Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach

We use variational approach to analyze the evolution of the dynamics of a neutral s-wave superconductor between BCS and BEC regimes. We consider 2D case, when BCS–BEC crossover occurs already at weak coupling and is governed by the ratio of the two scales — the Fermi energy EF and the bound state en...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Физика низких температур
Datum:2018
Hauptverfasser: Chubukov, A.V., Mozyrsky, D.
Format: Artikel
Sprache:English
Veröffentlicht: Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України 2018
Schlagworte:
Online Zugang:https://nasplib.isofts.kiev.ua/handle/123456789/176148
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Назва журналу:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Zitieren:Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach / A.V. Chubukov, D. Mozyrsky // Физика низких температур. — 2018. — Т. 44, № 6. — С. 684-690. — Бібліогр.: 42 назв. — англ.

Institution

Digital Library of Periodicals of National Academy of Sciences of Ukraine
id nasplib_isofts_kiev_ua-123456789-176148
record_format dspace
spelling Chubukov, A.V.
Mozyrsky, D.
2021-02-03T19:01:53Z
2021-02-03T19:01:53Z
2018
Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach / A.V. Chubukov, D. Mozyrsky // Физика низких температур. — 2018. — Т. 44, № 6. — С. 684-690. — Бібліогр.: 42 назв. — англ.
0132-6414
PACS: 74.20.Fg, 74.25.Вt, 74.78.–w
https://nasplib.isofts.kiev.ua/handle/123456789/176148
We use variational approach to analyze the evolution of the dynamics of a neutral s-wave superconductor between BCS and BEC regimes. We consider 2D case, when BCS–BEC crossover occurs already at weak coupling and is governed by the ratio of the two scales — the Fermi energy EF and the bound state energy for two fermions in a vacuum, E₀. BCS and BEC regimes correspond to EF >> E₀ and EF << E₀, respectively. We compute the spectrum of low-energy bosonic excitations and show that the velocity of phase fluctuations remains υF / √2 through BCS–BEC crossover. We also discuss the topological Aφ ̇ term in the effective action.
We thank I. Aleiner, J.J. Dziarmaga, I. Martin, M. Stone, J. Sauls, D. Solenov, and G.E. Volovik for valuable discussions. The work at Los Alamos was performed under the auspices of the United States Department of Energy under Contract DE5AC52-06NA25396 and supported by LDRD and BES Grant No’s. 20170460ER and E3B7, respectively. The work by AVC was supported by the Office of Basic Energy Sciences U.S. Department of Energy under award DE-SC0014402. AVC is thankful to KITP at UCSB, where part of the work has been done. KITP is supported by NSF grant PHY-1125915.
en
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
Физика низких температур
Специальный выпуск К 90-летию со дня рождения A.A. Абрикосова
Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach
Article
published earlier
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
collection DSpace DC
title Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach
spellingShingle Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach
Chubukov, A.V.
Mozyrsky, D.
Специальный выпуск К 90-летию со дня рождения A.A. Абрикосова
title_short Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach
title_full Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach
title_fullStr Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach
title_full_unstemmed Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach
title_sort evolution of the dynamics of neutral superconductors between bcs and bec regimes: the variational approach
author Chubukov, A.V.
Mozyrsky, D.
author_facet Chubukov, A.V.
Mozyrsky, D.
topic Специальный выпуск К 90-летию со дня рождения A.A. Абрикосова
topic_facet Специальный выпуск К 90-летию со дня рождения A.A. Абрикосова
publishDate 2018
language English
container_title Физика низких температур
publisher Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
format Article
description We use variational approach to analyze the evolution of the dynamics of a neutral s-wave superconductor between BCS and BEC regimes. We consider 2D case, when BCS–BEC crossover occurs already at weak coupling and is governed by the ratio of the two scales — the Fermi energy EF and the bound state energy for two fermions in a vacuum, E₀. BCS and BEC regimes correspond to EF >> E₀ and EF << E₀, respectively. We compute the spectrum of low-energy bosonic excitations and show that the velocity of phase fluctuations remains υF / √2 through BCS–BEC crossover. We also discuss the topological Aφ ̇ term in the effective action.
issn 0132-6414
url https://nasplib.isofts.kiev.ua/handle/123456789/176148
citation_txt Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach / A.V. Chubukov, D. Mozyrsky // Физика низких температур. — 2018. — Т. 44, № 6. — С. 684-690. — Бібліогр.: 42 назв. — англ.
