Senna leaf derived nitrogen-doped fluorescent carbon dots for Hg 2+ detection

Mercuric ions (Hg 2+) are among the most hazardous heavy metal pollutants, posing severe threats to the human reproductive and central nervous systems even at trace concentrations. Although various fluorescent probes for Hg 2+ have been reported, most rely on a “turn-off” (fluorescence quenching) mechanism, which often suffers from low sensitivity and susceptibility to interference. Novel nitrogen-doped fluorescent carbon dots (FL-CDs) were synthesized for the first time via a one-pot hydrothermal method, using natural senna leaf as a green carbon source, and m-phenylenediamine as both the nitrogen dopant and functional monomer. This unique precursor combination endows the as-prepared FL-CDs with excellent optical properties. The obtained FL-CDs are spherical with an average size of 2.19 nm, exhibit bright green fluorescence at 518 nm, and possess a high quantum yield of 23.54%. Strikingly, in contrast to the common quenching effect of most heavy metal sensors, the fluorescence of FL-CDs was significantly enhanced upon Hg 2+ addition, achieving a desirable “turn-on” response. The probe exhibited exceptional selectivity toward Hg 2+ over other metal ions, with a detection limit of 10.65 μmol/L. This work not only introduces a new biomass precursor for the fabrication of carbon dots but also provides a simple, sensitive, and sustainable “turn-on” sensing platform for trace Hg 2+ detection. The proposed strategy offers a promising alternative for environmental monitoring and food safety analysis, and also contributes to the rational design of advanced functional materials derived from natural resources.

The Mercury is the only liquid and volatile metal at room temperature. It is also one of the most toxic and widely distributed heavy metal pollutants. Mercury can exist in the environment for a long time and migrate easily, and it can also accumulate continuously in organisms, posing a great threat to ecosystems and human health (Wu et al., 2024). Mercury mainly exists in three forms, namely elemental mercury (Hg 0), inorganic mercury (Hg 2+) and organic mercury, among which methylmercury (MeHg) is the most typical organic mercury. These forms of mercury can cycle and transform among the atmosphere, soil and water. Inorganic mercury in water can be converted into more toxic methylmercury by microorganisms, and methylmercury can be continuously enriched through the food chain (Charkiewicz et al., 2025). Mercury is a recognized neurotoxin, which causes particularly severe damage to the early neurodevelopment of children (Dack et al., 2022). In addition, long-term mercury exposure may induce cardiovascular diseases such as hypertension, coronary heart disease and myocardial infarction (Pan et al., 2024). Therefore, it is necessary to pay attention to the identification and selective detection of Hg 2+ in environmental and biological samples.

At present, common Hg 2+ detection methods include atomic absorption spectrometry and chromatography. These traditional techniques are relatively mature, but most of them require expensive large-scale instruments and involve complicated operation steps, which cannot meet the demand for on-site rapid detection (Raina et al., 2024). In recent years, sensors based on nanomaterials have attracted extensive attention. Such sensors feature fast response, simple operation, visual detection and low cost, and are suitable for the detection of trace Hg 2+ (Zavalishin et al., 2024). Researchers have constructed fluorescent probes using carbon dots, precious metal nanoparticles, composite nanomaterials and nanozymes. These probes exhibit strong recognition ability for Hg 2+ in water samples and environmental samples with very low detection limits, providing a feasible technical method for on-site environmental monitoring and health risk assessment.

Fluorescent probes have the advantages of high sensitivity, fast response and simple operation, and are common tools in the fields of analytical sensing and fluorescence imaging (Azam et al., 2021). Fluorescence spectrometry is a specific and sensitive detection method with the advantages of low sample consumption, fast detection speed and simple sample pretreatment, and has been widely used in analytical chemistry and medical detection (Yang et al., 2023). In recent years, nanofluorescent materials with excellent optical properties have attracted considerable attention, and quantum dots are particularly popular in analytical chemistry and biology (Xiao et al., 2021; Sargazi et al., 2022).

Among many fluorescent nanomaterials, semiconductor quantum dots such as CdSe, CdS, Ag 2 S and PbSe have been widely studied (Kumar et al., 2025; García de Arquer et al., 2021). These quantum dots possess unique photoelectric properties and have been extensively and deeply investigated by researcher. However, most of these materials contain heavy metals with high potential toxicity and poor environmental friendliness, making them difficult to be applied in biology-related fields (Yao et al., 2022). In contrast, FL-CDs are novel carbon-based fluorescent nanomaterials with the advantages of wide raw material sources, high fluorescence quantum yield, low toxicity, good biocompatibility, low production cost and stable fluorescence performance, and have thus rapidly become a research hotspot (Qureshi et al., 2024; Molaei, 2019). The ultra-small FL-CDs developed in recent years are smaller than 10 nm with almost no biological toxicity, making them ideal candidates to replace traditional toxic quantum dots and showing broad application prospects in the biomedical field (Yang et al., 2023). In addition, carbon quantum dots possess adjustable photoluminescence properties and strong chemical stability, and also exhibit unique value in environmental analysis and metal ion detection (Ateia et al., 2024). Carbon quantum dots are easy to operate, respond rapidly and can realize real-time detection, providing strong support for the development of a new generation of environmental heavy metal detection technologies (Torres et al., 2022). The surface of carbon quantum dots is rich in functional groups such as carboxyl, amino and hydroxyl groups (Chang et al., 2022). These functional groups can specifically interact with many metal ions, leading to obvious changes in fluorescence intensity such as fluorescence quenching or enhancement, thereby achieving high-sensitivity and high-selectivity detection of target metal ion. In recent years, many studies have demonstrated that carbon quantum dots can be used to detect various heavy metal ions such as Hg 2+, Cr 6+ and Fe 3+. Ren et al. (2023) synthesized nitrogen-doped spherical fluorescent carbon dots via a one-step hydrothermal method and constructed a fluorescent “on-off” sensor to detect Cr 6+ based on the fluorescence quenching mechanism. Cr 6+ can chelate with nitrogen-doped carbon dots and promote charge transfer from the excited state of carbon dots to Cr (VI). Such carbon dots exhibit a wide detection range for Cr (VI), and the fluorescence quenching degree shows a linear relationship with Cr 6+ concentration. This study provides a reference for the detection of metal ions using carbon dots. Rishabh et al. (2024) prepared nitrogen-doped carbon dots via a green synthesis method and constructed a single-probe fluorescent sensing platform for the specific detection of Fe 3+ with a detection limit as low as 0.184 µM Fe 3+ can specifically quench the fluorescence of such carbon dots, which can be practically used for Fe 3+ monitoring in water samples. Mathew and Mathew (2023) employed green-source heteroatom-doped carbon dots as nanoprobes and constructed a fluorescent-electrochemical dual-signal sensing platform based on the internal filtration effect (IFE) for the detection of Co 2+ and Cu 2+, with the detection limits as low as 0.11 μM and 0.03 nM, respectively. This study provides a new idea for the detection of various heavy metal ions in water samples.

At present, most carbon dot probes for Hg 2+ detection suffer from poor selectivity and anti-interference ability in complex water samples and lack targeted structural design. As a key target in environmental monitoring, Hg 2+ still requires the development of more efficient detection methods.

Based on the above, this study proposes a function-oriented design strategy for carbon dots. Senna leaves rich in anthraquinone structure (which can enhance the binding ability with metal ions) are used as the carbon skeleton raw material, and m-phenylenediamine is added as a highly active nitrogen source. FL-CDs are synthesized via a one-pot hydrothermal method. This design aims to construct a recognition interface with strong affinity for Hg 2+ by utilizing the coordination ability of the anthraquinone structure and abundant surface functional groups (such as -NH 2) introduced by nitrogen doping. This study systematically optimizes the synthesis conditions, analyzes the optical properties of the obtained FL-CDs in detail, and focuses on evaluating the fluorescent sensing performance of FL-CDs for Hg 2+ with high selectivity and high sensitivity. This study not only provides a high-performance fluorescent probe for the detection of trace Hg 2+, but also offers a new idea for the rational design and functionalization of carbon dots through the combined strategy of natural products and controllable doping.

Experimental materials and main instruments

All chemical reagents were of analytical grade, and deionized water was used throughout the experiments. No further purification was performed on the reagents before use.

Senna leaves and m-phenylenediamine were purchased from Shanghai Anno Finechem Co., Ltd. Standard stock solutions (100 μmol/L) of metal ions including Al 3+, Ca 2+, Co 2+, Cr 3+, Cu 2+, Fe 3+, Fe 2+, Hg 2+, K +, Li +, Mg 2+, Mn 2+, Na + and Zn 2+ were obtained from Shanghai Macklin Biochemical Co., Ltd. Phosphate-buffered saline (PBS) buffer solution (pH 7.8) was purchased from Beijing Bioaugust Technology Co., Ltd. Nylon syringe filters with a pore size of 0.22 μm were supplied by Nantong Filterbio Membrane Co., Ltd. All the above reagents were of analytical grade and used directly without any pretreatment.

The microstructure of the fluorescent carbon dots (FL-CDs) was observed using a JEM-2100F Transmission Electron Microscope (TEM, JEOL, Japan). Fourier Transform Infrared (FT-IR) spectra of the FL-CDs were recorded on a Nicolet iS5 FT-IR Spectrometer (Thermo Fisher Scientific, United States). The elemental composition of the FL-CDs was characterized via X-ray Photoelectron Spectroscopy (XPS) using an ESCALAB 250Xi spectrometer (Thermo Fisher Scientific, United States). Fluorescence (FL) spectra of the FL-CDs were acquired with a Hitachi F-4600 Fluorescence Spectrophotometer (Hitachi, Japan). Ultraviolet-visible (UV-Vis) absorption spectra of the FL-CDs were measured using a Shimadzu UV 2600 UV-Vis Spectrophotometer (Shimadzu, Japan). Time-resolved fluorescence lifetimes of the FL-CDs were determined on an FL3-P-TCSPC Time-Resolved Fluorescence Lifetime Spectrometer (Horiba Jobin Yvon, France).

All UV–Vis absorption and fluorescence spectroscopic measurements were carried out using standard 1 cm path length quartz cuvettes.

Synthesis of FL-CDs

Fluorescent carbon dots (FL-CDs) were synthesized via a facile one-pot hydrothermal method. The detailed synthetic procedure was described as follows: 1.0 g of senna leaf powder and a certain amount of m-phenylenediamine were accurately weighed at mass ratios of 5:0, 5:1, 5:2, 5:3 and 5:4, respectively. The mixture was transferred into a 50 mL polytetrafluoroethylene (PTFE)-lined stainless steel autoclave, followed by the addition of 10 mL of deionized water. The mixture was thoroughly stirred to form a homogeneous suspension. The autoclave was sealed tightly and then transferred to a temperature-programmable oven for hydrothermal reaction.

To optimize the synthetic conditions, the effects of reaction time and temperature on the properties of FL-CDs were systematically investigated: The hydrothermal reaction was carried out at 200 °C for different durations (2, 4, 8, 10 and 12 h);

The reaction was conducted at 180 °C with reaction durations set as 2, 4, 8 and 10 h, respectively;

The reaction time was fixed at 6 h, while the reaction temperature was varied in the range of 120 °C–220 °C (120 °C, 140 °C, 160 °C, 200 °C and 220 °C).

After the reaction was completed, the autoclave was allowed to cool down naturally to room temperature, and the resulting mixture was collected. The mixture was first centrifuged at 8000 rpm for 15 min to remove the insoluble residues, and the supernatant was collected subsequently. Then, the supernatant was transferred into a dialysis bag with a molecular weight cutoff (MWCO) of 1000 Da and dialyzed against deionized water for 24 h, with water replacement every 6 h. Finally, the external dialysis solution was collected to obtain the purified aqueous dispersion of FL-CDs, which was stored at 4 °C for further use.

Determination of optical stability of FL-CDs

To evaluate the practical application potential of the synthesized FL-CDs in complex environments and detection systems, this study systematically investigated their optical stability, including photostability, pH stability, ionic strength stability, temperature stability and long-term storage stability.

Photostability test

An aqueous solution of FL-CDs with a certain concentration was prepared and irradiated continuously under a 365 nm UV lamp with constant power. Samples were taken at 0, 1, 2, 3, 4 and 5 h of irradiation, respectively. The fluorescence emission intensity at the optimal excitation wavelength was measured quickly. The photobleaching resistance of FL-CDs was evaluated by analyzing the curve of fluorescence intensity changing with irradiation time.

pH stability test

Dilute HCl and NaOH solutions were used to adjust the same batch of FL-CDs dispersion to a series of different pH values (such as pH 2–12). After equilibration for 10 min at each pH, the fluorescence emission spectra were measured. The suitable pH range for FL-CDs to maintain stable fluorescence performance was determined by investigating the relationship between the maximum fluorescence intensity and pH values.

Ionic strength stability test

NaCl solutions with different concentrations (such as 0–500 mM) were added into the aqueous FL-CDs solution to simulate the change of ionic strength in the environment. After adding the electrolyte and mixing evenly, the fluorescence intensity was measured immediately. The robustness of the optical performance of FL-CDs in high-salt or complex ion environments was evaluated by analyzing the change degree of fluorescence signals under different ionic strengths.

Temperature stability test

The aqueous FL-CDs solution was placed at a series of different temperatures (such as 4 °C–90 °C) for a period of time. After the system temperature was stable, the fluorescence emission intensity was measured. The tolerance of the fluorescence performance of FL-CDs to temperature changes was evaluated by comparing the changes of fluorescence intensity at different temperatures.

Long-term storage stability test

The aqueous FL-CDs solution was stored at room temperature and away from light for up to 2 months. Samples were taken at 0, 7, 15, 30 and 60 day of storage, respectively. The fluorescence intensity at the optimal excitation wavelength was measured. The optical stability of FL-CDs after long-term storage at room temperature was judged according to the change of fluorescence intensity with storage time.

Selectivity for metal ions and Hg 2+ detection performance of FL-CDs

To evaluate the recognition performance of the synthesized FL-CDs toward target analytes, selectivity experiments and sensitivity measurements were carried out in sequence.

Metal ion selectivity test

200 μL of 30-fold diluted FL CD stock solution synthesized under optimal conditions was mixed with 2.0 mL of different metal ion solutions (100 μmol/L), including K +, Na +, Ca 2+, Mg 2+, Zn 2+, Cu 2+, Fe 3+, Cr 3+, Pb 2+, Cd 2+, Cl −, SO 4 2-, NO 3 − and Hg 2+. The mixture was vortexed for full homogeneity and kept standing at room temperature for 15 min. Then the fluorescence emission spectra were measured at an excitation wavelength of 469 nm. The selective recognition ability of FL-CDs toward Hg 2+ was investigated by comparing the changes (enhancement or quenching) of fluorescence intensity at the characteristic emission peak in the presence of each metal ion.

Determination of detection sensitivity and limit of detection for Hg 2+

A series of Hg 2+ standard solutions with different concentrations were prepared (concentration range: 0–2 nmol/L). 500 μL of Hg 2+ solution at each concentration was taken and mixed with 500 μL of the diluted FL CD solution. Then 2.0 mL of PBS buffer at pH 7.8 was added, and the volume was adjusted to 3.0 mL. After mixing and standing at room temperature, the fluorescence emission spectra of the mixture were measured. The standard working curve was plotted with Hg 2+ concentration as the abscissa and fluorescence intensity (or fluorescence quenching efficiency F 0 /F) at the characteristic emission peak of FL-CDs as the ordinate. According to the method recommended by the International Union of Pure and Applied Chemistry (IUPAC), the limit of detection of this method for Hg 2+ was calculated as the ratio of three times the standard deviation of the blank signal (σ) to the slope of the working curve (k) (3σ/k).

Effect of pH value

The pH value of the solution may affect the surface charge state of FL-CDs and their complexation with Hg 2+, and further affect the detection sensitivity. To determine the optimal detection acidity, the effect of pH on the sensing system was first investigated. The specific method is as follows: 500 μL of diluted FL-CDs solution, 500 μL of Hg 2+ standard solution with fixed concentration (e.g., 50 nmol/L), and 2.0 mL of Britton-Robinson (BR) or phosphate (PBS) buffers with different pH values (range: 4.0–10.0) were mixed. After standing at room temperature for 15 min, the fluorescence intensity of the mixture at the characteristic emission wavelength was measured.

This section may be divided by subheadings. It should provide a concise and precise description of the experimental results, their interpretation, as well as the experimental conclusions that can be drawn.

Optimal synthesis conditions of FL-CDs

This study analyzed the effects of reaction time, reaction temperature and precursor mass ratio on the fluorescence properties of FL-CDs, with the results shown in Figures 1a–d. At 180 °C, the fluorescence intensity of FL-CDs increased gradually with reaction time from 2 h to 10 h, and the intensity at 8 h was close to that at 10 h. At 200 °C, the fluorescence intensity increased continuously with reaction time from 2 h to 8 h and decreased slightly at 10 h. When the reaction time was fixed at 6 h, the fluorescence intensity increased significantly with reaction temperature from 120 °C to 200 °C and decreased obviously at 220 °C. The mass ratio of senna leaf to m-phenylenediamine directly affected the fluorescence performance of FL-CDs, and the fluorescence intensity reached the highest level at the mass ratio of 5:1. Thus, the optimal synthesis conditions of FL-CDs were determined as reaction temperature 200 °C, reaction time 6 h, and mass ratio of senna leaf to m-phenylenediamine 5:1. FL-CDs prepared under these conditions were used in all subsequent experiments.

Morphology and structural composition of fluorescent carbon dots

This study characterized the micromorphology and size characteristics of carbon dots by transmission electron microscopy. Figure 2a shows that the carbon dots are regularly spherical with good dispersibility and no obvious agglomeration, clear particle edges without burrs and good overall morphological uniformity. The Gaussian distribution curve fitting result in Figure 2b shows that the carbon dots have a size range of 0.36–2.62 nm and an average particle size of about 2.19 nm with concentrated particle size distribution and no obvious abnormal particles. The average particle size is larger than that of FL-CDs reported in previous literature. The HR-TEM image in Figure 2c clearly presents the internal lattice network structure of carbon dots with clear and continuous lattice fringes and no obvious defects, and the lattice spacing is about 0.206 nm corresponding to the (100) crystal plane characteristic of graphitic carbon. The core of carbon dots is composed of graphitic and amorphous carbon structures.

The surface functional groups of FL-CDs were characterized by FT-IR spectroscopy, and the results are shown in Figure 3. The broad absorption peak in the range of 3150–3450 cm -1 corresponds to the stretching vibrations of O-H and N-H bonds on the surface of FL-CDs. The characteristic absorption peak at 2937 cm -1 is assigned to the stretching vibration of C-H bonds. The absorption peak at 1595 cm -1 is a typical signal for the stretching vibration of C=O bonds. The absorption peak at 1408 cm -1 corresponds to the stretching vibration of C=C bonds. The absorption peak at 1300 cm -1 matches the stretching vibration of C-N bonds. Based on the characteristic absorption peaks of the FT-IR spectrum, the presence of -NH 2 groups on the surface of FL-CDs can be confirmed.

The chemical composition of FL-CDs was determined by XPS. The full XPS spectrum in Figure 4a shows characteristic peaks of C 1s, N 1s and O 1s at binding energies of 284.93 eV, 399.33 eV and 531.36 eV, respectively, confirming that the main elements on the surface of FL-CDs are carbon, nitrogen and oxygen. Peak fitting analysis was performed on the above characteristic peaks, and the fitting results correspond to the high-resolution spectra in Figures 4b–d. In the C 1s high-resolution spectrum (Figure 4b), the characteristic peak at 284.8 eV is assigned to C-C/C=C groups, and the peak at 287.8 eV is attributed to C-O-C/O-C=O groups. In the N 1s high-resolution spectrum (Figure 4c), the peak at 399.4 eV corresponds to amino-type -N (-NH 2, -NHR or -NR 2) and C-N bonds, and the peak at 401.5 eV is ascribed to pyrrolic -N (C 4 H 4 N) or -NO 2 groups. In the O 1s high-resolution spectrum (Figure 4d), the characteristic peaks at 532.4 eV and 531.1 eV correspond to C=O groups and C-O-C/C-OH groups, respectively. The overall XPS results verify the presence of amino groups on the surface of FL-CDs, which is consistent with the FT-IR analysis.

As shown in Figure 5, the XRD pattern of pristine FL-CDs presents a broad amorphous peak at 2θ ≈ 20.23°, attributed to irregular stacking of carbon sheets. After mixing with Hg 2+, the diffraction peak slightly shifts to larger 2θ alongside reduced peak intensity. This result proves that Hg 2+ coordinates with surface −OH/−COOH of FL-CDs, compressing carbon interlayer spacing and raising structural disorder. Meanwhile, high atomic number of Hg enhances X-ray scattering, resulting in a modest upward shift of the whole baseline for the CDs + Hg 2+ sample.

UV absorption and fluorescence properties of FL-CDs

The optimal excitation wavelength of FL-CDs was determined to be 469 nm with a corresponding emission wavelength of 518 nm by three-dimensional wavelength scanning using a fluorescence spectrophotometer. Figure 6 shows the UV-Vis absorption spectrum and fluorescence spectrum of FL-CDs. The characteristic absorption peak at 383 nm in the UV-Vis absorption spectrum is attributed to the n–π* electron transition of groups such as carbonyl and amino groups on the FL-CD surface. The fluorescence spectrum exhibits a characteristic emission peak at 518 nm under excitation at 469 nm. The inset of Figure 6 shows that FL-CDs produce bright green fluorescence under blue light excitation at 469 nm. The relative fluorescence quantum yield of FL-CDs was calculated to be 23.54% according to Equation 1.

Symbols: Φ: Relative fluorescence quantum yield of the as-prepared CDs; Φ R: Fluorescence quantum yield of the reference standard; I, I R: Integrated fluorescence intensities of the sample and reference standard, respectively; A, A R: Absorbance values of the sample and reference standard at the excitation wavelength, respectively; η, η R: Refractive indices of the solvents for the sample and reference standard, respectively.

Stability test of FL-CDs

The photostability, pH stability, temperature stability, long-term storage stability, ionic strength stability and fluorescence lifetime of FL-CDs were systematically inves-tigated in this study. Figure 7a shows that the fluorescence intensity of FL-CDs slowly decreases from approximately 4.8 a. u. to about 4.6 a. u. under UV irradiation for 0–5 h, with a very small variation range. Figure 7b shows that the fluorescence intensity of FL-CDs gradually increases with the increase of pH in the range of pH 2–13, reaches the maximum at pH 10, and then slightly decreases with further increase of pH, maintaining a high level overall. Figure 7c shows that the fluorescence intensity of FL-CDs gradually decreases with the increase of temperature in the range of 15 °C–95 °C, remains basically stable in the range of 15 °C–30 °C, and the downward trend becomes more obvious when the temperature exceeds 50 °C. Figure 7d shows that the fluores-cence intensity of FL-CDs slowly decreases from approximately 4.7 a. u. to about 4.55 a. u. during long-term storage for 0–60 days, and the intensity tends to be stable without obvious attenuation after 30 days. Figure 7e shows that the fluorescence intensity of FL-CDs gradually decreases with the increase of ionic strength in the range of 0–500 mM NaCl, and still maintains a detectable level overall. Figure 7f shows that the fluo-rescence lifetime of FL-CDs decays rapidly with time, and the signal drops from ap-proximately 10,000 a. u. to nearly 0 within 0–70 ns.

Sensing performance of FL-CDs toward Hg 2+

The selective response of FL-CDs to various metal ions and the regulation effect of pH on their fluorescence performance were investigated in this study. Figure 8a shows that the fluorescence intensity of FL-CDs remains basically stable in the presence of metal ions such as Li +, Mg 2+, K +, Fe 3+, Al 3+, Na +, Fe 2+, Ca 2+, Cr 3+, Co 2+, Mn 2+, Zn 2+ and Cu 2+, and only Hg 2+ can significantly enhance the fluorescence intensity of FL-CDs. Figure 8b shows that the fluorescence intensity of FL-CDs reaches the highest level in the pH range of 9–11 without Hg 2+. With Hg 2+ added, the fluorescence intensity of FL-CDs is significantly enhanced at pH 2–7, and the enhancement amplitude decreases obviously when pH > 7. This study confirms that FL-CDs exhibit sensitive response to Hg 2+ in acidic to neutral environments.

The linear response range and detection limit of the FL-CDs sensing system toward Hg 2+ were evaluated under optimized experimental conditions. Figure 9 shows that the fluorescence intensity of the system gradually increases with the increase of Hg 2+ con-centration. The relative change in fluorescence intensity F/F 0 −1 exhibits a good linear relationship with Hg 2+ concentration in the range of 20–100 μmol/L, with a linear correlation coefficient R 2 of 0.9566, which conforms to the Stern–Volmer type response behavior. The fitted regression equation is y = 0.0002x − 0.002. The detection limit of FL-CDs for Hg 2+ was calculated using the formula LOD = 3σ/k, where σ is the standard deviation of the fluorescence signal of the blank solution and k is the fitting slope. Substituting σ = 0.00071 and k = 0.0002 μL/mol, the detection limit was determined to be 10.65 μmol/L. All measurements were conducted in triplicate, and RSD values of all test data were lower than 5%, confirming favorable experimental reproducibility.

To further verify the practical applicability of the as-prepared CDs sensor, tap water and lake water were adopted as real samples. Three spiked concentrations (12, 50, 90 μM) were designed: 12 μM near the LOD was used to evaluate trace detection performance, while 50 and 90 μM fell within the linear range for quantitative verification. As summarized in Table 1, recoveries were between 99.96% and 100.83%, and all RSD values were below 2.2%, demonstrating favorable accuracy and precision in real aqueous matrix.

To elucidate the fluorescence enhancement mechanism, the changes in UV Vis absorption spectrum and fluorescence lifetime of carbon dots before and after the ad-dition of Hg 2+ were investigated in this study. Figure 10a shows that the pristine carbon dots exhibit a broad absorption band near 350 nm, which is assigned to the n π* elec-tron transition of surface groups such as carbonyl and amino groups. After the addition of Hg 2+, the absorption intensity decreases in the range of 300–450 nm, while no ob-vious shift is observed in the position and profile of the absorption peak. The time re-solved fluorescence decay curve in Figure 10b shows that the decay traces of pristine carbon dots and CDs + Hg 2+ complex are highly overlapped. The fluorescence lifetime fitting results in Table 2. Show that the average fluorescence lifetime of pristine carbon dots is 8.62 ns, and that of the CDs + Hg 2+ complex is 9.08 ns, with a negligible difference between them.

Optimal synthesis conditions of FL-CDs

The variation in fluorescence intensity of FL-CDs under different synthesis conditions is related to precursor carbonization, surface passivation, and doping regulation. In the range of 180 °C–200 °C, increasing temperature and prolonging reaction time can accelerate the carbonization process of senna leaf and m-phenylenediamine precursors, regularize the carbon skeleton, achieve surface passivation, and introduce more functional groups such as O-H and N-H to form more efficient fluorescent luminescence centers, thereby enhancing the fluorescence intensity of FL-CDs (Li and Dong, 2018). Excessively high temperature beyond 200 °C or overly long reaction time will cause agglomeration of FL-CDs and loss of surface active functional groups, reducing their fluorescence intensity (Piasek et al., 2023). As a nitrogen source, m-phenylenediamine can introduce abundant nitrogen-containing functional groups such as -NH 2 onto the surface of FL-CDs via nitrogen doping. Proper addition improves the fluorescence quantum yield of FL-CDs, while excessive addition generates excessive structural defects and non-radiative recombination centers, weakening their fluorescence performance. When the mass ratio of senna leaf to m-phenylenediamine is 5:1, an optimal balance among carbonization degree, nitrogen doping level and surface functional group content is achieved, leading to the maximum fluorescence intensity of FL-CDs (Ren et al., 2025).

Morphology and structural composition of FL-CDs

Various functional group absorption peaks in the FT-IR spectrum fully reflect the surface chemical composition of FL-CDs. The existence of O-H and N-H bonds endows FL-CDs with good hydrophilicity and water dispersibility, which is an important basis for the application of FL-CDs in aqueous solution detection (Jorns and Pappas, 2021). C=C and C=O bonds jointly form the conjugated carbon skeleton of FL-CDs, which is the core premise for the fluorescence performance of FL-CDs (Yorozuya et al., 2025). The characteristic absorption peaks of C-N and N-H bonds directly verify the successful doping of nitrogen into the FL-CDs structure. As an active functional group, -NH 2 exists on the surface of FL-CDs and can provide binding sites for the specific interaction between FL-CDs and Hg 2+, which is the key structural basis for the application of FL-CDs in Hg 2+ fluorescence sensing (Zheng et al., 2025).

The characteristic peaks of C, N and O in the full XPS spectrum are consistent with the N, O co-doped synthesis strategy of FL-CDs. The peak-fitting results of the C 1s high-resolution spectrum reveal that FL-CDs possess both a conjugated carbon skeleton and oxygen-containing functional groups, which form the structural basis for the fluorescence properties of FL-CDs (Zhao et al., 2022). The characteristic peak of amino-type -N in the N 1s high-resolution spectrum directly confirms the successful doping of nitrogen into the FL-CDs structure in the form of amino groups. The peak-fitting results of the O 1s spectrum further supplement the types of oxygen-containing functional groups on the FL-CDs surface and improve the understanding of its surface chemical structure. The consistency of XPS and FT-IR in confirming the existence of amino groups provides cross-validation for the surface functional groups of FL-CDs. As key active sites, the confirmed presence of amino groups offers direct structural evidence for the specific recognition between FL-CDs and Hg 2+ (Kasif et al., 2024). XRD results further verify the coordination effect from the bulk stacking structure. Peak shift and intensity change after Hg 2+ treatment agree well with FT-IR and XPS analysis, confirming the structural variation induced by Hg 2+ binding.

UV absorption and fluorescence properties of FL-CDs

The UV absorption peak of FL-CDs at 383 nm originates from the n–π* electron transition of surface groups such as carbonyl and amino groups, and this transition characteristic is directly related to the chemical composition of the surface functional groups of FL-CDs (Huang et al., 2023). The fluorescence emission peak at 518 nm under excitation at 469 nm is a typical fluorescence feature of FL-CDs, corresponding to the radiative transition process of the internal luminescent centers of FL-CDs (Bai et al., 2021). The bright green fluorescence under 469 nm blue light excitation intuitively reflects the fluorescence emission performance of FL-CDs. The relative fluorescence quantum yield of 23.54% is at a high level, which provides a performance guarantee for the practical application of FL-CDs in the field of fluorescence sensing. A higher quantum yield means that FL-CDs can produce stronger fluorescence signals during detection, which is beneficial to improving detection sensitivity (Xu et al., 2022).

Stability test of FL-CDs

The UV irradiation results in Figure 10a indicate that FL-CDs have good photostability, and their structure is not easy to be photobleached or damaged by UV irradiation, which can meet the demand of long-term optical detection (Rao et al., 2022). The pH variation in Figure 10b is directly related to the protonation and deprotonation of surface functional groups of FL-CDs (Dilshener et al., 2024). The surface charge and functional group state of FL-CDs at pH 10 are most favorable for fluorescence emission, and this pH range provides a suitable acidity condition for subsequent sensing applications. The decrease in fluorescence intensity at high temperatures in Figure 10c is due to the increased probability of non-radiative transitions (Leng et al., 2022), and the stable range of 15 °C–30 °C ensures that FL-CDs can achieve stable fluorescence output at room temperature. The long-term storage results in Figure 10d confirm that FL-CDs have excellent storage stability, with stable intensity and no obvious attenuation after 30 day, which is suitable for long-term storage and detection of actual samples. The fluorescence intensity decreases slightly with the increase of ionic strength in Figure 10e, while FL-CDs still maintain detectable signals, indicating that FL-CDs have a certain resistance to ion interference. The rapid decay of fluorescence lifetime in Figure 10f is a typical fluorescence dynamic characteristic of FL-CDs, and short-lifetime fluorescence can reduce the interference of background signals and improve the detection signal-to-noise ratio (Zhang et al., 2023).

Sensing performance of FL-CDs toward Hg 2+

The fact that FL-CDs show significant fluorescence enhancement only toward Hg 2+ reflects their specific recognition ability for Hg 2+. Other metal ions cannot form effective interactions with the surface functional groups of FL-CDs and thus exert no influence on the fluorescence signal (Gao et al., 2022). The high fluorescence intensity at pH 9–11 without Hg 2+ originates from the optimal protonation state of the surface functional groups of FL-CDs, and the weakly alkaline environment is more conducive to the stability of the optical properties of FL-CDs (Chang et al., 2020). The pH-dependent response in the presence of Hg 2+ is directly related to the protonation behavior of amino groups on the FL-CDs surface. In acidic to neutral environments, amino groups are fully protonated to -NH 3 +, which can form strong complexation with Hg 2+ and induce fluorescence enhancement. In alkaline environments, amino groups are deprotonated to -NH 2, and the complexation ability with Hg 2+ is weakened, thus inhibiting the fluorescence enhancement effect. The sensitive response characteristic in acidic to neutral environments provides an application basis for the detection of Hg 2+ in actual water samples by FL-CDs.

The good linear relationship in the range of 20–100 μmol/L indicates that FL-CDs can realize quantitative detection of Hg 2+ within this concentration range, and the goodness of fit with R 2 = 0.9566 verifies the reliability of the response behavior. The detection limit of 10.65 μmol/L reflects the detection sensitivity of FL-CDs for Hg 2+, which meets the basic requirements for Hg 2+ detection in some actual water samples. Compared with reported mercury ion sensing materials, the FL-CDs sensor in this study has a wider linear detection range at a similar detection limit level. A wider linear range means that FL-CDs can cover the detection of Hg 2+ samples with more concentration gradients, improving the adaptability in practical application scenarios. Practical sample recovery experiments with tap water and lake water further supported its application potential. Spiked concentrations of 12, 50 and 90 μM covered near-LOD trace level and mid/high concentrations of the linear region, and satisfactory recoveries and low RSD proved the sensor is resistant to common water matrix interference. Compared with previously reported Hg 2+ fluorescent probes, the present CDs possess equivalent LOD and broader linear range. FL-CDs can be used as an effective supplementary material in the field of mercury ion sensing and provide a reference for the development of high-performance metal ion fluorescence sensors.

The favorable linear relationship from 20 to 100 μmol/L demonstrates that FL-CDs enable quantitative determination of Hg 2+ within this range, and the correlation coefficient R 2 = 0.9566 confirms the reliability of the fluorescence response. The limit of detection (LOD) of 10.65 μmol/L reveals the sensing sensitivity toward Hg 2+, satisfying the basic demands for Hg 2+ quantification in practical water samples. Compared with previously reported Hg 2+ sensors, the as-prepared FL-CDs exhibit a wider linear dynamic range while maintaining a comparable LOD. In contrast, our single-component N-doped FL-CDs avoid tedious mixing of two different carbon dots. Despite a relatively higher LOD of 10.65 μmol/L, our sensor achieves a much broader linear range from 20 to 100 μmol/L, which is suitable for direct quantification of moderately/heavily polluted wastewater without multiple dilution treatments, presenting distinctive superiority for practical field detection. Such an extended linear interval allows the sensor to quantify Hg 2+ across diverse concentration levels and improves its practicability for on-site detection.

The favorable linear relationship spanning 20–100 μmol/L demonstrates that FL-CDs enable quantitative determination of Hg 2+, and the correlation coefficient R 2 = 0.9566 confirms reliable fluorescence response. The limit of detection (LOD = 10.65 μmol/L) meets the requirement of practical Hg 2+ detection in environmental water. As reported by Guo et al. (Guo et al., 2025), the ratiometric sensor relied on two separately synthesized carbon dots to realize an ultralow LOD at the cost of narrow linear range. Different from their dual-CDs system, our single-component N-doped FL-CDs eliminate complicated component blending. Although our LOD is relatively higher, the resultant wide linear window (20–100 μmol/L) permits direct measurement of moderately or heavily polluted wastewater without repeated dilution, greatly improving the practicability for field sample analysis. All measurements were conducted in triplicate, and RSD values of all test data were lower than 5%, confirming favorable experimental reproducibility.

Spiked recovery assays using tap and lake water further validate its real-world applicability. Three fortified concentrations (12, 50, and 90 μM) were selected: 12 μM was close to the LOD for trace-level detection, whereas 50 and 90 μM fell inside the calibrated linear range. The obtained favorable recoveries and small RSD values confirm strong anti-interference capacity against complex water matrixes. In summary, FL-CDs serve as a promising alternative for mercury sensing and offer valuable guidance for fabricating high-performance fluorescent metal-ion sensors.

The absence of an obvious shift in the position and profile of the absorption peak in Figure 8a indicates that the interaction between Hg 2+ and FL-CDs mainly affects the surface electronic states without changing the graphitic structure of the carbon core and the quantum confinement effect. The negligible difference in fluorescence lifetime suggests that the fluorescence enhancement does not originate from the change in the radiative recombination rate of the excited state, and the luminescent centers of the carbon core remain essentially unchanged. The fluorescence enhancement induced by Hg 2+ is mainly attributed to the inhibition of the surface photoinduced electron transfer (PET) process (Khan et al., 2025). In the absence of Hg 2+, the lone pair electrons of nitrogen atoms on the FL-CDs surface can transfer to the empty orbitals of excited FL-CDs via the PET process, causing non-radiative transitions and fluorescence quenching. After the coordination of Hg 2+ with nitrogen atoms, the electron-donating ability of nitrogen atoms is weakened, the PET process is inhibited, and the radiative recombination pathway becomes dominant, finally leading to fluorescence enhancement. Hg 2+ regulates the non-radiative transition pathways of FL-CDs through surface coordination, thus realizing the effective modulation of fluorescence properties.

The PET inhibition mechanism is fully supported by multiple lines of evidence: (1) unchanged absorption spectra indicate no alteration of the carbon core; (2) nearly constant fluorescence lifetime confirms unchanged radiative recombination; (3) surface amino groups identified by XPS and FT-IR serve as electron donors for PET; and (4) pH-dependent response verifies the key role of amino protonation in Hg 2+ coordination. Together, these results strongly demonstrate that Hg 2+ suppresses the PET process and recovers fluorescence.

In this work, green-emitting fluorescent carbon dots (FL-CDs) were synthesized via a facile hydrothermal route, and synthetic parameters were optimized by single-factor experiments to obtain the optimal preparation condition of 200 °C for 6 h. Characterizations with TEM, FT-IR, XPS, XRD and UV–Vis spectroscopy proved that the resultant FL-CDs possessed uniform particle size, favorable water dispersibility and abundant surface hydroxyl/carboxyl groups, which provided sufficient binding sites for target ion combination and guaranteed excellent sensing performance. Ion selectivity tests demonstrated outstanding specific fluorescence response of FL-CDs toward Hg 2+, whereas common coexisting metal ions barely interfered with the detection signal. The FL-CDs-based fluorescent sensor displayed reliable linear fluorescence enhancement against Hg 2+ ranging from 20 to 100 μmol/L with R 2 = 0.9566 and an LOD of 10.65 μmol/L, which is competent for trace Hg 2+ analysis in environmental samples. Mechanism investigations using fluorescence lifetime and UV–Vis titration confirmed that Hg 2+ inhibited the PET process of FL-CDs and suppressed non-radiative decay, thereby triggering obvious fluorescence enhancement. Furthermore, satisfactory recoveries (99.96%–100.83%) and low RSD values (<2.2%) from tap and lake water spiking experiments verified favorable anti-matrix-interference ability and practical usability of the probe. Compared with previously reported Hg 2+ fluorescent sensors, the as-prepared FL-CDs achieved comparable detection sensitivity together with a broader linear detection window. Consequently, the developed FL-CDs are promising fluorescent probes for practical Hg 2+ monitoring and provide a feasible strategy for developing high-performance metal ion fluorescent sensors.