N-doped carbon dots from pericarpium citri reticulatae for wide linear range sensing of hydroquinone

Abstract Hydroquinone (H2Q) present in water is becoming a growing problem in both daily life and industry, with various harmful effects on human health. Therefore, the detection of H2Q is very important. The Carbon dot-based sensor is considered an effective method to detect H2Q. In this work, Pericarpium citri reticulatae carbon dots (P-CDs) were synthesized by a one-pot hydrothermal method for wide linear sensing of H2Q. An orthogonal test was used to expand the active site and fluorescence properties of P-CDs. Under optimal conditions, the P-CDs exhibit an extremely stable blue fluorescence and a quantum yield (QY) of up to 15.4%. In the range of 5–1000 μM, compared with the currently reported carbon dot-based H2Q sensors, the fluorescence intensity (FL) of P-CDs shows good linearity with the variation of H2Q concentration and a higher linear range. Moreover, the sensor demonstrates a low detection limit of 0.058 μM. The possible detection mechanism is based on the synergistic action of static quenching and IFE. This method has been successfully applied to the determination of H2Q in real samples, and the P-CDs-based sensor has the potential to detect H2Q in various complex scenes. Graphical abstract In this work, high quantum yield (15.4%) biomass-based carbon dots by orthogonal experimental were synthesized and a wide linear range of 5–1000 uM for the detection of hydroquinone was developed. P-CDs displayed the highest linear range of hydroquinone sensing to date compared to carbon point-based sensors and were successfully used in real water samples. Moreover, we expanded the possible detection mechanism based on the previous study. HIGHLIGHTS N-doped biomass carbon dots are facilely derived from pericarpium citri reticulatae powder and urea. Doping of exogenous N is decisive for the sensing of hydroquinone by P-CDs. Exogenous N-doped P-CDs have a high QY of 15.4% and enable sensing over a wide linear range of H2Q from 5 to 1000 μM, the detection limit is as low as 0.058 μM. The fluorescence detection of hydroquinone is based on the synergistic action of static quenching and IFE. P-CDs based sensor has been successfully used for real sample detection.


Introduction
Since phenolic compounds are highly toxic and refractory organic pollutants, [1] organizations like the European Union and the United States Environmental Protection Agency have strictly limited their presence in products and water. Hydroquinone (H2Q) is often used to make black and white developers, anthraquinone dye, azo dye, rubber anti-ageing agents, stabilizers and antioxidants. H2Q has a whitening effect, but long-term use of H2Q can also cause exogenous leukocytosis and brown, yellow disease. Thus, the European Union banned the addition of H2Q in cosmetics in 2001. [2] As one of the primary pollutants of industrial wastewater, it is difficult to remove them entirely through conventional water treatment processes. According to research, drinking tap water with trace amounts of H2Q for a long time may affect the metabolic changes of the human body and hematopoietic system. Besides, it may even cause cell cancer [3,4] which has been rated as a class 3 carcinogen by the World Health Organization International Agency for Research on Cancer. More seriously, it has been reported that taking medicines containing high levels of H2Q can even cause death within a short period of time. [5] The reporting H2Q detection methods mainly include liquid chromatography [6] and spectrophotometry. [7,8] Liquid chromatography is expensive and inconvenient to carry, and the accuracy of spectrophotometry still needs to be improved.
In contrast, the fluorescence method is a very simple and rapid detection method with the advantages of high sensitivity, low detection limit, and good selectivity. Currently, the emergence of new fluorescent nanomaterials has endowed the fluorescence method with potential application prospects. The preparation and application of fluorescent nanomaterials, especially functionalized materials such as carbon dots (CDs), silicon dots (SiDs), and metal nanoclusters (MNCs), are receiving increasing attention. Among them, CDs have their unique advantages, such as low biotoxicity, good water dispersibility, and biocompatibility. Therefore, CDs have attracted the extensive attention of many researchers since their discovery in 2004 [9] and have been widely used in catalysis, [10,11] optical devices, [12][13][14] biological imaging, [15,16] medicine, [17] environment monitoring [18][19][20] and other fields. In the past decade, CDs and their composites prepared from different synthetic materials and methods were widely used for the detection of anions, [21,22] organic small molecules [23][24][25] and macromolecules. [26,27] In particular, several instances of H2Q sensors have been disclosed. Ni et al. [28] prepared fluorescent CDs sensitive to H2Q by hydrothermal method with sodium citrate and ammonium bicarbonate as raw materials, realizing H2Q sensing in 0.1-50 mmol/L aqueous solution. Liu et al. [29] synthesized Sidoped CDs using N- [3-(trimethoxysilyl) propyl]-ethylenediamine as a raw material with a linear range of 1-40 mM. Wang et al. [30] synthesized N/S/P co-doped CDs using isocarbophos as a precursor, reaching a linear range of 0.56-375 mM. These techniques could detect the majority of low concentrations of H2Q with low detection limits (less than 1 mM), but the linearity ranges are constrained, making it challenging to detect H2Q in a variety of situations. According to previous studies, hydrophilic groups containing -OH and -COOH enhance the water solubility of CDs and the insertion of nitrogen-containing groups may confer excellent optical and potentially chemical properties to them. In addition, H2Q is known to be susceptible to addition reactions between -NH 2 and -OH of alcohols, which provides the basis for H2Q sensor design. Whereas, the high -NH 2 and -OH content provides more active sites for the selective binding of hydroquinone, which is an important factor for achieving a wide range of H2Q sensing. A large number of nitrogen-containing groups form different types of N bonds during the synthesis of CDs, such as pyridine N, pyrrole N, amino nitrogen and graphitic N control the fluorescent properties and active structure of CDs. Pericarpium citri reticulatae (PP), as a kind of medicinefood homologous substance, enriched with flavonoids, alkaloids, volatile oils, and polysaccharides, is an optimal carbon-containing precursor which provides a huge number of aromatic rings and groups like hydroxyl, carboxyl and carbonyl groups. Urea is a neutral nitrogen-containing compound with a nitrogen content of 47% and often used as a nitrogen dopant in organic synthesis. These two materials meet the basic requirements for the design of wide-range H2Q sensing systems.
Herein, N-doped fluorescent CDs were prepared by hydrothermal method using PP and urea as precursors. The orthogonal experiments were designed to optimize the raw material content, urea and PP content, time and temperature of the reaction to broaden the linear detection range. Pericaipium citri reticulata carbon dots (P-CDs) prepared by reacting 1.5 g of PP powder and 0.6 g of urea at 220 C for 10 hours were found to have the highest fluorescence with a fluorescence quantum yield (QY) of 15.4%. The P-CDs had suitable sensing for H2Q in the range of 5-1000 lM and successfully realized the detection of H2Q in mineral water and tap water. The effect of urea doping on CDs has been explained.

Chemicals and materials
The PP used in this experiment was obtained from Bozhou, China. Na 2 SO 4 , NaCl, and Na 2 CO 3 were purchased from Shanghai Reagent Factory (Shanghai, China). KI and K 2 CO 3 were purchased from Shanghai Myrell Chemical Technology Co., LTD. (Shanghai, China). Quinine sulfate, H2Q, and catechol (CC) were obtained from Aladdin Industrial Corporation (Shanghai, China). Resorcinol (RC) and phenol were supplied by Tianjin Bodi Chemical Co., LTD. (Tianjin, China). Benzoic acid was purchased from Shanghai Lingfeng Chemical Reagent Co., LTD. (Shanghai, China). All reagents were analytically pure and without other treatment before use. Distilled water was used twice in all experiments.

Synthesis of P-CDs
The water-soluble fluorescent CDs were prepared by a onestep hydrothermal reaction method using PP and urea as precursors. The synthesis process is shown in Scheme 1 and can be described as follows: PP (1.5 g) and urea (0.6 g) were rapidly stirred and dissolved in 50 mL of deionized water. The solution was then carefully transferred to a polytetrafluoroethylene-lined autoclave (100 mL) and heated at 220 C for 10 h. After the reaction was completed, the reaction kettle was naturally cooled to room temperature. The resulting solution was filtered through a 0.22 lm microporous membrane and then purified through an MW:100D dialysis bag for 24 h. The outer solution was collected and diluted 12 times and stored in a refrigerator at 4 C for subsequent experiments. P-CDs solution was obtained (1.6 mgÁmL À1 ).
The P-CDs screening was carried out by orthogonal experiment. The detailed information was shown in Electronic Supporting Material (ESM), and the synthesis method was the same as above.

