A new approach to the modeling of deactivation in the conversion of methanol on zeolite catalysts Ton V.W. Janssens * Haldor Topsøe A/S, Nymøllevej 55, DK-2800 Lyngby, Denmark article info Article history: Received 8 January 2009 Revised 18 February 2009 Accepted 8 March 2009 Available online 1 May 2009 Keywords: Conversion of methanol to hydrocarbons Zeolite Methanol to gasoline ZSM-5 Deactivation Kinetic model Catalyst lifetime Conversion capacity abstract The deactivation of a zeolite catalyst in the conversion of methanol to hydrocarbons is described as a reduction of the effective amount of catalyst with time on stream. With the assumptions that the conver- sion of methanol is a first-order reaction, and that the loss of active catalyst is proportional to the con- version, an expression for the conversion with time on stream is obtained, which describes the experimental data well. This expression contains the rate constant, that characterizes the activity, and a deactivation coefficient that describes the deactivation behavior as parameters. It is shown that active catalysts show a more sudden decrease in conversion, and that the deactivation rate determines the time at which the decrease in conversion is observed. If the initial conversion is close to 100%, the lifetime to 50% conversion does not depend on the activity, and the deactivation coefficient is directly derived from the experimental data, by dividing the measured lifetime to 50% conversion by the applied contact time. The lifetime to all other conversion levels is dependent on both deactivation and activity, which implies that a catalyst lifetime to breakthrough of methanol does not scale with the deactivation rate. Likewise, it is shown that the conversion capacity is a good characterization of the deactivation, and this can be read- ily calculated as the product of the space velocity of methanol (WHSV) and the lifetime to 50% conversion. The amount of converted methanol at other conversion levels depends on the deactivation, the activity, and applied contact time (space velocity), and is therefore less appropriate to use as a characterization of the deactivation behavior. Ó 2009 Elsevier Inc. All rights reserved. 1. Introduction The methanol to gasoline (MTG) reaction is the conversion of methanol over a zeolite-based catalyst to light olefins (C 2 –C 4 ) and liquid products in the boiling point range of gasoline, which typically occurs in the temperature range of 300–400 °C. In the TIGAS (Topsøe Integrated GASoline) process, the MTG reaction is combined with the methanol synthesis, and in this way an efficient process converting synthesis gas to gasoline is obtained [1,2]. The MTG reaction can be regarded as a sequential reaction, con- sisting of the following steps [3]: Methanol ¢ DME ! light olefins ! gasoline products and this is known to occur on a variety of zeolites [4]. The ZSM-5 zeolite, however, is the most commonly applied one, since it ap- pears to be superior for this reaction. The MTG reaction over a ZSM-5-based catalyst is, like many other hydrocarbon reactions over zeolite catalysts, always accompanied by coke formation, which leads to deactivation of the catalyst. The deactivation by coke formation occurs through blocking the access to the active acid sites, either by deposition of carbonaceous compounds directly on the acid sites itself and in the micropore channels of the zeolite (internal coke), or by blocking the entrance to the micropores, thereby preventing the diffusion of methanol molecules into the zeolite structure (external coke) [5–11]. Loss of activity due to the coke formation is, in principle, reversible, and the catalytic activity can be restored by a regeneration that removes the deposited coke. A common procedure is to burn off the coke with oxygen at 500– 600 °C [7,8,12,13]. Regenerated catalysts often show a somewhat lower activity than fresh ones, possibly due to dealumination of the zeolite, which typically occurs at elevated temperatures (>500 °C) in the presence of water. As a consequence of the deactivation, the total amount of meth- anol that can be converted over a ZSM-5 catalyst is limited, and it is therefore important to know the deactivation behavior of the cat- alyst. To characterize catalyst deactivation, it is required to de- scribe how the catalytic activity decreases during the catalyst lifetime. Different approaches for such a description have been developed. The approach often encountered in the literature is based on a description of carbon formation as a function of contact time, e.g. by the empirical Voorhies equation [11,14,15]. The cata- lyst activity is then related to the carbon content in the catalyst, using (semi-) empirical relations, resulting in the description of 0021-9517/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jcat.2009.03.004 * Fax: +45 4527 2999. E-mail address: tvj@topsoe.dk Journal of Catalysis 264 (2009) 130–137 Contents lists available at ScienceDirect Journal of Catalysis journal homepage: www.elsevier.com/locate/jcat