Abstract
Localized corrosion sites or pits form on stainless steels exposed to atmospheric conditions when a defect forms in the passive layer, leaving the material susceptible to corrosion and the development of local acidified environments. These pits can continue to propagate and serve as initiation sites for cracking, which can then cause failure of the material. A method to predict the maximum pit size that could develop on an alloy under certain environmental conditions is therefore desirable for the many applications of stainless steels. Recently, a computational model has been developed that can predict the maximum pit radius by analyzing the localized corrosion site as a galvanic couple, with the pit as the anode and the surrounding surface as the cathode. The amount of current that is supplied by the cathode, the amount of current necessary to maintain the anode, and the ohmic drop between the anode and the cathode all control stability of the localized corrosion site and place a limit on the size to which a pit can grow. Predicted model values have compared well to literature values of outdoor exposures out to twenty-six years.
The goal of this study was to evaluate the computational model by comparing predicted model values to pit depths observed on laboratory exposures, in which environmental conditions, such as relative humidity and temperature were controlled. In conjunction, the sensitivity of the model to several input parameters also was investigated.
Exposures of four ferrous alloys 304L, 316L, Custom 465, and Aermet with loading densities of 240 and 600 μg/cm2 deposited sodium chloride and relative humidity 90 and 95% were performed. After a one year exposure, pit sizes were < 2 um for the stainless steel alloys. On the Aermet exposures, large pit sizes > 200 μm were observed along with the occurrence of general corrosion.
To enhance the pitting process stainless steel alloys 304L, Custom, and 316L were studied with deposited ferric chloride and exposed to three relative humidity values of 98, 85, and 64%. The geometry of the thin electrolyte was studied by using several deposition methods. Smaller pit depths were found under drops of ferric chloride, while pit depths closer to the predicted pit values were observed under thin films. With decreasing relative humidity, general corrosion appeared on the 304L samples along with localized corrosion.
The input parameters examined include the pit stability product, the Tafel slope, the corrosion potential, the repassivation potential, and the deliquescence properties of the mixed salt. The repassivation potential and the deliquescence properties of the mixed salt were found to have the greatest effect on predicted model values.
