Can a deep-learning model make fast predictions of vacancy formation in diverse materials?

Choudhary, Kamal; Sumpter, Bobby G.

doi:10.1063/5.0135382

Can a deep-learning model make fast predictions of vacancy formation in diverse materials?

Journal Article · Wed Sep 06 00:00:00 EDT 2023 · AIP Advances

DOI:https://doi.org/10.1063/5.0135382· OSTI ID:2000421

^[1]; ^[2]

National Inst. of Standards and Technology (NIST), Gaithersburg, MD (United States)
Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). Center for Nanophase Materials Sciences (CNMS)

The presence of point defects, such as vacancies, plays an important role in materials design. Here, we explore the extrapolative power of a graph neural network (GNN) to predict vacancy formation energies. We show that a model trained only on perfect materials can also be used to predict vacancy formation energies (E_vac) of defect structures without the need for additional training data. Such GNN-based predictions are considerably faster than density functional theory (DFT) calculations and show potential as a quick pre-screening tool for defect systems. To test this strategy, we developed a DFT dataset of 530 E_vac consisting of 3D elemental solids, alloys, oxides, semiconductors, and 2D monolayer materials. We analyzed and discussed the applicability of such direct and fast predictions. We applied the model to predict 192 494 E_vac for 55 723 materials in the JARVIS-DFT database. Our work demonstrates how a GNN-model performs on unseen data.

View Accepted Manuscript (DOE)

Research Organization:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: USDOE Office of Science (SC), Basic Energy Sciences (BES). Scientific User Facilities (SUF)

Grant/Contract Number:: AC05-00OR22725

OSTI ID:: 2000421

Journal Information:: AIP Advances, Journal Name: AIP Advances Journal Issue: 9 Vol. 13; ISSN 2158-3226

Publisher:: American Institute of Physics (AIP)Copyright Statement

Country of Publication:: United States

Language:: English

References (47)

