Pseudogap regime of the unitary Fermi gas with lattice auxiliary-field quantum Monte Carlo in the continuum limit

S. Jensen Center for Theoretical Physics, Sloane Physics Laboratory, Yale University, New Haven, Connecticut 06520, USA Department of Physics, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801, USA    C. N. Gilbreth Quantinuum, Broomfield, Colorado 80021, USA    Y. Alhassid Center for Theoretical Physics, Sloane Physics Laboratory, Yale University, New Haven, Connecticut 06520, USA
Abstract

The unitary Fermi gas (UFG) is a strongly correlated system of two-species (spin-1/2) fermions with a short-range attractive interaction modeled by a contact interaction and has attracted much interest across different disciplines. The UFG is considered a paradigm for strongly correlated superfluids and has been investigated extensively, with generally good agreement found between theory and experiment. However, the extent of a pseudogap regime above the critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT for superfluidity, in which pairing correlations persist, is still debated both theoretically and experimentally. Here we study thermodynamic properties of the UFG across the superfluid phase transition using finite-temperature lattice auxiliary-field quantum Monte Carlo (AFMC) methods in the canonical ensemble of fixed particle numbers. We present results for the condensate fraction, a model-independent energy-staggering pairing gap, the spin susceptibility, and a free energy-staggering pairing gap. We extrapolate our lattice AFMC results to the continuous time and continuum limits, thus removing the systematic error associated with the finite filling factor of previous AFMC studies. Applying a finite-size scaling analysis to the condensate fraction results, we determine the critical temperature to be Tc=0.16(1)TFsubscript𝑇𝑐0.161subscript𝑇𝐹T_{c}=0.16(1)\,T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.16 ( 1 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. For the largest particle number studied N=114𝑁114N=114italic_N = 114, the energy-staggering pairing gap, which provides a direct signature of a pseudogap regime, is suppressed above a pairing scale temperature of T0.2TFsuperscript𝑇0.2subscript𝑇𝐹T^{*}\approx 0.2\,T_{F}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. The spin susceptibility displays moderate suppression above Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT with a spin gap temperature of Ts0.2TFsubscript𝑇𝑠0.2subscript𝑇𝐹T_{s}\approx 0.2\,T_{F}italic_T start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. We also calculate a free energy-staggering pairing gap, which shows substantially reduced statistical errors when compared with the energy-staggering gap, allowing for a clear signature of pairing correlations in the finite-size system. All results indicate that the pseudogap regime is narrow, with pseudogap signatures emerging at temperatures below T0.2TFsuperscript𝑇0.2subscript𝑇𝐹T^{*}\approx 0.2\,T_{F}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. The reduced statistical errors of the free energy gap enable an extrapolation at low temperatures, allowing an estimate of the zero-temperature pairing gap ΔE=0.576(24)ϵFsubscriptΔ𝐸0.57624subscriptitalic-ϵ𝐹\Delta_{E}=0.576(24)\,\epsilon_{F}roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT = 0.576 ( 24 ) italic_ϵ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT.

I Introduction

The unitary Fermi gas (UFG), with diverging s-wave scattering length, is of great interest in diverse areas of physics as it provides a well-defined paradigm for strongly correlated systems relevant to the study of nuclei, quark matter, neutron stars, and atomic Fermi gases. It undergoes a U(1)𝑈1U(1)italic_U ( 1 ) phase transition from a normal strongly correlated gas to an s-wave superfluid Leggett (2006); Bloch et al. (2008); Giorgini et al. (2008); Inguscio et al. (2008); Zwerger (2012); Levin et al. (2012); Svistunov et al. (2015); Pitaevskii and Stringari (2016) below a critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT. The UFG is realized in the middle of a crossover from a weakly interacting Fermi gas described by the Bardeen-Cooper-Schrieffer (BCS) Bardeen et al. (1957) theory to a weakly repulsive Bose-Einstein condensate (BEC) Zwerger (2012); Eagles (1969); Leggett (1980); Nozières and Schmitt-Rink (1985) (for T<Tc𝑇subscript𝑇𝑐T<T_{c}italic_T < italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT in the superfluid phase). The UFG has remarkable properties as one of the highest-temperature superfluids in units of the Fermi temperature TFsubscript𝑇𝐹T_{F}italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT Randeria (2010). It exhibits a nearly perfect fluid behavior Enss (2012); Bluhm et al. (2017); Nishida and Abuki (2005), and can be realized in ultracold atomic Fermi gas experiments with Li6superscriptLi6{}^{6}\textrm{Li}start_FLOATSUPERSCRIPT 6 end_FLOATSUPERSCRIPT Li and K40superscriptK40{}^{40}\textrm{K}start_FLOATSUPERSCRIPT 40 end_FLOATSUPERSCRIPT K Ketterle and Zwierlein (2008); Hu et al. (2022).

There is no known controlled expansion in a small parameter (around some nearby exact solution) to approach the superfluid transition of the UFG. It is therefore difficult to theoretically address this system on a quantitative level. The UFG, due in part to its simplicity and broad interest, has become a testing ground for strongly correlated many-body methods. Successful approaches in advancing our understanding of the UFG include the epsilon expansion Nishida and Son (2006), the Luttinger-Ward approach Haussmann et al. (2007, 2009); Zwerger (2016); Frank et al. (2018), conformal field theory methods Nishida and Son (2007a, b), quantum critical approaches Nikolić and Sachdev (2007); Enss (2012), non-self consistent diagrammatic resummation methods Perali et al. (2002a); Ohashi and Griffin (2002); Chen et al. (2005), and various quantum Monte Carlo methods Gilbreth and Alhassid (2013); Bulgac et al. (2006, 2008); Drut et al. (2012); Akkineni et al. (2007); Burovski et al. (2006a, b, 2008); Goulko and Wingate (2010); Van Houcke et al. (2011); Hu et al. (2010); Lee and Schäfer (2006); Endres et al. (2013).

The UFG has been extensively investigated experimentally in recent decades using ultracold Fermi gases  Regal et al. (2004); Zwierlein et al. (2004); Kinast et al. (2004); Chin et al. (2004); Zwierlein et al. (2005); Nascimbene et al. (2010); Nascimbène et al. (2011); Ku et al. (2012); Zwerger (2012); Mukherjee et al. (2017), but important open problems remain. A topic that has generated considerable interest is the existence and extent of a pseudogap regime above the critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT for superfluidity, in which there is no off-diagonal-long-range order (ODLRO) but pairing correlations persist with a gapped single-particle spectrum near the Fermi surface.

The extent of pairing correlations in the UFG has been debated in recent years. For low-temperatures (T0𝑇0T\to 0italic_T → 0), studies of pairing properties demonstrated good agreement between theory and experiment Shin et al. (2007); Gezerlis and Carlson (2008); Carlson et al. (2003); Chang et al. (2004); Carlson and Reddy (2005, 2008); Carlson et al. (2011). However, at finite temperature, and in particular across the superfluid critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT, our understanding is more limited. While there has been substantial progress in our understanding of the normal phase both theoretically Kashimura et al. (2012a, b); van Wyk et al. (2016); Perali et al. (2002b, 2011a); Palestini et al. (2012a); Tajima et al. (2016); Pini et al. (2019); Perali et al. (2002a, 2011b); Palestini et al. (2012b); Pantel et al. (2014); Tajima et al. (2014); Santos (1994); Jankó et al. (1997); Sewer et al. (2002); Chien et al. (2010); Chien and Levin (2010); Wlazłowski et al. (2013a); Magierski et al. (2009, 2011); Enss and Haussmann (2012); Jensen et al. (2020a) and experimentally Gaebler et al. (2010); Ku et al. (2012); Nascimbene et al. (2010); Nascimbène et al. (2011); Sommer et al. (2011); Wulin et al. (2011), our understanding of pairing correlations (even qualitatively) above Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT is still incomplete. For reviews of the pseudogap regime in the UFG, see Refs. Chen et al. (2005); Chen and Wang (2014); Mueller (2017); Jensen et al. (2019).

In this work, we calculate several thermodynamic properties of the spin-balanced homogeneous UFG using lattice auxiliary-field quantum Monte Carlo (AFMC) in the canonical ensemble of fixed particle numbers. In particular, we calculate the condensate fraction, a model-independent energy-staggering pairing gap, the spin susceptibility, and the free energy-staggering pairing gap for N=40,66𝑁4066N=40,66italic_N = 40 , 66 and 114114114114 particles. All properties are calculated in the continuum limit in both imaginary time and lattice spacing, thus removing systematic errors of a finite-filling factor of our previous work Jensen et al. (2020a) and allowing direct comparison with experiments.

Applying finite-size scaling to the condensate fraction data, we determine the superfluid critical temperature of the UFG to be Tc=0.16(1)TFsubscript𝑇𝑐0.161subscript𝑇𝐹T_{c}=0.16(1)T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.16 ( 1 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. The canonical ensemble framework enables the calculation of an energy-staggering pairing gap and a free energy-staggering pairing gap without the need for a numerically challenging analytic continuation. In particular, the free energy-staggering gap has remarkably small statistical errors, allowing for an extrapolation from finite temperature to estimate accurately the T=0𝑇0T=0italic_T = 0 energy-staggering pairing gap ΔE/εF=0.576(24)subscriptΔ𝐸subscript𝜀𝐹0.57624\Delta_{E}/\varepsilon_{F}=0.576(24)roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT / italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT = 0.576 ( 24 ) (where εFsubscript𝜀𝐹\varepsilon_{F}italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT is the Fermi energy). From the pairing gaps for our largest particle number simulated, N=114𝑁114N=114italic_N = 114, we estimate an upper bound on the pairing temperature scale associated with these observables to be T0.2TFsuperscript𝑇0.2subscript𝑇𝐹T^{*}\approx 0.2T_{F}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT (this pairing temperature decreases with increasing particle number N𝑁Nitalic_N). We find a similar value for the spin gap temperature of the spin susceptibility (for N=114𝑁114N=114italic_N = 114), Ts0.2TFsubscript𝑇𝑠0.2subscript𝑇𝐹T_{s}\approx 0.2T_{F}italic_T start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. These results for Tsuperscript𝑇T^{*}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT and Tssubscript𝑇𝑠T_{s}italic_T start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT indicate a narrow temperature pseudogap regime for the UFG.

The outline of this paper is as follows. In Sec. II we provide background material for the UFG and discuss progress toward understanding the pseudogap regime, with a focus on recent lattice Monte Carlo studies. In Sec. III we describe our canonical-ensemble AFMC method, and the recent optimization algorithms that allow for large lattice simulations. In Sec. IV we present the AFMC results of this work, and in Sec. V we provide concluding remarks.

II BCS-BEC crossover and the UFG

The BCS-BEC crossover in three spatial dimensions is a continuous crossover (at temperature T=0𝑇0T=0italic_T = 0) as a function of (kFa)1superscriptsubscript𝑘𝐹𝑎1(k_{F}a)^{-1}( italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT italic_a ) start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT (where a𝑎aitalic_a is the s-wave scattering length and kFsubscript𝑘𝐹k_{F}italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT is the Fermi wavenumber) from a weakly attractive Fermi gas described by BCS theory to a system of weakly interacting bosons described by BEC theory Zwerger (2012); Giorgini et al. (2008). As temperature is increased, the gas exhibits a U(1)𝑈1U(1)italic_U ( 1 ) (3D XY) superfluid phase transition into a normal disordered phase. The BCS regime corresponds to (kFa)1similar-tosuperscriptsubscript𝑘𝐹𝑎1(k_{F}a)^{-1}\sim-\infty( italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT italic_a ) start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT ∼ - ∞ where the normal phase can be described by Fermi liquid theory. The BEC regime, corresponding to (kFa)1+similar-tosuperscriptsubscript𝑘𝐹𝑎1(k_{F}a)^{-1}\sim+\infty( italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT italic_a ) start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT ∼ + ∞, is a weakly interacting Bose gas of tightly bound dimers with binding energy ϵb=2/(ma2)subscriptitalic-ϵ𝑏superscriptPlanck-constant-over-2-pi2𝑚superscript𝑎2\epsilon_{b}=\hbar^{2}/(ma^{2})italic_ϵ start_POSTSUBSCRIPT italic_b end_POSTSUBSCRIPT = roman_ℏ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT / ( italic_m italic_a start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ) where m𝑚mitalic_m is the mass of a single fermion. In the BCS regime, for T<Tc𝑇subscript𝑇𝑐T<T_{c}italic_T < italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT there are Cooper pairs of spatial extent much greater than the interparticle separation ρ1/3superscript𝜌13\rho^{-1/3}italic_ρ start_POSTSUPERSCRIPT - 1 / 3 end_POSTSUPERSCRIPT where ρ=N/V𝜌𝑁𝑉\rho=N/Vitalic_ρ = italic_N / italic_V is the particle density for N𝑁Nitalic_N fermions in a spatial volume V𝑉Vitalic_V. In the BEC regime, the pairs are tightly bound dimers of size much smaller than the interparticle separation. In the unitary limit, defined by (kFa)1=0superscriptsubscript𝑘𝐹𝑎10(k_{F}a)^{-1}=0( italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT italic_a ) start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT = 0 (i.e., infinite scattering length), the two-particle bound state is a zero-energy resonance. In this regime, the size of the Cooper pairs is comparable to the interparticle separation and the system is strongly correlated.

II.1 Scattering amplitude and the continuum model

We consider a dilute Fermi gas at low temperature. The scattering between two particles with wavevectors 𝕜1subscript𝕜1\mathbb{k}_{1}blackboard_k start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT and 𝕜2subscript𝕜2\mathbb{k}_{2}blackboard_k start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT is well described by the low-momentum s-wave expansion of the inverse of the scattering amplitude f(k)𝑓𝑘f(k)italic_f ( italic_k ),

f1(k)=1/aik+rek2/2+,superscript𝑓1𝑘1𝑎𝑖𝑘subscript𝑟𝑒superscript𝑘22f^{-1}(k)=-1/a-ik+r_{e}k^{2}/2+\ldots\;,italic_f start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT ( italic_k ) = - 1 / italic_a - italic_i italic_k + italic_r start_POSTSUBSCRIPT italic_e end_POSTSUBSCRIPT italic_k start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT / 2 + … , (1)

where 𝕜𝕜\mathbb{k}blackboard_k is the relative momentum wavevector 𝕜=𝕜1𝕜2𝕜subscript𝕜1subscript𝕜2\mathbb{k}=\mathbb{k}_{1}-\mathbb{k}_{2}blackboard_k = blackboard_k start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT - blackboard_k start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT, resubscript𝑟𝑒r_{e}italic_r start_POSTSUBSCRIPT italic_e end_POSTSUBSCRIPT the effective range, and \dots refers to higher-order corrections in k2superscript𝑘2k^{2}italic_k start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT. This expansion holds for 6Li and 40K, where the low-energy interaction at large distance is well described with the van der Waals potential V(r)=C6/r6𝑉𝑟subscript𝐶6superscript𝑟6V(r)=-C_{6}/r^{6}italic_V ( italic_r ) = - italic_C start_POSTSUBSCRIPT 6 end_POSTSUBSCRIPT / italic_r start_POSTSUPERSCRIPT 6 end_POSTSUPERSCRIPT with a microscopic interparticle interaction range of R0=(mC6/2)1/4subscript𝑅0superscript𝑚subscript𝐶6superscriptPlanck-constant-over-2-pi214R_{0}=(mC_{6}/\hbar^{2})^{1/4}italic_R start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = ( italic_m italic_C start_POSTSUBSCRIPT 6 end_POSTSUBSCRIPT / roman_ℏ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 1 / 4 end_POSTSUPERSCRIPT (Planck-constant-over-2-pi\hbarroman_ℏ is Planck’s constant) in the limits R0<<1/kF,λTmuch-less-thansubscript𝑅01subscript𝑘𝐹subscript𝜆𝑇R_{0}<<1/k_{F},\lambda_{T}italic_R start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT < < 1 / italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT , italic_λ start_POSTSUBSCRIPT italic_T end_POSTSUBSCRIPT. Here λTsubscript𝜆𝑇\lambda_{T}italic_λ start_POSTSUBSCRIPT italic_T end_POSTSUBSCRIPT is the thermal de Broglie wavelength λT=2π2/mkBTsubscript𝜆𝑇2𝜋superscriptPlanck-constant-over-2-pi2𝑚subscript𝑘𝐵𝑇\lambda_{T}=\sqrt{2\pi\hbar^{2}/mk_{B}T}italic_λ start_POSTSUBSCRIPT italic_T end_POSTSUBSCRIPT = square-root start_ARG 2 italic_π roman_ℏ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT / italic_m italic_k start_POSTSUBSCRIPT italic_B end_POSTSUBSCRIPT italic_T end_ARG (kBsubscript𝑘𝐵k_{B}italic_k start_POSTSUBSCRIPT italic_B end_POSTSUBSCRIPT is the Boltzmann constant), and the Fermi wavenumber is given by kF=(3π2ρ)1/3subscript𝑘𝐹superscript3superscript𝜋2𝜌13k_{F}=(3\pi^{2}\rho)^{1/3}italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT = ( 3 italic_π start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ρ ) start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT. See Ref. Khuri et al. (2009) for a detailed discussion of the low-momentum scattering expansion. Though the effective range parameter resubscript𝑟𝑒r_{e}italic_r start_POSTSUBSCRIPT italic_e end_POSTSUBSCRIPT is often of the order R0subscript𝑅0R_{0}italic_R start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT (for a broad Feshbach resonance |re|Rosimilar-tosubscript𝑟𝑒subscript𝑅𝑜|r_{e}|\sim R_{o}| italic_r start_POSTSUBSCRIPT italic_e end_POSTSUBSCRIPT | ∼ italic_R start_POSTSUBSCRIPT italic_o end_POSTSUBSCRIPT) this is not guaranteed to hold Castin and Werner (2012); Werner and Castin (2012). The unitary limit is reached when the modulus of the scattering amplitude is maximized, saturating the constraint imposed by unitarity, from the optical theorem, Imf(k)=k|f(k)|2Im𝑓𝑘𝑘superscript𝑓𝑘2\textrm{Im}f(k)=k|f(k)|^{2}Im italic_f ( italic_k ) = italic_k | italic_f ( italic_k ) | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT, and leading to f(k)=1/ik𝑓𝑘1𝑖𝑘f(k)=-1/ikitalic_f ( italic_k ) = - 1 / italic_i italic_k. We thus define the UFG by the conditions: R0<<1/kFmuch-less-thansubscript𝑅01subscript𝑘𝐹R_{0}<<1/k_{F}italic_R start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT < < 1 / italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT, kFa0subscript𝑘𝐹𝑎0k_{F}a\rightarrow 0italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT italic_a → 0, rekF<<1much-less-thansubscript𝑟𝑒subscript𝑘𝐹1r_{e}k_{F}<<1italic_r start_POSTSUBSCRIPT italic_e end_POSTSUBSCRIPT italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT < < 1, and R0<<λTmuch-less-thansubscript𝑅0subscript𝜆𝑇R_{0}<<\lambda_{T}italic_R start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT < < italic_λ start_POSTSUBSCRIPT italic_T end_POSTSUBSCRIPT.

