Well-dispersed gallium-promoted sulfated zirconia on mesoporous
MCM-41 silica
Wei Wanga, Chang-Lin Chena;*, Nan-Ping Xua, Song Hana, Tao Lib, Soo®n Chengb and Chung-Yuan Moub
aCollege of Chemical Engineering, Nanjing University of Technology, Nanjing 210009, China bDepartment of Chemistry, National Taiwan University, 1 Roosevelt Road, Section 4, Taipei 106, Taiwan
Received 11 April 2002; accepted 11 July 2002
Gallium-promoted sulfated zirconia (SZ) was con®ned inside pure-silica MCM-41 (abbreviated as SZGa/MCM-41), where the latter served as a host material. It was prepared by direct dispersion of metal sulfate in the as-synthesized MCM-41 materials, followed by thermal decomposition. The SZGa/MCM-41 catalysts were characterized by XRD, N2adsorption, HRTEM, DRIFT, NH3±TPD, and
TPR. The experimental results showed that the ordered porous host structure was still maintained in the catalyst. SZ was in meta-stable tetragonal phase and highly dispersed on the interior surface of MCM-41 even at a high loading of 50 wt%. Additionally, a small fraction of SZ nanoparticles on the external surface of MCM-41 was obtained. The catalytic activity of SZGa/MCM-41 was examined in n-butane isomerization. In comparison to SZ/MCM-41 without promoter, the catalytic activities of the Ga-promoted catalysts were greatly improved. The reason proposed for the higher activity of the Ga-promoted catalysts was that Ga enhances the oxidizing ability of the catalysts.
KEY WORDS: Ga-promoted sulfated zirconia; MCM-41; butane isomerization.
1. Introduction
Mesoporous materials with highly ordered pore structures have potential applications in catalysis, sorption, and as nanostructured host/guest compounds because of their high surface areas, large pore volume and tunable uniform pore structures [1]. However, the silica walls of mesoporous materials are quite neutral and the molecular sieves themselves have little use with-out proper modi®cation on the wall. Many eorts have been devoted to the introduction of some heteroatoms such as Al [2] into the silicate framework, or the intro-duction of encapsulation of solid acids such as hetero-poly acids [3] into MCM-41 channels to generate acidic sites.
In recent years, SZ and related materials have attracted increasing attention because they were found to be highly active in catalyzing reactions of industrial importance, such as hydrocarbon isomerization. How-ever, the non-uniform pore size and relatively small sur-face area of these catalysts may limit their applications for catalyzing bulky molecules. Therefore, a composite material of SZ and meso-porous silica should greatly expand the catalytic capabilities of the material. Several papers have been published on the preparation and applications of SZ supported on various meso-porous silica [4±9].
Recently, it was reported that doping the SZ with various metals such as Fe, Mn [10], and Al [11±19] would enhance their catalytic activity in the n-butane isomerization reaction. In the present work, the research is focused on the synthesis of Ga-promoted SZ by using MCM-41 as host materials. The method is by direct disper-sion of metal sulfate in the as-synthesized MCM-41 materials. The textural properties of SZGa/MCM-41 were characterized by XRD, N2 adsorption, and HRTEM. Experiments show that the procedure leads to highly dispersed SZ onto the external and internal surfaces of MCM-41. The catalytic activities of SZGa/MCM-41 in n-butane isomerization were investigated. In comparison with SZ/MCM-41, the catalytic activity of SZGa/MCM-41 was improved greatly. The reasons for the higher activity of the promoted catalysts were discussed.
2. Experimental
2.1. Catalyst preparation
As-synthesized pure siliceous MCM-41 was prepared according to the literature [20]. The as-synthesized meso-porous material was slurry impregnated with a desired amount of Zr(SO4)2 in methanol slurry and stirred at room temperature for about 10 h. The resulting sample was dried at 80 8C. Finally, it was calcined at a desired temperature for 3 h in air. Ga-promoted samples were prepared in the same way with the desired amounts of Zr(SO4)2and Ga2(SO4)3mixed as methanol slurry.