work_keys_str_mv AT chubukovav evolutionofthedynamicsofneutralsuperconductorsbetweenbcsandbecregimesthevariationalapproach
AT mozyrskyd evolutionofthedynamicsofneutralsuperconductorsbetweenbcsandbecregimesthevariationalapproach
first_indexed 2025-11-26T21:41:37Z
last_indexed 2025-11-26T21:41:37Z
_version_ 1850777784336515072
fulltext Low Temperature Physics/Fizika Nizkikh Temperatur, 2018, v. 44, No. 6, pp. 684–690 Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach Andrey V. Chubukov Department of Physics, University of Minnesota, Minneapolis MN 55455, USA E-mail: achubuko@umn.edu Dmitry Mozyrsky Theoretical Division (T-4), Los Alamos National Laboratory, Los Alamos NM 87545, USA E-mail: mozyrsky@lanl.gov Received January 24, 2018, published online April 25, 2018 We use variational approach to analyze the evolution of the dynamics of a neutral s-wave superconductor be- tween BCS and BEC regimes. We consider 2D case, when BCS–BEC crossover occurs already at weak coupling and is governed by the ratio of the two scales — the Fermi energy EF and the bound state energy for two fermi- ons in a vacuum, E0. BCS and BEC regimes correspond to EF >> E0 and EF << E0, respectively. We compute the spectrum of low-energy bosonic excitations and show that the velocity of phase fluctuations remains / 2Fv through BCS–BEC crossover. We also discuss the topological Aφ term in the effective action. PACS: 74.20.Fg Nanotubes; 74.25.Вt Thermodynamic properties; 74.78.–w Superconducting films and low-dimensional structures. Keywords: s-wave, superconductor, Berry phase. 1. Preface It is our great pleasure to present this article for a spe- cial issue in memory of Alexei Alexeevich Abrikosov. Alexei Alexeevich was one of the pioneers of field-theo- retical approach to superconductivity and the co-author of the book [1] which educated at least two generations of physicists. He is also the father of vortex physics in super- conductors. Our study is based on his works, and we hope it will be of interest to the readers of the special memorial issue of the Low Temperatures Physics. 2. Introduction In this paper, we discuss the evolution of the dynamic properties of a neutral superconductor between BCS re- gime, when bound pairs of fermions condense immediately once they form, and Bose–Einstein condensation (BEC) regime, when bound pairs of fermions form at a higher insT and condense at a smaller cT . Experimental evidence for preformed pairs has been reported for high- cT cuprates [2] and, more recently, for Fe-based superconductor 1eSe eF Tx x− (Ref. 3). We consider a 2D s-wave superconductor and assume a rotational symmetry and 2 / (2 )k m fermionic dispersion. In 2D BCS–BEC crossover can be analyzed already within weak coupling, when calculations are under control [4–6]. Indeed, in 2D systems, two fermions form a bound state already at arbitrary small attraction g , with energy 2/( )002 = 2 e N gE −Λ , where 0 = / (2 )N m π is the free parti- cle density of states per spin in 2D and Λ is the upper cut- off for the attraction [4–6]. In 3D systems, a bound state of two fermions emerges only once the interaction exceeds a certain cutoff, generally of the order of fermionic band- width [8]. The evolution of the static properties of a superconduc- tor between BCS and BEC regimes has been extensively discussed in the condensed matter context [4,5,7–23,42] and also for optical lattices of ultracold atoms [24,25]. The BCS–BEC crossover is determined by the interplay be- tween 0E and FE : the system is deep in the BCS regime when 0FE E and deep in the BEC regime when 0 FE E . In the BCS regime, the pairing instability temperature and superconducting cT are 1/2 0ins ( )FcT T E E≈  , and the pair- ing gap ∆ also scales as 1/2 0( )FE E . In the BEC regime © Andrey V. Chubukov and Dmitry Mozyrsky, 2018 Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach ins 0 0/ log ( / )FT E E E , c FT E , and ∆ still scales as 1/2 0( )FE E . The chemical potential 0( = 0) = FT E Eµ − , and it changes sign from positive to negative when 0E becomes larger than FE . Here we focus on the evolution of the dynamical prop- erties. We use variational approach, in which we assume that the order parameter ( , )∆ τr slowly fluctuates around its equilibrium value 0∆ . We obtain the quadratic form in the variations 0( , )∆ τ − ∆r and extract from it the spectrum of collective modes, including gapless Anderson–Bogo- lubov–Goldstone (AGB) phase mode. In the BCS regime, the velocity of AGB excitations, AGBv , is the same as the velocity of a sound wave in the normal state: = / 2AGB Fv v , where Fv is Fermi