Abstract

Bifunctional films of Keggin-type polyoxometalates H3PW12O40 (PW12), H4SiW12O40 (SiW12), H3PMo12O40 (PMo12), and TiO2 were successfully built on Fe3O4@SiO2 microspheres using a layer-by-layer method. The characterization by field emission scan electronic microscopy (FESEM) and energy dispersive X-ray spectroscopy (EDX) shows that TiO2 nanoparticles and polyoxometalate (POM) anions are successfully assembled. The photodegradation of methyl orange (MO) was used to test the photocatalytic efficiency of magnetic catalysts under UV irradiation. For MO decomposition, multilayer films that combine PW12 and SiW12 with TiO2 show high efficiency, which can be attributed to the synergistic effect between POMs and TiO2. The degradation of the model contaminant was also systematically checked under different conditions such as bilayer number deposited on magnetic microspheres, catalyst concentration, inorganic oxidants, and salts. The oxidation process of MO follows an apparent first-order reaction. Furthermore, the composite catalysts deposited on Fe3O4@SiO2 magnetic microspheres can be conveniently, quickly, and efficiently separated by an external magnet from a solution.

1. Introduction

The production and use of synthetic dyes contribute to a massive pollution problem as a result of their high toxicity, slow biodegradation, and potential carcinogenicity [1]. To solve environmental problems caused by dye pollutants, various advanced methods have been adopted. Among these methods, photocatalytic oxidization technology has attracted growing interest on account of its energy saving, highly efficient, nonselective, and environmentally friendly characteristics. Its origin can date back to the 70s of the previous century in the field of wastewater treatment [2]. During the past decades, the nanostructured TiO2 photocatalyst aroused much interest [3]. However, the fast recombination of photogenerated electrons and holes results in a decrease of photocatalytic efficiency and restricts the practical application of TiO2.

In contrast to TiO2, polyoxometalates (POMs) are often used as homogeneous photocatalysts in recent years [4, 5]. POMs are easily soluble in water and difficult to recycle from the reaction system for reuse. In order to overcome these shortcomings, heterogeneous catalysts containing POMs have been prepared by immobilizing POMs on solid supports such as functional microspheres [68], indium-tin oxide (ITO) [9], glassy carbon electrode [9, 10], and quartz slides [9, 11].

In recent years, significant attention was focused on the functional thin films containing POMs prepared by a layer-by-layer (LbL) self-assembly method. These films have potential applications in the fields of photoelectrochemistry [12, 13], photocatalysis [7, 1416], electrocatalysis [17, 18], photochromism [18], sensors [9, 19], and electrochromism [18, 20]. Compared with films assembled by single building blocks, composite films are employed for high performance. Sun et al. [12] reported that the incorporation of CdS nanoparticles into Dawson-type polyoxometalate (P2W18) films improved remarkably the photocurrent response and power conversion efficiency. Novais et al. [17] found that the {MG/Co(PW9)2}n multilayer films exhibited excellent electrocatalytic performance for nitrite and iodate reduction with very low detection limits. (PW12-TH) multilayer films deposited on the magnetic substrate of Fe3O4@SiO2 microspheres showed a better ability in terms of photocatalytic activity, stability, and reusability than quartz slides [7]. In particular, films can be efficiently separated by a magnet. The Bi2O3 nanoparticles assembled into [SiW9V3O40]7− endowed the films with specific properties. This film has a tendency toward electrocatalytic L-cysteine and nitrite as well as enhanced electrochromic and photochromic properties [18]. Moreover, the multifunctional films (PV6Mo6/MV) (PV6Mo6 = (NH4)5H4PV6Mo6O40·6H2O, MV = methyl violet) exhibited electrochemical activity for the reduction of NO2, BrO3, and ClO3 and remarkable antibacterial properties toward Escherichia coli[21]. However, less attention was paid to the photocatalytic performance of (TiO2/POM) composite films coated on Fe3O4@SiO2 magnetic microspheres.

In this work, (TiO2/POM) composite films with magnetic and photocatalytic performance were deposited on Fe3O4@SiO2 core-shell microspheres by the LbL method. MO was used as the model pollutant to investigate the photocatalytic properties of bifunctional films. The influence of bilayer number deposited on magnetic microspheres, catalyst concentration, inorganic oxidants, and salts on the degradation of the methyl orange solution was studied in detail.

