Computer Simulation Of Self-diffusion In Metals And Alloys

In solid materials, diffusion is the only way to achieve transmission of matter. Many phenomena in the process of manufacture or use of materials and some properties have correlation with diffusion, such as solid state phase transition, solidification, segregation, deposition, sintering of metallic powder, creep, annealing and oxidation etc. Thus, explanation for the macroscopical law and microcosmic mechanism of diffusion are very important. So a further investigation of diffusion has both theoretical and practical significance. In this paper, Combining computer simulation with modified analytic embedded-atom method (MAEAM) potential, the formation, migration and activation energies have been calculated for seven BCC transition metals Fe, W, Mo, Cr, Ta, Nb and V, four-kind migrations of Cu vacancy and three-kind migrations of Ag vacancy in Cu-Ag immiscible alloy system, Ni self-diffusion in intermetallic compound NiA1 for five diffusion mechanisms and a single vacancy in the first six planes of Pd (001) surface. From the energy minimization, the favorable diffusion mechanism of the vacancy in metals and alloys and vacancy on the uppermost and near layer of surface are discussed in detail. The results are shown in the following.(1) For each kind of three diffusion mechanisms nearest-neighbor, next-nearest-neighbor and third-nearest-neighbor in BCC transition metals Fe, W, Mo, Cr, Ta, Nb and V, the energy curve is symmetric and the maximum value of the energy appears at the middle point of the diffusion path. Compared the energies corresponding to three diffusion mechanisms, the NN diffusion needs the lowest activation energy (and thus the lowest migration energy). So that, the NN mono-vacancy diffusion is favorable in BCC transition metals.(2) The formation, migration and activation energies have been calculated for four-kind migrations of Cu vacancy and three-kind migrations of Ag vacancy in Cu-Ag immiscible alloy system. The equilibrium concentration of Cu vacancies is greater than that of Ag vacancies owing to the formation energy of Cu vacancy is lower than that of Ag vacancy. Comparing the migration or activation energy needed for four-kind migrations of Cu vacancy and three-kind migrations of Ag vacancy show that the favorable migration mechanism is the nearest-neighbor (NN) jump for Cu vacancy, while the straight [010] six-jump cycle (6JC) for Ag vacancy. Furthermore, the activation energy of the NN jump of Cu vacancy is lower than that of straight [010] 6JC of Ag vacancy also show that the former is more favorable. We conclude accordingly that the primary migration mechanism is the NN jump of an abundance of Cu vacancies. The formation, migration and activation energies of Ni self-diffusion in intermetallic compound NiA1 have been calculated for five diffusion mechanisms, NNN jump, [110] 6JC, straight [100] 6JC, bent [100] 6JC and triple-defect diffusion. The results show that the Ni self-diffusion is dominated by the triple-defect diffusion mechanism since it requires essentially the lowest migration or activation energy in the five diffusion mechanisms. This is consistent with the conclusion of Frank et al~[81].(3) The formation energy of a single vacancy is the lowest (much lower than the bulk value) for the vacancy in the first layer and the highest (slightly higher than the bulk value) for the vacancy in the second layer. For migration of a vacancy in intra-layer, the activation energy of self-diffusion increases in the sequence of the first-, second- and third-layer, of the third- or fourth-layer is slightly hiller than the bulk value and of the third-layer is the highest. For migration of a vacancy in inter-layer, the activation energy of self-diffusion of the vacancy migrating to upper layer decreases for the vacancy in the third layer to the first layer. Comparing the relative large of the activation energies of self-diffusion for the vacancy in the second-, third-, fourth- or fifth-layer to migrate in itself layer, to upper layer and to nether layer, we know that the vacancy in the second-layer is favorable to migrate to the first-layer. However, a single vacancy in the third-, fourth- or fifth layer is not easy to migrate to the upper layer due to its slightly higher the activation energy than that of the vacancy to migrate in intra-layer or to nether layer. Below the fifth-layer, the calculated activation energies of self-diffusion in intra-layer, upper-layer and inter-layer, are equal to the calculated bulk value means that the effects of the surface to the vacancy formation and migration are only for the vacancy in the first five layers.