Mechanistic Insight into Copper-Mediated Trifluoromethylation of Aryl Halides: The Role of CuI

Article abstract of DOI:10.1021/jacs.1c07408

The synthesis, characterization, and reactivity of key intermediates [Cu(CF3)(X)]-Q+ (X = CF3 or I, Q = PPh4) in copper-mediated trifluoromethylation of aryl halides were studied. Qualitative and quantitative studies showed [Cu(CF3)2]-Q+ and [Cu(CF3)(I)]-Q+ were not highly reactive. Instead, a much more reactive species, ligandless [CuCF3] or DMF-ligated species [(DMF)CuCF3], was generated in the presence of excess CuI. On the basis of these results, a general mechanistic map for CuI-promoted trifluoromethylation of aryl halides was proposed. Furthermore, on the basis of this mechanistic understanding, a HOAc-promoted protocol for trifluoromethylation of aryl halides with [Ph4P]+[Cu(CF3)2]- was developed.

Full text of DOI:10.1021/jacs.1c07408

pubs.acs.org/JACS
Article
Mechanistic Insight into Copper-Mediated Trifluoromethylation of
Aryl Halides: The Role of CuI
He Liu, Jian Wu, Yuxuan Jin, Xuebing Leng, and Qilong Shen*
Cite This: https://doi.org/10.1021/jacs.1c07408
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: The synthesis, characterization, and reactivity of key
intermediates [Cu(CF3)(X)]−Q+ (X = CF3 or I, Q = PPh4) in
copper-mediated trifluoromethylation of aryl halides were studied.
Qualitative and quantitative studies showed [Cu(CF3)2]−Q+ and
[Cu(CF3)(I)]−Q+ were not highly reactive. Instead, a much more
reactive species, ligandless [CuCF3] or DMF-ligated species
[(DMF)CuCF3], was generated in the presence of excess CuI.
On the basis of these results, a general mechanistic map for CuI-
promoted trifluoromethylation of aryl halides was proposed.
Furthermore, on the basis of this mechanistic understanding, a
HOAc-promoted protocol for trifluoromethylation of aryl halides with [Ph4P]+[Cu(CF3)2]− was developed.
plexes.13 More recently, copper-catalyzed trifluoromethylation
INTRODUCTION
■
of aryl halides under mild conditions was also reported.8c,14
Even though widespread applications of this strategy in the
preparation of trifluoromethylated drug candidates have been
achieved, mechanistic studies,15 especially the experimental
elucidation of the structures of active [CF3Cu] species in
copper-mediated/-catalyzed trifluoromethylation approaches,
lag far behind. In 1986, the group of Burton reported the first
spectroscopic studies of trifluoromethylcopper(I) species in
this process (Scheme 1A).15 A characteristic peak with a
chemical shift of −28.8 ppm (A) in the 19F NMR spectrum
was generated upon mixing a solution of trifluoromethylzinc/
cadmium in DMF/HMPA with copper(I) salt at −50 °C. In
the absence of HMPA, the peak disappeared at room
temperature, and two new peaks at −32.3 ppm (B) and
−35.5 ppm (C) in DMF were formed (Scheme 1A). It was
found that both (A) and (B) were able to react with aryl
iodides while species (C) was not active at all. Later on, these
signals were proposed to be [(CF3)Cu]·L (L = metal halide),
[(CF3)2Cu]−, and [(CF3)4Cu]−.16 In 2008, Kolomeitsev and
co-workers reported that similar [CF3Cu] species were
generated when a mixture of CuBr, Me3SiCF3, and KF in a
mixed solvent of DMF/DMI was stirred at 0 °C. In this case,
these species were assigned as CF3Cu·KBr (−28.8 ppm in
DMF/DMI), K+[Cu(CF3)2]− (−32.4 ppm in DMF/DMI),
Trifluoromethylated (hetero)arenes are privileged structural
motifs in a considerable number of prescribed drugs and drug
candidates in the pipelines of the pharmaceutical companies,1
owing to the unique steric and electronic properties of the
trifluoromethyl group that might allow us to fine tune the
molecule’s biological properties including metabolic stability
and lipophilicity.2 Consequently, in the past decade, a wide
variety of methods including transition-metal-catalyzed cross-
coupling reactions or C−H trifluoromethylation that are
capable of late-stage incorporation of the trifluoromethyl
group into (hetero)arenes have emerged.3 More specifically,
