Microporous and Mesoporous Materials 122 (2009) 264–269

Contents lists available at ScienceDirect

Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso

Structural effects of hierarchical pores in zeolite composite Jiajun Zheng a, Xiwen Zhang a, Yan Zhang b, Jinghong Ma a, Ruifeng Li a,* a b

Key Laboratory of Coal Science and Technology MOE, Taiyuan University of Technology, Taiyuan 030024, China College of Material Science and Technology, Taiyuan University of Technology, Taiyuan 030024, China

a r t i c l e

i n f o

Article history: Received 12 November 2008 Received in revised form 3 March 2009 Accepted 5 March 2009 Available online 13 March 2009 Keywords: Zeolite composite Hierarchical pores Diffusion Acidity accessibility Catalytic cracking

a b s t r a c t A zeolite composite with MFI and MOR zeolite structures (denoted as MMZ) was prepared by a two-step hydrothermal crystallization, and characterized by X-ray powder diffraction, nitrogen adsorption– desorption, scanning electron microscopy and in situ IR spectrometry of pyridine and 2,6-dimethyl pyridine. The acid catalysis of the composite MMZ was investigated during the catalytic cracking of n-octane and cumene. The results showed that a hierarchical pore system was created in the zeolite composite improving the accessibility of acid sites and diffusivity in the composite material. The diffusivity in the composite was about seven times higher than in the corresponding physical mixture, and the acid accessibility in the composite was 3.4 times as much as in the physical mixture when 2,6-dimethyl pyridine was used as the probe molecule. The conversion of cumene on H-MMZ was three times larger than that on the physical mixture at 423 K. Ó 2009 Elsevier Inc. All rights reserved.

1. Introduction Zeolites have unique properties in acid catalysis and shape selectivity. Besides providing shape selectivity, the intricate pore and channel systems of zeolites in the molecular size range (0.3– 1.5 nm) lead to diffusion limitations on reaction rates, due to the similarity between the size of the involved hydrocarbons and the micropore diameter [1–3]. Intracrystalline diffusion inside a zeolite micropore at a given temperature and pressure can not be increased without changing the internal pore architecture; limiting the effective utilization of acid sites by bulky molecules. In hydrocarbon transformation over zeolites, the conversion depends on both the time the reactant molecules spend inside zeolite crystals and the probability of the reactants accessing the acid sites. An alternative solution to minimize diffusion limitation is to reduce the intracrystalline diffusion path length by decreasing crystal size so as to improve catalytic performance. The beneficial effect of small zeolite crystals on overall reaction rate is twofold. Firstly, smaller crystals have shorter intracrystalline diffusion path length; hence reaction products are released more rapidly. Accordingly, less secondary reactions like coke formation and cracking are observed. Secondly, more micropore entrances per unit mass are present in zeolites with small crystals, which induces a higher accessibility into the zeolite crystals, and therefore, a net increase in overall activity. In the past decade, there have been a considerable number * Corresponding author. Address: Taiyuan University of Technology, Institute of Special Chemicals, 79# West Yingze Street, Taiyuan 030024, China. Tel./fax: +86 351 6010121. E-mail address: rfl[email protected] (R. Li). 1387-1811/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.micromeso.2009.03.009

of attempts to improve the micropore diffusion in zeolites by adjusting their crystal sizes to nanoscale ranges. However, filtering small colloidal particles is not easy, which severely hindered their practical applications. Moreover, the volume and surface area of the micropores in zeolites was decreased because the ordering of the three-dimensional network of the zeolite particles deteriorated when the crystal size decreased to the nanocrystalline region [3]. Therefore, the synthesis of new zeolite materials containing considerable intracrystalline or intercrystalline mesopores that provide a better diffusion transport has recently attracted attention of many researchers in the field [4–7]. A practicable route to prepare the materials with an enhanced accessibility is the combination of micropores and mesopores (or macropores) in the same material, since the diffusion in mesopores is several orders of magnitude faster than in micropores. For this goal, some recent studies have applied intercrystalline approaches by which zeolite materials are assembled into ordered mesoporous structures. A more generally applied strategy to attain the materials that combine zeolite micropores with mesopores is the intracrystalline approach, in which mesopores are created in the zeolite crystals. In this way the micropores of the zeolite are effectively shortened and their molecular accessibility is largely enhanced [8–10]. The creation of mesopores in zeolite crystals is equivalent to increasing the external surface area of the zeolite, in this respect a larger number of pore windows is made accessible to the reactants. The combination of micropores with mesopores in one crystal can be undertaken by steaming [11], acid [1,6] or base leaching [12]. Mesopores resulting from acid leaching was investigated in Pt/H-mordenite by Donk and co-workers [1]. The experiment results showed that the hydroisomerization activity on Pt/H-mordenite

