Interfacial charge separation in Cu₂O/RuO_x as a visible light driven CO₂ reduction catalyst

Abstract

We employ transient absorption spectroscopy to record the absorption spectrum of photogenerated charge carriers in Cu₂O. We have found that CO₂ reduction in Cu₂O is limited by fast electron–hole recombination. The deposition of RuO_x nanoparticles on Cu₂O results in a twofold increased yield of long-lived electrons, indicating partially reduced electron–hole recombination losses. This observation correlates with an approximately sixfold increase in the yield of CO₂ reduction to CO.

---

Photochemical reduction of CO₂ has the potential to convert this greenhouse gas into clean fuels or value-added chemicals. As such, it can contribute significantly to both renewable energy generation and CO₂ mitigation.¹ Photoreduction of CO₂ in heterogeneous systems was first reported using optical excitation of large bandgap semiconductor materials such as TiO₂, SrTiO₃, ZnO, or SiC with UV light irradiation.² Since this initial study, many reports have focused on the development of materials with improved light absorption properties in the visible region, such as narrow bandgap non-metal oxide semiconductors (e.g. ZnS, CdS, and GaP), or the functionalization or doping of wide bandgap semiconductors,³ although overall efficiencies reported for CO₂ photoreduction remain limited. These low efficiencies have often been attributed to fast charge recombination, requirements of high over-potentials and competition between CO₂ photoreduction reaction with H₂ evolution, as well as issues associated with catalyst/co-catalyst degradation or deactivation.⁴ Identifying the parameters that restrain the performance of catalysts towards CO₂ reduction and designing a system capable of overcoming these limitations is therefore a key challenge. In this context, transient absorption spectroscopy (TAS) is a technique that allows the assessment of competing kinetic processes in a semiconductor, namely charge separation and recombination, by monitoring the dynamics of charge carriers. We have, for example previously reported the signals corresponding to electrons and holes in Fe₂O₃ and other photoanodes for the related reaction of photochemical water-splitting.⁵

Herein, we report the dynamics of charge carriers in Cu₂O. To the best of our knowledge, the absorption spectrum and dynamics of photoexcited electrons in Cu₂O have not been identified. This narrow bandgap (ca. 2.2 eV) p-type semiconductor has been shown to be photocatalytically active for reduction of water⁶ and degradation of inorganic waste.⁷ Since the conduction band of Cu₂O lies above the reduction potential of CO₂ to CO,⁸ it is attracting increasing interest for the photoreduction of CO₂.⁹ As a p-type semiconductor, interfacial band-bending of Cu₂O, where present, is likely to favour interfacial reduction reactions but hinder oxidation reactions by presenting an energetic barrier for holes to approach the semiconductor surface. These surface reduction/oxidation reactions will be in kinetic competition with electron–hole recombination loss pathways (Fig. 1a). We have recently reported that the deposition of a charge acceptor (RuO_x, x ≤ 2) on top of Cu₂O improves its photocatalytic activity towards CO₂ reduction.^9b Herein, we focus on the study of the dynamics of photogenerated electrons in Cu₂O when forming an inorganic heterojunction with a material capable of (1) increasing the accessibility of photogenerated holes to the electrolyte, (2) having appropriate reaction sites to drive photo-oxidation reactions and (3) reducing the electron–hole recombination reaction on the Cu₂O surface (Fig. 1b). Ruthenium oxide (RuO_x) was selected as the overlayer material because its appropriate work function level can facilitate transfer of holes from Cu₂O, thereby potentially increasing the yield of long-lived electrons in Cu₂O to enable the reduction of CO₂. RuO_x is also less likely to react with Cu₂O to form an unwanted interfacial layer and has been shown to have the ability to oxidize many hole scavengers, including water.¹⁰

Fig. 1 Schematic of Cu₂O and the Cu₂O/RuO_x band diagram and electron and hole transfer reactions for the photocatalytic CO₂ reduction. Energy values were taken from ref. 8b and 12. E_F stands for Fermi level, E_CB and E_VB indicate the energy levels of the conduction band and valence band, respectively.