2. Experiments

2.1. Materials

The Fe3O4 magnetic microspheres were synthesized following the procedure reported by Sun et al. [22]. Silica-coated magnetic microspheres were prepared through the sol-gel method [7]. Titanium dioxide (TiO2) colloids were prepared according to literature [23]. H3PW12O40·xH2O (PW12), H4SiW12O40·xH2O (SiW12), and H3PMo12O40·xH2O (PMo12) were obtained from Sinopharm Chemical Reagent Co. Ltd, China. Methyl orange was purchased from Jining Chemical Engineering Research Institute. Commercial polyelectrolytes, poly(sodium 4-styrenesulfonate) (PSS, Mw ∼ 70,000) and poly(allylamine hydrochloride) (PAH, Mw ∼ 70,000) were obtained from Sigma-Aldrich Inc., USA. Perchloric acid was purchased from Shanghai Jinlu Chemical Co. Ltd, China. Sodium chloride and sodium sulfate were provided by Tianjin Kemiou Chemical Reagent Co. Ltd. Potassium chlorate, potassium bromate, and potassium iodate were obtained from Tianjin Kemiou Chemical Reagent Co. Ltd. Magnesium chloride, aluminium chloride, and sodium hydrogen phosphate were supplied by Aladdin Chemistry Co. Ltd. All reagents were used as received.

2.2. Buildup of (TiO2/POM) Composite Films on Magnetic Microspheres

In order to prepare (TiO2/POM) composite films, 0.2 g Fe3O4@SiO2 core-shell microspheres were first immersed in a poly(allylamine hydrochloride) (PAH) and poly(sodium 4-styrenesulfonate) (PSS) aqueous solution for 15 min to gain a surface with uniform charge. Then, the powders were magnetically separated and washed three times with pH 2.0 HClO4 aqueous solution. The concentration of PAH and PSS aqueous solution was fixed as 2.0 g/L containing 0.5 mol/L NaCl. The photocatalytic (TiO2/POM)(POMs = PW12, SiW12, and PMo12; is the number of TiO2-PW12 bilayers) films on magnetic microspheres were formed by alternative immersion in positively charged TiO2 (10 g/L, pH ∼ 2.0) and negatively charged POM solution (3 mmol/L). The assembly time was limited to 15 min to ensure the saturation adsorption of each component. Rinsing by HClO4 aqueous solution with a pH of 2.0 for three times was conducted after each deposition cycle. The Fe3O4@SiO2 microspheres coated with the desired deposition number of TiO2 and POMs were finally obtained. Then, the obtained catalysts were dried overnight at 60°C before further use. For comparison, (PSS/TiO2)10 and (PAH/PW12)10 composite films on Fe3O4@SiO2 magnetic microspheres were deposited in a similar way.

2.3. Characterization of Magnetic Microspheres

The morphology of magnetic microspheres was monitored using a field emission scanning electron microscope (FESEM, model: JSM-6700F, JEOL, Japan) with an accelerating voltage of 10 kV. The Fourier-transform infrared spectra of the products were recorded on a TR200 spectrometer (Thermo Electron Corporation). The chemical element of magnetic microspheres was detected by energy dispersive X-ray spectroscopy (EDX, model: Oxford X-Max 51-XMX1004, Oxford Instruments Co. Ltd., England). The magnetic property was measured on a MPMS (SQUID)-XL (Quantum Design) vibrating sample magnetometer at 300 K.

2.4. Evaluation of Photocatalytic Activity

The XPA-system photochemical reactor purchased from Xujiang Electromechanical Plant was used to conduct photocatalytic experiments. The photoreactor was equipped with a 300 W medium-pressure mercury lamp with a main emission wavelength of 365 nm. The lamp was horizontally arranged and surrounded by a quartz jacket in which cooling water was circulated to avoid overheating. There was a 20 cm distance between the light source and dye solution. A given amount of catalysts and 50 mL of methyl orange solution were placed in a 100 mL glass beaker thermostated using a water jacket. The temperature of the dye solution was controlled at 28 ± 1°C by adjusting the flow of cooling water. The reaction was performed at MO initial concentrations of 10 mg/L with the pH value of the solution at about 2.0 adjusted by HClO4. In the experiments, the catalyst concentration was fixed at 2 g/L except for a few runs to evaluate the effects of the catalyst concentration and inorganic oxidants. In order to establish the adsorption-desorption equilibrium between the MO molecule and the catalysts’ surface, the suspension was ultrasonically dispersed for 30 min in the dark. The photocatalytic decontamination of MO started after the mercury lamp became stable. At given intervals, a certain amount of the reaction solution was taken out and separated. The residual dyes were analyzed by a Shimadzu UV-2450 UV-vis spectrophotometer. The removal rate can be estimated by the following formula: where A0 is the original absorbance of the MO solution at the maximum absorption wavelength , and A is the residual absorbance of MO after UV-light irradiation.

