Scanning tunnelling microscopy was used to image low coverages of H on (001) surfaces at temperatures of between 300 and 700K. It was found that individual H atoms became mobile at about 570K. There was no apparent movement at temperatures below 500K and, at temperatures above 640K, the H was moving as fast as (or faster) than the scanning tip. The observed hopping rate along the dimer rows was consistent with an activation energy of 1.68eV; assuming an attempt frequency of 1013/s. Motion across the dimer rows was rarely observed, even at the higher temperatures. The diffusion barrier for motion along the dimer rows was estimated by using tight-binding and density-functional theory in the generalized gradient approximation. The calculated barrier was deduced to be equal to 1.65eV on the basis of tight binding considerations, and equal to 1.51eV according to the generalized gradient approximation. The experimental and theoretical results were in good agreement, and showed that thermal diffusion could not operate at temperatures below 500K. It was therefore suggested that a mobile precursor mechanism might be responsible for H pairing.

Hydrogen Diffusion on Si(001). J.H.G.Owen, D.R.Bowler, C.M.Goringe, K.Miki, G.A.D.Briggs: Physical Review B, 1996, 54[19], 14153-7