A simple 1D setup in Argon with one grounded and one HF excitated parallel plate. This setup has been motivated by Turner et al. (2013)¹⁾ knowing how unreliable PIC-MC modeling results can be without comprehensive testing of the underlying implementation.

The setup file contains four parameter files P1.par, P2.par, P3.par and P4.par which can be started with up to 8 CPUs each. Start the simulation (e.g. P1.par) with:

$ rvmmpi -picmc 8 P1.par

The simulation plots, among others, electron and Ar+ density profiles along domain axis which can be compared against accordant benchmark results: Turner13 - supplementary material. The comparison against these benchmark results can also be seen in the figures below. As a side node: There is a deviation for the low pressure case P1 when using the “wrong” electron energy partitioning model for ionization collisions. We prefer the simpler model 'one-takes-it-all' where the incident electron keeps the remaining energy whereas Turner et. al. rely on 'equal-share' in which the energy is equally divided between incident and secondary electron. The truth should be lying somewhere in between. However, we implemented a switch for each partitioning model so that the user can choose.

Figure 1: Comparison against benchmark results with only minor deviation (left). Different electron energy partitioning models for ionization events yield deviations in the low pressure case P1 (right).