Accepted Manuscript

Colorimetric detection of melamine in pretreated milk using silver nanoparticlesfunctionalized with sulfanilic acid

Juan Song, Fangying Wu, Yiqun Wan, Lihua Ma

PII: S0956-7135(14)00498-8

DOI: 10.1016/j.foodcont.2014.08.049

Reference: JFCO 4049

To appear in: Food Control

Received Date: 13 May 2014

Revised Date: 12 August 2014

Accepted Date: 27 August 2014

Please cite this article as: Song J., Wu F., Wan Y. & Ma L., Colorimetric detection of melamine inpretreated milk using silver nanoparticles functionalized with sulfanilic acid, Food Control (2014), doi:10.1016/j.foodcont.2014.08.049.

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

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

Colorimetric detection of melamine in pretreated milk using silver 1

nanoparticles functionalized with sulfanilic acid 2

Juan Song, Fangying Wu∗, Yiqun Wan, Lihua Ma 3

Department of Chemistry, Nanchang University, Nanchang 330031, China 4

5

Abstract Multiple methods have been published to detect melamine, only a few offer sensitivities 6

below 50 nM with complicated procedures and sophisticated equipments. We demonstrate here a 7

simple, rapid and lower-cost assay with high sensitivity for the melamine detection in milk 8

samples using sulfanilic acid-modified silver nanoparticles (SAA-AgNPs). Due to the special 9

chemical structures, SAA shows similar response to its analogues, which reveals that the 10

selectivity and sensitivity of SAA itself is poor. However, the formation of SAA-AgNPs 11

dramatically improves the selectivity of SAA and only melamine reacts with SAA-AgNPs. The 12

possible mechanism is discussed. The interaction between exocyclic amine of melamine and SAA 13

induces rapid aggregation of SAA-AgNPs accompanied by a naked-eye visible color change, 14

resulting in precise quantification of melamine that can be monitored by a simple UV-visible 15

spectrometer. Metal ions, amino acids and sugars that are common in milk have negligible 16

interference influences. The extinction ratio (A620nm/A390nm) is correlated with the melamine 17

concentration over the range of 0.1-3.1 µM. The detection limit is 10.6 nM, which is much lower 18

than the safety limits (8 µM for infant formula in China, 20 µM in both the USA and EU and 1.2 19

µM in the CAC review for melamine in liquid infant formula). The method is applied successfully 20

to determine melamine in pretreated milk products, indicating the potential practical use for the 21

products suspected of melamine exposure. 22

Key words: Colorimetric assay; Melamine; Silver nanoparticles; Sulfanilic acid; Milk samples. 23

24

1. Introduction 25

26

Melamine (1,3,5-triazine-2,4,6-triazine) is a trimer of cyanamide that is commonly found in 27

∗ Corresponding author: Fangying Wu, Tel: + 86 79183969882; Fax: + 86 79183969514; E-mail address: fywu@ncu.edu.cn

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

flame-resistant products, textile industries and the production of pesticides. In addition, although it is 28

not registered as a fertilizer in the U.S., melamine has been used as a fertilizer in some parts of the 29

world. In recent years, melamine has been intentionally and illegally adulterated in the milk, infant 30

formula and pet foods due to its low cost and high nitrogen content (66%, by mass) that can contribute 31

to the protein content measured by the conventional standard Kjeldajl and Dumas tests (Serdiuk, 32

Skryshevsky, Phaner-Goutorbe, & Souteyrand, 2010). Addition of 1% melamine in food causes protein 33

content to artificially boost more than 4%. The metabolism of melamine in vivo is inert and it has low 34

oral acute toxicity. However, melamine can be hydrolyzed to cyanuric acid which in turn associates 35

with melamine to form reticulations, resulting in the formation of kidney stones depending upon urine 36

pH that might cause organs failure and even death in humans and animals (Manzoori, Amjadi, & 37

Hassanzadeh, 2011). In 2007, US scientists confirmed that pet food adulteration with melamine was the 38

cause of illness and death of many cats and dogs (Tyan, Yang, Jong, Wang, & Shiea, 2009). In 2008, 39

consumption of melamine-contaminated infant formula and related dairy products in China likely 40

caused more than 51,900 infants and young children to suffer from urinary problems, including six 41

child deaths (Xu, et al., 2009). To assure food safety and protect public health, many countries have 42

restricted maximum residue limits (MRL) for melamine in various products. The European Union (EU) 43

set melamine MRL in dairy products and high-protein foods at 20 µM, whereas the US Food and Drug 44

Administration (FDA) set it as 2.0µM in milk and dairy products but stressed that infant formula sold 45

to US consumers must be completely free of melamine (Filazi, Sireli, Ekici, Can, & Karagoz, 2012). 46

The Ministry of Health of China published new dairy safety standards and emphasized that food should 47

not be contaminated with melamine (Guo, et al., 2011). 48

The most commonly used approaches that have been published are chromatography-based 49

methods including high performance liquid chromatography (HPLC) (Zhang, et al., 2014), 50

gas-chromatography mass spectrometry (GC–MS) (Li, Qi, & Shi, 2009), and LC–MS/MS (He, et al., 51

2014). Others also have tried immunoassay (Lei, et al., 2010), electrochemiluminescence (ECL) (Guo, 52

et al., 2011), fourier transform infrared spectroscopy (FTIR) (Jawaid, Talpur, Sherazi, Nizamani, & 53

Khaskheli, 2013), surface-enhanced Raman spectroscopic (SERS) (Zhao, et al., 2013) and molecular 54

imprinting (MIP) (Jin, Yu, Yang, & Ma, 2011) to detect melamine in foods. Some of the methods 55

possess high sensitivity and accuracy. Nevertheless, most of these technologies are time consuming, 56

labor-intensive, instrumental expensive and well-trained scientists are needed to perform sample 57

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3

pretreatment and to operate the instruments. 58

Owing to their unique chemical and physical properties, metal nanoparticles (NPs) have been 59

favorably selected to design novel sensors for detecting biologically or environmentally important 60

analytes by various optical methods such as absorption (Sun, Zhang, & Li, 2012), fluorescence (Su, 61

Chen, Sun, & Ai, 2012), resonant light scattering (RLS) (Navarro & Werts, 2013), surface-enhanced 62

raman scattering (SERS) (Zhou, et al., 2010) and so on. Due to the extremely high extinction 63

coefficient and strongly distance-dependent optical property, silver nanoparticles (AgNPs) were 64

utilized as an ideal color reporter for colorimetric sensors. Up to now, AgNPs-based colorimetric 65

methods have been established to detect proteins, ions and other small molecules (Li, Qiang, Vuki, Xu, 66

& Chen, 2011; Miao, et al., 2013; Yang, Gao, Zhang, Shen, & Wu, 2014). Melamine detection using 67

AgNPs without label has been published before and achieved only 2.32 µM LOD (Ping, et al., 2012). 68

In order to facilitate aggregation and increase the water solubility, the AgNPs are always covalently 69

modified with functional molecules (Lee, Lytton-Jean, Hurst, & Mirkin, 2007). Our team also reported 70

chromotropic acid-modified AgNPs for the visual detection of melamine with detection limit of 36 nM 71

(Song, Wu, Wan, & Ma, 2014). 72

Sulfanilic acid (SAA) is applied in quantitative analysis of nitrate and nitrite ions (Narayana & Sunil, 73

2009). It is commercially available, inexpensive, environmentally benign, stable and incompatible with 74

strong oxidizing agents. It shows similar UV-visible spectral responses to melamine and its structural 75

analogues. However, we found out that after the AgNPs was modified with SAA, the SAA-AgNPs 76

dramatically changed the selectivity and sensitivity of SAA and it only reacted with melamine among 77

the analogues. The affinity of melamine towards the functional group of SAA leads to a notable 78

aggregation of SAA-AgNPs and a clear and rapid color change from bright yellow to blue-green was 79

observed by the naked eye. Based on the above discovery, we established a colorimetric assay for 80

melamine (Scheme 1) with detection limit as low as 10.6 nM in aqueous solution and successfully 81

applied it to the milk products with satisfactory recoveries. 82

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

4

83

Scheme 1 Schematic illustration of possible mechanism for the colorimetric response of SAA-AgNPs 84

to melamine. 85

86

2. Experimental 87

88

2.1. Chemicals and Materials 89

90

Sulfanilic acid was purchased from the third Shanghai reagent factory. (Shanghai, China, 91

http://6379.cn.Toocle.com). Silver nitrate (AgNO3), sodium borohydride (NaBH4), trisodium citrate, 92

melamine and the other metallic ions (NaCl, MgCl2, ZnCl2, FeCl3, CaCl2, KNO3, Na2CO3, Na4P2O7) 93

were purchased from Shanghai Qingxi Technology Co. Ltd. (Shanghai, China, 94

www.ce-r.cn/sites/qingxi/). All the carbohydrate including D-fructose, D-glucose and sucrose were 95

purchased from Shanghai Lanji Technology Co. Ltd. (Shanghai, China, http://lj80414.bioon.com.cn/). 96

Aniline compounds and phenol compounds including o-phenylenediamine (o-PDA), 97

m-phenylenediamine (m-PDA), p-phenylenediamine (p-PDA), phloroglucinol, cyanuric acid, 98

trimethy-1,3,5-triazine, and amino acids including lysine (Lys), tryptophan (Try), methionine (Met), 99

leucine (Leu), isoleucine (Ile), phenylalanine (Phe) and valine (Val) were purchased from Shanghai 100

Jingchun Technology Co. Ltd. (Shanghai, China, http://www.aladdin-reagent.com/). 101

102

2.2 Instruments 103

104

UV-vis absorption spectra were recorded using UV-2550 spectrophotometer (Shimadzu, Kyoto, 105

Japan, http://www.shi-madzu.com) with a 1.0 cm quartz cell at room temperature. For the Transmission 106

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

5

electron microscopy (TEM) analysis, the solution of SAA-AgNPs was dropped onto the carbon-coated 107

copper grid and the microscope used was a JEM-2010 TEM (JEOL Ltd. Japan, http://www.jeol.cn). 108

The FTIR spectra were collected in Nicolet 380 FTIR Spectrometer (Nicolet, USA, 109

http://www.thermonicolet.com) at room temperature. For FTIR characterization, the solution of 110

SAA-AgNPs was centrifuged. Upon removal of the supernatant the solid deposit was rinsed with DI 111

water and then placed in a vacuum dryer (DZF-6030A) at 40°C and full pressure (133 pa) for 12 hours. 112

The dried AgNPs was mixed with KBr, and pelleted by macro KBr die kit. Pellets were loaded into a 113

Nicolet 380 FTIR spectrometer and the spectra were taken in the wavenumber range from 400 to 4000 114

cm-1 with a resolution of 4cm-1. For comparison, FTIR spectrum for pure SAA was also recorded. 115

