Highly Sensitive Ratiometric Fluorescent Sensor for Trinitrotoluene

Subscriber access provided by Southern Cross University Library

Article

A High Sensitive Ratiometric Fluorescent Sensor for TNT Based on InnerFilter Effect between Gold Nanoparticles and Fluorescent Nanoparticles Hongzhi Lu, Shuai Quan, and Shoufang Xu J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.7b03986 • Publication Date (Web): 16 Oct 2017 Downloaded from http://pubs.acs.org on October 18, 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.

Journal of Agricultural and Food Chemistry 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 32

Journal of Agricultural and Food Chemistry

1

A High Sensitive Ratiometric Fluorescent Sensor for TNT Based on

2

Inner-Filter Effect between Gold Nanoparticles and Fluorescent

3

Nanoparticles

4

Hongzhi Lua，Shuai Quanb, Shoufang Xu* b

5 6 7

a: School of Chemistry and Chemical Engineering, Linyi University, Linyi 276005,

8

China.

9

b: School of Materials Science and Engineering, Linyi University, Linyi 276005,

10

China.

11 12 13 14

*Corresponding author

15

Tel.: +86 539 7258661; fax: +86 539 7258661.

16

E-mail address: [email protected] (S.F. Xu)

17 18 19 20 21 22 1

ACS Paragon Plus Environment


23 24

ABSTRACT

25

In this work, we developed a simple and sensitive ratiometric fluorescent assay for

26

sensing TNT based on inner filter effect (IFE) between gold nanoparticles (AuNPs)

27

and ratiometric fluorescent nanoparticles (RFNs) which was designed by hybridizing

28

green emissive carbon dots (CDs) and red emissive quantum dots (QDs) into silica

29

sphere as fluorophore pair. AuNPs in its dispersion state can be a powerful absorber to

30

quench CDs while the aggregated AuNPs can quench QDs in the IFE-based

31

fluorescent assays due to complementary overlap between the absorption spectrum of

32

AuNPs and emission spectrum of RFNs. Due to the fact that TNT can induce the

33

aggregation of AuNPs, with the addition of TNT, the fluorescent of QDs can be

34

quenched while the fluorescent of CDs would be recovered. Then ratiometric

35

fluorescent detection of TNT is feasible. The present IFE-based ratiometric

36

fluorescent sensor can detect TNT ranging from 0.1nM to 270 nM with a detection

37

limit of 0.029 nM. In addition, the developed method was successfully applied to

38

investigate TNT in water and soil samples with satisfactory recoveries ranged from 95%

39

to 103% with precision below 4.5%. The simple sensing approach proposed here

40

could improve the sensitivity of colorimetric analysis by changing the UV analysis to

41

ratiometric fluorescent analysis, and promote the development of dual mode detection

42

system.

43 44

KEYWORDS: ratiometric fluorescent detection; inner filter effect; TNT; gold 2


Page 2 of 32

Page 3 of 32


45

nanoparticles.

46 47

INTRODUCTION

48

Fluorescence sensors have attracted great interest in practical applications because of

49

their high sensitivity, simplicity and cost-effective. Traditionally, the fluorescence

50

intensity (FI) of the sensor was response to target at a single wavelength, which could

51

be easily affected by a variety of factors, such as probe concentration, excitation

52

intensity, instrumental efficiency and measurement condition. Fortunately, these

53

problems can be solved by ratiometric fluorescence sensors, which possess two

54

fluorescence signals under single wavelength excitation. The ratiometric fluorescent

55

probes improve the sensitivity and accuracy by excludes numerous background

56

interferences. In addition, visualization of the analyte on the basis of two FI ratios can

57

be achieved. Therefore, taking the superior advantage of ratiometric fluorescent

58

techinique, many ratiometric fluorescent sensors have been designed. Ratiometric

59

fluorescent sensors can be divided into two groups. One was double fluorophore dual

60

emission ratiometric fluorescent probes1,2 which involved in two different fluorophore,

61

such as carbon dots (CDs), quantum dots (QDs), gold nanoclusters (AuNCs) or

62

fluorescent dyes. The two different fluorophores can be linked together by core-shell

63

structure3-5 (the references signal was encapsulated into silica particles while the

64

response signal was grafted onto the surface of the silica) or chemical coupling 6, 7 by

65

the functional groups on the surface of the fluorophores. The other was single

66

fluorophore dual emission probes using a special fluorophore which have two 3



Page 4 of 32

8-10

67

resolved fluorescent emission peak under single excitation wavelength

. When

68

detection the target, the FI of the response signal changed based on energy transfer or

69

electron transfer. However, rare reference reported inner filter effect based ratiometric

70

fluorescent sensor11.

71

Inner filter effect (IFE) is based on the absorption of the excitation or emission

72

light by absorbers in the detection system, and the key point is the absorption spectra

73

of the absorbers overlaps with the fluorescence excitation or emission spectra of

74

fluorophores. IFE based sensors can enhance the sensitivity and selectivity by

75

converting the absorption signals into fluorescence signals. Meanwhile, the

76

fabrication process is flexibility and simplicity because they do not require any

77

covalent linking between IFE acceptor and fluorophore. So IFE has emerged as an

78

efficient and useful strategy for design and development of novel sensors11-14.

79

Ascribed to the unique characteristics of the gold nanoparticles (AuNPs) that can

80

absorb the excitation or emission light of the fluorophore, many studies have reported

81

proposing sensors via the IFE between AuNPs and fluorescent nanoparticles 15-17.

82

In the meanwhile, AuNPs attracted much attention in colorimetric detection due

83

to their high extinction coefficient in the visible region and color-tunable behavior

84

that depends on the inter-particle distance18-20 or morphology change

85

display wine red color in dispersion state while display blue color in the aggregation

86

state, correspondingly, the maximum absorption peak of AuNPs showed a significant

87

red shift.

