There is about 2.37 billion ton textile wastewater discharged from Chinese textile industry every year, among which dying wastewater accounts for about 80%. The color of dyeing wastewater is particularly serious pollution, and it is difficult to remove by general biochemical methods. The recycling rate of wastewater after treatment in the whole textile industry is only about 10%. Therefore, the treatment and emission reduction of dyeing wastewater are urgent for improving the quality of water environment. Photocatalytic technology is a good application prospect in the fields of environmental purification, solar energy utilization and self-cleaning   .
CeO2 has attracted much attention in photocatalytic degradation of wastewater contaminants for its unique crystal structure, high oxygen storage, and oxygen release capacity. The activity and selectivity of pure CeO2 catalyst generally are improved by adjusting morphology and particle size of CeO2, so its catalytic oxidation technology is insufficiently pushed to a wide range of practical applications   . Researchers have tried to synthesize multi-component Ce-based composite oxide catalysts by doping transition metal M into CeO2. The lattice distortion of CeO2 results in defects when CeO2 interacts with transition metal M for its variable valence and excellent redox performance. Then it accelerates the flow of oxygen in the crystal phase and provides more active oxygen species for catalytic oxidation. So the catalytic activity and stability of the Ce-based composite oxide catalysts are greatly improved . Choi et al.  studied the effect of various metal ions on the photocatalytic activity of titanium dioxide, among which the effect of Fe3+ was the best. Machida et al.  found that the oxygen storage/release capacity of CeO2-Fe2O3 composites was better than that of pure components CeO2 and Fe2O3. They also pointed out that CeO2 was the entrance of oxygen storage. The high temperature reaction activity and thermal stability of the La modified Ni/CeO2-ZrO2 catalysts were improved because La3+ doped into Ni/CeO2-ZrO2 oxides could produce more oxygen vacancies during high temperature sintering, and then the oxygen mobility was enhanced . In this paper, Ce-based composite oxide catalysts were prepared with transition metals of Fe3+ and La3+ doped into CeO2, and their performances were also investigated.
The preparation methods of Ce-based composite oxide catalysts have shown great influence on their structure and morphology. The preparation methods are hydrothermal method , sol-gel method , combustion method , precipitation method , template method , and so on. In the paper, the La2O3/Fe2O3-CeO2 catalysts with different morphologies were prepared by coprecipitation method, sol-gel method and hydrothermal method. The effects of preparation methods on the structure and performance of the La2O3/Fe2O3-CeO2 catalysts were studied.
2. Experimental Sections
2.1. Preparation of La2O3/Fe2O3-CeO2 Catalyst
The reagents Ce(NO3)3·6H2O, Fe(NO3)3·9H2O, La(NO3)3·6H2O, polyethyleneglycol (PEG), ammonium bicarbonate, urea, and methylene blue from China National Medicines Corporation Ltd. were analytically pure. All the chemicals were AR grade, and used as received without further purification. Deionized water was used in all experiments.
Coprecipitation method: First, 0.868 g Ce(NO3)3·6H2O, corresponding Stoichiometric ratio Fe(NO3)3·9H2O and La(NO3)3·6H2O were dissolved in deionized water, then the solution was added into 250 ml three-port bottle. The polyethylene glycol (PEG) solution was slowly added into the three-port bottle with 0.21 g PEG as complexing agent dissolved in 40 ml water. Then the precipitator NH4HCO3 was added into to mixed solution to aid precipitate at 80˚C under agitation. After aging at 100˚C for 10 h for 24 h, the precipitate was filtrated and washed with deionized water. The La2O3/Fe2O3-CeO2 precursor samples were obtained after the precipitate being dried, followed by calcinating in a muffle furnace at 600˚C for 4 h. The corresponding La2O3/Fe2O3-CeO2 composite sample was denoted as sample-CP.
