We investigate the crystalline stationary states of a dipolar Bose-Einstein condensate in a planar trapping geometry. Our focus is on the ground state phase diagram in the thermodynamic limit, where triangular, honeycomb and stripe phases occur. We introduce a rigorous method to quantify the superfluidity of these states by analysing their nonclassical translational inertia, which allows us to identify favourable parameter regimes for observing supersolid ground states. The resulting superfluid fraction is a rank-2 tensor which is isotropic for the triangular and honeycomb states but anisotropic for the stripe case. We develop two simplified theories to approximately describe the ground states, and consider the relationship to roton softening in the uniform ground state. This also allows us to extend the phase diagram to the low density regime.
We investigate the crystalline stationary states of a dipolar Bose-Einstein condensate in a planar trapping geometry. Our focus is on the ground state phase diagram in the thermodynamic limit, where triangular, honeycomb and stripe phases occur. We introduce a rigorous method to quantify the superfluidity of these states by analysing their nonclassical translational inertia, which allows us to identify favourable parameter regimes for observing supersolid ground states. The resulting superfluid fraction is a rank-2 tensor which is isotropic for the triangular and honeycomb states but anisotropic for the stripe case. We develop two simplified theories to approximately describe the ground states, and consider the relationship to roton softening in the uniform ground state. This also allows us to extend the phase diagram to the low density regime.
We investigate the crystalline stationary states of a dipolar Bose-Einstein condensate in a planar trapping geometry. Our focus is on the ground-state phase diagram in the thermodynamic limit, where triangular, honeycomb, and stripe phases occur. We quantify the superfluid fraction by calculating the nonclassical translational inertia, which allows us to identify favorable parameter regimes for observing supersolid ground states. We develop two simplified theories to approximately describe the ground states and consider the relationship to roton softening in the uniform ground state. This also allows us to extend the phase diagram to the low density regime. While the triangular and honeycomb states have an isotropic superfluid response tensor, the stripe state exhibits anisotropic superfluidity.
Abstract We investigate the superfluid fraction of crystalline stationary states within the framework of mean-field Gross–Pitaevskii theory. Our primary focus is on a two-dimensional Bose–Einstein condensate with a non-local soft-core interaction, where the superfluid fraction is described by a rank-2 tensor. We then calculate the superfluid fraction tensor for crystalline states exhibiting triangular, square, and stripe geometries across a broad range of interaction parameters. Factors leading to an anisotropic superfluid fraction tensor are also considered. We also refine the Leggett bounds for the superfluid fraction of the 2D system. We systematically compare these bounds to our full numerical results, and other results in the literature. This work is of direct relevance to other supersolid systems of current interest, such as supersolids produced using dipolar Bose–Einstein condensates.
This work concerns a theoretical study of a two-component ultra-cold bosonic gas of polarised magnetic dipolar atoms. The atoms interact by short-ranged contact interactions and long-ranged dipole-dipole interactions. The two components can correspond to two different spin-states of a single atomic species or two different atomic species, as realised in recent experiments. Here, we develop a general theory for the ground states and excitations of a two-component dipolar Bose-Einstein condensate. The ground states are described by dipolar Gross-Pitaevskii equations and Bogoliubov-de Gennes equations describe the excitations. We apply these theories to systems in 3D harmonic traps and a uniform planar potential. Specialised numerical methods are developed to provide accurate solutions of these equations, and to deal with the singular dipole-dipole interaction kernel. These methods involve spectral representations based on plane waves and Bessel functions. We also discuss the methods we used for obtaining ground state solutions of the nonlinear Gross-Pitaevskii energy functional. Dipole-dipole interactions can favour spatial structure, as demonstrated in the recent observation of supersolid states of matter in a single-component dipolar gas. In the two-component system these interactions can also favour a ground state with a periodic order in the pseudo-spin density. A two-component system can be either miscible (components overlap) or immiscible (components are separated). Our focus here is in the spin-stripe state, which occurs in a flattened trap geometry and involves an immiscible stripe pattern