Theoretical Study of Energy Transfer between CO and CO₂ Based on Full-Dimensional Potential Energy Surface

We present a comprehensive study of the CO₂-CO interaction potential using a 9-dimensional (9D) potential energy surface (PES) constructed with the fundamental invariant-neural networks (FINN) method. The PES was generated from 65,330 CCSD(T)-F12a/aug-cc-pVTZ ab initio data points, with counterpoise correction, applied to eliminate basis set superposition error (BSSE). We performed quasi-classical trajectory simulations using this PES to investigate the rotational energy transfer dynamics. Our results reveal complex energy transfer mechanisms, with significant rotational excitation and relaxation dependent on collision energy and initial rotational states. The 9D PES provides a more accurate representation of the CO₂-CO system, offering new insights into molecular dynamics and interaction mechanisms.