| Summary: | An impressive array of astrophysical observations suggest that 83% of the matter in the universe is in a form of non-luminous, cold, collisionless, non-baryonic dark matter. Several extensions of the Standard Model of particle physics aimed at solving the hierarchy problem predict stable weakly interacting massive particles (WIMPs) that could naturally have the right cosmological relic abundance today to compose most of the dark matter if their interactions with normal matter are on the order of a weak scale cross section. These candidates also have the added benefit that their properties and interaction rates can be computed in a well defined particle physics model. A considerable experimental effort is currently under way to uncover the nature of dark matter. One method of detecting WIMP dark matter is to look for its interactions in terrestrial detectors where it is expected to scatter off nuclei. In 2007, the XENON10 experiment took the lead over the most sensitive direct detection dark matter search in operation, the CDMS II experiment, by probing spin-independent WIMP-nucleon interaction cross sections down to σχN ~ 5 × 10-44 cm2 at 30GeV/c2. Liquefied noble gas detectors are now among the technologies at the forefront of direct detection experiments. Liquid xenon (LXe), in particular, is a well suited target for WIMP direct detection. It is easily scalable to larger target masses, allows discrimination between nuclear recoils and electronic recoils, and has an excellent stopping power to shield against external backgrounds. A particle losing energy in LXe creates both ionization electrons and scintillation light. In a dual-phase LXe time projection chamber (TPC) the ionization electrons are drifted and extracted into the gas phase where they are accelerated to amplify the charge signal into a proportional scintillation signal. These two signals allow the three-dimensional localization of events with millimeter precision and the ability to fiducialize the target volume, yielding an inner core with a very low background. Additionally, the ratio of ionization and scintillation can be used to discriminate between nuclear recoils, from neutrons or WIMPs, and electronic recoils, from γ or β backgrounds. In these detectors, the energy scale is based on the scintillation signal of nuclear recoils and consequently the precise knowledge of the scintillation efficiency of nuclear recoils in LXe is of prime importance. Inspired by the success of the XENON10 experiment, the XENON collaboration designed and built a new, ten times larger, with a one hundred times lower background, LXe TPC to search for dark matter. It is currently the most sensitive direct detection experiment in operation. In order to shed light on the response of LXe to low energy nuclear recoils a new single phase detector designed specifically for the measurement of the scintillation efficiency of nuclear recoils was also built. In 2011, the XENON100 dark matter results from 100 live days set the most stringent limit on the spin-independent WIMP-nucleon interaction cross section over a wide range of masses, down to σχN ~ 7 x 10-45 cm2 at 50GeV/c2, almost an order of magnitude improvement over XENON10 in less than four years. This thesis describes the research conducted in the context of the XENON100 dark matter search experiment. I describe the initial simulation results and ideas that influenced the design of the XENON100 detector, the construction and assembly steps that lead into its concrete realization, the detector and its subsystems, a subset of the calibration results of the detector, and finally dark matter exclusion limits. I also describe in detail the new improved measurement of the important quantity for the interpretation of results from LXe dark matter searches, the scintillation efficiency of low-energy nuclear recoils in LXe.
|
|---|
