- Email: [email protected]
A label-free SERS probe for highly sensitive detection of Hg2+ based on functionalized Au@Ag nanoparticles
Talanta 162 (2017) 374–379
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
A label-free SERS probe for highly sensitive detection of Hg2+ based on functionalized Au@Ag nanoparticles
crossmark
⁎
Yi Zenga,1, Lihua Wanga,b,1, Lingwen Zengb, Aiguo Shena, , Jiming Hua a Key Laboratory of Analytical Chemistry for Biology and Medicine (Ministry of Education), College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072, PR China b Institute of Environmental and Food Safety, Wuhan Academy of Agricultural Science and Technology, Wuhan 430207, PR China
A R T I C L E I N F O
A BS T RAC T
Keywords: Label-free DMcT Hg2+ SERS Hg–N bond
Unbiodegradable toxic Hg2+ accumulates in ecological systems and results in contaminated food chain, exposing us to high level pollution and healthy risk. Here, monolayer 2,5-Dimercapto-1,3,4-thiadiazole (DMcT) functionalized Au@Ag nanoparticles (NPs) were controllably constructed as a label-free SERS probe. Based on them, we can perform rapid and easy Hg2+ identification with high sensitivity and selectivity over competing analytes. Remarkably, DMcT acted as both interior label and target acquisition. Since DMcT showed intrinsic Raman signal when attached to substrates surface, it can be employed for quantification instead of extra conventional signal label. Also, we’ve demonstrated that DMcT coordinated on the surface of Au@Ag NPs as bidentate ligand with both the thiocarbonyl sulfur atoms while the nitrogen atoms on the different sides of the molecule were devoted to Hg2+ recognition. Owing to the strong coordination between Hg2+ and nitrogen atoms, as low as 10 pM Hg2+ can be detected. The probe responded a good linear relationship ranging from 0.05 to 100 nM and the limit of detection is ~3 orders of magnitude lower than the United States Environmental Protection Agency (USEPA)-defined limit (10 nM) in drinkable water (EPA: Washington, DC, 2001). Furthermore, our suggested platform is highly effective to perform real samples detection. Utilizing the environmental water from East Lake, Wuhan, it resulted in better accuracy over the conventional standard method.
1. Introduction Mercury and most of its compounds are extremely toxic but are still continuously releasing into the ecosystem. Natural sources such as volcanic eruption are responsible, while fossil fuel burning, waste incineration, and gold mining caused by human also generated plenty mercury emissions. It permeates into soil, flows out into oceans, ultimately accumulates in human body via the food chain [1–4]. Actually, due to the poisonous unbiodegradable property, mercury's long-lasting toxic effects has already caused problems for endangered marine life. As for human beings, mercury exposure severely affects our endothelial and cardiovascular function, even at low doses. It also significantly induces changes in the central nervous system [2,3,5]. Thus, mercury poisoning is becoming a huge concern. Driven by the need, it is essential to develop convenient and simple platform for probing mercury distribution in both artificial and environmental sources, of course high sensitivity and selectivity are in demand. Specifically, the choice of real objects for analysis of mercury content
⁎
1
always goes to drinkable water along with where it come from in natural environment, for example, the river or lake water utilized for daily use. Till now, assays for mercury ions mainly lay in two fields. Traditional one is the instrumental analysis such as atomic absorption spectrometer (AAS) [6]. Besides, sensing platforms are widely used, such as resonance scattering spectrophotometry [7,8], electrochemistry [9,10], fluorimetry [11–13], and colorimetry [14,15]. Among all the above, surface enhanced Raman scattering (SERS) has been given considerable attention by the public. SERS possesses high sensitivity and narrow band, while providing spectroscopic fingerprints information, in addition it's anti-photobleaching and also nondestructive to sample [16,17]. Plenty of sensor platforms have been developed based on SERS. Ding et al. [18] synthesized gold nanoparticles (NPs)/ reduced graphene oxide hetero-junctions via a seed-assisted growth process, and utilized it for trace analysis of Hg2+ via thymine-Hg2+thymine coordination. Kang et al. [19] reported a single nanowire-onfilm SERRS sensor for Hg2+ detection based on structure-switching
Corresponding author. E-mail address: [email protected] (A. Shen). These authors contributed equally to the work.
http://dx.doi.org/10.1016/j.talanta.2016.09.062 Received 20 May 2016; Received in revised form 18 September 2016; Accepted 27 September 2016 Available online 28 September 2016 0039-9140/ © 2016 Elsevier B.V. All rights reserved.
Talanta 162 (2017) 374–379
Y. Zeng et al.
Scheme 1. Schematic illustration of the fabrication of Au@Ag-DMcT SERS probe and a sensing protocol for Hg2+.
double stranded DNAs. Binding of Hg2+ induces conformational changes of the dsDNAs and let a Raman reporter get close to the selective single nanowire-on-film structure, thereby turning on SERRS signal. Certainly they’ve made progress to some extent, but considering the probe construction part in those sensors, the procedure usually includes Raman active molecule labelling or DNA functionalization, which goes against convenience, simplicity and widely probe application. Because to some degree it makes detection more complicated and raises the whole cost, resulting in obstacles on the way of practical and extensive use. Hence, developing simple and rapid sensor for Hg2+ detection with low cost is of great significance. Herein, we presented a highly sensitive and selective monolayer 2,5-Dimercapto-1,3,4-thiadiazole (DMcT) functionalized Au@Ag nanoparticles (Au@Ag-DMcT) label-free probe for Hg2+ detection ( Scheme 1). Utilizing Au@Ag-DMcT as the SERS substrate, the platform mainly based on DMcT's original Raman signals and its coordination with metal ions. Facile quantification of Hg2+ in homogeneous solution was achieved by linking the amount of Hg2+ and ‘hot spot’ effect of SERS-based NPs. The incremental Raman intensity of DMcT was characteristically related to Hg2+ concentration over a range of 0.05– 100 nM. Actually, there has been several reports on Hg2+ detection based on DMcT capped noble metal NPs [11], but the reaction mode between DMcT and NPs are often overlooked. Generally, DMcT can coordinate as bidentate ligand and has mainly four donor sites: two nitrogen atoms and two sulfur atoms. In particular, according to our previous work [20] and via Raman spectra analysis, here we revealed that DMcT coordinated on the surface of Au@Ag NPs as disulfide salts, and the nitrogen atoms were left for Hg2+ recognition. Owing to the strong coordination between Hg2+ and nitrogen atoms, we can perform ultra-sensitive detection of Hg2+ low to 10 pM.
All other chemicals were of analytical grade or better quality, and used as received. 2.2. Apparatus The morphology and microstructure of the different NPs were characterized by using high resolution transmission electron microscope (TEM, JEM-2100, JEOL, Japan) operating at a 200 kV accelerating voltage. UV–vis adsorption spectra were recorded on a UV–vis spectrophotometer (UV-2550, Shimadzu, Japan) at room temperature using a 600 μL black body quartz cuvette with 1 cm path length. The Raman spectra were collected by Horbia Jobin-Yvon HR-800 Raman microspectrometer (System HR800, Horbia Jobin-Yvon, Villeneuve d′Ascq, France). 2.3. Synthesis of Au@Ag NPs A proper size of Au@Ag core–shell NPs as SERS substrates were synthesized according to our previous work [21]. First, 50 mL of 0.01% (w/w) HAuCl4 was reduced by sodium citrate solution (1%, w/w) at 100°C under vigorous magnetic stirring for 20 min. The as-prepared Au NPs were cooled down to room temperature and served as seed particles. Then, the Au@Ag NPs were synthesized as follows: aqueous solutions of sodium citrate (38.8 mM, 200 μL), ascorbic acid (0.1 M, 50 μL) and seeds (5 mL) were combined at room temperature; to this mixture, AgNO3 (0.01 M, 100 μL) was added drop wise immediately under continuous magnetic stirring. The solution colour immediately changed from burgundy red to orange within 2 min, indicating the formation of Au@Ag NPs. To make the reaction fully completed, the solution was stirred for 15 min and stored in 4°C.
2. Experimental section
2.4. Synthesis of Au@Ag-DMcT and detection of Hg2+
2.1. Reagents
Different volumes of freshly prepared 0.5 mg mL−1 DMcT aqueous solution (ultrasonically dispersed in phosphate buffer (PB) suspension (0.1 M, pH 7.4)) were added to the colloid solution of Au@Ag NPs with slow stirring, and the suspensions were allowed to incubate for several hours. The obtained NPs were then rinsed by centrifugation and resuspension with deionized water for three times. For Hg2+ detection, different concentration of Hg(NO3)2 solution were added into 500 μL probe solution, mixed for 1–2 min, and then Raman signal were
Tetrachloroauric(III) acid and sodium citrate were purchased from Sigma (USA). Silver nitrate was obtained from Beijing Chemical Reagents Company (Beijing, China). Ascorbic acid was supplied by Sinopbarm Chemical Reagent Co., Ltd. 2,5-Dimercapto-1,3,4-thiadiazole (DMcT, ≥ 98%) was a product of Alfa Aesar. Mercuric nitrate was purchased from Tongren Chemical Reagent Factory (Guizhou, China). 375
Talanta 162 (2017) 374–379
Y. Zeng et al.
Fig. 1. (A) The absorption spectra of Au@Ag NPs (solid line) and Au@Ag-DMcT (dotted line). The inset shows TEM image of Au@Ag-DMcT; (B) SERS spectra of Au@Ag NPs (solid line) and Au@Ag-DMcT (dotted line); (C) normalized SERS spectra of the disulfide salts (dotted line) and monosulfide salts (solid line) of DMcT; (D) SERS spectra of DMcT in solid state (solid line) and Au@Ag-DMcT after drying in the air (dotted line).
coordination [11]. Actually, DMcT has mainly four donor sites and can coordinate as bidentate ligand with (a) both the nitrogen atoms, (b) both the thiocarbonyl sulfur atoms, or (c) one nitrogen atom and one sulfur atom on either the same side or different sides of the molecule [20]. To figure out the reaction mode between DMcT and Au@Ag NPs in this case and further reveal the detection rationale, more complemented experiments were carried out. Initially, we found that monosulfide salts of DMcT possess 1360 and 1410 cm−1 belonging to C=N stretching vibration, while disulfide salts only show 1360 cm−1 (Fig. 1C), which is consistent with the previous work [23]. Moreover, a droplet concentrated probe solution was settled on tin foil to dry in air and the Raman spectra were further collected. As is shown Fig. 1D, the stretching vibration peak of S–H of DMcT at 2475 cm−1 was initially observed but disappeared after incubation with Au@Ag NPs; meantime, a S–Ag bond at ca. 220 cm−1 appeared. These data suggest that the S–H bond of DMcT was broken and the S–Ag bond was formed during the reaction between Au@Ag NPs and DMcT. Combination with the SERS spectra of Au@Ag-DMcT in Fig. 1B (only 1360 cm−1 peak shows in the C=N stretching vibration), we can safely infer that two S–H bond of DMcT have coordinated with Au@Ag NPs in our experiment.
collected. 3. Results and discussion 3.1. Characterization of Au@Ag-DMcT Spontaneous reaction between DMcT and Au@Ag NPs in homogeneous water solution was investigated. The UV–vis spectra of Au@ Ag NPs before and after incubation with DMcT were shown in Fig. 1A. There are two peaks at ca. 385 and 508 nm in the spectrum of the original Au@Ag NPs. These two peaks has been attributed to the surface Plasmon resonance (SPR) absorption of the Ag shell and Au core of Au@Ag NPs, respectively [21]. Related to our previous work [20], after incubation of the Au@Ag NPs with DMcT, the Ag peak declined because Ag has reacted with DMcT, meanwhile a little bit red shift can be observed for the Au absorption, mainly due to Au@Ag NPs surface condition change from Ag-DMcT reaction. The TEM characterization (the illustration in Fig. 1A) also implied the particles are welldispersed and uniform. To further confirm the successful functionalization of DMcT on the surface of Au@Ag NPs, we represented Raman spectra in Fig. 1B. It can be seen that the pure Au@Ag NPs showed no specific Raman signal, while two characteristic peaks at 664 and 1360 cm−1 appeared in the spectra of Au@Ag-DMcT, which should be assigned to the symmetric vibrations of C–S and C=N of DMcT anions, respectively [22]. As we know, previous research on the application of DMcT capped noble metal NPs on Hg2+ detection almost neglected the reaction mode of DMcT on NPs, which just oversimplified and assume it as sulfur
3.2. Detection rationale As we have confirmed DMcT existed as disulfide salts in Au@AgDMcT, donor sites left for metal ions must be the nitrogen atoms. The added Hg2+ coordinated with nitrogen atoms and induced aggregation of Au@Ag-DMcT, which consequently resulted in the enhancement of 376
Talanta 162 (2017) 374–379
Y. Zeng et al.
Fig. 2. (A) SERS spectra of the Au@Ag-DMcT based nanosensors in the absence (solid line) and presence (dotted line) of Hg2+(20 nM); (B) TEM image of Au@Ag-DMcT after addition of 20 nM Hg2+.
DMcT's SERS signals. Detection of Hg2+ in homogeneous solution was achieved by linking the amount of Hg2+ and ‘hot spot’ effect of SERSbased NPs. For feasibility, we compared Au@Ag-DMcT's Raman spectra before and after additive of 20 nM Hg2+. As Fig. 2A shown, DMcT responded obviously to Hg2+, typical DMcT peaks at 664 and 1360 cm−1 apparently increased after Hg2+ addition. The corresponding TEM images (Fig. 2B) are also demonstrated, indicating particles aggregation induced by Hg2+-DMcT coordination.
thick, SERS intensity became higher and higher till 200 μL. However, continuously increasing AgNO3 amount over that value resulted in turbid solution, which reflected that Au@Ag NPs turned to be unstable in this condition and easily aggregated or even got precipitated. Consequently, R0 increased and R-R0 declined in contrast, ending up with 200 μL as the most optimized. According to our previous research [20], when DMcT reacted with Au@Ag NPs, it firstly bound to the surface of Au@Ag NPs in the form of disulfide salts. If there are excessive DMcT in the system, the disulfide salts will further react with DMcT to form the monosulfide salts, which will interact with each other based on the bridge of Ag+ ions to form a heavy layer of infinite coordination polymers. Thus we further tested different incubation time of DMcT and Au@Ag NPs. As Fig. 3C shown, during 1–5 h, R-R0 reached maximum value at about 3 h. It indicated that DMcT disulfide salts had totally occupied the surface of Au@Ag NPs by monolayer adsorption in this period. With incubation time further increasing, polymer layer will get thicker and thicker, which will hinder aggregation of probes, resulting in R-R0 reduction. For buffer condition influence, we tested probe performance in different buffer pH. As shown in Fig. 3D, the developed SERS probe preferred mild acid and lower pH leaded to better performance. However, probe itself was unstable and easily get aggregated when buffer pH was 3.5. So we chose pH 4.5 as the best.
3.3. Optimization of experimental parameters For best probe performance, we optimized necessary experimental parameters. The amount of DMcT, composition ratio of Au@Ag NPs, incubation time of DMcT with Au@Ag NPs, and buffer pH were all taken into account. First of all, the amount of DMcT utilized belonged to key point as it largely affected the probe structure, subsequently performance. As Fig. 3A shown, upon the increase of DMcT volume, the corresponding Raman signal difference R-R0 (R0 and R represent Raman intensity before and after Hg2+ added, respectively) declined apparently. Especially, when the volume exceeded 10 μL, R-R0 barely changed then but achieved a steady low value instead. This phenomenon was reasonable. Since more DMcT were introduced to the system, the thicker the DMcT-Ag infinite coordination polymers shell would be [20]. This provided stronger protection to Au@Ag NPs that in consequence would not easily response to Hg2+ addition, so that the enhancement became too weak to be observed. As the thickness reached a certain extent, no more aggregation would happen with Hg2+ adding, and the Raman signal seemed to be steady. Moreover, we found that when the volume of DMcT was less than 1.25 μL, the structure formed was not stable. Thus we chose 1.25 μL DMcT as the best. As known to all, gold colloidal NPs have advantages including longterm stability, controllable size distribution, and high homogeneity. Conversely, the Raman enhancement factor of the gold NPs is ~100– 1000 times lower than that of silver. However the silver NPs also have got fatal flaws—very unstable, due to the limited solubility and aggregation properties [24,25]. Here we combined their both advantages and selected Au@Ag NPs as SERS substrates for satisfying stability and Raman enhancement owing to gold NPs core and silver shell, respectively [21]. Obviously, best molar ratio Ag/Au should be explored, which is carried out by controlling the diameter of Au NPs and changing the amount of AgNO3 during Au@Ag NPs synthesis. Au NPs with a diameter of 30 nm were employed as seeds owing to the stability and relatively high SERS enhancement. As shown in Fig. 3B, with the AgNO3 amount gradually increasing and silver shell getting
3.4. Quantitatively SERS detection of Hg2+ To explore the potential of Au@Ag-DMcT for quantitatively SERS detection of Hg2+, Raman spectroscopic homogeneous assay was performed in aqueous buffer solution under optimized condition. Enhanced SERS response of Au@Ag-DMcT was observed upon the addition of Hg2+ solution into SERS-based NPs colloids. Fig. 4A showed the corresponding Hg2+ concentration-dependent SERS spectra of Au@Ag-DMcT. With Hg2+ concentration increasing, since coordination between Hg2+ and DMcT caused probe aggregation, typical peak of DMcT were largely enhanced as highly recognizable SERS signals at 664 and 1360 cm−1 could be rapidly detected. Here, we picked out 1360 cm−1 for quantification. The trend for enhanced SERS intensity at 1360 cm−1 varying with Hg2+ concentration were shown in Fig. 4B. The SERS intensity displayed an excellent linear correlation with Hg2+ concentration ranging from 0.05 to 100 nM (R2 = 0.986). It is worth noting that Hg2+ as low as 10 pM can be detected (Supplementary data, Fig. S1), which is ~3 orders of magnitude lower than the United States Environmental Protection Agency (USEPA)defined limit (10 nM) in drinkable water (EPA: Washington, DC, 2001). Meanwhile, it was also remarkably superior to majority of 377
Talanta 162 (2017) 374–379
Y. Zeng et al.
Fig. 3. Optimization of the volume of 0.5 mg/mL−1 DMcT (A), the volume of 0.01 M AgNO3 (B), the incubation time of DMcT and Au@Ag NPs (C), and the value of pH (D). The error bars represent the standard deviations based on at least three independent measurements.
3.5. Selectivity of the SERS sensor for Hg2+
reported probes. For instance, there are electrochemical sensor for Hg2+ using amplification based on thymine−Hg2+−thymine base pairs and gold nanoparticles (0.5 nM) [10], SERS technique for Hg2+ detection utilizing interaction between silver nanoparticles and mercury ions (90.0 pM) [5], SERS sensor of Hg2+ by gold nanoparticles/ graphene hetero-junctions (0.1 nM) [18], and Hg2+ fluorescent probes using mesoporous cerium phosphonate nanostructured hybrid spheres (16 nM) [12], and so on. Moreover, the reproducibility of the SERS spectra was tested by acquiring several spectra from 8 times random test from 4 different batches of SERS NPs with 20 nM Hg2+, which are presented in Fig. S2. The percent coefficient of variation (CV%) is calculated as around 6.8, which indicates that the suggested probe possesses excellent reproducibility.
To evaluate the selectivity of this SERS platform, a series of control experiments between Hg2+ and other coexisting interfering metal ions were performed (Fig. 5). The potential competing substances were separately detected under the optimized conditions, including common ions in natural experiment such as Pb2+, Cd2+, Fe2+, Al3+, Fe3+, Cu2+, Na+, K+, Mn2+, Ni2+, Cr3+, Zn2+, Mg2+, Ca2+, and Co2+. The concentration were set as 1 μM Hg2+ and 0.1 mM for the others (Especially, the concentration of Na+ and K+ were 1 mM). It was clear that interference ions barely produced any SERS signal enhancement. The excellent selectivity of this platform mainly owe to coordination mechanisms between Hg2+ and DMcT. Actually, the nitrogen atom plays a role as main donor site here. Since both the two nitrogen atoms with lone pair of electrons could coordinate with Hg2+, quadridentate coordination
Fig. 4. (A) SERS spectra of Au@Ag-DMcT with different Hg2+ concentration; (B) SERS intensity change at 1360 cm−1 against concentration of Hg2+. The inset shows the linear plot of the relative Raman intensity against different Hg2+ concentrations. The error bars represent the standard deviations based on three independent measurements at least.
