Chemical Engineering Journal 243 (2014) 137–146

Contents lists available at ScienceDirect

Chemical Engineering Journal

journal homepage: www.elsevier .com/locate /cej

Synthesis, characterization and surface modification of PES hollow fibermembrane support with polydopamine and thin film composite forenergy generation

1385-8947/$ - see front matter � 2014 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.cej.2013.12.094

⇑ Corresponding author. Tel.: +82 42 860 3647.E-mail address: hklee@kier.re.kr (H.K. Lee).

Pravin G. Ingole, Wook Choi, Kee Hong Kim, Chul Ho Park, Won Kil Choi, Hyung Keun Lee ⇑Greenhouse Gas Research Center, Korea Institute of Energy Research (KIER), 71-2 Jang-dong, Yuseong-gu, Daejeon 305343, Republic of Korea

h i g h l i g h t s

� Implications of the results for power generation by PRO are evaluated and discussed.� Demonstrated the effect of MPD, PIP and TMC monomers on dopamine coated HFM for PRO performance.� The TFC-HF membranes show enhancements in both water flux and power density.� Made an overall comparison and discussion of hollow fiber aspect design.

a r t i c l e i n f o

Article history:Received 28 October 2013Received in revised form 20 December 2013Accepted 30 December 2013Available online 11 January 2014

Keywords:PolydopamineSurface modificationHydrophilicityPROPower density

a b s t r a c t

To develop a high flux and high power density the polyethersulfone (PES) membrane support was mod-ified by coating with polydopamine (PDA), based on the easy self-polymerization and strong adhesioncharacteristics of dopamine (DA) under mild conditions. After polydopamine coating the effect of thinfilm composite (TFC) on polydopamine layer was studied. The surface morphological changes were char-acterized using SEM. The influence of the modifying conditions such as coating time and DA concentra-tion on the membrane properties was investigated. It was found that the most reasonable modificationconditions to obtain the high power density performance are 0.2 g/l DA concentration with 1 h coatingtime. After PDA coating the TFC active layer was prepare on the surface of coated membrane. Therewas tremendous increment of the flux for the polydopamine coated thin film composite (PDA-TFC) mem-brane under the best conditions compared with that of the original membrane (PES substrate). The resultshowed that the PDA-TFC membrane has higher flux, excellent power density performance and good sta-bility. The results by using these membranes were promising, indicating that the modified membranesexhibited an increase in flux performance under testing conditions when compared to baseline controldata. This modification method, which is scalable, has the potential to enable the use of existing thin filmcomposite membranes for all engineered osmosis applications. This work provides a simple method ofmodifying PES membrane for better performance and potential application in pressure retarded osmosis(PRO).

� 2014 Elsevier B.V. All rights reserved.

1. Introduction

Power generation by using pressure retarded osmosis (PRO) of-fers the possibility for utilizing osmotic pressure gradients for awide range of applications. PRO is an emerging platform technol-ogy that has the potential to sustainably produce electric power.It has therefore garnered great interest amongst the membrane sci-ence community within the past half-decade [1–3]. Unlike hydrau-lically driven membrane processes, PRO exploits the naturalphenomenon of osmosis, which occurs when two solutions of

differing concentration are placed on two sides of a semi-perme-able membrane. The generated osmotic pressure difference drivesthe permeation of water across the membrane from the dilutesolution to the concentrated solution. As applications in direct os-motic concentration (DOC) for concentrating high-value solutes,forward osmosis (FO) for seawater desalination and PRO for elec-tric power generation [4–6].With fast economic development andremarkable increase in population, the world is facing distinctivechallenges of energy supply. There is a need to reduce green housegas (GHG) emissions from fossil fuels has focused attention onalternative energy sources such as solar, wind, geothermal energy,etc., for the global sustainable development and the option of theseproblems is the generation of osmotic power [7]. Its well known

138 P.G. Ingole et al. / Chemical Engineering Journal 243 (2014) 137–146

when two torrents with different salinities are mixed, a significantamount of energy is released. Loeb et al. were reported innovativeexperimental work on PRO in 1976 [8]. The osmotic power can besubsequently harvested in the form of electricity by running thepressurized draw solution through a hydroturbine generator[9,10]. Comprehensive reviews of the PRO technology have beenmade in recent publications [11,12]. The PRO technology, pio-neered by Leob and co-workers [9,13,14], has received significantinterests in past years [15–20]. While early PRO studies usingasymmetric reverse osmosis membranes observed extremely lowpower density due to their thick support layers [3,8,21–24], recentdevelopments in osmotic membranes and processes have shownpromising progresses [15–20,23–25]. The zeolites are also the goodmaterial for to increase the flux as well as useful in microsystemsfor chemical synthesis and energy generation [26–28].

The adhesive protein in mussels shows similar properties likenovel polymer polydopamine (PDA) material. The outstandingadhesive behaviors and adhesive mechanisms of PDA have been re-ported [29]. DA (3,4-dihydroxy-phenylalanine) is able to self-poly-merize in aqueous solution and form a PDA layer, which cansteadfastly attach to a wide range of substrates such as rocks, met-als, polymers and wood [30]. The polar groups in PDA layer, such ashydroxyl and amine groups, bequeath the substrates with im-proved hydrophilicity and anti-fouling ability [31]. Additionally,PDA has been successfully applied in modifying PE, PVDF and PSpolymer membranes for higher membrane fouling resistance[32–34]. However, so far, only limited information on the applica-tion of PDA modification is available.

