The Preparation and Intramolecular Radical Cyclisation Reactions of Chiral Oxime Ethers

O éter-oxima quiral 2 e o éster-oxima 4 foram preparados para alquilação e esterificação da oxina 1. Hidroxiamina racêmica 6 e a hidroxiamina 10 foram sintetisadas a partir de N-hidroxipucanimida e o álcool correspondente na presença de di etil azodicarboxilato, e os dois produtos convertidos nos éteres-oximas 7 e 11 respectivamente. As reações de ciclização radicalar intramoleculares desses éteres e ésteres-oxima foram estudadas; a formação das alquil hidróxi-aminas 3, 8 e 12 foi observada nas reações bem sucedidas.


Introduction
The first example of the intramolecular addition of a radical to an oxime ether was published in 1983 by Corey and Pyne 1 .In this work a silyloxyalkyl radical was produced by the reaction of zinc and trimethyl silyl chloride on a carbonyl group which then attacks the oxime ether.The alkoxyaminyl radical which results from the cyclisation seems to have special stability which may be due to stabilisation of the nitrogen radical centre by the adjacent oxygen lone pair.Since this early work extensive studies have taken place in which alkyl radicals have been produced by reaction of a halide with tributyl tin hydride, or a related reductive method, which then undergo addition to the oxime ether usually, but not always in an intramolecular reaction 2 .In 1990 Enholm et al., reported the intramolecular additon of an alkenyl radical to an oxime ether, the sp 2 radical was generated by addition of Bu3SnH to an alkyne 3 .Prompted by these studies we have recently reported a series of examples of intramolecular radical cyclisation reactions of oxime ethers 4 .An unusual rearrangement was also observed when the alkoxyamine products of the cyclisation were reduced with LiAlH4 5 .The next stage in this project was to modify the reaction to produce chiral products, we now wish to report our studies on this topic.The idea was to use a chiral hydroxylamine to produce a chiral oxime ether RCH=NOR*, nucleophilic attack on these compounds has received some attention in the literature 6 , and has been the subject of a recent publication 7 .However, we know of no previous work on the addition of radicals to chiral oxime ethers of this type.the chiral oxime ether 2 in 82% yield.Treatment of 2 with Bu3SnH and AIBN gave alkoxyamine 3 in 78% yield as a 1:1 mixture of diastereoisomers as measured from the 13 Cand 1 H-NMR of the product.The lack of diasteroselectivity is not surprising in view of the fact that the oxime carbon is five bonds away from the chiral centre on the menthyl group.Four of these connecting bonds are single and so we would expect a wide variety of conformations to be present hence making effective chirality transfer more difficult.
It is well known that esters have a prefered conformation in which the acyl and alkoxy substituents are trans.We therefore attempted to restrict the conformational mobility of the system by preparing oxime ester by reaction of the oxime 1 with camphanic chloride and pyridine to produce the oxime ester 4 in 93% yield.Oxime esters have previously been used in Beckmann type rearrangements 10 , radical cyclisation reactions have not been reported previously for the compounds.When the oxime ester 4 was treated with Bu3SnH the starting material was consumed but no identifiable product was obtained.Clearly the chirality in oxime ethers 2 and 4 is too far away from the oxime carbon atom and the two formally diastereotopic faces of the oxime are not sufficiently different for selectivity to occur in the addition reaction.The closest we can get the stereogenic carbon to the oxime is to have it directly attatched to the nitrogen of the oxime, in order to achieve this objective we need to us a chiral oxime ether.
Oxime ethers are most conveniently prepared from alcohols and N-hydroxy succinimide following the method of Grochowski and Jurczac 11 .We therefore need to apply this method to chiral alcohols to produce chiral hydroxylamines.The first example in Scheme 2 is the reaction of racemic 1-phenyl ethanol with N-hydroxy succinimide to give the adduct 5 in 62% yield.The succinimide group is then removed by reaction with hydrazine to furnish the racemic 1-phenyl ethyl hydroxylamine 6 which was converted directly into the oxime ether 7. Succesful intramolecular radical cyclisation of oxime 7 was achieved with Bu3SnH and AIBN to produce the alkoxy amine 8 in 62% yield as a 1:1 mixture of diastereoisomers.
The other example in Scheme 2 is the reaction of (S)-naphthyl ethanol with N-hydroxy succinimide and diethyl azodicarboxylate to furnish the (R) adduct 9 in 57% yield where inversion of configuration has occured in the substitution reaction.The succinimide group was again removed with hydrazine to yield the chiral hydroxylamine 10 which was converted directly into the chiral oxime ether 11 in 93% yield.Treatment of this oxime with Bu3SnH and AIBN gave radical cyclisation in 68% yield to give the alkoxy amine 12 as a 1:1 mixture of diastereoisomers.
In conclusion chiral and racemic oxime ether derivatives have been prepared and subjected to intramolecular radical cyclisation reactions.This is the first study of this topic to the best of our knowledge.The lack of diastereoselectivity in the process was disappointing, however it is hoped that in the future asymmetric induction will be possible in this reaction.Elemental analysis was carried out by Butterworth Laboratories, Teddington, Middlesex.Infra-red spectra were recorded on a Perkin-Elmer 298 spectrophotometer.Melting points were determined on a Kofler hotstage and are uncorrected.
Light petroleum (b.p. 40-60 °C) and ethyl acetate were distilled prior to use.THF was distilled from sodium metal in the presence of benzophenone.

Reaction of 2-(2-Bromoallyloxy)benzaldehyde O-camphanic oxime 4 with tributyltin hydride and AIBN
A solution of AIBN (23 mg, 0.014 mmol) in benzene (10 cm 3 ) was added dropwise over 8 h to a solution of 2-(3-bromobut-3-en-1-oxy)benzaldehyde O-camphanic oxime 4 (300 mg, 0.68 mmol) and tributyltin hydride (240 mg, 0.83 mmol) in dry deoxygenated benzene (34.4 cm 3 , 0.02 M dilution of 4) heated at reflux temperature.Heating was continued for a further 5 h.Benzene was evaporated under reduced pressure but analysis of the crude product mixture revealed in excess of ten components that could not be isolated.All starting material was consumed.