Removal of contaminants of emerging concern by membranes in water and wastewater: An updated review

Siye Kim; Byung-Moon Jun; Bongyeon Jung; Chang Min Park; Min Jang; Seong-Nam Nam; Yeomin Yoon

doi:10.4491/eer.2024.103

Environ Eng Res > Volume 30(1); 2025 > Article

Kim, Jun, Jung, Park, Jang, Nam, and Yoon: Removal of contaminants of emerging concern by membranes in water and wastewater: An updated review

Review

Environmental Engineering Research 2025; 30(1): 240103.

Published online: May 9, 2024

DOI: https://doi.org/10.4491/eer.2024.103

Removal of contaminants of emerging concern by membranes in water and wastewater: An updated review

Siye Kim¹, Byung-Moon Jun², Bongyeon Jung¹, Chang Min Park³, Min Jang⁴, Seong-Nam Nam^5,^†

, Yeomin Yoon^1,^†

¹Department of Environmental Science and Engineering, Ewha Womans University, 52 Ewhayeodae-gil, Seodaemun-gu, Seoul 03760, Republic of Korea

²Radwaste Management Center, Korea Atomic Energy Research Institute (KAERI), 111 Daedeok-Daero 989beon-gil, Yuseong-Gu, Daejeon 34057, Republic of Korea

³Department of Environmental Engineering, Kyungpook National University, 80 Daehak-ro, Buk-gu, Daegu 41566, Republic of Korea

⁴Department of Environmental Engineering, Kwangwoon University, 447-1 Wolgye-dong Nowon-gu, Seoul, Republic of Korea

⁵Military Environmental Research Center, Korea Army Academy at Yeongcheon, 495 Hoguk-ro, Gogyeong-myeon, Yeongcheon-si, Gyeongsangbuk-do, 38900, Republic of Korea

^†Corresponding author: E-mail: namsn76@gmail.com (S.-N.N.), yoony@ewha.ac.kr (Y.Y.), Tel: +82-54-330-4838 (S.-N.N.), +82-2-3277-2369 (Y.Y.), Fax: +82-54-330-4839 (S.-N.N.), +82-2-3277-2369 (Y.Y.), ORCID: 0000-0002-8664-5912 (S.-N.N.), 0000-0001-9893-0924 (Y.Y.)

Received February 20, 2024 Revised May 1, 2024 Accepted May 8, 2024

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

The current review covers removal of contaminants of emerging concern by membranes in water and wastewater since our much cited review in 2018. This review offers a wide-ranging examination of membrane technologies—in particular, forward osmosis, reverse osmosis, nanofiltration, ultrafiltration, and microfiltration—as potent solutions that can be used to target contaminants of emerging concern (CECs) in water and wastewater treatment. Emphasizing the urgency of preserving water quality amid increasing demand and CEC-related concerns, the current paper underscores the critical need to obtain a more holistic understanding of the impacts of CEC, effective strategies for their removal, and essential regulatory measures. The interplay between membrane properties, operating conditions, and contaminants underscores the importance of tailored membrane designs and optimization in achieving efficient CEC removal. The main purpose of this review was to synthesize the existing knowledge on membrane treatment of CECs and highlight future research directions. This review not only synthesizes the recent advancements that have been achieved but also highlights critical research avenues, including advanced surface modifications, novel materials, optimized operational parameters, and sustainability considerations. Achieving future strides in these areas would likely enhance the efficacy and sustainability of membrane technologies in combatting CECs in water systems.

Keywords: Contaminants of emerging concern, Forward osmosis, Nanofiltration, Reverse osmosis, Ultrafiltration, Water and wastewater treatment

Graphical Abstract

Keywords: Contaminants of emerging concern, Forward osmosis, Nanofiltration, Reverse osmosis, Ultrafiltration, Water and wastewater treatment

1. Introduction

The relentless growth of the global population, rapid industrial development, and increased human activity has led to an exponential surge in water demand [1]. This escalating demand has led to significant concerns about water quality, with a specific focus on the presence of contaminants of emerging concerns (CECs) [2]. Of these CECs, the ones that have drawn particular attention are endocrine-disrupting compounds (EDCs) and pharmaceuticals/personal care products (PPCPs) [3]. EDCs have been found to disrupt the normal synthesis and metabolism of hormones, thereby posing possible risks to both human health and the environment [4, 5]. PPCPs, due to their widespread usage, have also become a subject of growing interest, as conventional water treatment methods struggle to effectively remove them [1, 6]. Meanwhile, PPCPs, which include compounds such as ibuprofen and N,N-diethyl-meta-toluamide, are recognized to influence hormone function and the endocrine system, thereby posing further challenges to water quality and safety [7, 8]. These findings emphasize the urgent need for comprehensive and effective solutions to tackle the complex issue of water contamination to ensure the well-being of ecosystems and human communities.

The effect of CECs on human health and the environment is still partially understood, as these chemicals resist easy metabolization and treatment in water environments [9, 10]. This lack of knowledge poses a severe threat to aquatic systems that could potentially have long-term consequences. Although the effects of CECs on the aquatic ecosystem and human health are yet to be fully understood, it is clear that they can cause long-term damage to both. To address this concern, the United States Environmental Protection Agency initiated the Endocrine Disruptor Screening Program in 1998, which aimed to monitor the concentration and activities of EDCs in the environment and biological media [11]. Despite such efforts, many relevant chemicals remain unregulated. Certain substances, including disinfection byproducts, erythromycin, polyfluoroalkyl substances, estrone, 17b-estradiol, and estriol, have been listed as candidates for contamination that are known to occur in public water systems. It is essential to develop a comprehensive understanding of the sources and pathways of these contaminants. It is also important to take measures to minimize the discharge of CECs into the environment through improved wastewater treatment practices. A crucial element of such efforts is the development of cost-effective and reliable monitoring tools that can accurately detect and quantify the presence of CECs. Given the potential risks posed by CECs, it is of utmost importance that regulatory agencies take proactive measures in establishing effective regulations and guidelines for the management of these contaminants. Further research is imperative to completely comprehend the effect of CECs on the environment and human health, as it would allow for the development of targeted and efficient strategies to mitigate their negative effects. Continuous exploration of the behavior and fate of CECs in various water treatment processes is essential to ensure the efficacy of water treatment methods in removing these emerging contaminants.

Among the various technologies that are currently used in wastewater treatment, membrane processes play a pivotal role in addressing the challenges posed by CECs. The highly regarded membrane technologies that have shown the most promise in effectively removing CECs from water sources include forward osmosis (FO), reverse osmosis (RO), nanofiltration (NF), ultrafiltration (UF), and microfiltration (MF) [12]. Of these processes, FO stands out as a fascinating membrane separation method with diverse applications. In FO, water transport occurs osmotically, driven by a pressure difference across the FO membrane, which demands low hydraulic pressure, thus resulting in lower energy consumption and an often reversible outcome [13, 14]. On the other hand, RO has gained widespread usage in wastewater treatment due to its simplicity and high energy efficiency [15]. The high selectivity and effectiveness of RO in removing micropollutants also make it an effective solution for eliminating CECs from water sources [16]. Meanwhile, NF operates based on the principle of pressure-driven filtration with pores smaller than 0.001 μm, so multivalent ions and other contaminants are effectively rejected [17]. UF, which has historically been used for size separation of mixtures, has shown promising results in achieving higher selectivity and efficiently separating aquatic substances by employing low-pressure operations [18]. Lastly, MF is another low-pressure membrane filtration process that exhibits potential as an efficient technology for removing CECs from wastewater [19].

In 2018, we conducted an extensive review on membrane technologies that could be used to remove CECs. That review aimed to explore and update the existing knowledge on the removal of CECs using membranes in water and wastewater treatment [20]. By delving into the physicochemical characteristics, removal efficiency, and influencing factors of FO, RO, NF, UF, and MF membranes, we sought to provide practical recommendations for researchers and facilitate the design of more effective experiments. Through this endeavor, we aimed to contribute to ongoing global efforts to safeguard water quality and environmental well-being.

Overall, the dual phenomena of escalating demand for water and growing concerns about CECs necessitate the urgent development of comprehensive solutions to ensure water quality and safety. Efforts to better understand the effect of CECs on human health and the environment, coupled with the development of effective removal strategies and monitoring tools, are paramount for achieving such solutions. Membrane technologies offer promising approaches in the pursuit of cleaner water and a healthier environment. Continued research and proactive regulatory measures are essential aspects of effectively addressing the challenges posed by CECs and protecting our valuable water resources for future generations [20].

2. Membrane Treatment of Various CECs

2.1. Removal by FO Membranes

2.1.1. Effect of the physicochemical properties of CECs

The FO membrane separation process involves the use of draw solution with a high concentration and a great osmotic potential to draw water across a semi-permeable membrane from a feed source, where the key advantage of this process is that it involves relatively smaller energy than other separation processes, such as the RO process [21]. The physicochemical properties of chemicals encompass a range of characteristics, such as the freezing and boiling points, melting points, infrared spectra, electronic parameters, viscosity, molecular weight, and density. Certain properties, such as electronic parameters, molecular weight, and boiling/freezing points, have a direct influence on both environmental fate and health effects [22]. In the context of our study, which specifically examines the retention of emerging contaminants by membrane processes in water and wastewater treatment, it is essential to have a strong understanding of these physicochemical properties.