A hydrophilic polymer supported PDA coated thin film compos-ite membranes possessing the benefits of PRO membranes mightbe suitable for power regeneration processes. It is highly possiblethat this membrane still retain the high permselectivity as com-mercial RO membrane, and low resistance to mass transfer likecommercial FO membranes due to improved hydrophilicity ofthe support. There was great increment of the flux for the PDAmodified TFC membranes under the best conditions comparedwith that of the original membrane. The result showed that thePDA modified membrane has higher flux, excellent power densityperformance and good stability. In this research, PDA is used formembrane modification because of its highly hydrophilic natureand the stable polymerization of simple ingredients under mildreaction condition. We choose the PES hollow fiber membranefor PDA modification. The application performance of PDA coatedTFC membranes is also reported [35]. This study systematicallyinvestigated the effects of operating conditions i. e. effect of DAcoating, effect of TFC on PDA coating, effect of coating times, effectof pressures on the membrane performance and PRO performanceusing self making PES hollow fiber membranes as a substrate.These modified membranes show with reasonably high waterfluxes under the pressure retarded osmosis (PRO) mode using0.6 M NaCl as the draw solution and deionized water as the feed.The phenomenon of reverse solute diffusion and its adverse effecton PRO performance will also be discussed. Results obtained in thisstudy may provide significant insight into the PRO operation andPRO membrane design conditions.

2. Theory

2.1. Standard osmosis processes

The semi-permeable thin film membrane is use for only allowsthe passage of water but fully rejects other solute molecules/ions,the water flux in an osmosis process can be depicted as

Jw ¼ AðDp� DPÞ ð1Þ

where Jw is the volumetric water flux through the membrane, A isthe water permeability of the membrane, Dp and DP are the osmo-tic and hydrostatic pressure differences across the membrane,respectively. PRO is an intermediate osmosis process between theFO and RO, where the hydrostatic pressure of the draw solution islower than the osmotic pressure difference across the membrane,so that the water permeates from the feed (fresh water) side tothe draw (salty water) side. In terms of energy production or con-sumption, which is normally evaluated based on power density de-fined as the product of the trans-membrane hydrostatic pressure ofthe draw solution and the water flux permeating across the mem-brane, the power density (W) versus hydrostatic pressure differencebased on Eq. (2):

W ¼ JwDP ¼ AðDp� DPÞDP ð2Þ

It can be seen that pressure energy is produced in the PRO pro-cess by transferring the water from a low pressure side to a highpressure side means feed side to draw side. The energy densitymeans the amount of osmotic power produced per membrane areais a major performance indicator in PRO process, as it determinesthe amount of membrane area and thus the size of the PRO plantfor a given energy production capacity. There exists a maximumpower density when the hydrostatic pressure difference is equalto the half of the osmotic pressure difference, suggesting the opti-mal working condition for a PRO system.

3. Experimental

3.1. Materials

Polyethersulfone (PES, Ultrason� E6020P, BASF, Germany), usedas the base polymer, was purchased from General Electric Com-pany. N-methyl-2-pyrrolidion (NMP) with purity more than99.5% was purchased from Merck and was used as solvent withoutfurther purification. Lithium chloride (LiCl, Sigma Aldrich) wasused as pore former in dope solution.

Dopamine, diamine monomer m-phenylenediamine (MPD),piperazine (PIP) and acid chloride monomer trimesoyl chloride(TMC) were purchased from Sigma-Aldrich. Hexane, the solventfor TMC, was purchased from Fisher Scientific. Deionized water(DI) obtained from a Milli-Q ultrapure water purification system(Millipore) was used as the solvent for diamine monomers. Sodiumchloride was purchased from Fisher Scientific.

3.2. Methods

3.2.1. Preparation of PES hollow fiber membraneHollow fiber membranes were produced by a dry/wet phase

inversion method [36]. Commercially available PES was used asmembrane material. N-methylpyrrolidone and LiCl were used asthe solvent and additive, respectively. Distilled water was usedas internal coagulant and tap water as the external coagulant inthe hollow fiber spinning. Fig. 1 represents the schematic diagramof hollow fiber membrane spinning system.

Dope solution was mixed by a mechanical stirrer (160 rpm, at80 �C) after removing moisture in the PES at 80 �C over a three-day period. In order to remove bubbles that formed in the mixingprocess, the dope solution was left in a vacuum tank for one day.After that, the viscosity was measured with a viscometer (LVDV-II PRO, BK instrument, Denmark) at 25 �C. The dope solution wassupplied by using a gear pump, and 90 lm line-filter was set upto balance out the supply of the dope solution. Internal coagulant(DI water) was supplied by using an HPLC pump (Series II pump,Lab Alliance, USA). Dope solution and internal coagulant were

N2

a. Dope reservoir b. Inner coagulant reservoir c. Gear pump

d. HPLC pump e. Water bath f. Line filter

g. Chiller h. Spinneret i. 1st coagulation bath

j. 2nd coagulation bath k. Tension meter l. Winder

Fig. 1. Schematic diagram of HFM spinning system.

P.G. Ingole et al. / Chemical Engineering Journal 243 (2014) 137–146 139

passed through a double pipe spinneret with a 0.16/0.9 mm inner/outer diameter.

Hollow fiber passed through the first coagulation bath wherephase inversion occurred rapidly. After that, it moved to the secondcoagulation bath where hollow fiber was washed out and coiledaround the winder. The as-spun fibers were rinsed in a water bathfor six days to remove the remaining solvent. Then, fibers werepost-treated with methanol for 2 h to improve flux and were driedfor six days. The composition of dope solution and spinning condi-tion used in the preparation of PES hollow fiber membrane isshown in Table 1.

3.2.2. Preparation of membranes for dopamine coating on hollow fibermembrane

The cleaned PES membranes were immersed in the solution of2 mg/ml DA-HCl at pH 8.5 (PBS buffer). Since PDA has a tendencyto form free PDA particles at high dopamine concentration or tem-perature [37]. The newly prepared solution was shaken vigorouslyat 25 �C to avoid the formation of large PDA particles. After desiredtime of polymerization, the coated membranes were washed withdouble distilled water for 24 h to remove the redundantly darkbrown precipitates. The desired coating time periods for all mem-branes are shown in Table 2.