88

21-22

. AuNPs

Inspired by the above mentioned works, we attempt to demonstrate a new IFE 4


Page 5 of 32


89

based

90

ratiometric fluorescence sensors using AuNPs as IFE absorber. Considering that IFE

91

only occurs effectively when the absorption band of the absorber overlap well with

92

fluorophore. Choice the appropriate fluorophore is the key to this strategy.

93

Fluorescent nanoparticles including CDs

94

considerable attention in recent years owing to their remarkable advantages, such as

95

unique optical properties, good photostability and biocompatibility, and facile

96

preparation. It is feasible to choose the appropriate method to prepare fluorescent

97

nanoparticles with the specified emission wavelength.

23

, QDs

24, 25

, AuNCs

26

, have attracted

98

In this work, we demonstrated a new IFE based ratiometric fluorescence sensors

99

using AuNPs as IFE absorber and green emissive CDs and red emissive QDs as IFE

100

fluorophores. Trinitrotoluene (TNT), a highly explosive and environmentally

101

detrimental substance, which can induce the aggregation of AuNPs, was chosen as the

102

proof of concept

103

absorption at 520 nm can quench green emissive CDs based on IFE while the red

104

emissive QDs keep constant. By addition of TNT, the cysteamine modified AuNPs

105

aggregated induced by the donor–acceptor (D-A) interaction between TNT and

106

primary amines, and a new and strong absorption peak at 700 nm appeared.

107

Correspondingly, red emissive QDs were quenched while green emissive CDs were

108

recovered based on IFE. Using this strategy, colorimetric and fluorescent dual mode

109

detection can be build, also colorimetric analysis can be converted to highly sensitive

110

fluorescence detection. This study provided a new technology for construction of

18

. Without TNT, AuNPs in dispersion state with maximum

5



111

ratiometric fluorescence sensors.

112

MATERIALS AND METHODS

113

Materials

and

Chemicals.

Trisodium

citrate,

cysteamine,

Page 6 of 32

phenol,

114

ethylenediamine, tellurium powder (99.8%), CdCl2 (99%), NaBH4 (99%),

115

dinitrotoluene (DNT), 4-nitrophenol (4-NP) and tetraethyl orthosilicate (TEOS) were

116

purchased from Tianjin Reagent Plant (Tianjin, China). TNT standard solution (1.00

117

mg/mL in

118

3-mer-captopropionicacid (MPA) (99%) were purchased from J&K Technology Ltd.

119

(Beijng, China). TNT standard solution was diluted with ultrapure water for

120

fluorescent detection.

methanol),

hydrogen

tetrachloroaurate

(III)

hydrate

(HAuCl4),

121

Fluorescence measurements were carried out using a Fluoromax-4 Spectro-

122

fluorometer (Horiba Scientific). A transmission electron microscope (TEM,

123

JEM-2100F) were used for the morphological evaluation. A UV-3600 ultraviolet

124

spectrophotometry (Shimadzu, Japan) was employed for UV-Vis spectra recording.

125

An AGL-9406 portable UV lamp with 365 nm and 254nm emission was used for

126

taking photos.

127

Preparation of AuNPs, CDs and QDs. AuNPs were prepared by a classical

128

trisodium citrate reduction method as reported 27. Briefly, 38.8 mM trisodium citrate

129

(5 mL) was rapidly added to 1mM HAuCl4 (50 mL) boiling solution and the solution

130

was kept continually boiling for another 30 min to give a wine-red solution with

131

maximum absorbance at 510 nm. Then the citrate-stabilized AuNPs solution was

132

mixed with cysteamine (500 nM) and stirred at room temperature for 3 h to complete 6


Page 7 of 32


133

the substitution of ligands on the surface of AuNPs then stored in a refrigerator at 4℃

134

for further use.

135

CDs with maximum emissive wavelength at 520 nm were prepared by a typical

136

solvent thermal method using phenol and ethylenediamine as precursor 28. In a typical

137

way, phenol (25 mL, 0.1 M) and ethylenediamine (150 µL) was mixted, then

138

transferred into a 50 mL Teflon-lined stainless autoclave. The precursor solution was

139

heated to and maintained at 180 °C for 10 h to obtain dark yellow solution. After

140

centrifugation at 12,000 rpm for 10 minutes, the supernatant was collected and

141

dialyzed for 5 h (molecular porous membrane tubing, cutoff 1000) at room

142

temperature to remove the unreacted moieties.

143

Near infrared CdTe/Cds QDs with maximum wavelength at 700 nm was

144

prepared as follow29. First, 0.25 mM CdCl2 2.5H2O was dissolved in 50 mL ultrapure

145

water, followed by addition of 37 µL MPA, then the solution was adjusted to pH 12.2

146

by sodium hydroxide. The dissolved oxygen in the water was removed by N2 before

147

1.0╳10-5M freshly prepared NaHTe was injected. The mixture was reacted in a

148

refrigerator at 4 ℃ for one day, then followed by aged at 90 ℃ for 10 h to obtain light

149

yellow solution.

150

Preparation of Ratiometric Fluorescent Nanoparticles (RFNs). Ratiometric

151

fluorescent nanoparticles (RFNs) were prepared by embedding CDs and QDs into

152

SiO2 using simple one step reverse-microemulsion method

153

TX-100, 1.8 mL of n-hexanol, 7.5 mL of cyclohexane, 500 µL of QDs, 100 µL of CDs,

154

and 60 µL of ammonia to form a microemulsion. Then 100 µL of TEOS was added to 7


30

. Typically, 1.8 mL of


155

the microemulsion system to initiate the hydrolysis and the mixture was stirred for 10

156

h at room temperature. Finally, the microemulsion was broken by acetone, and the

157

resultant precipitate was washed in sequence with ethanol and water.

158

Detection of TNT. For the direct visualization of TNT with the naked eyes, different

159

concentrations of TNT and 1.0 mL of AuNPs was added to distilled water, with total

160

volume of the solution was 2.0 mL and the final concentrations of TNT ranging from

161

10 nM to 1500 nM. After 10 min, the dispersions were measured by UV/Vis

162

spectrometer or observed directly by naked eyes.

