This thesis presents work on Bose-Einstein condensates in non-harmonic optical potentials. First, a new trap design developed at the University of Queensland is presented that allows the creation of nearly arbitrary two-dimensional potential landscapes by spatially scanning a far red-detuned laser beam using a two-dimensional acousto-optic modulator. In conjunction with a feed-forward technique this trap is capable of producing optical traps which have the necessary stability to be used in ultra-cold atom research. Different geometries are presented. In particular toroidal trap geometries are discussed which are interesting because they offer the possibility for a multiply connected Bose-Einstein condensate. The trap also offers the possibility of dynamic potentials which have been employed to measure the critical velocity of superfluidity in Bose-Einstein condensates. Secondly, measurements on condensation dynamics are presented which use an optical dimple potential superimposed upon a harmonic magnetic trap. In the experiments the dimple potential is ramped on slowly or turned on suddenly for a range of dimple depths and widths and the condensate fraction and temperature are measured as a function of hold times. The measurements taken are compared to equilibrium thermodynamics and quantum kinetic theory and for the first time a quantitative model is presented that allows correct prediction of condensate fractions and temperatures in good agreement with the measurements. Also presented are first results in condensate rethermalization experiments where a dimple potential is suddenly switched off and the condensate left to rethermalize in a much wider magnetic potential. The measurements show a sudden breakdown in condensate fraction followed by a subsequent recovery. So far no theory exists to properly interpret the results as quantum kinetic theory cannot be applied to the problem. Lastly, progress on experiments to measure the critical velocity of superfluidity in Bose-Einstein condensates is reported. Calculations have been carried out that show the creation of solitons and vortices as mechanisms to dissipate energy leading to a breakdown of superfluid behaviour at velocities lower than that predicted by Landau’s theory of superfluidity, the first theory to describe superfluidity. Progress on an experiment is presented that uses the scanning beam trap developed as part of this project to create a line potential with a moving barrier to measure the critical velocity and show solitons and vortices as the mechanisms for energy dissipation as seen in the calculations.
This thesis presents work on Bose-Einstein condensates in non-harmonic optical potentials. First, a new trap design developed at the University of Queensland is presented that allows the creation of nearly arbitrary two-dimensional potential landscapes by spatially scanning a far red-detuned laser beam using a two-dimensional acousto-optic modulator. In conjunction with a feed-forward technique this trap is capable of producing optical traps which have the necessary stability to be used in ultra-cold atom research. Different geometries are presented. In particular toroidal trap geometries are discussed which are interesting because they offer the possibility for a multiply connected Bose-Einstein condensate. The trap also offers the possibility of dynamic potentials which have been employed to measure the critical velocity of superfluidity in Bose-Einstein condensates. Secondly, measurements on condensation dynamics are presented which use an optical dimple potential superimposed upon a harmonic magnetic trap. In the experiments the dimple potential is ramped on slowly or turned on suddenly for a range of dimple depths and widths and the condensate fraction and temperature are measured as a function of hold times. The measurements taken are compared to equilibrium thermodynamics and quantum kinetic theory and for the first time a quantitative model is presented that allows correct prediction of condensate fractions and temperatures in good agreement with the measurements. Also presented are first results in condensate rethermalization experiments where a dimple potential is suddenly switched off and the condensate left to rethermalize in a much wider magnetic potential. The measurements show a sudden breakdown in condensate fraction followed by a subsequent recovery. So far no theory exists to properly interpret the results as quantum kinetic theory cannot be applied to the problem. Lastly, progress on experiments to measure the critical velocity of superfluidity in Bose-Einstein condensates is reported. Calculations have been carried out that show the creation of solitons and vortices as mechanisms to dissipate energy leading to a breakdown of superfluid behaviour at velocities lower than that predicted by Landau’s theory of superfluidity, the first theory to describe superfluidity. Progress on an experiment is presented that uses the scanning beam trap developed as part of this project to create a line potential with a moving barrier to measure the critical velocity and show solitons and vortices as the mechanisms for energy dissipation as seen in the calculations.
The loading of ultra-cold atoms into two-dimensional, time-averaged optical potentials was demonstrated. The time-averaging scheme allows us to make arbitrary and two dimensional potentials, where the potential shape and depth can be dynamically controlled. Traps are produced by rapidly scanning a red-detuned laser focus in any desired pattern using a two-dimensional acousto-optical modulator (AOM). Smooth optical potentials are created when the scanning speed is much higher than any trapping frequency in system, resulting in a time-averaged dipole potential for the atoms. The depth of the potential can be controlled at any time and place by modulating the amplitude of the applied radio frequency to the AOM.
