<p>Due to the rapid development of nanophotonics, microring resonators suspended on a membrane holds promises for a scalable optical circuit with strong light-atom interaction. In this dissertation, I introduce a efficiently-coupled microring circuits for on-chip cavity QED with cold atoms and report my experimental efforts to integrate the optical chip into a ultrahigh-vacuum chamber with a magneto-optical trap for Rb atoms. My attempts to load single atoms into optical tweezers are also discussed.</p> <p>  </p> <p>  Although the loading of atom into optical tweezers above the top surface of resonator remains a challenge in experiment, I propose an alternative of cavity cooling based on cavity QED to facilitate the loading of atom into a two-color evanescent field trap around the waveguide. Assuming that the strong interaction between atoms and resonator modes is realized, I theoretically investigate the synthesis via photoassociation and the direct optical detection of a single ground-state cold molecule, whose corresponding excited-state has multiple decay channels. Similarly to the Purcell effect, the decay in a specific decay channel could be enhanced based on cavity QED, and therefore the synthesis efficiency can approach unity when the interaction between the resonator modes and a single cold molecule becomes stronger. In addition, for a single cold molecule without closed optical transition, the electromagnetically induced transparency is possible to be observed on our nanophotonic platform in the case of strong resonator-molecule coupling.</p>

This thesis describes the development of a new experimental technique for studying tunable-collision-energy, quantum state-selected, low-temperature ion-molecule reactions. This has been achieved through the combination of a Stark decelerator for neutral dipolar molecules, and a linear Paul ion trap. The Stark deceleration process for ND₃ was examined in detail, through the analysis of experimental data in combination with newly written molecular dynamics simulation programs. In order to prepare a sample of molecules appropriate for collision studies, additional beamline components were introduced after the decelerator. These components were: two hexapoles, to provide transverse focussing, maximising the molecular density; a molecular buncher, providing increased longitudinal velocity resolution; and a fast-opening shutter, to separate decelerated molecules from undecelerated molecules. The sympathetic-cooling of Xe⁺ ions and ND⁺₃ ions by laser-cooled, Coulomb crystallised ⁴⁰Ca⁺ ions with the ion trap was also studied. In particular, the stable trapping of Xe⁺ was demonstrated for the first time, and the experimental developments that led to this are discussed. The work in this thesis represents significant progress towards studying the reaction of tunable-energy ND₃ in the |j,mk> = |1,−1> quantum state with cold Xe⁺ ions. Ion-molecule reactions utilising ND₃ molecules electrostatically guided through the Stark decelerator were performed. It was observed that the main source of error in these experiments was in the calculation of the initial number of Xe⁺ ions that had been sympathetically cooled into the Coulomb crystal. The sensitivity of the crystal morphology to the number of Xe⁺ ions was evaluated using molecular dynamics simulations. Strategies have been developed to reduce this uncertainty in future studies. In addition to experimental work, the theory of low temperature ion-molecule reactions has been developed further. The temperature at which classical and quantum mechanical calculations diverge due to purely statistical effects has been investigated using different model intermolecular potentials, for closed-shell and open-shell species, and in the ground and rotationally excited states. From the results of these calculations, several promising candidate reactions have been suggested that might exhibit statistical quantum behaviour at experimentally achievable temperatures.