Fast machine learned $α$-Fe-H interatomic potential for hydrogen embrittlement (2511.21464v1)

Abstract: In this work, we present a machine-learned interatomic potential for the $α$-Fe-H system based on the tabulated Gaussian Approximation Potential (tabGAP) formalism. Trained on a Density Functional Theory (DFT) dataset of atomic configurations, energies, forces, and virials, the potential is designed to address the issue of H-induced acceleration of mechanical failure of metals, generally known as hydrogen embrittlement (HE). The proposed potential is shown to outperform the widely used classical and machine-learned interatomic potentials in fundamental properties of the $α$-Fe-H system. We show that the tabGAP model reproduces H-point defect properties, H-dislocation interaction, H-H interaction, and elastic constants with nearly DFT-level accuracy at a computational cost that is competitive with the efficient classical Embedded Atom Method (EAM) potentials. We further demonstrate the utility of the tabGAP model in molecular dynamics simulations of tensile tests of a perfect, and a $(111)[11\bar{2}]$-notched $α$-Fe structure with and without the load of H atoms. The simulations show evidence of HE via observation of accelerated decohesion of Fe atoms at the tip of the notch, and an increase in vacancy concentration driven by H-dislocation interactions. Hence, the results of the presented simulations support the hypotheses of hydrogen-enhanced decohesion (HEDE) alongside hydrogen-enhanced strain induced vacancies (HESIV) as important mechanisms for H-induced mechanical failure of iron systems.