Molecular dynamics simulation of the transformation of Fe-Co alloy by machine learning force field based on atomic cluster expansion

The force field describing the calculated interaction between atoms or molecules is key to accuracy of many molecular dynamics (MD) simulation results. Compared with traditional semi-empirical fields, machine learning fields have advantages faster speed and higher precision. We employed method atomic cluster expansion (ACE) combined first-principles density functional theory (DFT) calculations for learning, successfully obtained binary Fe-Co alloy. Molecular simulations alloy carried out using this ACE predicted correct phase transition range