Time-discretization of a plasma-neutral MHD model with a semi-implicit leapfrog algorithm

Interaction between plasma and neutral species can significantly alter the dynamic behavior of magnetically confined devices and incorporating the atomic physics associated with these interactions is required for an accurate plasma-neutral model. Previous work [Taheri, APS-DPP 2020] focused on incorporating the atomic physics in the semi-implicit leapfrog framework of the NIMROD code [Sovinec, JCP 2004] using Crank-Nicolson with Newton iteration for nonlinear terms and an operator-splitting method using Strang splitting. A battery of 0-D test cases showed the accuracy and efficiency of these methods. In this work we expand testing to consider 1-D cases that roughly represent tokamak-plasma edge fueling and z-pinch startup from a plasma-gun injector and show that a Douglas-Rachford inspired operator splitting can reduce the error, over Strang split to a level comparable to that of Crank-Nicolson with nonlinear iteration.