Reverse osmosis (RO) membranes based on cellulose acetate (CA), were prepared using a phase inversion technique. To improve the hydrophilicity, salt rejection and water flux of these membranes, a novel grafting of 2-acrylamido-2-methylpropanesulfonic acid (AMPSA) was added on the top surface of the CA-RO membranes. The grafted CA-RO membranes were characterized by Fourier transform infrared spectroscopy (FTIR), contact angle, and scanning electron microscopy techniques. It was found that the contact angles were 58° and 45° for pristine CA and 15 wt% grafted CA-RO membranes, respectively, which suggest an increase in the membrane surface hydrophilicity after grafting. The morphological studies of the surface of the pristine CA-RO membrane revealed a typical ridge-and-valley morphology and displayed a relatively high surface roughness of 337 nm, and a significant decrease at 15 wt% of grafted CA-RO membrane to 7 nm. The effect of the grafting percentages of AMPSA on the water flux and salt rejection was studied using a cross flow RO unit. The salt rejection and water flux of the grafted CA-RO membrane with 15 wt% were 99.03% and 6 L/m2h, respectively.

INTRODUCTION

Water scarcity is one of the most serious global challenges; one-third of the world's population are living in water-stressed countries; by the year 2030, this figure is predicted to rise to nearly two-thirds. The challenge of providing ample, safe drinking water is farther complicated by population growth, contamination of available freshwater resources, and climate change (Elimelech & Phillip 2011). Many decades of successful implementations demonstrate how desalination technology can provide supplementary or main water sources. The desalination process can be roughly categorized into two major types: thermal and membrane separation (Zhou et al. 2014). Reverse osmosis (RO) is currently the most widely used desalination technology due to continuous technological improvements and substantial cost reductions. This technology has the advantages of having modular construction and small carbon footprint, which allows for the combination of additional treatment processes. Presently, two main types of polymeric RO membranes exist in the market: cellulose acetate (CA) membranes and polyamide thin film composite (TFC) membranes. However, TFC membranes, which are the most commonly used type of RO membranes, cannot withstand chlorine exposure, even at low ppm concentrations. On the other hand, CA membranes exhibit some resistance against chlorine at concentrations of up to 5 mg Cl2/L and 0.2 mg Cl2/L for short and long exposure times, respectively. Consequently, CA membranes are used in most of the RO plants located in the Middle East Region while the elevated seawater temperature and water quality promote the risk of membrane biofouling (Khan et al. 2015).

Membrane desalination was improved by the development of an effective process to make defect-free CA membranes that had high fluxes and salt rejections where the resulting CA membrane contained a relatively dense ‘skin’ on the surface with a porous network support (Lonsdale 1982). The disadvantages of the CA membranes are limited pH operating range and lower salt rejections, due to possible biodegradation, hydrolysis by acids and alkalis, and narrow temperature limits. Surface modifications of RO membranes are a cost-effective technique that enhances the membrane performance while conserving its bulk characteristics. There are many surface modification techniques such as free radical, photochemical, radiation, coating and plasma-induced grafting (Savoji et al. 2013). Modification via graft polymerization has many advantages, such as its ease of use and controllable introduction of the graft chains to the surface with the bulk properties unchanged (Lee et al. 2011). Grafting of side chains can be performed in two ways: ‘grafting-from’ and ‘grafting-to’ methods. In the former method, the membrane surface consists of reactive radicals, while in the latter method grafting chains carry the reactive radicals for initiation (Azari et al. 2013). Grafting can be achieved using different chemical techniques, such as chemical oxidation, plasma discharge method, and radiation. Non-toxicity and non-biodegradability, are characteristics possessed by 2-crylamidopropane-2-methyl sulfonic acid (AMPSA), which makes it an ideal choice for use in the water purification industry (Thomas & Asheville 1980). AMPSA has been applied in the medical field as wound dressing material (Dasa et al. 2014). The distinguished properties of AMPSA are a result of high mobility associated with a conformationally flexible structure and water binding ability (Kato et al. 2003). The enhanced surface hydrophilicity of the membranes results in the improvement of the antifouling performance (Kang & Cao 2012). The grafted AMPSA coating aims at producing a surface that will decrease the nonspecific adsorption of various proteins due to its hydrophilicity and provides a highly sulfonated surface (Devi et al. 2014). It was indicated that AMPSA, grafted onto polyurethane membrane, has a negative charge due to the sulfonic acid groups (Coskun et al. 2006).

