Ultra-trace determination of gold nanoparticles in environmental water by surfactant assisted dispersive liquid liquid microextraction coupled with electrothermal vaporization-inductively coupled plasma-mass spectrometry

    Ultra-trace determination of gold nanoparticles in environmental water by surfactant assisted dispersive liquid liquid microextraction coupled with electrothermal vaporization-inductively coupled plasma-mass spectrometry Ying Liu, Man He, Beibei Chen, Bin Hu PII: DOI: Reference:

S0584-8547(16)30059-3 doi: 10.1016/j.sab.2016.04.009 SAB 5063

To appear in:

Spectrochimica Acta Part B: Atomic Spectroscopy

Received date: Revised date: Accepted date:

7 September 2015 2 March 2016 27 April 2016

Please cite this article as: Ying Liu, Man He, Beibei Chen, Bin Hu, Ultra-trace determination of gold nanoparticles in environmental water by surfactant assisted dispersive liquid liquid microextraction coupled with electrothermal vaporization-inductively coupled plasma-mass spectrometry, Spectrochimica Acta Part B: Atomic Spectroscopy (2016), doi: 10.1016/j.sab.2016.04.009

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Ultra-trace Determination of Gold Nanoparticles in Environmental Water by

2

Surfactant Assisted Dispersive Liquid Liquid Microextraction Coupled with

3

Electrothermal Vaporization-Inductively Coupled Plasma-Mass Spectrometry

4

Ying Liu, Man He, Beibei Chen, Bin Hu*

5

Key Laboratory of Analytical Chemistry for Biology and Medicine (Ministry of

6

Education), Department of Chemistry, Wuhan University, Wuhan 430072, P R China

7

Abstract: A new method by coupling surfactant assisted dispersive liquid liquid

8

microextraction (SA-DLLME) with electrothermal vaporization inductively coupled

9

plasma mass spectrometry (ETV-ICP-MS) was proposed for the analysis of gold

10

nanoparticles (AuNPs) in environmental water samples. Effective separation of

11

AuNPs from ionic gold species was achieved by using sodium thiosulphate as a

12

complexing agent. Various experimental parameters affecting SA-DLLME of AuNPs,

13

such as the organic solvent, organic solvent volume, pH of the sample, the kind of

14

surfactant,

15

centrifugation time, and different coating as well as sizes of AuNPs were investigated

16

carefully. Furthermore, the interference of coexisting ions, dissolved organic matter

17

(DOM) and other metal nanoparticles (NPs) were studied. Under the optimal

18

conditions, a detection limit of 2.2 ng L−1 and an enrichment factor of 152-fold was

19

achieved for AuNPs, and the original morphology of the AuNPs could be maintained

20

during the extraction process. The developed method was successfully applied for the

21

analysis of AuNPs in environmental water samples, including tap water, the East Lake

22

water, and the Yangtze River water, with recoveries in the range of 89.6-102%.

23

Compared with the established methods for metal NPs analysis, the proposed method

24

has the merits of simple and fast operation, low detection limit, high selectivity, good

25

tolerance to the sample matrix and no digestion or dilution required. It provides an

26

efficient quantification methodology for monitoring AuNPs' pollution in the

27

environmental water and evaluating its toxicity.

28

_______________________________

29

*

concentration,

vortex

time,

speed

of

centrifugation,

AC

CE P

TE

surfactant

D

MA

NU

SC R

IP

T

1

Corresponding author, tel: 86-27-68752162; fax: 86-27-68754067, email: [email protected]

ACCEPTED MANUSCRIPT 30

Keywords: gold nanoparticles; surfactant assisted dispersive liquid liquid

31

microextraction; electrothermal vaporization inductively coupled plasma mass

32

spectrometry; environmental water samples

IP

34

T

33

1. Introduction

Nanoparticles (NPs) have provided a unique basis for innovation in a wide

36

variety of fields such as chemistry, medicine, electronics, biology, and material

37

sciences. Among them, gold nanoparticles (AuNPs) have been applied in biomedical

38

imaging [1], cancer therapy and diagnostics [2], biological and chemical sensing [3].

39

Due to their large quantity of production and widespread applications, AuNPs will

40

inevitably be released into the environment. It is predicted that AuNPs are expected to

41

reach a concentration of 140 ng L-1 in environmental waters within the next 10 years

42

[4]. Therefore, the adverse effects of AuNPs are becoming one of the focuses of

43

current research. It has been reported that many properties including shape, size, and

44

coatings of the AuNPs have impact on the adverse effects and even toxicity in the

45

organisms [5-7], and the cytotoxic effects of AuNPs on model human skin [8], lung [9]

46

and stem cells [10] have been demonstrated. However, the information about the

47

occurrence, fate, and toxicity of AuNPs is very limited at present, partly due to the

48

lack of quantitative methodology for AuNPs analysis in environmental and biological

49

samples. Therefore, the development of simple, sensitive and accurate analytical

50

methods for rapid determination of trace AuNPs in water samples is of great

51

significance to the environmental pollution monitoring and the toxicity evaluation of

52

metal NPs.

53

Up to now, various analytical methods have been developed for the determination of

54

AuNPs, such as electrochemical methods [11], ultraviolet (UV) spectroscopy [12],

55

Raman spectroscopy [12], energy dispersive X-ray fluorescence [13], atomic

56

absorption spectrometry (AAS) [14] and inductively coupled plasma mass

57

spectrometry (ICP-MS)[15, 16]. Among them, ICP-MS is considered to be one of the

58

most powerful techniques for NPs analysis because of its high sensitivity, wide

59

dynamic linear range and multi-element capability. Recently, single particle (SP)

AC

CE P

TE

D

MA

NU

SC R

35

ACCEPTED MANUSCRIPT ICP-MS has been proposed for ultra-trace analysis of metal NPs [17, 18] with particle

61

size larger than 20 nm [19]. However, the direct determination of AuNPs in real-world

62

environmental samples is challenging task due to the expected very low concentration

63

level of AuNPs in the samples and the very complex sample matrix. Hence, a step of

64

separation and preconcentration of AuNPs is usually required prior to ICP-MS

65

analysis. Traditional sample pretreatment methods such as centrifugation [20],

66

filtration [21], and dialysis have been developed for separation and preconcentration

67

of NPs from aqueous phase, they are time-consuming, operation troublesome, and

68

easy to cause aggregation of NPs. Moreover, a size-selective approach like field-flow

69

fractionation [22] has been employed for the analysis of AuNPs. However, this

70

method suffers from incomplete recovery and analyte loss on instrument surfaces

71

[23].

MA

NU

SC R

IP

T

60

As a consequence, some new methods such as solid phase extraction (SPE) [24,

73

25], magnetic solid phase extraction (MSPE) [26], capillary microextraction (CME)

74

[16] and cloud point extraction (CPE) [27] have been developed for the separation

75

and preconcentration of metal NPs. CPE was firstly introduced into the separation and

76

preconcentration of silver nanoparticles (AgNPs) by Liu et al. [27]. By using ICP-MS

77

for quantification, a LOD of 6 ng L-1 were achieved for Ag NPs with recovery ranging

78

from 57 to 116% for spiked environmental samples. This strategy has the merits of

79

strong anti-interference ability, high enrichment factor and keeping the size and shape

80

of NPs. However, acid digestion and further dilution to 50-100 mL was used prior to

81

ICP-MS determination. This makes the analytical method laborious and

82

time-consuming. Later on, the CPE method was extended to the analysis of AuNPs

83

[14], AgNPs [28], iron oxide nanoparticles [29], zinc oxide nanoparticles [30] and

84

copper (II) oxide nanoparticles [31] in environmental waters. Schuster et al [14, 32]

85

reported a CPE method for the extraction of AgNPs [32] and AuNPs [14] with

86

electrothermal atomic absorption spectrometry (ETAAS) for quantification. Although

87

acid digestion is not required, two hour’s incubation is needed. Moreover, due to the

88

relatively high viscosity, the rich-surfactant phase should be dissolved with ethanol

89

before introducing into ETAAS. Leopold et al. [25] developed a new SPE method

AC

CE P

TE

D

72

ACCEPTED MANUSCRIPT followed by a ligand-assisted liquid extraction for separation and preconcentration of

91

AuNPs from aqueous samples. However, the desorption time was above 3 h. In our

92

previous work [26], we developed a MSPE-ICP-MS method for the simultaneous

93

analysis of AuNPs and Au ions in environmental water. With self-prepared Al3+

94

immobilized Fe3O4@SiO2@iminodiacetic acid nanoparticles as the adsorbent, AuNPs

95

and Au ions could be simultaneously retained on this adsorbent and their separation

96

was achieved by sequential elution of Au ions and AuNPs with Na2S2O3 and

97

NH3·H2O, respectively. This method is sensitive, faster, easy-to-operate and no

98

digestion required. Very recently, we proposed an online method by online coupling

99

poly(AA-VP-Bis) monolithic CME with ICP-MS for the analysis of trace AuNPs in

IP

SC R

NU

environmental water samples [16], and the sample throughput was 6 h-1.

MA

100

T

90

Dispersive liquid-liquid microextraction (DLLME), as an interesting and valid

102

alternative in liquid phase microextraction, is a simple and fast microextraction

103

technique based on a ternary component solvent system. The advantages of DLLME

104

include simple operation, rapidity, low cost, low consumption of organic solvents and

105

high enrichment factor. Since its introduction by Assadi and co-workers in 2006 [33],

106

DLLME has been successfully applied to the analysis of organic pollutant [34], trace

107

metals and their species [35,36] in environmental samples. In DLLME, an organic

108

solvent which is soluble in the extraction solvent and miscible with water is required

109

as a dispersant to assist the formation of fine oil droplets. However, the usage of

110

conventional dispersant could decrease the partition coefficient of analytes into the

111

extraction solvent [37], and increase the consumption of hazardous organic solvents.

112

Consequently, surfactant was used as dispersant or emulsifier for enhancing the

113

dispersion of extraction solvent in aqueous phase. In 2010, Yamini et al. [38] used

114

surfactant-assisted dispersive liquid-liquid microextraction (SA-DLLME) for sample

115

preparation in the analysis of chlorophenols in water samples. In this method, an

116

environmental friendly ionic surfactant is used to disperser solvent in water samples

117

instead of toxic organic solvent. Surfactants are organic compounds that contain both

118

hydrophobic and hydrophilic groups, and soluble in both organic solvent and water.

119

They can reduce the interfacial tension between the two phases by adsorbing at the

AC

CE P

TE

D

101

ACCEPTED MANUSCRIPT liquid-liquid interface and serve as an emulsifier to enhance the mass-transfer rate

121

from aqueous samples to the extraction solvent. Hence, SA-DLLME has been

122

successfully applied to the determination of organic compounds [39, 40], trace metals

123

and their species [41, 42] in various samples. However, to the best of our knowledge,

124

there is no report about the extraction of NPs by SA-DLLME so far.

IP

T

120

The aim of this work is to explore the applicability of SA-DLLME in the

126

separation and preconcentration of AuNPs from environmental water samples. By

127

combining SA-DLLME with electrothermal vaporization (ETV), a microamount

128

sample introduction technique, a method of SA-DLLME-ETV-ICP-MS was

129

developed for the determination of AuNPs in environmental samples, without acid

130

digestion and any further dilution. Experimental parameters affecting the extraction

131

efficiency of SA-DLLME and ETV-ICP-MS determination were studied in detail. The

132

proposed method was applied to the analysis of AuNPs in different environmental

133

water samples for validation.

134

2. Experimental

135

2.1. Instrumentation

CE P

TE

D

MA

NU

SC R

125

The determination of AuNPs was performed on an Agilent ICP-MS (7500a,

137

Tokyo, Japan) equipped with a modified commercially available WF-4C graphite

138

furnace (Beijing Second Optics, China) as electrothermal vaporizer. Details on the

139

modification of the graphite furnace and its connection with ICP-MS have been

140

described previously [43].The polyethylene tubing transfer line (6 mm i.d.) had a total

141

length of 70 cm. Optimization of the ICP-MS instrument was performed with a

142

conventional pneumatic nebulization (PN) sampling mode prior to connection with

143

the ETV device. Pyrolytic graphite coated graphite tubes were used throughout the

144

work. The operating conditions for ETV-ICP-MS and the temperature program for the

145

determination of AuNPs were summarized in Table 1. The transmission electron

146

micrograph (TEM) images of the AuNPs were captured on a JEM-2010 electron

147

microscope (Tokyo, Japan). Solution pH was adjusted with a Mettler Toledo 320-S

148

pH meter (Mettler Toledo Instruments Co. Ltd., Shanghai, China).

AC

136

ACCEPTED MANUSCRIPT 149

2.2. Standard solutions and reagents

150

HAuCl4·4H2O (Sigma-Aldrich, MO, USA) was used for the preparation of Au

151

NPs and a free Au ions standard solution. Sodium citrate was purchased from

152

Sigma-Aldrich.

153

bromide (CTAB), Triton X-100 (TX-100)，Na2S2O3·5H2O, HCl (38%, w/w) were

154

bought from Sinopharm Chemistry Reagent Co. Ltd (Shanghai, China). Humic acid

155

was purchased from Alfa Aesar (Tianjin, China). Mercaptosuccinic acid (MSA, 98%),

156

Triton X-114 (TX-114), sodium dodecyl sulfate (SDS, 99.5%) were purchased from

157

Aladdin (Shanghai, China). 11-Mercaptoundecanoic acid (MUA, 95%) was obtained

158

from J&K Chemical Ltd., China. The stock standard solutions for Au ions (1.000 g L-1)

159

were prepared by dissolving appropriate amounts of HAuCl4 (Sigma-Aldrich) in 1%

160

(v/v) dilute HCl. All reagents were of analytical grade unless otherwise noted.

161

High-purity deionized water (18.25 MΩ·cm, Milli-Q Element, Millipore, Mulheim,

162

France) was used throughout this work. To avoid the adsorption of AuNPs on the

163

vessels, borosilicate glass tubes were used for all experiments. The working

164

suspensions of AuNPs were prepared by diluting the stock suspension with ultrapure

165

water and quantified by ICP-MS daily.

166

2.3. Preparation of citrate-stabilized gold nanoparticles

K-30),

Cetyltrimethylammonium

T

(PVP,

AC

CE P

TE

D

MA

NU

SC R

IP

Polyvinylpyrrolidone

167

Citrate-stabilized AuNPs with different sizes were prepared by chemical

168

reduction in aqueous solution according to the method reported in ref. [44]. An

169

aqueous solution of HAuCl4 (50 mL, 0.01% (m/v)) was boiled under stirring. Then,

170

an appropriate amount of 1% (m/v) sodium citrate was rapidly added, and the color of

171

the solution changed from pale-yellow to wine-red. Boiling was continued for 10 min,

172

and then the heating source was removed. The obtained colloids were stirred and

173

allowed to reach room temperature, and AuNPs with different sizes (17-21, 47-64 and

174

80-108 nm, characterized by TEM) were obtained by adding 1.5, 0.4 and 0.25 mL

175

sodium citrate, respectively. The final solution containing AuNPs was stored in the

176

freezer in an amber bottle at 4 oC. The AuNPs with the particle size of 17-21 nm were

177

employed to optimize the SA-DLLME procedure. The concentration of the AuNPs

ACCEPTED MANUSCRIPT was about 40 mg L-1 (as Au), as confirmed by ICP-MS measurement without

179

digestion, and Au(III) ions prepared in aqueous solution was used for the preparation

180

of calibration standard series.

181

2.4. Preparation of AuNPs with different coatings

IP

T

178

The procedure for preparation of AuNPs with different coatings is based

183

on the procedure reported by Su et al. [26]. Firstly, the freshly prepared

184

citrate-stabilized AuNP solution (40 mg L-1) was diluted to 10 mg L-1 with high

185

purity water. Secondly, 1 mL of PVP (2.5 mmol L-1), MUA (2.5 mmol L-1), or

186

MSA (2.5 mmol L-1) was added to 4 mL of this freshly diluted AuNP solution,

187

respectively. The resulting solution was stirred at room temperature for 3 h and used

188

for further experiments.

189

2.5. Preparation of citrate-stabilized silver nanoparticles

D

MA

NU

SC R

182

The citrate-stabilized AgNPs were prepared according to Yeh et al. [45] with

191

minor modifications. Briefly, 25 mL 1 mmol L-1 silver nitrate was mixed with 4.4 mL

192

1% trisodium citrate under magnetic stirring. Then 5 mL of 10 mmol L-1 sodium

193

borohydride solution was added dropwise into the mixture at 0 oC in an ice-water bath.

194

With the addition of sodium borohydride, the colorless solution changed to bright

195

yellow immediately and the citrate-stabilized AgNPs were obtained with the diameter

196

of about 10 nm.

197

2.6. SA-DLLME procedure

AC

CE P

TE

190

198

5.0 mL aliquot of water samples was placed in a 10 mL screw cap glass tube

199

with conical bottom. 70 μL of 1, 2-dichloroethane as extraction solvent and 50 μL of

200

10% TX-114 as emulsifier were added into the sample solution, and the concentration

201

of TX-114 in sample solution was 0.1%. The tube was then vigorously shaken on a

202

vortex agitator for 1 min. During the vortex process, the solution became turbid due to

203

the dispersion of very fine 1, 2-dichloroethane droplets into the aqueous sample, and

204

finally formed a cloudy state. The organic droplets were collected by centrifugation at

205

2,500 rpm for 3 min, and the sedimented phase was directly injected into the

ACCEPTED MANUSCRIPT 206

ETV-ICP-MS for the determination of AuNPs.

207

2.7. Sample preparation Tap water, river water, and lake water were collected from the laboratory (Wuhan

209

University, Wuhan, China), the Yangtze River (Wuhan, China), and the East Lake

210

(Wuhan, China), respectively. Immediately after sampling, the water samples were

211

filtered through a 0.45 μm cellulose acetate membrane (Tianjin Jinteng Instrument

212

Factory, Tianjin, China), and then adjusted to pH 3.0 with concentrated HCl prior to

213

storage.

214

3. Results and discussion

215

3.1. Optimization of the SA-DLLME parameters

MA

NU

SC R

IP

T

208

In order to optimize the SA-DLLME extraction procedure for efficient

217

preconcentration of AuNPs, the effect of several parameters, such as the organic

218

solvent, organic solvent volume, pH of the sample, the kind of surfactant, surfactant

219

concentration, vortex time, speed of centrifugation, centrifugation time, and coating as

220

well as size of AuNPs, was studied.

221

3.1.1. Effect of extraction solvent

CE P

TE

D

216

The selection of an appropriate extraction solvent is of great importance in

223

SA-DLLME. The organic extraction solvent determines the partition coefficient

224

between the extraction phase and donor phase. In this method, a suitable solvent has

225

to meet the following requirements: (1) it should have a good extraction affinity for

226

the target analytes to ensure high enrichment; (2) it should be immiscible with water;

227

and (3) it should have a higher density than water. Herein, four common high-density

228

organic solvents were evaluated as extraction solvent including chloroform, carbon

229

tetrachloride, 1, 2-dichloroethane and chlorobenzene. In the experiment, 65 μL

230

chloroform, 38 μL carbon tetrachloride, 70 μL 1, 2-dichloroethane and 30 μL

231

chlorobenzene were employed, respectively, to obtain the sedimented phase with

232

similar volume (25 μL). As shown in Fig.1, 1, 2-dichloroethane presented the highest

233

extraction efficiency for AuNPs among the tested organic solvents, probably due to its

AC

222

ACCEPTED MANUSCRIPT better solubility toward AuNPs. Therefore, 1, 2-dichloroethane was selected as the

235

organic solvent for extraction of AuNPs by SA-DLLME.

236

3.1.2. Effect of the volume of extraction solvent

T

234

In SA-DLLME, the volume of extraction solvent is also an important parameter,

238

as it affects the enrichment factor. To study the effect of extraction solvent volume on

239

extraction, different volumes of 1, 2-dichloroethane (70, 80, 90, 100, 110 μL) were

240

added to 5 mL sample solution containing 0.1% (w/v) TX-114. It was found that the

241

signal intensity of AuNPs was decreased with the increase of volume of 1,

242

2-dichloroethane from 70 to 110 μL. However, it should be noted that when the

243

volume of 1, 2-dichloroethane was increased from 70 to 110 μL, the volume of the

244

sedimented phase was increased from 25 to 65 μL. And when the volume of 1,

245

2-dichloroethane is less than 70 μL, the volume of the obtained sedimented phase is

246

not enough for ETV sampling. Thus, 70 μL of 1, 2-dichloroethane were selected as

247

the extraction solvent for subsequent experiments.

248

3.1.3. Sample pH

CE P

TE

D

MA

NU

SC R

IP

237

The pH value is one of the most relevant parameters for SA-DLLME efficiency.

250

In this study, the sample pH was optimized in the range of pH 2.0-8.0 for the

251

extraction of AuNPs. As shown in Fig. 2, the maximum signal intensity was obtained

252

at pH 2.0-3.5, which is around the zero point charge pH (pHPZC) of the

253

citrate-stabilized AuNPs (~3.0). The possible reason is that electrostatically stabilized

254

AuNPs are beneficially extracted at pH values close to the pHPZC where the

255

coulombic repulsion forces responsible for their stabilization reach their minimum

256

value. Moreover, the reproducibility of signal intensity is not good at pH 2.0 due to

257

the agglomeration of AuNPs. Therefore, pH 3.0 was selected for the selective

258

separation of AuNPs.

259

3.1.4. Effect of the type of surfactant

AC

249

260

The surfactant nature is also of great importance for obtaining a satisfactory

261

extraction. The surfactant, which serves as an emulsifier, accelerates the

ACCEPTED MANUSCRIPT emulsification of the water-immiscible organic extraction solvent in the aqueous

263

solution under vortex mixing. After emulsification, the extraction solvent is dispersed

264

as fine droplets in the aqueous solution, which increases the mass transfer of the

265

analytes from the aqueous to the organic phase. In this work, various disperser

266

solvents were investigated and compared, including non-ionic surfactants (TX-100

267

and TX-114), cationic surfactant (CTAB), anionic surfactant (SDS), and commonly

268

used solvents (methanol and ethanol). Figure 3 displays the comparison of extraction

269

efficiency of AuNPs obtained with different disperser solvents. As can be seen, a

270

higher extraction efficiency was obtained with the use of surfactants as disperser

271

solvent (SA-DLLME) than that with methanol and ethanol as dispersive solvent

272

(conventional DLLME), probably due to the nonionic intermolecular forces formed

273

between AuNPs and the surfactant. And TX-114 gave the highest extraction efficiency

274

for AuNPs among the tested surfactants, because TX-114 might have a suitable

275

hydrophobicity for AuNPs, resulting in better extraction efficiency. Based on the

276

experimental results, TX-114 was selected as an appropriate surfactant for subsequent

277

studies.

278

3.1.5. Effect of the surfactant concentration

CE P

TE

D

MA

NU

SC R

IP

T

262

Surfactant concentration is another important factor for effective extraction

280

because it influences the emulsification and mass-transfer process. Then the influence

281

of the concentration of TX-114 ranging from 0 to 0.2 (w/v, %) on the extraction of

282

AuNPs in 5 mL aqueous solution were investigated by using 70 μL 1,

283

2-dichloroethane as the extraction solvent. As shown in Fig. S1 (Appendix A), the

284

analytical signal intensity of AuNPs increased with increasing TX-114 concentration

285

up to 0.02% (w/v) and then remained constant until 0.2%. Thus, 0.1% of TX-114 was

286

used in subsequent experiment.

287

3.1.6. Effect of vortex time

AC

279

288

Vortex time (duration of the vortexing) is one of the main factors in DLLME. It

289

affects the extraction equilibrium between the organic and aqueous phases and the

ACCEPTED MANUSCRIPT mass transfer of the analytes, thus influencing the extraction efficiency. For the

291

present study, the effect of the vortex time was studied over the time range of 30 s - 5

292

min. Figure S2 (Appendix A) shows the analytical signal intensity for AuNPs versus

293

vortex time. It can be observed that vortex time had no apparent effect on the

294

extraction of AuNPs, probably due to the fact that the contact surface between

295

extraction solvent and aqueous sample was greatly enhanced by the addition of

296

surfactant and the vortex agitation, which greatly speeded up the mass transfer. Thus,

297

1 min was selected for the subsequent experiments.

