We generate and analyze large amounts of material data to accelerate the discovery and design of new materials.
Advances in computational methods and the exponential increase in computer speed have made it possible to computationally generate large databases of material properties. Such high-throughput computing enables researchers to rapidly scan tens of thousands of candidate materials to identify those that hold the most promise for technological applications. Materials designers can analyze the data to identify important chemical and structural trends, providing new insights into how to create materials with desired properties. To fully leverage the power of high-throughput computing, it is necessary to have methods that are capable of rapidly and accurately predicting the relevant material property values. We develop such methods and use them in high-throughput calculations to design and discover new materials for advanced technologies.