Design of Metal Hydrides for Hydrogen Compression and Batteries: A Machine Learning and Simulation-Based Approach

• Grant number: 25/03614-7
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: January 01, 2026
• End date: December 31, 2029
• Field of knowledge: Engineering - Materials and Metallurgical Engineering - Physical Metallurgy
• Principal Investigator: Ricardo Floriano
• Grantee: Gabriel de Luccah Bueno Galliani Fontana
• Host Institution: Faculdade de Ciências Aplicadas (FCA). Universidade Estadual de Campinas (UNICAMP). Limeira , SP, Brazil

Abstract

Metal hydrides are one of the most versatile technologies for the development of hydrogen-based energy systems due to their wide range of applications. Recently, multicomponent alloys have gained prominence in the development of new hydrides as they offer a vast compositional space, providing numerous alloys with great potential to act as metal hydrides for several purposes. However, this enormous compositional space makes the initial selection of a multicomponent alloy for in-depth studies a difficult, costly and onerous process. The present study goal is to analyze, develop, and implement machine learning-based design methods to select multicomponent alloys, initially unreported in the literature, with a focus on compositions with high potential to act as metal hydrides that can be applied in different scenarios besides hydrogen storage, specifically, in hydrogen compressors and hydrogen batteries. The methods will be evaluated through comparisons of the predicted results with experimentally obtained results and through the comparison of the effectiveness of the developed methods with other methods already reported in the literature. It is expected that the results obtained can contribute not only to the scientific community but also to industries and other ventures that wish to implement hydrogen-based energy systems in their services and/or products. (AU)