Angewandte
Chemie
DOI: 10.1002/anie.200802164
Fluorination
Palladium-Mediated Fluorination of Arylboronic Acids**
Takeru Furuya, Hanns Martin Kaiser, and Tobias Ritter*
Fluorinated organic molecules have become increasingly
important as pharmaceuticals[1] and tracers for positron-
emission tomography (PET), a powerful technology for
noninvasive molecular imaging.[2] The nucleus of choice for
PET is fluorine-18 (18F), which is typically introduced into
PET tracers through the formation of carbon–fluorine bonds
using nucleophilic fluoride (18FÀ) under harsh reaction con-
ditions.[3] The short half-life of 18F (109 minutes) requires that
carbon-fluorine bond formation occurs at a late stage in the
PET tracer synthesis, ideally as the last step. Many promising
PET tracers for imaging are currently inaccessible owing to
the lack of suitable chemistry for the general, late-stage
introduction of fluorine into complex, functionalized mole-
cules.[3,4] Herein, we present a new strategy for carbon–
fluorine bond formation that relies on the fluorination of
arylboronic acids using palladium complexes [Eq. (1)]. The
understanding of basic, biomedical, and pharmaceutical
research through molecular imaging.[5–8]
Carbon–fluorine bond formation is a challenging chemical
transformation, and no general, functional-group-tolerant
fluorination reaction of arenes is currently available for the
synthesis of complex molecules. Simple fluoroarenes are
typically synthesized by pyrolysis of diazonium tetrafluoro-
borates,[9] direct fluorination using highly reactive elemental
fluorine,[10] or nucleophilic aromatic substitution reactions of
electron-poor aromatic molecules.[11,12] Common aromatic
organometallic compounds, such as aryl lithium and aryl
Grignard reagents, can afford arylfluorides if using electro-
philic fluorine sources; however, neither aryl lithium nor aryl
Grignard reagents can be used for the late-stage fluorination
of arenes bearing electrophiles, such as aldehydes, or protic
functionalities, such as alcohols, limiting their general util-
ity.[13] Organometallic compounds with lower basicity, such as
aryl zinc halides, aryl silanes, aryl stannanes, and aryl boronic
acids, afford fluorobenzenes in less than 10% yield (see the
Supporting Information). The electrophilic fluorination of
specific carbon–hydrogen bonds of phenylpyridine deriva-
tives and related structures was reported in 2006 by Sanford
et al., and uses catalytic palladium (II) acetate and N-
fluoropyridinium salts.[14] The reaction takes advantage of a
covalently attached pyridine directing group and affords
fluorinated aryl pyridine derivatives using microwave irradi-
ation (100–1508C, 1–4 h, 33–75% yield).
A different
approach, the reductive elimination of aryl fluorides from
palladium(II) fluoride complexes, would obviate the use of
directing groups, and has been investigated over the past
decade by Grushin and Yandulov.[15,16] Carbon–fluorine bond
formation to form aryl fluorides by reductive elimination
from a palladium(II) fluoride complex has not yet been
substantiated.[16,17] In general, all methods mentioned above
cannot be employed for late-stage fluorination of structurally
complex molecules owing to either harsh reaction conditions
or limited substrate scope.
We have sought a newregiospecific, late-stage fluorina-
tion reaction of arenes that encompasses a larger substrate
scope than is currently accessible, tolerates the presence of a
variety of functional groups, is not limited to a particular class
of arenes, and is not dependent on a directing group. Our
strategy is illustrated in Equation (1), and consists of the
synthesis of newaryl palladium complexes that react with the
electrophilic fluorination reagent selectfluor[18] to afford
fluoroarenes.
reaction permits a general, regiospecific late-stage formation
of carbon–fluorine bonds in the presence of a large variety of
functional groups found in biologically active molecules.
Ultimately, we anticipate our new fluorination reaction will
provide a chemical solution for the synthesis of currently
inaccessible PET tracers to increase both knowledge and
[*] T. Furuya, H. M. Kaiser, Prof. T. Ritter
Department ofChemistry and Chemical Biology
Harvard University, 12 Oxford Street, Cambridge, MA 02138 (USA)
Fax: (+1)617-496-4591
E-mail: ritter@chemistry.harvard.edu
index.html
Our initial investigations for the design of transition-metal
complexes that afford efficient fluorination was guided by the
observation that palladium has been successfully employed in
several carbon–heteroatom bond formations,[19,20] including
carbon–fluorine bonds for specific substrates.[14,21] Addition-
ally, the development of our methodology was directed by the
[**] We thank Merck & Co. and Amgen Inc. for unrestricted support, Eli
Lilly & Co. for a Graduate Fellowship for T.F., and the Degussa
Foundation for a fellowship for H.M.K. We thank Dr. Douglas M. Ho
for X-ray crystallographic analysis.
Supporting information for this article is available on the WWW
Angew. Chem. Int. Ed. 2008, 47, 5993 –5996
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5993