A Z-scheme photocatalyst constructed with an yttrium–tantalum oxynitride and a binuclear Ru(II) complex for visible-light CO₂ reduction

Abstract

An yttrium–tantalum oxynitride having a band gap of 2.1 eV (absorbing visible light at <580 nm) was applicable as a semiconductor component of a Z-scheme CO₂ reduction system operable under visible light, in combination with a binuclear Ru(II) complex that has strong absorption in the visible region (<600 nm). Excitation of this system with visible light under a CO₂ atmosphere induced photocatalytic formation of formic acid with very high selectivity (>99%).

---

In order to address the depletion of fossil fuels and the concomitant emission of CO₂, the chemical fixation of CO₂ as well as its conversion into useful compounds are currently important topics in the field of chemistry. Various reactions/schemes have been proposed to date for this purpose.^1–12 Among these, photocatalytic CO₂ reduction using a heterogeneous photocatalyst that consists of a semiconductor and a metal complex has great potential because it is a readily scalable process that utilizes only low-density sunlight.¹³

From the viewpoint of solar energy conversion, a semiconductor photocatalyst that harvests a wide range of visible light is highly desirable. Recently, it has been reported that certain metal oxides modified with Ag as a promoter achieved CO₂ reduction using water as the electron source.⁷ However, the band gaps of these metal oxides are too large to harvest visible light. While our group has developed heterogeneous photocatalysts for CO₂ reduction that consist of a visible-light-responsive semiconductor and a functional metal complex working as an oxidation and a reduction site, respectively, these novel semiconductors (e.g., TaON, CaTaO₂N and C₃N₄) can harvest visible light only at wavelengths shorter than 500 nm.^9–11 Unfortunately, there are very few semiconductors that have a band gap smaller than 2.5 eV (corresponding to 500 nm wavelength) and that can be applied for photocatalytic CO₂ reduction. Reducing the band gap of a semiconductor will reduce the reactivity of electrons and holes generated in the conduction and valence bands, respectively. At the same time, it will become difficult to satisfy thermodynamic requirements; that is, valence and conduction band potentials that straddle the redox potentials of a desired reaction. Because a CO₂ fixation system using metal-complex/semiconductor hybrid photocatalysts ultimately requires the use of water as the electron source, a semiconductor photocatalyst has to satisfy band edge potentials that straddle both the water oxidation and CO₂ reduction potentials.

Herein we report that an yttrium–tantalum oxynitride (YTON) having a band gap of 2.1 eV coupled with a binuclear Ru(II) complex (RuRu′, see Chart 1) works as a photocatalyst for reduction of CO₂ to formic acid under visible-light irradiation. This system works according to the Z-scheme principle; i.e., excitation of both YTON and the redox photosensitizer unit of RuRu′ with visible light induces reduction of CO₂ to HCOOH with >99% selectivity at room temperature and ambient atmospheric pressure.

Chart 1 Ruthenium(II) complexes used in this work.

The YTON powder was prepared by the conventional thermal ammonolysis of a corresponding Y–Ta oxide precursor that was previously synthesized using the polymerized complex (PC) method. The details of these two procedures are provided in the ESI†. The synthesis of single-phase pyrochlore Y₂Ta₂O₅N₂ was confirmed by X-ray diffraction (XRD) analysis (Fig. S1, ESI†). Energy-dispersive X-ray (EDX) spectroscopy and thermogravimetric (TG) analysis of the as-prepared YTON determined that the Y/Ta ratio was 0.95, and that the concentration of nitrogen in the material was 1.2 wt%, values that were below the theoretical values (Y/Ta = 1, N: 4.3 wt%), indicating a non-stoichiometric composition. TEM observations showed that the YTON consisted of 20–30 nm primary particles connected to one another to form larger secondary particles with porous structures (Fig. 1A). The specific surface area of the YTON was determined by nitrogen adsorption/desorption to be 16 m² g⁻¹, a result that can be attributed to its porous structure. As shown in Fig. 1B, the YTON had an absorption edge at approximately 580 nm, consistent with an earlier report.¹⁴ The band gap of the material was estimated to be about 2.1 eV based on the onset wavelength of its diffuse reflectance spectrum.

Fig. 1 (A) TEM image, and (B) UV-visible diffuse reflectance spectrum of the as-prepared YTON.

Li et al. have reported that yttrium–tantalum oxynitride exhibits photocatalytic activity for both water reduction and oxidation under visible light.¹⁴ However, the band-edge potentials of the material have not been reported and remain unclear. Therefore, we conducted photoelectrochemical analyses in order to investigate the band-gap structure.¹⁵ The actual band-edge positions were evaluated by measuring the onset potentials of photocurrents generated from an IrO₂/TiO₂/YTON/FTO electrode in electrolyte solutions with various pH values. In these experiments, the YTON was immobilized on the FTO by a conventional squeegee method and a TiO₂ layer was deposited to improve the inter-particle electron transfer and particle/substrate electron transfer.^15a,16 In addition, colloidal IrO₂ was deposited on the TiO₂/YTON to enhance the water oxidation.^16b

Current–voltage curves were acquired under chopped visible light irradiation (λ > 500 nm), scanning from negative to positive potential at 5 mV s⁻¹ in an aqueous Na₂SO₄ solution adjusted to pH 4.8, 8.7 or 11.7 by the addition of H₂SO₄ or NaOH. As shown in Fig. S2 (ESI†), an anodic photocurrent attributed to water oxidation was observed under all pH conditions examined, indicating that the YTON functions as an n-type semiconductor. The onset potential, corresponding to the flat-band potential (E_FB), was shifted to the negative direction with increasing pH. The energy difference between the conduction band edge potential (E_CB) and the E_FB is believed to be approximately 0.1–0.3 eV based on the n-type semiconducting character of the YTON.¹⁷ On the basis of the YTON band gap (2.1 eV), the valence band edge potential (E_VB) can be estimated to be 2.1 V more positive than E_CB. Fig. 2 presents the conduction and valence band edge potentials of the IrO₂/TiO₂/YTON/FTO electrode as a function of pH. The positions of the YTON conduction band minimum and the valence band maximum evidently vary with pH, and were shifted at a rate of approximately −17 mV per pH as the pH increased. It was thus shown that YTON had band-edge potentials suitable for water oxidation in all pH range examined. Under basic pH conditions, however, it is clear that the driving force for CO₂ reduction becomes insufficient.

Fig. 2 Dependence of conduction and valence band edge potentials for the as-prepared YTON on the pH of the electrolyte, as determined from photocurrent measurements (see Fig. S2, ESI†).

Prior to photocatalytic reaction, the YTON was modified with both Ag and RuRu′ having methylphosphonic-acid anchoring groups, which, respectively, work as a promoter for interfacial electron transfer and as another photocatalytic unit with high capability for CO₂ reduction.^10b,11 TEM observations indicated that metallic Ag nanoparticles approximately 10 nm in size were distributed on the YTON surface (Fig. 3A). The resulting lattice fringe had a period of 0.24 (Fig. 3B), consistent with the d spacing of Ag(111) planes. The successful immobilization of RuRu′ on Ag/YTON was verified by means of Fourier transform infrared (FT-IR) spectroscopy (Fig. S3, ESI†).

Fig. 3 TEM images of Ag(1.5 wt%)/YTON. Panel B shows a magnified image of the square in panel A.

Using the as-prepared YTON with Ag and/or RuRu′, CO₂ reduction was performed under visible light irradiation (λ > 400 nm). The reaction setup used in this work was identical to that we have reported previously.^10,11Table 1 summarizes the results. To avoid complications resulting from water oxidation catalysis, reactions were conducted in a mixed N,N-dimethylacetamide (DMA)/triethanolamine (TEOA) solution (4 : 1 v/v), where TEOA worked as a sacrificial electron donor.

Table 1 Results of visible-light CO₂ reduction using YTON with various modifications (λ > 400 nm)^a

Entry	Photocatalyst	Amount of HCOOH/nmol	TON	Selectivity
a Reaction conditions: photocatalyst, 8.0 mg (Ag 1.5 wt%); solution, a mixture of DMA and TEOA (4 : 1 v/v) 4.0 mL; light source, 400 W high-pressure Hg lamp (SEN) with a NaNO2 solution filter. Reaction time: 24 h. The amount of metal complex loaded in each case was 4.5 µmol g^−1. b Under an Ar atmosphere. c Without TEOA. d In the dark.
1	RuRu′/YTON	162	5	>99
2	RuRu′/Ag/YTON	630	18	>99
3	YTON	N.D.	—	—
4	Ag/YTON	N.D.	—	—
5	RuRu′ (6.0 µM)	N.D.	—	—
6^b	RuRu′/Ag/YTON	N.D.	—	—
7^c	RuRu′/Ag/YTON	N.D.	—	—
8^d	RuRu′/Ag/YTON	N.D.	—	—
9	Ru(PS)/Ag/YTON	N.D.	—	—
10	Ru(Cat)/Ag/YTON	N.D.	—	—

---

Combining YTON with RuRu′ resulted in clearly observable HCOOH production with very high selectivity (>99%) and catalytic turnover numbers (TON_HCOOH) of 5 based on the amount of loaded RuRu′ (entry 1). The Ag-modification increased the HCOOH formation (TON_HCOOH = 18, entry 2). This is attributed to the promotional effect of the Ag, which facilitates interfacial electron transfer from the semiconductor to the excited state of the binuclear Ru(II) complex,^10b,11 as discussed below. The generation of HCOOH over RuRu′/Ag/YTON was improved with increases in the amount of Ag loaded, up to 1.5 wt%, beyond which the yield began to decrease (Fig. S4, ESI†). By contrast, neither the YTON nor the Ag/YTON showed activity for the reaction (entries 3 and 4). Note that TEOA cannot reductively quench the photosensitizer unit of RuRu′, i.e., it is reasonable that only RuRu′ without YTON did not induce photocatalytic reduction of CO₂ (entry 5 in Table 1). In the absence of CO₂ or TEOA, no HCOOH production was observed (entries 6 and 7), nor was there any reaction under dark conditions (entry 8). It should be noted that neither CO nor H₂ gas evolution was observed in all cases. No C–C coupling product (e.g., ethane and ethanol) was detected as well. These results show that both YTON and RuRu′ are essential to realize HCOOH production, and that TEOA works as an electron donor to scavenge holes in the valence band of the YTON.

When using Ag/YTON modified with either a model mononuclear complex of the redox photosensitizer unit (Ru(PS)) or the catalytic model complex (Ru(Cat)), no HCOOH production could be detected (entries 9 and 10). Ru(Cat) has been shown to function as a cocatalyst for CO₂ reduction to generate HCOOH on certain semiconductors (such as C₃N₄ and CaTaO₂N),^9,10b but it was evidently not effective in conjunction with YTON, probably because of the lower conduction-band potential.

Fig. 4 presents an energy diagram for RuRu′/YTON based on the results of photoelectrochemical measurements (Fig. 2) and our previous study.^10a Upon visible light irradiation (λ > 400 nm), both the YTON and the photosensitizing unit of the RuRu′ undergo excitation. In this case, it should be noted that electron transfer from the excited state of RuRu′ (E_ox* = −1.30 V) to the conduction band of the YTON (−1.40 V) is energetically unfavourable. This energy diagram explains the observation that Ru(Cat)/YTON did not promote the photocatalysis of the CO₂ reduction (Table 1, entry 10), because the electron transfer from the conduction band of YTON (−1.40 V) to Ru(Cat) (−1.60 V) is energetically unfavourable. Conversely, the driving force for electron transfer from the conduction band of YTON to the excited state of RuRu′ (+0.17 V) is very large (i.e., ΔE = 1.57 V). This reductive electron transfer path appears to be the key to achieving CO₂ reduction. In fact, we were able to observe reductive electron transfer by investigating the lifetime of the excited state of Ru(PS) immobilized on the Ag/YTON surface. Fig. 5 presents the emission decay curves of Ru(PS) adsorbed on YTON with various loading amounts of Ag during excitation at λ_ex = 444 nm. In these experiments, both the YTON and Ru(PS) underwent photoexcitation, and increasing the loading amount of Ag accelerated the emission decay of the excited state of Ru(PS). It is noteworthy that oxidative quenching of the excited state of the Ru(PS) did not occur, as discussed above. Previously, we have confirmed that electron- and/or energy-transfer from the excited state of Ru(PS) to Ag is negligible.¹¹ Therefore, the evidently enhanced emission quenching is attributed to the acceleration of the reductive quenching of the excited state of Ru(PS) by YTON as well as facilitation of the interfacial electron transfer between the YTON and Ru(PS) by the added Ag.

Fig. 4 Energy diagram of RuRu′/YTON.

Fig. 5 Emission decay profiles of Ru(PS)/Ag/YTON with different Ag loading in MeCN.

In any photocatalytic CO₂ reduction experiments, it is essential to investigate the origin of the carbon source(s) in the reaction product(s), because these may result from surface contaminants such as hydrocarbons and organic acids.¹⁸ Therefore, we performed the reduction of ¹³C-labeled CO₂ in a similar manner to the previous trials. After 60 h of visible light irradiation, the reacted suspension was filtered and the supernatant was analysed by ¹H NMR spectroscopy. As shown in Fig. S5 (ESI†), a doublet centred at δ = 8.42 ppm with a coupling constant of J¹³_CH = 192 Hz was observed, attributed to a proton bound to the ¹³C atom in H¹³COOH. The observed coupling constant is consistent with that reported in our previous work.^9a In addition to the doublet, a singlet peak appeared at δ = 8.42 ppm, which was attributed to H¹²COOH. This singlet was also observed when the reaction was conducted in the dark, indicating that H¹²COOH detected was not generated photocatalytically but rather resulted from contaminants, as indicated by our previous study.^10b Here, approximately 50% of HCOOH generated came from CO₂. XRD analysis also showed that no change in the XRD pattern of RuRu/Ag/YTON before and after reaction (Fig. S1, ESI†). On the basis of these results, it is concluded that the production of HCOOH over RuRu′/Ag/YTON hybrid proceeds via CO₂ reduction driven by the two-step photoexcitation of both YTON and the light-harvesting Ru complex in RuRu′.

In summary, we developed a new semiconductor material for a Z-scheme photocatalyst coupled with a binuclear Ru(II) complex, i.e., an yttrium–tantalum oxynitride having a band gap of 2.1 eV. This hybrid photocatalyst can reduce CO₂ to HCOOH with a very high selectivity (>99%).

This work was supported by the PRESTO/JST program “Chemical Conversion of Light Energy” and a Grant-in-Aid for Young Scientists (A) (Project 16H06130). The support from a Grant-in-Aid for Scientific Research on Innovative Area “Artificial Photosynthesis (AnApple)” (JSPS), the Photon and Quantum Basic Research Coordinated Development Program (MEXT, Japan), and a CREST program (JST) is also gratefully acknowledged. Finally, K. M. wishes to thank the Noguchi Institute and the Murata Science Foundation for financial support.

