A Gas-permeable Membrane-based Conductivity ... - ACS Publications

Subscriber access provided by University of Massachusetts Amherst Libraries

Article

A Gas-permeable Membrane-based Conductivity Probe Capable of In Situ Real-time Monitoring Ammonia in Aquatic Environments Tianling Li, Jared Panther, Yuan Qiu, Chang Liu, jianyin huang, Yonghong Wu, Po Keung Wong, Taicheng An, Shanqing Zhang, and Huijun Zhao Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b03552 • Publication Date (Web): 25 Oct 2017 Downloaded from http://pubs.acs.org on October 25, 2017

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 35

Environmental Science & Technology

1

A Gas-permeable Membrane-based Conductivity

2

Probe Capable of In Situ Real-time Monitoring

3

Ammonia in Aquatic Environments

4

Tianling Lia, Jared Panthera,b, Yuan Qiua,c, Chang Liua,d, Jianyin Huanga,e, Yonghong Wuf, Po Keung

5

Wongg, Taicheng Anh*, Shanqing Zhanga*, Huijun Zhaoa*

6

a

7

Australia

8

b

Goulburn-Murray Water, VIC, 3616, Australia

9

c

Guangxi Vocational and Technical Institute of Industry, 37 Xiuling Road, Nanning, Guangxi,

Centre for Clean Environment and Energy, Gold Coast Campus, Griffith University, QLD, 4222,

10

530005, China

11

d

Department of Chemistry, Liaoning Medical University, 40 Songpo Road, Jinzhou, 121001, China

12

e

Division of Information Technology, Engineering and Environment, School of Natural and Built

13

Environment, Mason Lakes Campus, University of South Australia, SA, 5095, Australia

14

f

15

of Sciences, No.71, East Beijing Rd, Nanjing, 210008, China

16

g

School of Life Sciences, The Chinese University of Hong Kong, Shatin, NT, Hong Kong SAR, China

17

h

Institute of Environmental Health and Pollution Control, School of Environmental Science and

18

Engineering, Guangdong University of Technology, Guangzhou, 510006, China

State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy

19

1

ACS Paragon Plus Environment


20

*Corresponding Author

21

Prof. Huijun Zhao, Tel: +61-7-55528261, Fax: +61-7-55528067, E-mail: [email protected]

22

Prof. Shanqing Zhang, Tel: +61-7-55528155, Fax: +61-7-55528067, E-mail: [email protected]

23

Prof. Taicheng An, Tel: +86-20-2388 3536, Fax: +86-20-2388 3536, E-mail: [email protected]

Page 2 of 35

24

2


Page 3 of 35


25

ABSTRACT

26

Aquatic ammonia has toxic effects on aquatic life. This work reports a gas-permeable

27

membrane-based conductivity probe (GPMCP) developed for real-time monitoring of

28

ammonia in aquatic environments. The GPMCP innovatively combines a gas-permeable

29

membrane with a boric acid receiving phase to selectively extract ammonia from samples and

30

form ammonium at the inner membrane interface. The rate of the receiving phase

31

conductivity increase is directly proportional to the instantaneous ammonia concentration in

32

the sample, which can be rapidly and sensitively determined by the embedded conductivity

33

detector. A pre-calibration strategy was developed to eliminate the need for an ongoing

34

calibration. The analytical principle and GPMCP performance were systematically validated.

35

The laboratory results showed that ammonia concentrations ranging from 2 to 50,000 µg L-1

36

can be detected. The field deployment results demonstrated the GPMCP’s ability to obtain

37

high resolution continuous ammonia concentration profiles and the absolute average

38

ammonia concentration over a prolonged deployment period. By inputting the temperature

39

and pH data, the ammonium concentration can be simultaneously derived from the

40

corresponding ammonia concentration. The GPMCP embeds a sophisticated analytical

41

principle with the inherent advantages of high selectivity, sensitivity and accuracy, and it can

42

be used as an effective tool for long-term, large-scale, aquatic-environment assessments.

43

3



Page 4 of 35

44

INTRODUCTION

45

As the third most abundant nitrogen species, ammonia is widely present in aquatic

46

environments. In addition to uncontrollable natural sources, such as the decomposition of

47

organic matter, animal and human wastes, bacterial nitrogen fixation and atmospheric

48

deposition, aquatic ammonia also comes from agricultural land runoffs and municipal

49

discharges.1-4 Aquatic ammonia can exist in both NH3 and NH4+ forms, and it can transform

50

into other nitrogen species, such as NO2-, NO3- and amino acids, through microorganism

51

nitrification processes and living organism assimilation processes.5, 6 Similar to other nitrogen

52

species, elevated concentrations of ammonia in aquatic environments can cause

53

eutrophication, which is a common water pollution phenomenon.3, 7, 8 However, in contrast to

54

other aquatic nitrogen species, the presence of ammonia, even at low concentrations, can

55

have direct, toxic effects on aquatic life, leading to adverse ecological consequences.9-13 With

56

the projected increase in nitrogen fertilizer usage (to boost agriculture productivity)14-16 and

57

atmospheric ammonia precipitation (as a result of the increased ammonia emissions from

58

automotive, chemical and manufacturing industries),17-21 aquatic ammonia pollution is

59

expected to worsen in the foreseeable future, and this will require effective environmental

60

management practices to mitigate the impact of increased aquatic ammonia pollution.22, 23 In

61

this regard, developing field-based analytical technologies that are capable of in situ real-time

62

monitoring ammonia in a facile and economically viable fashion to enable long-term, large-

63

scale evaluations of aquatic ammonia distribution, migration and trends will directly benefit

64

environmental management practices.4, 24

65

The standard Nessler method is a traditional and widely used laboratory ammonia detection

66

method.25 This method is a colorimetric-based method with a high sensitivity, but its

67

accuracy can be affected by the presence of metal ions and other nitrogen species.26-28 4


Page 5 of 35


68

Although a number of ammonia flow-injection online detection systems based on automated

69

colorimetric methods have been developed,29-31 few of these systems have been used for

70

field-based applications due to the high costs for the equipment, installation, operation and

71

maintenance. Additionally, the inherent reliability and stability issues due to the complex

72

flow-injection system configuration make these systems unsuitable for field-based

73

applications. An ion-selective ammonia electrode is another commonly used ammonia

74

detection method. The electrode possesses an ideal sensor configuration and embodiment for

75

in situ analyses, but it often suffers from insufficient sensitivity and reproducibility.

76

Additionally, and more critically, the electrode only functions under strictly controlled

77

analytical environments that rarely exist in natural environments.32,

78

overcome the shortfalls of the colorimetric and electrochemical based online systems, Chow

79

and co-workers developed an elegant microdistillation flow injection ammonia detection

80

system and successfully applied to control the chlorination dosage for drinking water

81

treatment.34, 35 They demonstrated that ultrasensitive detection of ammonia can be achieved

82

by measuring the conductivity changes in a boric acid solution that result from adsorbing

83

ammonia from a water sample via microdistillation.36, 37 Despite the noticeable advantages of

84

this method, it is not suitable for large-scale, aquatic-environment information collection

85

because of its inability to perform in situ measurements and its complex system configuration.

