The Synergistic Effect of Adsorption-Photocatalysis for Removal of Organic Pollutants on Mesoporous Cu2V2O7/Cu3V2O8/g-C3N4 Heterojunction

Cu2V2O7/Cu3V2O8/g-C3N4 heterojunctions (CVCs) were prepared successfully by the reheating synthesis method. The thermal etching process increased the specific surface area. The formation of heterojunctions enhanced the visible light absorption and improved the separation efficiency of photoinduced charge carriers. Therefore, CVCs exhibited superior adsorption capacity and photocatalytic performance in comparison with pristine g-C3N4 (CN). CVC-2 (containing 2 wt% of Cu2V2O7/Cu3V2O8) possessed the best synergistic removal efficiency for removal of dyes and antibiotics, in which 96.2% of methylene blue (MB), 97.3% of rhodamine B (RhB), 83.0% of ciprofloxacin (CIP), 86.0% of tetracycline (TC) and 80.5% of oxytetracycline (OTC) were eliminated by the adsorption and photocatalysis synergistic effect under visible light irradiation. The pseudo first order rate constants of MB and RhB photocatalytic degradation on CVC-2 were 3 times and 10 times that of pristine CN. For photocatalytic degradation of CIP, TC and OTC, it was 3.6, 1.8 and 6.1 times that of CN. DRS, XPS VB and ESR results suggested that CVCs had the characteristics of a Z-scheme photocatalytic system. This study provides a reliable reference for the treatment of real wastewater by the adsorption and photocatalysis synergistic process.


Introduction
In recent decades, with the rapid development of industrialization, environmental pollution has become increasingly serious. More and more organic chemicals have been released into the environment. Water pollution has become one of the major obstacles to the sustainable development of human society [1]. Various organic pollutants, such as dyes and antibiotics, are seriously harmful to the ecological environment and human health [2]. Many traditional techniques have been developed to remove the organic pollutants in wastewater, including the bioelectrochemical method [3], electrochemical advanced oxidation processes [4], flocculent precipitation [5], physisorption [6], biological degradation, incineration, membrane filtration, etc. [7]. However, these traditional techniques suffer from the drawbacks, such as not being suitable for low concentrations of pollutant, high operation costs, low removal efficiency and secondary pollution. To date, the integration of adsorption and photocatalysis has been regarded as the most promising technology for the elimination of low-concentration contaminants [8][9][10]. This technology can combine the advantages of adsorption and photocatalysis, such as high efficiency, low cost, wide availability to adsorbates, superior recoverability and less secondary pollution. The adsorption process can ameliorate the accumulation of contaminants on the catalyst surface from the wastewater and conduce to improving the photocatalytic degradation efficiency. optical absorption from the UV to visible region with gap energy of 1.96 eV [44]. About 95% of RhB was removed during visible light irradiation for 120 min over CuV 2 O 6 :In 3+ . Contrastively, only 57% of RhB was eliminated over CuV 2 O 6 under the same conditions. Furthermore, copper vanadate-based heterojunctions, involving in Cu 2 V 2 O 7 /g-C 3 N 4 [45], Cu 2 V 2 O 7 /CoFe 2 O 4 /g-C 3 N 4 [46], Cu 2 O/Cu 2 V 2 O 7 [47], r-GO/β-Cu 2 V 2 O 7 /TiO 2 [48], β-Cu 2 V 2 O 7 /Zn 2 V 2 O 6 [49] and Cu 2 V 2 O 7 /Cu 3 V 2 O 8 [41] were manufactured to enhance photocatalytic activities. Although the relevant studies have shown important progress, more extensive efforts should be made to enhance photocatalytic efficiency of copper vanadates. Herein, Cu 2 V 2 O 7 /Cu 3 V 2 O 8 (CV) was loaded on ultrathin CN nanosheets to form CVC heterojunctions. The exfoliation of bulk CN increased the specific surface area and shortened the diffusion distance of photoinduced charge carriers. The construction of CVCs further accelerated the separation of charge carriers. Thus, the adsorption capability and photocatalytic degradation activity could be remarkably enhanced. The removal of CIP, TC, OTC, MB and RhB was investigated to discuss the synergistic effect of adsorption and photocatalysis.

