ICL
Developing charge-aware machine learning potential for electrochemical applications
Partially Funded
⏰ Closing Soon
🎓 Artificial Intelligence
🎓 Computational Chemistry
🎓 Computational Physics
🎓 Machine Learning
🎓 Materials Science
🎓 Physical Chemistry
🎓 Thermodynamics
AI
UK
DFT
Electrocatalysis
Electrochemistry
MLIP
Machine Learning Potentials
Materials Modelling
This PhD focuses on developing charge-aware machine learning interatomic potentials for electrochemical applications, enabling accurate modelling of electrocatalytic systems with long-range electrostatic interactions.
Project Description
In recent years, machine learning interatomic potentials (MLIPs) have emerged as a new frontier for materials modeling. They promise near-DFT accuracy with only a fraction of its cost, thereby significantly expanding the affordable time and spatial scales of the simulation. These developments open new doors for high-throughput materials discovery with more realistic structures and vast configurational space. However, there are still challenges to be overcome to use MLIPs for electrochemical reactions at surfaces and interfaces. The commonly used MLIPs determine the forces and energies of a given atom by its local atomistic environment within a certain cutoff, making the model inherently short-sighted. As a result, these MLIPs fail to capture the long-range electrostatic interaction between charged species, which is key in electrochemical catalysis.
The proposed project aims to develop an MLIP model with explicit control of electrochemical potential and charge state, and to establish a modeling framework utilizing charge-aware MLIP for predicting the thermodynamics and kinetics of electrocatalytic surfaces under explicit control of electrode potential. The project will involve training in electrochemistry, atomistic modeling, high-throughput calculation, and machine learning for materials.
Entry Requirements
Experience with density functional theory calculation preferred
Experience with machine learning preferred
Experience with machine learning preferred
How to Apply
Apply via Imperial College London application portal
Contact Dr Jing Yang for informal enquiries
Contact Dr Jing Yang for informal enquiries
Eligibility
UK/Home
EU
International
Supervisor Profile
DJ
Dr Jing Yang
Imperial College London, Department of Materials
Related Opportunities
MUM
MU
MU
CNR