298

3.1.7. Effect of centrifugation speed and time

NU

SC R

IP

T

290

Phase separation was accelerated by centrifugation. However, potential losses of

300

AuNPs by agglomeration and sedimentation at high speed have to be avoided. The

301

effect of centrifugation speeds ranging from 1,500 to 4,000 rpm on the extraction of

302

AuNPs was tested, and the results are shown in Fig. S3 (Appendix A). It can be seen

303

that the analytical signal intensity of AuNPs remained nearly constant when the

304

centrifugation speed was increased from 1500 to 3000 rpm and then decreased with

305

further increase of the centrifugation speed to 4000 rpm, probably due to the

306

agglomeration of AuNPs. Therefore, 2500 rpm was selected as the optimal

307

centrifugation speed.

AC

CE P

TE

D

MA

299

308

The effect of centrifugation time was studied in the range of 1-5 min. As shown

309

in Fig. S4 (Appendix A), the analytical signal intensity of AuNPs was increased with

310

increasing centrifugation time from 1 to 2 min then kept nearly constant with further

311

increase of the centrifugation time to 5 min. Thus, 3 min was selected as the optimal

312

centrifugation time.

313

3.2. Effects of particle size and coating of AuNPs

314

3.2.1. Particle size of AuNPs

315

In environmental samples, NPs possibly existed in various sizes. Therefore, the

316

effect of AuNPs’ size on their extraction was investigated. Firstly, AuNPs with

317

different sizes were prepared according to the method reported in ref. [41], and the

ACCEPTED MANUSCRIPT size of the self-prepared AuNPs were in the range of 17-21, 47-64 and 80-108 nm,

319

respectively, as shown in Fig. S5a, b and c (Appendix A). To investigate the size

320

effect of AuNPs, AuNPs with different particle size were determined directly by

321

conventional pneumatic nebulization (PN) ICP-MS and ETV-ICP-MS, respectively.

322

And no obvious difference of Au signal intensity was observed by PN/ETV-ICP-MS

323

detection for AuNPs with different sizes, in the same concentration (as Au). In other

324

words, no obvious size effect of AuNPs was observed in PN/ETV-ICP-MS detection.

325

After the SA-DLLME, AuNPs with different sizes were subjected to ETV-ICP-MS

326

detection, and the results are shown in Figure S6 (Appendix A). As can be seen, the

327

intensity of Au was decreased with the increase of AuNPs particle from 17-21 to

328

80-108 nm. It indicated that AuNPs within the size range of 17-108 nm can all be

329

extracted by the proposed method, and the extraction efficiency for AuNPs with larger

330

particle size was a little bit lower than that obtained for AuNPs with smaller particle

331

size.

332

3.2.2. Coating of AuNPs

CE P

TE

D

MA

NU

SC R

IP

T

318

The extraction efficiency for AuNPs with different coatings, namely, citrate,

334

MUA, PVP, MSA, respectively, was studied. As shown in Fig. 4, no obvious variation

335

was observed in the obtained signal intensity for AuNPs with different coatings,

336

indicating that the proposed method is robust and has the application potential for

337

separation/preconcentration of various coating stabilized AuNPs.

338

3.3. The variation of AuNPs in terms of size and shape after SA-DLLME

AC

333

339

To gain an insight in the size distribution of the nanoparticles in real samples, it

340

is necessary to know whether the size distribution of AuNPs is changing during the

341

extraction procedure or not. Figure 5a and b show TEM images of AuNPs before and

342

after extraction, in the aqueous and organic phases, respectively. It can be seen that

343

AuNPs were extracted into the organic phase without significant change in particle

344

size

345

separate/preconcentrate AuNPs, but also effectively preserve the original state of the

and

shape,

indicating

that

the

proposed

method

can

not

only

ACCEPTED MANUSCRIPT 346

AuNPs.

347

3.4. Fraction selectivity For selective determination of AuNPs in natural water samples, the separation of

349

AuNPs from gold ions is required. Hence, the extraction of Au(III) ions with the

350

developed SA-DLLME system was investigated under the optimal conditions. It was

351

found that Au(III) ions could be extracted with the developed SA-DLLME system. To

352

separate AuNPs from Au(III) ions with SA-DLLME system, the use of suitable

353

additive to change Au(III) ions into another form which cannot be extracted is a good

354

choice. Na2S2O3 was identified as a suitable ligand for the separation of Au(III) from

355

AuNPs [14], and it can reduce the Au(III) species to Au(I) which is the most stable

356

oxidation state of gold, ([Au(S2O3)2]3−), thus reduce the extraction efficiency of Au(III)

357

considerably. Therefore, Na2S2O3 was tested as the additive to reduce the Au(III)

358

species to Au(I) ([Au(S2O3)2]3−) in this work, and our preliminary experimental results

359

indicate that only 2% Au(III) ion was extracted after the addition of Na2S2O3. This

360

means that Na2S2O3 could effectively eliminate the interference of dissolved Au(III)

361

ions and the proposed method has the potential to selectively extract and

362

preconcentrate AuNPs even from an aqueous mixture containing Au ions.

CE P

TE

D

MA

NU

SC R

IP

T

348

The impact of Na2S2O3 concentration on the extraction efficiency of AuNPs and

364

Au(III) ions was investigated, respectively, and the results are shown in Fig. 6. As can

365

be seen, with the addition of 50 μL of 1 mol L-1 Na2S2O3 solution per 5 mL of sample

366

solution, the extraction efficiency of Au(III) ion was reduced to around 2%. In other

367

words, 10 mmol L-1 Na2S2O3 can effectively mask Au(III) ions. On the other hand, the

368

addition of Na2S2O3 hardly affected the extraction of AuNPs. Considering that the

369

high salt content may cause NP aggregation. 10 mmol L-1 Na2S2O3 was selected for

370

the sequential experiments.

AC

363

371

We also investigated the effect of co-existing Au(III) concentration on the

372

extraction of AuNPs by spiking various concentration of Au(III) (0-1 mg L−1) to a 1

373

μg L−1 AuNPs suspension with a presence of 10 mmol L-1 Na2S2O3, which was then

374

subjected to SA-DLLME. As shown in Fig. S7 (Appendix A), no apparent difference

ACCEPTED MANUSCRIPT 375

in the analytical signal intensity of AuNPs was observed, even with the presence of

376

Au(III) ions as high as 1 mg L−1. This indicates that high content of Au(III) ions will

377

not interfere with the analysis of AuNPs in the proposed method. To further investigate the feasibility of the proposed method for discriminating

379

ionic and NP-bound gold in real water sample containing dissolved organic matter

380

(NOM), the extraction of Au (III) ions in the presence of HA was processed with and

381

without the addition of Na2S2O3. The results are shown in Fig. S8 (Appendix A). As

382

can be seen, Au (III) ions could be extracted by the proposed DLLME system in the

383

presence of HA, while no obvious signal of Au ions was observed when Na2S2O3 was

384

spiked into the sample solution, indicating that the addition of Na2S2O3 could inhibit

385

the extraction of Au (III) ions in the presence of HA. Meanwhile, the addition of

386

Na2S2O3 would hardly affect the extraction of AuNPs by the proposed SA-DLLME

387

procedure in the presence of HA. It demonstrated the application potential of the

388

developed method to natural water samples.

389

3.5. The anti-interference ability of the proposed SA-DLLME method

CE P

TE

D

MA

NU

SC R

IP

T

378

The robustness and limitations of the proposed method for AuNPs separation and

391

preconcentration from environmental water samples was evaluated by investigating

392

the effect of coexisting ions, the effect of dissolved organic matter (DOM) and the

393

effect of other metal NPs (AgNPs), respectively.

394

3.5.1. Effect of coexisting ions

AC

390

395

To evaluate the selectivity of the developed method, the effect of common

396

coexisting ions, including K+, Na+, Ca2+, Mg2+, Cl−, SO42− and NO3−, on the extraction

397

and determination of AuNPs was studied. For this purpose, 5 mL of an aqueous

398

solution containing 1.0 μg L-1 of AuNPs and a certain amount of interfering ions was

399

subjected to the general procedure. The tolerance limit was defined as the largest

400

amount of coexisting ions, in the presence of which the recovery of the AuNPs could

401

be maintained in the range of 85-115%. The experimental results in Table 2 indicate

402

that 5000 mg L−1 K+, 5000 mg L−1 Na+, 1000 mg L−1 Ca2+, 1000 mg L−1 Mg2+, 8000

ACCEPTED MANUSCRIPT mg L−1 NO3−, 500 mg L−1 SO42− or 3000 mg L−1 Cl− had no significant effect on the

404

extraction and determination of AuNPs.

405

3.5.2. Effect of humid acid

T

403

In environmental water, AuNPs will be stabilized by interaction with dissolved

407

organic matter (DOM). Hence, it is critical to study the potential effect of DOM on

408

the extraction of AuNPs. To simulate the impact of DOM on the extraction of AuNPs,

409

commercially available humic acid (HA) was applied as the representative. The

410

influence of HA at different concentrations on the extraction of AuNPs was

411

investigated. As shown in Fig. S9 (Appendix A), HA in an environmentally relevant

412

concentration range from 0 to 30 mg L-1 has no obvious effect on the extraction of

413

AuNPs.

414

3.5.3. Effect of other nanoparticles

D

MA

NU

SC R

IP

406

Potential interferences from other metal NPs on the extraction of AuNPs were

416

studied. Here, AgNPs as the most commonly used metal NPs in consumer products

417

was exemplarily selected, and spiked into AuNPs solution followed by

418

SA-DLLME-ETV-ICP-MS procedure. As shown in Fig. S10 (Appendix A), no

419

obvious influence of AgNPs on the AuNPs (particle size, 17-21 nm) extraction was

420

observed in the presence of AgNPs up to 100 times that of AuNPs.

AC

CE P

TE

415

421

The above interference study demonstrated that the common coexisting ions,

422

humic acids and other metal NPs in environmental water will not interfere with the

423

analysis of AuNPs, and the developed method was suitable for the determination of

424

AuNPs in environmental water.

425

3.6. Optimization of ETV temperature program

426

To fully remove the extraction solvent from the furnace and prevent the analytes

427

signal loss at the same time, the effect of pyrolysis temperature on the signal intensity

428

was studied by varying pyrolysis temperature in the range of 150-600 oC, with

429

vaporization temperature fixed at 2200 oC and the vaporization time at 4 s. As can be

430

seen from Fig. 7a, the signal intensity of AuNPs was kept constant with increasing the

ACCEPTED MANUSCRIPT 431

pyrolysis temperature from 150 to 400 oC, then decreased from 400 to 600 oC. Thus, a

432

pyrolysis temperature of 300 oC was selected as the pyrolysis condition in this work. Under the selected pyrolysis condition, the effect of vaporization temperature on

434

the signal intensity of AuNPs was studied, and the result is shown in Fig. 7b. As can

435

be seen, the signal intensity of AuNPs was increased rapidly with the increase of the

436

vaporization temperature from 1800 to 2000 oC and then kept constant with further

437

increase of vaporization temperature from 2000 to 2600 oC. Therefore, a vaporization

438

temperature of 2200 oC was used for the subsequent experiments.

SC R

IP

T

433

Besides, the vaporization behavior of AuNPs with different particle sizes (17-21,

440

47-64 and 80-108 nm) and different coatings (citrate, MUA, PVP, MSA; 17-21 nm)

441

were investigated under the optimal ETV temperature program, and the experimental

442

results are shown in Fig. S11 (Appendix A). As can be seen, there is no apparent

443

difference in signal intensity of 20 μgL-1 AuNPs both with different particle sizes and

444

with different coatings, indicating the selected ETV heating program could be used

445

for the analysis of AuNPs with different particle sizes and coatings.

446

3.7. Analytical performance

CE P

TE

D

MA

NU

439

Under the optimized conditions, the analytical performance of the proposed

448

method was evaluated and the linear range, LOD, enrichment factor (EF) along with

449

the relative standard deviation (RSD) were listed in Table 3. According to the IUPAC

450

definition, the limit of detection (LOD, 3s) of the method, based on three times the

451

standard deviation of blank signal intensity in 11 runs, was 2.2 ng L−1 for AuNPs. The

452

relative standard deviation (RSD) for seven replicate analysis of 0.05 μg L-1 citrate

453

stabilized Au NPs (17-21 nm) was 9.3%. The enrichment factor calculated as the

454

slope ratio of the calibration with and without SA-DLLME, was about 152-fold for

455

AuNPs. The extraction efficiency of AuNPs by the proposed method was 76%.

AC

447

456

A comparison of the analytical performance of the proposed method with that of

457

several other methods for the determination of AuNPs is shown in Table 4. As can be

458

seen, the LOD of AuNPs obtained by this method was comparable with that reported

459

in other methods [14, 16] and higher than the method reported in ref. [26], however,

ACCEPTED MANUSCRIPT the proposed method has the fastest extraction kinetics (extraction time, 1 min).

461

Compared with ref. [14, 29], the proposed method did not require digestion or any

462

dilution, which is another advantage. SP-ICP-MS based method exhibited high

463

sensitivity for the analysis of AuNPs, and can provide the information of AuNP size,

464

size distribution, and particle concentration, with the ability of simultaneous detecting

465

dissolved and nanoparticulate forms of Au in a mixture. While the accurate

466

measurement of AuNPs by SP-ICP-MS requires careful experimental design and data

467

interpretation [18]. Comparatively, the proposed SA-DLLME-ETV-ICP-MS merits

468

simple operation and no need for NP reference materials, although the LOD for

469

AuNPs obtained by the proposed method is a little bit higher than those reported in

470

SP-ICP-MS based method [17-18].

471

3.8. Real sample analysis

MA

NU

SC R

IP

T

460

The proposed method was applied to the determination of AuNPs in different

473

environmental water samples including tap water, Yangtze River water and East Lake

474

water. The analytical results are given in Table 5. As can be seen, no AuNPs were

475

found in these natural environmental waters. In order to verify the accuracy of the

476

method, recovery test was conducted and the results are also listed in Table 5. It can

477

be seen that the recovery of AuNPs in three spiked water samples were in the range of

478

89.6-102% at spiked levels of 0.05-0.5 μg/L. To further investigate the loss effect

479

during filtering process, lake water sample has been spiked with AuNPs of 17-21 nm

480

before filtration, and then subjected to the proposed DLLME procedure. The results

481

(Table S1) showed that the recovery of AuNPs in spiked water samples were in the

482

range of 93.2-98.5% at spiked levels of 0.05-0.5 μg/L, indicating no obvious loss of

483

NPs

484

SA-DLLME-ETV-ICP-MS method has good application potential for the analysis of

485

AuNPs in natural water samples.

486

4. Conclusion

AC

CE P

TE

D

472

during

filtration.

These

results

indicate

that

the

proposed

ACCEPTED MANUSCRIPT A novel method of SA-DLLME-ETV-ICP-MS was reported for the selective

488

determination of trace AuNPs in environmental water samples. The developed method

489

is suitable for the determination of AuNPs with a size ranging from 17 to 108 nm,

490

regardless of the coating on AuNPs such as citrate, MUA, PVP or MSA. The

491

commonly co-existing ions, dissolved organic matter and other metal NPs such as

492

AgNPs in the environmental water have no remarkable interference with the

493

determination of AuNPs. Furthermore, AuNPs could preserve their morphology in

494

terms of the size and shape during the extraction process. The proposed

495

SA-DLLME-ETV-ICP-MS method has been successfully applied to the analysis of

496

AuNPs in environmental water samples, and it is featured with simple and fast

497

operation, good selectivity, and good tolerance to the sample matrix interference. It

498

can be extended to the analysis of other metal NPs such as AgNPs in water samples.

