High temperature oxidation of corrosion resistant alloys from machine learning

Christopher D Taylor,Brett M Tossey

doi:10.1038/s41529-021-00184-3

Abstract

Parabolic rate constants, kp, were collected from published reports and calculated from corrosion product data (sample mass gain or corrosion product thickness) and tabulated for 75 alloys exposed to temperatures between ~800 and 2000 K (~500–1700 oC; 900–3000 oF). Data were collected for environments including lab air, ambient and supercritical carbon dioxide, supercritical water, and steam. Materials studied include low- and high-Cr ferritic and austenitic steels, nickel superalloys, and aluminide materials. A combination of Arrhenius analysis, simple linear regression, supervised and unsupervised machine learning methods were used to investigate the relations between composition and oxidation kinetics. The supervised machine learning techniques produced the lowest mean standard errors. The most significant elements controlling oxidation kinetics were Ni, Cr, Al, and Fe, with Mo and Co composition also found to be significant features. The activation energies produced from the machine learning analysis were in the correct distributions for the diffusion constants for the oxide scales expected to dominate in each class.

Highlights

High-temperature oxidation is a significant contributor to the failure of energy generation system components, affecting combustion systems, heat exchangers, and furnaces[1]
Sato and co-workers in 2011 outlined a predictive model for oxidation kinetics[23], beginning with an expression for the rate precipitates induced by the oxidation
Data were collected for environments including lab air, ambient and supercritical carbon dioxide, supercritical water, and steam

Summary

INTRODUCTION

High-temperature oxidation is a significant contributor to the failure of energy generation system components, affecting combustion systems (turbines and engines), heat exchangers, and furnaces[1]. Sato and co-workers in 2011 outlined a predictive model for oxidation kinetics[23], beginning with an expression for the rate precipitates induced by the oxidation We will revisit this expression in the work that follows, as we extend the model to constant that is proportional to the free energy of oxide formation: a very broad class of high-temperature alloys and provide some kp. The above-reviewed papers provide a significant data set of high-temperature oxidation properties of alloys spanning a diverse range of materials: nickel-based superalloys, aluminides, low-Cr ferritic steels, and higher-Cr austenitic steels. To facilitate the present and future development of CAMD of alloys with high-temperature oxidation resistance, we have collected this data and evaluated several models for predicting the parabolic rate constants published in the literature (or in some cases, derived from the provided oxidation measurements) as a function of composition. Hou observed that the effect of alloying and shifting the majority base element (Fe, Cr or Ni) on

RESULTS AND DISCUSSION

METHODS

CODE AVAILABILITY