Effect of Fe doping on the catalytic performance of CuO–CeO₂ for low temperature CO oxidation

Abstract

Cu₁Ce₈Fe_x–O catalysts for low temperature CO oxidation were prepared by an improved citrate sol–gel method with incorporation of thermal treatment under N₂, and characterized by the nitrogen adsorption–desorption, XRD, CO pulse experiments, CO-TPR and CO-TPD. When x = 1, the Cu₁Ce₈Fe₁–O catalyst shows higher catalytic activity for CO oxidation, and the reaction temperature for 90% CO conversion (T₉₀) is only 50 °C. The presence of Fe in Cu₁Ce₈Fe₁–O can improve the catalytic activity for CO oxidation, which results from its higher surface area, smaller crystalline size, higher activity and larger amounts of surface oxygen species.

---

1. Introduction

As a toxic gas, CO exists widely in the utilization processes of any fuel burning appliance, vehicles, tools with fuel or other devices, and in many industrial processes, causing environmental pollution. Catalytic oxidation is one of the most effective methods to abate CO by converting CO to CO₂. So far, the catalysts used for CO oxidation include supported noble metal catalysts (e.g.Pt, Ru, and Au),^1–6 transition metal oxides (e.g.CuO, Cu–Mn–O, Co₂O₃)^7,8 and others. Although noble metal catalysts have high activity, their high cost and poor stability limits their application.⁶ Therefore, there is a strong demand to develop a novel, thermally stable and low cost catalyst for CO oxidation, and especially for CO oxidation at low temperature. It has been reported that the CeO₂ catalyst shows great oxygen storage/release properties and high activity for CO oxidation, and its activity can be improved significantly by adding other transition metal oxides, such as CuO,⁹ for instance, over the Ce₇Cu₃–O catalyst T₁₀ (the reaction temperature for 10% CO conversion) reached 76 °C. It has been reported that the CeO₂–CuO catalyst is equal or superior to platinum catalysts for the preferential oxidation of CO in excess hydrogen,^10–12 but its activity is relatively low compared with nano-gold catalysts.

Iron oxide has been used in many catalytic processes, such as the water–gas shift reaction, the oxidation of phenol in the liquid phase,^13,14 and selective removal of CO in methanol reformed gas.^15–19Iron oxide supported noble metal catalysts have been demonstrated to be very effective for low-temperature CO oxidation. Haruta et al.²⁰ firstly developed the Au–Fe₂O₃ catalyst for CO oxidation at room temperature. Deng et al.²¹ successfully prepared ferric hydroxide supported 4 wt% Pd catalyst, over which the complete oxidation of CO could be achieved at temperatures as low as −15 °C. It was reported that as a large amount of oxygen can adsorb on FeO_x, in the Pt–FeO_x catalyst FeO_x acting as an oxygen supply can provide sufficient active oxygen species for CO oxidation.²² Cao et al.²³ reported that CuO–Fe₂O₃ composite oxide catalysts with high surface area exhibited high catalytic activity and stability for low-temperature CO oxidation. Sirichaiprasert et al.²⁴ reported that the Cu–Fe–Ce–O composite oxide catalyst prepared by urea–nitrate combustion and a single-step citrate method was used in the selective oxidation of CO to CO₂ in a hydrogen stream, and using the Cu_0.15Ce₁Fe_0.5–O catalyst 50% CO conversion was obtained at ∼115 °C. We have found that adding an appropriate amount of Fe in CeO₂ can form Ce_1−xFe_xO_2−δ solid solutions (x ≤ 0.2) and the Ce_0.9Fe_0.1O₂ solid solution presents higher catalytic performance for CH₄ and CO oxidation.²⁵ Based on the research results reported, we have developed the Cu–Fe–Ce–O catalyst for low temperature CO oxidation, and are trying to improve its catalytic performance to satisfy the demands of purifying CO in air pollution, which has barely been reported.

Herein, the Cu–Ce–Fe–O composite oxide catalyst was prepared by the citrate sol–gel method, which shows the high catalytic activity for CO oxidation (its T₉₀ is 50 °C). The influence of Fe doping on the properties of the CuO–CeO₂ composite oxide was investigated by means of the nitrogen adsorption–desorption, XRD, CO pulse and CO-TPR and CO-TPD techniques, and the role of Fe in the Cu–Ce–O catalyst was discussed.

2. Experimental

2.1 Catalyst preparation

The Cu₁Ce₈Fe₁–O catalyst was prepared by the citrate sol–gel method. A mixture of Ce(NO₃)₃·6H₂O, Cu(NO₃)₂·3H₂O and Fe(NO₃)₃·9H₂O with a molar ratio of Cu : Ce : Fe = 1 : 8 : 1 was dissolved in de-ionized water. The obtained solution was stirred at 90 °C until a viscous gel was formed. Then this gel was dried at 100 °C overnight to form a spongy material of Cu₁Ce₈Fe₁–O citrate precursor, and finally calcined in N₂ at 600 °C for 2 h, and then calcined in air at 400 °C for 2 h. Changing the amount of Fe(NO₃)₃·9H₂O in the synthesis solution, a series of Cu₁Ce₈Fe_x–O catalysts with different Fe amounts were prepared, in which x = 0.5–2. The compositions of the catalysts were determined by ICP–AES technology (Varian 710), and the results are shown in Table 1 and are similar to the compositions of the synthesis solutions. The samples of CeO₂, Fe₂O₃ and CuO were also prepared with the same method above.

Table 1 Compositions of the catalysts measured by an ICP–AES instrument

Catalyst	Cu : Ce : Fe (mol)
Cu1Ce8Fe0.5–O	1 : 7.8 : 0.49
Cu1Ce8Fe0.95–O	1 : 8.0 : 0.91
Cu1Ce8Fe1–O	1 : 7.9 : 0.97
Cu1Ce8Fe1.05–O	1 : 7.8 : 1.02
Cu1Ce8Fe2–O	1 : 7.9 : 2.05

---

2.2 Catalyst characterization

Powder X-ray diffraction patterns were recorded on a Rigaku D/max 2250VB/PC diffractometer with Cu-Kα radiation (λ = 1.5406 Å) at scanning rate of 6° min⁻¹. The average crystalline size was determined by the Scherrer formula based on the CeO₂ (111) diffraction peak broadening. N₂ adsorption–desorption isotherms were measured at −196 °C on a NOVA 4200e surface area and pore size analyzer. The samples were outgassed at 180 °C for 4 h before testing. The Brumauer–Emmett–Teller (BET) method was used to calculate the specific surface areas of samples.

The CO temperature-programmed reduction (CO-TPR) was conducted in a conventional flow system with a quartz tube reactor packed 100 mg sample. Before testing, the catalyst was pre-treated at 400 °C for 40 min in the mixture gas of 20% O₂/He (50 ml min⁻¹) and then cooled down to room temperature, and swept with He for 1 h. CO-TPR was run in 5% CO/He flow (50 ml min⁻¹) at a heating rate of 10 °C min⁻¹ from room temperature to 500 °C. The effluent gas was monitored by on-line quadrupole mass spectrometry (MS, IPC 400, INFICON Co. Ltd.).

The CO pulse experiments were conducted on the quartz tube reactor and the effluent gas was monitored by on-line quadrupole MS. 100 mg catalyst was used and pretreated at 400 °C for 40 min in the mixture gas of 20% O₂/He (50 ml min⁻¹), and then cooled down to 60 °C in He atmosphere. After the system was reached the equilibrium, 35.6% CO/Ar was pulsed into the reactor system at an interval of 30 s with a loop volume of 73.7 μl.

The CO-TPD experiments were performed in a quartz tube reactor at atmospheric pressure. 100 mg catalyst was pretreated in the mixture gas of 20% O₂/He (50 ml min⁻¹) at 400 °C for 40 min to remove the surface impurities. After the reactor was cooled down to room temperature, 35.6% CO/Ar (20 ml min⁻¹) was introduced through the catalyst till the saturated adsorption, then He was purged instead of CO/Ar and was heated from room temperature to 400 °C at a heating rate of 10 °C min⁻¹. The effluent gas was monitored by an on-line quadrupole MS.

2.3 Testing of catalytic activity

The catalytic activities of all catalysts for CO oxidation were carried out in a quartz tube reactor packed 0.20 g catalyst at atmospheric pressure. The reagent gas was consisted of 1% CO–20% O₂/N₂ and the flow rate was 50 ml min⁻¹. A space velocity (SV) was 15 000 ml (g h)⁻¹. The reactants and products are analysed by on-line GC with flame ionization detector (FID).

