Ethylbenzene Disproportionation on HZSM-5 Zeolite: The Effect of Aluminum Content and Crystal Size on the Selectivity for p-Diethylbenzene

Abstract

The aim of this work was to verify the effect of MFI aluminum content and crystal size on the selectivity for para-diethylbenzene during ethylbenzene disproportionation. It was observed that the para-diethylbenzene selectivity increased as MFI crystal size increased. The increase in aluminum content caused a decrease in the selectivity for para-diethylbenzene. However, for crystals larger than 8 <FONT FACE="Symbol">m</FONT> m, the decrease in aluminum content had little influence on the selectivity for para-diethylbenzene. The results can be explained by the number of active aluminum sites on the external surface of the crystals.

Selectivity; crystal size; aluminum content

Ethylbenzene Disproportionation on HZSM-5 Zeolite: The Effect of Aluminum Content and Crystal Size on the Selectivity for p-Diethylbenzene

N.D. Velasco, M.S. Machado and D. Cardoso

Federal University of São Carlos, Chemical Engineering Department, P.O. Box 676,

13565-905 - São Carlos - SP, Brazil - [e-mail], [e-mail]

(Received: November 5, 1997; Accepted: March 24, 1998)

Abstract - The aim of this work was to verify the effect of MFI aluminum content and crystal size on the selectivity for para-diethylbenzene during ethylbenzene disproportionation. It was observed that the para-diethylbenzene selectivity increased as MFI crystal size increased. The increase in aluminum content caused a decrease in the selectivity for para-diethylbenzene. However, for crystals larger than 8 m m, the decrease in aluminum content had little influence on the selectivity for para-diethylbenzene. The results can be explained by the number of active aluminum sites on the external surface of the crystals.

Keywords: Selectivity, crystal size, aluminum content.

INTRODUCTION

The well-known ZSM-5 zeolite, whose structure was labeled MFI by the IZA committee, has been largely used as a catalyst for organic compound transformations, especially in petrochemistry. ZSM-5 differs from most other molecular sieves due to its micropore diameter (5.5 Å), which is approximately the size of an aromatic ring. This structural characteristic of the 10-ring ZSM-5 is responsible for its shape-selectivity properties, which have been utilized in the following processes: dewaxing of petroleum fractions, xylene isomerization, methanol-to-gasoline conversion and ethylbenzene production (Venuto, 1994; Hairston, 1996).

There are several methods for physicochemical characterization of a solid catalyst, such as by the spectroscopic analysis of the interaction of probe molecules with catalyst sites. By these methods one can determine the number of adsorption sites on a catalyst surface, but it must be born in mind that there is a distinction between adsorption sites and catalytically active sites. To determine catalyst activity, a test reaction can be performed under operational conditions similar to those the industrial process. Characterization of a catalyst by test reactions has also provided a wealth of information regarding its activity and product selectivity (Venuto, 1994).

The disproportionation of ethylbenzene is an interesting model reaction, first studied by Karge et al. to characterize zeolites in their acid form (Karge et al., 1982; Karge et al., 1983; Weitkamp et al., 1986). Using this test reaction, the authors were able to obtain information about the number of active acid sites in zeolites (Karge et al., 1982; Karge et al., 1983) and to discriminate between large (12-member ring) and medium-size (10-member ring) microporous zeolites (Weitkamp et al., 1986). During the disproportionation of ethylbenzene on large pore zeolites, the authors observed: (i) an induction period, when the conversion increased with time on stream until a maximum conversion was reached; (ii) during the induction period, the yield of benzene was significantly higher than that of the diethylbenzenes, although, under these reaction conditions since no dealkylation occurred they should be equal on the basis of reaction stoichiometry and (iii) during the induction period, the distribution of diethylbenzene isomers changed considerably and no shape selectivity was observed. During the reaction with medium micropore zeolites, the authors observed: (i) no induction period; (ii) a net deactivation of the catalyst and (iii) the shape-selectivity effect of the diethylbenzene isomers.