Characterization
The transmission electron micrographs were obtained by the talos F200x G2 (FEI, US) at 200 kV using ultrathin carbon film coated copper grid as substrates. The fluorescence spectra were recorded with an RF6000 fluorescence spectrophotometer (SHIMADZU, Japan). The fluorescence spectra were scanned at an excitation wavelength of 332 nm, and the emission spectral range was 365-500 nm. The emission slit and the excitation slit width were set to 3 and 5 nm. The fluorescence lifetime was measured with an FLS1000 spectrophotometer (Edinburgh, UK). The UV-Vis absorption spectra were recorded with an HB-7 spectrophotometer (Beijing, China). Fourier transform infrared (FT-IR) spectra were acquired on a BXII Spectrometer (PerkinElmer, US). X-ray Photoelectron Spectroscopy (XPS) was investigated using a Thermo Scientific K-Alpha spectrometer with a mono X-Ray source Al Ka excitation. The Zeta potential is carried out by Nano-ZS90 (Malvern, UK).

QY measurements
Quinine sulfate was used as a standard (QY ¼ 0.54 at 350 nm), and the fluorescence QY of the CDs was calculated using formula (1): where u is QY, I is the measured emission intensity, A is the absorbance at the excitation wavelength, and g is the refractive index solvent index. The subscripts "x" and "s" denote P-CDs and standard samples, respectively. For these aqueous solutions, g x /g s ¼ 1. To reduce the reabsorption effect, the absorbance at the excitation wavelength was kept below 0.05 at 350 nm.

Standard procedure for detecting H2Q
All assays for H2Q were performed at room temperature. The prepared H2Q stock solution of 3 mM was serially diluted to obtain different concentrations of H2Q. The P-CDs were first diluted 12-fold (1.6 mgÁmL À1 ), then 6 mL of P-CDs were mixed with 3 mL of different concentrations of H2Q and incubated at 30 C for 3 h. Finally, fluorescence measurements were carried out at an excitation wavelength of 332 nm and an excitation and emission slit width of 3 and 5 nm. The fitted functional equation was obtained by the variation of the fluorescence intensity (FL) of P-CDs before and after the addition of different concentrations of H2Q. The selectivity of P-CDs was verified by substituting H2Q with common metal ions, anions, and some phenolic organics (including H2Q, CC, RC, benzoic acid, phenol, and acetaminophen), where the concentrations of H2Q, CC and RC were 1 mM and all other substances were 250 lM. In the anti-interference experiment, metal ions, anions, and some phenolic organic compounds were added as interfering substances in the presence of H2Q. The concentration of H2Q was 1 mM, and the concentration of other substances was 250 lM.

Detection of real samples
Real samples (tap water, mineral water) were placed in a 50 mL beaker and ultrasound for 10 min before being heated Scheme 1. Schematic illustration of the one-step hydrothermal synthesis of P-CDs and detection of H2Q.
and boiled for 5 min. After allowing the solutions to cool to ambient temperature, the solutions were filtered over a 0.22 lm MCE membrane. Then, H2Q was added to the preanalysis samples at two levels of 10 lM and 50 lM, with 3 replicates of each added concentration level. The mixture was configured and tested according to the same procedure and conditions as in 2.5. At the end of the fluorescence test, calculations were performed based on the fitted equations obtained.

Optimization of synthesis conditions
The FL of P-CDs was used as a criterion to find the optimal synthesis conditions. The optimum preparation conditions were determined by an orthogonal design experiment (see supplementary materials for details), which gives the result as follows: 1.5 g PP powder (carbon source) and 0.6 g urea (N-dopant) reacted at 220 C for 10 h.
In combination with previous studies, the fluorescence performance and linear range of sensing may be related to N content, reaction temperature, N and C bond type, and so on, which will be discussed in detail in the following sections.

Characterization of the P-CDs
The characterization of the materials was carried out under optimal conditions. The morphology and nanostructure of P-CDs were identified from a typical TEM image. Figure 1a shows that the particle size is around 2-3 nm with distinct lattice stripes, indicating the successful preparation of P-CDs. Figure 1b shows spherical black areas with distinct streaks, this is the close proximity of the P-CDs to the thickness of the carbon film and the spatial stacking that occurs when photographed. The majority of P-CDs were subsequently captured for analysis with a lattice spacing of 0.31 nm, matching the graphite structure (002) plane, similar to previously reported biomass-derived CDs. [31,32] In addition, distinct from other biomass CDs, P-CDs have a smaller particle size distribution, which is probably due to the higher reaction temperature and relatively longer hydrothermal time allowing for more adequate dehydration and polymerization of PP. The particle size of CDs is closely related to the effective conjugation length or sp 2 domain size inside the carbon nucleus. The strong coupling between the p-electron system in the carbon nucleus and surface groups such as carboxyl and carbonyl groups thus modifies the energy gap of P-CDs, resulting in particle size-dependent fluorescence emission caused by the surface state, [33] which may be one of the main reasons for the excellent fluorescence performance of P-CDs.
The FT-IR spectra of undoped CDs and P-CDs are shown in Figure 2. The broad peaks of P-CDs near 3200 cm À1 are the stretching vibrations of C-OH and N-H, and the bending vibrations of N-H at 1567 cm À1 . The extra peak at 1094 cm À1 corresponds to the asymmetric stretching vibration of C-NH-C compared to the undoped CDs. The above shows the successful synthesis of hydrophilic groups containing hydroxyl and amino groups, which explains the good dispersibility of P-CDs. Heteroatom doping is also a common and effective method of modulating the intrinsic properties of CDs. [34] The bonding between nitrogen and carbon alters the carbon hexagonal ring, leading to  enhanced fluorescence properties of the CDs and the creation of emission energy traps for particles through radiative recombination excited by electron-hole pairs, thereby broadening the range of possible application scenarios. The introduction of nitrogen impurity allows excess electrons to enter the CDs, resulting in an upward shift of the Fermi level and a change in optical properties. [35] The amount of nitrogen doping was positively correlated with the fluorescence QY of CDs in a certain range, but the nitrogen content affected the viscosity of CDs and thus restricted the drying time, which was not conducive to practical applications. It is consistent with the results of the orthogonal experiment.
To further analyze the surface composition of P-CDs, XPS experiments were carried out. The XPS spectra of the undoped CDs are shown in Figure S1, with the percentages of C, N and O atoms being 68.23%, 2.67%, and 29.1%, respectively. As shown in Figure 3a, the percentages of C, N, and O atoms in the figure are 69.56%, 8.6%, and 21.84%, with the corresponding C1s, N1s, and O1s peaks occurring at 284.37 eV, 399.1 eV, and 531.64 eV. The significant increase in N content indicates that urea was successfully used for the doping of P-CDs. In Figure 3b, the high-resolution XPS spectra of C1s demonstrate the presence of C-C/ C ¼ C (284.61 eV), C-N (285.51 eV) and C ¼ O (288.01 eV). [36] In Figure 3c, the N1s spectrum of P-CDs consists of three peaks, which denotes the presence of N in three forms, corresponding to pyridine N (398.41 eV), 1/2 amino N (399.71 eV) and pyrrole N (400.91 eV). [37] In Figure 3d, the O1s spectrum of P-CDs consists of three peaks, with the presence of groups such as C ¼ O(531.08 eV), C-O(532.08 eV), and C-OH/C-O-C(532.68 eV). [38] The above results are consistent with the IR spectrum. Another important factor affecting the performance of P-CDs is the introduction of nitrogen atoms into the carbon lattice promotes physicochemical properties and provides an abundance of highly reactive sites as different types of nitrogen functional groups such as pyridine N, pyrrole N and graphitic N. [39] The strong electron-giving ability of nitrogen gives the carbon material a strong coordination capability, improving its electron transport properties [40] and providing more binding sites for subsequent small molecule sensing.
As displayed in Figure 4a, the P-CDs solution exhibits intense pale yellow under white light while appearing transparent and blue fluorescence when excited by a 365 nm UV lamp. At 272 nm and 300 nm, two obvious absorption peaks were observed, which were owing to the p-p Ã leap in the C ¼ C bond [41] and the n-p Ã leap in the C ¼ O bond, [42] respectively. In Figure 4b and c, the P-CDs fluoresced in the excitation wavelength range of 280-350 nm with an emission wavelength of 390-430 nm. When the excitation range was 300-350 nm, the emission wavelength increased with the excitation wavelength, showing the same excitation dependence as most CDs. P-CDs had maximum FL at 407 nm under 332 nm excitation, consistent with blue fluorescence under UV light, with a QY of 15.4%. In addition, the maximum excitation and emission wavelengths of undoped CDs were 345 nm and 433 nm, but the QY was only 2.48%. Compared to previously reported biomass CDs, the P-CDs had a higher QY. The results are listed in Table 1. Figure 4d showed the stability of P-CDs in aqueous solutions of different acidity and alkalinity. P-CDs have a high stability in the pH range of 4-11 and still less fluctuate under extreme conditions. In Figure 4e, the fluorescence intensity of P-CDs remains stable at high concentrations of NaCl. Therefore P-CDs have the potential to detect contaminants under complex conditions.

