1. Introduction

The oxides of Cu and its alloys are attractive p-type semiconductors for solar-driven chemical production via the hydrogen evolution reaction (HER) and CO₂ reduction reaction (CO₂RR) [1]. Among them, CuBi₂O₄ (CBO) has drawn attention due to the suitability of its optical properties and energy band structures for solar-driven chemical reactions. It has a direct band gap of 1.6–1.8 eV that corresponds to visible light in the solar spectrum and also has a high absorption coefficient of 10⁴ to 10⁵ cm⁻¹ for wavelengths shorter than 650 nm [2]. In addition, CBO has sufficiently large driving force for HER; its conduction band edge position is more negative than the thermodynamic redox potential of HER. It has onset potential > 1 V versus the reversible hydrogen electrode (RHE) for HER, while other p-type oxide semiconductors (such as Cu₂O, CuO, and CuFeO₂) show lower onset potential near 0.6–0.9 V (vs RHE) [3–5]. The optical band gap and band edge position of CBO make it an ideal candidate for a photocathode of a cost-effective and highly efficient solar-driven tandem cell for chemical production. Recent modelling predicted that the highest photoelectrochemical (PEC) water splitting efficiency could be achieved from a tandem cell comprising top cells with a band gap of 1.59 eV and bottom cells with a band gap of 0.92 eV with best catalysts such as Pt and IrO_x, implying that p-CBO/n-Si tandem PEC cells could achieve ~23% solar-to-hydrogen efficiency [6]. Currently there is no report of monolithic tandem cell using CBO photoanode. Yang et al. demonstrated overall water splitting with parallel illumination tandem PEC cell with a CBO photocathode and a TiO₂ photoanode [7]. However, a monolithic tandem PEC cell should be developed as it utilizes larger part of the sunlight spectrum, leading to higher overall efficiency [8].

Various synthetic processes including electrodeposition [9–12], hydrothermal [13], and sol-gel synthesis [2,14–17], have been used to form CBO films for the PEC water splitting reaction. However, previously reported CBO films often exhibit non-uniform, nanoporous morphology, which can result in reduced PEC performance. For example, the porous structure of CBO films can expose the underlying substrates to the electrolyte, which leads to reduction of the photovoltage and increase of the dark current. This can be further aggravated if the CBO surface is modified by co-catalysts. Increasing the thickness of nanoporous films can minimize exposure of the substrate to the electrolyte; however, it could inversely lower the photocurrent due to limited charge transport [18]. Although Wang et al. recently demonstrated that thin, dense CBO film could be fabricated by spray pyrolysis, the slow deposition rate when using spray pyrolysis (~2.3 nm min⁻¹) does not seem suitable for large-scale film fabrication [19]. Therefore, strong demand exists for a simple, fast synthetic process to form thin non-porous CBO films for large-scale PEC applications.

Herein, a facile spin coating process is presented by which to produce large-scale, thin, dense CBO films for PEC energy conversion. We demonstrate that the addition of polyvinylpyrrolidone (PVP) in the CBO precursor solutions is the key to forming non-porous, compact CBO films on fluorine-doped tin oxide (FTO)-coated glass substrates by spin coating. PVP additives have high viscosity and facilitate uniform distribution of the metal ions through PVP-metal ion interactions in a precursor film. This enables deposition of ~130 nm-thick, non-porous CBO films on FTO substrates with high temperature annealing. Improved optical and PEC properties of the CBO films were achieved with optimum metal ion concentrations in the precursor solutions, as well as by introducing two-step annealing processes. This highlights the importance of the homogeneous microstructure and dense morphology of the CBO films synthesized on FTO substrates.

2. Experiments

2.1 Fabrication of CuBi₂O₄ thin films

All chemicals were used as purchased without further purification. Precursor solutions for spin coating were prepared by dissolving Bi(NO₃)₃⋅5H₂O (Sigma Aldrich, 99.99%) and Cu(NO₃)₂⋅2.5H₂O (Sigma Aldrich, 99.99%) in a 2:1 molar ratio, and 0.024 M of PVP (Sigma Aldrich, average molecular weight (MW) = 10,000), sequentially into 5 mL of dimethylformamide (DMF, Merck, 99.8%) with sonication and vigorous vortexing. FTO-coated glasses (2 × 2 cm², 7 Ω cm⁻¹, Wooyang CMS Co.) or quartz (University Wafer Inc., USA) were used after cleaning in methanol, acetone, and isopropyl alcohol with sonication for 15 min in each solution. A portion of the CBO precursor solution (1.5 mL) was dispensed drop-by-drop onto a substrate rotating at 3000 rpm, which was followed by additional spinning for 30 s. The spin-coated CBO precursor films were then calcined using a two-step annealing process: annealing at 200 °C for 2 h, followed by annealing at 500 °C for 2 h in air. The temperature was ramped up at a rate of 5 °C min⁻¹. For comparison, one-step annealing at 500 °C for 2 h with the same ramp up rate was also carried out.

