Etching of unmodified Au@Ag nanorods: a tunable colorimetric visualization for the rapid and high selective detection of Hg²⁺

Abstract

A simple and cost-effective colorimetric approach based on unmodified Au@Ag nanorods (Au@Ag NRs) was developed for Hg²⁺ detection. Unmodified Au@Ag NRs with different Ag nanoshell thicknesses served for the signal readout because the Ag coating-induced a blueshift and enhancement of the longitudinal plasmon of Au NRs, resulting in abundant and tunable optical absorptions in the visible region. The etching sensing mechanism was revealed to be related to the redox reaction between Hg²⁺ and the Ag nanoshell of the Au@Ag NRs. The Ag nanoshell of the Au@Ag NRs was gradually etched from the ends as the Hg²⁺ concentration was gradually increased, and shoulder shapes were formed, and then disappeared. The Hg²⁺ concentration-dependent color of Au@Ag NRs with a thick Ag nanoshell thickness changed from brownish-red to light-red, light-violet, and finally to colorless. The limit of detection (LOD) and the detection range of Hg²⁺ became tunable as the Ag nanoshell thickness increased, and the lowest LOD was 10 nM. A dip located between two strong absorption peaks was observed when Au@Ag NRs with a thick Ag nanoshell thickness were used, and the change in this dip provided a new sensor parameter for Hg²⁺ detection on the basis of absorption spectra. The proposed method also showed a high selectivity toward Hg²⁺ over other metal ions. The Au@Ag NR detection system could detect even low Hg²⁺ concentrations in drinking water.

---

Introduction

Substantial sensing techniques based on noble metal nanoparticles (NPs) have attracted considerable attention because of their unique optical properties. Au NP-based optical sensing methods, such as colorimetry, light-scattering, and fluorescence, have been widely designed and applied for contaminant detection.^1,2 Colorimetric detection induced by localized surface plasmon resonance (LSPR) is commonly used because of its simplicity, convenience, and visibility to naked eye.^2–4 The color change of metal NP solutions associated with LSPR relies on the NPs' size, shape, interparticle distance, and local dielectric environment.^1–3,5 Given their transverse and longitudinal LSPR adsorption modes induced by the oscillation of conduction electrons along two directions, Au nanorods (Au NRs) enrich the color change of Au nanoprobes.^5–7 Recent studies have extensively applied Au NRs in the colorimetric sensing of metal ions, DNA, proteins, and small molecules based on the aggregation of Au NRs.^1–8 Compared with Au NPs, Ag NPs produce a much stronger and sharper plasmon resonance.^9,10 Moreover, the LSPR absorption band of Ag NPs with a well-controlled size changes more easily than Au NPs when exposed to special targets. The strong shape-dependent optical properties of Ag NPs allow the rapid, sensitive, and visualized detection of targets with a minimal consumption of materials.^4,11,12 Various shaped Ag NPs, such as nanoprisms, spherical, and nanoclusters, have been considered for the colorimetric detection of different target molecules on the basis of a morphology transition.^9–12 A Ag nanoprisms-based sensor has been applied to detect Hg²⁺ on the basis of its morphological transition from a nanoprism to a sphere after Hg²⁺ etching.¹² However, the precise controlling of Ag NPs' morphology remains a challenge.

Hg²⁺, a highly biologically toxic and ubiquitous heavy metal ion, is a stable inorganic form of Hg in the environment and in organisms.¹³ The traditional technologies used for Hg²⁺ analysis include instrumental and sensor methods. Instrumental analysis methods, such as atomic absorption spectroscopy, inductively coupled plasma mass spectrometry, and selective cold vapor atomic fluorescence spectrometry, are more precise than other methods, but they require expensive equipment, and involve a high operational cost, and laborious procedures.^14,15 Numerous remarkable sensors based on organic molecules, polymeric materials, biomaterials, and semiconductor nanocrystals have been recently developed for Hg²⁺ detection using optical and electrochemical signals.^{11,12,16–21} However, these sensors generally require complex material preparation or biomolecule conjugation processes.