86

In practice, to meet the needs of large-scale aquatic-environment ammonia assessments, an

87

analytical system should be: (i) cheap to build, easy to deploy, low maintenance and low cost

88

for operation; (ii) highly reliable, sensitive, accurate and interference-free; (iii) capable of in

89

situ measurements without the need for an additional sample delivery system (e.g., pumping

90

system). To date, although different types of ammonia detection methods (e.g., biosensors

91

and fluorescence and chemiluminescence methods) have been reported and successfully used

33

In an effort to

5



Page 6 of 35

92

for various applications,38-43 to the best of our knowledge, none of the reported ammonia

93

detection methods could meet the criteria listed above.

94

Herein, we report a gas-permeable membrane-based conductivity probe (GPMCP) that meets

95

all the required criteria for long-term, large-scale, in situ, real-time aquatic ammonia

96

monitoring applications. The developed GPMCP innovatively combines a gas-permeable

97

membrane with a homemade, four-point conductivity detector through a small volume of

98

boric acid solution. The probe embeds a simple but effective analytical principle to enable in

99

situ selective ammonia detection (see the Analytical Principle section for details). The simple

100

probe-shaped configuration makes the GPMCP cheap to build and easy to deploy. Unlike the

101

microdistillation flow injection ammonia detection system,34-37 where the ammonia

102

containing sample must be pumped into a micro-boiler and mixed with NaOH to extract the

103

ammonia from the sample, and the extracted ammonia must be condensed and adsorbed

104

before the detection, the GPMCP can directly and continuously extract ammonia from

105

samples through the gas-permeable membrane via a simple neutralization reaction of

106

ammonia with boric acid and simultaneously realize the detection. This unique feature of the

107

GPMCP makes in situ, real-time ammonia monitoring possible. The analytical principles that

108

enable GPMCP to continuously monitor instantaneous ammonia concentrations and

109

determine the average ammonia concentration over a given deployment period were proposed

110

and experimentally validated. A pre-calibration strategy was also developed to eliminate the

111

need for an ongoing calibration for increased practicality and minimized operational and

112

maintenance costs. The performance of the developed GPMCP was evaluated using synthetic

113

samples and field deployments.

6


Page 7 of 35


114

EXPERIMENTAL SECTION

115

Chemicals, Solutions and Sample Analysis. All the chemicals used in this work were of AR

116

grade and were purchased from Merck. All the solutions were prepared using deionized water

117

(Millipore Corp., 18 MΩ cm). Sodium hydroxide was used to adjust the testing solution pH.

118

Unless otherwise stated, the receiving phase used in this work was 0.500 mol L-1 boric acid.

119

The ammonia nitrogen (NH3-N) was determined by the APHA Standard Method,25 where

120

needed, and the standard distillation method was used to achieve sample pre-concentration.25

121

Apparatus and Methods. A two-compartment cell (Figure S1) containing 85.0 mL of 0.010

122

mol L-1 HCl and NaCl solutions separated by a PTFE gas-permeable membrane with an

123

exposed area of 1.8 cm2 was used to investigate the transport of H+ across the PTFE

124

membrane. A pH electrode was used to monitor the pH change of the NaCl solution.

125

Figure 1 shows the configuration and assembly of a GPMCP. The probe base consisted of a

126

receiving phase reservoir and a 4-electrode conductivity detector (EC detector). The EC

127

detector was constructed by directly inserting 4 stainless steel rods (1.5 mm in diameter) into

128

the probe base, and the control electronic circuit board was directly mounted onto the back of

129

the base and fully sealed to ensure it was waterproof. To minimize any electromagnetic

130

interferences, the analogue signal generated by the EC detector was in situ converted into a

131

digital signal by the control electronics (Figure 1b). A PTFE membrane purchased from

132

Merck (FluoroporeTM, pore size: 0.22 µm, porosity: 85%, diameter: 47 mm, thickness: 150

133

µm) was used as the gas-permeable membrane. The GPMCP was assembled by filling the

134

receiving phase reservoir with 2.00 mL of the 0.500 mol L-1 boric acid solution and clipping

135

the probe cap onto the base with the PTFE membrane and silicone spacers in between. The

136

exposed PTFE membrane area was 7.07 cm2. 7



Page 8 of 35

137

Figure S2a shows a typical sensing system setup for the laboratory experiments. In addition

138

to the GPMCPs, the system was composed of a data logger (Figure S2b), a wireless data

139

receiver, a control computer and a sample tank. Each data logger could host up to 8 sensors

140

(including temperature and pH sensors). A computer communicated with the data logger for

141

the data transfer/storage and operational control via the wireless data receiver. The control

142

software had an automatic temperature and pH correction function. Two sample tanks with

143

capacities to hold 4 and 25 L of sample solutions were used in this work. Unless otherwise

144

stated, for all the experiments, the temperature and pH sensors were deployed together with

145

the GPMCPs.

146

Field Deployment. All field deployments were carried out using a self-powered data logger

147

(Figure S2c) that can hold 6 GPMCPs, 1 temperature sensor and 1 pH sensor and

148

continuously operate for up to 30 days. The data logger was remotely controlled by a

149

computer. Figure S2d schematically illustrates the field deployment of the GPMCPs. The

150

recorded temperature and pH data were used for the data corrections.

151

For this work, the GPMCPs were deployed at three sites in Gold Cost City, Queensland State

152

in Australia. Site #1 was at the downstream of a creek entrance to the Pacific Ocean, was

153

partially covered by mangroves, and had conductivities that fluctuated between 28 and 45 mS

154

cm-1 depending on the tidal actions. Site #2 was located at the upstream of a freshwater creek

155

surrounded by light industry and had an almost constant conductivity of 2.3 mS cm-1 that was

156

not affected by tidal actions. Site #3 was a small freshwater lake that was located in an urban

157

residential area, and it had an almost constant conductivity of 1.3 mS cm-1. During the

158

deployment, water samples were collected every 2 h, preserved at < 4 °C and analyzed for

159

NH3-N within 12 h in the laboratory.

8


Page 9 of 35


160

161

Figure 1. GPMCP configuration and assembly.

162

RESULTS AND DISCUSSION

163

Analytical Principle. A typical GPMCP contained three key elements: a gas-permeable

164

membrane (GPM), an EC detector and a boric acid (H3BO3) receiving phase (Figure 2a). As

165

illustrated in Figure 2b, when a GPMCP was deployed in an aquatic sample, the dissolved

166

ammonia ({NH3}aq) diffused to the outer GPM interface, converted into gas phase ammonia

167

({NH3}g) at the interface via evaporation and permeated into the GPM. The permeated

168

{NH3}g inside the GPM rapidly diffused to the inner surface, and {NH3}g instantaneously and

169

stoichiometrically reacted with H3BO3 at the inner GPM interface to produce NH4+ and

170

B(OH)4- in the receiving phase (Equation 1). In effect, the rapidity of this acid-base reaction

171

can efficiently reduce and maintain the concentration of {NH3}g at the inner GPM surface to

172

essentially zero ([NH3] R≈0) and continuously drive the ammonia transport across the GPM.