3. Results and discussion

3.1 Testing of catalytic activity

Fig. 1 shows the catalytic performances of CeO₂, Fe₂O₃, CuO, Cu₁Ce₉–O and Cu₁Ce₈Fe₁–O for CO oxidation, and T₁₀ (the reaction temperature for 10% CO conversion), T₅₀ and T₉₀ are shown in Table 2. It can be seen that CO conversion on CeO₂ is the lowest, and the activity of CuO is similar to that of Fe₂O₃, for instance, T₉₀ (the reaction temperature for 90% CO conversion) on the CuO and Fe₂O₃ catalysts is 150 °C and 160 °C, respectively. When CeO₂ is doped by CuO to form Cu₁Ce₉–O mixed oxides, its activity can be improved significantly, for example, T₉₀ reaches to 70 °C. After Fe doping into Cu₁Ce₉–O, its catalytic activity can be further increased, T₉₀ reaches to 50 °C.

Fig. 1 Catalytic activities of CeO₂ (▲), Fe₂O₃(◆), CuO (□), Cu₁Ce₉–O (●) and Cu₁Ce₈Fe₁–O (■) for CO oxidation.

Table 2 BET surface areas (SA), crystallite sizes (d) and CO oxidation activities over CeO₂, Cu₁Ce₉–O and Cu₁Ce₈Fe₁–O

Catalyst	T 10 (°C)	T 50 (°C)	T 90 (°C)	d (nm)	SA (m^2 g^−1)
CeO2	70	180	240	7.8	25
Cu1Ce9–O	20	50	70	7.4	120
Cu1Ce8Fe1–O	10	30	50	5.2	164

---

Fig. 2 shows the effect of Fe amount on the catalytic activities of Cu₁Ce₈Fe_x–O catalysts for CO oxidation. When x = 1, the Cu₁Ce₈Fe₁–O catalyst behaves the best activity, T₁₀ = 10 °C, T₅₀ = 30 °C and T₉₀ = 50 °C. When x >1 or x < 1, the activities of catalysts declined obviously, compared with the activity of the Cu₁Ce₈Fe₁–O catalyst; even if x = 0.95 or x = 1.05, the catalytic activities of catalysts are lower than that of the Cu₁Ce₈Fe₁–O catalyst.

Fig. 2 Catalytic activities of Cu₁Ce₈Fe_0.5–O (●), Cu₁Ce₈Fe_0.95–O (★), Cu₁Ce₈Fe₁–O (■), Cu₁Ce₈Fe_1.05–O (☆) and Cu₁Ce₈Fe₂–O (▲) for CO oxidation.

The stability of Cu₁Ce₈Fe₁–O for CO oxidation has been tested in 1% CO–20% O₂/N₂ at 65 °C, and the results are shown in Fig. 3. The results show that 100% CO conversion over Cu₁Ce₈Fe₁–O can be maintained for ∼2 h, and then the CO conversion reduces gradually to 90% after which it hardly varied within 90 h.

Fig. 3 CO conversion as a function of the reaction time over the Cu₁Ce₈Fe₁–O catalyst at 65 °C. (1% CO–20% O₂/N₂ balance, WHSV 15 000 ml (gh)⁻¹).

The above results show that, doping with the appropriate amount of Fe in the Cu₁Ce₈–O catalyst can obviously improve its catalytic activity and increase the CO conversion; the Cu₁Ce₈Fe₁–O catalyst exhibits higher stability for CO oxidation.

