Grain boundary and lattice diffusion in nanocrystal α-iron: An atomistic simulation study

To obtain fundamental understanding on the effect of grain boundaries on the diffusion kinetics, molecular dynamics simulations (MD) were carried out on single crystal and nanocrystal (with a mean grain size of 2.5 nm) bcc iron using the second nearest-neighbor modified embedded atom method (2NN-MEAM) interatomic potential. Self-diffusion coefficient in single crystal and nanocrystal samples were calculated in the temperature range from 350 K to 1000 K. A temperature-dependence of the diffusion coefficient according to the Arrhenius law was obtained for both lattice and grain boundary diffusion. By doing so, activation energies as well as pre-exponential factors were derived from the diffusion coefficients and compared to experimental data. MD simulation results show that diffusion rate of iron atoms in nanocrystal sample is 6 to 28 orders of magnitude greater than single crystal. The trajectory of iron atoms during diffusion process verified that diffusion occurs mostly in the grain boundaries of nanocrystal iron; suggesting that grain boundary diffusion is dominant in nanocrystal iron. Based on the obtained results pure grain boundary diffusion coefficient was calculated.