Stereoselective substituted pyrrolidine and cyclic ether synthesis by PhS migration

Article abstract of DOI:10.1039/b303790h

The effect of substitution on heterocycle synthesis by a novel silica gel catalysed decarboxylative ring-closing reaction is shown to be stereospecific and dependent on the nature of the nitrogen substituent.

Full text of DOI:10.1039/b303790h

Stereoselective substituted pyrrolidine and cyclic ether synthesis by PhS
migration
Lorenzo Caggiano, John Davies, David J. Fox, David C. Moody and Stuart Warren*
University Chemistry Laboratory, Lensfield Road, Cambridge, UK CB2 1EW. E-mail: sw134@cam.ac.uk
Received (in Cambridge, UK) 4th April 2003, Accepted 30th May 2003
First published as an Advance Article on the web 12th June 2003
The effect of substitution on heterocycle synthesis by a novel
silica gel catalysed decarboxylative ring-closing reaction is
shown to be stereospecific and dependent on the nature of
the nitrogen substituent.
The novel silica gel catalysed ring-closing reaction recently
reported allows for the synthesis of nitrogen and oxygen
heterocycles by the intramolecular trapping of episulfonium
ions in neutral conditions (Scheme 1).1 We wished to extend
this methodology to the stereoselective synthesis of more highly
substituted heterocycles with a view to the synthesis of useful
compounds and as an aid to the understanding of the scope and
mechanism of the reaction. The strong acid catalysed sulfur
migration used in cyclic ether synthesis is a stereospecific
process,2 but the novel silica gel catalysed reaction occurs in
markedly different reaction conditions, and might not occur via
a stereochemically stable episulfonium ion. To this end, a pair
of complementary diastereoisomers of a cyclic carbamate was
synthesised (Schemes 2 and 3).
Scheme 3 Reagents and conditions: i, LDA, THF, 278 °C; 1, 15% (7), 57%
(8); ii, KOH, H2O, THF; iii, DCC, NHS, THF; NH3, 72% (2 steps); iv, BH3,
THF, D, 83%; v, CDI, CH3CN, D, 73%; vi, SiO2, CHCl3, D.
chloroform to produce the syn-pyrrolidine 10 in good yield,
along with a small quantity of allylic sulfide 11 (Scheme 3).
In this case the ring-closing transition state is destabilised by
a 1,3-diaxial relationship between the methyl and the cyclohex-
ane ring, therefore the competing elimination is more sig-
Scheme 1
nificant than in the anti-diastereomer. The conformational lock
A nitrile-aldol reaction with aldehyde3 1 produced an excess
in the cyclohexane ring tends to maximise this elimination,
of the anti-diastereoisomer 3 (anti-3 : syn-2 3 : 1) which could
be reduced and cyclised with carbonyldiimidazole (CDI) to give
the carbamate 4 as a single diastereoisomer (Scheme 2). The
stereochemistry was confirmed by X-ray crystallography (Fig.
1).† Treatment of the carbamate 4 with silica gel in refluxing
chloroform produced pyrrolidine 5 in excellent yield as a single
diastereomer as the only product (Scheme 2), the configuration
of which was also confirmed by X-ray crystallography of its
3,5-dinitrobenzamide 6 (Fig. 1).
compared to related compounds.2 The formation of allylic
sulfide does not occur in the reaction of the gem-dimethyl
carbamate 12a, where the ring closure is favoured by the gem-
dialkyl effect4 although a syn-relationship of substituents on the
ring is also produced (Scheme 4). A similar cyclisation via an
The syn-diastereoisomer of cyclic carbamate 9 was made via
ester 8, from a more syn-selective ester aldol reaction, followed
by amide coupling, reduction and cyclisation. This syn-
carbamate was also treated with silica gel in refluxing
Scheme 2 Reagents and conditions: i, LDA, THF, 278 °C; 1, 15% (2), 56%
(3); ii, LiAlH4, Et2O, 89%; iii, CDI, CH3CN, 63%; iv, SiO2, CHCl3, D,
97%.
Fig. 1 Molecular structure of anti-carbamate 4 and the 3,5-dinitrobenzamide
(6) of anti-pyrrolidine 5 determined by X-ray analysis.
1648
CHEM. COMMUN., 2003, 1648–1649
This journal is © The Royal Society of Chemistry 2003