173

{NH3}g + H3BO3 → NH4+ + B(OH)−4

(1).

174

Considering that ammonia concentrations in most aquatic environments are low (µg L-1 – mg

175

L-1), the sensing probe operates under ambient environmental temperatures, and the {NH3}g

176

ammonia diffuses inside the GPM and reacts with boric acid at the inner GPM interface

177

quickly, it is reasonable to assume that the conversion of {NH3}aq into {NH3}g at the outer 9



Page 10 of 35

178

GPM interface is the rate limiting step of the membrane process. Because this conversion

179

process is a continuous, dynamic evaporation process, and for a given GPM, temperature and

180

deployment time, the rate of {NH3}g evaporation is directly proportional to the {NH3}aq

181

concentration in a sample ([NH3]Sample), the flux of ammonia (J) permeating through GPM

182

can be given by:

J=

183

d ([ NH 3 ] g ) dt

= k[ NH 3 ]Sample

(2).

184

As the permeated NH3 from the sample is stoichiometrically converted into NH4+ via the

185

interfacial reaction (Equation 1), the rate of the NH4+ concentration increase in the receiving

186

phase can be expressed as:

d ([NH4+ ]R ) J k = = [ NH3 ]Sample dt VR VR

187

(3).

188

Where, [NH4+]R is the accumulated NH4+ concentration in the receiving phase, and VR is the

189

volume of the boric acid receiving phase. Because the conductivity of the receiving phase is

190

directly proportional to [NH4+]R, the rate of the receiving phase conductivity increase (RCI)

191

can be presented as:

192

RCI =

dσ d ([NH4+ ]R ) k =ρ = ρ [ NH3 ]Sample = K[ NH3 ]Sample dt dt VR

(4).

193

Where, ρ is the proportional constant of the EC detector. K depends on the property and

194

exposed area of the GPM, the boric acid receiving phase volume, the characteristics of EC

195

detector and the deployment temperature. For a given GPMCP with a temperature correction,

196

K is a probe constant that can be readily experimentally determined. That is, the

197

instantaneous ammonia concentration can be determined by simply measuring the RCI of the 10


Page 11 of 35


198

receiving phase. It is important to note that because K is a probe specific constant, once the K

199

value of a GPMCP is determined, an ongoing calibration is not required during deployment.

200

This is a distinct advantage for a field-based analytical technique because it reduces the

201

maintenance and operational costs and increases the reliability.42, 44, 45

202

203

Figure 2. Schematic diagrams illustrating (a) the GPMCP configuration and (b) the

204

conceptual view of the ammonia sensing principle.

205

The ability to determine the average pollutant concentration over a long time period can

206

provide useful information for environmental evaluations and impact predictions.46 However,

207

for most analytical techniques, determining an average concentration over a prolonged period

208

is not practical because a very high sampling frequency is required and results in a large

209

number of samples and a large assay task and costs. Similar to the diffusive gradients in thin

210

films (DGT) technique,8, 46, 47 the GPMCP is also an accumulative method and capable of

211

determining an average concentration over a deployment period. For a given deployment

212

time (t), Equation 4 can be rewritten as:

dσ = ρ • d ([ NH 4+ ]R ) = K ∫0 ([ NH3 ]Sample)dt t

213

(5).

11



Page 12 of 35

214

According to Equation 5, the average ammonia concentration ( [NH3]Sample) over a deployment

215

period of t can be determined by:

216

[ NH 3 ]Sample =

dσ Kt

(6a).

217

Where, dσ is the total conductivity increment of the receiving phase over the deployment

218

period. [NH3]Sample is an absolute average concentration because the accumulated [NH4+]R in

219

the receiving phase is the result of the integration of the instantaneous sample ammonia

220

concentrations ( ∫0t ([ NH 3 ] Sample

221

solution containing a fixed concentration of {NH3}aq, Equation 6a can be rewritten as:

222

) dt

) over the entire deployment period. For a deployment

[ NH 3 ]Sample =

RCI K

(6b).

223

NH3 and NH4+ always coexist in aquatic environments. For a given temperature, assuming

224

NH3 and NH4+ are in an equilibrium state, the NH4+ concentration in the sample ([NH4+]Sample)

225

can be calculated from the determined [NH3]Sample (Equation 7). Kb

226

[ NH 4+ ]Sample =

227

Where, Kb is the ammonia dissociation base constant.

228

Analytical Principle Validation. All the validation experiments were performed using the

229

GPMCPs shown in Figure 1 at a constant temperature of 25 °C unless otherwise stated. For

230

all the experiments, a temperature sensor and a pH sensor were deployed with the GPMCPs.

231

For all the conductivity response (σ) – deployment time (t) curves reported in this work, the

232

conductivity data were recorded at a frequency of 1 reading per minute. Figures 3a and S3

233

show 3 sets of original GPMCP σ – t profiles obtained from different {NH3}aq concentrations

[OH − ]

× [ NH 3 ]Sample

(7).

12


Page 13 of 35


234

ranging from 2 µg L-1 to 50 mg L-1. Perfect linear curves were observed for all the

235

investigated cases. For each [NH3]Sample, RCI can be derived from the slope of the

236

corresponding σ – t curve. For a given GPMCP and temperature, RCI should be directly

237

proportional to [NH3]Sample, according to Equation 4. Figure 3b shows a plot of RCI versus

238

[NH3]Sample. As predicted by Equation 4, a linear relationship was obtained over the entire

239

{NH3}aq concentration range investigated. It is important to note that the GPMCP probe

240

constant (K) equals the slope of the RCI – [NH3]Sample curve. The GPMCP used for these

241

experiments had a value of K (25 °C) = 1.35×10-3 µS cm-1 min-1 µg-1 L (denoted GPMCP#1).

