Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2019 Dec 24;5(1):683–691. doi: 10.1021/acsomega.9b03308

Light-Irradiated Electrochemical Direct Construction of Cu2O/CuO Bilayers by Switching Cathodic/Anodic Polarization in Copper(II)–Tartrate Complex Aqueous Solution

Masanobu Izaki †,*, Takayuki Koyama †, Pei Loon Khoo †, Tsutomu Shinagawa †,‡
PMCID: PMC6964298  PMID: 31956818

Abstract

graphic file with name ao9b03308_0009.jpg

p-CuO with a band gap energy of 1.5 eV, p-Cu2O with a band gap energy of 2.05 eV, and their bilayers were prepared by controlling the potential of anodic and cathodic polarization in a copper(II)–tartrate complex aqueous solution containing copper(II) sulfate hydrate and tartaric acid in the dark and under light irradiation. Electrochemical characteristics of the electrodeposition and the resultant CuO and Cu2O layers were investigated with cyclic voltammetry, chronoamperometry, and Mott–Schottky plots, and the structural and optical characterizations were performed with X-ray diffraction, scanning electron microscopy, and optical absorption spectra measurements. The CuO layer prepared at 0.4–0.7 V was composed of aggregates of granular grains with the monoclinic lattice, and the Cu2O layer composed of coarse grains with the cubic lattice was deposited at −0.4 to 0.6 V. The flat-band potentials were estimated to be 0.145 and −0.1 V (vs Ag/AgCl) for the CuO and Cu2O layers, respectively. The 0.4 μm CuO/1.1 μm Cu2O bilayers could be prepared by switching the electrodeposition potentials of 0.4 and −0.4 V, irrespective of the presence of light irradiation. The photoelectrodeposition under light irradiation enabled the preparation of continuous and dense 1.1 μm Cu2O/0.4 μm CuO bilayer by controlling the potential, while electrodeposition in the dark led to sparse, isolated, and coarse Cu2O grains being deposited. The mechanism for the photoelectrodeposition of the bilayers was discussed based on the energy band alignment at the heterointerface to the Cu–tartrate complex solution.

Introduction

Cupric oxide (CuO) and cuprous oxide (Cu2O) are p-type semiconductors with band gap energies of 1.51 and 2.0 eV,2 and they both have been employed as photocathodes in photoelectrochemical water-splitting system for generating hydrogen gas3,4 and photovoltaic layer in oxide solar cells.5,6 The photoactive performance of the photocathode and photovoltaic layers closely relates to their band gap energy,7 and combining two or more p-type semiconductors with different band gap energies is an excellent strategy to enhance the performance,8 as demonstrated by the 38% efficiency InGaP/GaAs multijunction solar cell.9 A CuO/Cu2O bilayer satisfying such a strategy was reported to act as a photocathode with superior performance compared to that of single CuO and Cu2O layers.10,11

The CuO and Cu2O layers have been prepared by several techniques of thermal oxidation12,13 and solution chemical process.14,15 The solution chemical process offers several advantages over thermal oxidation and gas phase processes, such as saving energy due to low-temperature preparation and scalability for mass production, as demonstrated by the preparation of the precursors for Cu(In,Ga)Se216 and Cu2SnZnS4 solar cells.17 Also, CuO/Cu2O bilayers prepared by chemical processes, followed by heating, were also reported,18,19 but nanopores and defects introduced during heating showed harmful effects for the photoactive performance.20 The possibility of both CuO and Cu2O layers being prepared by anodic and cathodic polarizations in an aqueous solution containing copper(II) sulfate and tartaric acid was reported by Poizot et al.21 However, stacking of Cu2O on CuO layers by electrodeposition was difficult because the electrons needed for the Cu2O electrodeposition were the minority carriers in the p-CuO layer.

Here, we report the electrochemical preparation of the CuO and Cu2O layers by anodic and cathodic polarization in a modified copper(II)–tartrate complex aqueous solution and the direct, continuous constructions of CuO/Cu2O and Cu2O/CuO bilayers by irradiating light at photon energies higher than the underlying layer’s band gap energies during electrodepositions. Electrochemical characterization was carried out for the preparation of CuO, Cu2O, and bilayers by cyclic voltammetry and chronoamperometry. Their structure and energy states including the band gap energies and flat-band potentials of the CuO and Cu2O layers were investigated with X-ray diffraction, scanning electron microscopy (SEM) observation, optical absorption spectra measurement, and Mott–Schottky plots measurement. Furthermore, the mechanism of the photoelectrodeposition of CuO/Cu2O bilayers was discussed based on the energy states of the CuO and Cu2O layers.

Results and Discussion

Electrochemical Preparation of CuO and Cu2O Layers by Anodic and Cathodic Polarization

The change in the appearance of the aqueous solution containing a 0.3 mol L–1 copper(II) sulfate and 0.3 mol L–1 tartaric acid with pH value is shown in Figure S1. The pH value was adjusted using a NaOH aqueous solution. The solution color changed from the precipitation-free light blue at pH 1.1 to diluted light blue, cloudy dark blue, and clear dark violet over pH 11. The white precipitation was formed in the solution from pH 1.5 and then was dissolved by raising the pH value, resulting in the formation of the precipitation-free dark violet solution at pH 13 by adding 1.5 mol L–1 NaOH; this solution was selected for the electrodeposition of CuO and Cu2O.

Figure 1 shows the cyclic voltammogram for the bare fluorine-doped tin oxide (FTO) substrate in the solution, and the cyclic voltammograms were recorded for the electrodeposition of Cu2O and CuO layers under cathodic and anodic polarizations. The red-colored Cu2O and black-colored CuO layers were deposited by the cathodic and anodic polarizations, respectively, and the electrochemical behavior closely related to the Cu–tartrate complex species and the electrochemical reactions, as demonstrated for the Cu2O electrodeposition in the Cu–lactate complex solution22 and CuO electrodeposition in the Cu–NH3–complex solution.15

Figure 1.

Figure 1

Open in a new tab

Cyclic voltammogram for the FTO substrate in aqueous solution containing copper(II) sulfate and tartaric acid, and the appearance of the CuO and Cu2O layers.

The electrolyte used in this study contains 0.3 mol L–1 copper(II) sulfate, 0.3 mol L–1l-tartaric acid (H2L = HOOCCH(OH)CH(OH)COOH), and 1.5 mol L–1 NaOH. The molar ratio of Cu2+/H2L/OH– is 1:1:5 and the pH of the resulting solution is ∼13. No precipitation occurred despite the strong alkaline solution due to the formation of Cu(II)–tartrate complex. Complex species in strongly alkaline solutions of Cu(II)–tartaric acid, also known as Fehling’s solution, have recently been investigated by Hörner et al.23 According to the literature, two complexes, [Cu2(H–2L)2]4– and [Cu(H–2L)2]6–, are proposed to be stable in the strong alkaline region above pH 10. The tartrate ligand, H–2L4– = –OOCCH(O–)CH(O–)COO–, has a structure in which four protons are lost from the two carboxyl groups and two hydroxyl groups. Considering the molar ratio of Cu2+ to H2L (1:1) in the prepared solution, the predominant complex present in the electrolyte would be [Cu2(H–2L)2]4–. The formation of this tartrate ligand requires at least 4 equiv NaOH to tartaric acid. The fact that the pH value ∼13 of the electrolyte containing 0.3 mol L–1 tartaric acid and 1.5 mol L–1 (5 equiv) NaOH is consistent with the formation of [Cu2(H–2L)2]4–. The expected structure of the [Cu2(H–2L)2]4– complex and the calculated fraction of complex species23 present depending on the solution pH are shown in Figure S2.

In the anodic polarization curve in Figure 1, the anodic current density increased from 0.27 V, and the black-colored CuO layer was deposited on the FTO substrate. The current densities were estimated to be 0.21, 0.65, 2.4, and 5.7 mA cm–2 at 0.4, 0.5, 0.6, and 0.7 V, respectively. It has already been reported that the CuO electrodeposition by the anodic polarization from the Cu–NH3 complex solution was induced by the oxygen gas generation reaction at the potential positive than that for the following reaction14

graphic file with name ao9b03308_m001.jpg

Since the potential of 0.27 V was close to 0.266 V for the oxygen gas generation reaction at pH 13, the formation of the CuO layer from the [Cu2(H–2L)2]4– complex was induced by the oxygen gas generation reaction like the Cu–NH3 complex solution.

When the potential was brought to the negative side, the cathodic current density gradually increased from −0.2 V and then rapidly increased from −0.6 V. The current densities were estimated to be −0.63, −0.65, −2.3, and −4.1 mA cm–2 at −0.4, −0.5, −0.6, and −0.7 V, respectively. The red-colored Cu2O layer was deposited from the [Cu2(H–2L)2]4– complex at a potential more negative than −0.2 V, although the equilibrium potential could not be calculated due to absence of data on the thermodynamic property of the complex. Also, the potential of −0.5 V was close to that for the following reaction at pH 13

graphic file with name ao9b03308_m002.jpg

Also, there was a possibility of the direct electrodeposition of metallic Cu from the [Cu2(H–2L)2]4– complex.

Figure 2 shows the X-ray diffraction patterns of the CuO and Cu2O layers prepared on FTO substrates by anodic polarizations at potentials from 0.4 to 0.7 V and cathodic polarizations at −0.4 to −0.7 V. Four peaks assigned as (002), (111̅), (111), and (200) planes with a characteristic monoclinic lattice24 could be observed for the CuO layer prepared at 0.4 V at 35.37, 35.59, 38.48, and 38.99°, respectively, in addition to those originating from SnO2 in the FTO substrate. The (002) peak disappeared and the (200) peak strengthened at potentials more positive than 0.5 V, indicating changes in the preferred orientation. The diffraction angles of (111̅) and (200) peaks were of almost constant values, irrespective of the potentials. The full width at half-maximum (FWHM) values of the CuO (111̅) plane were almost constant values of 0.32–0.34°, irrespective of the preparation potentials, which were smaller than 0.76° for a CuO layer prepared in a Cu–NH3 complex aqueous solution.15

Figure 2.

Figure 2

Open in a new tab

X-ray diffraction patterns for the CuO layers (A) prepared at 0.4 V (a), 0.5 V (b), 0.6 V (c), and 0.7 V (d) and the Cu2O layers (B) prepared at −0.4 V (e), −0.5 V (f), −0.6 V (g), and −0.7 V (h).

The red-colored Cu2O layers prepared at −0.4 to −0.6 V showed X-ray peaks assigned as the (110), (111), and (200) planes of the characteristic cubic lattice25 at 29.6, 36.04, and 42.23°, and the intensity ratios were almost constant, irrespective of the potentials. Additionally, two additional peaks assigned as the (111) and (200) planes of metallic Cu were observed at −0.7 V. The deposition potential of −0.7 V was more negative than −0.5 V for the reduction reaction of Cu2O to Cu, and it was possible that trace metallic Cu was incorporated in the Cu2O layer at −0.6 V. The FWHM values of the Cu2O (111) peak were almost constant values of 0.13°, irrespective of the preparation potentials.

Figure 3 shows the surface and cross-sectional structures of the CuO layers prepared at 0.4–0.7 V. The CuO layers were composed of aggregates of columnar grains with the size of approximately 0.48 μm, which were deposited over the entire surface of the FTO substrate without any defects, such as pores, on both the surface and cross-sectional structure. The grain morphology was significantly distinct from the fan-shaped grain of the CuO layer prepared on the FTO substrate in the Cu–NH3 complex aqueous solution.15 The thicknesses of CuO layers were estimated to be 0.80, 0.58, 0.39, and 0.36 μm at 0.4, 0.5, 0.6, and 0.7 V, respectively. The change in the thicknesses was consistent with the decrease in the height of the CuO(111̅) peak shown in Figure 2, indicating the change in the current efficiency by the potential.

Figure 3.

Figure 3

Open in a new tab

Field-emission scanning electron microscopy (FE-SEM) images of the surface morphologies (A) and cross-sectional structures (B) for CuO layers prepared at 0.4 V (a), 0.5 V (b), 0.6 V (c), and 0.7 V (d).

Figure 4 shows the surface and cross-sectional structures of Cu2O layers prepared at −0.4 to −0.7 V. The cubic-shaped and isolated Cu2O grains were deposited on the FTO substrate, with bare FTO substrate surface clearly observable between the Cu2O grains. The size of the cubic Cu2O grains decreased from 7.8 to 2.0 μm, and the grains’ density increased with change in potential from −0.4 to −0.6 V. At −0.7 V, 1.25 μm sized Cu2O cubic grains were deposited over the entire FTO substrate, and additional cauliflower-shaped grains (shown by the arrow in Figure 4A(d)) were speculated to be metallic Cu as detected by the X-ray diffraction. A dense Cu2O layer without any defect such as pores could be formed on the FTO substrates by cathodic polarization in a Cu-lactic acid aqueous solution.26 This shows that the solution formulation strongly affects the structure, including grain morphologies for both the CuO and Cu2O layers.

Figure 4.

Figure 4

Open in a new tab

FE-SEM images of the surface morphologies (A) and cross-sectional structures (B) for Cu2O layers prepared at −0.4 V (a), −0.5 V (b), −0.6 V (c), and −0.7 V (d).

Figure 5 shows the optical absorption spectra and the relationship between the absorption coefficient and photon energy for the CuO layers prepared at 0.4–0.7 V and Cu2O layers prepared at −0.4 to −0.7 V. The absorption coefficients were estimated from the absorbance and layer thickness, and the grain height was used alternative to the thickness for the Cu2O layers prepared at −0.7 V due to the vacancies between the Cu2O grains. The CuO layers possessed absorption edges at around 830 nm in wavelength, and the absorbance at 400 nm increased with increase in the CuO thickness. The absorption coefficient was almost constant at around 4 × 104 cm–1, irrespective of the potential. The band gap energy (Eg) can be estimated by using the following relationship between (αhν)n and photon energy (hν)

graphic file with name ao9b03308_m003.jpg

where α, h, and ν are the absorption coefficient, Plank’s constant, and frequency of light, respectively, and n is 2 or 1/2 for direct or indirect transition, respectively.27 The band gap energies were estimated to be 1.5 eV for all the CuO layers with the assumption of indirect transition, and this value agreed with the one already reported.15 The Cu2O layers possessed absorption edges at around 610 nm, irrespective of the potentials, and the band gap energies were estimated to be 2.05 eV with the assumption of direct transition. The absorbance increased by decreasing vacancies between Cu2O grains when the potential was brought to the negative side from −0.4 to −0.7 V. The Cu2O layer prepared at −0.7 V showed a high absorbance at wavelengths longer than 610 nm due to the incorporation of opaque metallic Cu.

Figure 5.

Figure 5

Open in a new tab

Optical absorption spectra and the relationship between the absorption coefficient and photon energy (insets) for CuO layers (A) prepared at 0.4 V (a), 0.5 V (b), 0.6 V (c), and 0.7 V (d) and for Cu2O layers (B) prepared at −0.4 V (e), −0.5 V (f), −0.6 V (g), and −0.7 V (h).

Figure 6 shows the flat-band potentials estimated from the Mott–Schottky plots (Figures S3 and S4) for CuO layers prepared at 0.4–0.7 V and Cu2O layers prepared at −0.4 to −0.7 V. The relationship between the space-change capacitance (C) and potential (E) in the Mott–Schottky plots is as follows

graphic file with name ao9b03308_m004.jpg

where ε, ε0, NA, Efb, K, T, and e are the permittivity of vacuum, dielectric constant of the semiconductor, carrier density, flat-band potential, Boltzmann constant, absolute temperature, and charge of electron, respectively.28 The flat-band potential is estimated by extrapolating the linear part to Inline graphic, and the carrier density corresponding to the acceptor density was estimated from the slope of the linear part. All of the CuO layers were p-type semiconductors, concluded from the negative slopes of the linear parts, irrespective of the potential. Also, the flat-band potentials (Efb) were estimated to be approximate constant values of 0.149 ± 0.003 V (vs Ag/Ag Cl) for all CuO layers, irrespective of the potential. It has been reported for CuO layers prepared by electrodeposition in an aqueous solution containing a copper(II) sulfate and glycine that the flat-band potential varied from 0.01 to 0.31 V (vs Ag/AgCl) depending on the solution pH,29 and the flat-band potential around 0.149 V was located within this variation. Also, a flat-band potential of 0.79 V (vs RHE) was reported for a CuO layer prepared by heating a Cu2O layer electrodeposited in a Cu–lactic acid aqueous solution,10 which is very close to 0.145 V (vs Ag/AgCl) obtained in this study. Also, all the Cu2O layers were identified as p-type semiconductors from the negative slopes of the linear parts for the Mott–Schottky plots, and the flat-band potentials (Efb) were estimated to be −0.106, −0.09, and −0.080 V (vs Ag/AgCl) at the deposition potentials of −0.4, −0.5, and −0.6 V, respectively. A flat-band potential of 0.55 V (vs RHE) was reported for the Cu2O layer prepared by electrodeposition in the Cu–lactic acid aqueous solution10 and was close to those obtained at −0.4 to −0.6 V. A flat-band potential of −0.009 V for the Cu2O layer prepared at −0.7 V was different compared to the values obtained at −0.4 to −0.6 V due to the incorporation of metallic Cu. The flat-band potential (Efb) closely relates to the Fermi level (EF) of the CuO and Cu2O semiconductors,30 and the location of Fermi level (EF) changed depending on the acceptor density (NA) in the p-type semiconductor, as represented by the following equation

graphic file with name ao9b03308_m006.jpg

where VBM and NV represent the valence band maximum and effective density of the state.31 The almost constant flat-band potentials reflected the almost constant acceptor densities for the CuO and Cu2O layers, except for the Cu-incorporated Cu2O layer prepared at −0.7 V. From the slope of the Mott–Schottky plots, the acceptor densities obtained for the CuO layers prepared at 0.4–0.7 V were in the orders of 1016 to 1017 cm–3 and for the Cu2O layers prepared at −0.4 to −0.6 V were in the order of 1016 cm–3. These values were close to those already reported for the CuO layers prepared by heating the Cu2O electrodeposits and for the Cu2O layers electrodeposited in the Cu(II)–lactic acid solution.10

Figure 6.

Figure 6

Open in a new tab

Effects of the deposition potentials on the flat-band potentials for CuO (A) and Cu2O layers (B).

The effect of the insertion of the n-ZnO layer on the structure of CuO and Cu2O layers prepared at 0.4 and −0.4 V is shown in Figure S5. The insertion of the n-ZnO layer between the Cu2O layer and FTO substrate induced the change in Cu2O grain structure from isolated coarse grains (Figure 4) to a dense and continuous layer. The degree of the preferred orientation was estimated by calculating Harris’s texture coefficient, Tc, by the following equation

graphic file with name ao9b03308_m007.jpg

where Im(hkl), I0(hkl), and n are the relative intensity of the measured (hkl) peak, the relative intensity of the (hkl) peak of the standard sample, and the total number of peaks used for the calculation, respectively.32 The Tc(111) for the Cu2O(111) plane was calculated by using the (110), (111), and (200) peaks and was estimated to be 0.7 and 1.4 for the Cu2O layers prepared on bare FTO and n-ZnO/FTO substrate. The n-ZnO layer possessed a (0001) preferred orientation from the X-ray diffraction pattern and maintains a lattice relationship of the (111) Cu2O//(0001) ZnO with a lattice mismatch of 7.1%. On the other hand, the orientation of the SnO2 layer in the FTO substrate was dispersed randomly. The n-ZnO layer and its orientation affected the grain structure and the preferred orientation of the resultant Cu2O layer. The insertion of the n-ZnO layer between the CuO layer and the FTO substrate showed no effects on the grain structure, thickness, and intensity ratio of the diffracted X-ray peaks of the CuO layer. There was a large lattice mismatch of over 15% for the (200) CuO/(0001) ZnO combination, and the lattice relationships between the monoclinic CuO and hexagonal ZnO lattices is a reason for no effects of the ZnO layer insertion on the electrochemical growth of the CuO layer. The potentials at 0.4 and −0.4 V were selected for the electrodeposition of CuO and Cu2O layers in bilayer constructions due to the small overpotential.

Photoelectrodeposition of CuO/Cu2O Bilayers

Figure 7 shows the chronoamperometry curves for the construction of Cu2O/CuO and CuO/Cu2O bilayers on the n-ZnO/FTO substrate by automatic switching of the potential in dark and under light irradiation from the SLG substrate side. The electric charge for the electrodeposition for both CuO and Cu2O layers was set at a constant absolute value of 1 C cm–2. In the construction of the CuO/Cu2O bilayer in dark, the cathodic current density increased from zero to a maximum value of −1.86 mA cm–2 when −0.4 V was applied for the Cu2O electrodeposition, and 662 s were needed for 1 C cm–2 (Figure 7A(a)). The following current density rapidly changed to the anodic side when 0.4 V was applied for the CuO electrodeposition and reached a plateau region at around 0.637 mA cm–2, which ended at 2,314 s in dark. The light irradiation induced an increase in the cathodic current density to −3.40 mA cm–2, which decreased with the passage of the deposition time till 388 s for the Cu2O electrodeposition, but no effect on the anodic current density was observed for the CuO electrodeposition (Figure 7A(b)). During the construction of the Cu2O/CuO bilayer in the dark, the anodic current density for the CuO electrodeposition gradually increased to 0.871 mA cm–2 at 2,816 s, and the current density rapidly shifted to zero when −0.4 V was applied for the Cu2O electrodeposition (Figure 7B(a)). The cathodic current density slowly increased to −0.848 mA cm–2 at 5,676 s, showing that a deposition time of 2,860 s was needed for the Cu2O electrodeposition of 1 C cm–2. The light irradiation gave a slight increase in anodic current density to 1.23 mA cm–2 and decreased slightly till 1,220 s (Figure 7B(b)). Applying −0.4 V for the Cu2O electrodeposition induced a rapid change to the cathodic side with a maximum value of −2.7 mA cm–2, and the deposition time for 1 C cm–2 shortened to 531 s, compared to 2,860 s in dark. The light irradiation strongly affected the Cu2O electrodeposition, but the effect was small for the CuO electrodeposition.

