Water splitting and CO₂ reduction over an AgSr₂Ta₅O₁₅ photocatalyst developed by a valence band control strategy

Abstract

Ag⁺ substitution was applied to a tungsten-bronze-type metal oxide. An AgSr₂Ta₅O₁₅ photocatalyst has emerged for water splitting and CO₂ reduction. DFT calculation and diffuse reflection spectra revealed that the Ag d-orbital formed a new valence band, leading to a narrow band gap (3.91 eV) compared to that of NaSr₂Ta₅O₁₅ (4.11 eV).

---

Photocatalytic water splitting and CO₂ reduction using water as an electron donor are attractive from the viewpoint of artificial photosynthesis.^1,2 Metal oxides have been widely studied as photocatalysts.¹ This is because most metal oxides are stable under photocatalytic reaction conditions. However, these metal oxide photocatalysts generally possess wide band gaps, resulting in limitation of their responses to irradiation lights with shorter wavelengths. In this context, many efforts have been made to narrow their band gaps.

Band engineering,³ including valence band control, doping, and solid solution formation, is an effective strategy for narrowing the band gap. Among them, the valence band control technique has indeed developed metal oxide photocatalysts with narrower band gaps. In this strategy, Cu⁺(3d¹⁰),^4–9 Ag⁺(4d¹⁰),^10–16 Sn²⁺(5s²),^17–20 Pb²⁺(6s²),^13,21 and Bi³⁺(6s²)^13,22–25 make new valence bands at shallower levels than O2p-based valence bands. Among the photocatalysts developed by this strategy, AgTaO₃¹⁰ and Na_0.5Bi_0.5TiO₃²⁴ photocatalysts with perovskite structures efficiently split water to H₂ and O₂ in stoichiometric amounts. However, the numbers of valence-band-controlled metal oxide photocatalysts for water splitting in a one-photon excitation mechanism are still limited to mainly the typical perovskites, resulting in a few crystal structures being useful for the valence band control strategy.

A tungsten-bronze-type crystal structure is one of the families of perovskite materials which are generally denoted as the chemical formula ABO₃. A notable feature in the crystal structure of tungsten bronze is that it is easy to replace A and A′ site cations of its chemical formula (AA′₂M₅O₁₅) with various cations. For example, the A site can contain Na⁺ and K⁺ cations, and the A′ sites can contain Ca²⁺, Sr²⁺, and Ba²⁺ cations, whereas the M sites can contain Nb⁵⁺ and Ta⁵⁺ cations. To date, it has been reported that KCaSrTa₅O₁₅^26,27 and KSr₂Ta₅O₁₅²⁸ show photocatalytic activities for water splitting and CO₂ reduction using water as an electron donor. A series of AA′₂Ta₅O₁₅ (A = K, Na; A′ = Sr, Ba)^29,30 and K₂RETa₅O₁₅ (RE = rare earth metal)³¹ are also tungsten-bronze-type photocatalysts. Therefore, these materials are expected to contribute to expanding the crystal structures for a valence band controlled photocatalyst.

In this study, we successfully prepared AgSr₂Ta₅O₁₅ with a tungsten-bronze-type structure and compared its band structure to that of NaSr₂Ta₅O₁₅ based on diffuse reflectance spectra and density functional theory (DFT) calculation. Moreover, the photocatalytic property of AgSr₂Ta₅O₁₅ for water splitting and CO₂ reduction was evaluated.

NaSr₂Ta₅O₁₅ was prepared by a polymerized complex method (PC method) that usually gives high photocatalytic performances compared with a conventional solid-state reaction.²⁷ AgSr₂Ta₅O₁₅ was prepared by a solid-state reaction because it would not be easy to prepare it by the PC method due to the isolation of metallic Ag. The obtained AgSr₂Ta₅O₁₅ was treated with an aqueous HNO₃ solution, if necessary. Their crystal phases were identified by X-ray diffraction. Their diffuse reflectance spectra were obtained using the Kubelka–Munk method. Particle shapes of AgSr₂Ta₅O₁₅ and NaSr₂Ta₅O₁₅ were observed using a scanning electron microscope. Photocatalytic CO₂ reduction and water splitting were conducted using a gas-flow system equipped with inner irradiation cells made of quartz and Pyrex, and a 400 W high-pressure mercury lamp. Density functional theory (DFT) calculations were performed using the CASTEP code to obtain band structures of AgSr₂Ta₅O₁₅ and NaSr₂Ta₅O₁₅. The detailed information of these experiments is summarized in the supporting information (including Fig. S1–S3 and Table S1, ESI†).

Fig. 1(A) shows X-ray diffraction patterns (XRDs) of AgSr₂Ta₅O₁₅ (not treated with an aqueous HNO₃ solution; denoted with “w/o HNO₃”) and NaSr₂Ta₅O₁₅. The diffraction peaks of both samples were well consistent with those of a powder diffraction file (PDF) of NaSr₂Ta₅O₁₅. This is because of the resembled ionic radii of Ag⁺ (115 pm) and Na⁺ (102 pm) in the 6 coordination number. For AgSr₂Ta₅O₁₅ (w/o HNO₃), AgTaO₃ and metallic Ag were not observed in the XRD pattern (Fig. S4, ESI†). Scanning electron microscopy (SEM) was performed to observe AgSr₂Ta₅O₁₅ and NaSr₂Ta₅O₁₅ particles (Fig. 1(B) and Fig. S5, ESI†). Sintered particles with a size of several hundred nm were observed. Diffuse reflectance spectra (DRS) of AgSr₂Ta₅O₁₅ and NaSr₂Ta₅O are shown in Fig. 1(C). The absorption spectrum red-shifted by replacing Na⁺ with Ag⁺, resulting in their band gaps of 3.91 eV for AgSr₂Ta₅O₁₅ and 4.11 eV for NaSr₂Ta₅O₁₅. The reason why the band gap was narrowed by containing Ag⁺ is due to a new valence band formed by the hybridization of O2p and Ag4d orbitals. These physicochemical analyses indicate that AgSr₂Ta₅O₁₅ of a tungsten-bronze-type crystal structure was successfully formed, and that its band gap was narrowed by Ag⁺ substitution as seen for NaTaO₃ and AgTaO₃.¹⁰ Density functional theory (DFT) calculation was performed to discuss the band structures of AgSr₂Ta₅O₁₅ and NaSr₂Ta₅O₁₅, as shown in Fig. 2. Conduction band minima of both samples were mainly formed by Ta d-orbital and O p-orbital, whereas both valence band maxima contained O p-orbital. Most importantly, the Ag d-orbital took part in making the valence band of AgSr₂Ta₅O₁₅, whereas the Na s-orbital did not make such a valence band. This is well consistent with the band structure of the AgTaO₃ photocatalyst.¹⁰ Thus, Ag⁺ substitution is effective to narrow the band gap of the tungsten-bronze-type metal oxide by forming the new valence band. The calculated band gaps of AgSr₂Ta₅O₁₅ and NaSr₂Ta₅O₁₅ were 3.25 and 3.28 eV, respectively. These values are small as compared to those experimentally obtained from DRS, even though HSE06 was used. This might be due to deviation of the optimized structures in the DFT calculation from the experimentally obtained samples; e.g., structural isomerism (Fig. S1 and S2, ESI†).^32–34

Fig. 1 (A) XRD patterns of (a) AgSr₂Ta₅O₁₅ (w/o HNO₃) and (b) NaSr₂Ta₅O₁₅, and PDF of (c) NaSr₂Ta₅O₁₅ (PDF No.: 39-1439). (B) SEM image of AgSr₂Ta₅O₁₅ (w/o HNO₃). (C) DRS of (d) AgSr₂Ta₅O₁₅ (w/o HNO₃) and (e) NaSr₂Ta₅O₁₅.

Fig. 2 DOS of (a) AgSr₂Ta₅O₁₅ and (b) NaSr₂Ta₅O₁₅. The dotted lines indicate their Fermi levels.

Photocatalytic performances of AgSr₂Ta₅O₁₅ and NaSr₂Ta₅O₁₅ were evaluated, as shown in Table 1. NaSr₂Ta₅O₁₅ with NiO and Ag cocatalysts showed higher photocatalytic activities for water splitting and CO₂ reduction than AgSr₂Ta₅O₁₅ (entries 1, 2, 5, 7). This is probably due to the enhancement of recombination between photogenerated e⁻ and h⁺ at a silver site, more or less. Next, let us see AgSr₂Ta₅O₁₅ newly obtained in the present study.

Table 1 Photocatalytic water splitting and CO₂ reduction over AgSr₂Ta₅O₁₅

Entry	Photocatalyst	HNO3 treatment	Cocatalyst (wt%)	Gas	NaHCO3 addition	Activity/μmol h^−1			SelCO %
						H2	O2	CO
Photocatalyst: 0.5 g, Reactant solution: water (350 mL), Reactor: a gas-flow system with an inner irradiation cell made of quartz, Light source: a 400 W high-pressure mercury lamp, Concentration of the aqueous NaHCO3 solution: 0.1 mol L^−1. Selectivity of CO formation (SelCO) was estimated as follows; SelCO (%) = (the CO formation rate)/[(the H2 formation rate) + (the CO formation rate)] × 100.
1	NaSr2Ta5O15	No	NiO(0.2)	Ar	No	1365	714	—	—
2	NaSr2Ta5O15	No	Ag(0.5)	CO2	Yes	25	29	29	54
3	AgSr2Ta5O15	No	—	Ar	No	80	34	0	0
4	AgSr2Ta5O15	No	NiO(0.2)	Ar	No	197	76	—	—
5	AgSr2Ta5O15	Yes	NiO(0.2)	Ar	No	457	204	—	—
6	AgSr2Ta5O15	Yes	Ag(0.5)	CO2	No	25	11	2.5	9.1
7	AgSr2Ta5O15	Yes	Ag(0.5)	CO2	Yes	17	12	13	43
8	AgSr2Ta5O15	Yes	Ag(3)	CO2	No	15	5.2	6.2	29
9	AgSr2Ta5O15	Yes	Ag(3)	CO2	Yes	6.2	9.3	21	77

---

Bare AgSr₂Ta₅O₁₅ split water into H₂ and O₂ (entry 3). Its performance was improved by loading NiO, which is a typical cocatalyst for water splitting (entry 4).³ Considering the literature about AgTaO₃,¹⁰ there is a possibility that Ag nanoparticles are segregated on the surface. Therefore, AgSr₂Ta₅O₁₅ was treated with an aqueous HNO₃ solution to remove the Ag nanoparticles, resulting in an improvement in the water splitting performance (entry 5). The treated AgSr₂Ta₅O₁₅ photocatalyst was applied to CO₂ reduction using water as an electron donor (entries 6–9). We have reported that an Ag cocatalyst loaded by impregnation works as an active site for CO₂ reduction to form CO.^2,35–37 The Ag-impregnated AgSr₂Ta₅O₁₅ produced not only H₂ and O₂ but also CO under CO₂ gas conditions. Because bare AgSr₂Ta₅O₁₅ did not give CO under Ar (entry 3), we can conclude that the evolved CO was generated from not contamination of carbonous species over the photocatalyst but CO₂. When the amount of Ag cocatalyst was increased, the CO formation rate and its selectivity were improved. The addition of NaHCO₃ was effective for CO₂ reduction.

Fig. 3 shows photocatalytic CO₂ reduction over Ag(3 wt%)-loaded AgSr₂Ta₅O₁₅ (Table 1 entries 8, 9). H₂, CO, and O₂ were formed from water dissolved with CO₂ gas (1 atm). The activity was improved by adding NaHCO₃. It is reported that the hydrogen carbonate enhances O₂ production in water splitting over a Pt/TiO₂ photocatalyst.³⁸ The selectivity for CO formation (Sel_CO) was also drastically improved up to 77%. This is because of the smooth supply of CO₂ molecules onto the active site due to a chemical equilibrium of the hydrogen carbonate.^2,36 The ratios of H₂/O₂ and (H₂ + CO)/O₂ were slightly beyond stoichiometry (i.e., beyond two). According to the literature about AgTaO₃,¹⁰ a part of the evolved O₂ might be adsorbed on the AgTaO₃ photocatalyst, resulting in the O₂ evolution being slightly less than the stoichiometry in an early period for water splitting. For the present CO₂ reduction over Ag/AgSr₂Ta₅O₁₅, the rates of gas evolution became gradually close to stoichiometry with the reaction time (Fig. S6, ESI†). Therefore, there is a possibility that the evolved O₂ was partly adsorbed on AgSr₂Ta₅O₁₅.

Fig. 3 CO₂ reduction using water as an electron donor over the Ag (3 wt%)-loaded AgSr₂Ta₅O₁₅ photocatalyst. Photocatalyst: 0.5 g, Reactant solution: water (350 mL), Reactor: a gas-flow system with an inner irradiation cell made of quartz, Light source: a 400 W high-pressure mercury lamp, Concentration of the aqueous NaHCO₃ solution: 0.1 mol L⁻¹.