2.2 Physical and PEC characterization of CBO thin films

The morphology of CBO films was characterized by environmental scanning electron microscopy (ESEM, FEI Co., Quanta FEG 650) without additional coating. Crystal structures of the CBO films were characterized by X-ray diffraction (XRD, RIGAKU, Ultima IV) with reference to the Joint Committee on Powder Diffraction Standards (JCPDS# 48-1886, # 46-1088, and # 27-0050 for CuBi₂O₄, SnO₂, and Bi₂O₃, respectively). The band gap of the CBO films was derived from the absorption spectra of the CBO films coated on quartz using a UV-Vis spectrometer (Perkin Elmer, Lambda 1050) equipped with an integrating sphere. The functional groups on PVP at the film surface were analyzed with a Fourier transform infrared spectrometer (FT-IR, ThermoFisher Sci., Nicolet 6700) and a Raman spectroscope (Nanobase, XperRam compact) with 532 nm diode-pumped laser.

CBO photoelectrodes for PEC measurements were prepared as described in the literature [20–23]. The edges of CBO films on FTO substrates were scribed to expose the underlying FTO and a Cu wire was connected to the exposed FTO surface using silver paste. The area of the photoelectrodes was defined by isolating the electrode surface with an epoxy resin (Loctite 9460). A 300 W Xenon lamp (Oriel Instrument, Model 6258) with AM 1.5 G and water infrared filters (Newport, Liquid filter 1.5 IN ALUM body model 61945) was used to simulate one-sun illumination. The light intensity was calibrated to AM 1.5G (100 mW cm⁻²) with a certified reference Si solar cell (Oriel Instrument, 91150V). All PEC measurements were conducted using a 3-electrode setup on a potentiostat (Bio-Logic, SP-150). A Pt coil and Ag/AgCl electrode (saturated) were used for the counter and reference electrode, respectively. All liner sweep voltammetry measurements were carried out at a scan rate of 20 mV s⁻¹ without stirring the electrolyte. The PEC oxygen reduction reaction (ORR) was chosen as a model PEC reaction due to the stable nature of CBO during ORR in alkaline solutions. Oxygen (99.995%) was bubbled through NaOH (0.1 M, pH 12.8; Sigma Aldrich, ≥ 98%) for ORR. Chronoamperometry shows that our CBO films were stable for more than 10 minutes at 0.4 V vs RHE in O₂ saturated 0.1 M NaOH under AM 1.5G illumination (Fig. 6 in Appendix). Electrochemical impedance spectroscopy (EIS) was performed under AM 1.5G illumination with a bias of 0.147 V vs Ag/AgCl while the frequency was swept from 10⁶ to 1 Hz with 10 mV AC amplitude.

3. Results and discussion

3.1 Preparation of CuBi₂O₄ film using spin coating

Figures 1(a)–1(d) shows SEM images of CBO films on FTO substrates, with and without PVP additives in the precursor solutions, after their formation by the two-step annealing process. The CBO precursor solutions contained 1 M Bi(NO₃)₃⋅5H₂O and 0.5 M Cu(NO₃)₂⋅2.5H₂O in DMF. When the PVP additives were not present in the precursor solution, islands of CBO particles formed and the underlying FTO substrate was exposed (Fig. 1(a)). Spin coating without PVP additives resulted in the deposition of non-uniform CBO films with uncontrollable thickness, which could be aggravated on a rough substrate such as FTO. The addition of 0.024 M PVP (average MW of 10,000) in the CBO precursor solution dramatically changes the morphology of CBO films and non-porous CBO films, as shown in Fig. 1(c). The typical average grain size of the CBO film is about 140 nm. In addition, uniformly ~130 nm-thick, smooth CBO film was formed even on a rough FTO substrate without exposure of the FTO surface, as presented in Fig. 1(d).

 

Fig. 1 (a) Plan view and (b) Tilted view scanning electron microscope (SEM) images of CuBi₂O₄ (CBO) films on fluorine doped tin oxide (FTO)-coated glass substrates, synthesized without PVP additives in a precursor solution; (c) Plan view and (b) Tilted view SEM images of CBO films on FTO coated glass substrates, synthesized with PVP additives in a precursor solution. The CBO precursor solution was 1 M Bi(NO₃)₃⋅5H₂O and 0.5 M Cu(NO₃)₂⋅2.5H₂O in DMF and 0.024 M PVP (average MW of 10,000) was used as an additive. The two-step annealing process was used to calcine the CBO films.

