Electrodeposition of ZnO/Cu2O Heterojunctions on Ni-Mo-P Electroless Coating

Electroless Ni-Mo-P coatings were deposited onto ceramic tiles in order to be employed as electrodes for the electrodeposition of ZnO and Cu2O heterojunction layers. Varying conditions, such as duration, annealing of the electroless coating and applied potential, and duration for ZnO electrodeposition were studied in order to optimize the properties of the ZnO/Cu2O heterojunctions toward improved photoelectrical performance. The coatings were evaluated in terms of morphology, crystalline structure, and by electrochemical and photoelectrical means. The obtained results indicated that a prolonged annealing treatment at low temperature is beneficial to improve the roughness and electrical conductivity of the Ni-Mo-P coating to further enhance the electrodeposition of ZnO. The morphology analysis revealed continuous and homogeneous Ni-Mo-P coatings. The formation of cube-like Cu2O crystals with larger grain size was induced by increasing the deposition duration of ZnO. The properties of ZnO layer are much improved when a higher cathodic potential is applied (−0.8 V) for 1 h, resulting in optimum photoelectric parameters as 1.44 mA·cm−2 for the JSC and 760.23 µV for the VOC value, respectively, for the corresponding heterojunction solar cell.


Introduction
Ceramic materials show excellent chemical stability, high hardness and wear resistance, and have higher melting temperature in comparison to other materials [1,2]. Owing to these properties, ceramic materials are employed in various fields such as heat exchangers, motor parts, electrical and electronic components, implants [1], as well as in electrical [3] and power electronic systems [4], as mechanical supports [5], and building-integrated photovoltaics (BIPV) [6].
The ceramics show an important drawback in some applications, namely they are poor conductors of electricity and heat and they exhibit high fragility [2]. In this regard, surface modification by methods including electroless deposition (ED) attracted attention in order to improve the electrical properties of these non-conductive surfaces (polymer, ceramics) without using external current [7][8][9][10]. The ED involves the following steps: (i) surface treatment, (ii) catalytic activation, (iii) electroless process, and (iv) annealing treatment [11]. The deposition bath is formed by a metal salt, a complexing and reducing agent, a stabilizer, an inhibitor, and others (pH regulators, wetting and polishing agents) [11,12]. Factors such as temperature, time, pH, and chemical composition directly influence the morphology, composition, and final structure of electroless coating, as well as the electrical and magnetic properties [12][13][14]. For example, thickness and size grain dependence were observed with

Electroless Plating of the Ceramic Surface
First, the ceramic surface was activated with palladium catalyst at 300 • C for 12 h [8]. Then, the electroless plating of Ni-Mo-P film was performed at 80 • C and 300 rpm stirring rate from a bath composed of: NiSO 4 , Na 2 MoO 4 , NaH 2 PO 2 , C 6 H 12 O 7 , and Na 2 C 4 H 4 O 6 in concentration (g L −1 ) of 7, 3.9, 10.0, 15.0, and 10, respectively. NH 4 OH was used to adjust the pH to 10. The metallized ceramic was further rinsed with water and air-dried. The plating time was varied from 30 to 300 min while the annealing treatment in vacuum conditions ranged from 160 • C for 16 h to 400 • C for 1 h (the metallized ceramic substrates were denoted as CS x-y where x is the plating time and y is the applied heat treatment). These substrates were used to fabricate the p-n heterojunctions solar cells.

Electrodeposition of ZnO on Ni-Mo-P/Ceramic Substrate
In order to fabricate the p-n heterojunction SCs using metallized ceramic substrates, electrodeposition (ELD) technique was selected. ZnO was electrodeposited onto Ni-Mo-P-coated ceramic substrates from 5 mM ZnCl 2 and 0.1 M KCl [29] in a deposition potential ranging from −0.6 to −0.8 V for 30 and 60 min at 75 • C in order to analyze the growth of ZnO layer and its effect on the performance of heterojunction SC using different substrates. In order to reference the electrodeposition of ZnO to the literature, a different batch of ceramic electrodes were obtained by coating the ceramic with an ITO layer with an electrical conductivity of about 1.8 µΩ cm and thickness of 300 nm.