Since the first Bose-Einstein Condensate (BEC) was demonstrated in a dilute atomic gas [1], rotating BECs have been of interest to researchers due to their quantized circulation. We propose a new method of creating controlled vortices in a BEC. Atoms are held in the focus of a scanning beam that rapidly scans to produce a time-averaged potential [2]. We have made theoretical calculations to measure the heating caused by a scanning beam trap and to find the regime where a scanning beam produces a time-averaged potential. The scanning beam splits the BEC into four parts then imprints a relative phase on each part such that, when the parts recombine, a vortex forms. We have simulated such a system using the Gross-Pitaevskii equation and have demonstrated the laser scanning pattern using a two-dimensional acousto-optic modulator (AOM). The optics for the scanning beam trap have been designed and the experimental trap dimensions are used in our simulations. To carry out the experimental part of this project, we are currently building and testing an all-optical BEC apparatus: we have made a magneto-optical trap (MOT) which cools the 87Rb atomsto 30uK. We are currently setting up an optical dipole trap to perform the final stages of cooling.
Bose-Einstein condensates are superfluids which when engineered under suitable laboratory conditions support dissipationless flows and facilitate the development of quantum enhanced applications. Realising those suitable conditions requires experimental technologies which enable the complete dynamical control of the superfluid density and velocity profiles. This thesis conceives and demonstrates several dynamical controls using optical potentials formed using commercially available spatial-light-modulator technologies, and constitutes important progress toward many future experiments and the realisation of practical applications.Experimental systems and control sequences are designed using complimentary acousto-optic deflection and digital micromirror spatial light modulator technologies, rather than magnetic field based approaches, given their inherently favourable spatial and temporal resolutions. An experimental apparatus which includes a time-averaged acousto-optic controlled potential is constructed to generate toroidal potentials suitable for developing inertial rotation sensors. Density and phase manipulation strategies are developed using that apparatus which prepare condensates with arbitrarily controllable wavefunctions. Complementary digital micromirror based adaptations are subsequently developed during this project.Corrugations across the condensate density distributions are reduced using an iteratively corrected spatial light modulator configuration. Rigorous and generalised scan requirements are established for engineering time-averaged potentials; numerical methods are conceived which facilitate the simulation of these systems. The combination of the feedforward density control and time-averaging phase control enable the deterministic preparation of persistent currents around multiply connected condensates. These capabilities enable experiments into many-body groundstates, superfluid turbulence, atomtronics and metrology.Spatial light modulator trapping sequences geared toward the study of Onsager vortices, vortex dipole optics, inertial rotation sensing and wake turbulence are specifically developed. Several upgraded experimental systems are finally designed using knowledge gained during this thesis, which enable these various applications in future. This thesis therefore constitutes significant progress toward several important experimental investigations using dynamically controllable optical trapping technologies.
Research on degenerate quantum gases of dilute atomic gases has become a well established field in modern physics and revealed many astonishing phenomena. The first experiments on Bose and Fermi gases at low temperatures aimed to prove the predictions of quantum mechanics on a macroscopic scale. Since then we make use of Bose and Fermi gases and mixtures of any combination to model other many body systems. A property of Bose Einstein condensates is superfluidity which we want to investigate further in toroidal potentials with a new apparatus which can produce an all-optical BEC. This means trapping thermal atoms in an optical dipole trap and cooling the sample by evaporation into the quantum degenerate regime. This new technique was first successfully demonstrated by M.D. Barrett et al. in 2001 and promises to be faster and simpler than the well established and common way of evaporation in magnetic traps. A new apparatus was built by the author and we report on the experimental details of the trapping of 87Rb atoms from a hot vapour and cooling them in a magneto-optical trap, and further evaporation to quantum degeneracy in a crossed optical dipole trap operating at a wavelength of l = 1064nm, with accurate control of the power in both beams via a feedback loop. The evaporation of neutral atoms of 87Rb in far red detuned optical dipole traps using linearly polarised laser light is spin independent, and mixed spinor condensates of the F=1 manifold can be formed. As an empirical technique we found by applying a magnetic gradient field during the final evaporation that we can selectively populate mF spin states or prepare mixtures. This intriguing mechanism was found earlier as well by M. S. Chang, but is yet not fully understood and subject of our future research. We can now routinely prepare an almost pure condensate containing up to 7000 atoms in the condensed phase purely in the mF = 0 spin state. This thesis contains the theoretical framework of Bose-Einstein condensation with in mean-field theory, the concepts of cooling of atoms in magneto-optical traps, the trapping and evaporation in optical dipole traps and the experimental realisation. In the future this experiment will be extended by a fast scanning dipole trap to apply arbitrary time averaged potentials of more complex geometries, such as toroidal traps to study the effects of superfluidity in a periodic potential. The production of toroidal traps and cold atoms in such a trap are outlined in this thesis.