Download Full Size | PDF

Addition of PVP in the precursor solutions has produced thick, dense, and stoichiometric metal oxide films when adopted during sol-gel synthesis [24–28]. PVP additives increase the viscosity of precursor solutions, which promotes thicker and more uniform deposition of the precursor solution on substrates [29,30]. PVP could also prevent crack formation in the film by relaxing tensile stress of the film during calcination process [31]. Moreover, PVP contains a tertiary amide group in the individual units that has a stronger affinity for metal ions [32] so that metal ions can be uniformly distributed in the coated precursor films. For example, the PVP additive used in this work has an average MW of 10,000 and approximately 90 tertiary amide groups per PVP molecule. In this case, the ratio between Cu²⁺ and Bi³⁺ ions and the tertiary amide groups of PVP additives in the CBO precursor solutions can be derived using the following equation.

According to Eq. (1), the PVP to metal ratio is about 0.7 for the CBO precursor films used in Fig. 1(d). Therefore, metal ions could be homogeneously distributed in the CBO precursor films through interaction with sufficiently large numbers of amide groups, which could lead to uniform nucleation and growth of CBO films upon annealing.

The XRD patterns of CBO films with and without PVP additives show characteristic peaks of polycrystalline CBO phases, as shown in Fig. 2(a). In the absence of PVP additive, the CBO film showed decreased crystallinity with reduced peak intensities. In addition, traces of β-Bi₂O₃ peaks appeared at 33.5°. The absence of PVP additives resulted in formation of a secondary phase due to the fast hydrolysis of Bi³⁺ ions in precursor solution [19], which can degrade the PEC property of the CBO films [33]. Strong affinity of a tertiary amide group of PVP for metal ions, may suppress the formation of Bi₂O₃.

 

Fig. 2 (a) X-ray diffraction (XRD) patterns of CBO films synthesized with (w/) and without (w/o) PVP on FTO substrates. (b) Tauc plot of the CBO films on quartz substrates. The inset shows the corresponding absorbance.

Download Full Size | PDF

In addition, PVP with higher molecular weight of 40,000 but same mass was used to decide the effect of chain length of PVP on the formation of pore-free CBO films. As shown in Fig. 7(a) (in Appendix), densely packed morphology of CBO films without pores were obtained when PVP. However, XRD in Fig. 7(b) (in Appendix) obviously confirmed that the film had various undesired phases such as Bi₂O₃ and CuO. These secondary phases might be attributed to the insufficient decomposition of PVP due to high molecular weight, which could suppress the metal ion supply. From these results, it can be inferred that there should be an optimum molecular weight of PVP to obtain dense CBO films, and thus, further studies are under way to find it.

As presented in Fig. 2(b), the optical band gap of the CBO films was estimated with a Tauc plot obtained from the UV-Vis absorption spectra (inset of Fig. 2(b)). The optical band gap of the CBO film synthesized in the presence of PVP is about 1.63 eV, which is in the appropriate band gap range of 1.6–1.8 eV from the literature [2]. Note that the theoretical current density [34] of a photocathode from a semiconductor having band gap of 1.63 eV is ~23 mA cm⁻² when it is illuminated with AM 1.5G.

The concentration of the metal salts in a precursor solution can influence the morphology and microstructure of a metal oxide film during a sol-gel process. Figure 3 displays digital photographs and SEM images of CBO films formed on FTO substrates with various concentrations of metal salts in sol-gel precursors. Note again that the 2:1 ratio of Bi:Cu salts and 0.024 M of PVP additives were maintained in all cases. Digital photographs highlight the uniform deposition of CBO films on FTO substrates through the PVP-assisted sol-gel synthesis (Figs. 3(a)–3(c)). CBO films become darker brown as the Bi³⁺ and Cu²⁺ ion concentrations increase. If the concentrations of Bi³⁺ and Cu²⁺ ions were further increased to 1.5 M and 0.75 M, the CBO film became slightly hazy (Fig. 3(c)). This apparent color change of CBO films due to different metal ion concentrations originates from the morphology of CBO films as revealed by the SEM images in Figs. 3(d)–3(i). When the precursor solution with the smallest amount of Bi³⁺ and Cu²⁺ was used, a CBO film with thickness of ~90 nm was formed with numerous voids. Void-free CBO films were formed when the concentration of the precursor solutions increased to 1 M and 0.5 M. This leads to higher photo-absorption and a darker appearance of the CBO film in Fig. 3(b). With 1.5 M/0.75 M Bi³⁺/Cu²⁺ ion concentrations, the thickness of CBO films increases to ~170 nm and several-micrometer sized hillocks, responsible for the hazy appearance, are formed (shown in Fig. 3(i) and inset of Fig. 3(f), respectively). Note that the grain of the CBO films from 1.5 M/0.75 M Bi³⁺/Cu²⁺ ion concentrations are significantly larger than those from lower Bi³⁺/Cu²⁺ ion concentrations. This larger grain size of CBO films may cause bad contacts on a rough FTO substrate surface (Fig. 3(i)).

 

Fig. 3 Photograph and SEM images of CBO films synthesized at different precursor solution concentrations; Digital photograph images of CBO films from precursor solutions with metal ion concentrations of (a) 0.5 M/0.25 M Bi³⁺/Cu²⁺, (b) 1 M/0.5 M Bi³⁺/Cu²⁺, and (c) 1.5 M/0.75 M Bi³⁺/Cu²⁺; (b)–(f) plan view and (g)–(i) cross-sectional view SEM images of the corresponding CBO films.

Download Full Size | PDF

The decomposition of PVP during calcination can also influence the microstructure and morphology of CBO films by interfering with nucleation and growth. It has been reported that fast decomposition of PVP in a sol-gel precursor film increased porosity in BaTiO₃ and Pb(Zr,Ti)O₃ films during PVP-assisted sol-gel synthesis [25,27,35]. To suppress the fast decomposition of PVP and the associated porosity in CBO films; we introduced a two-step annealing process for the CBO synthesis. In the two-step annealing process, the coated CBO precursor films were first subjected to annealing at 200 °C for 2 h, followed by high temperature annealing at 500 °C for 2 h in ambient air. One-step annealing was also carried out at 500 °C for 2 h for comparison.

Figures 4(a)–4(d) show SEM images of CBO films calcined with different calcination profiles. Clearly, the CBO films synthesized with the two-step annealing process have a compact morphology without voids in the films or at the interface. In addition, it is apparent that the CBO films synthesized using the two-step annealing process, have larger grain sizes than those formed by the one-step annealing process. XRD analysis indicates that the two-step annealed CBO films have crystallite size of about 36 nm, slightly larger than the ones obtained from one-step annealing (Fig. 8 in Appendix). This is attributed, in part, to slow decomposition of PVP and uniform nucleation of CBO particles during the first, low-temperature annealing step.

 

Fig. 4 Plan view SEM images of CBO film synthesized by (a) two- and (b) one-step annealing process; (c) and (d) are cross-sections of (a) and (b), respectively.

Download Full Size | PDF

As PVP is known to decompose above 150 °C when it is bound with metal ions, the first annealing step at 200 °C decomposes PVP monomers and breaks the amide bonds in the PVP monomers, which is called intramolecular decomposition [36]. The FT-IR spectra in Fig. 9(a) (in Appendix) show that absorption bands for the functional groups in PVP monomers such as C−N, CH₂, C = O, C−H, and O−H [37,38] disappear during the first annealing at 200 °C for 2 h, which confirms that PVP monomers are mostly decomposed during the annealing. Raman analysis also indicates that CBO particles are produced during the first annealing step as presented in Fig. 9(b) (in Appendix) [39]. The SEM analysis in Fig. 10 (in Appendix) further reveals that numerous CBO particles were formed uniformly at or near the FTO surface during the first annealing. Following the intramolecular decomposition, during the second annealing step at 500 °C, the remaining PVP was subject to intermolecular decomposition, after which no PVP peaks appeared in FT-IR spectra, and the densification of CBO film is proceeded from the nucleated CBO during the first annealing step. In contrast, the one-step annealing process drives fast decomposition of PVP molecules and inhomogeneous nucleation of CBO particles, which could lead to pore formation in the CBO film and bad contact with the FTO substrate. No matter what annealing process is involved, all the PVP was decomposed at the end after the completion of the annealing at 500 °C as no traces of PVP is found from the FT-IR spectra. Therefore, it can be concluded that the uniform distribution metal ions and nucleation of CBO near FTO is the key role for the synthesis of dense CBO films.

3.2 Photoelectrochemical properties of CBO films

As a model PEC reaction, the PEC ORR activity of CBO films grown by the PVP-assisted synthesis was investigated. Figure 5(a) shows linear sweep voltammetry (LSV) curves of CBO films with various precursor concentrations, of which the morphology and microstructures are shown in Fig. 3. The PEC ORR, as it takes place in O₂-saturated 0.1 M NaOH under simulated one-sun illumination, exhibits an obvious dependence upon morphology. The photocurrent density of the CBO films increases as the porosity of the CBO film decreases: the dense CBO film from the 1 M/0.5 M Bi/Cu precursor solution shows the highest photocurrent density (−0.4 mA cm⁻²) at 0.4 V. This is due to the enhanced light absorption of the non-porous, dense CBO film with ~130 nm thickness which is evident in the digital photograph images in Figs. 3(a) and 3(b). The CBO films synthesized without and with PVP has high contrast in porosity as shown in Figs. 1(a) and 1(c). Here the photocurrent density also shows same tendency with respect to porosity, as dense CBO film synthesized with PVP shows twice high photocurrent density compared to porous CBO film synthesized without PVP (Fig. 11 in Appendix). However, when the precursor concentration is further increased to 1.5 M/0.75 M Bi/Cu, the photocurrent density drops rather dramatically to < −0.4 mA cm⁻², although the resulting CBO film does not have voids in the film, as seen in Fig. 3(f). The decreased photocurrent of the 1.5 M/0.75 M Bi³⁺/Cu²⁺ CBO film originates in part from the formation of large voids at the CBO/FTO interface and the resultant suppressed charge transport to the FTO substrates. This notion can be further supported by the different PEC ORR behaviors with respect to the direction of illumination. In general, a metal oxide semiconductor with poor bulk charge transport exhibits higher PEC activity with backside illumination than with front-side illumination due to facile charge separation and collection at the semiconductor/back contact interface [40]. CBO films are also known for poor charge transport [2,9,19]. Thus, higher photocurrent from the back illumination is anticipated for CBO films. While the other CBO films show higher photocurrent density with backside illumination, the 1.5 M/0.75 M Bi³⁺/Cu²⁺ CBO film has similarly reduced photocurrent regardless of the illumination direction, as presented in Fig. 5(a). This suggests that the poor charge separation and collection at the CBO/FTO interface caused by bad contact, can limit the PEC ORR photocurrent density. The influence of the voids at the CBO/FTO interface can also be seen in Fig. 5(b), where the CBO film calcined using the two-step annealing process shows twice as high photocurrent than for the CBO film with the one-step annealing process. EIS can further elucidate the role of various interfaces in a photoelectrode on a PEC reaction [41]. Chronoamperometry shows that our CBO films were stable for more than 10 minutes at 0.4 V vs RHE in O₂ saturated 0.1 M NaOH under AM 1.5G illumination (Fig. 6). Poor charge transport from the numerous voids formation at the CBO/FTO interface in the one-step annealing process is clearly supported by EIS in Fig. 12 (in Appendix), where the one-step annealed CBO film has twice large resistance at the CBO/FTO interface (Table 1 in Appendix). Note that both films have similar UV-Vis absorption spectra (Fig. 13 in Appendix). Therefore, to achieve high PEC performance of the CBO, it is important to obtain an intimate contact with the adjacent layer and a pore-free microstructure and morphology of films by precisely controlling the precursor concentration and annealing profile of the PVP-assisted solution process. Recalling that highly dense and pore-less film is favorable for top absorber layer in monolithic tandem PEC cell, dense CBO film formed by this process can be combined with Si photoanode for high efficiency PEC devices.

 

Fig. 5 Linear sweep voltammetry scans of CBO films synthesized with (a) Various precursor solution concentrations, and with (b) Different annealing processes. All measurements were performed in 0.1 M NaOH (pH 12.8) saturated with O₂ under AM 1.5 G illumination (100 mW cm⁻²). Solid and dash-dot lines indicate back and front-side illumination, respectively. Dotted lines represent the dark current density.

Download Full Size | PDF

 

Fig. 6 Chronoamperometric measurement at 0.4 V vs RHE for the CBO thin film synthesized by two-step annealing process.

Download Full Size | PDF

Table 1. Resistances obtained from the EIS analysis of CBO photocathodes with different annealing profiles.

View Table

4. Conclusions

In this paper, we presented a facile solution-based process to form high density, void-free CBO films on FTO substrates for photoelectrochemical chemical conversions. The introduction of PVP molecules in a precursor solution promoted the formation of a compact CBO film on a rough FTO substrate and suppressed the formation of secondary Bi₂O₃ phase through the PVP-metal ion interaction in the precursor solution. The morphology of the synthesized CBO films on FTO substrates, was influenced by various processing parameters such as metal ion concentration and annealing temperature profile. A 130 nm-thick, non-porous CBO film without voids at the CBO/FTO interface was achieved with 1 M/0.5 M Bi³⁺/Cu²⁺ metal ions and 0.024 M of PVP in the precursor solution, and a two-step annealing process. The large grain-size of CBO films from un-optimized metal ion concentration, or fast decomposition of PVP molecules during annealing, resulted in void formation at a rough CBO/FTO interface. The PEC ORR of the CBO films revealed that it is critical to have intimate contact with the FTO substrates to enhance activity. The dense CBO photocathodes fabricated using this PVP-assisted solution process are expected to be integrated into Si bottom cells to create the most efficient PEC tandem cells for water splitting and CO₂RR.

Appendix

 

Fig. 7 (a) SEM image of CuBi₂O₄ (CBO) film synthesized with PVP molecular weight of 40,000. (b) XRD spectra of CBO film synthesized with PVP molecular weight of 40,000.

Download Full Size | PDF

 

Fig. 8 (a) X-ray diffraction (XRD) spectra of CBO films synthesized by two- and one-step annealing processes. (b) The crystallite sizes of the corresponding CBO films determined by the Scherrer equation.

Download Full Size | PDF

 

Fig. 9 (a) FT-IR spectra of PVP and CBO at different calcination conditions. (b) Raman spectra of CBO at different calcination conditions.

Download Full Size | PDF

 

Fig. 10 SEM images after calcination at 200 °C for 2 h.

Download Full Size | PDF

 

Fig. 11 Chopped light linear sweep voltammetry scans of CBO films synthesized without and with PVP. Light was illuminated from the backside of the film.

Download Full Size | PDF

 

Fig. 12 Electrochemical impedance spectroscopy (EIS) of CBO films synthesized by the two- and one-step annealing processes. The inset shows the equivalent circuit model of CBO/FTO photoelectrodes. C_FTO-CBO and C_CBO-El are the capacitive elements assigned to the CBO/FTO and CBO/electrolyte interfaces, respectively, and R_FTO-CBO and R_CBO-El are the resistance at the FTO/CBO and CBO/electrolyte interfaces, respectively. R_s is the series resistance.

Download Full Size | PDF

 

Fig. 13 UV-visible absorption spectra of CBO films synthesized by two- and one-step annealing processes.

Download Full Size | PDF

Funding

National Research Foundation of Korea (NRF) (2017R1A2B4008736).

References

1. D. Kang, T. W. Kim, S. R. Kubota, A. C. Cardiel, H. G. Cha, and K. S. Choi, “Electrochemical synthesis of photoelectrodes and catalysts for use in solar water splitting,” Chem. Rev. 115(23), 12839–12887 (2015). [CrossRef]   [PubMed]  

2. S. P. Berglund, F. F. Abdi, P. Bogdanoff, A. Chemseddine, D. Friedrich, and R. van de Krol, “Comprehensive evaluation of CuBi₂O₄ as a photocathode material for photoelectrochemical water splitting,” Chem. Mater. 28(12), 4231–4242 (2016). [CrossRef]  

3. M.-K. Son, L. Steier, M. Schreier, M. T. Mayer, J. Luo, and M. Grätzel, “A copper nickel mixed oxide hole selective layer for au-free transparent cuprous oxide photocathodes,” Energy Environ. Sci. 10(4), 912–918 (2017). [CrossRef]  

4. Y. J. Jang, J.-W. Jang, S. H. Choi, J. Y. Kim, J. H. Kim, D. H. Youn, W. Y. Kim, S. Han, and J. Sung Lee, “Tree branch-shaped cupric oxide for highly effective photoelectrochemical water reduction,” Nanoscale 7(17), 7624–7631 (2015). [CrossRef]   [PubMed]  

5. U. Kang, S. K. Choi, D. J. Ham, S. M. Ji, W. Choi, D. S. Han, A. Abdel-Wahab, and H. Park, “Photosynthesis of formate from CO₂ and water at 1% energy efficiency via copper iron oxide catalysis,” Energy Environ. Sci. 8(9), 2638–2643 (2015). [CrossRef]  

6. K. T. Fountaine, H. J. Lewerenz, and H. A. Atwater, “Efficiency limits for photoelectrochemical water-splitting,” Nat. Commun. 7(1), 13706 (2016). [CrossRef]   [PubMed]  

7. J. Li, M. Griep, Y. Choi, and D. Chu, “Photoelectrochemical overall water splitting with textured CuBi₂O₄ as a photocathode,” Chem. Commun. (Camb.) 54(27), 3331–3334 (2018). [CrossRef]   [PubMed]  

8. K. Zhang, M. Ma, P. Li, D. H. Wang, and J. H. Park, “Water splitting progress in tandem devices: moving photolysis beyond electrolysis,” Adv. Energy Mater. 6(15), 1600602 (2016). [CrossRef]  

9. D. Kang, J. C. Hill, Y. Park, and K.-S. Choi, “Photoelectrochemical properties and photostabilities of high surface area CuBi₂O₄ and Ag-doped CuBi₂O₄ photocathodes,” Chem. Mater. 28(12), 4331–4340 (2016). [CrossRef]  

10. Y. Nakabayashi, M. Nishikawa, and Y. Nosaka, “Fabrication of CuBi₂O₄ photocathode through novel anodic electrodeposition for solar hydrogen production,” Electrochim. Acta 125(3), 191–198 (2014). [CrossRef]  

11. N. T. Hahn, V. C. Holmberg, B. A. Korgel, and C. B. Mullins, “Electrochemical synthesis and characterization of p-CuBi₂O₄ thin film photocathodes,” J. Phys. Chem. C 116(10), 6459–6466 (2012). [CrossRef]  

12. D. Cao, N. Nasori, Z. Wang, Y. Mi, L. Wen, Y. Yang, S. Qu, Z. Wang, and Y. Lei, “p-Type CuBi₂O₄: An easily accessible photocathodic material for high-efficiency water splitting,” J. Mater. Chem. A Mater. Energy Sustain. 4(23), 8995–9001 (2016). [CrossRef]  

13. G. Sharma, Z. Zhao, P. Sarker, B. A. Nail, J. Wang, M. N. Huda, and F. E. Osterloh, “Electronic structure, photovoltage, and photocatalytic hydrogen evolution with p-CuBi₂O₄ nanocrystals,” J. Mater. Chem. A Mater. Energy Sustain. 4(8), 2936–2942 (2016). [CrossRef]  

14. H. S. Park, C.-Y. Lee, and E. Reisner, “Photoelectrochemical reduction of aqueous protons with a CuO|CuBi₂O₄ heterojunction under visible light irradiation,” Phys. Chem. Chem. Phys. 16(41), 22462–22465 (2014). [CrossRef]   [PubMed]  

15. J. Yang, C. Du, Y. Wen, Z. Zhang, K. Cho, R. Chen, and B. Shan, “Enhanced photoelectrochemical hydrogen evolution at p-type CuBi₂O₄ photocathode through hypoxic calcination,” Int. J. Hydrogen Energy 43(20), 9549–9557 (2018). [CrossRef]  

16. N. Xu, F. Li, L. Gao, H. Hu, Y. Hu, X. Long, J. Ma, and J. Jin, “Polythiophene coated CuBi₂O₄ networks: A porous inorganic–organic hybrid heterostructure for enhanced photoelectrochemical hydrogen evolution,” Int. J. Hydrogen Energy 43(4), 2064–2072 (2018). [CrossRef]  

17. N. Xu, F. Li, L. Gao, H. Hu, Y. Hu, X. Long, J. Ma, and J. Jin, “N,Cu-codoped carbon nanosheet/Au/CuBi₂O₄ photocathodes for efficient photoelectrochemical water splitting,” ACS Sustain. Chem.& Eng. 6(6), 7257–7264 (2018). [CrossRef]  

18. Y. Kuang, Q. Jia, H. Nishiyama, T. Yamada, A. Kudo, and K. Domen, “A front-illuminated nanostructured transparent BiVO₄ photoanode for >2% efficient water splitting,” Adv. Energy Mater. 6(2), 1501645 (2016). [CrossRef]  

19. F. Wang, A. Chemseddine, F. F. Abdi, R. van de Krol, and S. P. Berglund, “Spray pyrolysis of CuBi₂O₄ photocathodes: improved solution chemistry for highly homogeneous thin films,” J. Mater. Chem. A Mater. Energy Sustain. 5(25), 12838–12847 (2017). [CrossRef]  

20. S. Oh and J. Oh, “High performance and stability of micropatterned oxide-passivated photoanodes with local catalysts for photoelectrochemical water splitting,” J. Phys. Chem. C 120(1), 133–141 (2016). [CrossRef]  

21. S. Oh, H. Song, and J. Oh, “An optically and electrochemically decoupled monolithic photoelectrochemical cell for high-performance solar-driven water splitting,” Nano Lett. 17(9), 5416–5422 (2017). [CrossRef]   [PubMed]  

22. S. Oh, J. B. Kim, J. T. Song, J. Oh, and S.-H. Kim, “Atomic layer deposited molybdenum disulfide on Si photocathodes for highly efficient photoelectrochemical water reduction reaction,” J. Mater. Chem. A Mater. Energy Sustain. 5(7), 3304–3310 (2017). [CrossRef]  

23. J. T. Song, H. Ryoo, M. Cho, J. Kim, J.-G. Kim, S.-Y. Chung, and J. Oh, “Nanoporous Au thin films on Si photoelectrodes for selective and efficient photoelectrochemical CO₂ reduction,” Adv. Energy Mater. 7(3), 1601103 (2017). [CrossRef]  

24. Y. H. Rho and K. Kanamura, “Li⁺ ion diffusion in Li₄Ti₅O₁₂ thin film electrode prepared by PVP sol-gel method,” J. Solid State Chem. 177(6), 2094–2100 (2004). [CrossRef]  

25. H. Kozuka and M. Kajimura, “Single-step dip coating of crack-free BaTiO₃ films >1 μm thick: effect of poly(vinylpyrrolidone) on critical thickness,” J. Am. Ceram. Soc. 83(5), 1056–1062 (2000). [CrossRef]  

26. H. Kozuka, S. Takenaka, H. Tokita, and M. Okubayashi, “PVP-assisted sol-gel deposition of single layer ferroelectric thin films over submicron or micron in thickness,” J. Eur. Ceram. Soc. 24(6), 1585–1588 (2004). [CrossRef]  

27. S. Chao, B. Ma, S. Liu, M. Narayanan, and U. Balachandran, “Effects of pyrolysis conditions on dielectric properties of PLZT films derived from a polyvinylpyrrolidone-modified sol-gel process,” Mater. Res. Bull. 47(3), 907–911 (2012). [CrossRef]  

