Sn₃O₄: a novel heterovalent-tin photocatalyst with hierarchical 3D nanostructures under visible light

Abstract

Sn₃O₄ with hierarchical 3D nanostructures was synthesized for the first time by a facile template-free solvothermal method. The heterovalent photocatalyst is highly efficient and stable in the degradation of azo dyes under visible light.

---

Over the past years, tin oxides such as SnO₂ and SnO have attracted considerable research interest because of their wide range of applications in heterogeneous catalyses,^1a field-effect transitors,^1b chemical and biological sensors,^1c,d photonics and lithium-ion batteries,^1e,f sensitized solar cells,^1g,hetc. Besides the univalent tin oxides, heterovalent tin oxides (e.g. Sn₃O₄, Sn₂O₃, and Sn₅O₆) with a variety of coordination structures have been reported by applying theoretical calculations² and regulating Sn(II)/Sn(IV) ratio in experimental studies.³ Significantly, the heterovalent tin structures have been suggested to influence the properties of tin oxides. Park et al. have reported that the heterovalent tin oxide shows more sensitive for CH₂Cl₂ detection than either SnO or SnO₂ univalent oxides.⁴ As proposed by theoretical calculations, the Sn₃O₄ has been experimentally confirmed to possess absorption bands in visible-light region,⁵ which may have potential applications as the visible-light-driven photocatalyst in environmental remediation and energy conversion.

As it is known, nanosized catalyst can provide more promising properties than a conventional bulk material due to the high surface-to-volume ratio and more beneficial defects.⁶ Inspired by the close correlation between shape/morphology and performances of photocatalysts, fabrication of desired three-dimensional (3D) nanoarchitectures of Sn₃O₄ by employing low dimensional blocks could offer more opportunities for wide practical applications. Herein, we report a facile template-free solvothermal method for the fabrication of a hierarchical flower-like Sn₃O₄ nanomaterial. It reveals that the hierarchical Sn₃O₄ exhibits superior photoreactivity towards methyl orange (MO) and 4-phenylazophenol degradation under visible light irradiation. This work may contribute to the comprehensive understanding of heterovalent semiconductor oxides, and open up a new way to design and prepare other more practicable and efficient visible-light photocatalysts.

The crystallinity and phase purity of the as-prepared catalysts were determined by X-ray diffraction (XRD). All the diffraction peaks match the standard data for a triclinic Sn₃O₄ structure (JCPDS 16-0737) (Fig. 1a). For comparison, SnO₂ (JCPDS 41-1445) and SnO (JCPDS 6-0395) were also synthesized as reference samples (ESI† for Experimental details). It can be seen that no characteristic peaks of any other Sn-related impurities are detected in the patterns of the Sn₃O₄. Moreover, to confirm the nonexistence of the other mixtures, Raman spectroscopy was performed to clarify the crystalline compositions and structures of the above-mentioned samples. As shown in Fig. 1b, the Raman patterns of these samples are quite different from each other. For SnO, two fundamental Raman peaks are detected at 113 and 211 cm⁻¹, corresponding to the B_1g and A_1g vibration modes of the tetragonal structure, respectively.⁷ Several characteristic peaks of SnO₂ at 438, 472, 576 and 633 cm⁻¹ are observed. Among the levels, the peaks centered at 472 and 633 cm⁻¹ are assigned to E_g and A_1g phonon modes of SnO₂,⁸ respectively, while the peaks at 438, and 576 cm⁻¹ are characteristic peaks of nanobelts or nanotube arrays of SnO₂.⁶ Compared with SnO and SnO₂, the main patterns centered at 72, 90, 143, 170 and 238 cm⁻¹ are characteristic peaks of phonon modes of Sn₃O₄, consistent with the literature.⁹ Clearly, no other impurities could be found in the Sn₃O₄ as well, indicating the purity and high crystallinity of the catalyst.

Fig. 1 XRD patterns (a) and Raman spectra (b) of catalysts.

The morphology and microstructure of the Sn₃O₄ were examined by SEM and TEM. As shown in Fig. 2a and b, the low magnification SEM image indicates the product consists of a wealth of 3D flower-like microstructures with a scale size of approximately 0.8–2.0 μm in diameter, while the high magnification image shows each flower is composed of a large quantity of nanopetals with smooth surface. The nanopetals are ca. 8 nm in thickness assembled to form the flower-like hierarchical structure (Fig. 2c and d). Additionally, it shows the corresponding SAED patterns taken from leaves of flower (Fig. 2c). The diffraction image points marked (111), (−121) and (−210) index in accordance with the single crystalline triclinic Sn₃O₄. And the lattice spaces of the Sn₃O₄ crystallites are determined as 0.329 and 0.282 nm, also belonging to the (111) and (−210) crystallographic planes of triclinic Sn₃O₄ (JCPDS 16-0737), respectively (Fig. 2d). As the nanopetals do not contain pores, the aggregation of the self-assembled nanoplates may result in the mesopores (2.0–4.0 nm) formed between stacked nanoplates⁶ and give rise to a high BET surface area (87.66 m² g⁻¹) (Fig. S1, ESI†). Such a self-organized mesoporous architecture might be of importance in photocatalysis since it provides efficient transport pathways for reactant and product molecules.¹⁰

Fig. 2 SEM (a and b) and TEM (c and d) images of as-prepared Sn₃O₄. Inserts of (c and d) show the SAED patterns and magnifying HRTEM of Sn₃O₄, respectively.

The optical absorption of these samples was investigated using a UV-vis diffusion reflectance spectroscopy (DRS) spectrometer, as shown in Fig. 3a. The Sn₃O₄ sample presents the photoabsorption ability from UV light region to visible light with the wavelength shorter than 470 nm. A classical Tauc approach¹¹ was used to estimate the optical band gap for the catalysts. As a result, the optical band gaps are 2.65, 3.36 and 3.12 eV for Sn₃O₄, SnO₂, and SnO, respectively, which is in well accordance with the reported data.^5,12 Moreover, photocurrent response measurement was also carried out under visible light irradiation (λ > 420 nm) to confirm the generation of photoinduced charge carriers on the Sn₃O₄ (Fig. 3b). Compared to the SnO or SnO₂, the Sn₃O₄ exhibits a much stronger photocurrent response, implying that a more efficient separation of photoinduced carriers is obtained and a superior photocatalytic activity is expected. Moreover, the response switches promptly and reversibly when visible light is turned on and off, suggesting the high stability of the Sn₃O₄ in photocatalytic reactions. The flat-band potential derived from electrochemical analysis (Mott–Schottky plot) is around −1.10 V vs. NHE at pH 7 for the Sn₃O₄ (Fig. S2, ESI†). By combining with the band gap estimated from optical absorption, the corresponding valence band position locates at 1.55 V vs. NHE at pH 7. Note that the photoinduced electrons in the conduction band of Sn₃O₄ possess a large thermodynamic driving to reduce O₂ (O₂/O₂˙⁻, −0.33 V vs. NHE at pH 7).¹³ This feature would naturally make it possible for the O₂˙⁻ generation. To confirm the existence of O₂˙⁻ radicals, ESR spin-trap was employed to probe the nature of the reactive oxygen specied. As shown in Fig. S3,† under visible light irradiation, six strong and obvious peaks for DMPO-O₂˙⁻ species were observed in Sn₃O₄ methanolic dispersion, which is intimately correlated to the photocatalytic degradation of organic pollutants.

Fig. 3 (a) UV-vis DRS of catalysts. Insert shows the data plotted as transformed K–M function versus energy of light absorbed. (b) Photocurrent spectra of catalysts irradiated by visible light (λ > 420 nm).

The activities of the as-prepared catalysts were evaluated by monitoring the photodegradation of methyl orange (MO) under visible light (Fig. 4a). As expected, no degradation is observed for the SnO₂. Although the SnO can strongly absorb visible light, it is also inert for the photodegradation of MO because of its unfavorable band gap structure for the photoinduced carrier transfer and the subsequent photocatalytic reactions.¹⁴ The Sn₃O₄ exhibits a much superior activity in comparison with N-doped TiO₂. For example, the Sn₃O₄ can completely decompose MO within 20 min, while only about 10% MO can be decomposed by the N-doped TiO₂ under the same conditions. Moreover, the Sn₃O₄ displays good stability and maintains high photocatalytic performance during three reaction cycles (Fig. 4b). No variations could be detected for the Sn₃O₄ used before and after photocatalytic reaction (Fig. S4, ESI†). It demonstrates that the Sn₃O₄ catalyst is stable for the MO degradation, which is significant for the potential practical applications. In addition, the superior photoactivity of the Sn₃O₄ can be also confirmed by decomposing 4-phenylazophenol under visible light (Fig. S5, ESI†). The photodegradation rate of 4-phenylazophenol over the Sn₃O₄ is ca. 8 times higher than that of the N-doped TiO₂ under the same conditions. It seems that the hierarchical Sn₃O₄ photocatalyst is efficient for azo dye degradation under visible light irradiation.

Fig. 4 (a) Photocatalytic degradation of MO over catalysts under visible light irradiation (λ > 420 nm). Insert shows UV-vis spectral changes of MO as a function of reaction time. (b) Stability test for MO photodegradation. (c) Sn3d scans of the Sn₃O₄ irradiated by visible light in vacuum. (d) Content of Snⁿ⁺ (n = 0, 2 and 4) on the Sn₃O₄ as a function of irradiation time.

As it is known, Sn₃O₄ exists two tin oxidation states where one-third of the Sn atoms are in a Sn(IV) octahedral coordination to oxygen, and two-thirds are in a Sn(II) tetrahedral coordination to oxygen.^2b,15 When considering the heterovalent coordinations, an immediate question is whether there are any interactions between the two oxidation states or whether the both tin states contribute to the electric band gap structure of Sn₃O₄ and relate to the superior photocatalytic activity? To explore a possible contribution from the different oxidation states, in situ XPS was performed to investigate the changes of Sn-related states in the photophysical–photochemical processes (Fig. 4c). For Sn₃O₄ before irradiation, the prominent peak of Sn3d_5/2 level is deconvoluted into two peaks centered at 486.4 and 487.0 eV, attributed to Sn(II) and Sn(IV) configurations, respectively.¹⁵ And the corresponding Sn(II)/Sn(IV) ratio is in close proximity to the stoichiometric Sn₃O₄ (Table S1, ESI†). Upon visible-light irradiation, the Sn(II)/Sn_total atomic ratio almost remains unchanged, whereas an obvious variation is detected for the Sn(IV) oxidation state on the Sn₃O₄. It shows that the Sn(IV)/Sn_total ratio decreases in proportion from 0.338 to 0.156, while the Sn(0)/Sn_total ratio increases linearly from 0 to 0.215, suggesting that a faction of Sn₃O₄ is reduced to metallic tin by photoinduced electrons in the absence of electron trappers such as adsorbed oxygen molecules. In combination with the aforementioned characterization, neither SnO₂ nor SnO could be found in the as-prepared Sn₃O₄ sample. Given that SnO₂ or SnO is formed and below the minimum detection limit of the measurements, they are inactive for the visible-light photocatalytic degradation of azo dyes, as revealed by the photocatalytic tests. It is believed that there is a strong interaction between the Sn(IV)–O octahedron and Sn(II)–O tetrahedron, and the conduction band of the Sn₃O₄ could be resulted from the hybridization of Sn-5s and 5p states of the two tin atoms in the unit cell.^5,16 That is, the optical excitation of electrons from the valance band, i.e. O-2p states, to the conduction band would contribute to the Sn⁴⁺ → Sn²⁺ and Sn²⁺ → Sn⁰ redox steps, which are quite similar to the changes of heterovalent cobalt in the photoinduced Co₃O₄ system.¹⁷ In this way, the content of Sn²⁺-related species in the Sn₃O₄ will remain constant only if Sn²⁺-related species still exists, as confirmed by the XPS analysis. It should be noted that the used sample in vacuum will revive to the stoichiometric Sn₃O₄ when it is exposed to O₂-rich atmosphere, also suggesting the high stability of the Sn₃O₄ during the photocatalytic process in the presence of oxygen.

In summary, Sn₃O₄ photocatalyst with flower-like hierarchical nanostructures was successfully synthesized by a facile template-free solvothermal method. Results show that the heterovalent material possesses a band gap of 2.65 eV, with appropriate positions of conduction band (CB) and valance band (VB) [CB = −1.10 V, VB = 1.55 V vs. NHE, pH = 7]. The Sn₃O₄ exhibits a superior photoreactivity toward MO and 4-phenylazophenol degradation under visible light irradiation, and the catalyst is found to be stable and reusable, creating the promise of easily-available heterovalent materials for environmental applications.

This work was financially supported by the National Natural Science Foundation of China (21173047, 21073036 and 21373049), National Basic Research Program of China (973 Program, 2013CB632405).

Notes and references

1. (a) L. A. Shi and H. L. Lin, Langmuir, 2010, 26, 18718; (b) M. S. Arnold, P. Avouris, Z. W. Pan and Z. L. Wang, J. Phys. Chem. B, 2002, 107, 659; (c) X. Xu, J. Zhuang and X. Wang, J. Am. Chem. Soc., 2008, 130, 12527; (d) Y. Cheng, K. S. Chen, N. L. Meyer, J. Yuan, L. S. Hirst, P. B. Chase and P. Xiong, Biosens. Bioelectron., 2011, 26, 4538; (e) S. J. Ding, J. S. Chen, G. G. Qi, X. N. Duan, Z. Y. Wang, E. P. Giannelis, L. A. Archer and X. W. Lou, J. Am. Chem. Soc., 2011, 133, 21; (f) D. Deng and J. Y. Lee, Chem. Mater., 2008, 20, 1841; (g) M. A. Hossain, J. R. Jennings, Z. Y. Koh and Q. Wang, ACS Nano, 2011, 5, 3172; (h) J. Y. Liu, T. Luo, T. S. Mouli, F. L. Meng, B. Sun, M. Q. Li and J. H. Liu, Chem. Commun., 2010, 46, 472.
2. (a) H. J. Matzke, in Nonstoichiometric oxides, ed. O. T. Sorensen, Academic Press, New York, 1981, p. 2; (b) F. Lawson, Nature, 1967, 215, 955–956; (c) M. A. Maki-Jaskari and T. T. Rantala, Modell. Simul. Mater. Sci. Eng., 2004, 12, 33–41; (d) A. Seko, A. Togo, F. Oba and I. Tanaka, Phys. Rev. Lett., 2008, 100, 45702–45704.
3. M. S. Moreno, R. F. Egerton and P. A. Midgley, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 69, 233304.
4. S. H. Park, Y. C. Son, W. S. Willis, S. L. Suib and K. E. Creasy, Chem. Mater., 1998, 10, 2389.
5. O. M. Berengue, R. A. Simon, A. J. Chiquito, C. J. Dalmaschio, E. R. Leite, H. A. Guerreiro and F. E. G. Guimarães, J. Appl. Phys., 2010, 107, 033717.
6. K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquerol and T. Siemieniewska, Pure Appl. Chem., 1985, 57, 603–619.
7. X. Wang, F. X. Zhang, I. Loa, K. Syassen, M. Hanfland and Y. L. Mathis, Phys. Status Solidi B, 2004, 241, 3168–3178.
8. P. S. Peercy and B. Morosin, Phys. Rev. B: Condens. Matter Mater. Phys., 1973, 7, 2779–2786.
9. F. P. Wang, X. T. Zhou, J. G. Zhou, T. K. Sham and Z. F. Ding, J. Phys. Chem. C, 2007, 111, 18839–18843.
10. (a) F. Dong, W. K. Ho, S. C. Lee, Z. B. Wu, M. Fu, S. C. Zou and Y. Huang, J. Mater. Chem., 2011, 21, 12428–12436; (b) J. G. Yu, Y. R. Su and B. Cheng, Adv. Funct. Mater., 2007, 17, 1984–1990.
11. M. Arslan, H. Duymus and F. Yakuphanoglu, J. Phys. Chem. B, 2006, 110, 276–280.
12. (a) J. G. Lu, P. Chang and Z. Fan, Mater. Sci. Eng., R, 2006, 52, 49–91; (b) M. Z. Iqbal, F. Wang, T. Feng, H. Zhao, M. Y. Rafique, R. Din, M. H. Farooq, Q. U. A. Javed and D. F. Khan, Mater. Res. Bull., 2012, 47, 3902–3907.
13. A. J. Bard, L. R. Faulkner, in Electrochemical Methods Funda mentals and Applications, John Wiley & Sons, New York, 1980.
14. F. A. Frame, E. C. Carroll, D. S. Larsen, M. Sarahan, N. D. Browning and F. E. Osterloh, Chem. Commun., 2008, 2206–2208.
15. D. N. Wang, A. C. Miller and M. R. Notis, Surf. Interface Anal., 1996, 24, 127.
16. J. Köhler, J. W. Tong, R. Dinnebier and A. Simon, Z. Anorg. Allg. Chem., 2012, 638, 1970–1975.
17. L. Qiao, H. Y. Xiao, H. M. Meyer, J. N. Sun, C. M. Rouleau, A. A. Puretzky, D. B. Geohegan, I. N. Ivanov, M. Yoon, W. J. Weber and M. D. Biegalski, J. Mater. Chem. C, 2013, 1, 4618–4633.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details and characterization of the as-prepared samples. See DOI: 10.1039/c3ra45743e

---