Atomic emission spectra were recorded by ICP-AES OPTIMA 5300DV (PE, USA, 116

http://www.perkinelmer.com) 117

118

2.3. Preparation of SAA-modified AgNPs 119

120

The bare AgNPs were prepared by reducing AgNO3 according to the method reported previously 121

(Li, et al., 2010). All pieces of the experimental glassware used were cleaned in a bath of freshly 122

prepared aqua regia solution (HCl: HNO3, 3:1) and then rinsed thoroughly with milli-Q-purified 123

distilled water prior to use. Briefly, 2.0 mL of 0.01 M AgNO3 and 4.0 mL of 0.01 M sulfanilic acid 124

were added into 94 mL double distilled water under stirring. Then 8.8 mg of NaBH4 was quickly added 125

into the mixture solution under vigorous stirring. The resulting yellow silver colloidal solution was 126

stirred for 2 h in the dark at room temperature. The SAA-AgNPs was stored at 4°C in dark undergoing 127

no change within 7 days (shown in Fig. S1) and used directly in the following experiments. The 128

concentration of testing AgNPs solution was estimated to be 14 nM according to extinction coefficient 129

on particle diameter (ε=Adγ, for AgNPs which diameter is less or equal to 38 nm, A=2.3×105 M-1cm-1, 130

γ=3.48) (Navarro, & Werts, 2013). In our experiment, the average of AgNPs size is 6.7 nm and the 131

absorbance of testing solution is 2.4. 132

133

2.4. Milk Samples pretreatment 134

135

The milk samples were pretreated according to the general procedure (Ma, Niu, Zhang, & Cai, 136

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

6

2011). In short, proteins were removed by mixing with 1.5 mL 10% trichloroacetic acid, 5.0 mL 137

acetonitrile and then 375 µL 1.0 M Na2CO3 for 2.0 g milk. The mixture solution was transferred to 138

centrifugal tube to be sonicated for 10 min and then centrifuged at 12,000 rpm min-1 for 15 min. The 139

supernatant was filtered through a 0.22 µm membrane to remove lipids. The pH of filtrate was adjusted 140

to 6.8, and the filtrate was filtered again after centrifugation. The filtered liquid was diluted with 141

water to 10 mL for further analysis. The concentration of calcium ion before and after the treatment 142

were measured by the ICP-AES to be 1500 mg/L and 29 mg/L, respectively, indicating that this 143

procedure is suitable for the purpose of effective calcium ions removal. 144

145

2.5. General procedure of colorimetric detection of melamine 146

147

The colorimetric detection of melamine was performed at room temperature. A 1.0×10-4 M 148

aqueous solution of melamine was prepared. In a typical experiment, 50 µL of melamine solution with 149

various concentrations was added to the 3 mL SAA-AgNPs solution (14 nM), and the mixture was 150

incubated at room temperature for 5 min. The photographs of resulting solutions were taken and the 151

UV–visible absorption spectra of the mixtures were recorded immediately. The absorbance intensity 152

ratio of A620nm/A390nm acted as the quantitative index for melamine. The spiked-recovery detection of 153

melamine in pretreated samples was manipulated with the same procedure. 154

155

3. Results and discussion 156

157

3.1. Characterization of SAA-AgNPs 158

159

To elucidate the stabilization mechanisms and to gain further insight into the interactions between 160

various functional groups of SAA and AgNPs, FTIR measurements were carried out. Fig.S2 compared 161

the characteristic stretching frequencies for SAA (b) and SAA-AgNPs (a). It was noteworthy that the 162

S=O stretching vibration band at 1235-1109 cm-1 of sulfonic acid that is obvious for SAA disappears 163

for SAA-AgNPs. Furthermore, the bands at 1640-1560 cm–1 and 900–650 cm–1 those are the 164

characteristic peaks of primary aromatic amines N–H bending and wagging vibrations, respectively are 165

obviously shown in the spectra of both SAA-AgNPs (Fig.S2a) and SAA (Fig.S2b). The above 166

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

7

observation based on S=O and NH groups implied that SAA was bound to the surface of AgNPs via the 167

sulfonic acid and suggested that SAA-AgNPs were successfully synthesized. TEM was used to 168

characterize the size and shape of AgNPs. The TEM image of SAA-AgNPs in the absence of melamine 169

was shown in Fig.1. It was clear that the SAA-AgNPs were dispersed uniformly with spherical shape. 170

The average size was 6.7 nm with relative standard deviation as 14% (shown in Fig.S3). 171

172

3. 2 Recognition of melamine by AgNPs and the selectivity of the assay 173

In the aqueous media, the affinity of melamine towards the functional groups of SAA lead to a 174

notable aggregation of SAA-AgNPs and a color change (presented in Fig.1). The free SAA-AgNPs 175

with bright yellow color showed a typical peak around 390 nm (photographic image (a) and curve (a)) 176

resulting from the surface plasmon resonance with extremely high extinction coefficient. In the 177

presence of melamine, the SAA-AgNPs aggregated rapidly (Fig.1b), and the intensity of peak at 390 178

nm decreased and a broad peak at 620 nm appeared, accompanied by a visible color change from bright 179

yellow to blue-green (as shown in photographic image (b) and curve (b) of Fig.1). 180

181

Fig.1 UV-vis spectra (left) (insert image is colorimetric response) and TEM images (right) of 182

SAA-AgNPs in the absence (a) and presence (b) of 0.9 µM melamine. 183

184

In order to examine the specificity and selectivity of the interaction between melamine and 185

SAA-AgNPs, we performed two types of parallel experiments including the effect of the potentially 186

