A Study of Majorana Demonstrator Backgrounds with Bayesian Statistical Modeling
The Majorana Demonstrator was a neutrinoless double beta decay (0νββ) experiment consisting of around 30 kg of germanium detectors enriched to 88% in 76Ge and around 14 kg of natural germanium detectors. The detectors were divided between two cryostats and surrounded by a graded passive shield. The Demonstrator concluded in March 2021 and set a 0νββ half-life limit of T1/2 = 8.3×1025 years (90% C.L.) based on its full exposure of 64.5 kg·yr. The experiment achieved one of the lowest background rates in the region of the 0νββ Q-value, (6.23 ± 0.55) cnts/(keV·kg·yr). This background rate, however, was higher than the rate of (1.16 ± 0.04) cnts/(keV·kg·yr) projected by material assays and simulations. This discrepancy arises from an excess of events from the 232Th decay chain. Background model fits presented in this work aim to understand the observed 232Th excess and other deviations from assay-based projections, as well as allow a precision measurement of the 2νββ half-life. Comparisons of the data with simulations indicate the 232Th excess cannot arise from near-detector components, and likely originates from a component of the cryogenic system. This is an important finding related to the design and implementation of the LEGEND-200 experiment. The final results of the Demonstrator are presented along with this study of its backgrounds using Bayesian statistical inference methods. The efficacy of these inference methods is limited by the statistics of the Demonstrator dataset. However, these methods remain useful for supplementary background studies, and support the analysis to determine the location of the 232Th excess.
University of North Carolina at Chapel Hill