Electrochemical Synthesis of Highly Oriented, Transparent, and

Article pubs.acs.org/Langmuir

Electrochemical Synthesis of Highly Oriented, Transparent, and Pinhole-Free ZnO and Al-Doped ZnO Films and Their Use in Heterojunction Solar Cells Donghyeon Kang, Dongho Lee, and Kyoung-Shin Choi* Department of Chemistry, University of WisconsinMadison, Madison, Wisconsin 53706, United States S Supporting Information *

ABSTRACT: Electrochemical synthesis conditions using nonaqueous solutions were developed to prepare highly transparent (T > 90%) and crystalline ZnO and Al-doped ZnO (AZO) films for use in solar energy conversion devices. A focused effort was made to produce pinhole-free films in a reproducible manner by identifying a key condition to prevent the formation of cracks during deposition. The polycrystalline domains in the resulting films had a uniform orientation (i.e., the c-axis perpendicular to the substrate), which enhanced the electron transport properties of the films. Furthermore, electrochemical Al doping of ZnO using nonaqueous media, which was demonstrated for the first time in this study, effectively increased the carrier density and raised the Fermi level of ZnO. These films were coupled with an electrodeposited p-type Cu2O to construct p-n heterojunction solar cells to demonstrate the utilization of these films for solar energy conversion. The resulting n-ZnO/p-Cu2O and n-AZO/p-Cu2O cells showed excellent performance compared with previously reported n-ZnO/p-Cu2O cells prepared by electrodeposition. In particular, replacing ZnO with AZO resulted in simultaneous enhancements in short circuit current and open circuit potential, and the n-AZO/pCu2O cell achieved an average power conversion efficiency (η) of 0.92 ± 0.09%. The electrodeposition condition reported here will offer a practical and versatile way to produce ZnO or AZO films, which play key roles in various solar energy conversion devices, with qualities comparable to those prepared by vacuum-based techniques.

1. INTRODUCTION ZnO and Al-doped ZnO (AZO), in which Al doping increases the electrical conductivity of ZnO, have served as key components in various solar energy conversion devices. They are frequently used as n-type window or buffer layers for heterojunction solar cells.1−5 For example, Cu(In,Ga)Se2-based solar cells achieving a conversion efficiency greater than 20% employed both ZnO and AZO layers in their assemblies (e.g., AZO/ZnO/CdS/Cu(In,Ga)Se2).1 An AZO layer has also been used in the construction of Cu2O-based photocathodes where the AZO layer served as a buffer layer between the Cu2O photon absorber layer and the TiO2 protection layer, playing an important role in preventing photocorrosion and enhancing photocurrent of the p-type Cu2O photocathode during solar water reduction.6−9 For the applications mentioned above, ZnO and AZO need to be prepared as thin conformal layers that are highly transparent, pinhole free, and conductive. To satisfy these requirements, vacuum-based techniques such as chemical vapor deposition (CVD), atomic layer deposition (ALD), and radio frequency (RF) magnetron sputtering at elevated temperatures have predominantly been employed.10−15 However, mass production of large scale electrodes using these vacuum-based techniques may not be cost-effective. Therefore, it would be © XXXX American Chemical Society

desirable to develop easily scalable solution-based techniques that can produce ZnO and AZO as highly transparent, pinhole free, and crystalline films at ambient conditions. Herein, we report solution-based electrochemical routes to produce crystalline ZnO and AZO films that are highly transparent and pinhole free at mild conditions (at 85 °C under ambient pressure). Polycrystalline domains of the resulting films are found to uniformly align with the [002] direction perpendicular to the substrate. This is advantageous for improving the conductivity of the ZnO films because the electron mobility of ZnO is reported to be the greatest along the c-axis.16,17 In this study, in addition to reporting the synthesis and characterization of electrodeposited ZnO and AZO films, we demonstrate the construction of solid state p-n heterojunction solar cells by coupling these films with an electrodeposited p-type Cu2O absorber layer, as one example of utilizing these films for solar energy conversion. The resulting solar cells exhibit promising performances compared with previously reported all-electrodeposited n-ZnO/p-Cu2O cells. Comparing the performances of n-ZnO/p-Cu2O and n-AZO/ Received: May 18, 2016 Revised: August 14, 2016

A

DOI: 10.1021/acs.langmuir.6b01902 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

binding energies were calibrated with respect to the residual carbon 1s peak at 284.6 eV. The J−V measurements of the n-ZnO/p-Cu2O and n-AZO/p-Cu2O solar cells were performed using a SP-200 potentiostat/EIS (BioLogic Science Instrument) and simulated solar illumination obtained by passing light from a 300 W Xe arc lamp through neutral density filters, an AM 1.5G filter, and an IR filter (water filter) into an optical fiber. The power density of light was calibrated to be 100 mW/cm2 at the ITO surface (before the light passed through ITO) by using both a thermopile detector (International Light) and an NREL certified reference cell (Photo Emission Tech. Inc.). A blunt tip, spring-loaded Au probe was used to make contacts to the Pd top contact spots and copper tape attached to the back ITO contact. All ITO/ZnO/Cu2O/ Pd and ITO/AZO/Cu2O/Pd samples were back-illuminated through the ITO back contact to avoid shadowing of the solar cell by the top contact and the Au probe. The solar cell performance results are summarized in Table S2. At least three measurements using three different cells were used to obtain average values and standard deviations. Incident photon-to-current efficiency (IPCE) under the short circuit condition was measured using AM 1.5G illumination from a 300 W Xe arc lamp passed through an AM 1.5G filter and neutral density filters. Monochromatic light was generated using an Oriel Cornerstone 130 monochromator with a 10 nm bandpass, and the power density of light at each wavelength was measured with a photodiode detector (International Light). Mott−Schottky plots of ZnO and AZO films were obtained by measuring capacitance using a SP-200 potentiostat/EIS (BioLogic Science Instrument). Measurements were carried out with a typical three-electrode system composed of a sample working electrode, a Ag/ AgCl (4 M KCl) reference electrode and a Pt counter electrode. A sinusoidal modulation of 10 mV was applied at frequencies of 0.5 and 1 kHz. A 0.1 M phosphate buffer solution (pH 11) purged with N2 was used as the electrolyte. All electrodes were masked with epoxy resin to expose the same geometrical area (0.05 cm2). The conversion between potentials vs Ag/AgCl and vs NHE was performed using the equation below.

p-Cu2O cells, where the ZnO and the AZO layers were prepared using the same method, ensuring the same film morphologies and features, also allows us to unambiguously discuss the advantages of Al-doping in ZnO films.

