Carbon Quantum Dots-Based Fluorescent Hydrogel Hybrid Platform for Sensitive Detection of Iron Ions

In this study, we prepared novel ﬂ uorescent carbon quantum dots/hydrogel nanocomposite material (CQDsHG) with good adsorption and stable ﬂ uorescence detection of Fe 3+ . The materials were subsequently characterized according to their morphological features, chemical composition, adsorption, and optical properties. The carbon quantum dots (CQDs) were prepared using a microwave-assisted hydrothermal method in no more than 15min, and the as-prepared CQDs exhibited excellent water solubility, as well as emitted strong bright blue ﬂ uorescence with an ultrahigh quantum yield of 93.60%. The CQDs were then loaded into a hydrogel (HG) using the sol-gel method to obtain a functional CQDsHG. The CQDsHG exhibited high adsorption amounts (31.94 mg/g) and a good quenching response for Fe 3+ , thus, it could be used as a sensor to selectively detect Fe 3+ in the linear range of 0 – 150 μ M with a detection limit of 0.24 μ M. We observed minimal di ﬀ erence in the ﬂ uorescence lifetimes between the CQDsHG with and without a quencher (Fe 3+ ), with values of 5.816ns and 5.824 ns, respectively, con ﬁ rming that Fe 3+ was statically quenched on CQDsHG. The results indicated that the innovative combination of CQDs and HG can improve the synergistic performance of each component for the adsorption and quantitative detection of heavy metal ions in the aqueous environment.


Introduction
Iron (Fe 3+ ) is one of the most abundant and essential trace elements for humans and animals [1]. In humans, insufficient or excessive Fe 3+ levels can induce a variety of diseases including hemochromatosis, diabetes, liver damage, heart failure, anemia, and Parkinson's disease [2,3]. Therefore, Fe 3+ detection has attracted attention from researchers, and currently, a variety of well-established Fe 3+ detection methods have been reported, such as atomic absorption [4], inductively coupled plasma mass spectrometry (ICP-MS) [5] and electrochemical analysis [6]. However, most of these techniques require large instruments and complex procedures, which greatly limits their applications. In recent years, fluorescence analysis has attracted significant attention, as it offers advantages such as simple operation, high sensitivity, and a fast response time [7,8].
Recently, there has been an increase in the use of carbon quantum dots (CQDs) as fluorescent probes for chemical sensing, including Hg 2+ , Zn 2+ , Pb 2+ , and Fe 3+ [9][10][11][12], due to their unique optical properties, good biocompatibility, excellent water solubility, low cost, and low toxicity [13][14][15]. In addition, they can be used for a wide range of applications, including photocatalysis, bioimaging, and fluorescent textiles [15][16][17][18][19]. However, the main issue with carbon quantum dots-based sensors is that the fluorescent probes have to be mixed with aqueous solutions for subsequent fluorescence determination of the samples, which is not favorable for rapid in situ detection. Many CQDs applications can only be effectively implemented when they are embedded in solid matrices. Therefore, a quasi-solid platform that preserves the properties of CQDs in solution prevents fluorescence quenching due to self-aggregation of CQDs [20] and is used for CQDs dispersion and immobilization is necessary [21]. Among the various solid matrices, hydrogels are the preferred framework for CQDs, because they offer transparency, semi-wetting characteristics, simple processing, and a highly tunable 3D porous structure [22]. Furthermore, the high specific surface area allows the hydrogels to accommodate a variety of micro and nanoparticles.
This effectively prevents quantum dots clustering and greatly improves the fluorescence stability of the quantum dots. Several studies on the loading of CQDs into hydrogel networks have been reported [23][24][25][26], and CQDs/hydrogel composites have become important soft materials, with shared polymer and CQDs properties for use in interesting applications.
In this study, we fabricated a carbon quantum dotsbased fluorescent hydrogel, which was used as a solid sensor for optical Fe 3+ detection. The CQDs, with an ultrahigh fluorescence quantum yield of 93.60%, were prepared using a microwave-assisted hydrothermal method with citric acid. Then, the synthesized CQDs were introduced into the hydrogel framework via physical crosslinking. We then investigated the properties, structures, adsorption models, and photoluminescence characteristics of the CQDsHG, and the sensing properties of the CQDsHG for Fe 3+ were explored. The results showed that the CQDsHG had potential applications as an effective bifunctional sensor with Fe 3+ detection and adsorption properties.

2.2.
Apparatus. The CQDs were prepared using a microwave-assisted-hydrothermal synthesizer (MD20H, Oprah Technology Group, Inc.). X-ray powder diffractograms were obtained using an X-ray diffractometer (D8-FOCUS, Bruker, Germany). Materials chemical compositions of the materials were measured using an X-ray photoelectron spectrometer (ESCALAB Xi+, Thermo Fisher Scientific, USA). The infrared spectra in KBr were obtained using a Fourier transform infrared spectrometer (FTIR BXII, Perkin-Elmer, USA). Raman spectra were recorded using a Raman spectrometer (LabRAM HR Evolution, HORIBA Scientific, France) with a 514 nm laser beam. The morphologies were observed by high-resolution transmission electron microscopy (JEM-2100F, JEOL Ltd., Japan). Fluorescence experiments were performed on a fluorescence spectrophotometer (FL-7000, Hitachi, Japan), and samples fluorescence lifetime measurements were obtained using steady-state transient fluorescence spectrometry (FL-7100, Hitachi, Japan). The UV/Vis spectrum was determined by a spectrophotometer (T6, Pu-Analysis General Co., Ltd., China and Cary series UV-Vis-NIR, Agilent Technologies, Inc.). Moreover, a flame atomic absorption spectrophotometer (FAAS, A-6300C, Shimadzu, Japan) and an inductively coupled plasma mass spectrometer (ICP-MS, 2030LF, Shimadzu, Japan) were used to detect the metal ions. Photographs were obtained with a quadruple UV analyzer (WFH-203C, Shanghai JingKe Industrial Co., Ltd., China). Lastly, compression tests were performed on HG and CQDsHG using a microcomputer-controlled electronic universal testing machine (104B-EX, Shenzhen Wan Xiao Testing Equipment Co., Ltd., China).

Design and Analysis of the Orthogonal Experiment.
The orthogonal experiments were used to investigate the fluorescent quantum yield of the CQDs (details are presented in supplementary materials). The three-level four-factor orthogonal experiment was designed to investigate the effects of various reaction conditions, such as the citric acid dosage, reaction temperature, reaction time, and microwave power on the carbon quantum dots, and optimize the basic conditions (Tables S1-S3). Secondly, we designed the orthogonal experiments to determine the effects of nitrogen doping type and nitrogen doping amount on the fluorescence quantum yield (Tables S4-S6).

Synthesis of Carbon Quantum Dots (CQDs).
In this study, the CQDs were synthesized using the microwaveassisted hydrothermal method. The citric acid (0.005 mol) was dissolved in water (10 mL) while stirring, followed by the addition of ethylenediamine (0.005 mol). The mixture was ultrasonicated for 3 min and then transferred to a 100 mL microwave digestion tank. The reaction was carried out at 700 W and 160°C for 15 min in a microwavehydrothermal synthesizer, and the resulting CQDs solution was cooled to room temperature and filtered using a 0.22 μm filter membrane to remove any macromolecular impurities. Afterward, the solution was dialyzed for 9 h using a dialysis bag (MW500) and freeze-dried for 36 h to obtain the light-yellow N-doped CQDs.
2.5. Synthesis of the Fluorescent Hydrogel (CQDsHG). In this study, 2.30 g of SDBS and 12.80 g of AM were added to 60 mL (0.6 g/L) of CQDs solution. Then, 2.44 mL of AA and 1.15 g of SMA were added to the above solution, the mixture was stirred in a water bath at 40°C until the SMA was completely dissolved, and then, the initiator KPS was added (0.5 wt% of the total monomer mass). After stirring for another 10 min at room temperature, the solution was injected into a custom-made sealed model, and nitrogen bubbling was employed for 10 min to eliminate oxygen from the solution. Finally, the prepared solutions were placed in a water bath at 60°C for 12 h to obtain the fluorescent hydrogel (CQDsHG), and the resulting hydrogel was repeatedly rinsed with ultrapure water to remove excess reactants and a neutral pH was obtained. For the control, the blank hydrogel (HG) was synthesized following the same procedures as described above, but without the addition of CQDs during synthesis. The CQDsHG and HG were then stored in the dark before use.
2.6. Determination of Fluorescence Quantum Yield. Fluorescence measurements were performed under the following conditions. The slit widths of the emission and excitation gap were 2.5 nm and 5 nm, respectively, the scan rate was 240 nm/min, and the photomultiplier tube voltage was 400 V. The fluorescence quantum yield (QY) of the CQDs was examined based on the comparative method using quinine sulfate as the reference, which was calculated using Equation (1): where φ is the fluorescence quantum yield, x represents the sample, std refers to the standard quinine sulfate ( φ std = 0:54), η is the solvent refractive index (1.33 for aqueous solution), A is the absorbance at the excitation wavelength (366 nm), and I is the integrated fluorescence intensity at the fluorescence emission spectrum. To minimize reabsorption effects, the optical absorbency values were below 0.1 at the excitation wavelength [10].
2.7. Pressure Performance Test. The CQDsHG was fabricated into a cylindrical specimen, and the diameter and original thickness of the specimen were measured with vernier calipers. The specimen was compressed at a uniform speed, with a load of 500 N and a speed of 50 mm/min, until the deformation reached 85%. The data were then saved to produce a graph. 3+ . The CQDsHG or HG samples were sliced into 0.9 cm thick discs with a diameter of 1 cm, and the mass of each disc was measured. The HG discs were placed in Fe 3+ solutions, and Fe 3+concentrations in the solutions were analyzed before and after adsorption using ICP-MS. The adsorption amount was calculated according to Equation (2) [27]:

Hydrogel Adsorption Experiments for Fe
where q t is the adsorption amount at adsorption time t, (mg/ g); V is the solution volume, (mL); C 0 and C t are the concentrations of Fe 3+ in the solution before and after adsorption by the hydrogel samples, (mg/L); and m denotes the weight of the original dry hydrogel used to test, (g). The adsorption kinetics experiments were carried out at different adsorption times (2.5, 5, 10, 20, 30, 60, 90, and 150 min), with a fixed initial concentration (100 mg/L). The quasi-first-order kinetics Equation (3) and quasi-secondorder kinetics Equation (4) were fitted to the adsorption data, respectively [28]: where q e and q t (mg/g) are the adsorption amounts at equilibrium and at time t, respectively, and k 1 (min -1 ) and k 2 (g/ mg·min -1 ) denote the equilibrium rate parameter of the quasi-first-order and quasi-second-order adsorption interactions.
The isothermal adsorption experiments were carried out at different initial concentrations (20,40,60,80, and 100 mg/ L). Then, the adsorption data were fitted using the Langmuir isothermal adsorption model (Equation (5)) and the Freun-dlich isothermal adsorption model (Equation (6)) [29,30]: ln q e = 1 n where C e (mg/L) is the equilibrium Fe 3+ concentration, q max (mg/g) denotes the theoretical maximum adsorption amount, b (L/mg) is the Langmuir adsorption constant, K F (L/g) is the Freundlich constant that indicates the adsorption  Journal of Chemistry amount, and n is the heterogeneity factor representing the adsorption intensity.
2.9. Real Sample Analysis. To evaluate the practicality of the CQDsHG for real sample analysis, Fe 3+ concentrations in tap water and lake water were analyzed using the spiked recovery method with CQDsHG as a sensor. The lake water was sampled from the local Beichuan Lake, and all water samples were filtered through a 0.22 μm membrane before analysis.

Results and Discussion
3.1. Preparation Route of CQDsHG. The preparation route of the fluorescent hydrogel is shown in Figure 1. Citric acid and ethylenediamine were mixed to create N-doped CQDs with a uniform size using the microwave-assisted hydrothermal method in a remarkable 15 min. The quantum yield (QY) of the as-prepared CQDs reached 93.60%, which was higher than the reported results shown in Table S7. In order to find the mechanism for the ultrahigh quantum yield, we studied the synthetic condition of the CQDs by designing Then, the CQDsHG was successfully prepared using irregular free radical copolymerization, where the CQDs were used as the fluorescence source; AA and AM were the hydrophilic monomers; SMA was the hydrophobic monomers; SDBS was the surfactant, and KPS was the initiator. The CQDs could be dispersed and immobilized in the hydrogel through extensive hydrogen bonding to avoid leakage in water. Thus, the resulting CQDsHG exhibited synergistic performance of polymers and CQDs and emitted blue light under UV irradiation (365 nm) as shown in Figure 1. Follow-up studies showed that the CQDsHG could be used for quantitative detection of Fe 3+ based on fluorescence quenching.

Characterization of the CQDs and CQDsHG.
The obtained CQDs were almost spherical in morphology, with a particle size distribution mostly between 2 and 3 nm, and an average diameter of about 4.30 nm (Figure 2(a)). Furthermore, the HRTEM image showed a crystalline surface spacing of 0.35 nm (Figure 2(b)). As shown in Figure 2(c), the XRD pattern for the CQDs had a broad diffraction peak at 2θ = 25:1°, which corresponded to the (002) crystalline spacing of graphitic carbon, indicating that the synthesized CQDs had an amorphous structure. The calculated layer spacing d for the CQDs was 0.35 nm, which was consistent with the HRTEM result. The Raman spectrum of the CQDs ( Figure S1) showed the characteristic G band (related to a crystalline sp 2 carbon network) at 1588 cm -1 and the D band (related to disordered graphite or glassy carbon) at 1348 cm -1 , with an intensity ratio I D /I G of 1.00. The intensity ratio indicated defects of the CQDs with a partially disordered crystal structure, arising from the small sp 2 cluster size [31,32]. In the UV/Vis spectra of the CQDs aqueous solution (Figure 2(d)), two typical absorption peaks at 238 nm and 346 nm were observed, corresponding to the π − π * transition of the aromatic sp 2hybridized domains and n − π * transition of the sp 3hybridized domains, respectively [23]. Upon excitation at 370 nm, the fluorescence spectra of the aqueous CQDs exhibited a strong blue emission at 450 nm, which confirmed that the CQDs had a larger Stokes shift [25].
XPS was used to determine the functional groups and elemental composition on the CQDs surfaces. In the widescan XPS spectrum of the CQDs (Figure 3(a)), three major peaks at 284.6, 399.9, and 532.4 eV were observed, which were attributed to C1s, N1s, and O1s, respectively. The contents of carbon, nitrogen, and oxygen were calculated to be 57.48%, 11.82%, and 30.70%. Furthermore, the highresolution XPS spectra of C1s (Figure 3(b)) had four constitute of carbon bonds, corresponding to C-C at 284.8 eV, C-N at 285.7 eV, C-O at 286.8 eV, and C=O at 288.1 eV, respectively [33]. As shown in Figure 3(c), the N1 spectra could be curve fitted with three peak components at 399.4, 400.3, and 400.8 eV, which were attributed to the electron binding energies of C-N-C, C-N, and N-H [34]. In the O1 spectra (Figure 3(d)), the two peaks at 532.2 and 534.7 eV were attributed to the electron binding energies of C=O and C-OH [33,35]. Thus, the XPS results confirmed that the synthesized CQDs contained a large number of hydrophilic groups, which enhanced the hydrogen-bonding interaction between carbon dots and hydrogels.
The effect of CQDs addition on the mechanical strength of the hydrogel was examined by a pressure test and the results were shown in Figure 4(b). The hydrogel was  7 Journal of Chemistry compressed by applying pressure using an electronic universal compressor. Approximately 180 kPa and 120 kPa of pressure were required to compress the CQDsHG and HG from 1 to 0.5 cm (50% deformation), respectively. When the pressure was removed, both hydrogels recovered completely, indicating that the CQDsHG and HG exhibited good elasticity. With increasing pressure, the HG produced irreparable cracks at 85% deformation, but the CQDsHG did not. This indicated that the addition of CQDs improved the mechanical strength of the CQDsHG to a certain extent [25]. Figure 5(a)    Journal of Chemistry amount of the hydrogel [39]. This was because CQDs addition increased the crosslinking point of the polymer, allowing more functional groups to chelate and absorb Fe 3+ . Therefore, the hydrogel can provide mechanical and chemical stability to CQDs. Conversely, the introduction of CQDs also improved the structure and performance of the hydrogel to some extent. To better understand the adsorption mechanism of the hydrogel, quasi-first-order kinetics Equation (3) and quasi-second-order kinetics Equation (4) were fitted to the adsorption data, respectively. The fitting results are shown in Figures 5(b) and 5(c), and the calculated parameters are given in Table 1. The results showed that Fe 3+ adsorption on the CQDsHG could be modeled using the quasi-second-order kinetic model, and that the adsorption control mechanism of CQDsHG for Fe 3+ was chemical adsorption.

Fe 3+ Ion Adsorption Experiments.
We also investigated the effects of initial Fe 3+ concentration on the adsorption amount of the hydrogel (Figure 5(d)).
In the solutions with different Fe 3+ concentrations, the CQDsHG showed a higher adsorption amount than the HG. The adsorption data were then fitted using the Langmuir isothermal adsorption model (Equation (5)) and the Freundlich isothermal adsorption model (Equation (6)), respectively. According to the determination coefficients ( Figures 5(e) and 5(f)), the Freundlich adsorption isotherm model was more suitable for the Fe 3+ adsorption isotherm of CQDsHG, indicating that the adsorption of Fe 3+ by CQDsHG was limited by multilayer coverage.

Fluorescent
Measurements of the CQDsHG. Figure 6(a) displays the optimal excitation and emission trait of the CQDsHG, under 370 nm excitation, where the strongest emission peak appeared at 450 nm. We observed that the emission of the CQDsHG was dependent on the excitation wavelength, which shifted toward the long-wavelength direction with increasing excitation wavelength   Figure 6(b)). This phenomenon is common in fluorescent carbon materials, which is due to the surface state affecting the energy bandgap of CQDs [40,41].
The photographs of the HG and CQDsHG under day light and ultraviolet light (Figure 6(c)) showed that the HG had no obvious fluorescence emission under 370 nm excita-tion, while the CQDsHG displayed a strong blue emission band centered at 450 nm, indicating that the CQDs exhibited fluorescent behavior even if they were crosslinked into the polymer. The highest fluorescence intensity was obtained when the CQDsHG thickness was 9 mm (Figure 6(d)). The CQDsHG was stored in a sealed dark environment at 4°C,    (Figures 7(a) and 7(b)), suggesting that the CQDsHG could be used as a fluorescence sensor for Fe 3+ detection. We also studied the fluorescence intensity of the CQDsHG in a 0.5 mM Fe 3+ solution (25 mL) at different times (Figure 7(c)).
The fluorescence intensity decreased with prolonged time and stabilized between 15 and 60 min. Therefore, in the following experiments, the detection time of the CQDsHG for Fe 3+ is controlled to 15 min. The relationship between the fluorescence intensity of the CQDsHG and Fe 3+ concentration (0 to 1000μM) was measured by Fe 3+ titration experiments, and the results are shown in Figure 8. The fluorescence intensity steadily decreased with increasing Fe 3+ concentration, and fluorescence quenching did not cause an emission peak shift. We also observed a good linear relationship (0-150μM, R 2 = 0:993) between the Fe 3+ concentration and fluorescence quenching ratio (F/F 0 ) of CQDsHG, where F 0 denotes the initial fluorescence intensity of the solution without Fe 3+ , and F denotes the fluorescence intensity after Fe 3+ addition. The limit of detection (LOD) was calculated as 0.24μM, according to the triple standard deviation rule [42]. A comparison of CQDsHG with various published materials is presented in Table 2, showing that CQDsHG had better detection and removal capabilities.
To further verify the practical applications of the prepared CQDsHG, the Fe 3+ contents in tap water and river water were analyzed using the spiked recovery method with CQDsHG as the fluorescence sensor. The detection results were compared with those obtained by a flame atomic absorption spectrometer (FAAS). As shown in Table 3, the recovery rates obtained from fluorescence spectrometry were between 104.60% and 108.84%, which has no significant difference compared to the FAAS method. This showed that the fluorescence sensor could be used for certain practical applications. a An average response value from three measurements for each sample.
3.6. Fluorescence Quenching Mechanism. The quenching mechanism was investigated by the Stern-Volmer Equation (7) [9]:  Figure S2), when the concentration of Fe 3+ ranged from 0 to 150 μM, the curve of F 0 /F fitted well to the Stern-Volmer relationship. K SV was 388.56 L/mol, and k q was calculated to be 6:6 × 10 10 L/(mol‧s) much larger than 2 × 10 10 L/(mol‧s), implying a static quenching [50]. In order to explore the quenching mechanism, the average fluorescence lifetime of the CQDsHG without (τ 0 ) and with (τ 1 ) the addition of Fe 3+ was also examined ( Figure S3), and the results were τ 0 = 5:816 ns and τ 1 = 5:824 ns. Apparently, the addition of the quencher (Fe 3+ ) did not significantly shorten the fluorescence lifetime of the CQDsHG, which proved that the existence of the static quenching process [15,51].
In addition, The UV/Vis absorption spectrum of the CQDsHG was also investigated to further confirm the fluorescence quenching mechanism. As shown in Figure S4, when Fe 3+ was added to the CQDsHG, the absorption peak of the CQDsHG at 329 nm disappeared. This phenomenon may be due to the combination of Fe 3+ and functional groups (amino, hydroxyl, and carboxyl groups) on the surface of the CQDsHG. The difference further proves the mechanism of static quenching [37,51].

Conclusions
In this study, a fluorescent hydrogel nanocomposite with Fe 3+ adsorption and sensing capability was fabricated based on CQDs. The easily prepared hydrogel nanocomposite improved the synergistic performance of the CQDs and HG. The CQDs were dispersed and immobilized in the hydrogel through extensive hydrogen bonding, which avoided leakage in water. The hydrogel nanocomposite was stable and exhibited a higher adsorption property and mechanical strength than the HG. Furthermore, the fluorescence intensity of the CQDsHG at an excitation wavelength of 370 nm was significantly quenched upon the addition of Fe 3+ , and it could be used to quantitatively detect Fe 3+ in the range of 0-150 μM, with the detection limit of 0.24 μM. Thus, the satisfactory detection results in real samples demonstrated that the CQDsHG could be used as a bifunctional platform to qualitatively detect and remove trace metal ions.

Data Availability
The data used to support the findings of this study are included within the article, and any further information is available from the corresponding author upon request.

Supplementary Materials
The following are available online at http://www.com/xxx/ s1. Table S1: the factors and levels of the orthogonal experiment (basic conditions); Table S2: the orthogonal experiment scheme (basic conditions); Table S3: the numerical of orthogonal tests (basic conditions); Table S4: the factors and levels of the orthogonal experiment (doping conditions); Table S5: the orthogonal experiments scheme (doping conditions); Table S6: the numerical of orthogonal tests (synthetic conditions); Table S7: comparison of nitrogen doping amount and fluorescence quantum yield under various methods for the preparation of carbon quantum dots; Figure S1: Raman spectra of the CQDs; Figure S2: the calibration curves and linear equation for fluorescent quenching ratio and concentration of Fe 3+ ; Figure S3: Fluorescence decay traces of the CQDsHG with and without the addition of Fe 3+ (1 mM); Figure S4: UV-Vis absorption spectra of the CQDsHG (black), Fe 3+ (blue), and CQDs+Fe 3+ system (red), respectively. (Supplementary Materials)