(167h) High-Throughput Catalysis and Information Management

Experimental teams developing catalysts produce hundreds of samples per year. Without a proper information management system, information on how they were prepared and how they performed can become inaccessible, or at worst, completely missing. This results in increased cost and lower product performance. A data management tool together with predictive analytics can increase R&D efficiency and reduce the number of experiments that need to be performed. In this presentation we explore some of the tools available such as recursive partitioning, principal component analysis, and neural networks; and demonstrate how these techniques have been used in catalyst discovery. Enriching the available data with the results of molecular modeling provides valuable information that can improve the reliability of predictive models.