To engineer the wavefunction of a Bose–Einstein condensate is to exert control over both the density and phase of the Bose–Einstein condensate order parameter. Having the ability to engineer the condensate order parameter down to the smallest length scale relevant to condensate dynamics—the healing length scale—would enable the study of new combinations of topological defects and may pave the way to using Bose–Einstein condensates as versatile, precise quantum simulators. In this thesis, we present a new wavefunction engineering technique which reaches the sub-optical healing length scale. Influenced by magnetic resonance imaging, we name this technique magnetic resonance control. This technique uses time-varying coupling between internal spin states of a spinor Bose–Einstein condensate within a magnetic field gradient to address spatial regions of the condensate, enabling control over both the phase and the density of the condensate order parameter down to the healing length scale. Techniques already exist to engineer condensate wavefunctions, but not with such a fine degree of control. These techniques primarily rely on either the intensity variation of a laser beam, which limits the resolution to the diffraction limit (larger than the typically sub-optical healing length scale), or the adiabatic inversion of a magnetic trapping potential, which can not be easily changed to produce a variety of structures in the condensate wavefunction. To develop our magnetic resonance control technique, we simulate a spinor condensate in one dimension with time-dependent coupling between spin states and time-dependent external magnetic field gradients using the Gross–Pitaevskii equation. We show that magnetic resonance control can engineer a single black soliton using experimentally feasible parameters. A black soliton is an ideal target state to select for this demonstration because engineering such a state requires control over both the phase and density of the condensate with healing-length resolution. We demonstrate that magnetic resonance control can be extended to engineer more complicated target states by simulating the creation of multiple solitons in a condensate, with control over the initial positions and trajectories of the solitons. When magnetic resonance control is applied to Bose–Einstein condensates in the laboratory, it will be necessary to have an imaging system capable of resolving the fine structures created. As an alternative to high-cost, custom-manufactured lenses, and in-vacuum optical systems, I have designed and bench-tested an objective lens with a high numerical aperture (0.36) and a long working distance (35 mm) consisting of standard catalogue lenses. Using 780 nm light, suitable for imaging rubidium condensates, this objective can achieve a resolution of 1.3 μm across a diffraction-limited field of view of 360 μm through a 5 mm thick glass window of a science cell. By changing the spacing between the lens elements, this objective lens can compensate for the aberrations produced by a glass window up to 15 mm thick, and by changing the aperture size the objective becomes suitable for diffraction-limited monochromatic imaging on the D2 line of all the alkalis. Before performing proof-of-principle magnetic resonance control experiments on real Bose–Einstein condensates, we needed to construct an experimental apparatus capable of producing spinor Bose–Einstein condensates. In this thesis I summarise my main contributions to this group endeavour, including: constructing the ultra-high vacuum system; supervising the bakeout of our vacuum system; designing and aligning the optical systems to produce laser beams of different, tunable frequencies to trap, cool, and image rubidium gas; trapping a cloud of rubidium atoms in a magneto-optical trap; constructing laser beam shutters, photodetectors, and photodetector signal filters; and designing and constructing our side imaging and top imaging systems. Using our spinor Bose–Einstein condensate apparatus, I performed the first proof-of-principle magnetic resonance control experiments. With a pair of condensates side-by-side, separated by 200 μm, we can use magnetic resonance control to invert the spin of one condensate only, while leaving the other condensate unaffected. We anticipate magnetic resonance control being used in the laboratory to engineer the first black soliton in a Bose–Einstein condensate. Looking beyond solitons, magnetic resonance control could have applications to the field of magnon spintronics, and extending the technique to higher dimensions could enable the study of exotic topological defects such as spin knots in a quantum fluid.
