The Electrochemistry of Copper Release from Stainless Steels and Its Role in Localized Corrosion

Abstract

In this investigation, the role of copper in MnS dissolution was studied in a series of lab-made austenitic stainless steels (SS) with varying Cu content. The base composition of these samples was that of SS 303 and the Cu content was varied between 0.02 and 0.80 wt%. In potentiodynamic polarization experiments, it was found that Cu deposition passivated the MnS inclusions in all except the 0.02 wt% specimen. The critical potential for this passivation, from potentiostatic experiments, was found to be associated with the onset of metastable pitting. The "apparent" pitting potential in the specimen with 0.02 wt% Cu content was approximately 200 mV more negative than the other specimens. This apparent potential was attributed to MnS dissolution, due to a lack of Cu deposition/passivation, and not pitting. With respect to pit repassivation, at concentrations equal to and greater than 0.2 wt% Cu, repassivation potentials were on the order or +0.10 V SCE. In comparison, for the 0.02 wt% Cu specimen, the repassivation potential was less than the OCP (−0.125 V vs. Ag/AgCl) indicating Cu reduction inside the pit plays a role in the measured repassivation potential. Cu release was quantified using a rotating ring disk electrode. In these experiments, oxidation peaks for Cu(I) and Cu(II) were detected. In these experiments Cu was released at low potentials and low, passive, current densities but the resulting near surface concentrations of Cu(I) (0.02 mM) were insufficient to passivate MnS. Dissolution of manganese sulfide (MnS) inclusions is frequently reported as one of the key steps in the pitting corrosion of austenitic stainless steels (SS). MnS is less noble than the bulk stainless steel surface and more prone to oxidation in chloride media. Recently, proposed mechanisms describing the role of MnS in the stainless steel pitting process have focused on preferential matrix dissolution at MnS/stainless steel boundary. An investigation performed by Chiba et al. with the use of high resolution optical microscopy in a micro-electrochemical system, imaged trenching at the MnS/SS boundary during anodic polarization of the specimen in chloride media. 1 Real-time observations clearly revealed initiation of metastable and stable pits from these trenches. In another study carried out by Chiba et al., selective boundary dissolution was attributed to a synergistic effect of chloride ions and elemental sulfur. 2 It was confirmed that no composition change or anomalous phase existed at the boundary of inclusions and stainless steel. Lillard et al. found that trenches at the MnS/SS boundary only occurred after ennoblement of the inclusion due to copper (Cu) deposition. 3 They proposed pitting at the MnS/stainless steel boundary occurred in a sequence of four steps. The most important step in this process was Cu deposition on MnS inclusions, resulting in the inclusions acting as local cathodes thus accelerating localized corrosion at the MnS/SS interface much as it occurs in other alloys as discussed below. In a complementary study, Zhou et al. focused on atomic scale interaction between alloying copper and MnS inclusions by in-situ ex-environment transmission electron microscopy (TEM). 4 Energy dispersive spectroscopy (EDS) analysis of the MnS inclusion revealed copper enrichment especially at the periphery. Further analysis revealed that the composition of the layer was Cu 2 S. Based on high resolution TEM observations, a high density of vacancies was found at the Cu rich region, and it was proposed that this could provide a fast diffusion path for Cu. Further, the proposed mechanism for Cu deposition was: 1-Replacement of Mn 2+ by Cu + ions based on a cation exchange reaction, which is feasible due to the same crystal lattice and high solubility difference of Cu 2 S and MnS. 2-Dissolution of MnS followed by the release of S and subsequent formation and deposition of Cu 2 S on the inclusion surface, which was proposed by Sourisseau et al. 5 The presence of Cu on the surface of stainless steel accompanied by sulfur has been observed by other investigators including Zakipour * Electrochemical Society Member. z E-mail: lillard@uakron.edu and Leygraf, who investigated surface composition of AISI 304 and 316 stainless steel by means of Auger Electron Spectroscopy (AES) in chloride media. 6 They detected Cu 2 S at pH between 4-8 and elemental sulfur at pH = 1. Copper enriched MnS inclusions have also been reported by others though they proposed that a protective layer of Cu 2 S on the inclusions decrease in pitting rate. 7,8 Flower-like Cu compound deposition on MnS surface was imaged using SEM/EDS by Ke and Alkire. 7 Hermas et al. proposed that addition of 2% Cu to stainless steel decreased corrosion rate but polarization measurements failed to clarify its effect on passivation and pitting corrosion. 9 Copper dissolution during pitting corrosion is not unique to stainless steels and has been observed for other alloy systems including Al-Cu, Al-Cu-Mg, and also Al-Cu-Li alloys. 10-12 Pits that form on Al-Cu and Al-Cu-Mg alloys are frequently associated with Cu deposits that result from dealloying of Al 2 CuMg. 13 Although Al 2 CuMg is noble with respect to the alloy microstructure, selective oxidation of Al (e.g. dealloying) and subsequent Cu release results in a decrease in the intermetallic compound (IMC) corrosion potential and its dissolution. It has been proposed that during the dealloying process , the IMC coarsens to lower its surface energy resulting in the mechanical release of Cu clusters, which then deposit onto the electrode surface. As the OCP (open circuit potential) of the Al matrix is low relative to Cu, the galvanic interaction between the metallic copper deposit results in secondary pitting at this location. Buchheit et al. investigated the mechanism of copper release from Al 2 CuMg in high-strength Al-Cu and Al-Cu-Mg alloys by means of a rotating ring disk electrode (RRDE) experiments and stripping voltammetry (SV). 11 They compared the results from a pure Cu disk with that from a Al 2 CuMg specimen. The results were discussed within their framework for non-Faradic copper release similar to the Al-Cu system. Non-Faradic release of Cu has also been proposed by Li et al. for Mg-0.5Cu and AZ91 alloys. 14 Accumulation of imbedded noble elements in the outer columnar oxide layer of magnesium and also surface concentration of these noble metals, including copper, were reported by Taheri et al. and Birbilis et al. 15,16 Based on these findings, Li et al. by means of RRDE method, confirmed non-Faradic release of Cu form AZ91. Though Cu deposition on MnS in SS is well known, questions still remain about the mechanism of its release. For example, at what potentials does Cu release from SS occur, does it occur during passive dissolution? Is metastable pitting required? Stable pitting? Does Cu concentration in a SS alloy influence the pitting and repassivation potentials? The goal of this paper is to shed light on these questions.) unless CC License in place (see abstract). ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see