Characterization and Performance Relationships for a Commercial

Sep 17, 2015 - Characterization and Performance Relationships for a Commercial Thin Film Composite Membrane in Forward Osmosis Desalination and Pressu...
1 downloads 11 Views 3MB Size
Subscriber access provided by UNIV OF NEBRASKA - LINCOLN

Article

Characterization and Performance Relationships for a Commercial Thin Film Composite Membrane in Forward Osmosis Desalination and Pressure Retarded Osmosis Jason T Arena, Seetha S Manickam, Kevin Reimund, Pavel A Brodskiy, and Jeffrey McCutcheon Ind. Eng. Chem. Res., Just Accepted Manuscript • Publication Date (Web): 17 Sep 2015 Downloaded from http://pubs.acs.org on September 17, 2015

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.

Industrial & Engineering Chemistry Research 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 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

1

2

Characterization and Performance Relationships for a Commercial

3

Thin Film Composite Membrane in Forward Osmosis Desalination

4

and Pressure Retarded Osmosis

5 6

Jason T. Arena1, Seetha S Manickam1, Kevin K. Reimund1, Pavel Brodskiy2, Jeffrey R.

7

McCutcheon1*

8 9

1

Department of Chemical and Biomolecular Engineering, University of Connecticut,

10

Storrs, CT, 06269, USA

11

2

12

Notre Dame, IN, 46556, USA

Department of Chemical and Biomolecular Engineering, University of Notre Dame,

13

Submitted to Ind. Eng. Chem. Res.

14 15 16

*Corresponding Author

17

email: [email protected]

18

phone: 860-486-4601

ACS Paragon Plus Environment

Page 1 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 40

19

Abstract:

20

This paper presents performance data for an early generation thin film composite

21

membrane from Oasys Water. These tests sought to measure the membrane’s basic

22

chemical and morphological characteristics with additional testing to measure benchmark

23

performance in both seawater-riverwater pressure retarded osmosis and forward osmosis

24

desalination using the ammonia-carbon dioxide draw solution. In pressure retarded

25

osmosis the membrane exhibited compaction which increased structural parameter. While

26

in forward osmosis desalination, high cation flux was observed using the ammonia-

27

carbon dioxide draw solution.

28 29

ACS Paragon Plus Environment

Page 2 of 40

Page 3 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

30

Industrial & Engineering Chemistry Research

1. Introduction

31

Forward osmosis (FO) is a suite of technologies that harness chemical potential

32

gradients, developed from concentration differences, between two aqueous systems.1,2 FO

33

processes can take the form of waste or solution concentration, energy production, or

34

water purification.2,3,4,5 At the heart of these processes is a semi-permeable membrane

35

that mediates the concentration gradient between a concentrated draw solution and dilute

36

feed solution, allowing osmosis. Membranes in FO processes can be subjected to a

37

variety of differing conditions ranging from alkaline,6,7,8 acidic,9 high temperatures,10

38

high transmembrane pressure,11,12 or charged organic solvents.13 This necessitates the

39

development and use of a chemically tolerant membrane platform for these processes.

40

The current most widely available FO membrane is produced by Hydration

41

Technology Innovations (HTI). This membrane is made from cellulose triacetate, formed

42

through a Loeb-Sourirajan type wet casting process.14 While this has produced a

43

membrane with sufficient permeance, selectivity and chemical resilience to operate in a

44

number of processes;1,15,7,4,16,17,18,19,11 cellulose acetates are vulnerable to hydrolysis

45

which results in the replacement of acetyl groups with hydroxyl degrading membrane

46

selectivity.20,21 Additionally, membranes made from cellulose acetates characteristically

47

tend to have lower water permeance than a similar thin film composite membrane.20

48

These limitations encouraged the development of alternative membrane chemistries for

49

FO. Thin film composite (TFC) membranes have been considered as the logical

50

replacement for cellulose derived membranes in FO; however, early studies observing

51

commercial TFC reverse osmosis membranes in forward osmosis reported low water

52

fluxes.7,16 TFC membranes typically employ a cross-linked polyamide selective layer

ACS Paragon Plus Environment

Page 3 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 40

53

that, while susceptible to degradation by hypochlorous acid and hypochlorite salts,

54

exhibits stability over a broader pH range than cellulose acetate based membranes.20

22

55

The advantages of TFC membranes have driven the development of new membranes

56

adapted for the unique requirements of FO processes where the support layer has been

57

designed to minimized solute diffusion limitations. A common approach to FO

58

membrane characterization is most often performed using a simple model draw solute,

59

typically sodium chloride, under low pressure operating conditions;23,24,25,26,27,28,29,30,31,32

60

however, this testing methodology fails to capture membrane performance under more

61

realistic operating conditions whether it may be separating two differing electrolyte

62

solutions, as in a forward osmosis (FO) process,33,34,7,6,13,8 or under substantial hydrostatic

63

pressures, as in a pressure retarded osmosis (PRO) process.35,12,11,36,37,38

64

This objective of this paper is to characterize an early generation TFC FO membrane

65

from Oasys Water. Tests will be performed to examine the membrane’s surface and pore

66

structure. Additionally, the membranes will be tested to measure its intrinsic transport

67

properties in reverse osmosis and forward osmosis using sodium chloride. Finally the

68

membrane will be characterized under conditions inspired by FO processes specifically

69

seawater-river water PRO process and FO desalination using the ammonia-carbon

70

dioxide (NH3-CO2) draw solution. The usage of the NH3-CO2 FO process is significant in

71

both its previously demonstrated capacity for desalination using HTI’s CTA membrane7,8

72

and its use in Oasys Water’s osmotic brine concentrator.34

73

2. Materials and Methods

74

2.1. Materials

ACS Paragon Plus Environment

Page 4 of 40

Page 5 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

75

A proprietary TFC membrane, later referred to as the TFC, was provided by Oasys

76

Water (Boston, MA) in July 2012. The continuous rapid evolution of TFC FO

77

membranes means this particular membrane has since been superseded.

78

available patent literature the membrane is likely a polyamide TFC built upon a

79

polysulfone (PSu)/ polyethylene terephthalate (PET) supporting layer.39 Sodium chloride,

80

ammonium bicarbonate, and ammonium hydroxide were purchased from Fisher

81

Scientific (Pittsburgh, PA). Sodium tetraphenyl boron, potassium chromate, and silver

82

nitrate were purchased from Acros Organics (Geel, Belgium). Isopropanol was purchased

83

from J.T. Baker (Center Valley, PA). Water used in this study was ultrapure Milli-Q

84

(18.2 MΩ) water produced by a Millipore Integral 10 water system, (Millipore

85

Corporation, Billerica, MA).

86

2.2. Reverse osmosis characterization

Based on

87

The water permeance was measured in a lab scale reverse osmosis (RO) testing

88

system at pressures between 8.6 and 29.3 bar at a temperature of 20°C. Rejection tests

89

were carried out following the measurement of water permeance at 15.5 bar with a 2000

90

ppm sodium chloride feed at 20°C with a cross flow velocity of 0.25 m·s-1. Sodium

91

chloride rejection was measured using conductivity.

92

conditions of the system and empirical data40,41 intrinsic rejections were determined from

93

a Sherwood number correlation.1,42 Intrinsic rejection was used to determine the sodium

94

chloride permeability for this membrane and calculated from Eq. (1).1

95

B=

Based upon hydrodynamic

( 1 − R) ( 1 − R) A( ∆P − ∆π ) = J w (1) R R

96

Here B is the solute permeability (L·m-2·hr-1), R is the rejection, A is the water

97

permeance of the membrane (L·m-2·hr-1·bar-1), ∆P is the transmembrane hydrostatic

ACS Paragon Plus Environment

Page 5 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 40

98

pressure (bar), ∆π is the transmembrane osmotic pressure (bar), and Jw is the water flux

99

of the rejection measurement (L·m-2·hr-1).

100

2.3. Membrane structure evaluation

101

2.3.1 Scanning electron microscopy

102

The TFC membrane PSu layer pore structures were imaged with a FEI Phenom

103

scanning electron microscope (SEM) (FEI Company Hillsboro, OR). These samples

104

were prepared using a freeze fracturing technique on fresh membrane samples which

105

have had their PET removed. The PET was removed by taking advantage of the

106

stretchability of the loose PET this membrane was built upon. Upon stretching, the PSu

107

and polyamide of the membrane break and separate from the stretched PET layer. This

108

separation was used to start a clean separation of the membrane and PET. A small sample

109

of membrane with the PET removed was cut approximate five centimeters from the

110

starting tear and freeze fractured by immersing it in liquid nitrogen and snapping it in

111

half. The freeze fracturing technique has been used elsewhere to image the cross-sections

112

of membranes and allows for a clean, straight break preserving the internal pore structure

113

for observation.24 The TFC membrane’s polyamide selective layer was imaged with a

114

JEOL 6335F Field Emission SEM. The surface of the selective layer was imaged to

115

observe morphology of the membrane selective layer.

116

2.3.2. Mercury intrusion porosimetry

117

An AutoPoreIV mercury intrusion porosimeter (Micromeritics Instrument

118

Corporation, Norcross, GA) was used to characterize the membranes for pore diameter

119

and total pore volume. In mercury intrusion porosimetry (MIP), a sample chamber

120

containing dried membrane samples is vacuum evacuated and mercury is intruded into

ACS Paragon Plus Environment

Page 6 of 40

Page 7 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

121

the membrane pores. Intrusion pressures ranging from 0.14 to 1380 bar (2 to 20,000 psi)

122

were used for the pore diameter measurements, measuring pores with diameters of 90 µm

123

to 20 nm. This technique can detect both through and blind pores but not closed pores.43,6

124

2.3.3. Structural parameter calculation

125

The structural parameter is the transport property of membranes most different

126

between those developed for hydrostatic pressure driven processes when compared to

127

those developed for osmotic pressure driven membrane separations due to the formers

128

need for thick fabric layers to mechanically support the TFC membrane. The structural

129

parameter is represented within the governing equations for water fluxes in both of the

130

commonly referenced membrane orientations the PRO mode (draw solution in contact

131

with the membrane selective layer) and the FO mode (draw solution in contact with the

132

membrane support layer). Eq. (2) and Eq. (3) show one the more rigorous forms of the

133

governing equation for water flux in both the PRO mode (Eq. (2))44 and FO mode (Eq.

134

(3)).45

135

  J S   J   π d,b exp − w k  − π f,b exp w D     f ,b     J w = A − ∆P  (2)   J 1 + B exp J w S  − exp − w    D f ,b  k   J w    

136

  J S  J   π d,b exp − w D  − π f,b exp w k    d ,b     J w = A  (3) B     J J S   1 +  exp w  − exp − w Dd ,b    J w   k   

137

In these equations Jw is water flux, A is water permeance, πd,b is the bulk draw osmotic

138

pressure, k is the external boundary layer mass transfer coefficient (1.90·10-5 m·s-1 for

ACS Paragon Plus Environment

Page 7 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 40

139

PRO and 1.82·10-5 m·s-1 for FO), πf,b is the bulk feed osmotic pressure, S is the structural

140

parameter for the membrane, Df,b is the bulk feed diffusivity (1.23·10-9 m2·s-1), B is the

141

solute permeance, ∆P is the transmembrane pressure, and Dd,b is the bulk draw diffusivity

142

(1.30·10-9 m2·s-1). Diffusivities calculate from experimental data41 and mass transfer

143

coefficient calculated from established correlations.1,42

144

The structural parameter of a membrane describes the effective diffusion limited

145

distance with the membrane support structure and ideally relates to the morphology of the

146

membrane’s support structure, and this relationship is commonly expressed by Eq.

147

(4).1,29,26 S=

148

t⋅τ (4) ε

149

In Eq. (4), t is the thickness of the membrane support layer, τ is the tortuosity of the

150

support layer, and ε is the porosity of the support layer. Despite this relationship the

151

structural parameter of a membrane is most commonly calculated from Eq. (2) and Eq.

152

(3) by a numerical solution for S with specified experimental conditions (temperature,

153

flow channel, cross-flow velocity, draw and feed solutions), measured water flux, and

154

membrane selective layer properties (water permeance and solute permeability). In

155

addition, Eq. (2) can be used to simulate membrane performance in a PRO process and

156

various derivations of this equation have served as the standard benchmark for

157

experimental membrane performance in PRO.11,36,12,46

158

2.4. Membrane surface properties

159

2.4.1. Fourier transform infrared spectroscopy

ACS Paragon Plus Environment

Page 8 of 40

Page 9 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

160

The TFC membrane was tested in Fourier transform infrared (FTIR) spectroscopy to

161

examine the surface functional groups of the membranes’ selective layers. A Thermo

162

Scientific (Waltham, MA) Nicolet iS10 FTIR spectrophotometer with Smart iTR

163

attachment was used to perform these measurements on a dried membrane.

164

Measurements were taken on the selective layer using 64 scans with a resolution of

165

4 cm-1.

166

2.4.2. Surface hydrophilicity by contact angle goniometry

167

The support layer contact angles of a neat membrane and a membrane with the PET

168

removed were measured using the sessile drop method,24 with air as the light phase and

169

DI water as the heavy phase, on a CAM 101 series contact angle goniometer (KSV

170

Company Linthicum Heights, MD). The values were taken as an average of at least four

171

points with a volume of 7 ± 1 µL.

172

2.5. Osmotically driven membrane process performance

173

2.5.1. Pressure retarded osmosis testing

174

Membranes were tested in triplicate using fresh membrane samples on a bench scale

175

pressure retarded osmosis test system at an operating temperature of 20°C. As in other

176

studies working with various iterations of this TFC membrane, it was tested following a

177

short soak in a 50% 2-propanol/water solution (i.e. < 1 min) to wet out the membrane.23,33

178

The configuration of this system has been described in prior study.35,47 A draw solute

179

concentration 0.5 M NaCl was used. The draw solution was circulated co-currently

180

against a deionized water feed with cross-flow velocities of 0.25 m·s-1 for both the draw

181

and feed solutions. A tricot RO permeate spacer was used to support the membrane in the

182

PRO system cell. Pressure gauges located on the inlet and outlet of the PRO cell served

ACS Paragon Plus Environment

Page 9 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 40

183

to measure the pressure drop through the cell so transmembrane pressure could be

184

accurately determined. For the tightly packed feed spacer a pressure drop of 1.7 bar (24

185

psi) was observed with an inlet pressure of 1.9 bar (27 psi) and an outlet pressure of 0.2

186

bar (3 psi). For the purposes of this study an average feed pressure of 1 bar (the rounded

187

linear average of the inlet and outlet pressures) was assumed for transmembrane pressure

188

determination, making the transmembrane pressure equal to the draw solution pressure

189

minus 1 bar. No pressure drop was observed within the draw solution channel.

