of any two indicators @xmath10 and @xmath56 ( @xmath16 ) must meet the condition @xmath57 for any values @xmath58 and @xmath59 .
any collection of the indicators can then be termed a pairwise negative quadrant dependent ( pnqd ) sequence @xcite .
weighted sums of pnqd sequences are asymptotically normally distributed as the number of items in the sequence goes to infinity @xcite .
the number of unique items present from the subset of a category is a weighted sum of the indicators , and so its asymptotic distribution is normal as the number of items present and items drawn jointly approach infinity . the number of items in a category must approach infinity too , as the number of possible values that the sum can take must also go to infinity for its discrete distribution to converge to a continuous one .
this is guaranteed in the case where each category contains a fixed fraction of the total number of items . to show the limit of the joint distribution of the numbers of unique items from the categories is multivariate normal
, we can consider the set of weighted sums that give equal weight to all items within a category . if the weight given to items in category @xmath60 is @xmath61 and we add the constraint @xmath62 , then the distributions of this set still include all possible projections of the multivariate distribution of @xmath63 .
as all projections of this distribution are normal , then it itself must be multivariate normal @xcite .
we now consider when it is appropriate to approximate the distribution of the number of unique items from a subset of the data with a normal distribution .
we consider the case where the original items has @xmath0 items of which @xmath50 belong to a particular subset of fixed size , and where the number of items drawn is @xmath20 . in the case where @xmath64
, we have the heuristic developed in [ heuristic ] .
as @xmath21 while @xmath50 remains fixed , the correlation between the indicator functions of the items of category @xmath60 goes to zero and the distribution of @xmath53 approaches a binomial distribution with @xmath50 trials and a probability of success @xmath65 .
we argue that if both the distribution with @xmath66 and that with @xmath21 are well approximated by the binomial , it is highly likely that the intermediate stages are too .
this suggests the following rule for normal approximation of @xmath53 : if the number of samples drawn is proportional to the number of items ( @xmath20 ) and the expected number drawn from the category @xmath60 is @xmath67 , then if both the rules for approximation of the binomial limit ( with @xmath68 and @xmath69 ) are met , and the rule for the single category case with @xmath0 and @xmath1 substituted for @xmath50 and @xmath70 respectively , then a normal approximation should be appropriate for @xmath53 for any value of @xmath0 .
the resulting rule requires four inequalities to be met , which makes it a little unintuitive and tedious to apply .
as the inequalities due to the consideration of the binomial limit only have a small effect on the acceptance region near the limit of lowest @xmath50 / highest @xmath71 , and eventually become redundant as @xmath50 and @xmath71 increase , it is possible to account for them with a simple offset of the rule for the single category case .
a small strip of @xmath71@xmath50 parameter space is then sacrificed for the sake of simplicity .
we find a normal approximation to be permissible when @xmath72 this has no solutions with @xmath73 or @xmath74 .
we have aimed to make a clear and concise answer to the titular question readily available to machine learning researchers .
we have summarised the key properties of this distribution , and provided practical information about when a normal approximation is appropriate in the form of a heuristic , allowing others to justify its use . with particular consideration to classification problems
, we have considered the generalisation of this distribution to the scenario where items come from multiple categories . in this case , we have provided a theoretical guarantee of asymptotic normality and a considered heuristic rule for approximation based its limit distributions .
we hope our results rovide a useful resource to researchers interested in understanding or modifying the number of unique items to appear under random sampling with that replacement .
this work is funded by ucl ( code elcx ) , a case studentship with the epsrc and ge healthcare , epsrc grants ( ep / h046410/1 , ep/ h046410/1 , ep / j020990/1 , ep / k005278 ) , the mrc ( mr / j01107x/1 ) , the eu - fp7 project vph - dare@it ( fp7-ict-2011 - 9 - 601055 ) , the nihr biomedical research unit ( dementia ) at ucl and the national institute for health research university college london hospitals biomedical research centre ( nihr brc uclh / ucl high impact initiative ) . 10 a. abadie and g. w. imbens . on the failure of the bootstrap for matching estimators . , 76(6):15371557 , 2008 .
g. box , w. g. hunter , j. s. hunter , et al . .
john wiley and sons new york , 1978 .
l. breiman .
out - of - bag estimation . technical report , citeseer . l. breiman . bagging predictors . , 24(2):123140 , 1996 . l. breiman . random forests .
, 45(1):532 , 2001 .
p. bchlmann and b. yu . analyzing bagging . ,
pages 927961 , 2002 .
b. efron .
bootstrap methods : another look at the jackknife . , pages 126 , 1979 .
b. efron .
estimating the error rate of a prediction rule : improvement on cross - validation .
, 78(382):316331 , 1983 .
b. efron and r. tibshirani .
improvements on cross - validation : the 632 + bootstrap method .
, 92(438):548560 , 1997 . a. gut .
springer , 2009 .
n. l. johnson and s. kotz . .
wiley , 1977 .
estimating classification error rate : repeated cross - validation , repeated hold - out and bootstrap .
, 53(11):3735 3745 , 2009 .
d. e. knuth , r. graham , and o. patashnik . .
addison , 1989 .
li and j .- f .
an application of stein s method to limit theorems for pairwise negative quadrant dependent random variables .