3.2.3. Preparation of thin film composite polymer layer on hollow fibermembrane

The thin film composite membrane was prepared by coatingselective layer in situ on the surface of polydopamine coated hol-low fiber membrane by interfacial polymerization of m-phenylene-diamine (MPD) with trimesoyl chloride (TMC) as well as piperazine(PIP) with TMC. The polydopamine coated hollow fiber membranewas immersed in an aqueous solution of MPD for 3–5 min followedby draining off for 2–5 min to remove excess solution. It was thenimmersed into hexane solution of TMC of desired concentration for60 s followed by draining off excess solution. The polymerizationreaction occurs on the surface of polydopamine coated hollow fibermembrane resulting in the formation of an ultrathin layer of

Table 1Composition of dope solution and spinning condition for the preparation of PEShollow fiber membrane.

CompositionPES 18.0 wt.%NMP 77.0 wt.%LiCl 5.0 wt.%

Spinning conditionInternal coagulant Distilled waterInjection rate of dope solution 5.5 mL/minInjection rate of internal coagulant 2.5 mL/minWinding speed 18 m/min

cross-linked co-polyamide. The composite membrane so obtainedwas cured in hot air circulation at 70–80 �C for 5 min wherebypolymer layer attains chemical stability [38]. After heat treatmentdry the membrane at room temperature for 2 h and then stored itin DI water until next use. Similar procedure is followed in the caseof piperazine, TMC for the preparation of thin film compositemembrane. The Table 2 discloses the compositions of solutionsand reaction times used for the interfacial polymerization.

3.2.4. Membrane performance evaluationPDA coated thin film composite membranes were evaluated for

permeates flux and rejection on a PRO test kit. A schematic dia-gram of the bench-scale PRO setup is shown in Fig. 2. A high pres-sure pump was used to transport and pressurize the draw solutionthat passed through the active surface side of the hollow fibermembrane module. The active layer of the membrane was alwaysfacing the draw solution for the PRO tests. Membrane permeateflux (i.e., volumetric flux of water) was determined at predeter-mined time intervals by measuring the weight changes of the feedtank with a digital mass balance connected to a computer data log-ging system. Testing was done with NaCl solution at different oper-ating pressures. Hollow fiber membrane module was preparedwith effective membrane area was around 7.2 cm2 and the lengthof module is around 20.0 cm. A standardized digital conductivitymeter of UTV, US was used to measure the salt concentrations inthe feed and product water for determining membrane selectivity.The volume of permeate collected was used to describe flux interms of liter per square meter of active membrane area per hour(L/m2 h).

3.2.5. Membrane characterization3.2.5.1. Attenuated total reflection-Fourier transforms infraredspectroscopy (ATR-FTIR). Information about the presence of specificfunctional groups on the membrane surfaces was obtained byATR-FTIR. FTIR spectra for surface chemistry of the fibers wereperformed by ALPHA-P Spectrometer with a diamond ATR cell(Bruker) in the range of 600–4000 cm�1. A total of 30 scans wereperformed at a resolution of 4 cm�1 with a germanium crystal attemperature of 25 ± 1 �C. A program written for the V2 softwarefrom Bruker was used to record the spectra and for the selectionof the corresponding backgrounds.

3.2.5.2. Scanning electron microscopy (SEM) analysis. The Fiber wascharacterized by Scanning Electron Microscopy (SEM, S-4700, Hit-achi). After being dried off completely, the membrane was coatedwith polydopamine and thin film composite layer on polydop-amine coated hollow fiber membrane. A scanning electron micro-scope was used to determine the asymmetric structure and thedimensions of the fibers. Membrane samples were first immersedin distilled water, fractured in liquid nitrogen and then sputtered

Table 2Dopamine coating time and compositions of aqueous and non-aqueous solutions and reaction times used for the preparation of selective layer.

Entry Membrane code Dopamine coating (h) Composition Reaction time (s)

MPD/PIP (%w/w) TMC (%w/w)

1 PDA-M1 1 – – –2 PDA-M2 2 – – –3 PDA-M3 3 – – –4 PDA -TFC (MPD + TMC) 1 1.0 0.1 605 PDA � TFC (PIP + TMC) 1 1.0 0.1 60

Fig. 2. Pressure retarded osmosis (PRO) testing unit.

PES substrate

PIP+TMC thin film on PDA

MPD+TMC thin film on PDA

PDA Coating

% T

rans

mit

tanc

e (a

.u.)

140 P.G. Ingole et al. / Chemical Engineering Journal 243 (2014) 137–146

with a thin layer of gold using a Balzers Union SCD 040 sputteringapparatus.

3.2.5.3. Atomic force microscopy (AFM) analysis. Surface roughnessof the different coated membranes was measured by AFM. The sur-face roughness values of the composite membranes were obtainedusing a Nanoman AFM system (Veeco) in tapping mode. Experi-ments were conducted in Korea Advanced Institute of Scienceand Technology (KAIST), Daejeon, South Korea. Small strip of mem-branes was placed on specific sample holder and 3 lm � 3 lmareas were scanned. Mean roughness (Ra), root mean square Z val-ues (Rms), and maximum vertical distance between the highest andlowest data points (Rmax) were used to quantify the surface topol-ogy of membranes.

3.2.5.4. Water contact angle of membranes. Surface hydrophilicity ofmembrane substrates was evaluated by contact angle drop shapegeometry (DSA100, Germany) using Milli-Q deionized water asthe probe liquid at room. To minimize the experimental error,the contact angle was randomly measured at more than 15 differ-ent locations for each sample and the average value was reported.

3500 3000 2500 2000 1500 1000 500

Wavelength (cm-1)

Fig. 3. ATR-FTIR spectra of PES-HF membrane after thin film coating using differentconcentration of monomers (peaks around 1105, 1149, 1240, 1290, 1321, 1484,1540, 1577, 1660 cm�1 correspond to CAO stretching, symmetric stretching ofO@S@O, aromatic ether, CAN stretching, asymmetric O@S@O stretching, aromaticnature (C@C stretching), amide II (NAH), aromatic nature (benzene ring), amide I(C@O) respectively).