Fabrication of Cu 2 O/ZnO Heterojunction Solar Cells on Ni-Mo-P/Ceramic Substrate
Following the electrodeposition of ZnO layer, a Cu 2 O layer was electrodeposited at 35 • C, −0.6 V for 300 min from an electrolyte composed of 0.4 M CuSO 4 , 3 M C 3 H 6 O 3 , and 4 M NaOH [55]. The final device was sealed with conductive carbon cement (Leit-C, Agar Scientific, Essex, UK) at the Cu 2 O side [29,30].

Methods
All the electrochemical measurements were performed using a conventional three-electrode glass cell and potentiostat (PGSTAT 101 AUTOLAB, Metrohm, Madrid, Spain). The working electrode was the modified ceramic substrate while a Pt foil and an Ag/AgCl in saturated KCl electrode served as counter-electrode and reference electrode, respectively.
The deposition of ITO layer as reference electrode was achieved by DC magnetron sputtering (K675X, Emitech, Fall River, MA, USA) technique in Ar atmosphere, by using an In 2 O 3 /SnO 2 90/10 wt % target (Kurt J. Lesker, East Sussex, UK). An annealing treatment was further performed in a tubular oven (Carbolite) at 260 • C for 30 min in N 2 with 4% H 2 atmosphere.
The structural properties of the electroless coating and ZnO/Cu 2 O heterojunction synthetized onto ceramic substrates were investigated by X-ray diffraction (XRD) using an Ultima IV diffractometer (Rigaku, Barcelona, Spain) in the Bragg-Bentano configuration with CuKα radiation (1.54 Å). Atomic force microscopy (AFM, Multimode 8, Bruker, Madrid, Spain) and field emission scanning electron microscopy (FESEM, Bruker, Madrid, Spain working voltage 2 kV) techniques were used to determine the roughness and morphology of coatings. The electrical resistivity of coatings was determined by a four-point probe. The electrical properties of the solar cells were analyzed with a Sun 2000 Solar Simulator (ABET Technologies, Madrid, Spain) working with 800 W·m −2 irradiation on the ZnO window in agreement to the scheme in Figure 1 [45,56,57].

Electrodeposition Study
The first approach in the study of ZnO electrodeposition onto the ceramic substrate considered an ITO coating onto the ceramic surface in order to have a reference point of view on ZnO electrodeposition with respect to the literature. The ZnO electrodeposition onto ITO/ceramic substrate was performed by potentiostatic mode at the deposition potential −0.8 V vs. Ag/AgCl, as indicated by previous studies [29]. The related chronoamperogram is further presented in Figure 2A. The evolution of the process is found to agree with reported results, and the nucleation process is reached in about 100 s (see inset) [58] while the current plateau value establishes at about 0.75 mA cm −2 . These results indicate that the properties of ITO coating onto the ceramic surface allow the nucleation and growth of ZnO structure by offering the necessary nucleation centers. Therefore, the ceramic substrate could be successfully employed as support for the fabrication of ZnO/Cu2O heterojunction upon coating with a conductive layer.
Further, Ni-Mo-P coating was deposited onto the ceramic surface in order to be employed as substrate for the electrodeposition of ZnO films. Given that morphology and electrical properties of the substrate affect the nucleation and growth of an electrodeposited film [59], the effect of plating duration and annealing treatment of Ni-Mo-P coating onto the ZnO electrodeposition were studied as shown in Figure 2B-D. First, a cathodic linear sweep was performed in order to investigate the deposition process of ZnO-see Figure 2B. It can be observed that the reduction current increases above −0.4 V and a peak is formed at about −0.6 V which is attributed to the reduction process and formation of ZnO layer. The reduction peak is followed by a steady increase in the cathodic current and a steep increase in current above −1 V which is attributed to the deposition of Zn metal [60]. Therefore, it is indicated that the optimum electrodeposition potential of ZnO onto Ni-Mo-P coating ranges from −0.6 V to −0.8 V vs. Ag/AgCl. The other conditions for deposition of the Ni-Mo-P coating resulted in similar evolution of deposition potential for ZnO, however differences in the current response were recorded as described further.
The increase in Ni-Mo-P electroless deposition duration (30 to 300 min) was observed to induce a decrease in electrical resistivity from 10.6 µΩ·cm to 0.843 µΩ·cm [61]. Consequently, the cathodic current plateau value for ZnO electrodeposition at −0.8 V on corresponding substrates (see Figure  2C) increased because of an improved morphology and more active sites created at the surface of the Ni-Mo-P coating [62].
On the other hand, an annealing treatment was indicated to improve the coating conductivity [63]. In this study, the electrical resistivity of the electroless coating obtained at 300 min further decreased upon annealing treatment at 160 °C for 16 h, reaching a value of 0.210 Ω·sq −1 . The ZnO nucleation and growth process were observed to be better defined on the annealed Ni-Mo-P coating and as such, the current plateau increased [63]. It is evident that the nucleation is faster in the case of annealed Ni-Mo-P coating (about 250 s), while the current plateau is more stable, as well, and it indicates a more homogeneous growth of ZnO.

