PIC-MCC simulations of a GEC reactor using WarpX on GPUs

ORAL

Abstract

Detailed investigations of plasma physics in low-temperature plasma reactors requires fully-kinetic simulations. We will leverage WarpX to perform PIC-MCC simulations for a GEC reference cell configuration. WarpX was created under the Exascale Computing Project and won the 2022 Gordon Bell Prize in recognition of the outstanding accomplishment in HPC. It is performance portable on heterogeneous architectures. Although WarpX was developed primarily to model plasma-based particle accelerators, it has been generalized to solve a wide range of applications, including LTP regimes. Our proposed work will use the multigrid Poisson solver provided by AMReX for the electrostatic application. We will present our results validated for a 1D capacitive discharge with benchmark results from literature. We will also present results obtained for a 2D 3V GEC reactor cell.

Presenters

  • Revathi Jambunathan

    Lawrence Berkeley National Laboratory

Authors

  • Revathi Jambunathan

    Lawrence Berkeley National Laboratory

  • Remi Lehe

    LBNL

  • Andrew Myers

    LBNL

  • Jean-Luc Vay

    Lawrence Berkeley National Laboratory

  • Weiqun Zhang

    LBNL