Surface modification of seawater desalination reverse osmosis membranes: Characterization studies & performance evaluation

Desalination 343 (2014) 128–139

Contents lists available at ScienceDirect

Desalination journal homepage: www.elsevier.com/locate/desal

Surface modification of seawater desalination reverse osmosis membranes: Characterization studies & performance evaluation Asif Matin a,⁎, H.Z. Shafi a, Zafar Khan a, Mazen Khaled b, Rong Yang c, Karen Gleason c, Faizur Rehman d a

Depatment of Mechanical Engineering, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia Department of Chemistry, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia Department of Chemical Engineering, Massachusetts Institute of Technology Cambridge, MA, USA d Research Institute, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia b c

H I G H L I G H T S • thin film deposition using all-dry technique at ambient temperature. • minimal compromise on membrane performance. • presence of coating confirmed by multiple techniques.

a r t i c l e

i n f o

Article history: Received 7 July 2013 Received in revised form 2 October 2013 Accepted 21 October 2013 Available online 20 November 2013 Keywords: Reverse osmosis Surface modification Copolymer Amphiphilic

a b s t r a c t In this work we report surface modification of commercial reverse osmosis membranes by depositing ultrathin copolymer coatings, which could potentially enhance the biofouling resistance of RO membranes. Hydrophilic monomer hydroxyethyl methacrylate (HEMA) and a hydrophobic monomer, perfluorodecyl acrylate (PFDA) were copolymerized directly on the active layer of commercial aromatic polyamide reverse osmosis (RO) membranes using an initiated Chemical Vapor Deposition (iCVD) technique. Attenuated total reflective Fourier transform infrared spectra (ATR-FTIR) verified the successful modification of the membrane surfaces as a new FTIR adsorption band around 1730 cm−1 corresponding to carbonyl groups in the copolymer film appeared after the deposition. X-ray Photoelectron spectroscopy (XPS) analysis also confirmed the presence of the copolymer film on the membrane surface by showing strong fluorine peaks emanating from the fluorinated alkyl side chains of the PFA molecules. Contact angle measurements with deionized water showed the modified membrane surfaces to be initially very hydrophobic but quickly assumed a hydrophilic character within few minutes. Atomic Force Microscopy (AFM) revealed that the deposited films were smooth and conformal as the surface topology of the underlying membrane surface remained virtually unchanged after the deposition. FESEM images of the top surface also showed that the typical ridge-and-valley structure associated with polyamide remained intact after the deposition. Short-term permeation tests using DI water and 2000 ppm NaCl water showed that the deposited copolymer coatings had negligible effect on permeate water flux and salt rejection. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Seawater desalination by reverse osmosis (SWRO) is becoming increasingly popular due to its ease of operation, lower operational and maintenance costs, and environmental friendliness [1,2]. The heart of the process is a thin-film composite membrane composed of a nonwoven fabric, a porous support layer and a nonporous ultrathin selective layer [3,4]. However, despite possessing several advantages over contemporary techniques, the reverse osmosis process faces a major challenge of membrane fouling [5]. Fouling, which is defined as the irreversible ⁎ Corresponding author. Tel.: +966 3 8605054; fax: +966 3 8602949. E-mail address: [email protected] (A. Matin). 0011-9164/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.desal.2013.10.023

adsorption of solutes on the membrane surface and pores, results in a significant decline in permeate water flux as well as deterioration of permeate water quality. The economic consequences of membrane fouling are a substantial increase in both operation and maintenance costs due to the need for applying higher pressures as well as frequent membrane cleaning and replacement [6]. Among the different fouling types, organic and biofouling have been identified as the most problematic and hence the most common [7,8]. In fact, biofouling is commonly referred to as the Achilles heel of membrane processes [9]. Fouling due to organic and inorganic components and microorganisms can occur simultaneously, and results in the formation of a biofilm that causes operational problems [10–12]. Membrane surface modification is an attractive technique for the control and prevention of bio fouling. The main idea behind this

A. Matin et al. / Desalination 343 (2014) 128–139

approach is to modify the membrane surface characteristics to lower the affinity of foulants to the surface of the selective layer. Deposition of antifouling coatings is a simple way of membrane surface modification and has recently been the focus of several researchers. Kim and Lee [13] coated nanofiltration and reverse osmosis membranes with neutral polyvinyl alcohol (PVA) polymer and showed that the modified membrane had significantly reduced fouling levels. Louie et al. [14] used a hydrophilic polyethylene-polyamide block copolymer and found very encouraging results. Yu et al. [15] used the thermo responsive copolymer poly(N-isopropylacrylamide-co-acrylic acid) to modify commercial polyamide membranes. Amphiphilic copolymers are another class of material that have been explored in the context of membrane fouling. Asatekin et al. [16] grafted a comb-like amphiphilic copolymer on PAN ultrafiltration membranes that consisted of hydrophilic brushes attached to a hydrophobic backbone. These membranes showed excellent resistance to biopolymers as well as bacterial adhesion [17]. Similarly, Park et al. [18] grafted of the hydrophobic polysulfone and the hydrophilic polyethylene glycol copolymers on polysulfone membranes with considerable success. In the comb-like amphiphilic copolymers, localization of the additive at the membrane surface during immersion precipitation casting results in the formation of a dense hydrophilic “brush” of copolymer side chains, while the hydrophobic backbone intermixes with the membrane base component, serving as an anchor for the additive. The antifouling properties come mainly from the hydrophilic part (PEO/PEG) that is an integral part of the copolymer. PEO, for example, has been shown to be a very effective material to prevent adhesion of biomacromolecules and bacterial cells due to its hydrophilicity, large excluded volume, and unique ability to coordinate surrounding water molecules in an aqueous medium [19,20]. Similarly, PEG provides resistance to protein adsorption and cell adhesion by lowering the polymerwater interfacial energy [21,22]. Randomly amphiphilic copolymers are another class of amphiphilic copolymers with differences in chemistry at the molecular-scale level. Such a copolymer has hydrophilic and hydrophobic moieties located next to each other but with a distribution that is random. Due to the random nature, such a surface would be expected to discourage the adsorption of a wide variety of foulants. Baxamusa and Gleason [23] copolymerized the hydrophilic hydroxyethyl methacrylate (HEMA) and the hydrophobic perfluorodecyl acrylate (PFDA) using an initiated CVD technique to create such a surface. The films were deposited on silicon wafers and adsorption studies with a model protein, BSA, were also carried out. It was found that less protein adsorbed onto the copolymer as compared to the pure homopolymers and that a minimum adsorption occurred for amphiphilic chemistry (60% HEMA–40% PFDA). Since protein adsorption and bacterial adhesion are closely interrelated events [24], surfaces that interfere with protein adsorption can be expected to resist the attachment of bacteria as well. Given that bacterial adhesion is a critical early stage event in the overall process of Biofouling [25], this surface may ultimately prove effective in the control/ prevention of membrane Biofouling. Also, the technique used for the deposition of these copolymer films, initiated CVD, is solvent-free, all dry and carried out at ambient temperatures and thus suitable for RO membranes that are polymer-based. In light of the above, it was thought to explore the feasibility of these coatings on standard seawater desalination membranes. This is the first part of a series of studies planned to investigate the antifouling potential of these films on RO membranes. The amphiphilic copolymer film was

129

deposited using the initiated CVD technique on three different commercial membranes. The modified and virgin membranes were then characterized extensively with FESEM, AFM, FTIR, XPS and contact angle goniometer. Short-term cross flow permeation tests were carried out with a synthetic NaCl solution to determine the effect of the coating on the permeate flux and salt rejection. 2. Materials & methods 2.1. Membranes Flat sheets (12 in by 12 in) commercial seawater desalination membranes were purchased from Sterlitech Corp. (USA). Table 1 provides the performance details of the membranes as provided by the manufacturer from tests performed using a water feed with NaCl concentration of 32 g/L (pH = 8) at a temperature of 25 °C and a pressure of 800 psi (55 bar). 2.2. Materials The two monomer species, perfluorodecyl acrylate (97%) and hydroxyethyl methacrylate (99%), and the initiator Ter-butyl-Peroxide (98%), were purchased from Sigma Aldrich, USA and used without further purification. 2.3. Thin-film deposition The two monomers, perfluorodecyl acrylate and hydroxyethyl methacrylate were heated in separate crucibles to 80 °C and 70 °C respectively, while the initiator was kept at room temperature. Vapors of the two monomers and initiator were metered into the iCVD reactor chamber through mass flow controllers (MKS, 1152). The relative flow rates of the monomer gases and of the initiator were adjusted to obtain a composition of 40% PFA on the copolymer. The vapors met and mixed at a common manifold prior to entering the reactor. A throttling butterfly valve (MKS, 653B) was used to control the pressure inside the chamber at 200 mtorr (2.7 × 10−3 bar). The filaments in the reactor were heated to approximately 220 °C to promote radicalization of the initiator. The commercial RO membrane samples were placed on the reactor stage with active layer facing up. The membrane sample was adhesively taped on all sides to prevent any deposition on its backside. The membrane sample was held at a 30 °C by keeping its backside in contact with the temperature-controlled stage maintained at the constant 30 °C. The copolymerization took place directly onto the membrane active layer and resulted in the formation of the co-polymer film with a target thickness of around 20 nm. Film growth was monitored in situ by laser interferometry with the laser focused on a single point on a silicon wafer placed adjacent to the membrane sample. The film deposition was terminated once the laser interferometry indicated attainment of the desired thickness on the silicon wafer. After the depositions, the film thickness and composition on the silicon wafer were determined by Variable Angle Spectroscopy Ellipsometry (VASE, JA Woollam M-2000). Spectroscopic data were obtained at three different angles (65, 70 and 75°) with the wavelength range of 315–700 nm. A Cauchy–Euler model was obtained to fit the data. The uncertainty in measuring the film thickness is approx. ± 5 nm. 2.4. Membrane characterization

Table 1 Manufacturer information for the membranes used in this study. Manufacturer

Code

Service

Flux (L/m2h)

Rejection (%)

Toray GE Osmonics Dow FilmTec

80B AD SW30HR

Seawater RO Seawater RO Seawater RO

27 26 27

99.8 99.5 99.7

ATR-FTIR spectra were obtained using a Nicolet 8700 FTIR spectrometer coupled to a germanium crystal operated at 45° using OMNIC 6.2 software (Thermo Electron Corp., Hampton, NH). The active layer of the membrane was pressed somewhat tightly against the crystal plate. Carbon dioxide and water vapor were continuously purged out during

130

A. Matin et al. / Desalination 343 (2014) 128–139

the measurements. At least two replicates from each membrane sample were measured and 5 readings were taken from different points. Each spectrum represents an average of 16 scans collected in the range 600 to 4000 cm−1 at a resolution of 1 cm−1. X-ray Photoelectron Spectroscopy (XPS) analysis was used to confirm the presence of the copolymer film on the modified membranes. Square-shaped coupons with approximate dimensions 1 cm × 1 cm were cut from each of the coated membranes and mounted on the XPS stage. The analyses were conducted on an Axis Ultra DLD system under ultra-high vacuum conditions (1.6 × 10−12 bar) in the Advanced Nanofabrication Imaging & Characterization Laboratory at King Abdullah University of Science and Technology (KAUST), Saudi Arabia. The survey scan was performed in the binding energy range 0–1000 eV with a resolution of 1 eV. High resolution scans of C 1 s, F 1 s and O 1 s were conducted under similar conditions with 0.05 eV steps, pass energy 20 eV. Surface roughness was determined quantitatively using an atomic force microscope (Agilent Technologies) equipped with a standard SiN cantilever. Dry membrane samples of dimensions 1 cm × 1 cm were mounted on a glass plate using double-sided tape. For better accuracy and precision, measurements were performed at different locations and for variable scan areas. Surface roughness was reported in terms of the average plane roughness (Ra), root-mean-square (RMS) and the relative surface area [26]. Contact angle measurements were performed using an Interfacial Tension/Contact Angle Measurement Device (Kyowa Interface Science Company, Japan) with FAMAS Interface Measurement & Analysis System, version 3.1.3. The sessile drop method was used to measure the contact angle of a 20 μL water droplet placed carefully on the flat membrane surface. A total of 8 measurements at different locations were carried out for each sample. Receding angle measurements were also performed for membranes coated with copolymer films to observe the effects of surface reconstruction. Surface morphology and cross-sections were examined using JEOL 6301 F Field Emission Scanning Electron Microscopy (FESEM). The samples for SEM were coated with a thin layer of gold to make them conducting. For cross-section analysis, samples were submerged in liquid nitrogen to cause embrittlement and then cut very carefully with a sharp blade. 2.5. Filtration studies A cross flow rig was set up as shown in Fig. 1. This consisted of a Hydra-Cell G10 pump (Warner Engineering, Minneapolis, MN, USA); a 30 L feed tank and a Sterlitech CF042 stainless steel membrane cell. A Polystat chiller (Cole Parmer, Inc.) was used to maintain the feed

water temperature around 25 °C. A pulsation dampener was placed on the feed side to minimize pressure fluctuations. The retentate and permeate were circulated back to the feed tank through back pressure regulator valves (Swagelok BP-60 series). For analysis purposes, the permeate was passed through another route with a digital flowmeter (Cole Parmer, USA) and a closed cell for conductivity measurements. A pH meter (Orion, Inc.) with a probe dipped into the feed tank was used for pH and temperature measurements. For consistency, all experiments were conducted at a constant applied pressure of 50 Bar (700 psig) and the temperature of the feed was maintained at 25 °C. The solution pH was unbuffered, but remained constant around 6. 3. Results and discussion 3.1. Surface morphology Fig. 2 (a–f) show SEM images of the different membranes in both the surface-modified as well as the unmodified condition. Two different morphologies are clearly visible in the images: leaf-like and nodular. The SW30HR and TF-RO-AD membranes (both in the virgin and modified state) exhibit leaf-like morphology. On the other hand, the UTC80B membranes from Toray Industries Inc. show more of a rather nodular like morphology. The apparent difference in morphology can be explained by the different processing conditions used by different manufacturers. Any commercial RO membrane consists of a very thin polyamide layer on top of much thicker and porous supporting layers (polysulfone and polyester). The polyamide layer is synthesized by interfacial polymerization that involves a reaction between a polyfunctional amine, and an acid chloride which are dissolved in water and hydrocarbon respectively [27,28]. The polyamide layer formation is influenced by several factors such as monomer types and concentrations, reaction time, curing temperature, solvents and additives [29,30]. The more important observation here is the strikingly similar surface morphology before and after the thin film deposition on all three membranes. This shows that iCVD technique is able to deposit very conformal and continuous coatings on a given substrate. This unique feature of the iCVD vapor phase ambient temperature ultrathin film deposition capability makes it a better choice than several other contemporary thin film deposition techniques. FESEM images of the membrane cross sections were also taken to identify the different layers of the RO membrane and if possible identify the HEMA/PFA copolymer film on the modified membrane surfaces. Fig. 3 a & b show the cross-section of the virgin membrane at lower magnifications, 2 k and 10 k respectively. All the three layers (polyamide,

Fig. 1. Schematic of permeation setup.

A. Matin et al. / Desalination 343 (2014) 128–139

131

Fig. 2. SEM images of the surface morphology of the membranes in the unmodified and modified states; SW30HR (a & b), TF-RO-AD (c & d), and UTC-80B (e & f). Scale bar is 1 μm in all the figures.

polysulfone and polyester) are visible simultaneously in Fig. 3a. The polysulfone layer thickness is estimated to be around 40–50 μm which is in good agreement with values found in the literature [31,32]. Fig. 3c and d are high magnification (75 k) images of the virgin and coated membrane respectively. It becomes a lot easier to differentiate between the polyamide and polysulfone layers at this magnification. The polyamide layer is much denser and nonporous as compared to the polysulfone substrate that is microporous. The polyamide layer thickness was estimated from these figures with the help of the scale bar and it was found to be 250 nm with an uncertainty of ± 25 nm. This value is consistent with the active layer thickness reported in literature for different commercial membranes [33,34]. 3.2. Chemical structure ATR-FTIR was used to identify the different chemical bonds and linkages present in the different layers of the membranes. A broad range (600–4000 cm− 1) of spectra was covered in the scan. After

obtaining the complete picture, specific regions of the spectra, esp. where peaks associated with the HEMA-PFA copolymer film and the polyamide layer could be identified, were focused for further clarity on the details. Fig. 4 shows the complete spectra for a TF-RO-AD membrane in both states i.e. virgin (orange) and coated (blue and green). It can be seen that both are quite similar, though not identical in all respects. Almost all the characteristic peaks associated with the polyamide and polysulfone layers are present in the spectra of both samples. Peaks around 1541, 1609 and 1663 cm−1 (Fig. 6 for more detail) are assigned to amide 2 band, aromatic amide and amide 1 band, respectively [35]. Similarly, the spectra presented characteristic peaks of polysulfone around 1584 and 1243 cm−1 [36]. The most notable difference is the presence of a peak around 1730 cm− 1 in the spectra for the modified membrane that is absent in the other spectra. The differences become more obvious upon focusing on the particular region (Fig. 5). The peak present in the spectra of the modified SW30HR membrane around 1730 cm− 1

132

A. Matin et al. / Desalination 343 (2014) 128–139

(a)

(b) polyamide

polysulfone polysulfone

polyester 20 µm

(c)

(d) polyamide

polysulfone

polysulfone 1000 nm

Fig 3. FESEM images at different magnifications of the cross-section of an RO membrane (a) virgin at 2 k, (b) virgin at 10 k (c) virgin at 75 k, and (d) coated at 75 k.

can be readily associated with the carbonyl groups (C_O) present in the HEMA-PFA copolymer. Such a peak is also observed in some commercial RO membranes (SE and SG membranes from GE Osmonics, Inc.) that is assignable to C_O stretching observed in esters [34]. The esterification reaction occurs in the presence of PVA during interfacial polymerization where in the OH groups of the carboxyl moiety and those of the PVA interact to give a somewhat hydrophobic surface. Choi et al. [37] performed surface modification of seawater reverse osmosis (SWRO) membranes with methyl methacrylate–hydroxy poly(oxyethylene) methacrylate (MMAHPOEM) comb-copolymer for improved fouling resistance. They observed a similar peak at 1730 cm− 1 for the modified membrane and attributed it to a carbonyl group in the copolymer coating. The other observation that indicates or rather confirms the presence of the copolymer film on the membranes is the spotlight on the region around 1200 cm−1 (Fig. 6). Upon increasing the scale resolution several times, a peak is observed at exactly 1205 cm−1 for both the coated and uncoated membranes. However, it is evident from the relevant figure that this peak is more pronounced in the modified membrane as compared to the commercial one. This peak can be associated to the symmetric stretches of the pendant fluoroalkyl chain present in the PFA molecule [38]. A question though arises. There are some other peaks assignable to the HEMA-PFA copolymer that are also observed in the spectra of the

uncoated membrane. For example, peaks corresponding to asymmetric stretches of the pendant fluoroalkyl chain around 1240 cm−1 and a peak centered around 1152 cm−1 corresponding to CF2–CF3 stretching frequency at the end of pendant group [38]. The answer is that the above-mentioned peaks are dwarfed by peaks at similar wavenumbers originating from the polysulfone support layer of the RO membrane. For instance, a peak around 1240 cm−1 has been identified with the aromatic ether band present in the PS support layer [39]. Likewise, a peak around 1150 cm−1 apparently seems to overlap with the peak associated with CF2–CF3 stretching frequency from the PFA molecule. In addition to confirm the presence of the copolymer film on the polyamide membrane, the FTIR spectra also confirm that the polymers were true copolymers, not physical blends of homopolymers. A single carbonyl peak centered around 1730 cm−1 indicates a random copolymer with a single effective carbonyl bonding environment. The locations of the carbonyl stretches for the pure homopolymers are 1727 cm−1 for HEMA and 1741 cm−1 for PFDA [40]. Had the polymers been codeposited, rather than copolymerized, a split carbonyl peak would be observed because of two distinct carbonyl bonding environments [41]. 3.3. Surface composition Since XPS probes the ultra-shallow depth of any surface (~5 nm depth), it was also used to confirm the presence of the deposited film

133

0.08 0.06 0.04 0.00

0.02

Absorbance Units

0.10

A. Matin et al. / Desalination 343 (2014) 128–139

4000

3500

3000

2500

2000

1500

1000

Wavenumber cm-1 Fig. 4. FTIR spectra of SW30HR membrane over the complete range. The spectra in orange correspond to the unmodified membrane whereas the blue and green belong to the sample coated with the HEMA-PFA copolymer.

peaks indicating the dominating presence of the fluorinated alkyl side chains (CF2–CF3) on the very top surface. For intermediate/amphiphilic chemistry, this represents a preferential outward orientation for the fluorinated moieties. Similar fluorine enrichment at surfaces has been observed previously in copolymer thin films [42]. High resolution C 1 s scans were also done for both the virgin and modified polyamide. For the unmodified membrane, a single peak centered around 283 eV is observed (Fig. 8 a). On the other hand, a secondary peak of reduced intensity around 288 eV is also seen for the coated membranes (Fig. 8 b). The primary peak around 283 eV corresponds to a weak electron-withdrawing environment and is assigned to the vinyl (C\C) and carbon–nitrogen (C\N) bonds in the copolymer film and polyamide respectively [43]. The secondary peak around 288 eV is

0 -2500 -2000 -1500 -1000 -500

Transmittance [%]

500

1000 1500

by determining the elemental composition of membrane surfaces coated with the HEMA-PFA copolymer film. The main constituents of the copolymer are carbon, oxygen and fluorine and the concentrations of these elements were determined using survey scans. Table 2 shows the results of a survey scan for the coated as well as the virgin UTC80B membrane. The presence of fluorine confirms the deposition of the HEMA-PFA copolymer film on the membrane surface. Fig. 7 shows the result generated from a survey scan performed on a UTC80B membrane in both conditions, i.e. coated and uncoated. The binding energies are approximately 285, 396, 532 and 688 eV for C 1 s, N 1 s, O 1 s and F 1 s respectively. The main difference is the presence of the fluorine peak in the coated membrane spectra (Fig. 8b) that has an intensity roughly 4 times that of both carbon and oxygen

C=O

1900

1800

1700

1600

1500

Wavenumber cm-1 Fig. 5. ATR-FTIR spectra of unmodified (pink) and modified (red) SW30HR membranes in the finger print region.

A. Matin et al. / Desalination 343 (2014) 128–139

0.014 0.012 0.008

0.010

Absorbance Units

0.016

0.018

134

1220

1215

1210

1205

1200

1195

1190

1185

Wavenumber cm-1 Fig. 6. Portion of FTIR spectra showing the peak associated with the pendant fluoroalkyl chain of the PFA molecule around 1205 cm−1. Note the significant bump in the spectra of coated membranes (blue and green) as compared to the relatively flat trajectory for the commercial one (orange).

associated with carbon atoms in a relatively higher electron withdrawing environment (C–F).

desalination using reverse osmosis as firstly, it ensures uniform permeate flux throughout and secondly, it guarantees an antifouling layer of identical thickness everywhere.

3.4. Surface roughness 3.5. Contact angle Atomic Force Microscopy (AFM) was utilized to analyze the surface topology of the commercial membranes in the virgin and modified states (Fig. 9). Table 3 shows the average and root-mean-square roughness values for the three commercial membranes both in the virgin and modified state. It can be seen that although made from the same material (polyamide), the commercial membranes have different roughness values from each other that are in very good agreement with a similar study carried out by Widjaya et al. [44]. Several researchers have studied the surface properties of different commercial membranes from the same manufacturer and from different manufacturers [44,45]. It was found that the different morphological properties of the RO membranes are mainly the result of differences in the synthesis conditions: 1. using different monomer types, 2. subjecting the membrane to post-fabrication treatments, such as coating the membrane with an alcohol rich aliphatic polymer [46]. The other and perhaps more relevant observation is the negligible change in surface roughness after the deposition of the HEMA-PFA copolymer film. This is attributed to the unique nature of the iCVD technique that allows for the synthesis and deposition of such smooth and conformal coatings. As has been previously observed with deposition in trenches, the film is deposited equally on the ridges and valleys of the RO membranes, thus ensuring uniform thickness everywhere. This has important implications for seawater

Table 2 Elemental composition (atomic percent) of the very top surface of SW30HR membrane coated with the copolymer film of amphiphilic composition. Sample

C

O

N

F

Virgin PA Coated PA

69.53 ± 0.5 60.5 ± 0.5

20.64 ± 0.5 12.5 ± 0.5

9.83 ± 0.4 –

– 26.4 ± 0.6

Contact angle is a measure of the hydrophilicity/hydrophobicity of a surface; the lower the angle, the higher the hydrophilicity and vice versa. Researchers have found a strong correlation between the surface hydrophilicty and fouling rate; with hydrophilic surfaces being less prone to fouling [47]. Table 4 gives the average contact angle values for the different membranes both in the virgin and modified states. The average values for the unmodified membranes are in good agreement with those of Widjaya et al. [44] and consistent with the range of 43–49° noted by Tang et al. as being characteristic of uncoated polyamide membranes [35]. The UTC-80B membrane has a somewhat lower contact angle that can be associated with either a coating layer or modified polyamide chemistry [35]. Table 4 also shows the mean values of the receding angle measured approximately 30 min after placing the water droplet on the membrane surface and allowing it to evaporate. It can be seen that there is a decrease in the angle which is more pronounced in the case of membranes with the copolymer film. This observation is attributed to the reorientation of chemical moieties present on the surface: in dry air or vacuum, the relatively hydrophobic entities such as NH2 from polyamide and CF2 from the PFA molecule, point outwards, whereas the more hydrophilic ones orient inwards. Upon exposure to water, the trend reverses and causes a decrease in the contact angle. A similar trend was also observed by Baxamusa and Gleason [23] who deposited HEMA-PFA copolymer films of varying chemistries on Si wafers. The more significant change in angle for the coated membranes as opposed to the virgin ones can be explained by the greater hydrophobicity of the fluoroalkyl groups in HEMA-PFA as compared to the amide groups in the polyamide layer of the RO membrane. The commercial membranes are already somewhat hydrophilic and one would not expect major surface reconstruction upon exposure to water. The surface reconstruction has important consequences from the point of view of membrane biofouling. There are several stages in the

A. Matin et al. / Desalination 343 (2014) 128–139

(a)

(b) F 1s

Fig. 7. Results of a survey scan on UTC-80B membrane (a) without coating and (b) with the copolymer film.

135

136

A. Matin et al. / Desalination 343 (2014) 128–139

Fig. 8. High resolution C 1 s scans for the UTC80B membrane (a) virgin and (b) modified.

formation of a Biofilm that ultimately causes the undesired deterioration in membrane performance. The formation of a conditioning film (consisting of biopolymers and organic macromolecules) and the

subsequent adhesion of microorganisms are initial steps in the overall process and may take several hours to materialize. Given the rapid transformation of the modified membrane surfaces, it is the amphiphilic

A. Matin et al. / Desalination 343 (2014) 128–139

137

Fig. 9. AFM images of the surface of UTC-80B membrane (a) virgin and (b) modified with copolymer film.

chemistry and not the hydrophobic one that the bacteria have to encounter. 3.6. Permeation studies Table 5 shows the mean values for the permeate water flux for all the three membranes in the virgin as well as the modified condition. The average flux for the commercial membranes is in the range of 20–50 L/m2h which are the typical range for SWRO membranes [44]. The differences between the membranes are attributed to the differences in permeability that can be altered either by the application of a coating on top of the polyamide, or alternatively through changes in the extent of polymer crosslinking. The percentage decline in permeate flux due to the presence of the copolymer film is calculated as follows: Jv−Jc  100 Jv where Jv and Jc are the permeate water fluxes in the virgin and modified conditions respectively. Table 5 also shows the flux change for the commercial membranes upon deposition of a HEMA-PFA copolymer film with a thickness of 20 nm. It is observed that the flux decline is within 10% of the original value for all the three membranes. Just as permeate water flux determines the efficiency of the RO process, percentage salt rejection is a good indicator of the permeate water quality. In fact, the latter is what makes the water fit for human consumption. For this reason, salt rejection was also determined for all the commercial membranes both with and without the coating. The salt rejection was calculated as follows: "

Cp 1− Cf

!#  100

Table 6 shows the salt rejection values as well as the effect of the copolymer film on the salt rejection for the three membranes. Since these are designed for seawater desalination, it is normal to observe salt rejection N99%. However, the more relevant observation here is the negligible change in rejection values after the deposition of HEMA-PFA copolymer films. Assuming an approximate thickness of 250 nm for the polyamide layer (from the cross-sectional FESEM images above, Fig. 3), an addition of a 20 nm coating is equivalent to an increase in thickness of the barrier layer by around 10%. Although the copolymer film does swell upon exposure to water, but for amphiphilic chemistry (~40% PFDA), the rate of water uptake is very slow and the swelling is negligible [23]. Therefore, the hydrated thickness can be assumed to be nearly the same as the thickness without water exposure. This implies that an increase in total thickness of around 10% is causing a flux decline by approximately the same amount. Given that the flux is inversely proportional to the thickness, we can conclude that the copolymer films have water permeability comparable to that of the polyamide. Together with the rejection results, the presence of these coatings does not seem to adversely affect the membrane performance in a significant manner.

4. Conclusions Thin copolymer films of hydroxyethyl methacrylate and perfluorodecyl acrylate of an intermediate chemistry (40% PFA) were synthesized and deposited on commercial RO membranes using an initiated CVD technique. FTIR and XPS analyses confirmed the presence of the copolymer film on the membrane surface. FESEM images and AFM measurements prove the conformality and smoothness of these coatings as there is negligible change in the membrane surface morphology and topology. Contact angle measurements show the modified surfaces to be initially very hydrophobic but rapidly

where Cf and Cp are salt concentrations in the feed and permeate, respectively. Table 3 roughness values for the three commercial membranes in the modified as well as unmodified state. Membrane

SW30HR TF-RO-AD UTC-80B

Virgin state

Table 4 Average contact angle values for the commercial membranes in the modified and unmodified states. The receding angles were measured approx. 30 min after placing the water droplet on the surface. Note the significant reduction in the angle after a brief exposure to water. Membrane

Modified state

Ra

Rms

Ra

Rms

60 ± 3 52 ± 2 35 ± 2

75 ± 5 64 ± 4 47 ± 3

63 ± 2 50 ± 5 36 ± 4

78 ± 4 63 ± 7 48 ± 5

SW30HR TFROAD UTC80B

Virgin

Coated with 40% PFA

Static

Receding

Δ

Static

Receding

Δ

43.5 ± 2.5 46.4 ± 2.7 34.5 ± 0.6

12.5 ± 1.7 23.9 ± 2.1 10.1 ± 1.8

31.0 22.5 24.4

84 ± 2.3 86 ± 2.0 102 ± 2.1

25 ± 1.4 20 ± 1.5 44 ± 1.8

59.0 66.0 58.0

138

A. Matin et al. / Desalination 343 (2014) 128–139

Table 5 Change in the flux values for the commercial membranes in the presence of the copolymer film. Membrane

SW30HR TFROAD UTC80B

Permeate water flux (L/m2h) Jv

Jc

ΔJ (%)

37 25 47

34 22.5 42

8.1 10 10.6

Table 6 Salt rejection values for the membranes in the virgin and modified states and also the difference. Membrane

Salt rejection (%) Rv

Rc

ΔR (%)

SW30HR TFROAD UTC80B

99.5 99.2 99.3

99.3 99.4 99.7

−0.2 +0.2 +0.4

transforming to an amphiphilic one. Permeation tests reveal the film to be as permeable as the polyamide layer of the RO membrane and do not cause a significant decline in water flux. To summarize, the deposition technique and the antifouling coatings are quite compatible with the RO membranes. Further studies need to be conducted to prove the effectiveness of these films in controlling or preventing membrane Biofouling. For this purpose, long-term Biofouling tests in an experimental setup that simulates actual conditions prevalent in industrial RO plants are strongly recommended. Acknowledgment The authors would like to thank the King Fahd University of Petroleum and Minerals (KFUPM) in Dhahran, Saudi Arabia, for funding through the Center for Clean Water and Clean Energy at MIT and KFUPM under PROJECT NUMBER R5-CW-08. We would also like to thank Faisal Wali and Mohammad Nejib Hedhili of King Abdullah University of Science and Technology (KAUST), Saudi Arabia for the performing the XPS analyses. In addition, the authors would like to thank Owais Badr-uz-Zaman (KFUPM) for his continuous assistance with the permeation tests and Sadaqat and Luqman (KFUPM) for their help with the FESEM. References [1] M.H. Li, Reducing specific energy consumption in Reverse Osmosis (RO) water desalination: an analysis from first principles, Desalination 276 (2011) 128–135. [2] C. Liu, K. Rainwater, L.F. Song, Energy analysis and efficiency assessment of reverse osmosis desalination process, Desalination 276 (2011) 352–358. [3] W. Zhou, L. Song, Experimental study of water and salt fluxes through reverse osmosis membranes, Environ. Sci. Technol. 39 (2005) 3382–3387. [4] W.S. Ang, S. Lee, M. Elimelech, Chemical and physical aspects of cleaning of organic-fouled reverse osmosis membranes, J. Membr. Sci. 272 (2006) 198–210. [5] X. Zhu, M. Elimelech, Colloidal fouling of reverse osmosis membranes: measurements and fouling mechanisms, Environ. Sci. and Technol. 31 (1997) 3654–3662. [6] M. Al-Ahmad, F.A. Abdul Aleem, A. Mutiri, A. Ubaisy, Biofouling in RO membrane systems. Part 1: fundamentals and control, Desalination 132 (2000) 173–179. [7] E.G. Darton, M. Fazel, A statistical review of 150 membrane autopsies, Proc. Int. Water Conference (2001) 157–163. [8] J. Xu, G. Ruan, X. Chu, Y. Yao, B. Su, A pilot study of UF pretreatment without any chemicals for SWRO desalination in China, Desalination 207 (2007) 216–226. [9] H.C. Flemming, G. Schaule, T. Griebe, J. Schmitt, A. Tamachkiarowa, Biofouling — the Achilles heel of membrane processes, Desalination 113 (1997) 215–225. [10] H. Flemming, A. Tamachkiarowa, J. Klahre, J. Schmitt, Monitoring of fouling and biofouling in technical systems, Water Sci. and Technol. 38 (1998) 291–298.

