Organic Letters
Letter
arylmetals, including moisture and functional group sensitive
Grignard reagents3d (Scheme 1b). The direct C−H bond
trifluoroethylation of (hetero)arenes has also been reported,
but a directing group or expensive trifluoroethylating reagent is
needed (Scheme 1c).11 Notably, the trifluoroethylating
reagents used in these methods, such as CF3CH2I and
CF3CH2OTs, are derived from CF3CH2Cl.10,11 Inspired by
our nickel-catalyzed cross-coupling between industrial chem-
ical ClCF2H and aryl chlorides,3e we report herein the first
example of nickel-catalyzed cross-coupling of inexpensive and
widely available (hetero)aryl bromides and chlorides with
CF3CH2Cl (Scheme 1d). This protocol has the advantages of
being inexpensive and synthetically convenient without the
preparation of arylmetal reagents and having a broad substrate
scope, including pharmaceuticals, thus providing cost-efficient
access for applications in medicinal chemistry.
7), but the use of MgCl2 as the additive could provide 3 in 71%
yield without observation of 4 and 5 (entry 8). Other additives
were also tested but were inferior to MgCl2 (for details, see the
facilitate the reduction of the [NiII] complex by Zn to generate
an active nickel species.3e,12b Finally, the optimal reaction
conditions were identified by decreasing the loading amount of
Zn to 2.2 equiv with 1.0 equiv of MgCl2 as the additive,
providing 3 in 73% yield (entry 10). Essentially, the absence of
nickel, ligand, and reductant led to no product (entries 11−13,
respectively), thus demonstrating the essential role of [Ni/L]
and Zn in promoting the reaction.
With the viable reaction conditions in hand, a wide range of
aryl bromides were examined to ascertain the substrate scope
of this protocol (Scheme 2). Generally, moderate to good
yields were obtained. Aryl bromides bearing electron-donating
or electron-withdrawing substituents underwent the current
nickel-catalyzed trifluoroethylation smoothly (10−17). The
reaction exhibited good functional group tolerance. Versatile
synthetic handles, such as formyl, enolizable ketone, ester,
nitrile, and trimethylsilyl moieties, were compatible with the
reaction conditions (13−18, 20, and 21). Notably, hydroxyl
group and boronate-containing substrates (19, 22, and 23) did
not affect the reaction efficiency and provided the correspond-
ing products efficiently. In particular, the organoboron
compatibility of this protocol provides a good platform for
subsequent derivatization, which is in sharp contrast to the
previous palladium-catalyzed trifluoroethylation reactions
using arylborons as the coupling partner,10b−d,f,g thus
demonstrating the advantage of this protocol. In addition to
the aryl bromides, aryl chlorides could also be applied to the
reaction (3, 7, 9, 10, 15−19, 22, and 24) and even higher
yields were obtained in some cases (16 and 17), while previous
reports of trifluoroethylation of aromatic compounds demon-
strate that aryl chlorides make up a class of unreactive
substrates.10b Although aryl chlorides are the inexpensive and
widely available feedstock, the reductive cross-coupling
reactions with aryl chlorides are limited and only rare examples
of nickel-catalyzed cross-coupling of aryl chlorides with alkyl
chlorides have been reported thus far.3e,14 Thereby, the current
nickel-catalyzed process provides a cost-efficient route for
application of aryl chlorides in organic synthesis.
The reaction was not restricted to aryl halides, as heteroaryl
halides were also suitable substrates. A series of quinoline-,
isoquinoline-, indole-, and benzoxazole-derived heteroaryl
bromides and chlorides furnished the corresponding trifluor-
oethylated products efficiently (26−33). However, simple
pyridine derivatives led to poor yields.
The high functional group compatibility and broad substrate
scope of this method encourage us to examine the
trifluoroethylation of a series of aryl chloride-containing
pharmaceuticals. Clofibrate 34a and fenofibrate 34b used to
treat hyperlipidemia were competent coupling partners,
providing the trifluoroethylated compounds 35a and 35b,
respectively, in high yields. Importantly, N-heterocycle-
containing substrates did not interfere with the reaction
efficiency. For instance, chlormezanone 34c and indomethacin-
derived ester 34d efficiently furnished products 35c and 35d,
respectively; even pyridine-containing H1 receptor antagonist
loratadine 34e was a competent coupling partner. Because aryl
chlorides are quite common in pharmaceuticals and natural
products, this method also provides a facile route for drug
discovery and development.
We began our studies by choosing bromobenzene 2a as the
model substrate to test the nickel-catalyzed reductive cross-
coupling with ClCH2CF3 1 (Table 1). After extensive efforts
Table 1. Representative Results for Optimization of Ni-
a
Catalyzed Reductive Cross-Coupling between 1 and 2a
b
entry
[Ni]
additive (x)
none
TMSCl (0.5)
MgCl2 (0.5)
NaI (0.5)
KI (0.5)
TMEDA (0.5)
TMEDA
MgCl2 (0.5)
MgCl2 (1.0)
MgCl2 (1.0)
MgCl2 (1.0)
MgCl2 (1.0)
MgCl2 (1.0)
M (y)
yield (%) (3/4/5)
1
2
3
4
5
6
7
8
NiBr2
NiBr2
NiBr2
NiBr2
NiBr2
NiBr2
NiBr2
NiBr2
NiBr2
NiBr2
none
Mn (3)
Mn (3)
Mn (3)
Mn (3)
Mn (3)
Mn (3)
Zn (3)
Zn (3)
Zn (3)
Zn (2.2)
Zn (2.2)
Zn (2.2)
none
32/4/0
10/4/3
2/5/0
34/4/1
46/4/0
50/10/3
14/3/0
23/4/1
71/0/0
73/0/0
0/0/0
9
10
11
12
c
NiBr2
NiBr2
3/15/0
0/0/0
13
a
Reaction conditions (unless otherwise specified): 1 (0.2 mmol, 1.0
b
equiv), 2a (1.5 equiv), DMA (2 mL), 80 °C, 12 h. Determined by
c
19F NMR using fluorobenzene as an internal standard. Reaction
conducted without ligand.
the combination of NiBr2 (10 mol %) with 4,4′-ditBu-bpy (10
mol %) in the presence of Mn (3.0 equiv) in DMA at 80 °C
could afford the desired product 3 in 32% yield (entry 1).
Under these reaction conditions, an only 4% yield of
hydrodechlorinated byproduct 4 was formed and no gem-
difluoroethylene 5 was observed. Encouraged by these results,
we examined a series of additives previously demonstrated to
have a beneficial effect on the reaction efficiency (entries 2−5).
12
TMSCl and MgCl2 was less reactive (entries 2 and 3,
respectively), and the addition of NaI13 to the reaction mixture
led to 3 in a comparable yield (entry 4). The use of KI as the
additive could improve the yield of 3 to 46% (entry 5). In
addition to KI, TMEDA also improved the reaction efficiency,
providing 3 in 50% yield, but severe formation of byproducts 4
and 5 was observed (entry 6). Switching the reductant from
Mn to Zn (3.0 equiv) dramatically diminished the yield (entry
1401
Org. Lett. 2021, 23, 1400−1405