Answers are grounded in this paper and cite their sources.
From a measurement of two-dimensional displacement distributions in random-walk diffusion of Ir atoms on the Ir(001) surface, we find that the adatom jumps along the 〈001〉 direction instead of along the smoother closely packed 〈110〉 atomic-channel direction. The sites visited by one diffusing adatom therefore form a c(2×{}2) net of the substrate lattice. The displacement distributions measured at various temperatures agree with theoretical distributions for the discrete nearest-neighbor random walk on the c(2×{}2) lattice. This diffusion most probably occurs by an atomic-replacement mechanism of the adatom with a substrate atom. On the Ir(110) surface, an Ir adatom can displace along the [110] surface channel direction as well as across these channels. The latter displacements are known to occur by an atomic-replacement mechanism. The visited-site lattice agrees with the (1×{}1) net of the substrate. From the displacement distributions we conclude that {~}80% of the cross-channel jumps are in the 〈112〉 direction and {~}20% are in the 〈001〉 direction. The low-temperature displacement distributions agree best with Monte Carlo distributions for the discrete nearest-neighbor random walk, but at {~}260 K a significant number of the along-channel jumps extend beyond the nearest-neighbor distance. The energies needed for these atomic replacements on the (001) and (110) surfaces are found to be 0.84±{}0.05 and 0.71±{}0.02 eV, respectively. We also find a close correlation between atomic-relpacement self-diffusions and surface reconstructions.
No takes yet. Share an insight, caveat, or question.
Tsong et al. (1991) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: