Published on Web 05/19/2006
Three-Component Coupling of Benzyne: Domino Intermolecular
Carbopalladation
†
‡
,†
Jaclyn L. Henderson, Andrew S. Edwards, and Michael F. Greaney*
School of Chemistry, UniVersity of Edinburgh, Joseph Black Building, King’s Buildings, West Mains Road,
Edinburgh EH9 3JJ, U.K., and Organon Laboratories Ltd., Newhouse, Lanarkshire, ML1 5SH, U.K.
The introduction of 2-(trimethylsilyl)phenyl triflate, 1, as a
benzyne precursor that operates under mild reaction conditions has
enabled the development of a variety of transition-metal-mediated
aryne carbon-carbon bond forming processes.1 Intermolecular
carbopalladation, in particular, stands out as a powerful methodol-
ogy for the construction of diverse 1,2-functionalized arenes through
multicomponent coupling processes (Scheme 1). Treatment of 1
with a fluoride source, usually CsF in acetonitrile, generates benzyne
which can undergo carbopalladation with an organopalladium
species to produce the arylpalladium intermediate 3. Coupling with
a third component gives the 1,2-functionalized benzene, 4.4
Pioneering work from Yamamoto established that π-allylpalla-
dium complexes were particularly effective for intermolecular
carbopalladation of benzyne and could be employed in three-
component coupling (TCC) with alkynes to produce naphthalenes
or with a second benzyne equivalent to give phenanthrenes.5
Subsequently, Cheng demonstrated that initial carbopalladation with
π-allylpalladium chlorides could be followed by a second inter-
molecular Stille coupling with alkynyl or allenyl stannanes to
Scheme 1. Carbopalladative Three-Component Coupling of
Benzyne
-3
Scheme 2. Three-Component Coupling of Benzyne, Allyl
Chloride, and Methyl Acrylatea
a
Ratio of reactants 1:5:6 ) 1:2.4:1.2.
It was envisaged that changing the allyl chloride for a heteroallyl
halide, such as methyl bromoacetate, might improve the efficiency
of the domino carbopalladative TCC further, as benzyne/benzyne/
allyl-derived phenanthrene side product formation would no longer
be feasible. A brief catalyst screen proved this to be the case, with
benzyne, methyl bromoacetate, and methyl acrylate undergoing
6
produce 1-allyl-2-alkynylbenzenes or a second Suzuki coupling
7
to give 1-allyl-2-arylbenzenes. Larock has described the synthesis
of polycyclic aromatics using the two-component coupling of
benzyne with 2-halobiaryls.8
We are interested in developing new multicomponent benzyne
carbopalladation processes for the rapid assembly of diverse 1,2-
functionalized arenes. We wish to address two key points: first,
the current lack of versatility in the electrophile component used
in existing benzyne TCC reactions, where allyl chlorides predomi-
nate; and second, the introduction of new reactions to the second
carbon-carbon bond forming step that can better exploit the vast
array of nucleophiles available for palladium cross-coupling. This
paper presents our preliminary results in the area, describing our
development of a new benzyne TCC using carbopalladation
followed by an intermolecular Heck reaction.
2
clean reaction at 50 °C in DME using 5 mol % of PdCl dppf to
give the TCC product 12a in 75% yield as the only benzyne-derived
product (Chart 1).
The reaction proved effective for a small selection of acrylates
as well as for a methyl-substituted benzyne precursor 9. Here, the
isolation of TCC product 12c as a 1:1 mixture of regioisomers
supports the intermediacy of benzyne in the TCC mechanism, as
the unsymmetrical aryne undergoes nonselective carbopalladation
to produce equal amounts of regioisomeric o-palladium intermedi-
ates for Heck reaction.
Encouraged by the success of the double carbopalladation
reaction in TCC, we decided to introduce benzyl halides to the
reaction. The resulting phenylbenzyl products would contain a
versatile enoate group for further manipulation in addition to
displaying a variety of aryl functionality derived from commercially
available benzyl halides. Application of the optimized TCC reaction
conditions developed previously was not immediately successful.
Treatment of 1, tert-butyl acrylate, and benzyl bromide with Pd-
We began by studying the feasibility of the Heck reaction in the
TCC of allyl chloride, benzyne, and methyl acrylate using CsF in
MeCN, as shown in Scheme 2. Initial screening of catalysts revealed
5c
the formation of the phenanthrene product 8 at the expense of
TCC product 7 to be a significant problem. For example, the Pd-
(
5
OAc)
8% of 8 as the major product (based on benzyne).
The competitive formation of phenanthrenes is a major obstacle
2
/dppe catalyst system produced 7 in 28% yield along with
2
(dppf)Cl and CsF in DME produced the desired TCC benzylphenyl
to the development of any palladium-mediated benzyne TCC
strategy involving alkene components. To reduce the amount of 8
formed, we attempted to slow the rate of benzyne generation by
switching to a solvent with reduced CsF solubility. A solvent screen
established DME as the best choice, minimizing the formation of
product 15 in low yields with comparable amounts of a phenan-
threne derived from benzyne/benzyne/acrylate insertion as a difficult
to remove contaminant (Chart 2). A screen of catalyst systems
showed the bidentate ligands dppb, dppp, and xantphos to be
ineffective when used in combination with Pd(OAc)
2
. The com-
8
and producing the TCC product 7 in an encouraging 50% yield.
bination of dppe and Pd(OAc) , however, proved efficacious,
2
†
providing adduct 15a in 65% yield with much reduced levels of
the phenanthrene byproduct. It proved convenient to isolate the TCC
University of Edinburgh.
Organon Laboratories Ltd.
‡
7426
9
J. AM. CHEM. SOC. 2006, 128, 7426-7427
10.1021/ja0615526 CCC: $33.50 © 2006 American Chemical Society