Notes and references

1. (a) T. Ohkawara, K. Suzuki, K. Nakano, S. Mori and K. Nozaki, J. Am. Chem. Soc., 2014, 136, 10728–10735; (b) T. Suga, H. Mizuno, J. Takaya and N. Iwasawa, Chem. Commun., 2014, 50, 14360–14363.
2. G. Qi, Y. Wang, L. Estevez, X. Duan, N. Anako, A. A. Park, W. Li, C. W. Jones and E. P. Giannelis, Energy Environ. Sci., 2011, 4, 444–452.
3. (a) H. Hori, F. Johnson, K. Koike, O. Ishitani and T. Ibusuki, J. Photochem. Photobiol., A, 1996, 96, 171–174; (b) J. Smieja and C. P. Kubiak, Inorg. Chem., 2010, 49, 9283–9289; (c) L. W. Christina, C. Jim and W. K. Matthew, Nature, 2014, 508, 504–507.
4. (a) T. Arai, S. Sato, T. Kajino and T. Morikawa, Energy Environ. Sci., 2013, 6, 1274–1282; (b) G. Sahara, R. Abe, M. Higashi, T. Morikawa, K. Maeda, K. Ueda and O. Ishitani, Chem. Commun., 2015, 51, 10722–10725; (c) U. Kang, S. K. Choi, D. J. Ham, S. M. Ji, W. Choi, D. S. Han, A. Abdel-Wahab and H. Park, Energy Environ. Sci., 2015, 8, 2638–2643.
5. C. Liu, J. J. Gallagher, K. K. Sakimoto, E. M. Nichols, C. J. Chang, M. C. Y. Chang and P. Yang, Nano Lett., 2015, 15, 3634–3639.
6. (a) J. Hawecker, J. M. Lehn and R. Ziessel, J. Chem. Soc., Chem. Commun., 1983, 9, 536–538; (b) H. Takeda, K. Koike, H. Inoue and O. Ishitani, J. Am. Chem. Soc., 2008, 130, 2023–2031; (c) S. Matsuoka, K. Yamamoto, T. Ogata, M. Kusaba, N. Nakashima, E. Fujita and S. Yanagida, J. Am. Chem. Soc., 1993, 115, 601–609; (d) H. Takeda, H. Koizumi, K. Okamoto and O. Ishitani, Chem. Commun., 2014, 50, 1491–1493; (e) H. Takeda, K. Ohashi, A. Sekine and O. Ishitani, J. Am. Chem. Soc., 2016, 138, 4354–4357.
7. (a) K. Iizuka, T. Wato, Y. Miseki, K. Saito and A. Kudo, J. Am. Chem. Soc., 2011, 133, 20863–20868; (b) K. Teramura, X. Wang, S. Hosokawa, Y. Sakata and T. Tanaka, Chem. – Eur. J., 2014, 20, 9906–9909.
8. S. Sato, T. Morikawa, S. Saeki, T. Kajino and T. Motohiro, Angew. Chem., Int. Ed., 2010, 49, 5101–5105.
9. (a) K. Maeda, K. Sekizawa and O. Ishitani, Chem. Commun., 2013, 49, 10127–10129; (b) K. Maeda, R. Kuriki, M. Zhang, X. Wang and O. Ishitani, J. Mater. Chem. A, 2014, 2, 15146–15151; (c) R. Kuriki, K. Sekizawa, O. Ishitani and K. Maeda, Angew. Chem., Int. Ed., 2015, 54, 2406–2409; (d) K. Maeda, R. Kuriki and O. Ishitani, Chem. Lett., 2016, 45, 182–184; (e) R. Kuriki, O. Ishitani and K. Maeda, ACS Appl. Mater. Interfaces, 2016, 8, 6011–6018.
10. (a) K. Sekizawa, K. Maeda, K. Domen, K. Koike and O. Ishitani, J. Am. Chem. Soc., 2013, 135, 4596–4599; (b) F. Yoshitomi, K. Sekizawa, K. Maeda and O. Ishitani, ACS Appl. Mater. Interfaces, 2015, 7, 13092–13097; (c) A. Nakada, T. Nakashima, K. Sekizawa, K. Maeda and O. Ishitani, Chem. Sci., 2016 10.1039/C6SC00586A.
11. R. Kuriki, H. Matsunaga, T. Nakashima, K. Wada, A. Yamakata, O. Ishitani and K. Maeda, J. Am. Chem. Soc., 2016, 138, 5159–5170.
12. (a) S. Wang, W. Yao, J. Lin, Z. Ding and X. Wang, Angew. Chem., Int. Ed., 2014, 53, 1034–1038; (b) S. Wang and X. Wang, Angew. Chem., Int. Ed., 2016, 55, 2308–2320; (c) J. Qin, S. Wang, H. Ren, Y. Hou and X. Wang, Appl. Catal., B, 2015, 179, 1–8.
13. (a) K. Maeda and K. Domen, J. Phys. Chem. Lett., 2010, 1, 2655–2661; (b) Y. Yamazaki, H. Takeda and O. Ishitani, J. Photochem. Photobiol., C, 2015, 25, 106–137.
14. M. Liu, W. You, Z. Lei, G. Zhou, J. Yang, G. Wu, G. Ma, G. Luan, T. Takata, M. Hara, K. Domen and C. Li, Chem. Commun., 2004, 2192–2193.
15. (a) K. Maeda and K. Domen, J. Catal., 2014, 310, 67–74; (b) A. Ishikawa, T. Takata, J. N. Kondo, M. Hara and K. Domen, J. Phys. Chem. B, 2004, 108, 11049–11053; (c) K. Ogisu, A. Ishikawa, Y. Shimodaira, T. Takata, H. Kobayashi and K. Domen, J. Phys. Chem. C, 2008, 112, 11978–11984.
16. (a) N. Nishimura, B. Raphael, K. Maeda, L. Le Gendre, R. Abe, J. Kubota and K. Domen, Thin Solid Films, 2010, 518, 5855–5859; (b) K. Maeda and K. Domen, Angew. Chem., Int. Ed., 2012, 51, 9865–9869.
17. (a) D. E. Scaife, Sol. Energy, 1980, 25, 41–54; (b) Y. Matsumoto, J. Solid State Chem., 1996, 126, 227–234.
18. T. Yui, A. Kan, C. Saitoh, K. Koike, T. Ibusuki and O. Ishitani, ACS Appl. Mater. Interfaces, 2011, 3, 2594–2600.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6cc03627a

---