当前位置:首页 >> 能源/化工 >>

Intensified cleaning of organic-fouled reverse osmosis membranes by thermo-responsive polymer (TRP)


Journal of Membrane Science 392–393 (2012) 181–191

Contents lists available at SciVerse ScienceDirect

Journal of Membrane Science
journal homepage: www.elsevier.com

/locate/memsci

Intensi?ed cleaning of organic-fouled reverse osmosis membranes by thermo-responsive polymer (TRP)
Sanchuan Yu a,b,? , Zhihai Chen a,b , Jingqun Liu a,b , Guohua Yao a,b , Meihong Liu a,b , Congjie Gao c
a b c

Engineering Research Center for Eco-Dyeing & Finishing of Textiles, Ministry of Education of China, Zhejiang Sci-Tech University, Hangzhou 310018, People’s Republic of China Key Laboratory of Advanced Textile Materials and Manufacturing Technology of Education Ministry, Zhejiang Sci-Tech University, Hangzhou 310018, People’s Republic of China The Development Center of Water Treatment Technology, SOA, Hangzhou 310012, People’s Republic of China

a r t i c l e

i n f o

a b s t r a c t
The role of thermo-responsive polymer (TRP) in the cleaning of organic-fouled polyamide-based reverse osmosis membranes was systematically investigated in this study. Fouling and cleaning experiments were performed employing a laboratory-scale cross-?ow test unit. Following accelerated organic fouling runs with bovine serum albumin (BSA), cleaning experiments were conducted using de-ionized water and de-ionized water containing TRP with low critical solution temperature under various conditions, respectively. The separation performances of the fresh, fouled and cleaned membranes were characterized through permeation tests. It was found that the phase transition of the TRP that had diffused into the fouling layer would facilitate the removal of foulants located on membrane surface, and thereby improve the cleaning ef?ciency. The ef?ciency of the intensi?ed cleaning with TRP solution was largely affected by the type and concentration of TRP, as well as the soaking time of TRP solution. The ef?ciency of the intensi?ed cleaning could be enhanced by increasing the concentration and/or prolonging the soaking time of TRP solution. Moreover, membrane surface characterization via scanning electron microscopy, attenuated total re?ectance infrared and surface contact angle measurements also con?rmed the bene?cial effect of TRP on the removal of the deposited foulants from the membrane surface. ? 2011 Elsevier B.V. All rights reserved.

Article history: Received 5 July 2011 Received in revised form 18 December 2011 Accepted 21 December 2011 Available online 29 December 2011 Keywords: Membrane cleaning Reverse osmosis membrane Thermo-responsive polymer Organic fouling Intensi?ed cleaning

1. Introduction Reverse osmosis (RO) process, which uses polymeric semipermeable membranes to achieve molecular separation under the driving force of hydraulic pressure, is a well accepted technique for converting feed aqueous stream into a puri?ed permeate, and has gained extensive attention in separation, water treatment and reclamation because of its advantages such as high permeate quality, ease of operation, minimal chemical addition, as well as low energy requirement [1–3]. However, the successful utilization of RO technology is greatly limited by membrane fouling, which is a major obstacle for the application of membrane technology [3,4]. The major fouling mechanisms of reverse osmosis membranes are scaling, organic fouling and biofouling [5,6]. When fouling occurs on the membrane surface, the performance of membrane separation process, such as ef?ciency (e.g. ?ux, membrane permeability, permeate recovery, as

? Corresponding author at: Engineering Research Center for Eco-Dyeing & Finishing of Textiles, Ministry of Education of China, Zhejiang Sci-Tech University, Hangzhou 310018, People’s Republic of China. Tel.: +86 571 86843217; fax: +86 571 86843217. E-mail addresses: yuschn@mail.hz.zj.cn, yuschn@163.com (S. Yu). 0376-7388/$ – see front matter ? 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.memsci.2011.12.025

well as permeate quality) and effectiveness (e.g. rejection of selectivity) are typically reduced, leading to an increase in production cost due to the increased energy demand and chemical cleaning, reduced membrane life and additional labor for maintenance [7,8]. In order to mitigate the problem of membrane fouling and promote the application of RO technology, strategies including pretreatment [9,10], design of special modules and devices [11,12] and development of anti-fouling RO membranes [13–16] have been examined. However, these preventive strategies can only slow the fouling rate, membrane fouling is inevitable and membrane cleaning is an essential step in maintaining the performance of the membrane process. Many studies have been carried out in the ?eld of membrane cleaning to develop new cleaning techniques as well as to improve the cleaning ef?ciency. For example, Madaeni et al. [17] studied the chemical cleaning of the fouled polyamide-based reverses osmosis membranes using acid, alkaline, surfactant and detergent solutions, respectively. They reported that the cleaning ef?ciency depending on the type of the cleaning agent and its concentration, and that operating conditions such as cross-?ow velocity, turbulence in the vicinity of the membrane surface, temperature, pH and cleaning time also played an important role in the cleaning process. Feng et al. [18] investigated the on-line ultrasonic defouling of the polyamide-based reverse osmosis membranes. It was found that the permeate ?ux

182

S. Yu et al. / Journal of Membrane Science 392–393 (2012) 181–191

Fig. 1. Schematic diagram of the cleaning of fouled membrane using thermo-responsive polymer.

of the membrane increased signi?cantly with virtually no decrease in rejection in the presence of ultrasonication, and ultrasound could effectively remove both the inorganic foulants of CaSO4 and Fe(OH)3 and the organic foulant carboxymethyl cellulose on the membrane surface. Ang et al. [19] focused their study on the role of chemical and physical interactions in cleaning of organic-fouled reverse osmosis (RO) membranes. They reported that the cleaning of organic-fouled RO membranes was accomplished by the chemical reaction between cleaning agents and foulants in the fouling layer, followed by the mass transfer of loosened foulants from the fouling layer to the bulk solution. The cleaning ef?ciency of the organic-fouled reverse osmosis was largely affected by both the chemical reaction between the cleaning agent and the foulants in the fouling layer and the mass transfer of chemical agents (from the bulk phase to the fouling layer) and foulants (from the fouling layer to the bulk phase). In summary, optimum membrane cleaning requires understanding, at a fundamental level, the interactions between the foulants and the membrane as well as the effects of the cleaning procedure on the removal of the deposited foulants and the membrane performance. The success of chemical cleaning methods depends on many factors including membrane material, foulant nature, cleaning agent, temperature, pH, concentration of the cleaning chemical, contact time between the chemical solution and the membrane, and operation conditions such as cross-?ow velocity and pressure. Therefore, extensive research work is certainly needed to investigate and explore new ideas and techniques in the ?eld of membrane cleaning and restoration for improved ef?ciency. Thermo-responsive polymer (TRP) with low critical solution temperature (LCST) is soluble in water and exists in an extended random coil conformation that is fully hydrated at temperatures lower than its LCST, and the polymer chains will hydrophobically fold as a result of dehydration and assemble to form a phase separating state at temperatures above LCST [20]. A novel idea was introduced in the present study to intensify the cleaning of organicfouled reverse osmosis membranes by using thermo-responsive polymers with low critical solution temperature. Fig. 1 shows a general schematic representation of how the thermal-responsive polymers facilitate the removal of foulants deposited on the membrane surface. At temperatures below LCST, the thermal-responsive polymers are soluble and will diffuse into the fouling layer on the membrane surface. When the diffused thermo-responsive polymers are made insoluble by raising the temperature of the soaking solution above the LCST of the diffused thermo-responsive polymers, the fouling layer on the membrane surface will become

structurally loose and thereby can be easily washed off from the membrane surface through rinsing. The objective of this study was to evaluate and demonstrate the availability of thermo-responsive polymer in enhancing the cleaning ef?ciency of fouled reverse osmosis membrane. To achieve this purpose, polymers such as poly(N-isopropylacrylamide) (P(NIPAm)), poly(N-isopropylacrylamide-co-acrylic acid) (P(NIPAm-co-AAc)) and poly (N-isopropylacrylamide-co-acrylamide) (P(NIPAm-co-Am)) were used as the thermo-responsive polymer with LCST, and the polyamide thin-?lm composite reverse osmosis membranes with bovine serum albumin (BSA) fouling layer were used as the fouled RO membranes. The procedure adopted to clean the BSA-fouled reverse osmosis membranes included: soaking the fouled membrane with aqueous solution containing TRP at temperature below LCST, making the TRP diffused into the fouling layer insoluble by exposing the fouled membrane with warm aqueous solution of temperature higher than LCST, and ?nally rinsing the membrane with de-ionized water of room temperature. For comparison, the BSA-fouled membranes were also cleaned only using de-ionized water following the same procedure. The cleaning ef?ciency was studied by comparing the salt rejection and water ?ux of the fresh, fouled and cleaned membranes through permeation tests. Additionally, SEM, ATR-FTIR and surface contact angle measurements were also used to characterize the surfaces of the fresh, fouled and cleaned membranes to investigate the changes of membrane surface properties. 2. Theory 2.1. Solution-diffusion model Transport through reverse osmosis membranes is typically described using a solution-diffusion model [21]. The equation describing water ?ux across the reveres osmosis membrane is: Jw = Lp ( P ? ?) (1)

where Jw is volumetric water ?ux (l/(m2 h), Lp is water permeability coef?cient of the membrane (l/m2 h MPa), P is the applied trans-membrane pressure difference (MPa), and is the osmotic pressure difference between the feed and the permeate solutions (MPa). The equation describing the salt transport across the reverse osmosis membrane is: Js = B(Cf ? Cp ) (2)

S. Yu et al. / Journal of Membrane Science 392–393 (2012) 181–191

183

O C

O H C N

H N

O C

O C

H N

H N

n
C N O H COOH

1-n

N

H

Scheme 1. Chemical structure of the barrier layer of the polyamide thin-?lm composite reverse osmosis membrane used in this study.

where Js is salt ?ux (mg/m2 h), B is the salt permeability coef?cient of the membrane (l/(m2 h)), and Cf and Cp are the salt concentrations in the feed and permeate streams, respectively (mg/l). 2.2. Resistance-in-series model Resistance-in-series model has been widely applied to describe the permeation ?ux of the membrane processes. In this model, membrane hydraulic resistance (Rm ) and hydraulic resistances of fouling layer (Rf ) on membrane surface are incorporated, the water permeability coef?cient of a fouled membrane, Lp , is de?ned as 1/ w (Rm + Rf ), and the volumetric water ?ux (Jw ) of a fouled membrane is given by [22,23]: Jw = where P?
w (Rm w

3.2. Flat-sheet cross-?ow permeation test unit A laboratory-scale cross-?ow test unit as shown schematically in Fig. 2 was used to carry out the membrane fouling and cleaning experiments, as well as the permeation tests in a total recycle mode. It consists of two thin channel rectangular ?at-sheet stainless steel test cells in parallel, a high-pressure pump, a feed reservoir equipped with stirrer and a temperature controller. The desired pressure and feed ?ow rate were achieved by adjusting the bypass needle valve and back-pressure regulator. The cross-?ow velocity can also be tuned by varying the thickness of the feed channel of the test cell. The effective membrane area for permeation cell was 120.0 cm2 . 3.3. Membrane water ?ux and salt rejection tests Reverse osmosis performance in terms of permeate water ?ux and salt rejection for the fresh, fouled and cleaned membranes were tested using 500 mg/l sodium chloride (NaCl) aqueous solution at 1.0 MPa, 25.0 ? C and pH 6.8 employing the ?at-sheet cross-?ow test unit described above. The membranes were operated for at least 1 h before any data were collected. The permeate water ?ux was determined by measuring the permeate volume collected over a certain period in terms of liter per square meter per hour (l/m2 h):

? + Rf )

(3)

is the water viscosity (Pa s).

3. Experimental 3.1. Materials and reagents The commercial ?at sheet thin-?lm composite polyamide low-pressure reverse osmosis membrane used in this study was supplied by the Development Center of Water Treatment Technology (Hangzhou, China). It was produced by interfacial polymerization of m-pheylenediamine with trimesoyl chloride on a polysulfone porous substrate. The surface of the thin-?lm composite polyamide membrane is expected to contain cross-linked structure and non cross-linked structure containing carboxylic acid groups and its chemical structure is described in Scheme 1. The reverse osmosis membranes received were rinsed thoroughly with de-ionized water and stored wetly before use and characterization. Such thermo-responsive polymers with LSCT as poly(Nisopropylacrylamide) (P(NIPAm)), poly(N-isopropylacrylamideco-acrylic acid) (P(NIPAm-co-AAc)) and poly(N-isopropylacrylamide-co-acrylamide) (P(NIPAm-co-Am)) were synthesized by free radical copolymerization in our laboratory according to the methods as described in our previous studies [24,25]. The chemical structure, low critical solution temperature (LCST) and weight-average molecular weights (Mw ) of the thermo-responsive polymers used in this study are shown in Table 1. Bovine serum albumin (BSA) was purchased from Sigma–Aldrich and used as the organic foulant in this study. De-ionized water with a conductivity of less than 3 s/cm was used as the solvent for preparing aqueous solution as well as for soaking and rinsing the membranes. All other chemicals were analytic purity grade and used as received.

