It was pointed out that a major problem persisted in understanding diffusion in alumina: in that all such measurements referred to extrinsic diffusion but the experimental activation energy was larger, than any of the theoretical predictions, by a factor of 2. The situation tended to be even worse in the case of grain boundary diffusion. The most extensive set of experimental data suggested that the activation energy for boundary diffusion was 50% higher than that for bulk diffusion. A strategy for resolving these problems was considered here.

Experiment and Theory of Diffusion in Alumina. J.H.Harding, K.J.W.Atkinson, R.W.Grimes: Journal of the American Ceramic Society, 2003, 86[4], 554-9