3. Results and Discussion

3.1. Characterization of Magnetic Microspheres

The surface morphology of magnetic microspheres was examined by FESEM. It is clearly visible from Figure 1 that the Fe3O4 microspheres modified by the hydrolyzation of tetraethyl orthosilicate exhibit a typical spherical form, and the average particle sizes are determined to be 150–200 nm. The shape of the magnetic microspheres has no prominent change when TiO2 colloidal nanoparticles and PW12 nanoclusters assembled alternately on the outer surface of Fe3O4@SiO2 microspheres. It also shows that magnetic microspheres are aggregated to a certain extent because of the magnetic dipolar interaction among the particles. Moreover, the composition of magnetic photocatalysts was also confirmed by EDX spectroscopy (Figure 1(c)). EDX analysis reveals the presence of Ti, O, W, P, and Fe on the surfaces of Fe3O4@SiO2 magnetic microspheres. This proves the successful assembly of TiO2 nanoparticles and polyoxometalate (POM) anions on the surface of Fe3O4@SiO2 microspheres. Meanwhile, a dramatic Pt peak can be seen. Pt comes from the plated platinum used to increase the conductivity of sample.

The assembly process of magnetic microspheres was monitored by FT-IR spectra. The results are shown in Figure 2. The band that appeared at 584 cm−1 is assigned to the vibration of the Fe-O bond for Fe3O4 particles. After coating a SiO2 shell by the sol-gel method, three obvious bands at 1100 cm−1, 798 cm−1, and 469 cm−1 are assigned to the symmetric and antisymmetric stretching vibration of the Si-O bond. After an alternative deposition of TiO2 and PW12 multilayer films through the LbL method, the intensity of the Fe-O and Si-O bands decreases. This phenomenon is ascribed to the effective screening of TiO2 and PW12 multilayer films on Fe-O and Si-O bands. Meanwhile, the new bands at 960 cm−1, 895 cm−1, and 804 cm−1 appear, which associate with typical Keggin skeletal vibration bands for PW12. This is consistent with the previous report [14]. However, the stretching vibration of the P-O bond at 1080 cm−1 is covered by a strong peak of the Si-O bond. The FT-IR spectrum of Fe3O4@SiO2@(TiO2/PW12)10 microspheres indicates that the Keggin structure of PW12 still remains intact.

Figure 3(a) shows the magnetization of Fe3O4, Fe3O4@SiO2, and Fe3O4@SiO2@(TiO2/PW12)10 microspheres as a function of an applied magnetic field, which was measured by a vibrating sample magnetometer at 300 K. The samples all exhibit a superparamagnetic characteristic. The magnetizations of samples increase with the increase of the applied magnetic field and nearly reach saturation with saturation magnetization (Ms) values of 48.5, 15.8, and 6.0 emu/g, respectively. Compared to Fe3O4@SiO2 microspheres, the Ms values of composite photocatalysts slightly decrease. This is ascribed to the deposition of (TiO2/PW12)10 on core-shell Fe3O4@SiO2. Figure 3(b) illustrates the magnetic separation of Fe3O4@SiO2@(TiO2/PW12)10 microspheres from the aqueous suspension. The photocatalysts possess a strong magnetic response under an external magnetic field due to the existence of an Fe3O4 core. They can be easily removed from the MO solution after photodegradation by an external magnet. Then, a nearly colorless solution is obtained (see Figure 3(b)). The strong magnetic response of photocatalysts ensures their convenient, fast, and efficient separation from the reaction media for reuse.

