ES22-Georgescu

2022 Workshop on Recent Developments in Electronic Structure (ES22) Poster Session

View Poster

Author: Georgescu, Alexandru Bogdan - Northwestern University, McCormick School of Engineering, Department of Materials Science and Engineering

Title: Machine Learning Assisted Quantum Materials Discovery: Metal-Insulator Transition Compounds

Abstract: Metal–insulator transition (MIT) compounds are materials that may exhibit metallic or insulating behavior, depending on the physical conditions, and are of immense fundamental interest owing to their potential applications in emerging microelectronics. Nonetheless, only around 60 materials with a thermally-driven metal-insulator transition are known, and their computational discovery is difficult due to the non-equilibrium nature of the transition, and the complexity of the many-body problem. To address this issue, we have built the first database of all known thermally-drivem metal-insulator transition compounds, as well as stoichiometrically related compounds, and a machine-learning based classifier tool to accelerate their discovery - and provided both to the wider public, with no installation required. We also present here possible new metal-insulator transition oxide compounds identified through a combination of machine learning and DFT calculations, which may be of interest to the scientific community. 

This work was supported in part by the National Science Foundation (NSF) under award number DMR-1729303 and the Advanced Research Projects Agency-Energy (ARPA-E), U.S. Department of Energy, under Award Number DE-AR0001209.  

References:  [1] A.B. Georgescu, P. Ren, A.R. Toland, S. Zhang, K.D. Miller, D.W. Apley, E.A. Olivetti, N. Wagner, J.M. Rondinelli, ‘Database, Features, and Machine Learning Model to Identify Thermally-Driven Metal-Insulator Transition Compounds’, Chem. Matter, 2021, 33, 14, 5591-5605  [2] MIT material database: https://mtd.mccormick.northwestern.edu/mit-classification-dataset/  [3] ML Classifier: tinyurl.com/mit-classifiers

Other authors: Rondinelli, James & Northwestern University, McCormick School of Engineering, Department of Materials Science and Engineering

Machine Learning Assisted Quantum Materials Discovery: Metal-Insulator Transition Compounds