Sulfonium ylide epoxidation reactions: Methylene transfer

Article abstract of DOI:10.1039/b316653h

Using a D-mannitol derived chiral sulfide, terminal epoxides are formed in up to 76% ee; the first example of double asymmetric induction in a sulfonium methylide epoxidation is reported and an improved method of generating sulfonium ylides is detailed.

Full text of DOI:10.1039/b316653h

Sulfonium ylide epoxidation reactions: methylene transfer
Benjamin R. Bellenie and Jonathan M. Goodman*
Unilever Centre for Molecular Science Informatics, Department of Chemistry, Lensfield Road, Cambridge,
UK CB2 1EW. E-mail: j.m.goodman@ch.cam.ac.uk; Fax: (+44) 1223 336362; Tel: (+44) 1223 336434
Received (in Cambridge, UK) 22nd December 2003, Accepted 10th March 2004
First published as an Advance Article on the web 1st April 2004
Using a
D
-mannitol derived chiral sulfide, terminal epoxides are
used to minimise signal broadening. Enantiomeric excess of 48%
was observed. According to the literature for the optical rotation of
styrene oxide in chloroform,6 the major isomer was assigned as S.
This is consistent with chiral shift NMR.7
formed in up to 76% ee; the first example of double asymmetric
induction in a sulfonium methylide epoxidation is reported and
an improved method of generating sulfonium ylides is de-
tailed.
A range of aromatic aldehydes were investigated as substrates for
this reaction (Table 1). Most enantioselectivities obtained are in the
range of 34–49%, similar to those obtained by Aggarwal using
disulfide 2.3 The more sterically demanding 2,4-dichlorobenzalde-
hyde was found to be comparatively very reactive under these
conditions—reaction was complete after just one day stirring at
ambient temperature, and gave isolated epoxide in almost quantita-
tive yield. The enantioselectivity of this reaction was found to be
much higher than for the other cases studied, 78% being obtained
on the first attempt. Repeating this reaction gave consistently high
levels of enantioselectivity—between 72% and 80%.
The synthesis of terminal epoxides in high enantiomeric excess has
proven very challenging. A kinetic resolution step is normally
required to achieve high enantioselectivity.1 The reaction of
sulfonium ylides with carbonyl compounds offers a great deal of
potential for forming this important functionality,2 however in
general low enantiomeric excesses have been obtained. Aggarwal’s
variant of the sulfur ylide epoxidation reaction, using Simmons–
Smith conditions to directly form the sulfur ylide from the sulfide
via a carbene addition, has led to the highest enantioselectivities
observed to date for the formation of styrene oxide by methylene
transfer: 16% ee using sulfide 1 and 47% ee using disulfide ligand
2 (Fig. 1).3
Crystalline, odour-free sulfide 34 has been shown to mediate aryl
transfer epoxidation reactions in reasonable yield and high
enantiomeric excess.5 Diaryl epoxides are of limited synthetic
utility and so it was desirable to extend this work to the more
synthetically useful monosubstituted epoxides. Treatment of 3 with
methyl iodide then base did not produce the product. The
Aggarwal/Simmons–Smith conditions were used successfully in
reactions with sulfide 3 and preliminary results are detailed here.
The reaction of sulfide 3 (2 eq.) with chloroiodomethane (2 eq.),
benzaldehyde (1 eq.) and diethylzinc (1 eq.) led to the formation of
styrene oxide after 2 days stirring at room temperature (Scheme 1),
with an isolated yield of 61%. Recovery of sulfide is straightfor-
ward: the reaction is quenched with water, extracted with
dichloromethane and extracts are filtered through a short silica
column. Epoxide and remaining aldehyde are rapidly eluted; as
sulfide 3 is considerably more polar, it can be recovered simply by
washing the column with further dichloromethane. Where the
reaction did not go to completion, flash chromatography was
necessary to separate epoxide from starting aldehyde.
Reducing the reaction temperature to 0 °C had a small beneficial
effect on selectivity: 53% ee was obtained for the formation of
para-nitrostyrene oxide. Reactivity was reduced, with a yield of
62% being observed.
One major disadvantage of this protocol is the use of two
equivalents of the chiral sulfide. This problem is minimised by the
ease of recovery of sulfide. Between 70% and 100% of sulfide is
recovered from the reaction; the amount of sulfide recovered
decreases where a longer reaction time is required. Recycled sulfide
appears to work just as selectively as freshly prepared, giving an ee
of 75% in the formation of ortho,para-dichlorostyrene oxide.
The possible active species proposed for the Simmons–Smith
epoxidation reaction are noted to be unstable at room temperature.8
This means that the addition of all reagents at the start of an
extended reaction time is unlikely to be the best method for
formation of epoxide product. The addition of dimethoxyethane
was found to stabilise zinc complexes, and extend the lifetime of
the active species. Also, generating the active species continually
by syringe-pump addition in the presence of dimethoxyethane
allows considerably shorter reaction times than previously re-
ported. The reaction of chloroiodomethane (3 eq.), tetrahy-
drothiophene (1 eq.) and aldehyde (1 eq.) in dimethoxyethane (1
eq.), and dichloromethane, with addition of diethylzinc (1.2 eq.) by
syringe pump over 2.5 h gives complete conversion of aldehyde to
styrene oxide. This compares favourably with previously observed
reaction times of one to two days,2,3 and only one equivalent of
sulfide is required. However, the use of the syringe pump is less
convenient than leaving the reaction to stir.
The enantiomeric excess of this, and the other examples detailed
below, was determined by chiral shift NMR using the (+) isomer of
Eu(hfc)3. Low field NMR spectrometers (200 or 250 MHz) were
The use of diethylzinc has also been problematic. Diethylzinc
was used as a solution in hexane : diethylzinc solution from
Table 1 Yield and enantioselectivity observed using different aldehydes
Enantiomeric
excess
of epoxide
Fig. 1 Enantioselectivities achieved in the formation of styrene oxide.3
Isolated yield
of epoxide
R (RCHO)
Phenyl
2-Naphthyl
4-Nitrophenyl
3,4-Dichlorophenyl
2,4-Dichlorophenyl
2,6-Dichlorophenyl
61%
54%
86%
48%
96%
92%
48%
34%
49%
49%
76%
47%
Scheme 1 Terminal epoxidation mediated by sulfide 3.
C h e m . C o m m u n . , 2 0 0 4 , 1 0 7 6 – 1 0 7 7
1076
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

