Corrosion SIMS profiles

Dynamic SIMS is used for depth profile analysis of mainly inorganic samples. The objective is to measure the distribution of a certain compound as a function of depth. At best the resolution in this direction is < 1 nm, that is, considerably better than the lateral resolution. Depth profiling of semiconductors is used, for example, to monitor trace level elements or to measure the sharpness of the interface between two layers of different composition. For glass it is of interest to investigate slow processes such as corrosion, and small particle analyses include environmental samples contaminated by radioisotopes and isotope characterization in extraterrestrial dust. 

Aqueous corrosion profiles have also been measured on soda-lime-silica glasses using SIMS.(18) Both negative and positive oxygen primary ions were used for sputtering in combination with a metal diaphragm to minimize charging. 

Because hydrogen often is incorporated into corroded glass surfaces, the characterization of H profiles is important to understanding corrosion behavior. SIMS and SIPS are useful for characterizing H in near surface (<1 Jim) layers (e.g., see Reference 23). 

Both of the above sectioning techniques have an important place in the measurement of film thickness. For films that are much thicker than 1 pm, polishing seems to be the preferred approach. However, for thinner corrosion films (even as thin as a few hundred nm) fracturing at cryogenic temperatures can often produce thickness values that are more accurate than depth profiling using AES, XPS or SIMS, and is certainly faster. 

AES depth profiling has often been used to measure oxide film thickness. As indicated above, other cross-section methods exist which may be more accurate and faster. In Fig. 8 the two depth profiles have the position of the oxide/metal interface indicated as I" the position itself is usually taken to be the depth at which the metal abundance becomes >5091. In Fig. 8 there appears to be little doubt about the position of the interface as, so defined. However, in many profiles of significant depth, the interfaces become broadened to the point that the temt "interface" lo.ses its meaning. This is particularly true for films arising from aqueous corrosion. It follows that there is little likelihood of observing significant compositional changes in the inner oxide of such films, and, for that reason, other methods such as SIMS must be used if such information is desired. 

SIMS has not been much used as a tool for studying corrosion. The main reason for this seems to be that SIMS spectra and depth profiles are frequently difficult to interpret in terms of clearly defined diffusion or deposition mechanisms. Spectra from oxides contain a complex (and frequently overlapping) series of molecular ion masses secondary-ion depth profiles are complicated in the oxide and interface region by matrix-effected changes in ion yield... 

SIMS systems capable of imaging can be used to. study corrosion samples in plan view or in cross section. The choice of primary ion should be of one that enhances the detection of the major products Cs is the most suitable since it enhances the production of electronegative species such as O. H and oxide molecular fragments. The beam currents used should be such that the entire oxide thickness can be profiled in a reasonable time typical sputter erosion rates in oxides are 3-5 pm/h for a I pA beam rastered over a 250 pm- area. Ion images collected should reflect the possible oxide combinations, e.g.. FeO for a simple Fe oxide, or FeCrO" for an oxide believed to contain both Fe and Cr. 

SIMS studies of these same films, however, did add useful information. The film thicknesses were too low to be able to use a high-current Cs primary beam for depth profiling. An Ar beam was used instead at low currents, which would not have introduced many chemical perturbations into the film. The results of such profiling are shown in Figs. 19a-c for samples of alloy A after exposures in pH 10 solution at the corrosion potential for periods of 3, 12, and 24 h. The intensities of the oxide secondary ions NiO" and CuO" are shown as functions of equivalent sputter time. The depths profiled in these instances were so shallow that it was impossible to gauge them by profilometry thus only the product of sputter time and current density is given on the abscissa. As before, the RSFs for the two ions are considered to be approximately equal. 

---