Surface engineering studies using argon ion bombardment of compound semiconductors including photovoltaic semiconductors

Abstract

Ion bombardment of compounds often leads to dramatic changes in the surface properties of the compounds. This has important implications for industry and for surface analytical techniques. In the latter, low energy (0.5 to 15 ke V) argon bombardment is universally used. In most analytical techniques (such as AES and XPS) ion beam sputtering is used either as a depth profiling mechanism or for cleaning a sample surface from the ubiquitous oxide and contamination layers. Thus, for these techniques, ion bombardmentinduced surface changes will have significant implications for quantitative surface analyses. Compound semiconductors are often used in devices for specialized applications in modern technology. One such field is in photovoltaics where compound semiconductor-based systems have superior characteristics above the elemental semiconductor devices. Ion implantation for doping purposes and surface analytical techniques are commonly employed in these technologies. The latter is used to develop and characterize such devices. Any detrimental effect of ion bombardment is, thus, of paramount importance in these technologies. The theory of quantification of Auger electron spectra is well established and there are different practical methods to do this quantification. This study concentrates on a method based on the introduction of a couple of correction factors together with the Auger transition intensities measured experimentally. These correction factors are dependent on the physical parameters of the system being analysed. One of these correction factors (which is often ignored) is the sputter correction factor. This factor is needed in the analysis of multicomponent materials, where the various components have different sputter rates, leading to a non-stoichiometric composition at steady-state conditions. As this factor is difficult or sometimes unpractical to measure experimentally, a theoretical method must be used to evaluate it. Different methods based on existing sputter models were used to calculate the sputter correction factor for compound semiconductors. In this way, the relevant applicability of the sputter models was tested, by comparing the model predictions with averaged experimental values. The experimental values were taken as representative, as most of them were obtained using internal standards. The comparison procedure thus indicated a sputter model which produced good predictions for compound semiconductors. The underlying mechanism leading to compositional changes can be deduced from this specific model, which in the case of compound semiconductors is purely of a preferential sputtering nature. This model also provided a method of calculating the sputter correction factor. However, this method of obtaining the sputter correction factor is a generalization. For some materials, like lnP, the resulting surface composition is dependent on the ion beam angle of incidence. As this is difficult to calculate, the dependency was experimentally measured for lnP(lO0) by making use of internal standards. This investigation also examined any dependency on the ion species of the noble gas ion group. In general, this dependence proved to be almost negligible. Thus, a specific method of calculating the sputter correction factor for lnP as a function of the ion beam angle of incidence is provided.