Tailoring catalysts for reactions involving multifunctional molecules using machine learning
Event details
| Date | 14.04.2026 |
| Hour | 15:15 › 16:15 |
| Speaker | Assistant Professor Tej Choksi |
| Location | |
| Category | Conferences - Seminars |
| Event Language | English |
Abstract
First principles methods and more recently machine learning, have opened new avenues to systematize catalyst design. Yet, computational efforts of the last 2+ decades have primarily focused on reactions involving small molecules (C1-2, O1, N1) that are activated by high-symmetry sites of extended surfaces at low surface coverages. In contrast, most industrial processes involve large multi-functional molecules (e.g., cyclic hydrocarbons, nitroarenes etc.) that are activated by low-symmetry sites of nanoclusters under high surface coverages. Machine learning methods have emerged as a powerful approach to bridge this gap between first principles models and reality. Using physics-inspired machine learning trained on small datasets by active learning, we identify next-generation bimetallic Pt catalysts for dehydrogenating liquid organic hydrogen carriers like methylcyclohexane. Using machine learned potentials, we rationalize why functionalized nitroarenes adsorb in self-assembled monolayer-like arrangements on Pd catalysts, consistent with in situ surface characterization. We distil our learnings into a forward-looking viewpoint of the do’s and don’ts in using machine learned potentials to predict the surface structure and reactivity of multi-functional molecules.
Biography
Tej has been a tenure-track Assistant Professor at Nanyang Technological University, Singapore since Dec. 2019. He obtained his PhD in chemical engineering from Purdue University in Dec. 2017 (advisor: Prof. Jeffrey Greeley) and thereafter completed his postdoctoral training at Stanford University (advisor: Dr. Frank Abild-Pedersen). His group employs first principles methods and kinetic modelling to understand how catalysts operate at the atomic scale. In addition to managing his research programme, Tej is a Faculty Fellow of the NTU Honors College and the Director of the undergraduate programme in chemical and biomolecular engineering. Tej has received the Outstanding Young Principal Investigator of the AIChE, Singapore Local Section (2022) and the Nanyang Education Award (2022) for excellence in teaching. Tej holds early career advisory board roles in the Journal of Catalysis, ChemSusChem, and Applied Catalysis A: General.
First principles methods and more recently machine learning, have opened new avenues to systematize catalyst design. Yet, computational efforts of the last 2+ decades have primarily focused on reactions involving small molecules (C1-2, O1, N1) that are activated by high-symmetry sites of extended surfaces at low surface coverages. In contrast, most industrial processes involve large multi-functional molecules (e.g., cyclic hydrocarbons, nitroarenes etc.) that are activated by low-symmetry sites of nanoclusters under high surface coverages. Machine learning methods have emerged as a powerful approach to bridge this gap between first principles models and reality. Using physics-inspired machine learning trained on small datasets by active learning, we identify next-generation bimetallic Pt catalysts for dehydrogenating liquid organic hydrogen carriers like methylcyclohexane. Using machine learned potentials, we rationalize why functionalized nitroarenes adsorb in self-assembled monolayer-like arrangements on Pd catalysts, consistent with in situ surface characterization. We distil our learnings into a forward-looking viewpoint of the do’s and don’ts in using machine learned potentials to predict the surface structure and reactivity of multi-functional molecules.
Biography
Tej has been a tenure-track Assistant Professor at Nanyang Technological University, Singapore since Dec. 2019. He obtained his PhD in chemical engineering from Purdue University in Dec. 2017 (advisor: Prof. Jeffrey Greeley) and thereafter completed his postdoctoral training at Stanford University (advisor: Dr. Frank Abild-Pedersen). His group employs first principles methods and kinetic modelling to understand how catalysts operate at the atomic scale. In addition to managing his research programme, Tej is a Faculty Fellow of the NTU Honors College and the Director of the undergraduate programme in chemical and biomolecular engineering. Tej has received the Outstanding Young Principal Investigator of the AIChE, Singapore Local Section (2022) and the Nanyang Education Award (2022) for excellence in teaching. Tej holds early career advisory board roles in the Journal of Catalysis, ChemSusChem, and Applied Catalysis A: General.
Practical information
- Informed public
- Free
Organizer
- Philippe Schwaller
Contact
- Sarina Kopf