Session F18: Invited Session: Frontiers in Fermi Gasses

Observation of Antiferromagnetic Correlations in the Hubbard Model with Ultracold Atoms

Ultracold atoms on optical lattices form a versatile platform for studying many-body physics, with the potential of addressing some of the most important issues in strongly correlated matter. Progress, however, has been stymied by an inability to create sufficiently low temperatures in an optical lattice. In this talk, I will present our experimental results on the characterization of the three-dimensional Hubbard model near half-filling, realized using two spin-states of fermionic atomic lithium ($^6$Li). We have developed a compensated optical lattice that has enabled, for the first time, the achievement of temperatures that are below the tunneling energy, $t$. We use \emph{in-situ} imaging to extract the central density of the gas, and to determine its local compressibility. For intermediate to strong interactions, we observe the emergence of a density plateau and a reduction of the compressibility, indicative of the formation of a Mott insulator. Comparisons to state-of-the-art numerical simulations of the Hubbard model over a wide range of interactions set an upper limit for the temperature $T < t$.\footnote{P.M. Duarte \emph{et al.}, arXiv:1409.8348} The Hubbard model is known to exhibit antiferromagnetism at temperatures below the N\'{e}el temperature $T_N$. We have detected antiferromagnetic correlations in this system by spin-sensitive Bragg scattering of light. We deduce the temperature of the atoms in the lattice by comparing the light scattering to determinantal quantum Monte Carlo and numerical linked-cluster expansion calculations to find that $T/t = 0.51 \pm 0.06$, corresponding to $1.4\, T_N$.\footnote{R.A. Hart \emph{et al.}, arXiv:1407.5932} Further refinement of the compensated lattice may produce even lower temperatures which, along with light scattering thermometry, have important implications for achieving other novel quantum states of matter.   

Spin Transport in a Unitary Fermi Gas

We study spin transport in a quantum degenerate Fermi gas of $^{40}$K near an s-wave interaction resonance. The starting point of our measurements is a transversely spin-polarized gas, where each atom is in a superposition of the lowest two Zeeman eigenstates. In the presence of an external gradient, a spin texture develops across the cloud, which drives diffusive spin currents. Spin transport is described with two coefficients: $D_0^\perp$, the transverse spin diffusivity, and $\gamma$, the Leggett-Rice parameter. Diffusion is a dissipative effect that increases the entropy of the gas, eventually creating a mixture of spin states. $\gamma$ parameterizes the rate at which spin current precesses around the local magnetization. Using a spin-echo sequence, we measure these transport parameters for a range of interaction strengths and temperatures. At unitarity, for a normal-state gas initially at one fifth of the Fermi temperature, we find $D_0^\perp = 2.3(4)\,\hbar/m$ and $\gamma = 1.08(9)$, where $m$ is the atomic mass. In the limit of zero temperature, $\gamma$ and $D_0^\perp$ are scale-invariant universal parameters of the unitary Fermi gas. The value of $D_0^\perp$ reveals strong scattering and is near its proposed quantum limit, such that the inferred value of the transport lifetime $\tau_\perp$ is comparable to $\hbar/\epsilon_F$. This raises the possibility that incoherent transport may play a role. The nonzero value of $\gamma$ tells us that spin waves in unitary Fermi gas are dispersive, or in other words, that the gas has a spin stiffness in the long-wavelength limit. Time permitting, we will also discuss a time-resolved measurement of the contact, through which we observe the microscopic transformation of the gas from ideal to strongly correlated.   

Fulde-Ferrell Superfluids in Degenerate Fermi Gases with Synthetic Gauge Fields

Fulde-Ferrell-Larkin-Ovchinnikov (FFLO) phase was first predicted in two-dimensional superconductors about 50 years ago, but so far unambiguous experimental evidence is still lacking. The recently experimentally realized spin-imbalanced Fermi gases may potentially unveil this elusive state, but they require very stringent experimental conditions. In this talk, I will discuss a new route for generating stable FF phases through single particle band dispersion engineering using the recently experimentally generated synthetic gauge fields, such as artificial spin-orbit coupling or band hybridization in driven optical lattices. I will show that such FF superfluids can support topological quantum excitations such as Majorana and Weyl fermions.\\[4pt] [1] Z. Zheng, M. Gong, X. Zhou, C. Zhang, and G.-C. Guo, Phys. Rev. A 87, 031602 (2013).\\[0pt] [2] C. Qu, Z. Zheng, M. Gong, Y. Xu, L. Mao, X. Zou, G.-C. Guo, C. Zhang, Nature Communications 4, 2710 (2013).\\[0pt] [3] Y. Xu, C. Qu, M. Gong, C. Zhang, Phys. Rev. A 89, 013607 (2014).\\[0pt] [4] Y. Xu, R. Chu, C. Zhang, Phys. Rev. Lett. 112, 136402 (2014) Editors' Suggestions\\[0pt] [5] C. Qu, M. Gong, C. Zhang, Phys. Rev. A 89, 053618 (2014).\\[0pt] [6] Y. Xu, C. Zhang, arXiv:1407.3483\\[0pt] [7] Z. Zheng, C. Qu, X. Zou, C. Zhang, arXiv:1408.5824\\[0pt] [8] Z. Zheng, C. Qu, X. Zou, C. Zhang, Spin imbalanced Fulde-Ferrell superfluids in 3D driven fermionic optical lattices, in preparation.   

Pairing phenomena in quasi-2D Fermi gases

Quasi-two-dimensional Fermi systems are both of fundamental interest and technological importance. Recent advances in cold-atom experiments have now made it possible to investigate model quasi-2D Fermi gases in a controlled manner. In this talk, I will discuss the different pairing regimes in the attractive Fermi gas and how these can be dramatically modified by the finite transverse width of the quasi-2D system. In particular, I find that the critical temperature for pairing and superfluidity can be enhanced by relaxing the transverse confinement and perturbing away from the 2D limit. I will also discuss the exotic phases that may be generated when the spin populations are imbalanced.   

Strongly Interacting Fermi Gases of Atoms and Molecules

In recent years, ultracold gases of fermionic atoms have become a new platform for the realization of paradigmatic forms of strongly interacting matter. Feshbach scattering resonances allow to tune the interactions between atoms at will and to realize the crossover from Bose-Einstein condensation of molecules to Bardeen-Cooper-Schrieffer superfluidity of long-range Cooper pairs. On resonance, we encounter the unitary Fermi gas, with universal properties that closely correspond to those of dilute neutron matter in the crust of neutron stars, and to nuclear matter. I will present our recent study of solitonic excitations in this novel superfluid, the creation of planar solitons and the subsequent cascade into vortex rings and solitonic vortices. In the presence of spin imbalance, solitons are predicted stabilize, a hallmark of the Larkin-Ovchinnikov phase. To induce strong interactions one may also quench the atoms' kinetic energy in optical lattices. Of great interest here is the realization of the Fermi-Hubbard model, believed to hold the key to understanding high-temperature superconductors. We recently realized imaging of fermionic atoms with single-site resolution in optical lattices, an important step towards the direct observation of magnetic order. Finally, strong, long-range dipolar interactions can lead to novel states of fermionic matter such as topological superfluids. We have created chemically stable, strongly dipolar fermionic molecules, opening up prospects for observing a strongly interacting degenerate Fermi gas with dominant dipolar interactions.