Olefin complexation, stereoselection

The most famous mechanism, namely Cossets mechanism, in which the alkene inserts itself directly into the metal-carbon bond (Eq. 5), has been proposed, based on the kinetic study [134-136], This mechanism involves the intermediacy of ethylene coordinated to a metal-alkyl center and the following insertion of ethylene into the metal-carbon bond via a four-centered transition state. The olefin coordination to such a catalytically active metal center in this intermediate must be weak so that the olefin can readily insert itself into the M-C bond without forming any meta-stable intermediate. Similar alkyl-olefin complexes such as Cp2NbR( /2-ethylene) have been easily isolated and found not to be the active catalyst precursor of polymerization [31-33, 137]. In support of this, theoretical calculations recently showed the presence of a weakly ethylene-coordinated intermediate (vide infra) [12,13]. The stereochemistry of ethylene insertion was definitely shown to be cis by the evidence that the polymerization of cis- and trans-dideutero-ethylene afforded stereoselectively deuterated polyethylenes [138]. 

Yttrium-catalyzed cascade cyclization/hydrosilylation of 3-(3-butynyl)-l,5-hexadienes was stereospecific, and syn-19 (R =Gy, R = OGPh3) underwent cascade cyclization/hydrosilylation to form 80b (R = Gy, R = OGPh3) in 97% yield as a single diastereomer (Scheme 20). The regio- and stereoselective conversion of syn-19 to 80b was proposed to occur through an initial 5- x -intramolecular carbometallation via a chairlike transition state that resembles alkenyl olefin eomplex syn- m. followed by S-exo intramolecular carbometallation via a boatlike transition state that resembles alkyl olefin complex boat-llm. The second intramolecular carbometallation presumably occurs via a boatlike transition state to avoid the unfavorable 1,3-interaction present in the corresponding chairlike transition state (Scheme 20). 

A prerequisite for effective asymmetric hydrogenation is that the prochiral olefin is bound stereoselectively to metal at the rate-determining transition-state (Scheme 1). It is therefore of interest to consider stable metal-olefin complexes which may exist as diastereomers by virtue of alternative modes of prochiral olefin complexation. Most work has been done with comparatively simple asymmetric sulfur or nitrogen ligands, and selectivity is usually low. With simple olefins this is not surprising, since discrimination depends on rather small differences in steric bulk in the absence of polar interactions. 

The biggest challenge that remains in the field is control over the stereoselective formation of cis or trans olefins. Further experimental and theoretical mechanistic studies are needed to identify the stereodetermining steps of the reaction, and to discern the structures of important intermediates, such as the olefin complex B and... 

The origin of the remarkable stereoselectivities displayed by chiral homogeneous catalysts has occasioned much interest and speculation. It has been generally assumed, using a lock-and-key concept, that the major product enantiomer arose from a rigid preferred initial binding of the prochiral olefin with the chiral catalyst. Halpren 48) on the basis of considerable evidence, reached the opposite conclusion the predominant product enantiomer arises from the minor, less stable diastereomer of the olefin-catalyst adduct, which frequently does not accumulate in sufficient concentration to be detected. The predominant adduct is in essence a dead-end complex for it hydrogenates at a much slower rate than does the minor adduct. 

A related stereoselective route to glutamic acid derivatives consists of the addition of the nickel complex 5 to various activated olefins, i.e., 2-propenonitrile, 2-propenal and a,/ -unsaturated esters. 

Perhaps the most important chemical property of these complexes is their potential as catalysts, particularly of the early transition metal isoleptic compounds for a-olefin polymerization. This arises because unlike the methyls, they are sufficiently stable to be used at temperatures where polymerization rates are adequate. Some data are summarized in Table VIII ( 9) TT-acceptor ligands are clearly disadvantageous. It will be seen that some of the systems are more active than Ziegler types, although stereoselectivity is poorer. 

An enantioselective variant of the diene cydization reaction has been developed by application of chiral zirconocene derivatives, such as Brintzinger s catalyst (12) [10]. Mori and co-workers demonstrated that substituted dial-lylbenzylamine 25 could be cyclized to pyrrolidines 26 and 27 in a 2 1 ratio using chiral complex 12 in up to 79% yield with up to 95% ee (Eq. 4) [ 17,18]. This reaction was similarly applied to 2-substituted 1,6-dienes, which provided the analogous cyclopentane derivatives in up to 99% ee with similar diastereoselectivities [19]. When cyclic, internal olefins were used, spirocyclic compounds were isolated. The enantioselection in these reactions is thought to derive from either the ate or the transmetallation step. The stereoselectivity of this reaction has been extended to the selective reaction of enantiotopic olefin compounds to form bicyclic products such as 28, in 24% yield and 59% ee after deprotection (Eq. 5) [20]. 

A similar, although less marked difference characterizes the cyclopropanation of olefins 41 and 42. In the presence of either copper or copper complexes whose chelating ligands contain an azomethine moiety derived from an a-amino acid, no stereoselectivity was observed with diene 41, whereas the cyclopropanes derived from 42 occur with cisjtrans ratios of 57 43 to 69 31, depending on the catalyst93). 

Inhibition of diazoester decomposition by a large excess of olefin speaks in favor of intermediarily liberated W(CO)5 as direct metal precursor of425. Stereoselectivities in the cyclopropanation reaction are very similar to those observed in the Rh2(OAc)4 catalyzed version, which underlines once more the close relationship of tungsten and rhodium carbene complexes. 

---