Enhanced photocatalytic performance by ZnO/Graphene heterojunction grown on Ni foam for methylene blue removal

: ZnO nanostructures were obtained by electrodeposition on Ni foam where graphene was previously grown by chemical vapor deposition (CVD). The resulting heterostructures were characterized by X-ray diffraction and SEM microscopy and their potential application as a catalyst for the photodegradation of methylene blue (MB) was evaluated. The incorporation of graphene to the Ni substrate increases the amount of deposited ZnO at low potentials in comparison to bare Ni. SEM images show homogeneous growth of ZnO on Ni/G but not on bare Ni foam. A percent removal of almost 60% of MB was achieved by the Ni/G/ZnO sample, which represents a double quantity than the other catalysts proved in this work. The synergistic effects of ZnO-graphene heterojunctions play a key role in achieving better adsorption and photocatalytic performance. The results demonstrate the ease of depositing ZnO on seedless graphene by electrodeposition. The use of the film as a photocatalyst delivers interesting and competitive removal percentages for a potentially scalable degradation process enhanced by a non-toxic compound such as graphene.


Introduction
It is known that water pollution is an environmental problem that worsens year after year.Human activities, such as industry and agricultural production, affect bodies of water and, consequently, human health [1].It is believed that of all wastewaters generated by human activities, 80% are discharged without prior treatment.Poor quality drinking water is a problem that is associated with 80% of childhood illnesses and 50% of child deaths worldwide [2].Among water contaminants, organic dyes play a leading role: around 100,000 different types of dyes are produced annually, totaling more than 700,000 tons, with approximately 100 tons being dumped [3,4].In addition to many of them being carcinogenic and dangerous for humans [5,6], they have good stability in environmental conditions [7] and cause, among other drawbacks, loss of transparency, reduction in the penetration of sunlight, retard biological activity of plants and animals, increase chemical oxygen demand (COD) and biochemical oxygen demand (BOD), etc. [8,9].In this context, organic dyes contribute significantly to this environmental problem, since they are one of the main sources of contamination of surface and groundwater [10].One of the most commonly used dyes is methylene blue (MB).MB, from the thiazine class, is a heterocyclic aromatic compound, see Figure 1 [11].This cationic dye is commonly used in the textile industry, to dye wool, cotton, etc., where 15% is transferred to wastewater during this process, as an industrial pollutant [12][13][14].These dyes have also long been used in medicine and scientific purposes, such as in microscopy or as redox indicators [15].Although this dye can be used as a drug against some diseases, such as malaria, it is also toxic to human health and the environment and can cause vomiting, nausea, irritation and tachycardia among other diseases in humans.This pollutant can inhibit plant growth and reduce the pigment and protein content of algae [16][17][18].For these reasons, the elimination of these pollutants from water is now one of the main areas of study [3].The removal mechanisms of MB and other contaminants from wastewater are frequently studied [19][20][21].The photodegradation via photocatalysis using semiconductors is a preferred way to eliminate contaminants in various sources of water [22,23].One of the main advantages from photocatalysis relies on the generation of non-harmful products, such as CO2, H2O and inorganic salts after activating the process with light [24,25].Although there are various kinds of semiconductors that adapt well to these systems, including Fe2O3, CdS and ZnS [3]; TiO2 and ZnO are the most used in photocatalysis applications [10].Most likely due to their low toxicity, outstanding thermal and chemical stability and relatively low cost [26].In particular, ZnO presents higher percentages of contaminants degradation in water than TiO2 [27], which could be due to the higher exciton binding energy and the higher electrical conductivity of ZnO compared to TiO2, in addition to the ZnO band potentials (of valence and conduction) lower than TiO2 ones.This combination of properties allows an increase in the effectiveness of the degradation reactions [28,29].In this context, the use of thin supported ZnO films is most desired, since it allows a very simple separation process, which consists of removing the catalyst from the solution, also facilitating the recycling of the catalyst.On the other hand, the use of ZnO powder involves separation by some subsequent process.In addition to this, it is known that the photodegradation efficiency increases as the specific surface increases [30].This is why the use of porous supports, such as nickel foams, can obtain very high performance, comparable to powders.
ZnO thin films can be obtained through various methods, such as metal-organic chemical vapor deposition (CVD) [31,32], magnetron sputtering [33,34], molecularbeam epitaxy [35,36], sol-gel [37], spray pyrolysis [38,39] and electrodeposition [40][41][42].What is interesting about the latter is the simple scalability, film control, and the possibility of depositing structures in many different shapes and sizes without using high temperatures or expensive equipment.Growth can be carried out through various solutions that use Zn 2+ salts with different anions, whether chlorides, nitrates or sulfates, as precursors [43][44][45][46][47][48].By varying the type, concentration or pH of these solutions, as well as the applied potential, the deposition time or the current circulating in the cell, various types of ZnO films and morphologies can be obtained [49][50][51][52][53][54].However, it is desirable to use low currents or voltages, low-concentrated salts with non-toxic precursors and low times so that the process is profitable and easily scalable.
One of the ways to improve the properties of ZnO is by forming heterostructures with graphene.CVD Graphene presents outstanding properties such as transparency, flexibility, high carrier mobility and mechanical stability [55][56][57].When graphene is grown in 3D structures the carbon atoms are exposed on the surface, resulting in a large surface area and giving a large number of active sites [58].The three-dimensional ZnO-Graphene heterojunction provides synergetic properties and improves charge separation by ZnO followed by charge transport offered by highly conductive graphene, which dramatically reduces the probability of electron-hole recombination [59,60].This happens because the work function of graphene (−4.5 eV) is lower than the conduction band of the semiconductor oxide (−4.1 eV), so the excited electrons in ZnO are transported toward graphene, which has a high carrier mobility, thus inhibiting recombination in ZnO and consequently, increasing the photocatalysis degradation efficiency of MB [61].Researchers have shown that the photocatalytic activity of ZnO improves whenever graphene is used as a co-catalyst [62].In addition, graphene could extend the absorption range of ZnO to longer wavelengths further from the ultraviolet region [59].These exceptional properties offered by the ZnO-Graphene heterojunction have been applied in various fields including photocatalysis, photodetection, solar cells, etc.
In this work we propose the electrodeposition of ZnO on graphene already grown on a Ni foam by CVD to be used in photodegradation applications.This heterojunction is interesting both as an adsorbent material and for photocatalysis due to the high specific surface area provided by using Ni foam as a substrate, added to the improvement in conductivity and the decrease in electron/hole recombination provided by graphene.

