Faculty of Science, Technology and Medicine
Faculty of Law, Economics and Finance
Faculty of Humanities, Education and Social Sciences
Interdisciplinary Centre for Security, Reliability and Trust
Luxembourg Centre for Systems Biomedicine
Luxembourg Centre for Contemporary and Digital History
Luxembourg Centre for European Law
Luxembourg Centre for Socio-Environmental Systems
The batteries powering electric vehicles and storing energy from solar panels are central to solving one of the biggest challenges of our time: moving away from fossil fuels. Yet designing better battery materials remains complex and slow. Researchers spend years testing new materials, only to discover critical flaws late in the process, often after significant investment has already been made.
An approach combining AI and physis-based simulations
A new approach by Prof. Alexandre Tkatchenko combines artificial intelligence with physics-based simulations to dramatically speed up how new battery materials are discovered and evaluated. The planned “BATMAT” platform just received an ERC Proof of Concept Grant andwill screen large numbers of battery materials used in rechargeable systems, including solid electrolytes and electrode components. The platform identifies the most promising candidates based on stability, ion mobility, and resistance to degradation.
‟ Europe desperately needs focused efforts on bringing its frontier science towards valorisation. The BATMAT ERC project will develop a platform for the design of novel solid electrolytes that are used in consumer batteries. I thank the ERC and look forward to new technologies coming out of our efforts.”
Professor in Theoretical Condensed Matter Physics
Speed without sacrificing accuracy
What makes BATMAT distinctive is not just that it uses AI, it’s how. The platform is built as a closed loop: AI models rapidly propose and evaluate promising materials, while the most interesting candidates are automatically passed to more detailed physics-based simulations for verification.
The system is expected to make decisions roughly 100 times faster than conventional approaches, without sacrificing scientific accuracy. “For battery research, where a single simulation can take weeks to years, this is a meaningful shift to decarbonate our society”, explains Prof. Tkatchenko.
To achieve this, the project leverages the advanced computing power of the frontier High Performance Computing (HPC) facility of the Vlog for cutting-edge simulations, as well as national .
By enabling earlier identification of failure risks, BATMAT aims to reduce costly trial-and-error and accelerate the development of safer, longer-lasting energy storage systems. For Alexandre, the next step is to develop a commercial platform within 18 months, with the goal of putting this tool directly in the hands of both researchers and industrial R&D teams.