378
Talanta 162 (2017) 374–379
Y. Zeng et al.
References [1] X.J. Xue, F. Wang, X.G. Liu, One-step, room temperature, colorimetric detection of mercury (Hg2+) using DNA/nanoparticle conjugates, J. Am. Chem. Soc. 130 (2008) 3244–3245. [2] E.M. Nolan, S.J. Lippard, Tools and tactics for the optical detection of mercuric ion, Chem. Rev. 108 (2008) 3443–3480. [3] M.M. Veiga, J.A. Meech, N. Oñate, Mercury pollution from deforestation, Nature 368 (1994) 816–817. [4] C. Díez-Gil, R. Martínez, I. Ratera, T. Hirsh, A. Espinosa, A. Tárraga, P. Molina, O.S. Wolfbeis, J. Veciana, Selective picomolar detection of mercury(II) using optical sensors, Chem. Commun. 47 (2011) 1842–1844. [5] W. Ren, C.Z. Zhu, E.K. Wang, Enhanced sensitivity of a direct SERS technique for Hg2+ detection based on the investigation of the interaction between silver nanoparticles and mercury ions, Nanoscale 4 (2012) 5902–5909. [6] A.Q. Shah, T.G. Kazi, J.A. Baig, H.I. Afridi, M.B. Arain, Simultaneously determination of methyl and inorganic mercury in fish species by cold vapour generation atomic absorption spectrometry, Food Chem. 134 (2012) 2345–2349. [7] Y. Chen, L.H. Wu, Y.H. Chen, N. Bi, X. Zheng, H.B. Qi, M.H. Qin, X. Liao, H.Q. Zhang, Y. Tian, Determination of mercury(II) by surface-enhanced Raman scattering spectroscopy based on thiol-functionalized silver nanoparticles, Microchim. Acta 177 (2012) 341–348. [8] D. Han, S.Y. Lim, B.J. Kim, L.L. Piao, T.D. Chung, Mercury(II) detection by SERS based on a single gold microshell, Chem. Commun. 46 (2010) 5587–5589. [9] Y. Zhang, G.M. Zeng, L. Tang, J. Chen, Y. Zhu, X.X. He, Y. He, Electrochemical sensor based on electrodeposited graphene-Au modified electrode and nanoAu carrier amplified signal strategy for attomolar mercury detection, Anal. Chem. 87 (2015) 989–996. [10] Z.Q. Zhu, Y.Y. Su, J. Li, D. Li, J. Zhang, S.P. Song, Y. Zhao, G.X. Li, C.H. Fan, Highly sensitive electrochemical sensor for mercury(II) ions by using a mercuryspecific oligonucleotide probe and gold nanoparticle-based amplification, Anal. Chem. 81 (2009) 7660–7666. [11] N. Vasimalai, S. John, Mercaptothiadiazole capped gold nanoparticles as fluorophore for the determination of nanomolar mercury(II) in aqueous solution in the presence of 50000-fold major interferents, Analyst 137 (2012) 3349–3354. [12] Y.P. Zhu, T.Y. Ma, T.Z. Ren, Z.Y. Yuan, Mesoporous cerium phosphonate nanostructured hybrid spheres as label-free Hg2+ fluorescent probes, ACS Appl. Mater. Interfaces 6 (2014) 16344–16351. [13] D.B. Liu, S.J. Wang, M. Swierczewska, X.L. Huang, A.A. Bhirde, J.S. Sun, Z. Wang, M. Yang, X.Y. Jiang, X.Y. Chen, Highly robust, recyclable displacement assay for mercuric ions in aqueous solutions and living cells, ACS Nano 6 (2012) 10999–11008. [14] J.J. Du, B.W. Zhu, X.D. Chen, Urine for plasmonic nanoparticle-based colorimetric detection of mercury ion, Small 9 (2013) 4104–4111. [15] G. Sener, L. Uzun, A. Denizli, Lysine-promoted colorimetric response of gold nanoparticles: a simple assay for ultrasensitive mercury(II) detection, Anal. Chem. 86 (2014) 514–520. [16] J. Ando, K. Fujita, N.I. Smith, S. Kawata, Dynamic SERS imaging of cellular transport pathways with endocytosed gold nanoparticles, Nano Lett. 11 (2011) 5344–5348. [17] Y. Wei, C. Cao, R.C. Jin, C.A. Mirkin, Nanoparticles with Raman spectroscopic fingerprints for DNA and RNA detection, Science 297 (2002) 1536–1540. [18] X.F. Ding, L.T. Kong, J. Wang, F. Fang, D.D. Li, J.H. Liu, Highly sensitive SERS detection of Hg2+ ions in aqueous media using gold nanoparticles/graphene heterojunctions, ACS Appl. Mater. Interfaces 5 (2013) 7072–7078. [19] T. Kang, D.S.M. Yoo, D.I. Yoon, S. Lee, J. Choo, S.Y. Lee, P.B. Kim, Au nanowireon-film SERRS sensor for ultrasensitive Hg2+ detection, Chem. Eur. J. 17 (2011) 2211–2214. [20] L.H. Wang, A.G. Shen, X.C. Li, Y. Zeng, X.D. Zhou, Inclusion of guest materials in aqueous coordination network shells spontaneously generated by reacting 2,5dimercapto-1,3,4-thiadiazole with nanoscale metallic silver, RSC Adv. 4 (2014) 34294–34302. [21] A.G. Shen, L.F. Chen, W. Xie, J.C. Hu, A. Zeng, R. Richards, J.M. Hu, Triplex AuAg-C core-shell nanoparticles as a novel Raman Label, Adv. Funct. Mater. 20 (2010) 969–975. [22] V.T. Joy, T.K.K. Srinivasan, Ft-SERS studies on 1,3-thiazolidine-2-thione, 2,5dimercapto-1,3,4-thiadiazole and 2-thiouracil adsorbed on chemically deposited silver films, J. Raman Spectrosc. 32 (2001) 785–793. [23] J.M. Pope, T. Sato, E. Shoji, D.A. Buttry, T. Sotomura, N. Oyama, Spectroscopic identification of 2,5-dimercapto−1,3,4-thiadiazole and its lithium salt and dimer forms, J. Power Sources 68 (1997) 739–742. [24] K.A. Bosnick, J. Jiang, E. Brus, Fluctuations and local symmetry in single-molecule rhodamine 6G Raman scattering on silver nanocrystal aggregates, J. Phys. Chem. B 106 (2002) 8096–8099. [25] S. Lee, S. Kim, J. Choo, S.Y. Shin, Y.H. Lee, H.Y. Choi, S. Ha, K. Kang, C.H. Oh, Biological imaging of HEK293 cells expressing PLCγ1 using surface-enhanced Raman microscopy, Anal. Chem. 79 (2007) 916–922. [26] Q.F. Lü, M.R. Huang, X.G. Li, Synthesis and heavy-metal-ion sorption of pure sulfophenylenediamine copolymer nanoparticles with intrinsic conductivity and stability, Chem. Eur. J. 13 (2007) 6009–6018. [27] X.F. Wang, Y.H. Shen, A.J. Xie, S.K. Li, Y. Cai, Y. Wang, H.Y. Shu, Assembly of dandelion-like Au/PANI nanocomposites and their application as SERS nanosensors, Biosens. Bioelectron. 26 (2011) 3063–3067. [28] E.M. Nolan, S.J. Lippard, A “turn-on” fluorescent sensor for the selective detection of mercuric ion in aqueous media, J. Am. Chem. Soc. 125 (2003) 14270–14271. [29] G.K. Darbha, A.K. Singh, U.S. Rai, E. Yu, H.T. Yu, P.C. Ray, Selective detection of mercury (II) ion using nonlinear optical properties of gold nanoparticles, J. Am. Chem. Soc. 130 (2008) 8038–8043.
Fig. 5. Selectivity of the SERS nanosensors for Hg2+ over other cations, where R and R0 are the Raman intensity of band at 1360 cm−1 in the presence and absence of cations. The concentration of Hg2+ is 1 μM while the competing cations are set as 0.1 mM. Especially, the concentration of Na+ and K+ are 1 mM.
could be formed [26,27]. As reported, when compared with other common ions, Hg2+ possesses larger ionic radius, so the polarization and deformation happen more easily when interacting with nitrogen atom. So the covalent bond of between nitrogen atoms and Hg2+ is more stable [27]. Also, according to literature, the stability constant of mercury–N complex is much more higher than that of other interfering metal ions [28,29]. 3.6. Practical application Notably, real sample detection is carried out by spiking mercury ions into environmental water. The lake water used in this experiment is collected from East Lake, Wuhan, Hubei province of China. After filtered through 0.22 µm Millipore filters to remove any particulate suspension, the river water was treated with different concentration of mercury ions. The as-prepared real samples were then determined by both Raman spectrometer and AAS, in order to compare our developed SERS probe with conventional standard AAS method. The recovery from 95.5% to 108.4% was obtained (Table S1). As a result, the alkynecoded SERS test kit provides better correlation with spiked statistics. 4. Conclusion In summary, we have successfully demonstrated for the first time a label-free, highly sensitive and selective SERS assay for Hg2+ recognition in 10 pM level in aqueous solution. We convinced that DMcT binding on the surface of Au@Ag NPs as disulfide salts and nitrogen atoms were left for coordination with Hg2+. Utilizing the original signal of DMcT as the internal label for quantification, this platform can be successfully applied for label-free Hg2+ detection with a LOD as low as 10 pM. Along with the excellent selectivity, this strategy finally proved to be practical by highly sensitive and selective Hg2+ probing during real environmental monitoring. Acknowledgement We gratefully acknowledge the financial support from National Natural Science Foundation of China (Nos. 21475100, 81471696, 41273093 and 21175101). Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.talanta.2016.09.062.
379
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
A label-free SERS probe for highly sensitive detection of Hg2+ based on functionalized Au@Ag nanoparticles
crossmark
⁎
Yi Zenga,1, Lihua Wanga,b,1, Lingwen Zengb, Aiguo Shena, , Jiming Hua a Key Laboratory of Analytical Chemistry for Biology and Medicine (Ministry of Education), College of Chemistry and Molecular Sciences, Wuhan University, Wuhan 430072, PR China b Institute of Environmental and Food Safety, Wuhan Academy of Agricultural Science and Technology, Wuhan 430207, PR China
A R T I C L E I N F O
A BS T RAC T
Keywords: Label-free DMcT Hg2+ SERS Hg–N bond
Unbiodegradable toxic Hg2+ accumulates in ecological systems and results in contaminated food chain, exposing us to high level pollution and healthy risk. Here, monolayer 2,5-Dimercapto-1,3,4-thiadiazole (DMcT) functionalized Au@Ag nanoparticles (NPs) were controllably constructed as a label-free SERS probe. Based on them, we can perform rapid and easy Hg2+ identification with high sensitivity and selectivity over competing analytes. Remarkably, DMcT acted as both interior label and target acquisition. Since DMcT showed intrinsic Raman signal when attached to substrates surface, it can be employed for quantification instead of extra conventional signal label. Also, we’ve demonstrated that DMcT coordinated on the surface of Au@Ag NPs as bidentate ligand with both the thiocarbonyl sulfur atoms while the nitrogen atoms on the different sides of the molecule were devoted to Hg2+ recognition. Owing to the strong coordination between Hg2+ and nitrogen atoms, as low as 10 pM Hg2+ can be detected. The probe responded a good linear relationship ranging from 0.05 to 100 nM and the limit of detection is ~3 orders of magnitude lower than the United States Environmental Protection Agency (USEPA)-defined limit (10 nM) in drinkable water (EPA: Washington, DC, 2001). Furthermore, our suggested platform is highly effective to perform real samples detection. Utilizing the environmental water from East Lake, Wuhan, it resulted in better accuracy over the conventional standard method.
1. Introduction Mercury and most of its compounds are extremely toxic but are still continuously releasing into the ecosystem. Natural sources such as volcanic eruption are responsible, while fossil fuel burning, waste incineration, and gold mining caused by human also generated plenty mercury emissions. It permeates into soil, flows out into oceans, ultimately accumulates in human body via the food chain [1–4]. Actually, due to the poisonous unbiodegradable property, mercury's long-lasting toxic effects has already caused problems for endangered marine life. As for human beings, mercury exposure severely affects our endothelial and cardiovascular function, even at low doses. It also significantly induces changes in the central nervous system [2,3,5]. Thus, mercury poisoning is becoming a huge concern. Driven by the need, it is essential to develop convenient and simple platform for probing mercury distribution in both artificial and environmental sources, of course high sensitivity and selectivity are in demand. Specifically, the choice of real objects for analysis of mercury content
⁎
1
always goes to drinkable water along with where it come from in natural environment, for example, the river or lake water utilized for daily use. Till now, assays for mercury ions mainly lay in two fields. Traditional one is the instrumental analysis such as atomic absorption spectrometer (AAS) [6]. Besides, sensing platforms are widely used, such as resonance scattering spectrophotometry [7,8], electrochemistry [9,10], fluorimetry [11–13], and colorimetry [14,15]. Among all the above, surface enhanced Raman scattering (SERS) has been given considerable attention by the public. SERS possesses high sensitivity and narrow band, while providing spectroscopic fingerprints information, in addition it's anti-photobleaching and also nondestructive to sample [16,17]. Plenty of sensor platforms have been developed based on SERS. Ding et al. [18] synthesized gold nanoparticles (NPs)/ reduced graphene oxide hetero-junctions via a seed-assisted growth process, and utilized it for trace analysis of Hg2+ via thymine-Hg2+thymine coordination. Kang et al. [19] reported a single nanowire-onfilm SERRS sensor for Hg2+ detection based on structure-switching
Corresponding author. E-mail address: [email protected] (A. Shen). These authors contributed equally to the work.
http://dx.doi.org/10.1016/j.talanta.2016.09.062 Received 20 May 2016; Received in revised form 18 September 2016; Accepted 27 September 2016 Available online 28 September 2016 0039-9140/ © 2016 Elsevier B.V. All rights reserved.
Talanta 162 (2017) 374–379
Y. Zeng et al.
Scheme 1. Schematic illustration of the fabrication of Au@Ag-DMcT SERS probe and a sensing protocol for Hg2+.
double stranded DNAs. Binding of Hg2+ induces conformational changes of the dsDNAs and let a Raman reporter get close to the selective single nanowire-on-film structure, thereby turning on SERRS signal. Certainly they’ve made progress to some extent, but considering the probe construction part in those sensors, the procedure usually includes Raman active molecule labelling or DNA functionalization, which goes against convenience, simplicity and widely probe application. Because to some degree it makes detection more complicated and raises the whole cost, resulting in obstacles on the way of practical and extensive use. Hence, developing simple and rapid sensor for Hg2+ detection with low cost is of great significance. Herein, we presented a highly sensitive and selective monolayer 2,5-Dimercapto-1,3,4-thiadiazole (DMcT) functionalized Au@Ag nanoparticles (Au@Ag-DMcT) label-free probe for Hg2+ detection ( Scheme 1). Utilizing Au@Ag-DMcT as the SERS substrate, the platform mainly based on DMcT's original Raman signals and its coordination with metal ions. Facile quantification of Hg2+ in homogeneous solution was achieved by linking the amount of Hg2+ and ‘hot spot’ effect of SERS-based NPs. The incremental Raman intensity of DMcT was characteristically related to Hg2+ concentration over a range of 0.05– 100 nM. Actually, there has been several reports on Hg2+ detection based on DMcT capped noble metal NPs [11], but the reaction mode between DMcT and NPs are often overlooked. Generally, DMcT can coordinate as bidentate ligand and has mainly four donor sites: two nitrogen atoms and two sulfur atoms. In particular, according to our previous work [20] and via Raman spectra analysis, here we revealed that DMcT coordinated on the surface of Au@Ag NPs as disulfide salts, and the nitrogen atoms were left for Hg2+ recognition. Owing to the strong coordination between Hg2+ and nitrogen atoms, we can perform ultra-sensitive detection of Hg2+ low to 10 pM.
All other chemicals were of analytical grade or better quality, and used as received. 2.2. Apparatus The morphology and microstructure of the different NPs were characterized by using high resolution transmission electron microscope (TEM, JEM-2100, JEOL, Japan) operating at a 200 kV accelerating voltage. UV–vis adsorption spectra were recorded on a UV–vis spectrophotometer (UV-2550, Shimadzu, Japan) at room temperature using a 600 μL black body quartz cuvette with 1 cm path length. The Raman spectra were collected by Horbia Jobin-Yvon HR-800 Raman microspectrometer (System HR800, Horbia Jobin-Yvon, Villeneuve d′Ascq, France). 2.3. Synthesis of Au@Ag NPs A proper size of Au@Ag core–shell NPs as SERS substrates were synthesized according to our previous work [21]. First, 50 mL of 0.01% (w/w) HAuCl4 was reduced by sodium citrate solution (1%, w/w) at 100°C under vigorous magnetic stirring for 20 min. The as-prepared Au NPs were cooled down to room temperature and served as seed particles. Then, the Au@Ag NPs were synthesized as follows: aqueous solutions of sodium citrate (38.8 mM, 200 μL), ascorbic acid (0.1 M, 50 μL) and seeds (5 mL) were combined at room temperature; to this mixture, AgNO3 (0.01 M, 100 μL) was added drop wise immediately under continuous magnetic stirring. The solution colour immediately changed from burgundy red to orange within 2 min, indicating the formation of Au@Ag NPs. To make the reaction fully completed, the solution was stirred for 15 min and stored in 4°C.
2. Experimental section
2.4. Synthesis of Au@Ag-DMcT and detection of Hg2+
2.1. Reagents
Different volumes of freshly prepared 0.5 mg mL−1 DMcT aqueous solution (ultrasonically dispersed in phosphate buffer (PB) suspension (0.1 M, pH 7.4)) were added to the colloid solution of Au@Ag NPs with slow stirring, and the suspensions were allowed to incubate for several hours. The obtained NPs were then rinsed by centrifugation and resuspension with deionized water for three times. For Hg2+ detection, different concentration of Hg(NO3)2 solution were added into 500 μL probe solution, mixed for 1–2 min, and then Raman signal were
Tetrachloroauric(III) acid and sodium citrate were purchased from Sigma (USA). Silver nitrate was obtained from Beijing Chemical Reagents Company (Beijing, China). Ascorbic acid was supplied by Sinopbarm Chemical Reagent Co., Ltd. 2,5-Dimercapto-1,3,4-thiadiazole (DMcT, ≥ 98%) was a product of Alfa Aesar. Mercuric nitrate was purchased from Tongren Chemical Reagent Factory (Guizhou, China). 375
Talanta 162 (2017) 374–379
Y. Zeng et al.
Fig. 1. (A) The absorption spectra of Au@Ag NPs (solid line) and Au@Ag-DMcT (dotted line). The inset shows TEM image of Au@Ag-DMcT; (B) SERS spectra of Au@Ag NPs (solid line) and Au@Ag-DMcT (dotted line); (C) normalized SERS spectra of the disulfide salts (dotted line) and monosulfide salts (solid line) of DMcT; (D) SERS spectra of DMcT in solid state (solid line) and Au@Ag-DMcT after drying in the air (dotted line).
coordination [11]. Actually, DMcT has mainly four donor sites and can coordinate as bidentate ligand with (a) both the nitrogen atoms, (b) both the thiocarbonyl sulfur atoms, or (c) one nitrogen atom and one sulfur atom on either the same side or different sides of the molecule [20]. To figure out the reaction mode between DMcT and Au@Ag NPs in this case and further reveal the detection rationale, more complemented experiments were carried out. Initially, we found that monosulfide salts of DMcT possess 1360 and 1410 cm−1 belonging to C=N stretching vibration, while disulfide salts only show 1360 cm−1 (Fig. 1C), which is consistent with the previous work [23]. Moreover, a droplet concentrated probe solution was settled on tin foil to dry in air and the Raman spectra were further collected. As is shown Fig. 1D, the stretching vibration peak of S–H of DMcT at 2475 cm−1 was initially observed but disappeared after incubation with Au@Ag NPs; meantime, a S–Ag bond at ca. 220 cm−1 appeared. These data suggest that the S–H bond of DMcT was broken and the S–Ag bond was formed during the reaction between Au@Ag NPs and DMcT. Combination with the SERS spectra of Au@Ag-DMcT in Fig. 1B (only 1360 cm−1 peak shows in the C=N stretching vibration), we can safely infer that two S–H bond of DMcT have coordinated with Au@Ag NPs in our experiment.
collected. 3. Results and discussion 3.1. Characterization of Au@Ag-DMcT Spontaneous reaction between DMcT and Au@Ag NPs in homogeneous water solution was investigated. The UV–vis spectra of Au@ Ag NPs before and after incubation with DMcT were shown in Fig. 1A. There are two peaks at ca. 385 and 508 nm in the spectrum of the original Au@Ag NPs. These two peaks has been attributed to the surface Plasmon resonance (SPR) absorption of the Ag shell and Au core of Au@Ag NPs, respectively [21]. Related to our previous work [20], after incubation of the Au@Ag NPs with DMcT, the Ag peak declined because Ag has reacted with DMcT, meanwhile a little bit red shift can be observed for the Au absorption, mainly due to Au@Ag NPs surface condition change from Ag-DMcT reaction. The TEM characterization (the illustration in Fig. 1A) also implied the particles are welldispersed and uniform. To further confirm the successful functionalization of DMcT on the surface of Au@Ag NPs, we represented Raman spectra in Fig. 1B. It can be seen that the pure Au@Ag NPs showed no specific Raman signal, while two characteristic peaks at 664 and 1360 cm−1 appeared in the spectra of Au@Ag-DMcT, which should be assigned to the symmetric vibrations of C–S and C=N of DMcT anions, respectively [22]. As we know, previous research on the application of DMcT capped noble metal NPs on Hg2+ detection almost neglected the reaction mode of DMcT on NPs, which just oversimplified and assume it as sulfur
3.2. Detection rationale As we have confirmed DMcT existed as disulfide salts in Au@AgDMcT, donor sites left for metal ions must be the nitrogen atoms. The added Hg2+ coordinated with nitrogen atoms and induced aggregation of Au@Ag-DMcT, which consequently resulted in the enhancement of 376
Talanta 162 (2017) 374–379
Y. Zeng et al.
Fig. 2. (A) SERS spectra of the Au@Ag-DMcT based nanosensors in the absence (solid line) and presence (dotted line) of Hg2+(20 nM); (B) TEM image of Au@Ag-DMcT after addition of 20 nM Hg2+.
DMcT's SERS signals. Detection of Hg2+ in homogeneous solution was achieved by linking the amount of Hg2+ and ‘hot spot’ effect of SERSbased NPs. For feasibility, we compared Au@Ag-DMcT's Raman spectra before and after additive of 20 nM Hg2+. As Fig. 2A shown, DMcT responded obviously to Hg2+, typical DMcT peaks at 664 and 1360 cm−1 apparently increased after Hg2+ addition. The corresponding TEM images (Fig. 2B) are also demonstrated, indicating particles aggregation induced by Hg2+-DMcT coordination.
thick, SERS intensity became higher and higher till 200 μL. However, continuously increasing AgNO3 amount over that value resulted in turbid solution, which reflected that Au@Ag NPs turned to be unstable in this condition and easily aggregated or even got precipitated. Consequently, R0 increased and R-R0 declined in contrast, ending up with 200 μL as the most optimized. According to our previous research [20], when DMcT reacted with Au@Ag NPs, it firstly bound to the surface of Au@Ag NPs in the form of disulfide salts. If there are excessive DMcT in the system, the disulfide salts will further react with DMcT to form the monosulfide salts, which will interact with each other based on the bridge of Ag+ ions to form a heavy layer of infinite coordination polymers. Thus we further tested different incubation time of DMcT and Au@Ag NPs. As Fig. 3C shown, during 1–5 h, R-R0 reached maximum value at about 3 h. It indicated that DMcT disulfide salts had totally occupied the surface of Au@Ag NPs by monolayer adsorption in this period. With incubation time further increasing, polymer layer will get thicker and thicker, which will hinder aggregation of probes, resulting in R-R0 reduction. For buffer condition influence, we tested probe performance in different buffer pH. As shown in Fig. 3D, the developed SERS probe preferred mild acid and lower pH leaded to better performance. However, probe itself was unstable and easily get aggregated when buffer pH was 3.5. So we chose pH 4.5 as the best.
