Detection of ferric ions by nitrogen and sulfur co-doped potato-derived carbon quantum dots as a fluorescent probe

This paper reports the detection of ferric ions (Fe3+) based on nitrogen and sulfur co-doped carbon quantum dots. These nitrogen and sulfur co-doped carbon quantum dots were synthesized via a hydrothermal route using northern Shaanxi potatoes as carbon sources and ammonium sulfate as nitrogen and sulfur sources. The quantum yields of the carbon quantum dots were found to be 16.96% and 4.23% with and without doping, respectively. The structural details, morphology, and optical properties of carbon quantum dots were analyzed using Fourier transform infrared spectroscopy (FT-IR), x-ray photoelectron spectroscopy (XPS), transmission electron microscopy (TEM), ultraviolet-visible absorption spectroscopy (UV–vis), and fluorescence spectroscopy. The as-prepared co-doped carbon quantum dots were utilized as a fluorescent probe for detecting Fe3+ ions, where the fluorescence intensity of carbon quantum dots was remarkably quenched in the presence of Fe3+ ions. A good linear relationship for Fe3+ ion detection was obtained from 0 to 500 μmol/L with a detection limit as low as 0.26 μmol/L. Furthermore, the proposed method also provided satisfactory results in the tap water.


Introduction
Ferric ion (Fe 3+ ) is one of the essential metal ions in living systems, which plays indispensable roles in many physiological processes, such as oxygen transport and storage, electron transfer, enzymatic catalysis, etc However, an overdose Fe 3+ metal ions in the body can cause liver damage, kidney failure, or even death.Therefore, the rapid, sensitive, and reliable detection of Fe 3+ ions is critical concerning various health risk hazards [1,2].In recent years, numerous analytical techniques and detection methods for Fe 3+ ions have been developed, including voltammetry, atomic absorption spectrometry, and so forth [3][4][5].However, several inherent shortcomings limit these methods as they are time-consuming, expensive, and require complicated instrumentation or synthetic processes.In this regard, the fluorescence quenching approach has been explored for the analysis as it is simple, portable, has short response times, provides higher sensitivities, and is operational in solution and solid phases.
Among various nanomaterials that have been explored and designed for metal ion detection, carbon quantum dots (CQDs) are emerging as a fluorescence-sensing platform for sensing applications.They are zerodimensional carbon-based nanomaterials with good water solubility, stable photoluminescence, wide excitation range, and low cytotoxicity.Moreover, the surface of CQDs can be easily functionalized with other molecules giving them an edge over conventional semiconductor quantum dots and organic dyes [6,7].Recently, CQDs have been employed as fluorescent probes for various metal ion detection [8][9][10][11].It has been reported that CQDs can be prepared using biomass in one step by different synthetic methods, such as candle burning, in situ dehydration reactions, laser ablation methods, etc [12][13][14].This work employed a hydrothermal method to synthesize CQDs and nitrogen and sulfur co-doped CQDs (NSCQDs).This technique is easy to handle and yields high quality crystals with controllable particle shape and size [15].The rationale behind doping CQDs with heteroatoms such as nitrogen and sulfur was to achieve high quantum yields and stable properties without additional functionalization.In previous reports, N-doped CQDs were observed to have higher quantum yields than their pure counterpart [16].Moreover, other heteroatoms, such as boron, phosphorus, etc, manipulate the electronic structure and surface defects of CQDs and form more n-or p-carriers, resulting in a change in optical properties [17].Nevertheless, nitrogen and sulfur co-doped CQDs from biomass have rarely been used as fluorescent probes and thus need further investigation.
In this regard, the readily available potatoes in northern Shaanxi Province were employed as a precursor for the hydrothermal synthesis of CQDs and NSCQDs.Affordable potatoes have a high annual yield and are rich in carbohydrates, fats, cellulose, and minerals; thus, they are a great source of carbon content for synthesis.As mentioned earlier, the optical properties are highly dependent on elemental doping; therefore, nitrogen and sulfur are doped in CQDs to achieve a high quantum yield.The effect of different temperatures, time, doping content, and the quantity of carbon source was studied to determine the optimal synthesis conditions.Furthermore, the as-synthesized NSCQDs were explored for their potentiality as a fluorescent probe for the selective and sensitive detection of Fe 3+ ions.The detection was based on the fluorescence method by observing the change in intensity at 440 nm, which was dramatically quenched in the presence of Fe 3+ ions.

Preparation of NSCQDs
Firstly, potatoes were pre-treated to synthesize the nitrogen and sulfur co-doped CQDs.Potatoes were mixed with the saturated ammonium sulfate solution in a certain ratio.The solution was poured into a Teflon container with 100 ml capacity, and then the different reactions were set at a certain temperature for different times hydrothermally [18].Then, the autoclave used for the reaction was cooled to room temperature, and the yellow-brown solution was centrifuged at 5000 rpm for 30 min.The optimal reaction conditions were the temperature of 200 °C with a reaction time of 12 h.Later, the product was filtered with a 0.22 μm microporous membrane to obtain the NSCQDs solution.The schematics for the synthesis process as shown in figure 1.The obtained solution was stored in a refrigerator at 4 °C until further use.

Structure characterization
The specific structural details and morphology were obtained by using a JEM-2100F transmission electron microscope operated at an acceleration voltage of 200 keV.The XPS spectra were recorded on a Kratos Axis ULTRA x-ray photoelectron spectrometer with Al Kα as the x-ray source to provide further evidence of the product's composition.The FTIR spectra were recorded on an IR Prestige-21 spectrometer in the range of 400 cm -1 to 4000 cm -1 .The Fluorescence measurements were performed using a Shimadzu RF-5301PC spectrometer equipment with a 1 cm quartz cell, while the UV-vis absorption spectra were taken on a Shimadzu UV-2450.The fluorescence quantum yield (QY) of CQDs was determined in 0.1 M aqueous sulfuric acid solution by using quinine sulfate as the standard with a QY of 0.54 at an absorbance below 0.05.The pH values were measured with a PHS-3C pH meter.

Detection of Fe 3+
Firstly, 1 ml of NSCQDs was taken, and diluted to make up 100 ml solution.Then, 3 ml of the diluted solution was taken in 4 different test tubes, where 3 ml of a buffer solution with pH = 8 was added.This step was followed by adding 3 ml of Fe 3+ solution with a certain concentration into the test tube named 1, 2, and 3, respectively.Test tube 4 was marked as blank and thus was added with water.The whole solution was shaken at room temperature for 5 min for the fluorescence measurements.The change in fluorescence intensity of NSCQDs was recorded at the excitation wavelength, λ ex = 335 nm, with an emission at 440 nm.

Surface characterization of NSCQDs
The morphology, microstructure, and particle size of as-prepared NSCQDs were characterized by TEM, as shown in figure 2(a).As observed, the approximately spherical (quasi-spherical) particles of NSCQDs were prepared, which showed good uniformity and dispersity as there were no signs of aggregation.The particle size distribution calculated from the TEM images is presented in figure 2(b), showing the distribution in the range of 3 to 8 nm with an average particle size of 5 nm.
The optical properties of NSCQDs are closely related to their functional groups, such as hydroxyl and carbonyl groups, which can be confirmed through FT-IR.As shown in figure 3(a), the spectra showed the band within the range of 2950-3300 cm -1 , ascribed to the typical stretching vibration of O-H/N-H bonds.The peaks at 2975 cm -1 and 1685 cm -1 corresponded to -CH 3 /-CH 2 and C=O stretching vibrations, respectively.Furthermore, the absorption peak at 1442 cm -1 was attributed to the C-N bending mode, whereas the peaks at 1115 cm -1 and 680 cm -1 corresponded to C-O and C-S stretching vibrations, respectively [19].The above results confirmed the presence of hydrophilic functional groups and nitrogen and sulfur doping in the system.The composition and chemical state of the synthesized NSCQDs were also determined by XPS measurements.The survey spectrum confirmed the presence of C (77.26%), N (4.06%),O (18.43%), and S (0.25%) in the material, as shown in figure 3(b).

Optical properties of NSCQDs
The optical properties of the NSCQDs are shown in figure 3. The NSCQDs show a strong UV absorption peak at about 275 nm (figure 4(a)), which is attributed to the π-π * transition of the sp2 C=C bonds.The microabsorption peak at about 325 nm was also due to the n-π * transition induced by nitrogen doping [21].
The as-prepared NSCQDs showed excitation-dependent fluorescence spectra, where the excitation wavelength (λ ex ) was increased from 305 nm to 375 nm, and the corresponding emission spectra were recorded.It was observed that the strongest emission spectra of NSCQDs were obtained at an emission wavelength, λ em = 440 nm, when NSCQDs were excited at 335 nm.With the increasing excitation wavelength from 335 nm to 375 nm, emission peaks red-shifted with the decreasing fluorescence intensity; however, the red shift of the emission is relatively small due to the uniform particle size of NSCQDs.The overall Stokes shift from excitation to emission wavelength was 105 nm, which indicated the excitation-dependent behavior of NSCQDs [22].The stability of the fluorescence intensity of NSCQDs was investigated at different pH values and over time.It was found that the fluorescence intensity of NSCQDs gradually increased when the pH was increased from 1.0 to 8.0, followed by a decrease in emission intensity when the pH was further increased to 13.0, as shown in figure 4(c), the test time was 1 h.The observed change in intensity under strongly acidic and basic conditions can be attributed to the destruction of functional groups, such as carboxyl (-COO-) and amino (-NH 2 ) groups, on the surface of the CQDs [23,24].Under extreme pH conditions, excess H + or OH -ions induced significant changes in the functional groups caused by the protonation or deprotonation, thus, changing the optical properties of NSCQDs, leading to quenching.The highest fluorescence intensity of NSCQDs was observed at pH 8.0, indicating the stability of particles in a weak alkaline environment.In addition, the influence of incubation time on the fluorescence intensity was also studied, as shown in figure 4(d).Initially, there was a decrease in the fluorescence intensity of NSCQDs at room temperature with the change in time.Later after three days, the intensity declined more slowly than expected.Also, there was no notable change even after ten days showing the stability of the as-prepared NSCQDs.
Additionally, the quantum yield (QY) of NSCQDs was determined by the typical slope method by using the quinine sulfate dihydrate dissolved in 0.1 M H 2 SO 4 as the standard (QY = 0.54).The fluorescence emission  spectra of NSCQDs (excited at 335 nm) with quinine sulfate solution at different concentrations and their corresponding absorption spectra (UV-vis spectra) were recorded.The integrated fluorescence intensities and the absorbance values (at 335 nm) of the sample with the reference sample were plotted against each other to calculate the slope values [25].In order to minimize the reabsorption effect, the absorbance was maintained below 0.05.The QY was then calculated according to the following equation: where j represents the fluorescence quantum yield, m represents the standard slope of the integrated intensities of the emission spectra plotted against the absorbance values, and η is the refractive index of the solvent (refractive index η x /η s is 1).'s' and 'x' represent the quinine sulfate standard and NSCQDs sample, respectively.The calculated quantum yields of CQDs and NSCQDs were 4.23% and 16.96%, respectively, indicating the effect of heteroatom doping on the quantum yield of CQDs, as shown in figure 5.

Optimization of synthesis conditions of NSCQDs
The effects of reaction time, temperature, amount of potato, and the volume of ammonium sulfate on the fluorescence intensity of NSCQDs were investigated in detail.One can see in figure 6(a) that initially, the fluorescence intensity of NSCQDs increased and then decreased with the reaction time, with the highest intensity occurring at 12 h.Thus, 12 h was selected as an optimal reaction time.Furthermore, other parameters were kept constant to see the effect of temperature on the reaction conditions.In this case, the fluorescence intensity of NSCQDs first increased and then decreased with the increasing synthesis temperature and reached its maximum at 200 °C.As a result, 200 °C was chosen as the synthesis temperature.As shown in figure 6(c), when the amount of potato was increased, the emission intensity increased at first, reaching the maximum at 15 g, and later on decreased with a further increase in the amount.Therefore, 15 g was the optimal amount of potato for synthesis.Finally, the effect of the volume of ammonium sulfate on the reaction was studied, as shown in figure 6(d).When the volume of saturated ammonium sulfate was increased, the fluorescence intensity of NSCQDs increased, then decreased slightly and reached its maximum at a volume of 30 ml, which was optimal for effective doping.It was concluded that the optimal conditions for synthesizing NSCQDs were as follows: a reaction time of 12 h, a reaction temperature of 200 °C, 15 g of potato, and 30 ml of saturated ammonium sulfate solution.

Fe 3+ detection and sample analysis 3.3.1. Selectivity and interference test
The impact of different metal ions on the fluorescence intensity of NSCQDs was also studied to check the selectivity of the as-prepared probe [26].The measurements were performed at an excitation wavelength of 335 nm.Based on figure 7(a), the as-prepared NSCQDs showed strong fluorescence quenching when Fe 3+ ions were added.Despite the presence of other metal ions, quenching of the fluorescence intensity due to the Fe 3+ ion was found, having no interference from other ions, as depicted in figure 7(b).Therefore, the sensor showed high selectivity for determining Fe 3+ ions.
To assess the possibility of the analytical application of the as-prepared probe regarding the sensitivity, the impact of different concentrations of Fe 3+ ions was analyzed.The different concentrations used in this experiment were 0, 3,10,15,20,25,35,45,100,150,200,250,300,360,400,450, and 500 μM.Under the optimum conditions, fluorescence quenching by Fe 3+ ions was analyzed using the plot between the fluorescence intensity of NSCQDs at 440 nm and the concentration of Fe 3+ ions.As displayed in figure 8(b), a significant linear correlation (F/F o = 0.9983-0.0015C; R 2 = 0.9974) existed between them in the 0-40 μmol/L range.Similarly, the linear relationship was also observed in the 40-500 μmol/L range following the equation F/F o = 0.9706-0.0011C with R 2 = 0.9971.It should be mentioned here that the detection limit was 0.26 μmol/L, revealing the high sensitivity of the as-synthesized NSCQDs.It also meets the detection level limit of Fe 3+ ions [0.3 mg L −1 (5.36 μmol/L)] stipulated in GB 5749-2022 Hygienic Standard for drinking water [27].

Real sample detection
In order to further demonstrate the practicality of the proposed sensor, NSCQDs were used to detect Fe 3+ ions in simulated water samples using the standard addition method [28].The results shown in table 1 revealed the recoveries of Fe 3+ ions in two kinds of water were between 95.07% and 103.1%, with a relative standard deviation (RSD, n = 3) of 1.0% − 4.5%.Comparison of the sensitivities and linear ranges of other systems for the detection of Fe 3+ , the result of table 2 shown this work was better than others.

Fluorescence quenching mechanism
The synthesized NSCQDs have many hydroxyl and carboxyl groups on their surface, and Fe 3+ ions have a stronger affinity for these groups to form metal-complex structures [32].The time-resolved PL decay profile of NSCQDs is shown in figure 9.In addition, the corresponding lifetimes are calculated by fitting to exponential functions with the use of iterative reconvolution.The fluorescence lifetime of NSCQDs τ1 is 69.09 ns, τ2 is 14.90 ns, and the average lifetime is 41.99 ns.Compared to the average lifetime of 6.72 ns of pristine S-N-C-dots [33], dramatic longer fluorescence lifetimes of both τ1 and τ2 were obtained on our sample.It has been reported  that τ2 fluorescence lifetime revealed recombination nature of excitations [34,35].These structures may induce electron transfer from NSCQDs to the unfilled orbitals of Fe 3+ ions, resulting in non-radiative electron-hole recombination and, thus, in the quenching of the fluorescence intensity of NSCQDs [36].The quenching of intensity in NSCQDs can also be ascribed to the presence of C-S in the structure, the mechanism of Fe 3+ detection by NSCQDs as the fluorescent probe is shown in figure 10.Since sulfur has low electronegativity and large atomic radii; thus, they are more prone to lose valence electrons to Fe 3+ ions.This lead to fluorescence quenching produced by an internal filtration effect due to Fe 3+ ions [37].

Conclusion
In this work, biomass potato is used as carbon source, ammonium sulfate is used as nitrogen and sulfur dopant, and one-step hydrothermal method is used to synthesize NSCQDs.The excellent sensing activity of these biomass-derived quantum dots has opened up a route to devise eco-friendly sensors from plant matter, the NSCQDs being in the nanosecond regime suggest that the synthesized NSCQDs are very suitable for probe in optoelectronic and biological applications.

Figure 1 .
Figure 1.The schematics for the synthesis process.

Figure 4 .
Figure 4. UV-vis absorption spectra, fluorescence excitation, emission spectra of NSCQDs (a); Change in the emission wavelength at different excitation wavelengths showing the excitation dependent behavior (b); fluorescence intensity change of NSCQDs with pH value (c); emission intensity change at different times (d), illustrated with normalized treatment.

Figure 5 .
Figure 5.The relationship of fluorescence intensities and the absorbance values.

Figure 6 .
Figure 6.Effect of different reaction conditions, such as (a) reaction time, (b) reaction temperature, (c) potato quantity, (d) ammonium sulfate volume on the preparation of NSCQDs.

Figure 7 .
Figure 7. (a) Selectivity of NSCQDs for different metal ions; (b) Ratio of fluorescence intensity of NSCQDs solution with the addition of other metal cations along with Fe 3+ ions.

Figure 8 .
Figure 8.(a) Fluorescence of NSCQDs at different Fe 3+ ions concentration; (b) linear fitting graph of fluorescence quenching rate of NSCQDs in the presence of Fe 3+ ions.

Table 1 .
Determination results of Fe 3+ in actual samples.

Table 2 .
Comparison of the sensitivities and linear ranges of other systems for the detection of Fe 3+ .