Integrated computational materials design frameworks such as AFLOW can be used to automatically generate large quantities of materials properties data [1, 2, 3]. Data from ensembles of calculations can be used by AFLOW to evaluate thermomechanical properties [4] and to predict the synthesizability of new materials using the thermodynamic density of states [5]. This AFLOW data becomes a powerful tool for materials design when combined with machine learning models to predict properties ranging from electronic properties such as superconductivity [6] and band gap to thermomechanical properties such as bulk modulus and Debye temperature. These machine learning models are now available for use via interactive online applications on the aflow.org web portal. [1] S. Curtarolo et al., Comp. Mat. Sci. 58, 218 (2012); [2] S. Curtarolo et al., Comp. Mat. Sci, 58, 227 (2012); [3] R. H. Taylor et al., Comp. Mat. Sci. 93, 178 (2014); [4] C. Toher et al., Phys. Rev. B 90, 174107 (2014); [5] E. Perim et al., Nat. Commun. 7, 12315 (2016); [6] O. Isayev et al., Chem. Mater. 27, 735 (2014).

Automated materials property prediction using thermodynamic density of states and machine learning

253rd National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.