To engineer the wavefunction of a Bose–Einstein condensate is to exert control over both the density and phase of the Bose–Einstein condensate order parameter. Having the ability to engineer the condensate order parameter down to the smallest length scale relevant to condensate dynamics—the healing length scale—would enable the study of new combinations of topological defects and may pave the way to using Bose–Einstein condensates as versatile, precise quantum simulators. In this thesis, we present a new wavefunction engineering technique which reaches the sub-optical healing length scale. Influenced by magnetic resonance imaging, we name this technique magnetic resonance control. This technique uses time-varying coupling between internal spin states of a spinor Bose–Einstein condensate within a magnetic field gradient to address spatial regions of the condensate, enabling control over both the phase and the density of the condensate order parameter down to the healing length scale. Techniques already exist to engineer condensate wavefunctions, but not with such a fine degree of control. These techniques primarily rely on either the intensity variation of a laser beam, which limits the resolution to the diffraction limit (larger than the typically sub-optical healing length scale), or the adiabatic inversion of a magnetic trapping potential, which can not be easily changed to produce a variety of structures in the condensate wavefunction. To develop our magnetic resonance control technique, we simulate a spinor condensate in one dimension with time-dependent coupling between spin states and time-dependent external magnetic field gradients using the Gross–Pitaevskii equation. We show that magnetic resonance control can engineer a single black soliton using experimentally feasible parameters. A black soliton is an ideal target state to select for this demonstration because engineering such a state requires control over both the phase and density of the condensate with healing-length resolution. We demonstrate that magnetic resonance control can be extended to engineer more complicated target states by simulating the creation of multiple solitons in a condensate, with control over the initial positions and trajectories of the solitons. When magnetic resonance control is applied to Bose–Einstein condensates in the laboratory, it will be necessary to have an imaging system capable of resolving the fine structures created. As an alternative to high-cost, custom-manufactured lenses, and in-vacuum optical systems, I have designed and bench-tested an objective lens with a high numerical aperture (0.36) and a long working distance (35 mm) consisting of standard catalogue lenses. Using 780 nm light, suitable for imaging rubidium condensates, this objective can achieve a resolution of 1.3 μm across a diffraction-limited field of view of 360 μm through a 5 mm thick glass window of a science cell. By changing the spacing between the lens elements, this objective lens can compensate for the aberrations produced by a glass window up to 15 mm thick, and by changing the aperture size the objective becomes suitable for diffraction-limited monochromatic imaging on the D2 line of all the alkalis. Before performing proof-of-principle magnetic resonance control experiments on real Bose–Einstein condensates, we needed to construct an experimental apparatus capable of producing spinor Bose–Einstein condensates. In this thesis I summarise my main contributions to this group endeavour, including: constructing the ultra-high vacuum system; supervising the bakeout of our vacuum system; designing and aligning the optical systems to produce laser beams of different, tunable frequencies to trap, cool, and image rubidium gas; trapping a cloud of rubidium atoms in a magneto-optical trap; constructing laser beam shutters, photodetectors, and photodetector signal filters; and designing and constructing our side imaging and top imaging systems. Using our spinor Bose–Einstein condensate apparatus, I performed the first proof-of-principle magnetic resonance control experiments. With a pair of condensates side-by-side, separated by 200 μm, we can use magnetic resonance control to invert the spin of one condensate only, while leaving the other condensate unaffected. We anticipate magnetic resonance control being used in the laboratory to engineer the first black soliton in a Bose–Einstein condensate. Looking beyond solitons, magnetic resonance control could have applications to the field of magnon spintronics, and extending the technique to higher dimensions could enable the study of exotic topological defects such as spin knots in a quantum fluid.
Controlling quantum fluids at their fundamental length scale will yield superlative quantum simulators, precision sensors, and spintronic devices. This scale is typically below the optical diffraction limit, precluding precise wave-function engineering using optical potentials alone. We present a protocol to rapidly control the phase and density of a quantum fluid down to the healing length scale using strong time-dependent coupling between internal states of the fluid in a magnetic field gradient. We demonstrate this protocol by simulating the creation of a single stationary soliton and double soliton states in a Bose-Einstein condensate with control over the individual soliton positions and trajectories, using experimentally feasible parameters. Such states are yet to be realized experimentally, and are a path towards engineering soliton gases and exotic topological excitations.
