[Paper Review] Exploring material compositions for synthesis using oxidation states

Maung Thway, Andy Paul Chen+10 more|arXiv (Cornell University)|May 14, 2024

Catalysis and Oxidation ReactionsChemical Engineering3 citations

TL;DR

This paper introduces the Oxidation State Probability (OSP) method, a data-driven approach that predicts synthesizability of ternary materials by evaluating the likelihood of elements adopting required oxidation states for charge neutrality. Applied to the Cu-In-Te system, OSP successfully predicted and guided the experimental synthesis of CuIn₃Te₅, demonstrating its potential to prioritize viable compositions from large candidate sets.

ABSTRACT

Recent advances in machine learning techniques have made it possible to use high-throughput screening to identify novel materials with specific properties. However, the large number of potential candidates produced by these techniques can make it difficult to select the most promising ones. In this study, we develop the oxidation state probability (OSP) method which evaluates ternary compounds based on the probability (the OSP metric) of each element to adopt the required oxidation states for fulfilling charge neutrality. We compare this model with Roost and the Fourier-transformed crystal properties (FTCP)-based synthesizability score. Among the top 1000 systems with the most database entries in Materials Project (MP), more than 500 systems exhibit an attested compound among the top 3 compositions when ranked by the OSP metric. We find that the OSP method shows promising results for certain classes of ternary systems, especially those containing nonmetals, s-block, or transition metals. When applied to the Cu-In-Te ternary system, an interesting system for thermoelectric applications, the OSP method predicted the synthesizability of CuIn$_3$Te$_5$ without prior knowledge, and we have successfully synthesized CuIn$_3$Te$_5$ in experiment. Our method has the potential to accelerate the discovery of novel compounds by providing a guide for experimentalists to easily select the most synthesizable candidates from an arbitrarily large set of possible chemical compositions.

Motivation & Objective

• To address the challenge of selecting promising material compositions from high-throughput machine learning screenings, where vast candidate pools overwhelm experimental feasibility.
• To develop a computationally efficient metric that leverages oxidation state trends to estimate the likelihood of successful synthesis.
• To improve the prioritization of ternary compounds—especially those with nonmetals, s-block, or transition metals—by incorporating chemical intuition into a quantitative scoring system.
• To provide experimentalists with a reliable, interpretable guide to focus synthesis efforts on the most plausible compositions.

Proposed method

• The Oxidation State Probability (OSP) metric computes the probability of each element in a ternary compound adopting the oxidation states required for overall charge neutrality.
• OSP is derived from statistical analysis of known compounds in the Materials Project database, using historical oxidation state distributions for each element.
• The method ranks ternary compositions based on the product of individual element oxidation state probabilities, favoring combinations where elements are likely to adopt stable, charge-balancing oxidation states.
• OSP is compared against Roost and FTCP-based synthesizability scores to evaluate predictive performance on known materials.
• The approach is applied to the Cu-In-Te system, where it predicted CuIn₃Te₅ as a highly probable candidate despite no prior experimental reports.
• The method is designed to be interpretable and scalable, enabling rapid screening of arbitrarily large composition spaces.

Experimental results

Research questions

• RQ1 Can oxidation state trends be quantitatively leveraged to predict the synthesizability of ternary inorganic compounds?
• RQ2 How does the OSP metric compare in performance to existing synthesizability predictors like Roost and FTCP?
• RQ3 Does the OSP method effectively prioritize experimentally accessible compositions, particularly in systems with nonmetals, s-block, or transition metals?
• RQ4 Can the OSP method guide successful experimental synthesis of a previously unreported compound?
• RQ5 To what extent does the OSP metric identify known, attested compounds among the top-ranked compositions?

Key findings

• Among the top 1000 ternary systems with the most entries in the Materials Project database, over 500 had at least one attested compound ranked in the top 3 compositions by the OSP metric.
• The OSP method successfully predicted the synthesizability of CuIn₃Te₅ in the Cu-In-Te system, a compound not previously reported experimentally.
• The experimental synthesis of CuIn₃Te₅ was successfully achieved, validating the predictive power of the OSP model.
• The OSP method showed particular strength in predicting compounds containing nonmetals, s-block elements, and transition metals.
• The OSP metric outperformed baseline methods in identifying known, stable compounds among high-scoring candidates, especially in systems with complex oxidation state behavior.
• The method provides a reliable, interpretable, and scalable framework for guiding experimental synthesis in high-throughput materials discovery.