Many research groups (Kaeding et al., 1981; Young et al., 1982; Bezouhanova, 1986; Paparatto et al., 1988) studied the shape-selectivity properties of zeolites with different structures. The para-selectivity observed was quantitatively related to some key catalyst properties, i.e., activity (aluminum content in the structure), crystal size and diffusional limitation within zeolite channels. Richter (1989) studied the effect of MFI aluminum content on selectivity for p-xylene during m-xylene isomerization. He observed that para-selectivity depends only on m-xylene conversion. This result agreed with that obtained by Paparatto et al. (1987) but disagreed with the results of Bezouhanova (1986). Bezouhanova reported that aluminum content is a parameter that influences the para-selectivity of the MFI zeolite catalyst. Several ways to modify MFI catalysts to induce para-selectivity have been described (Young et al., 1982; Kaeding et al., 1981). Young et al. (1982) studied the transformation of some alkylbenzenes on MFI catalysts. They reported that para-selectivity increased as the aluminum content of the catalyst decreased. Paparatto et al. (1988) investigated the role of external surface sites on MFI catalysts in reaction selectivity. According to them, samples with lower external surface areas were more para-selective due to the lower extent of isomerization reactions. A critical point of view in the literature shows that these conclusions were based on experiments conducted on no more than three catalyst samples. Moreover, composition and crystal size of the zeolites were not the same and these two parameters influence zeolite selectivity. Indeed, some authors compared the catalyst selectivities obtained at different conversions, instead of at the same conversion of the reactant.

The present work addresses ethylbenzene disproportionation on the MFI zeolite in a differential reactor. It encompasses the effects of aluminum content and crystal size of the MFI zeolite on para-diethylbenzene.

EXPERIMENTAL

Samples of MFI were produced from sodium trisilicate, aluminum sulfate and n-butylamine as template, using an experimental procedure described previously (Machado et al., 1994). Three samples with Si/Al ratios of 25, 42 and 75 were synthesized, and for each series, four samples with crystal sizes between 5 and 32 m m where obtained. Variation in the aging of the precursor gel resulted in crystallization of samples of different crystal sizes. It was also possible to obtain samples with the same crystal size but with different aluminum contents and to obtain samples with the same aluminum content but with different crystal sizes.

The MFI properties of the synthesized samples are shown in Table 1. The samples were characterized by scanning electron microscopy for crystallite size and morphology and X-ray powder diffraction (XRD) for zeolite phase purity. The crystal size shown in Table 1 is the average of the highest hexagonal dimension for at least thirty crystals.

The template occluded in the pores of the zeolite crystals was removed by calcination of the sample at 775 K for 12 hours. The acid form of the MFI zeolite was obtained by repeated ion exchange with a 1 M aqueous solution of ammonia nitrate and a further calcination at 675 K for 10 hours.

Prior to use, ethylbenzene (>99 wt%, Fluka) was percolated through a column containing a -Al₂O₃ to eliminate oxygenated impurities, which have an inhibiting effect on aromatic transformations (Weitkamp, 1986). The catalytic test reactions were carried out in a continuous flow differential microreactor at 575 K and atmospheric pressure. Inert particles were mixed with the catalyst to minimize interparticle gradients throughout the differential reactor (Rose, 1977). Thus, a mixture of 200 mg of the MFI zeolite and 300 mg of ground glass (between 250 - 177 µm) was placed into the glass reactor and activated in situ at 675 K over dry nitrogen flow (4 l.h^-1.g^-1) for 4 hours. The reactant was supplied by bubbling a carrier gas (nitrogen) through a saturator containing ethylbenzene maintained at a constant temperature. The contact time of ethylbenzene W/F was varied between 3.8 and 9.1 g.h/mol (W is mass of the catalyst and F is molar flow of ethylbenzene). Analysis of the reactor effluent was achieved by on-line sampling with a Varian 3400 gas chromatograph.

Table 1:
Properties of the synthesized MFI samples

Gel Aging Time (Days)	Crystal Size (L) and Zeolite Crystallinity (C)
	Si/Al = 25		Si/Al = 42		Si/Al = 75
	L (m m)	C (%)	L (m m)	C (%)	L (m m)	C (%)
0	11.4	89	32.3	92	-	-
1	9.6	95	14.6	96	15.5	93
2	7.4	98	8.4	98	8.6	95
5	3.9	100	4.4	100	4.2	95

C: MFI phase purity determined by XRD

RESULTS AND DISCUSSION

In Figure 1 ethylbenzene conversion with time on stream on an MFI catalyst is shown. Rapid catalyst deactivation occurs during the first hours of reaction, after which the conversion decreases slowly. This behavior was also observed by Weitkamp (1986) and is characteristic of medium-size micropore zeolites such as the 10-member ring MFI. Special care was taken to evaluate the catalysts at the same degree of conversion in order to achieve a reliable comparison of zeolite selectivity. Conversion was measured after the rapid deactivation period of the catalyst and after 3 hours on stream.

During all the experiments, ortho-diethylbenzene (o-DEB) yield was lower than 2 wt% and was detected only during the first few minutes of the reaction. Only benzene (Bz), para-diethylbenzene (p-DEB) and meta-diethylbenzene (m-DEB) were detected in the product stream. Of the diethylbenzene isomers, o-DEB is the least favored by thermodynamic equilibrium and is only 17.5% at 575 K. In addition, formation of o-DEB in the MFI channels is not favored due to the steric constraints of the two ethyl groups.

The effects of ethylbenzene conversion on the yield of benzene and diethylbenzenes are shown in Figure 2. At the beginning of the catalytic runs, diethylbenzenes yield was lower than benzene yield. Since no triethylbenzenes were found in the product stream and no dealkylation was observed under these reactions conditions, Weitkamp (1986) suggests that the deficiency of diethylbenzenes in the product can be attributed to an accumulation of this product inside the catalyst channels.

Figure 3 shows the effect of MFI crystal size and aluminum content on the conversion of ethylbenzene. The results show that the conversion of ethylbenzene decreases as crystal size increases and that this behavior is more pronounced in samples with higher aluminum contents, i. e., for the samples with Si/Al ratio of 25. Kaeding (1981) observed similar results using the Y zeolite. The higher conversion in samples with small crystals can be explained by the fact that their internal sites are more easily accessible to the reactant. For the catalysts with a larger crystal size, diffusion limitations of reactants and products in the internal channels result in a lower conversion of ethylbenzene.

Figure 4 presents the effects of crystal size on para-selectivity, and for comparison, the calculated value at thermodynamic equilibrium at 575 K (Stull et al., 1969) is also shown. It can be seen that selectivity for p-DEB increases as crystal size increases. For samples with the same crystal size, these data also show that selectivity for p-DEB decreases as the aluminum content in the crystals increases; this effect is very noticeable in samples with higher aluminum contents. Extrapolation of the curve for samples with a Si/Al ratio of 25 shows that to achieve greater than 90% para-selectivity, it would be necessary to use a sample whose crystal size was 40 m m or more.

The selectivity for diethylbenzene isomer formation on an MFI catalyst with a Si/Al ratio of 75 and crystal size of 4.2 m m is presented in Figure 5. The results corroborate that p-DEB is the primary product of ethylbenzene disproportionation on the MFI catalyst, since selectivity for p-DEB approximates 100% as conversion tends towards zero.

Under the steric constraints of the MFI channels, para-selectivity can be described in terms of the following catalyst properties:

i) Inside zeolite channels: by the rate of isomerization (k_I) of the primary product, by the steric constraints of the channels and by the length of the diffusion path (crystal size);

ii) On the external surface: by the number of external sites and also by the isomerization (k_I) rate.

Independent of the Si/Al ratio, the observed increase selectivity for p-DEB for increasing crystal size is due to the increasing diffusion path length or the decreasing number of active sites on the external surface of the crystals. The main question remains, which of these two factors is more important for achieving high selectivity for p-DEB?

Das et al. (1993) reported that the selective poisoning of the external surface of zeolite crystallites by the chemical vapor deposition (CVD) of silica, employing tetraethyl orthosilicate Si(OC₂H₅)₄, improved the para-selectivity property of the MFI zeolite. The authors suggested that the improvement in para-selectivity was due to the suppression of isomerization of the para-isomer. So, increasing the number of external active sites increases the isomerization ratio, decreasing selectivity for p-DEB, and the final product composition approximates thermodynamic equilibrium.

To explain the effect of crystal size on selectivity for p-DEB, some results are presented in Figure 6. With crystals of the same size, these results show that by increasing the Si/Al ratio, and thus lowering overall aluminum content, there was higher para-selectivity during ethylbenzene disproportionation. This can be explained by the decreasing isomerization rate k_I of p-DEB due to a lower aluminum content. It’s interesting to observe the small increase in selectivity for p-DEB for samples with both Si/Al > 40 and crystal dimensions higher than 8m m. In this case, an increase in selectivity for p-DEB will be only possible with thre complete elimination of external sites of the MFI crystals that cause isomerization of the primarily produced para-isomer.