120

-1

0

15

-1

-1

2448 µg L -1 1632 µg L-1 816 µg L-1 408 µg L-1 204 µg L

60 0

30

σ (µ S cm )

-1

-1

RCI (µS cm-1 min-1)

(b) 100

-1

5001 µg L

180

RCI (µS cm min )

-1

σ (µ S cm )

(a)

163 µg L

6 4

80 60

0.12

y=0.00135x R2=0.9993

0.08 0.04 0.00

0

40

30

60

90

[NH3] (µg L-1)

y=0.00135x R2=0.9995

-1

85.0 µg L-1 40.8 µg L-1 20.4 µg L-1 10.2 µg L -1 5.10 µg L-1 2.04 µg L

2 0 0

15

20 0

30

0

Time (min)

10

20

30

40

50

[NH3] (mg L-1)

(c) 180

(d) 3.0

3.0

2.4

2.4

1.8

1.8

1.2

1.2

0.6

0.6

0.0

0.0

-1

-1

0

-4

-1

0.0

+

1.5

60

-4

-1

3.0

[NH3] (mg L )

σ (µ S cm )

120

[NH4 ] (10 mol L )

[NH3] (10 mol L )

4.5

-1.5 0

10

20

30

40

50

60

4

6

Time (min)

8

10 pH

12

14

242

243

Figure 3. (a) Original σ – t profiles obtained using GPMCP#1 in deployment solutions

244

containing different concentrations of NH3; (b) RCI – [NH3] relationship derived from Figures

245

3a and S3; (c) Original conductivity response obtained from a deployment solution with

246

successively increasing NH3 concentrations and the corresponding [NH3]Samplevalues (blue lines) 13



Page 14 of 35

247

determined using Equation 6b; (d) The [NH3]Sample (o) and [NH4+]Sample (∆) values determined

248

using Equations 4 and 7 in a synthetic freshwater deployment solution containing 0.240 µmol

249

L-1 of NH4Cl at different pH values. The red and blue dashed curves show the theoretical

250

distribution of NH3 and NH4+ for the pH values. All the experiments were carried out at

251

25 °C.

252

GPMCP#1 was used to obtain a set of conductivity responses from a synthetic freshwater

253

deployment solution with successively increasing NH3 concentrations (Figure 3c). The blue

254

lines shown in Figure 3c are the corresponding [NH3]Sampledetermined by Equation 6b. Table

255

S1 summarizes the measured RCI values for each investigated {NH3}aq concentration derived

256

from Figure 3c and the [NH3]Samplevalues determined using Equation 6b. If Equation 6b is

257

correct, the determined [NH3]Sample values should be equal to the added ammonia concentration.

258

The results show that the deviations between the determined [NH3]Sample values and the added

259

ammonia concentrations are less than ±5% for all the investigated cases, and this signifies the

260

validity of Equation 6 for determining the absolute average concentration of {NH3}aq over a

261

given deployment period.

262

A set of experiments was performed to validate Equation 7. For all the experiments, a 25 L of

263

a synthetic freshwater solution containing 0.240 µmol L-1 of NH4Cl was adjusted to different

264

pH values between 5.90 and 13.94 and was used as the deployment solution. Figure S4 shows

265

the recorded σ – t curves from the deployment solutions with different pH values. The

266

[NH3]Sample values of the deployment solutions with different pH values were determined

267

using Equation 4, and these values were used to calculate the corresponding [NH4+]Sample

268

values in accordance with Equation 7. Figure 3d shows the plots of both the [NH3]Sample and

269

[NH4+]Sample values against the pH, and the theoretical distributions of [NH3] and [NH4+] with 14


Page 15 of 35


270

the pH. Kb (25 °C) = 1.75×10-5 was used for all the calculations. The determined [NH3]Sample

271

and [NH4+]Sample values well agreed with the theoretical values within the pH range from 5.90

272

to 13.94. Considering the pH range in aquatic environments is normally between 6 and 9, the

273

results shown in Figure 3d confirm that Equation 7 is applicable for the determination of

274

[NH4+] in aquatic environments. In fact, the [NH3] and [NH4+] in aquatic environments can

275

be simultaneously monitored using Equations 4 and 7, respectively. Furthermore, the absolute

276

average ammonium concentration over a given deployment period ( [ NH 4+ ]Sample ) can be

277

obtained based on the determined [NH3]Sample value in accordance with Equation 6.

278

Sensitivity is important for environmental monitoring applications because the targeted

279

species are often present in low concentrations. The results shown in Figure 3a suggest that

280

the GPMCP is capable of directly determining a concentration of 2 µg L-1 of free ammonia.

281

In fact, the sensitivity can be further increased by simply extending the deployment time

282

because the GPMCP is an accumulative method. Additionally, Figure 3b shows an upper

283

linear range of 50 mg L-1, which can be readily extended by reducing the deployment time.

284

Furthermore, the sensitivity, linear range and maximum deployment time can be tuned by

285

altering the GPMCP design. A GPMCP with a higher ratio of the exposed membrane area to

286

the boric acid receiving phase volume will possess a higher sensitivity with a lower upper

287

linear range and shorter maximum deployment time.

288

The selectivity is an important performance indicator in many analytical applications,

289

especially for field-based environmental monitoring applications in which the sample

290

matrixes can be highly diverse and complex. The selectivity of the GPMCP was therefore

291

investigated. H+ was selected as the probing ion because it is the smallest ion in aqueous

292

media. The transport of H+ across the PTFE membrane was investigated using a two-

293

compartment cell (Figure S1) with 0.01 mol L-1 HCl and NaCl solutions separated by a PTFE 15



Page 16 of 35

294

gas-permeable membrane. Figure S5 shows a typical pH – time profile for the NaCl solution.

295

A measurable decreasing pH trend was not observed over the test period of 5 days, which

296

means that the transport of H+ across the membrane did not occur during the test period. This

297

also means that ionic species cannot permeate through the PTFE membrane. This is not a

298

supervising outcome because of the hydrophobic nature of the PTFE membrane. Based on

299

their acid – base properties, dissolvable gases in aquatic environments can be classified as

300

neutral (i.e., O2, N2, etc.), acidic (i.e., CO2, H2S, SO2, etc.) and basic (i.e., NH3) gases.

301

According to the GPMCP sensing principle (Figure 2b), only basic gases can continuously

302

permeate through the membrane via the acid – base reaction at the inner membrane interface,

303

and neutral and acidic gases cannot permeate because they do not react with boric acid. To

304

the best of our knowledge, ammonia is the only dissolvable basic gas in aquatic environments

305

with an appreciable solubility. Hence, GPMCP is specific for ammonia detection regardless

306

of the type of aquatic environment, and this is demonstrated in Figure S6 where the

307

determined [NH3]Sample and [NH4+]Sample values in a synthetic seawater matrix over a wide pH

308

range are almost identical to those obtained in a freshwater matrix (Figure 3d).

309

GPMCP Calibration. An analytical technique that requires no ongoing calibration is highly

310

desirable for field-based environmental monitoring applications. According to Equation 4, K

311

is a probe specific constant. For GPMCP#1, the probe constant at 25 °C was determined to be

312

K(25 °C) = 1.35×10-3 µS cm-1 min-1 µg-1 L (Figure 3b). However, for a given probe, K also

313

depends on the temperature and must be corrected for practical use. A two-step correction

314

strategy was developed for the GPMCP calibration: (i) the temperature effect on the EC

315

detector of the GPMCP was corrected to normalize the conductivities measured at different

316

temperatures (σT) to the ‘standard’ conductivity value at 25 °C (σ25); (ii) the σ25 data were

317

used to obtain the K – temperature dependent relationship of the GPMCP and to determine

318

the K (T) for practical use. 16


Page 17 of 35


319

Figure 4a shows the plots of the measured conductivity changes with the temperature for

320

GPMCP#1 with a 0.500 mol L-1 boric acid receiving phase. The slope and intercept values of

321

1.08 µs cm-1 oC-1 and 27.3 µs cm-1 were obtained, respectively. Therefore, the σT value at a

322

temperature (T, oC) measured by the EC detector in GPMCP#1 can be converted to σ25 by: σ 25 = σ T + 1 . 08 × ( 25 − T )

323

(8).

