The postulates of the eigenstate thermalization hypothesis (ETH) expresses that the thermalization occurs due to the individual eigenstate of the system’s Hamiltonian. But the ETH sheds no light on the dynamics that lead toward the thermalization. In this paper, we observe the thermalization of a Bose–Einstein condensate (BEC) confined in an optical lattice potential that is embedded on the harmonic trap. Such optical lattice potential offers local friction to the oscillating BEC. The spread in the temporal density plot of BEC shows the thermalization of the BEC. Moreover, we observe that the presence of a parity–time (PT)-symmetric potential greatly influences the BEC dynamics and the thermalization of the system. The presence of a PT-symmetric potential offers a way to manipulate the mean position of the BEC to a desired location and for a desired length of time.
The postulates of the eigenstate thermalization hypothesis (ETH) expresses that the thermalization occurs due to the individual eigenstate of the system’s Hamiltonian. But the ETH sheds no light on the dynamics that lead toward the thermalization. In this paper, we observe the thermalization of a Bose–Einstein condensate (BEC) confined in an optical lattice potential that is embedded on the harmonic trap. Such optical lattice potential offers local friction to the oscillating BEC. The spread in the temporal density plot of BEC shows the thermalization of the BEC. Moreover, we observe that the presence of a parity–time (PT)-symmetric potential greatly influences the BEC dynamics and the thermalization of the system. The presence of a PT-symmetric potential offers a way to manipulate the mean position of the BEC to a desired location and for a desired length of time.
Abstract If and how an isolated quantum system thermalizes despite its unitary time evolution is a long‐standing, open problem of many‐body physics. The eigenstate thermalization hypothesis (ETH) postulates that thermalization happens at the level of individual eigenstates of a system's Hamiltonian. However, the ETH requires stringent conditions to be validated, and it does not address how the thermal state is reached dynamically from an initial non‐equilibrium state. We consider a Bose‐Einstein condensate (BEC) trapped in a double‐well potential with an initial population imbalance. We find that the system thermalizes although the initial conditions violate the ETH requirements. We identify three dynamical regimes. After an initial regime of undamped Josephson oscillations, the subsystem of incoherent excitations or quasiparticles (QP) becomes strongly coupled to the BEC subsystem by means of a dynamically generated, parametric resonance. When the energy stored in the QP system reaches its maximum, the number of QPs becomes effectively constant, and the system enters a quasi‐hydrodynamic regime where the two subsystems are weakly coupled. In this final regime the BEC acts as a grand‐canonical heat reservoir for the QP system (and vice versa), resulting in thermalization. We term this mechanism dynamical bath generation (DBG).
Dynamics and properties of nonlinear matter waves in a trapped BEC subject to a PT-symmetric linear potential, with the trap in the form of a super-Gaussian potential, are investigated via a variational approach accounting for the complex nature of the soliton. In the process, we address how the shape of the imaginary part of the potential, that is, a gain-loss mechanism, affects the self-localization and the stability of the condensate. Variational results are found to be in good agreement with full numerical simulations for predicting the shape, width, and chemical potential of the condensate until the PT breaking point. Variational computation also predicts the existence of solitary solution only above a threshold in the particle number as the gain-loss is increased, in agreement with numerical simulations.
By direct numerical simulation of the time-dependent Gross-Pitaevskii equation, we study different aspects of the localization of a noninteracting ideal Bose-Einstein condensate (BEC) in a one-dimensional bichromatic quasiperiodic optical-lattice potential. Such a quasiperiodic potential, used in a recent experiment on the localization of a BEC [Roati et al., Nature (London) 453, 895 (2008)], can be formed by the superposition of two standing-wave polarized laser beams with different wavelengths. We investigate the effect of the variation of optical amplitudes and wavelengths on the localization of a noninteracting BEC. We also simulate the nonlinear dynamics when a harmonically trapped BEC is suddenly released into a quasiperiodic potential, as done experimentally in a laser speckle potential [Billy et al., Nature (London) 453, 891 (2008)]. We finally study the destruction of the localization in an interacting BEC due to the repulsion generated by a positive scattering length between the bosonic atoms.
This thesis includes theoretical studies regarding stability and manipulation of Bose-Einstein condensates (BECs) of ultra-cold atoms in 2D trapping geometry, as well as formation of steady states of exciton-polariton Bose-Einstein condensates created in solid states. We analyze and numerically model the dynamics and localization of the condensates using mean-field model. Chapter 1 contains an introduction to the physics of ultra-cold atom BEC and excitonpolariton BEC which provides a framework for the work presented in later chapters. In Chapter 2, we consider a method for achieving dynamically controllable transport of highly mobile matter-wave solitons in an ultra-cold atom BEC with attractive interparticle interaction loaded into a driven two-dimensional optical lattice. Our numerical analysis based on the mean-field model and the theory based on the effective particle approach demonstrate that fast, time-periodic rocking of the two-dimensional optical lattice enables efficient stabilization and manipulation of spatially localized matter wave packets via induced reconfigurable mobility channels. Chapter 3 consists of an investigation of the instability - collapse of a BEC with attractive interactions. In this chapter we explore the influence of an orbital angular momentum on the collapse of vortex-free elliptic clouds of Bose-Einstein condensates trapped in a radially symmetric harmonic potential or a rotating elliptic potential. The results of our analysis show that the number of trapped ultracold atoms corresponding to the collapse threshold can be radically increased for such rotating nonlinear matter waves in a radially harmonic trap. The results corresponding to a BEC cloud confined in a rotating elliptic trap show that the elongated stationary states can be parallel or perpendicular to the long axis of the trap and display bistable nature. In Chapter 4, we examine spatial localization and dynamical stability of Bose-Einstein condensates of exciton-polaritons in microcavities under the condition of off-resonant spatially inhomogeneous optical pumping both with and without a harmonic trapping potential. We employ the open-dissipative Gross-Pitaevskii model for describing an incoherently pumped polariton condensate coupled to an exciton reservoir. We reveal that spatial localization of the steady-state condensate occurs due to effective self-trapping created by the polariton flows, regardless of the presence of the external potential. A ground state of the polariton condensate with repulsive interactions between the quasiparticles represents a dynamically stable bright dissipative soliton. We also investigate the conditions for sustaining spatially localized structures, with nonzero angular momentum, in the form of single-charge vortices. Chapter 5 consider the existence of novel spatially localized states of exciton-polariton Bose-Einstein condensates in semiconductor microcavities with fabricated periodic inplane potentials. Our theory shows that, under the conditions of continuous nonresonant pumping, localization is observed for a wide range of optical pump parameters due to effective potentials self-induced by the polariton flows in the spatially periodic system. We show that the self-localization of exciton-polaritons in the lattice may occur both in the gaps and bands of the single-particle linear spectrum, and is dominated by the effects of gain and dissipation rather than the structured potential, in sharp contrast to the conservative condensates of ultra-cold alkali atoms.
A PT-symmetric Bose-Einstein condensate can be theoretically described using a complex optical potential, however, the experimental realization of such an optical potential describing the coherent in- and outcoupling of particles is a nontrivial task. We propose an experiment for a quantum mechanical realization of a PT-symmetric system, where the PT-symmetric currents of a two-well system are implemented by coupling possibly time-dependently varied additional wells to the system, which act as particle reservoirs and thus form a Hermitian system. We map the time-dependence of the amplitudes of a frozen Gaussian variational ansatz to a Schrödinger equation with a Hamiltonian matrix. This relates the parameters of a realistic external potential to the matrix elements of a matrix model. On one side, we can use the matrix model as a computationally cheap model to obtain results, on the other hand we can then map the time-dependence of the matrix elements back to the parameters of the potential, which would serve as an input for an experimental setup. In terms of these simple matrix model we derive conditions under which two wells of the Hermitian multi-well system behave exactly as the two wells of the PT-symmetric system. It turns out that the matrix elements of the Hermitian system must be time-dependent to fulfill the conditions for a sufficient time. These results are applied to calculate the time-dependencies of the matrix elements, first, by means of a four-well system, then consequently building up until we arrive at a system with a large number of wells, which can be analyzed and interpreted in terms of optical lattices. As a second method we use a full time-dependent Gaussian variational ansatz, where every variational parameter of the Gaussian functions is chosen to be time-dependent. This should give more accurate results in that this ansatz can describe significant more degrees of freedom. We derive conditions analogous to that of the matrix models that must be fulfilled such that PT-symmetry can be realized. This method is applied and results are obtained for the important case of an adiabatic current ramp which is proposed as a realistic experimental method to create a PT-symmetric ground state. Finally, the results to both methods are compared and in particular the approximations that lead to the simple matrix models are justified.
A Bose-Einstein condensate (BEC) is a macroscopic ensemble of atoms that collectively behaves as a single quantum-mechanical object. In this thesis I describe an apparatus built to create and study a BEC, as well as initial experiments carried out using a one-dimensional optical standing wave (optical lattice) to manipulate the BEC. In the first experiment, a weak optical lattice Bragg-reflects a BEC moving in a nearly harmonic trap. The Bragg reflection combined with the harmonic trapping potential cause the BEC to oscillate, much as light oscillates in an optical cavity. In the second experiment, a deep optical lattice applied to an atom cloud for a short duration is used to split and recombine the atomic wavefunction. By doing this we can create two different types of atom interferometer: a small-separation atom interferometer which works with either a BEC or a thermal gas, and a large-separation atom interferometer which works only with a BEC and whose interferometric arms are, in principle, separately addressable. In the third experiment, an optical lattice potential is slowly turned on. Under the influence of this lattice potential, the BEC loses its superfluid properties, but fully regains them when the lattice is turned back off. This behavior is consistent both with a reversible transition to a 1-D Mott insulator state and with mean-field effects in a deep lattice; both interpretations are discussed. In the latter part of this thesis I describe an apparatus designed to accommodate experiments involving both a BEC and an ion. The interactions between an ion and a BEC have never before been experimentally investigated. Our apparatus includes both the means to localize an ion by actively shielding against stray electric fields and to spatially image ions. This should allow future experiments in which a single ion is embedded in a BEC. In addition to imaging ions, the apparatus provides the capability to create high-resolution images of distributions of neutral atoms by photoionizing them and then quickly imaging them. I conclude the thesis with a discussion of future experiments exploring ion-BEC interactions.