Results and Discussion
The XRD patterns of CN, CVC-2, CVC-5, CVC-10 and CVC-20 heterojunctions are depicted in Figure 1a. All samples presented two peaks at 13.0 • and 27.6 • , corresponding to (110) and (022) crystal planes of CN. They demonstrated the existence of CN in these samples [16]. The diffraction peaks of CV were undetectable in CVC-2, CVC-5 and CVC-10. This may have resulted from the low dosage of CV in these heterojunctions, which could not cause the change of the chemical skeleton and structure of CN. This result was previously clarified by the literature [50,51]. Moreover, the diffraction peak intensity of CV in CVC-2, CVC-5 and CVC-10 were weaker while CV nanoparticles were covered by CN nanosheets [51]. Comparatively, the diffraction peaks of both CN and CV were all found in CVC-20. In fact, the existence of CV in CVC-2, CVC-5 and CVC-10 could be corroborated by EDS and elemental mapping results. SEM images and corresponding EDS of CVC-2 are displayed in Figure S1. It was revealed that the mass percentage of CV in CVC-2 was about 2.11%, which was well in accordance with the value calculated from the content of Cu(NO 3 ) 3 and NH 4 VO 3 in the precursor. The chemical composition and uniformity of CVC-2 were confirmed by the elemental mappings of C, N, O, Cu and V. All elements can be observed in Figure S2b Contrastively, only 57% of RhB was eliminated over CuV2O6 under the same conditio Furthermore, copper vanadate-based heterojunctions, involving in Cu2V2O7/g-C3N4 [4 Cu2V2O7/CoFe2O4/g-C3N4 [46], Cu2O/Cu2V2O7 [47], r-GO/β-Cu2V2O7/TiO2 [48], Cu2V2O7/Zn2V2O6 [49] and Cu2V2O7/Cu3V2O8 [41] were manufactured to enhance pho catalytic activities. Although the relevant studies have shown important progress, m extensive efforts should be made to enhance photocatalytic efficiency of copper va dates. Herein, Cu2V2O7/Cu3V2O8 (CV) was loaded on ultrathin CN nanosheets to fo CVC heterojunctions. The exfoliation of bulk CN increased the specific surface area a shortened the diffusion distance of photoinduced charge carriers. The construction CVCs further accelerated the separation of charge carriers. Thus, the adsorption capabi and photocatalytic degradation activity could be remarkably enhanced. The removal CIP, TC, OTC, MB and RhB was investigated to discuss the synergistic effect of adsorpt and photocatalysis.

Results and Discussion
The XRD patterns of CN, CVC-2, CVC-5, CVC-10 and CVC-20 heterojunctions depicted in Figure 1a. All samples presented two peaks at 13.0° and 27.6°, correspond to (110) and (022) crystal planes of CN. They demonstrated the existence of CN in th samples [16]. The diffraction peaks of CV were undetectable in CVC-2, CVC-5 and CV 10. This may have resulted from the low dosage of CV in these heterojunctions, wh could not cause the change of the chemical skeleton and structure of CN. This result w previously clarified by the literature [50,51]. Moreover, the diffraction peak intensity CV in CVC-2, CVC-5 and CVC-10 were weaker while CV nanoparticles were covered CN nanosheets [51]. Comparatively, the diffraction peaks of both CN and CV were found in CVC-20. In fact, the existence of CV in CVC-2, CVC-5 and CVC-10 could be c roborated by EDS and elemental mapping results. SEM images and corresponding E of CVC-2 are displayed in Figure S1. It was revealed that the mass percentage of CV CVC-2 was about 2.11%, which was well in accordance with the value calculated from content of Cu(NO3)3 and NH4VO3 in the precursor. The chemical composition and u formity of CVC-2 were confirmed by the elemental mappings of C, N, O, Cu and V. elements can be observed in Figure S2b-f. The uniformity of C, N, O, Cu and V demo strated the homogeneous distribution of CV on CN nanosheets. In addition, the XRD p terns of CV conformed well to the monoclinic phase Cu2V2O7 (PDF#73-1032) and mon clinic Cu3V2O8 (PDF#74-1503), indicating that CV was the composite of Cu2V2O7 a Cu3V2O8 (Figure 1b). These XRD results definitely proved the successful formation CVCs. The FTIR spectra of CN, CVC-2, CVC-5, CVC-10 and CVC-20 are shown in Figure 1c. The sharp adsorption peaks at 806 and 886 cm −1 were assigned to breathing vibration and N-H deformation vibration mode of CN s-triazine, respectively. The characteristic adsorption band in the range of 1200-1700 cm −1 was ascribed to the stretching vibrations of C-N and C=N of aromatic CN heterocycles [52]. The broad adsorption band at 2900-3500 cm −1 was attributed to N-H stretching vibration of residual unpolymerized amino groups. This adsorption band of CN centered at 3165 cm −1 shifted to 3423 cm −1 while CV was loaded on the CN nanosheets. This change of the peak position implied the strong interaction between CV nanoparticles and CN nanosheets [32].