copper-mediated/-catalyzed trifluoromethylation of aryl hal-
ides represents one of the most practical approaches for the
construction of the trifluoromethylated (hetero)arene units,
largely ascribing to the copper catalyst’s high reactivity, low
cost, mild conditions, and broad functional group compati-
bility.4
The first copper-mediated trifluoromethylation of aryl
iodides by using CF3I as the trifluoromethyl source was
reported by McLoughlin and Thrower5 and later on by
Kobayashi and Kumadaki in 1969.6 Mechanistically, it was
proposed that reaction of copper powder with CF3I generates
[CF3Cu], which then reacts with aryl iodide to give the
corresponding CF3Ar, even though the identity of [CF3Cu]
was not determined. From then on, a plethora of approaches
for the generation of active [CF3Cu] species were developed,
Received: July 18, 2021
including using trifluoromethyl metal reagent (MCF3),7
a
combination of fluoride with in situ generated difluorocarbene
from difluorocarbene precursors,8 Ruppert−Prakash reagent
(Me3SiCF3),9 CF3CO2Na,10 fluoroform (HCF3),11 etc.,12 as
well as a few well-defined trifluoromethylcopper(I) com-
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
© XXXX American Chemical Society
A

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Scheme 1. Previous Efforts in Elucidation the Key Intermediates in Copper-Mediated Trifluoromethylation of Aryl Halides
and an organocopper(III) species K+[Cu(CF3)4]− (−35.7
ppm) (Scheme1B).17 Likewise, Yagupolskii and co-workers
also reported, in 2010, that similar signals in the 19F NMR
spectrum that corresponded to the [CF3Cu] species were
observed from treatment of CuBr with a solution of
Zn(CF3)Br·DMF in a molar ratio of 2:1 in DMF at room
temperature. In this work, the signals were assigned as CuCF3
(−32.0 ppm in DMF) and Cu[Cu(CF3)2] (−28.4 ppm in
DMF) and Cu[CuIII(CF3)4] (−35.1 ppm in DMF) (Scheme
1c).7b Notably, while the assignment of the signal at −35.1
ppm in 19F NMR spectra of the [CF3Cu] species from three
groups was consistent, the assignment of the other two signals
was controversial, and the unambiguous structures of these
species remained elusive. Moreover, Goossen and co-workers
reported that when equimolar CuI and K+[CF3B(OMe)3]− in
DMF was stirred at room temperature, a species with a
chemical shift of −28.1 ppm was generated, which was
assigned as [Cu(CF3)(I)]−. Nevertheless, again, this species
was not isolated and characterized.14b
To elucidate the structure of the observed [CF3Cu] species,
several groups have tried to characterize the proposed structure
by independent synthesis. In 2008, Vicic and co-workers
reported the synthesis and characterization of a well-defined
cuprate complex, [(SIMes)2Cu]+[Cu(CF3)2]−, with a chemical
shift (−31.5 ppm in THF) close to those observed previously
(Scheme 1D).13a However, [(SIMes)2Cu]+[Cu(CF3)2]− re-
acted with aryl iodides sluggishly, and it was converted to more
active neutral Cu(I) species [(SIMes)Cu(CF3)] that reacted
with aryl iodides efficiently (Scheme 1).18 In 2019, the groups
trifluoromethyl (hetero)arenes in good to excellent yields.20
́
DFT calculation by Perez-Temprano and co-workers proposed
that the active copper species in the reaction was [(DMF)-
Cu(CF3)], which was generated by thermal decomposition of
[Cu(CF3)(I)]−.19
Despite these advances in attempting to elucidate the active
species in the copper-mediated/-catalyzed trifluoromethylation
of aryl halides, several questions remain. (1) Is the species at
−28.4 ppm in DMF [Cu(CF3)(I)]−? If so, is it the active
copper(I) species that promotes the trifluoromethylation with
aryl iodides? (2) Is the species at −32.4 ppm in DMF/DMI
[Cu(CF3)2]− since previous studies showed that it is not
reactive to aryl iodides? If it is not the active species, how is it
converted into the active species? (3) If both [Cu(CF3)(I)]−
and [Cu(CF3)2]− are not the active species in the reaction,
what is the active species? (4) What is the role of CuI in
promoting the trifluoromethylation with [Cu(CF3)2]−?
