Analysis of Stability\nand Mechanical Properties of\nAl-Cr-Fe-Co-Ni-Cu HEA Based on Machine Learning

A machine\nlearning (ML) interatomic potential was developed to\npredict the behavior of complex high-entropy alloys (HEAs) containing\nAl, Cr, Fe, Co, Ni, and Cu. This ML potential accurately reproduces\nDFT calculations, allowing for extensive molecular dynamics simulations.\nWe analyzed the mechanical properties of face-centered cubic (FCC),\nbody-centered cubic (BCC), and polycrystalline HEA phases at different\nelement concentrations and temperatures. The ML potential accurately\npredicts key system parameters such as lattice constants and elastic\nmoduli and describes the stress–strain material behavior. Strain\nanalysis shows a complex phase transformation and dislocation formation\nin the FCC phase under tension. High Co and Ni concentrations improve\nthe mechanical properties of the FCC phase. Polycrystalline structures\nform BCC-HCP mixtures, highlighting the dependence of their stability\non the element concentration. Temperature-dependent simulations show\na transition from the FCC phase to the amorphous phase at 1200 K.\nInterestingly, Cu-rich clusters play a crucial role in stabilizing\nthe FCC phase at lower temperatures. This study demonstrates the efficiency\nof the ML potential for studying HEAs and unveils the complex relations\nbetween element composition and the resulting mechanical properties.