Effect of Cu doping on the structural, optical and electrical properties of ZnO thin films grown by RF magnetron sputtering: application to solar photocatalysis

RF sputtered undoped and Cu doped ZnO (CZO) thin films were deposited on unheated glass substrates using a mixed Cu2O and ZnO powders target at different Cu concentrations of 0, 1, 2, 3 and 4 wt%. The effects of copper concentration on the structural, electrical, optical and photocatalytic properties of CZO films have been studied. From XRD and Raman spectroscopy studies, it was found that the deposited films were polycrystalline with a predominant hexagonal wurtzite structure along the c-axis perpendicular to the substrate surface. The presence of multiple interference fringes in the transmittance and reflectance spectra shows the good homogeneity of the films. All the films are highly transparent with transparency reaching 80% indicating the possibility to use these films as an optical window. The absorption tail gradually shifted towards a higher wavelength side, which resulted in the decrease of bandgap energy from 3.35 to 3.26 eV. All the sputtered films are highly conductive with a conductivity reaching 104 S.cm−1.The effect of Cu-doping on the photocatalytic activity of ZnO thin films for the degradation of methylene blue (MB) dye was studied under sunlight irradiation and the results showed that the Cudoping provokes appreciable degradation of MB and reached a maximum for the 1 wt% Cu doped ZnO film.


Introduction
Over the past three decades, oxide semiconductor thin films, and particularly zinc oxide have been the subject of a great deal of research (Kennedy et al. (2014); Kennedy et al. 2013;Kennedy et al. 2010;Kennedy et al. 2006;Kaviyarasu et al. 2019). Zinc Oxide (ZnO) is a semiconductor with interesting electrical and optical properties. The importance of its binding exciton energy (60 meV), and the width of its bandgap (3.37 eV), makes it a good candidate for applications in different technological sectors, notably in optoelectronics as transparent electrodes and ultraviolet detectors or laser diode emitting in blue or ultraviolet, in photovoltaic for production of solar cells, in piezoelectric for mechanical detectors, in acoustic wave devices and environmental applications (Linhua et al. 2020a(Linhua et al. , b, 2019Yang et al. 2019;Kaźmierczak-Bałata et al. 2020). Doping of suitable elements in ZnO shows different functions and improved, in particular, its electrical and optical properties, which facilitate the development of many electronic and optoelectronic devices (Andolsi and Chaabouni 2020;Li et al. 2020;Ali et al. 2020). It is expected that doping can enhance the diversity of applications of this oxide. Among the doping impurity elements, Copper Cu is an attractive dopant due to its high conductivity and being relatively abundant on the Earth's crust. Also, doping with Cu helps in obtaining p-type ZnO. Photocatalytic properties of ZnO can be tailored by doping with foreign elements in ZnO such as Cu (Narayanan and Deepak 2018), Fe (Nithyaa Sree et al. 2019), Ni , Co (Poornaprakash et al. 2020), and N (Numan Salah et al. 2016). Among these doping elements, copper Cu is one of the doping metals of interest because it is widely used as a catalyst for many reactions and can have a significant influence on the photocatalytic performance of materials. Different technological processes can be used to deposit ZnO thin films, including Spray pyrolysis, Sol-Gel, Laser Ablation and Magnetron Sputtering (Benramachea et al. 2014;Vishwasa et al. 2012;Alex 2012;Al-Hardan et al. 2011).
In this study, our research efforts are focused on the investigation of the effect of Cu doping on the structural, optical, electrical and photocatalytic efficiency of ZnO films deposited on a glass substrate at room temperature by a simple and inexpensive RF sputtering technique using raw powder targets as an alternative to the typical sputter target, where a relatively expensive ZnO sintered target was used. The photocatalytic activities were tested for the degradation of methylene blue (BM) under sunlight irradiations. All obtained results are correlated, and the enhanced photocatalytic activity is interpreted based on structural, optical and electrical investigations.

