Materials Informatics Applied to the Discovery of High Entropy Oxides for Solid Oxide Fuel Cells

• Grant number: 25/06024-6
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: August 01, 2025
• End date: August 31, 2028
• Field of knowledge: Physical Sciences and Mathematics - Physics - Condensed Matter Physics
• Principal Investigator: James Moraes de Almeida
• Grantee: Arthur da Silva Sousa Santos
• Host Institution: Centro de Ciências Naturais e Humanas (CCNH). Universidade Federal do ABC (UFABC). Ministério da Educação (Brasil). Santo André , SP, Brazil
• Associated research grant: 23/09820-2 - Materials by design: from quantum materials to energy applications, AP.TEM

Abstract

The development of solid oxide fuel cell (SOFC) technology reveals promising potential for decarbonizing various sectors. From the perspective of energy independence from fossil fuels, Brazil strategically stands out on the global stage and can assume a leadership role in the future, provided that technical challenges are overcome. This is due to the abundance of natural resources and an energy matrix recognized as one of the cleanest in the world. SOFCs prove to be allies to intermittent energy sources, allowing the conversion of electricity generated during periods of high solar or wind availability into hydrogen. This hydrogen can then be used to generate electricity through SOFCs during periods of low sun or wind incidence. However, the dissemination of this technology in the Brazilian market still faces significant challenges, with the main ones being the high manufacturing cost and the limited lifespan of the devices.In this research project, we propose to investigate an emerging class of advanced materials-high entropy oxides (HEOs)-with the objective of evaluating their applicability in SOFC cathodes. The literature indicates that these materials possess unique characteristics, making them promising for energy storage and conversion applications. Due to the effects associated with high entropy, HEOs can both extend the lifespan of cathodes and increase catalytic activity in the oxygen reduction reaction, the primary determinant of cathode performance.To address these complex challenges, we propose an interdisciplinary methodological approach. Initially, large language models (LLMs) will be used to generate, from the processing of scientific articles, a database containing HEO compositions and their respective phases. With this database ready, classification models will be trained and used to perform a screening to discriminate compositions with the greatest potential to form the perovskite phase, which is the phase usually used in SOFC cathodes. This group of potential high entropy perovskite compositions will be explored through active learning, involving calculations of the Gibbs free energy of hydrogen adsorption, which is the primary descriptor of cathode efficiency, using Density Functional Theory (DFT) and the computational hydrogen electrode model. The most promising compositions will be synthesized via the solid-state reaction route and characterized using Impedance Spectroscopy. (AU)