g-C3N4–Co3O4 Z-Scheme Junction with Green-Synthesized ZnO Photocatalyst for Efficient Degradation of Methylene Blue in Aqueous Solution

A simple wet chemical ultrasonic-assisted synthesis method was employed to prepare visible light-driven g-C3N4-ZnO-Co3O4 (GZC) heterojunction photocatalysts. X-ray diffraction (XRD), scanning electromicroscopy (SEM), Fourier-transform infrared spectroscopy (FTIR), Brunauer–Emmett–Teller (BET), ultraviolet (UV), and electrochemical impedance spectroscopy (EIS) are used to characterize the prepared catalysts. XRD confirms the homogenous phase formation of g-C3N4, ZnO, and Co3O4, and the heterogeneous phase for the composites. The synthesized ZnO and Co3O4 by using cellulose as a template show a rod-like morphology. The specific surface area of the catalytic samples increases due to the cellulose template. The measurements of the energy band gap of a g-C3N4-ZnO-Co3O4 composite showed red-shifted optical absorption to the visible range. The photoluminescence (PL) intensity decreases due to the formation of heterojunction. The PL quenching and EIS result shows that the reduction of the recombination rate and interfacial resistance result in charge carrier kinetic improvement in the catalyst. The photocatalytic performance in the degradation of MB dye of the GZC-3 composite was about 8.2-, 3.3-, and 2.5-fold more than that of the g-C3N4, g-C3N4-ZnO, and g-C3N4-Co3O4 samples. The Mott–Schottky plots of the flat band edge position of g-C3N4, ZnO, Co3O4, and Z-scheme g-C3N4-ZnO-Co3O4 photocatalysts may be created. Based on the stability experiment, GZC-3 shows greater photocatalytic activity after four recycling cycles. As a result, the GZC composite is environmentally friendly and efficient photocatalyst and has the potential to consider in the treatment of dye-contaminated wastewater.


Introduction
Today's society faces environmental challenges due to the ever-increasing demand for energy, the excessive use of fossil fuels, and related complications such as greenhouse gas emissions. Organic dyes are chemical pollutants that contaminate water. Tese contaminants in wastewater are very hazardous, can cause carcinogens, and are dangerous to humans, animals, and entire ecosystems [1]. Tere are approximately 10,000 varieties of commercially used dyes, with an annual production of around 0.7 billion tons [2]. About 20% of these colors are lost during dyeing processes and discharged as textile efuents [3]. Various methods, in particular, ion exchange [4], coagulation-focculation [5], oxidation [6], electrochemical treatment [7], and membrane-based flters have been employed in the removal of organic contaminants from polluted solution [8].
Photocatalysis has emerged as an intriguing degradation mechanism among the existing potential methods of removing organic efuents from the solution due to its low cost, nontoxicity, safety, and renewable nature. TiO 2 , which was used in 1972 by Fujishima for photoelectrochemical water splitting, is the most extensively used photocatalyst today [9]. Te typical TiO 2 catalyst, on the other hand, is stimulated by UV light that takes up less than 5% of the whole solar spectrum. Tis has prompted researchers to create innovative materials with lower bandgap energy (E g ) to improve sensitivity to an increasingly plentiful visible light photon [10][11][12][13][14][15]. Terefore, the heterojunction of cobalt(II, III) oxide and bismuth oxyiodides efciently remove nitrophenol from solutions due to the high specifc surface area and density of the photogenerated charge carrier resulting in its visible light-sensitivity of the catalyst [16,17]. Acid Blue 25 dyes were removed efectively by tetraphenylporphyrin/tungsten (VI) oxide/reduced graphene oxide photocatalyst [18]. Moreover, mesoporous dendritic silica supports TiO 2 to improve the catalytic performance for the degradation of carbamazepine due to the reduction of the charge recombination rate, charge kinetics improvement, and increased density of the active site for the analyte adsorption [19].
Graphitic carbon nitride (g-C 3 N 4 ) has recently gained a lot of interest because it possesses -conjugated planar layers akin to graphite, giving it great thermal and chemical stability as well as an attractive electrical structure [20]. As a result, it could be used as a direct semiconductor catalyst in sustainable chemistry. Te g-C 3 N 4 has distinct properties such as a good electrical and optical structure, as well as strong photochemical stability, and is considered a potential photocatalyst [21][22][23]. g-C 3 N 4 has been shown to function well in photo-degrading various organic dyes when exposed to visible light [24,25]. Nevertheless, due to its small surface area, limited visible light absorption, and the electron-hole recombination rate, the photocatalytic efciency of g-C 3 N 4 is far from optimal. Several ways have been used to overcome these limitations. Many methods, including chemical and physical exfoliation methods, have been used to enhance the surface area of g-C 3 N 4 . Exfoliation of g-C 3 N 4 provides not only a high surface area but also a shorter difusion length which might result in a low recombination rate [26][27][28][29]. Doping with metals and nonmetals has also been found to be an efective method for increasing visible light absorption using bandgap engineering [30]. Qiao's group demonstrated that phosphorus doping porous g-C 3 N 4 nanosheets can signifcantly lower the bandgap from 2.98 to 2.66 eV which creates more photo-excitation of electrons and holes [31][32][33][34]. Concerning the high electron-hole recombination rate of g-C 3 N 4 , it can be reduced by loading noble metals on its surface or creating junction with diferent other semiconductors such as TiO 2 /g-C 3 N 4 [35], graphene oxide/g-C 3 N 4 [36], Au/g-C 3 N 4 [37]/g-C 3 N 4 /α-Fe 2 O 3 [38], g-C 3 N 4 / Ag 2 O/TiO 2 [39], g-C 3 N 4 /Bi 2 WO 6 [40], Ag/g-C 3 N 4 [41], g-C 3 N 4 /Co 3 O 4 /V 2 O 5 [42], Cu doped ZnO/g-C 3 N 4 [43], g-C 3 N 4 /Co 3 O 4 [44], BiOBr-NFs/g-C 3 N 4 -SAF [45], and Co 3 O 4 @CeO 2 [46], and ZnO is an inexpensive and nontoxic semiconductor material that has been utilized as a photocatalyst; however, due to its high band gap (3.2-3.3 eV), it is only active in the UV region of the solar spectrum [47][48][49]. Several literatures demonstrate that the ZnO-coupled g-C 3 N 4 composite has shown to increase the oxidation potential and removal efciency of inorganic and organic contaminants. However, due to reduced visible light absorption in the solar spectrum, the application of the g-C 3 N 4 -ZnO composite has restricted activity in the visible light area. To solve this issue, a p-type inorganic spinal cobalt oxide (Co 3 O 4 ) semiconductor material with great thermal durability, nontoxicity, and excellent optical property energy bandgap (2.1 eV) can be used [50]. Few studies have combined all of the above-mentioned approaches for increasing the photocatalytic properties of g-C 3 N 4 . Herein, the incorporation of Co 3 O 4 into g-C 3 N 4 was used to increase absorbance in the visible portion of the solar spectrum. Chemical exfoliation of bulk g-C 3 N 4 was used to reduce the number of stacked layers while increasing the active surface area. We used cellulose extracted from the local plants as a template for metal oxide synthesis because the hydroxyl groups on cellulose act as efcient hydrophilic substrates for metal oxide nucleation and growth, resulting in rod-like morphology which results in a high surface area. Te Z-scheme system is designed to enhance the charge carrier density and collection efciency of the photocatalyst [51]. As a result, the current article presents a thorough examination of the efects of combining all of the approaches on enhancing the performance of g-C 3 N 4 regarding methylene blue degradation. Under visible light illumination, the composite outperformed pure g-C 3 N 4 in photodegradation of methylene blue by over 8.2 times. Using various characterization techniques, the role of each approach in enhancing photocatalytic activity of g-C 3 N 4 -ZnO-Co 3 O 4 compared to g-C 3 N 4 is discussed.

