Search

We simulate a molecular Bose-Einstein condensate in the strongly dipolar regime, observing the existence of self-bound droplets, as well as their splitting into multiple droplets by confinement-induced frustration. Our quantum Monte Carlo approach goes beyond the limits of the established effective mean-field theories for dipolar quantum gases, revealing small droplets produced by strong dipolar interactions outside known stable regimes. The simulations include realistic molecular interactions and therefore have direct relevance for current and future experiments., Comment: 6 + 1 pages text + Supplementary Material (appendix). 3 figures + 1 table

We present a quantum Monte Carlo based density functional to describe droplet formation and supersolidity in dipolar systems. The usual Lee-Huang-Yang term, accounting for quantum correlations in the conventional extended Gross-Pitaievskii equation (eGPE), has been substituted by the correlation energy evaluated with Quantum Monte Carlo. We demonstrate the ability of the new functional to reproduce existing experimental data for the minimum critical number of atoms $N_\mathrm{c}$ required for droplet formation. $N_\mathrm{c}$ is a challenging quantity for theoretical predictions, and the eGPE provides only a qualitative description of it, mainly when it is applied to Dysprosium. We also use the new approach to characterize the BEC-supersolid transition. The quantum Monte Carlo based functional can be easily implemented in any existing eGPE code, improving the description of dipolar systems without increasing the computational cost., Comment: 4 pages, 3 figures

We study the quasiparticle properties of a dipolar impurity immersed in a two-dimensional dipolar bath. We use the ab-initio Diffusion Monte Carlo technique to determine the polaron energy, effective mass and quasiparticle residue. We find that both the polaron energy and quasiparticle residue follow a universal behaviour with respect to the polarization angle when properly scaled in terms of the scattering length. This trend is maintained over a wide range of values of the gas parameter, even in the highly correlated regime. Instead, the effective mass shows growing anisotropy as the tilting angle is increased, which is induced, mainly, by the anisotropy of the impurity-boson dipole-dipole interaction. Surprisingly, the effective mass is larger in the direction of minimum inter-particle repulsion. Finally, we use our Monte Carlo results to check the accuracy of perturbative approaches and determine their range of validity in terms of the gas parameter., Comment: 10 pages, 8 figures

Future fusion nuclear reactors will produce sustainable energy form the fusion of deuterium and tritium. In order to do so, the reactors will need to produce their own tritium through the neutron bombardment of lithium. Such reaction will produce tritium and helium inside the breeding blanket of the reactor. Helium can trigger nucleation mechanisms due to its very low solubility inside liquid metals. Consequently, the knowledge and understanding of the microscopic processes of helium nucleation is crucial to improve the efficiency, sustainability and safety of the fusion energy production. The formation of helium bubbles inside the liquid metal used as breeding material may be a serious issue that has yet to be fully understood. We provide further insight on the behavior of lithium and helium mixtures at experimentally corresponding operating conditions (800~K and pressures between 1 and 100 bar) using a suitable microscopic model able to describe the helium and lithium atomic interactions, in excellent agreement with available experimental data. The simulations predict the formation of helium bubbles with radii around 10 Angstroem at ambient pressure and with surface tension values between 0.6-1.0 N/m, with a dependency of the concentration of helium. We also report cohesive energies of helium as well as a quantitative estimation of the Hildebrand and Kumar cohesion parameters., Comment: 32 pages, 10 figures

We investigate the phases of a Bose-Einstein condensate of dipolar atoms restricted to move in a two-dimensional plane. The dipole moments are all aligned in a direction tilted with respect to the plane normal. As a result of the attractive and repulsive components of the dipole-dipole interaction, the dipolar gas has a self-bound phase, which is stabilized by quantum fluctuations. Furthermore, tilting the dipoles tunes the anisotropy of the dipole-dipole interaction, which can trigger a spatial density modulation. In this work we study these two aspects and investigate the conditions for the formation of a self-bound and striped phase, which has been realized in experiments with dipolar droplets. We use a variational method based on the hypernetted-chain Euler-Lagrange optimization of a Jastrow-Feenberg ansatz for the many-body wave function to study the ground state properties. This method takes into account quantum fluctuations in a non-perturbative way and thus can be used also for strongly correlated systems., Comment: 9 pages, 9 figures

We study the dynamics of a one-dimensional Bose gas at unit filling in both shallow and deep optical lattices and obtain the dynamic structure factor ${S(k,\omega)}$ by monitoring the linear response to a weak probe pulse. We introduce a new procedure, based on the time-dependent variational Monte Carlo method (tVMC), which allows to evolve the system in real time, using as a variational model a Jastrow-Feenberg wave function that includes pair correlations. Comparison with exact diagonalization results of ${S(k,\omega)}$ obtained on a lattice in the Bose-Hubbard limit shows good agreement of the dispersion relation for sufficiently deep optical lattices, while for shallow lattices we observe the influence of higher Bloch bands. We also investigate non-linear response to strong pulses. From the power spectrum of the density fluctuations we obtain the excitation spectrum, albeit broadened, by higher harmonic generation after a strong pulse with a single low wave number. As a remarkable feature of our simulations we furthermore demonstrate that the full excitation spectrum can be retrieved from the power spectrum of the density fluctuations due to the stochastic noise inherent in any Monte Carlo method, without applying an actual perturbation., Comment: 12 pages, 3 figures

