Room temperature hydrogen storage properties of Ti-Zr-Mn-Fe-Co high-entropy alloys designed by semi-empirical descriptors , thermodynamic calculations and machine learning

Autor(es)

Edalatia, Payam ; Andrade, Gaspar ; Strozi, Renaco B. ; Dangwal, Shivam ; Edalati, Kaveh ; Floriano, Ricardo

• Tipo de documento: Artigo Científico
• Fonte: Journal of Alloys and Compounds; v. 1022, p. 10-pg., 2025-04-10.

Resumo

One of the most significant challenges in using hydrogen as a substitute for fossil fuels is designing alloys that can store hydrogen in solid form. These alloys must facilitate reversible reactions over long cycles, with fast kinetics at room temperature. Given the potential of high entropy alloys (HEAs) for safe, and high-density hydrogen storage, this study combines semi-empirical descriptors (atomic size mismatch, atomic radius ratio of hydrideforming to non-hydride-forming elements, and valence electron concentration), thermodynamic calculations via CALPHAD, and machine learning (ML) to design two novel alloys Ti20Zr20Mn20Fe20Co20 and Ti15Zr22Mn27Fe26Co10. While empirical descriptors and CALPHAD were used to select alloys with a high tendency to form the C14 Laves phase, ML was employed to predict the enthalpy of hydride formation, a critical thermodynamic parameter for hydrogen storage. The synthesis of these two HEAs and subsequent experimental investigations, using pressure-composition-temperature (PCT) isotherm and absorption kinetic measurements, revealed that the Ti20Zr20Mn20Fe20Co20 and Ti15Zr22Mn27Fe26Co10 alloys absorb 1.5 wt% of hydrogen with fast kinetics and reversibility at room temperature for at least ten cycles. The predicted enthalpy by ML closely matches the measured enthalpy, demonstrating the potential of artificial intelligence approaches in estimating fundamental thermodynamic parameters to design new hydrogen storage materials. (AU)

• Processo FAPESP: 23/07884-3 - Processamento e caracterização das propriedades mecânicas e da biocompatibilidade de ligas multicomponentes dos sistemas (TiZrNbTa)90Cu10-xAgx e(TiZrNbMo)90Cu10-xAgx (para X = 0, 2.5, 5.0, 7.5 and 10 at%) produzidas para as aplicações biomédicas.
• Beneficiário: Payam Edalati
• Modalidade de apoio: Bolsas no Brasil - Doutorado
• Processo FAPESP: 22/03024-7 - Estrutura, processamento e propriedade de ligas multicomponentes avançadas para as aplicações biomédicas e armazenagem de energia
• Beneficiário: Ricardo Floriano
• Modalidade de apoio: Auxílio à Pesquisa - Projeto Inicial