Designing a Stable g-C3N4/BiVO4-Based Photoelectrochemical Aptasensor for Tetracycline Determination

The excessive consumption of tetracycline (TC) could bring a series of unpredictable health and ecological risks. Therefore, it is crucial to develop convenient and effective detection technology for TC. Herein, a “signal on” photoelectrochemical (PEC) aptasensor was constructed for the stable detection of TC. Specifically, the g-C3N4/BiVO4 were used to promote the migration of photo-generated charges to an enhanced photocurrent response. TC aptamer probes were stably fixed on the g-C3N4/BiVO4/FTO electrode as a recognition element via covalent bonding interaction. In the presence of TC, the aptamer probes could directly recognize and capture TC. Subsequently, TC was oxidized by the photogenerated holes of g-C3N4/BiVO4, causing an enhanced photocurrent. The “signal on” PEC aptasensor displayed a distinguished detection performance toward TC in terms of a wide linear range from 0.1 to 500 nM with a low detection limit of 0.06 nM, and possessed high stability, great selectivity, and good application prospects.


Introduction
Nowadays, various hazardous small molecules (i.e., antibiotics, insecticides, environmental pollutants, pesticide residues, and hormones) have become significant factors threatening environmental health. Tetracycline (TC), a common hazardous small molecule, has been widely used as a feed additive in agriculture and livestock breeding on account of its economic advantage, broad-spectrum activity against bacteria, and favorable oral absorption [1,2]. However, the abuse of TC can cause residues in food-producing animals and environmental media, and the consumption of such food or exposure to contaminated soil/water can cause serious harm to the human body, such as allergic reactions, skeletal dysplasia, and hepatotoxicity [3,4]. Additionally, it can cause the spread of super bacteria with drug resistance and reduce the efficient treatment of bacterial infections [5]. Therefore, the development of efficient platforms for monitoring tetracycline or other hazardous small molecules is an objective and realistic demand. Currently, the main methods of TC detection are chromatographic detection, bioassay, and immunoassay, e.g., high-performance liquid chromatography (HPLC), fluorescence assay, lateral flow assay, and mass spectrometry [6][7][8][9][10]. Although these traditional techniques are able to detect TC, there are shortcomings, including high testing costs, complicated operations, long detection time, poor reproducibility, and so on. Thus, an efficient strategy for highly reliable TC detection is critical.
Indeed, various sensors have been rising to the challenge in becoming a feasible technique to detect TC, on account of their simple operation, high selectivity, and low costs, such as biomimetic electrochemical sensor [11], electrochemical sensor [12], electrochemical immunosensor [13], fluorescent sensor [14], etc. Alternatively, the photoelectrochemical (PEC) aptasensor has been drawing broad attention, owing to its low background noise, Toxics 2023, 11, 17 2 of 10 great sensitivity, and fast response [15,16]. It is worth noting that the PEC aptasensor can also directly realize the expression of the response signal without extra signal indicators, which is so practical compared to other sensors [17][18][19]. In PEC construction, the selection of appropriate photoactive materials is favorable to obtain a high efficiency and stable signal output [20]. Many photoactive materials (e.g., inorganic semiconductors and organic materials) have been developed to construct the PEC aptasensor, due to their unique PEC properties [21]. Among them, graphitic carbon nitride (g-C 3 N 4 ) has been widely explored owing to its feasible photoactivity, low cost, and appropriate bandgap [22]. Nevertheless, pure g-C 3 N 4 shows poor PEC properties due to its quick recombination of photogenerated electrons/holes, which restricts the construction of a high-performance PEC aptasensor [23]. Therefore, it is vital to effectively address this issue. In fact, a number of methods, including element doping, morphological control, and heterojunction system have been reported [24,25]. Inspiringly, the g-C 3 N 4 -based heterojunctions as an uncomplicated method are able to significantly improve the PEC performance of g-C 3 N 4 [26]. For instance, a proposed PEC sensor based on ZnIn 2 S 4 /g-C 3 N 4 heterojunction was developed for the sensitive detection of bisphenol A [27]. BiVO 4 as a magnetic metal oxide possesses considerable merits, i.e., high visible light response, low cost, and low toxicity [28]. Researchers have pointed out that BiVO 4 was suitable for decorating g-C 3 N 4 owing to their well-matched energy-level structure, and the designed heterojunction has promoted the fast transfer of electrons and exhibited attractive and stable photocurrent signal [29]. Thence, BiVO 4 /g-C 3 N 4 heterojunction could be expected to improve an increased PEC performance and be applied to the design of a PEC aptasensor for detecting TC.
Herein, we fabricated a reliable PEC aptasensor based on g-C 3 N 4 /BiVO 4 as an electrode material for TC determination. Additionally, the TC aptamer probes could be tightly assembled on the surface of g-C 3 N 4 /BiVO 4 via covalent bonding interaction, to prevent it from falling off the electrode and further obtain a more stable and reliable signal response. When there is presence of TC, the aptamer probes could specially capture the TC and be oxidated by the accumulated holes on the photoactive materials and impede the electronhole recombination, causing an enhanced photocurrent signal [30]. This work provided a promising method for a feasible analysis platform with excellent stability, selectivity, and detectability.
A scanning electron microscope (SEM, ZEISS Sigma 300, Oberkochen, Germany) was used to conduct morphology of the prepared materials. X-ray diffractometer (Panaco X' Pert-PRO Bruker AXS D8 Advances, Almelo, Netherland) was employed for X-ray diffraction (XRD) analysis. X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha, Waltham, America) was carried out to evaluate the chemical compositions of the material. LAMBDA 750 UV/Vis/NIR Spectrophotometer was run to evaluate the absorbance properties of the samples. Microinfrared spectroscopy (Nicolet iN10 MX Infrared Imaging Microscope) was performed to determine the structure of the materials. A CHI 660E workstation (CH Instruments Ins., Shanghai, China) was operated for PEC and electrochemical measurement.

Synthesis of g-C 3 N 4
A ceramic crucible with 10 g melamine was placed in a muffle furnace and subsequently heated to 550 • C for 4 h. Secondly, the resulting product was further heated at 550 • C for 2 h to obtain pure g-C 3 N 4 nanosheets [32].

Synthesis of g-C 3 N 4 /BiVO 4
The g-C 3 N 4 /BiVO 4 with different weight ratios were prepared by the same procedure [33]. Specifically, 5 g-C 3 N 4 was dissolved in 100 mL water and agitated for 30 min. Then, Bi(NO 3 ) 3 ·5H 2 O and NH 4 VO 3 with a molar ratio of 1:1 were dropped into the above solution, followed by stirring for 30 min. Then, the mixture was further rinsed and dried at 80 • C for 12 h and calcined to 400 • C for 2 h.

Fabrication of the PEC Aptasensor
First, 200 µL of 4 mg mL −1 g-C 3 N 4 /BiVO 4 suspension was added to the fluorinedoped tin oxide (FTO) electrode to obtain the working electrode (g-C 3 N 4 /BiVO 4 /FTO electrode). Afterwards, 20 µL each of 0.1 wt% chitosan (CS), 5 wt% glutaraldehyde (GA), and 4 µM TC aptamer solution was added to the surface of the working electrode, so that the TC aptamer probes could be tightly immobilized to the surface of electrode via covalent bonding interaction. Finally, 20 µL of 1 wt% BSA was dropped on the above electrode to prevent non-specific adsorption [34,35].

Detection of TC
Prior to the PEC detection, 20 µL of TC solution with different concentrations (0, 0.1, 1, 10, 50, 100, 500 nM) was dropped on the prepared electrode and incubated at room temperature for 60 min, followed by rinsing with 0.01 M PBS and testing PEC signal in Na 2 SO 4 (0.1 M) solution [15,32].

Working Principle of the PEC Aptasensor
Scheme 1 depicted the working principle of the proposed PEC aptasensor. In this progress (Scheme 1A), an FTO electrode was sequentially decorated by the g-C 3 N 4 /BiVO 4 , CS, GA, TC aptamer probes, BSA, and TC. Here, the g-C 3 N 4 /BiVO 4 heterojunction was designed successfully owing to the well-matched energy level of g-C 3 N 4 and BiVO 4 (Scheme 1B), which exhibited enhanced separation of photogenerated carriers and an improved, attractive, and stable photocurrent signal, indicating that it is feasible to fabricate more an accurate PEC aptasensor [29,36]. Moreover, the aptamer probes were tightly immobilized onto the surface of g-C 3 N 4 /BiVO 4 /FTO, due to the covalent bonding interaction (GA was connected to both the amino-rich CS and the amino-functionalized TC aptamer probes). Remarkably, π-π stacking and covalent bond interaction are common methods to fix aptamer probes onto the sensor surface. The π-π stacking have merits of simple operation, low cost, and short reaction time. However, to a certain degree, the covalent bond interaction appears to be more stable, because this method could make the aptamer probes tightly bonded to the electrode and not easily fall off, so that the prepared aptamer sensor is able to present relatively stable results [5,15]. Notably, after the decoration of the TC aptamer probes, the photocurrent was decreased due to the steric hindrance of biomolecules. Then, the BSA was dropped onto the modified electrode to prevent nonspecific adsorption. When there is existence of TC, the formation of aptamer-TC complexes brought about an increased photocurrent due to the oxidation by the holes accumulated on the g-C 3 N 4 /BiVO 4 , and further facilitated electrons transfer [30]. Specifically, the TC molecules could be oxidized directly by photogenerated holes in the valence band (VB) of BiVO 4 . Subsequently, the holes were consumed, which inhibited the recombination of photogenerated electrons/holes, resulting in the accumulation of more free electrons on the CB of g-C 3 N 4 , which were transferred to the electrode to generate an increased Toxics 2023, 11, 17 4 of 10 photocurrent signal (Scheme 1B). Therefore, a "signal on" PEC aptasensor for highly stable determination of TC was constructed.

Detection of TC
Prior to the PEC detection, 20 μL of TC solution with different concentrations (0, 0.1, 1, 10, 50, 100, 500 nM) was dropped on the prepared electrode and incubated at room temperature for 60 min, followed by rinsing with 0.01 M PBS and testing PEC signal in Na2SO4 (0.1 M) solution [15,32].

Characterization of the Prepared Materials
The morphologies of the synthesized g-C 3 N 4 /BiVO 4 were observed by SEM. As seen from Figure 1A, the g-C 3 N 4 showed the appearance of smooth sheet structure. After the introduction of BiVO 4 , the irregular flower-like BiVO 4 was directly anchored onto the surface of g-C 3 N 4 , confirming that the smooth surface of g-C 3 N 4 could provide a suitable platform for the decoration BiVO 4 , and further indicating that the electrode material was successfully prepared. Figure 1A also exhibited the XRD patterns of g-C 3 N 4 and g-C 3 N 4 /BiVO 4 to characterize the crystalline features. As for g-C 3 N 4 , two main peaks around 13.1 • and 27.3 • were observed, which could be attributed to the (100) in-plane structural packing of heptazine units and (002) interlayer stacking of the conjugated aromatic segment [29]. After the BiVO 4 decoration, the peaks of g-C 3 N 4 /BiVO 4 were matched with the XRD standard card of BiVO 4 (PDF#14-0688). Additionally, the diffraction peaks at 18.7 • , 28.  (1 5 2) diffraction phases of BiVO 4 , respectively [29]. Additionally, the peaks of g-C 3 N 4 in the g-C 3 N 4 /BiVO 4 were still retained but noticeably weakened owing to its low XRD intensity, and the existence of g-C 3 N 4 in the composite could be verified by other characterization.
XPS was employed to investigate the existence of g-C 3 N 4 and the interface interaction between g-C 3 N 4 and BiVO 4 in the g-C 3 N 4 /BiVO 4 composites. Figure 1B presented the full-scan XPS spectrum of the composites, showing C, N, Bi, V, and O elements were major compositions in g-C 3 N 4 / BiVO 4 . In Figure 1C, the spectrum C 1s displayed two peaks located at 287.98 eV and 284.58 eV, which were attributed to the tertiary carbon C-(N) 3 and the C-C bonds [37]. Four characteristic peaks at 398.48 eV, 400.10 eV, 401.18 eV, and 404.18 eV in the N 1s ( Figure 1D) might be due to the sp 2 hybrid nitrogen C=N-C, tertiary nitrogen N-(C) 3 , the nitrogen in the amino functional group C-N-H, and the πexcitation of C-N heterocycles, respectively [38]. The XPS analysis of C 1s and N 1s demonstrated that a triazine heterocyclic ring exists in the g-C 3 N 4 /BiVO 4 composite. The peaks of Bi 4f ( Figure 1E) at 159.03 eV and 164.31 eV were assigned to Bi 4f 7/2 and Bi 4f 5/2 , respectively [29]. The results indicated that Bi 3+ exists in the g-C 3 N 4 /BiVO 4 composite. Additionally, in Figure 1F, the V 2p at 516.70 eV and 523.98 eV could be associated with V 2p 3/2 and V 2p 1/2 , respectively [39]. Overall, these results illustrated that the g-C 3 N 4 / BiVO 4 was successfully prepared. 1E) at 159.03 eV and 164.31 eV were assigned to Bi 4f7/2 and Bi 4f5/2, respectively [29]. The results indicated that Bi 3+ exists in the g-C3N4/BiVO4 composite. Additionally, in Figure  1F, the V 2p at 516.70 eV and 523.98 eV could be associated with V 2p3/2 and V 2p1/2, respectively [39]. Overall, these results illustrated that the g-C3N4/ BiVO4 was successfully prepared. UV/vis diffuse reflectance spectrum (DRS) was performed to determine the optical absorption behaviors of the photoactive materials. In Figure S1A, the absorption edge of g-C3N4 was about 464 nm. After introducing BiVO4, the absorption edge of the complex was enhanced on account of the formation of a heterojunction between these two semiconductors [40]. Obviously, the g-C3N4/ BiVO4 could embrace more visible light which is of benefit to the PEC performance. Furthermore, Miro-FTIR spectra was adopted to confirm the chemical structures of the materials. As shown in Figure S1B, the g-C3N4/BiVO4 composites had similar absorption peaks of g-C3N4, suggesting no obvious structural change of g-C3N4 and g-C3N4/BiVO4 [38]. In conclusion, the g-C3N4/BiVO4 was successfully prepared, which could boost more charge transfer and will be helpful for the construction of the PEC sensor.

Fabrication of the PEC Aptasensor
To confirm the successful fabrication process of the PEC aptasensor, photocurrent measurement was conducted. In Figure 2A, the g-C3N4/BiVO4/FTO (curve b) displayed maximum photocurrent compared to the g-C3N4/FTO (curve a), owing to the appealing PEC properties of g-C3N4/BiVO4. However, after introducing aptamer probes and BSA, the photocurrent was certainly decreased due to their nonconductivity (curve c and curve d). Upon introduction of TC (curve e), the PEC response increased, which was attributed to the oxidation of TC by the accumulated holes on the prepared materials [41]. Electrochemical impedance spectroscopy (EIS) analysis can also be used to explain the feasibility of the PEC sensor. As illustrated in Figure 2B, g-C3N4/BiVO4/FTO electrode (curve b) presented a much smaller Rct compared to g-C3N4/FTO (curve a), suggesting that BiVO4 UV/vis diffuse reflectance spectrum (DRS) was performed to determine the optical absorption behaviors of the photoactive materials. In Figure S1A, the absorption edge of g-C 3 N 4 was about 464 nm. After introducing BiVO 4 , the absorption edge of the complex was enhanced on account of the formation of a heterojunction between these two semiconductors [40]. Obviously, the g-C 3 N 4 / BiVO 4 could embrace more visible light which is of benefit to the PEC performance. Furthermore, Miro-FTIR spectra was adopted to confirm the chemical structures of the materials. As shown in Figure S1B, the g-C 3 N 4 /BiVO 4 composites had similar absorption peaks of g-C 3 N 4 , suggesting no obvious structural change of g-C 3 N 4 and g-C 3 N 4 /BiVO 4 [38]. In conclusion, the g-C 3 N 4 /BiVO 4 was successfully prepared, which could boost more charge transfer and will be helpful for the construction of the PEC sensor.

Fabrication of the PEC Aptasensor
To confirm the successful fabrication process of the PEC aptasensor, photocurrent measurement was conducted. In Figure 2A, the g-C 3 N 4 /BiVO 4 /FTO (curve b) displayed maximum photocurrent compared to the g-C 3 N 4 /FTO (curve a), owing to the appealing PEC properties of g-C 3 N 4 /BiVO 4 . However, after introducing aptamer probes and BSA, the photocurrent was certainly decreased due to their nonconductivity (curve c and curve d). Upon introduction of TC (curve e), the PEC response increased, which was attributed to the oxidation of TC by the accumulated holes on the prepared materials [41]. Electrochemical impedance spectroscopy (EIS) analysis can also be used to explain the feasibility of the PEC sensor. As illustrated in Figure 2B, g-C 3 N 4 /BiVO 4 /FTO electrode (curve b) presented a much smaller R ct compared to g-C 3 N 4 /FTO (curve a), suggesting that BiVO 4 enhanced the transfer of the electrons. After the aptamer probes and BSA were modified on the electrode (curve c and curve d), the R ct increased gradually on account of the increased steric hindrance of the electrode interface [30]. During the presence of TC (curve e), the semicircles further increased, which was similar to the photocurrent results. These results confirmed successful modification of the PEC aptasensor. enhanced the transfer of the electrons. After the aptamer probes and BSA were modified on the electrode (curve c and curve d), the Rct increased gradually on account of the increased steric hindrance of the electrode interface [30]. During the presence of TC (curve e), the semicircles further increased, which was similar to the photocurrent results. These results confirmed successful modification of the PEC aptasensor.

Analytical Performance of the PEC Aptasensor
To confirm the detection capacity of the PEC aptasensor, various concentrations (0, 0.1, 1, 10, 50, 100, 500 nM) of TC were studied under the optimal conditions (the specific optimization results were detailed in Figure S2). As shown in Figure 3A, the photocurrent intensity of the PEC aptasensor was enhanced with the concentration of TC increasing. Additionally, the linear equation could be fitted to I (nA) = 29.0622 lg (C/nM) + 242.209 (R 2 = 0.989) in the range of 0.1-500 nM with a limit of detection (LOD) of 0.06 nM (S/N = 3) [42], which was below the maximum residue limits (MRLs) of TC (900 nM) in milk set by the US Food and Drug Administration [43]. Additionally, compared with most other methods reported previously, such as colorimetric, fluorescence, electrochemistry, and electrochemiluminescence (Table 1), the PEC aptasensor displayed excellent performance with a relatively lower detection limit and convenient operation.

Analytical Performance of the PEC Aptasensor
To confirm the detection capacity of the PEC aptasensor, various concentrations (0, 0.1, 1, 10, 50, 100, 500 nM) of TC were studied under the optimal conditions (the specific optimization results were detailed in Figure S2). As shown in Figure 3A, the photocurrent intensity of the PEC aptasensor was enhanced with the concentration of TC increasing. Additionally, the linear equation could be fitted to I (nA) = 29.0622 lg (C/nM) + 242.209 (R 2 = 0.989) in the range of 0.1-500 nM with a limit of detection (LOD) of 0.06 nM (S/N = 3) [42], which was below the maximum residue limits (MRLs) of TC (900 nM) in milk set by the US Food and Drug Administration [43]. Additionally, compared with most other methods reported previously, such as colorimetric, fluorescence, electrochemistry, and electrochemiluminescence (Table 1), the PEC aptasensor displayed excellent performance with a relatively lower detection limit and convenient operation. enhanced the transfer of the electrons. After the aptamer probes and BSA were modified on the electrode (curve c and curve d), the Rct increased gradually on account of the increased steric hindrance of the electrode interface [30]. During the presence of TC (curve e), the semicircles further increased, which was similar to the photocurrent results. These results confirmed successful modification of the PEC aptasensor.

Analytical Performance of the PEC Aptasensor
To confirm the detection capacity of the PEC aptasensor, various concentrations (0, 0.1, 1, 10, 50, 100, 500 nM) of TC were studied under the optimal conditions (the specific optimization results were detailed in Figure S2). As shown in Figure 3A, the photocurrent intensity of the PEC aptasensor was enhanced with the concentration of TC increasing. Additionally, the linear equation could be fitted to I (nA) = 29.0622 lg (C/nM) + 242.209 (R 2 = 0.989) in the range of 0.1-500 nM with a limit of detection (LOD) of 0.06 nM (S/N = 3) [42], which was below the maximum residue limits (MRLs) of TC (900 nM) in milk set by the US Food and Drug Administration [43]. Additionally, compared with most other methods reported previously, such as colorimetric, fluorescence, electrochemistry, and electrochemiluminescence (Table 1), the PEC aptasensor displayed excellent performance with a relatively lower detection limit and convenient operation.  Various interfering substances, such as OTC, KAN, TOB, and CIP, were chosen to evaluate the selectivity of the PEC aptasensor. As shown in Figure 3C, the photocurrent differences (∆I) of TC (10 nM) were different to the photocurrent differences of the four other interferents (10 nM), reflecting a high TC selectivity of the PEC sensor. Furthermore, the PEC aptasensor showed a wonderful stability under the eight on-off irradiation cycles Toxics 2023, 11, 17 7 of 10 ( Figure 3D). Additionally, the long-term stability of the designed PEC aptasensor was determined. The results showed that when the PEC aptasensor was kept at 4 • C for 16 days, the current of the prepared electrode had no obvious change ( Figure S3). These results demonstrate that the PEC aptasensor has favorable potential for TC detection.

Real Samples Analysis
A standard addition method was used to confirm the analytical feasibility of the PEC aptasensor in real samples. Table 2 exhibited that the recoveries of the PEC aptasensor ranged from 98.23-98.92%, and the RSD was 0.54-1.48%, which was essentially consistent with the results obtained by HPLC. Therefore, the fabricated PEC aptasensor has excellent application prospects.

Conclusions
In this work, we successfully fabricated a "signal on" PEC aptasensor based on the g-C 3 N 4 /BiVO 4 for the highly stable determination of TC. The g-C 3 N 4 /BiVO 4 acted as a photoactive material, which provided a stable and strong photocurrent. Moreover, TC aptamer probes were stably assembled on the surface of g-C 3 N 4 /BiVO 4 via covalent bonding interaction to obtain a more reliable signal response. Due to the oxidation of the aptamer-TC complex by the accumulated holes on the g-C 3 N 4 /BiVO 4 , the "signal on" PEC aptasensor displayed satisfactory stability and acceptable satisfactory detectability. Under the optimal conditions, the PEC aptasensor for TC detection can manifest a linear range of TC from 0.1 to 500 nM with a detection limit of 0.06 nM. Obviously, the proposed sensor presents a trend of accurate detection capability and early warning function and has the potential to control the environmental and health risks of such pollutants. Additionally, we hope to promote this detection method in the future with actual samples.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/toxics11010017/s1, Real samples preparation; Validation by HPLC; Figure S1: (A) UV/vis diffuse reflectance spectrum and (B) Miro-FTIR spectra of the prepared materials; Figure S2: The effect of PEC performance on (A) Proportion of composites, (B) Incubation concentration of aptamer probes; Figure S3: The long-term stability of the developed PEC aptasensor.