Sol-gel method: First, 0.868 g Ce(NO3)3·6H2O, corresponding Stoichiometric ratio Fe(NO3)3·9H2O and La(NO3)3·6H2O were dissolved in deionized water to obtain mixed solution, then the mixed solution was added into 250 ml three-port bottle. The polyethylene glycol (PEG) solution was slowly added into the three-port bottle with 0.21 g PEG as complexing agent dissolved in 40 ml water. Then the mixed solution was agitated at 80˚C to obtain sol. The scarlet La2O3/Fe2O3-CeO2 gel precursor was obtained after the sol was dried at 100˚C for 10 h. The CeO2 compound oxide samples were acquired by calcinating CeO2 gel in a muffle furnace at 600˚C for 4 h. The corresponding La2O3/Fe2O3-CeO2 composite sample was denoted as sample-SG.
Hydrothermal method: First 0.21 g PEG and 1.0 g urea were dissolved in distilled water, and continued stirring to make the mixture solution. Then 0.868 g Ce(NO3)3·6H2O, corresponding Stoichiometric ratio Fe(NO3)3·9H2O and La(NO3)3·6H2O were added into the mixture solution and stirred for 0.5 h at 80˚C. Finally, the mixed solution was transferred to a polytetrafluoroethylene high-pressure hydrothermal kettle of 100 ml for 24 hours at 140 C. The precipitate was filtrated and washed with deionized water. The La2O3/Fe2O3-CeO2 precursor samples were obtained after the precipitate being dried, followed by drying in oven at 80 C for 10 h then calcinating in a muffle furnace at 600˚C for 4 h. The corresponding La2O3/Fe2O3-CeO2 composite sample was denoted as sample-HT.
The molar ratio of La3+:Fe3+ was 1:2, and the molar doping amount of La2O3 and Fe2O3 in La2O3/Fe2O3-CeO2 was 3% in the experiment.
2.2. Catalytic Degradation Experiments
Catalytic degradation experiments were performed in a double beaker of 250 ml with water cooling. A portion of catalyst (10 mg) and H2O2 (6 ml) were introduced into 100 ml of aqueous dye solution containing methylene blue at a concentration of 100 mg·L−1. Sodium hydroxide and hydrochloric acid solution were used to adjust the pH value of the solution. Adsorption equilibrium was attained by the reaction solution after magnetic stirring for 30 min in the dark. The degradation reaction was then performed using 300 W xenon lamps as visible light. Liquid samples were removed at regular 10 min intervals for analysis of the absorbance. To evaluate the catalytic activity of the catalysts, the dye removal efficiency (η) was calculated, as shown below:
where c0 is the initial after the dark equilibrium, and ct is different times absorbance values of methylene blue, respectively.
2.3. Catalyst Characterization
A thermogravimetric (TG/DSC) analysis was conducted on an STD Q600 instrument at a rate of 5˚C·min−1 under nitrogen protection. The crystalline phases of the as-prepared CeO2 composite oxides were studied by x-ray diffraction (XRD) with a Bruker/AXS D8 Advance X-ray diffractometer operated at 40 kV and 200 mA with Cu Ka radiation (λ = 0.15418 nm). The morphology and dispersion of the powders were observed with the use of a transmission electron microscope (TEM, JEOL-6360LV). Textural properties of the samples were measured by the Brunauer-Emmett-Teller (BET) model on a Quantachrome NOVA instrument, with Aras the carrier gas and N2 as the adsorbent at −196˚C. The visible light absorbance at the characteristic wavelength of 664 nm for methylene blue dye was measured with a UV-2500 spectrophotometer.
3. Results and Discussions
3.1. TG/DSC Analysis
Figure 1 shows the TG/DSC curves of the dried La2O3/Fe2O3-CeO2 precursors with a heating rate of 5˚C·min−1. DSC curves display endothermic peaks and exothermic peaks of the dried precursors. The about 150˚C endothermic peak can be explained by the removal of adsorbed water and crystal water. The about 200˚C exothermic peaks can be explained by the burning and decompose of PEG in the precursor powders. Major mass loss of the precursor samples occurs below 500˚C. The overall weight loss is 20% - 50%. The thermal decomposition of precursor from the coprecipitation method was showed in Figure 1(A). The main mass loss occurred below 300˚C. A small endothermic peak at 150˚C could be attributed to the loss of adsorbed water and crystal water, which accounting for about 10% of the total mass loss. There are two exothermic peaks between 200˚C and 300˚C, which were caused by the decomposition of Ce(OH)3 and Ce2(CO3)3 precursors, accounting for about 40% of the total mass loss. The different interaction between PEG and metal ions leads to varied decomposition temperature. When the temperature is about 520˚C, there is another exothermic
Figure 1. TG/DSC curves of precursors synthesized by different methods: (A) CP, (B) SG, (C) HT.
