Per- and polyfluoroalkyl substance separation by NF and RO membranes: a critical evaluation of advances and future perspectives

Abstract

Per- and polyfluoroalkyl substances (PFAS), dubbed “forever chemicals”, are synthetic compounds containing strong carbon–fluorine bonds. They are widely used in various industrial processes and products, and as a result, PFAS pollution is pervasive and has led to persistent contamination of surface and groundwater sources. Due to the adverse impact of PFAS exposure on health, there have been growing concerns among the public, the scientific community, and regulatory bodies, and treating water to an adequate level is essential. Nanofiltration (NF) and reverse osmosis (RO) are two of the candidate technologies for separating PFAS from water. NF and RO systems are easy to operate and require little use of chemicals. In contrast, other water treatment technologies (e.g., chemical oxidation, adsorption, ion exchange, and photocatalytic degradation) are often unsatisfactory due to slow reaction kinetics, generation and release of harmful by-products, or high operating costs. Despite the advantages of NF and RO, a concentrated residual stream is produced which contains high levels of PFAS. This concentrate, which typically accounts for 10 to 20% of the feedwater volume and is 5 to 10 times more concentrated with PFAS, must be managed or further treated appropriately to prevent environmental contamination. In this review, the NF/RO systems for the treatment of PFAS-contaminated water are discussed, focusing on the factors that affect their effectiveness and the mechanisms by which they remove PFAS. Also, advances in NF/RO membranes and systems as well as technical challenges at present are discussed along with an introduction to a total management plan for concentrated residual streams using a novel combination of NF/RO processes coupled with other state-of-the-art methods.

---

Water impact

PFAS are pervasive environmental pollutants, found in the far reaches of the Arctic to urban rainwater. This paper provides a critical review of NF and RO systems for the separation of PFAS from water. Advances in NF/RO membranes and the challenges to existing NF or RO are presented along with an introduction to a total management plan.

1. Introduction

Per- and polyfluoroalkyl substances (PFAS) are a group of synthetic organic compounds dubbed “forever chemicals” because of their inherent chemical stabilities.¹ These compounds have been in commercial production since the 1950s and are ubiquitously present in the environment.² PFAS are composed of a hydrophobic chain containing fluorinated alkyl on one end, and a hydrophilic functional group on the other end.³ Perfluoroalkyl substances may be represented by the chemical formula C_nF_2n+1–R, where C_nF_2n+1 defines the length of the perfluoroalkyl chain tail and R represents the functional group head attached to the chain. PFAS are known for their remarkable chemical and thermal stabilities, which are attributed to their strong carbon–fluorine bonds.^4–6 These properties, along with the lipophobic and hydrophobic groups, make PFAS ideal for oil and water repellants and friction resistant.^6,7 As a result, they have found a broad range of applications (e.g., fire-fighting foams, surfactants, and the production of cosmetics, packaging materials, and textiles).^6,8Fig. 1(A and B) present a detailed summary of PFAS manufacturing from source to consumer use where the quest for fluorine in PFAS starts from mining fluorite (CaF₂) mineral deposits¹ and ends up in the environment in the form of forever-growing PFAS contamination (Fig. 1C). Currently, there are more than 4000 PFAS compounds in use globally, comprising both conventional (e.g., perfluorooctane sulfonic acid or PFOS, and perfluorooctanoic acid or PFOA) and alternative PFAS (consisting of short-chain PFAS and newly recognized fluorinated substitutes for traditional PFAS).⁹

Fig. 1 (A) PFAS manufacturing; PFAS contain fluorine obtained from mineral deposits of fluorite. The mineral is processed through digestion to create hydrofluoric acid (HF).¹ (B) Venn diagram showing the target application areas of PFAS. (C) Schematic illustration of the distribution of PFAS in the environment through various sources, including industrial processes, recycling, and landfilling. These substances can move through the air, soil, water, and sediment compartments, leading to human exposure. PFAS can also enter the food chain through bioaccumulation. (D) The types of PFAS: non-polymeric, including PFOA and PFOS (I & II), which have a long tail and different polar head groups, and polymeric, including the fluoropolymer subgroup e.g., polytetrafluoroethylene (PTFE) also known as Teflon (III) and the perfluoropolyether subgroup e.g., Krytox also known as a lubricant (IV). Abbreviations: HCFO, hydrochlorofluoroolefin; HFO, hydrofluoroolefin; HFC, hydrofluorocarbon; HFE, hydrofluoroether; PASF, perfluoroalkanesulfonyl fluoride.

PFAS have been detected in drinking water sources across the globe at levels differing based on their source and location.¹⁰ For example, in the US, PFOA in groundwater ranges from 10 to 2305 ng L⁻¹, while in surface water from less than 5 to 821 ng L⁻¹. Similarly, PFOS concentrations in groundwater are between 4 and 59 ng L⁻¹, while in surface water PFOS ranges from less than 1 to 69 ng L⁻¹, respectively.¹¹ Because of the extensive prevalence and harmful health consequences of PFAS, lower acceptable PFAS limits for drinking water have been implemented or proposed globally.¹² In February 2023, the European Chemicals Agency (ECHA) published a proposal that could result in the largest-ever crackdown on their chemical production worldwide. The proposal was submitted by the environmental agencies of Denmark, Germany, the Netherlands, Norway, and Sweden to impose restrictions on the production of over 12 000 PFAS substances.¹³ The U.S. Environmental Protection Agency (U.S. EPA) also announced its plan in 2021, to regulate PFOA and PFOS under the Safe Drinking Water Act.¹⁴ Additionally, the agency has recently updated the Lifetime Health Advisory levels for several chemicals, including PFBS, HFPO-DA or GenX (a marketing name), PFOS, and PFOA, and the revised levels have been set at 2000, 10, 0.02, and 0.004 ng L⁻¹, respectively.¹⁵ Moreover, the European Union Water Framework Directive has suggested a maximum limit of 100 ng L⁻¹ for 20 PFAS, including PFOS, PFOA, and their shorter-chain homologs.¹⁶

The presence of PFAS in water, coupled with increased regulations from government agencies, has created a pressing need to find treatment processes capable of removing PFAS to reach acceptable levels to reduce their health risk to humans and aquatic life.^17,18 Analytical methods to quantify PFAS at such low levels are also challenging due to their complex chemistry and the introduction of new short-chain PFAS to the environment. At present, the U.S. EPA recommends two standardized techniques (known as U.S. EPA Methods 537.1 and 533) for testing drinking water samples for PFAS based on solid-phase extraction.¹⁹ The U.S. EPA method 533 complements method 537.1 by including an additional 11 compounds and excluding 4 compounds found in 537.1. Method 533 specifically targets PFAS compounds with shorter carbon chains, ranging from C4 to C12. The primary distinction between the two methods lies in the solid-phase extraction media employed. Method 533 utilizes polystyrene divinylbenzene with positively charged diamino ligands and isotope dilution, while method 537.1 utilizes styrene–divinylbenzene (SDVB) media.²⁰

Treating PFAS-contaminated water is challenging due to the strong carbon–fluorine bond.¹ Many attempts have been made to develop techniques for treating PFAS-contaminated water, including destructive techniques (e.g., chemical oxidation and reduction, ultrasonication, and photocatalytic degradation) and non-destructive techniques (e.g., adsorption, ion exchange, and membrane separation).²¹ Among them, advanced oxidation/reduction techniques based on electrochemical,²² photocatalytic,²³ or sonochemical²⁴ processes have been favored due to their potential to degrade the PFAS compounds. However, the slow reaction kinetics, generation and release of harmful by-products, and high operating cost (high chemical and energy consumption) have limited their use.²⁵ In addition to advanced oxidation/reduction, adsorption can also effectively remove PFAS from water but the requirement of high dosages of adsorbent and subsequent waste handling after the treatment are the main concerns. Thus, effective, low-cost sustainable treatment technologies for PFAS removal from water are still in pursuit.

Membrane filtration processes, in particular, nanofiltration (NF) and reverse osmosis (RO), are candidate technologies for separating PFAS from water due to their easy operation, high separation efficiency, and minimal chemical usage.²⁶ Both NF and RO processes have demonstrated promising results in removing PFAS from water by as much as 90 to 99%.^27,28 Typically, using NF and RO to remove PFAS from contaminated feedwater results in a concentrated retentate stream that is 5 to 10 times more concentrated with PFAS, with the concentrate volume being 10 to 20% of the feedwater.^17,29 The concentrate must be further treated or managed appropriately to prevent environmental contamination.¹² NF and RO processes require periodic system downtime for membrane cleaning or replacement, which affects the process productivity and operating costs.³⁰

This paper intends to critically review current and next-generation NF and RO processes for treating PFAS-contaminated water. A detailed discussion of the factors that affect the separation performance and the mechanisms by which PFAS are removed is also provided. Subsequently, current limitations or challenges encountered by NF and RO membranes when treating such feed streams are elaborated, followed by an introduction to recent advances in nano-enabled membranes and integrated membrane processes for enhanced PFAS removal. The last section of the article attempts to highlight how the technical limitations and needs may be addressed using state-of-the-art technologies and emerging techniques.

2. Physicochemical characteristics and environmental occurrence of PFAS

PFAS are soluble in water and consist of both hydrophobic fluorinated alkyl chains and hydrophilic functional groups (Fig. 1D).^1,31 The most found functional groups are carboxylic or sulfonic acid, which can dissociate into ionic forms in appropriate aqueous solutions. The resulting ions can be anions, cations, or zwitterions, depending on the specific functional group, with anions being the most prevalent in the environment (Table 1). These different ionic forms display distinct behaviors in environmental settings.³² The carbon–fluorine bond present in PFAS is exceptionally strong, due to the small size and high electronegativity of the fluorine atom, making it one of the strongest bonds in organic chemistry.^1,33 PFAS can exist as linear or branched isomers and can be classified based on their chain length.³⁴

Table 1 Physical and chemical properties of both short and long-chain PFAS

PFAS types	Names	Chemical formula	M W (g mol^−1)	V m (cm^3 mol^−1)	pKa	Vapor pressure (Pa)
Abbreviations: MW; molecular weight, Vm: molar volume, pKa; acidity constant, PFBA; perfluorobutanoic acid, PFPnA; perfluoropentanoic acid, PFBS; perfluorobutanesulfonic acid, PFHxA; perfluorohexanoic acid, PFHpA; perfluoroheptanoic acid, PFHxS; perfluorohexanesulfonic acid, PFOA; perfluorooctanoic acid, PFNA; perfluorononanoic acid, PFOS; perfluorooctanesulfonic acid, FOSA; perfluorooctance sulfonamide, PFDA; perfluorodecanoic acid, PFUnA; perfluoroundecanoic acid, PFDS; perfluorodecanesulfonic acid, PFDoA; perfluorododecanoic acid, PFTA; perfluorotetradecanoic acid. The data were obtained from ref. 35–38. The molar volume was calculated by dividing the MW by the density of the PFAS compound at 25 °C. Note: the pKa values might be reported slightly differently by other studies at 25 °C.
Short-chain PFAS	PFBA	CF3–(CF2)2–COO–	212	130	0.394	851
	PFPnA	CF3–(CF2)3–COO–	263	154	0.569	—
	PFBS	CF3–(CF2)3–SO3^−	299	163	0.14	32
	PFHxA	CF3–(CF2)4–COO–	313	178	0.84	240
	PFHpA	CF3–(CF2)5–COO–	363	213	−0.19	20.89
Long-chain PFAS	PFHxS	CF3–(CF2)5–SO3^−	399	217	0.14	—
	PFOA	CF3–(CF2)6–COO–	413	230	0.5	4.17
	PFNA	CF3–(CF2)7–COO–	463	258	2.575	1.29
	PFOS	CF3–(CF2)7–SO3^−	499	277	−3.3	3.3 × 10^−4
	FOSA	CF3–(CF2)7–SO2–NH2	499	277	6.52	—
	PFDA	CF3–(CF2)8–COO–	513	293	2.606	0.23
	PFUnA	CF3–(CF2)9–COO–	563	319	3.128	0.1
	PFDS	CF3–(CF2)9–SO3^−	599	333	0.14	—
	PFDoA	CF3–(CF2)10–COO–	613	347	−0.21	0.008
	PFTA	CF–(CF2)12–COO–	713	396	−0.21	270

---

PFAS are widely present in the environment and have been found in groundwater and surface water, wastewater, solid waste, and sediments.³² The release of PFAS into the environment occurs via two major sources: (1) point sources and (2) non-point sources. Point sources are specific locations such as industrial facilities, landfills, firefighting training grounds, and wastewater treatment plants (WWTPs). WWTPs are the most thoroughly investigated point sources for PFAS. The non-point sources, on the other hand, are diffuse and often originate from unknown sources or locations, e.g., atmospheric transport of volatile PFAS compounds, surface run-off, precipitation, and the breakdown of PFAS in consumer products.³⁹Fig. 1C is an illustration of the PFAS distribution in the environment through various sources, including industrial processes, recycling, and landfilling.

3. Overview of NF and RO processes

NF and RO utilize permselective membranes that selectively transport certain components more easily than others (Fig. 2).^40,41 It involves mass transport through the membrane under a hydraulic pressure gradient from the feed to the permeate side across the membrane.²⁵ The membrane allows certain molecules to pass through the membrane, resulting in a retentate stream that contains the rejected components. The efficiency of the NF/RO process relies on both the operating pressure and the properties of the membrane used for separation. RO membranes possess small pores ranging from 1 to 12 Å (Fig. 2III), which enable the separation of small ions from water. NF membranes have a looser structure (i.e., bigger pores) than RO membranes, and thus RO requires higher pressures to operate than NF, which increases energy costs. To obtain a significant transmembrane flux, operating pressures ranging from 10–100 bar are often needed, depending on the osmotic pressure of the feed mixture. Solution–diffusion is the primary mechanism for RO to separate compounds with different diffusivities and solubilities.⁴² The rejection characteristics of NF is often measured in terms of molecular weight cut-off (MWCO), and NF MWCO is typically in the range of 200 to 2000 Da, placing them somewhere between RO and ultrafiltration (UF) membranes.⁴³ The pore size/voids of NF fall between RO and UF (Fig. 2III) and usually operate at pressures ranging from 5 to 40 bar. Additionally, NF membranes may possess significant surface charges that can contribute to the separation of charged solutes and are mainly used to separate small compounds from water through size exclusion mechanisms.⁴⁴

Fig. 2 Schematic illustration of NF and RO membranes for water treatment: (I) nanofiltration (NF) and (II) reverse osmosis (RO). Each membrane has varying pore sizes (III) and capabilities for filtering out particles and microorganisms. RO membranes exclude particles and many low molar mass species like salt ions and organics. NF membranes are relatively new among the membrane processes and have pores smaller than UF membranes but larger than RO membranes, exhibiting performance between the range of UF and RO.⁴⁵

Many factors affect the separation performance of RO and NF membranes, including size exclusion, electrostatic repulsion, and solute–membrane affinity.⁴⁶ The solute–membrane affinity is influenced by the chemical and physical properties of both the solutes and the membrane materials, such as hydrophobicity, hydrogen bonding, and dipole interactions.⁴⁷ In addition, membrane fouling may become a major issue that limits the separation efficiency, depending on the specific feed mixtures to be separated. Membrane fouling happens when macromolecules or colloidal substances in the feed accumulate on the surface of the membrane, resulting in concentration polarization and the formation of a gel or cake layer if concentration polarization is significant enough. The solute physical and chemical interactions with the membrane, such as adsorption onto the pore walls and pore plugging, also contribute to membrane fouling. The degree of membrane fouling determines how often cleaning is required, as well as the cost, lifespan, and operation of the membrane plant.⁴⁴

4. Application of nanofiltration and reverse osmosis for PFAS separation

This section provides a detailed discussion of the NF and RO systems used for treating PFAS-contaminated water. It attempts to evaluate the effectiveness of NF and RO for removing PFAS and examine the factors that impact PFAS compound separation from water.

4.1. PFAS filtration mechanisms

The driving force for the permeation of water and PFAS through dense NF and RO membranes is the chemical potential gradient across the membrane. Thus, membrane separation of PFAS from water originates from the unparalleled migration of PFAS (slow) and water molecules (fast) through the membrane. According to the classical solution–diffusion (SD) model, the migration process involves the partitioning of PFAS/water molecules from the feed to the membrane surface, their diffusion through the membrane matrix, and the release of the permeant from the membrane to the permeate liquid phase. Each step can be affected by different factors. Collectively, several mechanisms may be attributed to the rejection of PFAS by virgin NF and RO membranes, including electrostatic exclusion, steric (size) exclusion, and the exclusion associated with hydrophobic solute–solute and solute–membrane interactions (Fig. 3).^35,44 Typically, large molecules are retained through the size exclusion mechanisms (Fig. 3A), particularly in RO. However, linear PFAS molecules may also penetrate through the pores of NF/RO membranes, depending on the molecular dimensions of PFAS and the pore size of the membrane. For instance, PFOS is a rigid linear PFAS molecule with a small cross-section due to the strength of its carbon–fluorine bonds and the lack of rotation at the carbon–carbon bonds, compared to non-fluorinated chemicals with aliphatic or aromatic rings.^44,48 Additionally, for uncharged PFAS, the steric exclusion mechanism is often dominant, and the rejection efficiency increases with a decrease in the MWCO of the membrane and an increase in the size of PFAS molecules or entities.

Fig. 3 NF/RO membranes for the removal of PFAS compounds involve various mechanisms. These include (A) size exclusion, which restricts the passage of PFAS molecules based on their size; (B) electrostatic interactions, which involve electrostatic repulsion between the negative membrane surface and the negatively charged PFAS molecules and electrostatic attraction of PFAS molecules to the charged membrane surface; and (C) hydrophobic interactions, which involve the affinity of PFAS molecules for the nonpolar regions of the membrane.

Some PFAS can exist in neutral or anionic form depending on the pH level of water (Table 1).³¹ This is important to the NF/RO-based treatment process because the electrostatic repulsion plays a crucial role in the separation of negatively charged solutes by likely-charged membranes. Modifying the surface and pore charges of the membrane can greatly alter PFAS removal through electrostatic repulsion (Fig. 3B). To achieve a higher solute rejection via electrostatic repulsion, it is plausible to increase the negative charge density of the membrane surface.⁴⁹ Hydrophobic interaction can also significantly contribute to the rejection of PFAS by NF/RO membranes (Fig. 3C). For instance, Wang et al. reported that PFAS molecules strongly adsorb onto natural organic matter (NOM) through hydrophobic interactions, resulting in simultaneous retention of both PFAS and NOM during membrane filtration.²⁸ PFAS can be adsorbed, accumulated, and retained on membrane surfaces during the filtration process due to the hydrophobic interactions between their uncharged per-fluorocarbon chains and the membrane. When the concentration of PFAS exceeds its critical micelle concentration (CMC), the hydrophobic interactions will lead to the formation of PFAS micelles, which favors their retention by the membrane. Recently, Mohona et al. have also reported that PFOS molecules form self-aggregates due to their preferential interactions when introduced to a hydrophobic surface. This can lead to the formation of a micellar structure that influences mass transport through the membrane.⁵⁰ In addition, the occurrence of non-columbic interactions between the membrane surface and PFAS molecules may also form a bilayer structure that favors PFAS rejection (Fig. 3C). When inorganic ions are present in water, additional factors will come into play, including reduction of the electrical double-layer, the bridging effect of divalent cations, neutralization of surface charge, salting-out, reduction in solubility, and competitive adsorption.^18,44

Numerous studies have shown that the rejection rates of PFOS and FOSA can be reduced by “foulant (cake)-enhanced concentration polarization (CP)”.^37,51 The fouling layer leads to a decrease in the back-diffusion of PFAS, resulting in lower PFAS rejection. A simplified model that considers cake-enhanced CP may be used to predict the rejection of fouled membranes. Membrane fouling is a complex phenomenon that involves different stages based on various mechanisms and can severely inhibit the performance of NF/RO systems. Therefore, improving the antifouling characteristics in NF/RO membranes via the control of material properties such as morphology, surface charge, and hydrophilicity can greatly enhance the performance and reduce the operating costs of these systems.

