A global push for a circular bioeconomy has accelerated bioplastics adoption. These materials offer a sustainable alternative to fossil-based plastics. However, this progress has exposed a critical vulnerability in our recycling system. This issue was a central topic at the Bioplastics Recycling Cluster Talk #1.
Many waste sorting plants rely on Near-Infrared (NIR) sensors. These sensors identify plastics by reading their spectral signatures. A significant problem arises with bioplastics, for example PLA (polylactic acid). Its spectral fingerprint looks very similar to PET, a cornerstone of traditional plastic recycling. This spectral confusion leads to a critical error. Optical sorters misidentify PLA as PET and send it into the wrong stream.
This mis-sorting causes major contamination. Even a small amount of bioplastics can degrade the quality of recycled fossil-based materials. This, in turn, lowers the value of recycled products. It can also lead recyclers to reject entire loads. These factors create economic losses and undermine efforts toward a circular economy. The volume of bioplastics in the waste stream is projected to grow. Therefore, this “scaling problem” will worsen. We must develop smarter solutions for Bioplastics Recycling Europe.
The MoeBIOS solution: smarter sorting with AI
The MoeBIOS project is tackling this challenge head-on. We are advancing waste management through technological innovation. Our solution is not to rebuild plants. Instead, we make existing sorting equipment smarter through a groundbreaking combination of advanced imaging and artificial intelligence.
First, let’s understand the technology. Conventional NIR works by reflecting light off a material to read its chemical fingerprint. MoeBIOS improves on this with Hyperspectral Imaging (HSI). This advanced technology captures a series of images at different light wavelengths, from visible blue to near-infrared. The process creates a detailed data structure called a “hypercube”. In this hypercube, every single pixel has a unique spectral fingerprint. This level of detail allows us to accurately distinguish between even chemically similar plastics like PLA and PET, where a standard NIR sensor would fail.
Our project integrates HSI with AI-powered digital libraries developed by our partner, LEITAT. These libraries are a comprehensive database. They teach sorting machines to recognize new bioplastics with high efficiency and accuracy. Our partner, PICVISA, then implements this software into real-world sorting equipment. This software-based approach is cost-effective because it avoids hardware replacement. We are also exploring the use of Laser-Induced Breakdown Spectroscopy (LIBS). This is a complementary solution for black plastics, which are notoriously difficult for NIR to detect.
A practical roadmap for bioplastics recycling Europe
Ultimately, the goal of MoeBIOS is to make bioplastics a solution for a sustainable future. We are implementing and validating our technology across three key waste streams: plastic packaging, synthetic textiles, and agricultural plastics. This approach ensures our solution is robust and scalable. Our collaboration with partners like LEITAT and PICVISA ensures that innovations move beyond the lab and into real-world applications. By enabling accurate and efficient sorting, our project will help keep recycling streams cleaner and more reliable. This practical roadmap for Bioplastics Recycling Europe will significantly improve waste management efficiency, reduce environmental impact, and advance the transition to a truly circular bioeconomy.
Learn more
To follow MoeBIOS’s journey toward sustainable bioplastics recycling:
- Visit our project website
- Follow us on LinkedIn
- Join us on X (formerly Twitter)
Stay informed, stay engaged—and help us redesign the future of sustainable packaging.