For a two-component ultracold Fermi gas with unequal spin populations confined in a harmonic trap, mean-field theory in the local-density approximation (Bighin, Mazzarella, Dell'Anna and Salasnich, 2014) predicts the pair-condensate fraction and the superfluid-core/normal-shell phase boundary across the BCS–BEC crossover, but treats the polarized normal shell as a non-interacting Fermi gas, omits finite-temperature Gaussian pair fluctuations, and neglects possible FFLO (Fulde–Ferrell–Larkin–Ovchinnikov) phases; it overestimates the superfluid boundary (critical gap Δ0,c≈1.4ζ\Delta_{0,c}\approx 1.4\zeta versus the ≈ζ\approx\zeta suggested by quantum Monte Carlo and experiment) and cannot reproduce the loss-affected condensate fractions measured on the BEC side. The open problem is a quantitatively reliable theory of pair condensation in the trapped imbalanced gas — condensate fraction and profiles plus phase boundary as functions of interaction strength and polarization — that includes interactions in the normal shell and confronts the 2006 MIT measurements on the BCS/unitarity side, where the data are not loss-dominated, while assessing the roles of the further limitations that the source paper itself argued are subleading: finite-temperature Gaussian pair fluctuations, a possible FFLO contribution in three dimensions, and the loss-affected BEC side. The balanced-gas counterparts are solved, but the finite-polarization condensate fraction beyond mean field remains untouched; the problem remains open.

Background

When a two-component Fermi gas of atoms in two spin states is brought to a strong magnetic field near a Feshbach resonance (a tunable scattering resonance that lets the interaction strength be varied continuously), the s-wave interaction between unlike spins can be tuned from weak attraction to strongly bound molecules. This connects the Bardeen–Cooper–Schrieffer (BCS) regime of overlapping Cooper pairs to the Bose–Einstein-condensate (BEC) regime of tightly bound dimers — the BCS–BEC crossover; the middle of the crossover, where the scattering length diverges, is called unitarity. If the two spin populations are unequal (a spin-polarized or imbalanced gas), matching Fermi surfaces is impossible, and superfluidity survives only up to a critical polarization; in a harmonic trap the cloud separates into an inner superfluid core with paired atoms and an outer normal shell of the majority atom, a shell structure treated within the local-density approximation (LDA), which replaces the trap potential by a spatially varying local chemical potential. Bighin, Mazzarella, Dell'Anna and Salasnich (2014) computed, within extended BCS mean-field theory plus LDA, the number of condensed Cooper pairs via off-diagonal long-range order (ODLRO, the criterion for a macroscopically occupied pair state measured through the long-distance limit of the two-particle density matrix), obtaining the condensate fraction and density profiles versus interaction strength, polarization, and trap shape, and the position of the superfluid–normal interface. They found their mean-field scheme overestimates the superfluid boundary: pairing survives until the local gap reaches Δ0,c≈1.4ζ\Delta_{0,c}\approx 1.4\zeta, where Δ\Delta is the pairing gap, ζ=(μ↑−μ↓)/2\zeta=(\mu_\uparrow-\mu_\downarrow)/2 is the imbalance chemical potential, whereas fixed-node quantum Monte Carlo (QMC, Carlson and Reddy 2005) and the MIT experiment (Zwierlein, Schirotzek, Schunck and Ketterle 2006) indicate the superfluid core disappears already around Δ0,c≈ζ\Delta_{0,c}\approx\zeta. The paper's own discussion attributes this to modeling the normal shell as non-interacting: an interacting normal shell (equation of state computed by Lobo, Recati, Giorgini and Stringari 2006) would lower the normal free energy and shrink the superfluid region. Three further acknowledged gaps remain. First, finite-temperature Gaussian pair fluctuations — the leading beyond-mean-field correction obtained by expanding the thermodynamic potential to quadratic order in pairing-field fluctuations, which are known to be essential for the critical temperature of the balanced gas (Fukushima, Ohashi, Taylor and Griffin 2007 computed the Gaussian-fluctuation condensate fraction for the balanced gas; Mulkerin, Liu and Hu 2016 and Mulkerin, Yao, Ohashi, Liu and Hu 2026 pushed thermodynamics beyond Gaussian order; He, Li, Lu and Lee 2020 benchmarked the balanced unitary condensate fraction with auxiliary-field lattice QMC) — have never been evaluated for the imbalanced gas's condensate fraction, although the source paper judged their zero-temperature effect on the condensate fraction undetectable given the experimental sensitivities. Second, exotic inhomogeneous superfluid phases, the Fulde–Ferrell–Larkin–Ovchinnikov (FFLO) states pairing atoms at different Fermi momenta into a condensate with spatially oscillating order parameter, remain unobserved in three-dimensional ultracold gases and of unsettled stability (Mizushima, Ichioka and Machida 2007 found them only in a narrow BCS-side region in a full nonlocal Bogoliubov–de Gennes treatment of the trap; Pirolo, Pisani and Pieri 2026 revisit the FFLO transition and its quantum criticality). Third, on the BEC side the experimental condensate fractions of Zwierlein et al. (2006) drop rapidly with polarization because of inelastic molecule losses, so theory–experiment comparison there requires loss modeling; a decade later the trapped phase diagram of Olsen, Revelle, Fry, Sheehy and Hulet (2015) still disagreed with QMC critical polarizations on the BEC side while agreeing on the BCS/unitarity side. These elements frame the single comparison that defines the problem: a quantitatively faithful account of the measured pair condensation of the trapped polarized gas across the crossover, whose source-identified missing ingredient is the interacting normal shell.

Problem Statement

Construct a quantitatively reliable theory of pair condensation in a harmonically trapped two-component Fermi gas with unequal spin populations, across the BCS–BEC crossover, that improves on the extended-BCS mean-field plus local-density-approximation (LDA) treatment of Bighin, Mazzarella, Dell'Anna and Salasnich (2014) by incorporating: (a) fermion–fermion interactions in the polarized normal shell (e.g. via the QMC equation of state of Lobo et al. or equivalent), so that the superfluid–normal phase boundary and condensate density profiles are no longer overestimated (the mean-field critical gap Δ0,c≈1.4ζ\Delta_{0,c}\approx 1.4\zeta should move toward the ≈ζ\approx\zeta benchmark); and (b) a quantitative confrontation with the measured condensate fractions of Zwierlein et al. (2006) on the BCS/unitarity side, where the source paper itself notes the data are not dominated by inelastic losses. The deliverable is the condensate fraction of pairs and the superfluid–normal boundary as functions of interaction strength (e.g. 1/kFa1/k_Fa, where kFk_F is the Fermi momentum and aa the s-wave scattering length) and spin polarization, together with profiles through the trap, demonstrated to be consistent with the zero-temperature QMC benchmarks — the Carlson–Reddy critical gap Δ0,c≈ζ\Delta_{0,c}\approx\zeta and the Pilati–Giorgini phase separation — and with the BCS/unitarity-side experimental condensate fractions.

The source paper acknowledges three further limitations that it argued are subleading for this comparison; these remain open extensions rather than mandatory completion conditions: Gaussian pair fluctuations, whose effect on the condensate fraction it judged undetectable at zero temperature given the experimental sensitivities but which matter at finite temperature, where its finite-temperature condensate-number formula already applies; the possible FFLO contribution in three dimensions, omitted because it occupies a very small parameter region, has not been observed in 3D, and its stability is unsettled; and the loss-affected BEC side of the experimental data, excluded from the source's comparison because of inelastic losses. A corrected theory that quantifies these contributions — or substantiates the source's negligibility arguments for them — addresses the same open question.

References

  1. G. Bighin, G. Mazzarella, L. Dell'Anna, L. Salasnich (2014). Pair condensation of polarized fermions in the BCS–BEC crossover. Journal of Physics B: Atomic, Molecular and Optical Physics 47, 195302. DOI: 10.1088/0953-4075/47/19/195302. arXiv:1404.3178, https://arxiv.org/abs/1404.3178.
  2. C. Lobo, A. Recati, S. Giorgini, S. Stringari (2006). Normal State of a Polarized Fermi Gas at Unitarity. Physical Review Letters 97, 200403. DOI: 10.1103/PhysRevLett.97.200403. https://doi.org/10.1103/PhysRevLett.97.200403
  3. J. Carlson, S. Reddy (2005). Asymmetric Two-Component Fermion Systems in Strong Coupling. Physical Review Letters 95, 060401. DOI: 10.1103/PhysRevLett.95.060401. https://doi.org/10.1103/PhysRevLett.95.060401
  4. M. W. Zwierlein, A. Schirotzek, C. H. Schunck, W. Ketterle (2006). Fermionic Superfluidity with Imbalanced Spin Populations. Science 311, 492–496. DOI: 10.1126/science.1122318. https://doi.org/10.1126/science.1122318
  5. S. Pilati, S. Giorgini (2008). Phase Separation in a Polarized Fermi Gas at Zero Temperature. Physical Review Letters 100, 030401. DOI: 10.1103/PhysRevLett.100.030401. https://doi.org/10.1103/PhysRevLett.100.030401
  6. T. Mizushima, M. Ichioka, K. Machida (2007). Imbalanced Superfluid Phase of a Trapped Fermi Gas in the BCS–BEC Crossover Regime. Journal of the Physical Society of Japan 76, 104006. DOI: 10.1143/JPSJ.76.104006. https://doi.org/10.1143/JPSJ.76.104006
  7. N. Fukushima, Y. Ohashi, E. Taylor, A. Griffin (2007). Superfluid density and condensate fraction in the BCS-BEC crossover regime at finite temperatures. Physical Review A 75, 033609. DOI: 10.1103/PhysRevA.75.033609. https://doi.org/10.1103/PhysRevA.75.033609
  8. B. C. Mulkerin, X.-J. Liu, H. Hu (2016). Beyond Gaussian pair fluctuation theory for strongly interacting Fermi gases. Physical Review A 94, 013610. DOI: 10.1103/PhysRevA.94.013610. https://doi.org/10.1103/PhysRevA.94.013610
  9. B. C. Mulkerin, X. Yao, Y. Ohashi, X.-J. Liu, H. Hu (2026). Beyond Gaussian pair fluctuation theory for strongly interacting Fermi gases II: the broken-symmetry phase. AAPPS Bulletin 36, 15. DOI: 10.1007/s43673-026-00195-9. https://doi.org/10.1007/s43673-026-00195-9
  10. R. He, N. Li, B.-N. Lu, D. Lee (2020). Superfluid condensate fraction and pairing wave function of the unitary Fermi gas. Physical Review A 101, 063615. DOI: 10.1103/PhysRevA.101.063615. https://doi.org/10.1103/PhysRevA.101.063615
  11. F. Matera, M. F. Wagner (2017). Low-energy modes of spin-imbalanced Fermi gases in BCS phase. The European Physical Journal D 71, 293. DOI: 10.1140/epjd/e2017-80156-0. https://doi.org/10.1140/epjd/e2017-80156-0
  12. S. Aziz, S. Jamil, A. Rahman, I. Ashraf (2020). Comparative study of the condensate fraction of balanced and imbalanced uniform fermionic systems in the BEC-BCS crossover. Journal of the Optical Society of America B 37, 533. DOI: 10.1364/JOSAB.377423. https://doi.org/10.1364/JOSAB.377423
  13. B. A. Olsen, M. C. Revelle, J. A. Fry, D. E. Sheehy, R. G. Hulet (2015). Phase diagram of a strongly interacting spin-imbalanced Fermi gas. Physical Review A 92, 063616. DOI: 10.1103/PhysRevA.92.063616. https://doi.org/10.1103/PhysRevA.92.063616
  14. F. Pirolo, L. Pisani, P. Pieri (2026). FFLO transition and quantum criticality in polarized Fermi gases. arXiv:2607.28335. https://arxiv.org/abs/2607.28335