Comparative Long-Term Monitoring of Microplastics in the Effluent of Three Different Wastewater Treatment Plants with Two, Three, and Four Treatment Stages

Academic Editors: Chengtao Li and Xueqiang Lu Received: 6 February 2025 Revised: 24 February 2025 Accepted: 26 February 2025 Published: 28 February 2025 Citation: Sturm, M.T.; Argyropoulou, D.; Myers, E.; Korzin, A.; Ronsse, P.; Zernikel, O.; Schober, D.; Schuhen, K. Comparative LongTerm Monitoring of Microplastics in the Effluent of Three Different Wastewater Treatment Plants with Two, Three, and Four Treatment Stages. Water 2025,17, 711. https://doi.org/ 10.3390/w17050711 Copyright: © 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/ licenses/by/4.0/). Article Comparative Long-Term Monitoring of Microplastics in the Effluent of Three Different Wastewater Treatment Plants with Two, Three, and Four Treatment Stages Michael Toni Sturm 1, Daphne Argyropoulou 2, Erika Myers 1, Anika Korzin 1, Pieter Ronsse 1, Oleg Zernikel 1, Dennis Schober 1and Katrin Schuhen 1,* 1Wasser 3.0 gGmbH, Neufeldstr. 17a-19a, 76187 Karlsruhe, Germany 2 Sanitary Engineering Laboratory, Department of Water Resources and Environmental Engineering, School of Civil Engineering, National Technical University of Athens, 15780 Athens, Greece *Correspondence: [email protected]; Tel.: +49-721-15-65-95-93 Abstract: Wastewater treatment plants (WWTPs) are important point sources for microplastics (MPs) in the environment. For effective mitigation measures and regulations, it is important to monitor their release into the environment and understand the level of MPs in the WWTP effluents based on different treatment technologies. In this study, we compare the MP levels in the effluents of three different municipal WWTPs which each use a different treatment concept: a conventional three-stage WWTP, one with an additional fourth cleaning stage using powdered activated carbon, and a two-stage WWTP utilizing a membrane bioreactor (MBR). Long-term monitoring was performed on the WWTP effluents using the same standardized methods for sample collection, preparation, and detection, based on fluorescent staining. Despite the various advanced treatment processes, there are no significant differences in the resulting MP contamination in the investigation of WWTP effluents. The average MP concentrations in the effluents were 21.8 MPs/L for the conventional three-stage WWTP, 15.1 MPs/L for the four-stage WWTP, and 15.1 MPs/L for the MBR. Further, the MP contamination in all effluents shows a strong fluctuation over time. These findings highlight the need for standard MP monitoring at WWTPs, to gain a better understanding of the MP emission in different treatment processes. Further, it highlights the need for a fourth treatment stage that specifically targets MP removal to effectively prevent the MP release from WWTPs into the environment. Keywords: microplastics; wastewater treatment; WWTP; microplastic detection; microplastic analysis; microplastic analytics; fluorescent dyes; long-term monitoring 1. Introduction Microplastics (MPs), defined as plastic particles less than 5 mm in diameter, have emerged as a significant environmental pollutant, posing risks to aquatic and terrestrial ecosystems [ 1 ]. Wastewater treatment plants (WWTPs) are critical nodes in the pathway of MPs from urban environments to natural water bodies [ 2 , 3 ]. Despite the high MP removal efficiency in WWTPs, due to the high volumes of treated wastewater, substantial quantities still escape into the environment, raising concerns about their ecological and health impacts (Figure 1) [1,4]. MP can be ingested by a wide range of organisms, leading to physical harm, disturbance of cell functions and potential release or bioaccumulation of toxic substances [ 1 ]. For humans, MP have been found in various tissues and are linked to Water 2025,17, 711 https://doi.org/10.3390/w17050711 Water 2025,17, 711 2 of 14 potential health risks, including immunotoxicity and endocrine disruption, although more research is needed to fully understand their long-term impacts [5]. Water 2025, 17, x FOR PEER REVIEW 2 of 14 quantities still escape into the environment, raising