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

8

interfering cation, anion, some amino acids, sugars (shown in Fig. 2) and the responses of SAA-AgNPs 187

to other chemical analytes (arginine, histidine, guanidine hydrochloride, cyanuric acid, 188

trimethy-1,3,5-triazine, phloroglucinol, m-dihydroxybenzene, o-phenylenediamine(o-PDA), 189

m-phenylenediamine(m-PDA), p-phenylenediamine(p-PDA) shown in Fig.3A), the structures of which 190

were illustrated in Fig.3B. As shown in Fig.2, the presence of following amounts of ions and 191

compounds resulted in less than ±10 % error: 50 times the concentration of Lys, Try, Met, Leu, Ile, Phe, 192

Val, glucose, fructose, Sucrose, 300 times the concentration of Ca2+, 500 times the concentration of 193

Mg2+, Zn2+, Fe3+and 1000 times the concentration of Na+ and NO3-, pyrophosphate, citrate, CO3

2-, 194

EDTA. And the SAA-AgNPs were closely non-responsive to melamine analogues as shown in Fig.3A. 195

But His and Arg interfered the assay to some extent. Because these two amino acids are not common in 196

liquid milk and formula, together with the different color change, the influence can be ignored in the 197

practical use. 198

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 210.0

0.2

0.4

0.6

0.8

1.0

Substance

A62

0nm/A

390n

m

199

Fig.2 The intensity ratio (A620nm/A390nm) of SAA-AgNPs upon addition of 0.9µM melamine with 200

the coexistence of interference (1-21: control, 50 times Lys, Try, Met, Leu, Ile, Phe, Val, 1000 times 201

NO3-, pyrophosphate, citrate, CO3

2-, EDTA, 300 times of Ca2+, 500 times of Mg2+, Zn2+, Fe3+, 1000 202

times of Na+, glucose, fructose, sucrose, respectively 203

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

9

204

Fig.3 (A) The intensity ratio (A620nm/A390nm) and photographs of SAA-AgNPs in the presence of 0.9 205

µM melamine and other analytes. (B) Structures of melamine analogues and amino acids His and Arg. 206

207

3.3. Optimization of the assay for melamine 208

209

3.3.1. Effect of the SAA concentration 210

211

Since the concentration of SAA could have effect on SAA-AgNPs detection capability, the 212

stoichiometric ratio between SAA and AgNO3 were explored. In our work, SAA-AgNPs were 213

synthesized based on four different molar ratio (nAgNO3 : nSAA), intensity ratios (A620nm/A390nm) of which 214

were compared (insert in Fig.4). The result indicated that SAA-AgNPs prepared at lower molar ratio 215

was more sensitive to melamine. However, the concentration of SAA can’t be too high because the 216

excess of SAA may lead to SAA-AgNPs aggregation to some extent as shown in the absorption spectra 217

of Fig.4. When the molar ratio was 1:4, the absorbance at 390 nm was weak and became insensitive to 218

melamine. Therefore, molar ratio 1: 2 that gave the best sensitivity to melamine was selected as the 219

optimized condition. 220

221

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

10

Fig.4 Absorption spectra of SAA-AgNPs prepared with various molar ratio (nAgNO3 : nSAA). Insert 222

is the absorption intensity ratio (A620nm/A390nm) of these different types of SAA-AgNPs as a 223

function of melamine concentrations. 224

225

3.3.2. Effect of pH and reaction time 226

227

Influence of the media pH (4.0-11.5) and reaction time on the assay were also investigated. pH 9.5 228

was found to be the optimal condition for the further experimental. It was noticed that the reaction took 229

place in less than several seconds and underwent slight increase in 5 min. To obtain the maximum 230

response, 5 min was selected as the incubation time for the following experiments. 231

232

3.4. Analytical applications 233

234

3.4.1. Colorimetric assay of melamine in water 235

236

Under the optimized detection conditions and at room temperature, the sensitivity of the sensor 237

was investigated. Several calibration samples with various concentrations of melamine were analyzed. 238

As shown in Fig. 5A, with increasing the melamine concentration, the absorbance of SAA-AgNPs at 239

390 nm decreased gradually and the absorbance at 620 nm increased. The variation could be visualized 240

by the naked eyes as shown in photographs in Fig. 5A. Two linear ranges were observed, resulting in 241

two calibration equations, one was y = 0.00669 [C] + 0.00644 when melamine concentration was in the 242

range of 0 to 0.9µM, the other was y = 0.02697 [C] – 0.2112 when melamine concentration was from 243

0.9 to 3.1 µM (Insets a and b of Fig. 5B). The limit of detection (LOD) is defined by the equation 244

LOD=3S0/K, where S0 is the standard deviation of blank measurements (n=10) and K is the slope of 245

calibration line. The LOD of the proposed method was 10.6 nM according to the lower concentration 246

range. 247

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

11

248

Fig.5 (A)Absorbance spectra of the SAA-AgNPs in the presence of melamine with various 249

concentrations at 0, 0.1, 0.3, 0.5, 0.7, 0.9, 1.1, 1.3, 1.5, 1.7, 1.9, 2.1, 2.3, 2.5, 2.7, 2.9 and 3.1 µM, 250

respectively. Inset is the selected photograph. (B) The plot of intensity ratio (A620nm/A390nm) of 251

SAA-AgNPs versus the concentration of melamine. Insets are the two linear fitting curves. 252

253

3.4.3. Detection of melamine in milk samples 254

255

To validate the reliability of the proposed colorimetric method in practical applications, the 256

detection of melamine in milk products was examined. A big challenge for detecting melamine in milk 257

products is how to decrease the potential interference from macromolecular compounds (such as 258

protein). In our study, milk products samples were pretreated by a sample extract procedure and these 259

samples were spiked with different concentrations of melamine. In the pretreatment process, proteins 260

were precipitated by trichloroacetic acid, and lipid micelles were removed by repeating filtrations. 261

In order to increase the tolerance of Ca in the assays, we use Na2CO3 to get rid of most calcium. 262

The Ca concentration was measured by ICP-AES before and after treatment. The detailed 263

procedures were shown in the milk samples pretreatment (seen part of 2.4). Even though there 264

may be trace calcium exist, the lower amount of calcium would not interfere with the response of 265

melamine to the SAA-AgNPs sensor. As shown in Fig.S4, the UV-vis spectra of SAA-AgNPs, the 266

intensity ratio plot (A620nm/A390nm), the calibration equations in the presence of melamine were similar 267

in different matrix such as liquid and solid milk, suggesting that this colorimetric assay can detect 268

melamine without being affected by the complex environments. The recoveries for liquid and solid 269

milk were from 96% to 109% and 97% to 108%, respectively (Table 1). Therefore, it turns out to be a 270

promising method for rapid screening of melamine contamination in milk products with high 271

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

12

reproducibility. 272

273

Table 1 Recovery test results of the assay in liquid and solid milk samples with spiked melamine 274

Sample Added (10-7M) Average (10-7M) RSD (% , n=3) Recovery (%)

Liquid milk 9.0 9.8 1.9 109

21.0 21.6 1.3 103

Solid milk 9.0 9.7 1.5 108

21.0 20.3 0.9 97

275

3.5. Possible Mechanism of the sensor 276

The UV-visible spectra, visible color changes, TEM images explained the aggregation of 277

SAA-AgNPs in the presence of melamine. Based on the FTIR, we concluded that SAA was modified 278

on the surface of AgNPs through the sulfonic acid and therefore it was the -NH2 group on the outer 279

surface of the SAA-AgNPs that was responsible for sensing melamine based on hydrogen bonding (Ma, 280

Niu, Zhang, & Cai, 2011). The results implied that the selectivity and the sensitivity of the 281

SAA-AgNPs were significantly enhanced as compared with the free SAA (Fig.3 and Fig.S5), which 282

were possibly attributed to both hydrogen bonding and the shape complementarity between 283

SAA-AgNPs and melamine (Liu et al., 2011). By anchoring on the surface of AgNPs, multiple SAA 284

may assemble to provide more hydrogen binding sites and leave more spaces that are favored by 285

melamine molecules. It has been reported that melamine has a specific structure which could form 286

ordered structures, it might be reasoned that the neighbor melamine coated AgNPs could be 287

cross-linked with melamine molecules (Guan, Yu, & Chi, 2013; Silly, et al., 2008) (as shown in 288

Scheme1), which in turn significantly improve the responses of SAA-AgNPs to melamine compared 289

with that of free SAA. 290

291

4. Conclusions 292

293

In summary, an ultrasensitive and selective detection of melamine using SAA capped AgNPs 294

based on the interaction between melamine and –NH2 group of SAA was proposed. The presence of 295

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

13

melamine induced the aggregation of SAA-AgNPs through cooperative hydrogen bondings, resulting 296

in a color change from yellow to blue-green that is easily observed with the naked eye or the UV-vis 297

spectrometer. At the optimized conditions, the SAA_AgNPs based detection sstem has an excellent 298

selectivity to melamine comparing with other analogues. This method is easy to be operated with lower 299

cost (0.5$ per sample). In particular, the proposed low-cost and rapid-reaction method is successfully 300

applied to determine melamine in milk products with LOD of 10.6nM, which is lower than most 301

existing methods without the aid of expensive instruments (as shown in Table S1), allowing the 302

potential use of this system for fast quality screening of milk products. 303

304

Acknowledgement 305

This work was financially supported by Natural Science Foundation of China (No.201365014), Jiangxi 306

Province Science and Technology University Ground Plan project (KJLD No.14007) and Jiangxi 307

Province Natural Science Foundation (JXNSF No. 20132BAB203011). 308

309

References 310

311

Filazi, A., Sireli, U. T., Ekici, H., Can, H. Y., & Karagoz, A. (2012). Determination of melamine in milk 312

and dairy products by high performance liquid chromatography. Journal of Dairy Science, 95, 313

602-608. 314

Guan, H., Yu, J., & Chi, D. (2013). Label-free colorimetric sensing of melamine based on 315

chitosan-stabilized gold nanoparticles probes. Food Control, 32, 35-41. 316

Guo, Z., Gai, P., Hao, T., Wang, S., Wei, D., & Gan, N. (2011). Determination of melamine in dairy 317

products by an electrochemiluminescent method combined with solid-phase extraction. 318

Talanta, 83, 1736-1741. 319

He, D., Zhang, X., Gao, B., Wang, L., Zhao, Q., Chen, H., Wang, H., & Zhao, C. (2014). Preparation of 320

magnetic molecularly imprinted polymer for the extraction of melamine from milk followed 321

by liquid chromatography-tandem mass spectrometry. Food Control, 36, 36-41. 322

Jawaid, S., Talpur, F. N., Sherazi, S. T. H., Nizamani, S. M., & Khaskheli, A. A. (2013). Rapid 323

detection of melamine adulteration in dairy milk by SB-ATR–Fourier transform infrared 324

spectroscopy. Food Chemistry, 141, 3066-3071. 325

Jin, G.P., Yu, B., Yang, S.Z., & Ma, H.H. (2011). Extremely sensitive electrode for melamine using a 326

kind of molecularly imprinted nano-porous film. Microchimica Acta, 174, 265-271. 327