velocity. We argue that the value AGBv remains / 2Fv also in the BEC lim- it. Within this approach, we also identify the linear in fre- quency term, which corresponds to the ( , )i d d Aτ φ τ∫ r r term in the effective action. Such term is often referred to in the literature as the Berry phase term [26–31]. We obtain the variation of the prefactor A with respect to the variation of ∆. This allows us to express A as = / 2A n C+ , where n is the density (which may depend on time), up to a con- stant C. A constant C is irrelevant if the phase φ is defined globally, i.e., has a certain value at any point in space, be- cause then a prefactor can be pulled out from the d dτ∫ r , and the remaining term ( , )d dτφ τ∫ r r reduces to an irrele- vant boundary term, which does not affect the equation of motion. However, if the phase is not defined globally, as in the case of a moving Abrikosov vortex with coordinates ( )X τ and ( )Y τ , ( , ) ( )d d XY YXτφ τ ∝ −∫ r r   does not reduce to a boundary term and does affect the equation of motion. In particular, it gives rise to an effective Magnus force act- ing on a vortex [26–37]. To obtain A exactly, one needs an alternative approach, in which one expands the effective action in terms of time derivatives of the order parameter [38]. This approach yields = 0C in the absence of impurity scattering. The paper is organized as follows: in the next section we obtain, as a warm-up exercise, the expression for the condensation energy in the crossover between BCS and BEC regimes. In Sec. 4 we introduce the effective action of a superconductor in terms of its fluctuating order pa- rameter ( , )∆ τr . In Sec. 5 we derive the dispersion relation for the AGB mode by expanding in small variations of ∆ relative to its equilibrium value. In Sec. 6 we obtain the term linear in the time derivative of the phase of the order parameter (the Berry phase term). We summarize our re- sults in Sec. 7. 3. The condensation energy We first consider the case when the order parameter ( , )∆ τr is a constant 0∆ . In this situation the pairing Hamil- tonian has a conventional form ( )† † 0 , ,, ,= k k kk k k H c c c c↑ ↓↑ ↓ ε + +∑ ( )† † †* 0 0 ,, , , ,kk k k k c c c c ↑↑ − ↓ − ↓ + ∆ + ∆∑ (1) where operators † ,c σk are Fourier transforms of † ( )σψ r , etc. and = =−ξ ξ ε − µk k k , 2= / 2mεk k . The difference be- tween the ground state energy 0H〈 〉 in a superconductor (with the chemical potential 0= FE Eµ − ) and in a normal state (with the chemical potential 0 = FEµ ) is the conden- sation energy condE . The ground state energy is often separated into kinetic and potential energy parts, kin pot=scE E E+ , although this splitting is a bit elusive for a quantum system of interacting fermions as, e.g., fermionic self-energy, which is a potential energy of a fermion in a field created by other fermions, contributes to the fermionic density † ,,( ) = kkn k c c ααα 〈 〉∑ and, hence, to the kinetic energy. We follow earlier works and define the kinetic energy as kin = ( )kkE n kε∑ . The kinetic part of the ground state energy can be expressed via the normal Green’s function as kin 2 2 2 , 0 = 2 , iE ω  ω + ξ − ε   ω + ∆ + ξ   ∑ k k k k (2) where at = 0T and in thermodynamic limit 2 3 0 , = / (2 ) = / (2 )S d kd SN d d ω ω π ε ω π∑ ∫ ∫ k , where S is the area of a 2D sample. The potential energy is 0 pot 0 2 2 2 , 0 = .E ω ∆ −∆ ω + ∆ + ξ ∑ k k (3) For a system with weak attraction g , the self-consistent equation on 0∆ is 0 0 2 2 2 , 0 1= . g S ω ∆ ∆ ω + ∆ + ξ ∑ k k (4) Carrying out integration over the Matsubara frequency and over kξ up to the cutoff Λ , and using 0 0= exp ( 2/ )E N gΛ − , we obtain the algebraic equation [4–6] 2 2 0 0= 2 ,Eµ + ∆ − µ (5) which defines 0∆ in terms of the actual chemical potential µ and the two-particle bound state energy 0E . The self- consistency equation for µ follows from the condition that the total number of fermions, including bound pairs, is conserved [5]. This yields Low Temperature Physics/Fizika Nizkikh Temperatur, 2018, v. 44, No. 6 685 Andrey V. Chubukov and Dmitry Mozyrsky 2 2 0 = 2 .FEµ + ∆ + µ (6) The solution of Eqs. (5), (6) is 0= ,FE Eµ − 0 0= 2 .FE E∆ (7) Like we said, BCS–BEC crossover occurs when 0E be- comes comparable to the Fermi energy, FE . BCS behavior is realized when the bound state energy 0E is much smaller than FE , and BEC behavior is realized when 0 FE E . A negative µ at 0<FE E implies that the Fermi momentum Fk , defined as position of the minimum of the fermionic dispersion 2 2 0= ( )k kE ε − µ + ∆ , is zero [12]. When the particle density is small, FE is small compared to Λ , and the crossover occurs at small 0 /E Λ, i.e., already at weak coupling, when 0 1N g  . To obtain the condensation energy in BCS–BEC cross- over regime, we write 2 0 0 2 2 2 0 norm 0 0 2 ( ( )) = , (2 ) ( ) 2= . (2 ) ( ) sc id dE N S d dE N S i ε ω + ε − µ + ∆ε ω − π ω + ∆ + ε − µ ε ω ε π ω − ε − µ ∫ ∫ (8) Evaluating the integrals separately for 0>FE E , when > 0µ , and for 0<FE E , when < 0µ , and in both cases using Eq. (7) to relate µ with 0∆ , we obtain after a straight- forward algebra that the condensation energy 2 0 cond 0 0= , ,E SN f µµ − ∆  ∆ ∆  (9) where 2 21 ( 1 )( , ) = . 