These conditions are met with cold atomic Fermi gases using Feshbach resonance techniques to tune an external magnetic field so that a closed channel bound state becomes resonant with open channel scattering (see Refs. Chin et al. (2004, 2010); Giorgini et al. (2008)). Near a broad Feshbach resonance, many properties of the gas can be understood with a single-channel model using a contact interaction between spin-up and spin-down fermions (corresponding to an atomic hyperfine state manifold with two relevant states). More specifically, we model a system of Nsubscript𝑁N_{\uparrow}italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT spin-up, and Nsubscript𝑁N_{\downarrow}italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT spin-down, fermions in a three-dimensional cubic box of side length L𝐿Litalic_L with periodic boundary conditions. The continuum Hamiltonian of this system is given by

H^=szd3rψ^sz(𝕣)(222m)ψ^sz(𝕣)+V0d3rψ^(𝕣)ψ^(𝕣)ψ^(𝕣)ψ^(𝕣),^𝐻subscriptsubscript𝑠𝑧superscript𝑑3𝑟subscriptsuperscript^𝜓subscript𝑠𝑧𝕣superscriptPlanck-constant-over-2-pi2superscript22𝑚subscript^𝜓subscript𝑠𝑧𝕣subscript𝑉0superscript𝑑3𝑟subscriptsuperscript^𝜓𝕣subscriptsuperscript^𝜓𝕣subscript^𝜓𝕣subscript^𝜓𝕣\hat{H}=\sum_{s_{z}}\int d^{3}r\hat{\psi}^{\dagger}_{s_{z}}(\mathbb{r})\left(-% \frac{\hbar^{2}\nabla^{2}}{2m}\right)\hat{\psi}_{s_{z}}(\mathbb{r})\\ +V_{0}\int d^{3}r\hat{\psi}^{\dagger}_{\uparrow}(\mathbb{r})\hat{\psi}^{% \dagger}_{\downarrow}(\mathbb{r})\hat{\psi}_{\downarrow}(\mathbb{r})\hat{\psi}% _{\uparrow}(\mathbb{r})\;,start_ROW start_CELL over^ start_ARG italic_H end_ARG = ∑ start_POSTSUBSCRIPT italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT ∫ italic_d start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT italic_r over^ start_ARG italic_ψ end_ARG start_POSTSUPERSCRIPT † end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( blackboard_r ) ( - divide start_ARG roman_ℏ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ∇ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG start_ARG 2 italic_m end_ARG ) over^ start_ARG italic_ψ end_ARG start_POSTSUBSCRIPT italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( blackboard_r ) end_CELL end_ROW start_ROW start_CELL + italic_V start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT ∫ italic_d start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT italic_r over^ start_ARG italic_ψ end_ARG start_POSTSUPERSCRIPT † end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT ( blackboard_r ) over^ start_ARG italic_ψ end_ARG start_POSTSUPERSCRIPT † end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT ( blackboard_r ) over^ start_ARG italic_ψ end_ARG start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT ( blackboard_r ) over^ start_ARG italic_ψ end_ARG start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT ( blackboard_r ) , end_CELL end_ROW (2)

where ψ^sz(𝕣)subscriptsuperscript^𝜓subscript𝑠𝑧𝕣\hat{\psi}^{\dagger}_{s_{z}}(\mathbb{r})over^ start_ARG italic_ψ end_ARG start_POSTSUPERSCRIPT † end_POSTSUPERSCRIPT start_POSTSUBSCRIPT italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( blackboard_r ) and ψ^sz(𝕣)subscript^𝜓subscript𝑠𝑧𝕣\hat{\psi}_{s_{z}}(\mathbb{r})over^ start_ARG italic_ψ end_ARG start_POSTSUBSCRIPT italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( blackboard_r ) are, respectively, creation and annihilation operators for a fermion at position 𝕣𝕣\mathbb{r}blackboard_r with spin projection szsubscript𝑠𝑧s_{z}italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT. The contact interaction model is ill-defined without regularization. The model can be regularized with an ultra-violet (UV) cutoff ΛΛ\Lambdaroman_Λ in momentum space.

II.2 The pseudogap regime

A pseudogap regime is known to exist in high-Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT superconductors. For instance, the specific heat is strongly suppressed above Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT in contrast to conventional superconductors which show Fermi liquid behavior for low temperatures above Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT Dagotto (1994); Timusk and Statt (1999). The mechanism leading to the pseudogap regime of the cuprates is still not fully understood Randeria (2010); Dagotto (1994); Timusk and Statt (1999) as the complexity, competing orders, and strongly correlated nature of the problem are difficult to address with theoretical approaches. The BCS-BEC crossover is also strongly correlated for |kFa|>1subscript𝑘𝐹𝑎1|k_{F}a|>1| italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT italic_a | > 1 but the ordering is well understood to be a U(1)𝑈1U(1)italic_U ( 1 ) superfluid phase transition with no competing orderings and the appropriate microscopic model is simple. Pseudogap phenomena in this system can be attributed to precursor pairing correlations. One can define a temperature scale Tsuperscript𝑇T^{*}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT as the temperature at which signatures of pairing first appear as the temperature is lowered from the normal phase. There are different definitions of Tsuperscript𝑇T^{*}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT that depend on the choice of observable used to determine the signatures of pairing correlations Mueller (2017); Jensen et al. (2019). Precursor pairing develops in the BCS-BEC crossover for T>Tc𝑇subscript𝑇𝑐T>T_{c}italic_T > italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT as the coupling changes from the BCS regime where T=Tcsuperscript𝑇subscript𝑇𝑐T^{*}=T_{c}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT = italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT Fetter and Walecka (1971) to the BEC regime where T>>Tcmuch-greater-thansuperscript𝑇subscript𝑇𝑐T^{*}>>T_{c}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT > > italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT and is the dissociation temperature scale for breaking the two-body bound dimers due to thermal fluctuations. However, it is not clear how this precursor pairing emerges. It is a challenging many-body problem to calculate the extent of pairing signatures above Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT for |kFa|>1subscript𝑘𝐹𝑎1|k_{F}a|>1| italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT italic_a | > 1.

Many different techniques have been applied to the UFG and the calculation of observables that are expected to show pairing structure. The AFMC studies of Refs. Magierski et al. (2009, 2011); Wlazłowski et al. (2013a), where a spherical cutoff was used in the lattice model, found signatures of pairing above the critical temperature. Refs. Chien et al. (2010); Chien and Levin (2010); Perali et al. (2011b); Kashimura et al. (2012b); Palestini et al. (2012c, b); Tajima et al. (2014); Wulin et al. (2011); Pini et al. (2019, 2020) applied non-self-consistent diagrammatic approaches and found varying degrees of pairing above the critical temperature and gapped signatures in support of a pronounced pseudogap regime for the UFG. In contrast to these works, the self-consistent Luttinger-Ward diagrammatic approach of Refs. Haussmann et al. (2009); Enss and Haussmann (2012) did not find a pronounced pseudogap regime and found that Tsuperscript𝑇T^{*}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT and Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT are very close for the UFG (see Refs. Pini et al. (2019, 2020) for comparisons between self-consistent and non-self-consistent approaches). On the experimental side, the results also show disagreement. The works of Refs. Wulin et al. (2011); Gaebler et al. (2010) found pseudogap signatures whereas Refs. Ku et al. (2012); Nascimbene et al. (2010); Nascimbène et al. (2011); Sommer et al. (2011) saw no gapped signatures.

In the UFG, the value of Tsuperscript𝑇T^{*}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT (TTcsuperscript𝑇subscript𝑇𝑐T^{*}\geq T_{c}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ≥ italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT) is still debated in the literature. Discrepancies exist even when the same pairing observables are used to determine Tsuperscript𝑇T^{*}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT, as the temperature dependence of these observables differ substantially between various works. In Sec. IV we present controlled AFMC results for two pseudogap signatures: the temperature dependence of the spin susceptibility and the energy-staggering pairing gap across the superfluid phase transition.

II.3 Lattice Model

The continuum model described by the Hamiltonian in Eq. (2) can be replaced by a discrete lattice model and the continuum limit is then recovered by extrapolating to zero filling factor.

To construct the lattice model, we discretize a cubic box of volume V=L3𝑉superscript𝐿3V=L^{3}italic_V = italic_L start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT with NL3superscriptsubscript𝑁𝐿3N_{L}^{3}italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT lattice sites forming a three-dimensional cubic lattice with lattice spacing δx=L/NL𝛿𝑥𝐿subscript𝑁𝐿\delta x=L/N_{L}italic_δ italic_x = italic_L / italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT. The lattice of linear size NL+subscript𝑁𝐿superscriptN_{L}\in\mathbb{Z}^{+}italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT ∈ blackboard_Z start_POSTSUPERSCRIPT + end_POSTSUPERSCRIPT has lattice sites at positions x=0,±δx,±2δx,,±Mδx𝑥0plus-or-minus𝛿𝑥plus-or-minus2𝛿𝑥plus-or-minus𝑀𝛿𝑥x=0,\pm\delta x,\pm 2\delta x,...,\pm M\delta xitalic_x = 0 , ± italic_δ italic_x , ± 2 italic_δ italic_x , … , ± italic_M italic_δ italic_x when NL=2M+1subscript𝑁𝐿2𝑀1N_{L}=2M+1italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT = 2 italic_M + 1 is odd. The lattice Hamiltonian for a contact on-site attractive interaction is given by

H^=𝐤,szϵ𝐤a^𝐤,sza^𝐤,sz+g𝐱n^𝐱,n^𝐱,,^𝐻subscript𝐤subscript𝑠𝑧subscriptitalic-ϵ𝐤subscriptsuperscript^𝑎𝐤subscript𝑠𝑧subscript^𝑎𝐤subscript𝑠𝑧𝑔subscript𝐱subscript^𝑛𝐱subscript^𝑛𝐱\hat{H}=\sum_{{\bf k},s_{z}}\epsilon_{\bf{k}}\hat{a}^{\dagger}_{{\bf k},s_{z}}% \hat{a}_{{\bf k},s_{z}}+g\sum_{\bf{x}}\hat{n}_{\bf{x},\uparrow}\hat{n}_{\bf{x}% ,\downarrow}\;,over^ start_ARG italic_H end_ARG = ∑ start_POSTSUBSCRIPT bold_k , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT italic_ϵ start_POSTSUBSCRIPT bold_k end_POSTSUBSCRIPT over^ start_ARG italic_a end_ARG start_POSTSUPERSCRIPT † end_POSTSUPERSCRIPT start_POSTSUBSCRIPT bold_k , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_a end_ARG start_POSTSUBSCRIPT bold_k , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT + italic_g ∑ start_POSTSUBSCRIPT bold_x end_POSTSUBSCRIPT over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT bold_x , ↑ end_POSTSUBSCRIPT over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT bold_x , ↓ end_POSTSUBSCRIPT , (3)

where a^𝐤,szsubscriptsuperscript^𝑎𝐤subscript𝑠𝑧\hat{a}^{\dagger}_{{\bf k},s_{z}}over^ start_ARG italic_a end_ARG start_POSTSUPERSCRIPT † end_POSTSUPERSCRIPT start_POSTSUBSCRIPT bold_k , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT and a^𝐤,szsubscript^𝑎𝐤subscript𝑠𝑧\hat{a}_{{\bf k},s_{z}}over^ start_ARG italic_a end_ARG start_POSTSUBSCRIPT bold_k , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT are, respectively, creation and annihilation operators for discrete quasi-momentum 𝕜Planck-constant-over-2-pi𝕜\hbar\mathbb{k}roman_ℏ blackboard_k and spin projection szsubscript𝑠𝑧s_{z}italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT. In this work we use the quadratic single-particle dispersion relation ϵ𝕜=2k2/2msubscriptitalic-ϵ𝕜superscriptPlanck-constant-over-2-pi2superscript𝑘22𝑚\epsilon_{\mathbb{k}}=\hbar^{2}k^{2}/2mitalic_ϵ start_POSTSUBSCRIPT blackboard_k end_POSTSUBSCRIPT = roman_ℏ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_k start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT / 2 italic_m though other dispersion relations have also been used (see Ref. Werner and Castin (2012) for a discussion of dispersion relations). The lattice operators obey the anticommutation relations {a^𝐤,sz,a^𝐤,sz}=δ𝐤,𝐤δsz,szsubscript^𝑎𝐤subscript𝑠𝑧subscriptsuperscript^𝑎superscript𝐤superscriptsubscript𝑠𝑧subscript𝛿𝐤superscript𝐤subscript𝛿subscript𝑠𝑧superscriptsubscript𝑠𝑧\{\hat{a}_{{\bf k},s_{z}},\hat{a}^{\dagger}_{{\bf k}^{\prime},s_{z}^{\prime}}% \}=\delta_{{\bf k},{\bf k}^{\prime}}\delta_{s_{z},s_{z}^{\prime}}{ over^ start_ARG italic_a end_ARG start_POSTSUBSCRIPT bold_k , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT , over^ start_ARG italic_a end_ARG start_POSTSUPERSCRIPT † end_POSTSUPERSCRIPT start_POSTSUBSCRIPT bold_k start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT } = italic_δ start_POSTSUBSCRIPT bold_k , bold_k start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT italic_δ start_POSTSUBSCRIPT italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT. The number operator at lattice site 𝕩𝕩\mathbb{x}blackboard_x and spin projection szsubscript𝑠𝑧s_{z}italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT is n^𝐱,sz=ψ^𝐱,szψ^𝐱,szsubscript^𝑛𝐱subscript𝑠𝑧subscriptsuperscript^𝜓𝐱subscript𝑠𝑧subscript^𝜓𝐱subscript𝑠𝑧\hat{n}_{{\bf x},s_{z}}=\hat{\psi}^{\dagger}_{{\bf x},s_{z}}\hat{\psi}_{{\bf x% },s_{z}}over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT bold_x , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT = over^ start_ARG italic_ψ end_ARG start_POSTSUPERSCRIPT † end_POSTSUPERSCRIPT start_POSTSUBSCRIPT bold_x , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_ψ end_ARG start_POSTSUBSCRIPT bold_x , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT with the lattice field operators obeying {ψ^𝐱,sz,ψ^𝐱,sz}=δ𝐱,𝐱δsz,szsubscript^𝜓𝐱subscript𝑠𝑧subscriptsuperscript^𝜓superscript𝐱superscriptsubscript𝑠𝑧subscript𝛿𝐱superscript𝐱subscript𝛿subscript𝑠𝑧superscriptsubscript𝑠𝑧\{\hat{\psi}_{{\bf x},s_{z}},\hat{\psi}^{\dagger}_{{\bf x}^{\prime},s_{z}^{% \prime}}\}=\delta_{\bf{x},\bf{x}^{\prime}}\delta_{s_{z},s_{z}^{\prime}}{ over^ start_ARG italic_ψ end_ARG start_POSTSUBSCRIPT bold_x , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT end_POSTSUBSCRIPT , over^ start_ARG italic_ψ end_ARG start_POSTSUPERSCRIPT † end_POSTSUPERSCRIPT start_POSTSUBSCRIPT bold_x start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT } = italic_δ start_POSTSUBSCRIPT bold_x , bold_x start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT italic_δ start_POSTSUBSCRIPT italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT , italic_s start_POSTSUBSCRIPT italic_z end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT end_POSTSUBSCRIPT. The constant g=V0/(δx)3𝑔subscript𝑉0superscript𝛿𝑥3g=V_{0}/(\delta x)^{3}italic_g = italic_V start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT / ( italic_δ italic_x ) start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT in Eq.(3) is the rescaled coupling constant. The bare coupling constant V0subscript𝑉0V_{0}italic_V start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT is determined by the requirement that the physical s𝑠sitalic_s-wave scattering length a𝑎aitalic_a (a𝑎a\rightarrow\inftyitalic_a → ∞ for the UFG) is reproduced on the lattice. This leads to

1V0=m4π2aBd3k(2π)32ϵ𝐤,1subscript𝑉0𝑚4𝜋superscriptPlanck-constant-over-2-pi2𝑎subscript𝐵superscript𝑑3𝑘superscript2𝜋32subscriptitalic-ϵ𝐤\frac{1}{V_{0}}=\frac{m}{4\pi\hbar^{2}a}-\int_{B}\frac{d^{3}k}{(2\pi)^{3}2% \epsilon_{\mathbf{k}}}\;,divide start_ARG 1 end_ARG start_ARG italic_V start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT end_ARG = divide start_ARG italic_m end_ARG start_ARG 4 italic_π roman_ℏ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_a end_ARG - ∫ start_POSTSUBSCRIPT italic_B end_POSTSUBSCRIPT divide start_ARG italic_d start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT italic_k end_ARG start_ARG ( 2 italic_π ) start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT 2 italic_ϵ start_POSTSUBSCRIPT bold_k end_POSTSUBSCRIPT end_ARG , (4)

where the integral in (4) is over the first Brillouin zone B𝐵Bitalic_B of the lattice. Relation (4) can be derived by solving the Lippmann-Schwinger equation on an infinite lattice.

The UFG is recovered by solving the lattice model for a𝑎a\rightarrow\inftyitalic_a → ∞ and then taking the continuum limit when the filling factor on the lattice ν=N/NL30𝜈𝑁superscriptsubscript𝑁𝐿30\nu=N/N_{L}^{3}\rightarrow 0italic_ν = italic_N / italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT → 0 for given particle number N𝑁Nitalic_N, followed by the thermodynamic limit N𝑁N\rightarrow\inftyitalic_N → ∞. The approach to the continuum limit was first addressed in the works of Refs. Burovski et al. (2006b); Pricoupenko and Castin (2007). Following these works, we expect the many-body energies to be linear in ν1/3superscript𝜈13\nu^{1/3}italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT for a low filling factor ν𝜈\nuitalic_ν. In the present work (see Sec. IV), we take the continuum limit ν0𝜈0\nu\rightarrow 0italic_ν → 0 for various thermodynamic observables using a linear fit in ν1/3superscript𝜈13\nu^{1/3}italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT for a given particle number N𝑁Nitalic_N.

II.4 Lattice AFMC studies

Lattice AFMC methods are among the most accurate, but the main challenges have been (i) taking the continuum limit δx0𝛿𝑥0\delta x\to 0italic_δ italic_x → 0 (or equivalently ν0𝜈0\nu\to 0italic_ν → 0), and (ii) taking the thermodynamic limit of large particle number N𝑁Nitalic_N. There have been several works using large-scale lattice AFMC methods to study signatures of pairing correlations in the UFG at finite temperature. These studies have improved the understanding of finite-temperature properties of the UFG. However, satisfactory control of systematic errors has not been achieved in these studies. We summarize the current status and recent works in this section.

Spherical cutoff studies.— In Refs. Bulgac et al. (2006, 2008); Burovski et al. (2008); Magierski et al. (2009, 2011); Wlazłowski et al. (2012, 2013a, 2013b, 2015) the UFG was studied using a spherical cutoff in the single-particle quasi-momentum space. Ref. Burovski et al. (2008) studied a continuum model where the single-particle quasi-momentum states in the lab frame were restricted to have magnitude less than a cutoff ΛPlanck-constant-over-2-piΛ\hbar\Lambdaroman_ℏ roman_Λ with a renormalization condition similar to Eq. (4) but where the integration region is determined by a spherical cutoff in quasi-momentum space (i.e., ϵ𝕜=2k2/(2m)subscriptitalic-ϵ𝕜superscriptPlanck-constant-over-2-pi2superscript𝑘22𝑚\epsilon_{\mathbb{k}}=\hbar^{2}k^{2}/(2m)italic_ϵ start_POSTSUBSCRIPT blackboard_k end_POSTSUBSCRIPT = roman_ℏ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_k start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT / ( 2 italic_m ) for |𝕜|<Λ𝕜Λ|\mathbb{k}|<\Lambda| blackboard_k | < roman_Λ and ϵ𝕜=subscriptitalic-ϵ𝕜\epsilon_{\mathbb{k}}=\inftyitalic_ϵ start_POSTSUBSCRIPT blackboard_k end_POSTSUBSCRIPT = ∞ for |𝕜|>Λ𝕜Λ|\mathbb{k}|>\Lambda| blackboard_k | > roman_Λ). This model does not allow scattering to lab frame quasi-momenta states with |𝕜|>Λ𝕜Λ|\mathbb{k}|>\Lambda| blackboard_k | > roman_Λ. Refs. Bulgac et al. (2006, 2008); Magierski et al. (2009, 2011); Wlazłowski et al. (2012, 2013a, 2013b, 2015) used a similar model but on a periodic lattice. Pseudogap signatures at unitarity were observed with this model in the simulations of Refs. Magierski et al. (2009, 2011); Wlazłowski et al. (2013a) for temperatures below 0.25TFabsent0.25subscript𝑇𝐹\approx 0.25\;T_{F}≈ 0.25 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT.

