Highly Efficient Photocatalysis by Zinc Oxide-Reduced Graphene Oxide (ZnO-rGO) Composite Synthesized via One-Pot Room-Temperature Chemical Deposition Method

We synthesized zinc oxide-reduced graphene oxide (ZnO-rGO) composites using a one-pot chemical deposition method at room temperature. Zinc powder and graphene oxide (GO) of different mass ratios (1 :1, 1 : 2, 1 : 5, 1 :10, and 1 : 20 GO to Zn) were used as precursors in a mildly alkaline solution. UV-Vis spectroscopy was used to study the photocatalytic efficiency of the samples through the photodegradation of methylene blue (MB). UV-Vis measurements show the fast decomposition of methylene blue under UV light illumination with the best degradation efficiency of 97.7%within one hour, achieved with sample ZG2 (1 GO : 2 Zn mass ratio). &e corresponding degradation rate was kZG2 � 0.1253min , which is at least 5.5 times better than other existing works using hydrothermal methods. We argue that the excellent photodegradation of MB by ZG2 is due to the efficient charge separation brought about by the electronic interaction of the rGO with the ZnO and the formation of a Zn-O-C bond, as supported by XRD and Raman spectroscopy measurements.


Introduction
e discovery of new materials has led to the development of various applications in the field of environmental remediation. Among the immediate concern for remediation is the treatment of pollutants in waterways and bodies of water as increasing awareness to limited water resources arise. Methylene blue (MB) is one of the most extensively researched dyes for photodegradation. MB dye is used in textile and paper industries and in the medical field. Exposure to high doses and photoactivated MB has been found to have adverse effects to biological species [1].
For water pollution treatment, one of the most studied photocatalytic materials is zinc oxide (ZnO). Its relative large energy bandgap of 3.3 eV generates a sufficient number of electrons and holes upon illumination that facilitates the degradation of certain pollutants. e disadvantage of using ZnO by itself is that it corrodes under illumination [2], which limits its practical utilization. By synthesizing a ZnOgraphene composite, photocorrosion suppression of ZnO nanoparticles is achieved [3]. e surface hybridization of ZnO with carbon layers not only suppressed the photocorrosion but also enhanced the photocatalytic activity of the ZnO. ey reported an 85% degradation of MB when irradiated with UV light for 75 minutes. Many reports have also shown improved photocatalytic performance using ZnO and graphene composites. An enhanced photocatalytic degradation of MB using reduced graphene oxide (rGO) and ZnO composite, synthesized via a simple one-pot hydrothermal method, was reported [4]. e method resulted in rGO sheets covering the ZnO flowers, resulting in the enhancement of photocatalysis as much as 97% with a rate constant of 0.0395 min − 1 . e enhancement was attributed to the decrease in the size of the ZnO flower and to the apparent interaction between the RGO sheets and the ZnO flowers. e limitation in potentially attaining maximum photocatalytic performance was the amount of rGO that can wrap the ZnO flowers. e relationship between the size of the ZnO flower and the photocatalytic performance has also been reported [5].
e ZnO-rGO composite was synthesized also with a onepot hydrothermal method by mixing ultrasonically exfoliated graphene oxide with an amine compound and an inorganic zinc compound in solution. is led to the sufficient removal of oxygen functionalities leaving graphene sheets, where the ZnO nanoparticles then grow. e results are comparable to the work described in the previous paragraph, owing to the synergistic interaction between the ZnO nanoflowers and the graphene and also to the high conduction of charges through graphene sheets.
In another work [6], G/ZnO nanocomposites were synthesized through a two-step electrochemical technique. Photocatalytic activity of the annealed G/ZnO was found to be improved as compared to ZnO and as-synthesized G/ ZnO. rough the modified Kubelka-Munk function, energy band gaps have also been reported to be less than 3.37 eV. Similar to mentioned studies, enhanced photocatalytic activity was traced to the interaction of ZnO with graphene sheets.
Liu et al. [7] used one-pot hydrothermal synthesis of ZnO-reduced graphene oxide composites to study photocatalysis, by utilizing Zn powder as both reducing agent and precursor of ZnO and a surfactant, which reportedly promotes the reduction and dispersion of GO. ey exhibited a MB degradation of 100% in 90 minutes for the ZnO-rGO composite, even without the use of the surfactant.
In this work, we utilized a one-pot chemical method for the synthesis of ZnO-reduced graphene oxide using zinc powder as a precursor without any thermal treatment. Such method has been used before for the development of supercapacitors [8,9]. e method is simple and straightforward, does not require the use of any complex zinc precursors, and can be done at room temperature. We will also show that the oxidation of zinc and the simultaneous reduction of graphene oxide lead to a high-quality zinc oxide and reduced graphene oxide composite that acts as an excellent and highly efficient photocatalytic material. Furthermore, we will demonstrate that the bonding of zinc to the oxygen in graphene oxide and the synergistic effect between ZnO and the carbon layers is mainly responsible for the photocatalytic enhancement.

Materials and Methods
e method described below is simple and straightforward, requires relatively little processing time, and is easily scalable. It is also an environmentally friendly method as no thermal treatment was used to reduce the graphene oxide and no zinc compounds and inorganic compounds were used as zinc oxide precursors.
Graphene oxide (GO) was synthesized via modified Hummer's method [10]. It was then dispersed in 100 mL of distilled water (1.5 mg GO/mL H 2 O) and ultrasonicated for 5 minutes to homogenize the dispersion. At ambient conditions, 100 mL of 6 weight % ammonia solution (Qualikems) was added and mixed with the GO suspension. Afterwards, zinc powder was slowly added to the suspension and mixed, in varying GO to Zn ratios. As the Zn powder (HiMedia) was added, the color of the suspension gradually changed from brown to black. is is a clear indication of the reduction of GO [9]. e samples were named ZG1, ZG2, ZG3, ZG4, and ZG5 corresponding to a Zn : GO mass ratio of 1 : 1, 2 : 1, 5 : 1, 10 : 1, and 20 : 1, respectively. e resulting mixture was then ultrasonicated for 1 hr. Subsequently, potassium hydroxide (KOH) pellets (RCI Labscan, 85%) were added to the suspension without stirring and left 24 hours at room temperature.
e resulting precipitate was then filtered and washed with distilled water and ethanol (RCI Labscan, 99.9%), and the filtrates were dried in an oven at 60°C for 24 hrs.
Scanning electron microscopy with a JEOL JSM-6010 LV was used to study the morphology of the powder samples. e crystallinity and degree of reduction was analyzed with a Rigaku SmartLab 9 kW X-ray diffractometer with Cu Kα1 (λ �1.54056Å) at the National Institute for Material Science in Japan and Raman spectroscopy using an Argon laser source (λ � 514.5 nm) at the National Institute of Physics at the University of the Philippines, Diliman, Quezon City, Philippines.
e photocatalytic activity of the ZnO-rGO composites was evaluated by the decomposition of methylene blue (MB) (Ajax Finechem) under ultraviolet light. e reactor was prepared by adding 20 mg of the synthesized ZnO-rGO to 50 mL of MB (20 μM). Prior to UV irradiation, the solution has been stirred using a magnetic stirrer for 30 minutes in the dark to attain its adsorption-desorption equilibrium. Afterwards, the solution was irradiated with UV light with continuous magnetic stirring inside an aluminum foil-covered chamber. A commercially available UV light source (black light type) was used with an approximate power of 12 W. A Shimadzu UV-mini 1240 UV-Vis spectrophotometer was used to obtain the spectra of MB from 200 nm to 700 nm, beginning at the time when it achieved its adsorption-desorption equilibrium and every 30 minutes thereafter for two hours. At each time, 3 mL of the solution was collected from the reactor. is was done for all five samples of ZnO-rGO composites and for the GO sample. Figure 1 shows the photocatalytic degradation of methylene blue (MB) by UV light using zinc oxide-reduced graphene oxide composite as a function of time.

Photocatalytic Degradation of Methylene Blue.
For comparison, commercially available zinc oxide was also used in the experiment. To quantify the photocatalytic activity, the following expression for efficiency was used [11]: Journal of Nanotechnology where A 0 is the absorbance of MB in the dark and A t is the absorbance of MB under UV illumination for a time t (mins). e absorbance of MB at 664 nm was used as reference for measuring the initial concentration of the solution.
roughout the tests, the absorbance peak of MB shifts to a lower wavelength which suggests dye degradation resulting in altered chemical composition.
is change in the chemistry of the system is due to redox reactions facilitated by photogenerated holes and electrons. Table 1 shows the location of the absorbance peak of MB for each time interval. e most efficient photocatalytic degradation of MB was observed with ZG2 having degraded 97.7% of the dye within one hour. To the best of our knowledge, this is a record efficiency for photocatalytic degradation of MB with ZnO-rGO composites synthesized through an environmentally friendly, low-cost, and highly-scalable method. e efficiency for sample ZG1 was nearly the same at 96.4% within one hour. After the first hour, samples ZG1 and ZG2 had very little response to UV illumination.
ere was a remarkable decrease in efficiency within one hour for ZG3, ZG4, and ZG5 considering that the zinc precursor content was significantly increased as compared to ZG1 and ZG2. Zn (4.534 m 2 /g) [12] has a smaller effective surface area as compared to previously reported ZnO-graphene and ZnOgraphene oxide composites, with specific surface area ranging from 16 m 2 /g to 158 m 2 /g [13]. Increasing the amount of Zn precursor decreases the effective surface area of the samples, which in turn decreases the efficiency of the samples ZG3, ZG4, and ZG5 for photocatalysis.
Moreover, in the work reported by Mills et al. [14] on photobleaching of methylene blue (MB) using TiO 2 , they identified the absorbance peak for colorless leuco methylene blue (LMB) in aqueous solution at λ � 262 nm. In this study, they have observed a decrease in the intensity peaks of MB (660 nm and 292 nm) and an increasing intensity in the absorbance peak of LMB which, according to the authors, indicates the reduction of MB to LMB. In our study, no peak was recorded for LMB at any interval of UV irradiation. is indicates that MB underwent complete degradation. e complete absorbance spectra of MB for different samples are shown in Supplementary Figure 1S.
Using a first-order kinetic model relating the concentration to the reaction time as depicted in the following equation [15], the photocatalytic rate constant, k, was determined: e kinetic rates corresponding to each sample were obtained directly from a regression analysis of the linear  relation of ln (C 0 /C) versus t, as shown in Figure 1(b). Sample ZG2 exhibited the highest rate constant followed closely by sample ZG1 (see the Supplementary Table 1S for the list of the rate constants for all samples). As the zinc precursor was increased further, the kinetic rate decreased considerably by forty times. is was followed by a gradual increase in the kinetic rate as more zinc metal was introduced.
It is interesting to observe that an efficient photocatalytic reaction was obtained at an optimum 33 wt% GO in sample ZG2. A previous investigation done by Xu et al. [4] showed that increasing the GO content from 1 wt% to 3 wt% enhances the photocatalytic degradation of MB to 95% in 80 minutes corresponding to a rate constant of 0.0395 min − 1 .
ey attribute this achievement to the decreasing ZnO flower size with increasing GO content and to the electronic interaction between the reduced graphene oxide sheet and the ZnO flowers. However, their method prevents further photocatalytic degradation since rGO already covers the underlying ZnO flowers at 3 wt% GO. Following their results, increasing the GO content may lead to an increase in the kinetic rate and to further enhancement of photocatalytic performance. In our experiments, increasing the GO content was made possible by introducing the zinc metal onto the underlying GO substrate. However, the dependence of the photocatalytic efficiency to the GO content is counterintuitive. As the GO content was increased from 4.8 wt% (ZG5) to 9.1 wt% (ZG4) to 17 wt% (ZG3), the efficiency after 2 hours decreased from 29.2% to 18.0% to 13.9%, respectively. e proposed mechanism is a combination of photodegradation and adsorption to rGO/GO sheets. As reported, methylene blue is strongly adsorbed on the GO surface [16]. e first few intervals of UV irradiation of MB solution using ZG1 and ZG2 was dominated by the adsorption of MB molecules on the photocatalyst, as indicated by the minimal redshift of the absorbance peak for MB but accompanied by a large difference in absorbance intensity. From these observations, a two-step photodegradation process for the degradation of MB is proposed. is mechanism shows two steps in the photodegradation of MB by ZG1 and ZG2: (1) adsorption of MB molecules on the catalyst and (2) reaction of hydroxyl radicals with MB molecules to decompose or degrade the dye. is proposed mechanism is in accordance with an earlier report where graphene sheet acts as new reaction sites [17].

Scanning Electron Micrographs. At low magnification
(not shown), GO appears as compacted sheets. Upon closer inspection at a higher magnification, each GO sheet appears to be relatively smooth (Figure 2(a)). e surface of the zinc powder (Figure 2(b)) is also relatively smooth but contains smaller zinc particles.
e SEM image of sample ZG2 (Figure 2(c)) shows a thin layer of rGO covering, which appears to be a zinc metal particle. Small rod-like structures, approximately 0.2 µm in diameter, appear on the rGO sheet. Even smaller particles are present on the surface of the zinc metal, which does not appear on the surfaces of the zinc powder. e morphology of sample ZG5 (Figure 2(d)) exhibits the same smaller particles present on the surface of the zinc powder in ZG2, as well as larger particles, which are presumably zinc oxide. ere is no indication of any rGO found on these surfaces for ZG5.

X-Ray Diffractometry and Raman Spectroscopy Analysis.
In order to understand the origin of the enhanced photocatalysis, we analyze the observations from XRD and Raman spectroscopy. Figure 3 shows the Raman spectra for all samples. Raman active peaks for both ZnO and GO can be observed in Figure 3(a). e most prominent peak at 426-429 cm − 1 for zinc-containing samples ZG1 to ZG5 is attributed to the E2 (high) vibrational mode corresponding to vibrations along the oxygen sublattices [18]. e presence of this peak clearly indicates the formation of hexagonal wurtzite ZnO in all five samples, which agrees with the XRD results. e shift to lower wave numbers from the frequency of 437 cm − 1 corresponding to the E2 (high) mode in bulk ZnO may be due to the effect of defects as a result of the finite size of the crystalline domain [18]. e peak at 575 cm − 1 is also observed for all zinc-containing samples. is corresponds to the A1 longitudinal optical phonon mode for ZnO [A1(LO)].
Two more peaks are seen in Figure 3(a) that is also attributed to ZnO vibrational modes. e peak found in all samples between 329 and 333 cm − 1 corresponds to the difference, E2 (high)-E2 (low) [19,20]. e peak at 1140 cm − 1 found in all samples, except ZG2 where it is found at 1129 cm − 1 , is assigned to the second-order longitudinal optical phonon mode [2-LO] characteristics of II-IV semiconductors [21,22] (see Supplementary Figure 2S for the Raman spectra of the samples for the range corresponding to ZnO vibrational modes). In general, the relative intensities of these peaks for ZnO relative to rGO follow an increasing trend with increasing zinc content for all samples except ZG2. In addition to the ZG2 redshift, there is an abrupt increase in relative intensity from ZG1 to ZG2 and then a decrease in relative intensity from ZG2 to ZG3. A study on the first-and second-order Raman spectra using density functional theory (DFT) [20] shows that the 2-LO phonon modes should be found at 1158 cm − 1 .
All samples exhibited a redshift, as shown in Figure 4. It was suggested that the redshift might be due to the presence of defects and impurities. Since the relative intensity of the E2 (high) mode increased, the redshift cannot be attributed to intrinsic defects but rather implies the presence of extrinsic defects. Such redshift has been reported before in previous works where Mn was substituted for Zn [23] and Zr for Zn [24]. In this work, the anomalous behaviour exhibited by sample ZG2 may be similarly explained.
e Raman active peaks for GO are observed at 1359 cm − 1 and 1603 cm − 1 corresponding to the D-and G-bands, respectively [25][26][27]. e blue shift of the G-band from the ideal 1580 cm − 1 for graphite corresponds to the formation of nanocrystalline graphite [28]. It is possible that the G and D' bands overlap. e D' band is associated with crystal defects [29]. Following the assumption that the G and D' bands overlap, then as zinc was added, the G-band shifted to lower frequencies with the peak at around 1577 cm − 1 , while the D' band peak increases slightly to 1611 cm − 1 for samples ZG1 to ZG4. e G and the D' band peaks then redshifts, respectively, to 1570 cm − 1 and 1600 cm − 1 for sample ZG5. e separation between the G and D' bands, shown in Figure 2(b), may be attributed to either strain on the graphene sheet [30,31] or chemical doping [32,33]. We also observe the presence of the 2D and S3 band peaks at 2685 cm − 1 and 2930 cm − 1 , respectively, for sample ZG1. e 2D peak, sometimes referred to as the G' peak, and the S3 peak are both associated with defects in graphene. ese defects are due to the edges of the graphene planes as a result of the reduction of graphene oxide [25,26], which is consistent with the SEM results. e ratio of the intensities of the peaks of the D-band and G-band, i.e., I D /I G , describes the quality of graphene relative to the presence of in-plane defects. From ZG1 to ZG2, the ratio decreased implying an improved crystalline quality of graphene as the zinc content was increased twice. Increasing the zinc metal precursor further from ZG2 to ZG3 increases the intensity ratio indicating a decreased crystalline quality of graphene. As the zinc content was further increased from ZG3 to ZG5, the I D /I G ratio decreased. However, this does not necessarily mean improved crystallinity. As the zinc content was increased, the full-width at half maximum of the D-band increased and the intensities of the D-and G-bands relative to the zinc oxide peak intensities decreased. e most probable reason for this is that as the zinc metal precursor is increased, more and more zinc oxide is formed on the surface of the zinc metal that is covering the GO resulting in the decreased Raman response from the rGO sheets.
X-ray diffraction measurements were performed in order to confirm the presence of ZnO in our samples and to further study the reduction of graphene oxide. Figure 5 shows the characteristic peaks at 31.7°, 34.4°, 36.2°, 47.5°, and 56.5°for hexagonal wurtzite ZnO (JCPDS PDF Card 00-036-1451) corresponding to reflections from the 100, 002, 101, 102, and 110 planes. e peak at 56.5°corresponding to the 110 plane is observed to have a higher intensity for ZG4 which suggests that the thickness of ZnO is increased for this sample. e peaks were sharp suggesting the very good crystallinity of the synthesized zinc oxide, consistent with the Raman spectroscopy measurements. e characteristic peaks for graphene oxide (GO) at around 2θ � 9.8°and 19.2°for the 001 and 002 planes are also observed in Figure 5. e synthesized GO also contained a small amount of graphite, as evidenced by the peak at 2θ � 42.1° [34]. is peak became less evident with increasing zinc content, indicating that the graphene sheets became functionalized. e complete list of lattice parameters a and c for ZnO and the interlayer spacing for GO 001 and rGO was calculated using Bragg's equation and is shown in Table 2. e lattice constant decreased, as the zinc metal precursor is increased, approaching a value corresponding to that of single crystal ZnO. On the contrary, the lattice constant c has a maximum value for sample ZG2. is seems to suggest a direct chemical bond between the ZnO and GO, but this does not explain the increasing GO lattice spacing with increasing zinc content.

Discussion
ree unusual behaviours in the reduction of GO are discussed. In typical reduction processes, such as thermal [35,36], hydrothermal [7] or chemical [9,16,27], GO is completely reduced as indicated by the shift to higher values of a single characteristic 2θ peak. In this work, two sets of peaks are observed at around 6.9°to 8.6°and 23.7°to 24.2°in the spectra of samples ZG1 to ZG5. In general, the rst set (peaks assigned with open circles) is attributed to GO but shifted towards lower values of 2θ with increasing zinc content. is implies a gradual increase in the interplanar spacing of the GO sheets from 9.05Å to 12.26Å. e second set (peaks assigned with closed circles) is ascribed to reduced graphene oxide (rGO), consistent with previous results [25,37]. is is attributed to the formation of the graphene layers as a result of the breaking of the Van der Waals forces between graphene sheets in the composites and to the removal of oxygen functionalities and adsorbed water [9,10,34]. is supports the idea that Zn acts a reducing agent for GO. e second atypical behaviour is the presence of a small peak at 2θ 13.0°for samples ZG2 and ZG3. is may be attributed to the presence of water intercalated between the GO sheets, which imply a partial reduction of the GO for these two samples [34]. e third is the abrupt increase of the GO interlayer spacing to 12.76Å in sample ZG2. However, the presence of adsorbed water molecules between the GO sheets cannot explain this behaviour exhibited by sample ZG2 and the corresponding excellent photocatalytic activity in Figure 1.
As seen in Figure 1, ZG2 and ZG3 have an e ciency of 97.7% and 5.7%, respectively, after one hour. It is natural to expect that the e ciency should increase with increasing amount of zinc precursors. is discontinuous trend can be explained by taking note that as more zinc is added, more of the GO is covered with it. With a relatively small zinc content, ZnO is formed when zinc bonds with oxygen in GO. e role of GO is important in the formation of ZnO, as evidenced the existence of the two XRD peaks corresponding to both GO and rGO. e oxygen in GO bonds with the zinc in the precursor results in the rst 2θ peak, and the apparent reduction of GO results in the second 2θ peak. ese two peaks are clearest in samples ZG1 and ZG2. e reason for the slightly reduced photocatalytic e ciency of sample ZG1 is due to the presence of defects as evidenced by the existence of the 2D and S3 Raman peaks and by the higher I D /I G ratio.
Raman measurements also show a ZnO phase decrease and GO phase increase with increasing relative GO content. Since the amount of GO was the same in all the samples, the decrease in the detection of the XRD re ections from the GO and rGO planes and of the Raman vibrational modes due to GO and rGO from ZG3 to ZG5 is attributed to increasing amount of zinc covering the underlying GO. e increased amount of zinc also led to an increased amount of ZnO, as exhibited in both Raman and XRD measurements. Furthermore, the 001 GO peak shifts slightly towards higher values of 2θ as the zinc content is decreased (Table 1). It can also be noticed that the ratio I D /I G in the Raman spectroscopy measurements increases with decreasing zinc content.
is implies increasing number of defects with decreasing zinc metal precursor, which consequently decreased the probability of formation of ZnO. In addition, with high formation probability follows an increase in the interlayer spacing of GO. ese analyses suggest three things. e rst is that the zinc chemically bonds with the oxygen on the surface of graphene forming the rGO. e idea that zinc oxide is formed through a chemical bond with oxygen-containing functional groups (OFGs) on graphene was rst presented by Ma et al. [38]. e OFGs served as nucleation site for a ZnO crystallite. Eventually, each of the ZnO crystallites combines to form a ZnO lm. In this experiment, zinc on the metal surface readily bonded with the OFGs, forming Zn-O-C. Secondly, the photocatalytic degradation of MB is primarily due to the e ect of ZnO. We surmise that the presence of rGO only improves the degradation e ciency by presumably preventing electron-hole recombination. e third point is that the formation of the Zn-O-C bond greatly enhances the photocatalysis of ZnO.
is speculation requires further investigation.
One possible disadvantage in mechanical mixing is that when ZnO is formed in the solution, it may not directly bind with graphene. e interaction of ZnO with graphene is perhaps more of a dipole-dipole interaction rather than a direct charge transfer. It has been shown that both Znterminated polar surface of ZnO induces a dipole in the charge distribution of graphene [39]. While previous works describing photocatalysis of ZnO-graphene composites show that is comparatively e cient with this work, the kinetic rate could bene t more from a direct charge transfer rather than charge induction. is hypothesis needs to be veri ed further through a more detailed experiment in the future.
e GO enhancing the photocatalytic activity of ZnO also hints toward a better material for dye-sensitized solar cells. e Zn-O-C bonding may yet prove to perform better in terms of electron transfer kinetics between the metal oxide and the conducting electrode. To verify whether there is an indication of such chemical bonds, we performed thermal gravimetry (TG) and di erential thermal analysis (DTA) on sample ZG2 and compared it with the zinc metal powder used as the precursor and a commercially available ZnO. e results for the TG and DTA measurements are shown in Figure 6. All samples exhibited a mass loss below 100°C, which corresponds to the removal of water molecules. Two endothermic peaks can be observed in the DTA measurements for the zinc metal sample. Two endothermic peaks at 253°C and 428°C are attributed to the phase transformation of zinc from solid to liquid. After the second endothermic process, even as the temperature is increased, we observe no mass loss. Instead, an increase in the sample mass was detected starting at 600°C. is exothermic process may be attributed to the possible formation of ZnO.
e ZnO sample did not show any signi cant weight loss.
is is understandable due its very high melting point near 2000°C. e exothermic peak exhibited at T > 600°C corresponds to the further crystallization of the hexagonal wurtzite phase.
is is consistent with results from the existing literature [40]. e mass loss of about 11% for sample ZG2 between 100°C and 300°C corresponds to the phase transformation of zinc at 244°C and the combustion of GO  crystalline carbon sheets with lattice defects [34]. Further increase in temperature exhibits the phase transition of zinc at 423°C and the combustion of carbon between 500°C and 600°C, which agree very well with literature [7]. At T > 600°C, the remaining 33% was zinc oxide, as indicated by the constant mass and the exothermic process leading to the crystallization of ZnO.
ese results imply two things. First is that the reduction of GO resulted in the oxidation of zinc, as suggested by the remaining 33% ZnO. Sample ZG2 was synthesized with precursors composed of 67% zinc and 33% GO. e formation of ZnO, or presumably Zn-O-C, is therefore more favorable when GO is used as an oxidation substrate, compared to its formation in a basic solution. is interpretation supports previous work, where it was shown that GO acted as a nucleation seed and adhesion promotion agent for the formation of ZnO films [35]. Secondly, the idea that GO promotes ZnO formation in turn suggests that ZnO may be intercalated between graphene sheets, supported by XRD and Raman results.
Published work also suggests that the encapsulation of graphitic layers by ZnO suppresses the photocorrosion of ZnO [3]. e photocorrosion occurs when h + reacts with ZnO forming Zn (2+) and oxygen. With the carbon layers, the  Figure 5: (a) e characteristic peaks for graphene oxide ( ), reduced graphene oxide ( ), and water intercalated between graphene sheets ( * ) with increasing zinc content are shown. e spectra for samples ZG2 to ZG5 have been magnified for clarity, as indicated by the numbers on the rightmost part. (b) e characteristic spectra for zinc oxide exhibit a hexagonal wurtzite structure. e peaks for zinc precursor are also present.

Conclusions
We synthesized zinc oxide-reduced graphene oxide via onepot chemical deposition using zinc powder as precursor at room temperature and characterized the photocatalytic performance of the composite. e most efficient photocatalytic degradation of MB was observed with sample ZG2 having degraded 97.7% of the dye within one hour, which corresponds to a kinetic rate of degradation of 0.125/min. To the best of our knowledge, this is a record efficiency for photocatalytic degradation of MB with ZnO-rGO composites.
We argue that the possible formation of Zn-O-C and the synergistic interaction between ZnO and the graphene layers led to the enhancement of photocatalytic performance.

Data Availability
e characterization data and experimental results used to support the findings of this study are included within the article and in the supplementary file.