Solar light sensitive samarium-doped ceria photocatalysts: microwave synthesis, characterization and photodegradation of Acid Orange 7 at atmospheric conditions and in the absence of any oxidizing agents

Abstract

Novel, high surface area, mesoporous and crystalline samarium-doped ceria (CeO₂:Sm³⁺) nanopowders were successfully synthesized by combining the excellent properties of both microwave heating and surfactants and were used as remarkably efficient new photocatalysts for the degradation of a representative azo dye, Acid Orange 7 (AO7), in an aqueous medium under natural sunlight without the addition of any external reagents like peroxide, acid or base. The synthesized nanopowders were characterized by X-ray diffraction (XRD), transmission electron microscopy (TEM), Brunauer–Emmett–Teller (BET) analysis and UV-vis diffuse reflectance spectroscopy. The effects of calcination temperature, pH of the medium, catalyst dosage and irradiation time on the decolorization of AO7 were investigated and are discussed in this paper. Sm³⁺ doping in CeO₂ narrowed the band gap and significantly enhanced the photocatalytic degradation of the azo dye. The photocatalytic degradation of AO7 was also investigated by using certain radical scavengers and the results suggest that under solar light irradiation predominantly positive holes and superoxide radicals (O₂˙⁻) act as the active species in the degradation process. Our results suggest that the materials developed here are a promising alternative solar light sensitive photocatalyst.

---

1. Introduction

Nowadays, tremendous efforts are dedicated to water treatment research to ensure there is sufficient water supply as the global consumption of water is increasing. Wastewater generated from the textile, leather, paper, plastic, pharmaceutical and food industries extensively contains azo dyes, which are largely non-biodegradable and toxic under aerobic conditions and reduced to potential carcinogenic aromatic amines under anaerobic conditions. As a result, azo dyes are regarded as a major source of ecological and environmental problems.^1,2 Acid Orange 7 [p-(2-hydroxy-1-naphthylazo) benzene sulfonic acid] (AO7), which has a –N=N– unit as the chromophore in its molecular structure, is a popular water-soluble dye that poses a severe health threat to humans with effects including eye, skin, mucous membrane, and upper respiratory tract irritation, severe headaches, nausea, methemoglobinemia, tumors, water-borne diseases such as dermatitis etc.³ Therefore, the development of a simple and effective technology for the purification of dye-containing wastewater is imperative.

For degrading organics and toxic materials, many treatment methods require high pressure (0.5–20 Mpa) and high temperature (80–320 °C),⁴ which limit their practical application. To resolve this problem, heterogeneous photocatalysis using TiO₂ and ZnO photocatalysts in particular has emerged as a promising advanced oxidation process for the purification of dye-containing wastewater and also provides solutions to many problems related to environmental pollution.^5–7 Mostly these processes involve UV/H₂O₂, or UV/O₃ for the oxidative degradation of dyes.^6–8 Although H₂O₂ is an environmentally friendly oxidant and widely used for environmental remediation, disinfection, etc. the current industrial process for H₂O₂ production requires toxic solvents and a high energy input, and therefore is not environmentally benign.^9,10 Photo-assisted oxidation by solar irradiation has been found to be a cost effective and ideal approach, as it is readily available and free and can be used directly to oxidize or degrade hazardous organic chemicals.⁵ The advantage of CeO₂ (band gap E_g ∼ 2.9 eV) in comparison with TiO₂ (E_g ∼ 3.0 to 3.2 eV for anatase and absorbs less than 5% UV light of the sunlight) is that it absorbs over a larger fraction of the UV spectrum and the onset of absorption of CeO₂ is ∼440 nm.¹¹ CeO₂ nanopowder has been reported to be a more efficient photocatalyst than commercial TiO₂ P25.^11–13 So, recently, CeO₂ has gained immense interest as a photocatalyst for the photocatalytic degradation of various dyes as it is chemically stable, inexpensive and can be easily synthesized through various reproducible solution-based synthesis routes.^14–18 It has already been used in various catalytic applications such as fluid cracking, purification of harmful gases in three way automotive catalytic converters, water splitting for the generation of H₂ gas, as well as in biomedicine, solar cells, and inorganic phosphors.^19–23

The key to the wide range of catalytic applications of ceria-based materials is that CeO₂, by shifting some Ce⁴⁺ to Ce³⁺ ions, can easily produce oxygen vacancies, which act as sources for the oxygen involved in reactions taking place on the catalyst surface.^15,18,24 The effectiveness of a CeO₂ photocatalyst can be enhanced by introducing other metal ions into it by forming a solid solution.^16,18 It has also been reported that lanthanide ions including La³⁺, Nd³⁺, Pr³⁺, Sm³⁺, or Eu³⁺ doped in TiO₂ photocatalysts improved the separation rate of photo-induced charge carriers and greatly enhanced the photocatalytic activity of TiO₂ over pure TiO₂.²⁵ Cai et al. observed improved degradation of AO7 in the dark as well as under visible light irradiation in a CeO₂–H₂O₂ system after Fe³⁺ doping.¹⁸

It is well known that the concentration of oxygen vacancies can be remarkably enhanced after doping with Sm, due to Sm³⁺ → Ce⁴⁺ substitutions.²⁰ Sm³⁺ doping also induces the least distortion of the parent lattice when oxygen vacancies are created in the CeO₂ lattice for charge compensation. Many studies have reported on the Sm_0.2Ce_0.8O_1.9 system showing the highest electrical conductivity (specific to ionic conductivity) required for solid oxide fuel cell applications.²² Also, samarium-doped ceria nanoparticles are very important in developing new luminescence devices.²⁰ But no investigation on Sm³⁺-doped CeO₂ as a photocatalyst has been reported. Therefore, in this study, we aimed to obtain an efficient photocatalyst by doping CeO₂ with Sm³⁺ ions.

