Multi-Variable Multi-Metric Optimisation of Self-Assembled Photocatalytic CO2 Reduction Performance using Machine Learning Algorithms
Santiago Rodriguez Jimenez a, Erwin Reisner a, Shannon Bonke a, Giovanni Trezza b, Luca Bergamasco b, Leif Hammarström c, Eliodoro Chiavazzo b
a Yusuf Hamied Department of Chemistry, University of Cambridge, Lensfield Road, CB2 1EW, Cambridge, UK
b Department of Energy, Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Turin, Italy.
c Department of Chemistry, Ångström Laboratory, Uppsala University, Box 523, 75120 Uppsala, Sweden.
Materials for Sustainable Development Conference (MATSUS)
Proceedings of MATSUS Spring 2024 Conference (MATSUS24)
#PhotoMat - Advances in Photo-driven Energy Conversion and Storage: From Nanoscale Materials to Sustainable Solutions
Barcelona, Spain, 2024 March 4th - 8th
Organizers: Michelle Browne, Bahareh Khezri and Katherine Villa
Invited Speaker, Shannon Bonke, presentation 219
DOI: https://doi.org/10.29363/nanoge.matsus.2024.219
Publication date: 18th December 2023

Photocatalytic systems for the reduction of CO2 into fuels and platform chemicals are multi-variable multi-metric systems. As such, long-established optimisation approaches are poorly suited maximising system performance because they aim to optimise one performance metric while sacrificing the others and thereby limit overall system performance. I will present work to address this multi-metric challenge by defining a metric for holistic system performance that takes all figures of merit into account, and employing a machine learning algorithm to efficiently guide our experiments through the large parameter matrix to make holistic system optimisation accessible for human experimentalists. The test platform is a five-component system that self-assembles into photocatalytic micelles for CO2-to-CO reduction, which we experimentally optimised to simultaneously improve yield, quantum yield, turnover number, and frequency while maintaining high selectivity. Leveraging the dataset with machine learning algorithms allows quantification of each parameter’s effect on overall system performance, revealing that the buffer concentration is the dominating parameter for optimal performance with nearly four times more importance than the catalyst concentration. The expanded use and standardisation of this methodology to define and optimise holistic performance will accelerate progress in different areas of catalysis by providing unprecedented insights into performance bottlenecks, enhancing comparability, and taking results beyond comparison of subjective figures of merit.

© FUNDACIO DE LA COMUNITAT VALENCIANA SCITO
We use our own and third party cookies for analysing and measuring usage of our website to improve our services. If you continue browsing, we consider accepting its use. You can check our Cookies Policy in which you will also find how to configure your web browser for the use of cookies. More info