XPS and SIMS Study of Anhydrous  HF  and UV/Ozone‐Modified Silicon (100) Surfaces

Abstract

X-ray photoelectron spectroscopy and secondary ion mass spectrometry are used to study the surface chemistry of silicon single-crystal wafers. Anhydrous HF and UV/ozone processes modify the surface chemistry by altering the concentration and bonding environments of the most abundant surface atoms: Si, O, C, H, and F. Anhydrous HF in conjunction with water vapor [HF(g)] removes the silicon native oxide. Native oxides with initial thicknesses of 1 lfi~, four monolayers, are exposed to a HF(g)-H~O rinse, HF~g), and an aqueous HF-H20 rinse and are reduced to 0.4, 0.2, and 0.1 monolayers, respectively. The carbon concentration exhibits an inverse relationship with the submonolayer oxide coverage. The HF(g) process results in less than a monolayer of silicon fluorides. The silicon fluoride species decrease with air exposure resulting in mostly oxyfluorides after lh. Rinsing the fluorinated surface with H20 reduces the fluorine concentration, removes trace metal impurities initially bonded to the native oxide, and promotes initial S iO2 formation. UV/ozone exposure reduces the hydrocarbon concentration on the HF-etched surface while producing 8A, three monolayers, of SiO2. Exposing the HF(g)-treated surface to UV/ozone removes the silicon fluoride species; only oxyfluorides are observed with the ozoneinduced SiO2. The cleanest silicon surface with respect to metallic and hydrocarbon impurities was achieved with a HF etch-H20 rinse-UV/ozone oxidation process.) unless CC License in place (see abstract). ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 169.230.243.252 ABSTRACT Crystallographic faceting of unmasked monocrystalline SiC (100) thin film surfaces during plasma etching in SF6 has been observed using scanning electron microscopy. Although the faceting produced a rough surface, Auger electron spectroscopy showed it to be very chemically clean, having less native oxide than unetched SiC. The facets were determined to be {210} planes using a combination of selected area channeling and small angle tilt measurements. The facet nucleation appears to be surface sensitive and is caused by either crystal defects or compounds formed by reaction of SF6 with adsorbed gases on the sample and chamber wall. It is proposed that the subsequent facet growth is due to a surface reconstruction that causes the {210} planes to etch slower.

Transfer of InP epilayers by wafer bonding

“…Before bonding the native oxide is removed from the surfaces to have a crystal-to-crystal bonding. HF acid completely removes both Si and InP native oxides [45,46], as was seen also from XPS spectra in this work [47]. However, the hydrophobic bonding (HB) is very weak at room temperature and after low-temperature annealing, because of the hydrogen-terminated surface.…”

X-ray photoemission spectroscopy study of silicidation of Ti on BF2+-implanted polysilicon

“…Figure 4 shows that some or all of the shifted O 1s peaks at 531.6, 532.6, and 533.4 eV can be identified, depending on the annealing temperature. These peaks are associated with TiO x , 40,41 SiO x F y , [37][38][39] and SiO 2 , 20,37,42,43 respectively. Except for TiO x , their intensity increased with temperature up to 600°C, beyond which they decreased with temperature until they became undetectable again at 700°C.…”

Photo-oxidation effects of light-emitting porous Si

“…[38][39][40][41] The SiO x -related peaks posses in the energy range between 102 and 104 eV. 39,40,[42][43][44][45] We can never find such an oxide-related peak in the as-prepared spectrum ͓Fig. 7͑b͔͒; however, the large broadening of the Si 2p peak is evident.…”