Recently, Sun et al. reported a 97.6% degradation efficiency of AO7 by nitrogen-doped monocrystalline CeO₂ (N : Ce molar ratio of 0.3) nanoparticles synthesized solvothermally at 120 °C for 24 h in an autoclave.¹⁴ They investigated the degradation of a 20 mg L⁻¹ AO7 solution (at pH = 3.0) containing 1 g L⁻¹ of the as-synthesized sample, which had air bubbled through it at 60 °C at a flow rate of 300 mL min⁻¹, using a domestic 10 W compact fluorescent lamp.¹⁴ Salker et al. reported the low solar light assisted photocatalytic activity of CeO₂ for the degradation of the textile dye Naphthol Blue Black.²⁶ But a significant increase in the photocatalytic activity was observed by substitution of 30 mol% Mn⁴⁺ in the CeO₂ crystal lattice due to the enhancement of the redox couple Ce⁴⁺–Ce³⁺ and decrease in the band gap energy.²⁶ A few studies have also demonstrated the potential use of CeO₂,^12,13,15 or doped CeO₂ (ref. 16–18) for the photocatalytic degradation of dyes under visible light irradiation (mostly λ > 420 nm).^13,16,18 However, in most cases they are successful in the presence of high concentrations of oxidizing agents,^15,18 and/or under specific pH conditions.¹⁴ These conditions limit the industrial application of nanomaterials in environmental remediation. There is a need to develop general, simple and economical routes for designing novel nanomaterials which can efficiently remove various contaminants in the environment under solar illumination. Due to higher reaction rates and selectivities, microwave chemistry has recently been shown to be a fast-growing research area with immense potential.^27–29 Microwave-assisted methods have been effectively employed to synthesize a great many novel nanostructures with various shapes.^27–29

However, the application of microwave heating with the addition of surfactants to synthesize mesostructured nanomaterials has hardly been exploited until now. Here, we used a surfactant-assisted microwave heating method, which enables the convenient low temperature preparation of efficient CeO₂ photocatalysts with a higher surface area and higher crystallinity. In the present study, we demonstrated a new photochemical remediation method for dye-polluted waters by using samarium-doped ceria nanoparticles and natural sunlight only. The effects of the calcination temperature, pH and the amount of nanopowder were also investigated. Though several studies on the mechanistic details of dye degradation on TiO₂ have been reported under UV as well as visible light, some insight of the active species for the photodegradation process on CeO₂ under solar light irradiation is given here for the first time.

2. Materials and method

2.1 Synthesis of Sm³⁺-doped CeO₂ materials

All chemicals were of analytical grade and used as-received without further purification. (NH₄)₂Ce(NO₃)₆ and Sm₂O₃ were used as the sources of cerium and samarium, respectively. Samarium (1 mol%) doped ceria (CeO₂:Sm³⁺) nanopowders were synthesized by a novel surfactant-assisted microwave heating route using sodium dodecyl sulfate as the surfactant and NH₄OH as a precipitating agent. First, an 18 wt% sodium dodecyl sulfate solution was prepared by dissolving sodium dodecyl sulfate in water–ethanol (molar ratio 3 : 1) and stirring the solution for 2 h. Samarium nitrate was prepared by dissolving the requisite amount of samarium oxide in the minimum volume of concentrated nitric acid. Samarium nitrate was then dissolved separately in distilled water to a 0.5 M concentration. This solution was mixed with a 0.5 M aqueous solution of ammonium ceric ammonium nitrate and stirred for 30 min. The surfactant solution was added dropwise to the mixture solution containing the inorganic salts under vigorous stirring conditions. Under magnetic stirring, an NH₄OH solution (30 vol%) was added dropwise to the above solution to obtain a precipitate at a pH of 9. The mixture was aged at room temperature and then washed with water. The precipitate was dispersed in a certain volume of water and finally treated in a microwave oven (National, 2450 MHz, 300 W), for 30 min at 90 °C. The final precipitate was filtered, washed with distilled water and ethanol several times and dried at 80 °C. The as-prepared sample was then calcined at higher temperatures in air for 2 h. It is important to mention that the X-ray diffraction analysis of the as-prepared powder indicated the formation of a single phase of Sm³⁺:CeO₂ with a cubic fluorite structure (see Fig. 1a).

Fig. 1 XRD patterns of the as-prepared CeO₂:Sm³⁺ powders (a) and the powders calcined at (b) 500 °C, and (c) 800 °C for 2 h.

2.2 Characterization

X-Ray diffraction patterns were obtained on a Philips X’Pert Pro diffractometer using 0.154056 nm CuKα radiation. Microstructures were measured on a JEOL-JEM 2100 transmission electron microscope. Nitrogen adsorption–desorption isotherms and BJH pore size distributions were obtained at 77 K on a Quantachrome Autosorb-1 apparatus after degassing the samples at 200 °C for 3 h. UV-visible (UV-vis) diffuse reflectance spectra were measured using a Shimadzu spectrometer (UV-2450) using BaSO₄ as a blank. FT-IR spectra of the dye before and after photodegradation under solar irradiation were recorded on a PerkinElmer FTIR spectrometer. In order to confirm the final concentration of Sm³⁺ present in the CeO₂, EDS analysis was performed during the SEM study (not shown here) on a Nova NanoSEM 450/FEI equipped with an EDAX detector.

2.3 Evaluation of the photocatalytic activity

The photocatalytic efficiency of the CeO₂:Sm³⁺ samples was evaluated by the extent of degradation of Acid orange 7 (AO7) in an aqueous solution under solar light irradiation at the location of 22°15′N 84°54′E during the months of May and June (sunny days). Solar radiation was measured using a Pyranometer (with a global radiation sensor). The intensity of the sunlight (0.2 kW m⁻² at 7 am) reached its highest up to 1.0 kW m⁻² at noon and again diminished gradually to 0.3 kW m⁻² by 5:00 pm. From 10:00 am to 3:00 pm, the average solar intensity was ∼0.9 kW m⁻² and the intensity fluctuations were minimal. No steps were taken to maintain the intensity of the sunlight during subsequent reactions. In a typical experiment, a 20 mL AO7 solution (concentration: 2 × 10⁻⁴ M, 70 mg L⁻¹) was taken in a 100 mL beaker, to which 20 mg (1 g L⁻¹) of the photocatalyst was added separately. To achieve adsorption–desorption equilibrium, the suspension was stirred in the dark for 30 min, then exposed to sunlight at room temperature with vigorous stirring to ensure the mixing of the catalyst. The experiments using solar light were carried out between 7:00 am and 5:00 pm and stirred for a period ranging from 1 h to 10 h without any adjustment of the pH and then centrifuged. Sample aliquots were withdrawn from the reaction mixture at regular time intervals and centrifuged. After that the dye concentration in the residual solution was analyzed using a Shimadzu UV-2450 spectrophotometer. Changes in the concentration of AO7 were observed from its characteristic absorption band maximum at 484 nm. The decolorization efficiencies of the dyes were estimated by the equation: where C_i and C_t represent the concentration of dye in solution before and after irradiation for time t, respectively. To find out the effect of the pH on the photocatalytic degradation efficiency the pH value of the obtained suspension was adjusted in the range of 3 to 9 using dilute NH₄OH and HNO₃. The effects of calcination temperature and dosage concentration on the decolorization of AO7 were also investigated. To investigate the mechanism of the photodegradation process, different chemical scavengers – isopropanol (90 mM, ˙OH scavenger), CrO₃ (1.5 mM, e⁻ scavenger), sodium oxalate (60 mM, h⁺ scavenger), and 1,4-benzoquinone (2.0 mM, O₂˙⁻ scavenger) were used. To determine the adsorption behavior of the Sm³⁺ ion doped and pure CeO₂ catalysts, adsorption tests in the dark were performed. 20 mg of the photocatalyst was added to 20 mL of a 2 × 10⁻⁴ M AO7 solution. The mixture was well dispersed, and put in the dark for 10 h. The AO7 concentration in the suspension before and after the adsorption test was analyzed and the adsorbed amount of AO7 on the catalysts was calculated.