265

J. Zheng et al. / Microporous and Mesoporous Materials 122 (2009) 264–269

was four times higher than on untreated Pt/H-mordenite, because it gave rise to an acceleration of the uptake of n-hexane under reaction conditions due to a shorter intracrystalline diffusion path length resulting from the mesoporous structure created by acid leaching. Despite intracrystalline mesopores created by acid leaching approach was also reported by other researchers [6], alkali-leaching approach has been the most frequently employed way to create intracrystalline mesopores in zeolite materials. Groen and co-workers reported that considerable mesoporosity created in zeolites ZSM-5, BEA, MOR and FER was due to the enlargement of micropores resulting from the silicon extraction by the alkali leaching [2]. Jung also reported that the silicon extraction from micropores by the alkaline treatment created significant mesopores in ZSM-5 [13]. Intercrystalline mesopores resulting from polycrystalline accumulation were reported in zeolite A [3], which was successfully synthesized by using a template method with resorcinol–formaldehyde aerogels. This provides a new way to prepare novel zeolite materials with intercrystalline mesopores. In the present paper, we report a new method for the synthesis of hierarchically porous zeolites by employing the as-synthesized Mordenite zeolite as silica–alumina source. Intracrystalline mesopores are created in Mordenite zeolites by the treatment of basic solution and higher hydrothermal temperature during the second-step synthesis. In addition, intercrystalline macropores with 50–100 nm are formed because of the polycrystalline aggregation based on crystal growing of zeolite ZSM-5 around Mordenite zeolite particles.

2. Experiments 2.1. Synthesis First, Mordenite zeolite was prepared with the molar composition of the gel: 6Na2O:30SiO2:Al2O3:780H2O. An aliquot of 1.28 g of sodium aluminate (41 wt% Al2O3, 35 wt% Na2O) and 1.89 g of sodium hydroxide (96 wt%) were mixed in 45 ml water to form a clear solution, then 26.5 ml of silica sol (29 wt%) was slowly added to the solution with vigorous stirring. The mixture was stirred at room temperature for 2 h and transferred into 100 ml autoclave and kept at 443 K for 18–22 h without stirring. The reacted mixture contained the pre-synthesized Mordenite zeolite was directly used in the second-step synthesis. An aliquot of 0.63 g of sodium aluminate, 6.2 ml of ethylene diamine (EDA) and 0–1.0 ml of concentrated sulfuric acid (98 wt%) were added to the reacted mixture, and stirred for 2 h at room temperature. The final molar composition in the mixture of the second-step was: (2.2–4.5)Na2O:19SiO2:Al2O3:12EDA:525H2O. The new mixture was then loaded into an autoclave for hydrothermal treatment at 453 K for 48– 72 h under autogenous pressure. The as-synthesized solid product was recovered by filtration, washed with water, dried in air at 373 K overnight and denoted as MMZ. A physical mixture of Mordenite and ZSM-5 zeolites was denoted as Z + M. The NHþ 4 form of samples was obtained by repeating three times ion exchange at 353 K with 0.5 M NH4 NO3 solution, for 4 h every time. The protonic form was then obtained by calcining the NHþ 4zeolite at 823 K for 5 h and marked as H-MMZ and H-(Z + M).

NOVA 1200e gas sorption analyzer to study the micro- and mesoporosity in the zeolite crystals. The mesopore size distribution was calculated using the Barret–Joyner–Halenda (BJH) pore size model applied to the adsorption branch of the isotherm. The microporous structure was obtained from the t-plot analysis of the adsorption branch of the isotherm. Infrared spectra of pyridine and 2,6-dimethyl pyridine adsorption were obtained on a SHIMADZU FTIR8400 spectrometer. About 12 mg of sample was pressed into a self-supporting wafer of 10 mm in diameter. The wafer was first evacuated in situ in an IR cell at 573 K for 2 h, and the IR spectra were recorded at room temperature. Pyridine or 2,6-dimethyl pyridine was then introduced into the cell at room temperature until the saturated adsorption was reached. Finally, desorption of pyridine or 2,6-dimethyl pyridine was performed at increasing temperatures under 3  103 Pa of pressure and the spectra were recorded at various temperatures. 2.3. Catalytic activity All catalytic experiments were conducted under atmospheric pressure in a fixed-bed quartz tube (i.d. 6 mm) low micro reactor. Prior to each experiment, the H-zeolite was pressed into a cylinder and then broken into 20–40 meshes particles and activated under flowing N2 (50 ml/min) at 823 K for 2 h and then kept at desired reaction temperature. n-Octane cracking was carried out at 673 K. The reactant stream of n-octane in N2 (molar ratio of 0.018) was fed into the reactor containing 200 mg of catalyst. Cumene cracking was carried out at 423–673 K over 100 mg of catalyst, at a molar ratio of cumene and N2 kept at 0.006. The total gas flow at the reactor inlet was kept constant at 50 ml/min, and the lines were kept at 433 K by heating tapes. Cracking products were analyzed by an online gas chromatograph (Agilent1790) with a flame ionization detector (FID) and GDX-101 column. 3. Results and discussion 3.1. X-ray diffraction (XRD) Fig. 1 shows XRD patterns of as-synthesized samples MMZ, Mordenite and ZSM-5. The characteristic diffraction peaks of Mordenite and ZSM-5 zeolites can be observed to occur simultaneously, indicating the co-existence of Mordenite and ZSM-5 zeolite phases in the zeolite composite.

Mordenite

MMZ

2.2. Characterization The XRD patterns were recorded using a Rigaku Dmax X-ray diffractometer, which employed Ni-filtered CuKa radiation and was operated at 40 kV and 80 mA. Crystal size, morphology and macroporosity of the composite MMZ and the corresponding physical mixture Z + M were investigated on a JSM-6301F scanning electron microscope. N2 adsorption at 77 K was performed in a

ZSM-5 5

10

15

20

25

30

35

2Theta/degree Fig. 1. The XRD pattern of the as-synthesized MMZ, Mordenite and ZSM-5 zeolites.

266

J. Zheng et al. / Microporous and Mesoporous Materials 122 (2009) 264–269

3.2. Scanning electron microscopy (SEM)

3.3. N2 adsorption–desorption isotherms

Fig. 2 displays SEM images of the zeolite composite MMZ and the physical mixture Z + M. The SEM images of MMZ differ from those of Mordenite or ZSM-5 zeolites. The Mordenite and ZSM-5 zeolite crystals show slab-like particles with the sizes of about 3 lm  10 lm and 1 lm  2 lm, respectively (Fig. 2D). The sizes of both Mordenite and ZSM-5 zeolite crystals are smaller than that of the as-synthesized zeolite composite MMZ (about 7 lm  12 lm). The relatively bigger crystals of the zeolite composite MMZ resulted from the fact that zeolite crystals of zeolite ZSM-5 were grown around Mordenite zeolite particles. From Fig. 2A it can be seen that the as-synthesized sample MMZ shows an uneven morphology, indicating an uneven growth rate of ZSM-5 zeolite on Mordenite crystals. ZSM-5 zeolite grows relatively faster on the two terminals ends of Mordenite particles than on middle part because of the presence of more lattice defects on these two terminals ends resulting in the dumbbell-like morphology (Fig. 2B). As shown in Fig. 2C, the crystal has a rough surface. Macropores with the sizes of 50–100 nm can be obviously observed, which can be assigned to the porous structures composed of ZSM-5 crystals (with the sizes of 50–200 nm) grown onto Mordenite zeolite surface during the second-step crystallization.