View Online
different suppliers, and in some cases different batches from the
same supplier has led to inconsistent yields and reactivities.
The use of an isolable and analysable zinc complex removes the
inconsistencies between different supplies of diethylzinc. Stable
halomethylzinc complexes have been found by Charette et al. to be
effective in cyclopropanation reactions and in the alkylation of
sulfur.9 Therefore, the use of a stable complex in epoxidation
reactions was investigated.
Bis(chloromethyl)zinc : bipyridine 4 was prepared by addition of
a solution of 2,2A-dipyridyl to a solution of bis(chloromethyl)zinc,9
formed in situ from a 3 : 1 mixture of chloroiodomethane :
diethylzinc at low temperature (Scheme 2).
The bis(chloromethyl)zinc : bipyridyl complex 4 was found to be
effective in forming ylides for the sulfonium ylide epoxidation
reaction.† When 1.2 equivalents of complex 4 were used with one
equivalent of tetrahydrothiophene, styrene oxide was formed in
60% yield in just two hours. Disappointingly, no further reaction
was observed after this time indicating that all of the reagent has
been used up or decomposed. The use of two equivalents of the
bis(chloromethyl) complex 4 led to complete conversion in under
14 h. It is unclear why two equivalents of this reagent are
required.
To investigate the possibility that this reaction may work
catalytically, 0.2 eq. of tetrahydrothiophene were used. Only 18%
yield was obtained, indicating that stoichiometric sulfide is
required under these conditions. Nevertheless, this is an improve-
ment on the two equivalents required for this reaction using the
previous protocol.3
The use of complex 4 also proved effective in epoxidation
reactions with chiral sulfide 3 (Scheme 3), although reaction times
were longer than for tetrahydrothiophene. For the reaction
involving benzaldehyde, 56% yield was obtained after 20 hours,
and enantiomeric excess for styrene oxide formation was 57%. This
is the highest reported enantioselectivity for the direct formation of
styrene oxide by a sulfonium methylide epoxidation reaction.
To further test the effectiveness of the terminal epoxidation
conditions developed, they were applied to the synthesis of a useful
reaction intermediate. Amino acid-derived epoxide 5 is an
intermediate in the synthesis of the protease inhibitor amprena-
vir.2,9–11
Aldehyde 6 was formed from phenylalanine, using the synthetic
route developed by Reetz.12 This material was used immediately in
the epoxidation reactions described below (Scheme 4). Using a
racemic sulfide (tetrahydrothiophene) and literature epoxidation
conditions,2,3 the desired epoxide was found to be formed in a 5 :
1 syn : anti ratio, in agreement with published work.2 Using sulfide
3, a mixture of epoxides was formed in 57% conversion—a
pleasing result indicating that this sulfide is reactive in the
formation of aliphatic as well as aromatic epoxides. The ratio of syn
and anti epoxides (calculated by integration of an NMR sample of
the crude mixture) was 12 : 1.
Scheme 4 Combination of substrate and reagent control in epoxidation
reactions using sulfide 3.
sulfide used, the enantiomer of aldehyde 6 was synthesised, again
using the Reetz protocol12 but starting with the unnatural
-(R)-
D
phenylalanine. 45% conversion to product 7 was observed, with a
diastereomeric ratio of 2.7 : 1 (syn : anti). This case is
‘mismatched’, with the stereochemical induction of the sulfide
working against the substrate bias. The syn epoxide is the major
product, indicating that the reagent control is not powerful enough
to overturn the substrate selectivity. The use of preformed zinc
complex 4 was again more effective: 76% conversion to epoxides
was observed, in a similar ratio of 3.3 : 1.
To our knowledge, this is the first example of double asymmetric
induction in a sulfonium ylide epoxidation reaction, and further
demonstrates the potential utility of chiral sulfide 3 in the synthesis
of useful epoxide intermediates.
Notes and references
† Sample experimental procedure for epoxidation using zinc complex: To a
suspension of complex 4 (2 eq.) in dichloromethane (10 mL/mmol
aldehyde) was added sulfide (1 eq.) and aldehyde (1 eq.). The mixture was
stirred at 20 °C under argon for 2–24 h. Water (10 mL/mmol aldehyde) was
added, and the mixture extracted with dichloromethane (3 3 7 mL/mmol
aldehyde). Organic phases were filtered through a short plug of silica and
solvent removed under reduced pressure. Purification by flash column
chromatography (SiO2, 1 : 1 dichloromethane : hexane).
1 E. N. Jacobsen, Acc. Chem. Res., 2000, 33, 421.
2 V. K. Aggarwal, A. Ali and M. P. Coogan, J. Org. Chem., 1997, 62,
8628.
3 V. K. Aggarwal, M. P. Coogan, R. A. Stenson, R. V. H. Jones, R.
Fieldhouse and J. Blacker, Eur. J. Org. Chem., 2002, 319.
4 C. L. Winn and J. M. Goodman, Tetrahedron Lett., 2001, 42, 7091.
5 C. L. Winn, B. R. Bellenie and J. M. Goodman, Tetrahedron Lett., 2002,
43, 5427.
To investigate whether this is a genuine double stereoinductive
effect and not merely an artefact of the increased steric bulk of the
6 V. Capriati, S. Florio, R. Luisi and A. Salomone, Org. Lett., 2002, 4,
2445.
7 H. J. C. Yeh, S. K. Balani, H. Yagi, R. M. E. Greene, N. D. Sharma, D.
R. Boyd and D. M. Jerina, J. Org. Chem., 1986, 51, 5439.
8 A. B. Charette and J.-F. Marcoux, J. Am. Chem. Soc., 1996, 118, 4539;
S. E. Denmark, J. P. Edwards and S. R. Wilson, J. Am. Chem. Soc.,
1992, 114, 2592.
Scheme 2 Formation of an isolable bis(chloromethyl)zinc complex.
9 A. B. Charette, J.-F. Marcoux, C. Molinaro, A. Beauchemin, C. Brochu
and E. J. Isabel, J. Am. Chem. Soc., 2000, 122, 4508.
10 R. D. Tung, US Patent, 5585397, 1996.
11 J. Barluenga, B. Baragaña and J. M. Concellón, J. Org. Chem., 1995, 60,
6696.
12 M. T. Reetz, M. W. Drewes and R. Schwickardi, Org. Synth., 1998, 76,
110.
Scheme 3 Terminal epoxidation using sulfide 3 and complex 4.
C h e m . C o m m u n . , 2 0 0 4 , 1 0 7 6 – 1 0 7 7
1077