Experimental details
Undoped and Cu-doped ZnO (CZO) thin films were deposited at room temperature onto the glass substrates by RF magnetron sputtering technique using a raw powder target. The 99.99% pure Cu 2 O and ZnO powders with particle sizes in the range of 100 nm to 1 µm were mixed with the mass ratio of 1, 2, 3 and 4 wt% and blended in a rotatable drum. The blended powders were spread onto a copper plate and lightly tamped down to produce a uniform thickness and surface to target distance. No further processes (such as sintering) were involved in target production. The chamber was evacuated to a base pressure lower than 10 -4 Pa and then backfilled with argon gas to a pressure of 0.6 Pa. The sodium glass slides used as substrates were RF sputter cleaned in situ at 200 W for 15 min before deposition. The substrates were rotated (15 rd/mn) and the target-substrate separation was fixed at 65 mm.
The crystalline structure was analyzed by X-ray diffraction using a Siemens D500 diffractometer with monochromatic CuK α radiation (λ = 1.54 Å). On the other hand, Raman scattering experiments were carried out using a Renishaw micro-Raman system at room temperature. The optical properties of the deposited films were measured at normal incidence in the wavelength range from 300 to 1800 nm using a double-beam UV-VIS-NIR spectrophotometer (Shimadzu). The photoluminescence (PL) measurements were performed at room temperature using a He-Cd laser operating at 325 nm.
Finally, electrical resistivity, Hall mobility and carrier concentration were measured at room temperature by a Hall measurement system with the Van der Pauw method.

X-ray diffraction and Raman scattering study
The highly crystalline, c-axis oriented and columnar structured Cu-doped ZnO thin film was confirmed by X-ray diffraction as shown in Fig. 1. The XRD patterns of the CZO films show mainly a (002) and (004) diffraction lines of ZnO hexagonal wurtzite structure (JCPDS File No: 00-036-1451), i.e., indicating that all films have the c-axis orientation perpendicular to the substrate, which has been confirmed by a previous study . No other phases for Cu and their oxides can be detected in the films, indicating that Cu doping does not alter the hexagonal structure and confirm the uniformity of Cu-doped ZnO thin films. It is noted that increasing the Cu-doping concentration reduced the intensity of the (002) peak, suggesting that Cu-doping can degrade the (002) preferential orientation of CZO thin films. This can be explained by the fact that the additional Cu ions, apart from replacing the Zn ions, may occupy the interstitial positions in the ZnO lattice where they form neutral defects and become ineffective as dopant impurities. The crystalline size of CZO thin films was evaluated using the fullwidth at half-maximum (FWHM) of the (002)-peak according to Scherrer's equation (Cullity and Stock 2001) where λ (1.541 Å) is the wavelength of Cu K α radiation, θ is the Bragg angle, and β is the half-width at half maximum (HWHM) of the (002) peak. The values of average crystallite size are summarized in Table 1. Figure 1c shows that the crystalline size of the CZO thin films decreased with increasing Cu concentration from 33 to 22 nm. This finding indicated that Cu doping deteriorated the crystal quality of the ZnO thin films. Moreover, Fig. 1b shows that the (002) peak slightly shifted toward lower angles with increased Cu concentration, which resulted in an increase in the lattice parameter according to Bragg's formula (Cullity and Stock 2001): where d hkl denotes the crystalline plane distance of (h, k, l) indices. Since the Cu 2+ (0,72 Å) ionic radius is smaller than the Zn 2+ ion (0,074 Å) (Chen et al. 2018), and hence the substitution of Cu in the Zn site increases the lattice parameter as showing in Table 1.
