Advanced Anti-Fouling Membranes for Osmotic Power Generation

Subscriber access provided by Kaohsiung Medical University

Energy and the Environment

Advanced Anti-Fouling Membranes for Osmotic Power Generation from Wastewater via Pressure Retarded Osmosis (PRO) Gang Han, Jiang Tao Liu, Kangjia Lu, and Tai-Shung Chung Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b05933 • Publication Date (Web): 09 May 2018 Downloaded from http://pubs.acs.org on May 10, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

Environmental Science & Technology

1

Advanced Anti-Fouling Membranes for Osmotic

2

Power Generation from Wastewater via Pressure

3

Retarded Osmosis (PRO)

4

Gang Han,# Jiang Tao Liu,# Kang Jia Lu, Tai-Shung Chung*

5 6

Department of Chemical and Biomolecular Engineering, National University of Singapore,

7

Singapore 117585

8 9

Correspondence to: Tai-Shung Chung (Email: [email protected])

10

Tel: +65-65166645; Fax: +65-67791936

11

#

The first two authors contributed equally to this work

12 13

ABSTRACT: A facile and versatile approach was demonstrated for the fabrication of low-

14

fouling pressure retarded osmosis (PRO) membranes for osmotic power generation from highly

15

polluted wastewater. A water-soluble zwitterionic random copolymer with superior

16

hydrophilicity and unique chemistry was molecularly designed and synthesized via a single-step

17

free-radical polymerization between 2-methacryloyloxyethyl phosphorylcholine (MPC) and 2-

18

aminoethyl methacrylate hydrochloride (AEMA). The P[MPC-co-AEMA] copolymer was then

19

chemically grafted onto the surface of PES&Torlon hollow fibers via amino groups coupling of

20

poly(AEMA) with the polyimide structures of Torlon, leaving the zwitterions of poly(MPC) in

21

the feed solution. Due to the outstanding hydrophilicity, unique cationic and anionic groups and 1 ACS Paragon Plus Environment


Page 2 of 28

22

electrical neutrality of the zwitterionic brush, the newly developed membrane showed great

23

resistances to both inorganic scaling and organic fouling in PRO operations. When using a real

24

wastewater brine comprising multi-foulants as the feed, the P[MPC-co-AEMA] modified

25

membrane exhibits a much lower flux decline of 37% at ∆P=0 bar after 24-h tests and a smaller

26

power density decrease of 28% at ∆P=15 bar within 12-h tests, compared to 61% and 42%

27

respectively for the unmodified one. In addition to the low fouling tendency, the modified

28

membrane shows outstanding performance stability and fouling reversibility, where the flux is

29

almost fully recovered by physical backwash of water at 15 bar for 0.5 h. This study would

30

provide valuable insights and strategies for the design and fabrication of effective antifouling

31

materials and membranes for PRO osmotic power generation.

32 33

TOC Art

34 35 36

1. INTRODUCTION

37

The rapid growth in energy consumption and CO2 emission have stimulated worldwide search

38

for alternative energy sources. Salinity-gradient energy that is released from the mixing of water

39

streams with different salinities is an unexploited resource of clean energy.1-3 By employing a

40

semipermeable membrane to control the mixing process, salinity-gradient energy can be 2 ACS Paragon Plus Environment

Page 3 of 28


41

harvested via pressure retarded osmosis (PRO).4-6 In a typical PRO process, driven by the

42

osmotic pressure gradient, water permeates through the membrane from the feed solution into the

43

draw solution with a higher salinity against an uphill trans-membrane hydraulic pressure.1-7 The

44

salinity energy is therefore converted to osmotic power by releasing the hydraulic head generated

45

by the permeating water through an energy exchange device such as a hydro-turbine or pressure

46

exchanger. Because the global potential of osmotic power is projected to be huge with negligible

47

chemical usage or CO2 emissions, PRO becomes an important strategic thrust in solving global

48

energy and sustainability puzzles.4-9

49

In theory, power density (i.e., osmotic energy output per unit membrane area) of a PRO

50

membrane is determined by the product of the hydraulic pressure applied on the draw solution

51

and the water flux across the membrane.3-9 As a result, the concentrated brine from seawater

52

desalination and discharged effluents from municipal wastewater plants are recently utilized as

53

the feed pairs for PRO processes because of their high salinity gradient and merits of converting

54

waste to energy.10-14 The integration of PRO and seawater desalination also alleviates the

55

disposal problem of high-salinity brine and makes the desalination process less energy dependent

56

and more sustainable.15,16

57

In addition to find suitable feed pairs with a large salinity gradient, development of high

58

performance membranes with a high water flux and great mechanical stability is another focus in

59

PRO research. Many advances on the design and fabrication of effective PRO membranes have

60

been made using deionized water as the feed.6,17 However, fouling on PRO membranes is a big

61

challenge which significantly limits the process viability and membrane efficiency particularly

62

when using highly polluted solutions as the feeds.18-24 The unique water transport in PRO

63

processes tends to drag foulants toward the porous substrate underneath the dense-selective layer

3 ACS Paragon Plus Environment


Page 4 of 28

64

of PRO membranes. Using the state-of-the-art PRO membranes, thin-film composite (TFC)

65

membranes,25-30 as an example. Since they comprise a polyamide dense-selective skin and a

66

porous substrate layer, and the porous substrate faces the feed solution in PRO operations,

67

foulants such as inorganic salts, colloids, and organic matters in the feed are easily accumulated

68

inside the porous substrate by the water flow across the membranes. Not only do they

69

dramatically enhance internal concentration polarization (ICP) but also critically deteriorate the

70

effective osmotic pressure gradient for power generation.31-33 In addition, the formation of

71

fouling deep inside the substrate makes membrane cleaning sophisticated and difficult because of

72

high tortuosity in the substrate. As a result, how to effectively mitigate membrane fouling and

73

improve fouling reversibility within the porous substrate layer remains a major challenge for the

74

commercialization of PRO technology.

75

Therefore, this study aims to develop effective fouling resistant materials and antifouling

76

membranes to maximize the PRO efficiency for osmotic power generation. Zwitterionic

77

materials consisting of positively and negatively charged groups with the overall electrically

78

neutral show the potential.34-38 Electrostatically induced hydration of the zwitterions would form

79

a highly hydrated surface and results in outstanding hydrophilicity to inhibit the adsorption of

80

foulants.35,36 In order to take advantages of such zwitterionic materials in a simple and

81

straightforward way, a novel random copolymer comprising 2-methacryloyloxyethyl

82

phosphorylcholine (MPC) and 2-aminoethyl methacrylate (AEMA) was molecularly designed

83

and synthesized via a one-pot free-radical polymerization. As illustrated in Figure 1, the P[MPC-

84

co-AEMA] copolymer possesses a zwitterionic poly(MPC) segment that contains of positively

85

charged ammonium cations and negatively charged phosphate anions and a poly(AEMA)

86

segment with terminated amino groups. The poly(MPC) side chains perform as the fouling-


Page 5 of 28


87

resistant polymer brushes while the amino groups of the poly(AEMA) segment function as

88

anchors to efficiently graft the copolymer onto the surface of the PES&Torlon hollow fiber

89

membrane through covalent bonds with the polyimide groups of Torlon.

90

91 92

Figure 1. Schematic procedures for the fabrication of zwitterionic copolymer modified TFC-

93

PES&Torlon-P[MPC-co-AEMA] hollow fiber membranes with mitigated propensity for PRO.

94 95

The structure and chemistry of the synthesized P[MPC-co-AEMA] copolymer would be

96

systematically characterized. After that, the morphology, hydrophilicity and transport properties

97

of the pristine and copolymer modified TFC hollow fiber membranes as well as their PRO

98

performance would be evaluated. Lastly, the fouling behaviors, antifouling performance and

99

fouling reversibility of the newly developed membranes for osmotic power generation would be



Page 6 of 28

100

investigated using a real wastewater brine as the feed. This work may open up a new avenue in

101

the fabrication of zwitterionic materials functionalized antifouling membranes for PRO osmotic

102

power generation.

103 104

2. EXPERIMENTAL AND METHODS

105

2.1 Synthesis of the Zwitterionic Copolymer P[MPC-co-AEMA]

106

The zwitterionic random copolymer P[MPC-co-AEMA] was synthesized via a modified free-

107

radical polymerization between 2-methacryloyloxyethyl phosphorylcholine (MPC) and 2-

108

aminoethyl methacrylate hydrochloride (AEMA) using 2,2'-azobis(2-methylpropionitrile) (AIBN)

109

as the initiator,39 as illustrated in Figure S1. The detailed specifications of chemicals and

110

procedures for copolymer synthesis and purification are disclosed in the Supporting Information

111

(SI).

112 113

2.2 Fabrication of TFC-PES&Torlon Hollow Fiber Membranes

114

A PES&Torlon hollow fiber substrate was firstly prepared via a dry-jet wet spinning process.29,40

115

After that, a polyamide dense-selective thin film was formed on the inner surface (i.e., the lumen

116

side) of the hollow fibers via interfacial polymerization between m-phenylenediamine (MPD)

117

and trimesoyl chloride (TMC). Table S1 summarizes the specific spinning parameters, and the

118

detailed experimental procedures of fiber preparation, module fabrication and interfacial

119

polymerization reaction are similar to our previous works.29,41

120 121

2.3 Modification of TFC-PES&Torlon Membranes by the Zwitterionic Copolymer


Page 7 of 28


122

The outer surface of TFC-PES&Torlon hollow fiber membrane was modified by chemically

123

grafting the synthesized P[MPC-co-AEMA] zwitterionic copolymer as follows. A 0.8 wt %

124

coating solution was prepared by dissolving the copolymer in a 9:1 (wt %) mixture of 2-propanol

125

and water at room temperature. The P[MPC-co-AEMA] coating solution was slowly flowed

126

through the shell side of the module at 75 °C for 1.5 h. After the modification, the fibers were

127

carefully rinsed with deionized water to remove excess agents and then kept in deionized water

128

for further characterizations. The P[MPC-co-AEMA] copolymer modified TFC-PES&Torlon

129

membrane was termed as TFC-PES&Torlon-P[MPC-co-AEMA].

130 131

2.4 Characterizations

132

Fourier transform infrared (FTIR) spectroscopy was performed by a Bio-Rad FTS 135 Fourier

133

transform infrared spectrophotometer and the diffuse reflectance spectra were scanned over the

134

range of 400–4000 cm-1. Gel permeation chromatography (GPC) tests were carried out on a

135

Waters GPC system. Ultrapure water was used as the eluent and a calibration curve was

136

generated using polyethylene oxides as molecular weight standards. Membrane morphology was

137

observed by a field-emission scanning electron microscope (FESEM JEOL JSM-6700LV). The

138

contact angle of water on the outer surface of hollow fibers was carried out on a KSV Sigma 701

139

Tensiometer (KSV Instruments Ltd., Finland). Surface chemistry of the membranes was

140

analyzed by X-ray photoelectron spectroscopy (XPS, Kratos AXIS Ultra DLD spectrometer,

141

Kratos Analytical Ltd.) with a monochromatized Al Kα X-ray source (1486.71 eV photons).

142

The effective mean pore size, pore size distribution and molecular weight cut-off (MWCO) of

143

the hollow fiber substrates were measured via solute rejection experiments using polyethylene

144

glycols (PEG) as the neutral rejection probes.42,43 Mass transport characteristics, including pure



Page 8 of 28

145

water permeability coefficient (A, in L m-2 h-1 bar-1) and NaCl rejection (R, %) of the TFC hollow

146

fiber membranes were characterized by testing the membranes under the RO mode via a

147

circulating RO filtration apparatus. The detailed procedures to measure A, R, salt permeability

148

coefficient (B) and membrane structural parameter (S) were described in SI and our previous

149

works.41,44

150 151

2.5 Membrane Performance and Fouling Tests in PRO Operations

152

A laboratory-scale crossflow PRO system was used for PRO performance tests and membrane

153

fouling evaluation.41 NaCl solutions with various concentrations were used as the draw solutions.

154

Deionized water was firstly utilized as the feed to obtain the benchmark PRO performance of the

155

membranes. Then, a real wastewater brine collected from a local municipal wastewater recycling

156

plant and synthetic solutions that contain single foulants were used as the feeds to investigate the

157

membrane antifouling performance. The chemistry of the real wastewater brine was reported in

158

our previous works,33,45 and the compositions of the synthetic feed solutions are tabulated in

159

Table S2.

160

For the long-term fouling tests, the draw solution concentration was adjusted to achieve a

161

similar initial water flux for different fibers. The draw solution concentration was maintained

162

constant throughout each test by monitoring the solution conductivity. The reduction of water

163

flux or power density as a function of testing duration was monitored. Membrane cleaning was

164

performed by backwashing with water from the lumen into the shell side of the hollow fiber

165

module at 15 bar for 30 min. Each PRO test was carried out three times and the averaged data

166

were reported. In order to make the figures clearly visible, the error bar was not included in the

167

reported figures.


Page 9 of 28

168 169


The water permeation flux, Jw (in L m−2 h−1, abbreviated as LMH), was calculated from the feed volume change as:

Jw =

170

∆V Sm∆t

(1)

171

where ∆V (in L) was the permeation water collected over a predetermined testing duration ∆t (in

172

h). Membrane power density, W (in W/m2), was then computed as:

173

W = ∆P × J w

(2)

174

where ∆P (in bar) was the hydraulic pressure difference across the membrane. The detailed

175

experimental setup, operating conditions, and the determination of water flux and membrane

176

power density can be found in SI.

177 178

3. RESULTS AND DISCUSSION

179

3.1 Synthesis and Characterizations of the P[MPC-co-AEMA] Random Copolymer

180

As shown in Figure S1, the zwitterionic random P[MPC-co-AEMA] copolymer was prepared

181

through

182

phosphorylcholine (MPC) and 2-aminoethyl methacrylate hydrochloride (AEMA) in the

183

presence of 2,2'-azobis(2-methylpropionitrile) (AIBN) as the radical initiator. The synthesized

184

P[MPC-co-AEMA] copolymer has a number average molecular weight (Mn) and a weight

185

average molecular weight (Mw) of 41308 Da and 57034 Da, respectively, as summarized in

186

Table S3. It has a polydispersity index (PDI) of 1.38, suggesting that the copolymer has a

187

narrowly dispersed molecule weight distribution. The FTIR spectra confirm the chemistry of the

188

P[MPC-co-AEMA] copolymer. As displayed in Figure 2 (a), the strong peak around 1728 cm−1

189

is attributed to the C=O of the ester group, while the peaks at 1240, 1089 cm-1 and 970 cm−1 are

the

single-step

free-radical

polymerization

between

2-methacryloyloxyethyl



Page 10 of 28

190

from the −POCH2− and −N+(CH3) functional groups, respectively, suggesting the presence of the

191

poly(MPC) segments. The absorptions at 1621 cm-1 and 700 cm-1 are due to the -NH2 groups

192

from the poly(AEMA) segments.46 Because of the zwitterionic chemistry of the P[MPC-co-

193

AEMA] copolymer, it shows excellent solubility in water and IPA.

194

195 196

Figure 2. (a) FTIR spectra of the P[MPC-co-AEMA] copolymer, (b) XPS wide scan spectra of

197

the outer surfaces of PES&Torlon and PES&Torlon-P[MPC-co-AEMA] hollow fiber

198

membranes, and (c) an enlarged P2p peak of the XPS spectra.

199 200

3.2. Characterizations of the Pristine and Copolymer Modified PES&Torlon Hollow Fiber

201

Substrates

202

The PES&Torlon hollow fiber membrane was designed and fabricated as the substrate for the

203

thin-film composite (TFC) PRO membrane. The incorporation of a certain amount of Torlon into

204

the PES polymer aims to improve the reactivity of the fiber because the polyimide groups of

205

Torlon can react with the amino groups (-NH2) of the poly(AEMA) segments of the copolymer.

206

As a result, a zwitterionic functional layer can be covalently grafted onto the PES&Torlon fiber

207

without using a transition layer.


Page 11 of 28


208

209 210

Figure 3. FESEM images of (a-e) the as-spun PES&Torlon hollow fiber substrate and (f) the

211

P[MPC-co-AEMA] copolymer modified outer surface.

212 213

Figure 3 (a-e) shows the membrane morphology of the as-spun PES&Torlon hollow fiber

214

substrate. It has a highly concentric structure with an inner diameter of 680 µm and a cross-

215

section thickness of 160 µm. Due to the fast phase inversion behavior induced by the water bore

216

fluid, the fiber possesses a smooth and dense inner surface without any large pores. This surface

217

characteristic is critical for interfacial polymerization to form a stable and less defective

218

polyamide rejection layer on top of the substrate.47 The strong water bore fluid also induces a

219

highly porous cross-section structure with a fully open cell microstructure and some finger-like

220

macrovoids, which facilitate water and salt transportation across the membrane. In order to

221

further reduce transporting resistance and internal concentration polarization (ICP), an open

222

porous outer surface is fabricated by employing a dope-solvent co-extrusion technology during



Page 12 of 28

223

which a pure NMP solvent is fed at the outer channel of the tri-orifice spinneret.41,44 Table 1

224

summarizes the mean pore diameter (µp), pure water permeability (PWP), molecular weight cut-

225

off (MWCO) and water contact angle at the outer surface of the PES&Torlon hollow fiber, while

226

Figure S2 shows its pore size distribution. It has a relatively narrow pore size distribution with a

227

mean pore diameter of 17.8 nm. As a result, a high PWP of 126.8 L m-2 h-1 bar-1 and a MWCO of

228

292.5 KDa are obtained. Since both PES and Torlon polymers are relatively hydrophobic, the

229

PES&Torlon fiber has a water contact angle of 70.9°.