2. EXPERIMENTAL SECTION Electrodeposition. The electrodeposition was carried out in an undivided cell using a VMP2 multichannel potentiostat (Princeton Applied Research). A typical three-electrode system composed of an indium tin oxide (ITO) (sheet resistance of 8−12 Ω) working electrode, a Ag/AgCl (4 M KCl) reference electrode, and a Pt counter electrode was used. The Pt counter electrode was prepared by depositing 20 nm of Ti followed by 100 nm of platinum on clean glass slides by e-beam evaporation. For the electrodeposition of ZnO films, a DMSO (≥99.0% purity, Sigma-Aldrich) solution containing 50 mM Zn(NO3)2·6H2O (98% purity, Acros) was used as the plating solution. When necessary, varying amounts of high purity water (resistivity >18 MΩ) (1−8 mL) were added to 40 mL of a DMSO plating solution. Depositions were carried out at 85 °C and by passing 0.3 C/cm2 at −1.0 V vs Ag/AgCl. Passing a total charge of 0.3 C/cm2 resulted in the deposition of ∼300 nm thick ZnO films. When a more negative potential than −1.0 V vs Ag/AgCl was used, Zn metal was codeposited along with ZnO. On the other hand, when a more positive potential than −1.0 V vs Ag/AgCl was used (−1.0 V < E ≤ −0.8 V), films of a comparable quality could be obtained but it took a much longer time to pass 0.3 C/cm2 (e.g., ∼ 12 min at −1.0 V, 1 h at −0.9 V, and several hours at −0.8 V). The concentration of the precursor salt, Zn(NO3)2· 6H2O, was not varied in this study. For the deposition of Al-doped ZnO (AZO) films, 1 mL of high purity water containing varying amounts of Al(NO3)3·9H2O (≥98.0%, Sigma-Aldrich) was added to 40 mL of a DMSO solution containing 50 mM Zn(NO3)2·6H2O. The concentrations of Zn and Al in the plating solutions investigated in this study are listed in Table S1. Depositions were carried out at 85 °C and −1.0 V vs Ag/AgCl by passing 0.3 C/cm2, which is the same condition used for the deposition of undoped ZnO films. The as-deposited ZnO and AZO films were crystalline, and no post deposition heat treatment was performed. The n-ZnO/p-Cu2O and n-AZO/p-Cu2O cells were constructed by a sequential electrodeposition of p-Cu2O on the n-ZnO and n-AZO films. The Cu2O layer was cathodically electrodeposited using an aqueous solution containing 20 mM CuSO4 (anhydrous, 98% purity, Alfa Aesar) and 0.4 M L-(+)-lactic acid (85−90% aqueous solutions, Alfa Aesar). The solution pH was adjusted to 11 using NaOH. Electrodepositions of Cu2O were carried out at 60 °C by passing 1 C/ cm2 at −0.4 V vs Ag/AgCl. After deposition of Cu2O, Pd metal contacts were deposited by dc magnetron sputtering at 100 W using an Anatech USA, Hummer 8.3 Sputtering system with magnetron cathodes (US Inc., model MAK-2) under an Ar atmosphere. The Pd target was purchased from ACI Alloys (99.95% purity). The size of the Pd contact was 0.024 cm2. Characterization. The purity, crystallinity, and orientation of ZnO, AZO, and Cu2O films were examined by powder X-ray diffraction (XRD) (Bruker D8 Advanced PXRD, Ni-filtered Cu Kα radiation λ = 1.5418 Å) at room temperature. Scanning electron microscopy (SEM) images were taken on a LEO 1530 system operated at 5 kV. The atomic ratios of all films were obtained by the same scanning electron microscope equipped with an energy dispersive X-ray spectrometer (EDS) (Noran System Seven, Thermo Fisher) at an accelerating voltage of 10 kV. UV−vis absorption spectra were recorded using a Cary 5000 UV−vis-NIR spectrophotometer (Agilent). For the absorption measurements, the sample electrodes were placed in the center of an integrating sphere to collect all reflected and transmitted light to accurately estimate absorbance. For the transmittance measurement, the conventional setting measuring only the transmitted light was used. X-ray photoelectron spectroscopy (XPS) spectra were measured using a K-Alpha X-ray photoelectron spectrometer (Thermo Scientific) equipped with Al Kα excitation. The

E (vs NHE) = E (vs Ag/AgCl) + EAg/AgCl (reference)

(EAg/AgCl (reference) = 0.1976 V vs NHE at 25° C) Calculations. The carrier densities of ZnO and AZO were calculated using the following Mott−Schottky equation.

⎛ k T⎞ 1 2 ⎜V − Vfb − B ⎟ = 2 2 e ⎠ C eεZnOε0A ND ⎝ where C is the interfacial capacitance, e is the electronic charge (1.6 × 10−19 C), εZnO is the dielectric constant of ZnO (10), ε0 is the vacuum permittivity (8.85 × 10−12 F/m), A is the exposed electrode area (0.05 cm2), ND is the carrier density, V is the applied voltage, Vfb is the flat band potential, kB is the Boltzmann constant, and T is the absolute temperature.18,19 The Fermi levels (EF) of ZnO and AZO were calculated using the flatband potentials (Efb) obtained from the Mott−Schottky plots, the point of zero zeta potential (PZZP) of ZnO reported in the literature (8.8),20 and the pH of the solution where the Efb values were measured using the following equations.21,22

E F (in eV vs vacuum) = − Efb (in V vs NHE) + VH (in V) − 4.5 VH (in V) = 0.0592(PZZP − pH)

3. RESULTS AND DISCUSSION Electrodeposition of ZnO has been extensively studied using electrochemical generation of hydroxide ions in aqueous media.23−33 For example, reduction of nitrate to nitrite can be used to produce OH− (eq 1), which decreases the solubility of Zn2+ on the working electrode surface, triggering the precipitation of Zn2+ as zinc hydroxide or zinc oxide (eq 2). B

