Discovery of bimetallic catalysts via machine learning algorithms.
Project objectives: This PhD project will involve the development, optimisation, and benchmarking of machine learning models to predict the binding energies of key CO2RR intermediates using the thermodynamic data calculated by other partners or taken from the literature.
Working context: DC#6 will be hosted and supervised by Prof. M. García-Melchor in Trinity College Dublin (Ireland). DC#6 will enrol in the PhD programme (DubChem) of the same institution and receive the necessary training via the courses offered therein and the TCD team. Complementary training in the modelling of the reaction thermodynamics and kinetics of the CO2RR on metallic surfaces will also be provided via secondments at POLITO and UI, respectively. In addition, part of the proposed activities will be carried out via secondment periods at TME to expand DC#6’s knowledge of machine learning.
- The main outcome of this PhD project would be a machine learning model able to predict binding energies of key CO2RR intermediates while exhibiting a comparable accuracy to state-of-the-art density functional theory methods. The developed machine learning model will be used to screen for novel Cu-based materials (and possibly other bimetallic alloys) for the CO2RR.