A tandem carbanion addition/carbon-carbon bond cleavage yields alkynyl ketones

Article abstract of DOI:10.1021/ja050663m

Carbanion addition to vinylogous acid triflates triggers carbon-carbon bond cleavage to form alkynyl ketones under mild conditions. Mechanistic factors affecting the cleavage event and its relationship to complementary fragmentations are discussed. A rang

Full text of DOI:10.1021/ja050663m

Published on Web 03/15/2005
A Tandem Carbanion Addition/Carbon-Carbon Bond Cleavage Yields Alkynyl
Ketones
Shin Kamijo and Gregory B. Dudley*
Department of Chemistry and Biochemistry, Florida State UniVersity, Tallahassee, Florida 32306
Received February 1, 2005; E-mail: gdudley@chem.fsu.edu
Carbon-carbon bonds are the foundation of organic chemistry.
Tools for making such bonds are indispensable for the construction
of complex molecules, and intense efforts are continually directed
toward developing novel and selective bond-forming reactions. In
contrast, C-C bond cleavage reactions receive less attention.
Scheme 1. Fragmentation of Epoxy-Hydrazones
1
2
Oxidative cleavage, sigmatropic rearrangement, and alkene me-
3
tathesis are key examples on a short list of synthetic methods that
Scheme 2. Preparation of Enones from Vinylogous Esters
involve C-C bond cleavage. Chemists justifiably focus on bond-
forming reactions; however, judicious use of bond cleavage can
reveal frameworks that are otherwise difficult to prepare.
The Eschenmoser-Tanabe fragmentation (Scheme 1)4,5 is a
fundamental bond cleavage reaction, important from both a
pedagogical and synthetic standpoint, although it is often overlooked
in favor of other methods for preparing alkynyl ketones. Generally
in this process, epoxy-hydrazones (1) are prepared from cyclic
enone oxides in protic media (acetic acid or methanol); they then
decompose (1 f 2) with loss of nitrogen and a sulfinic acid. Scheme
Table 1. Initial Screening of Addition/Fragmentation Conditionsa
2
illustrates a common method for preparing cyclic enones (4):
nucleophilic addition to 3, followed by acidic hydrolysis.6
The presumed intermediates in Schemes 1 and 2 (A and B) are
strikingly similar, yet they generally decompose by distinct,
unrelated mechanisms. Inducing B to fragment under mild condi-
tions in an aprotic solvent would have important synthetic and
mechanistic implications. We achieved this crossover in reaction
pathways (B f 2) using vinylogous acid triflates, and this
Communication describes our preliminary investigations.
Table 1 summarizes our screening of reaction protocols, using
phenylmagnesium bromide (PhMgBr) as the carbanion source.
Entries 1-3 reveal a clear advantage for the ethereal solvent (THF)
over DMPU or toluene for production of 8a. Alternative nucleofuges
entry
X
solvent
observed products
1
2
OTf (5a)
OTf (5a)
THF
toluene
8a (78-83% isolated)
7 (66% isolated),
b
8a (trace)
8a (ca. 50% )
c
3
4
5
OTf (5a)
OMs (5b)
Br (5c)
DMPU
THF
THF
6
6
a
See Supporting Information for details. b 6 presumably hydrolyzes to
1
7
when X ) OTf. c Estimated by H NMR spectroscopy.
tion with the developing carbonyl.13 The heteroaromatic nucleophile
in entry 6 yielded the expected product (8f, 63%) without any
difficulty.
We also examined the C-C bond cleavage reaction employing
organolithium nucleophiles (entries 7-10). The more ionic lithium
alkoxide fragments more readily. In the case of phenyllithium
entries 47 and 5 ) did not fragment under our solution-phase
8
(
conditions.9 Varying the reaction concentration did not have a
,10
discernible effect.
Entries 1 and 2 illustrate that choice of solvent enables one to
choose between the two reaction pathways (3 f 2 or 4). This
provides an additional level of diversity with respect to the synthesis
of small molecule libraries.11 No attempt was made to optimize
the yield of 7 (entry 2).
We then explored the reaction of enol triflate 5a with a variety
of Grignard reagents (Table 2, entries 1-6). We prepared 5a by a
modified procedure12 that provides efficient and inexpensive access
to a range of stable triflates in good yields.
As mentioned above, PhMgBr provided the corresponding
alkynyl ketone 8a in 80% yield (entry 1, average of two runs).
The reaction with para-anisylmagnesium bromide afforded 8b
smoothly in 86% yield (entry 2). In contrast, the para-chlorophenyl
adduct (entry 5) fragmented more slowly; mild heating provided
the best yield of 8e. It thus appears that electron-donating groups
accelerate C-C bond cleavage, perhaps due to favorable conjuga-
(
PhLi), appreciable bond cleavage occurred at 0 °C, with the
resulting ketone then being susceptible to over-addition. Initial
cooling to -78 °C suppressed formation of a tertiary alcohol
impurity and provided alkynyl ketone 8a in 93% yield (average of
two runs). The meta- and ortho-anisyl reagents afforded the
corresponding products (8c and 8d) in 78 and 57% yields,
respectively (entries 8 and 9). This is an improvement over
Grignard-derived yields (entries 3 and 4). Methyllithium is also
applicable as a nucleophilic reagent in this transformation, yielding
14
alkynyl ketone 8g (65%, entry 10).
Having established a reasonable range of nucleophilic partners,
we next examined the scope of cyclic enol triflates. Mild heating
proved optimal in the case of the unsubstituted enol triflate (9; n
2
3
) 1; R , R ) H), affording the corresponding alkynyl ketone
(10a)15 in 65% yield. When the reaction was quenched at room
6
temperature, a large amount of 3-phenyl-2-cyclohexenone (analo-
5028
9
J. AM. CHEM. SOC. 2005, 127, 5028-5029
10.1021/ja050663m CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Scope with Respect to Nucleophilesa
to vinylogous carboxylic acid triflates. Tethered keto-alkynes can
thus be obtained efficiently by a unified and operationally simple
approach. The present reaction proceeds under mild conditions, and
the enol triflate substrates are easily prepared in one step from the
corresponding diketones. Further studies on the scope and synthetic
applications will be reported in due course.
entry
R1
−M
conditions
8
yield (%)b
Acknowledgment. This research was supported by an award
from Research Corporation and by the FSU Department of
Chemistry and Biochemistry. G.B.D. acknowledges an award from
the Oak Ridge Associated Universities. S.K. acknowledges the
Japan Society for the Promotion of Science (JSPS) for the
Postdoctoral Fellowship for Research Abroad (2004). We thank
all of these agencies for their support, and we thank Dr. Joseph
Vaughn and Dr. Umesh Goli for assistance with NMR and mass
spectrometry, respectively.
80c
1
2
3
4
5
6
7
8
9
Ph-MgBr
0 °C to rt
0 °C to rt
0 °C to rt
0 °C to rt
0-60 °C
0-60 °C
-78 °C to rt
-78 °C to rt
-78 °C to rt
-78 °C to rt
8a
8b
8c
8d
8e
8f
8a
8c
8d
8g
p-MeO-C6H4-MgBr
m-MeO-C6H4-MgBr
o-MeO-C6H4-MgBr
p-Cl-C6H4-MgBr
2-thienyl-MgBr
Ph-Li
86
57
34
61
63
93
78
57
65
c
m-MeO-C6H4-Li
o-MeO-C6H4-Li
Me-Li
10
Supporting Information Available: Experimental procedures,
characterization data, and copies of H NMR spectra for compounds 8
and 10. This material is available free of charge via the Internet at
http://pubs.acs.org.
a
Typical procedure: enol triflate 5a (0.55 mmol) in 2 mL of cold THF
1
1
was treated with R -M (0.50 mmol). All reactions were complete within
b
9
0 min. See Supporting Information for details. Isolated yield. c Average
of two runs.
Table 3. Scope with Respect to Cycloalkenonesa
References
(
1) Lee, D. G.; Chen, T. Cleavage Reactions. In ComprehensiVe Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press: Elmsford,
NY, 1991; Vol. 7, p 541.
(
2) Hill, R. K. Cope, Oxy-Cope and Anionic Oxy-Cope Rearrangements. In
ComprehensiVe Organic Synthesis; Trost, B. M., Fleming, I., Eds.;
Pergamon Press: Elmsford, NY, 1991; Vol. 5, p 785.
(
(
3) Recent review: Grubbs, R. H. Tetrahedron 2004, 60, 7117.
4) (a) Eschenmoser, A.; Felix, D.; Ohloff, G. HelV. Chim. Acta 1967, 50,
7
08. (b) Felix, D.; Shreiber, J.; Ohloff, G.; Eschenmoser, A. HelV. Chim.
Acta 1971, 54, 2896. (c) Tanabe, M.; Crowe, D. F.; Dehn, R. L.
Tetrahedron Lett. 1967, 3943. (d) Tanabe, M.; Crowe, D. F.; Dehn, R.
L.; Detre, G. Tetrahedron Lett. 1967, 3739.
(
5) (a) Weyerstahl, P.; Marschall, H. Fragmentation Reactions. In Compre-
hensiVe Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon
Press: Elmsford, NY, 1991; Vol. 6, pp 1041-1070. (b) Grob, C. A.;
Schiess, P. W. Angew. Chem., Int. Ed. Engl. 1967, 6, 1. (c) Grob, C. A.