Lee, J.S., Lytton-Jean, A. K. R., Hurst, S. J., & Mirkin, C. A. (2007). Silver 328

nanoparticle−oligonucleotide conjugates based on DNA with triple cyclic disulfide moieties. 329

Nano Letters, 7, 2112-2115. 330

Lei, H., Shen, Y., Song, L., Yang, J., Chevallier, O. P., Haughey, S. A., Wang, H., Sun, Y., & Elliott, C. 331

T. (2010). Hapten synthesis and antibody production for the development of a melamine 332

immunoassay. Analytica Chimica Acta, 665, 84-90. 333

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

14

Li, H., Li, F., Han, C., Cui, Z., Xie, G., & Zhang, A. (2010). Highly sensitive and selective tryptophan 334

colorimetric sensor based on 4,4-bipyridine-functionalized silver nanoparticles. Sensors and 335

Actuators B: Chemical, 145, 194-199. 336

Li, H., Qiang, W., Vuki, M., Xu, D., & Chen, H.-Y. (2011). Fluorescence Enhancement of Silver 337

Nanoparticle Hybrid Probes and Ultrasensitive Detection of IgE. Analytical Chemistry, 83, 338

8945-8952. 339

Li, J., Qi, H. Y., & Shi, Y. P. (2009). Determination of melamine residues in milk products by zirconia 340

hollow fiber sorptive microextraction and gas chromatography-mass spectrometry. Journal 341

Chromatogr A, 1216(29), 5467-5471. 342

Liu, Y., Liu, Y., Liang, Z., Li, X., Liu, S., & Yu, J. (2011). Enhanced Sensitivity and Selectivity of 343

Chemosensor for Malonate by Anchoring on Gold Nanoparticles. Chinese Journal Chemistry. 344

29, 531-538. 345

Ma, Y., Niu, H., Zhang, X., & Cai, Y. (2011). One-step synthesis of silver/dopamine nanoparticles and 346

visual detection of melamine in raw milk. Analyst, 136, 4192-4196. 347

Manzoori, J., Amjadi, M., & Hassanzadeh, J. (2011). Enhancement of the chemiluminescence of 348

permanganate-formaldehyde system by gold/silver nanoalloys and its application to trace 349

determination of melamine. Microchimica Acta, 175, 47-54. 350

Miao, P., Liu, T., Li, X., Ning, L., Yin, J., & Han, K. (2013). Highly sensitive, label-free colorimetric 351

assay of trypsin using silver nanoparticles. Biosensors and Bioelectronics, 49, 20-24. 352

Narayana, B., & Sunil, K. (2009). A Spectrophotometric Method for the Determination of Nitrite and 353

Nitrate. Eurasian Journal of Analytical Chemistry, 4, 204-214. 354

Navarro, J. R. G., & Werts, M. H. V. (2013). Resonant light scattering spectroscopy of gold, silver and 355

gold-silver alloy nanoparticles and optical detection in microfluidic channels. Analyst, 138, 356

583-592. 357

Ping, H., Zhang, M., Li, H., Li, S., Chen, Q.i, Sun, C., & Zhang, T. (2012). Visual detection of 358

melamine in raw milk by label-free silver nanoparticles. Food Control, 23, 191-197. 359

Serdiuk, T., Skryshevsky, V. A., Phaner-Goutorbe, M., & Souteyrand, E. (2010). Monitoring of 360

melamine contamination in fat watery milk by the photoluminescence analysis. Talanta, 82, 361

1543-1547. 362

Silly, F., Shaw, A. Q., Castell, M. R., Briggs, G. A. D., Mura, M., Martsinovich, N., & Kantorovich, L. 363

(2008). Melamine Structures on the Au(Ⅲ) Surface. The Journal of Physical Chemistry C, 364

112, 11476-11480. 365

Song, J., Wu, F., Wan, Y., & Ma, L.H. (2014). Visual test for melamine using silver nanoparticles 366

modified with chromotropic acid. Microchimica Acta, 181,1267–1274 367

Su, H., Chen, L., Sun, B., & Ai, S. (2012). Fluorescence detection of cyromazine using gallic 368

acid-reduced gold nanoparticles. Sensors and Actuators B: Chemical, 174, 458-464. 369

Sun, Y., Zhang, L., & Li, H. (2012). Chiral colorimetric recognition of amino acids based on silver 370

nanoparticle clusters. New Journal of Chemistry, 36, 1442-1444. 371

Tyan, Y.C., Yang, M.H., Jong, S.B., Wang, C.K., & Shiea, J. (2009). Melamine contamination. 372

Analytical and Bioanalytical Chemistry, 395, 729-735. 373

Xu, X.M., Ren, Y.P., Zhu, Y., Cai, Z.X., Han, J.L., Huang, B.F., & Zhu, Y. (2009). Direct determination 374

of melamine in dairy products by gas chromatography/mass spectrometry with coupled 375

column separation. Analytica Chimica Acta, 650, 39-43. 376

Yang, N., Gao, Y., Zhang, Y., Shen, Z., & Wu, A. (2014). A new rapid colorimetric detection method of 377

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

15

Al 3+ with high sensitivity and excellent selectivity based on a new mechanism of aggregation 378

of smaller etched silver nanoparticles. Talanta, 122, 272-277. 379

Zhang, S., Yu, Z., Hu, N., Sun, Y., Suo, Y., & You, J. (2014). Sensitive determination of melamine 380

leached from tableware by reversed phase high-performance liquid chromatography using 381

