SBIR Phase I: A Data Aggregation and Mining Platform for Materials Research and Development

The broader impact/commercial potential of this project is to create a data-driven electronic infrastructure for rapidly developing new advanced materials. Advanced materials are crucial to a $100 billion cross-section of the American economy, from clean energy (catalysts, batteries, solar cells), to aerospace & defense (heat shields, armor), to automotive (lightweight alloys), to electronics (semiconductors, LEDs). Unfortunately, inventing better materials is very challenging?the process can take 10-20 years?and a key part of the problem is a lack of consolidated, searchable, analytics-friendly materials databases that can inform research efforts. This project will build an advanced database of catalyst materials and their properties, which will allow for the use of the same data-driven strategy to accelerate the pace of innovation in many other materials applications. This project can immediately strengthen the United States? competitive position in research and manufacturing in the $14 billion global catalyst market, and position the nation for a future information technology advantage in other materials-related industries. This Small Business Innovation Research (SBIR) Phase I project addresses a critical technical problem plaguing the materials industry at large: a lack of comprehensive databases of materials and their properties. Without access to such data, the materials industry cannot systematically exploit trends in those data to invent next-generation materials. This particular project focuses on catalyst materials, but will be readily extensible to other areas of materials development in the future. Cost-effectively collecting, storing, and searching data on tens of thousands of materials is a major technical challenge, which this project plans to address by developing cutting-edge text mining algorithms and database infrastructure. Such technologies, if successful, will enable rapid harvesting and aggregation of materials data from text, tables, and smaller databases. These technologies could be readily applied to build out materials property databases in many other areas beyond catalysis, and thereby fuel the materials industry with an enormous infusion of data.