Machine Learning Representations of the Three Lowest Adiabatic Electronic Potential Energy Surfaces for the ArH₂ ⁺ Reactive System

In this work, we present Gaussian process regression machine learning representations of the three lowest coupled 2A' adiabatic electronic potential energy surfaces of the ArH2 + reactive system in full dimensionality. Additionally, the nonadiabatic coupling matrix elements were calculated. These adiabatic potentials and their nonadiabatic couplings are necessary ingre-dients in the theoretical investigation of the nonadiabatic reaction dynamics of the Ar + H-2( +) ? ArH+ + H and Ar+ + H-2 ? ArH+ + H reactions, as well as the competing charge transfer process, Ar + H-2(+)? Ar+ + H-2. Accurate ab initio electronic structure calculations (ic-MRCI+Q/aug-cc-pVQZ), wher e b y the effect of spin-orbit coupling in Ar+ has been accounted for through the state interaction method, serve as input for the machine learning training process. The potential energy surfaces are fitted with high accuracies, with root-mean-square errors on the order of 10(-7 )eV for the three surfaces, which meet the requirements for chemical dynamics at low temperature. It was found that quite a large number of training points (of the order of 5000 ab initio points) are needed in order to achieve these accuracies due to the complex topography of these electronic surfaces.