Herein, we report mechanistic studies of the copper-
mediated trifluoromethylation of aryl iodides and our
discoveries, including: (1) [Ph4P]+[Cu(CF3)(I)]− was pre-
1
pared and fully characterized, including H and 19F NMR
spectroscopic study, X-ray diffraction of its single crystals, and
elementary analysis. (2) 19F NMR spectroscopic study showed
that [Cu(CF3)(I)]− and [Cu(CF3)2]− were the species
previously observed with a chemical shift at −28.8 and
−32.3 ppm, respectively. (3) Both [Cu(CF3)(I)]− and
[Cu(CF3)2]− are not the most active trifluoromethylation
species since stoichiometric reactions of both species with
electron-rich aryl iodides gave the trifluoromethylated arenes
in moderate yield even at 100 °C. However, kinetic studies
showed that the reactivity of [Cu(CF3)2]− was about 2.3 times
higher than that of [Cu(CF3)(I)]−. These reactivity studies
also suggested that the active species was not generated from
thermal decomposition of [Cu(CF3)(I)]−. (4) The most active
Cu(I) species was determined to be ligandless [Cu(CF3)] or
DMF-ligated [(DMF)Cu(CF3)], which was formed by
reaction of [Cu(CF3)2]− with CuI, thus elucidating the role
of CuI in the reaction. (5) Finally, an acetic acid promoted
trifluoromethylation of aryl iodides by [Ph4P]+[Cu(CF3)2]−
was developed and various (hetero)aryl halides were
́
of Perez-Temprano and Shen independently reported the
preparation, characterization, and reactivity studies of crystal-
line ionic cuprate complexes Cs+[Cu(CF3)2]− (−31.1 ppm in
DMF)19 and [Ph4P]+[Cu(CF3)2]− (−31.4 ppm in THF or
CDCl3).20 It was found that both complexes reacted with aryl
iodides slowly at 50 or 100 °C (Scheme 1D). However, these
reactions were accelerated dramatically when 1 equiv of CuI
was added, and the scope of the reaction investigated by Shen’s
group showed that not only aryl iodides but also active
heteroaryl bromides or aryl chlorides with an ortho-ester or
-nitro directing group reacted to give the corresponding
B
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
trifluoromethylated in good to high yields under acidic
conditions, which represents the first copper-mediated
trifluoromethylation of aryl halides under acidic conditions.
[Cu(CF3)(I)]− and [Cu(CF3)2]−, as well as CF3H (−79
ppm in THF). was observed (method A in Scheme 2). The
failure in the isolation of the desired [Cu(CF3)(I)]− by
reaction of CuI with excess TMSCF3 and KF led us to consider
an alternative strategy by transmetalating one of the
trifluoromethyl groups from [Cu(CF3)2]− to CuI since
previously we discovered that the addition of CuI dramatically
accelerated the trifluoromethylation reaction. It was found that
stoichiometric reaction of [Ph4P]+[Cu(CF3)2]− and CuI in
DMF at 100 °C after 1 h generated a new peak at −27.0 ppm
in 53% yield. However, elongating the reaction time or
changing the ratios of two reactants did not lead to full
conversion of [Ph4P]+[Cu(CF3)2]−. Consequently, efforts to
isolate [Cu(CF3)(I)]− from the mixture failed (method B in
Scheme 2).
RESULTS AND DISCUSSION
■
Preparation and Characterization of Trifluoromethyl
Organocuprate(I) [Ph4P]+[Cu(CF3)(X)]− (X = Cl, I, or CF3).
