The fate and toxicity of nanoplastics (NPs) in the environment is largely determined by their stability. We explored how water composition, nanoplastic size, and surface carboxyl group density influenced the aggregation of polystyrene (PS) NPs in fresh water. Unfunctionalized 200, 300, 500, and 1000 nm PS NPs and 310 nm carboxylated PS NPs with carboxyl group densities of 0.35 and 0.6 mmol g⁻¹ were used to simulate pristine and aged NPs. Natural water matrices tested in this study include synthetic surface water (SSW), water from the Schie canal (Netherlands) and tap water. Suwannee River Natural Organic Matter (SRNOM) was included to mimic organic matter concentrations. In CaCl₂, we found PS NPs are more stable as their size increases with the increase of the critical coagulation concentration (CCC) from 44 mM to 59 mM and 77 mM for NP sizes of 200 nm, 300 nm and 500 nm. Conversely, 1000 nm PS NPs remained stable even at 100 mM CaCl₂. Increasing the carboxyl group density decreased the stability of NPs as a result of the interaction between Ca²⁺ and the carboxyl group. These results were consistent with the mass of Ca²⁺ adsorbed per mass of NPs. The presence of SRNOM decreased the stability of PS NPs via particle bridging facilitated by SRNOM. However, in SSW, Schie water and tap water with low divalent cation concentrations, the hydrodynamic size of PS NPs did not change, even at prolonged durations up to one week. We concluded that PS NPs are unlikely to aggregate in water with low divalent cation concentrations, like natural freshwater bodies. Ecotoxicologists and water treatment engineers will have to consider treating PS NPs as colloidally stable particles as the lack of aggregation in fresh surface water bodies will affect their ecotoxicity and may pose challenges to their removal in water treatment.

Nanoplastics are detected in surface water, yet accurately quantifying their particle number concentrations remains a significant challenge. In this study, we tested the applicability of a gold-labelling method to quantify nanoplastics in natural organic matter (NOM) containing water matrices. Gelatin-coated gold nanoparticles (Au-gel NPs) form conjugates with nanoplastics via electrostatic interaction which produces peak signals which can be translated into particle number concentration using single-particle inductively coupled plasma–mass spectrometry (SP-ICP-MS). We used water samples with various NOM concentrations, with and without the addition of 1 10 ⁷ particle ^–1 nanoplastics. Our results indicate that nanoplastics in low NOM samples (,1 mg·C L ¹) could be successfully quantified. However, in high NOM samples (.15 mg·C L ¹), only 13–19% of added nanoplastics were successfully quantified. Further digestion to remove NOM yielded only 10% of spiked nanoplastics. This discrepancy in high NOM samples could likely be attributed to the competition between nanoplastics and NOM existing in the water sample to bind with Au-gel NPs. Our study highlights the suitability of the Au-gel labelling method for quantifying nanoplastics in low NOM water samples. Nevertheless, further optimization, including pre-digestion steps, is essential to apply this method for high NOM water samples effectively.