Establishment of sensing model for H2Q
The synthesis of P-CDs with good water solubility prompted the exploration of their potential application as contaminant probes. Intriguingly, P-CDs were found to be sensitive to H2Q, and the addition of H2Q resulted in a significant quenching of the strong fluorescence of P-CDs. As two isomers of H2Q, CC and RC had a slight or almost no effect on the quenching of P-CDs. Moreover, the response of undoped CDs to H2Q was tested ( Figure S2). Differently, the undoped CDs showed little change before and after the addition of H2Q. The above phenomena suggest that the introduction of urea provides a binding site for the sensing of H2Q. The reaction mechanism of P-CDs will be discussed in detail later. Therefore, additional P-CDs-based fluorescent probes were constructed for the rapid identification and detection of H2Q.
First, the ability of P-CDs to quantitatively detect H2Q was evaluated. Since the maximum emission wavelength of P-CDs is known to be 407 nm and the position of the maximum emission peak is almost unchanged by the addition of H2Q. Thus, 332 nm and 407 nm were selected as the excitation and emission wavelengths for the subsequent tests. The preparation of the test solution was carried out according to the procedure in 2.4, and H2Q was divided into 16 concentration gradients between 5 and 1000 mM. After the preparation of the test solution, fluorescence tests were performed, and the obtained spectra are shown in Figure 5a. The FL of P-CDs at 407 nm decreased with increasing H2Q concentration. When the H2Q concentration reached 1 mM, the quenching rate was up to 60%. The F 0 /F values showed good linearity with H2Q concentrations in the range of 5 to 1000 lM (Figure 5b). The linearity was fitted using the Stern-Volmer equation and can be expressed as F 0 /F ¼ 0.00197c þ 1.00313 (R 2 ¼ 0.995). The detection limit was calculated as 3r/k (r is the standard deviation of the blank signal and k is the slope of the standard curve plot) and gave a result of 0.058 lM (0.0064 mg/L), which is below the US Environmental Protection Agency permitted emission standard (0.5 mg/L).
In addition, the method has a low detection limit and an exceptionally wide linear range compared to the other analytical methods in Table 2, indicating that the sensor has the excellent analytical capability.   [43] Dunaliella salina Hydrothermal 200 C, 3 h 415 nm Blue 8% [44] Lychee waste Solvothermal 180 C, 5 h 443 nm Blue 23.5% [45] Wool Microwave 200 C, 1 h 450 nm Blue 16.3% [46] C. pyrenoidosa powder Microwave 210 C, 2 h 425 nm Blue 2.31% [47] Tang et al. [50] mentioned that the surface of nitrogendoped carbon dots contains a large number of functional groups such as hydroxyl, carboxyl and amino groups, and the proportion of O, H and N increases accordingly, which can increase their water solubility and reactive sites. Hence, the ultra-wide linear range of P-CDs sensing may be attributed to the introduction of N atoms. The first use of N-GQDs as a fluorescence sensing platform was shown to broaden the linear concentration range used for detection. [51] N-GQDs were prepared by hydrothermal treatment of already synthesized GQDs by adding hydrazine (30%). This phenomenon can be explained by the fact that pyridine-N provides the active site for reduction reaction (ORR) by creating Lewis bases and lowering the Fermi energy level of P-CDs, and that pyridine-N bonding increases and improves the oxygen ORR activity of P-CD. [52]