In 2019, we reported the preparation of [Ph4P]+[Cu(CF3)2]−
(1a) by reaction of CuCl with excess Me3SiCF3 (3.5 equiv) in
the presence of excess KF (6.5 equiv) in THF, followed by
cationic exchange with [Ph4P]+Cl− at room temperature.20 It
was found that the reaction time (30 min) was key for the
high-yielding formation of the complex 1a. Carefully
monitoring the reaction by 19F NMR spectroscopy showed
that a peak at −28.0 ppm in THF was initially formed within
5.0 min. Upon stirring the suspension at room temperature for
another 15 min, the peak at −31.8 ppm in THF, which
corresponded to [Cu(CF3)2]−, emerged, suggesting that we
might be able to isolate the species at −28.0 ppm by quenching
the reaction within 15 min. Indeed, by controlling the reaction
time to 13 min, the species at −28.0 ppm was isolated as
[Ph4P]+[Cu(CF3)(Cl)]− (1b)21 after recrystallization, albeit in
low 7% yield (Scheme 2). The structure of [Cu(CF3)(Cl)]−
was unambiguously confirmed by X-ray crystallography of its
single crystals (see the Supporting Information for details).
Previously, Vicic and co-workers reported that even though
the reactivity of [Cu(CF3)2]− was not high, it reacted with aryl
iodides slowly.18 We thus envisaged that pure [Cu(CF3)(I)]−
could be prepared by consumption of [Cu(CF3)2]− in the
mixture via its reaction with an appropriate aryl halide. As
shown in Figure 1, addition of 1.0 equiv of iodobenzene to the
Scheme 2. Strategies for the Preparation of Ionic Cuprate
Cu(I) Complex [Ph4P]+[Cu(CF3)(I)]− 1c
Figure 1. (a) Monitoring the stoichiometric reaction of [Ph4P]+[Cu-
(CF3)2]− (1a) and CuI in DMF at 100 °C by 19F NMR spectroscopy.
(b) Monitoring the stoichiometric reaction of [Ph4P]+[Cu(CF3)2]−
(1a) and CuI in the presence of 1.0 equiv of iodobenzene in DMF at
100 °C by 19F NMR spectroscopy. (c) Monitoring the stoichiometric
reaction of [Ph4P]+[Cu(CF3)2]− (1a) and CuI in the presence of 1.0
equiv of 2-bromopyridine in DMF at 100 °C by 19F NMR
spectroscopy.
reaction mixture led to the consumption of both species.
Nevertheless, addition of 1.0 equiv of 2-bromopyridine
resulted in the complete conversion of [Cu(CF3)2]− into the
corresponding 2-trifluoromethylated pyridine after 1.0 h at 100
°C, while [Cu(CF3)(I)]− remained intact under these
conditions (method C in Scheme 2). Further optimization of
the reaction conditions enabled us to isolate the well-defined
organocuprate(I) complex [Ph4P]+[Cu(CF3)(I)]− (1c) in 13%
yield after recrystallization, for the very first time, by treating
[Ph4P]+[Cu(CF3)2]− with 1.0 equiv of CuI and 2.0 equiv of 2-
bromo-6-trifluoromethylpyridine in DMF at 100 °C for 2 h
(Scheme 3).
Considering that one of the potentially active species with a
chemical shift of −28.8 ppm in THF in the copper-mediated
trifluoromethylation of aryl iodides was proposed to be
[Cu(CF3)(I)]−, and the only difference between this species
and complex 1b is the anionic halogen ligand, we therefore
focused our initial efforts to synthesize [Ph4P]+[Cu(CF3)(I)]−
by employing the same strategy. Unfortunately, our attempts
to synthesize [Ph4P]+[Cu(CF3)(I)]− by reaction of CuI with
excess TMSCF3 and KF in various solvents at room
temperature or at low temperature failed. It was found that
CF3H (−79 ppm in THF) was observed as the major product
by monitoring of the reactions by 19F NMR spectroscopy when
THF was used as the solvent. In other solvents such as DMF
or DMSO, the formation of an inseparable mixture of
Ionic cuprate complex [Ph4P]+[Cu(CF3)(I)]− (1c), a white
crystalline solid, is sensitive to air and moisture. A solution of
1c in CDCl3 has a characteristic peak with a chemical shift at
C
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
NMR spectroscopy should overlap and show a single peak.
Otherwise, we would observe two different signals. As shown
in Figure 3, upon treatment of CuI (1.2 equiv) with Me3SiCF3
(1.0 equiv) and KF (1.0 equiv) in DMF/DMI (v/v = 10:1) at
0 °C for 15 min, two proposed species, CF3Cu·KX and
[Cu(CF3)2]− with a chemical shift at −28.9 and −31.4 ppm,
respectively, were observed by 19F NMR spectroscopy,
respectively, despite a slight deviation of chemical shifts of
the two species reported in Kolomeitsev’s work due to the
replacement of CuBr to CuI17 (Figure 3a). To the mixture was
added a solution of complex [Ph4P]+[Cu(CF3)(I)]− (1c) in
DMF at 0 °C. Clearly, a superimposed peak with a chemical
shift at −28.2 ppm in DMF/DMI was observed, indicating the
proposed CF3Cu·KX has the authentic structure of [Cu(CF3)-
(I)]− (Figure 3b). Likewise, addition of a solution of complex
[Ph4P]+[Cu(CF3)2]− 1a in DMF at 0 °C also showed a
superimposed peak at −31.4 ppm in DMF/DMI, suggesting
[Cu(CF3)2]− is another key intermediate in copper-catalyzed/
mediated trifluoromethylation (Figure 3c). These results
provided strong evidence that [Cu(CF3)2]− and [Cu(CF3)-
(I)]− are the key intermediates observed previously in copper-
mediated trifluoromethylation of (hetero)aryl halides.
Scheme 3. Optimized Reaction Conditions for the
Preparation of Ionic Cuprate Cu(I) Complex
[Ph4P]+[Cu(CF3)(I)]− 1c
−27.8 ppm in 19F NMR spectroscopy, which is identical to
that of [Ph4P]+[Cu(CF3)(Cl)]−. In DMF, the peak slightly
shifts to −27.0 ppm. The 31P NMR spectrum of complex 1c in
CDCl3 showed a singlet peak with a chemical shift of −24.3
ppm. To further establish the structure of complex 1c, single
crystals of complex 1c was obtained by slow diffusion of
methyl tert-butyl ether to a solution of complex 1c in THF at
−35 °C. X-ray diffraction studies of the single crystals showed
the bond angle of C−Cu−I is 171.4°, suggesting that the ionic
cuprate complex adopts approximately a linear geometry. The
bond length of Cu−C(CF3) in complex 1c is 2.10 Å, which is
much longer than the Cu−C(CF3) bond length in complex
[Cu(CF3)2]− (1.93 Å) and [Cu(CF3)Cl]− (1.66 Å),
respectively. The longer bond length of Cu−C(CF3) in
complex 1c than those in complexes 1a and 1b is due to the
stronger trans influence of the iodide anion than chloride and
trifluoromethyl anion (Figure 2).
Comparison of the Reactivities of [Cu(CF3)(I)]−,
[Cu(CF3)2]−, and [Cu(CF3)2]− + CuI. To evaluate the
reactivities of complexes [Cu(CF3)(I)]− and [Cu(CF3)2]−,
as well as the combination of [Cu(CF3)2]− with CuI, we
studied stoichiometric reactions of these complexes with aryl
halides with different substituents including electron-rich 4-
methoxylphenyl iodide, electron-neutral phenyl iodide, elec-
tron-poor methyl 4-iodobenzoate, and a heteroaryl bromide 2-
bromopyridine. Specifically, a set of parallel experiments were
conducted in which aryl halides were allowed to react with 1.0
equiv of cuprate complex [Cu(CF3)2]− 1a or [Cu(CF3)(I)]−
1c as well as the combination of [Cu(CF3)2]− with CuI in
DMF at 100 °C for 3.0 h. The yields of the corresponding
trifluoromethyl (hetero)arenes were then determined by 19F
NMR spectroscopy in the presence of an internal standard. As
shown in Table 1, it was found that complex [Cu(CF3)(I)]−
1c reacted in lower yields than complex [Cu(CF3)2]− 1a, while
a combination of [Cu(CF3)2]− 1a with CuI reacted in much
higher yields than both [Cu(CF3)2]− 1a and [Cu(CF3)(I)]−
1c. For instance, reaction of complex [Cu(CF3)(I)]− 1c with
4-methoxylphenyl iodide in DMF occurred in less than 5%
yield after 3 h at 80 °C and in 22% yield after 3 h at 100 °C,
while reaction of [Cu(CF3)2]− 1a with 4-methoxylphenyl
iodide gave 4-methoxy-1-trifluoromethylbenzene in 30% yield
after 3 h at 80 °C and 57% yield after 3 h at 100 °C.