, 67(1):110 , 2008 .
m. w. mitchell .
bias of the random forest out - of - bag ( oob ) error for certain input parameters .
, 1:205 , 2011 .
j. muoz - garca , r. pino - mejias , j. munoz - pichardo , and m. cubiles - de - la - vega .
identification of outlier bootstrap samples .
, 24(3):333342 , 1997 .
r. pino - mejas , m .- d .
jimnez - gamero , m .- d .
cubiles - de - la vega , and a. pascual - acosta . reduced bootstrap aggregating of learning algorithms .
, 29(3):265271 , 2008 . c. r. rao , p. pathak , and v. koltchinskii
. bootstrap by sequential resampling .
, 64(2):257 281 , 1997 .
m. schader and f. schmid .
two rules of thumb for the approximation of the binomial distribution by the normal distribution . , 43(1):2324 , 1989 .
i. weiss . limiting distributions in some occupancy problems .
, pages 878884 , 1958 .
before proceeding , we shall need two intermediate results .
* add one to the upper argument .
* subtract one from the lower argument and multiply by minus one .
* multiplication by offset between @xmath2 and the second argument of the stirling number ( i.e. , the initial @xmath78 plus the number of @xmath79 operations that have occurred to the number thus far ) . by multiplying by @xmath80
, we sequentially apply all these operations @xmath75 times . in a way similar to pascal s triangle
, we can see this triplication as carrying a `` charge '' of stirling numbers down to the next level of a pyramid ( see _ a ) _ in fig .
[ pyramid ] .
locations in the pyramid can be described by the number of times each operator has been applied to all charge that reached them .
we shall call these coordinates @xmath81 , @xmath82 and @xmath31 .
any given selection of @xmath75 populates the pyramid down to the level described by @xmath83 . at any point in the pyramid ,
the arguments of the stirling numbers are determined only by the coordinates .
if these operators commuted , we could consider a flow of charge exactly the same as in pascal s triangle , and apply the effects of the operators on the coefficients of the stirling numbers separately at the end .
as @xmath79 and @xmath84 do not commute with one another , we must consider their interaction and how they change the charge passing through them .
@xmath85 commutes with the other operators and does not change the coefficient of the charge , so we can `` factorise '' the pyramid .
that is , we need only consider the number of ways to select @xmath86 non-@xmath85 operations , and then to sum over all possible orders of choosing @xmath82 and @xmath81 operations of @xmath79 and @xmath84 respectively .
the amount of charge at a location in the pyramid will be @xmath87 , where @xmath88 is the coefficient from the charge triangle of the operators @xmath79 and @xmath84 alone . to determine what this is
, we use the fact that the @xmath78 in eq .
is initially zero . see _
b _ in fig . [ pyramid ] for illustration .
this gives us the initial conditions @xmath89 and @xmath90 for @xmath91 .
together with the recurrence @xmath92 , these are enough to uniquely specify @xmath93 .
this then gives us eq . .
we begin with the standard relation @xcite @xmath95 from eq . , it is clear that @xmath96 for values of @xmath2 higher than @xmath0 .
consistent with their recurrence relation and combinatorial meaning , the stirling numbers @xmath6 can be defined as zero for positive values of @xmath2 outside @xmath97 @xcite . as
one of these two factors is zero whenever @xmath2 exceeds either @xmath1 or @xmath0 , we can truncate the sum in [ standardsum ] to the smaller of those without effect .
we can therefore write @xmath98 if we combine this with the knowledge that @xmath99 we can then produce eq . . if we take the definition of the moment as in the first line of eq .
, apply eq .
, reverse the order of summation , and then apply eq .
, we are then able to derive the final form for the raw integer moments seen in the second line ( of eq . [ moments ] ) . | sampling with replacement occurs in many settings in machine learning , notably in the bagging ensemble technique and the .632 + validation scheme .
the number of unique original items in a bootstrap sample can have an important role in the behaviour of prediction models learned on it . indeed , there are uncontrived examples where duplicate items have no effect .
the purpose of this report is to present the distribution of the number of unique original items in a bootstrap sample clearly and concisely , with a view to enabling other machine learning researchers to understand and control this quantity in existing and future resampling techniques .
we describe the key characteristics of this distribution along with the generalisation for the case where items come from distinct categories , as in classification . in both cases
we discuss the normal limit , and conduct an empirical investigation to derive a heuristic for when a normal approximation is permissible .