4. Result and discussion

4.1. Membrane physicochemical characteristics

4.1.1. ATR-FTIRFig. 3 shows the ATR-FTIR spectra of different PES membranes

before and after polydopamine coating and interfacial

polymerization forming a polyamide (PA) thin film coating. FTIRspectra displaying peaks at around 1107, 1151, 1243, 1291, 1322,1487 and 1578 cm�1 are characteristics of the PES membranematerial. In particular peaks around 1487 and 1574 cm�1 are thecharacteristics of PES. The FTIR spectrum of the PDA coated

P.G. Ingole et al. / Chemical Engineering Journal 243 (2014) 137–146 141

PES-1 h substrate is composed of bands attributed to both the PDAlayer and PES gallows. For PDA coated PES-1 h, the peaks at1574 cm�1 is assigned to the deformation vibration of N–H [39],and the peak at 1650 cm�1 is attributed to stretching vibration ofC@O which is formed by the oxidation of the catechol groups intoquinine during the self-polymerization [40,41]. After surface coat-ing of PDA, compared with the PES membrane, several new absorp-tion signals appeared, a broad peak between 3600 and 3000 cm�1

was attributed to the stretching vibrations of the NAH/OAH andthe intensive peak at 1610 cm�1 was the overlapped peaks ofCAC vibrations of the aromatic ring and the NAH bending vibra-tions. These changes of the characteristic peaks indicated the exis-tence of PDA on the membrane surface, and the PDA coating wassuccessful. The specific chemical bonds for PDA obtained fromATR-FTIR further confirm the formation of a PDA layer that canfunctionalize the membrane surface. In addition to these the pres-ence of two peaks around 1668 and 1547 cm�1 corresponding toamides I (C@O) and II (NAH) respectively in all remaining twospectra are the indication of PA film formation. The penetrationdepth for IR was calculated using the equation reported by Oldaniand Schock [42]. Assuming the refractive index of the polymer tobe 1.5, IR incident angle 45� and wavelength region 2–20 lm, theradiation penetration depth is dependent on the refractive indexof the IR spectrometer crystal. For germanium crystal, therefractive index is 4.00. Thus the radiation penetration depth is0.11–1.1 lm. The thickness of the thin film (t) for each membranewas determined by multiplying the depth of penetration of IRradiation corresponding to the amide I frequency (�1668 cm�1;for specific membranes) by the intensity of the peak (i.e. percentabsorption). The thickness of the thin film (t) coated on all themembranes varied in a range of 0.1–0.2 lm.

4.1.2. Scanning electron microscopyThe morphology of composite membrane was observed in scan-

ning electron microscope. The top and transverse section of com-posite membrane (Fig. 4) indicates clearly a top thin dense layer(selective layer) and porous thick bottom layer (support layer).Fig. 4a–j shows the SEM images of the selected PES membranes be-fore (Fig. 4a and b) and after coating of polydopamine and forma-tion of PA thin film coating on polydopamine coated membranes(Fig. 4c–j). The polydopamine coated membranes (Fig. 4c–f) showsdispersed granular surface with visible open pore structures. Thethickness of coated membrane is clearly seen in the figures, its in-creases with respect to the reaction time or called as polydopaminecoating times. The cross-sectional morphology of the resultant TFCcoated on polydopamine hollow fiber membranes is shown inFig. 4g and h and polyamide membranes dispersed granular sur-face with visible open pore structures were shows as a Fig. 4i(MPD + TMC) and Fig. 4j (PIP + TMC) respectively. Since the activepolyamide layer was synthesized on the outer surface of the poly-dopamine coated hollow fiber substrate, the structure of the mem-brane substrate is different much before and after interfacialpolymerization. Fig. 4g and h displays the evolution of top surfaceand cross-section morphology with PDA coating duration underthe same interfacial polymerization conditions. This thin PA layeris the functional selective layer whose nature primarily determinesthe water permeability and power density of the resulting TFC-PROmembrane. The MPD + TMC PA selective layers formed on PDAcoated HF substrates are thinner than that formed by usingPIP + TMC membrane materials its clearly shown in Fig. 4i andFig. 4j. However, the thickness of the former monotonically in-creases with increasing PDA coating time it is shown in Fig. 4c-f.Since the conditions for substrate fabrication as well as interfacialpolymerization are the same, obviously the physico-chemicalproperties of the intermediate PDA layer may play the most impor-tant role in determining the PRO performance of these TFC

membranes. The portion structure indicating crooked, wormlikestrands connected with each other was attempted to correlate withthe steric hindrance effect of MPD and TMC as well as PIP and TMCbecause compounds containing aromatic rings.

The evolution of color change with time of the PDA functional-ized PES hollow fiber membranes. Clearly, the dopamine has suc-cessfully self-polymerized and coated onto the PES hollow fibermembrane top surface and the PDA layer thickness increases withan increase in coating time and it is proved by taking SEM imagesafter and before coating of PDA and thin film on PDA coated mem-brane [43].

The morphology of PES substrates, PDA coated on PES and PDA-TFC PES hollow fiber membranes are shown in Fig. 4, it is clearlyseen that after coating the thickness is increases and that is indi-cated in the figure. Consistent with the SEM images shown inFig. 4, this phenomenon is mainly due to the coated PDA layerwhich partially blocks some surface pores at first and then formsa denser PDA thin film layer when the coating time is long enough[35]. Interestingly, the PWP firstly increases up to the highest valueof 27.91 L/m2 h after 1 h PDA coating due to the increased mem-brane hydrophilicity. However, the PWP begins to decrease whenfurther increasing the coating time to 2 h and 3 h. Although, the in-creased membrane hydrophilicity can compensate part of the in-creased resistance due to the reduced pore size, the latergradually plays the dominant role when a further increase in coat-ing time.

4.1.3. Surface morphologies of the thin-film layer by AFMThe roughness of the outer surface of the hollow fiber mem-

branes was determined by AFM. Fig. 5 shows the 3-dimensional(3D) micrograph indicating the surface topology of the substrateand the TFC membrane. From the three-dimensional AFM imagesin Fig. 5, it can be observed that PES substrates coated by PDAfor different durations have different top surface roughness. Table 3summarizes the surface roughness parameters in terms of Ra, Rms

and Rmax for substrate PES, PDA 1 h coated membrane along withboth types of TFC membranes. The mean roughness (Ra) andmean-square surface roughness (Rms) decrease with an increasein PDA coating time [35]. Rmax also follows a similar trend excepta slight variation for the sample coated for 3 h. Generally, the mod-ified membranes have improved hydrophilicity, smoothened sur-face, smaller surface pores and narrower pore size distributioncompared to the uncoated substrate [44–46]. These improvementsmay affect the formation of the selective polyamide layer andinfluence the PRO performance of the resultant TFC membranes.

