Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 22, 2001 5359
hydrolytic deboronation. This resulted in consumption of the
boronic acid before the reaction could proceed to completion.
To overcome this undesirable side reaction, the use of phase
transfer reagents was examined. Addition of 0.5 equiv of sodium
dodecyl sulfate (SDS) as a phase transfer agent served to
accelerate the reaction and reduce the amount of deboronation.
For example, while reaction of 1a with 2b and 2c does not result
in complete conversion in the absence of a phase transfer agent
(Entries 6 and 8), in the presence of 0.5 equiv of SDS, they react
efficiently to generate 4b and 4c in 85 and 73% yield, respectively
(entries 7 and 9).5 More sterically hindered systems are also
compatible as exemplified by reaction of 1a with 2e to give 4e
in 86% yield. Of particular interest is the use of the 3-iodophe-
nylboronic acid 2d to produce 4d in 72% yield (entry 10),
illustrating that insertion of rhodium into the aryliodide bond is
not a competitive process. The chemoselectivity exhibited by this
system could be used to generate compounds capable of partici-
pating in subsequent metal-catalyzed coupling reactions.
Varying degrees of electron donation are tolerated in the olefin
moiety as well. For example, 1b and 1c react with 2a to produce
4b and 4g in 76 and 81% yields, respectively. We also examined
the use of nonstyrenyl olefin 1d. While complete consumption
of 1d occurred, olefin isomerization was a competing process
resulting in complex reaction mixtures and lower isolated yields
of the coupled product 4h (Entry 15).
When 2-vinylpyridine 3a was reacted under the standard
conditions, a dramatic change in reactivity was observed. Instead
of obtaining the Heck-type product, 5a was generated exclusively
arising from an addition-hydrolysis pathway (Table 1, entry 1).
This pattern was found to be general for a variety of arylboronic
acids (entries 2-5). Again, competitive insertion into the aryl-
bromide bond was not observed when 2-bromopheylboronic acid
2g was used (entry 5). In addition to the use of 3a, several other
vinyl heteroaromatics 3b-d reacted analogously to produce the
addition products 5f-h.
use up to 10 equiv of the boronic acid. When the coupling
reactions described herein are run in neat water with SDS, only
3 equiv of the boronic acids are required, indicating that this side-
process is not accelerated to a noticeable degree and does not
preclude the use of water as a solvent as may have been
anticipated.6
The mechanisms of these transfomations are the focus of
continuing study. The sequence of steps in the Heck-type additions
is particularly intriguing. If the olefin is regenerated via a
â-hydride elimination in an analogous fashion to the palladium-
catalyzed Heck reaction, the intermediacy of a rhodium hydride
must be involved. How this species can be converted to an active
catalyst capable of undergoing transmetalation with the arylbo-
ronic acids is not presently known.7 The reaction mechanism for
the addition reactions is likely similar to those proposed for the
conjugate addition reactions whereby the presence of an enolizable
nitrogen functionality allows isomerization of the C-bound
rhodium species to an N-bound form that can then undergo
hydrolysis and regenerate the active catalyst complex.8
In conclusion, we have developed an aqueous rhodium-
catalyzed coupling reaction of arylboronic acids and olefins. Use
of water as the solvent is crucial since these substrates do not
react under analogous conditions in organic solvents. Two types
of reactivity are observed. First, when styrenyl olefins are used,
the reaction proceeds to give the Heck-type products. This type
of reactivity is unprecedented in rhodium catalysis. When an
enolizable functionality is present within the aromatic ring, the
addition-hydrolysis pathway occurs to give the hydrophenylated
compounds. These results should broaden the scope of rhodium-
catalyzed addition reactions and have the potential to open the
door to new possibilities for asymmetric transformations.
Acknowledgment. We would like to thank NSERC, the ORDCF,
and the University of Toronto for valuable support of our programs. A.R.
and K.F. would like to thank NSERC for postgraduate scholarships. B.M.-
M. thanks the MEC (Ministerio de Educacio´n y Ciencia) for a predoctoral
fellowship. We also thank Takeda Chemical Industries for providing a
research leave for K.F.
The use of water as the solvent for these transformations
deserves comment. In each case, addition of other cosolvents to
the reaction mixture or use of organic solvents with the sulfonated
ligands or with triphenylphoshine resulted in either no reaction
after 15 h or less than 10% conversion. This clearly illustrates
the differences in reactivity that can be obtained by changing to
an aqueous environment. We were also gratified to learn that
competitive rhodium-catalyzed hydrolytic deboronation of the
arylboronic acid can be kept at a minimum. In previous reports
of additions to activated olefins, where organic solvents containing
as little as 10% water were used, degradation of the boronic acid
nucleophiles could become problematic and result in the need to
Supporting Information Available: Experimental details and char-
acterization data including 1H and 13C NMR, IR, and mass spectroscopy
data (PDF). This material is available free of charge via the Internet at
JA010402M
(6) Experimental evidence for the deboronation process being mediated
by water was obtained by running the reactions in D2O. In these cases, the
only product detected (in addition to the desired coupling product) was the
deboronated aryl species where the boron had been exchanged for a deuterium
atom.
(7) We have attempted to address whether the arylboronic acid is being
used as a sacrificial hydride acceptor via an oxidative addition of the Rh-H
species into the aryl-boron bond with subsequent reductive elimination of
an aryl-H compound. This was done by running the reaction in D2O. In this
case, only deuterated aryl compounds were produced from deuteriolytic
deboronation of the boronic acid. If the arylboronic acid were acting as a
hydride acceptor, 1 equiv of the aryl-H compound should have been produced.
(8) Preliminary mechanistic investigations support this hypothesis. When
the reaction is run in D2O, deuterium is incorporated quantitatively at the
benzylic position of the product adjacent to the heteroaromatic ring.
(5) Representative experimental procedure: To a mixture of [Rh(COD)-
Cl]2 (4.3 mg, 2 mol %) and TPPDS (17.5 mg, 8 mol %) in H2O (2.2 mL, 0.2
M) at room temperature was successively added phenylboronic acid (133 mg,
1.09 mmol), Na2CO3 (97 mg, 0.916 mmol), SDS (63 mg, 0.228 mmol), and
styrene (50 µL, 0.436 mmol). The reaction mixture was heated at 80 °C for
15 h. After cooling to room temperature the colored solution was poured into
Et2O (25 mL) and the reaction flask was carefully rinsed with Et2O. The
heterogeneous mixture was vigorously stirred at room temperature for 2 h.
The two phases were separated, the aqueous phase was extracted with Et2O
(3 × 30 mL), and the combined organic layers were dried over MgSO4, filtered,
and evaporated to dryness. The crude residue was purified by flash chroma-
tography.
Lautens
