Angewandte
Chemie
DOI: 10.1002/anie.201400080
À
C H Activation
Diastereoselective Carbocyclization of 1,6-Heptadienes Triggered by
À
Rhodium-Catalyzed Activation of an Olefinic C H Bond**
Christophe Aꢀssa,* Kelvin Y. T. Ho, Daniel J. Tetlow, and Marꢁa Pin-Nꢂ
Abstract: The use of a,w-dienes as functionalization reagents
for olefinic carbon–hydrogen bonds has been rarely studied.
Reported herein is the rhodium(I)-catalyzed rearrangement of
prochiral 1,6-heptadienes into [2,2,1]-cycloheptane derivatives
with concomitant creation of at least three stereogenic centers
and complete diastereocontrol. Deuterium-labeling studies and
the isolation of a key intermediate are consistent with a group-
À
directed C H bond activation, followed by two consecutive
migratory insertions, with only the latter step being diastereo-
selective.
T
he metal-catalyzed activation of otherwise inert carbon–
À
hydrogen (C H) bonds is increasingly recognized as a power-
ful synthetic method, and numerous transformations involv-
[1]
Scheme 1. Proposed catalytic cycle for a diastereoselective carbocyliza-
tion of 1,6-heptadienes into [2,2,1]-cycloheptane derivatives.
À
ing aromatic C H bonds have been described. In contrast,
besides hydrovinylation reactions using ethylene,[2] the metal-
À
catalyzed functionalization of an olefinic C H bond with
a different alkene has been reported on far fewer occasions.[3]
Specifically, besides the addition of 2-isopropenyl-pyridine to
1,5-hexadiene[3h] and the addition of various olefins to 1,3-
dienes,[2,4] the use of a,w-dienes as functionalization reagents
heptadienes, which are moderately diastereoselective, when
the two terminal olefins of the substrate are linked by
a prochiral tether.[6,7] Herein, we report that the implementa-
tion of our reaction design leads to the formation of II as
a single diastereoisomer and we propose a refinement of the
initially envisioned mechanism.
Although initial investigations with the model substrate
1a validated our reaction design, the desired compound 2a
was contaminated by 3a (Figure 1). This problem was solved
by rendering the catalyst cationic through the addition of
AgBF4.[8] Control experiments confirmed that AgBF4 alone
cannot act as the catalyst, and that phosphine-free rhodium
species are not catalytically competent. Importantly, 2a was
obtained as a single diastereomer, even in the crude reaction
mixtures. The stereochemistry of 2a was confirmed by
NOESY and by X-ray crystallography of its hydrochloride
salt (Figure 1).[9]
À
of olefinic C H bonds has not yet been explored.
In this context, we anticipated that the treatment of 1,6-
heptadiene I with a rhodium catalyst would lead to the [2,2,1]-
cycloheptane derivative II by undergoing the following
À
elementary steps (Scheme 1): a) pyridine-directed C H
bond activation, b) migratory insertion of the first terminal
olefin into the metal hydride thus generated, c) migratory
insertion of the second olefin,[5] and d) final reductive
elimination. Significantly, the success of this design would
enable a strong increase of molecular complexity, as mea-
sured by the number of sp3-hybridized stereogenic centers
created in the rearrangement. Specifically, it would contrast
with the classical metal-catalyzed cycloisomerizations of 1,6-
With these results in hand, we examined the generality of
this transformation and observed in all cases the formation of
2 as a single diastereomer in good to excellent yields upon
isolation (Scheme 2). The stereochemistry of the compounds
2 was confirmed by NMR spectroscopy and by X-ray
crystallography in the case of 2g (see the Supporting
Information). Naphthyl (1b), electron-poor and electron-
rich phenyl (1c–f), 1-oxa-3-indenyl (1g), acetal (1h), pro-
tected amine (1i), and alkyl (1j–m) groups were all tolerated.
The selectivity between 2 and 3 remained excellent (ꢀ 95:5)
except in the case of substrates having a nonbulky substituent
(R), such as a linear ether chain (1l) or a simple methyl group
(1m). Hence, the Thorpe–Ingold effect appears important for
the selective formation of 2. Further confirming this rationale,
the reaction of 1 when R is a hydrogen atom led to the
[*] Dr. C. Aꢀssa, K. Y. T. Ho, Dr. D. J. Tetlow, M. Pin-Nꢁ
Department of Chemistry, University of Liverpool
Crown Street, L69 7ZD (UK)
E-mail: aissa@liverpool.ac.uk
[**] Financial support from EPSRC (DTA studentship to K.Y. T.H.), the
University of Liverpool (Studentship to M.P.), the Leverhulme Trust
(RPG198), and AstraZeneca is gratefully acknowledged. We thank
Johnson-Matthey for the loan of rhodium salts and Dr. C. M.
Robertson for X-Ray crystallographic analysis of 2a·HCl and 2g.
Supporting information for this article is available on the WWW
ꢂ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co.
KGaA. This is an open access article under the terms of the Creative
Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly
cited.
Angew. Chem. Int. Ed. 2014, 53, 4209 –4212
ꢀ 2014 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4209