.
Angewandte
Communications
DOI: 10.1002/anie.201209569
Gold Catalysis
Rational Design of a Gold Carbene Precursor Complex for a Catalytic
Cyclopropanation Reaction**
David H. Ringger and Peter Chen*
We report herein the rational design of an isolable gold
carbene precursor complex and its reactivity in stoichiometric
and catalytic cyclopropanation reactions. This work is another
example of combined gas- and solution-phase studies for the
rational discovery of new catalytic transformations.[1]
Cyclopropane motifs play a prominent role in chemistry
and biology, as they are present in natural products and are
found in biologically active compounds, such as pharmaceut-
ical agents.[2] Cyclopropanes are interesting building blocks
owing to their rigid structure and geometric similarities to
olefins.[3] The challenging formation of highly strained cyclo-
alkanes[4] can be accomplished using transition metals,[3,5]
including gold-promoted carbon–carbon bond-forming reac-
tions.[6] Most of these transformations presumably involve
a gold carbene exhibiting carbocationic character.[7]
Our group has recently reported the first example of the
mass spectrometric (ESI-MS/MS) detection of a cationic N-
heterocyclic carbene (NHC) gold(I) benzylidene complex,
generated from a phosphonium ylide.[8] Gas-phase character-
Scheme 1. Design, synthesis, and fragmentation of gold carbene
precursor 2.
ization experiments have shown that these carbenes cyclo-
propanate electron-rich olefins.[8,9] This reactivity was not
observed in solution under thermal conditions. This was
attributed to a too-slow formation of gold benzylidene
compared to other decay products, although the carbon–
phosphorus bond was the first bond to break under collision-
induced dissociation (CID) conditions in the gas phase.
Gai et al. have used sulfone-based compounds in nick-
el(II)-catalyzed cyclopropanation reactions. However, the
procedure requires slow addition of methyllithium to the
refluxing reaction mixture. This complicates the reaction
setup and limits the scope to base-tolerant olefins. Further-
more, the exact mechanism of these transformations and the
nature of the intermediates remain unresolved.[10] Other
Based on these findings, we envisioned that appropriate
modifications on the gold alkyl complex would give access to
an isolable gold(I) carbene precursor complex 2 (Scheme 1),
which would cyclopropanate olefins.[6a,12] The isolation of such
a gold carbene precursor complex would not only allow for
full characterization but also produce insights concerning the
nature of gold carbenes and their cyclopropanation reaction
mechanism.[7,13] Lastly, we sought to eliminate the need for
addition of stoichiometric amounts of a strong base.
We reasoned that an electron-rich NHC ligand[14] and
a methoxy substituent in the para position on the phenyl ring
would stabilize the carbocationic gold carbene 3 and thus
facilitate its formation. The SO2-imidazolium moiety acts as
a leaving group, which can dissociate to give SO2 and
imidazolylidene 6. The latter, produced in situ, serves as
a base to deprotonate another molecule of imidazolium
sulfone salt 1 in the putative catalytic cycle (Scheme 2).
Furthermore, the imidazolium moiety carries a charge,
making it possible to manipulate the gold carbene precursor
2 not only in solution, but also in the gas phase. The presence
of the SO2 moiety is crucial as it: 1) acidifies the a-position in
the imidazolium sulfone salt 1 owing to electronic stabiliza-
tion; and 2) leaves the reaction as a gas that does not interfere
reports have shown that sulfone and structurally similar
I
À
compounds are capable of ligating Ph3P Au to give stable
gold alkyl complexes.[11] The complexes were, however,
unreactive in cyclopropanation.
[*] D. H. Ringger, Prof. Dr. P. Chen
Laboratorium fꢀr Organische Chemie
Eidgençssische Technische Hochschule (ETH)
Zꢀrich (Switzerland)
E-mail: peter.chen@org.chem.ethz.ch
[**] The support from ETH Zꢀrich and Swiss Nationalfonds is gratefully
acknowledged. We thank: Dr. Bernd Schweizer for the crystal
structure determination, Dr. Daniel Serra and Dr. Tim den Hartog
for useful discussions, and Armin Limacher and Dino Wu for
synthetic support.
with further steps.[15] The exact order of the C1 S1 and C2 S1
bond cleavage (Figure 1) is still unclear and could not be
determined through gas- or solution-phase experiments.
However, preliminary theoretical density functional theory
(DFT) calculations suggest that either both bonds are cleaved
À
À
Supporting information for this article is available on the WWW
4686
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 4686 –4689