4.2. Nanofiltration for PFAS separation

NF has been extensively used in water and wastewater treatment for decades,⁴³ and for removing PFAS from water. These membranes have been proven to be highly effective to selectively filtrate small molecules while retaining larger ones, resulting in the nearly complete removal of a variety of PFAS.¹² However, the effectiveness of the NF membrane in removing PFAS depends on the chemical and physical characteristics of the individual PFAS compound as well.⁵² Therefore, for removing PFAS by NF, various factors such as hydrophobicity, functional groups, molecular weight/geometry, and pK_a need to be considered.⁵³ Other water quality parameters, including pH, co-existing ions, NOM, and temperature, along with membrane properties and operating conditions (e.g., pressure) can also influence the separation of PFAS by NF membranes.^2,25,49,54 For instance, NF can effectively separate PFOA from water through size exclusion and electrostatic interactions due to its high molecular weight (414 g mol⁻¹) and negative charge (low pK_a value of PFOA).⁵⁵ Several studies have reported the high retention of PFAS by commercially available NF membranes (Table 2). For example, Tang et al. studied the removal of PFOS compounds using three different NF membranes, and a removal rate ranging from 90% to 99% was achieved.⁵⁶ Another study by Zhao et al. evaluated the NF270 membrane for the removal of PFOS from simulated surface water containing calcium ions, and it was found that the membrane was able to remove up to 99% of PFOS, indicating the potential of NF for treating PFAS-contaminated water.⁵⁷ Liu et al. studied the rejection of 42 PFAS compounds in groundwater using the NF90 membrane, demonstrating that the rejection of PFAS was above 98% for almost all operating conditions and water matrices evaluated.⁵⁸ They reported that the rejection of PFAS by NF increases with increasing molecular weight (Fig. 4a). However, the NF process may not be as effective in removing short-chain PFAS compared to long-chain PFAS due to their small molecular size. Some studies also reported less than 80% rejection of short-chain anionic PFAS (e.g., perfluorobutanoic acid (PFBA), perfluoropentanoic acid (PFPeA) and perfluorohexanoic acid (PFHxA)) and neutral PFAS (e.g., perfluorooctane sulfonamide (FOSA)).^28,37

Table 2 Removal performance of commercially available NF and RO as reported in the literature

Membranes	Model	Membrane properties						Membrane performance			References
		Selective layer	MWCO^a (Da)	Pore size (nm)	Contact angle (°)	Isoelectric point (pH)	Zeta potential (mV)	Permeability (L m^−2 h^−1 bar^−1)	Salt rejection (%)	PFAS rejection (%)
Note: the data on membrane properties and performances were obtained from various sources.a Molecular weight cut-off.
Commercial NF	NF90	Polyamide	90–200	0.4	45.4–73.5	4.1	24.6 (at pH 4)	7.3–10.5	94.4 (NaCl)	PFHxA (72–99)	2, 37, 49, 54, 56, 59, 60, 61
										PFOA (90–99.3)
										PFBA (>98)
	NF270	Polyamide	155–300	0.3–0.4	14.4–32.6	4–5	−5.6 (at pH 4–5)	12.2–16.6	97.0 (MgSO4)	PFOA (90–99)	2, 54, 56
									30.0–58.0 (NaCl)
										PFHxA (>95)
									50.0–75.0 (CaCl2)	PFHpA (>95)
										PFNA (>95)
									90.0 (NaSO4)
										PFDA (>95)
										PFUnA (>95)
										PFTA (>95)
	DK	Polyamide	150–300	0.7	45.1	4–5	7.4 (at pH 4)	2.8	98.0 (MgSO4)	PFHxA (>95)	37, 56, 62, 63
									56.0–94.4 (NaCl)	PFHpA (>95)
										PFNA (>95)
										PFDA (>95)
										PFUnA (>95)
										PFTA (>95)
	DL	Polyamide	150–300	0.7	42.0–51.0	4–5	−58.0 (at pH 7)	4.0	96.0 (MgSO4)	PFHxA (>95)	37, 62, 64
									50.0 (NaCl)	PFHpA (>95)
										PFNA (>95)
										PFDA (>95)
										PFUnA (>95)
										PFTA (>95)
	NF200	Polyamide	200–1000	1.0–2.0	26.0–30.3	4–5	−16.5 (at pH 7)	4.2	97.0 (MgSO4)	98.5 (FOSA)	37, 63
									60.0 (NaCl)	PFHxA (>95)
										PFHpA (>95)
										PFNA (>95)
										PFDA (>95)
										PFUnA (>95)
										PFTA (>95)
	XN45	Polyamide	200	NA	57.0	NA	−25 (at pH 7)	10.8	62.0 (MgSO4)	PFOA (95)	65
									41.3 (NaCl)
	2540-ACM5-TSF	Polyamide	200	NA	NA	NA	NA	NA	98.5 (MgSO4)	PFOA (>99)	66
	HYDRACORE	Sulphonated polyethersulfone	1000	NA	NA	NA	−52.3 (at pH 7)	3.1	60.0 (MgSO4)	PFOA (55–86)	67
	NE70	Polyamide	350	0.8	45.6–54.1	4.0	−11.0 (at pH 7)	4.9	58.9 (NaCl)	PFOA (38)	68, 69
									98.9 (MgSO4)	PFOS (38)
	NTR-7450	Sulphonated polyethersulfone	600–800	NA	69.6	NA	−16.6 (at pH 7)	5.7–10.9	55.0 (NaCl)	PFHxA (96)	49, 70
	ESNA1-K1	Polyamide	286	0.5	24.0	NA	−32.0 (at pH 7)	NA	NA	PFOA (94–97)	71, 72
	NFG	Polyamide	150–800	2.1	31.0	NA	NA	12.3	NA	PFOS (49.0)	73
										PFOA (38.1)
										PFHxS (44.4)
										PFHxA (35.5)
										PFBS (48.4)
										PFBA (22.3)
Commercial RO	SG	Polyamide	100	0.1	72.9	NA	−12.3 (at pH 4)	2.3	95.2 (NaCl)	POFA (>99)	56, 74, 75
	LFC1	Polyamide	100	0.1	76.5	3.1	4.0 (at pH 4)	4.0	97.3 (NaCl)	POFA (>99)	56, 75, 76
	LFC3	Polyamide	<100	0.1	35.0	NA	3.0–3.8 (pH at 4)	2.8	98.5 (NaCl)	POFA (>99)	56, 75, 77
	BW30	Polyamide	98	0.1	43.7	NA	3.0–3.6 (at pH 4)	4.0	97.9 (NaCl)	POFA (>99)	2, 61, 73, 75
										PFBA (>99)
							−10.0 (at pH 7)
	ESPA3	Polyamide	NA	0.1	NA	NA	14.7 (at pH 4)	7.5	94.9 (NaCl)	POFA (>99)	56, 75
							15.0 (at pH 4)
	ESPA-2540	Polyamide	100	0.1	NA	pH < 4	NA	5.9	99.0 (NaCl)	PFAAs (>99)	59, 78
									99.0 (MgSO4)
	XLE	Polyamide	98	0.1	67.2	NA	−30.0 (at pH 4)	4.8	NA	PFHxA (>99)	2, 79

---

Fig. 4 Relationship between PFAS molecular weights and their rejection by (a) NF and (b) RO. Principal component analysis data of the intrinsic factors (e.g., PFAS molecular weight, membrane pore size) and extrinsic factors (e.g., pH value, initial concentration, co-cation concentration) for PFAS rejection by RO (c) and NF (d). The PCA plot displays the loadings (represented by arrows) and scores (represented by dots) of the experimental data. An arrow with a longer projected length on the axis of PC1 or PC2 indicates that the variable has a larger loading and a stronger relationship to PC1 or PC2. The dots represent data points taken from conditions. Fig. (a) was reproduced with permission from ref. 58, copyright 2021, Elsevier, and Fig. (b–d) were adopted and modified from ref. 61, copyright 2024, John Wiley & Sons.

Adsorption of PFAS compounds onto NF membranes can impact the system's performance. For example, Kwon et al. investigated the adsorption behavior of perfluorononanoic acid (PFNA) and PFOS on three different TFC membranes (including two NF membranes NF90 and NF270, and RO membrane BW30),⁸⁰ and found that the adsorption rates of PFNA and PFOS were dependent on the membrane type and water quality conditions. The semi-aromatic piperazine-based NF270 membrane showed higher adsorption of PFOS and PFNA than the fully-aromatic MPD-based membranes (NF90 and BW30). This is because of the specific chemistry of the piperazine-based polyamide, which provides more favorable interaction sites for PFAS adsorption. The BW30 membrane exhibited lower adsorption than NF90 because of the presence of aliphatic carbon and hydroxyl groups in BW30 that reduced its interactions with PFOS and PFNA, resulting in decreased adsorption. However, water quality parameters such as low pH conditions and higher ionic strength can exacerbate this phenomenon due to a decrease in electrostatic repulsion of the membrane surface and PFAS compounds. Similar results were also reported recently by Soriano et al. where NF270 and NF90 membranes showed lower rejection to PFAS than BW30 membranes at acidic pH and higher ionic concentration due to the charge-shielding effects, which reduce the electrostatic repulsion.^2,27

In addition to laboratory studies, several pilot-scale studies have also been conducted to evaluate the effectiveness of NF for PFAS removal. Among the different NF membranes, the NF270 membrane is widely studied as a commercial membrane for removing PFAS from water. It has shown promising results in removing most PFAS species from both simulated and natural water samples under environment-relevant pH conditions.^2,59,68 As mentioned earlier, it may not be as effective in removing short-chain PFAS with a molecular weight that is close to the MWCO of NF270 (which ranges from 155 to 300 Da).³⁷ For instance, one study found that the NF270 membrane only rejected 72% of PFPeA (with M_W of 263 Da), while the removal of the same membrane for other PFAS homologous with higher M_W was >95%. On the other hand, the NF90 membrane (molecular weight cut-off of 90 to 200 Da) showed higher rejections (94% to 99%) for PFPeA (MW 263 Da).⁸¹ The variation in PFPeA rejection between the NF270 and NF90 membranes can be attributed to the lower MWCO of NF90. Hence, selecting a suitable membrane for separating a particular PFAS compound is critical as it can greatly enhance throughput and reduce operating costs. Other commercially available NF membranes, including NTR7450, DK, NF200, ESNA1-K1, and 2540-ACM5-TSF, have been used for PFAS removal and have shown higher rejections (Table 2).⁸² Though NF has shown promising results in removing PFAS, there are still other issues to address, including membrane fouling and the management of concentrated residual streams after filtration.

