The growth of renewable energy makes efficient and cost-effective storage technologies essential to balance fluctuations between generation and consumption and to ensure a reliable energy supply during extended periods of low wind and solar output. One promising option is the redox flow battery, which stores energy in liquid electrolytes and can release it rapidly when needed. Its performance depends on suitable electrolyte materials, the development of which has so far been resource-intensive and time-consuming.

This is where the Marie Skłodowska-Curie doctoral network “PREDICTOR – Predictive Platform for Electrochemical Energy Storage Materials” comes in. Creating a fully integrated, AI-supported development platform is at the heart of the work. This platform will cover the entire materials development process: from the computer-aided identification of new electrolyte materials to automated synthesis and characterization to validation in battery demonstrators.The project uses high-throughput methods, well known from pharmaceutical research. It applies them for the first time in a fully integrated way to electrochemical energy storage, seamlessly linking modeling, experimental testing, and data management.

The PREDICTOR network brings together seven core partners and 15 associated organizations and companies from eleven European countries. Within this consortium, 17 doctoral researchers are working on a new generation of high-performance organic electrolytes for redox-flow batteries, using advanced modeling methods, simulations, machine learning, and self-optimizing laboratory systems.

Fraunhofer SCAI’s Virtual Material Design business area leads the work package on developing semantic concepts and data-driven tools for design optimization. This includes building an ontology to integrate modeling and characterization data, and developing AI-based prediction models and tools for semantic search and data analysis. The goal is to systematically exploit the large amount of data generated to identify promising material candidates more quickly and refine them in a targeted manner.

The project is funded by the Marie Skłodowska-Curie Actions of the European Commission.

Project duration: 09/2024 to 08/2028