324

Equation 8 is probe specific and can only be used to correct the EC detector for GPMCP#1.

325

In practice, a temperature correction formula needs to be determined for each probe. In this

326

work, all the measured σ values were real-time corrected to σ25 values based on the measured

327

temperature and the predetermined EC detector temperature correction formula (e.g.,

328

Equation 8) of the probe using the sensor control software. All the conductivity values

329

reported in this work are σ25 values unless otherwise stated.

(a) 80

(b) 0.003 y=1.08x+27.3 2 R =0.999

-1

σ (µ S cm )

70

-5

-4

y=2.6510 x+6.8510 R2=0.9199

0.002

K

60 50 0.001 40 30

0.000 5

10

15

20

25

30

35

40

15

20

o

330

Temperature ( C)

25

30

35

40

o

Temperature ( C)

331

Figure 4. (a) The temperature correction curve for the EC detector in GPMCP#1; (b) The

332

temperature correction curve for the GPMCP#1 constant.

333

Figure S7 shows 6 groups of typical σ – t curves and the corresponding RCI – [NH3] curves

334

determined using GPMCP#1 for a set of deployment solutions containing different

335

concentrations of NH3 at different temperatures between 15 and 40 °C. Each group of the σ –

336

t curves was measured at a constant temperature and was used to obtain the RCI values to 17



Page 18 of 35

337

determine the probe constant (K(T)) at the corresponding temperature. The probe constant

338

and temperature dependent relationship can then be determined from the K(T) – T curve

339

shown in Figure 4b: K (T) = 2.65 × 10−5 T ( oC ) + 6.85 × 10−4

340

(9).

341

Equation 9 is also probe specific and can only be used to correct the temperature-induced

342

probe constant changes for GPMCP#1. In practice, each GPMCP must be calibrated to obtain

343

the temperature correction equation. Again, the sensor control software automatically

344

corrected for the temperature effect.

345

In theory, once a GPMCP is calibrated, the obtained probe constant should not change over

346

time, and an ongoing calibration is not needed. Figure S8 shows a set of σ – t curves obtained

347

by CPMCP#1 over a period of 142 days in a deployment solution containing 1020 µg L-1 of

348

added NH3. Over the entire testing period, other than obtaining the results shown in Figure S8,

349

the same probe was used for various other experiments without additional calibration. The

350

determined [NH3]Samplevalues with the probe constant shown in Equations 8 and 9 are almost

351

identical to those for the added NH3 concentrations, demonstrating that a calibrated GPMCP

352

can be used over a prolonged period without the need for additional calibration. The other

353

GPMCPs (Table S2) used in this work were calibrated using the same method described

354

above.

355

Field Deployment. GPMCPs were deployed at three sites to evaluate the system performance

356

for field-based monitoring applications. For all the sites, a GPMCP was deployed with a pH

357

probe and a temperature sensor. Site #1 was selected to represent typical estuary waters.

358

GPMCP#1 (see Table S2, σ 25 = σ T + 1 . 08 × ( 25 − T ) ; K (T) = 2.65×10−5 T (oC) + 6.85×10−4 ) was

359

deployed at Site #1. During a 24 h deployment period, the conductivity, pH and temperature 18


Page 19 of 35


360

of the site water varied between 28 – 45 mS cm-1, 7.90 – 8.09 and 21.2 – 24.8 °C,

361

respectively. These changes were strongly associated with the tidal actions. Higher

362

conductivity and pH values and a lower temperature were observed during the high tide

363

periods. Figure 5a shows the recorded σ – t profile and the corresponding instantaneous NH3

364

concentration changes that were determined in real-time using Equation 4. The continuous

365

NH3 concentration profile shows the changes in the NH3 levels. During the deployment, two

366

high tides occurred at approximately 10 am and 10 pm, and two low tides occurred at 5 pm

367

and 4 am the next day. Higher NH3 concentrations were observed around these high and low

368

tide points. This could be due to the tidal actions stirring up precipitated NH3/ NH4+ in the

369

mangrove wetlands and creek sediments. Furthermore, over the 24 h deployment period, the

370

total conductivity increment (dσ) derived from Figure 5a was 7.1 µS cm-1, and the absolute

371

average NH3 concentration ( [ NH3 ]Sample ) of 5.25 µg L-1 was obtained from Equation 6a.

372

Figure 5b shows the NH4+ profile over the deployment period determined by Equation 7. The

373

absolute average NH4+ concentration ( [ NH 4+ ]Sample ) of 70.3 µg L-1 was obtained via Equation 7

374

using [ NH3 ]Sample = 5.25 µg L-1. Kb (25 °C) = 1.75×10-5 was used for all the calculations.

375

Considering that the temperatures varied between 21.2 and 24.8 °C during the deployment

376

period, the errors introduced by using Kb (25 °C) should be less than 5% because a 5 °C

377

temperature change will cause a < 5% change in the Kb value.48 During the deployment,

378

water samples were collected every 2 h for laboratory determination of the ammoniacal

379

nitrogen ([NH3-N]STD) content using a standard method.25 In our case, the determined [NH3-

380

N]STD was equivalent to the total concentration of NH3 and NH4+. Figure 5c shows a plot of

381

the total NH3 and NH4+ concentrations ([NH3-N]GPMCP) determined by GPMCP against [NH3-

382

N]STD. The near unity slope value suggests an excellent agreement between the two methods.

19



(a) 20

(c) 160

(b)

58

2

R =0.981 slope=1.09 n=12

10

-1

52 5

200

+

54

[NH 4 ] (µ g L-1)

-1

[NH3-N]GPMCP (µg L )

56 σ (µS cm )

[NH3] (µ g L-1)

300

15

100

50

Page 20 of 35

120

80

40

0

0 05 AM 11 AM

05 PM

11 PM

48 05 AM 11 AM

05 AM 11 AM

Time

05 PM

11 PM

05 AM 11 AM

0 0

Time

383

40

80

120

-1

160

[NH3-N]STD (µg L )

384

Figure 5. (a) The recovered σ – t profile and the corresponding instantaneous NH3

385

concentration – time profile; (b) NH4+ concentration – time profile (Red arrows indicating

386

sampling points for laboratory analysis); (c) A plot of [NH3-N]GPMCP against [NH3-N]STD.