4.3. Reverse osmosis for PFAS separation

RO driven by a high pressure has been used for water desalination for decades and has dominated the global market since the 1980s, owing to its exceptional performance.⁸³ RO has also been utilized to treat PFAS-contaminated water but has not been investigated as extensively as for NF membranes. The SD model has been widely used to describe the transport of salt and water in RO by assuming that water and salt diffuse through the membrane polymer matrix independently.⁸⁴ The transport of organic compounds in RO membranes may also be described by diffusion and advection through the membrane pores.⁴⁶ Several studies have reported on the use of RO for PFAS removal from water (Table 2). For instance, Tang et al. investigated the efficiency of four RO membranes (including LFC3, ESPA3, SG, and BW30) for the separation of PFOS from semiconductor wastewater. They found that all the membranes performed well in removing PFOS at different concentrations (from 0.5 to 1600 mg L⁻¹). At lower concentrations of 0.5 to 1 mg L⁻¹, all four membranes achieved over 99% retention of PFOS. As the PFOS concentrations increased, the retention ability of all the membranes improved but the flux decreased. Notably, the BW30 and LFC3 membranes demonstrated a remarkable 99.9% retention of PFOS at concentrations exceeding 100 mg L⁻¹.⁷⁵ As stated in section 4.1, when the concentration of PFAS exceeds the critical micelle concentration (CMC) in water, micelles will be formed that enhance its rejection through size exclusion. Similar results were also reported regarding the influence of the initial solute concentration on the retention of pharmaceutical compounds (e.g., diclofenac, phenacetin, and primidone) by RO membranes.⁸⁵ Large PFOS (with a molecular weight of 538 g mol⁻¹) can be removed through a size exclusion mechanism by loosely structured RO membranes at a higher permeability than dense RO membranes.⁷⁵Fig. 4b shows the relationship between molecular weights of PFAS and their rejection by the RO membrane. It is generally considered that RO membranes primarily remove organic compounds through size exclusion.⁸⁶ However, other factors may also play a role in the rejection of PFAS by RO, and therefore we must look into the different rejection mechanisms (e.g., electrostatic repulsion, size exclusion, and adsorption onto the membrane) relevant to the systems.

Electrostatic repulsion has been found to enhance the rejection of negatively charged organic compounds through membranes with negatively charged surfaces.⁸⁷ However, this may not always apply to PFOS rejection. For example, membranes used by Tang et al. showed varying solute retentions, with ESPA3 displaying the highest rejections, followed by BW30 and LFC3, and SG showing the lowest rejection.⁷⁵ Although both size exclusion and electrostatic repulsion are important mechanisms in removing PFOS by RO, the membrane performance can still be affected by the adsorption of the foulant. The same study also revealed that the membrane flux decreased as the PFOS concentration increased due to membrane fouling caused by PFOS adsorption onto the membrane surface.⁷⁵ This suggests that a given membrane with certain surface characteristics will respond differently to the retention of various PFAS compounds present in water.

The pilot-scale studies for PFAS removal using RO are currently rather limited as compared to the NF process. Thompson et al. conducted a study on the effectiveness of RO to remove PFOS from drinking water treatment plants in Australia. They found that PFOS was significantly removed from drinking water to reach a concentration level of several ng L⁻¹ in the permeate.⁸⁸ Later, similar results were also reported for studies in the US and Spain.^89,90 Recently, Pang et al. studied the effectiveness of RO membranes in removing contaminants of emerging concerns (CECs) and PFAS in a full-scale recycled water plant in Australia. They collected samples at different stages of the plant, and 44 PFAS compounds were found to be present in the feed, while only 8 PFAS compounds had concentrations greater than the method of deduction limit after membrane filtration treatment, indicating that RO membranes are effective in removing PFAS compounds.⁹¹ However, there is still a lack of sufficient research concerning the specific innovations required for handling the concentrated RO retentate. At present, large-scale applications of RO for PFAS removal from water are still limited due to the high costs related to installation, maintenance, operation, and management of the concentrate.^35,92 In light of the current regulatory restrictions and the health and environmental impacts of PFAS, further enhancement and modification of the RO membranes and systems for increased permeability to remove PFAS from drinking water appear to be a direction for future research.

4.4. Factors affecting the rejection of PFAS by NF/RO

4.4.1. Effect of water pH. The feed pH significantly impacts the efficiency of NF/RO membranes to remove PFAS from water.⁹³ PFAS molecules with fluorinated carbon chains attached to functional groups such as carboxylic or sulfonic acid groups (e.g., PFOA and PFOS) are negatively charged in neutral pH water. The electrostatic repulsion between these molecules and the membrane surface having negative charges can enhance their rejection. However, at low pH values, these functional groups can become protonated, which will potentially decrease their rejection by the negatively surface-charged membrane.⁹⁴ Many PFAS compounds exhibit low pK_a values across a broad range of pH. Therefore, the surface charge of the membrane plays a crucial role in removing these compounds from water.³⁵ The degree to which negatively charged membranes retain anionic PFOA molecules may be greatly influenced by the pH of the solution. For example, Hang et al. investigated the effects of feed pH (ranging from 4 to 10) on the transmembrane pressure required for a given flux and the PFOA rejection with NF. They found that at a low pH (4), PFOA molecules adsorbed onto the membrane surface, leading to membrane fouling, which would need a higher transmembrane pressure to achieve the same flux. In contrast, at higher pH conditions (from 6 to 10), both PFOA and the membrane were negatively charged, and the electrostatic repulsion helped prevent the adsorption of PFOA onto the membrane.⁵⁴ However, the study found that PFOA retention at pH 10 was significantly lower than at other pH values, presumably due to pore enlargement of the membrane. Another study by Lu and Wan also found that the functional groups of the polymer chains completely dissociated at a high pH, resulting in the repulsion between negatively charged polymer chains of the membrane. This, in turn, led to the opening of membrane pores and thus a lower rejection.⁹⁵

Several other studies have shown that removing PFAS from water through NF and RO with a variety of membranes is highly dependent on the solution pH.^2,49,54,60 For example, Ma et al. carried out a principal component (PC) analysis to understand the relation between different factors and their impacts on PFAS rejection by NF and RO (Fig. 4c and d). They found that increasing the pH value or decreasing the membrane zeta potential could enhance the PFAS rejection rate due to the improved electrostatic exclusion.⁶¹ Soriano et al. demonstrated that RO membranes with more negatively charged surfaces (e.g., the XLE membrane) exhibited better rejection of PFHxA than those membranes with less negatively charged surfaces under neutral pH conditions.² Most TFC membranes have an active surface layer that carries negative charges at a neutral pH of water, which minimizes PFAS adsorption onto the membrane surface. Increasing the pH of the water results in electrostatic exclusion enhancement between PFAS and the membrane, thus improving PFAS removal.⁵¹ According to Steinle-Darling and Reinhard, the effective MWCO of polyamide membranes can be altered by the pH of water. They reported that the 90% rejection cut-off point of the NF270 membrane was below 300 g mol⁻¹ at pH 5–6, while the same membrane at pH 2.8 showed a cut-off point between 500 and 550 g mol⁻¹. This suggests that a change in the feed pH can alter the MWCO of the membranes to some extent.³⁷ This may be attributed to the hydrolysis of carboxyl functional groups on the membrane surface, which expands the PA layer and loosens the membrane pores, causing a decrease in PFSA rejection.⁵⁴ Recently, membranes with sulfonate groups (–SO₃H) (e.g., NTR-7450 membrane) have exhibited exceptional PFAS rejection (>94.0%) in a wide range of pH (3.3 to 10) due to the changes in membrane structure with pH.⁴⁹ Therefore, efforts to control the surface chemistry and fabricate membranes that can adapt to pH changes (like pH-responsive membranes) would be highly desirable for separating PFAS from water under various pH conditions.

4.4.2. Effect of co-existing ions. The presence of co-existing ions can have varying effects on PFAS rejection by NF or RO membranes, and there are discrepancies in the results reported. There are reports that certain co-existing cations (including Fe³⁺, Pb²⁺, Mg²⁺, Ca²⁺, and Na⁺) can improve PFAS removal by the membrane, particularly cations of high valence and at high concentrations.^{51,57,71,96,97} For example, a density functional theory (DFT) analysis by Zhao et al. showed that divalent ions (e.g., Mg²⁺) can form a chemical bond with the negative functional groups of PFOS through electrostatic interactions,⁹⁶ and this enhances the steric hindrance, which results in improved rejection.^57,96 Additionally, cations can act as a bridge between two PFOS molecules to form complexes, which increases the molecular size of the PFOS species and promotes the sieving mechanism of membranes.⁹⁶ However, Soriano et al. also found that the cations may neutralize the negative surface charges of NF/RO membranes, which causes a reduction in the electrostatic shielding effect. This can lead to a decrease in PFAS removal by hindering electrostatic exclusion.⁹⁸

As one may expect, the presence of co-existing anions such as Cl⁻, SO₄²⁻, and PO₄³⁻ will enhance PFAS retention. It should be noted, however, that prior research is mainly concentrated on the effects of various anions on the rejection of PFOS but not all PFAS compounds. Among the co-existing anions studied, PO₄³⁻ has been found to have the most significant influence on PFOS rejection due to its highest electronegativity, which promotes the electrostatic repulsion between PFOS molecules and the membrane surface.^57,71,99 Therefore, the presence of anions should be considered very carefully when designing a NF/RO process for PFAS removal. The types and concentrations of the co-existing anions in the feed water can affect the performance and efficiency of the membrane system. Optimal operating conditions and suitable membranes should be carefully considered to maximize the separation of PFAS while minimizing the impact of co- and counter-ions.

4.4.3. Effect of natural organic matter (NOM). The presence of natural organic matter (NOM) in water can influence the removal of PFAS by affecting both the target compounds and membrane properties. It has been shown that NOM may form a layer on the membrane surface during filtration, which can thus alter the surface chemistry and morphology of the membrane.^28,96,100 Wang et al. found that PFAS and NOM tend to be adsorbed to each other due to their hydrophobic interactions, leading to a great molecular size, which hinders their ability to pass through the membrane.²⁸ In addition, the fouling layer formed by NOM on the membrane surface can enhance the negative surface charge on the membrane surface, which can lead to high electrostatic repulsion to PFAS.¹⁰⁰

However, the effects of NOM on PFAS rejection may vary, depending on both the type of NOM present in water and the chemical structure of the PFAS compound to be removed. For instance, a recent study by Liu et al. showed that membrane fouling due to sodium alginate (SA), bovine serum albumin (BSA), and humic acid (HA) have different effects on the rejection of PFAS by NF. It was found that SA formed a dense and smooth fouling layer that reduced water flux but increased the transmission of PFAS. The dense fouling layer also intensified the concentration polarization which prevented reverse solute diffusion and led to increased PFAS transmission. On the other hand, membrane fouling by BSA and HA was found to increase the PFAS rejection by NF due to the looser and thicker fouling layer. The latter enhanced steric hindrance and electrostatic repulsion, resulting in a higher PFAS rejection. However, the effects of the presence and types of NOM were less significant for PFAS with longer perfluorocarbon chains and higher molecular weights. The steric hindrance and hydrophobic interactions were found to be important to PFAS rejection by NF membranes.¹⁰¹ Wang et al. also reported that SA and BSA present in the feed had different effects on the rejection of PFOS (C8, containing eight perfluorinated carbon atoms) and PFBS (C4, containing four perfluorinated carbon atoms) from water.²⁸ Both SA and BSA increased PFOS (C8) rejection. The rejection was found to be dominated by adsorption before fouling occurred, whereas size exclusion played a key role in the rejection of PFOS (C8) after the membrane was considerably fouled. On the other hand, the removal of PFBS (C4) was not consistently affected by BSA, and the primary retention mechanism for PFBS (C4) was electrostatic repulsion.²⁸ In addition, Zhao et al. found that the presence of HA at concentrations ranging from 5 to 20 mg L⁻¹ did not increase PFOS rejection. Interestingly, the co-existence of HA with Mg²⁺ increased PFOS rejection from 94% to 98%, indicating that the bridging effects facilitated by the presence of Mg²⁺ could contribute to the improvement of PFOS removal.⁹⁶ The removal of PFAS compounds from water is a complex process due to their complicated chemistry and relation to various water quality parameters. Therefore, further research is necessary to better understand the interactions in order to optimize the PFAS removal processes.

