Achieving DFT accuracy in short range ordering and stacking fault energy using moment tensor potential for CoCrFeNi and CoCrNi

Medium-entropy alloys (MEAs) such as CoCrFeNi and CoCrNi are promising structural materials owing to their outstanding mechanical and thermal properties, which arise from complex chemical disorder and atomic-scale interactions. Although density functional theory (DFT) has provided fundamental insights into these systems, its high computational cost limits exploration of large-scale phenomena. Classical interatomic potentials have been used to address this gap but often lack the fidelity needed to capture many-body interactions and chemical short-range ordering (CSRO) effects. In this work, we developed a machine-learned Moment Tensor Potential (MTP) to bridge accuracy and efficiency. The MTP was trained on a comprehensive DFT database spanning unary to quaternary configurations and reproduces energies, forces, and stresses with near-DFT accuracy across diverse structural and chemical environments. It accurately predicts elastic properties and recovers compositional trends in bulk and shear moduli in agreement with DFT. Hybrid Monte Carlo/molecular dynamics simulations capture CSRO, reproducing key DFT-reported features including Cr-Cr and Fe-Fe repulsion and Ni-Cr ordering. Stacking fault energetics were modeled, yielding ISF energies near 54 mJ/m2 for CoCrNi and 36 mJ/m2 for CoCrFeNi, consistent with DFT predictions. Local chemical environment effects on stacking faults were also resolved: Co-rich planes reduce, whereas Cr- or Fe-rich planes increase, the stacking fault energy. By enabling large-scale, high-fidelity simulations at a fraction of DFT's cost, the developed MTP provides a robust framework for predictive modeling of thermodynamic stability, defect behavior, and mechanical response of FCC MEAs.