The materials science domain has undergone significant transformation after introducing so-called high-throughput computational materials design [1], resulting in generation of huge amounts of data, e.g., AFLOW[2] or MaterialsProject[3] initiatives. In particular, thermodynamic properties of alloys are in the focus of researchers’ interest, including the relatively young topic of high-entropy alloys [4], gaining a lot of attention among researchers due to potentially very promising properties of such alloys.

Despite large amounts of data already generated and available, the principles of data linking and data-driven decision making have not yet established their place in this field, despite this would be an irreplaceable support for human intuition, which certainly reaches its limit when it comes to exploration of the combinatorially huge space of multi-principle alloys [4].

 The recent ontology review work [5] elevates the necessity of alignment between different domain-level ontologies (DLOs), to provide a unified ontology framework for data interchange and interoperability in materials science. One of the challenges is that existing DLOs lack formal axiomatic conceptualization, tending to provide only basic taxonomy, thus hampering even pairwise alignment of the related DLOs [5].

We aim to tackle this challenge by proposing a semantically-rich ontology for alloy thermodynamics, defining relevant domain terms like phase diagram, alloy, etc. through class axioms based on more elementary classes, down to periodic elements, thus providing the semantic backbone for understanding the alloy research results. This ontology will be mapped to BFO-2020 [7] and PMDco [8] (version 3.0) and designed in a way, enabling integration of the existing thermodynamic alloy data (either computational or experimental) into the corresponding alloy knowledge graph. Which, in turn, can be integrated into the prospective MSE Knowledge Graph version [6], as it shares the same principle of semantically-rich data annotation. Along with alignment of this ontology with the other relevant DLOs, this will foster data interoperability in the material science domain.

This thermodynamic alloy ontology aims to formalize answering to the corresponding competency questions by domain experts, involving referring to more basic underlying concepts and their relations.

References:

1. Curtarolo, Stefano, et al. "The high-throughput highway to computational materials design." Nature materials 12.3 (2013): 191-201.
2. Curtarolo, S. et al. AFLOWLIB.ORG: A distributed materials properties repository from high-throughput ab initio calculations. Comp. Mater. Sci. 58, 227–235 (2012)
3. Jain, A. et al. A high-throughput infrastructure for density functional theory calculations. Comp. Mater. Sci. 50, 2295–2310 (2011).
4. Miracle, Daniel B., and Oleg N. Senkov. "A critical review of high entropy alloys and related concepts." Acta Materialia 122 (2017): 448-511.
5. De Baas, Anne, et al. "Review and Alignment of Domain-Level Ontologies for Materials Science." IEEE Access 11 (2023): 120372-120401.
6. https://go.fzj.de/mse-kg
7. Otte, J. Neil, John Beverley, and Alan Ruttenberg. "BFO: Basic formal ontology." Applied ontology 17.1 (2022): 17-43.
8. B. Bayerlein, M. Schilling, H. Birkholz, M. Jung, J. Waitelonis, L. Mädler, H. Sack, Materials & Design 2024, 237 112603.