ReaxFF Molecular Dynamics Simulations of Thermal Reactivity of Various Fuels in Pyrolysis and Combustion

Abstract

The methodology development and applications of ReaxFF molecular dynamics (ReaxFF MD) in unraveling the complex reactions and kinetics for pyrolysis and oxidation of organic systems are reviewed. Particular attention is given to the large-scale ReaxFF MD simulation method of similar to 10,000 atoms and practical simulation strategies to overcome the temporal-spatial limits as much as possible. High-performance computing codes running on CPU cluster/supercomputers and on GPU were overviewed. GPU-enabled code like GMD-Reax is revolutionizing large-scale ReaxFF MD simulations to run mainly on a single GPU. A step-forward for reaction analysis was achieved in the code of VARxMD to reveal detailed reactions based on the identification of bond types and unique species that allows for categorization of reaction species and pathways through structure searching of reaction sites and reactants/products. Efforts for extracting rate constants and kinetics modeling from unperturbed kinetics of ReaxFF MD simulations are reviewed, and challenges remain. The important factors of model scale effects, elevated simulation temperature, and possible validation of ReaxFF MD reaction simulation results theoretically and experimentally are illustrated and discussed. The novel hybrid simulation strategies proposed recently that may expand applications of ReaxFF MD to connect with realistic scenarios of engine combustion and relevant chemical effects on aerospace vehicle materials are briefed. The simulation practices of large-scale ReaxFF MD in understanding the similarities and differences of reactivity and reaction mechanisms in pyrolysis of liquid hydrocarbon fuels, solid fuels (coal, biomass, polymer), and energetic materials of CL-20 and its cocrystals are briefly described, which indicates that large-scale ReaxFF MD simulations are a feasible and straight means to capture the dynamics of an almost entire reaction process and to evaluate reactivity of organic systems computationally. There is still a great deal of work to be done to push the boundaries of ReaxFF MD simulations forward. With the constant improvement of the ReaxFF force field and reaction analysis capability, ReaxFF MD playing a key role in understanding reaction mechanisms and its potential in kinetics modeling of pyrolysis and oxidation of various fuels can be expected.