Stochastic A Priori Dynamics for Complex Reactive Chemical Environments

The first-principles modeling of energetic chemical environments with use-inspired complexities is an ongoing challenge. We describe a stochastic framework for predicting the chemistry of complex systems that is scalable and well-suited for leveraging high performance computing resources. Elementary chemical events are treated using detailed semiclassical theories that include treatments for nonadiabatic transitions, tunneling, and zero-point energy maintenance combined with high-accuracy potential energy surfaces constructed automatically via ab initio permutationally invariant polynomials. Competition between elementary events is modeled stochastically to enable simulations of complex networks of reactions and to access long timescales. Applications of the stochastic model include quantifying the effects of rare nonthermal events in many body gas phase systems relevant to combustion, where highly reactive but transient energized species are formed, as well as low temperature atmospheric chemistry characterizing the reactivity of transient complexes.