A user-friendly Windows program has been developed with the capability to assess corrosion of commercial alloys in complex gases at high temperatures, Thermochemical predictions of the stable alloy corrosion products are used to infer the likely corrosion mechanism. Regression analysis is used to produce correlations of alloy corrosion with exposure conditions. A large data collection has been compiled and represents nearly 4.7 million hours of exposure time for 71 alloys. The data represent total metal penetration (sound metal loss by surface scale plus internal corrosion) and have been collected from public plus private sources representing a time period of nearly 40 years. The data originate from well-defined exposures with known concentrations of H2, H2S, H2O, CO, CO2, COS, O2, N2, and CH4, temperature, exposure time, and alloy. A large amount of information is now available to predict corrosion for the conditions of interest. The capabilities of the program are reviewed in this presentation for corrosion by oxidation, sulfidation, oxidation/sulfidation, and carburization.
Corrosion in high temperature processes often limits the useful lifetimes of equipment, and corrosion specialists need comprehensive methods of corrosion problem diagnosis, corrosion data retrieval, analysis, and corrosion prediction. The complex interactions between gas composition, alloy composition, temperature, and exposure time influence corrosion rates, and must be considered in selecting suitable materials of construction. An enormous amount of data has been generated and compiled for a diversity of exposure conditions, corrosion mechanisms and alloys. The information most useful to a corrosion engineer is the total metal penetration (sound metal loss) which has been influenced by the corrosive exposure conditions. An example of the traditional surface scale corrosion assessment and the total metal penetration (including internal corrosion product formation) is illustrated in Figure 1. Measurements of the total penetration for corroded specimens is the information which has been incorporated into the computer program discussed in this presentation. Various sources and formats of corrosion information (such as technical journals, reports, and symposia volumes) contain data, which has not been previously collected into a useable format. The use of this type of data traditionally depends upon the expertise of a corrosion engineer to acquire and use the data.
We have created a comprehensive computer program to provide corrosion information for complex, high temperature, corrosive gases. Both theoretical and empirical data analyses are used in providing corrosion information, The approach includes: thermochemically analyzing the corrosive environments, determining corrosion product stabilities, archiving of corrosion data, statistically correlating corrosion with exposure conditions, and predicting corrosion. The functionality of the program predicts dominant corrosion mechanisms and alloy corrosion rates in high temperature gases that contain mixtures of H2, H2O, H2S, CO, CO2, COS, O2, and CH4. The approach has been evaluated for substantial variations in gas compositions, exposure times, temperatures, and commercial alloys containing significant concentrations of Fe-Cr-Ni-Co-W-Mo-Si-Al. Data from long-term exposure (durations of thousands of hours) have been emphasized because of the interest in accurate corrosion predictions for commercial equipment lifetimes.
