EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Chemical Fingerprints of New vs Weathered Plastics: A Machine Learning Approach

Huy Nguyen, Eric Fries, David Jones, David Megson iD, Alexa Zonderman iD, Lars Hildebrandt, Scott Coffin iD, Daniel Pröfrock, Roxana Sühring iD

Environmental Science & Technology · 2026

DOI: 10.1021/acs.est.6c03575 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Environmental Science & Technology, 2026. Verfasst von Huy Nguyen und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemistry and Biology, Toronto Metropolitan University, TorontoM5B 2K3, Canada. Hauptthemen laut MEDLINE: Machine Learning, Plastics. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 74 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Canada Innovation Fund, MITACS Accelerate Fellowship, CIHR, NSERC and SSHRC through the New Frontiers in Research Fund - Exploration.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Regulatory initiatives to mitigate plastic pollution face challenges in identifying microplastics and their sources. Recent advances in computational fingerprinting offer opportunities for the forensic investigation of plastic pollution sources. Here, we present an open-source computational workflow integrating multi-instrumental, nontargeted (organic and inorganic) data from three distinct mass spectrometry methods to evaluate two approaches for potential plastic pollution source tracking. The workflow includes a novel data imputation, normalization, and feature selection strategy with a Random Forest classifier applied to data sets of 42 pristine (store-bought) and 21 weathered (environmentally collected) plastics from U.S. retail products and California beaches. To develop a workflow for source identification, we first, evaluated an “Universal Fingerprint” approach, which tested whether chemical profiles from pristine, store-bought plastic products could be used to identify weathered environmental counterparts based on their chemical additive patterns. This approach yielded low classification accuracy due to high geographic and manufacturer-specific formulation variability for plastic products that the random forest model could not sufficiently resolve. Second, we evaluated a “Suspect–Source Comparison” approach that simulated a scenario where the chemical fingerprint of an environmentally sampled plastic was matched to the fingerprint of a suspect source. Under this paradigm, the workflow achieved a high classification accuracy across all platforms (up to 1.00 MCC for ICP-MS/MS and 0.79 for HPLC-QToF-MS). Additionally, we found that the different analytical techniques functioned best as complementary data sets to distinguish product categories, while combining all raw data sets prior to classification introduced excessive variability that hindered the classification. Our results demonstrate the feasibility of using relative chemical similarity to connect environmental plastics to potential local pollution sources. The presented workflow establishes a methodological foundation for localized source tracking, environmental litigation, and targeted regulatory monitoring.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningPlasticsRandom ForestEnvironmental MonitoringMicroplasticsMass Spectrometry

Stichwörter der Autor:innen

clustering analysesfeature selectionmicroplasticsmulti-instrumental nontargeted fingerprintingplastic additivesrandom foreststatistical analysissupervised machine learning

Substanzen

PlasticsMicroplastics

Autor:innen und Institutionen

  • Department of Chemistry and Biology, Toronto Metropolitan University, TorontoM5B 2K3, Canada.
  • Department of Natural Science, Manchester Metropolitan University, ManchesterM1 5GD, U.K.
  • Chemistry Matters Inc., CalgaryT2N 3P7, Canada.
  • Department for Inorganic Environmental Chemistry, Helmholtz-Zentrum Hereon, Geesthacht21502, Germany.
  • California State Water Resources Control Board, Sacramento, California95814, United States.

ORCID-Profile: David Megson, Alexa Zonderman, Scott Coffin, Roxana Sühring

Förderung

  • Canada Innovation Fund
  • MITACS Accelerate Fellowship
  • CIHR, NSERC and SSHRC through the New Frontiers in Research Fund - Exploration

Zitieren

APA

Nguyen, H., Fries, E., Jones, D., Megson, D., Zonderman, A., Hildebrandt, L., Coffin, S., Pröfrock, D., & Sühring, R. (2026). Chemical Fingerprints of New vs Weathered Plastics: A Machine Learning Approach. Environmental Science & Technology. https://doi.org/10.1021/acs.est.6c03575

Vancouver

Nguyen H, Fries E, Jones D, Megson D, Zonderman A, Hildebrandt L, et al. Chemical Fingerprints of New vs Weathered Plastics: A Machine Learning Approach. Environmental Science & Technology. 2026. doi:10.1021/acs.est.6c03575

BibTeX
@article{Nguyen2026,
  author = {Nguyen, Huy and Fries, Eric and Jones, David and Megson, David and Zonderman, Alexa and Hildebrandt, Lars and Coffin, Scott and Pröfrock, Daniel and Sühring, Roxana},
  title = {Chemical Fingerprints of New vs Weathered Plastics: A Machine Learning Approach},
  journal = {Environmental Science & Technology},
  year = {2026},
  doi = {10.1021/acs.est.6c03575}
}
Kostenlos starten →