Background

The main cause of mercury poisoning in humans was polluted natural waters [1]. Hg2+ ion metabolism by aquatic microbes produces methyl mercury, which was a potent neurotoxin associated with cognitive and motion disorders [2]. Therefore, mercury detection methods that are rapid, cost-effective, facile, and applicable to complex environments are necessary. Particularly, nanomaterials with unique optical properties can be employed to develop optical sensors with high sensitivity and selectivity [3]. Semiconductor quantum dots (QDs), fluorescent metal nanoclusters (NCs), noble metal nanoparticles (NPs), and carbon nanodots (CDs) were commonly used in the design of Hg2+ optical sensors because of their distinct properties, such as easy synthesis, high stability, functionalization, and biocompatibility. Many fluorescent sensors for Hg2+ had been reported [4,5,6,7,8]. For example, Huang et al. [9] developed a time-gated Förster resonance energy transfer (FRET) sensor for Hg2+ detection. Moreover, different FRET systems had been developed for the detection for Hg2+ [10,11,12]. Notably, FRET systems could be similarly built using nanoparticles, such as QDs, as well as organic and inorganic NPs [13,14,15]. Among the nanoparticles, g-C3N4 nanosheets had attracted widespread interest [16, 17]. Although g-C3N4 nanosheets have been applied as sensors, a FRET detection system with g-C3N4 nanosheets and CdSe QDs for metal ions has not been reported. FRET-based fluorescence sensing systems offer multiple advantages [18].

In the present study, a new FRET-based fluorescence sensor was developed to detect mercury ions in aqueous media by using g-C3N4 nanosheets and CdSe QDs particles as vehicles. The proposed mechanism was illustrated in Fig. 1.

Fig. 1
figure 1

The mechanisms of FRET-based detection for mercury ions

Methods

Materials

Mercury (II) chloride (HgCl2) was purchased from Tong Ren Chemical Research Institute (Guizhou, China). Urea and CdSe QDs were purchased from Aladdin Reagent Company (Shanghai, China). Other reagents and chemicals were of analytical reagent grade and used without further purification. All solutions were prepared using purified water from a Milli-Q gradient water purification system (Millipore Inc., USA; nominal resistivity 18.2 MÙ cm).

Characterization

An X-ray diffractometer (Rigaku D/max-2400) was used to obtain diffraction patterns. Ultraviolet–visible (UV–vis) spectra were recorded on a UV–vis 800 spectrophotometer at room temperature. Fourier-transform infrared (FTIR) spectra were recorded on a Nicolet-nexus670 spectrometer using KBr. Fluorescence measurements were performed at room temperature with an RF-5301PC fluorescence spectrometer. X-ray photoelectron spectroscopy (XPS) measurements were performed using a multifunctional spectrometer (Thermo Scientific).

Construction of FRET Sensor between the g-C3N4 Nanosheets and CdSe QD Particles

In a typical procedure, g-C3N4 (125 mg, which was synthesized according to our previous report [19]) was dispersed in 250 mL of water (1:1) and ultrasonicated for 5 h at ambient temperature. Then, CdSe QDs (1.838 g, 0.0216 mol) were dissolved in the solution by sonication for 2 h. Given that the amine group on the g-C3N4 nanosheets and CdSe QDs had a carboxyl group, g-C3N4 nanosheets and CdSe QDs nanoparticles would be combined by electrostatic interaction. All solutions were prepared in Milli-Q gradient water (pH = 7). The CdSe QDs/g-C3N4 nanosheet conjugate emission spectra were recorded. All samples were excited at 334 nm, which was near the minimal acceptor absorption.

Fluorescence Detection of Hg2+

Hg2+ was quenched at room temperature in water. During a typical operation, 10 μL of the CdSe QDs/g-C3N4 nanosheet conjugates was added to 3 mL of ultrapure water, and then the calculated amount of Hg2+ was added. The emission spectra of the CdSe QDs/g-C3N4 nanosheet conjugates were recorded 2 mins later at room temperature.

Interference and Competition Analyses

The response of the FRET nanoprobe to other metal ions (Na+, Mg2+, Ca2+, Pb2+, Cr3+, Cd2+, Zn2+, and Cu2+) was studied through fluorescence spectroscopy. Studies were carried out using the CdSe QDs/g-C3N4 nanosheet conjugates emitting at 450 nm. The conjugate solution was placed in a 1-cm optical path quartz fluorescence cuvette. Fluorescence intensity was measured at emission wavelength of 450 nm under excitation wavelength of 334 nm in the presence of each possible interference (32 nM). Competition assays were also performed for all the possible interferences previously analyzed. For competition experiments, 32 nM Hg2+ aqueous solutions were prepared.