The main objective of this work is to functionalize and hydrophilize CA-RO membrane surfaces with novel grafting of AMPSA via chemical oxidation. The grafted CA-RO membranes will be characterized by Fourier transform infrared spectroscopy (FTIR), contact angle, atomic force microscopy (AFM), and scanning electron microscopy (SEM) techniques. The performance of the CA-RO membranes with different grafting percentages will be assessed for water desalination using water flux and salt rejection measurements.

MATERIALS AND METHODS

Materials

CA (molecular weight of 100,000 g/mol and 39.8 wt% acetyl) was received from Aldrich and 1,4-dioxane was supplied by Panreac Quimica S.A (Barcelona, Spain). Methanol (purity>99.5%) and acetone (purity>99%) were received from Labsolve (Lisbon, Portugal). The acetic acid (purity>99.8%) was supplied by BDH Anala R (UK) and 2-acrylamido-2-methylpropanesulfonic acid was obtained from Fisher, and was used as received. Potassium persulfate, NaCl and NaOH were purchased from Merck, Aldrich and the Egyptian Petrochemical Company, respectively.

Preparation of CA-RO membranes

The CA-RO membrane was prepared using a mixture of dioxane (27.62 g), acetone (10.57 g), and acetic acid (5.07 g) as solvents while methanol (8.45 g) was used as a nonsolvent in addition to CA (8.45 g). This mixture of solution was left under stirring for 24 h at room temperature, until the CA completely dissolved. The solution was put into an ultrasonic bath for 30 min to remove the air bubbles entrapped in the polymer solution. The RO membranes were obtained through the spreading of the solution onto a glass plate using the knife of an automatic applicator, at room temperature and 42% relative humidity. The thickness of the film was previously selected (250 μm), and spread at a constant speed (10 mm/s) with an automatic film applicator (Zehntner 2300-Swiss). After casting the solution with evaporation time of the solvent of 60 s, the CA membrane (cast onto the glass plate) was immersed for 15 min in a deionized water ice bath. The formed CA membrane was then placed in a water bath at about 4 °C for 2 h to eliminate the effect of capillary pressure and then washed with distilled water to completely remove the residual solvents (Han & Bhattacharyya 1995; Werner et al. 2011). The prepared CA-RO were then post-treated for 10 min at about 80–85 °C. Then these membranes were soaked in deionized water for 24 hours and air-dried for 24 h before characterization.

Grafting of AMPSA onto the CA-RO membrane surface

Sodium hydroxide (0.04%) was poured on the top surface of the CA-RO membranes for 5 min to partially deacetylate the CA-RO membrane. After washing the membranes several times with deionized water, potassium persulfate (1.5 wt%) was added drop-wise onto the CA-RO membrane top surface for 10 min to form a free radical as shown in Figure 1. Different weight percentages (10, 15, 20 and 25) of AMPSA were placed onto the CA-RO membrane surface for 10 min at room temperature to allow the grafting reaction to take place. The functionalized CA-RO membranes were washed again by deionized water several times to remove unreacted chemicals. Finally, the membranes were thermally heated at 75 °C for 30 min.
Figure 1

The grafting process of CA-RO membrane.

Figure 1

The grafting process of CA-RO membrane.

Grafting mechanism of AMPSA onto CA-RO surface membrane

The grafting of AMPSA was conducted on the top surface of the CA-RO membranes according to the schematic diagram of Figure 2. The presence of the hydroxyl groups in the CA backbone structure makes it vulnerable to grafting via forming free radicals on the membrane surface. Potassium persulfate solution (1.5%) was used to produce free radicals on the CA-RO membrane surface via chemical oxidation and abstraction of hydrogen atoms from the hydroxyl groups (Akahiro & Long 2013). This free radical attacks the weak double bond at the end group of AMPSA to form another free radical. In addition, the two free radicals on AMPSA and the CA membrane surface react with each other to from a single bond. Chain propagation from AMPSA free radicals took place on the membrane surface.
Figure 2

Schematic diagram of grafting polymerization mechanism of AMPSA onto CA-RO membranes. R represents AMPSA monomer and * represents the formation of free radicals.

Figure 2

Schematic diagram of grafting polymerization mechanism of AMPSA onto CA-RO membranes. R represents AMPSA monomer and * represents the formation of free radicals.

Characterization techniques

FTIR was used to characterize the membranes. The spectra were recorded (Spectrum BX 11 Infrared spectrometer FTIR LX 18-5255 Perkin Elmer) in the wavenumber range of 4,000–400 cm−1 for the prepared CA-RO membranes. To produce cross sectional, surface, and bottom images of the membranes, SEM XL 30 JEOL was used. The morphological images of the functionalized CA-RO membrane cross sections were snapped after liquid nitrogen treatment that was used to give a generally consistent and clean break of the membrane. The membranes were sputter-coated with a thin film of gold and were mounted on a brass plate using double-sided adhesion tape in a lateral position. The surface topography and roughness were analyzed using AFM; a Dimension 3100 Digital Instrument equipped with manufactured Nanoscope IV Scanning Probe Microscope with software version 6.12 and used in resonant (tapping) mode. Tapping mode was utilized to determine roughness analyses of the samples. Square samples of (1 cm2) were attached to the AFM stub with carbon tape. The contact angle of the CA-RO membranes surfaces was measured using a contact angle goniometer from Ramé Hart, Instrument Company, France. A drop of distilled water (2 μL) was placed on the CA-RO membrane surface (3 × 2 cm) using a microsyringe (Hamilton Company, Reno, NV, USA). The contact angle was measured within 10 sec after the water drop was placed and the contact angles were measured at five different positions. The performance tests (salt rejection and water flux) were conducted for the CA-RO membranes (area 42 cm2) for three different samples using a cross flow RO unit (CF042, Sterling, USA) equipped with hydra pump, pressure control valve and gauge through rejection line, with variable frequency drive (SV015IG5A-4) and flow meter (F-550 USA). Saline salt solutions of NaCl of 10,000 ppm with pH 7 were used. The determination of the total dissolved salt of the permeate water was measured with a pH and conductivity meter (430 portable, Jenway, UK). The water flux (F) and salt rejection (R) values were obtained using Equations (1) and (2) (Ulbricht et al. 1996; Chen et al. 2010): 
formula
1
where V is the total volume of water passing through the membrane (L), A is the membrane area (m2) and t is the time (hour). 
formula
2
C0 is the conductivity of salty water at the pressure side at the beginning of the experiment and Cmemb. is the conductivity of water that goes through the membrane.

RESULTS AND DISCUSSION

Structure investigation of grafted CA-RO membranes

The chemical structure of CA-RO grafted membranes was investigated using infrared spectroscopy. Figure 3 shows the FTIR spectra of pristine CA-RO membrane, AMPSA and the grafted CA-RO membranes with different weight percentages. The band at 3,434 cm−1 is attributed to the stretching of the hydroxyl group and the band at 1,752 cm−1 is attributed to the C = O bond of the carboxylic group. In addition, the H-O-H bending group of the absorbed water appeared at 1,660 cm−1. The peak at 1,380 cm−1 corresponded to the C-H bending and the peak at 1,250 cm−1 was attributed to the C-O stretching bands, while the bands at 1,060 cm−1 correspond to the stretching modes of the C-O-C single bond of esters and to the C-O-C pyranose ring skeletal vibration (Li et al. 2009; Azari et al. 2013).
Figure 3

FTIR spectra of (a) pristine CA-RO membrane and AMPSA, (b) grafted CA-RO membranes with different w% percentages of grafting.

Figure 3

FTIR spectra of (a) pristine CA-RO membrane and AMPSA, (b) grafted CA-RO membranes with different w% percentages of grafting.

Conversely, the FTIR spectrum of AMPSA showed the transmission band, also called the amide I mode, which resonated at 1,672 cm−1, while the band at 1,641 cm−1 was assigned to the alkene C = C stretch, and the band at 1,765 cm−1 for the C = O, and the N–H stretch, amide II mode, absorbed around 3,470 cm−1, and the C–O stretch absorbed at 1,222 cm−1. The 1,036 cm−1 band of the sulfonic group of AMPSA was also observed. The bands at 1,560 cm−1 and at 634 cm−1 were assigned to C-N- and S-O groups, respectively (Durmaz & Okay 2000; Coskun et al. 2006).

The FTIR spectra of the grafted CA with different weight percentages of AMPSA are shown in Figure 3(b). The broad absorption band around 3,496 cm−1 was ascribed to the overlapping peaks of NH and OH groups. The bands at 2,960 and 2,884 cm−1 were attributed to CH3 and CH2 stretching. The band observed at 1,430 cm−1 is the C-N bend. The disappearance of the band at 1,640 cm−1 indicated the grafting formation at the C = C sites (Yetimoglu et al. 2007).

Morphological properties of grafted CA-RO membranes

The grafted CA-RO membranes with different wt% of AMPSA SEM images are shown in Figure 4. For pristine CA membranes, the cross section images revealed an asymmetric structure. During the first step of desolvation by solvent evaporation, a thin skin layer of CA was formed instantly at the top of the cast film due to the loss of solvent. In the solvent–nonsolvent exchange process, the nonsolvent diffused into the membrane, while the solvent diffused out of the polymer solution through the thin solid layer (Strathmann & Kock 1977). This thin layer became the top skin layer, which governed the selectivity and the flux of the membrane, while the porous structure that formed during the solvent–nonsolvent extraction step became the porous sublayer (Smolders et al. 1992). Generally, large finger-like cavities in the membrane were formed when the cast solution precipitated rapidly, while the pore structure was reduced. Conversely, the porous sponge structure of the membrane was formed when the precipitation process was slow. The finger-like RO membrane had a low flux and significant rejection to the micro-solutes, and was generally used in high pressure RO processes. In addition, a typical ridge-and-valley morphology was observed for the surface membrane (Han et al. 2012). According to the bottom surface images for the pristine CA-RO membrane, pinholes have appeared due to the exchange of solvent and nonsolvent across the interface that caused a phase separation in the polymer film (Strathmann & Kock 1977).
Figure 4

SEM images of surface, bottom and cross section of CA-RO membrane with different w% of AMPSA grafting.

Figure 4

SEM images of surface, bottom and cross section of CA-RO membrane with different w% of AMPSA grafting.

The thickness of the dense layer at the bottom of the CA-RO membrane that is grafted with 10 wt% AMPSA decreased; meanwhile the finger-like structure did not change (Figure 4(b). Conversely, the top layer of this membrane became smoother than the pristine CA membrane. From the cross section of the CA-RO membrane grafted with 15 wt%, it was clear that the thickness of the top dense layer increased, while the bottom layer decreased. The finger-like structure became distorted with tear-shaped macrovoids that extended from the compact skin layer towards the permeate side (Richards et al. 2012). In addition, the top layer of this membrane revealed an irregular rhombohedral structure (as shown in inset of Figure 4(c)) (Abdul et al. 2014). This led to the decrease of water flux and increase of the operation pressure. The cross sections of the CA-RO membranes grafted with 20 and 25 wt% indicated that the thickness of the top and the bottom dense skin layer increased (Figure 4(d) and 4(e)). The surfaces of these highly grafted membranes became smoother and the pores in the bottom layer were blocked by the grafting of AMPSA inside these pores. The monomer had penetrated through the active layer of the membrane, particularly for higher degrees of grafting (Dai et al. 2004).

AFM analysis

The three-dimensional AFM images of pure CA and grafted CA-RO membranes with different weight percentages of grafting are presented in Figure 5. The grafted CA-RO membranes possess nodule valley-like structures. The nodules were seen as bright high peaks and the pores were seen as dark depressions. It was noted that the morphology of the membranes changed with the increase in the amount of grafting of the CA membranes. The surface of the pristine CA membrane had no nodules, and as the grafting percentage increased, the numbers of nodules increased while their sizes decreased. Analysis of AFM images was performed in order to measure the roughness of the membrane surface from the arithmetic mean roughness value. Roughness of the membrane surface was attributed to the presence of hills and valleys which increased the surface area and provided more binding sites. The pristine CA-RO membrane displayed a relatively high surface roughness of 337 nm. The significant decrease in the surface roughness was noted at 15 wt% of grafted CA-RO membrane of 7 nm, and 4.2 nm for the 20 wt% grafted CA-RO membrane. The curing and grafting of the membranes improved the smoothness of the surface by filling the valleys on the membrane surface (Bhattacharya & Misra 2004). Fouling was often linked to the intrinsic membrane properties, indicating that fouling rates increase with membrane surface roughness increases because foulant particles are more likely to be entrained by rougher topologies than by smoother membrane surfaces (Kochkodan & Sharma 2012).
Figure 5

AFM images of pure CA and grafted CA-RO membranes with different w% of grafting.

Figure 5

AFM images of pure CA and grafted CA-RO membranes with different w% of grafting.

Hydrophilic properties of CA-RO grafted membranes

To evaluate the hydrophilic properties of the AMPSA grafted membranes, the contact angles were measured. The contact angles are 58° and 45° for pristine CA and the 15 wt% grafted CA-RO membranes, respectively, which suggested a slight increase in the membrane surface hydrophilicity after grafting (Worthley et al. 2011). The enhanced hydrophilic properties were due to the presence of three hydrophilic groups (sulfonic, carboxylic and amide) of AMPSA. At higher grafting content ratios of 20 and 25%, pore clogging took place due to the agglomeration of monomer particles through the active dense layer of the membrane, with some hydrophobic structure as indicated by the increase of water contact angle (Figure 6) (Dai et al. 2004).
Figure 6

Contact angle versus wt% of AMPSA in grafting of CA-RO membranes.

Figure 6

Contact angle versus wt% of AMPSA in grafting of CA-RO membranes.

Salt rejection and water flux of grafted CA-RO membranes

Figures 7 and 8 depict the salt rejection and water flux versus the operating pressure for CA-RO membranes grafted with different wt% of AMPSA. The salt rejection for the pristine CA-RO membrane was 92.3% at 14 bar, which decreased to 84.0% at 26 bar. This can be attributed to the effect of the concentration polarization, where a boundary layer is formed near the membrane surface, in which the salt concentration exceeds the salt concentration in the bulk solution (Li & Wang 2010). High salinity at the membrane surface leads to salt transportation through the membrane and the local osmotic pressure. In addition, at high pressure, the salt rejection decreased dramatically due to the increase of the osmotic pressure along the feed channel (Lee et al. 2011). Conversely, it was observed that the water fluxes of CA-RO membranes increased with increased operating pressure. According to the solution and diffusion model, the flux was proportional to the net differential pressure across the membrane, using the transport equation: 
formula
3
where Jv is flux, A is the water permeation coefficient, ΔP is the trans-membrane pressure and Δπ is the osmotic pressure.
Figure 7

Salt rejections versus feed pressure of CA-RO membrane grafted with different w% of AMPSA.

Figure 7

Salt rejections versus feed pressure of CA-RO membrane grafted with different w% of AMPSA.

Figure 8

Water flux versus feed pressure of CA-RO membranes grafted with different wt% of AMPSA.

Figure 8

Water flux versus feed pressure of CA-RO membranes grafted with different wt% of AMPSA.