MA

NU

SC R

IP

T

487

D

499

Acknowledgements

501

Financial support from the National Nature Science Foundation of China (No.

502

21175102, 21375097), Science Fund for Creative Research Groups of NSFC (No.

503

20921062), the National Basic Research Program of China (973 Program,

504

2013CB933900) and Large-scale Instrument and Equipment Sharing Foundation of

505

Wuhan University are gratefully acknowledged.

506

Appendix A. Supplementary data

507 508 509

AC

CE P

TE

500

Supplementary data to this article can be found online at http://dx.doi.org/

ACCEPTED MANUSCRIPT

References

511

[1] J. Lim, S. A. Majetich, Composite magnetic-plasmonic nanoparticles for

512

biomedicine: manipulation and imaging, Nano Today, 8 (2013), 98-113.

513

[2] N. Khlebtsov, V. Bogatyrev, L. Dykman, B. Khlebtsov, S. Staroverov, A. Shirokov,

514

L. Matora, V. Khanadeev, T. Pylaev, N. Tsyganova, G. Terentyuk, Analytical and

515

theranostic applications of gold nanoparticles and multifunctional nanocomposites,

516

Theranostics, 3 (2013), 167-180.

517

[3] P. K. Jain, X. Huang, I. H. El-Sayed, M. A. El-Sayed, Noble metals on the

518

nanoscale: Optical and photothermal properties and some applications in imaging,

519

sensing, biology, and medicine, Acc. Chem. Res., 41 (2008), 1578-1586.

520

[4] B. F. d. Silva, S. Pérez, P. Gardinalli, R. K. Singhal, A. A. Mozeto, D. Barceló,

521

Analytical chemistry of metallic nanoparticles in natural environments, TrAC-Trend.

522

Anal. Chem., 30 (2011), 528-540.

523

[5] X. D. Zhang, D. Wu, X. Shen, P. X. Liu, N. Yang, B. Zhao, H. Zhang, Y. M. Sun,

524

L. A. Zhang, F. Y. Fan, Size-dependent in vivo toxicity of PEG-coated gold

525

nanoparticles, Int. J. Nanomed., 6 (2011), 2071-2081.

526

[6] W. S. Cho, M. Cho, J. Jeong, M. Choi, H. Y. Cho, B. S. Han, S. H. Kim, H. O.

527

Kim, Y. T. Lim, B. H. Chung, J. Jeong, Acute toxicity and pharmacokinetics of

528

13 nm-sized PEG-coated gold nanoparticles, Toxicol. Appl. Pharmacol., 236 (2009),

529

16-24.

530

[7] X. D. Zhang, H. Y. Wu, D. Wu, Y. Y. Wang, J. H. Chang, Z. B. Zhai, A. M. Meng,

531

P. X. Liu, L. A. Zhang,F. Y. Fan, Toxicologic effects of gold nanoparticles in vivo by

532

different administration routes, Int. J. Nanomed., 5 (2010), 771-781.

533

[8] N. M. Schaeublin, L. K. Braydich-Stolle, E. I. Maurer, K. Park, R. I. MacCuspie,

534

A. R. M. N. Afrooz, R. A. Vaia, N. B. Saleh, S. M. Hussain, Does shape matter?

535

Bioeffects of gold nanomaterials in a human skin cell model, Langmuir, 28 (2012),

536

3248-3258.

537

[9] S. Y. Choi, S. Jeong, S. H. Jang, J. Park, J. H. Park, K. S. Ock, S. Y. Lee, S. W. Joo,

538

In vitro toxicity of serum protein-adsorbed citrate-reduced gold nanoparticles in

AC

CE P

TE

D

MA

NU

SC R

IP

T

510

ACCEPTED MANUSCRIPT human lung adenocarcinoma cells, Toxicol. In Vitro, 26 (2012), 229-237.

540

[10] J. H. Fan, W. T. Li, W. I. Hung, C. P. Chen, J. M. Yeh, Cytotoxicity and

541

differentiation effects of gold nanoparticles to human bone marrow mesenchymal

542

stem cells, Biomed. Eng.-Appl. Basis Commun., 23 (2011), 141-152.

543

[11] A. Ambrosi, M. T. Castañeda, A. J. Killard, M. R. Smyth, S. Alegret, A. Merkoçi,

544

Double-codified gold nanolabels for enhanced immunoanalysis, Anal. Chem., 79

545

(2007), 5232-5240.

546

[12] A. I. López-Lorente, B. M. Simonet, M. Válcarcel, Rapid analysis of gold

547

nanoparticles in liver and river water samples, Analyst, 137 (2012), 3528-3534.

548

[13] H. D. Fiedler, E. E. Drinkel, B. Orzechovicz, E. C. Leopoldino, F. D. Souza, G. I.

549

Almerindo, C. Perdona, F. Nome, Simultaneous nondestructive analysis of palladium,

550

rhodium, platinum, and gold nanoparticles using energy dispersive X-ray fluorescence,

551

Anal. Chem., 85 (2013), 10142-10148.

552

[14] G. Hartmann, M. Schuster, Species selective preconcentration and quantification

553

of gold nanoparticles using cloud point extraction and electrothermal atomic

554

absorption spectrometry, Anal. Chim. Acta., 761 (2013), 27-33.

555

[15] B. Franze, C. Engelhard, Fast separation, characterization, and speciation of gold

556

and silver nanoparticles and their ionic counterparts with micellar electrokinetic

557

chromatography coupled to ICP-MS, Anal. Chem., 86 (2014), 5713-5720.

558

[16] L. Zhang, B. B. Chen, M. He, X. L. Liu, B. Hu, Hydrophilic polymer monolithic

559

capillary microextraction online coupled to ICPMS for the determination of carboxyl

560

group-containing gold nanoparticles in environmental waters, Anal. Chem., 87 (2015),

561

1789-1796.

562

[17] R. Peters, Z. Herrera-Rivera, A. Undas, M. van der Lee, H. Marvin, H.

563

Bouwmeester, S. Weigel, Single particle ICP-MS combined with a data evaluation

564

tool as a routine technique for the analysis of nanoparticles in complex matrices, J.

565

Anal. At. Spectrom., 30 (2015), 1274-1285

566

[18] J. Y. Liu, K. E. Murphy, R. I. MacCuspie, M. R. Winchester. Capabilities of

567

single particle inductively coupled plasma mass spectrometry for the size

568

measurement of nanoparticles: A case study on gold nanoparticles, Anal. Chem., 86

AC

CE P

TE

D

MA

NU

SC R

IP

T

539

ACCEPTED MANUSCRIPT (2014) 3405-3414.

570

[19] Y.B Dan, W.L. Zhang, R.M. Xue, X.M. Ma, C. Stephan, H.L. Shi,

571

Characterization of gold Nanoparticle uptake by tomato plants using enzymatic

572

extraction followed by single-particle inductively coupled plasma-mass spectrometry

573

analysis. Environ. Sci. Technol., 49 (2015) 3007-3014.

574

[20] J. P. Novak, C. Nickerson, S. Franzen, D. L. Feldheim, Purification of

575

molecularly bridged metal nanoparticle arrays by centrifugation and size exclusion

576

chromatography, Anal. Chem., 73 (2001), 5758-5761.

577

[21] J. C. Trefry, J. L. Monahan, K. M. Weaver, A. J. Meyerhoefer, M. M.

578

Markopolous, Z. S. Arnold, D. P. Wooley, I. E. Pavel, Size selection and concentration

579

of silver nanoparticles by tangential flow ultrafiltration for SERS-based biosensors, J.

580

Am. Chem. Soc., 132 (2010), 10970-10972.

581

[22] H. Hagendorfer, R. Kaegi, J. Traber, S. F. L. Mertens, R. Scherrers, C. Ludwig, A.

582

Ulrich, Application of an asymmetric flow field flow fractionation multi-detector

583

approach for metallic engineered nanoparticle characterization-Prospects and

584

limitations demonstrated on Au nanoparticles, Anal. Chim. Acta., 706 (2011),

585

367-378.