Electrodeposition Study
The first approach in the study of ZnO electrodeposition onto the ceramic substrate considered an ITO coating onto the ceramic surface in order to have a reference point of view on ZnO electrodeposition with respect to the literature. The ZnO electrodeposition onto ITO/ceramic substrate was performed by potentiostatic mode at the deposition potential −0.8 V vs. Ag/AgCl, as indicated by previous studies [29]. The related chronoamperogram is further presented in Figure 2A. The evolution of the process is found to agree with reported results, and the nucleation process is reached in about 100 s (see inset) [58] while the current plateau value establishes at about 0.75 mA cm −2 . These results indicate that the properties of ITO coating onto the ceramic surface allow the nucleation and growth of ZnO structure by offering the necessary nucleation centers. Therefore, the ceramic substrate could be successfully employed as support for the fabrication of ZnO/Cu 2 O heterojunction upon coating with a conductive layer.
Further, Ni-Mo-P coating was deposited onto the ceramic surface in order to be employed as substrate for the electrodeposition of ZnO films. Given that morphology and electrical properties of the substrate affect the nucleation and growth of an electrodeposited film [59], the effect of plating duration and annealing treatment of Ni-Mo-P coating onto the ZnO electrodeposition were studied as shown in Figure 2B-D. First, a cathodic linear sweep was performed in order to investigate the deposition process of ZnO-see Figure 2B. It can be observed that the reduction current increases above −0.4 V and a peak is formed at about −0.6 V which is attributed to the reduction process and formation of ZnO layer. The reduction peak is followed by a steady increase in the cathodic current and a steep increase in current above −1 V which is attributed to the deposition of Zn metal [60]. Therefore, it is indicated that the optimum electrodeposition potential of ZnO onto Ni-Mo-P coating ranges from −0.6 V to −0.8 V vs. Ag/AgCl. The other conditions for deposition of the Ni-Mo-P coating resulted in similar evolution of deposition potential for ZnO, however differences in the current response were recorded as described further.
The increase in Ni-Mo-P electroless deposition duration (30 to 300 min) was observed to induce a decrease in electrical resistivity from 10.6 µΩ·cm to 0.843 µΩ·cm [61]. Consequently, the cathodic current plateau value for ZnO electrodeposition at −0.8 V on corresponding substrates (see Figure 2C) increased because of an improved morphology and more active sites created at the surface of the Ni-Mo-P coating [62].
On the other hand, an annealing treatment was indicated to improve the coating conductivity [63]. In this study, the electrical resistivity of the electroless coating obtained at 300 min further decreased upon annealing treatment at 160 • C for 16 h, reaching a value of 0.210 Ω·sq −1 . The ZnO nucleation and growth process were observed to be better defined on the annealed Ni-Mo-P coating and as such, the current plateau increased [63]. It is evident that the nucleation is faster in the case of annealed Ni-Mo-P coating (about 250 s), while the current plateau is more stable, as well, and it indicates a more homogeneous growth of ZnO.
Furthermore, the electrodeposition of ZnO was studied with the deposition potential at the surface of Ni-Mo-P coated ceramic that performed best, namely the one coated with Ni-Mo-P film deposited for 300 min and annealed at 160 • C for 16 h. Figure 2D shows the current transients corresponding to the formation of ZnO onto such substrates. As it can be seen, the current plateau increases with the applied potential, which is in agreement with literature, indicating an improved deposition rate at an overpotential value [64]. The nucleation process was observed to take place faster and change from a progressive trend toward an instantaneous one with the increase in the deposition potential. Furthermore, the electrodeposition of ZnO was studied with the deposition potential at the surface of Ni-Mo-P coated ceramic that performed best, namely the one coated with Ni-Mo-P film deposited for 300 min and annealed at 160 °C for 16 h. Figure 2D shows the current transients corresponding to the formation of ZnO onto such substrates. As it can be seen, the current plateau increases with the applied potential, which is in agreement with literature, indicating an improved deposition rate at an overpotential value [64]. The nucleation process was observed to take place faster and change from a progressive trend toward an instantaneous one with the increase in the deposition potential.

