Published on Web 08/01/2006
The Selective Reaction of Aryl Halides with KOH: Synthesis of Phenols,
Aromatic Ethers, and Benzofurans
Kevin W. Anderson, Takashi Ikawa, Rachel E. Tundel, and Stephen L. Buchwald*
Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
Received June 6, 2006; E-mail: sbuchwal@mit.edu
Table 1. Pd-Catalyzed Synthesis of Phenols from Aryl Halidesa
Phenols are structural constituents of pharmaceuticals, polymers,
and naturally occurring compounds in addition to serving as
versatile synthetic intermediates.1 Traditional non-oxidative methods
for their preparation include nucleophilic aromatic substitution of
activated aryl halides and copper-promoted conversion of diaz-
oarenes, as well as benzyne protocols, all of which are limited with
respect to available starting materials and, in some cases, the
requisite harsh reaction conditions.2,3 A recent means of synthesizing
phenols is the aromatic C-H activation/borylation/oxidation se-
quence reported by Smith and Maleczka.4 This method is particu-
larly useful for the preparation of non-ortho-substituted phenols.
The palladium-catalyzed formation of C-O bonds has emerged
as an effective method for the construction of diaryl and alkyl aryl
ethers from aryl halides and phenols or aliphatic alcohols.5 The
use of bulky, monodentate ligands which facilitate C-O reductive
elimination has provided the best results.5b Interestingly, the use
of hydroxide salts (e.g., KOH, NaOH) as nucleophiles for this
process to form phenols directly has never been reported.6,7
While attempting the Pd-catalyzed coupling of water-soluble
nucleophiles with aryl halides using a solvent of aqueous 1,4-
a Isolated yields (average of two runs). b GC-yield. c In toluene/H2O (1:
dioxane and KOH as the base, we observed the formation of
significant amounts of phenol with only trace quantities of the
corresponding diaryl ether. We report herein that reactions with
catalyst systems derived from Pd2dba3 and ligands L1 or L2 (Table
1) provide an effective route for the preparation of phenols from
neutral, electron-rich, ortho-substituted and functionalized aryl/
heteroaryl bromides and chlorides.
1). d Isolated as the methyl ester. e 80 °C, 12 h.
efficient in the coupling of aryl chlorides and heteroaryl halides
with KOH providing the phenols in excellent yields. We postulate
that the increased reactivity and stability of Pd/L2 can be attributed
to a faster rate in C-O reductive elimination of its L1PdAr(OH)
species. Of importance, functional groups, such as a nitrile,
carboxylic acid, methyl ketone, and aldehyde, were well tolerated
under the reaction conditions employed.
Initial study of the reaction of 3-bromoanisole and KOH as the
nucleophile revealed that catalytic systems based on Pd2dba3 and
ligands L1 or L2 in 1,4-dioxane/water at 100 °C for 1.5 h provided
3-methoxyphenol as the exclusive product in 94% yield. While the
reaction of aryl bromides could be carried out at 80 °C, the
successful transformation of aryl chlorides generally required higher
temperatures (100 °C). Reactions could be carried out in water,
with no cosolvent (in these cases the aryl halide is the organic phase)
with results to that seen with the use of 1,4-dioxane. Toluene was
a less effective cosolvent, except in the reaction of 3-bromoben-
zonitrile where 3-hydroxybenzonitrile is formed in 80% yield. The
catalyst system based on Pd/L1 proved to be the most effective
with the di-ortho-substituted 2-bromomesitylene and 2-chloro-m-
xylene, where the catalyst derived from Pd/L2 resulted in <10%
conversion of the aryl halide. In these cases, C-O reductive
elimination (with L1) is enhanced by the ortho substitutents on the
aryl halide. In contrast, with L2 the reaction is inefficient. However,
catalyst systems based on Pd/L2 are generally more stable under
the reaction conditions (Pd-black rapidly forms using L1) and thus,
could be carried out with lower catalyst loadings (1% Pd). This
result is counterintuitive, since the bulkier ligand, L2, would be
expected to undergo more facile dissociation from Pd, leading to
decomposition of the catalyst. Also, Pd/L2 was generally more
To further exploit this new method, we developed a one-pot
protocol for the conversion of aryl halides to alkyl aryl ethers, where
the initially formed phenoxide is treated with various alkyl halides
facilitated by the phase-transfer catalyst cetyltrimethylammonium
bromide (Table 2).8 The current process represents a conceptually
different means for the synthesis of alkyl aryl ethers from aryl
halides, obviating issues of â-hydride elimination that plague the
direct addition of alkoxides using Pd-catalyzed methodology.
Further, the best results that are seen in the coupling of electron-
rich and hindered aryl halides even with primary aliphatic alcohols
require quite complex ligands.5b In contrast, L1 and L2 are, or soon
will be, commercially available. Most notably 2-chloroanisole, 2-
bromo-(isopropyl)benzene, and 4-chloroanisole are converted ef-
ficiently to their respective phenoxide, then react with secondary
alkyl halides providing alkyl aryl ether products in good yields
(Table 2, entries 2, 3, 5). The best previous results obtained for the
preparation of these types of alkyl aryl ethers, from the reaction of
aryl halides and secondary alcohols using Pd-catalyzed protocols,
provide low yields (9-46%) due to competing â-hydride elimina-
tion.5b,d Further, a wider range of alkyl aryl ethers are accessible
using this protocol (Table 2, entries 6, 7, 9). Thus, the current
9
10694
J. AM. CHEM. SOC. 2006, 128, 10694-10695
10.1021/ja0639719 CCC: $33.50 © 2006 American Chemical Society