Premium
The effect of hydrogen content on silicon coordination in silicon oxynitrides: An Auger parameter study
Author(s) -
Rivière J. C.,
Crossley J. A. A.
Publication year - 1990
Publication title -
surface and interface analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.52
H-Index - 90
eISSN - 1096-9918
pISSN - 0142-2421
DOI - 10.1002/sia.740160125
Subject(s) - analytical chemistry (journal) , silicon oxynitride , silicon , hydrogen , auger , plasma enhanced chemical vapor deposition , chemical vapor deposition , chemistry , ionization , argon , oxygen , coordination number , oxide , ion , atomic physics , silicon nitride , physics , organic chemistry , chromatography
Silicon oxynitride films prepared by plasma‐enhanced chemical vapour deposition (PECVD), with compositions in the range O/N = 0 to O/N = 0.89, and containing amounts of hydrogen in the range H(H + O + N) = 0.07–0.31, have been studied by measurement of the Auger parameter α = E K (Si KLL) – E K (Si 2s). A pure thermal oxide was also included. Depth profiles using 4 keV argon ions showed that the films were uniform in both composition and silicon coordination. Peak widths of the 2p, 2s and KLL features were measured as a function of oxygen fraction O/(H + O + N); in contrast to the films prepared by low‐pressure chemical vapour deposition (LPCVD) studied by Rivière et al. , the widths of the 2p and 2s peaks were constant. However, the width of the KLL Auger peak went through a maximum near O/(H + O + N) = 0.5, as observed previously. The 2p and 2s width invariance is interpreted as being due to the presence of hydrogen, in that the weak hydrogen bonding allows strain due to varying bond angles and distances to be taken up more easily than in SiO and SiN bonds. The KLL variation, on the other hand, is suggested as arising from the effect of the core hole in the L shell, causing instantaneous local strain around the ionized Si atom associated with SiO and SiN bonds. The dependence of the Auger parameter on the oxygen fraction O/(H + O + N) was different from that found for the LPCVD films in that it was accurately linear. As a result, the PECVD films do not obey the random bonding model of progressive substitution of N by O, but consist of a mixture of two phases, in agreement with Claassen and Kuiper who suggested the phases SiO 2 and Si x N y H z . The linear decrease in the Auger parameter is thus due to the proportionate decrease in the Si x N y H z content. Plots of the Auger parameter as a function of ( n 2 − 1)/ n 2 were again linear, confirming the direct relationship between the parameter and the polarization energy found previously.