Preview

PDF, English Download (24MB) | Terms of use

Abstract

The Germanium Detector Array (Gerda) experiment, located underground at the INFN Laboratori Nazionali del Gran Sasso (LNGS) in Italy, deploys high-purity germanium detectors to search for the neutrinoless double beta-decay (0vbb) of 76Ge. An observation of this lepton number violating process, which is expected by many extensions of the Standard Model, would not only generate a fundamental shift in our understanding of particle physics, but also unambiguously prove the neutrino to have a non-vanishing Majorana mass component. A first phase of data recording lasted from November 2011 to May 2013 - resulting in a total exposure (defined as the product of detector mass and measurement time) of 21.6 kg*yr. Within this thesis a thorough study of this data with special emphasis on the development and scrutiny of an active background suppression technique by means of a signal shape analysis has been performed. Among several investigated multivariate approaches, particularly a selection algorithm based on an artificial neural network is found to yield the best performance; i.a. the background index close to the Q-value of the 0vbb- decay could be suppressed by 45% to 1*10^(-2) cts/(keV*kg*yr), while still retaining a considerably high signal survival fraction of (83+/-3)% leading to a significant improvement of the experimental sensitivity. The efficiency is derived by a simulation and further validated by substantiated consistency checks availing themselves of measurements taken with different calibration sources and physics data. No signal is observed and a new lower limit of T(90%C.L.)>2.2*10^25 yr for the half-life of neutrinoless double beta-decay of 76Ge is established.