Machine Learning Guided Synthesis of Multinary Chevrel Phase Chalcogenides

The Chevrel phase (CP) is a class of molybdenum chalcogenides that exhibit compelling properties for next-generation battery materials, electrocatalysts, and other energy applications. Despite their promise, CPs are underexplored, with only ∼100 compounds synthesized to date due to the challenge of identifying synthesizable phases. We present an interpretable machine-learned descriptor (H_δ) that rapidly and accurately estimates decomposition enthalpy (ΔH_d) to assess CP stability. To develop H_δ, we first used density functional theory to compute ΔH_d for 438 CP compositions. We then generated >560 000 descriptors with the new machine learning method SIFT, which provides an easy-to-use approach for developing accurate and interpretable chemical models. From a set of >200 000 compositions, we identified 48 501 CPs that H_δ predicts are synthesizable based on the criterion that ΔH_d < 65 meV/atom, which was obtained as a statistical boundary from 67 experimentally synthesized CPs. The set of candidate CPs includes 2307 CP tellurides, an underexplored CP subset with a predicted preference for channel site occupation by cation intercalants that is rare among CPs. We successfully synthesized five of five novel CP tellurides attempted from this set and confirmed their preference for channel site occupation. Our joint computational and experimental approach for developing and validating screening tools that enable the rapid identification of synthesizable materials within a sparse class is likely transferable to other materials families to accelerate their discovery.

Stabilizing Hydrogen Adsorption through Theory-Guided Chalcogen Substitution in Chevrel-Phase Mo6X8 (X=S, Se, Te) Electrocatalysts

In this work, we implement a facile microwave-assisted synthesis method to yield three binary Chevrel-Phase chalcogenides (Mo₆X₈; X = S, Se, Te) and investigate the effect of increasing chalcogen electronegativity on hydrogen evolution catalytic activity. Density functional theory predictions indicate that increasing chalcogen electronegativity in these materials will yield a favorable electronic structure for proton reduction. This is confirmed experimentally via X-ray absorption spectroscopy as well as traditional electrochemical analysis. We have identified that increasing the electronegativity of X in Mo₆X₈ increases the hydrogen adsorption strength owing to a favorable shift in the p-band position as well as an increase in the Lewis basicity of the chalcogen, thereby improving hydrogen evolution reaction energetics. We find that Mo₆S₈ exhibits the highest hydrogen evolution activity of the Mo₆X₈ series of catalysts, requiring an overpotential of 321 mV to achieve a current density of 10 mA cm^-2_ECSA, a Tafel slope of 74 mV per decade, and an exchange current density of 6.01 × 10^-4 mA cm^-2_ECSA. Agreement between theory and experiment in this work indicates that the compositionally tunable Chevrel-Phase chalcogenide family is a promising framework for which electronic structure can be predictably modified to improve catalytic small-molecule reduction reactivity.

Linear correlation models for the redox potential of organic molecules in aqueous solutions

In this study, we use the molecular orbital energy approximation (MOEA) and the energy difference approximation (EDA) to build linear correlation models for the redox potentials of 53 organic compounds in aqueous solutions. The molecules evaluated include nitroxides, phenols, and amines. Both the MOEA and EDA methods yield similar correlation models, however, the MOEA method is less computationally expensive. Correlation coefficients (R²) below 0.3 and mean absolute errors above 0.25 V were found for correlation models built without solvent effects. When explicit water molecules and a continuum solvent model are added to the calculations, correlation coefficients close to 0.8 are reached, and mean absolute errors below 0.18 V are obtained. The incorporation of solvent effects is necessary for good correlation models, particularly for redox processes of charged molecules in aqueous solutions. A comparison of the correlation models from different methodologies is provided. Graphical abstract.