Introduction

Nowadays, the pollution caused by heavy metals in nature is a great concern due to potentially toxic effects on living creatures. Cadmium, a nonessential element for life, is widely used in fertilizers, pesticides, nickel-cadmium batteries, dyes, pigments, and coating of steel and various alloys, resulting in the widespread contamination of it in air, soil, and water (Dobson 1992; McLaughlin and Singh 1999). Cadmium has been recognized as a highly toxic heavy metal ion and is enlisted by the US Environmental Protection Agency, Disease Registry, and Agency for toxic substances as one of the superior pollutants. It accumulates in the human body with a long biological half-life of two to three decades. As a consequence of the food chain system, cadmium exposure can cause anemia, abdominal pain, neurological and adverse developmental effects, kidney damage, hypertension, and changes in vitamin D metabolism (Daher 1995; Darwish and Blake 2001; Davis et al. 2006; Friberg et al. 1992; Martinez and Blasco 2012). In addition, chronic exposure to Cd2+ has implicated a cause of cancer in lungs, prostate, pancreas, and kidney and even increased risk of cancer (Akesson et al. 2008; Jane et al. 2006). Therefore, the accurate determination of Cd2+ is critical.

Scientists have been dedicated to developing the facile detection of Cd2+ for the purpose of public health. The techniques currently available for the detection of Cd2+ include flame atomic absorption spectrometry (Mirabi et al. 2015), electrochemical analysis (Willemse et al. 2011), inductively coupled plasma atomic emission spectrometry (Matsumoto et al. 2010), atomic fluorescence spectrometry (Wan et al. 2006), inductively coupled plasma mass spectrometry (Guo et al. 2010), and molecular fluorescence spectroscopy (Yunus et al. 2008). Although these methods offer excellent sensitivity and multielement analysis, they are expensive, time-consuming, intricate, and arduous for on-site field analysis. Hence, the interest in swapping instrumental analytical tools with suitable, selective, sensitive, and low-cost Cd2+ sensors has been increasing.

To date, optical sensors, especially the colorimetric ones, have always been the most convenient analytical tool and gained a lot of interest because they provide a simple and economical assay without the aid of expensive instruments (Chansuvarn et al. 2015; Ratnarathron et al. 2015; Zhan et al. 2015). Compared with the well-developed chromophoric chemosensors, metal nanoparticle-based colorimetric assays have recently emerged as a fascinating research field (Chen et al. 2015a, 2015b; Mehta et al. 2015; Sharif et al. 2015; Yang et al. 2014). Among them, gold and silver nanoparticles (AuNPs, AgNPs) are of particular interest as they have played significant roles in environmental monitoring and biological sensing (Chen et al. 2015a, 2015b; Medley et al. 2008). One feature that makes them particularly outstanding is that AuNPs and AgNPs possess high extinction coefficients in the visible region. The colors of the dispersed and aggregated nanoparticle solution are different. For example, the dispersed AgNP solution is bright yellow, while the highly aggregated AgNP solution is red or blue (Zhang and Ye 2011). Many metal nanoparticles have been recently used for the determination of Cd2+. For example, Zhang’s group (Zhang et al. 2012) developed a peptide-modified AuNP-based colorimetric method for the parallel detection of Cd2+, Ni2+, and Co2+ in river water. Guo et al. (2014) reported a novel method for Cd2+ based on label-free AuNPs with the aid of high salt concentration and glutathione. Most recently, Manjumeena et al. (2015) also developed a glutathione functionalized green AuNP probe for the precise colorimetric detection of Cd2+ contamination in the environment. Kumar and Anthony (2014) reported a selective colorimetric sensor for Cd2+, Hg2+, and Pb2+ ions using costabilizing agents. Obviously, the distinctive properties of AuNPs/AgNPs make them very useful for the development of colorimetric sensing for Cd2+, but it is still demanding to develop novel colorimetric methods to detect Cd2+ with better sensitivity and selectivity.

Herein, we report a simple, rapid, selective, and sensitive colorimetric method for the detection of Cd2+ in various real samples using 1-amino-2-naphthol-4-sulfonic acid (ANS)-AgNPs as a probe. This sensing strategy is based on the aggregation of ANS-AgNPs induced by Cd2+, which resulted in a red shift in the surface plasmon resonance (SPR) absorption band and an appreciable color change from bright yellow to reddish-brown. The bare AgNPs were functionalized with ANS and used as sensors for the colorimetric detection of Cd2+ by taking advantage of metal chelation of Cd2+ with the hydroxyl group (–OH) and amino group (–NH2) of ANS on the surface of AgNPs (Scheme 1).

