Classification of Induction Signals for the EXO-200 Double Beta Decay Experiment

THE EXO-200 experiment searches for double beta decay in Xenon-136, in both the 2-neutrino and 0-neutrino modes. The EXO-200 detector has a set of wires that collect all charge released within the detector volume. Aside from the double beta decay events that occur within the detector, there are other types of background events that can release charge, such as compton scattering. Some of these types of events are single-site, whereas others are multi-site. Being able to classify single-site from multi-site events is important for accurately fitting backgrounds which obscure the double beta decay signal. Multi-site events are simply classified as events where multiple charge clusters are collected by the wires (with certain timing constraints). An issue is, however, that there are some instances where a wire collects charge from a true single-site event, but the closely packed adjacent wires may also show an induction signal, which can be mistaken as an additional charge collection in Event Reconstruction, and can lead to mistakenly classifying the event as multi-site. Removing these induction signals during the Event Reconstruction would improve the single-site/multi-site fraction by ∼ 10% and the total reconstruction efficiency by ∼ 30%, two critical improvements for increasing our sensitivity to detecting 0neutrino double beta decay for the experiment. The ultimate goal is to find a model that can be trained in one sitting and then can be incorporated into Event Reconstruction to classify all signals in the experimental data.