A previous study investigated the retention of several pharmaceuticals by an FO membrane to elucidate the separation mechanisms [23]. That study examined both the individual and combined performances of four specific trace PPCPs (ciprofloxacin, sulfamethoxazole, acetaminophen, and carbamazepine) during the FO process. 0.1 mol NaCl 1 L was used as the draw solute in this process. The mean retention rates followed a pattern based on molecular weight, with ciprofloxacin having the highest retention rate of 94.8% and acetaminophen having the lowest such rate of 29.1%. These findings reveal that, for single pharmaceuticals, their retention rates correlate with molecular size, with larger molecules being retained more effectively. In binary mixtures, the presence of additional molecules impacts retention differently. While the membrane surface charge was negative at pH 7 and positive at pH 5, negatively charged sulfamethoxazole at pH 7 gained high retention by negatively charged membrane. However, positively charged ciprofloxacin at pH 5 gained relatively weak electrostatic repulsion. Charged pharmaceuticals exhibit more complex behavior due to electrostatic interactions, while neutral pharmaceuticals rely primarily on molecular collisions and size considerations. pH can also have a significant impact on the retention of charged pharmaceuticals by altering their charge interactions with both the membrane surface and with each other. Fig. 1 shows the removal mechanism of ciprofloxacin and sulfamethoxazole [23].

A separate study found that the adsorption onto the FO membrane surface is primarily driven by hydrophobic interactions and hydrogen bonding [24]. Metronidazole, phenazone, sulfamethoxazole, diclofenac, bezafibrate, amitriptyline, trimethoprim, acetaminophen, caffeine, carbamazepine, linuron, and triclosan were selected as the target CECs. The hydrophobic neutral CECs, such as triclosan and linuron, showed substantial affinities with the membrane surface, thus enhancing partitioning, and stimulating transportation and strong adsorption to the membranes. Both negatively (amitriptyline and trimethoprim) and positively charged CECs (metronidazole, phenazone, sulfamethoxazole, diclofenac, and bezafibrate) showed relatively higher rejection than neutral CECs, which was attributed to the electrostatic repulsion and charge attraction. The results also showed that the removal and adsorption of neutral CECs were enhanced with increasing molecular weight and hydrophobicity. Fig. 2 Shows the schematic of the target CECs’ removal [24].

Altogether, the various studies above show that the physicochemical properties of contaminants significantly influence their adsorption onto FO membranes. Hydrophobic interactions primarily drive the removal of various CECs for membrane surfaces. The rejection and adsorption of CECs have been shown to increase with their molecular weight and hydrophobicity [24]. Studies on pharmaceutical retention also reveal correlations between retention rates and molecular size, as well as the complexities arising from the behavior of charged pharmaceuticals and the interaction with pH [23].

2.1.2. Effect of water quality conditions

To fully leverage the potential of FO for water treatment, appropriate draw solutes must be used [25]. The removal of CECs also depends on solution conditions such as pH, total dissolved solids, ionic strength, and temperature [4]. High draw solute concentration was shown to increase initial flux while causing quick decrease and unfavorable recoverability of FO membrane flux, while low draw solute concentration resulted in low reactor efficiency due to the long hydraulic retention time (HRT) [26]. The accumulation of salt, protein, polysaccharide, and humic acid was observed in the process, which was somewhat unfavorable to constant long-term operation. 1 M NaCl was employed as an optimal draw solute based on membrane fouling and anaerobic FO membrane bioreactor performance, and it is important to apply effective techniques to relieve salinity build-up to ensure stable long-term running [26]. In a study examining its utility for the purification of medical radioactive liquid waste, pH did not have a significant effect on external concentration polarization [27]. Iodine was considered as a target contaminant in the treatment of radioactive liquid waste. Findings indicated that FO demonstrated effective removal rates for both natural and radioactive iodine (¹²⁵I), reaching up to 99.3%. Such efficiency in removal was attained at elevated pH levels, primarily attributed to the electrical repulsion between iodine and the membrane. However, the results of that study showed that the main separation mechanism for ions is electrostatic repulsion. The FO membrane surface charge appears to be negative at pH 7 and 10 whereas it is neutral at pH 4. This results in natural iodine ions being rejected at pH 7 or 10 due to electric repulsion, while they can be transported to the draw solute at pH 4 [27].

A study using a baffled osmotic membrane bioreactor-microfiltration hybrid process under oxic/anoxic conditions to remove organic micropollutants (caffeine, atrazine, and atenolol) utilized NaCl, KCl, and sodium acetate (NaOAc) as draw solutes to compare the system’s performance [28]. Water flux was shown to be an important parameter during the FO process. The process achieved great retention of organic micropollutants and nutrients, with the maximum FO membrane rejection observed for atenolol, along with higher removal of organic micropollutants with organic draw solutes compared to inorganic solutes. Atrazine showed significant anoxic removal under different redox conditions, thus pointing to the presence of diverse microbial consortia that are responsible for enzyme secretion [28].

2.1.3. Effect of membrane properties and operating conditions

A submerged membrane bioreactor was used to remove and degrade carbamazepine from wastewater through a combination of membrane filtration and activated sludge biodegradation [29]. The removal rates were around 94.8-97.5% for chemical oxygen demand (COD), 93.6-99.4% for NH⁴⁺-N, and 88.2-94.5% for carbamazepine. Higher carbamazepine concentrations were found to be correlated with better COD and NH⁴⁺-N removal, but very high carbamazepine levels hindered such removal due to microbe inhibition. Oxidation, hydroxylation, and decarboxylation were identified as the carbamazepine degradation steps, and the specific bacteria named Delftia was determined to be involved [29].

Soluble microbial products and extracellular polymeric substances have been shown to play crucial roles in membrane fouling [28]. In particular, liquid chromatography-organic carbon detection analysis of mixed liquor supernatant in an osmotic membrane bioreactor-microfiltration (OMBR) revealed different biopolymer and humic substance ratios for various draw solutes. NaOAc led to higher biofilm growth and soluble microbial products formation than both NaCl and KCl. Mixed liquor extracellular polymeric substances had a higher protein-to-polysaccharide ratio, which impacted microbial floc hydrophobicity and contributed to fouling. Confocal laser scanning microscopy analysis showed that the biofilms were in descending thickness of NaOAc > NaCl > KCl, which was attributed to acetate ions promoting microbial growth. Fouled membranes contained 5-12% biopolymers and humic substances, thus aligning with the layer thickness trends. Larger sludge granules were linked to reduced fouling, and NaOAc led to smaller flocs but thicker biofilm due to decreased food-to-mass ratio [28].

2.2. Removal by RO Membranes

2.2.1. Effect of the physicochemical properties of CECs

The retention of pesticides by RO membranes is affected by various factors, such as molecular weight, projection area, size/steric exclusion, electrostatic repulsion, and hydrophobicity adsorption [30]. Tributyl phosphate, despite having a smaller molecular weight than other pesticides, has shown high rejection rates of 98-99% for all three RO membranes having similar NaCl rejection values of 99.0–99.8%, thus indicating the occurrence of effective removal through size exclusion and other mechanisms. Flutriafol, which has a slightly higher molecular weight and projection area than tributyl phosphate, exhibited lower NaCl rejection rates of 99.0% by one of the RO membranes, indicating that factors other than size exclusion may play a significant role in its removal. Dicofol, which has the highest molecular weight among the pesticides, demonstrated highly efficient removal by membranes, thus emphasizing the significant role played by size exclusion. Lower rejections were observed for irgarol, which is likely due to its lower molecular weight and hydrophobicity interactions with polyamide membranes, in addition to size exclusion effects. Comparing the removal using GE-AD membrane (pressure 20 bar), dicofol having the highest molecular weight showed the rejection of 99.1%, while irgraol showed the least removal of 95.3%. Overall, the retention of pesticides by RO membranes is the result of a complex interplay of multiple factors that govern their removal efficiency [30].

Triclosan demonstrated remarkably high removal efficiency of >99% when treated with RO membranes under various conditions [16]. This exceptional removal rate was primarily attributed to triclosan’s significant molecular weight and stokes molecular radius, which make size repulsion the dominant removal mechanism. Importantly, the removal of triclosan remained consistently high and was unaffected by changes in pH levels, thus distinguishing it from other studied PPCPs like ibuprofen and carbamazepine, which were shown to be influenced by both electrostatic interactions and pH variations during the RO process. Overall, the combination of size repulsion and electrostatic repulsion played a significant role in the final removal [16].

The removal of CECs was investigated using a pressure-driven RO process at various flux rates and recovery rates [31]. The results showed that RO was highly effective in removing CECs, particularly those with molecular weights above approximately 200 g/mol. CECs with molecular weights between 100 and 560 g/mol exhibited good removal rates, but compounds with molecular weights below 300 g/mol showed a sharp drop in removal efficiency. Negatively charged CECs were generally rejected more effectively than neutral or positively charged ones, and charged compounds were consistently removed by more than 90% [31].

2.2.2. Effect of water quality conditions

Once the removal efficiencies of ibuprofen, carbamazepine, and triclosan by RO were compared under various water quality conditions, the retention rate was found to increase with higher influent concentrations and pre-membrane pressures [16]. As shown in Fig. 3, the pH of the solution influenced the removal efficiencies of ibuprofen, carbamazepine, and triclosan due to its impact on electrostatic interactions between contaminants and the membrane. Unlike the removal of triclosan, the retention rates of ibuprofen and carbamazepine were influenced by electrostatic interactions and pH variations. The removal rate of ibuprofen depended on its ionization at pH levels above its pKa, which enhanced electrostatic repulsion with negatively charged membranes, thus resulting in increased removal. By contrast, carbamazepine had slightly lower removal rates than triclosan, and its removal was affected by both size repulsion and electrostatic repulsions. Overall, the differences in molecular properties and dominant removal mechanisms explained the varying removal rates of ibuprofen and carbamazepine during the RO process. Moreover, by using factorial design optimization, the highest removal rates achieved for ibuprofen, carbamazepine, and triclosan were 98.9, 97.5, and 99.0%, respectively, with an initial concentration of 500 μg/L, pre-membrane pressure of 16 bar, and pH 10 [16].

