Prof. Heather Kulik
Title: Machine Learning for Open Shell Transition Metal Complex and Metal-Organic Framework Discovery
Abstract: I will discuss our efforts to use machine learning (ML) to accelerate the computational tailoring and design of open shell transition metal homogeneous catalysts as well as metal-organic framework (MOF) materials for catalysis and gas separations/storage. One limitation in a challenging materials space such as open shell transition metal chemistry present in the open metal sites of many homogeneous catalysts and most catalytically active MOFs is that ML models and ML-accelerated high-throughput screening traditionally rely on density functional theory (DFT) for data generation, but DFT is both computationally demanding and prone to errors that limit its accuracy in predicting new MOFs. In the realm of MOFs, I will describe how we have curated a dataset of thousands of MOFs that have been experimentally synthesized and used this data to train ML models to predict experimentally reported measures of stability. These models predict experimental thermal stability and activation stability, which would be extremely difficult to predict using computational modeling. I will also describe how we have used these models to accelerate the discovery of novel stable MOFs, creating a dataset enriched with stability and diversity 1-2 orders of magnitude beyond what is typically included in most hypothetical MOF datasets. In the realm of open shell transition metal chemistry, I will describe our efforts to accelerate the discovery of transition metal chromophores and single site catalysts with machine learning. We have used our open-source toolkit molSimplify to accelerate the discovery of candidate catalysts with ML. We exploit active learning to optimize catalysts and chromophores in spaces of millions of candidate materials. Finally, time permitting, I will talk about how we’ve used this approach to design materials from the nanoscale to the macroscale, identifying polymers crosslinked with transition metal complexes with unprecedented toughness.
Bio: Professor Heather J. Kulik is a tenured Professor in the Departments of Chemical Engineering and Chemistry at MIT. She received her B.E. in Chemical Engineering from the Cooper Union in 2004 and her Ph.D. from the Department of Materials Science and Engineering at MIT in 2009. She completed postdoctoral training at Lawrence Livermore and Stanford, prior to joining MIT as a faculty member in November 2013. Her research has been recognized by an Office of Naval Research Young Investigator Award, DARPA Young Faculty Award and Director’s fellowship, NSF CAREER Award, a Sloan Fellowship in chemistry, an AIChE Computational and Molecular Simulation Engineering Forum Impact Award, and a Hans Fischer Senior Fellowship from the Technical University of Munich, among others.
Keywords: Density functional theory, Machine learning, Transition metal chemistry, Catalysis, Open shell transition metals
Host: Prof. JR Schmidt