Scheme 1
scheme 1

Schematic representation for the detection of Cd2+ by ANS-modified AgNPs

Full size image

Materials and methods

Chemicals and materials

All chemicals used were of analytical grade or of the highest purity available. Silver nitrate, sodium borohydride, ANS, and metal salts were purchased from Shanghai Qingxi Technology Co., Ltd. (Shanghai, China, www.ce-r.cn/sites/qingxi/). All glassware was cleaned thoroughly with freshly prepared aqua regia (3:1 (v/v) HCl/HNO3) and rinsed thoroughly with doubly distilled water prior to use. Doubly distilled water was used to prepare all the solutions in this study.

Instruments

UV–Vis absorption spectra were examined on a UV-2550 spectrophotometer (Shimadzu, Kyoto, Japan) using a 1.0-cm quartz cell at room temperature. The infrared spectra were measured in KBr pellets with a Nicolet 5700 FTIR Spectrum (Nicolet, USA). Transmission electron micrograph (TEM) was recorded by a JEM-2100 transmission electron microscope (JEOL Ltd., Japan). The data of dynamic light scattering (DLS) were obtained on an NPA152 nanoparticle size analyzer (Microtrac Inc., USA). Photographs were taken with Sony DSC-TX5 camera (Tokyo, Japan).

Surface modification of AgNPs or AuNPs with ANS

AgNPs were prepared by the reduction of AgNO3 with NaBH4 according to a previously reported method (Li et al. 2010). Briefly, in a flask, 2.0 mL of 0.01 M AgNO3 was added into 97.5 mL doubly distilled water. Then, 8.8 mg of NaBH4 was quickly added into the above mixture solution under vigorous stirring for 30 min at room temperature (25 ± 2 °C). To obtain ANS-AgNPs, 0.5 mL of 0.01 M ANS was added into the above aqueous solution and then stirred for 2 h at room temperature to functionalize the surface of AgNPs with ANS. Finally, the as-prepared silver colloidal solution was stored at 4 °C in the dark before use.

Citrate-capped AuNPs were prepared by means of the chemical reduction of HAuCl4 as follows. A one hundred-milliliter aqueous solution of 1.0 mM HAuCl4 was brought to a boil with vigorous stirring in a round-bottom flask fitted with a reflux condenser. Trisodium citrate (38.8 mM, 10 mL) was then added rapidly to the solution, and the mixture was heated under reflux for another 15 min. The solution was cooled to room temperature while being stirred continuously. The obtained wine-red solution was stored at 4 °C for further use. For surface modification of AuNPs with ANS, 0.5 mL of 0.01 M ANS was added into the above aqueous solution. The resulted mixture was equilibrated at room temperature for 2 h. Finally, the as-prepared silver colloidal solution was stored at 4 °C in the dark before use.

Colorimetric detection of Cd2+

The following procedure was used for colorimetric detection of Cd2+ using ANS-AgNPs as a probe. Briefly, different concentrations of Cd2+ solutions were added into the 2.0-mL ANS-AgNP solution. The mixed solution was vortexed for 30 s and then allowed to stand for 5 min at room temperature. Absorption spectra were recorded accordingly. The magnitude of absorbance ratio (A580nm/A390nm) has been used as the index parameter for Cd2+ detection, and the numerator and the denominator represent the degree of aggregation and dispersion, respectively.

Results and discussion

Characterization of ANS modified AgNPs

The formation of ANS-AgNPs is estimated by the color change of the solution and further confirmed by the FT-IR spectra of ANS and ANS-AgNPs. It could be observed that the color of the reaction solution turned bright yellow from colorless. Figure S1 displays FT-IR spectra of ANS and ANS-AgNPs. It was noteworthy that the bands in the range of 900–650 cm−1 which are the out-of-plane bending vibration band of the amino group and the band at about 3230 cm−1 corresponding to –NH asymmetric stretching vibrations appeared in the spectra of both ANS and ANS-AgNPs, although the peak slightly shifted. It was also found that the wave number at 1350 cm−1 is the bending vibration band of –OH located in the spectra of both ANS and ANS-AgNPs. For pure ANS, the peaks at 1162 and 1200 cm−1 are assigned to the characteristic stretching vibration absorption of the –SO3H group, which were hardly to be seen in the spectrum of ANS-AgNPs, indicating that ANS had been successfully bound onto the surface of the AgNPs via the sulfonic group of ANS. Besides, from the TEM image of ANS-AgNPs in Fig. S2, it was found that a dark core was surrounded by a light shell for the AgNPs, and the corresponding high-resolution TEM (HRTEM) image showed that the core had the lattice fringe with interplanar spacings of 0.235 nm ascribed to the (111) plane of Ag, indicating the successful modification of ANS as the ligand on the AgNP surface. In addition, in Fig. S3, for the ANS-AgNPs, the full-scale XPS spectrum showed the characteristic peaks of C 1s, N 1s, O 1s, S 2p, and Ag 3d, further suggesting that ANS was successfully anchored on the AgNP surface. It should be noted that the peaks of the N and S elements were relatively weak, which was possibly because their contents were very low compared with other elements like C and O in ANS. These results revealed that ANS molecules were successfully attached onto the surface of the AgNPs.

To evaluate surface capacity of the ANS on the AgNPs, we compared the fluorescence intensities at ca. 440 nm of supernatant solution of ANS-AgNPs and pure ANS solution before functionalization (Fig. S4). It was found that approximately 54 % of ANS was bound onto the AuNPs, from which we can calculate the amount of ligands to be about 1931 ANS molecules per AgNP.

Selective response of ANS-AgNPs to Cd2+

When the ANS-AgNPs are well dispersed in solution, the colloids exhibit a characteristic SPR band at 390 nm in the UV–Vis absorption spectrum and a bright yellow color. However, upon addition of Cd2+, the absorbance at 390 nm decreased sharply, along with the emergence of a long wavelength band at ca. 580 nm (Fig. 1a). Meanwhile, the color of the ANS-AgNPs immediately changed from yellow to reddish-brown, as depicted in Fig. 1b. Both the spectral and color changes indicated the aggregation of AgNPs induced by Cd2+, which will also be demonstrated below.

Fig. 1
figure 1

a UV-visible absorption spectra and b a photograph of ANS-AgNPs in the presence of different metal ions (Cu2+, Cd2+, Co2+, Ni2+, Hg2+, Fe3+, Al3+, Cr3+, Ba2+, Pb2+, Ca2+, Mg2+, Zn2+, Mn2+, and Ag+; the concentrations of metal ions are all 5.0 μM)

Full size image

In order to investigate the recognition ability of ANS-AgNPs toward metal ions, the specificity of this method was also evaluated with other metal ions (Cu2+, Co2+, Ni2+, Hg2+, Fe3+, Al3+, Cr3+, Ba2+, Pb2+, Ca2+, Mg2+, Zn2+, Mn2+, and Ag+). The study revealed that the spectrum of ANS-AgNPs had little change in the presence of almost common metal ions. As Ni2+, Co2+, and Pb2+ had a slight effect on the spectrum of ANS-AgNPs, methylimidazole was applied as the masking agent. The result showed that after masking, Ni2+, Co2+, and Pb2+ did not affect the detection of Cd2+ (Fig. S5). It should be noted that although methylimidazole was introduced, the spectra were pretty much the same as that of pure ANS-AgNPs for other metal ions and that the response to Cd2+ was not influenced, either. This was directly evidenced from Fig. 1b, in which all colorimetric tubes demonstrate bright yellow color after the masking except that Cd2+ induced the color change of ANS-AgNPs. So, the developed strategy was highly selective toward Cd2+.

Mechanism of the sensing system

The spectral and color changes of ANS-AgNPs upon addition of Cd2+ could be well explained by the aggregation of ANS-AgNPs through the coordination interaction between Cd2+ and ANS. ANS-AgNPs were stabilized in solution because ANS on the surface of AgNPs protected them from aggregation. ANS has –NH2, −SO3, and –OH groups, which can be used to bind with metal ions (Jaliehvand et al. 2009; Xue et al. 2011). As shown in Scheme 1, ANS is linked with AgNPs through –SO3H, and owing to the interactions between the O atom in the –OH group and the N atom in the –NH2 group of ANS and Cd2+, the neighboring AgNPs became closer and finally aggregated.

DLS analysis (Fig. 2a) clearly exhibited that Cd2+ can cause the aggregation of ANS-AgNPs. The intensity contribution versus diameters of ANS-AgNPs in the absence and presence of Cd2+, around 12 and 28 nm, respectively, also suggested that Cd2+ could result in the effective aggregation of ANS-AgNPs. Furthermore, the TEM images revealed monodispersed ANS-AgNPs in the absence of Cd2+ and a significant aggregation in the presence of Cd2+, as seen in Fig. 2b. The as-prepared ANS-AgNPs tended to aggregate due to the strong coordination bond between Cd2+ and ANS, which could be easily confirmed by the addition of powerful chelating agent ethylenediaminetetraacetic acid (EDTA). The presence of EDTA recovered the absorption spectrum of ANS-AgNPs almost to the original state (Fig. S6).

Fig. 2
figure 2

Size distribution (a) and TEM images (b) of the ANS-AgNPs in the absence (a) and presence of 5.0 μM Cd2+ (b)

Full size image

Optimization of the sensing conditions

Effect of ANS concentration

Parameter ascertainment and optimization of the present method are key factors to its performance in terms of effectiveness and sensitivity, which strongly depends on the ANS concentration. Since the concentration of ANS could have an effect on ANS-AgNP detection capability, the stoichiometric ratios between ANS and AgNPs were explored. In our work, it was found that the absorbance ratio (A580nm/A390nm) increased with various molar ratios of AgNPs and ANS (2:1, 4:1, 8:1). As seen in Fig. 3, the lower the molar ratio of AgNPs and ANS was, the worse the capability of Cd2+ detection was. In contrast, when the molar ratio of AgNPs and ANS was higher than 4:1, the response of Cd2+ to ANS-AgNPs became worse. Based on above results, the molar ratio of 4:1 between AgNPs and ANS was considered to be optimal for Cd2+ detection.

Fig. 3
figure 3

Influence of the concentration of ANS on the absorbance ratio (A580nm/A390nm) of ANS-AgNPs

Full size image

Effect of pH and reaction time

The effect of pH on the interaction between ANS-AgNPs and Cd2+ was studied over the pH range from 4.0 to 12 (Fig. S7A). When the pH value was less than 4.0, ANS-AgNPs were unstable and aggregated easily. High pH value more than 12 is not beneficial since Cd2+ may react with OH− to form the corresponding metal hydroxides, which would affect the accuracy to some extent. It was found that the absorbance ratio reached the maximum value at alkaline media upon addition of Cd2+. Since the pH value of the as-prepared ANS-AgNPs was 9.8, to save the time and cost, the original solution (pH around 9.8), which was the condition of the as-prepared ANS-AgNPs without any pH adjustments, was selected.

To estimate the kinetics on the aggregation of ANS-AgNPs induced by Cd2+, we studied the time-dependent A580nm/A390nm values from the UV–Vis absorption spectra of ANS-AgNPs in the presence of Cd2+ at different times from 0 to 10 min. As shown in Fig. S7B, the absorbance ratio increased very rapidly within 3 min and then changed subtly, and the equilibrium was achieved in 5 min. Therefore, 5 min was chosen as the reaction time in the following tests for Cd2+.

Sensitivity of the colorimetric assay for Cd2+

In order to evaluate the sensitivity of the developed colorimetric method, under the above optimized conditions, the absorption spectra of ANS-AgNPs to different concentrations of Cd2+ were recorded. As the concentration of Cd2+ increased, the UV–Vis absorbance curves showed a red shift and broadened gradually and a new absorption peak emerged slowly at a longer wavelength (580 nm). Furthermore, with the increase of the concentration of Cd2+, the color of the ANS-AgNPs gradually changes from yellow to orange and finally to reddish-brown (Fig. 4), suggesting the Cd2+ concentration-dependent aggregation of ANS-AgNPs. For quantitative analysis of Cd2+, the absorbance ratio at 580 and 390 nm was used. A linear correlation existed between A580nm/A390nm and the Cd2+ concentration in the range from 1.0 to 10 μM (as shown in the inset in Fig. 4). The calibration equation obtained was y = 0.0987 [C]/μM + 0.0159 with a correlation coefficient (R) of 0.997, where y is the absorbance ratio A580nm/A390nm and [C] is the concentration of Cd2+. The limit of detection (LOD) was calculated to be 87 nM defined by the equation LOD = 3 S 0 / k, where S 0 is the standard deviation of blank measurements (n = 10) and k is the slope of the calibration curve. Compared with existing reported colorimetric methods of Cd2+, our assay gave comparable detection limits (Table S1).

Fig. 4
figure 4

Photographs and UV-visible spectra of ANS-AgNP solutions with various concentrations of Cd2+ in the range from 0 to 10 μM. Inset is the linear calibration curve between absorbance ratio A580nm/A390nm and the concentrations of Cd2+

Full size image

We also investigated the response of ANS-AuNPs to Cd2+. Unlike that of ANS-AgNPs, ANS-AuNPs were testified to be insensitive to Cd2+ (Fig. S8). For this result, we explained the following: on the one hand, it is because the AuNP solution is acidic (pH ~5.5), in which the sulfonic group of ANS was too difficult to be deprotonated to bind to the Au atom. However, in the case of ANS-AgNPs, the basic environment of the solution (pH ~9.8) is very favorable for the deprotonation of the sulfonic group leading to the strong binding to the Ag atom. On the other hand, the as-synthesized AuNPs were capped with the citrate, which, to some extent, would block the contact between the Au atom and ANS. While for ANS-AgNPs, the bared AgNPs were prepared by NaBH4 reduction, which is easier to be directly modificated with ANS due to the absence of any other ligand.

The effect of coexisting components

In order to examine the specific detection of Cd2+ using ANS-AgNPs, we investigated the potentially interfering components which may exist in milk, serum, and lake water. As shown in Fig. 5, the presence of the following amounts of foreign species compared with the concentration of Cd2+ resulted in less than ±10 % error: 100 times of Ag+ and Ba2+; 50 times of Mg2+; 40 times of K+; 35 times of Ca2+; 30 times of Cu2+; 15 times of Fe3+; 10 times of Zn2+, Al3+, and Cr3+; 5 times of Na+ and Ni2+; 2 times of Hg2+; 1.5 times of Mn2+; and 1 time of Pb2+ and Co2+. Meanwhile, the effect of surface-active agents was also inspected. The existence of 50 times nonionic surfactant (polyoxyethylene nonylphenol ether, OP emulsifier) and 10 times anionic surfactant (sodium dodecyl sulfate, SDS) did not interfere. However, 1 time of cationic surfactant (cetyl trimethyl ammonium bromide, CTAB) interfered with the determination of Cd2+, indicating that ANS-AgNPs had a negative charge. These results showed that the present method had good anti-interfering ability toward Cd2+ against coexisting components.

Fig. 5
figure 5

Selectivity of the ANS-AgNPs for Cd2+ detection. Blue bars represent the absorbance ratio of ANS-AgNPs in the presence of Cd2+ or the competing ions. The concentrations of metal ions are all 5.0 μM. Red bars are the responses of ANS-AgNPs for a mixture of Cd2+ (5.0 μM) and maximum amounts of competing ions (color figure online)

Full size image

Practical application

To evaluate the applicability of the colorimetric assay to real samples, lake water, milk powder, and serum were spiked with Cd2+ and tested in the assay. The detailed pretreatments of real samples were presented in Electronic Supplementary Materials. As summarized in Table 1, the mean recoveries of these samples are between 88.3 and 106 %. Such excellent results suggested that other chemical species present in samples did not interfere with the quantification of Cd2+. These results clearly indicate that the proposed method is able to detect Cd2+ from various samples with high sensitivity and excellent selectivity.

Table 1 Determination of Cd2+ in real samples
Full size table

Conclusions

In this paper, a facile and rapid colorimetric method has been developed for the detection of Cd2+ using ANS-AgNPs as the colorimetric probe. The ANS-AgNPs showed specific recognition to Cd2+ accompanied by the color change from bright yellow to reddish-brown, which could be observed with the naked eye. It was demonstrated that the aggregation of ANS-AgNPs induced by Cd2+ was attributed to cooperative metal-ligand interaction between Cd2+ and ANS. The absorbance ratio A580nm/A390nm was linear with the concentration of Cd2+ ranging from 1.0 to 10 μM with a correlation coefficient of 0.997, and LOD was as low as 87 nM. Finally, this developed method was successfully applied to determinate Cd2+ in milk powder, serum, and lake water with satisfactory results. Therefore, this assay opens up new possibilities for developing analytical methods using ANS-AgNPs as a colorimetric probe for the monitoring of Cd2+ in real samples.