The effect of feedwater pH on the retention of aromatic amine pollutants by a RO membrane was investigated with both freshwater and brackish water [32]. In freshwater, increasing feedwater pH led to improved rejection of aromatic amine pollutants. For instance, at a low pressure (4.8 bar), rejection increased from 45.0% at pH 4 to 67.8% at pH 10. Higher pressure on the freshwater side further enhanced aromatic amine pollutants rejection, reaching a maximum of 78.4% at pH 10 and 7.6 bar. By contrast, brackish water showed a different trend, where the highest aromatic amine pollutants rejection consistently occurred at the alkaline pH of 10.0 and the lowest rejection consistently occurred at pH 7.0. The RO membrane performed well in removing charged aromatic amine pollutants in both freshwater and brackish water, while its capability to remove neutral aromatic amine pollutants was significantly reduced in brackish water at neutral pH, thus raising concerns for its practical applicability in water treatment. The RO membrane showed stable NaCl with pH and salt concentration having minimal impacts, thus suggesting that solute rejection by the RO membrane involves factors beyond electrostatic effects [32]. In a separate study, transmembrane pressure affected a slight difference in removing micropollutants [33]. By decreasing the feed pressure from 15.5 bar to 5 bar, the removal of micropollutants decreased. This exhibits that lower feed pressure led to lower micropollutant removal in the RO process. The effect of pressure on micropollutant removal is well shown in Fig. 4.

2.2.3. Effect of membrane properties and operating conditions

Evaluating a newly developed titania nanotube-incorporated RO membrane was focused in terms of its behavior in the removal of bisphenol A and caffeine at low concentrations [34]. That study used a thin-film nanocomposite membrane with a 0.01% titania nanotube, which showed improved permeabilities of 50 L/m²·h· bar for bisphenol A and 49 L/m²·h·bar for caffeine and achieved a satisfactory bisphenol A rejection of 89.0% and a caffeine rejection of 97.9%. The incorporated titania nanotube improved surface wettability, created nanochannels through the polyamide layer, and smoothened the membrane surface, thereby facilitating water passage and enhancing the permeability and rejection of bisphenol A and caffeine. It also showed promising antifouling properties due to its enhanced hydrophilicity and smooth surface [34].

The behavior of three RO membranes (BW30-LE, SW30-XLE, and GE-AD) was investigated in removing different pesticides from treated wastewater [30]. All three membranes had the polymer structure of polyamide. BW30-LE showed 99.0% NaCl rejection, 63-78/17 L/m²·h·bar pure water flux/transmembrane pressures, and a 49° contact angle. SW30-XLE showed 99.6% NaCl rejection, 30-40/55 L/m²·h·bar pure water flux/transmembrane pressures, and a 48° contact angle. GE-AD showed 99.75% NaCl rejection, 20-32/55 L/m²·h·bar pure water flux/transmembrane pressures, and a 62° contact angle. All membranes showed over 95% retention for all pesticides at different pressures. The BW30-LE membrane had the highest rejection for phosphate, while the GE-AD membrane performed best for others. Increasing the pressure did not significantly affect rejection. The deionized water flux increased with pressure, and the most hydrophilic BW30-LE membrane showed the highest flux. Contrary to expectations, the more hydrophobic GE-AD membrane had lower flux decline due to fouling, which was ascribed to the effects of surface roughness and porosity on fouling behavior. The three membranes differed in terms of their hydrophilicity, fouling tendencies, clean and wastewater flux behaviors, and their unexpected performances in relation to these properties [30].

2.3. Removal by NF Membranes

2.3.1. Effect of the physicochemical properties of CECs

The retention of various pesticides, pharmaceuticals, and industrial chemicals was investigated in a submerged membrane bioreactor process [35]. That study examined the effects of varying sludge retention times (SRT) and HRT. Hydrophobic pollutants generally showed high removal efficiencies exceeding 97.5%, while hydrophilic pollutants—and atrazine in particular—had lower removal rates. SRT of at least 15 days was required to achieve the effective removal of 4-tert-octylphenol; for fluoxetine, an SRT of 45 days or more was needed. Moreover, penconazole, a hydrophobic and moderately biodegradable pollutant, showed better removal rates than, which could possibly be attributed to its hydrophobic nature. That study also examined the impact of two different HRT (12 and 6 h) on the retention of target micropollutants. Most micropollutants, aside from fluoxetine, atrazine, and penconazole, were not significantly influenced by changes in HRT. Decreasing HRT from 12 to 6 hours reduced the removal efficiency of fluoxetine, which was attributed to its biodegradation resistance, but increased the removal efficiencies of atrazine and penconazole. The hydrophobic nature of penconazole likely contributed to its relatively higher removal rate compared to that of atrazine. Malathion, which is a hydrophilic organophosphorus pesticide, displayed high removal rates that were unaffected by HRT or SRT [35]. A different study investigated the performance of R- and S-enantiomers of ibuprofen in aqueous solution using nanofiltration membranes [36]. Enantioselective behavior was noted, with S-ibuprofen adsorbing more onto metal surfaces than R-ibuprofen. This result is attributed to molecular geometry, as supported by density functional theory computations. The difference in dipole moments between enantiomers correlated with their adsorption selectivity. Other factors like hydrogen bonding and molecular interactions might also contribute to the distinct enantiomer behavior. Fig. 5 shows that the removal efficiency was affected by SRT and HRT [36].

2.3.2. Effect of water quality conditions

The effectiveness of the Trisep TS80 NF membrane was investigated in removing two commonly found pharmaceutical compounds (sulfamethoxazole and diclofenac) from water [37]. At low pH, hydrophobic interactions dominate, resulting in high retention of sulfamethoxazole, while increasing pH leads to electrostatic repulsion and high rejections for both compounds. In a pH range around 4-6, the simultaneous presence of sulfamethoxazole and diclofenac reduces their individual rejections. Binary solutions show complex behavior, with pH influencing different mechanisms. The study concludes that pH optimization is vital for efficient pharmaceutical compound removal through nanofiltration, regardless of the water source or contaminant mixture, and that toxicity reduction is achieved by the treatment [37].

The effect of feedwater pH on the retention of aromatic amine pollutants by a NF membrane was evaluated in both freshwater and brackish water [32]. In freshwater, as the pH increased from acidic to neutral, there was only a slight improvement in aromatic amine pollutants rejection whereas there was a significant increase in alkaline pH. For example, at an applied pressure of 4.8 bar, rejection was 30.8% at pH 4.0 and 52.5% at pH 10.0. When the pressure was increased to 7.6 bar, rejection was increased to 43.9% at pH 4 and to 74.5% at pH 10. The great removal at alkaline pH was ascribed to electrostatic repulsion between the negatively charged membrane and deprotonated aromatic amine pollutants molecules. In brackish water, the screening effect of high NaCl concentration reduced electrostatic rejection, resulting in similar rejection levels at pH 4.0, 7.0, and 10.0, ultimately indicating the better performance of the NF membrane in rejecting neutral solutes in brackish water. For the NF membrane, the removal of NaCl was influenced by feedwater pH, with lower rejection at acidic pH and higher rejection at neutral and alkaline pH. The increase in NaCl removal with pH was attributed to electrostatic exclusion as a result of membrane charge. Unlike the RO membrane, the contradiction between aromatic amine pollutants and NaCl rejection in brackish water indicated that the rejection of aromatic amine pollutants deviated from simple electrolyte removal by the NF membrane [32].

In a study investigating the behavior of the R- and S-enantiomers of ibuprofen, it was found that, at pH values below the pKa of ibuprofen, both the equipment and membrane surfaces exhibited adsorption [36]. Up to 23% of the drug was adsorbed onto stainless steel equipment, primarily driven by S-ibuprofen. Additional adsorption occurred on the membrane surfaces, as confirmed by increased contact angles. The adsorption process can be explained by Freundlich and Langmuir isotherms for equipment and membrane, respectively. At pH 4.0 and racemic concentrations, the NF retention ranged from 34.5% to 49.5%, with S-ibuprofen rejected more than R-ibuprofen. Adsorption influenced rejection by 18.9% to 27.3%, and as pH increased, retention increased whereas adsorption was reduced [36].

2.3.3. Effect of membrane properties and operating conditions

During the removal process of estriol using polymeric NF, a significant observation was made concerning the thickness of the NF, which directly influenced the rate of estriol removal [2]. In particular, it was found that NF with a thinner thickness exhibited a higher area of contact due to an increased number of fibers, thus leading to improved efficiency in the removal of estriol. Conversely, NF with greater thickness resulted in a smaller area of contact, thus leading to reduced efficiency in the removal of the compound [2]. In a study using mixed matrix membranes to separate diclofenac sodium from an aqueous solution using manganese nanoparticles (MnO₂, Mn₂O₃) in polysulfide casting solutions, the addition of nanoparticles was found to enhance membrane properties, such as permeate flux, diclofenac removal efficiency, hydrophilicity, and mechanical strength [38]. The polysulfone/Mn₂O₃ membrane showed the highest diclofenac removal and antifouling properties. Incorporating a small amount of MnO₂ and Mn₂O₃ into the polysulfone casting solutions improves the membrane’s porosity, water flux, diclofenac rejection, and antifouling characteristics [38]. Another study investigated the adsorption behavior of ibuprofen enantiomers on flat-sheet equipment and membranes [36]. In that study, experiments with a range of initial feed concentrations from 100 to 1.50 mg/L showed that ibuprofen adsorption increased with concentration, following adsorption isotherms. The additional adsorption of 19.6 to 39.2% with membranes was observed, as confirmed by the increased hydrophobicity of the membrane [36].

