Research Projects

Hyperspectral Imaging

SpectralAIge

KI gestützte, spektroskopische Ermittlung der Alterung von Kunststoffen beim mechanischen Recycling

AI-supported, spectroscopic determination of the ageing of plastics during mechanical recycling

The increasing use of plastics requires a sustainable circular economy, i.e. the cyclical reuse of plastics after the end-of-life of products.

The associated challenges in waste management require innovative solutions. The SpectralAIge joint project aims to develop the world’s first spectroscopic measurement systems that can assess the process-related material ageing of plastics quickly and precisely. With the support of artificial intelligence (AI), the best possible decisions are to be made for the processing and use of secondary plastics in order to recognize and sort plastic material that has aged to varying degrees and to feed it into the optimal recycling path.

The knowledge gained from the project should make it possible to identify degraded materials in the closed PET deposit bottle cycle and in plastic sorting systems. This offers the option of making the best possible decisions for the processing and use of secondary plastics. Sorting out excessively degraded materials at an early stage ensures a higher quality of recyclate. This is essential, for example in the production of food packaging such as returnable bottles.

Challenges due to material ageing

Plastics are repeatedly exposed to high temperatures and mechanical stresses during the recycling process, usually in the extrusion and injection molding process. This can lead to ageing or degradation of the materials. If the concentration of heavily aged material in the manufacturing process is too high, the quality of the product could deteriorate. This can lead to the product no longer being allowed to be placed on the market due to quality defects. Although additives can compensate for these ageing processes to a certain extent, the only option for heavily aged material is downcycling or chemical recycling. However, there is currently a lack of industrial processes for quickly and reliably analyzing and evaluating the ageing of plastic waste.

Research project pursues two possible approaches
In the first project approach, hyperspectral camera systems (HSI), which are already established in plastics sorting, will be expanded to include LED-based light sources for fluorescence excitation. In addition to the classic absorption spectra, excited fluorescence spectra can also be recorded. The analysis of this spectral data using AI enables the precise determination of the material ageing of plastic waste.

The second approach focuses on the development of AI-supported, inexpensive inverted spectrometers. To this end, relevant wavelengths are extracted from broadband measurement results from HSI measurements and an inverse spectrometer is built using this data. This should allow conclusions to be drawn about the process-related ageing of plastic materials. In contrast to a spectrometer, which covers a complete wavelength range, an inverse spectrometer only covers specific individual wavelength bands. This allows production costs to be reduced and a wide range of industrial applications to be achieved.

Great economic potential
“The knowledge gained from the research project can be used to promote a wide range of economic applications and enable SMEs to move towards a circular economy through the optimal use of secondary plastics,” says Frank Stüpmann, Managing Director of Silicann Systems GmbH.

“We hope that the packaging sector, among others, will benefit greatly from the developments in the project. This generates the largest proportion of post-consumer (PC) plastic waste and there are short product life cycles and consequently frequent processing,” adds Dr. Linda Mittelberg, Head of Quality and Life Cycle at SKZ. “The results can be used in PET deposit bottle cycles and plastic sorting plants in particular.”

Key Facts

Project duration: 10/2024 – 09/2027
Funding measure: KMU innovativ – Production research in the program “Future of Value Creation” by the Federal Ministry of Education and Research (BMBF)
Project partners: Silicann Systems GmbH, HAIP Solutions GmbH, SKZ – KFE GmbH, Fraunhofer-Institut für Fabrikbetrieb und -automatisierung IFF