Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Moving closer to experimental level materials property prediction using AI

Journal Article · · Scientific Reports
Abstract

While experiments and DFT-computations have been the primary means for understanding the chemical and physical properties of crystalline materials, experiments are expensive and DFT-computations are time-consuming and have significant discrepancies against experiments. Currently, predictive modeling based on DFT-computations have provided a rapid screening method for materials candidates for further DFT-computations and experiments; however, such models inherit the large discrepancies from the DFT-based training data. Here, we demonstrate how AI can be leveraged together with DFT to compute materials properties more accurately than DFT itself by focusing on the critical materials science task of predicting “formation energy of a material given its structure and composition”. On an experimental hold-out test set containing 137 entries, AI can predict formation energy from materials structure and composition with a mean absolute error (MAE) of 0.064 eV/atom; comparing this against DFT-computations, we find that AI can significantly outperform DFT computations for the same task (discrepancies of $$>0.076$$ > 0.076 eV/atom) for the first time.

Research Organization:
Northwestern Univ., Evanston, IL (United States)
Sponsoring Organization:
National Science Foundation (NSF); US Department of Commerce; USDOE; USDOE Office of Science (SC)
Grant/Contract Number:
SC0019358; SC0021399
OSTI ID:
1875973
Alternate ID(s):
OSTI ID: 1904338
Journal Information:
Scientific Reports, Journal Name: Scientific Reports Journal Issue: 1 Vol. 12; ISSN 2045-2322
Publisher:
Nature Publishing GroupCopyright Statement
Country of Publication:
United Kingdom
Language:
English

References (63)

