Peroxo-niobium oxyhydroxide sensitized TiO₂ crystals

Abstract

We report a facile method to produce active photocatalysts by treating niobium oxyhydroxide/TiO₂ composites with H₂O₂. The peroxo-niobium groups generated on the oxyhydroxide make the composite active under visible light. The interaction between the niobium oxyhydroxide and TiO₂ enhanced the photogenerated charge separation on the composite, and consequently its photocatalytic activity.

---

Since the first work by Fujishima and Honda¹ on the UV-induced redox chemistry on TiO₂, several studies focusing on the development of novel photocatalysts based on metal oxides, oxysulfides, and oxynitrides have been reported.^2–4 Among these materials, TiO₂ is an excellent candidate as a photocatalyst due to its chemical and physical stability, availability, and non-toxicity.⁵ However, because of its large bandgap energy (E_g = 3.2 eV), TiO₂ does not work as a photocatalyst under visible light, which makes its use of solar light limited.

To improving the photocatalytic performance of TiO₂ under visible light irradiation strategies including doping of TiO₂ with transition metal (Fe, Cr, Mn, Zn, W and V)⁶ or with non-metal (C, F, S and N),^6–8 organic dyes sensitizing,⁹ and noble metal deposition on the TiO₂ surface¹⁰ have been used. However, these strategies present some drawbacks for the developing efficient photocatalysts, as the doping of TiO₂ with transition metals or non-metals usually creates electron–hole recombination sites, the dyes used as sensitizers can be consumed during the photocatalytic process, and noble metals are expensive. Alternatively, TiO₂ can be sensitized with an inorganic semiconductor with small bandgap.^11,12

In this communication, we report on the design of a visible-light-harvesting photocatalyst based on the peroxo-niobium oxyhydroxides sensitized TiO₂ crystals. The peroxo-niobium groups on the amorphous oxyhydroxide surfaces allow the photocatalyst to be activated by visible light, and the coupling between TiO₂ and the niobium oxyhydroxide facilitates the charge separation on the semiconductor, thus improving its photocatalytic efficiency.

The photocatalysts were prepared by precipitating niobium oxyhydroxide on the TiO₂ surface (sample NbTi), followed by treatment with H₂O₂ to produce peroxo-niobium oxyhydroxides/TiO₂ composites (sample NbTi//H₂O₂). The photocatalysts were characterized by scanning electron microscopy (SEM) coupled on an energy dispersive X-ray analyzer, X-ray diffraction (XRD), and UV-Vis spectroscopy with diffuse reflectance geometry (UV-Vis DRS). The photocatalytic activity of the catalysts was evaluated by monitoring the oxidation of methylene blue (MB) dye under a Xe lamp (λ > 400 nm) on a UV-Vis spectrophotometer. The products of MB oxidation were detected by positive ion mode electron spray ionization mass spectroscopy (ESI-MS). The reactive photocatalytic species were investigated by scavenger tests. Details on the synthesis procedure, characterization, and photocatalytic tests are given in the ESI.†

Fig. 1 shows the SEM images and the respective EDS spectra of the prepared photocatalysts. The TiO₂ image (Fig. 1a) exhibited a 3D morphology with crystal aggregates of irregular sizes. After formation of the composite (Fig. 1b and c), the micrographs displayed similar 3D morphologies with crystal aggregates before and after the treatment with H₂O₂. The EDS spectrum of sample TiO₂ (Fig. 1d) showed only signals of the Ti element. In addition to the Ti signals, the respective EDS spectra of samples NbTi and NbTi/H₂O₂ (Fig. 1e and f) presented signals due to the Nb element, confirming the impregnation of niobium (red line) on the crystals of TiO₂. The Nb content in the composites, determined by ICP-MS, was approximately 6 wt%.

Fig. 1 SEM images of the samples (a) TiO₂, (b) NbTi, and (c) NbTi/H₂O₂. EDS spectra of the photocatalysts (d) TiO₂, (e) NbTi, and (f) NbTi/H₂O₂.

XRD was used to investigate the crystallographic structure of the photocatalysts. The structure, crystallite size and crystallinity of TiO₂ play an important role in the photocatalytic activity, and many studies have confirmed that among the polymorphs of TiO₂, anatase exhibits higher photocatalytic activity than brookite and rutile.¹³ Anatase in the samples TiO₂, NbTi, and NbTi//H₂O₂ (Fig. S1†) was identified by its diffraction maxima (101), (103), (004), (112), (200), (105) and (211) according to JPCDS # 4-477. Any evidence of the existence of rutile or brookite could not be identified in the diffraction patterns of these samples, indicating that anatase is the single polymorph of TiO₂ in these samples. In addition to anatase, the samples NbTi and NbTi//H₂O₂ exhibited an amorphous phase in the range from 20 to 35°, which can be related to niobium oxyhydroxides dispersed on TiO₂. The crystallite average apparent sizes of TiO₂ estimated from Scherrer equation were 38(8), 13(2) and 12(2) nm for the samples TiO₂, NbTi, and NbTi//H₂O₂, respectively.

The bandgap energies of the photocatalysts were determined from the UV-Vis DRS data (Fig. 2). The estimated E_g for the pure TiO₂ was found to be 3.22 eV (λ = 385 nm), which is consistent with the values reported in the literature.¹⁴ The composites NbTi and NbTi//H₂O₂ presented bandgap energies of 3.13 (396 nm) and 3.03 eV (409 nm), respectively. It is interesting to note that the sample NbTi/H₂O₂ absorbs visible light up to 465 nm, which may be due to the formation of peroxo groups on the surface of the niobium oxyhydroxide. Our findings suggest that the treatment of the composite with H₂O₂ shifts the absorption edge of the composite to the visible range. Thus, the NbTi//H₂O₂ may be attractive as a visible-light-harvesting photocatalyst. A recent publication from our group¹⁵ identified that the treatment of Nb₂O₅ with H₂O₂ produced a yellow compound due to the formation of peroxo groups on its surface.

Fig. 2 Diffuse reflectance spectra of the photocatalysts.

The photocatalytic activity of the catalyst was evaluated in a standard reaction of methylene blue oxidation under a solar simulator (Fig. 3). The control experiment (only the dye and light, no photocatalyst) showed that there was no significant discoloration even after 270 min of reaction. Upon light irradiation, approximately 25% of the total MB was removed by both TiO₂ and NbTi photocatalysts. In the presence of the composite treated with H₂O₂ (sample NbTi//H₂O₂), the highest discoloration capacity with approximately 83% removal after 270 min of reaction was achieved. These results clearly showed that the treatment of the composite with H₂O₂ was crucial to producing an active photocatalyst. The rate constants for the TiO₂, NbTi, and NbTi//H₂O₂ were 0.00058(8), 0.00071(5), and 0.00638(9) min⁻¹. It means that the treatment with H₂O₂ produced a photocatalyst approximately 10 times more active that the samples TiO₂ or NbTi.

Fig. 3 Kinetics of MB oxidation under Xe lamp (λ > 400 nm). [MB] = 10 mg mL⁻¹, volume of the solution = 80 mL, mass of catalyst = 60 mg, initial pH = 6.03.

Discoloration measurements by UV-Vis spectroscopy did not provide any information on the reaction intermediates of methylene blue dye oxidation. To confirm that the methylene blue was oxidized rather than only adsorbed on the photocatalyst surface, we carried out ESI-MS analysis. Identification of intermediates was monitored using spectra of the reaction after 60 min with TiO₂, NbTi, and NbTi//H₂O₂ under ultraviolet light (15 W germicide lamp) (Fig. S2a†) or sunlight simulator (Fig. S2B†). In the methylene blue oxidation by photocatalysts using sunlight simulator or ultraviolet light, many organic intermediates were observed after 60 min reaction. These results suggest the methylene blue oxidation via a photocatalytic mechanism (signals at m/z = 300 and 316, referring to the first and second hydroxylations). The successive oxidation leads to ring tension followed by ring-opening reactions (Fig. S3†), as demonstrated by the appearance of intermediates with m/z ratios lower than 284. Peaks at m/z = 148, 184 and 202 were indicative of ring rupture and probable subsequent mineralization, as confirmed by total organic carbon analysis. This mechanism was in agreement with the high catalytic activity previously discussed for these samples.

TOC removal analysis showed that after 60 min under UV light and TiO₂, NbTi or NbTi/H₂O₂ samples, the carbon removal were 35, 42 and 56%, respectively. In the same conditions and using sunlight simulator instead of UV light, the carbon removal were 21, 34 and 41% when the TiO₂, NbTi or NbTi/H₂O₂ samples were used respectively. It can be seen that the reaction under ultraviolet light was more efficient in TOC removal for all materials. Furthermore, it was interesting that the modified composite (NbTi//H₂O₂) presented the highest removal capacity, suggesting that pretreatment with hydrogen peroxide might enhance the efficiency of the photocatalytic system.

The recyclability tests of the NbTi/H₂O₂ sample (Fig. S4†) indicated that the MB removal capacity decreased from 83% in the fresh photocatalyst to 53 and 47% in the first and second reuse cycle. It was probably due to the MB adsorbed on the photocatalyst surface was not completely mineralized. Nevertheless, even at the first and second cycle of reuse, the photocatalyst was more active than the single TiO₂ or NbTi.

It is accepted that organic compounds can be degraded via photocatalytic oxidation by different species, e.g. hydroxyl (˙OH) and superoxide (⁻˙O₂) radicals, and holes (h⁺). Thus, by adding various scavengers to the reaction media to remove the corresponding reactive species, the role of different reactive species in the photocatalytic process can be evaluated based on the changes in the photocatalytic efficiency. Fig. S5† shows the removal capacity of NbTi//H₂O₂ in the presence of different scavengers. Without the addition of scavengers, the decolorization of MB was 80% after 120 min of irradiation. By adding isopropanol to suppress ˙OH, the decolorization efficiency of MB decreased to 62%, which indicated that the ˙OH affected slightly the capacity of MB removal. On the other hand, when ammonium oxalate was used as hole-trap, the photodegradation of MB decreased to 15%, indicating that the photogenerated hole are critical to the photocatalytic process on the NbTi//H₂O₂. In the presence of benzoquinone, the MB removal substantially reduces to 9%. Based on these findings, the order of reactive species in the photocatalytic reaction mediated by NbTi//H₂O₂ was ⁻˙O₂ > h⁺ > ˙OH.

Based on the photocatalytic results it was proposed a mechanism of MB photodegradation by NbTi//H₂O₂ (Fig. 4). By this mechanism, the composite absorbs visible light, and the electron in the valence band of the niobium oxyhydroxide is excited to the conduction band. The holes in the valence band of the niobium oxyhydroxide can directly oxidize the MB to produce non-colored compounds, or then oxidize water to produce hydroxyl radicals, which can further oxidize the MB molecule. The electron in the conduction band of the niobium oxyhydroxide can be injected into the conduction band of TiO₂, thus enhancing the reactive charge separation. The electrons in the conduction band of TiO₂ can then reduce the O₂ to produce superoxide radicals, which can further oxidize the MB molecule. Thus, the NbTi//H₂O₂-assisted photodegradation of MB may occur via two primary processes including a photocatalytic and dye-photosensitized process. Both the photocatalytic and the photosensitized process most likely occurs concurrently. However, the photocatalytic process may be the predominant process due to photodegradation rate of MB in the absence of photocatalyst is small.

Fig. 4 Schematic representation of TiO₂ sensitization by peroxo-niobium oxyhydroxide.

Conclusions

The treatment of niobium oxyhydroxide composites with H₂O₂ produced a visible-light active photocatalyst due to the formation of peroxo-niobium groups on the oxyhydroxide surface. The sensitization of TiO₂ with peroxo-niobium oxyhydroxide allowed the formation of superoxide radicals, which was the main reactive specie in this photocatalytic system. The possibility of using solar energy to activate the photocatalyst makes this material promising for use in low-cost photocatalytic systems for the treatment of wastewaters.

Acknowledgements

This work was supported by the CNPq, FAPEMIG, and CAPES.

Notes and references

1. A. Fujishima and K. Honda, Nature, 1972, 238, 37.
2. M. C. Pereira, E. M. Garcia, A. C. Silva, E. Lorençon, J. D. Ardisson, E. Murad, J. D. Fabris, T. Matencio, T. C. Ramalho and M. V. J. Rocha, J. Mater. Chem., 2011, 21, 10280.
3. F. Zhang, K. Maeda, T. Takata, T. Hisatomi and K. Domen, Catal. Today, 2012, 185, 253.
4. A. Fuertes, J. Mater. Chem., 2012, 22, 3293.
5. P. G. W. A. Kompio, A. Brückner, F. Hipler, G. Auer, E. Löffler and W. Grünert, J. Catal., 2012, 286, 237.
6. G. Liu, C. Han, M. Pelaez, D. Zhu, S. Liao, V. Likodimos, N. Ioannidis, A. G. Kontos, P. Falaras, P. S. M. Dunlop, J. A. Byrne and D. D. Dionysiou, Nanotechnology, 2012, 23, 294003.
7. M. V. Dozzi, S. Livraghi, E. Giamello and E. Selli, Photochem. Photobiol. Sci., 2011, 10, 343.
8. J. H. Pan, G. Han, R. Zhou and X. S. Zhao, Chem. Commun., 2011, 47, 6942.
9. S. Ito, S. M. Zakeeruddin, R. Humphry-Baker, P. Liska, R. Charvet, P. Comte, M. K. Nazeeruddin, P. Péchy, M. Takata, H. Miura, S. Uchida and M. Grätzel, Adv. Mater., 2006, 18, 1202.
10. A. Gallo, M. Marelli, R. Psaro, V. Gombac, T. Montini, P. Fornasiero, R. Pievo and V. D. Santo, Green Chem., 2012, 14, 330.
11. A. K. L. Sajjad, S. Shamaila, B. Tian, F. Chen and J. Zhang, Appl. Catal., B, 2009, 91, 397.
12. S. Zhan, J. Yang, Y. Liu, N. Wang, J. Dai, H. Yu, X. Gao and Y. Li, J. Colloid Interface Sci., 2011, 355, 328.
13. D. Gumy, A. G. Rincon, R. Hajdu and C. Pulgarin, Sol. Energy, 2006, 80, 1376.
14. R. López and R. Gómez, J. Sol-Gel Sci. Technol., 2012, 61, 1.
15. L. C. A. Oliveira, M. F. Portilho, A. C. Silva, H. A. Taroco and P. P. Souza, Appl. Catal., B, 2012, 117–118, 29.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra07343j

---