Artificial neural network molecular mechanics of iron grain boundaries

This study reports grain boundary (GB) energy calculations for 46 symmetric-tilt GBs in α-iron using molecular mechanics based on an artificial neural network (ANN) potential and compares the results with density functional theory (DFT), embedded atom method (EAM), modified EAM (MEAM). The by ANN are excellent agreement those of DFT (5% average), while MEAM significantly differ from (about 27% average). In a uniaxial tensile calculation ∑3(11¯2) GB, reproduced brittle fracture tendency GB observed mistakenly showed ductile behaviors. These demonstrate effectiveness calculating boundaries iron, which is high demand modern industry.