A critical review on recent polymeric and nano-enhanced membranes for reverse osmosis

Abstract

In this paper, current and recent advances in polymeric and nano-enhanced membrane development for reverse osmosis have been reported in terms of membrane performance and fouling. Graphene, zeolites, carbon nanotubes, silica, silver, and titanium dioxide are the predominantly tested nanoparticles in current and recent investigations. Membranes from graphene, zeolites, and carbon nanotubes have all been shown to enhance membrane water permeability. Silica has been observed to exhibit high affinity for water and improve the hydrophilicity of RO membranes. Silver and titanium dioxide have strong antimicrobial properties and can be included in RO membranes to reduce biofouling. However, the use of nanomembranes for commercial and industrial RO applications is still under development as their scalability is still a challenge. Polymeric membranes, such as cellulose acetate and polyamide, and their integration with other polymers or nanoparticles have also been presented in this paper. Overall, the choice of membrane materials for future RO processes would depend largely on the required permselectivity and the targeted foulants. However, membrane performance and antifouling features would have to be taken into consideration for sustainability of the type of RO membrane desired for a specific application.

---

1. Introduction

Over the last few years, the market share of reverse osmosis (RO) has witnessed continued domination in desalination when compared with other technologies.¹ RO tends to be the most preferred choice because it is simple and requires relatively low specific energy consumption; therefore, more than half of the total installed desalination plants in the world today use RO technology.² Progress in RO desalination has also been directed towards RO membrane synthesis in recent times. It seems RO is indispensable for today’s pure water production and many membrane production firms have sprung up lately. Although commercial RO membranes are known for their high salt rejection rates currently,³ it is worth noting here that fresh water scarcity is still prevalent and many people still do not have access to usable water as a result of natural and anthropogenic factors.^4,5 The RO market is currently a multibillion dollar market and a huge amount of money is being spent yearly on the research and development of new membrane materials and technology in order to reduce the fresh water supply shortfall.² Advancements in RO membrane materials in terms of performance and fouling resistance are necessary to further improve the contribution of RO to the world’s pure water production. The currently available RO membrane materials still suffer from fouling and regular replacement, leading to a high lifecycle cost of the RO process.^6,7 Therefore, materials that can maintain high salt rejection, high water flux and productivity, and low vulnerability to fouling are needed to produce RO membranes at low cost so that fresh water can be easily made available. Improved technologies are needed for the fabrication, testing, and commercial applications of these RO membranes so that enhanced performance can be guaranteed.

RO is a pressure driven process that permits the flow of water through a semipermeable membrane by a solution-diffusion mechanism.^1,8,9 Because high pressure is needed to stimulate water diffusion and subdue the osmotic pressure of the saline feed, the properties of RO membranes are quite important as factors that significantly determine the process performance and cost. Recently, some state-of-the-art RO membranes have been prepared from nanomaterials, polymers and their combinations in order to overcome the challenges of limited water flux, salt rejection and the propensity for membrane fouling. The most recent articles on RO membrane materials have presented novel fabrication methods, improved membrane development conditions, newer composites or combinations of materials to integrate their competitive advantages, advanced membrane structural modification methods, and challenges and/or drawbacks of the latest RO membranes. Nanomaterials are dominating the current wave of novel membrane material development because of their inherent specific physicochemical features that make them suitable for the RO process.^10,11 Buonomenna¹² has indicated that the use of nano-enhanced RO membranes for desalination has become attractive recently as a result of the tendency of nanomaterials to improve the flux of fresh water through the membranes and reduce the energy required for transmembrane transport. Buonomenna¹² emphasized that cost savings might be achieved from nano-enhanced RO membranes as a result of reduced energy consumption, the antifouling characteristics of some of these membranes, and the requirement for lower quantities of cleaning chemicals for fouling control. These benefits would translate to the reduction of the environmental impact associated with RO desalination. Although the commercially feasible use of nanomembranes for practical large-scale RO desalination is yet to be achieved, the commercial scale-up of effective small-scale nanomembrane systems would be attained if the performance improvements that have been recorded recently are sustained and can be reproduced in real-life desalination.

Graphene, zeolites, carbon nanotubes, silica, silver, and titanium dioxide are the predominantly tested nanoparticles in current and recent investigations. Membranes from graphene, zeolites, and carbon nanotubes have all been shown to enhance membrane water permeability.^13–15 Silica has been shown to exhibit high affinity for water and improve the hydrophilicity of RO membranes.¹⁶ Silver and TiO₂ have strong antimicrobial properties and can be used to develop RO membranes that are less susceptible to biofouling.^17,18 However, the use of nanomembranes for commercial and industrial RO applications is still under development and scaling membranes from nanoparticles is still a challenge.^10,19

Lately, some advances in polymeric membrane development have been reported in the literature. The integration of commonly used polymeric materials such as cellulose acetate and polyamide with other polymers or nanoparticles has ensured an improvement in the membrane performance. Therefore, this review presents the most recent RO membranes, their fabrication, properties, applications, and performance characteristics. The benefits and drawbacks of these membranes are also explained. The modification of the membrane surface features through material selection has also been presented. The impact of the operating conditions on the performance of recently developed RO membranes and state-of-the-art methods that have been devised to curtail membrane fouling are also discussed in this review.

2. Nano-enhanced membranes and polymer nanocomposites

2.1 Graphene and graphene oxide

Today, graphene could be an interesting ultra-permeable membrane for RO due to its atomic thickness and its precise sieving properties.^13,20–26 However, there is limited information regarding the ideal conditions for the use of graphene in terms of the operating pressure in the literature.²⁷ Meanwhile, the water permeability of nanoporous graphene (NPG) under different pressures has been studied for the RO process.²⁷ A molecular simulation through visual molecular dynamics (VMD) was applied to see if real-world RO pressures (1000–2000 bars) could have an impact on the ultrahigh water permeability of NPG. The data showed that NPG maintained its ultrahigh permeability at high pressures. A permeability of 1041 ± 20 L m⁻² h⁻¹ bar⁻¹ was achieved for NPG, assuming a nanopore density of 1.7 × 10¹³ cm⁻². This work proved that NPG could be a high-performing membrane for RO desalination. However, this is only a theoretical study; an experimental investigation would be required to confirm the simulation results obtained. On the other hand, the successes recorded from the incorporation of graphene oxide (GO) into polymeric membranes for water permeability improvements have continued lately. Some recent research has been directed towards the incorporation of GO and poly(allylamine hydrochloride) (PAH) via electrostatic interactions on a porous poly(acrylonitrile) support in a layer-by-layer assembly.²⁸ Characterization techniques revealed that the mass of GO was 2–5 times higher than that of PAH in each GO–PAH bilayer of 16.5 nm thickness. The transport of water and selected solutes were studied for this membrane. It was shown that the water permeability of the GO membrane was about one order of magnitude higher than that of the commercial polymeric membrane. The GO membrane retained a tight structure in solutions of low ionic strength and exhibited a high rejection of sucrose (∼99%) with a membrane weight cutoff of around 342 or an equivalent pore size of around 1 nm. However, this membrane was only well suited for applications in the pressure retarded osmosis (PRO) or forward osmosis (FO) modes. The performance of different kinds of RO membranes for FO desalination has been discussed elsewhere.²⁹

Safarpour et al. has also prepared a novel thin film nanocomposite (TFN) RO membrane via interfacial polymerization (IP) of m-phenylenediamine (MPD) and trimesoyl chloride (TMC) monomers, and introduced a reduced hydrophilic graphene oxide–titanium dioxide (rGO–TiO₂) nanocomposite in the membrane’s active PA layer to enhance the membrane water flux and salt rejection.²⁰ The rGO–TiO₂ nanocomposite was prepared by employing a hydrothermal method, and was then distributed into the MPD solution before being embedded into the PA layer. The three-dimensional atomic force microscopy (AFM) images showed that the modified membranes had a rough surface (Fig. 1). It was observed from contact angle measurements that the hydrophilicity of the membranes could be increased by increasing the concentration of rGO–TiO₂ because of the existence of several negatively charged and hydrophilic functional groups on the rGO–TiO₂ nanocomposite material. The best performance for the modified membranes was observed using 0.02 wt% rGO–TiO₂ to obtain a water flux of 51.3 L m⁻² h⁻¹ (LMH) and salt rejection of 99.45%. A GO thin film composite (TFC) membrane has also been prepared through fractionation and dispersing in a solution of MPD.³⁰ The fabricated membrane was then characterized and changes in the hydrophilicity, surface charge, surface roughness and thickness were noticed. It was observed that the GO embedding enhanced the water permeability and anti-biofouling property by 80 and 98%, respectively. Furthermore, with a chlorine concentration at 48 000 ppm, a high salt rejection was still retained. Overall, the GO layer enhanced the performance of the TFC membrane.

Fig. 1 Three-dimensional images of rGO–TiO₂ enhanced TFN membranes.²⁰

A rGO–TiO₂ nanocomposite has also been integrated into the active polyamide (PA) layer of PA TFC RO membranes in order to impart antifouling properties to the membranes. Filtration of bovine serum albumin (BSA) solution was used to test the antifouling and chlorine-resistant properties of the developed membranes.²⁰ The best antifouling ability was demonstrated by the RO membrane with 0.02 wt% rGO–TiO₂. Moreover, the chlorine tests revealed that the salt rejection declined by 30% for the pristine TFC RO membranes, and declined by only 3% for a TFC RO membrane integrated with 0.02 wt% rGO–TiO₂. It was concluded that the developed nanocomposite has better antifouling and chlorine-resistant efficiency when compared with the pristine PA TFC membranes.

2.2 Zeolites

Zeolites as nanomaterials have also been used for RO membrane development. Zeolites are chemically stable and have the necessary characteristics required for rejecting ions.^10,31–34 In order to prevent the expensive pretreatment required for current polymeric RO membranes, Zhu et al. have recently proposed the implementation of mordenite framework inverted (MFI)-type zeolite membranes for recycled saline wastewater desalination.³² A collective method of rubbing and secondary hydrothermal growth was employed to develop the MFI-type zeolite membrane on a tubular α-alumina substrate. The zeolite membrane demonstrated a water flux of 4 LMH and a salt rejection of 80% when tested at a pressure of 7 MPa, utilizing recycled wastewater from a local treatment plant situated in Melbourne, Australia as the feed. The zeolite membranes also attained a rejection of 90% or higher for divalent calcium and magnesium ions, and a rejection greater than 70% for monovalent potassium and sodium ions when tested at 21 °C with applied pressures of 3 MPa and above. The modified membranes also eliminated more than 90% of the organic compounds comprising protein-like and aromatic substances at 21 °C with applied pressures of 3 MPa and above. Batch concentration runs with saline wastewater were conducted for a test period of 48 hours, and the results demonstrated a water recovery of 43%. Also, a marginally lower salt rejection was observed when the zeolite membrane was tested for RO performance using saline wastewater as the feed. Additionally, the water flux remained approximately at 2 LMH during the test period of 48 hours, which indicated the resistance of the zeolite membrane to organic fouling. Moreover, the long-term chlorine stability tests conducted for 7 days using strong hypochlorite cleaning (chlorine exposure of 168 000 mg L⁻¹ h) revealed that the zeolite membranes did not significantly retard the water flux or salt rejection, thereby proving the outstanding chemical resilience of the zeolite membranes. Hence, easy biofouling control and cleaning techniques were also observed. However, there is a need to enhance the water flux and salt rejection so that the proposed zeolite membranes can compete with the current polymer RO membranes to achieve water of higher purity. Zeolites may be affected by the pH of treatment. The chemical resistance and acid stability of Linde Type A (LTA) zeolites during the treatment of synthetic low pH wastewater brine prepared by the United States’ National Aeronautics and Space Administration (NASA) has also been investigated in order to evaluate the effectiveness of LTA zeolites in producing potable water.³⁵ It was observed that the LTA zeolites disintegrated the synthetic wastewater brine quickly. The effects of monobasic potassium phosphate, sulfuric acid, chromium(VI) oxide and monobasic sodium phosphate dihydrate, which are acid producing components present in wastewater brine, were studied precisely to identify the acidic components that caused degradation. The impact of the acids on the surface morphology, bonding, solid phase composition and crystal structure of the LTA zeolites was examined. It was observed that the zeolites only incurred some damage after 24 hours when exposed to the synthetic wastewater brine at a pH of 1.94. However, the LTA zeolites were degraded extensively after 128 hours of exposure to synthetic wastewater brine at a pH of 1.94. The LTA particles’ crystal structure was damaged completely by all acidic solutions and phosphate solutions with pH values below 1 and 5, respectively. It was observed that the elimination of aluminum from the framework of the zeolite caused the structural degradation. Moreover, phosphate removed the silicon from zeolite framework and formed precipitates of aluminum phosphate, thereby altering the zeolite bonding. Furthermore, it was revealed that the acidic anions, rather than the pH, played the most important role in damaging the LTA structures present in the solution containing dihydrogen phosphate. In conclusion, the experimental results indicated that the LTA zeolites were not suitable for applications in low pH environments, and especially in the presence of phosphate.

The integration of zeolites into a PA membrane has also resulted in an impressive membrane performance. For example, TFN membranes incorporated with NaY zeolite nanoparticles have been fabricated on polysulfone supports by IP of MPD and TMC.³¹ These zeolite nanoparticles were characterized as having cubical shapes with a particle size range of 100–200 nm (Fig. 2). It was observed from the experimental results that a higher salt rejection could be obtained by increasing the IP reaction time in order to obtain a denser zeolite–PA layer. A zeolite loading of 0.15 wt% was found to be the optimal loading. The inclusion of NaY zeolite nanoparticles in the TFN membranes enhanced the water flux from 39.63 to 74.32 LMH with a salt rejection of 98.8% under the optimal conditions of 25 °C, 1.55 MPa, and 2000 ppm NaCl solution. The water flux was further enhanced to 86.05 LMH, but with a salt rejection of 98.4% after the post-treatment of the TFN membranes using the optimal solution composition of sodium lauryl sulfate, camphorsulfonic acid–triethylamine salt and glycerol. The water flux of post-treated TFN membranes with NaY zeolite nanoparticles was two times higher than that obtained using a TFC membrane devoid of zeolite nanoparticles.

Fig. 2 SEM image of NaY zeolite nanoparticles.³¹

The feasibility of employing zeolite incorporated membranes for seawater desalination using RO has been investigated in another study.³⁶ Two kinds of zeolite – hydrophilic faujasite (FAU) and hydrophobic MFI – with distinct wetting properties were used for RO membrane fabrication. A salt rejection of almost 100% was achieved using both membranes. Additionally, the membrane permeability was significantly improved to 720 L m⁻² h⁻¹ bar⁻¹ for a membrane thickness of less than 3.5 nm, which is 100 times higher than the permeability of commercial RO membranes. Moreover, the pressure drop in the FAU membrane was found to be less due to its hydrophilicity, whereas a larger pressure drop was observed in the hydrophobic MFI zeolite membrane due to capillary resistance. It was concluded that the pressure drop during desalination could be sustained by placing the nanoscale zeolite membrane on a porous substrate and optimizing the ratio of the nanomembrane thickness to the pore radius of the supporting substrate. NaX zeolite nanoparticles have also been used in TFN membranes for an improved membrane performance.³⁷ The TFN membranes comprising NaX zeolite nanoparticles were synthesized on porous polyethersulfone (PES) ultrafiltration (UF) supports by IP of MPD and TMC. The results obtained demonstrated that the NaX zeolite TFN membranes have higher water permeability and greater thermal stability compared to the pure PA membranes. In addition, increasing the content of NaX zeolite nanoparticles reduced the film thickness of the PA membrane and enhanced its surface properties, water flux, and pore size. Increasing the concentration of MPD and TMC in the IP of the TFN membranes resulted in a higher water flux but reduced solute rejection. The optimum membrane performance was achieved for TFN membranes comprising approximately 0.2% (w/v) NaX zeolite nanoparticles, 2% (w/v) MPD and 0.1% (w/v) TMC. The optimum water flux through the hybrid membrane was 1.8 times greater than that of the pure PA membrane with no change in the salt rejection.

Interesting results have also been reported for a hybrid membrane integrated with zeolitic imidazolate framework-8 (ZIF-8). The effect of introducing ZIF-8, which is a microporous and hydrophobic hybrid material, as filler into the PA selective layer of TFN RO membranes has been investigated experimentally.¹⁴ The incorporation of ZIF-8 nanoparticles enhanced the water permeability by up to 162% at 0.4% (w/v) loading compared to pristine PA membranes. Salt rejection, however, remained at 98% for simulated brackish water (2000 ppm sodium chloride solution) desalination with a hydraulic pressure of 15.5 bars. Additionally, the inclusion of ZIF-8 caused the TFN selective layer surface to become more hydrophilic and less cross-linked compared to that of the pristine PA membrane. In general, a higher water permeability was obtained using microporous hydrophobic fillers (silicalite-1, carbon nanotubes and ZIF-8) compared to hydrophilic fillers (zeolite 4A). This is because of the greater compatibility of ZIF-8 with the PA matrix, and faster transport of water within the hydrophobic framework due to less attraction of water to the pore wall compared to the classic zeolite hydrophilic framework. In addition, the performance of the TFN RO membranes can be improved further by reducing the particle sizes of ZIF-8 and including a metal–organic framework with an identical surface hydrophobicity but marginally bigger pores. Another study on a nanocomposite zeolite membrane with aminated template free zeolite nanoparticles (aTMA) has given impressive water fluxes.³⁸ Kim et al. fabricated nanocomposite TFC RO membranes from a sulfonated poly(arylene ether sulfone) material comprising aminated template free zeolite nanoparticles (aTMA) and amino groups (aPES) to improve the chlorination resistance and membrane performance.³⁸ The proposed membranes were synthesized via IP. Based on the performance test, the nanocomposite membranes had a water flux and salt rejection of 37.8 LMH and 98.8%, respectively. After chlorination, the water flux increased by 2.5 LMH but the salt rejection was reduced by 12.7% compared to that of pristine PA. An improved performance of the hybrid RO membrane resulted from the nanoparticle loading because of the change in the three-dimensional structure of the PA network. As a result of this structural modification, a high degree of cross-linking leading to stiffness of the copolymer chain in the RO membranes was ensured. The aminated nanoparticle composition protected the underlying active layer from chlorine attack. Consequently, aPES and aTMA facilitated an increase in water permeability and improved the chlorine resistance. The active layer of the RO membrane was also protected from degradation.

