The Deal-Grove equation for the oxidation of silicon is examined. The evidence that O2 dissolves in the oxide layer and diffuses through it is examined; if 18O is used, there appears to be no exchange of atoms between the molecule and the network already formed. A distinction is made between interstitial sites, through which the molecule diffuses, and special low energy sites which determine the solubility. The latter are not relevant to the rate of oxidation. For oxidation in steam, the evidence is that the mechanism is similar, H2O diffusing through the network and hydrogen being liberated at the Si/SiO2 interface. The solubility of H2O is determined by the formation of non-bridging OH radicals. This does not affect the rate of oxidation, but does give a mechanism for the exchange of 18O with the network. The work of Jorgensen using a platinum electrode is discussed, and it is suggested that if electrons can pass through or round the oxide film, a charged species, perhaps a silicon vacancy, moves through the oxide. Finally a discussion is given of the nature of charged centres in the oxide layer, which are thought to be responsible for Anderson localization in the inversion layer.
No takes yet. Share an insight, caveat, or question.
Nevill Mott (1981) studied this question.