Enolate anions conformation

The proportion of the /rans-O-alkylated product [101] increases in the order no ligand < 18-crown-6 < [2.2.2]-cryptand. This difference was attributed to the fact that the enolate anion in a crown-ether complex is still capable of interacting with the cation, which stabilizes conformation [96]. For the cryptate, however, cation-anion interactions are less likely and electrostatic repulsion will force the anion to adopt conformation [99], which is the same as that of the free anion in DMSO. This explanation was substantiated by the fact that the anion was found to have structure [96] in the solid state of the potassium acetoacetate complex of 18-crown-6 (Cambillau et al., 1978). Using 23Na NMR, Cornelis et al. (1978) have recently concluded that the active nucleophilic species is the ion pair formed between 18-crown-6 and sodium ethyl acetoacetate, in which Na+ is co-ordinated to both the anion and the ligand. 

The stereoselectivity of the trifluoromethylation varied with the bulkiness of the boron reagent used. Thus, with enolate anion 64, the product ratio of the thermodynamically less stable )8-CF3 isomer 66 versus the more stable a-CF3 isomer 65 increased with the bulkiness in the order 59 < 60 < 61 (Eq. 37). This was explained by the conformation of the intermediate complexes. 

Substitution of a carbon monoxide ligand of complexes, such as 1, by the more electron-donating triphenylphosphane group (see Section 1.1.1.3.4.1.3.) provides chiral monophos-phane complexes, such as 3. Monophosphane complexes in general lack sufficient electrophilic-ity to react with amines or thiols, but react readily with amine anions at the /J-position, producing enolate anions such as 4, which may be quenched stereoselectively at the a-carbon by electrophiles46 (see Section 1.1.1.3.4.1.3.). The conformational and stereochemical issues involved are essentially identical to those already discussed in this section for the 1,4-additions of carbon nucleophiles. 

Whatever the explanation, the effect of acids is less marked than the selectivffy in alkaline solutions, where a attack is largely suppressed. The effect of alkali may depend upon the formation and selective reduction of enolate anions. The A2 4-dienolate anion, which is the major product of kinetically-controlled enolisation by bases (see p. 156) is seen from a molecular model to have a somewhat "folded conformation of the A/B ring system (ii). The convex / -face of the A/B ring system and the absence of an axial 2jS-proton should favour approach to the catalyst from this direction, whereas the a-face of the A -bond is severely hindered by the axial hydrogens at C(7) and C<9>. 

A similar conformational analysis has been done with formamide derivatives, with secondary amides, and for hydroxamide acids. It is known that thioformamide has a larger rotational barrier than formamide, which can be explained by a traditional picture of amide resonance that is more appropriate for the thioformamide than formamide itself. Torsional barriers in a-keto amides have been reported, and the C—N bond of acetamides, thioa-mides, enamides carbamates (R2N—C02R), and enolate anions derived... 

The preference for the lowest energy conformation of the enolate anion is seen in larger ring systems as well (sec. 1.5.B,C), leading to good selectivity in alkylation and condensation reactions. The methyl group provides only small steric encumbrance to approach of the electrophile in enolate 505 (derived from lactone 504 and LDA). The preferred mode of attack for this relatively stable conformation was from the top face (path a, pseudoequatorial attack) and gave the syn diastereomer (506) with >99 1 selectivity.- ... 

The argument for selectivity in the enolate displacement is probably correct but the structure shown does not indicate this. The enolate anion could, in principle, react with the bromine or the acetate, if they are properly oriented. This depends of the conformational preferences of the molecules. Some attempt to show the three dimensional representation of the molecule must be made in order to address these points. 

The erythro isomer cannot eliminate entirely from an anti conformation because only the cis olefin would be produced. Of the two hydrogen atoms which could be abstracted by a base, the more acidic atom is removed and the elimination probably proceeds through an enolate anion (32). These anions however, have the same configuration as their precursors as isomerisation of the erythro to threo substrate before elimination is not observed, presumably due to steric interactions which cause rotation about the 2-3 carbon-carbon bond to be a slower process than proton capture from the solvent. To minimise steric interactions in the erythro dichloride, the bulky aryl residues will separate to a maximum by occupying the "anti positions (32). 

---