Material Characterization.
Te phase composition and crystallinity of the prepared photocatalyst were measured by using XRD, ShimadzuXRD-7000, with CuK α radiation. Fourier-transform infrared (FTIR) analysis was performed using Spectrum 65 FTIR (Perkin Elmer) in the range of 4000−400 cm −1 using KBrpellets. Te morphologies of the samples were examined by feld-emission scanning electron microscopy (FESEM, JSM 6500F, and JEOL). Te optical absorption spectra were measured using a Shimadzu 3600 UV-Vis-NIR spectrophotometer in the wavelength range of 200-800 nm using BaSO4 as a reference. Brunauer, Emmett, and Teller (BET), ASAP 2020 HD88 surface area analyzer, was used to measure the specifc surface area of the samples by N 2 adsorption. A spectrophotometer, PE-LS55, USA, a xenon lamp light source, and an exciton wavelength of 326 nm were used to measure the photoluminescence spectra of the sample.
EIS and Mott-Schottky were measured using a threeelectrode system with an Autolab PGSTAT 302 N electrochemical test system (Metrohm Autolab B.V.), with Pt wire and Ag/AgCl electrode as the counter and reference electrodes, respectively.

Synthesis of Cellulose.
Cellulose from enset was synthesized, as shown in Figure S1 [40]. Te plant fbre was washed and dried in an electric oven at 80°C to eliminate reaming dust. Te raw material was boiled with a 2% NaOH (1 : 40) W/V ratio for 2 hr, then dried in an electric oven at 70°C, and chopped. After the frst alkaline treatment, the second alkaline pretreatment was carried out on a hot plate at 90°C for 1.5 hr in 10% NaOH solutions with a 1 : 10 (W/V) solid-to-liquid ratio of the dry material. Te resulting pulps were centrifuged, dried, and subsequently treated with 20% of CH 2 O 2 /20% of CH₃COOH/7.5% of H 2 O 2 (2 : 1 : 2) solution at 90°C temperature of hotplate for 1.5 hr. Te dignifed pulps were fltered and washed with hot water to separate the cooking liquid (which contains lignin and hemicellulose) from the cellulose. Before the bleaching procedure, the pulps were centrifuged and dried. Te pulps were bleached for 30 min at 70°C with 7.5% H 2 O 2 and 10% NaOH solutions. Te pulps were washed several times with distilled water to eliminate residuals before drying at 60°C in an oven. Finally, 5 g cellulose was added to a 250 ml 50% H 2 SO 4 aqueous solution for 30 min at 60°C before being diluted with distilled water. After centrifuging and washing with distilled water until the pH was neutral, the cellulose gel was ultrasonically dispersed in 1 L of distilled water and stored for later use.

Synthesis of ZnO.
A simple chemical precipitation synthesis method was used to prepare ZnO, as shown in Figure S2 [41]. In a nutshell, 0.1 M of zinc nitrate hexahydrate was dissolved in 200 ml of dispersed cellulose solution under vigorous stirring for 0.5 hrs, and then the pH of the solution was adjusted to 10 by adding 1 M NaOH drop wise with continuous stirring for precipitation. Ten, the solution was stirred for 2 hr at 60°C and later aged for 24 hr. Te precipitate was centrifuged and rinsed with distilled water and ethanol. After washing, the precipitate sample was recovered and dried in an electric oven at 100°C for 2 hrs. Te dried sample was then calcined in a mufe furnace at 500°C for 2 hrs. After being ground to a powder, the calcined ZnO was recovered and used in future studies. Figure S3 [42]. In a nutshell, 0.1 M of cobalt nitrate hexahydrate was dissolved in 200 ml of dispersed cellulose solution under vigorous stirring for 0.5 hrs, and then the pH of the prepared solution was raised to 10 by adding 1 M NaOH drop wise with continuous stirring for precipitation. Ten, the solution was stirred for 2 hrs at 60°C and then aged for 24 hrs. Te precipitate was centrifuged and washed with distilled water and ethanol. After washing, the precipitate sample was recovered and dried in an electric oven at 100°C for 2 hrs. Te dried sample was then calcined in a mufe furnace at 400°C for 2 hrs. Te calcined Co 3 O 4 was collected and ground to powder and used in future studies.

Synthesis of g-C 3 N 4 .
To synthesize g-C 3 N 4 , a thermal pyrolysis method was used, as shown in Figure S4 [31,32]. 10 g of urea fake was placed in a crucible tightly wrapped in aluminum foil and heated in a mufe furnace at the rate of 2°C min -1 up to 550°C and held for 4 hrs. After naturally cooling to room temperature, a pale-yellow sponge-like powder product was collected. 300 mg of as-prepared powder was mixed in 15 ml of H 2 SO 4 and vigorously stirred at room temperature for 24 hrs to break the bonds between layered stacked structures of g-C 3 N 4 . Tis solution was diluted with 400 ml of distilled water and ultrasonicated for 6 hrs to exfoliate the g-C 3 N 4 layers. Te suspension was fltered from the residual solution and washed multiple times with distilled water and ethanol to clean any residual acid from the sample. Finally, at 80°C, the mixture was dried to obtain a g-C 3 N 4 powder.

X-Ray Difraction Analysis.
A simple wet chemical ultrasonic-assisted synthesis method was employed to prepare the g-C 3 N 4 -ZnO-Co 3 O 4 composite, as shown in Figure S5. XRD was used to examine the phase purity and the crystalline structure of samples. Figure 1 (A-K) depicts XRD results of bulk g-C 3 N 4 , exfoliated g-C 3 N 4 , Co 3 O 4 , ZnO, g-C 3 N 4 -Co 3 O 4 , g-C 3 N 4 -ZnO, and g-C 3 N 4 -ZnO-Co 3 O 4 . Figure S6 shows that the XRD result shows peaks at 16°, 22°, and 34°at (2θ), respectively, with the assigned crystallographic planes of (110), (200), and (400), indicating the formation of cellulose I, as well as no doublet in the intensity of the peaks, indicating the absence of cellulose II [52]. Tere are two distinct peaks for bulk g-C 3 N 4 with a strong signal at 27.5°and 13.2°, which correspond to the (002) and (100) planes of g-C 3 N 4 (#JCPDS-87-1526) [53] (see Figure 1 A). Tri-s-triazine units and aromatic interlayer stacking are the indication of (002) and (100) crystal planes, respectively [52]. Te peak intensity of plane (002) is weak for exfoliated g-C 3 N 4 , as shown in Figure 1 (B), which implies a reduction in the number of stacked layers in g-C 3 N 4 due to exfoliation using sulfuric acid [41]. Te metal oxides XRD result is also depicted in Figure 1 [55]. Te XRD patterns of g-C 3 N 4 -ZnO, g-C 3 N 4 -Co 3 O 4 , and prepared g-C 3 N 4 -ZnO-Co 3 O 4 composite catalysts with diferent weight percent contents of zinc oxide and cobalt oxide to graphitic carbon nitride were illustrated in Figure 1(E-K). Te g-C 3 N 4 -ZnO-Co 3 O 4 composites have matched with original difraction peaks of g-C 3 N 4 , the wurtzite hexagonal ZnO phase, and the cubic Co 3 O 4 phase, with no additional peaks present, indicating that the composite has successfully formed. In the composites, a similar peak shape and position as seen in g-C 3 N 4 , ZnO, and Co 3 O 4 were observed that indicates that the interaction of ZnO and Co 3 O 4 with g-C 3 N 4 does not afect the original lattice structure of g-C 3 N 4 .
