Occurrence and fate of microplastics at two different drinking water treatment plants within a river catchment

Highlights

• MPs ≥ 1 μm were analysed at different stages of drinking water treatment at two DWTPs.

• Scanning electron microscopy and micro-Raman spectroscopy were employed.

• MP content ranged from <20 to >1200 L⁻¹, while fragments <10 μm always prevailed.

• The number of MPs varied both between the DWTPs and along the treatment chain.

• Current water treatment technology is capable of removing almost 90% of MPs ≥ 1 μm.

Abstract

Microplastics (MPs) are emerging globally distributed pollutants of aquatic environments, and little is known about their fate at drinking water treatment plants (DWTPs), which provide a barrier preventing MPs from entering water for human consumption. This study investigated MPs ≥ 1 μm in raw and treated water of two DWTPs that both lie on the same river, but the local quality of water and the treatment technology applied differ. In the case of the more complex DWTP, MPs were analysed at 4 additional sampling sites along the treatment chain. The content of MPs varied greatly between the DWTPs. There were 23 ± 2 and 14 ± 1 MPs L⁻¹ in raw and treated water, respectively, at one DWTP, and 1296 ± 35 and 151 ± 4 MPs L⁻¹ at the other. Nevertheless, MPs comprised only a minor proportion (<0.02%) of all detected particles at both DWTPs. With regard to size and shape of MPs, the majority (>70%) were smaller than 10 μm, and only fragments and fibres were found, while fragments clearly prevailed. The most frequently occurring materials were cellulose acetate, polyethylene terephthalate, polyvinyl chloride, polyethylene, and polypropylene. Much higher total removal of MPs was achieved at the DWTP with a higher initial MP load and more complicated treatment (removal of 88% versus 40%); coagulation-flocculation-sedimentation, deep-bed filtration through clay-based material, and granular activated carbon filtration contributed to MP elimination by 62%, 20%, and 6%, respectively. Additionally, results from this more complex DWTP enabled to observe relationships between the removal efficiency and size and shape of MPs, particularly in the case of the filtration steps.

Introduction

Microplastics (MPs) are being detected in various aquatic environments worldwide, including both seawater and freshwater (Cole et al., 2011; Novotna et al., 2019). Some studies have also revealed the occurrence of MPs in raw water that supplies drinking water treatment plants (DWTPs) (Pivokonsky et al., 2018; Mintenig et al., 2019; Wang et al., 2020). High MP quantities were reported in the case of surface water resources. Raw water originating from a large dam, a small dam, and a river (all located within the Czech Republic) comprised 1473 ± 34, 1812 ± 35, and 3605 ± 497 MPs L⁻¹ on average, respectively (Pivokonsky et al., 2018), and even more MPs, 6614 ± 1132 MPs L⁻¹, were found in raw water from the Yangtze River (China) (Wang et al., 2020). Both studies determined MPs down to the size of 1 μm, and the vast majority of MPs were <10 μm (Pivokonsky et al., 2018; Wang et al., 2020). By contrast, negligible amounts of MPs, with a maximum of 7 MPs m⁻³, were detected in raw water from groundwater wells of the Oldenburg-East-Frisian water board (Germany), but only MPs > 20 μm were determined (Mintenig et al., 2019). In addition to the diversity of water sources, the differences in MP content might have arisen from the variations in analytical approaches, particularly the different lower size limits of detected MPs, and analysing MPs in the μm size range is presumably important to avoid underestimation of the results (Novotna et al., 2019). Nevertheless, studies on MPs in raw water are too scarce to derive any clear conclusions.

MPs have also appeared in water intended for human consumption, including DWTP-treated water (Pivokonsky et al., 2018; Mintenig et al., 2019; Wang et al., 2020), water from tap or public drinking fountains (Shruti et al., 2020; Tong et al., 2020), and bottled water (Mason et al., 2018; Oßmann et al., 2018; Schymanski et al., 2018). The MP numbers in drinking water varied greatly, as did the lower size boundaries of analysed MPs. The overall highest MP content was observed in bottled water in a study where MPs as small as 1 μm were determined, up to 6292 ± 10,521 MPs L⁻¹ (Oßmann et al., 2018). MPs raise health concerns not only due to possible physical effects associated with particle ingestion but also owing to the content of monomers or additives in plastic materials or the capability of MPs to adsorb and desorb toxic chemicals (Wang et al., 2018; World Health Organisation, 2019). However, most toxicological studies have been devoted to aquatic organisms (de Sá et al., 2018; Triebskorn et al., 2019), and despite some studies reporting that MPs interact with human cells (Schirinzi et al., 2017; Triebskorn et al., 2019), the possible effects of MPs on human health are still largely unknown (Wright and Kelly, 2017; Revel et al., 2018; World Health Organisation, 2019).

Currently, no limitations for MP content in drinking water exist, but DWTPs are of great interest as barriers preventing MPs from entering public drinking water supplies (Novotna et al., 2019). Interestingly, when focusing on treated water originating from surface water bodies with high MP contents, the MP numbers were by 70–86% lower than those in the corresponding raw water samples. The average values for treated water ranged from 338 ± 76 to 930 ± 71 MPs L⁻¹ (Pivokonsky et al., 2018; Wang et al., 2020).

In the study by Pivokonsky et al. (2018), the influence of the employed water treatment steps on MP removal was proposed, despite MPs being analysed only in raw and treated water. The overall MP removal was higher (81–83%) at DWTPs that performed two-step separation of aggregates and granular activated carbon (GAC) filtration in addition to conventional coagulation-flocculation; a DWTP with one-step separation of aggregates and no GAC filtration removed 70% of MPs. Wang et al. (2020) measured MPs after each treatment step at one DWTP and found that coagulation-flocculation-sedimentation, sand filtration, and GAC filtration were all involved in MP removal. Contradictory findings were reported by Zhang et al. (2020), who did laboratory investigations with particles of 0.18–125 μm and observed only minor MP removal by coagulation-flocculation-sedimentation for all particle sizes, while filtration was very effective. Additionally, some other laboratory studies on MP removal were conducted, but the MP characteristics or concentrations were far from real conditions (Ma et al., 2019a, Ma et al., 2019b; Skaf et al., 2020), and thus, the results are difficult to compare. In general, the current knowledge of MP removal at DWTPs operating under ordinary conditions is scarce, and more research is required in this field, which is also mentioned in the report on MPs published by the World Health Organization (World Health Organisation, 2019). Moreover, despite the significant removal of MPs at DWTPs, possible MP enrichment during treatment due to the utilization of plastic materials has also been proposed (Pivokonsky et al., 2018; Wang et al., 2020) and deserves further investigation (Novotna et al., 2019; World Health Organisation, 2019).

MPs are emerging water pollutants and have attracted considerable attention, and research addressing the fate of MPs at DWTPs is very limited so far. This study provides unique insight into the occurrence of MPs at two different DWTPs that both lie on the same river, separated by a distance of approximately 90 km by water. Quantification and characterization of MPs was performed not only in raw and treated water but also after each technological treatment step. MPs as small as 1 μm were analysed, and their categorization according to size, shape, and material was performed.