Fig. 2: Calculation accuracy of the constructed machine learning interatomic potential for the Fe-H binary system. | Communications Materials

Fig. 2: Calculation accuracy of the constructed machine learning interatomic potential for the Fe-H binary system.

From: Machine learning interatomic potential reveals hydrogen embrittlement origins at general grain boundaries in α-iron

Fig. 2

a, b Hydrogen solution energy at tetrahedral sites in α-Fe under uniaxial strain and hydrostatic pressure, respectively. c Segregation energy of hydrogen to the easy core structure of screw dislocation. d (112) generalized stacking fault energy when hydrogen is positioned at tetrahedral sites on generalized stacking faults. e Interaction energy between a single vacancy and multiple hydrogen atoms. f Segregation energy of hydrogen for the most stable site on a low-index surface. g Bond length dependence of the energy of hydrogen molecules. h Interaction energy between hydrogen atoms at tetrahedral sites in the bulk. For comparison, results obtained from DFT, the MLIP of previous study, and EAM are also shown. In a, b, c, e, and h, the positions of hydrogen atoms corresponding to the horizontal axis are also shown.

Back to article page