230 231

Table 1. Summary of the mean effective pore size (µp), PWP, MWCO, outer surface water

232

contact angle and dimension of the hollow fiber substrates Fiber ID

µp

σp

MWCO

Contact

OD/ID

[L/(m ·bar·h)]

(KDa)

angle (°)

(µm)

2

(nm)

233

PWP

PES&Torlon

17.8

1.67

126.8

292.5

70.9

1000/680

PES&TorlonP[MPC-co-AEMA]

16.6

1.55

185.4

220.0

40.0

1000/680

The tests were performed from the outer surface to the inner one of hollow fibers at 1 bar.

234 235

After grafting the P[MPC-co-AEMA] copolymer, a uniform coating layer is formed on the

236

outer surface of the modified PES&Torlon-P[MPC-co-AEMA] hollow fiber, as depicted in

237

Figure 3 (f). Compared to the pristine PES&Torlon fiber, the outer surface also becomes

238

relatively smooth and dense. The XPS data presented in Figure 2 (b) and Table S4 show the

239

unique P2s and P2p peaks of the P element on the modified outer surface. The significant

240

increases in N and O element concentrations and the decreases in C and S element contents

241

further confirm the successful graft of the P[MPC-co-AEMA] copolymer. Since a more 12 ACS Paragon Plus Environment

Page 13 of 28


242

hydrophilic zwitterionic copolymer layer is introduced onto the outer surface, the membrane

243

hydrophilicity is dramatically improved. As a result, the water contact angle drops from 70.9° to

244

40° after the modification (Table 1). The P[MPC-co-AEMA] coating also induces a slight

245

reduction in membrane pore size. As shown in Figure S2, the mean pore diameter drops to 16.6

246

nm and the pore size distribution becomes narrower. Consequently, the MWCO of PES&Torlon-

247

P[MPC-co-AEMA] fiber decreases to 220 KDa. Nevertheless, the PWP value increases to 185.4

248

L m-2 h-1 bar-1 because of the enhanced hydrophilicity and the lowered transport resistance. In

249

summary, the P[MPC-co-AEMA] modification significantly enhances the membrane surface

250

hydrophilicity, reduces the transport resistance, slightly lowers the membrane pore size and

251

narrows the pore size distribution.

252 253

3.3. Characterizations of TFC Hollow Fiber Membranes

254

Figure 4 shows the morphology of the polyamide dense-selective layer of the TFC hollow fiber

255

membranes that was synthesized on the inner surface of the hollow fiber substrate via interfacial

256

polymerization. The polyamide thin film has a typical “ridge-and-valley” morphology with a low

257

thickness of around 180 nm. This thin and rough structure is crucial for the TFC membrane to

258

achieve high water permeation and high power density.

259



Page 14 of 28

260 261

Figure 4. FESEM images of the TFC-PES&Torlon-P[MPC-co-AEMA] hollow fiber membrane

262

with an inner polyamide selective skin and a copolymer coating layer on the outer surface.

263 264

Table 2 summarizes the intrinsic properties of the newly developed TFC hollow fiber

265

membranes in terms of pure water permeability (A), salt permeability coefficient (B), and

266

membrane structural parameter (S). The TFC-PES&Torlon and TFC-PES&Torlon-P[MPC-co-

267

AEMA] membranes possess very similar A values of 1.1 and 1.2 L m-2 h-1 bar-1 with ultralow B

268

values of 0.020 and 0.023 L m-2 h-1, respectively. The slightly improved water permeability of

269

the PES&Torlon-P[MPC-co-AEMA] membrane is mainly resulted from the enhanced substrate

270

hydrophilicity. Since the two TFC membranes have the same polyamide selective layer, the

271

improvement is not significant. In addition, the structural parameters of TFC-PES&Torlon and

272

TFC-PES&Torlon-P[MPC-co-AEMA] membranes are 1147 µm and 1731 µm, respectively. The

273

slightly increased structural parameter of the latter is likely due to the smaller pore size and

274

narrower pore size distribution of the substrate fiber as shown in Table 1 and Figure S2.

275 14 ACS Paragon Plus Environment

Page 15 of 28

276


Table 2. Transport properties and structural parameters of the TFC hollow fiber membranes Membrane ID

Water permeability, A 2

[L/(m ·bar·h)] at 1 bar

Salt permeability, B −2

−1

(L m h )

Km 5

S −1

(×10 s m )

(µm)

TFC-PES&Torlon

1.1

0.020

7.7

1147

TFC-PES&Torlon-

1.2

0.023

11.7

1731

P[MPC-co-AEMA] 277 278

The mechanical stability of the two TFC hollow fiber membranes was assessed by measuring

279

their PWP values and NaCl rejections at various hydraulic pressures. As displayed in Figure S3,

280

both fibers can withstand a high pressure of larger than 20 bar. Compared to the PWP measured

281

at 1 bar, the PWP values of both fibers slightly increase with an increase in operating pressure.