387

Site #2 was selected to represent a typical freshwater creek not affected by tidal actions.

388

GPMCP#2 (see Table S2, σ 25 = σ T + 1.10× (25 − T ) ; K(T) = 5.13×10−5T (oC) +1.78×10−4 ) was

389

deployed at Site #2. During the deployment period, the temperature of the site water varied

390

between 23.5 – 26 °C, and the conductivity (2.3 mS cm-1) and pH (7.57) were nearly constant.

391

In contrast to Site #1, the continuous NH3 and NH4+ concentration profiles (Figures S9a and

392

9b) recorded over the deployment period showed high NH3 and NH4+ concentrations during

393

the daytime and low concentrations during the night, suggesting that the high daytime NH3

394

and NH4+ levels could be attributed to discharges into the creek from the surrounding

395

industries. The [ NH3 ]Sample ,

396

deployment period were 5.95, 280 and 286.1 µg L-1, respectively, and the [ NH3 ]Sample and

397

[ NH 4+ ]Sample

398

13.21 to 2.22 and 622.7 to 104.9 µg L-1, respectively. The plot of [NH3-N]GPMCP against

399

[NH3-N]STD is given in Figure S9c, and the slope value is near unity, suggesting that the NH3-

400

N concentrations determined by both methods are essentially the same.

[ NH 4+ ]Sample

and average [NH3-N]GPMCP values over the entire

values during the day (11 am - 6 pm) and night (6 pm – 8.30 am) varied from

20


Page 21 of 35


401

Site #3 was selected to represent immobile water. GPMCP#3 (see Table S2,

402

σ 25 = σ T + 1 . 12 × ( 25 − T ) ; K(T) = 3.50×10−5T(oC) + 2.85×10−4 ) was deployed at Site #3. During the

403

deployment period, the water temperatures varied between 20.6 – 22.5 °C, and the

404

conductivity (1.25 mS cm-1) and pH (6.63) remained almost the same. Figures S10a and b

405

show the NH3 and NH4+ concentration profiles recorded over the deployment period. In a

406

strong contrast to Sites #1 and #2, only trivial variations in the NH3 and NH4+ concentrations

407

were observed over the deployment period. The maximum and minimum NH3 and NH4+

408

concentrations were 2.48, 1.71 and 1017, 705 µg L-1, respectively. This could be attributed to

409

the immobile nature of the lake. The [ NH3 ]Sample ,

410

values over the deployment period were 2.11, 865 and 867 µg L-1, respectively. Figure S10c

411

shows the plot of [NH3-N]GPMCP against [NH3-N]STD, and the slope value is near unity,

412

confirming the validity of the GPMCP measurement.

413

The system maintenance requirement is an important practical issue for long-term field-based

414

monitoring applications. In this regards, except periodically replacing boric acid receiving

415

solution, the GPMCP barely requires any other maintenance during long-term deployment.

416

The above results demonstrate that the GPMCP reported in this work can be used for in situ,

417

real-time monitoring of ammonia and ammonium in aquatic environments in a continuous

418

fashion for a prolonged time period without the need for ongoing calibration. The GPMCP

419

can also be used to easily and conveniently determine the absolute average ammonia and

420

ammonium concentrations over a given deployment period. The GPMCP uses a simple yet

421

sophisticated analytical principle with the inherent advantages of a high selectivity,

422

sensitivity and accuracy and does not require an ongoing calibration. Additionally, the

423

GPMCP has a simple configuration, and it is cheap to build, easy to massively produce and

424

convenient for field deployment with minimal maintenance. These advantageous features

[ NH 4+ ]Sample

and average [NH3-N]GPMCP

21



Page 22 of 35

425

make the GPMCP an effective monitoring tool for long-term, large-scale environmental

426

assessments.

427

ACKNOWLEDGMENTS

428

The authors gratefully acknowledge the financial supports from Australia Government

429

Flagship Collaboration Research Fund, Sensor Systems for Analysis of Aquatic

430

Environments.

431

SUPPORTING INFORMATION

432

Figures: Schematic diagram of H+ permeation experiments; Photographs of ammonia sensing

433

system setup for the laboratory experiments and schematic diagrams of field deployment of

434

the GPMCPs; σ – t profile for high concentrations of {NH3}aq; σ – t profile for freshwater

435

deployment solutions with different pH values; pH measurement against time for NaCl

436

compartment; σ – t profile for seawater deployment solutions with different pH values and

437

corresponding determined NH3 and NH4+ concentration; σ – t curves and corresponding RCI –

438

[NH3] curves determined using GPMCP#1 at different temperatures; σ – t curves obtained

439

using CPMCP#1 over a period of 142 days; Ammonia and ammonium monitoring results for

440

Site #2 and Site #3. Tables: Calculation of RCI, K and [NH3]Sample; Temperature correction for

441

EC detector and GPMCP.

442

REFERENCES

443

1.

444

Soil Use Manage. 2017, 33, (2), 263-275.

445

2.

446

of in-canopy ammonia sources and sinks in a fertilized Zea mays field. Environ. Sci. Technol. 2010,

447

44, (5), 1683-1689.

Bittman, S.; Sheppard, S.; Hunt, D., Potential for mitigating atmospheric ammonia in Canada.

Bash, J. O.; Walker, J. T.; Katul, G. G.; Jones, M. R.; Nemitz, E.; Robarge, W. P., Estimation

22


Page 23 of 35


448

3.

Clarisse, L.; Clerbaux, C.; Dentener, F.; Hurtmans, D.; Coheur, P.-F., Global ammonia

449

distribution derived from infrared satellite observations. Nat. Geosci. 2009, 2, (7), 479-483.

450

4.

451

Managing nitrogen for sustainable development. Nature 2015, 528, (7580), 51-59.

452

5.

453

Kiese, R.; Wolf, B., The nitrogen cycle: A review of isotope effects and isotope modeling approaches.

454

Soil Biol. Biochem. 2017, 105, 121-137.

455

6.

456

agriculture and the role of biological nitrification inhibition. Nat. Plants 2017, 3, 17074.

457

7.

458

wastewater by the ion exchange resin. Nature Environment and Pollution Technology 2017, 16, (1),

459

261.

460

8.

461

Films” techniques provide representative time-weighted average measurements of inorganic nutrients

462

in dynamic freshwater systems. Environ. Sci. Technol. 2016, 50, (24), 13446-13454.

463

9.

464

excretion as a strategy for survival in a fish living in a very alkaline environment. Nature 1989, 337,

465

(6203), 165-166.

466

10.

467

electrolysis-photocatalysis on water treatment and fish performance in saline recirculating aquaculture

468

system. Sci Rep 2017, 7, 45066.

469

11.

470

(USA) ictalurid catfish ponds and the possible risk of ammonia toxicity. Aquaculture 2015, 437, 263-

471

269.

472

12.

