Exploring quantum phase slips in 1D bosonic systems

The goal of my thesis is to experimentally investigate the phenomenon of phase slips in one dimensional (1D) Bose-Einstein condensates, by probing the supercurrent in the presence of an obstacle in different ranges of velocities, interactions and temperature. Phase slips are the primary elementary excitations of the order parameter due to thermal or quantum fluctuations in 1D superfluids and superconductors, in the presence of an obstacle for the superflow and supercurrent. They control the dissipation of nominally frictionless systems by inducing a finite resistance and a finite dissipation in 1D superconductors and superfluids, respectively. Due to the fact that 1D systems are more fragile and more vulnerable to the presence of perturbation and fluctuations than higher dimensional systems, phase slips are more easily detected in 1D systems. Several theoretical models concerning the phenomenon of phase slips in ultracold quantum gases have been built, but an experimental exhaustive picture of quantum phase slips in ultracold superfluids has remained elusive. The system I use is a 1D Bose-Einstein condensate of 39K atoms, in the presence of a 1D optical lattice along the axis of the system, which acts as an obstacle. The Bose Einstein condensate, being a superfluid, should flow without dissipation, even in the presence of an obstacle. Anyway, by performing transport measurements, I observe a finite dissipation, due to phase slips that cause the superfluity breakage. During my Ph.D I focused my attention on this dissipation phenomenon. In the first part of my work, I focused my attention on the system dissipation and I investigated the system oscillation for different values of the velocity. More in detail, I tuned the system velocity in a range between vC/5 and vC, where vC is the critical velocity for breaking superfliduity via the so-called dynamical instability. Depending on the velocity, the systems behaves differently: close to the occurrence of the dynamical instability, I observed an overdamped motion, which is a consequence of the divergence of phase slips. For velocity smaller than vc, instead, I observed how phase slips act on the system: in this situation, the system oscillates with a damping due to the presence of phase slips. I measured the damping rate G due to the presence of phase slips for different values of interaction, velocity and temperature, far from the dynamical instability. In this situation the system never enters in the unstable regime but keeps oscillating with a finite dissipation. The damping rate G, which is related to the phase slips nucleation rate, according to the theory should behave differently depending on the nature of phase slips: in the presence of phaseslips due to thermal fluctuations, the damping rate depends on the temperature, whereas it depends on velocity if the phase slips are due to quantum fluctuations. These observations appear consistent with the theoretically predicted crossover from a regime where the nucleation of phase slips is due to thermal effect to a regime of quantum phase slips and provides the first experimental evidence of quantum phase slips in a 1D atomic superfluid. In the second part of my work, I employed a different method to study the dissipation, by performing transport measurements at constant velocity, also for velocities lower than vC/5, both in the regime of shallow lattices and deep one. When the system is in the superfluid phase, the system dissipation is related to the presence of phase slips, both thermal and quantum depending on the interaction value. Surprisingly, I observed a finite dissipation also when the system is in the insulating phase and it should not dissipate. This dissipation may be due to two different phenomena. The First one is related to the coexistence of a superfluid and a Mott insulating phase, due to the inhomogeneity of our system, whereas the second phenomenon is related to the excitation of the gapped Mott phase. In the presence of a weak optical lattice, it is difficult to discriminate which one of the two effects dominates the observed dissipation. As a consequence, I have repeated the same measurements in the presence of a deep lattice and also in this case I observed a dissipation both in the superfluid phase and in the insulating phase. As in the case of a weak optical lattice, when the system is in the superfluid phase, the system dissipation is related to the presence of phase slips. By comparing the gap of the Mott insulator and the energy acquired during the harmonic potential trap displacement, I found that the Mott insulator gap was an order of magnitude larger than the energy due to the trap shift. As a consequence, I excluded that the finite damping was due to the excitation of the Mott insulator; it seems due to the dissipation of the superfluid phase coexistent with the insulating one. These results open the way for a deeper understanding of the intriguing phenomenon of phase slips, which is still an open topic. There are, in fact, a lot of open questions regarding this kind of excitation, and Bose Einstein condensates may be the ideal system to investigate this phenomenon thanks to their ample tunability and to the relative ease of modelling. For example, with the technique used in the first part of my work, it would be interesting to excite the dipole oscillations in the system in the presence of individual defects or controlled disorder. Moreover, with the technique used in the second part of my work, further studies of the system dissipation during a shift of the harmonic trap at constant velocity, in the absence of the Mott insulator, may give information about the phase slips phenomenon in the very strongly interacting regime.