163

For ratiometric fluorescent measurements, RFN was first added to 1.0 mL of

164

AuNPs, and then different concentrations of TNT was added to the dispersion with

165

total volume of the solution was 2.0 mL. The final concentrations of TNT in ranging

166

from 0.1 nM to 300 nM while the final concentrations of RFN as 30 mg/L.

167

Selectivity and Interference Experiments. In order to test the selectivity of this

168

strategy, the fluorescent response to other explosives, such as 4-NP and DNT were

169

examined by the similar procedure mentioned above. In the meanwhile, metal ions,

170

such as Cu2+, Hg2+, Ag+, Pb2+, which may quench fluorescent nanoparticles, and Na+，

171

K+，which may induce the aggregation of AuNPs at high concentration were detected.

172

For interference experiment, 4-NP, DNT, Cu2+, Hg2+, Ag+, Pb2+ as interferences were

173

simultaneously added while TNT and interferences fixed at 100 nM.

174

Real Sample Detection. River water and soil samples were employed to verify the

175

feasibility of the method in the real sample detection. River water sample (Yi River,

176

Linyi) was filtered through 0.45 µm membrane to remove suspended particles. Soil 8


Page 8 of 32

Page 9 of 32


177

samples were extracted by ACN to retrieve TNT according to the methods reported

178

previously18, 31. For detection of TNT in water and soil samples, pure and TNT-spiked

179

samples were added to an aqueous dispersion of AuNPs and RNF with the

180

concentrations of AuNPs and RNF essentially the same as those for the direct

181

detection of TNT.

182

RESULTS AND DISCUSSION

183

Preparation of RFNs. It’s well known that AuNPs have been widely used in

184

various research based on the color change caused by the controllable change in their

185

dispersion/aggregation states. In the dispersion state, AuNPs display wine red color

186

with absorbance peak at 520 nm. In the aggregation state, aggregated AuNPs display

187

blue color and a new and strong absorbance peak at 700 nm. In this present work, we

188

propose a novel ratiometric fluorescent sensor based on IFE between AuNPs and

189

fluorescent nanoparticles. The detection strategy is shown in Scheme 1. The RFNs

190

proposed two fluorescent nanoparticles, one was CDs with maximal emission

191

wavelength at 520 nm, and the other was CdTe/CdS QDs with maximal emission

192

wavelength at 700 nm. When the AuNPs in dispersion state, the FI of RFNs at 520 nm

193

can be significantly quenched by AuNPs based on IFE. When the AuNPs in

194

aggregation state, the FI of RFNs at 700 nm can be significantly quenched by

195

aggregated AuNPs. During this process, the color change of AuNPs can be

196

specifically induced by target molecule TNT. By detection the ratio of FI at 520 nm

197

and 700 nm, the concentration of target TNT can be measured. Using this method,

198

colorimetric and fluorescent dual mode detection can be build, also colorimetric 9



199

analysis can be converted to higher sensitive fluorescence detection.

200

Scheme 1

201

In order to achieve ratiometric fluorescent detection based on IFE between

202

AuNPs and fluorescent nanoparticles, to prepare fluorescent nanoparticles which

203

emission spectra overlapping with AuNPs absorption spectrum is the foundation. So,

204

CDs with maximal emission wavelength at 520 nm and CdTe/CdS QDs with maximal

205

emission wavelength at 700 nm were prepared as reported, and their fluorescence

206

spectrum and photographs under UV illumination were showed in Figure 1A.

207

Many strategies have been reported to propose ratiometric fluorescent sensor,

208

such as core-shell structure or covalently bonding methods. In this work, we hope that

209

the FI change is caused by the change of AuNPs dispersion state, rather than the direct

210

interaction of the target with the fluorescent substance, so that the two fluorescent

211

nanoparticles were both encapsulated into SiO2 by reverse-microemulsion method.

212

The fluorescence spectrum and TEM image of RFNs were shown in Figure 1. From

213

Figure 1A we can see that RFNs displayed two well resolved emission bands with

214

maximal emission wavelength at 520 nm and 700 nm under a single wavelength

215

excitation (excitation at 380 nm). When excited under 365 nm UV illumination, the

216

RFNs showed a strong yellow emission (c), differing from that of the original green

217

CDs (a) and red QD (b). From the TEM image of RFNs we can see that RFNs were

218

spherical in shape with the sizes of about 50 nm, and the fluorescence nanoparticles

219

can be observed clearly (the black dots in SiO2). The silica shell ensure the photo

220

stability of fluorescence nanoparticles and the small particle size of RFNs ensures its 10


Page 10 of 32

Page 11 of 32


221

stability during the detection process. Furthermore, with the protection of silica shell,

222

a variety of material interference can be effectively avoided.

223

Figure 1

224

IFE between RFNs and AuNPs. Reference18 reported that TNT can induce the

225

aggregation of the cysteamine-stabilized AuNPs through the D–A interaction between

226

TNT and primary amine. The nature of the aggregation of AuNPs induced by TNT

227

was discussed in detail in the previous work 18. The stabilizer for AuNPs played the

228

key role. Two functional groups were essential to the stabilizer. One, typically a

229

mercapto group, can act with gold nanoparticles with Au-S covalent bond, and the

230

other was amino group, which can interact with TNT through D–A interactions. From

231

the chemical structure we can see that cysteamine was an ideal choice, so cysteamine

232

was used as the stabilizer for Au NPs to facilitate the D–A interaction between TNT

233

and the primary amine for direct visualization of TNT. Similarly, glutathione, cystein

234

and homocysteine also have SH- and –NH2 groups. So, the effect of glutathione,

235

cystein and homocysteine also were studied as stabilizer of Au NPs. Experiment

236

results indicated that under the same experiment condition, using glutathione, cystein

237

and homocysteine as stabilizer, TNT cannot induce the aggregation of AuNPs

238

effectively. This can be explained from the chemical structure of glutathione, cystein

239

and homocysteine. For cystein and homocysteine, in addition to mercapto and amino

240

groups, there is a carboxyl group present at the end of the chain, which may interfere

241

the interaction of TNT and amino group by steric effect or electronic effect.

