Disclaimer:

    The views, information, or opinions expressed in the Industry News RSS feed belong solely to the author and do not necessarily represent those of IDEX Health & Science and its employees.

    Hyperspectral Imaging Could Automate, Improve Plastics Recycling

    Article obtained from Photonics RSS Feed.

    A new method using NIR hyperspectral imaging (HSI) and chemometrics could make it possible to sort between different types of plastic and between different flame retardants added to plastic — a necessity for recycling plastics more economically. The research took place at the University of Extremadura and the University of Copenhagen.

    Three classification models were developed based on a combination of chemometrics techniques and HSI. All three methods were found suitable for classifying the plastic samples, but the best results were achieved with decision tree (DT) as the classification technique.

    Near-infrared hyperspectral imaging of samples of plastics and their classification as a function of type of flame retardant: 1,2,5,6,9,10-hexabromo-cyclododecane, HBCD (red samples); 3,5-tetrabromobisphenol A, TBBPA (yellow samples); Pentabromophenyl ether, Deca-BDE (green samples); and reference, REF (blue samples). Courtesy of D. Caballero, M. Bevilacqua, and J.M. Amigo.
    Images obtained from NIR reflectance spectroscopy hyperspectral imaging (NIRS-HSI) were collected in the wavelength range of 1100 to 2250 nm with a spectral resolution of 4.85nm (115 bands). The samples were illuminated with diffuse white light, and the final pixel resolution was 300 µm. The spectra were pre-processed to remove outliers and noise. The training samples were evaluated and classified by applying three different classification models — DT, partial least square-discriminant analysis, and hierarchical model (HM) — in a pixel-by-pixel fashion and by analyzing sample-by-sample. Once the best classification model was obtained, this model (DT) was evaluated on the NIRS-HSI of real samples of plastic in order to evaluate the polymers and the flame retardants doping waste plastics of different common brands in use. 

    The applications of the models on real samples led to correct classification of 100 percent, in spite of the differences in texture, shape, and orientation of these samples. The researchers concluded that the application of the DT classification technique with HSI could be used to sort plastic samples with respect to the type of polymer and the flame retardant used with a high degree of accuracy in an automated way, potentially saving the plastic and waste recycling industries time and money.

    Professor José Amigo at the University of Copenhagen said, “Recycling plastics has been studied for many years. Indeed, some commercial cameras separating a limited number of plastic types have been available for some time. However, in this research, we wanted to go a step further to separate plastics containing flame retardants. Moreover, the proposed methodology was tested with real samples that can be found in current recycling lines.”

    The research was published in the Journal of Spectral Imaging (https://www.impopen.com/subs/jsi/v8/I08_a1.pdf).

    READ MORE
    Jan, 24 2019 |

    IDEX Health & Science is the global authority in fluidics and optics, bringing to life advanced optofluidic technologies with our products, people, and engineering expertise. Intelligent solutions for life.