Efficient Microplastic Detection in Environmental Matrices: Portable NIR Spectroscopy for Soil Analysis

The current framework for microplastic identification lacks practical and feasible methodologies for Tier 1 (Screening Method) and Tier 2 (Bulk Quantification) processes. Near-Infrared (NIR) spectroscopy offers promising features that could address both tiers, particularly with portable devices enabling on-site analysis.

This study investigates the use of miniaturized/handheld and benchtop NIR spectroscopy instruments to distinguish microplastics in soil samples, providing insights into their potential for rapid on-site screening in environmental monitoring. The performance of several handheld instruments, such as the MicroNIR 1700ES, NeoSpectra Scanner, microPHAZIR, Hefei FT-NIR, NIR-S-G1, and SCiO sensor, was compared to the reference benchtop FT-NIR NIRFlex N-500. Their ability to detect and differentiate polymers—including PMMA, HDPE, LDPE, POM, PS, EVAC, SBR, PA6, and ASA—at concentrations as low as 0.75% (w/w) without sample preparation was evaluated. Results revealed considerable variability between instruments, with the NIRFlex N-500 and NeoSpectra Scanner showing the highest accuracy compared to other handheld options. Covariance and variance analysis provided further insights into the instrumental factors influencing performance.

Quantitative analysis was performed to determine the limit of detection (LOD) using partial least squares regression (PLSR). This analysis focused on specific polymers and the instrument-dependent LOD ranges, with the goal of detecting concentrations below 0.1% w/w (less than 1000 ppm) using high-performance benchtop FT-NIR devices with sample spatial averaging. Polymer type and instrument variability were key factors influencing these ranges. The study highlights the importance of sensor selection, precise calibration, and model optimization for accurate microplastic quantification, essential for effective environmental monitoring and remediation.

This research was supported by the Austrian Science Fund (FWF) [V1014-NBL].