3.2. Photocatalytic Activity of Magnetic Microspheres

In the photocatalytic procedure, methyl orange (MO) was used as the model organic molecule to evaluate the photocatalytic activity of composite films assembled on Fe3O4@SiO2 magnetic microspheres. The photocatalytic activity of samples toward azo dye MO is shown in Figure 4. In the dark, the degradation of MO is hardly observed even though magnetic films are presented. Only 13.30% of MO can be degraded by direct photolysis after UV-light irradiation for 100 min. After being exposed at the same time, the removal rates of MO in the presence of (TiO2/PW12)10, (TiO2/PSS)10, and (PAH/PW12)10 composite films deposited on Fe3O4@SiO2 microspheres are 83.91%, 28.66%, and 16.44%, respectively. The fast disappearance of MO in the presence of the Fe3O4@SiO2@(TiO2/PW12)10 photocatalyst owes to the synergistic effect of TiO2 and PW12. Since photogenerated electrons can be transferred from the TiO2 conduction band to the empty orbit of tungstophosphate acid, quantum efficiency can be improved. Accordingly, reactive oxygen species (ROSs) are easy to generate through the reaction of electrons and holes with hydroxyl ion, adsorbed water, and oxygen. These ROSs and active holes are without selectivity and can completely oxidize a MO molecule to CO2, H2O, and inorganic salts. However, without the synergistic effect, the fast recombination of photoinduced holes and electrons cannot be avoided in pure TiO2 or PW12 films, which leads to the decrease in photocatalytic efficiency.

Moreover, the photocatalytic efficiency of Fe3O4@SiO2@(TiO2/SiW12)10 and Fe3O4@SiO2@(TiO2/PMo12)10 magnetic films was also examined for comparison with Fe3O4@SiO2@(TiO2/PW12)10. Under the same experimental conditions, the photodegradation of MO by the abovementioned films reaches 62.07% and 13.55%, respectively. These results shown in Figure 5 indicate that (TiO2/PW12)10 and (TiO2/SiW12)10 composite films deposited on Fe3O4@SiO2 can lead to effective degradation of MO. In particular, a higher efficiency can be obtained from (TiO2/PW12)10 films. However, the photocatalytic activity of Fe3O4@SiO2@(TiO2/PMo12)10 magnetic microspheres is very low. This is because the reoxidization of reduced PMo12 is very difficult.

The effect of (TiO2/PW12) bilayer number coated on magnetic microspheres on MO decoloration was also investigated. As shown in Figure 6, the removal rate is increased significantly by increasing the (TiO2/PW12) bilayer number. After UV-light irradiation for 100 min, the degradation efficiency of MO with 2, 4, 6, 8, and 10 bilayers is 30.46%, 41.14%, 69.75%, 78.79%, and 83.91%, respectively. The highest degradation rate is obtained when the bilayer number of (TiO2/PW12) n grows to 10. This study reveals that the photocatalytic activity strongly depends on the bilayer number of (TiO2/PW12) films. The increase of photocatalytic efficiency is attributed to more TiO2 and PW12 nanoparticles effectively deposited on Fe3O4@SiO2 core-shell microspheres. It should be noted that Fe3O4@SiO2 microspheres have no photocatalytic activity. Thus, the reason why (TiO2/PW12)10 magnetic microspheres have a higher photocatalytic efficiency is that the contents of TiO2 and PW12 in (TiO2/PW12)10 are higher than in other catalysts.

Figure 7 compares the photodegradation of MO under various concentrations of Fe3O4@SiO2 microspheres coated with (TiO2/PW12)10 films. The catalyst concentration was selected as 0.2, 0.4, 0.8, 1.4, and 2.0 g/L. It is observed that the degradation rate increases gradually with the growth of the catalyst concentration. In the presence of (TiO2/PW12)10 magnetic microspheres, the degradation efficiency of MO with 0.2 g/L of the catalyst is 69.37% after UV-light irradiation for 100 min. When the concentration of the catalyst is 2.0 g/L, (TiO2/PW12)10 magnetic microspheres exhibit the excellent degradation capability for MO and around 83.91% of MO molecules are removed. The inset shows the plot of the rate constant (k) versus a sample concentration. It is indicated that the photocatalytic efficiency increased with the catalyst concentration and the optimal concentration was 2.0 g/L. The enhancement of photocatalytic efficiency is mainly relevant to the increase of catalyst concentration, which generates massive active sites. Thus, high efficiency can be obtained when more catalysts are added.