2.4. Removal by UF Membranes

2.4.1. Effect of the physicochemical properties of CECs

In a study evaluating the removal efficiencies of diclofenac, paracetamol, and metronidazole using carbon-polymeric ultrafiltration membranes, the presence of powdered activated carbon to the membrane matrix appeared to significantly improve the removal of these pharmaceutical compounds from water [3]. Specifically, the removal efficiency increased by 34% for paracetamol and 28% for metronidazole when compared to conventional polymeric ultrafiltration membranes. The removal efficiencies of diclofenac, paracetamol, and metronidazole differ due to their molecular weights and hydrophobicity (log K_ow). The carbon-polymeric membranes used have a molecular weight cutoff of 35,000 Da, while the pharmaceuticals are much smaller. Pharmaceuticals with low log K_ow values have high hydrophilicity and are less effectively removed, while those with high log K_ow values are more hydrophobic and better adsorbed on the membrane surface, thus leading to higher removal rates [3].

UF is widely used in combination with other systems; e.g., a dual-functional ultrafiltration membrane. A dual-functional ultrafiltration membrane is a combination of hollow mesoporous carbon nanospheres and UF membrane that was examined in a prior study [39]. The target micropollutants of the study were tetracycline, 17β-estradiol, estrone, testosterone, and progesterone. Among the target contaminants, tetracycline and 17β-estradiol showed the highest removal efficiencies of 97% and 94%, respectively. The main removal mechanism is based on the adsorption of hollow mesoporous carbon nanospheres, which can be subclassified into hydrophobic interactions and π-π interactions. When the hormone has a large hydrodynamic diameter, the adsorption rate is reduced [39].

It is widely known in studies analyzing the presence of various CECs in treated wastewater that chemical properties affect the removal of CECs [40]. Pharmaceuticals such as carbamazepine, diclofenac, and ibuprofen, personal care products such as triclosan, and endocrine-disrupting compounds such as 17β-estradiol were treated by ozonation-UF combined system in that study. The results showed that the combined process of ultrafiltration with ozonation was effective in removing most of the CECs present in the secondary urban wastewater, although some CECs, such as carbamazepine and diclofenac, were not effectively removed by the treatment process. The removal rates of carbamazepine and diclofenac were lower than 50%. This result is attributed to the physicochemical properties of these compounds. Carbamazepine is a relatively hydrophobic compound, which means it has a low solubility in water and tends to adsorb onto surfaces. On the other hand, diclofenac is a relatively polar compound, which means it is more soluble in water but that it can be resistant to degradation by some treatment processes [40].

2.4.2. Effect of water quality conditions

A prior study used a graphene oxide chitosan hydrophilic nanoporous UF composite membrane for a sidestream membrane bioreactor process to remove the toxic pesticide atrazine [41]. The membrane showed excellent atrazine removal efficiency (> 95%) at optimized conditions. The mechanism of membrane fouling in that study was analyzed using Hermia’s model, and pneumatic back pulsing was applied to minimize fouling. Mixed liquor suspended solid was shown to affect both atrazine biodegradation and membrane fouling, with mixed liquor suspended solid dosage optimization enhancing atrazine removal. The membrane bioreactor process achieved about 95% atrazine removal along with complete removal of other contaminants. Water quality conditions play a pivotal role in UF membrane performance. Parameters like mixed liquor suspended solids dosage significantly impact both biodegradation and membrane fouling, in turn influencing contaminant removal [41].

2.4.3. Effect of membrane properties and operating conditions

In a previous study examining CoFe₂O₄-based particles blended with polyethersulfone polymer UF, these membranes were found to effectively degrade 70% of naproxen in batch experiments, with the best performance achieved at a CoFe₂O₄ concentration of 2.0% [42]. Catalytic UF membranes with longer residence times show higher naproxen degradation efficiency. Moreover, the introduction of a layer-by-layer assembly of polymers on the membrane surface enhances stability [42]. The main removal mechanisms of sulfamethoxazole, ciprofloxacin, trimethoprim, caffeine, and acetaminophen in the experiment were biodegradation and sorption to sludge and membrane surface [43]. The removal rates of the micropollutants were affected by the operating conditions of the UF, such as the water flux. A flux of 10 L/m²·h resulted in great removal efficiencies of micropollutants [43].

A separate study explored the influences of transmembrane pressure, time, and crossflow velocity on the efficiency of a developed UF membrane [44]. These parameters were all found to have crucial effects on permeate flux. Increasing the transmembrane pressure resulted in higher flux, which reached its peak at 5 bar, which was primarily attributed to the significant role played by transmembrane pressure as the driving force behind membrane filtration. Crossflow velocity also exhibited a synergistic effect on flux. The results of ANOVA analysis established that time significantly influenced flux. Similarly, transmembrane pressure was found to have a considerable impact on flux. Conversely, crossflow velocity showed no substantial impact on flux. In terms of ciprofloxacin removal, that study revealed that increasing transmembrane pressure led to a continuous rise in removal efficiency. While crossflow velocity had a minor effect on removal, with a modest increase from 96.7 to 98.9 between 1 L/min and 2 L/min crossflow velocity, time exhibited a positive impact on removal. The removal efficiency increased to 99 within the first hour, decreased to 90 at 90 minutes, and settled at 92.1 up to 180 minutes, which was attributed to adsorption-desorption processes. This study has highlighted time as a significant factor affecting removal efficiency, while transmembrane pressure and crossflow velocity had negligible effects [44].

In another study using a dual-functional ultrafiltration membrane, the removal rates of tetracycline and 17β-estradiol were also found to be affected by the operating conditions [39]. The best removal was shown at the low operation pressure of 0.12 bar. Water flux increment decreased the removal of both tetracycline and 17β-estradiol: Tetracycline removal in the permeate steadily decreased from 97% (50 L·h⁻¹·m⁻²) to 88 % (300 L·h⁻¹·m⁻²), while the 17β-estradiol removal ratio also decreased from 94 % (64 L·h⁻¹·m⁻²) to 40 % (480 L·h⁻¹·m⁻²). The water flux determines the contact time of the hormones, and the study results show that long contact times are required for efficient micropollutant adsorption and removal [39].

2.5. Removal by MF Membranes

2.5.1. Effect of the physicochemical properties of CECs

A prior study utilized experiments using a sericin-coated hollow fiber MF membrane to evaluate the removal efficiency of various PPCPs [45]. The membrane exhibited high removal efficiencies for ibuprofen, diclofenac, amoxicillin, and ciprofloxacin due to charge-based interactions and hydrogen bonding with functional groups on the membrane surface. However, for unionized steroid drugs such as estrone and β-estradiol, which exist largely in an uncharged species at neutral pH, the retention efficiency was lower, suggesting that hydrophobic interactions were not the main mechanism governing their adsorption. The presence of specific functional groups on the drugs and the membrane surface played a crucial role in their adsorption onto the membrane. The retention efficiencies of drugs and contaminants by MF membranes are influenced by their molecular properties and functional groups. Charge-based interactions and hydrogen bonding on membrane surfaces significantly affect the removal of various drugs, including ibuprofen, diclofenac, and antibiotics. Hydrophobic interactions play a smaller role in the adsorption of certain unionized steroid drugs [45].

2.5.2. Effect of water quality conditions

The effect of common background ions on the drug retention efficiency of the sericin coated membrane was examined using NaCl as monovalent salt and CaCl₂ and MgSO₄ as divalent salts at a concentration of 100 mg/L, along with drug solutions without salts [45]. The presence of salts, and of divalent salts in particular, led to a reduction in the removal efficiencies of ibuprofen, amoxicillin, and estrone. The inhibitory effect was attributed to the screening effect of the salts on the membrane surface, where charged anions/cations accumulated and obstructed binding sites for drug particles. Divalent salts affected the removal of amoxicillin and estrone, which was possibly attributed to complex formation and interference with hydrogen bonding between the drugs and the membrane surface. However, monovalent salt did not exhibit an inhibitory effect on drug adsorption [45].

The removal of organic microcontaminants was investigated using an innovative approach involving an OMBR-MF hybrid system under different redox conditions [28]. Three different draw solutes (NaCl and KCl as inorganic draw solutes along with NaOAc as an organic draw solute) and three model organic micropollutants (atenolol, caffeine, and atrazine) were examined. As shown in Fig. 6, the OMBR-MF system effectively removed organic micropollutants, with the highest FO membrane rejection achieved for atenolol (100%). The removal efficiency varied with different draw solutes, with inorganic draw solutes showing high removal rates. Atenolol exhibited the highest FO membrane rejection of 100% (NaCl, KCl, NaOAc) due to its positive charge and higher molar mass. Meanwhile, caffeine showed the highest removal rates with inorganic DS, while atrazine demonstrated highest removal with applying NaOAc as a draw solute [28].