Fig. 4 shows the effect of the wavelength of the irradiation light on photocatalytic CO₂ reduction over AgSr₂Ta₅O₁₅ without HNO₃ treatment (w/o HNO₃). AgSr₂Ta₅O₁₅ (w/o HNO₃) was active for CO₂ reduction even without additional Ag cocatalyst loading (1st run). The CO formation rate and its selectivity were improved by adding NaHCO₃ (2nd run). This result implies that a small amount of Ag nanoparticles would be segregated on the surface of AgSr₂Ta₅O₁₅ prepared by a solid-state reaction as seen in AgTaO₃, though it was not detected by XRD and DRS. The segregated Ag worked as the active site for CO₂ reduction. Additionally, it is noteworthy that the AgSr₂Ta₅O₁₅ (w/o HNO₃) produced H₂, CO, and O₂ under irradiation of light with wavelength longer than 300 nm. This is a feature of the AgSr₂Ta₅O₁₅ possessing a red-shifted absorption edge as shown in Fig. 1(C) being different from NaSr₂Ta₅O₁₅. AgSr₂Ta₅O₁₅ is superior to NaSr₂Ta₅O₁₅ in this photoresponse. Thus, AgSr₂Ta₅O₁₅ (w/o HNO₃) is a photocatalyst that is active not only for water splitting but also for CO₂ reduction using water as an electron donor.

Fig. 4 CO₂ reduction using water as an electron donor over the AgSr₂Ta₅O₁₅ (w/o HNO₃) photocatalyst without any additional cocatalyst loading. Photocatalyst: 0.5 g, Reactant solution: water (350 mL), Reactor: a gas-flow system with an inner irradiation cell made from quartz (λ > 250 nm) or Pyrex (λ > 300 nm), Light source: a 400 W high-pressure mercury lamp, Concentration of an aqueous NaHCO₃ solution: 0.1 mol L⁻¹.

In conclusion, AgSr₂Ta₅O₁₅ was prepared by a conventional solid-state reaction based on a valence band control strategy using Ag⁺ substitution. Its crystal structure was almost the same as that of NaSr₂Ta₅O₁₅ due to the resembled ionic radii of Ag⁺ and Na⁺. This Ag⁺ substitution resulted in red-shift of the absorption spectrum from that of NaSr₂Ta₅O₁₅. The band gap of AgSr₂Ta₅O₁₅ was 3.91 eV, while that of NaSr₂Ta₅O₁₅ was 4.11 eV. DOS estimated by DFT calculations revealed that the Ag d-orbital took part in making the new valence band. Furthermore, AgSr₂Ta₅O₁₅ was a new photocatalyst for water splitting and CO₂ reduction. This study demonstrated the usefulness of the valence band control strategy for the development of new metal oxide photocatalysts, resulting in the AgSr₂Ta₅O₁₅ photocatalyst with a tungsten-bronze type crystal structure. This finding will contribute to liberating the valence band control strategy from the material design space mainly limited in typical perovskite-type metal oxides for the development of active photocatalysts for water splitting and CO₂ reduction.

This work was supported by JSPS KAKENHI, Grants-in-Aid for Scientific Research (A) 23H00248 and Grant Numbers 17H06433 and 17H06440 in Scientific Research on Innovative Areas “Innovations for Light-Energy Conversion (I⁴LEC)”.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. Q. Wang and K. Domen, Chem. Rev., 2020, 120, 919.
2. S. Yoshino, T. Takayama, Y. Yamaguchi, A. Iwase and A. Kudo, Acc. Chem. Res., 2022, 55, 966.
3. A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253.
4. H. Kato, A. Takeda, M. Kobayashi, M. Hara and M. Kakihana, Catal. Sci. Technol., 2013, 3, 3147.
5. K. Iwashina, A. Iwase and A. Kudo, Chem. Sci., 2015, 6, 687.
6. H. Kato, T. Fujisawa, M. Kobayashi and M. Kakihana, Chem. Lett., 2015, 44, 973–975.
7. K. Iwashina, A. Iwase, S. Nozawa, S. Adachi and A. Kudo, Chem. Mater., 2016, 28, 4677.
8. I. Sullivan, B. Zoellner and P. A. Maggard, Chem. Mater., 2016, 28, 5999.
9. B. Zoellner, S. Stuart, C.-C. Chung, D. B. Dougherty, J. Jones and P. A. Maggard, J. Mater. Chem. A, 2016, 4, 3115.
10. H. Kato, H. Kobayashi and A. Kudo, J. Phys. Chem. B, 2002, 106, 12441.
11. Y. Hosogi, H. Kato and A. Kudo, J. Mater. Chem., 2008, 18, 647.
12. J. Boltersdorf and P. A. Maggard, ACS Catal., 2013, 3, 2547.
13. J. Boltersdorf, T. Wong and P. A. Maggard, ACS Catal., 2013, 3, 2943.
14. H. Horie, A. Iwase and A. Kudo, ACS Appl. Mater. Interfaces, 2015, 7, 14638.
15. S. Zong, C. Cheng, J. Shi, Z. Huang, Y. Hu, H. Yang and L. Guo, Chem. – Asian J., 2017, 12, 882.
16. K. Watanabe, A. Iwase, S. Nozawa, S. Adachi and A. Kudo, ACS Sustainable Chem. Eng., 2019, 7, 9881.
17. Y. Hosogi, H. Kato and A. Kudo, J. Phys. Chem. C, 2008, 112, 17678.
18. Y. Hosogi, Y. Shimodaira, H. Kato, H. Kobayashi and A. Kudo, Chem. Mater., 2008, 20, 1299.
19. J. Boltersdorf, I. Sullivan, T. L. Shelton, Z. Wu, M. Gray, B. Zoellner, F. E. Osterloh and P. A. Maggard, Chem. Mater., 2016, 28, 8876.
20. D. Noureldine and K. Takanabe, Catal. Sci. Technol., 2016, 6, 7656.
21. Y. Shimodaira, H. Kato, H. Kobayashi and A. Kudo, Bull. Chem. Soc. Jpn., 2007, 80, 885.
22. A. Kudo, K. Omori and H. Kato, J. Am. Chem. Soc., 1999, 121, 11459.
23. H. Fujito, H. Kunioku, D. Kato, H. Suzuki, M. Higashi, H. Kageyama and R. Abe, J. Am. Chem. Soc., 2016, 138, 2082.
24. K. Watanabe, Y. Iikubo, Y. Yamaguchi and A. Kudo, Chem. Commun., 2021, 57, 323.
25. D. Ozaki, H. Suzuki, O. Tomita, Y. Inaguma, K. Nakashima, H. Kageyama and R. Abe, J. Photochem. Photobiol., A, 2021, 408, 113095.
26. T. Takayama, K. Tanabe, K. Saito, A. Iwase and A. Kudo, Phys. Chem. Chem. Phys., 2014, 16, 24417.
27. T. Takayama, A. Iwase and A. Kudo, Bull. Chem. Soc. Jpn., 2015, 88, 538.
28. Z. Huang, K. Teramura, S. Hosokawa and T. Tanaka, Appl. Catal., B, 2016, 199, 272.
29. Z. Huang, S. Yoshizawa, K. Teramura, H. Asakura, S. Hosokawa and T. Tanaka, ACS Omega, 2017, 2, 8187.
30. Z. Huang, S. Yoshizawa, K. Teramura, H. Asakura, S. Hosokawa and T. Tanaka, ACS Sustainable Chem. Eng., 2018, 6, 8247.
31. Z. Huang, K. Teramura, H. Asakura, S. Hosokawa and T. Tanaka, Catal. Today, 2018, 300, 173.
32. A. Kubo, G. Giorgi and K. Yamashita, J. Phys. Chem. C, 2017, 121, 27813.
33. A. Kubo, G. Giorgi and K. Yamashita, Chem. Mater., 2017, 29, 539.
34. M. Kaneko, M. Fujii, T. Hisatomi, K. Yamashita and K. Domen, J. Energy Chem., 2019, 36, 7.
35. K. Iizuka, T. Wato, Y. Miseki, K. Saito and A. Kudo, J. Am. Chem. Soc., 2011, 133, 20863.
36. H. Nakanishi, K. Iizuka, T. Takayama, A. Iwase and A. Kudo, ChemSusChem, 2017, 10, 112.
37. T. Takayama, H. Nakanishi, M. Matsui, A. Iwase and A. Kudo, J. Photochem. Photobiol., A, 2018, 358, 416.
38. K. Sayama and H. Arakawa, J. Chem. Soc., Faraday Trans., 1997, 93, 1647.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3cc01481a

‡ Present address: His current affiliation is Graduate School of Science and Technology, Division of Materials Science, Nara Institute of Science and Technology, 8916-5 Takayama, Ikoma, Nara 630-0192, Japan.

§ Present address: His current affiliation is Department of Applied Chemistry, School of Science and Technology, Meiji University, Kanagawa 214-8571, Japan.

---