Detecting Polystyrene Nanoparticles in Environmental Samples: A Comprehensive Quantitative Approach Based on TD-PTR-MS and Multivariate Standard Addition

Abstract

Submicrometer-sized plastic particles (nanoplastic; NP) have been detected in a large variety of different ecosystems. They occur in small quantities within a complex organic matrix comprising a plethora of compounds. A robust quantification of the NP concentration thus requires the development of a comprehensive analytical workflow to handle potential interferents. Thermal desorption–proton-transfer reaction–mass spectrometry (TD-PTR-MS) creates the necessary chemical selectivity to distinguish NP signals from the organic matrix. Nevertheless, the recorded raw mass spectra are too complex for direct interpretation, and further signal clustering/scoring is required for a more in-depth analysis. Here, we resolved this problem in a novel workflow, which combines non-negative matrix factorization (NMF) and multivariate standard addition (MSA). This allows us to mathematically separate the NP’s signature from the mixture, as showcased for polystyrene nanoparticles. The method produces an unequivocal and matrix-corrected NP fingerprint for identification and quantification. MSA and NMF enabled us to quantify polystyrene NP in different environmental samples in the lower nanogram range. The mass concentration of polystyrene NP in Waal River water sampled close to Nijmegen, the Netherlands, was 4.7 ± 0.65 ng/mL and 39 ± 0.70 ng/g in sand samples from the river’s shore. A sand sample from a local playground in Nijmegen exhibited a higher concentration of 129 ± 1.1 ng/g.

The proposed novel workflow is built on sensitive mass spectrometry and a machine learning approach to data interpretation that enables identification and precise quantification of nanoplastic concentrations in complex environmental samples. This method will allow a deeper understanding of nanoplastic contamination in the environment.