DOI: 10.1021/acs.langmuir.6b01902 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir When the deposition temperature is above 50 °C, simultaneous dehydration occurs and crystalline ZnO films can be deposited.34 NO3− + H 2O + 2e− → NO2− + 2OH−

(1)

Zn 2 + + 2OH− → Zn(OH)2 → ZnO

(2)

When aqueous plating solutions are used, ZnO deposits are typically composed of well faceted crystals such as hexagonal rods or hexagonal plates.19,30,35−40 The growth of these wellfaceted crystals, however, is not advantageous for the formation of a uniform conformal ZnO layer when ZnO needs to be deposited on already existing photon absorber layers.41 The use of DMSO as a plating medium to produce ZnO films can effectively suppress the development of crystal facets and result in the formation of ZnO films composed of featureless polycrystalline domains that are densely packed. This phenomenon was first reported by Gal et al. in their study where ZnO was deposited in DMSO using O2 reduction to generate OH− (eq 3).42 O2 + 2H 2O + 2e− → H 2O2 + 2OH− −

O2 + 2H 2O + 4e → 4OH

−

or (3)

However, Gal et al. used RF sputtering to deposit AZO when AZO needed to be deposited on the electrodeposited ZnO layer to assemble a Cu(In,Ga)Se2-based solar cell. Electrodeposition of AZO in nonaqueous media has not been investigated prior to this study. We also experienced that the formation of cracks during ZnO deposition in nonaqueous media is a common problem, and we present a simple solution to overcome this issue in this study. We have explored the use of the nitrate reduction reaction shown in eq 1 to deposit both ZnO and AZO in DMSO. While water is necessary for nitrate reduction, as-purchased DMSO typically contains a trace amount of water and thus ZnO can be deposited via nitrate reduction. However, we found that the amount of water in DMSO is not sufficient to generate stable cathodic current (Supporting Information, Figure S1), which results in the formation of cracks during the deposition of ZnO (Figure 1a). We suspect that the deposition of ZnO using O2 reduction in DMSO may suffer from the same issue because the generation of OH− from O2 reduction also requires a supply of water (eq 3). We discovered that adding a small amount of water to the DMSO solution ensures that the nitrate reduction occurs at a constant rate (Figure S1), which in turn ensures the formation of uniform and crack free ZnO films. Figure 1b shows a ZnO film deposited using the same condition used to prepare the ZnO film shown in Figure 1a (−1.0 V vs Ag/AgCl at 85 °C) except that 1 mL of water was added to the plating solution (40 mL of a DMSO solution containing 50 mM of Zn(NO3)2). The dramatic effect of adding water to produce crackless films is clearly evident from these images. The amount of H2O added to remove the cracks needs to be carefully optimized because adding more water than necessary will trigger the formation of faceted crystals, as in the case of deposition in aqueous solutions. The effect of water on the faceted growth of ZnO crystals is demonstrated in Figure 1c−f, where high magnification SEM images of ZnO films deposited with gradually increasing water content are shown. The formation of ZnO crystals with a more pronounced hexagonal rod shape, as the water content increases, suggests that only

Figure 1. SEM images of ZnO films deposited (a) without and (b) with adding 1 mL of water to 40 mL of a DMSO plating solution. High magnification SEM images show the effect of water on the ZnO deposition; with (c) 1 mL, (d) 2 mL, (e) 4 mL, and (f) 8 mL of water. (g) Side-view and (h) cross-sectional view SEM images of ZnO films deposited with adding 1 mL of water.

(0001) and (01−10) planes can be stabilized in the presence of water. The side view and cross sectional view SEM images of ZnO films deposited with 1 mL of water are shown in Figure 1g and h, which reveal the uniform, pinhole free nature of the film. The thickness of the ZnO layer appears to be the same throughout the film and is estimated to be ∼300 nm. AZO was prepared using the same deposition conditions except for the fact that the DMSO solution also contained Al(NO3)3 as the Al source. To find an optimal Al doping content, the Zn:Al molar ratio in the plating solution was varied from 200:1 to 50:1 by increasing the Al content while keeping the concentration of Zn constant. The amounts of Zn and Al present in the plating solutions and in the resulting asdeposited AZO films are summarized in Table S1. A ZnO film containing 1 atomic % of Al in the Zn site showed the best performance and will be discussed primarily in the rest of this study. (The AZO film hereafter refers to the ZnO film containing 1% Al unless otherwise noted.) X-ray diffraction patterns confirmed that both as-deposited ZnO and AZO films are crystalline (Figure 2a). In addition, both ZnO and AZO films exhibit a strong preferential orientation along the [002] direction, which makes (002) reflection dominant in the XRD pattern. The only other hkl peak observed was the (103) peak, which is reasonable as the normal directions of the (103) and (002) planes are similar. This result suggests that all the polycrystalline domains in the C

DOI: 10.1021/acs.langmuir.6b01902 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 2. (a) XRD patterns and (b) comparison of the (002) peak of ZnO (black) and AZO with 1% Al-doping (red). The (002) peak in (b) appears as a doublet because the (002) reflections by Cu Kα1 (left peak) and Kα2 (peak shown as a shoulder on the right side) radiations are resolved. The peaks originated from ITO are denoted by asterisks.

Figure 3. (a) Zn 2p XPS of ZnO film (black) and AZO film containing 1% Al (red); (b,c) Mott−Schottky plots of (b) ZnO film and (c) AZO film containing 1.0% Al measured in a 0.1 M phosphate buffer solution (pH 11) at 1 kHz (filled circle) and at 0.5 kHz (open circle).