Results and Discussion

Characterization

The structure and morphology of g-C3N4 nanosheets were characterized by TEM, XPS, and XRD. The TEM image in Fig. 2a showed that the g-C3N4 nanosheet possessed a graphene-like morphology that mainly consists of a few layers [19]. Figure 2a showed the XRD patterns of the g-C3N4 nanosheets. The strong XRD peak centered at 27.4° corresponded to the typical graphitic interlayer stacking (002) peak of g-C3N4. The small peak at 13.1° corresponded to the periodic in-plain structural packing feature within the sheets [20, 21]. XPS measurement was used to analyze the valence states of g-C3N4 nanosheets. The XPS spectrum in Fig. 2c showed the C–C bonded to N at 284.8 and 288.0 eV, and the N 1 s spectrum was at 397.04 eV. In Fig. 2d, the peak at 811 cm−1 was attributed to the vibration of the triazine ring. The peaks around 1000 cm−1 represented the stretching modes of CN heterocycles, and the peak at 1800 cm−1 corresponded to C–NH–C. The peaks at 300–3600 cm−1 corresponded to N–H and O–H stretching vibrations [22].

Fig. 2
figure 2

Characterization of the as-prepared g-C3N4 nanosheet. a TEM image. b XRD image. c XPS spectrum. d FTIR spectrum

UV-vis and Fluorescence Properties of CdSe QDs/g-C3N4 Nanosheets

Fluorescence and UV-vis absorption spectra were obtained to evaluate the optical properties of CdSe QDs/g-C3N4 nanosheets. As shown in Fig. 3a, a large peak at approximately 334 nm was observed in the UV–vis absorption spectrum. Moreover, the fluorescence emission and excitation peaks were observed at 452 and 334 nm in the synchronous fluorescence spectroscopy in Fig. 3b and were associated with the emission fluorescence and ultraviolet light excitation of nanosheets. The emission peaks showed a shift compared with the pure g-C3N4 nanosheets at 14–16 nm (emission and excitation peaks were observed at 438 and 310 nm as presented in Fig. 3c), which could be ascribed to the FRET. The influence of excitation wavelengths on fluorescence intensities was also confirmed.

Fig. 3
figure 3

The UV-vis absorption and fluorescence spectrum of CdSe QDs/g-C3N4 conjugates

Effect of pH to the Fluorescence of the CdSe QDs/g-C3N4 Nanosheet Conjugates

Figure 4 showed the fluorescence of the CdSe QDs/g-C3N4 nanosheet conjugates at different pH values. The pH value increased from 3 to 7 with the fluorescence intensity. However, the fluorescence intensity gradually decreased when the pH value varied increased from 7 to 10, which could be attributed to the effect of pH on the change in the surface charge owing to protonation–deprotonation due to the existence of amino groups in the structure of g-C3N4 nanosheets. In this study, the CdSe QDs/g-C3N4 nanosheet conjugates were conducted for the detection of Hg2+ ions, and the pH value of 7 was selected as the optimum pH value. The fluorescence emissions were measured at pH 7 containing different concentrations of NaCl to obtain the stability of the CdSe QDs/g-C3N4 nanosheet conjugates under high ionic strength circumstances. Only a slight change was observed under high ionic strength in the fluorescence intensities of the CdSe QDs/g-C3N4 nanosheet conjugates. The result showed that high ionic strength had minimal effects to the fluorescence intensities of the conjugates.

Fig. 4
figure 4

The effect of pH to the fluorescence of the CdSe QDs/g-C3N4 nanosheets conjugates

Selectivity of CdSe QDs/g-C3N4 Nanosheet FRET System in Detecting Mercury Ion

Selectivity is an important parameter of a new sensing system. The selectivity of the CdSe QDs/g-C3N4 nanosheet FRET sensor was evaluated using various metal ions (e.g., Cu2+, Mg2+, Na+, Ca2+, Hg2+, Cr3+, Pb2+, Cd2+, and Zn2+); the results were shown in Fig. 5a. Compared with the blank sample without ions, the fluorescence ratio of Hg2+ increased obviously, while the fluorescence intensity of other metal ions changed slightly or remains the same. These results indicated that the FRET sensor showed more selectivity than the others (Fig. 5b). Thus, the CdSe QDs/g-C3N4 showed high selectivity toward Hg2+. This phenomenon was distinct in comparison with pure g-C3N4 nanosheet, which was selective for Cu2+ and Hg2+ [23, 24].