Zhang, X.; Davidson, E. A.; Mauzerall, D. L.; Searchinger, T. D.; Dumas, P.; Shen, Y.,

Denk, T. R.; Mohn, J.; Decock, C.; Lewicka-Szczebak, D.; Harris, E.; Butterbach-Bahl, K.;

Coskun, D.; Britto, D. T.; Shi, W.; Kronzucker, H. J., Nitrogen transformations in modern

Qiu, M.; Hu, C.; Liu, J.; Chen, C.; Lou, X., Removal of high concentration of ammonia from

Huang, J.; Bennett, W. W.; Welsh, D. T.; Li, T.; Teasdale, P. R., “Diffusive Gradients in Thin

Randall, D.; Wood, C.; Perry, S.; Bergman, H.; Maloiy, G.; Mommsen, T.; Wright, P., Urea

Ye, Z.; Wang, S.; Gao, W.; Li, H.; Pei, L.; Shen, M.; Zhu, S., Synergistic effects of micro-

Zhou, L.; Boyd, C. E., An assessment of total ammonia nitrogen concentration in Alabama

Randall, D.; Tsui, T., Ammonia toxicity in fish. Mar. Pollut. Bull. 2002, 45, (1), 17-23.

23



Page 24 of 35

473

13.

Huszár, H.; Pogány, A.; Bozóki, Z.; Mohácsi, Á.; Horváth, L.; Szabó, G., Ammonia

474

monitoring at ppb level using photoacoustic spectroscopy for environmental application. Sens.

475

Actuator B-Chem. 2008, 134, (2), 1027-1033.

476

14.

477

emissions of ammonia, nitrous oxide and on crop response—A review. Agric. Ecosyst. Environ. 2010,

478

137, (1–2), 39-46.

479

15.

480

ESA Working paper Rome, FAO, 2012.

481

16.

482

sustainability and intensive production practices. Nature 2002, 418, (6898), 671-677.

483

17.

484

sources of atmospheric NH3 in an urban environment in northern China during wintertime. Environ.

485

Sci. Technol. 2017, 51, (12), 6839-6848.

486

18.

487

H. W.; Leong, Y. J.; Yang, M. M., Vehicle emissions as an important urban ammonia source in the

488

United States and China. Environ. Sci. Technol. 2017, 51, (4), 2472-2481.

489

19.

490

characterized by mobile, open-path measurements. Environ. Sci. Technol. 2014, 48, (7), 3943-3950.

491

20.

492

review on emission sources, atmospheric chemistry and deposition on terrestrial bodies. Environ. Sci.

493

Pollut. Res. 2013, 20, (11), 8092-8131.

494

21.

495

22, (1), 63-70.

496

22.

497

point in enhanced coagulation–sedimentation for treating micro ammonia polluted raw water. Desalin.

498

Water Treat. 2016, 57, (5), 2142-2151.

Webb, J.; Pain, B.; Bittman, S.; Morgan, J., The impacts of manure application methods on

Alexandratos, N.; Bruinsma, J. World agriculture towards 2030/2050: the 2012 revision;

Tilman, D.; Cassman, K. G.; Matson, P. A.; Naylor, R.; Polasky, S., Agricultural

Teng, X.; Hu, Q.; Zhang, L.; Qi, J.; Shi, J.; Xie, H.; Gao, H.; Yao, X., Identification of major

Sun, K.; Tao, L.; Miller, D. J.; Pan, D.; Golston, L. M.; Zondlo, M. A.; Griffin, R. J.; Wallace,

Sun, K.; Tao, L.; Miller, D. J.; Khan, M. A.; Zondlo, M. A., On-road ammonia emissions

Behera, S. N.; Sharma, M.; Aneja, V. P.; Balasubramanian, R., Ammonia in the atmosphere: a

Sapek, A., Ammonia Emissions from Non-Agricultural Sources. Pol. J. Environ. Stud. 2013,

Liao, Z. L.; Chen, H.; Zhu, B. R.; Li, H. Z., Analysis and selection of powdered zeolite dosing

24


Page 25 of 35


499

23.

Li, Y.; Xu, E. G.; Liu, W.; Chen, Y.; Liu, H.; Li, D.; Liu, Z.; Giesy, J. P.; Yu, H., Spatial and

500

temporal ecological risk assessment of unionized ammonia nitrogen in Tai Lake, China (2004–2015).

501

Ecotox. Environ. Safe. 2017, 140, 249-255.

502

24.

503

Sensors 2012, 12, (7), 9635-9665.

504

25.

505

wastewater. American Public Health Association (APHA): Washington, DC, USA, 2005.

506

26.

507

substances using the modified Nessler method. J. Chem. 2013, 2013, 1-9.

508

27.

509

Analysis; Leo M. L. Nollte, Leen S. P. De Gelder, Eds.; CRC Press: Boca Raton 2013; pp 249.

510

28.

511

for ammonium/ammonia achieved by Berthelot's reaction by use of solid-phase extraction coupled to

512

diffuse reflectance spectroscopy. Anal. Chim. Acta 2005, 534, (2), 327-334.

513

29.

514

nitrogen in water by flow injection analysis based on NH3-o-phthalaldehyde-Na2SO3 reaction.

515

Analytical Chemistry Research 2016, 10, 1-8.

516

30.

517

141, (7), 2085-2139.

518

31.

519

determination of trace ammonium in seawater with a long-path liquid waveguide capillary cell and

520

spectrophotometric detection. Mar. Chem. 2014, 162, 114-121.

521

32.

522

analysis. A review. Anal. Chim. Acta 2012, 714, 8-19.

523

33.

524

analysis. Anal. Chim. Acta 2009, 648, (1), 7-22.

Liu, X.; Cheng, S.; Liu, H.; Hu, S.; Zhang, D.; Ning, H., A survey on gas sensing technology.

Federation, W. E.; Association, A. P. H., Standard methods for the examination of water and

Jeong, H.; Park, J.; Kim, H., Determination of NH4+ in environmental water with interfering

Galvão, J. A.; et al. Determination of ammonia in water samples. In Handbook of Water

Moliner-Martinez, Y.; Herraez-Hernandez, R.; Campins-Falco, P., Improved detection limit

Liang, Y.; Yan, C.; Guo, Q.; Xu, J.; Hu, H., Spectrophotometric determination of ammonia

Trojanowicz, M.; Kołacińska, K., Recent advances in flow injection analysis. Analyst 2016,

Zhu, Y.; Yuan, D.; Huang, Y.; Ma, J.; Feng, S.; Lin, K., A modified method for on-line

Melchert, W. R.; Reis, B. F.; Rocha, F. R. P., Green chemistry and the evolution of flow

Mesquita, R. B. R.; Rangel, A. O. S. S., A review on sequential injection methods for water

25



Page 26 of 35

525

34.

Davey, D. E.; McLeod, S.; Chow, C. W.; Ostrowski, J.; Duker, P.; Bustamante, H.; Vitanage,

526

D.; Meli, T. Development of an on-line nitrogen monitoring system using Microdistillation Flow

527

Analysis, Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP), Seventh

528

International Conference 2011, pp 180-183.