Fig. 3 reveals N2 adsorption–desorption isotherms and the corresponding BJH pore size distribution curves (inset) of the zeolite composite MMZ and physical mixture Z + M. The adsorption– desorption of nitrogen on the mixture Z + M is a type-I isotherm, indicating the presence of micropores only. However, a curve combining type-I and type-IV isotherms is observed for MMZ. A larger hysteresis loop occurs after p/p0  0.45 in the adsorption–desorption isotherm of the zeolite composite MMZ, which indicates not only the presence of mesopores but a broad pore size distribution also. Moreover, significant increase in N2 adsorption is also observed from p/p0  0.8 for the as-synthesized sample MMZ. According to the SEM results mentioned above, zeolite composite MMZ displays agglomerates of very small crystallites with sizes of 50–200 nm. The sharp condensation in the range of p/p0 = 0.8– 1.0 should therefore be attributed to capillary condensation [14] in open mesopores obtained by filling the interparticle spaces [3,15]. As shown in Fig. 3 (inset), the pore size distribution of the composite MMZ illustrates the existence of a mesopore structure with pore size centered around 3.7 nm. However, the BJH pore size distribution derived from the adsorption branch of the isotherm

Fig. 2. SEM images of the zeolite composite MMZ with different magnification (A–C) and physical mixture of Mordenite and ZSM-5 zeolites (D).

267

J. Zheng et al. / Microporous and Mesoporous Materials 122 (2009) 264–269

A

150 140

Adsorbed amount/ (ml/g)

160

MMZ Z+M

dv/dlogd (ml/g)

N2 adsorbed amount (ml/g)

170

10

Pore diameter(nm)

130 120

16 14 12 10 8

MMZ Z+M

6 0

110 0.0

20

40

60

80

100

120

Time/ min

0.2

0.4 0.6 Relative pressure(p/p0)

0.8

1.0

B

Fig. 3. N2 adsorption–desorption isotherms and the corresponding BJH pore size distribution curves (inset) of the zeolite composite MMZ and physical mixture Z + M.

3.4. Adsorption results of n-octane The volume of adsorbed n-octane per unit weight of zeolite versus time for the as-synthesized zeolite composite MMZ along with that of the physical mixture Z + M is depicted in Fig. 4A. Comparing these results, we can find that the adsorption capacity of the zeolite composite MMZ sample is about 16 ml/g, higher than 14 ml/ g, the adsorption capacity of the physical mixture Z + M. The presence of meso- and macropores in the MMZ sample can contribute to the higher adsorption capacity and to its higher mesoporous volume and lower bulk density. The uptake curves of n-octane on the two samples also show that adsorption equilibrium could be reached more quickly on the zeolite composite MMZ than on the mixture of zeolites. This could be attributed to the lower diffusion resistance resulting from the existence of meso- and macropores in the MMZ zeolite composite. The formation of hierarchical pores increases the rate of diffusion of n-octane in the MMZ zeolite composite. About 90 min is

1.5 ln[ 0.61qm /( qm-qt ) ]

does not show the similar distribution in the mixture Z + M. The pore centered around 3.7 nm can be attributed to the contribution of alkali leaching on Mordenite zeolite. During the second-step synthesis, the framework of Mordenite zeolite was etched by the synthesis solution and silica was partially solved, which caused the enlargement of micropores, resulting in the formation of mesopores [13,16]. The value of the BET surface area of the zeolite composite MMZ (SBET = 358 m2/g) is lower than that of the mixture Z + M (SBET = 395 m2/g). The decrease in the BET surface area is most probably because the decrease in the crystal size of ZSM-5 zeolite particles to the nanocrystalline region [3], causes the deterioration of the ordering of the three-dimensional network of ZSM-5 zeolite particles. The dissolution of Mordenite zeolite may also play a role, which results in the breakdown of part micropores [17]. However, the external surface area of the composite MMZ (SEXT = 69 m2/g) is much larger than that of the physical mixture Z + M (SEXT = 18 m2/ g). These results show that a hierarchical pore system is created in the zeolite composite MMZ, namely, micropore from Mordenite and ZSM-5 zeolites, mesopores centered around 3.7 nm, and macropores with the sizes of 50–100 nm.