Figure 7.

Figure 7

Open in a new tab

Chronoamperometry curves for the photoelectrochemical deposition of CuO/Cu2O (A) and Cu2O/CuO bilayers (B) on ZnO/SLG substrates in dark (a) and under light irradiation (b). Potential for the electrodepositions of the CuO and Cu2O was set at 0.4 and −0.4 V vs Ag/AgCl.

Figure 8 shows the surface and cross-sectional images of CuO/Cu2O and Cu2O/CuO bilayers prepared by switching the potential in the dark and under light irradiation. The CuO and Cu2O layers in bilayers possessed characteristic monoclinic and cubic lattices with characteristic band gap energies of 1.50 and 2.05 eV, respectively, as shown in Figures S6 and S7. The CuO/Cu2O bilayers prepared in the dark and under light irradiation were almost the same in terms of the grain structures and thicknesses for both the CuO and Cu2O layers. The thickness of the CuO and Cu2O layers were 0.4 and 1.0 μm, respectively, and the CuO layers composed of granular grains were deposited over the entire Cu2O layer, irrespective of the light irradiation.

Figure 8.

Figure 8

Open in a new tab

FE-SEM images of surface morphologies (A) and cross-sectional structures (B) for CuO/Cu2O (a) and Cu2O/CuO bilayers (b) prepared on the n-ZnO/FTO substrate in the dark and under irradiation.

The light irradiation showed a drastic effect on the electrodeposition and structure of the Cu2O layer on the CuO layer, although the structure and thickness of the CuO layer deposited on the n-ZnO/FTO substrate were almost the same in the dark and under light irradiation. The isolated Cu2O grains approximately 2 μm in size were deposited sparsely on the CuO layer, and the underlying CuO surface could be observed between the Cu2O grains when deposited in the dark. The light irradiation contributes to the formation of continuous and dense Cu2O layer composed of the 2 μm size coarse grains without any pores.

Figure 9 shows the schematic illustration of the mechanism for photoelectrodeposition of the CuO layer on the Cu2O layer and vice versa. The flat-band potentials were estimated to be 0.145 and −0.106 V referenced to Ag/AgCl electrode for the CuO layer prepared at 0.4 V and Cu2O layer prepared at −0.4 V. In the anodic deposition of the CuO layer on the Cu2O layer, the holes (h+), that are the majority carriers, are swept toward the Cu(II)–tartrate complex solution by the electric field formed at the heterointerface at an applied potential of 0.4 V. Also, the ejected holes cause the anodic deposition (AD) of a layer from the alkaline Cu(II)–tartaric acid solution (pH ∼13), which is mainly composed of [Cu2(H–2L)2]4–, by the plausible mechanism as follows

graphic file with name ao9b03308_m008.jpg
graphic file with name ao9b03308_m009.jpg
graphic file with name ao9b03308_m010.jpg
graphic file with name ao9b03308_m011.jpg

The anodic potential applied to the FTO substrate decomposes H2O to generate H+, decreasing the local pH on the substrate surface (AD-i). The [Cu2(H–2L)2]4– complex is partially protonated to give [Cu2(H–1L)2]2–, which is unstable in the alkaline region,22 causing the precipitation of Cu(OH)2 on the substrate (AD-ii, iii). Subsequently, CuO crystals grow through the dehydration of Cu(OH)2 (AD-iv). A similar mechanism has been proposed for the anodic deposition of Ag2O from an alkaline solution containing [Ag(NH3)2]+ complex.33 The irradiated light at photon energy over the band gap energy induces the excitation of electrons from the valence band to the conduction band, resulting in the increase in the amount of electrons, which are the minority carriers, but the excited electrons are confined in the Cu2O layer due to the energy barrier formed at the heterointerface. Additional holes were generated by the excitation of electrons, but the amount was much less compared to the intrinsic amount of hole that is the majority carrier in the Cu2O layer. The effects of the light irradiation were limited to the electrodeposition and structure of the CuO layer, as shown in Figure 8.

Figure 9.

Figure 9

Open in a new tab

Schematic illustrations for the photoelectrochemical deposition of CuO on Cu2O layer (a) and Cu2O on CuO layer (b).

During the Cu2O deposition on the CuO layer, the energy barrier in the CuO layer bends downward at the heterointerface to the Cu–tartrate complex solution at −0.4 V for the Cu2O electrodeposition. The cathodic deposition (CD) mechanism of Cu2O from an alkaline Cu(II)–tartaric acid solution (pH ∼ 13) is mainly composed of [Cu2(H–2L)2]4– and can be explained as follows, similar to the cathodic Cu2O deposition from a Cu(II)–lactic acid solution34

graphic file with name ao9b03308_m012.jpg
graphic file with name ao9b03308_m013.jpg
graphic file with name ao9b03308_m014.jpg

The cathodic potential applied to the FTO substrate reduces [Cu2(H–2L)2]4– to generate unstable Cu+, which is then hydroxylated to precipitate CuOH on the substrate (CD-i, ii). The Cu2O crystals grow through the dehydration of CuOH (CD-iii). As such, since the electrons needed for the Cu2O electrodeposition are the minority carriers in the CuO layer, the number of electrons is quite low just after −0.4 V was applied and were then gradually transported to the heterointerface to the Cu–tartrate complex solution, resulting in the formation of isolated and coarse Cu2O grains in dark. The light irradiation at the photon energies higher than the band gap energy induced the increase in the number of electrons as minority carriers by excitation from the valence band to the conduction band, which were then swept down to the Cu–tartrate complex solution, resulting in the formation of a continuous Cu2O layer. This consideration is consistent with the difference in the electrodeposition and structure of the Cu2O layer prepared on the CuO layer in the dark and under light irradiation.

Conclusions

The p-CuO layers with the 1.5 eV band gap energy and monoclinic lattice, p-Cu2O layers with the 2.05 eV band gap energy and cubic lattice, and bilayers consisting of both have been prepared by controlling the potential in an aqueous solution containing copper(II) sulfate hydrate and tartaric acid in the dark and under light irradiation. The CuO layers were composed of granular grains on the FTO substrate with and without the n-ZnO layer, irrespective of the light irradiation. The isolated and coarse Cu2O grains were deposited sparsely on the FTO substrate, but the insertion of the n-ZnO layer induced the change to a continuous and dense layer, irrespective of the light irradiation. The flat-band potential was estimated to be at almost constant values of 0.149 V versus Ag/AgCl for CuO layers prepared at 0.4 to 0.7 V and approximately −0.10 V for Cu2O layers prepared at −0.4 to −0.6 V. The 0.4 μm CuO/1.1 μm Cu2O bilayers with characteristic grain structures could be prepared by cathodic polarization at −0.4 V, followed by an anodic polarization at 0.4 V, irrespective of the light irradiation. The electrodeposition of the Cu2O layer on the CuO layer was strongly affected by the light irradiation and continuous and defect-free 1.1 μm Cu2O/0.4 μm CuO bilayers could be prepared by anodic polarization at 0.4 V, followed by cathodic polarization at −0.4 V under the light irradiation. The mechanism for the photoelectrodeposition of CuO/Cu2O and Cu2O/CuO bilayers was discussed based on the energy band alignments drawn based on the flat-band potentials and applied potentials for the electrodeposition. The results demonstrated here reveals the ability to construct nanostructured oxide layers including the bilayers and further multilayers.

Experimental Section

An aqueous solution containing 0.3 mol L–1 copper(II) sulfate hydrate, 0.3 mol L–1 tartaric acid, and 1.5 mol L–1 sodium hydroxide was used for the electrodeposition of CuO, Cu2O, and bilayers, respectively, with a solution pH of 13.0. The electrodepositions were performed for CuO layers at potentials of 0.4, 0.5, 0.6, and 0.7 V and for Cu2O layers at −0.4, −0.5, −0.6, and −0.7 V, respectively, referenced to Ag/AgCl electrode for an electric charge of 1 C cm–2 with a Pt counter electrode using a polarization system (Hokuto Denko, HSV-110) connected with a Coulomb meter (Hokuto Denko, HF301), and the solution temperature was kept at 323 K. The CuO/Cu2O and Cu2O/CuO bilayers were fabricated by automatically switching the potential with the polarization system in dark and under light irradiation. The light was irradiated from the side of the glass substrate using a high-pressure mercury lamp (USHIO, OPTICAL-MODULEX, 500 W) without any optical filter. The ZnO layer was prepared by electrodeposition on the FTO substrate in an aqueous solution containing 80 mmol L–1 zinc nitrate hydrate at −0.8 V referenced to an Ag/AgCl electrode and 335 K for an electric charge of 0.5 C cm–2 using a potentiostat (Hokuto Denko, HAL 3000) connected with a Coulomb meter (Hokuto Denko, HF 301). The solution was prepared using reagent grade chemicals and distilled water purified with Milli Pore Ellix-UV-Advantage. F:SnO2/soda-lime glass (FTO, AGC Fabritech, Co. Ltd., 10 Ω sq–1) was used as the substrate. Prior to the electrodeposition, the FTO substrates were degreased by anodic polarizations in a 1 mol L–1 sodium hydroxide aqueous solution.

Cyclic voltammetry and chronoamperometry measurements were performed using the polarization system (Hokuto Denko, HSV-110). The X-ray diffraction patterns were recorded by a θ/2θ scanning technique with monochromated Cu Kα radiation operated at 40 kV and 200 mA using a Rigaku RINT2500. The optical absorption spectra were measured by recording the absorbance using a UV–vis–NIR spectrophotometer (Hitachi, U4100) with an integrated sphere, and the bare substrate was used as the reference. Electron microscopy observations were carried out at an accelerating voltage of 5 kV using field-emission scanning electron microscopy (FE-SEM, Hitachi, SU8000). An electrochemical impedance measurement was performed in a 0.1 mol L–1 sodium sulfate aqueous solution at pH 6.0 at room temperature by using VersaSTAT 3-100 (Princeton Applied Research), and Mott–Schottky plots were drawn by using a VersaStudio software (Princeton Applied Research).

Acknowledgments

This work was supported in part by a Grant-in-Aid for Scientific Research (19H02810).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.9b03308.

  • Solution appearance; complex species; Mott–Schottky plots; X-ray diffraction patterns; absorption spectra for bilayers (PDF)

Author Contributions

M.I. conceived the project, designed the experiments, and analyzed data. T.K., P.L.K., and T.S. carried out the material preparation and most of characterizations. All authors discussed the results and commented on the manuscript.

The authors declare no competing financial interest.

Supplementary Material

ao9b03308_si_001.pdf (4.7MB, pdf)

References

  1. Koffyberg F. P.; Benko F. A. A photoelectrochemical determination of the position of the conduction and valence band edges of p-type CuO. J. Appl. Phys. 1982, 53, 1173–1177. 10.1063/1.330567. [DOI] [Google Scholar]
  2. Snoke D. W.; Shields A. J.; Cardona M. Phonon-absorption recombination luminescence of room-temperature excitons in Cu2O. Phys. Rev. B. 1992, 45, 11693–11697. 10.1103/PhysRevB.45.11693. [DOI] [PubMed] [Google Scholar]
  3. Paracchino A.; Laporte V.; Sivula K.; Gratzel M.; Thimsen E. Highly active oxide photocathode for photoelectrochemical water reduction. Nat. Mater. 2011, 10, 456–461. 10.1038/nmat3017. [DOI] [PubMed] [Google Scholar]
  4. Masudy-Panah S.; Moakhar R. S.; Chua C. S.; Kushwaha A.; Dalapati G. K. Stable and efficient CuO based photocathode through oxygen-rich composition and Au-Pd nanostructure incorporation for solar hydrogen production. ACS Appl. Mater. Interfaces 2017, 9, 27596–27606. 10.1021/acsami.7b02685. [DOI] [PubMed] [Google Scholar]
  5. Izaki M.; Shinagawa T.; Mizuno K.; Ida Y.; Inaba M.; Tasaka A. Electrochemically constructed p-Cu2O/n-ZnO heterojunction diode for photovoltaic devices. J. Phys. D: Appl. Phys. 2007, 40, 3326–3329. 10.1088/0022-3727/40/11/010. [DOI] [Google Scholar]
  6. Khoo P. L.; Nagai M.; Watanabe M.; Izaki M. Electrodeposited CuO photovoltaic device with polysiloxane layer. J. Surf. Finish. Soc. Jpn. 2018, 69, 469–471. 10.4139/sfj.69.469. [DOI] [Google Scholar]
  7. Loferski J. J. Theoretical considerations governing the choice of the optimum semiconductor for photovoltaic solar energy conversion. J. Appl. Phys. 1956, 27, 777–784. 10.1063/1.1722483. [DOI] [Google Scholar]
  8. Yamaguchi M. III-V compound multi-junction solar cells: present and future. Sol. Energy Mater. Sol. Cells 2003, 75, 261–269. 10.1016/S0927-0248(02)00168-X. [DOI] [Google Scholar]
  9. Takamoto T.; Kaneiawa M.; Imaizumi M.; Yamaguchi M. InGaP/GaAs-based multijunction solar cells. Prog. Photovolt. Res. Appl. 2005, 13, 495–511. 10.1002/pip.642. [DOI] [Google Scholar]
  10. Yang Y.; Xu D.; Wu Q.; Diao P. Cu2O/CuO bilayered composite as a high-efficiency photocathode for photoelectrochemical hydrogen evolution reaction. Sci. Rep. 2016, 35158 10.1038/srep35158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Ho-Kimura S.; Moniz S. J.; Tang J.; Parkin I. P. A method for synthesis of renewable Cu2O junction composite electrode and their photoelectrocheimical properties. ACS Sustainable Chem. Eng. 2015, 3, 710–717. 10.1021/acssuschemeng.5b00014. [DOI] [Google Scholar]
  12. Minami T.; Nishi Y.; Miyata T. Efficiency enhancement using a Zn1-xGexO thin film as an n-type window layer in Cu2O-based heterojunction solar cells. Appl. Phys. Express 2016, 9, 052301 10.7567/APEX.9.052301. [DOI] [Google Scholar]
  13. Wang P.; Zhao X.; Li B. ZnO-coated CuO nanowire arrays: fabrications, optoelectronic properties, and photovoltaic applications. Opt. Express 2011, 19, 11271–11279. 10.1364/OE.19.011271. [DOI] [PubMed] [Google Scholar]
  14. Mizuno K.; Izaki M.; Murase K.; Shinagawa T.; Chigane M.; Inaba M.; Tasaka A.; Awakura Y. Structural and electrical characterizations of electrodeposited p-type semiconductor Cu2O films. J. Electrochem. Soc. 2005, 152, C179–C182. 10.1149/1.1862478. [DOI] [Google Scholar]
  15. Izaki M.; Nagai M.; Maeda K.; Fariza B. M.; Motomura K.; Sasano J.; Shinagawa T.; Watase S. Electrodeposition of 1.4-eV-bandgap-p-copper (II) oxide film with excellent photoactivity. J. Electrochem. Soc. 2011, 158, D578–D584. 10.1149/1.3614408. [DOI] [Google Scholar]
  16. Lincot D.; Guillemoles J. F.; Taunier S.; Guimard D.; Sicx-Kurdi J.; Chaumont A.; Roussel O.; Ramadani O.; Hubert C.; Fauvarque J. P.; Bodereau N.; Parrisi L.; Panheleux P.; Fanouillere P.; Naghavi N.; Grand P. P.; Benfarah M.; Mogensen P.; Kerrec O. Chalcopyrite thin film solar cells by electrodeposition. Sol. Energy 2004, 77, 725–737. 10.1016/j.solener.2004.05.024. [DOI] [Google Scholar]
  17. Ahmed S.; Reuter K. B.; Gunawan O.; Guo L.; Romankiw L. T.; Deligianni H. A high efficiency electrodeposited Cu2ZnSnS4 solar cell. Adv. Energy. Mater. 2012, 2, 253–259. 10.1002/aenm.201100526. [DOI] [Google Scholar]
  18. Wijesundera R. P.; Hidaka M.; Koga K.; Choi J.-Y.; Sung N. E. Structural and electronic properties of electrodeposited heterojunction of CuO/Cu2O. Ceramics 2010, 54, 19–25. [Google Scholar]
  19. Kim A.-Y.; Kim M. K.; Cho K.; Woo J.-Y.; Lee Y.; Han S.-W.; Byun D.; Choi W.; Lee J. K. One-step catalytic synthesis of CuO/Cu2O in a grapitized porous C matric derived from the Cu-based metal-organic framework for Li- and Na-ion batteries. ACS Appl. Mater. Interfaces 2016, 8, 19514–19523. 10.1021/acsami.6b05973. [DOI] [PubMed] [Google Scholar]
  20. Izaki M.; Fukazawa K.; Sato K.; Khoo P.-L.; Kobayashi M.; Takeuchi A.; Uesugi K. Defect structure and photovoltaic characteristics of internally stacked CuO/Cu2O photoactive layer prepared by electrodeposition and heating. ACS Appl. Energy Mater. 2019, 2, 4833–4840. 10.1021/acsaem.9b00514. [DOI] [Google Scholar]
  21. Poizot P.; Hung C.-J.; Nikiforov M. P.; Bohannan E. W.; Switzer J. A. An electrochemical method for CuO thin film deposition from aqueous solution. Electrochem. Solid State Lett. 2003, 6, C21–C25. 10.1149/1.1535753. [DOI] [Google Scholar]
  22. Chen T.; Kitada A.; Seki Y.; Fukami K.; Usmanov D. T.; Chen L. C.; Hiraoka K.; Murase K. Identification of copper(II)-lactate complexes in Cu2O electrodeposition baths; Deprotonation of α-hidroxyl group in highly concentrated alkaline solution. J. Electrochem. Soc. 2018, 165, D444–D451. 10.1149/2.0831810jes. [DOI] [Google Scholar]
  23. Hörner T. G.; Klüfers P. The Species of Fehling’s Solution. Eur. J. Inorg. Chem. 2016, 2016, 1798–1807. 10.1002/ejic.201600168. [DOI] [Google Scholar]
  24. Joint Committee on Powder Diffraction Standards . Powder Diffraction File; International Data for Diffraction Data: Swarthmore, PA, 1992; pp 5–667. [Google Scholar]
  25. Joint Committee on Powder Diffraction Standards . Powder Diffraction File; International Data for Diffraction Data: Swarthmore, PA, 1992; 2–1041. [Google Scholar]
  26. Mizuno K.; Izaki M.; Murase K.; Shinagawa T.; Chigane M.; Inaba M.; Tasaka A.; Awakura Y. Structural and electrical characterizations of electrodeposited p-type semiconductor Cu2O films. J. Electrochem. Soc. 2005, 152, C179–C182. 10.1149/1.1862478. [DOI] [Google Scholar]
  27. Luque A.; Hegedus S.. Handbook of Photovoltaic Science and Engineering; John Wiley & Sons: West Sussex, 2003; pp. 70–74. [Google Scholar]
  28. Govindaraju G. V.; Wheeler G.; Lee D.; Choi K.-S. Methods for electrochemical synthesis and photoelectrochemical characterization for photoelectrodes. Chem. Mater. 2017, 29, 355–370. 10.1021/acs.chemmater.6b03469. [DOI] [Google Scholar]
  29. Nakaoka K.; Ueyama J.; Ogura K. Photoelectrochemical behaivior of electrodeposited CuO and Cu2O thin films on conducting substrates. J. Electrochem. Soc. 2004, 151, C661–C665. 10.1149/1.1789155. [DOI] [Google Scholar]
  30. Sivula K.; von de Krol R. Semiconducting materials for photoelectrochemical energy conversion. Nat. Rev. Mater. 2016, 1, 15010 10.1038/natrevmats.2015.10. [DOI] [Google Scholar]
  31. Sze S. M.; Ng K. K.. Physics of Semiconductor Devices; 3ed ed.; John Wiley & Sons: New Jersey, 2007; pp. 79–90. [Google Scholar]
  32. Shinagawa T.; Watanabe M.; Mori T.; Tani J.-I.; Chigane M.; Izaki M. Oriented transformation from layered zinc hydroxides to nanoporous ZnO: A comparative study of different anion types. Inorg. Chem. 2018, 57, 13137–13149. 10.1021/acs.inorgchem.8b01242. [DOI] [PubMed] [Google Scholar]
  33. Ida Y.; Watase S.; Shinagawa T.; Watanabe M.; Chigane M.; Inaba M.; Tasaka A.; Izaki M. Direct Electrodeposition of 1.46 eV Bandgap Silver (I) Oxide Semiconductor Films by Electrogenerated Acid. Chem. Mater. 2008, 20, 1254–1256. 10.1021/cm702865r. [DOI] [Google Scholar]
  34. Shinagawa T.; Ida Y.; Mizuno K.; Watase S.; Watanabe M.; Inaba M.; Tasaka A.; Izaki M. Controllable Growth Orientation of Ag2O and Cu2O Films by Electrocrystallization from Aqueous Solutions. Cryst. Growth Des. 2013, 13, 52–58. 10.1021/cg300813z. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao9b03308_si_001.pdf (4.7MB, pdf)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES