Chromium Depletion around Sulfide Inclusions in
Stainless Steels Remains a Mystery
K. Chandra and Vivekanand Kain
Materials Science Division, Bhabha Atomic Research Centre, Mumbai 400 085
Stainless steels are used in many industries due to their corrosion resistance. Although they have extremely good general corrosion resistance, they are susceptible to localized corrosion such as pitting. Many contested models for pitting corrosion exist, but one undisputed aspect is that manganese sulfide (MnS) inclusions play a critical role. Majority of pitting events are found to occur at, or adjacent to, such second phase particles. It is commonly accepted that pits in stainless steel (SS) are initiated by either chemical or electrochemical dissolution of the sulfides to form an aggressive species that accelerates localized attack.
Research work in the last few years indicated that the chemical changes in and around sulfide inclusions take place during the fabrication of the stainless steel itself. When the steel is poured, the metal solidifies before the sulfide inclusions do, and as it cools, the composition of the inclusion itself and the metal zone around it will change. It was proposed (Ref 1) that the replacement of Mn by Fe and Cr within the sulfide would occur. This was postulated to lead to formation of a Cr depleted zone around the inclusion. So far, there had been no data to support such a model. This is primarily because of the difficulty of making analytical measurements with the required resolution.
Recently, Ryan, et al. (Ref 2) used focused ion beam (FIB) with secondary ion mass spectroscopy attachment (SIMS) to provide the characterization of the material chemistry as a function of the proximity to the inclusion. The specimen was type 316F SS (high sulphur stainless steel with a high inclusion content). It was polished up to 0.05 μm Al2O3 or 0.25 μm diamond, with careful ultrasonic cleaning between each stage of polishing. A reduction in the Cr:Fe ratio as a function of distance from the matrix to the edge of an inclusion was observed at several inclusion locations on the sample surface. A rather wide Cr depleted zone (about 0.5 μm) was found around the MnS inclusions. The alloy next to the inclusion had a Cr content as low as 10 at.%. Ryan, et al. suggested (Ref 2) the following sequence of events for pitting corrosion due to sulfide inclusions:
1) composition changes are induced in and around inclusion during the processing of steels from melting temperature leading to formation of Cr depletion zone around the inclusions,
2) the Cr depleted zones provide rapid, high-rate metal dissolution that causes a narrow trench to form at the inclusion edge and that changes the local environment,
3) in this environment, the inclusion is itself unstable to dissolution,
4) sulfur products form a crust around the former inclusion providing an occluded environment of altered chemistry in which the steel matrix is unstable,
5) stable pitting ensues
This opened up a possibility of making stainless steels resistant/immune to localized corrosion. It was proposed that heating at a low and suitable temperature would heal or erase out the chromium depletion around the sulfide inclusions making the stainless steel resistant to pitting corrosion. However, presence of chromium depletion around sulfide inclusions is not a commonly accepted feature.
An investigation by Meng, et al. (Ref 3) was undertaken to confirm the presence of chromium depletion zone around sulfide inclusions using different characterization techniques. They investigated several different materials including a commercial grade of type 304SS, type 316F SS in the as-cast and annealed conditions, and in collaboration with Ryan and Williams, the same piece of type 316F used in their study. They characterized the regions between MnS inclusions and the matrix in the above steels using high resolution scanning transmission electron microscopy (STEM) and line profiling, in combination with FIB sectioning and SIMS mapping. STEM line profiling for Mn, S, Cr, Fe & Ni were performed across the MnS/matrix interface in horizontal, diagonal and vertical directions. The Cr/Fe intensity ratios for these three line scans were also measured. The probe size was < 2nm with a step size of 10nm. The Cr/Fe ratios were higher in the sulfide inclusions than in the matrix region where it was ~0.3. No trend of a decrease in the Cr/Fe ratio near the inclusion was evident. The line profile of the individual elements also revealed a compositionally sharp interface of the MnS inclusion and matrix, which indicated that no Cr depletion zone existed around the MnS inclusion in 304SS. SIMS mapping was also performed on a different MnS inclusion using the same parameters as used by Ryan et al. The lateral resolution of the SIMS mapping was about 150nm. Several 10 μm by 10 μm regions were mapped using both positive and negative modes to detect positive and negative secondary ions respectively. These observations further indicated no Cr depletion zone around MnS inclusions in type 304SS. However, in some cases a manganese oxide (MnO) particle was observed near the MnS inclusion. Same results were observed in the case of as-cast and annealed 316F SS i.e. no Cr depletion around the MnS inclusions. This was true for the sample used by Ryan et al also. In this case, a majority of the sulfide inclusions were connected to oxides. Energy dispersive spectroscopy (EDS) line scans for the elements were performed along the line that crosses the boundary between the MnS inclusion and the matrix. The results were identical to that of the analyses performed on sulfide inclusions from the other SS described above i.e. no Cr depleted zone. Another scan was made along the line, from the sulfide, through the oxide, and then into the matrix. The resulting Cr profile had a Cr depleted region, but this was the effect of the oxide particle. The Fe content in the oxide decreased more than the Cr content relative to the bulk concentration, resulting in a higher Cr/Fe concentration ratio in the oxide. The depleted zone reported by Ryan, et al. cannot be rationalized by the effects of oxides adjoining the sulfide particles.
The as-cast type 316F sample and the sample 316F used by Ryan, et al. had vastly different Mn contents, which might have affected the interaction between the Cr in the alloy and the sulfide. However, findings by Meng, et al. indicated (Ref 3) that MnS inclusions in neither alloy exhibited Cr depleted zones. Therefore, Mn content is not critical to the formation of these zones. In a recent response (Ref 4) to the findings of this study, Ryan and Williams indicated that only about 20% of the 25 inclusions they studied from the type 316F sample exhibited a Cr depleted region and about 20% exhibited no depletion. The data from the rest were complicated by sputtering of the inclusion itself. They suggested that this variation in behavior might explain why all MnS inclusions do not generate pits. However, Meng, et al. (Ref 3) studied 17 inclusions from three alloy sources by STEM and SIMS, and no evidence of Cr depletion was found.
The presence of a chromium depletion region around a sulfide inclusion remains a controversial topic. If it is shown conclusively that there is chromium depletion around such inclusions, the possibilities of improving the corrosion resistance of stainless steels are immense. However, it has been commonly accepted that sulfide inclusions in stainless steels get attacked (dissolved) whereas adjacent regions of such inclusions in steels are the sites where pitting corrosion attack initiates. Sulfide inclusions in the form of stringers have also been shown to be responsible for other forms of corrosion e.g. intergranular corrosion in applications involving nitric acids (Ref 5). Establishing the presence/absence of a chromium depletion zone around these inclusions in stainless steels would change our understanding of localized corrosion.
REFERENCES
1. D.E. Williams, Y.Y. Zhu: J. Electrochem. Soc., Vol. 147, 2000, p.1763
2. M.P. Ryan, D.E. Williams, R.J. Chater, B.M. Hutton, D.S. McPhail: Nature, Vol. 415, 2002, p. 770
3. Q. Meng, G.S. Frankel, H.O. Colijn, S.H. Goss: Corrosion, Vol. 60, 2004, p. 346
4. M.P. Ryan, D.E. Williams: Nature, Vol. 424, 2003, p.390
5. V. Kain, S.S. Chouthai, H.S. Gadiyar: Br. Corr. J., Vol. 27, 1992, p. 59