4.1.4. Contact angleTable 4 shows the water contact angle of the top surface of PDA

coated PES substrates as well as TFC membranes. Consistent withthe observation by Arena et al. [3], an increase in PDA coating timeresults in a rapid decrease in water contact angle from 107� (±5) ofthe PES to 58� (±4) of the PDA coated PES-3 h and 51� to 54� (±4)for TFC membranes. Furthermore, dopamine may penetrate intothe pores inside the substrate and attach onto the pore wall viaself-polymerization reaction during the PDA modification process,which can enhance their hydrophilicity [30].

4.2. Determination of membrane water and salt permeabilities

Pure water permeability (PWP) and salt permeability of thepolydopamine coated and modified TFC-polydopamine coated hol-low fiber membranes by using PRO mode were evaluated in a lab-oratory-scale cross flow PRO test unit. The effective membranearea was 7.2 cm2. The HF membrane module was first compactedwith DI water at an applied pressure, DP, of 1.0 bar. The appliedpressures were changes up to 7 bars.

Fig. 4. SEM images of the selected PES membranes before (Fig. 4a and b) and after coating of polydopamine (Fig. 4c–f) and after formation of polyamide thin film coating(Fig. 4g–j).

142 P.G. Ingole et al. / Chemical Engineering Journal 243 (2014) 137–146

Salt rejection was characterized by keeping the applied pres-sures up to 7 bar and measuring rejection of 500 ppm NaCl solutionusing a calibrated conductivity meter. Observed NaCl rejection, R,

was determined from the difference in feed (Cf) and permeate(Cp) salt concentrations,

Fig. 5. AFM images of the selected PES membranes before (Fig. 5a) and after formation of polyamide (Fig. 5b–d) thin film coating.

Table 3Surface roughness values of the thin film composite membranes.

Membranes Rms (nm) Ra (nm) Rmax (nm)

PES 11.37 7.50 69.94PDA-1 h coated 11.08 7.49 68.40Composite M1 10.12 7.39 53.53Composite M2 4.86 3.34 37.77

Table 4Contact angles of support as well as coated membranes.

Entry Membrane name (time) Contact angle (�)

1 PES-HF membrane substrate 107 ± 52 PDA coating – 1 h 68 ± 43 PDA coating – 2 h 63 ± 44 PDA coating – 3 h 58 ± 45 PDA coating (1 h)-TFC (MPD + TMC) 51 ± 46 PDA coating (1 h)-TFC (PIP + TMC) 54 ± 4

P.G. Ingole et al. / Chemical Engineering Journal 243 (2014) 137–146 143

R ¼ 1� Cp=Cf ð3Þ

The rejection values for each sample are the average of threedifferent measurements collected over a �10 min period. The tem-perature of the system was maintained at 25 �C throughout theexperiment. Fig. 6 and Fig. 7 were shows the variation in purewater flux and Fig. 8 and Fig. 9 were show the variation in saltrejection (500 ppm) with operation pressure respectively. Withthe increase in dopamine concentration NaCl rejection by the PA

thin film increases. The maximum NaCl rejection and flux was98.5% and 24.91 L/m2 h by using TFC-M1 (MPD + TMC) on 1 hcoated PDA and 98.7% and 19.57 L/m2 h by using TFC-M2(PIP + TMC) on 1 h coated PDA hello fiber membrane respectively.The thin film displayed high free volume and was therefore ex-pected to display high flux. In our case experimental results slowsthat increase in flux by using MPD as monomer crosslinked withTMC could be due to the increased free volume within the thinfilm.

4.3. Water flux and power density

By comparing the PRO performance of variation of PDA coatedTFC-M1 (MPD + TMC) PES HF membranes using a 0.6 M NaCl solu-tion as the draw solution and deionized water as the feed solutiontested under PRO modes. The TFC on PDA coated PES HF mem-branes made from the PIP + TMC shows a poor water flux as lowas 5.84 L/m2 h under the PRO mode with 1.09 W/m2 power densityby using 3 h PDA coated TFC membrane and by using 1 h PDAcoated TFC membrane we got 8.45 L/m2 h with 1.58 W/m2 fluxand power density respectively at 7 bar (Fig. 10 and Fig. 11). How-ever, the water flux dramatically increases when the PA activelayer is formed on the PDA coated membrane by using MPD + TMC,and it reaches to the lowest value of 9.22 L/m2 h with 1.81 W/m2

power density by using 3 h PDA coated TFC membrane and byusing 1 h PDA coated TFC membrane we got 15.42 L/m2 h and2.98 W/m2 flux and power density respectively at 7 bar (Fig. 12and Fig. 13).

Fig. 6. Variation in pure water flux using TFC-M1 (MPD + TMC) membrane withoperation pressure.

Fig. 7. Variation in pure water flux using TFC-M2 (PIP + TMC) membrane withoperation pressure.

Fig. 8. Variation in salt rejection using TFC-M1(MPD + TMC) membrane withoperation pressure.

Fig. 9. Variation in salt rejection using TFC-M2 (PIP + TMC) membrane withoperation pressure.

Pressure (Δ P) bar0 1 2 3 4 5 6 7 8

Wat

er F

lux

(L/m

2 h)

4

6

8

10

12

14

16PDA-1 h- TFC (PIP+TMC)PDA-2 h- TFC (PIP+TMC)PDA-3 h- TFC (PIP+TMC)

Fig. 10. Variation in water flux using TFC-M2 (PIP + TMC) membrane withoperation pressure.

