(Invited) Complementary Experimental and Modeling Approaches to Elucidate Critical Conditions for Pit Stability and Repassivatio

Abstract

Robust life prediction models for metallic assets exposed to corrosive environments generally employ a maximum damage size criterion as a design parameter. Such models are particularly useful for use in systems such as compressor blades on carrier-based aircraft, offshore platforms, and radioactive waste storage containers1. Stainless steel, which is often used in these cases, is resistant to general corrosion in such environments, but susceptible to localized corrosion due to the galvanic couple that occurs between the actively dissolving pit and the cathode surrounding it. These galvanic characteristics allow for the critical conditions for each electrode to be analyzed separately as functions of corrosion geometry and then combined through the conservation of charge to determine the maximum damage size2,3. It is imperative therefore to accurately identify and measure the critical conditions for both the anode and the cathode, so that the resulting prediction correctly estimates the maximum damage size for a given system. Galvele’s stability criterion4, which describes the steady-state flux arising out of the competition between metal dissolution and diffusion in the pit, governs the anodic stability conditions, and expresses the anodic current density as a function of the pit depth. For the cathodic reaction, the physical chemistry of the electrolyte and details of the immersion conditions (fully immersed or thin film) specify the constraints leading to the assessment of ohmic drop and cathodic kinetics on the metal surface. When combined, these expressions permit the estimation of the maximum size a pit that initiates in this system is likely to grow to, its propagation limited by the cathodic current available to support it. A comprehensive, quantitative electrochemical framework that unifies the critical conditions describing pit stability and repassivation necessitates a relationship between the measured parameters to the chemical conditions at the corroding surface. Diffusion modeling of a one-dimensional pit employed for this purpose shows that the concentration of the cation at the corroding surface can be expressed as a fraction of the concentration at saturation in terms of time and pit depth. Correspondingly, the time to dilution of the surface from 100% saturation to any fraction f can also be calculated from this expression. When this time variable is compared to the experimental time between the surface states of the salt film and repassivation and then juxtaposed with the minimum depth at which the repassivation potential reaches a constant plateau value, an estimate of the critical surface concentration is obtained.5,6 The artificial pit can also be utilized to determine the pH associated with the critical surface concentration. Rapid anodic kinetics were performed at different surface concentrations as determined by the time to dilution from model calculations following the attainment of a predetermined pit depth. These experiments provide data that show how the dissolution kinetics change as the surface concentration approaches the critical value, where a distinct active-passive transition is observed. Tafel extrapolation of the anodic kinetics at this critical surface concentration in combination with cathodic kinetics for the hydrogen evolution reaction provides an estimate of the local cathodic reaction’s contribution towards repassivation. A map of the local cathodic reaction fraction versus pH can be obtained from an analysis of the varying degrees of hydrolysis to estimate the pH at the critical surface concentration. Such an analysis permits an estimate of the critical pH as the pit transitions from active dissolution to repassivation. Chemical simulations based on solution thermodynamics indicate that repassivation at the critical pH occurs as a result of the nucleation of specific oxides, which provide quantitative support to the view formulated by Okada7 and Sridhar et al.8 This presentation will describe the quantitative framework, the experimental and modeling results that support it, and its utility for the long-term prediction of pitting damage.