Fig. 5
figure 5

The selectivity experiments for the CdSe QDs/g-C3N4 nanosheets FRET sensor

Feasibility of the FRET Fluorescence Process in Detecting Hg2+

To study the practicability of the FRET sensor, the CdSe QDs/g-C3N4 nanosheet fluorescence detection of Hg2+ was performed. The presence of Hg2+ resulted in decreased fluorescence intensity as shown in Fig. 6, which illustrated that Hg2+ could effectively quench the FRET sensor. In order to study the sensitivity, the response of the sensor to different Hg2+ concentrations was further evaluated by fluorescence spectroscopy, and the results were shown in Fig. 6a. The fluorescence intensity of g-C3N4 nanosheets gradually decreased with increasing of Hg2+ concentrations. Figure 6b explained that the I/I0 was dependent on the concentration of Hg2+, where I0 and I were the fluorescence intensity in the absence and presence, respectively, of Hg2+. Moreover, the relation of I/I0 between concentrations of Hg2+ was linear, and the equation of linear regression was I = − 9.6 × 107+ 550.5(R2 = 0.9882), as shown in the inset of Fig. 6b. Compared with recently reported luminescence methods, the proposed method had lower detection limit and higher sensitivity [25, 26]. The g-C3N4 nanosheets and CdSe QDs displayed no obvious quenching response to other metal ions apart from Hg2+, which suggested a relatively high selectivity for this method.

Fig. 6
figure 6

The sensing properties of the CdSe QDs/g-C3N4 nanosheets conjugates and the dependence of the fluorescence intensity to the concentration of Hg2+(CHg2+: 1, 0 nM; 2, 4 nM; 3, 8 nM; 4, 12 nM; 5, 16 nM; 6, 20 nM; 7, 24 nM; 8, 28 nM; 9, 32 nM)

The other coexisting cations that affect the detection of mercury ion were detected as well. The response of the CdSe QDs/g-C3N4 nanosheet-based sensing system toward Hg2+ ion in the presence of alkali, alkaline earth, and other transition metal ions was shown in Table 1. The coexistence of most of the metal ions did not interfere with the binding of Hg2+, which indicated that the interference of these coexisting ions on the Hg2+ sensor was negligible.

Table 1 Effect of interfering species on the fluorescence intensity of FRET sensor solution under optimum conditions

In addition, long-term stability is a superior property of sensors. The absorbance and the fluorescence during the continuous investigation every 3 days within 2 weeks indicated that the activity of CdSe QDs/g-C3N4 nanosheets remained above 92% of the initial efficiency though they were stored at ambient environment. The results indicated that the CdSe QDs/g-C3N4 nanosheets as FRET sensors had good long-term stability.

Compared with previous reports concerning fluorescence assays for Hg2+ (results are listed in Table 2), the CdSe QDs/g-C3N4 nanosheet fluorescence probe based on FRET with the concentration of Hg (II) in the range of 0–32 nmol/L at pH = 7 exhibited a limit of detection at 5.3 nmol/L. Thus, our method obtained a superior detection limit and linear range.

Table 2 The comparison of the previously reported and the present flurescence method for mercury ions

Application of the FRET Sensor

The CdSe QDs/g-C3N4 nanosheets as a FRET sensor successfully provided a good platform for detecting Hg2+ in real samples because of their sensitivity and selectivity. Well, lake, and tap waters were selected as real samples for analysis in which the recovery of Hg2+ were in the range of 95.4–101.6% (Table 3). The relative standard deviation (RSD) of Hg2+ was in the range of 0.64–1.72%. The result stated clearly that the designed method can be efficiently used to detect Hg2+ in practical applications. The acceptable values of RSD and relative error confirmed the high sensitivity, high precision, and high reliability of the proposed FRET sensor for Hg2+ determination in practical applications.

Table 3 The results of the determination for mercury ions in practical samples

Conclusions

A FRET-based system was developed for detecting Hg2+ within g-C3N4 nanosheets/CdSe QDs. The detection limit for Hg2+ ion was 5.3 nM, with a linear response ranging from 0 to 32 nM. The applicability of this sensor was demonstrated by measuring the content of Hg2+ in real samples. Given the long-term stability, low cost, and facile preparation of the CdSe QDs/g-C3N4 nanosheet conjugates, the fluorescence assay could be used as an environmental protection sensor. This strategy would provide an alternative approach for constructing FRET-based sensors for Hg2+ in aqueous media, including environmental and biological samples.

Highlights

  1. 1.

    Fluorescence resonance energy transfer (FRET) system was constructed between CdTe quantum dots (QDs) (donor) and g-C3N4 (acceptor) in the presence of Hg2+ for the first time.

  2. 2.

    The nanocomposites of g-C3N4 supported by CdSe QDs (CdSe QDs /g-C3N4) were fabricated through a simple electrostatic interaction route in an aqueous solution.

  3. 3.

    The feasibility of the FRET system as a sensor was demonstrated for detecting Hg (II) in water solution. At pH 7, a linear relationship was observed between the quenched fluorescence intensity of the concentration of Hg (II) in the range of 0–32 nmol/L. The detection limit was 5.3 nmol/L.

  4. 4.

    The novel FRET-based fluorescence detection may provide an attractive assay platform for quantifying Hg2+ in complex water solutions.