Memristors Based on 2D Materials as an Artificial Synapse for Neuromorphic Electronics Huh, Woong; Lee, Donghun; Lee, Chul‐Ho Advanced Materials, Vol. 32, Issue 51 https://doi.org/10.1002/adma.202002092	journal	September 2020
Vacancy‐Induced Synaptic Behavior in 2D WS ₂ Nanosheet–Based Memristor for Low‐Power Neuromorphic Computing Yan, Xiaobing; Zhao, Qianlong; Chen, Andy Paul Small, Vol. 15, Issue 24 https://doi.org/10.1002/smll.201901423	journal	May 2019
Prediction of the Electron Density of States for Crystalline Compounds with Atomistic Line Graph Neural Networks (ALIGNN) Kaundinya, Prathik R.; Choudhary, Kamal; Kalidindi, Surya R. JOM, Vol. 74, Issue 4 https://doi.org/10.1007/s11837-022-05199-y	journal	March 2022
Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set Kresse, G.; Furthmüller, J. Computational Materials Science, Vol. 6, Issue 1, p. 15-50 https://doi.org/10.1016/0927-0256(96)00008-0	journal	July 1996
A comprehensive first-principles study of pure elements: Vacancy formation and migration energies and self-diffusion coefficients Shang, Shun-Li; Zhou, Bi-Cheng; Wang, William Y. Acta Materialia, Vol. 109 https://doi.org/10.1016/j.actamat.2016.02.031	journal	May 2016
Development of a ReaxFF reactive force field for intrinsic point defects in titanium dioxide Huygh, Stijn; Bogaerts, Annemie; van Duin, Adri C. T. Computational Materials Science, Vol. 95 https://doi.org/10.1016/j.commatsci.2014.07.056	journal	December 2014
Vacancy formation energies in metals: A comparison of MetaGGA with LDA and GGA exchange–correlation functionals Medasani, Bharat; Haranczyk, Maciej; Canning, Andrew Computational Materials Science, Vol. 101 https://doi.org/10.1016/j.commatsci.2015.01.018	journal	April 2015
A computational framework for automation of point defect calculations Goyal, Anuj; Gorai, Prashun; Peng, Haowei Computational Materials Science, Vol. 130 https://doi.org/10.1016/j.commatsci.2016.12.040	journal	April 2017
Convergence and machine learning predictions of Monkhorst-Pack k-points and plane-wave cut-off in high-throughput DFT calculations Choudhary, Kamal; Tavazza, Francesca Computational Materials Science, Vol. 161 https://doi.org/10.1016/j.commatsci.2019.02.006	journal	April 2019
Vacancy formation energy and its connection with bonding environment in solid: A high-throughput calculation and machine learning study Cheng, YingXing; Zhu, Linggang; Wang, Guanjie Computational Materials Science, Vol. 183 https://doi.org/10.1016/j.commatsci.2020.109803	journal	October 2020
Graph neural network predictions of metal organic framework CO2 adsorption properties Choudhary, Kamal; Yildirim, Taner; Siderius, Daniel W. Computational Materials Science, Vol. 210 https://doi.org/10.1016/j.commatsci.2022.111388	journal	July 2022
PyDII: A python framework for computing equilibrium intrinsic point defect concentrations and extrinsic solute site preferences in intermetallic compounds Ding, Hong; Medasani, Bharat; Chen, Wei Computer Physics Communications, Vol. 193 https://doi.org/10.1016/j.cpc.2015.03.015	journal	August 2015
Vacancy formation energy and size effects Yu, Xiaohua; Zhan, Zhaolin; Rong, Ju Chemical Physics Letters, Vol. 600 https://doi.org/10.1016/j.cplett.2014.03.046	journal	April 2014
Universal machine learning framework for defect predictions in zinc blende semiconductors Mannodi-Kanakkithodi, Arun; Xiang, Xiaofeng; Jacoby, Laura Patterns, Vol. 3, Issue 3 https://doi.org/10.1016/j.patter.2022.100450	journal	March 2022
High-Throughput Computational Screening of Perovskites for Thermochemical Water Splitting Applications Emery, Antoine A.; Saal, James E.; Kirklin, Scott Chemistry of Materials, Vol. 28, Issue 16 https://doi.org/10.1021/acs.chemmater.6b01182	journal	July 2016
On the Dopability of Semiconductors and Governing Material Properties Goyal, Anuj; Gorai, Prashun; Anand, Shashwat Chemistry of Materials, Vol. 32, Issue 11 https://doi.org/10.1021/acs.chemmater.9b05126	journal	May 2020
