Phenyl cyclodextrin catalysis

In these equations, Dmax is the larger of the summed values of STERIMOL parameters, Bj, for the opposite pair 68). It expresses the maximum total width of substituents. The coefficients of the ct° terms in Eqs. 37 to 39 were virtually equal to that in Eq. 40. This means that the a° terms essentially represent the hydrolytic reactivity of an ester itself and are virtually independent of cyclodextrin catalysis. The catalytic effect of cyclodextrin is only involved in the Dmax term. Interestingly, the coefficient of Draax was negative in Eq. 37 and positive in Eq. 38. This fact indicates that bulky substituents at the meta position are favorable, while those at the para position unfavorable, for the rate acceleration in the (S-cyclodextrin catalysis. Similar results have been obtained for a-cyclodextrin catalysis, but not for (S-cyclodextrin catalysis, by Silipo and Hansch described above. Equation 39 suggests the existence of an optimum diameter for the proper fit of m-substituents in the cavity of a-cyclodextrin. The optimum Dmax value was estimated from Eq. 39 as 4.4 A, which is approximately equivalent to the diameter of the a-cyclodextrin cavity. The situation is shown in Fig. 8. A similar parabolic relationship would be obtained for (5-cyclodextrin catalysis, too, if the correlation analysis involved phenyl acetates with such bulky substituents that they cannot be included within the (5-cyclodextrin cavity. 

An important feature of the cyclodextrins is that they can also accelerate chemical reactions, and therefore serve as models for the catalytic as well as the binding properties of enzymes. The rapid reaction is not catalysis, since the dextrin enters reaction but is not regenerated presumably it arises from approximation, where complex formation forces the substrate and the cyclodextrin into intimate contact. In particular, cyclodextrins can increase the rate of cleavage of phenyl pyrophosphate by factors of as much as 100 (Cramer, 1961). More recent work has improved upon this early example. 

The situation is complex. In another study we examined the cyclization of compound 54 catalyzed by cyclodextrin bis-imidazoles [140]. This dialdehyde can perform the intramolecular aldol reaction using the enol of either aldehyde to add to the other aldehyde, forming either 55 or 56. In solution with simple buffer catalysis both compounds are formed almost randomly, but with the A,B isomer 46 of the bis-imidazole cyclodextrin there was a 97 % preference for product 56. This is consistent with the previous findings that the catalyst promotes enolization near the bound phenyl ring, but in this case the cyclization is most selective with the A,B isomer 46, not the A,D that we saw previously. Again the enolization is reversible, and the selectivity reflects the addition of an enol to an aldehyde group. The predominant product is a mixture of two stereoisomers, 56A and 56B. Both were formed, and were racemic despite the chirality of the cyclodextrin ring. 

In catalysis the excess of a phosphine ligand is often necessary because it preserves the active species in the medium [2a]. However, it retards to some extent the co-ordination of the alkene to the metal center. Recent studies, performed by Monflier and coworkers, have shown that the water-soluble TPPTS ligand could reduce the rate of the reaction by another effect. Indeed, TPPTS can be included partially in the cyclodextrin hydrophobic cavity [53,54] NMR measurements, observation by UV-visible spectroscopy and circular dichroism, as well as scanning tunneling microscopy are consistent with a 1 1 inclusion complex in which the phosphorus atom would be incorporated into the torus of the /S-CD. NMR investigations carried out on (m-sulfonatophenyl)diphenylphosphine have shown that a phenyl group is incorporated [55]. Thus, the phosphorus ligand could modify the association constant of the alkene with the cyclodextrin so that the mass transfer between the two phases could be decreased. 

---