The high degree of control on ultracold gases allows us to precisely manipulate their internal state. When the gas is made of atoms in two different internal states, it can be considered as a two-component spin mixture. Below a critical temperature, the gas becomes a superfluid mixture, never realized before with any other platform, and therefore interesting to study per se, but it also constitutes a promising and versatile platform for applications in spintronic devices or to study phenomena belonging to very different fields, such as magnetism, high-energy physics or gravitation. Here, I will revisit ground-state properties and excitations of a binary bosonic superfluid, and then introduce a coherent coupling between the states and treat the global state of the atoms as a spin in the presence of a variable external field.
The high degree of control on ultracold gases allows us to precisely manipulate their internal state. When the gas is made of atoms in two different internal states, it can be considered as a two-component spin mixture. Below a critical temperature, the gas becomes a superfluid mixture, never realized before with any other platform, and therefore interesting to study per se, but it also constitutes a promising and versatile platform for applications in spintronic devices or to study phenomena belonging to very different fields, such as magnetism, high-energy physics or gravitation. Here, I will revisit ground-state properties and excitations of a binary bosonic superfluid, and then introduce a coherent coupling between the states and treat the global state of the atoms as a spin in the presence of a variable external field.
Abstract Ultracold atomic spin mixtures develop rich and intriguing magnetic properties when external radiation coherently couples different spin states. In particular, the coupled mixture may acquire a critical behavior when the spin interactions equal the coupling energy. However, atomic mixtures generally feature a relatively high sensitivity to magnetic fields that can set a limitation to the observable phenomena. In this article, we present an overview of experimental studies of magnetism based on superfluid multicomponent gases in an ultrastable magnetic field environment, which recently became available.
Just a couple decades ago bosonic low-temperature physics was synonymous with one particular liquid at a temperature of several degrees Kelvin. This liquid, helium-4, and its famous Helium-II phase, has revealed remarkable quantum properties such as quantized vortices and second sound. Yet, despite many things learned by investigating helium, its study has certain limitations. On the one hand, helium is a strongly interacting liquid, which makes first-principles theoretical research challenging. To name one example, establishing a connection between the non-interacting condensate state described by Bose and Einstein, and the helium properties observed in the lab took decades. Furthermore, helium is a specific chemical element, and hence there is no surprise that its physical properties are fixed by nature. However, every particular chemical element has particular physical properties, so for a long time it seemed that it was as good as it gets. The picture has changed drastically when the first Bose-Einstein condensate (BEC) was observed in an ultracold atomic gas. “Low temperature” has become synonymous with “ultracold”, and the latter implies temperatures on the order of nanokelvins. Most strikingly, the relationship between what is fixed by nature and what is experimentally changeable has evolved. In particular, systems of different particle statistics, spin, and interactions have been engineered using ultracold vapours of various alkali atoms. This versatile manipulation not only makes emulating various condensed-matter physics models in a controlled system without impurities possible, but also opens a path towards novel phenomena that are at present not achievable in any other manner. In this thesis, we have explored some of the novel phenomena that arise in systems of bosonic particles with internal degrees of freedom, such as (pseudo)spin and electric dipole moment. In the introduction, we describe a ferromagnetically coupled spin-1/2 Bose gas with contact interactions in the mean- field approximation. In particular, we compute and discuss the phase diagram of this gas, since it is a simple system which nevertheless has two order parameters. In the chapters that follow, the ideas touched upon in the introduction are developed in more detail. In Chapter 2 we develop a hydrodynamic description of the ferromagnetic spin-1/2 Bose gas at arbitrary temperatures. We study magnetization relaxation and geometric forces. In particular, we consider the topological Hall effect due to the presence of a skyrmion. In Chapter 3 we investigate the miscible (non-ferromagnetic) spin-1/2 Bose gas at arbitrary temperatures, construct its hydrodynamic description, calculate the thermodynamic properties, and study the collective modes of this system. In Chapter 4 we discuss the influence of the off-diagonal Berry curvature on the Bose-Einstein condensation temperature. Finally, in Chapter 5 we study a Bose-Einstein condensate of dipolar molecules in a weak electric field and find it to be described by a quantum rotor model. Moreover, we show that the molecular Bose-Einstein condensate is a ferroelectric material that is fully disordered by quantum fluctuations.