2.3 Silica

Silica nanoparticles belong to another class of nanomaterials that have been tested recently for improved RO membrane performance. Recent studies have recommended the incorporation of silica with composite membranes for improved water permeation.^39–44 Pure silica membranes can exhibit microporous structures but would lose stability during desalination because the water permeating though the micropores would increase the pore sizes of the silica film by interacting with the silane groups.⁴² Therefore, the incorporation of silica in polymeric nanocomposites may be necessary for chemical and mechanical stability. A modified acetate–polyethylene glycol (PEG) polymer has been combined with silica to produce different nanocomposite RO membranes with different levels of performance.⁴⁰ This work entailed the optimization of the salt rejection and the permeation properties of the modified acetate–PEG membranes with varying amounts of silica for RO. A 2-stage phase inversion protocol with varying cellulose acetate (CA)/PEG ratios was followed in order to optimize the membranes. The incorporation of silica into the membranes increased the permeation properties of the fabricated membranes in terms of: water flux (from 0.35 to 2.46 LMH), salt rejection (increase of 11.41%), and mechanical stabilization (from 1 to 4%). The improved trend observed for all the measured membrane performance properties resulting from the addition of silica showed the need to optimize the silica particle loading because the silica had a negative effect on the hybrid RO membrane properties beyond a critical load. Peyki et al. has modified the surfaces of PA TFC RO membranes by incorporating silica nanoparticles on the PA active layer.¹⁶ Different amounts of silica nanoparticles were added to the amine solution. The hydrophilicity of the modified membrane was higher than the pristine membranes as observed from contact angle measurements. The hydrophilicity of the membrane surface was found to increase with increasing silica content in the aqueous solution. Additionally, a significant alteration in the membrane surface structure occurred due to the incorporation of silica nanoparticles. The roughness of the membrane surfaces increased with an increase in the silica concentration in the amine solution. Moreover, cross-flow permeation tests were used to test the performance of the modified membranes at an operating pressure of 44 bars with aqueous sodium chloride solution. The short-term experimental results indicated that the water flux of the modified membranes progressively improved at lower nanoparticle levels, up to 0.1 wt%, and then reduced steadily. Furthermore, an increase in sodium chloride rejection was observed for lower silica concentrations (0.005 and 0.01 wt%) followed by a reduction in salt rejection with increasing silica concentration in amine solution. Long-term experiments revealed a lower water flux decline for the modified RO membranes compared to pristine TFC RO membranes. Lastly, the modified membranes also had better anti-fouling properties than the unmodified membranes.

Silica nanoparticles have also been synthesized on a CA TFC RO membrane to impart antibacterial properties to the membrane. This work studied the effect of the development of dry-cast CA TFC RO membranes by swelling in room-temperature water baths before the annealing treatment.⁴⁵ After the swelling in water, the performance of the membranes was also studied, as well as the membrane’s physical properties, by ellipsometry, contact angles, and SEM. It was found that the silica–CA TFC membrane could reduce the bacterial density in both the feed solution and on the surface of the membrane for biofouling control. The CA TFC membrane seemed to be a good platform to study the CA selective layer in the RO membrane and the results obtained also showed that a combined swelling and annealing treatment improved the desalination performance with a salt rejection of 94% for CA TFC membranes compared to 55% obtained for asymmetric CA RO membranes without deterioration of flux. The water swelling time improved the permeability and the selectivity was slightly increased. Hydrophilic water-swollen TFC membranes can also be used for other applications such as the removal of water vapor, carbon dioxide, and hydrogen sulfide simultaneously from agro biogas.⁴⁶ Swelling of RO membranes has also been used lately to detect a new fouling mechanism in the membranes.⁴⁷

2.4 Other nano-enhanced membranes

Recent insight on the integration of nanoparticles with polymer-based membranes has revealed the antifouling properties of these nanoparticles. Today, silver nanoparticles are largely considered to be biocidal but the loading of silver nanoparticles onto membranes is still a real challenge.^{11,19,48–51} Meanwhile, a new procedure aimed at loading silver nanoparticles on TFC RO membranes has been developed.¹⁹ The method involved the reaction of the silver salt with a reducing agent on the membrane to form a properly-bound nanocomposite. The results showed that water permeability through the nanocomposite membrane was affected by silver nanoparticle functionalization.¹⁹ Water permeability decreased by up to 17%, whereas improvements in the salt selectivity, hydrophilicity, zeta potential, and surface roughness were achieved. Interestingly, a strong antibacterial activity was reported with a reduction of more than 75% in the number of live bacteria on the membrane and with a 41% reduction in the total biological volume. The simplicity of the method, short reaction time, ability to load the silver nanoparticles on-site, and strong imparted antibacterial activity highlight the potential of this method for real-world RO membrane applications. Novel surface coatings have also been developed on PA TFC membranes using biocidal silver nanoparticles and antifouling polymer brushes via polyelectrolyte layer-by-layer (LBL) self-assembly.¹⁷ The polyelectrolyte LBL films contained poly(acrylic acid) (PAA) and poly(ethylene imine) (PEI). The PEI was used either as pure PEI or in the form of silver nanoparticles coated with PEI (Ag–PEI). Because PA TFC membranes are prone to biofouling due to their characteristic physicochemical surface features, the membrane surfaces were modified by coating in order to enhance their antifouling properties for this study. Then the surface energy and hydrophilic properties were studied to show the microbial adhesion to the surfaces. The results showed that in general, modified surfaces containing silver nanoparticles demonstrated a low adhesion in a microbial adhesion test with E. coli and the membranes coated with PAA–Ag–PEI could remove 95% of the bacteria attached to the surface with 1 hour of contact time. Both antifouling and antimicrobial studies proved that these novel coated membranes could be a potential solution to prevent and control biofouling in PA TFC RO membranes.

The influence of integrating a polyhedral oligomeric silsesquioxane (POSS) nanofiller and PA on water flux and salt rejection improvements has also been investigated.⁵² This study involved the investigation of the influence of adding a POSS nanofiller in a PA selective layer. POSS is a hybrid intermediate silicone-based compound which behaves like a nanostructured fiber when mixed with a polymer.⁵³ Four different POSS materials (P-8NH₂, P-8phenyl, P-1NH₂ and P-8NH₃Cl) were incorporated into the PA selective layer during IP by either chemical cross-linking or physical blending. The fabricated membranes were analyzed at a pressure of 15.5 bar using sodium chloride solution with a concentration of 2000 ppm. Compared to the pristine PA membrane, the water flux increased by 65% and the salt rejection remained at 98% when 0.4% (w/v) of P-8phenyl was added to the organic phase and physically blended into the selective layer. Similar results were obtained when P-8NH₂ and P-8NH₃Cl were added to the organic phase. An increased water flux was also obtained from the chemical fixation of P-8NH₂ and P-8NH₃Cl into the selective layer. However, the water flux was unsatisfactory for chemical cross-linking of P-1NH₂ in the organic phase due to the hydrophobicity of the membrane produced.

The recent integration of carbon nanotubes into polymeric membranes for the RO process has also yielded positive results.^2,10,54–58 Zhao et al. has enhanced the performance of nanocomposite PA RO membranes incorporated with carboxy-functionalized multi-walled carbon nanotubes (MWCNTs).¹⁵ Mixed acids were used to pretreat the virgin MWCNTs prior to modifying them with diisobutyryl peroxide to improve their chemical activity and dispersity. The synthesized membranes had a skin layer of thickness 100–300 nm within which the modified MWCNTs were inserted (Fig. 3). It was observed that the nanocomposite membrane surface was more negatively charged compared to the pristine PA membranes. Additionally, the morphology of the membrane was noticeably altered when the carbon nanotube loading was increased, thereby causing the permeate flux to increase significantly from 14.86 to 28.05 LMH. However, only a marginal decrease in salt rejection was observed. Moreover, the modified nanocomposite membranes demonstrated better anti-oxidative and anti-fouling properties than the PA membranes free of MWCNTs under similar testing conditions. Nanotubes have the ability to facilitate fast water transport through membranes⁵⁹ and the incorporation of functionalized MWCNTs into TFN RO membranes is capable of enhancing the water flux obtained from these membranes.⁶⁰

Fig. 3 (A and B) surface morphology of the unmodified PA composite/PA–MWCNT nanocomposite from SEM; (C and D) cross-section of the unmodified PA composite/PA–MWCNT nanocomposite from transmission electron microscopy (TEM).¹⁵

Thin film bipolar RO desalination membranes have also been synthesized from PAA and PAH through a spin assisted layer by layer (SA-LbL) coating technique.⁶¹ Polysulfone UF membranes and silicon wafer substrates were used for depositing these polyionic coatings. The thickness of the film could be regulated by altering the number of layers, varying the pH or changing the spin speed. The thickest film was made when the low pH of PAA was combined with the high pH of PAH, and vice versa for the formation of the thinnest film. The membrane performance test conducted with feed water at a pressure of 700 psig and a sodium chloride concentration of 15 000 ppm showed that 35 bi-layers of PAH–PAA on polysulfone UF membranes demonstrated a water permeability of 0.22 L m⁻² h⁻¹ bar⁻¹ and a salt rejection of 88%. The morphology of the membrane film was observed to change after the permeation/performance test due to the application of high pressure that enabled salts to diffuse further into the film and change the polymer chain conformation. The SA-LbL method was found to be very effective for fabricating RO membranes for membrane thickness regulation.

3. Recently developed polymeric and hybrid polymer membranes

3.1 Fabrication and performance studies

A commonly used polymeric membrane material for the RO process in recent times is PA.^39,62–65 This has prompted the molecular study of a PA RO membrane to give a better comprehension of the separation process that can be achieved through the membrane at the molecular level.⁶⁶ A computational molecular simulation was used to build a fully aromatic PA (FAPA) RO membrane made from identical monomers, such as those found in commercial membranes.⁶⁶ The cross-linking degree and solid–liquid interface of the membrane were controlled by this simulation. This study confirmed the previous data available regarding the water structure, water dielectric properties, and water dynamics through the membrane. Indeed, the confinement inside the membrane cavities would not disturb the overall structure of the water and the relaxation of water dipoles was much slower inside the membrane than in the bulk phase. In addition, the dielectric properties of confined water were severely modified with respect to the bulk phase. This work showed that a FAPA RO membrane could play a significant role in salt rejection and proposals for future research to establish the relationship between the investigated molecular dynamics and salt rejection were suggested.

Another study was performed to attempt to develop a model to simulate the transport of water molecules across highly cross-linked PA RO membranes.⁶⁷ The mechanism behind ion transport through highly cross-linked PA membranes at the molecular level was also investigated in this study. Fujioka et al. studied the internal structure of the PA RO membranes’ selective skin layer using positron annihilation spectroscopy (PAS) to obtain a better understanding of the water and solute transport, which will help in introducing further enhancement to RO membranes.³ PAS showed that the RO membranes contain a very thin selective skin layer with free volume as subnanometer sized holes. Previous studies using data from positron annihilation lifetime spectroscopy (PALS) reported that the radius of average free volume hole for commercial RO membranes was between 0.20–0.29 nm in the selective skin layer with a thickness of around 100 nm. The experimental results obtained using PALS showed that the size of the free volume hole is the most significant parameter in governing rejection (of boron). Additionally, the selective skin layer’s free volume fraction was found to affect the solute rejection. The solute transport could also be influenced by the thickness of the selective layer. The measurements of the shape of the free volume hole, the free volume fraction and the distribution of the hole size limit the possibility of further investigating the solute transport. Using molecular dynamics simulations in addition to the free volume hole size measured by PALS can help to achieve a better understanding of solute rejection mechanisms through membrane polymer chains.

Cellulose triacetate (CTA) is another polymeric membrane material whose continued use for RO has been sustained in recent times.^62,68 Although CTA membranes are in a distant second position to PA TFC membranes in terms of their market share in recent applications, they are preferred most especially in the Middle East because of their superior ability to withstand chlorine attack.^62,68 Fujioka et al. evaluated the performance of CTA hollow fiber RO membranes for reusing potable water by observing its rejection behavior for forty one trace organic chemicals (TrOCs). The results obtained were compared with those achieved using PA membranes.⁶⁹ The results obtained for the CTA membranes were largely similar to those obtained for the TrOC rejection behavior of PA membranes. However, compared to PA membranes, TrOC rejection by CTA membranes was found to be greatly influenced by hydrophobic interactions and less dependent on electrostatic interactions. No variation was observed in the rejection of positively and negatively charged TrOCs when CTA membranes were used. Nonetheless, there was great variation in neutral TrOC rejection by CTA membranes (unlike PA membranes) because the neutral TrOC rejection was observed to depend on their minimum projection area (molecular size) which strongly correlated with the PA membrane structure. Additionally, hydrophobicity was found to greatly influence the rejection of N-nitrosamines by CTA membranes, whereas the molecular size influenced the rejection of N-nitrosamines by PA membranes. Finally, the charged TrOCs exhibited much greater rejection by CTA membranes than the neutral TrOCs irrespective of the molecular size. N-Nitrosamine rejection has also been evaluated using a spiral wound PA nanocomposite RO membrane from Hydranautics (ESPA2-4040). In this study, a mathematical model was developed using hydrodynamic calculations and the principle of irreversible thermodynamics to precisely describe N-nitrosamine rejection by the spiral wound RO membrane systems for a set of operating conditions.⁷⁰ The N-nitrosamine rejections obtained from the mathematical model were validated with experimental results from a pilot RO system. The modeled results showed that the rejection of N-nitrosamines with low molecular weight could be increased by increasing the permeate flux, and decreased by increasing the system recovery. Also, a reduction in localized rejection was observed along the flow path in the membrane vessel. The developed model can be beneficial for use in regulatory monitoring and system design. In general, these studies confirm the appropriateness of RO membranes for rejecting micropollutants or carcinogenic disinfection products and trace organics from water.⁷¹

The incorporation of hyperbranched polyesters (HBPE) into CA for improved membrane performance has been tested. CA membranes with different compositions of HBPE blended into the membrane structure were prepared via phase inversion.⁷² HBPEs were investigated because of the hydrophilic nature of the polymers. The addition of HBPE improved certain morphological features of the membrane, such as the filtration properties, wettability, thermal behavior and hydrophilicity. The membranes were tested for 30 minutes at a pressure of 500 kPa and pure water fluxes were measured at 400 kPa. The results obtained showed that a membrane prepared from pure CA has the lowest pure water flux because of its dense low-hydrophilicity surface structure. The main reasons for the formation of such a structure include the slow phase separation, time required for uniform dispersion, and nucleus growth on the surface. Hybrid CA–HBPE membranes (CA–HBPEG3) with small dimensions showed higher fluxes. The pure water flux of these membranes increased from 24.6 to 37.6 LMH when the HBPEG3 content was increased from 2.5 to 10 wt%. This was due to the formation of a greater number of pores on the membrane surface at higher HBPEG3 contents. Another set of hybrid CA–HBPE membranes (CA–HBPEG4) was developed with larger pores formed by partial phase separation. However, these membranes had relatively small water fluxes despite having larger pores. This effect was possibly related to the high permeation resistance caused by the honeycomb-like voids in the membrane matrix. This study also indicated that the molecular weight cutoff (MWCO) values for CA–HBPEs are low but can be slightly increased through the controlled increase of surface porosity. The incorporation of CA with poly(vinylidene fluoride) (PVDF) with TiO₂ loading has also been investigated by Liu et al.⁷³ TiO₂ nanoparticles are antimicrobial agents and they can curtail organic fouling.¹⁸ A novel dual-layer CA–PVDF hollow fiber membrane was fabricated via a two-step thermally induced phase separation (TIPS) and non-solvent induced phase separation (NIPS) process.⁷³ The morphology and porosity of the synthesized membranes were found to be influenced by the outer layer’s dope solution composition: co-solvent (acetone) ratio, polymer concentration, and TiO₂ nanoparticle loading. Increasing the acetone ratio and TiO₂ nanoparticle loading resulted in a narrower pore size distribution (Fig. 4) and a thinner outer layer that significantly improved the rejection of electrolytes and dextran, and maintained the pure water permeability of the hollow fiber membrane. Increasing the loading of TiO₂ nanoparticles also helped to improve the hydrophilicity, anti-bacterial and anti-biofouling properties of the dual-layer hollow fiber membrane. The mechanical strength of the dual-layer membrane was observed to be greater than 8 MPa, in addition to the sodium sulfate rejection of 90–95% and pure water permeability of 1 to 4 L m⁻² h⁻¹ bar⁻¹. The possibility of using the resultant membrane for treating RO concentrates was verified from its performance test that yielded a TOC removal greater than 90% and a rejection of total dissolved salts of less than 60%. Permeate from the fabricated hybrid membrane was safe for discharge to underground and surface waters as a result of the antibacterial effect of TiO₂.

Fig. 4 Pore size distribution of CA–PVDF and pure CA flat sheet membranes.⁷³

It has also been demonstrated that the integration of polyvinyl alcohol (PVA) with CA could lead to an improvement in the CA RO membrane performance for desalination of brackish water and seawater. Muhammed et al. have fabricated and characterized a PVA–CA composite membrane by using different concentrations of maleic acid and varying the reaction times to cross-link the PVA layer.⁷⁴ Thin films comprising PEG and CA were interfacially polymerized over PVA support membranes having different chemistry and pore structures. Large pores were added onto the hydrophobic skin layer resulting in membranes with higher water permeability. A more rough composite membrane surface was achieved from a smaller percentage of PVA formation with the membrane pores. As a result of this approach, the overall path length became shorter for solute and water transport from the feed to the permeate side of the PVA layer.

Most RO membranes are composites because they combine the properties of the different components.^75,76 PA TFC membranes have found the most important niche in the current RO industry.^43,77–81 They provide the benefits of high flux and high salt rejection when compared with asymmetric CTA membranes.^77,78,80 However, in contrast to their impressive water permeation and salt rejection features and greater resistance to biodegradability, PA TFCs are more susceptible to chlorine attack because they contain chlorine-sensitive nitrogen sites.^80–82 More developments in the fabrication and evaluation of the performance of PA TFC RO membranes for a high-temperature selection process have also been reported lately.⁸³ Apart from susceptibility to chlorine deformation, the influence of temperature changes on the performance of a PA TFC membrane has been tested in a recent work.⁸⁴ In this study, gas separation tests were used to evaluate the impact of increased temperature on TFC RO membranes after different pretreatment procedures. It was shown that high permeance results can be obtained for a maximum He/N₂ selectivity at temperatures up to 150 °C. The temperature-induced changes in the polymer structure and in the transport of compounds suggested the appropriateness of the use of PA membranes for high-temperature separation processes. The use of a PA TFC RO membrane to remove substances like fluorinated surfactants, which are resistant to chemical or thermal attack, has also been studied.⁸⁵ Fluorinated surfactants can reduce the surface tension of fire-fighting waters. Therefore, a PA thin film membrane has been used to significantly reduce the concentration of fluorinated surfactants in fire-fighting water from 30 to 0.1 mg L⁻¹. The results obtained also showed consistent permeability over several days (0.5 L m⁻² h⁻¹ bar⁻¹) and the retention rate achieved was around 99.4 to 99.9%. A recent study of PA TFC RO membranes has shown that the performance of these membranes in terms of the water flux and salt rejection can be further improved through the post-treatment of the membrane with dimethylformamide (DMF).⁸⁶ The group fabricated and investigated the performance of PA TFC RO membranes that were prepared by IP of MPD and TMC, and casted on a cross-linked polyimide support. The effect of using the activating solvent, DMF, on the performance of the membrane was studied. The results demonstrated an increase in water flux by eight times after post-treating the membrane with DMF. An improvement in salt (sodium chloride) rejection was also observed upon post-treatment with DMF, especially for membranes with initial salt rejections of less than 50%, resulting in a peak rejection of 94%.