ZnO and AZO films are uniformly oriented with the c-axis perpendicular to the substrate. Substitutional replacement of Zn2+ ions with Al3+ ions in the AZO film could also be confirmed by the XRD study. The (002) peak of the AZO film was shifted to a higher two theta value compared to that of the ZnO film (Figure 2b). This indicates that the Al3+ ion (radius = 0.53 Å for coordination number = 4), which is smaller than the Zn2+ ion (radius = 0.74 Å for coordination number = 4),43 is partially replacing the Zn2+ ion in the ZnO lattice. If Al3+ ions were stabilized in interstitial sites in the ZnO lattice or were present as impurity phases, such as Al(OH)3 or Al2O3, between the ZnO grains this peak position shift would not be observed. X-ray photoelectron spectroscopy (XPS) was used to compare the Zn 2p peaks of the ZnO and AZO films to examine the effect of Al doping on the oxidation state of Zn2+ in the ZnO lattice. The results show that the Zn 2p peaks of AZO are shifted to lower binding energy values compared to those of pure ZnO (Figure 3a). For example, the positions of the Zn 2p3/2 peaks of ZnO and AZO are 1022.4 and 1022.2 eV, respectively. This result agrees well with previous XPS study of ZnO (1022.4 eV) and 2% Al-doped ZnO (1021.9 eV).44 This suggests that in order to compensate for the charge imbalance caused by replacing Zn2+ with Al3+, the average oxidation state of Zn2+ in the ZnO lattice is reduced. This data not only confirms that Al3+ ions substitute Zn2+ ions in the ZnO lattice but also suggests that Al doping increases the majority carrier density in ZnO. The Al 2p peak of AZO and the O 1s peaks of ZnO and AZO can be found in Figure S2. The carrier densities of ZnO and AZO films were compared using Mott−Schottky plots obtained in a 0.1 M phosphate

buffer solution (pH 11) (Figure 3b). The slope for the AZO was approximately one tenth of the slope for the ZnO, suggesting that 1% Al doping increases the carrier density by a factor of 10. Although the slopes of the Mott−Schottky plots show a slight frequency dependence, we calculated the majority carrier densities of ZnO and 1% AZO using the slopes obtained at 1 kHz, which were 2.7 × 1019 and 3.0 × 1020 cm−3, respectively. From the Mott−Schottky plots, the flatband potentials of ZnO and AZO were estimated to be −0.36 and −0.39 V vs NHE at pH 11, respectively. The fact that the flatband potential of AZO is more negative than that of ZnO by ca. 30 mV agrees well with the observed increase in carrier density. The Mott−Schottky plots of AZO films containing higher Al contents can be found in Figure S3. These films do not show any further increase in carrier density, suggesting that although these films contain more Al ions, Al ions in these films are not incorporated into the ZnO lattice. The Fermi level energies (EF) of ZnO and AZO can also be calculated from their flatband potential values using the equation shown in eq 4. The pH of the solutions used for the Mott−Schottky measurements was 11, and the point of zero zeta potential (PZZP) of ZnO is reported to be 8.8.20 The calculated EF values of ZnO and AZO are −4.27 and −4.24 eV vs vacuum, respectively (calculation details shown in the Experimental Section). E F (in eV vs vacuum) = −Efb (in V vs NHE) + VH (in V) − 4.5 D

(4)

DOI: 10.1021/acs.langmuir.6b01902 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir The transparencies of these films are critical when they are used as window or protection layers and the light must pass through the ZnO layer before reaching the main light absorber. The UV−vis spectra show that the transmittances for both ZnO and AZO films prepared in this study are greater than 90% in the range of visible light (400−800 nm) (Figure 4a).

deposited ZnO layer without worrying about annealing processes affecting the properties of ZnO or the properties of the ZnO/Cu2O interface. Additionally, there are many previous studies on all-electrodeposited ZnO/Cu2O solar cells, and so the performance of the ZnO and AZO layers prepared in this study can be easily evaluated by comparison with previously reported n-ZnO/p-Cu2O cells.4,36,39,40,54−61 To build n-ZnO/p-Cu2O and n-AZO/p-Cu2O heterojuections, a 1.3−1.4 μm thick p-type Cu2O layer was electrodeposited on a 300 nm thick ZnO or AZO film using a previously reported procedure (Figure 5a).22,62 Detailed

Figure 4. (a) Transmittance and (b) absorbance of ZnO (black) and AZO (red) films measured in the UV−vis region. The inset in (a) shows photographs of ZnO and AZO films deposited on ITO, while the inset in (b) shows Tauc plots. Seal of the University of Wisconsin used with permission from the University of Wisconsin.

Figure 5. (a) Cross-sectional view and (b) top view SEM images of the ITO/ZnO/Cu2O/Pd solar cell. (c) Current density−voltage (J− V) curves of ITO/ZnO/Cu2O/Pd (black) and ITO/AZO/Cu2O/Pd (red) cells under 1 sun illumination with an AM 1.5G filter.

Achieving this level of transmittance typically requires the use of vacuum-based deposition techniques at an elevated temperature (200−380 °C), such as magnetron sputtering.10,45−47 The absorbance and corresponding Tauc plots of these films are shown in Figure 4b. The band gap energies of both ZnO and AZO were estimated to be ca. 3.36 eV. The band gap of our ZnO is wider than that of typical undoped ZnO, which is in the range of 3.28−3.29 eV.48,49 Our electrodeposition condition appears to produce ZnO that is intrinsically heavily doped (e.g., presence of oxygen deficiencies) even wtihout Al doping when the carrier density of our ZnO (2.7 × 1019 cm−3) is compared with those of regular undoped ZnO films (10 17 −10 18 cm−3).48−50 Therefore, the observed band gap widening may be caused by the Burstein−Moss effect.48−52 The fact that AZO has the same band gap as ZnO suggests that the carrier density of AZO (3.0 × 1020 cm−3) is above the Mott critical value and the Burstein−Moss effect is canceled out by the band gap narrowing due to many body effects.48−51,53 To evaluate the performance of electrodeposited ZnO and AZO films as electron transport layers in solar cell devices, heterojunction n-ZnO/p-Cu2O and n-AZO/p-Cu2O solar cells were constructed. We chose to couple ZnO and AZO films with a p-Cu2O layer because a high quality crystalline p-Cu2O layer can be deposited easily using electrodeposition methods. Since additional annealing is not necessary to form a crystalline p-Cu2O layer, the investigation of n-ZnO/p-Cu2O and n-AZO/ p-Cu2O solar cells allows us to study the properties of the as-