5. Advances in NF/RO technologies for PFAS separation

Despite the promising results obtained with NF/RO for PFAS rejection, challenges remain from the viewpoint of engineering applications. Therefore, research efforts to explore the various approaches to improving the performance of NF/RO-based treatment of PFAS-contaminated water are needed. Two major technical routes may be distinguished in these efforts, i.e., the development of better membranes and the integration of NF/RO with other processes. Some notable advances in both aspects are highlighted below.

5.1. Better NF/RO membranes

Most studies reported previously have been focused on enhanced removal of long-chain PFAS such as PFOA and PFOS by NF/RO. The main strategy was the modification of membrane structures to promote electrostatic and/or size exclusion of PFAS. According to Nadagouda and Lee, increasing the density of carboxyl groups in the selective layer of a polyamide membrane can enhance the partitioning/sorption of negatively charged PFAS compounds on the membrane, thereby improving the PFAS rejection.¹⁰² In another study, MXene was embedded into the polyamide selective layer of an NF membrane, which resulted in increased water flux and greater rejection for PFOS. The enhanced removal of PFOS was primarily promoted by two factors: electrostatic repulsion and size selectivity, resulting from the negative surface charge of MXene and the enhanced extended nodular structure of PA, respectively when MXene nanosheets were incorporated into the PA layer during the IP process.¹⁰³ In another study, Meragawi et al. prepared a polyethyleneimine (PEI) functionalized GO membrane, which effectively removed PFOA (400 Da) from water with improved permeability. The GO-PEI membrane was able to reject 96.5% of PFOA and had a high water permeance of 15.9 L m⁻² h⁻¹ bar⁻¹. The PEI, which is rich in electrons, deoxygenates GO, leading to a smaller interlayer spacing and increased surface hydrophilicity. These properties resulted in increased PFOA retention through steric hindrance and enhanced water permeation.¹⁰⁴ In addition, anti-scaling NF/RO membranes have been developed for the treatment of PFAS-contaminated water. For instance, Boo et al. prepared a loose and negatively charged NF membrane, which showed a higher rejection of PFOA than a commercial NF membrane with a substantially reduced scaling potential.¹⁰⁵ Ma et al. prepared MXene-functionalized NF membrane with adjustable membrane surface morphology and charge, and short-chain PFAS were rejected effectively by the nano-modified membrane due to increased membrane hydrophilicity and charge density.¹⁰⁶

Prior studies on PFAS removal have been mainly concentrated on PFOA and PFOS treatment. Additional PFAS compounds (e.g., short-chain compounds) that are being used in industries and consumer products could also add to the PFAS levels in drinking water and make the treatment process more difficult. Therefore, further research is needed to design advanced membranes for the separation of short-chain PFAS from water due to limited studies on the topic available in the literature.

5.2. Hybrid processes for PFAS removal

A major drawback in NF/RO processes for PFAS removal is that PFAS compounds are not destroyed and they end up in a concentrated waste stream (retentate). A single-stage NF/RO can recover a significant amount of water while concentrating PFAS compounds in the retentate (Fig. 5). If the retentate stream is not properly managed, it will be a potential secondary source of PFAS for PFAS to release to the environment again. Therefore, it must be managed or treated appropriately to prevent further environmental contamination. To address this issue, hybrid processes incorporating different technologies appear to be a potential solution for the total management of NF/RO concentrated residual streams. This may involve integrating NF/RO systems with other innovative techniques to achieve complete remediation of PFAS-contaminated water, for example, A) coupling NF/RO with a defluorination process, or B) coupling NF/RO with an additional concentration technology (Fig. 5).

Fig. 5 A novel combined approach for treating PFAS-contaminated water. Due to the strong C–F bond, it is challenging to fully mineralize PFAS, but NF/RO membranes can concentrate them effectively. To manage the residual stream, a combined approach of membrane and other methods should be developed. Typically, NF/RO membranes recover 80–90% of water, as illustrated in the figure. The remaining 10–20% concentrate should be treated with coupled techniques, which can recover a substantial amount of water. Sequestration of the PFAS concentrate can be a cost-effective management strategy, but it is a temporary solution because PFAS cannot be naturally degraded, and it will reintroduce PFAS back into the environment if not fully recovered.¹

A. Coupling NF/RO with defluorination processes. For the destruction of PFAS, the combination of NF/RO with other processes such as electrochemical oxidation, photocatalytic degradation, plasma treatment, or other destructive processes may be utilized.¹⁷ In a recent study, Li et al. used plasma treatment with the addition of a cationic surfactant (cetrimonium bromide, CTAB) for the degradation of PFAS in the RO concentrate. It was found that the addition of CTAB significantly enhanced the removal of short- and long-chain PFAS in RO-rejected water below a detection limit, except for PFBA which had a removal rate of 93%. The removal of ultrashort-chain PFAAs in the RO concentrate was less effective in the presence of CTAB compared to lab-prepared water, with a removal rate ranging from 29% to over 64%. The lower removal of ultrashort-chain PFAS can be attributed to the role of water conductivity. For instance, a high solution conductivity favors the transport of the CTAB–PFBS complexes to the interface but reduces the efficiency of ultrashort-chain PFAAs due to competitive interactions with chloride ions.¹⁰⁷ Soriano et al. studied the destruction of perfluorocaproic acid (PFHxA) in the concentrated stream generated from NF treatment of an industrial process water. It was found that the presence of sulfate and the high conductivity of the NF concentrate stream facilitated the electrochemical oxidation of PFHxA.¹⁰⁸ The combination of NF with photocatalytic oxidation was also exploited for the removal of PFOA from the concentrate produced during nanofiltration of groundwater.¹⁰⁹ These studies have shown that coupling the NF/RO membrane filtration systems with other defluorination technologies can significantly minimize PFAS concentration. While some chemical oxidation techniques can be effective in destroying PFAS, it is important to consider the potential by-products produced by the reaction process (e.g., short-chain PFAS, which exhibit similar toxicity to their long-chain counterparts).^110–112

B. Coupling NF/RO with another concentration process. To further concentrate the retentate stream from the membrane processes, other separation techniques such as adsorption, electrocoagulation, and foam fractionation can be coupled with the NF/RO systems.¹⁷ Recently, McCleaf et al. carried out pilot-scale investigations into the coupling of NF with foam fractionation (FF) for the treatment of AFFF contaminated groundwater. The NF process removed 98% of PFAS, yielding a permeate water stream with a concentration of 1.4 ng L⁻¹, a decrease from 77 ng L⁻¹. Afterward, FF was used to further concentrate the NF retentate at a concentration of 3000 ng L⁻¹, which removed 94% of short-chain PFAS when cationic co-surfactants were used in the foam fractionation process. In addition, adsorption and anion exchange (AIX) can also be used to treat the membrane concentrated stream. For example, Franke et al. used GAC and AIX to remove PFAS from the NF concentrated stream, resulting in a 2.6 and 4.1-fold enrichment compared with the raw water, respectively.⁷² These studies have shown that additional concentration technologies can significantly reduce the volume of the membrane concentrated stream. However, it is essential to understand that such additional concertation processes will only further enrich the concentrated waste but not convert PFAS into harmless products. If not recovered for reuse, the enriched PFAS stream still needs to be subjected to destruction or long-term sequestration. Note that the post-NF/RO concentration process represents additional investment and treatment costs. Therefore, further pilot or full-scale studies are necessary to validate the cost-effectiveness and technological readiness of integrated processes for PFAS pollution control.

6. Challenges and prospects

• NF and RO membranes have been shown effective in removing both short- and long-chain PFAS from water. However, the PFAS rejection is primarily based on size exclusion, while other factors are also important to the final performance of the membrane processes. The role of membrane chemistry, including membrane materials and surface properties, has not been thoroughly studied in the context of PFAS removal. Therefore, additional research is needed to investigate the overall impact comprehensively.

• Commercially available membranes have achieved outstanding results. However, the exploration of novel polymeric materials and innovative designs can help expand the potential range of membrane properties that can be controlled for enhanced PFAS separation. Membrane preparation conditions must also be optimized, considering the emerging nanotechnology approaches, to fine-tune the physicochemical properties of membranes (e.g., pore size, hydrophilicity, surface roughness and charge) to improve the permselectivity of NF/RO systems.

• To treat PFAS-contaminated water, it is worthwhile to explore other emerging membrane technologies, including forward osmosis (FO) and membrane distillation (MD), which are currently less used due to their performance, cost, and lack of technology maturity. New applications of these membrane processes may be unlocked for treating PFAS-contaminated water and improving the overall separation efficiency.

• As mentioned above, RO and NF are effective in removing PFAS and other dissolved species from water, but they do not destroy PFAS compounds. This leaves a concentrated retentate stream, which if not fully recovered for reuse, will require proper management either through long-term sequestration or destructive defluorination.

• The combination of NF/RO with other techniques such as electrochemical advanced oxidation, photocatalytic degradation, or foam fractionation can significantly contribute to the management of the concentrated retentate stream from the membrane process (shown in Fig. 4).

• If the PFAS in the retentate stream from the membrane process is not recovered for reuse, a technology that can destroy PFAS would be an ideal option to minimize ecological or human health impacts caused by the potential reintroduction of PFAS into the environment.

• Each process component should be evaluated based on its potential efficacy in removing PFAS, the formation of any breakdown products, and the impacts of background water constituents on the technology applicability.

• Effective PFAS removal will likely require a combination of different complementary treatment technologies and approaches, including source reduction, treatment, and disposal.

