Morphology Effect of Bismuth Vanadate on Electrochemical Sensing for the Detection of Paracetamol

Morphology-control, as a promising and effective strategy, is widely implemented to change surface atomic active sites and thus enhance the intrinsic electrocatalytic activity and selectivity. As a typical n-type semiconductor, a series of bismuth vanadate samples with tunable morphologies of clavate, fusiform, flowered, bulky, and nanoparticles were prepared to investigate the morphology effect. Among all the synthesized samples, the clavate shaped BiVO4 with high index facets of (112), (301), and (200) exhibited reduced extrinsic pseudocapacitance and enhanced redox response, which is beneficial for tackling the sluggish voltammetric response of the traditional nanoparticle on the electrode surface. Benefiting from the large surface-active area and favorable ion diffusion channels, the clavate shaped BiVO4 exhibited the best electrochemical sensing performance for paracetamol with a linear response in the range of 0.5–100 µmol and a low detection limit of 0.2 µmol. The enhanced electrochemical detection of paracetamol by bismuth vanadate nanomaterials with controllable shapes indicates their potential for applications as electrochemical sensors.


Introduction
Drug detection is essential in the monitoring of drug molecules in bio-fluids and plays an important role in drug quality control [1,2]. Paracetamol, also known as acetaminophen, is one of the most popular analgesics/antipyretics and has been applied in effective treatment of pain and fever in adults and children [3,4]. Paracetamol distributes rapidly after oral administration and is easily excreted in the urine. Unlike other analgesic drugs, paracetamol does not produce gastrointestinal damage or untoward cardiorenal effects [5,6]. However, the hypersensitivity or overdose of paracetamol can lead to formation of some liver and nephrotoxic metabolites, such as acute liver necrosis [7]. Moreover, the hydrolytic degradation product of paracetamol is 4-amino-phenol that can be found in pharmaceutical preparations and can cause teratogenic effect and nephrotoxicity [8].
It is desirable to develop an efficient electrochemical catalyst for paracetamol for the quality control of pharmaceuticals, physiological function, and diagnosis in clinical medicine [9]. Semiconductors have been taken as effective photocatalytic and electrochemical sensors for direct detection of paracetamol [10][11][12][13]. Therein, transition metal oxide BiVO 4 , with an excellent charge transport property (hole diffusion length L p = 70 nm) [14,15], has emerged as a highly promising electrocatalytic material with good chemical stability, environmental inertness, and low cost [16,17]. Medeiros et al. reported that BiVO 4 nanoparticles could be used as a highly efficient and sensitive photoelectrochemical sensor for paracetamol detection [18]. Generally speaking, morphology optimization can further enhance the electrocatalytic performance of material oxides. Control of the size and shape of material oxides is essential to optimize their active areas and favorable ion diffusion channels [19]. As a result, many efforts have been made to engineer metal oxides on the nanoscale that have led to the understanding of their fundamental size-and shape-dependent properties [20]. For example, porous BiVO 4 with a larger surface area and more reactive sites compared with nanoparticle shape could afford a faster electron transfer rate, as well as higher stability and reproducibility of the sensor [21]. The ability to control the particle morphology can provide a means to tune so-called structure-sensitive catalytic reactions [22,23]. It is highly desirable to be able to synthesize electrocatalytic materials with different morphologies in a facile and controllable manner for the purpose of improving the sensitivity of paracetamol detection, as well as investigating the morphology effect on the electrochemical sensing.
In this work, the BiVO 4 electrodes with different morphologies were prepared for the purpose of investigating the influence of morphology on sensitivity of electrochemical sensing. The BiVO 4 samples were synthesized by the microwave approach at 180 • C for 30 min. The addition of additives and the adjustment of the pH value of solutions could change the morphology of the BiVO 4 samples effectively. Further, the influence of different morphology of BiVO 4 on the electrochemical detection of paracetamol was explored. Among all synthesized BiVO 4 samples, the clavate morphological of the BiVO 4 electrode exhibited the best electrochemical sensing performance on paracetamol with the widest linear detection range (0.5-100 µM) and lowest detection limit (0.2 µM).