Morphology Analysis
Beside the electrical properties, the substrate roughness is also known to highly affect the properties of an electrodeposited film [65]. Therefore, morphology analysis was performed on the coated substrates by AFM technique. Figure 3 presents the AFM micrographs of the ceramic

Morphology Analysis
Beside the electrical properties, the substrate roughness is also known to highly affect the properties of an electrodeposited film [65]. Therefore, morphology analysis was performed on the coated substrates by AFM technique. Figure 3 presents the AFM micrographs of the ceramic substrates upon coating with ITO and Ni-Mo-P films in terms of amplitude and topography. Initially, the bare ceramic substrate was found to exhibit a roughness value (Rq) of 412 ± 40 nm (not shown). Upon coating with ITO the ceramic surface showed a slight decrease in the roughness down to 396.8 ± 20 nm [48]-see Figure 3a,b. Coatings 2020, 10, x FOR PEER REVIEW 6 of 14 Figure 3c-f shows the appearance of the Ni-Mo-P coatings obtained with a palladium activation process at 300 °C for 12 h and autocatalytic chemical reduction at 80 °C for 30 and 300 min, respectively. Under these conditions, the coatings show an increment of roughness from ~76 to ~223 nm, attributed to the increase of the electroless deposition duration. Nevertheless, the AFM micrographs show the Ni-Mo-P coatings are homogeneous. Figure 4 shows the effect of annealing on the Ni-Mo-P coating properties by depicting the FE-SEM micrographs of the ceramic substrates with a Ni-Mo-P coating in different conditions. As observed in Figure 4a, Ni-Mo-P coating annealed at low temperature as 160 °C for 16 h shows uniform, continuous, and compact nodular morphology. In order to obtain a smoother coating, the annealing temperature was increased to 400 °C, however a network of cracks breaking the continuous structure of the layer was observed (Figure 4b). This defect may be due to the increase in the crystallization degree with temperature [66] or thermal expansion coefficient mismatch between coating and substrate. For such a temperature value, only 1 h of annealing was studied, given the detrimental effects on the continuity of the coating. In this context, other authors indicate that the application of a thermal treatment with a temperature higher than 350 °C can negatively affect the morphology and topography of the surface of the coating [67,68]. The annealing treatment slightly Figure 3c-f shows the appearance of the Ni-Mo-P coatings obtained with a palladium activation process at 300 • C for 12 h and autocatalytic chemical reduction at 80 • C for 30 and 300 min, respectively. Under these conditions, the coatings show an increment of roughness from~76 to~223 nm, attributed to the increase of the electroless deposition duration. Nevertheless, the AFM micrographs show the Ni-Mo-P coatings are homogeneous. Figure 4 shows the effect of annealing on the Ni-Mo-P coating properties by depicting the FE-SEM micrographs of the ceramic substrates with a Ni-Mo-P coating in different conditions. As observed in Figure 4a, Ni-Mo-P coating annealed at low temperature as 160 • C for 16 h shows uniform, continuous, and compact nodular morphology. In order to obtain a smoother coating, the annealing temperature was increased to 400 • C, however a network of cracks breaking the continuous structure of the layer was observed (Figure 4b). This defect may be due to the increase in the crystallization degree with temperature [66] or thermal expansion coefficient mismatch between coating and substrate. For such a temperature value, only 1 h of annealing was studied, given the detrimental effects on the continuity of the coating. In this context, other authors indicate that the application of a thermal treatment with a temperature higher than 350 • C can negatively affect the morphology and topography of the surface of the coating [67,68]. The annealing treatment slightly reduced the surface resistivity from 0.210 Ω·sq −1  The evolution of roughness and morphology of the Ni-Mo-P coating with plating duration and annealing treatment is in high agreement with the electrochemical results on ZnO growth. It is known that the density and growth orientation of ZnO nanostructures further affect the morphology properties of Cu2O. Figure 5 depicts the effect of ZnO electrodeposited at −0.8 V on the morphology of Cu2O films by SEM. While the presence of ZnO structures is not appreciated in the images, a good Cu2O coverage is observed which is indicative of good interaction and crystallographic quality of the Cu2O/ZnO interface [69,70]. The morphology of the Cu2O layer is markedly influenced by ZnO electrodeposition time, namely with an increase in electrodeposition duration from 30 min to 60 min, cubic structures of Cu2O with a larger grain size are obtained, which is indicative of a thicker layer of Cu2O ( Figure 5b) and it is expected to improve the light absorption because of the fact that there are fewer grain edges [70].

Structure Analysis
The structure and orientation of the electrodeposited heterojunctions were further analyzed by XRD measurements. First, the diffraction patterns of Ni-Mo-P coating obtained for varying duration are depicted in Figure 6A. The results indicate the metallic Ni-Mo-P coatings present a Ni crystalline structure because of the presence of the reflection peaks of Ni (111), corresponding to the cubic phase centered on the face of Ni [71] and Ni (200) which appear around 44° and 51.3° [72]. Furthermore, the increase in the duration of electroless plating results in an increase in crystallite size which is evidenced in the reduction in full width of half maximum (FWHM) of the predominant peak [73].
The corresponding XRD diffraction patterns of the Cu2O/ZnO heterojunctions electrodeposited on Ni-Mo-P films plated for varying duration with or without annealing are further shown in Figure  (Figure 5b) and it is expected to improve the light absorption because of the fact that there are fewer grain edges [70].
Coatings 2020, 10, x FOR PEER REVIEW 7 of 14 reduced the surface resistivity from 0.210 Ω·sq −1 (coating annealed at 160 °C for 16 h) to 0.185 Ω·sq −1 (coating annealed at 400 °C for 1 h) being attributed to improved crystallinity [63]. The evolution of roughness and morphology of the Ni-Mo-P coating with plating duration and annealing treatment is in high agreement with the electrochemical results on ZnO growth. It is known that the density and growth orientation of ZnO nanostructures further affect the morphology properties of Cu2O. Figure 5 depicts the effect of ZnO electrodeposited at −0.8 V on the morphology of Cu2O films by SEM. While the presence of ZnO structures is not appreciated in the images, a good Cu2O coverage is observed which is indicative of good interaction and crystallographic quality of the Cu2O/ZnO interface [69,70]. The morphology of the Cu2O layer is markedly influenced by ZnO electrodeposition time, namely with an increase in electrodeposition duration from 30 min to 60 min, cubic structures of Cu2O with a larger grain size are obtained, which is indicative of a thicker layer of Cu2O ( Figure 5b) and it is expected to improve the light absorption because of the fact that there are fewer grain edges [70].