Ref. Werner and Castin (2012) calculated the two-particle scattering in the continuum limit using a spherical cutoff model and found that the spherical cutoff does not produce the desired two-body scattering amplitude for finite center-of-mass momentum 𝕂𝕂\mathbb{K}blackboard_K. The low-energy inverse scattering amplitude was found to have the form f1(k,K)=1/aik+K/(2π)+superscript𝑓1𝑘𝐾1𝑎𝑖𝑘𝐾2𝜋f^{-1}(k,K)=-1/a-ik+K/(2\pi)+\ldotsitalic_f start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT ( italic_k , italic_K ) = - 1 / italic_a - italic_i italic_k + italic_K / ( 2 italic_π ) + … where K=|𝕂|𝐾𝕂K=|\mathbb{K}|italic_K = | blackboard_K | (typical k𝑘kitalic_k and K𝐾Kitalic_K are of order kFsubscript𝑘𝐹k_{F}italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT for scattering of the degenerate UFG). Comparing this with Eq. (1), the extra center-of-mass term, K/(2π)𝐾2𝜋K/(2\pi)italic_K / ( 2 italic_π ), shifts the scattering length to an effective value, 1/aeff=1/aK/(2π)1subscript𝑎eff1𝑎𝐾2𝜋1/a_{\textrm{eff}}=1/a-K/(2\pi)1 / italic_a start_POSTSUBSCRIPT eff end_POSTSUBSCRIPT = 1 / italic_a - italic_K / ( 2 italic_π ) away from unitarity (1/a=01𝑎01/a=01 / italic_a = 0) towards the BCS side. This argument suggests the spherical cutoff model underestimates pseudogap effects. However, collision effects for varying center-of-mass momentum in the spherical cutoff model lead to enhanced pairing effects Jensen et al. (2019, 2020a).

Ref. Jensen et al. (2020a) performed a many-body calculation with and without the spherical cutoff. The comparison presented there is not sufficient to address differences at the many-body level in the continuum limit. The effective range parameter resubscript𝑟𝑒r_{e}italic_r start_POSTSUBSCRIPT italic_e end_POSTSUBSCRIPT and the center-of-mass parameter Resubscript𝑅𝑒R_{e}italic_R start_POSTSUBSCRIPT italic_e end_POSTSUBSCRIPT are different when using a spherical cutoff, as opposed to including the full first Brillouin zone (i.e., they approach the continuum limit differently). To address this, simulations were performed in Ref. Jensen et al. (2019) for the spin susceptibility for N=20𝑁20N=20italic_N = 20 particles with multiple lattice sizes to approach the continuum limit. These studies, however, do not resolve the magnitude of the discrepancy in the thermodynamic limit, which has not been addressed.

The spherical cutoff model reduces the single-particle model space and allows for larger lattice simulations within the AFMC framework to more closely approach the continuum and thermodynamic limits (for given computational resources). The early works utilizing this truncation provided the first lattice AFMC investigation of finite temperature pairing signatures. However, recent algorithm advances Drut et al. (2011); Jensen et al. (2019); Richie-Halford et al. (2020); Gilbreth et al. (2021) and increased availability of computational resources have allowed for large-lattice simulations using the full first Brillouin zone for the single-particle model space.

Full first Brillouin zone studies.— There have been a number of studies with finite-temperature AFMC using the full first Brillouin zone with the quadratic single-particle dispersion relation. The early work of Ref. Lee and Schäfer (2006) performed finite filling factor calculations to find a number of observables including the thermal energy and spin susceptibility. This study was performed for smaller lattice sizes up to NL=6subscript𝑁𝐿6N_{L}=6italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT = 6 and down to temperatures of T/TF=0.14𝑇subscript𝑇𝐹0.14T/T_{F}=0.14italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT = 0.14 providing the first lattice AFMC estimate of the Bertsch parameter 0.07ξ0.440.07𝜉0.440.07\leq\xi\leq 0.440.07 ≤ italic_ξ ≤ 0.44. Ref. Drut et al. (2011) produced the first large-scale (NL>13subscript𝑁𝐿13N_{L}>13italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT > 13) finite-temperature AFMC calculations for the UFG using the full first Brillouin zone (see Ref. Drut and Nicholson (2013) for details on algorithm advances). However, this work presented results for the contact and did not focus on pseudogap properties.

Recently, Ref. Richie-Halford et al. (2020) calculated pairing observables for N30𝑁30N\approx 30italic_N ≈ 30 for multiple values of the interaction parameter 1/(kFas)1subscript𝑘𝐹subscript𝑎𝑠1/(k_{F}a_{s})1 / ( italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT italic_a start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ) along the BCS-BEC crossover and obtained estimates for Tsuperscript𝑇T^{*}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT as a function of 1/(kFas)1subscript𝑘𝐹subscript𝑎𝑠1/(k_{F}a_{s})1 / ( italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT italic_a start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ). This work reached much smaller filling factors compared with the simulations of Ref. Jensen et al. (2020a) and provided estimates throughout the entire strongly correlated regime of the BCS-BEC crossover. Ref. Richie-Halford et al. (2020) restricted the study to a moderate particle number N30𝑁30N\approx 30italic_N ≈ 30 and did not take the lattice continuum limit.

In Refs. Jensen et al. (2020a, 2019), the current authors used the canonical ensemble algorithm discussed in Sec. III, to produce a large-scale study at finite filling factor ν0.6𝜈0.6\nu\approx 0.6italic_ν ≈ 0.6 for pairing observables using the full first Brillouin zone. These works presented results for the spin susceptibility, condensate fraction, and the energy-staggering pairing gap. Results were presented for multiple particle numbers up to N=130𝑁130N=130italic_N = 130. This work found a negligible pseudogap regime based on the analysis of the pairing gap and spin susceptibility for large particle number. However, these results had sizable lattice effects with an effective range parameter kFre0.41similar-to-or-equalssubscript𝑘𝐹subscript𝑟𝑒0.41k_{F}r_{e}\simeq 0.41italic_k start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT italic_r start_POSTSUBSCRIPT italic_e end_POSTSUBSCRIPT ≃ 0.41 and cannot be directly compared with experiment or other UFG results. This presents a large systematic error relative to the UFG with vanishing effective range parameters as defined in Sec. II.1. The current work (see results in Sec. IV) addresses the systematic errors of these previous studies.

III Finite-temperature lattice AFMC

We calculate thermodynamic properties of the system with the Hamiltonian of Eq. (2) using a discrete lattice of size NL3superscriptsubscript𝑁𝐿3N_{L}^{3}italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT and total particle number N=N+N𝑁subscript𝑁subscript𝑁N=N_{\uparrow}+N_{\downarrow}italic_N = italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT + italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT with controllable systematic errors as discussed in Sec. II.3. In particular, we have implemented finite-temperature auxiliary-field quantum Monte Carlo (AFMC) methods in the canonical ensemble of fixed particle numbers Nsubscript𝑁N_{\uparrow}italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT and Nsubscript𝑁N_{\downarrow}italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT.

III.1 Hubbard-Stratonovich transformation

The Gibbs operator eβH^superscript𝑒𝛽^𝐻e^{-\beta\hat{H}}italic_e start_POSTSUPERSCRIPT - italic_β over^ start_ARG italic_H end_ARG end_POSTSUPERSCRIPT at inverse temperature β=1/kBT𝛽1subscript𝑘𝐵𝑇\beta=1/k_{B}Titalic_β = 1 / italic_k start_POSTSUBSCRIPT italic_B end_POSTSUBSCRIPT italic_T can be thought of as the propagator in imaginary-time β𝛽\betaitalic_β. Decomposing the Hamiltonian into H^=K^+V^^𝐻^𝐾^𝑉\hat{H}=\hat{K}+\hat{V}over^ start_ARG italic_H end_ARG = over^ start_ARG italic_K end_ARG + over^ start_ARG italic_V end_ARG, where K^^𝐾\hat{K}over^ start_ARG italic_K end_ARG is the kinetic energy operator and V^^𝑉\hat{V}over^ start_ARG italic_V end_ARG is the two-body interaction, we use a symmetric Trotter decomposition

eβH^=[eΔβK^/2eΔβV^eΔβK^/2]Nτ+O(Δβ2),superscript𝑒𝛽^𝐻superscriptdelimited-[]superscript𝑒Δ𝛽^𝐾2superscript𝑒Δ𝛽^𝑉superscript𝑒Δ𝛽^𝐾2subscript𝑁𝜏𝑂Δsuperscript𝛽2e^{-\beta\hat{H}}=[e^{-\Delta\beta\hat{K}/2}e^{-\Delta\beta\hat{V}}e^{-\Delta% \beta\hat{K}/2}]^{N_{\tau}}+O(\Delta\beta^{2})\;,italic_e start_POSTSUPERSCRIPT - italic_β over^ start_ARG italic_H end_ARG end_POSTSUPERSCRIPT = [ italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β over^ start_ARG italic_K end_ARG / 2 end_POSTSUPERSCRIPT italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β over^ start_ARG italic_V end_ARG end_POSTSUPERSCRIPT italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β over^ start_ARG italic_K end_ARG / 2 end_POSTSUPERSCRIPT ] start_POSTSUPERSCRIPT italic_N start_POSTSUBSCRIPT italic_τ end_POSTSUBSCRIPT end_POSTSUPERSCRIPT + italic_O ( roman_Δ italic_β start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ) , (5)

where we have divided the imaginary time into Nτ=β/Δβsubscript𝑁𝜏𝛽Δ𝛽N_{\tau}=\beta/{\Delta\beta}italic_N start_POSTSUBSCRIPT italic_τ end_POSTSUBSCRIPT = italic_β / roman_Δ italic_β discrete time slices of length ΔβΔ𝛽\Delta\betaroman_Δ italic_β. The number operator, n^𝕩subscript^𝑛𝕩\hat{n}_{\mathbb{x}}over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT, at site 𝕩𝕩\mathbb{x}blackboard_x is given by n^𝕩=n^𝕩,+n^𝕩,subscript^𝑛𝕩subscript^𝑛𝕩subscript^𝑛𝕩\hat{n}_{\mathbb{x}}=\hat{n}_{\mathbb{x},\uparrow}+\hat{n}_{\mathbb{x},\downarrow}over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT = over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x , ↑ end_POSTSUBSCRIPT + over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x , ↓ end_POSTSUBSCRIPT. For fermions, each interaction term in Eq. (2) can be written as gn^𝕩,n^𝕩,=12g(n^𝕩2n^𝕩)𝑔subscript^𝑛𝕩subscript^𝑛𝕩12𝑔superscriptsubscript^𝑛𝕩2subscript^𝑛𝕩g\hat{n}_{\mathbb{x},\downarrow}\hat{n}_{\mathbb{x},\uparrow}=\frac{1}{2}g(% \hat{n}_{\mathbb{x}}^{2}-\hat{n}_{\mathbb{x}})italic_g over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x , ↓ end_POSTSUBSCRIPT over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x , ↑ end_POSTSUBSCRIPT = divide start_ARG 1 end_ARG start_ARG 2 end_ARG italic_g ( over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT - over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT ). Completing the squares, we obtain the Hubbard-Stratonovich transformation Hubbard (1959); Stratonovich (1957),

eΔβgn^𝕩2/2=Δβ|g|2π𝑑σ𝕩eΔβ|g|σ𝕩2/2eΔβsgσ𝕩n^𝕩,superscript𝑒Δ𝛽𝑔superscriptsubscript^𝑛𝕩22Δ𝛽𝑔2𝜋superscriptsubscriptdifferential-dsubscript𝜎𝕩superscript𝑒Δ𝛽𝑔superscriptsubscript𝜎𝕩22superscript𝑒Δ𝛽𝑠𝑔subscript𝜎𝕩subscript^𝑛𝕩e^{-\Delta\beta g\hat{n}_{\mathbb{x}}^{2}/2}=\sqrt{\frac{\Delta\beta|g|}{2\pi}% }\int_{-\infty}^{\infty}d\sigma_{\mathbb{x}}e^{-\Delta\beta|g|\sigma_{\mathbb{% x}}^{2}/2}e^{-\Delta\beta sg\sigma_{\mathbb{x}}\hat{n}_{\mathbb{x}}}\;,italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β italic_g over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT / 2 end_POSTSUPERSCRIPT = square-root start_ARG divide start_ARG roman_Δ italic_β | italic_g | end_ARG start_ARG 2 italic_π end_ARG end_ARG ∫ start_POSTSUBSCRIPT - ∞ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ∞ end_POSTSUPERSCRIPT italic_d italic_σ start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β | italic_g | italic_σ start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT / 2 end_POSTSUPERSCRIPT italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β italic_s italic_g italic_σ start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT end_POSTSUPERSCRIPT , (6)

where s=1𝑠1s=1italic_s = 1 for g<0𝑔0g<0italic_g < 0 and s=i𝑠𝑖s=iitalic_s = italic_i for g>0𝑔0g>0italic_g > 0. This is repeated for each of the Nτsubscript𝑁𝜏N_{\tau}italic_N start_POSTSUBSCRIPT italic_τ end_POSTSUBSCRIPT time slices and NL3superscriptsubscript𝑁𝐿3N_{L}^{3}italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT lattice sites, and leads, in the limit Δβ0Δ𝛽0\Delta\beta\to 0roman_Δ italic_β → 0, to the functional integral form

eβH^=D[σ]GσU^σ.superscript𝑒𝛽^𝐻𝐷delimited-[]𝜎subscript𝐺𝜎subscript^𝑈𝜎e^{-\beta\hat{H}}=\int D[\sigma]G_{\sigma}\hat{U}_{\sigma}\;.italic_e start_POSTSUPERSCRIPT - italic_β over^ start_ARG italic_H end_ARG end_POSTSUPERSCRIPT = ∫ italic_D [ italic_σ ] italic_G start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT . (7)

The integration in Eq. (7) is over auxiliary fields σ𝕩(τn)subscript𝜎𝕩subscript𝜏𝑛\sigma_{\mathbb{x}}(\tau_{n})italic_σ start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT ( italic_τ start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT ) defined for each lattice site 𝕩𝕩\mathbb{x}blackboard_x and imaginary time slice τn=nΔβsubscript𝜏𝑛𝑛Δ𝛽\tau_{n}=n\Delta\betaitalic_τ start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT = italic_n roman_Δ italic_β (n=0,1,,Nτ𝑛01subscript𝑁𝜏n=0,1,\ldots,N_{\tau}italic_n = 0 , 1 , … , italic_N start_POSTSUBSCRIPT italic_τ end_POSTSUBSCRIPT), and the integration measure D[σ]=𝐱,n[dσ𝐱(τn)Δβ|g|/2π]𝐷delimited-[]𝜎subscriptproduct𝐱𝑛delimited-[]𝑑subscript𝜎𝐱subscript𝜏𝑛Δ𝛽𝑔2𝜋D[\sigma]=\prod_{{\bf x},n}\left[d\sigma_{\bf x}(\tau_{n})\sqrt{{{\Delta}\beta% |g|/{2\pi}}}\right]italic_D [ italic_σ ] = ∏ start_POSTSUBSCRIPT bold_x , italic_n end_POSTSUBSCRIPT [ italic_d italic_σ start_POSTSUBSCRIPT bold_x end_POSTSUBSCRIPT ( italic_τ start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT ) square-root start_ARG roman_Δ italic_β | italic_g | / 2 italic_π end_ARG ]. U^σsubscript^𝑈𝜎\hat{U}_{\sigma}over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT is a time-ordered product for the propagator of non-interacting fermions in the imaginary-time dependent fields σ𝐱(τ)subscript𝜎𝐱𝜏\sigma_{\bf x}(\tau)italic_σ start_POSTSUBSCRIPT bold_x end_POSTSUBSCRIPT ( italic_τ ), given explicitly by

U^σ=neΔβK^/2eΔβh^σ(τn)eΔβK^/2,subscript^𝑈𝜎subscriptproduct𝑛superscript𝑒Δ𝛽^𝐾2superscript𝑒Δ𝛽subscript^𝜎subscript𝜏𝑛superscript𝑒Δ𝛽^𝐾2\hat{U}_{\sigma}=\prod_{n}e^{-\Delta\beta\hat{K}/2}e^{-\Delta\beta\hat{h}_{% \sigma}(\tau_{n})}e^{-\Delta\beta\hat{K}/2}\;,over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = ∏ start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β over^ start_ARG italic_K end_ARG / 2 end_POSTSUPERSCRIPT italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β over^ start_ARG italic_h end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ( italic_τ start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT ) end_POSTSUPERSCRIPT italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β over^ start_ARG italic_K end_ARG / 2 end_POSTSUPERSCRIPT , (8)

where h^σ(τn)=g𝐱σ𝐱(τn)n^𝐱gN^/2subscript^𝜎subscript𝜏𝑛𝑔subscript𝐱subscript𝜎𝐱subscript𝜏𝑛subscript^𝑛𝐱𝑔^𝑁2\hat{h}_{\sigma}(\tau_{n})=g\sum_{\bf x}\sigma_{\bf x}(\tau_{n})\hat{n}_{\bf x% }-g\hat{N}/2over^ start_ARG italic_h end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ( italic_τ start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT ) = italic_g ∑ start_POSTSUBSCRIPT bold_x end_POSTSUBSCRIPT italic_σ start_POSTSUBSCRIPT bold_x end_POSTSUBSCRIPT ( italic_τ start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT ) over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT bold_x end_POSTSUBSCRIPT - italic_g over^ start_ARG italic_N end_ARG / 2 and Gσ=enΔβ|g|σ𝐱(τn)/2subscript𝐺𝜎superscript𝑒subscript𝑛Δ𝛽𝑔subscript𝜎𝐱subscript𝜏𝑛2G_{\sigma}=e^{-\sum_{n}\Delta\beta|g|\sigma_{\bf x}(\tau_{n})/2}italic_G start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = italic_e start_POSTSUPERSCRIPT - ∑ start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT roman_Δ italic_β | italic_g | italic_σ start_POSTSUBSCRIPT bold_x end_POSTSUBSCRIPT ( italic_τ start_POSTSUBSCRIPT italic_n end_POSTSUBSCRIPT ) / 2 end_POSTSUPERSCRIPT is a Gaussian weight. The operator U^σsubscript^𝑈𝜎\hat{U}_{\sigma}over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT describes the many-particle propagator of the non-interacting spin-balanced system of fermions in external time-dependent one-body auxiliary fields σ𝕩(τ)subscript𝜎𝕩𝜏\sigma_{\mathbb{x}}(\tau)italic_σ start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT ( italic_τ ).

III.2 Thermal expectation values of observables and Gaussian quadratures

The thermal expectation value O^delimited-⟨⟩^𝑂\langle\hat{O}\rangle⟨ over^ start_ARG italic_O end_ARG ⟩ of an observable is given by

O^=Tr(O^eβH^)Tr(eβH^)=D[σ]O^σWσΦσD[σ]WσΦσ,delimited-⟨⟩^𝑂Tr^𝑂superscript𝑒𝛽^𝐻Trsuperscript𝑒𝛽^𝐻𝐷delimited-[]𝜎subscriptdelimited-⟨⟩^𝑂𝜎subscript𝑊𝜎subscriptΦ𝜎𝐷delimited-[]𝜎subscript𝑊𝜎subscriptΦ𝜎\langle\hat{O}\rangle=\frac{\textrm{Tr}(\hat{O}e^{-\beta\hat{H}})}{\textrm{Tr}% (e^{-\beta\hat{H}})}=\frac{\int D[\sigma]\langle\hat{O}\rangle_{\sigma}W_{% \sigma}\Phi_{\sigma}}{\int D[\sigma]W_{\sigma}\Phi_{\sigma}}\;,⟨ over^ start_ARG italic_O end_ARG ⟩ = divide start_ARG Tr ( over^ start_ARG italic_O end_ARG italic_e start_POSTSUPERSCRIPT - italic_β over^ start_ARG italic_H end_ARG end_POSTSUPERSCRIPT ) end_ARG start_ARG Tr ( italic_e start_POSTSUPERSCRIPT - italic_β over^ start_ARG italic_H end_ARG end_POSTSUPERSCRIPT ) end_ARG = divide start_ARG ∫ italic_D [ italic_σ ] ⟨ over^ start_ARG italic_O end_ARG ⟩ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT italic_W start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT roman_Φ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG start_ARG ∫ italic_D [ italic_σ ] italic_W start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT roman_Φ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG , (9)

where Wσ=Gσ|TrU^σ|subscript𝑊𝜎subscript𝐺𝜎Trsubscript^𝑈𝜎W_{\sigma}=G_{\sigma}|\textrm{Tr}\hat{U}_{\sigma}|italic_W start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = italic_G start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT | Tr over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT | is a positive-definite weight function, Φσ=TrU^σ/|TrU^σ|subscriptΦ𝜎Trsubscript^𝑈𝜎Trsubscript^𝑈𝜎\Phi_{\sigma}=\textrm{Tr}\hat{U}_{\sigma}/|\textrm{Tr}\hat{U}_{\sigma}|roman_Φ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = Tr over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT / | Tr over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT | is the sign function, and O^σ=Tr(O^U^σ)/TrU^σsubscriptdelimited-⟨⟩^𝑂𝜎Tr^𝑂subscript^𝑈𝜎Trsubscript^𝑈𝜎\langle\hat{O}\rangle_{\sigma}=\textrm{Tr}(\hat{O}\hat{U}_{\sigma})/\textrm{Tr% }\hat{U}_{\sigma}⟨ over^ start_ARG italic_O end_ARG ⟩ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = Tr ( over^ start_ARG italic_O end_ARG over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ) / Tr over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT is the expectation value for a single configuration σ𝜎\sigmaitalic_σ of the auxiliary fields. At each site 𝕩𝕩\mathbb{x}blackboard_x and time slice, we approximate the integral over σ𝕩subscript𝜎𝕩\sigma_{\mathbb{x}}italic_σ start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT by Gaussian quadrature

Δβ|g|2π𝑑σ𝕩eΔβ|g|σ𝕩2/2eΔβsgσ𝕩n^𝕩=j=1,0,1wjeΔβsgσ𝕩,jn^𝕩+O(Δβ3),Δ𝛽𝑔2𝜋superscriptsubscriptdifferential-dsubscript𝜎𝕩superscript𝑒Δ𝛽𝑔superscriptsubscript𝜎𝕩22superscript𝑒Δ𝛽𝑠𝑔subscript𝜎𝕩subscript^𝑛𝕩subscript𝑗101subscript𝑤𝑗superscript𝑒Δ𝛽𝑠𝑔subscript𝜎𝕩𝑗subscript^𝑛𝕩𝑂Δsuperscript𝛽3\sqrt{\frac{\Delta\beta|g|}{2\pi}}\int_{-\infty}^{\infty}d\sigma_{\mathbb{x}}e% ^{-\Delta\beta|g|\sigma_{\mathbb{x}}^{2}/2}e^{-\Delta\beta sg\sigma_{\mathbb{x% }}\hat{n}_{\mathbb{x}}}\\ =\sum_{j=-1,0,1}w_{j}e^{-\Delta\beta sg\sigma_{\mathbb{x},j}\hat{n}_{\mathbb{x% }}}+O(\Delta\beta^{3})\;,start_ROW start_CELL square-root start_ARG divide start_ARG roman_Δ italic_β | italic_g | end_ARG start_ARG 2 italic_π end_ARG end_ARG ∫ start_POSTSUBSCRIPT - ∞ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ∞ end_POSTSUPERSCRIPT italic_d italic_σ start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β | italic_g | italic_σ start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT / 2 end_POSTSUPERSCRIPT italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β italic_s italic_g italic_σ start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT end_POSTSUPERSCRIPT end_CELL end_ROW start_ROW start_CELL = ∑ start_POSTSUBSCRIPT italic_j = - 1 , 0 , 1 end_POSTSUBSCRIPT italic_w start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT italic_e start_POSTSUPERSCRIPT - roman_Δ italic_β italic_s italic_g italic_σ start_POSTSUBSCRIPT blackboard_x , italic_j end_POSTSUBSCRIPT over^ start_ARG italic_n end_ARG start_POSTSUBSCRIPT blackboard_x end_POSTSUBSCRIPT end_POSTSUPERSCRIPT + italic_O ( roman_Δ italic_β start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ) , end_CELL end_ROW (10)

where σ𝕩,±1=±3/(Δβ|g|)subscript𝜎𝕩plus-or-minus1plus-or-minus3Δ𝛽𝑔\sigma_{\mathbb{x},\pm 1}=\pm\sqrt{3/(\Delta\beta|g|)}italic_σ start_POSTSUBSCRIPT blackboard_x , ± 1 end_POSTSUBSCRIPT = ± square-root start_ARG 3 / ( roman_Δ italic_β | italic_g | ) end_ARG and σ𝕩,0=0subscript𝜎𝕩00\sigma_{\mathbb{x},0}=0italic_σ start_POSTSUBSCRIPT blackboard_x , 0 end_POSTSUBSCRIPT = 0. The weights wjsubscript𝑤𝑗w_{j}italic_w start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT are given by w±1=1/6subscript𝑤plus-or-minus116w_{\pm 1}=1/6italic_w start_POSTSUBSCRIPT ± 1 end_POSTSUBSCRIPT = 1 / 6 and w0=2/3subscript𝑤023w_{0}=2/3italic_w start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = 2 / 3. We sample the discrete auxiliary fields σ𝕩,j(τ)subscript𝜎𝕩𝑗𝜏\sigma_{\mathbb{x},j}(\tau)italic_σ start_POSTSUBSCRIPT blackboard_x , italic_j end_POSTSUBSCRIPT ( italic_τ ) according to the distribution Wσsubscript𝑊𝜎W_{\sigma}italic_W start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT using the Metropolis algorithm Metropolis et al. (1953).

III.3 Particle-number projection

The particle-number projection operator for N𝑁Nitalic_N particles and Nssubscript𝑁𝑠N_{s}italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT single-particle states can be written as a discrete Fourier transform Ormand et al. (1994)

P^N=eβμNNsm=1NseiφmNe(βμ+iφm)N^,subscript^𝑃𝑁superscript𝑒𝛽𝜇𝑁subscript𝑁𝑠superscriptsubscript𝑚1subscript𝑁𝑠superscript𝑒𝑖subscript𝜑𝑚𝑁superscript𝑒𝛽𝜇𝑖subscript𝜑𝑚^𝑁\hat{P}_{N}=\frac{e^{-\beta\mu N}}{N_{s}}\sum_{m=1}^{N_{s}}e^{-i\varphi_{m}N}e% ^{(\beta\mu+i\varphi_{m})\hat{N}}\;,over^ start_ARG italic_P end_ARG start_POSTSUBSCRIPT italic_N end_POSTSUBSCRIPT = divide start_ARG italic_e start_POSTSUPERSCRIPT - italic_β italic_μ italic_N end_POSTSUPERSCRIPT end_ARG start_ARG italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT end_ARG ∑ start_POSTSUBSCRIPT italic_m = 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT end_POSTSUPERSCRIPT italic_e start_POSTSUPERSCRIPT - italic_i italic_φ start_POSTSUBSCRIPT italic_m end_POSTSUBSCRIPT italic_N end_POSTSUPERSCRIPT italic_e start_POSTSUPERSCRIPT ( italic_β italic_μ + italic_i italic_φ start_POSTSUBSCRIPT italic_m end_POSTSUBSCRIPT ) over^ start_ARG italic_N end_ARG end_POSTSUPERSCRIPT , (11)

where φm=2πmNssubscript𝜑𝑚2𝜋𝑚subscript𝑁𝑠\varphi_{m}=\frac{2\pi m}{N_{s}}italic_φ start_POSTSUBSCRIPT italic_m end_POSTSUBSCRIPT = divide start_ARG 2 italic_π italic_m end_ARG start_ARG italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT end_ARG (m=1,,Ns𝑚1subscript𝑁𝑠m=1,\ldots,N_{s}italic_m = 1 , … , italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT) are quadrature points, and μ𝜇\muitalic_μ is a chemical potential introduced to ensure the numerical stability of the Fourier sum. μ𝜇\muitalic_μ is chosen so as to reproduce the correct number of particles N𝑁Nitalic_N in a grand-canonical average in the ensemble Uσsubscript𝑈𝜎U_{\sigma}italic_U start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT, i.e., Tr(N^U^σeβμN^)/Tr(U^σeβμN^)=NTr^𝑁subscript^𝑈𝜎superscript𝑒𝛽𝜇^𝑁Trsubscript^𝑈𝜎superscript𝑒𝛽𝜇^𝑁𝑁{\rm Tr}(\hat{N}\hat{U}_{\sigma}e^{\beta\mu\hat{N}})/{\rm Tr}(\hat{U}_{\sigma}% e^{\beta\mu\hat{N}})=Nroman_Tr ( over^ start_ARG italic_N end_ARG over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT italic_e start_POSTSUPERSCRIPT italic_β italic_μ over^ start_ARG italic_N end_ARG end_POSTSUPERSCRIPT ) / roman_Tr ( over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT italic_e start_POSTSUPERSCRIPT italic_β italic_μ over^ start_ARG italic_N end_ARG end_POSTSUPERSCRIPT ) = italic_N. We can calculate grand-canonical traces over the many-particle Fock space by using the matrix representation 𝕌σsubscript𝕌𝜎\mathbb{U}_{\sigma}blackboard_U start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT of the many-particle propagator U^σsubscript^𝑈𝜎\hat{U}_{\sigma}over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT in the single-particle space. For example

Tr[e(βμ+iφm)N^U^σ]=det[11+e(βμ+iφm)𝕌σ],Trdelimited-[]superscript𝑒𝛽𝜇𝑖subscript𝜑𝑚^𝑁subscript^𝑈𝜎detdelimited-[]11superscript𝑒𝛽𝜇𝑖subscript𝜑𝑚subscript𝕌𝜎\textrm{Tr}[e^{(\beta\mu+i\varphi_{m})\hat{N}}\hat{U}_{\sigma}]=\textrm{det}[1% \!\!1+e^{(\beta\mu+i\varphi_{m})}\mathbb{U}_{\sigma}]\;,Tr [ italic_e start_POSTSUPERSCRIPT ( italic_β italic_μ + italic_i italic_φ start_POSTSUBSCRIPT italic_m end_POSTSUBSCRIPT ) over^ start_ARG italic_N end_ARG end_POSTSUPERSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ] = det [ 1 1 + italic_e start_POSTSUPERSCRIPT ( italic_β italic_μ + italic_i italic_φ start_POSTSUBSCRIPT italic_m end_POSTSUBSCRIPT ) end_POSTSUPERSCRIPT blackboard_U start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ] , (12)

where 11111\!\!11 1 is the identity matrix in the single-particle space. Thus the canonical trace of U^σsubscript^𝑈𝜎\hat{U}_{\sigma}over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT at fixed particle number N𝑁Nitalic_N is given by

TrNU^σ=eβμNNsm=1NseiφmNdet[11+e(iφm+βμ)𝕌σ].subscriptTr𝑁subscript^𝑈𝜎superscript𝑒𝛽𝜇𝑁subscript𝑁𝑠superscriptsubscript𝑚1subscript𝑁𝑠superscript𝑒𝑖subscript𝜑𝑚𝑁delimited-[]11superscript𝑒𝑖subscript𝜑𝑚𝛽𝜇subscript𝕌𝜎\textrm{Tr}_{N}\hat{U}_{\sigma}=\frac{e^{-\beta\mu N}}{N_{s}}\sum_{m=1}^{N_{s}% }e^{-i\varphi_{m}N}\det[1\!\!1+e^{(i\varphi_{m}+\beta\mu)}\mathbb{U}_{\sigma}]\;.Tr start_POSTSUBSCRIPT italic_N end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = divide start_ARG italic_e start_POSTSUPERSCRIPT - italic_β italic_μ italic_N end_POSTSUPERSCRIPT end_ARG start_ARG italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT end_ARG ∑ start_POSTSUBSCRIPT italic_m = 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT end_POSTSUPERSCRIPT italic_e start_POSTSUPERSCRIPT - italic_i italic_φ start_POSTSUBSCRIPT italic_m end_POSTSUBSCRIPT italic_N end_POSTSUPERSCRIPT roman_det [ 1 1 + italic_e start_POSTSUPERSCRIPT ( italic_i italic_φ start_POSTSUBSCRIPT italic_m end_POSTSUBSCRIPT + italic_β italic_μ ) end_POSTSUPERSCRIPT blackboard_U start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ] . (13)

In our AFMC calculations, we choose the single-particle basis to be the single-particle states with good quasi-momentum 𝕜Planck-constant-over-2-pi𝕜\hbar\mathbb{k}roman_ℏ blackboard_k and spin projection σ𝜎\sigmaitalic_σ. The single-particle propagator matrices 𝕌σsubscript𝕌𝜎\mathbb{U}_{\sigma}blackboard_U start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT are then evaluated in momentum space. When calculating the short-time propagator with the linearized interaction term, we obtain a speedup for the local interaction in coordinate space using a fast Fourier transform approach Batrouni et al. (1986); Davies et al. (1988); Bulgac et al. (2006); Drut and Nicholson (2013). All observables are calculated with two number projections for fixed Nsubscript𝑁N_{\uparrow}italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT and fixed Nsubscript𝑁N_{\downarrow}italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT with the exception of the spin susceptibility. The spin susceptibility uses only one number projection for total particle number N=N+N𝑁subscript𝑁subscript𝑁N=N_{\uparrow}+N_{\downarrow}italic_N = italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT + italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT as discussed in Section IV.4.

III.4 Monte Carlo sign

The spin-balanced Fermi gas with an attractive contact interaction has a good Monte Carlo sign, i.e., the canonical trace of U^σsubscript^𝑈𝜎\hat{U}_{\sigma}over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT is positive for each sample σ𝜎\sigmaitalic_σ of the auxiliary fields. To show this, we note that in general

TrN,NU^σ=TrNU^σTrNU^σ.subscriptTrsubscript𝑁subscript𝑁subscript^𝑈𝜎subscriptTrsubscript𝑁superscriptsubscript^𝑈𝜎subscriptTrsubscript𝑁superscriptsubscript^𝑈𝜎{\rm Tr}_{N_{\uparrow},N_{\downarrow}}\hat{U}_{\sigma}={\rm Tr}_{N_{\uparrow}}% \hat{U}_{\sigma}^{\uparrow}\;{\rm Tr}_{N_{\downarrow}}\hat{U}_{\sigma}^{% \downarrow}\;.roman_Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = roman_Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ↑ end_POSTSUPERSCRIPT roman_Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ↓ end_POSTSUPERSCRIPT . (14)

For an attractive interaction, the propagators U^σsuperscriptsubscript^𝑈𝜎\hat{U}_{\sigma}^{\uparrow}over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ↑ end_POSTSUPERSCRIPT and U^σsuperscriptsubscript^𝑈𝜎\hat{U}_{\sigma}^{\downarrow}over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ↓ end_POSTSUPERSCRIPT are invariant under time reversal and both particle-number projected traces on the right-hand-side of Eq. (14) are real. For the spin-balanced system N=Nsubscript𝑁subscript𝑁N_{\uparrow}=N_{\downarrow}italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT = italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT, we have TrN,NU^σ=(TrNU^σ)2subscriptTrsubscript𝑁subscript𝑁subscript^𝑈𝜎superscriptsubscriptTrsubscript𝑁superscriptsubscript^𝑈𝜎2{\rm Tr}_{N_{\uparrow},N_{\downarrow}}\hat{U}_{\sigma}=\left({\rm Tr}_{N_{% \uparrow}}\hat{U}_{\sigma}^{\uparrow}\right)^{2}roman_Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = ( roman_Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT ↑ end_POSTSUPERSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT and thus the canonical trace is positive for any sample σ𝜎\sigmaitalic_σ.

III.5 Numerical stabilization

The calculation of the propagator matrix 𝕌σsubscript𝕌𝜎\mathbb{U}_{\sigma}blackboard_U start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT requires the multiplication of Nτ=β/Δβsubscript𝑁𝜏𝛽Δ𝛽N_{\tau}=\beta/\Delta\betaitalic_N start_POSTSUBSCRIPT italic_τ end_POSTSUBSCRIPT = italic_β / roman_Δ italic_β matrices. In this section we use the simplified notation U𝕌σ=i=1NτUi𝑈subscript𝕌𝜎superscriptsubscriptproduct𝑖1subscript𝑁𝜏subscript𝑈𝑖U\equiv\mathbb{U}_{\sigma}=\prod_{i=1}^{N_{\tau}}U_{i}italic_U ≡ blackboard_U start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = ∏ start_POSTSUBSCRIPT italic_i = 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_N start_POSTSUBSCRIPT italic_τ end_POSTSUBSCRIPT end_POSTSUPERSCRIPT italic_U start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT where Uisubscript𝑈𝑖U_{i}italic_U start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT denotes the ithsuperscript𝑖𝑡i^{th}italic_i start_POSTSUPERSCRIPT italic_t italic_h end_POSTSUPERSCRIPT imaginary-time step propagator. The condition number of U𝑈Uitalic_U is defined by κ=UU1𝜅norm𝑈normsuperscript𝑈1\kappa=||U||||U^{-1}||italic_κ = | | italic_U | | | | italic_U start_POSTSUPERSCRIPT - 1 end_POSTSUPERSCRIPT | |, where Unorm𝑈||U||| | italic_U | | is a matrix norm Meyer (2000); Golub and Van Loan (2013) such as the largest eigenvalue of U𝑈Uitalic_U or the Frobenius norm,

U=i=1Nsj=1Ns|Uij|2.norm𝑈superscriptsubscript𝑖1subscript𝑁𝑠superscriptsubscript𝑗1subscript𝑁𝑠superscriptsubscript𝑈𝑖𝑗2||U||=\sqrt{\sum_{i=1}^{N_{s}}\sum_{j=1}^{N_{s}}|U_{ij}|^{2}}\;.| | italic_U | | = square-root start_ARG ∑ start_POSTSUBSCRIPT italic_i = 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT end_POSTSUPERSCRIPT ∑ start_POSTSUBSCRIPT italic_j = 1 end_POSTSUBSCRIPT start_POSTSUPERSCRIPT italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT end_POSTSUPERSCRIPT | italic_U start_POSTSUBSCRIPT italic_i italic_j end_POSTSUBSCRIPT | start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT end_ARG . (15)

As Nτsubscript𝑁𝜏N_{\tau}italic_N start_POSTSUBSCRIPT italic_τ end_POSTSUBSCRIPT becomes large and we approach low temperatures (large β𝛽\betaitalic_β), the condition number κ𝜅\kappaitalic_κ can become very large and the product U𝑈Uitalic_U is ill-conditioned, i.e., we lose relevant information regarding the separation of large and small scales.

This problem is resolved in AFMC calculations by decomposing the propagator into a singular-value decomposition (SVD) Loh et al. (1989); Loh Jr and Gubernatis (1992); Loh et al. (2005); Gubernatis et al. (2016), or a QDR decomposition, in order to keep the separation of scales in a diagonal matrix and avoid the loss of information from mixing scales. In both cases, the matrix U𝑈Uitalic_U is decomposed into a product U=ADB𝑈𝐴𝐷𝐵U=ADBitalic_U = italic_A italic_D italic_B, where D𝐷Ditalic_D is a diagonal matrix with the different scales and A𝐴Aitalic_A and B𝐵Bitalic_B are well-conditioned matrices Loh Jr and Gubernatis (1992); Koonin (1997); Gilbreth and Alhassid (2015); Gubernatis et al. (2016). This can be described schematically by

U=(XXXXXXXXXXXXXXXX)(XXXX)(XXXXXXXXXXXXXXXX),𝑈𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋missing-subexpressionmissing-subexpressionmissing-subexpressionmissing-subexpression𝑋missing-subexpressionmissing-subexpressionmissing-subexpressionmissing-subexpression𝑋missing-subexpressionmissing-subexpressionmissing-subexpressionmissing-subexpression𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋𝑋U=\left(\begin{array}[]{cccc}\scriptscriptstyle{X}&\scriptscriptstyle{X}&% \scriptscriptstyle{X}&\scriptscriptstyle{X}\\ \scriptscriptstyle{X}&\scriptscriptstyle{X}&\scriptscriptstyle{X}&% \scriptscriptstyle{X}\\ \scriptscriptstyle{X}&\scriptscriptstyle{X}&\scriptscriptstyle{X}&% \scriptscriptstyle{X}\\ \scriptscriptstyle{X}&\scriptscriptstyle{X}&\scriptscriptstyle{X}&% \scriptscriptstyle{X}\end{array}\right)\left(\begin{array}[]{cccc}X&&&\\ &\scriptstyle{X}&&\\ &&\scriptscriptstyle{X}&\\ &&&\scriptscriptstyle{X}\end{array}\right)\left(\begin{array}[]{cccc}% \scriptscriptstyle{X}&\scriptscriptstyle{X}&\scriptscriptstyle{X}&% \scriptscriptstyle{X}\\ \scriptscriptstyle{X}&\scriptscriptstyle{X}&\scriptscriptstyle{X}&% \scriptscriptstyle{X}\\ \scriptscriptstyle{X}&\scriptscriptstyle{X}&\scriptscriptstyle{X}&% \scriptscriptstyle{X}\\ \scriptscriptstyle{X}&\scriptscriptstyle{X}&\scriptscriptstyle{X}&% \scriptscriptstyle{X}\end{array}\right)\;,italic_U = ( start_ARRAY start_ROW start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL end_ROW start_ROW start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL end_ROW start_ROW start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL end_ROW start_ROW start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL end_ROW end_ARRAY ) ( start_ARRAY start_ROW start_CELL italic_X end_CELL start_CELL end_CELL start_CELL end_CELL start_CELL end_CELL end_ROW start_ROW start_CELL end_CELL start_CELL italic_X end_CELL start_CELL end_CELL start_CELL end_CELL end_ROW start_ROW start_CELL end_CELL start_CELL end_CELL start_CELL italic_X end_CELL start_CELL end_CELL end_ROW start_ROW start_CELL end_CELL start_CELL end_CELL start_CELL end_CELL start_CELL italic_X end_CELL end_ROW end_ARRAY ) ( start_ARRAY start_ROW start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL end_ROW start_ROW start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL end_ROW start_ROW start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL end_ROW start_ROW start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL start_CELL italic_X end_CELL end_ROW end_ARRAY ) , (16)

where the size of the characters represents the numerical scales.

In our implementation, we use a QDR decomposition, i.e., U=QDR𝑈𝑄𝐷𝑅U=QDRitalic_U = italic_Q italic_D italic_R, where Q𝑄Qitalic_Q is a unitary matrix, R𝑅Ritalic_R is upper right triangular, and D𝐷Ditalic_D is diagonal. To diagonalize U𝑈Uitalic_U, one cannot multiply out the factors QDR𝑄𝐷𝑅QDRitalic_Q italic_D italic_R in this order without losing information contained in the smaller scales of D𝐷Ditalic_D. Using the standard method to calculate TrN(U^σ)subscriptTr𝑁subscript^𝑈𝜎\textrm{Tr}_{N}(\hat{U}_{\sigma})Tr start_POSTSUBSCRIPT italic_N end_POSTSUBSCRIPT ( over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ) stabily, which involves a QDR decomposition for each Fourier sum component in the canonical trace, would lead to scaling with system size of O(Ns4)𝑂superscriptsubscript𝑁𝑠4O({N_{s}}^{4})italic_O ( italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT ) (as opposed to O(Ns3)𝑂superscriptsubscript𝑁𝑠3O({N_{s}}^{3})italic_O ( italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ) for grand canonical calculations). This is costly as the number of single-particle states Nssubscript𝑁𝑠N_{s}italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT grows larger. Instead, one can diagonalize the matrix S=DRQ𝑆𝐷𝑅𝑄S=DRQitalic_S = italic_D italic_R italic_Q obtained from U𝑈Uitalic_U by a similarity transformation, and from this calculate number-projected observables. This procedure reduces the number of operations required for numerically stabilized canonical-ensemble calculations from O(Ns4)𝑂superscriptsubscript𝑁𝑠4O(N_{s}^{4})italic_O ( italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 4 end_POSTSUPERSCRIPT ) (without diagonalization) to O(Ns3)𝑂superscriptsubscript𝑁𝑠3O(N_{s}^{3})italic_O ( italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ) Gilbreth and Alhassid (2015). This method is used in our code. Recently, a new algorithm in the canonical ensemble was introduced in Ref. Shen et al. (2023) using a recursive method which was shown to be numerically stable. It will be interesting to compare both methods in future work.

III.6 Model space truncation

In our simulations we use the model space truncation method first introduced in Ref. Jensen et al. (2019) and presented in detail in Ref. Gilbreth et al. (2021). This allows for substantial savings in computational time.

The main idea of the method is to reduce the dimension of the one-body propagator in the decomposition U=QDR𝑈𝑄𝐷𝑅U=QDRitalic_U = italic_Q italic_D italic_R from Nssubscript𝑁𝑠N_{s}italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT to Noccsubscript𝑁occN_{\rm occ}italic_N start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT, where Noccsubscript𝑁occN_{\rm occ}italic_N start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT, the number of significantly occupied single-particle states, is of the order of particle number N𝑁Nitalic_N. The entries of the diagonal matrix D𝐷Ditalic_D can be sorted in an appropriate descending order and decompose D=Docc+Dε𝐷subscript𝐷occsubscript𝐷𝜀D=D_{\rm occ}+D_{\varepsilon}italic_D = italic_D start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT + italic_D start_POSTSUBSCRIPT italic_ε end_POSTSUBSCRIPT, where Doccsubscript𝐷occD_{\rm occ}italic_D start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT is obtained by zeroing out the diagonal elements for row i>Nocc𝑖subscript𝑁occi>N_{\rm occ}italic_i > italic_N start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT (and the discarded terms are in Dεsubscript𝐷𝜀D_{\varepsilon}italic_D start_POSTSUBSCRIPT italic_ε end_POSTSUBSCRIPT). This leads to

U=QDoccR+QDεR=Uocc+E,𝑈𝑄subscript𝐷occ𝑅𝑄subscript𝐷𝜀𝑅subscript𝑈occ𝐸U=QD_{\rm occ}R+QD_{\varepsilon}R=U_{\rm occ}+E\;,italic_U = italic_Q italic_D start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT italic_R + italic_Q italic_D start_POSTSUBSCRIPT italic_ε end_POSTSUBSCRIPT italic_R = italic_U start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT + italic_E , (17)

where Uocc=QDoccRsubscript𝑈occ𝑄subscript𝐷occ𝑅U_{\rm occ}=QD_{\rm occ}Ritalic_U start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT = italic_Q italic_D start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT italic_R represents the occupied part of U𝑈Uitalic_U and E=QDεR𝐸𝑄subscript𝐷𝜀𝑅E=QD_{\varepsilon}Ritalic_E = italic_Q italic_D start_POSTSUBSCRIPT italic_ε end_POSTSUBSCRIPT italic_R is a small perturbation. All observables are then calculated using the reduced-rank matrix Uoccsubscript𝑈occU_{\rm occ}italic_U start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT. The error made is represented by the matrix E𝐸Eitalic_E with norm Enorm𝐸||E||| | italic_E | |. Our algorithm implementation uses the Frobenius norm and we choose Noccsubscript𝑁occN_{\rm occ}italic_N start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT such that the error matrix E𝐸Eitalic_E satisfies E<εnorm𝐸𝜀||E||<\varepsilon| | italic_E | | < italic_ε, where ε𝜀\varepsilonitalic_ε is a given small parameter. Ref. Gilbreth et al. (2021) provides an error analysis and further details of the truncation method.

The benefit of the truncation is to replace the original O(NτNs3)𝑂subscript𝑁𝜏superscriptsubscript𝑁𝑠3O(N_{\tau}N_{s}^{3})italic_O ( italic_N start_POSTSUBSCRIPT italic_τ end_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ) time complexity of computing the partition function for a given field configuration σ𝜎\sigmaitalic_σ of Eq. 12 in the complete single-particle model space with O(NτNsNocc2)𝑂subscript𝑁𝜏subscript𝑁𝑠superscriptsubscript𝑁occ2O(N_{\tau}N_{s}N_{\rm occ}^{2})italic_O ( italic_N start_POSTSUBSCRIPT italic_τ end_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ) complexity. This reduction in complexity becomes particularly large at low temperatures where Noccsubscript𝑁occN_{\rm occ}italic_N start_POSTSUBSCRIPT roman_occ end_POSTSUBSCRIPT is comparable to particle number N𝑁Nitalic_N. Further optimizations to reduce the O(Ns3)𝑂superscriptsubscript𝑁𝑠3O(N_{s}^{3})italic_O ( italic_N start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT ) complexity of the QR𝑄𝑅QRitalic_Q italic_R decompositions were introduced in the work of Ref. He et al. (2019) and applied in grand-canonical simulations of the two-dimensional BCS-BEC crossover He et al. (2022). These additional optimizations will be useful for reaching even larger lattice sizes and, in particular, lower temperatures in future studies of the UFG.

IV Results and analysis

In Ref. Jensen et al. (2020a) we performed simulations for N=20,40,80𝑁204080N=20,40,80italic_N = 20 , 40 , 80 and 130130130130 particles for a filling factor of ν0.06𝜈0.06\nu\approx 0.06italic_ν ≈ 0.06. The finite filling factor introduces a systematic error as discussed in Sec. II.4. In Ref. Jensen et al. (2020b), we eliminated this systematic error when calculating the temperature dependence of Tan’s contact across the superfluid phase transition for N=40,66𝑁4066N=40,66italic_N = 40 , 66 and 114114114114 particles. This was accomplished by extrapolating to the continuum limit using results obtained for lattice sizes of NL3=53,73,93,113,133superscriptsubscript𝑁𝐿3superscript53superscript73superscript93superscript113superscript133N_{L}^{3}=5^{3},7^{3},9^{3},11^{3},13^{3}italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT = 5 start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT , 7 start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT , 9 start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT , 11 start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT , 13 start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT and 153superscript15315^{3}15 start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT. Here we extend the simulations of Ref. Jensen et al. (2020b) to extrapolate to the continuum limit for several other thermodynamic observables including the condensate fraction, the energy-staggering pairing gap, the spin susceptibility, and the free energy-staggering pairing gap. We also performed simulations for N=54𝑁54N=54italic_N = 54 particles for temperatures within the critical region to reduce the statistical error of our estimate for the critical temperature calculated by finite-size scaling.

In our calculations, for each particle number N𝑁Nitalic_N, temperature T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT, and lattice size NL3superscriptsubscript𝑁𝐿3N_{L}^{3}italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT, we also carry out simulations with several (typically four) values of the time slice ΔβΔ𝛽\Delta\betaroman_Δ italic_β and extrapolate to the continuous imaginary-time limit Δβ0Δ𝛽0\Delta\beta\to 0roman_Δ italic_β → 0.

In Appendix A, we discuss further details of the extrapolations to continuous time Δβ0Δ𝛽0\Delta\beta\to 0roman_Δ italic_β → 0 and to the continuum limit ν0𝜈0\nu\to 0italic_ν → 0 and present typical extrapolation plots for the condensate fraction and free energy staggering pairing gap.

IV.1 Condensate Fraction

We determine the superfluid critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT by analyzing the condensate fraction n𝑛nitalic_n as a function of temperature T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT and particle number N𝑁Nitalic_N in the continuum limit of the lattice model. We define the condensate fraction n𝑛nitalic_n by

n=λ(N/2),𝑛𝜆𝑁2n=\frac{\lambda}{(N/2)}\;,italic_n = divide start_ARG italic_λ end_ARG start_ARG ( italic_N / 2 ) end_ARG , (18)

where λ𝜆\lambdaitalic_λ is the largest eigenvalue of the two-body density matrix ψ^𝐤1,ψ^𝐤2,ψ^𝐤3,ψ^𝐤4,delimited-⟨⟩subscriptsuperscript^𝜓subscript𝐤1subscriptsuperscript^𝜓subscript𝐤2subscript^𝜓subscript𝐤3subscript^𝜓subscript𝐤4\langle\hat{\psi}^{\dagger}_{\mathbf{k}_{1},\uparrow}\hat{\psi}^{\dagger}_{% \mathbf{k}_{2},\downarrow}\hat{\psi}_{\mathbf{k}_{3},\downarrow}\hat{\psi}_{% \mathbf{k}_{4},\uparrow}\rangle⟨ over^ start_ARG italic_ψ end_ARG start_POSTSUPERSCRIPT † end_POSTSUPERSCRIPT start_POSTSUBSCRIPT bold_k start_POSTSUBSCRIPT 1 end_POSTSUBSCRIPT , ↑ end_POSTSUBSCRIPT over^ start_ARG italic_ψ end_ARG start_POSTSUPERSCRIPT † end_POSTSUPERSCRIPT start_POSTSUBSCRIPT bold_k start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT , ↓ end_POSTSUBSCRIPT over^ start_ARG italic_ψ end_ARG start_POSTSUBSCRIPT bold_k start_POSTSUBSCRIPT 3 end_POSTSUBSCRIPT , ↓ end_POSTSUBSCRIPT over^ start_ARG italic_ψ end_ARG start_POSTSUBSCRIPT bold_k start_POSTSUBSCRIPT 4 end_POSTSUBSCRIPT , ↑ end_POSTSUBSCRIPT ⟩, and N/2𝑁2N/2italic_N / 2 is the maximal number of pairs. For fermions, λ𝜆\lambdaitalic_λ can be shown to satisfy

λN(MN/2+1)/(2M)N/2,𝜆𝑁𝑀𝑁212𝑀𝑁2\lambda\leq{N(M-N/2+1)}/(2M)\leq N/2\;,italic_λ ≤ italic_N ( italic_M - italic_N / 2 + 1 ) / ( 2 italic_M ) ≤ italic_N / 2 , (19)

where M=NL3𝑀superscriptsubscript𝑁𝐿3M=N_{L}^{3}italic_M = italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT is the number of lattice points. Having a large eigenvalue λ𝜆\lambdaitalic_λ that scales with the system size is referred to as off-diagonal long-range order (ODLRO) Yang (1962).

Refer to caption
Figure 1: AFMC condensate fraction n𝑛nitalic_n in the continuum limit for N=40𝑁40N=40italic_N = 40 (solid blue squares), N=66𝑁66N=66italic_N = 66 (solid orange circles), and N=114𝑁114N=114italic_N = 114 (solid green down triangles) particles. We also include results for N=54𝑁54N=54italic_N = 54 particles (solid red diamonds) in the critical regime. For reference we show the T=0𝑇0T=0italic_T = 0 quantum Monte Carlo result of Ref. Astrakharchik et al. (2005) (open orange pentagon), the T=0𝑇0T=0italic_T = 0 QMC result of Ref. He et al. (2020) (purple cross), the recent low-T experimental result of Ref. Kwon et al. (2020) (open up yellow triangle), and the finite temperature rapid ramp experimental result of Ref. Ku et al. (2012) (black open circles). The grey band represents the critical temperature Tc=0.16(1)TFsubscript𝑇𝑐0.161subscript𝑇𝐹T_{c}=0.16(1)\,T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.16 ( 1 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT determined from finite-size scaling.

In Fig. 1 we show the condensate fraction n𝑛nitalic_n (after the continuous time and continuum extrapolations discussed in Appendix A.1) as a function of temperature T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT for N=40𝑁40N=40italic_N = 40 (solid blue squares), N=66𝑁66N=66italic_N = 66 (solid orange circles), and N=114𝑁114N=114italic_N = 114 (solid green down triangles) particles. We also show results for N=54𝑁54N=54italic_N = 54 particles (solid red diamonds) in the vicinity of the critical temperature. There is a noticeable difference between the results of Ref. Jensen et al. (2020a) and our current continuum results. This is expected and demonstrates the importance of the continuum limit extrapolations in obtaining reliable results from lattice simulations. We also show the T=0𝑇0T=0italic_T = 0 quantum Monte Carlo result of Ref. Astrakharchik et al. (2005) (open orange pentagon) where a diffusion Monte Carlo calculation was carried out for the full BCS to BEC crossover, and the T=0𝑇0T=0italic_T = 0 quantum Monte Carlo (QMC) result of Ref. He et al. (2020) (purple cross) where a more complex lattice interaction is introduced to accelerate the convergence to the continuum limit. We compare with the recent low-temperature Josephson current experimental results of Ref. Kwon et al. (2020) (open up yellow triangle) and the finite temperature rapid ramp experimental result of Ref. Ku et al. (2012) (open black circles).

The grey band represents our estimate for the critical temperature Tc=0.16(1)TFsubscript𝑇𝑐0.161subscript𝑇𝐹T_{c}=0.16(1)T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.16 ( 1 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT from a finite-size scaling analysis (see Sec. IV.2). We observe that the condensate fraction decreases with particle number N𝑁Nitalic_N in the normal phase as expected when the thermodynamic limit is approached in a finite-size system, but it does not show a noticeable trend with N𝑁Nitalic_N below the critical temperature. Our results for the condensate fraction are larger than the rapid ramp experimental results of Ref. Ku et al. (2012) for all temperatures. Comparing our T<<Tcmuch-less-than𝑇subscript𝑇𝑐T<<T_{c}italic_T < < italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT results with the T=0𝑇0T=0italic_T = 0 results of other works, we find good agreement with the diffusion Monte Carlo result of Ref. Astrakharchik et al. (2005) (open orange pentagon) and with the experimental result of Ref. Kwon et al. (2020) (open up yellow triangle). Our low-temperature results are slightly higher than the improved lattice interaction result of Ref. He et al. (2020) (purple cross).

