1 Introduction

In recent years, nanostructured materials have been extensively explored owing to their unique features for various significant applications and the development of low-cost and high-quality nanostructured materials is of great interest [1,2,3,4,5,6,7]. Zinc oxide (ZnO) with a direct wide energy bandgap and a relatively high exciton binding energy at room temperature is one of the functional semiconductor materials [8,9,10]. Nanostructured ZnO materials with various morphologies and sizes have gained a considerable interest over the last decade due to their high technology applications in microelectronics and optoelectronic devices [4, 11,12,13,14,15,16,17,18,19]. A wide variety of physical and chemical methods with and without a vacuum system used in the manufacturing process of ZnO materials are available in the literature. Among them, electrochemical deposition (ED) process without a vacuum system is one of the most effective chemical methods for high-quality ZnO materials owing to its unique advantages such as cost effectiveness, versatility, low-temperature operation, relatively high growth rates, flexibility, and environmentally friendly and large scale deposition [14, 15, 19, 20]. In addition, various shapes of nanostructured ZnO materials with different physical properties such as nanobelts, nanosprings, nanosheets, nanorods (NRs), nanowires, nanocombs and nanorings can be fabricated by ED process without using templates or metal catalyst [6, 20,21,22]. It is well known that the crystallinity, orientation, size and morphology of nanostructured materials play a very important role in semiconductor nanotechnology [22, 23]. For this purpose, the seed layers prepared on the substrate surfaces prior to actual deposition of subsequent ZnO particles have been used to achieve a higher structural and morphological quality compared to the ZnO particles without the seed layer. In this case, it becomes important to reveal the effect of seed layer deposition conditions (cathode potential and/or current density and electroplating time) on the ZnO samples. It was shown that the seed layers prepared by cathodic ED process at different cathode potentials played a significant role on the optical and structural characteristics, size and density of ZnO particles [17, 18, 24]. The studies also reported that the structural, morphological and optical properties of ZnO platelets grown from a zinc nitrate (Zn(NO3)2) [14] and ZnO NRs prepared from zinc chloride (ZnCl2) [21] aqueous electrolyte solutions were considerably affected by the seed layer current density. Based on the literature, it is clear that the controlling of the diameter, density, orientation and distribution of ZnO particles grown by ED technique via seed layer growth process is still under investigation. However, there are very limited studies about the effect of the tseed layer on the ZnO particles in the literature. The study investigated the influence of the tseed layer on the ZnO samples prepared from Zn(NO3)2 aqueous electrolyte solution on the indium tin oxide (ITO) coated glass substrates with etching process in diluted HCl by the two-step ED method [15]. In that study, the potentiostatic step (– 1.2 V versus saturated calomel electrode, SCE) was used to produce ZnO seed layers while the galvanostatic step (– 0.5 mA/cm2) was applied to fabricate subsequent ZnO particles. The effect of the tseed layer on the properties of subsequent ZnO NRs prepared from Zn(NO3)2 aqueous electrolyte solution on fluorine tin oxide (FTO) coated glass substrates by the two-step potentiostatic electrolysis has been also studied in a very recent paper [20]. In that paper, the seed layers were deposited under an electrochemical potential of – 1.3 V while the subsequent ZnO NRs were electroplated on the seed layers at the electrochemical potential of – 1 V versus SCE. In our present work, a constant current density of – 3 mA/cm2 was applied for the seed layer growth process while the subsequent ZnO NRs were fabricated on the seed layers at a constant cathode potential of – 0.72 V versus a silver/silver chloride (Ag/AgCl) electrode. We studied the effect of the tseed layer on the structural properties, optical energy bandgap, diameter, growth orientation and surface roughness of the ZnO NRs electrochemically produced from ZnCl2 aqueous electrolyte solutions onto ITO-covered glass substrates. The results showed that the properties of the subsequent ZnO NRs can be controlled by the tseed layer value.

