Highly efficient Pd/Al₂O₃-Ce_0.6Zr_0.4O₂ catalyst pretreated by H₂ for low-temperature methanol oxidation

Abstract

We report an interesting finding that the catalytic performance of Pd/Al₂O₃-Ce_0.6Zr_0.4O₂ catalyst toward complete oxidation of methanol at low temperature could be greatly enhanced by prereduction in hydrogen. The light-off temperature (T₅₀) could be reached rapidly at 67 °C. TPR patterns showed that the reducibility is highly increased due to the formation of stronger interactions between PdO_x and CeO₂ after hydrogen pretreatment. It was confirmed by XPS analysis, and the generation of higher charged Pd^δ+ (δ > 2) species were discovered. As a result, it was induced that the presence of stronger interactions between PdO_x and CeO₂ after prereduction mainly contributes to the remarkably enhanced activity.

---

1. Introduction

Recently, air pollution caused from the gasoline engine powered vehicles along with the large consumption of petroleum has attracted worldwide attention. One satisfactory and efficient solution is to use alternative automotive fuels, such as methanol.^1–3 Methanol has several advantages relative to conventional fuel, e.g., non-sooting fuel, high oxygen content, improved energy utilization and higher engine output.^4,5

The use of methanol as a fuel can reduce the emission of conventional pollutants like CO, THC, NO_x. Nevertheless, the vehicles operating on methanol have a tendency to emit significant amount of partial oxidation products such as formaldehyde and unburned methanol vapor during the engine cold-start,⁶ which are harmful to human health and considered as major contributors of environmental pollution. It is, therefore, very essential and urgent to develop efficient complete oxidation catalysts at low temperature.

Currently, a number of metals (Pt, Pd, Rh, Ag, etc.) and metal oxides (CuO, CuO-Cr₂O₃, CuO-ZnO-Cr₂O₃, etc.) supported on γ-alumina have been proposed as catalysts for complete oxidation of methanol.^6–9 As to the activity and selectivity of the catalysts for catalytic oxidation, the supported noble metals have been generally regarded as the most desirable catalysts. It was concluded that either Pt or Pd catalysts gave the highest activity and lowest HCHO yields without CO in the feed.^6,10,11 Brewer et al.¹² reported that a monolithic catalyst containing only Pd on a γ-alumina substrate represented the best activity for methanol oxidation because of its greater durability compared with base metal, and it cost less than a platinum-only catalyst.

Until now, γ-alumina is the most commonly used support for complete oxidation of methanol. In view of the advantages of palladium and high low-temperature activity, we introduce the Al₂O₃-50 wt% Ce_0.6Zr_0.4O₂ (ACZ) supported palladium catalyst. Cerium oxide is a well-known promoter in noble metal-based combustion catalysts.¹³ In addition, it is established that Zr⁴⁺ incorporation greatly enhanced the redox property, resulting in superior catalytic oxidation. Nevertheless, summarizing the literature, the mechanism over these promoted properties in catalytic oxidation of methanol is little. On the other hand, the reason why some catalysts were pretreated by H₂ before measurement in previous researches remains obscure. In this paper, the work deals with a combined investigation of the effect of pretreatment by hydrogen on the structural/textural properties of catalyst and on its catalytic performance of methanol.

2. Experimental

2.1 Catalyst preparation

The ACZ support was prepared by conversional coprecipitation method. First, an aqueous salt solution containing the required amounts of nitrates was prepared. Then, the buffer solution of NH₃ and (NH₄)₂CO₃ was dropped into the salt solution until the pH was kept at 9. The precipitate was centrifuged using distilled water until no pH change could be detected, and subsequently dried at 105 °C overnight. The obtained cakes were calcined at 500 °C for 4 h in a muffle furnace, resulting in the final support. The total content of Ce and Zr as oxide state (CeO₂ + ZrO₂) is 50 wt%, and the molar ratio of Ce/Zr is 3 : 2.

The Pd (0.5 wt%) was deposited on the as-synthesized support with sieving size of 30–80 meshes using an aqueous solution of Pd(NO₃)₂ by incipient impregnation method. During this process, the impregnated sample was kept at room temperature for 12 h, dried at 80 °C for 4 h and calcined at 500 °C for 4 h in the muffle furnace. The obtained catalyst is labeled as Pd/ACZ, and the catalyst pretreated by pure H₂ at 300 °C is designated as Pd/ACZ-H₂.

2.2 Characterization techniques

X-ray powder diffraction (XRD) data were recorded on a Panalytical X'Pert Pro diffractometer at 40 kV and 40 mA with a step size of 0.0167°, using Co-Kα radiation and then revised by Cu-Kα.

The textural properties were carried out at −196 °C on a TriStar 3000 apparatus using nitrogen adsorption/desorption method. The samples were degassed under vacuum for 3 h at 300 °C prior to measurements.

The temperature–programmed reduction of H₂ (H₂-TPR) was conducted on an AutoChem 2920 instrument equipped with a TCD, and 50 mg sample was used in each measurement. The samples were first pretreated under He (30 ml min⁻¹) at 300 °C for 1 h, and then cooling down to ca. 5 °C. After that, the temperature of the samples was heated up to 800 °C at a constant rate of 10 °C min⁻¹ in 10% H₂-90% Ar stream.

The temperature-programmed desorption of O₂ (O₂-TPD) was performed using the same apparatus as for H₂-TPR. First, 50 mg of the sample was heated in pure O₂ from room temperature to 500 °C and held for 0.5 h, subsequently cooled to room temperature in He atmosphere. Then the sample was heated from room temperature to 800 °C in helium at a heating rate of 10 °C min⁻¹.

XPS analysis was performed on Physical Electronics Quantum 2000. A monochromatic Al-Kα source (Kα = 1,486.6 eV) and a charge neutralizer were equipped in the instrument. And the surface charging effect was referenced to the C 1s peak at a binding energy of 284.6 eV.

2.3 Catalytic activity measurement

Catalytic tests were carried out using a fixed-bed continuous-flow reactor packed with 0.1 g of catalyst. The complete oxidation of methanol over the catalysts was conducted in the stream of the feed-gas mixture with a composition of CH₃OH/O₂/N₂ = 0.2/1.0/98.8 (molar ratio) at GHSV = 24 000 h⁻¹. The reactants and products were analyzed by an on-line GC equipped with a FID, and the trend of the effluent gas was monitored by mass spectrum simultaneously. The catalyst activity was expressed by the temperatures T₅₀ and T₉₀ (°C) at which methanol conversion reached 50 and 90%, respectively.

3. Results and discussion

3.1 Catalytic methanol oxidation

Fig. 1 presents the results of the catalytic oxidation of methanol over the catalysts. As shown in the figure, it was found that the catalytic activity could be considerably improved by prereduction in H₂. The T₅₀ of Pd/ACZ-H₂ was achieved at a low temperature as low as 67 °C while the T₉₀ is reached rapidly at 111 °C, which may resolve the cold-start problem effectively in practice. Notably, The T₅₀ and T₉₀ of Pd/ACZ-H₂ shift sharply ca. 100 and 87 °C respectively below the corresponding temperature observed on Pd/ACZ catalyst, which is likely due to the change of palladium species.¹⁴

Fig. 1 Conversion activity of CH₃OH oxidation and the trend of HCHO and CO₂ in the exit-gas at different temperatures over the catalyst of (a) Pd/ACZ; (b) Pd/ACZ-H₂.

In addition, the level of HCHO emitted is increased below 140 °C for Pd/ACZ catalyst, which implies the unsatisfied complete oxidation of methanol at low temperature. Contrarily, the prereduction in H₂ shows an apparent inhibiting effect on the elimination of partial oxidation product HCHO. All in all, the catalyst Pd/ACZ-H₂ exhibits the excellent behavior of complete oxidation of methanol at low temperature.

3.2 XRD and N₂ physisorption

In order to understand the enhanced catalytic performance with pretreatment by H₂, we investigated the textural and structural properties of ACZ, Pd/ACZ and Pd/ACZ-H₂. The nitrogen adsorption/desorption isotherms are shown in Fig. 2, and the corresponding values including BET surface area, pore volume and average pore radius are listed in Table 1.

Fig. 2 Nitrogen adsorption/desorption isotherms of ACZ, Pd/ACZ and Pd/ACZ-H₂.

Table 1 Textural and structural properties of the support and its catalyst

Samples	Average pore radius/nm	Pore volume/cm^3 g^−1	SBET/m^2 g^−1	Lattice parameter^a/nm	Crystallite size^b/nm
a Lattice parameter is derived from (111), (200), (220) and (311) planes based on Bragg' law. b Crystallite size is derived from (111) peaks based on Scherrer Equation.
ACZ	1.9	0.29	250.6	0.5333	11.5
Pd-ACZ	2.3	0.26	168.9	0.5342	9.5
Pd-ACZ-H2	2.3	0.26	167.4	0.5333	10.2

---

From Fig. 2, it can be seen that the nitrogen adsorption/desorption isotherms researched are all corresponding to type IV with an H2-type hysteresis loop, indicating the ink-bottle shaped pore in detail.¹⁵ Meanwhile, there is a subtle difference between Pd/ACZ and Pd/ACZ-H₂, which implies that the prereduction affects the textural properties slightly. It can be also confirmed from the data listed in Table 1, the surface area of Pd/ACZ and Pd/ACZ-H₂ is ca. 168.0 m² g⁻¹, as well as the same value of pore volume and average pore radius. It should be pointed out here that the decrease in surface area and pore volume for catalysts, indicating that the palladium was impregnated in the channel of support.

The XRD patterns of the samples are depicted in Fig. 3. For all the samples, the diffraction peaks could be indexed to (111), (200), (220), (311) and (331) crystal faces, corresponding to the mainly cubic fluorite-structured of Zr_0.4Ce_0.6O₂ (JCPDS Card No. 38-1439). No tiny peaks related to palladium species can be observed, suggesting that either palladium species are finely dispersed on the support or they are too small to be detected by XRD analysis.¹⁶ To be mentioned, XRD pattern of support ACZ presents an unsymmetrical peak shape on (111) crystal face, indicating some kind of inhomogeneous solid solution. For comparison, Pd/ACZ and Pd/ACZ-H₂ exhibit the better symmetrical peak shape, which illustrates the enhanced homogeneity by the doped Pd. It implies that an interaction between Pd particles and support may be present in the catalysts. Moreover, the lattice parameter and the crystallite size of Pd/ACZ-H₂ change with respect to Pd/ACZ, reflecting the structural variation, such as the status of cerium and palladium. Thus, it is suggested that the prereduction will affect the strength of interaction between Pd particles and support.

Fig. 3 XRD characterization of ACZ, Pd/ACZ and Pd/ACZ-H₂.

3.3 H₂-TPR

H₂-TPR profiles of support and Pd-supported samples are shown in Fig. 4, and the corresponding values of H₂ consumption and peak temperature maxima are listed in Table 2. Regarding the support, it features one broad peak with maxima at ca. 480.7 °C and a shoulder at ca. 340 °C. The relatively low intensity of this shoulder suggests its attribution to some kind of inhomogeneity, which is well consistent with the XRD result of support mentioned above. Meanwhile, the main broad peak centered at ca. 480.7 °C could be assigned to the release of bulk and surface oxygen.

Fig. 4 H₂-TPR profiles of ACZ, Pd/ACZ and Pd/ACZ-H₂.

Table 2 H₂-TPR data of different samples

Samples	Peak position (°C)			H2 consumption/μmol g^−1
	T1	T2	T3	n1	n2	n3
ACZ	—	—	480.7	—	—	341.5
Pd-ACZ	115.7	311.7	—	477.9	15.1	—
Pd-ACZ-H2	60.7	297.2	—	407.1	13.0	—

---

For the catalyst, Pd/ACZ exhibits two dominant reduction peaks at 115.7 and 311.7 °C, respectively. The peak located at 115.7 °C can be attributed to the stable PdO species dispersed on ZrO₂ crystallites.¹⁷ The tiny peak at higher temperature is associated with the reduction of surface oxygen.¹⁸ It is noteworthy that the H₂ consumption at lower temperature is much greater than expected. Stoichiometrically, 1 g catalyst requires 47.0 μmol H₂ for the reduction of PdO to Pd with a Pd loading of 0.5 wt% by assuming the following complete reduction process

PdO+H₂ → Pd+H₂O

From Table 2, it can be seen that the H₂ consumption of Pd/ACZ at lower temperature is 477.9 μmol g⁻¹, much larger related to 47.0 μmol g⁻¹. It can be speculated that the partial reduction of Ce⁴⁺, intimately contacted with Pd, should overlap with the reduction of PdO. Previous researches reported that the presence of Pd promotes the reduction of CeO₂ via the spilling of hydrogen.^19,20 In other words, there is a strong interaction between PdO and the support, verifying the assumption in the XRD analysis.

In the case of Pd/ACZ-H₂, it can be noticed that two reduction peaks shift to lower temperature, indicating the better redox properties after prereduction. Especially, the peak at lower temperature shifts from 115.7 °C down to 60.7 °C, which may explain the great promotion on oxidation behaviour after hydrogen pretreatment. Previous studies reported that the PdO species in Pd/Ce-Zr catalysts could be reduced at 60–90 °C.²¹ Thus, it can be concluded that the pretreatment by H₂ will increase the interface between PdO_x species and CeO₂, such as Pd_xCe_1−xO₂ (1 1 1), leading to the least reaction energy.¹⁴ It is no wonder that a bit smaller H₂ consumption of Pd/ACZ-H₂ compared to that of Pd/ACZ is due to the reduction of part Ce⁴⁺, which is confirmed by XPS analysis.

3.4 O₂-TPD

Fig. 5 shows the O₂-TPD profiles of the support and its supported palladium catalyst with or without pretreatment by H₂. As shown in Fig. 5, it can be clearly seen that all the samples display two broad peaks of oxygen desorption at about 340 and 616 °C, respectively.

Fig. 5 O₂-TPD profiles of ACZ, Pd/ACZ and Pd/ACZ-H₂.

The α desorption peak appeared at T < 500 °C is ascribed to the oxygen chemically adsorbed on the surface, and the β desorption peak appeared at 500 < T < 800 °C corresponds to the oxygen chemically adsorbed on the oxygen vacancy.²²

For α desorption peak, the area decreases with the loading of palladium, which may be due to the loss of surface area. It was suggested that desorption of surface oxygen is closely related to the surface area.²³ The O₂-TPD profiles of catalysts show the similar oxygen desorption peaks in the temperature range investigated, indicating the similar adsorption/desorption behavior to oxygen in spite of prereduction. Furthermore, the large amount of surface and vacancy oxygen desorped for both support and catalysts verifies the benefit of introduction of ceria-zirconia oxide in this work.

3.5 XPS studies

XPS was performed to investigate the effect of prereduction on the surface of catalyst. The Ce 3d, O 1s and Pd 3d XP spectra of catalysts are presented in Fig. 6a–c, respectively. As illustrated in Fig. 6a, The curve of Ce 3d spectra comprises eight peaks corresponding to four pairs of spin–orbit doublets. Clearly, the strong peaks at 885.2 eV (v′) and 903.1 eV (u′) are typical for Ce³⁺, while the main features of Ce⁴⁺ are at 882.7 eV (v), 888.4 eV (v′′), 898 eV (v′′′), 900.7 eV (u), 906.4 eV (u′′) and 916.9 eV (u′′′).²⁴ It may be noted that the relative intensity of v′ and u′ peaks increase from Pd/ACZ to Pd/ACZ-H₂, indicating that the content of Ce³⁺ increases due to the Ce⁴⁺ reduction. This can be also proved by the percentage of u′′′ in the whole Ce 3d multiplet for its well-defined peak only attributed to Ce⁴⁺.²⁵ The value was found to be 12.8% and 11.6% for Pd/ACZ and Pd/ACZ-H₂, respectively. It indicates that the tiny reduction of Ce⁴⁺ occurs, which is in good agreement with the TPR results.

Fig. 6 XPS spectra of Pd/ACZ and Pd/ACZ-H₂. (a) Ce 3d, (b) O 1s and (c) Pd 3d.

Further information can be obtained from the O 1s spectra as shown in Fig. 6b. For each sample, the O 1s spectrum contains a main peak at ca. 531.2 eV (O_I) assignable to a mixture of hydroxyl groups and adsorbed water on the surface of the catalyst^26,27 along with a distinct shoulder at ca. 529.5 eV (O_II) characteristic of the lattice oxygen.²⁶ For quantitative determination, the value of O_II/O_I for Pd/ACZ is 0.22 while it is 0.21 after prereduction. Although the difference is small, it implies the reduced species of lattice oxygen after prereduction. In other words, the surface oxygen vacancies increase, proving the reduction of ceria is accompanied by the formation of oxygen vancancies.

Fig. 6c presents the XPS spectra for Pd 3d of Pd/ACZ and Pd/ACZ-H₂. The unreduced catalyst exhibits a weak doublet centering at 337.4 and 342.4 eV, attributed to Pd 3d_5/2 and Pd 3d_3/2, respectively. In addition, the reduced catalyst appears at relatively higher BE of 337.8 and 342.8 eV, respectively. The weak signal may be due to the combination effects of low loading and the distraction of Zr 3p. Nevertheless, the observed BE values are between that of PdO (336.3 eV)²⁸ and PdO₂ (337.9 eV),²⁹ indicating the high charged Pd^δ+ (2 < δ < 4) species exist in the catalyst. Moreover, the peak positions are shifted towards higher BE by 0.4 eV for Pd/ACZ-H₂, indicating stronger metal-support interaction (SMSI) after prereduction. That is to say, the Pd-O bonding does not belong to Pd-O-Pd but rather to Pd-O-Ce in the Pd-CeO₂ interface.³⁰ It is further confirmed that the prereduction will enhance the interface between PdO_x species and CeO₂, which should be a key to the high activity of methanol oxidation.

4. Conclusions

Highly efficient complete oxidation of methanol at low temperature could be achieved over the Al₂O₃-Ce_0.6Zr_0.4O₂ supported palladium catalyst pretreated by H₂. The T₅₀ was as low as 67 °C accompanied with small emission of HCHO, which may resolve the cold-start problems in the methanol-fueled vehicles. The results of N₂ physisorption and O₂-TPD indicated that the prereduction has little effect on the textural properties and oxygen desorption behaviour of catalysts. The analysis of XRD implied that the interaction between noble metal and support may be present. Moreover, H₂-TPR results showed that stronger interaction between PdO_x and CeO₂ exists after prereduction. It was verified by XPS analysis and the generation of higher charged Pd^δ+ (δ > 2) after H₂ pretreatment should play a positive role in the enhancement of catalytic activity.

Acknowledgements

This work was financially supported by Key Projects in the National & Technology Pillar Program, Grant No. 2007BAE08B01.

