GPU parallelization of spin-tracking simulations for the SNS nEDM experiment

The neutron's electric dipole moment (nEDM) is measured in the Spallation Neutron Source (SNS) nEDM experiment by detecting the spin-dependent capture events of polarized $^3$He and ultracold neutrons in a measurement cell with parallel magnetic and electric fields. Simulations tracking particles' spin have been performed to better understand the systematic effects present in the experiment. However, spin-tracking on CPUs can be slow and computationally expensive. An additional constraint is maintaining high accuracy for long durations with accumulating rounding errors. GPU parallelization can be used to track many particles simultaneously and improve solver efficiency. We will present an overview of the approach and its advantages and limitations, as well as preliminary results of systematic studies from our simulation to track the spin of particles in the measurement cell.