Development of ciprofloxacin sensor using iron-doped graphitic carbon nitride as transducer matrix: Analysis of ciprofloxacin in blood samples

In the present work, we have synthesized an iron-decorated graphitic carbon nitride (Fe@g-C 3 N 4 ) composite and employed it for electrochemical sensing of ciprofloxacin (CFX). The physicochemical characteristics of the Fe@g-C 3 N 4 composite were analyzed with X-ray diffraction (XRD), scanning electron microscopy (SEM), and energy-dispersive X-ray diffraction (EDX) spectroscopy methods. Further, the pencil graphite electrode (PGE) was modified with Fe@g-C 3 N 4 composite to get PGE/Fe@g-C 3 N 4 electrode and characterized the resultant electrode by cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS). Differential pulse voltammetry (DPV) was employed to determine the effect of concentration and interferents. The modified PGE/Fe@g-C 3 N 4 electrode demonstrated the exceptional electrochemical performance for CFX identification and quantification with a LOD of 5.4 nM, a wide linear range of 0.001-1.0 µM, and high sensitivity of 0.0018 µA mM -1 cm -2 . Besides, Fe@g-C 3 N 4 modified PGE showed remarkable recovery results in qualitative analysis of CFX in human blood specimens. This research advocates that the Fe@g-C 3 N 4 composite acts as an excellent transducer material in the electrochemical sensing of CFX in blood and standard samples. Further, the proposed strategy deduces that the PGE/Fe@g-C 3 N 4 sensor can be a prospective candidate for the dynamic determination of CFX in blood serum and possibly ratified as an exceptional drug sensor for therapeutic purposes.


Introduction
The quinolones are a class of broad spectrum of drugs due to their excellent activity against gramnegative pathogens. They are the choice for patients with intra-abdominal infections together with anti-anaerobic agents [1][2][3]. Ciprofloxacin (CFX) continues to be the most efficient quinolone in veterinary and human medication [4]. An immoderate dosage of CFX residues can purport significant antagonistic effects besides causing the ailments such as skin and respiratory infections, chronic bacterial prostatitis, and nosocomial pneumonia [5]. The European Union has set the maximum residue level of ciprofloxacin in milk [6] to be 100 ng mL -1 . The presence of CFX and other antimicrobials in the environment is a ground for attention, owing to the possible genesis of resistance to antibiotics. Therefore, developing alternate sensitive and precise sensors to examine the antibiotics in the biological samples (blood) is a dynamic field of the probe [7].
Iron (Fe) nanoparticles are acknowledged for their excellent electrocatalytic activity, low cost, non-toxicity, high stability, and fair conductivity, indicating that combining Fe with g-C3N4 might improve the sensitivity of the intended sensor [31]. The combination of iron and g-C3N4 can potentially magnify the electrochemical sensing characteristics of the electrode surface. The decoration of Fe on the g-C3N4 surface can demonstrate a stable sensing platform with good electron transfer characteristics, which ideally suit the construction of novel electrochemical sensing systems. An electrochemical sensor was described with this motivation by modifying a pencil graphite electrode (PGE) with Fe@g-C3N4 nanomaterials.
This work demonstrates a new promising electroactive drug sensor for the qualitative and quantitative detection of CFX using g-C3N4 and Fe@g-C3N4. The proposed sensor manifested an extensive linear detection range, high sensitivity, low detection limit, and selectivity concerning the detection of CFX. Moreover, the sensor considers high accuracy, extensive shelf-life, and reproducibility, indicating that the Fe@g-C3N4 is a proper matrix for the sensor fabrication. The realtime application of the advanced sensor is further validated with the analysis of CFX in human blood specimens. Finally, a comprehensive comparison with earlier sensors highlighted the performance of PGE/Fe@g-C3N4. Therefore, it is anticipated that the drug sensor can be a useful tool for biomedical and diagnostic applications.

Synthesis of Fe@g-C3N4
The synthesis of g-C3N4 was carried out by urea pyrolysis (20 g) using a lid crucible (Isotemp Programmable Muffle Furnace 650-750 Series, Fisher Scientific) in a muffle furnace at 550 °C for 3 h [31]. Finally, the doping of Fe on g-C3N4 nanosheets was made by mixing calculated quantities of ferric(II) chloride and 0.6 g of g-C3N4 nanosheets in 50 mL of acetone with constant stirring. The resulting suspension was agitated for two hours at ambient temperature, centrifuged for 15 min at around 6000 rpm, and rinsed many times with acetone to eliminate aggregates. The color of the withered specimens slightly shifted from pale yellow to imperceptibly reddish-brown upon the doping of iron [29].