2.5.3. Effects of membrane properties and operating conditions

A prior study found that using polyethersulfone MF led to the successful and efficient removal of 17β-estradiol from water [12]. By coating the membranes with hydrophilic amide functional groups, the wettability of the membrane surfaces was significantly enhanced, thus leading to more efficient 17β-estradiol removal. Moreover, by comparing functionalities, it was possible to determine that hydrophobic interactions played a minor role in the 17β-estradiol adsorption process [12]. Hydrophobic ceramic microfiltration membranes were created using chromium-laden waste from the tannery industry [46]. These membranes were effective in removing atrazine, showing efficiency of over 95%. The performance of the membranes appears to vary depending on various operating parameters, including transmembrane pressure, cross-flow velocity, and filtration time. Higher transmembrane pressure increased permeate flux but reduced atrazine rejection, while cross-flow velocity had a marginal impact on rejection but increased flux. Further, filtration time affected rejection due to concentration polarization. Atrazine removal improved with increasing concentration, and membrane fouling was mitigated by the pneumatic back pulsing mechanism of atrazine removal, which was attributed to hydrophobic and π-π interactions. Overall, these cost-effective ceramic membranes have the potential to efficiently remove toxic contaminants from water in wastewater treatment applications [46].

Experiments using ibuprofen as a model drug have shown that the retention rate of the sericin-coated hollow fiber MF membrane was influenced by the feed flow rate [45]. At lower flow rates (5 L/h and 10 L/h), the complete removal of ibuprofen was achieved for specific volumes of the solution filtered (1750 mL and 750 mL, respectively). However, as the flow rate increased to 15 L/h and 20 L/h, the removal rate was reduced significantly, owing to the existence of a shorter residence time, thus causing the drug particles to interact less with the membrane surface and pass through to the permeate. It also appeared that the increase in feed concentration led to earlier saturation of binding sites on the membrane surface, which caused the drug to appear more quickly in the permeate water [45].

Another study examined the impact of the hydrophilicity of the fabricated membrane [47]. That study used porous polyether sulfone MF, which has good water flux, porosity, and stability. During the pharmaceutical removal process, the adsorption of proteins on the membrane leads to a huge product loss and membrane fouling. That study showed that atom transfer radical polymerization increased the hydrophilicity of the membrane and reduced the surface roughness, thus resulting in the existence of less fouling caused by protein [47]. A different study investigated the removal rate of bisphenol A by MF with different microdomains [48]. Polyamide, polyvinylidene fluoride, nitrocellulose, and polytetrafluoroethylene were used as microdomains. Polyamide-MF showed the best adsorption capacity of 161 mg/g, while polyvinylidene fluoride-MF showed the best adsorption rate of 2.373 1/h. The hydrophilic microdomain offered the necessary energy to lower the energy barrier of the process while the hydrophobic microdomain was important for the driving mechanisms. The hydro-attractions of the microdomains resulted in different interactions. It was found that neither a highly hydrophobic membrane nor a highly hydrophilic membrane was effective for the retention of bisphenol A [48].

The retention of sulfonamide antibiotics was investigated using three types of graphene-based catalytic membranes [49]. Peroxydisulfate, peroxymonosulfate, and hydrogen peroxide were used for membrane fabrication. Sulfamethoxazole, sulfadiazine, sulfamerazine, and sulfamethazine were chosen as sulfonamide antibiotics. Peroxydisulfate- and peroxymonosulfate-based membranes worked by surface-active species and ¹O₂ as the main reactive oxygen species through non-radical pathways, while hydrogen peroxide-based membrane removed sulfonamide antibiotics by ^•OH reactions. Among the three chemicals considered, hydrogen peroxide has the lowest ability to drive the removal process. The peroxydisulfate-based system showed the maximum efficiency in removing sulfadiazine and sulfamethoxazole during continuous 10 h filtration, having removal rates exceeding 98% and 93%, respectively. The peroxymonosulfate-based system showed better removal for removing sulfamerazine and sulfamethazine, with respective removal rates of over 99% and 90% [49]. The removal rates of the micropollutants are listed in Table 1. In addition, a retention diagram of organic CECs during membrane treatments based on solute and membrane properties is presented in Fig. 7.

3. Conclusions and Areas of Future Research

In the pursuit of safeguarding water quality amidst the dual pressures of escalating demand and concerns about CECs, it is imperative to find comprehensive solutions. The present review underscores the urgency of understanding CEC impacts on water systems, advocating for effective removal strategies, and regulatory measures. Membrane technologies—FO, RO, NF, UF, and MF—emerge as promising tools for water quality preservation, thus necessitating ongoing research and proactive measures to secure this vital resource for future generations. It is essential to understand the pivotal factors influencing the effectiveness of FO membranes in CEC removal. Physicochemical properties significantly determine CEC adsorption onto FO membranes, where hydrophobic interactions dominate, particularly for neutral CECs. Factors like molecular weight and hydrophobicity correlate with enhanced rejection and adsorption. Moreover, water quality conditions, such as pH and solute concentrations, notably impact membrane performance and fouling. Membrane properties and operational conditions, including draw solute selection and microbial activities, also affect CEC removal and membrane fouling dynamics. These complexities highlight the need for meticulous considerations in system design and operational parameters for optimizing FO processes. The efficiency of RO membranes in removing CECs hinges upon various influencing factors: For example, molecular properties such as size, hydrophobicity, and charge significantly dictate CEC removal mechanisms, including size exclusion, hydrophobic interactions, and electrostatic repulsions. Water quality conditions, including pH, influent concentrations, and pressures, also play crucial roles in CEC removal efficiency, thus affecting electrostatic interactions and size-based removal. Further, membrane properties like surface modifications and differing hydrophilicity impact permeability, rejection rates, and fouling tendencies. RO membranes exhibit intricate behavior shaped by structure, hydrophilicity, and surface characteristics, thus shaping their effectiveness in removing diverse CECs. The efficacy of NF membranes in removing CECs is reliant on multiple factors. While hydrophobic pollutants generally display high removal rates, hydrophilic compounds exhibit varying efficiencies that are affected by parameters like SRT and HRT. Existing studies emphasize the enantioselective behavior of compounds like ibuprofen, showcasing the significance of molecular geometry in surface adsorption. Water quality conditions, particularly pH, notably influence pharmaceutical removal, thereby impacting hydrophobic interactions and electrostatic repulsion. Moreover, NF membrane properties, such as thickness and nanoparticle integration, significantly impact efficiency, flux, and rejection rates, thereby underscoring the crucial importance of membrane design and composition in CEC removal.

UF membranes demonstrate the ability to efficiently remove diverse CECs, influenced by molecular properties and membrane design. Incorporating powdered activated carbon in membrane matrices notably enhances the removal of pharmaceutical compounds, particularly hydrophobic ones, thus showcasing the impact of membrane modifications. Moreover, a number of studies demonstrate the importance of water quality conditions like mixed liquor suspended solids and pH, which govern both biodegradation and membrane fouling, which are the critical factors that affect contaminant removal in UF processes. Moreover, the interplay between membrane properties like composition and structure, in addition to operating conditions such as transmembrane pressure and flux rates, substantially affects removal efficiency, thereby highlighting the need for tailored membrane designs and operational optimization.

MF membranes prove effective in removing diverse CECs through molecular interactions and membrane modifications. Charge-based interactions and hydrogen bonding significantly contribute to the high removal efficiency of pharmaceuticals, while hydrophobic interactions exhibit varying significance, particularly for unionized steroid drugs. Moreover, water quality conditions notably impact MF efficiency. Further, salts, and particularly divalent salts, hinder drug adsorption due to surface obstruction and interference with hydrogen bonding. Membrane properties, including surface functionalities and materials, also affect CEC removal. Hydrophilic modifications enhance the removal of specific contaminants, while cost-effective ceramic membranes have shown promise for use toxic contaminant removal. Operational parameters like feed flow rate and concentration influence removal efficiency, where lower flow rates and optimized concentrations enhance removal, but higher flow rates reduce efficiency due to shorter residence times. Novel membrane designs offer varied adsorption capacities for targeted contaminants, thus showcasing the potential utility of tailored MF membranes in efficient CEC removal across different contexts.

Directions for future research in membrane technology include studies exploring advanced surface modifications to enhance selectivity and fouling resistance, investigating novel materials and nanotechnology applications for membrane fabrication, optimizing operational parameters for energy-efficient processes, and elucidating the environmental impact and long-term sustainability of membrane-based water treatment systems. In addition, a comprehensive evaluation of the environmental and economic implications through lifecycle analyses is imperative for assessing the sustainability of membrane-based water treatment systems in comparison to conventional methods. This analysis will provide deeper insights into the economic viability of membrane technologies and their long-term effects on resource allocation, energy consumption, and waste generation. It is essential to examine how advancements in membrane technology may influence regulatory frameworks and policy development in the field of water and wastewater treatment. By scrutinizing the potential impacts of technological advancements on regulatory landscapes, we can better anticipate and shape forthcoming regulations and industry standards. Continued research and development in these areas will advance the effectiveness and sustainability of membrane technologies in addressing the challenges posed by CECs in water systems.

Acknowledgements

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. RS-2023-00217228, No. RS-2023-00272059, and No. RS-2023-00282898).

Notes

Author Contributions

S.K. (Undergraduate Student) wrote the manuscript. B.J. (Principal Researcher), B.J. (Research Professor), C.P. (Associate Professor), M.J. (Professor) helped in developing the conceptualization and reviewed the manuscript. S.N. (Assistant Professor) provided valuable research insights into the study. Y.Y. (Professor/Supervisor) provided valuable research insights into the study and helped with publishing.

