Introduction

Ascorbic acid (AA), commonly known as vitamin C, is an essential nutrient involved in numerous biological processes, including the formation of blood vessels, scar tissue, and cartilage. It serves as a cofactor for several enzymes and functions as a water-soluble antioxidant, protecting cells from oxidative stress associated with chronic diseases such as cancer and atherosclerosis1. AA also plays a crucial role in the biosynthesis of collagen, ATP, and neurotransmitters like dopamine. A deficiency in AA leads to scurvy, a condition characterized by impaired tissue repair and growth2,3,4,5,6,7,8. Although excessive intake of AA may cause side effects such as kidney stones, the recommended daily intake is 90 mg for men and 75 mg for women9. Therefore, monitoring AA levels in food and biological fluids is essential for health maintenance10. Accurate and rapid quantification of AA is vital in both clinical diagnostics and food quality control.

Iron (Fe3+), an essential trace element, is equally critical to human physiology. It is involved in oxygen transport, DNA synthesis, and numerous enzymatic reactions. However, both iron deficiency and iron overload pose serious health risks. Insufficient iron can lead to anemia, cognitive impairments such as reduced memory and learning capacity, and a weakened immune system. In contrast, excess iron is associated with conditions like hemochromatosis, kidney damage, increased infection risk, and heightened cancer susceptibility11. Thus, there is a pressing need for rapid, reliable, and selective detection methods for both Fe3+ and AA.

Several analytical techniques have been developed for the detection of AA and Fe3+, including electrophoresis12, spectrophotometry13, fluorometry, electrochemistry14, and chromatography. Despite their utility, these methods often suffer from drawbacks such as complex operation, high cost, low sensitivity, and interference from background signals. Fluorescence-based methods offer advantages such as high sensitivity and real-time detection; however, they too can be affected by background noise and often require sophisticated instrumentation15. Therefore, the development of simple, cost-effective, and sensitive fluorescence-based sensing platforms remains a significant research priority.

Recent advances have explored carbon-based nanomaterials for fluorescence sensing. For instance, Gao et al. reported lignin-derived carbon quantum dots (CQDs) for the dual detection of AA and Fe3+ in living cells16,17,18,19. Yin et al. synthesized bright blue fluorescent polydopamine nanoparticles (PDANPs) through a single-step process and demonstrated their utility as "turn-off" sensors for Fe3+ detection via coordination between Fe3+ and phenolic hydroxyl groups on the nanoparticles20. Despite their potential, the synthesis of these materials can be complex, and the fluorescence performance may be compromised by UV-induced background interference. Hence, there remains a need for simpler and more effective sensing systems for AA in complex biological environments.

Fluorescence-based optical sensing is particularly appealing due to its sensitivity, selectivity, and operational simplicity. These systems often rely on fluorescence quenching, where emission intensity decreases in the presence of a target analyte. The fluorescence can be selectively restored using a competitive ligand, enabling a reversible "turn-off/turn-on" detection mechanism21. This strategy not only reduces false positives but also allows efficient recovery of fluorescence through competitive bidding, making it highly suitable for biological applications. Ideal fluorophores for such systems should exhibit strong emission, high photostability, and ease of quenching and recovery.

Carbon dots (CDs), due to their high-water solubility, excellent optical properties, biocompatibility, and low cytotoxicity, have emerged as promising candidates for fluorescence sensing. Their tunable fluorescence, cost-effective synthesis, and ease of surface functionalization make them attractive for detecting pH, metal ions, and biomolecules22,23,24,25,26,27,28. However, despite growing interest, "turn-on" fluorescence sensing platforms based on CDs remain underexplored29, leaving ample room for innovation.

In this study, we report a simple one-step hydrothermal synthesis of blue-emitting CDs using ethylene glycol tetraacetic acid (EGTA) as the sole precursor, eliminating the need for additional surface passivation. By adjusting the synthesis temperature, we successfully tuned the emission properties of the CDs. EGTA was selected due to its rich carbon content and the presence of functional groups (–OH and –NH–) that can influence the surface chemistry and optical behavior of the resulting CDs. In this method, EGTA plays a dual role-as a carbon source and an intrinsic functionalizing agent-simplifying the synthesis while reducing cost, processing time, and environmental burden.

The unique molecular structure of EGTA promotes carbonization and the formation of surface functional groups, resulting in CDs with high fluorescence quantum yield and excitation-dependent emission. These CDs show excellent performance in the selective fluorescence quenching by Fe3+ and subsequent recovery by AA, establishing a robust “turn-off/turn-on” sensing platform. Our approach is not only energy-efficient and scalable but also offers superior selectivity, high reproducibility, and low toxicity making it highly suitable for biosensing and bioimaging applications. By leveraging EGTA’s multifunctionality, this work introduces a cost-effective and practical route for the development of advanced fluorescence sensors.

