UCLA

Direct dark matter detectors are currently probing the favored supersymmetric theoretical phase space for weakly interacting massive particles (WIMPs) as part of a larger electroweak sector particle search. The XENON100 detector has recently improved upon their world-best upper limits on WIMP-nucleon scattering cross sections from 2011 with new results presented in 2012, which have further ruled out much of the predicted regions for dark matter. In the low mass regime, the results shown so far have been conservative in addressing the claims of a WIMP detection by CoGeNT, DAMA and CRESSTII.

This thesis discusses a different approach to analyzing the XENON100 data with a profile likelihood statistical method using the ionization channel to improve both energy reconstruction and energy resolution and probe the low WIMP mass region. A Monte Carlo simulation using a combination of detector geometry and underlying statistical features of signal production has been developed to determine the ionization yield, which has only a few direct measurements, and to model WIMP interactions as input to the statistical technique. The resulting profile likelihood analysis, which includes systematic uncertainties in the energy scales and background and signal models, has been able to improve the current upper limits by a factor of 10 in the low mass region (6-10 GeV/c²) and about a factor of 2 up to 50 GeV/c². The discovery potential of the dataset is also studied, which has produced a 2.3σ significance for a 7.5 GeV/c² particle at a cross section of σ_χ-N=1.8×10^-43 cm² with a 95% confidence interval of [1.74×10^-44,7.76×10^-43] cm².