Growth and characterization of electrodeposited Cu₂O thin films

The electrodeposition potential required for a synthesis in the potentiostatic mode is usually unknown. In this situation, cyclic voltammetric measurements help one to identify oxidation reduction processes potentially undergone by the system of interest and to choose an appropriate potential. As shown in figure 1(a), cyclic voltammetry (CV) reveals two reduction reactions for Cu²⁺ ions: one leads to Cu⁺ ions and the other to metallic Cu. Figure 1(a) shows a typical cathodic scan performed between 0.4 and −0.8 V at a scan rate of 20 mV s⁻¹. The beginning of the current decrease was detected at −0.2 and −0.74 V versus SCE, which is characteristic of the overpotential deposition process of reduction process of Cu²⁺ to Cu⁺ and the Cu⁺ to Cu⁰, respectively [14]:

$$\begin{equation} {\rm Cu}^{2 + } + {\rm e}^ - \to {\rm Cu}^ + \end{equation} \tag{ 1 }$$

$$\begin{equation} {\rm Cu}^ + + {\rm e}^ - \to {\rm Cu}{\rm .} \end{equation} \tag{ 2 }$$

The produced Cu⁺ ions react with OH⁻ ions in the solution to form Cu₂O [16]:

$$\begin{equation} 2{\rm Cu}^ + + 2{\rm OH}^ - \to {\rm Cu}_{\rm 2} {\rm O}_ \downarrow + {\rm H}_{\rm 2} {\rm O}{\rm .} \end{equation} \tag{ 3 }$$

In the reverse scan, the anodic stripping peak due to the dissolution of copper is observed at a potential around 0 V.

Zoom In Zoom Out Reset image size

The current–time transients of the films produced at the −0.4, −0.5 and −0.6 V for the first 150 s of deposition temps was shown in figure 1(b). The shape of the film growth curve can clearly indicate the nucleation-growth mechanism [17]. As seen from this figure, at the beginning of the applied potentials, a high cathodic current is seen for a short time of 2 s compared to the deposition of 150 s. It indicates the formation of Cu₂O on the surface of substrate. As the cathode potential is increased the current also increases. The curves i–t show also a normal dependence with the applied potential, where an increase of the current density with the applied potential was observed and the process of the electrodeposition becomes faster. It is seen that Cu₂O thin films deposited at the different deposition potentials have the same type of growth mode and the films were deposited correctly and orderly with the constant current values.

The capacitance measurement on the electrode/electrolyte was also employed to determine the flat band (E_fb) and carrier density (N_A), which can be obtained in a Mott–Schottky (MS) plot with 1/C² versus potential at a fixed frequency of 20 kHz. The capacitance-potential measurements are presented as an MS plot following the equation below [18]:

$$\begin{equation} \frac{1}{{C^2 }} = \frac{2}{{N_A \varepsilon \varepsilon _0 e}}\left[ {\left( {E - E_{fb} - \frac{{kT}}{e}} \right)} \right]. \end{equation} \tag{ 4 }$$

In equation (4), C is the interfacial capacitance (i.e., capacitance of the semiconductor depletion layer), ε is the dielectric constant of Cu₂O (taken as 7.6, [19]), ε₀ is the permittivity of free space (8.85 × 10⁻¹² F m⁻¹), N_A is the number density (cm⁻³) of acceptors in Cu₂O (doping level), E is the applied potential, E_fb is the flat band potential, T is the absolute temperature (298 K), k is the Boltzmann constant (1.38 × 10⁻²³ J K⁻¹), and e is the electron charge (1.6 × 10⁻¹⁹ C). The temperature term is generally small and can be neglected. The capacitance values of the space-charge region obtained at different applied potentials are shown in figure 2. According to the MS equation, a linear relationship of $\frac{1}{{C_{CS}^2 }}$ versus E can be observed (figure 2). The p-type conductivities of the films are also justified, by the negative slopes of the straight lines [20, 21]. From the slope $\big( { = \frac{2}{{\varepsilon \varepsilon _0 eN_A }}} \big)$ and intercept at C = 0, the acceptor density and the flat-band potential of a p-type semiconductor can be obtained, respectively.

Zoom In Zoom Out Reset image size

Thus, from figure 2, the flat band potential is found to be about 0.167, −0.07 and −0.004 V for films deposited at −0.4, −0.5 and −0.6 V, respectively. Also, from this latter figure the acceptor densities in the film were 2.41 × 10¹⁸, 2.85 × 10¹⁸ and 5.38 × 10¹⁸ cm⁻³ at three precedent deposition potentials. These values, although high for a typical semiconductor, agree with the previously reported values for Cu₂O films synthesized through different methods [22, 23]. Films obtained at high deposition potential present higher crystallites sizes than those obtained at low deposition potential, which eventually might decrease the number of defects in the films (see XRD section 3.3). However, considering the electrochemical process taking place at the electrode surface, an increase in the deposition potential can also increase the number of stoichiometric defects as a consequence of a larger interdiffusion of oxygen into the films. This effect would predominate over the crystallites size effect. Thus, at low potential one expects to obtain films with a higher defect concentration than the films obtained at a high potential. The latter is consistent with the results found in the optical analyses of the films obtained at different potentials [24].

According to the above reactions, our experiments were employed at different cathodic potentials. The morphological characteristics of the Cu₂O thin films were investigated by AFM. Figure 3 shows the AFM images of the Cu₂O films electrodeposited on the FTO substrate at different deposition potentials. In these images, it can be observed that their morphology differs from the one of the potential employed. The film obtained at −0.4 V has a globular morphology with a grain size apparently smaller than those obtained at −0.5 and −0.6 V. Thus, the grains observed at −0.4 V in the AFM image, correspond mainly to granular material. On the other hand, at −0.5 and −0.6 V the films are crystalline and the morphology of the obtained films is different from those prepared at −0.4 V. The qualitative comparison between surface roughnesses for three electrodeposited Cu₂O films is shown in table 1. Each value was the average of four independent layers, and at each layer four determinations at different locations were made. It is indicated that the surface of the film obtained from films prepared at −0.4 V versus SCE consists of smaller feature grains than for that obtained at −0.5 and −0.6 V versus SCE, respectively, and has the best surface morphology in agreement with the XRD analysis. It is clear that the deposition potential controls both the crystal nucleation and the growth rate, which in turn determine the grain size and surface roughness.

Zoom In Zoom Out Reset image size

The structural state of the Cu₂O electrodeposited on the conductive substrate FTO was characterized by XRD. Figure 4 shows the XRD pattern of Cu₂O thin films deposited at −0.4, −0.5 and −0.6 V, respectively. Besides the diffraction peaks from the FTO conductive film on glass, all other diffraction peaks can be identified as due to the standard cubic cuprous oxide with polycrystalline structure. We did not find any peaks of other phases such as CuO and elemental Cu in the detection limit of our technique. Effectively, all the peaks correspond to the reflections from (1 1 1), (2 0 0), (2 2 0) and (3 1 1) planes of cubic phase of Cu₂O (JCPDS file no.05-0667). Hence, the XRD patterns show that the Cu₂O film is of a single phase with a preferential orientation along the (1 1 1) plane. The intensity of this peak is found to increase with increasing deposition potential. In fact, at −0.5 V and −0.6 V, the intensity of the (1 1 1) peak is very strong and their width at half maximum is small, indicating a good crystallization state through a large crystallites size.

