When
Monday, September 28, 2026, at 10:00 a.m.
Thomas Purcell
Assistant Professor
Department of Chemistry & Biochemistry
University of Arizona
"Automating Computational Materials and Molecule Design with Explainable AI"
Harshbarger 118A-A1
ABSTRACT
High-throughput density functional theory (DFT) calculations have the capability to quickly screen thousands of materials to identify the top candidates for myriad energy and sustainability applications. However, despite its impressive efficiency relative to experiments, computational screening is still incapable of exploring the entirety of materials space, which is needed to find the optimal candidate structures. Incorporating artificial intelligence (AI) models into these frameworks would further accelerate these searches by focusing on the most promising candidates as early as possible, but these models are often limited by a scarcity of available data. Here I will present the Purcell Lab's recent efforts to develop new high-throughput workflows to describe various material properties for solid-state and polymer materials. In particular, I will focus on how incorporating explainable AI into these frameworks allows them to not only focus the searches on the optimal regions of materials space, but also extract design rules to guide future experimental studies. Finally, I will showcase how these workflow architectures can be adapted for verifying the consistency of DFT calculations across different implementations.
ABOUT THE SPEAKER
Thomas Purcell is an assistant professor in chemistry and biochemistry at the University of Arizona. He received his BS in chemistry at New York University, before getting his PhD at Northwestern University under Prof. Tamar Seideman. During his PhD, he developed classical and semi-classical methods to describe the coupling between quantum emitters and plasmonic nanoparticles using the finite-difference time-domain and Maxwell Liouville algorithms. He then joined the Theory Department (now the NOMAD Laboratory) of the Fritz-Haber-Institut der Max-Planck-Gesellschaft (FHI) as an Alexander von Humboldt Postdoctoral Fellow, and later became a group leader in the same department. At the University of Arizona, Purcell focuses on merging explainable AI into high-throughput computational workflows for materials discovery applications. As a part of these efforts, he developed and still maintains the FHI-vibes and SISSO++ software packages for modeling the vibrational properties of a material and finding analytical expressions for a material's properties using the sure-independence screening and sparsifying operator, respectively.