Abstract
The repassivation of a commercial grade 18% Cr 8% Ni austenitic stainless steel in neutral deoxygenated water at 313, 353, 393, 433 and 473°K was studied by XPS, obtaining composition profiles by argon ion etching. Evaluation of the technique revealed the negligible effect of the sample transfer through air from the spectrometer to the corrosion cell, but photoelectron attenuation and reduction in peak definition resulted from organic carbon pick-up from water and spectrometer vacuum. Where necessary multiplet analysis by curve synthesis was carried out to determine the spectral peak heights of overlapping peaks. Quantitative analysis was based upon spectral peak normalization using experimentally determined sensitivity factors. For the experiments in the temperature range 393 to 473°K, a PTFE corrosion cell with a built-in heater was constructed which was used in a nitrogen filled pressure vessel. XPS analysis revealed three types of passive layer: (1) An outer water-rich contaminent layer (2) A passive film consisting of Cr3+ Fe2+ Fe3+ O= and OH- (3) An interfacial zone consisting of internally oxidised regions of the steel. The outer contaminent layer was most prominent in the 313 to 353°K temperature range. It consisted of water possibly hydrogen-bonded to acreated organic molecules. At 313 and 353°K, the unreacted nickel of the alloy showed an attenuation of the 2p3/2 signal with exposure, which presented a unique opportunity to calculate film thickness. This was only possible at these lower temperatures where no interfacial zones were observed. The film thickness was estimated at approximately 2.0 nm. At the higher temperatures it was possible to consider the etch-time to be proportional to the film thickness. The films formed at 393 and 433°K were found to be of a comparable thickness to those formed at 313 and 353°K. The passive films formed at 313 and 353°K revealed chromium-rich and iron-rich areas. Slight chromium enrichment was found at 353°K. The more comprehensive analysis at 393 and 433°K found evidence of a distinct iron-rich, possibly spinel region in the outer layer, and an inner chromium-rich region, possibly of a corrundum structure. There was no strong evidence of selective oxidation or dissolution in these outer layers. The films in this range of temperature were concluded to conform to the Mott-Cabrera, field-induced growth theory for the low temperature oxidation of metals. The outer layer was thought to correspond to the initial barrier film. The inner layer was thought to be the result of a solid-state reduction/oxidation reaction between chromium metal and magnetite, and the original metal/film interface, resulting in Cr2O3 and iron. Such a reaction could be responsible for the chromium enrichment of films formed at 313 and 353°K. This reaction would precede the internal oxidation of the steel at these higher temperatures where anion diffusion would be sufficient to penetrate these passive films. On passing through breaks in the chromium-rich inner-layer the oxygen and hydroxyl anions were then able to react with the underlying steel, along grain boundaries and surface imperfections, resulting in an interfacial zone. The thermodynamically more favourable direct oxidation of the chromium metal would tend to arrest the former reaction. At 473°K, the higher cation diffusion resulted in considerable film growth, corresponding to approximately 7 and 18 nm after 1 and 3 h respectively. The iron-rich and chromium-rich zones were again in evidence at this temperature. No selective dissolution or oxidation of the outer layers was observed. However, a considerable concentration of nickel ions as well as traces of copper ions were found to be enriched in the outer layers of the passive films. The nickel was thought to be incorporated in the film as NiFe2O4, an extremely stable spinel. The concentration profile of nickel ions was assumed to be a function of the rate of film, growth. The overall film growth appeared to correspond to a logarithmic growth rate. Whereas the thickness of the interfacial zones formed at 393 and 433°K were comparable to the passive film thickness, at 473°K the greater rate of cation diffusion compared to anion diffusion resulted in interfacial zones to a depth of a fraction of the passive film thickness. Throughout this work the suitability of XPS to a study of this kind was evaluated, and various aspects of the analysis have been closely examined including the procedure for multiplet analysis and the experimental determinations of peak sensitivity factors for quantitative analysis.