propyl enolate 5a (84% de). The diastereoselectivities for all of
the enolate alkylation reactions investigated were calculated
from integration of the resonances corresponding to both the
major (4S,2AS)-diastereoisomers 6a–c and minor (4S,2AR)-
diastereoisomers 7a–c in the 1H NMR spectrum of each crude
reaction mixture. Authentic pure samples of both the major
(4S,2AS)-diastereoisomers 6a–c and minor (4S,2AR)-diastereo-
isomers 7a–c were obtained via chromatographic purification of
the three pairs of diastereoisomers 6a–c and 7a–c obtained from
N-acylation of the lithium anions of each of the parent
oxazolidin-2-ones with racemic 2-phenylpropanoyl chloride,
and fully characterised.
Scheme 3 Reagents and conditions: (i) LDA, MeI, THF, 230 °C.
studies were shown to be stable over the period of the
experiment via subsequent treatment of both solutions of
enolates 5a and 5c with MeI to afford the major diastereo-
isomers 6a and 6c with diastereoselectivites identical to those
obtained previously.
Having clearly demonstrated that methylation of the 5,5-di-
methyl-4-iso-propyl enolate 5c with MeI at 0 °C occurred in
higher de than methylation of the corresponding 4-iso-propyl
enolate 5a, 1H nOe NMR spectroscopic analysis on the enolates
5a and 5c was carried out in order to probe directly their
conformation in solution. Enolates 5a and 5c were generated via
treatment of 4a and 4c with 1 equiv. of LHMDS at 278 °C in
d8-THF, followed by warming the resulting solution to 0 °C.
The 1H NMR spectra of both 5a and 5c were entirely consistent
with the proposed (Z)-enolate structures since enolate 5a
showed a singlet resonance at d 4.58, while enolate 5c exhibited
a singlet resonance at d 4.68, corresponding to the C(2A) vinylic
proton of the enolate functionality in each case. Examination of
the resonances corresponding to the iso-propyl CH(Me)2
protons in the 1H NMR spectrum of enolates 5a and 5c revealed
small vicinal coupling constants of 3.2 and 2.3 Hz respectively
between the iso-propyl CH(Me)2 proton and the oxazolidin-
2-one H(4) proton. This is consistent with both enolates 5a and
5c adopting conformations in which the CH(Me)2 protons of
their iso-propyl groups lie approximately syn- or anti-periplanar
to the C(4)–C(5) bond of the oxazolidin-2-one ring, with both
methyl groups of their iso-propyl units directed either towards,
or away from the attached enolate fragment. Qualitative 1H nOe
NMR spectroscopic analysis of the 4-iso-propyl enolate 5a in
d8-THF at 0 °C revealed a strong enhancement between the
C(2A) vinylic proton of the enolate fragment and the oxazolidin-
2-one iso-propyl CH(Me)2 proton, with no nOe enhancement
with either of the iso-propyl CH(CH3)2 groups. Further strong
nOe enhancements were observed between the pro-(S) H(5)
proton and both of the iso-propyl CH(CH3)2 methyl groups.
These nOe enhancements are consistent with a major enolate
conformer 5a in which both of the iso-propyl methyl groups are
directed away from the attached enolate fragment (Fig. 2). In
In order to confirm the enhanced facial selectivity observed
for alkylation of N-acyl-5,5-dimethyl-4-iso-propyl enolates,
methylation of the enolate of N-butyryl-5,5-dimethyl-4-iso-
propyloxazolidin-2-one 2c was carried out under the original
conditions described by Evans et al. for N-butyryloxazolidin-
2-ones 2a and 2b. Thus, methylation of the enolate of 2c gave
the major diastereoisomer (4S,2AS)-3c in 96% de (Scheme 3),
which is significantly higher than that obtained for methylation
of the enolate of N-butyryl-4-iso-propyloxazolidin-2-one 3a of
82% de, and compares favourably with that observed for
methylation of the enolate of N-butyryl-4-tert-butyryloxazoli-
din-2-one 3b of 97% de.
