Electrochemical sensing of gatifloxacin using Ag₂S/RGO nanocomposite

2.1 Materials and reagents

Gatifloxacin hydrochloride, reduced graphene oxide (RGO), silver diethyldithiocarbamate (Ag (DDTC)), oleic acid (OA, 90 %), octadecylamine (ODA, 90 %), 1-octadecene (ODE, 90 %), and N, N-dimethylformamide (DMF, 99 %) are purchased from Sigma-Aldrich company (Merck, Darmstadt, Germany) without further purification. Potassium dihydrogen phosphate, purchased from Sinopharm Group Co., Ltd. (Beijing, China), is used to prepare 0.1 mol·L^-1 phosphate buffer solution (PBS buffer). Ultra-pure water (Millipore, Merck, Darmstadt, Germany) was used throughout the experiments, and all experiments are carried out at room temperature.

2.2 Material synthesis and sensor preparation

0.1 mmol/L of Ag (DDTC), 10 mmol/L of OA, 10 mmol/L of ODA, and 20 mmol/L of ODE were added to a 100 mL three-necked flask as a solvent. To remove any water and oxygen, the solution was continuously stirred and heated to 100℃ until it became transparent. The solution was heated to 200℃ (with a heating rate of 15 ℃/min) and kept in N₂ atmosphere for 10 min, 25 min, and 45 min. After cooling to room temperature, add excessive ethanol into the reaction solution. Then, wash and centrifuge the solution at 12000 rpm for 10 min. Repeat the above operation for 5 times, and finally dry the solid in air at 60℃. The resulting nanocrystals are easily redistributed in nonpolar organic solvents, i.e., cyclohexane. The yield of nano-crystalline is about 60–70 % (Du et al., 2010). In addition, the synthesis of Ag₂S/RGO is the same as that of Ag₂S, except that 1 mg of RGO dissolved in 1 mL DMF is added to the three neck flask, and the reaction time is 10 min.

The bare glassy carbon electrode (GCE) was polished with alumina powder (0.1 μm) and alumina powder (0.05 μm) successively, then washed with deionized water twice, washed with ethanol for 1 min, and dried and cooled for use. A certain amount of Ag₂S or Ag₂S/RGO (2 mg) was mixed with 50 μL chitosan (1 mg/mL) and 80 μL chloroform, and then 4.0 μL of the mixed solution with Ag₂S/RGO nanocomposites was dropped on the surface of the treated GCE. The brief diagram for the fabrication process of Ag₂S/RGO composites was illustrated in Scheme 1. After drying, the modified electrode was washed thoroughly with secondary distilled water. When not in use, grind and polish the modified electrode with alumina powder in turn, immerse it in PBS (pH 7.0, 0.1 mol/L), and keep it at constant temperature of 4 ℃, which can make the reaction liquid junction be moist and smooth and is very helpful for the modified electrode potential to maintain stable.

2.3 Instrument analysis

The electrochemical experiments were carried out on a CHI 660E electrochemical workstation (Chinstruments, Shanghai, China) (Yao et al., 2024). The SEM images were recorded with a Regulus 8100 (Hitachi, Tokyo, Japan), and the energy dispersive spectrometer (EDS) spectra were obtained with an EDAX EDS Element (AMETEK Inc., Berwyn, USA). XRD results was measured with a D/max-2600PC (Rigaku Corporation, Tokyo, Japan) via ceramic-monochromatic Cu K, which was operated at 45 kV and 20 mA with a scanning range of 20 – 70° and a scanning rate of 5°/min in 2θ. The transmission electron microscopy (TEM) was performed on a Talos F200S (FEI 2 SDD, Thermo Scientific, USA).