concerns about their ecological and health impacts (Figure 1) [1,4]. MP can be ingested by a wide range of organisms, leading to physical harm, disturbance of cell functions and potential release or bioaccumulation of toxic substances [1]. For humans, MP have been found in various tissues and are linked to potential health risks, including immunotoxicity and endocrine disruption, although more research is needed to fully understand their long-term impacts [5]. Figure 1. Overview of the pathway of MP exposure from WWTPs to human, animal, and environment and subsequent negative impacts [1,5]. Recent studies have highlighted the prevalence of MPs in wastewater effluents and sludge. For instance, a meta-analysis by Azizi et al. (2022) revealed removal efficiencies of conventional three-stage treatment plants ranging from 64% to over 99%, depending on the study [6]. A 2024 review by Miino et al. discusses the MP removal between the first, second, and third cleaning stages, as well as for a polishing treatment (e.g., filtrate membrane disc filter, disinfection by UV, ozone, or chlorination), with the majority of MPs being removed during the primary treatment [7]. However, long-term studies providing absolute numbers obtained from comparable sampling methods are not available. The remaining MPs, particularly those smaller than 150 µm, are often discharged into aquatic systems or accumulate in sewage sludge, which is frequently applied to agricultural lands [8]. The fate and transport of MPs in WWTPs are influenced by various factors, including the physical and chemical properties of the plastics, the design and operational parameters of the treatment processes, and the environmental conditions. Jose et al. (2024) reviewed the transformation and fate of MPs in WWTPs, emphasizing the need for standardized methodologies to accurately assess their removal and environmental release [9]. The recent revision of the EU Urban Wastewater Directive (UWWTD) introduces several key measures to address MPs [10]. The updated directive is a first step towards the systematic monitoring of MPs in wastewater treatment plants, although the current proposal is not sufficient for obtaining representative data on the actual MP concentrations [11]. Additionally, the directive aligns with the EU’s zero-pollution action plan, aiming to prevent harmful substances, including MPs, from being released into the environment. This revision also emphasizes the “polluter pays” principle, requiring industries responsible for MP pollution to bear the costs of advanced treatment processes. These changes reflect a significant step towards mitigating the environmental impact of MPs and enhancing the overall effectiveness of wastewater treatment in Europe. Figure 1. Overview of the pathway of MP exposure from WWTPs to human, animal, and environment and subsequent negative impacts [1,5]. Recent studies have highlighted the prevalence of MPs in wastewater effluents and sludge. For instance, a meta-analysis by Azizi et al. (2022) revealed removal efficiencies of conventional three-stage treatment plants ranging from 64% to over 99%, depending on the study [ 6 ]. A 2024 review by Miino et al. discusses the MP removal between the first, second, and third cleaning stages, as well as for a polishing treatment (e.g., filtrate membrane disc filter, disinfection by UV, ozone, or chlorination), with the majority of MPs being removed during the primary treatment [ 7 ]. However, long-term studies providing absolute numbers obtained from comparable sampling methods are not available. The remaining MPs, particularly those smaller than 150 µ m, are often discharged into aquatic systems or accumulate in sewage sludge, which is frequently applied to agricultural lands [8]. The fate and transport of MPs in WWTPs are influenced by various factors, including the physical and chemical properties of the plastics, the design and operational parameters of the treatment processes, and the environmental conditions. Jose et al. (2024) reviewed the transformation and fate of MPs in WWTPs, emphasizing the need for standardized methodologies to accurately assess their removal and environmental