Pressure regulating valve FI Flow meter FI Three-way valve

Test cell

Pressure gauge PI

PI FI Concentrate Feed FI Permeate

By pass

Heat exchanger Feed tank

Pump

Fig. 2. Schematic diagram of the ?at-sheet cross-?ow permeation test unit.

184

S. Yu et al. / Journal of Membrane Science 392–393 (2012) 181–191

Table 1 Chemical structure, LCST and weight-averaged molecular weight of the thermo-responsive polymers used in this study. Thermo-responsive polymers Chemical structure LCSTa (? C) Weight-averaged molecular weightb (Mw )

( CH2
P(NIPAm)c

CH C NH CH

)m
O
32.5 4.5 × 104

H3C

CH3 CH C

( CH2
P(NIPAm-co-AAc)d

)m ( CH2
O

CH ) C OH

n

O
35.0 4.8 × 104

NH CH H3C CH3 CH C

( CH2
P(NIPAm-co-Am)e

)m ( CH2
O

CH ) C NH2

n

O
38.5 5.0 × 104

NH CH H3C CH3

Low critical solution temperature was determined using 0.5 wt.% aqueous polymer solution according to the method described in Ref. [24]. Weight-averaged molecular weight was measured by GPC with water as the mobile phase at a ?ow rate of 0.5 ml/min. Prepared by free radical copolymerization of monomer N-isopropylacrylamide (NIPAm) in aqueous solution at 40 ? C using potassium persulfate and sodium sul?te as the initiators. d Prepared by free radical copolymerization of monomers N-isopropylacrylamide (NIPAm) and acrylamide (Am) (weight ratio of NIPAm/Am = 90/10) in aqueous solution at 40 ? C using potassium persulfate and sodium sul?te as the initiators. e Prepared by free radical copolymerization of monomers N-isopropylacrylamide (NIPAm) and acrylic acid (AAc) (mole ratio of NIPAm/AAc = 98/2) in 1,4-dioxane at 70 ? Cusing AIBN as the initiator.
b c

a

Jw =

Q A t

(4)

where Jw is the volumetric permeate water ?ux, A is the effective area of the membrane for permeation, and Q is the volume of permeation over a time interval t. The salt rejection was evaluated using the following equation: R (%) = 100 × 1 ? Cp Cf (5)

in which Cp and Cf are the salt concentrations in permeate and feed, respectively. The salt concentration was determined by measuring the electrical conductivity of the salt solution using a conductance meter (DDSJ-308A, Cany Precision Instruments Co. Ltd., China) and comparing the calibration plot drawn between salt concentration and conductivity. Reported ?uxes and salt rejections are averaged values from four samples for each membrane type. 3.4. Fouling experiments Fouling experiments were performed with organic foulant bovine serum albumin (BSA) employing the laboratory-scale cross?ow test unit described above. Fresh sheet of the ?at polyamide reverse osmosis membrane was cut into rectangular sections of approximately 18.0 cm × 7.0 cm. All membrane samples were cut from a single sheet of membrane for all the ?ltration experiments. All the membranes were immersed in de-ionized water for 24 h before the fouling experiments. Prior to the fouling tests, the

reverse osmosis performance of the fresh membrane was determined as described in Section 2.3. After that, a stock BSA solution, which was freshly prepared to the required concentration prior to each fouling experiment by dissolving powdered BSA in de-ionized water followed by vigorous stirring of solution, was added into the feed tank to achieve the desired concentration of about 100 mg/l. Then, cross-?ow membrane ?ltrations of the BSA aqueous solution were carried out under pH 6.8, 25.0 ? C and 0.6 MPa until the permeate ?ux of the tested membrane reached a steady state. Periodical measurements were carried out to check the permeate ?ux and the BSA concentration of the feed solution, which was used to estimate the amount of the BSA deposited on the membrane surface. The fouled membranes were remained in the test cells for the following cleaning experiments. 3.5. Cleaning experiments In order to systematically investigate the effect of thermoresponsive polymer on the cleaning ef?ciency of organic fouled reverse osmosis membrane, cleaning experiments were conducted in situ with the fouled membranes using de-ionized water with and without thermal-responsive polymer, respectively. Four steps were employed in each cleaning experiment: (1) measuring the water ?ux and salt rejection value for the fouled membrane before cleaning according to the method described in Section 2.3 using 500 mg/l NaCl aqueous solution; (2) alternate circulation and soaking with de-ionized water or de-ionized water containing certain amount of thermal-responsive polymer for a certain period

S. Yu et al. / Journal of Membrane Science 392–393 (2012) 181–191

185

4

Normalized flux J/J0

0.9

3
0.8

0.7

2

0.6

1
Normalized flux Deposited BSA

3.6. Membrane characterization All the membrane samples were dried under vacuum at for 24 h before characterization. Attenuated total re?ectance Fourier transform infrared (ATRFTIR) spectroscopy was employed to analysis the surface composition of the fresh, fouled and cleaned thin-?lm composite polyamide reverse osmosis membranes, respectively. The ATR-FTIR spectra were recorded on a Nicolet Aratar 370 FTIR spectrometer with a ZnSe crystal as the internal refection element with an angle of incidence of 45? . The surface morphological structure of the fresh, fouled and cleaned reverse osmosis membranes were also investigated by using a ?eld emission scanning electron microscopy (FE-SEM) (Hitachi S-4800, Japan). Magni?cation up to 30,000 was obtained. Contact angle measurements were performed with a DSA10MK2 contact angle analyzer (KRUSS BmbH Co, Germany). The sessile drop method was used to measure the contact angles of de-ionized water (about 3 l) on the dried surfaces of the membranes at 25.0 ? C. Images were captured 5 s after introducing the drop and the contact angles were calculated. At least ten measurements on different locations of the membrane sample were performed and averaged to obtain the contact angle of the measured membrane sample. All the results presented were an average data from four membrane samples with standard deviation of the measured values. 4. Results and discussion 4.1. Membrane fouling using BSA aqueous solution Fig. 3 presents the changes of the normalized ?ux and the estimated amount of BSA deposited on the membrane surface with the running time in ?ltration 100 mg/l BSA aqueous solution under pH 6.8 and 0.6 MPa. Experimental results reveal that the permeate ?ux declines continuously with the ?ltration time due to the accumulation of BSA on the membrane surface, which was resulted from the adsorption of BSA on the membrane surface through the interaction between BSA molecule and membrane surface followed by the deposition of BSA on the membrane surface through the interaction between the adsorbed BSA and BSA in bulk solution [26]. The favorable interaction between the BSA in bulk solution and BSA adsorbed on the membrane surface leads to the formation of a signi?cant BSA fouling layer on the membrane and thereby a signi?cant ?ux decline as the ?ltration time prolongs. 4.2. Diffusion and phase transition of thermo-responsive polymer
4000 3500 3000 2500 2000
-1

