Speakers

Alkaline membrane electrolysis (AEM) and high-temperature steam electrolysis (SOEC) are important technologies for the emerging production of green hydrogen from seawater. While both have been proven effective for conventional electrolysis, many processes relevant to catalytic activity and to the degradation of functional electrode/electrolyte interfaces remain insufficiently understood at the atomistic scale, even in cells operated with water of the highest purity. This lack of understanding severely hampers a rapid, knowledge-driven transfer of the technology to seawater use cases, where tolerance to residual trace species will be a major factor determining future large-scale adoption.

Due to the structural and dynamic complexity of electrified catalytic solid–liquid and solid–solid interfaces in AEM and SOEC under technical operating conditions, insights from theoretical modeling of simplified systems have, in the past, been limited in predictive power. I will present ML-enabled methods co-developed by our group over recent years and their application to fundamental problems in AEM HER and SOEC OER, showcasing the growing scope of emerging theoretical modeling capabilities for realistic models of complex functional interfaces.