Bose-Einstein condensates (BECs), a quantum state of matter formed when bosonic atoms are cooled close to absolute zero, have become the premier platform for investigating fundamental physics with atomic vapours. Experiments on Bose-Einstein condensates now achieve exquisite control over many aspects of the system, including interactions, trapping potential, and dynamics. This has precipitated a new wave of research into many-body quantum phenomena and, in particular, solitons. These structures are fundamental excitations of an interacting non-linear medium, of interest to a multitude of scientific disciplines from non-linear optics to financial markets. The highly controllable environment of BECs form an attractive playground for the study of solitons allowing the non-linearity to be dynamically tuned, facilitating deeper investigations into these structures. Consistently generating and analysing solitons in BEC experiments continues to be problematic. In particular, the non-linear dynamics of BECs, though required for the generation of solitons, produce particularly challenging control and optimization problems. These control problems must be solved before further investigations into the fundamental physics of soliton dynamics can be answered. This thesis makes three important advances in the control and measurement of BECs that will lead to better generation and observation of solitons. (1) a theoretical model for a control scheme capable of highly precise wavefunction engineering, (2) the experimental implementation of a machine learning algorithm for online optimisation, and (3) a continuous non-destructive imaging system capable of directly observing soliton dynamics in real-time. Together, these advances provide a suite of tools for manipulating and exploiting solitons in Bose-Einstein condensates. A novel technique was developed theoretically, offering control of the macroscopic wavefunction of a Bose-Einstein condensate with unprecedented spatial resolution and speed. The ability to control the atomic wavefunction at the fundamental length scale is key to the advancement of many quantum technologies such as quantum simulators. The magnetic resonance control scheme is demonstrated through simulation of a 87Rb condensate with the exemplar model generating a single dark soliton with corresponding π phase kink. The soliton represents a structure at the fundamental length scale of the system, and demonstrates the potential of the scheme for precision state engineering. The scheme is extended to generate higher-order soliton modes which are yet to be experimentally realised. A machine learning algorithm based on Gaussian processes was developed and implemented on the evaporative cooling stage of the production of a 87Rb Bose-Einstein condensate, successfully demonstrating fast optimisation to condensation. The Gaussian process develops a statistical model based on the data that enables the characterisation of the relationship between the experimental controls and resultant quality of the BEC. This relationship is often obfuscated through technical details of the apparatus, frustrating the use of theoretical models to design optimal evaporation ramps. These models often only consider ergodic dynamics with two-body s-wave interactions and no other loss rates with better ramps likely exploiting more complex interactions. The internal model generated from the Gaussian process utilised uncertainty in measured data, making the optimisation more robust to experimental noise than alternate methods. The algorithm is shown to produce high quality Bose-Einstein condensates in 10 times fewer experimental iterations than previously used online optimisation techniques. By exploiting information on the sensitivity of each control, the model can be used to aid experimental design. The convergence of the optimisation is further improved by eliminating a superfluous parameter identified by the model. The general usefulness of machine learning compared with bespoke optimisation algorithms has seen machine learning approach ubiquity. Finally, an experimentally straightforward technique for continuous non-destructive imaging of matter-wave solitons was developed and implemented, facilitating measurements of stochastic phenomena. The technique is readily practicable on any ultracold atom experiment with an existing absorption imaging system, simply requiring the probe laser be far-detuned from resonance. With a signal-to-noise of ∼ 33 at 1.25 GHz detuning, the technique is capable of producing 100 images with no observable heating or atom loss. Coupled with a fast optical phase locked loop, the technique can be used in conjunction with absorption imaging to generate a series of non-destructive images followed by a final high signal-to-noise absorption image solely through moving the laser on resonance for the final image. The high performance and utility of this imaging setup make it a powerful tool for ultra-cold atom experiments.
This thesis presents a novel magnetic resonance imaging (MRI) technique for spinor Bose-Einstein condensates (BEC) with the paramagnetic Faraday effect. This quantum-photon interface couples the spin of the BEC to the polarisation of an off-resonant laser, which in a magnetic field gradient allows a 1D density profile of the BEC to be reconstructed. As opposed to conventional imaging techniques that are destructive or diffraction limited, this technique is minimally destructive to the BEC and not diffraction limited. Multishot in situ imaging of a single condensate is therefore possible, paving the way for time-resolved studies and 2D/3D reconstructions. A theoretical model for Faraday imaging is developed using the tensor polarisability formalism for the atom-light interaction, and the signal-to-noise ratio is derived. Resolution limits induced by Stern-Gerlach separation are considered, encouraging rapid measurements with strong magnetic field gradients. Analysing the trapping potential reveals that the `magic-zero' wavelength 790nm produces no dipole force and enables a tightly focused probe laser to be used without perturbing the trap. Polarisation-maintaining fibers were seen to impart large fluctuations depending on the incident polarisation, which were minimised. The broad diode laser emission background is seen to reduce the lifetime of the BEC, which was corrected with an interference filter. Short-time Fourier transforms are used to process the photodetected signal, showing the evolution of the polarisation rotation frequency and amplitude. Frequency-modulation is observed at the power line frequency, corresponding to magnetic field fluctuations induced by nearby equipment. Birefringence of optical elements results in elliptical polarisation of the probe and an effective magnetic field, which is cancelled using a quarter-waveplate. The quadratic Zeeman effect causes rapid amplitude modulation, and is eliminated by applying a microwave dressing field. Radiative spin echo is achieved in a magnetic field gradient, proving that dephasing is coherent, and that multishot imaging possible. Magnetic resonance imaging (MRI) is performed on a partially-evaporated atom cloud, distinguishing the thermal and condensed fractions. A split dipole trap is used to create two spatially separated BECs, which are individually resolved during MRI. The resolution is only limited by the strength of the applied magnetic field gradient, and is not subject to the diffraction limit. Custom coils will enable MRI of condensates at the sub-micron scale.