* keywords : * bootstrap , resampling , sampling with replacement , bagging |
one of the most transparent pictures of the scrdinger s eigenvalue equation in 1-dimension is obtained by reinterpreting the space coordinate @xmath1 as the `` time '' ( @xmath2 ) , and the values of the wave function @xmath0 and its derivative @xmath3 as the position and momentum of a classical point particle @xcite .
the resulting classical trajectory obeys hamilton equations which can be further simplified by introducing the prfer s angle on the poincare plane @xcite .
the integration of the 1-st order `` angular equation '' then turns out one of the most efficient methods to determine the energy spectra @xcite . in this note
we show that the classical algorithm works even better for the helmholtz problem @xcite ; in addition , it exhibits a new aspect of the helmholtz spectra .
given an interval @xmath4 \subset { \bf r}$ ] and a real function @xmath5 \rightarrow { \bf r}$ ] , the _ helmholtz equation _ is a particular case of a sturm - liouville problem ( see @xcite ) : @xmath6 and the _ helmholtz spectrum _ is defined as the set of all @xmath7 for which ( 2.1 ) has a non trivial solution @xmath8 vanishing on both ends : @xmath9 the trajectory picture @xcite applies quite naturally to the helmholtz eigenproblem , and the spectral condition is much simpler than in the schrdinger s case .
let @xmath10 be any real solution of ( 2.1 ) .
define : @xmath2 , @xmath11 , @xmath12 ; eq . ( 2.1 ) then becomes : @xmath13 with the spectral condition : @xmath14 observe , that ( 2.3 ) are the canonical equations for a classical point - particle with the hamiltonian : @xmath15 each integral trajectory of ( 2.3 ) portraits a wave function @xmath0 and its first derivative .
we shall further assume that @xmath16 is a piece - wise continuous function in @xmath4 $ ] .
the attractive oscillator force in ( 2.3 ) then drives the classical trajectory clockwise around the origin of the phase plane ( fig .
when @xmath17 grows to @xmath18 , the end point @xmath19 must cross the @xmath20 axis an infinite number of times ( fig .
each time this happens , @xmath17 belongs to the point spectrum of ( 2.1 ) . in this way
the helmholtz eigenproblem ( 2.1 ) has a discrete spectrum in the form of an infinite ladder of positive eigenvalues determined by the classical trajectories ( compare the theorems @xcite ) .
the calculation of the spectral values can be carried out by introducing the polar variables on the phase plane : @xmath21 , @xmath22 .
the variable @xmath23 then separates and ( 2.3 ) reads @xcite : @xmath24 with the initial and spectral conditions : @xmath25 some exact solutions of ( 2.1 - 2 ) were recently studied by goyal et al to check the validity of the wkb and langer approximations @xcite . for curiosity
, we applied the angular method to the same collection of cases .
( 2.6 ) was integrated numerically for a net of @xmath17-values by applying the runge - kutta method and the condition ( 2.9 ) selected the eigenvalues @xmath17 quoted in table 1 . [ cols= " <
, < , < , < , < , < " , ] * table 1 : * helmholtz eigenvalues known in the literature compared with the results of the angular method . in addition we have examined @xmath26 in @xmath27 $ ] .
the last column gives the _optical information_ , disregarded in the literature .
as one can see , for the low eigenvalues the ranking is won by the almost ignored angular algorithm ( both , in number of digits and simplicity of operation ) .
its additional advantage is that it uncovers the links of the helmholtz spectrum with the techniques of generating the squeezed states .
in a typical sturm - liouville problem , one looks for a solution of a differential equation which fulfills a definite _ boundary condition .
the second independent solution is dimissed as unphysical .
however , we shall show that for ( 2.1 ) the second solution too carries an important message . as eqs .
( 2.3 ) are linear , the corresponding @xmath28 , @xmath29 depend linearly on the initial values @xmath30 , @xmath31 : _ @xmath32 where @xmath33 is the simplectic _ evolution matrix _ @xcite ( also : the _ transfer matrix _ ) .
the existence of a non - trivial solution of ( 2.2 ) with @xmath34 means that the evolution matrix @xmath35 in the interval @xmath4 $ ] maps : @xmath36 and since @xmath35 is simplectic , it must be of the form : @xmath37 once this is known , one can deduce not only the spectral orbit of ( 2.3 ) , but also the form of other trajectories .
consider therefore the congruence of trajectories which diverge at @xmath38 from the initial point ( @xmath39 ) , where @xmath40 is fixed and @xmath41 varies . if @xmath17 is an eigenvalue of ( 2.1 ) and @xmath42 is of the form ( 3.3 ) , then at @xmath43 : @xmath44 i.e. , all orbits departing from the same @xmath40 ( with different momenta p ) meet at @xmath45 ( with momenta @xmath46 ) .
thus , the eigenvalues @xmath17 of ( 2.1 ) are the real numbers for which the time evolution ( 2.3 ) focuses any congruence of classical orbits diverging from any common initial @xmath40 .
the optical constants @xmath47 , @xmath48 can be read from the second ( disregarded ) solution of ( 2.1 ) .
note that the focusing , in general , is not the _ scale transformation @xcite , ( as the matrix ( 3.3 ) is triangular ) , though it is an example of squeezing @xcite .
the scale transformation ( @xmath49 ) is an exceptional effect which can happen only for some @xmath50 s ( see @xcite ) , whereas the _ focusing _ is a typical phenomenon which must occur for any positive @xmath51 , on an infinite ladder of @xmath17 s .
the converse problem is as well interesting . instead of fixing an interval and looking for @xmath17 ,
fix any @xmath52 and @xmath53 and consider an expanding time interval @xmath54 $ ] . if @xmath55 mantains some positive average in @xmath56 , the final point @xmath57 , as @xmath58 must cross infinity of times the @xmath41 -axis , each time generating our focusing effect no matter the exact form of @xmath59 .
