Ultrasensitive turn-on fluorescent detection of trace thiocyanate based on fluorescence resonance energy transfer

Author's Accepted Manuscript

Ultrasensitive turn-on fluorescent detection of trace thiocyanate based on fluorescence resonance energy transfer Juan Song, Fang-Ying Wu, Yi-Qun Wan, Li-Hua Ma

www.elsevier.com/locate/talanta

PII: DOI: Reference:

S0039-9140(14)00851-0 http://dx.doi.org/10.1016/j.talanta.2014.10.023 TAL15162

To appear in:

Talanta

Received date: 13 August 2014 Revised date: 5 October 2014 Accepted date: 10 October 2014 Cite this article as: Juan Song, Fang-Ying Wu, Yi-Qun Wan, Li-Hua Ma, Ultrasensitive turn-on fluorescent detection of trace thiocyanate based on fluorescence resonance energy transfer, Talanta, http://dx.doi.org/10.1016/j. talanta.2014.10.023 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 galley proof before it is published in its final citable 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.



Ultrasensitive turn-on fluorescent detection of trace thiocyanate based on fluorescence resonance energy transfer Juan Song, Fang-Ying Wu , Yi-Qun Wan, Li-Hua Ma Department of Chemistry, Nanchang University, Nanchang 330031, China Tel: + 86 79183969882; Fax: + 86 79183969514. E-mail address: [email protected]



Corresponding authors: Department of Chemistry, Nanchang University, Nanchang 330031, China, Tel: + 86 79183969882; Fax: + 86 79183969514. E-mail address: [email protected]; [email protected]. Present address: ALS Ellington, 1414 lumpkin Road, Houston, Texas, 77043



1



Ultrasensitive turn-on fluorescent detection of trace thiocyanate based on fluorescence resonance energy transfer Juan Song, Fang-Ying Wu , Yi-Qun Wan, Li-Hua Ma* Department of Chemistry, Nanchang University, Nanchang 330031, China

Abstract Thiocyanate (SCNԟ) is a small anion byproduct of cyanide metabolism. Several methods have been reported to measure SCNԟ above the micromolar level. However, SCNԟ is derived from many sources such as cigarettes, waste water, food and even car exhaust and its effect is cumulative, which makes it necessary to develop methods for the detection of trace SCNԟ. In this paper, a simple and ultrasensitive turn-on fluorescence assay of trace SCNԟ is established based on the fluorescence resonance energy transfer (FRET) between gold nanoparticles (AuNPs) and fluorescein. The detection limit is 0.09 nM, to the best of our knowledge, which has been the lowest detection LOD ever without the aid of costly instrumentation. The fluorescence of fluorescein is significantly quenched when it is attached to the surface of AuNPs. Upon the addition of SCNԟ, the fluorescence is turned on due to the competition action between SCNԟ and fluorescein towards the surface of AuNPs. Under an optimum pH, AuNPs size and concentration, incubation time, the fluorescence enhancement efficiency [(IF-I0)/I0] displays a linear relationship with the concentration of SCNԟ in the range of 1.0 nM to 40.0 nM. The fluorescein-AuNP sensor shows absolutely high selectivity toward SCNԟ than other 16 anions. The common metal ions, amino acids and sugars have no obvious interference effects. The accuracy and precision were evaluated based on the recovery experiments. The cost effective sensing system is successfully applied for the determination of SCNԟ in milk products and saliva samples. Key words: Gold nanoparticles  Fluorescein  Fluorescence resonance energy transfer  Thiocyanate  Milk  Saliva

1.

Introduction

As an important chemical raw material, thiocyanate (SCNԟ) is widely employed in many industrial processes such as pesticide production, textile dyeing, electroplating, printing, photofinishing and hydrometallurgy. Compared with the lethal cyanide ion, SCNԟ is a less toxic anion that can be found at M in human bodies through the digestion of glucosinolate-containing brassica vegetables (broccoli, cauliflower, cabbage, etc.), or by ingesting cheese and milk,



Corresponding authors:Department of Chemistry, Nanchang University, Nanchang 330031, China, Tel: + 86 79183969882; Fax: + 86 79183969514. E-mail address: [email protected]; [email protected]. Present address: ALS Ellington, 1414 lumpkin Road, Houston, Texas, 77043



2



which contain SCNԟ [1]. SCNԟ ranks after perchlorate as a potent inhibitor of iodide uptake by the thyroid but may be more concentrated in some food items such as milk products [2]. And it is more serious for pregnant woman, infant or population in iodine deficiency area [3]. Besides, another significant source of SCNԟ in the body is from smoker. Cyanide in tobacco, once absorbed, is eliminated through its conversion to SCNԟ by the enzyme rhodanese, which is found in the mitochondria of liver and kidney cells [4]. Therefore, the concentration levels of SCNԟ in human body liquids are considered to be an important indicator for assessing habitual smoking behavior as well as distinguishing nonsmokers from smokers [1]. For the general population, low levels of serum SCN may predispose them to inflammatory or inflammation-mediated diseases [5]. All of these aspects underscore the importance of SCN determination.. Several methods for the SCN detection have been developed over the recent years, including ion chromatography [6], gas chromatography/mass spectrometry [7], colorimetry [4, 8, 9], electrophoresis [10], electrochemistry [11], fluorimetry [12], and surface-enhanced Raman scattering (SERS) [13, 14]. However, some of these methods lack sufficient sensitivity to detect SCNԟ at low levels (5-8.5 mg L-1) [15], which correspond to the SCNԟ levels typically found in milk samples. And most of them are time-consuming and costly or require sophisticated instrumentation and professional staffs. Hence, exploring an accurate, rapid, sensitive and low-cost method to detect SCNԟ in milk and human body liquids still remains a challenge. Fluorescence resonance energy transfer (FRET) has been widely exploited as an extremely useful tool in analytical and sensing applications [16, 17]. FRET is a kind of non-radiation energy transfer form, and the energy of the donors is transferred to the acceptors by the interaction of electric dipoles. The efficiency of FRET is determined by many factors including the overlap extent between the emission spectrum of the donors and the absorption spectrum of the acceptors, the distance between the donors and acceptors, the relative orientation of the electric dipoles of the donors and acceptors and so on. Gold nanoparticles (AuNPs) have been widely exploited for chemical and biochemical sensing [18-21] because of their high surface area-to-volume ratios, size-dependent optical properties, and easy to be modified or functionalized chemically. Particularly, AuNPs have been explored as the efficient energy acceptors to substitute for organic acceptors [22] and selected for establishing a FRET system as a fluorescence quencher [23], which have been developed for detection of DNA [24] in the presence of Rhodamine B, protein [25] with aptamer as part of the biosensor, iodide and iodate [26] based on fluorescein-5-isothiocyannate-modified AuNPs, thiols [27] based on Nile Red-Absorbed AuNPs, medicine captopril [28], glyphosate [29] based on the FRET between charged CdTe quantum dots and AuNPs, metallic ions[30] using quantum dots and AuNPs co-sensing system. It was reported that fluorescein can be absorbed on the surface of AuNPs, resulting in fluorescence quenching due to FRET between fluorescein and the AuNPs to detect 

3



acetylthiocholine with the aid of enzyme [31] and melamine [32]. Inspired by these works, we have examined the potential of FRET between fluorescein and AuNPs as a novel and sensitive probe for the the determination of SCNԟ in milk products and saliva. Upon addition of SCNԟ that competed with fluorescein molecules to absorb on the surface of the AuNPs, fluorescein molecules were freed from the surface of AuNPs and the fluorescence was recovered. We have tested another 16 different kinds of anions and only the presence of SCN- was able to restore fluorescence, allowing the ultrasensitive and selective determination of SCN- concentration within 5 min. The aim of present work is to propose an alternative methodology with economic aspects, on site, real time and operating simplicity to conventional techniques for SCNԟ detection in water, milk products and saliva samples of smokers and nonsmokers. (Scheme 1) 2. Experimental 2.1. Materials

Chemicals were of analytical grade and used without further purification unless otherwise stated. Hydrogen tetrachloroaurate hydrate (HAuCl44H2O) was bought from Sinopharm Chemical Reagent Co.Ltd. (Shanghai, China). Amino acids including Arginine (Arg), Histidine (His), Lysine (Lys), Tyrosine (Tyr), Valine (Val) and Threonine (Thr) were purchased from Shanghai Jingchun Technology Co. Ltd. (Shanghai, China). Na2SO4, Na2SO3, Na2S2O3, Na2S, Na2CO3, Na2C2O4, CH3CO2Na, NaNO3, NaNO2, KClO4, NaF, NaCl, kBr, KI, FeCl3, MgCl2, CaCl2, ZnCl2, and NH4Cl were purchased from Shanghai Qingxi Technology Co. Ltd. (Shanghai, China). All the carbohydrates including D-fructose, D-glucose and sucrose were purchased from Shanghai Lanji Technology Co. Ltd. (Shanghai, China). The pH value of the solution was adjusted by mixing different volume ratio of NaH2PO4 and Na2HPO4.

2.2. Apparatus

All fluorescence measurements were carried out on F-4600 spectrofluorimeter (Hitachi, Japan) equipped with a xenon lamp source and a 1.0 cm quartz cell, and the scan speed was 12,000 nm min1. UV-vis absorption spectra were recorded using UV-2550 spectrophotometer (Shimadzu, Kyoto, Japan) with a 1.0 cm quartz cell at room temperature. Transmission electron microscopy (TEM) was carried out with a JEM-2010 transmission electron microscope (JEOL Ltd. Japan). The samples were prepared by drop-coating the AuNPs solution onto the carbon-coated copper grid and were loaded onto a specimen holder for the purpose of TEM.

2.3. Preparation of colloidal Au NPs 

4



All pieces of the experimental glassware used were cleaned in a bath of freshly prepared aqua regia solution (HCl: HNO3, 3:1) and then rinsed thoroughly with H2O and finally dried in the oven at 100°C prior to use. Citrate capped gold nanoparticles with different sizes was synthesized based on the well-documented Frens’ method [33] (trisodium citrate was used as reducing agent). In this method, it was possible to control the size of the particles by varying [Au(III)]/[citrate] ratio during the reduction step as listed in Table S1. A standard procedure for the preparation of AuNPs with size 13 nm was summarized as follows. A 50 mL aqueous solution of HAuCl4 (0.25 mM) was heated to boiling point and 1.3 mL of trisodium citrate (1%) was added. The end point of the reaction was right after the solution reached a wine red color. The obtained wine-red solution was stored at 4 °C for further use. The UV–vis spectral characteristics and sizes for different sets of gold nanoparticles are summarized in Table S1 and results are in agreement with Frens’ method. The molar concentration of 13 nm sized AuNPs was calculated according to Beer’s law [34] (the molar extinction coefficient for 13 nm AuNPs is 2.78×108 (mol L-1)-1cm-1 at 520 nm).

2.4. Preparation of real samples

The pretreatment of liquid milk samples was carried out following the general procedure [35]. Briefly, 2 g milk product was added into 1.5 mL of 10% trichloroacetic acid and 5.0 mL of acetonitrile mixture to remove proteins in milk samples. The mixture solution was transferred to centrifugal tube to undergo sonication for 10 min and then centrifuged at 12,000 rpm min-1 for 15 min. The supernatant was filtered through a 0.22 m membrane filter to remove lipids. The pH of filtrate was adjusted to 6.8, and the filtrate was filtered through 0.22 m membrane filter again after centrifugation. The filtered liquid was diluted with water to 10 mL for further analysis. In the case of infant formula, about 0.6 g sample was used for the determination. Firstly, sample was dissolved with 2.0 mL water in a centrifuge tube, 1.5 mL of 10% trichloroacetic acid and 5.0 mL acetonitrile mixture were added to undergo 15 min ultrasonic treatment. Then, the mixture was centrifuged at 12,000 rmp min-1 for 15 min. The rest of operation was the same as the pretreatment of liquid milk. Saliva is a cleaner matrix. Saliva samples from healthy smoking and non-smoking volunteers were collected and refrigerated for 30 min at 4ºC, then centrifuged at 3000 rpm for 10 min. After centrifugation, 100 μL of the supernatant solution was transferred into 10.0 mL volumetric flask and diluted to the mark with deionized water.

2.5. Detection of SCNԟ 

5



Typically, a stock solution of 10 M fluorescein was prepared with Milli-Q water for the further use. 30 L of 10 M fluorescein was added into 200 L AuNPs (15 nM), the resulted mixture was equilibrated at the room temperature for 10 min. Then, 770 L of 0.01 M (pH 7.0) PBS (NaH2PO4-Na2HPO4) containing various concentrations of SCNԟ was mixed with above mixture solution and further incubated for 10 min. The fluorescence intensities at 531 nm (excitation wavelength was 459 nm) of the mixture in the presence (IF) and absence (I0) of SCNԟ were recorded, respectively.

2.6 Accuracy and Precision

A recovery study was conducted by adding known amounts (5 nM, 10 nM, 15 nM) of the SCN- to the real samples, followed by pre-treatment. Each concentration was replicated in triplicate. Degree of accuracy expresses the concordance between the accepted value and the value found. The evaluation of inter-day precision (intermediate precision) of the assay was performed on two different days by different analysts. The repeatability was evaluated on the same day for intra-day precision in 3 vessels used for the dissolution test. The relative standard deviation (RSD) and the relative mean square error prediction (RMSEP) from the results was calculated.

3. Results and discussion

3.1. Characterization of AuNPs

Transmission electron microscope datum was used to characterize the size and shape of AuNPs. We synthesized various sized AuNPs and chose the AuNPs sized 13nm as an example shown in Fig. 1 A, the prepared AuNPs were dispersed uniformly with spherical shape and the sizes were measured to be an average diameter of 13.1 nm with relative standard deviation as 24% (shown in Fig. S1). The characterizations of other sized AuNPs were listed in Table S1. (Fig. 1) 3.2. Interaction between SCN– and fluorescein-AuNPs assembly

The fluorescence quenching of fluorophores by AuNPs was dominated by the mechanisms of FRET, the inner filter effect and electron transfer [31]. FRET requires good overlap between the emission of the fluorophores and the SPR band of AuNPs. The emission spectrum of fluorescein (peak at 531 nm) was partially overlapped with the absorption spectrum 

6



of AuNPs in different sizes (as shown in Fig.2A). It has been reported that the fluorescein can be adsorbed onto the surface of citrate-stabilized AuNPs to form a fluorescein-AuNPs assembly due to the electrostatic interaction. The overlap integral areas of absorption spectrum of different sized AuNPs with emission spectrum of fluorescein were shown in Table S2. The fluorescence of fluorescein exhibits a maximum wavelength at 531nm and the SPR peak of AuNPs with size 13 nm is 520 nm, indicating the largest overlap integral area (4063) between the fluorescence of fluorescein and the SPR band of 13 nm AuNPs. Fig.S2A shows the fluorescence intensity change at 531 nm of the AuNPs-fluorescein system following successive addition of different size AuNPs and the Fig.S2B shows the resulting emission spectral change of addition of 13 nm sized AuNPs. It is clear that the fluorescence emission intensity decreased gradually with increasing the AuNPs concentration for all four different sized AuNPs. The fluorescence quenching efficiency can be quantified by the Stern-Volmer equation:

I0 I

1  K sv [Q]

(1)

where I0 and I are the fluorescence intensity in the absence and presence of the quencher (AuNPs), respectively. Ksv is the quenching constant and Q is the concentration of quencher. The Stern-Volmer plots (see Fig.2B) show that the different quenching efficiency with the different size of the AuNPs. The quenching constants derived from the slope of the different sized AuNPs were shown in Table S2. It is obvious that the 13 nm sized AuNPs had the largest quenching efficiency. The result is consistent with the conclusion that this type of AuNPs possess maximal FRET overlapping integral area. We also investigated the effect of the different size of AuNPs on the fluorescent enhancement efficiency with the addition of SCN– (as shown in Fig.S3). The result indicated that when the SCN– was added to the fluorescein-AuNPs with different sizes, the 13 nm sized fluorescein-AuNPs showed the maximum level of fluorescence recovery. The results may be due to that SCN– had good ability to interact with 13nm sized AuNPs. To sum up, the 13 nm sized AuNPs was selected as the FRET acceptor for the further experiments. Fluorescein molecules adsorbed on the surface of AuNPs can be released by introducing another compound, which can interact with the AuNPs, resulting in the recovery of the fluorescence [26-28]. To validate the competition between fluorescein and SCN–, the effect of SCN– on the fluorescence of the fluorescein-AuNPs mixture was performed and shown in Fig.2C. The fluorescence spectrum of fluorescein (curve ‘a’ in Fig.2C) was almost identical to that of the mixture of fluorescein and SCN– (curve ‘b’ in Fig.2C). The results provided clear evidence that there was no interaction between fluorescein and SCN–. When fluorescein was mixed with AuNPs sized 13nm, the fluorescence was significantly quenched (curve ‘c’ in Fig.2C) due to the FRET between fluorescein and AuNPs. Upon the addition of SCN– to the



7



solution of fluorescein-AuNPs, the fluorescein was detached from the AuNPs surface, leading to the recovery or turn-on of the fluorescence (curve ‘d’ in Fig.2C). In addition, we found that the SPR peak was slightly shifted after the addition of 3.0 M SCN– (Fig.S4). As shown in Figure 1B and 1C, 0.3 M SCN– can make the AuNPs aggregate slightly, which became stronger in the presence of 3.0 M of SCN– (Fig.1C). (Fig. 2)

3.3. Effect of AuNPs concentration and pH

To improve the sensitivity of the sensor, we investigated the effect of the AuNPs concentration and the pH of the solution on the displacement of fluorescein from the surface of AuNPs by SCNԟ. Fig.S5A shows the fluorescent enhancement efficiency [(IF-I0)/I0] in the presence of 15 nM SCNԟ as a function of AuNPs concentration. The efficiency was gradually increased with increasing the concentration of AuNPs and reached maximum when the concentration was up to 3 nM. The fluorescent enhancement efficiency started decreasing when the concentration was beyond 3 nM, which may be attributed to the fact that high concentration of the AuNPs increased the collisional encounter, leading to an enhancement in collisional quenching of released fluorescein. Thus, 3 nM of AuNPs was chosen for the following experiment. The effect of pH was also investigated in the range from pH 5.0 to 9.0. As shown in Fig.S5B, the fluorescent enhancement efficiency is highly dependent on pH value of solution and the highest point was obtained at pH 7.0. Therefore, the optimized pH value for SCNԟ detection system is 7.0.

3.4. Effect of incubation time

The effect of the incubation time of SCNԟ was investigated as well. Fig.S6 shows the plot of fluorescence intensity variation over the different concentrations of SCNԟ. It can be deduced that the fluorescence recovery was almost 90% completed within 10 min. Therefore, all the fluorescence measurements were performed 10 min after initiation of the reaction.

3.5. Effect of potentially interfering ions

To assess the recognition ability of fluorescein-AuNPs probe, the fluorescent enhancement efficiency responses from metal ions (including Ca2+, Mg2+, Zn2+, Fe3+, Na+, K+), anions (including SO42–, S2O32–, S2O32–, S2–, CO32–, C2O42–, AC–, 

8



NO3–, NO2–, EDTA2–, F–, Cl–, Br–, I–, ClO4–), sugars (glucose, fructose, sucrose) and amino acid (L-Histidine, L-Arginine, L-Lysine, L-Tyrosine,

L-Valine, L-Threonine), which possibly existed in the real samples were performed. As shown in

Fig.3 (Blue histograms), the addition of SCN– to a solution of 3.0 nM fluorescein-AuNPs resulted in a dramatic increase in the fluorescent enhancement efficiency [(IF-I0)/I0], while the other ions and compounds exhibited no such change. It turns out that this fluorecein-AuNPs probe showed excellent selectivity of SCN– over the other ions and compounds. To expand the field of application, the interference from these ions and compounds were further investigated as shown in Fig.3 (Red histograms). This was conducted through the analysis of a standard solution of SCN– (0.1 M) in the presence of the excess amounts of coexisting compounds. Ca2+, Mg2+, Zn2+, Fe3+, Na+, K+, SO42–, S2O32–, S2O32–, S2–, CO32–, C2O42–, AC–, NO3–, NO2–, EDTA, F–, Cl–, Br–, ClO4–, glucose, fructose, sucrose, L-Tyrosine, L-Valine, and L-Threonine did not interfere at the concentrations of 50 M, and L-Histidine, L-Arginine, L-Lysine did not interfere at the concentrations of 10 M, and 0.1 M of I– did not interference. In practical assays, the potential interference from amino acids may be insignificant and could be eliminated during sample pretreatment by removing the protein from the samples. (Fig. 3) 3.6. The sensitivity for the detection of SCNԟ

Under the optimized conditions, the capability of this sensing system for quantitative detection of SCNԟ was evaluated. The fluorescent spectra of fluorescein-AuNPs in the presence of different amounts of SCNԟ were shown in Fig.4A. A linear relationship between the fluorescent enhancement efficiency [(IF-I0)/I0] and the concentration of SCNԟ (CSCNԟ, nM) was observed in the range of 1.0 nM to 40.0 nM (as shown in Fig.4B) with a detection limit of 0.09 nM (3s/k). The calibration equation was: Y = -0.8974+1.172CSCNԟ with the satisfactory correlation coefficient (R2=0.998), where Y referred to the fluorescent enhancement efficiency [(IF-I0)/I0] and X was the concentration of SCNԟ. (Fig. 4) 3.7. Analysis of real samples

In order to validate the applicability of the fluorescein-AuNPs probe for SCN– sensing, a recovery study was conducted by adding known amounts of SCN– to the real samples. Each concentration was replicated in triplicate. That is to say, none and various concentration of SCN– were spiked into the milk and saliva samples according to the procedure described in Section 2.4. The accuracy and precision evaluation of the results of such measurements were then processed with the use of recoveries (%) and the relative mean square error prediction (RMSEP)[36] (Table 1). Degree of accuracy 

9



expresses the concordance between the accepted value and the value found. The measured recovery is typically between 90% and 110% with RSD lower than 10%, indicating the good accuracy of the method. The precision results of the proposed method in real samples were evaluated by analyzing repeatablility and intermediate precision. The assay evidenced a low RSD ranging from 0.9% to 8.6% for intra-day precision and from 0.3% to 8.5% for inter-day precision performed by the different analysts. The t-value of 2.57 is obtained with 95% significance. Analytical data in Table 1 showed that the recoveries varied from 102% to 104% in the spiked liquid milk sample, 100% to 107% in the spiked infant milk sample, 100% to 101% in the non-smoker saliva sample, and 101% to 108% and 101% to 105% Smoker1’ saliva and Smoker2’ saliva, respectively. The good recoveries, lower RSD and RMSEP (shown in Table 1 and Table S3) demonstrate the accuracy and reliability of the present fluorescence method for detecting SCN– in practical applications. (Table 1)

4. Conclusions

In summary, a sensitive turn-on fluorescent detection method for SCN– based on FRET from fluorescein to AuNPs is developed. It has been found that the 13 nm sized AuNPs exhibit higher quenching efficiency than other sized AuNPs, which is rationalized that this type of AuNPs shows the maximal overlap between its SPR absorption and the emission of fluorescein. This method offers a couple of advantages over the published SCN– detection techniques (as shown in Table S4). Firstly, this method can be operated without expensive instrument, which simplifies operations and reduces the associated costs. Secondly, it allows detection of concentrations as low as 0.09 nM, resulting in the rapid and sensitive detection of SCN– in milk and saliva samples. Finally, this sensor exhibits unbelievable selectivity for SCN– over the other ions and compounds. We will focus on the practical applications in the future research.

Acknowledgments

This work is financially supported by Jiangxi Province Scientific and Technological support major projects planned (No. 20133ACG70002, Jiangxi Province Natural Science Foundation (No. 20132BAB203011), and Natural Science Foundation of China (No. 21365014).

References



10



[1] K.A. O'Connell, M.M. Gerkovich, B.A. Fears, M.R. Cook, Addict. Behav. 13 (1988) 383-386. [2] R.A. Niemann, D.L. Anderson, J. Chromatogr. A 1200 (2008) 193-197. [3] K.K. Banerjee, P. Marimuthu, P. Bhattacharyya, M. Chatterjee, Brit. J. Nutr. 78 (1997) 679-681. [4] J. Zhang, C. Yang, X. Wang, X. Yang, Anal. Bioanal. Chem. 403 (2012) 1971-1981. [5] Y. Xu, S. Szép, Z. Lu, Proceedings of the National Academy of Sciences, 106 (2009) 20515-20519. [6] I. Demkowska, . Polkowska, J. Namienik, J. Chromatogr. B 875 (2008) 419-426. [7] J.S. Torano, H.J.M. van Kan, Analyst 128 (2003) 838-843. [8] Z. Zhang, J. Zhang, C. Qu, D. Pan, Z. Chen, L. Chen, Analyst 137 (2012) 2682-2686. [9] H.H. Deng, C.L. Wu, A.L. Liu, G.W. Li, W. Chen, X.H. Lin, Sens. Actuators, B 191 (2014) 479-484. [10] J.E. Prest, S.J. Baldock, M.S. Beardah, S.P. Doyle, P.R. Fielden, N.J. Goddard, B.J. Treves Brown, Analyst 136 (2011) 3170-3176. [11] M. Sancy, J. Pavez, M.A. Gulppi, I.L. de Mattos, R. Arratia-Perez, C. Linares-Flores, M. Paez, T. Nyokong, J.H. Zagal, Electroanal 23 (2011) 711-718. [12] A. Banerjee, A. Sahana, S. Lohar, I. Hauli, S.K. Mukhopadhyay, D.A. Safin, M.G. Babashkina, M. Bolte, Y. Garcia, D. Das, Chem. Commun. 49 (2013) 2527-2529. [13] X. Lin, W.L.J. Hasi, X.T. Lou, S. Lin, F. Yang, B.S. Jia, Y. Cui, D.X. Ba, D.Y. Lin, Z.W. Lu, J. Raman Spectrosc. 45 (2014) 162-167. [14] P. Pienpinijtham, X.X. Han, S. Ekgasit, Y. Ozaki, Anal. Chem. 83 (2011) 3655-3662. [15] Report of an FAO/WHO technical meeting, Benefits and potential risks of the lactoperoxidase system of raw milk preservation,http://apps.who.int/iris/handle/10665/43553. [16] C.A. Chen, C.C. Wang, Y.L. Chen, Y.J. Jong, S.M. Wu, Chem. Commun. 50 (2014) 8069-8072. [17] C. Wang, Y. Tang, Y. Liu, Y. Guo, Anal. Chem. 86 (2014) 6433-6438. [18] Z. Chen, G. Zhang, X. Chen, J. Chen, J. Liu, H. Yuan, Biosens. Bioelectron. 41 (2013) 844-847. [19] X. Liu, Y. Wang, P. Chen, Y. Wang, J. Zhang, D. Aili, B. Liedberg, Anal. Chem. 86 (2014) 2345-2352. [20] Y. Wu, S. Zhan, L. Wang, P. Zhou, Analyst 139 (2014) 1550-1561. [21] G. Lu, P. Tan, C. Lei, Z. Nie, Y. Huang, S. Yao, Talanta128 (2014) 360-365. [22] M. Li, S.K. Cushing, Q. Wang, X. Shi, L.A. Hornak, Z. Hong, N. Wu, J. Phys. Chem. Lett. 2 (2011) 2125-2129. [23] H. Zhang, L. Wang, W. Jiang, Talanta 85 (2011) 725-729. [24] L. Dyadyusha, H. Yin, S. Jaiswal, T. Brown, J.J. Baumberg, F.P. Booy, T. Melvin, Chem. Commun. (2005) 

11



3201-3203. [25] W. Wang, C. Chen, M. Qian, X.S. Zhao, Anal. Biochem. 373 (2008) 213-219. [26] Y.M. Chen, T.L. Cheng, W.L. Tseng, Analyst 134 (2009) 2106-2112. [27] S.J. Chen, H.T. Chang, Anal. Chem. 76 (2004) 3727-3734. [28] M.R. Hormozi-Nezhad, H. Bagheri, A. Bohloul, N. Taheri, H. Robatjazi, J. Lumin. 134 (2013) 874-879. [29] J. Guo, Y. Zhang, Y. Luo, F. Shen, C. Sun, Talanta 125 (2014) 385-392. [30] X. Wang, X. Guo, Analyst 134 (2009) 1348-1354. [31] S.Y. Lim, J.H. Kim, J.S. Lee, C.B. Park, Langmuir 25 (2009) 13302-13305. [32] L. Guo, J. Zhong, J. Wu, F. Fu, G. Chen, Y. Chen, X. Zheng, S. Lin, Analyst 136 (2011) 1659-1663. [33] G. Frens, Nature (Physical Science) 241 (1973) 20-22.

[34] J. R. G. Navarro, M. H. V Werts. Analyst 138(2013), 583-592. [35] Y. Ma, H. Niu, X. Zhang, Y. Cai, Analyst 136 (2011) 4192-4196.

[36] Y. Ni, Z. Xia, S. Kokot, J. Hazard. Mater. 192 (2011) 722-729.



12



Captions Table 1 The application of the proposed method for analysis of milk and saliva samples spiked with different amounts of SCN– and the relevant statistic calculation results.

Scheme1. Schematic illustration for the sensing of SCNԟ based on FRET.

Fig. 1. TEM images of (A) AuNPs, (B) AuNPs with fluorescein (0.3 M) and low concentration of SCN– (0.5 M), and (C) AuNPs with fluorescein (0.3 M) and a high concentration of SCN– (3.0 M). The concentration of AuNPs is 3.0 nM with 13 nm size.

Fig. 2. (A) Emission spectra of fluorescein (0.3M) and absorption spectra of AuNPs (3.0 nM) with different sizes (8, 13, 41 and 55 nm), respectively; (B) Stern-Volmer plots. The quenching efficiencies of AuNPs as a function of particle sizes; (C) Fluorescence spectra of (a) fluorescein, (b) fluorescein and SCN– mixture, (c) fluorescein and AuNPs sized 13 nm, (d) AuNPs sized 13 nm, fluorescein and SCN– in pH 7.0 buffer (0.01M Na2HPO4-NaH2PO4). Fluorescein, 0.3M; SCN–, 0.2M; AuNPs, 3nM.

Fig. 3. The fluorescent enhancement efficiency [(IF/I0)/I0] of fluorescein (0.3 M)-AuNPs (3.0 nM) in the presence of different kinds of foreign ions and compounds (0.1 M) without (Blue bars) and with SCNԟ (0.1 M)(Red bars) in pH 7.0 buffer (0.01M Na2HPO4-NaH2PO4). The concentrations of testing substances were 50 M for Ca2+, Mg2+, Zn2+, Fe3+, Na+, K+, SO42–, S2O32–, S2O32–, S2–, CO32–, C2O42–, AC–, NO3–, NO2–, EDTA, F–, Cl–, Br–, ClO4–, glucose, fructose, sucrose, L-Tyrosine, L-valine, and L-threonine, 10 M for L-histidine, L-arginine, L-lysine and 0.1 M for I–, respectively.

Fig. 4. (A) Fluorescent spectra of fluorescein-AuNPs in the presence of SCNԟ at various concentrations in pH 7.0 buffer (0.01M Na2HPO4-NaH2PO4). Amounts of SCNԟ were 1.0, 3.0, 5.0, 10.0, 15.0, 20.0, 25.0, 30.0, 35.0, 40.0, 50.0, 100.0 and 200.0 nM from bottom to top, respectively. (B) Effect of the SCNԟ concentration on the fluorescent enhancement efficiency. Inset is the calibration curve.



13



Tables

Table 1 The application of the proposed method for analysis of milk and saliva samples spiked with different amounts of SCN–. Sample liquid Milk Infant milk Non-smo ker saliva Smoker1’ saliva Smoker2’ saliva

Determination

Not found

Not found

Not found

Not found

Not found

Added

Amount found

RSD

Recovery

RMSEPb

(10-9 M)

(10-9M)a

(%)

(%, n=6)

(%)

5.0

5.1±0.3

c

5.2

102

10.0

10.3±0.4

3.3

103

15.0

15.6±1.2

6.4

104

5.0

5.0±0.4

7.0

100

10.0

10.7±0.7

5.8

107

15.0

15.8±1.0

5.5

106

5.0

5.0±0.5

8.6

100

10.0

10.1±0.6

5.6

101

15.0

15.0±0.4

2.4

100

5.0

5.2±0.4

6.2

103

10.0

10.7±0.2

1.5

108

15.0

15.2±0.2

0.9

101

5.0

5.1±0.1

2.0

102

10.0

10.5±0.7

5.6

105

15.0

15.1±0.9

5.3

101

3.7

5.7

0.5

4.5

2.9

a The range amount found is evaluated by [ r W V , the value of t0.95,5 is 2.57. Q b RMSEP means relative mean square error prediction, 506(3

¦  [ IRXQG  [ DGGHG ¦  [ DGGHG 





c Average ±standard deviation (SD) of the three replicate measurements for each test sample. 



14



 Highlights

z

A fluorescent off-on probe for trace thiocyanate (SCNԟ) was exploited.

z

The assay for SCNԟ is ultrasensitive with detection limit 0.09nM.

z

The method cost is 0.2$ per sample with the aid of fluorescence spectrometer.

z

The operation is ultraselective and relatively fast response time within10 min.





15

*Graphical Abstract (for review)

Graphical abstract

A simple and ultrasensitive turn-on fluorescence assay of trace thiocyanate (SCN‒) is established based on the fluorescence resonance energy transfer (FRET) between the gold nanoparticles (AuNPs) and fluorescein. The limit of detection is 0.09nM which has been the lowest ever reported without the expensive instrumentaion to our best of knowledge.

Scheme 1

Fig 1

Fig 2

Fig 3

Fig 4