(514g) Development of an Efficient Nonempirical Tight Binding Theory and Its Applications to Catalytic Systems

Computational predictions of structures and chemical reactivities of molecules and materials are limited by the high cost of ab initio quantum mechanical methods, such as density functional theory (DFT) and coupled cluster theory. Low-cost empirical reactive force fields, machine-learned interatomic potentials, and semiempirical tight binding (TB) models promise to accelerate the rate of discovery but suffer from large and expensive training dataset requirements and limited transferability of the empirical parameters across the chemical space. Recently, we introduced the nonempirical tight binding (NTB) theory that uses no NTB demonstrated high accuracy in calculating equilibrium bond lengths, binding energies, potential energy curves, and vibrational frequencies of model H_x systems.

In this work, we discuss our efforts toward the generalization of the NTB theory to main group elements in the pursuit of accurate and low-cost calculations of thermochemistry and energy barriers in model chemical reactions. We describe our approaches in implementing charge transfer, core-core interactions, and sp-hybridization effects and discuss their role in chemical bonding. We demonstrate the accuracy of NTB for describing bond dissociation in homonuclear and heteronuclear molecules and report on the progress toward generalizing the method to describe energy barriers in catalytic hydrogen-deuterium exchange, H₂ dissociation, and OH dissociation model reactions. We anticipate that the development of NTB will enable rapid computational characterization of a wide array of materials and reaction mechanisms, accelerating discoveries in chemistry and catalysis.

References:

1. Mironenko, A. V. (2023). "Analytical and Parameter-Free Huckel Theory Made Possible for Symmetric H(x) Clusters." J Phys Chem A 127(37): 7836-7843.
2. Mironenko, A.V. (2024). “Self-Consistent Equations for Nonempirical Tight Binding Theory”. Under review.
3. Leung, A. & Mironenko, A. V. (2024). In preparation.