Answers are grounded in this paper and cite their sources.
We report a systematic study of the early oxidation process in a wide temperature range (20--700?deC) for the Si(111)-(7×{}7) annealed surface. Oxygen uptake data obtained by Auger-electron spectroscopy indicate that the surface oxygen uptake increases and then gradually saturates with O₂ exposure at 20?deC. At intermediate temperatures, there is no saturation and the oxygen uptake rate is enhanced. At 700?deC, the uptake rate is enhanced in the high-exposure (or high-O₂-pressure) region, but is significantly reduced in the low-exposure region, presumably because of oxygen desorption as volatile SiO.Low-energy electron diffraction pattern transition from 7×{}7 to no pattern through a 1×{}1 intermediate pattern seems to be governed only by the amount of surface oxygen at any temperature. Using photoelectron spectroscopy we found that O₂ exposures corresponding to this 1×{}1 pattern result in the disappearance of surface states centered about 0.6 eV below the valence-band maximum. A chemical-shift study of the Si 2p core level shows that not only the oxygen uptake rate, but also the way oxygen bonds to Si is strongly dependent on temperature. Our results show that at room temperature (20?deC) all the Si atoms are oxidized at more or less the same rate, so that the adsorbed oxygen atoms are spread out uniformly over the entire Si surface. At high temperatures, certain Si atoms are being oxidized faster (i.e., have more oxygen atoms bonded to them) than others, so that the distribution of adsorbed oxygen atoms is highly nonuniform; we find microscopic regions of SiO₂-like material interdispersed with regions of Si-like material. Therefore, in the early stage of oxidation, heating the substrate causes nonuniform oxidation with microscopic phase separation of SiO₂ and Si in the oxide film. Tight-binding calculations reported in the literature predict the oxygen 2p orbital to produce these valence states: O {π} (nonbonding), O {σ} ({σ} bonding), and O i (intermediate nature between nonbonding and {σ} bonding). By studying the O i--O {π} peak height ratio as a function of oxidation temperature and referring to theoretical calculations, we again come to the conclusion of nonuniform oxidation at high temperatures.
No takes yet. Share an insight, caveat, or question.
Tabe et al. (1986) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: