Catalytic decarbonylation of epoxyaldehydes: Applications to the preparation of terminal epoxides

Article abstract of DOI:10.1055/s-0029-1217562

A catalytic decarbonylation reaction for epoxyaldehydes is reported. This reaction may be sequentially used with known asymmetric methods to access optically active mono- and disubstituted terminal epoxides, as illustrated for a key example. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-0029-1217562

2076
LETTER
Catalytic Decarbonylation of Epoxyaldehydes: Applications to the
Preparation of Terminal Epoxides
Decarbonylation
i
of Epox
l
yaldeh
l
ydes Morandi, Erick M. Carreira*
Laboratorium für Organische Chemie, ETH Zürich, 8093 Zürich, Switzerland
Fax +41(1)6321328; E-mail: carreira@org.chem.ethz.ch
Received 2 March 2009
ly, we sought a complementary strategy for the prepara-
tion of optically active epoxides by combining existing
methods for the preparation of optically active epoxy-
aldehydes and catalytic decarbonylation processes
(Scheme 1). Herein we document for the first time the de-
carbonylation reaction of epoxides, a reaction lacking in
precedence. The work extends the approach we have pre-
viously reported involving the use of removable groups as
a strategy for the preparation of optically active building
blocks.2
Abstract: A catalytic decarbonylation reaction for epoxyaldehydes
is reported. This reaction may be sequentially used with known
asymmetric methods to access optically active mono- and disubsti-
tuted terminal epoxides, as illustrated for a key example.
Key words: asymmetric synthesis, decarbonylation, epoxides, ho-
mogenous catalysis, rhodium
The catalytic decarbonylation of aldehydes has become a
well-known reaction since its discovery by Tsuji and
Ohno.1 We previously used this reaction in a new strategy
to access chiral building blocks in high optical purity, us-
ing the aldehyde as a removable steering group.2 More re-
cently, Madsen has used the catalytic decarbonylation of
unprotected aldoses to generate polyols.3 Herein we de-
scribe a procedure for the decarbonylation of epoxyalde-
hydes that permits preparation of a range of terminal
epoxides. Moreover, this method, coupled with known
asymmetric procedures for the preparation of optically ac-
tive epoxyaldehydes, allows access to optically active ter-
minal epoxides, which can otherwise be difficult to
prepare.
1) Sharpless
2) oxidation
R1
R2
OH
R1
R2
O
R1
R2
O
O
*
*
*
decarbonylation
O
O
O
O
Jørgensen
*
*
R1
R1
R1
*
Scheme 1 Decarbonylation strategy for the preparation of optically
active terminal epoxides
The implementation of the decarbonylation approach with
epoxyaldehydes is not without potential pitfalls. Thus, for
example, epoxides themselves are known to form
rhodaoxetanes with certain rhodium complexes.12 Of ad-
ditional concern, it was not clear if it would be possible to
decarbonylate epoxyaldehydes without having the prod-
uct oxirane suffer ring opening.
Several approaches have been described for the enantio-
selective synthesis of epoxides. Sharpless pioneered the
field with a titanium-mediated epoxidation of allylic alco-
hols, which still remains one of the most useful reactions
in organic synthesis.4 Jacobsen has described the selective
kinetic resolution of terminal epoxides with a cobalt–
salen complex.5 Important additional advances for the
enantioselective preparation of epoxides have also ap-
peared.6 Among these, Jørgensen has described an enanti-
oselective epoxidation reaction of unsaturated aldehydes
with hydrogen peroxide and a proline-derived catalyst.7
Shi has used a sugar-derived organocatalyst in combina-
tion with oxone to epoxidize unfunctionalized olefins in
high ee’s.8 More recently, List has disclosed a comple-
mentary strategy for the enantioselective epoxidation of
unsaturated aldehydes.9 Shibasaki has documented a cat-
alytic, enantioselective approach to 2,2-disubstituted
terminal epoxides through the addition of dimethylsulfo-
nium methylide to ketones.10
We initiated the project by identifying optimal conditions
for the decarbonylation of racemic citral oxide (Table 1,
entry 1), which is easily accessible by aqueous epoxida-
tion of the commercially available citral. Screening of a
variety of solvents, reaction temperatures, and ligands
with Rh(I) catalysts, afforded an optimal yield of 73% for
this substrate. The major side-product of the reaction
could be identified as 6-methyl-5-hepten-2-one, which is
likely to form via a rhodaoxetane intermediate derived
from the oxidative addition of the Rh-catalyst to the prod-
uct epoxide that subsequently undergoes fragmentation to
give the ketone and a rhodium–methylene complex.12
Nonetheless, encouraged by this result, we investigated
the scope of the reaction (Table 1).13
Despite the number of available methods for enantioselec-
tive epoxidation, the asymmetric synthesis of terminal ep-
oxides, and mainly disubstituted terminal epoxides,
remains a challenge for organic chemistry.11 Consequent-
The various substrates undergo decarbonylation in 40–
73% yield. In all cases, the corresponding rearrangement
product, the methylketone or corresponding aldehyde,
was isolated as a by-product. The reaction shown in
Table 1, entry 5 was problematic and represents a limita-
tion to the method because, under the standard conditions
SYNLETT 2009, No. 13, pp 2076–2078
x
x
.x
x
.2
0
0
9
Advanced online publication: 15.07.2009
DOI: 10.1055/s-0029-1217562; Art ID: D06509ST
© Georg Thieme Verlag Stuttgart · New York

