A simple method for the selective quantification of polyethylene, polypropylene, and polystyrene plastic debris in soil by pyrolysis-gas chromatography/mass spectrometry

The lack of adequate analytical methods for the quantification of plastic debris in soil challenges a better understanding of their occurrence and fate in the terrestrial environment. With this proof-of-principle study, we developed a simple and fast method for the selective quantification of the three most environmentally relevant polymers polyethylene (PE), polypropylene (PP), and polystyrene (PS) in soil using pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS). In order to facilitate the preparation of calibration series and to better account for the heterogeneity of soil matrix, polymers were dissolved in 1,2,4-trichlorobenzene (TCB) at 120 °C. Thereby, liquid sample aliquots from up to 4 g of solid sample became amenable to pyrolysis without further preparation. To evaluate the performance of this approach, three reference soils with 1.73–5.16 % organic carbon (C_org) were spiked at 50 and 250 μg g⁻¹ of each polymer and extracted with TCB. A prior cleanup step with methanol, flocculation with KAl(SO₄)₂, and Fenton digestion were tested for their suitability to reduce potentially interfering C_org. Calibration curves responded linearly (adj. R² > 0.996) with instrumental detection limits of 1–86 ng corresponding to method detection limits of 1–86 μg g⁻¹. The measurement repeatability was 3.2–7.2 % relative standard deviation. Recoveries of 70–128 % were achieved for plastic contents of 250 μg g⁻¹ extracted with TCB without prior cleanup from soils with less than 2.5 % C_org. A higher C_org particularly interfered with the quantification of PE. The addition of non-target polymers (polyethylene terephthalate, polyvinyl chloride, poly(methyl methacrylate), and tire wear particles) did not interfere with the quantification of the analytes highlighting the selectivity of the method. Further research is needed to determine low plastic contents in soils exceeding 2.5 % C_org. With 1–3 h processing time per sample, our method has the potential for routine analyses and screening studies of agricultural systems to be complemented with microspectroscopic techniques for additional information on particle shapes and sizes.