190

The maximum hydrostatic pressure tested was statistically near the flux reversal point

191

(i.e. the error bars overlap 0 L—m-2—h-1 water flux). The flux reversal point can be defined

192

as the hydrostatic pressure at which water flux is zero. Tests were started with a draw

193

solution hydrostatic pressure of 2.8 bar (40 psi) and increased in 2.8 bar (40 psi)

194

increments until the flux reversal point. After data was collected near the flux reversal

195

point, pressures were decreased back to 2.8 bar, again in 2.8 bar (40 psi) increments. The

196

sequence of pressure increases and pressure decreases are later referenced as the

197

ascending pressure ramp and descending pressure ramp, respectively. Running a PRO

198

test in both ascending and descending pressure ramps allows for the detection of

199

permanent damage to the membrane selective layer, which would result in an increase in

200

reverse solute flux during the descending pressure ramp due to loss of selectivity from

201

selective layer damage.

202

Following collection of experimental data (water and solute flux) membrane

203

structural parameter, theoretical water flux, and power density were calculated. Structural

204

parameters were calculated using Eq. (2) with PRO experimental data and assuming

205

constant selective layer properties (i.e. no selective layer damage). This calculation seeks

ACS Paragon Plus Environment

Page 10 of 40

Page 11 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

206

to illustrate how mass transport through the selective layer is impacted by membrane

207

compaction from the increasing applied hydrostatic pressure.

208

Theoretical water fluxes as a function of changing draw solution hydrostatic pressure

209

were calculated from Eq. (2). These calculations were performed using three differing

210

sets of assumptions, with considerations towards both external and internal concentration

211

polarization. The water fluxes were calculated without either ECP or ICP, with ECP and

212

without ICP, and without ECP with variable ECP. The assumptions and additional details

213

included within the calculation of these theoretical water fluxes are in Table 1.

214

ACS Paragon Plus Environment

Page 11 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 40

215 Table 1: Assumptions for the numerical simulation of membrane PRO performance with a 0.5 M NaCl draw solution at 20°C and 0.25 m/s.

Descriptor Simulated

216 217 218

Assumptions • Constant A • Constant B • Constant S • Negligible ECP

Notes • A and B values calculated from RO • S value calculated from zero transmembrane pressure tests by Eq. (2).

Simulated w/ compaction

• • • •

Constant A Constant B Variable S w/ pressure Negligible ECP

Simulated w/ ECP

• • • •

Constant A Constant B Constant S ECP w/ constant k

• A and B values calculated from RO • Uses S values assumed to be variable with pressure and calculated from experimental data and finding numerical solutions for S using Eq. (2). • A and B values calculated from RO • k calculated from Sherwood correlation based upon system hydrodynamic conditions. • S value calculated from zero transmembrane pressure tests by Eq. (2).

The theoretical power density can be calculated from a known transmembrane pressure and water flux using Eq. (5).36

W = η ⋅ J w ⋅ ∆P (5)

219 220

In Eq. (5), W is the power density of the membrane, η is the turbine efficiency (for the

221

purpose of this study assumed to be 1), Jw is the water flux and ∆P is the transmembrane

222

pressure.

223

2.5.2. Forward osmosis desalination testing

224 225

226

The TFC membrane was assessed for sodium chloride rejection in forward osmosis. In a FO process the rejection can be calculated by Eq. (6).6,7,8

 J s ,i   Jw   × 100% (6) % Rejection = 1 − c f ,i     

ACS Paragon Plus Environment

Page 12 of 40

Page 13 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

227

In Eq. (6) Js,i is the forward solute flux (flux of a solute in the same direction as water

228

flux) of component i, Jw is the water flux across the membrane, and cf,i is the

229

concentration of component i in the feed solution.

230

These tests were performed using a sodium chloride feed solution with a NH3-CO2

231

based draw solution. The salts formed from NH3 and CO2 gases in solution are highly

232

soluble, and some formulations of the NH3-CO2 have sufficient osmotic pressure to

233

dewater feeds solutions which possess a high concentration of dissolve solids. Its

234

viability has been demonstrated in its use in Oasys Water’s produced water brine

235

concentrator which has been shown concentrating brines up to 180,000 mg·L-1 TDS.34

236

The NH3-CO2 desalination tests were performed in a laboratory scale osmosis test

237

systems using a 2.0 M (carbon basis) draw solution with feed solutions of deionized

238

water and 0.25 M NaCl.6 The ammonia to carbon dioxide ratio was varied for these tests

239

to observe what effect if any this would have on desalination performance. The ratios

240

used for this study were 1.2:1, 1.5:1 and 2:1 NH3:CO2 on a molar basis. These solutions

241

were circulated counter-current with a cross flow velocity of 0.25 m·s-1 at 23±1 °C. The

242

membrane support layer was in contact with the NH3-CO2 draw solution (FO mode).

243

2.5.4. Draw solution speciation

244

The NH3-CO2 draw solution comprises a varied mixture of chemical species within

245

solution. Within the draw solution there are dissolved NH3 and CO2 gases in addition to

246

ammonium (NH4+) cations and bicarbonate (HCO3-), carbonate (CO32-), and carbamate

247

(NH2COO-) anions. The solution is typically alkaline having pHs > 7.8,6 Many studies on

248

the equilibrium relationship between NH3 and CO2 gases within solution have been

ACS Paragon Plus Environment

Page 13 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 40

249

performed.48,49,50,51 Draw solute speciation is affected by 5 chemical equilibria, mass

250

balances upon the nitrogen species, carbon species, and solution electroneutrality.49

251 252

ACS Paragon Plus Environment

Page 14 of 40

Page 15 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

253 254

Table 2: Relationships governing the speciation of NH3-CO2 draw solution

Chemical Equilibria +

-

NH3+H2O↔NH4 +OH +

-

CO +H O↔H +HCO 2

2

3

-

+

2-

HCO ↔H +CO 3

3

-

+

-

NH +HCO ↔H +NH COO 3

3

2

+

-

H2O↔H +OH Mass Balances mtotal-N = mNH₃ + mNH₄⁺ + mNH₂COO⁻ mtotal-C = mCO₂ + mHCO₃⁻ + mCO₃²⁻ + mNH₂COO⁻ Electroneutrality mNH₄⁺ = mHCO₃⁻ + 2mCO₃²⁻ + mNH₂COO⁻ 255 256

The concentration of each species is dependent on the concentration of aqueous

257

ammonia and carbon dioxide. Using the relationships shown in Table 2 a numerical

258

determination for the concentration of ionic and neutral species within solution can be

259

obtained. Draw solute speciation was determined numerical using Mathematica

260

accounting for ion, and molecular interactions parameters given by Edwards et al.,49 NH3

261

and CO2 equilibrium constants from Kawazuishi and Prausnitz,48 and water self-

262

dissociation equilibrium constants from Robinson and Stokes.52 The source code of this

263

program has been provided as part of the supplementary material. In solving for the

ACS Paragon Plus Environment

Page 15 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 40

264

speciation of the NH3-CO2 draw solution a direct calculation of the osmotic pressure of

265

these solution can be obtain from the water activity in solution by Eq. (7).52,53,54 π=−

266

1 ln (a w )RT (7) vw

267

Here π is the osmotic pressure of the solution in bar, R is the ideal gas constant (0.08314

268

L·bar·mol-1·K-1), T is the absolute temperature, vw is the molar volume of water

269

(0.018018 L·mol-1) and aw is the molal activity of water.

