Rational Design of Carbon Dots Featuring High Spectral Overlap with Analyte for Effective Detection of Metronidazole via Inner Filter Effect

The development of fluorescent probe with high spectral overlap with analyte is favorable for the sensing applications via inner filter effect, which requires facile tuning of their structure and photophysical properties. Herein, the rational design of blue emissive carbon dots (B‐CDs) is reported for effective detection of metronidazole (MNZ) in both deionized water and real water samples. The design strategy is based on the elaborate choice of precursors that allows precise control over the nitrogen‐doping degree and graphitization content of B‐CDs. As a result, the as‐prepared B‐CDs exhibit a specific and quick fluorescence quenching response to MNZ via the IFE mechanism because of their good spectral overlap in absorbance. This approach shows a broad linear range (0.5–160 µm), low detection limit (0.1 µm), and moderate selectivity, suggesting its promise for practical applications in environmental sensing.


Introduction
The rapid development of society and economy has greatly improved people's living standards, but also brings the problem of environmental pollution that poses a great threat to the ecological environment and human health.[3] Therefore, the research on antibiotic analysis and detection technology is extremely important.Fluorescence DOI: 10.1002/adsr.202300185[9] Different from these response mechanisms, the inner filter effect (IFE), which refers to the emission intensity decrease caused by competitive light absorption of the analyte, does not require tedious synthesis or modification of fluorescence probe. [10,11]Thus, IFE-based fluorescence assays have recently attractive interests in sensing and detecting applications due to their low cost, simple operation, and good generality.
[14][15][16] Therefore, carbon dots are widely used as fluorescent probes for pollutant detection.Dai's group [17] reported on surface functionalization of carbon dots for the detection of CS 2 in the environment.Radhakrishnan's group [18] reported a dual-mode detection of arsenic (III) and ferrous ions.Recently, several groups have reported on the synthesis of fluorescent carbon dots to detect environmental pollutants through the IFE mechanism. [19,20]Ding's group reported the non-conjugated polymer carbon dots for fluorometric detection of antibiotic.Chen's group reported the synthesis of fluorescent CDs-based probe for p-aminoazobenzene detection.However, in these works, the spectral overlap (e.g., absorption) between CDs and analytes, which determines the efficiency of IFE sensing, remains to be improved.As well documented, the absorption and fluorescence spectra of CDs are dependent on their graphitization content and surface structure.Hence, development of design strategy that can precisely control the structure and photophysical properties of CDs is urgently desired.
Herein, we have reported a blue emissive CDs (B-CDs) that can effectively detecting metronidazole (MNZ), a synthetic nitroimidazole antibiotic, via IFE mechanism.The B-CDs was synthesized by hydrothermal treatment of aspartic acid and 1,2-cyclohexanediamine (DCH).Among the precursors, DCH was employed to improve the nitrogen-doping degree of carbon core and thus promote the n→* transition of N-containing domains.Meanwhile, aspartic acid was used to introduce carboxyl groups to the CDs surface, facilitating their contact with electrondeficient MNZ and thus enhancing the IFE.In addition, considering the main absorbance of MNZ (≈320 nm) in the UV A region, the graphitization degree of CDs was strictly controlled by avoiding using aromatic precursors.Based on these considerations, the as-prepared B-CDs features an intense shoulder absorption peak at 324 nm, which exhibits good spectral overlap with that of MNZ.On this basis, obvious fluorescence quenching was observed for B-CDs upon addition of MNZ, with a good linear relationship in a broad concentration range of 0.5-160 μm.The correlation coefficient R 2 was determined to be 0.995, indicating B-CDs an ideal fluorescent probe for MNZ.Then, comparison of the absorption profile and emission lifetime between B-CDs and B-CDs/MNZ validated the IFE as the emission quenching mechanism.Furthermore, the application potential of B-CDs was demonstrated by the good performance in detecting MNZ in tap water and lake water.