Density Functional Theory Calculations of Oxygen Vacancy Formation and Subsequent Molecular Adsorption on Oxide Surfaces Hinuma, Yoyo; Toyao, Takashi; Kamachi, Takashi The Journal of Physical Chemistry C, Vol. 122, Issue 51 https://doi.org/10.1021/acs.jpcc.8b11279	journal	November 2018
Intrinsic Material Properties Dictating Oxygen Vacancy Formation Energetics in Metal Oxides Deml, Ann M.; Holder, Aaron M.; O’Hayre, Ryan P. The Journal of Physical Chemistry Letters, Vol. 6, Issue 10 https://doi.org/10.1021/acs.jpclett.5b00710	journal	May 2015
Descriptor-Based Approach for the Prediction of Cation Vacancy Formation Energies and Transition Levels Varley, Joel B.; Samanta, Amit; Lordi, Vincenzo The Journal of Physical Chemistry Letters, Vol. 8, Issue 20 https://doi.org/10.1021/acs.jpclett.7b02333	journal	October 2017
Machine Learning-Enabled Design of Point Defects in 2D Materials for Quantum and Neuromorphic Information Processing Frey, Nathan C.; Akinwande, Deji; Jariwala, Deep ACS Nano, Vol. 14, Issue 10 https://doi.org/10.1021/acsnano.0c05267	journal	September 2020
Factors Governing Oxygen Vacancy Formation in Oxide Perovskites Wexler, Robert B.; Gautam, Gopalakrishnan Sai; Stechel, Ellen B. Journal of the American Chemical Society, Vol. 143, Issue 33 https://doi.org/10.1021/jacs.1c05570	journal	August 2021
Large variation of vacancy formation energies in the surface of crystalline ice Watkins, M.; Pan, D.; Wang, E. G. Nature Materials, Vol. 10, Issue 10 https://doi.org/10.1038/nmat3096	journal	September 2011
Machine learned features from density of states for accurate adsorption energy prediction Fung, Victor; Hu, Guoxiang; Ganesh, P. Nature Communications, Vol. 12, Issue 1 https://doi.org/10.1038/s41467-020-20342-6	journal	January 2021
The joint automated repository for various integrated simulations (JARVIS) for data-driven materials design Choudhary, Kamal; Garrity, Kevin F.; Reid, Andrew C. E. npj Computational Materials, Vol. 6, Issue 1 https://doi.org/10.1038/s41524-020-00440-1	journal	November 2020
Evolutionary computing and machine learning for discovering of low-energy defect configurations Arrigoni, Marco; Madsen, Georg K. H. npj Computational Materials, Vol. 7, Issue 1 https://doi.org/10.1038/s41524-021-00537-1	journal	May 2021
Benchmarking graph neural networks for materials chemistry Fung, Victor; Zhang, Jiaxin; Juarez, Eric npj Computational Materials, Vol. 7, Issue 1 https://doi.org/10.1038/s41524-021-00554-0	journal	June 2021
Atomistic Line Graph Neural Network for improved materials property predictions Choudhary, Kamal; DeCost, Brian npj Computational Materials, Vol. 7, Issue 1 https://doi.org/10.1038/s41524-021-00650-1	journal	November 2021
Recent advances and applications of deep learning methods in materials science Choudhary, Kamal; DeCost, Brian; Chen, Chi npj Computational Materials, Vol. 8, Issue 1 https://doi.org/10.1038/s41524-022-00734-6	journal	April 2022
Designing high-TC superconductors with BCS-inspired screening, density functional theory, and deep-learning Choudhary, Kamal; Garrity, Kevin npj Computational Materials, Vol. 8, Issue 1 https://doi.org/10.1038/s41524-022-00933-1	journal	November 2022
Zn Vacancy Formation Energy and Diffusion Coefficient of CVT ZnO Crystals in the Sub-Surface Micron Region Parmar, Narendra S.; Boatner, Lynn A.; Lynn, Kelvin G. Scientific Reports, Vol. 8, Issue 1 https://doi.org/10.1038/s41598-018-31771-1	journal	September 2018
Defect graph neural networks for materials discovery in high-temperature clean-energy applications Witman, Matthew D.; Goyal, Anuj; Ogitsu, Tadashi Nature Computational Science, Vol. 3, Issue 8 https://doi.org/10.1038/s43588-023-00495-2	journal	August 2023
High-throughput DFT calculations of formation energy, stability and oxygen vacancy formation energy of ABO3 perovskites Emery, Antoine A.; Wolverton, Chris Scientific Data, Vol. 4, Issue 1 https://doi.org/10.1038/sdata.2017.153	journal	October 2017