Importantly, reaction of [Cu(CF3)2]− 1a with 4-methoxyl-
phenyl iodide in the presence of 1.0 equiv of CuI occurred in
much higher yields of 82% and quantitative yield, respectively,
under the same reaction conditions. A similar trend was also
observed for reactions of phenyl iodide and 2-bromopyridine.
In addition, electron-poor aryl iodides generally reacted much
faster and in higher yields than electron-rich aryl iodides, which
is consistent with previous observations in copper-mediated
trifluoromethylation of aryl halides. For instance, reaction of
electron-poor methyl 4-iodobenzoate with complex [Cu-
(CF3)2]− 1a occurred to full conversion after 3.0 h at 80 °C
to give the methyl 4-trifluoromethylbenzoate in quantitative
yield. On the contrary, reactions of electron-neutral phenyl
iodide or electron-rich 4-methyl phenyl iodide with complex
[Cu(CF3)2]− 1a occurred in less than 70% yields after 3.0 h at
100 °C. Furthermore, we also observed that reaction of 2-
Figure 2. ORTEP diagram of [Ph4P]+[Cu(CF3)(I)]− 1c. Ellipsoids
are shown at the 30% level, and selected bond lengths and bond
angles: Cu(1)−C(25): 2.111(18) Å; Cu(1)−I(1): 2.2646(14); C−
Cu−I: 171.4°.
Are [Cu(CF3)2]− and [Cu(CF3)(I)]− the Key Intermedi-
ates in the Copper-Mediated Trifluoromethylation of
Aryl Halides? Having successfully prepared and characterized
both [Cu(CF3)2]− and [Cu(CF3)(I)]−, we set out to study
whether these two species were the proposed key inter-
mediates in copper-mediated trifluoromethylation of aryl
halides. To determine whether the species is one of the
proposed key intermediates, we decided to add the isolated
cuprate complex [Cu(CF3)2]− 1a and [Cu(CF3)(I)]− 1c to a
mixture of previously proposed active copper species that was
generated from reaction of CuI and TMSCF3 in the presence
of KF and then examine the mixture by 19F NMR
spectroscopy. If complex [Cu(CF3)2]− 1a or [Cu(CF3)(I)]−
1c is the proposed key intermediate, the signals in the 19F
D
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 3. (a) 19F NMR spectrum for reaction of CuI/KF/Me3SiCF3 in DMF/DMI at 0 °C for 15 min. (b) 19F NMR spectrum of the reaction
mixture with [Cu(CF3)(I)]− 1c. (c) 19F NMR spectrum of the reaction mixture with [Cu(CF3)2]− 1a.
Table 1. Comparison of the Reactivities of [Cu(CF3)(I)]−, [Cu(CF3)2]−, and [Cu(CF3)2]− in the Presence of 1.0 equiv of CuI
a,b
with a Variety of (Hetero)aryl Halides
a
Reaction conditions: (hetero)aryl halide (0.025 mmol), Cu(I) complex (0.025 mmol), DMF (0.50 mL) at 80 or 100 °C for 3 h under argon.
