*2.4. Photocatalytic Experiment*

Acetaminophen were used as the target degradation contaminants to evaluate the photocataytic activity of the prepared catalysts. Before illumination, the solution containing 150 mL of 10 mg L−<sup>1</sup> acetaminophen and 0.15 g photocatalysts was sonicated and stirred for 30 min in dark to ensure an adsorption/desorption equilibrium. Then the above suspension was irradiated by 300 W Xe arc lamp (PLS-SXE 300, Perfectlight Co. Ltd, Beijing, China) with a UV-cutoff filter (λ > 400 nm). At a given time interval of irradiation, 2 mL aliquots were collected from the suspension and centrifuged. The residual concentration of organics in the aliquots was measured by a TU-1901 UV-vis spectrophotometer (Purkinje General Instrument Co., Ltd., Beijing, China). The concentration of acetaminophen was determined by high performance liquid chromatography (LC-20AT, Shimadzu, Kyoto, Japan) with an Agela Venusil MP C18 (0.46 μm × 250 mm) reverse-phase column equipped with UV-Vis detector (SPD-20A, Shimadzu, Kyoto, Japan) at 254 nm. The mobile phase was methanol: water (35:65, v/v) and the flow rate was 0.8 mL min−1. All the photocatalytic experiments were performed in triplicates and the mean values are reported.

#### **3. Results and Discussion**

#### *3.1. Characterization of the BiVO4*/*RGO Composites*

BiVO4/RGO composites were synthesized employing covalent chemistry to achieve BiVO4 samples in situ growth on graphene surface. SEM and AFM images were taken to characterize the microscopic morphology and structure. Figure 1c and Figure S1 display the prepared BiVO4 nanosheet exhibiting two-dimensional sheet-like morphology with 500–1000 nm in width and 8–30 nm in thickness. The products layer on the RGO sheet platform, which forms the homogenous 2D-2D interfacial contact. Figure 1b indicated that the as-prepared BiVO4 nanotubes had a typical nanotubular structure and paved well on the RGO sheet to form 1D-2D heterostructures. During the synthesis, the formation of the 1D and 2D BiVO4 samples was controlled by the pH value of the solution. It is reported that an increase of the pH value will decelerate the nucleation kinetics and provide a more suitable condition for anisotropic growth of 2D or 1D m-BiVO4 nanostructures [21]. Similarly, the SEM images of BiVO4 nanoparticles/RGO (0D-2D) exhibit the well-dispersed BiVO4 nanoparticles on the RGO sheet in Figure 1a.

**Figure 1.** SEM images of BiVO4/ reduced graphene oxide (RGO): (**a**) BiVO4 nanoparticles/RGO; (**b**) BiVO4 nanotubes/RGO; (**c**) BiVO4 nanosheets/RGO.

The RGO content of the above three composites are determined at ~2 wt% by TGA test in Figure S2. According to the literature, the amount of reduced graphene oxide was determined by the weight loss from 200 to 600 ◦C [23].

The phase and crystal structure of the as-prepared samples were examined by XRD. As shown in Figure 2a, the XRD patterns of BiVO4 and BiVO4/RGO samples all agree with the monoclinic scheelite type BiVO4 (JCPDS No.14-0688). Compared with BiVO4 nanoparticles and nanotubes samples, the dominant 004 diffraction peak suggests that BiVO4 nanosheets and BiVO4 nanosheets/RGO have a preferred orientation along the (001) planes [24]. Notably, for the sample GO of Figure S3, the peak at 2θ of 10.3◦ is attributed to the (002) reflection of stacked GO sheets. However, no diffraction peak of GO is observed in the composites, attributing to the disappearance of layer-stacking regularity after redox of graphite [8].

**Figure 2.** (**a**) XRD patterns BiVO4 and BiVO4/ RGO composites. (**b**) Raman spectra of graphene oxide (GO) and BiVO4/RGO composites in the 1200–1800 cm<sup>−</sup><sup>1</sup> region.

The monoclinic scheelite phase of BiVO4 in the composites is further confirmed by a typical Raman band at 126, 210, 325, 367, 710 and 828 cm<sup>−</sup><sup>1</sup> in Figure S4 [25]. In addition, the D band centered at 1350 cm<sup>−</sup><sup>1</sup> (disorder band) and the G band at 1580 cm<sup>−</sup><sup>1</sup> (tangential vibration band) are present, indicating that RGO has been successfully loaded on the surface of BiVO4 [26,27]. Furthermore, the *ID*/*IG* ratio is inversely proportional to the average size of the sp2-hybridized graphene domains [28]. As is shown in Figure 2b, after the solvothermal process, the *ID*/*IG* ratio of BiVO4 nanosheets/RGO decreased from 1.04 to 0.99, indicating that the reduction of GO increased the average size of the graphene domains and reduced the defect density in the composite [29]. Differently, an increase in the

*ID*/*IG* ratio of BiVO4 nanoparticles/RGO and BiVO4 nanotubes/RGO are observed. The result shows that more numerous sp<sup>2</sup> domains have been formed in the composites [28,30].

The chemical state of BiVO4 nanosheets and RGO in the composites is investigated by XPS in Figure 3. The survey spectrum in Figure 3a indicated the existence of Bi, V, O and C elements in BiVO4 nanosheets/RGO. In the C1s spectrum of BiVO4 nanosheets/RGO (Figure 3b), the peak centered at 283.6 eV binding energy indicates the existence of C–C bond from graphene. The peak located at 284.2 eV is attributed to the formation of C–O bond, suggesting the combination of BiVO4 nanosheets and RGO. The weak O=C–O bond centered at 287.1 eV indicates that the oxygenated functional groups of reduced graphene oxide are weakened during the hydrothermal reaction, along with the reduction of GO to RGO [31]. In Figure 3c, the two peaks at 158.87 and 164.47 eV are attributed to the orbital of Bi 4f7/2 and Bi 4f5/2, indicating the existence of Bi3<sup>+</sup> in the pure BiVO4 nanosheets. However, in the BiVO4 nanosheets/RGO composites, the binding energy of Bi 4f7/2 is blue shifted to lower values by 0.5 eV, owing to the change of the chemical environment surrounding Bi element under the influence of RGO. The same results occurred in V 2p in Figure 3d. The observations in the XPS spectra further demonstrates the intense interaction between BiVO4 nanosheets and RGO.

**Figure 3.** (**a**) The XPS survey spectrum of BiVO4 nanosheets/RGO and (**b**) C 1s band of BiVO4 nanosheets/RGO, (**c**) Bi 4f band of BiVO4 nanosheets and BiVO4 nanosheets/RGO. (**d**) V 2p band of BiVO4 nanosheets and BiVO4 nanosheets/RGO.

The optical properties of the composites were analyzed by DRS. The UV-vis di ffuse reflectance spectra can be used to determine the absorption edge information and the width of the forbidden band of catalysts. As shown in Figure 4a, the introduction of graphene enhanced the absorbance in the visible light region for the as-prepared BiVO4/RGO, especially for BiVO4 nanosheets/RGO. The optical band gap energy (*E*g) can be determined from the plot between *E* = <sup>1240</sup>/λAbsorp.Edge and [*F(R* ∞*)h*υ] 1/2 [32]. As shown in Figure 4b, the band gap energy of BiVO4 nanosheets/RGO is 2.45 eV, which is lower than that of pure BiVO4 nanosheets (0.21eV). These results demonstrate that the introduction of graphene in BiVO4 nanosheets/RGO nanocomposites can directly produce more excited charge transfer under visible-light irradiation, which is the premise of excellent photocatalytic performance.

**Figure 4.** (**a**) UV-vis diffuse reflectance spectra of BiVO4 and BiVO4/RGO composites. (**b**) The relative band gap energy of the prepared samples.

#### *3.2. Photocatalytic Performance of BiVO4*/*RGO Samples*

As a common analgesic and antipyretic drug, acetaminophen is heavily used all over the world and detected in surface water, ground water and sewage effluents [33,34]. Once acetaminophen has overdosed, it may cause potential liver damage and even death [35,36]. Thus, it is particularly urgen<sup>t</sup> to provide an efficient method to enhance the degradation of acetaminophen in wastewater.

Hence, acetaminophen was used as a target pollutant to evaluate the photocatalytic properties of the prepared materials under visible light irradiation (λ > 400 nm). The chromatogram corresponding to the acetaminophen standard sample is shown in Figure S5 with the retention time at about 5.6 min. For comparison, BiVO4 nanoparticles (0D), BiVO4 nanotubes (1D), BiVO4 nanosheets (2D), BiVO4 nanoparticles/RGO (0D/2D), BiVO4 nanotubes/RGO (1D/2D), and BiVO4 nanosheets/RGO (2D/2D) were all examined. Figure S6 displayed the adsorption of acetaminophen on the nanomaterials reached an equilibrium state within 30 min in the dark. In Figure 5a, the photolysis performance of acetaminophen under visible light irradiation without any photocatalyst indicates that the self-degradation of acetaminophen is negligible under the visible light irradiation. The photodegradation curves of acetaminophen were fitted by pseudo-first-order reaction kinetics. As is shown in Figure 5a and Table S1, the addition of RGO can obviously improve the visible light performance of BiVO4 photocatalysts, indicating the heterojunction structure between BiVO4 and RGO contributed remarkably to the photocatalytic degradation rate. In particular, the BiVO4 nanosheets/RGO composites (*k* = 0.0141 min−1) exhibited the optimal performance compared with the corresponding pure BiVO4 (*k* = 0.0080 min−1). This demonstrates that the 2D-2D heterojunction structure is more beneficial to the photocatalytic activity than the other dimensional heterojunctions. When the 2D flake BiVO4 and the thin slice of RGO are parallel to the space, it is beneficial to maintain high photoelectron transport efficiency and reduce the recombination of the electron hole, improving the catalytic efficiency [37].

The amphoteric behaviour of the solution influences the surface charge of the photocatalyst [38]. The role of pH on the photodegradation efficiency was studied in the pH range 3–11. As is shown in Figure S7, the photodegradation efficiency of BiVO4/RGO samples increases with the increasing of pH and the maximum rate was at pH 11. That may be ascribed to major contribution of electrostatic interaction on mass transfer rate [39]. It is considered that under alkaline conditions, there is a large quantity of OH− in the solution, which favours the formation of •OH. The strong oxidation of •OH plays an important role in the process of photocatalytic degradation [40]. Compared with BiVO4 nanotubes/RGO and BiVO4 nanoparticles/RGO, BiVO4 nanosheets/RGO showed more excellent catalytic performance under neutral conditions, indicating that the pH application range of flaky BiVO4/RGO was wider.

**Figure 5.** (**a**) Time-online photocatalytic performance of acetaminophen over the as prepared photocatalysts under visible light irradiation. (**b**) Stability experiments of BiVO4 nanosheets/RGO.

The stability of the BiVO4 nanosheets/RGO was evaluated by performing the recycling experiments. In Figure 5b, the photocatalysts can still maintain excellent degradation efficiency after four cycles. Figure 6a displays the smooth surface of BiVO4 nanosheet after visible light irradiation. Compared with the fresh sample, no obvious discrepancy in the XRD pattern of the recycled sample was observed in Figure 6b. As is shown in Figures 3b and 6c, the XPS spectra of the recycled BiVO4 nanosheets/RGO exhibited a slight decrease of the C−O and O=C–O peak intensity, possibly attributed to the further photocatalytic reduction of GO to RGO during the photocatalytic degradation [20]. However, the above changes did not affect the photocatalytic performance after recycling experiments. It is proved that the BiVO4 nanosheets/RGO composites prepared exhibit relatively high stability.

**Figure 6.** (**a**) SEM, (**b**) XRD and (**c**) XPS results of C 1s band of BiVO4 nanosheets/RGO after four cycles irradiation.

#### *3.3. Photocatalytic Mechanism of BiVO4*/*RGO*

It is commonly accepted that a series of reactive species, such as hole (h+), hydroxyl radical (•OH), and superoxide radial (O2•−), usually govern the photocatalytic degradation reactions of organic pollutants [41]. In order to investigate the main reactive species responsible for the photocatalytic activity of BiVO4/RGO samples, the radical trapping experiment was carried out. Figure 7 displays the photocatalytic degradation curves of acetaminophen over BiVO4 nanotubes/RGO, BiVO4 nanoparticles/RGO, and BiVO4 nanosheets/RGO with the addition of ROS scavengers. In this experiment, isopropanol (IPA) was used to quench •OH, formic acid for h<sup>+</sup>, and p-benzoquinone (BQ) for O2•−. For comparison, the radical trapping results of the pure BiVO4 samples was demonstrated in Figure S8.

As depicted in Figure 7b,c, acetaminophen degradation process was obviously depressed by IPA, verifying •OH plays the most important role in the photocatalytic reaction over BiVO4 nanosheets/RGO and BiVO4 nanotubes/RGO. In addition, when the scavenger for h+ was added into the photocatalytic solution in Figure 7c, the degradation of acetaminophen was also depressed. It illustrates that h+ was involved as minor radical species in the photocatalytic process of BiVO4 nanosheets/RGO. As for the system based on BiVO4 nanoparticles/RGO, the reaction process was a little different. The degradation rate of acetaminophen in Figure 7a showed a certain decrease in the presence of formic acid, indicating h+ is the key reactive species responsible for the photodegradation over BiVO4 nanoparticles/RGO. According to the results mentioned above, we can preliminarily conclude that for BiVO4 nanoparticles/RGO photocatalytic process, h+ is the main radical species. In the BiVO4 nanotubes/RGO reaction system, •OH plays an important role, while in the BiVO4 nanosheets/RGO based reaction system, •OH and h+ as the major and minor radical species are all produced and participated in the acetaminophen degradation process.

**Figure 7.** Free radical inhibition experiment of BiVO4/RGO: (**a**) BiVO4 nanoparticles/ RGO; (**b**) BiVO4 nanotubes/RGO; (**c**) BiVO4 nanosheets/RGO.

Different from the BiVO4/RGO photocatalytic systems, pure BiVO4 samples take on the similar characteristics in Figure S8. For the pure BiVO4 nanoparticles, nanotubes and nanosheets, h+ is the main radical species participating the photocatalytic processes.

The involved ROS is further confirmed by ESR experiments under visible light. As is shown in Figure 8b, after the catalysts were exposed to visible light for 10 min, the characteristic peaks of h+ increased obviously for both BiVO4 nanoparticles/RGO and BiVO4 nanosheets/RGO than in dark condition, and the signals for BiVO4 nanotubes/RGO could nearly be ignored. Notably, the peak intensity of h+ referring to the BiVO4 nanosheets/RGO was much higher than that of BiVO4 nanoparticles/RGO. The result validated that the generation amount of h+ of this 2D-2D system was more than other nanocomposites. Moreover, from Figure 8c, obvious signals of Hydroxy-5, 5-dimethyl-1-pyrrolidinyloxy (DMPO-•OH) were also observed for BiVO4 nanotubes/RGO and BiVO4 nanosheets/RGO in the measurement under visible light irradiation, implying that •OH was produced in the above two reaction systems and took part in the photocatalytic process. It is noticeable that although stronger intensity DMPO- O2•− and DMPO-•OH adducts were found in BiVO4 nanotubes/RGO system, BiVO4 nanosheets/RGO displayed higher photocatalytic efficiency for acetaminophen. This suggests that the photogenerated valence band hole of BiVO4 nanosheets/RGO can oxidize water to generate •OH and participated in the acetaminophen degradation process [42].

**Figure 8.** Electron spin resonance spectra of radical in BiVO4/RGO composites under visible light: (**a**) DMPO-O2•−, (**b**) h+ and (**c**) DMPO-•OH.

For comparison, the electron spin resonance spectra of the pure BiVO4 samples was demonstrated in Figure S9. Although the stronger signals of h+ and DMPO-O2•<sup>−</sup> were observed with BiVO4 nanosheets, the photocatalytic degradation e fficiency is similar to the BiVO4 nanotubes in Figure 5a and Table S1. It can be inferred that, although the photocatalytic activity of BiVO4 materials decreases dramatically after addition of formic acid, •OH dominants the acetaminophen degradation process. The inhibition of holes reduces the amount of reactive hydroxyl radicals, thereby reducing the photocatalytic activity of the system. Thus, this is to say, h+ as well as •OH was the main active species participating in the pure BiVO4 photocatalytic system.

It was reported that higher separation e fficiency of electron-hole pairs plays a vital role in the photocatalytic degradation of pollutants [43,44]. According to the radical trapping experiment and ESR analysis, the reaction mechanism of 1D and 2D BiVO4/RGO heterojunctions for degrading organic pollutants was proposed (Scheme 1). It is assumed that 2D-2D heterojunction between BiVO4 nanosheets and graphene can facilitate the photogenerated electron transfer. That may promote the direct participation of holes in the photocatalytic process or the reaction with OH− to generate •OH. 1D-2D heterojunction interfaces have the ability to yield more photo-generated electrons in the degradation process and promotes the oxygen molecules to generate O2•− and then oxidized to ge<sup>t</sup> •OH. Besides, the stronger intensity of h<sup>+</sup>, O2•−, and •OH in 2D-2D system also demonstrates the intense interface facilitates more e fficient charge separation and transfer.

**Scheme 1.** Schematic image of electron-hole separation mechanism for BiVO4/RGO.

#### *3.4. Photoinduced Electron Transfer Properties of BiVO4*/*RGO Composites*

In a photo-degradation process, the higher photocatalytic e fficiency demands that the electron transfer is faster than the recombination [45]. The prolonged lifetime of the photogenerated electrons can be supported by Photoluminescence (PL) spectra in Figure S10. Under an excitation wavelength of 420 nm, the main emission peak of BiVO4 was detected at around 521 nm, owing to the recombination of electrons in the conduction band and holes in the valence band [46]. The introduction of graphene quenched the PL intensity of excited BiVO4 nanocomposites. The orders of the detected PL intensities were: BiVO4 nanoparticles > BiVO4 nanotubes > BiVO4 nanosheets > BiVO4 nanoparticles/RGO > BiVO4 nanotubes/RGO > BiVO4 nanosheets/RGO, which was in good accordance with the photocatalytic behaviors. Furthermore, the lower PL intensity of BiVO4 nanosheets/RGO suggests that the recombination of the photogenerated electron-hole pairs is e fficiently inhibited by the two-dimensional heterjunction interface and the charge carriers separation rate is promoted.

The enhanced charge transfer rate of BiVO4 nanosheets/RGO was further demonstrated by the transient photocurrent responses. As displayed in Figure 9, the photocurrent density of the BiVO4/RGO composites is much higher than that of the pure BiVO4 samples, especially for BiVO4 nanosheets/RGO. That is in good accordance with the result of photocatalytic performances. It is indicated that the enhanced photocurrent response of the BiVO4 nanosheets/RGO represents higher e fficiency of charge separation and lower recombination rate in 2D-2D heterojunction interface [47].

**Figure 9.** Photocurrent responses of the prepared nanocomposites in 0.5 M Na2SO4 solution during the repeated on-off cycles under visible light irradiation.

On the basis of the above results, the charge transfer mechanism could be proposed as follows. Due to the intimate contact of 2D-2D interface, the favorable transfer of electrons from BiVO4 nanosheets to graphene can reduce the recombination of electron–hole pairs. The enhanced generation of ROS, especially h+ and •OH, accelerate the photocatalytic degradation process of acetaminophen.
