1 Introduction

Potassium permanganate (KMnO4) is a strong oxidant and is usually used as a disinfectant and antiseptic; it is widely applied in daily life and industry, e.g., treating polluted water and diseases in fish [1, 2]. MnO4 also is toxic, corrosive, and carcinogenic [3], and excess MnO4 can lead to skin irritation, neurological disorder, respiratory damage, gastrointestinal distress, and even genetic mutation [3, 4], and thus excess MnO4 can be seriously harmful to human health [5]. Accordingly, selective and sensitive determination of MnO4 in polluted water is critical for environmental evaluation [6]. A variety of detection methods have been developed such as flame atomic absorption spectrophotometry [7, 8], inductively coupled plasma-mass spectrometry [9, 10], and electrochemistry [11]. However, most of these techniques require sophisticated equipment and a skilled operator and are time-consuming. By contrast, fluorescent methods merited with simple operations and a fast response. Hence, novel fluorescent sensors to detect trace MnO4 in simulation samples are urgently needed. Currently, the MnO4 determination can be done by precious metal nanoclusters [12], organic layer [13], metal organic framework [14, 15], coordination polymer [16, 17], etc. However, the above fluorescent probes are usually toxic and harmful, complicated in synthesis process or expensive. To overcome the above limitations, it is necessary to seek a convenient, low toxicity, economic fluorescent material with high fluorescence quantum yield.

Carbon dots (CDs) have attracted widespread attention due to their low cytotoxicity, high biocompatibility, good chemical and photo-stability, easy preparation methods, and tunable emission [18,19,20]. They have been widely used in drug delivery [21], sensing [22,23,24], bioimaging [25], photocatalysis [26], electromagnetic composites [27,28,29,30,31,32], and optoelectronic devices [33,34,35,36]. Nitrogen doping is one of the most widely used strategies to improve the optical properties of CDs. Precursor materials for the preparation of nitrogen-doped CDs (N-CDs) include biomass and synthetic chemicals. Biomass stands out due to its low cytotoxicity, favorable biocompatibility, renewability, cost-effectiveness, and environment-friendly nature (relative to synthetic chemicals) [37,38,39,40,41]. Biomass is rich in proteins and carbohydrates, which are sources of nitrogen and carbon for N-CD preparation. Accordingly, N-CDs can be easily prepared from biomass without the addition of nitrogen and passivators. Numerous N-CDs have been reported which were prepared using natural biomass as precursor, including oyster mushroom [42], palm powder [43], leek [44], rice husk [45], and Aegle Marmelos [46].

In this work, a facile one-step hydrothermal method was reported for the synthesis of N-CDs using Auricularia auricula (L.ex Hook.) Underw as the sole precursor. The N-CDs were successfully used in the determination of MnO4 in polluted water based on the inner filter effect (IFE), as displayed in Scheme 1.

Scheme 1
scheme 1

Preparation method for the N-CDs and the detection of MnO4 based on the IFE

2 Experimental

2.1 Materials

The Auricularia auricula (L.ex Hook.) Underw used here was purchased from a local supermarket. Quinine sulfate was purchased from Aladdin. Other analytical-grade chemicals were from Xilong Chemical Co. Ltd. without further purification. Ultrapure water was used throughout. Polluted water was collected from Bosi lake on our campus. Before use, the polluted water was centrifuged at 5000 rpm for 10 min, and then filtered through a 0.2-μm filtration membrane.

2.2 Instrument

Transmission electron microscopy (FEI f 20, USA) was used to obtain the morphological features of the N-CDs. A Bruker D8 Advance X-ray diffractometer (Bruker, Germany) was used to evaluate the crystalline structure. A PerkinElmer Fourier transform infrared spectrometer (PerkinElmer, USA) was used to analyze the surface functional groups. XPS spectra were acquired on an EscaLab 250Xi X-ray photoelectron spectroscopy system (Thermo Fisher Scientific, USA). UV–vis absorption spectra and fluorescence spectra were obtained using a Shimadzu 2550 UV–vis spectrometer (Shimadzu, China) and a Hitachi FL-7000 fluorescence spectrophotometer (Hitachi, Japan), respectively. Fluorescence lifetime measurements were obtained with a Horiba FluoroMax-4 fluorescence spectrophotometer (Horiba, Japan). The analysis of the above data was conducted on origin 9.0.