Figure 1:
Conversion of ethylbenzene with time on stream.

Figure 2:
Effect of conversion on benzene yield and diethylbenzenes/ benzene ratio.

Figure 3:
Effect of MFI crystal size on conversion of ethylbenzene.

Figure 4: Effect of crystal size on selectivity for p-DEB at 3% conversion of ethylbenzene.

Figure 5:
Diethylbenzenes formation on an MFI sample.

Figure 6: Effect of both MFI crystal size and Si/Al ratio on selectivity for p-DEB.

CONCLUSIONS

Very high product selectivity on the MFI catalyst can be achieved during ethylbenzene disproportionation by controlling both crystal size and aluminum content. The results corroborate that para-diethylbenzene is the primary product of ethylbenzene disproportionation on the MFI zeolite. On MFI samples with high aluminum contents, the results showed that to achieve para-selectivity greater than 90% it would be necessary to use samples whose crystal sizes were 40 m m or larger. The best results were obtained using an MFI crystal size of 14 m m and a Si/Al ratio greater than 40.

ACKNOWLEDGMENTS

The authors wish to acknowledge the financial support received from Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). Nelson D. Velasco would like to thank the Chemical Engineering Department of Universidade Federal de São Carlos for the research position.

REFERENCES

Bezouhanova, C., Dependence of p-xylene selectivity of pentasils on the Si/Al ratio, Appl. Catal., 21, 149 (1986).

Das, J.; Blat, Y.S. and Halgari, A.B., Ethylbenzene dealkylation and realkylation over pore size regulated MFI zeolite, Ind. Eng. Chem. Res., 32, 2525 (1993).

Hairston, D.W., Zealous Zeolites, Chemical Engineering, 103, no. 7, 57 (1996).

Kaeding, W.W.; Chu, C.; Young, L.B.; Winstein, B. and Butter, S.A., Selective alkylation of toluene with methanol to produce para-Xylene, J. Catal., 67, 159 (1981).

Karge, H.G.; Ladebeck, J.; Sarbak, Z. and Hatada, K., Conversion of alkylbenzenes over zeolite catalysts I. Dealkylation and disproportionation of ethylbenzene over mordenites, Zeolites, 2, 94 (1982).

Karge, H.G.; Hatada, K. and Zheng, Y. and Fiedorow, R., Conversion of alkylbenzenes over zeolite catalysts II. Disproportionation of ethylbenzene over Faujasite-type zeolites, Zeolites, 3, 13 (1983).

Machado, M.S.; Cardoso, D. and Silva, M.A.P., Anais do 11º Congresso Brasileiro de Engenharia e Ciência dos Materiais, Águas de São Pedro, 949 (1994).

Paparatto, G.; Moreti, E.; Leofanti, G. and Gatti, F., Toluene ethylation on ZSM zeolites, J. Catal., 105, 227 (1987).

Paparatto, G.; Alberti, G. de; Leofanti, G. and Padovan, M., the role of the external surface in reactions with zeolite catalysts, Stud. Surf. Sci. Catal., 44, 255 (1988).

Richter, M., The effect of MFI aluminum content on selectivity for p-xylene, Zeolites, 9, 238 (1989).

Rose, H.F., Chemical Reactor Design for Process Plants, John Wiley & Sons Inc., New York, (1977).

Stull, D.R.; Westrum, E.F. and Sinke Jr., G.C., The Chemical Thermodynamics of Organic Compounds, John Wiley & Sons Inc., New York, (1969).

Venuto, P.B., Organic catalysis over zeolites: a perspective on reactions with zeolite catalysts, Microporous Materials, 2, 297 (1994).

Weitkamp, J.; Ernst, S.; Jacobs; P.A. and Karge, H.G., ZSM-type and related zeolites as catalysts in the disproportionation of ethylbenzene, Erdöl, Kohle-Erdgass-Petrochem, 39, 13 (1986).

Young, L.B.; Butter, S.A. and Kaeding, W.W., Shape selective reactions with zeolite catalysts III. Selectivity in xylene isomerization, toluene-methanol alkylation and toluene disproportionation over ZSM-5 zeolite catalysts. J. Catal., 76, 418 (1982).

Publication Dates

• Publication in this collection
  09 Oct 1998
• Date of issue
  June 1998

History

• Accepted
  24 Mar 1998
• Received
  05 Nov 1997