The behavior of a commercial PA composite RO membrane has also been tested under gamma rays in the presence of water and oxygen in order to detect the stability of the membrane under irradiation.⁸⁷ The PA composite membrane used was tested before and after two doses of gamma rays (0.1 and 1 MGy). The membrane’s performance in terms of its NaCl rejection and water permeability was obtained. Commercial PA TFC membranes are usually composed of an aromatic PA active layer with a polysulfone microporous support layer and a bottom structure made up of non-woven polyester. Membrane texture analysis and chemical composition were carried out to characterize the membrane by advanced analysis techniques. It was observed from AFM images that the membrane degraded after irradiation at 1 MGy by structural modifications resulting from increased roughness and chain breakage at different layers (Fig. 5). The drop of the permselectivity properties of the membrane was a direct consequence of this structural modification. Irradiation at 0.1 MGy seemed to have only a minor effect. It was concluded that PA composite RO membranes can be resistant to gamma rays until a dose level of 0.1 MGy is reached. Structural modifications of the membrane surface would occur around 1 MGy. Further research is ongoing to show the performance of RO membranes when exposed to different levels of irradiation and radioactive materials in water.^87–89

Fig. 5 Atomic force microscopy (AFM) images of (a) a virgin PA composite RO membrane, and membranes irradiated at (b) 0.1 and (c) 1 MGy gamma irradiation.⁸⁷

Operating conditions have also been shown to play a significant role on the efficiency of PA TFC membranes for the RO process.⁹⁰ For example, the behavior of Dow’s BW30LE PA TFC RO membranes for brackish water desalination has been examined using varying salinity and pH conditions in both experimental and numerical modeling studies.⁹¹ It was found experimentally that the membrane charge, thickness and pore size were dependent on both the salinity and pH levels. In particular, the membrane thickness increased with increasing feed salinity and pH. Additionally, the flux of hydroxide and hydronium ions were inspected using similar experimental conditions to examine the membrane performance. An increase in the rejection of hydroxide and hydronium ions was observed with rising surface charge density. The flux was not affected by the applied pressure at zero salinity. However, negative rejection of both ions was observed after adding salt in order to sustain the electroneutrality in the permeate solution. The flux of hydronium ions increased significantly when the permeate flux was increased. In order to predict hydroxide and hydronium ion transport through the RO membrane, modeling was carried out using the extended Nernst–Planck equation and the interface partitioning was evaluated based on steric hindrance and Donnan equilibrium. A good agreement was obtained with experimental results without adjusting any parameters.

Richards et al. also used Dow’s PA TFC membrane to demonstrate a safe operating window (SOW) for a RO desalination system powered by renewable energy.⁹² This study was the first to develop the methodology to experimentally determine the SOW for a renewable energy powered membrane filtration system used for brackish water desalination. The SOW for the wind-powered membrane filtration system was dependent on the following factors: available power of the pump (pressure and flow rate), highest suggested recovery and the osmotic pressure of feed water. The RO membrane modules (and brackish feed water salinity) used in this study were: BW30-4040 (5500 and 10 000 mg L⁻¹) and aged BW30-4040 (5500 mg L⁻¹). The maximum recovery of 30% was found to be the main limiting factor at lower salinity values, whereas the osmotic pressure was the chief constraint at higher salinity values. Constant recovery was observed to be the best operating approach as it yielded the maximum water flux and sustained good retention at a specific power consumption, which resulted in the lowest specific energy consumption (SEC). Nonetheless, a ‘constant set-point’ mode was the preferred the operating strategy over the constant recovery approach as this is challenging to apply in practice. The mode of ‘constant set-point’ offered an effective and robust solution for water provision in remote areas with a slight decrease in the system performance. Overall, it was possible to power the suggested approach by solar (photovoltaics) or wind energy due to its low SEC (approximately 3 kW h m⁻³). This allowed the system to function over a wide power range (70–280 W) in order to attain the anticipated pressure range (5–11.5 bar). Generally, this SOW methodology allows the performance assessment of a large range of factors for PA TFC RO membrane filtration systems. This is against the backdrop that energy efficiency is a crucial factor for RO desalination.^93–96

Cross-flow velocity has also been studied as an operating parameter that influences the concentration layer of PA TFC membranes during RO desalination. An experimental investigation and numerical modelling has been carried out recently to investigate the development of a concentration polarization (CP) layer during cross-flow in a RO membrane slit channel.⁹⁷ The development of a CP layer in the slit was directly visualized as an interference fringe pattern using a digital holographic interferometry (Fig. 6). It was observed that the polarization layer along the slit-type channel continuously increased irrespective of the cross-flow velocity, except near the cell outlet because of the edge effect. However, the growth of the polarization layer was found to be bigger at a lower Reynold’s number but the permeate flux was not influenced noticeably. Nonetheless, the permeate flux was significantly affected by the concentration of the feed. Overall, the polarization layer development was found to be dependent on the driving force, flux resistance and hydrodynamics of the process. The simulation results obtained from commercial CFD software were observed to be in good agreement with the experimental results.

Fig. 6 Concentration profile from digital holographic interferometry.⁹⁷

Novel TFC RO membranes with high permeate fluxes have also been fabricated for seawater desalination by adding the hydrophilic o-aminobenzoic acid–triethylamine (o-ABA–TEA) salt onto the PA TFC RO membrane during IP of MPD and TMC on a polysulfone support.⁹⁸ Several membrane fabrication conditions, such as the o-ABA–TEA salt and isopropanol concentration, extra drying time of the amine and the hydrocarbon removal time, were enhanced based on membrane performance tests. The optimal additive concentration was found to be 1.0 wt%. The modified membranes were more hydrophilic than the pristine membranes due to the incorporation of hydrophilic materials. Performance tests were conducted using a synthetic sodium chloride solution of 3.28 wt% at 5.52 MPa and 25 °C. The modified membranes had a salt rejection of 99.41% and a water flux of 83.5% (75.42 LMH), higher than the pristine membranes. The hydrophilic additive produced an added route for water transport enhancement and delivered a charge repulsion that improved the salt rejection. The synthesized membranes were further tested for their performance under the same conditions using seawater from Port Hueneme in California, United States. A lower water flux was attained using the seawater due to the presence of a higher solid content (3.45%) in the tested seawater. Moreover, the stability test conducted using this seawater as the feed solution demonstrated a stable desalination performance for a period of 30 days.

The surface and structural features of PA TFC RO membranes also play a crucial role in the membrane performance in terms of water permeability and salt rejection.^99–103 Liu et al. investigated the influence of surface properties and the structure of several TFC PA RO membranes on the performance of these membranes.⁹⁹ These membranes, with slight structural variances, were synthesized via the reaction of 5-chloroformyloxy-isophthaloyl chloride (CFIC), 5-isocyanato-isophthaloyl chloride (ICIC) and TMC with MPD separately through the IP technique on a polysulfone supporting mesh. The performance of the fabricated RO membranes was evaluated based on their water permeability and salt rejection. The properties of the membrane surfaces were characterized using XPS, SEM, ATR-IR, AFM, streaming potential and contact angle measurements. The stability of chlorine was examined by evaluating the performance of the membrane before and after exposing it to hypochlorite. The results indicated that the performance, surface properties and chlorine stability of the RO composite membranes were strongly influenced by the structure of polyacyl chloride. Incorporation of the isocyanato group into polyacyl chloride was found to improve the water permeability, surface smoothness and hydrophilicity but reduce the chlorine stability of the TFC membrane. On the other hand, integration of a chloroformyloxy group into polyacyl chloride was found to increase the salt rejection and increase the surface roughness but reduce the water permeability of the TFC membrane. The most hydrophilic PA TFC membrane was MPD–ICIC and the least hydrophilic was MPD–CFIC. Among the three membranes, the surface of MPD–CFIC was found to be the roughest followed by MPD–TMC, and MPD–ICIC had the smoothest membrane surface. The coefficient of pure water permeability was 2.87, 3.53, and 4.05 L m⁻² h⁻¹ bar⁻¹ for MPD–CFIC, MPD–TMC, and MPD–ICIC, respectively. The salt rejection of the three TFC PA membranes was greater than 98% when tested at 1.6 MPa using a sodium chloride feed solution with a concentration of 2000 mg L⁻¹. The rate of salt rejection was highest for MPD–CFIC and lowest for MPD–ICIC. Moreover, the MPD–CFIC membrane was verified to be the most suitable for processing highly concentrated aqueous salt solutions. The MPD–ICIC membrane was the least chlorine resistant among the tested membranes.

The functionalization of PA TFC RO membranes with acyl chloride monomers has also been tested in another recent investigation as a method of improving the membrane performance. Wang et al. has synthesized three novel monomers of polyacyl chloride – 2,3′,4,5′,6-biphenyl pentaacyl chloride (BPAC), 2,4,4′,6-biphenyl tetraacyl chloride (BTAC) and 2,2′,4,4′,6,6′-biphenyl hexaacyl chloride (BHAC) – to investigate the impact of their functionality on the PA TFC RO membrane properties.¹⁰³ The TFC RO membranes were fabricated on a polysulfone support via IP of TMC, BPAC, BTAC and BHAC with MPD. The characterization and separation performances of the synthesized membranes were evaluated using SEM, ATR FT-IR spectroscopy, AFM, XPS, contact angle, streaming potential measurements and cross-flow RO performance tests. The results indicated that the RO membrane properties were significantly influenced by the acyl chloride monomer functionality. It was observed that the skin layer of the prepared membrane became thinner, more negatively charged and smoother as the acyl chloride monomer functionality increased. This occurred because of the higher degree of cross-linking of the nascent PA film with severe resistance to amine diffusion into the organic phase throughout the process of IP. Nonetheless, no trend was observed for the hydrophilicity change due to the combined influences of the carboxylic acid group content and surface roughness. Moreover, the RO performance tests yielded salt rejection rates that were close for all the four membranes. However, it was suggested that the quantity of the acyl chloride functionality needs to be optimized for an improved membrane performance. An uncontrolled increase in acyl chloride functionality was observed to decrease the water flux as a result of reduced surface roughness, greater number of carboxylic acid groups on the membrane surface, and reduced flexibility of the cross-linked PA chains.

The modification of the PA active layer of a RO membrane by coating it with a hyperbranched poly(amido amine) (PAMAM) hydrophilic polymer has also been studied for membrane performance improvement.¹⁰⁰ PAMAM was coated on the PA active layer of the RO membrane via chemical coupling. The RO membrane was coated by spraying a solution containing 10 wt% PAMAM on the membrane surface using either water or methanol as solvent. Membrane modification using water and methanol as solvents improved the water flux by 19.6 and 24.8%, respectively. The increase in water flux was due to the suppression of cross-linking following the final step of curing. A stable membrane surface modification occurred under filtration conditions. It was observed that a smooth thin PAMAM film was produced with almost complete surface coverage when water was used as a solvent (Fig. 7). However, using methanol as solvent caused PAMAM to deposit as clusters on the membrane surface. Surface modification with PAMAM increased the water permeability by 20–25% compared to the uncoated membrane, but did not influence the salt rejection. However, a lower salt rejection resulted for the modified membrane when methanol was used as a solvent. It was also observed that the modified membranes coated with aqueous PAMAM solution exhibited low protein fouling when compared with the pristine membranes and those that were coated with methanolic PAMAM solution. The enhanced performance of the modified membranes with aqueous PAMAM solution was due to the development of a second homogenous hydrophilic hydrogel layer.

Fig. 7 AFM pictures of (a) unmodified TFC–PAMAM, (b) TFC–PAMAM with methanol as solvent and (c) TFC–PAMAM with water as solvent.¹⁰⁰

The pore structure of the surfaces of PA TFC RO membranes could also contribute to the performance of these membranes. Therefore, it is necessary to understand the mechanism involved during pore formation on the film surfaces of PA TFC RO membranes. Yan et al. characterized and studied the complex porous structures of commercial FAPA RO membranes using TEM and SEM.¹⁰¹ Experimental results showed that the formation of pores on the FAPA film surface always took place on the side facing the aqueous phase. A novel FAPA film structure model was proposed for the RO membranes using the observations obtained from the TEM and SEM images of the FAPA film cross-section. The RO membrane’s main barrier layer was the dense selective layer that formed on the top surface of the PA. The pores and the dense layer on the FAPA film’s back surface were linked through the asymmetrical interconnecting tunnels and voids of the ridges and protrusions. These tunnels and voids offered passages to permit MPD to move through the FAPA film during the process of IP formation and allowed water to penetrate through during RO filtration. The partial replacement of the conventional MPD monomer in PA TFC RO membranes with the linear monomer, 1,3-diamino-2-hydroxypropane (DAHP) has also been studied to evaluate the effect of this substitution on the membrane performance. TFC PA RO membranes were fabricated via IP in the aqueous phase to increase the water flux.¹⁰² The linear DAHP monomer was used to partially replace the conventional MPD monomer and the permeability increased by approximately 22% (2.67 ± 0.09 L m⁻² h⁻¹ bar⁻¹) for the synthesized membrane at an optimal DAHP/MPD ratio of 12.8%. However, no significant change in salt rejection was observed, which was maintained from 96 to 98%. The permeability increased to 2.18 ± 0.08 L m⁻² h⁻¹ bar⁻¹ when the support surface was washed to remove protective surface coatings before the membrane development. In general, an enhanced water flux occurred for both washed and unwashed supports when DAHP units were incorporated into the PA network. This achievement was reported because the integrated DAHP changed the membrane surface structure of the PA film from large ridge and valley to enlarged nodular structures. The introduction of DAHP to the PA TFC membrane also reduced the thickness of the selective layer, reduced the surface roughness and increased the hydrophilicity of the membrane that caused an increase in the water flux.

To improve the performance of asymmetric RO membranes, Stevens et al. used 20% disulfonated biphenol and an aryl sulfone (BPS) copolymer to develop an optimal recipe.⁸¹ The membranes were fabricated via the phase inversion of BPS solutions. In addition to optimizing the recipe, a number of post-treatment steps were included for the formation of a dense rejection layer and healing of the membrane defects in order to optimize the salt passage. The post-treatment involved boiling the prepared membrane in concentrated electrolytes to break down the pores and dehydrate the membranes so as to produce almost defect-free skins. This step was followed by in situ poly(vinyl alcohol) coating to repair any leftover holes or defects so that a highly salt rejecting membrane surface could be produced. An enhanced water flux and salt rejection resulted from the post-treatment of the membranes. The final salt passage was observed to be strongly dependent on the temperature and ionic strength of the annealing solution. The salt passage ranged from 0.34 to 1.81% across a water flux range of 2.5 to 20 LMH when tested at a pressure of 15.5 bar with 2000 ppm sodium chloride solution. Although the performance of the membrane was not good enough to compete with commercial PA RO membranes, the obtained results demonstrated that the BPS membranes have the potential to be used for RO desalination.

RO membrane performance is also strongly dependent on the properties of the membrane support. Several studies have estimated the dependence of the structural parameters of the membrane on its performance.^104,105 Conventional methods, such as SEM, have been used to characterize the tortuosity and porosity of the membrane support layer. However, these methods are difficult to implement and cannot be used to extensively visualize the pore structure. Manickam et al. used X-ray computed tomography for imaging the pore structure and calculating the tortuosity and porosity.¹⁰⁶ The structural parameters of two commercial TFC RO membrane support layers were measured using imaging (X-ray microscopy (XRM)) and analytical (mercury intrusion porosimetry (MIP)) methods. The structural parameters found using XRM images and MIP were compared with the ones obtained from the conventional model using empirical data. The structural parameters obtained using the different techniques were found to differ significantly, which indicates that revisions are required to the customary methods of characterizing membrane structures. The advantage of XRM over other techniques was its ability to calculate the porosity of the membrane as a function of depth. The tests which have been carried out to evaluate the performance of recently developed RO membrane materials and the achievements reported from these tests are generally summarized in Table 1.