586

[23] B. Schmidt, K. Loeschner, N. Hadrup, A. Mortensen, J. J. Sloth, C. Bender Koch,

587

E. H. Larsen, Quantitative characterization of gold nanoparticles by field-flow

588

fractionation coupled online with light scattering detection and inductively coupled

589

plasma mass spectrometry, Anal. Chem., 83 (2011), 2461-2468.

590

[24] L. Li, K. Leopold, M. Schuster, Effective and selective extraction of noble metal

591

nanoparticles from environmental water through a noncovalent reversible reaction on

592

an ionic exchange resin, Chem. Commun., 48 (2012), 9165-9167.

593

[25] L. Li, K. Leopold, Ligand-assisted extraction for separation and preconcentration

594

of gold nanoparticles from waters, Anal. Chem., 84 (2012), 4340-4349.

595

[26] S. W. Su, B. B. Chen, M. He, Z. W. Xiao, B. Hu, A novel strategy for sequential

596

analysis of gold nanoparticles and gold ions in water samples by combining magnetic

597

solid phase extraction with inductively coupled plasma mass spectrometry, J. Anal. At.

598

Spectrom., 29 (2014), 444-453.

AC

CE P

TE

D

MA

NU

SC R

IP

T

569

ACCEPTED MANUSCRIPT [27] J. F. Liu, J. B. Chao, R. Liu, Z. Q. Tan, Y. G. Yin, Y. Wu,G. B. Jiang, Cloud point

600

extraction as an advantageous preconcentration approach for analysis of trace silver

601

nanoparticles in environmental waters, Anal. Chem., 81 (2009), 6496-6502.

602

[28] J. B. Chao, J. F. Liu, S. J. Yu, Y. D. Feng, Z. Q. Tan, R. Liu, Y. G. Yin, Speciation

603

analysis of silver nanoparticles and silver ions in antibacterial products and

604

environmental waters via cloud point extraction-based separation, Anal. Chem., 83

605

(2011), 6875-6882.

606

[29] G. Z. Tsogas, D. L. Giokas,A. G. Vlessidis, Ultratrace determination of silver,

607

gold, and iron oxide nanoparticles by micelle mediated preconcentration/selective

608

back-extraction coupled with flow injection chemiluminescence detection, Anal.

609

Chem., 86 (2014), 3484-3492.

610

[30] S. M. Majedi, H. K. Lee, B. C. Kelly, Chemometric analytical approach for the

611

cloud point extraction and inductively coupled plasma mass spectrometric

612

determination of zinc oxide nanoparticles in water samples, Anal. Chem., 84 (2012),

613

6546-6552.

614

[31] S. M. Majedi, B. C. Kelly, H. K. Lee, Evaluation of a cloud point extraction

615

approach for the preconcentration and quantification of trace CuO nanoparticles in

616

environmental waters, Anal. Chim. Acta, 814 (2014), 39-48.

617

[32] G. Hartmann, C. Hutterer, M. Schuster, Ultra-trace determination of silver

618

nanoparticles in water samples using cloud point extraction and ETAAS, J. Anal. At.

619

Spectrom., 28 (2013), 567-572.

620

[33] M. Rezaee, Y. Assadi, M. R. M. Hosseini, E. Aghaee, F. Ahmadi, S. Berijani,

621

Determination of organic compounds in water using dispersive liquid-liquid

622

microextraction, J. Chromatogr. A, 1116 (2006), 1-9.

623

[34] J. Xiong, B. Hu, Comparison of hollow fiber liquid phase microextraction and

624

dispersive liquid-liquid microextraction for the determination of organosulfur

625

pesticides in environmental and beverage samples by gas chromatography with flame

626

photometric detection, 2008, J. Chromatogr. A, 1193 (2008), 7-18.

627

[35] H. M. Jiang, Y. C. Qin, B. Hu, Dispersive liquid phase microextraction (DLPME)

628

combined with graphite furnace atomic absorption spectrometry (GFAAS) for

AC

CE P

TE

D

MA

NU

SC R

IP

T

599

ACCEPTED MANUSCRIPT determination of trace Co and Ni in environmental water and rice samples, Talanta, 74

630

(2008), 1160-1165.

631

[36] Y. Zhang, J. K. Duan, M. He, B. B. Chen, B. Hu, Dispersive liquid liquid

632

microextraction combined with electrothermal vaporization inductively coupled

633

plasma mass spectrometry for the speciation of inorganic selenium in environmental

634

water samples, Talanta, 115 (2013), 730-736.

635

[37]

636

surfactant-enhanced-emulsification liquid-liquid microextraction combined with gas

637

chromatography-mass spectrometry for the fast determination of phthalate esters in

638

bottled water, J. Chromatogr. A, 1274 (2013), 28-35.

639

[38] M. Moradi, Y. Yamini, A. Esrafili, S. Seidi, Application of surfactant assisted

640

dispersive liquid-liquid microextraction for sample preparation of chlorophenols in

641

water samples, Talanta, 82 (2010), 1864-1869.

642

[39] M. Behbahani, F. Najafi, S. Bagheri, M. K. Bojdi, M. Salarian, A. Bagheri,

643

Application of surfactant assisted dispersive liquid-liquid microextraction as an

644

efficient sample treatment technique for preconcentration and trace detection of

645

zonisamide and carbamazepine in urine and plasma samples, J. Chromatogr. A, 1308

646

(2013), 25-31.

647

[40] Y. Li, P. S. Chen, S. D. Huang, Water with low concentration of surfactant in

648

dispersed solvent-assisted emulsion dispersive liquid-liquid microextraction for the

649

determination of organochlorine pesticides in aqueous samples, J. Chromatogr. A,

650

1300 (2013), 51-57.

651

[41] Q. W. Deng, M. H. Chen, L. M. Kong, X. Zhao, J. Guo, X. D. Wen, Novel

652

coupling of surfactant assisted emulsification dispersive liquid-liquid microextraction

653

with spectrophotometric determination for ultra trace nickel, Spectrochim. Acta, 104

654

A (2013), 64-69.

655

[42] S. M. Yousefi, F. Shemirani, Selective and sensitive speciation analysis of Cr(VI)

656

and Cr(III) in water samples by fiber optic-linear array detection spectrophotometry

657

after ion pair based-surfactant assisted dispersive liquid-liquid microextraction, J.

658

Hazard. Mater., 254-255 (2013), 134-140.

H.

K.

Lee,

SC R

Zhang,

Low-density

solvent-based

vortex-assisted

AC

CE P

TE

D

MA

NU

Y.

IP

T

629

ACCEPTED MANUSCRIPT [43] S. Q. Li, B. Hu, Z. C. Jiang, Direct determination of trace impurities in niobium

660

pentaoxide solid powder with slurry sampling fluorination assisted electrothermal

661

vaporization inductively coupled plasma mass spectrometry, J. Anal. At. Spectrom.,

662

19 (2004) 387-391.

663

[44] G. Frens, Controlled nucleation for the regulation of the particle size in

664

monodisperse gold suspensions, Nature-Phys. Sci., 24 (1973), 20-22.

665

[45] C. H. Yeh, W. T. Chen, H. P. Lin, T. C. Chang,Y. C. Lin, Development of an

666

immunoassay based on impedance measurements utilizing an antibody-nanosilver

667

probe, silver enhancement, and electro-microchip, Sens. Actuat. B Chem., 139 (2009),

668

387-393.

NU

SC R

IP

T

659

MA

669

AC

CE P

TE

D

670

ACCEPTED MANUSCRIPT Figure captions

672

Fig. 1. Effect of extraction solvent on SA-DLLME of AuNPs. Conditions: AuNPs, 1 μg L-1;

673

sample pH, 3.0; sample volume, 5 mL; TX-114 concentration, 0.1% (w/v); extraction time, 1 min;

674

centrifugation speed, 2500 rpm; centrifugation time, 3 min; determination: ETV-ICP-MS; error

675

bars represent the standard deviation (SD) of parallel runs (n=3).

676

Fig. 2. Effect of pH on SA-DLLME of AuNPs. Conditions: AuNPs, 1 μg L-1; sample volume, 5

677

mL; 1,2-dichloroethane, 70 L; TX-114 concentration, 0.1% (w/v); extraction time, 1 min;

678

centrifugation speed, 2500 rpm; centrifugation time, 3 min; determination: ETV-ICP-MS; error

679

bars represent the SD of repeated runs (n=3).