144 P.G. Ingole et al. / Chemical Engineering Journal 243 (2014) 137–146

To evaluate the impact of PDA coating on the PES hollow fibermembranes for PRO applications, the membranes were tested forosmotic flux in the PRO mode. The both TFC on PDA coated mem-branes were compared. The PDA modified membrane exhibited

substantial flux improvement, indicative of an increasing thehydrophilicity of the membrane support layer. This increasedhydrophilic nature promotes water transport through the supportlayer and to the interior interface of the polyamide layer.

4.4. Effect of coating time on the performance of hollow fibermembranes

As we seen after the coating of polydopamine on the surface ofhollow fiber membranes there is tremendous effect towards mem-branes performances. We study the series of coated membrane byusing two types of monomers with trimesoyl chloride. With the in-crease of PDA coating time, the corresponding membranes showdecreased fluxes as well as power density. The water permeabilitycoefficient of PDA coated PES HF membranes and TFC on PDAcoated PES HF membranes firstly increases and then decreaseswith the raising of PDA coating time. The PDA coated PES HF mem-branes and TFC on PDA coated PES HF membranes made from aPDA coating time-1 h shows the highest water permeability andpower density. Since the membrane constructed on a PDA coatingtime-1 h has the most balanced PRO performance using NaCl solu-tions with 0.6 M concentration as draw solution and deionizedwater as the feed solution. The water permeability coefficient ofTFC-PRO membranes firstly increases and then decreases with

0 1 2 3 4 5 6 7 8

Pow

er d

ensi

ty (W

/m2 )

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8PDA-1 h- TFC (PIP+TMC)PDA-2 h- TFC (PIP+TMC)PDA-3 h- TFC (PIP+TMC)

Pressure (Δ P) bar

Fig. 11. Variation in power density using TFC-M2 (PIP + TMC) membrane withoperation pressure.

0 1 2 3 4 5 6 7 8

Wat

er F

lux

(L/m

2 h)

8

10

12

14

16

18

20PDA-1 h- TFC (MPD+TMC)PDA-2 h- TFC (MPD+TMC)PDA-3 h- TFC (MPD+TMC)

Pressure (Δ P) bar

Fig. 12. Variation in water flux using TFC-M1 (MPD + TMC) membrane withoperation pressure.

0 1 2 3 4 5 6 7 8

Pow

er d

ensi

ty (W

/m2 )

0.0

0.5

1.0

1.5

2.0

2.5

3.0PDA-1 h- TFC (MPD+TMC)PDA-2 h- TFC (MPD+TMC)PDA-3 h- TFC (MPD+TMC)

Pressure (Δ P) bar

Fig. 13. Variation in power density using TFC-M1 (MPD + TMC) membrane withoperation pressure.

P.G. Ingole et al. / Chemical Engineering Journal 243 (2014) 137–146 145

the raising of PDA coating time. The TFC-PRO membrane madefrom a PDA-1 h coated PES hollow fiber substrate shows the high-est water permeability at 7 bar. In addition, a tedious decrease insalt permeability coefficient is observed for TFC-PRO membranes

with the increase of PDA coating time. However, the water fluxdramatically decreases when the coating time increases, and itreaches to the lowest value of 9.22 L/m2 h with 1.81 W/m2 powerdensity by using 3 h PDA coated TFC membrane and by using 1 hPDA coated TFC membrane we got 15.42 L/m2 h and 2.98 W/m2

flux and power density respectively at 7 bar as shown in Fig. 12and Fig. 13.

The polyamide film formation is described to take place inthree steps: embryonic film formation which is a fast process fol-lowed by slow down in polymerization depending upon the per-meability of the initial film formed; finally shifting to a diffusioncontrolled process. The initial layer formed during the embryonicfilm formation is the actual barrier layer controlling the separationcharacteristics of the thin film and divides the film in two regions;each region is rich in one type of monomer and end group. In thediffusion controlled step film growth takes place until the mono-mers diffusing through the film get consumed by other monomersand/or unreacted functional groups of the film [47]. Furthermoreincrease in film density can result in further increase in powerdensity observed in our case at MPD concentrations (1.0%). Furthermembranes prepared using 1.0% concentration of MPD and PIPsolutions with 0.1% TMC concentration exhibited higher flux withhigh power density.

The interfacial polymerization between MPD (as well as PIP)and TMC occurs in the organic side; the reaction is diffusion-con-trolled and exists in a self-limiting phenomenon. The reaction timeplays an important role in determining the extent of polymeriza-tion, and thereby the cross-linking degree and thickness of top skinlayer as well as the resulting membrane performance [48,49]. Thebest reaction time of 60 s. was selected due to the high power den-sity at high pressure.

5. Conclusions

In the present study, an improved procedure has been demon-strated to fabricate the thin film composite PRO (TFC-PRO) mem-branes, which was facilely processed by surface coating on thetop surface of PES hollow fiber substrates with polydopamine(PDA) before the conventional reaction of interfacial polymeriza-tion. This improved fabrication process can provide a new para-digm for the preparation of high performance TFC-PROmembranes. The modified membranes must possess toughmechanical properties to with stand the compression, trim andelongational stresses during the high pressure PRO process withoutshowing structural damage and sacrificing much water flux. Thefollowing conclusion can be made from this study:

1. This improved process resulted in TFC-PRO membranes withsimultaneous enhancements in both water permeability andsalt rejection properties. A high water flux and power densityhas been achieved when using a 0.6 M NaCl as the draw solu-tion and deionized water as the feed solution in the active layerfacing the draw solution configuration at room temperature.

2. The PDA surface coating plays a positive role in the preparationof the PA selective layer, which is realized by producing ahydrophilic smooth membrane surface with smaller pore sizesand a narrower pore size distribution for carrying on the inter-facial polymerization reaction, as well as enhancing the hydro-philicity of the pore-wall inside the substrate layer. In thisstudy, the optimized PDA coating time is around 1 h in termsof the best PRO performance.

3. The coated PDA functional layer could actively interact with theTMC monomer during the interfacial polymerization reaction.This is favorable for the formation of less defective PA layerfor the enhancing of water flux and power density.

