SELECTIVE DISPROPORTIONATION OF TOLUENE TO PRODUCE BENZENE AND P-XYLENE

Abstract

Toluene disproportionation reaction to produce benzene and xylenes is of considerable importance due to the increasing demand of benzene and xylenes in chemical industries. The demand of toluene is found to be the lowest, whereas, the para-isomer of xylene is of greater significance due to its use in the manufacture of polyester fiber and many petrochemicals. The commercial process for p-xylene production from toluene involves the vapor phase disproportionation of toluene over a variety of acid zeolites. Toluene disproportionation yields benzene and a mixture of xylenes as initial products inside the zeolite pores. The products subsequently diffuse out of the zeolite pores. Owing to its smaller size and very high diffusivity, p-xylene diffuses out of the pores at a very high rate in comparison to o- and m-xylene. The ortho- and meta-xylenes undergo isomerization reaction within the zeolite. Several researchers have shown enhanced para-selectivity of zeolites by the use of modifier agents such as Si, B, P and Mg. These modifier agents have been reported to (i) partially block the pores, increase tortuosity and thereby delay the exit of larger molecules of o- and m-xylenes, and (ii) block the unselective active sites and inactivate the external active sites for secondary isomerization. The modification of the zeolites by ion exchange has been reported to increase the acid strength when pretreated by La3+ ion and increasing the initial toluene conversion and also the hydrothermal stability. Based on the fact that the ion exchange of a zeolite may lead to (i) selectivity of the reactions based on the nature of the cation, (ii) blockage of the zeolite pore mouth in proportion to the size of the exchanged ion, leading to shape selectivity for the desired product (p-xylene in the present case), and also (iii) greater stability of the exchanged zeolite than the modified form, the present work has been undertaken. The present work deals with the preparation, characterization and activity tests of the following catalysts for toluene disproportionation: (a) ZSM-5 zeolite ion-exchanged with several elements representing various groups of the Periodic Table. (b) P-zeolites exchanged with Ni, Co, Ce, and H. (c) Y-zeolites exchanged with H, Ni. (d) ZSM-5 exchanged simultaneously with two different elements. (e) 13X zeolite exchanged with Ce, La and also simultaneously with two different elements, namely BaCe, CrCe, CrLa, FeCe, CrNi, ZnNi and ZnCe. (f) Perovskite type oxide, namely LaCo03, LaNi03, BaFe03, LaFe03 .BaNi 03 (g) Ni/Ti02, Ni/MgO, Ni/Zr02. The ion-exchange of the zeolite powder was performed by mixing the powder with the respective metal nitrate solutions having a metal content. The solution was heated at 50°C for about 3h with constant stirring. The contents were further refluxed under atotal condenser at 95±5 °C for 15 h. The solid powder was then filtered, dried at 110 °C overnight and calcined at 700°C for 10 h. The dried powder, in order to make it suitable for tests in the reactor, was pelletized at 104 bar pressure, and then broken and sieved to get the particles in the specific size range of 0.33 to 0.52 mm. The catalytic activity tests were carried out in adown flow vertical silica glass tubular reactor. The reactor tube having 1cm inside diameter and 50 cm length was heated electrically and the temperature was controlled to within ±1°C. The catalyst particles were mixed with quartz particles of similar size range and packed in the middle of the reactor tube. Above the catalyst bed quartz particles (0.95 - 1.14 mm size) were packed serving the dual purpose of preheating as well as mixing the reactants. Passing air through the reactor at 500 °C for lh activated the catalyst. The catalyst bed was cooled to the reaction temperature in the nitrogen atmosphere before carrying out reaction tests. The toluene feed was pumped into the reactor at a regulated rate using a metering pump. The nitrogen gas was simultaneously introduced at the top of the in reactor tube through a Yjunction and a specified weight hourly space velocity (WHSV) of the toluene feed was maintained. The products from the reactor were passed through a set of glass condensers fitted in series and cooled by ice-cold water. The liquid and gaseous products at each reaction temperature were collected in sampling bottles for further analysis. The flow rate of both the liquid and vapour were also recorded for overall mass balance. The liquid product was analyzed by agas-liquid chromatograph using a4mlong stainless steel column of 3.2 mm outside diameter packed with 7% Bentone-34 ±5% Dinonyl phthalate on Celite 545, 60-80 mesh. The oven temperature was kept at 90°C, and the flow rate of nitrogen gas as the carrier was maintained at 30 ml min"1. The nitrogen to toluene molar ratio was maintained at 5.9 in all the runs. The zeolite samples were characterized by XRD for phase formation and percent crystalUnity, SEM for particle size, and atomic absorption spectrometer and inductively coupled plasma (ICP) for elemental analysis. Infrared spectrograms of the samples were also taken to estimate the relative strengths of Bronsted acid sites (strong, weak and medium) and Lewis acid sites. Results and Discussion The XRD pattern ofthe zeolite samples indicated the crystalline structure of the zeolite. However, the percent crystallinity calculated after ion exchange was found to decrease due to partial dealumination of the zeolite. Catalyst systems were studied for toluene disproportionation and the effects of various parameters were observed. Effects of reaction temperature in the range of 300-600°C, time-on-stream studies at a fixed temperature of 550° for 5 h, silica/alumina ratio (SAR) of the ZSM-5 zeolite catalysts at a fixed temperature, carrier gas (N2, H2, Ar), type of zeolite (ZSM-5 with SAR-60, 90 &235, P- and Y-), percentage ion exchange, acidity of the catalysts and