Theoretical investigation of the excited-state dynamics in the CH4 + O(3P) → CH3 + OH reaction

In combustion processes, particularly in systems involving reactive oxygen species (ROS), the role of electronically excited states has become a critical research frontier. This study employs the benchmark combustion reaction CH4 + O(3P) → CH3 + OH as a prototypical system. We constructed potential energy surfaces (PESs) for the three lowest-lying electronic states using high-level ab initio calculations and achieved efficient PES interpolation via machine learning (embedded atom neural network). By combining quasi-classical trajectory simulations with trajectory surface hopping for non-adiabatic transitions, we developed a multistate dynamical framework to quantify reaction rate constants inclusive of excited-state contributions. Key findings reveal that excited-state participation enhances the CH4 + O(3P) → CH3 + OH rate constant by approximately one order of magnitude (e.g., from 6.62 × 1010 to 2.08 × 1011 cm3 mol-1 s-1), in good agreement with available experimental data. This study not only confirms the crucial involvement of excited states in combustion processes but also delivers innovative theoretical frameworks and computational approaches for high-fidelity combustion kinetics modeling.