Complex Nanofeatures in Crystals: Theory and Experiment Meet in the Cloud
The project will combine computation to predict, synthesis to make, and x-ray and neutron local structure characterization to validate the predictions, an approach that embodies the Materials Genomics philosophy and applies it to polymorphous network materials (PNMs). Quantum mechanical density functional theory (DFT) calculations will be applied to supercells of transition metal oxides and chalcogenides that are sufficiently large to support the PNM effect, to see if nanostructured distortions can lower the total energy. These will be applied to classes of known materials, such as hybrid organic-inorganic halides, to search for and characterize the nature of the PNM distortions. The most promising materials will be synthesized and characterized using pair distribution function (PDF) analysis, a diffraction method sensitive to the local distortions. A computational infrastructure will be built that will save results, both theoretical and experimental, to databases for later mining. The infrastructure will be made available to the community to carry out their own computational 'experiments'.
An AI-driven, Cloud-based, Materials Discovery Platform for Nanomaterial Structure: PDFitc
S. Billinge, Q. Du, D. Hsu (Columbia U.)