2.4 Electrochemical detection

Linear Sweep Voltammetry (LSV) is a common electrochemical analytical technique with three electrodes that studies the redox behavior of electroactive species in solution by applying a linearly varying voltage (sweep voltage) to the working electrode and measuring the resulting current. Moreover, LSV method has the advantages of simple operation, high sensitivity, and wide application range, and has been widely used in analytical chemistry, materials science, biochemistry, environmental science and other fields. The kinetic parameters of the reaction can be determined by analyzing the slope and shape of the current–potential curve. Therefore, in this work, LSV analysis with a typical three-electrode cell, including a platinum wire as the auxiliary electrode, a modified GCE as the working electrode (glassy carbon disk electrode with 3 mm diameter and 7.065 mm² geometric area), and a saturated calomel electrode (SCE) as the reference electrode, was applied to detect GAT. The optimization of pH was carried out with the addition of acetate in the PBS buffer. When performing the electrochemical testing, the modified electrodes were first placed in acetate buffer (pH = 4.0) containing different concentrations of GAT and then deposited using the current–time method to achieve electrodeposition. After stirring and letting stand, the dissolution peak of GAT was recorded via oxidative dissolution measurements with LSV, in which the step increments is 4 mV, and the scanning potential range is −1.0 to −0.2 V. After each test, the potential was returned to 0.1 V and the electrodes were washed for 120 s for clean-up. In the anti-interference and stability experiments, the same conditions were carried out. Besides, the CV measurements were carried out in a mixture solution of 20 mmol·L^-1 K₃[Fe(CN)₆]/K₄[Fe(CN)₆] and 0.1 mol·L^-1 KCl with a scan rate of 100 mV·s⁻¹. Three kinds of real samples, including shrimp, crucian carp, chicken meat, were collected from Shanghai market. The pretreatment was based on the standard method as reported before (GB/T 20366–2006). 5.0 g of sample, 20 mL of formic acid-acetonitrile solution were added to a 50 ml polyfluoroethylene centrifuge tube. Then, centrifuge the solution at 4000 rpm for 5 min. In simple terms, the sample is placed in a petri dish and different concentrations of gatifloxacin are added to the sample. The samples were stored at 4 ℃ and tested with Ag₂S/RGO electrochemical sensors. Before the analysis of GAT, the pH of extracted samples was modified to 4.0 using acetate buffer solution. The concentrations of GAT in the real samples was fabricated with the standard addition method (10, 100, and 200 μmol⋅L^-1), and then LSV analysis was carried out for three parallels.

3.1 Material characterization

X-ray diffraction (XRD), transmission electron microscopy (TEM), and scanning electron microscopy-x-ray energy dispersive spectrometer (SEM-EDS) were used to characterize the crystal phase, morphology, structure and element content of the materials. Ag₂S was synthesized in the organic phase of diethyldithiocarbamate (Ag (DDTC)), oleic acid (OA), octadecylamine (ODA) and 1-octadecene (ODE). Ag₂S nanoparticles with different particle size were prepared by controlling the reaction time. X-ray diffraction (XRD) was applied to determine and analyze the crystal structure of the synthetic Ag₂S nanoparticles. The results in Fig. S1 (Supplementary Material) indicate that the material prepared by organic phase method is monoclinic Ag₂S because of the similar pattern of the synthesized materials to the standard JCPDS 14–0072 (Jang et al., 2021). In details, the angle, height and shape of the corresponding peaks in the two patterns are basically the same, and the characteristic diffraction peaks (28.97°, 31.52°, 34.39°, 40.74°) generally coincide (Ghayeb et al., 2016).

Based on the XRD data of Ag₂S and Scherrer formula, Ag₂S particle size with a reaction time of 10 min can be estimated to be 13.6 nm. $$D=\frac{K\lambda }{\beta cos\theta }=\frac{0.89\times 0.15405}{\frac{0.60}{180}\times \pi \times \mathrm{c}\mathrm{o}\mathrm{s}(\frac{31.52}{2})}=13.6nm$$

As shown in Fig. S1, the XRD spectra of the fabricated Ag₂S is well consistent with the Ag₂S standard card, indicating the successful synthesis of Ag₂S. Besides, the successful fabrication of Ag₂S can also be proved by the EDS results in Fig. S2, in which the Ag peak is clearly observed. As shown in Fig. S2 and Table S1 in the Supplementary Material, the Scanning electron microscope X-ray energy spectrum (SEM-EDS) once again confirmed the successful fabrication of Ag₂S since the atomic proportion of Ag and S is approximately 2:1 in the element distribution of the synthetic materials, taking the reaction for 10 min as an example (Riojas et al., 2019; Wong et al., 2022).