should @xmath59 be a random variable performing errating oscillations , the motions ( 2.3 ) will be chaotic too ; yet , for any congruence of trajectories , a clean echo effect of the initial configuration ( an amplified or reduced image ) must emerge now and then against the chaotic forces .
note also that the effect must have a quantum equivalent .
consider the hamiltonian @xmath60 in form ( 2.5 ) , where @xmath61 and @xmath41 are position and momentum operators . since ( 2.5 )
is cuadratic , the quantum evolution follows faithfully the classical motions .
the techniques of the classical `` piloting trajectory '' gives specially good results for the gaussian wave packets @xcite .
we shall show , that it provides also an explicit expression for the evolution operator @xmath62 : @xmath63 with this end define : @xmath64 where @xmath65 is the traditional oscillator hamiltonian and @xmath66 is a real , differentiable function .
due to the well known transformation rules : @xmath67 , @xmath68 , the operator @xmath69 fulfills : @xmath70 w(t ) \nonumber\end{aligned}\ ] ] notice , that the time dependent term coupled to @xmath71 corresponds exactly to the prfer equation .
thus , if @xmath66 is the prfer angle ( 2.6 ) , then ( 4.3 ) reduces to : @xmath72 w(t)\end{aligned}\ ] ] and integrates explicitely in terms of the prfer s radious @xmath73 : @xmath74 inserting ( 4.5 ) into ( 4.4 ) and remembering that the first exponent in ( 4.4 ) is the scale operation , one gets : @xmath75 and returning to ( 4.2 ) one has an explicit formula for the unitary operator @xmath62 with three exponential terms piloted by prfer variables ( 2.6 - 7 ) : @xmath76 if in addition the number @xmath17 is the eigenvalue of the sturm - liouville problem ( 2.1 ) @xmath77 ( @xmath78 ) , then the first factor in ( 4.7 ) becomes @xmath79 ( where @xmath80 ) and @xmath81 transforms any initial wave packet @xmath82 into : @xmath83 ( x ) = { 1 \over \sqrt{\sigma } } e^{i \eta { x^2 \over 2 \sigma^2 } } \psi ( x / \sigma)\ ] ] where @xmath84 .
the probability density @xmath85 is simultaneously mapped into its own diffractionless ] it thus turns out that the authors looking for the squeezing parameters were indeed solving the xix - century spectral problem ! the relative simplicity of the three term formula ( 4.7 ) ( compare @xcite ) is due to the `` bottom structure '' ( 2.6 - 9 ) .
its additional implication concerns the `` potential kicks '' @xmath86 considered recently as a method of controlling and cooling quantum systems @xcite .
while the `` kicks '' are mathematically simple , they are overidealized ( no @xmath87-pulses available in laboratories ! ) .
notice , however , that if @xmath17 is an eigenvalue of ( 2.1 ) and if @xmath55 is symmetric @xmath88 in a symmetric operation interval @xmath89 $ ] , then the eigenvector @xmath8 ( and consistenly the phase trajectory @xmath90 is even / odd in @xmath89 $ ] , implying the trivial squeezing @xmath91 , and the only non trivial part of ( 4.7 - 8 ) is @xmath92 . for @xmath93 , this is the unitary operator describing the results of a sharp kick of the oscillator potential @xmath94 .
thus , the spectrum of ( 2.1 ) tells in addition how to `` imitate softly '' the effects of @xmath87-kicks converting the results on kicked oscillators @xcite into a branch of realistic technology
. 0.5 cm
0.5 cm where * n * is a constant unit vector and @xmath97 is the magnetic field intensity , @xmath98 defining the amplitude and @xmath99 the shape of the magnetic pulses . if * n * point the @xmath100-axis , the schrdinger s wave packet obeys the hamiltonian : in ( 5.2 ) the term @xmath107 means the free propagation along the @xmath100-axis , the @xmath108 generates rotations in the @xmath109-plane . the last term ( a 2-dimensional oscillator ) , splits into two twin 1-dimensional oscillators of the form : where @xmath111 means either @xmath112 or @xmath113 .
quite evidently , ( 5.4 ) is a model of ( 2.5 ) with @xmath114 and @xmath115 . if @xmath116 , the spectrum of the helmholtz equation ( 2.1 ) forms a ladder of eigenvalues , and if @xmath117 is one of them , both oscillators yield the simultaneous focus of the trajectory congruencies ( in classical case ) or probability patterns ( in quantum case ) . to check this ,
we have asked the computer to draw congruencies of classical trajectories diverging from a common initial point , with a common initial @xmath118 , for @xmath114 coinciding with one of the eigenvalues of ( 2.1 ) and with @xmath119 taken from table 1 . in all cases
we have obtained quite precise optical images on planes perpendicular to the solenoid axis ( see fig .
the dependence of the phenomenon on the sturm - liouville spectrum is illustrated by sequences of amplified focuses corresponding exactly to the nodal points of the wave function ( 2.1 - 2 ) .
henceforth , the integration of the first order differential equation of prfer ( for one variable ) offers the simplest algorithm to predict the _ optical echo _ ( squeezing ) for charged particle beams .
the `` reduced images '' might imply new schemes to produce narrow particle beams .
the `` amplfied images '' are no less interesting .
they can certainly be used to separate the wave packet components in the quantum state engineering @xcite ) .
the magnetic pulses @xmath120 with @xmath121 symmetric in the operation interval and with @xmath117 in the spectrum of ( 2.1 ) , provide a magnetic model for systems under the influence of oscillator kicks .
a. malkin , v. i. manko and d. a. trifonow , j. math .
* 14 * , 576 ( 1973 ) ; v.v .
dodonov , i. a. malkin and v. i. manko , j. phys . * a 8 * , 19 ( 1975 ) ; int .
* 14 * , 37 ( 1975 ) ; n. a. ansari , v. i. manko , phys . lett . * a 223 * , 31 ( 1996 ) .
b. mielnik , j. math .
* 27 * , 2290 ( 1986 ) ; v.v .
dodonov , o.v . manko and v.i .
manko , phys . lett . * a 175 * , 1 ( 1993 ) ; b. mielnik and d.j .
fernndez c. , iv icssur , nasa conf .
3322 , ed .
et al _ , maryland ( 1996 ) , pp .
241 - 251 ; h. ammann and n. christensen , phys .
* 78 * , 2088 ( 1997 ) .
* fig . 1.- * the orbit of the classical oscilator ( 2.5 ) for @xmath122 , @xmath123 .
the points of intersection with the @xmath41-axis determine the time moments @xmath124 , and the amplifications @xmath125 of the optical images .
the similar phenomenon would be generated by a chaotic @xmath126 .
* fig . 2.- * the orbits of ( 2.3 ) and the helmholtz spectrum of ( 2.1 ) .
if the end point of the orbit with q(a)=0 places itself on the @xmath41-axis , @xmath17 is an eigenvalue of the helmholtz equation ( 1 ) . above : a sequence of orbits reaching @xmath127 after multiple intersections with the @xmath41-axis illustrates the sequence of helmholtz eigenstates for @xmath128 , @xmath4= [ 0 , 1]$ ] .
the corresponding optical constants are reported in table 1 . *
* the optical effect of the helmholtz spectrum observed in the solenoid for sinusoidally oscilating @xmath129 . _
a ) _ when @xmath130 is an eigenvalue of ( 2.1 ) , the congruence of particle trajectories forms an exact optical focus at @xmath131 ( @xmath132 ) ( in quantum case , an amplified diffractionless image ) with @xmath133 .
the integration of the first order differential equation of prfer for one angular variable turns out the simplest method to predict the effect .
_ b ) _ for @xmath134 the congruence forms @xmath135 increasingly amplified focuses ( images of the initial point ) in exact correspondence to the @xmath135 nodal points of the wave function @xmath0 . | the classical interpretation of the wave function @xmath0 reveals an interesting operational aspect of the helmholtz spectra .
it is shown that the traditional sturm - liouville problem contains the simplest key to predict the squeezing effect for charged particle states .
pacs number(s ) : 02.70.hm , 41.75.-i , 03.65.ge _ preprint _ cinvestav : ciea - fis/97 - 07 |
in a hall - filled hubbard model @xcite with only nearest - neighbor ( nn ) hopping , the spin exchange between nn electrons is antiferromagnetic , leading to the antiferromagnetic insulating ground state , the nel state , on a square lattice . when the next - nearest - neighbor ( nnn ) hopping is present , there is also antiferromagnetic spin exchange between nnn electrons , which is destructive to the antiferromagnetic order .
this spin frustration leads to a transition from the nel state to a paramagnetic state at a critical on - site repulsion @xcite . a metallic antiferromagnetic
phase exists in a narrow region near this transition @xcite .
another possible ground state found in numerical studies @xcite due to spin frustration is the collinear antiferromagnetic ( caf ) state in which the spin configuration is antiferromagnetic along one axis and ferromagnetic along the other axis .
the transition from the nel state to the caf state in a heisenberg model on a square lattice with both nn spin exchange @xmath0 and nnn spin exchange @xmath1 was proposed at @xmath2 @xcite , and the exact phase diagram of this model is still under investigation . in the @xmath3-hubbard model ,
properties of the caf state are largely unknown . in this work ,
we study properties of the caf state in a half - filled hubbard model on a square lattice at zero temperature .
we find that there is a metal - insulator transition in the caf state .
the magnetic moment displays a kink at this transition point .
when @xmath4 is small , the caf state is metallic , and a van hove singularity may be close to the fermi energy . when @xmath4 is large , the caf state is an insulator .
the paramagnetic , antiferromagnetic , and caf states are ground states in different regions of the phase space .
there is a second - order transition between caf and paramagnetic phases , and a first - order transition between caf and antiferromagnetic phases .
based on these results , a zero - temperature phase diagram is obtained , in comparison with previous studies @xcite .
the hamiltonian of the hubbard model with both nn and nnn hoppings is given by @xmath5 where @xmath6 and @xmath7 are electron annihilation and creation operators at site @xmath8 with spin @xmath9 , and @xmath10 is the number operator . the first two terms on r .- h .- s .
( [ hubbard ] ) describe nn and nnn hoppings , and the last term describes the on - site repulsion .
the square lattice has total @xmath11 sites and the lattice constant is given by @xmath12 . in the large @xmath4 limit
, the nn hopping leads to antiferromagnetic spin exchange @xmath13 between nns , and similarly @xmath14 between nnns .
the energy per site of the classical antiferromagnetic state is @xmath15 , while it is @xmath16 in the classical caf state .
thus intuitively when @xmath17 or equivalently @xmath18 , the caf state is preferred over the antiferromagnetic state .
however , much is unknown when the repulsion @xmath4 is of the same order as @xmath19 or @xmath20 , which will be investigated in this work . in both antiferromagnetic and caf states
, the square lattice can be divided into two sublattices .
the order parameter is the same on each sublattice , but opposite on different sublattices .
it can be generally written as @xmath21 where @xmath22 is the coordinate of site @xmath8 and @xmath23 is the magnetic moment . in the antiferromagnetic state ,
the wavevector @xmath24 is given by @xmath25 ; in the caf state , @xmath26 or @xmath27 .
when @xmath28 , the order parameter is zero , and the system is in a paramagnetic state . in the mean - field approach , the on - site repulsion term in the hamiltonian can be approximated by @xmath29 and the mean - filed hamiltonian of the antiferromagnetic or caf states can be written as @xmath30 , \end{aligned}\ ] ] where @xmath31 , @xmath32 at half filling .
the mean - field hamiltonian can be diagonalized by a standard canonical transformation @xmath33
-{un\over 4}\left(n^2-m^2\right),\end{aligned}\ ] ] where the @xmath34-summation is over the first brillouin zone of a sublattice , the quasi - particle operators are given by @xmath35 , @xmath36 , @xmath37 , @xmath38 , and the coefficients are given by @xmath39,\ ] ] with @xmath40 .
the quasi - particles form two bands , with energies given by @xmath41 the magnetic moment @xmath42 can be determined self - consistently , @xmath43,\ ] ] where @xmath44 is the chemical potential .
this self - consistency equation can be further written as , @xmath45 equation ( [ m - eq ] ) can be solved together with the density equation @xmath46.\end{aligned}\ ] ] the total energy per site @xmath47 is given by @xmath48 -{u\over 4}(n^2-m^2 ) .
\label{energy}\end{aligned}\ ] ] the self - consistency equation ( [ m - eq ] ) is equivalent to the energy - extreme condition @xmath49 .
for @xmath50 , ( b ) magnetic moment @xmath42 versus repulsion @xmath4 for @xmath50 .
one kink of @xmath42 appears at the transition point @xmath51 , and the other kink appears at @xmath52 where there is a van hove singularity at the fermi surface.,width=321 ] in a caf state at half filling , e. g. with @xmath27 , from eq .
( [ band ] ) quasi - particle energies are given by @xmath53 the minimum of the upper band is given by @xmath54 at @xmath55 , and the maximum of the lower band is given by @xmath56 at @xmath57 .
therefore when @xmath58 , the caf state is metallic ; when @xmath59 , the system is an insulator with a band gap @xmath60 .
for @xmath50 , this metal - insulator transition takes place at a critical repulsion @xmath61 , as shown from band structures in fig .
[ band_mi](a ) . when @xmath62 , the caf state is an insulator ; when @xmath63 , it is a metal . at the metal - insulator transition point
, the magnetic moment @xmath42 displays a kink as shown in fig .
[ band_mi](b ) .
the other kink of @xmath42 in fig .
[ band_mi](b ) is due to the van hove singularity . from eq .
( [ band2 ] ) , when @xmath64 , there is always a van hove singularity in each band , located at @xmath65 and @xmath66 in upper and lower bands respectively , @xmath67 $ ] .
the density of states ( dos ) diverges logarithmically at these points . in the insulating caf state
, the chemical potential is always between the two bands at half filling , and van hove singularities are not important .
however , in the metallic caf state , the chemical potential can reach the van hove singularity in the upper band at certain repulsion @xmath68 , whereas the energy of the van hove singularity in the lower band is always less than chemical potential . at @xmath69 ,
the magnetic moment @xmath42 can exhibit either a kink as shown in fig .
[ van - hove ] ( a ) , or a jump as shown in fig .
[ van - hove ] ( b ) . for @xmath70 ,
the magnetic moment @xmath42 is continuous near @xmath69 , because the total energy @xmath47 as a function of @xmath42 has only one local minimum , as shown in the inset of fig .
[ van - hove ] ( a ) .
in contrast , for @xmath71 , two local minima of @xmath47 appear , as shown in the inset of fig .
[ van - hove ] ( b ) , and at @xmath69 the magnetic moment is discontinuous .
the van hove singularity at @xmath69 can be clearly seen from the band structure and the dos divergence in fig .