Materials and methods
We proposed to grow ZnO nanostructures by electrodeposition on Ni foam, a low cost and simple technique that had proved to be successful in obtaining ZnO films on different flat substrates like ITO and FTO [40][41][42][43].

Graphene growth in nickel foam
Nickel foam with 1.6 mm thickness and porosity of 87% was purchased from MTI Corp (Richmond, CA, USA).Graphene was grown on Ni foam by CVD method on Ni foam following the protocol described by Messina et al. [63].Briefly, the bare Ni foam was sonicated in acetone for 20 min and placed in a quartz tube under a vacuum pressure of 8 × 10 −5 torr.During the synthesis, a constant flow of H2 was maintained at a rate of 75 mL min −1 .At 950 ℃ a flow of CH4 was introduced at a rate of 35 mL min −1 for 5 min leading to the growth of graphene on the entire surface of Ni.Finally, the furnace was cooled down at a rate of 16 ℃ per minute until reaching room temperature.
The electrolytic cell is a three-electrode set-up where Ni foam, Pt wire, and Ag/AgCl electrodes immersed in 3.0 M KCl acted as working, counter, and reference electrodes, respectively.Figure 2 shows a scheme of the cell.Before electrodeposition, the working electrode was washed with soapy water, sonicated in distilled water for 10 min, and then sonicated in isopropyl alcohol for 10 min.Electrodeposition was performed at constant temperature ( 70

Sample characterization
Cyclic voltammetry and subsequent amperometry were performed using a Teq4 potentiostat from NanoTeq Co., Argentina.X-Ray Diffraction (XRD) patterns were acquired using a Philips PW1710, Panalytical X'Pert PRO diffractometer at 45 kV and 45 mA with monochromatized CuKα radiation in the range of 20° ≤ 2θ ≤ 40° with a step of 0.02 °/s and a grazing angle of 3°.Optical images were obtained through a Leica DM IL LED optical microscope.SEM images were obtained from an FEI Quanta 250 environmental scanning microscope (ESEM) at an operating voltage of 20 kV.The equipment has an X-ray detector, EDAX, through which the chemical composition was studied qualitatively using energy dispersive spectroscopy (EDS).

