Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:cond-mat/0311374 · Physical Review B

Berezinskii-Kosterlitz-Thouless transition and BCS-Bose crossover in the two-dimensional attractive Hubbard model

N. Dupuis

Published Oct 13, 2004• 26 citations
arXivHTMLPDF

Abstract

We study the two-dimensional attractive Hubbard model using the mapping onto the half-filled repulsive Hubbard model in a uniform magnetic field coupled to the fermion spins. The low-energy effective action for charge and pairing fluctuations is obtained in the hydrodynamic regime. We recover the action of a Bose superfluid where half the fermion density is identified as the conjugate variable of the phase of the superconducting order parameter. By integrating out charge fluctuations, we obtain a phase-only action. In the zero-temperature superconducting state, this action describes a collective phase mode smoothly evolving from the Anderson-Bogoliubov mode at weak coupling to the Bogoliubov mode of a Bose superfluid at strong coupling. At finite temperature, the phase-only action can be used to extract an effective $XY$ model and thus obtain the Berezinskii-Kosterlitz-Thouless (BKT) phase transition temperature. We also identify a renormalized classical regime of superconducting fluctuations above the BKT phase transition, and a regime of incoherent pairs at higher temperature. Special care is devoted to the nearly half-filled case where the symmetry of the order parameter is enlarged to SO(3) due to strong $\mathbf{q}=(\ensuremath{\pi},\ensuremath{\pi})$ charge fluctuations. The low-energy effective action is then an SO(3) nonlinear $\ensuremath{\sigma}$ model with a (symmetry breaking) magnetic field proportional to the doping. In the strong-coupling limit, the attractive Hubbard model can be mapped onto the Heisenberg model in a magnetic field, which reduces to the quantum $XY$ model (except for a weak magnetic field, i.e., in the low-density limit of the attractive model). In the low-density limit, the Heisenberg model allows one to recover the action of a Bose superfluid, including the ${(\mathbf{\ensuremath{\nabla}}\ensuremath{\rho})}^{2}$ term (with $\ensuremath{\rho}$ the density), and in turn the Gross-Pitaevskii equation.

DOI
10.1103/physrevb.70.134502

Related papers

Effective action for superfluid Fermi systems in the strong-coupling limit

N. Dupuis

We derive the low-energy effective action for three-dimensional superfluid Fermi systems in the strong-coupling limit, where superfluidity originates from Bose-Einstein condensation of composite bosons. Taking into account density and pairing fluctuations on the same footing, we show that the effective action involves only the fermion density ${\ensuremath{\rho}}_{\mathbf{r}}$ and its conjugate variable, the phase ${\ensuremath{\theta}}_{\mathbf{r}}$ of the pairing order parameter ${\ensuremath{\Delta}}_{\mathbf{r}}$. We recover the standard action of a Bose superfluid of density ${\ensuremath{\rho}}_{\mathbf{r}}∕2$, where the bosons have a mass ${m}_{B}=2m$ and interact via a repulsive contact potential with amplitude ${g}_{B}=4\ensuremath{\pi}{a}_{B}∕{m}_{B},{a}_{B}=2a$ ($a$ the $s$-wave scattering length associated to the fermion-fermion interaction in vacuum). For lattice models, the derivation of the effective action is based on the mapping of the attractive Hubbard model onto the Heisenberg model in a uniform magnetic field, and a coherent state path integral representation of the partition function. The effective description of the Fermi superfluid in the strong-coupling limit is a Bose-Hubbard model with an intersite hopping amplitude ${t}_{B}=J∕2$ and an on-site repulsive interaction ${U}_{B}=2Jz$, where $J=4{t}^{2}∕U$ ($t$ and $\ensuremath{-}U$ are the intersite hopping amplitude and the on-site attraction in the (fermionic) Hubbard model, $z$ the number of nearest-neighbor sites).

