Skip to content
‹ Back to News
23 October 2026

DandeLiion and STFC Hartree Centre collaborate to advance LFP battery modelling using AI

Cahn-Hilliard phase-field simulations showing interconnected regions from spinodal decomposition on the left and isolated droplets on the right
Cahn-Hilliard phase-field simulations illustrating spinodal decomposition (left) and droplet formation (right). Images courtesy of STFC researchers, from previous work.

DandeLiion Technologies has launched a six-month collaborative research project with the Science and Technology Facilities Council (STFC) Hartree Centre, part of UK Research and Innovation (UKRI), to develop AI-accelerated physics-based modelling techniques for lithium iron phosphate (LFP) batteries.

LFP is an increasingly important battery chemistry for electric vehicles and stationary energy storage, offering high safety, long cycle life and relatively low material costs. However, accurately modelling the complex phase transitions and lithium distribution within LFP electrode particles requires computationally demanding simulations.

The project, funded by Innovate UK, will investigate the use of the Cahn-Hilliard model to describe lithium concentration and phase separation in LFP materials. Researchers will combine experimental data, atomistic modelling and high-performance computing (HPC) to develop and parameterise the model, generating detailed simulation datasets. These datasets will then be used to train and evaluate AI-based surrogate models capable of reproducing key physical quantities at a significantly reduced computational cost.

The collaboration brings together DandeLiion's expertise in fast, physics-based battery simulation and the Hartree Centre's capabilities in scientific computing, computational modelling and artificial intelligence.

The project aims to lay the groundwork for extending DandeLiion's simulation capabilities to phase-changing electrode materials such as LFP. By combining detailed physical models with AI-based acceleration, this approach has the potential to enable faster battery design and optimisation while retaining the physical insight needed for reliable engineering decisions.