Conflicts of Interests Statement

The authors declare that they have no conflict of interest.

References

1. Aguinaco A, Beltran FJ, Garcia-Araya JF, Oropesa AL. Photocatalytic ozonation to remove the pharmaceutical diclofenac from water: Influence of variables. Chem. Eng. J. 2012;189:275–282. https://doi.org/10.1016/j.cej.2012.02.072

2. Nectoux AD, Medeiros LF, Rodrigues RDB, Soares RMD, Fernandes AN. Electrospun nanofibrous membranes for solid-phase extraction of estriol from aqueous solution. J. Appl. Polym. Sci. 2019;136(11)47189. https://doi.org/10.1002/app.47189

3. Nadour M, Boukraa F, Benaboura A. Removal of diclofenac, paracetamol and metronidazole using a carbon-polymeric membrane. J. Environ. Chem. Eng. 2019;7(3)103080. https://doi.org/10.1016/j.jece.2019.103080

4. Bayode AA, dos Santos DM, Omorogie MO, et al. Carbon-mediated visible-light clay-Fe2O3-graphene oxide catalytic nanocomposites for the removal of steroid estrogens from water. J. Water Process. Eng. 2021;40:101865. https://doi.org/10.1016/j.jwpe.2020.101865

5. Bayode AA, Vieira EM, Moodley R, et al. Tuning ZnO/GO p-n heterostructure with carbon interlayer supported on clay for visible-light catalysis: Removal of steroid estrogens from water. Chem. Eng. J. 2021;420:127668. https://doi.org/10.1016/j.cej.2020.127668

6. Esplugas S, Bila DM, Krause LGT, Dezotti M. Ozonation and advanced oxidation technologies to remove endocrine disrupting chemicals (EDCs) and pharmaceuticals and personal care products (PPCPs) in water effluents. J. Hazard. Mater. 2007;149(3)631–642. https://doi.org/10.1016/j.jhazmat.2007.07.073

7. Chen CQ, Chen DZ, Xie SS, Quan HY, Luo XB, Guo L. Adsorption behaviors of organic micropollutants on zirconium metal-organic framework UiO-66: Analysis of surface interactions. ACS Appl. Mater. Interfaces. 2017;9(46)41043–41054. https://doi.org/10.1021/acsami.7b13443

8. Rana D, Narbaitz RM, Garand-Sheridan AM, et al. Development of novel charged surface modifying macromolecule blended PES membranes to remove EDCs and PPCPs from drinking water sources. J. Mater. Chem. A. 2014;2(26)10059–10072. https://doi.org/10.1039/c4ta01530d

9. Loos R, Carvalho R, Antonio DC, et al. EU-wide monitoring survey on emerging polar organic contaminants in wastewater treatment plant effluents. Water Res. 2013;47(17)6475–6487. https://doi.org/10.1016/j.watres.2013.08.024

10. Padhye LP, Yao H, Kung’u FT, Huang CH. Year-long evaluation on the occurrence and fate of pharmaceuticals, personal care products, and endocrine disrupting chemicals in an urban drinking water treatment plant. Water Res. 2014;51:266–276. https://doi.org/10.1016/j.watres.2013.10.070

11. USEPA. Endocrine disruptor screening program report to congress [Internet]. 2000. Available from: https://www.epa.gov/sites/default/files/2015-12/documents/reporttocongress0800.pdf

11. USEPA. Environmental Protection Agency – Endocrine disruptor screening program. Report to congress. 2000;

12. Niavarani Z, Breite D, Prager A, Abel B, Schulze A. Estradiol removal by adsorptive coating of a microfiltration membrane. Membranes. 2012;11(2)99. https://doi.org/10.3390/membranes11020099

13. Chekli L, Phuntsho S, Kim JE, et al. A comprehensive review of hybrid forward osmosis systems: Performance, applications and future prospects. J. Membr. Sci. 2016;497:430–449. https://doi.org/10.1016/j.memsci.2015.09.041

14. Xu L, Ma X, Niu JF, Chen J, Zhou CZ. Removal of trace naproxen from aqueous solution using a laboratory-scale reactive flow-through membrane electrode. J. Hazard. Mater. 2019;379:191–209. https://doi.org/10.1016/j.jhazmat.2019.05.085

15. Taniguchi Y. An overview of pretreatment technology for reverse osmosis desalination plants in Japan. Desalination. 1997;110(1-2)21–35. https://doi.org/10.1016/s0011-9164(97)00081-7

16. Liu JN, Duan L, Gao QS, Zhao Y, Gao F. Removal of typical PPCPs by reverse osmosis membranes: optimization of treatment process by factorial design. Membranes. 2023;13(3)355. https://doi.org/10.3390/membranes13030355

17. Anand A, Unnikrishnan B, Mao JY, Lin HJ, Huang CC. Graphene-based nanofiltration membranes for improving salt rejection, water flux and antifouling-A review. Desalination. 2018;429:119–133. https://doi.org/10.1016/j.desal.2017.12.012

18. Feins M, Sirkar KK. Highly selective membranes in protein ultrafiltration. Biotechnol. Bioeng. 2004;86(6)603–611. https://doi.org/10.1002/bit.20069

19. Fiksdal L, Leiknes T. The effect of coagulation with MF/UF membrane filtration for the removal of virus in drinking water. J. Membr. Sci. 2006;279(1-2)364–371. https://doi.org/10.1016/j.memsci.2005.12.023

20. Kim S, Chu KH, Al-Hamadani YAJ, et al. Removal of contaminants of emerging concern by membranes in water and wastewater: A review. Chem. Eng. J. 2018;335:896–914. https://doi.org/10.1016/j.cej.2017.11.044

21. Johnson DJ, Suwaileh WA, Mohammed AW, Hilal N. Osmotic’s potential: An overview of draw solutes for forward osmosis. Desalination. 2018;434:100–120. https://doi.org/10.1016/J.desal.2017.09.017

22. Rostvall A, Zhang W, Dürig W, et al. Removal of pharmaceuticals, perfluoroalkyl substances and other micropollutants from wastewater using lignite, xylit, sand, granular activated carbon (GAC) and GAC+Polonite® in column tests - Role of physicochemical properties. Water Res. 2018;137:97–106. https://doi.org/10.1016/j.watres.2018.03.008

23. Cao Q, Yang XX, Yang WY, Wang QH, Hao XD. Separation of trace pharmaceuticals individually and in combination via forward osmosis. Sci. Total Environ. 2020;718:137366. https://doi.org/10.1016/j.scitotenv.2020.137366

24. Jang D, Jeong S, Jang A, Kang S. Relating solute properties of contaminants of emerging concern and their rejection by forward osmosis membrane. Sci. Total Environ. 2018;639:673–678. https://doi.org/10.1016/j.scitotenv.2018.05.078

25. Ge QC, Lau CH, Liu MH. A novel multi-charged draw solute that removes organic arsenicals from water in a hybrid membrane process. Environ. Sci. Technol. 2018;52(6)3812–3819. https://doi.org/10.1021/acs.est.7b06506

26. Gao Y, Fang Z, Chen C, et al. Evaluating the performance of inorganic draw solution concentrations in an anaerobic forward osmosis membrane bioreactor for real municipal sewage treatment. Bioresour. Technol. 2020;307:123254. https://doi.org/10.1016/j.biortech.2020.123254

27. Lee S, Kim Y, Park J, Shon HK, Hong S. Treatment of medical radioactive liquid waste using Forward Osmosis (FO) membrane process. J. Membr. Sci. 2018;556:238–247. https://doi.org/10.1016/j.memsci.2018.04.008

28. Pathak N, Li S, Kim Y, et al. Assessing the removal of organic micropollutants by a novel baffled osmotic membrane bioreactor-microfiltration hybrid system. Bioresour. Technol. 2018;262:98–106. https://doi.org/10.1016/j.biortech.2018.04.044

29. Yao MC, Duan L, Wei J, Qian F, Hermanowicz SW. Carbamazepine removal from wastewater and the degradation mechanism in a submerged forward osmotic membrane bioreactor. Bioresour. Technol. 2020;314:1–8. https://doi.org/10.1016/j.biortech.2020.123732

30. Ates N, Uzal N, Yetis U, Dilek FB. Removal of pesticides from secondary treated urban wastewater by reverse osmosis. Environ. Sci. Pollut. Res. 2023;30(4)8732–8745. https://doi.org/10.1007/s11356-022-20077-5

31. Shad MF, Juby GJG, Delagah S, Sharbatmaleki M. Evaluating occurrence of contaminants of emerging concerns in MF/RO treatment of primary effluent for water reuse - Pilot study. J. Water Reuse Desalination. 2019;9(4)350–371. https://doi.org/10.2166/wrd.2019.004

32. Shah IA, Ali S, Yang Z, Ihsanullah I, Huang HO. Effects of water matrix on the rejection of neutral pharmaceutically active compound by thin-film composite nanofiltration and reverse osmosis membranes. Chemosphere. 2022;303:135211. https://doi.org/10.1016/j.chemosphere.2022.135211

33. Alonso E, Sanchez-Huerta C, Ali Z, Wang YG, Fortunato L, Pinnau I. Evaluation of nanofiltration and reverse osmosis membranes for efficient rejection of organic micropollutants. J. Membr. Sci. 2024;693:122357. https://doi.org/10.1016/j.memsci.2023.122357