Experimental section

Chemicals

The supplier of EGTA (99%) was Guangfu Reagent Company in Tianjin, China. Heowns (Tianjing, China) supplied the Na2S, NaI, , (CH3COO)2Zn, FeSO4, AgNO3, Hg(NO3)2, and chlorides of (K+, Mg2+, Fe3+, Cu2+, Li+, Ca2+, Cr3+, Ni2+, Al3+, NH4+, Mn2+, Na+ and Co2+). Heowns (Tianjing, China) provided the following reagents reduced L-cysteine (Cys), D-aspartic acid (Asp), reduced L-glutathione (GSH), L-ascorbic acid, dopamine hydrochloride (DA), L-glutamic Acid (Glu), L-isoleucine (lle), ß-alanine (Ala), L-histidine (Hls), L-tyrosine (Tyr), L-asparagine (Asn), L-proline (Pro), L-valine (Val), L-phenylalanine (Phe), glycine (Gly), L-serine (Ser), L-methionine (Met), L-tryptophan (Trp), L-arginine (Arg), L-lysine (Lys), L-glutamine (Gln), L-leucine (Leu), and L-threonine (Thr). All reagents are of analytical grade. Ultrapure water obtained from a Milli-Q ultrapure (18.2 MΩ cm) system was used in all experiments.

Instrument

XRD measurement was performed on X-ray diffractometer (Philips X’Pert Holland) with Cu Kα radiation (λ = 1.54059), with opening voltage and current at 40 kV and 35 mA. The transmission electron microscopy (TEM) was performed on a JEM-2100 transmission electron microscopy at an accelerating voltage of 200Kev. Dynamic light scattered (DLS) was got on a BI-200SM (USA Brookhaven) at room temperature. Freeze drying of CDs product was carried out in benchtop freeze dryer (Labconco) at -50 °C and 0.05 atm pressure. The Fourier transform infrared Spectroscopy (FTIR) spectra were measured by Nicolet 360 FTIR spectrometer with the KBr pellet technique. X-ray photoelectron spectra (XPS) was measured on a PHI-550 spectrometer by using Mg Ka radiation (hv = 1253.6) photoemission spectroscopy with a base vacuum operated at 300W. Elemental analysis for CDs was analyzed by inductively coupled plasma atomic emission spectrometry (ICP-AES) (Thermo Jarrel Ash,Franklin, MA, USA).

Steady-state and time-resolved fluorescence spectroscopy and UV–Vis absorption

The UV–Vis absorption spectra were recorded using a Varian UV-Cary100 spectrophotometer. Corrected steady-state fluorescence emission spectra were obtained on an Edinburgh Instruments FLS920 spectrofluorometer with quantum yields determined by comparing the integrated emission band of the sample with that of quinine sulfate in 0.5 M H2SO4 (quantum yield, Φ = 0.54)[30] . Fluorescence lifetimes were measured using time-correlated single-photon counting (TCSPC) across 2048 channels at room temperature, with excitation provided by a 320 nm pulsed diode laser.

The LUDOX scatter solution was used to acquire instrument response functions, and sample decays were histogrammed until peak channel counts reached ~ 1.0 × 104. The decay profiles were analyzed by fitting the data to multi-exponential functions using a Gaussian-weighted nonlinear least squares method implemented in the instrument’s software, based on the Marquardt–Levenberg minimization algorithm. The pre-exponential factors and corresponding decay times were extracted by minimizing the reduced chi-square (χ2). The quality of the fits was also evaluated graphically by plotting weighted residuals as 2D “carpet” surfaces against channel numbers. All fitted decay profiles exhibited χ2 values less than 1, indicating good agreement between experimental data and the fitting model.

Synthesis of Photoluminescent CDs

Photoluminescent carbon dots (CDs) were synthesized via a hydrothermal carbonization method. Optimization was conducted by varying the carbonization temperatures (160, 180, 200, 300, and 350 °C) while maintaining a constant reaction time. The optimal conditions involved heating 0.4 g of EGTA in 25 mL of ultrapure water in a 50 mL Teflon-lined stainless-steel autoclave at 200 °C for 3 h without any surface-passivating agents. Upon natural cooling to room temperature, a yellowish solution exhibiting photoluminescence was obtained.

The resulting solution was subjected to evaporation overnight to remove excess water, followed by freeze-drying. The dried CDs were redispersed in 100 mL of water and centrifuged at 10,000 rpm for 30 min to eliminate unreacted materials and larger by-products. The final aqueous CD dispersion had a concentration of approximately 120 µg/mL and was buffered to pH 7.4 using HEPES (12.5 mM). For further purification, the supernatant was dialyzed overnight using dialysis membranes with a 1000 Da molecular weight cut off. The purified CDs were subsequently diluted in 100 mL of HEPES buffer (12.5 mM) to prepare the stock solution used in further evaluations.

The environmental stability of the synthesized CDs was investigated under varying pH and temperature conditions. The fluorescence intensity of the CDs remained consistent at pH 7.4, indicating their applicability in biological and environmental samples with moderate pH fluctuations. Furthermore, thermal stability studies revealed that the CDs retained their optical properties up to 200 °C, affirming their robustness for biosensing and environmental monitoring applications. These findings demonstrate the adaptability and reliability of the EGTA-derived CDs under realistic operational conditions.

Sensing properties of CDs for quantitative detection for Fe3+ and AA

To evaluate the sensing capabilities of the synthesized carbon dots (CDs) at ambient temperature, photoluminescence (PL) spectra were recorded using a fluorescence spectrophotometer at an excitation wavelength of 320 nm. All sensing experiments were conducted in HEPES buffer (12.5 mM, pH 7.4). For Fe3+ detection, varying volumes of a 0.01 M Fe3+ solution (pH 7.4) was added to 2500 µL of the aqueous CD solution. The mixtures were sonicated for 1 min to ensure homogeneous dispersion before fluorescence measurements. The final Fe3+ concentrations ranged from 0 to 320 µM. The CDs demonstrated a strong and selective response toward Fe3+ ions, which prompted the evaluation of selectivity against other metal ions using a similar procedure.

To assess the detection of ascorbic acid (AA), 280 µM of Fe3+ was first introduced into 2500 µL of the CD solution, forming a CDs-Fe3+ complex. Subsequently, different concentrations of AA (0–300 µM) were added to this complex solution, followed by sonication for 1 min before fluorescence measurements. This system enabled the fluorescence “turn-on” response toward AA, illustrating the potential of the CDs–Fe3+ ensemble for selective AA sensing. A similar experimental approach was applied to evaluate the CDs’ response to other amino acids, confirming their selectivity for AA.

Fluorescence imaging, cytotoxicity assay and cell culture

Baby hamster kidney (BHK-2) cells were obtained from the Cell Bank of the Shanghai Science Academy. Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% calf serum, 100 µg·mL−1 neomycin, 100 µg·mL−1 streptomycin, and 100 U·mL−1 penicillin. The cells were maintained in a fully humidified incubator at 37 °C with 5% CO2 until approximately 80% confluence and appropriate morphology were achieved, at which point they were used for further experiments.

To investigate intracellular fluorescence behavior, the culture medium was removed, and the cells were incubated with 1.0 mL of fresh medium containing CDs for 2 h. After incubation, the cells were rinsed three times with phosphate-buffered saline (PBS) to remove excess CDs. For fluorescence imaging of Fe3+ and AA detection, cells preloaded with CDs or CDs–Fe(III) complexes were further incubated for 30 min at 37 °C. Subsequently, 160 µM Fe3+ and 280 µM ascorbic acid (AA) were added to the medium. Fluorescence changes were observed using an Olympus FV1000-IX81 laser confocal microscope with an excitation wavelength of 320 nm.

The cytotoxicity of CDs was evaluated using the Cell Counting Kit-8 (CCK-8) assay. BHK-2 cells were seeded into 96-well plates at a density of 1 × 104 cells per well in 100 µL of medium and cultured for 24 h. Cells were then exposed to varying concentrations of CDs (0, 10, 20, 40, 80, and 100 µg·mL−1) in serum-free medium for 24 h. After treatment, the medium was removed, and the cells were rinsed three times with PBS. Then, 10% CCK-8 solution in serum-free medium was added to each well and incubated for 2 h. The optical absorbance at 450 nm was recorded using a Berthold Mithras 2LB943 microplate reader. Control experiments using culture medium without CDs were conducted to determine baseline absorbance.

Results and discussion

Characterization of CDs

The morphology and size of the synthesized CDs were examined by transmission electron microscopy (TEM). As shown in Fig. 1a,b, the CDs are well-dispersed and exhibit a quasi-spherical shape. Dynamic light scattering (DLS) analysis further confirmed the average hydrodynamic diameter to be approximately 4.8 nm, consistent with previously reported values for similar CDs31, as shown in the size distribution plot (Fig. 1d). High-resolution transmission electron microscopy (HRTEM) was employed to evaluate the crystallinity of the CDs. The HRTEM images revealed lattice spacings of 0.24 nm and 0.23 nm, which correspond to the (102) plane of sp2 graphitic carbon and the (103) plane of sp3 hybridized carbon, respectively, suggesting the coexistence of graphitic and diamond-like domains within the CDs. These features are illustrated in Fig. 1c.

Fig. 1
figure 1

The CDs as prepared are displayed in representative TEM (a,b) and HRTEM (c) images. Image (c) shows the crystalline lattice. Histogram (d) shows the size distribution of the CDs.

Full size image

To further probe the structural characteristics, XRD analysis was conducted. The broad diffraction peak centered at 2θ ≈ 23° (Fig. 2) indicates the predominantly amorphous nature of the CDs. This result supports the notion that the CDs consist of an amorphous carbon shell with a disordered internal core, consistent with previous reports on carbon-based nanomaterials.

Fig. 2
figure 2

XRD pattern of CDs.

Full size image

Fourier-transform infrared (FTIR) spectroscopy was employed to elucidate the surface functional groups and chemical composition of the synthesized CDs. The FTIR spectrum of CDs is shown in Fig. 3. A broad absorption band observed in the region of 3100–3500 cm−1 corresponds to the stretching vibrations of O–H and N–H groups. The increased sharpness of this band suggests partial dehydration during the hydrothermal carbonization process. These observations indicate the presence of hydroxyl and amide functionalities, which play a crucial role in enhancing the aqueous dispersibility and stability of the CDs through surface passivation. A weaker absorption band at approximately 3033 cm−1 is attributed to C–H stretching vibrations. The strong band in the region of 1600–1700 cm−1 is assigned to C=O stretching vibrations, indicative of surface carboxylic acid groups (–COOH). Additionally, the distinct peak at 1724 cm−1 is characteristic of amide linkages (–CONH), further supporting the presence of nitrogen-containing surface functionalities. Other notable absorption bands include those at 1243, 1128, and 1067 cm−1, corresponding to the C–O–C stretching vibrations, likely derived from the ethylene glycol moiety, residual carboxylic groups, and ether linkages. The band near 1400 cm−1 is attributed to CH2 bending vibrations. Collectively, these findings confirm the presence of multiple polar functional groups on the CD surface, which not only improve hydrophilicity but also offer potential sites for analyte interaction in sensing applications.

Fig. 3
figure 3

The FTIR spectra of (a) pure EGTA, (b) CDs.

Full size image

X-ray photoelectron spectroscopy (XPS) was employed to analyze the surface elemental composition and chemical states of the synthesized CDs. The survey spectrum (Fig. 4a) exhibits three prominent peaks centered at ~286 eV (C 1s), ~408 eV (N 1s), and ~530 eV (O 1s), indicating that carbon, nitrogen, and oxygen are the predominant elements present in the CDs. Elemental quantification (Table S2) reveals that the CDs consist of 60.91% carbon, 10.82% nitrogen, and 28.27% oxygen, corroborating the presence of heteroatoms introduced during synthesis. To further investigate the chemical bonding environment of each element, high-resolution XPS (HR-XPS) spectra were analyzed (Figures 4b–d, ). The C 1s spectrum was resolved into peaks at 284.5 eV (C–C/C = C, sp2/sp3 carbon), 285.3 eV (C–N), 286.2 eV (C–OH), 286.8 eV (C–O–C/C = N), and 288.3 eV (C = O), reflecting the diverse carbon bonding states within the CDs. The deconvoluted N 1s spectrum shows two o Gaussian peaks located at 399.6 eV (N–H), and 401.6 eV (graphitic N, N–(C)₃), suggesting the incorporation of both amino and graphitic nitrogen species. The O 1s spectrum exhibits peaks at 531.1 eV (*O = C–O), 532.2 eV (C–O), and 533.2 eV (C=O/C–O–C), confirming the presence of carboxyl, carbonyl, and ether functionalities. These XPS findings suggest that the CDs consist of a carbon-rich core with abundant surface functional groups, including hydroxyl, carboxyl, amide, and nitrogen-containing species. The presence of such groups is favorable for aqueous dispersibility and contributes to the CDs’ ability to interact selectively with metal ions and biomolecules in sensing applications.

Fig. 4
figure 4

Full XPS spectrum of CDs (a), and High resolution XPS spectra of C 1s (b), (c) N 1s and (d) O 1s for the CDs.

Full size image

Optical properties of CDs

The optical properties of the synthesized carbon dots (CDs) were thoroughly investigated, including quantum yield (QY), fluorescence decay, emission/excitation spectra, and UV/vis absorption.

As shown in Figure S1 (Supporting Information), the UV/Vis absorption spectrum of CDs in 12.5 mM HEPES buffer at pH 7.4 exhibits broad absorption features with a shoulder at approximately 316 nm, which corresponds to the n-π* transition of CDs. Additionally, a peak at 236 nm is observed, attributed to the π-π* transition of C = O/C–OH/C = C/C = N bonds, consistent with the previously reported absorption characteristics of CDs31,32.

To explore the photoluminescent behavior of the CDs, emission spectra were recorded at various excitation wavelengths, ranging from 280 to 480 nm, as shown in Fig. 5. The maximum emission was observed at 407 nm when excited at 320 nm, indicating that the CDs exhibit excitation-dependent fluorescence. The emission spectrum exhibited a noticeable shift as the excitation wavelength was varied from 280 to 480 nm, demonstrating typical excitation-dependent behavior observed in other carbon nanoparticles. When exposed to UV light at 365 nm, the CDs in aqueous solution emitted a strong blue fluorescence, visible to the naked eye (Figure S2 in Supporting Information). The fluorescence stability of the CDs was also evaluated, and it was found that the fluorescence intensity remained nearly constant after six months, highlighting the exceptional stability of the CDs. This remarkable stability makes the CDs suitable for use as long-term fluorescent probes in various applications. When excited at 320 nm, the quantum yield of the current CDs was found to be 50% ± 0.05 using quinine sulphate (quantum yield = 54%) as a standard.

Fig. 5
figure 5

Fluorescence emission spectra of CDs in HEPES (12 mM) buffer at pH = 7.4 (excitation wavelength (λex) from 280 to 480 nm).

Full size image

Fluorescence turn-off of CDs in the presence of Fe3+

Iron (Fe3+) is an essential trace element for all living organisms, playing a pivotal role in numerous biological and chemical processes, including oxygen transport, storage, metabolism, electron transfer in enzymatic reactions, and mitochondrial respiration. However, excessive iron accumulation in the body can lead to toxicity, causing damage to vital organs such as the kidneys and liver11. The U.S. Environmental Protection Agency (EPA) sets the maximum permissible concentration of iron in drinking water at 0.3 mg·L⁻1 (~ 5.357 μM). As a result, accurate and rapid detection of Fe3+ is of utmost importance for both biological systems and environmental monitoring.

Recent studies have focused on carbon dots (CDs) as fluorescence probes, with most fluorescence detection systems relying on the "turn-off" model33. In these systems, Fe3+ quenches the fluorescence of the fluorophore, but interference from ions like Cu2⁺, Pb2⁺, Hg2⁺, and Ag⁺ can cause poor selectivity24,26,34. Consequently, developing new fluorescence probes for Fe3+ detection that are both highly sensitive and selective remains a challenge. In this study, we introduce a "turn-on" sensing mechanism, wherein Fe3+ first quenches the CDs’ fluorescence and then restores it, offering a promising approach for Fe3+ detection. To explore this system, a variety of metal cations were tested for their quenching effects on the blue-emitting CDs. Transition metal complexation significantly influences the fluorescence of many organic compounds, as the intersystem crossing rate from singlet to triplet is often higher in species with unpaired electron spins. Paramagnetic transition metal ions, such as Fe3+, Ni2⁺, Cu2⁺, and Co2⁺, may quench fluorescence through intramolecular energy transfer.

As shown in Fig. 6 and Fig. S2 (Supporting Information), CDs exhibit strong electrostatic interactions and electron transfer with Fe3+, making them an efficient probe for Fe3+ detection. Functional groups such as –OH and –COOH on the CD surface bind Fe3+ through coordination or electrostatic interactions, demonstrating a significantly higher binding affinity for Fe3+ than for other tested metal ions like Zn2⁺, Cu2⁺, or Na⁺. This selective binding quenches the fluorescence, while other metal ions do not exhibit the same effect. The synergistic action of these functional groups enables the formation of stable complexes with Fe3+, distinguishing it from other metal ions. Even at concentrations up to 450 µM, ions such as Hg2⁺, Mg2⁺, Al3+, Co2⁺, Li⁺, Ni2⁺, Zn2⁺, K⁺, Mn2⁺, Na⁺, Cr3+, NH4⁺, Fe2⁺, Ag⁺, Cu2⁺, and Ca2⁺ did not significantly affect the UV/Vis absorption or fluorescence spectra, confirming the CDs’ strong selectivity for Fe3+.

Fig. 6
figure 6

Fluorescence spectra of CDs in HEPES (12 mM) buffer at pH 7.4 with varying concentrations of Fe3+. Top to bottom Fe3+ concentration: (0, 0.4, 2, 4, 6, 12, 16, 20, 24, 28, 32, 36, 40, 48, 56, 64, 72, 78, 84, 90, 96, 100, 106, 112,120, 128, 136, 144, 152,160, 170, 180, 190, 200, 220, 240, 280, and 320 µM) (λex = 320 nm).

Full size image

The high selectivity of the CDs extends to ascorbic acid (AA), which efficiently displaces Fe3+ from the CD surface due to its reducing properties. This displacement restores the fluorescence in a specific "on–off–on" mechanism. This response is highly sensitive to AA, with no observed effect from other biomolecules such as dopamine or amino acids. Tests with various metal ions and biomolecules showed minimal interference, further confirming the CDs’ exceptional selectivity for both Fe3+ and AA. This characteristic makes the CDs highly effective for sensitive and specific detection in complex biological and environmental samples.

As the concentration of Fe3+ increased, the fluorescence intensity of the CDs decreased dramatically, as shown in Fig. 6. However, no significant spectral changes (such as the appearance of new emission bands or spectral shifts) were observed as the Fe3+ concentration continued to rise. At higher Fe3+ concentrations, there was a significant decrease in both fluorescence intensity and fluorescence quantum yield. The CDs-Fe (III) nanocomposite efficiently controlled charge recombination and accelerated excitation transmission, leading to a clear quenching of fluorescence. The fluorescence emission spectra of CDs as a function of Fe3+ concentration are depicted in Fig. 6. At Fe3+ concentrations approaching 200 µM, the fluorescence at 320 nm was nearly completely quenched, with 98% of the fluorescence intensity muted. The fluorescence quantum yield at [Fe3+] = 200 µM was measured to be 0.002, indicating almost complete fluorescence quenching. This concentration of Fe3+ was chosen to ensure a complete "turn-off" of the fluorescence for subsequent detection of AA.

Furthermore, the fluorescence intensity decreased in a nearly linear fashion at 320 nm with increasing Fe3+ concentration (0–64 µM) in HEPES buffer at pH 7.4. This relationship allows for the quantitative analysis of Fe3+, as illustrated in Fig. S3 (Supporting Information), which shows a strong correlation between the quenching efficiency (F) and the concentration of Fe3+ in the 0–64 µM range. A calibration curve for Fe3+ concentration in HEPES buffer at pH 7.4 is shown in Fig. S3 (Supporting ). After adding Fe3+ to HEPES buffer, the fluorescence peak intensity is represented by the letter F, providing a reliable tool for Fe3+ detection in complex samples.

$${\text{F}} = {8500} - {56}{\text{.5}} \times \left[ {{\text{Fe}}^{3 + } } \right]({\text{r}} = 0.9992,\left[ {{\text{Fe}}^{3 + } } \right] = 0 - 64\mu {\text{M}}).$$

The calibration equation provides a quantitative basis for determining Fe3+ in aqueous solutions. Compared to the method reported by Gong et al., our approach requires a lower concentration of Fe3+ (200 µM) to fully "turn-off" the fluorescence. This improvement can be advantageous for the development of sensors, as it may reduce production costs while maintaining high sensitivity. Additionally, it is important to note that the presence of Fe3+ in the sample may interfere with the quantification of ascorbic acid (AA), which is the primary target analyte in this study. After the CDs are dissolved in HEPES buffer solution and exposed to Fe3+ and AA, the UV/Vis absorption spectra are shown in Fig. S2 of the Supporting Information.

The following equation was used to estimate the detection limit (DL):

$${\text{DL }} = { 3} \times \sigma /{\text{k}}.$$

whereby k represents the calibration curve’s slope and σ is the blank sample’s standard deviation. In an aqueous solution, at 320 nm, the DL of Fe3+ was 23 nM.

The sensitivity and specificity of the CDs for Fe3+ detection were rigorously evaluated in the presence of various potential interfering metals, including Hg2⁺, Mg2⁺, Al3+, Co2⁺, Li⁺, Ni2⁺, Zn2⁺, K⁺, Mn2⁺, Na⁺, Cr3+, NH4⁺, Fe2⁺, Ag⁺, Cu2⁺, and Ca2⁺, under the same excitation conditions (320 nm) as Fe3+ (100 µM). Specificity tests demonstrated that fluorescence quenching occurred exclusively in the presence of Fe3+, even at a higher concentration of 500 µM, as shown in Fig. 7. This result highlights the CDs’ exceptional hypersensitivity and hyperselectivity toward Fe3+, likely due to the strong binding affinity between Fe3+ and the CDs’ surface functional groups, such as carboxyl and hydroxyl groups.

Fig. 7
figure 7

CDs’ (120 µg/mL) fluorescence response to various metals (500 µM), Fe3+ (80 µM), and HEPES (12 mM) buffer at pH 7.4 (λex = 320 nm).

Full size image

35

Fluorescence “turn-on” of CDs-Fe (III) in the presence of AA

The observed fluorescence increase occurs if the quencher cation’s capacity to enhance the rate of intersystem crossing through mechanisms like a shift in metal oxidation state is compromised, or if the quencher cation’s interaction with the fluorophore is disturbed. This principle forms the basis of our AA detection "turn-on" model. Ascorbate anions, the main form of AA, are stabilized primarily through electron delocalization. Due to its reducing and antioxidant properties, AA readily interacts with Fe3+, causing a reduction in the oxidation state from + 3 to + 2. This reduction weakens the complexation between CDs and Fe3+, which potentially restores the fluorescence of the CDs. To investigate the efficiency of this sensing mechanism, we employed micro-titration. The original CDs-Fe(III) solution exhibited minimal fluorescence due to the strong quenching effect of Fe3+. However, when AA was added, there was an immediate increase in fluorescence intensity, and the emission spectrum was measured after 1 min. As the Fe3+ concentration was raised from 24 µM to 280 µM in the presence of 300 µM AA, the fluorescence intensity increased 10 to 20 times at 320 nm. The fluorescence intensity progressively recovered with increasing AA concentrations from 0 to 300 µM at 320 nm (Fig. 8). This suggests that, in the presence of AA, the redox-induced release of Fe3+ from the CD surface leads to the recovery of the CDs-Fe(III) PL intensity. A strong linear correlation (Fig. S4 in Supporting Information) was observed between the recovered fluorescence intensity (F) of CDs-Fe(III) and the AA concentration at 320 nm, ranging from 0 to 52 µM. The estimated detection limit of 21 nM is significantly lower than those reported in previous fluorescence probes (Table S1 in Supporting Information). These results demonstrate that our sensing system exhibits exceptional selectivity, making this probe highly sensitive for detecting AA.

Fig. 8
figure 8

Fluorescence spectra of CDs-Fe (III) at pH 7.4 in HEPES (12 mM) buffer containing varying concentrations of AA. AA concentration in ascending order: (0, 2, 4, 6, 8, 10, 12, 14, 16, 24, 32, 40, 52, 64, 76, 88, 100, 120, 140, 160, 180, 200, 260 and 300 µM) (λex = 320 nm).

Full size image

Selectivity of the CDs-Fe (III) system and interference measurements

The selectivity of a high-quality sensing probe must be guaranteed and thoroughly assessed to ensure it is free from external interferences and false positives. To evaluate the selectivity of the CDs-Fe (III) system for AA detection, fluorescence responses to dopamine and other amino acids at the same concentration were also examined. A volume of 2500 µL of CDs-Fe (III) was combined with 500 µM of dopamine and various amino acids. As shown in Fig. 9, no significant fluorescence change was observed upon the addition of dopamine or amino acids, suggesting that no competitive reaction occurred between these substances and AA. These findings confirm that the CDs-Fe (III) system is highly selective for AA, making it a promising candidate for intracellular AA detection.

Fig. 9
figure 9

CDs-Fe (III) fluorescence response to various anions and amino acids (HEPES (12 mM) buffer at pH 7.4) (λex = 320nm). The anions and amino acids are all concentrated at 500 µM.

Full size image

Possible sensing mechanism

To gain deeper insight into the quenching mechanism, we calculated the quenching equation using the Stern–Volmer equation: F0/F = 1 + ksv[Q] where F0 is the initial fluorescence intensity (Fe3+) without the quencher, Ksv is the static Stern–Volmer equilibrium constant, and F is the fluorescence intensity observed in the presence of the quencher. The quenching behavior of Fe3+ in the CDs system displayed a good linear Stern–Volmer relationship, as shown in Fig. S6 in the supporting information. The addition of Fe3+ and AA to the CDs aqueous solution did not affect the fluorescence or lifetime of the CDs, suggesting that the quenching process is attributed to the formation of a stable, non-fluorescent complex between the CDs and Fe3+, following a static quenching mechanism.

To further investigate the fluorescence dynamics of the CDs, the single-photon timing technique was employed to record the fluorescence decay traces at 320 nm33. In pure CDs water, the fluorescence decay exhibited a tri-exponential behavior, with lifetimes of approximately 1.55 ns (14.49%), 5.48 ns (44.01%), and 14.86 ns (41.50%). Upon the addition of 280 µM Fe3+ and 300 µM AA, the fluorescence decay time remained unchanged (Fig. S8 in supporting information). The recorded lifetimes in the presence of CDs-Fe (III) or AA were as follows: τ₁ = 1.16 ns (10.28%), τ2 = 4.51 ns (45.66%), τ₃ = 13.61 ns (44.06%) for Fe3+, and τ₁ = 1.14 ns (11.04%), τ2 = 4.24 ns (45.65%), τ₃ = 11.58 ns (43.31%) for AA. These values were consistent with the fluorescence lifetimes of pure CDs (τ₁ = 1.55 ns, 14.49%; τ2 = 5.48 ns, 44.01%; τ₃ = 14.86 ns, 41.50%), indicating that the addition of Fe3+ and AA did not significantly affect the fluorescence decay dynamics (Table S4 in supporting information).

After Fe3+ binding, the CDs fluorophore undergoes electron transfer (ET) to the metal center, leading to fluorescence quenching. As the concentration of Fe3+ increases, the proportion of the non-fluorescent complex also rises, resulting in a corresponding decrease in fluorescence intensity when compared to the fluorescent form of CDs. Notably, the emission maximum and fluorescence lifetime of this non-fluorescent complex remain unchanged upon Fe3+ addition, further supporting the static quenching mechanism.

Recent literature has reported several fluorescent CD-based chemosensors for Fe3+ and AA detection35. Tables S1 and S2 in the Supporting Information provide a comparative overview of the performance of our system against previously reported CD-based fluorescence methods for detecting Fe3+ and AA. These comparisons demonstrate that our CDs exhibit superior sensitivity and selectivity, highlighting their potential as highly effective optical sensors for Fe3+ in aqueous media.

Thus, the CDs system described here offers a robust and selective platform for Fe3+ and AA detection. A schematic illustration of the interaction mechanism between CDs, Fe3+, and AA is presented in Fig. 10.

Fig. 10
figure 10

Schematic representation of CDs, how Fe3+ and AA bind to CDs, and how Fe3+and AA quench on–off-on CDs.

Full size image

Application of CDs vitamin C tablets and in human kidney cell

Application to real sample

This interaction allows the CDs to act as a Fe3+-selective turn-off probe, with fluorescence quenching occurring, making them highly suitable for real-world applications.Our CDs-based system demonstrates superior or comparable sensitivity and selectivity when compared to commercial assays and other fluorescent probes. The detection limits for Fe3+ (23 nM) and ascorbic acid (AA) (21 nM) are among the lowest reported for fluorescence-based detection of these analytes. In contrast, commercial fluorescent probes for Fe3+ typically exhibit detection limits in the micromolar range, significantly higher than those achieved with our CDs. For Fe3+, the detection limit is well below the World Health Organization’s (WHO) recommended limit of 300 µg/L (~ 5.37 µM) for iron in drinking water, making the CDs ideal for environmental monitoring of trace iron levels[35]. In human serum, Fe(III) concentrations typically range from 1–2 µM, meaning the detection limit of 23 nM is sufficient for practical use in biological systems. Similarly, the detection limit for AA aligns with the range required for its detection in biological and pharmaceutical applications, where typical AA concentrations in blood plasma are between 30 and 60 µM. Moreover, other sensors for AA typically rely on complex modifications, such as functionalizing agents or co-precursors. In contrast, our approach utilizes a simple one-step hydrothermal synthesis, eliminating the need for surface passivation or additional reagents. This makes the CDs not only highly sensitive but also cost-effective and user-friendly for practical applications.The high selectivity of our system for Fe3+ and AA, even in complex biological and environmental samples, is attributed to the specific interactions between the CDs’ functional groups and target analytes. This ensures reliable detection with minimal interference from other metal ions or biomolecules, a challenge for many commercial assays and fluorescent probes that often suffer from cross-reactivity or require extensive sample preparation.Moreover, our approach offers practical advantages. The CDs are synthesized via a low-cost, straightforward procedure that requires no additional reagents or functionalization steps, making the method affordable and scalable for large-scale applications. This combination of simplicity, sensitivity, and selectivity positions our CDs as a highly competitive alternative to existing commercial assays and fluorescent probes for detecting Fe3+ and AA in real-world applications.We evaluated the practical applicability of the proposed CDs-based fluorescent "turn-on" sensing probe for AA detection using a real-world sample: the commercially available water-soluble vitamin C tablet, Novartis CaC-1000 Plus. This supplement, commonly consumed by the general public, including pregnant women, contains 1000 mg of calcium lactate gluconate (Novartis Specs), 327 mg of CaCO₃, 500 mg of Vitamin C (ascorbic acid, Ph.Eur), 100 IU of Vitamin D₃ (Ph.Eur), and 10 mg of Vitamin B₆ (B.P.).

The ascorbic acid content in the tablet was determined using the CDs-based sensing probe and compared with results obtained via inductively coupled plasma atomic emission spectroscopy (ICP-AES). As shown in Table 1, the recoveries obtained by the fluorescence method were in good agreement with those from ICP-AES, indicating the reliability of the CDs probe.

Table 1 AA content measurement in vitamin C tablets*
Full size table

Importantly, the presence of other components in the tablet, such as calcium and vitamin D₃, did not interfere with the fluorescence response. This was confirmed by the selectivity study shown in Fig. 7, where Ca2⁺ ions exhibited negligible influence on the probe’s fluorescence. These results validate the use of the CDs-Fe3+ complex-based "off–on" fluorescence sensing system for real sample analysis, highlighting its high selectivity and applicability in complex matrices.

Evaluation of Fe3+ and AA in BHK-2 cells

BHK-2 cells were used as a real biological model to evaluate the performance of the proposed CDs-based molecular probe for detecting Fe3+ and AA in cellular environments an essential step toward practical biomedical applications. Confocal microscopy was employed to assess the ability of CDs to monitor intracellular AA levels in living cells. Initially, BHK-2 cells were incubated with CDs, followed by exposure to either 140 μM AA or  μM Fe3+. The bright-field microscopy image of the untreated BHK-2 cells is shown in Fig. 11(1a). Under UV excitation, cells incubated with CDs displayed strong blue fluorescence (Fig. 11(2b)), indicating effective uptake and internalization of CDs. As seen in the image, the fluorescence is predominantly localized in the cytosol and along the cell membrane, with minimal accumulation in the nucleus. Upon incubation with Fe3+ for 30 min at 37 °C, a significant decrease in intracellular fluorescence was observed (Fig. 11(3b)), suggesting successful quenching by Fe3+. This confirms that CDs can permeate the cell membrane likely via endocytosis and effectively interact with intracellular Fe3+ ions. Subsequent treatment of the CDs/Fe3+-loaded cells with AA led to a noticeable restoration of fluorescence (Fig. 11(4b)). This fluorescence "turn-on" effect is attributed to the strong affinity of AA for Fe3+, which leads to the formation of a Fe3+ AA complex. As AA competes with CDs for Fe3+ binding, the disruption of the CDs–Fe3+ interaction results in fluorescence recovery. Therefore, after 30 min of treatment with AA, enhanced intracellular fluorescence was clearly observed (Fig. 11b), demonstrating the effectiveness of this CDs-based probe as an intracellular “off–on” sensor for AA via competitive interaction with Fe3+.

Fig. 11
figure 11

Live BHK-2 cell confocal fluorescence images. (1) 1a: Fluorescence image of pure BHK-2 cells stimulated by UV light; 1b: Bright field microscopy image of pure BHK-2 cells. 1c: an ultraviolet light-excited layer of pure BHK-2 cells. (2) BHK-2 cells cultured with CDs for one hour at 37 °C were shown in a fluorescence image. (3) Fluorescence picture of BHK-2 cells stained with CDs after 30 min at 80 μM Fe3+. (4) Fluorescence picture of CDs- Fe3+ stained BHK cells after 30 min at 140 μM AA.

Full size image

Bright-field transmission imaging confirmed the viability of BHK-2 cells treated with CDs and Fe3+ during the imaging experiments (Fig. 11(2a,3a)). These results are consistent with the findings from the steady-state fluorescence experiments discussed in the previous section. Furthermore, the cytotoxicity of CDs was evaluated using the MTS assay in BHK-2 cell lines. As shown in Fig. S7 (Supporting Information), CDs exhibited low cytotoxicity, suggesting their good biocompatibility. The results indicate that CDs can be safely used for intracellular applications. Even in the presence of potentially interfering substances, such as proteins and amino acids, the CDs-based probe effectively captured and responded to analytes, highlighting its potential for reliable bioimaging and sensing in complex biological environments.

Conclusion

This study demonstrates the successful synthesis and application of EGTA-derived carbon dots (CDs) for the sensitive and selective detection of ascorbic acid (AA) and Fe3+. The CDs were synthesized through a one-step hydrothermal procedure, eliminating the need for functionalizing or surface passivating agents and simplifying the production process. These CDs exhibit excitation-dependent fluorescence, enabling their use as label-free fluorescent nanoprobes for detecting various analytes with high accuracy. They facilitate the quantitative detection of AA and Fe3+ with detection limits of 23 nM and 21 nM, respectively, using a straightforward "on–off-on" fluorescence switching technique in an aqueous medium, demonstrating their sensitivity for trace-level detection. The CDs also show excellent selectivity at 320 nm, ensuring reliable detection of target analytes in complex samples compared to existing fluorescent probes. Fluorescence imaging studies in living cells confirm their practical utility, highlighting their potential for real-time biological applications, particularly in cellular monitoring. Importantly, the CDs are fast, sensitive, and specific, requiring no additional chemical modifications, which simplifies the process and reduces costs. However, this study has certain limitations. For instance, while the CDs show promising results in controlled laboratory settings, their performance in real-world applications, such as environmental or clinical samples, has not been fully validated.