30.0 ? C

0.5

0

5

10

15

20

25

30

0

Filtration time (h)
Fig. 3. Changes of the normalized ?ux ( ) and the estimated amount of BSA deposited on the membrane surface ( ) with ?ltration time. J0 is the pure water ?ux of the fresh membrane tested at 0.6 MPa.

attached molecules on the membrane surface, and then characterized by ATR-FTIR and surface contact angle measurement to investigate the diffusion and phase transition of the thermoresponsive polymer, respectively. Fig. 4 gives the ATR-FTIR spectra of the BSA-fouled and TRPsoaked membranes. The enhancements of the peaks between 3100–3400, 2800–2900 and at 1540 cm?1 , which are assigned to the NH stretching, CH stretching and N H in plane bending, respectively, with the increasing P(NIPAm) content in the soaking solution suggest that P(NIPAm) has diffused into the BSA layer deposited on the membrane surface after soaking and the amount of the diffused P(NIPAm) increases with the increasing P(NIPAm) content in the soaking solution. Additionally, the shift of the peak of amide I from about 1658 cm?1 to a lower frequency of around 1646 cm?1 , which is due to the formation of hydrogen bonds between the C O groups with more N H groups [27], also con?rms the existence of P(NIPAm) in the BSA layer after soaking. Surface contact angle measurements were conducted under different environmental temperatures between 20 ? C and 50 ? C. The

a. 0.0 mg/l P(NIPAm) b. 40.0 mg/l P(NIPAm) c. 60.0 mg/l P(NIPAm) d. 100.0 mg/l P(NIPAm)

Absorbance

d c b a
1500 1000

Soaking tests were carried out by espousing the BSA-fouled membrane samples that remained in the test cells with de-ionized water containing different contents of P(NIPAm) for 1.0 h at room temperature. The membrane samples were taken out of the test cells and rinsed with de-ionized water to remove any weakly

Wavenumber cm

Fig. 4. ATR-FTIR spectra of the BSA-fouled membrane soaked with de-ionized water containing different contents of P(NIPAm): (a) 0.0 mg/l, (b) 40.0 mg/l, (c) 60.0 mg/l, (d) 100.0 mg/l.

BSA deposited on membrane surface g/m

at temperature lower than the LCST of the thermal-responsive polymer and 0.17 MPa feed pressure; (3) alternate circulation and soaking with warm de-ionized water of temperature higher than the LCST of the thermal-responsive polymer for a certain time at 0.17 MPa feed pressure; (4) rinsing with de-ionized water of room temperature (25.0 ± 1.0 ? C) and repeating step (1) to get the post cleaning reverse osmosis performance. Cleaning experiments were conducted by varying the type of thermo-responsive polymer, the concentration of TRP solution, the soaking time of TRP solution (step (2)), as well as the cross-?ow velocity of the circulation of warm water (step (3)).

5
1.0
2

186

S. Yu et al. / Journal of Membrane Science 392–393 (2012) 181–191

90

Sureface contact angle( °)

80

Fresh membrane TRP-soak ed membrane BSA-fouled membrane

70

60

50

20

25

30

35

40

45

50

Temperature(?C)
Fig. 5. Changes of contact angle with environmental temperature for the fresh ( ), BSA-fouled ( ) and TRP-soaked ( ) membranes.

membrane samples were stabilized for 30 min before measurement at each test temperature. The changes of the measured contact angles with environmental temperature for the fresh membrane, the BSA-fouled membrane, as well as the BSA-fouled membrane after soaking with 100 mg/l P(NIPAm) aqueous solution for 1.0 h are demonstrated in Fig. 5. The distinctive change behavior of the surface contact angle of the TRP-soaked membrane with environmental temperature con?rms the existence and occurrence of phase transition of P(NIPAm) in the BSA layer deposited on the membrane surface. 4.3. Membrane cleaning using different thermal-responsive polymers Membrane cleaning experiments were ?rstly investigated by using different thermal-responsive polymers under the following cleaning conditions: alternate circulation and soaking with aqueous solution containing 50 mg/l thermal-responsive polymer for 1 h at temperature of 25 ? C and feed pressure of 0.17 MPa; alternate circulation and soaking with warm de-ionized water of 45.0 ? C for 1 h at 0.17 MPa feed pressure and 0.5 m/s cross-?ow velocity; rinsing with de-ionized water at 25 ? C for 0.5 h. The results of the water ?uxes and salt rejections of the fouled and cleaned membranes are presented in Table 2.

It can be seen that, in all the cases using thermalresponsive polymer, the water ?ux and salt rejection of the cleaned membrane are higher than those of the membrane cleaned only using de-ionized water. The salt rejection of the cleaned membrane increases following the order of de-ionized water-cleaned RO membrane (96.5%) < P(NIPAm-co-Am) solutioncleaned RO membrane (96.9%) < P(NIPAm) solution-cleaned RO membrane (97.2%) < P(NIPAm-co-AAc) solution-cleaned RO membrane (98.3%), while the water ?ux is increased by 6.6, 13.2, 15.3, and 22.4% for membranes cleaned with de-ionized water, P(NIPAm-co-Am) solution, P(NIPAm) solution, and P(NIPAm-coAAc) solution, respectively. The enhanced cleaning ef?ciency may be explained in terms of the phase transition of the thermal-responsive polymer that occurs during the cleaning process. During the cleaning step of alternate circulation and soaking with aqueous solution containing thermal-responsive polymer at temperature of 25 ? C, all the thermo-responsive polymers (TRPs) used in this study are soluble in de-ionized water and exist in extended random coil conformation, and will diffuse into the fouling layer on the surface of the fouled membrane. When the treated membrane is soaked with warm water of 45.0 ? C, the thermal-responsive polymers that has diffused into the fouling layer will be insoluble and change into fold conformation. The phase transition of the thermal-responsive polymers will weaken the structural integrity of the BSA fouling layer on the membrane surface, and thereby promote the removal of BSA molecules from the membrane surface. Considering the cleaning ef?ciency, P(NIPAm-co-AAc) was selected as thermal-responsive polymer in the following study. 4.4. Membrane cleaning under different concentrations of TRP The effect of thermal-responsive polymer concentration on the cleaning ef?ciency was studied by using aqueous solution containing different contents of P(NIPAm-co-AAc) under the following cleaning conditions: alternate circulation and soaking the fouled membrane with P(NIPAm-co-AAc) aqueous solution for 1 h at temperature of 25 ? C and feed pressure of 0.17 MPa; alternate circulation and soaking with warm de-ionized water of 45.0 ? C for 1 h at 0.17 MPa feed pressure and 0.5 m/s cross-?ow velocity; rinsing with de-ionized water at 25 ? C for 0.5 h. The results of the water ?uxes and salt rejections of the cleaned membranes are shown in Fig. 6, which also presents the resistance to water permeation of the cleaned membrane. It can be seen clearly from the ?gure that the concentration of the thermal-responsive polymer has a signi?cant effect on the cleaning ef?ciency. With the P(NIPAm-co-AAc) concentration increasing from 0 to 80 mg/l, the salt rejection and water ?ux of the cleaned membrane increase from 96.4 to 98.5% and 42.0 to 49.5 l/m2 h, respectively, while the resistance to water permeation declines from 9.121 × 1013 to 7.739 × 1013 m?1 , and then all almost level off. The behaviors of the observed water ?ux and salt rejection as well as the calculated resistance to water permeation of the cleaned membrane with P(NIPAm-co-AAc) concentration may be explained in terms of the mass transfer of the thermal-responsive polymer from the bulk soaking solution to fouling layer that occurs during the second step of soaking and circulation. When the fouled membrane was soaked with aqueous solution containing thermalresponsive polymer under temperature lower than the LCST, the polymer will diffuse into the fouling layer located on the membrane surface through molecular motion under the chemical potential gradient of the thermal-responsive polymer between the bulk soaking solution and the fouling layer, and the diffusion transport of TRP from bulk solution to the fouling layer will increase as the concentration of TRP in bulk solution increases [28]. Thus, larger

Table 2 Water ?uxes and salt rejections of the fresh, fouled and cleaned reverse osmosis membranes. Membrane Fresh RO membrane BSA-fouled RO membrane De-ionized water-cleaned RO membrane P(NIPAm) solution-cleaned RO membrane P(NIPAm-co-AAc) solution-cleaned RO membrane P(NIPAm-co-Am) solution-cleaned RO membrane Water ?uxa (l/m2 h) 60.5 ± 0.6 38.0 ± 0.5 40.5 ± 0.6 43.8 ± 0.8 46.5 ± 0.7 Salt rejectiona (%) 98.5 ± 0.2 95.4 ± 0.3 96.5 ± 0.2 97.2 ± 0.3 98.3 ± 0.2

43.0 ± 0.5

96.9 ± 0.2

a Test conditions: feed = 500 mg/l NaCl aqueous solution; pressure = 1.0 MPa; temperature = 25.0 ? C and pH = 6.8.

S. Yu et al. / Journal of Membrane Science 392–393 (2012) 181–191

187
101

(a)
54 52 50

100

54 52

Water flux Salt rejection
100 99 98 97

98
50

Water flux l/m h

2 Water flux l/m h

Salt rejection (%)

96 48 46 44 42 40 0 20 40 60 80 100 120 94

2

48 46 44 42 40 5

96 95 5 0 30 60 90 120 150 180

Water flux Salt rejection

92

90

0

Soaking time (min)
Fig. 7. Effects of soaking time of P(NIPAm-co-AAc) aqueous solution on the salt rejection ( ) and water ?ux ( ) of the cleaned membrane tested with 500 mg/l NaCl aqueous solution at 1.0 MPa, 25 ? C and pH 6.8. The water ?ux and salt rejection of fouled membrane are 38.5 l/m2 h and 95.2%, respectively.

P(NIPAm-2AAc) concentration (mg/l)

(b) 10
m -1 13
9

10

Resistance to water permeation

8

7

6 1 0 0 20 40 60 80 100 120

P(NIPAm-2AAc) concentration (mg/l)
Fig. 6. Effects of P(NIPAm-co-AAc) concentration on the performance (a) in terms of salt rejection ( ) and water ?ux ( ), and resistance to water permeation (b) of the cleaned membrane tested with 500 mg/l NaCl aqueous solution at 1.0 MPa, 25 ? C and pH 6.8. The water ?ux and salt rejection of fouled membrane are 39.5 l/m2 h and 95.6%, respectively.

amount of P(NIPAm-co-AAc) molecules will diffuse into the BSA fouling layer on the membrane surface at higher concentration of thermo-responsive polymer. As a result, when these diffused TRP molecules were made insoluble, the fouling layer will be loosed to a great extent and can be removed from the membrane surface more easily, and the cleaned membrane will exhibit relative higher salt rejection and water ?ux and lower resistance to water permeation. Further increase in the concentration, however, tends to have almost no more effect on the diffusion of the thermo-responsive polymer into the fouling layer and thereby has less effect on the separation performance and permeation resistance of the cleaned membrane. 4.5. Membrane cleaning under different soaking time of TRP solution The effect of the soaking time of thermal-responsive polymer aqueous solution (second step) on the cleaning ef?ciency was further studied under the following cleaning conditions: alternate circulation and soaking the fouled membrane with 80 mg/l P(NIPAm-co-AAc) aqueous solution for a period from 30 to 180 min at temperature of 25 ? C and feed pressure of 0.17 MPa; alternate circulation and soaking with warm de-ionized water of 45.0 ? C for 1 h at 0.17 MPa feed pressure and 0.5 m/s cross-?ow velocity; rinsing with de-ionized water at 25 ? C for 0.5 h. The results of the water

?uxes and salt rejections of the cleaned membranes are shown in Fig. 7. As shown in the ?gure, with the soaking time increasing from 30 to 120 min, the salt rejection of the cleaned membrane increases appreciably from 97.8 to 98.6% and the water ?ux ascends obviously from about 45.5 to 50.5 l/m2 h, and then both change slightly as the soaking time is further prolonged. The mass transfer of the thermo-responsive polymer from the bulk soaking solution to the fouling layer is also affected by the soaking time of thermal-responsive polymer aqueous solution under temperature lower than the LCST. As the soaking time extends from 30 to 120 min, the amount of thermal-responsive polymer diffused into the foulant layer on the surface increases, as discussed in the previous section, the increased amount of TRP will result in an augment in cleaning ef?ciency. However, as the amount of thermal-responsive polymer in the fouling layer increases to a certain value, further diffusion of the thermal-responsive polymer from bulk solution to the fouling layer will be retarded and the amount of TRP in the foulant layer will remain constant for further increase in soaking time [28]. This may be the reason for the slight changes in both the water ?ux and salt rejection for the cleaned membranes soaked with TRP solution for more than 120 min. 4.6. Membrane cleaning under different cross-?ow velocities of warm water circulation The impact of thermo-responsive polymer on the cleaning of fouled reverse osmosis membrane was also studied under different cross-?ow velocities of warm water circulation (third step). Two series of cleaning experiments were conducted using de-ionized water with and without thermo-responsive polymer P(NIPAm-coAAc), respectively. For the cleaning experiments with TRP solution, the concentration and soaking time of the TRP aqueous solution of the ?rst cleaning step were ?xed at 80 mg/l and 2.0 h, respectively. The cross-?ow velocity of the warm water circulation step was varied from 0.2 to 1.0 m/s. The water ?uxes and salt rejections of the membranes cleaned using de-ionized water with and without TRP under different cross-?ow velocities are presented in Fig. 8. It can be seen from the ?gures that both the water ?uxes and salt rejections of the cleaned membranes increase with the increasing cross-?ow velocity. However, the increasing rates of salt rejection and water ?ux of the membrane cleaned using TRP aqueous solution are higher than those of the membrane cleaned only using de-ionized water. As the cross-?ow velocity increases from 0.2 to