34. Ahmad NA, Goh PS, Azman N, et al. Enhanced removal of endocrine-disrupting compounds from wastewater using reverse osmosis membrane with titania nanotube-constructed nanochannels. Membranes. 2022;12(10)12100958. https://doi.org/10.3390/membranes12100958

35. Caglak A, Chormey DS, Bakirdere S, Engin GO. Performance evaluation of ceramic membrane bioreactor: effect of operational parameters on micropollutant removal and membrane fouling. Environ. Sci. Pollut. Res. 2022;29(45)68306–68319. https://doi.org/10.1007/s11356-022-20612-4

36. Higgins CJ, Duranceau SJ. Removal of enantiomeric ibuprofen in a nanofiltration membrane process. Membranes. 2020;10(12)383. https://doi.org/10.3390/membranes10120383

37. Gomes D, Cardoso M, Martins RC, Quinta-Ferreira RM, Gando-Ferreira LM. Removal of a mixture of pharmaceuticals sulfamethoxazole and diclofenac from water streams by a polyamide nanofiltration membrane. Water Sci. Technol. 2020;81(4)732–743. https://doi.org/10.2166/wst.2020.166

38. Abdallah H, Gaballah S, Mansour MS, Farag H, Farid YH. Separation of diclofenac sodium using polysulfone membranes incorporated with manganese nanoparticles. Chem. Eng. Technol. 2022;46:600–613. https://doi.org/10.1002/ceat.202200320

39. Liao ZP, Nguyen MN, Wan GJ, et al. Low pressure operated ultrafiltration membrane with integration of hollow mesoporous carbon nanospheres for effective removal of micropollutants. J. Hazard. Mater. 2020;397:122779. https://doi.org/10.1016/j.jhazmat.2020.122779

40. Graça CAL, Ribeirinho-Soares S, Abreu-Silva J, et al. A pilot study combining ultrafiltration with ozonation for the treatment of secondary urban wastewater: Organic micropollutants, microbial load and biological effect. Water. 2020;12(12)3458. https://doi.org/10.3390/w12123458

41. Mukherjee D, Dewanjee A, Ghosh S, Majumdar S. Development of graphene oxide/chitosan composite membrane on ceramic support for atrazine remediation by MBR process. Environ. Sci. Pollut. Res. 2018;25(33)33334–33352. https://doi.org/10.1007/s11356-018-3255-9

42. Wang T, de Vos WM, de Grooth J. CoFe2O4-peroxymonosulfate based catalytic UF and NF polymeric membranes for naproxen removal: The role of residence time. J. Membr. Sci. 2022;646:120209. https://doi.org/10.1016/j.memsci.2021.120209

43. Isik O, Erbil MC, Abdelrahman AM, et al. Removal of micropollutants from municipal wastewater by membrane bioreactors: Conventional membrane versus dynamic membrane. J. Environ. Manage. 2022;303:114233. https://doi.org/10.1016/j.jenvman.2021.114233

44. Bhattacharya P, Mukherjee D, Dey S, Ghosh S, Banerjee S. Development and performance evaluation of a novel CuO/TiO2 ceramic ultrafiltration membrane for ciprofloxacin removal. Mater. Chem. Phys. 2019;229:106–116. https://doi.org/10.1016/j.matchemphys.2019.02.094

45. Verma VK, Subbiah S. Sericin-coated polymeric microfiltration membrane for removal of drug-based micropollutants. J. Chem. Technol. Biotechnol. 2019;94(11)3625–3636. https://doi.org/10.1002/jctb.6168

46. Mukherjee D, Kar S, Mandal A, Ghosh S, Majumdar S. Immobilization of tannery industrial sludge in ceramic membrane preparation and hydrophobic surface modification for application in atrazine remediation from water. J. Eur. Ceram. Soc. 2019;39(10)3235–3246. https://doi.org/10.1016/j.jeurceramsoc.2019.04.008

47. Yun KH, Sharma K, Kim HU, Bae TH. Modification of a PES microfiltration membrane to enhance sterile filtration by inhibiting protein adsorption. J. Ind. Eng. Chem. 2023;123:311–319. https://doi.org/10.1016/j.jiec.2023.03.048

48. Sun JX, Jiang XT, Zhou Y, Fan JX, Zeng GM. Microfiltration membranes for the removal of bisphenol A from aqueous solution: Adsorption behavior and mechanism. Water. 2022;14(15)2306. https://doi.org/10.3390/w14152306

49. Xia X, Yan Y, Luo JP, Liu TT, Wu BD, Qian FY. Effects of peroxide types on the removal performance and mechanism of sulfonamide antibiotics using graphene-based catalytic membranes. Process Saf. Environ. Prot. 2023;179:362–372. https://doi.org/10.1016/j.psep.2023.09.022

50. Yao MC, Duan L, Song YH, Hermanowicz SW. Degradation mechanism of ibuprofen via a forward osmosis membrane bioreactor. Bioresour. Technol. 2021;321:124448. https://doi.org/10.1016/j.biortech.2020.124448

51. Hidalgo AM, Gomez M, Murcia MD, et al. Ibuprofen removal by graphene oxide and reduced graphene oxide coated polysulfone nanofiltration membranes. Membranes. 2022;12(6)562. https://doi.org/10.3390/membranes12060562

52. Mukherjee D, Bhattacharya P, Jana A, et al. Synthesis of ceramic ultrafiltration membrane and application in membrane bioreactor process for pesticide remediation from wastewater. Process Saf. Environ. Prot. 2018;116:22–33. https://doi.org/10.1016/j.psep.2018.01.010

53. Lou YY, Tan FJ, Zeng R, Wang MG, Li P, Xia SJ. Preparation of cross-linked graphene oxide on polyethersulfone membrane for pharmaceuticals and personal care products removal. Polym. 2020;12(9)1921. https://doi.org/10.3390/polym12091921

54. Romdhani M, Aloulou W, Aloulou H, Duplay J, Charcosset C, Ben Amar R. Performance assessment of a new flat sepiolite clay-based ultrafiltration membrane for the removal of paracetamol and indigo blue dyes from two synthetic aqueous solutions. Sustainability. 2024;16(5)1860. https://doi.org/10.3390/su16051860

55. Kamaz M, Wickramasinghe SR, Eswaranandam S, et al. Investigation into micropollutant removal from wastewaters by a membrane bioreactor. Int. J. Environ. Res. Public Health. 2019;16(8)1363. https://doi.org/10.3390/ijerph16081363

56. Bellona C, Drewes JE, Xu P, Amy G. Factors affecting the rejection of organic solutes during NF/RO treatment - a literature review. Water Res. 2004;38(12)2795–2809. https://doi.org/10.1016/j.watres.2004.03.034

Fig. 1

Various mechanisms on FO separation of binary trace pharmaceuticals: (a) collision between pharmaceutical molecules without charge; (b) repulsion between pharmaceutical molecule and membrane with same charge; (c) attraction between pharmaceutical molecule and membrane with opposite charge; (d) attraction between pharmaceutical molecules with opposite charge [23].

Fig. 2

The proposed schematic for the mechanism of CECs rejection by the FO membrane [24].

Fig. 3

Removal rate variation of various PPCPs under different conditions regarding (a) Concentration; (b) Membrane pressure, and (c) pH [16].

Fig. 4

Impact of transmembrane pressure on organic micropollutant rejection by a commercial RO seawater membrane (RO4) and a defect-free piperazine-based NF membrane (PIP-KRO), The cross-flow operating conditions were as follows: initial organic micropollutant concentration, 12 μg/L; background electrolyte, 1 mM (84 mg/L) NaHCO₃; temperature, 23°C; pH 8; and cross-flow velocity, 0.37 m/s The error bars indicate standard deviation of the measurements obtained from the samples [33].

Fig. 5

Calculated removal efficiencies of micropollutant at different SRTs and HRTs [35].

Fig. 6

Variations of OMPs removal in baffled OMBR-MF hybrid system using different DS [28].

Fig. 7

Retention diagram for organic CECs during membrane treatment based on solute and membrane properties modified from [20, 56].

Table 1

Summary of selected CEC removal by FO, RO, NF, UF, and MF membranes.

