Expanding Materials Selection Via Transfer Learning for High-Temperature Oxide Selection

Abstract

Materials with higher operating temperatures than today’s state of the art can improve system performance in several applications and enable new technologies. Under most scenarios, a protective oxide scale with high melting temperatures and thermodynamic stability as well as low ionic diffusivity is required. Thus, the design of high-temperature systems would benefit from knowledge of these properties and related ones for a large number of oxides. While some properties of interest are available for many oxides (e.g., elastic constants exist for > 1000 oxides), the melting temperature is known for a relatively small subset. The determination of melting temperatures is time consuming and costly, both experimentally and computationally; thus, we use data science tools to develop predictive models from the existing data. Since the relatively small number of available melting temperature values precludes the use of standard tools, we use a multi-step approach based on transfer learning where surrogate data from first principles calculations are leveraged to develop models using small datasets. We use these models to predict the desired properties for nearly 11,000 oxides and quantify uncertainties in the space.