TEM and HRTEM images of CVC-2 are displayed in Figure 2. As shown in Figure 2a,b, there was an ultrathin layered structure of CN nanosheets. It profited from the thermal etching effect in the reheating synthesis process. Especially, the porous structure of CVC-2 could also be observed (Figure 2a,b). The pore diameter was between several and tens of nanometers [50]. This porous structure might originate from the generation of NH 3 and HCl in the polymerization process of dicyandiamine and NH 4 Cl [53]. The ultrathin and porous structure increased the specific surface area of CVC-2, which was proved by BET results. The increased specific surface area could improve the adsorption capability of CVC-2. CV nanoparticles with evident aggregation were also observed in Figure 2a,b. Therefore, the diameters of CV nanoparticles could not be measured accurately from the TEM image. They were estimated to be 50-100 nm in diameter. From Figure 2a,b, it can also be seen that CV nanoparticles were located on CN nanosheets. The spacing of lattice fringes of 0.360 and 0.325 nm are distinctly observed in HRTEM images illustrated in Figure 2c Figure 2d, demonstrating the formation of Cu 2 V 2 O 7 /Cu 3 V 2 O 8 composite. Based on the above XRD, FTIR, TEM and HRTEM results, it was concluded that CVC heterojunctions were successfully constructed [54].
The adsorption and photocatalysis performance of dyes (MB and RhB) over CN and CVCs were investigated and the results are depicted in Figure 3. As shown in Figure 3a,d, the adsorption-desorption equilibrium was achieved within 30 min for all samples. The adsorption capacities of MB and RhB on CVC-2 were 3 times that of CN. The enhanced adsorption capacity of CVC-2 indicated the stronger interaction between CVCs and dyes. All CVCs had much higher adsorption capacity of MB than CN. There was no necessary relation between the adsorption capacity and CV content, suggesting that the changed surface charge that resulted from the addition of CV nanoparticles was not the main factor to affect the interaction between CVCs and dyes. The larger specific surface area profited from the thermal etching effect in the reheating synthesis process which improved the adsorption capacity. The adsorption and photocatalytic degradation results revealed in Figure 3a,d manifest that CVC-2 exhibited the best overall performance for removal of dyes, although it removed MB mainly by adsorption and eliminated RhB primarily via photocatalysis (Figure 3c,f). The kinetic data of MB and RhB photocatalytic degradation on CN and CVCs were well fitted by a pseudo first order rate equation (Figure 3b,e). The degradation rate constants of MB and RhB on CVC-2 were, respectively, 0.036 min −1 and 0.061 min −1 , which were about 3 times and 10 times that of pristine CN. The total removal efficiency of MB and RhB on CVC-2 was 96.2% and 97.3%, respectively. It was much higher than that of CN (55.3% for MB and 34.3%for RhB). This could be considered as the result of the synergistic effect of adsorption and photocatalysis. The adsorption and photocatalysis performance of dyes (MB and RhB) over CN and CVCs were investigated and the results are depicted in Figure 3. As shown in Figure 3a,d, the adsorption-desorption equilibrium was achieved within 30 min for all samples. The adsorption capacities of MB and RhB on CVC-2 were 3 times that of CN. The enhanced adsorption capacity of CVC-2 indicated the stronger interaction between CVCs and dyes. All CVCs had much higher adsorption capacity of MB than CN. There was no necessary relation between the adsorption capacity and CV content, suggesting that the changed surface charge that resulted from the addition of CV nanoparticles was not the main factor to affect the interaction between CVCs and dyes. The larger specific surface area profited from the thermal etching effect in the reheating synthesis process which improved the adsorption capacity. The adsorption and photocatalytic degradation results revealed in Figure 3a,d manifest that CVC-2 exhibited the best overall performance for removal of dyes, although it removed MB mainly by adsorption and eliminated RhB primarily via photocatalysis (Figure 3c,f). The kinetic data of MB and RhB photocatalytic degradation on CN and CVCs were well fitted by a pseudo first order rate equation (Figure 3b,e). The degradation rate constants of MB and RhB on CVC-2 were, respectively, 0.036 min −1 and 0.061 min −1 , which were about 3 times and 10 times that of pristine CN. The total removal efficiency of MB and RhB on CVC-2 was 96.2% and 97.3%, respectively. It was much higher than that of CN (55.3% for MB and 34.3%for RhB). This could be considered as the result of the synergistic effect of adsorption and photocatalysis. Antibiotics (CIP, TC and OTC) were also selected as the target contaminants to further investigate the synergistic removal effect of adsorption and photocatalysis on CVCs. The results are displayed in Figure 4. As presented in Figure 4a,d,g, the adsorption-desorption equilibrium was achieved within 30 min for all samples. The adsorption capacity of CVCs was much higher than that of pristine CN. The larger specific surface area improved the adsorption capacity. CVC-2 exhibited the best overall performance for removal of antibiotics, although it removed TC mainly by adsorption and eliminated CIP and OTC primarily via photocatalysis (Figure 4c,f,i). The adsorption capacity of CVC-2 was 16.9 (CIP), 58.6 (TC) and 4.2 times (OTC) that of CN, respectively. The total removal efficiency of CIP, TC and OTC on pristine CN was 30.2%, 17.5% and 28.0% during the adsorption and photocatalytic degradation process. In contrast, it was 83.0%, 86.0% and 80.5% for CVC-2, respectively. This was considered as the result of the synergistic effect of adsorption and photocatalysis. All CVCs exhibited considerably higher total removal efficiency than CN, indicating the excellent adsorption and photocatalysis performance of CVCs. From Figure 4c,f,i, we found that CVC-2 possessed highest photocatalytic activity. By comparison, the photocatalytic activity of CVC-5, CVC-10 and CVC-20 was decreased with the increase in CV content. This might be derived from the reaction active sites on the surface of CN nanosheets that were excessively occupied by CV nanoparticles in CVC-5, CVC-10 and CVC-20 [16,32]. Moreover, the removal efficiency of dyes and antibiotics on CVC-2 was compared with various recent reported results. As depicted in Table S1, the diversity degradation activities were higher than literature values [25,28,52,[55][56][57][58][59], signifying that CVC-2 is probably a valuable catalyst of practical application in environmental sewage treatment. The kinetic data of CIP, TC and OTC photocatalytic degradation on CN and CVCs were well fitted by a pseudo first order rate equation (Figure 4b,e,h). The degradation rate constants of CVC-2 were 0.017 min −1 (CIP), 0.0049 min −1 (TC) and 0.014 min −1 (OTC). These were 3.6, 1.8 and 6.1 times that of pristine CN, respectively. This result indicated that the loaded CV nanoparticles on the surface of CN nanosheets could improve the photocatalytic activity.  Antibiotics (CIP, TC and OTC) were also selected as the target contaminants to further investigate the synergistic removal effect of adsorption and photocatalysis on CVCs The results are displayed in Figure 4. As presented in Figure 4a,d,g, the adsorption-desorption equilibrium was achieved within 30 min for all samples. The adsorption capacity signifying that CVC-2 is probably a valuable catalyst of practical application in environmental sewage treatment. The kinetic data of CIP, TC and OTC photocatalytic degradation on CN and CVCs were well fitted by a pseudo first order rate equation (Figure 4b,e,h). The degradation rate constants of CVC-2 were 0.017 min −1 (CIP), 0.0049 min −1 (TC) and 0.014 min −1 (OTC). These were 3.6, 1.8 and 6.1 times that of pristine CN, respectively. This result indicated that the loaded CV nanoparticles on the surface of CN nanosheets could improve the photocatalytic activity.  The BET surface area, pore volume and average pore diameter of CN and CVC-2 were investigated by N 2 adsorption-desorption isotherms. As displayed in Figure 5a, the adsorption-desorption isotherms possessed the features of type IV curves, suggesting the samples had a mesoporous structure [60]. The H3 hysteresis loop at high P/P 0 manifested that the mesopores of CN and CVC-2 were irregular. The BET surface area, pore volume and average pore diameter of CVC-2 were evidently greater than those of CN (Figure 5a, inset). They were, respectively, around 4.4, 15.0 and 1.1 times those of CN, which might result from the thermal etching in the reheating synthesis process [61]. The BJH pore size distribution of CN and CVC-2 revealed the wide pore size distribution from 20 to 60 nm, which probably resulted from the aggregation of CN in the reheating synthesis process as displayed previously in the SEM results ( Figure S1a). The pore size between 10 and 20 nm originated from the porous structure of CN, which was observed previously in TEM images (Figure 2a,b). Therefore, the increased specific surface area, pore volume and average pore diameter could be instrumental in providing more adsorption and photocatalytic reaction active sites, and thus finally improve the synergistic effect of adsorption and photocatalysis. originated from the porous structure of CN, which was observed previously in TEM im ages (Figure 2a,b). Therefore, the increased specific surface area, pore volume and avera pore diameter could be instrumental in providing more adsorption and photocataly reaction active sites, and thus finally improve the synergistic effect of adsorption and ph tocatalysis. UV-Vis diffuse reflectance spectra of CN and CVC-2 are exhibited in Figure 6a. T absorption edge of pristine CN was at 457 nm, corresponding to a band gap of 2.71 eV was in agreement with the result reported previously [32,52], confirming that pristine C had proper band gap for visible light harvesting. The absorption edge of CVC-2 was 488 nm. The red shift of the absorption edge of CVC-2 suggested that the incorporation CV nanoparticles onto CN nanosheets was instrumental in extending the visible light a sorption range. From the DRS spectra in Figure 6a, it could be observed distinctly th CVC-2 possessed stronger light absorption in the wavelength range of 400-800 nm th pristine CN nanosheets. This could result in the improvement of the visible light-driv photocatalytic activity, which was clarified by the adsorption and photocatalysis degr dation of CIP, TC and OTC depicted in Figure 4. The photoluminescence spectra (PL) we used to investigate the separation and recombination process of photogenerated char carriers in CN and CVC-2 ( Figure 6b). The higher PL intensity commonly indicates th photogenerated charge carriers have lower separation efficiency and faster recombinati rate [32]. The PL emission peak intensity of CVC-2 was remarkably weaker than that CN. It corroborated that the separation and recombination of photogenerated charge ca riers was effectively ameliorated while CVC heterostructure was formed by incorporati CV nanoparticles into CN nanosheets. It could boost the photocatalytic activity of CVC The property of separation and recombination of the charge carriers in CN and CVC could be further acquired from the photochemical measurements. TPC spectra and E Nyquist plots of CN and CVC-2 are exhibited in Figure 6c,d. The TPC spectra reveal that CVC-2 had better photostability and higher photocurrent density than CN (Figu UV-Vis diffuse reflectance spectra of CN and CVC-2 are exhibited in Figure 6a. The absorption edge of pristine CN was at 457 nm, corresponding to a band gap of 2.71 eV. It was in agreement with the result reported previously [32,52], confirming that pristine CN had proper band gap for visible light harvesting. The absorption edge of CVC-2 was at 488 nm. The red shift of the absorption edge of CVC-2 suggested that the incorporation of CV nanoparticles onto CN nanosheets was instrumental in extending the visible light absorption range. From the DRS spectra in Figure 6a, it could be observed distinctly that CVC-2 possessed stronger light absorption in the wavelength range of 400-800 nm than pristine CN nanosheets. This could result in the improvement of the visible light-driven photocatalytic activity, which was clarified by the adsorption and photocatalysis degradation of CIP, TC and OTC depicted in Figure 4. The photoluminescence spectra (PL) were used to investigate the separation and recombination process of photogenerated charge carriers in CN and CVC-2 ( Figure 6b). The higher PL intensity commonly indicates that photogenerated charge carriers have lower separation efficiency and faster recombination rate [32]. The PL emission peak intensity of CVC-2 was remarkably weaker than that of CN. It corroborated that the separation and recombination of photogenerated charge carriers was effectively ameliorated while CVC heterostructure was formed by incorporating CV nanoparticles into CN nanosheets. It could boost the photocatalytic activity of CVCs. The property of separation and recombination of the charge carriers in CN and CVC-2 could be further acquired from the photochemical measurements. TPC spectra and EIS Nyquist plots of CN and CVC-2 are exhibited in Figure 6c,d. The TPC spectra revealed that CVC-2 had better photostability and higher photocurrent density than CN (Figure 6c). The photocurrent density of CVC-2 was about 2.6 times that of pristine CN. It suggested that the photogenerated charge carriers possessed higher transfer rate in CVC-2, which was beneficial to the migration of the charge carriers to the surface of CVC-2 and consequently to improving its photocatalytic activity. EIS Nyquist plots of CN and CVC-2 are shown in Figure 6d. The arc radius of CN and CVC-2 under visible light irradiation was smaller than that detected in the dark. It demonstrated that the photoelectrode conductivity would be increased under the light irradiation condition. In addition, the arc radius of CVC-2 was less than that of pristine CN in either case, indicating that CVC-2 had higher photogenerated charge transfer efficiency. It was consistent with the previous PL and TPC results, demonstrating that the formation of CVC heterostructure could effectively facilitate the separation and transfer of the photogenerated charge carriers, suppress their recombination and finally improve the photocatalytic activity. The deduction obtained from the DRS, PL, TPC and EIS measurements conformed to the photocatalytic degradation experimental results of dyes and antibiotics. radius of CVC-2 was less than that of pristine CN in either case, indicating that CVC-2 had higher photogenerated charge transfer efficiency. It was consistent with the previous PL and TPC results, demonstrating that the formation of CVC heterostructure could effectively facilitate the separation and transfer of the photogenerated charge carriers, suppress their recombination and finally improve the photocatalytic activity. The deduction obtained from the DRS, PL, TPC and EIS measurements conformed to the photocatalytic degradation experimental results of dyes and antibiotics. The reactive species involved in the photocatalytic degradation reaction were evaluated by radical scavenger experiments. NaNO3, ammonium oxalate(AO), isopropyl alcohol(IPA) and p-benzoquinone (pBQ) were adopted as the scavengers to trap e − , h + , •OH and •O2 − in the photocatalytic degradation of CIP over CVC-2, respectively. As revealed in Figure 7a, the removal efficiency of CIP was reduced dramatically after adding AO and pBQ, suggesting that h+ and •O2 − were the major reactive species in the photocatalytic process. Comparatively, just a slight reduction of degradation efficiency was observed with the addition of NaNO3 and IPA. It indicated that eand •OH hardly participated in the photocatalytic reaction. To further confirm the production of •O2 − , ESR spectra of DMPO-•O2 − over CVC-2 in the CIP photocatalytic system were detected under different light irradiation times and the result is presented in Figure 7b. No obvious ESR signals of •O2 − were found in the dark, but they appeared after 15 min of visible light irradiation. The reactive species involved in the photocatalytic degradation reaction were evaluated by radical scavenger experiments. NaNO 3 , ammonium oxalate(AO), isopropyl alcohol(IPA) and p-benzoquinone (pBQ) were adopted as the scavengers to trap e − , h + , •OH and •O 2 − in the photocatalytic degradation of CIP over CVC-2, respectively. As revealed in Figure 7a, the removal efficiency of CIP was reduced dramatically after adding AO and pBQ, suggesting that h+ and •O 2 − were the major reactive species in the photocatalytic process. Comparatively, just a slight reduction of degradation efficiency was observed with the addition of NaNO 3 and IPA. It indicated that eand •OH hardly participated in the photocatalytic reaction. To further confirm the production of •O 2 − , ESR spectra of DMPO-•O 2 − over CVC-2 in the CIP photocatalytic system were detected under different light irradiation times and the result is presented in Figure 7b. No obvious ESR signals of •O 2 − were found in the dark, but they appeared after 15 min of visible light irradiation. These results demonstrated that •O 2 − was generated under the visible light irradiation and participated in the photocatalytic degradation process of CIP.
The stability was assessed by four cycles of the adsorption and photocatalysis synergistic removal experiments of MB and CIP on CVC-2. CVC-2 was centrifuged and desorbed in deionized water several times to eliminate absorbed MB and CIP after each cycle experiment. As shown in Figure 8a,b, the removal efficiency of MB and CIP was well maintained after four degradation runs, confirming that CVC-2 had superior stability and excellent potential application prospects in wastewater treatment. TEM, XRD and FTIR were used to characterize the differences of structure and morphology of CVC-2 before and after the synergistic removal experiment. As shown by the TEM, XRD and FTIR results depicted in Figures S3-S5, the phase structure and morphology characteristics were not changed evidently after four degradation runs, which further illustrated the excellent stability of CVC-2.
These results demonstrated that •O2 − was generated under the visible light irradiation and participated in the photocatalytic degradation process of CIP. The stability was assessed by four cycles of the adsorption and photocatalysis synergistic removal experiments of MB and CIP on CVC-2. CVC-2 was centrifuged and desorbed in deionized water several times to eliminate absorbed MB and CIP after each cycle experiment. As shown in Figure 8a,b, the removal efficiency of MB and CIP was well maintained after four degradation runs, confirming that CVC-2 had superior stability and excellent potential application prospects in wastewater treatment. TEM, XRD and FTIR were used to characterize the differences of structure and morphology of CVC-2 before and after the synergistic removal experiment. As shown by the TEM, XRD and FTIR results depicted in Figures S3-S5, the phase structure and morphology characteristics were not changed evidently after four degradation runs, which further illustrated the excellent stability of CVC-2. The energy band structure of CN and CV was obtained by DRS and XPS VB spectra. The DRS spectrum of CV is displayed in Figure S6. It exhibited the absorption edge of 628 nm, corresponding to the band gap of 1.97 eV. It was in agreement with thatβ-Cu2V2O7in the literature [62]. XPS VB spectra of CN and CV are shown in Figure S7. The EVB of CN and CV was 1.58 eV and 2.20 eV, respectively. The ECB of CN and CV could be estimated based on the following equation: Eg = EVB − ECB. It was -1.13 eV and 0.23 eV, respectively. The ECB of CV was more positive than the redox potential of O2/•O2 − (−0.33 V) [52],  The stability was assessed by four cycles of the adsorption and photocatalysis synergistic removal experiments of MB and CIP on CVC-2. CVC-2 was centrifuged and desorbed in deionized water several times to eliminate absorbed MB and CIP after each cycle experiment. As shown in Figure 8a,b, the removal efficiency of MB and CIP was well maintained after four degradation runs, confirming that CVC-2 had superior stability and excellent potential application prospects in wastewater treatment. TEM, XRD and FTIR were used to characterize the differences of structure and morphology of CVC-2 before and