10-methyl-acridone-2-sulfonyl chloride as a pre-column fluorescent labeling reagent. Food 382

Control, 39, 25-29. 383

Zhao, Y., Luo, W., Kanda, P., Cheng, H., Chen, Y., Wang, S., & Huan, S. (2013). Silver deposited 384

polystyrene (PS) microspheres for surface-enhanced Raman spectroscopic-encoding and rapid 385

label-free detection of melamine in milk powder. Talanta, 113(15), 7-13. 386

Zhou, X., Xu, W., Wang, Y., Kuang, Q., Shi, Y., Zhong, L., & Zhang, Q. (2010). Fabrication of 387

Cluster/Shell Fe3O4/Au Nanoparticles and Application in Protein Detection via a SERS 388

Method. The Journal of Physical Chemistry C, 114, 19607-19613. 389

390

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

16

Captions 391

392

Scheme 1 Schematic illustration of possible mechanism for the colorimetric response of SAA-AgNPs 393

to melamine. 394

395

Fig.1 UV-vis spectra (left) (insert image is colorimetric response) and TEM images (right) of 396

SAA-AgNPs in the absence (a) and presence (b) of 0.9 µM melamine. 397

398

Fig.2 The intensity ratio (A620nm/A390nm) of SAA-AgNPs upon addition of 0.9µM melamine with 399

the coexistence of interference (1-21: control, 50 times Lys, Try, Met, Leu, Ile, Phe, Val, 1000 times 400

NO3-, pyrophosphate, citrate, CO3

2-, EDTA, 300 times of Ca2+, 500 times of Mg2+, Zn2+, Fe3+, 1000 401

times of Na+, glucose, fructose, sucrose, respectively. 402

403

Fig.3 (A) The intensity ratio (A620nm/A390nm) and photographs of SAA-AgNPs in the presence of 0.9 404

µM melamine and other analytes. (B) Structures of melamine analogues and amino acids His and Arg. 405

406

Fig.4 Absorption spectra of SAA-AgNPs prepared with various molar ratio (nAgNO3 : nSAA). Insert 407

is the absorption intensity ratio (A620nm/A390nm) of these different types of SAA-AgNPs as a 408

function of melamine concentrations. 409

410

Fig.5 (A)Absorbance spectra of the SAA-AgNPs in the presence of melamine with various 411

concentrations at 0, 0.1, 0.3, 0.5, 0.7, 0.9, 1.1, 1.3, 1.5, 1.7, 1.9, 2.1, 2.3, 2.5, 2.7, 2.9 and 3.1 µM, 412

respectively. Inset is the selected photographs. (B) The plot of intensity ratio (A620nm/A390nm) of 413

SAA-AgNPs versus the concentration of melamine. Insets are the two linear fitting curves. 414

415

416

Table 1 Recovery test results of the assay in liquid and solid milk samples with spiked melamine. 417

418

419

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Research Highlights

An-easy-operate colorimetric assay for melamine(MA) based on AgNPs was presented.

The color change can be observed via naked-eyes when only 1.0µM MA exists.

The assay was applied to detect MA in milk sample.

The detection limit as 10.6 nM for MA is much lower than the safety limits.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

Electronic Supporting Materials 1

Colorimetric detection of melamine in pretreated milk using silver 2

nanoparticles functionalized with sulfanilic acid 3

List of Contents page number 4

5

Fig. S1 The UV-vis spectra of SAA-AgNPs stored at 4°C in different time 2 6

Fig. S2 IR spectra of SAA-AgNPs (a) and pure SAA (b) 2 7

Fig. S3 Histograms of size distribution of SAA-AgNPs 2 8

Fig. S4 Relationship between the absorbance ratio (A620nm/A390nm) of SAA-AgNPs and the 9

concentration of melamine in water and milk samples. 3 10

Fig. S5 The UV-vis spectra of SAA in the presence of melamine and its analogues 3 11

Table S1 Comparison of different methods for melamine determination 4 12

Reference 4-5 13

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

14

15

Fig.S1 The UV-vis spectra of SAA-AgNPs stored at 4°C in different time 16

17

18

Fig.S2 IR spectra of SAA-AgNPs (a) and pure SAA (b) 19

20

Fig.S3 Histograms of size distribution of the sulfanilic acid-AgNPs 21

22

23

24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3

25

26

Fig.S4 Relationship between the absorbance ratio (A620nm/A390nm) of SAA-AgNPs and the 27

concentration of melamine with water (black), liquid milk (red) and solid milk (blue) samples. 28

29

30

Fig.S5 The UV-vis spectra of SAA (2.0×10-4 M) in the presence of melamine and its analogues 31

(18µM). 32

33

34

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

4

Table S1 Comparison of linear range and detection of limit of melamine obtained by different 35

methods. 36

Methods Detection limit

(mol L-1)

Linear range

(mol L-1)

Reference

HPLC 3.2×10-9 4.0×10-9-1.6×10-6 (Zhang, et al., 2014)

LC-MS 2.1×10-8 7.9×10-8 -8.0×10-6 (He, et al., 2014)

GC-MS 8.0×10-9 8.0×10-9-7.9×10-3 (Li, et al., 2009)

FTIR 2.0×10-8 4.0×10-8-7.9×10-8 (Jawaid, et al., 2013)

ECL 2.4×10-9 7.9×10-8 -7.9×10-6 （Guo, et al., 2011）

SERS 1.9×10-8 1.0×10-8-1.0×10-3 (Zhao, et al., 2013)

MIP-based sensors 1.3×10-9 1.0×10-8-9.0×10-7 (Jin, et al., 2011)

Citrate capped AgNPs 2.3×10-6 4.0×10-6-1.7×10-4 (Ping, et al., 2012)

p-nitroaniline modified AgNPs 7.9×10-7 5.0×10-8-1.0×10-4 (Han & Li, 2010)

Dopamine modified AgNPs 7.9×10-8 7.9×10-8-1.0×10-5 (Ma, et al., 2011)

Crown ether modified AuNPs 4.7×10-8 7.9×10-8 -4.0 ×10-6 (Kuang, et al., 2011)

Cysteamine modified AuNPs 7.9×10-6 7.9×10-6 -1.6×10-3 (Liang, et al., 2011)

Chitosan-stabilized AuNPs

CTA capped AgNPs

4.8×10-8

3.6×10-8

7.9×10-8-7.9×10-5

1.0×10−7–1.5×10−6

(Guan, et al., 2013)

(Song, et al., 2014)

SAA capped AgNPs 1.1×10-8 1.0×10-7-3.1×10-6 Our method

37

References 38

39

Guan, H., Yu, J., & Chi, D. (2013). Label-free colorimetric sensing of melamine based on 40

chitosan-stabilized gold nanoparticles probes. Food Control, 32(1), 35-41. 41

Guo, Z., Gai, P., Hao, T., Wang, S., Wei, D., & Gan, N. (2011). Determination of melamine in dairy 42

products by an electrochemiluminescent method combined with solid-phase extraction. 43

Talanta, 83(5), 1736-1741. 44

Han, C., & Li, H. (2010). Visual detection of melamine in infant formula at 0.1 ppm level based on 45

silver nanoparticles. Analyst, 135(3), 583-588. 46

He, D., Zhang, X., Gao, B., Wang, L., Zhao, Q., Chen, H., Wang, H., & Zhao, C. (2014). Preparation of 47

magnetic molecularly imprinted polymer for the extraction of melamine from milk followed 48

by liquid chromatography-tandem mass spectrometry. Food Control, 36(1), 36-41. 49

Jawaid, S., Talpur, F. N., Sherazi, S. T. H., Nizamani, S. M., & Khaskheli, A. A. (2013). Rapid 50

detection of melamine adulteration in dairy milk by SB-ATR–Fourier transform infrared 51

spectroscopy. Food Chemistry, 141(3), 3066-3071. 52

Jin, G.-P., Yu, B., Yang, S.-Z., & Ma, H.-H. (2011). Extremely sensitive electrode for melamine using a 53

kind of molecularly imprinted nano-porous film. Microchimica Acta, 174(3-4), 265-271. 54

Kuang, H., Chen, W., Yan, W., Xu, L., Zhu, Y., Liu, L., Chu, H., Peng, C., Wang, L., Kotov, N. A., & 55

Xu, C. (2011). Crown ether assembly of gold nanoparticles: Melamine sensor. Biosensors and 56

Bioelectronics, 26(5), 2032-2037. 57

Li, J., Qi, H. Y., & Shi, Y. P. (2009). Determination of melamine residues in milk products by zirconia 58

hollow fiber sorptive microextraction and gas chromatography-mass spectrometry. J 59

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

5

Chromatogr A, 1216(29), 5467-5471. 60

Liang, X., Wei, H., Cui, Z., Deng, J., Zhang, Z., You, X., & Zhang, X. E. (2011). Colorimetric detection 61

of melamine in complex matrices based on cysteamine-modified gold nanoparticles. Analyst, 62

136(1), 179-183. 63

Ma, Y., Niu, H., Zhang, X., & Cai, Y. (2011). One-step synthesis of silver/dopamine nanoparticles and 64

visual detection of melamine in raw milk. Analyst, 136(20), 4192-4196.Ma, Y., Niu, H., 65

Zhang, X., & Cai, Y. (2011). One-step synthesis of silver/dopamine nanoparticles and visual 66

detection of melamine in raw milk. Analyst, 136(20), 4192-4196. 67

Ping, H., Zhang, M., Li, H., Li, S., Chen, Q., Sun, C., & Zhang, T. (2012). Visual detection of 68

melamine in raw milk by label-free silver nanoparticles. Food Control, 23(1), 191-197. 69

Song, J., Wu, F., Wan, Y., & Ma, L.-H. (2014). Visual test for melamine using silver nanoparticles 70

modified with chromotropic acid. Microchimica Acta, 1-8. 71

Zhang, S., Yu, Z., Hu, N., Sun, Y., Suo, Y., & You, J. (2014). Sensitive determination of melamine 72

leached from tableware by reversed phase high-performance liquid chromatography using 73

10-methyl-acridone-2-sulfonyl chloride as a pre-column fluorescent labeling reagent. Food 74

Control, 39, 25-29. 75

Zhao, Y., Luo, W., Kanda, P., Cheng, H., Chen, Y., Wang, S., & Huan, S. (2013). Silver deposited 76

polystyrene (PS) microspheres for surface-enhanced Raman spectroscopic-encoding and rapid 77

label-free detection of melamine in milk powder. Talanta, 113(0), 7-13. 78

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 1 Recovery test results of the assay in liquid and solid milk samples with spiked melamine

Sample Added (10-7M) Average (10-7M) RSD (% , n=3) Recovery (%)

Liquid milk 9.0 9.8 1.9 109

21.0 21.6 1.3 103

Solid milk 9.0 9.7 1.5 108

21.0 20.3 0.9 97