IV.2 Finite-size scaling and critical temperature

To determine the superfluid critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT we apply a finite-size scaling analysis to the condensate fraction n𝑛nitalic_n for N=40,54,66𝑁405466N=40,54,66italic_N = 40 , 54 , 66 and N=114𝑁114N=114italic_N = 114 (the N=54𝑁54N=54italic_N = 54 simulations were performed only in the critical regime to improve our estimate of Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT). In Refs. Burovski et al. (2006a, b); Akkineni et al. (2007); Goulko and Wingate (2010) a finite-size scaling analysis, discussed in Ref. Binder (1981), was applied to the 3D XY phase transition of the UFG to determine the critical temperature. The critical behavior of the continuum limit condensate fraction n(L,T)𝑛𝐿𝑇n(L,T)italic_n ( italic_L , italic_T ) is captured by the scaling ansatz

L1+ηn(L,T)=f(x)[1+CLω+],superscript𝐿1𝜂𝑛𝐿𝑇𝑓𝑥delimited-[]1𝐶superscript𝐿𝜔L^{1+\eta}n(L,T)=f(x)\ [1+CL^{-\omega}+...\ ]\;,italic_L start_POSTSUPERSCRIPT 1 + italic_η end_POSTSUPERSCRIPT italic_n ( italic_L , italic_T ) = italic_f ( italic_x ) [ 1 + italic_C italic_L start_POSTSUPERSCRIPT - italic_ω end_POSTSUPERSCRIPT + … ] , (20)

where x=(L/ξ)1/νc𝑥superscript𝐿𝜉1subscript𝜈𝑐x=(L/\xi)^{1/\nu_{c}}italic_x = ( italic_L / italic_ξ ) start_POSTSUPERSCRIPT 1 / italic_ν start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT end_POSTSUPERSCRIPT, ξ𝜉\xiitalic_ξ the correlation length, f(x)𝑓𝑥f(x)italic_f ( italic_x ) is a universal scaling function, and C𝐶Citalic_C is a non-universal coefficient of the term for the leading correction to scaling with an exponent ω𝜔\omegaitalic_ω. The exponents of the U(1)𝑈1U(1)italic_U ( 1 ) universality class have previously been calculated with νc0.67similar-to-or-equalssubscript𝜈𝑐0.67\nu_{c}\simeq 0.67italic_ν start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT ≃ 0.67, η0.038similar-to-or-equals𝜂0.038\eta\simeq 0.038italic_η ≃ 0.038 with ω0.8𝜔0.8\omega\approx 0.8italic_ω ≈ 0.8 for the exponent of the leading irrelevant field Guida and Zinn-Justin (1998); Campostrini et al. (2006); Pelissetto and Vicari (2002).

Refer to caption
Figure 2: Continuum limit condensate fraction n(L,T)𝑛𝐿𝑇n(L,T)italic_n ( italic_L , italic_T ) scaled by L1+ηsuperscript𝐿1𝜂L^{1+\eta}italic_L start_POSTSUPERSCRIPT 1 + italic_η end_POSTSUPERSCRIPT as a function of temperature T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT for N=40𝑁40N=40italic_N = 40 (solid blue squares), N=54𝑁54N=54italic_N = 54 (solid red diamonds), N=66𝑁66N=66italic_N = 66 (solid orange down triangles), and N=114𝑁114N=114italic_N = 114 (solid green down triangles). Inset: the crossing temperatures T(Li,Lj)𝑇subscript𝐿𝑖subscript𝐿𝑗T(L_{i},L_{j})italic_T ( italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) as a function of g~(Li,Lj)~𝑔subscript𝐿𝑖subscript𝐿𝑗\tilde{g}(L_{i},L_{j})over~ start_ARG italic_g end_ARG ( italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) (discussed in the main text) and the fit (straight purple line) to extract the critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT. We find Tc=0.16(1)TFsubscript𝑇𝑐0.161subscript𝑇𝐹T_{c}=0.16(1)T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.16 ( 1 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT (solid black circle) indicated by the vertical grey band in the main figure.

In Fig. 2 we plot the condensate fraction in the continuum limit n(L,T)𝑛𝐿𝑇n(L,T)italic_n ( italic_L , italic_T ) scaled by L1+ηsuperscript𝐿1𝜂L^{1+\eta}italic_L start_POSTSUPERSCRIPT 1 + italic_η end_POSTSUPERSCRIPT as a function of T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT for the different particle numbers. L𝐿Litalic_L does not represent a simulated lattice size but the size of the physical system with N𝑁Nitalic_N particles. We arbitrarily choose L=1superscript𝐿1L^{\prime}=1italic_L start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT = 1 for N=40superscript𝑁40N^{\prime}=40italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT = 40 particles in Fig. 1(b) and the L𝐿Litalic_L values for the remaining particle numbers are determined through the relation N/L3=N/L3superscript𝑁superscript𝐿3𝑁superscript𝐿3N^{\prime}/L^{\prime 3}=N/L^{3}italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT / italic_L start_POSTSUPERSCRIPT ′ 3 end_POSTSUPERSCRIPT = italic_N / italic_L start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT. If the corrections to the scaling ansatz are vanishing the crossings of the curves for different particle number N𝑁Nitalic_N scaled by L1+ηsuperscript𝐿1𝜂L^{1+\eta}italic_L start_POSTSUPERSCRIPT 1 + italic_η end_POSTSUPERSCRIPT will occur at the superfluid critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT. However, the corrections to the scaling relation can be significant for the finite particle numbers simulated in the current work. We follow Ref. Burovski et al. (2006a, b); Goulko and Wingate (2010) and expand f(x)𝑓𝑥f(x)italic_f ( italic_x ) to leading order (f(x)𝑓𝑥f(x)italic_f ( italic_x ) is analytic at x=0𝑥0x=0italic_x = 0), where the correlation length is ξ|t|νcproportional-to𝜉superscript𝑡subscript𝜈𝑐\xi\propto|t|^{-\nu_{c}}italic_ξ ∝ | italic_t | start_POSTSUPERSCRIPT - italic_ν start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT end_POSTSUPERSCRIPT in the critical regime where the reduced temperature is t=(TTc)/Tc𝑡𝑇subscript𝑇𝑐subscript𝑇𝑐t=(T-T_{c})/T_{c}italic_t = ( italic_T - italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT ) / italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT. We then obtain the relation between the crossing temperature for system sizes Lisubscript𝐿𝑖L_{i}italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT and Ljsubscript𝐿𝑗L_{j}italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT, denoted by T(Li,Lj)𝑇subscript𝐿𝑖subscript𝐿𝑗T(L_{i},L_{j})italic_T ( italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ), and the thermodynamic critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT

|T(Li,Lj)Tc|=κg(Li,Lj)𝑇subscript𝐿𝑖subscript𝐿𝑗subscript𝑇𝑐𝜅𝑔subscript𝐿𝑖subscript𝐿𝑗|T(L_{i},L_{j})-T_{c}|=\kappa g(L_{i},L_{j})| italic_T ( italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) - italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT | = italic_κ italic_g ( italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) (21)

with

g(Li,Lj)=[LiωLjω][(Lj1/νcLi1/νc)+C(Lj1/νcωLi1/νcω)].𝑔subscript𝐿𝑖subscript𝐿𝑗delimited-[]superscriptsubscript𝐿𝑖𝜔superscriptsubscript𝐿𝑗𝜔delimited-[]superscriptsubscript𝐿𝑗1subscript𝜈𝑐superscriptsubscript𝐿𝑖1subscript𝜈𝑐𝐶superscriptsubscript𝐿𝑗1subscript𝜈𝑐𝜔superscriptsubscript𝐿𝑖1subscript𝜈𝑐𝜔g(L_{i},L_{j})=\frac{\ [L_{i}^{-\omega}-L_{j}^{-\omega}\ ]}{\ [(L_{j}^{1/\nu_{% c}}-L_{i}^{1/\nu_{c}})+C(L_{j}^{1/\nu_{c}-\omega}-L_{i}^{1/\nu_{c}-\omega})\ ]% }\;.italic_g ( italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) = divide start_ARG [ italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT start_POSTSUPERSCRIPT - italic_ω end_POSTSUPERSCRIPT - italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT start_POSTSUPERSCRIPT - italic_ω end_POSTSUPERSCRIPT ] end_ARG start_ARG [ ( italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 / italic_ν start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT end_POSTSUPERSCRIPT - italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 / italic_ν start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT end_POSTSUPERSCRIPT ) + italic_C ( italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 / italic_ν start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT - italic_ω end_POSTSUPERSCRIPT - italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 / italic_ν start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT - italic_ω end_POSTSUPERSCRIPT ) ] end_ARG . (22)

Here κ𝜅\kappaitalic_κ absorbs constants in the expansion of the universal function f(x)𝑓𝑥f(x)italic_f ( italic_x ). Refs. Burovski et al. (2006a, b) dropped the second term in the denominator of Eq. (22) to obtain a simplified relation

g~(Li,Lj)=[LiωLjω][(Lj1/νcLi1/νc)].~𝑔subscript𝐿𝑖subscript𝐿𝑗delimited-[]superscriptsubscript𝐿𝑖𝜔superscriptsubscript𝐿𝑗𝜔delimited-[]superscriptsubscript𝐿𝑗1subscript𝜈𝑐superscriptsubscript𝐿𝑖1subscript𝜈𝑐\widetilde{g}(L_{i},L_{j})=\frac{\ [L_{i}^{-\omega}-L_{j}^{-\omega}\ ]}{\ [(L_% {j}^{1/\nu_{c}}-L_{i}^{1/\nu_{c}})\ ]}.over~ start_ARG italic_g end_ARG ( italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) = divide start_ARG [ italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT start_POSTSUPERSCRIPT - italic_ω end_POSTSUPERSCRIPT - italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT start_POSTSUPERSCRIPT - italic_ω end_POSTSUPERSCRIPT ] end_ARG start_ARG [ ( italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 / italic_ν start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT end_POSTSUPERSCRIPT - italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 1 / italic_ν start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT end_POSTSUPERSCRIPT ) ] end_ARG . (23)

In this work the data is not precise enough to perform the analysis of Ref Goulko and Wingate (2010) and instead we extract Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT with the simplified crossings function g~(Li,Lj)~𝑔subscript𝐿𝑖subscript𝐿𝑗\widetilde{g}(L_{i},L_{j})over~ start_ARG italic_g end_ARG ( italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) . In the inset of Fig. 2 we show the crossing temperatures T(Li,Lj)𝑇subscript𝐿𝑖subscript𝐿𝑗T(L_{i},L_{j})italic_T ( italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) (open black circles) as a function of g~(Li,Lj)~𝑔subscript𝐿𝑖subscript𝐿𝑗\widetilde{g}(L_{i},L_{j})over~ start_ARG italic_g end_ARG ( italic_L start_POSTSUBSCRIPT italic_i end_POSTSUBSCRIPT , italic_L start_POSTSUBSCRIPT italic_j end_POSTSUBSCRIPT ) for N=40,54,66𝑁405466N=40,54,66italic_N = 40 , 54 , 66 and 114114114114 particles where the crossings are determined from straight line fits to the Monte Carlo data for each particle number N𝑁Nitalic_N. We have performed a generalized least square analysis to determine the critical temperature Tc=0.16(1)TFsubscript𝑇𝑐0.161subscript𝑇𝐹T_{c}=0.16(1)\,T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.16 ( 1 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT (solid black circle). This is larger than the estimate Tc=0.130(15)TFsubscript𝑇𝑐0.13015subscript𝑇𝐹T_{c}=0.130(15)\,T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.130 ( 15 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT of Ref. Jensen et al. (2020a), where simulations were performed at a finite filling factor of ν0.06𝜈0.06\nu\approx 0.06italic_ν ≈ 0.06. We observe that the critical temperature extracted from the AFMC data in the continuum limit is closer to the experimental value Tc=0.167(13)TFsubscript𝑇𝑐0.16713subscript𝑇𝐹T_{c}=0.167(13)T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.167 ( 13 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT of Ref. Ku et al. (2012) and to the continuum limit diagrammatic Monte Carlo estimates Tc=0.152(7)TFsubscript𝑇𝑐0.1527subscript𝑇𝐹T_{c}=0.152(7)\,T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.152 ( 7 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT of Ref. Burovski et al. (2006a) and Tc=0.173(6)TFsubscript𝑇𝑐0.1736subscript𝑇𝐹T_{c}=0.173(6)\,T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.173 ( 6 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT of Ref. Goulko and Wingate (2010). The path integral Monte Carlo work of Ref. Akkineni et al. (2007) found a higher critical temperature Tc=0.245TFsubscript𝑇𝑐0.245subscript𝑇𝐹T_{c}=0.245\,T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.245 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. There are several other theoretical estimates for the critical temperature in Refs. Bulgac et al. (2006, 2008); Lee and Schäfer (2006); Nishida (2007); Haussmann et al. (2007); Abe and Seki (2009); Akkineni et al. (2007); Floerchinger et al. (2010); Boettcher et al. (2014). Experimental results for Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT were obtained in the works of Refs. Ku et al. (2012); Nascimbene et al. (2010); Horikoshi et al. (2010). For reference we list the critical temperature estimates of various experiments and theoretical calculations in Table  1.

Table 1: Estimates of the critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT (in units of the Fermi temperature TFsubscript𝑇𝐹T_{F}italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT)
Method Tc(TF)subscript𝑇𝑐subscript𝑇𝐹T_{c}(T_{F})italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT ( italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ) error
MIT experiment Ku et al. (2012) 0.167 0.013
experiment Nascimbene et al. (2010) 0.157 0.015
experiment Horikoshi et al. (2010) 0.17 0.01
lattice DiagMC Burovski et al. (2006a) 0.152 0.007
lattice DiagMC Goulko and Wingate (2010) 0.173 0.006
lattice AFMC Richie-Halford et al. (2020) 0.16 0.02
lattice AFMC Bulgac et al. (2008) less-than-or-similar-to\lesssim 0.15 0.01
lattice AFMC Bulgac et al. (2006) 0.23 0.02
low-density neutron Abe and Seki (2009) 0.189 0.012
RPIMC Akkineni et al. (2007) 0.245absent0.245\approx 0.245≈ 0.245
lattice Monte Carlo Lee and Schäfer (2006) <<< 0.14
FRG Floerchinger et al. (2010) 0.248
ϵitalic-ϵ\epsilonitalic_ϵ-expansion matching from d=2,4 Nishida (2007) 0.183 0.014
Luttinger-Ward Haussmann et al. (2007) 0.183 0.014
lattice AFMC (this work) 0.16 0.01

IV.3 Energy-staggering pairing gap

Refer to caption
Figure 3: (a) AFMC energy-staggering pairing gap ΔEsubscriptΔ𝐸\Delta_{E}roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT (in units of the Fermi energy εFsubscript𝜀𝐹\varepsilon_{F}italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT) for N=40𝑁40N=40italic_N = 40 (solid blue squares), N=66𝑁66N=66italic_N = 66 (solid orange circles), and N=114𝑁114N=114italic_N = 114 (solid green down triangles) particles as a function of T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. For reference, we show the low-T experimental result of Ref. Schirotzek et al. (2008) (open down pink triangle), the low-T experimental result of Ref. Hoinka et al. (2017) (open up black triangle), and the low-T hybrid QMC experimental result of Refs. Carlson and Reddy (2008) (open purple diamond). The AFMC energy-staggering pairing gap vanishes above a temperature of T0.2TFsuperscript𝑇0.2subscript𝑇𝐹T^{*}\approx 0.2\,T_{F}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT for the largest particle number simulated. The gray band is our estimate of the critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT (in units of TFsubscript𝑇𝐹T_{F}italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT). (b) The AFMC spin susceptibility χssubscript𝜒𝑠\chi_{s}italic_χ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT (in units of χ0subscript𝜒0\chi_{0}italic_χ start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT) for different particle numbers as a function of T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. The convention of the symbols for different particle numbers is as in panel (a). For comparison we also show the spin susceptibility calculated in the Luttinger-Ward theory Enss and Haussmann (2012) (solid orange line), the t𝑡titalic_t-matrix results of Ref. Palestini et al. (2012b) (dotted line), the extended T𝑇Titalic_T-matrix result of Refs. Tajima et al. (2014, 2016) (dashed line), and the self-consistent Nozieres and Schmitt-Rink results of Ref. Pantel et al. (2014) (dashed-dotted line).
Refer to caption
Figure 4: AFMC free energy gap ΔFsubscriptΔ𝐹\Delta_{F}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT (in units of the Fermi energy εFsubscript𝜀𝐹\varepsilon_{F}italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT) for N=40𝑁40N=40italic_N = 40 (solid blue squares), N=66𝑁66N=66italic_N = 66 (solid orange diamonds), and N=114𝑁114N=114italic_N = 114 particles (solid green up triangles). We also show the linear extrapolations in T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT for N=40𝑁40N=40italic_N = 40 (blue dashed line) and N=66𝑁66N=66italic_N = 66 (pink dashed line), and the T=0𝑇0T=0italic_T = 0 energy-staggering pairing gap ΔE/εFsubscriptΔ𝐸subscript𝜀𝐹\Delta_{E}/\varepsilon_{F}roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT / italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT from the N=40𝑁40N=40italic_N = 40 (solid blue up triangle) and N=66𝑁66N=66italic_N = 66 (solid red up triangle) extrapolations. The gray band shows the critical temperature Tc=0.16(1)TFsubscript𝑇𝑐0.161subscript𝑇𝐹T_{c}=0.16(1)T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.16 ( 1 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT determined from a finite-size scaling analysis. For reference, we show the low-temperature gap from the QMC results of Refs. Carlson and Reddy (2005) (open orange circle) and Carlson and Reddy (2008) (open blue square), the lattice estimator QMC result of Ref. Lee et al. (2011) (solid yellow up triangle), and the experimental values of Ref. Schirotzek et al. (2008) (open down green triangle) and Ref. Hoinka et al. (2017) (open up black triangle). Inset: The free energy gap ΔFsubscriptΔ𝐹\Delta_{F}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT on a log-linear scale, showing more clearly the increase in ΔFsubscriptΔ𝐹\Delta_{F}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT as the temperature decreases below a certain temperature which we associate with a pairing temperature scale Tsuperscript𝑇T^{*}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT. For the largest particle number N=114𝑁114N=114italic_N = 114 we estimate an upper bound of T0.2TFsuperscript𝑇0.2subscript𝑇𝐹T^{*}\leq 0.2\,T_{F}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ≤ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT.

A signature of a pseudogap regime is the non-vanishing of a pairing gap above the critical temperature. We define a model-independent energy-staggering pairing gap ΔEsubscriptΔ𝐸\Delta_{E}roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT at temperature T𝑇Titalic_T for an even number of particles N𝑁Nitalic_N by

ΔE=[2E(N/2+1,N/2)E(N/2+1,N/2+1)E(N/2,N/2)]/2,subscriptΔ𝐸delimited-[]2𝐸𝑁21𝑁2𝐸𝑁21𝑁21𝐸𝑁2𝑁22\Delta_{E}=[2E(N/2+1,N/2)-E(N/2+1,N/2+1)\\ -E(N/2,N/2)]/2\;,start_ROW start_CELL roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT = [ 2 italic_E ( italic_N / 2 + 1 , italic_N / 2 ) - italic_E ( italic_N / 2 + 1 , italic_N / 2 + 1 ) end_CELL end_ROW start_ROW start_CELL - italic_E ( italic_N / 2 , italic_N / 2 ) ] / 2 , end_CELL end_ROW (24)

where E(N,N)𝐸subscript𝑁subscript𝑁E(N_{\uparrow},N_{\downarrow})italic_E ( italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT ) is the thermal energy for Nsubscript𝑁N_{\uparrow}italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT spin-up particles and Nsubscript𝑁N_{\downarrow}italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT spin-down particles. This observable was first calculated for the UFG at T=0𝑇0T=0italic_T = 0 in Ref. Carlson et al. (2003) and at finite temperature in the works of Refs. Jensen et al. (2019, 2020a).