3.2 XRD and BET surface area

The XRD patterns of the catalysts are shown in Fig. 4. It can be found that no diffraction peaks of CuO and Fe₂O₃ are observed in the XRD patterns of Cu₁Ce₉–O and Cu₁Ce₈Fe₁–O, indicating that CuO and Fe₂O₃ are finely dispersed in the catalysts to form a solid solution or their crystal size is too small to be detected. The crystalline size of Cu₁Ce₉–O is 7.4 nm, which is smaller than that of pure CeO₂ (7.8 nm). After doping Fe into Cu₁Ce₉–O, its crystalline size further diminishes, for instance, the crystalline size of Cu₁Ce₈Fe₁–O is only 5.2 nm, resulting in its highest catalytic activity. It can also be seen that the diffraction peaks of CeO₂ in the XRD patterns of Cu₁Ce₉–O and Cu₁Ce₈Fe₁–O hardly shift but rather its diffraction peaks intensify in a comparable way to that of pure CeO₂, which has rarely been reported. Fig. 5 shows the TEM images of CeO₂, Cu₁Ce₉–O and Cu₁Ce₈Fe₁–O samples. It can be seen that after doping Cu or Cu–Fe–O into CeO₂, the crystalline sizes of sample become obviously smaller, and are similar results obtained by XRD (Table 2).

Fig. 4 XRD patterns of (1) Fe₂O₃, (2) CuO, (3) CeO₂, (4) Cu₁Ce₉–O and (5) Cu₁Ce₈Fe₁–O.

Fig. 5 TEM images of (a) CeO₂, (b) Cu₁Ce₉–O, and (c) Cu₁Ce₈Fe₁–O.

The data in Table 2 show that the BET surface area of CeO₂ is 25 m² g⁻¹, and the surface area of Cu₁Ce₉–O reaches 120 m² g⁻¹. After doping Fe₂O₃ into Cu₁Ce₉–O, the BET surface area of Cu₁Ce₉Fe₁–O further increases to 164 m² g⁻¹. It is very obvious that, the higher surface area of Cu₁Ce₉Fe₁–O compared to that of Cu₁Ce₉–O and CeO₂ is one of the reasons that Cu₁Ce₉Fe₁–O possesses the excellent catalytic performance, because the catalyst with higher surface area can provide more active sites to enhance its catalytic activity.^26,27

3.3 CO-TPR

The CO-TPR (CO₂ produced) profiles of CuO, Fe₂O₃, CeO₂, Cu₁Ce₉–O and Cu₁Ce₉Fe₁–O samples are shown in Fig. 6, in which the curves of CO₂ produced are the same as the curves of CO consumption. The results show that there is a broad and asymmetric peak of CO₂ desorption at 160–500 °C (top temperature, 255 °C) in the CO-TPR profile of CuO, which can be assigned to direct reduction of CuO to metallic copper.^28,29 In the CO-TPR profile of Fe₂O₃ there is a CO₂ peak at ∼315 °C and a broad band at >350 °C, and in the CO₂ desorption curve of CeO₂, no obvious desorption peak of CeO₂ can be observed at <500 °C, which indicates that CeO₂ is hardly reduced by CO at <500 °C. It can also be seen that there are two CO reduction peaks at 103 °C and 182 °C in the CO-TPR profile of Cu₁Ce₉–O, and the former should be assigned to reduction of the surface oxygen species,^30,31 and the latter corresponds to the reduction of lattice oxygen.

Fig. 6 CO₂ (m/z = 44) evolution curves in CO-TPR-MS of (1) CuO, (2) Fe₂O₃, (3) CeO₂, (4) Cu₁Ce₉–O and (5) Cu₁Ce₈Fe₁–O.

After Cu₁Ce₉–O is modified with Fe, its reduction property is promoted obviously, for instance, the first reduction peak falls down from 103 °C to 74 °C and second reduction peak falls from 182 °C to 162 °C, indicating that the surface oxygen and lattice oxygen of Cu₁Ce₈Fe₁–O are more active than that of Cu₁Ce₉–O. The fact that the peak area of low temperature CO reduction of Cu₁Ce₈Fe₁–O is larger than that of Cu₁Ce₉–O indicates that Cu₁Ce₈Fe₁–O possesses a much larger amount of surface oxygen species. The results above may deduce that more highly active and large amounts of the surface oxygen species are one of the main reasons for Cu₁Ce₈Fe₁–O displaying high catalytic activity and better reducibility by CO.

CO-TPR of Cu₁Ce₈Fe₁–O was repeatedly measured three times and the results are shown in Fig. 7. The results exhibit that the CO-TPR profiles of Cu₁Ce₈Fe₁–O obtained the second and third time are similar to that of the fresh catalyst, indicating that the reduction temperature of Cu₁Ce₈Fe₁–O is hardly changed after redox cycles.

Fig. 7 The CO₂ (m/z = 44) curves in CO-TPR-MS of Cu₁Ce₈Fe₁–O (1), second run (2) (after first testing of CO-TPR, this catalyst was treated at 500 °C under 20% O₂/He for 1 h, and then cooled down to room temperature), and third run (3) (repeat of the second run).

3.4 CO pulse experiments

The surface oxygen properties of Cu₁Ce₉Fe₁–O and Cu₁Ce₉–O were investigated by the CO pulse experiment, and the results are shown in Fig. 8. It can be seen that the intensities of CO signals (m/z = 28) at the first several pulses on Cu₁Ce₉Fe₁–O are much lower than that on Cu₁Ce₉–O, as the red line indicates. The amounts of CO consumption on Cu₁Ce₉Fe₁–O and Cu₁Ce₉–O were measured and are 3.13 mmol g⁻¹_cat and 2.27 mmol g⁻¹_cat, respectively. These results indicate that more CO molecules was reduced by the active oxygen species of Cu₁Ce₉Fe₁–O and further supports the results of CO-TPR above, that is, Cu₁Ce₈Fe₁–O has a larger amount of the active oxygen species than Cu₁Ce₉–O.

Fig. 8 CO pulse experiment at 60 °C on (1) Cu₁Ce₉–O, (2) Cu₁Ce₈Fe₁–O.

3.5 CO-TPD

The CO-TPD technique was used to investigate the CO adsorption on the catalysts. As shown in Fig. 9, there are three peaks located at 120 °C, 160 °C and 192 °C in the CO-TPD (CO₂ produced) profile of Cu₁Ce₉–O, indicating that there are three kinds of CO adsorption sites on its surface. Similarly, there are also two larger peaks of CO₂ desorption at 84 °C and 154 °C in the CO-TPD profile of Cu₁Ce₈Fe₁–O. Compared with that of Cu₁Ce₉–O, the desorption peaks of CO₂ in the CO-TPD profile of Cu₁Ce₈Fe₁–O shift to lower temperature, which implies that CO adsorbed on Cu₁Ce₈Fe₁–O is oxidized more easily by its surface oxygen species to CO₂ at lower temperature than on Cu₁Ce₉–O. This is a reason that the catalytic performance of Cu₁Ce₈Fe₁–O is higher than that of Cu₁Ce₉–O for CO oxidation at low temperature.

Fig. 9 CO₂ (m/z = 44) evolution curves in CO-TPD-MS of (1) Cu₁Ce₉–O and (2) Cu₁Ce₈Fe₁–O.

3.6 Discussion about role of Fe in the Cu–Ce–O catalyst

The testing of the catalytic activities of catalysts (Fig. 1) shows that adding Fe in Cu₁Ce₉–O can improve obviously its catalytic activity for CO oxidation, which can be attributed to following reasons: Firstly, after doping Fe into Cu₁Ce₉–O, the crystallite size of Cu₁Ce₈Fe₁–O was decreased from 7.8 nm to 5.2 nm, and its BET surface area was significantly increased from 120 m² g⁻¹ to 164 m² g⁻¹. Secondly, the surface oxygen species on the Cu₁Ce₈Fe₁–O catalyst is much more active than that on Cu₁Ce₈–O, for instance its reduction peak of surface oxygen species (Fig. 6) and its CO₂ desorption peaks (Fig. 9) shift to lower temperature. Also, the amount of surface oxygen species on Cu₁Ce₈Fe₁–O is more than that on Cu₁Ce₈–O, for instance the reduction peak intensity of surface oxygen species of Cu₁Ce₈Fe₁–O is much stronger (Fig. 6) and the intensities of CO signals (m/z = 28) at the first several pulses on Cu₁Ce₉Fe₁–O are lower (Fig. 8) than that on Cu₁Ce₉–O. These may be due to the smaller particle size and higher surface area of Cu₁Ce₉Fe₁–O.

The CO pulse experiment (Fig. 8) shows that after the first several pulses, the pulsed CO is mainly converted by the lattice oxygen species migrated from the sub-surface or bulk of catalyst, and the migration rate of lattice oxygen in Cu₁Ce₈Fe₁–O is faster than in Cu₁Ce₉–O. The change slope of the CO signals is obviously milder in the spectrum of Cu₁Ce₉Fe₁–O than in that of Cu₁Ce₉–O with an increase in the time or pulses of CO, which is induced by different amounts of lattice oxygen or different rates of oxygen migration.

For the CO oxidation over Cu–Ce–O catalyst, there is the Mars–van Krevelen mechanism⁹ or Langmuir–Hinshelwood mechanism³² or both concurrences that the Langmuir–Hinshelwood mechanism was the main reaction mechanism and the Mars–van Krevelen mechanism also existed.³³ We thought⁹ the Langmuir–Hinshelwood mechanism may exist as the incidental mechanism in the CO oxidation over the Cu–Ce–O catalyst, and it follows the Mars–van Krevelen mechanism as follows:

CO + * → CO* (1)

CO* + O_L → COO_L* (2)

COO_L* → * + CO₂ + V_L (3)

O₂ + V_L → 2O_L (4)

In eqn (1)–(4), “*” represents an active site, and “O_L” and “V_L” denote the lattice oxygen and oxygen vacancy in the catalyst, respectively. In the Cu₁Ce₈Fe₁–O catalyst, the Cu species exists as Cu⁺, Cu²⁺ and Cu⁰, and among them the Cu⁺ has the strongest ability to adsorb CO.^34,35 Ceria has unique oxygen storage properties; adding ceria into CO oxidation catalysts can obviously improve their catalytic performance. When x ≤ 0.2 Ce_1−xFe_xO_2−δ solid solutions can be formed, and the presence of Fe in CeO₂ can obviously improve the lattice oxygen migration, formation of oxygen vacancies and reducibility of the CeO₂ catalyst, resulting in an increase of its catalytic activity for CO oxidation.²⁵ Like Ce₉Fe₁–O solid solution, the presence of Fe in Cu₁Ce₈Fe₁–O catalyst can improve the lattice oxygen migration, formation of oxygen vacancies and its reducibility, resulting in the acceleration the reactions of eqn (2) and (3).

Conclusions

In summary, the Cu₁Ce₈Fe₁–O catalyst has been prepared by improved citrate sol–gel method with incorporation of thermal treatment under N₂, and displays excellent catalytic activity for CO oxidation, T₉₀ is 50 °C. Compared with the Cu₁Ce₉–O catalyst, the presence of Fe can increase the surface area of catalyst and diminish the crystalline size. The surface oxygen species on the Cu₁Ce₈Fe₁–O catalyst is much more active than that on Cu₁Ce₈–O, and the amount of surface oxygen species on the former is more than that on the latter. The higher catalytic activity of Cu₁Ce₈Fe₁–O for CO oxidation is due to its higher surface area and smaller crystalline size, and the more highly active and much larger amounts of the surface active oxygen species on its surface.

Acknowledgements

This project was supported financially by the National Basic Research Program of China (2010CB732300), the Fundamental Research Funds for the Central Universities, and the “Shu Guang” Project (10GG23) and Leading Academic Discipline Project (J51503) of Shanghai Municipal Education Commission and Shanghai Education Development Foundation.

