The generation of single enantiomers is vitally important in the synthesis of
natural products and synthetic analogues with interesting biological properties;
different enantiomers can exhibit totally different biological activity. There are many cyclopropanes that occur naturally as single enatiomers and show key biological properties. The aim of this project was to use readily available cyclopropane containing starting materials and their derivatives to prepare such single enantiomers to complement the use of starting materials from the chiral pool and from enzymatic resolutions and desyrnmetrisations.
Three methods were examined.
Polymer chemistry.
Attempts were made to couple cyclopropane-1,2-dicarboxylic acid anhydride
and trans-cyclopropane-1,2-dicarbonyl chloride to a solid support with a view to using polymer bound chiral auxiliaries to desymmetrise these compounds. Succinic anhydride and succinoyl dicarbonyl chloride were used as model reagents. The acid chloride route allowed the mono-protection of succinoyl dicarbonyl chloride followed by reaction with an amide, cleavage and solution phase esterification to yield amido esters, although poor yields were obtained. Procedures for the monitoring of polymer bound reactions by mass were developed leading to a clear indication of the loading of resins obtained and the completion of reaction. Concern over the low yields and slow progress meant that this strategy was abandoned in favour of a solution based approach to desymmetrising these compounds before work progressed to the application of polymer bound chiral functionalities.
Chiral bicyclic imides.
Reaction of cyclopropane-1,2-dicarboxylic acid anhydride with a chiral
amine afforded a diastereomeric mixture of the corresponding a.mi do acids, which
when refluxed in acetic anhydride cyclised to a single enantiomer of the bi cyclic
imide. Deprotonation of these compounds with LDA in the presence ofTMSCl led
to mildly diastereoselective reaction to the mono silylated product as well as a single enantiomer of the di silylated species. The best results were obtained with 3-(2-hydroxy-l-isopropyl-ethyl)-3-aza-bicyclo[3. l .0]hexane-2,4-dione which was prepared in 3 steps from the anhydride and valinol. 3-(2-Trimethylsilanyl-l -isopropyl-ethyl)-l-trimethylsilanyl-3-aza-bicyclo[3 .1.0]hexane-2,4-dione was
obtained with a 6 : 10 diasteriomeric ratio, by reaction of this imide. Co-ordination oflithium species to the hydroxyl functionality is thought to be responsible for the degree of selectivity obtained as both 3-(2-Trimethylsilanyl-1-isopropyl-ethyl)-3-azabicyclo[3 .1. 0]hexane-2,4-dione and acetic acid 2-(2,4-dioxo-3-azabicyclo[3. 1.0]hex-3-yl)-2-isopropyl-ethyl ester exhibited a lower degree of diastierioselectivity when silylated. The modest selectivity of these reactions and concern over the general low yields, both in the synthesis and reaction of these compounds, caused the focus of research to again be shifted.
Mannitol derived chiral cyclopropane esters.
(1 R,2R,4'S)-2-[(tert-Butyldiphenylsilyloxy)methyl]-1-(2',2'-dimethyl-1 ',3'-
dioxolan-4'-yl )cyclopropane and ( 1 R,2S,4'S)-2-[ ( tert-butyldi pheny lsi lyloxy )methyl]-1 -(2' ,2'-dimethyl-1 ',3 '-dioxo lan-4'-yl)cyclopropane can be prepared in 6 steps from Mannitol with very high selectivity. From these, novel chiral esters (2R,3S)-4-(tertbutyldiphenylsilyloxy)-2,3-methanobutanoic acid methyl ester and (2R,3R)-4-(tertbutyldiphenylsilyloxy)-2,3-methanobutanoic acid methyl ester were prepared in 3 steps with a view to alkylating them to produce single enantiomers of novel trisubstituted cyclopropanes. However attempts to deprotonate them with LDA and quench with TMSCl or D20 failed with both molecules.