2 Experimental

In this work, the ZnO NRs were produced using ED technique. The ED process of ZnO NRs consisted of two steps. The galvonostatic step was applied to produce the seed layers while the potentiostatic step was applied to deposit the subsequent ZnO NRs. A constant current density of – 3 mA/cm2 was applied during the seed layer process and three different tseed layer values of 25 s, 50 s and 100 s were chosen for the ED process of the seed layers. After the seed layer growth process, the subsequent ZnO NRs were fabricated on the seed layers at a constant cathode potential of – 0.72 V versus Ag/AgCl electrode at the bath temperature of 80 °C for electroplating time of 1800s. For comparison, the ZnO sample was also electroplated on the ITO-covered glass substrates without a seed layer (tseed layer = 0 s) under the same experimental conditions. Before starting the ED process of the ZnO NRs (30 min ago) and during the experiments, the oxygen was also introduced into the aqueous electrolyte solution by bubbling. The ED experiments were performed in a conventional electrochemical cell with three electrodes. The working electrode was conductive ITO-coated glass substrate. A platinum sheet with an area of 1.5 cm2 and the Ag/AgCl electrode were used as a counter electrode and the reference electrode, respectively. The sheet resistance of the conductive ITO substrate was about 8–12 Ω/sq and the working area was about 1 cm2. The thickness of the magnetron sputtered ITO thin films utilized as a working electrode in the experiments was 160–180 nm. Prior to starting, the ED process of the samples carried out by the VersaSTAT 3 Potansiostat/Galvanostat, the substrate surfaces were first cleaned in the medium for 10 min in acetone and then in the medium for 10 min in ethanol. After that, they washed with deionized water for 15 min using an ultrasonic instrument cleaner. The aqueous electrolyte solution utilized in the ED process contained ZnCl2 (0.003 M) and KCl (0.1 M) purchased from Merck without further purification. The purity of using materials was greater 99%. Apart from these reactants, further additives and complexing agents were not introduced in the electrolyte solution. The pH of the freshly prepared electrolyte solutions was 6.9 ± 0.1. Before the characterization of the resultant ZnO samples, they were washed with deionized water and dried in air.

The morphologies of the subsequent ZnO NRs in dependence of the tseed layer were characterized using a MultiMode V SPM of Veeco atomic force microscopy (AFM) in tapping mode and a Tescan MAIA3 XMU scanning electron microscopy (SEM). An operating voltage of 5 kV was applied during the SEM analysis and the SEM images were taken at low and high magnifications of 30 kx and 50 kx, respectively. The surface roughness of the ZnO samples was obtained using a WSxM 5.0 develop 8.2 software package [25]. To evaluate the crystal structure, preferential growth orientation and mean crystallite size with respect to the tseed layer, the XRD spectra of the ZnO samples were collected by a Rigaku Smart LabTM X-ray diffractometer (XRD). During the XRD measurements, the scans were done in the 2θ range of angle between 20° and 65° with a step of 0.01° using a Cu–Kα radiation (0.15406 nm). The optical absorption spectra of the samples were obtained by means of an ultra violet visible-near infrared (UV–VIS–NIR) spectrophotometer (Shimadzu 3600 Plus) in a wavelength range from 350 to 600 nm. All characterization techniques were applied at ambient temperature and pressure.