[ van - hove ] ( c ) and ( d ) . in our numerical calculation , the discontinuity of @xmath42 at @xmath69 appears when @xmath72 .
when @xmath73 , the magnetic moment @xmath42 exhibits a kink at @xmath69 . versus repulsion @xmath4 near @xmath69 in the caf state for ( a ) @xmath70 and ( b ) @xmath71 .
total energies per site versus @xmath42 are in the insets , where @xmath74 is the energy per site of the paramagnetic state .
band structures and dos at @xmath75 for ( c ) @xmath70 and ( d ) @xmath71.,width=325 ]
the caf state is not always energetically favored over antiferromagnetic or paramagnetic states .
it is well known that when there is only nn hopping , the ground state is the antiferromagnetic state .
when the nnn hopping is finite , electron spins are frustrated . in the strong coupling limit , @xmath76 and @xmath77 , in both caf and antiferromagnetic states ,
the lower band is fully occupied and the upper band is empty . from eq .
( [ band ] ) , the energy per site of caf state is approximately @xmath78 , and in the antiferromagnetic state @xmath79 .
therefore , when @xmath80 , the caf state has lower energy than the antiferromagnetic state , consistent with the @xmath81heisenberg - model result in ref .
@xcite . in the opposite limit
, @xmath82 and @xmath83 , the ground state is a metallic paramagnetic state .
a continuous phase transition between paramagnetic and caf states can take place at a critical repulsion @xmath84 which can be determined from eq .
( [ m - eq ] ) by setting @xmath85 , @xmath86 where @xmath87 .
the transition point between paramagnetic and antiferromagnetic states can also be determined similarly . when @xmath88 , the paramagnetic - antiferromagnetic phase transition takes place first with the increase of the repulsion @xmath4 ; when @xmath89 , the paramagnetic - caf transition occurs first . .
( b ) the energy of the antiferromagnetic state @xmath90 as a function of the magnetic moment @xmath91 for different repulsion strength for @xmath92 , with @xmath93 at the transition .
( c ) @xmath90 for @xmath94 , with @xmath95 at the transition .
the reference energy @xmath74 is the energy of the paramagnetic state for @xmath93.,width=325 ] the paramagnetic - antiferromagnetic transition was first found by lin and hirsch @xcite . in later studies , it was recognized that this transition can be either a first - order or second - order transition depending on the hopping ratio @xmath96 @xcite .
a narrow region of a metallic antiferromagnetic state may exist near the transition @xcite .
our numerical results show that the second - order transition takes place not only in the region with @xmath97 @xcite , but also in the region with @xmath98 , as can be seen from the vanishing magnetic moment @xmath91 of the antiferromagnetic state at the transition in fig .
[ neel_uc ] . when @xmath99 , the magnetic moment @xmath91 is finite at the transition , indicating that it is a first - order phase transition . near the second - order paramagnetic - antiferromagnetic transition
, there is a tiny region of metallic antiferromagnetic state ; near the first order transition , the antiferromagnetic state is always an insulator . -hubbard model at half filling .
afi denotes antiferromagnetic insulator , afm denotes antiferromagnetic metal , cafi denotes collinear antiferromagnetic insulator , cafm denotes collinear antiferromagnetic metal , and pm denotes paramagnetic metal.,width=302 ] by comparing energies of paramagnetic , antiferromagnetic , and caf states , we have obtained the mean - field phase diagram of the @xmath3-hubbard model at half filling as shown in fig .
[ phase_diagram ] .
these three phases meet at a tricritical point @xmath100 .
the paramagnetic state is always the ground state when the repulsion is small enough . when @xmath101 , there is a transition between paramagnetic and antiferromagnetic phases at a critical repulsion , which can be seen from the energy comparison shown in fig .
[ energy - fig](a ) .
when @xmath102 or @xmath103 , it is a second order transition ; when @xmath104 , it is a first - order transition , as shown in fig .
[ neel_uc ] . and antiferromagnetic - state energy @xmath90 for : ( a ) @xmath105 , ( b ) @xmath106 and ( c ) @xmath70 , where @xmath107 is the paramagnetic - state energy.,width=245 ] in the region with @xmath108 and @xmath109 , caf and antiferromagnetic states have close energies , and there is a reentrance from the caf state to the antiferromagnetic state as can be seen from the energy comparison shown in fig .
[ energy - fig](b ) .
there is a transition between the metallic caf and antiferromagnetic states at a critical repulsion @xmath110 larger than @xmath111 .
however the metallic antiferromagnetic region is very small .
when the repulsion further increases , the antiferromagnetic state becomes insulating , and eventually a transition between insulating antiferromagnetic and caf states takes place at another critical repulsion @xmath112 .
the transitions between antiferromagnetic and caf states at @xmath112 and @xmath110 are the first - order transitions , because antiferromagnetic and caf order parameters are discontinuous .
since the parameters @xmath4 , @xmath19 and @xmath20 are of the same order at these transitions , fluctuations may modify these phase boundaries . in the phase diagram in ref .
@xcite , the antiferromagnetic and caf phases are separated by a very small paramagnetic region in between . in ref .
@xcite , the caf - antiferromagnetic transition was found near @xmath113 without reentrance .
the detail of this transition needs to be resolved in future studies . for @xmath114 and @xmath109
, the caf state has always lower energy than the antiferromagnetic state , as shown in fig .
