The Role of Oxygen on the Stability of Crevice Corrosion M. K. Sawford, a B. G. Ateya,* ,b A. M. Abdullah,** and H. W. Pickering*** ,z Department of Materials Science and Engineering, The Pennsylvania State University, University Park, Pennsylvania 16802, USA Crevice corrosion can initiate in a spontaneously active metal/electrolyte system even with some appreciable oxygen concentration in the crevice electrolyte, as indicated by the measured instantaneous milliampere current and active E x=L potential values at the bottom of the crevice. However, for crevice corrosion to occur, the oxidant concentration has to be below a certain level, set by the metal/electrolyte system and the experimental conditions. Similarly, the oxidant concentration needs to be increased to some higher level to terminate an on-going crevice corrosion process. Reinitiation of crevice corrosion on a passive crevice wall is much more difficult than the original initiation. This is because the passive crevice wall provides a small ionic current which generates only a modest IR voltage that cannot shift the potential far enough on the crevice wall to reach the active region. Oxidant reduction at the anodic sites on the crevice wall resulted in a significant reduction in the ionic current, I , and hence in IR. This decrease in IR caused E x=L to increase to a value in the passive region which resulted in passivation of the entire crevice wall and termination of the crevice corrosion process. Conversely, removal of the oxidant caused I to increase and E x=L to shift negatively back to the active region resulting in reactivation of crevice corrosion. © 2002 The Electrochemical Society. DOI: 10.1149/1.1470655 All rights reserved. Manuscript submitted April 6, 2001; revised manuscript received December 10, 2001. Available electronically April 12, 2002. It is generally believed that oxygen depletion may be a prereq- uisite for the initiation and stabilization of crevice corrosion. 1-4 Within this context, Yao et al. 5 found that by increasing the concen- tration of dissolved oxygen in the electrolyte within a crevice in titanium, the incubation time increased. In an earlier work, Ruskol and Klinov 6 found that oxygen depletion increased with time, depth into the crevice, and decrease in the crevice gap dimension. Crevice corrosion is characterized by a highly constricted geom- etry, which hinders the transport of the ionic species in and out of the crevice, and leads to the separation of the anodic and cathodic reactions and resulting large IR voltages, and to changes in the com- position of the crevice electrolyte, e.g., localized acidification and chloride ion accumulation. The theory of achieving a critical com- position within the crevice electrolyte, e.g., localized acidification, attributes the initiation and stabilization of crevice corrosion to the breakdown of the passive film on the crevice wall. 2,3 On the other hand, the IR theory stipulates that crevice corrosion is initiated and stabilized when the electrode potential, E ( x ), at some distance into the crevice is in the active region of the polar- ization curve that exists on the crevice wall. 7 An induction period is often required to achieve this condition, i.e., the active peak is ini- tially too small, or nonexistent as in spontaneously passive metal/ electrolyte systems. During this induction time, the composition of the crevice electrolyte needs to change in the direction of destabi- lizing the passive film and forming, and increasing the size of, an active peak in the polarization curve of the crevice electrolyte. The induction period ends and crevice corrosion starts when the IR volt- age associated with the ionic current flowing through the crevice electrolyte places the bottom of the crevice in the potential region of the growing active peak. 7 Thus, when the active peak reaches the critical size for the given crevice aspect ratio, solution resistivity, and polarization conditions, crevice corrosion begins and is stabi- lized by the IR voltage. The latter maintains the active region of the polarization curve on the crevice wall as described elsewhere. 7 Acidification and/or chloride ion accumulation in the crevice elec- trolyte initiates and enlarges an active peak for many spontaneously passive systems. This IR theory 7 has been adopted by others to describe the localized corrosion behavior of spontaneously passive systems. 8,9 In the case of spontaneously active metal/electrolyte systems, for conditions in which the aspect ratio, AR, of the crevice is greater than the critical aspect ratio, AR c , crevice corrosion occurs imme- diately upon polarizing the outer surface of the sample into the passive region. 10 Thus, no change in composition of the crevice electrolyte is needed for the start of crevice corrosion in these cases. This immediate onset of crevice corrosion when AR  AR c was experimentally demonstrated in several spontaneously active systems. 11-15 Since the induction time was zero in these experiments, acidifi- cation or chloride ion accumulation could not occur in the crevice electrolyte prior to the start of crevice corrosion. Furthermore, a specially designed cell/procedure was used that maintained either the pH or the entire crevice electrolyte at essentially the same pH or composition as the bulk electrolyte, allowed visual observation of the crevice wall, and permitted measurement of the E ( x ) profile. 11-15 As a result, it was also determined that crevice corrosion continued indefinitely in the region of the crevice wall where the E ( x ) poten- tial was in the active region of the polarization curve existing on the crevice wall. From these proof-of-concept experiments, it was clear that crevice corrosion was solely caused by the IR voltage in these spontaneously active systems when AR  AR c . The objective of this paper is to provide an additional critical test of the IR mechanism of crevice corrosion in spontaneously active systems and to investigate the role of oxidants in the crevice corro- sion process. We are presenting here the results of novel experi- ments in which various oxidants were suddenly added to the crevice electrolyte in order to quickly shift the E ( x ) values within actively corroding crevices from the active to the passive regions, and to see if the crevice corrosion process was simultaneously terminated. The experiments were designed to simultaneously measure the potential at the bottom of the crevice, E x =L , and the crevice corrosion ionic current, I , flowing out of the crevice, before and after an oxidant was added to the crevice electrolyte. The results of these measure- ments also have an important bearing on the questions of oxygen depletion and the separation of the anodic and cathodic partial reac- tions as a requirement for the onset of IR-induced crevice corrosion. Experimental The iron 99.95% sample was placed into a Teflon mount so that only two of its surfaces crevice wall: 5  10 mm, and external surface: 5  20 mm were exposed to the electrolyte. These sur- faces were polished with 0.5 m alumina to a mirror-like finish and * Electrochemical Society Active Member. ** Electrochemical Society Student Member. ** Electrochemical Society Fellow. a Present address: Crucible Research, Pittsburgh, PA. b Present address: Department of Chemistry, Faculty of Science, Cairo University, Egypt. z E-mail: pick@ems.psu.edu Journal of The Electrochemical Society, 149 6 B198-B205 2002 0013-4651/2002/1496/B198/8/$7.00 © The Electrochemical Society, Inc. B198