Oxygen vacancy and hole conduction in amorphous TiO ₂ Pham, Hieu H.; Wang, Lin-Wang Physical Chemistry Chemical Physics, Vol. 17, Issue 1 https://doi.org/10.1039/c4cp04209c	journal	January 2015
Unified graph neural network force-field for the periodic table: solid state applications Choudhary, Kamal; DeCost, Brian; Major, Lily Digital Discovery, Vol. 2, Issue 2 https://doi.org/10.1039/d2dd00096b	journal	January 2023
Chalcogen vacancies in monolayer transition metal dichalcogenides and Fermi level pinning at contacts Guo, Y.; Liu, D.; Robertson, J. Applied Physics Letters, Vol. 106, Issue 17 https://doi.org/10.1063/1.4919524	journal	April 2015
Machine learning substitutional defect formation energies in ABO ₃ perovskites Sharma, Vinit; Kumar, Pankaj; Dev, Pratibha Journal of Applied Physics, Vol. 128, Issue 3 https://doi.org/10.1063/5.0015538	journal	July 2020
Chemical accuracy for the van der Waals density functional Klimeš, Jiří; Bowler, David R.; Michaelides, Angelos Journal of Physics: Condensed Matter, Vol. 22, Issue 2 https://doi.org/10.1088/0953-8984/22/2/022201	journal	December 2009
Charge optimized many-body (COMB) potential for Al ₂ O ₃ materials, interfaces, and nanostructures Choudhary, Kamal; Liang, Tao; Chernatynskiy, Aleksandr Journal of Physics: Condensed Matter, Vol. 27, Issue 30 https://doi.org/10.1088/0953-8984/27/30/305004	journal	July 2015
Charge optimized many-body (COMB) potential for dynamical simulation of Ni–Al phases Kumar, Aakash; Chernatynskiy, Aleksandr; Liang, Tao Journal of Physics: Condensed Matter, Vol. 27, Issue 33 https://doi.org/10.1088/0953-8984/27/33/336302	journal	August 2015
High-throughput assessment of vacancy formation and surface energies of materials using classical force-fields Choudhary, Kamal; Biacchi, Adam J.; Ghosh, Supriyo Journal of Physics: Condensed Matter, Vol. 30, Issue 39 https://doi.org/10.1088/1361-648x/aadaff	journal	September 2018
Real time scissor correction in TD-DFT Wang, C-Y; Elliott, P.; Sharma, S. Journal of Physics: Condensed Matter, Vol. 31, Issue 21 https://doi.org/10.1088/1361-648x/ab048a	journal	March 2019
Projector augmented-wave method Blöchl, P. E. Physical Review B, Vol. 50, Issue 24, p. 17953-17979 https://doi.org/10.1103/physrevb.50.17953	journal	December 1994
Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set Kresse, G.; Furthmüller, J. Physical Review B, Vol. 54, Issue 16, p. 11169-11186 https://doi.org/10.1103/physrevb.54.11169	journal	October 1996
Vacancy-Formation Energies in Metals from Positron Annihilation McKee, B. T. A.; Triftshäuser, W.; Stewart, A. T. Physical Review Letters, Vol. 28, Issue 6 https://doi.org/10.1103/physrevlett.28.358	journal	February 1972
Effects of Pressure on Diffusion and Vacancy Formation in MgO from Nonempirical Free-Energy Integrations Ita, Joel; Cohen, Ronald E. Physical Review Letters, Vol. 79, Issue 17 https://doi.org/10.1103/physrevlett.79.3198	journal	October 1997
First-principles calculations of oxygen vacancy in CaO crystal Wu, Kaili; Liu, Tingyu; Sun, Ruxi The European Physical Journal D, Vol. 74, Issue 10 https://doi.org/10.1140/epjd/e2020-10299-8	journal	October 2020
Machine Learning Based Methodology to Predict Point Defect Energies in Multi-Principal Element Alloys Manzoor, Anus; Arora, Gaurav; Jerome, Bryant Frontiers in Materials, Vol. 8 https://doi.org/10.3389/fmats.2021.673574	journal	June 2021

Similar Records

Accelerating defect predictions in semiconductors using graph neural networks

Journal Article · Wed Mar 27 00:00:00 EDT 2024 · APL Machine Learning · OSTI ID:2564484

Related Subjects

2D materials
36 MATERIALS SCIENCE
Artificial neural networks
Crystalline properties
Crystallographic defects
Density functional theory
Energy data
Energy forecasting
Machine learning
Memristor
Reactive force field

Can a deep-learning model make fast predictions of vacancy formation in diverse materials?

Citation Formats

References (47)

Similar Records

Related Subjects