[ energy - fig](c ) . in the metallic caf state
, the magnetic moment @xmath42 has a kink for @xmath115 and a discontinuity for @xmath72 , when the van hove singularity is at the fermi surface .
the metal - insulator transition takes place inside the caf state at repulsion @xmath116 where also the magnetic moment @xmath42 has a kink .
the location of the caf phase in the mean - field phase diagram is in qualitative agreement with those obtained in other approaches @xcite .
we would like to stress that the mean - field method is more accurate when the on - site repulsion is relatively small .
it is inadequate for treating spin - liquid or superconducting states @xcite due to strong correlation .
spin frustration can significantly change the magnetic excitations in the antiferromagnetic state@xcite . beyond half filling , ferromagnetic state , incommensurate spin - density - wave state , and phase separation @xcite between these states
can also appear . in conclusion , we studied the caf state in a half - filled @xmath3 hubbard model with nn and nnn hoppings at zero temperature .
we found that a metal - insulator transition takes place inside the caf phase at a critical on - site repulsion @xmath116 . in the metallic caf state
, there is a kink or discontinuity in the magnetic moment when the van hove singularity is at the fermi surface . at zero temperature ,
the caf , antiferromagnetic , and paramagnetic states meet at a tricritical point @xmath100 .
there is a first - order caf - antiferromagnetic phase transition , and a second - order caf - paramagnetic phase transition .
the mean - field phase diagram of the @xmath3 hubbard model is obtained .
this work is supported by nsfc under grant no . 10674007 and 10974004 , and by chinese most under grant number 2006cb921402 .
j. hubbard , proc .
london , ser .
a * 276 * , 238 , ( 1963 ) .
h. q. lin and j. e. hirsch , phys . rev .
b * 35 * , 3359 ( 1987 ) . h. kondo and t. moriya , j. phys .
. jpn . * 65 * , 2559 ( 1996 ) ; h. kondo and t. moriya , _ ibid .
_ , * 67 * , 234 ( 1998 ) .
w. hofstetter and d. vollhardt , ann .
d. physik * 7 * , 48 ( 1998 ) .
d. duffy and a. moreo , phys .
b * 55 * , r676 ( 1997 ) .
i. yang , e. lange , and g. kotliar , phys .
b * 61 * , 2521 ( 2000 ) .
r. chitra , and g. kotliar , phys .
* 83 * , 2386 ( 1999 ) .
t. kashima and m. imada , j. phys .
soc . jpn . *
70 * , 3052 ( 2001 ) ; h. morita , s. watanabe , and m. imada , _ ibid_. * 71 * , 2109 ( 2002 ) ; t. mizusaki and m. imada phys .
b * 74 * , 014421 ( 2006 ) .
l. f. tocchio , f. becca , a. parola , and s. sorella , phys .
b * 78 * , 041101(r ) ( 2008 ) . a. h. nevidomskyy , c. scheiber , d. senechal , and a. m. s. tremblay , phys
b * 77 * , 064427 ( 2008 ) ; see also , s. r. hassan , b. davoudi , b. kyung , and a. m. s. tremblay , phys . rev .
b * 77 * , 094501 ( 2008 ) .
p. chandra , p. coleman and a. i. larkin , phys .
64 * , 88 ( 1990 ) .
q. p. li and r. joynt , phys .
b. * 49 * , 1632 ( 1994 ) .
b. valenzuela , m. a. h. vozmediano , and f. guinea , phys .
b * 62 * , 11312 ( 2000 ) .
a. avella , f. mancini , and r. mnzner , phys .
b * 63 * , 245117 ( 2001 ) .
h. taniguchi , y. morita , and y. hatsugai , phys .
b * 71 * 134417 ( 2005 ) .
h.yokoyama , m. ogata , and y. tanaka , j. phys .
. jpn . * 75 * , 114706 ( 2006 ) ; s. onari , h. yokoyama , and y. tanaka , physcia c , * 463 * -*465 * , 120 ( 2007 ) . g. i. japaridze , r. m. noack , d. baeriswyl , and l. tincani , phys .
b * 76 * , 115118 ( 2007 ) .
r. peters and t. pruschke , new j. phys .
* 11 * , 083022 ( 2009 ) .
p. delannoy , m. j. p. gingras , p. c. w. holdsworth , and a .- m .
s. tremblay , phys .
b * 79 * , 235130 ( 2009 ) .
p. a. igoshev , m. a. timirgazin , a. a. katanin , a. k. arzhnikov , and v. yu .
irkhin , phys . rev .
b * 81 * , 094407 ( 2010 ) . | in a half - filled hubbard model on a square lattice , the next - nearest - neighbor hopping causes spin frustration , and the collinear antiferromagnetic ( caf ) state appears as the ground state with suitable parameters .
we find that there is a metal - insulator transition in the caf state at a critical on - site repulsion .
when the repulsion is small , the caf state is metallic , and a van hove singularity can be close to the fermi surface , resulting in either a kink or a discontinuity in the magnetic moment . when the on - site repulsion is large , the caf state is a mott insulator . a first - order transition from the caf phase to the antiferromagnetic phase and a second - order phase transition from the caf phase to the paramagnetic phase
are obtained in the phase diagram at zero temperature . |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.