1
In conclusion, H NMR nOe spectroscopic studies confirm
that the superior enolate alkylation diastereoselectivities ob-
served for (S)-5,5-dimethyl-4-iso-propyl enolate 5c over (S)-
4-iso-propyl enolate 5a is a result of steric interactions between
the C(5)-gem-dimethyl and the iso-propyl group of 5c which
direct both methyl groups of the iso-propyl stereocontrolling
group towards the enolate fragment. We believe that the
strategy outlined above of employing adjacent gem-dimethyl
groups to control the conformation of an iso-propyl group such
that it mimics a tert-butyl group is a general one, and we are
currently investigating the development of a wide range of
novel ligands and auxiliaries containing this structural motif,
which is easily prepared from valine, as replacements for the
corresponding tert-leucine derived analogues.7
We thank Dr Barbara Odell for her technical assistance with
the qualitative 1H nOe NMR spectroscopic studies on enolates
5a and 5c.
1
contrast, qualitative H nOe NMR spectroscopic analysis of
Notes and references
enolate 5c revealed both a medium and small enhancement
between the C(2A) vinylic proton of the enolate fragment and
each of the methyl groups of the iso-propyl CH(CH3)2 group. A
further strong nOe enhancement was observed between the
CH(Me)2 proton and one of the C(5)-gem-dimethyl groups.
These nOe enhancements are consistent with the proposed
model for 5,5-dimethyl-4-iso-propyl enolate 5c in which the
major conformer in solution has both methyl groups of the
stereocontrolling iso-propyl group directed towards the at-
tached enolate fragment (Fig. 3). Importantly, both samples of
1 D. A. Evans, Aldrichim. Acta, 1982, 15, 23; D. A. Evans, in Asymmetric
Synthesis, ed. J. D. Morrison, 1984, vol. 3, Academic Press, New York;
D. A. Evans, T. C. Britton and J. A. Ellman, Tetrahedron Lett., 1987, 28,
6141.
2 D. A. Evans, K. T. Chapman, D. T. Hung and A. T. Kawaguchi, Angew.
Chem., Int. Ed. Engl., 1987, 26, 1184.
3 L-tert-Leucine is commercially available at £16.50 per gram while the
corresponding
Company.
D enantiomer is £101.60 per gram from Aldrich Chemical
4 Energy minimisation on enolate 5c was carried out using MOPAC with
a MM2 forcefield calculation.
1
enolates 5a and 5c in d8-THF at 0 °C used in these H NMR
5 S. G. Davies and H. J. Sanganee, Tetrahedron: Asymmetry, 1995, 6, 671;
S. D. Bull, S. G. Davies, S. Jones, M. E. C. Polywka, R. S. Prasad and
H. J. Sanganee, Synlett, 1998, 519; S. D. Bull, S. G. Davies, S. Jones and
H. J. Sanganee, J. Chem. Soc., Perkin Trans 1, 1999, 387; S. G. Davies,
H. J. Sanganee and P. Szolcsanyi, Tetrahedron, 1999, 55, 3337.
6 Similar arguments have been proposed to explain the high facial
selectivity observed for a 2,2-dimethyloxazolidine derived chiral auxil-
iary in controlling (a) the [3 + 2] cycloaddition of dipolarophiles with
alkenes, K. Onimura and S. Kanemasa, Tetrahedron, 1992, 48, 8631; (b)
the [4 + 2] cycloaddition of dienes with singlet oxygen, W. Adam, M.
Güthlein, E.-M. Peters, K. Peters and T. Wirth, J. Am. Chem. Soc., 1998,
120, 4091; (c) and a 5,5-diphenyloxazolidin-2-one for enolate alkyla-
tions, T. Hintermann and D. Seebach, Helv. Chim. Acta, 1998, 81,
2093.
Fig. 2 nOe study on Evans’ enolate 5a. Other nOe enhancements omitted for
clarity.
7 For a review on the use of the advantages of employing homochiral tert-
leucine in asymmetric synthesis see A. S. Bommarius, M. Schwarm, K.
Stingl, M. Kottenhahn, K. Huthmacher and K. Drauz, Tetrahedron:
Asymmetry, 1995, 12, 2851.
Fig. 3 nOe study on SuperQuat enolate 5c. Other nOe enhancements
omitted for clarity.
1722
Chem. Commun., 2000, 1721–1722