Predicting Properties from Near-Infrared Spectra with Machine Learning for Improved Polyolefin Differentiation

Published

January 29, 2026

Author(s)

Shuaijun Li, Robert Ivancic, Bradley Sutliff, Derek Huang, Enrique Blázquez-Blázquez, Tyler Martin, Kalman Migler, Debra Audus, Sara Orski

Abstract

The rapid increase in plastic waste necessitates innovative strategies to advance plastic recycling. As currently used, the industrial state-of-the-art sorting technology, near-infrared (NIR) spectroscopy, cannot effectively differentiate polyolefins, the single largest class of polymers by volume. Chemical similarity combined with architectural diversity in polyolefins stymie subclass delineation, such as differentiating low-density polyethylene from high-density polyethylene, due to their spectral similarity and chemical overlap. To address this challenge, we use machine learning (ML) to directly predict density, crystallinity, and short-chain branching from NIR spectra, enabling property-based sorting for more effective recycling. After testing a variety of ML models, we find that Partial Least Squares Regression provides high prediction accuracy with model simplicity. Since the resulting model leverages the correlated intensities, we develop a method to enhance interpretability by identifying the most important wavenumbers for property prediction, which we then relate to known polyolefin CH₃ NIR vibrational absorption bands. This approach provides a linkage between ML model predictions and the underlying polyolefin chemistry and confirms that our models effectively capture structure-property relationships in polyolefins, reinforcing the fundamental role of polymer chain structure in determining physical properties. These findings significantly contribute to the understanding of polyolefin differentiation using NIR spectroscopy, which could inform future advancements in property-based sorting strategies for plastic recycling efficiency.

Citation

ACS Polymers Au

Volume

6

Issue

1

Pub Type

Journals

Download Paper

https://doi.org/10.1021/acspolymersau.5c00131

Local Download

Keywords

machine learning, polyolefins, structure-property relationships, property prediction, spectroscoyp

Citation

Li, S. , Ivancic, R. , Sutliff, B. , Huang, D. , Blázquez-Blázquez, E. , Martin, T. , Migler, K. , Audus, D. and Orski, S. (2026), Predicting Properties from Near-Infrared Spectra with Machine Learning for Improved Polyolefin Differentiation, ACS Polymers Au, [online], https://doi.org/10.1021/acspolymersau.5c00131, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=960098 (Accessed September 5, 2026)

Additional citation formats