Structure Analysis
The structure and orientation of the electrodeposited heterojunctions were further analyzed by XRD measurements. First, the diffraction patterns of Ni-Mo-P coating obtained for varying duration are depicted in Figure 6A. The results indicate the metallic Ni-Mo-P coatings present a Ni crystalline structure because of the presence of the reflection peaks of Ni (111), corresponding to the cubic phase centered on the face of Ni [71] and Ni (200) which appear around 44° and 51.3° [72]. Furthermore, the increase in the duration of electroless plating results in an increase in crystallite size which is evidenced in the reduction in full width of half maximum (FWHM) of the predominant peak [73].
The corresponding XRD diffraction patterns of the Cu2O/ZnO heterojunctions electrodeposited on Ni-Mo-P films plated for varying duration with or without annealing are further shown in Figure

Structure Analysis
The structure and orientation of the electrodeposited heterojunctions were further analyzed by XRD measurements. First, the diffraction patterns of Ni-Mo-P coating obtained for varying duration are depicted in Figure 6A. The results indicate the metallic Ni-Mo-P coatings present a Ni crystalline structure because of the presence of the reflection peaks of Ni (111), corresponding to the cubic phase centered on the face of Ni [71] and Ni (200) which appear around 44 • and 51.3 • [72]. Furthermore, the increase in the duration of electroless plating results in an increase in crystallite size which is evidenced in the reduction in full width of half maximum (FWHM) of the predominant peak [73]. the crystallite size decreases upon growing the ZnO layer onto a Ni-Mo-P coating annealed at 160 °C, which suggests a greater number of p-n junctions with Cu2O. However, when applying a higher temperature annealing treatment (400 °C for 1 h) to the Ni-Mo-P substrate, the ZnO (100) peak is shifted probably because of stress in the crystal lattice. On the other hand, cubic structure of Cu2O layer was identified by the presence of planes (110), (111), (200), and (211) (JCPDS card 00-005-0667) [45]. The effect of the electroplating time of the ZnO layer was further investigated. Ni-Mo-P coating annealed at 160 °C for 16 h was employed as substrate, as it exhibited the best performance for ED. The corresponding diffractograms of the Cu2O/ZnO heterojunctions are shown in Figure 6C. The diffraction patterns are similar to the previous one, with wurtzite hexagonal crystalline structure and a preferential orientation in the plane (101) for the ZnO layer. The characteristic peak ZnO (100) (see inset) slightly changed in terms of FWHM and position which is indicative of smaller crystallite size [77] with the increase in the electrodeposition time to 60 min, which suggests an improved density of heterojunctions [29,78,79]. Regarding the synthesized Cu2O layers, they showed similar pattern as previously shown with the previous samples. The corresponding XRD diffraction patterns of the Cu 2 O/ZnO heterojunctions electrodeposited on Ni-Mo-P films plated for varying duration with or without annealing are further shown in Figure 6B. Ni-Mo-P coating peaks (labeled with *) appear at 44 • and 52 • , the other peaks are typical of the ceramic substrate. The presence of ZnO layer is clearly evidenced in agreement with the JCPDS identification card 00-036-1451 [43], corresponding to a wurtzite-type hexagonal crystalline structure with a (101) preferential orientation which was attributed to branched ZnO nanowires [74] as an effect of applied temperature [75]. This peak is overlapped with a Cu 2 O one, namely (111), in agreement with other reports [76]. The analysis of the characteristic peak ZnO (100) (inset) indicates the crystallite size decreases upon growing the ZnO layer onto a Ni-Mo-P coating annealed at 160 • C, which suggests a greater number of p-n junctions with Cu 2 O. However, when applying a higher temperature annealing treatment (400 • C for 1 h) to the Ni-Mo-P substrate, the ZnO (100) peak is shifted probably because of stress in the crystal lattice. On the other hand, cubic structure of Cu 2 O layer was identified by the presence of planes (110), (111), (200), and (211) (JCPDS card 00-005-0667) [45].
The effect of the electroplating time of the ZnO layer was further investigated. Ni-Mo-P coating annealed at 160 • C for 16 h was employed as substrate, as it exhibited the best performance for ED. The corresponding diffractograms of the Cu 2 O/ZnO heterojunctions are shown in Figure 6C. The diffraction patterns are similar to the previous one, with wurtzite hexagonal crystalline structure and a preferential orientation in the plane (101) for the ZnO layer. The characteristic peak ZnO (100) (see inset) slightly changed in terms of FWHM and position which is indicative of smaller crystallite size [77] with the increase in the electrodeposition time to 60 min, which suggests an improved density of heterojunctions [29,78,79]. Regarding the synthesized Cu 2 O layers, they showed similar pattern as previously shown with the previous samples.

