Abstract
We consider the two-dimensional Fermi gas at finite temperature with attractive short-range interactions. Using the virial expansion, which provides a controlled approach at high temperatures, we determine the spectral function and contact for the normal state. Our calculated spectra are in qualitative agreement with recent photoemission measurements [M. Feld et al., Nature 480, 75 (2011)], thus suggesting that the observed pairing gap is a feature of the high-temperature gas rather than being evidence of a pseudogap regime just above the superfluid transition temperature. We further argue that the strong pair correlations result from the fact that the crossover to bosonic dimers occurs at weaker interactions than previously assumed.
Introduction: The Pseudogap Debate in 2D
Understanding the mechanism of particle pairing is fundamental to condensed matter physics, particularly in two-dimensional (2D) systems where standard mean-field theories often fail. A central question in the study of ultra-cold 2D Fermi gases is the nature of the "pseudogap"—a gap in the energy spectrum that persists above the superfluid transition temperature ($T_c$). Recent experiments observed such a gap, sparking debate: is this a true many-body pseudogap, or simply the signature of two-body dimer (molecule) formation at high temperatures?
Methodology: The Virial Expansion
To address this, we employ the virial expansion, a controlled theoretical framework valid at temperatures $T \gtrsim \epsilon_F$ (the Fermi temperature). This approach allows for a systematic calculation of thermodynamic and spectral properties by expanding in powers of the fugacity.
- Diagrammatic Expansion: We calculated the second and third virial coefficients ($b_2, b_3$) using Feynman diagrams to capture interactions between pairs and triplets.
- T-matrix Approach: A non-perturbative T-matrix calculation was used to derive the spectral function, explicitly incorporating two-body correlations.
- Contact: We determined the "contact," a universal parameter quantifying short-range correlations.
Spectral Signatures and the Pairing Gap
Our results show that the "pairing gap" observed in experiments is a robust feature of the high-temperature thermal gas. The calculated spectral functions $A(k, \omega)$ exhibit two distinct branches: one corresponding to bound dimers and another to unpaired atoms.
As shown in Figure 4, the spectra display a characteristic "back-bending" dispersion, often misinterpreted as a many-body pseudogap. Our analysis suggests this is instead a consequence of short-range two-body physics. The persistence of this gap at high temperatures ($T \approx 5\epsilon_F$) strongly supports the dimer interpretation over a precursor to superfluidity.
Universal Contact and the Crossover Shift
A key finding is that the BCS-to-BEC crossover in 2D occurs at significantly weaker interactions than previously thought. While the community often identifies the crossover at $\ln(k_F a_{2D}) = 0$, our calculations (supported by Quantum Monte Carlo data) show the chemical potential $\mu$ vanishes at $\ln(k_F a_{2D}) \approx 0.5$. This implies that many experimental regimes are already deep in the molecular (BEC) side.
Figure 3 illustrates the temperature dependence of the contact. Notably, in the BCS regime, the contact increases with temperature initially. This non-monotonic behavior arises because thermal energy overcomes Pauli blocking, allowing fermions to approach each other more closely and form pairs.
Conclusion and Outlook
The virial expansion provides a clear picture of the 2D Fermi gas: the observed pairing gap is primarily a high-temperature effect driven by dimer formation. While this model is highly accurate at $T \gtrsim \epsilon_F$, it is an approximation that breaks down near $T_c$. Future work should focus on the very weak interaction limit ($\ln(k_F a_{2D}) \gg 1$) to search for genuine many-body pseudogap signatures and explore complex three-body physics in mass-imbalanced systems.
Summary updated