Zoom In Zoom Out Reset image size

Assuming a homogeneous strain across crystallites, the average crystallites size can be estimated from the full width at half maximum (FWHM) values of the diffraction peaks. An average size of the crystallites in the direction perpendicular to the plane of the films could be obtained using the Scherrer equation [25]:

$$\begin{equation} D = \frac{{k\lambda }}{{\beta \cos \,\theta }} \end{equation} \tag{ 5 }$$

where D, θ, and λ are the mean crystallite size, Bragg angle, and wavelength of the incident x-ray (1,54 056 nm). K is a shape factor and usually takes a value of 0.94. The calculation shows that the average size of the crystallites of as-grown Cu₂O samples electrodeposited at −0.4, −0.5 and −0.6 V versus SCE estimated from (1 1 1) diffraction peak are 47, 94 and 80 nm, respectively, and hence the crystallite size was confirmed to the nanoscale. It is known that the term 'crystallite size' means the dimensions of the coherent diffracting domain. Therefore, this formula is applicable to thin films where lattice strain is absent. Electrodeposited thin films can possess some strains, which could also contribute to peak widening, thereby affecting the estimation of the crystallite size. Therefore, the size of the crystalline domains determined from the XRD peak widths is used only as a comparative parameter among samples [26].

Using these values of crystallite size, the dislocation density δ defined as the length of dislocation lines per unit surface of the crystal can be calculated through the following relation [27, 28]:

$$\begin{equation} \delta = \frac{n}{{D^2 }} \end{equation} \tag{ 6 }$$

where δ is the dislocation density, n is a factor, when equal unity gives minimum dislocation density and D is the crystallite size. According to this, the dislocation density varies with the deposition potential from 4.52 × 10⁻¹⁴, 1.11 × 10⁻¹⁴ to 1.55 × 10⁻¹⁴ lines m⁻² for deposition potential of −0.4, −0.5 and −0.6 V versus SCE, respectively.

Analysis of the photoluminescence (PL) of semiconductor materials is a powerful tool to obtain information about the structure of energy bands and the crystalline quality. The room temperature (RT) PL spectra of Cu₂O film grown on FTO substrates were also studied using the 355 nm line of Nd-YAG laser, as shown in figure 5 for sample deposited at −0.4, −0.5 and −0.6 V, respectively. From this figure, the visible emission centered at about 530 nm observed for sample deposited at −0.5 V was commonly attributed to the deep-level defects such as oxygen vacancies and copper interstitials [28, 29]. Such a band is usually observed in ZnO films and is also attributed to the same type of defects [30]. The absorption–transmission properties of Cu₂O films deposited at −0.4, −0.5 and −0.6 V were measured by UV-visible (UV-vis) spectrophotometry, as shown in figure 6. From this figure, high transparency is observed throughout the visible wavelength range with an average transmission of above 75% for all Cu₂O films. In the UV region, all the films exhibit a sharp fundamental absorption edge. The inset of figure 6 shows the UV-vis transmittance spectra of the Cu₂O thin films. The absorption at wavelengths longer than 360 nm is due to the absorption of Cu₂O.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The optical absorption studies are important in material characterization for determining the absorption in the semiconducting materials, direct and indirect optical band gaps of the semiconductors because the absorption coefficient and energy band gap play an important role in understanding the optoelectronic properties of semiconducting materials.

The energy band gap (E_g) for Cu₂O thin films was evaluated by using the Tauc plot [31]. The Tauc plot is the plotting of (αhν)^1/n as a function of photon energy (hν) for different values of n (n = 1/2 and 2). A linear relationship between (αhν)^1/n and hν for n = 1/2 and 2, respectively, ensures the direct and indirect allowed transitions in the solid. The value of E_g is determined from the intercept of the straight-line portion of the Tauc plots at the horizontal axis when α = 0. This method is known to be accurate for the estimation of the E_g of Cu₂O thin films [32]. The relationship of (αhν)² and photon energy hν for Cu₂O thin film deposited at different potentials is shown in figure 7. The plots of (αhν)² versus photon energy (hν) give straight lines, indicating that the deposited Cu₂O thin films were of direct forbidden band gaps. Using the linear extrapolation method, the values of band gap energy of Cu₂O films deposited at −0.4, −0.5 and −0.6 V are estimated to be 2.07, 2.36 and 2.49 eV, respectively. These values of optical band gap of Cu₂O thin film obtained by electrodeposition is similar to other reports [32–35].

Zoom In Zoom Out Reset image size