MB degradation
The photodegradation of MB was evaluated by immersing Ni, Ni/G, Ni/ZnO, and Ni/G/ZnO in 1.3 ppm methylene blue (MB) solution.The solution was kept in the dark for 15 min to reach adsorption equilibrium.Then, it was irradiated with an 8 W fluorescent lamp (BTE Lighting, Argentina) with a wavelength ranging from 250 to 600 nm.The UV lamp was kept on for 165 min and the experiment was carried out at room temperature.Figure 3 shows the experimental set-up.At the indicated times, the sample was removed from the solution and the absorbance of the remaining solution was measured using a Shimadzu UV-2600 spectrometer.

Results
In this section details of the electrodeposition method are presented along with the structural and morphological characterization of the obtained heterostructures.Also, the use of these samples on the MB removal is analyzed.

Cyclic voltammetry (CV) and amperometry
CVs were run from negative from 0 to −1.1 V and back to 0 V at 100 mV/s scan rate.Figure 4 shows only the cathodic sweep that is of interest for the ZnO deposition.The CV reveals significant differences based on the electrode substrate.Our group successfully electrodeposited ZnO on different substrates, including ITO and FTO, by applying potentials between −700 mV and −1000 mV.We identified −800 mV as the optimal potential.Therefore, we applied −800 mV which resulted in a noticeable increase in steep curve for Ni/G as compared to the naked Ni electrode shown in Figure 4. Due to the increase in slope observed in both cases, this potential was chosen and applied for 1 h.After amperometry, the Ni/G sample exhibited a white deposit to the naked eye, while no apparent change in color was observed for Ni foam.

X-Ray diffraction (XRD)
Figure 5 shows the diffractograms for Ni/ZnO and Ni/G/ZnO samples measured at angles between 20° to 40°, a region where the characteristic diffraction peaks of ZnO become more evident after smoothing the original signal using the Savitzky-Golay method.Amperometry was performed for 1 h on both samples.As it can be seen, no characteristic peak for ZnO was detected within the Ni/ZnO sample.This does not necessarily rule out the deposition of ZnO since there may be a low amount of mass that is below the limit of detection.On the other hand, for Ni/G/ZnO samples three peaks clearly evolved corresponding to (100), (002), and (101) crystallographic orientations that matched the peaks for ICSD no.01-080-0074.This clearly confirms the successful deposition of ZnO semiconductor onto the Ni/G sample.Furthermore, there is no evidence of preferential growth in the (002), as the intensity of this peak (at 34.335°) is not dominant.This is consistent with the literature which has also shown predominant (101) peaks using electrodeposition and other techniques [30,[64][65][66].These results suggest that the as-deposited ZnO film does not display a nanocolumnar morphology [67].Additionally, no other diffraction peaks are detected, as Ni exhibits diffraction peaks for 2θ > 40° [68].The technique also allows us to confirm that no other crystalline material was deposited in large quantities, since the diffractograms do not present other peaks, nor amorphous ones, since no bands were observed at low angles.

SEM-EDS
Figure 7 shows SEM images for Ni (A), Ni/G (B), Ni/ZnO (C), and Ni/G/ZnO (D).Notable differences in the surfaces are observed between the electrodeposition of ZnO on bare Ni and on Ni/G using the same deposition parameters.Figure 7B clearly shows the growth of graphene on the surface of bare Ni.It can be seen there is a large amount of graphene in the center of the foam and that, close to the edge of the pores, that decreases.To confirm that it was a different material, it was analyzed with backscattered electrons, since in this type of images the contrast is based on the atomic number.Figure 7D shows a homogeneous growth of ZnO on the Ni/G substrate.EDS measured at various points in the sample indicated 27% and 31% by weight of Zn and O, respectively.In addition, a similar amount of C (31% by weight) was detected, consistent with the growth of graphene.This confirms the successful deposition of ZnO and indicates that graphene remains even after the voltage application in the synthesis.Figure 9 shows the Ni/G/ZnO sample with a higher magnification, where an increase in the exposed surface can be seen due to the irregularity of the deposit.This type of morphology is comparable to those reported in the literature, despite using a different ZnO deposition technique [64,65].Finally, Figure 10 shows a color map obtained by EDS, which shows the location of the evaluated elements.It is confirmed that ZnO is deposited throughout the foam unlike graphene, which is mostly located in the center of it.

Removal of MB from a solution
An absorption spectrum was initially obtained after immersing the samples in 1.3 ppm MB solution for 15 min under dark conditions.In this way, the percentage of MB adsorption in the first min of the experiment was obtained.Then, the lamp was turned on and absorbance measured at 45 and 165 min.To calculate the percentage removal of MB in the solution, we choose to measure the area under the curve for the absorption band located at 662 nm.Finally, Equation ( 1) was applied to obtain the MB removal percentages: MB Removal [%] = (1 − At/A0) 100 (1) where At is the absorption measured at a certain time t and A0 is the initial absorbance.Figure 11 shows the absorbance spectra for each sample measured at the indicated times.Table 1 shows the MB removal percentages for each sample for two different periods of time: while the light was off, that is, the black and red curves indicated in Figure 11, and then once the light was turned on, that is, between the red and green curves of the same figure.It is clear that the Ni/G/ZnO sample exhibits a higher removal percentage than the other samples (almost double) in both time periods, thus improving both adsorption and degradation of the contaminant.The Ni foam and Ni/G foam are characterized by their abundance of microscopic-sized pores, which provides a large surface area leading to high contaminants adsorption capacity.The significant enhancement in adsorption observed in the samples with G/ZnO can be attributed to two concurrent factors.Firstly, morphology of the deposited ZnO increases the contact surface, thereby facilitating increased adsorption.This fact is corroborated by SEM images (Figures 7 and 9) where the nanostructured feature of ZnO is clearly appreciated.Secondly, the pH of the MB solution is 6, which results in the formation of Zn(OH) + on the ZnO surface.In contrast, the functional groups of MB carry a negative charge (−SO3 − ).This allows the formation of an ionic bond that increases the adsorption capacity [69].
Figure 12 shows percentages of MB removal from the solution achieved for each sample after irradiation with UV light represented by the black and green curves in Figure 11.Numerous studies [18,70,71] have corroborated that ZnO serves as a photocatalyst in this reaction.It is particularly noteworthy that the sample composed of Ni/G/ZnO exhibits the greatest efficiency achieving about 60% removal of MB from the solution.
Table 1.MB removal percentages from the solution for each sample before and after light irradiation.

Discussion
This work reports on the growth of ZnO on Ni and Ni/G foams by electrodeposition and its possible application in the removal of dyes in water bodies.In the case of bare Ni foam, ZnO deposition is evidenced by the presence of Zn in EDS analysis, although ZnO growth could not be observed by XRD.On the other hand, when the Ni/G foam is used as a substrate, a large amount of ZnO deposition is observed by both techniques used.Unlike the work of Fei et al. [30], here, the growth of ZnO was achieved in a single step, without the need of adding seeds and performing heat treatments.This behavior could be due to the presence of defects and wrinkles in the graphene, which act as nucleation sites for ZnO.Then growth occurs in all directions.
It is worth highlighting that our starting solution has a low concentration of precursors, compared to the growth solution used by Lv et al. [64].The salt that contains the Zn 2+ , Zn(NO3)2, has a concentration 20 times lower, while the supporting electrolyte KCl has a concentration 48 times lower.This makes our process more affordable and accessible.Furthermore, the application of a potential difference of −800 mV in our synthesis results in a more efficient and less demanding method in terms of equipment, compared to the use of −10 V in the aforementioned study.
Regarding the incorporation of graphene, two benefits were found: in addition to allowing nucleation at lower applied voltages, it increases the percentage of MB removal from the solution (57%).In this context it is interesting to note that graphene is an organic, non-toxic and inexpensive material.The results obtained in this work are close to those obtained by various authors who grow ZnO on different substrates [3,72].Fei et al. [30] do photocatalysis and photoelectrocatalysis using the Ni/ZnO and Ni/ZnO/MoS2 heterojunction.The results obtained with both techniques with the Ni/ZnO electrode are less efficient than our findings doing photocatalysis with the Ni/G/ZnO heterostructure.However, Fei and co-workers achieved higher removal when using the Ni/ZnO/MoS2 heterostructure, although it should be noted that graphene is a no toxic and cheaper compound.
On the other hand, Kulis-Kapuscinska et al. [10] have studied the photodegradation of MB by growing ZnO films by sputtering on Si(100) with a subsequent thermal treatment and have managed to remove 64% of MB from the solution in 540 min.That is, with a more complicated growth method they have achieved a result similar to that obtained in this work with a substantial difference in time: their experiment lasts 9 h while ours lasts 3 h, reducing the cost involved in maintaining the light on.This result is very important from the point of view that the degradation rate is usually a limiting factor for the selection of the photocatalyst.This could be due, in addition to the graphene, to the chosen substrate, since the Ni foam, being so porous, has a large active surface.
The improvements occur mainly due to the use of a nickel foam that increases the active surface compared to other flat substrates [30], added to the incorporation of graphene, which inhibits the recombination of electrons and holes according to the mechanism that can be seen in Figure 13.The conduction band of ZnO (−4.05 eV vs. vacuum) is aligned with the graphene in such a way that the electron excited in the semiconductor can be transferred to the graphene (also taking advantage of the high electron mobility), separating the charges and reducing, thus way, recombination.Then, the electron and the hole interact with the water in the solution: the molecular oxygen is reduced, generating the superoxide anion O2 − and the water is oxidized to obtain hydroxyl radicals (OH*), which finally degrades the MB and CO2 and H2O are obtained as products of this [59,62,73].

Conclusion
In this study, ZnO was successfully electrodeposited on Ni/G foams substrates to obtain new materials with high specific surface area.Large amounts of ZnO were observed on Ni/G due to the presence of graphene whose defects acted as nucleation sites.SEM images confirmed a homogeneous distribution of ZnO nanorods along the substrate.Instead scarce deposition was observed for the bare Ni foam.The combined presence of ZnO and graphene increased the MB removal capacity, giving around 60% higher than what is achieved with the other heterostructures.The percentage improvement in terms of adsorption is due to the ionic bond that occurs between the ZnO surface and the negatively charged functional groups of the MB, while in photocatalysis it is probably due to a decrease in electron/hole recombination that leads to better charge separation Furthermore, the electrodeposition technique is a highly scalable method, so the synergistic effects of G/ZnO heterojunctions on a porous substrate becomes an interesting alternative for outperforming in areas of water cleaning and environmental remediation.
This study highlights the importance of the efficient synthesis of ZnO on Ni/G substrates via electrodeposition, not only for its applications in environmental remediation but also, for its contribution to the development of sustainable practices within the framework of green chemistry.
Author contributions: Conceptualization, LCD; methodology, LCD, LFM, MVG; validation, LCD, MVG and LFM; formal analysis, LFM and LJ; investigation, LFM; resources, LCD, FJI and MM; data curation, LFM, MVG and LJ; writing-original draft preparation, LCD and LFM; writing-review and editing, LCD, FJI, LFM and MVG; visualization, LFM; supervision, LCD and FJI; project administration, LCD and MM; funding acquisition, LCD and FJI.All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by CONICET, grant numbers PIP0901 and PIP 0001 and UNLP grant numbers 11X933 and X-887.
℃) and stirring (200 rpm) to ensure homogenous distribution.The applied voltage was −800 mV (chosen after performing Cyclic Voltammetry on the studied system) for 60 min.The initial pH of the solution is 5.3.The samples obtained were called Ni/G (Ni foam + graphene), Ni/ZnO (ZnO on Ni foam) and Ni/G/ZnO (ZnO on Ni foam with graphene).

Figure 2 .
Figure 2. Scheme represents the three-electrode set-up under the parameters used in this work.

Figure 3 .
Figure 3. Scheme of the experimental set-up used to measure MB adsorption and degradation.

Figure 5 .
Figure 5. XRD Diffractogram comparing Ni/ZnO and Ni/G/ZnO samples.Blue lines that correspond to card no.01-080-0074 are used as a comparison.

Figure
Figure 6A exhibits multiple pores within the naked Ni foam structure.No apparent changes were detected for graphene grown on Ni foam (not shown) most likely due to the poor resolution of the optical microscope.Figure 6B,C clearly exhibits the as-deposited ZnO, however the amount of deposit seems not to be uniform along the entire Ni/G/ZnO sample.
Figure 6A exhibits multiple pores within the naked Ni foam structure.No apparent changes were detected for graphene grown on Ni foam (not shown) most likely due to the poor resolution of the optical microscope.Figure 6B,C clearly exhibits the as-deposited ZnO, however the amount of deposit seems not to be uniform along the entire Ni/G/ZnO sample.

Figure 8
Figure8shows a SEM image of the Ni/G sample obtained by backscattered electrons.In these images the contrast is achieved with the difference in atomic number, which allows us to confirm that they are two different materials, since the Ni substrate appears bright while the graphene deposit appears dark.This confirms that

Figure 8 .
Figure 8. SEM image of the Ni/G sample obtained by backscattered electrons.

Figure 9 .
Figure 9. SEM image of the Ni/G/ZnO sample at higher magnification, showing the uniformity of the deposit.

Figure 10 .
Figure 10.EDS color map showing the location of Ni, C, Zn, and O elements within the Ni/G/ZnO sample.

Figure 12 .
Figure 12.Percent removal of MB achieved by the different samples.

Figure 13 .
Figure 13.Scheme of the process with the ZnO and graphene bands.