Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Machine learning descriptors for predicting the high temperature oxidation of refractory complex concentrated alloys

Akhil Bejjipurapu, Alejandro Strachan, Kenneth H. Sandhage, Michael S. Titus

Abstract

Refractory Complex Concentrated Alloys (RCCAs) can exhibit exceptional high-temperature strength, making such alloys promising candidates for high-temperature structural applications. However, current RCCAs do not possess the high-temperature oxidation resistance required to survive in oxidizing environments for more than a few hours at or above 1000$^\circ$C, without relying primarily on an environmental barrier coating. Here, we present a machine-learning framework designed to predict the oxidation-induced specific mass changes of RCCAs exposed for 24 h at 1000$^\circ$C in air, in order to support the search for oxidation-resistant alloys over a wide range of compositions. A database was constructed of experimental specific mass change data, upon oxidation at 900-1000$^\circ$C for 24 h in air, for 77 compositions comprised of simple elements, binary alloys, and higher-order elemental systems. We then developed a Gaussian Process Regression (GPR) model with physics-informed descriptors based on oxidation products, capturing the fundamental chemistry of oxide formation and stability. Application of this GPR model to the database yielded a MAE (mean absolute error) test score of 5.78 mg/cm$^2$, which was a significant improvement in accuracy relative to models only utilizing traditional alloy-based descriptors. Our model was used to screen over 5,100 quaternary RCCAs, revealing compositions with significantly lower predicted specific mass changes compared to existing literature sources. Overall, this work establishes a versatile and efficient strategy to accelerate the discovery of next-generation RCCAs with enhanced resistance to extreme environments.

Machine learning descriptors for predicting the high temperature oxidation of refractory complex concentrated alloys

Abstract

Refractory Complex Concentrated Alloys (RCCAs) can exhibit exceptional high-temperature strength, making such alloys promising candidates for high-temperature structural applications. However, current RCCAs do not possess the high-temperature oxidation resistance required to survive in oxidizing environments for more than a few hours at or above 1000C, without relying primarily on an environmental barrier coating. Here, we present a machine-learning framework designed to predict the oxidation-induced specific mass changes of RCCAs exposed for 24 h at 1000C in air, in order to support the search for oxidation-resistant alloys over a wide range of compositions. A database was constructed of experimental specific mass change data, upon oxidation at 900-1000C for 24 h in air, for 77 compositions comprised of simple elements, binary alloys, and higher-order elemental systems. We then developed a Gaussian Process Regression (GPR) model with physics-informed descriptors based on oxidation products, capturing the fundamental chemistry of oxide formation and stability. Application of this GPR model to the database yielded a MAE (mean absolute error) test score of 5.78 mg/cm, which was a significant improvement in accuracy relative to models only utilizing traditional alloy-based descriptors. Our model was used to screen over 5,100 quaternary RCCAs, revealing compositions with significantly lower predicted specific mass changes compared to existing literature sources. Overall, this work establishes a versatile and efficient strategy to accelerate the discovery of next-generation RCCAs with enhanced resistance to extreme environments.

Paper Structure

This paper contains 11 sections, 1 equation, 12 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methods
  3. Data Collection and Curation
  4. Descriptor Calculations
  5. Machine Learning for Specific Mass Change
  6. Alloy Design Space
  7. Distributions of Oxide-Related Descriptors Across Alloy Design Space
  8. SHAP Analysis
  9. Results and Discussion
  10. Conclusions
  11. Supplementary Information

Figures (12)

  • Figure 1: Integrated machine learning framework for predicting the high-temperature oxidation behavior of Refractory Complex Concentrated Alloys (RCCAs)
  • Figure 1: (a) Bar chart showing the distribution of different alloy systems and oxidation temperatures present in the entire dataset. (b) Bar chart showing the distribution of specific mass gain values for the selected 77 alloys oxidized at 900–1000°C
  • Figure 2: Left: Prediction of the multi-layered structure of the oxidation products formed upon oxidation of a Al35Nb5Ti50Cr10 alloy at 1000$^{\circ}$C in air, with the most oxygen-rich layer (highest oxygen chemical potential) shown at the top. Right: For the first layer containing 30 vol percent oxide, descriptors are shown for characteristics associated with the entire layer and for characteristics associated with the oxide mixture in this layer.
  • Figure 2: (a) Bar chart showing the number of alloys (out of a total of 77 alloys) containing a given element (present at any concentration) in the dataset that had been filtered for oxidation at 900–1000°C. (b) Box plot showing the atomic fraction distributions of elements in this dataset.
  • Figure 3: (a) Full design space (mapped into 2D via MDS) and (b) design space restricted to alloys with only BCC phases processable via arc melting.
  • ...and 7 more figures