Enamide complexes

Fig. 12.13 Simultaneous -PHIP-NMR spectra of dihydrides from enamide complexes of the Dl PHOS-containing Rh-catalyst.

In the presence of the dehydroamino acid reactant, bidentate complexation as an enamide occurs. With a chiral diphosphine ligand, two diastereomeric forms of the enamide complex are observed in equilibrium that differ in the... 

At low temperatures, only the disfavored enamide complex reacts with dihy-drogen, forming an alkylhydride complex that decomposes to form the hydrogenation product above -50 °C. No evidence for an intermediate dihydride could be established, although its involvement was assumed. All of the described intermediates give well-defined and distinctive NMR spectra in which 1H, 31P and 13C (with enrichment) are all informative. 

Dihydrogen addition to the enamide complex is rate-limiting and irreversible. With para-enriched hydrogen, there is no ortho-para equilibration in a dehydroamino acid turnover system until hydrogenation is complete [18] (this last precept has come under recent close scrutiny). 

The status quo provided by the early mechanistic studies described above was incomplete in several respects. Although the affinity of the solvate complex for dihydrogen was known to be low, the addition product remained uncharacterized. Likewise, the putative H2 addition intermediate between the enamide complex and the transient, but observable, alkylhydride had not been characterized. 

Fig. 31.12 (a) X-ray structure of the enamide complex that corresponds to the correct" hand of product (from [58]). Solvent hydrogens and counterion are omitted for clarity, (b) Structure of the cation in (a), (c) Application of the Quadrant Rule by these authors. 

The problems which remain in understanding and interpreting rhodium asymmetric hydrogenation arise from a persistent lack of information on the presumed rhodium dihydride without which the pathway between the enamide complex and the turnover limiting TS for H2 addition (i.e. the step in which the enantioselectivity of the reaction is set) remains opaque, and hence the overall understanding is elusive. [52]... 

The difference between this catalytic system and Wilkinson s catalyst lies in the sequence of the oxidative addition and the alkene complexation. As mentioned above, for the cationic catalysts the intermediate alkene (enamide) complex has been spectroscopically observed. Subsequently oxidative addition of H2 and insertion of the alkene occurs, followed by reductive elimination of the hydrogenation product. 

The dynamic behavior of the model intermediate rhodium-phosphine 99, for the asymmetric hydrogenation of dimethyl itaconate by cationic rhodium complexes, has been studied by variable temperature NMR LSA [167]. The line shape analysis provides rates of exchange and activation parameters in favor of an intermo-lecular process, in agreement with the mechanism already described for bis(pho-sphinite) chelates by Brown and coworkers [168], These authors describe a dynamic behavior where two diastereoisomeric enamide complexes exchange via olefin dissociation, subsequent rotation about the N-C(olefinic) bond and recoordination. These studies provide insight into the electronic and steric factors that affect the activity and stereoselectivity for the asymmetric hydrogenation of amino acid precursors. 

Brown and coworkers [128] have studied the exchange process for the enamide complex 103, using magnetization transfer techniques (Figure 1.28). Compound 103 represents the catalytic resting state in the asymmetric homogeneous hydroge-... 

Unlike the Rh-based hydrogenation of a-(acylamino)acrylates, the corresponding Ru chemistry has not been studied extensively. Ru complexes of (S)-BINAP and (S,S)-CHIRAPHOS catalyze the hydrogenation of (Z)-a-(acylamino)cinnamates to give the protected ( -phenylalanine with 92% ee [74] and 97% ee [75], respectively. It is interesting that the Rh and Ru complexes with the same chiral diphosphines exhibit an opposite sense of asymmetric induction (Scheme 1.6) [13,15,56,74,75]. This condition is due primarily to the difference in the mechanisms the Rh-catalyzed hydrogenation proceeds via Rh dihydride species [76], whereas the Ru-catalyzed reaction takes place via Ru monohydride intermediate [77]. The Rh-catalyzed reaction has been studied in more detail by kinetic measurement [78], isotope tracer experiments [79], NMR studies [80], and MO calculations [81]. The stereochemical outcome is understandable by considering the thermodynamic stability and reactivity of the catalyst-enamide complexes. 