Characteristics of B-CDs
The B-CDs was prepared by hydrothermal treatment of aspartic acid and 1,2-cyclohexanediamine (DCH) precursors at 160°C for 7 h (Figure 1a).After the reaction was completed, the crude solution was first filtered and then purified using a 500-1000 dialysis bag.The synthesis of B-CDs was repeated at least three times, with a production yield of 15.36 ± 0.096 wt.%.The TEM image of B-CDs shows well-dispersed particles with uniform quasispherical morphology (Figure 1b).The average particle size is determined to be 1.66 nm (Figure S1, Supporting Information).Clear lattice fringes were detected from the HRTEM image (insert of Figure 1b), suggesting the formation of graphitic carbon core.The fringes spacing of 0.21 nm could be ascribed to the (100) crystal plane of graphite carbon (Figure S2, Supporting Information). [21]The presence of graphitic structures in B-CDs is also reflected by XRD pattern (Figure 1c), which shows a broad peak at 21°. [22] Notably, this peak is much broadened than that of traditional graphene materials.It indicates that amorphous carbon or nitrogen, which could be derived from the DCH precursor, was incorporated to the carbon core of B-CDs and resulting in the coexistence of sp 2 and sp 3 domains. [23]This assumption is further supported by Raman spectra (Figure S3, Supporting Information), where the peaks at 1396 and 1582 cm −1 are assigned to the D-band (disordered structure comprised of sp 3 domain) and G-band (graphitized structure comprised of sp 2 domain), respectively. [24]The ratio of the D-band to G-band intensity (I D /I G ) is determined to be 0.79, indicating that the degree of graphitization is moderate.
The surface functional groups of B-CDs were then evaluated by FT-IR spectrum (Figure 1d).The broad absorption at 3430 cm −1 and the sharp peak at 1590 cm −1 indicates the presence of O─H and C═O groups, respectively.It suggests that the surface of B-CDs is mainly decorated by carboxyl (─COOH) groups. [25]These carboxyl groups should be derived from the aspartic acid and contribute to the good water dispersity of B-CDs.In addition, a strong signal from C─N species could be found at 1440 cm −1 , indicative of the covalent linkage between carbon core and surface groups. [26]X-ray photoelectron spectroscopy (XPS) was further used to analyze the elemental composition of B-CDs (Figure 2).The XPS full survey spectrum showing three main peaks corresponding to C 1s (285 eV), N 1s (400 eV), and O 1s (531 eV) and their element proportions are 69.86%,10.55%, and 19.59%, respectively (Figure 2a).Hence, besides the presence of C and O, B-CDs have a high degree of N doping.For the C 1s spectrum, four peaks at 284.6, 285.1, 286.4,and 288.3 eV are corresponding to the sp 2 -C, sp 3 -C, C─O/C─N, and C═O species, respectively (Figure 2b). [27]The ratio of C sp3-C /C sp2-C (1.02) indicates the moderate graphitization degree of B-CDs, which agrees with the results from Raman spectra.The N 1s spectrum of B-CDs exhibits three peaks at 399.4, 400.2, and 401.4 eV, which are assigned to pyridinic-N, pyrrolic-N, and graphitic-N, respectively (Figure 2c). [28]The O 1s spectra of B-CDs have two peaks at 531.5 and 532.2 eV attributable to C═O and C─O, respectively (Figure 2d).According to these results, we could conclude that the DCH and aspartic acid are primarily responsible for the formation of carbon core and surface state, respectively.More importantly, the N doping content is substantially improved by using N-containing molecules as precursors, which is of great importance for tuning the spectroscopic behaviors of CDs. [29]

