Microwave-assisted synthesis of BSA-stabilised gold nanoclusters for the sensitive and selective detection of lead(II) and melamine in aqueous solution

Abstract

In this study, a sensitive and selective fluorescence assay using the microwave-assisted synthesis of bovine serum albumin-stabilised gold nanoclusters (MW_BSA-AuNCs) was proposed for the determination of lead(II) and melamine in aqueous solution. The fluorescence of the MW_BSA-AuNCs was quenched through a Pb²⁺-mediated interparticle aggregation mechanism, and thus the change in fluorescence at 660 nm was dependent on the Pb²⁺ ion concentration. In addition, the fluorescence of the MW_BSA-AuNCs was also quenched on the basis of the high-affinity metallophilic Hg²⁺–Au⁺ interaction, whereas the fluorescence of the MW_BSA-AuNCs was restored when melamine was added and coordinated with Hg²⁺. Under optimal conditions, the limits of detection for Pb²⁺ ions and melamine at a signal-to-noise ratio of 3 were 4.8 nM and 2.94 μM, respectively. Our present approach is simpler and more cost-effective than the existing techniques for the detection of Pb²⁺ ions in seawater and aqueous urine samples and of melamine in milk samples.

---

Introduction

Exposure to lead (Pb) has been associated with behavioural abnormalities and impairments in learning, hearing, and cognitive functions in humans and experimental animals.^1–5 Pb can inhibit brain development in children, and the U.S. Food and Drug Administration (FDA) has suggested a Pb concentration of 2.5 μM (518 μg L⁻¹) as an “action level” for products intended for children.^6–9 Inductively coupled plasma mass spectrometry (ICP-MS), X-ray fluorescence spectrometry, atomic absorption spectrometry, and anodic stripping voltammetry are effective techniques for determining metal ion concentrations; however, they are expensive and thus unsuitable for on-site analysis.^10–13 Therefore, developing sensitive, selective, and reliable analytical techniques for determination of Pb is crucial.

Melamine is a nitrogen-rich (66.7% by weight) industrial chemical compound that has been extensively used in fire retardant pigment and manufacturing of plastic, insecticides, and fertilisers.¹⁴ Recently, melamine was found using the Kjeldah method to have been used in milk products; the illegal addition was made to increase nitrogen content, yielding products that appear to have higher protein content at a reduced cost. High levels of melamine easily form high molecular weight complexes, which have poor aqueous solubility, leading to precipitation in renal tubules. Consequently, they cause urinary system damage and kidney stones, and eventually prove fatal in some cases. According to the FDA regulations for the interim safety and risk assessment of melamine, a maximum level of 2.5 ppm (20 μM) melamine can be present in food for human consumption.^15–17 Although several efficient methods such as capillary electrophoresis, mass spectrometry, gas chromatography/mass spectrometry, surface-enhanced Raman scattering spectrometry, electrochemistry, and immunosensing have been used for the determination of melamine, they are unsuitable for on-site analysis because of their long time requirements, high cost, and complex procedures for sample pretreatment.^18–22

Fluorescent gold nanoparticles (AuNPs), often called Au nanoclusters (AuNCs), provide advantages such as favourable water solubility, low toxicity, remarkable photophysical properties, and facile surface modification, rendering them suitable for sensors and bioimaging.^23–26 Increasing attention is being paid to bovine serum albumin (BSA)-based synthesis approaches for preparing fluorescence AuNCs.^27–30 During the preparation process, BSA first coordinates with Au³⁺ through cysteine and histidine residues acting as templates. The tyrosine residue of the protein under basic conditions (pH value > 10.0) is then considered to favour the reduction of Au³⁺ ions to Au atoms; that is, the proteins act as reducing and stabilising agents. Because the as-prepared AuNCs possess favourable photophysical properties such as a long fluorescence lifetime (>20 ns), large Stoke's shift (>100 nm), and high quantum yield, they can be used to develop sensitive fluorescent sensors.^31–33 Xie and colleagues first demonstrated a simple method for the preparation of fluorescent BSA-AuNCs.²³ The prepared AuNCs consisted of 25 Au atoms with red emission fluorescence. The surface of the BSA-AuNCs was stabilised with a small amount of Au⁺, which has a high affinity metallophilic interaction with Hg²⁺.²⁴ Consequently, it effectively quenched the fluorescence of BSA-AuNCs. The BSA-AuNCs showed a remarkably high sensitivity for Hg²⁺ with a limit of detection (LOD) as low as 0.5 nM. Chen's group prepared Au clusters (Au₂₅) protected by BSA through the same procedure with a slight modification.³⁴ They first observed the electrogenerated chemiluminescence from BSA-AuNCs, then proposed the electrogenerated chemiluminescence mechanism and demonstrated the potential application of the method for Pb²⁺ detection. Liu's group proposed that the fluorescence quenching of the BSA-AuNCs was induced by Cu²⁺ ions, which chelated with glycine in the BSA to generate BSA-AuNCs–Cu²⁺, whereas the fluorescence restoration of the BSA-AuNCs was caused by the pyrophosphate ion chelating with Cu²⁺, resulting in Cu²⁺ being removed from the BSA-AuNCs surface.³⁵ Thus, the fluorescence turn-off sensing of Cu²⁺ and turn-on sensing of pyrophosphate ions were demonstrated. A BSA-AuNCs fluorescent sensor for cyanide in aqueous solution on the basis of cyanide etching-induced fluorescence quenching was reported by Lu's group.³⁶ With this sensor, the LOD for cyanide ions could be reduced to 200 nM, which is approximately 14 times lower than the maximum level (2.7 μM) of cyanide in drinking water permitted by the World Health Organization (WHO). Although BSA-AuNCs are sensitive and selective for the detection of cations and anions, their preparation requires a long reaction time.³⁷ Since 2011, microwave (MW)-assisted synthesis of BSA-AuNCs has attracted considerable attention because of its uniform heating and rapid reaction time; the reaction time can be shortened from several hours to minutes.³⁸ The prepared BSA-AuNCs display a strong red emission that is quenched by Cu²⁺ ions on the basis of the aggregation of BSA-AuNCs induced by the interaction between BSA and the Cu²⁺ ions. In addition, the fluorescence quenching of the MW-assisted synthesis of HSA-AuNCs occurs through nitrogen oxides, demonstrating its great potential in determining the intracellular concentration of nitrogen oxides by Xiao's group.³⁹ Wu's group also demonstrated a one-step MW-assisted method for the synthesis of Au₁₆NCs@BSA and their fluorescence-enhanced sensing of Ag⁺ ions.^40,41 However, expensive instruments were required for this synthesis.

In this study, MW-assisted synthesis of fluorescence BSA-AuNCs (MW_BSA-AuNCs) was demonstrated along with their fluorescence turn-off sensing of Pb²⁺ and turn-on sensing of melamine in an aqueous solution. The red emitting BSA-AuNCs that could be synthesised by a domestic MW oven (120 W) with a 4 min MW programme are the highlight of this work. UV-Vis spectroscopy, time-resolved fluorescence spectroscopy, fluorescence spectroscopy, transmission electron microscope (TEM), energy dispersive X-ray spectroscopy (EDS), and a dynamic light scattering spectrophotometer (DLS) were used to characterise the MW_BSA-AuNCs in the absence and presence of Pb²⁺ ions. Finally, the potential use of MW_BSA-AuNCs for determining Pb²⁺ and melamine in aqueous solution was investigated.

Experimental sections

Chemicals

All the chemicals used were of analytical grade or of the highest available purity. Hydrogen tetrachloroaurate(III) trihydrate (HAuCl₄·3H₂O), BSA, sodium hydroxide (NaOH), sodium hydrogen phosphate (NaH₂PO₃), disodium hydrogen phosphate (Na₂HPO₃), quinine sulphate, sulfuric acid (H₂SO₄), ethylenediaminetetraacetic acid (EDTA), adenine, thymine, glycine (Gly), histidine (His), cysteine (Cys), Pb(NO₃)₂, Ni(NO₃)₂·6H₂O, NaCl, CaCl₂·2H₂O, Sr(NO₃)₂, BaCl₂·6H₂O, Cd(NO₃)₂·4H₂O, Mg(NO₃)₂·6H₂O, FeCl₃, ZnCl₂, HgCl₂, Cu(NO₃)₂·2.5H₂O, and KCl were obtained from Sigma Aldrich (St. Louis, MO, USA). Melamine was purchased from Acros Organic (New Jersey, USA). NaH₂PO₃ (0.05 M) and Na₂HPO₃ (0.05 M) were prepared in phosphate buffer solution (PBS; 0.05 M; pH 7.0). Deionised water (18.2 MΩ cm) was used to prepare all the aqueous solutions.

MW_BSA-AuNCs synthesis

The MW_BSA-AuNCs was synthesized based on the earlier report with modifications.²⁷ In an MW-assisted synthesis experiment, 5.0 mL of 65 mg mL⁻¹ BSA solution was added to 5.0 mL of 10 mM HAuCl₄ solution, followed by 1.0 mL of 1.0 M NaOH solution. The mixture solution was temporarily paused and heated using a domestic MW oven (120 W) for 4 min, which turned the solution from light yellow to dark brown. Under UV illumination, the solution showed red fluorescence emission, indicating the formation of MW_BSA-AuNCs. For the sake of simplicity, the concentration of the as-prepared MW_BSA-AuNCs is presented as 1×.

Instrumentation and characterisation

A UV-Vis spectrometer (Evolution 200; Thermo Fisher, Waltham, MA, USA) and a fluorescence spectrometer (Varian, CA, USA) were used to measure the optical properties of the MW_BSA-AuNCs. A JEOL-1200EX II (JEOL, Tokyo, Japan) TEM system was used to measure the size of the MW_BSA-AuNCs. A dynamic light scattering (DLS) spectrophotometer (SZ-100, Horiba, Kyoto, Japan) was used to measure the hydrodynamic radius of the MW_BSA-AuNCs. The time-resolved fluorescence spectra of the MW_BSA-AuNCs were measured using a time-correlated photon counting spectrometer, the details for which have been reported previously.^42,43 For the fluorescence lifetime measurement, the polarisation of the excitation light was set at the vertical position (relative to the optical table) and the angle of the emission polariser was set to 54.7° relative to the excitation light.

General procedure for sensing Pb²⁺ ions

The MW_BSA-AuNCs was added to 25 mM PB buffer solutions (pH 7.0) containing Pb²⁺ ions (0–1.0 μM) and other metal ions (10.0 μM). The final concentration of the BSA-AuNCs was 0.1×. The mixtures were equilibrated for 5 min and then were subjected to fluorescence, TEM, EDS (Oxford Instruments, Abingdon, UK) measurements.

For practicality, determination of Pb²⁺ ions concentration in the complicated samples (seawater, and urine) is conducted by the proposed MW_BSA-AuNCs. For a biological sample, all ethics statements and consent were written and signed. We declared that all experiments were performed in compliance with the relevant laws and institutional guidelines (DOH Regulation no. 0950206912, Taiwan) and carried out according to approved protocol by the Bioethical committee of the National Changhua University of Education. A statement that informed consent was obtained from a male donor. A urine donor was healthy according to WHO standards.

Results and discussion

MW-assisted synthesis and characterisation of BSA-AuNCs

The concentrations of BSA, HAuCl₄, and NaOH being the same as the earlier report were used to synthesis MW_BSA-AuNCs in this study. The MW programme was modified. We found that the increased pause time from 1 min to 2 min not only increased the numbers of MW irradiation but also improved the fluorescence intensity of the MW_BSA-AuNCs. As a result, the fluorescence intensity of the MW_BSA-AuNCs in this study was higher than that of the MW_BSA-AuNCs in the earlier report (1.5 times).²⁷ Next, different synthesis procedures of BSA-AuNCs are discussed.

Fig. 1 displays the optical properties of the BSA-AuNCs synthesised under different synthesis procedures. The black, red, and blue curves in Fig. 1 represent the BSA-AuNCs synthesised under ambient conditions for 5 h (RT_BSA-AuNCs), heating in an oven at 70 °C for 20 min (H_BSA-AuNCs), and MW irradiation at 120 W for 4 min (MW_BSA-AuNCs), respectively. The reaction mixture changed from yellow to dark brown after MW irradiation. The optical absorption spectra show a gradual rise at 500 nm with a continuous increase in absorbance afterward (Fig. 1A). The MW_BSA-AuNCs exhibit comparatively higher electronic absorption features than do the RT_BSA-AuNCs and H_BSA-AuNCs. The absence of the typical surface plasmon resonance peak for larger AuNPs (520 nm) indicates that most of these BSA-AuNCs had dimensions of less than 15 nm. Fig. 1B displays the fluorescence spectra of these BSA-AuNCs synthesised using different procedures. In this study, the MW irradiation was temporarily paused to prevent overheating of the reaction mixture. MW_BSA-AuNCs was synthesised using a 4 min MW programme which consisted of five 2.0 min pause between six 40 s MW irradiation. The final temperature of MW_BSA-AuNCs was 75 °C. A higher reaction temperature yields higher-quality BSA-AuNCs,³⁷ and higher emission intensity was observed for the MW_BSA-AuNCs than for the RT_BSA-AuNCs and H_BSA-AuNCs. When a reaction temperature higher than 75 °C was reached, BSA denaturation occurred and formed some insoluble gel such that a lower yield of fluorescent BSA-AuNCs was found. Therefore, an MW procedure for the preparation of the MW_BSA-AuNCs was selected for further investigation.

Fig. 1 (A) Absorption and (B) fluorescence spectra of BSA-AuNCs synthesised (a) in ambient conditions for 5 h (RT_BSA-AuNCs, black curve), (b) through heating in an oven at 70 °C for 20 min (H_BSA-AuNCs, red curve) and (c) using MW irradiation at 120 W for 4 min (MW_BSA-AuNCs, blue curve). The inset displays photographs of (a) RT_BSA-AuNCs, (b) H_BSA-AuNCs, and (c) MW_BSA-AuNCs under daylight and UV light (λ_ex: 365 nm), respectively.

Fig. 2 represents the fluorescence intensity of the BSA-AuNCs synthesised with different reaction times. Whereas RT_BSA-AuNCs and H_BSA-AuNCs required at least 5 h and 20 min to complete, respectively, MW_BSA-AuNCs required less than 5 min. The fluorescence intensity of the MW_BSA-AuNCs peaked approximately 4 min after the reaction mixtures were heated at 75 °C using the MW irradiation method. The results indicate that the established MW irradiation method yields a rapid and high elevation of temperature and homogeneous heating, which favours the reduction of the metal precursors and the nucleation of AuNCs, consequently producing a markedly enhanced fraction of fluorescent AuNC species.^38,39 In summary, we concluded that MW irradiation for BSA-AuNCs synthesis has the advantages of saving time, high fluorescence intensity, and being more convenient than other methods.

Fig. 2 Plot of time-dependent fluorescence intensity of MW_BSA-AuNCs at 660 nm for different synthesis routes: (●) under ambient conditions, (■) heating in an oven at 70 °C, and (▲) MW irradiation at 120 W. The inset shows an expanded view of the results found with MW exposure time.

In this study, the as-prepared MW_BSA-AuNCs exhibited red emission (λ_em 660 nm) when excited at 360 nm, which showed a longer stroke shift (300 nm). The quantum yield (QY) of the as-prepared MW_BSA-AuNCs was determined to be 2.1% according to a comparative method using quinine sulphate as a reference.²⁹ Fig. S1† shows the photographic images of precipitating MW_BSA-AuNCs in an acidic acetate buffer (pH 4.7) and its redissolution in an alkaline PB buffer (pH 9.0). Both the precipitate and solution of MW_BSA-AuNCs exhibit red emission under UV irradiation. The fluorescence intensity of the MW_BSA-AuNCs solution maintains 90% of its initial solution, indicating that the precipitate and solution BSA-AuNCs can be easily obtained by controlling the pH of the solution (Fig. S1†). This property is beneficial for maintaining the MW_BSA-AuNCs in the required form.

