Zeolites transition-state selectivity

Figure C2.12.10. Different manifestations of shape-selectivity in zeolite catalysis. Reactant selectivity (top), product selectivity (middle) and transition state selectivity (bottom).

Many chemical reactions, especially those involving the combination of two molecules, pass through bulky transition states on their way from reactants to products. Carrying out such reactions in the confines of the small tubular pores of zeolites can markedly influence their reaction pathways. This is called transition-state selectivity. Transition-state selectivity is the critical phenomenon in the enhanced selectivity observed for ZSM-5 catalysts in xylene isomerization, a process practiced commercially on a large scale. 

It is not the catalytic activity itself that make zeolites particularly interesting, but the location of the active site within the well-defined geometry of a zeolite. Owing to the geometrical constraints of the zeolite, the selectivity of a chemical reaction can be increased by three mechanisms reactant selectivity, product selectivity, and transition state selectivity. In the case of reactant selectivity, bulky components in the feed do not enter the zeolite and will have no chance to react. When several products are formed within the zeolite, and only some are able to leave the zeolite, or some leave the zeolite more rapidly, we speak about product selectivity. When the geometrical constraints of the active site within the zeolite prohibit the formation of products or transition states leading to certain products, transition state selectivity applies. 

The metal complexes in an SIB catalyst are confined to separate supercages. Consequently, the formation of inactive dimers is no longer possible. Shape-selectivity is another feature of SIB catalysts that follows from the restricted space inside the zeolite pore system. This can be simply due to discrimination in size of the reactant molecules (a large reactant molecule is excluded from the zeolite) or to a constrained orientation of the reactant at the catalytic site (transition state selectivity). 

There are four widely accepted theories of shape selectivity reactant shape selectivity (RSS), product shape selectivity (PSS), transition state selectivity (TSS) (Figure 12.2), and concentration effect all of them are based on the hypothesis that the reactions occur within the zeolite micropores only. As indicated earlier, this hypothesis is often verified, the external surface area of the commonly used zeolites being much lower (one to two orders of magnitude) than their internal surface area. ... 

The use of zeolites can overcome many of these limitations and provide new controlled entries into these oxidized hydrocarbons and new materials. For example, some of the most valuable industrial intermediates are terminally oxidized hydrocarbons, snch as n-hexanol or adipic acid, that are not readily available in free-radical chain processes. The ability of zeolites to function as shape-selective catalysts can, in principle, be used to restrict access, by reactant or transition state selectivity, to sites not normally attacked by oxidants [3]. 

Transition-state selectivity is sometimes difficult to distinguish from product shape selectivity. A recent study by Kim et al. (8) shows that the high para-selectivity for the alkylation of ethylbenzene with ethanol in metallosilicates (MeZSM-5) is not due to product selectivity alone. They conclude that the primary product of the alkylation on ZSM-5 type metallosilicates is p-diethylbenzene which isomerizes further inside the cavity of ZSM-5 to other isomers. As the acid sites of zeolites becomes weaker (achieved by substituting different metals into the framework of the zeolite), the isomerization of the primarily produced p-isomer is suppressed. Although Kim et al. attribute this suppression of the isomerization activity to restricted transition-state selectivity, it is more likely that this suppression is due to the decrease in acid strength. 

Direct observation of transition-state selectivity has been observed from the low-temperature cyclization of dienes inside H-mordenite and H-ZSM-5 (9). By using electron spin resonance (ESR) spectroscopy, it has been possible to explore radical formation upon the sorption of dienes on H-mordenite and H-ZSM-5. From the analysis obtained, it was found that the dienes are not very reactive for oligomerization inside H-mordenite channels. Heating H-mordenite with presorbed 1,4-pentadiene or 1,5-hexadiene yields selective cyclization of molecules via cycloalkenie radicals inside the H-mordenite channel. However, in the smaller pores of H-ZSM-5 (although die nature of both acid and redox sites in both zeolites are the same) no eyclo-olefinie radicals are formed as shown by the ESR spectrum. These experiments illustrate the reality of transition-state selectivity inside the pores of zeolites. 

Ti-Beta was also applied in the selective transformation of a-pinene oxide to camphenolenic aldehyde.78 A selectivity of over 98% was observed in the gas-phase reaction that was explained as a combination of a Lewis catalysed reaction in the absence of Brpnsted acid sites. Furthermore the concentration of a-pinene oxide in the zeolite pores was found to be an important factor not ruling out additional transition state selectivity as well. 

Supported non-framework elements, as well as substituted or doped framework atoms, have been important for zeolite catalyst regeneration. By incorporating metal atoms into a microporous crystalline framework, a local transition state selectivity can be built into the active site of a catalytic process that is not readily attainable in homogeneous catalysis. The use of zeolites for carrying out catalysis with supported transition metal atoms as active sites is just beginning. The local environment of transition metal elements as a function of reaction parameters is being defined by in situ Mossbauer spectroscopy, electron spin echo measurements, EXAFS, and other novel spectroscopic techniques. This research is described in the second part of this text. 

H. van Bekkum et al. (72) studied a number of catalysts in the Fischer synthesis starting from l-phenyl-2-butanone 40 (with R, = Ph, R2 = CH3) and phenylhydrazine. The isomeric products are the bulky 2-ethyl-3-phenylindole 45 (with R, = Ph, R2 = CH3) and the linear 2-benzyl-3-methylindole 46 (with R, = Ph, R2 = CH3). Catalysis of the inolization of 40 by soluble as well as solid (e.g. Amberlyst 15) catalysts typically yielded a mixture of the two isomers in a bulky/linear ratio of about 75/25. Zeolite BEA reverses this bulky/linear ratio giving 75% of the linear isomer 46, a result interpreted in terms of restricted transition-state selectivity. Although in zeolite BEA the intraporous formation of 45 is largely suppressed, it is in fact probably not completely inhibited. 

The selective Fischer synthesis of 2-benzyl-3-methylindole from l-phenyl-2-butanone and phenylhydrazine and catalyzed by zeolite BEA can be considered as an interesting example of restricted transition state selectivity. 

As noted above, one of the most important characteristics of zeolites is their ability to discriminate between molecules solely on the basis of their size. This feature, which they share with enzymes, is a consequence of them having pore dimensions close to the kinetic diameter of many simple organic molecules. Hence, zeolites and zeotypes have sometimes been referred to as mineral enzymes. This so-called shape selectivity can be conveniently divided into three categories substrate selectivity, product selectivity and transition state selectivity. Examples of each type are shown in Fig. 2.11. 

Then, we investigated a reaction for which transition state selectivity is important. It appears that several alternative reaction pathways can be followed among which some lead to a minimization of the steric constraints. These steric constraints are strongly dependent on the local topology of the zeolite framework as well as on the geometry of the transition state. 

We also mentioned stereospecificity of metal-catalyzed reactions inside zeolite cavities. In acid catalysis by zeolites it is well known that shape selectivity can be imposed by (1) selective admission of reactants fitting into zeolite pores, (2) selective release of products able to diffuse through zeolite channels, while larger molecules are retained, and (3) transition state selectivity, favoring, e.g., a monomolecular transition state over a bimolecular state in a narrow cavity. New tools that have conceptually been added to this arsenal include the collimation of molecules diffusing through well-defined pores, which then hit an active site preferentially via one particular atom or group. 

One of the most in ortant properties of zeolites is their ability to cany out shqie selective reactions [5]. These can be cl sified as, firstfy, product shape selective reactions in which the only products formed are those which can diffiise out of e pores of die zeolite, second, reactant shape selective reactions which occur when some of the molecules in a reactant mbcture are too large to diffiise through the catalyst pores, and, thirdfy, restricted transition-state selective reactions in which the only reactions which occur are those in which qiace exists in the pores or cavities to allow the formation of the activated transition state con lex. In some cases where the zeolite is three dimensional the gze of the channel intersections will also be a determining ictor. This unique catalytic property is related to the pore size of the zeolite and has led to the synthesis of zeohtes with a very w e range of pore gzes. 

Shape selectivity can be induced by differences in the diffusivities of the reactants and/or the products or by steric constraints of the transition state. A schematic representation of the three types of shape selectivity, i.e., the limitations of the access of some of the reactants to the pore system (reactant selectivity), the limitation of the diffusion of some of the products out of the pores (product selectivity) and constraints in forming certain transition states (transition state selectivity) are given in Fig. 8. Differentiation between the latter two is difficult as the kinetic results may be disguised when the overall rate is influenced by the rates of diffusion. In situ IR and NMR spectroscopy have contributed much to our understanding of these complex phenomena. The aspects of shape selectivity have been extensively discussed and excellent reviews exist [242,243,244]. The examples given here should only illustrate what can be achieved by employing a zeolite and why the pathway of a particular reaction is influenced. 

One of the most discussed cases of shape selectivity involving transition state selectivity or product diffusional constraints is the production of p-xylene over chemically modified MFI zeolites [248]. Several processes exist which utilize the shape selectivity of these zeolites, for example the alkylation of toluene with methanol [249], xylene isomerization [250] and selective toluene disproportionation [251]. The first two of these examples shall be used to describe in detail the principal possibilities to tailor the reaction pathway by shape selectivity. 

The isomerization of m-xylene is a good example of transition state selectivity [256]. Irrespective of the temperature and coverage, (a particular sample of) MFI showed a product ratio of p- to o-xylene of 2 1. In the zeolite pores, only m-xylene was found to be sorbed in appreciable quantities. Thus, the reaction rate was concluded not to be influenced by the preferred retention of one of the products. The selectivity must be governed by the differences in the transition states of the two products. The constant selectivity with varying reaction temperature indicates an identical apparent energy of activation for the formation of p- and o-xylene. Thus, the different selectivities must be caused by differences in the... 

Restricted transition state selectivity is observed when the transition state leading to one product is more bulky than that leading to another. In this case the product from the less-restricted transition state will be favored. As an example, the disproportion of o-xylene to trimethyl benzene and toluene requires a bulky diaryl species as the transition state but isomerization to m- or p-xylene simply requires successive 1-2 methyl shifts, a process having a smaller transition state. As the size of the zeolite cavity decreases, the ratio of the rate of disproportionation to the rate of isomerization decreases because the larger transition state is not as easily accommodated in the smaller cavities. 2... 

Running the Fisher indole synthesis on an unsymmetrical phenyl hydrazone gives a mixture of 2,3-disubstituted indoles. For example, reaction of the phenyl hydrazone, 34, with acid can give both 35 and 36 (Eqn. 22.26). Soluble acids and Amberlyst-15 give these two products in a 75 25 ratio at 100% conversion. With an H-M catalyst they are formed in a 65 35 ratio but over a dealuminated H-beta zeolite, the selectivity is reversed and 36 is produced in an 82% yield at 100% conversion.62 n was proposed that the preferential formation of 36 over the H-beta catalyst was the result of a restricted transition state selectivity. ... 

The most important consequence of restricted transition state selectivity is that ZSM-5 and many other medium-pore zeolites deactivate much slower than most other crystalline and amorphous catalysts. The difference is not trivial. In most acid catalyzed reactions large-pore zeolites deactivate within minutes or in hours, whereas the activity of ZSM-5 ranges from weeks to years. Most of the coke in large-pore zeolites is formed within the pores. In ZSM-5 most of the coke is deposited on the outer surface of the crystals like an eggshell over an egg [23] because coke precursors cannot form in the pores of pentasil molecular sieves. The resistance of ZSM-5 to coking makes a number of industrial processes economical. 

Olson and Haag observed a high selectivity for xylene isomerization versus disproportionation in the presence of ZSM-5 zeolite, and ascribed it to a transition state selective process rather than shape-selective diffusion. They were also able to produce p-xylene in high selectivity (up to 80%) from toluene disproportionation by using suitable modifications of the ZSM-5 catalyst. This selectivity was said to... 

As most of the acid sites are located in pores of molecular size the rate and the selectivity of catalytic reactions depend not only on the intrinsic properties of the sites but also on the pore structure. A zeolite catalyst selects the reactant or the product by their ability to diffuse to and from the active sites (reactant and product selectivity). Steric constraints in the environment of the sites limit or inhibit the formation of intermediates or transition states (restricted transition state selectivity) [24,25]. The strong polarizing interaction between zeolite crystallites and adsorbed molecules leads to an unusually high concentration of the reactants in the pores. This concentration effect causes an enhancement of the rates of bimolecular reaction steps over monomolecular reaction steps [26]. 

Zeolite titanium beta has been tested in the liquid- and gas-phase Meerwein-Ponndorf-Verley reduction of cyclohexanones and the Oppenauer oxidation of cyclohexanols. A high selectivity towards the thermodynamically unfavourable cis-alcohol was observed, which has been ascribed to transition-state selectivity in the pores of the zeolite. Under gas-phase conditions the dehydration of alcohols to cycloalkenes is observed as a side reaction. The catalyst was found to be active even in the presence of water and ammonia. 

High space velocity was found to favour a nearly exclusive formation of the para-isomer (figure 2), and the para-selectivity was also improved over each zeolite as the reaction temperature was decreased. We therefore suggest that only the para-isomers are formed within the pores initially due to transition state selectivity. Secondary isomerization reactions will, however, lead to the appearance of all three isomers in the product. This proposal is supported by the fact that the para-yield increased with increasing size of the zeolite crystals. When the extent of... 

---