X-ray spectroscopy of the oxidation of 6H-SiC(0001)

Abstract

Chemically abrupt ${\mathrm{SiO}}_2/6\mathrm{H}-\mathrm{S}\mathrm{i}\mathrm{C}\mathrm{}\left(0001\right)\mathrm{}$ interfaces have been obtained after thin (≅5 nm) oxide thermal growth of the Si-terminated face. The originality of the paper resides in a simultaneous use of standard technological conditions for this oxidation (dry ${\mathrm{O}}_2$ oxidation at 1 atmosphere and 1000 °C) and in control and characterization of the starting substrates taken from in situ, ultrahigh vacuum-surface techniques. The oxidation of $3\times 3$ Si-rich and $6\sqrt{3}\times 6\sqrt{3}R30\mathrm{}°$ C-rich reconstructed surfaces is compared. In both cases no C-C bonds or C enrichment are observed at the ${\mathrm{SiO}}_2/\mathrm{S}\mathrm{i}\mathrm{C}$ interface by x-ray photoelectron spectroscopy $\mathrm{C}1s$ analysis, the C-C bonds of the initially graphitized C-rich surface being removed during the first nanometer oxide growth. The interfacial suboxide components with binding energy between the substrate and ${\mathrm{SiO}}_2\mathrm{}\mathrm{Si}2p$ features remain below our detection limit. X-ray photoelectron diffraction analyses indicate that, below the oxide layer, the ordering and polarity of the 6H-SiC(0001) are maintained. These results allow us to conclude that there is probably no fundamental contraindication to obtain nearly ideal ${\mathrm{SiO}}_2/\mathrm{S}\mathrm{i}\mathrm{C}$ interfaces on flat SiC terraces. Two growth regimes are observed in the oxide growth kinetics. They are discussed to determine the possible reasons of carbon exodiffusion or previously observed C enrichments at the ${\mathrm{SiO}}_2/\mathrm{S}\mathrm{i}\mathrm{C}$ interface.