Spectroscopic measurements were performed using Cu₂O films deposited onto FTO glass. Cu₂O cuboid films with a nanoparticle size of ca. 100 nm were prepared by a chemical solution method using CuSO₄, Na₂S₂O₃ and NaOH, as reported previously.¹¹ RuO_x was subsequently deposited in the form of nanoparticles <10 nm on the Cu₂O surface.¹³ The RuO_x deposition did not alter the bandgap absorption of Cu₂O of approx. 2.2 eV. Only Cu₂O diffraction peaks were detected in Cu₂O–RuO_x samples, and Cu(I) species were confirmed by XPS analysis, albeit coexisting with small amounts of Cu(II) (Fig. S4, ESI†). Trace amounts of Ru were detected on the Cu₂O–RuO_x heterojunction by elemental analysis during TEM and XPS investigations, but not X-ray diffraction due to the low loading amount and small crystallite size of RuO_x. TAS spectra were recorded after UV excitation (355 nm) of Cu₂O and Cu₂O/RuO_x, using an experimental setup described previously.¹⁴ Details of synthetic procedures, material characterisation and experimental setup for transient absorption measurements can be found in the ESI† (Fig. S1–S4).

The absorption spectrum of charge carriers in a semiconductor can be characterized by TAS, by employing relevant fast hole and electron scavengers.¹⁵ Thus, the transient absorption spectrum of Cu₂O films deposited onto the FTO glass was obtained (i) in a N₂ purged aqueous solution (ii) following the addition of Na₂SO₃ as a hole scavenger or (iii) AgNO₃ as an electron scavenger. Photoexcitation of the Cu₂O semiconductor in the absence of charge carrier scavengers results in the appearance of a ground state bleach signal between 475 and 750 nm and two positive photoinduced absorption peaks at wavelengths <475 nm and >800 nm (Fig. 2). The signals at wavelengths >850 nm are significantly enhanced by the addition of Na₂SO₃ and weakened by the addition of AgNO₃. Given that Na₂SO₃ can effectively scavenge holes, while Ag⁺ scavenges electrons, we can assign the transient signals above 850 nm to Cu₂O photoexcited electrons. On the other hand, we observed only the signal corresponding to photoexcited holes in Cu₂O at wavelengths <475 nm when using AgNO₃ as an electron scavenger. Thus, the direct probing of holes under the current experimental photocatalytic conditions is hindered by the partial overlap with the bleach signal of the ground state and cannot be directly monitored. Since these signals corresponding to photoexcited electrons and holes in the Cu₂O are already present in the 10 μs timescale, our measurements indicate that electron–hole scavenging by the relevant chemicals are very fast (<10 μs), beyond the instrument's limit of response.

Fig. 2 Transient absorption spectrum of Cu₂O recorded under N₂, in the presence of a 0.1 M Na₂SO₃ and 0.01 M AgNO₃ aqueous solutions. The samples were excited at λ_ex = 355 nm, with a laser intensity of ∼1.2 mJ cm⁻².

The formation of inorganic heterojunctions or the addition of sacrificial agents which are able to scavenge charge carriers have been reported to improve the performance of the photocatalytic systems for organic waste decomposition or water splitting reactions by decreasing the recombination of photogenerated electrons and holes in a semiconductor.^4b,16 Thus, transient absorption spectroscopy experiments were also employed to assay the effect of RuO_x deposition on the Cu₂O charge carrier dynamics. Typical kinetic data of the photoinduced electron absorption signals taken at 950 nm are plotted in Fig. 3, showing the decay dynamics on the micro- to second timescales. We note that we only observed significant long-lived signal amplitudes with relatively intense excitation densities, indicative of the presence of significantly fast (<10 μs) electron–hole recombination in these films. The presence of Na₂SO₃ resulted in an increase in the amplitude of this Cu₂O electron signal (Fig. 2), although this effect is relatively minor, indicative of inefficient hole scavenging/fast recombination, and consistent with the expected band bending impeding access of photogenerated holes to the Cu₂O surface (Fig. 1a). The deposition of RuO_x on the Cu₂O surface results in a significant increase in the amplitude of the long-lived electron signal, independent of the presence of Na₂SO₃, and indicative of a more efficient reduction in fast electron–hole recombination losses. These results qualitatively suggest that RuO_x deposition results in a significant increase in the yield of long-lived (>100 μs) Cu₂O electrons. This is attributed to a reduction in fast electron–hole recombination losses due to formation of the inorganic junction, most probably due to hole transfer from Cu₂O to RuO_x, thereby increasing the spatial separation of electrons and holes and facilitating the photooxidation reaction by holes. The lack of absorption of RuO in the near IR region of the spectrum indicates that the increase in absorbance observed at 950 nm is not due to the reduction of RuO_x by Cu₂O photoexcited electrons, and therefore confirms our peak assignation.¹⁷ We want to note that a Cu₂O–Al:ZnO–TiO₂–RuO_x heterojunction for H⁺ reduction, where the Cu₂O and RuO_x interfaces are connected through an interfacial (AZO–TiO₂) layer, has been reported by Tilley et al.^6d Contrary to our approach, where RuO_x accepts the holes from Cu₂O, charge separation in Cu₂O–AZO–TiO₂–RuO_x takes place through electron transfer from Cu₂O to RuO_x. We believe that, while in the Cu₂O–AZO–TiO₂–RuO_x there is a favorable downhill electron transfer cascade from Cu₂O to RuO_x, the AZO–TiO₂ layer blocks the hole transfer to RuO_x.

Fig. 3 TAS decays of photo-excited electrons of Cu₂O films (black trace) and Cu₂O–RuO_x films (red trace) in N₂ purged water, and in the presence of a N₂ purged 0.1 M Na₂SO₃ aqueous solution (Cu₂O films: grey trace, Cu₂O–RuO_x films: pink trace). Data obtained at λ_probe = 950 nm with an excitation intensity of 1.6 mJ cm⁻² at λ_ex = 355 nm.

The effect of enhanced charge separation by the Cu₂O–RuO_x junction strategy on CO₂ photoreduction was tested using analogous bare Cu₂O and Cu₂O–RuO_x nanoparticulate suspensions in CO₂-saturated deionised H₂O in the presence of a Na₂SO₃ hole scavenger upon irradiating under the full arc of a 150 W Xe lamp (Fig. 4). In agreement with the results observed upon irradiating the samples with visible light λ ≥ 420 nm,^9b the coupling of Cu₂O with RuO_x results in a nearly sixfold increase of the initial rate of CO production, from 0.16 μmol g⁻¹ on bare Cu₂O to around 0.88 μmol g⁻¹ on the Cu₂O–RuO_x junction after 1 h irradiation using a 150 W Xe lamp. Trace amounts of methanol (0.01 μmol g⁻¹ h⁻¹) and methane (<0.001 μmol g⁻¹ h⁻¹) were also observed in the Cu₂O–RuO_x sample during the reaction. Quantum yield measurements were carried out using a near monochromatic visible light 400 nm bandpass filter (Δλ < 10 nm, filtered light output ca. 600 ± 40 μW cm⁻²). The apparent initial quantum yield for CO generation (20 min) for the Cu₂O–RuO_x heterojunction was determined to be ca. 1.6%. It is also noted that the CO evolution rate is reduced after the first hour. These results can be explained in terms of the instability of Cu₂O towards the presence of increasing amounts of SO₄²⁻ formed upon scavenging the holes by Na₂SO₃.¹⁸ Another possible explanation for non-linear CO evolution could be the strong interaction between CO and Cu₂O, which will result in lowered catalytic activity over time.¹⁹ Studies to address this instability by using alternative hole scavengers and further modifying material protection layers are ongoing.

Fig. 4 CO₂ reduction to CO on bare and heterojunction based Cu₂O core photocatalysts under full arc 150 W Xe lamp irradiation. Photoreduction reactions were undertaken in 3 mL of CO₂ saturated H₂O in the presence of a hole scavenger (ca. 0.7 M Na₂SO₃).

RuO_x has been reported to be able to either reduce or oxidize water,^6d,20 but is unable to reduce CO₂ due to its work function being more positive than the E_red(CO₂/CO) reduction potential. We did not observe any water oxidation and oxygen evolution upon the addition of RuO_x, as holes are preferentially scavenged by Na₂SO₃. Thus, the increase in the CO₂ photoreduction yield is consistent with the efficient charge separation due to the inorganic heterojunction, where holes are transferred to RuO_x and CO₂ reduction takes place at the surface of Cu₂O.

In order to provide evidence that the formation of CO is due to the photoreduction of CO₂ by Cu₂O, three different control experiments were carried out: (1) traces of organic contaminants were removed by treating the aqueous suspension system under strong light irradiation without CO₂ until there was no increase in the CO or CO₂ amount as detected by gas chromatography before the photocatalytic measurements. (2) Control experiments under similar conditions but in the absence of photocatalyst, CO₂ or light, and irradiating the Cu₂O with longer wavelength light (λ > 668 nm, much lower than Cu₂O bandgap requirement) demonstrated that surface carbon gasification is not a major contributing factor (Fig. S5, ESI†). (3) Gas chromatography – Mass spectrometry isotope labelling analysis showed the sole formation of ¹³CO (m/z 29) (Fig. S6, ESI†). These results strongly indicate that the CO observed by GC is due to photochemical reduction of CO₂ rather than from the reduction of organic contaminants present on the surface of Cu₂O–RuO_x.

Conclusions

In summary, we have assigned the transient absorption spectrum of Cu₂O photoexcited electrons to appear at wavelength λ > 800 nm, and holes at λ < 475 nm. In bare Cu₂O films, the major loss pathway limiting their photocatalytic activity is the fast electron–hole recombination. Our results demonstrate that the Cu₂O–RuO_x heterojunction strategy is effective in partially suppressing this electron–hole recombination loss through the transfer of photogenerated holes from Cu₂O to RuO_x. This is shown by a significant increase in the yield of long-lived Cu₂O electrons as observed by transient absorption spectroscopy. This electron yield and a lifetime increase is translated into a sixfold photocatalytic reaction increase by Cu₂O–RuO_x coupling in favour of CO₂ reduction.

Acknowledgements

Financial support from the EPSRC (EP/H046380/1) and ERC (project Intersolar to J. D.) is gratefully acknowledged. A.R. thanks the European Commission Marie Curie CIG (PhotoCO2) and the Spanish Ministry of Education for the EX2010-0479 postdoctoral fellowship. E. P. also thanks the Spanish Ministry of Education, the University of Valencia and the EU for the Erasmus scholarship. The authors thank Dr M. Ardakani (Imperial College London) for TEM and elemental analyses and Prof. P. Cumpson (NEXUS, Newcastle University) for XPS analyses.

Notes and references

1. (a) E. V. Kondratenko, G. Mul, J. Baltrusaitis, G. O. Larrazabal and J. Perez-Ramirez, Energy Environ. Sci., 2013, 6, 3112; (b) S. C. Roy, O. K. Varghese, M. Paulose and C. A. Grimes, ACS Nano, 2010, 4, 1259; (c) G. Centi, E. A. Quadrelli and S. Perathoner, Energy Environ. Sci., 2013, 6, 1711; (d) Y. Izumi, Coord. Chem. Rev., 2013, 257, 171.
2. T. Inoue, A. Fujishima, S. Konishi and K. Honda, Nature, 1979, 277, 637.
3. (a) B. Kumar, M. Llorente, J. Froehlich, T. Dang, A. Sathrum and C. P. Kubiak, Annu. Rev. Phys. Chem., 2012, 63, 541; (b) S. Navalón, A. Dhakshinamoorthy, M. Álvaro and H. Garcia, ChemSusChem, 2013, 6, 562; (c) S. N. Habisreutinger, L. Schmidt-Mende and J. K. Stolarczyk, Angew. Chem., Int. Ed., 2013, 52, 7372; (d) T. W. Woolerton, S. Sheard, E. Reisner, E. Pierce, S. W. Ragsdale and F. A. Armstrong, J. Am. Chem. Soc., 2010, 132, 2132; (e) S. Sato, T. Arai, T. Morikawa, K. Uemura, T. M. Suzuki, H. Tanaka and T. Kajino, J. Am. Chem. Soc., 2011, 133, 15240; (f) K. Sekizawa, K. Maeda, K. Domen, K. Koike and O. Ishitani, J. Am. Chem. Soc., 2013, 135, 4596.
4. (a) A. Corma and H. Garcia, J. Catal., 2013, 308, 168; (b) A. J. Cowan and J. R. Durrant, Chem. Soc. Rev., 2013, 42, 2281.
5. M. Barroso, S. R. Pendlebury, A. J. Cowan and J. R. Durrant, Chem. Sci., 2013, 4, 2724.
6. (a) C.-C. Hu, J.-N. Nian and H. Teng, Sol. Energy Mater. Sol. Cells, 2008, 92, 1071; (b) P. E. de Jongh, D. Vanmaekelbergh and J. J. Kelly, Chem. Commun., 1999, 1069; (c) A. Paracchino, V. Laporte, K. Sivula, M. Gratzel and E. Thimsen, Nat. Mater., 2011, 10, 456; (d) S. D. Tilley, M. Schreier, J. Azevedo, M. Stefik and M. Graetzel, Adv. Funct. Mater., 2013, 24, 303; (e) C.-Y. Lin, Y.-H. Lai, D. Mersch and E. Reisner, Chem. Sci., 2012, 3, 3482.
7. (a) Y. Zhang, B. Deng, T. Zhang, D. Gao and A.-W. Xu, J. Phys. Chem. C, 2010, 114, 5073; (b) C. H. Kuo, C. H. Chen and M. H. Huang, Adv. Funct. Mater., 2007, 17, 3773.
8. (a) E. Ruiz, S. Alvarez, P. Alemany and R. A. Evarestov, Phys. Rev. B: Condens. Matter Mater. Phys., 1997, 56, 7189; (b) H. Raebiger, S. Lany and A. Zunger, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 045209.
9. (a) K. Tennakone, A. H. Jayatissa and S. Punchihewa, J. Photochem. Photobiol., A, 1989, 49, 369; (b) A. D. Handoko and J. Tang, Int. J. Hydrogen Energy, 2013, 38, 13017; (c) K. Rajeshwar, N. R. de Tacconi, G. Ghadimkhani, W. Chanmanee and C. Janáky, ChemPhysChem, 2013, 14, 2251.
10. (a) M. Neumann-Spallart, K. Kalyanasundaram, C. Grätzel and M. Grätzel, Helv. Chim. Acta, 1980, 63, 1111; (b) J. Kiwi, M. Grätzel and G. Blondeel, J. Chem. Soc., Dalton Trans., 1983, 2215.
11. M. Ristov, G. Sinadinovski and I. Grozdanov, Thin Solid Films, 1985, 123, 63.
12. (a) J. Deuermeier, J. Gassmann, J. Brotz and A. Klein, J. Appl. Phys., 2011, 109, 113704; (b) L.-S. Wang, H. Wu, S. R. Desai and L. Lou, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 53, 8028; (c) A. Klein, Thin Solid Films, 2012, 520, 3721; (d) Y. L. Chueh, C. H. Hsieh, M. T. Chang, L. J. Chou, C. S. Lao, J. H. Song, J. Y. Gan and Z. L. Wang, Adv. Mater., 2007, 19, 143.
13. M. Kohno, T. Kaneko, S. Ogura, K. Sato and A. Yasunobu Inoue, J. Chem. Soc., Faraday Trans., 1998, 94, 89.
14. A. Reynal, F. Lakadamyali, M. A. Gross, E. Reisner and J. R. Durrant, Energy Environ. Sci., 2013, 6, 3291.
15. (a) J. Tang, J. R. Durrant and D. R. Klug, J. Am. Chem. Soc., 2008, 130, 13885; (b) M. Barroso, C. A. Mesa, S. R. Pendlebury, A. J. Cowan, T. Hisatomi, K. Sivula, M. Grätzel, D. R. Klug and J. R. Durrant, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 15640.
16. (a) S.-I. In, D. D. Vaughn and R. E. Schaak, Angew. Chem., Int. Ed., 2012, 51, 3915; (b) H. G. Kim, P. H. Borse, J. S. Jang, E. D. Jeong, O.-S. Jung, Y. J. Suh and J. S. Lee, Chem. Commun., 2009, 5889; (c) S. C. Yan, S. X. Ouyang, J. Gao, M. Yang, J. Y. Feng, X. X. Fan, L. J. Wan, Z. S. Li, J. H. Ye, Y. Zhou and Z. G. Zou, Angew. Chem., Int. Ed., 2010, 49, 6400.
17. G. L. Zimmerman, S. J. Riviello, T. A. Glauser and J. G. Kay, J. Phys. Chem., 1990, 94, 2399.
18. M. J. Siegfried and K.-S. Choi, J. Am. Chem. Soc., 2006, 128, 10356.
19. (a) A. V. Larin, Langmuir, 1987, 3, 318; (b) H. Yamashita, M. Matsuoka, K. Tsuji, Y. Shioya, M. Anpo and M. Che, J. Phys. Chem., 1996, 100, 397; (c) X. Wang, J. C. Hanson, A. I. Frenkel, J.-Y. Kim and J. A. Rodriguez, J. Phys. Chem. B, 2004, 108, 13667.
20. A. Harriman, I. J. Pickering, J. M. Thomas and P. A. Christensen, J. Chem. Soc., Faraday Trans. 1, 1988, 84, 2795.

---

Footnotes

† Electronic supplementary information (ESI) available: Detailed experimental procedures, characterisation of Cu₂O and Cu₂O–RuO_x powder and films, and control experimental data. See DOI: 10.1039/c4cp00102h

‡ Present address: Department of Chemistry, National University of Singapore, 3 Science Drive 3, 117543, Singapore.

§ Present address: Stephenson Institute for Renewable Energy, Department of Chemistry, University of Liverpool, Crown Street, Liverpool, L69 7ZD, UK.

---