1011-372X/02/1100-0281/0 # 2002 Plenum Publishing Corporation * To whom correspondence should be addressed.
2.2. Catalyst characterization
XRD patterns of the samples were obtained on a Bruker D8 Advance instrument with Cu K radiation at 40 kV and 30 mA. The surface-area and pore-size distribution measurements were carried out on a Micro-meritics ASAP 2010 automatic adsorption instrument using N2 as the analysis gas. DRIFT spectra of the samples were acquired as previously described [18]. HRTEM was performed on a CM-200 Philips, with an operating voltage of 200 keV. TPR and NH3-TPD experiments were performed on a CHEMBET-3000, equipped with a thermal conductivity detector (TCD). TPR pro®les were recorded in 10% H2/Ar of 30 ml/min from 100 8C to 700 8C at 10 8C/min, while NH3±TPD pro®les were recorded in an He ¯ow of 30 ml/min from 100 8C to 600 8C at 10 8C/min. Sulfur content in the catalysts was detected by a chemical method. The sulfate was turned into BaSO4and determined by a gravimetric method.
2.3. Catalytic experiments
The catalytic activity of samples was tested in an n-butane isomerization reaction using a ®xed-bed continu-ous ¯ow reactor. The reactor was operated at atmospheric pressure. 0.6 g of the catalyst was loaded into the reactor and then pretreated in ¯owing dry air (20 ml/min) at 450 8C for 3 h. The reactor temperature was then lowered to the reaction temperature of 250 8C. The feed gas n-butane/H2mixtures (1:10 v/v) ¯owed through the catalyst bed at an n-butane WHSV of 0.52 hÿ1. An on-line SP6800A gas chromatograph equipped with FID was used to analyze the reaction products.
3. Results and discussion 3.1. X-ray diraction
Figure 1 shows the powder XRD patterns of SZGa/ MCM-41 after dierent heating treatment. All samples show a diraction peak due to (100) re¯ection at low 2 ranges, which are characteristic of an ordered porous structure. An increase in the intensity of the (100) re¯ection with an increase in calcination tempera-ture was observed (®gure 1). This may be due to the removal of the organic component and decomposition of metal sulfate in the pores. The XRD patterns (®gure 1(1)) for uncalcined samples within the range of 10± 70 8C show no peaks. This indicates that the metal sulfate may have been rather homogeneously dispersed onto the interior surfaces of MCM-41. Figure 1(2), (3), and (4) show that very weak and broad peaks in the higher 2 range appeared, which can be indexed into the presence of the tetragonal ZrO2 crystalline phase. This suggests that very small SZ particles might be formed outside
the pore structure when the sample was further heated at or above 630 8C. Since particles of these dimensions are observable by XRD, the very small peaks of ZrO2 observed in the region of 10±708 implied that their con-centration was relatively low, and most of the SZ was highly dispersed on the interior surface of MCM-41 at a high loading of 50 wt% SZ.
3.2. Physico-chemical properties of the samples
Figure 2 illustrates the N2adsorption/desorption iso-therm and pore-size distribution of SZGa/MCM-41 in comparison to those of pristine MCM-41. The N2 adsorption amount decreased and the in¯ection point of the respective steps is shifted to lower values of P=P0, as expected for smaller pores. It is also noticeable that the pore-size distribution remains narrow for SZGa/ MCM-41. These results indicate that the supported SZ has been well dispersed onto the interior surface of MCM-41. Table 1 shows the sulfur content, BET surface area, and total pore volume of series SZGa/MCM-41 samples. It is noticed that the BET surface area and pore volume reduce with the deposition of SZ-modi®ed samples in comparison to that of parent MCM-41. Figure 1. XRD patterns of the composites of SZGa/MCM-41 (ZrO2:
50 wt%, Ga: 1.7 wt%) calcined at various temperatures. (1) 80 8C; (2) 630 8C, 3 h; (3) 680 8C, 3 h; (4) 720 8C, 3 h.
Figure 2. N2adsorption/desorption isotherms and pore-size distribution
curves of (a) MCM-41, and (b) calcined SZGa/MCM-41 (ZrO2: 50 wt%,
However, the surface area and pore volumes of SZGa/ MCM-41 samples are still much larger than those of the traditional SZ (around 100±120 m2/g [21]). Both Ga content and calcination temperature aect the sur-face area. For SZGa/MCM-41 catalysts calcined at 680 8C, the BET surface area increases to 480 m2/g with the addition of a proper amount of Ga. It can also be seen that the sulfur content decreases as the calcination temperature increases. In contrast, the sulfur content increases with the addition of Ga. It may be explained that the addition of Ga may help to stabilize the surface sulfate species during calcinations at 680 8C.
Figure 3 shows the HRTEM picture of SZGa/MCM-41 after calcination at 680 8C. From the HRTEM micro-graph of SZGa/MCM-41, well-ordered channels with continuous walls are clearly observed. On the other hand, some external SZ particles were in fact present and the dispersed particles were observed around 10 nm. It is clear that the sizes of SZ particles on the outside surface exceeded the sum of the pore diameter and the wall thickness. It can be concluded that the use of elevated temperature leads to the premature
formation of large SZ particles outside of the pores. However, the dispersed particles or layers formed in the channels of MCM-41 are very small. Porous MCM-41 gives a physical constraint to prevent the for-mation of large SZ agglomerates. To con®rm the pre-sence of SZ in the channels of MCM-41, the DRIFT spectra of MCM-41 and SZGa/MCM-41 calcined at 680 8C were shown in ®gure 4. The spectra were taken after the samples were heated at 300 8C for 1 h. Two strong absorption peaks at 1344 cmÿ1 and at 3740 cmÿ1 are observed for MCM-41, which correspond to the Si±O stretching vibration and the O±Hvibration asso-ciated with the free SiO±Hgroups, respectively. Upon supporting SZ on MCM-41, the intensity of both peaks was markedly reduced. The decreased absorbance is presumably due to the dispersion of SZ on MCM-41 and the chemical interaction of SZ with silanols. 3.3. Acidity measurements
The NH3±TPD is a simple method widely used to inves-tigate both the strength and the number of acid sites present on the surface of acid solid. The NH3±TPD pro-®les of the SZ/MCM-41 and SZGa/MCM-41 samples after calcination at 680 8C are shown in ®gure 5. They present only one wide peak of desorption at about 250 8C. Clearly, the strength of acid sites present on SZ/ MCM-41 is not aected by the presence of Ga.
Table 1
Physico-chemical properties of the catalysts and support.
Catalysts Ga content
(wt%) Calc. temp.(8C) S content(wt%) BET S.A.(m2/g) Pore volume(ml/g)
SZ/MCM-41a 0.0 680 0.91 442 0.31 SZGa/MCM-41a 0.57 680 1.26 438 0.32 SZGa/MCM-41a 1.7 680 1.45 480 0.36 SZGa/MCM-41a 2.84 680 1.63 357 0.27 SZGa/MCM-41a 1.7 630 2.19 442 0.31 SZGa/MCM-41a 1.7 720 0.63 350 0.26 MCM-41 ± 680 ± 1010 1.10
aSamples contain 50 wt% ZrO 2.
Figure 3. HRTEM image of SZGa/MCM-41 calcined at 680 8C (ZrO2:
50 wt%, Ga: 1.7 wt%).
Figure 4. DRIFT spectra of samples calcined at 680 8C after the powder samples were heated at 300 8C for 1 h. (a) Calcined MCM-41; (b) SZGa/ MCM-41 (ZrO2: 50 wt%, Ga: 1.7 wt%).