For the CA-RO membrane grafted with 10 wt%, the salt rejection increased to 97.2% at 14 bar and then decreased to 85.6% at 20 bar. Meanwhile, for the CA-RO membranes grafted with 15 wt%, the salt rejection increased to 99.03% at 11 bar and decreased to 96.9% at 15 bar. This was due to the higher thickness of the active top dense layer while the bottom layer decreased in thickness. At high grafting weight percentages of 20 and 25, the operating pressure increased to 16 and 24 bar, respectively, to produce a salt rejection decrease from 66.6% to 36.3%. These results are in agreement with those obtained from the SEM images, where the thickness of the top active dense skin layer was deformed and the bottom layer increased. The grafting process may destroy or add flaws to the membrane surface, resulting in an increased water flux (Bentvelzen et al. 1973).

Figure 9 presents the salt rejection and water flux versus the grafting polymerization wt% of AMPSA onto the CA-RO membranes. The CA-RO membrane grafted with 15 wt% AMPSA had the highest rejection of 99.03% and 6 L/m2h water flux. However, at higher weight percentages of grafting, the pores have been plugged, which reduced the salt rejection and increased the operating pressure, as previously stated (Tang et al. 2007; Savoji et al. 2013).
Figure 9

Salt rejection and water flux versus grafting wt% of AMPSA onto CA-RO membranes.

Figure 9

Salt rejection and water flux versus grafting wt% of AMPSA onto CA-RO membranes.

The effect of salinity concentration on the salt rejection and water flux of CA-RO membranes grafted with 15 wt% is depicted in Figure 10. It was observed that an increase in feed water salinity enhances the membrane salt passage. The salt rejection decreased from 99.03% to 92.0% at 10,000 ppm and 35,000 ppm, respectively. The salt passage was affected by the total dissolved solids concentration due to interactions between the ions and the membrane surface. The RO membranes have an overall negative surface charge and repelled negatively charged ions or molecules. As negative ions were repelled, more cations than anions were present near the membrane surface; this phenomenon created an electric potential known as the Donnan potential. The Donnan potential helped to repel ions from the membrane. An increase in salinity decreases the Donnan potential effect on the membrane salt rejection (Greenlee et al. 2009).
Figure 10

Salt rejection and water flux vs. pressure onto CA-RO membranes.

Figure 10

Salt rejection and water flux vs. pressure onto CA-RO membranes.

The data for the CA-RC membrane grafted with 15% AMPSA showed great promise for large scale water desalination. Future work will focus on a pilot scale study for commercial applications.

CONCLUSIONS

The surface of the CA-RO membranes was successfully modified by grafting a hydrophilic AMPSA monomer. The cross section of the CA-RO membrane grafted with 15 wt% of AMPSA showed that the thickness of the top dense layer had increased while the bottom layer had decreased. The dominating finger-like structure was distorted to form tear-shaped macrovoids. The surfaces of theses membranes became smoother and the pores in the bottom layer were blocked. Moreover, the CA-RO membrane surface hydrophilicity was enhanced to have a contact angle of 45° when grafted with 15 wt% of AMPSA as three hydrophilic groups (sulfonic, carboxylic and amide) were present. From the atomic force images, it was noted that the pristine CA-RO membranes displayed a relatively high surface roughness of 337 nm and showed a significance decrease at 15 wt% of grafted CA-RO membrane to 7 nm. The optimum salt rejection and water flux of 99.03% and 6 l/m2h were obtained for the grafted CA-RO membrane with 15 wt% of AMPSA. These data showed great promise for designing an effective RO system for large scale applications.

ACKNOWLEDGEMENTS

This work is supported by the Science and Technology Development Fund in Egypt (project ID 3988).

