Studies on the formation of a tricyclic C2-symmetric sulfide

Article abstract of DOI:10.1016/S0040-4039(01)01455-1

Attempts to synthesise chiral sulfide 1 from the dimesylate precursor 5, via a cyclisation reaction using sodium sulfide in dimethyl sulfoxide, led to the unexpected formation of an unsymmetrical furan derivative 8. Studies were made into the effect of changing the leaving group, and changing the solvent of reaction. The desired compound could be synthesized in good yields from the xanthate 9 using a radical-mediated procedure.

Full text of DOI:10.1016/S0040-4039(01)01455-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 7091–7093
Studies on the formation of a tricyclic C2-symmetric sulfide
Caroline L. Winn and Jonathan M. Goodman*
The University Chemical Laboratory, Lensfield Road, University of Cambridge, Cambridge CB2 1EW, UK
Received 23 May 2001; revised 30 July 2001; accepted 7 August 2001
Abstract—Attempts to synthesise chiral sulfide 1 from the dimesylate precursor 5, via a cyclisation reaction using sodium sulfide
in dimethyl sulfoxide, led to the unexpected formation of an unsymmetrical furan derivative 8. Studies were made into the effect
of changing the leaving group, and changing the solvent of reaction. The desired compound could be synthesised in good yields
from the xanthate 9 using a radical-mediated procedure. © 2001 Elsevier Science Ltd. All rights reserved.
Compound 1 was the target molecule in a study of
chiral sulfur compounds, which have been shown to be
useful in asymmetric synthesis,1 and was related to the
C2-symmetric polyhydroxylated pyrrolidine (2), the syn-
thesis and potential applications of which have been
reported in the literature2 (Fig. 1).
expected for 1, except that some of the peaks were
doubled, showing that it was not symmetrical.
The unsymmetrical nature of the product could have
been due to epimerisation at one of the benzylidene
acetal centres. However, DIBALH reduction of the
benzylidene acetals afforded an unsymmetrical product,
as did reduction of the analogous para-methoxybenzyl-
idene acetals, which were prepared by a similar route.
Therefore, it could be concluded that epimerisation had
not occurred in this position, but that there had been a
retention of stereochemistry at one of the carbon cen-
tres adjacent to the mesylate during the cyclisation
reaction.
The synthetic route towards 1 initially involved diben-
zylidene protection of
D
-mannitol 3 followed by mesy-
lation of the resulting diol, 4. Dimesylate 5 was then
expected to undergo cyclisation via an SN2 substitution
reaction in the presence of sodium sulfide nonahydrate,
with inversion occurring at each reacting centre to
afford the enantiopure compound 1.
Acetal formation was based on the method of Baggett
and Stribblehill,3 although it was shown that the yield
could be improved to give 86% product without loss of
selectivity on increasing the reaction time to 5 days
(after 3 days, as described in the literature, the yield
was approximately 50%). Mesylation also proceeded in
good yield (82% after 16 h) but cyclisation using
sodium sulfide nonahydrate in dimethyl sulfoxide at
80°C for 16 h afforded a compound in 43% yield, and
approximately 20% of the starting material. This new
Mass spectrometry and elemental analysis showed that
the compound was not an unsymmetrical thiophene
derivative, but a furan derivative, 8, in which there had
been retention of configuration at one of the stereocen-
tres adjacent to the oxygen in the five-membered ring
(Scheme 1).
Full NMR characterisation4 showed that the product
was furan derivative 8 in which the stereochemistry at
one of the carbon centres adjacent to the oxygen had
undergone inversion, while the other had retained its
stereochemistry. It was shown that sodium sulfide nona-
hydrate was necessary for this reaction to occur; when
the reaction was repeated in the absence of the sulfide,
the starting material was recovered in near quantitative
yield. It is proposed, therefore, that the mechanism of
reaction involves attack of the sulfide anion onto the
sulfur atom of the mesylate, leaving an oxygen anion
which can undergo intramolecular attack onto the car-
bon atom bearing the second mesylate group in an SN2
fashion.
1
compound had an H NMR spectrum similar to that
Ph
Ph
Ph
Ph
O
O
O
O
O
O
O
O
S
N
H
1
2
Figure 1.
* Corresponding author.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01455-1

7092
C. L. Winn, J. M. Goodman / Tetrahedron Letters 42 (2001) 7091–7093
Ph
Ph
Ph
OR
O
O
OH
O
O
Ph
OH OH OH
OH OH OH
O
O
a
c
b
O
O
O
O
OR
O
O
OH
O
Ph
5 R = Ms; 6 R = Ts; 7 R = Tf
Ph
3
4
8
Scheme 1. Reagents and conditions: (a) PhCHO, DMF, H2SO4, rt, 120 h (86%); (b) either: Et3N, DCM, MsCl, 0°C, 16 h (82%
of 5); or: TsCl, Py, rt, 16 h (70% of 6); or: Tf2O, Py, THF, rt, 16 h (62% of 7); (c) Na2S·9H2O, DMSO, 80°C, 16 h (43% from
5, 51% from 6, 6% from 7).
in toluene with a,a-diazoisobutyronitrile as an initiator,
following the method of Rama Rao.6,7 The xanthate
could be prepared in excellent yields (90%) from the
dibenzylidene acetal. This reaction affords a single C2-
symmetric, diastereomeric product in which there is
inversion at each of the carbon centres adjacent to the
sulfur atom (Scheme 2).
Investigations were then carried out to see whether
different leaving groups resulted in the formation of the
same product. The corresponding ditosylate (6)3 and
ditriflate (7)5 were prepared according to literature
methods, and cyclisation afforded the same unsymmet-
rical product, 8, in 51% and 6% respectively, with none
of the expected compound, 1.
MacroModel MM2* calculations8,9 showed that the
symmetrical sulfide, 1, was the thermodynamically pre-
ferred diastereoisomer (Fig. 2, structure A), and this
may explain the single diastereomeric product from the
radical cyclisation. The global minimum conformation
of 1 is shown in Fig. 3.
Solvent effects were also examined. Cyclisation of the
mesylate was carried out in a 9:1 mixture of etha-
nol:iso-propanol at 80°C for 16 h, and afforded the
unsymmetrical furan compound in 67% yield. On heat-
ing with sodium sulfide in dimethylformamide at 150°C
for 40 h, it was possible to isolate the desired sulfide, 1,
in 10% yield, recovering approximately 40% of the
starting material. The harsh reaction conditions led to
decomposition of the starting material with some ben-
zaldehyde being recovered from the reaction mixture.
The tetrahydrofuran diastereoisomer which formed, 8
(Fig. 2, structure B), was of a much higher energy than
the symmetrical diastereoisomer, and so cannot form
under thermodynamic control. The strain energies in
the sulfides, relative to tetrahydrothiophene, are higher
than in the furans, relative to tetrahydrofuran, for
diastereoisomers A and B.
The sulfide, 1, could be synthesised by a radical-medi-
ated route from the xanthate using tributyl tin hydride
S
Ph
Ph
Ph
Ph
SMe
O
S
MeS
OH
O
O
O
O
O
O
O
Ph
O
O
Ph
O
O
Ph
O
O
a
b
O
O
O
O
OH
O
O
O
O
O
S
S
SMe
S
MeS
Ph
Ph
Ph
O
1
9
4
Scheme 2. Reagents and conditions: (a) NaH, THF, CS2, MeI, rt, 19 h (90%); (b) Bu3SnH, AIBN, toluene, 80°C, 16 h (78%). The
modified Barton–McCombie mechanism illustrated is based on that described in Rama Rao’s papers.6,7
Figure 2. Energies of the diastereomers.

C. L. Winn, J. M. Goodman / Tetrahedron Letters 42 (2001) 7091–7093
7093
2. Esposito, A.; Falorni, M.; Taddei, M. Tetrahedron Lett.
1998, 39, 6543–6546.
3. Baggett, N.; Stribblehill, P. J. Chem. Soc., Perkin Trans.
1 1977, 1123–1126.
4. Data for compound 8: [h]D −106.8 (c 0.81, CHCl3); mp
133–135°C; 1H NMR (C6D6, 400 MHz): l (4H, dd, J=
7.3, 1.4, PhH
OCHOPh), 5.30 (1H, s, OCH
CHO), 4.35 (1H, dd, J=4.5, 9.6, H
J=4.4, 12.1, Heq), 3.99 (1H, dd, J=6.0, 9.7, CH
(1H, dd, J=5.1, 12.1, CHaxO), 3.66–3.60 (2H, m, CHax
CHO), 3.39 (1H, td, J=4.5, 9.9, CH
O); 13C NMR
6
), 7.31–7.21 (6H, m, PhH
OPh), 4.47 (1H, t, J=6.4,
eqO), 4.11 (1H, dd,
O), 3.81
6 ), 5.48 (1H, s,
6
6
6
6
6
6
6
6
,
6
6
(C6D6, 400 MHz): l 139.5, 137.8, 128.8, 128.7, 128.2,
128.0, 126.6, 126.2, 102.4, 98.1, 87.3, 75.9, 73.8, 70.9,
70.8, 65.0; HRMS (+EI) calcd for C20H20NaO5:
363.1209; found: 363.1208 (MNa+). Elemental analysis
found: C, 70.39; H, 5.87. C20H20O5 requires: C, 70.56;
H, 5.93%.
Figure 3. Global minimum conformation of 1.
Acknowledgements
5. Shing, T. K. M. Tetrahedron 1988, 44, 7261–7264.
6. Rama Rao, A. V.; Ashok Reddy, K.; Gurjar, M. K.;
Kunwar, A. C. J. Chem. Soc., Chem. Commun. 1988,
1273–1274.
This work was supported by grants from the Royal
Society and the EPSRC (Engineering and Physical Sci-
ences Research Council).
7. Rama Rao, A. V.; Ashok Reddy, K.; Srinivas, N. R.;
Gurjar, M. K.; Padmaja, N.; Ramakumar, S.; Viswami-
tra, M. A.; Swapna, G. V. T.; Jagannadh, B.; Kunwar,
A. C. J. Chem. Soc., Perkin Trans. 1 1993, 1255–1259.
8. Mohamadi, F.; Richards, N. G. J.; Guida, W. C.;
Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.; Hen-
drickson, T.; Still, W. C. J. Comp. Chem. 1990, 11, 440–
467.
References
1. (a) Durst, T.; Breau, L. Phosphorus Sulfur Silicon Relat.
Elem. 1993, 74, 215–232; (b) Julienne, K.; Metzner, P.;
Greiner, A.; Henryon, V. Phosphorus Sulfur Silicon
Relat. Elem. 1999, 153, 341–342; (c) Li, A.-H.; Dai, L.-
X.; Aggarwal, V. K. Chem. Rev. 1997, 97, 2341–2372.
9. Allinger, N. L. J. Am. Chem. Soc. 1977, 99, 8127–8134.