Omniphobic Polyvinylidene Fluoride (PVDF) Membrane for

Oct 20, 2016 - Superhydrophobic electrospun nanofiber membrane coated by carbon nanotubes network for membrane distillation. Kang-Kang Yan , Lei .... ...
0 downloads 12 Views 5MB Size
Subscriber access provided by Univ. of Tennessee Libraries

Article

Omniphobic Polyvinylidene Fluoride (PVDF) Membrane for Desalination of Shale Gas Produced Water by Membrane Distillation Chanhee Boo, Jongho Lee, and Menachem Elimelech Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b03882 • Publication Date (Web): 20 Oct 2016 Downloaded from http://pubs.acs.org on October 28, 2016

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 free 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 accessible to all readers and 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.

Environmental Science & Technology 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 25

Environmental Science & Technology

Omniphobic Polyvinylidene Fluoride (PVDF) Membrane for Desalination of Shale Gas Produced Water by Membrane Distillation

Chanhee Boo, Jongho Lee, and Menachem Elimelech*

Department of Chemical and Environmental Engineering, Yale University, New Haven, Connecticut 06520-8286, United States

* Corresponding author: email: [email protected]; Tel. +1 (203) 432-2789

1 ACS Paragon Plus Environment

Environmental Science & Technology

1

ABSTRACT

2

Microporous membranes fabricated from hydrophobic polymers such as polyvinylidene fluoride

3

(PVDF) have been widely used for membrane distillation (MD). However, hydrophobic MD

4

membranes are prone to wetting by low surface tension substances, thereby limiting their use in

5

treating challenging industrial wastewaters, such as shale gas produced water. In this study, we

6

present a facile and scalable approach for the fabrication of omniphobic polyvinylidene fluoride

7

(PVDF) membranes that repel both water and oil. Positive surface charge was imparted to an

8

alkaline-treated PVDF membrane by aminosilane functionalization, which enabled irreversible

9

binding of negatively charged silica nanoparticles (SiNPs) to the membrane through electrostatic

10

attraction. The membrane with grafted SiNPs was then coated with fluoroalkylsilane

11

(perfluorodecyltrichlorosilane) to lower the membrane surface energy. Results from contact

12

angle measurements with mineral oil and surfactant solution demonstrated that overlaying SiNPs

13

with ultralow surface energy significantly enhanced the wetting resistance of the membrane

14

against low surface tension liquids. We also evaluated desalination performance of the modified

15

membrane in direct contact membrane distillation with a synthetic wastewater containing

16

surfactant (sodium dodecyl sulfate) and mineral oil, as well as with shale gas produced water.

17

The omniphobic membrane exhibited a stable MD performance, demonstrating its potential

18

application for desalination of challenging industrial wastewaters containing diverse low surface

19

tension contaminants.

20 21

TOC Art

22 23

2 ACS Paragon Plus Environment

Page 2 of 25

Page 3 of 25

Environmental Science & Technology

INTRODUCTION

24 25 26

Natural gas from unconventional shale plays has the potential to secure energy supply and reduce

27

carbon dioxide emissions by minimizing the dependence on fossil fuel energy production.1

28

Breakthroughs in horizontal drilling and hydraulic fracturing technologies have allowed access

29

to vast natural gas resources from shale formation.2 However, environmental impacts of shale

30

gas production, including wastewater management and potential groundwater contamination,

31

remain major challenges for efficient exploitation of this emerging energy resource.3, 4

32

The Environmental Protection Agency (EPA) estimates 25,000−30,000 new shale gas wells

33

were developed annually in the United States between 2011 and 2014, with two to four million

34

gallons of water used for drilling and hydraulic fracturing of each well.5, 6 Shale gas produced

35

water comprises water trapped in underground formation and flowback from hydraulic

36

fracturing.7 This waste stream is characterized by high total dissolved solids (TDS)

37

concentrations and hydrocarbon content (i.e., oil and grease) due to the nature of formation water

38

and high levels of chemical additives used in hydraulic fracturing.8 Shale gas produced water is

39

challenging to treat by conventional water treatment practices due to its high salinity and

40

complex physicochemical composition.6

41

Membrane distillation (MD) is an emerging technology that shows promise for efficient

42

desalination of high salinity industrial wastewaters.9, 10 MD desalination is driven by a vapor

43

pressure gradient between a hot feed stream and a cold permeate (distillate) stream. Performance

44

of this thermal-based separation process is only slightly sensitive to the feed salinity, rendering

45

MD suitable for desalination of highly saline shale gas produced water.6, 11 The temperature of

46

shale gas produced waters, which originates from geothermal heat sources at well sites, can be as

47

high as 100 ºC.6 Hence, the elevated temperature of produced water could be exploited to drive

48

MD desalination.

49

Despite the advantages of applying MD to produced water desalination, potential wetting of

50

the MD membrane remains a practical shortcoming that hinders its implementation. Shale gas

51

produced water contains high levels of low surface tension contaminants, such as oil and grease,

52

organic solvents, and surfactants.12 These low surface tension substances can potentially wet

53

conventional hydrophobic MD membranes, resulting in failure of the MD desalination process.

3 ACS Paragon Plus Environment

Environmental Science & Technology

54

Omniphobic membranes resisting wetting to both water and low surface tension liquids hold

55

promise for MD application in treatment of shale gas produced water.13, 14 Omniphobicity can be

56

realized by constructing surfaces with ultralow interfacial energy and re-entrant structures that

57

together allow a metastable Cassie-Baxter state for the liquid−solid−vapor interface.15 Numerous

58

approaches have been attempted to create surfaces that feature a re-entrant structure using

59

electrochemical16 and plasma etching,17 anodization,18 and photolithography19, and to achieve

60

ultralow surface energy by silanization19 and coating20, 21 of fluoroalkyl molecules. However,

61

these techniques have mostly focused on fabricating omniphobic surfaces for impermeable

62

prototype substrates (e.g., silicon wafer)22 or fabric materials.23

63

Microporous polyvinylidene fluoride (PVDF) membranes have been widely used for MD

64

applications because of their hydrophobic nature, excellent chemical compatibility, and physical

65

robustness.24 The PVDF membrane structure properties that are ideal for MD, such as high

66

porosity and sub-micrometer pore size, can be easily controlled by the well-established phase

67

separation fabrication technique.25 However, conventional hydrophobic PVDF membranes are

68

prone to wetting by low surface tension substances. Furthermore, surface engineering of PVDF

69

membranes to achieve desired chemical and physical properties is challenging because PVDF is

70

chemically inert and possesses no surface groups.

71

This study presents a scalable approach to fabricate omniphobic membranes for desalination

72

of shale gas produced water by membrane distillation. Specifically, we demonstrate surface

73

modification of a conventional hydrophobic PVDF microporous substrate to produce an

74

omniphobic membrane that exhibits surface wetting resistance to low surface tension substances.

