WO₃/BiVO₄ photoanode coated with mesoporous Al₂O₃ layer for oxidative production of hydrogen peroxide from water with high selectivity

Abstract

A WO₃/BiVO₄ photoanode coated with various metal oxides demonstrated high selectivity (faradaic efficiency) for hydrogen peroxide (H₂O₂) generation from water (H₂O) under irradiation of simulated solar light in a highly concentrated hydrogen carbonate (KHCO₃) aqueous solution. A mesoporous and amorphous aluminium oxide (Al₂O₃) layer significantly facilitated inhibition of the oxidative degradation of generated H₂O₂ into oxygen (O₂) on the photoanode, resulting in unprecedented H₂O₂ selectivity (ca. 80%) and the accumulation (>2500 μM at 50C).

---

Chemical conversions using light energy have been performed in various fields since the discovery of the Honda–Fujishima effect.^1–25 Significant efforts have recently been devoted to H₂ production by water splitting using inexhaustible light for clean energy conversion processes.^1–4,8–29 Photoelectrode systems are widely recognised as a promising technology for H₂ production because they operate at an electrolysis voltage lower than the theoretical electrolysis voltage of water (<1.23 V).^1,8–29 Visible light-responsive oxide photoanodes with a narrow bandgap energy, such as WO₃, BiVO₄ and Fe₂O₃, are desirable for the efficient utilisation of solar light and economical synthetic processes.^8–29 Most importantly, numerous efforts have been focused on BiVO₄ photoanodes capable of utilising a wide range of light energy (∼520 nm) and achieving efficient O₂ generation by water splitting.^{9–21,28,29,32} A WO₃/BiVO₄ photoanode that combines BiVO₄ with a WO₃ underlayer for the efficient transfer of excited electrons on BiVO₄ to the F-doped SnO₂ conductive glass (FTO) substrate shows exceptional photoelectrochemical performance for water splitting into H₂ and O₂.^{10–13,17–20,28,29,32} However, most previous investigations, containing electrochemical reactions, focused solely on the recovery of H₂ energy generated on the cathode and little attention was paid to the recovery of the oxidation products simultaneously evolved during water splitting.^24–31

H₂O₂ is an especially versatile and clean oxidation product having the potential to generate instead of O₂ from H₂O (eqn (1)).

2H₂O → H₂O₂ + 2H⁺ + 2e⁻ (E(H₂O₂/H₂O) = +1.77 V vs. RHE) (1)

However, the accumulation of H₂O₂, generated oxidatively is extremely difficult because degradation of H₂O₂ into O₂ also occurs easily and oxidatively in a conventional photoelectrochemical system i.e. the redox potential of H₂O₂ degradation is more negative than the redox potential of H₂O₂ production from H₂O (eqn (1) and (2)), resulting in low selectivity for oxidative H₂O₂ generation.

H₂O₂ → O₂ + 2H⁺ + 2e⁻ (E(O₂/H₂O₂) = +0.68 V vs. RHE) (2)

Recently, we reported that a photoelectrochemical system combining the WO₃/BiVO₄ photoanode and aqueous electrolyte of KHCO₃ under CO₂ bubbling could achieve simultaneous generation and accumulation of H₂O₂ and H₂ from H₂O (eqn (3)).^28,29 In this system, the aqueous electrolyte of KHCO₃ acts as an excellent oxidative catalyst for generating H₂O₂ from H₂O. Moreover, H₂O₂ could be produced at no external bias on both a WO₃/BiVO₄ photoanode (from H₂O) and an Au cathode (from O₂) via a two-photon process (eqn (4)).²⁹

2H₂O → H₂O₂ + H₂ (two-photon process) (3)

2H₂O + O₂ → 2H₂O₂ (two-photon process) (4)

Although the selectivity (faradaic efficiency: η(H₂O₂)) of reductive H₂O₂ production from O₂ on cathodes such as Au was very high, almost 100%, the maximum selectivity (η(H₂O₂)) for oxidative H₂O₂ production on WO₃/BiVO₄ photoanodes was still low, only ca. 54%. The design of novel photoanodes capable of achieving efficient H₂O₂ generation and inhibiting oxidative degradation of generated H₂O₂ is absolutely imperative for building a clean and breakthrough technology, by accumulating H₂O₂ and H₂ with unprecedented H₂O₂ selectivity using only H₂O as the raw material.

Here, we focused on a surface modification of the metal oxide (MeO_x) layers on the WO₃/BiVO₄ photoanode surface to achieve excellent selectivity of generation and accumulation of H₂O₂ in the KHCO₃ aqueous solution under simulated solar light irradiation (Fig. 1). The MeO_x layers were prepared by spin-coating of metal organic solutions and calcination. Introducing a porous Al₂O₃ layer was found to specifically permit oxidative H₂O₂ generation and accumulation with exceptional selectivity in an aqueous KHCO₃ electrolyte because of the blocking effect of oxidative degradation of the generated H₂O₂ into O₂ on the photoanode.

Fig. 1 Pattern and energy diagrams for photoelectrochemical H₂O₂ generation from H₂O on WO₃/BiVO₄/MeO_x photoanodes under solar light irradiation.

Details regarding experimental procedures for preparation and photoelectrochemical reactions of photoanodes are provided in the ESI†.

The effects of MeO_x layers, modified on the WO₃/BiVO₄ photoanode, for oxidative H₂O₂ generation properties were investigated at an applied electric charge of 0.9C (900 s at steady photocurrent of 1 mA) in a 0.5 M KHCO₃ aqueous electrolyte. As shown in Fig. 2, all MeO_x-coated photoanodes, except CoO_x, enhanced the oxidative H₂O₂ generation compared to a bare WO₃/BiVO₄ photoanode, and the enhanced effect, ranked by the modified metal oxide, was Al₂O₃ > ZrO₂ > TiO₂ > SiO₂ ≫ CoO_x. Little H₂O₂ was observed on the CoO_x coated photoanode, because CoO_x probably decomposed the generated H₂O₂ quickly, or O₂ may be evolved on CoO_x directly. It should be noted that the Al₂O₃ modification on the WO₃/BiVO₄ photoanode achieved roughly twice the oxidative H₂O₂ generation compared to the bare WO₃/BiVO₄ photoanode. The Al₂O₃ uniformly, smoothly and flatly covered the entire area of the WO₃/BiVO₄ photoanode as shown in the SEM images (Fig. 3), whereas other MeO_x were granularly and uniformly supported on that and possessed any crack holes (Fig. S1; ESI†). It was also confirmed, from XRD measurement (Fig. S2; ESI†), that no diffraction peaks derived from MeO_x were observed in all WO₃/BiVO₄/MeO_x photoanodes, suggesting that all tried MeO_x modified on WO₃/BiVO₄ photoanode possess amorphous-like structure. As shown in Fig. S3; ESI,† little change of the light harvesting efficiency (LHE) was also confirmed in tried all photoanodes, suggesting that these MeO_x introduced on the WO₃/BiVO₄ have little effect to light absorption efficiency on WO₃/BiVO₄ photoanode. The time courses of voltages applied between photoanode and a counter electrode of Pt mesh at steady photocurrent of 1 mA (Fig. 2) in oxidative H₂O₂ generation reaction were also confirmed (Fig. S4; ESI†). The voltages for applying steady photocurrent of 1 mA slightly increased by introducing MeO_x on the WO₃/BiVO₄ photoanode. In particular, WO₃/BiVO₄/Al₂O₃ photoanode, coated uniformly, smoothly and flatly at Al₂O₃ compared to other MeO_x, required highest applied voltage. In order to confirm the effect introducing the Al₂O₃ on the photoanode in more detail, the photocurrent property of the WO₃/BiVO₄/Al₂O₃ photoanode was investigated in a 0.5 M KHCO₃ aqueous solution (Fig. S5; ESI†). The bare WO₃/BiVO₄ photoanode exhibited excellent photocurrent property in all applied voltage ranges as with our past reported example,^11,12,28,29 and the photocurrent property slightly decreased by introducing the Al₂O₃ layer. However, it should be noted that the decreasing degree of the photocurrent property was slight, only ca. 9% and 5% at +1.23 V and +1.77 V vs. RHE, respectively, although the Al₂O₃, having an insulation property, covered the entire area of the WO₃/BiVO₄ photoanode. A similar phenomenon has also been observed in O₂ and H₂ generation through water splitting on a photoanode coated with amorphous-like Ta₂O₅.³² In addition, it was confirmed, from the N₂ absorption and desorption measurement of MeO_x particles (Fig. S6; ESI†), that almost all MeO_x possess mesoporous structure at a pore size of ca. 4–20 nm. In particular, a pore size of the Al₂O₃ was ca. 4.7 nm. The thicknesses of Al₂O₃ calculated from the coating amount on the WO₃/BiVO₄ photoanode by XRF measurement were ca. 100 nm (0.055 mg cm⁻²). In order to also investigate the effects of dense Al₂O₃ on WO₃/BiVO₄ on the oxidative H₂O₂ generation, increasing Al₂O₃ amount on WO₃/BiVO₄ photoanode was performed by decreasing the spin coating number (500 rpm) of precursor solution of EMOD solved in butyl acetate containing ethylcellulose when introducing Al₂O₃ layers. The thickness of Al₂O₃ introduced at 500 rpm calculated from the XRF measurement was ca. 127 nm (0.070 mg cm⁻²), suggested that the thickness increases with decreasing the spin coating number. As shown in Fig. S7; ESI,† little change of the H₂O₂ generation amounts was observed on these WO₃/BiVO₄/Al₂O₃ photoanodes prepared at 500 and 1000 rpm, indicating that increasing Al₂O₃ on WO₃/BiVO₄ photoanode has little effect on the oxidative H₂O₂ generation. In subsequent experiments, WO₃/BiVO₄/Al₂O₃ photoanode prepared at 1000 rpm was utilized as the photoanode. These results indicate that the specific effect enhancing oxidative H₂O₂ generation property was achieved on the WO₃/BiVO₄ though the mesoporous and amorphous Al₂O₃ layer covered uniformly, smoothly and flatly the entire area.

Fig. 2 Oxidative H₂O₂ generation on photoanodes (WO₃/BiVO₄/MeO_x) modified various metal oxides on a WO₃/BiVO₄ at an electric charge of 0.9C (900 s at steady photocurrent of 1 mA) in a 0.5 M KHCO₃ aqueous electrolyte (35 mL) in an ice bath (below 5 °C) under simulated solar light.

Fig. 3 SEM images of (A) WO₃/BiVO₄ and (B) WO₃/BiVO₄/Al₂O₃ photoanodes.

To track the specific performance enhancing effect of generating H₂O₂ by introducing the Al₂O₃ layer, the concentration dependency of KHCO₃ aqueous electrolytes on the oxidative H₂O₂ generation property was investigated at an applied electric charge of 0.9C (Fig. 4(A)). We have already reported that the oxidative H₂O₂ generation property on the WO₃/BiVO₄ photoanode was improved with increasing concentration of KHCO₃, which acts as an effective catalyst for H₂O₂ generation via the two-electron oxidation of H₂O.²⁸ Even in the case of using the WO₃/BiVO₄/Al₂O₃ photoanode, the selectivity (η(H₂O₂)) for H₂O₂ generation was significantly enhanced with increasing concentration of KHCO₃, and the η(H₂O₂) in the 2.0 M KHCO₃ aqueous solution reached ca. 80% at 0.9C, whereas that using the bare WO₃/BiVO₄ photoanode was ca. 54%. It should be noted that the selectivity (η(H₂O₂) = ca. 53%) on the WO₃/BiVO₄/Al₂O₃ photoanode in lowly concentrated KHCO₃ (0.1 M) was comparable to that (ca. 54%) on the bare WO₃/BiVO₄ photoanode in highly concentrated KHCO₃ (2.0 M). This suggests that the Al₂O₃ could effectively be contributing to oxidative H₂O₂ generation from H₂O even in the lowly concentrated KHCO₃. Moreover, as shown in Fig. 4(B), the excellent H₂O₂ generation property on the WO₃/BiVO₄/Al₂O₃ photoanode compared to the WO₃/BiVO₄ photoanode was significantly maintained even at high electric charge up to 50C. As a result, the accumulation amount, using the WO₃/BiVO₄/Al₂O₃ photoanode, reached >2500 μM at 50C, while that using the bare WO₃/BiVO₄ photoanode was >1300 μM at 50C. The dependency of the applied voltage on the oxidative H₂O₂ generation was investigated to confirm the effect of the Al₂O₃ coating in detail (Fig. S8; ESI†). A small change in H₂O₂ generation performance was observed in all ranges of applied voltages (0.8–1.8 V), suggesting that the enhanced effect of introducing an Al₂O₃ layer is independent of the voltages applied between a photoanode as the working electrode and a Pt mesh as counter electrode using the aqueous electrolyte of the KHCO₃.

Fig. 4 (A) Oxidative H₂O₂ generation in KHCO₃ aqueous electrolytes (35 mL) of different concentrations at applied electric charges of 0.9C (900 s at steady photocurrent of 1 mA) under simulated solar light and (B) accumulation of oxidative H₂O₂ generation in a 2.0 M KHCO₃ aqueous solution (35 mL) under visible light irradiation (λ > 420 nm) using an intense Xe lamp at an applied voltage of 1.5 V in an ice bath (below 5 °C) on a (a) bare WO₃/BiVO₄ and (b) WO₃/BiVO₄/Al₂O₃ photoanodes.

Although little development with regards to highly selective H₂O₂ generation via two-photon oxidation of H₂O and accumulation using photoanodes has been reported, our method of Al₂O₃ coating on the WO₃/BiVO₄ photoanode produced tremendous improvement in selective H₂O₂ generation and accumulation from H₂O in a KHCO₃ aqueous electrolyte. It is speculated that the specific enhancement of selectivity for H₂O₂ generation on the WO₃/BiVO₄/Al₂O₃ photoanode may be caused by a blocking effect, on the mesoporous Al₂O₃ layer, that inhibits oxidative H₂O₂ degradation into O₂ on the BiVO₄. To investigate the blocking effect on the Al₂O₃ layer, a degradation property test of H₂O₂ was performed in a 2.0 M KHCO₃ aqueous solution containing H₂O₂ (550 μM) in the presence of the bare WO₃/BiVO₄ or WO₃/BiVO₄/Al₂O₃ photoanodes in presence or absence of simulated solar light irradiation in an ice bath (below 5 °C). In both cases, as shown in Fig. 5, almost all the initial amount of H₂O₂ was maintained in the dark condition, however, the H₂O₂ amount drastically decreased with irradiation by simulated solar light, suggesting that the H₂O₂ was decomposed by photocarriers (excited electrons and holes) produced on the BiVO₄. It should be noted that the H₂O₂ degradation in the presence of the WO₃/BiVO₄/Al₂O₃ photoanode was dramatically inhibited compared to the degradation in the presence of a bare WO₃/BiVO₄ photoanode. The oxidative H₂O₂ generation test was also confirmed in a 2.0 M KHCO₃ aqueous electrolyte, initially containing H₂O₂ (210 μM) on the bare WO₃/BiVO₄ and WO₃/BiVO₄/Al₂O₃ photoanodes, to track the generated H₂O₂ degradation behaviour in more detail (Fig. 6). The generated rates of H₂O₂ were reduced by the initial addition of H₂O₂ in both cases of presence or absence of Al₂O₃. However, the decreasing rate of H₂O₂ generation was significantly inhibited, from ca. 61% to ca. 39%, by introducing the Al₂O₃ layer on the WO₃/BiVO₄ photoanode. These results suggest that introducing the Al₂O₃ layer significantly contributed to the highly selective H₂O₂ generation and accumulation from H₂O, with a high photocurrent property, by a blocking effect that inhibited the oxidative degradation of generated H₂O₂. The mechanism of blocking effect is proposed that the H₂O₂ generated on the BiVO₄ in the WO₃/BiVO₄/Al₂O₃ photoanode diffuses in electrolyte of KHCO₃ aqueous solution through mesoporous of the Al₂O₃, and contact of the H₂O₂ diffused in electrolyte with the BiVO₄ covered uniformly and smoothly Al₂O₃ may be significantly inhibited compared with that with bare BiVO₄, resulting in the formation of effective inhibition of oxidative H₂O₂ degradation. Furthermore, there may be other possible mechanisms such as a blocking effect of a direct O₂ evolution site via a 4-photon process covering by Al₂O₃, or an enrichment effect resulting from the increasing KHCO₃ concentration around the photoanode based on the acid–base adsorption between HCO₃⁻ (a weak base) and the weakly acidic sites on the Al₂O₃ surface, related to the good η(H₂O₂) in lower KHCO₃ concentration, as shown in Fig. 4(A). The tracking and contribution of these other mechanisms, on the Al₂O₃ layer, is currently under investigation.

Fig. 5 Degradation properties of H₂O₂ (550 μM) initially added in a 2.0 M KHCO₃ aqueous solution in an ice bath (below 5 °C) under CO₂ bubbling and simulated solar light irradiation in the presence of a (a) WO₃/BiVO₄ and (b) WO₃/BiVO₄/Al₂O₃ photoanodes at no applied voltage.

Fig. 6 Comparison of oxidative H₂O₂ generation in a 2.0 M KHCO₃ aqueous electrolyte (a) in the absence of or (b) containing initially-added H₂O₂ (210 μM) in an ice bath (below 5 °C) on a (A) WO₃/BiVO₄ and (B) WO₃/BiVO₄/Al₂O₃ photoanodes under simulated solar light irradiation at steady photocurrent of 1 mA.

Conclusions

In summary, various metal oxides were coated onto a WO₃/BiVO₄ photoanode to enhance the selectivity (faradaic efficiency) of oxidative H₂O₂ generation, in an aqueous electrolyte of KHCO₃, from water under solar light irradiation. Among the various metal oxides, the Al₂O₃ coating, which produced a mesoporous and amorphous structure on the WO₃/BiVO₄ photoanode, achieved excellent oxidative H₂O₂ generation at a selectivity of ca. 80% and an accumulation of >2500 μM (50C). Interestingly, the Al₂O₃-coated WO₃/BiVO₄ photoanode dramatically inhibited oxidative degradation of H₂O₂ generated on the WO₃/BiVO₄ photoanode after introducing the Al₂O₃ layer. This study contributes to developing a promising design for a clean H₂O₂ production system that uses only water as the raw material under solar light irradiation. More effective dreamy H₂O₂ generation, at an excellent selectivity close to 100%, can be expected by modifying the surface-treatment technology, and it is currently under investigation.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The present work was partially supported by JSPS KAKENHI Grant Number 26810105 and the International Joint Research Program for Innovative Energy Technology. We thank Dr Etsuko Fujita (Brookhaven National Laboratory) for helpful discussions.

Notes and references

1. A. Fujishima and K. Honda, Nature, 1972, 238, 37.
2. Z. G. Zou, J. H. Ye, K. Sayama and H. Arakawa, Nature, 2001, 414, 625.
3. A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253.
4. K. Maeda, K. Teramura, D. Lu, T. Takata, N. Saito, Y. Inoue and K. Domen, Nature, 2006, 440, 295.
5. K. Fuku, K. Hashimoto and H. Kominami, Chem. Commun., 2010, 46, 5118.
6. K. Fuku, T. Kamegawa, K. Mori and H. Yamashita, Chem.–Asian J., 2012, 7, 1366.
7. K. Fuku, R. Hayashi, S. Takakura, T. Kamegawa, K. Mori and H. Yamashita, Angew. Chem., Int. Ed., 2013, 52, 7446.
8. B. D. Alexander, P. J. Kulesza, I. Rutkowska, R. Solarska and J. Augustynski, J. Mater. Chem., 2008, 18, 2298.
9. K. Sayama, A. Nomura, Z. Zou, R. Abe, Y. Abe and H. Arakawa, Chem. Commun., 2003, 2908.
10. P. Chatchai, Y. Murakami, S. Kishioka, A. Y. Nosaka and Y. Nosaka, Electrochim. Acta, 2009, 54, 1147.
11. R. Saito, Y. Miseki and K. Sayama, Chem. Commun., 2012, 48, 3833.
12. I. Fujimoto, N. Wang, R. Saito, Y. Miseki, T. Gunji and K. Sayama, Int. J. Hydrogen Energy, 2014, 39, 2454.
13. X. Shi, I. Y. Choi, K. Zhang, J. Kwon, D. Y. Kim, J. K. Lee, S. H. Oh, J. K. Kim and J. H. Park, Nat. Commun., 2014, 5, 4775.
14. T. W. Kim and K. S. Choi, Science, 2014, 343, 990.
15. Y. Liu, Y. Guo, L. T. Schelhas, M. Li and J. W. Ager III, J. Phys. Chem. C, 2016, 120, 23449.
16. L. H. Hess, J. K. Cooper, A. Loiudice, C. M. Jiang, R. Buonsanti and I. D. Sharp, Nano Energy, 2017, 34, 375.
17. J. Su, L. Guo, N. Bao and C. A. Grimes, Nano Lett., 2011, 11, 1928.
18. P. M. Rao, L. Cai, C. Liu, I. S. Cho, C. H. Lee, J. M. Weisse, P. Yang and X. Zheng, Nano Lett., 2014, 14, 1099.
19. I. Grigioni, K. G. Stamplecoskie, D. H. Jara, M. V. Dozzi, A. Oriana, G. Cerullo, P. V. Kamat and E. Selli, ACS Energy Lett., 2017, 2, 1362.
20. K. Mase, M. Yoneda, Y. Yamada and S. Fukuzumi, ACS Energy Lett., 2016, 1, 913.
21. F. M. Toma, J. K. Cooper, V. Kunzelmann, M. T. McDowell, J. Yu, D. M. Larson, N. J. Borys, C. Abelyan, J. W. Beeman, K. M. Yu, J. Yang, L. Chen, M. R. Shaner, J. Spurgeon, F. A. Houle, K. A. Persson and I. D. Sharp, Nat. Commun., 2016, 7, 12012.
22. J. Y. Kim, G. Magesh, D. H. Youn, J. W. Jang, J. Kubota, K. Domen and J. S. Lee, Sci. Rep., 2013, 3, 2681.
23. Q. Mi, A. Zhanaidarova, B. S. Brunschwig, H. B. Gray and N. S. Lewis, Energy Environ. Sci., 2012, 5, 5694.
24. J. C. Hill and K. S. Choi, J. Phys. Chem. C, 2012, 116, 7612.
25. K. Ueno and H. Misawa, NPG Asia Mater., 2013, 5, e61.
26. K. Fuku, N. Wang, Y. Miseki, T. Funaki and K. Sayama, ChemSusChem, 2015, 8, 1593.
27. H. G. Cha and K. S. Choi, Nat. Chem., 2015, 7, 328.
28. K. Fuku and K. Sayama, Chem. Commun., 2016, 52, 5406.
29. K. Fuku, Y. Miyase, Y. Miseki, T. Funaki, T. Gunji and K. Sayama, Chem.–Asian J., 2017, 12, 1111.
30. K. Fuku, Y. Miyase, Y. Miseki, T. Gunji and K. Sayama, ChemistrySelect, 2016, 1, 5721.
31. T. Shiragami, H. Nakamura, J. Matsumoto, M. Yasuda, Y. Suzuri, H. Tachibana and H. Inoue, J. Photochem. Photobiol., A, 2015, 313, 131.
32. R. Saito, Y. Miseki, W. Nini and K. Sayama, ACS Comb. Sci., 2015, 17, 592.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental section, SEM images, XRD spectra, LHE spectra, applied voltage properties, I–V characteristic of photoanodes, pore size distribution of the MeO_x particles, effect of Al₂O₃ amount on WO₃/BiVO₄ and dependence of applied voltage. See DOI: 10.1039/c7ra09693c

---