Heterogeneous photocatalysis has become an alternative treatment method for degradation of organic pollutants from wastewater, which has the ability to mineralize organic compounds. In fact, photocatalysis can greatly contribute to the remediation of those environmental pollutants into environmental friendly species: CO₂, H₂O or other oxides, halide ion, phosphate (Sakthivel et al., 2005; Brijesh et al., 2008; Gomathi et al., 2009). The semiconducting metal oxides like iron oxide (Fe₂O₃) nanomaterials exhibit promising photocatalytic activities due to their environmental friendly behavior, low catalyst cost, high specific surface area, high crystallinity and solar energy application (Leland and Bard, 1987; Zhao et al., 2009; Bharathi et al., 2010) and thus, could be an alternative material for environmental application and wastewater treatment (Rhoton et al., 2002; Lei et al., 2006, 2007; Wang et al., 2009). However, the photocatalytic activity of the iron oxide is depending on the particle size, which is difficult to synthesize nano-sized iron oxide by conventional method and to control its crystal size in the photocatalyst (Park et al., 2010). This is due to the agglomeration of nano-particles in the aqueous solution, which causes the reduction of photocatalytic efficiency. This therefore warrants some smart strategy for use of iron oxide catalyst. One certain way to overcome this drawback is to apply innovative synthetic method of iron oxide nanoparticles of the catalysts that can be easily dispersed in organic medium and homogenously loaded on to the supported materials.

Many studies have continuously tried to improve the photocatalytic activity of iron oxide by coupling of different semiconductor oxide nanoparticles (Baldrian et al., 2006; Laat and Le, 2006; Kitis and Kaplan, 2007; Zelmanov and Semiat, 2008). Thus Al₂O₃/Fe₂O₃ supported photocatalyst has shown a better photocatalytic activity (El-Moselhy, 2009; Li et al., 2010; Pattanayak, 2010; Jia-feng et al., 2011). These new composite allows for the decreasing of the particle size of the catalyst (Karakassides and Gournis, 2003). It is an important material for the application of heterogeneous photocatalysis (Cordier et al., 2006). However, the photocatalytic activity of Al₂O₃/Fe₂O₃ nanocomposite is still not satisfactory for degradation of organic pollutants in wastewater because of its low electron-hole separation. This may be due to similar ionic radii and same charge of the materials. Subsequently complete substitutions occur between aluminum (III) oxide and iron (III) oxide nanoparticles and no structural defect could be expected (Neiva et al., 2009). Currently, nano-manganese (IV) oxide (MnO₂) has received a great potential application in environmental protection and degradation of organic pollutants in wastewater (Mitta et al., 2009; Su, 2010). It is a promising material for heterogeneous photocatalyst as a new generation environmental friendly catalyst. This is due to its high specific surface area, crystallinity, ability to disintegrate water molecule into hydrogen and hydroxide ions (Kang et al., 2007; Chu and Zhang, 2009). Based on these advantages MnO₂ semiconductor was chosen as an assistant component of photocatalyst to improve the photocatalytic activities of binary Al₂O₃/Fe₂O₃ nanocomposite. Therefore, in this study the triple oxide photocatalyst of MnO₂/Al₂O₃/Fe₂O₃ was used in photodegradation of MG dye in aqueous solution by using atomic absorption spectroscopy (AAS), X-ray diffraction (XRD), Fourier transform infrared (FTIR) and UV/Vis spectrometric techniques. A schematic representation of the photocatalytic process is represented in Figure 1.

 

Synthesis of photocatalyst

The MnO₂/Al₂O₃/Fe₂O₃ ternary mixed nanoxide powder was prepared by sol-gel method. The sol corresponded to total volume ratio of metal, butanol, deionized water and nitric acid ratio of 1:20:4:0.1. In each case, ferric nitrate nanohydrate and aluminum nitrate nanohydrate were dissolved in stoichiometric amounts of water, 69% HNO₃ and butanol then mixed with vigorous stirring. KMnO₄ and Na₂S₂O₃.5H₂O were mixed with each other and dissolved in double dionized water to form dark-brown MnO₂ solution. Subsequently, the dark brown MnO₂ solution was added drop wise into the mixture of aluminum oxide/iron oxide solution through stirring for 30 min at room temperature. The prepared sol was left to stand for the formation of gel. After the gelation was completed, the gel was aged for 5 days at room temperature and sample was dried at 75°C for 36 h. After grinding the dried samples, they were calcined at 400 and 600°C for 3 h at the different molar ratios. The different molar ratios of MnO₂:Al₂O₃:F₂O₃ are summarized in Table 1.

 

 

Characterization

Elemental composition

The elemental compositions of the as-synthesized powder were analyzed by flame atomic absorption spectrophotometer. 0.01 g of the as-synthesized powders were digested with conc. HNO₃ (7 ml), conc. HCl (4 ml) and H₂O₂ (2 ml) using acid digestion tube till clear solution appeard. The samples of Fe (III) and Mn (IV) solutions were transferred to 100 and 50 ml volumetric flasks and brought to volume using de-ionized water; triplicate solutions of iron and manganese were read from 100 and 50 ml volumetric flasks, respectively. Stock standard solutions of 1000 mg/L Fe and Mn were prepared by dissolving 0.14 and 0.36 g of corresponding salts in de-ionized water, respectively and series standard solutions were prepared to plot the calibration curves of the metals by appropriate dilution.

X-ray diffraction study

In order to determine the crystal phase composition of as-synthesized photocatalysts, all powder catalysts were ground to fine particles and analyzed by a BRUKER D8 Advance XRD, AXS GMBH, Karlsruhe, West Germany X-ray diffractometer (XRD) equipped with a Cu target for generating a Cu Kα radiation (λ = 0.15406 nm) at GSE. The accelerating voltage and the applied current were 40 kV, 30 mA, respectively. The instrument was operated under step scan mode with step time and degree (2θ) of 1 s and 0.020°, respectively for the range of 4 to 64°.

UV-Visible diffuse absorbance

For the estimation of absorption edge of the as-synthesized photocatalyst, UV-Visible diffuse absorption was measured using SP65 spectrophotometer at Addis Ababa University research laboratory. As-synthesized sample was dissolved in hot methanol then the UV-Visible diffuse absorbance of the powder was recorded at 200 to 800 nm range to determine the band gap of the photocatalyst.

Fourier transform infrared study

As-synthesized nanopowder was characterized using FTIR (SHIMIDAZU) instrument. Ten mg (dry mass) of the photocatalyst was thoroughly mixed with 100 mg (dry mass) of KBr and ground to a fine powder. A transparent disc was formed by applying a pressure in moisture-free atmosphere. The IR absorption spectrum was recorded between 400 and 4000 cm^-1.

Photocatalytic degradation studies

Photocatalytic activities of the as-synthesized powder were evaluated by decolorization of malachite green dye in aqueous solution. The experiments were carried out in the presence of UV and Visible light irradiation without any catalyst (blank), with catalyst in dark and in the presence of MnO₂/Al₂O₃/Fe₂O₃ photocatalyst. The photocatalytic reactor consists of a Pyrex glass beaker with an inlet tube for provision of air purging during photocatalysis and outlet tube for the collection of samples from the beaker as shown in Figure 2. Reaction was set up by adding 0.15 g of the as-synthesized powder into 100 ml of MG solution (25 mg/L) in the Pyrex glass beaker of 250 ml volume and the suspension was magnetically stirred in dark for 30 min to obtain adsorption/desorption equilibrium before irradiating the light in the beaker. Before illumination of the samples by UV or visible radiations, air/oxygen was purged into the solution with the help of a porous tube at hand purging in order to keep the suspension of the reaction homogenous. During the reaction, the solution was maintained at room temperature and the distance of the lamp from the solution was 9 cm and its intensity was recorded to be 8.25 MW/cm². Then, the light source was activated; 10 ml of the sample was withdrawn at 20 min time interval over irradiation time for 180 min. The suspension was centrifuged at 3000 rpm for 10 min and filtered to remove the catalyst particles before measuring absorbance. The absorbance of the clear solution was measured at a λ_max of 620 nm for quantitative analysis. The UV lamp with a definite power 12 W, 230 V and 50 Hz frequency was employed as UV light source, and incandescent bulb was used as visible light source with a definite power of 40 W, 220 V and 60 Hz frequency. Percentage degradation of MG dye was calculated using the following relation:

Where: A₀ is absorbance of dye at initial stage, A_t is absorbance of dye at time t.

 

 

Characterization

Elemental composition

Elemental compositions of the as-synthesized samples were analyzed through flame atomic absorption spectrophotometer. Triplicate sample solutions were carried out together with blank solution on each sample as shown in Table 2. The standard solutions of iron and manganese metals were analyzed and the calibration curves were plotted on standard solution versus absorbance. Accordingly, the percentage composition of iron and manganese oxide analyzed by AAS was similarly to those expected from the initial concentration of the solution that was measured during the synthesis.

 

 

X-ray diffraction analysis

The crystallinity of the Al₂O₃/Fe₂O₃ and MnO₂/Al₂O₃/Fe₂O₃ photocatalysts synthesized by the sol-gel method at different molar ratios were studied by XRD after calcination at 400 and 600°C. The broad diffraction peaks presented in Figures 3 and 4 clearly revealed that all as-synthesized samples were nanosized crystals with a rhombohedral structure of hematite. Similarly, weak diffraction peaks presented in all as-synthesized samples at 2θ values of 24.1, 35.6, 40.8, 49.3, 54.0 and 62.3 were due to Fe₂O₃, and these correlated with the reported data of hematite (Park et al., 2010). However, the XRD data did not show any presence of Al₂O₃ and MnO₂ particles in the ternary mixed oxide system which confirmed that Al (III) and Mn (IV) enter into the Fe₂O₃ lattice substitution. In fact, ionic radius of Mn (IV) of 53 pm and Al (III) of 54 pm is similar to that of ionic radius (55 pm) of Fe (III) thus the substitution in the matrix of Fe₂O₃ is a favorable process. As a result, Al (III) ions and Mn (IV) ions can substitute Fe (III) lattice site to form a stable solid solution. The average crystallite sizes of the as-synthesized powders were calculated using Debye Scherer equation (Yan et al., 2010):

 

 

Where D is the crystal size (nm), λ is X-ray wavelength corresponding to the Cu target Kα irradiation (0.15406 nm), and β is full width at half maximum (FWHM) of the peak in radian, K = 0.90 is a constant coefficient of spherical shape and θ is the (Bragg angle) corresponding to the diffraction angle. The calculated average crystallite sizes (D) of the photocatalysts are given in Tables 3 and 4.

As shown from the tables, all the as-synthesized powders were nanosized crystal and sample T₃ of Mn-1.66/Al-5.64/Fe-30.38 calcined at 400°C has the smallest crystalline size and hence a high surface area. This sample was selected for UV-Visible and FTIR characterization and photodegradation experiment of malachite green.

 

 

 

FTIR analysis of photocatalysts

The FTIR spectra of the nanocomposite MnO₂/Al₂O₃/Fe₂O₃ photocatalyst prepared by sol-gel method before and after calcination are presented in Figures 5 and 6. The intense band at 3384 cm^–1 may be due to the stretching modes of -OH group from adsorbed water in the sample. The band observed at 1631 cm^-1 can be assigned to the bending vibration of free water molecule; while bands observed at 2344, 1448, 1384 and 1307 cm^-1 may be attributed to -CH, -CH₂ and -CH₃ functional groups. Sharp peaks at 1196, 1087 and 1004 cm^-1 may be due to C-O stretching vibration of primary alcohol, which was used to support the sol gel synthesis. The absorption peak observed at 628 cm^-1 could be associated with the presence of Fe-O bond in the catalyst structure. The band at 511 cm^-1 may indicate the presence of iron oxide in MnO₂/Al₂O₃/Fe₂O₃. Broad band at 3440 cm^-1 may be due to the stretching vibration of -OH from adsorbed water in the crystal sample. The band observed at 1623 cm^-1 could be assigned to the bending vibration mode of free water molecule. Weak bands observed at 2925, 2850 and 1384 cm^-1 may be due to C-H stretching vibrations of primary alcohol. The peaks at 486 and 553 cm^-1 may be attributable to the Fe-O vibration bond of hematite in the MnO₂/Al₂O_3/Fe₂O₃ nanocomposite of rhombohedral structure (Li et al., 2007).

 

 

 

Generally, from the FTIR spectra shown in the Figure 6, it can be concluded that intensities of most of the vibration peaks have eventually decreased during the conversion process of the precursors into the MnO₂/Al₂O₃/Fe₂O₃ photocatalyst. This change may be attributed to the decomposition of organic substances present before the calicination and the formation of crystalline phase during the process of heat treatment. As a consequence, only the vibration band corresponding to a Metal-Oxygen bond could be detected below 700 cm^-1 (Torres-Martinez et al., 2010).

 

UV-Visible diffuse absorption edge

The optical absorption property of the material and migration of the light induced electrons and holes are considered as the key factor controlling the photocatalytic reaction. Of course, these migrations are relevant to the change in the electronic structure and characteristic of the material. The UV-Vis diffuse reflectance spectra of the as-synthesized MnO₂/Al₂O₃/Fe₂O₃ nanopowder obtained by the sol gel method and calcined at 400°C for 3 h was subjected to strong photoabsorption in the visible light region as shown in Figure 7. On the basis of this, the band gap energy (E_g) of the material can be calculated using the following formula (Chien-Tsung, 2007):

 

 

 

The higher surface area also facilitates the absorption of dye molecules on the surfaces of MnO₂/Al₂O₃/Fe₂O₃ photocatalyst. Under visible light irradiation, MG molecules are absorbed on the surfaces of nanocomposite and produced electrons. These electrons are captured by the surface adsorbed O₂ molecules to yield O₂•− and HO₂• radicals, which makes more chance to touch with dye molecules and giving a faster reaction speed then, the MG molecules could be mineralized in time by the super oxide radical ions. Therefore, it can be concluded that the smaller crystalline size of nanocomposite are favorable for the reduction of O₂ and oxidation of H₂O molecules by trapping electrons and holes, which improves the photocatalytic activity of the nanocomposites photocatalyst under visible light region.

The second reason could be the presumption that some of the Al (III) ions and Mn (IV) ions in the photocatalyst might be substituted by Fe (III) lattice site to form a stable solid solution. Such replacements of Al (III) and Mn (IV) ions by the Fe (III) ions can obviously create a charge imbalance (Ali Ismail, 2005). The charge imbalance must be satiated as Mn (IV) ions were reduced to Mn (III) ions. The existence of the Mn (IV) ions can thus inhibit the recombination of the photogenerated electron-hole pairs.

As a result, more OH^- ions would be adsorbed on the surface of the catalyst to overcome the charge balance. These OH^- ions present on the surface of the photocatalyst would accept the holes that are generated by the visible light illumination and thus converted to hydroxyl radicals (•OH), the radicals so formed can oxidize very easily the MG dye solution. On the other hand, the same visible light irradiation could be able to create oxygen vacancies on the structure of the photocatalyst structure. The oxygen vacancies formed as such can trap the electrons that are excited from the VB to the CB and thereby converted into superoxide radicals (O•₂⁻) (Zhang et al., 2009). These O^•₂⁻ radicals are active enough to promote the oxidation of organic substances and effectively inhibited the electron-hole pair recombination. Therefore, these results show that the composite nanomaterial has a good photocatalytic performance in the degradation of MG dye in aqueous solution under visible light irradiation. The photodegradation values of MG dye as a function of time under UV and Visible light irradiations and without light irradiation are presented. Plot of percentage degradation as a function of time under UV, Visible and without irradiation are also shown in Figure 9. 

 

 

Kinetic studies of photocatalytic degradation of MG

The adsorption of MG under without irradiation and its photocatalytic degradation under UV and Visible irradiations follow the pseudo first- order reaction kinetics expressed by the equation:

Where k is the reaction rate constant, C₀ is the initial concentration of MG solution and C_t is the concentration of MG solution at the reaction time t. The linear plot of lnC_o/C_t versus irradiation time t is shown in Figure 10 and values of rate constant (k) as a function of time is presented in Table 5. The calculated adsorption rate constant for MG using MnO₂/Al₂O₃/Fe₂O₃ photocatalyst without irradiation is 4 × 10^-4 min^-1. The calculated rate constants for the degradation of this dye under the UV and visible light irradiations using the same catalyst were found to be 6.56 × 10^-3 and 1.0 × 10^-2 min^-1, respectively. Determination coefficients (R²) of the pseudo-first order reaction without light irradiation, under UV and visible light irradiations were found to be 0.856, 0.976 and 0.995, respectively.

 

 

 

Comparison of binary and ternary photocatalytic activities

In order to explore the photocatalytic activity of the binary and ternary mixed oxides, various studies had been carried out to search ideal semiconductor photocatalyst for degradation of organic pollutants. The result is shown in Table 6. It can be seen that the photocatalytic activities of the ternary mixed oxide photocatalyst displays significantly high degradation ability for organic pollutants in aqueous solution. The improvement may be explained in terms of synergetic effect on the specific adsorption property and the efficient electron-hole separation at the coupled ternary mixed oxides photocatalyst interface. However, up till now no MnO₂/Al₂O₃/Fe₂O₃ coupled ternary photocatalyst has been reported for degradation of organic pollutants. These nanocomposite photocatalyst were found to be more photoeffective than binary Al₂O₃/Fe₂O₃ photocatalyst for oxidation of organic compound.

 

New composite nanosized MnO₂/Al₂O₃/Fe₂O₃ photocatalyst with a rhombohedral structure and oxygen vacancies was successfully synthesized by the sol-gel method by using metal salts as a starting materials. The as-synthesized materials were characterized by AAS, XRD, FTIR and UV/Visible spectroscopic techniques. The XRD pattern suggested that the as-synthesized nanopowders have a crystal structure with a good nonosized range of 20 to 26 nm particle sizes. Smallest particle size of catalyst was taken for characterization and photodegradation study of MG dye. Its band gap energy was about 1.97 eV, which is an important band gap for improving photocatalytic degradation of organic dyes in the visible region. The FTIR spectra before and after calcination are shown as -OH and Fe-O vibration bond in the as-synthesized photocatalyst calcinied at 400°C. Photocatalytic degradation of malachite green in aqueous solution was carried out under different conditions. It was included in the absence of the catalyst, under both UV and visible light irradiations (in blank), without light irradiation (in dark), and under both UV and visible light irradiations in the presence of the photocatalyst. The experimental results showed that higher decolorization efficiency was obtained under visible light irradiation in the presence of the catalyst. This may be due to the larger specific surface area, formation of oxygen vacancies and mixed valence states of manganese in the structure of the MnO₂/Al₂O₃/Fe₂O₃ photocatalyst.

The authors have not declared any conflict of interests.