75

Desalination performance of the omniphobic membrane was evaluated in direct contact

76

membrane distillation with synthetic wastewaters containing low surface tension contaminants,

77

including surfactants and mineral oil. Produced water from shale gas industry was further

78

employed to evaluate MD performance of the omniphobic membrane. Implications of the results

79

for the use omniphobic membranes in MD to treat challenging industrial wastewaters are

80

evaluated and discussed.

81 82

MATERIALS AND METHODS

4 ACS Paragon Plus Environment

Page 4 of 25

Page 5 of 25

Environmental Science & Technology

83

Materials and Chemicals. ACS grade sodium hydroxide (NaOH, J.T. Baker, Phillipsburg,

84

NJ), (3-Aminopropyl)triethoxysilane (99%, APTES) (Sigma-Aldrich, St. Louis, MO), anhydrous

85

ethanol (100%, Decon Labs, Inc., PA), silica nanoparticles (SiNPs) with an average diameter of

86

0.1 µm (AngstromSphere Monodispersed Silica Powder, Fiber Optic Center Inc., MA), and

87

(heptadecafluoro-1,1,2,2-tetrahydrodecyl)trichlorosilane (FDTS, Gelest Inc., PA) were used for

88

the surface modification of the PVDF substrate.

89

Surface Modification of PVDF Membrane. A flat sheet polyvinylidene fluoride

90

(PVDF) membrane with a nominal pore size of 0.45 µm and an average thickness of 125 µm was

91

used as a substrate (HVHP, Millipore, Billerica, MA). The PVDF substrate was immersed in a

92

7.5 M NaOH solution at ~70 °C for three hours to generate hydroxyl surface functionality. To

93

ensure complete soaking of PVDF substrate in the NaOH solution, the membrane was prewetted

94

by ethanol. An alkaline treated PVDF substrate was then soaked in 1% v/v APTES in anhydrous

95

ethanol for one hour under gentle stirring to functionalize the surface with amine terminal groups,

96

rendering the PVDF substrate positively charged. The APTES-functionalized PVDF substrate

97

was immersed in an aqueous SiNP solution for one hour under gentle mixing. APTES

98

functionalization and SiNP coating were executed with only the top surface of the membrane in

99

contact with the suspension. The aqueous SiNP solution was prepared by dispersing 0.05 wt%

100

SiNPs in acetate buffer with an ionic strength of ~1 mM. The pH of the suspension was adjusted

101

to 4 to promote electrostatic attraction between the negatively charged SiNPs and the positively

102

charged PVDF substrate.

103

After attaching SiNPs to the PVDF substrate, the membrane was coated with

104

perfluorodecyltrichlorosilane (FDTS) via vapor-phase silanization to lower the membrane

105

surface energy. An SiNP-coated PVDF membrane was fixed on the wall of a glass bottle (Septa

106

jars, Thermo Fisher Scientific, Inc., MA) using heat resistant tape (Kapton® Polyimide, Uline,

107

WI). The bottle was nitrogen purged for 20 minutes through the septa cap. After nitrogen

108

purging, 100 µL FDTS was placed in a small petri dish inside the bottle. Then, the SiNP-coated

109

PVDF membrane was silanized in a vacuum oven (vacuum pressure at ~80 kPa and temperature

110

at ~100 °C) for 17 hours.

111

The stability of surface coating with SiNPs and FDTS was assessed by reevaluation of SEM

112

images and contact angles after the modified membranes were subjected to harsh chemical and 5 ACS Paragon Plus Environment

Environmental Science & Technology

113

physical stresses. Chemical stress was applied by exposing the modified PVDF membrane for 20

114

minutes to a pH 2 solution (HCl), a pH 12 solution (NaOH), or a 1.0 M NaCl solution, followed

115

by thorough rinse with DI water. Physical stress was exerted by immersing the membranes in a

116

sonicating water bath (Fisher Scientific F60) for 20 minutes.

117

Membrane Characterization. Surface morphology of the modified PVDF membranes

118

was investigated by scanning electron microscopy (SEM, Hitachi SU-70). Before imaging,

119

membrane samples were sputter-coated with a chromium layer (BTT-IV, Denton Vacuum, LLC,

120

Moorestown, NJ). To obtain cross-section images, membranes were dried at room temperature

121

and freeze-fractured using liquid nitrogen before the chromium coating. Acceleration voltage of

122

5.0 kV was applied to image all membrane samples.

123

Zeta potential of membrane surface was evaluated by a streaming potential analyzer utilizing

124

an asymmetric clamping cell (EKA, Brookhaven Instrument, Holtsville, NY) as described

125

elsewhere.26 Measurements were performed with a solution containing 1 mM KCl and 0.1 mM

126

KHCO3. Electrophoretic mobility of the silica nanoparticles (SiNPs) was measured in a

127

background electrolyte of 1 mM NaCl using a Zeta Potential Analyzer (ZetaPals, Brookhaven

128

Instruments, Holtsville, NY) at different pH values. The zeta potentials of SiNPs were calculated

129

from electrophoretic mobility measurement by using the tabulated data of Ottewill and Shaw.27

130

The elemental composition of the membranes was investigated by Fourier transform infrared

131

(FTIR) spectroscopy with attenuated total reflectance (ATR). Absorbance spectra were measured

132

with 32 scans of each sample at a spectral resolution of 1 cm-1. Background measurements in air

133

were collected before each membrane sample measurement.

134

Contact angles of the modified PVDF membranes for liquids with a wide range of surface

135

tensions, including water (γ = 72.8 mN/m), 0.1 mM SDS in 1.0 M NaCl (γ ≈ 38 mN/m), and

136

mineral oil (γ ≈ 31 mN/m) were measured by a goniometer (OneAttension, Biolin scientific

137

instrument) using the sessile drop method. A 5-µL liquid droplet was placed on the membrane

138

sample and photographed using a digital camera for three minutes. The left and right contact

139

angles were analyzed from the digital images by a post-processing software (OneAttension

140

software). The measurements were conducted on a minimum of two random locations with three

141

different membrane samples, and the data were averaged.