Angew. Chem., Int. Ed. Engl. 1969, 8, 535.
a
Enol triflates 9 (0.55 mmol) in 2 mL of cold THF were treated with
(
6) (a) Woods, G. F.; Tucker, I. W. J. Am. Chem. Soc. 1948, 70, 2174. (b)
Zimmerman, H. E.; Nesterov, E. E. J. Am. Chem. Soc. 2003, 125, 5422.
7) Lee, S.; Fuchs, P. L. Tetrahedron Lett. 1991, 32, 2861.
PhLi (0.50 mmol). All reactions were complete within 90 min. See
Supporting Information for details. Isolated yield.
b
(
(
(
8) Shih, C.; Swenton, J. S. J. Org. Chem. 1982, 47, 2825.
Scheme 3. Mechanistic Hypothesis
9) Flash vacuum pyrolysis conditions for such fragmentations have been
reported: Coke, J. L.; Williams, H. J.; Natarajan, S. J. Org. Chem. 1977,
42, 2380.
(
2
10) Note that preformed hydroxy vinyl selenones of type 6 (X ) SeO Ph,
prepared in three steps from 5c) fragment upon subsequent treatment with
base (i.e., NaH): Shimizu, M.; Ando, R.; Kuwajima, I. J. Org. Chem.
1981, 46, 5246.
(
11) (a) A Practical Guide to Combinatorial Chemistry; Czarnik, A. W.,
DeWitt, S. H., Eds.; American Chemical Society: Washington, DC, 1997.
(b) Schreiber, S. L. Science 2000, 287, 1964.
gous to 7) was obtained. The six-membered enol triflate bearing a
16
(12) (a) See Supporting Information for details. (b) Hori, M.; Mori, M. J. Org.
geminal dimethyl group afforded 10b in excellent yield, indicating
that there is no significant “reverse Thorpe-Ingold effect” in the
fragmentation. Five- and seven-membered enol triflates also af-
forded the desired products (10c and 10d) in 61 and 79% yields,
respectively. Capillon (10c) is a natural product isolated from the
Japanese shrub, Artemisia capillaris.17
Our working mechanistic hypothesis (Scheme 3) includes several
guiding assumptions: (1) decomposition of intermediate C is the
rate-limiting step; (2) lithium triflate is released as dissociated ions
that subsequently combine;18 (3) increasing the ionic character of
C promotes fragmentation; and (4) the stability of the product
Chem. 1995, 60, 1480.
(13) We postulate a late transition state with significant carbonyl character.
(14) Compound 8g has also been synthesized by the Eschenmoser-Tanabe
strategy in four steps (32% overall): Harding, C. E.; King, S. L. J. Org.
Chem. 1992, 57, 883.
(
15) Compound 10a has been made previously by two methods. (a) Benzoyl-
ation of a pentynylzinc reagent: Knoess, H. P.; Furlong, M. T.; Rozema,
M. J.; Knochel, P. J. Org. Chem. 1991, 56, 5974. (b) Propargylation of
an organoiron nucleophile: Itoh, K.; Nakanishi, S.; Otsuji, Y. Bull. Chem.
Soc. Jpn. 1991, 64, 2965.
(
16) Compound 10b has been prepared in two steps (50% overall) by an
allenylstannane conjugate addition approach: Haruta, J.; Nishi, K.;
Matsuda, S.; Akai, S.; Tamura, Y.; Kita, Y. J. Org. Chem. 1990, 55, 4853.
17) (a) Harada, R. Nippon Kagaku Zasshi 1956, 77, 990; Chem. Abstr. 1957,
(
5
1, 18489. (b) Harada, R. Nippon Kagaku Zasshi 1956, 77, 1036; Chem.
ketone and of the dissociated ion pair are reflected in the transition
Abstr. 1959, 53, 21773.
state.1
3,19
Note that epoxy-hydrazones (1) fragment to afford
(18) Lithium triflate forms an intimate ion pair in THF: Saito, Y.; Yamamoto,
H.; Kageyama, H.; Nakamura, O.; Miyoshi, T.; Matsuoka, M. J. Mater.
Sci. 2000, 35, 809-812.
(19) Nonconcerted fragmentation pathways should also be considered. We thank
a referee for pointing this out.
regioisomers of the alkynyl ketones described herein. Experiments
aimed at elucidating further mechanistic details are planned.
In conclusion, we report a facile C-C bond cleavage reaction
induced by the nucleophilic addition of Grignard or lithium reagents
JA050663M
J. AM. CHEM. SOC.
9
VOL. 127, NO. 14, 2005 5029