peak which shows crystal formation. The mass loss of precursor samples by coprecipitation method mainly occurs in two regions of 200˚C - 250˚C and 300˚C - 460˚C, with a total loss of about 25%. The corresponding exothermic peak at 200˚C - 250˚C can be attributed to thermal decomposition of residual PEG in precursor sample-SG. The strong endothermic peak at 300˚C - 460˚C can be decomposed of PEG additives (shown in Figure 1(B). While at about 350˚C - 550˚C, there occurs an exothermic peak of sample-HT precursors in the heating process which implies the decomposition of precipitates of precursor Ce(HCOO)3 and the formation of CeO2 crystal phases, with a total weight loss of about 20% (Figure 1(C). This is proved by the XRD results in Figure 2. From the above results, the intermolecular and ionic interactions in the precursor precipitates prepared by different methods are different. So the thermal stability of precursors and the decomposition temperature of samples are different.
3.2. XRD Analysis
The XRD patterns of La2O3/Fe2O3-CeO2 samples are exhibited in Figure 2. The peaks around 2θ = ~28.5˚, ~33.0˚, ~47.5˚, ~57.5˚ and ~69.5˚ in agreement with the (111), (200), (220), (311) and (400) planes of the cubic fluorite structure of CeO2, respectively. The crystal form is consistent with the standard JCPDS (34-0394) spectra . The results show that the main active component of the catalyst is crystallized CeO2 with cubic fluorite structure and this structure remains after calcinating process. However, the crystallinity of the samples varies greatly with different synthesis methods. Sample-CP shows the best crystallinity and strongest diffraction peak intensity. To resume, the crystallinity of catalyst samples is in the following order: sample-CP > sample-SG > sample-HT. According to Scherer formula, the particle sizes of La2O3/Fe2O3-CeO2 samples prepared by the three methods were calculated to be about 100 nm (sample-CP), 32.56 nm (sample-SG) and 5.23 nm (sample-HT), respectively.
From the diffraction peak in Figure 2, we can see that there is no impurity phase peak, and the samples display high purity, while the diffraction peak of the sample obtained by hydrothermal method slightly shifts red. The doping elements La and Fe in the samples are not detected by XRD, indicating that La3+ and Fe3+ cations have been well inserted into the CeO2 lattice, or La2O3-based and Fe2O3-based clusters are well dispersed on the surface of CeO2 particles .
Figure 2. XRD of La2O3/Fe2O3-CeO2 catalysts from different methods.
3.3. TEM Analysis
Figure 3 shows TEM photographs of La2O3/Fe2O3-CeO2 samples synthesized by different methods. It can be seen that the rod-like sample, with a diameter of about 100 nm, obtained from coprecipitation is well dispersed from Figure 3(A). The sample-SG is spherical and well dispersed, with a diameter of 20 - 40 nm. The samples obtained by hydrothermal method are fine particles, some of which accumulate into electronic clouds. The results consist of the results of XRD. Compared with (A), (B) and (C) in Figure 3, it is obvious that the morphologies of samples synthesized by different methods vary greatly.
During the formation of sol-gel, the spherical structure is formed by nanoparticles self-assembly through hydrogen bonds between ethylene glycol molecules. When the gel is dried and calcined at high temperature, the nuclei grow along the isotropic state, and nanospheres are acquired . There are Ce(OH)3 and Ce2(CO3)3 precipitators formed during coprecipitation. While Ce(OH)3 and Ce2(CO3)3 are polyhedrons constituted of a series of polyhedrons consisting of Ce3+ and OH－, . During coprecipitation, Ce3+ binds with OH－and then form nanocrystals with very small size. Nanorods are formed when these nanocrystals grow along the crystal plane with high chemical activity . Ce-OCH2-CH2-OH are formed between Ce(HCOO)3 and polyethylene glycol in the hydrothermal reaction process, because polyethylene glycol was adsorbed on the positive charge Ce(HCOO)3 crystal surface with high chemical activity. Which hindered the growth of Ce(HCOO)3 along the crystal surface, resulting in fine nanoparticles. When calcined at high temperature, some of these fine nanoparticles aggregated. The morphology and particle size of La2O3/Fe2O3-CeO2 samples directly affect their texture properties.