Table 1 Performance tests of recently developed RO membrane materials and reported achievements

Performance test	Notable achievements	Reference
CA hollow fiber	Removal of micropollutants or TrOCs such as charged N-nitrosamines	69
CA with HBPE	Enhancement of membrane filtration properties, wettability, thermal behavior and hydrophilicity	72
CA with PVDF and TiO2	Enhancement of membrane hydrophilicity, sodium sulfate and TOC rejection, anti-bacterial, and anti-biofouling properties	73
CA with HBPE, PEG, PVDF, or PVA	Better productivity and higher fouling resistance	40, 54 and 72–74
CA with TiO2 nanoparticles	Mechanical strength improvement	73
CA with PVA	Addition of large pores on the skin layer and increase in water permeability of membrane	74
Spiral wound PA nanocomposite	Rejection of low molecular weight TrOCs	70
TFC for fluorinated surfactant rejection from fire-fighting water	More than 99.4–99.9% rejection with consistent flux of 0.5 L m^−2 h^−1 bar^−1	85
TFC with o-ABA–TEA salt	High salt rejection of 99.41% and water flux of 1.81 m^3 m^−2 d^−1 for seawater desalination	98
Incorporation of isocyanato group into polyacyl chloride during TFC fabrication	Improvement in the water permeability, surface smoothness and hydrophilicity of membrane but reduction in chlorine stability	99
Incorporation of chloroformyloxy group into polyacyl chloride during TFC fabrication	Increase in membrane salt rejection and surface roughness but decrease in water permeability	99
TFC with acyl chloride monomers such as BPAC, BTAC, and BHAC	Enhancement of water flux	103
TFC with PAMAM	Enhancement of water flux	100
Partial replacement of conventional MPD monomer DAHP in TFC	Increase in membrane hydrophilicity and water flux	102
TFC with NaY zeolite	High flux of 50.6 gfd	31
TFC with silica nanoparticles	Biocidal properties imparted on the membranes	19
TFC with amine groups (PAMAM, L-DOPA amino acid, urethane, DMMPD, PNIPAm)	Enhancement of water flux, salt rejection, and anti-biofouling property	100 and 107–111
TFC with acrylates (PFDA, HEMA)	Antifouling enhancement	112 and 113
TFC with barium sulfate	Antifouling enhancement	114
TFC with SA and QA post-treatment	Antifouling enhancement	115
Phase inversion of BPS solutions	Low salt passage (0.34–1.81%) and high water flux (2.5–20 LMH) during brackish water desalination	81

---

3.2 Fouling studies, mitigation, and detection

3.2.1 Detection and monitoring of membrane fouling. Fouling remains a serious drawback for membrane application to RO desalination.^6,116–120 Fouling increases the operating cost of the RO process and reduces the lifetime of RO membranes.^116,118,121 Recently, biofouling has been proven to cause some of the most serious limitations of membrane performance, most especially during commercial water desalination.^{117,119,120,122} The inability to detect biofouling early enough for quick control may lead to even worse deleterious effects on membranes. A recent investigation on biofouling detection through the use of Earth’s Field Nuclear Magnetic Resonance (EF NMR) suggested that early biofouling detection is possible. This work involved the analysis of the feasibility of using EF NMR for non-destructive biofouling detection at its onset within a commercial RO spiral-wound membrane module.¹²³ EF NMR is an appropriate technique for early biofouling detection as it can assess the existence of biofouling very quickly by evaluating the change in total EF NMR signals caused by variations in the fluid flow field due to biofouling within the membrane module. Likewise, the EF NMR measurement technique can be used for high operational flow rates and pressures, and can be located anywhere along the RO membrane module. In a related study, an online feed fouling monitor (FFM) with a salt tracer response technique (STRT) has been investigated for the prediction and monitoring of RO membrane fouling under steady flux filtration.¹²⁴ The foulant loads were determined using the FFM, whereas the CP development was measured using the STRT to evaluate the contribution of cake-enhanced osmotic pressure (CEOP). The RO fouling trends were predicted using a developed model that included both the CEOP effect and cake resistance because of cake formation. The model foulants that were used to represent inorganic and organic fouling were colloidal silica and humic acid, respectively, in sodium chloride solutions. The FFM fouling and RO experiments were run at varying constant fluxes at a cross-flow velocity of 0.1 m s⁻¹ and with the same feed solutions. It was observed that higher fluxes augmented the fouling rate. Moreover, the higher flux significantly increased the cake resistance contribution to RO fouling, which was more apparent for the experiments using humic acid as the model foulant. Dead end filtration was used to study the impact of salt concentration on the specific cake resistance. It was found that the cake layer’s specific cake resistance increased for humic acid over a wide range of sodium chloride concentrations due to a decline in the particle–particle repulsion and an alteration in the packing of the foulant layer when the salt concentration was increased. The results demonstrated that FFM can be used together with STRT to reasonably assess the trends of RO fouling for both humic acid and colloidal silica. Furthermore, the substantial influence of CEOP on RO fouling can be observed from the predicted profiles of transmembrane pressure.

In a recent study, diffusiophoresis has also been shown to contribute considerably to fouling mechanisms in low salinity RO systems.¹²⁵ This study showed through experimentation and modeling that diffusiophoresis is a colloidal fouling mechanism in RO which significantly adds to cake formation and particle formation on RO membranes. The particle flux decline varied for the different salts tested in this study. It was also observed that cake-enhanced CP increased particulate fouling over time in a positive feedback loop resulting from diffusiophoresis. Improved fouling control has also been ensured through the redesign of the spacer in the RO membrane module configuration. Koutsou and Karabelas studied a new retentate-spacer with a net-type structure that has parallelogram unit cells formed from symmetrically joining spherical nodes with cylindrical filaments.¹²⁶ The most important characteristic of the proposed configuration was the presence of small contact regions (contact points) between the nodes and the bounding spiral wound membranes that prevented dead flow zones. Numerical simulations of mass transfer and flow field were conducted directly in detail within narrow channels. It was observed from the results of numerical simulation that the newly developed spacers helped to alleviate the fouling phenomena and reduce CP in the membrane process.

The possibility of early detection of fouling in hybrid polymer membranes would reduce the vulnerability of the membranes to a short life span. Electrical impedance spectroscopy (EIS) is a recent method that has been tested for real time monitoring of foulants in RO systems.^127,128 EIS has been employed for real time monitoring of electrical properties and early detection of calcium sulfate scale formation in RO systems on a lab scale.¹²⁸ The laboratory experiments were performed using commercial PA TFC membranes from Dow (BW30-FR Filmtec). The EIS signals and conductance obtained at a frequency range of 10⁻¹ to 10⁵ Hz were associated with a permeate flux decline for batch and recirculation modes. It was observed that the rate of change of conductance was greater than the permeate flux decline. For the experiments conducted in recirculation mode, a significant change in the value of capacitance was observed at a frequency of 73 Hz. For the batch mode experiments, significant changes were observed at all frequencies but the change in conductance at a frequency of 38 Hz was found to be the most effective and it corresponded to the coating layer on the membrane surface. Within the specified range of frequency, the proposed scale monitoring method was found to be effective in detecting scale formation at its initial stage before observing any considerable decline in permeate flux. Hence, EIS could be attached to a full scale RO module as a side stream for detecting scale formation before any noticeable decline in permeate flux can occur. It is also possible to comprehend the separation performance of membranes by characterizing the membranes’ electrical surface properties.¹²⁹ In one study, experimental measurements of streaming current were performed in the tangential mode for several RO membranes in an inert nitrogen environment, to provide a controlled environment, and with degassed background solutions.¹³⁰ This was done to obtain precise electrokinetic measurements even within the basic pH range. It was first demonstrated in this work that the streaming current technique is an effective method to reveal the existence of additional coating layers on commercial PA TFC membrane surfaces. It was further observed that conducting the experiments under a controlled atmosphere was much better than the alternative attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) technique, as it allowed clear differentiation between a coated and uncoated membrane surface. This is due to the sensitivity of the streaming current measurements to the ionization of the functional groups that are found on membrane’s coating layer. Moreover, the experiments that were conducted in the presence and absence of a nitrogen environment showed specific adsorption of carbonate and bicarbonate ions on the RO PA membrane surfaces.

Lutskiy et al. described a robust assay for quantifying biofilm growth on the surface of different types of PA TFC RO membranes in the initial stage.¹³¹ The procedure involved the attachment of bacteria on the PA TFC membrane, biofilm development on the membrane surface, quantitative estimation of bacteria, and detachment of bacteria from the membrane surface through 30–60 seconds of sonication. Bacterial growth and adhesion on the membrane surfaces were investigated with respect to time under static conditions, along with continual elimination of planktonic bacteria. The proposed assay gave a more quantitative result compared to using fluorescence microscopy for observing bacterial growth. The extent of biofilm growth on these membranes was found to be similar. The proposed assay can be used to predict the vulnerability of RO membranes to bacterial adhesion and biofilm growth. This assay can also be used to screen microbial and biofilm resistant membrane coatings. Additionally, a deeper understanding of the procedure of surface tethered antimicrobial agents can be obtained from this assay.

3.2.2 Dependence of membrane fouling on the RO operating conditions. The RO operating conditions are known to influence membrane fouling.^132–135 The dependence of filtration resistance on the RO operating conditions using a wide range of polymeric membranes has also been investigated. This work was a study of the performance of RO membranes for inhibitor separation from sugars, and for concentrating sugars from a biomass hydrolysate to increase the performance efficiency.¹³² A variety of commercial RO membranes with varying pore sizes, porosity and membrane chemistrya were tested for their performance based on inhibitor separation, permeate flux, sugar yield and ease of cleaning. The materials of these members include PA (from Dow and Toray), thin film (from GE Osmonics), and PA TFC (from Koch). The performance of the different membranes tested varied and the yield of sugar in the retentate ranged from under 20% to approximately 100%. In general, lower sugar yields resulted from greater inhibitor separation factors. Moreover, a reduction in permeate flux mainly occurred due to reversible fouling, and from increased osmotic pressure caused by the retention of soluble compounds. It was observed that greater porosity, higher hydrophilicity and lower surface roughness of membranes resulted in reduced membrane surface fouling and a sustained elevated permeate flux for the RO membranes during the processing of enzymatic hydrolysate. In a related study, Nguyen et al. screened some commercial FAPA RO membranes on a flat sheet to remove toxic compounds from lignocellulosic hydrolysates and prevent inhibition of the fermentation process.¹³³ These membranes included CPA2, CPA3, and ESPA2 (from Hydranautics), XLE Filmtec (from Dow), and SG (from GE Osmonics). These membranes were tested for their capacity to isolate C5 and C6 sugars from a model solution containing furfural, acetic acid, vanillin and 5-hydroxymethyl furfural. The highest sugar rejection obtained was greater than 97% but the transmission of inhibitors was found to be low.

The fouling behavior of gypsum (CaSO₄·2H₂O) scalants and sodium alginate on PES TFC membranes under different operating conditions has been investigated in several studies.¹³⁴ Gypsum and sodium alginate are two of the most critical foulants hindering membrane performance and causing a membrane flux decline. The effect of the applied hydraulic pressure (up to 18 bars) on the extent of membrane fouling was analyzed using SEM. It was observed that the alginate fouling was minimal in the absence of applied pressure but increased with higher hydraulic pressures. However, gypsum scaling was observed in the absence of applied pressure and was negligible between 8 bars and 18 bars. Combined fouling was severe with the co-existence of gypsum crystals and alginate in the absence of applied pressure. The study concluded that the removal of alginate-type foulants from the feed water stream may become essential for the success of PES TFC membranes at high pressures. Khan et al. conducted a comparative investigation on fouled PA TFC RO membranes that were collected from seawater RO pilot plants to study the influence of the operating parameters and water quality on membrane fouling.¹³⁵ The pilot plants were situated in Vilaseca (East Spain) and used UF as a pretreatment method. From the results obtained using chemical, ultrastructural and microbiological analyses on fouling layers, it was observed that the seawater RO train was greatly affected by bio/organic fouling both at the lead and terminal position elements. It was discovered that the microbial population causing bio/organic fouling largely comprised alpha and gamma proteobacteria, whereas the inorganic fouling layer mostly comprised calcium, sulfur and iron elements. It was verified from these results that the composition of the RO fouling layer is strongly dependent on the water quality of the source.

3.2.3 Dependence of membrane fouling on the membrane surface properties. For PA membranes, the chemical structure of the membrane surface might play a key role in determining the fouling propensity of these membranes.¹³⁶ The chemical heterogeneity of the surface of PA RO membranes and its effect on membrane fouling has been studied using AFM.¹³⁷ The AFM colloidal probes were made using positively charged aliphatic amine latex (AAL) and negatively charged carboxyl modified latex (CML). As a result of the negatively charged membrane surfaces, the adhesive forces were observed to be considerably stronger with the AAL probe compared to the CML probe. It was further observed that for the RO membranes having similar average surface properties, it was possible for them to have surfaces that are chemically heterogeneous. The results obtained from fouling experiments conducted using two commercial RO membranes with different chemical heterogeneities showed that fouling was more prominent for membranes with higher chemical heterogeneity. This verified that the fouling of PA RO membranes is dependent on the surface charge distribution. Hence, the surface chemical heterogeneity can be used as an important parameter for synthesizing RO membranes that are resistant to fouling. Wu et al. studied the impact of specific surface functional groups found on a common PA RO membrane surface on alginate membrane fouling under solution conditions relevant to seawater desalination.¹³⁸ Self-assembled monolayers (SAM) with desired terminal functional groups such as –COOH, –NH₂, –CONH₂ and –OH were prepared and alginate adsorption and removal was studied using a quartz crystal microbalance with dissipation monitoring. The initial adsorption rate and the reversibility of the adsorbed layer were studied. It was discovered that the –NH₂ group has a strong affinity for alginate; however, the –COOH group showed the highest adsorption rate for simulated seawater. Furthermore, the reversibility was shown at high ionic strength and calcium ion concentration where alginate–alginate interactions and alginate aggregation played an important role for its adsorption behavior. Specific functional groups seemed to have no significant influence on the equilibrium adsorption of alginate; however, it would be interesting to further investigate the role of surface heterogeneity in membrane fouling. Biofouling in PA RO membranes has also been controlled by immobilizing the membranes with lysozyme, which is an antibacterial enzyme.¹³⁹ The antibacterial enzyme was immobilized on the PA layer of RO membranes through 6-amino caproic acid molecules by using the method of IP and amine coupling. An increase in the surface density of the immobilized lysozyme was observed (Fig. 8) as the concentration of 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide and N-hydroxysuccinimide was increased in the amine coupling reaction. The highest surface density of the immobilized lysozyme was found to be around 0.1 μg cm⁻². The modified membranes caused a reduction in water permeability but sustained the same salt rejection ratio as the original PA RO membrane. Moreover, it was found from biofouling experiments that when the fabricated antibacterial PA RO membrane was exposed to the feed solution containing bacteria, it efficiently inhibited biofilm formation which could have resulted in adverse biofouling.

Fig. 8 Analyzed images of unmodified and lysozyme-immobilized membranes from COMSTAT software.¹³⁹

PA RO membranes with high chlorine tolerance properties and good permeation have also been prepared on microporous polysulfone membrane supports via conventional IP of different diamine compounds with TMC or isophthaloyl chloride.¹⁴⁰ The diamine compounds used for polymer preparation were metaphenylene diamine, 2,4-diaminotoluene, 2,6-diaminotoluene, 3,4-diaminotoluene, N,N′-diphenylethyldiamine, 4-chlorometaphenylene diamine and 2,4-diaminoanisol. All the polymers prepared using the different diamine compounds were tested for their hydrophilicity and chlorine tolerance. Of the diamines, only metaphenylene diamine and 2,6-diaminotoluene were used for the preparation of TFC RO membranes via IP with trimesoyl chloride. It was observed from the chlorine tolerance that the TFC membranes synthesized from 2,6-diaminotoluene demonstrated better chlorine tolerance compared to those synthesized from metaphenylene diamine as an amine monomer. However, performance tests showed that the water permeability and salt rejection of both the membranes were similar.

The surface composition and properties of hybrid polymer membranes are crucial determining factors of the fouling propensities of the membranes.^141–145 Initial biofouling in polymeric composites could be attributed to the presence of defects on the membrane surface.¹⁴⁶ Up to 30% of the membrane surface could be covered by bacteria due to defects in the membrane surface.¹⁴⁶ The areas of heterogeneity on the membrane surface can provide niches for promoting biofilm growth. Therefore, the application of a conditioning film can be adopted in order to reduce the impact of the membrane heterogeneity on the RO membrane performance. Although there are some reports that membrane surface properties such as hydrophilicity, surface charge, and roughness could affect biofilm formation, Baek et al. argued that biofilm formation and flux reduction are experienced despite differences in the surface properties.¹⁴⁷ This observation suggests that the current research interest directed towards the manipulation of membrane surface conditions to reduce biofouling needs to be reconsidered. The mitigation of biofilm formation through the addition of anti-biofouling membrane surfaces or effective pretreatment might prove to be more worthwhile.

The influence of the surface features of four commercial PA composite RO membranes on membrane fouling has been investigated in a recent study.¹⁴⁸ These membranes included SW22 (from Vontron Technology Inc.), SW30 (from Film Tec Inc.), SWC1 (from Hydranautics), and TM810 (from Toray Bluestar Membrane Inc.). The surface features and microstructures of these commercial RO membranes were characterized and investigated for seawater desalination. It was found from experimental results that only SW22 and SW30 had hydrophilic coatings. It was further observed that the main role in the desalting process was played by the mid-section of the separation layer, which differed for each membrane because of the different approaches for the membrane synthesis. The RO membranes were also tested for their resistance to fouling using aqueous solutions containing different foulants with a sodium chloride concentration of 32 000 ppm. It was discovered that SW22 had the highest resistance to fouling from humic acid due to the existence of substantial hydrophilic groups on the surface of the membranes. SWC1 strongly resisted fouling from silica colloids and BSA because the membrane was negatively charged. The antifouling properties of the RO membranes greatly depended on their surface features like the surface charge properties, functional groups and surface roughness. A smooth membrane surface would improve the anti-fouling properties of the membranes by hindering the foulants’ deposition on membranes under hydraulic shear forces during the desalting process.¹⁴⁹ Furthermore, the fouling mechanism of BSA and humic acid was observed to vary significantly from that of silica colloids. Another work was an investigation of the modification of a PA RO membrane with L-3,4-dihydroxyphenylalanine (L-DOPA) amino acid for minimizing membrane fouling.¹⁰⁷ This study presented a comparison between a virgin PA membrane and an L-DOPA modified PA membrane. The effects of this modification on these membranes and fouling agents in terms of physicochemical interactions were examined. The extended Derjaguin–Landau–Verwey–Overbeek (XDLVO) theory and the calculation of adhesion free energies were used to show the impact of the fouling agents (BSA and alginate) on the membranes. The results obtained showed that the adhesion of both BSA and alginate on the membranes was mainly due to acid–base interactions. L-DOPA seemed to affect the acid–base component of the adhesion free energy for the modified membrane. The results indicated that the modification of the virgin PA membrane with L-DOPA could restrict the membrane fouling propensity. However, L-DOPA needs to be further explored for industrial and commercial anti-fouling processes.

