Machine Learning Energies of 2 M Elpasolite (ABC$_2$D$_6$) Crystals

TL;DR

This paper presents a machine learning model that accurately predicts formation energies of approximately 2 million elpasolite crystal structures, revealing bonding trends and identifying stable compounds on the convex hull.

Contribution

The study introduces a scalable machine learning approach to estimate DFT-quality energies for a vast crystal space, uncovering key bonding insights and stable structures.

Findings

Model achieves 0.1 eV/atom accuracy with 10k training crystals.

Fluoride best fits D site coordination, lowering formation energy.

90 structures predicted on the convex hull, including novel compounds.

Abstract

Elpasolite is the predominant quaternary crystal structure (AlNaK$_2$F$_6$ prototype) reported in the Inorganic Crystal Structure Database. We have developed a machine learning model to calculate density functional theory quality formation energies of all $\sim$2 M pristine ABC$_2$D$_6$ elpasolite crystals which can be made up from main-group elements (up to bismuth). Our model's accuracy can be improved systematically, reaching 0.1 eV/atom for a training set consisting of 10 k crystals. Important bonding trends are revealed, fluoride is best suited to fit the coordination of the D site which lowers the formation energy whereas the opposite is found for carbon. The bonding contribution of elements A and B is very small on average. Low formation energies result from A and B being late elements from group (II), C being a late (I) element, and D being fluoride. Out of 2 M crystals, 90 unique structures are predicted to be on the convex hull---among which NFAl$_2$Ca$_6$, with peculiar stoichiometry and a negative atomic oxidation state for Al.