242

Glutathione, have more complex structure, is not available the D-A interaction 11



243

between TNT and the primary amine. So in this work, cysteamine was adopted to

244

facilitate the TNT induced Au NP aggregation for visualization of TNT. The

245

feasibility of the strategy was verified by adding 1 µM TNT into AuNPs solutions.

246

From Figure 2A we can see that AuNPs displayed wine red color with absorbance

247

peak at 520 nm. After addition of 1 µM TNT, the solution changed from wine red to

248

blue, accompanied by the decrease of absorbance at 520 nm, and a new and strong

249

absorbance peak appeared at 700 nm, which ascribed to the absorbance of the

250

aggregated AuNPs. The TEM images of AuNPs before (a) and after (b) addition of

251

TNT confirmed the aggregation of AuNPs. It should be noted that many factors can

252

induce the aggregation of AuNPs, such as environment polarity and viscosity,

253

temperature, buffering ion concentration. So in order to achieve the detection of TNT

254

with high selectivity, the experimental conditions need to be optimized to ensure that

255

the agglomeration of gold is due to TNT. The potential interference from salts, such

256

as sodium chloride, and surfactants were studied detail in the previous work18. The

257

experimental results indicated that sole addition of sodium chloride with final

258

concentrations higher than 10 mM to the AuNPs suspension could lead to the

259

aggregation of the AuNPs. However, the co-existence of sodium chloride with final

260

concentration lower than 3 mM in the suspension did not interfere with TNT detection.

261

For the surfactants studies revealed that the co-existence of sodium dodecyl sulfate at

262

0.1mM did not interfere with TNT detection. For the temperature, the AuNPs were

263

stable for more than 2 month under room temperature (lower than 50 ℃).

264

Then the IFE between AuNPs and RNFs was confirmed. The remarkable overlap 12


Page 12 of 32

Page 13 of 32


265

between the fluorescence emission spectrum and the absorption spectrum of the

266

AuNPs was an indispensable factor for IFE. From Figure 2B we can see that the

267

fluorescence emission spectrum of CDs (line 1) was overlapped well with the

268

absorption spectrum of the dispersion AuNPs (line 2), and the FI of CDs decreased

269

with the addition of AuNPs (line 3). With the increase of the amount of AuNPs, the FI

270

of CDs gradually decreased. In the meanwhile, the fluorescence emission spectrum of

271

QDs (line 4) was overlapped well with the absorption spectrum of the aggregated

272

AuNPs (line 5), and the FI of QDs decreased with the addition of aggregated AuNPs

273

(line 6). Figure 2A confirmed that TNT can induce the aggregation of AuNPs, and

274

Figure 2B verified the feasibility of IFE between AuNPs and CDs and QDs. So we

275

can draw the conclusion safely that ratiometric fluorescence detection of TNT based

276

on IFE between AuNPs and RFNs was feasibility.

277

Figure 2

278

TNT Detection. The experimental conditions for TNT detection, such as

279

concentration of AuNPs and RFNs, reaction time were optimized in this study. The

280

amount of AuNPs has an important effect on the detection process. On the one hand,

281

the amount of AuNPs affects the quenching effect of fluorescent nanoparticles. On the

282

other hand, the amount of AuNPs affects the sensitivity and linear range of TNT

283

colorimetric analysis. We first investigated the effect of the amount of AuNPs on TNT

284

colorimetric analysis. From Figure S1, the UV spectrum indicated that at lower

285

AuNPs concentration (10%-20%), the change of absorbance at 520 nm was weak,

286

accompanied by prominent change of absorbance at 645 nm (It should be noted that 13



287

the concentration of AuNPs is higher, the red shift of the absorbance peak is more

288

obvious after it changed from dispersion state to aggregation state. When the

289

concentration of AuNPs is 10%, the maximum absorbance peak at 645 nm at

290

aggregated state, while the maximum absorbance peak at 710 nm at aggregated state

291

when the AuNPs used directly without dilution). When increased the concentration of

292

AuNPs (40%-60%), the change of absorbance both at 520 nm and 700 nm was

293

remarkable. Further increase the concentration of AuNPs to 80%, the change of

294

absorbance both at 520 nm and at 700 nm without obvious improvement. From the

295

photos in Figure S1A we can see that when the concentration of AuNPs is higher, the

296

solution displayed dark red color under sunlight. When addition of TNT, the color

297

change from red to blue is not obvious, implying the sensitivity is relatively lower.

298

When the concentration of AuNPs is lower, the solution displayed lighter red color

299

under sunlight. When addition of the same concentration of TNT, the color change

300

from red to blue is obvious, indicating the higher sensitivity. In the same time, when

301

increase the concentration of AuNPs from 10%-50%, an obviously quench of FL at

302

520 nm can be observed (Figure S1B). Further increase the concentration of AuNPs

303

to 80%, the fluorescent intensity almost be quenched. Considering from the two

304

aspect, AuNPs was used after diluted one times (50%).

305

The dosage of RFNs was optimized due to the parameter has a great impact on

306

the detection sensitivity. From Figure S2 we can see that the FI both at 520 nm and

307

700 nm increased obviously with increasing of the amount of RFNs from 10 mg/L to

308

30 mg/L. However, further increasing the amount of RFN from 30 mg/L to 50 mg/L 14


Page 14 of 32

Page 15 of 32


309

gave no prominent help to increase the FI both at 520 nm and 700 nm. Addition of the

310

same amount of Au NPs resulted in the decrease of FI at 520 nm. After calculated the

311

quenching amount, defined as (F0-F)/F0 at 520 nm, we can see that when the amount

312

of RFN fixed at 30 mg/L, the probe displayed the highest sensitivity.

