Machine-Learning Guided Discovery of Novel and Earth Abundant Electrolyser Materials
The future of our energy system lies in green hydrogen - a versatile and highly demanded energy source that complements the intermittent nature of renewable energy sources like solar and wind power. A promising technology for producing green hydrogen is the polymer electrolyte membrane (PEM) electrolyser, but it relies on scarce platinum and iridium as electrocatalysts. Within this Energy in Transition Hub seed-funding project, the research team developed cutting-edge Artificial Intelligence (AI) and machine learning (ML) algorithms, coupled with advanced computational models for hydrogen and oxygen splitting reactions. Using quantum-chemistry data from public databases like the Open Catalyst Project, the team created an ML model that connects quantum-chemical properties and catalytic performance, considering not only the catalytic activity of the materials but also their price and abundance.
The collected data on the costs, availability, and electrochemical properties (including catalytic activity and selectivity) of the materials have been collected in a central repository. Going forward, the methodology will be further expanded to search for protective coatings to prevent catalyst corrosion in seawater. The team has also been exploring other types of electrolysers, such as anion-exchange membrane electrolysers. All data produced within the project will be shared with the scientific community in the spirit of UU Open-Science, so that the scientific community can work together to create a more sustainable energy future.
Contact
For more information contact Dr. Artrith Nong, Debye Institute for Nanomaterials Science - Materials Chemistry and Catalysis.