Benzene as a by-product of toluene with methanol transformation on the basic zeolite catalysts and its presumable origin

Introduction

Reaction of toluene/methanol mixture on basic zeolite catalysts is very complex (Sivasankar and Vasudevan, 2010). It yields styrene and ethylbenzene as the products of side-chain alkylation of toluene with methanol whereas ring alkylation produces isomers of xylene. Methanol decomposes deeply toward carbon monoxide in consequence of the major side reaction. But very few authors report on benzene in the products of the toluene and methanol transformation on these catalysts without forming xylenes (Borgna et al., 2004, 2005; Itoh et al., 1980; Sooknoi and Dwyer, 1995; Yashima et al., 1972), and the origin of this product is not discussed.

Simultaneous formation of benzene and xylenes during toluene conversion is a commonplace (Borgna et al., 2004). This indicates the acidic nature of active sites in charge of the reaction. Acid-activated formation of benzene occurs in number of carbonium ion reactions with the participation of aromatics. Toluene disproportionation is one of them, producing also xylenes. But it might be excluded in the case of above mentioned benzene appearing because of absence of xylenes. Dealkylation of aromatics, giving benzene as a secondary product of decomposition of ethylbenzene or styrene, for instance, might be excluded too because it requires high temperature at least—923 K (Tsai et al., 1999) or 818 K in presence of CO₂ (Zhao et al., 2015). In the work (Itoh et al., 1980), authors recorded the appearing and intensification of benzene formation in toluene with methanol conversion on KX zeolite with addition of O₂ into feed and increasing O₂/methanol ratio. Then benzene may arise from toluene oxidation products, and be a secondary product too.

The main goal of this research was to clarify the nature of benzene forming in the base catalyzed side-chain alkylation of toluene with methanol. To verify the mechanisms leading to the formation of one or another reaction product, due to the high activity of the intermediates, methods of products’ identification at the very beginning of transformations must be used. One of such methods is catalytic transformation in vacuum, which enables to reveal primary catalytic processes due to the negligible contact time of the reactants and products with the catalyst’s active site. In the case of side-chain alkylation, it may be interesting to compare the results of alkylation in vacuum and under the atmospheric pressure when all possible transformations have time to come to pass. Another way to identify the primary reaction products is detection of compounds formed at the lowest possible transformation degree, which may be achieved also by reducing the concentration of catalyst’s active sites to the amount when such concentration becomes a limiting factor for reaction passage. Then, the number of transformation acts per time unit including secondary ones will decrease even in the case of long enough residence time of reactants on the catalyst.

Experimental procedure

We studied the conversion of toluene with methanol on native NaX and Cs-exchanged zeolite X impregnated with caesium nitrate. Cs-form of catalysts has been obtained in two ways. First technique was a conventional ion exchange (CIE) of NaX sample (grain size of 0.5–1 mm) with 1 N CsNO₃ (solid to liquid ratio (S:L) 1:5) by successive twofold treatment at 423 K during 2 hours under hydrothermal conditions. The sample was washed thoroughly between and after treatments with distilled water and dried in a drying oven at 373 K.

Second technique was solid state ion exchange (SSIE) of HX with CsCl. Hydrogen form has been obtained from ammonium one by calcination in muffle at 823 K during 3 hours. By-turn, NH₄-form has been synthesized treating NaX sample with 3 N NH₄NO₃ (S:L = 1:5) in water bath at 358 K during 3 hours followed by distilled water washing and drying at the end. After that zeolite sample and thoroughly grinded caesium chloride in stoichiometric ratio to ion exchange capacity of zeolite were suspended in hexane under continuously mixing during 2 hours. Then hexane was evaporated, dry product was pressed and crushed to grain of 0.5–1 mm size. The sample obtained was temperature treated in flow set-up (temperature rise 1–5 K/min, carrier gas helium, 30 cm³/min) at 823 K during 3 hours.

Dried (653 K, 2 hours) ion-exchanged zeolite samples both CsX/Cs (CIE) and SSCsX/Cs (SSIE) were impregnated with CsNO₃ solution in ratio of 1 mol of Cs in solution per 10 mol of charge compensating Cs⁺ in zeolite.

Adsorption/desorption isotherms of nitrogen (T = 77 K) were measured on Quantachrome® Autosorb NOVA-2200e set-up after samples evacuation at 453 K during 20 hours. Pore structure parameters (surface area on Brunauer–Emmett–Teller (S_BET), micropore and mesopore surface areas (S_micro and S_meso) and micropore volume (V_micro) on t-method, total pore volume (V)) were calculated by means of ASiQwin™ software.

Catalysts were tested in side-chain toluene with methanol alkylation at 698 K under flow conditions without using a carrier gas (p = 101.325 kPa, weight hourly space velocity (WHSV)—1 h⁻¹, toluene to methanol molar ratio—2) and in vacuum using the same toluene/methanol mixture.

Results and discussion

Catalytic results are shown in Figure 1(a) to (d). First of all note that methanol converts very active and almost completely in all cases except the case of catalyst NaX in vacuum (Figure 1(a)). In this case, methanol conversion is also sufficiently high. Such high conversions under short residence time of reactants on the catalyst coupled with low aromatics yields indicate an instant side reaction of deep decomposition of alcohol toward carbon monoxide.

OPEN IN VIEWER

Figure 1.

Parameters of flow alkylation under atmospheric pressure and in vacuum on the catalysts NaX and CsX/Cs: conversion of methanol (a) and toluene (b), yields of major liquid products (c) and benzene (d). X_M: conversion of methanol; X: conversion of toluene; Y_MP: yields of major liquid products (xylenes for NaX, total styrene and ethylbenzene for CsX/Cs) referred to methanol fed; Y_B: yields of benzene referred to methanol fed.

Toluene conversions on the NaX catalyst both in vacuum and under atmospheric pressure (101.325 kPa) are low and very close (Figure 1(b)). In vacuum reaction occurs to a great extent (∼50%) toward the formation of xylenes. Yield of major products (xylenes) remains unchanged as a contact time of reactants with a catalyst increases (Figure 1(c)). That is, whole potential for a ring alkylation is realized already at the initial stages. Catalyst is nonselective in goal reaction due to low basicity of framework oxygen atoms (Palomares, 1997; Vayssilov et al., 2000).

At the same time, the increase in contact time for the catalyst CsX/Cs lead to an order of magnitude growth of conversion (Figure 1(b)). The catalyst is selective; a ring alkylation has been diminished on it from the very beginning. Products of side-chain alkylation in vacuum amount to about 83% of liquid alkylate. Under atmospheric pressure their yield increases in the same manner as toluene conversion (Figure 1(c)) unlike to major products’ yield on the catalyst NaX. All these indicate that short contact time hinders to fulfill completely the catalyst’s potential for side-chain alkylation, differently from the potential for ring alkylation; the latter, so, is easier realized. In support note that total yield of styrene and ethylbenzene on the catalyst NaX increases (in 1.5 times) too as a contact time increases.

Coupled with alkylation products we observed benzene formation under short contact time in vacuum both on nonselective NaX and selective CsX/Cs (Figure 1(d)). This shows the primary nature of benzene occurrence. Benzene content in the liquid products was quite noticeable and made up 9.29 and 17.17 mol% for NaX and CsX/Cs, respectively. That is, benzene is formed more intensively independent of xylenes in the case of CsX/Cs (xylenes is absent in the liquid products) than together with them on the catalyst NaX (xylenes content in the liquid products is 49.2%). Thus, formation of benzene and xylenes are not correlated one with another, and CsX/Cs catalyst is more suitable for the by-reaction. Yield of benzene referred to methanol fed on the NaX doubles as a contact time increases, while on the CsX/Cs it halves, reflecting consequence of concurrency between the reaction of formation of this compound and the side-chain alkylation. Indeed, growth of CsX/Cs activity toward side-chain alkylation gives rise to such concurrency unlike the case of NaX catalyst (Figure 1(c)). On the whole, benzene yield is limited by values of 0.2–0.4 mol% (Figure 1(d)) apparently due to improper reaction conditions, for instance low temperature (earlier we have observed in micro pulse toluene methylation sharp intensification of benzene formation when a temperature is raised in the range of 673–773 K, especially on the catalysts in Rb- or Cs-form unlike their Na- or K-form (unpublished data)).

The results obtained are supplemented with data for toluene alkylation on the catalyst SSCsX/Cs in vacuum. This catalyst possesses pore structure parameters drastically reduced (Table 1) apparently as a result of procedure of SSIE. Consequently, a content of active sites in the catalyst has been reduced too. Small quantity of sites for interaction of reactants’ molecules especially favors detection of initial reaction stages.

OPEN IN VIEWER

Table 1.

Parameters of pore structure of caesium-modified catalysts according to low temperature adsorption of nitrogen (T = 77 K).

Catalyst	SBET (m2/g)	Smicro (m2/g)	Smeso (m2/g)	V (cm3/g)	Vmicro (cm3/g)	Vmicro/V
CsX/Cs	382.6	374.7	7.88	0.172	0.141	0.82
SSCsX/Cs	10.61	11.25 a	10.61	0.020	0.004 a	0.20

SSCsX/Cs demonstrates in vacuum twice lesser values of conversions of methanol and toluene as well as value of the yield of major products (ethylbenzene in this case) comparing with CsX/Cs. From absence of xylenes one may state sufficiently high basicity of catalyst and as a result—its selectivity in side-chain alkylation. Yield of benzene on it makes up 0.11 mol% that three times smaller than on CsX/Cs. Decrease in value of this parameter under decrease in content of active sites indicates that benzene formation is not a gaseous phase reaction.

It has been shown in Palomares et al. (1997) that less basic alkali exchanged zeolites (e.g., NaX) preferentially sorb methanol over toluene. More basic zeolites (i.e., CsX) on the contrary preferentially sorb toluene over methanol. The following may be assumed considering conclusions of the work cited and that benzene is formed more intensively in the cases of more basic catalysts: the same sites as acting in side-chain toluene alkylation (probably basic framework oxygen atoms) are responsible for benzene formation, and toluene is a raw material for this. Because benzene is detected among products even under short contact time in spite of inappropriate conditions realized the reaction appears to be monomolecular.

Conclusion

Conversion of toluene with methanol on basic catalysts NaX and CsX impregnated with caesium nitrate has been studied under flow conditions without using a carrier gas and in vacuum. Benzene formation has been observed in the very beginning of conversion of toluene/methanol mixture that let to believe it as a primary product. Dynamics of alteration of benzene content in reaction products has been explained. The nature of sites responsible for benzene formation has been suggested taking into account preferences in toluene or methanol adsorption on the catalyst of different basicity.