New symmetry-breaking deprotonation reactions of cyclic imides using a chiral lithium amide base

Article abstract of DOI:10.1039/a803574a

Treatment of various N-substituted succinimides, possessing additional three-, four- or five-membered ring-fusion, with a mixture of chiral lithium amide base and Me₃SiCl, enables enantioselective silylation to give products in up to 95% ee.

Full text of DOI:10.1039/a803574a

View Article Online / Journal Homepage / Table of Contents for this issue
New symmetry-breaking deprotonation reactions of cyclic imides using a chiral
lithium amide base
David J. Adams, Nigel S. Simpkins*† and Torben J. N. Smith
Department of Chemistry, The University of Nottingham, University Park, Nottingham, UK NG7 2RD
Treatment of various N-substituted succinimides, possessing
additional three-, four- or five-membered ring-fusion, with a
Ph
N
Li
4
Ph
H
H
H
SiMe3
O
mixture of chiral lithium amide base and Me3SiCl, enables
enantioselective silylation to give products in up to 95%
ee.
LiCl, Me3SiCl
ca. –100 °C
O
O
O
N
Ph
N
Ph
5 80% (95% ee)
3
During the past few years there have been great developments
in the application of chiral lithium amide base reactions to
asymmetric synthesis.1 Most of the enantioselective deprotona-
tion reactions of these chiral bases described to date involve
discrimination between enantiotopic hydrogens activated by a
single common functional group. The best developed chemistry
of this type is the asymmetric enolisation of cyclic ketones,
although somewhat analogous situations involving metallation
of sulfoxides and prochiral organometallics have also been
described.2
We considered that a considerable broadening in the scope
for applications of chiral lithium amides would be possible if
multifunctional substrates could be employed. In a general
sense this could involve desymmetrisation of a prochiral
molecule having a number of functional groups deployed on a
suitable conformationally constrained template (most likely a
carbocyclic or saturated heterocyclic or polycyclic system). In
this type of chiral base reaction the key step would involve
kinetically controlled discrimination between hydrogens which
are activated by separate functional groups. For a situation in
which just two functional groups are involved (as the activating
groups for two enantiotopic methine hydrogens), the trans-
formation can be illustrated as the conversion of 1 into 2
(Scheme 1).3
Scheme 2
Imide 5 proved to be crystalline, which enabled us to
determine the absolute configuration shown by single crystal
X-ray structure determination.§ Our delight at this excellent
result was tempered by the observation that only the use of
Me3SiCl as an in situ electrophile appears to enable clean
substitution, with the attempted reaction with alkylating agents
such as MeI resulting only in the destruction of starting
material. Although this is a drawback, we were able to
circumvent this apparent limitation by substitution of the
remaining acidic hydrogen by conventional means (Table 1).
Table 1 Substitution of 5 using LDA
E
SiMe3
O
LDA–LiCl, THF, –78 °C
electrophile
5
O
N
Ph
6
Compound
Electrophile
Yield of 6 (%)
6a
6b
6c
6d
6e
6f
MeI
AllylBr
BnBr
PhSSPh
PhCHOa
PhCOCl
93
73
67
71
61
57
R
R
H
X
H
X
H
X
E
X
chiral lithium amide
electrophile - E +
*
*
(
*
n
(
)
n
)
a Formed as a ca. 1:1 ratio of diastereoisomers.
1
2
Scheme 1
Thus, introduction of a range of different substituents was
possible in good to excellent yield by use of LDA–LiCl as the
base.5 In the case of 6a we have confirmed that removal of the
silicon group is possible, to give the expected methylated imide.
As an alternative approach to the problem we have found that
fluoride-mediated substitution of the Me3Si group of 5 is
possible by a narrow range of electrophiles (PhCHO, PhSSPh
and PhCOF), to give directly the imide products 7 (Scheme 3).6
We expect that no loss of ee takes place in this process, although
this has not been confirmed to date.
The two activating groups X could be typical carbonyl types,
including ketone, ester or amide, or other types such as sulfone.
These two functions could be close together or somewhat
remote (variation in n) and the overall framework could bear
substituents R as long as the overall symmetry is retained in
substrate 1. As can be seen, the chiral base transformation,
involving selective substitution at one activated position would
be expected to generate chiral products having a number of
asymmetric centres (marked *).
Herein we describe our initial results in this new area of chiral
lithium amide base chemistry in which we have examined the
desymmetrisation of a number of ring–fused imides, i.e. n = 0
and X = amide (X groups linked to form a five-membered ring
imide).
Initial work showed that treatment of cyclopropane deriva-
tive 3 with chiral base 4 at low temperature in the presence of
Me3SiCl resulted in the formation of the desired product 5 in
80% yield and 95% ee (Scheme 2).4‡
H
E
CsF, 18-C-6, THF
electrophile
5
O
O
N
Ph
7a (68%) (1:1 dr)
7b (61%)
PhCHO
PhSSPh
PhCOF
7c (44%)
Scheme 3
Chem. Commun., 1998
1605