270

2.5.4. Dissolved species quantification

271

The flux of ammonia species, sodium, and chloride were measured using techniques

272

previously illustrated by Arena et al.6 All fluxes were determined based on the change in

273

concentration of the solute of interest (i.e. ammonia species present with in the feed

274

solution, sodium and chloride ions within the draw solution) over the duration of the test

275

and the membrane surface area. The concentration of ammonium species was measured

276

gravimetrically using sodium tetraphenyl boron to precipitate ammonia as ammonium

277

tetraphenyl borate.55,56 The concentration of chloride was determined from the Modr

278

titration55 on a sample of rehydrated draw solution from which the water, dissolved

279

ammonia, and dissolved carbon dioxide has been boiled off. The concentration of sodium

280

was determine using a Perkin-Elmer 3100 Atomic Absorption Spectrometer (Perkin-

281

Elmer, Waltham, MA) equipped with a sodium hollow cathode lamp (Perkin-Elmer

282

Intensitron Part# 303-6065, Perkin-Elmer Waltham, MA).

283

3. Results and discussion Table 3: Experimentally determined transport properties for the TFC FO membrane.

Water Permeance 2000 ppm Intrinsic Sodium Chloride Rejection (%) Sodium Chloride Permeability

4.25 ± 0.04 L·m-2·hr-1·bar-1 99.2 ± 0.2 % 0.38 ± 0.11 L·m-2·hr-1

ACS Paragon Plus Environment

Page 16 of 40

Page 17 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Effective Structural Parameter

483 ± 79 Μm

284 285

ACS Paragon Plus Environment

Page 17 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 40

286

Figure 1: Water flux (a) and sodium chloride reverse flux (b). NaCl draw, DI feed, 0.25 m/s, and 20°C.

287 288

3.1 Membrane performance

289

Basic benchmark values for this membrane are shown in Table 3. Water permeance,

290

sodium chloride rejection, and sodium chloride permeability were measured directly

291

using reverse osmosis. The membranes effective structural parameter was calculated

292

based on the observed osmotic water fluxes in both the PRO and FO membrane

293

orientations using Eq. (2) and Eq. (3) respectively. Comparing these to published values

294

for other TFC membranes, the TFC membrane was observed having a higher water

295

permeance compared to literature values for other TFC FO membranes’ that were shown

296

having a solute permeability less than 0.5 L·m-2·hr-1

297

osmosis between 20 – 25°C. This membrane presents comparatively high observed water

298

flux greater than 50 L·m-2·hr-1 and 20 L·m-2·hr-1 in the PRO and FO orientations as

299

shown in Figure 1a.

300

3.2. Membrane structural properties

57,58,24

ACS Paragon Plus Environment

when tested in reverse

Page 18 of 40

Page 19 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

301

Industrial & Engineering Chemistry Research

3.2.1. Scanning electron micrographs

302

ACS Paragon Plus Environment

Page 19 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 40

Figure 2: Scanning electron microscope images of the TFC.

303

ACS Paragon Plus Environment

Page 20 of 40

Page 21 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

304

SEMs of the TFC membrane, shown in Figure 2, show the selective layer

305

morphology and internal structure of this membrane. The selective layer morphology

306

resembles the commonly noted ridge and valley structure consistent with an aromatic

307

polyamide.59,60 The PSu support of this membrane is thin, having a thickness of around

308

35 µm. The bottom 15 – 20 µm of the membrane support’s pore structure has macrovoids

309

with a spongey pore structure consisting of most of the upper half of the support layer.

310

The porous polymer support layer for this membrane is noticeably thinner than other flat

311

sheet TFC membranes engineered for FO typically encountered in literature.29,58 Thin

312

support layers are important factor in minimizing membrane structural parameter, which

313

limits water flux in engineered osmosis processes from increased severity of ICP. Only

314

TFC membranes which built upon nanofibrous supports have been shown to have thinner

315

porous mid-layers.26,35,30

316

3.2.2. Membrane support porosity and pore diameters

317

ACS Paragon Plus Environment

Page 21 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 40

Figure 3: MIP data for the TFC membrane’s support layer (a) shows the porosity of the support layer with the PET fabric layer, without the PET fabric layer, and of only the PET fabric layer. (b), (c), and (d) show the pore diameter distributions for this membrane’s support layer with the PET fabric layer, without the PET fabric layer, and of only the PET fabric layer respectively.

318 319

Support layer porosity, measured using mercury MIP, is shown in Figure 3a. The

320

measured porosity of complete support layer of the TFC membrane is approximately

321

65%. For comparison, the porosity of the TFC’s PSu support layer was measured with the

322

PET layer removed. Also the porosity of just the PET was measured coinciding with the

323

porosity measured for the complete membrane structure (Figure 3d). This is an

ACS Paragon Plus Environment

Page 22 of 40

Page 23 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

324

interesting finding as it suggests that the porosity of the PET is dominant in the porosity

325

of the complete structure. It necessary to note that this method may have an inherent bias

326

toward lower porosities given that the high pressures can cause compaction of soft

327

materials and the throttling of intruded mercury whereby larger pores are registered as

328

smaller pores by the ink-bottle effect.43

329

Figures 3b, 3c, and 3d illustrate pore volume contributions to pore diameter for this

330

membrane’s support layers. It is noteworthy that upon removal of the PET layer the

331

membranes pore diameter distribution fails to change significantly, except for a slight

332

increase in pore volume contribution for pore diameters from 1-2 µm. This would suggest

333

that the PET layer does not significantly enhance the porosity of the TFC membrane.

334

Such a contribution would be presented as a spike to the pore volume contributions for

335

pore diameters of 20-50 µm for the complete membrane structure (Figure 3b) when

336

compared to the membrane with the PET removed (Figure 3c). The opposite behavior can

337

be observed in Figure 3a where the porosity of the PET appears to decrease or at least

338

have a dominant contribution to the porosity of the complete support layer.

339

3.3. Membrane surface properties and chemistry

340

3.3.1. Membrane surface contact angles

341

ACS Paragon Plus Environment

Page 23 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 40

Figure 4: Contact angle of differing membrane layer’s using the sessile drop method with deionized water and a droplet size of 7±1 µL.

342 343

Contact angles for the membrane surfaces are shown in Figure 4. The selective layer

344

of this membrane is the most hydrophilic surface in the structure. This is to be expected

345

as a result of hydrogen bonding sites for water from carboxylic acid functional groups

346

and the amide bonds within the chemical structure of the selective layer. 61 The support

347

layers both with and without the PET layer are more hydrophobic. A lower contact angle

348

for the membrane with the PET layer attached is likely the result of some wicking into

349

the large pores of the PET layer. The PSu layer displays the hydrophobic character

350

reported by others24 for TFC membranes. This inherent hydrophobicity explains the need

351

for pre-wetting the membrane with an alcohol prior to FO and PRO testing as illustrated

352

in this study and others who have worked with varying iterations of this structure.23,33

353

3.3.2. FTIR spectra

354

ACS Paragon Plus Environment

Page 24 of 40

Page 25 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Figure 5: FTIR spectra of the membrane’s selective layer from 3900 to 2400 cm-1 and 1800 to 600 cm1 .