6 ACS Paragon Plus Environment

Page 6 of 25

Page 7 of 25

Environmental Science & Technology

142

Model and Shale Gas Produced Water Feed Solutions. NaCl (1.0 M) solution at

143

60 °C and DI water at 20 °C were employed as feed and permeate streams, respectively, for the

144

initial two hour DCMD runs. After stabilization, sodium dodecyl sulfate (SDS) or mineral oil

145

was added to the feed solution every two hours during the DCMD experiment to evaluate

146

wetting resistance of the control and the modified PVDF membranes. The SDS concentrations in

147

the feed solution after sequential SDS addition were 0.05, 0.1, and 0.2 mM. Surface tensions

148

with varying SDS concentrations in DI water and 1.0 M NaCl solution were determined by a

149

tensiometer (Sigma 300, KSV Instrument, Helsinki) using the Wilhelmy Plate method. Mineral

150

oil-in-water emulsion at concentration of 1% v/v was prepared by adding 0.1% v/v Tween80®

151

(non-ionic surfactant) (Sigma-Aldrich, St. Louis, MO) as an emulsifier. The mineral oil

152

concentrations in the feed solution after addition of emulsion every two hour DCMD were 0.001,

153

0.005, and 0.01% v/v, which correspond to 8, 40, and 80 mg/L. Size of mineral oil droplets in the

154

feed was analyzed by dynamic light scattering (DLS) measurements (ZetaPals, Brookhaven

155

Instruments, Holtsville, NY). The feed solution (initial volume of 1 L) was replenished every 50

156

mL loss to maintain variation of SDS or mineral oil concentration within 5%.

157

Shale gas produced water from the Permian Basin in Texas (provided by Gradiant

158

Corporation) was pre-filtered through 16-µm filter (Ashless Grade 44, GE Whatman, PA) to

159

remove large particulate and suspended matters before they were supplied to DCMD

160

experiments. The major composition and key properties of the pre-filtered shale gas produced

161

water were analyzed by a third-party analyzing laboratory (Environmental Service Lab, Inc., PA)

162

(Table S1). The biological activity of the shale gas wastewater was minimized by storing the

163

sample at 4 °C. The DCMD experiments with shale gas wastewater were conducted until the

164

feed solution (initial volume of 1 L) was concentrated by 50% (or until the permeate volume

165

reached 500 mL).

166

Membrane Distillation Performance Tests. The MD performance of the control and

167

modified PVDF membranes was evaluated by a laboratory-scale direct contact membrane

168

distillation (DCMD) unit. The membrane was inserted into a custom-built transparent acrylic cell

169

with channel dimensions of 77-mm in length, 26-mm in width, and 3-mm in depth. The effective

170

membrane area exposed to feed and permeate (distillate) streams was 20.0 cm2. Spacers were

171

inserted in both feed and permeate channels to support and maintain the membrane geometry in

7 ACS Paragon Plus Environment

Environmental Science & Technology

172

the cell. The temperatures of feed and permeate solutions were maintained at 60 and 20 °C,

173

respectively. A slightly higher cross-flow rate for the feed stream, 0.4 L/min (cross-flow velocity

174

of 8.5 cm/s), than for the permeate stream, 0.35 L/min (cross-flow velocity of 7.5 cm/s), was

175

used to facilitate the detection of MD membrane wetting.

176

The water vapor flux, Jw, across the membrane was determined by measuring the increase in

177

permeate weight. Electric conductivity of the permeate solution was monitored to assess the

178

NaCl concentration in the permeate stream, CP. The cumulative salt mass in the permeate stream,

179

VPCP , was divided by the volume of water vapor flux to calculate the permeate salt

180

concentration, from which the salt (NaCl) rejection, RNaCl, was determined using

181

 V C /J A t R NaCl = 1 − P P w m  100 CF  

182

where Vp, is the volume of permeate stream, CF is the initial NaCl concentration in the feed (1.0

183

M), Am is the membrane area, and t is time.

184 185

RESULTS AND DISCUSSION

186

Surface Properties of Modified Omniphobic Membrane. Silica nanoparticles (SiNPs)

187

were used for the omniphobic surface modification of a PVDF substrate because SiNP is

188

abundant in hydroxyl functional groups that allow surface fluorination via well-established silane

189

chemistry to lower the membrane surface energy.28 Further, spherical SiNP is a geometry that

190

provides a re-entrant structure for substrate surface, which is critical for achieving surface

191

omniphobicity.29

(1)

192

SiNPs possessing negative surface charge were attached to aminosilane functionalized PVDF

193

substrate via electrostatic attraction (Figure 1). First, a PVDF substrate was treated with high

194

concentration NaOH solution (7.5 M) under high temperatures (~70 °C) to generate hydroxyl

195

functional groups on PVDF polymer chains. Mechanisms for the formation of hydroxyl groups

196

on a PVDF chain by alkaline treatment are well described in previous studies.30, 31 After alkaline

197

treatment, the PVDF substrate was soaked in an aminosilane solution (APTES 1% v/v in

198

anhydrous ethanol) to functionalize the surface with amine terminal groups. The APTES

199

covalently binds to the hydroxyl groups on the PVDF substrate via hydrolysis and condensation, 8 ACS Paragon Plus Environment

Page 8 of 25

Page 9 of 25

Environmental Science & Technology

200

rendering the substrate surface positively charged. Positive surface charge of the APTES-

201

functionalized PVDF membrane enables binding of the negatively charged SiNPs via

202

electrostatic attraction. Grafted SiNPs were then coated with perfluorodecyltrichlorosilane

203

(FDTS) to lower the surface energy of the membrane.

204

Zeta potentials of the control, alkaline treated, and amine functionalized PVDF membranes

205

were evaluated to confirm a binding mechanism between the SiNPs and the PVDF substrate

206

(Figure 2A). The control PVDF substrate was found to be negatively charged over the entire pH

207

range investigated, consistent with other reported data.32, 33 Because the control PVDF membrane

208

does not possess a fixed surface charge, the observed negative zeta potential may be attributed to

209

the adsorption of anions, such as OH- and Cl- from the background electrolyte, onto the

210

membrane surface, as proposed elsewhere.34, 35 Zeta potential of the PVDF substrate became

211

more negative and stable over the whole pH range after treatment with alkaline solution (7.5 M

212

NaOH). The PVDF membrane became hydrophilic after alkaline treatment due to the formation

213

of hydroxyl groups.30,

214

substrate is likely because the hydrophilic surface offers closer proximity for anions (i.e., OH-

215

and Cl-) from the background electrolyte. Functionalization of the substrate surface with APTES

216

reduced the negative zeta potential (i.e., more positive charge) of the PVDF membrane with an

217

extrapolated isoelectric point of ~5.2 (Figure 2A). The SiNP coating was performed under

218

suspension pH at 4, where the zeta potential of the amine functionalized PVDF substrate is

219

positive while that of the SiNPs is negative.

31

Increased negative surface charges of the alkaline treated PVDF

220

Surface morphology of the PVDF substrates before and after SiNP coating was investigated

221

by scanning electron microscope (SEM). For comparison, top-down and cross-section SEM

222

images of the control PVDF and the SiNP-coated PVDF membranes are shown in Figure 2C.

