Comparison of hyperspectral imaging and FTIR spectroscopy for microplastic polymer identification: Proposal of a...

In this study, hyperspectral imaging (HSI) was systematically compared with Fourier-transform infrared spectroscopy (FTIR/micro-FTIR) for microplastic (MP) polymer identification using environmental samples collected during a 12-month monitoring campaign at six stations along 300 km of the Po River (Italy), following the Marine Strategy Framework Directive (MSFD, 2008/56/EC) guidelines adapted to freshwater environments. In total, 11,828 MP particles (330 μm-5 mm) were collected and analyzed to assess MP abundance, morphology, color and polymer composition. Across the monitored sites and sampling period, MP concentrations ranged from 0.03 to 12.7 n°/m³, generally lower than values reported for other major world rivers. Fragments and foam dominated the MP categories (56 % and 24 %, respectively), and white was the dominant color (57 %), while PE, PP and EPS were the most abundant polymers (44.9 %, 28.8 % and 22.5 %, respectively). The particle-by-particle comparison between HSI in the short-wave infrared range (1000-2500 nm) and FTIR or micro-FTIR, depending on particle size, yielded an overall polymer identification agreement exceeding 99 %. HSI significantly improved analytical efficiency, requiring only 8 min for classification of 100 particles compared to 300-500 min using FTIR/micro-FTIR spectroscopy. Black MP particles (8 % of the total) were systematically excluded from HSI analysis and characterized exclusively by FTIR/micro-FTIR. The results support the proposal of a new analytical protocol using HSI as the primary tool for polymer classification, with FTIR/micro-FTIR reserved for black or unclassified particles, enabling faster, scalable and reliable MP monitoring.