PA-urethane composites form another class of RO membranes that have been investigated lately because of their improved mechanical properties.^108,109 An attempt has also been made to endow antifouling and chlorine resistance properties on PA-urethane TFC RO membranes by modifying the membranes with a functional monomer. Liu et al. has fabricated a PA-urethane RO composite membrane by a two-step IP technique using dimethyl-m-phenylenediamine (DMMPD) as the main functional monomer on a polysulfone support layer.¹⁰⁸ The modified RO TFC membranes, poly(amide-urethane@imide), were more chlorine resistant due to the inclusion of an ultrathin polyimide film on the outermost surface of the synthesized TFC RO membrane. The chlorine tolerant properties of the poly(amide-urethane@imide) membranes were verified by a combination of XPS and infrared (IR) analyses. In addition, the modified membranes prepared via the two-step IP technique demonstrated enhanced antifouling properties, as they possessed much smoother and more hydrophilic surfaces than the conventional poly(amide/imide-urethane) membranes that are prepared by a one-step IP technique. However, the water flux and salt rejection of the poly(amide-urethane@imide) were marginally lower than the conventional poly(amide/imide-urethane) membrane and the commercial PA membrane. A PA-urethane TFC RO membrane has been fabricated in another study for seawater desalination via IP of MPD and CFIC.¹⁰⁹ However, in this study, only the membrane performance enhancement was investigated. The impacts of utilizing mixed cross-linking agents, choosing organic solvents of varying diffusivity and/or solubility for MPD, changing the temperature of the organic solvent, implementing a modified curing process, and post-treatment on the performance of the modified TFC RO membrane were investigated. The experimental results demonstrated that water flux through the membrane improved effectively. The salt rejection remained unchanged or increased under the following conditions: addition of diacyl chloride, selection of an organic solvent with higher amine diffusivity, synthesis of a membrane at a reasonably higher temperature, curing the nascent membrane via a two-stage curing process, soaking the membrane in aqueous hypochlorite solution. The modified membrane’s water flux increased from around 35 LMH to approximately 42 LMH for seawater desalination and the salt rejection was always observed to be greater than 99.4% for a feed solution containing 3.5 wt% sodium chloride at a pressure of 5.5 MPa. Moreover, the PA–urethane TFC RO membrane with the most efficient performance comprised a distinct cross-linked selective skin layer and a comparatively hydrophobic surface. This study proved that the modification of the fabrication and process conditions for the PA-urethane TFC RO membrane could improve the membrane performance in addition to its antifouling benefits.¹⁰⁹

Zwitterionic polymers have also been reported and confirmed to exhibit antifouling properties.^110,150,151 Choi et al. prepared and applied amphiphilic zwitterionic carboxylated polyethyeleneimine (carboxylated PEI) and comb-like hydroxyl poly(oxyethylene) methacrylate homopolymer (HPOEM) anti-fouling layers on the surface of commercial seawater RO membranes to enhance the fouling resistance of these membranes.¹⁵¹ No significant changes occurred to the morphologies of the modified membranes because of the preparation of the ultrathin coating layer under mild conditions. The performances of the modified membranes were tested in both seawater and brackish water conditions. Fouling tests were conducted on the modified membranes using BSA and alginate as model foulants. Fouling tests at high salt concentrations revealed a salting-in effect for the membranes coated by zwitterionic carboxylated PEI, and a salting-out effect for the membranes coated by HPOEM. The zwitterionic carboxylated PEI coated membrane demonstrated a lower affinity for deionized water than for sodium chloride solution, and showed better fouling resistance in seawater conditions. Hence, an efficient fouling resistant layer made from zwitterionic materials can be proposed for seawater desalination. The HPOEM coated membranes also showed improved fouling resistance under the conditions of brackish water. Meng et al. attempted to minimize fouling on a commercial PA TFC RO membrane by grafting a salt-responsive zwitterionic polymer poly(4-(2-sulfoethyl)-1-(4-vinylbenzyl)pyridinium betaine) (PSVBP) on the membrane.¹¹⁰ The investigation of the synthesized membrane revealed that the PSVBP grafting enhanced the membrane hydrophilicity and caused the membrane surface to be negatively charged. It was observed from RO performance tests that the PSVBP grafting resulted in an increase in salt rejection from 98.0 to 99.7%; however, the permeation flux decreased by 20%. Additionally, the fabricated membrane was found to have good antifouling properties over the short term, and was able to recover 90% of the initial permeate flux after being rinsed with brine. The enhanced antifouling properties of the membrane resulted from the suppression of hydrophobic–hydrophobic interactions and negatively charged organic foulants in the surface water. Moreover, the PSVBP membrane’s salt-responsive properties provided the driving force for easy elimination of protein foulants from the membrane surface to improve its cleaning efficiency, as shown in Fig. 9.

Fig. 9 Synthesis and easy cleaning of a salt-responsive PA TFC membrane.¹¹⁰

The imparting of antifouling properties on a commercial PA TFC RO membrane (LE-400 Film Tec from Dow) through membrane surface modification has also been investigated by Nikkola et al.¹⁵² In this study, the surface properties of the membrane were modified by coating the membrane surface via an inorganic trimethylaluminium-based atomic layer deposition (ALD) process. The processing temperatures and number of ALD cycles were varied systematically and observed to influence membrane surface properties such as surface polarity, hydrophilicity and surface roughness. The ALD coatings were found to enhance the antibacterial and antifouling properties of the RO membrane. Among all the ALD-coated membranes produced, the colony forming unit (CFU) test results showed that the lowest bacterial adhesion took place on the membrane with the most polar and hydrophilic surface. In addition, the RO membrane performance in terms of water permeability, salt permeability and salt rejection was observed to be affected by the processing temperatures and the number of ALD cycles. Generally, less ALD cycles and low temperature contributed to the improved RO performance of the ALD-coated membranes in terms of the permeability and antifouling properties. The grafting of imidazolidinyl urea (IU) on a commercial aromatic PA RO membrane (RE4021-TL from Woongjin Chemical Co., Korea) using the carbodiimide induced method has also been shown to be a way to improve the anti-biofouling and chlorine resistant properties of the membrane.¹¹¹ The grafted IU worked like an anti-microbial agent to enhance the anti-biofouling properties of the membrane surfaces, and worked as the precursor of N-halamine to provide lasting and regenerable anti-biofouling and chlorine resistant properties to the membrane. An improved hydrophilicity was observed for the modified membrane and the membrane surface’s isoelectric point was found to shift towards a higher value. No changes were observed in the morphology and roughness of the membrane surface. Compared to pristine PA RO membranes, which do not have anti-biofouling properties, the IU modified membrane demonstrated very high anti-biofouling properties and eliminated all the possible E. coli cells from the IU-modified membrane surface. The water flux of the pristine membrane was almost negligible in the case of severe biofouling, whereas the salt rejection and water flux of the IU-modified membrane decreased by 2.7 and 16.0%, respectively. Additionally, the IU-modified membrane’s anti-biofouling properties were observed to improve considerably after chlorination because the grafted IU on the surface of the membrane worked as sacrificial pendant groups to avoid the chlorination of PA. Moreover, the IU-modified membrane’s anti-biofouling and the chlorine resistant properties retained a high regeneration capacity due to reversible transition between the IU and chlorinated IU. Furthermore, results from the long term investigation of this method demonstrated that the lifetime of the commercial RO membrane can be increased for practical applications with the inclusion of IU modification along with occasional free chlorine pretreatment in the RO systems.

The deposition of N-isopropylacrylamide-co-acrylic acid copolymers (P(NIPAm-co-AAc)) on the surface of a PA TFC RO membrane is another way to improve the antifouling properties of the membrane. Yu et al. has modified the surface of interfacially prepared commercial aromatic PA TFC RO membranes by depositing P(NIPAm-co-AAc) on the membrane surface.¹⁵³ Free radical copolymerization was used to prepare P(NIPAm-co-AAc) copolymers and varying concentrations of their aqueous solutions were used for PA membrane modification through sedentary surface coating technique. Fouling experiments using an aqueous solution of BSA demonstrated that the modified membrane had significantly higher fouling resistance to BSA compared to unmodified membranes due to the increased hydrophilicity and negative charge of the membrane surface at neutral pH. It was revealed from the washing experiments using deionized water that the P(NIPAm-co-AAc) coating layer enhanced the cleaning efficiency of the membrane. Furthermore, the elimination of foulants situated on the membrane surface would be possible when the coating layer’s phase transition occurs at temperatures higher than its low critical solution temperature. The membrane surface became more hydrophilic and the surface charge increased at neutral pH due to hydrophilic P(NIPAm-co-AAc) layer deposition; but the water permeation resistance increased. It was observed from permeation tests that the sodium chloride and sodium sulfate permeability decreased under neutral and alkaline conditions due to membrane modification. The decline in permeability was higher for sodium sulfate (divalent anion) compared to sodium chloride (monovalent anions).

The surface modification of commercial RO membranes by the deposition of ultrathin copolymer films on the membrane surfaces could also be used to promote biofouling resistance. In a recent study, four commercial seawater RO membranes made from PA were tested: the defunct 80B (from Toray), AD (from GE Osmonics), and SW30HR (from Dow Film Tec).¹¹² The initiated Chemical Vapor Deposition (iCVD) technique was used for the direct deposition of the two copolymerized monomers (perfluorodecyl acrylate (PFDA) and hydroxyethyl methacrylate (HEMA)) with tert-butyl-peroxide as the initiator, forming an amphiphilic surface measured by the contact angles. Several analyses were used to confirm the presence and smoothness of the film on the membrane surface, such as FTIR, XPS and AFM. Short-term permeation tests were conducted to show the impact of this coating on the salt rejection. Analysis of the membranes showed that the surface was smooth and that the thin film was well-deposited on the membrane surface. The permeation tests revealed that the film was as permeable as the PA layer and that it would not cause a significant decline in water flux. Moreover, this deposition technique could be used to prevent biofouling because the bacterial composition around the membrane surfaces would be hampered by their amphiphilic chemistry, rather than hydrophobic chemistry. However, further studies are needed to evaluate the long-term viability of this approach. In a related study, commercial PA TFC RO membranes were modified by depositing copolymer films on the membrane surfaces to decrease bacterial adhesion and control biofouling.¹¹³ The iCVD technique was used to deposit the copolymerized hydrophobic PFDA and hydrophilic HEMA monomers on several substrates. The thickness and chemistry of the deposited films were controlled by regulating the comparative gas flow rates of the two monomers. It was observed from Quartz Crystal Microscopy (QCM) that the amphiphilic chemistry copolymer film exhibited relatively less foulant adsorption than the pure homopolymers. E. coli cells were used for short-term batch studies to test for the coated and virgin membranes’ resistance to microbial adhesion. The minimum bacterial adhesion was observed for the copolymer film membrane with amphiphilic chemistry (Fig. 10). Moreover, it was discovered from the quantitative analysis of the optimized modified membrane and virgin membrane that the existence of the amphiphilic copolymer film on the membrane surface reduces the bacterial adhesion by around 100 times. The copolymer films can therefore be used as antifouling coatings on RO membranes as they can effectively prevent bacterial adhesion and foulant adsorption.

Fig. 10 Number of E. coli cells attachments to amphiphilic and virgin PA TFC membranes for images across the membrane surfaces.¹¹³

McVerry et al. proposed the modification of commercial PA TFC RO membranes using the perfluorophenyl azide (PFPA) photochemistry method to enhance their anti-fouling properties in addition to retaining the scalability and ensuring a roll-to-roll manufacturing process for RO membranes.¹⁵⁴ The commercial RO membranes were first immersed into an aqueous solution containing PFPA-terminated poly(ethyleneglycol) (PEG) species, followed by UV light exposure under ambient conditions. Contact angle measurements and ATR-IR spectroscopy confirmed the effective covalent modification of the RO membranes. Moreover, XPS demonstrated that the PFPAs went through nitrene addition generated by UV and were attached to the RO membrane via aziridine linkage. Furthermore, the results from fouling experiments revealed that the RO membranes modified with PFPA-PEG derivatives had high resistance to fouling. However, RO performance tests demonstrated that hydrophilic polymers on the modified membrane surfaces reduced the water permeability and increased the salt (sodium chloride) rejection compared to the original commercial RO membranes because of steric hindrance. When the molecular weight of hydrophilic PEG brush polymers was systematically increased, the water flux and salt rejection of the modified membranes were also significantly affected because of bigger flexible polymer chains. Despite a reduction in the initial water flux of PFPA-PEG modified membranes, the modified membranes provided higher water fluxes than most commercial RO membranes with a similar sodium chloride salt rejection. The anti-fouling experiments on the modified and unmodified membranes were conducted by fluorescent microscopy using E. coli as the model bacteria. It was revealed that the bacterial adhesion on the modified membranes was significantly less than for the unmodified membranes. Moreover, the anti-fouling resistance of the modified RO membranes improved with increasing PEG molecular weight. Overall, the production of anti-fouling RO membranes can reduce the energy and high maintenance costs associated with RO desalination by eliminating biofouling on the membrane surface.

The enhancement of the antifouling properties of commercial PA TFC RO membranes has also been investigated in a recent study by dip-coating a synthesized terpolymer on the membrane surface.¹⁵⁵ In this work, the novel terpolymer poly(methylacryloxyethyldimethyl benzyl ammonium chloride-r-acrylamide-r-2-hydroxyethyl methacrylate) [P(MDBAC-r-Am-r-HEMA)] was prepared by free radical polymerization. The synthesized terpolymer was dip-coated and cross-linked on the surface of commercial TFC PA RO membranes (LCLE and BW30 Film Tec from Dow) to enhance their antimicrobial properties and chlorine resistance (Fig. 11). The chlorine resistance of the coated membranes was investigated via cross-flow filtration of sodium hypochlorite solution, and the membranes were observed to have a tolerance to chlorine exposure of greater than 16 000 ppm, which was 7 to 8 times higher than that of uncoated membranes. Additionally, the antifouling performance of the coated membranes was assessed via protein solution cross-flow filtration and cell-culture experiments. It was found that the coated membranes were able to efficiently maintain a stable permeate flux under protein filtration and reduced bacterial growth was observed on the coated membrane surfaces. This was because the membrane surface coating served as a sacrificial and protective layer in protecting the PA film from chlorine attack. The PMDBAC and polyacrylamide components on the membrane coating material were found to be vital for enhancing the antimicrobial properties, and the heightened surface hydrophilicity of the membranes augmented their antifouling performance. Moreover, a modest salt rejection and permeate flux were shown for the coated membrane through performance tests.

Fig. 11 Schematic of (a) terpolymer P(MDBAC-r-Am-r-HEMA) and (b) surface-modified RO membrane.¹⁵⁵

A novel surface mineralization of a commercial PA TFC RO membrane has proven to be an effective antifouling technique lately.¹¹⁴ The aim of this work was to enhance the performance and anti-fouling properties of the commercial PA TFC RO membrane by depositing a barium sulfate-based mineral coating on the membrane surface through alternating soaking of the membrane in aqueous solutions of barium chloride and sodium sulfate. The degree of mineralization on the PA RO membranes was varied by changing the number of alternating soaking process cycles. The characterization of the modified membrane revealed that the mineral coating was evenly dispersed on the membrane surface. Additionally, it was observed that the mineralized membrane surface was more negatively charged and hydrophilic than the original membrane. Moreover, the membrane demonstrated an increased salt rejection and pure water permeability simultaneously under the specific conditions of modification. The improved water permeability resulted from increased repulsion forces between the surface of the modified TFC PA RO membrane and the chloride ions. The modified membrane’s permeation performance could also be regulated by changing the number of the alternating soaking process cycles. The fouling experiment utilized BSA as a model foulant, and it was found that the mineralized membrane was more resistant to BSA fouling. The enhanced fouling resistance was a result of the barium sulfate-based surface coating layer which efficiently alleviated the foulant molecules’ deposition and adsorption on the surface of the membrane by reducing hydrophobic interactions and improving the electrostatic repulsion between the membrane surface and the BSA molecules. Furthermore, the mineralized membrane could be cleaned easily by simple hydraulic washing.

3.2.4 Antiscalants, chemical post-treatments, and their effects. Antiscalants have been used widely for fouling prevention and control in RO desalination for decades. However, a recent study has suggested that antiscalants may not be appropriate for the long term sustainability of fouling mitigation in RO membranes due to their side effects. Sweity et al. investigated the side effects of two polyphosphonate and polyacrylate-based antiscalants on the biofouling of ESPA-2 RO membranes (a spiral wound composite PA membrane from Hydranautics) using brackish water for the RO desalination process.¹⁵⁶ Pseudomonas fluorescens was deposited on the membrane and fed with brackish water with and without antiscalants. The nutritional contribution of antiscalants to the biofilm was investigated by growing the biofilms in packed-bed biofilm reactors using cross-flow RO filtration. The RO membrane modules were tested in industrial and pilot desalination plants. It was shown that antiscalants would enhance the RO membrane biofouling and play a role in increasing the flux decline and salt passage. Polyacrylate-based antiscalants enhanced membrane biofouling by altering the surface physicochemical properties of the membrane to promote the initial attachment and deposition of the bacterial cells. Polyphosphonate-based antiscalants also contributed to membrane biofouling by serving as sources of nutrients, especially phosphorous, for the bacterial cells. It was concluded that the selection of the type and dosage of antiscalants should be taken into account when determining the associated contribution of antiscalants to the membrane’s biofouling propensity. Another recent study has reported the impact of preservative chemicals on the performance of a PA RO membrane with respect to water flux and solute rejection.¹⁵⁷ To study the effect of preservatives on the membrane, a cross-flow membrane filtration system was used to test the membrane performance and then the membrane was submerged at pH 3 and 7. The preservative solutions selected for this research were formaldehyde, sodium metabisulfite, and 2,2-dibromo-3-nitrilopropionamide. Water permeability and salt rejection of the PA membrane were tested before and after chemical preservation. The results obtained showed that chemical preservation may alter the membrane surface properties (i.e. surface charge and hydrophobicity) and consequently result in a negative impact on the water permeability and solute rejection. It was also shown that the pH of the solution in the system and the preservatives has a direct impact on the membrane performance. The results suggested that the adverse effect resulting from the use of formaldehyde and sodium metabisulfite as preservative chemicals can be minimized if the solution is maintained at near-neutral pH. It was concluded that the choice of chemicals for membrane preservation may have a negative impact on water permeability and solute rejection. Hence, it is important to properly select both the preservatives and the solution pH.

Post-treatments could also be employed to enhance the overall effectiveness of RO membranes.¹⁵⁸ The post-treatment of a PA TFC RO membrane with salicylaldehyde (SA) and quaternary ammonium cations (QACs) could also improve the membrane’s performance efficiency.¹¹⁵ Zhang et al. modified the surface of a commercial aromatic PA TFC RO membrane via post-reaction with SA and QACs to improve the membrane performance for RO desalination.¹¹⁵ SA and QACs were used in the membrane modification because they are bacteria contact-killers and antimicrobial agents, and could provide the membrane with lasting anti-biofouling properties. In addition, SA is a chlorine consumer and can help to increase the chlorine resistance of the membrane, whereas QA is a hydrophilic material that can increase the hydrophilicity of the membrane. It was found from experimental results that the higher density of SA and QACs on the surface of the membrane enhanced the chlorine resistance, hydrophilicity and anti-biofouling properties of the membrane compared to nascent membranes. A negligible change in water flux occurred for the modified membrane after biofouling and chlorination. The results confirmed that the synthesized modified membrane had perfect stability in salt rejection and water flux, in addition to demonstrating strong resistance to chlorine and bacterial adhesion.

The fouling tests which have been carried out to reduce the tendency of RO membranes to foul in recent studies and the notable findings reported in these studies are summarized in Table 2 below.