2.0

1.0

0.5 MMZ Z+M

0.0 0

500

1000

1500

2000

2500

3000

t/s Fig. 4. The adsorption kinetics curves of n-octane over physical mixture Z + M and zeolite composite MMZ. (A) n-Octane uptake curves of the zeolite composite MMZ and physical mixture Z + M. (B) Uptake profiles in a long time domain. Straight lines correspond to fitting by Eq. (2).

required for the saturation of n-octane on the physical mixture Z + M. Whereas it takes only 50 min on the MMZ zeolite composite even though its uptake amount is larger than that of the physical mixture Z + M. Further quantification of the diffusion properties in the different samples is attained using the classic Fick’s law of diffusion [18], which describes the change of the concentration of molecules inside the zeolite crystals as a function of time.

1  qt =qm  ð6p2 Þ  expðp2 Dt=r 2 Þ

ð1Þ

or

ln½0:61qm =ðqm  qt Þ  p2 Dt=r 2

ð2Þ

where qt and qm are instant adsorption capacity and equilibrium adsorption capacity, respectively, D is diffusivity, r is characteristic diffusion length, and t is time. The characteristic diffusion length r of the crystals in the zeolite composite MMZ and in the physical mixture Z + M is derived from SEM and estimated as 1 lm and 0.5 lm (r = Vp/Sp for slab-like particles, where Vp and Sp are the volume and external surface area of the particle [19], respectively). Plotting ln½0:61qm =ðqm  qt Þ versus time t for the zeolite composite MMZ and the physical mixture Z + M leads to two straight lines, as shown in Fig. 4B. This fitting adequately describes the experimental data covering the uptake curve up to qt/qm = 0.8 and leads to a characteristic diffusion time of r2/D = 7.35  103 s and 1.48  104 s for

J. Zheng et al. / Microporous and Mesoporous Materials 122 (2009) 264–269

the zeolite composite MMZ and the physical mixture Z + M, respectively. Combined with r = 1 lm (the zeolite composite MMZ) and 0.5 lm (the physical mixture Z + M), yields a diffusivity D of approximately 1.36  1016 m2 s1 for the zeolite composite MMZ and 1.70  1017 m2s1 for the physical mixture (Z + M). Showing a lower diffusion resistance for the MMZ zeolite composite. Comparing with the mixture Z + M, the characteristic diffusion path length in the composite is shortened because of the presence of meso- and macropores, which causes the improved gas transport properties [19] and thus faster diffusion of n-octane. 3.5. Accessibility of Brönsted-acid sites To investigate the accessibility of Brönsted-acid sites, 2,6-dimethyl pyridine (DMPy) is selected as a probe molecule because it is widely used as a sterically hindered base. DMPy has already been used in the investigation of the acid sites located on the external surface of zeolites [20] as well as in the characterization of the strength of Brönsted-acid sites [21]. The size of DMPy molecular is about 6.7 Å, which is bigger than the dimensions of pores in ZSM-5 (5.3  5.6 Å) and close to the dimensions of pores in Mordenite (6.5  7.0 Å). Theoretically the DMPy molecule can enter the micropores of Mordenite zeolite, but diffusion inside the pores would be very limited, and therefore not all the acid sites inside the crystal are accessible. The creation of meso- and macropores in the MMZ zeolite composite should facilitate the diffusion of bulky molecule and increase the accessibility of the sites. The aim of the DMPy IR results is to verify whether the hierarchical porosity in the zeolite composite could lead to the improvement of the accessibility of the Brönsted-acid sites. Fig. 5 summarizes IR results of pyridine (Py) and DMPy adsorption at saturation in terms of the integrated absorbance (IA) of the bands at 1545 cm1 (Py), 1630 and 1650 cm1 (DMPy) which gives the Brönsted site (B) accessibility for Py and DMPy in H-MMZ and H-(Z + M). A high IA indicates high accessibility of Brönsted for Py observed on both samples. However, when DMPy is used as probe molecule, higher amount of accessible sites is only found on HMMZ. In comparison with Py as the probe molecule, the accessible Brönsted-acid site of DMPy in H-MMZ zeolite composite is about 63 percent, however, it is only 17 percent in the physical mixture H-(Z + M). The Brönsted-acid site accessibility of Py in the physical mixture H-(Z + M) is slightly higher than that in the zeolite composite H-MMZ. However, when DMPy is used as probe molecule, the acid accessibility in H-MMZ is 3.4 times as much as in H-(Z + M). That

