Promising Alloys for Hydrogen Storage in the Compositional Space of (TiVNb)100–x(Cr,Mo)x High-Entropy Alloys

This study reports on the search for the most promising alloys in the compositional space of (TiVNb)80Cr20-xMox (x = 5, 10, and 15) and (TiVNb)75Cr25-xMox (x = 5, 10, 15, and 20) high-entropy alloys. First, data-driven machine learning applied to these systems predicts that increasing the Mo content destabilizes the enthalpy of the hydride phases. Second, experimental and density functional theory (DFT) validations were performed. The as-prepared alloys have single-phase bcc lattices and rapidly absorb hydrogen to form fcc-type hydrides with a high capacity between 1.6 and 2.0 H/M. Despite a positive effect on the thermodynamics of the hydride phases, increasing the Mo content in these alloys has a negative effect on the maximum capacity. The cycling experiments highlight the need to balance the reversible capacity, cycle life, and crystalline stabilities of these phases. Therefore, considering all these results, the most promising alloy with trade-off properties within the targeted compositional space has been identified to be (TiVNb)75Cr5Mo20 that shows a maximum capacity of 2.6 wt % (1.8 H/M), a reasonable enthalpy of hydride formation (-38.6 kJ/mol H2), and a notable gravimetric reversible capacity of 1.42 wt % at room temperature. To identify the most promising high-entropy alloys for this application, integrated machine learning predictions followed by experimental and DFT validations proved to be an effective strategy.