Notes and references

1. Z. Huang, H. Lu, D. Jiang, K. Zeng, B. Liu and J. Zhang, et al. , Proc. Inst. Mech. Eng., Part D, 2004, 218, 435–447.
2. M. Canakci, C. Sayin and M. Gumus, Energy Fuels, 2008, 22, 3709–3723.
3. C. Sayin, K. Uslu and M. Canakci, Renewable Energy, 2008, 33, 1314–1323.
4. Z. Huang, H. Lu, D. Jiang, K. Zeng, B. Liu and J. Zhang, et al. , Proc. Inst. Mech. Eng., Part D, 2004, 218, 1011–1024.
5. H. Bayraktar, Fuel, 2008, 87, 158–164.
6. R. W. McCabe and P. J. Mitchell, Appl. Catal., 1986, 27, 83–98.
7. H. K. Plummer Jr, W. L. H. Watkins and H. S. Gandhi, Appl. Catal., 1987, 29, 261–283.
8. R. W. McCabe and P. J. Mitchell, Ind. Eng. Chem. Proc. Res. Dev., 1983, 22, 212–217.
9. R. W. McCabe and P. J. Mitchell, Ind. Eng. Chem. Prod. Res. Dev., 1984, 23, 196–202.
10. R. W. McCabe and P. J. Mitchell, J. Catal., 1987, 103, 419–425.
11. R. W. McCabe and P. J. Mitchell, Appl. Catal., 1988, 44, 73–93.
12. T. F. Brewer and M. A. Abraham, Ind. Eng. Chem. Res., 1994, 33, 526–533.
13. L. F. Yang, C. K. Shi, X. E. He and J. X. Cai, Appl. Catal., B, 2002, 38, 117–125.
14. A. D. Mayernick and M. J. Janik, J. Catal., 2011, 278, 16–25.
15. G. Leofanti, M. Padovan, G. Tozzola and B. Venturelli, Catal. Today, 1998, 41, 207–219.
16. Z. Hua, Q. Zhang, S. Wen, S. Wen and W. Jian, J. Catal., 2004, 225, 267–277.
17. G. Wang, M. Meng, Y. Zha and T. Ding, Fuel, 2010, 89, 2244–2251.
18. Q. Wang, G. Li, B. Zhao, M. Shen and R. Zhou, Appl. Catal., B, 2010, 101, 150–159.
19. N. Hickey, P. Fornasiero, J. Kašpar, J. M. Gatica and S. Bernal, J. Catal., 2001, 200, 181–193.
20. J. Kašpar, P. Fornasiero and N. Hickey, Catal. Today, 2003, 77, 419–449.
21. M. Luo and X. Zheng, Appl. Catal., A, 1999, 189, 15–21.
22. N. Yamazoe, Y. Teraoka and T. Seiyama, Chem. Lett., 1981, 12, 1767–1770.
23. B. Zhao, C. Yang, Q. Wang, G. Li and R. Zhou, J. Alloys Compd., 2010, 494, 340–346.
24. J. P. Holgodo, R. Alvaarez and G. Munuera, Appl. Surf. Sci., 2000, 161, 301–315.
25. M. Romeo, K. Bak, J. E. Fallah, F. L. Normand and L. Hilaire, Surf. Interface Anal., 1993, 20, 508–512.
26. S. D. Gardner, G. B. Hoflund, M. R. Davidson, H. A. Laitinen, D. R. Schryer and B. T. Upchurch, Langmuir, 1991, 7, 2140–2145.
27. S. J. Lee, A. Gavriilidis, Q. A. Pankhurst, A. Kyek, F. E. Wagner, P. C. L. Wong and K. L. Yeung, J. Catal., 2001, 200, 298–308.
28. R. Gopinath, N. Lingaiah, B. Sreedhar, I. Suryanarayana, P. S. S. Prasad and A. Obuchi, Appl. Catal., B, 2003, 46, 587–594.
29. M. P. Seah, Quantification of AES and XPS, in Practical Surface Analysis: Auger and X-Ray Photoelectron Spectroscopy, ed. D. Briggs and M. P. Seah, Wiley, Chichester, 1990, pp. 201-255.
30. W. J. Shen, Y. Ichihashi and Y. Matsumura, Catal. Lett., 2002, 79, 125–127.

---