Fabrication of CFX sensor with Fe@g-C3N4
A cylindrically shaped pencil graphite rod with a diameter of 3 mm (surface area: 0.07068 cm 2 ) was utilized as a working electrode. Further, the electrode was modified by taking a tiny portion of the rod and polished along one face. A copper wire was fastened to secure the electrical contact flanked by the electrode and the potentiostat. To obtain a shiny surface, PGE was polished using emery paper (80 and 300 Grit) and consequently, with the electrode polishing solution containing alumina and silica using electrode polishing tool kit (PK-3 brand kit). Then, the polished PGE was sonicated and eventually washed with milli-Q water and dried at room temperature to eliminate loosely bounded shreds. The prepared bare PGE was further deposited with 3 µl of 5 mg mL -1 stock solution of Fe@ g-C3N4 by drop-casting method and eventually, the electrode was dried at ambient temperature to get the working electrode PGE/Fe@g-C3N4.

Electrochemical studies
The electrochemical investigations were performed in a three-electrode cell, where PGE was used as a working electrode (surface area: 0.07068 cm 2 ), saturated calomel electrode as the reference electrode, and platinum wire as the counter electrode.

Preparation of sample for real analysis
The serum samples collected from normal individuals (taking their inscribed consent) were refrigerated till examination. 5 mL of serum was treated with an equivalent volume of methanol as a serum desaturating and precipitating agent. The conduits were vortexed for 10 min and then centrifuged for 40 min at 5000 rpm to eliminate the protein residues. The supernatants were diluted up to 10 mL with the 0.1 M PBS buffer solution of pH 7.0. The standard addition method was employed for calculating the recoveries of the spiked CFX in human serum. The percentage recovery and detection precision were computed based on the known amount of spiked CFX (R) and empirical values (E) using equations 1 and 2. Figure 1A shows XRD patterns of the nanosheets of the integrated g-C3N4 and Fe@g-C3N4. A strong diffraction peak at 27.3 o demonstrates strong interlayer interactions of aromatic rings, indexed as the (002) planes for g-C3N4. The smaller diffraction signal at around 13.1 o , listed as (100), is associated with the in-plane structural perpetual motif, i.e., the continuous tri-s-triazine structures [32]. Besides, the depth of the (002) peak had substantially diminished and expanded for g-C3N4 nanosheets [33]. The resulting doping with Fe evidenced no change of the crystal phase of g-C3N4. The location of diffraction peaks for nanosheets Fe@g-C3N4 switched to a steadily higher angle for nanosheets of g-C3N4. The peak intensity diminished, and the diffraction peak width broadened for iron content, implying the presence of excess Fe species caused the host-guest interactions and polymeric condensation inhibition. It is further evident that the iron is chemically coordinated to g-C3N4 via Fe-N bonds [34].

X-ray diffraction and FT-IR studies
.