Hubbard-corrected DFT energy functionals: The LDA+U description of correlated systems journal July 2013
Adaptive machine learning framework to accelerate ab initio molecular dynamics journal December 2014
The Fourth Paradigm – Data-Intensive Scientific Discovery book January 2012
Data-Driven Model for Estimation of Friction Coefficient Via Informatics Methods journal May 2012
Materials Design and Discovery with High-Throughput Density Functional Theory: The Open Quantum Materials Database (OQMD) journal September 2013
Recent progress in metallurgical thermochemistry journal January 1969
Transfer learning for solvation free energies: From quantum chemistry to experiments journal August 2021
A high-throughput infrastructure for density functional theory calculations journal June 2011
AFLOWLIB.ORG: A distributed materials properties repository from high-throughput ab initio calculations journal June 2012
Matminer: An open source toolkit for materials data mining journal September 2018
Transfer learning for materials informatics using crystal graph convolutional neural network journal April 2021
Atomistic calculations and materials informatics: A review journal June 2017
Ultra-fast and accurate binding energy prediction of shuttle effect-suppressive sulfur hosts for lithium-sulfur batteries using machine learning journal March 2021
Extracting Grain Orientations from EBSD Patterns of Polycrystalline Materials Using Convolutional Neural Networks journal October 2018
High-Throughput Machine-Learning-Driven Synthesis of Full-Heusler Compounds journal October 2016
Predicting the Thermodynamic Stability of Solids Combining Density Functional Theory and Machine Learning journal May 2017
How Chemical Composition Alone Can Predict Vibrational Free Energies and Entropies of Solids journal July 2017
Deep learning journal May 2015
Machine-learning-assisted materials discovery using failed experiments journal May 2016
Accelerated search for materials with targeted properties by adaptive design journal April 2016
Quantum-chemical insights from deep tensor neural networks journal January 2017
Universal fragment descriptors for predicting properties of inorganic crystals journal June 2017
The Open Quantum Materials Database (OQMD): assessing the accuracy of DFT formation energies journal December 2015
A general-purpose machine learning framework for predicting properties of inorganic materials journal August 2016
Enhancing materials property prediction by leveraging computational and experimental data using deep transfer learning journal November 2019
Predicting materials properties without crystal structure: deep representation learning from stoichiometry journal December 2020
Cross-property deep transfer learning framework for enhanced predictive analytics on small materials data journal November 2021
Machine learning in materials informatics: recent applications and prospects journal December 2017
Machine learning modeling of superconducting critical temperature journal June 2018
High-throughput Identification and Characterization of Two-dimensional Materials using Density functional theory journal July 2017
ElemNet: Deep Learning the Chemistry of Materials From Only Elemental Composition journal December 2018
Plasma Hsp90 levels in patients with systemic sclerosis and relation to lung and skin involvement: a cross-sectional and longitudinal study journal January 2021
Experimental formation enthalpies for intermetallic phases and other inorganic compounds journal October 2017
Computational screening of high-performance optoelectronic materials using OptB88vdW and TB-mBJ formalisms journal May 2018
A predictive machine learning approach for microstructure optimization and materials design journal June 2015
A Statistical Learning Framework for Materials Science: Application to Elastic Moduli of k-nary Inorganic Polycrystalline Compounds journal October 2016
MoleculeNet: a benchmark for molecular machine learning journal January 2018
Density functional calculations of nuclear magnetic shieldings using the zeroth-order regular approximation (ZORA) for relativistic effects: ZORA nuclear magnetic resonance journal April 1999
A combined DFT and restricted open-shell configuration interaction method including spin-orbit coupling: Application to transition metal L-edge X-ray absorption spectroscopy journal May 2013
Commentary: The Materials Project: A materials genome approach to accelerating materials innovation journal July 2013
Perspective: Materials informatics and big data: Realization of the “fourth paradigm” of science in materials science journal April 2016
SchNet – A deep learning architecture for molecules and materials journal June 2018
Properties of intrinsic point defects in silicon determined by zinc diffusion experiments under nonequilibrium conditions journal December 1995
Formation enthalpies by mixing GGA and GGA + U calculations journal July 2011
Combinatorial screening for new materials in unconstrained composition space with machine learning journal March 2014
Predicting density functional theory total energies and enthalpies of formation of metal-nonmetal compounds by linear regression journal February 2016
Representation of compounds for machine-learning prediction of physical properties journal April 2017
Including crystal structure attributes in machine learning models of formation energies via Voronoi tessellations journal July 2017
Elastic properties of bulk and low-dimensional materials using van der Waals density functional journal July 2018
Machine Learning Energies of 2 Million Elpasolite ( A B C 2 D 6 ) Crystals journal September 2016
Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties journal April 2018
Diffusion of Ge below the Si(100) Surface: Theory and Experiment journal March 2000
Matrix- and tensor-based recommender systems for the discovery of currently unknown inorganic compounds journal January 2018
Machine learning with force-field-inspired descriptors for materials: Fast screening and mapping energy landscape journal August 2018
Developing an improved crystal graph convolutional neural network framework for accelerated materials discovery journal June 2020
Nobel Lecture: Electronic structure of matter—wave functions and density functionals journal October 1999
A Survey on Transfer Learning journal October 2010
Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning journal May 2016
BRNet: Branched Residual Network for Fast and Accurate Predictive Modeling of Materials Properties book January 2022
IRNet: A General Purpose Deep Residual Regression Framework for Materials Discovery
  • Jha, Dipendra; Ward, Logan; Yang, Zijiang
  • KDD '19: The 25th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining https://doi.org/10.1145/3292500.3330703
conference July 2019
Materials Informatics: The Materials “Gene” and Big Data journal July 2015
Phase Diagrams of the Elements book December 1991
Materials science with large-scale data and informatics: Unlocking new opportunities journal May 2016

Similar Records

Enhancing materials property prediction by leveraging computational and experimental data using deep transfer learning
Journal Article · Thu Nov 21 19:00:00 EST 2019 · Nature Communications · OSTI ID:1619594

Measuring the structure and equation of state of polyethylene terephthalate at megabar pressures
Journal Article · Thu Jun 17 20:00:00 EDT 2021 · Scientific Reports · OSTI ID:1797603