Machine Learning Energies of 2 Million ElpasoliteCrystals
Top Cited Papers
Open Access
- 20 September 2016
- journal article
- research article
- Published by American Physical Society (APS) in Physical Review Letters
- Vol. 117 (13) , 135502
Abstract
Elpasolite is the predominant quaternary crystal structure ( prototype) reported in the Inorganic Crystal Structure Database. We develop a machine learning model to calculate density functional theory quality formation energies of all pristine elpasolite crystals that can be made up from main-group elements (up to bismuth). Our model’s accuracy can be improved systematically, reaching a mean absolute error of for a training set consisting of crystals. Important bonding trends are revealed: fluoride is best suited to fit the coordination of the site, which lowers the formation energy whereas the opposite is found for carbon. The bonding contribution of the elements and is very small on average. Low formation energies result from and being late elements from group II, being a late (group I) element, and being fluoride. Out of crystals, 90 unique structures are predicted to be on the convex hull—among which is , with a peculiar stoichiometry and a negative atomic oxidation state for Al.
Keywords
All Related Versions
Funding Information
- Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (PP00P2_138932)
- Air Force Office of Scientific Research
- U.S. Air Force (FA9550-15-1-0026)
- Vetenskapsrådet (621-2011-4249)
- Swiss National Supercomputing Centre (mr14)
- Swedish e-Science Research Centre
- Swedish National Infrastructure for Computing
- National Center of Competence in Research for Computational Design and Discovery of New Materials
This publication has 38 references indexed in Scilit:
- Accelerating materials property predictions using machine learningScientific Reports, 2013
- Machine learning of molecular electronic properties in chemical compound spaceNew Journal of Physics, 2013
- Semiconductor thermochemistry in density functional calculationsPhysical Review B, 2008
- Improved grid‐based algorithm for Bader charge allocationJournal of Computational Chemistry, 2007
- Voronoi deformation density (VDD) charges: Assessment of the Mulliken, Bader, Hirshfeld, Weinhold, and VDD methods for charge analysisJournal of Computational Chemistry, 2003
- The SIESTA method forab initioorder-Nmaterials simulationJournal of Physics: Condensed Matter, 2002
- From ultrasoft pseudopotentials to the projector augmented-wave methodPhysical Review B, 1999
- Projector augmented-wave methodPhysical Review B, 1994
- Special points for Brillouin-zone integrationsPhysical Review B, 1976
- Self-Consistent Equations Including Exchange and Correlation EffectsPhysical Review B, 1965