282

The small increases in PWP at high pressures are possibly attributed to the effects of high hoop

283

stresses on the inner polyamide dense-selective layer. With the rise in pressure, the rejections to

284

NaCl of the two fibers continuously increase and reach 98-99% at 20 bar, suggesting superior

285

membrane strength and stability. These transport characteristics and mechanical strength make

286

the two TFC hollow fibers suitable for PRO applications.48

287 288

3.4 Implications of the TFC Hollow Fiber Membranes for Osmotic Power Generation

289

The newly developed TFC hollow fiber membranes were evaluated for osmotic power

290

generation using deionized water and a real wastewater brine as the feeds. Figure 5 compares

291

their water fluxes (Jw) and power density (W) as a function of operating pressure (∆P) using a 1.0

292

M NaCl synthetic brine as the draw solution. The TFC-PES&Torlon and TFC-PES&Torlon-

293

P[MPC-co-AEMA] membranes have Jw values of 32.0 and 22.5 LMH at ∆P=0 bar, respectively,



Page 16 of 28

294

when using deionized water as the feed (Figure 5 (a)). The difference in Jw is mainly due to their

295

variation in membrane structural parameter (S) (Table 2). With an increase in ∆P, Jw shows a

296

rapid decrease because of the lowered driving force and membrane compaction. Since the

297

membrane power density (W) is the product of Jw and ∆P, W rapidly increases with an increase

298

in ∆P and reaches the highest values of 13.5 and 10.2 W/m2 at ∆P=20 bar for the TFC-

299

PES&Torlon and TFC-PES&Torlon-P[MPC-co-AEMA] membranes, respectively (Figure 5 (b)).

300

301 302

Figure 5. Water flux (Jw) and power density (W) of the TFC hollow fiber membranes as a

303

function of operating pressure using (a, b) deionized water and (c, d) real wastewater brine as

304

feed solutions. (Draw solution was 1 M NaCl solution and the testing duration at each pressure

305

was 30 min).

306 16 ACS Paragon Plus Environment

Page 17 of 28


307

However, when the feed is changed from deionized water to the real wastewater brine, both

308

membranes exhibit dramatic decreases in Jw and W and the drop is more significant for the TFC-

309

PES&Torlon membrane. As depicted in Figure 5 (c) and (d), Jw of the TFC-PES&Torlon

310

membrane at ∆P=0 bar drops from 32.0 to 20.5 LMH, while its W at ∆P=20 bar decreases from

311

13.5 to 6.8 W/m2, implying that fouling caused by the wastewater brine is significant and

312

happens immediately with the permeation of water. In contrary, the TFC-PES&Torlon-P[MPC-

313

co-AEM] membrane displays mild declines in Jw and W when changing the feed to the

314

wastewater brine. It achieves a W of 8.6 W/m2 at ∆P=20 bar which is even higher than that of the

315

TFC-PES&Torlon membrane. This demonstrates the good antifouling properties of the

316

zwitterionic copolymer modified TFC-PES&Torlon-P[MPC-co-AEM] membrane.

317 318

3.5 Antifouling Performance of TFC-PES&Torlon-P[MPC-co-AEMA] Membranes

319

The PRO fouling behaviors of TFC-PES&Torlon and TFC-PES&Torlon-P[MPC-co-AEMA]

320

membranes were further assessed by conducting long-term tests. The initial water fluxes of the

321

two membranes were controlled to be almost the same by adjusting the draw solution

322

concentration. Figure 6 (a, b) shows the variations of the normalized Jw of the two TFC

323

membranes as a function of testing duration at ∆P=0 bar using deionized water and real

324

wastewater brine as the feeds, respectively. Both membranes display very mild flux declines of

325

less than 10% even after 24-h tests when using deionized water as the feed. Since the draw

326

solution concentration was maintained throughout the tests, the flux decrease is mainly resulted

327

from the increased feed salinity and ICP caused by the reverse salt diffusion.

328



Page 18 of 28

329 330

Figure 6. Normalized water fluxes and power density of TFC-PES&Torlon and TFC-

331

PES&Torlon-P[MPC-co-AEMA] hollow fiber membranes as a function of time at (a, b) ∆P=0

332

and (c, d) ∆P=15 bar using deionized water and real wastewater brine as feeds, respectively.

333

(The initial water fluxes for the two fibers at each pressure were maintained almost the same by

334

adjusting draw solution concentration).

335 336

However, when the real wastewater brine was utilized as the feed, the TFC-PES&Torlon

337

membrane shows a fast and dramatic flux decline in a very short duration. Its Jw decreases to 80%

338

of the initial value during the first 100 min, suggesting that fouling and its induced ICP play

339

primary roles in flux drop in the early stage of PRO tests as they happen very fast. In contrary, Jw

340

of the TFC-PES&Torlon-P[MPC-co-AEMA] membrane remains a high value and the flux

341

decline during such a short period is less than 7%. This indicates that fouling is effectively 18 ACS Paragon Plus Environment

Page 19 of 28


342

retarded by the zwitterionic copolymer coating. With a further increase in testing duration, both

343

Jw values of TFC-PES&Torlon and TFC-PES&Torlon-P[MPC-co-AEMA] membranes rapidly

344

decrease to 52% and 74% of the initial fluxes when the testing duration reaches 400 min, and

345

then slightly drop to 39% and 63% at a high duration of 1440 min, respectively. In other words,

346

the TFC-PES&Torlon-P[MPC-co-AEMA] membrane shows a much lower flux drop of 37%,

347

compared to 61% of the TFC-PES&Torlon membrane during 24-h tests.

348

A similar phenomenon was observed for both membranes at a ∆P=15 bar, but the decreases of

349

Jw and W are less than those at ∆P=0 bar, mainly because of the lower initial water flux at a

350

higher pressure. As presented in Figure 6 (c, d), the TFC-PES&Torlon-P[MPC-co-AEMA]

351

membrane displays slower and less Jw and W declines than the TFC-PES&Torlon membrane

352

particularly at a relatively short testing duration. After a 12-h operation, a small drop of 28% in

353

Jw and W is achieved by the TFC-PES&Torlon-P[MPC-co-AEMA] membrane; however, Jw and

354

W of the TFC-PES&Torlon membrane decrease to 58% of the initial values. Again, these data

355

confirm the superior fouling resistance provided by the zwitterionic copolymer coating to the

356

real wastewater brine even at high operating pressures for PRO.