Table 2 Recent RO membrane fouling tests and reported notable findings

Fouling tests	Notable findings	Reference
Variations in membrane surface chemical heterogeneity	Less fouling with lower chemical heterogeneity	137
PA immobilization with lysozyme	Increase in surface density with around 0.1 μg m^−2 as highest surface density; inhibition of biofilm formation; water permeability reduction but same salt rejection ratio	139
Variations of diamine compounds in TFC	TFC membranes synthesized from 2,6-diaminotoluene demonstrated better chlorine tolerance	140
Modification of PA with L-DOPA amino acid	Lowering of membrane fouling propensity	107
Modification of PA–urethane TFC with DMMPD	Increase in chlorine resistance	108
Modification of PA–urethane TFC with CFIC	Increase in water flux from about 35 to 42 LMH for seawater desalination	109
Zwitterionic carboxylated PEI and HPOEM layers on commercial membranes	Development of antifouling ultrathin coating layers and enhancement of fouling resistance	151
ALD coatings on commercial TFC membranes	Enhancement of the antibacterial and antifouling properties	152
Grafting of IU on commercial PA membrane	Improvement in anti-biofouling and chlorine resistant properties	111
Modification of TFC with P(NIPAm-co-AAc)	Improvement in membrane antifouling properties and increase in membrane hydrophilicity	153
Deposition of PFDA and HEMA on commercial TFC membrane	Amphiphilic surface formation and biofouling prevention	112
Modification of commercial TFC membranes through PFPA photochemistry	Enhancement of antifouling properties	154
Dip-coating of P(MDBAC-r-Am-r-HEMA) on commercial TFC membranes	Enhancement of antifouling properties and chlorine resistance	155
Deposition of barium sulfate-based mineral coating on commercial TFC membrane	Enhancement of fouling resistance and increase in salt rejection	114
Post-treatment of commercial TFC membranes using SA and QACs	Improvement of chlorine resistance, hydrophilicity and anti-biofouling properties	115

---

4. Concluding remarks

Recent evolutions into new and improved members aimed at tackling RO process constraints, such as low water flux and salt rejection, membrane fouling and regular membrane replacement, are important developments. The sustenance of research activities in this direction would help to improve the productivity of RO and preserve RO membranes for durable use. Nanomaterials including graphene, zeolites, silica, CNTs, TiO₂ and silver have gained a lot of attention recently because they exhibit excellent physicochemical properties that permit high permeability and act as antifouling agents on membrane surfaces. Graphene is ultrapermeable and the hybridization of membrane composites with graphene oxide has significantly improved membrane performance in recent times.^23,30 However, the scalability of graphene incorporation into membranes is still a challenge. Nanocomposite membranes containing graphene need to be optimized so that they can function effectively for real-world high-pressure RO applications. Currently, information about the ideal experimental conditions for graphene use in commercial RO applications in terms of the operating pressure is limited.²⁷ Zeolites also present a promising opportunity for improved RO processes in the future because of their chemical stability, affordability, and high water permeability.^10,31–33 Impressive results have been reported for the performance of hybrid membranes integrated with zeolites, such as TFN membranes incorporated with NaX, NaY, FAU, MFI, ZIF-8, LTA and aTMA zeolites.^{14,31,32,35–38} However, the use of zeolites in hybrid membranes is still not feasible for commercial application.^159,160 The incorporation of RO membranes with MWCNTs has also yielded interesting results in terms of high water permeation.^2,15,54,161 Additionally, silica nanoparticles can be included in RO composite membranes to impart biocidal properties on the membranes.^11,19,49 However, the viability of loading these nanomaterials onto RO membranes for large-scale water production is still difficult.¹⁹ Nonetheless, it is expected that recent works aimed at improving the stability and long-term effectiveness of nanocomposites for RO water production would enhance the use of these hybridized membranes in the near future.

For polymeric membranes, CA is fast diminishing in the RO market. Although CA membranes are known for their cheap price, they exhibit poor mechanical strength and low water flux because of their low hydrophilicity surface structure, as compared to PA TFC membranes.^40,62,68 Apart from these drawbacks, CA membranes show high vulnerability to temperature and chemical (acid and alkali) attack resulting from membrane hydrolysis, poor salt rejection, and high operating cost.^2,40 However, the incorporation of other polymers with CA has been shown to improve the hydrophilicity of dense CA membrane surfaces, increase the pore number, reduce the thickness of the membrane outer layer, and improve the resistance of CA membranes to chlorine attack. As a result, CA membranes might be resurrected for wide RO applications.^72–74 The incorporation of cellulose acetate with TiO₂ has also been shown to be an effective way of improving the mechanical strength of CA RO membranes.⁷³ In addition, CA membranes have higher resistance to chlorine than PA TFC membranes.² CA membranes have also been shown to exhibit a higher rejection of micropollutants and trace organic compounds from water.⁶⁹ However, despite recent improvements in membrane design and fabrication, CA-based membranes still exhibit a low salt rejection because of irregularities resulting from inherent polymer impurities in CA membranes during the fabrication process.¹⁶² The inability of CA-based membranes to maintain a high flux and high salt rejection at the same time still poses a challenge.² The development of thin film CA membranes that can exhibit a very high salt rejection and water flux could shift the RO market preferences for membrane types in the near future. Recent efforts geared towards this objective have yielded some positive outcomes.^162,163

PA TFC is currently dominating the RO world. Remarkable water permeation, enhanced salt rejection properties and better resistance to biodegradability are some of the impressive characteristics of PA TFC when compared with asymmetric CA membranes. In addition, PA TFC has been useful for the treatment of water containing surfactants, the purification of water containing radioactive substances to a particular threshold, and the treatment of saline water with a wide range of pH.^{77,85,87–89} Modification of the surface functionality of PA TFC membranes by testing other aromatic chlorides apart from TMC (or incorporating isocyanato or chloroformyloxy groups into polyacyl chloride) has enhanced either the water flux or salt rejection.^99,109 The investigation of other amine groups apart from MPD in PA TFC membranes (such as DAHP) and the post-treatment of these membranes with organic solvents (such as DMF) might also be necessary for determining the sustainability of PA TFC membranes for improved RO membrane performance.^86,102 Other attempts to include PAMAM, L-DOPA amino acid, urethane, DMMPD, PSVBP, IU, PNIPAm, PEG, PFDA, HEMA, PMDBAC, barium sulfate, SA and QAC in TFC membranes have either improved the water flux, enhanced salt rejection, increased the antifouling properties of the developed membrane, or enriched the capacity of the fabricated membrane to act as a biocide.^{100,107–112,114,153–155,158} However, operating conditions still play a vital role in the efficiency of PA TFC membranes.^90,91 Additionally, these membranes are still greatly affected by chlorine.^81,82 To ensure an enhanced water flux and continued use of PA TFC membranes, current and future research needs to be directed towards the stability of these membranes in the presence of chlorine. Also, research should be continually directed towards the determination of the SOW of these membranes for specific applications.^92–94 A membrane, RO reactor and spacer design aimed at minimizing the effect of the cross-flow velocity and dead zones would be an interesting area to look at in the near future for process optimization.

In order to further reduce the fouling propensity of recently developed membranes, the early detection of fouling mechanisms is highly necessary. Recently, some approaches have been developed to ensure rapid and online monitoring of RO membrane fouling. For example, EF NMR, FFM, STRT, EIS, and streaming current techniques have been presented and shown to be effective approaches for early prediction and real time monitoring of RO membrane fouling in order to foresee the occurrence of any noticeable fouling effect.^{123,124,128,130} The surface modification of RO membranes through enzymatic immobilization or coatings via varied ALD cycles or iCVD are other recent approaches that have shown promising results in terms of membrane fouling minimization.^{112,113,139,152} However, as a result of steric effects, reductions in water permeation or salt rejection have been reported as a consequence of some attempts which have been directed towards imparting antifouling or anti-chlorine properties on RO membranes via membrane surface modification.^{110,153,154,157} Therefore, more research is needed to ensure that improving the fouling resistance of RO membranes via the alteration of the surface characteristics is not counterproductive. In addition, improved methods are needed to characterize the membrane support structure so that properties such as porosity and tortuosity can be easily calculated and extensively visualized as an early membrane deformation detection approach.

Overall, it appears that the choice of membrane materials for future RO processes would depend largely on the desired permselectivity and the targeted foulants. For instance, research can be targeted at amine polymers that can promote the reduction of the membrane selective layer thickness or reduction of the membrane surface roughness through the creation of nodular structures so that the water permeability can be improved. In addition, the functionality of the membrane surface could be targeted if the focus is to curtail specific membrane foulants. In this case, hydrophilic groups would restrict humid acid foulants whereas negatively-charged groups would curb colloidal foulants. PSVBP grafting or (P(NIPAm-co-AAc)) copolymer incorporation could enhance both the membrane hydrophilicity and the negative charge properties of the membrane so that organic foulants can be curtailed. However, continued investigation is still required to ratify the long-term feasibility of copolymerization for RO membrane enhancement. If the focus is to prevent the deposition of foulants arising from hydraulic shear forces on the membrane surface, a smooth membrane surface would be required. Post-treatment could be necessary for membranes that exhibit strong resistance to chlorine and bacterial attack. Therefore, the peculiarities of RO operations are a strong determining factor for the type of membrane enhancement needed. However, for sustainability of the type of RO membrane desired for a specific application, both the membrane performance and antifouling features would have to be taken into consideration.

Acknowledgements

The authors wish to appreciate Masdar Institute of Science and Technology (Abu Dhabi, UAE) for providing a research platform where information on current and recent research works regarding polymeric and nano-enhanced RO membranes are accessed.

