Abstract Samples of steel from two different sources were examined. The materials had nominally the same bulk composition but different samples from each batch had been hardened under two slightly different conditions. The surface oxide films were analysed by both Auger emission spectroscopy (AES) and X-ray photoelectron spectroscopy (XPS) after a number of ion bombardments, and the variation in composition with depth was established. In general the outermost oxide layers were chromium deficient and XPS results suggested the presence of Fe 2 O 3 at the surface, together with atmospheric contamination. After ion bombardment the proportion of chromium oxide increased and both iron metal and oxide (Fe 3 O 4 ) were present. The amount of chromium oxide reached a maximum at the steel oxide interface and on further bombardment was replaced by chromium metal. AES and XPS results were in agreement qualitatively and also quantitatively after measurement with a curve analyser of the areas under the peaks of certain elements. The present investigation has shown that AES and XPS can give very similar analyses provided some simple corrections are applied. However, the use of the higher resolution of XPS can provide additional information which cannot be obtained by AES. Thus, using expanded XPS scans, together with suitable curve analysis techniques, it is relatively simple to separate signals from oxide and from free metal regions of the sample, and to study in detail the changes in composition of the oxide from the oxide-air to the oxide-metal interfaces. The study of the four steel samples in this manner has shown that there is a thin region rich in chromium oxide adjacent to the metal. The high level of chromium falls off fairly rapidly nearer the surface and appears to stabilize at a very low level. This is particularly obvious in the thicker oxides (PH), where there is a thick surface layer which is principally iron oxide. It would appear that under the conditions of the heat treatment given to the present samples the thickening of the oxide is due almost entirely to iron oxide. Other work has shown that the low temperature air-formed oxide on these steels is chromium rich and it is suggested that the thickening is due to diffusion of iron through the original film with little movement of the chromium. In view of the high chromium deficiency in the outer layers of the oxides examined it is apparent that the supposedly protective chromium oxide film on chrome steel is situated at the oxide-metal interface and not on the outer surface. The present investigation has shown the existence of two unreported iron satellites. One would appear to be associated with the mixed valence oxide Fe 3 O 4 and another with the purely trivalent oxide Fe 2 O 3 . The satellites could be associated with a strong plasma loss mechanism, or be due to shake-up phenomena as in the case of copper and nickel oxides 14 As might be expected from the normal oxidation behaviour of iron, the present oxide films appear to consist of a thin layer of Fe 2 O 3 overlying the main Fe 3 O 4 oxide. The removal of this outermost Fe 2 O 3 layer is rapid and is accompanied by the changes in the satellites mentioned above, together with changes in the position of the Fe 2p 3 2 oxide peak. In conclusion, it is clear that a detailed XPS examination can not only provide information on overall compositional changes, as can be obtained by AES, but can also provide a comprehensive picture of changes in oxide composition, including those due to oxides of one metal in different valency states.
Read full abstract