Apparatus
A microwave chemistry working platform (Model: TOPEX+, PreeKem Scientific Instruments Co., Ltd., Shanghai, China) and high-performance liquid chromatography (Model: Essentia CTO-16L, Shimadzu, Shanghai, China) with a column of 2.1 mm × 10 cm (Model: ZORBAX SB-C18, Agilent, CA, USA) were used. Scanning electron microscopy (SEM) images were obtained by Apreo (Thermo Scientific, Waltham, MA, USA). Field emission transmission electron microscopy (TEM) images were recorded using a TecnaiTM G2 F30 (FEI Co. Ltd., Hillsboro, OA, USA). X-ray diffraction (XRD) characterization was carried out on a D2 PHASER (Bruker, Karlsruhe, Germany) with Cu-Kα as the radiation source (λ = 0.154 nm). X-ray photoelectron spectroscopy (XPS, K-Alpha + , Thermo Scientific, MA, USA) was used. The C1s binding energy of adventitious carbon contamination with 284.6 eV was selected as the reference. UV-Vis diffuse reflectance spectra were recorded with a UV-Vis spectrophotometer (Model: Frontier, PerkinElmer Inc., MA, USA). All the electrochemical experiments were carried out using an electrochemical workstation (CH Instrument 660E, Shanghai Chenhua Instrument Co., Ltd., Shanghai, China). The electrochemical experiment was performed using a conventional three-electrode system with the prepared BiVO 4 as the working electrode, a graphite rod as the counter electrode, and a saturated Ag/AgCl electrode as the reference electrode.

Preparation of Bismuth Vanadate
The amounts of 5 mM Bi(NO 3 ) 3 and 5 mM NH 4 VO 3 were dissolved in 10 mL ultrapure water. Different amounts of NaH 2 PO 4 , Na 2 HPO 4 , Na 3 PO 4 , and Na 2 CO 3 were used as additives. The amount of 1.0 M NaOH and 36% HCl solutions were used to adjust the pH values of solutions. Then 20 mL amounts of prepared solutions with different pH values were transferred to 100 mL Teflon reactors. The microwave reaction was carried out at 180 • C for 30 min. Different morphologies of BiVO 4 powders were collected by centrifugation at 12,000 rpm and dried in an oven at 80 • C for 24 h. The BiVO 4 electrodes with different morphologies were prepared by the spin coating method; 5 mg BiVO 4 powder was dispersed in 1 mL DI water by ultrasound for 10 min. Then the BiVO 4 suspension was transferred to FTO substrates by spin coating. The as-prepared BiVO 4 /FTO electrodes were annealed in a muffle furnace with 200 • C for 2 h for stabilization. Finally, the BiVO 4 /FTO electrodes with different morphologies were obtained.

Determination of Paracetamol in Tablets
High performance liquid chromatography (HPLC) was used for the estimation of the content of paracetamol in standard drug of paracetamol tablets. The method was carried out on a Hichrom C18 (25 cm × 4.6 mm i.d., 5 µm) column with a mobile phase consisting of methanol and icy ultra-pure water containing formic acid (volume ratio of 50%/49.9%/0.1%) at a flow rate of 0.2 mL min −1 . Detection was carried out at 257 nm. Standard stock solutions of 0, 10, 20, 50, and 100 µmol of paracetamol were prepared in PBS (pH = 7.4), respectively. The injection volume of solution was 50 µL.

Characterization of BiVO 4 with Different Morphologies
The synthesis procedures with different conditions to obtain different morphologies of BiVO 4 are summarized in Figure S1. The morphology of bismuth vanadate could be controlled with different additives, including NaH 2 PO 4 , Na 2 HPO 4 , Na 3 PO 4 , and Na 2 CO 3 .
The pH values of the solutions were adjusted in the range of 2 to 10 with HCl or NaOH. Due to the differences in ionization and hydrolysis of phosphate, carbonate, and alkaline, the morphology of bismuth vanadate could form nanosheet, tetrahedron, cuboid, sphere, or irregular shapes. In this article, several representative morphologies of bismuth vanadate were selected, which were denoted as clavate, fusiform, flowered, bulky, and particle BiVO 4 , as illustrated in Figure 1. Certain bismuth vanadate samples were chosen as representatives to discuss the effect of morphology on electrochemical properties of BiVO 4 materials.    The particle BiVO 4 was obtained by adding Na 3 PO 4 to the precursor solution ( Figure 1E). Then HCl was added drop by drop to adjust the pH value to 4. The Bi 3+ ions and VO 3 − ions combined to form nanoparticles. The size of BiVO 4 nanoparticles ranged from 30 to 300 nm. The clavate and fusiform topography were obtained by adding Na 2 HPO 4 ( Figure 1A,B). The pH values of solutions were adjusted to 9.5 and 3.6, respectively. The pH value of 0.1 M Na 2 HPO 4 solution was about 9 (Ka 1 = 7.1 × 10 −3 ). Due to the similar pH values of solution, the clavate morphology BiVO 4 was formed directly and rapidly. By contrast, the fusiform BiVO 4 was formed by the aggregation of nanoparticles ( Figure S2), which was due to the partial dissolution and structural reorganization during the process of adjusting the pH value of the solution. The flowered BiVO 4 was formed with the addition of NaH 2 PO 4 . NaH 2 PO 4 solution is acidic (Ka 1 = 6.2 × 10 −8 ). NaOH was added slowly to change the pH value of solution to 7.7. The BiVO 4 gradually assumed a cross-linked flowered-sphere structure with a diameter of ca. 5 µm ( Figure 1C). The bulky topography of BiVO 4 could be obtained by adding NaHCO 3 and adjusting the pH value to 6 ( Figure 1D). The bulky BiVO 4 showed a cube structure with a hole in the center.
Due to the different hydrolysis and ionization rates of carbonate and phosphate, the pH values of the solution were different, which further affected the nucleation rate of the Bi + in solution. A further change in the pH of the solution would lead to the dissolution or reshaping of BiVO 4 samples. As a result, the morphology control of BiVO 4 could be achieved with different additives and pH values.
The selected BiVO 4 samples were characterized with a powder X-ray diffractometer ( Table 1. The surface chemical states of selected BiVO 4 samples were further investigated by XPS. As for the O 1s spectrum ( Figure 2F), the peaks around 529.2 eV were clearly shown in particle, bulky, and flowered BiVO 4 samples, which could be attributed to the lattice oxygen (O 2− ) in BiVO 4 [24]. High-resolution XPS spectra of O 1s after peak fitting were shown in Figure S3. The peak at 529.2 eV was not observed in clavate and fusiform BiVO 4 samples, while a new peak at 530.4 eV appeared, which could be ascribed to the lattice oxygen in bismuth oxide, suggesting the surface of clavate and fusiform BiVO 4 were oxidized. The existence of the surface oxide layer in clavate and fusiform BiVO 4 samples was further confirmed by high-resolution XPS spectra of Bi 4f ( Figure 2E) and V 2p ( Figure 2D). The Bi 4f of particle, bulky, and flowered BiVO 4 samples showed two characteristic peaks at 164.8 eV and 159.1 eV that were attributed to Bi 4f 5/2 and Bi 4f 7/2 , respectively. The Bi 4f of clavate and fusiform BiVO 4 samples showed an apparent shift to high binding energy, attributed to the bismuth oxide. A similar phenomenon could be observed in the V 2p spectra. The V 2p peaks were not shown in clavate and fusiform BiVO 4 samples due to the surface oxide layer that covered the single V orbit. XRD results show that the clavate and fusiform BiVO 4 , which were both synthesized with Na 2 HPO 4 , were much more unstable. The VO 4 unit in BiVO 4 easily formed a new bismuth oxide unit as well as oxygen vacancies on the surface of BiVO 4 . Nanomaterials 2022, 12, 1173 5 of 12 V 2p spectra. The V 2p peaks were not shown in clavate and fusiform BiVO4 samples due to the surface oxide layer that covered the single V orbit. XRD results show that the clavate and fusiform BiVO4, which were both synthesized with Na2HPO4, were much more unstable. The VO4 unit in BiVO4 easily formed a new bismuth oxide unit as well as oxygen vacancies on the surface of BiVO4.

Optical Analysis of BiVO4 Samples
The BiVO4 powders obtained from different precursors showed different colors (the inset of Figure 2B). The flowered, bulky, and particle BiVO4 showed yellow color with the band edge of optical absorption at around 550 nm. The clavate and fusiform BiVO4 exhibited a white color with a shift of optical absorption edge to about 400 nm ( Figure 2B). The change of the color was due to the surface oxide layer of BiVO4. The UV-vis DRS data were combined with the Kubelka-Munk (K-M) relation to study the association of diffused reflectance with the absorption coefficient: F(R) = (1 − R) 2 /2R, where F(R) is the Kubelka-Munk function, and R is the absolute reflectance of the sample. The optical band gap of the prepared samples is calculated using Tauc's equation: F(R) hυ = A(hυ − Eg) n , where n = 2 for a directly allowed transition, and n = 1/2 for an indirectly allowed

Optical Analysis of BiVO 4 Samples
The BiVO 4 powders obtained from different precursors showed different colors (the inset of Figure 2B). The flowered, bulky, and particle BiVO 4 showed yellow color with the band edge of optical absorption at around 550 nm. The clavate and fusiform BiVO 4 exhibited a white color with a shift of optical absorption edge to about 400 nm ( Figure 2B). The change of the color was due to the surface oxide layer of BiVO 4 . The UV-vis DRS data were combined with the Kubelka-Munk (K-M) relation to study the association of diffused reflectance with the absorption coefficient: F(R) = (1 − R) 2 /2R, where F(R) is the Kubelka-Munk function, and R is the absolute reflectance of the sample. The optical band gap of the prepared samples is calculated using Tauc's equation: F(R) hυ = A(hυ − E g ) n , where n = 2 for a directly allowed transition, and n = 1/2 for an indirectly allowed transition, and A is a constant and hυ is photon energy [25]. According to the calculation, the band gap (E g ) of the selected BiVO 4 was summarized in Table 1. The band gap of intrinsic BiVO 4 is around 2.4 eV [26]. The particle, bulky, and flowered BiVO 4 showed a comparable value of E g with intrinsic BiVO 4 , while clavate and fusiform BiVO 4 had a larger value of E g due to the surface oxide layer. Generally speaking, BiVO 4 shows excellent photoelectrochemical performance due to its suitable band gap. To investigate the influence of the solar light in photo-assisted detection of paracetamol, differential pulse voltammetry (DPV) of clavate BiVO 4 on determination of paracetamol in the dark and under illumination was studied.
As displayed in Figure S4, the photoresponse current on BiVO 4 electrode increased from 17.5 to 21.0 mA cm −2 , which indicates that solar light has a positive effect on improving the sensitivity of paracetamol detection.