Colorimetric sensors designed by generating a nanoshell on the surface of the inner core exhibit more brilliant color changes than colorimetric detection systems based on the aggregation of Au NRs and the morphology transition of Ag NPs.^7,22 The Ag coating-induced blueshift and enhancement of the longitudinal LSPR of Au NRs result in abundant and tunable optical absorptions in the visible region, making Au@Ag core–shell NRs (Au@Ag NRs) a feasible candidate for colorimetric sensing.^7,8,19 Au@Ag NRs are easy to prepare and their LSPR properties can be easily controlled by changing the size and shape of the core and the thickness of the shell. Considering the previously reported interaction mechanism between Hg²⁺ and Ag NPs,^11,12 we assumed that Au@Ag NRs would be highly suitable for Hg²⁺ detection because of their controllable monodispersity and aspect ratio, broad plasmon resonance tenability from the near-UV to IR region, and increased sharpness and strength of their longitudinal SPR bands.^7,8,19,20 Although a general method has been developed to tune the dynamic range of biosensors for the detection of heavy metal ions, no such method has been reported for colorimetric nanosensors. The current study thus developed a simple, rapid, sensitive and selective colorimetric assay of Hg²⁺ based on unmodified Au@Ag NRs, wherein Ag nanoshells with different thicknesses were introduced into Au NRs to confer the assay a tunable dynamic range. The etching mechanism of Au@Ag NRs by Hg²⁺ was analyzed through UV-vis spectroscopy, high-resolution transmission electron microscopy (HR-TEM), and energy dispersive X-ray spectroscopy (EDS). The developed method was also successfully used to detect Hg²⁺ in drinking water samples.

Experimental section

Reagents and apparatus

Gold chloride trihydrate (HAuCl₄), cetyltrimethyl ammonium bromide (CTAB), sodium borohydride (NaBH₄), silver nitrate (AgNO₃), and ascorbic acid (AA) were purchased from Sigma-Aldrich. Hydrochloric acid (HCl), sodium hydroxide (NaOH), Pb(NO₃)₂, MnCl₂, AlCl₃, CuCl₂, FeSO₄, Hg(NO)₃, CdCl₂, CoCl₂, CaCl₂, and Mg(NO₃)₂ were purchased from Beijing Chemical Works (China). All the reagents were of analytical grade and were used without further purification. Ultrapure water with a resistivity of 18.2 MΩ, obtained from a Millipore water purification system (Milli-Q, Millipore, USA), was used in all experiments.

All the absorption spectra were obtained on a Shimadzu UV-3150 spectrometer (Japan) or a NanoDrop 2000 (Thermo, USA). HR-TEM images with an accelerating voltage of 200 kV and EDS spectra were obtained using a JEM-2100F transmission electron microscope. The samples were dispersed at an appropriate concentration and cast onto a carbon-coated copper grid. A PB-10 pH meter (Sartorius, Germany) was employed to measure the pH values of all the aqueous solutions.

Preparation of Au@Ag NRs with different Ag nanoshell thicknesses

Au NRs were synthesized using the silver ion-assisted, seed-mediated method as previously described.²³ In brief, the seed solution was initially prepared by mixing 5 mL of 0.1 M CTAB solution with 42 μL of 29 mM HAuCl₄, and 0.3 mL of 10 mM NaBH₄ with vigorous stirring for 10 min. Then, the mixture of 0.4 mL of 10 mM AgNO₃ and 40 mL of 0.1 M CTAB solution was added to with 0.8 mL of 29 mM HAuCl₄. After 0.32 mL of 0.1 M ascorbic acid was added with gentle mixing, 130 μL of the seed solution was added. The mixture was maintained at 30 °C overnight without any further stirring. The as-prepared Au NRs (Fig. S1†) were further purified twice via centrifugation at 9000 rpm for 10 min to remove any excess reagents, and were then used to prepare of Ag@Au NRs. The synthesized Au NRs had a UV-vis absorbance at 830 nm (Fig. S1†).

Au@Ag NRs were prepared as previously described with some modifications.²⁴ In brief, 2 mL of Au NR solution was added to 4 mL of 0.04 M CTAB aqueous solution with vigorous stirring at 28 °C. Up to 130 μL of 0.1 M ascorbic acid, varying amounts of 1 mM AgNO₃, and 240 μL of 0.1 M NaOH were then added sequentially. Au@Ag NRs with various Ag nanoshell thicknesses were prepared by tuning the amount of AgNO₃ added. The color of the solution gradually changed in 2 min, indicating the formation of Au@Ag NRs. The Au@Ag NR solution was purified and then concentrated to 2 mL with deionized water.

Analysis of Hg²⁺ based on unmodified Au@Ag NRs

Various Hg²⁺ concentrations (0–267 μM) were added to 1 mL of Au@Ag NR solution, and a different volume of ultrapure water was added to ensure that the total volume was the same. The resulting solution was stored at room temperature for 5 min. A quantitative analysis was performed, and the absorption spectra of the mixture were obtained under the same conditions. All the experiments were performed in triplicate.

To evaluate the potential matrix effects of environmental samples on Hg²⁺ detection, spiked samples of tap water and commercially available bottled water were tested at concentrations of 0.6, 1.0, and 2 μM. The specificity of the sensor was assessed by evaluating its responses to potentially interfering metal ions, such as Cu²⁺, Mg²⁺, Cd²⁺, Al³⁺, Co²⁺, Mn²⁺, Pb²⁺, Ca²⁺, Zn²⁺, Fe²⁺, Fe³⁺, and Ag⁺, at concentrations up to 1 mM.

Results and discussion

Characterization of Au@Ag NRs with different Ag nanoshell thicknesses

Au NRs with a longitudinal SPR peak at 830 nm and a transverse peak at 576 nm were selected as the core, which had a uniform size distribution and good dispersity (Fig. S1†). For colorimetric detection, the target induced spectral shifts that lead to a visually detectable color change were the primary consideration, and the most sensitive region of color perception for the naked eye was at 500–600 nm.^25,26 Although the longitudinal SPR of the Au NRs was not within this range, Ag coating triggered a blueshift of the longitudinal LSPR of the Au NRs. Furthermore, the plasmonic line width of the Au@Ag NRs was narrower than that of the original Au NRs.²⁷ This phenomenon, known as ‘plasmonic focusing’, makes Au@Ag NRs more suitable for a high-quality colorimetrical sensor.²⁷ The thickness of the Ag nanoshell can be easily controlled by tuning the amounts of silver nitrate and ascorbic acid. In this study, Au@Ag NRs with thin, moderate, and thick Ag nanoshell thickness were prepared, and denoted as Au@Ag-NR1 (∼2.1 nm), Au@Ag-NR2 (∼5.8 nm), and Au@Ag-NR3 (∼9.5 nm), respectively (Fig. 1a). As shown in the insets of Fig. 1b, the color of the Au@Ag NR colloid gradually changed from dark-yellow to green to brownish-red, and the plasmon resonance of the Au@Ag NRs relocated from 687 nm to 572 nm as the Ag nanoshell thickness was increased. Four SPR peaks could be observed and were designated as peaks 1–4 from long to short wavelength. Peak 1 and peak 3, with a remarkable intensity and fine tenability, corresponded to the longitudinal and transverse SPR peak of NRs, respectively. As the Ag nanoshell thickness on the Au NRs increased, peak 1 exhibited a remarkable blueshift, accompanied with enhanced absorbance intensity, whereas peak 3 exhibited a remarkable redshift, accompanied with enhanced absorbance intensity. Peaks 2 and 3 of Au@Ag-NR3 merged, and a deep dip between peaks 1 and 3 appeared. Fig. 1b–d illustrate the change in thickness of the Ag nanoshell in the Au@Ag NRs.

Fig. 1 (a) Absorption spectra of Au@Ag NRs with different Ag nanoshell thicknesses. Typical TEM images of Au@Ag NRs with (b) thin, (c) moderate, and (d) thick Ag nanoshells, respectively. Insets: the corresponding colors of the Au@Ag NRs colloid with different Ag nanoshell thicknesses.

Sensing mechanism of Hg²⁺ detection based on Au@Ag NR etching

Synthesized Au@Ag NRs present excellent optical properties because of the distinct surface plasmon resonance (SPR) absorption band in the visible region, which is beneficial for the colorimetric detection of targets. The proposed etching mechanism of Au@Ag NRs by Hg²⁺ is demonstrated in Fig. 2a. The etching process was initially inspected using UV-vis spectrometry. Au@Ag-NR3 was used as a model to investigate the interaction between Hg²⁺ and Au@Ag NRs. The solution color was altered from brownish-red to light-red after adding the Hg²⁺ solution to the as-prepared Au@Ag-NR3 solution. The color continued to change from light-red to light-violet in response to a further increase in the Hg²⁺ concentration (inset of Fig. 2b–d). The color of the Au@Ag-NR2 colloid was also gradually changed as the Hg²⁺ concentration increased (Fig. S2†). As shown in Fig. 3b, the absorbance intensity decreased and the LSPR peak slightly blueshifted in the presence of low Hg²⁺ concentrations in the solution. The Au@Ag-NR3 solution turned to light-violet when the Hg²⁺ concentration exceeded 87 μM. This results indicated that the absorbance intensity obviously decreased and the LSPR peak slightly redshifted. The spectra shift and intensity decrease of peak 1 can be attributed to the presence of Hg²⁺ and reveal the SPR change of the Au@Ag NRs, which greatly contribute to the color change. Hg²⁺ and Au@Ag NRs could interact in a short time (<2 min) on the surface of Au@Ag NRs. After incubating the mixture of Hg²⁺ and Au@Ag NRs for 2 min, the color of the mixture did not change again, even when it was stored at room temperature for a week (data not shown). The mixture stability is essential for the accurate and visualized detection of Hg²⁺. The following redox reaction occurs between zero-valent Ag and Hg²⁺ with standard potentials of 0.8 V (Ag⁺/Ag) and 0.85 V (Hg²⁺/Hg).

Ag_n + Hg²⁺ ↔ Ag_n−2Hg + 2Ag⁺ (1)

Fig. 2 Etching mechanism of Au@Ag NRs by Hg²⁺. (a) Scheme of the Au@Ag NR etching mechanism. (b) Absorption spectra of Au@Ag NRs without Hg²⁺ and with different concentrations of Hg²⁺ (5 μM or 65 μM). TEM images of: (c) original Au@Ag NRs, (d) Au@Ag NRs etched by a low concentration of Hg²⁺ (5 μM), and (e) Au@Ag NRs etched by a high concentration of Hg²⁺ (65 μM), respectively. Insets: the corresponding colors of the Au@Ag NR solution with different Hg²⁺ concentrations.

Fig. 3 Effect of the Ag nanoshell thickness of Au@Ag NRs on Hg²⁺ detection. Typical absorbance spectra of Au@Ag NRs with a: (a) thin, (b) moderate, and (c) thick Ag nanoshell thickness mixed with different Hg²⁺ concentrations, respectively. Insets: the corresponding colors of the Au@Ag NR solution with different Hg²⁺ concentrations. (d) Dose–response curves of Hg²⁺ detection with Au@Ag-NR1, Au@Ag-NR2, and Au@Ag-NR3. The presented values are the average of three independent experimental results.

This redox reaction led to the etching of nano-Ag and the formation of Ag–Hg nanoalloy on the surface of the Au@Ag NRs. The etching mechanism was analyzed via TEM and EDS in addition to via UV-vis spectroscopy (Fig. 2c–e). The TEM images demonstrate that the color changes of Au@Ag NRs can be attributed to their morphological transition as the etching process progresses. This observation was demonstrated by the shoulder shapes (Fig. 2d, with 5 μM Hg²⁺), and etching-cylindrical shapes (Fig. 2e, with 65 μM Hg²⁺) formed in the presence of Hg²⁺ compared to the original cylindrical shapes (Fig. 2c) in the absence of Hg²⁺. Interestingly, the Ag nanoshell of the Au@Ag NRs was gradually etched from the end of the NRs (Fig. 2c–e) as the Hg²⁺ concentration increased. The morphology transition of the Au@Ag NRs can be ascribed to the following reasons: first, the active Ag atoms at both ends of the Au@Ag NRs can easily be coordinated with Hg²⁺ and separated from the original nanostructure. On basis of the Gibbs–Thomson effect, a convex surface has a higher surface energy than a flat surface.²⁷ Both ends of the Au@Ag NRs have a higher surface energy than their lateral sides. On the other hand, the Ag atoms at the end areas have a higher coordination number than the lateral sides, which results in higher surface energy in these areas.¹² Therefore, the ends of the Au@Ag NRs are more prone to etching over other areas. On the other hand, the as-prepared Au@Ag NRs were surrounded by a small amount of positively charged CTAB molecules. More CTAB molecules were on the lateral side than on the end; as a result, more Hg²⁺ ions were adsorbed on the latter than on the former because of electrostatic repulsion, thus accelerating the etching of the ends of the Au@Ag NRs, which caused the distinct changes in the absorption spectra.

Freshly generated Hg atoms can strongly bond on the Ag surface, which accounts for the slight blueshift of the SPR band of the Ag NPs.²⁸ The shape change of the Au@Ag NRs in the presence of Hg²⁺ indicates the reduction of Hg(II) to Hg(0) and thus the formation of an amalgam of Hg and Ag wrapped around the Au@Ag NRs.²⁹ To further verify the interaction between Au@Ag NRs and Hg²⁺, the EDS in STEM was used to characterize the elemental identity of the Au@Ag NRs after adding a low concentration (1 μM) and high concentration (30 μM) of Hg²⁺. The EDS has been recently proven to be a powerful technique in analyzing the elemental identity and location of atomic columns in nanomaterials at atomic resolution.³⁰ The results of the EDS elemental analyses are summarized in Table 1 and shown in Fig. S3.† Hg²⁺ appeared on the surface of the Au@Ag NRs after Hg²⁺ reacted with Au@Ag NRs. The amount of Hg²⁺ increased and the amount of Ag decreased on the surface of Au@Ag NRs. This finding is consistent with the TEM images and UV-vis spectra of Au@Ag NRs. These observations also agree with those of previous results, which demonstrated the formation of a Hg nanoshell on the surface of a Ag nanoshell after Hg²⁺ action, which was attributed to the formation of the amalgam of Ag and Hg.¹² However, the etching mechanism of the Au@Ag NRs by Hg²⁺ has not been reported to date. In accordance with the proposed etching mechanism of Hg²⁺, the change in the absorption spectra caused the color change of the Au@Ag NR colloids in the presence of Hg²⁺. Furthermore, we proved that the effect of pH on the absorption spectra of the Au@Ag NRs was insignificant when the pH was higher than 2 (Fig. S4†). Thus, the colorimetric detection of Hg²⁺ can be realized.

Table 1 EDS spectral analysis of Au@Ag NRs upon the addition of Hg²⁺ in low concentration (5 μM) and high concentration (30 μM)

Hg^2+ concentration	Element	Weight (%)	Atomic (%)	Uncert. (%)	Correction	k-Factor
Low	Ag(K)	21.47	33.32	1.43	0.98	6.491
	Au(L)	73.69	62.64	1.93	0.75	5.653
	Hg(L)	4.83	4.03	0.70	0.75	5.824
High	Ag(K)	13.23	21.82	0.86	0.98	6.491
	Au(L)	76.24	68.85	1.55	0.75	5.653
	Hg(L)	10.51	9.32	0.65	0.75	5.824

---

Dose–responses of Hg²⁺ using Au@Ag NRs with different Ag nanoshell thicknesses

The absorption spectra of Au@Ag-NR1, Au@Ag-NR2, and Au@Ag-NR3 upon the addition of different Hg²⁺ concentrations were compared to investigate the effect of the Ag nanoshell thickness of the Au@Ag NRs on Hg²⁺ detection. Fig. 3a–c display the absorption spectra of the interaction between Au@Ag NRs and Hg²⁺ at concentrations ranging from 0 to 267 μM. For Au@Ag-NR1 and Au@Ag-NR2, the absorbance intensity gradually reduced as the Hg²⁺ concentration was gradually increased and the wavelength of peak 1 blueshifted when the Hg²⁺ concentration was not too high. However, the wavelength of peak 1 redshifted when the Hg²⁺ concentration exceeded 33 μM. This phenomenon should contribute to the LSPR of Au NRs caused by the complete etching of the Ag nanoshell of the Au@Ag NRs. After the addition of Hg²⁺ solution to the prepared Au@Ag-NR1 solution, the color of the Au@Ag-NR1 solution changed from orange-yellow to light-violet, and its absorbance intensity decreased as the Hg²⁺ concentration was increased (inset of Fig. 3a). The Hg²⁺ detection system based on Au@Ag-NR1 showed a linear response at 0.6–20 μM and a detection limit of 10 nM based on the 3σ/slope (where σ is the standard deviation of the blank samples) (Fig. 3d). This detection limit is comparable to that of some reported colorimetrical sensors for Hg²⁺,^3,11,12,31 and satisfies the requirement for drinking water standards in the US.³² The introduction of Hg²⁺ to Au@Ag-NR2 decreased the absorbance intensity, and changed the color (inset of Fig. 3b). The linear response ranged from 2.0 μM to 30 μM, and the detection limit was 200 nM (Fig. 3d).

The absorbance intensity of peak 3 was slightly higher than that of peak 1 in Au@Ag-NR3 as compared with Au@Ag-NR1 and Au@Ag-NR2. Fig. 3c displays the absorption spectra of the interaction between Au@Ag-NR3 and Hg²⁺ at 0–267 μM. The addition of Hg²⁺ significantly affected the absorbance intensity and peak position of Au@Ag-NR3. The linear response ranged from 5.0 μM to 200 μM and the detection limit was 500 nM (Fig. 3d). The results indicate that the sensitivity of the Au@Ag NR-based chemosensor decreases with the increasing Ag nanoshell thickness. Interestingly, the color drastically transitioned from brownish-red to light-red, light-violet, and finally to colorless (inset of Fig. 3c). This color range in Au@Ag-NR3 allows visualization of the color change. Therefore, from the point of view of macroscopic colorimetry, Au@Ag-NR3 is more suitable for direct readout visualization than Au@Ag-NR1 and Au@Ag-NR2. Except for the decrease in the absorbance intensity at 580 nm with the increasing Hg²⁺ concentration, the absorbance intensity of 412 nm was Hg²⁺ concentration-dependent (Fig. S5†). However, what attracts us most is the change of the dip located between peak 1 and peak 3. As the Hg²⁺ concentration was increased, the dip wavelength slightly blueshifted and the absorbance intensity gradually decreased. When the Hg²⁺ concentration was 267 μM, peaks 1 and 3 were merged, and the dip between the two original peaks disappeared and ultimately transformed into a wide peak. The peak wavelength and absorbance intensity of LSPR are commonly used parameters to detect heavy metal ions on the basis of NP absorption spectral analyses. However, the LSPR absorption spectral changes induced by noble metal nanostructures with different morphologies are influenced by other parameters. Abundant spectral signals, such as the dip between the two peaks, the integration of adjacent peaks, and the relative intensity change of different peaks, can be applied to quantify the concentration of targets. These enhance the overall performance of the sensors. Therefore, further exploring the normal spectral lines is essential to obtain new sensor parameters. As demonstrated in Fig. 3d, the absorbance intensity of the dip was Hg²⁺ concentration-dependent. The dip is a new sensor parameter located between two strong absorption peak positions and can be used to quantify the target. The Hg²⁺ concentration detection can be detected using the change in the position and intensity of the dip.

Several studies have reported on the application of Au@Ag NPs as sensors to detect small molecules and metal ions.^3,22,33 However, our proposed approach is different from the previously reported Au@Ag core–shell nanomaterial-based sensors and their sensing properties. The proposed approach is convenient and efficient and does not need complicated instruments. Only one absorption spectrometer after only 2 min incubation is needed for the proposed approach. Moreover, the proposed sensor can achieve a tunable dynamic range by adjusting the Ag nanoshell thickness of Au@Ag NRs. This is important as a practical sensor needs to have a tunable dynamic range that matches the concentration ranges for different locations, because most analytes of interest have varied concentration ranges at different locations in the environment.³⁴

Selectivity of Ag@Au NR-based sensor

To assess the selectivity of the unmodified Ag@Au NR-based sensor, other metal ions (Cu²⁺, Mg²⁺, Cd²⁺, Al³⁺, Co²⁺, Mn²⁺, Pb²⁺, Ca²⁺, Zn²⁺, Fe²⁺, Fe³⁺, and Ag⁺) at concentrations of up to 1 mM were added in to Ag@Au-NR3 solution under the same conditions. Fig. 4 showed the interaction between freshly prepared Ag@Au NRs and various metal ions, and their color change. The solution contacting Hg²⁺ changes from brownish-red to colorless, whereas other alkaline earth metals (Mg²⁺, Ca²⁺) and transition-metal ions (Ni²⁺, Mn²⁺, Cu²⁺, Zn²⁺, Co²⁺, Cd²⁺, Fe²⁺, Fe³⁺, and Ag⁺) exerted negligible effects on the color and SPR band of the Ag@Au NR solution. This result indicates that the Ag@Au NR-based assay approach is highly selective toward Hg²⁺ but not to other transition-metal and alkaline metal ions under similar conditions. The specific Hg²⁺ detection can be mainly attributed to the specific etching ability of Hg²⁺-Ag@Au NRs. Different Hg types, including Hg, Hg (OH)₂, HgO, CH₃Hg⁺ and CH₃HgCl, can be transformed into Hg²⁺ ions using a digestive method.³¹ Thus the proposed probe may offer great promise as a colorimetric detection method for total Hg forms.

