Ab-Initio Collision Models for DSMC and Their Applications to Reacting Flows

The direct simulation Monte Carlo (DSMC) method is an atomistic-level technique for modeling nonequilibrium flows appearing in the fields of gas dynamics and physical chemistry. Development of DSMC collision models, in its vast majority, has traditionally focused on high-enthalpy and high-Knudsen number reentry flows. As pointed out in this work, standard DSMC approaches for post-reaction energy redistribution may not satisfy detailed balance when recombination/exchange reactions play an important role in the flow energy balance. This issue can be even more critical in reacting mixtures involving polyatomic species such as combustion. As the first goal of this dissertation, we address this issue and propose new strategies for post-reaction energy redistribution that ensure reacting mixtures relax to complete thermochemical equilibrium. Second, we apply these strategies to model atmospheric low-speed combustion problems with DSMC. In particular, we model the 1-D laminar flame structure of H2-O2 premixed systems. This is also intended to illustrate how evolution of high-performance computational platforms can be used to extend the conventional DSMC range of applicability. The third and last part of the dissertation focuses on the implementation of compact high-fidelity collision models, based on ab-initio data, to accurately describe strong nonequilibrium flows. These new reaction and energy exchange models are consistently compared with the standard DSMC phenomenological framework