References

1. P. Zhang, P. Knötig, S. Gray and M. Duke, Scale reduction and cleaning techniques during direct contact membrane distillation of seawater reverse osmosis brine, Desalination, 2015, 374, 20–30,  DOI:10.1016/j.desal.2015.07.005.
2. S. S. Shenvi, A. M. Isloor and A. F. Ismail, A review on RO membrane technology: Developments and challenges, Desalination, 2015, 368, 10–26,  DOI:10.1016/j.desal.2014.12.042.
3. T. Fujioka, N. Oshima, R. Suzuki, W. E. Price and L. D. Nghiem, Probing the internal structure of reverse osmosis membranes by positron annihilation spectroscopy: Gaining more insight into the transport of water and small solutes, J. Membr. Sci., 2015, 486, 106–118,  DOI:10.1016/j.memsci.2015.02.007.
4. C. A. Quist-Jensena, F. Macedonioa and E. Driolia, Membrane technology for water production in agriculture: Desalination and wastewater reuse!, Desalination, 2015, 364, 17–32,  DOI:10.1016/j.desal.2015.03.001.
5. S. M. J. Zaidi, F. Fadhillah, Z. Khan and A. F. Ismail, Salt and water transport in reverse osmosis thin film composite seawater desalination membranes, Desalination, 2015, 368, 202–213,  DOI:10.1016/j.desal.2015.02.026.
6. S.-J. Kim, B. S. Oh, H.-W. Yu, L. H. Kim, C.-M. Kim and E.-T. Yang, et al., Foulant characterization and distribution in spiral wound reverse osmosis membranes from different pressure vessels, Desalination, 2015, 370, 44–52,  DOI:10.1016/j.desal.2015.05.013.
7. C. Sun, L. Xie, X. Li, L. Sun and H. Dai, Study on different ultrafiltration-based hybrid pretreatment systems for reverse osmosis desalination, Desalination, 2015, 371, 18–25,  DOI:10.1016/j.desal.2015.05.020.
8. G. M. Geise, D. R. Paul and B. D. Freeman, Fundamental water and salt transport properties of polymeric materials, Prog. Polym. Sci., 2014, 39, 1–24,  DOI:10.1016/j.progpolymsci.2013.07.001.
9. L. Ni, J. Meng, G. M. Geise, Y. Zhang and J. Zhou, Water and salt transport properties of zwitterionic polymers film, J. Membr. Sci., 2015, 491, 73–81,  DOI:10.1016/j.memsci.2015.05.030.
10. S. Daer, J. Kharraz, A. Giwa and S. W. Hasan, Recent applications of nanomaterials in water desalination: a critical review and future opportunities, Desalination, 2015, 367, 37–48,  DOI:10.1016/j.desal.2015.03.030.
11. P. S. Goh and A. F. Ismail, Review: is interplay between nanomaterial and membrane technology the way forward for desalination?, J. Chem. Technol. Biotechnol., 2015, 90, 971–980,  DOI:10.1002/jctb.4531.
12. M. G. Buonomenna, Nano-enhanced reverse osmosis membranes, Desalination, 2013, 314, 73–88,  DOI:10.1016/j.desal.2013.01.006.
13. A. Aghigh, V. Alizadeh, H. Y. Wong, M. S. Islam, N. Amin and M. Zaman, Recent advances in utilization of graphene for filtration and desalination of water: A review, Desalination, 2015, 365, 389–397,  DOI:10.1016/j.desal.2015.03.024.
14. J. Duan, Y. Pan, F. Pacheco, E. Litwiller, Z. Lai and I. Pinnau, High-performance polyamide thin-film-nanocomposite reverse osmosis membranes containing hydrophobic zeolitic imidazolate framework-8, J. Membr. Sci., 2015, 476, 303–310,  DOI:10.1016/j.memsci.2014.11.038.
15. H. Zhao, S. Qiu, L. Wu, L. Zhang, H. Chen and C. Gao, Improving the performance of polyamide reverse osmosis membrane by incorporation of modified multi-walled carbon nanotubes, J. Membr. Sci., 2014, 450, 249–256,  DOI:10.1016/j.memsci.2013.09.014.
16. A. Peyki, A. Rahimpour and M. Jahanshahi, Preparation and characterization of thin film composite reverse osmosis membranes incorporated with hydrophilic SiO₂ nanoparticles, Desalination, 2015, 368, 152–158,  DOI:10.1016/j.desal.2014.05.025.
17. M. S. Rahaman, H. Thérien-Aubin, M. Ben-Sasson, C. K. Ober, M. Nielsen and M. Elimelech, Control of biofouling on reverse osmosis polyamide membranes modified with biocidal nanoparticles and antifouling polymer brushes, J. Mater. Chem. B, 2014, 2, 1724,  10.1039/c3tb21681k.
18. A. M. A. El-Aassar, Improvement of reverse osmosis performance of polyamide thin-film composite membranes using TiO₂ nanoparticles, Desalin. Water Treat., 2014, 1–12 , http://www.tandfonline.com/doi/abs/10.1080/19443994.2014.940206?journalCode=tdwt20#.VbPj8fk_zIY, accessed July 25, 2015.
19. M. Ben-Sasson, X. Lu, E. Bar-Zeev, K. R. Zodrow, S. Nejati and G. Qi, et al., In situ formation of silver nanoparticles on thin-film composite reverse osmosis membranes for biofouling mitigation., Water Res., 2014, 62, 260–270,  DOI:10.1016/j.watres.2014.05.049.
20. M. Safarpour, A. Khataee and V. Vatanpour, Thin film nanocomposite reverse osmosis membrane modified by reduced graphene oxide/TiO₂ with improved desalination performance, J. Membr. Sci., 2015, 489, 43–54,  DOI:10.1016/j.memsci.2015.04.010.
21. D. Cohen-Tanugi and J. C. Grossman, Nanoporous graphene as a reverse osmosis membrane: recent insights from theory and simulation, Desalination, 2015, 366, 59–70,  DOI:10.1016/j.desal.2014.12.046.
22. W. Choi, J. Choi, J. Bang and J.-H. Lee, Layer-by-layer assembly of graphene oxide nanosheets on polyamide membranes for durable reverse-osmosis applications., ACS Appl. Mater. Interfaces, 2013, 5, 12510–12519,  DOI:10.1021/am403790s.
23. H. M. Hegab and L. Zou, Graphene oxide-assisted membranes: fabrication and potential applications in desalination and water purification, J. Membr. Sci., 2015, 484, 95–106,  DOI:10.1016/j.memsci.2015.03.011.
24. P. S. Goh and A. F. Ismail, Graphene-based nanomaterial: the state-of-the-art material for cutting edge desalination technology, Desalination, 2014, 356, 115–128,  DOI:10.1016/j.desal.2014.10.001.
25. Y. Liu and X. Chen, Mechanical properties of nanoporous graphene membrane, J. Appl. Phys., 2014, 115, 034303,  DOI:10.1063/1.4862312.
26. D. Cohen-Tanugi and J. C. Grossman, Mechanical strength of nanoporous graphene as a desalination membrane, Nano Lett., 2014, 14, 6171–6178,  DOI:10.1021/nl502399y.
27. D. Cohen-Tanugi and J. C. Grossman, Water permeability of nanoporous graphene at realistic pressures for reverse osmosis desalination, J. Chem. Phys., 2014, 141, 074704,  DOI:10.1063/1.4892638.
28. M. Hu and B. Mi, Layer-by-layer assembly of graphene oxide membranes via electrostatic interaction, J. Membr. Sci., 2014, 469, 80–87,  DOI:10.1016/j.memsci.2014.06.036.
29. N. Akther, A. Sodiq, A. Giwa, S. Daer, H. A. Arafat and S. W. Hasan, Recent advancements in forward osmosis desalination: a review, Chem. Eng. J., 2015, 281, 502–522,  DOI:10.1016/j.cej.2015.05.080.
30. H.-R. Chae, J. Lee, C.-H. Lee, I.-C. Kim and P.-K. Park, Graphene oxide-embedded thin-film composite reverse osmosis membrane with high flux, anti-biofouling, and chlorine resistance, J. Membr. Sci., 2015, 483, 128–135,  DOI:10.1016/j.memsci.2015.02.045.
31. H. Dong, L. Zhao, L. Zhang, H. Chen, C. Gao and W. S. Winston Ho, High-flux reverse osmosis membranes incorporated with NaY zeolite nanoparticles for brackish water desalination, J. Membr. Sci., 2015, 476, 373–383,  DOI:10.1016/j.memsci.2014.11.054.
32. B. Zhu, D. T. Myat, J.-W. Shin, Y.-H. Na, I.-S. Moon and G. Connor, et al., Application of robust MFI-type zeolite membrane for desalination of saline wastewater, J. Membr. Sci., 2015, 475, 167–174,  DOI:10.1016/j.memsci.2014.09.058.
33. B. Zhu, Z. Hong, N. Milne, C. M. Doherty, L. Zou and Y. S. Lin, et al., Desalination of seawater ion complexes by MFI-type zeolite membranes: temperature and long term stability, J. Membr. Sci., 2014, 453, 126–135,  DOI:10.1016/j.memsci.2013.10.071.
34. C. Perego, R. Bagatin, M. Tagliabue and R. Vignola, Zeolites and related mesoporous materials for multi-talented environmental solutions, Microporous Mesoporous Mater., 2013, 37–49,  DOI:10.1016/j.micromeso.2012.04.048.
35. H. L. Jamieson, H. Yin, A. Waller, A. Khosravi and M. L. Lind, Impact of acids on the structure and composition of Linde Type A zeolites for use in reverse osmosis membranes for recovery of urine-containing wastewaters, Microporous Mesoporous Mater., 2015, 201, 50–60,  DOI:10.1016/j.micromeso.2014.09.017.
36. Y. Liu and X. Chen, High permeability and salt rejection reverse osmosis by a zeolite nano-membrane., Phys. Chem. Chem. Phys., 2013, 15, 6817–6824,  10.1039/c3cp43854f.
37. M. Fathizadeh, A. Aroujalian and A. Raisi, Effect of added NaX nano-zeolite into polyamide as a top thin layer of membrane on water flux and salt rejection in a reverse osmosis process, J. Membr. Sci., 2011, 375, 88–95,  DOI:10.1016/j.memsci.2011.03.017.
38. S. G. Kim, D. H. Hyeon, J. H. Chun, B.-H. Chun and S. H. Kim, Nanocomposite poly(arylene ether sulfone) reverse osmosis membrane containing functional zeolite nanoparticles for seawater desalination, J. Membr. Sci., 2013, 443, 10–18,  DOI:10.1016/j.memsci.2013.03.065.
39. S. G. Kim, J. H. Chun, B.-H. Chun and S. H. Kim, Preparation, characterization and performance of poly(aylene ether sulfone)/modified silica nanocomposite reverse osmosis membrane for seawater desalination, Desalination, 2013, 325, 76–83,  DOI:10.1016/j.desal.2013.06.017.
40. A. Ahmad, S. Waheed, S. M. Khan, S. e-Gul, M. Shafiq and M. Farooq, et al., Effect of silica on the properties of cellulose acetate/polyethylene glycol membranes for reverse osmosis, Desalination, 2015, 355, 1–10,  DOI:10.1016/j.desal.2014.10.004.
41. C. X. C. Lin, L. P. Ding, S. Smart and J. C. D. da Costa, Cobalt oxide silica membranes for desalination., J. Colloid Interface Sci., 2012, 368, 70–76,  DOI:10.1016/j.jcis.2011.10.041.
42. S. Wijaya, M. C. Duke and J. C. Diniz da Costa, Carbonised template silica membranes for desalination, Desalination, 2009, 236, 291–298,  DOI:10.1016/j.desal.2007.10.079.
43. N. Misdan, W. J. Lau and A. F. Ismail, Seawater Reverse Osmosis (SWRO) desalination by thin-film composite membrane—Current development, challenges and future prospects, Desalination, 2012, 287, 228–237,  DOI:10.1016/j.desal.2011.11.001.
44. S. Y. Park, S. G. Kim, J. H. Chun, B.-H. Chun and S. H. Kim, Fabrication and characterization of the chlorine-tolerant disulfonated poly(arylene ether sulfone)/hyperbranched aromatic polyamide-grafted silica composite reverse osmosis membrane, Desalin. Water Treat., 2012, 43, 221–229,  DOI:10.1080/19443994.2012.672175.
45. D. H. N. Perera, S. K. Nataraj, N. M. Thomson, A. Sepe, S. Hüttner and U. Steiner, et al., Room-temperature development of thin film composite reverse osmosis membranes from cellulose acetate with antibacterial properties, J. Membr. Sci., 2014, 453, 212–220,  DOI:10.1016/j.memsci.2013.10.062.
46. P. Dolejš, V. Poštulka, Z. Sedláková, V. Jandová, J. Vejražka and E. Esposito, et al., Simultaneous hydrogen sulphide and carbon dioxide removal from biogas by water-swollen reverse osmosis membrane, Sep. Purif. Technol., 2014, 131, 108–116,  DOI:10.1016/j.seppur.2014.04.041.
47. W. Ying, R. Kumar, M. Herzberg and R. Kasher, Diminished swelling of cross-linked aromatic oligoamide surfaces revealing a new fouling mechanism of reverse-osmosis membranes, Environ. Sci. Technol., 2015, 49, 6815–6822,  DOI:10.1021/es504325d.
48. J. Yin, Y. Yang, Z. Hu and B. Deng, Attachment of silver nanoparticles (AgNPs) onto thin-film composite (TFC) membranes through covalent bonding to reduce membrane biofouling, J. Membr. Sci., 2013, 441, 73–82,  DOI:10.1016/j.memsci.2013.03.060.
49. A. Ronen, S. Lerman, G. Z. Ramon and C. G. Dosoretz, Experimental characterization and numerical simulation of the anti-biofuling activity of nanosilver-modified feed spacers in membrane filtration, J. Membr. Sci., 2015, 475, 320–329,  DOI:10.1016/j.memsci.2014.10.042.
50. A. Ananth, G. Arthanareeswaran and Y. S. Mok, Effects of in situ and ex situ formations of silica nanoparticles on polyethersulfone membranes, Polym. Bull., 2014, 71, 2851–2861,  DOI:10.1007/s00289-014-1226-y.
51. J.-H. Kim, P. Lee, Y.-J. Lee and C.-K. Shin, Reverse osmosis membrane including nano-silver wire layer and fabrication method thereof, US 20140299537 A1, 2014.
52. J. Duan, E. Litwiller and I. Pinnau, Preparation and water desalination properties of POSS–polyamide nanocomposite reverse osmosis membranes, J. Membr. Sci., 2015, 473, 157–164,  DOI:10.1016/j.memsci.2014.09.022.
53. S.-W. Kuo and F.-C. Chang, POSS related polymer nanocomposites, Prog. Polym. Sci., 2011, 36, 1649–1696,  DOI:10.1016/j.progpolymsci.2011.05.002.
54. A. Sabir, M. Shafiq, A. Islam, A. Sarwar, M. R. Dilshad and A. Shafeeq, et al., Fabrication of tethered carbon nanotubes in cellulose acetate/polyethylene glycol-400 composite membranes for reverse osmosis, Carbohydr. Polym., 2015, 132, 589–597,  DOI:10.1016/j.carbpol.2015.06.035.
55. W. J. Lau, S. Gray, T. Matsuura, D. Emadzadeh, J. P. Chen and A. F. Ismail, A Review on Polyamide Thin Film Nanocomposite (TFN) Membranes: History, Applications, Challenges and Approaches, Water Res., 2015, 80, 306–324,  DOI:10.1016/j.watres.2015.04.037.
56. A. Khalid, A. A. Al-Juhani, O. C. Al-Hamouz, T. Laoui, Z. Khan and M. A. Atieh, Preparation and properties of nanocomposite polysulfone/multi-walled carbon nanotubes membranes for desalination, Desalination, 2015, 367, 134–144,  DOI:10.1016/j.desal.2015.04.001.
57. T.-Y. Liu, Y. Tong, Z.-H. Liu, H.-H. Lin, Y.-K. Lin and B. van der Bruggen, et al., Extracellular polymeric substances removal of dual-layer (PES/PVDF) hollow fiber UF membrane comprising multi-walled carbon nanotubes for preventing RO biofouling, Sep. Purif. Technol., 2015, 148, 57–67,  DOI:10.1016/j.seppur.2015.05.004.
58. J. Park, W. Choi, S. H. Kim, B. H. Chun, J. Bang and K. B. Lee, Enhancement of Chlorine Resistance in Carbon Nanotube Based Nanocomposite Reverse Osmosis Membranes, Desalin. Water Treat., 2010, 15, 198–204,  DOI:10.5004/dwt.2010.1686.
59. K. Zhao and H. Wu, Fast Water Thermo-pumping Flow Across Nanotube Membranes for Desalination, Nano Lett., 2015, 15, 3664–3668,  DOI:10.1021/nl504236g.
60. L. Zhang, G.-Z. Shi, S. Qiu, L.-H. Cheng and H.-L. Chen, Preparation of high-flux thin film nanocomposite reverse osmosis membranes by incorporating functionalized multi-walled carbon nanotubes, Desalin. Water Treat., 2011, 34, 19–24,  DOI:10.5004/dwt.2011.2801.
61. F. Fadhillah, S. M. Javaid Zaidi, Z. Khan, M. Khaled and P. T. Hammond, Reverse osmosis desalination membrane formed from weak polyelectrolytes by spin assisted layer by layer technique, Desalin. Water Treat., 2011, 34, 44–49,  DOI:10.5004/dwt.2011.2856.
62. K. P. Lee, T. C. Arnot and D. Mattia, A review of reverse osmosis membrane materials for desalination-development to date and future potential, J. Membr. Sci., 2011, 370, 1–22,  DOI:10.1016/j.memsci.2010.12.036.
63. A. Matin, Z. Khan, S. M. J. Zaidi and M. C. Boyce, Biofouling in reverse osmosis membranes for seawater desalination: phenomena and prevention, Desalination, 2011, 281, 1–16,  DOI:10.1016/j.desal.2011.06.063.
64. V. T. Do, C. Y. Tang, M. Reinhard and J. O. Leckie, Degradation of polyamide nanofiltration and reverse osmosis membranes by hypochlorite, Environ. Sci. Technol., 2012, 46, 852–859,  DOI:10.1021/es203090y.
65. A. Rodríguez-Calvo, G. A. Silva-Castro, F. Osorio, J. González-López and C. Calvo, Novel membrane materials for reverse osmosis desalination, Hydrol.: Curr. Res., 2014, 5, 167,  DOI:10.4172/2157-7587.1000167.
66. M. Ding, A. Ghoufi and A. Szymczyk, Molecular simulations of polyamide reverse osmosis membranes, Desalination, 2014, 343, 48–53,  DOI:10.1016/j.desal.2013.09.024.
67. M. Ding, A. Szymczyk, F. Goujon, A. Soldera and A. Ghoufi, Structure and dynamics of water confined in a polyamide reverse-osmosis membrane: a molecular-simulation study, J. Membr. Sci., 2014, 458, 236–244,  DOI:10.1016/j.memsci.2014.01.054.
68. M. T. Khan, P.-Y. Hong, N. Nada and J. P. Croue, Does chlorination of seawater reverse osmosis membranes control biofouling?, Water Res., 2015, 78, 84–97,  DOI:10.1016/j.watres.2015.03.029.
69. T. Fujioka, S. J. Khan, J. A. McDonald and L. D. Nghiem, Rejection of trace organic chemicals by a hollow fibre cellulose triacetate reverse osmosis membrane, Desalination, 2015, 368, 69–75,  DOI:10.1016/j.desal.2014.06.011.
70. T. Fujioka, S. J. Khan, J. A. McDonald, A. Roux, Y. Poussade and J. E. Drewes, et al., Modelling the rejection of N-nitrosamines by a spiral-wound reverse osmosis system: mathematical model development and validation, J. Membr. Sci., 2014, 454, 212–219,  DOI:10.1016/j.memsci.2013.12.008.
71. K. Doederer, M. J. Farré, M. Pidou, H. S. Weinberg and W. Gernjak, Rejection of disinfection by-products by RO and NF membranes: influence of solute properties and operational parameters, J. Membr. Sci., 2014, 467, 195–205,  DOI:10.1016/j.memsci.2014.05.029.
72. H. Mahdavi and T. Shahalizade, Preparation, characterization and performance study of cellulose acetate membranes modified by aliphatic hyperbranched polyester, J. Membr. Sci., 2015, 473, 256–266,  DOI:10.1016/j.memsci.2014.09.013.
73. T.-Y. Liu, C.-K. Li, B. Pang, B. van der Bruggen and X.-L. Wang, Fabrication of a dual-layer (CA/PVDF) hollow fiber membrane for RO concentrate treatment, Desalination, 2015, 365, 57–69,  DOI:10.1016/j.desal.2015.02.020.
74. A. Muhammed, M. A. El-Hashash, M. A. Mekewi, D. B. Guirguis, R. Ramadan and M. A. H. Hassanien, Polyvinyl alcohol–cellulose acetate composite reverses osmosis membranes: I. synthesis and characterization, Hydrol.: Curr. Res., 2012, 3, 131,  DOI:10.4172/2157-7587.1000131.
75. H. Minehara and K. Nakatsuji, Composite semipermeable membrane and method for producing the same, US8739977 B2, 2014, http://www.google.com/patents/US8739977, accessed July 25, 2015.
76. O. Hayashi, K. Kawamura and T. Fukumura, Method for Producing Composite Semipermeable Membrane, US 2012201972, 2012.
77. W. Gao, F. She, J. Zhang, L. F. Dumée, L. He and P. D. Hodgson, et al., Understanding water and ion transport behaviour and permeability through poly(amide) thin film composite membrane, J. Membr. Sci., 2015, 487, 32–39,  DOI:10.1016/j.memsci.2015.03.052.
78. S. Hong, I.-C. Kim, T. Tak and Y.-N. Kwon, Interfacially synthesized chlorine-resistant polyimide thin film composite (TFC) reverse osmosis (RO) membranes, Desalination, 2013, 309, 18–26,  DOI:10.1016/j.desal.2012.09.025.
79. M. Liu, Z. Chen, S. Yu, D. Wu and C. Gao, Thin-film composite polyamide reverse osmosis membranes with improved acid stability and chlorine resistance by coating N-isopropylacrylamide-co-acrylamide copolymers, Desalination, 2011, 270, 248–257,  DOI:10.1016/j.desal.2010.11.052.
80. G.-R. Xu, J.-N. Wang and C.-J. Li, Strategies for improving the performance of the polyamide thin film composite (PA-TFC) reverse osmosis (RO) membranes: Surface modifications and nanoparticles incorporations, Desalination, 2013, 328, 83–100,  DOI:10.1016/j.desal.2013.08.022.
81. D. M. Stevens, B. Mickols and C. V. Funk, Asymmetric reverse osmosis sulfonated poly(arylene ether sulfone) copolymer membranes, J. Membr. Sci., 2014, 452, 193–202,  DOI:10.1016/j.memsci.2013.10.042.
82. M. Aiba, T. Tokuyama, H. Matsumoto, H. Tomioka, T. Higashihara and M. Ueda, Effect of primary structure on permselectivity of ultrathin semipermeable polybenzimidazole membrane, J. Appl. Polym. Sci., 2014, 132,  DOI:10.1002/app.41531.
83. Y.-H. La, J. Diep, R. Al-Rasheed, M. Nassar, E. Idil Mouhoumed and A. Szymczyk, et al., The effect of cross-contamination in the sequential interfacial polymerization on the RO performance of polyamide bilayer membranes, J. Membr. Sci., 2014, 466, 348–356,  DOI:10.1016/j.memsci.2014.05.005.