146 P.G. Ingole et al. / Chemical Engineering Journal 243 (2014) 137–146

4. ATR-FTIR and SEM characterization have revealed that a thinnerand smoother polyamide layer with a larger free volume istherefore produced, which leads to a higher water flux, bettermechanical stability and greater power density.

5. Due to the dopamine coating the stability between the PA activelayer and the substrate membrane may also be enhanced.

Acknowledgements

This work was conducted under the framework of Research andDevelopment Program of the Korea Institute of Energy Research(KIER) (B3-2441-02).

References

[1] X. Song, Z. Liu, D.D. Sun, Energy recovery from concentrated seawater brine bythin-film nano fiber composite pressure retarded osmosis membranes withhigh power density, Energy Environ. Sci. 6 (2013) 1199–1210.

[2] C. Klaysom, T.Y. Cath, T. Depuydt, I.F.J. Vankelecom, Forward and pressureretarded osmosis: potential solutions for global challenges in energy and watersupply, Chem. Soc. Rev. 42 (2013) 6959–6989.

[3] J.T. Arena, B. McCloskey, B.D. Freeman, J.R. McCutcheon, Surface modificationof thin film composite membrane support layers with polydopamine: enablinguse of reverse osmosis membranes in pressure retarded osmosis, J. Membr. Sci.375 (2011) 55–62.

[4] J.R. McCutcheon, M. Elimelech, Influence of concentrative and dilutive internalconcentration polarization on flux behavior in forward osmosis, J. Membr. Sci.284 (2006) 237–247.

[5] J.R. McCutcheon, R.L. McGinnis, M. Elimelech, A novel ammonia—carbondioxide forward (direct) osmosis desalination process, Desalination 174 (2005)1–11.

[6] J.R. McCutcheon, R.L. McGinnis, M. Elimelech, Desalination by ammonia–carbon dioxide forward osmosis: influence of draw and feed solutionconcentrations on process performance, J. Membr. Sci. 278 (2006) 114–123.

[7] M.I. Hoffert, K. Caldeira, G. Benford, D.R. Criswell, C. Green, H. Herzog, A.K. Jain,H.S. Kheshgi, K.S. Lackner, J.S. Lewis, H.D. Lightfoot, W. Manheimer, J.C.Mankins, M.E. Mauel, L.J. Perkins, M.E. Schlesinger, T. Volk, T.M.L. Wigley,Advanced technology paths to global climate stability: energy for a greenhouseplanet, Science 298 (5595) (2002) 981–987.

[8] S. Loeb, F. Van Hessen, D. Shahaf, Production of energy from concentratedbrines by pressure-retarded osmosis. II. Experimental results and projectedenergy costs, J. Membr. Sci. 1 (1976) 249–269.

[9] S. Loeb, Production of energy from concentrated brines by pressure retardedosmosis. I. Preliminary technical and economic correlations, J. Membr. Sci. 1(1976) 49–63.

[10] S. Loeb, Large-scale power production by pressure-retarded osmosis, usingriver water and sea water passing through spiral modules, Desalination 143(2002) 115–122.

[11] A. Achilli, A.E. Childress, Pressure retarded osmosis: from the vision of SidneyLoeb to the first prototype installation – review, Desalination 261 (2010) 205–211.

[12] G.Z. Ramon, B.J. Feinberg, E.M.V. Hoek, Membrane-based production of salinitygradient power, Energy Environ. Sci. 4 (2011) 4423–4434.

[13] S. Loeb, Energy production at the Dead Sea by pressure-retarded osmosis:challenge or chimera?, Desalination 120 (1998) 247–262

[14] S. Loeb, One hundred and thirty benign and renewable megawatts from GreatSalt Lake? The possibilities of hydroelectric power by pressure-retardedosmosis, Desalination 141 (2001) 85–91.

[15] Y. Xu, X. Peng, C.Y. Tang, Q.S. Fu, S. Nie, Effect of draw solution concentrationand operating conditions on forward osmosis and pressure retarded osmosisperformance in a spiral wound module, J. Membr. Sci. 348 (2010) 298–309.

[16] N.Y. Yip, A. Tiraferri, W.A. Phillip, J.D. Schiffman, L.A. Hoover, Y.C. Kim, M.Elimelech, Thin-film composite pressure retarded osmosis membranes forsustainable power generation from salinity gradients, Environ. Sci. Technol. 45(2011) 4360–4369.

[17] S. Chou, R. Wang, L. Shi, Q. She, C.Y. Tang, A.G. Fane, Thin-film compositehollow fiber membranes for pressure retarded osmosis (PRO) process withhigh power density, J. Membr. Sci. 389 (2012) 25–33.

[18] R.L. McGinnis, J.R. McCutcheon, M. Elimelech, A novel ammonia–carbondioxide osmotic heat engine for power generation, J. Membr. Sci. 305 (2007)13–19.

[19] T. Thorsen, T. Holt, The potential for power production from salinity gradientsby pressure retarded osmosis, J. Membr. Sci. 335 (2009) 103–110.

[20] A. Achilli, T.Y. Cath, A.E. Childress, Power generation with pressure retardedosmosis: an experimental and theoretical investigation, J. Membr. Sci. 343(2009) 42–52.

[21] A. Tiraferri, N.Y. Yip, W.A. Phillip, J.D. Schiffman, M. Elimelech, Relatingperformance of thin-film composite forward osmosis membranes to supportlayer formation and structure, J. Membr. Sci. 367 (2011) 340–352.

[22] J. Wei, C. Qiu, C.Y. Tang, R. Wang, A.G. Fane, Synthesis and characterization offlat-sheet thin film composite forward osmosis membranes, J. Membr. Sci. 372(2011) 292–302.

[23] S. Zhang, P. Sukitpaneenit, T.-S. Chung, Design of robust hollow fibermembranes with high power density for osmotic energy production, Chem.Eng. J. 241 (2013) 457–465.

