J. E. Castle

THE INITIATION OF PITTING CORROSION AT MNS INCLUSIONS

Abstract The dissolution of MnS inclusions in stainless steel exposed in 0.5 M NaCl + 0.5 M H 2 SO 4 + 0.08% H 2 O 2 at the free corrosion potential for 10 s has been studied using simultaneous Auger/EDX analysis and imaging and electron microscopy. Inclusions at different stages of dissolution have been examined. As dissolution of the inclusion progressed, Auger/EDX imaging showed the resulting cavity to have a local solution composition rich in MnCl 2 and the exposed bare metal surface was found to repassivate. It is proposed that metallic corrosion and stabilisation of pit growth is critically dependent on the precipitation of a MnCl 2 salt film on to the exposed metal surface preventing repassivation. A critical concentration of MnCl 2 within the cavity must be attained before salt film precipitation will occur. If the critical MnCl 2 concentration is not reached during dissolution of the sulphide inclusion, repassivation of the cavity will result. The probability of salt film formation and consequent pit propagation will be dependent on the size and geometrical shape of the inclusion.

Studies by auger spectroscopy of pit initiation at the site of inclusions in stainless steel

Abstract In this paper a high resolution scanning Auger microscope fitted with an X-ray analysis system is used to investigate the surface chemistry of inclusion groups in stainless steel 316 relative to the bulk composition. Changes in the inclusions resulting from corrosion in sodium chloride solution are observed over a sequence of several successive exposures. The work demonstrates that careful control of experimental conditions allows the progress of initiation of pits to be observed systematically. The role of inclusions as initiators of pitting corrosion has been studied from the very inception of attack. Detailed observations have been taken from separate manganese suphide as well as mixture of manganese sulphide and multi-element oxide at various pH values of solutions to trace the behaviour of these inclusions during exposures. It is suggested from this investigation that corrosive attack starts on MnS inclusions or on small MnS precipitates associated with oxide inclusions. The observations showed that sulphur precipitated on the area surrounding the inclusion group in acidic NaCl solution. It is believed that this deposition stems from the cathodic reduction of a transient oxidised species of sulphur produced during the acid hydrolysis of the MnS. Following the deposition of sulphur the formation of an iron oxide cap to the pit and inclusion is observed. Deposition of sulphur does not occur in alkaline solutions.

The use of in the x-ray photo-electron spectroscopy analyses of passive layers on stainless steel

The repassivation of a commercial grade 18% Cr 8% Ni austenitic stainless steel in low temperature, neutral, de-oxygenated water was studied by XPS, obtaining composition profiles by argon ion milling. Evaluation of the technique revealed negligible effect of sample transfer through air from the spectrometer to the corrosion cell, but photo-electron attenuation and reduction in peak definition resulted from organic carbon pick-up both from water and spectrometer vacuum. The passive layers contained chromium-rich and iron-rich areas. The Cr3+, Fe2+ and Fe3+ cations are incorporated in layers containing both OH− and O2− ions. The unreacted nickel of the alloy showed an attenuation of the 2p32 signal with exposure which presented a unique opportunity to calculate film thickness. Increases in refreshment rate (1, 3 and 6 ml/h) at 313 K and 353 K respectively increased the film thickness (1.5–3.0 nm); possibly due to increased carbon accreation, incorporating hydrogen-bonded water into the film. The outer 33% of the film revealed (Fe2+ + Fe3+)/(Cr3+) corresponding to alloy proportions, suggesting no overall selective oxidation or dissolution.

Curve-fitting in XPS using extrinsic and intrinsic background structure

Abstract It has been shown by Tougaard that the extrinsic background to a photoelectron peak is very small at the position of the peak. Most of the background step generated by the peak arises from other forms of energy loss which we have suggested is intrinsic to the photo-excitation process. We have defined this intensity, using a specific modified Shirley background in the immediate vicinity of the peak, and hence fixed it for a given element and chemical state by means of a shape parameter. In this paper we describe an examination of the passive film on stainless steel by ARXPS in which the fixed intrinsic backgrounds are used in fitting oxide and metal for iron and chromium and also for oxygen and carbon. By using peaks of identical shape, position and background for each angle it becomes easy to spot the behaviour of the residual extrinsic background, which as expected depends on angle. In this work we demonstrate the use of a polynomial function, that approximates the Tougaard background, included in the fitting process. This inclusion allows to define a parameter of the polynomial which represents the background slope and use this to assign different background tails to each component in a multiple peak. The behaviour of the extrinsic background can, by this means, be compared with the changing ratio of peak areas as a function of the take-off angle. This work on the behaviour of the extrinsic component of the background within one or two peaks widths of the peak enables surface structures to be placed in depth-wise order with only a small extension to the peak-fitting routines that are normally employed in X-ray photo-electron spectroscopy. We conclude that within the background close to the peak there lies information on the layer thickness and that this might be developed into a reliable method for the estimation of overlayer thickness.