3 Results and discussion

Figure 1 presents the XRD spectra of the ZnO samples electroplated at different tseed layer values. As noted in the XRD spectra, the (100), (002), (101), (102), (110) and (103) and diffraction peaks appeared on the XRD patterns, reflecting the polycrystalline nature of the ZnO samples. These visible diffraction peaks are quite consistent with those of the standard wurtzite crystal structure of ZnO (PDF code no: 01-079-0207). Except for ZnO, no characteristic diffraction peaks were observed due to impurities and unreacted species in the XRD spectra. Besides, there was a negligible shift in the 2θ position of the diffraction peaks with the tseed layer, reflecting that the effect of the tseed layer on the ZnO lattice distortion was insignificant. On the other hand, compared to other diffraction peaks, the (002) and (101) diffraction peaks are the most intense for the ZnO samples electroplated at the tseed layer values of 25, 50 and 100 s. However, the (100) and (101) diffraction peaks are the strongest for the ZnO sample without the seed layer. On the other hand, the intensity (101) diffraction peak decreased gradually with the increase of the tseed layer value. The intensity of the (002) diffraction peak increased considerably as the tseed layer increased from 0 to 50 s and then decreased with the tseed layer. Contrary to the case of (002) diffraction peak, the intensity (100) diffraction peak decreased noticeably with increasing tseed layer from 0 to 50 s and then increased with the tseed layer. Furthermore, the relative integral intensity of the (101) and (002) diffraction peaks changed depending on the tseed layer value. The integral intensity ratio of (002) to (101), I(002)/I(101), was estimated from the XRD data to determine the preferential growth orientation of the ZnO NRs. The obtained values are demonstrated in Fig. 2 against the tseed layer. For comparison, the standard value of the I(002)/I(101) for ZnO powder is also shown in Fig. 2 [14]. The I(002)/I(101) value for the ZnO sample without the seed layer was close to the standard value of the ZnO powder. This indicated that the ZnO NRs without the seed layer has a random growth orientation. However, the I(002)/I(101) values were found to be higher than that of the ZnO powder for the ZnO NRs fabricated at the tseed layer values of 25 s, 50 s and 100 s. This revealed the creation of the preferential growth orientation along the [002] direction, implying that the ZnO NRs fabricated on the seed layers grow along the c-axis direction being perpendicular to the substrate surface. Furthermore, the I(002)/I(101) ratio reached its maximum value at the tseed layer of 50 s (Fig. 2). This indicated the improved growth of the subsequent ZnO NRs along the c-axis orientation of the wurtzite structure, which makes the ZnO sample fabricated at the tseed layer of 50 s a promising candidate for applications in optoelectronics [13]. Thus, we concluded that the [002] preferential growth orientation can be controlled by the tseed layer. In a previous study [15], all of the ZnO particles without and with the seed layers grown at different tseed layer values showed the preferential growth orientation along c-axis. The best c-axis orientation was observed at the tseed layer of 15 s. In another recent study [20], the ZnO NRs exhibited the c-axis orientation at the tseed layer values of 10 s, 30 s and 60 s while the ZnO NRs with 90 s of tseed layer showed no preferential growth orientation. The best c-axis orientation was detected at the tseed layer of 60 s. Compared to our results, the differences may be ascribed to different parameters applied in ED process of the ZnO samples such as precursor type, solution pH, substrate type and applied current density or deposition cathode potential. Similar improvements in the strength of the [002] preferential growth orientation were also achieved via seed layer current density in former studies [14, 21].

Fig. 1
figure 1

XRD spectra of the ZnO NRs without and with the seed layer for different tseed layer values. The diffraction peaks related to the ITO-covered glass substrate are indicated with asterisks (*)

Full size image
Fig. 2
figure 2

The I(002)/I(101) integral intensity ratio of the ZnO samples against the tseed layer value. tseed layer = 0 s shows the sample without the seed layer

Full size image

The mean crystallite size, D, for different tseed layer values was estimated by means of the Scherrer relation [26]:

$$ D = [0.9\lambda /\beta \cos \theta ] \times [180^\circ /\pi ] $$

where λ represents the wavelength of X-ray utilized in the XRD measurements, θ corresponds the Bragg diffraction angle, β is the full width at half maximum (FWHM) values of the diffraction peaks. The FWHM and θ values of the diffraction peaks were estimated by Gaussian fitting. The variation of the mean crystallite size is demonstrated in Fig. 3 as a function of the tseed layer. As seen from Fig. 3, the mean crystallite size of the samples first decreased from 56 to 50 nm as the tseed layer increased from 0 to 25 s. Subsequently, the mean crystallite size enhanced from 50 to 55 nm with increasing tseed layer from 25 to 100 s (Table 1).

Fig. 3
figure 3

The mean crystallite size of the ZnO samples with respect to the tseed layer value. tseed layer = 0 s indicates the sample without the seed layer

Full size image
Table 1 The structural, morphological and optical characteristics of the ZnO NRs without (tseed layer = 0 s) and with the seed layers obtained at different tseed layer values
Full size table

The microstructures of the resultant ZnO samples were investigated by taking top-view SEM images. The low and high magnification SEM images of the electrochemically produced ZnO samples without and with the seed layers produced at different tseed layer values are shown in Fig. 4. The diameter of the ZnO NRs ranged from several tens of nanometers (~ 60 nm) to several hundreds of nanometers (~ 380 nm). However, the mean diameter, orientation, shape and density of the ZnO NRs were noticeably affected by the tseed layer value. As distinctly noticed from Fig. 4a, the ZnO sample without the seed layer had a broad size distribution and the NRs were not uniformly distributed (randomly oriented) over the entire ITO-coated glass substrate. The top–view SEM image of the sample without the seed layer shown in Fig. 4a indicated that the ZnO NRs were mostly hexagonal in shape. However, spherical-ended NRs were also detected in some regions on the sample surface. The average rod diameter for this ZnO sample was found to be around 270 nm. Compared to the SEM image of the sample without the seed layer, more homogeneously distributed ZnO NRs with a smaller diameter were obtained at the tseed layer of 25 s. Their average rod diameter was about 170 nm (Fig. 4b). When the tseed layer was increased from 25 to 50 s, the diameter of the ZnO NRs decreased more. The most compact surface morphology with a good vertical alignment and a narrow size distribution was obtained at the tseed layer of 50 s. The average rod diameter for this sample was determined to be ~ 120 nm (Fig. 4c). Furthermore, the shape of the ZnO NRs obtained at the tseed layer values of 25 s and 50 s was completely hexagonal. Further increase in the tseed layer from 50 to 100 s resulted in a decrement in the homogeneity and an increment in the diameter of the NRs, indicating a decrease in the grain refining effect of the seed layer process. Also, there were spherical-ended NRs in some regions on the sample surface and the NRs showed a broad size distribution. The ZnO sample grown at the tseed layer of 100 s consisted of NRs with an average diameter of ~ 200 nm (Fig. 4d). In a very recent study [20], the SEM results showed that an increment in the tseed layer from 10 to 90 s gives rise to an increase in the ZnO NRs density and a gradual decrease in the rod diameter. However, poor alignment was obtained in the film deposited at the tseed layer of 90 s compared to the film prepared at the tseed layer of 60 s. In another former study [15], the SEM analysis indicated that the ZnO film deposited with 15 s in tseed layer exhibited the highest density of the ZnO particles with the smallest size compared to other films without and with the seed layer deposited at the tseed layer values of 5 s and 60 s. The average rod diameter determined for the ZnO sample grown at the tseed layer of 50 s in this study is considerably smaller than those of the ZnO samples [21, 27] deposited on the ITO-coated glass substrates and comparable with the results previously reported [28].

Fig. 4
figure 4

Top-view SEM images of the ZnO samples a without (tseed layer = 0 s) and with the seed layers electrochemically grown at different tseed layer values b 25 s, c 50 s and d 100 s, respectively