The photocatalytic efficiency of MO by (TiO2/PW12)10 films coated on Fe3O4@SiO2 microspheres was enhanced when halate was introduced. The comparative results with different halate concentrations are shown in Figure 8. It displays a highly positive role in MO degradation due to the addition of halate. The highest degradation rate of 97.32% is obtained at 1 mM BrO3 after 40 min of irradiation. Without BrO3, only 52.51% of the MO removal rate is attained within the same amount of time. The enhanced removal rate by the addition of KIO3, KBrO3, and KClO3 may be attributed to the fact that all of them can act as electron acceptors in the aqueous solution, thus inhibiting electron-hole recombination on the catalysts’ surface. The efficiency of BrO3 is always higher than that of IO3 and ClO3. This is probably related to the higher redox potential of BrO3, which easily reacts with the conduction band electron [24]. The degradation by IO3 is accelerated straightforward by the extra radicals such as IO2∙ and HIO3∙. However, the effect of I as the end product of reduction should not be overlooked. The accumulated I can result in a significant deterioration of the MO degradation rate [25].

To determine the effect of inorganic salts which always coexists with dye wastewater on the degradation process, different experiments were carried out. Figure 9 shows the results when inorganic salts such as NaCl, MgCl2, AlCl3, Na2SO4, and Na2HPO4 were added. These salts play different roles in the process of MO photocatalytic degradation. The results shown in Figure 9 reveal that metal chloride has a significant accelerating effect on MO degradation. AlCl3 illustrates the highest photocatalytic efficiency, and the removal rate reaches 97.32% after being irradiated for 60 min. Na2SO4 has no obvious impact on MO removal. However, the addition of Na2HPO4 results in a great decrease in the MO degradation rate, which is less than 11.45% after UV-light irradiation for 100 min. The presence of Cl ions which leads to the reduction of photocatalytic efficiency has been stated by many previous reports [2629]. The photocatalytic activity inhibits by Cl anion mainly attributed to competitive adsorption between Cl and pollutant. However, the positive effect of chloride was achieved in our experiments. This enhancement should be attributed to the efficient formation of chlorine and dichloride anion radicals by scavenging holes and ∙OH. This process promotes the separation of photogenerated electron-hole pairs [26]. In addition, the higher efficiency of AlCl3 may be related to the charge density δ of the metal cation calculated through the ionic radii. The higher charge density of the metal cation favors the photocatalytic degradation of MO by inhibiting recombination of the intermolecular relaxation of dye molecules with electrons [30]. The decrease of the removal rate by Na2HPO4 is due to the addition of Na2HPO4 which results in a pH increase of the MO solution. At high pH, H3PW12O40 will be decomposed [15].

The reaction kinetics of MO photodegradation in the presence of halate by Fe3O4@SiO2@(TiO2/PW12)10 films were investigated. It can be found in Figure 10 that the plots of lnc0/c (where c0 stands for the initial concentration and c is the residual concentration at time t) versus the reaction time yields straight lines. It indicates that the photodecomposition of the diluted MO solution follows apparent first-order kinetics. The first-order rate constants (k) shown in Table 1 were determined from the regression curves of lnc0/c versus the reaction time. The constants are 0.01578 min−1 without halate, 0.02592 min−1 for KClO3, 0.08717 min−1 for KBrO3, and 0.07117 min−1 for KIO3. The rate constant increases along with the addition of halate, and the greatest rate constant reaches by 1 mM BrO3. It also confirms that photocatalytic efficiency is enhanced, which ascribes to the effective reception of photogeneration electrons.

4. Conclusions

Photocatalytic (TiO2/POM) film coated on magnetic Fe3O4@SiO2 microspheres were prepared based on the electrostatic LbL method. Degradation of methyl orange under UV irradiation was investigated to compare their photocatalytic activity. The presence of PW12 and SiW12 leads to the effective degradation of MO due to the synergistic effect by combining POMs with TiO2. However, the reoxidization of reduced PMo12 is very difficult, which causes the decline of the catalysts’ effectiveness. The degradation of methyl orange is significantly enhanced by increasing the bilayer number and catalyst concentration. The combination with inorganic oxidants enhances significantly the photodecoloration rate of the dye. It is found that 97.32% of MO is removed after 40 min of irradiation by the addition of KBrO3. Analyses on the primary role of inorganic salts demonstrate that the photocatalytic efficiency increases drastically by metal chloride, but strongly decreases by Na2HPO4. The degradation reactions of MO follow pseudo first-order kinetics. It is found that the use of magnetic support guarantees a photocatalyst with facile, clean, fast, and efficient separation.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was financially supported by the National Natural Science Foundation of Shandong Province Combination Research Projects (ZR2013BL001, ZR2015BL032, and ZR2014BL001).