[24] D. Emadzadeha, W.J. Laua, T. Matsuura, M. Rahbari-Sisakht, A.F. Ismail, A novelthin film composite forward osmosis membrane prepared from PSf–TiO2

nanocomposite substrate for water desalination, Chem. Eng. J. 237 (2014) 70–80.

[25] N. Widjojo, T.-S. Chung, M. Weber, C. Maletzko, V. Warzelhan, A sulfonatedpolyphenylenesulfone (sPPSU) as the supporting substrate in thin filmcomposite (TFC) membranes with enhanced performance for forwardosmosis (FO), Chem. Eng. J. 220 (2013) 15–23.

[26] J.L.H. Chau, A.Y.L. Leung, K.L. Yeung, Zeolite micromembranes, Lab Chip 3(2003) 53–55.

[27] S.M. Kwan, K.L. Yeung, Zeolite micro fuel cell, Chem. Commun. (2008) 3631–3633.

[28] K.L. Yeung, S.M. Kwan, W.N. Lau, Zeolites in microsystems for chemicalsynthesis and energy generation, Top. Catal. 52 (2009) 101–110.

[29] H. Lee, N.F. Scherer, P.B. Messersmith, Single-molecule mechanics of musseladhesion, Science 103 (2006) 12999–13003.

[30] H. Lee, S.M. Dellatore, W.M. Miller, P.B. Messersmith, Mussel-inspired surfacechemistry for multifunctional coatings, Science 318 (2007) 426–430.

[31] L.P. Zhu, J.H. Jiang, B.K. Zhu, Y.Y. Xu, Immobilization of bovine serum albuminonto porous polyethylene membranes using strongly attached polydopamineas a spacer, Colloids Surf. B 86 (2011) 111–118.

[32] Z.Y. Xi, Y.Y. Xu, L.P. Zhu, Y. Wang, B.K. Zhu, A facile method of surfacemodification for hydrophobic polymer membranes based on the adhesivebehavior of poly(DOPA) and poly(dopamine), J. Membr. Sci. 327 (2009) 244–253.

[33] B.D. McCloskey, H.B. Park, H. Ju, B.W. Rowe, D.J. Miller, B.J. Chun, K. Kin, B.D.Freeman, Influence of polydopamine deposition conditions on pure water fluxand foulant adhesion resistance of reverse osmosis, ultrafiltration, andmicrofiltration membranes, Polymer 51 (2010) 3472–3485.

[34] J.H. Jiang, L.P. Zhu, X.L. Li, Y.Y. Xu, B.K. Zhu, Surface modification of PE porousmembranes based on the strong adhesion of polydopamine and covalentimmobilization of heparin, J. Membr. Sci. 364 (2010) 194–202.

[35] G. Han, S. Zhang, X. Li, N. Widjojo, T.S. Chung, Thin film composite forwardosmosis membranes based on polydopamine modified polysulfone substrateswith enhancements in both water flux and salt rejection, Chem. Eng. Sci. 80(2012) 219–231.

[36] K.H. Kim, P.G. Ingole, J.H. Kim, H.K. Lee, Separation performance of PEBAX/PEIhollow fiber composite membrane for SO2/CO2/N2 mixed gas, Chem. Eng. J. 233(2013) 242–250.

[37] B. Fei, B.T. Qian, Z.Y. Yang, R.H. Wang, W.C. Liu, C.L. Mak, J.H. Xin, Coatingcarbon nanotubes by spontaneous oxidative polymerization of dopamine,Carbon 46 (2008) 1795–1797.

[38] P.G. Ingole, H.C. Bajaj, K. Singh, Optical resolution of racemic lysinemonohydrochloride by novel enantioselective thin film compositemembrane, Desalination 305 (2012) 54–63.

[39] J.F. Qu, J.Q. Wang, S. Liu, J.F. Zhou, S.R. Yang, Self-assembly and tribologicalproperty of a novel 3-layer organic film on silicon wafer with polydopaminecoating as the interlayer, J. Phys. Chem. C 113 (2009) 20429–20434.

[40] L. Shi, S.R. Chou, R. Wang, W.X. Fang, C.Y. Tang, A.G. Fane, Effect of substratestructure on the performance of thin-film composite forward osmosis hollowfiber membranes, J. Membr. Sci. 382 (2011) 116–123.

[41] L.P. Yang, S.L. Phua, J.H. Teo, X.H. Lu, A biomimetic approach to enhancinginterfacial interactions: polydopamine-coated clay as reinforcement for epoxyresin, ACS Appl. Mater. Interf. 3 (2011) 3026–3032.

[42] M. Oldani, G. Schock, Characterization of ultrafiltration membranes by infraredspectroscopy, ESCA and contact angle measurements, J. Membr. Sci. 43 (1989)243–258.

[43] E.L. Wittbecker, P.W. Morgan, Interfacial polycondensation 1, J. Polym. Sci. PartA: Polym. Chem. 34 (1996) 521–529.

[44] P.G. Ingole, K. Singh, H.C. Bajaj, Enantioselective polymeric compositemembrane for optical resolution of racemic mixtures of a-amino acids, Sep.Sci. Technol. 46 (2011) 1898–1907.

[45] P.G. Ingole, K. Singh, H.C. Bajaj, Enantioselective permeation of a-amino acidisomers through polymer membrane containing chiral metal–Schiff basecomplexes, Desalination 281 (2011) 413–421.

[46] K. Singh, H. Bajaj, P. Ingole, A method for preparation of enantioselectivecomposite membrane, WO Patent 2,010,109,490.

[47] R. Nadler, S. Srebnik, Molecular simulation of polyamide synthesis byinterfacial polymerization, J. Membr. Sci. 315 (2008) 100–105.

[48] P.W. Morgan, S.L. Kwolek, Interfacial polycondensation. II. Fundamentals ofpolymer formation at liquid interfaces, J. Polym. Sci. Part A: Polym. Chem. 34(1996) 531–559.

[49] A.L. Ahmad, B.S. Ooi, Properties–performance of thin film compositesmembrane: study on trimesoyl chloride content and polymerization time, J.Membr. Sci. 255 (2005) 67–77.