In this thesis, both experimental studies and numerical simulations of ultracold atoms in optical potentials are presented in a collection of nine scientific papers. In particular, noise-induced transport in dissipative optical lattices and superfluid properties of Bose-Einstein condensates have been studied. Noise is usually regarded as a complication to most systems and as something that needs to be minimized. However, in a series of experiments at Umeå University, noise has been shown to play a key role for laser-cooled cesium atoms trapped in dissipative optical lattices. By using a combination of two dissipative optical lattices, where the relative spatial phase between them can be controlled, a so-called Brownian motor can be realized, where energy can be extracted from the inherent noise. In the experiment, this energy is used to control the transport of the laser-cooled atoms in real time and along pre-designed paths. This thesis also presents a way to characterize this system in terms of energy conversion efficiency and coherence of the transport, which may allow for a more straightforward comparison with other systems that rely on noise rectification. In the studies, it is also shown that the noise triggers a downward drift due to gravity, even though the optical potential should support the atoms. Further investigation of this might help to understand the underlying principles of laser cooling, as well as showing that the system might be suitable as a flexible test bed for statistical physics. In close relation to the experimental system, two numerical simulations are also presented, one in which different ways to induce asymmetries between two periodic potentials are investigated, and one in which a proposal for detecting quantum walks is explored. In the second part of the thesis, a work from the Joint Quantum Institute is presented, where a long-lived persistent current in a toroidal Bose-Einstein condensate, held in an all-optical trap, is created. The critical velocity of the superflow is measured in the presence of a tunable barrier. The system can be seen as a first realization of an elementary closed-loop atom circuit. Finally a theoretical study of the crossover between one- and two-dimensional systems is presented, in particular the transition between a two-dimensional superfluid to a one-dimensional Mott insulator is investigated.
In this work we investigate the dynamics of Bose-Einstein condensates (BECs) in inhomogeneous potential landscapes. As this research field continues to develop, more attention will focus on non-equilibrium systems, on potential applications that use condensates, and on the integration of cold atoms with other physical systems. This thesis covers all of these areas. We begin by recapping the historical background of condensate physics, with a definition of the condensed phase and discussion of various analytical quantities of relevance to this work. The Landau picture of supefluidity and predictions of its breakdown, given by the Landau criterion, is particularly pertinent to the results on supersonic flow in an inhomogeneous system. After outlining current experimental procedures, we present a computationally efficient modelling technique, used in our numerical simulations of atomic condensates. We then use this technique to study the dynamics of supersonic condensate flow, in the presence of a perturbing potential. Normally one would expect this situation to introduce disturbances, known as Landau excitations into the system, potentially destroying it. However, we find, under certain circumstances, complete suppression of Landau excitations: a behaviour that has not, to our knowledge, been previously observed. The efficiency of our chosen modeling technique allowed the possibility to conduct the large phase space campaigns necessary to find these special circumstances. On investigation, the mechanism resulting in the suppression of these Landau excitations is continuously related to the presence of transmission resonances in an equivalent linear quantum system. This demonstration of a link between linear and non-linear quantum regimes is of great interest in understanding possible behaviour in other non-equilibrium superfluid systems. Finally, we consider the magnetic fields from small scale (~ 1 µm) quantum electronic devices fabricated within a two-dimensional electron gas (2DEG). We demonstrate that atomic condensates provide a powerful tool for imaging these fields, or indeed similar fields created by other structures. Using a Fourier method, we show that the field profile that would be measured by the condensate can be used to recreate the current density of the 2DEG structure. The spatial resolution of this current mapping technique is limited only by the separation of the condensate from the current-carrying structure. We also show that quantum electronic conductors in 2DEGs are well suited to form a new generation of atom chips capable of trapping atoms < 1 µm away, thereby reducing both the size and power requirements of chip-trap potentials.