Figure 3. TEM photography of La2O3/Fe2O3-CeO2 catalyst from different methods: (A) CP, (B) SG, (C) HT.
3.4. BET Analysis
The texture structure of the samples was conducted by N2 adsorption and desorption method. The specific surface area, pore volume and average pore diameter results of the samples from three methods are shown in Table 1. Sample-CP has the highest surface area of 76.21 m2·g−1 and the highest pore volume of 0.091 cm3·g−1. The smallest surface area of the sample-SG sample are 32.66 m2·g−1, and the largest pore diameter is 13.84 nm. While the surface area of sample-HT intervene sample-CP and sample-SG, the pore volume and pore diameter were the smallest, the values of them are 53.48 m2·g−1, 0.038 cm3·g−1 and 3.95 nm, respectively. This may be due to the different decomposition temperature, gas release rate and heat release rate of the precursors during thermal decomposition, which results in obvious difference in pore structure and surface area.
The pore size of the sample-SG shows obvious growth tendency, indicating that the sample-SG possesses better aging resistance and stronger adsorption capacity. During the sol-gel process, the organic dispersant PEG has been kept in the precursor. When calcined at high temperature, PEG decomposes and releases a large amount of gas, resulting in an increase in the pore size of the sample. However, its surface area is the smallest because the particle shape of the sample-SG is spherical. Most PEG is lost following the filtrate during precipitate filtrating for coprecipitation process and hydrothermal process. At the same time, the particle size is small and the particles agglomerated seriously of the sample-HT, which reduces the pore volume and pore size. The result is consistent with the thermal analysis results.
Figure 4 shows the adsorption and desorption isotherms of the La2O3/Fe2O3-CeO2 samples. We can find that three kinds of samples present adsorption and desorption isotherms of mesoporous (2 - 50 nm) materials characteristics. There are also hysteresis loops in the isotherms due to incomplete uniformation of pore shapes and pore size, as well as phenomenon of capillary condensation. Of which, the hysteresis loops of sample-CP are the most obvious, and the hysteresis loops of sample-HT are the smallest. Whatmore, the pore shape of samples is bottle-shaped. Hysteresis loops are formed due to the phenomenon of capillary condensation and incomplete pore size . The adsorption curves of the samples prepared by coprecipitation and sol-gel method increase sharply at the relative pressure (P/P0) of 0.6 - 1.0. It may be due to the larger pore size and relatively concentrated pore size distribution of the sample, and the pore size is mainly concentrated in the mesoporous size. The results are consistent with the pore size distribution curves. It can also be seen from Figure 4 that among the three samples, desorption of sample-HT is the strongest, and the hysteresis loop formed by adsorption-desorption curve is the most obvious. The adsorption desorption value of the sample-CP is the largest, followed by the sol-gel method. The value of the hydrothermal method is the smallest, which is consistent with the analysis results of the surface area and pore structure of the La2O3/Fe2O3-CeO2 sample. These textural properties will directly affect the photocatalytic degradation performance of La2O3/Fe2O3-CeO2.
Table 1. Texture properties of different La2O3/Fe2O3-CeO2 photocatalyst samples.
Figure 4. N2 adsorption-desorption isotherms and pore volume distribution of La2O3-Fe2O3-CeO2 catalyst samples: (A) CP, (B) SG, (C) HT.
3.5. La2O3-Fe2O3-CeO2 Photocatalytic Degradation Analysis
Figure 5 shows the time dependence of the methylene blue removal efficiency by different La2O3/Fe2O3-CeO2 catalysts under 300 W Xenon lamps. The methylene blue removal efficiencies are shown in Figure 5 for a degradation process with a catalyst concentration of 100 mg·L−1 at solution pH = 10.0. The sample-CP catalyst shows the best catalytic activity during the catalytic degradation process. The degradation of methylene blue is almost complete at 50 min for La2O3/Fe2O3-CeO2/γ-Al2O3, which represents an increase of 5.58% and 9.54%, compared with the efficiencies of sample-SG and sample-HT under the same reaction conditions, respectively. The removal efficiency of methylene blue by sample-CP achieves a maximum value of 99.58% at 50 min. The high catalytic activity is attributed to the texture and morphology of the catalyst sample.
The sample-CP has good adsorption and more active sites on surface for its large surface area. So the catalytic reaction effect is better than the other two. This indicates that catalytic oxidation process is major dependent on adsorption-desorption, and catalytic oxidation also plays an important role. The pore size and pore size of sample-CP are large, which increases the defects inside the lattice and decreases the band gap. It provides more reactive oxygen species for catalytic oxidation and improves the catalytic activity . The sample-HT has small particle size and there is agglomeration among the particles, which leads to the reduction of the catalytic effect.
The degradation reaction of methylene blue can be described by the first-order reaction kinetics: ln (c0/ct) = kt + qe, where c0 and ct are the concentration of methylene blue at the initial time and each interval time during irradiation, respectively, k is the first-order rate constant, t is the irradiation time and qe is a constant. A high k-value usually implies the fast reaction rate of efficient catalysts . The calculated k value (in Figure 6) is 0.074, 0.063 and 0.060 min−1 for sample-CP, sample-SG and sample-HT, respectively, which indicates that the sample-CP catalyst exhibits the highest photocatalytic activity. The result is consistent with the results of Figure 5.
Figure 5. Degradation methylene blue results for different La2O3/Fe2O3-CeO2 catalyst samples under Xenon light.
Figure 6. Linear fits of ln (c0/ct) versus time.
3.6. Mechanism Analysis
The electron-hole pairs on the surface of the catalyst react with H2O, H2O2 and O2, then form •OH and •O2. In the degradation process, methylene blue is first adsorbed on the surface of the catalyst, then reacts with •OH and •O2 and decomposed into CO2 and H2 . Researchers found that doped with transition metals, electron-hole pairs on the surface of CeO2 will be promoted  . When transition metal La3+ is doped into CeO2, some of Ce4+ lattices are substituted by La3+ and produce a large amount of reactive oxygen species, which greatly improves the catalytic activity of CeO2 . The doping of Fe3+ into TiO2 results in the decrease of its energy level, catalytic reaction shifted to 400 - 700 nm in the visible region, and the catalytic activity of TiO2 catalyst was improved . Therefore, the doping of La3+ and Fe3+ into CeO2 will result in lattice distortion and defect, decrease energy level, accelerate the flow of oxygen in the crystal phase, and provide more active oxygen species for catalytic oxidation. Therefore, the catalytic activity of La2O3/Fe2O3-CeO2 samples is enhanced under visible light excitation.
Usually, the visible wavelength is 380 - 760 nm and its corresponding energy is 3.10 - 1.63 ev. CeO2 has a large band gap of 3.2 ev, which makes it sensitive to ultraviolet radiation only . The band gap between the valence and conduction bands of the CeO2 and La2O3/Fe2O3-CeO2 samples is determined using UV-vis diffuse reflectance spectra. As shown in Figure 7(A), the large absorption at wavelengths shorter than 400 nm can be attributed to intrinsic band-to-band absorption of La2O3/Fe2O3-CeO23. Compared to the CeO2, the La2O3/Fe2O3-CeO2 compound oxide exhibit significantly increased absorption at above 430 nm, indicating that the loaded La3+ and Fe3+ may be helpful for utilizing visible light more effectively. The diffuse reflectance was transferred via Kubelka-Munk function into Tauc plots and the band gap energy of the samples was defined via Tauc plots that are shown in Figure 7(B). The band gap La2O3/Fe2O3-CeO2 exhibits the energy gaps of sample-CP (2.50 ev), sample-SG (2.68 eV) and sample-HT (2.82 eV) respectively. It is deduced that the decrease
Figure 7. UV–Vis diffuse reflectance spectra of different La2O3/Fe2O3-CeO2 catalyst samples (A); (ahv)2 versus energy (hv) plots of different La2O3/Fe2O3-CeO2 catalyst samples (B).
of band gap increases the light absorption, and electrons through absorbing lower photon energy in the valence band more easily jump to the conduction band . Thereby, optical results of La2O3/Fe2O3-CeO2 imply more efficient processes for visible-light catalytic degradation of methylene blue. Sample-CP shows the best catalytic degradation effect for methylene blue.
