Mechanisms, Descriptors, and Data-Driven Discovery for Rational, Accelerated Design of Catalytic Materials
DIPC Seminars
- Speaker
-
Max García-Melchor
CIC energIGUNE - When
-
2026/02/20
12:00 - Place
- DIPC Josebe Olarra Seminar Room
- Host
- Ricardo Díez Muiño
- Add to calendar
-
iCal
The development of sustainable technologies for energy storage and conversion is essential for mitigating climate change and reducing reliance on fossil resources. A central challenge in this transition is the rational design of catalytic materials capable of efficiently converting renewable electricity into chemical fuels and value-added products under realistic operating conditions. Addressing this challenge requires approaches that combine mechanistic understanding, physically meaningful descriptors, and data-driven strategies implemented within scalable computational workflows.
This talk will first discuss how elucidating reaction mechanisms and identifying robust activity descriptors provide a foundation for rational catalyst design. Using the oxygen evolution reaction (OER) as a representative example, recent work from our group elucidating activity trends and scaling relations across molecular and heterogeneous catalysts will be presented, with particular emphasis on how these insights can be embedded into automated screening pipelines to accelerate the discovery of cost-effective catalysts for green hydrogen production.
The importance of modeling realistic catalyst states under operando conditions will then be highlighted, with a focus on electrified solid–liquid interfaces. In particular, this part will demonstrate how the catalyst resting state and the distribution of interfacial hydrogen govern selectivity in electrochemical hydrogenation reactions, and how rational interface engineering enables control over surface coverages and reaction pathways using earth-abundant materials.
The final part of the talk will address the accelerated rational design of catalysts for the electrochemical reduction of CO2 toward C2+ products. Two complementary examples will be presented: a machine learning–assisted screening of Cu-based bimetallic catalysts grounded in adsorption energetics, and a statistical framework that captures the configurational complexity of disordered alloy surfaces. These efforts are supported by modular, data-centric workflows that enable the systematic generation, curation, and reuse of chemically consistent datasets, providing a natural interface with machine learning and AI tools for catalyst discovery. The presentation will conclude with a concise synthesis of these results and an outlook on future directions for rational, operando-aware catalyst design.
Zoom: https://dipc-org.zoom.us/j/
YouTube: https://youtube.com/live/