Ion implantation, and exposure to a plasma at 250C, were used to incorporate 2H. In the implanted samples, some 50% of the H was lost from the surface during annealing (500C, 1200s) and was essentially all gone before reaching 700C. This H retention was considerably less than that in materials such as SiC and sapphire. The in-diffusion of 2H from a plasma was much faster for LiAlO2, with an apparent diffusion coefficient of about 10-13cm2/s at 250C. This value was about 2 orders of magnitude greater than that for LiGaO3. It was concluded that H out-diffusion from these materials, used as substrates during nitride epitaxy, would not cause difficulties because the H was expected to have left these materials at temperatures that were lower than the epitaxial layer growth temperature.

R.G.Wilson, B.L.H.Chai, S.J.Pearton, C.R.Abernathy, F.Ren, J.M.Zavada: Applied Physics Letters, 1996, 69[25], 3848-50