Comparison of Coulomb Collision Models in Simulations of Odd-Parity Field Reversed Configuration

Accurately computing Coulomb interactions in collisional plasmas is a key component of the kinetic description achieved in particle-in-cell (PIC) simulations. Two common approaches for solving the Landau–Fokker–Planck collision operator are the binary method that performs energy and momentum conserving pairwise collisions, and the grid-based Langevin equations method that computes the drag and diffusion on particles in response to the local field. These Monte Carlo techniques are computationally expensive because of the large number of simulation particles required to sufficiently resolve velocity space and the scattering process. In this talk, we investigate a low-noise PIC method that uses Gauss–Hermite quadrature to initialize particles and compute collisions. This reduces the number of particles needed to resolve velocity space and, similar to the quiet direct simulation technique, eliminates the strong dependence on dense random number sampling for accuracy. We report on validation efforts of the new model through electrical conductivity and stopping-power measurements. We will also discuss a comparison of the methods in simulations of an odd-parity field reversed configuration that demonstrates electron heating by rotating magnetic fields.