Membrane class	CEC class	Experimental condition	C_o and water type	Key removal (%)	Key finding	Ref.
FO	Atenolol, caffeine, atrazine	Baffled osmotic membrane bioreactor-microfiltration 0.0264 m² area	SWW	89–96 94–100 16–40	Since atenolol has the largest molecular weight among these three organic micropollutants, it has the greatest detention time. Bioflim of NaOAc has resulted in great retention.	[28]
	Ibuprofen	NH₄⁺-N COD	50~200 μg/L	> 98 (NH₄⁺-N) 96.32–98.53 (COD)	Since ibuprofen has small molecular weight, the retention effectiveness largely depends on biodegradation efficacy of the system.	[50]
	Ciprofloxacin, sulfamethoxazole, acetaminophen, carbamazepine	Cross-flow CTA-ES FO Charged binary trace pharmaceuticals 0.1 M NaCl 1 L	50 μg/L ciprofloxacin + 50 μg/L sulfamethoxazole, 50 μg/L sulfamethoxazole + 50 μg/L acetaminophen	90.1 (ciprofloxacin + sulfamethoxazole) 85.7 (sulfamethoxazole + acetaminophen)	Negatively charged pharmaceuticals have higher retention while positively charged pharmaceuticals weaken electrostatic repulsion and increased retention. The collision effects, electrostatic interactions, and the change of the total molecular weight has lowered the retention charged binary trace pharmaceuticals.	[23]
	Carbamazepine	Dead-end	50~200 μg/L	88.20–94.45	COD and NH₃⁺-N were related with the concentration of the carbamazepine and influenced the retention rate. Also, high carbamazepine levels hindered its own removal due to inhibiting microbes.	[29]
	Ciprofloxacin, sulfamethoxazole, acetaminophen, carbamazepine	Cross-flow Single-pharmaceuticals	100 μg/L 1000 μg/L SDW	94.8 86.4 29.1 83.0	In single pharmaceutical eliminating system, the size of molecule majorly affected the retention and removal rate.	[23]
	Metronidazole, phenazone, sulfamethoxazole, diclofenac, bezafibrate, amitriptyline, trimethoprim, acetaminophen, caffeine, carbamazepine, linuron, triclosan	Cross-flow Flow rate 702 m/min Gear pump pH 7 ± 0.2	100 μg/L	4.5–5 μg/cm² ^a (triclosan) 2.5–3 μg/cm² ^a (linuron) > 1.0 μg/cm² ^a (metronidazole, phenazone, sulfamethoxazole, diclofenac, bezafibrate) 1.0–1.5 μg/cm² ^a (amitriptyline, trimethoprim)	The removal rate was influenced by size, electrostatic repulsion, and hydrophobicity. The retention rate increased with increasing the molecular weight of CECs. In addition, charged CECs showed a relatively high removal rate. Negatively charged CECs exhibited a greater removal rate than positively charged CECs/	[24]
RO	Ibuprofen	16 bar pH 10	50~1000 μg/L	98.9	The retention rate of the ibuprofen differed owing to the size repulsion and electrostatic repulsion.	[16]
	Carbamazepine	16 bar pH 10	50~1000 μg/L	97.5	Retention rate was influenced highly owing to its initial concentration of the carbamazepine and increased promotionally.	[16]
	Triclosan	16 bar pH 10	50~1000 μg/L	99.0	Due to triclosan’s large molecular weight, the steric exclusion was the dominant factor of its retention rate.	[16]
	Bisphenol A, caffeine	Cross-flow Polyamide thin-film nanocomposite membranes	10 mg/L	> 75 > 93	The size exclusion made the significant difference of the retention rate between bisphenol A and caffeine. Also, there was the hydrogen bond interaction between membrane surface and bisphenol A molecule.	[34]
	Acetaminophen, primidone, bisphenol A, carbamazepine, estrone, triclosan, ethinyl estradiol	Cross-flow 1 mM NaHCO₃ 23°C 15.5 bar transmembrane pressure pH 8	12 μg/L	98.4 98.9 98.3 98.7 99.4 98.6 99.4	For non-ionic contaminants, the retention was mainly driven by size exclusion and hydrophobic/hydrophilic interactions. The retention efficiency was affected by the size of the solutes and their interactions with the membrane surface properties	[33]
	Ibuprofen, naproxen, mefenamic acid, gemfibrozil, ketoprofen, diclofenac	Cross-flow 1 mM NaHCO₃ 23°C 15.5 bar transmembrane pressure pH 8	12 μg/L	99.4 99.5 99.6 99.4 99.5 99.5	For ionic hydrophobic contaminants, the rejection was affected by electrostatic repulsion interactions. Membrane surface charge played an important role in enhancing the removal through electrostatic interactions between the charged compounds and the negatively charged membrane surface.	[33]
NF	Diclofenac	Dead-end Manganese-PSF	30–120 mg/L SDW	71.6 (MnO₂-PSF), 98.4 (MnO₃-PSF)	Mechanical testing revealed that higher tensile strength was obtained with embedding MnO₂ nanoparticles in the polysulfone membrane matrix	[38]
	Ibuprofen	Polysulfone pH 7 Pressure 15 bar	7.5 mg/L	77	By coating polysulfone membrane surface with nanoparticles, its pore size became smaller.	[51]
	Eristol	Polymer membrane	1.0 mg/L NSW	77–80	Since PCL polymer membrane has less thick nanofibers, its contact area is higher than PA-6 polymer membrane. The difference of contact area affects removal efficiency of eristol.	[2]
	Acetaminophen, primidone, bisphenol A, carbamazepine, estrone, triclosan, ethinyl estradiol	Cross-flow 1 mM NaHCO₃ 23°C 15.5 bar transmembrane pressure pH 8	12 μg/L	95.7 99.1 99.2 98.7 99.5 98.4 99.4	For non-ionic contaminants, the retention was mainly driven by size exclusion. NF membranes show varying degrees of retention for non-ionic compounds, with removal percentages typically ranging between 2% and 26%.	[33]
	Ibuprofen, naproxen, mefenamic acid, gemfibrozil, ketoprofen, diclofenac	Cross-flow 1 mM NaHCO₃ 23°C 15.5 bar transmembrane pressure pH 8	12 μg/L	99.2 99.3 99.6 99.5 99.2 99.5	For ionic hydrophobic contaminants, the greater rejection was obtained comparing to non-ionic compounds. The higher removal of was attributed owing to electrostatic interactions, particularly through Donnan exclusion mechanisms. The negative surface charge of the membranes at pH 8 improves the removal through electrostatic repulsion effects.	[33]
UF	Atrazine	60 mill	1 mg/L IWW	97	UF-MBR process leads to greater removal and lower fouling compared to MF-MBR.	[52]
	Ibuprofen, gemfibrozil, triclosan	Dead-end Cross-linked GO-PPD pH 3		34.8 52.8 99	GO added to the membrane made the space between membrane layers smaller, which causes greater removal rate. Among three CECs in this experiment, triclosan exhibited the maximum removal rate owing to its benzene ring and steric hindrance, and hydrophobicity.	[53]
	Paracetamol, metronidazole	2.5% PAC 0.5% MC M3	10 mg/L	34 28	By adding activated carbon on the membrane, removal of pharmaceutical compounds was enhanced.	[3]
	Sulfamethoxazole, ciprofloxacin, trimethoprim, caffeine, acetaminophen	10 L/m² · h average flux APAP > 0.1 μg/L 55 d 0.05 μm pore	> 0.1 μg/L Raw MWW	> 68.9 > 72.2 > 65.2 > 99.6 > 99.7	The high average removal of caffeine and acetaminophen was achieved by their well-biodegradable chemical structure.	[43]
	Tetracycline, 17β-estradiol,	Dual-functional UF Mesoporous carbon nanosphere loaded 0.12 bar (>0.15 bar)	100 ng/L Milli-Q water 1 mmol/L NaHCO3, 10 4mmol/L NaCl	97 (50 L · h⁻¹ · m⁻²) 94 (64 L · h⁻¹ · m⁻²)	Among diverse hormones, tetracycline and 17β-estradiol showed high removal rate owing to their small hydrodynamic diameter. Also, the retention rate of tetracycline and 17β-estradiol decreased as the water flux increased.	[39]
	Paracetamol	Dead-end Sepiolite membrane sintered at 650 °C 50 L/m² · h permeate flux 25 °C 3 bar	40 mg/L	100	The functional groups on the surface of sepiolite membrane effectively absorbed paracetamol, exhibiting decent fouling resistance. With decent flux recovery ratio of 96.7%, the membrane also exhibited great permeate flux recovery, which leads to cost effectiveness.	[54]
MF	Acetaminophen, amoxicillin, atrazine, estrone, triclosan	PVDF	SWW	100 100 < 25 98 100	Removal rate of each micropollutants varied due to adsorption degrees and biodegradation rerates.	[55]
	Atrazine	SRT 90 d, HRT 12 h SRT 45 d, HRT 6 h	1.0 μL SWW	17.5 (SRT 90 d, HRT 12 h) 42.5 (SRT 45 d, HRT 6 h)	Microcontaminants were released back during the process of SRT 45 d–90 d.	[35]
	Fluoxetine	Ceramic SRT 15 d, HRT 12 h SRT 90 d, HRT 12 h	1.0 μL SWW	84.9 (SRT 15 d, HRT 12 h) 97.5 (SRT 90 d, HRT 12 h)	Since fluoxetine is persistent to biodegradation, its removal rate is related to sorption onto solids.	[35]
	Atrazine	Cross-flow contact angle 141° 2 bar	0.5–5 mg	94.5–98.5	The removal performance was influenced by transmembrane pressure and filtration time. Retention mechanism involved hydrophobic and π-π interactions, and the membranes showed robustness.	[46]
	Bisphenol A	Room temperature Polyamide, polyvinylidene fluoride, nitrocellulose, polytetrafluoroethylene	5–50 mg/L	161.29 mg/g^a (Polyamide)	Neither very hydrophobic membrane nor very hydrophilic membrane was not effective for the retention of bisphenol A. And hydrophobicity and hydrophilicity effected to the adsorption capacity and the removal rate.	[48]

CA = cellulose acetate; CTA-ES = cellulose triacetate with embedded polyester screen support; C₀ = CEC initial concentration; GAC = granular activated carbon; MC = methylcellulose; NOM = natural organic matter; COD = chemical oxygen demand; PAC = powdered activated carbon; PCL = nanofiber membranes of poly(ɛ-carprolactone); PIM = polymer inclusion membranes; PPD = p-phenylenediamine; PSF = polysulfone; PVDF = polyvinylidene difluoride; SDW = synthetic drinking water; NSW = natural surface water; IWW; industrial wastewater; MWW = municipal wastewater; NGW = natural groundwater; SWW = synthetic wastewater; WWE: WW effluent;

maximum adsorption capacity.