Executive Summary : | Electrochemical water splitting, a method for generating molecular hydrogen and oxygen, has potential to address the energy crisis and fossil fuel depletion. However, its efficiency is limited by the high overpotential of the four-electron transfer kinetics of oxygen evolution reaction (OER). Ir/Ru-based oxides are considered the most effective electrocatalysts for OER due to their intrinsic activity but high cost. Metal-organic frameworks (MOFs) have gained interest for energy-related applications due to their tunable porosity, high specific surface area, and diversity of metal centers and organic linkers. Machine learning (ML) algorithms will be used to accelerate catalyst discovery and identify efficient MOF-based nanomaterials for OER catalysis. The team will use domain-specific ML methods, such as symbolic regression and compressed sensing, to find descriptive parameters and optimize catalytic performance. The goal is to improve MOF catalysts' long-term stability and hydrogen production efficiency. |