In order to analyze the morphology and crystal phase of the composite, the transmission electron microscopy (TEM) was carried out and the results are shown in Fig. 1. Fig. 1a-c exhibits that the generated Ag₂S nanoparticles in every group have uniform particle size and good dispersibility. Moreover, with the prolonging of the reaction time, the average size of Ag₂S nanoparticles kept increasing with the diameter ranging from ∼ 10 nm to ∼ 30 nm. From the TEM images of Ag₂S/RGO as illustrated in Fig. 1d, it can be seen that the Ag₂S was round and successfully attached to the RGO surface, and the agglomeration of the nanomaterials was effectively prevented. Moreover, the spherical morphology of silver sulfide increased the specific surface area of the composite material, which led to the enhancement of the light absorption ability. Therefore, the photocatalytic activity of the composites was improved (Shafi et al., 2019). Besides, the inset histogram for the particle size distribution in Fig. 1d also demonstrated the major diameter of ∼ 10 nm. To further confirm the structure of the fabricated materials, high resolution TEM studies were performed as depicted in Fig. 1e. After comparing with the standard cards, it can be determined to be (111) crystal surface and the lattice spacing is 0.2836 nm (Chih and Yang 2014), which is consistent with monoclinic crystal system. These results confirm that the observed nanocrystals on the RGO layer are Ag₂S nanoparticles with a uniform spherical structure.

Close

Fig. 1

TEM images of Ag₂S nanoparticles with (a) 10 min reaction time, (b) 25 min reaction time, and (c) 45 min reaction time; (d) TEM images of Ag₂S/RGO materials at 10 min reaction time with inset of the histogram for the particle size distribution; (e) HR-TEM images of Ag₂S nanoparticles on the RGO; (f) CV plots of bare GCE, Ag₂S/RGO/GCE and Ag₂S/GCE in 20 mmol·L^-1 [Fe(CN)₆]^3-/4- solution.

Export to PPT

Fig. 1f displays the electrochemical cyclic voltammetry (CV) of Ag₂S and Ag₂S/RGO modified electrodes in 0.1 mol L^-1 KCl solution of 20 mmol L^-1 [Fe(CN)₆]^3-/4-. Firstly, although the current intensity of Fe²⁺/Fe³⁺ on Ag₂S/GCE has no obvious advantage compared with that on the bare electrode, the current response with the reaction time of 10 min is the largest and the redox peak potential difference decreases, which proves that Ag₂S has good electrocatalytic performance and increases the reversibility of the electrochemical reaction. For Ag₂S modified systems, it can be seen from Fig. 1f that the Fe²⁺/Fe³⁺ redox peak current increased on Ag₂S modified electrode (Ag₂S/GCE) with the shorter reaction time. The results can be explained that the smaller the size of nanoparticles, the larger the specific surface area, and the faster the electron transfer. In order to further improve the electrochemical properties of modified electrode and the potential response to gatifloxacin, the Ag₂S/RGO modified electrode (Ag₂S/RGO/GCE) was constructed (10 min reaction time). Fig. 1f shows that the peak current of Fe²⁺/Fe³⁺ on the Ag₂S/RGO/GCE is the largest. The combination of silver sulfide and RGO effectively prevents the agglomeration of nanomaterials and increases the specific surface area of the composite due to the Ag₂S nanoparticles, thus enhancing the photocatalytic activity. The observed higher current at Ag₂S/RGO composites and the synergistically enhanced electrochemical properties might speculatively be attributed to the following three aspects: (i) the abundant oxygen-containing groups and high adsorption capacity of RGO might do favor to the touching of targeted GAT to the modified electrode; (ii) the large specific surface area of Ag₂S nanoparticles might provide more abundant active sites for the GAT reaction; and (iii) the Ag₂S/RGO composites may further improve the good conductivity of the electrode material as mentioned above.

Electrochemical impedance spectroscopy (EIS) can provide useful information regarding impedance changes on the electrode surface during the fabrication process. The charge-transfer resistance was related to the electron-transfer kinetics of the redox probe ([Fe(CN)₆]^3-/4-) at the electrode interface and could be estimated from the diameter of the semicircular part of the EIS curve. The typical Nyquist plots obtained from bare GC, Ag₂S/GC, RGO/GC, and Ag₂S/RGO/GC (with RCT values of about 360, 440, 290, 60, respectively) in aqueous KCl (0.1 mol/L) containing 20 mmol/L [Fe(CN)₆]^3-/4- were illustrated in Fig. S6. A large diameter of the semicircle part was observed for GC. The diameter of the semicircle was increased when the GC surface was coated with Ag₂S nanoparticles due to poor electrical conductivity. The semicircle diameter showed the decrease in RGO/GC due to the excellent electron transfer ability. The electrochemical impedance spectrum of Ag₂S/RGO/GC had the smallest semicircle diameter, which proved that the fastest electron-transfer kinetics of [Fe(CN)₆]^3-/4- occurred in Ag₂S/RGO/GC rather than in other modified electrodes. So the EIS results were in full agreement with the CV results. These results generally indicate that Ag₂S/RGO has good electrocatalytic ability and extremely strong conductivity, which can provide strong support for the subsequent detection of gatifloxacin.

3.2 Optimization of Ag₂S/RGO/GCE based electrochemical sensor

Fig. 2a illustrated the LSV curves of 200 μmol·L^-1 gatifloxacin on bare GCE and Ag₂S/RGO/GCE with a sweep rate of 0.1 V·s⁻¹. It can be observed that Ipa of gatifloxacin on Ag₂S/RGO/GCE is significantly increased, which is about 2.5 times of that on the bare electrode. It turns out that Ag₂S/RGO with large specific surface area and good conductivity has a good catalytic effect on the electrochemical oxidation of gatifloxacin and accelerates the electron transfer on the electrode. Thus, the Ag₂S/RGO modified GCE was used in designing the electrochemical sensors for gatifloxacin detection. As reported before, the electrochemical reaction of GAT has been reported to be a two electron-transferred oxidation process accompanied by two proton releasing due to GAT’s redox mechanistic properties (Li et al., 2024; Zhu et al., 2020). Thus, the mechanism herein may possibly be the addition of hydroxyl group on the N4 atom with the formation of a hydroxylamine during electro-oxidation, which is a process of two electron transfer and two proton loss, similar to previous publications (Li et al., 2024; Zhu et al., 2020). To make it clear, the speculative reaction equation was depicted in Scheme 1 with the possible electron transfer process and the potential reaction product.

Close

Fig. 2

(a) LSV plots of Ag₂S/RGO/GCE and bare GCE in 200 µmol·L^-1 gatifloxacin solution; (b) CV plots of bare GCE and Ag₂S/RGO/GCE in 200 µmol·L^-1 gatifloxacin solution; (c) LSV of 40 µmol·L^-1 gatifloxacin at Ag₂S/RGO/GCE electrode in the different buffer solution with various pH values; (d) Cyclic voltammograms of Ag₂S/RGO/GCE electrode at different scan rates in 100 µmol·L^-1 gatifloxacin solution.

Export to PPT

Fig. 2b displays the cyclic voltammogram (CV) profiles of 200 µmol/L gatifloxacin at the bare glassy carbon electrode and the Ag₂S/RGO/GCE with a sweep rate of 0.1 V·s⁻¹. The electrochemical oxidation of gatifloxacin at the Ag₂S/RGO/GCE shows a significant increase of about 2.5-fold compared to the bare electrode, along with a negative shift of the peak potential (−0.01 V). This suggests that Ag₂S/RGO/GCE, with its large specific surface area and good conductivity, has a positive catalytic effect and accelerates electron transfer on the electrode.

Apart from that, Fig. S3 in the Supplementary Materials illustrated the linear sweep voltammetry (LSV) curves of the bare GCE and Ag₂S/GCE prepared at different reaction times in gatifloxacin solution of 200 μmol·L^-1, with a scanning rate of 0.1 V/s. It was observed that the oxidation peak potential is near 1.1 V, and the electrochemical response of gatifloxacin on Ag₂S/GCE was slightly different with different reaction time. As mentioned before, different reaction time will affect the size of Ag₂S nanoparticles. The results show that the shorter the reaction time is, the smaller the grain size is, and the better electrochemical signal is. Sheng et al. (2014) reported a monocrystalline 3D TiO₂ branching nanowire array consisting of one-dimensional branching that grows epitaxial from the backbone. The 3D branched-chain nanoarrays exhibited an increase in the specific surface area, along with characteristics of rapid charge transfer and high absorption capacity (Sheng et al., 2014). Therefore, the probable reason might be that the specific surface area is larger and the electron transfer is faster for the small-size Ag₂S nanoparticles, which consequently result in the better electrochemical performance. Therefore, the Ag₂S with the shortest reaction time (10 min) and the smallest particle size is selected as the initial modified electrode material. Due to the imino group of gatifloxacin molecules, pH in PBS buffer solution might affect the electrochemical response of gatifloxacin. Fig. 2c shows the LSV response of gatifloxacin on Ag₂S/RGO electrode in PBS solution with different pH values. pH variations in PBS buffer solution will affect the electrochemical response of gatifloxacin. We could observe that with the increase of pH, I_pa firstly increased and then began to decrease. It can be concluded that the best pH of PBS solution is 4 for the electrolyte. Moreover, with the decrease of pH, the oxidation peak potential of gatifloxacin at Ag₂S/RGO electrode was found negatively shifted. It exhibited a linear relationship between the potential and pH, as shown in Fig. S4, with the linear regression equation of E_pa(V) = -0.047pH + 1.300 (R² = 0.9886). This tendency indicated the protons participate in the oxidation of gatifloxacin. The slope in this relationship (−47 mV/pH) approximated the theoretical value of −59 mV/pH (Jiang et al. 2016; Zhang et a., 2013b; Li et al., 2024; Zhu et al., 2020). This indicated that the proton number and electron transfer number involved in the oxidation of gatifloxacin on Ag₂S/RGO electrodes were equal, implying that it possibly followed the 2-electron-2-proton mechanism (Jiang et al. 2016; Zhang et a., 2013b; Li et al., 2024; Zhu et al., 2020). Generally, the current result was in accordance with the reported value and the probable mechanism of the reaction was surmised as shown in Scheme 1.

Scanning rate is of great importance for the sensing performance of an electrochemical sensor, so the effect of scanning rate on the electrochemical current response was also investigated next. As shown in Fig. 2d, when the scanning rate increases gradually from 20 mV·s⁻¹ to 300 mV·s⁻¹, Ipa of GAT kept increasing with the Epa moving forward gradually. The results suggested that the current reaction is an electrochemical process controlled by adsorption of GAT on Ag₂S/RGO electrode and the oxidation reaction of GAT is irreversible in this process. It is worth noting that Ipa of GAT decreased with the decrease of v, and thus, the sensitivity could also be weakened. Meanwhile, with the increase of v, Ipa also increased, but Epa shifted positively. Therefore, it is necessary to select the relatively optimal scanning rate on consideration of both the current response and the sensitivity. Finally, after comprehensive consideration, the optimized scanning rate was determined as 100 mV·s⁻¹ for the following electrochemical sensing tests.

3.3 Linear range and sensitivity

After the multiple optimization, Ag₂S/RGO/GCE accomplished LSV detection of gatifloxacin under optimal experimental conditions (pH 4.0, v of 100 mV·s⁻¹), and the working curve was obtain as shown in Fig. 3. Generally speaking, for both concentration range, the Ipa continuously increased with the rise of gatifloxacin concentration in Fig. 3a and 3b. Furthermore, as shown in Fig. 3c and 3d, in the concentration range of 0.2 μmol·L^-1 ∼ 20 μmol·L^-1 and 20 μmol·L^-1 ∼ 250 μmol·L^-1, there are good linear relationships between Ipa and concentration of gatifloxacin on Ag₂S/RGO/GCE. At low concentration, the linear regression equation is: Ipa (μA) = 0.07212c (μmol·L^-1) + 0.0216 (R² = 0.9947). At high concentration, the linear regression equation is: Ipa (μA) = 0.0973c (μmol·L^-1) − 1.6688 (R² = 0.9983). The detection limit was calculated to be 6.67 × 10^-8 mol·L^-1 base on the formula of Signal/Noise being 3:1 (Liu et al., 2021).

