Physical-Chemistry-Inspired AI for Catalyst Design
Chemistry plays a double-edged role in the climate change battle: It is both a facilitator of the problem and a contributor to a solution. The very science that drove global industrial advancements, leading to increased carbon emissions, also holds the keys to innovations such as CO2 capture and storage, CO2 reduction catalysts, photovoltaics and batteries, sustainable plastics or water splitting and the hydrogen economy. Computational chemistry has—fueled by the recent incredible progress of artificial intelligence (AI)—been particularly relevant to these innovations. By simulating chemical processes, we can explore vast molecular landscapes and reactions without the need for comparably slow real-life lab experiments that often involve hazardous substances. The scalability, automatability, and efficiency of the AI-enhanced methods are almost unprecedented. One promising approach to combating climate change is the development of efficient catalysts for water splitting, a process that generates hydrogen, a clean and renewable energy carrier:
2 H2 O → 2 H2 + O2
Artificial photosynthesis aims to imitate the natural process by which plants convert sunlight, water, and carbon dioxide into energy-rich molecules. However, replicating this complex mechanism is a huge challenge due to the intricate electronic and structural properties of the catalysts involved, often transition metals.