La2O3/Fe2O3-CeO2 composite oxide catalysts were prepared by coprecipitation method, sol-gel method and hydrothermal method. The catalyst was CeO2 phase with cubic fluorite structure. The samples with different morphology were obtained due to varied interaction forces between ions and molecules in the precursors, and different decomposition processes. The rod-shaped sample from coprecipitation method was about 100 nm particles with good dispersivity, 20 - 40 nm spheres sample obtained by sol-gel method, and it was small agglomerated particles by hydrothermal method. The morphology and particle size of the sample directly affected its texture properties of surface area, pore volume and pore diameter. Catalytic degradation performance depends not only on structure and texture properties, but also on band gap of the catalyst. The catalytic degradation rate of methyl blue under visible light is in the following order: sample-CP > sample-SG > sample-HT.
This project was financially supported by the projects of “Qian Jiao He KY”  114 and “Talents of Qian Ke He Platform” (5727-11).
 Selvakumar, S., Manivasagan, R. and Chinnappan, K. (2013) Biodegradation and Decolourization of Textile Dye Wastewater Using Ganoder malucidum. 3 Biotech, 3, 71-79.
 Pitre, S.P., Yoon, T.P. and Scakano, J.C. (2017) Titanium Dioxide Visible Light Photocatalysis: Surface Association Enablesphotocatalysis with Visible Light Irradiation. Chemical Communications, 53, 4335-4338.
 Saravanan, R., Agarwal, S., Gupta, V.K., Khan, M.M., Gracia, F., Mosquera, E., Narayan, V. and Stephen, A. (2018) Line Defect Ce3+ Induced Ag/CeO2/ZnO Nanostructure for Visible-Light Photocatalytic Activity. Journal of Photochemistry and Photobiology A: Chemistry, 353, 499-506.
 Fazal, A., Javed, I., Tariq, J., Noor, B., Qaisar, M. and Muhammad, I. (2016) Structural, Morphological, Raman, Optical, Magnetic, and Antibacterial Characteristics of CeO2 Nanostructures. International Journal of Minerals, Metallurgy, and Materials, 23, 102-108.
 Thaís, M.N., Tuanny, S.F., Evelyn, C.C.D.S., Marcos, A.G. and Vanessa, B.M. (2017) An Investigation into an Alternative Photocatalyst Based on CeO2/Al2O3 in Dye Degradation. Environmental Technology & Innovation, 8, 349-359.
 Tan, J.P.Y., Tan, H.R., Boothroyd, C., Foot, Y.L., Het, C.B. and Lint, M. (2011) Three-Dimensional Structure of CeO2 Nanocrystals. Journal of Physical Chemistry C, 115, 3544-3551.
 Choi, J., Park, H. and Hoffmann, M.R. (2010) Effects of Single Metal-Ion Doping on the Visible-Light Photoreactivity of TiO2. Journal of Physical Chemistry C, 114, 783-792.
 Machida, M., Kawada, T., Fujii, H. and Hinokuma, S. (2015) The Role of CeO2 as a Gateway for Oxygen Storage over CeO2-Grafted Fe2O3 Composite Materials. Journal of Physical Chemistry C, 119, 24932-24941.
 Tarak, M., Nidhi, K., Ritesh, M. and Kamal, K.P. (2016) Catalytic Steam Reforming of Model Oxygenates of Bio-Oil for Hydrogen Production Over La Modified Ni/CeO2-ZrO2 Catalyst. Topics in Catalysis, 59, 1343-1353.
 Liu, W., Feng, L.J., Zhang, C., Yang, H.X., Guo, J.X., Liu, X.F., Zhang, X.Y. and Yang, Y.Z. (2013) A Facile Hydrothermal Synthesis of 3D Flowerlike CeO2 via a Cerium Oxalate Precursor. Journal of Materials Chemistry A, 1, 6942-6948.
 Román, R.L., Ignacio, E., Esaú, E., Rodrí, G.M., Silverio, M.L. and Abelardo, I.F.V. (2017) Al2O3-CeO2 Sol-Gel Synthesis and Addition of Rh to Improve the Oxygen Mobility of Mixed Support. Journal of Sol-Gel Science and Technology, 81, 214-219.
 Sun, C.W., Xie, Z., Xia, C.R., Li, H. and Chen, L.Q. (2006) Investigations of Mesoporous CeO2-Ｒu as a Reforming Catalyst Layer for Solid Oxide Fuel Cells. Electrochemistry Communications, 8, 833-838.
 Jeong, M., Nunotani, N., Moriyama, N. and Imanaka, N. (2017) Introduction of NiO in Pt/CeO2-ZrO2/γ-Al2O3 Catalysts for Removing Toluene in Indoor Air. Materials Letters, 208, 43-45.
 Liu, C.B., Sun, H., Qian, J.C., Chen, Z.G., Lv, Y.F., Chen, F., Lu, X.W. and Wu, Z.Y. (2017) Biotemplating Synthesis and Photocatalytic Activities of N-doped CeO2 Microcapsule Tailored by Hemerocallis Pollen. Advanced Powder Technology, 28, 2741-2746.
 Satyanarayana Reddy, A., Chen, C.Y., Chen, C.C., Chen, S.H., Lin, C.J., Lin, K.H., Chen, C.L. and Chang, S.C. (2010) Synthesis and Characterization of Fe/CeO2 Catalysts: Epoxidation of Cyclohexene. Journal of Molecular Catalysis A: Chemical, 318, 60-67.
 Li, Y.F., He, Y.H., Liu, G.Q. and Zeng, L.W. (2018) Influence of La Precursors on Structure and Properties of CeO2-ZrO2-Al2O3 Composite Oxides. Transactions of Nonferrous Metals Society of China, 28, 739-747.
 Yu, X.X., Dong, Z.F., Yang, X., Dong, B. and Wu, Y. (2017) Facile Synthesis of ZnO Nanoparticles for the Photocatalytic Degradation of Methylene Blue. Journal of Sol-Gel Science and Technology, 82, 167-176.
 Zhang, C., Zhang, X.Y., Wang, Y.C., Xie, S.L., Liu, Y., Lu, X.H. and Tong, Y.X. (2014) Facile Electrochemical Synthesis of CeO2 Hierarchical Nanorods and Nanowires with Excellent Photocatalytic Activities. New Journal of Chemistry, 38, 2581.
 Liu, Y. and Sun, D.Z. (2007) Effect of CeO2 Doping on Catalytic Activity of Fe2O3/γ-AL2O3 Catalyst for Catalytic Wet Peroxide Oxidation of Azo Dyes. Journal of Hazardous Materials, 43, 448-454.
 Lamdab, U., Wetchakun, K., Phanichphanrs, S., Kangwansuamonkon, W. and Wetchakun, N. (2015) Highly Efficient Visible Light-Induced Photocatalytic Degradation of Methylene Blue over InVO4/BiVO4 Composite Photocatalyst. Journal of Materials Science, 50, 5788-5798.
 Choudhury, B., Chetri, P. and Choudury, A. (2014) Oxygen Defects and Formation of Ce3+ Affecting the Photocatalytic Performance of CeO2 Nanoparticles. RSC Advances, 4, 4663-4671.
 Chen, H., Saysri, A., Adaot, A. and Larachi, F. (2001) Composition-Activity Effects of Mn-Ce-O Composites on Phenol Catalytic Wet Oxidation. Applied Catalysis B: Environmental, 32, 195-204.
 Neri, G., Pistone, A., Milone, C. and Galvagno, S. (2002) Wet Air Oxidation of Pcoumaric Acid over Promoted Ceria Catalysts. Applied Catalysis B: Environmental, 38, 321-329.
 Reina, T.R., Ivanova, S., Centenom, A. and Odriozola, J.A. (2015) Catalytic Screening of Au/CeO2-MOx /Al2O3 Catalysts (M1/4 La, Ni, Cu, Fe, Cr, Y) in the CO-PrOx Reaction. International Journal of Hydrogen Energy, 40, 1782-1788.