Experimental and atomistic simulations were made of grain boundary mobility, as a function of temperature and boundary misorientation, by using a geometry that ensured steady-state curvature-driven boundary migration. Molecular dynamics simulations were performed by using Lennard-Jones potentials on a triangular lattice. This was the first systematic study of the dependence of intrinsic grain boundary mobility upon misorientation. The experiments focussed upon high-purity Al having <111> tilt boundaries which were isomorphic to those used in the simulations. Excellent agreement between simulations and experiment was obtained with regard to almost every aspect. The boundary velocity was found to be a linear function of the curvature, and the mobility was observed to be an Arrhenius function of the temperature. The activation energies for boundary migration varied, with misorientation, by more than 40% in simulations and by 50% in experiments. In both simulations and experiments, the activation energies and the logarithm of the pre-exponential factor for mobility exhibited very similar variations with misorientation. This included the presence of distinct cusps at low Σ-misorientations. The activation energy for boundary migration was a logarithmic function of the pre-exponential factor for mobility within a small misorientation range that was centred around low Σ-misorientations.

Misorientation Dependence of Intrinsic Grain Boundary Mobility - Simulation and Experiment. M.Upmanyu, D.J.Srolovitz, L.S.Shvindlerman, G.Gottstein: Acta Materialia, 1999, 47[14], 3901-14