Metastable pitting corrosion of stainless steel and the transition to stability

Abstract

The evolution of corrosion pits on stainless steel immersed in chloride solution occurs in three distinct stages: nucleation, metastable growth and stable growth. This paper describes the growth of metastable corrosion pits on stainless steel immersed in chloride solution, and their transition to stability. The rate of growth of individual corrosion pits is controlled by diffusion of the dissolving metal cations from the pit interior, the surface of which is saturated with the metal chloride. This process is independent of electrode potential. Analysis of the diffusion yields a critical value of the product of the pit radius and its dissolution current density (termed the ‘pit stability product’) below which the pit is metastable and may repassivate, and above which the pit is stable. The critical value of the pit stability product for stainless steel in chloride solution is 0.3 A m -1 . All pits, whether metastable, or destined to become stable, grow initially in the metastable condition, with a pit stability product which increases linearly with time, but below the critical value. Metastable growth requires a perforated cover over the pit mouth to provide an additional barrier to diffusion, enabling the aggressive pit anolyte to be maintained. In this state pits grow at a constant mean current density which is maintained by periodic partial rupture of the cover. Stable pit growth is then achieved when the cover is no longer required for continued propagation, and the pit depth is itself a sufficient diffusion barrier; stability is characterized by a constant mean pit stability product above the critical value. If the cover is lost prematurely, before the critical pit stability product is achieved, the pit anolyte is diluted and repassivation is inevitable. In contrast to the growth rate of individual pits, the distribution of pitting current transients is dependent on electrode potential: the pit nucleation site, particularly its geometry, is exclusively responsible for this potential distribution. It is proposed th at shallower, more-open sites are activated only at higher potential and higher current density, and are consequently more likely to achieve stability.