characterization of electrodeposited p-Cu2O layers can be found in these previously reported studies.22,62 A Pd contact was then deposited on top of the Cu2O layer by sputter coating. The top view SEM image of an electrodeposited Cu2O layer coated with Pd is shown in Figure 5b. Since the Pd contact layer is very thin, the top-view SEM image essentially shows the surface morphology of the underlying Cu2O layer. The XRD patterns of Cu2O deposited on ITO and on ZnO are shown in Figure S4. The Cu2O layer is crystalline as deposited, and it shows a preferential orientation along the [111] direction. Among all electrodeposited n-ZnO/p-Cu2O solar cells, one of the best performances was achieved by a cell composed of ZnO nanorods covered with a Cu2O layer, which provided a large n-ZnO/p-Cu2O junction area.40 We chose to compare this n-ZnO/p-Cu2O solar cell with our n-ZnO/p-Cu2O solar cell because the high surface area ZnO nanorods obtained from electrodeposition in aqueous media provide a good contrast to the uniform planar ZnO layers we prepared using nonaqueous media, while the Cu2O layers used in both solar cells look comparable. The nanorod n-ZnO/p-Cu2O cell generated a short circuit current density (JSC) as high as 8.2 mA/cm2.40 However, like most nanostructured solar cells, it suffered from low open circuit potential (VOC) possibly due to recombination losses at the interfacial states in the nanojunction area.54,63 The VOC, fill E

DOI: 10.1021/acs.langmuir.6b01902 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir factor (ff), and η of the best nanorod n-ZnO/p-Cu2O cell were reported to be 0.29 V, 36%, and 0.88%, respectively.40 Our n-ZnO/p-Cu2O cell achieved a JSC of 5.31 ± 0.65 mA/ cm2 even though it has a planar junction structure (Figure 5c). This indicates that the ZnO films deposited from DMSO, which exhibit high crystallinity and are highly oriented, have excellent electron transport properties. As mentioned in the introduction section, ZnO films oriented with the c-axis perpendicular to the substrate are expected to possess greater conductivity than ZnO films with other orientations because the electron mobility is the highest along the c-axis.16,17 The high transparency of our ZnO films (T > 90%) should be an additional factor that increases the JSC, as it can enable more visible photons to reach the Cu2O absorber layer. What is even more exciting is that our n-ZnO/p-Cu2O cell generated a VOC of 0.378 ± 0.012 V, which is significantly higher than that achieved by the best nanorod n-ZnO/p-Cu2O cell. This suggests that the dense and uniform electrodeposited ZnO film created a high quality n-ZnO/p-Cu2O junction, eliminating leakage (or shunt) pathways and minimizing recombination losses at the interface. With a ff of 33.7 ± 2.7%, our n-ZnO/p-Cu2O cell achieved an average η of 0.70 ± 0.08%. The J−V plots of n-ZnO/p-Cu2O cells prepared with ZnO layers shown in Figure 1c−f, which were electrodeposited by adding more than 1 mL of water, are presented in Figure S5. These results also show that when ZnO layers containing more faceted crystals are used, the JSC increases most likely due to an increase in the junction area between the n-ZnO and p-Cu2O layers but the VOC decreases, lowering η. When the ZnO layer was replaced by an AZO layer, the resulting n-AZO/p-Cu2O cell showed substantially enhanced VOC and JSC, 0.406 ± 0.008 V and 6.33 ± 0.71 mA/cm2, respectively. As a result, an average η of 0.92 ± 0.09% was achieved with a ff of 35.7 ± 2.1%. This 20% increase in the JSC clearly shows that Al doping can effectively improve the electrical conductivity of ZnO. The VOC increased by 0.03 V because Al doping raises the EF of ZnO by ∼0.03 V. Since the Fermi level of electrodeposited p-Cu2O is known (−4.66 eV vs vacuum),22 the maximum achievable VOC can be calculated using the Fermi levels of the ZnO and AZO materials produced in this study. The calculated VOC values for the n-ZnO/p-Cu2O and n-AZO/p-Cu 2 O devices are 0.390 and 0.423 V, respectively. The measured VOC values, 0.378 V for n-ZnO/ p-Cu2O and 0.406 V for n-AZO/p-Cu2O, are very close to these expected values. This proves that high quality n-ZnO/pCu2O and n-AZO/p-Cu2O junctions are created by electrodeposited ZnO and AZO films from DMSO. The average and best performances obtained from the n-ZnO/p-Cu2O and nAZO/p-Cu2O cells are summarized in Table S2. Incident photon-to-current efficiencies (IPCEs) of the n-ZnO/p-Cu2O and AZO/Cu2O cells measured under the short circuit condition can be found in Figure S6. The n-ZnO/p-Cu 2 O and n-AZO/p-Cu2 O solar cells discussed above were constructed by depositing a Cu2O layer on a ZnO or AZO layer. Therefore, our ability to conformally deposit a uniform ZnO layer has not been fully utilized for the construction of n-ZnO/p-Cu2O and n-AZO/p-Cu2O solar cells. However, we believe that the synthesis conditions reported in this study will be found to be even more useful when ZnO needs to be deposited as a window layer or a protection layer on top of an absorber layer.

4. CONCLUSIONS In summary, we developed optimum conditions to prepare crack-free ZnO and AZO films using electrodeposition in nonaqueous media at 85 °C and ambient pressure. In particular, the effect of adding water to nonaqueous media to prevent crack formation during deposition was identified for the first time. The resulting films were highly transparent (T > 90%) and crystalline with a uniform orientation (i.e., the c-axis perpendicular to the substrate), which enhanced the electron transport of the films. Electrochemical Al doping of ZnO using nonaqueous media, which was investigated in this study for the first time, effectively increased the carrier density and raised the Fermi level of ZnO. As a result when an n-AZO/p-Cu2O cell was constructed, VOC = 0.406 ± 0.008 V, JSC = 6.33 ± 0.71 mA/cm2, and ff =35.7 ± 2.1% were achieved, resulting in an average η of 0.92 ± 0.09%. This is one of the best performances reported for an all electrodeposited ZnO/Cu2O-based cell, although our n-AZO/p-Cu2O cell was based on a planar bilayer junction. We believe that the electrodeposition condition reported in this study will offer a facile, practical, and diverse way to conformally deposit pinhole-free ZnO or AZO layers on various semiconductor absorber films to serve as buffer or protection layers.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.6b01902. Table showing the Zn:Al ratios in solutions and films, a table listing the performances of the n-ZnO/p-Cu2O and n-AZO/p-Cu2O solar cells, deposition current−time profiles, Al 2p XPS of AZO and O 1s XPS of ZnO and AZO, Mott−Schottky plots of ZnO films containing varying amounts of Al, XRD of an electrodeposited pCu2O film and an ITO/ZnO/Cu2O/Pd solar cell, J−V curves of ITO/ZnO/Cu2O/Pd cells using various ZnO layers, IPCE data of ITO/ZnO/Cu2O/Pd and ITO/ AZO/Cu2O/Pd cells (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences of the U.S. Department of Energy through Grant DESC0008707.

■

REFERENCES

(1) Jackson, P.; Hariskos, D.; Lotter, E.; Paetel, S.; Wuerz, R.; Menner, R.; Wischmann, W.; Powalla, M. New World Record Efficiency for Cu(In,Ga)Se2 Thin-Film Solar Cells beyond 20%. Prog. Photovoltaics 2011, 19, 894−897. (2) Lee, Y. S.; Chua, D.; Brandt, R. E.; Siah, S. C.; Li, J. V.; Mailoa, J. P.; Lee, S. W.; Gordon, R. G.; Buonassisi, T. Atomic Layer Deposited Gallium Oxide Buffer Layer Enables 1.2 V Open-Circuit Voltage in Cuprous Oxide Solar Cells. Adv. Mater. 2014, 26, 4704−4710.

F

DOI: 10.1021/acs.langmuir.6b01902 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir (3) Park, S.; Tark, S. J.; Lee, J. S.; Lim, H.; Kim, D. Effects of Intrinsic ZnO Buffer Layer Based on P3HT/PCBM Organic Solar Cells with Al-Doped ZnO Electrode. Sol. Energy Mater. Sol. Cells 2009, 93, 1020− 1023. (4) Hsu, Y.-K.; Lin, H.-H.; Wu, J.-R.; Chen, M.-H.; Chen, Y.-C.; Lin, Y.-G. Electrochemical Growth and Characterization of a p-Cu2O Thin Film on n-ZnO Nanorods for Solar Cell Application. RSC Adv. 2014, 4, 7655−7659. (5) Minami, T.; Nishi, Y.; Miyata, T.; Nomoto, J.-i. High-Efficiency Oxide Solar Cells with ZnO/Cu2O Heterojunction Fabricated on Thermally Oxidized Cu2O Sheets. Appl. Phys. Express 2011, 4, 062301. (6) Paracchino, A.; Laporte, V.; Sivula, K.; Grätzel, M.; Thimsen, E. Highly Active Oxide Photocathode for Photoelectrochemical Water Reduction. Nat. Mater. 2011, 10, 456−461. (7) Morales-Guio, C. G.; Tilley, S. D.; Vrubel, H.; Grätzel, M.; Hu, X. Hydrogen Evolution from a Copper(I) Oxide Photocathode Coated with an Amorphous Molybdenum Sulphide Catalyst. Nat. Commun. 2014, 5, 3059. (8) Morales-Guio, C. G.; Liardet, L.; Mayer, M. T.; Tilley, S. D.; Grätzel, M.; Hu, X. Photoelectrochemical Hydrogen Production in Alkaline Solutions Using Cu2O Coated with Earth-Abundant Hydrogen Evolution Catalysts. Angew. Chem. 2015, 54, 674−677. (9) Tilley, S. D.; Schreier, M.; Azevedo, J.; Stefik, M.; Graetzel, M. Ruthenium Oxide Hydrogen Evolution Catalysis on Composite Cuprous Oxide Water-Splitting Photocathodes. Adv. Funct. Mater. 2014, 24, 303−311. (10) Martinez, M. A.; Herrero, J.; Gutiérrez, M. T. Deposition of Transparent and Conductive Al-Doped ZnO Thin Films for Photovoltaic Solar Cells. Sol. Energy Mater. Sol. Cells 1997, 45, 75−86. (11) Janotti, A.; Van de Walle, C. G. Fundamentals of Zinc Oxide as a Semiconductor. Rep. Prog. Phys. 2009, 72, 126501. (12) Min, Y.-S.; An, C.-J.; Kim, S.-K.; Song, J.-W.; Hwang, C.-S. Growth and Characterization of Conducting ZnO Thin Films by Atomic Layer Deposition. Bull. Korean Chem. Soc. 2010, 31, 2503− 2508. (13) Wang, Y.; Li, X.; Jiang, G.; Liu, W.; Zhu, C. Origin of (103) Plane of ZnO Films Deposited by RF Magnetron Sputtering. J. Mater. Sci.: Mater. Electron. 2013, 24, 3764−3767. (14) Chen, M.; Pei, Z. L.; Sun, C.; Wen, L. S.; Wang, X. Surface Characterization of Transparent Conductive Oxide Al-Doped ZnO Films. J. Cryst. Growth 2000, 220, 254−262. (15) Hu, J.; Gordon, R. G. Textured Aluminum-Doped Zinc Oxide Thin Films from Atmospheric Pressure Chemical-Vapor Deposition. J. Appl. Phys. 1992, 71, 880−890. (16) Kaga, H.; Kinemuchi, Y.; Yilmaz, H.; Watari, K.; Nakano, H.; Nakano, H.; Tanaka, S.; Makiya, A.; Kato, Z.; Uematsu, K. Orientation Dependence of Transport Property and Microstructural Characterization of Al-Doped ZnO Ceramics. Acta Mater. 2007, 55, 4753−4757. (17) Kaga, H.; Kinemuchi, Y.; Tanaka, S.; Makiya, A.; Kato, Z.; Uematsu, K.; Watari, K. Preparation and Thermoelectric Property of Highly Oriented Al-Doped ZnO Ceramics by a High Magnetic Field. Jpn. J. Appl. Phys. 2006, 45, L1212−L1214. (18) Gelderman, K.; Lee, L.; Donne, S. W. Flat-Band Potential of a Semiconductor: Using the Mott-Schottky Equation. J. Chem. Educ. 2007, 84, 685−688. (19) Aragonès, A. C.; Palacios-Padrós, A.; Caballero-Briones, F.; Sanz, F. Study and Improvement of Aluminium Doped ZnO Thin Films: Limits and Advantages. Electrochim. Acta 2013, 109, 117−124. (20) Compton, R. G. Electrode Kinetics: Reactions; Elsevier Science Publishers B.V.: Amsterdam, 1987; Vol. 27. (21) Nozik, A. J. Photoelectrochemistry: Applications to Solar Energy Conversion. Annu. Rev. Phys. Chem. 1978, 29, 189−222. (22) McShane, C. M.; Choi, K.-S. Junction Studies on Electrochemically Fabricated p−n Cu2O Homojunction Solar Cells for Efficiency Enhancement. Phys. Chem. Chem. Phys. 2012, 14, 6112. (23) Peulon, S.; Lincot, D. Cathodic Electrodeposition from Aqueous Solution of Dense or Open-Structured Zinc Oxide Films. Adv. Mater. 1996, 8, 166−170.

(24) Lu, X.-H.; Wang, D.; Li, G.-R.; Su, C.-Y.; Kuang, D.-B.; Tong, Y.-X. Controllable Electrochemical Synthesis of Hierarchical ZnO Nanostructures on FTO Glass. J. Phys. Chem. C 2009, 113, 13574− 13582. (25) Elias, J.; Tena-Zaera, R.; Lévy-Clément, C. Effect of the Chemical Nature of the Anions on the Electrodeposition of ZnO Nanowire Arrays. J. Phys. Chem. C 2008, 112, 5736−5741. (26) Liu, R.; Vertegel, A. A.; Bohannan, E. W.; Sorenson, T. A.; Switzer, J. A. Epitaxial Electrodeposition of Zinc Oxide Nanopillars on Single-Crystal Gold. Chem. Mater. 2001, 13, 508−512. (27) Limmer, S. J.; Kulp, E. A.; Switzer, J. A. Epitaxial Electrodeposition of ZnO on Au(111) from Alkaline Solution: Exploiting Amphoterism in Zn(II). Langmuir 2006, 22, 10535−10539. (28) Pauporté, T.; Lincot, D. Heteroepitaxial Electrodeposition of Zinc Oxide Films on Gallium Nitride. Appl. Phys. Lett. 1999, 75, 3817. (29) Pauporté, T.; Lincot, D. Hydrogen Peroxide Oxygen Precursor for Zinc Oxide Electrodeposition I. Deposition in Perchlorate Medium. J. Electrochem. Soc. 2001, 148, C310−C314. (30) Yoshida, T.; Komatsu, D.; Shimokawa, N.; Minoura, H. Mechanism of Cathodic Electrodeposition of Zinc Oxide Thin Films from Aqueous Zinc Nitrate Baths. Thin Solid Films 2004, 451−452, 166−169. (31) Xu, L.; Guo, Y.; Liao, Q.; Zhang, J.; Xu, D. Morphological Control of ZnO Nanostructures by Electrodeposition. J. Phys. Chem. B 2005, 109, 13519−13522. (32) Goux, A.; Pauporté, T.; Chivot, J.; Lincot, D. Temperature Effects on ZnO Electrodeposition. Electrochim. Acta 2005, 50, 2239− 2248. (33) Izaki, M.; Omi, T. Transparent Zinc Oxide Films Prepared by Electrochemical Reaction. Appl. Phys. Lett. 1996, 68, 2439. (34) Peulon, S.; Lincot, D. Mechanistic Study of Cathodic Electrodeposition of Zinc Oxide and Zinc Hydroxychloride Films from Oxygenated Aqueous Zinc Chloride Solutions. J. Electrochem. Soc. 1998, 145, 864−874. (35) Elias, J.; Tena-Zaera, R.; Lévy-Clément, C. Electrodeposition of ZnO Nanowires with Controlled Dimensions for Photovoltaic Applications: Role of Buffer Layer. Thin Solid Films 2007, 515, 8553−8557. (36) Zoolfakar, A. S.; Rani, R. A.; Morfa, A. J.; Balendhran, S.; O’Mullane, A. P.; Zhuiykov, S.; Kalantar-zadeh, K. Enhancing the Current Density of Electrodeposited ZnO−Cu2O Solar Cells by Engineering Their Heterointerfaces. J. Mater. Chem. 2012, 22, 21767− 21769. (37) Cheng, W.-Y.; Lin, Y.-F.; Lu, S.-Y. Nanowires Improved Charge Separation and Light Utilization in Metal-Oxide Solar Cells. Appl. Phys. Lett. 2011, 99, 063107. (38) Lévy-Clément, C.; Elias, J. Optimization of the Design of Extremely Thin Absorber Solar Cells Based on Electrodeposited ZnO Nanowires. ChemPhysChem 2013, 14, 2321−2330. (39) Hsu, Y.-K.; Lin, H.-H.; Chen, M.-H.; Chen, Y.-C.; Lin, Y.-G. Polarity-Dependant Performance of p-Cu2O/n-ZnO Heterojunction Solar Cells. Electrochim. Acta 2014, 144, 295−299. (40) Cui, J.; Gibson, U. J. A Simple Two-Step Electrodeposition of Cu2O/ZnO Nanopillar Solar Cells. J. Phys. Chem. C 2010, 114, 6408− 6412. (41) Lee, Y. S.; Heo, J.; Siah, S. C.; Mailoa, J. P.; Brandt, R. E.; Kim, S. B.; Gordon, R. G.; Buonassisi, T. Ultrathin Amorphous Zinc-TinOxide Buffer Layer for Enhancing Heterojunction Interface Quality in Metal-Oxide Solar Cells. Energy Environ. Sci. 2013, 6, 2112−2118. (42) Gal, D.; Hodes, G.; Lincot, D.; Schock, H. W. Electrochemical Deposition of Zinc Oxide Films from Non-Aqueous Solution: a new Buffer/Window Process for Thin Film Solar Cells. Thin Solid Films 2000, 361−362, 79−83. (43) Shannon, R. D. Revised Effective Ionic Radii and Systematic Studies of Interatomic Distances in Halides and Chalcogenides. Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr. 1976, 32, 751−767. G

DOI: 10.1021/acs.langmuir.6b01902 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

(63) Yuhas, B. D.; Yang, P. Nanowire-Based All-Oxide Solar Cells. J. Am. Chem. Soc. 2009, 131, 3756−3761.

(44) Islam, M. N.; Ghosh, T. B.; Chopra, K. L.; Acharya, H. N. XPS and X-Ray Diffraction Studies of Aluminum-Doped Zinc Oxide Transparent Conducting Films. Thin Solid Films 1996, 280, 20−25. (45) Yoo, J.; Lee, J.; Kim, S.; Yoon, K.; Park, I. J.; Dhungel, S. K.; Karunagaran, B.; Mangalaraj, D.; Yi, J. High Transmittance and Low Resistive ZnO:Al Films for Thin Film Solar Cells. Thin Solid Films 2005, 480−481, 213−217. (46) Tanuševski, A.; Georgieva, V. Optical and Electrical Properties of Nanocrystal Zinc Oxide Films Prepared by DC Magnetron Sputtering at Different Sputtering Pressures. Appl. Surf. Sci. 2010, 256, 5056−5060. (47) Jeong, W. J.; Kim, S. K.; Park, G. C. Preparation and Characteristic of ZnO Thin Film with High and Low Resistivity for an Application of Solar Cell. Thin Solid Films 2006, 506−507, 180−183. (48) Jain, A.; Sagar, P.; Mehra, R. M. Band Gap Widening and Narrowing in Moderately and Heavily Doped n-ZnO Films. Solid-State Electron. 2006, 50, 1420−1424. (49) Lu, J. G.; Fujita, S.; Kawaharamura, T.; Nishinaka, H.; Kamada, Y.; Ohshima, T.; Ye, Z. Z.; Zeng, Y. J.; Zhang, Y. Z.; Zhu, L. P.; He, H. P.; Zhao, B. H. Carrier Concentration Dependence of Band Gap Shift in n-type ZnO:Al Films. J. Appl. Phys. 2007, 101, 083705−8. (50) Saw, K. G.; Aznan, N. M.; Yam, F. K.; Ng, S. S.; Pung, S. Y. New Insights on the Burstein-Moss Shift and Band Gap Narrowing in Indium-Doped Zinc Oxide Thin Films. PLoS One 2015, 10, e0141180−17. (51) Roth, A. P.; Webb, J. B.; Williams, D. F. Absorption Edge Shift in ZnO Thin Films at High Carrier Densities. Solid State Commun. 1981, 39, 1269−1271. (52) Kim, K. J.; Park, Y. R. Large and Abrupt Optical Band Gap Variation in In-Doped ZnO. Appl. Phys. Lett. 2001, 78, 475. (53) Roth, A. P.; Webb, J. B.; Williams, D. F. Band-Gap Narrowing in Heavily Defect-Doped ZnO. Phys. Rev. B: Condens. Matter Mater. Phys. 1982, 25, 7836−7839. (54) Musselman, K. P.; Marin, A.; Schmidt-Mende, L.; MacManusDriscoll, J. L. Incompatible Length Scales in Nanostructured Cu2O Solar Cells. Adv. Funct. Mater. 2012, 22, 2202−2208. (55) Marin, A. T.; Munoz- Rojas, D.; Iza, D. C.; Gershon, T.; Musselman, K. P.; MacManus-Driscoll, J. L. Novel Atmospheric Growth Technique to Improve Both Light Absorption and Charge Collection in ZnO/Cu2O Thin Film Solar Cells. Adv. Funct. Mater. 2013, 23, 3413−3419. (56) Musselman, K. P.; Marin, A.; Wisnet, A.; Scheu, C.; MacManus Driscoll, J. L.; Schmidt - Mende, L. A Novel Buffering Technique for Aqueous Processing of Zinc Oxide Nanostructures and Interfaces, and Corresponding Improvement of Electrodeposited ZnO-Cu2O Photovoltaics. Adv. Funct. Mater. 2011, 21, 573−582. (57) Shinagawa, T.; Shibata, K.; Shimomura, O.; Chigane, M.; Nomura, R.; Izaki, M. Solution-Processed High-Haze ZnO Pyramidal Textures Directly Grown on a TCO Substrate and the Light-Trapping Effect in Cu2O Solar Cells. J. Mater. Chem. C 2014, 2, 2908−2917. (58) Musselman, K. P.; Wisnet, A.; Iza, D. C.; Hesse, H. C.; Scheu, C.; MacManus-Driscoll, J. L.; Schmidt-Mende, L. Strong Efficiency Improvements in Ultra-Low-Cost Inorganic Nanowire Solar Cells. Adv. Mater. 2010, 22, E254−E258. (59) Izaki, M.; Shinagawa, T.; Mizuno, K.-T.; Ida, Y.; Inaba, M.; Tasaka, A. Electrochemically Constructed p-Cu2O/n-ZnO Heterojunction Diode for Photovoltaic Device. J. Phys. D: Appl. Phys. 2007, 40, 3326−3329. (60) Izaki, M.; Ohta, T.; Kondo, M.; Takahashi, T.; Mohamad, F. B.; Zamzuri, M.; Sasano, J.; Shinagawa, T.; Pauporté, T. Electrodeposited ZnONanowire/Cu2O Photovoltaic Device with Highly Resistive ZnO Intermediate Layer. ACS Appl. Mater. Interfaces 2014, 6, 13461− 13469. (61) Fujimoto, K.; Oku, T.; Akiyama, T. Fabrication and Characterization of ZnO/Cu2O Solar Cells Prepared by Electrodeposition. Appl. Phys. Express 2013, 6, 086503. (62) McShane, C. M.; Siripala, W. P.; Choi, K.-S. Effect of Junction Morphology on the Performance of Polycrystalline Cu2O Homojunction Solar Cells. J. Phys. Chem. Lett. 2010, 1, 2666−2670. H

DOI: 10.1021/acs.langmuir.6b01902 Langmuir XXXX, XXX, XXX−XXX