Machine learning driven methodology for enhanced nylon microplastic detection and characterization

Abstract

In recent years, the field of microplastic (MP) research has evolved significantly; however, the lack of a standardized detection methodology has led to incomparability across studies. Addressing this gap, our current study innovates a reliable MP detection system that synergizes sample processing, machine learning, and optical photothermal infrared (O-PTIR) spectroscopy. This approach includes examining high-temperature filtration and alcohol treatment for reducing non-MP particles and utilizing a support vector machine (SVM) classifier focused on key wavenumbers that could discriminate between nylon MPs and non-nylon MPs (1077, 1541, 1635, 1711 cm−1 were selected based on the feature importance of SVM-Full wavenumber model) for enhanced MP identification. The SVM model built from key wavenumbers demonstrates a high accuracy rate of 91.33%. Results show that alcohol treatment is effective in minimizing non-MP particles, while filtration at 70 °C has limited impact. Additionally, this method was applied to assess MPs released from commercial nylon teabags, revealing an average release of 106 particles per teabag. This research integrates machine learning with O-PTIR spectroscopy, paving the way for potential standardization in MP detection methodologies and providing vital insights into their environmental and health implications.

Assessing the inconsistency of microplastic measurements in foods and beverages

“…Hernandez et al (2019) argued that they applied NTA to SEM image analysis as a validation method, and NTA counts provided similar results to the counts estimated by SEM. In our recent study (Yang et al, 2024), our findings were substantially lower than those reported by Busse et al (2020), even though both studies performed particle-based analysis. We attribute the potential misidentification of non-MP as MPs to the use of Raman.…”

“…We attribute the potential misidentification of non-MP as MPs to the use of Raman. For instance, PE particles identified by Busse in the leachate of nylon teabags could also be behenamide, a common slip additive used in PE plastic (Yang et al, 2024). In addition, a rigorous spectra analysis was absent in Hernandez et al (2019) and Busse et al (2020).…”

“…The relationship between MPs number and filter pore size is plotted in Figure 5. The correlation coefficient between the abundance of MPs and the pore size of the filter was measured for salt studies (Iniguez et al, 2017;Karami et al, 2017;Yang et al, 2015), canned fish studies (Akhbarizadeh et al, 2020;Diaz-Basantes et al, 2022;Karami et al, 2018), BW studies (Mason et al, 2018;Oßmann et al, 2018;Schymanski et al, 2018;Zuccarello et al, 2019;Qian et al, 2024), and teabag studies (Busse et al, 2020;Hernandez et al, 2019;Yang et al, 2024), with the result of −0.7, −0.53, −0.46, and −0.87, respectively. The negative correlation coefficient indicates an inverse relationship between the variables.…”

“…The inconsistency between studies (Table 1) on teabags was interpreted by considering MPs larger than 1 µm, as this was the common size class investigated in all four studies (Afrin et al, 2022;Busse et al, 2020; Yang et al, 2024). Hernandez et al (2019) examined nylon and PET teabags, and they reported the average result of released MPs from a single teabag.…”

“…However, they did not indicate the specific result for nylon or PET teabags in this class size (> 1 µm). Busse et al (2020) and Yang et al (2024) restricted their study to teabags made of nylon, and Afrin et al (2022) did not mention the plastic material of commercial teabags. There were great variations in the reported number of released MPs between studies, but it was doubtful that the results in all reviewed studies were MPs released from the same type of plastic teabag.…”