911³Ô¹Ï

Synopsis

Event reconstruction and background discrimination are studied for the ARGO liquid-argon (LAr) dark matter experiment, a proposed 400-tonne single-phase detector instrumented with tens of thousands of silicon photomultipliers (SiPMs). The study is performed using Monte Carlo (MC) simulations based on the packages GEANT4 and RAT, which model particle interactions and detector response. Within this framework, pulse shape discrimination (PSD) between electron and nuclear recoils is optimized using the prompt light fraction, and the resulting leakage of beta backgrounds into an energy region similar to that expected from Weakly Interacting Massive Particle (WIMP) interactions is evaluated. In parallel, attention-based machine-learning techniques are developed to reconstruct interaction positions from SiPM charge patterns and to define fiducial volumes that reduce surface and external backgrounds. Radiogenic neutrons from the AV are also studied, and their contribution to the expected ARGO background rates is estimated. Together, these studies address key aspects of event reconstruction and background suppression in large LAr dark matter detectors. These studies enable the assessment of ARGO’s sensitivity to WIMP interactions.

For Zoom link info, please contact Lindiwe Coyne at physgrad@sfu.ca.