7. Conclusions

PFAS have been detected in water, and this poses a risk to human health and aquatic life. Traditional water treatment processes are often ineffective due to the recalcitrant nature of PFAS, leading to the need for effective advanced treatment approaches. NF/RO membranes have emerged as a promising technology for PFAS removal, offering excellent retention of both short- and long-chain PFAS. Understanding the removal mechanisms of various PFAS in the advanced treatment processes is essential, given the challenges posed by the physicochemical characteristics of PFAS. However, NF/RO systems produce concentrated PFAS streams that need to be managed appropriately. As PFAS is not destroyed in the membrane processes, if the PFAS in the concentrate stream is not recovered for reuse, there is a potential for the PFAS to go back into water sources. To address this issue, hybrid processes combining NF/RO with other techniques (e.g., electrochemical advanced oxidation, photocatalytic degradation, or foam fractionation) can be employed in order to effectively manage the concentrated PFAS stream from the membrane process. A good option would be a technology that can destroy PFAS in the concentrate stream, which would allow for the reintroduction of the water in the concentrate stream into the water supply without concerns of ecological or human health impacts.

Conflicts of interest

The authors declare no competing interest.

Acknowledgements

The authors acknowledge the research support received from the Natural Sciences and Engineering Research Council (NSERC) of Canada. The first author (SA) acknowledges the Vanier Canada Graduate Scholarship, funded by the Government of Canada.

