.
Angewandte
Communications
because of the formation of the protonated compound
HCF2PO(OEt)2 and other unidentified byproducts. After
extensive investigations, 3a was still only obtained in mod-
erate yield (53%, determined by 19F NMR spectroscopy)
when phenyl bromide 1 was treated with 2 and the
[Pd(PPh3)4]/Xantphos (Pd/L = 1:2) catalytic system in toluene
at 808C.[12] To further improve the reaction efficiency, we
À
envisioned that if the oxidative addition of the C Br bond of
compound 4 to [Pd0] was feasible, then the utilization of the
moisture-stable and commercially available bromodifluoro-
methylphosphonate 4 as a coupling partner would benefit the
reaction. With this alternative strategy, protonation can be
suppressed (Scheme 1b). To the best of our knowledge, such
a palladium-catalyzed process has not been reported thus far.
Even though transition-metal-catalyzed reactions between
aryl metal species and alkyl halides are well established,[13]
similar fluoroalkylation processes of aryl metal species with
À
fluoroalkyl halides (Rf X) that are catalyzed by transition
metals have remained underdeveloped, and represent an
ongoing challenge.[10b,14]
On the basis of the above hypothesis, air-stable phenyl-
boronic acid (5a) was chosen as the coupling partner
[Eq. (1)]. To our delight, under modified reaction conditions
Scheme 2. Palladium-catalyzed phosphonyldifluoromethylation of aryl
boronic acids 5 with bromodifluoromethylphosphonate 4. Reaction
conditions (unless otherwise specified): 5 (0.3 mmol), 4 (2.0 equiv),
dioxane (2 mL), 24 h. Yields of isolated products are given.
[a] [PdCl2(PPh3)2] (5 mol%). [b] [PdCl2(PPh3)2] (10 mol%), Xantphos
(20 mol%), molecular sieves (3 ꢀ). [c] [PdCl2(PPh3)2] (7.5 mol%), Xant-
phos (15 mol%). [d] Molecular sieves (3 ꢀ) were added.
[e] [PdCl2(PPh3)2] (10 mol%), Xantphos (20 mol%).
(Scheme 1a), 3a was afforded in 16% yield by using the
[Pd(PPh3)4]/Xantphos/K2CO3 catalytic system in toluene at
808C (for details, see the Supporting Information). Encour-
aged by this result, several reaction parameters were inves-
tigated to further improve the reaction efficiency (for details,
see Supporting Information). The nature of the ligand, the
base, and the solvent were critical to the reaction efficiency,
and the use of the bidentate ligand Xantphos, K2CO3, and
dioxane was found to provide the most efficient reaction,
providing 3a in 89% yield, as determined by 19F NMR
spectroscopy [Eq. (1)]. Any changes in these three factors led
to a low reaction efficiency or no conversion. Among the
tested palladium sources, [PdCl2(PPh3)2] was also a suitable
pre-catalyst, providing 3a in good yield (76%, determined by
19F NMR spectroscopy, Eq. (1)).
a beneficial effect on some other substrates (3d, 3 f, 3h, and
3o). However, their role in the present reaction remains
elusive. Importantly, versatile functional groups (ester, thio-
ether, trimethylsilyl) were tolerated rather well (3j, 3m, 3n).
It is noteworthy that the difluoromethylated phosphonate 3o,
a protected protein phosphatase inhibitor,[15] was obtained
with high efficiency. Sterically hindered aromatic boronic
acids 5 f and 5p were also suitable substrates, providing
compounds 3 f and 3p in synthetically useful yields. Further-
more, a carbazole-derived boronic acid could also be con-
verted into the corresponding difluoromethylated product in
moderate yield (3q). However, vinyl-substituted aryl boronic
acids and vinyl boronic acids all failed to provide the desired
products.
To demonstrate the substrate scope of this method,
reactions with a variety of aryl boronic acids were examined
(Scheme 2). In many cases, [PdCl2(PPh3)2] was superior to
[Pd(PPh3)4], providing a wide range of aryldifluoromethyl-
phosphonates 3 with high efficiency. Generally, electron-rich
aryl boronic acids 5 provided the corresponding difluorome-
thylated phosphonates in high yields (3b–e, 3g–i). This is in
sharp contrast to previous results, in which the reactions of
aromatic derivatives that bear an electron-donating group
proceeded with low reaction efficiency.[8e,10c] Electron-defi-
cient substrates 5 were also smoothly transformed when 3 ꢀ
molecular sieves (MS) were employed as an additive (3j–l).
Interestingly, addition of these molecular sieves also had
After the generality of this catalytic system had been
demonstrated, the preparation of aryldifluoroacetates 7 by
the present strategy was also explored (Scheme 3). To our
delight, in the presence of a catalytic amount of CuI
(5 mol%) as a co-promoter,[16] a large variety of difluoroace-
tate derivatives 7 could be generated with this method by
employing bromodifluoroacetate (6) as the coupling partner.
Both electron-rich and electron-deficient aryl boronic acids 5
furnished the corresponding difluoroacetates in high effi-
ciency. Many versatile functional groups, such as aldehyde,
ketone, ester, thioether, amine, or silyl moieties, were all
tolerated under the reaction conditions (7i–l, 7n–p), which
highlights the advantages of the present reaction. It should be
1670
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1669 –1673