Angewandte
Chemie
DOI: 10.1002/anie.200907162
À
Catalytic C F Activation
Titanium-Catalyzed C–F Activation of Fluoroalkenes**
Moritz F. Kꢀhnel and Dieter Lentz*
The thermodynamic and kinetic inertness of carbon–fluorine
bonds has proven to be a mixed blessing to mankind. Ideally
suited for practical applications, organofluorine compounds
have found their way into everyday life in chemically stable
polymers like polytetrafluoroethene (PTFE) and in modern
pharmaceutical chemistry, where fluorine substitution has
beneficial effects on the physicochemical and physiological
properties of organic molecules.[1] However, halofluorocar-
bons have also found their way into the upper atmosphere,
where their long-time persistence has had a deleterious effect
on the ozone layer and has contributed to global warming.[2]
Consequently, there is a long-standing interest in selectively
activating the inert carbon–fluorine bond, although the
number of catalytically active systems for such applications
is still very sparse.[3,4] Most of these catalysts are based on late
transition metals, but recently they have been competing with
first-row transition metals as well as with Lewis acidic main-
group species.[5,6]
As reported previously,[13] treatment of a solution of air-
stable titanocene difluoride (1) with silanes 2a–c results in the
formation of the titanium(III) hydride species 3. Upon
addition of hexafluoropropene (4), the green color of the
metal complex changes to red within minutes and 1,2,3,3,3-
pentafluoropropene (5a, b) is obtained in high yields
(Scheme 1).
Scheme 1. Hydrodefluorination of hexafluoropropene (4) to give (Z)-
pentafluoropropene (5a) and (E)-pentafluoropropene (5b);
R3SiH=Ph2SiH2 (2a), PhSiH3 (2b), poly(methylhydrosiloxane) (PMHS,
2c).
The few examples of catalytic activation of fluoroalkenes
have a common drawback, as they are either costly or slow
and inefficient in terms of turnover frequencies (TOFs) and
turnover numbers (TONs), not to mention that the high
sensitivity of most catalysts limits possible practical applica-
tions.[7] Complexes of the less expensive Group 4 metals are
reactive towards fluoroalkenes, as shown by Jones et al. in
stoichiometric hydrodefluorination (HDF) employing zirco-
nium and hafnium hydrides.[8] However, C–F activation
catalyzed by Group 4 metals is known only for fluoroarenes.[9]
We were surprised to find that titanium, despite its broad
application in homogeneous catalysis, has rarely been used in
C–F activation.[10] In fact only two examples of titanium-
catalyzed C–F activation can be found in the literature.[11]
Richmond et al. have shown that titanocene dihalides under
reducing conditions can catalytically defluorinate perfluori-
nated cycloalkanes to give perfluoroarenes and their HDF
products. Herein, we report the first titanium-catalyzed HDF
of fluoroalkenes to give hydrofluoroalkenes at room temper-
ature as part of our studies on the hydrometalation of
fluorinated substrates.[12]
The reaction is very fast: TOFs of up to 26 minÀ1 have
been observed at 208C; it even proceeds at À258C with a
TOF of 0.1 minÀ1, and more than 125 turnovers are possible
(Table 1). These values dramatically exceed the scarce
published data on comparable reactions:
A TOF of
0.05 minÀ1 at 1008C and a TON of less than 10 have been
reported for the HDF of 4 employing a b-diketiminate
iron(II) fluoride catalyst;[7g] a TOF of 0.2 minÀ1 at 358C and a
TON of 8 were observed for the related HDF of fluoro-
ethylene using Wilkinsonsꢀs catalyst.[7e] However, the recently
published rhodium-catalyzed functionalization of 4 to give
fluoroalkyl boronates achieves a TOF of 12.5 minÀ1 at room
temperature with a TON of up to 250.[7a] The high chemo-
selectivity of the titanium catalyst is demonstrated by the
absence of any 1,1,3,3,3-pentafluoropropene (6) in the
reaction mixture.
To expand the scope of titanium-catalyzed C–Factivation,
we subjected the commercially relevant 1,1,3,3,3-pentafluor-
opropene (6) and 3,3,3-trifluoropropene (8) to similar reac-
tion conditions (Scheme 2). Hydrodefluorination of 6 pro-
ceeded smoothly at room temperature, leading to 1,3,3,3-
tetrafluoropropene (7a,b) and 1,1,3,3-tetrafluoropropene
(7c). The reaction was significantly slower with a TOF of
0.69 minÀ1, but highly selective leading to predominantly the
E-isomer 7a (90%) along with small amounts of the Z-isomer
7b (6%) and 7c (4%; Scheme 2).
Although 8 does not contain any olefinic fluorine
substituents, HDF is possible, but the reaction proceeds less
smoothly. A drastically lowered TOF of 0.04 minÀ1, formation
of large amounts of the hydrogenation product 9b, and the
generation of the secondary HDF products 9c,d prior to
complete consumption of the starting material make this
[*] M. F. Kꢀhnel, Prof. Dr. D. Lentz
Freie Universitꢁt Berlin
Institut fꢀr Chemie und Biochemie, Anorganische Chemie
Fabeckstrasse 34–36, 14195 Berlin (Germany)
Fax: (+49)30-838-52440
E-mail: lentz@chemie.fu-berlin.de
[**] We thank Solvay Germany for donating hexafluoropropene and
Hoechst for donating trifluoropropene. This work was supported
financially by the Deutsche Forschungsgemeinschaft as part of the
graduate school program GRK 1582/1 “Fluor als Schlꢀsselelement”.
We thank S. Matthies, D. Nitsch, and. M. Sparenberg for their
contributions.
Angew. Chem. Int. Ed. 2010, 49, 2933 –2936
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2933