Machine Learning and Computational Design of Materials for Energy Applications

About this project

Project description

Recent progress in integrating machine learning (ML) methods with data obtained from density functional theory (DFT) calculations has opened up a myriad of possibilities to explore a whole new way of high-throughput catalyst screening. In this study, machine learning models will be developed to estimate ‘reactivity descriptors’ on a catalyst surface, calculating turn over frequencies of the target reaction.

Accelerated prediction of descriptors will provide a platform for computational design of a heterogeneous catalyst. Qunatum mechanical simulations of the reactions on the catalyst surface will be performed using CP2K and VASP software packages on high performance computing clusters available at IIT Delhi and UQ. Applications of the project will lead towards the development of catalytic materials for renewable energy and CO2 mitigation.


Good computational skills

Information for applicants

Essential capabilities

Good computational skills

Desireable capabilities

good knowledge of chemistry, reaction mechanisms

Expected qualifications (Course/Degrees etc.)

B.Tech/ M.Tech or M.Sc. in Sciences

Project supervisors

Principal supervisors

UQ Supervisor

Professor Debra Bernhardt

Australian Institute for Bioengineering and Nanotechnology (AIBN)
IITD Supervisor

Associate Professor M. Ali Haider

Department of Chemical Engineering