Notes and references

1. J. L. Ayastuy, M. P. Gonzalez-Marcos, J. R. Gonzalez-Velasco and M. A. Gutierrez-ortiz, Appl. Catal., B, 2007, 70, 532–541.
2. Y. H. Wang, J. L. Zhu, J. C. Zhang, L. F. Song, J. Y. Hu, S. L. Ong and W. J. Ng, J. Power Sources, 2006, 155, 440–446.
3. S. Y. Chin, O. S. Alexeev and M. D. Amiridis, Appl. Catal., A, 2005, 286, 157–166.
4. L. H. Chang, N. Sasirekha and Y. W. Chen, Catal. Commun., 2007, 8, 1702–1710.
5. F. Mariño, C. Descorme and D. Duprez, Appl. Catal., B, 2004, 54, 59–66.
6. C. N. Costa, S. Y. Christou, G. Georgiou and A. M. Efstathiou, J. Catal., 2003, 219, 259–272.
7. M. F. Luo, J. M. Ma, J. Q. Lu and Y. J. Wang, J. Catal., 2007, 246, 52–59.
8. J. J. Shao, P. Zhang, X. F. Tang, B. C. Zhang, W. Song, Y. D. Xu and W. J. Shen, Chin. J. Catal., 2007, 28, 163–169.
9. Y. Liu, C. Wen, Y. Guo, G. Z. Lu and Y. Q. Wang, J. Phys. Chem. C, 2010, 114, 9889–9897.
10. P. G. Harrison, I. K. Ball, W. Azalee, W. Daniell and D. Goldfarb, Chem. Mater., 2000, 12, 3715–3725.
11. S. Hocevar, U. O. Krasovec, B. Orel, A. S. Arico and H. Kim, Appl. Catal., B, 2000, 28, 113–125.
12. P.-O. Larsson and A. Andersson, Appl. Catal., B, 2000, 24, 175–192.
13. J. Carriazo, E. Guélou, J. Barrault, J. M. Tatibouët, R. Molina and S. Moreno, Catal. Today, 2005, 107–108, 126–132.
14. J. G. Carriazo, L. M. Martınez, J. A. Odriozola, S. Moreno, R. Molina and M. A. Centeno, Appl. Catal., B, 2007, 72, 157–165.
15. Y. P. Hu, H. F. Jin, J. R. Liu and D. S. Hao, Chem. Eng. J., 2000, 78, 147–152.
16. X. Wang and R. J. Gorte, Appl. Catal., A, 2003, 247, 157–162.
17. I. L. Junior, J. M. M. Millet and M. Aouine, Appl. Catal., A, 2005, 283, 91–98.
18. N. P. Hua, H. T. Wang, Y. K. Du, M. Shen and P. Yang, Catal. Commun., 2005, 6, 491–496.
19. K. Sekizawa, S. Yano, K. Eguchi and H. Arai, Appl. Catal., A, 1998, 169, 291–297.
20. M. Haruta, N. Yamada, T. Kobayashi and S. Iijima, J. Catal., 1989, 115, 301–309.
21. B. T. Qiao, L. Q. Liu, J. Zhang and Y. Q. Deng, J. Catal., 2009, 261, 241–244.
22. L. Liu, F. Zhou, L. Wang, X. J. Qi, F. Shi and Y. Q. Deng, J. Catal., 2010, 274, 1–10.
23. J. L. Cao, Y. Wang, X. L. Yu, S. R. Wang, S. H. Wu and Z. Y. Yuan, Appl. Catal., B, 2008, 79, 26–34.
24. K. Sirichaiprasert, A. Luengnaruemitchai and S. Pongstabodee, Int. J. Hydrogen Energy, 2007, 32, 915–926.
25. D. S. Qiao, G. Z. Lu, X. H. Liu, Y. Guo, Y. Q. Wang and Y. L. Guo, J. Mater. Sci., 2011, 46, 3500–3506.
26. M. F. Luo, Y. P. Song, X. Y. Wang, G. Q. Xie, Z. Y. Pu, P. Fang and Y. L. Xie, Catal. Commun., 2007, 8, 834–838.
27. W. H. Shen, X. P. Dong, Y. F. Zhu, H. R. Chen and J. L. Shi, Microporous Mesoporous Mater., 2005, 85, 157–162.
28. Hornes, P. Bera, A. L. Camara, D. Gamarra, G. Munuera and A. Martinez-Arias, J. Catal., 2009, 268, 367–375.
29. X. Q. Wang, J. C. Hanson, A. I. Frenkel, J. Y. Kim and J. A. Rodriguez, J. Phys. Chem. B, 2004, 108, 13667–13673.
30. H. B. Zou, S. Z. Chen and W. M. Lin, J. Nat. Gas Chem., 2008, 17, 208–211.
31. H. Q. Zhu, Z. F. Qin, W. J. Shan, W. J. Shen and J. G. Wang, J. Catal., 2004, 225, 267–277.
32. W. Liu and M. Flytzani-Stephanopoulos, J. Catal., 1995, 153, 317–332.
33. J. B. Wang, D.-H. Tsai and T.-J. Huang, J. Catal., 2002, 208, 370–380.
34. M. Tada, R. Bal, X. D. Mu, R. Coquet, S. Namba and Y. Iwasawa, Chem. Commun., 2007, 4689–4691.
35. H. Q. Wan, Z. Wang, J. Zhu, X. W. Li, B. Liu, F. Gao, L. Dong and Y. Chen, Appl. Catal., B, 2008, 79, 254–261.

---