LETTER
Decarbonylation of Epoxyaldehydes
2077
1) ref 14.
92%, 85% ee
Me
Me
Me
Me
O
2) TPAP (6 mol%)
NMO
CH2Cl2, 77%
Me
OH
Me
CHO
2
1
[Rh(cod)Cl]2 (2.5 mol%)
rac-BINAP (10 mol%)
1,2-dichlorobenzene, sealed tube
140 °C, 14 h, 73%
Me
Me
O
85% ee
Me
3
Scheme 2 Preparation of an optically active epoxide
unfunctionalized olefins are based on electrophilic re-
agents and are, thus, unable to selectively epoxidize elec-
tron-poor double-bonds in the corresponding precursors.
Table 1 Decarbonylation of Racemic Epoxyaldehydes
O
[Rh(cod)Cl]2 (2.5 mol%)
R1
R2
O
R1
R2
O
R1
R2
rac-BINAP (10 mol%)
+
O
After having explored the scope of the reaction on racem-
ic substrates, we prepared an optically active epoxyalde-
hyde in order to test whether the decarbonylation of chiral
epoxyaldehydes would afford optically active epoxides.
Sharpless asymmetric epoxidation14 of 1, followed by
TPAP/NMO oxidation, furnished 2 in 71% yield and 85%
ee (Scheme 2). Subjecting this epoxyaldehyde to decarbo-
nylation conditions provided 3 in 73% yield and 85% ee.
Thus, no significant loss of optical purity was observed in
the course of the decarbonylation reaction.
1,2-dichlorobenzene
sealed tube
A
B
135–145 °C, 14 h
Entry R1
R2
Yield A YieldB
(%)a
(%)a
1
Me
73
23
Me
Me
BnO
Ph
2
Me
H
45
63
64
40
64
40
3
26
In conclusion, we have developed a new catalytic decar-
bonylation reaction, which was not previously known for
epoxyaldehydes. When coupled with known asymmetric
transformations that yield optically active epoxyalde-
hydes, decarbonylation provides access to optically active
terminal epoxides 3 in good ee without any observable ra-
cemization, as determined for a test substrate.
4
n-Heptyl
Ph
H
23b
20b,d
30
5c
6
Me
Me
Me
Me
Me
Me
7
Me
54
34
Acknowledgment
OTIPS
We are grateful to the Swiss National Science Foundation and SCCI
for funding this project.
a Isolated yield.
b NMR yields.
c Reaction performed in p-cymene with (S)-BINAP as ligand.
References and Notes
d 35% of 2-phenylpropanal was isolated.
(1) Ohno, K.; Tsuji, J. J. Am. Chem. Soc. 1969, 90, 99.
(2) Fessard, T.; Andrews, S. P.; Motoyoshi, H.; Carreira, E. M.
Ang. Chem. Int. Ed. 2007, 46, 9331.
(3) Monrad, R. N.; Madsen, R. J. Org. Chem. 2007, 72, 9782.
(4) For a review, see: Pfenninger, A. Synthesis 1986, 89.
(5) Schaus, S. E.; Brandes, B. D.; Larrow, J. F.; Tokunaga, M.;
Hansen, K. B.; Gould, A. E.; Furrow, M. E.; Jacobsen, E. N.
J. Am. Chem. Soc. 2002, 124, 1307.
(6) For a general review on organocatalysis, see: Dalko, P. I.;
Moisan, L. Angew. Chem. Int. Ed. 2004, 43, 5138.
(7) Marigo, M.; Franzen, J.; Poulsen, T. B.; Zhuang, W.;
Jorgensen, K. A. J. Am. Chem. Soc. 2005, 127, 6964.
(8) Wang, Z.-X.; Tu, Y.; Frohn, M.; Zhang, J.-R.; Shi, Y. J. Am.
Chem. Soc. 1997, 119, 11224.
(9) Wang, X.; List, B. Angew. Chem. Int. Ed. 2007, 47, 1119.
(10) Sone, T.; Yamaguchi, A.; Matsunaga, S.; Shibasaki, M.
J. Am. Chem. Soc. 2008, 130, 10078.
described above, this reaction gave 2-phenylpropanal as
the major product without any traces of a-methylstyrene
oxide. The formation of 2-phenylpropanal might be ratio-
nalized by a Pinacol-type rearrangement of the corre-
sponding decarbonylated epoxide. However, the use of a
less polar solvent, p-cymene, provided moderate yields of
the desired product epoxide. (S)-BINAP was used in this
case since precipitation was observed to occur with the
rac-BINAP in p-cymene, whereas the complex formed
with (S)-BINAP produced a homogenous solution
throughout the reaction course. To our knowledge, this
difference in behavior between the racemic and the enan-
tiopure BINAP has not been previously noted.
(11) Arends, I. W. C. E. Angew. Chem. Int. Ed. 2006, 45, 6250.
(12) For more information on rhodaoxetane, see: Calhorda, M. J.;
Galvao, A. M.; Uenaleroglu, C.; Zlota, A.; Frolow, F.;
Milstein, D. Organometallics 1993, 12, 3316.
The reaction conditions are compatible with trisubstituted
double bonds (Entries 1 and 6). This last result is of rele-
vance, since known asymmetric epoxidation methods of
Synlett 2009, No. 13, 2076–2078 © Thieme Stuttgart · New York

2078
B. Morandi, E. M. Carreira
LETTER
Triisopropyl{3-methyl-2-[2-(2-methyloxiran-2-
(13) General procedure for the decarbonylation reaction:
[Rh(cod)Cl]2 (3.8 mg, 0.0077 mmol) and rac-BINAP (18.5
mg, 0.03 mmol) were dissolved in o-dichlorobenzene (3 mL)
in a 25 mL sealed tube under an argon atmosphere. After 15
min of stirring at r.t., a homogenous dark-red solution
formed. The corresponding a,b-epoxyaldehyde (0.3 mmol)
was then added to the stirring mixture, the tube was
evacuated and filled with argon three times and finally
sealed under vacuum. The reaction mixture was then
immerged in a pre-heated oil bath at the corresponding
temperature (140 °C). After 14 h, the mixture was cooled to
r.t. and directly loaded onto a column packed with silica gel.
Prior elution with pure pentane to remove the o-dichloro-
benzene, followed by elution with the corresponding solvent
mixture (pentane–Et2O) afforded the pure compound as an oil.
Caution: careful work-up is necessary since some of the
products are highly volatile!
yl)ethyl]but-3-enyloxy}silane (Table 1, entry 7, A):
1H NMR (300 MHz, CDCl3): d (mixture of isomers) = 4.81
(br s, 1 H), 4.70 (br s, 1 H), 3.62–3.58 (m, 2 H), 2.60–2.57
(m, 2 H), 2.19 (m, 1 H), 1.63 (s, 3 H), 1.60–1.40 (m, 4 H),
1.30 (s, 3 H), 1.05 (m, 21 H). 13C NMR (75 MHz, CDCl3): d
(mixture of isomers) = 145.4, 112.1, 66.4, 57.0, 54.1, 53.8,
49.8, 34.4, 25.0, 21.1, 21.0, 20.3, 20.2, 18.2, 12.1. IR (neat):
2924 (m), 2865 (m), 1462 (m), 1106 (m), 882 (s), 680 (s)
cm–1. HRMS (ESI): m/z calcd for C19H38O2SiNa+: 349.2534;
found: 349.2534.
6-Methyl-5-[(triisopropylsilyloxy)methyl]hept-6-en-2-
one (Table 1, entry 7, B): 1H NMR (300 MHz, CDCl3): d =
4.81 (s, 1 H), 4.70 (s, 1 H), 3.62–3.58 (m, 2 H), 2.43–2.38
(m, 2 H), 2.19 (m, 1 H), 2.12 (s, 3 H), 1.82 (m, 1 H), 1.66 (s,
3 H) 1.60 (m, 1 H), 1.05 (m, 21 H). 13C NMR (75 MHz,
CDCl3): d = 209.1, 145.3, 112.3, 66.2, 49.2, 41.5, 29.8, 23.4,
19.8, 17.9, 11.9. IR (neat): 2942 (m), 2865 (m), 1719 (s),
1112 (m), 881 (s), 679 (s) cm–1. HRMS (ESI): m/z calcd for
C18H36O2SiNa+: 335.2377; found: 335.2376.
(14) Nishiguchi, G. A.; Little, R. D. J. Org. Chem. 2005, 70,
5249.
Synlett 2009, No. 13, 2076–2078 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.