along one of the weakly trapped or untrapped directions. We examine the dynamical stability conditions of the uniform miscible state in planar confinement, and produce an associated stability phase diagram. We identify the various instabilities and analytical criteria for estimating these. We focus on the occurrence of, and conditions that favour, density-roton or spin-roton excitations, anticipating that the softening of these excitations will lead to structured ground states. To characterize these excitations we introduce density and spin dynamical structure factors, which can be measured in experiments using Bragg spectroscopy. Building on the stability phase diagram we compute ground states where the immiscible state is unstable to spin-rotons, finding triangular and stripe immiscible states with a length scale set by the confinement and interactions. We find that the transition to the stripe state is continuous and coincides with the spin-roton softening in the uniform miscible state. We study the excitations of the spin stripe state which exhibits three gapless excitation branches arising from the spontaneous broken symmetries. We identify a balanced parameter regime, where the excitations exhibit a nonsymmorphic symmetry (a symmetry is nonsymmorphic when it is a combination of point-group operations with nonprimitive lattice translations, i.e. not an integer multiple of lattice constants), which is revealed in the density and spin dynamical structure factors. We also consider the dynamics of the system following a quench from the uniform to immiscible stripe states. In the early time dynamics, a spin-stripe state is formed in a such way that the stripe order only extends over small domains. In subsequent dynamics, these domain sizes grow as the system phase-orders towards its equilibrium state. We identify the relevant topological defects such as disclinations, dislocations and grain boundaries. Using ensembles of large scale simulations we show that dynamical scaling holds for the growth of order.
Recent breakthrough experiments on dipolar condensates have reported the creation of supersolids, including two-dimensional arrays of quantum droplets. Droplet arrays are, however, not the only possible nontrivial density arrangement resulting from the interplay of mean-field instability and quantum stabilization. Several other possible density patterns may occur in trapped condensates at higher densities, including the so-called honeycomb supersolid, a phase that exists, as it is also the case of a triangular droplet supersolid, in the thermodynamic limit. We show that compared to droplet supersolids, honeycomb supersolids have a much-enhanced superfluid fraction while keeping a large density contrast, and constitute in this sense a much better dipolar supersolid. However, in contrast to droplet supersolids, quantized vortices cannot be created in a honeycomb supersolid without driving a transition into a so-called labyrinthic phase. We show that the reduced moment of inertia, and with it the superfluid fraction, can be however reliably probed by studying the dynamics following a scissorslike perturbation.
The behavior of dipolar Bose-Einstein condensates in planar geometries is investigated, focusing on the effects of the polarization orientation. While perpendicular polarization produces a phase diagram with hexagonal, stripes, and honeycomb phases ending at a single critical point, the presence of an in-plane polarization component transforms the critical point into three critical lines, separating two phases at a time and changing radically the appearance of the phase diagram. All transition lines contain first- and second-order regions, while the phase diagram itself shows a resemblance with those displayed by quasi-one-dimensional dipolar systems. Finally, we investigate the effect of introducing an in-plane polarization on the structural properties of the phases and determine the superfluid fraction. Our results show that this process induces an axial deformation on the hexagonal and honeycomb phases, resulting in an anisotropic behavior in the long distance properties of the system like superfluidity. We expect that the rich phenomenology observed provides motivation for new experiments and theoretical works.
We study the ground states of rotating atomic Bose-Einstein condensates with dipolar interactions. We present the results of numerical studies on a periodic geometry which show vortex lattice ground states of various symmetries: triangular and square vortex lattices, ``stripe crystal,'' and ``bubble crystal.'' We present the phase diagram (for systems with a large number of vortices) as a function of the ratio of dipolar to contact interactions and of the chemical potential. We discuss the experimental requirements for observing transitions between vortex lattice ground states via dipolar interactions. We finally investigate the stability of mean-field supersolid phases of a quasi-two-dimensional nonrotating Bose gas with dipolar interactions.