3.4. TPR of SZ/MCM-41 andSZGa/MCM-41
Two representative examples of TPR pro®les are shown in ®gure 6. The SZ/MCM-41, calcined at 680 8C, displays a reduction peak starting at 550 8C with the maximum centered at 660±670 8C. Conversely, SZGa/MCM-41 calcined at 680 8C displays a reduction peak starting at 450 8C with the maximum centered at 528 8C. Blank experiments con®rmed no thermal decom-position of sulfates before a temperature of 700 8C. This suggests that the presence of Ga in¯uences the oxidizing ability of SZGa/MCM-41.
3.5. Catalytic activity in n-butane isomerization
Figure 7 showed the eect of various amounts of Ga incorporated in SZ/MCM-41 on n-butane isomerization at 250 8C. The activity of the SZ/MCM-41 was much lower. The selectivity to isobutane for Ga-free catalyst is about 78%. The addition of a small amount of Ga can greatly improve the catalytic activity, and the selec-tivity to isobutene for all the Ga-promoted catalysts increases to around 88%. The n-butane conversion
increased with the Ga content up to 1.7 wt%, and then decreased as the Ga content was further increased. It may be explained that excess amounts of Ga may cover the zirconia surface and reduce the catalytically active sites. It is also observed that the steady activity values of these SZGa/MCM-41 samples are still higher than that of the SZ/MCM-41, though the initial deactivation was fast. The catalysts after reaction on stream for 6 h can be regenerated in dry air at 450 8C for 3 h. It seems that the deactivation is due to the coking deposited on the catalyst surface.
The calcination temperature used for preparation of SZGa/MCM-41 catalysts has a signi®cant eect on the catalytic activities of n-butane isomerization, as shown in ®gure 8. The initial conversion of the catalyst calcined at 680 8C was higher than those of the catalysts calcined at 630 8C and 720 8C. When the calcination temperature is at 630 8C, which is lower than the decomposition temperature of Zr(SO4)2 and Ga2(SO4)3, Zr(SO4)2and Ga2(SO4)3are partly decomposed. When the calcination temperature is at 720 8C, the amount of surface sulfated ions decreases markedly.
We try to relate the characterization study of these catalysts with the catalytic properties in order to under-stand the Ga promoter eect for n-butane isomerization. Figure 5. TPD pro®les of samples calcined at 680 8C. (1) SZ/MCM-41
(ZrO2: 50 wt%); (2) SZGa/MCM-41 (ZrO2: 50 wt%, Ga: 1.7 wt%).
Figure 6. TPR pro®les of samples calcined at 680 8C. (1) SZ/MCM-41 (ZrO2: 50 wt%); (2) SZGa/MCM-41 (ZrO2: 50 wt%, Ga: 1.7 wt%).
Figure 7. Catalytic activity of catalysts calcined at 680 8C at constant ZrO2
content 50 wt% with various Ga contents. (a) 0.0 wt%; (b) 0.57 wt%; (c) 1.7 wt%; (d) 2.84 wt%.
Figure 8. Eect of calcination temperature on butane conversion at time-on-stream of 5 min over SZGa/MCM-41 at 250 8C (ZrO2: 50 wt%, Ga:
In the literature, there are several attempts to correlate the acidity with the catalytic performance of SZ. In our case, nevertheless, the acidity characterization of the samples shows no dierence in strength of the acid sites between SZ/MCM-41 and SZGa/MCM-41. The TPR pro®les indicate that Ga enhances the oxidizing ability of SZGa/MCM-41. These results would favor the idea that SZ/MCM-41 is a bifunctional catalyst for n-butane isomerization, containing acid functionality and redox functionality [22]. It is worth noting, from TPR data, that the sulfate reduction does not occur in the temperature range where the n-butane reaction occurs. The remarkable activity of SZGa/MCM-41 is due to the enhanced redox capability.