357

In order to understand the fouling behavior of the real wastewater brine and the antifouling

358

properties of the newly developed membranes, a series of fouling tests were performed by using

359

synthetic solutions that contain various types of foulants with the same or higher concentrations

360

than those in wastewater brine as the feeds (Table S2). As displayed in Figure S4, the flux

361

declines induced by alginate organic fouling, silica colloidal fouling and CaSO4 scaling are mild,

362

while the CaHPO4 scaling leads to a fast and dramatic flux drop. Clearly, the latter induces the

363

predominated fouling in previous PRO tests when using wastewater brine as the feed. Consistent

364

with the performance illustrated in Figure 6, the TFC-PES&Torlon-P[MPC-co-AEMA]



Page 20 of 28

365

membrane shows superior fouling resistance not only to organic foulants but also to inorganic

366

scaling. Three reasons may be involved for this superior antifouling property. Firstly, a strong

367

hydration layer is formed above the P[MPC-co-AEMA] coating layer due to the excellent

368

hydrophilicity of the zwitterionic brush. Secondly, the electrostatic interactions between the

369

membrane surface and the foulants in the feed would be reduced by the net electrical neutrality

370

of the zwitterions.37,49 Lastly, the phosphonate functional groups of the zwitterions might act as

371

powerful chelating agents that form stable complexes with free Ca2+ ions and thus suppress

372

inorganic scaling.50

373 374

3.6 Multiple Fouling and Cleaning Cycles in PRO

375

Multiple PRO performances of the TFC-PES&Torlon-P[MPC-co-AEMA] membrane were

376

evaluated at ∆P=15 bar using the wastewater brine as the feed. Each cycle was performed for

377

720 min (or 12 h) and the fouled membrane was regenerated by physical backwashing of

378

freshwater from the lumen to the shell side of the module at 15 bar for 30 min. As presented in

379

Figure 7, the TFC-PES&Torlon-P[MPC-co-AEMA] membrane shows quite stable PRO

380

performance for the second and third cycles. During the three PRO cycles of 2160 min in total,

381

the decreases of Jw and W are less than 30% of the initial values, much lower than those of the

382

pristine TFC-PES&Torlon membrane. In addition, the fouled TFC-PES&Torlon-P[MPC-co-

383

AEMA] membrane can be easily regenerated by water backwashing. The initial water flux can

384

be almost fully restored back to its original value with a great recovery of 95-98%, surpassing

385

that of 77-82% for the unmodified membrane. It is believed that the hydraulic pressure induced

386

cross flow during the backwash could help carry the accumulated salts and foulants out of the

387

substrate while the hydrophilic copolymer coating makes this process easy and efficient. Given


Page 21 of 28


388

the ease of membrane fabrication and the promising antifouling performance, this work may

389

provide valuable insights for the design and fabrication of effective fouling-resistant materials

390

and membranes for PRO osmotic power generation.

391

392 393

Figure 7. Normalized water fluxes and power density vs. time of the TFC hollow fiber

394

membranes at ∆P=15 bar in PRO performance cycles using real wastewater brine as the feed.

395

(Membrane cleaning was performed by physical backwashing of freshwater at 15 bar for 0.5 h.

396

The numbers in the figure shows the flux recoveries after cleaning.).

397 398

Supporting Information

399

Materials and chemicals; membrane characterizations; synthetic route for the P[MPC-co-AEMA]

400

random copolymer (Figure S1); hollow fiber spinning conditions (Table S1); compositions of the

401

synthetic feed solutions (Table S2); GPC results (Table S3); pore size distribution of the hollow

402

fiber substrates (Figure S2) and XPS data (Table S4) of the hollow fibers; mechanical stability of 21 ACS Paragon Plus Environment


Page 22 of 28

403

the TFC hollow fiber membranes (Figure S3); long-term PRO performance at ∆P=0 bar using

404

synthetic feed solutions (Figure S4); and PRO operating conditions and performance evaluation.

405 406

ACKNOWLEDGMENTS

407

This work is granted by the Singapore National Research Foundation under its Environmental

408

&Water Research Programme and administered by PUB, Singapore’s national water agency. It is

409

funded under the projects entitled "Membrane Development for Osmotic Power Generation, Part

410

1. Materials Development and Membrane Fabrication" (1102-IRIS-11-01) and NUS Grant no. R-

411

279-000-381-279; "Membrane Development for Osmotic Power Generation, Part 2. Module

412

Fabrication and System Integration" (1102-IRIS-11-02) and NUS Grant no. R-279-000-382-279.

413

The authors would also like to thank the undergraduate students Miss Yuyan Wang and Mr.

414

Shuai Jin for all their kind help on experiments work.

415 416

REFERENCES

417

(1) Loeb, S.; Norman, R. S. Osmotic power plant. Science 1975, 189, 654–655.

418

(2) Thorsen, T.; Holt, T. The potential for power production from salinity gradients by pressure

419 420 421 422 423

retarded osmosis. J. Membr. Sci. 2009, 335, 103–110. (3) Achilli, A.; Cath, T. Y.; Childress, A. E. Power generation with pressure retarded osmosis: an experimental and theoretical investigation. J. Membr. Sci. 2009, 343, 42–52. (4) Ramon, G. Z.; Feinberg, B. J.; Hoek, E. M. V. Membrane-based production of salinity gradient power. Energy Environ. Sci. 2011, 4, 4423–4434.


Page 23 of 28


424

(5) Chung, T. S.; Li, X.; Ong, R. C.; Ge, Q.; Wang, H.; Han, G. Emerging forward osmosis (FO)

425

technologies and challenges ahead for clean water and clean energy applications. Curr. Opin.

426

Chem. Eng. 2012, 1, 246–257.

427 428 429 430 431 432

(6) Logan, B. E.; Elimelech, M. Membrane-based processes for sustainable power generation using water. Nature 2012, 488, 313–319. (7) Han, G.; Zhang, S.; Li, X.; Chung, T. S. Progress in pressure retarded osmosis (PRO) membranes for osmotic power generation. Prog. Polym. Sci. 2015, 51, 1–27. (8) Alsvik, I. L.; Hägg, M. B. Pressure retarded osmosis and forward osmosis membranes: materials and methods. Polymers 2013, 5, 303–327.