Electrochemical Response at Various BiVO 4 Electrodes
Electrochemical techniques have been widely explored in the detection of paracetamol in biological fluids and tablets due to their simple pretreatment procedure, high sensitivity, low time of analysis, and low costs over other analytical methods [27,28]. To understand the effect of morphology control on electrochemical performance, different morphological BiVO 4 were examined in 0.01 M PBS with and without 100 µM paracetamol using CV. All the electrochemical experiments were carried out under dark conditions in order to exclude the influence of other factors. Figure 3A-E shows the CV response of the paracetamol re-dox process on different BiVO 4 electrodes. The CV curves of all the BiVO 4 electrodes displayed a strong anodic peak (E pa ) at 0.48 V. The different peaks at BiVO 4 electrodes could be observed in the process of backward scan with two small cathodic peaks (E pc ), which were registered at 0.4 and 0.1 V, respectively. The chemical reaction was coupled to the electrochemical product for the oxidation of acetaminophen, i.e., N-acetyl-p-benzoquinoneimine (NAPQI) [29]. The competition between two forms of NAPQI (protonated and unprotonated species) has been proposed by Kissinger et al. [30]. The anodic peak currents in CV curves were higher than the cathodic ones, where the most probable coupled chemical reaction was the hydration of the NAPQI molecule, leading to a lower cathodic current. The limited cathodic current illustrates that the reduction of the radical intermediate is controlled by dynamics. In addition, the low cathodic current suggests that the oxidation of paracetamol on the surface of the BiVO 4 electrode is more efficient than its reduction.  In order to investigate the influence of morphology on the electrochemical performance, the compared CV results of BiVO4 with clavate, fusiform, flowered, and bulky, as well as the reference morphology of particle shape are shown in Figure 3A-E. The reference morphology of particle BiVO4 exhibited high electrochemical performance of 75 µ A cm 2 with 100 µ M paracetamol when the applied potential was 0.5 V. However, the particle BiVO4 showed large capacitance and a sluggish voltammetric response. The In order to investigate the influence of morphology on the electrochemical performance, the compared CV results of BiVO 4 with clavate, fusiform, flowered, and bulky, as well as the reference morphology of particle shape are shown in Figure 3A-E. The reference morphology of particle BiVO 4 exhibited high electrochemical performance of 75 µA cm 2 with 100 µM paracetamol when the applied potential was 0.5 V. However, the particle BiVO 4 showed large capacitance and a sluggish voltammetric response. The pseudocapacitance could affect the redox peak during the voltammetric response, and the extrinsic pseudocapacitance arose at the electrode surfaces along with the gradual BiVO 4 nanonization [31]. The bulky and flower BiVO 4 exhibited low current density due to their low specific surface area and poor crystallinity, respectively ( Figure 3C,D). It is noted that the fusiform BiVO 4 with orthorhombic structure and clavate BiVO 4 with tetragonal structure were both covered by bismuth oxides. However, the fusiform BiVO 4 ( Figure 3B) showed weaker catalytical performance compared with that of clavate BiVO 4 , illustrating the tetragonal structure is beneficial for catalytic performance. Compare to BiVO 4 with other morphologies, the clavate BiVO 4 ( Figure 3A) exhibited the highest electrochemical performance with a maximum redox peak current of 100 mA cm −2 in the 0.1 M PBS solution containing 100 µM paracetamol, owing to the large specific surface area and surface oxygen vacancies, which could effectively improve the charge transfer characteristics and increase the charge diffusion coefficient of BiVO 4 . Moreover, the clavate BiVO 4 could afford favorable channels for ion diffusion, which further increased the contact between the carriers and the drug molecules.
In order to suppress the influence of charging current and obtain higher sensitivity, the DPV was introduced into electrochemical analysis. The DPV experimental parameters were optimized at 50 ms pulse width, 50 mV pulse amplitude. The DPV curves of different BiVO 4 samples in different concentrations of paracetamol solutions are shown in Figure S5, and the electrochemical performances on morphology-dependent BiVO 4 electrodes in 0.01 M PBS without and with 100 µM paracetamol from DPV results are summarized in Figure 3F. Accordingly, clavate BiVO 4 with high crystallinity showing the highest electrochemical signal was chosen to systematically investigate its paracetamol sensing performance.

Structural Characterization of Clavate BiVO 4
TEM characterizations were performed on the clavate BiVO 4 with the length of several micrometers ( Figure 4A,B). Figure 4D surface could show good electrocatalytic performance. The high index lanes refer to a facet where one of the indexes is greater than 1 in (h, k, l). The lattice fringes with an interlayer spacing of 0.315 nm (white color) corresponded to the (111) crystal plane of BiO2, suggesting oxidization at the surface of BiVO4. The EDS data in Figure 4C depict the uniform distribution of Bi, V, and O elements. The TEM characterization proved the tetragonal structure of clavate BiVO4, and the surface of BiVO4 was partially covered by BiO2 nanoparticles.

Electrochemical Investigation of Paracetamol on Clavate BiVO4
The effect of the clavate BiVO4 electrode on the electrochemical detection of paracetamol was further investigated. Cyclic voltammograms at the clavate BiVO4 electrode in 100 μM paracetamol with scan rates of 1 to 480 mV s −1 are shown in Figure  5A. The anodic peak potentials shifted with the increase of scan rate, and the anodic peak currents presented a linear dependence on the scan rate ( Figure 5B). A linear regression equation was adopted for the anodic peak, logIpa = 0.421 logυ + 0.905, giving the correlation coefficient values of R 2 = 0.999. It demonstrated that oxidation of paracetamol is an irreversible redox process with diffusion-controlled mass transport. The regression equation was obtained as Epa = 0.0493 logυ + 0.404 (R 2 = 0.996, Figure 5C). The number of

Electrochemical Investigation of Paracetamol on Clavate BiVO 4
The effect of the clavate BiVO 4 electrode on the electrochemical detection of paracetamol was further investigated. Cyclic voltammograms at the clavate BiVO 4 electrode in 100 µM paracetamol with scan rates of 1 to 480 mV s −1 are shown in Figure 5A. The anodic peak potentials shifted with the increase of scan rate, and the anodic peak currents presented a linear dependence on the scan rate ( Figure 5B). A linear regression equation was adopted for the anodic peak, logI pa = 0.421 logυ + 0.905, giving the correlation coefficient values of R 2 = 0.999. It demonstrated that oxidation of paracetamol is an irreversible redox process with diffusion-controlled mass transport. The regression equation was obtained as E pa = 0.0493 logυ + 0.404 (R 2 = 0.996, Figure 5C). The number of electrons involved in the reaction and the charge transfer coefficient could be calculated as 2 and 0.5, respectively. These results indicate that two electrons are involved in the electrochemical redox process of paracetamol. The paracetamol exhibited sluggish voltammetric response at the traditional electrode surface, which restricted the sensitivity of the electrochemical sensor. To solve this problem, DPV was used to improve the detection sensitivity. DPV curves of the clavate BiVO 4 electrode for paracetamol determination are shown in Figure 5D. The oxidation DPV peak of paracetamol was observed at about +0.35 V on the clavate BiVO 4 electrode. A linear correlation between the current density and the paracetamol concentration was obtained in the range from 5 × 10 −7 to 1 × 10 −5 M, which could be represented by a regression equation as follows: I = 4.65 × 10 −2 c + 5.51 (R 2 = 0.997). The detection limit of the sensor was calculated as 2 × 10 −7 M (S/N = 3). The obtained DPV data revealed that the clavate BiVO 4 electrode had more competitive analytical performance and a much lower detection limit compared with BiVO 4 electrodes of other morphologies ( Figure S4). The comparisons between the clavate BiVO 4 electrode and some reported electrodes for paracetamol determination are summarized in Table S1. The clavate BiVO 4 electrode exhibited analytical performances with acceptable sensitivity and a wide linear range. To evaluate the selectivity and stability of the clavate BiVO 4 electrode, metal interference ions such as K + , Na + , Ni + , Zn 2+ , Co 2+ , Mg 2+ , Cd 2+ , Fe 2+ , Pd 2+ , Al 3+ , and Fe 3+ were each added into a standard solution containing 100 µM paracetamol. As shown in Figure 6A, the addition of 0.10 M metal ion species did not affect the DPV current response of paracetamol on the clavate BiVO 4 electrode. The result illustrates that the clavate BiVO 4 electrode has excellent selectivity, even in the presence of a 1000-fold concentration of interference species.
BiVO4 electrode and some reported electrodes for paracetamol determination are summarized in Table S1. The clavate BiVO4 electrode exhibited analytical performances with acceptable sensitivity and a wide linear range. To evaluate the selectivity and stability of the clavate BiVO4 electrode, metal interference ions such as K + , Na + , Ni + , Zn 2+ , Co 2+ , Mg 2+ , Cd 2+ , Fe 2+ , Pd 2+ , Al 3+ , and Fe 3+ were each added into a standard solution containing 100 µ M paracetamol. As shown in Figure 6A, the addition of 0.10 M metal ion species did not affect the DPV current response of paracetamol on the clavate BiVO4 electrode. The result illustrates that the clavate BiVO4 electrode has excellent selectivity, even in the presence of a 1000-fold concentration of interference species.  The mechanism of the electrochemical oxidation and reduction of paracetamol is shown in Figure 6B. The applied voltage promotes the separation of electrons and holes in BiVO4. The oxidation peak of paracetamol could vary with the pH of the solution. Paracetamol converts to intermediate NAPQI easily when the pH of solution is 7.4. NAPQI can stably exist in the solution in a deprotonated form. The CV of paracetamol shows an oxidation peak and a relatively weak reduction peak. With the progress of the reaction, NAPQI gradually transforms into benzoquinone through other intermediates. The reduction peak of benzoquinone could be observed in CV.

Determination of Paracetamol in Pharmaceutical Samples
The analytical applicability of the clavate BiVO4 electrode was tested by determining the concentration of paracetamol in compound paracetamol tablets (Ⅱ) (250 mg per pill). One tablet was dissolved in 0.01 M PBS solution (pH = 7.4). The concentration of paracetamol in the measured solution with a calculated concentration of 45.0 µ M was detected by HPLC and DPV methods, respectively. Compared with the HPLC results, the values obtained by the DPV method exhibited high credibility. Moreover, the DPV method showed good reliability and repeatability during three times tests. The recoveries The mechanism of the electrochemical oxidation and reduction of paracetamol is shown in Figure 6B. The applied voltage promotes the separation of electrons and holes in BiVO 4 . The oxidation peak of paracetamol could vary with the pH of the solution. Paracetamol converts to intermediate NAPQI easily when the pH of solution is 7.4. NAPQI can stably exist in the solution in a deprotonated form. The CV of paracetamol shows an oxidation peak and a relatively weak reduction peak. With the progress of the reaction, NAPQI gradually transforms into benzoquinone through other intermediates. The reduction peak of benzoquinone could be observed in CV.

Determination of Paracetamol in Pharmaceutical Samples
The analytical applicability of the clavate BiVO 4 electrode was tested by determining the concentration of paracetamol in compound paracetamol tablets (II) (250 mg per pill). One tablet was dissolved in 0.01 M PBS solution (pH = 7.4). The concentration of paracetamol in the measured solution with a calculated concentration of 45.0 µM was detected by HPLC and DPV methods, respectively. Compared with the HPLC results, the values obtained by the DPV method exhibited high credibility. Moreover, the DPV method showed good reliability and repeatability during three times tests. The recoveries with 96.1-101.9% are given in Table 2, indicating that the fabricated clavate BiVO 4 sensor is accurate and sensitive enough for detecting paracetamol in pharmaceutical tablets. The above results show that the DPV method is a simple and reliable method to detect the content of paracetamol in drugs. The comparison between our work and some reported sensors for paracetamol determination were shown in Table S1. The clavate BiVO 4 electrode shows good stability and reliability in electrochemical detection of paracetamol.

Conclusions
In summary, the ability to control the morphology of BiVO 4 has great development prospects in high sensitivity electrochemical analysis. The BiVO 4 samples with different morphologies were obtained by adding different additives, and changing the pH values of the solution. The electrochemical properties of different morphology BiVO 4 were systematically studied in this article. The clavate BiVO 4 with high index lanes at the surface could solve the problem of the sluggish voltammetric response of the traditional nanoparticle on the electrode surface. The clavate BiVO 4 also could afford favorable channels for ion diffusion, increasing the contact between the carriers and the drug molecules. Therefore, the clavate BiVO 4 with a tetragonal structure exhibited the highest sensitivity and lowest detection limit among all selected BiVO 4 electrodes. In the DPV mode, the clavate BiVO 4 electrode showed linear responses over the concentration range of 0.5-100 µM (R 2 = 0.998) for paracetamol detection, and the LOD value was found to be 0.2 µM. The study of the relationship between morphology and electrocatalytic performance could provide very important information on the reaction activity and selectivity on target.