3. Results and discussion

3.1 Lattice structure and morphologies of the prepared photocatalysts

Fig. 1 shows the XRD patterns of the as-prepared precursor powders of CeO₂:Sm³⁺ (a) and the powders calcined at 500 °C (b), and 800 °C (c) for 2 h. All the peaks of the as-prepared as well as the calcined Sm³⁺-doped CeO₂ nanopowders, in the 20–80° range of the 2θ-value, are indexed to the single crystalline cubic-fluorite structure of pure CeO₂ (JCPDS 34-0394). Further, no diffraction peaks that could be attributed to cerium hydroxide or samarium oxide were observed for any sample. Unlike previous reports for the synthesis of ceria particles at low temperature, neither of these are poorly crystalline, nor contain any appreciable amount of undecomposed precursor.^27,30 The results imply that the present aqueous-based microwave-assisted synthesis route is advantageous for preparing nanocrystalline CeO₂ powders at temperatures as low as 90 °C and reducing the reaction time for the hydrothermal process to only 30 min. These make this aqueous-based route a low-cost and time saving one. Godinho et al. also demonstrated that microwave heating during hydrothermal treatment drastically decreased the required treatment time to 30 min to obtain crystalline gadolinium-doped ceria nanorods at 130 °C.²⁸

With increasing calcination temperature, the peaks become narrower and their intensity increases due to the increase in the crystallinity of the Sm³⁺-doped CeO₂ crystallites. The average crystallite sizes of the Sm³⁺-doped CeO₂ samples were calculated from the X-ray line broadening of the (111), (200), (220) and (311) reflections using Scherrer’s equation (peak width obtained after correction for the instrumental broadening).³¹ The instrumental peak width (β_i) was subtracted from the measured peak width (β_m) at half maximum intensity; i.e. where the subscript “s” indicates the contribution from the sample. The calculated average crystallite sizes were 7.9, 8.3, and 22.0 nm for the as-prepared sample and the samples heated at 500 °C and 800 °C, respectively. It is interesting to observe that a very slow increase of crystallite size occurs in the early stage of calcination from 100 °C to 500 °C. While an activated growth of crystallite size from 8 to 28 nm was observed in pure CeO₂ on increasing the annealing temperature from 100 to 800 °C. This signifies that the Sm³⁺ doping in CeO₂ stabilizes the solid solution against crystal growth during calcination.³²

The lattice parameter (a) value (0.5429 nm) of the as-prepared mesoporous Sm³⁺-doped CeO₂ sample is higher than that of the mesoporous as-prepared CeO₂ sample (a = 0.53929 nm), prepared under the same experimental conditions, but very close to that of bulk cubic CeO₂ (0.5411 nm). Due to the small size of the nanocrystallites, the formation energy of oxygen vacancy is reduced and the two electrons from the oxygen atom are captured by two lattice site Ce⁴⁺ ions and hence favors the formation of reducing Ce⁴⁺ ions to the Ce³⁺ state. Using Kroger–Vink notation, the vacancy formation reaction is represented as The presence of Ce³⁺ ions allows the increase of the lattice parameter due to the fact that the Ce³⁺ ion has a larger ionic radius compared to Ce⁴⁺ and the concentration of the Ce³⁺ ion decreases with increase in calcination temperature.²² Again, the ionic radius of Sm³⁺ (0.1079 nm) is larger than Ce⁴⁺ (0.0971 nm), hence little lattice expansion is expected due to incorporation of Sm³⁺ into the CeO₂ lattice, and the results are in good agreement with previous reports.^21,22 In order to balance the charge compensation, Sm³⁺ additives create oxygen vacancies in the CeO₂ lattice, for example A decrease in the a value to 0.5420 nm was observed when the sample was calcined at 500 °C. This lattice contraction observed in the mesoporous structure (as evidenced from the HRTEM results) may be due to the strains imposed by the pore wall structure, and also the oxidation of Ce³⁺ to Ce⁴⁺ occurring during the calcination process. As expected, on the further increase of the calcination temperature to 800 °C the lattice parameter value decreased to 0.5381 nm due to the increase in the crystallite size. The consistency of the lattice parameter infers that the introduction of Sm³⁺ ions into CeO₂ can stabilize the host lattice at high temperatures and no new phase is formed.

The textural properties of the Sm³⁺-doped CeO₂ photocatalysts were investigated by nitrogen sorption analysis. The nitrogen sorption isotherms and the corresponding pore size distribution (in the inset) of the (a) as-prepared and (b) heated at 500 °C powders are shown in Fig. 2. Both the samples show a type IV isotherm characterized by the presence of a hysteresis loop covering a broad relative pressure range, indicating a mesoporous structure containing some micropores. The surface area and total pore volume of the as-prepared sample were as high as 112 m² g⁻¹ and 0.095 cm³ g⁻¹ respectively, and increased to 135 m² g⁻¹ and 0.16 cm³ g⁻¹ when the sample was calcined at 500 °C. Subsequent thermal treatment not only improves the crystallinity but also maintains a high specific surface area, which is an important criterion for good photocatalysts. The BJH pore size distributions (see the inset of Fig. 2) reveal that the as-prepared sample has a monomodal pore size distribution in the range of 2 to 30 nm, with an average pore size of 5.5 nm, while the sample calcined at 500 °C shows a similar pore size distribution but has a larger average pore size of 6.2 nm, in line with the TEM results. As shown in Table 1, the surface area of the Sm³⁺-doped CeO₂ sample is higher than that of the pure CeO₂ (82 m² g⁻¹ at 500 °C) heat treated at a particular temperature. This may be due to Sm³⁺ doping in the ceria stabilizing the solid solutions and decreasing the crystallite size. Similarly, increases in surface area on doping some trivalent cations such as Fe³⁺ and Sm³⁺ in CeO₂ have also been reported.^18,22

Fig. 2 N₂ adsorption–desorption isotherms (with the BJH pore size distribution plots in the inset) of the CeO₂:Sm³⁺ powders (a) as-prepared and (b) calcined at 500 °C for 2 h.

Table 1 Physical parameters of the CeO₂ and CeO₂:Sm³⁺ photocatalysts calcined at 500 °C and their photodegradation efficiency (%) under solar light irradiation for 1 h

Sample	Crystallite size (nm)	Lattice volume (nm^3)	Surface area (m^2 g^−1)	Pore volume (cm^3 g^−1)	Band gap (eV)	%photodegradation
CeO2	8.6	0.1582	82	0.15	2.86	60
CeO2:Sm^3+	8.3	0.1592	135	0.16	2.79	82

---

On calcining the Sm³⁺-doped CeO₂ material at the higher temperatures of 600 °C and 800 °C, the BET surface area decreased to 70 and 30 m² g⁻¹, respectively. This is because the mesoporous structure collapses due to the structural shrinkage of the skeleton and increased material density during the calcinations. Natile et al. reported a surface area value of 72 m² g⁻¹ for pure CeO₂ calcined at 250 °C and prepared by the microwave-assisted heating of an aqueous solution of [(NH₄)₂Ce(NO₃)₆] containing 1 wt% poly(ethylene glycol) 2000 (PEG) and 1 wt% sodium acetate (NaOOCCH₃).²⁹ These mesoporous Sm³⁺:CeO₂ materials having large pores and high surface areas are expected to show high activity in adsorption and photocatalytical degradation.

The TEM image shown in Fig. 3 further confirms the mesoporous structure of the Sm³⁺-doped CeO₂ calcined at 500 °C for 2 h. It is evident from the HRTEM image in Fig. 3b that the particles were uniform in shape and also crystalline in nature as the lattice fringes could be observed. The estimated particle size was around 8 nm, which is consistent with the XRD results. The selected-area electron diffraction (SAED) patterns (inset of Fig. 3a) consist of single spots superimposed on rings, indicating that the crystal domains within the pore walls are quite small. As microwave heating leads to uniform heating of the whole synthesis mixture, it led to a homogeneous microstructure as is evident from the micrographs. From the EDS spectrum (inset of Fig. 3b), the expected peaks of Sm, Ce and O were observed from the CeO₂:Sm³⁺ sample. The addition of 1.0 mol% Sm₂O₃ into the reaction solution led to the incorporation of 2.2 at% Sm in the CeO₂:Sm³⁺, as confirmed by EDS analysis.

Fig. 3 (a) TEM image, (b) HRTEM image and SAED pattern (inset of a) of the CeO₂:Sm³⁺ powders calcined at 500 °C for 2 h. The EDS of the sample is shown as the inset of (b).

3.2 UV-vis diffuse reflectance spectroscopic studies

UV-vis diffuse reflectance spectroscopic studies have been conducted for various metal oxides to obtain information on the surface coordination and different oxidation states of the metal ions. DR spectra of the Sm³⁺-doped CeO₂ samples as-prepared and calcined at different temperatures are shown in Fig. 4. All the samples showed very broad intense absorption in the UV region (<400 nm) of the spectrum. Careful observation in the region revealed three peaks at 255, 280 nm and 325 nm.³² The absorption of ceria in the UV-vis region is due to the charge transfer transition between O 2p and Ce 4f bonds. The absorption band at 255 nm corresponds to Ce³⁺ variety present which is caused by Ce³⁺ ← O²⁻ in the system.³³ The peak at 325 nm arises due to inter-band transitions in ceria, and the band at 280 nm is assigned to Ce⁴⁺ ← O²⁻charge transfer transitions.³³

Fig. 4 UV-vis diffuse reflectance spectra of CeO₂:Sm³⁺ (a) as-prepared (compared with that of the pure as-prepared CeO₂ in the inset), and the powders calcined at (b) 500 °C, (c) 600 °C, and (d) 800 °C for 2 h.

The band gap energies (E_g) of the samples were calculated by fitting the absorption data to the equation E_g = 1240/λ_{Absorption Edge}. The bandgap energies were calculated to be 2.85 eV for the pure as-prepared CeO₂, and 2.78, 2.79, 2.81, and 2.91 eV for the Sm³⁺:CeO₂ as-prepared, and calcined at 500, 600 and 800 °C, respectively. These values are smaller than the values obtained for the bulk CeO₂ (3.08 eV),³⁴ CeO₂ microspheres (2.81 eV), and CeO₂ microflowers (3.08 eV).³⁵ This shows that Sm³⁺ doping can shift the absorption edge of CeO₂ to the visible light range and narrow the band gap as can be seen from the inset of Fig. 4. Modification of the band gap is expected to improve the efficiency of the utilization of sunlight especially in the visible region. Thus, narrowing of the band gap is beneficial for improving the photoabsorption performance of Sm³⁺-doped CeO₂ and enhances its photocatalytic performance. The redshift of the optical band gap of cerium oxide (CeO_x) films was correlated with the increase of Ce³⁺ content at the grain boundary.³⁶ Chen et al. also observed a blueshift of the band gap due to the valence transition of Ce³⁺–Ce⁴⁺ induced by thermal annealing.³⁷ The quantum-size effect is expected to enhance the band gap of the materials with decreasing particle size due to the higher localization of the energy bands.³⁶ The content of Ce³⁺ and oxygen vacancies from the surface progressively decreased with the growth of the CeO₂ nanoparticles during annealing in an air atmosphere. The decrease of Ce³⁺ concentration and the corresponding decrease of vacancies (defects) content will eliminate some localized defect energy states within the band gap, and thus result in an increase in the band gap. So, the blueshift of E_g in our Sm³⁺:CeO₂ samples with increasing calcination temperatures is due to the Ce³⁺ content decrease, not due to the results of the quantum-size effect. The results also suggest that the Sm³⁺ doping in CeO₂ has induced a higher concentration of Ce³⁺. These defects provided many activity sites on the CeO₂ in the photocatalytic process.¹⁷

3.3 Evaluation of the photocatalytic activity of the samples

AO7 photocatalytic degradation experiments were performed to investigate the photocatalytic activity of Sm³⁺-doped CeO₂ under solar light illumination.

3.3.1 Effect of catalyst dosage. In order to study the effect of the Sm³⁺-doped CeO₂ catalyst dosage on the photodegradation efficiency we carried out experiments varying the amount of nanopowder from 1 to 4 g L⁻¹ using a 2 × 10⁻⁴ M stock solution of AO7 over a period of 1 h of sunlight irradiation, without any further adjustment of the pH of the solutions and at atmospheric conditions. As can be seen from Fig. 5, the photodegradation efficiency increased from 82% to 94% with increasing catalyst dosage from 1 g L⁻¹ to 4 g L⁻¹ at 1 h solar irradiation time. This was due to the increased number of adsorption and active sites available on the catalyst surface, which in turn increased the number of hydroxyl groups and superoxide radicals. The total active surface area increased with increasing catalyst dosage. At the same time, due to an increase in the turbidity of the suspension, there was a decrease in sunlight penetration as a result of the increased light scattering and screening effect and hence the photo-activated volume of the suspension decreased.³⁸ When the catalyst dosage was varied between 1 g L⁻¹ and 2 g L⁻¹, the dye removal performance varied only∼6%. As it is important to keep the treatment expenses low for industrial use and also taking the operational error of the photocatalytic test into consideration, we regarded 1 g L⁻¹ as the optimum catalyst dosage in the present work.

Fig. 5 Effect of CeO₂:Sm³⁺ (calcined at 500 °C) catalyst amount on the UV-vis spectra of AO7 (photodegradation% in the inset) during the decolorization process at a solar irradiation time of 1 h and a neutral pH.

3.3.2 Effect of pH of the medium. The solution pH is an important parameter in photocatalytic degradation reactions and the wastewaters from textile industries usually have a wide range of pH values. pH has a significant influence on the surface charge of the metal oxide catalyst and therefore on the adsorption ability of pollutants onto the surface of the catalyst. A higher adsorption of dye could lead to a higher degree of degradation. The effect of the pH of the medium on the photocatalytic degradation efficiency of AO7 under sunlight irradiation for 10 h was examined in the pH range of 2.9 to 9.1 at a fixed dose of Sm³⁺:CeO₂ (1 g L⁻¹) and keeping all other experimental parameters constant.

Fig. 6 shows the direct influence of the initial solution pH values of 2.9, 4.2, 6.9, and 9.1 on the degradation efficiency of the dye on the catalyst. It is interesting to observe that the Sm³⁺:CeO₂ photocatalyst was very efficient in degrading (99.5% to 95%) the dye over the pH range from 2.9 to 9.1. At a neutral pH of 6.9, the photodegradation of AO7 was 98%. The photodegradation efficiency of AO7 was slightly higher in acidic solutions than in alkaline solutions. The zero point charge of ceria is at pH 4.2, which is close to the neutral pH of 6.9 of the solution. The AO7 dye acts as a Lewis base due to the presence of a negatively charged acidic sulfonate group on it and can easily adsorb on the positively charged Ce–OH₂⁺ catalyst surface in an acidic pH, and hence increase the degradation rate. In an alkaline medium, maybe due to the increased OH⁻ ion concentration with the increased pH, there is a competitive adsorption between hydroxyl groups and the dye molecules. In addition, the negatively charged catalyst surface repels the dye, thereby the degradation is a bit slower.

Fig. 6 Effect of initial pH value on the photocatalytic degradation efficiency of AO7 in 10 h.

3.3.3 Effect of calcination temperature. Calcination temperature is a vital parameter for the preparation of a catalyst and has prominent influence on the activity of the prepared photocatalysts. Fig. 7 shows the effect of temperature ranging from 90 (as-prepared) to 800 °C on the photocatalytic activity under sunlight irradiation for 1 h at normal pH condition. The results revealed that the photocatalytic activity rapidly increased from 37% to 82% as the calcination temperature increased from 90 (as-prepared) to 500 °C and was optimum at 500 °C. Reduced photodegradation efficiencies of 79% and 36% were observed for the catalysts calcined at 600 and 800 °C, respectively. Therefore, the catalyst calcined at 500 °C has the most favorable physical properties for optimal photo-activity.

Fig. 7 Effect of calcination temperature of the CeO₂:Sm³⁺ catalyst on the UV-vis spectra of AO7 (photodegradation% in the inset) during the decolorization process at a solar irradiation time of 1 h and a neutral pH.

Generally, the specific surface area, surface states and crystallinity play key roles in the degradation efficiency.³⁸ The catalyst calcined at 600 °C possesses a lower surface area than the as-prepared catalyst though it shows a prominently 42% higher photodegradation efficiency. This indicates that surface area is not the only determining factor to cause the different efficiencies seen in this study. The as-prepared catalyst also contains a certain amount of SDS molecules, which are attached to the inorganic framework and are difficult to remove by just a simple washing process. The SO⁴⁻ ions of the SDS molecules present in the as-prepared sample also decrease the adsorption of the negatively charged dye on the surface, and thereby decrease the photodegradation efficiency. Carbonaceous species formed during the sunlight irradiation and low crystallinity could be possible reasons for the unusually low activity of the as-prepared catalyst despite its high surface area. Wang et al. found that organic impurities (carbonaceous species) present in the well-known photocatalyst TiO₂ diminished the photocatalytic activities by accelerating the electron and hole recombination by acting as intermediate levels under UV light.³⁹

As the decomposition temperature of SDS is ∼210 °C, the catalyst calcined at 500 °C is surfactant free, as was also evident from IR analysis (not shown here). This calcination temperature is high enough to develop well the crystalline structure of the catalyst. The highest photocatalytic activity of the catalyst calcined at 500 °C may be attributed to the high surface area, pore volume and mesoporous morphology. Also the catalyst can absorb light in the near UV and also absorbs a larger fraction of the solar spectrum than the others. As the temperature reaches above 500 °C, the crystallites grow larger and the surface area and total pore volume go down, leading to the poor performance of the samples. The photocatalytic activity reduction for the catalyst calcined at 800 °C may be attributed to the extensive decrease of the specific surface area and its nonporous morphology.

3.3.4 Effect of irradiation time. Fig. 8 displays the evolution of the UV-vis absorption spectra of the AO7 solution prior to and during the photocatalytic degradation using CeO₂:Sm³⁺ calcined at 500 °C as the catalyst under solar irradiation as a function of irradiation time at natural pH conditions. The absorption spectrum of AO7 has three characteristic absorption peaks at 228, 310, and 484 nm, and two shoulders at 254 nm and 415 nm. The peaks at 228 and 310 nm and the shoulder at 254 nm are attributed to the π → π* transitions of the benzene and naphthalene rings, respectively. The peak at 484 nm and the shoulder at 415 nm correspond to the n → π* transition of the hydrazone form and azo, respectively.¹ In the photo-decoloration process, it was found that the major absorption band at 484 nm declined rapidly until no peak was further observed with increasing irradiation time. This indicates the cleavage of the chromophore (N=N or C–N bond) responsible for the characteristic color of AO7 and the representative photograph (Fig. 9) for the AO7 solution using the catalyst clearly shows the decoloration process of the AO7 solutions along with irradiation time.

Fig. 8 Evolution of the UV-vis spectra with irradiation time for the photocatalytic degradation of AO7 in aqueous solution in the presence of the CeO₂:Sm³⁺ catalyst calcined at 500 °C. pH neutral.

Fig. 9 A photograph of the AO7 solutions in the presence of the CeO₂:Sm³⁺ calcined at 500 °C as the catalyst under solar irradiation at different time intervals (h) as marked therein.

It was observed that after the first 1 h of solar irradiation the absorbance peaks including the shoulder in the visible region decreased by over 82% from their initial values and almost complete decolorization (98%) was achieved after 10 h of photocatalysis. The peaks at 228, 254 and 310 nm decreased significantly. This indicates that the benzene and naphthalene rings of AO7 were broken into smaller species.

3.3.5 FT-IR study on the dye before and after photodegradation. The degradation of the dye solution was further confirmed with FT-IR spectroscopy. Fig. 10 presents the FT-IR spectra of the pure dye (AO7), the AO7 adsorbed over the Sm³⁺:CeO₂ catalyst calcined at 500 °C, and finally after photocatalytic treatment under solar irradiation for 2 and 6 h. The IR spectrum between 1000 and 1900 cm⁻¹ in the fingerprint area of the pure AO7 molecule exhibited several characteristic intense and prominent peaks at 1005, 1035, 1122, 1185, 1210 and 1508 cm⁻¹.⁴⁰ The IR spectrum of AO7 exhibits characteristic bands of aromatic skeletal vibrations at 1620, 1598, 1568, 1553 and 1452 cm⁻¹. The intense band at 1508 cm⁻¹, which is attributed to the vibrations of the –N=N– bond or aromatic ring sensitive to interaction with the azo bond, or to the N–H bending vibration mode δ(N–H) of the hydrazone form of the azo dye, nearly disappears after 6 h of degradation indicating the breakdown of the azo-chromophore. The most striking feature is that after 2 h of degradation, the important band of ν_as(SO₃) at 1185 cm⁻¹ is shifted to 1165 cm⁻¹, while the band of ν_s(SO₃) at 1122 cm⁻¹ is slightly shifted to 1120 cm⁻¹. It is interesting to observe that the two weak bands at 1035 and 1122 cm⁻¹ due to the coupling between the benzene mode and ν_s(SO₃) can still be detected by FTIR, when 94.5% of the dye was already degraded after 6 h of photodegradation. During the photodegradation all the peaks decreased in intensity until they almost disappear after 6 h of solar irradiation, indicating the destruction of the AO7 dye molecule adsorbed over the catalyst. Even after 30 h of irradiation, a certain amount of chemicals containing phenyl and sulfo groups still exist on the surface of the CeO₂ powder prepared by the precipitation method.¹³

Fig. 10 FT-IR spectra of AO7 (a) isolated, (b) adsorbed on the catalyst, and after photodegradation at (c) 2 h and (d) 6 h of irradiation.

3.3.6 Active species to attack dye molecules. Chemical scavengers were employed to understand the role of photogenerated radical species for the photocatalytic oxidation of AO7 over the CeO₂:Sm³⁺ under solar light irradiation. We performed a series of control experiments adding different scavengers: isopropanol for ˙OH, sodium oxalate (SOX) for holes (h⁺), CrO₃ for electrons (e⁻), 1,4-benzoquinone (BQ) for superoxide radicals (˙OOH/O₂˙⁻). As shown in Fig. 11, the e⁻ and ˙OH scavengers did not affect the photodegradation experiments. However, when BQ was added to the reaction system, a significant inhibition effect on the decolorization of the AO7 solution was observed. A remarkable inhibition effect on the photodegradation of AO7 was also observed when sodium oxalate was used as a scavenger for holes. The experiments clearly suggest that h⁺ play a predominant role in the photocatalytic destruction of AO7. It is worth noting that though e⁻ in bulk solution play a negligible role in photodegradation, superoxide radicals appear to play secondary roles in the photocatalytic destruction of AO7.

Fig. 11 Control experiments for the photocatalytic degradation of AO7 with the addition of different radical scavengers: isopropanol (scavenger for hydroxyl radicals), CrO₃ (scavenger for electrons), sodium oxalate (SOX, scavenger for holes), and benzoquinone (BQ, scavenger for superoxide radicals), over the optimum CeO₂:Sm³⁺ calcined at 500 °C under solar light irradiation for 6 h.

A tentative photocatalytic reaction mechanism for the degradation of AO7 over the CeO₂:Sm³⁺ can be proposed as the following. Under solar light irradiation, electron–hole pairs are generated from CeO₂. Part of the photogenerated e⁻ also gets trapped by oxygen vacancies (F centres) before moving to the conduction band.²⁰ Adsorbed surface oxygen first interacts with the Ce³⁺ and oxygen vacancies to generate superoxide radicals and then forms H₂O₂, ˙OOH and ˙OH. Due to the short lifetime of the resulting ˙OH on the photocatalyst, the contribution of the oxidizing species in the bulk solution could be insignificant. Active h⁺ and O₂˙⁻ generated in the bulk solution is accepted to be produced by the following equations:

CeO₂ + hν → h⁺ + e⁻

Charge carrier recombination: h_vb⁺ + e_cb⁻ → heat

Ce⁴⁺ + e⁻ → Ce³⁺

Ce³⁺ + O_2(ads) → Ce⁴⁺ + O₂˙⁻

O_2(ads) + F → O₂˙⁻ + F⁺

H₂O + h⁺ → ˙OH + H⁺

O₂˙⁻ + H⁺ → ˙OOH

˙OOH + H⁺ + e⁻ → H₂O₂

H₂O₂ + e⁻ → ˙OH + OH⁻

h⁺ + dye → degraded product

O₂˙⁻ + dye → degraded product.

The adsorption experiments indicated that AO7 adsorbed strongly on the CeO₂ surface. Two oxygen atoms from the sulfonate group of AO7 are linked with two Ce surfaces and form an inner-sphere complex. AO7 molecules, being strong electron donors due to the oxygen atoms of SO₃²⁻, are able to directly interact with valence band holes. The valence band holes are primarily captured by the adsorbed AO7 molecules rather than by the adsorbed water or hydroxyl groups as the oxidative potential of AO7 (0.76 V versus NHE),⁴¹ is much lower than that of the photogenerated holes. The pathway for photodegradation under solar light as presented in our study is different from the dye self-sensitization pathway implicated during visible light irradiation (λ > 420 nm) using pure CeO₂ as the photocatalyst.¹⁶ Another study reported that active h⁺ play a major role in AO7 degradation using P-25 under UV irradiation.

Kondarides and coworkers reported that it takes about 50 h to totally decolorize AO7 solutions with a concentration of up to 100 mg L⁻¹ by TiO₂ under visible light irradiation.⁴² A Cu₂O/CeO₂ composite photocatalyst (1 g L⁻¹) with a surface area of 53 m² g⁻¹ degraded about 96.2% of AO7 (35 mg L⁻¹) after 4 h of visible light irradiation.¹⁶ When 0.05 g of Co-doped CeO₂ nanocubes (catalyst) was added with 15 mL of AO7 (0.3 mmol) and exposed under UV illumination (λ_max = 365 nm) for 8 h, a specific change in the color of the dye was observed due to its decomposition.⁴³ In particular, it is interesting to evoke some reasons why the mesoporous CeO₂:Sm³⁺ prepared through the microwave route exhibited enhanced solar light photocatalytic activities. The first explanation is that the shifts of the absorption edge of CeO₂ to the visible-light range narrow the band gap and have strong absorption in the UV region as well as definite absorptions in the visible region. This decrease in band gap may help to facilitate the propagation of electrons to the conduction band initiating the photolysis reaction.⁴⁴ Furthermore, the photocatalytic activity can be enhanced by the high crystallinity of these new mesoporous Sm³⁺:CeO₂ catalysts. Mesoporous ceria structures with a high specific surface area, large pore size and thermal stability have several advantages for high adsorption, which is a key factor for the degradation rate and photocatalysis. The mesoporous structure space effectively enhances the spatial dispersion, which results in a higher surface area and also mass transportation of molecules both into and out of the pore structure.⁴⁵ It has been suggested that the degradation of an organic dye in the ceria/H₂O₂ system relies significantly on its adsorption on the surface of the CeO₂.¹⁵ We found that the adsorption effects of AO7 in the dark after 11 h by pure CeO₂ and 1 mol% Sm³⁺-doped CeO₂ are 52.7 and 65%, respectively. The very high adsorption of AO7 by CeO₂ is attributed to the high surface area and pore volume of the mesoporous catalysts, high oxygen vacancies on the surface originating from the facile Ce³⁺/Ce⁴⁺ redox cycle,¹⁵ and to the ability of cerium ions to form complexes with the electron-rich sulfonate group (SO₃⁻) of AO7.¹³ Xiao et al. also demonstrated that Sm³⁺-TiO₂ had a higher methylene blue adsorption capacity than undoped TiO₂.⁴⁶ Sm³⁺ doping further improved the concentration of Ce³⁺ or, in other words, the concentration of oxygen vacancies in CeO₂ samples as well as enhanced the adsorption capacity of AO7, which resulted in better catalytic performances. Tang et al. reported that the oxygen vacancies can not only act as impurity levels in the band structure of ZnO but can also work as electron traps to accept the photogenerated electrons.⁴⁷ Oxygen vacancies on the CeO₂ catalyst surface can easily trap the photogenerated electrons and efficiently benefit the separation of excited electron–hole pairs and thus can play a more functional and effective role in photocatalytic applications.

In our study, the optimum Sm³⁺ concentration in the CeO₂ was 1.0 mol%, at which the recombination of photo-induced electrons and holes could be effectively inhibited, and thereby the highest photocatalytic activity was achieved. At natural pH conditions, the photodegradation of AO7 under 1 h of solar light irradiation by the CeO₂ photocatalysts (calcined at 500 °C) with 0, 0.5, 1.0, 2.0 and 5.0 mol% Sm³⁺ doping were 60, 66, 82, 78, and 72%, respectively. It is well documented that a higher adsorption capacity with a higher Sm³⁺ content does not lead to a higher photocatalytic activity in TiO₂.⁴⁶ Cai et al. also showed that doping Fe³⁺ improved the defect concentration and thus favored the catalytic activity of CeO₂ at low levels, but retarded the degradation of AO7 at high levels, with an optimal Fe/Ce ratio of 1/100.¹⁸ Probably at higher Sm³⁺ concentrations the Sm may be present at the surface of particles and form clusters, which are detrimental for photocatalytic activity carried out at the surface.⁴⁸

4. Conclusions

In summary, mesoporous Sm³⁺-doped CeO₂ photocatalysts with high crystallinity and high surface area were successfully synthesized by a facile and low-cost surfactant-assisted microwave heating method using (NH₄)₂Ce(NO₃)₆, Sm₂O₃ and SDS. Structural, spectroscopic, and electron microscopy techniques were used to characterize the photocatalysts. The resultant Sm³⁺-doped CeO₂ exhibits excellent photocatalytic activities for the degradation of Acid Orange 7 (AO7) under natural sunlight irradiation in the absence of any oxidising agents and over a broad range of pH values from 2.9 to 9.1. The results indicated that the degree of AO7 degradation was influenced by the calcination temperature and the amount of photocatalyst. The main features that are responsible for making CeO₂:Sm³⁺ a promising material for use as an active solar light sensitive photocatalyst are its specific surface area, crystallinity, enhanced dye adsorption capacity, importantly, high UV-vis absorption capability and the reduced band gap of the material due to the presence of a higher concentration of Ce³⁺ in the CeO₂. It was found that besides holes, which are the main active species in the photocatalytic destruction of AO7, O₂˙⁻ played secondary roles in the photocatalytic degradation process. This new material is a promising candidate as a robust solar light active photocatalyst over the broad pH range of dye polluted water without the addition of any oxidizing agents.

Acknowledgements

We gratefully thank the Council of Scientific and Industrial Research, India, for financial support.