433

(9) Skilhagen, S. E.; Dugstad, J. E.; Aaberg, R. J. Osmotic power—power production based on

434

the osmotic pressure difference between waters with varying salt gradients. Desalination

435

2008, 220, 476–482.

436

(10)

Saito, K.; Irie, M.; Zaitsu, S.; Sakai, H.; Hayashi, H.; Tanioka, A. Power generation with

437

salinity gradient by pressure retarded osmosis using concentrated brine from SWRO system

438

and treated sewage as pure water. Desalination Water Treat. 2012, 41, 114–21.

439

(11)

Kim, J.; Park, M.; Snyder, S. A.; Kim, J. H. Reverse osmosis (RO) and pressure retarded

440

osmosis (PRO) hybrid processes: model-based scenario study. Desalination 2013, 322, 121–

441

30.

442

(12)

Achilli, A.; Prante, J. L.; Hancock, N. T.; Maxwell, E. B.; Childress, A. E. Experimental

443

results from RO−PRO: a next generation system for low-energy desalination. Environ. Sci.

444

Technol. 2014, 48, 6437–43.



445

(13)

Page 24 of 28

Han, G.; Zuo, J.; Wan, C. F.; Chung, T. S. Hybrid pressure retarded osmosis−membrane

446

distillation (PRO−MD) process for osmotic power and clean water generation. Environ. Sci.:

447

Water Res. Technol. 2015, 1, 507–515.

448

(14)

Yip, N. Y.; Brogioli, D.; Hamelers, H. V. M.; Nijmeijer, K. Salinity gradients for

449

sustainable energy: primer, progress, and prospects. Environ. Sci. Technol. 2016, 50,

450

12072−12094.

451 452 453 454 455

(15)

Wan, C. F.; Chung, T. S. Energy recovery by pressure retarded osmosis (PRO) in

SWRO–PRO integrated processes. Appl. Energy 2016, 162, 687–698. (16)

Sakai, H.; Ueyama, T.; Irie, M.; Matsuyama, K.; Tanioka, A.; Saito, K.; Kumano, A.

Energy recovery by PRO in sea water desalination plant. Desalination 2016, 389, 52–57. (17)

Klaysom, C.; Cath, T. Y.; Depuydt, T.; Vankelecom, I. F. J. Forward and pressure

456

retarded osmosis: potential solutions for global challenges in energy and water supply. Chem.

457

Soc. Rev. 2013, 42, 6959–6989.

458

(18)

Liu, Y.; Mi, B. Combined fouling of forward osmosis membranes: Synergistic foulant

459

interaction and direct observation of fouling layer formation. J. Membr. Sci. 2012, 407–408,

460

136–144.

461 462 463 464 465

(19)

Thelin, W. R.; Sivertsen, E.; Holt, T.; Brekke, G. Natural organic matter fouling in

pressure retarded osmosis. J. Membr. Sci. 2013, 438, 46–56. (20)

Zhang, M.; Hou, D.; She, Q.; Tang, C. Y. Gypsum scaling in pressure retarded osmosis:

experiments, mechanisms and implications. Water Res. 2014, 48, 387–395. (21)

Kim, Y.; Elimelech, M.; Shon, H. K.; Hong, S. Combined organic and colloidal fouling

466

in forward osmosis: fouling reversibility and the role of applied pressure. J. Membr. Sci.

467

2014, 460, 206–212.


Page 25 of 28

468


(22)

Chen, S. C.; Amy, G. L.; Chung, T. S. Membrane fouling and anti-fouling strategies

469

using RO retentate from a municipal water recycling plant as the feed for osmotic power

470

generation. Water Res. 2016, 88, 144–155.

471

(23)

Kim, D. I.; Kim, J.; Shon, H. K.; Hong, S. Pressure retarded osmosis (PRO) for

472

integrating seawater desalination and wastewater reclamation: energy consumption and

473

fouling. J. Member. Sci. 2015, 483, 34–41.

474

(24)

Han, G.; Cheng, Z. L.; Chung, T. S. Thin-film composite (TFC) hollow fiber membrane

475

with double-polyamide active layers for internal concentration polarization and fouling

476

mitigation in osmotic processes. J. Membr. Sci. 2017, 523, 497–504.

477

(25)

Yip, N. Y.; Tiraferri, A.; Phillip, W. A.; Schiffman, J. D.; Hoover, L. A.; Kim, Y. C.;

478

Elimelech, M. Thin-film composite pressure retarded osmosis membranes for sustainable

479

power generation from salinity gradients. Environ. Sci. Technol. 2011, 45, 4360–4369.

480

(26)

Sun, S. P.; Chung, T. S. Outer-selective pressure-retarded osmosis hollow fiber

481

membranes from vacuum-assisted interfacial polymerization for osmotic power generation.

482

Environ. Sci. Technol. 2013, 47, 13167–13174.

483

(27)

Chou, S.; Wang, R.; Shi, L.; She, Q.; Tang, C.; Fane, A. G. Thin-film composite hollow

484

fiber membranes for pressure retarded osmosis (PRO) process with high power density. J.

485

Membr. Sci. 2012, 389, 25–33.

486 487 488 489

(28)

Bui, N. N.; McCutcheon, J. R. Nanofiber supported thin-film composite membrane for

pressure-retarded osmosis. Environ. Sci. Technol. 2014, 48, 4129–4136. (29)

Zhang, S.; Sukitpaneenit, P.; Chung, T. S. Design of robust hollow fiber membranes with

high power density for osmotic energy production. Chem. Eng. J. 2014, 241, 457–465.



490

(30)

Page 26 of 28

Ingole, P. G.; Choi, W.; Kim, K. H.; Park, C. H.; Choi, W. K.; Lee, H. K. Synthesis,

491

characterization and surface modification of PES hollow fiber membrane support with

492

polydopamine and thin film composite for energy generation. Chem. Eng. J. 2014, 43, 137–

493

146.

494

(31)

Li, X.; Cai, T.; Chung, T. S. Anti-fouling behavior of hyper-branched polyglycerol

495

grafted polyethersulfone hollow fiber membranes for osmotic power generation. Environ. Sci.

496

Technol. 2014, 48, 9898–9907.

497

(32)

Hu, M.; Zheng, S.; Mi, B. Organic fouling of graphene oxide membranes and its

498

implications for membrane fouling control in engineered osmosis. Environ. Sci. Technol.

499

2016, 50, 685–693.

500

(33)

Xiong, J. Y.; Cheng, Z. L.; Wan, C. F.; Chen, S. C.; Chung, T. S. Analysis of flux

501

reduction behaviors of PRO hollow fiber membranes: experiments, mechanisms, and

502

implications. J. Membr. Sci. 2016, 505, 1–14.

503

(34)

Vaisocherová, H.; Yang, W.; Zhang, Z.; Cao, Z.; Cheng, G.; Piliarik, M.; Homola, J.;

504

Jiang, S. Ultralow fouling and functionalizable surface chemistry based on a zwitterionic

505

polymer enabling sensitive and specific protein detection in undiluted blood plasma. Anal.

506

Chem. 2008, 80, 7894–7901.

507

(35)

Yu, H. Y.; Kang, Y.; Liu, Y.; Mi, B. Grafting polyzwitterions onto polyamide by click

508

chemistry and nucleophilic substitution on nitrogen: a novel approach to enhance membrane

509

fouling resistance. J. Membr. Sci. 2014, 449, 50–57.

510 511

(36)

Zhao, Y. H.; Wee, K. H.; Bai, R. Highly hydrophilic and low-protein-fouling

polypropylene membrane prepared by surface modification with sulfobetaine-based


Page 27 of 28


512

zwitterionic polymer through a combined surface polymerization method. J. Membr. Sci.

513

2010, 362, 326–333.

514

(37)

Cai, T.; Li, X.; Wan, C.; Chung, T. S. Zwitterionic polymers grafted poly(ether sulfone)

515

hollow fiber membranes and their antifouling behaviors for osmotic power generation. J.

516

Membr. Sci. 2016, 497, 142–152.

517

(38)

Le, N. L.; Quilitzsch, M.; Cheng, H.; Hong, P. Y.; Ulbricht, M.; Nunes, S. P.; Chung, T.

518

S. Hollow fiber membrane lumen modified by polyzwitterionic grafting. J. Membr. Sci. 2017,

519

522, 1–11.

520

(39)

Chen, L.; Tan, L.; Liu, S.; Bai, L.; Wang, Y. Surface modification by grafting of

521

poly(SBMA-co-AEMA)-g-PDA coating and its application in CE. J. Biomater. Sci. Polym.

522

Ed. 2014, 25, 766–785.

523

(40)

Xia, Q. C.; Wang, J.; Wang, X.; Chen, B. Z.; Guo, J. L.; Jia, T. Z.; Sun, S. P. A

524

hydrophilicity gradient control mechanism for fabricating delamination-free dual-layer

525

membranes. J. Memb. Sci. 2017, 539, 392–402.

526

(41)

Han, G.; Wang, P.; Chung, T. S. Highly robust thin-film composite pressure retarded

527

osmosis (PRO) hollow fiber membranes with high power densities for renewable salinity-

528

gradient energy generation. Environ. Sci. Technol. 2013, 47, 8070–8077.

529 530 531 532 533 534

(42)

Van der Bruggen, B.; Vandecasteele, C. Modelling of the retention of uncharged

molecules with nanofiltration. Water Res. 2002, 36, 1360–1368. (43)

Aimar, P.; Meireles, M.; Sanchez, V. A contribution to the translation of retention curves

into pore size distributions for sieving membranes. J. Membr. Sci. 1990, 54, 321–338. (44)

Han, G.; Chung, T. S. Robust and high performance pressure retarded osmosis hollow

fiber membranes for osmotic power generation. AIChE J. 2014, 60, 1107–1119.



535

(45)

Page 28 of 28

Wan, C. F.; Chung, T. S. Osmotic power generation by pressure retarded osmosis using

536

seawater brine as the draw solution and wastewater retentate as the feed. J. Membr. Sci. 2015,

537

479, 148–158.

538

(46)

Goda, T.; Konno, T.; Takai, M.; Ishihara, K. Photoinduced phospholipid polymer

539

grafting on Parylene film: Advanced lubrication and antibiofouling properties. Colloids Surf

540

B Biointerfaces 2007, 54, 67–73.

541

(47)

Ghosh, A. K.; Hoek, E. M. V. Impacts of support membrane structure and chemistry on

542

polyamide–polysulfone interfacial composite membranes. J. Membr. Sci. 1990, 336, 140–

543

148.

544

(48)

Touati, K.; Tadeo, F. Study of the Reverse Salt Diffusion in pressure retarded osmosis:

545

Influence on concentration polarization and effect of the operating conditions. Desalination

546

2016, 389, 171–186.

547

(49)

Zhao, D.; Qiu, G.; Li, X.; Wan, C.; Lu, K. Tai-Shung Chung, Zwitterions coated hollow

548

fiber membranes with enhanced antifouling properties for osmotic power generation from

549

municipal wastewater. Water Res. 2016, 104, 389–396.

550

(50)

Qin, J. J.; Wai, M. N.; Oo, M. H.; Kekre, K. A.; Seah, H. Impact of anti-scalant on

551

fouling of reverse osmosis membranes in reclamation of secondary effluent. Water Sci.

552

Technol. 2009, 60, 2767–2774.

553


Advanced Anti-Fouling Membranes for Osmotic Power Generation

Recommend Documents