12. October 2022

New at HZB: Tomography lab for AI-assisted battery research

Federal Ministry of Research and Education is funding the "TomoFestBattLab" project with 1.86 million euros

X-ray tomography of a battery cathode © M. Osenberg, I. Manke/ HZB / Binder/ KIT

X-ray tomography of a battery cathode, virtually disassembled into its components. © M. Osenberg, I. Manke/ HZB / Binder/ KIT

At HZB, a laboratory for automated X-ray tomography on solid-state batteries is being set up. The special feature: 3D data during charge/discharge processes (operando) can be evaluated quickly and in a more versatile way using artificial intelligence (AI) methods. The Federal Ministry of Research and Education is funding the "TomoFestBattLab" project with 1.86 million euros.

X-ray tomography allows a direct glimpse into a battery's inner structures during discharging and charging. "For example, when the lithium moves back and forth between the anode and cathode during charging and discharging, the lithium storage material may expand or chemical transformation processes may take place," explains tomography expert Dr Ingo Manke. The three-dimensional imaging of these structural changes can reveal weak points in terms of performance and durability, for example ageing processes. X-ray tomography can map these structural changes and has therefore also become an indispensable measurement technique in battery research – similar to medicine.

HZB is now setting up an automated tomography laboratory that is specifically geared to the needs of solid-state batteries. The evaluation of tomographic measurements is extremely time-consuming because the data volumes are huge and require complex 3D algorithms. Therefore, large parts of the 3D evaluations are to be fully automated with the help of artificial intelligence (or machine learning) methods. This is supported by a special high-performance computer with which so-called "digital twins" of the real batteries can be generated.

This combination of artificial intelligence methods and tomography measurement techniques is an innovative approach with a pilot function for equipping future laboratories. "The project helps us to digitalise battery research with regard to the requirements of Industry 4.0 and to accelerate the development of batteries," says project coordinator Manke.

The new laboratory will support working groups at university and non-university research institutions as well as industrial companies in developing and improving new battery technologies.

Funded until 2024

The project "Machine Learning supported automated laboratory for multi-dimensional Operando Tomography of solid-state batteries under real operating conditions" (TomoFestBattLab, FKZ 03XP0462) is funded by the Federal Ministry of Education and Research (BMBF) as part of the initiative to expand the national research infrastructure in the field of battery materials and technologies (ForBatt). The project is funded from 01.09.2022 to 31.08.2024.