References

1. F. S. Freyria, B. Bonelli, R. Sethi, M. Armandi, E. Belluso and E. Garrone, J. Phys. Chem. C, 2011, 115, 24143.
2. A. Özcan, M. A. Oturan, N. Oturan and Y. Şahin, J. Hazard. Mater., 2009, 163, 1213.
3. V. K. Gupta, A. Mittal, V. Gajbe and J. Mittal, Ind. Eng. Chem. Res., 2006, 45, 1446.
4. G. Neri, A. Pistone, C. Milone and S. Galvagno, Appl. Catal., B, 2002, 38, 321.
5. B. Qiu, M. Xing and J. Zhang, J. Am. Chem. Soc., 2014, 136, 5852.
6. Y.-G. Peng, J.-L. Ji, Y.-L. Zhang, H.-X. Wan and D.-J. Chen, Environ. Prog. Sustainable Energy, 2013, 33, 123.
7. J. B. Joo, I. Lee, M. Dahl, G. D. Moon, F. Zaera and Y. Yin, Adv. Funct. Mater., 2013, 23, 4246.
8. P. V. Nidheesh, R. Gandhimathi and S. T. Ramesh, Environ. Sci. Pollut. Res., 2013, 20, 2099.
9. G. Moon, W. Kim, A. D. Bokare, N. Sung and W. Choi, Energy Environ. Sci., 2014, 7, 4023.
10. S. Shibata, T. Suenobu and S. Fukuzumi, Angew. Chem., Int. Ed., 2013, 52, 12327.
11. P. Ji, J. Zhang, F. Chen and M. Anpo, J. Phys. Chem. C, 2008, 112, 17809.
12. T. Feng, X. Wang and G. Feng, Mater. Lett., 2013, 100, 36.
13. P. Ji, J. Zhang, F. Chen and M. Anpo, Appl. Catal., B, 2009, 85, 148–154.
14. D. Sun, M. Gu, R. Li, S. Yin, X. Song, B. Zhao, C. Li, J. Li, Z. Feng and T. Sato, Appl. Surf. Sci., 2013, 280, 693.
15. F. Chen, X. Shen, Y. Wang and J. Zhang, Appl. Catal., B, 2012, 121–122, 223.
16. S. Hu, F. Zhou, L. Wang and J. Zhang, Catal. Commun., 2011, 12, 794.
17. S. Yuan, Q. Zhang, B. Xu, Z. Jin, Y. Zhang, Y. Yang, M. Zhang and T. Ohno, RSC Adv., 2014, 4, 62255.
18. W. Cai, F. Chen, X. Shen, L. Chen and J. Zhang, Appl. Catal., B, 2010, 101, 160.
19. A. L. Gal and S. Abanades, J. Phys. Chem. C, 2012, 116, 13516.
20. V. V. Ursaki, V. Lair, L. Żivković, M. Cassir, A. Ringuede and O. Lupan, Opt. Mater., 2012, 34, 1897.
21. N. Sutradhar, A. Sinhamahapatra, S. Pahari, M. Jayachandran, B. Subramanian, H. C. Bajaj and A. B. Panda, J. Phys. Chem. C, 2011, 115, 7628.
22. Y.-P. Fu, S.-B. Wen and C.-H. Lu, J. Am. Ceram. Soc., 2008, 91, 127.
23. J. Chen, S. Patil, S. Seal and J. F. McGinnis, Nat. Nanotechnol., 2006, 1, 142.
24. X. Liu, K. Zhou, L. Wang, B. Wang and Y. Li, J. Am. Chem. Soc., 2009, 131, 3140.
25. Y. Q. Wang, H. M. Cheng, Y. Z. Hao, J. M. Ma, W. H. Li and S. M. Cai, Thin Solid Films, 1999, 349, 120.
26. P. Borker and A. V. Salker, Mater. Chem. Phys., 2007, 103, 366.
27. C.-Y. Cao, Z.-M. Cui, C.-Q. Chen, W.-G. Song and W. Cai, J. Phys. Chem. C, 2010, 114, 9865.
28. M. Godinho, C. Ribeiro, E. Longo and E. R. Leite, Cryst. Growth Des., 2008, 8, 384.
29. M. M. Natile, G. Boccaletti and A. Glisenti, Chem. Mater., 2005, 17, 6272.
30. H. Wang, J.-J. Zhu, J.-M. Zhu, X.-H. Liao, S. Xu, T. Dinga and H.-Y. Chen, Phys. Chem. Chem. Phys., 2002, 4, 3794.
31. M. Tabatabee, R. Shaikhalishahi and R. Mohammadinasab, Res. Chem. Intermed., 2015, 41, 113.
32. D. Jampaiah, K. M. Tur, S. J. Ippolito, Y. M. Sabri, J. Tardio, S. K. Bhargava and B. M. Reddy, RSC Adv., 2013, 3, 12963–12974.
33. M. Guo, J. Lu, Y. Wu, Y. Wang and M. Luo, Langmuir, 2011, 27, 3872.
34. C. Ho, J. C. Yu, T. Kwong, A. C. Mak and S. Lai, Chem. Mater., 2005, 17, 4514.
35. H. Wu and L. Wang, Catal. Commun., 2011, 12, 1374.
36. P. Patsalas, S. Logothetidis, L. Sygellou and S. Kennou, Phys. Rev. B: Condens. Matter Mater. Phys., 2003, 68, 35104.
37. M. Y. Chen, X. T. Zua, X. Xianga and H. L. Zhang, Phys. B, 2007, 389, 263.
38. U. G. Akpan and B. H. Hameed, J. Hazard. Mater., 2009, 170, 520.
39. J. Wang, J. Wang, Q. Sun, W. Wang, Z. Yan, W. Gong and L. Min, J. Mater. Chem., 2009, 19, 6597.
40. C. Bauer, P. Jacques and A. Kalt, Chem. Phys. Lett., 1999, 307, 397.
41. K. Vinodgopal and P. V. Kamat, Environ. Sci. Technol., 1995, 29, 841.
42. M. Stylidi, D. I. Kondarides and X. E. Verykios, Appl. Catal., B, 2004, 47, 189.
43. N. S. Arul, D. Mangalaraj, P. C. Chen, N. Ponpandian and C. Viswanathan, Mater. Lett., 2011, 65, 3320.
44. J. Schneider, M. Matsuoka, M. Takeuchi, J. Zhang, Y. Horiuchi, M. Anpo and D. W. Bahnemann, Chem. Rev., 2014, 114, 9919.
45. H. Song, G. Zhou, C. Wang, X. Jiang, C. Wu and T. Li, Res. Chem. Intermed., 2013, 39, 747.
46. Q. Xiao, Z. Si, J. Zhang, C. Xiao and X. Tan, J. Hazard. Mater., 2008, 150, 62.
47. Y. Tang, H. Zhou, K. Zhang, J. Ding, T. Fan and D. Zhang, Chem. Eng. J., 2015, 262, 260.
48. D. de la Cruz, J. C. Arevalo, G. Torres, R. G. Bautista Margulis, C. Ornelas and A. Aguilar-Elguezabal, Catal. Today, 2011, 166, 152.

---