To evaluate the effect of Cu-doping on vibrational properties, room temperature Raman scattering analysis ranging from 100 to 1000 cm −1 was performed (Fig. 2). The wurtzite ZnO belongs to the space group of C 6v , with six active Raman modes of specific peak corresponding to the E 2 high-frequency E 2 (high) branch at wave number around 439 cm −1 was observed in the Raman spectrum of the undoped and Cu doped ZnO sample. As the Cu concentration increased, the E 2 (high) mode was gradually decreased and broadened. From the previous works, the E 2 (high) phonon mode is typically related to the crystalline quality, phase orientation, and strain present in the ZnO matrix (Horzum and Iyikanat 2019;Gautam 2019;Hu et al. 2010). So, the decrease of E 2 (high) peak intensity as the Cu concentration increased was attributed to the deterioration of the crystalline structure of ZnO thin film, as confirmed from the degradation of the c-axis growth orientation discussed in the XRD analysis. In addition to the E 2 (high) mode, A 1 (TO) and A 1 (LO) modes were observed at wavenumbers around 380 and 580 cm −1 in the Raman spectra of all samples. The A1(LO) and A 1 (TO) modes are reported to be linked to the abundant shallow donor defects such as oxygen vacancy (V O ) or interstitial Zn (Zn i ) in ZnO material (Gautam 2019 (Zhu et al. 2010). Also, the presence of this mode may be due to Raman diffusion indicating the presence of the structural disorder in all deposited films. So, photoluminescence spectra analysis was performed to confirm these results and to understand the effect of Cu doping on the optoelectrical properties. Figure 3 shows the optical transmittance and reflectance spectra of undoped and Cu doped ZnO thin films over a spectral range of 300-1800 nm. The fringes associated with interference effects confirm the optical homogeneity and the excellent surface quality of the deposited films. All prepared films exhibit good transparency in the visible range (> 80%) and a sharp fundamental absorption edge around a wavelength of 360 nm, indicating that these materials are suitable for use as a transparent electrode. The absorption edge of direct where α is the absorption coefficient, hν is the photon energy, A is a constant and E g is the optical bandgap energy. Figure 4 shows the plot of (αhν) 2 versus hν of undoped and Cu doped ZnO thin films deposited at various Cu concentrations. The straight-line portion of the curve, when extrapolated to zero, gives the optical band gap E g . The bandgap energy values E g for the deposited films are summarized in Table 1. Remarkably, the bandgap energy values of CZO thin films decreased with increasing Cu concentrations. According to a previous report (Elilarassi and Chandrasekaran 2010), the decrease in the bandgap energy may be due to thesp-d spin-exchange interactions between the band electrons and the localized d electrons of the metal ion (Zn) substituting the Cu ion. Moreover, the difference in the electronegativity between Cu and Zn atoms may induce some defects resulting in a decrease of bandgap energy (Muthukumaran and Gopalakrishnan 2012). Room temperature photoluminescence (PL) measurements were carried out for a more detailed investigation of defects that existed in the Cu-doped ZnO films. The PL spectra of Cu-doped ZnO films have been recorded by using He-Cd laser at an excitation wavelength of 325 nm. Figure 5 shows the deconvolution of PL spectra of undoped and Cu doped ZnO thin films at various Cu contents. From this deconvolution, it is seen that four emission peaks were observed at around 386 (3.21 eV), 415 (2.98 eV) 484 (2.56 eV) and 530 nm (2.33 eV) for all deposited samples and are in good agreement with previous reports (Sreedhar 2016). A strong UV emission peak located at about 386 nm is related to a near band-edge transition of ZnO, namely, the recombination of the free excitons (Ouarez et al. 2018). As shown in Fig. 6, the UV emission peak shifted to a higher wavelength with increasing the Cu doping concentration. Indeed, ZnO is a kind of n-type semiconductor and Cu atom is a p-type dopant. So, this shift of the UV emission peak can be attributed to the sp-d spin-exchange interactions between the band electrons and the localized d electrons of the metal ion (Zn) substituting the Cu ion (Elilarassi and Chandrasekaran 2010). Amol R. et al. (Nimbalkar and Patil 2017) reported also that the value of bandgap energy (E g ) should decrease after Cu doping into ZnO crystal lattice.PL results are in agreement Fig. 4 The plots of (ahν) 2 against hν of