Salt rejection (%)

188

S. Yu et al. / Journal of Membrane Science 392–393 (2012) 181–191

(a)

100

Membrane cleaned with TRP
99

Membrame cleaned without TRP

a. Fresh b. BSA-fouled c. De-ionized water-cleaned d. TRP solution-cleaned

Salt rejection (%)

98

Absorbance

d c
1730 1645

97

b
96
1660

1610

a
95 0.2 0.4 0.6 0.8 1.0 2000 1800 1600 1400 1200
-1

1000

800

Cross-flow velocity (m/s)

Wavenumbers cm

(b)

60

Membrane cleaned with TRP Membrame cleaned without TRP
56

Fig. 9. ATR-FTIR spectra of the fresh (a), BSA-fouled (b), de-ionized water-cleaned (c) and P(NIPAm-co-AAc) solution-cleaned (d) membranes.

Water flux l/m h

2

water ?ux and salt rejection of the cleaned membrane will change distinctly with the increasing cross-?ow velocity.
52

4.7. Membrane resistance to water permeation
48

44

40 0.2 0.4 0.6 0.8 1.0

Cross-flow velocity (m/s)
Fig. 8. Effects of the cross-?ow velocity of warm water circulation on the salt rejection (a) and water ?ux (b) of the membranes cleaned with ( ) and without P(NIPAm-co-AAc) ( ). Testing conditions employed were: 500 mg/l NaCl aqueous solution at 1.0 MPa, 25 ? C and pH 6.8. The water ?ux and salt rejection of fouled membrane are 40.0 l/m2 h and 95.8%, respectively.

1.0 m/s, the salt rejection and water ?ux of the membrane cleaned using TRP solution increase obviously from 97.6 to 98.7% and 50.0 to 54.0 l/m2 h, respectively. While the salt rejection and water ?ux of the membrane cleaned only using de-ionized water ascend slightly from 96.4 to 96.6% and 42.0 to 43.0 l/m2 h, respectively. For the fouled membrane cleaned using de-ionized water, the BSA fouling layer on the membrane surface cannot be loosed during the step of soaking and circulation with warm water, and the structurally integrity of the fouling layer is still very much intact. The increase in cross-?ow velocity which results in an augment in the shear rate will not enhance the mass transfer of the foulants in the fouling layer to the bulk solution [19]. Thus, the cross-?ow velocity of the circulation of the warm de-ionized water has little effect on cleaning ef?ciency, and the water ?ux and salt rejection of the cleaned membrane will change slightly with the increasing cross-?ow velocity. While for the fouled membrane cleaned using de-ionized water containing TRP, the BSA fouling layer on the membrane surface will be loosed due to the phase transition of the TRP diffused into the foulant layer, and the augment in the shear rate resulting from the increased cross-?ow velocity will promote the mass transfer of the foulants in the fouling layer to the bulk solution [19]. As a result, the

The pure water ?uxes of the fresh, BSA-fouled, TRP-soaked, as well as cleaned membranes were measured under 0.5 MPa and temperatures of 25.0 and 45.0 ? C, respectively. The resistance to water permeation was calculated from Eq. (3) by ignoring the osmotic pressure difference and the results are shown in Table 3. It can be seen that the membrane hydraulic resistance (Rm ), is nearly constant and independent of temperature, and the fouling layer of BSA on the membrane surface results in an additional hydraulic resistance (Rf ) of about 3.181 × 1013 m?1 at 25.0 ? C, which ascends sharply to 5.265 × 1013 m?1 at 45.0 ? C due to the densi?cation of the BSA fouling layer under high temperature [29]. The diffusion of P(NIPAm-co-AAc) into the BSA fouling layer leads to a slight increment of 0.149 × 1013 m?1 and a sharp decline of 1.2 × 1013 m?1 in resistance to water permeation at 25.0 and 45 ? C, respectively, suggesting that the phase transition of TRP at higher temperature looses the fouling layer and thereby decreases the permeate resistance. Additionally, the relatively higher value of the resistance of the cleaned membrane compared to the fresh membrane reveals that part of the deposited BSA still remains on the membrane surface after cleaning. 4.8. Membrane surface characterization In order to further study the impact of thermo-responsive polymer on the cleaning of fouled reverse osmosis membrane, membrane surface characterization including ATR-FTIR and SEM analysis, and contact angle measurements were also performed with the fresh, fouled and cleaned membranes, respectively. 4.8.1. ATR-FTIR ATR-FTIR was employed to investigate the change of surface chemical structure of the composite membranes before and after cleaning. The ATR-FTIR spectra of the fresh, BSA-fouled, de-ionized water-cleaned, as well as the TRP solution-cleaned membranes are shown in Fig. 9. As shown in the ?gure, the peaks at 1660 cm?1 and 1610 cm?1 , which are the C O stretching and hydrogen-bonded carbonyl of the amide of the fresh aromatic polyamide reverse osmosis membrane

S. Yu et al. / Journal of Membrane Science 392–393 (2012) 181–191 Table 3 Water ?uxes and resistances to water permeation of the fresh, BSA-fouled, TRP-soaked and cleaned membranes. Membrane sample Water ?ux (l/m2 h)a At 25 C Fresh RO membrane BSA-fouled membrane P(NIPAm-co-AAc)-soaked membrane Membrane cleaned with de-ionized water Membrane cleaned with TRP solution
a ?

189

Resistance to water permeation (m?1 ) At 45 C 44.0 24.2 27.0 29.1 39.6
?

