The IGF‑Vorhaben 21841 BG was a joint effort between the Papiertechnische Stiftung (PTS) and the Fraunhofer‑Gesellschaft e.V. – Institut für Integrierte Schaltungen IIS. Funded by the German Federal Ministry of Education and Research under the Innovation Fund, the project aimed to extend the technical incoming inspection of paper waste bales by applying computed tomography (CT). The collaboration combined PTS’s expertise in paper technology and waste handling with Fraunhofer IIS’s advanced CT hardware, software and data‑analysis capabilities.
The scientific work was organised into ten work packages. Work package 1 defined the technical requirements for CT and X‑ray systems, while work package 2 produced a set of synthetic paper‑bale patterns that could be scanned under controlled conditions. In work package 3 a software, database and IT framework was built to store, process and analyse the large volumes of CT data. Work package 4 established a modular reference sample set and performed a reference evaluation. Work packages 5 and 6 varied the measurement modes and increased the complexity of the samples, respectively. Work package 7 enlarged the sample and measurement volumes, and work package 8 developed mathematical models linking CT attenuation to material composition. Work package 9 validated the approach on real paper bales, and work package 10 produced a concept study for industrial implementation.
The technical results are the core of the report. The team demonstrated that real paper bales could be imaged with a spatial resolution better than 400 µm while still achieving sufficient contrast to distinguish between paper, filler, plastic and metal. Measurement times were optimised, and both single‑scan and dual‑scan modes were evaluated, increasing the range of obtainable information. Importantly, the study found that a sample volume of only a few cubic centimetres was adequate for most analyses, a significant reduction compared with conventional bulk sampling. Using the synthetic patterns, the researchers established clear relationships between material groups, density, moisture content, filler content and internal structure. Standard reference patterns covering the attenuation spectrum of paper and related products were defined, proving reproducibility across different CT systems. Image‑processing methods were developed to manipulate CT data, especially spatial operations that align the data with the preferred layering direction of the bales. Moisture was shown to have no discernible effect on the CT images, simplifying the interpretation of results.
Quantitative composition estimates from the real bale scans revealed that 75 % of the material was corrugated cardboard, 9 % graphic paper and less than 2 % other fractions, meeting the norm requirement of at least 70 % corrugated cardboard for the 1.04.01 grade. Moisture measurements by AP500 and PBSII gave 7–8 % and 8.6 %, confirming a dry state. Plastic content was measured at 0.5 % by PBSII. The remaining fraction consisted of 25 % office folders with metal, 64 % plastics in various forms, and 10 % wood. Although the CT data were acquired at a single energy, the comparison with standard materials allowed estimation of air and corrugated cardboard fractions. The mathematical model developed in work package 8 was validated against these real‑world measurements, confirming its predictive power.
The final concept study outlined how the developed CT‑based inspection could be integrated into existing paper‑waste handling lines, highlighting the potential for real‑time, non‑destructive quality control. The project’s outcomes provide a robust, reproducible framework for assessing paper‑waste composition, paving the way for more efficient sorting, recycling and resource recovery in the paper industry.