Erratum: Superfluid density of states and pseudogap phenomenon in the BCS-BEC crossover regime of a superfluid Fermi gas [Phys. Rev. A82, 043630 (2010)]

Abstract

In our paper, the effects of the superfluid order parameter were underestimated in the crossover region due to a computational mistake in the treatment of the off-diagonal self-energy. The corrected calculation gives a larger superfluid gap E SF , so that Fig. 4(a) should be replaced by the corrected one shown below.In the weak-coupling BCS regime and the strong-coupling BEC regime, the density of states ρ(ω), as well as the spectral weight A p (ω), are almost unaltered by this correction, because pairing fluctuations are less important there. On the other hand, Fig. 6, which shows the detailed competition between the pseudogap and superfluid gap in the unitarity regime, should be replaced by the corrected one presented here. In our paper, we reported that the superfluid gap started to open below 0.9T c , after the pseudogap almost disappears. In the correct result, however, since effects of the superfluid order are more remarkable, the superfluid gap starts to develop below T c , before the pseudogap disappears completely, as shown in the corrected Fig. 6. As expected, far below T c , the ordinary BCS-type superfluid density of states is obtained, as shown in the corrected Fig. 6(a4). FIG. 4. (Color online) (a) Comparison of the pseudogap size E PG at T c and superfluid gap size E SF at T = 0 evaluated from ρ(ω). For comparison, we also show the energy gap E G in the BCS-Leggett crossover theory [1] (which equals when μ > 0, and equals μ 2 + 2 when μ < 0). * rwatanab@rk.phys.keio.ac.jp FIG. 6. (Color online) Temperature dependence of DOS ρ(ω) and intensity of spectral weight A( p,ω) in the crossover regime [(k F a s ) −1 = 0 (unitarity limit)].In our paper, we have introduced the pseudogap temperature T * at which the superfluid gap structure starts to appear in ρ(ω). However, since the corrected Fig. 6 indicates that the superfluid gap starts to develop without the disappearance of the pseudogap, the previous simple definition for T * is insufficient to discuss effects of pairing fluctuations below T c . Thus, to definitely determine T * , in this erratum we redefine this pseudogap temperature as the temperature at which the superfluid density of states at ω = 0 [ρ(ω = 0)] decreases by 50% compared with the value at T c [ρ c0 ≡ ρ(ω = 0,T = T c )]. Below this temperature, the superfluid density of states ρ(ω) around ω = 0 is remarkably suppressed, which means the suppression of pairing fluctuations due to the development of superfluid order. The resulting pseudogap regime in Fig. 8(b) is almost unaltered (although the definition of T * is different). We briefly note that T * is a crossover temperature without being accompanied by any phase transition, so that it depends on the choice of the value of R ≡ ρ(ω = 0)/ρ c0 . For the readers' convenience, we show T * for various values of R in Fig. 8(c) as supplementary information. 039908-1 1050-2947/2012/85(3)/039908 (2)

Pseudogap temperature and effects of a harmonic trap in the BCS-BEC crossover regime of an ultracold Fermi gas

“…However, because of the complexity of this system, other scenarios have been also proposed, such as antiferromagnetic spin fluctuations [24,25], and a hidden order [26]. In contrast, in cold Fermi gases, because of the simplicity of the system, the origin can be uniquely identified as strong pairing fluctuations [27][28][29][30][31][32][33][34][35][36][37]. Thus, this system would be very useful for the study of the preformed-pair scenario.…”

Heteropairing and component-dependent pseudogap phenomena in an ultracold Fermi gas with different species with different masses

“…In our recent paper [42], we showed that the ordinary (non-self-consistent) T -matrix approximation (TMA) [70], which has been extensively used to successfully clarify various interesting BCS-BEC crossover physics in the mass-balanced case [53,54,[71][72][73][74][75][76], breaks down in the presence of mass imbalance, to unphysically give double-valued T c in the crossover region, as shown in Fig. 1.…”

Spin Susceptibility and Effects of Inhomogeneous Strong Pairing Fluctuations in a Trapped Ultracold Fermi Gas

“…We briefly note that the ordinary T -matrix approximation 22,25,27 is immediately reproduced by simply replacing the ETMA Green's function G −σ in Eq. (4) with the bare one,…”

“…In a uniform Fermi gas, it has been shown that ETMA correctly describes the BCS-BEC crossover behavior of the spin susceptibility 18,21 , which makes us expect that this strong-coupling theory is also valid for the trapped case. We briefly note that the ordinary T -matrix approximation 22,25,27 , as well as the strong-coupling theory developed by Nozières and Schmitt-Rink 6,7 , are known to unphysically give negative spin susceptibility in the BCS-BEC crossover region, although these theories have successfully explained various many-body phenomena in the BCS-BEC crossover region. Using the combined ETMA with LDA, we calculate the local spin susceptibility χ(T, r) in the normal state near T c .…”

“…Theoretically, the possibility of the so-called spin-gap phenomenon has been discussed in the crossover region near the superfluid phase transition temperature T c 17,18,19,20,21 , where the spin susceptibility is anomalously suppressed by preformed spin-singlet Cooper pairs. Since preformed Cooper pairs also cause the pseudogap phenomenon 22,23,24,25,26 (where the single-particle density of states exhibits a gap-like structure even in the normal state), the spin-gap phenomenon and pseudogap phenomenon are deeply related to each other in the cold Fermi gas system. So far, the spin susceptibility has theoretically been discussed in a uniform Fermi gas 18,19,20,21 , although a real ultracold Fermi gas is always prepared in a trap potential.…”