Sensing approach

Curve (a) in Fig. 3A displays the fluorescence spectrum of the MW_BSA-AuNCs in 25 mM PB buffer solution (pH 7.0); the maximum wavelengths of the fluorescence appear at 660 nm. We then investigated the effect of Pb²⁺ on the fluorescence of MW_BSA-AuNCs in PB buffer solution. As shown in curve (b) in Fig. 3A, the fluorescence quenching of the MW_BSA-AuNCs was detected through a Pb²⁺-mediated interparticle aggregation mechanism, considering that the residual amino, carboxylic, mercapto groups of BSA (17 glycine, 35 cysteine, and 59 glutamate residues) could easily interact with Pb²⁺.^44–47 Fig. S2† shows that EDTA could recover the quenched fluorescence of MW_BSA-AuNCs through competitive interaction with Pb²⁺. This indicates that the fluorescence quenching of MW_BSA-AuNCs toward Pb²⁺ is related to Pb²⁺–BSA interaction. In addition, we found high electronic absorption features and strong scattering for the MW_BSA-AuNCs in the presence of Pb²⁺ ions (Fig. 3B). Moreover, comparing the TEM images (Fig. 3C) of the MW_BSA-AuNCs and MW_BSA-AuNCs/Pb²⁺ revealed that the average size of MW_BSA-AuNCs/Pb²⁺ (35.3 ± 9.3 nm) is clearly larger than that of MW_BSA-AuNCs (5.8 ± 1.1 nm). The EDS spectrum displayed in Fig. 3D peaked from Au and Pb, thus proving the interaction of MW_BSA-AuNCs and Pb²⁺ ions. The DLS results also showed that the average hydrodynamic diameter of MW_BSA-AuNCs increased from 15.1 ± 2.1 nm to 63.4 ± 3.6 nm upon addition of Pb²⁺, thus proving the possibility that the Pb²⁺-induced aggregation of MW_BSA-AuNCs was responsible for the fluorescence quenching.

Fig. 3 (A) Fluorescence, (B) absorption spectra, and (C) TEM images of MW_BSA-AuNCs (a) without and (b) with Pb²⁺ ions (1.0 μM). The inset displays photographs of MW_BSA-AuNCs prepared in the same conditions (λ_ex: 365 nm). (D) EDS spectrum of MW_BSA-AuNCs with Pb²⁺ ions.

The quenching constant (K_sv) is calculated by the Stern–Volmer equation.

I_F0/I_F = 1 + K_sv[Q]

where I_F0 and I_F represent the fluorescence intensities at 660 nm in the absence and presence of Pb²⁺ ions, respectively. K_sv is the Stern–Volmer quenching constant and Q is the concentration of Pb²⁺ ions. The Stern–Volmer plot shows a straight line (Fig. S3†). According to the Stern–Volmer equation, the estimated K_sv of Pb²⁺ ions of 6.42 × 10⁶ M⁻¹ confirms a strong fluorescence quenching of MW_BSA-AuNCs in the presence of Pb²⁺ ions. Moreover, the fluorescence decay experiments showed that the lifetime of MW_BSA_AuNCs remained essentially unchanged upon addition of Pb²⁺ (τ₁: 4.2/4.2 ns, τ₂: 54/56 ns, τ₃: 1.23/1.25 μs), thus suggesting that the quenching is static.

Sensing system optimisation

The effects of several factors, including the buffer system, pH, and concentration of the buffer solution on the sensing system containing 75 nM Pb²⁺ were assessed to optimise the assay. The fluorescence quenching of MW_BSA-AuNCs by Pb²⁺ was measured under different buffer solution systems including HEPES–NaOH, PB, and Tris–HCl buffers. Fig. S4A† shows the effect of the buffer system on the values of (I_F0 − I_F)/I_F0, where I_F and I_F0 represent the fluorescence at 660 nm of BSA-AuNCs in the presence and absence of Pb²⁺, respectively. The maximum values of (I_F0 − I_F)/I_F0 for Pb²⁺ ion detection were obtained when the PB buffer was used. Therefore, the PB buffer system was selected for further study. The influence of pH value for the PB buffer solution was then investigated over the range of 5.0–9.0. As shown in Fig. S4B,† the values of (I_F0 − I_F)/I_F0 for the MW_BSA-AuNCs remain constant over the pH range from 5.0 to 9.0. Therefore, a PB buffer solution at pH 7.0 was used for all subsequent experiments. In addition, the influence of the PB concentration on the system responses was investigated from 5.0 to 100 mM (Fig. S4C†). The value of (I_F0 − I_F)/I_F0 of the MW_BSA-AuNCs solution decreased with an increase in the concentration of PB buffer, probably due to the MW_BSA-AuNCs aggregation at higher buffer concentration. Therefore, a PB concentration of 25 mM was selected as the optimal value in the follow study.

Sensing system selectivity and sensitivity

To test the selectivity of the MW_BSA-AuNC probe towards Pb²⁺ ions, we conducted experiments similar to those used to obtain Fig. 4A, but with the addition of various other metal ions including Na⁺, K⁺, Ba²⁺, Mg²⁺, Ca²⁺, Sr²⁺, Cd²⁺, Ni²⁺, Cu²⁺, Hg²⁺, Fe³⁺, and Zn²⁺ (each at a concentration of 10 μM). When compared with the effect ((I_F0 − I_F)/I_F0 = 0.89) of 1.0 μM Pb²⁺, the changes induced by the other metal ions were much smaller except for that of Hg²⁺ (white bar in Fig. 4A). The result indicated that the fluorescence of MW_BSA-AuNCs could be quenched through metallophilic Hg²⁺–Au⁺ interactions.²⁴ Greater selectivity of the MW_BSA-AuNC probe towards Pb²⁺ ions was readily achieved in the presence of the masking agent (melamine, 100 μM). Melamine, which has a multi-nitrogen heterocyclic ring, is prone to coordinate with Hg²⁺.^48–51 To ensure improved masking and the formation of stable complexes with Hg²⁺, melamine was added to each MW_BSA-AuNCs solution at a minimal concentration of approximately 10 times greater than that (10 μM) of Hg²⁺. As depicted by the black bar in Fig. 4A, MW_BSA-AuNCs in 25 mM PB buffer (pH 7.0) containing 100 μM melamine responded selectively towards Pb²⁺ ions relative to the other metal ions. We also measured the fluorescence quenching kinetics of MW_BSA-AuNCs after addition of different metal ions. Fig. S5† shows that plots of the time dependent fluorescence ratios (I_F/I_F0) over 30 min of MW_BSA-AuNCs with metal ions and melamine were obtained. As shown in Fig. S5,† 10 μM of Na⁺, K⁺, Ba²⁺, Mg²⁺, Ca²⁺, Sr²⁺, and Hg²⁺ did not change the fluorescence of the MW_BSA-AuNCs. In contrast, Zn²⁺ and Cd²⁺ slightly enhanced the fluorescence of the MW_BSA-AuNCs. This may be because Zn²⁺ and Cd²⁺, with similar outer electronic orbit with Au⁺, have filled their d-orbitals and then repaired the lattice defect of MW_BSA-AuNCs.^44,52 For the other ions (Cu²⁺, Fe³⁺, Ni²⁺, and Pb²⁺), the fluorescence intensities were decreased with the incubation times. According to literature, Cu²⁺ ions also could chelated with glycine in BSA and generated BSA-AuNCs–Cu²⁺, leading to fluorescence quenching through a Cu²⁺-mediated interparticle aggregation.^35,38,53 In all ions, Pb²⁺ ions exhibited an obvious fluorescence decreasing within 5 min comparing to that of the other metal ions. We believe that the explanation of selectivity is probably because the fluorescence quenching rates of Pb²⁺ ions were faster than that of other ions.

Fig. 4 (A) Selectivity and (B) tolerance of the MW_BSA-AuNC probe in the absence and presence of masking agent (melamine, 100 μM) for different metal ions (10 μM) and Pb²⁺ ions (1.0 μM) in 25 mM PB buffer (pH 7.0). The error bars represent the standard deviations from three repeated experiments. Inset: photographs of the MW_BSA-AuNC probes with different metal ions (λ_ex: 365 nm).

To further test the practicality of using the probe, we conducted measurements in mixtures containing 1.0 μM Pb²⁺ and various interfering ions (Na⁺, K⁺, Ba²⁺, Mg²⁺, Ca²⁺, Sr²⁺, Cd²⁺, Ni²⁺, Cu²⁺, Hg²⁺, Fe³⁺, and Zn²⁺, each at a concentration of 10 μM; i.e., 10-fold more concentrated) as shown in Fig. 4B. The results confirm the high selectivity and tolerance of the MW_BSA-AuNCs toward Pb²⁺ ions. Other metal ions had minor or negligible quenching effects on the fluorescence intensity of the MW_BSA-AuNCs.

The fluorescence spectra of the MW_BSA-AuNCs solutions containing different Pb²⁺ concentrations under optimal conditions were also recorded for a quantitative assay. The fluorescence intensity of the MW_BSA-AuNCs clearly decreased as Pb²⁺ concentration increased (Fig. 5A). A linear relationship was found between the quenching effect and the Pb²⁺ concentration over the range 10–100 nM, and the linear correlation coefficient was 0.959 (Fig. 5B). The LOD for Pb²⁺, defined as the concentration giving a signal-to-noise ratio of 3, was found using this technique to be 4.8 nM, suggesting that the probe is well suited for monitoring Pb²⁺ ions in environmental water samples because the highest Pb²⁺ concentration permitted in drinking water by the U.S. Environmental Protection Agency is 75 nM.⁵⁴ The MW_BSA-AuNCs probe possesses some attractive features comparing to previous method: (1) environmental friendly reagents—toxic reagents (tetrakis(hydroxymethyl)phosphonium chloride and Cd²⁺ ions) are not used;^45,47 (2) low cost—expensive enzymes, DNA, and other recognition elements are not required.^55–59

Fig. 5 (A) Fluorescence spectra and (B) values of (I_F0 − I_F)/I_F0 for the responses of MW_BSA-AuNCs for different concentrations of Pb²⁺ ions. Inset: photographs of the BSA-AuNC probes with different concentrations of Pb²⁺ ions (λ_ex: 365 nm). The error bars represent the standard deviations from three repeated experiments.

Because of the high affinity of Hg²⁺ to melamine, the fluorescence quenching by Hg²⁺ ions can be inhibited in the presence of melamine. Moreover, in the absence of Hg²⁺ ions, the fluorescence intensity of the MW_BSA-AuNCs was not influenced by the addition of melamine. In addition, melamine is unable to recover the fluorescence of the MW_BSA-AuNCs when it is added to the MW_BSA-AuNC solution premixed with Hg²⁺. Therefore, the melamine detection assay procedure includes the addition of MW_BSA-AuNCs to the blended solution containing Hg²⁺ and melamine. To get the maximum response for melamine detection in aqueous samples, the concentration of Hg²⁺ and solution pH were optimised as 20 μM and pH 7.0, respectively. Fig. 6 shows that the fluorescence intensity of MW_BSA-AuNCs at 660 nm increased with melamine concentration, indicating that the addition of melamine led to the coordination with Hg²⁺, preventing the fluorescence quenching induced by Hg²⁺. As shown in Fig. 6, the MW_BSA-AuNC–Hg²⁺ probe provided a favourable linear range of 10–200 μM (the linear correlation coefficient was 0.992) and LOD (signal-to-noise ratio of 3) of 2.94 μM for melamine, suggesting that the probe is well suited for monitoring melamine in real samples because the safety level for melamine permitted by the FDA is 2.5 ppm (20 μM).¹⁵ The effects of several factors could be discussed to improve the linear range and LOD for melamine detection, such as concentration of Hg²⁺, reaction time and temperature, pH effect and buffer system selection.

Fig. 6 Fluorescence spectra for the responses of MW_BSA-AuNCs–Hg²⁺ for different concentrations of melamine. Inset: values of (I_F − I_F0)/I_F0 and photographs of the MW_BSA-AuNC–Hg²⁺ probes with different concentrations of melamine (λ_ex: 365 nm), where I_F and I_F0 represent the fluorescence at 660 nm of MW_BSA-AuNCs–Hg²⁺ probe in the presence and absence of melamine, respectively. The error bars represent the standard deviations from three repeated experiments.

The selectivity of the MW_BSA-AuNC–Hg²⁺ probe towards melamine was performed. As shown in Fig. S6,† the nitrogen-containing organic molecules (adenine, thymine, Gly, and His) also shows the ability to coordinate with Hg²⁺. Compared with the values of (I_F − I_F0)/I_F0 for melamine, these organic molecules coordinated with Hg²⁺ is weaker, but they could also affect the recovery of fluorescence. In addition, Cys with sulfydryl and amine groups has a strong coordination interaction with Hg²⁺ and thus the value of (I_F − I_F0)/I_F0 is much higher than that of other interferences. Because the proteins can be removed by the pretreatment of milk samples, the interferences (Cys and other nitrogen-containing molecules) can be avoided for melamine detection in milk samples.

Applications in real samples

We predicted that our MW_BSA-AuNCs-based probe would have great potential for use in the analysis of Pb²⁺ in environmental water and biological samples. We filtered a water sample from the coast in Taichung City in Central Taiwan and a urine sample from a healthy male through a 0.2 μm membrane, and then subjected the filtrate to ICP-MS analysis; no contamination of Pb²⁺ was determined. These trace Pb²⁺ ion concentrations in seawater and urine were below the detection limit of our present method when 0.5 mL of a water sample was added to 0.5 mL of 50 mM PB buffer solution containing MW_BSA-AuNC probes. To demonstrate the feasibility of this approach for detecting Pb²⁺ ions in complicated seawater and urine matrices, we applied a standard addition method to determine the concentration of Pb²⁺ in the samples. A linear correlation existed between the (I_F0 − I_F)/I_F0 value and the spiked Pb²⁺ concentration in each sample over the range of 10–100 nM. The recoveries of these measurements were 96.5–110%, and the LOD (for a signal-to-noise ratio of 3) for Pb²⁺ in the seawater and urine matrices was 5.2 and 6.5 nM, respectively. Our results suggest that this probe can be useful for detecting environmentally and biologically relevant Pb²⁺ concentrations. To prove the practicality of the MW_BSA-AuNC–Hg²⁺ probe for analysing melamine in milk samples, we spiked different amounts of melamine standard solution (25 and 50 μM) into real samples. The favourable recoveries ranged from 92% to 113% with an RSD of approximately 3.5%, indicating the accuracy and reliability of the MW_BSA-AuNC probe for the detection of melamine in practical applications.

Conclusions

We prepared fluorescent MW_BSA-AuNCs by using a domestic MW oven for 4 min. To prevent overheating, the MW irradiation was temporarily paused. The fluorescent MW_BSA-AuNCs were sensitive for the detection of Pb²⁺ ions through a Pb²⁺-mediated interparticle aggregation mechanism, which effectively statically quenched the fluorescence of AuNCs. In addition, the fluorescence of the MW_BSA-AuNCs was also quenched on the basis of the high-affinity metallophilic Hg²⁺–Au⁺ interaction, whereas the fluorescence of the MW_BSA-AuNCs was restored when melamine was added and coordinated with Hg²⁺. Our results show that MW_BSA-AuNCs possess great potential for the detection of Pb²⁺ ions and melamine in real aqueous solutions such as seawater, urine, and milk. Further research examining the application of this simple MW-assisted synthesis for the formation of other protein-AuNCs, such as in the detection of antibodies and receptors, is currently under way in our laboratory.

Acknowledgements

This study was supported by Ministry of Science and Technology under contracts (MOST 103-2113-M-018-001-MY2) and (MOST 105-2113-M-018-007). We thank Wallace Academic Editing for the English language editing.

Notes and references

1. L. Y. Cabrera, J. Tesluk, M. Chakraborti, R. Matthews and J. Illes, Environ. Health, 2016, 15, 1–5.
2. Y. Yang, X. Z. Zhang, Y. W. Fu and H. Yang, Toxicol. Lett., 2014, 227, 172–178.
3. P. Grandjean and P. J. Landrigan, Lancet Neurol., 2014, 13, 330–338.
4. M. Meadows-Oliver, J. Pediatr. Health Care, 2012, 26, 213–215.
5. M. P. Iqbal, Pak. J. Pharm. Sci., 2012, 25, 289–294.
6. A. Pena-Fernandez, M. C. Lobo-Bedmar and M. J. Gonzalez-Munoz, Environ. Int., 2014, 72, 170–175.
7. J. H. Liu and G. Lewis, J. Environ. Health, 2014, 76, 130–138.
8. W. J. Luo, D. Y. Ruan, C. H. Yan, S. T. Yin and J. Y. Chen, Neurotoxicology, 2012, 33, 862–871.
9. C. Sobin, N. Parisi, T. Schaub, M. Gutierrez and A. X. Ortega, Arch. Environ. Contam. Toxicol., 2011, 61, 521–529.
10. M. Trzcinka-Ochocka, R. Brodzka and B. Janasik, J. Clin. Lab. Anal., 2016, 30, 130–139.
11. Q. Shan, X. L. Zhang, Y. Zhang, W. B. Jia, Y. S. Ling, D. Q. Hei and S. N. Chu, Spectrosc. Lett., 2016, 49, 188–193.
12. U. A. Barbosa, I. F. dos Santos, A. M. P. dos Santos and S. L. C. Ferreira, Anal. Lett., 2016, 49, 799–807.
13. Y. W. Xu, W. Zhang, J. Y. Shi, X. B. Zou, Z. H. Li and Y. D. Zhu, Food Chem., 2016, 201, 190–196.
14. K. Sharma and M. Paradakar, Food Security, 2010, 2, 97–107.
15. J. W. Choi, K. M. Min, S. Hengoju, G. J. Kim, S. I. Chang, A. J. deMello, J. Choo and H. Y. Kim, Biosens. Bioelectron., 2016, 80, 182–186.
16. J. Peng, Y. Feng, X. X. Han and Z. N. Gao, Anal. Methods, 2016, 8, 2526–2532.
17. Y. J. Dong, S. C. Zhao, Z. M. Li, H. Yue, Y. Wang, L. G. Deng and S. Q. Zhang, Int. J. Electrochem. Sci., 2016, 11, 1765–1776.
18. Y. L. Zhang, L. J. Chen, C. Zhang, S. T. Liu, H. K. Zhu and Y. M. Wang, Talanta, 2016, 150, 375–387.
19. Y. M. Fan, X. G. Ma, Z. Y. Li and M. Q. Chen, Food Analytical Methods, 2016, 9, 16–22.
20. Q. T. Zhou, G. W. Meng, N. Q. Wu, N. N. Zhou, B. S. Chen, F. D. Li and Q. Huang, Sens. Actuators, B, 2016, 223, 447–452.
21. K. Rovina and S. Siddiquee, Food Control, 2016, 59, 801–808.
22. H. L. Guo, X. H. Zhou, Y. Zhang, B. D. Song, J. X. Zhang and H. C. Shi, Food Chem., 2016, 197, 359–366.
23. J. P. Xie, Y. G. Zheng and J. Y. Ying, J. Am. Chem. Soc., 2009, 131, 888–889.
24. J. P. Xie, Y. G. Zheng and J. Y. Ying, Chem. Commun., 2010, 46, 961–963.
25. J. J. Li, J. J. Zhu and K. Xu, Trends Anal. Chem., 2014, 58, 90–98.
26. L. B. Zhang and E. K. Wang, Nano Today, 2014, 9, 132–157.
27. N. Y. Hsu and Y. W. Lin, New J. Chem., 2016, 40, 1155–1161.
28. X. X. Chen, J. T. Zhao, W. S. Cong, X. X. Li, H. Y. Fan, J. Z. Sun, J. Lin, B. Li, Y. X. Gao, C. Qin and Y. F. Li, J. Nanosci. Nanotechnol., 2016, 16, 772–776.
29. X. Le Guevel, B. Hotzer, G. Jung, K. Hollemeyer, V. Trouillet and M. Schneider, J. Phys. Chem. C, 2011, 115, 10955–10963.
30. D. H. Hu, Z. H. Sheng, P. Gong, P. F. Zhang and L. T. Cai, Analyst, 2010, 135, 1411–1416.
31. R. Chib, S. Butler, S. Raut, S. Shah, J. Borejdo, Z. Gryczynski and I. Gryczynski, J. Lumin., 2015, 168, 62–68.
32. P. P. Xu, R. P. Li, Y. F. Tu and J. L. Yan, Talanta, 2015, 144, 704–709.
33. M. L. Cui, J. M. Liu, X. X. Wang, L. P. Lin, L. Jiao, Z. Y. Zheng, L. H. Zhang and S. L. Jiang, Sens. Actuators, B, 2013, 188, 53–58.
34. Y. M. Fang, J. Song, J. A. Li, Y. W. Wang, H. H. Yang, J. J. Sun and G. N. Chen, Chem. Commun., 2011, 47, 2369–2371.
35. J. M. Liu, M. L. Cui, S. L. Jiang, X. X. Wang, L. P. Lin, L. Jiao, L. H. Zhang and Z. Y. Zheng, Anal. Methods, 2013, 5, 3942–3947.
36. Y. L. Liu, K. L. Ai, X. L. Cheng, L. H. Huo and L. H. Lu, Adv. Funct. Mater., 2010, 20, 951–956.
37. P. C. Chen, C. K. Chiang and H. T. Chang, J. Nanopart. Res., 2013, 15, 1336–1345.
38. D. F. He, Y. Xiang, X. Wang and X. F. Yu, Mater. Res. Bull., 2011, 46, 2418–2421.
39. L. Yan, Y. Q. Cai, B. Z. Zheng, H. Y. Yuan, Y. Guo, D. Xiao and M. M. F. Choi, J. Mater. Chem., 2012, 22, 1000–1005.
40. H. W. Li, Y. Yue, T. Y. Liu, D. M. Li and Y. Q. Wu, J. Phys. Chem. C, 2013, 117, 16159–16165.
41. Y. Yue, T. Y. Liu, H. W. Li, Z. Y. Liu and Y. Q. Wu, Nanoscale, 2012, 4, 2251–2254.
42. K. H. Wang and C. W. Chang, Phys. Chem. Chem. Phys., 2015, 17, 23140–23146.
43. H. Y. Weng, K. M. Lee, Y. S. Chen and C. W. Chang, Phys. Chem. Chem. Phys., 2013, 15, 16935–16940.
44. M. Wang, Q. S. Mei, K. Zhang and Z. P. Zhang, Analyst, 2012, 137, 1618–1623.
45. Z. Q. Yuan, M. H. Peng, Y. He and E. S. Yeung, Chem. Commun., 2011, 47, 11981–11983.
46. F. Chai, C. A. Wang, T. T. Wang, L. Li and Z. M. Su, ACS Appl. Mater. Interfaces, 2010, 2, 1466–1470.
47. E. M. Ali, Y. G. Zheng, H. H. Yu and J. Y. Ying, Anal. Chem., 2007, 79, 9452–9458.
48. G. X. Cao, X. M. Wu, Y. M. Dong, Z. J. Li and G. L. Wang, Microchim. Acta, 2016, 183, 441–448.
49. B. Y. Jiang, L. Yu, F. Z. Li and J. Q. Xie, Sens. Actuators, B, 2015, 212, 446–450.
50. H. C. Dai, Y. Shi, Y. L. Wang, Y. J. Sun, J. T. Hu, P. J. Ni and Z. Li, Biosens. Bioelectron., 2014, 53, 76–81.
51. J. J. Du, S. Y. Yin, L. Jiang, B. Ma and X. D. Chen, Chem. Commun., 2013, 49, 4196–4198.
52. H. Shi, M. Y. Ou, J. P. Cao and G. F. Chen, RSC Adv., 2015, 5, 86740–86745.
53. M. Shamsipur, F. Molaabasi, M. Shanehsaz and A. A. Moosavi-Movahedi, Microchim. Acta, 2015, 182, 1131–1141.
54. Y. J. Lan and Y. W. Lin, Anal. Methods, 2014, 6, 7234–7242.
55. L. Q. Guo, D. D. Nie, C. Y. Qiu, Q. S. Zheng, H. Y. Wu, P. R. Ye, Y. L. Hao, F. F. Fu and G. N. Chen, Biosens. Bioelectron., 2012, 35, 123–127.
56. M. Y. Liu, X. H. Lou, J. Du, M. Guan, J. Wang, X. F. Ding and J. L. Zhao, Analyst, 2012, 137, 70–72.
57. C. L. Li, K. T. Liu, Y. W. Lin and H. T. Chang, Anal. Chem., 2011, 83, 225–230.
58. Y. W. Lin, C. W. Liu and H. T. Chang, Talanta, 2011, 84, 324–329.
59. T. Li, S. J. Dong and E. K. Wang, J. Am. Chem. Soc., 2010, 132, 13156–13157.

---

Footnote

† Electronic supplementary information (ESI) available: Fig. S1. Fluorescence spectra of (a) freshly prepared and (b) redispersed solutions of MW_BSA-AuNCs. The inset displays photographs of MW_BSA-AuNCs under different conditions. Fig. S2. Fluorescence spectra of the MW_BSA-AuNCs after adding 1 μM Pb²⁺, and followed by adding 100 μM EDTA. Fig. S3. Stern–Volmer plot for fluorescence quenching of MW_BSA-AuNCs in the presence of Pb²⁺ ions. Fig. S4. Values of (I_F0 − I_F)/I_F0 for the responses of (A) different buffer solutions, (B) the pH values, and (C) the concentrations of buffer solutions (PB buffer) for Pb²⁺ ion sensing. Fig. S5. Plots of time-dependent fluorescence ratios (I_F/I_F0) of MW_BSA-AuNCs with metal ions. Fig. S6. The selectivity of the MW_BSA-AuNCs–Hg²⁺ toward melamine. See DOI: 10.1039/c6ra16043c

---