[11] H. Flemming, Reverse Osmosis membrane biofouling, Exp. Thermal Fluid Sci. 14 (1997) 382–391. [12] A. Matin, Z. Khan, S.M.J. Zaidi, M.C. Boyce, Biofouling in reverse osmosis membranes for seawater desalination: phenomena and prevention, Desalination 281 (2011) 1–16. [13] I.C. Kim, K.H. Lee, Dyeing process wastewater treatment using fouling resistant nanofiltration and reverse osmosis membranes, Desalination 192 (2006) 246–251. [14] J.S. Louie, I. Pinnau, I. Ciobanu, K.P. Ishida, A. Ng, M. Reinhard, Effects of polyether-polyamide block copolymer coating on performance and fouling of reverse osmosis membranes, J. Membr. Sci. 280 (2006) 762–770. [15] S.C. Yu, Z.H. Lu, Z.H. Chen, X.S. Liu, M.H. Liu, C.J. 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. 371 (2011) 293–306. [16] A. Asatekin, E.A. Olivetti, A.M. Mayes, Fouling resistant, high flux nanofiltration membranes from polyacrylonitrile-graft-poly(ethylene oxide) J, Membr. Sci. 332 (2009) 6–12. [17] A. Adout, S. Kang, A. Asatekin, A.M. Mayes, M. Elimelech, Ultrafiltration membranes incorporating amphiphilic comb copolymer additives prevent irreversible adhesion of bacteria, Environ. Sci. and Technol. 44 (2010) 2406–2411. [18] J.Y. Park, M.H. Acar, A. Akthakul, W. Kuhlman, A.M. Mayes, Polysulfone-graftpolyethylene glycol graft copolymers for surface modification of polysulfone membranes, Biomaterials 27 (2006) 856–865. [19] P. Hamilton-Brown, T. Gengebach, H.J. Griesser, L. Meagher, End terminal, poly(ethylene oxide) graft layers: surface forces and protein adsorption, Langmuir 25 (2009) 9149–9156. [20] K. Bergstrom, E. Osterberg, K. Holmberg, A.S. Hoffman, T.P. Schuman, A. Kozlowski, J.M. Harris, Effects of branching and molecular-weight of surface-bound poly(ethylene oxide) on protein rejection, J. Biomater. Sci. Polym. Ed. 6 (1994) 123–132. [21] S. Krishnan, C.J. Weinman, C.K. Ober, Advances in polymers for anti-biofouling surfaces, J. Mater. Chem. 18 (2008) 3405–3413. [22] L.P. Zhu, Y.Y. Xu, X.Z. Wei, B.K. Zhu, Hydrophilic modification of poly(phthalazine ether sulfone ketone) ultrafiltration membranes by the surface immobilization of poly(ethylene glycol) acrylates, Desalination 242 (2009) 96–109. [23] S.H. Baxamusa, K.K. Gleason, Random copolymer films with molecular-scale compositional heterogeneities that interfere with protein adsorption, Adv. Funct. Mater. 19 (2009) 3489–3496. [24] R.G. Chapman, E. Ostuni, M.N. Liang, G. Meluleni, E. Kim, L. Yan, G. Pier, H.S. Warren, G.M. Whitesides, Polymeric thin films that resist the adsorption of proteins and the adhesion of bacteria, Langmuir 17 (2001) 1225–1233. [25] A. Matin, Z. Khan, S.M.J. Zaidi, M.C. Boyce, Biofouling in reverse osmosis membranes for seawater desalination: phenomena and prevention, Desalination 28 (2011) 1–16. [26] M. Liu, D. Wu, S. Yu, G. Congjie, 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. 326 (2009) 205–214. [27] N.Y. Yip, A. Tiraferri, W.A. Philip, J.D. Schiffman, M. Elimelech, High performance thin-film composite forward osmosis membrane, Environ. Sci. and Technol. 44 (2010) 3812–3818. [28] V. Ramachandhiran, A.K. Ghosh, S. Prabhakar, P.K. Tewari, Preparation and separation performance studies on composite polyamide membranes using different amine systems and support membranes, J. Polym. Mater. 26 (2009) 177–185. [29] Y. Jin, Z. Su, Effects of polymerization conditions on hydrophilic groups in aromatic polyamide thin films, J. Membr. Sci. 330 (2009) 175–179. [30] A.K. Ghosh, B.-H. Jeong, X. Huang, E.M.V. Hoek, Impacts of reaction and curing conditions on polyamide composite reverse osmosis membrane properties, J. Membr. Sci. 311 (2008) 34–45. [31] A.K. Ghosh, R.C. Bindal, S. Prabhakar, P.K. Tewari, Composite polyamide reverse osmosis (RO) membranes — recent developments and future directions, BARC Newsletter 321 (2011) 43–51. [32] B. Mi, O. Coronell, B.J. Mari˜nas, F. Watanabe, D.G. Cahill, I. Petrov, Physico-chemical characterization of NF/RO membrane active layers by Rutherford backscattering spectrometry, J. Membr. Sci. 282 (2006) 71–81. [33] M. Fathizadeh, A. Aroujalian, A. Raisi, Effect of lag time in interfacial polymerization on polyamide composite reverse osmosis membrane with different hydrophilic sub layers, Desalination 284 (2012) 32–41. [34] O. Akin, F. Temelli, Probing the hydrophobicity of commercial reverse osmosis membranes produced by interfacial polymerization using contact angle, XPS, FTIR, FE-SEM and AFM, Desalination 278 (2011) 387–396. [35] C.Y. Tang, Y.N. Kwon, J.O. Leckie, Effect of membrane chemistry and coating layer on physicochemical properties of thin film composite polyamide RO and NF membranes: I. FTIR and XPS characterization of polyamide and coating layer chemistry, Desalination 242 (2009) 168–182. [36] Y.-N. Kwon, S. Hong, H. Choi, T. Tak, Surface modification of a polyamide reverse osmosis membrane for chlorine resistance improvement, J. Membr. Sci. 415–416 (2012) 192–198. [37] H. Choi, J. Park, T. Tak, Y.-N. Kwon, Surface modification of seawater reverse osmosis (SWRO) membrane using methyl methacrylate–hydroxy poly(oxyethylene) methacrylate (MMA-HPOEM) comb-copolymer and its performance. [38] D. Lin-Vien, N.C.W. Fately, J. Grasselli, The Handbook of Infrared and Raman Characteristic Frequencies of Organic Molecules, Academic Press, San Diego, CA, 1991. [39] H. Susanto, M. Ulbricht, Characteristics, performance and stability of polyethersulfone ultrafiltration membranes prepared by phase separation method using different macromolecular additives, J. Membr. Sci. 327 (2009) 125–135. [40] M. Gupta, K.K. Gleason, Initiated chemical vapor deposition of poly (1H, 1H, 2H, 2Hperfluorodecyl acrylate) thin films, Langmuir 22 (2006) 10047–10052.

A. Matin et al. / Desalination 343 (2014) 128–139 [41] K.K.S. Lau, K.K. Gleason, All-dry synthesis and coating of methacrylic acid copolymers for controlled release, Macromol. Biosci. 7 (2007) 429–434. [42] Y. Mao, K.K. Gleason, Vapor-deposited fluorinated glycidyl copolymer thin films with low surface energy and improved mechanical properties, Macromolecules 39 (2006) 3895. [43] K. Boussu, J. De Baerdemaeker, C. Dauwe, M. Weber, K.G. Lynn, D. Depla, S. Aldea, I.F.J. Vankelecom, C. Vandecasteele, B. Van der Bruggen, Physico-chemical characterization of nanofiltration membranes, Chem. Phys. Chem. 8 (2007) 370–379. [44] A. Widjaya, T. Hoang, G.W. Stevens, S.E. Kentish, A comparison of commercial reverse osmosis membrane characteristics and performance under alginate fouling conditions, Sep. Purif. Technol. 89 (2012) 270–281.

139

[45] S.Y. Kwak, S.G. Jung, Y.S. Yoon, D.W. Ihm, Details of surface features in aromatic polyamide reverse osmosis membranes characterized by scanning electron and atomic force microscopy, J. Polym. Sci. B Polym. Phys. 37 (1999) 1429–1440. [46] C.Y. Tang, Y.-N. Kwon, J.O. Leckie, Probing the nano- and micro-scales of reverse osmosis membranes — a comprehensive characterization of physicochemical properties of uncoated and coated membranes, J. Membr. Sci. 287 (2007) 147–156. [47] K. Boussu, Y. Zhang, J. Cocquyt, P. van der Meeren, A. Vololdin, C. Van Haesendonck, J.A. Martens, B. Van der Bruggen, Characterization of polymeric nanofiltration membranes for a systematic analysis of membrane performance, J. Membr. Sci. 278 (2006) 418–427.