release [9]. The recent revision of the EU Urban Wastewater Directive (UWWTD) introduces several key measures to address MPs [ 10 ]. The updated directive is a first step towards the systematic monitoring of MPs in wastewater treatment plants, although the current proposal is not sufficient for obtaining representative data on the actual MP concentrations [ 11 ]. Additionally, the directive aligns with the EU’s zero-pollution action plan, aiming to prevent harmful substances, including MPs, from being released into the environment. This revision also emphasizes the “polluter pays” principle, requiring industries responsible for MP pollution to bear the costs of advanced treatment processes. These changes reflect a significant step towards mitigating the environmental impact of MPs and enhancing the overall effectiveness of wastewater treatment in Europe. In the UWWTD, it is stated that WWTPs serving populations greater than 150,000 PE (population equivalent) must conduct MP sampling at least twice annually, ensuring no more than six months between samples [ 10 ]. Additionally, WWTPs with capacities exceeding 150,000 PE must monitor MPs in sewage sludge. In contrast, plants serving over Water 2025,17, 711 3 of 14 10,000 PE are required to sample only once every two years and are not required to monitor the sludge. A standardized methodology for monitoring and detection must be established within 30 months of the directive’s enactment. Once approved by the EU Parliament, this methodology must be incorporated into national legislation. Monitoring MPs involves three main steps: sampling, sample preparation, and detection [ 12 ]. Currently, there are no standardized methods for these steps, leading to variations in methodologies across different studies, which complicates result comparisons [ 13 ]. Additionally, detecting MPs is often labor-intensive and expensive due to the intricate sampling and preparation processes, as well as the costly detection methods that require long processing times, resulting in low sample throughput [14]. The fluorescence staining method used in this study, in combination with the standardized sampling and sample preparation protocols, offers a quick, comparable, and user-friendly alternative [15,16]. The focus of this study is the comparison of three municipal wastewater with two, three, and four stages regarding their MP emissions, applying the same sampling, sample preparation, and detection method. Compared are a conventional three-cleaning-stage WWTP in Germany, a four-stage WWTP in Germany that uses powdered activated carbon (PAC) and cloth filtration, and a two-stage WWTP in Greece, using a membrane bioreactor (MBR). MBRs and cloth filtration are often discussed in the scientific literature as efficient methods for MP removal in WWTP [ 17 ]. This comparison should clarify if the WWTPs show significant differences in their MP emissions or if additional treatment steps are necessary. 2. Materials and Methods 2.1. Demo Site Descriptions The following three WWTPs were selected because they use different treatment technologies, which are compared in terms of their MP release. WWTP-A is a conventional three-stage municipal WWTP, while WWTP-B is a conventional WWTP with an additional fourth treatment stage using PAC and cloth filtration. WWTP-C has 2 cleaning stages, including a modern MBR. MBRs and cloth filtration are often discussed in science as solutions for the MP emission from WWTPs into the environment, which is why these WWTPs are particularly interesting for this investigation [17]. 2.1.1. WWTP-A: Conventional 3-Stage WWTP WWTP-A is a municipal WWTP with a capacity of 80,000 population equivalents (PE). The WWTP receives wastewater from approximately 51,000 people, along with wastewater from industrial and commercial enterprises, two hospitals, and agriculture—primary viticulture. An overview of the WWTP can be seen in Figure 2. The WWTP runs three treatment stages with a cleaning performance of 95–97% for COD: 1. A primary treatment using rakes, a sand trap, and a fat separator; 2. Secondary biological treatment (nitrification and denitrification); 3. Chemical treatment: tertiary phosphate elimination. The flow rate ranges from 9000 to 13,000 m 3 /d during dry weather periods and up to 40,000 m 3 /d during rain events. The average hydraulic retention time (HRT) is 24 h, and the average sludge retention time is 12–14 d. The treated wastewater in the effluent has an average chemical oxygen demand (COD) of 20 mg/L, 6.5 mg/L nitrate, <1 mg/L ammonia, and a total phosphorus concentration of 0.3 mg/L Water 2025,17, 711 4 of 14 Water 2025, 17, x FOR PEER REVIEW 4 of 14 an average chemical oxygen demand (COD) of 20 mg/L, 6.5 mg/L nitrate, <1 mg/L ammonia, and a total phosphorus concentration of 0.3 mg/L Figure 2. Image and flow diagram of WWTP-A, with conventional 3-stage treatment. 2.1.2. WWTP-B: Conventional WWTP with Additional 4th Cleaning Stage The municipal WWTP-B is a 4-stage WWTP equipped with primary treatment using rakes, a sand trap, and a fat separator, a secondary biological treatment (nitrification and denitrification), and a tertiary phosphate elimination. For the fourth treatment stage, a powdered activated carbon (PAC) treatment is used, followed by a cloth filter for separation of the PAC and TSS reduction (Figure 3). When WW volumes are too high, e.g., during rain events, the capacity of the 4th cleaning stage is exceeded, and a bypass discharges the treated water after the 3rd cleaning stage directly into the receiving water. The catchment area consists of residential and industrial areas and accounts for 250,000 PE. The dry weather inflow ranges from 30,000 m3/d to 65,000 m3/d; the max. rainwater inflow is 135,000 m3/d. The average sludge retention time is 9 d. The treated effluent has an average CSB of 12 mg/L and the following concentrations: nitrate 11 mg/L, ammonia < 0.1 mg/L, nitrite < 0.1 mg/L, and a total phosphorus concentration of 0.2 mg/L. Figure 3. Image and flow diagram of WWTP-B, with 4 treatment stages. Figure 2. Image and flow diagram of WWTP-A, with conventional 3-stage treatment. 2.1.2. WWTP-B: Conventional WWTP with Additional 4th Cleaning Stage The municipal WWTP-B is a 4-stage WWTP equipped with primary treatment using rakes, a sand trap, and a fat separator, a secondary biological treatment (nitrification and denitrification), and a tertiary phosphate elimination. For the fourth treatment stage, a powdered activated carbon (PAC) treatment is used, followed by a cloth filter for separation of the PAC and TSS reduction (Figure 3). Water 2025, 17, x FOR PEER REVIEW 4 of 14 an average chemical oxygen demand (COD) of 20 mg/L, 6.5 mg/L nitrate, <1 mg/L ammonia, and a total phosphorus concentration of 0.3 mg/L Figure 2. Image and flow diagram of WWTP-A, with conventional 3-stage treatment. 2.1.2. WWTP-B: Conventional WWTP with Additional 4th Cleaning Stage The municipal WWTP-B is a 4-stage WWTP equipped with primary treatment using rakes, a sand trap, and a fat separator, a secondary biological treatment (nitrification and denitrification), and a tertiary phosphate elimination. For the fourth treatment stage, a powdered activated carbon (PAC) treatment is used, followed by a cloth filter for separation of the PAC and TSS reduction (Figure 3). When WW volumes are too high, e.g., during rain events, the capacity of the 4th cleaning stage is exceeded, and a bypass discharges the treated water after the 3rd cleaning stage directly into the receiving water. The catchment area consists of residential and industrial areas and accounts for 250,000 PE. The dry weather inflow ranges from 30,000 m3/d to 65,000 m3/d; the max. rainwater inflow is 135,000 m3/d. The average sludge retention time is 9 d. The treated effluent has an average CSB of 12 mg/L and the following concentrations: nitrate 11 mg/L, ammonia < 0.1 mg/L, nitrite < 0.1 mg/L, and a total phosphorus concentration of 0.2 mg/L. Figure 3. Image and flow diagram of WWTP-B, with 4 treatment stages. Figure 3. Image and flow diagram of WWTP-B, with 4 treatment stages. When WW volumes are too high, e.g., during rain events, the capacity of the 4th cleaning stage is exceeded, and a bypass discharges the treated water after the 3rd cleaning stage directly into the receiving water. The catchment area consists of residential and