References

1. M. G. Evich, M. J. B. Davis, J. P. McCord, B. Acrey, J. A. Awkerman, D. R. U. Knappe, A. B. Lindstrom, T. F. Speth, C. Tebes-Stevens, M. J. Strynar, Z. Y. Wang, E. J. Weber, W. M. Henderson and J. W. Washington, Per- and polyfluoroalkyl substances in the environment, Science, 2022, 375(5680) DOI:10.1126/science.abg9065.
2. Á. Soriano, D. Gorri and A. Urtiaga, Selection of High Flux Membrane for the Effective Removal of Short-Chain Perfluorocarboxylic Acids, Ind. Eng. Chem. Res., 2019, 58, 3329–3338.
3. O. S. Arvaniti and A. S. Stasinakis, Review on the occurrence, fate and removal of perfluorinated compounds during wastewater treatment, Sci. Total Environ., 2015, 524–525, 81–92.
4. P. Meng, X. Jiang, B. Wang, J. Huang, Y. Wang, G. Yu, I. T. Cousins and S. Deng, Role of the air-water interface in removing perfluoroalkyl acids from drinking water by activated carbon treatment, J. Hazard. Mater., 2020, 386, 121981.
5. V. Mulabagal, L. Liu, J. Qi, C. Wilson and J. S. Hayworth, A rapid UHPLC-MS/MS method for simultaneous quantitation of 23 perfluoroalkyl substances (PFAS) in estuarine water, Talanta, 2018, 190, 95–102.
6. B. Trang, Y. L. Li, X. S. Xue, M. Ateia, K. N. Houk and W. R. Dichtel, Low-temperature mineralization of perfluorocarboxylic acids, Science, 2022, 377(6608), 839–845.
7. C. F. Kwiatkowski, D. Q. Andrews, L. S. Birnbaum, T. A. Bruton, J. C. DeWitt, D. R. U. Knappe, M. V. Maffini, M. F. Miller, K. E. Pelch, A. Reade, A. Soehl, X. Trier, M. Venier, C. C. Wagner, Z. Y. Wang and A. Blum, Scientific Basis for Managing PFAS as a Chemical Class, Environ. Sci. Technol. Lett., 2020, 7, 532–543.
8. J. Li, B. D. Xi, G. H. Zhu, Y. Yuan, W. J. Liu, Y. Gong and W. B. Tan, A critical review of the occurrence, fate and treatment of per- and polyfluoroalkyl substances (PFASs) in landfills, Environ. Res., 2023, 218, 9.
9. S. P. Lenka, M. Kah and L. P. Padhye, A review of the occurrence, transformation, and removal of poly- and perfluoroalkyl substances (PFAS) in wastewater treatment plants, Water Res., 2021, 199, 22.
10. H. R. Li, A. L. Junker, J. Y. Wen, L. Ahrens, M. Sillanpaa, J. Y. Tian, F. Y. Cui, L. Vergeynst and Z. S. Wei, A recent overview of per- and polyfluoroalkyl substances (PFAS) removal by functional framework materials, Chem. Eng. J., 2023, 452, 14.
11. B. C. Crone, T. F. Speth, D. G. Wahman, S. J. Smith, G. Abulikemu, E. J. Kleiner and J. G. Pressman, Occurrence of per- and polyfluoroalkyl substances (PFAS) in source water and their treatment in drinking water, Crit. Rev. Environ. Sci. Technol., 2019, 49, 2359–2396.
12. P. McCleaf, W. Stefansson and L. Ahrens, Drinking water nanofiltration with concentrate foam fractionation – a novel approach for removal of per- and polyfluoroalkyl substances (PFAS), Water Res., 2023, 232, 119688.
13. X. Z. Lim, Could the world go PFAS-free? Proposal to ban “Forever Chemicals”, fuel debate, Nature, 2023, 620, 24–27.
14. U.S. EPA (U.S. Environmental Protection Agency), Announcement of final regulatory determinations for contaminants on the fourth drinking water contaminant candidate list, Document number: 2021–04184, 2021, pp. 12272–12291.
15. U.S. EPA (U.S. Environmental Protection Agency), Technical Fact Sheet: Drinking Water Health Advisories for Four PFAS (PFOA, PFOS, GenX chemicals, and PFBS), EPA 822-F-22-002, Office of Water, 2022.
16. Z. Y. Wang, A. Alinezhad, S. Nason, F. Xiao and J. J. Pignatello, Enhancement of per- and polyfluoroalkyl substances removal from water by pyrogenic carbons: Tailoring carbon surface chemistry and pore properties, Water Res., 2023, 229, 10.
17. E. W. Tow, M. S. Ersan, S. Kum, T. Lee, T. F. Speth, C. Owen, C. Bellona, M. N. Nadagouda, A. M. Mikelonis, P. Westerhoff, C. Mysore, V. S. Frenkel, V. deSilva, W. S. Walker, A. K. Safulko and D. A. Ladner, Managing and treating per- and polyfluoroalkyl substances (PFAS) in membrane concentrates, AWWA Water Sci, 2021, 3, 1–23.
18. D. Banks, B.-M. Jun, J. Heo, N. Her, C. M. Park and Y. Yoon, Selected advanced water treatment technologies for perfluoroalkyl and polyfluoroalkyl substances: A review, Sep. Purif. Technol., 2020, 231, 115929.
19. L. J. Winchell, M. J. M. Wells, J. J. Ross, X. Fonoll, J. W. Norton, S. Kuplicki, M. Khan and K. Y. Bell, Analyses of per- and polyfluoroalkyl substances (PFAS) through the urban water cycle: Toward achieving an integrated analytical workflow across aqueous, solid, and gaseous matrices in water and wastewater treatment, Sci. Total Environ., 2021, 774, 15.
20. Y. F. Wang, J. H. Kim, C. H. Huang, G. L. Hawkins, K. Li, Y. S. Chen and Q. G. Huang, Occurrence of per- and polyfluoroalkyl substances in water: a review, Environ. Sci.: Water Res. Technol., 2022, 8, 1136–1151.
21. S. Yin and D. Villagran, Design of nanomaterials for the removal of per- and poly-fluoroalkyl substances (PFAS) in water: Strategies, mechanisms, challenges, and opportunities, Sci. Total Environ., 2022, 831, 15.
22. T. X. H. Le, H. Haflich, A. D. Shah and B. P. Chaplin, Energy-Efficient Electrochemical Oxidation of Perfluoroalkyl Substances Using a Ti4O7 Reactive Electrochemical Membrane Anode, Environ. Sci. Technol. Lett., 2019, 6, 504–510.
23. S. Verma, B. Mezgebe, E. Sahle-Demessie and M. N. Nadagouda, Photooxidative decomposition and defluorination of perfluorooctanoic acid (PFOA) using an innovative technology of UV-vis/Zn(x)Cu(1-x)Fe(2)O(4)/oxalic acid, Chemosphere, 2021, 280, 130660.
24. V. L. Gole, R. Sierra-Alvarez, H. Peng, J. P. Giesy, P. Deymier and M. Keswani, Sono-chemical treatment of per- and poly-fluoroalkyl compounds in aqueous film-forming foams by use of a large-scale multi-transducer dual-frequency based acoustic reactor, Ultrason. Sonochem., 2018, 45, 213–222.
25. T. Lee, T. F. Speth and M. N. Nadagouda, High-pressure membrane filtration processes for separation of Per- and polyfluoroalkyl substances (PFAS), Chem. Eng. J., 2022, 431, 134023.
26. S. Ali, S. A. U. Rehman, H. Y. Luan, M. U. Farid and H. Huang, Challenges and opportunities in functional carbon nanotubes for membrane-based water treatment and desalination, Sci. Total Environ., 2019, 646, 1126–1139.
27. A. Soriano, C. Schaefer and A. Urtiaga, Enhanced treatment of perfluoroalkyl acids in groundwater by membrane separation and electrochemical oxidation, Chem. Eng. J. Adv., 2020, 4, 100042.
28. J. X. Wang, L. Wang, C. Q. Xu, R. Zhi, R. Miao, T. Liang, X. L. Yue, Y. T. Lv and T. T. Liu, Perfluorooctane sulfonate and perfluorobutane sulfonate removal from water by nanofiltration membrane: The roles of solute concentration, ionic strength, and macromolecular organic foulants, Chem. Eng. J., 2018, 332, 787–797.
29. V. Franke, M. Ullberg, P. McCleaf, M. Wålinder, S. J. Köhler and L. Ahrens, The Price of Really Clean Water: Combining Nanofiltration with Granular Activated Carbon and Anion Exchange Resins for the Removal of Per- And Polyfluoralkyl Substances (PFASs) in Drinking Water Production, ACS ES&T Water, 2021, 1, 782–795.
30. C. C. Murray, H. Vatankhah, C. A. McDonough, A. Nickerson, T. T. Hedtke, T. Y. Cath, C. P. Higgins and C. L. Bellona, Removal of per- and polyfluoroalkyl substances using super-fine powder activated carbon and ceramic membrane filtration, J. Hazard. Mater., 2019, 366, 160–168.
31. S. Rayne and K. Forest, Perfluoroalkyl sulfonic and carboxylic acids: a critical review of physicochemical properties, levels and patterns in waters and wastewaters, and treatment methods, J. Environ. Sci. Health, Part A: Toxic/Hazard. Subst. Environ. Eng., 2009, 44, 1145–1199.
32. J. N. Meegoda, J. A. Kewalramani, B. Li and R. W. Marsh, A Review of the Applications, Environmental Release, and Remediation Technologies of Per- and Polyfluoroalkyl Substances, Int. J. Environ. Res. Public Health, 2020, 17, 8117.
33. P. A. Rice, J. Cooper, S. E. Koh-Fallet and S. V. Kabadi, Comparative analysis of the physicochemical, toxicokinetic, and toxicological properties of ether-PFAS, Toxicol. Appl. Pharmacol., 2021, 422, 115531.
34. OECD, Reconciling Terminology of the Universe of Per-and Polyfluoroalkyl Substances: Recommendations and Practical Guidance, ENV/CBC/MONO, 2021, vol. 25.
35. T. Jin, M. Peydayesh and R. Mezzenga, Membrane-based technologies for per- and poly-fluoroalkyl substances (PFASs) removal from water: Removal mechanisms, applications, challenges and perspectives, Environ. Int., 2021, 157, 106876.
36. M. F. Rahman, S. Peldszus and W. B. Anderson, Behaviour and fate of perfluoroalkyl and polyfluoroalkyl substances (PFASs) in drinking water treatment: a review, Water Res., 2014, 50, 318–340.
37. E. Steinle-Darling and M. Reinhard, Nanofiltration for trace organic contaminant removal: Structure, solution, and membrane fouling effects on the rejection of perfluorochemicals, Environ. Sci. Technol., 2008, 42, 5292–5297.
38. G. Ding and W. J. G. M. Peijnenburg, Physicochemical Properties and Aquatic Toxicity of Poly- and Perfluorinated Compounds, Crit. Rev. Environ. Sci. Technol., 2013, 6, 598–678.
39. S. Kurwadkar, J. Dane, S. R. Kanel, M. N. Nadagouda, R. W. Cawdrey, B. Ambade, G. C. Struckhoff and R. Wilkin, Per- and polyfluoroalkyl substances in water and wastewater: A critical review of their global occurrence and distribution, Sci. Total Environ., 2022, 809, 19.
40. M. M. Pendergast and E. M. V. Hoek, A review of water treatment membrane nanotechnologies, Energy Environ. Sci., 2011, 4, 1946–1971.
41. M. Elimelech and W. A. Phillip, The Future of Seawater Desalination: Energy, Technology, and the Environment, Science, 2011, 333, 712–717.
42. F. E. Ahmed, R. Hashaikeh and N. Hilal, Hybrid technologies: The future of energy efficient desalination - A review, Desalination, 2020, 495, 114659.
43. K. P. Wang, X. M. Wang, B. Januszewski, Y. L. Liu, D. Y. Li, R. Y. Fu, M. Elimelech and X. Huang, Tailored design of nanofiltration membranes for water treatment based on synthesis-property-performance relationships, Chem. Soc. Rev., 2022, 51, 672–719.
44. T. F. Mastropietro, R. Bruno, E. Pardo and D. Armentano, Reverse osmosis and nanofiltration membranes for highly efficient PFASs removal: overview, challenges and future perspectives, Dalton Trans., 2021, 50, 5398–5410.
45. R. W. Baker, in Membrane Technology and Applications, 2004, Wiley, pp. 1–14.
46. C. Bellona, J. E. Drewes, P. Xu and G. Amy, Factors affecting the rejection of organic solutes during NF/RO treatment - a literature review, Water Res., 2004, 38, 2795–2809.
47. C. Bellona, M. Marts and J. E. Drewes, The effect of organic membrane fouling on the properties and rejection characteristics of nanofiltration membranes, Sep. Purif. Technol., 2010, 74, 44–54.
48. K. L. Tu, L. D. Nghiem and A. R. Chivas, Coupling effects of feed solution pH and ionic strength on the rejection of boron by NF/RO membranes, Chem. Eng. J., 2011, 168, 700–706.
49. C. H. Zeng, S. Tanaka, Y. Suzuki and S. Fujii, Impact of feed water pH and membrane material on nanofiltration of perfluorohexanoic acid in aqueous solution, Chemosphere, 2017, 183, 599–604.
50. T. M. Mohona, Z. Ye, N. Dai and P. C. Nalam, Adsorption behavior of long-chain perfluoroalkyl substances on hydrophobic surface: A combined molecular characterization and simulation study, Water Res., 2023, 239, 120074.
51. T. Wang, C. W. Zhao, P. Li, Y. Li and J. Wang, Fabrication of novel poly(m-phenylene isophthalamide) hollow fiber nanofiltration membrane for effective removal of trace amount perfluorooctane sulfonate from water, J. Membr. Sci., 2015, 477, 74–85.
52. A. Garg, N. P. Shetti, S. Basu, M. N. Nadagouda and T. M. Aminabhavi, Treatment technologies for removal of per- and polyfluoroalkyl substances (PFAS) in biosolids, Chem. Eng. J., 2023, 453, 18.
53. F. Leniz-Pizarro, R. J. Vogler, P. Sandman, N. Harris, L. E. Ormsbee, C. Q. Liu and D. Bhattacharyya, Dual-Functional Nanofiltration and Adsorptive Membranes for PFAS and Organics Separation from Water, ACS ES&T Water, 2022, 2, 863–872.
54. X. Hang, X. Chen, J. Luo, W. Cao and Y. Wan, Removal and recovery of perfluorooctanoate from wastewater by nanofiltration, Sep. Purif. Technol., 2015, 145, 120–129.
55. K.-U. Goss, The pKa Values of PFOA and Other Highly Fluorinated Carboxylic Acids, Environ. Sci. Technol., 2008, 42, 56–458.
56. C. Y. Tang, Q. S. Fu, C. S. Criddle and J. O. Leckie, Effect of flux (transmembrane pressure) and membrane properties on fouling and rejection of reverse osmosis and nanofiltration membranes treating perfluorooctane sulfonate containing wastewater, Environ. Sci. Technol., 2007, 41, 2008–2014.
