Quantum Electron Dynamics of Large--Scale, Highly-Excited Systems using the Gaussian Hartree Approximated eFF kernel (eFF-GHA)

Our original eFF methodology enabled important contributions to modeling explicit electron large-scale non-adiabatic dynamics phenomena for systems containing low-Z elements of the periodic table (Z<6), at the cost of a force field. Yet, the simplicity of its Gaussian basis formulation leads to several limitations, including 1) the omission of the true electron exchange between same spin electrons, and 2) the lack of cusp conditions and nodal structures, which complicates scaling to higher Z’s. This leads to incorrect descriptions for double and triple bonds, lone pairs, and aromatic pi electron systems. We have now developed the Gaussian Hartree Approximated (GHA) kernel of the eFF concept to address these problems rigorously. GHA incorporates formal energy penalties for singlet and triplet pairs and angular momentum projected effective core pseudo-potentials that account for cusp conditions and missing nodal structures. I will present the underlying theory behind the new kernel and demonstrate its application to modeling materials in extremes.