View Article Online
Lawesson's
reagent
Notes and References
H
SiMe3
O
† E-mail: nigel.simpkins@nottingham.ac.uk
toluene
reflux
S
N
‡ In a typical asymmetric silylation, a solution of the chiral lithium amide
base 4 mixed with LiCl was added dropwise to a mixture of starting imide
and Me3SiCl (excess) in THF at ca. 2100 °C (internal temperature). The
mixture was then allowed to warm slowly (4 h) to ambient temperature
before standard aqueous work-up and chromatography on silica gel. All
products have been fully characterised by spectroscopic methods and give
satisfactory elemental analysis and/or HRMS results. Enantiomeric excess
values were established by HPLC (UV detection) using PriOH–hexane as
eluent, using a Chiralcel OD column for 5, 11 and 12, and a Chiralcel OJ
column for 10.
Ph
8 (50%)
SiMe3
5
H
DIBAL
CH2Cl2, –78 °C
HO
O
N
Ph
9 (70%)
Scheme 4
§ The absolute configuration of 5 was established by the collection of low
temperature data, including Friedel equivalents, and by refinement of a
Flack parameter, value 0.05(19), see H. D. Flack, Acta Crystallogr., Sect. A,
1983, 39, 876. The absolute stereochemistry of product 12 was likewise
established; Flack parameter 20.12(14). We thank Dr A. J. Blake of this
Department for these determinations; full details will be published
elsewhere.
Since one of the main objectives of this new chiral base
chemistry is to enable the synthesis of varied enantiomerically
enriched products, it was important to establish if the initial
enantioselective substitution to give 5 enabled subsequent
regiocontrolled reactions of the imide. In line with our
expectations, imide 5 undergoes highly regioselective thiona-
tion and reduction reactions, to give 8 and 9 respectively
(Scheme 4).7,8
Taken in combination with the highly enantioselective access
to compounds of type 6 and 7, this chemistry begins to indicate
some of the potential of our new approach for the synthesis of
lactams, lactones, pyrrolidines, etc.
¶ These reactions have not been optimised; the absolute configurations
shown for 10 and 11 are based on analogy with 5 and 12.
1 For reviews, see P. O’Brien, J. Chem. Soc., Perkin Trans. 1, 1998, 1439;
P. J. Cox and N. S. Simpkins, Tetrahedron: Asymmetry, 1991, 2, 1; N. S.
Simpkins, Advances in Asymmetric Synthesis, ed. G. R. Stephenson,
Blackie Academic, 1996; K. Koga, Pure Appl. Chem., 1994, 66, 1487.
2 For leading references in these two areas, see A. J. Blake, S. M.
Westaway and N. S. Simpkins, Synlett, 1997, 919 (sulfoxides); R. A.
Ewin, D. A. Price, N. S. Simpkins, A. M. MacLeod and A. P. Watt,
J. Chem. Soc., Perkin Trans. 1, 1997, 401 (organometallics).
3 We have recently described an enantioselective elimination reaction akin
to the process outlined in Scheme 1, see C. D. Jones, N. S. Simpkins and
G. M. P. Giblin, Tetrahedron Lett., 1998, 39, 1023.
Finally, we were able to extend this new asymmetric process
to several other imide systems, the silylated products 10–12
4 The starting imide 5 was prepared via reaction of the corresponding
anhydride (J. J. Tufariello, A. S. Milowsky, M. Al-Nuri and S. Goldstein,
Tetrahedron Lett., 1987, 28, 267) with aniline, see A. L. J. Beckwith and
D. R. Boate, J. Org. Chem., 1988, 53, 4339. See also A. Mustafa,
S. M. A. D. Zayed and S. Khattab, J. Am. Chem. Soc., 1956, 78, 145 for
an alternative synthesis, and ref. 9.
H
SiMe3
O
H
SiMe3
O
H
SiMe3
O
O
O
O
N
Bn
N
Bn
N
Ph
5 In the past we have observed increased rates of metallation by addition of
LiCl to lithium amides, see for example D. A. Price, N. S. Simpkins,
A. M. MacLeod and A. P. Watt, Tetrahedron Lett., 1994, 35, 6159; B. J.
Bunn, N. S. Simpkins, Z. Spavold and M. J. Crimmin, J. Chem. Soc.,
Perkin Trans. 1, 1993, 3113. Recently, little effect of LiCl on the rate of
deprotonation of a simple ketone by LDA was observed, see M.
Majewski and P. Nowak, Tetrahedron Lett., 1998, 39, 1661.
6 These reactions were modelled on analogous substitutions of episulfones
that we examined previously, see A. P. Dishington, R. E. Douthwaite, A.
Mortlock, A. B. Muccioli and N. S. Simpkins, J. Chem. Soc., Perkin
Trans. 1, 1997, 323.
10 (37%) (94% ee)
11 (72%) (91% ee)
12 (65%) (93% ee)
being isolated in excellent levels of ee from reaction of the
appropriate symmetrical imide starting material with a mixture
of 4 and Me3SiCl at low temperature.9¶
These results show that the concept outlined in Scheme 1 can
be realised, and the products clearly have some potential for
target synthesis. However, these asymmetric transformations of
ring-fused imides form but a small sub-group of the chemistry
implicit in Scheme 1, and efforts to realise further examples of
this diverse group of reactions are underway.
7 M. J. Milewska, M. Gdaniec and T. M. Polonski, J. Org. Chem., 1997, 62,
1860.
8 T. Mukaiyama, H. Yamashita and M. Asami, Chem. Lett., 1983, 385.
9 For details of the preparation and enantioselective reduction of these
types of meso-imide, see M. Ostendorf, R. Romagnoli, I. C. Pereiro, E. C.
Roos, M. J. Moolenaar and W. N. Speckamp and H. Hiemstra,
Tetrahedron: Asymmetry, 1997, 8, 1773.
We are grateful to the EPSRC and Chiroscience Limited
(Cambridge Science Park, Milton Road, Cambridge, UK
CB4 4WE) for support of T. J. N. S under the CASE scheme.
We are also very grateful to Professor H. Hiemstra of the
University of Amsterdam for samples of certain ring-fused
imides.
Received in Liverpool, UK, 12th May 1998; 8/03574A
1606
Chem. Commun., 1998