28. H. Kozuka, S. Takenaka, H. Tokita, T. Hirano, Y. Higashi, and T. Hamatani, “Stress and cracks in gel-derived ceramic coatings and thick film formation,” J. Sol-Gel Sci. Technol. 26(1), 681–686 (2003). [CrossRef]  

29. K. Sivaiah, K. N. Kumar, V. Naresh, and S. Buddhudu, “Structural and optical properties of Li⁺: PVP & Ag⁺: PVP polymer films,” Mater. Sci. Appl. 2(11), 1688–1696 (2011). [CrossRef]  

30. P. Qiu and C. Mao, “Viscosity gradient as a novel mechanism for the centrifugation-based separation of nanoparticles,” Adv. Mater. 23(42), 4880–4885 (2011). [CrossRef]   [PubMed]  

31. H. Kozuka, M. Kajimura, T. Hirano, and K. Katayama, “Crack-free, thick ceramic coating films via non-repetitive dip-coating using polyvinylpyrrolidone as stress-relaxing agent,” J. Sol-Gel Sci. Technol. 19(1-3), 205–209 (2000). [CrossRef]  

32. M. Sivakumar, S. Kanagesan, V. Umapathy, R. Suresh Babu, and S. Nithiyanantham, “Study of CoFe₂O₄ particles synthesized with various concentrations of PVP polymer,” J. Supercond. Nov. Magn. 26(3), 725–731 (2013). [CrossRef]  

33. M. K. Hossain, G. F. Samu, K. H. Gandha, S. Santhanagopalan, J. P. Liu, C. Janáky, and K. Rajeshwar, “Solution combustion synthesis, characterization, and photocatalytic activity of CuBi₂O₄ and its nanocomposites with CuO and α-Bi₂O₃,” J. Phys. Chem. C 121(15), 8252–8261 (2017). [CrossRef]  

34. Z. Chen, T. F. Jaramillo, T. G. Deutsch, A. Kleiman-Shwarsctein, A. J. Forman, N. Gaillard, R. Garland, K. Takanabe, C. Heske, M. Sunkara, E. W. McFarland, K. Domen, E. L. Miller, J. A. Turner, and H. N. Dinh, “Accelerating materials development for photoelectrochemical hydrogen production: standards for methods, definitions, and reporting protocols,” J. Mater. Res. 25(1), 3–16 (2010). [CrossRef]  

35. A. Yamano and H. Kozuka, “Effects of the heat-treatment conditions on the crystallographic orientation of Pb(Zr,Ti)O₃ thin films prepared by polyvinylpyrrolidone-assisted sol-gel method,” J. Am. Ceram. Soc. 90(12), 3882–3889 (2007).

36. Y. Cheng, B. Zou, J. Yang, C. Wang, Y. Liu, X. Fan, L. Zhu, Y. Wang, H. Ma, and X. Cao, “Fabrication of CoFe₂O₄ hollow fibers by direct annealing of the electrospun composite fibers and their magnetic properties,” CrystEngComm 13(7), 2268–2272 (2011). [CrossRef]  

37. Y. K. Du, P. Yang, Z. G. Mou, N. P. Hua, and L. Jiang, “Thermal decomposition behaviors of PVP coated on platinum nanoparticles,” J. Appl. Polym. Sci. 99(1), 23–26 (2006). [CrossRef]  

38. Y. Borodko, S. E. Habas, M. Koebel, P. Yang, H. Frei, and G. A. Somorjai, “Probing the interaction of poly(vinylpyrrolidone) with platinum nanocrystals by UV-Raman and FTIR,” J. Phys. Chem. B 110(46), 23052–23059 (2006). [CrossRef]   [PubMed]  

39. S. Yuvaraj, K. Karthikeyan, D. Kalpana, Y. S. Lee, and R. K. Selvan, “Surfactant-free hydrothermal synthesis of hierarchically structured spherical CuBi₂O₄ as negative electrodes for Li-ion hybrid capacitors,” J. Colloid Interface Sci. 469(1), 47–56 (2016). [CrossRef]   [PubMed]  

40. F. F. Abdi and R. Van De Krol, “Nature and light dependence of bulk recombination in Co-Pi-catalyzed BiVO₄ photoanodes,” J. Phys. Chem. C 116(17), 9398–9404 (2012). [CrossRef]  

41. U. Sim, J. Moon, J. An, J. H. Kang, S. E. Jerng, J. Moon, S.-P. Cho, B. H. Hong, and K. T. Nam, “N-doped graphene quantum sheets on silicon nanowire photocathodes for hydrogen production,” Energy Environ. Sci. 8(4), 1329–1338 (2015). [CrossRef]