Effect of the addition of acidic and basic compounds on the catalytic behavior of exchanged zeolites in the alkylation of toluene with methanol

Effect of the addition of acidic and basic compounds on the catalytic behavior of exchanged zeolites in the alkylation of toluene with methanol

Studies in Surface Science and Catalysis 130 A. Corma, F.V. Melo, S. Mendioroz and J.L.G. Fierro (Editors) C 2000 Elsevier Science B.V. All fights res...

403KB Sizes 1 Downloads 5 Views

Studies in Surface Science and Catalysis 130 A. Corma, F.V. Melo, S. Mendioroz and J.L.G. Fierro (Editors) C 2000 Elsevier Science B.V. All fights reserved.

2621

Effect of the addition of acidic and basic compounds on the catalytic behavior of exchanged zeolites in the alkylation of toluene with methanol A. Borgna*, J. SepOlveda, S. Magni and C. Apesteguia. INCAPE, FIQ-UNL-CONICET, Sgo. del Estero 2654, (3000)-Santa Fe, Argentina FAX: +(54)-342-4531068. e-mail: [email protected] The alkylation of toluene with methanol over alkali and alkaline-earth exchanged Y zeolites was studied. While Mg(Ca)-exchanged zeolites produced the selective alkylation of the benzene ring, alkali exchanged zeolites were highly selective for side-chain alkylation. "in situ" poisoning experiments were performed by doping the reactant mixture with an acid or a base. Alkalineearth exchanged zeolites were only deactivated by adding base compounds. By the contrast, alkali exchanged zeolites were strongly deactivated by the addition of either acid or base compounds, demonstrating that a cooperative action of acid/base pairs is required for side chain alkylation. 1. INTRODUCTION The alkylation of toluene with methanol is readily catalyzed on synthetic zeolites, but the product selectivity strongly depend on the catalyst surface acid-base properties. Acid zeolites promote the aromatic-ring alkylation [1 ] while the side-chain alkylation occurs preferentially over basic zeolites [2,3]. The side-chain alkylation of toluene with methanol for producing a mixture of styrene and ethylbenzene offers economical advantages compared with the conventional homogeneously catalyzed Friedd-Cratts process, which use ethylene and benzene as reactants [4]. Due to its potential as a novel route for obtaining styrene, this alkylation reaction has been widely studied [5], but both the reaction mechanism and the active sites requirements for enhancing the selectivity toward styrene are not completely understood yet. Zeolites are attractive materials for using in alkylation reactions because their acid-basic properties can be modified by dealumination, isomorphous substitution or ionic-exchange. The role of the acid-base properties of zeolite catalysts on the product distribution of aromatic alkylation reactions has been previously studied [6]. Particularly, the side-chain alkylation of toluene with methanol requires basic sites for activating the carbon atom in the methyl group of toluene [7] and also for dehydrogenating methanol to formaldehyde, which is the alkylating agent to form styrene [8]. Acid sites catalyze preferentially ring alkylation reactions, but surface acidity is also necessary for efficiently promoting side-chain alkylations, via combined acid-base pathways. However, the exact requirements of acid site density and strength in the side-chain alkylation reaction mechanism is still debated. In this work we have studied the effect that cofeeding acid and basic compounds has on the catalyst activity and selectivity for the alkylation of toluene with methanol over ionexchanged Y zeolites. The goal was to establish the catalyst surface acid-base requirements for efficiently promoting the reaction and, as a consequence, to obtain more insight on the reaction mechanism pathways.

2622 2. EXPERIMENTAL The catalysts were prepared by exchanging a commercial NaY zeolite (Si/AI: 2.5) with alkaline and alkaline earth metals. Ion exchanges were carried out at 80 ~ in four consecutive steps. Following each individual exchange step, the zeolites were filtered, washed with hot water, dried at 120 ~ and finally calcined in flowing air at 450 ~ The exchange degree was determined by chemical analysis. The BET surface areas (S o were measured by N2 adsorption at -196 ~ The exchanged zeolites were characterized by X-ray diffraction. The acid-basic properties were characterized by infi,ared spectroscopy of adsorbed CO2. The gas phase alkylation of toluene with methanol was carried out in a fixed-bed tubular reactor at 1 atm. A mixture of toluene/methanol of 1:1 molar ratio was vaporized in a preheating section and delivered to the reactor. The reaction was carried out in a temperature range between 400 y 500 ~ employing a space velocity (WHSV) of 2 hl. Toluene conversion ( X ~ ) was calculated as: Xtol(%) = [YTolf (YYj + YTol.)]. 100, where EYj is the molar fractions of the aromatic reaction products, including benzene, and YT~ is the molar fi-action of toluene in the products. The selectivity to product j was determined as: Sj (%) - [Yj/'ZYj].100. The SEt-Bzselectivity includes the sum of ethylbenzene and styrene, which are the side-chain alkylation products. In-situ poisoning experiments were carried out by doping the toluene/methanol mixture with either acetic acid or basic compounds (butyl-amine, 3,5-dimethyl pyridine and pyridine) in a concentration range between 0-15000 ppm. 3. RESULTS AND DISCUSSION 3.1 Catalyst characterization

Table 1 summarizes the main characteristics of ion-exchanged Y zeolites. The exchange degree, as determined by chemical analysis, was always higher than 65 %. The specific surface areas of exchanged zeolites were lower than that of the parent NaY zeolite. Other authors have also reported that addition of alkaline metals by using the ion-exchange method may modify the physical properties of the zeolite [9,10]. Table 1 shows that the Sg values decrease with the exchanged-cation size. This suggests that the exchange with large cations reduce the volume inside the pores of the zeolites thereby decreasing both the BET surface area and the pore volume. Table 1 Main characteristics of ion-exchanged Y zeolites and catalytic results for alkylation of toluene with methanol. Zeolite ED c~) Sg Xtol. Selectivities (%) (%) (m2/g) (%) SEt-Bz Sx, SC9+ SBz NaY -950 0.9 6.0 88.5 5.5 -KY 92 840 2.0 64.6 31.4 3.2 0.8 CsY 68 430 7.5 88.0 12.0 . . . . MgY 81 540 52.2 -54.8 43.9 1.3 CaY 88 400 26.6 -44.9 53.7 1.4 HY 89 660 58.8 -58.5 18.8 22.7 (b) Exchange Degree, [(% Nam~ - % Nard) / % Name]* 100

2623

NaY

MgY . ,,,,q

CsY

10

20

30 40 20 (deg) Figure 1. X-ray diffraction patterns.

50

Figure 1 shows the X-ray dif~action patterns of NaY, CsY and MgY zeolites. Significant changes in the relative ~ o n line intensities are observed by comparing the XRD patterns of NaY with those of ionexchanged zeolites. Moreover, a strong decrease of the diffraction peak intensities was observed on Cs-exchanged zeolites. These results are attributed to changes in the structure factors and in X-ray absorption coefficients of the parent NaY zeolite caused by the incorporation of alkaline metal cations bigger than Na+ [11]. Nevertheless, the XRD structure of the NaY zeolite was preserved. FTIR spectroscopy of adsorbed C02 showed that the Cs ion-exchanged zeolite contains the highest basic site density, while Kexchanged zeolites displays an intermediate basicity, between Na and Cs-exchanged zeolites.

3.2 Catalytic tests with pure reactants. The catalytic results for the re.action of toluene with methanol over modified Y zeolites are shown in Table 1. The HY zeolite, which is employed here as a reference sample, was active and selective for benzene-ring alkylation. The xylenes isomers were the predominant products and the isomer distribution was close to the thermodynamic equilibrium distribution: p-Xy: 22.5 %, m-Xy: 50.0 % and o-Xy: 27.5 %. Formation of benzene and higher alkylated compounds (C9+), such as tri and tetramethylbenzenes, was also observed. Mg(Ca)Y zeolites also promoted selectively the benzene-ring alkylation. Both samples were highly active for producing C9+ compounds and, as a consequence, exhibited lower benzene selectivities (SBz) as compared with the HY zeolite. The xylene distribution on CaY sample (p-Xy: 31.6 %, m-Xy: 33.0 % and o-Xy: 35.4 %) was far from the thermodynamic equilibrium value. All these results show that alkaline earth ion-exchanged zeolites are highly selective for the acid-catalyzed benzene ring alkylation. Because the concentration of Br6mted acid sites increases as the cation size decreases [12], it is expected that the CaY zeolite would be less acidic than the MgY sample. This assumption is consistent with our results which show that the formation of m-Xy via the acid-catalyzed isomerization of o- or pxylenes is lower on CaY as compared to MgY. The catalytic behavior of alkali-exchanged zeolites is clearly different as compared to Mg(Ca)-exchanged zeolites. Alkali-exchanged samples present poor activities for toluene/methanol conversion reactions but are highly selective for producing ethylbenzene and styrene. For example, the selectivity to (ethylbenzene + styrene) was about 88% on CsY zeolite (Table 1). The S~-Bz values increased in the order NaY < KY < MgY, which is consistent with the basieity trend measured by IR spectroscopy of adsorbed CO2. Regarding the xylene distribution, Cs(K)Y zeolites

2624 favored the formation of o-xylenes; on both samples the relative concentration of o-Xy in the xylene mixture was higher than 50%. 3.3 Catalytic tests with doped reactants

In situ poisoning experiments were carried out on MgY and CsY zeolites by doping the toluene/methanol mixture either with acetic acid or with a basic compound (pyridine, 3,5dimethyl pyridine and butyl-amine). Fresh catalysts were initially tested for about 3 h using a pure toluene/methanol mixture before introducing the doped feed. Figure 2 shows the activity and selectivity decay curves obtained as a function of time on MgY zeolite when doping the feed with acetic acid. The activity is defined as a = r(t)/ro, where ro a n d r(t) are the reaction rates at zero time and time t, respectively. Figure 2 shows that the MgY activity diminishes with time on stream when using undoped reactants, probably because of the formation of carbonaceous deposits. The coinjection of 5000 ppm of acetic acid did not change the slope of the deactivation curve obtained with pure reactants, thereby revealing that the presence of acetic acid does not inhibit the toluene conversion reactions. Similarly, the zeolite selectivity was not affected by cofeeding acetic acid. 1.0 - * ~

100

Acetic Acid *

~

ppm

-

0.8 . Sxy 9

9 ia,

9 9

9

MgY - T = 400 ~

1.0 ~

80

p

,,

9

.~ 0.6

m

0.8 6o

9~ ~ _ 0 0

o.6

ppm

8700 ppm/

0

Sc~f

< 0.4 i

----

9

in -

40 ~" "~ 0.4

9

20 SEt.Bz m.-~r---.m:•177177

0.0

0

100

14700 ppm

m

0.2

200 300 400 Time (min)

0.2 3,5 dimethyl pyridine

SBz

500

Figure 2. Evolution of the alkylation activity and selectivity as a function of time. Effect of an acid compound. MgY zeolite.

"~

0.0

,

0

I

100

,

I

,

I

200 300 Time (min)

,

400

Figure 3. Evolution of the alkylation activity as a function of time and poison concentration. Effect of a basic compound. MgY zeolite.

In contrast, the addition of a basic compound to the reactants caused the rapid deactivation of the MgY zeolite. Figure 3 illustrates the time on stream behavior of the MgY zeolite during the toluene alkylation with methanol feed both with and without 3,5 dimethyl pyridine. It is observed that the activity decay increased with the concentration of 3,5 dimcthyl pyridine in the feed. Qualitatively, similar activity decay curves were obtained when the reactant mixture was doped with butyl-amine or pyridine. The zeolite deactivation increased with the pKa value of the basic dopant, i.e. in the order butyl-amine > 3,5 dimethyl pyridine > pyridine. The selectivity pattern exhibited by the MgY zeolite was not significantly changed by

2625 poisoning in any case. This suggests that the same type of active sites, probably strong and weak Br0nsted acid sites [13], are involved in the formation of xylenes and C9§ alkylate compounds. However, the eoinjeetion of butyl-amine changed the xylene distribution on MgY zeolite by decreasing the relative formation of m-xylene (Figure 4). Alkylation of the benzene ring occurs mainly in otto and para positions, and m-xylene is produced via consecutive isomerization pathways [14]. The injection of basic compounds decreases the surface acidity and, as a consequence, also decreases the MgY zeolite activity for the isomerization reactions leading to m-xylene. This result also suggests that isomerization to m-xylene takes place on strong Br0nsted acid sites, which would be preferentially poisoned in the presence of gaseous basic compounds. Figure 5 shows the time evolution of the activity for toluene alkylation with methanol on CsY zeolite during in situ poisoning experiments with different 3,5 dimethyl pyridine concentrations. The catalyst activity was practically constant before adding the basic compound, thereby suggesting that coke formation was not significant on CsY zeolite in the reaction conditions used in this work. Alter addition of 3,5 dimethyl pyridine, a strong deactivation occurs, as it was also observed when butyl-amine or pyridine were added to the reactants. The addition of acetic acid also produced a strong deactivation of the CsY zeolite. In spite of the strong deactivation caused by the presence of either basic or acidic compounds, no significant changes in the selectivity pattern were observed. However, the m-xylene yield increased at expenses of o-xylene when acetic acid was added to the reactant mixture. As discussed above, m-xylene is produced from o-xylene via isomerization pathways, probably involving strong BrOnsted acid sites. The presence of gaseous acetic acid would increase the surface acidity of the CsY zeolite thereby increasing the isomerization activity for m-xylene formation. 1 O0

Z $0

o

60

o

,o 40

_

1.0

'

. ~ ~V~

V v V

V

~

~/,

0

100

,

I

V

CsY - T-- 400 ~

o 0.6 <

,

200

I

300

V

0 ppm

ppm

0.4 -

I

v

0.8

o-Xy

' 9~

,

V

~~

20 0

_ v

V

ii

m-Xy

_J

3,5 dimethyl pyridine

',41

MgY- T = 400 ~ butyl-amine 13000 ppm

0.2

,

400

Time (min) Figure 4. Evolution of the distribution ofxylenes as a function of time. Effect of a basic compound. MgY zeolite.

o\

I

100

.

I

200

,

I

300

,

l

400

i

500

Time (min) Figure 5. Evolution of the alkylation activity as a function of time and poison concentration. Effect of a basic compound. CsY zeolite.

2626 It has been suggested that the key step in the side chain alkylation of toluene with methanol is the surface reaction between adsorbed toluene and formaldehyde; the presence of base sites is required to activate the methyl group in the toluene molecule [7]. However, the presence of acid sites is also required for stabilizing the adsorbed toluene molecule. A combination of Lewis acid/base pairs would be therefore required for selectively promoting the side-chain alkylation reaction. Recently, Palomares et al. [15] reported that toluene on basic zeolites interacts with both the alkaline cation, via the electrons of the aromatic ring, and the lattice oxygen, via the hydrogen of the methyl group. It has also been suggested that the surface reaction of the adsorbed species requires a specitic configuration of add/base pairs [9]. This specific configuration allows the formation of an intermediate complex between methanol which reacts to produce formaldehyde with a strongly polarized C atom and the methyl group of adsorbed toluene, which strongly interacts with the basic framework oxygen atoms. The activated C atom of methanol may react with the activated C atom of the methyl group of toluene, producing styrene and ethyl benzene via an aldol-type condensation mechanism [15]. Beside, as has already been pointed out, toluene adsorption requires to be stabilized on a Lewis acid site. Therefore, the proximity of the acid/base sites is crucial to produce side-chain alkylation and the active site is actually an assembly of acid/base sites, which would be highly sensitive to the presence of either acid and base compounds. The assumption that combined Lewis acid/base pairs with a specific configuration are the active sites for the side-chain alkylation of toluene with methanol is supported by our catalytic results. In fact, the addition of either acidic or basic compounds strongly deactivated the side chain alkylation on Cs ion-exchanged zeolite, thereby indicating that both acid and base sites are involved in the rate limiting steps of the reaction mechanism. REFERENCES

1. Kaeding, W.W., Chu, C., Young, L.B., Weinstein B. and Butter, S.A., J.Catal., 67 (1981) 159. Yashima, T., Sato, K. Hayasaka, T. and Hara, N., J. Catal, 26 (1972) 303. 3. Engelhardt, J. Szanyi, J. and Valyon, J., J. Catal, 107 (1987) 296. 4 Brownstein, A.M., in "Catalysis of Organic Reactions (W.R. Moser, Ed.), M. Dekker, Vol. 5, 1981, p. 3. Palomaes, A.E., Eder-Mirth, G. Rep, M. and Lercher, J.A., J. Catal., 180 (1998) 56. Giordano, N., Pino, L., Cavallaro, S., Vitarelli, P. and Rao, B.S., Zeolites, 7 (1987) 131. Itoh, H., Miyamoto, A. and Murakami, Y., J. Catal., 64 (1980) 284. Hathaway, P.E. and Davis, M.E., J. Cayal., 119 (1989) 497. Wieland, W. Davis, R. J. and Garces J. M., J. Catal., 173 (1998) 490. 10. Bentes Jr., A.P., Veloso, C.O. and Monteiro, J.L., Proc. 15th Iberoam. Symp. Catal., C6rdoba, Argentina, Vol. 1, 1996, p. 275. 11. Weitkamp, J., Ernst, S., Hunger, M., R6ser, T., Huber, S., Schubert, U.A., Thomasson, P. and KnOzinger, H., Stud. Surf. Sci. and Catal. 101 (1996) 731. 12. Ward, J.H., J. Catal., 10 (1968) 34. 13. Vinek, H., Derewinski, M., Mirth, G. and Lercher, J.A., Appl. Catal., 68 (1991) 277. 14 Yashima, T., Ahmad, H., Yamakazi, K., Katsuta, M. and Hara, N., J. Catal., 16 (1970) 273. 15 Palomaes, A.E., Eder-Mirth, G. And Lercher, J.A., J. Catal., 168 (1997) 442. .