Ultracold atom experiments offer the unique opportunity to study mixing of different types of superfluid states. Our interest is in superfluid mixtures comprising of particles with different statistics---Bose and Fermi. Such scenarios occur naturally, for example, in dense quantum chromodynamics (QCD) matter. Interestingly, cold atomic experiments are performed in traps with finite spatial extent, thus critically destabilizing the occurrence of various homogeneous phases. Critical to this analysis is the understanding that the trapped system can undergo phase separation, resulting in a unique situation where phase transition in either species (bosons or fermions) can overlap with the phase separation between possible phases. In the present work, we illustrate how this intriguing interplay manifests in an interacting two-species atomic mixture---one bosonic and another fermionic with two spin components---within a realistic trap configuration. We further show that such interplay of transitions can render the nature of the ground state to be highly sensitive to the experimental parameters and the dimensionality of the system.
We consider ultracold imbalanced Fermi gases with strong attractive interactions, for which Cooper-pair formation plays an important role. The two-component mixtures consist either of identical fermionic atoms in two different hyperfine states, or of two different atomic species. In both cases, the number of atoms for each component is allowed to be different, leading to a spin imbalance or a polarization. In the second case, the two different species also give rise to a mass imbalance for the particles. Imbalanced Fermi mixtures play a fundamental role in condensed-matter, nuclear and astroparticle physics. Experimentally, the atomic mixtures are made strongly interacting with the use of two-channel Feshbach resonances. We start out by treating these in more detail. We review the relevant scattering theory, and apply it to the well-studied case of resonant s-wave collisions with zero angular momentum. Furthermore, we discuss the less-studied case of p-wave collisions with nonzero angular momentum, and the inhomogeneous case due to a trapping potential. Having discussed the two-body physics, we turn to the many-body physics. The pioneering experimental studies of Cooper pairing in lithium-6 mixtures with a spin imbalance were performed by Zwierlein et al. [Science 311, 492 (2006)] and Partridge et al. [Science 311, 503 (2006)]. We study the mean-field theory for the polarized Fermi mixture to obtain a qualitative understanding of these experiments. In the unitarity limit, where the scattering length of the interatomic interaction diverges, the theory gives rise to two exotic superfluid phases that are absent in the balanced mixture. These are the gapless superfluid Sarma phase and the phase-separated phase. By using renormalization group techniques to account for the effects of strong interactions, we obtain also quantitative agreement with Monte-Carlo calculations and experiments. We then generalize our knowledge of the solely spin-imbalanced mixture to the mass-imbalanced case, where in particular the mixture consisting of lithium-6 and potassium-40 is experimentally promising. We study the mean-field phase diagram for this mixture in the unitarity limit, where we not only find phase separation and Sarma superfluidity, but also a Lifshitz point. The latter signals an instability towards a supersolid phase. We also include the effects of fluctuations, which do not alter the topology of the phase diagram. However, fluctuation effects lower the critical temperatures with an experimentally relevant factor of three. Finally, we study superfluidity in a fully polarized gas of potassium-40 atoms near a p-wave Feshbach resonance. In particular, we consider coherent Josephson oscillations between the superfluid components in the two channels of the resonance, whose frequencies contain a clear signature of the quantum phase transition that occurs as a function of applied magnetic field.
Manifestations of Quantum Physics at the thermodynamical level are found in a broadrange of physical systems. A famous example is superfluidity, discovered at the beginningof the 20th century and found in many different situations, from liquid helium to neutronstars. Dilute ultracold gases offer a unique versatility to engineer quantum many-bodysystems, which can be directly compared with theory thanks to the controllability of theirenvironment. In this thesis we present several experimental investigations led on ultracoldlithium gases. Lithium provides the possibility to study ensembles of bosons andfermions, with controllable interactions between the constituents. We present experimentaltechniques for preparation and studies of degenerate gases of lithium, with prospects forimprovement of the existing methods. We first report on an investigation of three-bodyrecombination of bosons under a resonant two-body interaction. This study, quantitativelycompared with theory constitutes a benchmark for further studies of the unitary Bose gas.Finally, we present the first experimental realization of a mixture of a Bose superfluid witha Fermi superfluid. We demon- strate that both components are in the superfluid regime,and that the counter-flow motion between them possesses the characteristics of superfluidflow, with the absence of viscosity below a critical velocity, and an onset of friction above.Using collective oscillations of the mixture, we measure the coupling between the twosuperfluids in close agreement with a theoretical model.