arXiv:cond-mat/0412563•Physical Review A•Jul 11, 2005•5 citations

Effective action for superconductors and BCS-Bose crossover

S. De Palo, Claudio Castellani, C. Di Castro +1 more

A standard perturbative expansion around the mean-field solution is used to derive the low-energy effective action for superconductors at $T=0.$ Taking into account the density fluctuations at the outset we get the effective action where the density $\ensuremath{\rho}$ is the conjugated momentum to the phase $\ensuremath{\theta}$ of the order parameter. In the hydrodynamic regime, the dynamics of the superconductor is described by a time-dependent nonlinear Schr\"odinger (TDNLS) equation for the field $\ensuremath{\Psi}(x)=\sqrt{\ensuremath{\rho}/2}{e}^{i\ensuremath{\theta}}.$ The evolution of the density fluctuations in the crossover from weak-coupling (BCS) to strong-coupling (Bose condensation of localized pairs) superconductivity is discussed for the attractive Hubbard model. In the bosonic limit, the TDNLS equation reduces to the Gross-Pitaevskii equation for the order parameter, as in the standard description of superfluidity. The conditions under which a phase-only action can be derived in the presence of a long-range interaction to describe the physics of the superconductivity of ``bad metals'' are discussed.

arXiv:cond-mat/9901305•Physical review. B, Condensed matter•Jul 1, 1999•89 citations

BCS-BEC crossover and effects of density fluctuations in a two-component Fermi gas described by the three-dimensional attractive Hubbard model

Hiromasa Tamaki, Yoji Ohashi, K. Miyake

We investigate the superfluid phase transition in a gas of Fermi atoms loaded on a three-dimensional optical lattice. When the lattice potential is strong, this system can be well described by an attractive Hubbard model. In this model, we calculate the superfluid phase transition temperature ${T}_{\mathrm{c}}$, including both superfluid and (spin and charge) density fluctuations within the self-consistent $t$-matrix theory and fluctuation exchange approximation, respectively. Since we treat these fluctuations in a consistent manner, our theory satisfies the required particle-hole symmetry over the entire BCS--Bose-Einstein-condensation (BEC) crossover region. We show that charge density fluctuations compete against superfluid fluctuations near the half-filling, leading to the suppression of ${T}_{\mathrm{c}}$. As a result, the maximum ${T}_{\mathrm{c}}$ is obtained away from the half-filling. Since the strong density fluctuations originate from the nesting property of the Fermi surface at the half-filling (which is absent in a uniform gas with no lattice potential), our results would be useful in considering lattice effects on strong-coupling superfluidity.

arXiv:0803.2084•Physical Review A•Jun 18, 2008•22 citations

On the theory of 2D superconductivity: The peculiarities of the Berezinskii–Kosterlitz–Thouless transition and crossover from local to Cooper pairs (Review Article)

В. М. Локтев, Volodymyr Turkowski

We have analyzed superconducting properties of a strongly coupled two-dimensional (2D) superconductor with a boson-exchange attraction in the s- and d-wave pairing channels, with a focus on the properties near the Berezinskii–Kosterlitz–Thouless (BKT) transition, and with a more general goal to give a picture of the physics of cuprates at different dopings and temperatures. In the s-channel, we have found that superconductivity exists at all dopings and there is a crossover from the Bardeen–Cooper–Schrieffer (BCS) phase of large Cooper pairs to a state with local pairs, or Bose–Einstein condensation (BEC) phase, at all coupling strengths and bosonic frequencies. On the other hand, in the d-wave channel, we have found that superconductivity sets in at a finite doping and is suppressed at lower dopings due to a reduced, doping-dependent effective coupling. Moreover, the BCS–BEC crossover is absent in the d-wave channel. One rather unexpected result we found is an oscillatory dependence of the BKT critical temperature on doping in the d-pairing channel that is a result of an oscillatory dependence of the superconducting stiffness on the chemical potential. Obtained results may help to better understand superconducting properties of 2D materials, including cuprates, and might be used in modern technological applications.