Probabilistic models based on Restricted Boltzmann Machines (RBMs) imply the evaluation of normalized Boltzmann factors, which in turn require from the evaluation of the partition function Z. The exact evaluation of Z, though, becomes a forbiddingly expensive task as the system size increases. This even worsens when one considers most usual learning algorithms for RBMs, where the exact evaluation of the gradient of the log-likelihood of the empirical distribution of the data includes the computation of Z at each iteration. The Annealed Importance Sampling (AIS) method provides a tool to stochastically estimate the partition function of the system. So far, the standard use of the AIS algorithm in the Machine Learning context has been done using a large number of Monte Carlo steps. In this work we show that this may not be required if a proper starting probability distribution is employed as the initialization of the AIS algorithm. We analyze the performance of AIS in both small- and large-sized problems, and show that in both cases a good estimation of Z can be obtained with little computational cost.

This is a Reply to the Comment from F. Cinti and M. Boninsegni on our recent work on the Berezinskii-Kosterlitz-Thouless (BKT) phase transition in a two-dimensional dipolar system [R.Bomb\'in, F. Mazzanti and J. Boronat, Physical Review A 100, 063614 (2019)]. The main criticism about our work, expressed in that Comment, is that we did not explicitly report the two spatial contributions to the total superfluid fraction. Here, we analyze our results for a point of the phase diagram corresponding to the stripe phase, close to the gas to stripe transition line, and for a temperature below the BKT critical temperature. The scaling with the system size of the contribution to the superfluid fraction, coming from the direction in which spatial order appears, shows that it remains finite in the thermodynamic limit, as we already stated in our original work. This allow us to state that the stripe phase is superfluid at low temperatures. Furthermore, we offer some comments that help to understand where the differences between the results of Cinti and Boninsegni and ours comes from.

A two-dimensional quantum system of dipoles, with a polarization angle not perpendicular to the plane, shows a transition from a gas to a stripe phase. We have studied the thermal properties of these two phases using the path integral Monte Carlo (PIMC) method. By simulating the thermal density matrix, PIMC provides exact results for magnitudes of interest such as the superfluid fraction and the one-body density matrix. As it is well known, in two dimensions the superfluid-to-normal phase transition follows the Berezinskii-Kosterlitz-Thouless (BKT) scenario. Our results show that both the anisotropic gas and the stripe phases follow the BKT scaling laws. At fixed density and increasing the tilting angle, the transition temperature decreases in going from the gas to the stripe phase. Superfluidity in the perpendicular direction to the stripes is rather small close to the critical temperature but it becomes larger at lower temperatures, mainly close to the transition to the gas. Our results are in qualitative agreement with the supersolidity observed recently in a quasi-one-dimensional array of dipolar droplets.

We study the repulsive polaron problem in a two-component two-dimensional system of fermionic atoms. We use two different interaction models: a short-range (hard-disk) potential and a dipolar potential. In our approach, all the atoms have the same mass and we consider the system to be composed of a uniform bath of a single species and a single atomic impurity. We use the diffusion Monte Carlo method to evaluate polaron properties such as its chemical potential and pair distribution functions, together with a discussion on the deficit of volume induced by the impurity. We also evaluate observables that allow us to determine the validity of the quasi-particle picture: the quasi-particle residue and the effective mass of the polaron. Employing two different potentials allows us to identify the universality regime, where the properties depend only on the gas parameter $n a_s^2$ fixed by the bath density and the two-dimensional scattering length.

Dipolar quantum droplets are exotic quantum objects that are self-bound due to the subtle balance of attraction, repulsion and quantum correlations. Here we present a systematic study of the critical atom number of these self-bound droplets, comparing the experimental results with extended mean-field Gross-Pitaevskii equation (eGPE) and quantum Monte-Carlo simulations of the dilute system. The respective theoretical predictions differ, questioning the validity of the current theoretical state-of-the-art description of quantum droplets within the eGPE framework and indicating that correlations in the system are significant. Furthermore, we show that our system can serve as a sensitive testing ground for many-body theories in the near future.