Cu-doped ZnO thin films at various Cu contents with the UV-Vis spectroscopy studies and support the observation of a decrease in the bandgap energy on increasing Cu doping concentration. On the other hand, it was found that the progressive addition of Copper Cu reduces the absolute intensity of the observed  (Drmosh et al. 2013), N-doped ZnO films (Kim and Leem 2015) and Ni-doped ZnO films (Amanullah et al. 2011) and can be ascribed to the density of free exciton in ZnO thin films affecting the intensity of ultraviolet emission (Yan et al. 2011). The emissions observed in the visible region have previously been attributed to several defective states within bandgap energy, such as oxygen vacancy (V O ), zinc vacancy (V Zn ), oxygen interstitial (O i ) and zinc interstitial (Zn i ) (Hasabeldaim et al. 2019). In our study, the violet emission energy centered at 2.98 eV which may be assigned to the transition energy of Zn i defect level to the valence band maxima. According to the theoretical calculation, P. S. Xu et al. have previously investigated the violet emission energy of the ZnO and found it to be correlated with the density of Zn i defects (Xu et al. 2003). Himanshi Gupta et al. reported that the blue emission could be originated from the recombination of a delocalized electron in the conduction band with a hole in the singly ionized oxygen vacancies (influence of doping and energetic ion irradiation 2019). Based on the theoretical approach presented by Federico Gallino et al.the transition energy between the conduction band and the singly ionized V O (+ 1/0) level was 2.52 eV (Gallino et al. 2010). Thus, we proposed that in our films, the blue emission can be ascribed to the singly ionized oxygen vacancies. The emission peak centered at about 2.33 eV is attributed to green luminescence and can be assigned to the transition from the defect level of neutral zinc interstitial Zn i to the valance band maxima which is consistent with the results reported by L. T. Tseng et al. (Tseng et al. 2014).

Hall effect Study
Hall measurements were performed in a Van der Pauw four-point configuration at room temperature and the corresponding results are listed in Table 2. From previous reports, Cu doping introduces an acceptor (Cu + ) level in ZnO at 0.45 eV above the valence band (Suja et al. 2015;Agarwal et al. 2019). So, we can conclude from these reports that Cu may assume a valency of either + 1 or + 2 depending on its chemical environment. In our study, we found that all the deposited Cu doped ZnO films exhibit n-type conductivity. This result is surprising since the Cu ion is supposed to provide p-type ZnO. We can, therefore, think that the probable insertion in a substitution position of the Cu + ion in the ZnO matrix is not sufficient to induce p-type conduction and (or) insertion is not total and that the noninserted Cu ions are responsible for defects that can bring carriers of n-type conductivity. This behavior was observed by Chang Oh Kim et al. (Chang Oh Kim 2010) in their study on the Cu doped ZnO thin films, but also in the ZnO thin films doped with Nitrogen N (Pathak et al. 2016) or with Lithium Li (Wang et al. 2009).On the other hand, we can notice that the resistivity increases slightly with increasing the Cu doping concentrations while the carrier's mobility decreases slightly with increasing Cu doping concentration. This result is not surprising and supports the explanation given above. We have probably a competition between two opposite effects. The first is related to the insertion of the Cu ions into the ZnO matrix and is supposed to provide a p-type conductivity and therefore induce an increase in the resistivity of the films. The second effect is related to the increase in defects number due to the Cu ions not inserted in the ZnO matrix which can bring doping of n-type conductivity and therefore induce a decrease in the resistivity. This idea is supported by the decrease in the carrier's mobility which is intimately linked to the quality and purity of the deposited thin films. Indeed, the presence of ionized impurity and the existence of defects within the samples increases the probability of free carrier collisions. The vibration of the atoms of the crystal lattice around their equilibrium position is also an obstacle for free electrons. However, it should be borne in mind that the resistivity values before and after doping with Cu remain relatively low in the range of 10 -4 -10 −2 Ω.cm which shows that this system remains potentially interesting for photovoltaic applications.

