Organic Letters
Letter
a
Table 1. Optimization of the Reaction Conditions
Scheme 2. Substrate Scope of Bromotrifluoromethylation of
Unactivated Alkenes
a
Cu
catalyst
yield
b
entry
pyridine derivative Br source additive
(%)
1
2
3
4
5
6
7
8
9
CuI
CuBr
pyridine
pyridine
pyridine
pyridine
NaBr
NaBr
NaBr
NaBr
NaBr
NaBr
NaBr
NaBr
n-Bu4NBr
NaBr
NaBr
NaBr
NaBr
―
―
−
−
−
−
−
−
−
−
0
0
0
5
trace
trace
16
0
CuOTf
CuBr2
CuBr2
CuBr2
CuBr2
CuBr2
CuBr2
CuBr2
CuBr2
CuBr2
CuBr2
CuBr2
CuBr2
2-fluoropyridine
4-phenylpyridine
2,2′-bipyridine
f
phen
2,2′-bipyridine
2,2′-bipyridine
2,2′-bipyridine
2,2′-bipyridine
2,2′-bipyridine
2,2′-bipyridine
2,2′-bipyridine
−
6
10
11
Zn
Fe
Zn
Zn
Zn
Zn
34
18
60
62
55
78
c
12
d
13
14
c
c e
,
15
a
Reaction conditions: 1a (0.2 mmol), Tf2O (0.6 mmol), Br source
(0.4 mmol), Cu catalyst (0.02 mmol), pyridine derivative (0.6 mmol),
b
additive (0.2 mmol), DCE (2.0 mL), N2, rt, overnight. Yields were
determined by 19F NMR spectroscopy using trifluoromethylbenzene
c
d
as an internal standard. CuBr2 (0.1 mmol). CuBr2 (0.16 mmol).
e
f
Tf2O (0.8 mmol), 2,2′-bipyridine (0.8 mmol). phen = 1,10-
phenanthroline.
desired product 2a was detected in the presence of Cu(I)
catalysts, including CuI, CuBr, and CuOTf (entries 1−3). 2a
was formed in 5% yield when CuBr2 was used as the catalyst
(entry 4). The screening of a variety of pyridine derivatives
indicated that 2,2′-bipyridine was optimal, giving 2a in 16%
yield (entries 5−8). We reasoned that 2,2′-bipyridine probably
was used as both the activating reagent of Tf2O and the ligand
to coordinate to CuBr2. The reaction efficiency was not
improved when n-Bu4NBr instead of NaBr was used as
bromide source (entry 9). Subsequently, the yield of 2a was
increased in the presence of Zn or Fe powder as a reductant
(entries 10 and 11), with Zn leading to a higher yield. The
yield was improved to 60% by increasing the amount of CuBr2
to 0.5 equiv (entry 12). However, a further increase in
equivalents of CuBr2 had little effect on the yield (entry 13).
Notably, even without NaBr, 2a was formed in 55% yield
(entry 14). This result showed that CuBr2 played a dual role as
the catalyst and bromide source. Lastly, the yield of 2a was
improved to 78% when larger amounts of Tf2O and 2,2′-
bipyridine were employed (entry 15).
Having optimized the reaction conditions, we then explored
the substrate scope of this copper and zinc copromoted
bromotrifluoromethylation reaction (Scheme 2). A variety of
unactivated alkenes (1a−o) underwent this reaction smoothly
to give the bromotrifluoromethylated products (2a−o) in
moderate to excellent yields. The reaction of 1a was easily
scaled up to 1.0 mmol. The mild conditions tolerated a range
of functional groups, including ether, ester, amide, sulfonate,
sulfamide, chloro, bromo, and nitro. 1,1-Disubstituted alkene
1m was efficiently converted to the desired product 2m in
good yield. However, internal alkenes were not suitable
substrates. Alkenes derived from biologically relevant com-
a
Reaction conditions: 1 (0.4 mmol), Tf2O (1.6 mmol), 2,2′-
bipyridine (1.6 mmol), CuBr2 (0.2 mmol), Zn (0.4 mmol), DCE
(4.0 mL), N2, rt, overnight. Isolated yields are shown. The reaction
was performed on a 1.0 mmol scale.
b
pounds including, dicamba (1n) and estrone (1o), were also
compatible with the reaction conditions. In the case of
substrate 1o containing a cyclic ketone, electrophilic triflation
also took place to form enol triflate 2o.
To expand the scope of the method, styrenes and alkynes
were tested under the standard reaction conditions (Table 1,
entry 15). Styrenes 1p−t bearing different substituents on the
benzene ring furnished the corresponding products 2p−t in
moderate yields (Scheme 3). Bromotrifluoromethylation of
heterocyclic substrates such as 2-vinylthiophene and 2-
vinylpyridine failed to provide any of the desired products.
Furthermore, terminal and internal alkynes 3a−d were
smoothly converted to substituted CF3-containing alkenes
4a−d with excellent regio- and stereoselectivities (Scheme 4).
Subsequently, this copper and zinc copromoted protocol
was successfully extended to chlorotrifluoromethylation
reactions11 by replacing CuBr2 with CuCl2 and n-Bu4NCl
Scheme 5, alkenes 1a−o underwent this reaction smoothly to
give the chlorotrifluoromethylated products 5a−o in good
yields. Unfortunately, the analogous fluoro- and iodotrifluor-
omethylation reactions were unsuccessful.
Mechanistic experiments were then performed to gain
insight into the reaction mechanism. The reaction of 1a was
completely blocked when 2,2,6,6-tetramethylpiperidin-1-oxyl
(TEMPO) was added to the standard conditions, whereas the
347
Org. Lett. 2021, 23, 346−350