can be ascribed to the formation of meso- and macropores in the zeolite composite H-MMZ. As discussed previously, the presence of meso- and macropores in the zeolite composite crystal facilitates the diffusion of bulky molecules and increases the accessibility of Brönsted-acid sites. 3.6. Catalytic tests The effects of the existence of meso- and macropores on catalytic performances of zeolite composite H-MMZ were investigated by n-octane and cumene catalytic cracking, and compared with those of the physical mixture H-(Z + M), as shown in Figs. 6 and 7, respectively. It can be seen that H-MMZ has a higher activity than H-(Z + M) for n-octane catalytic cracking as well as for cumene cracking under the same reaction conditions. Fig. 6 shows that the initial conversion of n-octane on H-MMZ is about 73 wt% (1 h), higher than the 62 wt% conversion (1 h) on H-(Z + M); In addition, the conversion of cumene on H-MMZ (57 wt%) is about three times as much as on H-(Z + M) (about 20 wt%) at 423 K (Fig. 7). The higher conversion of n-octane or cumene on the zeolite composite H-MMZ could be caused by both the faster diffusion of reactant in its channel and higher accessibility of Brönsted-acid sites resulting from the existence of the meso- and macropores in the composite. The

100

Conversion of n-octane/ wt.%

268

H-MMZ H-(Z+M)

80

60

40

20

0 5

10

15

20

25

Time on stream/ h Fig. 6. The conversions of n-octane on H-MMZ and H-(Z + M) at 673 K.

Relative contents of B-acid

H-MMZ H-(Z+M)

Convesion of cumene/ wt.%

100 90 80 70 60 50 40 30

H-MMZ H-(Z+M)

20 Py

DMPy

Fig. 5. Relative contents of Brönsted sites accessibility for Py and DMPy in H-MMZ and H-(Z + M), determined from the areas of the peaks around the bands at 1545 cm1 (Py), 1630 and 1650 cm1 (DMPy).

450

500

550

600

Reaction temperature/ K Fig. 7. The conversions of cumene on H-(Z + M) and H-MMZ at different reaction temperatures. All the analysis was carried out after 5 min on stream.

J. Zheng et al. / Microporous and Mesoporous Materials 122 (2009) 264–269

hierarchical pore system in the composite makes the reactants easy to access acid sites [10]. The existence of meso- and macropores could also accelerate the elution of cracked products from the composite materials. The introduction of the hierarchical porosity in the zeolite composite has a major impact on catalytic activity. Higher activity of the H-MMZ zeolite composite results from the enhanced accessibility of Brönsted-acid sites, however, the largest beneficial effect is the alleviated diffusion limitation by the presence of mesoand macropores. 4. Conclusions In summary, a zeolite composite MMZ has been successfully synthesized by using Mordenite zeolite as silica–alumina source. The results from SEM observation reveal that the meso- and macropores presented in the composite MMZ have the porous windows of 50–100 nm, which is attributed to the polycrystalline aggregation of zeolite ZSM-5. N2 adsorption–desorption isotherms show that zeolite composite MMZ has a mesopore structure with pore size centered around 3.7 nm, which is associated with the alkali leaching on Mordenite zeolites. The existence of meso- and macropores in the zeolite composite facilitate the molecular transport by shortening its micropores and its acid sites are more accessible, which gives the catalyst a higher activity. Acknowledgment Author R. Li thanks Professor E.E. Wolf (University of Notre Dame) for his valuable advice and edit for the revised manuscript. This work is supported by the ‘‘973” project (No. 2005CB221204) and SinoPEC (No. 107009).

269

References [1] S. van Donk, A. Broersma, O.L.J. Gijzeman, J.A. van Bokhoven, J.H. Bitter, K.P. de Jong, J. Catal. 204 (2001) 272. [2] J.C. Groen, L.A.A. Peffer, J.A. Moulijn, J. Pérez-Ramírez, Micropor. Mesopor. Mat. 69 (2004) 29. [3] Y. Tao, H. Kanoh, K. Kaneko, Langmuir 21 (2005) 504. [4] K. Egeblad, M. Kustova, S.K. Klitgaard, K. Zhu, C.H. Christensen, Micropor. Mesopor. Mat. 101 (2007) 214. [5] B.T. Holland, Micropor. Mesopor. Mat. 89 (2006) 291. [6] M. Tromp, J.A. van Bokhoven, M.T. Garriga Oostenbrink, J.H. Bitter, K.P. de Jong, D.C. Koningsberger, J. Catal. 190 (2000) 209. [7] [a] J.C. Groen, L.A.A. Peffer, J.A. Moulijn, J. Pérez-Ramírez, Colloid Surface A 241 (2004) 53; [b] J.C. Groen, J.A. Moulijn, J. Pérez-Ramírez, Micropor. Mesopor. Mat. 87 (2005) 153; [c] J.C. Groen, A. Brückner, E. Berrier, L. Maldonado, J.A. Moulijn, J. PérezRamírez, J. Catal. 243 (2006) 212. [8] V.V. Ordomsky, V.Y. Murzin, Y.V. Monakhova, Y.V. Zubavichus, E.E. Knyazeva, N.S. Nesterenko, I.I. Ivanova, Micropor. Mesopor. Mat. 105 (2007) 101. [9] R. Mann, Catal. Today 18 (1993) 509. [10] N.S. Nesterenko, F. Thibault-Starzyk, V. Montouillout, V.V. Yuschenko, C. Fernandez, J.-P. Gilson, F. Fajula, I.I. Ivanova, Micropor. Mesopor. Mat. 71 (2004) 157. [11] K.-H. Lee, B.-H. Ha, Micropor. Mesopor. Mat. 23 (1998) 211. [12] M. Ogura, S. Shinomiya, J. Tateno, Y. Nara, E. Kikuchi, M. Matsukata, Chem. Lett. 8 (2000) 882. [13] J.S. Jung, J.W. Park, G. Seo, Appl. Catal. A Gen. 288 (2005) 149. [14] M. Bjørgen, F. Joensen, M.S. Holm, U. Olsbye, K.-P. Lillerud, S. Svelle, Appl. Catal. A Gen. 345 (2008) 43. [15] A. Simon-Masseron, J.P. Marques, J.M. Lopes, F. Ramôa Ribeiro, I. Gener, M. Guisnet, Appl. Catal. A Gen. 316 (2007) 75. [16] J.C. Groen, J.C. Jansen, J.A. Moulijn, J. Perez-Ramirez, J. Phys. Chem. B 108 (2004) 13062. [17] Y. Bouizi, L. Rouleau, V.P. Valtchev, Chem. Mater. 18 (2006) 4959. [18] H.G. Karge, J. Weitkamp, Molecular Sieves Science and Technology – Adsorption and Diffusion, Springer, Berlin, 2008. [19] J.C. Groen, W. Zhu, S. Brouwer, S.J. Huynink, F. Kapteijn, J.A. Moulijn, J. PérezRamírez, J. Am. Chem. Soc. 129 (2007) 355. [20] A. Corma, C. Rodellas, V. Fornes, J. Catal. 88 (1984) 374. [21] L. Oliviero, A. Vimont, J.-C. Lavalley, F.R. Sarria, M. Gaillard, F.O. Mauge, Phys. Chem. Chem. Phys. 7 (2005) 1861.