Abstract
Accurately identifying microplastics (MPs) poses a challenge because environmental and human activities, like thermal oxidation (common in wildfires, urban fires, and waste burning) and mechanical abrasion (common in riverine and marine environments), chemically modify the plastic structure. Conventional Fourier transform infrared (FTIR) libraries lack the accuracy to identify these plastics because degradation alters their spectra-thermal oxidation adds new peaks, and abrasion causes characteristic peaks to fade. This creates a knowledge gap, leading to MP misidentification and flawed risk assessments for plastic pollution. To address this, we developed a new, environmentally relevant spectral library that significantly improved identification, increasing match rates by 7.3% for thermally oxidized plastics and by 23.8% for mechanically abraded plastics. We did this by subjecting seven commercial polymers (polypropylene [PP], polystyrene, rayon, low-density polyethylene [LDPE], linear low-density polyethylene, high-density polyethylene, and polyethylene terephthalate) to controlled thermal oxidation (100-200°C for 1-24 h) and by exposing PP and LDPE preproduction pellets to 40 days of mechanical abrasion (33 rpm; 3.41 rad s - 1). We then collected attenuated total reflectance (ATR)-FTIR spectra from these altered plastics and incorporated them into two newly developed spectral libraries. We tested these new libraries on 62 environmental MP samples (thermally oxidized) and 15 lab-generated particles (abraded), confirming a significant improvement in identification accuracy. Our findings underscore the importance of incorporating spectra obtained from plastics exposed to environmentally weathering processes into FTIR libraries, which represent a scientific advancement and a practical tool for more accurate MP identification and monitoring in environmental research.