Full size image

To learn more about the influence of the tseed layer on the morphological properties of the ZnO samples, AFM images were also utilized for determining the surface roughness parameters of the subsequent ZnO NRs. Three-dimensional (3D) AFM images of the electrochemically deposited ZnO samples without and with the seed layers fabricated at various tseed layer values are presented in Fig. 5. As evident from the AFM images shown in Fig. 5, compared to the ZnO sample without the seed layer, the rod diameter decreased considerably as the sample was produced on the seed layer with the tseed layer of 25 s. As the tseed layer increased from 25 to 50 s, the rod diameter decreased more. A further increase in the tseed layer from 50 to 100 s led to an increase in the rod diameter. These findings obtained from the AFM analysis matched very well with those obtained from the SEM analysis. The average surface roughness (Ra) and root mean square surface roughness (Rq) values of the ZnO samples are presented in Fig. 6 with respect to the tseed layer. The results revealed that the surface roughness of the ZnO samples can be significantly reduced by adjusting the tseed layer. As seen from Fig. 6, the ZnO sample without the seed layer showed the highest Ra and Rq values (104.9 nm and 128.8 nm, respectively). The ZnO sample produced at the tseed layer of 25 s exhibited significantly lower Ra and Rq values of 62.3 nm and 75.9 nm, respectively, compared to the ZnO sample without the seed layer. The Ra and Rq values decreased more when the tseed layer increased from 25 to 50 s and thus the lowest Ra and Rq values (46.8 nm and 58.2 nm, respectively) were achieved for the ZnO sample electrochemically fabricated at the tseed layer of 50 s. The decrease of the surface roughness is due to the decrease of the rod size. However, a further increase in the tseed layer from 50 to 100 s caused an increase in the Ra and Rq values to 72.5 nm and 95.4 nm, respectively. As a result, the analysis showed that the ZnO sample grown at the tseed layer of 50 s exhibited the subsequent ZnO NRs with the smoothest top surface.

Fig. 5
figure 5

3D AFM images of the ZnO samples a without (tseed layer = 0 s) and with the seed layers electrochemically fabricated at different tseed layer values b 25 s, c 50 s and d 100 s, respectively

Full size image
Fig. 6
figure 6

The Ra and Rq values of the ZnO samples without (tseed layer = 0 s) and with the seed layers electrochemically deposited at different tseed layer values

Full size image

For the optoelectronic applications, the major characteristic of ZnO is its wide energy bandgap [15]. To reveal the effect of the tseed layer on the optical energy bandgap, the absorption spectra of the electrochemically fabricated ZnO samples were measured. Figure 7 presents the obtained optical absorption spectra with a wavelength range from 350 to 600 nm. All samples without and with the seed layer displayed a high absorption in ultraviolet region and a low absorption in the visible region due to the wide energy bandgap of ZnO. The energy bandgap values (Eg) with respect to the tseed layer were evaluated using the data acquired from the absorption spectra by applying the Tauc method. According to the Tauc method, the relationship between the absorption coefficient (α) and the photon energy (\(h\nu \)) for near-edge absorption in semiconductor materials is given by [14, 21, 23]:

$$ (\alpha h\nu )^{1/m} = k(h\nu - E_{{\text{g}}} ) $$
Fig. 7
figure 7

Optical absorption spectra of the electrochemically fabricated ZnO samples without and with the seed layers obtained at different tseed layer values