Selectivity and anti-interference capability of P-CDs fluorescence sensor for H2Q detection
Selectivity and anti-interference capability are important parameters to evaluate the performance of novel fluorescent probes before being applied to actual sample detection. Hence, the selectivity and anti-interference of this fluorescent sensor for P-CDs were investigated in this experiment. The effect of P-CDs on the analytical results of anions, cations and phenolic compounds including phenol, benzoic acid, H2Q, CC, RC, acetaminophen, K þ , Na þ , Cl À , I À , CO 3 2À , and SO 4 2À was probed. Figure 6a illustrates the impact of some phenolic compounds and anions on the analytical results, only H2Q exhibited a discernible quenching effect on the FL of P-CDs. The P-CDs@H2Q system and the impact of additional pollutants on the mixed system are shown in Figure 6b, the addition of CO 3 2À causes an increase in solution pH, while H2Q is more readily oxidized to p-benzoquinone under alkaline conditions, leading to further quenching of the P-CDs@H2Q system. In contrast, the fluorescence quench caused by RC is attributed to the formation of a red complex of RC with H2Q or p-benzoquinone, which had been illustrated to be present in water, ethanol and methanol solvent. [53] This red complex may have produced direct electron transfer with P-CDs or enhanced electron transfer between P-CDs and p-benzoquinone. The conversion of p-benzoquinone and RC to the red complex is as formula (2) p À benzoquinoneðAÞ þ RCðBÞ ! transfercomplexðABÞ (2) These results indicate that the H2Q sensing system has good selectivity and anti-interference performance which allows the detection of the presence of H2Q in water in a variety of complex situations.

Possible mechanism of H2Q detection based on P-CDs fluorescence probe
Fluorescence quenching are mostly static or dynamic quenching. In brief, static quenching is the interaction of CDs with the quencher to form a non-fluorescent ground state complex, which is characterized by an almost constant fluorescence lifetime of CDs and the appearance or change of characteristic peaks in the absorption spectrum of CDs. Dynamic quenching is the collision of excited fluorescent molecules with the quencher, which is distinguished by a change in the fluorescence lifetime of CDs with the quencher but no significant change in the absorption spectrum. [54] On this basis, the possible reaction mechanism of   [28] Si-CQDs N-[3-(trimethoxysilyl) propyl]ethylenediamine Fluorescence 1.0-40 0.077 [29] N/S/P codoped CD-Fe 3þ isocarbophos Fluorescence 0.56-375 0.160 [30] N-rGO/CuO GO,Cu(OAc) 2 electrochemistry 1-600 0.25 [48] CeO 2 NPs Ce(NO 3 ) 3 .6H 2 O electrochemistry 15-225 0.9 [49] P-CDs PP, urea Fluorescence 5-1000 0.058 This work the fluorescent probe was discussed. Research has indicated that H2Q and CC are easily oxidized to p-benzoquinone and o-benzoquinone, whereas RC is more stable. As the relative positions of the two hydroxyl groups on the benzene rings of H2Q, CC and RC are distinct, the distribution of charge density also differs, resulting in different rates of electron transfer. The charge density is greatest when the two hydroxyl groups are in opposite positions, followed by the neighboring position and the smallest interposition. The higher the charge density, the easier it is to be oxidized, which may be the main reason of P-CDs have the highest selectivity for H2Q. As an example, the possible reaction mechanism of H2Q was investigated, which has the highest degree of quenching. The color of the P-CDs solution clearly changed from pale yellow to reddish brown after the addition of H2Q or p-benzoquinone to the P-CDs (inset of Figure 7a). In contrast, the color of the solution did not change significantly before and after the addition of H2Q for undoped CDs ( Figure S3). Then, the UV ( Figure 7a) and FT-IR spectra (Figure 7b) of P-CDs@H2Q and P-CDs@pbenzoquinone were compared, revealing a high degree of agreement. Distinct from P-CDs, the additional absorption peaks at 1770 cm À1 and 1510 cm À1 in the FT-IR absorption spectrum corresponds to the stretching vibration of the ketone carbonyl group and the stretching vibration of the cyclic olefin C ¼ C, respectively; the disappearance of the absorption peak at 272 nm and the enhancement of the absorption peak around 210 nm in the UV absorption spectrum suggest the formation of the CDs-quinone complexes. [55] These observations imply that H2Q is first oxidized to p-benzoquinone and where a variety of reactions may follow ( Hydrogen bonding between p-benzoquinone and P-CDs. Based on the reaction types (1) and (2), the introduction of high N content increases the binding sites of type (1) P-CDs and H2Q, which is the main reason for the wide linear sensing of this system. To further explore the mechanism of fluorescence burst, fluorescence decay curves, cyclic voltammetry and zeta potential tests were performed. In Figure 9a, the excitation spectrum of the prepared P-CDs partially overlaps with the UV absorption spectrum of H2Q, which may affect the excitation process of the P-CDs, and the burst mechanism at this point may be attributed to IFE or FRET. As shown in     To verify whether the fluorescence burst is based on the PET mechanism, the HOMO and LUMO energy levels of P-CDs were determined by cyclic voltammetry. The scheme was as follows: an appropriate amount of P-CDs solution was applied dropwise to the surface of a carbon paper electrode, shade dried and placed in an electrolytic cell with an Ag/AgCl electrode and a Pt electrode in 0.1 M PBS buffer solution, followed by cyclic voltammetry. The reduction potential of P-CDs was measured to be À0.27 eV (Figure 9c) and E g to be 4.11 eV. E HOMO and E LUMO can be calculated according to Equations (3)-(5).
The E HOMO and ELUMO of P-CDs were calculated to be À8.24 eV and À4.13 eV, respectively, while the E HOMO and E LUMO of H2Q were À8.13 eV and À4.09 eV. Judging from Figure 10, the electrons in the excited state of P-CDs could not be transferred from the LUMO to the LUMO of H2Q, so the fluorescence burst of P-CDs was not generated by PET.
Subsequently, zeta potential tests were performed. The zeta potential of P-CDs was À5.32 mV, while the zeta potential of the mixture of H2Q and P-CDs was À17.33 mV, indicating that H2Q reacts with the positively charged groups of the P-CDs. The nitrogen doping causes some of the sp 2 carbon to be replaced by nitrogen in the form of pyridine nitrogen and pyrrole nitrogen, which dissociate in water making them positively charged and providing a binding site for H2Q. The carbon nucleus of P-CDs consists mainly of graphitic carbon (sp 2 structure) and disordered sp 3 carbon, which is partially replaced by nitrogen due to nitrogen doping. These nitrogen species exist in the form of pyridine and pyrrole nitrogen, which are positively charged by dissociation in water, providing a binding site for H2Q. Furthermore, the quenching effect can be well expressed by the Stern-Volmer equation [21] (6): K sv ¼k q s 0 is the Stern-Volmer quenching constant; k q is the quenching rate; s 0 is the fluorescence lifetime of the phosphor in the absence of the quencher, and [Q] is the concentration of the quencher. According to the slope of the regression line, K sv is about 1.97 Â 10 3 M À1 . Considering the fluorescence lifetime s 0 ¼ 5.978 ns of P-CDs, K q is 3.295 Â 10 11 M À1 Ás À1 , higher than the maximum dynamic quenching effect (1.0 Â 10 10 M À1 Ás À1 ). These results further suggest that the fluorescence burst is due to the synergistic effect of IFE and static bursts.

Real sample detection
The results show that the fluorescence sensor for H2Q detection based on P-CDs has the advantages of high selectivity and simplicity of operation. To further evaluate the feasibility of this method in real water samples, H2Q was dissolved in mineral water and tap water to make a standard solution mixed with the P-CDs solution. The experimental results are shown in Table 3. The results indicate that the method still has good accuracy in real water samples and can be used for the determination of H2Q content in real samples.

Conclusion
In summary, a biomass material is designed for wide-range H2Q sensing which uses pericarpium citri reticulatae powder and urea as precursors. The produced P-CDs are spherical in shape with an extremely narrow size distribution of about 5 nm and a QY of 15.4%. P-CDs show bright blue fluorescence under exposure to UV radiation. Under optimal conditions, P-CDs responds to H2Q through a synergistic action of IFE and static quenching. There was a linear relationship between the fluorescence quenching and varying H2Q concentrations in the range of 5-1000 lM, with a linear correlation coefficient (R 2 ) of 0.995 and a detection limit of 0.058 lM. This ultra-wide linear range may be attributed to the high nitrogen doping which increases the active sites on the surface of the P-CDs. As well, a practical implementation of the H2Q assay in mineral water and tap water showed 97.4-104.72% recoveries. This study offered a viable potential for the deployment of biomass-derived N-doped CDs as assays for H2Q in complex scenarios.

Disclosure statement
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.