Mid-Infrared Spectroscopy and Machine Learning for Nondestructive Detection of Inapparent Deterioration in Acrylic Waterborne Coatings for Wood - key study approach

Aug 5, 2026

Upload by Sage

Nondestructive detection of inapparent deterioration in waterborne acrylic coatings containing cellulose nanoﬁbers (CNFs) for wood is achieved by combining mid-infrared spectroscopy with machine learning. Film specimens are evaluated before and after 500 h of accelerated weathering, representing about one year outdoors, and show no obvious appearance change. Oxidative degradation is identified mainly in the acrylic resin, with CNFs appearing to mitigate damage. ATR-FTIR alone cannot clearly reveal pre-apparent chemical change, while ML using absorbance features from 400–4000 cm−1 enables classification, with SVM providing the best predictive accuracy and RF highlighting key wavenumber zones.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Introduction

Motivation: sustainable wood use and coating deterioration

Challenge: lack of methods for pre-apparent (inapparent) deterioration

Proposed method: infrared spectroscopy + machine learning classification

Experimental approach: accelerated weathering and spectral acquisition

[www.advsustainsys.com](www.advsustainsys.com) Mid-Infrared Spectroscopy and Machine Learning for Nondestructive Detection of Inapparent Deterioration in Acrylic Waterborne Coatings for Wood Yoshikuni Teramoto,* Takumi Ito, Chihiro Yamamoto, Toshiyuki Takano, and Hironari Ohki This study presents an approach for nondestructive detection of inapparent deterioration in waterborne acrylic coatings (containing cellulose nanoﬁbers (CNFs)) for wood by using mid-infrared spectroscopy and machine learning. The method evaluates ﬁlms that mimic coatings before and after 500 h of accelerated weathering, equivalent to roughly 1 year of outdoor exposure. No noticeable transformation in ﬁlm appearance is evident with a spectrophotometer following the accelerated weathering. Chemiluminescence analysis indicates oxidative degradation predominantly in the acrylic resin, an impact that the CNFs seem to mitigate. Whereas attenuated total reﬂectance (ATR)-Fourier transform infrared (FTIR) spectroscopy commonly identiﬁes chemical changes in visibly degraded coatings, it does not clearly discern prior, inapparent deterioration. In this context, machine learning algorithms (such as k-nearest neighbors, decision tree, random forest (RF), and support vector machine (SVM)) categorize these nuanced changes by using the absorbance from 400 to 4000 cm−1 as explanatory variables. The SVM model exhibits the highest predictive accuracy, and the RF recognizes crucial variables in some wavenumber zones. This approach has the potential for enhancing recoating schedules, cutting costs, and encouraging sustainable use of wood. appeal to woody materials. [1–3] In recent years, the increasing demand for sustainable products has led to a worldwide increase in the number of medium- and high-rise wooden buildings[4] ; thus the importance of wood coatings should increase. However, in general, long-term deterioration of coating barrier properties is not fully understood, and its elucidation continues to be of substantial scientiﬁc and industrial interest.[5] Regarding coated wood, chemical and physical degradation of the coating causes pinholes, cracks, or scratches that rapidly progress the biological degradation of the wood (cell wall degradation initiated by bacteria and fungi) .[6] It is therefore important to extend the period that precedes deterioration and learn how to predict it. However, current evaluation methods[7] mainly use apparent indicators that manifest during deterioration (such as cracks, blisters, and peeling of the paint ﬁlm) and comparisons of discoloration as well as fading 1. Introduction Wood coatings are widely used because they are relatively inexpensive and eﬀective in protecting as well as adding aesthetic Y. Teramoto, C. Yamamoto, T. Takano Division of Forest and Biomaterials Science, Graduate School of Agriculture Kyoto University Kitashirakawa Oiwake-cho, Sakyo-ku, Kyoto 6068502, Japan E-mail: [teramoto.yoshikuni.3e@kyoto-u.ac.jp](teramoto.yoshikuni.3e@kyoto-u.ac.jp) T. Ito, H. Ohki Gen Gen Corporation 74 Aza Nakano Ori, Kamori-cho, Tsushima city, Aichi 4960005, Japan The ORCID identiﬁcation number(s) for the author(s) of this article can be found under [https://doi.org/10.1002/adsu.202300354](https://doi.org/10.1002/adsu.202300354)[ ](https://doi.org/10.1002/adsu.202300354)© 2023 The Authors. Advanced Sustainable Systems published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. DOI: 10.1002/adsu.202300354 Adv. Sustainable Syst. 2024, 8, 2300354 2300354 (1 of 9) with product samples. It is thus diﬃcult to obtain information on preliminary (inapparent) stages before apparent deterioration. If it is possible to quickly and nondestructively diagnose chemical deterioration of wood coatings at a stage that cannot be visually recognized,

FAQs

What problem does the study address in wood coating maintenance?

The work targets the difficulty of detecting chemical deterioration before visible, apparent damage occurs, so recoating timing can be optimized and biodegradation prevented.

How are the coating samples tested for early-stage (inapparent) deterioration?

Acrylic waterborne coating-like films are subjected to accelerated weathering (500 h, approximating one year outdoors), and mid-infrared ATR-FTIR spectra are collected before and after treatment.

Why is machine learning needed instead of relying only on ATR-FTIR?

ATR-FTIR can identify chemical changes when coatings are visibly degraded, but it does not clearly discern earlier, inapparent deterioration. Machine learning uses full-spectrum absorbance (400–4000 cm−1) to classify nuanced changes.

Which machine learning model performs best and what does it contribute?

The SVM model shows the highest predictive accuracy, while the RF model recognizes important variables across specific wavenumber zones, indicating which spectral regions are most informative for detection.

Aug 5, 2026

Upload by Sage