Press release of June 10, 2026

Three HZDR AI projects receive Helmholtz funding

New AI methods for science and real-world applications

With the projects AIRE, INFUSE-X, and MLGREEN, the Helmholtz-Zentrum Dresden-Rossendorf (HZDR) has successfully secured funding for three projects in the Helmholtz AI Project Call 2025. Each project will receive up to €250,000 from the Helmholtz Association’s Impulse and Networking Fund (INF). The projects aim to develop new artificial intelligence (AI) approaches for scientific applications and thereby advance data-driven research across a range of disciplines. At the same time, they strengthen collaboration both within the Helmholtz Association and with partners from academia and industry.

The Helmholtz AI Project Call supports innovative research projects that develop new artificial intelligence methods or advance their application in scientific research. A total of 48 proposals were submitted in response to the 2025 call. The selection process was conducted by an international panel of experts, which evaluated the scientific quality, relevance, and innovative potential of the submitted projects. The three successful HZDR projects were particularly recognized for their contributions to the advancement of AI technologies within the Helmholtz Association and for their potential to generate scientific and societal impact.

AIRE: Beyond aluminum recycling

Although aluminum is highly recyclable, the recycling process is still challenged by small amounts of unwanted metals that can form microscopic structures during remelting and reduce material quality. Understanding the type, shape, and distribution of these structures is essential for developing alloys that are better suited for a circular economy. However, existing synchrotron X-ray imaging methods often struggle to distinguish between similar structures within recycled materials.

AIRE aims to overcome this limitation by developing a new AI approach that can automatically identify and analyze these structures in three-dimensional X-ray images. The project brings together a unique Helmholtz consortium, combining access to world-class synchrotron facilities (HZB), high-performance computing expertise (HZDR), and industrial know-how from Novelis, one of the world's leading companies in the aluminum industry.

The project will tackle major challenges in the analysis of complex 3D image data, including low image contrast, intricate material structures, and the large amount of expert input normally required to train AI systems. By building on a large existing dataset and advanced machine-learning methods, AIRE will enable the efficient analysis of vast amounts of imaging data while reducing the need for time-consuming manual data preparation.

Successful implementation will provide new quantitative insights into the microscopic structures that influence recycling performance and support the development of more sustainable aluminum alloys. In the long term, AIRE will help advance a truly circular aluminum economy.

Beyond aluminum recycling, the project will deliver an open-source software toolbox for the automated analysis of 3D scientific images. The resulting methods could benefit a wide range of research fields, including materials science, geoscience, and biomedical imaging, while laying the foundation for future AI models capable of learning from large volumes of scientific imaging data.

INFUSE-X: AI for Sustainable Raw Materials Exploration

The global transition to renewable energy, digital technologies, and sustainable industries is driving an unprecedented demand for critical raw materials. At the same time, there is a growing need to identify and assess new mineral resources in a more efficient and environmentally responsible way. The INFUSE-X project addresses this challenge by developing new artificial intelligence methods that combine and analyze geological data from multiple sources.

By integrating information from satellite observations, geophysical measurements, and high-resolution drone-based sensor systems, the project aims to improve the detection and characterization of mineral deposits and subsurface geological structures. The resulting AI-based tools will help make raw materials exploration more accurate, efficient, and sustainable.

INFUSE-X brings together the complementary expertise of the Helmholtz Institute Freiberg for Resource Technology (HZDR-HIF) and the GFZ Helmholtz Centre for Geosciences. Together, the partners will develop new methods for analyzing and integrating complex geoscientific data from multiple sources.

The new methods will be tested and validated at two mining sites with contrasting geological conditions: Rio Tinto in Spain and Roșia Poieni in Romania. These real-world case studies will ensure that the developed approaches are robust and transferable across different exploration environments.

In addition to advancing scientific understanding, the project will provide openly accessible tools and datasets for the research community. In the long term, INFUSE-X aims to support a new generation of data-driven mineral exploration technologies that contribute to a secure and sustainable supply of the raw materials needed for future technologies.

MLGREEN: Materials simulations beyond a few hundred atoms

Many materials are tested on computers long before they are produced in the laboratory. Yet even on high-performance computers (HPC), some of the most powerful methods in materials science remain limited to comparatively small systems far from realistic scales. The MLGREEN project plans to use AI to reduce the computational cost of the Korringa-Kohn-Rostoker (KKR) Green function method, making this highly accurate but demanding simulation approach applicable to much larger and more realistic materials systems.

Most material properties depend on electron behavior, including electrical conductivity, magnetism, certain optical properties, and superconductivity. So what material scientists, chemists and physicists want to know is, broadly speaking, where are the electrons, how do they move, and what energy do they have? The KKR Green function is a mathematical tool that describes how electrons move through a material while interacting with atomic nuclei that determine the material’s internal structure. The function captures all possible paths and scattering events in a compact way. It is therefore particularly useful for studying complex systems where regular atomic order is interrupted, for example in alloys, deliberately introduced impurities or defects of materials, material surfaces, or magnetic materials.

Despite its advantages, the KKR Green method is not used widely in science and industry. It is mathematically complex and is therefore harder to learn and implement than more standard methods. Another severe obstacle is computational cost. The application of the KKR Green method is – despite HPC – restricted to systems of only a few hundred atoms, precluding the simulation of realistic nanoscale systems where long-range interactions and complex geometries govern functionality. The MLGREEN project addresses the scaling challenge. Using neural networks, the aim is to reduce the computational cost from cubic to linear scaling with system size.

If successful, the project will help accelerate the discovery of novel materials for applications such as permanent magnets that do not rely on problematic rare-earth elements and spintronic devices – that is, devices that exploit not only the charge but also the spin of electrons.

The MLGREEN proposal was jointly submitted by the Center for Advanced Systems Understanding (CASUS) at Helmholtz-Zentrum Dresden-Rossendorf (HZDR) and the Forschungszentrum Jülich (FZJ).

Additional information:

AIRE

Dr. Peter Steinbach | Head Group of Artificial Intelligence
Department of Information Services and Computing
Phone: +49 351 260 3844 | Email: p.steinbach@hzdr.de

INFUSE-X

Prof. Dr. Richard Gloaguen | Department Head Exploration
Helmholtz Institute Freiberg for Resource Technology
Phone: +49 351 260 4424 | Email: r.gloaguen@hzdr.de

MLGREEN

Dr. Attila Cangi | Department Head Machine Learning for Materials Design
Center for Advanced Systems Understanding (CASUS)
Phone: +49 3581 375 23 52 | Email: a.cangi@hzdr.de

Media contact:

Simon Schmitt | Head
Communications and Media Relations at HZDR
Phone: +49 351 260 3400 | Mobile: +49 175 874 2865 | Email: s.schmitt@hzdr.de