New directions for the Morita-Baylis-Hillman reaction; homologous aldol adducts via epoxide opening

Article abstract of DOI:10.1039/b603510h

Under trialkylphosphine catalyzed Morita-Baylis-Hillman reaction conditions, epoxides react with enones to give rise to homologous aldol adducts. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b603510h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
New directions for the Morita–Baylis–Hillman reaction; homologous
aldol adducts via epoxide opening{
Marie E. Krafft* and James A. Wright
Received (in Bloomington, IN, USA) 8th March 2006, Accepted 22nd May 2006
First published as an Advance Article on the web 7th June 2006
DOI: 10.1039/b603510h
resulting in the first homologous intramolecular MBH reaction,
which represents a C(sp2)–C(sp3) coupling with concomitant
cyclization.
Under trialkylphosphine catalyzed Morita–Baylis–Hillman
reaction conditions, epoxides react with enones to give rise to
homologous aldol adducts.
Organocatalyzed reaction of epoxy enone 1 would be expected
to generate either alcohol 2 from exo opening of the epoxide or
alcohol 3 from the endo mode of opening (eqn (1)) via a traditional
MBH organocatalyzed mechanism (Scheme 1).
The Morita–Baylis–Hillman (MBH) reaction1 is an organocataly-
tic process that generates aldol adducts via reaction at the a-carbon
of an unsaturated carbonyl moiety. Existing conditions for the
MBH reaction now permit the use of a wide variety of alkenes,
electrophiles and nucleophilic catalysts, including both tertiary
phosphines and amines. The types of electrophiles that have been
shown to react under MBH conditions include aldehydes, ketones,
aldimines, arenes, unsaturated ketones2 and vinyl sulfones.3
Recently, we have extended the traditional MBH reaction to
include allylic halides4 and alkyl halides.5 Epoxides however, have
been overlooked as an electrophile in the MBH reaction despite
being considered an important organic functional group.
Nucleophilic epoxide-opening reactions play a key role in the
construction of both carbon–carbon and carbon–oxygen bonds,
essential components of organic synthesis. Reactions of epoxides
with enolates generate a chain extended homologous aldol
product.6 A number of examples illustrate effective epoxide
opening resulting in skeletal enhancement under either Lewis
acidic or base-catalyzed reaction conditions.7
ð1Þ
Substitution on the epoxide or the tether would be expected to bias
which cyclization mode will operate. Initial studies were conducted
with epoxy enones 4 and 5 that were prepared as described in
Scheme 2.13 Treatment of epoxide 4 with 1 equiv. of PBu3 in
t-BuOH at rt led to the isolation of 25% of cyclic enone 6
in addition to a significant amount of unrecognizable materials
In the presence of nucleophilic catalysts, epoxides have
demonstrated variable stability and reactivity. An early commu-
nication reported that epoxides failed to yield recognizable
products in the tertiary amine-catalyzed intermolecular MBH
reaction with acrylates8 whereas a,b-epoxy aldehydes, on the other
hand, underwent efficient DABCO catalyzed coupling giving a
traditional MBH aldol adduct.9 To date, there have been no
reports of attempted phosphine-catalyzed MBH reactions with
epoxides. Under tributylphosphine catalysis, epoxides have been
shown to react with deoxygenation (160 uC),10 or react with the
phosphine resulting in ring opening under mild conditions (rt A
refluxing t-BuOH).11 In addition, epoxides react with acetate
generated by reaction of Ac2O with tributylphosphine in refluxing
toluene, thus illustrating the stability of epoxides in the presence of
a trialkylphosphine when competing processes are possible.12
These examples demonstrate that the reactivity of epoxides in the
presence of trialkylphosphines or tertiary amines is highly
dependent on the reaction environment. Herein we report our
unprecedented work on the use of epoxides in the MBH reaction
Scheme 1
Department of Chemistry and Biochemistry, Florida State University,
Tallahassee, FL, 32306-4390, USA. E-mail: mek@chem.fsu.edu;
Fax: 850-644-7409; Tel: 850-644-2297
{ Electronic supplementary information (ESI) available: Experimental
details for the synthesis of compounds 4–6 and 9–27. See DOI: 10.1039/
b603510h
Scheme 2
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 2977–2979 | 2977

View Article Online
warranted. Optimization studies involving variations in both
Table 1
solvent concentration and equivalents of trialkylphosphine led to
conditions for efficient epoxide opening (Table 1, eqn (2)). These
results showed that PMe3 was a more effective organocatalyst than
PBu3. At higher concentrations (0.5 M), loss of material and
generation of unrecognizable products14 was a problem regardless
of whether catalytic or stoichiometric amounts of PMe3 or PBu3
were used. At lower concentrations (0.025 M) only PMe3 was an
effective organocatalyst giving rise to 65–67% of the desired
cyclized adduct. Both stoichiometric and catalytic conditions gave
the same result at 0.025 M concentration using PMe3, but the
stoichiometric reaction required less time. The use of higher
temperatures was not advantageous to the reaction outcome. It
was therefore beneficial to continue further studies using stoichio-
metric amounts of trialkylphosphine in order to keep the reaction
time to a minimum.
(2)
Concentration/M
PR3 (equiv.)
Time/h
Yield (%)
0.50
0.50
0.20
0.10
0.10
0.05
0.025
0.025
0.025
PBu3 (0.25)
PBu3 (1.0)
PMe3 (1.0)
PMe3 (0.10)
PMe3 (1.0)
PMe3 (0.10)
PBu3 (0.10)
PMe3 (0.10)
PMe3 (1.0)
72
30
30
72
30
72
72
72
30
25
25
38
41
44
48
n.r.
65
67
To ensure that direct reaction of the phosphine with the epoxide
was not a competing or interfering process, control reactions were
carried out. Reaction of epoxides 7 or 8 with 1 equiv. of either
PBu3 or PMe3 in t-BuOH (0.05 M) at rt for 20 h resulted in the
recovery of 90–95% of the epoxide (eqn (3)).
(eqn (2)). None of the endo epoxide opening adduct was observed.
While this was a promising result because it represented the first
epoxide opening using a zwitterionic enolate under MBH
conditions, further refinement of reaction conditions was
Table 2 Homologous Morita–Baylis–Hillman epoxide-opening reactionsa
Entry
Epoxide
Equiv. PMe3
Time
Product(s)
Yield (%)
1
2
4
5
R = Me
R = Ph
1.0
1.0
30 h
30 h
6
9
67
66
3
4
10
13
R = Me
R = Ph
1.0
1.0
18 h
18 h
11
14
1.7 : 1
1.7 : 1
12
15
65
73
5
6
16
18
R = Me
R = Ph
1.0
1.0
72 h
72 h
17
19
60
50
7
8
20
22
R = Me
R = Ph
1.0
1.0
18 h
18 h
21
23
76
70
9
10
a
24
26
R = Me
R = Ph
10.0
10.0
7 d
72 h
25
27
43
92
Reactions were conducted at a concentration of 0.025 M in tert-butanol at room temperature.
2978 | Chem. Commun., 2006, 2977–2979
This journal is ß The Royal Society of Chemistry 2006

View Article Online
ð3Þ
These results strongly suggest that the traditional MBH organo-
catalyzed mechanism1 described in Scheme 1 is operative and the
trialkylphosphine acts as a nucleophile adding to the enone and
not the epoxide. Conjugate addition of PMe3 to the enone gives
rise to a zwitterionic enolate which subsequently adds to the
epoxide giving the corresponding zwitterionic alkoxide.
Subsequent alkoxide induced elimination of Me3P gives rise to
the observed cyclic enone homoaldol adduct.
We have described an unprecedented opening of epoxides under
Morita–Baylis–Hillman reaction conditions. While epoxides are
often considered an important functional group in organic
synthesis, they have been overlooked as an electrophile in the
MBH reaction. Nucleophilic epoxide openings are key reactions in
the construction of carbon skeletons and this new reaction
enhances the synthetic utility of epoxides. The MBH epoxide
opening results in the formation of homologous aldol products
efficiently and embodies a C(sp2)–C(sp3) coupling with concomi-
tant cyclization. Further work on the scope of this reaction is
currently in progress.
A number of epoxy enones underwent effective cyclization
giving new homologous MBH adducts resulting from opening of
the epoxide by the zwitterionic enolate (Table 2). Unsubstituted
enone-epoxides 10 and 13 gave almost equimolar mixtures of the
endo and exo modes of opening due to minimal steric interactions
influencing one mode of opening. In the cyclizations of either
epoxide 10 or 13, the products of a 5-exo mode of opening,
alcohols 12 or 15 respectively, would be expected to be favored
kinetically. However, with the unsubstituted epoxide terminus,
the 6.5-endo mode of cyclization apparently competes favorably
giving rise to alcohols 11 or 14. Reaction at the unsubstituted
epoxide terminus giving endo selectivity is competitive in the
absence of any other overriding steric factors evident in
the examples in entries 5–10, Table 2. While reaction of the
unsubstituted epoxides 10 or 13 (entries 3 and 4, Table 2)
exhibited marginal regioselectivity in the epoxide opening, the
examples in entries 5–10 were highly selective. Introduction of
geminal substituents adjacent to the epoxide (entries 7 and 8) did
not have a detrimental effect on the cyclization, and ring opening
via the 6-endo mode was preferred due to significant steric
interactions generated between the geminal dimethyl groups and
the enolate as shown with transition state models 20x and 20n.
Reaction of the c-disubstituted enone was expected to be slow if
the reaction were to occur at all. Substitution adjacent to the site of
nucleophilic addition of trimethylphosphine should make the
addition difficult and thus decrease the zwitterion concentration.
Reaction of methyl ketone 24 was extremely slow requiring 7 d
and 10 equiv. of phosphine for complete consumption of starting
material. The corresponding phenyl ketone was consumed during
the course of 72 h although 10 equiv. of phosphine were also
required. However, when the reaction of enone 24 was conducted
at higher concentration (0.1 M) with 10 equiv. of PMe3 for 30 h,
enone 25 was still generated in only 44% yield. Opening to form
the five-membered ring via the exo mode was preferred over the
endo mode generating the cyclohexenol. At this point we have
been unable to promote cyclizations where 6-exo adducts are
expected.
The National Science Foundation and the MDS Research
Foundation supported this work.
Notes and references
1 D. Basavaiah, A. J. Rao and T. Satyanatayana, Chem. Rev., 2003, 103,
811; E. Ciganek, Org. React., 1997, 51, 201; P. Langer, Angew. Chem.,
Int. Ed., 2000, 39, 3049; D. Basavaiah, P. D. Rao and R. S. Hyma,
Tetrahedron, 1996, 52, 8001.
2 For examples of phosphine catalyzed reactions of enones, i.e., an
intramolecular vinylogous MBH or an intramolecular Rauhut–Currier
reaction, see: S. A. Frank, D. J. Mergott and W. R. Roush, J. Am.
Chem. Soc., 2002, 124, 2404; L.-C. Wang, A. L. Luis, K. Agapiou,
H.-Y. Jang and M. J. Krische, J. Am. Chem. Soc., 2002, 124, 2402, and
references cited therein.
3 A. L. Luis and M. J. Krische, Synthesis, 2004, 2579.
4 M. E. Krafft and T. F. N. Haxell, J. Am. Chem. Soc., 2005, 127, 10168.
For examples of organomediated MBH reactions of allylic carbonates
under Pd catalysis, see: B. G. Jellerichs, J. -R. Kong and M. J. Krische,
J. Am. Chem. Soc., 2003, 125, 7758.
5 M. E. Krafft, K. A. Seibert, T. F. N. Haxell and C. Hirosawa, Chem.
Commun., 2005, 5772.
6 S. K. Taylor, Tetrahedron, 2000, 56, 1149.
7 For examples, see: D. F. Taber, L. J. Silverberg and E. D. Robinson,
J. Am. Chem. Soc., 1991, 113, 6639; P. Crotti, V. D. Bussolo, L. Favero,
F. Macchia, M. Pineschi and E. Napolitano, Tetrahedron, 1999, 55,
5853; G. H. Posner, J. P. Maxwell and M. Kahraman, J. Org. Chem.,
2003, 68, 3049, and references cited therein; N. Asao, T. Kasahara and
Y. Yamamoto, Tetrahedron Lett., 2001, 42, 7903; P. Crotti,
V. DiBussolo, L. Favero, M. Pineschi and M. Pasero, J. Org. Chem.,
1996, 61, 9548; P. Van de Weghe and J. Collin, Tetrahedron Lett., 1995,
36, 1649.
8 J. S. Hill and N. S. Isaacs, J. Chem. Res. (M), 1988, 2641.
9 P. R. Krishna, K. R. Lopinti and V. Kannan, Tetrahedron Lett., 2004,
45, 7847.
10 G. Wittig, Chem. Ber., 1955, 88, 1654.
11 R.-H. Fan, X.-L. Hou and L.-X. Dai, J. Org. Chem., 2004, 69, 689.
12 R.-H. Fan and X.-L. Hou, Tetrahedron Lett., 2003, 44, 4411.
13 Other enones were prepared in a similar manner; see ESI{ for details.
14 Intermolecular enone head-to-tail coupling products, as mixtures of
dimers and perhaps trimers, appear to constitute a portion of the
remaining material.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 2977–2979 | 2979