223

The top-down SEM image shows that dense SiNP coating formed on the surface of the PVDF

224

substrate (Figure 2C-b). In addition, the cross-section SEM image of the SiNP-coated membrane

225

(Figure 2C-d) indicates that SiNPs penetrated into the porous substrate and created a 3-5 µm

226

thick SiNP coating layer. A re-entrant structure developed by a spherical or cylindrical surface

227

geometry creates a local kinetic barrier for the transition from the metastable Cassie-Baxter state

228

to the fully wetting Wenzel state for low surface tension liquids. Results from the SEM images

229

clearly show a re-entrant surface structure achieved by coating with spherical SiNPs. In addition,

9 ACS Paragon Plus Environment

Environmental Science & Technology

230

SiNP coating inside the pores (Figure 2C-d) is expected to offer multiple kinetic barriers, and

231

thus further increase membrane wetting resistance against low surface tension liquids.

Page 10 of 25

232

Surface coatings of the PVDF substrate with SiNPs and FDTS were confirmed by attenuated

233

total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy measurement. Figure 2B

234

shows the 1400−800 wavenumber region of the FTIR spectra, covering the characteristic

235

absorbances of the distinct functional groups in the control PVDF substrate and SiNP-, and

236

FDTS-coated PVDF membranes. After surface coating with SiNPs, a new peak at ~ 1110 cm-1

237

which corresponds to silanol groups on the SiNPs was detected. An increase in absorbance

238

intensity at ~1140 cm-1, which is ascribed to CF2 symmetric stretching mode, was observed after

239

coating with FDTS, substantiating successful surface fluorination via vapor-phase silanization.

240

Figure 1.

241

Figure 2.

242

Wetting Resistance of Modified Omniphobic Membrane. Contact angles for the

243

control and the modified PVDF membranes were measured to evaluate the surface wetting

244

resistance to liquids with a wide range of surface tensions, including water (γ = 72.8 mN/m), 0.1

245

mM sodium dodecyl sulfate (SDS) in 1.0 M NaCl (γ ≈ 38 mN/m), and mineral oil (γ ≈ 31 mN/m).

246

We have selected SDS and mineral oil as representative surface wetting agents found in shale

247

gas wastewater streams. The SDS solution was prepared with high salinity (1.0 M NaCl) to

248

mimic the high TDS shale gas wastewater. Also, we note that the surface tension of ionic

249

surfactant (i.e., SDS) solution is significantly reduced under high salt concentrations because

250

electrolytes screen the charge of the polar head groups, leading to a higher surfactant density at

251

the liquid-air interface as described in previous studies.36

252

Contact angles of liquids with different surface tensions on the control and the modified

253

PVDF membranes are shown in Figure 3. Due to the dynamic behavior of droplets of certain

254

testing liquids, the contact angles were monitored for three minutes. The control PVDF substrate

255

maintained a relatively high (~110°) and stable water contact angle during three minutes of

256

observation. However, the control PVDF substrate failed to resist wetting by surfactant (SDS)

257

solution and mineral oil (Figure 3A). The contact angle of 0.1 mM SDS solution reduced from

258

~100° to ~55° during the three minute measurement interval and complete wetting was found

10 ACS Paragon Plus Environment

Page 11 of 25

Environmental Science & Technology

259

two minutes after placing mineral oil on the control PVDF substrate. The modified PVDF

260

membrane exhibited superhydrophobicity with a water contact angle >150° (Figure 3B). The

261

observed significantly high water contact angle on the modified PVDF membrane is attributed to

262

vapor/air pockets developed between SiNPs attached on the substrate and ultralow surface

263

energy achieved by FDTS coating. Notably, the modified PVDF membrane exhibited high

264

(>130°) and stable (for three minute observation) contact angles with both SDS solution and

265

mineral oil. An omniphobic surface that repels both water and low surface tension liquids (e.g.,

266

oil and alcohol) is available only when the surface has features with a re-entrant structure and

267

ultralow surface energy. The observed wetting resistances of the modified PVDF membrane to

268

SDS solution and mineral oil demonstrate that substrate coatings with SiNP and FDTS

269

successfully imparted surface omniphobicity to the control PVDF membrane.

270

The irreversibility of SiNP and FDTS coatings on the substrate surface and subsequent

271

membrane wetting resistance were assessed by subjecting the modified PVDF membranes to

272

harsh chemical (i.e., pH 2, pH 12, or 1.0 M NaCl) and physical (i.e., bath sonication for 20

273

minutes) stresses. SEM micrographs taken after stress protocols showed no significant difference

274

compared to the results obtained from the SiNP-coated membrane not subjected to stresses (see

275

Figure S1 in the Supporting Information). The results indicated that electrostatic attraction

276

between the negatively charged SiNPs and the positively charged PVDF substrate was strong

277

enough to render the SiNP coating irreversible. Liquid contact angles can be used as a proxy to

278

appraise the stability of SiNP and FDTS coatings because the apparent contact angle is

279

determined by both surface morphology and chemistry. Figure 4 presents the contact angles of

280

water, 0.1 mM SDS in 1.0 M NaCl, and mineral oil on the modified PVDF membranes before

281

and after chemical and physical stresses. Contact angles did not change significantly for all

282

tested liquids, compared to those analyzed immediately after surface modification (indicated as

283

initial in Figure 4). We attribute the observed stable wetting resistance of the modified PVDF

284

membrane to the robust SiNP binding by electrostatic attraction and the strong covalent Si−O−Si

285

linking between FDTS and SiNP achieved via silanization (Figure 1). The confirmed stable

286

wetting resistance of the modified PVDF membrane ensures a long-term omniphobic

287

functionalization under typical MD operational conditions.

288

Figure 3.

11 ACS Paragon Plus Environment

Environmental Science & Technology

Page 12 of 25

Figure 4.

289 290 291

Desalination Performance with High Salinity, Low Surface Tension Feed Water.

292

To investigate the desalination performance of the control PVDF and modified omniphobic

293

membranes with feed water containing low surface tension contaminants, we performed direct

294

contact membrane distillation (DCMD) experiments using feed solutions with varying

295

concentrations of sodium dodecyl sulfate (SDS, anionic surfactant) and mineral oil. SDS and

296

mineral oil were sequentially introduced to the feed every two hours during the DCMD

297

experiment after an initial two hours stabilization period with a 1.0 M NaCl feed solution. Water

298

(vapor) flux and salt (NaCl) rejection were monitored for eight hours.

299

Surfactants are amphiphilic organic compounds comprising a hydrophilic polar head and a

300

hydrophobic non-polar tail.37 Below the critical micelle concentration (CMC), diffusion of

301

surfactant towards the air-liquid interface is triggered by the hydrophobic tail, which results in a

302

decrease of the solution surface tension. When surfactants are present in a solution of high

303

salinity, they significantly reduce the surface tension of the medium because the charge of the

304

surfactant polar head group is screened by counter ions, leading to a higher surfactant density at

305

the liquid-air interface.38 Surface tensions of the SDS solution measured using a tensiometer

306

were much lower in 1.0 M NaCl than in DI water (Figure S2-A, Supporting Information). We

307

added 0.05, 0.1, and 0.2 mM SDS into the feed solution (1.0 M NaCl at 60 °C) every two hours

308

during the DCMD experiments to progressively lower the feed surface tensions to ~53, ~42, and

309

~33 mN/m, respectively.36

310

A slight increase in water flux and a decrease in salt rejection were observed with the control

311

PVDF membrane after the addition of 0.05 mM SDS to the feed solution (Figure 5A). Because

312

we applied a slightly higher cross flow rate for the feed stream than the permeate stream (cross

313

flow velocities of 8.5 cm/s versus 7.5 cm/s for feed and permeate streams, respectively), the feed

314

solution penetrates through the MD membrane to the permeate side when pores are wetted. The

315

increase in water flux and the concomitant decrease in salt rejection observed for the control

316

PVDF membrane are evidence of pore wetting due to the reduced solution surface tension

317

resulting from the added surfactant (SDS) in the feed. Further SDS addition to the feed (i.e., 0.1

318

mM SDS) leads to a more drastic water flux increase and sudden decrease of salt rejection,

319

suggesting more severe or even complete membrane pore wetting. The modified omniphobic 12 ACS Paragon Plus Environment

Page 13 of 25

Environmental Science & Technology

320

membrane, on the other hand, showed stable water flux and complete salt rejection even after the

321

addition of 0.2 mM SDS to the feed, which corresponds to a solution surface tension of ~33

322

mN/m, demonstrating wetting resistance of the omniphobic membrane to the low surface tension

323

feed water.

324

The effect of oil in the feed solution on MD desalination performance of the control and

325

modified PVDF membranes is presented in Figure 5B. A homogeneous oil stock solution was

326

prepared by mixing mineral oil (1% v/v) and non-ionic surfactant (0.1% v/v) in water. The

327

prepared oil emulsion was sequentially added to the feed to achieve mineral oil concentrations of

328

8, 40, and 80 mg/L. The average diameter of the mineral oil droplets in the feed solution

329

measured by dynamic light scattering (DLS) was ~0.7 µm (Figure S2-B, Supporting

330

Information). The prepared feed solution simulates the wastewaters contaminated by emulsified

331

oils with droplet sizes below several micrometers, which are challenging to treat by conventional

332

pre-treatment practices, such as oil skimmers, coalescers, settling tanks, and depth filters.39

333

At a relatively low mineral oil concentration (0.001% v/v or 8 mg/L) in the feed, the control

334

PVDF membrane exhibited water flux decline without a significant change in salt rejection

335

(Figure 5B). Nonpolar oils easily adhere to the PVDF membrane by hydrophobic interactions,40

336

thereby causing membrane fouling.41 The observed water flux decline without significant change

337

in salt rejection at 0.001% v/v mineral oil concentration in the feed is likely because the water

338

and salt passage was blocked by the accumulated oils within the porous structure. At a higher

339

mineral oil concentration (0.005% v/v or 40 mg/L), however, the control PVDF membrane was

340

totally wetted by the feed solution as evidenced by drastically increased water flux and reduced

341

salt rejection, similar to the results obtained with the SDS feed solutions.

342

In contrast, the omniphobic membrane showed stable water flux and complete salt rejection

343

with feed solution mineral oil concentrations up to 0.01% v/v or 80 mg/L (Figure 5B). Mineral

344

oil droplets are expected to adhere to the omniphobic membrane via nonpolar-nonpolar

345

interaction as they adsorb on hydrophobic surfaces. However, contrary to the hydrophobic

346

control PVDF membrane, the modified omniphobic membrane resists wetting by mineral oils,

347

thereby preventing penetration of the oils into the porous substrate. Also, mineral oil droplets

348

adhered to the membrane can be readily removed by hydrodynamic shear force in cross-flow

349

DCMD operation due to the ultralow interfacial energy of the omniphobic surface.42, 43 13 ACS Paragon Plus Environment

Environmental Science & Technology

350

Page 14 of 25

Figure 5.

351

Desalination Performance with Shale Gas Produced Water. We further tested the

352

desalination performance of the modified omniphobic membrane by employing a brine from the

353

shale gas industry as a feed stream in DCMD. The shale gas brine has very high total dissolved

354

solids (TDS) concentration (101,000 mg/L) and contains substances which can potentially foul

355

and wet the MD membranes (Table S1, Supporting Information). In particular, oil and grease,

356

surfactants, and organic loading in the shale gas wastewater feed threaten the anti-wetting

357

properties of conventional hydrophobic membranes.

358

Figure 6 presents flux decline curves, expressed as normalized water flux, and permeate

359

conductivity obtained from DCMD experiments for the control PVDF and the omniphobic

360

membranes with the shale gas wastewater as feed solution (60 °C) and DI water as permeate

361

solution (20 °C). To allow meaningful comparison of membrane performance at different initial

362

water fluxes (23.5 and 13.6 L m-2 h-1 for the control PVDF and the omniphobic membranes,

363

respectively), data are presented as a function of cumulative permeate volume. Experiments were

364

continued until the feed recovery reached 50% (or permeate volume of 500 mL was attained),

365

which took 13.3 hours for experiments with the control PVDF membrane and 16.2 hours for the

366

omniphobic membrane. Other conditions were identical to those employed in the DCMD

367

experiments described in the previous subsection.

368

As presented in Figure 6, the omniphobic membrane experienced a much lower flux decline

369

and reduced salt passage compared to the control PVDF membrane. The observed significant

370

performance deterioration of the control PVDF membrane can be attributed to various factors,

371

including wetting, fouling, and scaling of the membrane, given the complex physicochemical

372

composition of the shale gas produced water used. Of these, membrane wetting is expected to be

373

the main factor that caused performance degradation of the control PVDF membrane in the early

374

DCMD stage because wetting is a phenomenon that affects MD performance much faster than

375

fouling and scaling.44, 45 Results obtained from a DCMD experiment using a clean 1.7 M NaCl

376

feed solution (salinity identical to the shale gas produced water) without surface wetting

377

substances further demonstrated a stable desalination performance of the control PVDF

378

membrane in the early MD stage (Figure S3-A of Supporting Information).

14 ACS Paragon Plus Environment

Page 15 of 25

Environmental Science & Technology

379

The omniphobic membrane exhibited stable MD performance with shale gas wastewater feed

380

(Figure 6B), which was comparable to the results obtained using a clean NaCl feed solution with

381

a salinity (~1.7 M NaCl) identical to the shale gas produced water (Figure S3-B of Supporting

382

Information). Although in both cases (i.e., shale gas produced water and 1.7 M NaCl solution

383

feed) a slight flux decline and a small conductivity increase in the permeate were observed in a

384

long-term DCMD run, likely caused by intrinsic membrane defects, the omniphobic membrane

385

exhibited much more stable desalination performance than the control PVDF membrane. The

386

observed stable desalination performance in MD with a shale gas produced water sample

387

containing a variety of low surface tension contaminants is attributable to the enhanced wetting

388

resistance of the omniphobic membrane achieved by surface modification with silica

389

nanoparticles and low surface energy materials (i.e., fluoroalkylsilane). Figure 6.

390 391

In summary, we demonstrated a facile and scalable surface engineering technique to fabricate

392

an omniphobic membrane that exhibits wetting resistance to low surface tension substances. Our

393

omniphobic membrane showed a stable performance in MD for desalination of shale gas

394

produced water, demonstrating the potential benefits of omniphobic membranes in treating high

395

salinity wastewaters that contain diverse low surface tension contaminants. Application of MD

396

with omniphobic membranes can be beneficial for a wide range of industries, considering the

397

increasing demand for efficient treatment of challenging industrial wastewaters and stricter

398

regulations for disposal of such wastewaters.

399 400

ASSOCIATED CONTENT

401

Major composition and key properties of pre-filtered shale gas produced water (Table S1); SEM

402

images of modified PVDF membranes taken after chemical and physical stresses (Figure S1);

403

surface tension of SDS solution in DI water and in 1 M NaCl (Figure S2-A) and size distribution

404

of mineral oil droplets (Figure S2-B); water (vapor) flux and permeate conductivity obtained

405

from DCMD experiments for unmodified PVDF and modified PVDF membranes with 1.7 M

406

NaCl feed solution and DI water permeate solution (Figure S3). This material is available free of

407

charge via the Internet at http://pubs.acs.org. 15 ACS Paragon Plus Environment

Environmental Science & Technology

408

ACKNOWLEGMENT

409

We acknowledge the support received from the National Science Foundation through the

410

Engineering Research Center for Nanotechnology-Enabled Water Treatment (ERC-1449500).

411

Facilities used were supported by the Yale Institute of Nanoscale and Quantum Engineering

412

(YINQE) and the Chemical and Biophysical Instrument Center (CBIC) at Yale.

413

16 ACS Paragon Plus Environment

Page 16 of 25

Page 17 of 25

414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459

Environmental Science & Technology

REFERENCES 1. Inman, M., The Fracking Fallacy. Nature 2014, 516, (7529), 28-30. 2. Veatch, R. W., Jr.; Moschovidis, Z. A., An Overview of Recent Advances in Hydraulic Fracturing Technology. In Society of Petroleum Engineers. 3. Vengosh, A.; Jackson, R. B.; Warner, N.; Darrah, T. H.; Kondash, A., A Critical Review of the Risks to Water Resources from Unconventional Shale Gas Development and Hydraulic Fracturing in the United States. Environ Sci Technol 2014, 48, (15), 8334-8348. 4. Getzinger, G. J.; O'Connor, M. P.; Hoelzer, K.; Drollette, B. D.; Karatum, O.; Deshusses, M. A.; Ferguson, P. L.; Elsner, M.; Plata, D. L., Natural Gas Residual Fluids: Sources, Endpoints, and Organic Chemical Composition after Centralized Waste Treatment in Pennsylvania. Environ Sci Technol 2015, 49, (14), 8347-8355. 5. https://fracfocus.org/. 6. Shaffer, D. L.; Chavez, L. H. A.; Ben-Sasson, M.; Castrillón, S. R.-V.; Yip, N. Y.; Elimelech, M., Desalination and Reuse of High-Salinity Shale Gas Produced Water: Drivers, Technologies, and Future Directions. Environ Sci Technol 2013, 47, 9569-9538. 7. Gregory, K. B.; Vidic, R. D.; Dzombak, D. A., Water Management Challenges Associated with the Production of Shale Gas by Hydraulic Fracturing. Elements 2011, 7, (3), 181-186. 8. Ahmadun, F. R.; Pendashteh, A.; Abdullah, L. C.; Biak, D. R. A.; Madaeni, S. S.; Abidin, Z. Z., Review of technologies for oil and gas produced water treatment. J Hazard Mater 2009, 170, (2-3), 530-551. 9. Tong, T.; Elimelech, M., The Global Rise of Zero Liquid Discharge for Wastewater Management: Drivers, Technologies, and Future Directions. Environ Sci Technol 2016, 50, (13), 6846-6855. 10. Lin, S. H.; Yip, N. Y.; Elimelech, M., Direct contact membrane distillation with heat recovery: Thermodynamic insights from module scale modeling. J Membrane Sci 2014, 453, 498-515. 11. Xu, P.; Cath, T. Y.; Robertson, A. P.; Reinhard, M.; Leckie, J. O.; Drewes, J. E., Critical Review of Desalination Concentrate Management, Treatment and Beneficial Use. Environ Eng Sci 2013, 30, (8), 502-514. 12. Shih, J. S.; Saiers, J. E.; Anisfeld, S. C.; Chu, Z. Y.; Muehlenbachs, L. A.; Olmstead, S. M. M., Characterization and Analysis of Liquid Waste from Marcellus Shale Gas Development. Environ Sci Technol 2015, 49, (16), 9557-9565. 13. Lee, J.; Boo, C.; Ryu, W.-H.; Taylor, A. D.; Elimelech, M., Development of Omniphobic Desalination Membranes Using a Charged Electrospun Nanofiber Scaffold. ACS Applied Materials & Interfaces 2016, 8, (17), 11154-11161. 14. Lin, S.; Nejati, S.; Boo, C.; Hu, Y.; Osuji, C. O.; Elimelech, M., Omniphobic membrane for robust membrane distillation. Environmental Science & Technology Letters 2014, 1, (11), 443-447. 15. Wang, Z. X.; Elimelech, M.; Lin, S. H., Environmental Applications of Interfacial Materials with Special Wettability. Environ Sci Technol 2016, 50, (5), 2132-2150. 16. Kumar, R. T. R.; Mogensen, K. B.; Boggild, P., Simple Approach to Superamphiphobic Overhanging Silicon Nanostructures. J Phys Chem C 2010, 114, (7), 2936-2940. 17. Aulin, C.; Yun, S. H.; Wagberg, L.; Lindstrom, T., Design of Highly Oleophobic Cellulose Surfaces from Structured Silicon Templates. Acs Applied Materials & Interfaces 2009, 1, (11), 2443-2452. 17 ACS Paragon Plus Environment

Environmental Science & Technology

460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505

Page 18 of 25

18. Fujii, T.; Aoki, Y.; Habazaki, H., Fabrication of Super-Oil-Repellent Dual Pillar Surfaces with Optimized Pillar Intervals. Langmuir 2011, 27, (19), 11752-11756. 19. Zhao, H.; Law, K. Y.; Sambhy, V., Fabrication, Surface Properties, and Origin of Superoleophobicity for a Model Textured Surface. Langmuir 2011, 27, (10), 5927-5935. 20. Ozbay, S.; Erbil, H. Y., Superhydrophobic and oleophobic surfaces obtained by graft copolymerization of perfluoroalkyl ethyl acrylate onto SBR rubber. Colloid Surface A 2015, 481, 537-546. 21. Zhang, G. W.; Lin, S. D.; Wyman, I.; Zou, H. L.; Hu, J. W.; Liu, G. J.; Wang, J. D.; Li, F.; Liu, F.; Hu, M. L., Robust Superamphiphobic Coatings Based on Silica Particles Bearing Bifunctional Random Copolymers. Acs Applied Materials & Interfaces 2013, 5, (24), 1346613477. 22. Zhao, H.; Park, K. C.; Law, K. Y., Effect of Surface Texturing on Superoleophobicity, Contact Angle Hysteresis, and "Robustness". Langmuir 2012, 28, (42), 14925-14934. 23. Leng, B. X.; Shao, Z. Z.; de With, G.; Ming, W. H., Superoleophobic Cotton Textiles. Langmuir 2009, 25, (4), 2456-2460. 24. Liu, F.; Hashim, N. A.; Liu, Y. T.; Abed, M. R. M.; Li, K., Progress in the production and modification of PVDF membranes. J Membrane Sci 2011, 375, (1-2), 1-27. 25. Nejati, S.; Boo, C.; Osuji, C. O.; Elimelech, M., Engineering flat sheet microporous PVDF films for membrane distillation. J Membrane Sci 2015, 492, 355-363. 26. Elimelech, M.; Chen, W. H.; Waypa, J. J., Measuring the Zeta (Electrokinetic) Potential of Reverse-Osmosis Membranes by a Streaming Potential Analyzer. Desalination 1994, 95, (3), 269-286. 27. Ottewill, R. H.; Shaw, J. N., Electrophoretic Studies on Polystyrene Latices. J Electroanal Chem 1972, 37, (Jun), 133-&. 28. Matsumoto, A.; Tsutsumi, K.; Schumacher, K.; Unger, K. K., Surface functionalization and stabilization of mesoporous silica spheres by silanization and their adsorption characteristics. Langmuir 2002, 18, (10), 4014-4019. 29. Boo, C.; Lee, J.; Elimelech, M., Engineering Surface Energy and Nanostructure of Microporous Films for Expanded Membrane Distillation Applications. Environ Sci Technol 2016, 50, (15), 8112-8119. 30. Ross, G. J.; Watts, J. F.; Hill, M. P.; Morrissey, P., Surface modification of poly(vinylidene fluoride) by alkaline treatment 1. The degradation mechanism. Polymer 2000, 41, (5), 1685-1696. 31. Zheng, Z. R.; Gu, Z. Y.; Huo, R. T.; Luo, Z. S., Superhydrophobic poly(vinylidene fluoride) film fabricated by alkali treatment enhancing chemical bath deposition. Applied Surface Science 2010, 256, (7), 2061-2065. 32. Liang, S.; Kang, Y.; Tiraferri, A.; Giannelis, E. P.; Huang, X.; Elimelech, M., Highly Hydrophilic Polyvinylidene Fluoride (PVDF) Ultrafiltration Membranes via Postfabrication Grafting of Surface-Tailored Silica Nanoparticles. Acs Applied Materials & Interfaces 2013, 5, (14), 6694-6703. 33. Han, M. J.; Barona, G. N. B.; Jung, B., Effect of surface charge on hydrophilically modified poly(vinylidene fluoride) membrane for microfiltration. Desalination 2011, 270, (1-3), 76-83. 34. Childress, A. E.; Elimelech, M., Effect of solution chemistry on the surface charge of polymeric reverse osmosis and nanofiltration membranes. J Membrane Sci 1996, 119, (2), 253268. 18 ACS Paragon Plus Environment

Page 19 of 25

506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535

Environmental Science & Technology

35. Elimelech, M.; Omelia, C. R., Effect of Electrolyte Type on the Electrophoretic Mobility of Polystyrene Latex Colloids. Colloid Surface 1990, 44, 165-178. 36. Matijevic, E.; Pethica, B. A., The Properties of Ionized Monolayers .1. Sodium Dodecyl Sulphate at the Air-Water Interface. Transactions of the Faraday Society 1958, 54, (9), 13821389. 37. Dominguez, A.; Fernandez, A.; Gonzalez, N.; Iglesias, E.; Montenegro, L., Determination of critical micelle concentration of some surfactants by three techniques. J Chem Educ 1997, 74, (10), 1227-1231. 38. Lu, J. R.; Marrocco, A.; Su, T. J.; Thomas, R. K.; Penfold, J., Adsorption of DodecylSulfate Surfactants with Monovalent Metal Counterions at the Air-Water-Interface Studied by Neutron Reflection and Surface-Tension. J Colloid Interf Sci 1993, 158, (2), 303-316. 39. Zhang, W. B.; Shi, Z.; Zhang, F.; Liu, X.; Jin, J.; Jiang, L., Superhydrophobic and Superoleophilic PVDF Membranes for Effective Separation of Water-in-Oil Emulsions with High Flux. Adv Mater 2013, 25, (14), 2071-2076. 40. Tanford, C., Interfacial Free-Energy and the Hydrophobic Effect. P Natl Acad Sci USA 1979, 76, (9), 4175-4176. 41. Lu, X. M.; Peng, Y. L.; Ge, L.; Lin, R. J.; Zhu, Z. H.; Liu, S. M., Amphiphobic PVDF composite membranes for anti-fouling direct contact membrane distillation. J Membrane Sci 2016, 505, 61-69. 42. Werber, J. R.; Osuji, C. O.; Elimelech, M., Materials for next-generation desalination and water purification membranes. Nature Reviews Materials 2016, 1, 16018. 43. Chen, W. J.; Su, Y. L.; Peng, J. M.; Zhao, X. T.; Jiang, Z. Y.; Dong, Y. A.; Zhang, Y.; Liang, Y. G.; Liu, J. Z., Efficient Wastewater Treatment by Membranes through Constructing Tunable Antifouling Membrane Surfaces. Environ Sci Technol 2011, 45, (15), 6545-6552. 44. Srisurichan, S.; Jiraratananon, R.; Fane, A. G., Humic acid fouling in the membrane distillation process. Desalination 2005, 174, (1), 63-72. 45. Razmjou, A.; Arifin, E.; Dong, G. X.; Mansouri, J.; Chen, V., Superhydrophobic modification of TiO2 nanocomposite PVDF membranes for applications in membrane distillation. J Membrane Sci 2012, 415, 850-863.

536

19 ACS Paragon Plus Environment

Environmental Science & Technology

537 538 539 540

Figure 1. Schematic illustrating the procedure for omniphobic surface modification of PVDF

541

membrane. (A) Hydroxyl groups are generated on the surface of the PVDF substrate by alkaline

542

treatment with a 7.5 M NaOH solution at 70 °C for three hours. (B) Hydroxyl groups are used

543

for grafting of (3-Aminopropyl)triethoxysilane (APTES). (C) Silica nanoparticles (SiNPs) bind

544

to APTES functionalized PVDF substrate via electrostatic attraction. (D) SiNPs are coated with

545

perfluorodecyltrichlorosilane (FDTS) via vapor-phase silanization.

546 547 548 549 550 551 552 553 554 555 556 557 558 559

20 ACS Paragon Plus Environment

Page 20 of 25

Page 21 of 25

Environmental Science & Technology

Zeta Potential (mV)

(A) 60 40 20 0 -20 -40 -60 -80

Amine Functionalized PVDF Control PVDF Alkaline Treated PVDF SiNPs (100 nm)

3

4

5

6

7

8

9

pH

Absorbance (a.u.)

(B)

1400

After FDTS Coating After SiNP Coating Control PVDF 1140 cm-1 1110 cm-1

1200

1000

Wavenumber (cm-1)

800

560 561 562

Figure 2. (A) Zeta potential as a function of pH of silica nanoparticles (SiNPs) and PVDF

563

membranes at various stages of surface modification. (B) ATR-FTIR spectra of control, SiNP-

564

coated, and FDTS-coated PVDF membranes. (C) SEM images depicting top-down and cross-

565

section of control and SiNP-coated PVDF membranes.

566

21 ACS Paragon Plus Environment

Environmental Science & Technology

Page 22 of 25

Water (72.8 mN m-1) 0.1 mM SDS in 1 M NaCl (~38 mN m-1) Mineral Oil (~30 mN m-1)

Contact Angle (°)

200

(A) Control PVDF

150 100

Complete Wicking

50 200

(B) Omniphobic PVDF

150 100 50 0

567 568

Complete Wicking

60

120

180

Time (sec)

569 570

Figure 3. Variation in static contact angle of (A) control PVDF and (B) modified omniphobic

571

PVDF membranes measured with DI water, 0.1 mM SDS in 1 M NaCl, and mineral oil. A 5-µL

572

liquid droplet was placed on the membrane surface and the contact angle was monitored for three

573

minutes (180 seconds). Error bars represent standard deviations of two contact angles from three

574

different membrane samples. Surface tensions of the liquids are indicated in legends.

575

22 ACS Paragon Plus Environment

Environmental Science & Technology

Contact Angle (°)

Page 23 of 25

Initial pH 2 (HCl) pH 12 (NaOH) 1.0 M NaCl Bath Sonicated

210 180 150 120 90

Water

0.1 mM SDS (in 1 M NaCl)

Mineral Oil

576 577

Figure 4. Contact angles of DI water, 0.1 mM SDS in 1 M NaCl, and mineral oil on the

578

membrane surface measured immediately after surface modification (indicated as initial) and

579

after the modified membranes were subjected to the indicated physical and chemical stresses.

580

Contact angles three minutes after placing 5 µL of the testing liquid were determined to be a

581

steady-state value. The variation in liquid volume during the three minute measurement was

582

within 0.5 µL. Error bars represent standard deviations of two contact angles from three different

583

membrane samples.

584 585 586 587 588 589 590 591 592 23 ACS Paragon Plus Environment

Environmental Science & Technology

Page 24 of 25

0.05

50

0.1

0.2

(A)

Control PVDF Omniphobic PVDF

100

40 30

90

20 10 0 0

Salt (NaCl) Rejection

Water Flux (L m-2 h-1)

Surfactant Concentration in the Feed (mM)

120

240

80 480

360

Time (min)

593

594

0.001

50

0.005

0.01

(B)

Control PVDF Omniphobic PVDF

100

40 30

90

20 10 0 0

Salt (NaCl) Rejection

Water Flux (L m-2 h-1)

Oil Concentration in the Feed (% v/v )

120

240

360

80 480

Time (min)

595 596

Figure 5. Water flux and salt rejection of control and modified PVDF membranes measured in

597

DCMD using 1 M NaCl at 60 °C with varying (A) SDS and (B) mineral oil concentrations as

598

feed solution and DI water at 20 °C as permeate solution. The SDS concentrations in the feed

599

after sequential additions every two hours were 0.05, 0.1, and 0.2 mM, and the corresponding

600

estimated surface tensions of the feed solutions were ~53, ~42, and ~33 mN/m, respectively. The

601

mineral oil concentrations in the feed after sequential additions every two hours were 0.001,

602

0.005, and 0.01% v/v, which correspond to 8, 40, and 80 mg/L. The feed solution (initial volume

603

of 1 L) was replenished after every 50 mL volume loss to maintain the variation of contaminant

604

(SDS and mineral oil) concentrations within less than 5%.

605 24 ACS Paragon Plus Environment

1.0

150

0.8

100 50

0.6 0

0 100 200 300 400 500

Permeate Volume (mL)

1.2 (B) Omniphobic PVDF

200

1.0

150

0.8

100 50

0.6 0

0 100 200 300 400 500

Permeate Volume (mL)

Permeate Conductivity (µS cm-1)

200

Permeate Conductivity (µS cm-1)

606 607

1.2 (A) Control PVDF

Nomalized Water Flux (J / J0)

Environmental Science & Technology

Nomalized Water Flux (J / J0)

Page 25 of 25

608

Figure 6. Desalination performances of (A) control PVDF and (B) modified omniphobic PVDF

609

membranes with brines from shale produced water. Water vapor flux and permeate conductivity

610

were measured in DCMD until the cumulative permeate volume reached 500 mL, using shale

611

gas wastewater at 60 °C as a feed stream and DI water at 20 °C as a permeate stream. Properties

612

of the pre-filtered (14 µm filter) shale gas wastewater are described in Table S1 (Supporting

613

Information). Initial water flux, J0, was determined by averaging the permeate flux from the

614

onset of the experiment until the permeate volume reached 25 mL. The obtained initial water

615

fluxes were 23.5 and 13.6 L m-2 h-1 for the control and omniphobic PVDF membranes,

616

respectively. Water flux performance is expressed as a normalized water flux, J/J0, based on the

617

initial water flux.

25 ACS Paragon Plus Environment