Low Temperature Physics•Nov 1, 2026•0 citations

The Berezinskii–Kosterlitz–Thouless transition in imbalanced 2D Fermi gases: The role of fluctuations

Jacques Tempere, Serghei N. Klimin, J. T. Devreese

Abstract The physics of two-dimensional (2D) quantum gases can be revealed in strongly confining optical lattices. Upon cooling, 2D bosonic quantum gases as well as Fermi gases where pairing is present, become superfluid. The superfluid-to-normal transition is no longer governed by the presence or absence of a condensate, but by the Berezinskii–Kosterlitz–Thouless (BKT) mechanism: above a critical temperature, vortices and antivortices proliferate and destroy phase coherence. We investigate the BKT transition for superfluid 2D Fermi gases in the whole range of the BCS–BEC crossover, from weakly bound Cooper pairing (the BCS state), up to strongly bound molecules (the BEC state). Using a path-integral description, we then focus on the case of imbalanced gases: when the number of ‘spin-up’ and ‘spin-down’ fermions that form the pair is no longer equal. When an excess of one spin species exists, pairing is frustrated and the vortex energetics is strongly affected, influencing the KT mechanism. In the present work we are concentrated on the effect of both phase and amplitude fluctuations on phase diagrams of the fermion system. The amplitude fluctuations only slightly influence the BKT phase transition temperature. However, they lead to a substantial modification of the complete phase diagram for the Fermi gas in 2D with respect to that obtained taking into account only phase fluctuations.

Physica C Superconductivity•Feb 24, 2010•3 citations

Renormalization of the superfluid density in the two-dimensional BCS-BEC crossover

Giacomo Bighin, Luca Salasnich

We analyze the theoretical derivation of the beyond-mean-field equation of state for two-dimensional gas of dilute, ultracold alkali-metal atoms in the Bardeen–Cooper–Schrieffer (BCS) to Bose–Einstein condensate (BEC) crossover. We show that at zero temperature our theory — considering Gaussian fluctuations on top of the mean-field equation of state — is in very good agreement with experimental data. Subsequently, we investigate the superfluid density at finite temperature and its renormalization due to the proliferation of vortex–antivortex pairs. By doing so, we determine the Berezinskii–Kosterlitz–Thouless (BKT) critical temperature — at which the renormalized superfluid density jumps to zero — as a function of the inter-atomic potential strength. We find that the Nelson–Kosterlitz criterion overestimates the BKT temperature with respect to the renormalization group equations, this effect being particularly relevant in the intermediate regime of the crossover.

arXiv:1710.11171•International Journal of Modern Physics B•Mar 7, 2018•7 citations
We study the two-dimensional attractive Hubbard model using the mapping onto the half-filled repulsive Hubbard model in a uniform magnetic field coupled to the fermion spins. The low-energy effective action for charge and pairing fluctuations is obtained in the hydrodynamic regime. We recover the action of a Bose superfluid where half the fermion density is identified as the conjugate variable of the phase of the superconducting order parameter. By integrating out charge fluctuations, we obtain a phase-only action. In the zero-temperature superconducting state, this action describes a collective phase mode smoothly evolving from the Anderson-Bogoliubov mode at weak coupling to the Bogoliubov mode of a Bose superfluid at strong coupling. At finite temperature, the phase-only action can be used to extract an effective XYXYXY model and thus obtain the Berezinskii-Kosterlitz-Thouless (BKT) phase transition temperature. We also identify a renormalized classical regime of superconducting fluctuations above the BKT phase transition, and a regime of incoherent pairs at higher temperature. Special care is devoted to the nearly half-filled case where the symmetry of the order parameter is enlarged to SO(3) due to strong q=(\ensuremathπ,\ensuremathπ)\mathbf{q}=(\ensuremath{\pi},\ensuremath{\pi})q=(\ensuremathπ,\ensuremathπ) charge fluctuations. The low-energy effective action is then an SO(3) nonlinear \ensuremathσ\ensuremath{\sigma}\ensuremathσ model with a (symmetry breaking) magnetic field proportional to the doping. In the strong-coupling limit, the attractive Hubbard model can be mapped onto the Heisenberg model in a magnetic field, which reduces to the quantum XYXYXY model (except for a weak magnetic field, i.e., in the low-density limit of the attractive model). In the low-density limit, the Heisenberg model allows one to recover the action of a Bose superfluid, including the (\ensuremath∇\ensuremathρ)2{(\mathbf{\ensuremath{\nabla}}\ensuremath{\rho})}^{2}(\ensuremath∇\ensuremathρ)2 term (with \ensuremathρ\ensuremath{\rho}\ensuremathρ the density), and in turn the Gross-Pitaevskii equation.
We derive the low-energy effective action for three-dimensional superfluid Fermi systems in the strong-coupling limit, where superfluidity originates from Bose-Einstein condensation of composite bosons. Taking into account density and pairing fluctuations on the same footing, we show that the effective action involves only the fermion density ${\ensuremath{\rho}}_{\mathbf{r}}$ and its conjugate variable, the phase ${\ensuremath{\theta}}_{\mathbf{r}}$ of the pairing order parameter ${\ensuremath{\Delta}}_{\mathbf{r}}$. We recover the standard action of a Bose superfluid of density ${\ensuremath{\rho}}_{\mathbf{r}}∕2$, where the bosons have a mass mB=2m{m}_{B}=2mmB​=2m and interact via a repulsive contact potential with amplitude ${g}_{B}=4\ensuremath{\pi}{a}_{B}∕{m}_{B},{a}_{B}=2a$ (aaa the sss-wave scattering length associated to the fermion-fermion interaction in vacuum). For lattice models, the derivation of the effective action is based on the mapping of the attractive Hubbard model onto the Heisenberg model in a uniform magnetic field, and a coherent state path integral representation of the partition function. The effective description of the Fermi superfluid in the strong-coupling limit is a Bose-Hubbard model with an intersite hopping amplitude tB=J∕2{t}_{B}=J∕2tB​=J∕2 and an on-site repulsive interaction UB=2Jz{U}_{B}=2JzUB​=2Jz, where J=4t2∕UJ=4{t}^{2}∕UJ=4t2∕U (ttt and $\ensuremath{-}U$ are the intersite hopping amplitude and the on-site attraction in the (fermionic) Hubbard model, zzz the number of nearest-neighbor sites).
A standard perturbative expansion around the mean-field solution is used to derive the low-energy effective action for superconductors at T=0.T=0.T=0. Taking into account the density fluctuations at the outset we get the effective action where the density $\ensuremath{\rho}$ is the conjugated momentum to the phase $\ensuremath{\theta}$ of the order parameter. In the hydrodynamic regime, the dynamics of the superconductor is described by a time-dependent nonlinear Schr\"odinger (TDNLS) equation for the field $\ensuremath{\Psi}(x)=\sqrt{\ensuremath{\rho}/2}{e}^{i\ensuremath{\theta}}.$ The evolution of the density fluctuations in the crossover from weak-coupling (BCS) to strong-coupling (Bose condensation of localized pairs) superconductivity is discussed for the attractive Hubbard model. In the bosonic limit, the TDNLS equation reduces to the Gross-Pitaevskii equation for the order parameter, as in the standard description of superfluidity. The conditions under which a phase-only action can be derived in the presence of a long-range interaction to describe the physics of the superconductivity of ``bad metals'' are discussed.
We investigate the superfluid phase transition in a gas of Fermi atoms loaded on a three-dimensional optical lattice. When the lattice potential is strong, this system can be well described by an attractive Hubbard model. In this model, we calculate the superfluid phase transition temperature Tc{T}_{\mathrm{c}}Tc​, including both superfluid and (spin and charge) density fluctuations within the self-consistent ttt-matrix theory and fluctuation exchange approximation, respectively. Since we treat these fluctuations in a consistent manner, our theory satisfies the required particle-hole symmetry over the entire BCS--Bose-Einstein-condensation (BEC) crossover region. We show that charge density fluctuations compete against superfluid fluctuations near the half-filling, leading to the suppression of Tc{T}_{\mathrm{c}}Tc​. As a result, the maximum Tc{T}_{\mathrm{c}}Tc​ is obtained away from the half-filling. Since the strong density fluctuations originate from the nesting property of the Fermi surface at the half-filling (which is absent in a uniform gas with no lattice potential), our results would be useful in considering lattice effects on strong-coupling superfluidity.
We have analyzed superconducting properties of a strongly coupled two-dimensional (2D) superconductor with a boson-exchange attraction in the s- and d-wave pairing channels, with a focus on the properties near the Berezinskii–Kosterlitz–Thouless (BKT) transition, and with a more general goal to give a picture of the physics of cuprates at different dopings and temperatures. In the s-channel, we have found that superconductivity exists at all dopings and there is a crossover from the Bardeen–Cooper–Schrieffer (BCS) phase of large Cooper pairs to a state with local pairs, or Bose–Einstein condensation (BEC) phase, at all coupling strengths and bosonic frequencies. On the other hand, in the d-wave channel, we have found that superconductivity sets in at a finite doping and is suppressed at lower dopings due to a reduced, doping-dependent effective coupling. Moreover, the BCS–BEC crossover is absent in the d-wave channel. One rather unexpected result we found is an oscillatory dependence of the BKT critical temperature on doping in the d-pairing channel that is a result of an oscillatory dependence of the superconducting stiffness on the chemical potential. Obtained results may help to better understand superconducting properties of 2D materials, including cuprates, and might be used in modern technological applications.
Abstract The physics of two-dimensional (2D) quantum gases can be revealed in strongly confining optical lattices. Upon cooling, 2D bosonic quantum gases as well as Fermi gases where pairing is present, become superfluid. The superfluid-to-normal transition is no longer governed by the presence or absence of a condensate, but by the Berezinskii–Kosterlitz–Thouless (BKT) mechanism: above a critical temperature, vortices and antivortices proliferate and destroy phase coherence. We investigate the BKT transition for superfluid 2D Fermi gases in the whole range of the BCS–BEC crossover, from weakly bound Cooper pairing (the BCS state), up to strongly bound molecules (the BEC state). Using a path-integral description, we then focus on the case of imbalanced gases: when the number of ‘spin-up’ and ‘spin-down’ fermions that form the pair is no longer equal. When an excess of one spin species exists, pairing is frustrated and the vortex energetics is strongly affected, influencing the KT mechanism. In the present work we are concentrated on the effect of both phase and amplitude fluctuations on phase diagrams of the fermion system. The amplitude fluctuations only slightly influence the BKT phase transition temperature. However, they lead to a substantial modification of the complete phase diagram for the Fermi gas in 2D with respect to that obtained taking into account only phase fluctuations.
We analyze the theoretical derivation of the beyond-mean-field equation of state for two-dimensional gas of dilute, ultracold alkali-metal atoms in the Bardeen–Cooper–Schrieffer (BCS) to Bose–Einstein condensate (BEC) crossover. We show that at zero temperature our theory — considering Gaussian fluctuations on top of the mean-field equation of state — is in very good agreement with experimental data. Subsequently, we investigate the superfluid density at finite temperature and its renormalization due to the proliferation of vortex–antivortex pairs. By doing so, we determine the Berezinskii–Kosterlitz–Thouless (BKT) critical temperature — at which the renormalized superfluid density jumps to zero — as a function of the inter-atomic potential strength. We find that the Nelson–Kosterlitz criterion overestimates the BKT temperature with respect to the renormalization group equations, this effect being particularly relevant in the intermediate regime of the crossover.