after the synergistic removal experiment. As shown by the TEM, XRD and FTIR results depicted in Figures S3-S5, the phase structure and morphology characteristics were not changed evidently after four degradation runs, which further illustrated the excellent stability of CVC-2. The energy band structure of CN and CV was obtained by DRS and XPS VB spectra. The DRS spectrum of CV is displayed in Figure S6. It exhibited the absorption edge of 628 nm, corresponding to the band gap of 1.97 eV. It was in agreement with thatβ-Cu2V2O7in the literature [62]. XPS VB spectra of CN and CV are shown in Figure S7. The EVB of CN and CV was 1.58 eV and 2.20 eV, respectively. The ECB of CN and CV could be estimated based on the following equation: Eg = EVB − ECB. It was -1.13 eV and 0.23 eV, respectively. The ECB of CV was more positive than the redox potential of O2/•O2 − (−0.33 V) [52], The energy band structure of CN and CV was obtained by DRS and XPS VB spectra. The DRS spectrum of CV is displayed in Figure S6. It exhibited the absorption edge of 628 nm, corresponding to the band gap of 1.97 eV. It was in agreement with thatβ-Cu 2 V 2 O 7 in the literature [62]. XPS VB spectra of CN and CV are shown in Figure S7. The E VB of CN and CV was 1.58 eV and 2.20 eV, respectively. The E CB of CN and CV could be estimated based on the following equation: E g = E VB − E CB . It was -1.13 eV and 0.23 eV, respectively. The E CB of CV was more positive than the redox potential of [52], meaning that O 2 adsorbed on CV could not be reduced to •O 2 − . Therefore, the reduction reaction of O 2 could only occur on the CB of CN. ESR spectra of DMPO-•OH were also measured under visible light irradiation with the presence of CVC-2 ( Figure S8). They confirmed the production of •OH although the radical scavenger experiments indicated that •OH radicals did not participate in the photocatalytic reaction. Furthermore, the E VB of CN was more negative than the redox potential of •OH/OH − (+1.99 V) and •OH/H 2 O (+2.37 V) [63]. The •OH radicals could not be generated on CN, and only formed on the VB of CV. These results clearly suggested that CVC heterojunctions possessed the characteristics of Z-scheme photocatalytic systems.
The possible photocatalytic mechanism of dyes and antibiotics on CVCs is consequently proposed in Figure 9. Firstly, the photoinduced electrons (e − ) were excited and transferred to the CB of CN and CV under visible light irradiation. The photoinduced holes (h + ) were still retained in the VB. Secondly, the electrons in the CB of CV transferred to CN and recombined with the holes located in the VB of CN. This recombination could significantly accelerate the separation of the photoinduced electrons and holes of CN and CV (Equation (1)), which was the advantage of Z-scheme heterojunctions [64,65]. Thirdly, the electrons on CN reduced oxygen to generate •O 2 − and then •O 2 − further oxidized dyes and antibiotics (Equations (2) and (3)). The holes on CV directly oxidized dyes and antibiotics to small molecules (Equation (4)). The photocatalytic degradation process of dyes and antibiotics could be elaborated as follows: Dyes and antibiotics Dyes and antibiotics + CV (h + ) → degraded products. (4) quently proposed in Figure 9. Firstly, the photoinduced electrons (e − ) were excited transferred to the CB of CN and CV under visible light irradiation. The photoind holes (h + ) were still retained in the VB. Secondly, the electrons in the CB of CV transf to CN and recombined with the holes located in the VB of CN. This recombination c significantly accelerate the separation of the photoinduced electrons and holes of CN CV (Equation (1)), which was the advantage of Z-scheme heterojunctions [64,65]. Th the electrons on CN reduced oxygen to generate •O2 − and then •O2 − further oxidized and antibiotics (Equations (2) and (3)). The holes on CV directly oxidized dyes and a otics to small molecules (Equation (4)). The photocatalytic degradation process of dye antibiotics could be elaborated as follows:

Synthesis of CVCs
Preparation of CN nanosheets: 1:1 mole ratio of dicyandiamine and NH4Cl mixed and calcinated for 4 h at 550 °C [61]. One hundred micrograms of obtained po was ground and dispersed in 100 mL deionized water. The mixture was tr

Synthesis of CVCs
Preparation of CN nanosheets: 1:1 mole ratio of dicyandiamine and NH 4 Cl was mixed and calcinated for 4 h at 550 • C [61]. One hundred micrograms of obtained powder was ground and dispersed in 100 mL deionized water. The mixture was treated ultrasonically for 8 h and then centrifuged for 10 min at 5000 rpm. The as-prepared CN nanosheets in the supernatant were separated by vacuum freeze-drying.
Preparation of CVCs: 100 mg of CN nanosheets, 1:1 mole ratio of Cu(NO 3 ) 3 and NH 4 VO 3 was dispersed in 100 mL of deionized water. Then, the water was evaporated and annealed at 500 • C for 2 h. A series of CVCs were synthesized by adjusting the mass ratio of CV and CN. The as-prepared CVC-2, CVC-5, CVC-10 and CVC-20 contained 2, 5, 10 and 20 wt% of CV, respectively. CV nanoparticles were synthesized under the same conditions except for the absence of CN nanosheets.

Catalyst Characterization
X-ray diffraction (XRD) patterns of CN, CV and CVCs were tested on a Rigaku Smartlab diffractometer equipped with a Cu-Kα radiation source. Transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) images of CVCs were taken on a JEOL-2100F system. Fourier-transform infrared spectra (FTIR) of CN and CVCs were detected on a Nicolet NEXUS 470 spectrometer in the range of 4000-500 cm −1 . Scanning electron microscopy (SEM) images, energy dispersion spectrum (EDS) and elemental mapping images of CVC-2 were characterized on a JSM-4800F scanning electron microscope. X-ray photoelectron spectroscopy (XPS) spectra of CN, CV and CVCs were obtained on a Thermo ESCALAB 250XI spectrometer equipped with an AlKα X-ray source. The BET specific surface area and N 2 adsorption desorption isotherms of CN and CVCs were recorded on a Micromeritics ASAP 2460 analyzer at 77 K. The ultraviolet-visible diffuse reflectance spectra (DRS) of CN and CVCs were taken on a Shimadzu UV-2401 spectrophotometer equipped with an integrating sphere accessory.

Adsorption and Photocatalytic Experiments
The adsorption of contaminants on CN, CV and CVCs was implemented in the dark before the photocatalysis process. Typically, 50 mg CN, CV or CVCs was dispersed in 100 mL antibiotic (CIP, TC and OTC) or dye (MB and RhB) solution. The pH value of the solution was adjusted using 0.1 M HCl or NaOH. The mixture was stirred continuously in the dark for 30 min to reach adsorption-desorption equilibrium of the contaminants on the catalysts. The photocatalytic degradation of contaminants was evaluated under the visible light irradiation of 40 W white LED. Five milliliters of solution was fetched out at intervals and the catalyst was removed through a 0.22 µm PTFE filter membrane. The contaminant concentration in adsorption solution was detected by a UV-Vis spectrophotometer at 277 nm (CIP), 357 nm (TC), 352 nm (OTC), 664 nm (MB) and 552 nm (RhB). The temperature was fixed at 25 • C in the adsorption and photocatalysis process. The tests were repeated three times.
The stability of CVC-2 was appraised by 4 cycles of adsorption photocatalytic degradation experiments. CVC-2 was centrifuged and desorbed in deionized water several times after each cycle experiment to eliminate absorbed contaminant. Then, CVC-2 was separated and freeze-dried for the usage in the next cycle experiment. The reproducibility of CVC-2 was evaluated by repeating the above process twice. NaNO 3 , AO, IPA and pBQ were selected as the scavengers to trap e − , h + , •OH and •O 2 − , respectively.

Photoelectrochemical Measurement
The photoelectrochemical measurements were conducted on the CHI 660E electrochemical workstation. Ag/AgCl was the reference electrode and Pt foil was the counter electrode. A 0.2 M Na 2 SO 4 solution was the electrolyte. For the preparation of working electrode, 5 mg of CN or CVCs was mixed with 0.2 mL Nafion and 1.8 mL ethanol ultrasonically. Then, the mixture was dropped on 1 cm 2 of FTO glass and dried naturally. The transient photocurrent response (TPC) was tested under the irradiation of 40 W white LED. The irradiation intervals were realized using a mechanical light chopper. The electrochemical impedance spectroscopy (EIS) was measured at a frequency from 100 to 0.01 Hz.

Conclusions
In summary, CVC heterojunctions with ultrathin structure were successfully prepared by the reheating synthesis process for the adsorption and photocatalysis synergistic removal of various dyes and antibiotics. The thermal etching process increased the specific surface area of CVCs. The formation of heterojunctions enhanced the visible light absorption and improved the separation efficiency of photoinduced charge carriers. These factors simultaneously ameliorated the adsorption capacity and photocatalytic degradation performance of CVCs. CVC-2 exhibited the best synergistic removal efficiency of MB (96.2%), RhB (97.3%), CIP (83.0%), TC (86.0%) and OTC (80.5%). These photocatalytic degradation processes followed the pseudo first order equation. The pseudo first order rate constants of MB, RhB, CIP, TC and OTC photocatalytic degradation on CVC-2 were 3, 10, 3.6, 1.8 and 6.1 times those of pristine CN, respectively. DRS, XPS VB and ESR results suggested that CVCs had the characteristics of Z-scheme photocatalytic systems. Moreover, superoxide radicals and photoinduced holes were proved to be the major active species in the photocatalytic degradation process. This work provides a reliable reference for environmental sewage treatment by the adsorption and photocatalysis synergistic process.  Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: The original data are available from the corresponding author upon reasonable request.

Conflicts of Interest:
The authors declare no conflict of interest.