Photoelectical Properties of Cu 2 O/ZnO Heterojunction Solar Cells Supported onto Ni-Mo-P-Coated Ceramic
The short circuit current density (J SC ) and open circuit voltage (V OC ) photoelectric parameters obtained for varying Cu 2 O/ZnO heterojunction solar cells supported onto Ni-Mo-P coated ceramic are shown in Table 1. It was observed that all the Cu 2 O/ZnO heterojunction solar cells show photoelectric properties, however, the Cu 2 O/ZnO heterojunction solar cell fabricated onto ITO coating exhibited the lowest J SC values which could be explained by the properties of the ITO layer that need optimizing. The photoelectric properties of the heterojunction SC induced by Ni-Mo-P plating conditions were analyzed. Thus, by increasing the plating duration for Ni-Mo-P film from 30 to 300 min, the heterojunction SC exhibited improved V OC value while the J SC values slightly decreased. This improvement is mainly attributed to the improved homogeneity and low resistivity of the electroless Ni-Mo-P coating. Upon annealing at 160 • C for 16 h of the Ni-Mo-P plated for 300 min, a significant improvement in the photoelectric properties is observed, that is the J SC markedly increased while the V OC slightly decreased, which could be explained by improved properties of ZnO layer derived from smaller lattice mismatch of the Ni-Mo-P coating and the ZnO layer as well as the reduction in resistivity of Ni-Mo-P. Clearly, an annealing temperature as high as 400 • C for the Ni-Mo-P coating negatively affects the photoelectric properties of the heterojunction SC as J SC markedly drops. This decrease may be due to the cracks in the coating as indicated by microscopy results that causes short circuits reducing the SC performance [80]. Such results show the optimum conditions among the parameters studied for Ni-Mo-P plating, which are duration of 300 min and the application of an annealing treatment at 160 • C for 16 h.
The photoelectric properties of the heterojunction SC supported on Ni-Mo-P-coated ceramic (electroless plating 300 min with heat treatment at 160 • C for 16 h) as induced by ZnO electrodeposition conditions were further investigated and presented in Table 2. It was observed that increasing the electrodeposition potential of ZnO layer markedly improves the J SC value of the SC while V OC decreases, which indicates good crystallinity of the ZnO layer. By further increasing the deposition duration, the J SC value reached 1.44 mA·cm −2 and V OC improved. This improvement in the photoelectric properties (J SC ) may be due to the good interconnection and crystallographic quality of the Cu 2 O/ZnO interface and improved properties of ZnO nanostructures such as density and aspect ratio that could improve charge transport [69]. The increase in V OC value is an indication of reduced grain limits and improved light absorption [70].

Conclusions
A ZnO/Cu 2 O heterojunction type solar cell was fabricated onto ceramic substrate by simple electrodeposition technique. The electrical functionalization of the ceramic surface has been achieved by electroless coating of a Ni-Mo-P film. Continuous and homogeneous Ni-Mo-P coatings were obtained with typical XRD reflection planes. The results indicated the morphology and electrical conductivity of Ni-Mo-P film can be easily tailored by adjusting the electroless coating conditions such as duration and by applying an appropriate annealing treatment, namely lower temperature for long time. The increase in applied potential and duration for ZnO electrodeposition, namely −0.8 V for 1 h resulted in the best performance of the ZnO/Cu 2 O heterojunction type solar cell, that is, 1.44 mA·cm −2 for the J SC and 760.23 µV for the V OC value, respectively, which demonstrates the importance of electrodeposition of ZnO layer toward improving the photoelectrical parameters of the heterojunction solar cell. Although the performance still need improvements of the photoelectric parameters, the obtained results show that an insulating substrate (in this case a ceramic tile) could be used for electrical power generation applications by applying a Ni-Mo-P electroless coating. The results of this work indicate that optimized electrochemical growth of the heterojunction layers, annealing treatment, and the tailoring of the ZnO/Cu 2 O interface are parameters of paramount importance for solar cell performance improvement.