crystallinity of the catalysts were investigated. IV Na-form of the zeolite was first converted to H-form and subsequently exchanged with different cations The activity tests have been compared group wise (belonging to the same group of the Periodic Table) as well as among the bests of different groups. The toluene conversion, in general, was found to increase with temperature upto 500 or 550°C and thereafter showed a decline. The maximum toluene conversion obtained was around 35% with NiZSM-5 at 500°C and one atmospheric pressure. The p-xylene in the xylenes product, in general, increased with the reaction temperature. It varied between 90 and 97% when conversion was less than 2% and 50-60% when conversion was between 15 and 30%. The percent pxylene yield, defined as mole p-xylene formed/mole toluene fed, also increased with reaction temperature upto 500°C and then decreased. In general, it varied between 3 and 4% and in a particular case with a certain catalyst it was found to be as high as 9%. Benzene/xylene (B/X) ratio varied from less than 1 to 3. Time-on-Stream studies were conducted at a 550°C for 5 hours and one atmospheric pressure in the presence of different carrier gases (N2, H2, and Ar). In general, the activity with all the three carrier gases remained almost unchanged with time excepting that for nickel exchanged catalysts. The presence of nickel, it seems, does not help in oxidizing the coke formed onthe catalyst. The toluene conversion was observed to be a little lower with nitrogen as the carrier gas than with hydrogen. Argon showed a trend in between the above two gases. However, p-xylene in the xylenes product was found to be the highest with nitrogen as the carrier gas, followed by argon and hydrogen, respectively. The sequence was found in reverse order for p-xylene yield, i.e., and hydrogen as the carrier gas was the best followed by argon and nitrogen in that order HZSM-5 with three SAR (60, 90 and 235) were tested for their catalytic performance. Para-xylene yield with nitrogen as the carrier gas followed the order SAR 60 > 90 > 235. Using Ni ZSM-5 catalyst and argon as the carrier gas a similar trend SAR 60 > 907>235 was observed. In some cases, the activity of the zeolite having SAR=235 was observed to be a little higher than for SAR the zeolite having 60 and 90. The acidity of the catalysts due to different cations present did not give a definite trend for toluene conversion. For example, toluene conversion at 500°C was found to increase in general with the increase in acidity (measured as mmol KOH/g) for most of the catalysts having less than 4% of metal ion-exchange. With higher exchange, however, the trend was reversed. At 550°C, the toluene conversion increased with acidity. The loading of nickel on a ZSM-5 was varied from 2% to 12%. Exchanged catalysts with 4% nickel solution showed greater toluene conversion in comparison to 6, 10 and 12 percent in that order. The p-xylene in the xylenes product was minimum with 4% loading, whereas, the p-xylene yield was the best obtained with this loading. The results, thus, indicate 4% as the optimum Ni-concentration for exchange. The results have also been confirmed with other catalytic components. ZSM-5, P-zeolite and Y-zeolite were used with and without exchange for the comparison of catalytic activity. The results with, for example, 4% nickel exchange catalysts showed toluene conversion in the order: ZSM-5 (SAR=60) > ZSM-5 (SAR=90) > p-zeolite > Y-zeolite. Para-xylene in the xylenes product for the same catalysts had the order: p-zeolite > ZSM-5 (SAR=235) > ZSM-5 (SAR=90) > Yzeolite, whereas, the p-xylene yield was obtained in the order: ZSM-5 (SAR 60) > p-zeolite >ZSM-5 (SAR=90) > ZSM-5 (SAR 235) > Y-zeolite. Many catalysts based on 13X zeolites (single as well as double exchange), perovskites and Ni/Ti02 etc. as mentioned above were tested to explore their activity for toluene disproportionation. The results, however, were found discouraging in comparison to the ZSM-5 and p-zeolites. The increase in the average crystallite size of ion exchanged ZSM-5 from 5 to 36 micron hardly modifies para-xylene selectivity. The toluene conversion was generally found to increase with an increase in the acidity of the catalyst. No VI correlation could be established between the BET surface area of the zeolite catalysts (ZSM-5 based) and their catalytic activity. As the silica-alumina ratio of the ZSM-5 zeolite decreased from 235 to 60; toluene conversion and the total p-xylene yield increased whereas p-xylene in the xylenes product decreased. Catalytic activity for toluene disproportionation increased with reaction temperature and the optimum temperature for most of the catalysts lied between 500 and 550°C at the experimental conditions of the present work. Among the catalysts tested, namely, ZSM-5, X-, Y- and Beta zeolite, the catalytic activity, selectivity and stability of ZSM-5 based catalysts were found to be the best. Para-xylene selectivity in the product using ion-exchanged ZSM-5 catalysts was found to be much higher than its equilibrium value of 24%. The catalytic activity of zeolite catalysts were found to increase after ion exchange. Time-on-stream studies with the exchanged ZSM-5 showed poorer performance at 550°C. Catalysts with higher acidity value, like, NiZSM-5, gave high toluene conversion. With beta zeolite toluene conversion activity decreased during time-on-stream tests. The performance of ZSM-5 with, or, without ion exchange showed that with H2 as the carrier gas, highest toluene conversion was obtained, followed by argon and nitrogen in that order. Para-xylene selectivity was found to be the highest with nitrogen followed by hydrogen and argon, whereas p-xylene yield was found to be maximum with nitrogen followed by hydrogen and argon. The activities ofperovskite based oxides, ion exchange 13X zeolites, Ni/Ti02, Ni/MgO and Ni/Zr02 catalysts for toluene conversion were extremely poor in the presence of nitrogen as the carrier gas.