This thesis describes the design, construction and optimisation of a new apparatus to produce Bose-Einstein condensates (BECs) of 87Rb atoms. The main aim in building this system was to include a high resolution imaging system capable of resolving single atoms. Optical access for the imaging system was created by including a stage of atom transport in which the atoms are magnetically transferred ~50 cm from a magneto-optical trap (MOT), where they are initially collected, to a glass science cell where experiments are carried out and imaging takes place. Two magnetic transport schemes have been demonstrated, based on approaches first used in other laboratories. First, a scheme in which the atoms are transferred in a moving pair of magnetic trapping coils. Second, a hybrid scheme where the atoms are translated part of the distance in the moving coils, and the rest of the way by switching the current in a chain of fixed coils. This second scheme was designed to allow optical access for a high numerical aperture microscope objective to be placed immediately next to the science cell for high resolution imaging.The atoms were first collected in a large pyramid MOT which can be loaded with 3 × 10^9 atoms in a time of 20 s. Around half of these atoms – those in the |F = 1, mF = −1> magnetic substate – were then magnetically trapped prior to transport. The typical fraction of the trapped atoms transferred to the science cell was ~30% and ~18% for the moving coils and hybrid schemes respectively.Evaporative cooling was carried out on the atom cloud following transport with the moving coils and loading into a time-orbiting potential trap. The optimised cooling sequence lasted for 28 s and consistently produced a pure condensate with 5 × 10^5 atoms. A BEC has also been produced by evaporative cooling following hybrid transport. The next experimental steps will be to optimise the hybrid transfer approach further and install the high resolution imaging system.The system is well-placed to continue an ongoing series of experiments in which ultracold atoms are trapped in RF-dressed potentials. These potentials will be used to study low-dimensional quantum gases as well as in experiments where small atom number BECs are rapidly rotated to enter the fractional quantum Hall regime.
Spinor Bose-Einstein condensates (BECs) are multi-component superfluids with magnetic interactions, formed from dilute gases of atoms in the Zeeman sublevels mF of a single hyperfine ground state F. Spinor BECs are inherently magnetic quantum objects, and exhibit a multitude of features such as spin domains, magnon excitations, and distinct magnetic phases. At the heart of the magnetic nature of a spinor BEC is the spin-dependent collisional interaction c, which determines the overall magnetic behaviour of the superfluid. In conjunction with the quadratic Zeeman shift q, the magnitude and sign of c give rise to an intricate magnetic phase diagram. The magnetic phase of a spinor BEC is tunable, and c and q are the knobs. In addition to realising quantum simulators of magnetism, spinor BECs are also precise magnetic sensors. Atomic magnetometry measures the Larmor precession frequency of atoms, which is proportional to the magnetic field strength. The sensitivity of an atomic magnetometer is principally dependent on the duration of the measurement (interrogation time) and number of atoms participating in the measurement. BECs are highly suitable as atomic magnetometers, allowing for long interrogation times and high atomic density, with microscopic sensor volumes. The latter is an important consideration, as measurement of small magnetic sources demand equally microscopic sensors. Established forms of magnetometry (such as warm vapour magnetometers) use large sensing volumes and are not trivially miniaturised. The exquisite sensitivity of a spinor BEC to magnetic fields becomes a liability when one studies low energy (~ 1 Hz) spinor physics over long evolution times (100 ms) in a typical laboratory environment. To first order, the spin-collisional interaction is independent of the magnetic field. However, spatially inhomogeneous magnetic fields are problematic, exerting mechanical forces on the magnetically sensitive spin components, suppressing spin collisions and possibly introducing relaxation mechanisms that confound the observation of intrinsic equilibration. Precise characterisation of magnetic field gradients and their cancellation is thus of paramount importance. In this thesis, the interaction of a spinor BEC with its environment is studied. The construction, operation and optimisation of a BEC machine is described in detail, as well as the quantum state preparation and control techniques used. Spin-mixing collisions were studied by observing population oscillations, the characteristic hallmark of such coherent phenomena. The observed population dynamics showed strong evidence of the deleterious effects of magnetic field gradients. A series of experiments were performed using spin-echo pulse sequences to decouple the evolving spinor condensate from the inhomogeneous magnetic environment with encouraging results, but which posed further questions. The fidelity of spin-echo pulses in the strong radiative coupling regime was also studied. Spinor condensates were then employed as sensitive magnetometers to characterise the magnetic landscape they inhabit. A new method of measuring magnetic field gradients was developed, using a pair of spinor BECs separated in space that are simultaneously addressed by a Ramsey interferometry sequence. Combined with a re-orientable magnetic bias field, the resulting gradiometer was used to measure the magnetic field gradient tensor in vacuo. This scheme can be extended to realise a high-precision vector co-magnetometer, with spatiotemporal sensitivity comparable to established forms of magnetometry. Vector light shifts (VLSs) are optically-induced energy shifts that result in effective magnetic fields, which in our case originate from the laser beams used to trap the atoms. The VLS vanishes for purely linearly polarised light, which is experimentally difficult to obtain in vacuo. Small elliptical polarisation imperfections originating from birefringent optics and the glass vacuum cell result in a non-zero shift that contaminates magnetometry measurements. The VLS is spatially varying due to the intensity gradient of the trapping light. Due to gravitational sag of the trapping potential, the spatial variation of the VLS is accentuated, resulting in an effective magnetic field gradient that contributes to dephasing and component separation. The simultaneous Ramsey interferometry experiments are then adapted to measure the VLS-induced Zeeman shift and diagnose its elimination. Up to 99.96% of the VLS is eliminated using this method, and extensions to the technique are described.
The possibility to mould and control pure quantum systems has been offered by the experimental observation of Bose-Einstein condensation, a unique phase of matter when macroscopic quantities of a gas occupy the lowest quantum state. Techniques for creating these degenerate gases vary from laboratory to laboratory; each offers an unique test bed for studying quantum physics on a macroscopic scale. This thesis reports on the experimental design, construction and performance of an apparatus to create two-component 87Rb and 41K condensates for studies of non-equilibrium dynamics. This thesis is broken down as follows. In Chapter 1, a brief overview of the history and status of Bose Einstein condensates is presented, to affirm the motivation behind our work. Chapter 2 then presents the fundamental theory and background information to understand how and why our experiment was built. Chapter 3 describes the bulk of the work undertaken at the beginning of this thesis. In particular it describes the design choices, construction and performance of the vacuum, and laser and magnetic coil systems that are the key structural elements of the apparatus. In particular the vacuum system is a two-component differentially pumped system designed to optimise the number and lifetime of trapped atoms. Another integral element of the vacuum chamber is the science cell that enables high optical access, and close physical, access to an atomic cloud. As a result, high resolution imaging, ≃ 980 nm resolution at 780 nm is expected with a commercial microscope objective lens. The laser system is carefully de- signed to best combine and deliver the seven different optical frequencies required to simultaneously trap and manipulate 87Rb and 41K atoms. The magnetic coil system also represents an integral component of the apparatus, responsible for trapping and transferring atoms in a quadrupole field. One pair of coils is able to have their current direction reversed in order to generate bias fields. This allows access to Feshbach resonances between the two species, once they have been condensed. The second part of this thesis, Chapter 4, describes the performance of the apparatus when used to simultaneously trap 41K and 87Rb in the 3D magneto- optical trap (MOT) and produce a condensate of 87Rb atoms. In particular 1 × 109 rubidium atoms are routinely trapped in the 3D-MOT and transferred to the science chamber. In the hybrid trap the atoms are evaporatively cooled via microwave radiation, later to be used to sympathetically cool 41K, and loaded into a hybrid optical dipole and magnetic trap. At this stage ≃ 5 × 106 atoms are evaporatively cooled, by lowering the trap depth of the optical dipole beam, until a condensate of 87Rb is formed, containing 1.5 × 105 atoms. Additionally we are able to produce 3D MOTs of 41K simultaneously with the 87Rb 3D MOTs.