Fig. 4 Selectivity of the Ag@Au NR-based sensing system. The concentration of Mg²⁺, Ca²⁺, Ni²⁺, Mn²⁺, Cu²⁺, Zn²⁺, Co²⁺, Cd²⁺, Fe²⁺, Fe³⁺, or Ag⁺ was 1 mM, and the concentration of Hg²⁺ was 200 μM. Insets: the corresponding colors of the Au@Ag NR solution upon the addition of different heavy metal ions.

Determination of Hg²⁺ in drinking water samples

The applications of the proposed colorimetric sensor based on unmodified Au@Ag NRs were evaluated to determine Hg²⁺ in real samples such as tap water and commercially available bottled water. The water samples were spiked with different Hg²⁺ concentrations. The results summarized in Table 2 agree with the expected values. The recovery of all measured samples was between 90% and 115%, and the parallel tests showed that the relativity coefficient (the relative ratio of the standard deviation σ to the mean μ) was within 1.78% to 5.4% (n = 2).³⁵ These results indicate that any possible interference from the different background compositions of water samples on the Au@Ag NR-based sensing system was negligible. Therefore, the developed method can be successfully applied to Hg²⁺ analysis in drinking water samples.

Table 2 Determination of Hg²⁺ in drinking water samples using the proposed method

Samples	Spiked concentration (μM)	Detection concentration (μM)	Recovery (%)	CV (%)
Bottled water	0.6	0.58	96.7	3.56
	1.0	0.92	92.0	3.78
	2.0	2.11	105.5	1.78
Tap water	0.6	0.67	111	3.56
	1.0	0.97	97.0	5.4
	2.0	2.16	108	3.23

---

In summary, we developed a simple and cost-effective colorimetrical approach for the rapid and highly selective detection of Hg²⁺ based on the etching mechanism of unmodified Au@Ag NRs by Hg²⁺. This simple and rapid method showed detection limits as low as 10 nM for Hg²⁺, as well as high selectivity toward Hg²⁺ over other metal ions. Our proposed approach has several advantages over other colorimetrical sensors for Hg²⁺ detection. First, the present method only requires unmodified Au@Ag NRs as the detection material. Second, the whole detection process is time-saving (<2 min), and the color changes of the Au@Ag NR solution upon the addition of Hg²⁺ are visible with the naked eye. Third, the detection system only requires one absorption spectrometer. The proposed sensor can also achieve a tunable dynamic range by adjusting the Ag nanoshell thickness of Au@Ag NRs. We believe that this method could provide new breakthroughs in Hg²⁺ detection in drinking water.

Acknowledgements

This study was financially supported by the National Natural Science Foundation of China (21077063, 21277173), the National Instrument Major Project of China (2012YQ3011105), the Special Fund of State Key Joint Laboratory of Environment Simulation and Pollution Control (14K01ESPCT), and the Basic Research Funds in Renmin University of China from the Central Government (13XNLJ01).

