Pile-up event identification and rejection in SNO+: enhancing the signal to noise ratio

This thesis has been developed in the context of the SNO+ experiment, a multipurpose neutrino detector whose goals include the search of the neutrinoless double beta decay, low energy solar neutrinos, supernova neutrinos and geological and reactor anti-neutrinos. To achieve these goals SNO+ requires a precise modelling of its backgrounds. Several techniques are employed to guaranty the purity of the samples and materials, as well as various methods for the identification and rejection of single event backgrounds.

Nevertheless, this is not a complete picture of all the backgrounds present in the experiment. The main goal is the study of one of these classes of backgrounds particularly difficult to identify: the pile-up events. These are events that happen in the same trigger window (460 ns) and are therefore reconstructed as a single event. As a consequence, low energy events can be shifted into the region of interest and/or reconstructed in inner volumes, impacting many analyses.

However, due to their nature, pile-up event's characteristics can differ from single events making the identification techniques, developed for those type of events, ineffective. Therefore, new identification and rejection techniques for pile-up events are required. The strategy developed and presented in this thesis will consist of three main steps (1) study of the type and amount of pile-up events, (2) identification of the variables that characterize these events, (3) comparison between Monte-Carlo simulation and data for two main phases of SNO+.