Abstract: Argonne National Laboratory is well known as a center of fundamental materials expertise and exceptional user facilities. A complementary expertise is that of materials manufacturing, where researchers aim to bring new materials capabilities to the market. The Materials Engineering Research Facility at Argonne is a locus of that effort and houses diverse manufacturing capabilities. Many of these processes are highly instrumented, resulting in a wealth of information that is currently underutilized. There is an opportunity to leverage this near real-time feedback to accelerate process optimization and result in better material outcomes. In this work, I present suitable artificial intelligence techniques to tackle this problem and demonstrate their application to two Argonne manufacturing capabilities, atomic layer deposition, and flame spray pyrolysis.
Bio: Noah Paulson is a computational materials scientist in the Applied Materials Division at Argonne National Laboratory. He investigates the intersection between statistics, machine learning, and materials science with applications including alloy design and nanoparticle synthesis. He joined Argonne as a postdoctoral researcher in 2017.