313

Next, the reaction time is examined. Since the process of IFE between AuNPs

314

and RFNs does not need time, the time is needed for TNT to induce the aggregation of

315

AuNPs. So use the ratio of A700/A520 in the UV absorption spectrum as index, the

316

reaction time was investigated. From Figure S3 we can see that as the reaction time

317

increases, the ratio of A700/A520 gradually increased and reached equilibrium at 10 min.

318

Therefore, 10 min was selected as the optimized reaction time. After 10 min of TNT

319

addition, the UV spectrum and fluorescence spectrum were measured.

320

Under the optimized condition, different amounts of TNT with concentrations

321

from 10 nM to 1500 nM were added to aqueous suspensions of AuNPs, and the results

322

were listed in Figure 3. From Figure 3A we can see that the increasing the

323

concentration of TNT resulted in a clear decrease in the absorbance at 520 nm and an

324

increase in the absorbance at 700 nm. And the color of the suspension gradually

325

changed from initially wine red to blue. Furthermore, the ratio of A700 to A520 was

326

found to be linear with the TNT concentration range from 10 nM to 1200 nM with the

327

regression equation is A700/A520=0.089+0.76 CTNT (µM), with the corresponding

328

regression coefficient is 0.994, and the limit of detection (LOD) for TNT is 2.63 nM.

329

In this work, the limit of detection (LOD) was calculated by the Bessel equation:

330

LOD=3s/b, s is the standard deviation of fluorescence intensity after repeated 15



331

determination for 12 times and b is the slope of the fitted curve.

332

In the meanwhile, TNT was detected by RFNs. To verify the feasibility of

333

fluorescent detection in improving the sensitivity, we reduced the concentration of

334

TNT. Different amounts of TNT ranging from 0.1 nM to 300 nM were added to

335

aqueous suspensions of AuNPs and RFNs, and the results were listed in Figure 3B.

336

With the addition of TNT, the FI at 700 nm decreased and FI at 520 nm increased.

337

And the fluorescence color of the suspension under 365 nm UV light changed from

338

initially red to yellow, then to green. The decrease of FI at 700 nm ascribe to the

339

increase of absorbance at 700 nm and the increase of FI at 520 nm ascribe to the

340

decrease of absorbance at 520 nm. We also found the ratio of F520 to F700 was found to

341

be linear with the TNT concentration range from 0.1 nM to 270 nM with the

342

regression equation is F520/F700=0.362+0.0154CTNT (nM), and the corresponding

343

regression coefficient is 0.998, and the limit of detection (LOD) for TNT is 0.029 nM.

344

At low concentrations of 0.1-10 nM, UV analysis did not detect the differences in

345

absorbance (Figure S4). When compared colorimetric analysis with fluorescent

346

detection, we can see that ratiometric fluorescent detection displayed higher

347

sensitivity and more distinguished color change.

348

Figure 3

349

What is more, a comparison between our method and other reported methods for 32-36

350

TNT detection were summarized in Table1

. Compared with other sensing

351

methods, this proposed ratiometric fluorescent method was most sensitive. These

352

excellent properties substantially enable the practical application of our method for 16


Page 16 of 32

Page 17 of 32


353

detection of TNT in real samples.

354

Selectivity of TNT Detection. In addition to the properties described above,

355

selectivity was an important index to evaluate the proposed method. Control

356

experiments with the addition of other explosives and derivatives, such as 4-NP, DNT

357

to the AuNPs suspension or AuNPs/RFNs mixture suspension were carried out to

358

evaluate the specificity of the sensor for detection of TNT. When fixed the

359

concentration of those controls at 500 nM, no obvious color change (under UV light

360

or under sunlight) was observed, and it did not result in a change in the absorbance

361

spectrum or fluorescence spectrum. The result indicated that these species do not

362

interfere with the detection of TNT. The excellent selectivity of the present method to

363

TNT may be due to that their electron deficiencies property. For 4-NP and DNT there

364

are only one and two electron-withdrawing nitro groups in each aromatic ring,

365

respectively, while TNT own three nitro groups in each aromatic ring. Considering

366

real applications, especially in environmental samples, some metal ions may be

367

interference the detection, especially Hg2+, Cu2+, Ag+, which have already been

368

reported can quench QDs heavily. The selectivity of the probes in the presence of

369

Hg2+, Cu2+, Ag+, Pb2+ was tested. From Figure 4 we can see that metal ions did not

370

displayed measurable effect on TNT detection, even when their concentration was 10

371

times higher than TNT. This may be ascribe to the fact that those CDs or QDs were

372

embedded into SiO2 particle, with the protection of SiO2 layer, the interaction of

373

electron transfer between QDs and metal ions were prevented. Reference reported that

374

AuNPs also can be induced to aggregation under the higher concentration of salts, so 17



375

we tested the effect of Na+, K+ on the probe. The experimental result indicated that

376

Na+ or K+ with the concentration of lower than 1 mM have no measurable effect on

377

TNT detection. Figure 4

378 379

Furthermore, competitive experiments were performed by adding interferences with

380

TNT simultaneously into AuNPs or RFNs. From Figure 4B we can see that the

381

addition of interferences did not cause obvious color change and absorbance change

382

for the UV detection system. When adding TNT into AuNPs solutions with

383

interferences, AuNPs still can be induce to aggregation state. The color change and

384

absorbance change caused by TNT in AuNPs or in AuNPs with interferences system

385

were almost the same. During the RFNs fluorescence detection process, the similar

386

results were obtained. Those interferences did not cause influence for the detection of

387

TNT in complicated matrix.

388

Real Sample Detection. To investigate the potential practical application of this

389

method, detection of TNT in river water and soil samples were carried out. No change

390

in the fluorescence color or no fluorescence responses of RFN system were observed.

391

Then spiked experiments were carried out. The average recovery test was made by

392

using the standard addition method, and the results are listed in Table 2. The

393

recoveries of TNT in samples were between 95% and 103%. The relative standard

394

deviation (RSD) was lower than 4.5%. When compared the ratiometric fluorescence

395

detection with UV detection, fluorescence detection displayed higher sensitivity and

396

accuracy. These results suggested the potential application of this method for TNT 18


Page 18 of 32

Page 19 of 32


397

detection in these environmental matrices.

398

In summary, we developed a novel ratiometric fluorescence sensor for TNT

399

detection based on the IFE between the RFNs and AuNPs, and also based on the fact

400

that TNT can induce the aggregation of AuNPs. This biosensor displayed many

401

advantages, such as highly sensitivity, high selectivity and simplicity in preparation.

402

The proposed method was successfully applied to determination of TNT in water and

403

soil samples with satisfactory results. Using this strategy, colorimetric and fluorescent

404

dual mode detection can be build, also colorimetric analysis can be converted to

405

highly sensitive fluorescence detection.

406

ACKNOWLEDGEMENTS

407

We thanks NSFC (NO. 21307052, 21777065), the Natural Science Foundation of

408

Shandong Province of China (ZR201610).

409 410

(Supporting Information: Optimization of detection conditions; The absorption spectra

411

of AuNPs upon the addition of different concentration of TNT)

412 413

REFERENCE

414

(1) Liu, C., Ning, D.H., Zhang, C., Liu, Z.J., Zhang, R.L., Zhao, J., Zhao, T.T., Liu, B.H., Zhang,

415

Z.P., Dual-colored carbon dot ratiometric fluorescent test paper based on a specific spectral energy

416

transfer for semiquantitative assay of copper ions. ACS Appl. Mater. Interfaces 2017, 9,

417

18897-18903.

418

(2) Wu, L., Guo, Q., Liu, Y., Sun, Q., Fluorescence resonance energy transfer-based ratiometric

419

fluorescent probe for detection of Zn2+ using a dual-emission silica-coated quantum dots mixture.

420

Anal. Chem. 2015, 87, 5318-5323. 19



421

(3) Yan, X., Li, H.X., Zheng, W.S., Su, X.G., Visual and fluorescent detection of tyrosinase

422

activity by using a dual-emission ratiometric fluorescence probe. Anal. Chem. 2015, 87,

423

8904-8909.

424

(4) Zhang, K., Zhou, H., Mei, Q., Wang, S., Guan, G., Liu, R., Zhang, J., Zhang, Z., Instant visual

425

detection of trinitrotoluene particulates on various surfaces by ratiometric fluorescence of

426

dual-emission quantum dots hybrid. J. Am. Chem. Soc. 2011, 133, 8424-8427.

427

(5) Zou, C.C., Foda, M.F., Tan, X.C., Shao, K., Wu, L., Lu, Z.C., Bahlol, H.S., Han, H.Y.,

428

Carbon-dot and quantum-dot-coated dual-emission core–satellite silica nanoparticles for

429

ratiometric intracellular Cu2+ imaging. Anal. Chem. 2016, 88, 7395-7403.

430

(6) Yan, Y.H., Sun, J., Zhang, K., Zhu, H.J., Yu, H., Sun, M.T., Huang, D.J., Wang, S.H.,

431

Visualizing gaseous nitrogen dioxide by ratiometric fluorescence of varbon nanodots−quantum

432

dots hybrid. Anal. Chem. 2015, 87, 2087-2093.

433

(7) Lan, M.H., Di, Y.F., Zhu, X.Y., Ng, T.W., Xia, J., Liu, W.M., Meng, X.M., Wang, P.F., Lee,

434

C.S., Zhang, W.J., A carbon dot-based fluorescence turn-on sensor for hydrogen peroxide with a

435

photo-induced electron transfer mechanism. Chem. Commun. 2015, 51, 15574-15577.

436

(8) Shangguan, J.F., He, D.G., He, X.X., Wang, K.M., Xu, F.Z., Liu, J.Q., Tang, J.L., Yang X.,

437

Huang, J., Label-free carbon dots based ratiometric fluorescence pH nanoprobes for intracellular

438

pH sensing. Anal. Chem. 2016, 88, 7837-7843.

439

(9) Qu, S.N., Chen, H., Zheng, X.M., Cao J.S., Liu, X.Y., Ratiometric fluorescent nanosensor

440

based on water soluble carbon nanodots with multiple sensing capacities. Nanoscale 2013, 5,

441

5514-5518.

442

(10) Qu, S.N., Wang, X.Y., Lu, Q.P., Liu, X.Y., Wang, L.J., A Biocompatible fluorescent ink based

443

on water-soluble luminescent carbon nanodots. Angew. Chem. Int. Ed. 2012, 51, 12215-12218.

444

(11) Yan, X., Li, H.X., Han, X.S., Su, X.G., A ratiometric fluorescent quantum dots based

445

biosensor for organo-phosphorus pesticides detection by inner-filter effect. Biosens. Bioelectron.

446

2015, 74, 277-283.

447

(12) Liu, H.J., Li, M., Xia, Y.N., Ren, X.Q., A turn-on fluorescent sensor for selective and

448

sensitive detection of alkaline phosphatase activity with gold nanoclusters based on inner filter

449

effect. ACS Appl. Mater. Interfaces 2017, 9, 120-126. 20


Page 20 of 32

Page 21 of 32


450

(13) Liu, J.J., Chen, Y.L., Wang, W.F., Feng, J., Liang, M.J., Ma, S.D., Chen, X.G., “Switch-on”

451

fluorescent sensing of ascorbic acid in food samples based on carbon quantum dots−MnO2 probe.

452

J. Agric. Food Chem. 2016, 64, 371-380.

453

(14) Liu, Y., Ouyang, Q., Li, H.H., Zhang, Z.Z., Chen, Q.S., Development of an inner filter

454

effects-based upconversion nanoparticles–curcumin nanosystem for the sensitive sensing of

455

fluoride ion. ACS Appl. Mater. Interfaces 2017, 9, 18314-18321.

456

(15) Chang, H., Ho, J., Gold nanocluster-assisted fluorescent detection for hydrogen peroxide and

457

cholesterol based on the inner filter effect of gold nanoparticles. Anal. Chem. 2015, 87,

458

10362-10367.

459

(16) Li, J.W., Li, X.M., Shi, X.J., He, X.W., Wei, W., Ma, N., Chen, H., Highly sensitive detection

460

of caspase-3 activities via a nonconjugated gold nanoparticle–quantum dot pair mediated by an

461

inner-filter effect. ACS Appl. Mater. Interfaces 2013, 5, 9798-9802.

462

(17) Zhang, L., Han, Y., Zhu, J., Zhai, Y., Dong, S., Highly sensitive detection of caspase-3

463

activities via a nonconjugated gold nanoparticle–quantum dot pair mediated by an inner-filter

464

effect. Anal. Chem. 2015, 87, 2033-2036.

465

(18) Jiang, Y., Zhao, H., Zhu, N.N., Lin, Y.Q., Yu, P., Mao, L.Q., A simple assay for direct

466

colorimetric visualization of trinitrotoluene at picomolar levels using gold nanoparticles. Angew.

467

Chem. Int. Ed. 2008, 47, 8601-8604.

468

(19) Xu, S.H., Ouyang, W.J., Xie, P.S., Lin, Y., Qiu, B., Lin, Z.Y., Chen, G.N., Guo, L.H., Highly

469

uniform gold nanobipyramids for ultrasensitive colorimetric detection of influenza virus. Anal.

470

Chem. 2017, 89, 1617-1623.

471

(20) Cao, G.M., Xu, F.J., Wang, S.L., Xu, K.L., Hou, X.D., Wu, P., Gold nanoparticle-based

472

colorimetric assay for selenium detection via hydride generation. Anal. Chem. 2017, 89,

473

4695-4700.

474

(21) Zhang, Z.Y., Chen, Z.P., Qu, C.L., Chen, L.X., Highly sensitive visual detection of copper

475

ions based on the shape-dependent LSPR spectroscopy of gold nanorods. Langmuir 2014, 30,

476

3625-3630.

477

(22) Lu, L.L., Xia, Y.S., Enzymatic reaction modulated gold nanorod end-to-end self-assembly for

478

ultrahigh sensitively colorimetric sensing of cholinesterase and organophosphate pesticides in

479

human blood. Anal. Chem. 2015, 87, 8584-8591. 21



480

(23) Miao, X., Yan, X.L., Qu, D., Li, D.B., Tao, F.F., Sun, Z.C., Sulfur, R.E., Red emissive sulfur,

481

nitrogen codoped carbon dots and their application in ion detection and theraonostics. ACS Appl.

482

Mater. Interfaces 2017, 9, 18549-18556.

483

(24) Gu, W., Pei, X., Cheng, Y., Zhang, C., Zhang, J., Yan, Y., Ding, C., Xian, Y., Black

484

phosphorus quantum dots as the ratiometric fluorescence probe for trace mercury ion detection

485

based on inner filter effect. ACS Sens. 2017, 2, 576-582.

486

(25) Xiao, S., Zhao, X., Chu, Z., Xu, H., Liu, G., Huang, C., Zhang, L., New off–on sensor for

487

captopril sensing based on photoluminescent MoOx quantum dots. ACS Omega. 2017, 2,

488

1666-1671.

489

(26) Han, L., Xia, J.M., Hai, X., Shu, Y., Chen, X.W., Wang, Protein-stabilized gadolinium

490

oxide-gold nanoclusters hybrid for multimodal imaging and drug delivery. ACS Appl. Mater.

491

Interfaces 2017, 9, 6941-6949.

492

(27) Ai, K., Liu, Y., Lu, L., Hydrogen-bonding recognition-induced color change of gold

493

nanoparticles for visual detection of melamine in raw milk and infant formula. J. Am. Chem. Soc.

494

2009, 131, 9496-9497.

495

(28) Li, G.L., Fu, H.L., Chen, X.J., Gong, P.W., Chen, G., Xia, L., Wang, H., You, J.M., Wu,

496

Y.N., Facile and sensitive fluorescence sensing of alkaline phosphatase activity with

497

photoluminescent carbon dots based on inner filter effect. Anal. Chem. 2016, 88, 2720-2726.

498

(29) Deng, Z.T., Schulz, O., Lin, S., Ding, B.Q., Liu, X.W., Wei, X.X., Ros, R., Yan, H., Liu, Y.,

499

The aqueous synthesis of zinc blende CdTe/CdS magic-core/thick-shell tetrahedral-shaped

500

nanocrystals with emission tunable to near-infrared. J. Am. Chem. Soc. 2010,132, 5592-5593.

501

(30) Xu, S.F., Lu, H.Z., Li, J.H., Song, X.L., Wang, A.X., Chen, L.X., Han, S.B., Dummy

502

molecularly imprinted polymers capped CdTe quantum dots for fluorescent sensing of

503

2,4,6-trinitrotoluene. ACS Appl. Mater. Interfaces 2013, 5, 8146-8154.

504

(31) Xu, S.F., Lu, H.Z., Ratiometric fluorescence and mesoporous structure dual signal

505

amplification for sensitive and selective detection of TNT based on MIPs@QDs fluorescence

506

sensor. Chem. Commun. 2015, 51, 3200-3203.

507

(32) Zhang, R., Li, N., Sun, J.Y, Gao, F., Colorimetric and phosphorimetric dual-signaling strategy

508

mediated by inner filter effect for highly sensitive assay of organophosphorus pesticides. J. Agric.

509

Food Chem. 2015, 63, 8947-8954. 22


Page 22 of 32

Page 23 of 32


510

(33) Xia, Y., Song, L., Zhu, C., Turn-on and near-infrared fluorescent sensing for

511

2,4,6-trinitrotoluene based on hybrid (gold nanorod)-(quantum dots) assembly. Anal. Chem. 2011,

512

83, 1401-1407.