529

35.

530

of chloramination control strategy based on free-ammonia concentration. J. Water Supply Res

531

Technol.-Aqua 2009, 58, (1), 29-39.

532

36.

533

based preconcentration technique for ammonia measurement. Analyst 1997, 122, (12), 1549-1552.

534

37.

535

flow-injection system for the determination of trace level ammonia. Laboratory Automation &

536

Information Management 1997, 33, (2), 129-136.

537

38.

538

nanoparticles. Anal. Chem. 2010, 82, (12), 5002-5004.

539

39.

540

sensing of water toxicants. Lab Chip 2014, 14, (7), 1270-1280.

541

40.

542

resonance energy transfer between peroxymonocarbonate and branched NaYF4: Yb3+/Er3+

543

nanoparticles. Anal. Chem. 2012, 84, (20), 8871-8879.

544

41.

545

fluorescent zwitterionic spirocyclic meisenheimer complex. Sci Rep 2017, 7, 40465.

546

42.

547

environmental monitoring. Sens. Actuator B-Chem. 2017, 243, 566-578.

548

43.

549

ammonia. Talanta 2008, 77, (1), 66-72.

Motzko, S.; Fabris, R.; Badalyan, A.; Henderson, R.; Chow, C. W.; Vitanage, D., Assessment

Lane, R.; Chow, C. K.; Davey, D. E.; Mulcahy, D.; McLeod, S., On-line microdistillation-

Chow, C. W.; Lane, R.; Yeow, T.; Davey, D.; Mulcahy, D., Development of an automated

Mader, H. S.; Wolfbeis, O. S., Optical ammonia sensor based on upconverting luminescent

Liu, F.; Nordin, A.; Li, F.; Voiculescu, I., A lab-on-chip cell-based biosensor for label-free

Chen, H.; Li, H.; Lin, J.-M., Determination of ammonia in water based on chemiluminescence

Das, T.; Pramanik, A.; Haldar, D., On-line ammonia sensor and invisible security ink by

Li, T.; Wu, Y.; Huang, J.; Zhang, S., Gas sensors based on membrane diffusion for

Waich, K.; Mayr, T.; Klimant, I., Fluorescence sensors for trace monitoring of dissolved

26


Page 27 of 35


550

44.

Ho, C. K.; Itamura, M. T.; Kelley, M.; Hughes, R. C., Review of chemical sensors for in-situ

551

monitoring of volatile contaminants. Sandia Report SAND2001-0643, Sandia National Laboratories

552

2001, 1-27.

553

45.

554

012004.

555

46.

556

waters using thin-film gels. Nature 1994, 367, (6463), 546-548.

557

47.

558

(DGT)–back to basics. Environ. Chem. 2012, 9, (1), 1-13.

559

48.

560

the base strength of ammonia. J. Res. Natl. Bur. Stand. 1949, 42, 419-430.

Hodgkinson, J.; Tatam, R. P., Optical gas sensing: a review. Meas. Sci. Technol. 2012, 24, (1),

Davlson, W.; Zhang, H., In situ speciation measurements of trace components in natural

Davison, W.; Zhang, H., Progress in understanding the use of diffusive gradients in thin films

Bates, R. G.; Pinching, G., Acidic dissociation constant of ammonium ion at 0 to 50 oC, and

561

562

27



Page 28 of 35

563

Table of Contents Graphic

564

Abstract art……………………………………………………….………………..……….3

565

Figure 1. GPMCP configuration and assembly…………...………………………...….…...9

566

Figure 2. Schematic diagrams illustrating (a) the GPMCP configuration and (b) the

567

conceptual

568

principle………………….…………………………….…....11

569

Figure 3. (a) Original σ – t profiles obtained using GPMCP#1 in deployment solutions

570

containing different concentrations of NH3; (b) RCI – [NH3] relationship derived from Figures

571

3a and S3; (c) Original conductivity response obtained from a deployment solution with

572

successively increasing NH3 concentrations and the corresponding [NH3]Sample values (blue

573

lines) determined using Equation 6b; (d) The [NH3]Sample (o) and [NH4+]Sample (∆) values

574

determined using Equations 4 and 7 in a synthetic freshwater deployment solution containing

575

0.240 µmol L-1 of NH4Cl at different pH values. The red and blue dashed curves show the

576

theoretical distribution of NH3 and NH4+ for the pH values. All the experiments were carried

577

out

578

25 °C……………………………………………………………………………..………….13

579

Figure 4. (a) The temperature correction curve for the EC detector in GPMCP#1; (b) The

580

temperature correction curve for the GPMCP#1 constant………………...………………...17

581

Figure 5. (a) The recovered σ – t profile and the corresponding instantaneous NH3

582

concentration – time profile; (b) NH4+ concentration – time profile (Red arrows indicate the

583

sampling points for laboratory analysis); (c) A plot of [NH3-N]GPMCP against [NH3-

584

N]STD…………………………………………………………………………………………20

view

of

the

ammonia

sensing

at

28


Page 29 of 35


585

29



Abstract art 84x47mm (300 x 300 DPI)


Page 30 of 35

Page 31 of 35


Figure 1. GPMCP configuration and assembly. 199x80mm (300 x 300 DPI)



Figure 2. Schematic diagrams illustrating (a) the GPMCP configuration and (b) the conceptual view of the ammonia sensing principle. 199x88mm (300 x 300 DPI)


Page 32 of 35

Page 33 of 35


Figure 3. (a) Original σ – t profiles obtained using GPMCP#1 in deployment solutions containing different concentrations of NH3; (b) RCI – [NH3] relationship derived from Figures 3a and S3; (c) Original conductivity response obtained from a deployment solution with successively increasing NH3 concentrations and the corresponding average [NH3] sample values (blue lines) determined using Equation 6b; (d) The [NH3]Sample (o) and [NH4+]Sample (∆) values determined using Equations 4 and 7 in a synthetic freshwater deployment solution containing 0.240 µmol L-1 of NH4Cl at different pH values. The red and blue dashed curves show the theoretical distribution of NH3 and NH4+ with the pH. All the experiments were carried out at 25 oC. 109x80mm (300 x 300 DPI)



Figure 4. (a) The temperature correction curve for the EC detector in GPMCP#1; (b) The temperature correction curve for the GPMCP#1 constant. 53x19mm (600 x 600 DPI)


Page 34 of 35

Page 35 of 35


Figure 5. (a) The recovered σ – t profile and the corresponding instantaneous NH3 concentration – time profile; (b) NH4+ concentration – time profile (Red arrows indicate the sampling points for laboratory analysis); (c) A plot of [NH3-N]GPMCP against [NH3-N]STD. 58x16mm (600 x 600 DPI)


A Gas-permeable Membrane-based Conductivity ... - ACS Publications

Recommend Documents