Scheme 4 Reagents and conditions: i, LDA, THF, 278 °C, 1; 78%; ii,
LiAlH4, Et2O, 98%; iii, CDI, CH3CN, 89%; iv, SiO2, CHCl3, D, 95%.
Scheme 7 Reagents and conditions: i, NaH, MeI or BnBr or iPrI, DMF; ii,
SiO2, CHCl3, D, see Table 1.
iodonium ion also favours 3,4-anti-substitution, resulting in the
formation of a single diastereoisomer.5
Cyclic carbonates also react in the presence of silica gel with
loss of carbon dioxide to form cyclic ethers.1 The gem-dimethyl
substituted cyclic carbonate 14 (Scheme 5) was much more
labile than carbonate 15 and the decarboxylation and cyclisation
were initiated by silica gel at ambient temperature during
purification. It is not clear why this substitution should increase
the reactivity of the carbonate intermediates.
Table 1 Yields for the alkylation of 12a and silica catalysed reaction
Allylic
Yield (%)
from 12a
Pyrrolidine
yield (%)
sulfide
yield (%)
Carbamate
R
12a
12b
12c
12d
H
CH3
CH2Ph
CH(CH3)2
—
94
96
91
> 95
77
54
< 5
14
30
88
The seven-membered carbonate 17, made from diol6 16, also
gave an ether (Scheme 6), but the higher temperature required
for seven-membered-ring carbonate formation provided harsh
enough conditions to initiate ether-ring closure and decarbox-
ylation. The only cyclic ether isolated from the reaction was
however the kinetically favoured unrearranged THF 18, not the
more stable rearranged THP 19. This route provides a
complementary method of forming THFs from 1,4-diols,
compared to the strong acid catalysed route that produces the
more thermodynamically stable THP.2,7 The similar iodonium
ion mediated reaction produces a kinetically controlled mixture
of the five-membered (13%) and six-membered (87%) cyclic
ethers.8
Substitution at the nitrogen in the pyrrolidine synthesis was
also investigated, and a range of N-alkyl carbamates 12b–d was
made by alkylation of the parent carbamate 12a (Scheme 7).1
Treatment of the carbamates with silica gel in refluxing
chloroform produced a range of results in which an increase in
the steric requirement of the N-alkyl substituent led to a
decrease in the amount of pyrrolidine 13a–d produced and an
increase in the proportion of allylic sulfide 20a–d formed by
loss of a proton from the episulfonium ion intermediate (Table
< 1
1). When the alkyl group is isopropyl the ring closure is
disfavoured enough that no cyclic amine is isolated.
Overall, this method is in contrast to other nitrogen
heterocycle syntheses involving the attack of iodine9 or
selenium10,11 reagents where stereochemical control is supplied
by other substituents or by reagents.
We thank the EPSRC for financial assistance towards the
purchase of the Nonius CCD diffractometer, and a studentship
(L. C.).
Notes and references
†
Crystal data for 4: C17H23NO2S, M = 305.42, monoclinic, space group
P21/n (no. 14), a = 8.5716(4), b = 6.5783(2), c = 28.464(2) Å, b =
92.190(2)°, U = 1603.8(2) Å3, Z = 4, m(Mo-Ka) = 0.206 mm21, 9478
reflections measured at 180(2) K using an Oxford Cryosystems Cryostream
cooling apparatus, 3642 unique (Rint = 0.044); R1 = 0.042, wR2 = 0.098.
The structure was solved with SHELXS-9712 and refined with SHELXL-
97.12 CCDC 207753.
For 6: C23H25N3O5S, M = 455.52, monoclinic, space group P21/c (no.
14), a = 12.2145(5), b = 7.8483(2), c = 23.4165(11) Å, b = 102.054(2)°,
U = 2195.3(2) Å3, Z = 4, m(Mo-Ka) = 0.188 mm21, 11499 reflections
measured at 180(2) K using an Oxford Cryosystems Cryostream cooling
apparatus, 3763 unique (Rint = 0.040); R1 = 0.039, wR2 = 0.090. The
structure was solved with SHELXS-9712 and refined with SHELXL-97.12
CCDC 207754. See http://www.rsc.org/suppdata/cc/b3/b303790h/ for crys-
tallographic data in CIF or other electronic format.
1 L. Caggiano, D. J. Fox and S. Warren, Chem. Commun., 2002, 2528.
2 D. J. Fox, D. House and S. Warren, Angew. Chem., Int. Ed., 2002, 41,
2462.
3 V. K. Aggarwal, I. Coldham, S. McIntyre and S. Warren, J. Chem. Soc.,
Perkin Trans. 1, 1991, 451.
4 A. J. Kirby, Adv. Phys. Org. Chem., 1980, 17, 183.
5 W. S. Lee, K. C. Jang, J. H. Kim and K. H. Park, Chem. Commun., 1999,
251.
Scheme 5 Reagents and conditions: i, LDA, THF, 278 °C; 1, 92%; ii,
LiAlH4, Et2O, 93%; iii, CDI, CH3CN, then SiO2 chromatography, 80%.
6 L. Djakovitch, J. Eames, D. J. Fox, F. H. Sansbury and S. Warren, J.
Chem. Soc., Perkin Trans. 1, 1999, 2771.
7 J. Eames, N. Kuhnert, F. H. Sansbury and S. Warren, Synlett, 1999,
1211.
8 K. Takaku, H. Shinokubo and K. Oshima, Tetrahedron Lett., 1996, 37,
6781.
9 A. D. Jones, D. W. Knight and D. E. Hibbs, J. Chem. Soc., Perkin Trans.
1, 2001, 1182.
10 T. Wirth, Angew. Chem., Int. Ed., 2000, 39, 3740.
11 R. Deziel and E. Malenfant, J. Org. Chem., 1995, 60, 4660.
12 G. M. Sheldrick, SHELXS-97/SHELXL-97, University of Göttingen,
Germany, 1997.
Scheme 6 Reagents and conditions: i, CDI, DMF, 100 °C, 32% (18).
CHEM. COMMUN., 2003, 1648–1649
1649