Description
Monte Carlo simulation of neutral particles in plasmas is explored using an unstructured mesh through Parallel Unstructured Mesh Infrastructure (PUMI) Tally [1]. Neutral particle transport is typically modelled using Monte Carlo simulation on constructive solid geometry (CSG) as in, for example Degas 2 [2] or EIRENE [3]. CSG allows for computationally efficient mathematical relations to determine particle locations but requires sophisticated processing to generate from computer aided design (CAD). The PUMI framework allows for efficiently parallelized Monte Carlo simulation with heterogeneous computing architecture on complex unstructured meshes as opposed to CSGs. This allows for handling more complex geometries with acceptable performance loss [1] and a simpler process for Monte Carlo geometry preparation directly from CAD models.
This work develops an atomic physics library that connects to the PUMI-Tally framework previously demonstrated for neutron transport. PUMI-Tally leverages PUMIPic’s sophisticated procedure for decomposing domains across multiple graphics processing units (GPUs) to allow for GPU parallelized Monte Carlo simulation [4]. This work implements electron impact ionization and charge exchange with weighting for variance reduction. We demonstrate that this reproduces the desired statistics of neutral-plasma interactions and compare the results of this framework to existing codes to demonstrate the viability of this approach. Further development of this approach will allow for modelling atomic transport in complex plasma geometries and improved workflows from CAD models.
Acknowledgements:
This work was supported in part by the U.S. Department of Energy, Office of Science, Office of Workforce Development for Teachers and Scientists (WDTS) under the Science Undergraduate Laboratory Internships Program (SULI). This work is supported by the US DOE Contract No. DE-AC02-09CH11466
References:
[1] Hasan, F. et. al. (2025), GPU Acceleration of Monte Carlo Tallies on Unstructured Meshes in OpenMC with PUMI-Tally. arXiv preprint arXiv:2504.19048.
[2] Stotler, D. and Karney, C. (1994), Neutral Gas Transport Modeling with DEGAS 2. Contrib. Plasma Phys., 34, 392-397.
[3] Borodin, D. et. al. (2022), Fluid, kinetic and hybrid approaches for neutral and trace
ion edge transport modelling in fusion devices. Nucl. Fusion 62.
[4] Diamond, G., Smith, C. W., Zhang, C., Yoon, E., and Shepard, M. S. (2021). PUMIPic: A mesh-based approach to unstructured mesh Particle-In-Cell on GPUs. Journal of Parallel and Distributed Computing, 157, 1-12.