Production

"blackValue: black's bright future"

blackValue is implemented with compact SMD technology.
© Photo Fraunhofer FHR

blackValue is implemented with compact SMD technology.

The terahertz line scan camera, which was developed by Fraunhofer FHR, allows the sorting of colored and black plastics according to type.
© Photo Fraunhofer FHR

The terahertz line scan camera, which was developed by Fraunhofer FHR, allows the sorting of colored and black plastics according to type.

The center console in the car. The filing trays in the office. The laptop casing. The telephone, the monitor, the keyboard. All of these objects are made of black plastic. But they are not destined to live forever. What happens to them then?

At present, approximately 50 % of plastic waste is recycled in Europe. The recycling method most frequently used is thermal recycling, where the waste is incinerated to generate energy. The recyclate produced during the material recycling process is used to produce a new product. The recycling process is, however, complicated by the large number of plastics as well as the different types of additives such as colorants, plasticizers, flame retardants or UV blockers. Products that are produced from mixed recyclate have poorer material properties than products made from new plastics. End products of equal value are, however, essential to achieve a closed recycling loop. Sorting systems that are capable of identifying materials and are used in plastics recycling must therefore also be capable of correctly classifying material types.

The most efficient plastics recycling techniques currently available use hyperspectral cameras. When sorting black or very dark plastics, the reflected part in the visible as well as in the infrared wavelength range is, however, insufficient as black almost completely absorbs the spectrum. The increased use of composite materials also poses as yet unresolved problems in the area of quality control due to the fact that the materials are identified solely on the basis of their surface.

Raw materials from scrap

The tonnage price for plastic waste with a high degree of purity is significantly higher than the price for low purity waste. Given that the economic potential of black plastic waste amounts to approx. one billion euro, recyclers are now using technically sophisticated methods in an attempt to sort the plastic fragments according to type. The Fraunhofer-Gesellschaft has set up a strategic alliance to exploit this potential. The alliance comprises the Fraunhofer institutes FHR, IAIS and IOSB. Under the management of Fraunhofer FHR, the research activities of the three participating institutes are pooled in the project blackValue®. The central aim of this initiative is the development of a real-time capable sensor and evaluation system for the identification of plastic materials, particularly for the recycling of black plastics. The system currently operates at up to 94 GHz. In the current development phase, the frequency has been raised to 120 GHz to achieve even better results. Here, the researchers are using a state-of-the-art radar camera in the terahertz range together with conventional sensors. The research conducted by Fraunhofer FHR in the areas of radar technology and chip design have led to the development of cost-effective, highly integrated radar systems on a silicon-germanium basis.

The secret to success: teamwork

A full spectroscopic analysis over a large terahertz range is not economically viable due to the large amount of hardware that would be needed. The researchers at Fraunhofer FHR therefore decided to use a little trick: they evaluate a frequency range with a very low bandwidth in the lower terahertz range and at the same time keep an eye on the availability of commercial system components. This optimization does, however, place special demands on the evaluation algorithms that are used.

To respond to these demands and ensure correct allocation, Fraunhofer IAIS uses algorithms from the area of machine learning. In this subfield of computer science, an algorithm is trained within the framework of an initial test phase. In the subsequent detection phase, the developer shows the system further samples and the system must allocate the samples to the various types on the basis of its training. The algorithm currently has a precision rate of 94-99 %.

A corresponding prototype for sorting plastic waste is being developed at Fraunhofer IOSB. The plastic waste is placed on a conveyor belt and passes the FHR camera at a speed of two to three meters per second, whereby the system needs just 35 milliseconds to decide whether or not the plastic particle should be blown out. Just two persons are needed to load and unload the system. The system has a throughput of approximately 30 tons per second.

With blackValue®, it is now possible to recycle black plastics on an industrial scale thanks to the clean separation of plastics according to type or color and increased economic efficiency.