Improved Reaction Energies and Electronic Structures in Charged Electrocatalytic Environments

Batteries, fuel cells, and electrolyzers are electrochemical technologies that are becoming an important part of the U.S. industrial infrastructure. Their key advantage is that they can convert back and forth between chemical energy and electricity. This provides a bridge between the energy sector and the chemical industry, and it allows intermittent power sources to be stored for later use. These devices rely on chemical reactions that occur at a charged surface. There, an applied voltage can "push" atoms and molecules into reactions that would not otherwise occur. New computer simulations can predict these reactions at the atomic level. This could allow researchers to design reactions that are not possible or practical today. However, today's simulations struggle to capture how the charge at a reacting surface changes as the reaction proceeds. This limits their accuracy and usefulness. This project will improve the accuracy of these simulations by improving the physical models of how charge moves in response to reactions at electrified surfaces. More accurate simulations will help researchers and engineers design better devices, supporting national goals for energy security and economic competitiveness. This project will take a three-pronged approach to improving both the reaction energies and electronic structure representation when charged, electrochemical environments are simulated using periodic calculations. The first approach will rely on an a posteriori treatment involving electrostatic modeling, resulting in both first- order improvements to reaction energy accuracies and physically intuitive models describing such interactions. The second approach will involve integration of these electrostatic techniques into the self-consistent field cycle of density functional theory codes, providing an a priori improvement to the calculated electronic structure in changing charge environments. The third approach will improve grand-canonical machine-learned interatomic potentials to account for long-range charge interactions. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.