3.3. Optimization of experimental parameters For best probe performance, we optimized necessary experimental parameters. The amount of DMcT, composition ratio of Au@Ag NPs, incubation time of DMcT with Au@Ag NPs, and buffer pH were all taken into account. First of all, the amount of DMcT utilized belonged to key point as it largely affected the probe structure, subsequently performance. As Fig. 3A shown, upon the increase of DMcT volume, the corresponding Raman signal difference R-R0 (R0 and R represent Raman intensity before and after Hg2+ added, respectively) declined apparently. Especially, when the volume exceeded 10 μL, R-R0 barely changed then but achieved a steady low value instead. This phenomenon was reasonable. Since more DMcT were introduced to the system, the thicker the DMcT-Ag infinite coordination polymers shell would be [20]. This provided stronger protection to Au@Ag NPs that in consequence would not easily response to Hg2+ addition, so that the enhancement became too weak to be observed. As the thickness reached a certain extent, no more aggregation would happen with Hg2+ adding, and the Raman signal seemed to be steady. Moreover, we found that when the volume of DMcT was less than 1.25 μL, the structure formed was not stable. Thus we chose 1.25 μL DMcT as the best. As known to all, gold colloidal NPs have advantages including longterm stability, controllable size distribution, and high homogeneity. Conversely, the Raman enhancement factor of the gold NPs is ~100– 1000 times lower than that of silver. However the silver NPs also have got fatal flaws—very unstable, due to the limited solubility and aggregation properties [24,25]. Here we combined their both advantages and selected Au@Ag NPs as SERS substrates for satisfying stability and Raman enhancement owing to gold NPs core and silver shell, respectively [21]. Obviously, best molar ratio Ag/Au should be explored, which is carried out by controlling the diameter of Au NPs and changing the amount of AgNO3 during Au@Ag NPs synthesis. Au NPs with a diameter of 30 nm were employed as seeds owing to the stability and relatively high SERS enhancement. As shown in Fig. 3B, with the AgNO3 amount gradually increasing and silver shell getting
3.4. Quantitatively SERS detection of Hg2+ To explore the potential of Au@Ag-DMcT for quantitatively SERS detection of Hg2+, Raman spectroscopic homogeneous assay was performed in aqueous buffer solution under optimized condition. Enhanced SERS response of Au@Ag-DMcT was observed upon the addition of Hg2+ solution into SERS-based NPs colloids. Fig. 4A showed the corresponding Hg2+ concentration-dependent SERS spectra of Au@Ag-DMcT. With Hg2+ concentration increasing, since coordination between Hg2+ and DMcT caused probe aggregation, typical peak of DMcT were largely enhanced as highly recognizable SERS signals at 664 and 1360 cm−1 could be rapidly detected. Here, we picked out 1360 cm−1 for quantification. The trend for enhanced SERS intensity at 1360 cm−1 varying with Hg2+ concentration were shown in Fig. 4B. The SERS intensity displayed an excellent linear correlation with Hg2+ concentration ranging from 0.05 to 100 nM (R2 = 0.986). It is worth noting that Hg2+ as low as 10 pM can be detected (Supplementary data, Fig. S1), which is ~3 orders of magnitude lower than the United States Environmental Protection Agency (USEPA)defined limit (10 nM) in drinkable water (EPA: Washington, DC, 2001). Meanwhile, it was also remarkably superior to majority of 377
Talanta 162 (2017) 374–379
Y. Zeng et al.
Fig. 3. Optimization of the volume of 0.5 mg/mL−1 DMcT (A), the volume of 0.01 M AgNO3 (B), the incubation time of DMcT and Au@Ag NPs (C), and the value of pH (D). The error bars represent the standard deviations based on at least three independent measurements.
3.5. Selectivity of the SERS sensor for Hg2+
reported probes. For instance, there are electrochemical sensor for Hg2+ using amplification based on thymine−Hg2+−thymine base pairs and gold nanoparticles (0.5 nM) [10], SERS technique for Hg2+ detection utilizing interaction between silver nanoparticles and mercury ions (90.0 pM) [5], SERS sensor of Hg2+ by gold nanoparticles/ graphene hetero-junctions (0.1 nM) [18], and Hg2+ fluorescent probes using mesoporous cerium phosphonate nanostructured hybrid spheres (16 nM) [12], and so on. Moreover, the reproducibility of the SERS spectra was tested by acquiring several spectra from 8 times random test from 4 different batches of SERS NPs with 20 nM Hg2+, which are presented in Fig. S2. The percent coefficient of variation (CV%) is calculated as around 6.8, which indicates that the suggested probe possesses excellent reproducibility.
To evaluate the selectivity of this SERS platform, a series of control experiments between Hg2+ and other coexisting interfering metal ions were performed (Fig. 5). The potential competing substances were separately detected under the optimized conditions, including common ions in natural experiment such as Pb2+, Cd2+, Fe2+, Al3+, Fe3+, Cu2+, Na+, K+, Mn2+, Ni2+, Cr3+, Zn2+, Mg2+, Ca2+, and Co2+. The concentration were set as 1 μM Hg2+ and 0.1 mM for the others (Especially, the concentration of Na+ and K+ were 1 mM). It was clear that interference ions barely produced any SERS signal enhancement. The excellent selectivity of this platform mainly owe to coordination mechanisms between Hg2+ and DMcT. Actually, the nitrogen atom plays a role as main donor site here. Since both the two nitrogen atoms with lone pair of electrons could coordinate with Hg2+, quadridentate coordination
Fig. 4. (A) SERS spectra of Au@Ag-DMcT with different Hg2+ concentration; (B) SERS intensity change at 1360 cm−1 against concentration of Hg2+. The inset shows the linear plot of the relative Raman intensity against different Hg2+ concentrations. The error bars represent the standard deviations based on three independent measurements at least.
378
Talanta 162 (2017) 374–379
Y. Zeng et al.
References [1] X.J. Xue, F. Wang, X.G. Liu, One-step, room temperature, colorimetric detection of mercury (Hg2+) using DNA/nanoparticle conjugates, J. Am. Chem. Soc. 130 (2008) 3244–3245. [2] E.M. Nolan, S.J. Lippard, Tools and tactics for the optical detection of mercuric ion, Chem. Rev. 108 (2008) 3443–3480. [3] M.M. Veiga, J.A. Meech, N. Oñate, Mercury pollution from deforestation, Nature 368 (1994) 816–817. [4] C. Díez-Gil, R. Martínez, I. Ratera, T. Hirsh, A. Espinosa, A. Tárraga, P. Molina, O.S. Wolfbeis, J. Veciana, Selective picomolar detection of mercury(II) using optical sensors, Chem. Commun. 47 (2011) 1842–1844. [5] W. Ren, C.Z. Zhu, E.K. Wang, Enhanced sensitivity of a direct SERS technique for Hg2+ detection based on the investigation of the interaction between silver nanoparticles and mercury ions, Nanoscale 4 (2012) 5902–5909. [6] A.Q. Shah, T.G. Kazi, J.A. Baig, H.I. Afridi, M.B. Arain, Simultaneously determination of methyl and inorganic mercury in fish species by cold vapour generation atomic absorption spectrometry, Food Chem. 134 (2012) 2345–2349. [7] Y. Chen, L.H. Wu, Y.H. Chen, N. Bi, X. Zheng, H.B. Qi, M.H. Qin, X. Liao, H.Q. Zhang, Y. Tian, Determination of mercury(II) by surface-enhanced Raman scattering spectroscopy based on thiol-functionalized silver nanoparticles, Microchim. Acta 177 (2012) 341–348. [8] D. Han, S.Y. Lim, B.J. Kim, L.L. Piao, T.D. Chung, Mercury(II) detection by SERS based on a single gold microshell, Chem. Commun. 46 (2010) 5587–5589. [9] Y. Zhang, G.M. Zeng, L. Tang, J. Chen, Y. Zhu, X.X. He, Y. He, Electrochemical sensor based on electrodeposited graphene-Au modified electrode and nanoAu carrier amplified signal strategy for attomolar mercury detection, Anal. Chem. 87 (2015) 989–996. [10] Z.Q. Zhu, Y.Y. Su, J. Li, D. Li, J. Zhang, S.P. Song, Y. Zhao, G.X. Li, C.H. Fan, Highly sensitive electrochemical sensor for mercury(II) ions by using a mercuryspecific oligonucleotide probe and gold nanoparticle-based amplification, Anal. Chem. 81 (2009) 7660–7666. [11] N. Vasimalai, S. John, Mercaptothiadiazole capped gold nanoparticles as fluorophore for the determination of nanomolar mercury(II) in aqueous solution in the presence of 50000-fold major interferents, Analyst 137 (2012) 3349–3354. [12] Y.P. Zhu, T.Y. Ma, T.Z. Ren, Z.Y. Yuan, Mesoporous cerium phosphonate nanostructured hybrid spheres as label-free Hg2+ fluorescent probes, ACS Appl. Mater. Interfaces 6 (2014) 16344–16351. [13] D.B. Liu, S.J. Wang, M. Swierczewska, X.L. Huang, A.A. Bhirde, J.S. Sun, Z. Wang, M. Yang, X.Y. Jiang, X.Y. Chen, Highly robust, recyclable displacement assay for mercuric ions in aqueous solutions and living cells, ACS Nano 6 (2012) 10999–11008. [14] J.J. Du, B.W. Zhu, X.D. Chen, Urine for plasmonic nanoparticle-based colorimetric detection of mercury ion, Small 9 (2013) 4104–4111. [15] G. Sener, L. Uzun, A. Denizli, Lysine-promoted colorimetric response of gold nanoparticles: a simple assay for ultrasensitive mercury(II) detection, Anal. Chem. 86 (2014) 514–520. [16] J. Ando, K. Fujita, N.I. Smith, S. Kawata, Dynamic SERS imaging of cellular transport pathways with endocytosed gold nanoparticles, Nano Lett. 11 (2011) 5344–5348. [17] Y. Wei, C. Cao, R.C. Jin, C.A. Mirkin, Nanoparticles with Raman spectroscopic fingerprints for DNA and RNA detection, Science 297 (2002) 1536–1540. [18] X.F. Ding, L.T. Kong, J. Wang, F. Fang, D.D. Li, J.H. Liu, Highly sensitive SERS detection of Hg2+ ions in aqueous media using gold nanoparticles/graphene heterojunctions, ACS Appl. Mater. Interfaces 5 (2013) 7072–7078. [19] T. Kang, D.S.M. Yoo, D.I. Yoon, S. Lee, J. Choo, S.Y. Lee, P.B. Kim, Au nanowireon-film SERRS sensor for ultrasensitive Hg2+ detection, Chem. Eur. J. 17 (2011) 2211–2214. [20] L.H. Wang, A.G. Shen, X.C. Li, Y. Zeng, X.D. Zhou, Inclusion of guest materials in aqueous coordination network shells spontaneously generated by reacting 2,5dimercapto-1,3,4-thiadiazole with nanoscale metallic silver, RSC Adv. 4 (2014) 34294–34302. [21] A.G. Shen, L.F. Chen, W. Xie, J.C. Hu, A. Zeng, R. Richards, J.M. Hu, Triplex AuAg-C core-shell nanoparticles as a novel Raman Label, Adv. Funct. Mater. 20 (2010) 969–975. [22] V.T. Joy, T.K.K. Srinivasan, Ft-SERS studies on 1,3-thiazolidine-2-thione, 2,5dimercapto-1,3,4-thiadiazole and 2-thiouracil adsorbed on chemically deposited silver films, J. Raman Spectrosc. 32 (2001) 785–793. [23] J.M. Pope, T. Sato, E. Shoji, D.A. Buttry, T. Sotomura, N. Oyama, Spectroscopic identification of 2,5-dimercapto−1,3,4-thiadiazole and its lithium salt and dimer forms, J. Power Sources 68 (1997) 739–742. [24] K.A. Bosnick, J. Jiang, E. Brus, Fluctuations and local symmetry in single-molecule rhodamine 6G Raman scattering on silver nanocrystal aggregates, J. Phys. Chem. B 106 (2002) 8096–8099. [25] S. Lee, S. Kim, J. Choo, S.Y. Shin, Y.H. Lee, H.Y. Choi, S. Ha, K. Kang, C.H. Oh, Biological imaging of HEK293 cells expressing PLCγ1 using surface-enhanced Raman microscopy, Anal. Chem. 79 (2007) 916–922. [26] Q.F. Lü, M.R. Huang, X.G. Li, Synthesis and heavy-metal-ion sorption of pure sulfophenylenediamine copolymer nanoparticles with intrinsic conductivity and stability, Chem. Eur. J. 13 (2007) 6009–6018. [27] X.F. Wang, Y.H. Shen, A.J. Xie, S.K. Li, Y. Cai, Y. Wang, H.Y. Shu, Assembly of dandelion-like Au/PANI nanocomposites and their application as SERS nanosensors, Biosens. Bioelectron. 26 (2011) 3063–3067. [28] E.M. Nolan, S.J. Lippard, A “turn-on” fluorescent sensor for the selective detection of mercuric ion in aqueous media, J. Am. Chem. Soc. 125 (2003) 14270–14271. [29] G.K. Darbha, A.K. Singh, U.S. Rai, E. Yu, H.T. Yu, P.C. Ray, Selective detection of mercury (II) ion using nonlinear optical properties of gold nanoparticles, J. Am. Chem. Soc. 130 (2008) 8038–8043.
Fig. 5. Selectivity of the SERS nanosensors for Hg2+ over other cations, where R and R0 are the Raman intensity of band at 1360 cm−1 in the presence and absence of cations. The concentration of Hg2+ is 1 μM while the competing cations are set as 0.1 mM. Especially, the concentration of Na+ and K+ are 1 mM.
could be formed [26,27]. As reported, when compared with other common ions, Hg2+ possesses larger ionic radius, so the polarization and deformation happen more easily when interacting with nitrogen atom. So the covalent bond of between nitrogen atoms and Hg2+ is more stable [27]. Also, according to literature, the stability constant of mercury–N complex is much more higher than that of other interfering metal ions [28,29]. 3.6. Practical application Notably, real sample detection is carried out by spiking mercury ions into environmental water. The lake water used in this experiment is collected from East Lake, Wuhan, Hubei province of China. After filtered through 0.22 µm Millipore filters to remove any particulate suspension, the river water was treated with different concentration of mercury ions. The as-prepared real samples were then determined by both Raman spectrometer and AAS, in order to compare our developed SERS probe with conventional standard AAS method. The recovery from 95.5% to 108.4% was obtained (Table S1). As a result, the alkynecoded SERS test kit provides better correlation with spiked statistics. 4. Conclusion In summary, we have successfully demonstrated for the first time a label-free, highly sensitive and selective SERS assay for Hg2+ recognition in 10 pM level in aqueous solution. We convinced that DMcT binding on the surface of Au@Ag NPs as disulfide salts and nitrogen atoms were left for coordination with Hg2+. Utilizing the original signal of DMcT as the internal label for quantification, this platform can be successfully applied for label-free Hg2+ detection with a LOD as low as 10 pM. Along with the excellent selectivity, this strategy finally proved to be practical by highly sensitive and selective Hg2+ probing during real environmental monitoring. Acknowledgement We gratefully acknowledge the financial support from National Natural Science Foundation of China (Nos. 21475100, 81471696, 41273093 and 21175101). Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.talanta.2016.09.062.
379