513

(34) Xiao, S., Zhao, X., Hu, P., Chu, Z., Huang, C., Zhang, L., Highly photoluminescent

514

molybdenum oxide quantum dots: one-pot synthesis and application in 2,4,6-trinitrotoluene

515

determination. ACS Appl. Mater. Interfaces 2016, 8, 8184-8191.

516

(35) Liu, J., Yang, S., Li, F., Dong, L., Liu, J., Wang, X., Pu, Q., Highly fluorescent polymeric

517

nanoparticles based on melamine for facile detection of TNT in soil. J. Mater. Chem. A 2015, 3,

518

10069-10076.

519

(36) Yang, X., Wang, J., Su, D., Xia, Q., Chai, F., Wang, C., Qu, F., Fluorescent detection of TNT

520

and 4-nitrophenol by BSA Au nanoclusters. Dalton Trans. 2014, 43, 10057-10063.

521 522 523 524

Figure captions

525

Scheme 1 Illustration of the ratiometric fluorescent detection of TNT through the IFE

526

effect of AuNPs on RFNs.

527

Figure 1 (A) The fluorescence spectra of CDs (green line), QDs (red line) and RFNs

528

(yellow line). The inset was the photos of CDs (a), QDs (b) and RFNs (c) under UV

529

365 nm light. (B) The TEM images of RFNs.

530

Figure2 (A) UV/Vis spectra of AuNPs without (red line) or with (blue line) the

531

addition of 5 nM TNT. The insets were the photo of AuNPs under sunlight and TEM

532

images of AuNPs without (a) or with (b) the addition of 1 µM TNT. (B) The IFE

533

mechanism between AuNPs and fluorescence nanoparticles. Line 1 fluorescence

534

spectrum of CDs; Line 2 UV absorption spectrum of dispersion AuNPs; Line 3 23



535

fluorescence spectrum of CDs+ dispersion AuNPs; Line 4 fluorescence spectrum of

536

QDs; Line 2 UV absorption spectrum of aggregated AuNPs; Line 3 fluorescence

537

spectrum of QDs+ aggregated AuNPs.

538

Figure 3 (A) The absorption spectra of AuNPs upon the addition of different

539

concentration of TNT from 10 nM to 1200 nM. Inset: the linear plot of A700/A520

540

versus the concentration of TNT, and the photo of Au NPs after addition of different

541

concentration of TNT. (B) The fluorescence spectra of RFNs upon the addition of

542

different concentration of TNT from 0.1nM to 300 nM. Inset: the linear plot of

543

F520/F700 versus the concentration of TNT, and the photo of RFN after addition of

544

different concentration of TNT.

545

Figure 4 (A) The selectivity of the ratiometric sensor to various interference. The

546

concentration of TNT was 100 nM, DNT and 4-NP was 500 nM, Cu2+, Hg2+, Ag+,

547

Pb2+ was 1000 nM, and Na+, K+ 1mM, respectively. The inset photo (a) was the photo

548

of AuNPs after addition of different interference under sunlight. And the inset photo

549

(b) was the photo of AuNPs+RFN+ interference under UV 365nm light. From left to

550

right was blank, Na+, K+, Hg2+, Cu2+, Ag+, Pb2+, 4-NP, DNT and TNT, respectively. (B)

551

The interference experiment for TNT detection. Red line was AuNPs, black line was

552

AuNPs + interferences (100 nM DNT+100 nM 4-NP+100 nM Cu2++100 nM Ag++

553

100 nM Hg2+ +100nM Pb2+); the blue line was AuNPs +100 nM TNT; the green line

554

was AuNPs+interferences+100 nM TNT. The inset photo (a) was the photo of AuNPs

555

under sunlight and the photo of AuNPs+RFNs under UV 365nm light (b). From left to

556

right was blank, blank + interferences, blank + TNT, blank+ interferences+ TNT. 24


Page 24 of 32

Page 25 of 32


557 558 559 560 561 562 563 564 565 566 567 568 569

Scheme 1

570

25



571 572 573 574 575 576 577 578 579 580 581 582

Figure 1

26


Page 26 of 32

Page 27 of 32


583

584 585 586 587

Figure 2

588 27



589

590 591 592 593 594 595

Figure 3 28


Page 28 of 32

Page 29 of 32


596

597 598 599 600 601 602

Figure 4 29



603

604 605 606 607 608 609

Table 1 Comparisons of analytical performances with other reported

610

fluorescence methods for TNT sensing. 30


Page 30 of 32

Page 31 of 32


611

Probe

Mechanism

Linear range

LOD

Ref.

N-rich CDs

charge transfer

0.01-1.5 uM

1 nM

32

AuNR-QDs

FRET

1.1-66 nM

0.01nM

33

MoOx QDs

IFE

0.5-240µM

0.12uM

34

Schiff base nanoparticles

Energy transfer

6-1100 nM

1.8 nM

35

BAS AuNCs

Electron transfer

10 nM-5µM

10 nM

36

AuNPs-RFNs

IFE

0.1-270 nM

0.029 nM

This work

612 613

Table 2 Spiked recoveries and relative standard deviations for detection of TNT in

614

spiked samples by fluorescence and UV detection system.

Sample

River water sample

Soil sample

RFN FL detection

UV absorbance detection

Recovery, mean±RSD (%)

Recovery, mean±RSD (%)

0.1

97.1±4.0

-

10

98.5±3.8

95.2±3.5

100

103.2±3.4

97.6±4.2

0.1

95.2±4.3

-

10

99.2±3.9

95.6±3.5

100

96.4±3.7

101.2±4.5

Added (nM)

615

31



Graphical abstract

Ratiometric fluorescent assay for sensing TNT was proposed based on inner filter effect between gold nanoparticles and ratiometric fluorescent nanoparticles which hybridizing green emissive carbon dots and red emissive quantum dots.


Page 32 of 32

Highly Sensitive Ratiometric Fluorescent Sensor for Trinitrotoluene

Recommend Documents