Synergistic combination of ab initio computations and machine learning to leverage the power of both represents the idea of Materials Genome approach, which refers to leveraging of big data obtained from high-throughput computations for predictive modeling.
Ab initio methods, specifically density functional theory (DFT) based methods, can be used to elucidate the structures and properties of small nanoclusters. Our research further focuses on utilizing electron density as a tool for elucidating bonding characteristics of nanoclusters. Having performed ab initio calculations and electronic density based analysis on several nanoclusters, we use the data obtained from ab initio calculations to develop cheminformatics models to predict the properties of larger nanoclusters. Such a combination of ab initio computations and machine learning helps to leverage the power of both and represents the future for the study of large and more challenging systems. This is also the idea of Materials Genome approach, which refers to leveraging of big data obtained from high-throughput computations on inorganic materials, and could be a valuable tool for creating a material science based informatics model.
Key publications:
- Pinaki Saha, Amol B. Rahane, Vijay Kumar, and N. Sukumar, Electronic Origin of the Stability of Transition Metal Doped B14 Drum Shaped Boron Clusters and Their Assembly in to a Nanotube. J. Phys. Chem. C, 121(20), 10728–10742 (2017). DOI: 10.1021/acs.jpcc.6b10838
- Pinaki Saha, Amol B. Rahane, Vijay Kumar, N. Sukumar, Analysis of the electron density features of small boron ring clusters and the effects of doping, Physica Scripta 91, 053005 (2016). DOI: 10.1088/0031-8949/91/5/053005
- Pinaki Saha, Amol B. Rahane, Vijay Kumar, and N. Sukumar, Double ring tubular structures of boron clusters stabilized by metal atom doping: M@B14 (M = Cr, Fe, and Ni), 253rd Nat. Meet. Amer. Chem. Soc., San Francisco, CA, April 2-6, 2017.
