Today 2026 March15 (Sun) 11:54 Etc/GMT-9

2026/04/08 00:30~2026/04/08 01:30

ChE Seminar: John Kitchin, Carnegie Mellon University, "Large machine-learned potentials for materials design with catalysis and energy applications"

 Sponsored by the Department of Chemical and Biomolecular EngineeringHost: AlexandraZagalskayaazagalskaya@umass.edu 413-545-7114John Kitchin Carnegie Mellon University “Large machine-learnedpotentials for materials design with catalysis and energy applications” Tuesday, April 7, 2026, 11:30 a.m.201 LGRT, UMass Amherst(Refreshments at 11:15 a.m.) Abstract:Designing multicomponent materials ischallenging with density functional theory (DFT) because it computationallyexpensive to evaluate all the ways that elements may combine and react. It isalso difficult to estimate free energy contributions to reactions and to locatereaction barriers with DFT. The Open Catalyst Project is developing machinelearned potentials (MLP) to mitigate these challenges. These MLPs are trainedon 100M+ DFT calculations spanning 55 different elements and 80+ adsorbatesthat are relevant in catalysis and energy applications. Nominally these modelswere trained to predict energy and forces, and from these one can derivereaction energies. We will show in this talk, however, that these models alsoshow great utility in computing reaction barriers, and in estimating freeenergy contributions to reactions. This opens the door to a post-scaling era ofcomputational catalysis where reaction barriers can be computed in complexreaction networks with near DFT accuracy rather than relying on less accuratelinear scaling relations. We will show some case studies of results and discussfuture research directions in this area. Finally, I will discuss the growingrole of generative AI in scientific research with examples from our most recentwork. Bio: John Kitchin completed his B.S. in Chemistryat North Carolina State University. He completed a M.S. in Materials Scienceand a PhD in Chemical Engineering at the University of Delaware in 2004 underthe advisement of Dr. Jingguang Chen and Dr. Mark Barteau. He received anAlexander von Humboldt postdoctoral fellowship and lived in Berlin, Germany for1 ½ years studying alloy segregation with Karsten Reuter and Matthias Schefflerin the Theory Department at the Fritz Haber Institut. Professor Kitchin began atenure-track faculty position in the Chemical Engineering Department atCarnegie Mellon University in January of 2006. He is currently the John E.Swearingen Professor. At CMU, Professor Kitchin works in the areas of alloycatalysis and molecular simulation. He was awarded a DOE Early Career award in2010 to investigate multifunctional oxide electrocatalysts for the oxygenevolution reaction in water splitting using experimental and computationalmethods. He received a Presidential Early Career Award for Scientists andEngineers in 2011. He completed a sabbatical in the Accelerated Science groupat Google learning to apply machine learning to scientific and engineeringproblems in 2018.

📍 LGRT 201