The calculation of ΔEsubscriptΔ𝐸\Delta_{E}roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT at finite temperature requires the use of the canonical ensemble of fixed particle numbers. To this end, we use the particle-number reprojection method of Ref. Alhassid et al. (1999). The expectation value of the Hamiltonian H^^𝐻\hat{H}over^ start_ARG italic_H end_ARG for fixed particle numbers Nsubscript𝑁N_{\uparrow}italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT and Nsubscript𝑁N_{\downarrow}italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT is given by

H^N,N=TrN,N(H^eβH^)TrN,N(eβH^)=D[σ]H^N,N,σWσΦσD[σ]WσΦσsubscriptdelimited-⟨⟩^𝐻subscript𝑁subscript𝑁subscriptTrsubscript𝑁subscript𝑁^𝐻superscript𝑒𝛽^𝐻subscriptTrsubscript𝑁subscript𝑁superscript𝑒𝛽^𝐻𝐷delimited-[]𝜎subscriptdelimited-⟨⟩^𝐻subscript𝑁subscript𝑁𝜎subscript𝑊𝜎subscriptΦ𝜎𝐷delimited-[]𝜎subscript𝑊𝜎subscriptΦ𝜎\langle\hat{H}\rangle_{N_{\uparrow},N_{\downarrow}}=\frac{\textrm{Tr}_{N_{% \uparrow},N_{\downarrow}}(\hat{H}e^{-\beta\hat{H}})}{\textrm{Tr}_{N_{\uparrow}% ,N_{\downarrow}}(e^{-\beta\hat{H}})}=\frac{\int D[\sigma]\langle\hat{H}\rangle% _{N_{\uparrow},N_{\downarrow},\sigma}W_{\sigma}\Phi_{\sigma}}{\int D[\sigma]W_% {\sigma}\Phi_{\sigma}}⟨ over^ start_ARG italic_H end_ARG ⟩ start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT = divide start_ARG Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( over^ start_ARG italic_H end_ARG italic_e start_POSTSUPERSCRIPT - italic_β over^ start_ARG italic_H end_ARG end_POSTSUPERSCRIPT ) end_ARG start_ARG Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( italic_e start_POSTSUPERSCRIPT - italic_β over^ start_ARG italic_H end_ARG end_POSTSUPERSCRIPT ) end_ARG = divide start_ARG ∫ italic_D [ italic_σ ] ⟨ over^ start_ARG italic_H end_ARG ⟩ start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT , italic_σ end_POSTSUBSCRIPT italic_W start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT roman_Φ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG start_ARG ∫ italic_D [ italic_σ ] italic_W start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT roman_Φ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG (25)

where

Wσ=Gσ|TrN,NU^σ|subscript𝑊𝜎subscript𝐺𝜎subscriptTrsubscript𝑁subscript𝑁subscript^𝑈𝜎W_{\sigma}=G_{\sigma}|\textrm{Tr}_{N_{\uparrow},N_{\downarrow}}\hat{U}_{\sigma}|italic_W start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = italic_G start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT | Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT | (26)

is a positive definite weight function used for the Monte Carlo sampling, Φσ=TrN,N(U^σ)/|TrN,N(U^σ)|subscriptΦ𝜎subscriptTrsubscript𝑁subscript𝑁subscript^𝑈𝜎subscriptTrsubscript𝑁subscript𝑁subscript^𝑈𝜎\Phi_{\sigma}=\textrm{Tr}_{N_{\uparrow},N_{\downarrow}}(\hat{U}_{\sigma})/|% \textrm{Tr}_{N_{\uparrow},N_{\downarrow}}(\hat{U}_{\sigma})|roman_Φ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ) / | Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ) | is the Monte Carlo sign, and

H^N,N,σ=TrN,N(H^U^σ)TrN,N(U^σ),subscriptdelimited-⟨⟩^𝐻subscript𝑁subscript𝑁𝜎subscriptTrsubscript𝑁subscript𝑁^𝐻subscript^𝑈𝜎subscriptTrsubscript𝑁subscript𝑁subscript^𝑈𝜎\langle\hat{H}\rangle_{N_{\uparrow},N_{\downarrow},\sigma}={\textrm{Tr}_{N_{% \uparrow},N_{\downarrow}}(\hat{H}\hat{U}_{\sigma})\over\textrm{Tr}_{N_{% \uparrow},N_{\downarrow}}(\hat{U}_{\sigma})}\;,⟨ over^ start_ARG italic_H end_ARG ⟩ start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT , italic_σ end_POSTSUBSCRIPT = divide start_ARG Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( over^ start_ARG italic_H end_ARG over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ) end_ARG start_ARG Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT ( over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ) end_ARG , (27)

is the expectation value of H𝐻Hitalic_H at fixed spin-up ad spin-down particle numbers for a given configuration σ𝜎\sigmaitalic_σ of the auxiliary fields. We use exact particle-projection operators given by Eq. (11).

To calculate H^N,Nsubscriptdelimited-⟨⟩^𝐻subscriptsuperscript𝑁subscriptsuperscript𝑁\langle\hat{H}\rangle_{N^{\prime}_{\uparrow},N^{\prime}_{\downarrow}}⟨ over^ start_ARG italic_H end_ARG ⟩ start_POSTSUBSCRIPT italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT for different particle numbers N,Nsubscriptsuperscript𝑁subscriptsuperscript𝑁N^{\prime}_{\uparrow},N^{\prime}_{\downarrow}italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT, we use the original Monte Carlo sampling for N,Nsubscript𝑁subscript𝑁N_{\uparrow},N_{\downarrow}italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT particles but reproject on the new particle numbers Alhassid et al. (1999).

H^N,N=H^N,N,σTrN,NU^σTrN,NU^σΦσWTrN,NU^σTrN,NU^σΦσW,subscriptdelimited-⟨⟩^𝐻subscriptsuperscript𝑁subscriptsuperscript𝑁subscriptdelimited-⟨⟩subscriptdelimited-⟨⟩^𝐻subscriptsuperscript𝑁subscriptsuperscript𝑁𝜎subscriptTrsubscriptsuperscript𝑁subscriptsuperscript𝑁subscript^𝑈𝜎subscriptTrsubscript𝑁subscript𝑁subscript^𝑈𝜎subscriptΦ𝜎𝑊subscriptdelimited-⟨⟩subscriptTrsubscriptsuperscript𝑁subscriptsuperscript𝑁subscript^𝑈𝜎subscriptTrsubscript𝑁subscript𝑁subscript^𝑈𝜎subscriptΦ𝜎𝑊\langle\hat{H}\rangle_{N^{\prime}_{\uparrow},N^{\prime}_{\downarrow}}={\left% \langle\langle\hat{H}\rangle_{N^{\prime}_{\uparrow},N^{\prime}_{\downarrow},% \sigma}\frac{\textrm{Tr}_{N^{\prime}_{\uparrow},N^{\prime}_{\downarrow}}\hat{U% }_{\sigma}}{{\textrm{Tr}_{N_{\uparrow},N_{\downarrow}}\hat{U}_{\sigma}}}\Phi_{% \sigma}\right\rangle_{W}\over\left\langle\frac{\textrm{Tr}_{N^{\prime}_{% \uparrow},N^{\prime}_{\downarrow}}\hat{U}_{\sigma}}{{\textrm{Tr}_{N_{\uparrow}% ,N_{\downarrow}}\hat{U}_{\sigma}}}\Phi_{\sigma}\right\rangle_{W}}\;,⟨ over^ start_ARG italic_H end_ARG ⟩ start_POSTSUBSCRIPT italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT = divide start_ARG ⟨ ⟨ over^ start_ARG italic_H end_ARG ⟩ start_POSTSUBSCRIPT italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT , italic_σ end_POSTSUBSCRIPT divide start_ARG Tr start_POSTSUBSCRIPT italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG start_ARG Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG roman_Φ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ⟩ start_POSTSUBSCRIPT italic_W end_POSTSUBSCRIPT end_ARG start_ARG ⟨ divide start_ARG Tr start_POSTSUBSCRIPT italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUPERSCRIPT ′ end_POSTSUPERSCRIPT start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG start_ARG Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG roman_Φ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ⟩ start_POSTSUBSCRIPT italic_W end_POSTSUBSCRIPT end_ARG , (28)

where we have introduced the notation

XσWD[σ]WσXσD[σ]Wσ.subscriptdelimited-⟨⟩subscript𝑋𝜎𝑊𝐷delimited-[]𝜎subscript𝑊𝜎subscript𝑋𝜎𝐷delimited-[]𝜎subscript𝑊𝜎\langle X_{\sigma}\rangle_{W}\equiv{{\int D[\sigma]W_{\sigma}X_{\sigma}\over% \int D[\sigma]W_{\sigma}}}\;.⟨ italic_X start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ⟩ start_POSTSUBSCRIPT italic_W end_POSTSUBSCRIPT ≡ divide start_ARG ∫ italic_D [ italic_σ ] italic_W start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT italic_X start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG start_ARG ∫ italic_D [ italic_σ ] italic_W start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG . (29)

In Fig. 3(a) we show the continuum limit energy-staggering pairing gap ΔEsubscriptΔ𝐸\Delta_{E}roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT (measured in units of the Fermi energy εFsubscript𝜀𝐹\varepsilon_{F}italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT) as a function of temperature T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT for N=40,66𝑁4066N=40,66italic_N = 40 , 66 and 114114114114 particles. We observe that for the largest particle number (N=114𝑁114N=114italic_N = 114), the pairing gap vanishes above a temperature of T0.2TFsuperscript𝑇0.2subscript𝑇𝐹T^{*}\approx 0.2\,T_{F}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. This suggests that the pseudogap regime, in which the pairing gap does not vanish above Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT is considerably narrower than previously believed.

IV.4 Spin susceptibility

The spin susceptibility χssubscript𝜒𝑠\chi_{s}italic_χ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT is suppressed by pairing correlations and is another signature of a pseudogap regime. It is given by

χs=βV(N^N^)2,subscript𝜒𝑠𝛽𝑉delimited-⟨⟩superscriptsubscript^𝑁subscript^𝑁2\chi_{s}=\frac{\beta}{V}\langle(\hat{N}_{\uparrow}-\hat{N}_{\downarrow})^{2}% \rangle\;,italic_χ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT = divide start_ARG italic_β end_ARG start_ARG italic_V end_ARG ⟨ ( over^ start_ARG italic_N end_ARG start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT - over^ start_ARG italic_N end_ARG start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ⟩ , (30)

where the expectation value on the r.h.s. is calculated for the spin-balanced system N^=N^delimited-⟨⟩subscript^𝑁delimited-⟨⟩subscript^𝑁\langle\hat{N}_{\uparrow}\rangle=\langle\hat{N}_{\downarrow}\rangle⟨ over^ start_ARG italic_N end_ARG start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT ⟩ = ⟨ over^ start_ARG italic_N end_ARG start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT ⟩ using only one particle projection on total number of particles N=N+N𝑁subscript𝑁subscript𝑁N=N_{\uparrow}+N_{\downarrow}italic_N = italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT + italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT in Eq. (30).

We calculated the spin susceptibility using the same Monte Carlo sampling at fixed numbers of spin-up and spin-down particles, i.e., according to Wσsubscript𝑊𝜎W_{\sigma}italic_W start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT defined in Eq. (26). We then have

χs=βV(N^N^)2σTrNU^σTrN,NU^σΦσTrNU^σTrN,NU^σΦσ,subscript𝜒𝑠𝛽𝑉delimited-⟨⟩subscriptdelimited-⟨⟩superscriptsubscript^𝑁subscript^𝑁2𝜎subscriptTr𝑁subscript^𝑈𝜎subscriptTrsubscript𝑁subscript𝑁subscript^𝑈𝜎subscriptΦ𝜎delimited-⟨⟩subscriptTr𝑁subscript^𝑈𝜎subscriptTrsubscript𝑁subscript𝑁subscript^𝑈𝜎subscriptΦ𝜎\chi_{s}=\frac{\beta}{V}{\left\langle\langle(\hat{N}_{\uparrow}-\hat{N}_{% \downarrow})^{2}\rangle_{\sigma}\frac{\textrm{Tr}_{N}\hat{U}_{\sigma}}{{% \textrm{Tr}_{N_{\uparrow},N_{\downarrow}}\hat{U}_{\sigma}}}\Phi_{\sigma}\right% \rangle\over\left\langle\frac{\textrm{Tr}_{N}\hat{U}_{\sigma}}{{\textrm{Tr}_{N% _{\uparrow},N_{\downarrow}}\hat{U}_{\sigma}}}\Phi_{\sigma}\right\rangle}\;,italic_χ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT = divide start_ARG italic_β end_ARG start_ARG italic_V end_ARG divide start_ARG ⟨ ⟨ ( over^ start_ARG italic_N end_ARG start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT - over^ start_ARG italic_N end_ARG start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ⟩ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT divide start_ARG Tr start_POSTSUBSCRIPT italic_N end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG start_ARG Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG roman_Φ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ⟩ end_ARG start_ARG ⟨ divide start_ARG Tr start_POSTSUBSCRIPT italic_N end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG start_ARG Tr start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT over^ start_ARG italic_U end_ARG start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG roman_Φ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT ⟩ end_ARG , (31)

where (N^N^)2σsubscriptdelimited-⟨⟩superscriptsubscript^𝑁subscript^𝑁2𝜎\langle(\hat{N}_{\uparrow}-\hat{N}_{\downarrow})^{2}\rangle_{\sigma}⟨ ( over^ start_ARG italic_N end_ARG start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT - over^ start_ARG italic_N end_ARG start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ⟩ start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT is the expectation value of (N^N^)2superscriptsubscript^𝑁subscript^𝑁2(\hat{N}_{\uparrow}-\hat{N}_{\downarrow})^{2}( over^ start_ARG italic_N end_ARG start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT - over^ start_ARG italic_N end_ARG start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT at fixed total particle number N𝑁Nitalic_N for a given field configuration σ𝜎\sigmaitalic_σ.

In Fig. 3(b) we show the continuum limit spin susceptibility χssubscript𝜒𝑠\chi_{s}italic_χ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT (in units of the T=0𝑇0T=0italic_T = 0 free Fermi gas susceptibility χ0=3ρ/2εFsubscript𝜒03𝜌2subscript𝜀𝐹\chi_{0}=3\rho/2\varepsilon_{F}italic_χ start_POSTSUBSCRIPT 0 end_POSTSUBSCRIPT = 3 italic_ρ / 2 italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT) as a function of T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT for particle numbers N=44,66𝑁4466N=44,66italic_N = 44 , 66 and 114114114114. We observe the strong suppression of χssubscript𝜒𝑠\chi_{s}italic_χ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT below the critical temperature due to pairing correlations. However, we also observe moderate suppression above Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT. While there is a clear particle number dependence and we have not reached the thermodynamic limit, we can put an upper bound on a spin gap temperature Ts0.2TFsubscript𝑇𝑠0.2subscript𝑇𝐹T_{s}\approx 0.2\,T_{F}italic_T start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT below which χssubscript𝜒𝑠\chi_{s}italic_χ start_POSTSUBSCRIPT italic_s end_POSTSUBSCRIPT starts to become suppressed. We observe that the value of the spin gap temperature is similar to that of Tsuperscript𝑇T^{*}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT.

Based on our results in Fig. 3 for the energy-staggering pairing gap and spin susceptibility, we estimate the pseudogap regime above Tc0.16TFsubscript𝑇𝑐0.16subscript𝑇𝐹T_{c}\approx 0.16\,T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT ≈ 0.16 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT to be below T0.2TFsuperscript𝑇0.2subscript𝑇𝐹T^{*}\approx 0.2\,T_{F}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT.

IV.5 Free energy gap and extraction of the T=0𝑇0T=0italic_T = 0 gap

The energy-staggering pairing gap is noisy, which makes the determination of an appropriate cutoff in filling factor ν𝜈\nuitalic_ν for the continuum extrapolation rather difficult. A much less noisy quantity to investigate pairing correlations is the free energy staggering pairing gap, which is defined by a formula analogous to Eq. (24) but with the thermal energy replaced by the free energy

ΔF=[2F(N/2+1,N/2)F(N/2+1,N/2+1)F(N/2,N/2)]/2.subscriptΔ𝐹delimited-[]2𝐹𝑁21𝑁2𝐹𝑁21𝑁21𝐹𝑁2𝑁22\Delta_{F}=[2F(N/2+1,N/2)-F(N/2+1,N/2+1)\\ -F(N/2,N/2)]/2\;.start_ROW start_CELL roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT = [ 2 italic_F ( italic_N / 2 + 1 , italic_N / 2 ) - italic_F ( italic_N / 2 + 1 , italic_N / 2 + 1 ) end_CELL end_ROW start_ROW start_CELL - italic_F ( italic_N / 2 , italic_N / 2 ) ] / 2 . end_CELL end_ROW (32)

Here F(N,N)=kBTlnZN,N𝐹subscript𝑁subscript𝑁subscript𝑘𝐵𝑇subscript𝑍subscript𝑁subscript𝑁F(N_{\uparrow},N_{\downarrow})=-k_{B}T\ln Z_{N_{\uparrow},N_{\downarrow}}italic_F ( italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT ) = - italic_k start_POSTSUBSCRIPT italic_B end_POSTSUBSCRIPT italic_T roman_ln italic_Z start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT and ZN,Nsubscript𝑍subscript𝑁subscript𝑁Z_{N_{\uparrow},N_{\downarrow}}italic_Z start_POSTSUBSCRIPT italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT , italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT end_POSTSUBSCRIPT are, respectively, the free energy and partition function for Nsubscript𝑁N_{\uparrow}italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT spin-up particles and Nsubscript𝑁N_{\downarrow}italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT spin-down particles.

Within the AFMC framework, this observable is calculated using the particle-number reprojection method. We rewrite the free energy gap as

ΔF=kBTln(ZN/2,N/2+1ZN/2,N/2ZN/2,N/2+1ZN/2+1,N/2+1).subscriptΔ𝐹subscript𝑘𝐵𝑇subscript𝑍𝑁2𝑁21subscript𝑍𝑁2𝑁2subscript𝑍𝑁2𝑁21subscript𝑍𝑁21𝑁21\Delta_{F}=-k_{B}T\ln\left(\frac{Z_{N/2,N/2+1}}{Z_{N/2,N/2}}\frac{Z_{N/2,N/2+1% }}{Z_{N/2+1,N/2+1}}\right)\;.roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT = - italic_k start_POSTSUBSCRIPT italic_B end_POSTSUBSCRIPT italic_T roman_ln ( divide start_ARG italic_Z start_POSTSUBSCRIPT italic_N / 2 , italic_N / 2 + 1 end_POSTSUBSCRIPT end_ARG start_ARG italic_Z start_POSTSUBSCRIPT italic_N / 2 , italic_N / 2 end_POSTSUBSCRIPT end_ARG divide start_ARG italic_Z start_POSTSUBSCRIPT italic_N / 2 , italic_N / 2 + 1 end_POSTSUBSCRIPT end_ARG start_ARG italic_Z start_POSTSUBSCRIPT italic_N / 2 + 1 , italic_N / 2 + 1 end_POSTSUBSCRIPT end_ARG ) . (33)

Each partition function ratio can then be calculated using particle-number reprojection, e.g.,

ZN/2,N/2+1ZN/2,N/2=TrN/2,N/2+1UσTrN/2,N/2UσW,subscript𝑍𝑁2𝑁21subscript𝑍𝑁2𝑁2subscriptdelimited-⟨⟩subscriptTr𝑁2𝑁21subscript𝑈𝜎subscriptTr𝑁2𝑁2subscript𝑈𝜎𝑊\frac{Z_{N/2,N/2+1}}{Z_{N/2,N/2}}=\left\langle\frac{{\rm Tr}_{N/2,N/2+1}U_{% \sigma}}{{\rm Tr}_{N/2,N/2}U_{\sigma}}\right\rangle_{W}\;,divide start_ARG italic_Z start_POSTSUBSCRIPT italic_N / 2 , italic_N / 2 + 1 end_POSTSUBSCRIPT end_ARG start_ARG italic_Z start_POSTSUBSCRIPT italic_N / 2 , italic_N / 2 end_POSTSUBSCRIPT end_ARG = ⟨ divide start_ARG roman_Tr start_POSTSUBSCRIPT italic_N / 2 , italic_N / 2 + 1 end_POSTSUBSCRIPT italic_U start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG start_ARG roman_Tr start_POSTSUBSCRIPT italic_N / 2 , italic_N / 2 end_POSTSUBSCRIPT italic_U start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT end_ARG ⟩ start_POSTSUBSCRIPT italic_W end_POSTSUBSCRIPT , (34)

where Wσ=GσTrN/2,N/2Uσsubscript𝑊𝜎subscript𝐺𝜎subscriptTr𝑁2𝑁2subscript𝑈𝜎W_{\sigma}=G_{\sigma}{\rm Tr}_{N/2,N/2}U_{\sigma}italic_W start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT = italic_G start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT roman_Tr start_POSTSUBSCRIPT italic_N / 2 , italic_N / 2 end_POSTSUBSCRIPT italic_U start_POSTSUBSCRIPT italic_σ end_POSTSUBSCRIPT is positive definite for N=N=N/2subscript𝑁subscript𝑁𝑁2N_{\uparrow}=N_{\downarrow}=N/2italic_N start_POSTSUBSCRIPT ↑ end_POSTSUBSCRIPT = italic_N start_POSTSUBSCRIPT ↓ end_POSTSUBSCRIPT = italic_N / 2.

In Fig. 4 we show the AFMC free energy gap ΔFsubscriptΔ𝐹\Delta_{F}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT (in units of the Femi gas energy εFsubscript𝜀𝐹\varepsilon_{F}italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT) as a function of temperature T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT for N=40,66𝑁4066N=40,66italic_N = 40 , 66 and 114114114114 particles. The statistical errors on the free energy gap are small and this quantity can thus be determined more accurately than the energy-staggering pairing gap. We see that the free energy gap is suppressed above the critical temperature Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT. To see more clearly the behavior of ΔFsubscriptΔ𝐹\Delta_{F}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT above Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT, we plot in the inset the free energy gap on a log-linear scale. We observe that as we decrease the temperature from above Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT, the free energy gap starts to increase below the pairing temperature scale Tsuperscript𝑇T^{*}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT that we identified in the energy-staggering pairing gap and the spin susceptibility in Fig. 3. Extrapolation to the thermodynamic limit will be important in future works to determine whether the pseudogap temperature scale T0.2TFsuperscript𝑇0.2subscript𝑇𝐹T^{*}\approx 0.2\,T_{F}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT is robust or will go down, reducing further the extent of the UFG pseudogap regime.

The free energy gap ΔFsubscriptΔ𝐹\Delta_{F}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT can be used to obtain a more precise value for the T=0𝑇0T=0italic_T = 0 energy-staggering pairing gap as follows. We write ΔFsubscriptΔ𝐹\Delta_{F}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT as ΔF=ΔETΔSsubscriptΔ𝐹subscriptΔ𝐸𝑇subscriptΔ𝑆\Delta_{F}=\Delta_{E}-T\Delta_{S}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT = roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT - italic_T roman_Δ start_POSTSUBSCRIPT italic_S end_POSTSUBSCRIPT, where ΔSsubscriptΔ𝑆\Delta_{S}roman_Δ start_POSTSUBSCRIPT italic_S end_POSTSUBSCRIPT is the entropy gap. Both ΔEsubscriptΔ𝐸\Delta_{E}roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT and ΔSsubscriptΔ𝑆\Delta_{S}roman_Δ start_POSTSUBSCRIPT italic_S end_POSTSUBSCRIPT saturate at low temperatures leading to a linear dependence of ΔFsubscriptΔ𝐹\Delta_{F}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT on T𝑇Titalic_T at low temperatures. In Fig. 4 we show linear extrapolations of the free energy gap to T=0𝑇0T=0italic_T = 0 for N=40𝑁40N=40italic_N = 40 (blue dashed line) and N=66𝑁66N=66italic_N = 66 (pink dashed line). We then estimate the T=0𝑇0T=0italic_T = 0 energy-staggering pairing gap ΔE/εFsubscriptΔ𝐸subscript𝜀𝐹\Delta_{E}/\varepsilon_{F}roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT / italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT by taking the average of the N=40𝑁40N=40italic_N = 40 (solid blue up triangle) and N=66𝑁66N=66italic_N = 66 (solid red up triangle) extrapolations to find ΔE=0.576(24)εFsubscriptΔ𝐸0.57624subscript𝜀𝐹\Delta_{E}=0.576(24)\,\varepsilon_{F}roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT = 0.576 ( 24 ) italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. We compare with the T=0𝑇0T=0italic_T = 0 QMC results of Ref. Carlson and Reddy (2005) (open orange circle), Ref. Carlson and Reddy (2008) (open blue square), Ref. Lee et al. (2011) (solid yellow up triangle), and the experimental results of Ref. Schirotzek et al. (2008) (open down green triangle) and Ref. Hoinka et al. (2017) (open up black triangle). For reference, in Table 2 we list theoretical and experimental results for the T=0𝑇0T=0italic_T = 0 pairing gap of the UFG.

Table 2: Estimates of the T=0𝑇0T=0italic_T = 0 pairing gap
Method ξ𝜉\xiitalic_ξ error
MIT low-T experiment Schirotzek et al. (2008) 0.44 0.03
Swinburne low-T experiment Hoinka et al. (2017) 0.47 0.03
Ground-state fixed-node Monte Carlo Chang et al. (2004) 0.594 0.024
Ground-state AFMC Carlson et al. (2003) 0.55 0.05
Ground-state AFMC Carlson and Reddy (2005) 0.50 0.03
QMC + MIT experiment Carlson and Reddy (2008) 0.45 0.05
Luttinger-Ward Haussmann et al. (2007) 0.46
Functional renormalization group Bartosch et al. (2009) 0.61
Functional renormalization group Floerchinger et al. (2010) 0.46
Lattice quantum Monte Carlo Lee et al. (2011) 0.52 0.01
AFMC (this work) 0.576 0.024

V Conclusion and outlook

There has been substantial progress in recent years towards understanding the BCS-BEC crossover and the UFG in the two-species Fermi gas with short-range interactions . Experimental advances have provided an understanding of several observables for this strongly correlated system DeMarco and Jin (1999); Modugno et al. (2002); Truscott et al. (2001); Schreck et al. (2001); Granade et al. (2002); Hadzibabic et al. (2002); Jochim et al. (2003); Zwierlein et al. (2005); Navon et al. (2010); Ku et al. (2012); Carcy et al. (2019); Mukherjee et al. (2019). However, from a theoretical perspective, there is no controlled analytic approach to the UFG. The current theoretical approaches that are controllable (and scale to study large system sizes) are quantum Monte Carlo methods.

In this work, we calculated the first lattice continuum limit of the condensate fraction at finite temperature for the homogeneous UFG. The continuum limit extrapolations require calculations for large lattices and they were made possible through the development of canonical-ensemble AFMC optimizations in Refs. Gilbreth et al. (2021); Gilbreth and Alhassid (2015). Our low-temperature results for the condensate fraction show remarkable agreement with the T=0𝑇0T=0italic_T = 0 QMC result of Ref. Astrakharchik et al. (2005). Using our finite-temperature results for the condensate fraction for multiple particle numbers N𝑁Nitalic_N, we used finite-size scaling to determine the superfluid critical temperature. Our result Tc=0.16(1)TFsubscript𝑇𝑐0.161subscript𝑇𝐹T_{c}=0.16(1)\;T_{F}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT = 0.16 ( 1 ) italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT is in agreement with experiment Ku et al. (2012) and with the QMC results of Refs. Burovski et al. (2006a); Goulko and Wingate (2010).

To address open questions regarding the existence of a pseudogap regime and its extent in the UFG, we calculated a model-independent energy-staggering pairing gap and the spin susceptibility in the continuum limit. The calculation of the energy-staggering pairing gap was made possible by the use of a canonical-ensemble AFMC formulation of fixed particle numbers. The pairing gap and spin susceptibility can be used to identify signatures of a pseudogap regime above Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT and below the pairing temperature scale Tsuperscript𝑇T^{*}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT. From the largest particle number N=114𝑁114N=114italic_N = 114 simulations in the continuum limit we find an upper bound of T0.2TFsuperscript𝑇0.2subscript𝑇𝐹T^{*}\approx 0.2\,T_{F}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. Thus we conclude that the pseudogap regime in the UFG is narrow. Compared with our previous finite filling factor study in Ref. Jensen et al. (2020a), we observe that in the continuum limit the pairing signatures increased for fixed temperature and particle number N𝑁Nitalic_N. This is consistent with the results of Ref. Gezerlis and Carlson (2008) where it was shown at T=0𝑇0T=0italic_T = 0 that a finite effective range interaction suppresses attractive pairing correlations.

We also calculated the free energy staggering pairing gap and found it to have significantly reduced statistical errors compared with the energy-staggering pairing gap. Our results for this observable supports a pairing temperature scale of T0.2TFsuperscript𝑇0.2subscript𝑇𝐹T^{*}\approx 0.2\;T_{F}italic_T start_POSTSUPERSCRIPT ∗ end_POSTSUPERSCRIPT ≈ 0.2 italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. In addition, extrapolating the low-temperature results for the free energy gap to zero temperature, we obtained an accurate estimate for the zero-temperature pairing gap of ΔE=0.576(24)εFsubscriptΔ𝐸0.57624subscript𝜀𝐹\Delta_{E}=0.576(24)\,\varepsilon_{F}roman_Δ start_POSTSUBSCRIPT italic_E end_POSTSUBSCRIPT = 0.576 ( 24 ) italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. Such extrapolation is not possible with the energy-staggering pairing gap because of its much larger statistical errors.

While we have reached the continuum limit for the first time for several thermodynamic observables that are useful signatures of pairing correlations for particle number as large as N=114𝑁114N=114italic_N = 114, we have not quite reached the thermodynamic limit of large particle number for the pairing gap and the spin susceptibility. It would be interesting to investigate the thermodynamic limit of these observables in future works.

Acknowledgements.
This work of was supported in part by the U. S. DOE grants Nos. DE-SC0019521, DE-SC0020177, and DE-FG02-00ER41132. The calculations presented here used resources of the National Energy Research Scientific Computing Center (NERSC), a U.S. Department of Energy Office of Science User Facility operated under Contract No. DE-AC02-05CH11231. We also thank the Yale Center for Research Computing for guidance and use of the research computing infrastructure.

Appendix A Continuous time and continuum extrapolations

We demonstrate the continuous time and continuum extrapolations for the condensate fraction in Sec. A.1. Typical continuum extrapolations for the free energy gap are shown in Sec. A.2.

Refer to caption
Figure 5: Extrapolation in ΔβΔ𝛽\Delta\betaroman_Δ italic_β for N=40𝑁40N=40italic_N = 40 particles and lattice size NL3=93superscriptsubscript𝑁𝐿3superscript93N_{L}^{3}=9^{3}italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT = 9 start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT or the condensate fraction n𝑛nitalic_n at temperatures (a) T/TF0.19𝑇subscript𝑇𝐹0.19T/T_{F}\approx 0.19italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.19, (b) T/TF0.15𝑇subscript𝑇𝐹0.15T/T_{F}\approx 0.15italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.15, and (c) T/TF0.12𝑇subscript𝑇𝐹0.12T/T_{F}\approx 0.12italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.12. The solid blue circles are the calculated condensate fraction for several discrete values of ΔβΔ𝛽\Delta\betaroman_Δ italic_β, We perform a straight line fit (solid purple line) in (ΔβεF)2superscriptΔ𝛽subscript𝜀𝐹2(\Delta\beta*\varepsilon_{F})^{2}( roman_Δ italic_β ∗ italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT (where εF=(2/2m)(3π2ρ)2/3)\varepsilon_{F}=(\hbar^{2}/2m)(3\pi^{2}\rho)^{2/3})italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT = ( roman_ℏ start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT / 2 italic_m ) ( 3 italic_π start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT italic_ρ ) start_POSTSUPERSCRIPT 2 / 3 end_POSTSUPERSCRIPT ) is the Fermi energy) in the range (ΔβεF)2<0.003superscriptΔ𝛽subscript𝜀𝐹20.003(\Delta\beta*\varepsilon_{F})^{2}<0.003( roman_Δ italic_β ∗ italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT < 0.003. The fitted line is shown to larger values of ΔβΔ𝛽\Delta\betaroman_Δ italic_β for reference (dashed green line).
Refer to caption
Refer to caption
Refer to caption
Figure 6: Continuum limit extrapolations ν0𝜈0\nu\rightarrow 0italic_ν → 0 (ν=NL3/N𝜈superscriptsubscript𝑁𝐿3𝑁\nu=N_{L}^{3}/Nitalic_ν = italic_N start_POSTSUBSCRIPT italic_L end_POSTSUBSCRIPT start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT / italic_N is the filling factor) of the condensate fraction for N=40𝑁40N=40italic_N = 40 particles (left column, open blue squares), N=66𝑁66N=66italic_N = 66 particles (middle column, open orange circles), and N=114𝑁114N=114italic_N = 114 particles (right column, open down green triangles) for several representative temperatures T/TF0.12𝑇subscript𝑇𝐹0.12T/T_{F}\approx 0.12italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.12 (top row), and T/TF0.15𝑇subscript𝑇𝐹0.15T/T_{F}\approx 0.15italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.15 (middle row). In the bottom row we show the extrapolations for T/TF0.19𝑇subscript𝑇𝐹0.19T/T_{F}\approx 0.19italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.19 (N=40𝑁40N=40italic_N = 40) and T/TF0.18𝑇subscript𝑇𝐹0.18T/T_{F}\approx 0.18italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.18 (N=66,114𝑁66114N=66,114italic_N = 66 , 114). We perform a straight line fit (solid line) in ν1/3superscript𝜈13\nu^{1/3}italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT in the range ν1/3<0.5superscript𝜈130.5\nu^{1/3}<0.5italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT < 0.5 for each temperature T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT and particle number N𝑁Nitalic_N. The fitted line is shown to larger filling factors ν1/3superscript𝜈13\nu^{1/3}italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT for reference (dashed line).
Refer to caption
Refer to caption
Refer to caption
Figure 7: Continuum limit extrapolations in ν1/3superscript𝜈13\nu^{1/3}italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT for the free energy-staggering pairing gap ΔFsubscriptΔ𝐹\Delta_{F}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT for N=40,66𝑁4066N=40,66italic_N = 40 , 66 and 114114114114 particles and for a few representative temperatures as in Fig. 6. We perform straight line fits (solid lines) in the range ν1/3<0.5superscript𝜈130.5\nu^{1/3}<0.5italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT < 0.5.

A.1 Condensate fraction

Extrapolations to continuous time.— The finite imaginary time slice ΔβΔ𝛽\Delta\betaroman_Δ italic_β introduces a systematic error in the AFMC calculations. As discussed in Sec. III.1, using a symmetric Trotter decomposition and a three-point Gaussian quadrature leads to a discretization error of O(Δβ2)𝑂Δsuperscript𝛽2O(\Delta\beta^{2})italic_O ( roman_Δ italic_β start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT ). We extrapolate Δβ0Δ𝛽0\Delta\beta\rightarrow 0roman_Δ italic_β → 0 by a linear fit in (εFΔβ)2superscriptsubscript𝜀𝐹Δ𝛽2(\varepsilon_{F}\Delta\beta)^{2}( italic_ε start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT roman_Δ italic_β ) start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT for sufficiently small values of ΔβΔ𝛽\Delta\betaroman_Δ italic_β.

In Fig. 5 we show representative extrapolations of the condensate fraction for N=40𝑁40N=40italic_N = 40 particles and lattice size of 93superscript939^{3}9 start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT at three temperatures (a) T/TF0.19𝑇subscript𝑇𝐹0.19T/T_{F}\approx 0.19italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.19, (b) T/TF0.15𝑇subscript𝑇𝐹0.15T/T_{F}\approx 0.15italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.15, and (c) T/TF0.12𝑇subscript𝑇𝐹0.12T/T_{F}\approx 0.12italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.12. The slope of the fit increases dramatically in magnitude as temperature is reduced, so the extrapolations become necessary at low temperatures (see for T/TF0.12𝑇subscript𝑇𝐹0.12T/T_{F}\approx 0.12italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.12 and T/TF0.15𝑇subscript𝑇𝐹0.15T/T_{F}\approx 0.15italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.15). We performed these ΔβΔ𝛽\Delta\betaroman_Δ italic_β extrapolations for all particle numbers and lattice sizes.

Continuum extrapolations.— After carrying out the ΔβΔ𝛽\Delta\betaroman_Δ italic_β extrapolations, we perform the continuum limit extrapolation using lattices of increasing size with ν0𝜈0\nu\to 0italic_ν → 0 for each temperature T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT and particle number N𝑁Nitalic_N.

In Fig. 6 we show the condensate fraction n𝑛nitalic_n as a function of ν1/3superscript𝜈13\nu^{1/3}italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT for particle numbers N=40,66𝑁4066N=40,66italic_N = 40 , 66 and 114114114114 for three temperatures in the critical regime near Tcsubscript𝑇𝑐T_{c}italic_T start_POSTSUBSCRIPT italic_c end_POSTSUBSCRIPT. We perform linear fits in ν1/3superscript𝜈13\nu^{1/3}italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT using the range ν1/30.50.6less-than-or-similar-tosuperscript𝜈130.50.6\nu^{1/3}\lesssim 0.5-0.6italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT ≲ 0.5 - 0.6. We use this range in the continuum extrapolations of the condensate fraction n𝑛nitalic_n for all temperatures and particle numbers. We note that the behavior of n𝑛nitalic_n for larger filling factors is significantly different for lower temperatures. At the lower temperature T/TF0.12𝑇subscript𝑇𝐹0.12T/T_{F}\approx 0.12italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT ≈ 0.12 (upper panels of Fig. 6), the largest filling factor results are below the continued straight line fit (dashed line) whereas for higher temperatures (middle and lower panels), the larger filling factor results are above the continued fit. The continuum extrapolation provides the final estimate for the condensate fraction n𝑛nitalic_n for a given particle number N𝑁Nitalic_N and temperature T/TF𝑇subscript𝑇𝐹T/T_{F}italic_T / italic_T start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT.

A.2 Free energy gap

In this section, we demonstrate the continuum limit (ν0𝜈0\nu\rightarrow 0italic_ν → 0) extrapolations for the free energy gap ΔFsubscriptΔ𝐹\Delta_{F}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT. These continuum extrapolations are carried out after the continuous time extrapolations Δβ0Δ𝛽0\Delta\beta\rightarrow 0roman_Δ italic_β → 0.

In Fig. 7 we show ΔFsubscriptΔ𝐹\Delta_{F}roman_Δ start_POSTSUBSCRIPT italic_F end_POSTSUBSCRIPT vs. ν1/3superscript𝜈13\nu^{1/3}italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT for several values of particle number and temperatures (as in Fig. 6). The straight lines are fits in the range ν1/3<0.5superscript𝜈130.5\nu^{1/3}<0.5italic_ν start_POSTSUPERSCRIPT 1 / 3 end_POSTSUPERSCRIPT < 0.5.

References