| 2023 | 2022 | 2021 | 2020 | 2019 | 2018 | 2017 | 2016 | 2015 | 2014 | 2013 | 2012 | 2011 | 2010 | 2009 | 2008 | 2007 | 2006 | 2005 | 2004 | 2003 | 2002 | 2001 | 2000 | 1999 | 1997

Predicting the thermodynamic stability of solids combining density functional theory and machine learning

Authors: J. Schmidt, J. Shi, P. Borlido, L. Chen, S. Botti, and M.A.L. Marques

Ref.: Chem. Mater. 29, 5090-5103 (2017)

Abstract: We perform a large scale benchmark of machine learning methods for the prediction of the thermodynamical stability of solids. We start by constructing a data set that comprises density functional theory calculations of around 250000 cubic perovskite systems. This includes all possible perovskite and anti-perovskite crystals that can be generated with elements from hydrogen to bismuth, and neglecting rare gases and lanthanides. Incidentally, these calculations already reveal a large number of systems (around 500) that are thermodynamically stable, but that are not present in crystal structure databases. Moreover, some of these phases have unconventional compositions and define completely new families of perovskites. This data set is then used to train and test a series of machine learning algorithms to predict the energy distance to the convex hull of stability. In particular, we study the performance of ridge regression, random forests, extremely randomized trees (including adaptive boosting), and neural networks. We find that extremely randomized trees give the best results, achieving errors in the test set of around 120 meV/atom when trained in 20000 samples. Surprisingly, the machine already works if we give it as sole input features the group and row in the periodic table of the three elements composing the perovskite. Moreover, we find that the prediction accuracy is not uniform across the periodic table, being worse for first-row elements and elements forming magnetic compounds. Our results point to the fact that machine learning can be successfully used to guide high-throughput density functional theory calculations to speed up by at least a factor of 5 systematic searches of new materials, without any degradation of the accuracy.

Citations: 182 (Google scholar)

DOI: 10.1021/acs.chemmater.7b00156

URL: pubs.acs.org

Download

Bibtex:

@article{Schmidt_2017,
	doi = {10.1021/acs.chemmater.7b00156},
	url = {https://doi.org/10.1021%2Facs.chemmater.7b00156},
	year = 2017,
	month = {jun},
	publisher = {American Chemical Society ({ACS})},
	volume = {29},
	number = {12},
	pages = {5090--5103},
	author = {Jonathan Schmidt and Jingming Shi and Pedro Borlido and Liming Chen and Silvana Botti and Miguel A. L. Marques},
	title = {Predicting the Thermodynamic Stability of Solids Combining Density Functional Theory and Machine Learning},
	journal = {Chemistry of Materials}
}