At 25 ? C 6.493 × 1013 9.674 × 1013 9.823 × 1013 8.706 × 1013 7.424 × 1013

At 45 ? C 6.455 × 1013 1.172 × 1012 1.052 × 1012 9.774 × 1013 7.445 × 1013

29.5 19.8 19.5 22.0 26.8

Tested with de-ionized water under 0.5 MPa.

(spectra a) [30], disappear in the spectra of the fouled membrane (spectra b), meanwhile a new peak of strong intensity at about 1645 cm?1 presents in the spectra. This new peak at 1645 cm?1 , which originates predominately from the C O stretching vibration of the peptide groups, is the characteristic peak of BSA and is generally inspected as proof of the presence of BSA on membrane surface [31]. Additionally, a new peak of weak intensity near 1730 cm?1 , which belongs to residuals of different fatty acid in BSA [32], also appears in the spectra of the fouled membranes. These results indicate that the surface of the fresh polyamide reverse osmosis membrane has been fouled and a deposition layer of BSA has formed on the membrane surface. After cleaning with de-ionized water, the featured peaks of BSA at about 1645 cm?1 and 1730 cm?1 can still be observed in the spectra (spectra c), indicating that some BSA molecules still remain on the surface of the de-ionized water-cleaned membrane. While for the membrane cleaned with TRP solution, the featured peaks of the BSA foulant disappear in the spectra (spectra d), meanwhile the characteristic peaks of the fresh polyamide membrane at about 1610 cm?1 and 1660 cm?1 reappear, indicating that the amount of BSA remained on the surface of the TRP-cleaned membrane is less than that on the de-ionized water-cleaned membrane.

4.8.2. SEM The change of membrane surface morphological structure before and after cleaning was investigated by using SEM. Representative SEM images of the surfaces of the fresh, BSA-fouled, de-ionized water-cleaned and TRP solution-cleaned membranes are provided in Fig. 10. It can be seen clearly that the surface feature of the fresh thin?lm composite polyamide reverse osmosis membrane appears to be relatively loose and exhibits a typical nodular structure (Fig. 10a). After ?ltration with BSA solution, the surface feature of the used polyamide thin-?lm composite membrane becomes denser and compacter (Fig. 10b), indicating that the membrane has been seriously fouled due to the deposition of BSA molecules on the membrane surface. After cleaning with de-ionized water (Fig. 10c), the surface of the fouled membrane becomes less compact than the fouled membrane, but it is still denser than the fresh membrane. This means that the deposited BSA molecules on the membrane surface have been partially removed and some BSA molecules still remain on the membrane surface. While for the membrane cleaned with TRP solution (Fig. 10d), the surface morphology feature becomes close to the fresh membrane after cleaning, indicating that the removal

Fig. 10. SEM images of the surfaces of the fresh (a), BSA-fouled (b), de-ionized water-cleaned (c) and P(NIPAm-co-AAc) solution-cleaned (d) membranes.

190

S. Yu et al. / Journal of Membrane Science 392–393 (2012) 181–191

100

80

60

40

As the concentration of P(NIPAm-co-AAc) increased from 10 to 80 mg/l, the salt rejection of the cleaned membrane was increased from 97.2 to 98.5%, which corresponds to an increase of water ?ux by 11.2%. The increase of soaking time of P(NIPAm-co-AAc) solution from 30 to 120 min resulted in an augment of 11.0% in water ?ux and an increase of salt rejection from 97.8 to 98.6% for the cleaned membrane. Moreover, with increasing cross-?ow velocity of the circulation of warm water from 0.2 to 1.0 m/s, the salt rejection and water ?ux of the membrane cleaned using P(NIPAm-co-AAc) solution ascended from 97.6 to 98.7% and 50.0 to 54.0 l/m2 h, respectively. Acknowledgments

Contact angle o

20

0 Fresh BSA-fouled TRP solutioncleaned De-ionized water-cleaned

Fig. 11. Surface contact angles of the fresh, BSA-fouled, de-ionized water-cleaned and P(NIPAm-co-AAc) solution-cleaned membranes.

of BSA molecules from the membrane surface is facilitated by using thermo-responsive polymer. 4.8.3. Surface contact angle measurements The surface contact angles of the fresh, BSA-fouled, de-ionized water-cleaned and TRP solution-cleaned membranes are provided in Fig. 11. It can be seen from the ?gure that the surface contact angle increases from about 58.0? for the fresh membrane to around 92.0? for the BSA-fouled membrane, which means that the membrane surface is completely covered with BSA after fouling. After cleaning with TRP solution, the surface contact angle decreases dramatically to a very low value of about 60.5? , while for the membrane cleaned only using de-ionized water, the surface contact angle only drops to 74.5? . The decrease of surface contact angle by a wide margin for the membrane cleaned with TRP solution also strongly suggests that the cleaning of fouled reverse osmosis membrane can be effectively intensi?ed by using thermo-responsive polymer. 5. Conclusions Cleaning experiments of BSA-fouled reverse osmosis membranes using de-ionized water with and without thermoresponsive polymer of low critical solution temperature have been conducted in this study. The bene?cial effect of TRP on the removal of the deposited foulant from the fouled membrane surface was demonstrated through testing separation performance in terms of salt rejection and water ?ux, and surface characterization via scanning electron microscopy (SEM), attenuated total re?ectance infrared (ATR-FTIR) and surface contact angle measurements. It was found that the diffusion of TRP into the BSA fouling layer deposited on the membrane surface leads to a slight increase in total resistance to water permeation of the BSA-fouled membrane at low temperature of 25.0 ? C, and a sharp decline in total resistance to water permeation of the BSA-fouled membrane at higher temperature of 45.0 ? C due to the loosing effect resulting from the phase transition of the TRP. The cleaning ef?ciency of BSA-fouled reverse osmosis membrane could be enhanced by the insolubilization of the TRP diffused into the fouling layer, and was largely affected by the type and concentration of TRP, the soaking time of TRP solution, and the cross-?ow velocity of the circulation of warm water. The cleaning ef?ciency in terms of water ?ux and salt rejection increased following the order of de-ionized water < P(NIPAmco-Am) solution < P(NIPAm) solution < P(NIPAm-co-AAc) solution.

