Multi-species modeling in the particle-based ESBGK method for monatomic gas species

TL;DR

This paper extends the particle-based ESBGK method to multi-species monatomic gases within the PICLas simulation suite, verifying accuracy against DSMC and highlighting efficient transport coefficient calculation methods.

Contribution

It implements and verifies multi-species modeling in the ESBGK method using Brull's model and Wilke's mixing rules, with a focus on transport coefficient determination.

Findings

Good agreement with DSMC results across test cases

Collision integral approach yields better accuracy

Transport coefficient calculation impacts computational efficiency

Abstract

Multi-species modeling is implemented for the particle-based ellipsoidal statistical Bhatnagar-Gross-Krook (ESBGK) for monatomic species in the open-source plasma simulation suite PICLas. After a literature review on available multi-species extensions of the kinetic model equations and approaches for the determination of the transport coefficients, Brull's model is implemented for the former and Wilke's mixing rules and collision integrals for the latter. The implementation is verified with three simulation test cases: a simple reservoir, a supersonic Couette flow and the hypersonic flow around a 70$^{\circ}$ blunted cone. The simulation results are compared with the Direct Simulation Monte Carlo (DSMC) method, where good overall agreement can be achieved. However, the determination of the transport coefficients through collision integrals offers better agreement with the DSMC results at acceptable computational cost. For the last test case, a comparison of the computational duration is presented.