4 2 x x xf x y y + + + − (10) Using 2 1 = 2x x y+ + (Eq. (6)) we obtain 1( , ) = ( ). 4 f x y y y x+ − (11) Using next 2 2 0 0 0( ) = ( ) / = / = 1/ 4Fy x y E E− µ µ − µ ∆ ∆ , we find 1 1 1( , ) = = , 4 4 2 f x y + (12) Substituting into (9), we obtain that 2 0 cond 0 0 0= (2 ) = 2FE SN E E SN ∆ − − (13) in the whole BCS–BEC crossover region. A remark is in order about the order of integration over frequency and dispersion. The order does not matter as long as the integral over kε is taken within finite limits, i.e., one can do d dω ε∫ in any order and the result will be the same. The situation gets more tricky when the integra- tion over ε is extended to infinite limit. The most extreme case here is BCS limit, where µ ∆ , and the integration over =ξ ε − µ can be formally extended to d ∞ −∞ ξ∫ . In the BCS limit, the difference between µ and 0µ is irrelevant, to leading order in /∆ µ , and condE can be expressed as 2 2 2 cond 0 0 2 2 2 2 2 0 = 2 ( )( ) d dE SN ω ξ ω − ξ − ∆ π ω + ξ ω + ξ + ∆∫ . (14) At small ω and ξ , the would be divergence of the inte- grand is regularized by 0∆ , but at large ω and ξ , the inte- grand scales as 2 2 2 2 2( ) / ( )ω − ξ ω + ξ , i.e., the 2D integral is marginal by power counting. Such an integral is non- singular, but its value depends on how the integration is performed. If we treat ω and ξ as “equivalent” variables and evaluate the integral using polar coordinates, the inte- gral vanishes because of 2 2ω − ξ in the numerator. If we integrate over ω first or over ξ first, both times in infinite limits, the result is finite, but obviously of a different sign. Because at = 0T the integration over Matsubara frequency ω truly goes from −∞ to +∞ , while the integration over ξ actually holds in finite limits, which can only approximate- ly be extended to ±∞ , the correct way to evaluate the inte- gral in (14) is to integrate over frequency first. Integrating, we obtain 2 2 0 cond 0 0 2 2 2 2 2 0 1= 2 (| | ) dE SN ∆ξ − ∆ ξ + ∆ ξ + ξ + ∆ ∫ . (15) The integrand scales as 31/ | |ξ at large | |ξ , i.e., the in- tegral converges and can be evaluated in infinite limits. By rescaling 0= xξ ∆ , one can immediately verify that the integral does not depend on 0∆ , and (15) reduces to 2 2 2 2 cond 0 0 0 02 0 1= 1 = 21 dxE SN x x SN x ∞  − ∆ + − − ∆   + ∫ . (16) This agrees with (13). One can further check that the variation of the energy is exactly the same as the variation of the chemical potential 2 0 0 0 0= ( ) = 2 ( ) / 2F F F Fn n E N E E E E Nδµ µ − + − ≡ ∆ . As a result, E N− µ (equal to the Grand potential at = 0T ) does not change between the normal and the superconducting state. Note in passing that the self-consistency analysis can be straightforwardly extended to a finite T . In the BCS re- gime, the onset temperatures pT for the pairing and cT for superconductivity (i.e., for pair coherence) are comparable and both scale as 0FE E (more precisely, pT almost coin- cides with the mean-field cT , while the actual cT of Berezinskii–Kosterlitz–Thouless transition is smaller by a numerical factor). In the BEC regime the two temperatures differ strongly: 0 0/ log ( / )p F FT E E E E  , while c FT E , i.e., the ratio /c pT T vanishes at 0FE → (see, e.g., Ref. 6). When 0FE E , 0 ins= 1.76T∆ , like in BCS theory, when 0FE E , 0∆ is much larger than cT , but much smaller than pT . 686 Low Temperature Physics/Fizika Nizkikh Temperatur, 2018, v. 44, No. 6 Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach 4. The effective action The effective action for an order parameter of an s-wave superconductor can be obtained by departing from a mi- croscopic model with local four-fermion attractive interac- tion g− ( > 0g ) and decoupling four-fermion interaction via Hubbard–Stratonovich transformation [39], by intro- ducing the pairing field ( , )r∆ τ . This procedure is well documented, and we just quote the results. The partition function Z is expressed via the integral over the Grassmann fields as [ , ]= e ,SZ d d − ψ ψψ ψ∫ (17) where = ( , )αψ ψ τr and = ( , )αψ ψ τr are spin-full coordi- nate and time dependent Grassmann fields, and ( )[ , ] = ( , ) ( , ) [ , ] .S d d Hα τ αψ ψ τ ψ τ ∂ ψ τ + ψ ψ∫ r r r (18) Here τ is the imaginary (Matsubara) time = itτ and 2 †[ , ] = ( , ) ( , ) 2 H mσ σ   ∇ ψ ψ ψ τ − − µ ψ τ −       r r † †( , ) ( , ) ( , ) ( , ).g ↓ ↑↑ ↓ − ψ τ ψ τ ψ τ ψ τr r r r (19) The four-fermion interaction is decoupled by Hubbard–Stra- tonovich transformation 22 22 1e = e . 2 y yxax a dy a    − +     π ∫ (20) In our case we introduce two Hubbard–Stratonovich fields ( , )∆ τr and *( , )∆ τr and rewrite the partition function as ** [ , , , ]= e ,SZ d d d d − ψ ψ ∆ ∆ψ ψ ∆ ∆∫ (21) where now *[ , , , ] =S ψ ψ ∆ ∆ 2 *| ( , ) |= ( , ) ( , ) [ , , , ] ,d d H gα τ α  ∆ τ τ ψ τ ∂ ψ τ + + ψ ψ ∆ ∆     ∫ rr r r (22) and 2 *[ , , , ] = ( , ) ( , ) 2 H mσ σ   ∇ ψ ψ ∆ ∆ ψ τ − − µ ψ τ +      r r ( , ) ( , ) ( , ) ( , ) ( , ) ( , )∗ ↑ ↓ ↓ ↑ + ∆ τ ψ τ ψ τ + ∆ τ ψ τ ψ τ r r r r r r . (23) The action *[ , , , ]S ψ ψ ∆ ∆ can be re-expressed in a more compact form by introducing Gorkov–Nambu spinor †= [ , ]T↑ ↓ ψ ψ ψ . Then 2 * | ( , ) |[ , , , ] =S d d g ∆ τ ψ ψ ∆ ∆ τ −∫ rr 1( , ) ( , ; , ) ( , )d d d 'd G ' '−′ ′ ′− τ τ ψ τ τ τ ψ τ∫ ∫r r r r r r , (24) where the inverse Green’s function 1( , ; , )G '− ′τ τr r is given by 1 ˆ ˆ( , ; , ) = [ ( ) ( , , )] ( ) ( ),G ' K− τ′ ′ ′τ τ −∂ − − ∆ τ λ δ − δ τ − τr r r r r r (25) with 2 2 (1/ 2 ) 0ˆ ( ) = , 0 (1/ 2 ) m K m  − ∇ − µ    ∇ + µ  r and 0 ( , )ˆ ( , ) = . ( , ) 0∗ ∆ τ  ∆ τ   ∆ τ   r r r Integrating over ψ and ψ we then obtain ** [ , ]= e SZ d d − ∆ ∆∆ ∆∫ (26) and 2 * | ( , ) |[ , ] =S d d g ∆ τ ∆ ∆ τ −∫ rr 1ˆlog ( , ; , ).d d d 'd Tr G '−′ ′− τ τ τ τ∫ ∫r r r r (27) In Fourier space (momentum k and Matsubara frequency ω) 1 ( , )ˆ ( , ) = . ( , ) k k i k G k k i ∗ − ∗  ω − ξ ∆ ω  ω  ∆ ω ω + ξ  5. The dynamics of phase fluctuations We now expand the action in small variations of the or- der parameter around a constant value, ( , ) = ( , )δ∆ τ ∆ τ −r r 0 ( ) ( ).− ∆ δ δ τr . The action to the second order in δ∆ has been obtained in the BCS–BEC crossover regime in Ref. 40, and our expression for the quadratic form in δ∆ in the action agrees with theirs. The authors of Ref. 40, how- ever, didn’t extract the spectrum of the ABG mode from their data, nor they extracted the linear in Ω (i.e., φ ) term in the action. Collecting all second-order contributions and Fourier transforming to momentum and Matsubara fre- quency space, we obtain the variation of  in the form ( )2 2 , 1= | | | | h.c. , 2 A A B+ + − − + − Ω  δ δ∆ + δ∆ + δ∆ δ∆ + ∑ q  (28) where = ( , ) = ( , ) e , = ( , ) = ( , ) e i i i i d d t d d t − + Ωτ + − Ωτ − δ∆ δ∆ Ω τ δ∆ δ∆ δ∆ − −Ω τ δ∆ ∫ ∫ qr qr q r r q r r (29) and Low Temperature Physics/Fizika Nizkikh Temperatur, 2018, v. 44, No. 6 687 Andrey V. Chubukov and Dmitry Mozyrsky /2 /2 2 2 2 2 2 2 , 0 /2 0 /2 2 0 2 2 2 2 2 2 , 0 /2 0 /2 ( ( / 2) )( ( / 2) )1= , (( / 2) )(( / 2) ) = , (( / 2) )(( / 2) ) k i i A g B ± − ± ± ω ± − ± ω + − + ω ± Ω + ξ − ω Ω + ξ − ε ω ± Ω + ∆ + ξ ω Ω + ∆ + ξ ∆ ω + Ω + ∆ + ξ ω − Ω + ∆ + ξ ∑ ∑ k q k q k k q k q k k q k q   (30) ______________________________________________ where, we remind, 2= = / (2 )k k k mξ ε − µ − µ. The excitation spectrum is obtained from the condition 2=A A B+ − , or, equivalently 2 = 0. 2 2 2 A A A A A AB B+ − + − + −+ + −    + − −          (31) Evaluating the frequency and momentum integrals, we obtain 20 2 2 20 0 = 1 2 8 NA A B+ −   + µ  − Ω + +   ∆ µ + ∆   22 0 2 2 2 2 0 0 1 = m    ∆µ   + µ + +    µ + ∆ µ + ∆    q 2 20 2 00 , 4 F F F N E E E E m   = Ω +  +∆    q 0 0 2 2 00 = 1 ... = ..., 2 2 F F N N EA A B E E + −  + µ + + + +   +µ + ∆  0 0 2 2 00 1 1= , = = , 2 4 4 F N NA A i A A E E + −− Ω − − +µ + ∆ (32) where dots stand for Ω and q-dependent terms in ( ) / 2A A B+ −+ + , which we do not need. Substituting this into (31), we obtain from (31) 2 2 2 20 0 0 = 0. 16 ( ) 2 F F N E E E   Ω +  +    qv (33) Converting this last expression into real frequencies ( iΩ → − Ω), we see that the excitation spectrum has a gap- less mode, whose velocity remains / 2Fv , no matter what the ratio of 0 / FE E is. To see that this mode corresponds to phase fluctuations, we split the complex variations +δ∆ and −δ∆ into real and imaginary parts ±′δ∆ and ±′′δ∆ and re-express the action in Eq. (28) as 2 2 2 , 1= ( ) 2 q B BA A A A+ + − − − + +Ω    ′ ′ ′δ δ∆ δ∆ + δ∆ − +        ∑ 2 2 2 , 1 ( ) . 2 q B BA A A A+ − − − − + +Ω    ′′ ′′ ′′+ δ∆ δ∆ + δ∆ −        ∑ (34) This expression shows that there are gapped and gap- less modes. Suppose that the equilibrium value of 0=∆ ∆ is real. Then longitudinal variation ( , )r tδ∆ in coordinate is real and transverse one is imaginary. Using Eq. (29) we find that for longitudinal gap fluctuations –=+′ ′δ∆ −δ∆ and =+ −′′ ′′δ∆ δ∆ , while for transverse (phase) fluctuations –=+′ ′δ∆ −δ∆ and =+ −′′ ′′δ∆ δ∆ . Using now the fact that at = 0Ω and = 0q , = =A A B+ − , we immediately find from (31) that longitudinal fluctuations are gapped and phase fluctua- tions are gapless, as they should be. The 2q term in the integrand in Eq. (28) for δ deter- mines the phase stiffness. Taking 2q term from A A+ −+ in (29) and expressing ( , )r∆ τ as 0( , ) = eir ∇φ∆ τ ∆ r , such that 0( , ) = ( )qδ∆ Ω ∆ δ − ∇φq , we obtain 2 20= ( ) = ( ) . 8 4 FN End m m δ ∇φ ∇φ∫ r (35) We see that the phase stiffness also does not change in BCS–BEC crossover and remains the same as in BCS limit [41]. The expression for δ , Eqs. (28) and (30), have been obtained in [40], but the excitation spectrum have not been obtained there, although the dispersion of phase fluctua- tions at arbitrary 0 / FE E can be extracted from Eq. (26) in that paper. 6. The term linear in τ∂ φ in the action The prefactors A+ and A− contain the piece linear in Ω , with opposite signs. The corresponding term in the action is 0 linear 2 2 0 = 8 N iδ − × µ + ∆  ( ) ( )2 2 2 2 – – , ( ) ( ) ( ) ( ) = q + + Ω  ′ ′ ′′ ′′× Ω δ∆ − δ∆ + δ∆ − δ∆  ∑ 0 2 2 08 N i= − × µ + ∆ [ ]– – – – , ( )( ) ( )( ) . q + + + + Ω ′ ′ ′ ′ ′′ ′′ ′′ ′′× Ω δ∆ + δ∆ δ∆ − δ∆ + δ∆ + δ∆ δ∆ − δ∆∑ (36) For definiteness, let’s assume that ±δ∆ are real, i.e., =± +′δ∆ δ∆ . The transverse gap variation –+′ ′δ∆ − δ∆ can be expressed as the time variation of the phase of super- conducting gap 0 0( , ) = e = (1 ...)ir iφτ∆ τ ∆ ∆ + φτ +   , where ( , )φ = φ τr and = /φ ∂φ ∂τ . Taking the Fourier transform we find 688 Low Temperature Physics/Fizika Nizkikh Temperatur, 2018, v. 44, No. 6 Evolution of the dynamics of neutral superconductors between BCS and BEC regimes: the variational approach – 0= 2 sin .d+′ ′ ′ ′ ′δ∆ − δ∆ − φ∆ τ Ωτ τ∫ (37) The longitudinal gap variation can be approximated by its = 0q and = 0Ω value = ( , )d d′ ′δ∆ τδ∆ τ∫ r r . Substituting this into Eq. (36) we obtain 0 0 linear 2 2 0 = ( , ), 2 N iQ d d ∆ ′δ τ φδ∆ τ µ + ∆ ∫ r r  (38) where = sin . 2 d dQ ′Ω τ ′ ′Ωτ Ωτ π∫ (39) The integral contains the universal contribution = (2 / )Q π × ( )sin / = 1d× Ω Ω Ω∫ and parasitic high-energy contribu- tion, which vanishes under proper regularization. Substitut- ing = 1Q into (38) we obtain 0 0 linear 2 2 0 = ( , ). 2 N i d d ∆ ′δ τ φδ∆ τ µ + ∆ ∫ r r  (40) We now observe that to leading order in ( , )′δ∆ τr , 2 2 0 0( , ) /′∆ δ∆ τ µ + ∆r is the variation of 2 2| ( , ) |µ + ∆ τr over ∆, keeping µ constant. As the consequence, Eq. (40) can be viewed as the variation of 2 20 linear 0 ( , )= ( ) = , 2 N i d d C iA d d∂φ τ τ µ + ∆ + µ τφ   ∂τ∫ ∫ rr r  (41) where 2 20 0= ( ) , 2 N A C µ + ∆ + µ   (42) and ( )C µ is some unknown constant, which depends on µ. Using 2 2 0 0 0= 2 =Fn N E N  µ + ∆ + µ    , we can rewrite (42) as = ( ), 2 nA C+ µ (43) where ( )C µ is some other constant. This constant cannot be obtained using the variational approach. In Ref. 38 we obtained a systematic expansion of the action in terms of the derivatives of the order parameter over the imaginary time. In this approach we computed the A term explicitly and found that = 0C in the limit when the energy differ- ence between discrete levels in the vortex core well ex- ceeds a scattering rate due to impurities. 7. Conclusion In this paper we analyzed the evolution of the low-fre- quency dynamics of collective excitations of an s-wave neutral superconductor at zero temperature, between BCS and BEC regimes. The two regimes correspond to small and large ratio of 0 / FE E , respectively, where FE is the Fermi energy, and 0E is the bound state energy for two particles. In 2D, bound state develops already at weak coupl- ing, what allows one to analyze the crossover without in- cluding strong coupling renormalizations. We found that the phase velocity of the collective excitations remains / 2Fv through the BCS–BEC crossover. The supercon- ducting stiffness (the prefactor for 2( )∇φ term in the ac- tion) also does not change through the BCS–BEC crosso- ver and remains equal to 0 / (4 ) = / (8 )FN E m n m , as in BCS limit. The action also contains the term linear in time derivative of φ — the Berry phase term. When φ is defined globally, i.e., has a certain value at any point in space, such term reduces to a boundary term and does not contribute to equation of motion, if the density is homogeneous. If = ( )n n τ , the linear in φ term in the action is linear = ( / 2) ( )i d d nτ τ φ∫ ∫r  , where = ( , )φ φ τr . For the case of a moving vortex with coordinates ( )X τ and ( )Y τ , φ is not uniquely defined at the center of the vortex core, i.e., for x X→ and y Y→ . In this situation ( )d d d XY YXτφ ∝ τ −∫ ∫r    , which does not reduce to the boundary term. Then the Berry phase term in the action contributes to equation of motion even when the prefactor is a constant, and it becomes relevant to find the exact prefactor = / 2A n C+ . The exact prefactor A has been computed by expanding the action in time derivatives of the order parameter [38]. This calculation yields = 0C . Our results for the expansion of the effective action in terms of ′δ∆ can be straightforwardly extended to other symmetries of the order parameter and to non-Galilean- invariant dispersion. One can also use our formulas to ob- tain terms with higher-order derivatives (these are terms with higher powers of Ω and q in the integrand of Eq. (28)). Acknowledgements We thank I. Aleiner, J.J. Dziarmaga, I. Martin, M. Stone, J. Sauls, D. Solenov, and G.E. Volovik for valuable discus- sions. The work at Los Alamos was performed under the auspices of the United States Department of Energy under Contract DE5AC52-06NA25396 and supported by LDRD and BES Grant No’s. 20170460ER and E3B7, respectively. The work by AVC was supported by the Office of Basic Energy Sciences U.S. Department of Energy under award DE-SC0014402. AVC is thankful to KITP at UCSB, where part of the work has been done. KITP is supported by NSF grant PHY-1125915. ________ 1. A.A. Abrikosov, L.P. Gor’kov, and I.E. Dzyaloshinski, Methods of Quantum Field Theory on Statistical Physics, Dover Publications, New York (1975). 2. see e.g., J.C. Campuzano, M.R. Norman, and M. Randeria, in: Physics of Superconductors, Vol. II, K.H. Bennemann and J.B. Ketterson (eds.), Springer, Berlin (2004), p. 167; Lu Li, Yayu Wang, and N.P. Ong, Phys. Rev. B 87, 056502 (2013) and references therein; D.S. Dessau, private communication. 3. K. Okazaki, Sci. Rep. 4, 4109 (2014); S. Kasahara, PNAS 111, 16309 (2014); S. Rinott, Sci. Adv. 3, e1602372 (2017). Low Temperature Physics/Fizika Nizkikh Temperatur, 2018, v. 44, No. 6 689 https://doi.org/10.1126/sciadv.1602372 Andrey V. Chubukov and Dmitry Mozyrsky 4. K. Miyake, Prog. Theor. Phys. 69, 1794 (1983). 5. M. Randeria, J.-M. Duan, and L.-Y. Shieh, Phys. Rev. Lett. 62, 981 (1989); Phys. Rev. B 41, 327 (1990). 6. A.V. Chubukov, I. Eremin, and D.V. Efremov, Phys. Rev. B 93, 174516 (2016). 7. V.M. Loktev, R.M. Quick, and S.G. Sharapov, Phys. Rep. 349, 1 (2001). 8. C.A.R. Sa de Melo, M. Randeria, and J.R. Engelbrecht, Phys. Rev. Lett. 71, 3202 (1993); J.R. Engelbrecht, M. Randeria, and C.A.R. Sa de Melo, Phys. Rev. B 55, 15153 (1997). 9. P. Noziéres and S. Schmitt-Rink, J. Low Temp. Phys. 59, 195 (1985). 10. M. Drechsler and W. Zwerger, Ann. Phys. 1, 15 (1992). 11. Y. Ohashi and A. Griffin, Phys. Rev. A 67, 033603 (2003) 12. Y.L. Loh, M. Randeria, N. Trivedi, C.-C. Chang, and R. Scalettar, arXiv:1507.05641. 13. see M.Y. Kagan, Modern Trends in Superconductivity and Superfluidity, Lecture Notes in Physics, 874, Springer, Heidelberg/Germany (2013) and references therein. 14. V.B. Geshkenbein, L.B. Ioffe, and A.I. Larkin, Phys. Rev. B 55, 3173 (1997). 15. B. Tobijaszewska and R. Micnas, Acta Physica Polonica A 97, 393 (2000). 16. L. Benfatto, A. Toschi, S. Caprara, and C. Castellani, Phys. Rev. B 64, 140506(R) (2001); L. Benfatto, S. Caprara, C. Castellani, A. Paramekanti, and M. Randeria, Phys. Rev. B 63, 174513 (2001); L. Benfatto, A. Toschi, and S. Caprara, Phys. Rev. B 69, 184510 (2004). 17. M. Iskin and C.A.R. Sa de Melo, Phys. Rev. B 74, 144517 (2006). 18. L. Fanfarillo, L. Benfatto, S. Caprara, C. Castellani, and M. Grilli, Phys. Rev. B 79, 172508 (2009). 19. D. Innocenti, N. Poccia, A. Ricci, A. Valletta, S. Caprara, A. Perali, and A. Bianconi, Phys. Rev. B 82, 184528 (2010); A. Bianconi, D. Innocenti, A. Valletta, and A. Perali, J. Phys.: Conf. Ser. 529, 012007 (2014). 20. J. Ranninger, J.M. Robin, and M. Eschrig, Phys. Rev. Lett. 74, 4027 (1995). 21. A. Levchenko, M.R. Norman, and A.A. Varlamov, Phys. Rev. B 83, 020506(R) (2011). 22. L. Fanfarillo, L. Benfatto, and C. Castellani, Phys. Rev. B 85, 024507 (2012). 23. A. Guidini and A. Perali, Supercond. Sci. Technol. 27, 124002 (2014). 24. Q. Chen, J. Stajic, S. Tan, and K. Levin, Phys. Rep. 412, 188 (2005); Y. Shin, C.H. Schunck, A. Schirotzek, and W. Ketterle, Nature 451, 689693 (2007); S. Giorgini, L.P. Pitaevskii, and S. Stringari, Rev. Mod. Phys. 80, 1215 (2008); F. Chevy and C. Mora, Rep. Prog. Phys. 73, 112401 (2010); Z. Shen, L. Radzihovsky, and V. Gurarie, Phys. Rev. Lett. 109, 245302 (2012); M. Randeria and E. Taylor, Ann. Rev. Condens. Matter Phys. 5, 209232 (2014). 25. J.P. Gaebler, Nat. Phys. 6, 569 (2010). 26. G.E. Volovik, The Universe in a Helium Droplet, Clarendon Press, Oxford (2003). 27. P. Ao and D.J. Thouless, Phys. Rev. Lett. 93, 090403-1 (2004); J.R. Aitchison, P. Ao, D.J. Thouless, and X.-M. Zhu, Phys. Rev. B 51, 6531 (1995). 28. J. Dziarmaga, Phys. Rev. B 53, 6572 (1996). 29. A. van Otterlo, M.V. Feigelman, V.B. Geshkenbein, and G. Blatter, Phys. Rev. Lett. 75, 3736 (1995); A. van Otterlo, D.S. Golubev, A.D. Zaikin, and G. Blatter, Eur. Phys. J. B 10, 131 (1999); 30. E. Simanek, Phys. Rev. B 52, 10336 (1995). 31. M. Stone, Phys. Rev. B 54, 13222 (1996). 32. N.B. Kopnin, G.E. Volovik, and U. Parts, Europhys. Lett. 32, 651 (1995). 33. J. Bardeen, R. Kummel, A.E. Jacobs, and L. Tewordt, Phys. Rev. 187, 556 (1969). 34. J. Bardeen and M.J. Stephen, Phys. Rev. 140, A 1197 (1965); P. Nozieres and W.F. Vinen, Philos. Mag. 14, 667 (1996). 35. N.B. Kopnin and V.E. Kravtsov, JETP Lett. 23, 578 (1976); N.B. Kopnin and M.M. Salomaa, Phys. Rev. B 44, 9667 (1991). 36. G.E. Volovik, JETP Lett. 65, 676 (1997). 37. F. Gaitan, Phys. Rev. B 51, 9061 (1995). 38. D. Mozyrsky, A.V. Chubukov, and M. Stone, to be submitted. 39. R.L. Stratonovich, Sov. Phys. Dokl. 2, 416 (1958); J. Hubbard, J. Phys. Rev. Lett. 3, 77 (1959); A. Altland and B. Simons, Condensed Matter Field Theory, University Press, Cam- bridge (2006). 40. L. He, H. Lu, G. Cao, H. Hu, and X.-J. Liu, Phys. Rev. A 92, 023620 (2015). 41. D.J. Scalapino, S.R. White, and S. Zhang, Phys. Rev. B 47, 7995 (1993); L. Benfatto, A. Toschi, S. Caprara, and C. Castellani, Phys. Rev. B 64, 140506(R) (2001); L. Benfatto, S. Caprara, C. Castellani, A. Paramekanti, and M. Randeria, Phys. Rev. B 63, 174513 (2001); S.G. Sharapov, V.P. Gusynin, and H. Beck, Eur. Phys. J. B 30, 45 (2002); L. Benfatto, A. Toschi, and S. Caprara, Phys. Rev. B 69, 184510 (2004). 42. Y. Lemonik, I.L. Aleiner, and B.L. Altshuler, arXiv:1505.04820. ___________________________ 690 Low Temperature Physics/Fizika Nizkikh Temperatur, 2018, v. 44, No. 6 https://doi.org/10.1143/PTP.69.1794 https://doi.org/10.1103/PhysRevLett.62.981 https://doi.org/10.1103/PhysRevB.41.327 https://doi.org/10.1103/PhysRevB.93.174516 https://doi.org/10.1016/S0370-1573(00)00114-9 https://doi.org/10.1103/PhysRevLett.71.3202 https://doi.org/10.1103/PhysRevB.55.15153 https://doi.org/10.1007/BF00683774 https://doi.org/10.1002/andp.19925040105 https://doi.org/10.1103/PhysRevA.67.033603 https://doi.org/10.1103/PhysRevB.55.3173 https://doi.org/10.1103/PhysRevB.55.3173 https://doi.org/10.12693/APhysPolA.97.393 https://doi.org/10.1103/PhysRevB.64.140506 https://doi.org/10.1103/PhysRevB.64.140506 https://doi.org/10.1103/PhysRevB.63.174513 https://doi.org/10.1103/PhysRevB.63.174513 https://doi.org/10.1103/PhysRevB.69.184510 https://doi.org/10.1103/PhysRevB.74.144517 https://doi.org/10.1103/PhysRevB.79.172508 https://doi.org/10.1103/PhysRevB.82.184528 https://doi.org/10.1088/1742-6596/529/1/012007 https://doi.org/10.1088/1742-6596/529/1/012007 https://doi.org/10.1103/PhysRevLett.74.4027 https://doi.org/10.1103/PhysRevB.85.024507 https://doi.org/10.1088/0953-2048/27/12/124002 https://doi.org/10.1016/j.physrep.2005.02.005 https://doi.org/10.1103/RevModPhys.80.1215 https://doi.org/10.1088/0034-4885/73/11/112401 https://doi.org/10.1103/PhysRevLett.109.245302 https://doi.org/10.1146/annurev-conmatphys-031113-133829 https://doi.org/10.1146/annurev-conmatphys-031113-133829 https://doi.org/10.1103/PhysRevLett.93.090403 https://doi.org/10.1103/PhysRevB.51.6531 https://doi.org/10.1103/PhysRevB.53.6572 https://doi.org/10.1103/PhysRevLett.75.3736 https://doi.org/10.1007/s100510050836 https://doi.org/10.1007/s100510050836 https://doi.org/10.1103/PhysRevB.52.10336 https://doi.org/10.1103/PhysRevB.54.13222 https://doi.org/10.1209/0295-5075/32/8/006 https://doi.org/10.1103/PhysRev.187.556 https://doi.org/10.1103/PhysRev.187.556 https://doi.org/10.1080/14786436608211964 https://doi.org/10.1103/PhysRevB.44.9667 https://doi.org/10.1134/1.567405 https://doi.org/10.1103/PhysRevB.51.9061 https://doi.org/10.1103/PhysRevLett.3.77 https://doi.org/10.1103/PhysRevA.92.023620 https://doi.org/10.1103/PhysRevB.64.140506 https://doi.org/10.1103/PhysRevB.63.174513 https://doi.org/10.1140/epjb/e2002-00356-9 https://doi.org/10.1103/PhysRevB.69.184510 1. Preface 2. Introduction 3. The condensation energy 4. The effective action 5. The dynamics of phase fluctuations 6. The term linear in in the action 7. Conclusion Acknowledgements