The authors gratefully acknowledge the ?nancial support of the National Nature Science Foundation of China (NNSFC) (Grant No. 20976167), Nature Science Foundation of Zhejiang Province (Grant No. Y4080355), Zhejiang Provincial Key Innovation Team (No. 2010R50038) and Foundation of Zhejiang Top Academic Discipline of Applied Chemistry and Eco-Dyeing & Finishing (Grant No. YR2011010). References
[1] L.F. Greenlee, D.F. Lawler, B.D. Freeman, B. Marrot, P. Moulin, Reverse osmosis desalination: water sources, technology, and today’s challenges, Water Res 43 (2009) 2317–2348. [2] C. Fritzmann, J. L?wenberg, T. Wintgens, T. Melin, State-of-the-art of reverse osmosis desalination, Desalination 216 (2007) 1–76. [3] R.W. Baker, Membrane Technology Applications, 2nd ed., John Wiley & Sons, Ltd., Chichester, 2004. ? [4] M.A. Shannon, P.W. Bohn, M. Elimelech, J.G. Georgiadis, B.J. Marinas, A.M. Mayes, Science and technology for water puri?cation in the coming decades, Nature 452 (2008) 301–310. [5] A. Al-Amoudi, R.W. Lovitt, Fouling strategies and the cleaning system of NF membranes and factors affecting cleaning ef?ciency, J. Membr. Sci. 303 (2007) 4–28. [6] M. Mulder, Basic Principles of Membrane Technology, 2nd ed., Kluwer Academic Publishers, Dordrecht, The Netherlands, 1996. [7] Z. Amjad, Reverse Osmosis, Van Nostrand Reihold, New York, 1993. [8] K.O. Agenson, T. Urase, Change in membrane performance due to organic fouling in nano?ltration (NF)/reverse osmosis (RO) applications, Sep. Purif. Technol. 55 (2007) 147–156. [9] J.-J. Qin, M.H. Oo, H. Lee, R. Kolkman, Dead-end ultra?ltration for pretreatment of RO in reclamation of municipal wastewater ef?uent, J. Membr. Sci. 243 (2004) 107–113. [10] A. Maartens, P. Swart, E.P. Jacobs, Feed-water pretreatment: methods to reduce membrane fouling by natural organic matter, J. Membr. Sci. 163 (1) (1999) 51–62. [11] A.J.B. van Boxtel, Z.E.H. Otten, H.J.L.J. van der Linden, Evaluation of process models for fouling control of reverse osmosis of cheese whey, J. Membr. Sci. 58 (1991) 89–111. [12] T.N. Shah, Y. Yoon, C.L. Pederson, R.M. Lueptow, Rotating reverse osmosis and spiral wound reverse osmosis ?ltration: a comparison, J. Membr. Sci. 285 (2006) 353–361. [13] S.H. Kim, S-Y. Kwak, B-H. Sohn, T.H. Park, Design of TiO2 nanoparticle selfassembled aromatic polyamide thin-?lm-composite (TFC) membrane as an approach to solve biofouling problem, J. Membr. Sci. 211 (2003) 157–165. [14] I.C. Kim, Y.H. Ka, J.Y. Park, K.H. Lee, Preparation of fouling resistant nano?ltration and reverse osmosis membranes and their use for dyeing wastewater ef?uent, J. Ind. Eng. Chem. 10 (2004) 115–121. [15] 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. [16] S. Belfer, Y. Purinson, R. Fainshtein, Y. Radchenko, O. Kedem, Surface modi?cation of commercial composite polyamide reverse osmosis membranes, J. Membr. Sci. 139 (1998) 175–181. [17] S.S. Madaeni, T. Mohamamdi, M.K. Moghadam, Chemical cleaning of reverse osmosis membranes, Desalination 134 (2001) 77–82. [18] D. Feng, J.S.J. van Deventer, C. Aldrich, Ultrasonic defouling of reverse osmosis membranes used to treat wastewater ef?uents, Sep. Purif. Technol. 50 (2006) 318–323. [19] W.S. Ang, S. Lee, M. Elimelech, Chemical physical aspects of cleaning of organicfouled reverse osmosis membranes, J. Membr. Sci. 272 (2006) 198–210. [20] X.M. Liu, L.I. Wang, L. Wang, J. Haung, C. He, The effect of salt and pH on the phase transition behaviors of temperature-sensitive copolymers based on Nisopropylacrylamide, Biomaterials 25 (2004) 5659–5666. [21] J.G. Wijmans, R.W. Baker, The solution-diffusion model: a review, J. Membr. Sci. 107 (1995) 1–21.

S. Yu et al. / Journal of Membrane Science 392–393 (2012) 181–191 [22] A.C. Sagle, E.M. VanWagner, H. Ju, B.D. McCloskey, B.D. Freeman, M.M. Sharma, PEG-coated reverse osmosis membranes: desalination properties and fouling resistance, J. Membr. Sci. 340 (2009) 92–108. [23] S. Mattaraj, W. Phimpha, P. Hongthong, R. Jiraratananon, Effect of operating conditions and solution chemistry on model parameters in cross?ow reverse osmosis of natural organic matter, Desalination 253 (2010) 38–45. [24] D. Wu, X. Liu, S. Yu, M. Liu, C. Gao, Modi?cation of aromatic polyamide thin-?lm composite reverse osmosis membranes by surface coating of thermoresponsive copolymers P(NIPAM-co-Am). I: preparation and characterization, J. Membr. Sci. 352 (2010) 76–85. [25] S. Yu, Z. Lü, Z. Chen, X. Liu, M. Liu, C. Gao, Surface modi?cation of thin-?lm composite polyamide reverse osmosis membranes by coating N-isopropylacrylamide-co-acrylic acid copolymers for improved membrane properties, J. Membr. Sci. 371 (2011) 293–306. [26] W.S. Ang, M. Elimelech, Protein (BSA) fouling of reverse osmosis membranes: Implications for wastewater reclamation, J. Membr. Sci. 296 (2007) 83–92.

191

[27] G. Socrates, Infrared Characteristic Group Frequencies, Wiley Interscience, Chichester, 1994. [28] J. Crank, The Mathematics of Diffusion, 2nd ed., Oxford University Press, 1975. [29] H. Mo, K.G. Tay, H.Y. Ng, Fouling of reverse osmosis membrane by protein (BSA): effects of pH, calcium, magnesium, ionic strength and temperature, J. Membr. Sci. 315 (2008) 28–35. [30] S. Belfer, Y. Purinson, O. Kedem, Surface modi?cation of commercial polyamide reverse osmosis membranes by radical grafting: an ATR-FTIR study, Acta Polym. 49 (1998) 574–582. [31] S. Belfer, R. Fainchtain, Y. Purinson, O. Kedem, Surface characterization by FTIRATR spectroscopy of polyethersulfone membranes-unmodi?ed, modi?ed and protein fouled, J. Membr. Sci. 172 (2000) 113–124. [32] D. Kuzmenko, E. Arkhangelsky, S. Belfer, V. Freger, V. Gitis, Chemical cleaning of UF membranes fouled by BSA, Desalination 179 (2005) 323–333.


相关文章:
更多相关标签: