One-Step Hydrothermal Synthesis of N, Fe-Codoped Carbon Dots as Mimic Peroxidase and Application on Hydrogen Peroxide and Glucose Detection

In this paper, N, Fe-codoped carbon dots (N, Fe-CDs) were synthesized from β-cyclodextrin, ethylenediamine, and ferric chloride for the first time using a convenient one-step hydrothermal method. The obtained N, Fe-CDs were characterized by various methods including transmission electron microscopy, X-ray photoelectron spectroscopy, and Fourier-transform infrared spectroscopy. The N, Fe-CDs exhibited better catalytic activity than horseradish peroxidase (HRP) and caused an evident color change for 3,3′,5,5′-tetramethylbenzidine in the presence of H2O2. Kinetic experiments show that the apparent Km value for the N, Fe-CDs with TMB (0.40mM) or H2O2 (0.35mM) as the substrate was lower than that of HRP (0.43 and 3.70mM), suggesting that the N, Fe-CDs have a much higher affinity for TMB and H2O2 than HRP. The Km/Vmax value for the N, Fe-CDs (21:74 × 103 · s for H2O2) is significantly lower than that for HRP (42:53 × 103 · s), suggesting that the N, Fe-CDs have a stronger catalytic efficiency for H2O2 than HRP. Furthermore, a highly efficient and sensitive colorimetric detection method for glucose was developed using the N, Fe-CDs as mimic peroxidase to detect the hydrogen peroxide generated by the oxidation of glucose by glucose oxidase. The limit of detection for H2O2 and glucose was found to be 0.52 and 3.0 μM, respectively. The obtained N, Fe-codoped carbon dots, which possess simulated peroxidase activity, can potentially be used in the field of biotechnology.


Introduction
Peroxidase such as horseradish peroxidase (HRP) is a kind of natural enzyme that catalyzes the decomposition of peroxides and prevents biological cells from being damaged by toxic substances [1]. However, the natural enzyme would suffer from degradation, denaturation, and inactivation in practical applications especially under harsh conditions. The extraction, purification, and storage processes of natural peroxidase are also fussy. High price further limits its applications [2][3][4]. Under this background, various enzyme mimics have been developed in order to address these issues.
In this paper, a new mimic peroxidase, N, Fe-codoped carbon dots (N, Fe-CDs) were synthesized from β-cyclodextrin, ethylenediamine, and ferric chloride using a convenient one-step hydrothermal method. The obtained N, Fe-CDs exhibited better catalytic activity than HRP. Combining with glucose oxidase (GOx), a good colorimetric method for glucose detection was developed.     3 Journal of Nanomaterials kinetic parameters were calculated from the double reciprocal Lineweaver-Burk equation:

Experimental Section
where V is the initial velocity, V max is the maximal reaction velocity, and Km is the Michaelis-Menten constant, and ½S is the concentration of the substrate. The initial velocity was obtained by calculating the slopes of absorbance changes at 652 nm within 100 s. After that, the absorption spectra of the reacted solution were recorded in the range of 500 nm to 800 nm.

RhB Oxidation
Glucose detection was carried out as follows: first, 20 μL of GOx (1 mg/mL), 20 μL of glucose at different concentrations, and 60 μL of phosphate-buffered saline (PBS, 0.1 M, pH 7.0) were mixed and incubated at a 37°C water bath for 60 min. Then, 200 μL of TMB (700 μM), 100 μL of N, Fe-CDs (150 μg/mL), and 600 μL of acetate buffer (0.1 M, pH 2.5) were successively added to the glucose reaction solution and incubated in a water bath at 35°C for 36 min. Finally, the UV-vis spectrum of the mixed solution was recorded using a UV-vis spectrometer. Control experiments, in which 0.5 mM glucose was replaced by 5 mM fructose, ascorbic acid, maltose, xylose, uric acid, α-lactose, and sucrose, were also performed. Each point was measured three times. The standard errors, defined as the average squared deviation of each number from its mean, were calculated using origin software.

Results and Discussion
3.1. Characterizations of the N, Fe-CDs. The TEM image in Figure 1(a) shows that the N, Fe-CDs had good dispersibility, and the size was in the range of 0.9-2.5 nm. FTIR was used to analyze the functional groups in the material. The FTIR spectrum ( Figure 1(b)) shows a broad peak at 3328 cm -1 , which was ascribed to the stretching vibrations of the O-H and N-H bonds. The intense band at 2920 cm -1 arose from symmetric C-H stretching vibrations, and the characteristic band at 1030 cm -1 was assigned to the vibration of the C-N bond.
XPS was performed to analyze the composition of the N, Fe-CDs, and the results are shown in Figure 2(a). Four elements (C, N, O, and Fe) were detected in the XPS spectrum of the N, Fe-CDs. The oxygen content was found to be as high as 31.78%, which further suggests that the N, Fe-CDs contained abundant functional groups. The main peak at 286.25 eV, attributed to C 1 s, was deconvoluted into four contributing peaks at 284.5, 285.0, 286.2, and 287.7 eV (Figure 2(b)), which indicates the presence of four carbon environments: sp 2 C=C or sp 3 C-C, C-N, sp 2 N-C=N, and C=O [28], respectively. The peak at 399.51 eV, attributed to N 1 s, was deconvoluted into three contributing peaks at 399.2, 399.8, and 401.5 eV (Figure 2(c)), which indicates the presence of three types of nitrogen environments: C-N, C=N, and C=N-H, respectively [29]. The peak at 532.61 eV was ascribed to O 1 s, which can be resolved into 532.0, 532.6, and 533.2 eV, corresponding to the C-O, C=O, and O-H bonds, respectively [30]. The weak peak at 716.57 eV was ascribed to Fe 2p. The content of Fe was 0.56%,  Journal of Nanomaterials indicating that Fe has been successfully incorporated into the nanoparticles.  (Figures 3(b) and 3(e)). The influence of the N, Fe-CDs and TMB concentrations, the pH value, the reaction temperature, and duration on the catalytic performance were further evaluated, and the results are shown in Figure 4. The absorbance is shown as a function of the N, Fe-CDs concentration in Figure 4(a), interestingly, low concentrations of N, Fe-CDs showed high activity towards TMB oxidation, but the absorbance changed little when the N, Fe-CDs concentration exceeded 150 μg/mL. Therefore, a N, Fe-CDs concentration of 150 μg/mL was used in the subsequent experiment. The absorbance increased continuously as the TMB concentration and reaction duration increased. The optimal TMB concentration and reaction  Journal of Nanomaterials duration were determined to be 700 μM (Figure 4(b)) and 36 min (Figure 4(e)), respectively. Figure 4(c) shows the change in absorbance caused by varying the pH from 2.0 to 7.0. The results indicate that catalytic activity of the N, Fe-CDs was highest at a pH of 2.5. The catalytic activity of the N, Fe-CDs was also increased when the temperature increased from 25 to 35°C, but it decreased temperatures higher than 35°C (Figure 4(d)).
In order to better understand the mechanism by which N, Fe-CDs catalyze TMB and H 2 O 2 oxidation, the apparent steady-state kinetics were measured. As shown in Figures 5(a) and 5(b), a range of TMB and H 2 O 2 concentrations were used in the catalytic reactions so that the kinetic parameters could be obtained using Lineweaver-Burk plots (Figures 5(C) and 5(D)). Table 1 compares the parameters for this system with those of HRP. A smaller Km value indicates a stronger affinity between the enzyme and the substrate and a higher catalyst efficiency. The Km value for N, Fe-CDs with TMB or H 2 O 2 as substrate was lower than that of HRP, which suggests that the N, Fe-CDs have a higher affinity for TMB and H 2 O 2 than HRP does. On the other hand, although the V max values of the N, Fe-CDs were smaller than those of HRP, the Km/V max value of the N, Fe-CDs with H 2 O 2 as substrate was smaller than those of HRP. The lower the value of Km/V max is, the higher the catalytic efficiency is. Considering their peroxidase mimetic activity, the N, Fe-CDs could be employed as a potential substitute for HRP.
The fitted Lineweaver-Burk lines were nearly parallel at different concentrations of TMB and H 2 O 2 . That means the  Journal of Nanomaterials N, Fe-CDs bind to and react with the first substrate, release the product, then recombine with the second substrate, react and release the second product in a manner similar to the ping-pong mechanism of HRP [31,32].

Mechanism of the Peroxidase-like Activity of N, Fe-CDs.
Generally, the catalytic pathway for peroxidase-like activity can be divided into two types, reactive oxygen species generation and an electron transfer process [33][34][35]. To evaluate the possible active intermediates in the present system, various fluorescent and colorimetric probes were employed. RhB was used to detect •OH as the presence of •OH would cause a decrease in the absorbance intensity of RhB. Figure 6 Figure 7(a). The magnitude of the peak at  A linear relationship between the absorbance and the H 2 O 2 concentration from 0 to 60 μM was obtained (R2 = 0:9932), and the limit of detection (LOD) was calculated to be 0.52 μM based on three times the standard deviation rule (LOD = 3Sd/k). Also, the color of the mixture changed (inset of Figure 7(a)), which indicates that H 2 O 2 can be detected with the naked eye. As glucose would be oxidized to H 2 O 2 and gluconic acid in the presence of glucose oxidase, this method can be used to detect glucose indirectly when combined with the catalytic properties of GOx and N, Fe-CDs. Figure 8(a) shows that the absorbance at the peak of 658 nm increased as the concentration of glucose increased. There exist two parts of linear range between the absorbance and the concentration of glucose in the range of 0 to 60 μM and 60 to 100 μM (Figures 8(b)-8(d)). The LOD was calculated to be 3.0 μM. As shown in Table 2, this method was comparable or more sensitive than other reported systems based on different nanoparticles as peroxidase mimics.
To further investigate the specificity of this method, five other sugars (fructose, maltose, xylose, α-lactose, and sucrose), ascorbic acid, and uric acid were selected as interfering substances for detection. Figure 9 shows the absorbance at 658 nm of solutions containing the other sugars, ascorbic acid, and uric acid at concentrations that were ten times higher than that of glucose (0.5 mM). The absorbance of the other substances differs little from that of the blank (glucose-free) sample, which shows that this method has good glucose selectivity.

Conclusions
As a conclusion, N, Fe-CDs were synthesized from β-cyclodextrin, ethylenediamine, and FeCl 3 •6H 2 O using a one-step hydrothermal method for the first time. The raw materials are easily available, and the preparation process is

Data Availability
Some or all data, models, or code generated or used during the study are available in a repository or online in accordance with funder data retention policies.

Conflicts of Interest
The authors have declared no conflict of interest.