Photocatalytic activity test
The photocatalytic activities of undoped and Cu doped ZnO thin films with different Cu concentrations were evaluated by monitoring the decomposition of methylene blue (MB) for 3 h under sunlight irradiation. The analysis has been done in February, where the temperature was about 25 degrees. Undoped and Cu doped ZnO thin films with an area of 2 × 2 cm 2 were placed in a methylene blue solution (3 mg. l −1 ) and this solution was stirred in the dark for 30 min to reach the adsorption-desorption equilibrium. Then, the MB decomposition testing was carried out by a UV-Visible spectrophotometer. Figure 7 shows the absorption spectra of MB after solar irradiation for 3 h for the 1 wt% Cu doped ZnO film. It can be seen from these absorption spectra, the decrease in the maximum absorption peak intensity is located at about 664 nm which indicates the degradation of the MB and the improvement of the photocatalytic activity of the films. For all the samples, they show the same tendency; that is to say, the MB concentration gradually decreased with the prolonging of irradiation time and the degradation efficiencies achieved 85%. To evaluate the influence of copper concentration on the photocatalytic properties of ZnO thin films, we have Fig. 7 The absorption spectra of MB for the 1 wt% Cu doped ZnO film plotted the absorption spectra of MB after solar irradiation after 3 h of undoped and Cu doped ZnO thin films at various Cu concentrations as showing in Fig. 8. Indeed, we can see an increase in the intensity of the absorption peak with the increase in Cu concentrations except for the 1 wt% Cu doped ZnO film, confirmed by the low absorption peak intensity indicating its great photocatalytic activity compared to the other layers. This may be due to the high mobility and the important density of the charge carrier of this film. Also, we can explain this by the fact that at higher Cu doping concentration, the Cu atoms cannot occupy zinc lattice sites and they have a tendency to occupy interstitial sites where they form neutral defects that will serve as recombination centers for photogenerated electrons and can hind the electron-hole recombination affecting thus the photocatalytic activity (Tabib et al. 2017).
The degradation efficiencies were evaluated by using the following relation (Nazarabad and Goharshadi 2017;Nasser et al. 2017): where A 0 is the initial (t = 0) absorbance of MB and A is the absorbance at time t. Figure 9 shows the degradation efficiencies of MB dye versus irradiation time. The photocatalytic activity of the 1 wt% Cu doped ZnO film is the highest and the degradation of MB is nearly up to 80% after 3 h of irradiation. So, we can notice that the ZnO films doped with a small amount of Cu could enhance the photocatalytic activity. Figure 10 shows the least fit curve of the logarithmic ratio of concentrations with time for the undoped and Cu doped films with different concentrations. The straight-line curve indicates that the rate of photo-degradation of MB dye obeys pseudo-first-order kinetics as stated by Langmuir-Hinshel Model (Vasanth Kumar et al. 2008): where A is the absorption peak intensity at the desired irradiation time interval, A 0 the initial absorption intensity of MB dye, k the rate constant and t the irradiation time. The values of k are determined from the slope of the fitted curves and they are presented in    Table 3. The kinetic rate constant is estimated to be about 3.03 × 10 -3 min −1 for MB in the absence of a photocatalyst. We can notice that the introduction of photocatalyst enhances the kinetic constant rate and improves the photocatalytic activity. The most important value of k is obtained for the 1 wt% Cu doped ZnO film and is about 8.58 × 10 -3 min −1 which confirms the best photocatalytic activity of this film.

Conclusions
In summary, we report ZnO thin films doped with varying Cu concentrations deposited by RF magnetron sputtering technique with recourse to a low-cost powder target. The structural analysis carried out by XRD suggested that the incorporation of Cu deteriorates the crystallinity of the host thin films. We have noticed from the UV-Vis analysis that all deposited films exhibit high transparency and increasing in Cu doping concentration leads to a decrease in the bandgap energy. PL spectra indicate that a high concentration of defects and impurities exist in the deposited films, which is in agreement with the results of Raman analysis. Finally, The photocatalytic activities of undoped and Cu doped ZnO thin films with different Cu concentrations were evaluated by monitoring the decomposition of methylene blue (MB) and we have concluded that the best photocatalytic activity was obtained for the 1 wt% Cu doped ZnO film.

Conflict of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.