4. Conclusions
In this work, Ga-promoted SZ can be well dispersed in the channels of MCM-41. All SZGa/MCM-41 samples display a narrow mesopore size distribution. On the other hand, under high ZrO2 loading, some SZ clusters are dispersed outside the MCM-41 structure. The proper addition of Ga enhanced the catalytic activity and product selectivity for n-butane isomerization. The best Ga load-ing in these catalysts calcined at 680 8C is around 1.7 wt%. The Ga promoter is postulated to improve the redox capability of SZGa/MCM-41.
Acknowledgment
We acknowledge the support of the Educational Department of Jiangsu Province project (00KJB530001 to C.-L.C.) and the Key Laboratory of Chemical Engi-neering and Technology of Jiangsu Province. We thank Dr. H.-P. Lin for conducting the HRTEM observations.
We also thank the CTCI foundation for doing the HRTEM observations.
References
[1] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli and J.S. Beck, Nature 359 (1992) 710.
[2] A. Corma, M.J. Climent, S. Iborra, M.C. Nawarro and J. Primi, J. Catal. 161 (1994) 569.
[3] I.V. Kozhevnikov, A. Sinnema, R.J.J. Jansen, K. Pamin and H. van Bekkum, Catal. Lett. 30 (1995) 241.
[4] T. Lei, W.M. Hua, Y. Tang, Y.H. Yue and Z. Gao, Chem. J. Chinese Univ. 21 (2000) 1240.
[5] Q.H. Xia, K. Hidajat and S. Kawi, Chem. Commun. (2000) 2229. [6] C.L. Chen, S. Cheng, H.P. Lin, S.T. Wong and C.Y. Mou, Appl.
Catal. A 215 (2001) 21.
[7] Q.H. Xia, K. Hidajat and S. Kawi, J. Catal. 205 (2002) 318. [8] W.M. Hua, Y.H. Yue and Z. Gao, J. Mol. Catal. A 170 ( 2001) 195. [9] H. Matsuhashi, M. Tanaka, H. Nakamura and K. Arata, Appl. Catal.
A 208 (2001) 1.
[10] C.Y. Hsu, C.R. Heimbuch, C.T. Armes and B.C. Gates, J. Chem. Soc. Chem. Commun. (1992) 1645.
[11] S. Baba, Y. Shibata, T. Kawamura, H. Takaoka, T. Kimura, K. Kou-saka, Y. Minato, N. Yokoyama, K. Lida and T. Imai, Eur. Patent 174836, 1985.
[12] M. Marella, M. Tomaselli, L. Meregalli and F. Pinna, Eur. Patent 983967, 2000.
[13] Z. Gao, Y. Xia, W. Hua and C. Miao, Topics Catal. 6 (1998) 101. [14] T. Lei, J.S. Xu, W.M. Hua, Y. Tang and Z. Gao, Catal. Lett. 61 (1999)
213.
[15] T. Lei, J.S. Xu, Y. Tang, W.M. Hua and Z. Gao, Appl. Catal. A 192 (2000) 181.
[16] E. Angelescu, M. Ropot, F. Contantinescu, D. Boiangiu and M. Pencea, Prog. Catal. 7 (1998) 5.
[17] J.A. Moreno and G. Poncelet, J. Catal. 203 (2001) 453.
[18] C.L. Chen, T. Li, S. Cheng, H.P. Lin, C. J. Bhongale and C.Y. Mou, Microporous Mesoporous Mater. 50 (2001) 201.
[19] C.L. Chen, T. Li, S. Cheng, N.P. Xu and C.Y. Mou, Catal. Lett. 78 (2002) 223.
[20] H.P. Lin, S. Cheng and C.Y. Mou, Microporous Mater. 10 (1997) 111. [21] B.H. Davis, R.A. Keogh and R. Srinivasan, Catal. Today 20 (1994)
219.
