Prediction of the pitting corrosion rates in AISI 304 steel heat exchangers in industrial circulating waters

Abstract

A predictive methodology was developed to estimate the pitting growth rate on AISI 304 steel surfaces operating in industrial circulating waters. The approach is based on the observation that most pits formed near oxide inclusions are metastable and repassivate within minutes; therefore, corrosion losses of Cr, Ni, and Fe are attributed primarily to stable pits. Stable pits were identified using selective dissolution coefficients for chromium (ZCr) and nickel (ZNi). Second-order regression models were established to correlate corrosion losses (ΔCr, ΔNi, ΔFe) with steel composition, structural features, and water parameters such as chloride concentration and pH. Results show that ΔCr is mainly influenced by chloride content and microstructural factors, including the number of oxides (1.98–3.95 μm), average austenite grain size, and δ-ferrite volume. Factors reducing ΔFe losses follow the order: chloride concentration < Ni content < acidity < grain size < number of 1.98–3.95 μm oxides; whereas smaller oxides, inter-oxide spacing, and δ-ferrite promote higher losses. ΔNi primarily depends on structural heterogeneity rather than chemical composition or water parameters. The resulting model enables prediction of average pitting growth rates on AISI 304 steel in circulating waters with an accuracy of ±19%, providing a practical tool for assessing corrosion resistance in heat exchanger applications.