355 356

The membranes selective layer has a FTIR spectrum, shown in Figure 5, attributable

357

to aromatic polyamide. The peaks visible from 1700-1500 cm-1 capture stretching

358

vibrations attributable to C=O, C−N, and −CO−NH− bonds existing within the

359

polyamide.62,63 Also notable are the peaks occurring between 3000-2500 cm-1 which can

360

be attributed to –OH stretching amongst carboxylic acid functional groups;6,63,64 however,

361

the peak at 2900 cm-1 peaks can also be attributed to the −CH3CCH3− groups within

362

polysulfone.62 Carboxylic acid functionality within the polyamide would be significant as

363

prior work by Arena et al. hypothesized the contribution of this functional group to a

364

cation exchange behavior observed for membrane containing a polyamide selective layer

365

when using electrolyte draw solutions.6 The peak at 2900 cm-1 that can be attributed to

366

the −CH3CCH3− and peaks within this spectra attributable to the –SO2− of a sulfone at

367

~1120 and ~1340 suggest that the membrane is built upon a polysulfone support

368

layer.62,63

ACS Paragon Plus Environment

Page 25 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

369

3.4. Osmotic performance

370

3.4.2. Pressure retarded osmosis testing

Page 26 of 40

371

Figure 6: Membrane performance in PRO showing water flux (a), power density (b), structural parameter (c), and sodium chloride reverse solute flux (d). Test conditions 0.5 M NaCl draw, DI feed, 0.25 m—s-1 draw cross-flow, and 20°C. Assumptions incorporated in the calculation of these values are in Table 1.

372 373

Experimentally measured water fluxes and supporting simulated values for PRO are

374

shown in Figure 6a. The power densities corresponding to those water fluxes

ACS Paragon Plus Environment

Page 26 of 40

Page 27 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

375

observed/calculated in Figure 6a are shown in Figure 6b. The measured water flux and

376

power density are much lower than those simulated for the ideal case of constant

377

structural parameter and no external polarization. Upon the inclusion ECP, numerical

378

simulation of water fluxes still overpredicts power densities observed experimentally,

379

suggesting that the significant difference between the experimental and simulated water

380

flux is from changing support layer structural parameter caused by compaction. Figure 6c

381

shows how the structural parameter sharply increases over a range of transmembrane

382

pressures. These changes are especially prominent at pressures above 4.5 bar. Prior

383

studies comparing experimental and theoretical PRO performance have also observed

384

rapid flux decline with increasing transmembrane pressure. 11,35

385

While this effect appears to rapidly drop the flux and increases the effective structural

386

parameter of the membrane it does not seem to be a permanent effect with the observed

387

water flux, power density and structural parameter (Figures 6a, 6b, and 6c) all returning

388

to near their initial values on a reduction of the applied pressure. Only the observed salt

389

flux changed as a result of membrane compaction (Figure 6d), being lower along the

390

descending pressure ramp as compared to the ascending pressure ramp. This could be

391

related to compaction/fouling of the selective layer which in RO has been observed to

392

increase the rejection of the membrane in long term tests.69

393

The observed increase in structural parameter (Figure 6c) is likely the result of the

394

applied hydrostatic pressure compacting the membrane’s support layer. This compaction

395

will reduce the thickness of the support layer and likely collapse some of the membrane’s

396

internal pore structure.37 The loss of support layer porosity is directly competitive with

397

the reduced thickness of the support.

ACS Paragon Plus Environment

Page 27 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 40

398

The impact of decreasing porosity on tortuosity is not clear because the classical

399

relationship between porosity and tortuosity, the Bruggemann relation, is an empirical

400

relationship shown in a generalized form in Eq.(8).65

τ = γε1−α (8)

401 402

γ and α are constants relating to the morphology of the material it is possible to

403

develop a conceptual understanding on how the tortuosity of porous membrane materials

404

are affected by compaction. In general values for both γ and α are >1.65 This suggests that

405

for the range of common values in the Bruggemann relation a loss of porosity within the

406

membrane support layers would result in increases to the support layer tortuosity (i.e. for

407

γ=1 and α=1.6 a decrease in porosity from 70% to 60% would increase tortuosity from

408

~1.24 to ~1.36). The common range of values for the Bruggeman relation should only

409

dictate the magnitude of tortuosity increase and not whether to tortuosity will increase

410

with a loss of porosity.

411

Improved compaction tolerance could be implemented through forming membrane

412

upon a stiffer support structure which will not compact in PRO or one which when under

413

compaction does not demonstrate a rapid increase in the membranes structural parameter.

414

Support layers with a spongey structure are viewed by some as having greater tolerance

415

to compaction making them more suitable for PRO.37,66

416 417

3.4.3. Ammonia-carbon dioxide water flux and rejection

418

ACS Paragon Plus Environment

Page 28 of 40

Page 29 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Figure 7: Water flux of the membrane against a 0.25 M NaCl feed using a 2 M NH3-CO2 (based on carbon species) draw of varying NH3:CO2.

419 420

ACS Paragon Plus Environment

Page 29 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 40

421 Table 4: Speciation of NH3-CO2 draw solution at 23°C. 1.2:1 2 MCO₂

1.5:1 2 MCO₂

2:1 2 MCO₂

pH

8.2

8.6

9.0

mtotal-N (mol/kgH₂O)

2.36

2.98

4.06

mtotal-C (mol/kgH₂O)

2.06

2.06

2.07

ρsolution (kg/L)

1.058

1.059

1.055

mNH₃ (mol/kgH₂O)

0.0612

0.213

0.678

mCO₂ (mol/kgH₂O)

0.0632

0.0163

0.00304

mNH₄⁺ (mol/kgH₂O)

2.04

2.08

2.10

mHCO₃⁻ (mol/kgH₂O)

1.77

1.34

0.758

mCO₃²⁻ (mol/kgH₂O)

0.0113

0.0250

0.0374

mNH₂COO⁻ (mol/kgH₂O)

0.255

0.686

1.27

Ionic strength

2.05

2.10

2.14

π (bar)

50.1

74.2

116

422 423

The water fluxes observed from desalination tests using the NH3-CO2 draw solution

424

with differing NH3:CO2 ratios are shown in Figure 7. When compared to similar studies

425

using this draw solution (with a 1.2:1 NH3:CO2) the observed water fluxes were higher

426

than those observed using polydopamine (PDA) modified TFC RO membranes.6,7,8

427

Coinciding with observations made by Arena, there was a substantial decrease in flux

428

when modest amounts of salt were added to the feed (0.25M).6 A decrease in water flux

429

was observed when the NH3:CO2 ratio what changed from 1.2:1 to 1.5:1. The observed

430

water flux increased for the 2:1 ratio compared to 1.2:1 ratio. The overall change in water

431

flux was approximately ±5 L·m-2·h-1 when using a feed solution of DI water for the three

432

NH3:CO2 ratios studied.

ACS Paragon Plus Environment

Page 30 of 40

Page 31 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

433

The drop in water flux for a 1.5:1 ratio is observed in spite of the higher osmotic

434

pressure of this draw solution (Table 4). This suggests that the increased osmotic pressure

435

cannot be effectively used by the membrane, possibly as a result of an uneven permeation

436

of draw solutes through the selective layer. Ammonia is present in substantial quantities

437

at higher ratios, and this species readily passes through the selective layer due to its

438

similarity with water in size and polarity.67 As ammonia that freely crosses the membrane

439

does not contribute to the osmotic driving force and could alter the speciation of the draw

440

solution at the membrane’s selective layer.

441

The multicomponent nature of this draw solution complicates even a qualitative

442

analysis of draw solute speciation. In general, the NH3:CO2 draw solution appears to

443

perform best at extremes with a minimal NH3:CO2 (close to 1:1, having a low amount of

444

aqueous NH3) or with a large excess of NH3 in solution, having more than 20% total

445

molality nitrogen species as aqueous NH3. At the low ratios, the most abundant nitrogen

446

species is NH4+, which will not easily cross the membrane. At higher NH3:CO2 ratios the

447

draw solution comprises increasing amounts of carbamate and dissolved ammonia. Large

448

amounts of dissolved ammonia contribute significantly to the draw solution’s osmotic

449

pressure even though some of the dissolved ammonia diffuses across the membrane.

450

From an operational perspective, having excess ammonia might be preferred to increase

451

the stability and solubility of the solution.

452

ACS Paragon Plus Environment

Page 31 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 40

Figure 8: Ion fluxes (a) and feed ion rejection (b) of a 0.25 M NaCl feed solution using a 2 M NH3CO2 draw solution of varying NH3:CO2 ratios. Ammonia species flux is the reverse draw solute flux, going from the draw solution to the feed solution, while the sodium flux and chloride flux are forward solute fluxes going from the feed solution into the draw solution in the same direction as the water flux.

453 454

The forward (sodium cations and chloride anions) and reverse (total ammonia

455

species, consisting of ammonia, ammonium, and carbamate) fluxes are shown in Figure

456

8a. For all NH3:CO2 ratios, cation fluxes (sodium and ammonium) were statistically

457

similar. In all instances the ammonium to sodium fluxes water greater than or statistically

458

identical to the sodium flux observed for all NH3:CO2 ratios. Ammonia species flux is

459

higher for the 1.5:1 NH3:CO2 draw solution. This further supports the hypothesis that the

460

low water fluxes occurs due to greater ammonia flux into the feed solution. For the 2:1

461

ratio, the ammonia species flux stays the same, indicating that the additional dissolved

462

ammonia effectively contributes to the osmotic driving force. The 1.2:1 NH3:CO2 draw

463

solution has the lowest dissolved ammonia (gas) concentration (less than 0.1 mol·kg-1)

464

and exhibits the lowest flux of ammonia species. Due to the low dissolved ammonia

ACS Paragon Plus Environment

Page 32 of 40

Page 33 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

465

concentration, nearly all of ammonia species flux must occur from the transport of

466

ammonium.

467

Large differences in the forward cation and anion fluxes are shown in Figure 8b.

468

Sodium flux is an order of magnitude higher than the chloride flux for all of the NH3:CO2

469

ratios tested. The lower chloride flux suggests that anion transport is not necessary to

470

maintain electroneutrality between the feed and draw solution, suggesting that sodium-

471

ammonium cation exchange is occurring. A similar result was reported in previous work

472

by Arena et al. on PDA modified TFC RO membranes used with this draw solution (only

473

in a 1.2:1 NH3:CO2).6 Lu et al. also observed the occurrence of ammonium/sodium

474

exchange across the selective layer of another iteration of this membrane.68 This work

475

also showed that the rate of ammonium and sodium transport can be reduce through a

476

surface modification of the selective layer which converts carboxylic acid functional

477

groups to amine reactive esters, which could then be reacted with aqueous

478

ethylenediamine,

479

CONHCH2CH2NH2.68

effectively

converting

the

carboxylic

acid

(−COOH)

to



480

The lower forward flux of anions to cations is a result of inherent negative surface

481

functionality common to TFC membrane polyamide layers due to the presence of

482

carboxylic acid functional groups which form from the hydrolysis of acid chloride groups

483

from the interfacial polymerization.6 This behavior is not unique to the NH3-CO2 draw

484

solution as shown in prior study by Coday et al. This study, which also used an iteration

485

of this membrane platform, employed a NaCl draw solution and a feed of mixed

486

electrolytes observed different rates of cation and anion transport through TFC

487

membranes.33

ACS Paragon Plus Environment

Page 33 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 40

488

These data show that while productivity (or water flux) is an important component in

489

forward osmosis desalination interactions between draw solutes, feed solutes, and the

490

membrane can lower product water quality and complicate draw solution recovery.

491

Mitigating the impact of sodium/ammonium cation exchange has been incorporated into

492

the design of Oasys Water’s osmotic brine concentrator.34 Their refinements include a

493

stripper for the feed solution to recover draw solutes that cross the membrane through

494

diffusion and ion exchange. Additionally, there is an RO polisher following draw solute

495

recovery to remove the cations that cross the membrane from ion exchange. The RO

496

permeate is clean water. The RO concentrate, containing nonvolatile cations and anions

497

from the feed and draw solution can be blended with the pretreated feed solution. Draw

498

solute anions blended with the feed solution are removed with draw solute cations in the

499

feed solution stripper.

500

4. Conclusions

501

This early generation of Oasys Water’s TFC membrane was demonstrated to have

502

high water flux in FO and PRO operation. Data for PRO conditions illustrated evidence

503

of membrane compaction under pressure. Evidence of ion exchange between ammonium

504

and sodium was found in FO using the NH3-CO2 draw solution. These findings suggest

505

that the need of low structural parameter support layers for FO and PRO now competes

506

with the development of membranes which are compaction tolerant and/or ion exchange

507

resistant.

508

Associated Content

509

Supporting information

ACS Paragon Plus Environment

Page 34 of 40

Page 35 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

510

The mathematica notebook with a sample output used to determine the speciation of

511

the dissolved ammonia-carbon dioxide draw solution. The Supporting Information is

512

available free of charge on the ACS Publications website at pubs.acs.org.

513

Acknowledgements

514

The authors acknowledge funding from the NSF CBET Chemical and Biological

515

Separations Program # 1160098 and #1160069, and USEPA (Project No. R834872). The

516

authors also acknowledge the NWRI-AMTA Fellowship for Membrane Technology and

517

National Science Foundation GK-12 Program, which provided support for Jason T.

518

Arena. The authors also wish to thank Oasys Water for providing the membranes used

519

for this study. Additionally, the authors would like to thank Dr. Abhay Vaze at the

520

University of Connecticut for permitted use of the atomic absorption spectrometer.

521 522

References (1) Cath, T. Y.; Childress, A. E.; Elimelech, M. Forward osmosis: Principles, applications, and recent developments. J. Membr. Sci. 2006, 281, 70-87. (2) Hoover, L. A.; Phillip, W. A.; Tiraferri, A. ; Yip, N. Y.; Elimelech, M. Forward with Osmosis: Emerging Applications for Greater Sustainability. Environ. Sci. Technol. 2011, 45, 9824–9830. (3) McGinnis, R. L.; Elimelech, M. Global challenges in energy and water supply: the promise of engineered osmosis. Environ. Sci. Technol. 2008, 42, 8625-8629. (4) Chung, T.-S. ; Zhang, S. ; Wang, K. Y.; Su, J. ; Ling, M. M. Forward osmosis processes: Yesterday, today, and tomorrow. Desalination 2012, 287, 78-81. (5) Klaysom, C. ; Cath, T. Y.; Depuydt, T. ; Vankelecom, I. F. J. Forward and pressure retarded osmosis: potential solutions for global challenges in energy and water supply. Chem. Soc. Rev. 2013, 42, 6959-5989. (6) Arena, J. T.; Manickam, S. S.; Reimund, K. K.; Freeman, B. D.; McCutcheon, J. R. Solute and water transport in forward osmosis using polydopamine modified thin film composite membranes. Desalination 2014, 343, 8-16. (7) McCutcheon, J. R.; McGinnis, R. L.; Elimelech, M. A novel ammonia-carbon dioxide forward (direct) osmosis desalination process. Desalination 2005, 174, 1-11. (8) McCutcheon, J. R.; McGinnis, R. L.; Elimelech, M. Desalination by ammoniacarbon dioxide forward osmosis: Influence of draw and feed solution concentrations