57. C. W. Zhao, J. Zhang, G. Z. He, T. Wang, D. Y. Hou and Z. K. Luan, Perfluorooctane sulfonate removal by nanofiltration membrane the role of calcium ions, Chem. Eng. J., 2013, 233, 224–232.
58. C. J. Liu, T. J. Strathmann and C. Bellona, Rejection of per- and polyfluoroalkyl substances (PFASs) in aqueous film-forming foam by high-pressure membranes, Water Res., 2021, 188, 116546.
59. C. J. Liu, T. J. Strathmann and C. Bellona, Rejection of per- and polyfluoroalkyl substances (PFASs) in aqueous film-forming foam by high-pressure membranes, Water Res., 2021, 188, 9.
60. C. Boo, Y. Wang, I. Zucker, Y. Choo, C. O. Osuji and M. Elimelech, High Performance Nanofiltration Membrane for Effective Removal of Perfluoroalkyl Substances at High Water Recovery, Environ. Sci. Technol., 2018, 52, 7279–7288.
61. Q. Q. Ma, Q. Lei, F. Z. Liu, Z. M. Song, B. Khusid and W. Zhang, Evaluation of commercial nanofiltration and reverse osmosis membrane filtration to remove per-and polyfluoroalkyl substances (PFAS): Effects of transmembrane pressures and water matrices, Water Environ. Res., 2024, 96(2), e10983.
62. A. Al-Amoudi, P. Williams, A. S. Al-Hobaib and R. W. Lovitt, Cleaning results of new and fouled nanofiltration membrane characterized by contact angle, updated DSPM, flux and salts rejection, Appl. Surf. Sci., 2008, 254, 3983–3992.
63. A. L. Ahmad, L. S. Tan and S. R. Abd Shukor, The role of pH in nanofiltration of atrazine and dimethoate from aqueous solution, J. Hazard. Mater., 2008, 154, 633–638.
64. B. Su, T. Wang, Z. Wang, X. Gao and C. Gao, Preparation and performance of dynamic layer-by-layer PDADMAC/PSS nanofiltration membrane, J. Membr. Sci., 2012, 423–424, 324–331.
65. R. Rattanaoudom and C. Visvanathan, Removal of PFOA by hybrid membrane filtration using PAC and hydrotalcite, Desalin. Water Treat., 2011, 32, 262–270.
66. A. Boonya-Atichart, S. K. Boontanon and N. Boontanon, Study of hybrid membrane filtration and photocatalysis for removal of perfluorooctanoic acid (PFOA) in groundwater, Water Sci. Technol., 2018, 2017, 561–569.
67. Y. Yu, C. Zhao, L. Yu, P. Li, T. Wang and Y. Xu, Removal of perfluorooctane sulfonates from water by a hybrid coagulation–nanofiltration process, Chem. Eng. J., 2016, 289, 7–16.
68. A. H. M. Anwar Sadmani and H. Toure, Nanofiltration of perfluorooctanoic acid and perfluorooctane sulfonic acid as a function of water matrix properties, Water Sci. Technol.: Water Supply, 2019, 19, 2199–2205.
69. B.-M. Jun, H. K. Lee and Y.-N. Kwon, Acid-catalyzed hydrolysis of semi-aromatic polyamide NF membrane and its application to water softening and antibiotics enrichment, Chem. Eng. J., 2018, 332, 419–430.
70. K. Boussu, B. Van der Bruggen, A. Volodin, C. Van Haesendonck, J. A. Delcour, P. Van der Meeren and C. Vandecasteele, Characterization of commercial nanofiltration membranes and comparison with self-made polyethersulfone membranes, Desalination, 2006, 191, 245–253.
71. C. W. Zhao, G. S. Hu, D. Y. Hou, L. Yu, Y. Zhao, J. Wang, A. X. Cao and Y. H. Zhai, Study on the effects of cations and anions on the removal of perfluorooctane sulphonate by nanofiltration membrane, Sep. Purif. Technol., 2018, 202, 385–396.
72. V. Franke, P. McCleaf, K. Lindegren and L. Ahrens, Efficient removal of per- and polyfluoroalkyl substances (PFASs) in drinking water treatment: nanofiltration combined with active carbon or anion exchange, Environ. Sci.: Water Res. Technol., 2019, 5, 1836–1843.
73. M. Li, F. Sun, W. Shang, X. Zhang, W. Dong, Z. Dong and S. Zhao, Removal mechanisms of perfluorinated compounds (PFCs) by nanofiltration: Roles of membrane-contaminant interactions, Chem. Eng. J., 2021, 406, 126814.
74. A. F. S. Foureaux, E. Reis, Y. Lebron, V. Moreira, L. Santos, M. Amaral and L. Lange, Rejection of pharmaceutical compounds from surface water by nanofiltration and reverse osmosis, Sep. Purif. Technol., 2019, 212, 171–179.
75. C. Y. Tang, Q. S. Fu, A. P. Robertson, C. S. Criddle and J. O. Leckie, Use of reverse osmosis membranes to remove perfluorooctane sulfonate (PFOS) from semiconductor wastewater, Environ. Sci. Technol., 2006, 40, 7343–7349.
76. Y.-N. Kwon and J. O. Leckie, Hypochlorite degradation of crosslinked polyamide membranes: I. Changes in chemical/morphological properties, J. Membr. Sci., 2006, 283, 21–26.
77. T. Fujioka, S. J. Khan, J. A. McDonald, A. Roux, Y. Poussade, J. E. Drewes and L. D. Nghiem, N-nitrosamine rejection by nanofiltration and reverse osmosis membranes: The importance of membrane characteristics, Desalination, 2013, 316, 67–75.
78. C. Conidi, F. Macedonio, P. Argurio, A. Cassano and E. Drioli, Performance of reverse osmosis membranes in the treatment of flue-gas desulfurization (FGD) wastewaters, Environments, 2018, 5, 71.
79. T. Y. Haan, M. S. H. Ghani and A. W. Mohammad, Physical and chemical cleaning for nanofiltration/reverse osmosis (NF/RO) membranes in treatment of tertiary palm oil mill effluent (POME) for water reclamation, J. Kejuruteraan, 2018, 24, 51–58.
80. Y.-N. Kwon, K. Shih, C. Tang and J. O. Leckie, Adsorption of perfluorinated compounds on thin-film composite polyamide membranes, J. Appl. Polym. Sci., 2012, 124, 1042–1049.
81. E. Steinle-Darling and M. Reinhard, Nanofiltration for trace organic contaminant removal: Structure, solution, and membrane fouling effects on the rejection of perfluorochemicals, Environ. Sci. Technol., 2008, 42, 5292–5297.
82. C. H. Liu, X. Q. Zhao, A. F. Faria, K. Y. D. Quinones, C. H. Zhang, Q. He, J. Ma, Y. Shen and Y. Zhi, Evaluating the efficiency of nanofiltration and reverse osmosis membrane processes for the removal of per- and polyfluoroalkyl substances from water: A critical review, Sep. Purif. Technol., 2022, 302, 14.
83. Z. Yang, Y. Zhou, Z. Y. Feng, X. B. Rui, T. Zhang and Z. E. Zhang, A Review on Reverse Osmosis and Nanofiltration Membranes for Water Purification, Polymers, 2019, 11(8), 1252.
84. L. Wang, T. C. Cao, J. E. Dykstra, S. Porada, P. M. Biesheuvel and M. Elimelech, Salt and Water Transport in Reverse Osmosis Membranes: Beyond the Solution-Diffusion Model, Environ. Sci. Technol., 2021, 55, 16665–16675.
85. K. Kimura, G. Amy, J. E. Drewes, T. Heberer, T.-U. Kim and Y. Watanabe, Rejection of organic micropollutants (disinfection by-products, endocrine disrupting compounds, and pharmaceutically active compounds) by NF/RO membranes, J. Membr. Sci., 2003, 227, 113–121.
86. Y. Yoon and R. M. Lueptow, Removal of organic contaminants by RO and NF membranes, J. Membr. Sci., 2005, 261, 76–86.
87. Y. Yoon, P. Westerhoff, S. A. Snyder and E. C. Wert, Nanofiltration and ultrafiltration of endocrine disrupting compounds, pharmaceuticals and personal care products, J. Membr. Sci., 2006, 270, 88–100.
88. J. Thompson, G. Eaglesham, J. Reungoat, Y. Poussade, M. Bartkow, M. Lawrence and J. F. Mueller, Removal of PFOS, PFOA and other perfluoroalkyl acids at water reclamation plants in South East Queensland Australia, Chemosphere, 2011, 82, 9–17.
89. T. D. Appleman, C. P. Higgins, O. Quinones, B. J. Vanderford, C. Kolstad, J. C. Zeigler-Holady and E. R. Dickenson, Treatment of poly- and perfluoroalkyl substances in U.S. full-scale water treatment systems, Water Res., 2014, 51, 246–255.
90. C. Flores, F. Ventura, J. Martin-Alonso and J. Caixach, Occurrence of perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA) in NE Spanish surface waters and their removal in a drinking water treatment plant that combines conventional and advanced treatments in parallel lines, Sci. Total Environ., 2013, 461, 618–626.
91. H. J. Pang, M. Allinson, K. Northcott, A. Schultz and P. J. Scales, Demonstrating removal credits for contaminants of emerging concern in recycled water through a reverse osmosis barrier-A predictive framework, Water Res., 2023, 244, 120427.
92. N. J. Herkert, J. Merrill, C. Peters, D. Bollinger, S. Zhang, K. Hoffman, P. L. Ferguson, D. R. U. Knappe and H. M. Stapleton, Assessing the Effectiveness of Point-of-Use Residential Drinking Water Filters for Perfluoroalkyl Substances (PFASs), Environ. Sci. Technol. Lett., 2020, 7, 178–184.
93. C. Y. Tang, Q. S. Fu, D. W. Gao, C. S. Criddle and J. O. Leckie, Effect of solution chemistry on the adsorption of perfluorooctane sulfonate onto mineral surfaces, Water Res., 2010, 44, 2654–2662.
94. A. Soriano, D. Gorri, L. T. Biegler and A. Urtiaga, An optimization model for the treatment of perfluorocarboxylic acids considering membrane preconcentration and BDD electrooxidation, Water Res., 2019, 164, 114954.
95. J. Q. Luo and Y. H. Wan, Effects of pH and salt on nanofiltration-a critical review, J. Membr. Sci., 2013, 438, 18–28.
96. C. W. Zhao, C. Y. Tang, P. Li, P. Adrian and G. S. Hu, Perfluorooctane sulfonate removal by nanofiltration membrane-the effect and interaction of magnesium ion/humic acid, J. Membr. Sci., 2016, 503, 31–41.
97. C. W. Zhao, T. Zhang, G. S. Hu, J. Ma, R. P. Song and J. D. Li, Efficient removal of perfluorooctane sulphonate by nanofiltration: Insights into the effect and mechanism of coexisting inorganic ions and humic acid, J. Membr. Sci., 2020, 610, 118176.
98. A. Soriano, D. Gorri and A. Urtiaga, Efficient treatment of perfluorohexanoic acid by nanofiltration followed by electrochemical degradation of the NF concentrate, Water Res., 2017, 112, 147–156.
99. Q. Luo, Y. X. Liu, G. X. Liu and C. W. Zhao, Preparation, characterization and performance of poly(m-phenylene isophthalamide)/organically modified montmorillonite nanocomposite membranes in removal of perfluorooctane sulfonate, J. Environ. Sci., 2016, 46, 126–133.
100. P. J. Zhao, X. H. Xia, J. W. Dong, N. Xia, X. M. Jiang, Y. Li and Y. M. Zhu, Short- and long-chain perfluoroalkyl substances in the water, suspended particulate matter, and surface sediment of a turbid river, Sci. Total Environ., 2016, 568, 57–65.
101. C. Liu, Y. Shen, X. Zhao, Z. Chen, R. Gao, Q. Zuo, Q. He, J. Ma and Y. Zhi, Removal of per- and polyfluoroalkyl substances by nanofiltration: Effect of molecular structure and coexisting natural organic matter, J. Hazard. Mater., 2023, 454, 131438.
102. M. N. Nadagouda and T. Lee, Cross-Flow Treatment of PFAS in Water: Materials Challenges and Potential Solutions, Acc. Mater. Res., 2021, 2, 129–133.
103. T. Le, E. Jamshidi, M. Beidaghi and M. R. Esfahani, Functionalized-MXene Thin-Film Nanocomposite Hollow Fiber Membranes for Enhanced PFAS Removal from Water, ACS Appl. Mater. Interfaces, 2022, 14, 25397–25408.
104. S. El Meragawi, A. Akbari, S. Hernandez, M. S. Mirshekarloo, D. Bhattacharyya, A. Tanksale and M. Majumder, Enhanced permselective separation of per-fluorooctanoic acid in graphene oxide membranes by a simple PEI modification, J. Mater. Chem. A, 2020, 8, 24800–24811.
105. C. Boo, Y. K. Wang, I. Zucker, Y. Choo, C. O. Osuji and M. Elimelech, High Performance Nanofiltration Membrane for Effective Removal of Perfluoroalkyl Substances at High Water Recovery, Environ. Sci. Technol., 2018, 52, 7279–7288.
106. J. Ma, Y. Y. Wang, H. Xu, M. M. Ding and L. Gao, MXene (Ti3T2CX)-reinforced thin-film polyamide nanofiltration membrane for short-chain perfluorinated compounds removal, Process Saf. Environ. Prot., 2022, 168, 275–284.
107. R. Li, O. F. Isowamwen, K. C. Ross, T. M. Holsen and S. M. Thagard, PFAS-CTAB Complexation and Its Role on the Removal of PFAS from a Lab-Prepared Water and a Reverse Osmosis Reject Water Using a Plasma Reactor, Environ. Sci. Technol., 2023, 57, 12901–12910.
108. A. Soriano, D. Gorri and A. Urtiaga, Efficient treatment of perfluorohexanoic acid by nanofiltration followed by electrochemical degradation of the NF concentrate, Water Res., 2017, 112, 147–156.
109. A. Boonya-Atichart, S. K. Boontanon and N. Boontanon, Study of hybrid membrane filtration and photocatalysis for removal of perfluorooctanoic acid (PFOA) in groundwater, Water Sci. Technol., 2018, 2017, 561–569.
110. J. N. Meegoda, J. A. Kewalramani, B. Li and R. W. Marsh, A Review of the Applications, Environmental Release, and Remediation Technologies of Per- and Polyfluoroalkyl Substances, Int. J. Environ. Res. Public Health, 2020, 17(21) DOI:10.3390/ijerph17218117.
111. A. P. Trzcinski and K. Harada, Adsorption of PFOS onto graphite intercalated compound and analysis of degradation by-products during electro-chemical oxidation, Chemosphere, 2023, 323, 138268.
112. T. A. Bruton and D. L. Sedlak, Treatment of Aqueous Film-Forming Foam by Heat-Activated Persulfate Under Conditions Representative of In Situ Chemical Oxidation, Environ. Sci. Technol., 2017, 51(23), 13878–13885.

---