2 / o
Wavenumber, cm -1 Figure 1. A -XRD data of g-C3N4 and Fe@g-C3N4, and B -IR spectrum of g-C3N4 and Fe@g-C3N4 Figure 1B shows the FT-IR spectra of the g-C3N4 and Fe@g-C3N4 nanosheets. From Figure 1B, the extended absorption band at nearly 3155 cm -1 , assigned to the stretching vibrational modes of surplus N-H components connected with uncondensed amino groups [35,36], can be identified. The peak at 1633 cm -1 is classified as vibrational stretching mode, whereas aromatic C-N stretching vibrations of heterocyclic rings matched with the bands at 1415, 1400, and 1234 cm -1 [26]. The specific particular peak at 807 cm -1 confirmed the s-triazine ring system [37]. Further, it is evidenced from Figure 1B that the intensity of the peaks reduced with an increase in Fe content in the Fe@g-C3N4, and the principal characteristic bands of g-C3N4 nanosheets change to smaller wavenumbers (redshift) intimates that the C-N and C=N bonds are weakened [29] SEM and EDS analysis The g-C3N4 and Fe@g-C3N4 synthesis was demonstrated by SEM and EDS study. Figures 2A and  2B denote the SEM image of integrated g-C3N4 and Fe@g-C3N4 nanocomposite. The micrographs were obtained at 3,000 magnifications by an expediting voltage of 5.0 kV LED. The sheets-like morphology evidences a greater surface area for the catalytic reactions between target and transducer interface [38]. Figure 2C shows the EDS spectrum of the synthesized material, and it reveals elemental composition. The EDS data illustrated the presence of carbon (C) 51.92 %, nitrogen (N) 31.24 %, oxygen (O) 2.96 %, and iron (Fe) 13.88 % in the Fe@g-C3N4 sample, indicating that Fe-doped g-C3N4 has a pretty high rate of purity, and it comprises solely four elements. These essential considerations made on EDS interpretation insinuate the purity of the substance. Energy, keV well-defined quasi reversible oxidation and reduction peaks with a peak-to-peak separation (ΔEp) of 201.6 mV ( Figure 3A, curve a). The peak current increased after the deposition of 3 µl Fe@g-C3N4, which resulted in peak separation (ΔEp) of 271.2 mV ( Figure 3A, curve b). The increase in current response and stability of the CV curve suggests the successful deposition of Fe@g-C3N4 on the PGE electrode surface.
EIS spectra of PGE and PGE/Fe@g-C3N4 recorded in 1 mM [Fe(CN)6] 4-/3-PBS solution Figure 3B are displayed in the Nyquist diagram. From the EIS data, the charge transfer resistance (Rct) could be calculated from the best fit of the Randles electrical equivalent circuit. The Rct for bare PGE is 78.73 kΩ, and after the addition of Fe@g-C3N4, Rct is reduced to 19.41 kΩ on account of the higher rate of electron transfer between the redox probe and electrode surface. Reduction in Rct values on the deposition of Fe@g-C3N4 is clearly in agreement with CV results, and it further confirms the successful electrode deposition.
Randles-Sevcik equation (Equation 3) was used to validate the improved catalytic response of modified PGE in terms of increased active electrode surface area [39].  [40]. From this equation, the PGE/Fe@g-C3N4 active electrode surface area was calculated to 3.94 cm 2 , ensuring the high electrocatalytic surface area for the modified electrode [41].

Figure 3. A -CVs of 5 mM [Fe(CN)6] 3-/4solution obtained held at (a) bare PGE, (b) PGE/Fe@g-C3N4(3 µl); B -Nyquist diagrams of EIS data of (a) bare PGE, (b) PGE/Fe@g-C3N4 (3 µl)
Effect of scan rate and pH Figure 4A shows the effect of the scan rate on the modified PGE/Fe@g-C3N4 for various scan rates like from 25 up to 375 mV s -1 in 10 μM solution of CFX. The oxidation peak current increases with the scan rate. Figure 4B shows the linear relationship between the current peak height and the square root of the scan rate with the regression coefficient of R 2 =0.9966. It is because the larger surface area facilitates faster electron transfer.
Further, the regression analysis of log Ip versus log ν plot gives a relation of log Ip=0.7601 log ν + + 1.7926; R 2 = 0.9973 with the slope close to 1. The value of slope of log Ip versus log v confirms that the electrode process is dominantly diffusion-controlled [41].
To know the effect of pH on the electrochemical properties of the proposed sensor, the electrochemical performance of the Fe@g-C3N4 decorated PGE electrode was investigated in different pHs in the range 3 -9) in the presence of 10 μM CFX in PBS buffer solution. From Figure 4C, the current response of the sensor at different pH values reveals that the oxidation peak reached the maximum at pH 7.0. Therefore, pH 7.0 was adopted for all further analyses. With the increase in the concentration of CFX, the oxidation peak current proportionately increases at the Fe@g-C3N4 modified electrode. Oxidation of CFX is the electrochemical reaction occurring at the electrode/electrolyte interface, as shown in Scheme 1. The anodic peak current increased linearly from 1 to 100 µM of CFX with a correlation coefficient (R 2 ) of 0.9675. Hence, this proves the authenticity of the sensor performance.
The DPV experiment was conducted in 0.1 PBS solution in the potential range -0.1 to 1.5 V at smaller concentrations of CFX. Fig 5B illustrates the DPV current response from 1 to 1000 nM of CFX, and it is clear from Fig 5B that there is a linear relationship between current and CFX concentration with a correlation coefficient of R 2 = 0.9968. The data from DPV experiments were used to compute the analytical quantities of the sensor i.e., the limit of detection (LOD), sensitivity, and quantification limit (LOQ) (equations 4, 5, and 6) [42]. (5) LOQ = 10 / S (6) Here, σ represents the standard deviation of the blank, and S indicates the slope of the calibration plot (Inset: Figure 5B). The sensitivity, LOQ, linear range, and LOD determined using the above experimental data are 0.0596 µA mM -1 cm -2 , 0.0018 µM, 0.001 to 1.0 µM, and 5.4 nM, respectively. The overall analytical performance of the suggested Fe@g-C3N4 based sensor is in accordance with those already reported in the literature (Table 1) [43][44][45][46][47]. The superior performance characteristics exhibited by the PGE/Fe@g-C3N4 sensor are explicitly related to the composite matrix's synergic effects [48]. This suggests that the as-developed sensor can be a promising tool in the analysis of CFX.