Optical Properties of B-CDs
The optical properties of B-CDs are studied by UV-vis absorption (Abs) and fluorescence (FL) spectroscopy (Figure 3a).As shown in the Figure S4 (Supporting Information), B-CDs possess two absorption band at 200-260 and 290-400 nm.[32] Upon excitation at 334 nm, bright blue emission with peak maximum of 403 nm could be observed from the aqueous solution of B-CDs.The photoluminescent excitation (PLE) spectra of B-CDs show primary contribution from the absorption in the 290-400 nm region, indicating that the sp 2 N-doping domain is responsible for the blue emission of B-CDs.The absolute quantum yield and lifetime is determined to be 11.83% and 4.24 ns, respectively (Figure S5, Supporting Information), demonstrating the fluorescence character of the emission  and its origin from singlet excited state.Moreover, the emission maximum of B-CDs is essentially independent on the excitation wavelength (Figure 3b).This suggests the single emission center of B-CDs, which is beneficial for its sensing applications due to the simplified working mechanism. [33]ubsequently, the fluorescence stability of B-CDs was evaluated.First, the aqueous solution of B-CDs was kept standing in ambient conditions (15-17 °C in air) for 2 weeks and no obvious variation was detected for the fluorescence profiles (Figure S6, Supporting Information), suggesting the high ambient stability of B-CDs.In addition, the fluorescence intensity of B-CDs hardly changes when the concentration of NaCl varies from 0 to 1 mm (Figure S7, Supporting Information), demonstrating that the B-CDs fluorescence shows good resistance to the interference of salt.Then, the influence of B-CDs concentration on its physical property is also explored.In particular, the fluorescence intensity of B-CDs increases linearly with its concentration increasing from 50 to 400 μg mL −1 (Figure S8, Supporting Information).Hence, no aggregation caused quenching (ACQ) was observed in this concentration range.We envisage that the abundant functional groups on the surface of B-CDs may lead to the mitigation of ACQ by creating a "molecular fence" between the emission centers from adjacent CDs.Moreover, the presence of these surface groups could also contribute to the good water dispersity and low toxicity of B-CDs, which is advantageous for the environmental sensing applications.

Fluorescent Sensing of MNZ
To investigate the fluorescence response of B-CDs toward MNZ, fluorescence spectra of the B-CDs with different concentrations of MNZ was recorded (Figure 4a).Specifically, the fluorescence intensity of B-CDs ( ex = 334 nm) shows a 55% decrease upon addition of trace amount of MNZ (80 μm) and the fluorescence changes instantly (<30 s) after the MNZ addition (Figure S9, Supporting Information), suggesting the sensitive and quick response of B-CDs to MNZ.When the concentration of MNZ increases from 0 to 160 μm, the fluorescence of B-CDs gradually quenches and the fluorescence quenching factor (F 0 /F, F 0 and F represent the fluorescence intensity of B-CDs before and after the MNZ addition, respectively) possess a linear relationship with the concentration of MNZ in the range of 0.5-160 μm (Figure 4b).On this basis, the efficiency of fluorescence quenching (K SV ) was obtained by the Stern-Volmer equation: F 0 /F = 1 + K SV C, where C represents the concentration of MNZ.By fitting the plot of F 0 /F versus C, the K SV value was obtained as 0.016 μm −1 with a correlation coefficient (R 2 ) of 0.995.Moreover, according to the 3/slope rule (where  represents the standard deviation of multiple blank samples), the limit of detection (LOD) is calculated to be 0.10 μm.These results demonstrate that B-CDs is a highly sensitive probe for MNZ detection.
To test the selectivity of B-CDs in MNZ detection, a series of ions (100 μm) including Ca 2+ , Cu 2+ , K + , Mg + , Mn 2+ , Na + , NH 4 + , Zn 2+ were added as interference.As shown in Figure S10 (Supporting Information), all these ions exert negligible compact on the fluorescence of B-CDs/MNZ solution.We also tested the fluorescence response of B-CDs to other antibiotics (Figure S11, Supporting Information), such as chloramphenicol, erythromycin and amoxicillin.In contrast to MNZ, these antibiotics do not induce notable change to the fluorescence of B-CDs.However, emission quenching was detected for ornidazole and secnidazole with the quenching factor (F 0 /F) were determined to be 1.88 and 1.82, respectively.These values are comparable with that of MNZ (2.32), suggesting that B-CDs is also capable of detecting ornidazole and secnidazole due to their similarities in structure and absorption.However, it would cause a decrease in sensing selectivity for B-CDs.This is a typical characteristic for the environmental sensing using inner filter effect, which usually features good versatility but moderate (or even low) selectivity. [34]In addition, after comparing the MNZ sensing performance of B-CDs with previously reported probes (Table S1, Supporting Information), they possess comparable sensing parameters such as linear concentration range and detection limit.Nevertheless, the easy preparation and low cost may make B-CDs more promising in practical applications.

Sensing Mechanism
Generally, the common fluorescence quenching mechanisms include electron transfer, fluorescence resonance energy transfer (FRET), and IFE.To reveal the exact quenching mechanism of B-CDs/MNZ, a series of experiments were performed (Figure 5).Upon mixing B-CDs and MNZ together in water, its UVvis absorption spectrum is essentially identical to the absorption spectra of B-CDs + MNZ (Figure 5a).It suggests that MNZ exerts minor compact on the photophysical properties of B-CDs.Thus, electron and charge transfer between B-CDs and MNZ are excluded. [35]Moreover, it is unlikely that the emission quenching arise from FRET since the emission lifetime of B-CDs hardly changes after addition of MNZ (4.24 and 4.05 ns for B-CDs and B-CDs/MNZ, respectively) (Figure 5c). [36]Hence, IFE should be responsible for the emission quenching of B-CDs/MNZ. [37,38]his is supported by the excellent overlap between the absorption spectrum of MNZ and the PLE spectrum of B-CDs (Figure 5b).More specifically, the addition of MNZ would act as a competitive photon absorber to reduce the number of photons for B-CDs excitation, resulting in the decline of emission intensity (Figure 5d).Additionally, the zeta potential of B-CDs was measured to be −4 mV (Figure S12, Supporting Information), indicative of the negatively charged surface.Considering the electrondeficient character of MNZ molecule, this could facilitate the electrostatic interaction between B-CDs and MNZ and thereby shortens their distance, which is beneficial for improving the emission quenching efficiency via IFE.

MNZ Detection in Real Water
For evaluating the practicability of fluorescence sensing method to analyze MNZ in actual samples, we conducted actual sample analysis using local tap water and lake water.Similar to the deionized water experiment, the B-CDs in these two samples show obvious fluorescence quenching upon addition of MNZ and the fluorescence intensity decreases gradually as a function of MNZ concentration (Figure 6a,c).On this basis, the Stern-Volmer equation was fitted as F 0 /F = 0.017C + 0.97 and F 0 /F = 0.016C + 0.96 for tap water and lake water (Figure 6b,d), respectively, both of which gives a linear relationship between the quenching factor (F 0 /F) and the concentration of MNZ (0.5-160 μm) with high correlation coefficient.These results demonstrate B-CDs a reliable analytical tool for MNZ detection in the environment and other fields.

Conclusion
In summary, we have reported a facile strategy to improve the IFE for environmental sensing application.The strategy is based on the precise control over the nitrogen-doping degree and graphitization content of CDs, resulting in good spectral overlap between the absorption band of B-CDs and MNZ analyte.On this basis, B-CDs exhibit a specific and quick fluorescence quenching response to MNZ via IFE mechanism.Moreover, quantitative analysis shows the broad linear range (0.5-160 μm), low LOD (0.1 μm), and moderate selectivity of B-CDs for MNZ detection in deionized water and real water samples (tap water and lake water).This work may provide a new perspective for designing efficient probes for sensing and imaging applications.