The application of X-ray photoelectron spectroscopy to the study of polymer-to-metal adhesion

The nature of polybutadiene to mild steel adhesion has been characterized by X-ray photoelectron spectroscopy (XPS) used in conjunction with a novel oxide stripping/ argon ion bombardment technique for interface analysis. It is concluded that the interface is more correctly described as an interphase region. The failure mode of the polybutadiene/steel couple has been determined for mechanical delamination, and corrosion induced failure; both at the free corrosion potential (FCP), and when cathodically protected. Cohesive failure of the polymer occurred in the two former instances, but when cathodically protected the film is observed to separate in two stages. These results are discussed in relation to underfilm pH and the possible importance of oxide reduction to coatings deiamination from cathodically protected structures.

XPS study of the adsorption of ethoxysilanes on iron

The uptake by polished iron surfaces of two R-ethoxysilanes, used commercially as adhesion promoters, has been studied by XPS. It is concluded that the presence of an active group in the organic radical may influence uptake. However, the presence of unsuspected impurities in the solvent used in preparation of the test media also had an important effect. Subsidiary data on the adsorption of butylamine from the gas phase which yield values for the escape depth of electrons in organic materials are also reported.

Chemical state information from the near-peak region of the X-ray photoelectron background

Abstract Following the work of Tougaard [Appl. Surf. Sci. 100/101 (1996) 1] it is now generally accepted that the contribution to the energy loss background of a photoelectron peak arising from the transport of electrons through the solid is very small at the position of the peak. Nevertheless, the sharp rise in background at the peak, as originally defined by Shirley [Phys. Rev. B 5 (1972) 4709] is very real and is probably the most easily recognised feature of the X-ray photoelectron spectrum. To undertake quantification in XPS analysis it is normal to remove a background which is based on the original Shirley algorithm, i.e. an integration of the peak lying above background using an integration constant which allows the background to merge with the experimental data at some point beyond the peak. At this point in the spectrum the extrinsic loss, defined by Tougaard, has become significant and varies with the thickness of any overlayer which might be present, including any surface contamination. Thus the Shirley scattering parameter, as the constant of integration has been called, contains a contribution from extrinsic losses and a contribution from the rise in background at the peak, which for clarity we have referred to as the intrinsic loss. We have developed a method to obtain a value for the intrinsic part of the background, defining it by a parameter, κ. The value of κ varies systematically from one element to another and from one chemical state to another. This behaviour is shown to provide an additional basis for interpretation of XPS spectra

The initiation of pitting corrosion of stainless steels at oxide inclusions

Abstract The initial stages of pit growth for stainless steel exposed in 0.5 M H 2 SO 4 + 0.5 M NaCl + 0.08% H 2 O 2 at the free corrosion potential for 10 s have been examined by Auger and simultaneous EDX mapping and electron microscopy. This investigation was focussed on pits which initiated at mixed Al/Ti/Mn/Cr oxide inclusions. Auger/EDX maps showed the strong presence of Al in corrosion deposits partially covering the pit, and traces of Mn and Ti in deposits adjacent to the inclusions, indicating that these elements have been leached from the inclusion. The reduction of pH within the pit has resulted in partial dissolution of the generally inert oxide inclusion. Activity-pH diagrams have been used to interpret the dissolution and redeposition of the oxide inclusion components.

Peak fitting of the chromium 2p XPS spectrum

Abstract An XPS investigation of loss feature and peak shape analysis of the chromium 2p region has been performed. The approach is based on the use of a modified Shirley background model and requires the estimation of an individual background for each component of the photopeak under examination. The results obtained from the analysis of clean and oxidised chromium metal and pure Cr 2 O 3 , notwithstanding possible shortcomings in the model, give evidence for the importance of a proper consideration of satellite structures in the peak synthesis of Cr 2p XPS spectra. The results also have important ramifications in the peak fitting of the 2p spectra of other d-band transition metals such as manganese, iron and cobalt.

A co-ordinated study of the passivation of alloy steels by plasma source mass spectrometry and x-ray photoelectron spectroscopy—1. characterization of the passive film

Samples of Fe17Cr and Fe15Cr4Mo alloy steels, prepared for use in a European “round robin” on surface analysis of passivating films, were passivated in 0.1 M H2SO4 at different potentials. The passivating solutions were analysed by Inductively Coupled Plasma Source Mass Spectrometry (ICP-MS), showing that Cr and/or Mo are significantly depleted in the solutions and therefore the enrichment of these elements should be expected in the passive films. The excess charge over the amount equivalent to the elements dissolved in the solution is related to the formation of ions precipitated within the passive film and hence to its thickness. Subsequent analysis of the passive films by X-Ray Photoelectron Spectroscopy gave results in very good agreement with the deductions made from solution analysis, showing that a complete characterization of the passivation process has been achieved.