ACS Paragon Plus Environment

Page 35 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 40

on process performance. J. Membr. Sci. 2006, 278, 114-123. (9) Garcia-Castello, E. M.; McCutcheon, J. R. Dewatering press liquor derived from orange production by forward osmosis. J. Membr. Sci. 2011, 372, 97-101. (10) McGinnis, R. L.; McCutcheon, J. R.; Elimelech, M. A novel ammonia–carbon dioxide osmotic heat engine for power generation. J. Membr. Sci. 2007, 305, 13-19. (11) She, Q. ; Jin, X. ; Tang, C. Y. Osmotic power production from salinity gradient resource by pressure retarded osmosis: Effects of operating conditions and reverse solute diffusion. J. Membr. Sci. 2012, 401-402, 262-273. (12) Straub, A. P.; Yip, N. Y.; Elimelech, M. Raising the Bar: Increased Hydraulic Pressure Allows Unprecedented High Power Densities in Pressure-Retarded Osmosis. Environ. Sci. Technol. Lett. 2014, 1, 55-59. (13) Stone, M. L.; Rae, C. ; Stewart, F. F.; Wilson, A. D. Switchable polarity solvents as draw solutes for forward osmosis. Desalination 2013, 312, 124-129. (14) Herron, J. Asymmetric forward osmosis membranes. United States Patent No. US 7,445,712, Nov. 4, 2008. (15) Cath, T. Y.; Adams, D. ; Childress, A. E. Membrane contactor processes for wastewater reclamation in space II. Combined direct osmosis, osmotic distillation, and membrane distillation for treatment of metabolic wastewater. J. Membr. Sci. 2005, 257, 111-119. (16) Cath, T. Y.; Gormly, S. ; Beaudry, E. G.; Flynn, M. T.; Adams, V. D.; Childress, A. E. Membrane contactor processes for wastewater reclamation in space Part I. Direct osmotic concentration as pretreatment for reverse osmosis. J. Membr. Sci. 2005, 257, 85-98. (17) Cath, T. Y.; Hancock, N. T.; Lundin, C. D.; Hoppe-Jones, C. ; Drewes, J. E. A multi-barrier osmotic dilution process for simultaneous desalination and purification of impaired water. J. Membr. Sci. 2010, 362, 417-436. (18) Ge, Q. ; Su, J. ; Chung, T.-S. ; Amy, G. Hydrophilic Superparamagnetic Nanoparticles: Synthesis, Characterization, and Performance in Foward Omosis Processes. Ind. Eng. Chem. Res. 2011, 50, 382-288. (19) Ling, M. M.; Chung, T.-S. Desalination process using super hydrophilic nanoparticles via forward osmosis. Desalination 2011, 278, 194-202. (20) Baker, R. W. Membrane Technology and Applications: Second Edition; John Wiley & Sons Ltd: West Sussex, 2004. (21) Vos, K. D.; Burris, F. O.; Riley, R. L. Kinetic Study of the Hydrolysis of Cellulose Acetate in the pH Range of 2-10. J. Appl. Polym. Sci. 1966, 10, 825-832. (22) Watters, J. C.; Klein, E. ; Fleischman, M. ; Roberts, J. S.; Hall, B. Rejection Spectra of Reverse Osmosis Membranes Degraded by Hydrolysis of Chlorine Attack. Desalination 1986, 60, 93-110. (23) Cath, T. Y.; Elimelech, M. ; McCutcheon, J. R.; McGinnis, R. L.; Achilli, A. ; Anastasio, D. ; Brady, A. R.; Childress, A. E.; Farr, I. V.; Hancock, N. T.; Lampi, J. ; Nghiem, L. D.; Xie, M. ; Yip, N. Y. Standard Methodology for Evaluating Membrane Performance in Osmotically Driven Membrane Processes. Desalination 2013, 312, 31-38.

ACS Paragon Plus Environment

Page 36 of 40

Page 37 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

(24) Arena, J. T.; McCloskey, B. ; Freeman, B. D.; McCutcheon, J. R. Surface modification of thin film composite membrane support layers with polydopamine: Enabling use of reverse osmosis membranes in pressure retarded osmosis. J. Membr. Sci. 2011, 375, 55-62. (25) Bui, N. N.; Lind, M. L.; Hoek, E. M. V.; McCutcheon, J. R. Electrospun nanofiber supported thin film composite membranes for engineered osmosis. J. Membr. Sci. 2011, 385-386, 10-19. (26) Bui, N. N.; McCutcheon, J. R. Hydrophilic Nanofibers as New Supports for Thin Film Composite Membranes for Engineered Osmosis. Environ. Sci. Technol. 2013, 47, 1761-1769. (27) Ren, J. ; McCutcheon, J. R. A new commercial thinfilm composite membrane for forward osmosis. Desalination 2014, 343, 187-193. (28) Wang, R. ; Shi, L. ; Tang, C. Y.; Chou, S. ; Qui, C. ; Fane, A. G. Characterization of novel forward osmosis hollow fiber membranes. J. Membr. Sci. 2010, 355, 158-167. (29) Yip, N. Y.; Tiraferri, A. ; Phillip, W. A.; Schiffman, J. D.; Elimelech, M. High Performance Thin Fillm Composite Forward Osmosis Membrane. Environ. Sci. Technol. 2010, 44, 3812-38-18. (30) Song, X. ; Lui, Z. ; Sun, D. D. Nano Gives the Answer: Breaking the Bottleneck of Internal Concentration Polarization with a Nanofiber Composite Forward Osmosis Membrane for a High Water Production Rate. Adv. Mater. 2011, 23, 3256-3260. (31) Huang, L. ; McCutcheon, J. R. Hydrophilic nylon 6,6 nanofibers supported thin film composite membranes for engineered osmosis. J. Membr. Sci. 2014, 457, 162-169. (32) Hoover, L. A.; Schiffman, J. D.; Elimelech, M. Nanofibers in thin-film composite membrane support layers: Enabling expanded application of forward and pressure retarded osmosis. Desalination 2013, 308, 73-81. (33) Coday, B. D.; Heil, D. M.; Xu, P. ; Cath, T. Y. Effects of Transmembrane Hydraulic Pressure on Performance of Forward Osmosis Membranes. Environ. Sci. Technol. 2013, 47, 2386-2393. (34) McGinnis, R. L.; Hancock, N. T.; Nowosielski-Slepowron, M. S.; McGurgan, G. D. Pilot Demonstation of the NH3/CO2 forward osmosis desalination process on high salinity brines. Desalination 2013, 312, 67-74. (35) Bui, N. N.; McCutcheon, J. R. Nanofiber Supported Thin-Film Composite Membrane for Pressure-Retarded Osmosis. Environ. Sci. Technol. 2014, 48, 41294136. (36) 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. (37) Han, G. ; Zhang, S. ; Li, X. ; Chung, T. S. High performance thinfilm composite pressure retarded osmosis (PRO) membranes for renewable salinity-gradient energy generation. J. Membr. Sci. 2013, 440, 108-121. (38) Song, X. ; Lui, Z. ; Sun, D. D. Energy recovery from concentrated seawater brine by thin-film nanofiber composite pressure retarded osmosis membranes with high power density. Energy Environ. Sci. 2013, 6, 1199-1210.

