A Moving-Trap Zeeman Decelerator

Abstract

This thesis reports on the design, construction and implementation of a moving-trap Zeeman decelerator which uses 3D magnetic traps to guide and decelerate paramagnetic particles from a supersonic beam. The decelera- tor will ultimately be used in a quantum simulator device, where a system of strongly-interacting quantum particles, where their interactions can be tuned, is formed using polar molecules. The decelerator is a potential load- ing stage for a molecular magneto-optical trap (MOT), where high densities of molecules can be cooled down to the sub milliKelvin temperatures. The molecules can then be further cooled sympathetically with laser-cooled atoms into the microKelvin regime, allowing their trapping in an optical lattice.
This thesis mainly describes the design, construction, and implementation of the decelerator. The technicalities of the beam machine, the decelerator coils, and the driving electronics are described in detail. A homebuilt fast- ionisation gauge (FIG) detector allows the characterisation and optimisation of pulsed beams produced by a cryogenically cooled pulsed valve. We produce supersonic beams of metastable argon atoms (Ar*), made by the electronic excitation of the atoms using a homebuilt pulsed electric discharge assembly, stabilised by a hot filament. This enables the discharge to operate more stably at voltages as low as 400 V and at discharge pulse durations as short as 20 μs, which combine to create a cold packet of Ar* atoms. An optimised slow beam of Ar* with a measured velocity of 306±8 m/s and a translational temperature of 4 K is formed. The decelerator has a detection chamber that allows different means of detecting atoms and molecules: michrochannel- plate (MCP), single-pass laser-induced fluorescence (LIF), cavity-enhanced laser-induced fluorescence (CELIF), and a quadrupole mass spectrometer (QMS) with the ability to photo-ionise. We demonstrate that the CELIF detection technique, which combines a cavity ring-down (CRD) setup and a LIF setup, using a standard UV pulsed dye laser, can be an effective detection method for molecules with fluorescence lifetimes on the order of hundreds of nanoseconds. Using CELIF, we measure the absolute density of SD radicals in a pulsed supersonic jet down to the limit-of-detection of 105 cm−3. In the 0.002 cm3 probe volume, this corresponds to ca. 200 molecules, and the quantum-noise-limited absorption coefficient is αmin = 7.9 × 10−11 cm−1 in 200 s of acquisition time.

The biggest advantage of this type of decelerator is in the fact that the para- magnetic atoms are confined in all three dimensions continuously throughout the length of the decelerator. We present our proof-of-principle experimen- tal results where we demonstrate, using a single deceleration stage with a length of 123mm, the manipulation of Ar* atoms in the 3P2 metastable state using 3D magnetic fields, and using continuously modulated magnetic fields which produce a travelling potential. It is successfully shown that the Ar* signal intensity is greatly increased, nearly by a factor two by using a 290mm long quadrupole magnetic guide which provides transverse confine- ment of the atoms. With the addition of the decelerator coils, magnetic confinement along the longitudinal beam axis is achieved, forming 3D-traps. The 3D-guiding of the low-field-seeking states of 3P2 state of Ar* is carried out at constant velocities ranging from 320 m/s up to 400 m/s along a single decelerator module. The longitudinal temperatures were ∼500mK. While attempting the deceleration of the traps, though no real deceleration was observed for this short decelerator length, the fields did show a manipulation effect. This gives us the confidence that with a longer decelerator, we will see very prominent bunching.
The work presented in this thesis is a major step forward in the demon- stration of an efficient Zeeman decelerator which can bring large numbers of molecules to low velocities. It will be an ideal loading step for a molecular MOT where high densities of molecules can be cooled down to the sub mil- liKelvin temperatures. A beam of CaF molecules from a buffer gas source, starting at an initial velocity of 150m/s, could be decelerate to a standstill using a 1 m long decelerator. This would only require eight decelerator mod- ules. Aside from its use as a loading step for a molecular MOT, for building a quantum simulator device, this new type of decelerator can be used for various applications. One of these is in cold chemistry. The methyl radi- cal is one of the most important and fundamental intermediates in chemical reactions. With regards to magnetic deceleration, the methyl radical has a similar magnetic-moment-to-mass ratio to argon, so with an appropriate choice of a seed gas we should be able reproduce the results we have so far with argon and demonstrate deceleration of our first molecule.