Predicting charge density distribution of materials using a local-environment-based graph convolutional network

Gong, Sheng; Xie, Tian; Zhu, Taishan; Wang, Shuo; Fadel, Eric; Li, Yawei; Grossman, Jeffrey C.

doi:10.1103/physrevb.100.184103

Cited by 45 publications

(52 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…83 Electronic energies, particularly if converted to formation energies, may provide insight into the relative stability of MOFs. 84 Machine learning the charge density 85 is a potential way to bypass a large portion of the calculations performed with Kohn-Sham DFT. [86][87][88] Both the charge density and density of states can provide insight into the electronic structure in addition to serving as promising features to predict a variety of other quantum-chemical properties.…”

Section: Resultsmentioning

confidence: 99%

Machine learning the quantum-chemical properties of metal–organic frameworks for accelerated materials discovery

Rosen

Iyer

Ray

et al. 2021

Matter

219

259

View full text Add to dashboard Cite

Using a new database of electronic structure properties derived from highthroughput density functional theory calculations for thousands of metal-organic frameworks (MOFs), we benchmark a variety of machine learning models to accurately and rapidly predict MOF band gaps. Unsupervised dimensionality reduction techniques are also used to map the MOF feature space and identify otherwise subtle structure-property relationships. We anticipate that machine learning models derived from this new database will accelerate the discovery of promising MOFs with targeted quantum-chemical properties.

show abstract

Section: Resultsmentioning

confidence: 99%

Machine learning the quantum-chemical properties of metal–organic frameworks for accelerated materials discovery

Rosen

Iyer

Ray

et al. 2021

Matter

219

259

View full text Add to dashboard Cite

show abstract

“…We connect a given atom with its neighbors, identified by the tessellations, and describe Si/Ge configurations as crystal graphs. Crystal graphs encode both atomic information and bonding environments [44][45][46] , and are being increasingly used in ML models for successful materials property prediction 42,[44][45][46] . Figure 2c shows representative crystal graphs G of a Si/Ge configuration.…”

Section: Resultsmentioning

confidence: 99%

First-principles prediction of electronic transport in fabricated semiconductor heterostructures via physics-aware machine learning

Pimachev

Neogi

2021

npj Comput Mater

View full text Add to dashboard Cite

First-principles techniques for electronic transport property prediction have seen rapid progress in recent years. However, it remains a challenge to predict properties of heterostructures incorporating fabrication-dependent variability. Machine-learning (ML) approaches are increasingly being used to accelerate design and discovery of new materials with targeted properties, and extend the applicability of first-principles techniques to larger systems. However, few studies exploited ML techniques to characterize relationships between local atomic structures and global electronic transport coefficients. In this work, we propose an electronic-transport-informatics (ETI) framework that trains on ab initio models of small systems and predicts thermopower of fabricated silicon/germanium heterostructures, matching measured data. We demonstrate application of ML approaches to extract important physics that determines electronic transport in semiconductor heterostructures, and bridge the gap between ab initio accessible models and fabricated systems. We anticipate that ETI framework would have broad applicability to diverse materials classes.

show abstract

“…Minimum Li adsorption energies from GCN and DFT. In this work, we choose graph convolutional networks (GCN) as the machine learning architecture for learning Li site energies at different adsorption sites, because it has been shown to encode atomic and geometric information with high transferability 34,43,45 , and has been utilized as a model form of interatomic potentials 30, 42 . In order to efficiently learn Li adsorption energies at different sites, we iteratively sample sites from each material with site energies calculated by DFT, and then train GCN on all the calculated energies.…”

Section: Resultsmentioning

confidence: 99%

Screening and understanding Li adsorption on 2-dimensional metallic materials by learning physics

Gong

Wang

Zhu

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Two-dimensional (2D) materials have received considerable attention as possible electrodes in Li-ion batteries (LIBs), although a deeper understanding of the Li adsorption behavior as well as broad screening of the materials space is still needed. In this work, we build a high-throughput screening scheme that incorporates a learned interaction. First, density functional theory and graph convolution networks are utilized to calculate minimum Li adsorption energies for a small set of 2D metallic materials. The data is then used to find a dependence of the minimum Li adsorption energies on the sum of ionization potential, work function of the 2D metal, and coupling energy between Li+ and substrate. Our results show that variances of elemental properties and density are the most correlated features with coupling. To illustrate the applicability of this approach, the model is employed to show that some fluorides and chromium oxides are potential high-voltage materials with adsorption energies < -7 eV, and the found physics is used as the design principle to enhance the Li adsorption ability of graphene. This physics-driven approach shows higher accuracy and transferability compared with purely data-driven models.

show abstract

Predicting charge density distribution of materials using a local-environment-based graph convolutional network

Cited by 45 publications

References 49 publications

Machine learning the quantum-chemical properties of metal–organic frameworks for accelerated materials discovery

Machine learning the quantum-chemical properties of metal–organic frameworks for accelerated materials discovery

First-principles prediction of electronic transport in fabricated semiconductor heterostructures via physics-aware machine learning

Screening and understanding Li adsorption on 2-dimensional metallic materials by learning physics

Contact Info

Product

Resources

About