ACS Paragon Plus Environment

Page 37 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 40

(39) McGinnis, R. ; McGurgan, G. Forward osmosis membranes. United States Patent No. US 8,181,794, May 22, 2012. (40) Comesana, J. F.; Otero, J. J.; Camesella, E. ; Correa, A. Densities and Viscosities of Ternary Systems of Water + Fructose + Sodium Chloride from 20 to 40 °C. J. Chem. Eng. Data 2001, 46, 1153-1155. (41) Lobo, V. M. M. Mutual Diffusion Coefficients in Aqueous Electrolyte Solutions. Pure Appl.Chem. 1993, 65, 2613-2640. (42) McCabe, W. L.; Smith, J. C.; Harriot, P. Unit Operations of Chemical Engineering, Seventh Edition ed.; McGraw-Hill: New York, 2005. (43) Manickam, S. S.; Gelb, J. ; McCutcheon, J. R. Pore structure characterization of asymmetric membranes: Non-destructive characterization of porosity and tortuosity. J. Membr. Sci. 2014, 454, 549-554. (44) Yip, N. Y.; Tiraferri, A. ; Phillip, W. A.; Schiffman, J. D.; Hoover, L. A.; Kim, Y. C.; Elimelech, M. Thin-Film Composite Pressure Retarded Osmosis Membranes for Sustainable Power Generation from Salinity Gradients. Environ. Sci. Technol. 2011, 45, 4360-4369. (45) Tiraferri, A. ; Yip, N. Y.; Straub, A. P.; Castrillon, S. R.-V.; Elimelech, M. A method for the simultaneous determination of transport and structural parameters of forward osmosis membranes. J. Membr. Sci. 2013, 444, 523-538. (46) Yip, N. Y.; Elimelech, M. Performance Limiting Effects in Power Generation from Salinity Gradients by Pressure Retarded Osmosis. Environ. Sci. Technol. 2011, 45, 10273-10282. (47) Anastasio, D. D.; Arena, J. T.; Cole, E. A.; McCutcheon, J. R. Impact of temperature on power density in closed-loop pressure retarded osmosis for grid storage. J. Membr. Sci. 2015, 479, 240-245. (48) Kawazuishi, K. ; Prausnitz, J. M. Correlation of Vapor-Liquid Equilibria for the System of Ammonia-Carbon Dioxide-Water. Ind. Eng. Chem. Res. 1987, 26, 14821485. (49) Edwards, T. J.; Maurer, G. ; Newman, J. ; Prausnitz, J. M. Vapor-Liquid Equilibria in Multicomponent Aqueous Solutions of Volatile Weak Electrolytes. AIChE J. 1978, 24, 966-976. (50) Pawlikowski, E. M.; Newman, J. ; Prausnitz, J. M. Phase Equilibria for Aqueous Systems of Ammonia and Carbon Dioxide. Ind. Eng. Chem. Proc. Des. Dev. 1982, 21, 764-770. (51) Buetler, D. ; Renon, H. Representation of NH3-H2S-H2O, NH3-CO2-H2O, and NH3-SO2-H2O Vapor-Liquid Equilibria. Ind. Eng. Chem. Proc. Des. Dev. 1978, 17, 220-230. (52) Robinson, R. A.; Stokes, R. H. Electrolyte Solutions, Second Revised ed.; Dover Publications: Mineola, 2002. (53) Prausnitz, J. M.; Lichtenthaler, R. N.; de Azevedo, E. G. Molecular Thermodynamics of Fluid-Phase Equilibria, 3rd ed.; Prentice-Hall, Inc.: Upper Saddle River, 1999. (54) Grattoni, A. ; Merlo, M. Osmotic Pressure beyond Concentration Restrictions. J.

ACS Paragon Plus Environment

Page 38 of 40

Page 39 of 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

Phys. Chem. Part B 2007, 111, 11770-11775. (55) Jeffrey, G. H.; Bassett, J. ; Mendham, J. ; Denny, R. C. Vogel's Textbook of Quantitative Chemical Analysis, Fifth Edition ed.; Longman Scientific & Tehcnical: Essex, 1989. (56) Harris, D. C. Quantitative Chemial Analysis, Sixth Edition ed.; W.H. Freeman and Company: New York, 2003. (57) Shi, L. ; Chou, S. R.; Wang, R. ; Fang, W. X.; Tang, C. Y.; Fane, A. G. Effect of substrate structure on the performance of thin-film composite forward osmosis hollow fiber membranes. J. Membr. Sci. 2011, 382, 116-123. (58) Tiraferri, A. ; Yip, N. Y.; Phillip, W. A.; Schiffman, J. D.; Elimelech, M. Relating performance of thin-film composite forward osmosis membranes to support layer formation and structure. J. Membr. Sci. 2011, 367, 340-352. (59) Kwak, S. Y.; Jung, S. G.; Yoon, Y. S.; Ihm, D. W. Details of Surface Features in Aromatic Polyamide Reverse Osmosis Membranes Characterized by Scanning Electron and Atomic Force Microscopy. J. Polym. Sci., Part B: Polym. Phys. 1999, 37, 1429-1440. (60) Petersen, R. J. Composite reverse osmosis and nanofiltration membranes. J. Membr. Sci. 1993, 83, 81-150. (61) Green, M. M.; Wittcoff, H. A. Organic Chemistry Principles and Industrial Practice, 1st ed.; Wiley-VCH: Weiheim, 2003. (62) Tang, C. Y.; Kwon, Y.-N.; Leckie, J. O. Effect of membrane chemistry and coating layer on physiochemical properties of thin film composite polyamide RO and NF membranes I. FTIR and XPS characterization of polyamide and coating layer chemistry. Desalination 2009, 242, 149-167. (63) Lambert, J. B.; Shurvell, H. F.; Lightner, D. A.; Cooks, R. G. Organic Structural Spectroscopy; Prentice-Hall, Inc.: Upper Saddle River, 1998. (64) Speight, J. G. Lange's Handbook of Chemistry, Sixteenth ed.; McGraw Hill: New York, 2005. (65) Thorat, I. V.; Stephenson, D. E.; Zacharias, N. A.; Zaghib, K. ; Harb, J. N.; Wheeler, D. R. Quantifying tortuosity in porous Li-ion battery materials. J. Power Sources 2009, 188, 592-600. (66) Widjojo, N. ; Chung, T. S.; Weber, M. ; Maletzko, C. ; Warzelhan, V. The role of sulphonated polymer and macrovoid-free structure in the support layer for thin-film composite (TFC) forward osmosis (FO) membranes. J. Membr. Sci. 2011, 383, 214223. (67) Hesske, H. ; Gloe, K. Hydration Behavior of Alkyl Amines and Their Corresponding Protonated Forms. 1. Ammonia and Methylamine. J. Phys. Chem. 2007, 111, 98489853. (68) Lu, X.; Boo, C. ; Ma, J. ; Elimelech, M. Bidirectional Diffusion of Ammonium and Sodium Cations in Forward Osmosis: Role of Membrane Active Layer Surface Chemistry and Charge. Environ. Sci. Technol. 2014, 48, 14369-14376.

524

ACS Paragon Plus Environment

Page 39 of 40

Industrial & Engineering Chemistry Research

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

525 526 527 528

Page 40 of 40

(69) Van Wagner, E. M.; Sagle, A. C.; Sharma, M. M.; Freeman, B.D. Effect of crossflow testing conditions, including feed pH and continuous feed filtration, on commercial reverse osmosis membrane performance. J. Membr. Sci. 2009, 345, 97109.

ACS Paragon Plus Environment

Page 40 of 40