BEAST: Beyond-DFT Electrochemistry with Accelerated and Solvated Techniques

Electrocatalytic systems are notoriously difficult to study due to the presence of significant configurational complexity and intricate electronic interactions between the molecules undergoing transformation and transition metals present in the vast majority of catalysts. The BEAST collaboration carries out various avenues of work to meet these modeling needs for electrocatalysis. In this talk, we discuss the BEAST team’s work on developing machine-learned functionals for the description of electrolytes in electrocatalytic systems, scalable and accurate beyond-DFT methods for the calculation of electronic structure and total energies, machine-learned models for beyond-DFT electronic structure and total energies, and a database of electrocatalytic properties computed with grand-canonical DFT (BEAST DB). We discuss the importance of describing both electronic and classical degrees of freedom accurately for understanding of electrocatalytic systems, and a vision for modeling of electrocatalytic systems moving forward.