A time saving and cost effective route for metal oxides activation

Abstract

Defects are the most reactive sites on the surface of metal oxides. Herein, thermal evacuation creates nonstoichiometric reduced forms of metals accompanying appropriate oxygen vacancies at specific optimized temperatures and timings without any doping or impregnation. Electron–hole recombination is reduced due to the presence of mixed valence states of metals. The generation of nonstoichiometric reduced forms of metals on the surface of bulk oxides is the promoting effect for higher photocatalytic activity and photostability.

---

Introduction

Chemistry that occurs at the surfaces of metal oxides (MO_x) is critical in a variety of industrial applications including catalysis, optical display technology, solar energy devices and corrosion prevention.¹ Metal oxide semiconductors are eminently attractive candidates for solar energy conversion applications particularly as photocatalysts for hydrogen generation from water.² Oxygen vacancies are the most reactive sites on the surfaces of metal oxides. Esch et al. reported an exciting study that clearly elucidated the formation of oxygen vacancies on a cerium oxide (CeO₂) surface.³ Unlike other candidate semiconductors such as gallium phosphide, indium phosphide or cadmium telluride oxides, the metal oxides like tungsten trioxide (WO₃), titania (TiO₂), zinc oxide (ZnO) and cerium oxide (CeO₂) do not contain precious metals or toxic elements. These are also chemically inert and have exceptional chemical and photo electrochemical stability in aqueous media. Titania (TiO₂) is an inert wide band gap insulator in its stoichiometric form; however, its applications are enabled by excess electrons originating from the defect state that is located within the band gap of reduced titania (TiO_2−x).⁴ Zinc oxide (ZnO) is an n-type semiconductor with a direct band gap. It is well-known that chemical doping, as well as intrinsic lattice defects, greatly influences electronic and optical properties of ZnO.⁵ Tungsten trioxide (WO₃) has proved to be an important material with broad technological applications.^6–9 The surface reactivity of WO₃ is connected with the presence of oxygen vacancies found in sub-stoichiometric compounds, since in the stoichiometric compound there are no cation d electrons available to be transferred to adsorbates. Understanding metal oxide reactivity thus requires an understanding of the nature of surface oxygen vacancies. The photo catalytic processes using metal oxide semiconductors have demonstrated that the limitations in achieving higher photo conversion efficiencies must overcome under solar irradiations. In photocatalysis, the recombination of electron–hole pairs must be inhibited to improve the overall quantum efficiency for interfacial charge transfer. Several approaches have been developed to achieve this goal, mostly involving the optimization of particle size/structure and surface area,¹⁰ surface modifications with redox couples or noble metals¹¹ and coupling two semiconductors with different electronic energy levels.¹²

Using thermal evacuation, we can shift this response even further toward the visible region. Importantly, in this approach, we show that thermal evacuation is demonstrated to be a versatile and energy efficient method for making the inorganic oxide semiconductors for photovoltaic or photocatalytic solar energy conversion. It is also possible to tune the optical characteristics of the oxide semiconductor (i.e., shift its response towards visible range of the electromagnetic spectrum). As a bonus, the resultant material shows enhanced surface properties relative to the sample obtained from commercial sources. Finally, this approach requires only very simple equipment. This simple and low cost approach is a step forward toward tailoring photocatalysts for various purposes and it can valuably contribute to photocatalyst design. After characterization, the photocatalytic performance of evacuated WO₃ was evaluated in comparison with that of pure WO₃ in the degradation of methylene blue (MB), methyl orange (MO) and 2,4-dichlorophenol (2,4-DCP). The generation of stable nonstoichiometric reduced forms of metals (MO_x−y/MO_x) on the surface of bulk oxides at specific temperature and timing is the promoting effect for the higher degradation rates of organics under visible light.

Experimental section

Materials

Tungsten trioxide (WO₃), zinc oxide (ZnO) and cerium dioxide (CeO₂) were supplied by Shanghai Sinopharm. Co., Ltd and Degussa P-25 titania (TiO₂) was supplied by Degussa Co. Ltd. Methylene blue, methyl orange and 2,4-dichloro phenol were of analytical grade received from Shanghai Ling Feng Chem., Sinopharm Chemical Reagent and Shanghai Chemical Reagent Co., Ltd Shanghai, respectively and used without any further purification. Double distilled water was used throughout the experiment.

Catalyst preparation

Pure WO₃, ZnO, CeO₂ and TiO₂ (Degussa P-25) are taken and thermally evacuated at 197.7 °C for 3 hours. The manufacturer of vacuum oven (DZF-6020) is Shanghai Hua-lion equipment Co. Ltd. China. Low cost and time saving thermal evacuation at specific temperature is adapted in the present study to tune the optical characteristics of the oxide semiconductor for the generation of modest oxygen vacancies. After the thermal evacuation, the powders change their colors in this manner.

Comm. WO₃ (pale yellow) to evacuated WO₃ (bluish green)

Comm. ZnO (white) to evacuated ZnO (light brownish)

Comm. CeO₂ (light yellow) to evacuated CeO₂ (orange)

Comm. TiO₂ (white) to evacuated TiO₂ (brownish)

Photocatalytic activity test

The photocatalytic activity was evaluated in terms of the degradation of methylene blue, methyl orange and 2,4-dichlorophenol. The photocatalytic reactions were carried out at 30 °C using a home-made reactor. The initial concentration of MB, MO and 2,4-DCP in a quartz reaction vessel was fixed at 10 and 15 mg L⁻¹, respectively, with a catalyst loading of 1.1 g L⁻¹. A 1000 W halogen lamp was used as the light source and the light from the lamp included beams from ultraviolet and visible light regions. The short-wavelength components (λ < 420 nm) of the light were cut off using a cut-off glass filter for visible photoreaction, mounted 10 cm away from the reaction solution. During the reaction, a water-cooling system cooled the water-jacketed photochemical reactor to maintain the solution at room temperature. The distance between the lamp and the centre of quartz tube was 10 cm. Prior to illumination, reaction mixture was sonicated for 20 min to homogenize and the suspension was magnetically stirred in darkness for 30 min to establish adsorption–desorption equilibrium at room temperature. During irradiation, stirring was maintained to keep the mixture in suspension. At regular intervals, samples were withdrawn and centrifuged to separate photocatalyst for analysis. Then filtered through a 0.22 μm Millipore filter to remove the photo catalyst. The photo activities for MB, MO and 2,4-DCP in dark in the presence of the photocatalyst and under visible light irradiation in the absence of the photocatalyst were also evaluated. The extent of MB and MO decomposition was determined by measuring the absorbance value at approximately 660 and 464 nm. The concentration of 2,4-DCP was calculated from the height of peak at 283 nm by using the calibration curve. The measurements were repeated for the catalyst three times and the experimental error was found to be within ±3%.

Characterization

UV-vis diffuse reflectance spectrum (DRS) was obtained with a Scan UV-vis-NIR spectrophotometer (Varian Cary 500) equipped with an integrated sphere assembly, using BaSO₄ as a reflectance sample. X-ray diffraction (XRD) measurements were carried out to investigate the crystallographic properties with a Rigaku D/Max 2550 VB/PC apparatus (Cu K_α1 radiation, λ = 1.54056 Å) at room temperature operated at 40 kV and 100 mA. Diffraction patterns were recorded in the angular range of 20–80°. The S_BET of the samples were determined through nitrogen physical adsorption at 77 K (Micromeritics ASAP 2010). All the samples were degassed at 473 K before the measurement. Raman spectra of the sample were recorded by Renishaw inVia Raman spectrometer at room temperature with the excitation wavelength of 514.6 nm. To investigate the chemical states of the photo catalysts, X-ray photoelectron spectroscopy (XPS) was recorded with PerkinElmer PHI 5000C ESCA system with Al K_α radiation operated at 250 W. The shift of binding energy due to relative surface charging was corrected using the C 1s level at 284.6 eV as an internal standard. Electron paramagnetic resonance (EPR) measurements for the catalyst powders were recorded at room temperature on a Bruker EMX-8/2.7 with the microwave frequency of 9.85 GHz and power of 6.35 mW. Thermogravimetric analysis (TGA) was performed on the Perkin-Elmer Pyris Diamond thermogravimetric analyzer. TGA trace for the pure and vacuum treated samples was carried out at a heating rate of 10 °C min⁻¹ from 40 to 800 °C in air. The UV-vis absorption spectra of the samples were recorded on Cary 100 UV/vis spectrophotometer.

Results and discussion

The first inkling that the optical response of the thermally evacuated samples are different from the commercial WO₃ is furnished by their visual appearance which are markedly darker blue than the yellow hue of the commercial WO₃ powder as shown in inset Fig. 1. Yellow color is indicative of stoichiometric WO₃ in which W has valency of +6. The appearance of blue color is evidence for the presence of reduced form as W⁵⁺. This is quantitatively borne out by the diffuse reflectance UV-visible spectrophotometer data of pure and evacuated metal oxides (Fig. 1 and S1†). The spectra for the evacuated sample (Fig. 1b) show stronger absorption at wavelengths longer than the band edge cut-off relative to the commercial sample without treatment (Fig. 1a). The DRS plot prepared by these afford the optical band gap (2.38 eV) for the thermally evacuated sample which show significant red shift relative to (2.61 eV) for the commercial sample. The value of 2.6–2.8 eV optical band gap for pure WO₃ is in accordance with the literature.¹³ The thermally evacuated WO₃ sample shows red shift due to the d–d transitions. These absorptions, responsible for blue colored sample, are associated with the reduction of W⁶⁺ to a lower oxidation state¹⁴ W⁵⁺ during evacuation process. The origin of this optical response shift was further probed by the characterization given below.

Fig. 1 (A) DRS plots and photograph of different samples of WO₃: (a) pure WO₃ without thermal evacuation, (b) WO₃ thermally evacuated obtained for diffused reflectance data shown in the inset.

Fig. 2A shows XRD patterns of pure and thermally evacuated WO₃. XRD profile of pure WO₃ is clearly attributed to monoclinic phase. The data demonstrate that the crystallite size of the bulk metal oxide is slightly decreased (43.0 vs. ∼39.0 nm) after vacuum activation processes due to the generation of reduced forms of metals; a higher number of unsaturated reduced metal sites are located on the surface. Fig. 2B shows the Raman spectra of pure WO₃ without treatment and vacuum treated WO₃. The Raman spectra are almost identical before and after the coloration, which means that no hydrogen tungsten bronze¹⁵ is formed under evacuation. Major Raman bands at 809, 718 (W–O stretching modes) and 271 cm⁻¹ (W–O bending modes) as shown in Fig. 2 are characteristic of monoclinic WO₃.¹⁶ Some other reflections of the surface states in the form of reduced W are also appeared in the evacuated samples. The reflections located at 445.8, 573.2, 640.3 and 673 cm⁻¹ may be considered as the surface defects in the reduced form of W.

Fig. 2 (A) Wide angle XRD patterns and (B) Raman spectra of different WO₃ samples: (a) pure WO₃, (b) evacuated WO₃.

These values are agreed well with the literature.¹⁷ The generation of oxygen vacancies increases the electron density near W; this makes the W–O bond longer. Thus one could expect the decrease in the stretching frequency. However, an increase in the W–O stretching frequency is observed in the Raman analysis (Fig. 2B(b)). To shed light on this discrepancy, it is important to take in account the change in the valence state of W ions, which works in the opposite direction, i.e., decreasing the length of W–O bonds.¹⁸ Therefore, for more ‘reduced’ W states a stronger bonding between tungsten and oxygen ligands is to be expected. It is the interplay of these two antagonistic effects that results in the behaviour observed experimentally.

A significant decrease in the surface area (5.01 versus 6.88 m² g⁻¹), using N₂ as a probe molecule, is observed after thermal evacuation (Table S1†). This decrease can be associated with the formation of smaller crystallites of reduced form of W and closing of pores due to the formation of smaller particles of reduced W on the surface of WO₃ (see ESI Fig. S2†).

Thermogravimetric analyses (TGA) of evacuated composite sample shows more weight loss as compared to pure sample due to the presence of more reduced forms of metals. From TGA, the stoichiometries of the WO_3−x, ZnO_1−x, CeO_2−x and TiO_2−x calculated are x = 0.60, 0.12, 0.18 and 0.41 (see ESI Fig. S3†). This type of surface defect and oxygen vacancies can cause a self-doping effect for the bulk WO₃. It can sensitize the bulk WO₃ towards visible light.

Fig. 3 shows the XPS patterns of pure WO₃ and thermally evacuated WO₃ samples, respectively. W4f occurs in doublet form W4f_7/2 and W4f_5/2. W4f of the evacuated sample is broader suggestive of more than one W oxidation state. These are favourably comparable with literature reports.¹⁹ The binding energies of W4f_7/2 of pure WO₃ without treatment show the binding energies at 35.4 eV which corresponds well with the W⁶⁺ while the binding energies of W4f_7/2 of evacuated sample appears a bit lower binding energies at 35.2 eV. This negative shift is due to the presence of W⁵⁺ which is generated as a result of thermal treatment in vacuum. The intensity of the W4f band increases with the thermal evacuation in consistent with XRD and Raman results. The new peak appeared at 36.4 eV in evacuated sample indicates the existence of new surface states. This result corresponds well to the report.²⁰ This peak is absent in W4f of pure WO₃ as clear from Fig. 3a. The binding energies of O1s of evacuated sample appear a bit lower binding energies at 530.2 eV as compared to 530.4 eV of pure WO₃ (Fig. 3B). O1s becomes broader and increase in intensity in thermally evacuated sample which shows the presence of more oxygen vacancies.²¹ The detailed respective metals XPS and deconvoluted O1s spectra of different pure and evacuated metal oxides are given in Fig. S4 and S5.†

Fig. 3 XPS spectra of W4f and O 1s of different samples: (a) pure WO₃, (b) evacuated WO₃.

In case of EPR spectra, pure WO₃ shows weak, less intensive and hyperfine signal (Fig. 4a). Thermal treatment in vacuum, which causes the loss of oxygen from the lattice, is accompanied by the formation of a number of EPR signals reflecting a variety of defects created during the process. Thermal treatment in vacuum on such samples led to very new weak and broad signals at g = 1.63 and 1.83 assigned to reduced form of tungsten²² (W⁵⁺) with oxygen vacancies which show the paramagnetic character as shown in Fig. 4b.

Fig. 4 EPR spectra of different samples of WO₃: (a) pure WO₃ without treatment, (b) thermally evacuated WO₃.

Methylene blue, a thiazine dye, is used as a probe of the surface and photo catalytic attributes of the evacuated metal oxide samples relative to the commercial specimen. This dye is a popular probe in the heterogeneous photocatalysis community and its subsequent decoloration and decomposition can be monitored via its visible-light absorption signature (at λ_max = 660 nm). Methyl orange is a stable azo dye and 2,4-DCP is toxic pollutant. The data was obtained from metal oxide dispersions equilibrated with the pollutants under visible-light illumination (Fig. 5). This can be interpreted by assuming that, with the formation of a significant number of surface states, in the form of oxygen vacancies are found to enhance charge transfer from the surface to the adsorbate, which increase the interactions between these. The photo catalytic activity of photocatalyst is determined by its ability of light absorbing, the efficiency of separation between photoelectrons and holes and the transfer rate of charge carriers. The particle size of the semiconductors for photocatalytic application is a crucial parameter influencing light adsorption and scattering, the photo induced charge carrier dynamics and the number of the active centres but also charge recombination centre on the surface of the photocatalyst. The degradation profile of MB, MO and 2,4-DCP on pure and evacuated WO₃ is compared in Fig. 5. High rates of MB degradation observed for evacuated metal oxides can be associated through relatively low band gaps with dye sensitization. MO is stable dye which can be much degraded by evacuated WO₃ as compared to pure metal oxides in 4 h (Fig. 5). The high rates of 2,4-DCP degradation are observed for evacuated samples compared to pure samples for 5 h. This can be attributed to the presence of surface active species on the bulk catalysts, owing to their nonstoichiometric nature (confirmed by TGA) and the lowering of band gap (as confirmed by DRS) via oxygen vacancies. The generation of oxygen vacancies is further confirmed by EPR, XPS and Raman data.

Fig. 5 Degradation profile of pure WO₃ without treatment and thermally evacuated pure WO₃ in case of MB, MO and 2,4-DCP samples.

The vacuum treatment of pure WO₃ for 3 h at 197.7 °C is optimized value for timings and temperature. Below this temperature, the samples activated in vacuum are lighter in color showing the less surface states. This sample shows less photocatalytic activity under visible light. Above this temperature, no more change in color of the samples but the photocatalytic activities are decreased. The temperature is optimized and the modest amount of oxygen vacancies is created at this temperature and timings. The sample activated in vacuum for 3 h shows higher activity than samples activated for lower time. Defects (oxygen vacancies) in the metal oxides play an important role in photocatalysis. These can trap the photo generated oxygen radicals, resulting in suppression of the recombination of electron–hole pairs and, thus, improved photo catalytic activity. Oxygen vacancies can reduce the electron–hole recombination as confirmed by the PLS spectra (see ESI Fig. S6†). More important, the number of reduced forms may vary with varying oxygen vacancies. The reduced form of the metals creates the energy levels below the conduction band of the metal oxides which results into the decrease of band gap (DRS) and thus increased the photocatalytic activity under visible light (Fig. S7†). The oxygen vacancies act as the additional impurity states or defect energy levels which contribute to the enhanced visible light absorption. Visible light illumination on thermally evacuated sample produces electron–hole pairs at the surface of the photo catalyst, the electrons are trapped by the oxygen vacancies and benefit the separation of the electron–hole pairs. The holes can produce OH radicals, or directly degrade the organic pollutants to CO₂ and H₂O. The proposed mechanism of degradation of organics over thermally evacuated metal oxides via generation of oxygen vacancies is shown in Fig. 6. Evacuated metal oxides shows high photo catalytic activity (Fig. S8 and S9†) due to the reduced band gap energy, electron–hole recombination and more absorbing ability of visible light which are the determining factors of photo catalyst efficiencies. It is also observed that this surface phenomenon is reversed at 500 °C for WO₃ in the air annealing time for 4 h and the annealing of the samples in nitrogen atmosphere is not able to reach the stoichiometry obtained by thermal evacuation.

Fig. 6 A degradation profile proposed reaction mechanism for visible light driven photo catalyst WO_3−x modified by thermal evacuation for the degradation of organics. C. B. conduction band, V. B. valence band. MO_x. pure metal oxides, MO_x−y. Thermally evacuated metal oxides, V⁰_O oxygen vacancies, V_O ionized oxygen vacancies, C. B. conduction band, V. B. valence band.

Herein, thermal evacuation can also be used as a simple, economic, versatile and energy-efficient method for tuning the characteristics of other metal oxide semiconductor photo catalysts. It creates nonstoichiometric reduced forms of metals accompanying appropriate oxygen vacancies at specific optimized temperature without any doping or impregnation. The band gap of metal oxides is decreased due to the transitions made available by reduced forms of metals. Different techniques confirm the presence of oxygen vacancies generated via thermal evacuation. The electron–hole recombination is reduced due to the mixed valence of metals. The detailed results obtained for different metal oxides (WO₃, TiO₂, CeO₂ and ZnO) are given in ESI.† Evacuated metal oxides show the photo stability due to the generation of reduced form of metals and oxygen vacancies via thermal evacuation. The catalyst's lifetime and stability are important parameters of the photo catalytic process so that it can be used for a longer period of time to reduce the cost of the treatment. For this reason, the catalyst was recycled three times which showed the least decrease in efficiency for the composite as shown in Fig. S10.† This loss can be due to filtration or fouling of catalyst. The photo stability of the composite is very clear from these experiments because of the generation of stable reduced states via thermal evacuation. The composite shows little color change after photo degradation reactions which show the photostability of the photo catalyst (Fig. S11†).

Conclusions

In conclusion, our results indicate that evacuated metal oxide materials are very promising as visible light efficient photo catalysts due to the generation of stable nonstoichiometric reduced forms of metals accompanying oxygen vacancies. Appropriate oxygen vacancies at specific temperature and timings can obviously improve the efficiency of the separation of electrons and holes and the absorbing ability for visible light, so as to improve the photo catalytic activity of the photo-catalysts. In this regard, it is a novel and economic route discovery that induces photo catalytic oxidation of organics without any dopant or impregnation under visible light.

Acknowledgements

This work has been supported by Higher Education Commission of Pakistan and East China University of Science and Technology, Shanghai China.

Notes and references

1. C. T. Campbell and C. H. F. Peden, Science, 2005, 309, 713.
2. K. Rajeshwar, J. Appl. Electrochem., 2007, 37, 765.
3. F. Esch, S. Fabris, L. Zhou, T. Montini, C. Africh, P. Fornasiero, G. Comelli and R. Rosei, Science, 2005, 309, 752.
4. M. Valden, X. Lai and D. W. Goodman, Science, 1998, 281, 1647.
5. S. Fujihara, Y. Ogawa and A. Kasai, Chem. Mater., 2004, 16, 2965.
6. C. Bittencourt, A. Felten, F. Mirabella, P. Ivanov, E. Liobet, M. A. P. Silva, L. A. O. Nunes and J. J. Pireaux, J. Phys.: Condens. Matter, 2005, 17, 6813.
7. S. Sajjad, S. A. K. Leghari and J. Zhang, RSC Adv., 2013, 3, 12678.
8. M. A. Butler, R. D. Nasby and R. K. Quinn, Solid State Commun., 1976, 19, 1011.
9. G. Hodes, D. Cahen and J. Manassen, Nature, 1976, 260, 312.
10. V. Chhabra, V. Pillai, B. K. Mishra, A. Morrone and D. O. Shah, Langmuir, 1995, 11, 3307.
11. S. Sajjad, S. A. K. Leghari, F. Chen and J. Zhang, Chem.–Eur. J., 2010, 16, 13795.
12. S. A. K. Leghari, S. Sajjad and J. Zhang, RSC Adv., 2013, 3, 15354.
13. C. G. Granqvist, Sol. Energy Mater. Sol. Cells, 2000, 60, 201.
14. G. Ramis, G. Busca, C. Cristiani, L. Lietti, P. Forzatti and F. Breganis, Langmuir, 1992, 8, 1744.
15. T. He, Y. Ma, Y. A. Cao, X. L. Hu, H. M. Liu, G. J. Zhang, W. S. Yang and J. N. Yao, J. Phys. Chem. B, 2002, 106, 12670.
16. Z.-X. Lu, S. M. Kanab and C. P. Tripp, J. Mater. Chem., 2002, 12, 983.
17. S. H. Lee, H. M. Cheong, J.-G. Zhang, C. E. Tracy, A. Mascarenhas, D. K. Benson and S. K. Deb, Appl. Phys. Lett., 1999, 74, 242.
18. A. Kuzmin, J. Purans, E. Cazzanelli, C. Vinegoni and G. Mariotto, J. Appl. Phys., 1998, 84, 5515.
19. S. O. Neill, I. P. Parkin, R. J. H. Clark, A. Mills and N. Elliott, Chem. Vap. Deposition, 2004, 10, 136.
20. C. Bittencourt, A. Felten, F. Mirabella, P. Ivanov, E. Llobet, M. A. P. Silva, L. A. O. Nunes and J. J. Pireaux, J. Phys.: Condens. Matter, 2005, 17, 6813.
21. C. Cantalini, H. T. Faccio, M. Palino, L. Lozzi and M. Passacantando, Sens. Actuators, B, 1996, 31, 81.
22. A. Fuerte, M. D. Hernández-Alonso, A. Iglesias-Juez, A. Martınez-Arias, J. C. Conesa, J. Soria and M. Fernandez-Garcıa, Phys. Chem. Chem. Phys., 2003, 5, 2913.

---

Footnote

† Electronic supplementary information (ESI) available: Full experimental, characterization and photocatalytic activity test details and XRD, Raman, BET, TGA, PLS, XPS results and spectral changes of degradation of different organics. See DOI: 10.1039/c3ra46518g

---