The sample then was cooled to -80°C and sealed under hydrogen. It was warmed to -50°C in the probe, and the P-31 NMR spectrum was monitored at this temperature. The minor diastereomer (10b) disappeared concomitant with the formation of Complex 11, while the concentration of Complex 10a remained constant within experimental error. In a second experiment formation of the enamide complex was effected at -20°C, at which temperature the equilibration of Complexes 10a and 10b is rapid and cooling the sample to -50°C provides a spectrum from which the latter is effectively absent. When this solution is blanketed with hydrogen, little observable change occurs in the... 

At this point mechanistic studies have reached an impasse. All of the observable intermediates have been characterized in solution, and enamide complexes derived from diphos and chiraphos have been defined by X-ray structure analysis. Based on limited NMR and X-ray evidence it appears that the preferred configuration of an enamide complex has the olefin face bonded to rhodium that is opposite to the one to which hydrogen is transferred. There are now four crystal structures of chiral biphosphine rhodium diolefin complexes, and consideration of these leads to a prediction of the direction of hydrogenation. The crux of the argument is that nonbonded interactions between pairs of prochiral phenyl rings and the substrate determine the optical yield and that X-ray structures reveal a systematic relationship between P-phenyl orientation and product configuration. 

The optimized structures of the diastereomeric catalyst-enamide complexes with the enamide bonded from its pro-R and pro-S faces have very similar structures to their nitrile counterparts. The calculated free energy difference between these two diastereomers is practically zero (0.07 kcal/ mol). 

Alcock, N. W. Brown, J. M. Maddox, P. J. Chem. Commun. 1986, 1532. Reaction between resolved iridium enamide complexes and racemic chiraphos mixture is highly enantioselective and permits in situ resolution for use in asymmetric catalysis. 

By the late 1970s the structures of the precursor complexes in asymmetric hydrogenation were known in principle. Kagan had speculated on the possible intervention of an enamide complex (alkene and amide carbonyl group bound)... 

A further step was taken when first Halpern [28] and then Brown [29] were able to identify a further intermediate, the rhodium alkyl hydride formed by addition of dihydrogen to the enamide complex with transfer of a single hydride to the benzylic carbon. For the simple dppe complex studied by Halpern, the interpretation of the experiment was straightforward, but the intermediate derived from DIPAMP by Brown and Chaloner provided a major surprise only the disfavored minor diastereomer of the enamide complex was reactive towards H2. The major/minor equilibrium is so strongly biased towards the former below -50 °C that reaction with H2 is undetected. Only when the solvate complex is allowed to react with the dehydroamino acid derivative under H2, well below -50 °C (under which conditions up to 35% of the minor diastereomer is initially observed) is the alkyl hydride observed, concomitant with disappearance of that minor diastereomer. This reactive intermediate was characterized by its H-NMR (hydride), the distinctive P-NMR and by both heteronuclear coupling and chemical shifts in the C-NMR spectra of alkyl hydrides derived from singly and doubly labeled dehydroamino esters. 

The corresponding iridium enamide complexes and their alkyl hydride counterparts are much more stable, and a full NMR characterization of the alkyl hydride proved possible in the DIPAMP series. Here, as in the corresponding rhodium chemistry, the presumed dihydride precursor proved to be elusive [31]. By employing a different approach to enamide complexes in which an iridium bis-enamide complex was allowed to react with the diphosphine (Fig. 6) both major and minor enamide complexes could be prepared separately the path to one of them is shown in Fig. 6. The trick was to employ menthyl esters so that stereo-chemically homogeneous Ir complexes were formed. Some additional structural features of the intermediates were derived from detailed NMR analysis, and especially the role of the OMe group in coordinating to iridium trans to the hydride [32]. 

Two separate diastereomers Ph of enamide complex separated by crystallisation. E = R-1-menthyl ester. 

Preferred diastereomer of enamide complex leading to R-enantiomer of product... 

Fig. 11 The quadrant approach and rationalization of the relative energies of the two diastere-omeric enamide complexes...

It was subsequently demonstrated that enamides displace solvent from this adduct, giving new species which are air-sensitive and highly reactive towards hydrogen. These enamide complexes have been characterized spectroscopically (15), and one of the more informative experiments was carried out with the asymmetric ligand DIPAMP and methyl Z-a-benzamidocinnamate (Figure 1). This shows that two diastereomeric enamide complexes are formed in a ratio of 10 1 at room temperature. The two species are related by binding of opposite prochiral faces of... 

---