2.3 Synthesis of N-CDs

Here, 1.0000 g of crushed Auricularia auricula (L.ex Hook.) Underw powder and 20 mL of ultrapure water were transferred into a 100-mL Teflon-lined autoclave and heated at 180 °C for 6 h. After the Teflon-lined autoclave was cooled to room temperature, the brown N-CD solution was first filtered with filter paper and then centrifuged at 5000 rpm for 10 min. The material was then filtered through a 0.2-μm filtration membrane and finally dialyzed with a 500-Da dialysis membrane for 24 h. The pure N-CD solution was dried in a vacuum oven, and a brown solid was obtained. The N-CDs were then dissolved in ultrapure water and stored at 4 °C for later use.

2.4 MnO4 detection based on N-CDs

MnO4 stock solution (3 mM) was prepared and quantitatively diluted with ultrapure water. Next, 200 μL of pure N-CD solution and different amounts of MnO4 solution were transferred into a 1-cm quartz cuvette, and the mixture was diluted with PBS buffer solution (pH = 2) to a final volume of 2 mL, this was then mixed thoroughly. The mixed solution was then incubated for 20 min at room temperature. The fluorescence spectra of the above solutions were recorded from 334 to 550 nm under 324 nm excitation. Each experiment was measured three times.

3 Results and discussion

3.1 Characterization of the N-CDs

N-CDs were synthesized from Auricularia auricula (L.ex Hook.) Underw via a one-step hydrothermal method. A high-resolution transmission electron microscope (HR-TEM) was used to characterize the morphology of the N-CDs. Figure 1a and b show that the N-CDs were spherical with good dispersion in aqueous solution and had a uniform particle size. The average particle size of the N-CDs is about 2.85 nm. There was a diffuse peak centered at 23.5 (2θ value) in the XRD pattern of the N-CDs (Fig. S1), suggesting the amorphous nature of the synthesized N-CDs [47].

Fig. 1
figure 1

a A typical HR-TEM image of N-CDs. b Diameter size distribution. c FT-IR spectrum of N-CDs. d The XPS survey spectrum

The surface functional groups of the N-CDs were identified by XPS and FT-IR. In the FT-IR spectrum of the N-CDs (Fig. 1c), the peaks at 3399 cm−1 and 2933 cm−1 were assigned to the stretching vibrations of O–H/N–H, and C-H. The peaks at 1731–1613 cm−1 and 1415–1376 cm−1 were related to C = O and C-N, respectively. Peaks at 1248 cm−1 and 1036 cm−1 were contributed to the characteristic absorption band of C-O. The surface chemical groups and elemental analysis of N-CDs were further identified by XPS. Figure 1d shows three peaks located at 284.8 eV, 399.8 eV, and 531.7 eV, which illustrated that the N-CDs mainly included C (64.69%), N (9.48%), and O (25.83%). The C1s spectrum of N-CDs (Fig. S2) revealed three peaks at 284.8 eV, 286.2 eV, and 287.8 eV due to C–C/C = C, C-O, and C = O, respectively. The N1s spectrum of N-CDs (Fig. S3) exhibited three fitted peaks at 399.2 eV, 399.8 eV, and 401.2 eV corresponding to N-(C)3, C-N, and N–H, respectively. The high-resolved O1s spectrum (Fig. S4) was well-fitted into two peaks located at 531 eV and 532.6 eV, thus demonstrating the presence of C = O and C-O on the surface of N-CDs. Results of XPS were well consistent with the FT-IR. The FT-IR and XPS data demonstrate that the surface of the N-CDs is rich in carboxyl, amino, hydroxyl, and other hydrophilic groups [48,49,50,51,52,53,54].

3.2 Optical properties of the N-CDs

The optical properties of the N-CDs were explored by UV–vis absorption and fluorescence spectra at room temperature. An obvious absorption peak centered at 284 nm in the UV–vis absorption spectrum (Fig. 2a) could be ascribed to the π-π* transition of C = C bond [55]. The N-CD solution was pale brown and transparent under sunlight and exhibited bright blue light under a 365-nm UV lamp, as shown in Fig. 2a (inset). When the excitation wavelength of N-CDs is at 324 nm, the maximum fluorescence emission peak is centered at 400 nm. The fluorescence spectra of N-CDs were recorded with the increase of the excitation wavelength from 294 to 354 nm in 10-nm increments. The N-CDs have an excitation-dependent emission behavior similar to most of the CDs reported in the literature (Fig. 2b) [48]. The N-CDs exhibited great anti-photobleaching property and photostability under high concentration NaCl solution. The fluorescence intensities of N-CDs changed slightly under 40 min UV irradiation (Fig. S5) and 1 mol·L−1 NaCl solution (Fig. S6). And the fluorescence properties of the N-CD solution stored at 4 °C were almost unchanged for 2 months. Excellent stability of N-CDs facilitates the fluorescence probe in complex matrixes.

Fig. 2
figure 2

a UV–vis absorbance (black line), fluorescence excitation (red line), and emission (blue line) spectra of N-CDs (inset: photographs of N-CDs in aqueous solution under visible light (left) and 365-nm UV lamp (right)). b FL spectra of N-CDs at different λex (294 nm–354 nm)

3.3 Fluorescence quenching mechanism of MnO4

Figure 3a shows considerable overlap between the excitation or emission spectrum of N-CDs and the UV absorption band of MnO4. The inner filter effect (IFE) may be formed between N-CDs and MnO4 leading to fluorescence quenching of N-CDs. The fluorescence lifetimes of N-CDs and N-CDs with 3 μM MnO4 were recorded to further explore the fluorescence quenching mechanism of MnO4. As shown in Fig. S7, the fluorescence lifetime was almost unchanged after the addition of MnO4, which suggested that the fluorescence quenching of N-CDs was not caused by dynamic quenching or fluorescence resonance energy transfer [55, 56]. In addition, the UV absorption spectra of N-CDs obviously changed (Fig. 3b): The intensity of the UV absorption peak (284 nm) of N-CDs increased with an obvious redshift. There were broad absorption peaks in the range of 450–600 nm in the UV absorbance spectrum of N-CDs with MnO4, which indicated the static complex or chelate was formed between MnO4 and N-CDs [49]. In conclusion, the fluorescence quenching of N-CDs was due to static quenching and IFE.

Fig. 3
figure 3

a The UV–vis absorbance of MnO4 (blue line), the fluorescence excitation (red line), and emission (black line) spectra of N-CDs. b The UV–vis absorbance of N-CDs, MnO4, and N-CDs after the addition of MnO4

3.4 Analytical performance of MnO4 detection

3.4.1 Selectivity of the fluorescent probe

To explore the anti-interference performance of the fluorescent probe to the sensing of MnO4, under the same experimental conditions, the effects of some metal cations (Fe3+, Hg2+, Cu2+, Ba2+, Cr3+, K+, Mg2+, Li+, Pb2+, Mn2+, Na+, and Ca2+) and anions (CO32−, SO42−, F, Cl, Br, I, C2O42−, HPO42−, S2O82−, and NO3) on the fluorescence intensity of N-CDs were investigated by adding the same concentrations (50 μM) of MnO4 and potential interference substances. Figure 4a and b show that the largest (F0 − F)/F0 was obtained upon addition of MnO4 where F0 and F were the fluorescence intensities of N-CDs without and with metal cations or anions, respectively. The fluorescence intensity of N-CDs decreased sharply upon addition of MnO4, but all other metal cations and anions had a negligible effect on the fluorescence intensity of N-CDs. These results suggested that the fluorescence probe based on N-CDs had excellent selectivity and strong tolerance to the detection of MnO4.

Fig. 4
figure 4

Effects of metal cations (a) and anion (b) on the FL probe of MnO4. Concentrations of the interference and MnO4 were all 50 μM

3.4.2 Optimization of detection conditions

The influence of pH and reaction time were optimized to obtain a high sensitivity (Fig. 5a and b). The fluorescence intensity with or without MnO4 was pH-dependent from 2 to 13. The highest quenching effect was obtained at pH 2. Furthermore, the fluorescence intensity of the N-CDs decreased immediately upon addition of 1.5 μM MnO4, and the F/F0 remained constant when the incubation time was 20 min. F0 and F were the fluorescence intensities of N-CDs without and with 1.5 μM MnO4, respectively. Hence, pH 2 and 20 min were used as the optimal pH and reaction time in the later work.

Fig. 5
figure 5

a Effect of pH on the fluorescence intensity of N-CDs before and after the addition of MnO4 (1.5 μM). b Effect of reaction time on the fluorescence intensity of N-CDs and MnO4 system

3.4.3 Fluorescence response to MnO4 -

Figure 6a shows that the fluorescence intensity of N-CDs decreased gradually with increasing MnO4 concentration. The fluorescence quenching effect of the system is linear with the concentration of MnO4 from 0.15 to 9.00 μM (Fig. 6b). A linear equation was thus established as (F0 − F)/F0 = 0.00615 + 0.058 \({\mathrm{C}}_{{\mathrm{MnO}}_{4}^{-}}\) (μM), R2 = 0.9976, where F0 and F were the fluorescence intensities of N-CDs without and with different concentrations of MnO4, respectively. The limit of detection (LOD) of MnO4 was determined to be 0.12 μM according to the equation LOD = 3σ/k, where σ is standard deviation of the blank solution (n = 11) and k is the slope of the regression line [47].

Fig. 6
figure 6

a Fluorescence spectra of N-CDs with different concentrations of MnO4. b Relationship between the ratio of (F0 − F)/F0 and the concentration of MnO4 from 0.15 to 9.00 μM

The results of the proposed fluorescent probe were also compared to the reported literature, as shown in Table 1. This novel probe offers high sensitivity and low detection limits for MnO4, suggesting that it has great potential for the detection of MnO4 in simulation samples.

Table 1 Comparison of the reported methods for the analysis of MnO4

3.4.4 Simulation sample analysis

The proposed fluorescent probe was successfully used for the detection of MnO4 in polluted water to confirm that the method is accurate and reliable. The simulation sample analysis was carried out as follows. For FL spectra, 200 μL of pure N-CD solution, 200 μL polluted water, and different amounts of MnO4 solution were successively transferred into a 1-cm quartz cuvette, and the mixture was diluted with PBS buffer solution (pH = 2) to a final volume of 2 mL, and this was then mixed thoroughly. The mixed solution was then incubated for 20 min at room temperature. The results are shown in Table 2. The recoveries were 99.4–101.2%, and the relative standard deviations (RSD) were in the range of 2.12 to 3.93%. Thus, the results suggest that the N-CDs can offer trace MnO4 detection in simulation samples.

Table 2 Analysis of MnO4 in polluted water

4 Conclusions

Auricularia auricula (L.ex Hook.) Underw was used to synthesize N-CDs via a facile hydrothermal method in this study. The FL intensity of the synthesized N-CDs could be selectively and sensitively quenched by MnO4. There was a good linear relationship between the FL response and the concentration of MnO4 from 0.15 to 9.00 μM with the detection limit of 0.12 μM. The fluorescence quenching of N-CDs was caused by static quenching and IFE. A fluorescent probe was developed and successfully used to quantitatively detect MnO4 in polluted water with good recoveries from 99.4 to 101.2%. The fluorescent probe is highly sensitive, selective, low cost, environmentally friendly, and easy to prepare. Thus, it has great potential for quantitative monitoring of MnO4 in simulation samples.