REFERENCES

REFERENCES
Abdul
R. I.
Jean
B.
Vuningoma
Y. H.
Hongtao
W.
Jun
L. R.
2014
Rapid carbonation for calcite from a solid-liquid-gas system with an imidazolium-based ionic liquid
.
Journal of Molecular Sciences
15
,
11350
11363
.
Bentvelzen
J. M.
Kimura-yeh
F.
Hopfenberg
H. B.
Stannett
V.
1973
Modification of cellulose acetate reverse osmosis membranes by radiation grafting
.
Journal of Applied Polymer Science
17
,
809
820
.
Bhattacharya
A.
Misra
B. N.
2004
Grafting: a versatile means to modify polymers techniques, factors and applications
.
Progress Polymer Science
29
,
767
814
.
Chen
J.
Li
J.
Zhan
X.
Han
X.
Chen
C.
2010
Effect of PEG additives on properties and morphologies of polyetherimide membranes prepared by phase inversion
.
Frontiers of Chemical Engineering China
4
(
3
),
300
306
.
Dai
Q.
Xu
Z.
Deng
H.
Liu
Z.
Wu
J.
Seta
P.
2004
Surface modification of microporous polypropylene membranes by graft polymerization of N,N-dimethylaminoethyl methacrylate
.
Chinese Journal Polymer Science
22
,
369
377
.
Dasa
M.
Devia
N.
Sarmab
J.
2014
Preparation, characterization, and water sorption study of 2-acrylamido-2-methylpropane sulfonic acid (AMPS) based hydrogel
.
Journal of Chemical and Pharmaceutical Research
6
(
10
),
800
806
.
Devi
N.
Sarma
J.
Das
M.
2014
Blood compatible hydrogel composed of starch, 2-acrylamido-2-methylpropane sulfonic acid and acrylamide
.
International Journal of Latest Research in Science and Technology
3
(
4
),
205
210
.
Greenlee
L. F.
Lawlerb
D. F.
Freemana
B. D.
Marrotc
B.
Moulinc
P
.
2009
Reverse osmosis desalination: water sources, technology, and today's challenges
.
Water Research
43
,
2317
2348
.
Han
G.
Chung
T.
Toriida
M.
Tamai
S
.
2012
Thin film composite forward membranes with Novel hydrophilic supports for desalination
.
Journal of Membrane Science
423–424
,
543
555
.
Kato
K.
Uchida
E.
Kang
E.
Uyama
Y.
Ikada
Y.
2003
Polymer surface with graft chains
.
Progress in Polymer Science
28
,
209
259
.
Khan
M. T.
Hong
P. Y.
Nada
N.
Croue
J.
Ph.
2015
Does chlorination of seawater reverse osmosis membranes control biofouling
.
Water Research
78
,
84
97
.
Kochkodan
V. M.
Sharma
K.
2012
Graft polymerization and plasma treatment of polymer membranes for fouling reduction
.
Journal of Environmental Science and Health Part A
47
,
1713
1727
.
Li
D.
Wang
H.
2010
Recent developments in reverse osmosis desalination membranes
.
Journal of Materials Chemistry
20
,
4551
4566
.
Lonsdale
H. K.
1982
The growth of membrane technology
.
Journal of Membrane Science
10
,
81
181
.
Richards
H. L.
Baker
P. G. L.
Iwuoha
E.
2012
Metal nanoparticle modified polysulfone membranes for use in wastewater treatment: a critical review
.
Journal of Surface Engineered Materials and Advanced Technology
2
,
183
193
.
Smolders
C. A.
Reuvers
A. J.
Boom
R. M.
Wienk
I. M.
1992
Microstructures in phase-inversion membranes, Part 1. Formation of macrovoids
.
Journal of Membrane Science
73
,
259
275
.
Strathmann
H.
Kock
A.
1977
The formation mechanism of phase inversion membranes
.
Desalination
21
,
241
255
.
Thomas
C. A.
Asheville
N. C.
1980
Highly absorbent fibers of rayon with sulfonic acid polymer incorporated
.
US Patent US4242242 A
.
Werner
B.
Ramesh
P.
Anna
D.
2011
A membrane and use thereof
.
European Patent, EP 2
305
368
A1
.
Worthley
C. H.
Constantopoulos
K. T.
Ginic-Markovic
M.
Pillar
R. J.
Matisons
J. G.
Clarke
S.
2011
Surface modification of commercial cellulose acetate membranes using surface-initiated polymerization of 2-hydroxyethyl methacrylate to improve membrane surface biofouling resistance
.
Journal of Membrane Science
385–386
,
30
39
.