This $150,000 National Science Foundation project grant supports collaborative research at the University of North Carolina at Chapel Hill to develop GPU-accelerated first-principles simulation techniques for modeling exciton dynamics in complex materials systems. Funded through NSF's Mathematical and Physical Sciences program (CFDA 47.049), the three-year award running from August 2022 to July 2025 will advance the scientific understanding needed to design novel photovoltaic and photocatalytic energy harvesting approaches.
Specifically, the university researchers will implement new quantum mechanical methods for describing electron-electron interactions and electron-ion dynamics, optimized for graphical processing units. They will apply these simulation methods to study long-term exciton behavior in complex heterogeneous systems of energy technology relevance. Project results including documentation and tutorials will be made freely available to researchers. The team will also host a recurring summer school to build expertise, especially among early career U.S. scientists, in computational modeling of exciton processes. Findings on the accuracy, applicability, and computational costs of exchange-correlation and non-adiabatic dynamics approximations should help overcome barriers in large-scale simulations of practical energy systems.