High-­throughput Materials Discovery through Materials Genomics

Discovering better materials is essential for tackling the enormous challenges in developing new renewable energy sources.  The experimental variables that must be considered to optimize a given property are too many and their relationships are too complex to allow anything but incremental improvements to be made.  So how do we think "outside the box" to find entirely new materials?  As part of the larger effort known as the "Materials Genome Initiative," approaches based on data-mining and materials informatics techniques can help screen new compounds with desired properties and features, at greatly accelerated rates, and provide insights into the design principles required to engineer improved materials.  These tools will be used, in particular, to find better photovoltaics (in collaboration with Buriak and Shankar groups) and catalysts for solar fuels (in collaboration with Bergens group).

Publications, Activities, and Awards

  • Accelerating the Discovery of Solid State Materials: From Traditional to Machine-Learning Approaches
  • Accelerating the Discovery of Solid State Materials: From Traditional to Machine-Learning Approaches
  • Accelerating the Discovery of Solid State Materials: From Traditional to Machine-Learning Approaches
  • Discovery of Intermetallic Compounds from Traditional to Machine-Learning Approaches
  • Disentangling Structural Confusion through Machine Learning: Structure Prediction and Polymorphism of Equiatomic Ternary Phases ABC
  • Excellence in Undergraduate Teaching 2017
  • Excellence in Undergraduate Teaching 2018
  • Faculty of Science Students' Choice Honour Roll
  • How to look for compounds
  • How to look for compounds
  • How to look for compounds
  • Prediction of Novel Compounds and Rapid Property Screening through a Machine Learning Approach
  • Searching for Missing Binary Equiatomic Phases: Complex Crystal Chemistry in the Hf–In System