Full size image

where k is a constant. In this relation, m was taken as 1/2 for an allowed direct energy bandgap and then (αhν)2 was plotted against hν. Finally, the direct energy bandgap values were estimated from the intercept of the straight–line portion of the (αhν)2 versus hν plot on the x-axis where y-axis was zero. The obtained plots and energy bandgap values are demonstrated in Fig. 8. The direct energy bandgap was found to be 3.318 eV for the ZnO NRs without the seed layer. Compared to the ZnO NRs without the seed layer, the subsequent ZnO NRs fabricated on the seed layers exhibited a wider energy bandgap value. Similar enlargement in the energy bandgap of the ZnO samples by applying seed layer pretreatment process was also reported in previous studies [14, 21]. It was also determined that the direct energy bandgaps for the tseed layer values of 25 s, 50 s and 100 s were close to each other (3.376, 3.368 and 3.364 eV, respectively). Those values of the direct energy bandgap found for the ZnO NRs fabricated in our present work are compatible with the relevant literature [9, 14, 29, 30]. The increase or decrease in the energy bandgap of the samples electroplated in the present study can be explained to the decrease or increase in the mean crystallite size of the samples, respectively (Table 1). The tseed layer dependence of the energy bandgap revealed in our present study has also been observed by Ait hssi et al. in a very recent study [20]. In that work, it was reported that all ZnO NRs grown at different tseed layer values (10 s, 30 s, 60 s and 90 s) displayed the energy bandgaps in the range of 3.24 ± 0.01 eV, which supports our findings. Accordingly, the results obtained in the present study showed that the ZnO NRs with a wider energy bandgap can be obtained by applying the seed layer process. It was also shown that the diameter and surface roughness of the electrochemically fabricated ZnO NRs can be noticeably reduced and the strength of the c-axis orientation can be considerably improved by controlling the tseed layer. The structural, morphological and optical characteristics of the samples are also listed in Table 1 as a function of the tseed layer.

Fig. 8
figure 8

Plots of (α \(h\nu \))2 versus \(h\nu \) and direct energy bandgaps for the ZnO samples without and with the seed layers electroplated at different tseed layer values

Full size image

4 Conclusions

In this paper, we studied the influence of the tseed layer on the structural properties, optical energy bandgap, diameter, growth orientation, and surface roughness of the ZnO NRs prepared by two-step ED technique. The ZnO seed layers were electroplated on the ITO-coated glass substrates under a constant current density of – 3 mA/cm2 while the subsequent ZnO NRs were electrochemically deposited at a constant cathode potential of – 0.72 V versus Ag/AgCl electrode. For comparison, the ZnO NRs was also deposited on the ITO-covered glass substrates without the seed layer. The results showed that the structural and morphological properties of the subsequent ZnO NRs are influenced greatly by the tseed layer. The shape of the ZnO NRs obtained at the tseed layer values of 25 s and 50 s was completely hexagonal. However, for the samples without the seed layer and with the seed layer grown at the tseed layer of 100 s, spherical-ended NRs were also detected in some regions on the sample surfaces. Compared to other samples, hexagonal-shaped ZnO NRs with the smallest mean diameter (~ 120 nm) and the lowest average surface roughness (46.8 nm) were obtained for the sample fabricated at the tseed layer of 50 s. The XRD spectra of the resultant samples revealed that all ZnO NRs have the hexagonal wurtzite crystal structure with no secondary or impurity diffraction peaks. The ZnO NRs without the seed layer had no specific growth orientation while the c-axis orientation was the preferential growth direction for the ZnO NRs with the seed layers. Furthermore, the ZnO NRs fabricated at the tseed layer of 50 s exhibited the highest I(002)/I(101) integral intensity ratio, indicating the creation of the best c-axis orientation being perpendicular to the ITO substrate surface. The results also showed that the samples have nano-sized crystallites ranging from 50 to 56 nm. The energy bandgap of the ZnO NRs without the seed layer was found to be 3.318 eV while the energy bandgaps of the ZnO NRs prepared on the seed layers were in the range of 3.370 ± 0.006 eV. It was also observed that the change of the energy bandgap of the samples was inversely proportional to the change of the mean crystallite size. In conclusion, the findings of the present study suggest that the tseed layer is one of the important key factors affecting the diameter, surface roughness and c-axis orientation of the ZnO NRs electrochemically fabricated from aqueous ZnCl2 electrolyte solution on the ITO-covered glass substrates. The ZnO NRs with 50 s of tseed layer exhibited the best vertical orientation, the smoothest top surface, the smallest mean diameter and a wide energy bandgap of 3.368 eV, making it a promising candidate for applications in optoelectronics. The improvements reported in this paper have high potential to contribute to many related investigations and applications of ZnO materials in the future.