industrial areas and accounts for 250,000 PE. The dry weather inflow ranges from 30,000 m 3 /d to 65,000 m 3 /d; the max. rainwater inflow is 135,000 m 3 /d. The average sludge retention time is 9 d. The treated effluent has an average CSB of 12 mg/L and the following concentrations: nitrate Water 2025,17, 711 5 of 14 11 mg/L , ammonia < 0.1 mg/L, nitrite < 0.1 mg/L, and a total phosphorus concentration of 0.2 mg/L. 2.1.3. WWTP-C: 2 Treatment Stages, with a Membrane Bioreactor WWTP-C (Figure 4) has been constructed on three distinct elevation levels. The administrative buildings and the pretreatment units are located at the highest level, while the biological treatment units are located on the second and third levels. The chlorination unit is downstream at the end of the field and receives the treated wastewater by gravity. The plant consists of two main lines that receive the same influent and operate in parallel: 1. A conventional activated sludge system with a capacity of 8400 m3/d; 2. A membrane bioreactor with a capacity of 10,800 m3/d. Water 2025, 17, x FOR PEER REVIEW 5 of 14 2.1.3. WWTP-C: 2 Treatment Stages, with a Membrane Bioreactor WWTP-C (Figure 4) has been constructed on three distinct elevation levels. The administrative buildings and the pretreatment units are located at the highest level, while the biological treatment units are located on the second and third levels. The chlorination unit is downstream at the end of the field and receives the treated wastewater by gravity. The plant consists of two main lines that receive the same influent and operate in parallel: 1. A conventional activated sludge system with a capacity of 8400 m3/d; 2. A membrane bioreactor with a capacity of 10,800 m3/d. In total, the WWTP has been designed to facilitate the needs of 48,000 PE. The influent wastewater in this system originates from two main sources: municipal wastewater collected through the sewer network of Mykonos and wastewater from septic tanks. Seasonal tourism has a strong influence on wastewater volumes, as in the off-season, the population can decline to under 10,000 PE. To ensure effective treatment, both streams initially pass through bar screening and grit removal to eliminate large debris and sand. After preliminary treatment, the combined wastewater stream is directed into two parallel treatment lines for further processes, i.e., biological treatment and disinfection (an overview of the WWTP processes is shown in Figure 4). It is noted that at the time of the samplings, the final filtration step (before chlorination) was not used by WWTP due to maintenance. The treated wastewater shows an average BOD5 of 15 mg/L, COD of 3, nitrate concentration of 10 mg/L, and total phosphorus of 5 mg/L. Figure 4. Image and flow diagram of WWTP-C, with two parallel treatment lines: a two-stage conventional treatment and a two-stage treatment with an MBR. 2.2. Microplastic Monitoring The MP monitoring follows a standard protocol using standardized sample collection by filtration, sample processing by hydrogen peroxide digestion, and MP detection by fluorescent staining and imaging [15,18,19]. The effectiveness of the method was evaluated using MPs and natural particles to determine recovery rates and risks for Figure 4. Image and flow diagram of WWTP-C, with two parallel treatment lines: a two-stage conventional treatment and a two-stage treatment with an MBR. In total, the WWTP has been designed to facilitate the needs of 48,000 PE. The influent wastewater in this system originates from two main sources: municipal wastewater collected through the sewer network of Mykonos and wastewater from septic tanks. Seasonal tourism has a strong influence on wastewater volumes, as in the off-season, the population can decline to under 10,000 PE. To ensure effective treatment, both streams initially pass through bar screening and grit removal to eliminate large debris and sand. After preliminary treatment, the combined wastewater stream is directed into two parallel treatment lines for further processes, i.e., biological treatment and disinfection (an overview of the WWTP processes is shown in Figure 4). It is noted that at the time of the