84. J. Albo, J. Wang and T. Tsuru, Gas transport properties of interfacially polymerized polyamide composite membranes under different pre-treatments and temperatures, J. Membr. Sci., 2014, 449, 109–118,  DOI:10.1016/j.memsci.2013.08.026.
85. C. Baudequin, Z. Mai, M. Rakib, I. Deguerry, R. Severac and M. Pabon, et al., Removal of fluorinated surfactants by reverse osmosis – role of surfactants in membrane fouling, J. Membr. Sci., 2014, 458, 111–119,  DOI:10.1016/j.memsci.2014.01.063.
86. P. Gorgojo, M. F. Jimenez-Solomon and A. G. Livingston, Polyamide thin film composite membranes on cross-linked polyimide supports: Improvement of RO performance via activating solvent, Desalination, 2014, 344, 181–188,  DOI:10.1016/j.desal.2014.02.009.
87. N. Combernoux, L. Schrive, V. Labed, Y. Wyart, E. Carretier and A. Benony-Rhodier, et al., Study of polyamide composite reverse osmosis membrane degradation in water under gamma rays, J. Membr. Sci., 2015, 480, 64–73,  DOI:10.1016/j.memsci.2015.01.019.
88. M. Meng, F. Pellizzari, S. O. B. Boukari, N. K. Vel Leitner and B. Teychene, Impact of e-beam irradiation of municipal secondary effluent on MF and RO membranes performances, J. Membr. Sci., 2014, 471, 1–8,  DOI:10.1016/j.memsci.2014.07.046.
89. T. Sasaki, J. Okabe, M. Henmi, H. Hayashi and Y. Iida, Cesium (Cs) and strontium (Sr) removal as model materials in radioactive water by advanced reverse osmosis membrane, Desalin. Water Treat., 2012, 51, 2013 , http://www.tandfonline.com/doi/abs/10.1080/19443994.2012.704696#.VbPnGPk_zIY, accessed July 25, 2015.
90. J. M. Gohil and A. K. Suresh, A statistical study of the effect of preparation conditions on the structure and performance of thin film composite reverse osmosis membranes, Desalin. Water Treat., 2014, 1–18,  DOI:10.1080/19443994.2014.983181.
91. K. Kezia, J. Lee, W. Ogieglo, A. Hill, N. E. Benes and S. E. Kentish, The transport of hydronium and hydroxide ions through reverse osmosis membranes, J. Membr. Sci., 2014, 459, 197–206,  DOI:10.1016/j.memsci.2014.02.018.
92. B. S. Richards, G. L. Park, T. Pietzsch and A. I. Schäfer, Renewable energy powered membrane technology: safe operating window of a brackish water desalination system, J. Membr. Sci., 2014, 468, 400–409,  DOI:10.1016/j.memsci.2014.05.053.
93. T. H. Chong, S.-L. Loo and W. B. Krantz, Energy-efficient reverse osmosis desalination process, J. Membr. Sci., 2015, 473, 177–188,  DOI:10.1016/j.memsci.2014.09.005.
94. H. D. Raval and S. Maiti, Ultra-low energy reverse osmosis with thermal energy recovery from photovoltaic panel cooling and TFC RO membrane modification, Desalin. Water Treat., 2014, 1–10,  DOI:10.1080/19443994.2014.993725.
95. A. Giwa, H. Fath and S. W. Hasan, Humidification–dehumidification desalination process driven by photovoltaic thermal energy recovery (PV-HDH) for small-scale sustainable water and power production, Desalination, 2016, 377, 163–171,  DOI:10.1016/j.desal.2015.09.018.
96. A. Abusharkh, A. Giwa and S. Hasan, Wind and geothermal energy in desalination: a short review on progress and sustainable commercial processes, Industrial Engineering and Management, 2015, 04, 175,  DOI:10.4172/2169-0316.1000175.
97. R. Salcedo-Díaz, P. García-Algado, M. García-Rodríguez, J. Fernández-Sempere and F. Ruiz-Beviá, Visualization and modeling of the polarization layer in crossflow reverse osmosis in a slit-type channel, J. Membr. Sci., 2014, 456, 21–30,  DOI:10.1016/j.memsci.2014.01.019.
98. L. Zhao and W. S. W. Ho, Novel reverse osmosis membranes incorporated with a hydrophilic additive for seawater desalination, J. Membr. Sci., 2014, 455, 44–54,  DOI:10.1016/j.memsci.2013.12.066.
99. M. Liu, D. Wu, S. Yu and C. Gao, Influence of the polyacyl chloride structure on the reverse osmosis performance, surface properties and chlorine stability of the thin-film composite polyamide membranes, J. Membr. Sci., 2009, 326, 205–214,  DOI:10.1016/j.memsci.2008.10.004.
100. D. Nikolaeva, C. Langner, A. Ghanem, M. A. Rehim, B. Voit and J. Meier-Haack, Hydrogel surface modification of reverse osmosis membranes, J. Membr. Sci., 2015, 476, 264–276,  DOI:10.1016/j.memsci.2014.11.051.
101. H. Yan, X. Miao, J. Xu, G. Pan, Y. Zhang and Y. Shi, et al., The porous structure of the fully-aromatic polyamide film in reverse osmosis membranes, J. Membr. Sci., 2015, 475, 504–510,  DOI:10.1016/j.memsci.2014.10.052.
102. D. H. N. Perera, Q. Song, H. Qiblawey and E. Sivaniah, Regulating the aqueous phase monomer balance for flux improvement in polyamide thin film composite membranes, J. Membr. Sci., 2015, 487, 74–82,  DOI:10.1016/j.memsci.2015.03.038.
103. T. Wang, L. Dai, Q. Zhang, A. Li and S. Zhang, Effects of acyl chloride monomer functionality on the properties of polyamide reverse osmosis (RO) membrane, J. Membr. Sci., 2013, 440, 48–57,  DOI:10.1016/j.memsci.2013.03.066.
104. S. J. Lee, K. Kim and S. Ahn, The Internal Structure of Macroporous Membranes and Transport of Surface-Modified Nanoparticles, Microsc. Microanal., 2015, 21, 936–945,  DOI:10.1017/S1431927615013719.
105. R. Safi, M. Karimi and A. Madhi, Structural transition of asymmetric poly(ether imide) membrane prepared by wet phase inversion, Polym. Bull., 2015, 72, 1763–1774,  DOI:10.1007/s00289-015-1369-5.
106. S. S. Manickam, J. Gelb and J. R. McCutcheon, Pore structure characterization of asymmetric membranes: non-destructive characterization of porosity and tortuosity, J. Membr. Sci., 2014, 454, 549–554,  DOI:10.1016/j.memsci.2013.11.044.
107. S. Azari, L. Zou and E. Cornelissen, Assessing the effect of surface modification of polyamide RO membrane by L-DOPA on the short range physiochemical interactions with biopolymer fouling on the membrane, Colloids Surf., B, 2014, 120, 222–228,  DOI:10.1016/j.colsurfb.2014.03.028.
108. L.-F. Liu, Z.-B. Cai, J.-N. Shen, L.-X. Wu, E. M. V. Hoek and C.-J. Gao, Fabrication and characterization of a novel poly(amide-urethane@imide) TFC reverse osmosis membrane with chlorine-tolerant property, J. Membr. Sci., 2014, 469, 397–409,  DOI:10.1016/j.memsci.2014.06.029.
109. S. Yu, M. Liu, X. Liu and C. Gao, Performance enhancement in interfacially synthesized thin-film composite polyamide-urethane reverse osmosis membrane for seawater desalination, J. Membr. Sci., 2009, 342, 313–320,  DOI:10.1016/j.memsci.2009.07.003.
110. J. Meng, Z. Cao, L. Ni, Y. Zhang, X. Wang and X. Zhang, et al., A novel salt-responsive TFC RO membrane having superior antifouling and easy-cleaning properties, J. Membr. Sci., 2014, 461, 123–129,  DOI:10.1016/j.memsci.2014.03.017.
111. J. Xu, Z. Wang, L. Yu, J. Wang and S. Wang, A novel reverse osmosis membrane with regenerable anti-biofouling and chlorine resistant properties, J. Membr. Sci., 2013, 435, 80–91,  DOI:10.1016/j.memsci.2013.02.010.
112. A. Matin, H. Z. Shafi, Z. Khan, M. Khaled, R. Yang and K. Gleason, et al., Surface modification of seawater desalination reverse osmosis membranes: characterization studies & performance evaluation, Desalination, 2014, 343, 128–139,  DOI:10.1016/j.desal.2013.10.023.
113. A. Matin, Z. Khan, K. K. Gleason, M. Khaled, S. M. J. Zaidi and A. Khalil, et al., Surface-modified reverse osmosis membranes applying a copolymer film to reduce adhesion of bacteria as a strategy for biofouling control, Sep. Purif. Technol., 2014, 124, 117–123,  DOI:10.1016/j.seppur.2013.12.032.
114. C. Zhou, D. Ye, H. Jia, S. Yu, M. Liu and C. Gao, Surface mineralization of commercial thin-film composite polyamide membrane by depositing barium sulfate for improved reverse osmosis performance and antifouling property, Desalination, 2014, 351, 228–235,  DOI:10.1016/j.desal.2014.07.040.
115. T. Zhang, C. Zhu, H. Ma, R. Li, B. Dong and Y. Liu, et al., Surface modification of APA-TFC membrane with quaternary ammonium cation and salicylaldehyde to improve performance, J. Membr. Sci., 2014, 457, 88–94,  DOI:10.1016/j.memsci.2014.01.024.
116. P. Ahmadiannamini, M. L. Bruening and V. V. Tarabara, Sacrificial polyelectrolyte multilayer coatings as an approach to membrane fouling control: disassembly and regeneration mechanisms, J. Membr. Sci., 2015, 491, 149–158,  DOI:10.1016/j.memsci.2015.04.041.
117. M. Ginic-Markovic, T. G. Barclay, K. T. Constantopoulos, E. Markovic, S. R. Clarke and J. G. Matisons, Biofouling resistance of polysulfobetaine coated reverse osmosis membranes, Desalination, 2015, 369, 37–45,  DOI:10.1016/j.desal.2015.04.024.
118. L. Gutierrez, C. Aubry, R. Valladares Linares and J.-P. Croue, Natural organic matter interactions with polyamide and polysulfone membranes: formation of conditioning film, Colloids Surf., A, 2015, 477, 1–8,  DOI:10.1016/j.colsurfa.2015.03.031.
119. A. J. Karabelas and D. C. Sioutopoulos, New insights into organic gel fouling of reverse osmosis desalination membranes, Desalination, 2015, 368, 114–126,  DOI:10.1016/j.desal.2015.01.029.
120. J. Wang, Z. Wang, J. Wang and S. Wang, Improving the water flux and bio-fouling resistance of reverse osmosis (RO) membrane through surface modification by zwitterionic polymer, J. Membr. Sci., 2015, 493, 188–199,  DOI:10.1016/j.memsci.2015.06.036.
121. S. Jamaly, N. N. Darwish, I. Ahmed and S. W. Hasan, A short review on reverse osmosis pretreatment technologies, Desalination, 2014, 354, 30–38,  DOI:10.1016/j.desal.2014.09.017.
122. W. Ying, N. Siebdrath, W. Uhl, V. Gitis and M. Herzberg, New insights on early stages of RO membranes fouling during tertiary wastewater desalination, J. Membr. Sci., 2014, 466, 26–35,  DOI:10.1016/j.memsci.2014.04.027.
123. E. O. Fridjonsson, S. J. Vogt, J. S. Vrouwenvelder and M. L. Johns, Early non-destructive biofouling detection in spiral wound RO membranes using a mobile earth’s field NMR, J. Membr. Sci., 2015, 489, 227–236,  DOI:10.1016/j.memsci.2015.03.088.
124. A. H. Taheri, L. N. Sim, T. H. Chong, W. B. Krantz and A. G. Fane, Prediction of reverse osmosis fouling using the feed fouling monitor and salt tracer response technique, J. Membr. Sci., 2015, 475, 433–444,  DOI:10.1016/j.memsci.2014.10.043.
125. R. Guha, X. Shang, A. L. Zydney, D. Velegol and M. Kumar, Diffusiophoresis contributes significantly to colloidal fouling in low salinity reverse osmosis systems, J. Membr. Sci., 2015, 479, 67–76,  DOI:10.1016/j.memsci.2015.01.024.
126. C. P. Koutsou and A. J. Karabelas, A novel retentate spacer geometry for improved spiral wound membrane (SWM) module performance, J. Membr. Sci., 2015, 488, 129–142,  DOI:10.1016/j.memsci.2015.03.064.
127. T. C. Chilcott, J. Cen and J. M. Kavanagh, In situ characterization of compaction, ionic barrier and hydrodynamics of polyamide reverse osmosis membranes using electrical impedance spectroscopy, J. Membr. Sci., 2015, 477, 25–40,  DOI:10.1016/j.memsci.2014.12.013.
128. Z. Hu, A. Antony, G. Leslie and P. le-Clech, Real-time monitoring of scale formation in reverse osmosis using electrical impedance spectroscopy, J. Membr. Sci., 2014, 453, 320–327,  DOI:10.1016/j.memsci.2013.11.014.
129. M. T. Darestani, T. C. Chilcott and H. G. L. Coster, Changing the microstructure of membranes using an intense electric field: Filtration performance, J. Membr. Sci., 2014, 449, 158–168,  DOI:10.1016/j.memsci.2013.08.027.
130. E. Idil Mouhoumed, A. Szymczyk, A. Schäfer, L. Paugam and Y. H. La, Physico-chemical characterization of polyamide NF/RO membranes: insight from streaming current measurements, J. Membr. Sci., 2014, 461, 130–138,  DOI:10.1016/j.memsci.2014.03.025.
131. M.-Y. Lutskiy, S. Avneri-Katz, N. Zhu, M. Itsko, Z. Ronen and C. J. Arnusch, et al., A microbiology-based assay for quantification of bacterial early stage biofilm formation on reverse-osmosis and nanofiltration membranes, Sep. Purif. Technol., 2015, 141, 214–220,  DOI:10.1016/j.seppur.2014.12.003.
132. A. Gautam and T. J. Menkhaus, Performance evaluation and fouling analysis for reverse osmosis and nanofiltration membranes during processing of lignocellulosic biomass hydrolysate, J. Membr. Sci., 2014, 451, 252–265,  DOI:10.1016/j.memsci.2013.09.042.
133. N. Nguyen, C. Fargues, W. Guiga and M.-L. Lameloise, Assessing nanofiltration and reverse osmosis for the detoxification of lignocellulosic hydrolysates, J. Membr. Sci., 2015, 487, 40–50,  DOI:10.1016/j.memsci.2015.03.072.
134. S. C. Chen, C. F. Wan and T.-S. Chung, Enhanced fouling by inorganic and organic foulants on pressure retarded osmosis (PRO) hollow fiber membranes under high pressures, J. Membr. Sci., 2015, 479, 190–203,  DOI:10.1016/j.memsci.2015.01.037.
135. M. T. Khan, M. Busch, V. G. Molina, A.-H. Emwas, C. Aubry and J.-P. Croue, How different is the composition of the fouling layer of wastewater reuse and seawater desalination RO membranes?, Water Res., 2014, 59, 271–282,  DOI:10.1016/j.watres.2014.04.020.
136. Y. Kim, S. Lee and S. Hong, Influence of solution chemistry on the surface heterogeneity of reverse osmosis membrane, Desalin. Water Treat., 2012, 43, 308–313,  DOI:10.1080/19443994.2012.672212.
137. Y. Kim, S. Lee, J. Kuk and S. Hong, Surface chemical heterogeneity of polyamide RO membranes: measurements and implications, Desalination, 2015, 367, 154–160,  DOI:10.1016/j.desal.2015.04.006.
138. J. Wu, A. E. Contreras and Q. Li, Studying the impact of RO membrane surface functional groups on alginate fouling in seawater desalination, J. Membr. Sci., 2014, 458, 120–127,  DOI:10.1016/j.memsci.2014.01.056.
139. D. Saeki, S. Nagao, I. Sawada, Y. Ohmukai, T. Maruyama and H. Matsuyama, Development of antibacterial polyamide reverse osmosis membrane modified with a covalently immobilized enzyme, J. Membr. Sci., 2013, 428, 403–409,  DOI:10.1016/j.memsci.2012.10.038.
140. S. H. Son and J. Jegal, Preparation and characterization of polyamide reverse-osmosis membranes with good chlorine tolerance, J. Appl. Polym. Sci., 2011, 120, 1245–1252,  DOI:10.1002/app.33111.
141. Q. Cheng, Y. Zheng, S. Yu, H. Zhu, X. Peng and J. Liu, et al., Surface modification of a commercial thin-film composite polyamide reverse osmosis membrane through graft polymerization of N-isopropylacrylamide followed by acrylic acid, J. Membr. Sci., 2013, 447, 236–245,  DOI:10.1016/j.memsci.2013.07.025.
142. S. H. Maruf, A. R. Greenberg, J. Pellegrino and Y. Ding, Fabrication and characterization of a surface-patterned thin film composite membrane, J. Membr. Sci., 2014, 452, 11–19,  DOI:10.1016/j.memsci.2013.10.017.
143. G. D. Vilakati, M. C. Y. Wong, E. M. V. Hoek and B. B. Mamba, Relating thin film composite membrane performance to support membrane morphology fabricated using lignin additive, J. Membr. Sci., 2014, 469, 216–224,  DOI:10.1016/j.memsci.2014.06.018.
144. H.-C. Yang, J. Luo, Y. Lv, P. Shen and Z.-K. Xu, Surface engineering of polymer membranes via mussel-inspired chemistry, J. Membr. Sci., 2015, 483, 42–59,  DOI:10.1016/j.memsci.2015.02.027.
145. G. Kang, Z. Liu, H. Yu and Y. Cao, Enhancing antifouling property of commercial polyamide reverse osmosis membrane by surface coating using a brush-like polymer containing poly(ethylene glycol) chains, Desalin. Water Treat., 2012, 37, 139–145,  DOI:10.1080/19443994.2012.661265.
146. A. Allen, A. J. C. Semião, O. Habimana, R. Heffernan, A. Safari and E. Casey, Nanofiltration and reverse osmosis surface topographical heterogeneities: do they matter for initial bacterial adhesion?, J. Membr. Sci., 2015, 486, 10–20,  DOI:10.1016/j.memsci.2015.03.029.
147. Y. Baek, J. Yu, S.-H. Kim, S. Lee and J. Yoon, Effect of surface properties of reverse osmosis membranes on biofouling occurrence under filtration conditions, J. Membr. Sci., 2011, 382, 91–99,  DOI:10.1016/j.memsci.2011.07.049.
148. Q. Li, J. Song, H. Yu, Z. Li, X. Pan and B. Yang, Investigating the microstructures and surface features of seawater RO membranes and the dependencies of fouling resistance performances, Desalination, 2014, 352, 109–117,  DOI:10.1016/j.desal.2014.08.015.
149. F. Zamani, J. W. Chew, E. Akhondi, W. B. Krantz and A. G. Fane, Unsteady-state shear strategies to enhance mass-transfer for the implementation of ultrapermeable membranes in reverse osmosis: a review, Desalination, 2015, 356, 328–348,  DOI:10.1016/j.desal.2014.10.021.
150. H. Choi, J. Park, T. Tak and Y. N. Kwon, Surface modification of seawater reverse osmosis (SWRO) membrane using methyl methacrylate-hydroxy poly(oxyethylene) methacrylate (MMA-HPOEM) comb-polymer and its performance, Desalination, 2012, 291, 1–7,  DOI:10.1016/j.desal.2012.01.018.
151. H. Choi, Y. Jung, S. Han, T. Tak and Y.-N. Kwon, Surface modification of SWRO membranes using hydroxyl poly(oxyethylene) methacrylate and zwitterionic carboxylated polyethyleneimine, J. Membr. Sci., 2015, 486, 97–105,  DOI:10.1016/j.memsci.2015.03.040.
152. J. Nikkola, J. Sievänen, M. Raulio, J. Wei, J. Vuorinen and C. Y. Tang, Surface modification of thin film composite polyamide membrane using atomic layer deposition method, J. Membr. Sci., 2014, 450, 174–180,  DOI:10.1016/j.memsci.2013.09.005.
153. S. Yu, Z. Lü, Z. Chen, X. Liu, M. Liu and C. Gao, Surface modification of thin-film composite polyamide reverse osmosis membranes by coating N-isopropylacrylamide-co-acrylic acid copolymers for improved membrane properties, J. Membr. Sci., 2011, 371, 293–306,  DOI:10.1016/j.memsci.2011.01.059.
154. B. T. McVerry, M. C. Y. Wong, K. L. Marsh, J. A. T. Temple, C. Marambio-Jones and E. M. V. Hoek, et al., Scalable antifouling reverse osmosis membranes utilizing perfluorophenyl azide photochemistry, Macromol. Rapid Commun., 2014, 35, 1528–1533,  DOI:10.1002/marc.201400226.
155. L. Ni, J. Meng, X. Li and Y. Zhang, Surface coating on the polyamide TFC RO membrane for chlorine resistance and antifouling performance improvement, J. Membr. Sci., 2014, 451, 205–215,  DOI:10.1016/j.memsci.2013.09.040.
156. A. Sweity, T. R. Zere, I. David, S. Bason, Y. Oren and Z. Ronen, et al., Side effects of antiscalants on biofouling of reverse osmosis membranes in brackish water desalination, J. Membr. Sci., 2015, 481, 172–187,  DOI:10.1016/j.memsci.2015.02.003.
157. K. L. Tu, A. Chivas and L. D. Nghiem, Effects of chemical preservation on flux and solute rejection by reverse osmosis membranes, J. Membr. Sci., 2014, 472, 202–209.
158. L. Ni, J. Wang, Y. Zhang, J. Meng and Y. Zhang, The performance improvement of hollow fiber composite reverse osmosis membranes by post-treatments, Desalin. Water Treat., 2012, 32–36 , http://www.tandfonline.com/doi/abs/10.5004/dwt.2011.2851#.VbPmE_k_zIY, accessed July 25, 2015.
159. Z. Hu, Y. Chen and J. Jiang, Zeolitic imidazolate framework-8 as a reverse osmosis membrane for water desalination: insight from molecular simulation, J. Chem. Phys., 2011, 134, 134705,  DOI:10.1063/1.3573902.
160. S. Turgman-Cohen, J. C. Araque, E. M. V. Hoek and F. A. Escobedo, Molecular dynamics of equilibrium and pressure-driven transport properties of water through LTA-type zeolites, Langmuir, 2013, 29, 12389–12399,  DOI:10.1021/la402895h.
161. Y. Shen, P. O. Saboe, I. T. Sines, M. Erbakan and M. Kumar, Biomimetic membranes: a review, J. Membr. Sci., 2014, 454, 359–381,  DOI:10.1016/j.memsci.2013.12.019.
162. S. Delagah, Cellulose acetate membrane, 2015, http://www.usbr.gov/research/projects/detail.cfm?id=1526.
163. A. P. Murphy, S. Delagah and R. L. Riley, Method for making high flux, high salt rejection cellulose desalting membranes, US 8123945 B2, 2012, http://www.google.com/patents/US8123945, accessed August 8, 2015.

---