Close

Fig. 3

LSV of different concentrations of gatifloxacin at Ag₂S/RGO/GCE: (a) 0.2 µmol/L − 20 µmol/L, (b) 20 µmol/L − 250 µmol/L; Calibration curves of the concentration and oxidation current of gatifloxacin: (c) 0.2 µmol/L − 20 µmol/L, (d) 20 µmol/L ∼ 250 µmol/L.

Export to PPT

In order to evaluate the detection performance of the constructed sensors more intuitively, we compared this study with the existing literatures about GAT sensors as listed in Table 1. In general, the proposed Ag₂S/RGO-based sensor shows relatively wide linear ranges for both low and high concentrations of GAT. This performance is, to some extent, at leading level among the existing GAT sensors whose linear ranges are often located at either low or high concentrations as shown in Table 1. Moreover, the limit of detection (LOD) of this sensor is relatively low (0.0667 μmol⋅L^-1), justifying the high sensitivity and remarkable advantage of the currently proposed sensor.

3.4 Repeatability, reproducibility, and selectivity

Repeatability and reproducibility are two essential indexes for a sensor. Therefore, the same Ag₂S/RGO/GCE was constructed for five times in the same concentration of gatifloxacin solution and the relative standard deviation (RSD) was calculated to be 1.9 %. Furthermore, five Ag₂S/RGO/GCE with the same conditions were prepared and tested in the same gatifloxacin solution with a RSD of 3.5 %. The bar diagram for the repeatability and reproducibility of the sensor was illustrated in Fig. S5. These results show that the fabricated Ag₂S/RGO/GCE sensor has good repeatability and reproducibility. Moreover, the long-term stability of the sensor is also an important factor which deserves consideration. The results show that after storage for one week at room temperature, the same sensor retained a current response of 97.5 % of the initial current, indicating the satisfactory stability of the developed electrochemical sensor.

Besides, in the actual environments, the sensing platforms are often co-existed with various probable interferences, so the interference resistance of the proposed sensors should be investigated. Herein, we investigated the influence of various possible interferences through mixing 40 μmol⋅L^-1 of different compounds, including ions, organic compounds, and other antibiotics (ciprofloxacin (CIP), enoxacin (ENX), norfloxacin (NFO), ascorbic acid (ASA), glucose (GLC), levodopa (L-dopa), uric acid (UA), K⁺, Mg²⁺, Fe³⁺, Zn²⁺) with 40 μmol⋅L^-1 GAT in the acetate buffer solution (pH = 4). As depicted in Fig. 4, none of the tested interferences interfered the accurate analysis of GAT, suggesting that the proposed GAT sensors in this study possess high selectivity and anti-interference. Overall, this electrochemical sensor based on Ag₂S/RGO/GCE has good selectivity, low detection limit, and wide range of linearity, for the sensing of GAT, and exhibits promising potential in the analysis of GAT in real samples.

Close

Fig. 4

Response of the Ag₂S/RGO/GCE sensor to GAT with or without interfering substances.

Export to PPT

3.5 Real sample analysis

In order to verify the applicability of Ag₂S/RGO/GCE sensor in real samples, the standard addition method is utilized to test the detection performance in various real samples and the results are listed in Table 2. It can be seen that the recovery rates are within the range of 91.8 % − 102.4 % after the addition of various GAT concentrations, suggesting that the proposed electrochemical sensor is effective and reliable for the gatifloxacin analysis in different real samples. Moreover, all the RSDs were lower than 4.0 % after 5 repetitions, revealing the excellent precision and stability of the developed sensors.