680

Fig. 3. Effect of surfactant on SA-DLLME of AuNPs. Conditions: AuNPs, 1 μg L-1; sample pH,

681

3.0; sample volume, 5 mL; 1,2-dichloroethane, 70 L; extraction time, 1 min; centrifugation speed,

682

2500 rpm; centrifugation time, 3 min; determination: ETV-ICP-MS; error bars denote the SD of

683

repeated runs (n=3).

684

Fig. 4. Effect of particle coating on SA-DLLME of AuNPs. Conditions: AuNPs, 1 μg L-1; sample

685

pH, 3.0; sample volume, 5 mL; 1,2-dichloroethane, 70 L; TX-114 concentration, 0.1% (w/v);

686

extraction time, 1 min; centrifugation speed, 2500 rpm; centrifugation time, 3 min; determination:

687

ETV-ICP-MS; error bars represent the SD of parallel runs (n=3).

688

Fig. 5. TEM images of AuNPs (a)17-21 nm before SA-DLLME and (b)17-21 nm after

689

SA-DLLME.

690

Fig. 6. Effect of Na2S2O3 concentration on SA-DLLME of AuNPs or Au(III). Conditions: AuNPs,

691

Au(III), 1 μg L-1; sample pH, 3.0; sample volume, 5 mL; 1,2-dichloroethane, 70 L; TX-114

692

concentration, 0.1% (w/v); extraction time, 1 min; centrifugation speed, 2500 rpm; centrifugation

693

time, 3 min; determination: ETV-ICP-MS; error bars represent the SD of parallel runs (n=3).

694

Fig. 7. ETV temperature program. (a) Effect of pyrolysis temperature on the signal intensity of

695

AuNPs. Conditions: AuNPs, 20 μg L-1; drying, 110 oC ramp 5 s, hold 5 s; vaporization, 2200 oC,

696

hold 4s; (b) Effect of vaporization temperature on the signal intensity of AuNPs. Conditions:

697

AuNPs, 20 μg L-1; drying, 110 oC ramp 5 s, hold 5 s; pyrolysis, 300 oC ramp 10 s, hold 15 s;

698

determination: ETV-ICP-MS; error bars represent the SD of parallel runs (n=3).

699 700

AC

CE P

TE

D

MA

NU

SC R

IP

T

671

ACCEPTED MANUSCRIPT 701

Table 1 Operating conditions of ETV-ICP-MS ICP-MS 1200 W

Plasma gas

15.0 L min-1

Auxiliary gas

0.9 L min-1

IP

T

RF power

0.7 L min-1

Carrier gas

SC R

Sampling depth Sampler/skimmer diameter orifice

7.0 mm

Nickel 1.0 mm/0.4 mm

Peak pattern

Peak-hop transient

NU

Dwell time Integration mode

MA

Isotopes for detection

50 ms Peak area 197

Au

Electrothermal vaporizer

Carrier gas

TE

Drying

D

Injection volume

0.4 L min-1 110 oC ramp 5 s, hold 5 s 300 oC ramp 10 s, hold 15 s

Vaporization

2200 oC, hold 4s

Cooling

100 oC, hold 5s

Cleaning

2400 oC, hold 3s

CE P

Pyrolysis

AC

702 703

10 μL

ACCEPTED MANUSCRIPT Table 2 Tolerance limits of foreign ions in determination of 1.0 μg L-1 AuNPs in water samples using SA-DLLME-ETV-ICP-MS Tolerance limit of ions (mg L-1)

K+ Na+ Ca2+ Mg2+ NO3SO42Cl-

5000 5000 1000 1000 8000 500 3000

IP

T

Coexisting ions

SC R

704 705

706

AC

CE P

TE

D

MA

NU

707

ACCEPTED MANUSCRIPT 708

Table 3 The analytical performance of SA-DLLME-ETV-ICP-MS for determination of AuNPs Analyte AuNPsa

709

a

710

merits.

711

b

Linear equation (μg L-1) y=486570x+2213.6

r 0.9991

Linear range (μg L-1) 0.01-10

LOD (ng L-1) 2.2

EF 152

RSD%b (n=7) 9.3

IP

T

AuNPs of 17-21 nm with citrate coating were employed for the evaluation of the figures of

SC R

c=0.05 μg L-1

712

AC

CE P

TE

D

MA

NU

713

ACCEPTED MANUSCRIPT

Particle size (nm)

Analytical time

LOD (ng L-1)

Ref.

SPE-GFAAS SPE-GFAAS MSPE-ICP-MS CME-ICP-MS CPE-ETAAS CPE-UV SP-ICP-MS SP-ICP-MS SP-ICP-MS

10-80 10 14-140 3 2-150 15-60 20-200 >20 nm 20-200 nm

>3 h >45 h 17 min 10 min >15 min 40 min -

0.31 3.97 5 1.1 pmol L-1 1a 0.65b 1.2 x 107 L-1 c

SA-DLLME-ETV-ICP-MS

17-108

[25] [24] [26] [16] [14] [29] [17] [19] [18] This work

SC R

NU

MA

a

IP

Method

4 min

2.2d

Au NPs of 60 nm; Au NPs of 40 nm; c particle number concentration for different size AuNPs which can be separated (not LOD); d AuNPs of 20 nm.

D

b

TE

716 717 718 719 720

Table 4 The comparison of analytical performance of this method with other methods for the determination of AuNPs

T

714 715

AC

CE P

721

ACCEPTED MANUSCRIPT Table 5 Analytical results (mean ± s.d., n = 3) for AuNPs in real water samples

East Lake water

Yangtze River water

Found (μg L-1)

Recovery (%)

0 0.05 0.5 0 0.05 0.5 0 0.05

ND 0.048±0.003 0.51±0.04 ND 0.046±0.004 0.47±0.04 ND 0.045±0.005

95.3 102 91.5 93.6 89.6

0.5

0.50±0.04

CE P

TE

D

MA

NU

ND: Not detected.

AC

723 724

a

T

Tap water

Added (μg L-1)

IP

Sample

SC R

722

101

ACCEPTED MANUSCRIPT

4x10

5

2x10

5

1x10

5

T

5

IP

3x10

0 CCl4

CHCl3

SC R

Signal intensity of 197Au

AuNPs

C2H4Cl2

C6H5Cl

Different extraction solvents

725

Fig. 1

NU

726 727

AC

CE P

TE

D

MA

728

4x10

5

3x10

5

2x10

5

1x10

5

AuNPs

T

5

IP

5x10

0 2

3

4

5

730

Fig. 2

6

NU

729

pH

SC R

Signal intensity of 197Au

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

731

7

8

3x10

5

2x10

5

1x10

5

AuNPs

T

5

IP

4x10

0 TX-100 TX-114 CTAB

SDS

732 Fig. 3

MeOH

NU

733

SC R

Signal intensity of 197Au

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

734

EtOH

4x10

5

3x10

5

2x10

5

1x10

5

AuNPs

T

5

IP

5x10

SC R

Signal intensity of 197Au

ACCEPTED MANUSCRIPT

0 Citrate

MSA

PVP

Different coatings

NU

735 736

Fig. 4

AC

CE P

TE

D

MA

737

MUA

738 739 740

Fig. 5

AC

CE P

TE

D

MA

NU

741

SC R

IP

T

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

AuNPs Au(III)

80

IP

T

60 40

SC R

Extraction efficiency(%)

100

20 0 0

5

10

Concentration of Na2S2O3 (mM)

743

Fig. 6

NU

742

MA

744

AC

CE P

TE

D

745

15

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

D

746

TE

747 748

AC

CE P

749

Fig. 7

ACCEPTED MANUSCRIPT Highlights

750

> SA-DLLME combined with ETV-ICP-MS was proposed for the analysis of AuNPs

752

in environmental water.

753

> AuNPs with both different sizes and different coatings could be extracted.

754

> AuNPs could preserve their morphology in terms of the size and shape during the

755

extraction process.

756

> The proposed method is simple, sensitive, selective, and no digestion or dilution

757

required.

AC

CE P

TE

D

MA

NU

SC R

IP

T

751