Atom chips provide a flexible and scalable platform for applications of cold atoms and Bose-Einstein condensates (BECs) in future quantum technologies including miniaturised atomic clocks, quantum information processing and quantum sensors. Our project studies the coherent properties of a two-component BEC prepared on a magnetic film atom chip using a two-photon microwave/radiofrequency transition. At a particular value 3.23 G of the magnetic field two hyperfine states |F =1, mF = −1 > and |F = 2, mF = 1 > in 87Rb exhibit a low differential Zeeman shift (431 Hz/G2) and can be trapped simultaneously with long dephasing times. The coherent superposition of these two states has a long coherence time and is attractive for applications in quantum atom devices.
Atom chips provide a flexible and scalable platform for applications of cold atoms and Bose-Einstein condensates (BECs) in future quantum technologies including miniaturised atomic clocks, quantum information processing and quantum sensors. Our project studies the coherent properties of a two-component BEC prepared on a magnetic film atom chip using a two-photon microwave/radiofrequency transition. At a particular value 3.23 G of the magnetic field two hyperfine states |F =1, mF = −1 > and |F = 2, mF = 1 > in 87Rb exhibit a low differential Zeeman shift (431 Hz/G2) and can be trapped simultaneously with long dephasing times. The coherent superposition of these two states has a long coherence time and is attractive for applications in quantum atom devices.
We report on progress towards experiments studying the coherence of a two-component Bose-Einstein condensate using a magnetic film atom chip. Our atom chip provides extremely low heating rates and offers advantageous conditions for increased coherence times. We currently trap 4 × 10787Rb atoms in the |F= 1, mF = −1> ground state and use forced evaporative cooling towards the degeneracy. The Ramsey method of separated fields will be used to interrogate the evolution of a Bose-Einstein condensate in a superposition of the |1, −1> and |2, 1> states.
Ultracold atoms and Bose-Einstein condensates represent a burgeoning and ever exciting area of research which has produced numerous breakthroughs in the last 15 years. Ultracold atoms at nano-Kelvin temperatures, isolated in atom traps from the hot, room temperature environment, exhibit ultimate quantum properties and can serve as pure samples for testing quantum mechanics and performing precision measurements. The research described in this thesis investigates, both theoretically and experimentally, the coherence and dynamical evolution of a two-component Bose-Einstein condensate (BEC). The studies are performed on the |1 › ≡ F = 1,mF = −1 › and |2 › ≡ |F = 2,mF = +1 › hyperfine ground states of a 87Rb BEC trapped on an atom chip. Conventional understanding has been that the fringe contrast in interferometric experiments on a BEC is limited by the mean-field dephasing due to strong interatomic interactions which lead to inhomogeneous collisional shifts. We have discovered using Ramsey interferometry a mean-field driven self-rephasing effect in a trapped twocomponent BEC. When combined with a spin-echo technique, we find that the selfrephasing leads to a coherence time of 2.8 s, the longest ever recorded for an interacting BEC. Secondly, we have developed a new technique based on periodic collective oscillations for precision measurements of the interspecies and intraspecies scattering lengths and derived an analytic mean-field theory for the phase and density dynamics in the twocomponent BEC. This technique has been applied to the measurements of the scattering lengths for the two components, |1i and |2i, in 87Rb with a precision of 0.016%. Additionally, the two-body loss coefficients for these states have been measured. Thirdly, we have developed and applied a new, interferometric method for calibrating the detection system, in order to determine the total atom number precisely for Ramsey interferometric measurements. The calibration coefficient is found to be in a good agreement with that obtained using a conventional calibration technique based on the critical BEC temperature. Finally, we have detected predicted RF-induced Feshbach resonances by monitoring changes in the two-body loss coefficients in the two-component BEC. Conventional Feshbach resonances have been used in many works to tune the s-wave interactions; however, there is no achievable magnetic Feshbach resonance for magnetically trappable states of 87Rb. The RF-induced resonances which we have detected can provide a way to tune the scattering lengths which is important for interferometry and entanglement experiments. The positions of the resonances may provide useful information for predicting the atomic scattering properties of 87Rb.
Atom chips are devices used to magnetically trap and manipulate ultracold atoms and Bose-Einstein condensates near a surface. In particular, permanent magnetic film atom chips can allow very tight confinement and intricate magnetic field designs while circumventing technical current noise. Research described in this thesis is focused on the development of a magnetic film atom chip, the production of Bose-Einstein condensates near the film surface, the characterisation of the associated magnetic potentials using rf spectroscopy of ultracold atoms and the realisation of a precision sensor based on splitting Bose-Einstein condensates in a double-well potential. The atom chip itself combines the edge of a perpendicularly magnetised GdTbFeCo film with a machined silver wire structure. A mirror magneto-optical trap collects up to 5 x 108 87Rb atoms beneath the chip surface. The current-carrying wires are then used to transfer the cloud of atoms to the magnetic film microtrap and radio frequency evaporative cooling is applied to produce Bose-Einstein condensates consisting of 1 x 105 atoms. We have identified small spatial magnetic field variations near the film surface that fragment the ultracold atom cloud. These variations originate from inhomogeneity in the film magnetisation and are characterised using a novel technique based on spatially resolved radio frequency spectroscopy of the atoms to map the magnetic field landscape over a large area. The observations agree with an analytic model for the spatial decay of random magnetic fields from the film surface. Bose-Einstein condensates in our unique potential landscape have been used as a precision sensor for potential gradients. We transfer the atoms to the central region of the chip which produces a double-well potential. A single BEC is formed far from the surface and is then dynamically split in two by moving the trap closer to the surface. After splitting, the population of atoms in each well is extremely sensitive to the asymmetry of the potential and can be used to sense tiny magnetic field gradients or changes in gravity on a small spatial scale.