samplings, the final filtration step (before chlorination) was not used by WWTP due to maintenance. The treated wastewater shows an average BOD5 of 15 mg/L, COD of 3, nitrate concentration of 10 mg/L, and total phosphorus of 5 mg/L. Water 2025,17, 711 6 of 14 2.2. Microplastic Monitoring The MP monitoring follows a standard protocol using standardized sample collection by filtration, sample processing by hydrogen peroxide digestion, and MP detection by fluorescent staining and imaging [ 15 , 18 , 19 ]. The effectiveness of the method was evaluated using MPs and natural particles to determine recovery rates and risks for false positives [ 15 , 18 ]. The advantage of this method is the fast and cost-effective application, which makes it efficient to process high numbers of samples [ 20 ]. The disadvantages are missing chemical characterization of polymer types and the risks of false positives with natural particles [20]. 2.2.1. Sample Collection Sampling in the WWTP effluents followed Sturm et al. (2023) [ 15 ]. Samples were taken in duplicates using the Wasser 3.0 Particle Sampling Unit (PSU, Wasser 3.0 gGmbH, Karlsruhe, Germany) [11]. The PSU combined a 10 µ m filter cartridge with a pump, tubing, valves, and a water meter for direct and easy on-site application [ 15 ]. The water was pumped from the inlet through the filter cartridge with a pressure of max. 4 bar. The filter retained all particles larger than 10 µm, including MPs. The materials used are listed in Table 1. Table 1. Materials used for the construction of the Wasser 3.0 PSU. Part Model No. Supplier Rotary Pump MG80B C-B-CMS1B Grundfos, Erkrath, Germany Filter cartridge (10 µ m) 01WTGD Wolftechnik Filtersysteme GmbH & Co., KG, Weil, Germany Water meter Zenner ETKD ZENNER International GmbH & Co. KG, Saarbrücken, Germany Tubing TUFLON-PTFE/NW25 Industriebedarf Castan GmbH, Freiberg, Germany The inlet and outlet tubes were connected to the PSU. The inlet tube needed to be placed before the outlet tube in the flow direction to prevent the inflow of prefiltered water. The inlet tube was made from black PVC, which was not detected by the fluorescent staining method and avoided contamination of the samples. Additionally, a 2 mm suction basket was applied to avoid clogging or damaging the pump. For each sample, 100 L of treated wastewater was run through the filter. Subsequently, the filter cartridge was removed, and the filtered solids were flushed into a 2.5 L glass bottle using a spray bottle filled with tap water. The 2.5 L glass bottle was used for sample storage and transport. 2.2.2. Sample Processing Sample processing was conducted using a hydrogen peroxide treatment followed by fluorescence staining [ 18 ]. Hydrogen peroxide treatment reduces the number of natural particles by chemical decomposition while MPs are not degraded and remain in the samples [ 13 , 21 ]. For samples from WWTP-C, the sampling preparation protocol was successfully transferred and implemented, with pretreatment being conducted in Greece and the remaining treatment and analysis conducted in Germany. First, 500 mL subsamples were filtered over a 10 µ m stainless steel filter disc (Wolftechnik Filtersysteme GmbH & Co., KG, Weil, Germany) using vacuum filtration (DURAN® Filtering Apparatus, Cat. No. 257106304, DWK Life Sciences GmbH, Mainz, Germany). The filter disc and retained solids were placed into a beaker, which was filled with 20 mL hydrogen peroxide (35%, AB171423, abcr GmbH, Karlsruhe, Germany) and 3–5 grains of iron(II)sulfate (AB203817, abcr GmbH, Karlsruhe, Germany). Subsequently, the beaker was placed on the heating plate and kept at 100 ◦ C for 1 h. After cooling down for 10 min, the Water 2025,17, 711 7 of 14 filter disc was removed from the beaker and rinsed with water into the same beaker. The filter disc was inserted again in the vacuum filtration, and the whole sample was filtered. The filter disc was removed, and all solids were flushed from the filter disc into the empty beaker. In the next step, the fluorescence dye abcr eco Wasser 3.0 detect mix MP-1 (AB930015, abcr GmbH, Karlsruhe, Germany) was added to the beaker. The fluorescent dye was developed to selectively stain MP, but not natural particles and residues [ 15 , 18 , 22 ]. In total, 25 µ L of the fluorescent dye was added to the sample (c = 0.25 mg/L) and kept for 1 h at 80 ◦ C. After staining, the sample was filtered through a black filter membrane (Metricel ® Black PES Membranfilter, 0.80 µm, Pall, Dreieich, Germany) and stored in a Petri dish. 2.2.3. Detection Fluorescent imaging was performed using a Zeiss Axiozoom.V16 (Carl Zeiss Microscopy Deutschland GmbH, Oberkochen, Germany). The microscope was equipped with a PlanNeoFluar Z 1.0 × Objective and an Axiocam 712 mono. For green fluorescence, a custom-made filter set from AHF Analystechnik AG (Tübingen, Germany) was used. The images were taken with the following parameters (Table 2). Table 2. Parameters for fluorescence imaging of MP samples. Parameter Setting Exposure 20 ms Aperture 100% Binning 3.3 Bit depth 14 Bit Zoom 10 Total magnification (ocular) 100 Scaling (per Pixel) 1.035 ×1.035 µm Depth of field 8.9 µm Excitation 430–480 nm Emission 500–570 nm Beam splitter 495 nm Image processing Stitching For analysis, five 3.9 × 3.8 mm squares of the 47 mm round filter were photographed and analyzed. For image analysis, the particle counting module for the software ZEN 3.8 (Carl Zeiss Microscopy Deutschland GmbH, Oberkochen, Germany) was used. The particle counting module (ZEN Toolkit 2D, Carl Zeiss Microscopy Deutschland GmbH, Oberkochen, Germany) can automatically detect MPs based on brightness threshold values. 2.2.4. Contamination Control For contamination control, all samples were handled in a separate laboratory that was used only for MP analytics. The laboratory was cleaned before each use with a lint-free cleaning rag, and lint-free clothing was worn (4510 M, 3 M Deutschland GmbH, Ness, Germany). Before entering the laboratory, the lint-free suit was rinsed using a lint roller with adhesive paper. Additionally, an air filter was operated, glass laboratory equipment was used if possible, and the samples were always covered with aluminum foil. Further, regular blanks were measured and subtracted from the detected MP concentrations. 3. Results and Discussion 3.1. Comparison of the WWTPs The MP contamination of the three WWTPs is compared in Figure 5and Table 3. In total, 84 samples were taken at WWTP-A, 13 at WWTP-B, and 41 at WWTP-C. Water 2025,17, 711 8 of 14 Water 2025, 17, x FOR PEER REVIEW 8 of 14 = −0.01, p = 0.99). It is notable that the relatively small sample size at WWTP-B affects the robustness of the t-test, which means that smaller differences between the MP contamination levels might not be recognized [23]. Figure 5. Boxplot of the MP contamination of the effluents of the three different WWTPs. Note: For each day, the mean of the duplicate-sample takes was calculated and counted as one sample. For WWTP-A, two outliers (432 MPs/L and 391.8 MPs/L) are not displayed. At all WWTPs, fragments were almost exclusively detected (compare Figure 6). In comparable studies at WWTPs, fibers released from synthetic textiles and fragments are typically the dominant morphologies of the MPs found [24]. The reasons for this could be the absence of textile fibers in the investigated WW or the relatively aggressive chemical hydrogen peroxide digestion which may decompose the fibers [25]. However, this pretreatment is necessary, as fluorescence staining as a detection method is more prone to false positives by natural particles [19]. Therefore, the number of MPs in the form of fibers might be underestimated. The effluent of the third CS at WWTP-B shows an elevated MP contamination with a mean of 36.2 MPs/L, compared to 15.1 MPs/L in the fourth CS effluent of WWTP-B. At WWTP-C, the MBR effluent had a mean of 15.1 MPs/L. These differences are statically significant (Welch’s t-test, two-sided, unpaired, WWTP-B third CS–WWTP-B fourth CS: t(12) = 3.55, p = 0.04, WWTP-B third CS–WWTP-C MBR effluent: t(40) = 8.39; p = 0.03 ). The difference between the third CS at WWTP-B and the third CS at WWTP-A is also statistically significant (Welch’s t-test, two-sided, t(83) = 3.18, p = 0.002). At WWTP-B, there was an average reduction of 36.9 ± 56.7 % from the effluent of the third CS to the effluent of the fourth CS. Despite the application of a fourth CS (PAC) and cloth filtration, the WWTP-B MP release is similar to that of WWTP-A with three conventional CSs. An investigation of MP removal by applying cloth filtration at the WWTP in Oldenburg in Germany by Mintenig et al. (2017) showed an MP reduction of 97% [26]. A laboratory study by Sembring et al. (2021) showed an MP removal efficiency of 34–85%, depending on the MP size and the pore size of the used cloth filter [27]. The cloth filter applied at WWTP-B has a pore size of 5 µm. Whether MPs larger than 5 µm can still pass through is unclear and requires further investigation. One possibility could be operational issues. The MBR at WWTP-C does not show a lower MP contamination compared to WWTP-A or WWTP-B. Bayo et al. (2022) investigated the effectiveness of MP removal of Figure 5. Boxplot of the MP contamination of the effluents of the three different WWTPs. Note: For each day, the mean of the duplicate-sample takes was calculated and counted as one sample. For WWTP-A, two outliers (432 MPs/L and 391.8 MPs/L) are not displayed. Table 3. Summarized key indicators of the MP monitoring in the three different WWTPs. Note: For each day, the mean of the duplicate-sample takes was calculated and counted as one sample. WWTP-A 3rd CS WWTP-B 3rd CS WWTP-B 4th CS WWTP-C MBR No. of samples 84 13 13 41 Mean [MPs/L] 21.8 36.2 15.1 15.1 S.D. [MPs/L] 62.6 30.1 14.7 11.4 Median [MPs/L] 7.9 25.3 13.0 13.6 Min [MPs/L] 0.0 7.7 0.8 1.0 Max [MPs/L] 432.4 105.1 58.7 61.2 The data show the MP contamination at all WWTPs, with the final effluents showing a similar MP concentration range and fluctuations. No statistically significant difference was detected in the final effluents (Welch’s t-test, two-sided, unpaired, WWTP-A–WWTPB: t(79) = 0.83 ,p= 0.41; WWT-A–WWTP-C: t(94) = 0.95, p= 0.34; WWTP-B–WWTP-C: t(17) = − 0.01, p= 0.99). It is notable that the relatively small sample size at WWTP-B affects the robustness of the t-test, which means that smaller differences between the MP contamination levels might not be recognized [23]. At all WWTPs, fragments were almost exclusively detected (compare Figure 6). In comparable studies at WWTPs, fibers released from synthetic textiles and fragments are typically the dominant morphologies of the MPs found [ 24 ]. The reasons for this could be the absence of textile fibers in the investigated WW or the relatively aggressive chemical hydrogen peroxide digestion which may decompose the fibers [ 25 ]. However, this pretreatment is necessary, as fluorescence staining as a detection method is more prone to false positives by natural particles [ 19 ]. Therefore, the number of MPs in the form of fibers might be underestimated. Water 2025,17, 711 9 of 14 Water 2025, 17, x FOR PEER REVIEW 9 of 14 an MBR in a Spanish WWTP (EDAR Águilas) and found a removal efficiency of 79% [28]. In contrast, a study by Egea-Corbacho et al. (2023) at the Søholt WWTP in Denmark found the MBR to effectively remove 99.69% of MPs [29]. As the MP influent concentration of the MBR was not measured, the removal efficiency cannot be specified. It is unclear if the MP emission is caused by high influent concentrations or unintentional release during MBR operation. To clarify the reason for MPs passing the MBR, further investigations would be needed. Figure 6. Example images of processed MP samples from the three WWTPs. Shown in the photos are green fluorescence and marked in red are particles detected as MPs from the detection software. Figure 6. Example images of processed MP samples from the three WWTPs. Shown in the photos are green fluorescence and marked in red are particles detected as MPs from the detection software. Note: The sample from the WWTP-B 4th CS contains residues from the PAC, which are visible as black particles in the color photo. The effluent of the third CS at WWTP-B shows an elevated MP contamination with a mean of 36.2 MPs/L, compared to 15.1 MPs/L in the fourth CS effluent of WWTP-B. At WWTP-C, the MBR effluent had a mean of 15.1 MPs/L. These differences are statically significant (Welch’s t-test, two-sided, unpaired, WWTP-B third CS–WWTP-B fourth CS: