Electron nuclear dynamics of time-dependent symmetry breaking in H+ + H2O at ELab = 28.5-200.0 eV: a prototype for ion cancer therapy reactions

Recurrent Neural Network/Machine Learning Predictions of Reactive Channels in H+ + C2H4 at ELab = 30 eV: A Prototype of Ion Cancer Therapy Reactions

We present a simplest-level electron nuclear dynamics/machine learning (SLEND/ML) approach to predict chemical properties in ion cancer therapy (ICT) reactions. SLEND is a time-dependent, variational, on-the-fly, and nonadiabatic method. In SLEND, nuclear and electronic parameters determine reactants-to-products trajectories in a quantum phase space; this establishes a mapping between reactants' initial conditions and products' properties. To accelerate simulations, SLEND/ML utilizes a modicum of SLEND trajectories to train ML methods on the aforesaid mapping and employs them to predict chemical properties. We employ SLEND/ML to predict reaction types and products' charges in H+ + C2H4 at ELab = 30 eV, a prototype of ICT reactions involving double-bonded compounds. For reaction predictions, a recurrent neural network (RNN) and k-nearest neighbor method are the best models with 98.23% and 95.13% accuracy. RNN correctly predicts frequent and infrequent reaction types and generalizes over data sets. For charge predictions, the RNN exhibits low mean absolute errors of 0.02-0.07.

Testing standard basis sets for direct ionizations: H+ + H at ELab = 0.1-100 keV

With the simplest-level electron nuclear dynamics (SLEND) method, we test standard Slater-type-orbital/contracted-Gaussian-functions (STO/CGFs) basis sets for the simulation of direct ionizations (DIs), charge transfers (CTs), and target excitations (TEs) in H+ + H at ELab  = 0.1-100 keV. SLEND is a time-dependent, variational, on-the-fly, and nonadiabatic method that treats nuclei and electrons with classical dynamics and a Thouless single-determinantal state, respectively. While previous tests for CTs and TEs exist, this is the first SLEND/STO/CGFs test for challenging DIs. Spin-orbitals with negative/positive energies are treated as bound/unbound states for bound-to-bound (CT and TE) and bound-to-unbound (DI) transitions. SLEND/STO/CGFs simulations correctly reproduce all the features of DIs, CTs and TEs over all the considered impact parameters and energies. SLEND/STO/CGFs simulations correctly predict CT integrals cross-sections (ICSs) over all the considered energies and predict satisfactory DI and TE ICSs within some energy ranges. Strategies to improve SLEND/STO/CGFs for DI predictions are discussed.