Ultra-cold neutral atoms and Bose-Einstein condensates (BECs) offer fascinating possibilities for quantum information processing. Long decoherence times, easy qubit addressing and the potential for scalability are some advantages. Storing and manipulating atoms in magnetic microtraps or 'Atom Chips' provides a promising base on which quantum logic operations could be performed. A primary source of decoherence in atom chip experiments arises from the interaction of the micro-Kelvin atoms with the room temperature chip surface. We have experimentally investigated these interactions and observe two interesting effects. Firstly, ultra-cold clouds are seen to fragment into clumps when brought within 100 microns of the chip surface. Secondly, the lifetime of atom clouds is seen to drop off near the surface due to spin flip transitions. These are caused by radio-frequency thermal fluctuations of the magnetic field as predicted but not previously observed.
The controlled generation of entanglement forms the basis for currently emerging ‘quantum technologies’, such as quantum simulation, computation, and metrology. In the field of quantum metrology, multi-particle entangled states, such as spin-squeezed states, are investigated as a means to improve measurement precision beyond the ‘standard quantum limit’. This limit arises from the quantum noise inherent in measurements on a finite number of uncorrelated particles and limits today’s best atomic clocks. Atom chips combine exquisite coherent control of ultracold atoms with a compact and robust setup, suggesting their use for quantum metrology with portable atomic clocks and interferometers. A severe limitation of atom chips, however, is that techniques to control atomic interactions and to generate entanglement have not been experimentally available so far. In this thesis, I present experiments where we generate for the first time multi-particle entanglement on an atom chip. We achieve this by controlling elastic collisional interactions with a state-dependent potential. We employ this novel technique to generate spin-squeezed states of a two-component Bose-Einstein condensate and show that they are a useful resource for quantum metrology, as they could be used to improve an interferometric measurement by 2.5 dB over the standard quantum limit. The state-dependent potential is created with the help of a coplanar microwave guide, which is integrated on our atom chip. In the vicinity of this waveguide a microwave nearfield is formed. When a Bose-Einstein condensate of 87Rb is brought into this near-field, the hyperfine energy levels of the atoms are shifted differentially due to the AC Zeeman effect. The strong gradients in the field can be used to state-selectively shift the minimum of a static magnetic atom trap and thus coherently split an ensemble of atoms which have been prepared in a superposition of two internal states. During this process, nonlinear atomic interactions lead to the formation of a spinsqueezed state. I tomographically analyze the produced state, reconstruct its Wigner function, and deduce that it is at least four-particle entangled. I compare our results with a dynamical multi-mode simulation which takes not only the atomic motion and internal state dynamics but also particle losses into account and find good agreement. Moreover, I use this comparison to identify technical noise sources in our experiment, which currently limit the achieved amount of squeezing, and make suggestions on how to eliminate them in future experiments. Our method can in principle create a very large amount of squeezing and entanglement and is applicable to a wide variety of atomic systems, in particular to those for which no convenient Feshbach resonance exists. We envisage the implementation of this technique in portable atomic clocks and interferometers operating beyond the standard quantum limit. Furthermore, it is a valuable tool for experiments on many-body quantum physics and could enable quantum information processing on atom chips.
In this thesis I discuss work that has been carried out on the dynamics of a Bose Einstein condensate of Rb 87 produced near an atom chip. A Bose Einstein Condensate (BEC) is a quantum state of matter where a single quantum state becomes occupied by a macroscopic number of identical Bosons. In our case this is achieved by cooling a system of trapped identical rubidium 87 atoms to its ground state. To reach temperatures of condensation we initially laser cool atoms from room temperature, before loading them into a magnetic trap. The magnetic trap is produced through a combination of uniform magnetic fields from coils outside our vacuum chamber and currents running through wires on an atom chip. The atom chip is a microfabricated device, produced by a coating a silicon chip with a thin layer of gold and etching wires into it. Together, these fields create a magnetic field minimum 120μm from the surface of the chip which can be used to confine low field seeking hyperfine states of the atom in an elongated harmonic trap. Once the atoms are confined in the magnetic trap we used force evaporative cooling out to reach the phase space densities required for Bose Einstein condensation. The BEC is used to investigate the relative dynamics between the fraction of the atoms in the condensate to those not in the condensate. Our atom chip provided a suitable environment to investigate this due to fragmentation of the magnetic potential close to the chip. Small imperfections in the wires on our atom chip mean that the trapping potential isn’t smooth. Small regions of higher trapping frequency - or fragments - are formed. Due to the small size of these fragments it is possible to find a position where a condensate can form in the fragment, and see a potential of high frequency, whereas a non-condensed atom will see a lower frequency potential. We exploit this to set the condensed fraction moving relative to the non condensed part and investigate the subsequent damping of their motion relative to each other.