Te difraction peaks of ZnO and Co 3 O 4 were steadily strengthened at the cost of the g-C 3 N 4 peak intensity while the ZnO and Co 3 O 4 concentrations increased. Furthermore, no additional peaks appeared in the samples for any possible impurities or phases. Figure 2, three separate bands in the FTIR spectra of the g-C 3 N 4 , ZnO, Co 3 O 4 , and GZC-3 samples were observed. Te heterocyclic aromatic bonds, C-N, are associated with the stretching vibration of C�N, C-N, and C-N-C. Te stretching vibration peak is predicted at the wavenumber range of 1200-1650 cm −1 , along with 1238, 1318, 1410, 1574, and 1640 cm −1 [56][57][58][59]. Te broad peak in the vicinity of 3000-3500 cm −1 is attributed to the N-H stretching of residual amine groups (-NH 2 and -NH), which might be a residue of the precursor urea. Tis broad peak might also attribute to the O-H stretching band due to moisture absorption of the sample from the environment [60]. Te typical characteristic peaks at around 808 cm −1 indicate the s-triazine ring structure of g-C 3 N 4 , which is evidence of proper phase formation [61]. In Figure 2, the FTIR spectra of the Co 3 O 4 absorption spectra band at 3416 cm −1 and 1635 cm −1 show O-H stretching and bending vibrations, respectively. It might be due to moisture absorption of the sample from the environment [62,63]. Furthermore, the absorbance bands from the FTIR spectra at 654 cm −1 and 558 cm −1 indicate vibrations of Co(III)-O bonds and Co-O stretching, confrming the formation of Co 3 O 4 [60,64,65]. In Figure 2, the FTIR spectra of the ZnO absorption spectra band at 3400 cm −1 and 1380 cm −1 show O-H stretching and bending vibrations, respectively. It might be due to moisture absorption of the sample from the environment [66]. Te absorbance bands from the FTIR spectra at 1110 cm −1 indicate vibrations of the Zn-O bond, confrming the formation of ZnO [67]. Te FTIR results support the g-C 3 N 4 , ZnO and Co 3 O 4 phase formation as claimed in XRD analysis. Te FTIR spectra of the GZC-3 composite include bands associated with g-C 3 N 4 , Co 3 O 4 , and ZnO functional groups, respectively, confrming the successful synthesis of the g-C 3 N 4 -ZnO-Co 3 O 4 composite as illustrated in Figure 2. 3.3. SEM Analysis. Scanning electron microscopy was used to examine the morphology of synthesized samples. Cellulose templates can bind metal cations and regulate the particle size of metal oxide during the synthesis process. When cobalt nitrate and zinc nitrate are mixed in a cellulose solution, the cobalt nitrate and zinc nitrate  hydrolyze and interact to generate a strong adhesion on the cellulose surface because cellulose is rich in hydrogen bonds and a hydroxyl macromolecule. Te physical space from cellulose makes it difcult for the formation of a large aggregation during the growth of these metal oxides. After calcination, the template is removed, resulting in dispersed ZnO and Co 3 O 4 . Figures 3(a) and 3(b) show the image of the SEM for the ZnO and Co 3 O 4 grown using cellulose as a template, which has a characteristic of rod-like morphology. Te rod-like morphology allows for faster kinetics of charge carriers to the catalyst's surface, which may reduce electron and hole recombination rates [68]. Figure 3(c) displays a sheet-like morphology for g-C 3 N 4 ; it is typical for g-C 3 N 4 synthesized by thermal polymerization followed by chemical exfoliation. Figure 3(d) shows the image of the scanning electron microscope of the GZC-3 composite and illustrates the typical morphology of ZnO and Co 3 O 4 across the matrix of g-C 3 N 4 . Te chemical composition of the GZC-3 composite was analyzed using EDX spectroscopy, with the fndings shown in Figure 3(e). Te EDX spectra of the GZC-3 composite contained peaks corresponding to the Zn, Co, O, C, and N elements. Tese results clearly show the successful formation of the GZC composite.

UV Analysis.
Te optical absorbance of synthesized samples g-C 3 N 4 , ZnO, Co 3 O 4 , and GZC composite samples were measured by using a UV-vis spectrometer, as shown in Figure 4(a). ZnO absorbs more in the UV range of the solar spectrum, less than 400 nm wavelength, as shown in the spectra designated by the black line in Figure 4(a) [69]. Te Co 3 O 4 exhibited a strong absorption tail at a wavelength longer than 500 nm, as shown in the spectra designated by the red line in Figure 4(a) [70]. Te g-C 3 N 4 displayed high absorption capabilities between 250 and 380 nm, with an absorption edge at 460 nm, which is still short for absorbing the visible portion of solar radiation and leaving the large part of the visible light spectrum unexplored. Te addition of Co 3 O 4 and ZnO samples in the g-C 3 N 4 matrix increased absorption intensity while also red-shifting the wavelength towards the visible light area (shifted to lower energy). Te appropriate form of a junction formed between g-C 3 N 4 , ZnO, and Co 3 O 4 might be the reason for the better absorbance to the visible area of the composites. Te Tauc equation was employed to determine the energy bandgap of the synthesized samples from UV-visible absorbance spectra, as shown in Figure 4( , g-C 3 N 4 , GZ, GC, GZC-1, GZC-2, GZC-3, GZC-4, and GZC-5 composites, respectively. Te energy band gap of the composite samples seems to obey the superposition principle, as shown in Figure 4(b).

BET Surface Area and Pore Size Analysis.
Te pore volume and the specifc surface area of metal oxide prepared by cellulose templated or not are summarized in Figure 5. Te specifc surface area of ZnO prepared without cellulose measured 484.707 m 2 /g compared to cellulose templated 538.563 m 2 /g. Te surface area of cobalt oxide prepared without cellulose is 48.673 m 2 /g, and with cellulose templated, it is 520.102 m 2 /g. For cellulose-templated ZnO and Co 3 O 4 , the total pore volume was 0.15 cm 3 /g and 0.16 cm 3 /g, respectively, whereas they were 0.11 cm 3 /g and 0.10 cm 3 /g for ZnO and Co 3 O 4 without cellulose.
Te results show that using cellulose as a template to prepare ZnO and Co 3 O 4 result in higher total pore volume and specifc surface area. Te pore volume and specifc surface area increment attributed to high-temperature cellulose decomposition and the creation of a porous structure of ZnO and Co 3 O 4 . Te bulk g-C 3 N 4 synthesized without the exfoliation process typically shows a poor surface area (<10 m 2 /g −1 ) [71]. Te sulfuric acid-treated g-C 3 N 4 surface area and pore volumes show 47.136 m 2 /g −1 and 0.09 cm 3 /g, respectively. Te surface area of the GZC-3 composite (60.578 m 2 /g) and the pore volume (0.12 cm 3 /g) increased from bulk g-C 3 N 4 , as shown in Figure 5.

Photocatalytic Performance.
Te photocatalytic MB dye degradation process takes place in the photoreactor with a 250 Watts halogen lamp as the light source. Typically, around 30 mg of the prepared sample is added to a 100 ml (10 ppm) aqueous solution of MB dye. Te catalyst-containing solution was stirred for 30 min in the dark before the photocatalytic degradation test commenced. Tat leads to achieving equilibrium for the adsorption and desorption of dye molecules on the photocatalyst surface, a dark reaction. To separate the solution from the photocatalyst powder, a 4 ml dye solution was taken and centrifuged to separate the catalyst and solution at each time interval. A UV-Vis spectrophotometer was used to measure the concentration of MB dye at each time interval. Te distance between the light source and the solution remained constant at 5 cm. Te MB dye degradation efciency (η) was determined using the following formula: η � ((C o − C t )/Co) * 100%. Te C o and C t are the MB dye concentrations at the initial and each time interval of light irradiation, respectively. All tests have been conducted again to validate the fndings.

Photoluminescence (PL) Analysis.
To observe the recombination of photogenerated charge carriers in the catalyst room temperature, PL spectra were measured at 326 nm excitation wavelength, as shown in Figure 6(a). Te g-C 3 N 4 sample has the maximum PL peak intensity and exhibits wider peak width emission at 419 [43], demonstrating a high rate of electron and hole recombination relative to the synthesized sample. Te formation of GZC composites results in photoluminescence quenching; it might boost photocatalytic activity. Figure 6 illustrates the PL emission spectra of the photocatalysts g-C 3 N 4 , showing the highest peak and the composite GZC's lowest peak. Te peak intensity is from highest to lowest in the order g-C 3 N 4 , GC, GZ, GZC-5, GZC-4, GZC-1, GZC-2, and GZC-3, as shown in Figure 6(a). Te GZC-3 composite shows the lowest PL emission peak intensity among all synthesized catalysts, implying the efective suppression of photogenerated charge carriers from recombination. Tis phenomenon will favor having a large number of electrons and holes engage in the redox process.

Te Nyquist Plots.
Electrochemical impedance spectroscopy (EIS) is a powerful technique for understanding the transport of interface charges [72]. We examined the charge transfer resistances and interfacial charge separation efciency of g-C 3 N 4 and GZC composites by EIS, as shown in Figure 6(b). During interfacial transport, the composites possess the smallest EIS semicircle radius compared to g-C 3 N 4 , implying the lowest impedance, which permits the rapid transfer of charges. Based on the EIS data, the GZC-3 composite catalyst has shown a small radius compared to other samples in this experiment. Small radii correspond to small charge transfer resistance, thereby increasing the effciency of photocatalysts, which complement the PL spectra somewhere above.
3.9. Photocatalytic Activity of the Samples. Te photocatalytic MB dye degradation performance of the synthesized samples was measured using the beer lambert law. Te catalyst-dyecontaining solution was irradiated, and the dye concentration was measured by UV-visible spectroscopy. Te absorbance vs. wavelength graph of the GZC-3 composite catalyst change is shown in Figure 7(a), and the concentration of MB dye over time changed in the presence of the catalyst.
Te blank test indicates that MB dye is only weakly degraded in the absence of a catalyst, noting that photolysis is not an option to degrade the MB dye, as shown in Figure 7 binary and ternary composites of g-C 3 N 4 , ZnO, and Co 3 O 4 exhibit improved photodegradation activity for MB dye degradation. In particular, the ternary composites g-C 3 N 4 / ZnO/Co 3 O 4 show promising performance for the photodegradation of MB. Te GZC-3 composite had the maximum photodegradation activity showing 97.4% MB dye degradation after 60 min of photoreaction.
Te degradation kinetic was studied to highlight the photodegradation activity of the materials, as shown in Figure 7(c). Te pseudo-frst-order kinetic equation -ln (C t / C 0 ) � k app was used to ft the degradation kinetic plots, where k app and t represent the apparent frst-order reaction rate constant and irradiation time, respectively. Table 1 shows the corresponding correlation coefcient values, which are all

Recyclability.
A catalyst for practical use has to examine its photostability in the efuent. Te GZC-3 composite catalyst sample measured the photocatalytic MB dye degradation activity for 60 min. Subsequently, 4.0 ml of the degraded solution has taken to measure the dye concentration, and the remaining degraded solution was centrifuged and separated the catalyst from the solution. Te separated catalyst was washed and cleaned with distilled water and dried in an electric oven at 60°C for 2 hrs. Te recyclability and photostability of the catalyst were measured for four cycles following the procedure. Te catalyst shows excellent stability in degrading MB dye in an aqueous solution, as shown in Figure 8 and Figure S7. However, after the four cycles of the degradation process, negligible fuctuation in the degradation performance was observed, which might be attributed to the loss of catalyst amount while collecting from the centrifuge tube, since the amount of the catalyst added in the dye solution is small. Extrapolating the linear section of the plot to the x-axis shows the fat band potentials (V FB ) of g-C 3 N 4 , ZnO, and Co 3 O 4 as −1.05 V, −0.83 V, and 1.6 V vs. RHE, respectively. Te conduction band edge potential of g-C 3 N 4 , ZnO, and Co 3 O 4 was approximately the same as the V FB [73]. Te valence band edge position of g-C 3 N 4 , ZnO, and Co 3 O 4 is determined to be ∼1.7 V, 2.3 V, and 3.4 V vs. RHE; therefore, the composite may create a junction band structure shown in Figure 9(d).
Up on irradiation to the catalyst in the dye solution, electrons from the valance bands of g-C 3 N 4 , ZnO, and Co 3 O 4 are excited to the conduction band of g-C 3 N 4 , ZnO, and Co 3 O 4 , which creates a large number of charge carriers in the catalyst valance and conduction band. Te electron on the conduction band of cobalt oxides will fall to the valance band of g-C 3 N 4 due to the potential variance; this could result in a Z-scheme electron transfer process [74,75], as shown in Figure 9(d).
Te excited electron on the conduction band of g-C 3 N 4 will transfer to the conduction band of the ZnO. Te Z-scheme improves the photogenerated charge carriers' transfer and separation efciency while maintaining the strong reduction capability of electrons in g-   Te electron in the conduction band of ZnO will participate in the reduction process to produce superoxide (O •− ) radicals, and the holes in the valance bands of ZnO and    Co 3 O 4 will primarily participate in the oxidation reaction to form hydroxyl (OH • ) radicals. ZnO is thermodynamically more feasible to undergo a reduction reaction to produce superoxide (O •− ) radicals due to high negative conduction band potential (−0.8 vs. NHE) compared to (O •− /O 2 ), whereas the hole in the ZnO and Co 3 O 4 valance band are thermodynamically more feasible to undergo oxidation process to hydroxyl (OH • ) radicals due to sufcient positive valance band potentials compared to H + /H 2 O and OH − /OH • [76][77][78]. Tis active species (O •− ) and OH• reacts with MB dye to break down the bond of the dye into nonpolluting molecules such as water, carbon dioxide, and others.

Conclusions
Te Z-scheme ZnO-g-C 3 N 4 -Co 3 O 4 composite was prepared by wet chemical and ultrasonicate-assisted synthesis methods. It has been observed that utilizing cellulose as a template to prepare metal oxide increases the specifc surface area with a distinct morphology that might enhance the density of the active site. Te photocatalytic activity of the GZC composite towards MB dye degradation was better than that of the pure and binary composites. Te GZC-3 composite demonstrated the highest degradation efciency in degrading MB dye, with 97.4% degradation efciency in 60 min under visible light irradiation. All the synthesized composite photocatalysts exhibit improved photocatalytic activity compared to the binary composite and pure ones due to the proper band alignment, the higher specifc surface area, better visible light absorbance, and low charge transfer resistance. Te GZC-3 composite demonstrates excellent stability and degradation efciency, suggesting that GZC is a feasible candidate photocatalytic material for MB dye degradation in an aqueous solution.

Data Availability
Te data used to support the fndings of this study are included within the article, and the supporting information/ supplementary fle is available from the Hindawi Online Library.

Conflicts of Interest
Te authors declare that there are no conficts of interest. Technology University, Ethiopia, for technical expertise support in doing experiments and Google inclusion research support. Figure S1: schematic illustration of the celluloses synthesis process. Figure S2: schematic illustration of the ZnO synthesis process. Figure S3: schematic illustrations of the Co 3 O 4 synthesis process. Figure S4: schematic illustration of the g-C 3 N 4 synthesis process. Figure S5: schematic illustration of the g-C 3 N 4 /ZnO/Co 3 O 4 composite synthesis process. Figure S6: XRD of prepared cellulose. (Supplementary Materials)