Effects of interferents
The proposed sensor was subjected to radical scavenging experiments to affirm its selectivity. The DPV responses were recorded after the additions of 50 µM of some common interferents like AA, UA, glucose, Ca 2+ , and Mg 2+ to 10 µM CFX in 0.1 PBS buffer solution. The obtained DPV responses are exhibited in Figure 6. Figure 6 discloses no significant variation in the current peaks despite the residence of interferents, implying the selectivity and robustness of the sensor for field purposes.

Repeatability, reproducibility, and stability
These experiments are crucial to argue on the sensor's practical applicability and reliability. The repeatability of the sensor performance was tested by measuring the CV response for 10 µM CFX in 0.1M PBS solution for twenty electro-analytical cycles between −0.1 to 1.5 V at a scan rate of 50 mV s -1 . Further, we examined the reproducibility of the PGE/Fe@g-C3N4 electrodes by preparing a set of five distinct electrodes using a method described in the experimental section. The current response of these electrodes was measured by CV in 0.1 M PBS comprising 10 µM CFX. The estimated magnitude of the current response under identical circumstances depicts comparable electrochemical properties for the sensor with a suitable shift in peak current, demonstrating an agreeable reproducibility as apparent from Fig   The storage stability of the PGE/Fe@g-C3N4 electrode had been determined steadily for up to twenty days, and the results are shown in Figure 7B. The current response for 10 µM CFX was monitored at regular intervals, and the developed sensor retained 100, 98.01, 97.13, 97.02, and 96 % of the initial current response after 0, 5, 10, 15, and 20 days of storage, respectively, suggesting that the developed sensor exhibits a high level of stability in the detection of CFX.

Real sample analysis
The potency of the proposed sensor in practical applications was ascertained by analyzing CFX in human blood specimens. The actual samples were diluted with PBS in equal proportions, accompanied by spiking a known quantity of CFX to them. The responses were measured, and the observed percentage of the recovery is registered in Table 2. As noted, the results attained are good, with insignificant errors and hence the developed sensor might be used for the determination of CFX in biological fluids. It ratifies the generated sensor for primary specimen analysis. The worthy administration characteristics buttressed by the PGE/Fe@g-C3N4 sensor are completely ascribed to the synergic effect of the composite matrix.

Conclusion
In the present work, we developed an electrochemical sensor using Fe@g-C3N4 composite as a working electrode matrix. The physical, chemical and electrochemical investigations of the Fe@g-C3N4 matrix confirmed the sensor's stability, conductivity, and electrocatalytic nature. It has the advantage of a low detection limit (5.4 nM) and a wide linear range (0.001-1.0 µM) in the detection of CFX. The offered sensor manifested an exceptional selectivity, sensitivity (0.0596 µA mM -1 cm -2 ) and reproducibility regarding the CFX determination. This sensor proved comparatively improved performance than various CFX sensors reported earlier in the literature. Remarkably, the success had interlaced by the cost-effective matrix and optimized material usage of the sensor. The parameters including storage stability, reproducibility, and repeatability were studied. The PGE/Fe@g-C3N4 sensor may be an alternative to the reported sensors for detecting and quantifying CFX in blood, environmental and industrial specimens.