Experimental Section
Materials: Aspartic acid, 1,2-diaminocyclohexane, metronidazole, chloramphenicol, ceftriaxone sodium, amoxicillin, florfenicol, Figure 6.FL spectra of the B-CDs upon addition of different concentrations of MNZ in a) tap water system and c) lake water system, the relationship between the fluorescence quenching factor F 0 /F and the concentration of MNZ in b) tap water system and d) lake water system.erythromycin, ornidazole, and secnidazole were purchased from Shanghai Macklin Biochemical Co., Ltd.; sodium chloride and sodium oxalate were purchased from Shanghai Yuanye Bio-Technology Co., Ltd.; ammonium chloride, zinc chloride, calcium chloride, potassium chloride, silver nitrate, ferric chloride, sodium sulfate, sodium acetate, barium chloride were purchased from Sinopharm Group Chemical Reagent Co., Ltd.; magnesium chloride, manganese (II) chloride, copper chloride dihydrate were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd.The water used in the experiment is ultrapure water.
Instrumentation and Characterization: The morphology was examined by using a transmission electron microscope (JEM-F200).The Xray diffraction (XRD) pattern was collected using a SmartLab 9 kW Xray polycrystalline diffractometer.Raman spectrum was recorded using a laser confocal micro-Raman spectroscopy (InVia-Reflex, Renishaw, Britain).Fourier transform infrared (FTIR) spectroscopy was collected using a Nexus 870 FTIR spectrometer.The X-ray photoelectron spectroscopy (XPS) was obtained with an ESCALAB250 XPS using Al K radiation (1486.6 eV).The UV-vis spectra was measured using a UV-2600i spectrophotometer (Shimadzu Corporation, Japan).Fluorescence (FL) spectra were collected on F-4700 fluorescence spectrophotometer (HITACHI, Japan).The HORIBA FLSP920 system was used to obtain the absolute quantum yield (QY) in the calibration sphere.Fluorescence lifetimes were collected on a Horiba Fluoro max plus (Horiba, Japan).
Synthesis of B-CDs: The B-CDs were synthesized by hydrothermal treatment of aspartic acid (AA) and 1,2-diaminocyclohexane (DCH).Briefly, aspartic acid (0.40 g) and 1,2-diaminocyclohexane (0.20 g) were dissolve in 30 mL of ultrapure water, then the solution was transfer to a 50 mL poly (tetrafluoroethylene)-lined autoclaves.After reacting at 160 °C for 7 h, the reaction solution was naturally cooled to room temperature.The obtained yellow solution was filtered (0.22 μm) and dialyzed in regenerated cellulose dialysis tubing with MWCO 500-1000 for 48 h to remove unreacted AA and DCH.After freeze-dried to remove the water, the obtained B-CDs were used for further characterizations and applications.
Detection of MNZ: In a typical experiment, 4 mg of B-CDs were dissolved in 20 mL of ultrapure water to form 200 μg mL −1 solution.After that, a series of MNZ solutions with different concentrations were mixed the above B-CDs solution (1 mL).The mixtures were equilibrated at room temperature for 30 s, then the fluorescence emission spectra were recorded with an excitation wavelength of 334 nm.
Detection of MNZ in Real Samples: For the detection of MNZ in real samples, a local tap water and a local lake water were gathered.The samples were purified by 0.22 μm filter membrane.After that, 4 mg of B-CDs were dissolved in 20 mL of real samples (tap water or lake water) to form 200 μg mL −1 solution.Then a series of MNZ solutions with different concentrations (dissolved in real samples) were mixed with the above B-CDs solution (1 mL).The test conditions were the same as above.
Statistical Analysis: Particle size statistics were performed in Nano Measurer software (Figure S1, Supporting Information).Crystalline lattice fringes with the spacing was calculated by using software GMS 3 version (Figure 1b; Figure S2, Supporting Information).The normalized UV-vis absorption, PLE, and PL spectra were plotted by Origin 2018 (Figures 3a  and 5b).The spectra (the relationship between the fluorescence quenching factor F 0 /F and the concentration of MNZ) were fitted by Origin 2018 (Figures 4b and 6b,d).All experiments were repeated at least 3 times and presented as mean ± SD.

Figure 1 .
Figure 1.a) Illustration of the synthesis and purification process of B-CDs; b) the TEM image (insert: HRTEM images), c) XRD patterns and d) FT-IR spectra of B-CDs.

Figure 2 .
Figure 2. a) Full survey XPS spectrum, b) High-resolution XPS C 1s, c) N 1s and d) O 1s spectra of B-CDs.

Figure 3 .
Figure 3. a) UV-vis absorption, PLE, FL spectra of B-CDs; and b) FL spectra under different excitation wavelengths of B-CDs.

Figure 4 .
Figure 4. a) Fluorescence spectra of the B-CDs upon addition of different concentrations of MNZ.b) The relationship between the fluorescence quenching factor F 0 /F and the concentration of MNZ.

Figure 5 .
Figure 5. a) The UV-vis absorption spectra of B-CDs, MNZ, B-CDs/MNZ, and the absorption of B-CDs + MNZ (C MNZ = 100 μm).b) The UV-vis absorption spectrum of MNZ, and the PLE, FL spectra of B-CDs.c) The FL decay profiles of B-CDs and B-CDs/MNZ.d) The sensing mechanism of B-CDs.