Evaluation of improvements in the separation of monolayer and multilayer films via measurements in transflection and application of machine learning approaches

Publikationen: Beitrag in FachzeitschriftArtikelForschung(peer-reviewed)

Standard

Evaluation of improvements in the separation of monolayer and multilayer films via measurements in transflection and application of machine learning approaches. / Koinig, Gerald; Kuhn, Nikolai; Barretta, Chiara et al.
in: Polymers, Jahrgang 14.2022, Nr. 19, 3926, 20.09.2022.

Publikationen: Beitrag in FachzeitschriftArtikelForschung(peer-reviewed)

Bibtex - Download

@article{8f1c4f011da5434e9fcdb31e2c15cd9c,
title = "Evaluation of improvements in the separation of monolayer and multilayer films via measurements in transflection and application of machine learning approaches",
abstract = "Small plastic packaging films make up a quarter of all packaging waste generated annually in Austria. As many plastic packaging films are multilayered to give barrier properties and strength, this fraction is considered hardly recyclable and recovered thermally. Besides, they can not be separated from recyclable monolayer films using near-infrared spectroscopy in material recovery facilities. In this paper, an experimental sensor-based sorting setup is used to demonstrate the effect of adapting a near-infrared sorting rig to enable measurement in transflection. This adaptation effectively circumvents problems caused by low material thickness and improves the sorting success when separating monolayer and multilayer film materials. Additionally, machine learning approaches are discussed to separate monolayer and multilayer materials without requiring the near-infrared sorter to explicitly learn the material fingerprint of each possible combination of layered materials. Last, a fast Fourier transform is shown to reduce destructive interference overlaying the spectral information. Through this, it is possible to automatically find the Fourier component at which to place the filter to regain the most spectral information possible.",
keywords = "Near-infrared spectroscopy (NIR), small film recycling, machine learning, multilayer films, sensor-based sorting, 2D plastic packaging",
author = "Gerald Koinig and Nikolai Kuhn and Chiara Barretta and Karl Friedrich and Daniel Vollprecht",
year = "2022",
month = sep,
day = "20",
doi = "10.3390/polym14193926",
language = "English",
volume = "14.2022",
journal = "Polymers",
issn = "2073-4360",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "19",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - Evaluation of improvements in the separation of monolayer and multilayer films via measurements in transflection and application of machine learning approaches

AU - Koinig, Gerald

AU - Kuhn, Nikolai

AU - Barretta, Chiara

AU - Friedrich, Karl

AU - Vollprecht, Daniel

PY - 2022/9/20

Y1 - 2022/9/20

N2 - Small plastic packaging films make up a quarter of all packaging waste generated annually in Austria. As many plastic packaging films are multilayered to give barrier properties and strength, this fraction is considered hardly recyclable and recovered thermally. Besides, they can not be separated from recyclable monolayer films using near-infrared spectroscopy in material recovery facilities. In this paper, an experimental sensor-based sorting setup is used to demonstrate the effect of adapting a near-infrared sorting rig to enable measurement in transflection. This adaptation effectively circumvents problems caused by low material thickness and improves the sorting success when separating monolayer and multilayer film materials. Additionally, machine learning approaches are discussed to separate monolayer and multilayer materials without requiring the near-infrared sorter to explicitly learn the material fingerprint of each possible combination of layered materials. Last, a fast Fourier transform is shown to reduce destructive interference overlaying the spectral information. Through this, it is possible to automatically find the Fourier component at which to place the filter to regain the most spectral information possible.

AB - Small plastic packaging films make up a quarter of all packaging waste generated annually in Austria. As many plastic packaging films are multilayered to give barrier properties and strength, this fraction is considered hardly recyclable and recovered thermally. Besides, they can not be separated from recyclable monolayer films using near-infrared spectroscopy in material recovery facilities. In this paper, an experimental sensor-based sorting setup is used to demonstrate the effect of adapting a near-infrared sorting rig to enable measurement in transflection. This adaptation effectively circumvents problems caused by low material thickness and improves the sorting success when separating monolayer and multilayer film materials. Additionally, machine learning approaches are discussed to separate monolayer and multilayer materials without requiring the near-infrared sorter to explicitly learn the material fingerprint of each possible combination of layered materials. Last, a fast Fourier transform is shown to reduce destructive interference overlaying the spectral information. Through this, it is possible to automatically find the Fourier component at which to place the filter to regain the most spectral information possible.

KW - Near-infrared spectroscopy (NIR)

KW - small film recycling

KW - machine learning

KW - multilayer films

KW - sensor-based sorting

KW - 2D plastic packaging

U2 - 10.3390/polym14193926

DO - 10.3390/polym14193926

M3 - Article

VL - 14.2022

JO - Polymers

JF - Polymers

SN - 2073-4360

IS - 19

M1 - 3926

ER -