We study a two-component mixture of fermionic dipoles in two dimensions at zero temperature, interacting via a purely repulsive $1/r^3$ potential. This model can be realized with ultracold atoms or molecules, when their dipole moments are aligned in the confinement direction orthogonal to the plane. We characterize the unpolarized mixture by means of the Diffusion Monte Carlo technique. Computing the equation of state, we identify the regime of validity for a mean-field theory based on a low-density expansion and compare our results with the hard-disk model of repulsive fermions. At high density, we address the possibility of itinerant ferromagnetism, namely whether the ground state can be fully polarized in the fluid phase. Within the fixed-node approximation, we show that the accuracy of Jastrow-Slater trial wave functions, even with the typical two-body backflow correction, is not sufficient to resolve the relevant energy differences. By making use of the iterative-backflow improved trial wave functions, we observe no signature of a fully-polarized ground state up to the freezing density., Comment: 12 pages, 6 figures

Recent experiments confirmed that fluctuations beyond the mean-field approximation can lead to self-bound liquid droplets of ultra-dilute binary Bose mixtures. We proceed beyond the beyond-mean-field approximation, and study liquid Bose mixtures using the variational hypernetted-chain Euler Lagrange method, which accounts for correlations non-perturbatively. Focusing on the case of a mixture of uniform density, as realized inside large saturated droplets, we study the conditions for stability against evaporation of one of the components (both chemical potentials need to be negative) and against liquid-gas phase separation (spinodal instability), the latter being accompanied by a vanishing speed of sound. Dilute Bose mixtures are stable only in a narrow range near an optimal ratio $\rho_1/\rho_2$ and near the total energy minimum. Deviations from a universal dependence on the s-wave scattering lengths are significant despite the low density., Comment: 5 pages, 5 figures

We study the superfluid response, the energetic and structural properties of a one-dimensional ultracold Bose gas in an optical lattice of arbitrary strength. We use the Bose-Fermi mapping in the limit of infinitely large repulsive interaction and the diffusion Monte Carlo method in the case of finite interaction. For slightly incommensurate fillings we find a superfluid behavior which is discussed in terms of vacancies and interstitials. It is shown that both the excitation spectrum and static structure factor are different for the cases of microscopic and macroscopic fractions of defects. This system provides a extremely well-controlled model for studying defect-induced superfluidity., Comment: 14 pages, 13 figures, published version

One-dimensional Bose gas with contact interaction in optical lattices at zero temperature is investigated by means of the exact diffusion Monte Carlo algorithm. The results obtained from the fundamental continuous model are compared with those obtained from the lattice (discrete) Bose-Hubbard model, using exact diagonalization, and from the quantum sine-Gordon model. We map out the complete phase diagram of the continuous model and determine the regions of applicability of the Bose-Hubbard model. Various physical quantities characterizing the systems are calculated and it is demonstrated that the sine-Gordon model used for shallow lattices is inaccurate., Comment: 5 figures

We present a method based on the Path Integral Monte Carlo formalism for the calculation of ground-state time correlation functions in quantum systems. The key point of the method is the consideration of time as a complex variable whose phase $\delta$ acts as an adjustable parameter. By using high-order approximations for the quantum propagator, it is possible to obtain Monte Carlo data all the way from purely imaginary time to $\delta$ values near the limit of real time. As a consequence, it is possible to infer accurately the spectral functions using simple inversion algorithms. We test this approach in the calculation of the dynamic structure function $S(q,\omega)$ of two one-dimensional model systems, harmonic and quartic oscillators, for which $S(q,\omega)$ can be exactly calculated. We notice a clear improvement in the calculation of the dynamic response with respect to the common approach based on the inverse Laplace transform of the imaginary-time correlation function., Comment: Accepted for publication on "Jornal of Chemical Physics"

Restricted Boltzmann Machines (RBMs) are general unsupervised learning devices to ascertain generative models of data distributions. RBMs are often trained using the Contrastive Divergence learning algorithm (CD), an approximation to the gradient of the data log-likelihood. A simple reconstruction error is often used to decide whether the approximation provided by the CD algorithm is good enough, though several authors (Schulz et al., 2010; Fischer & Igel, 2010) have raised doubts concerning the feasibility of this procedure. However, not many alternatives to the reconstruction error have been used in the literature. In this manuscript we investigate simple alternatives to the reconstruction error in order to detect as soon as possible the decrease in the log-likelihood during learning., Comment: 7 pages, 4 figures

We present calculations of the ground state and excitations of an anisotropic dipolar Bose gas in two dimensions, realized by a non-perpendicular polarization with respect to the system plane. For sufficiently high density an increase of the polarization angle leads to a density instability of the gas phase in the direction where the anisotropic interaction is strongest. Using a dynamic many-body theory, we calculate the dynamic structure function in the gas phase which shows the anisotropic dispersion of the excitations. We find that the energy of roton excitations in the strongly interacting direction decreases with increasing polarization angle and almost vanishes close to the instability. Exact path integral ground state Monte Carlo simulations show that this instability is indeed a quantum phase transition to a stripe phase, characterized by long-range order in the strongly interacting direction., Comment: 5 pages, 4 figures