The Use of Percolation Theory to Predict the Probability of Failure of Sensitized, Austenitic Stainless Steels by Intergranular Stress Corrosion Cracking

Abstract
A quantitative description of the degree of sensitization in austenitic stainless steels has been developed based on percolation theory. This is used to predict the likelihood of failure of sensitized stainless steel by intergranular stress corrosion cracking (IGSCC). The underlying premise is that, for a component to fail by IGSCC, a continuous pathway of susceptible (sensitized) grain boundary facets must exist through the grain structure. Calculation of the percentage of sensitized grain boundaries necessary to form a continuous pathway provides a criterion for predicting IGSCC. The theoretical results, obtained using a computer model is that material with less than 23% of sensitized grain boundaries will not fail by IGSCC, that material with between 23 and 89% of sensitized boundaries will show mixed ductile and brittle failure in a slow strain rate test, and that material with more than 89% sensitized boundaries will show only brittle ntergranular failures. A range of sensitized structures has been examined experimentally using the electrochemical potentiokinetic reactivation method (EPR). An accurate estimate of the degree of sensitization in terms of the percentage of sensitized grain boundaries was obtained using quantitative image analysis on micrographs of specimens from the EPR test. Slow strain rate IGSCC experiments in which the cracking was induced by very dilute solutions of sodium thiosulphate were used to test the predictions of the percolation model. These have demonstrated that the modified examination method provided a more reliable means of determining susceptibility to IGSCC than the conventional interpretation of EPR data, and have confirmed the predictions derived from percolation theory. The critical percentages of sensitized grain boundaries were obtained with sensitization heat treatments at 650°C of about 4 h (for 23%) and 12 h (for 89%).