References

1. P. C. Ray, Chem. Rev., 2010, 110, 5332.
2. K. Saha, S. S. Agasti, C. Kim, X. Li and V. M. Rotello, Chem. Rev., 2012, 112, 2739.
3. J. Du, L. Jiang, Q. Shao, X. Liu, R. S. Marks, J. Ma and X. Chen, Small, 2013, 9, 1467–1481.
4. K. Tan, G. Yang, H. Chen, P. Shen, Y. Huang and Y. Xia, Biosens. Bioelectron., 2014, 59, 227–232.
5. S. Jayabal, A. Pandikumar, H. N. Lim, R. Ramaraj, T. Sund and N. M. Huang, Analyst, 2015, 140, 2540–2555.
6. X. Huang, S. Neretina and M. A. El-Sayed, Adv. Mater., 2009, 21, 4880–4910.
7. T. K. Sau, A. L. Rogach, F. Jackel, T. A. Klar and J. Feldmann, Adv. Mater., 2010, 22, 1805.
8. F. Zhang, J. Zhu, J. J. Li and J. W. Zhao, J. Mater. Chem. C, 2015, 3, 6035–6045.
9. X. Yang, Y. Yu and Z. Gao, ACS Nano, 2014, 8, 4902–4907.
10. B. Xue, D. Wang, J. Zuo, X. Kong, Y. Zhang, X. Liu, L. Tu, Y. Chang, C. Li, F. Wu, Q. Zeng, H. Zhao, H. Zhao and H. Zhang, Nanoscale, 2015, 7, 8048–8057.
11. K. Farhadia, M. Forough, R. Molaei, S. Hajizadeh and A. Rafipour, Sens. Actuators, B, 2012, 161, 880–885.
12. L. Chen, X. Fu, W. Lu and L. Chen, ACS Appl. Mater. Interfaces, 2013, 5, 284–290.
13. H. H. Harris, I. J. Pickering and G. N. George, Science, 2003, 301, 1203.
14. S. Ashoka, B. M. Peake, G. Bremner, K. J. Hageman and M. R. Reid, Anal. Chim. Acta, 2009, 653, 191–199.
15. C. G. Yuan, J. Wang and Y. Jin, Microchim. Acta, 2012, 177, 153–158.
16. G. Aragay, J. Pons and A. Merkoci, Chem. Rev., 2011, 111, 3433–3458.
17. F. H. Wang, C. W. Cheng, L. C. Duan, L. Wu, M. Z. Xia and F. Y. Wang, Sens. Actuators, B, 2015, 206, 679–683.
18. D. N. Kumar, A. Rajeshwari, S. A. Alex, N. Chandrasekaran and A. Mukherjee, New J. Chem., 2015, 39, 1172–1178.
19. J. S. Lee, M. S. Han and C. A. Mirkin, Angew. Chem., Int. Ed., 2007, 46, 4093–4096.
20. X. Zhang and Y. Y. Zhu, Sens. Actuators, B, 2014, 202, 609–614.
21. L. Li, B. Yu and T. You, Biosens. Bioelectron., 2015, 74, 263–269.
22. J. R. Hao, B. Xiong, X. D. Cheng, Y. He and E. S. Yeung, Anal. Chem., 2014, 86, 4663.
23. N. Babak and A. E. Mostafa, Chem. Mater., 2003, 15, 1957–1962.
24. Y. J. Xiang, Langmuir, 2008, 24, 3465–3470.
25. B. Xiong, R. Zhou, J. Hao, Y. Jia, Y. He and E. S. Yeung, Nat. Commun., 2013, 4, 1708.
26. G. Park, C. Lee, D. Seo and H. Song, Langmuir, 2012, 28, 9003–9009.
27. J. Becker, I. Zins, A. Jakab, Y. Khalavka, O. Schubert and C. Sönnichsen, Nano Lett., 2008, 8, 1719–1723.
28. Y. Li, Z. Li, Y. Gao, A. Gong, Y. Zhang, N. S. Hosmane, Z. Shen and A. Wu, Nanoscale, 2014, 6, 10631.
29. L. Deng, Y. Li, X. Yan, J. Xiao, C. Ma, J. Zheng, S. Liu and R. Yang, Anal. Chem., 2015, 87, 2452–2458.
30. N. R. Lugg, G. Kothleitner, N. Shibata and Y. Ikuhara, Ultramicroscopy, 2015, 151, 150–159.
31. A. Fan, Y. Ling, C. Lau and J. Lu, Talanta, 2010, 82, 687–692.
32. Mercury Update: Impact of Fish Advisories, EPA Fact Sheet EPA-823-F-01–011, EPA, Office of Water, Washington, DC, 2001.
33. X. Miao, S. Zou, H. Zhang and L. Ling, Sens. Actuators, B, 2014, 191, 396–400.
34. Y. Xiang, A. Tong and Y. Lu, J. Am. Chem. Soc., 2009, 131, 15352–15357.
35. N. Yildirim, F. Long, C. Gao, M. He, H. Shi and Z. A. Gu, Environ. Sci. Technol., 2012, 46, 3288–3294.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra19627b

---