Fischer-Tropsch reaction selectivity

The induction of steric effects by the pore walls was first demonstrated with heterogeneous catalysts, prepared from metal carbonyl clusters such as Rh6(CO)16, Ru3(CO)12, or Ir4(CO)12, which were synthesized in situ after a cation exchange process under CO in the large pores of zeolites such as HY, NaY, or 13X.25,26 The zeolite-entrapped carbonyl clusters are stable towards oxidation-reduction cycles this is in sharp contrast to the behavior of the same clusters supported on non-porous inorganic oxides. At high temperatures these metal carbonyl clusters aggregate to small metal particles, whose size is restricted by the dimensions of the zeolitic framework. Moreover, for a number of reactions, the size of the pores controls the size of the products formed thus a higher selectivity to the lower hydrocarbons has been reported for the Fischer Tropsch reaction. 

Catalysts were tested for activity in the Fischer-Tropsch reaction using a fixed-bed reactor. The catalyst (0.4 g) was reduced in situ in flowing hydrogen at 425°C for 7 h prior to testing. The test was performed under 2/1 H2/CO at 20 bar total pressure. The initial flow was 64 ml/min, but this was reduced after 24 h to increase the conversion. A final reading of activity and selectivity was taken after 100 h on stream. 

The mannitol-modified catalyst showed significantly increased activity in the Fischer-Tropsch reaction (Table 1.3). After 20 h on stream, the mannitol-modified catalyst is 286% as active as the unmodified catalyst, and 262% as active after 100 h. The selectivity values of the catalysts are similar. Hence, the characteristics of a mannitol-modified catalyst are that it has a higher activity than but the same selectivity as an unmodified catalyst. Increased activity for the FT reaction... 

Five of the chapters in this volume can be considered directly related to this topic. First, Edd Blekkan, 0yvind Borg, Vidar Froseth, and Anders Holmen (Norwegian University of Science and Technology, Trondheim) review recent work on the effect of water on the Fischer-Tropsch reaction. Steam is both a reactant and product in this syngas-based process, and its effect on Co- and Fe-based catalysts is important in determining the activity and selectivity of the FT process. 

An ex-carbonyl K-promoted alumina-supported catalyst prepared from Ru3(CO),2 and decarbonylated under H2 at 450°C was more dispersed and more active and selective for C2-C5 olefins in the Fischer-Tropsch reaction than conventionally prepared samples [108]. 

Metal molybdates421 and cobalt-thoria-kieselguhr422 also catalyze the formation of hydrocarbons. It is believed, however, that methanol is simply a source of synthesis gas via dissociation and the actual reaction leading to hydrocarbon formation is a Fischer-Tropsch reaction. Alumina is a selective dehydration catalyst, yielding dimethyl ether at 300-350°C, but small quantities of methane and C2 hydrocarbons423 424 are formed above 350°C. Heteropoly acids and salts exhibit high activity in the conversion of methanol and dimethyl ether.425-428 Acidity was found to determine activity,427 130 while hydrocarbon product distribution was affected by several experimental variables.428-432... 

Modification of the zeolite appears to have affected the selectivity of Ru in these hydrogenation reactions. Exchange of K cations for Na cations in Y zeolite increases the basicity of the support (ref. 9). In Fischer-Tropsch reactions over similar catalysts, Ru/Y catalysts so modified yielded significant increases in the olefinic product fraction at the expense of paraffins. Olefins are believed to be primary products in F-T synthesis, with paraffins being produced from olefins in secondary hydrogenation reactions. In an analogous fashion, the Ru/KY catalyst used in the present study might also be expected to... 

These mixed metal systems have also been tested with the transient method for catalytic activity in the Fischer-Tropsch reaction. We would like to remark here that the nature of the cation, anion, and zeolite are all important factors in the Fischer-Tropsch reactions that we have studied. Further details of these catalytic studies can be found elsewhere (23). We do observe here, however, that some catalysts that are completely reduced to the metallic state are not necessarily the most active catalysts. Also, even though the Mossbauer experiments suggest that 400°C is sufficient for complete reduction, higher activation temperatures can increase the activity and selectivity of these reactions. We have also observed that the cation definitely changes the product distribution and the activity. 

As described below, this approach is being applied by several groups to analyze elementary reaction steps that are part of the Fischer-Tropsch reaction scheme. A detailed imderstanding of the relationship between activation energies and site structure is becoming possible hence, the factors that control activity as well as selectivity can be identified. 

Discriminating between the two proposed mechanisms for the Fischer-Tropsch reaction is essential because only when the mechanism is known, can the structure sensitivity of the Fischer-Tropsch reaction and its relationship with selectivity and stability be understood. 

This relationship indicates that the rate of dissociation of adsorbed CO has to be fast in comparison with the other two rates, which control the selectivity of the Fischer-Tropsch reaction. As follows directly from the expression that defines a (Equation (8b)), the value of a will be close to one only when the rate of hydrocarbon chain termination is small and thus rate limiting for the Fischer-Tropsch reaction. Because a is related to 0c(l), one can deduce the following ... 

Equations (12a) through (12c) are of great significance because they allow formulation of quantitative relationships between the activation energies of the elementary steps of the Fischer-Tropsch reaction that have to be satisfied for a high chain-growth selectivity. 

It is now widely accepted that the activation of CO is highly structure sensitive (II). The activation of CO on most of the transition metals has been investigated. The computational results for cobalt (6) and ruthenium (5) are of particular relevance to us because these elements in the metallic state are active for the Fischer-Tropsch reaction. These results can be compared with those obtained for rhodium (40), which selectively catalyzes the formation of alcohols from CO and H2, and for nickel (30), which is a methanation catalyst. 

Assuming BEP-type relationships to be valid, we can make a prediction of the selectivity of fhe Fischer-Tropsch reaction as a function of the M—C bond energy. In Figure 10, a schematic representation is given of the relative rates of production of particular groups of Fischer-Tropsch products as a function of fhe M—C interaction energy. Four types of reaction are compared coke or carbide formation, hydrocarbon chain growth, CH4 formation, and CO dissociation. 

After a short review of the historical developments, the basic features of the Fischer-Tropsch reaction will be dealt with and special attention will be given to the possibilities of product selectivity control. Finally, the various mechanistic proposals known so far will be discussed in respect to recent studies of catalyst... 

Investigations of functioning catalysts with Mossbauer spectroscopy have been performed for a wide range of samples and applications. The reactions include hydrodesulfuration 15), the Fischer-Tropsch reaction (20,180), selective oxidation or oxidative dehydrogenation (181-186), and acetonitrile synthesis (187). 

In the past 15 years increased interest in the Fischer-Tropsch reaction over iron has been evidenced by a targe quantity of work in this area. However, most studies have been directed mainly at understanding the FT reaction, e.g. measuring activity/selectivity or investigating surface species taking part in the reaction rather than understanding the species responsible for deactivation. 

When comparing the competing processes for making hydrocarbons from synthesis gas - the Fischer Tropsch CO hydrogenation and the MTG conversion -the process flow sheets show as the main difference the additional step of methanol synthesis for the MTG route. However, product selectivity is basically different for both the conversions. And from this point of view the one or the other route can be the more favourable option as fitting best the particular demand pattern. Selectivity differences fundamentally result from the different kinds of chemistry which are involved Hydrogenation on special metal type catalysts in case of the Fischer Tropsch reaction and a conversion via car-benium ion intermediates on acidic sites, which is additionally constrained by shape selectivity in case of the MTG process. 

Due to the known limitations of the world oil reserves, methane oxidation under fuel rich conditions will become increasingly important for the production of synthesis gas. which through methanol synthesis and Fischer-Tropsch reactions is the basis of many important petrochemical synthesis routes. Therefore, catalytic oxidation of methane has again become the focus of much basic and applied catalysis research in recent years. In this context, Schmidt and coworkers were able to show recently, that catalytic direct oxidation of methane over noble metal coated monoliths can yield CO and H2 with very high conversions and selectivities at the desired 1 2 CO H2 ratio (Hickman and Schmidt. 1992 and 1993a Torniainen and Schmidt. 1994 Bharadwaj and Schmidt. 1995). 

The kinetics of the Fischer-Tropsch reaction has been studied on both cobs lt and iron. These studies supply data for the rational de gn of reactors and selection of optimum operating conditions they also prcMde information on the mechanism of the synthesis, thus aiding development of catalysts and processes. 

Fischer-Tropsch activity, selectivity and deactivation data obtained in fixed bed reaction tests of Co/Si02 catalysts are summarized in Table 1. The turnover frequencies (TOFs) or site time yields based on H2 uptake and on rate measured after 20 hours of reaction agree within a factor of two with those reported for other cobalt catalysts [2, 3, 25-27]. CO conversion and methane selectivity versus time for Cab-O-Sil supported cobalt at both low and high space velocities are shown in Figure 1. It can be seen that at high conversion the catalyst deactivates rapidly while at low conversion the catalyst appears to be stable. The conversion is proportional to the water partial pressure thus water could be causing this deactivation. 

---