High resolution field imaging with atomic vapor cells

In this thesis, I report on the development of imaging techniques in atomic vapor cells. This is a relatively unexplored area, despite the ubiquitous use of imaging in experiments with ultracold atoms. Our main focus is in high resolution imaging of microwave near fields, for which there is currently no satisfactory established technique. We detect microwave fields through Rabi oscillations driven by the microwave on atomic hyperfine transitions. The technique can be easily modified to also image dc magnetic fields. In addition, we have developed techniques to image vapor cell processes such as atomic T1 and T2 relaxation. These provide a new window into vapor cell physics, which we have used to obtain spatially resolved information on Rb interactions with the cell walls, and to estimate the Rb relaxation probability in a collision with the cell wall.
As a first application of our imaging techniques, we imaged the dc and microwave magnetic fields inside a state-of-the-art vapor cell atomic clock. This new clock characterisation technique should lead to real improvements in clock performance, and is in the process of being adopted by the atomic clock community.
We have developed a widefield, high resolution imaging setup using a microfabricated vapor cell, which we have used to image microwave and dc magnetic vector fields. With the addition of a 480 nm laser, the setup can be configured to image microwave electric fields. Our camera-based imaging system records 2D images with a 6x6 mm2 field of view at a rate of 10 Hz. It provides up to 50 um spatial resolution, and allows imaging of fields as close as 150 um above structures, through the use of extremely thin external cell walls. This is crucial in allowing us to take practical advantage of the high spatial resolution, as feature sizes in near-fields are on the order of the distance from their source, and represents an order of magnitude improvement in surface-feature resolution compared to previous vapor cell experiments. We demonstrate a microwave magnetic field sensitivity of 1.4 uT/sqrt-Hz per 50x50x140 um3 voxel, at present limited by the speed of our imaging system. Since we image 120x120 voxels in parallel, a single scanned sensor would require a sensitivity of at least 12 nT/sqrt-Hz to produce images with the same sensitivity.
The spatial resolution, distance of approach, and sensitivity of our high resolution setup are sufficient for characterising 6.8 GHz microwave fields above a range of real world devices. However, frequency tunability is essential for wider applications of our imaging technique. Industry is particularly interested in techniques for imaging high frequency microwaves, above 18 GHz, where simulations become increasingly unreliable. I have shown that our technique can be extended to image microwaves of any frequency, in principle from dc to 100s of GHz, by using a large dc magnetic field to Zeeman shift the hyperfine ground state transitions to the desired frequency. I present results from a proof-of-principle setup, where we have used a 0.8 T solenoid to detect and image microwaves from 2.3 GHz to 26.4 GHz.