Yields were determined by 19F NMR spectra with 1-fluoronaphthalene as internal standard; nd = not detected, quant = quantitative yield.
b
bromopyridine generally occurred much more slowly and gave
Quantitative Comparison of the Reactivities of
[Cu(CF3)(I)]− 1c and [Cu(CF3)2]− 1a. To assess quantitatively
the reactivities of complexes [Cu(CF3)(I)]− 1c and [Cu-
(CF3)2]− 1a, we studied the kinetics of the complex
[Cu(CF3)(I)]− 1c and [Cu(CF3)2]− 1a with 10.0 equiv of
methyl 4-iodobenzoate. More specifically, a solution of
complex [Cu(CF3)2]− 1a with 10.0 equiv of methyl 4-
iodobenzoate in DMF in a J. Young NMR tube was placed
in the NMR probe which was preheated to 60 °C, and the
progress of the reaction was monitored by 19F NMR
spectroscopy (Figure 4A). The decay of the signal correspond-
much lower yields than that of aryl iodides, likely due to the
stronger C−Br bond in 2-bromopyridine than the C−I in aryl
iodides. These qualitative reactivity studies disclosed that (1)
complex [Cu(CF3)(I)]− 1c is less reactive than complex
[Cu(CF3)2]− 1a; (2) in the presence of 1.0 equiv of CuI, the
reactivity of [Cu(CF3)2]− 1a improved significantly and a new
active species might be generated; and (3) reactions of
(hetero)aryl halides with different electron properties might
proceed via different reaction pathways.
E
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 4. (a) Plot of the decay of the concentration of [Cu(CF3)2]− 1a over reaction time for the reaction of complex[Cu(CF3)2]− 1a (0.050 mM)
with methyl 4-iodobenzoate (0.5 mM) in DMF at 60 °C. (b) Effect of initial concentration of 1a on the rate constants. (c) Plot of initial rates
versus initial concentration of methyl 4-iodobenzoate for the reaction of [Cu(CF3)2]− 1a (0.050 mM) with methyl 4-iodobenzoate in DMF at 40
°C. (d) Plot of the decay of the concentration of [Cu(CF3)(I)]− 1c over reaction time for reaction of [Cu(CF3)(I)]− 1c (0.050 mM) with methyl
4-iodobenzoate (0.5 mM) in DMF at 60 °C.
ing to complex [Cu(CF3)2]− 1a over the reaction time was
then plotted, and it was found that the concentration of
[Cu(CF3)2]− decreased exponentially, indicating a pseudo-
first-order dependence of the rate on the concentration of
complex [Cu(CF3)2]− 1a. Accordingly, the rate constant kobs1
for this reaction was determined to be (1.26 0.12) × 10−3
s−1. In addition, we found that variation of the initial
average observed rate constant kobs2 was determined to be
(3.86 0.21) × 10−4 s−1, and the rate constant of the reaction
between [Cu(CF3)(I)]− 1c and methyl 4-iodobenzoate in
DMF at 60 °C was (3.86 0.21) × 10−3 L·(mol·s)−1 (Figure
4D). Thus, a quantitative comparison of the reactivities of
complexes [Cu(CF3)(I)]− 1c and [Cu(CF3)2]− 1a revealed
that complex [Cu(CF3)2]− 1a showed a 3.3-fold higher
reactivity than complex [Cu(CF3)(I)]− 1c when they reacted
with electron-poor aryl iodides ((1.26 0.12) × 10−2 L·(mol·
s)−1 vs (3.86 0.21) × 10−3 L·(mol·s)−1).
Mechanistic Studies of Trifluoromethylation of Aryl
Iodides with [Ph4P]+[Cu(CF3)2]−. Because it was found that
reaction of electron-poor aryl iodides with [Ph4P]+[Cu-
(CF3)2]− 1a does not required the addition of CuI, we wanted
to investigate the mechanism of this reaction in more detail.
Therefore, we focused on the kinetics of the reaction of
[Cu(CF3)2]− with excess methyl 4-iodobenzoate in DMF at 60
°C. It was found that reaction of [Cu(CF3)2]− with methyl 4-
iodobenzoate gave 4-trifluoromethylbenzoate as well as
[Cu(CF3)(I)]− slowly. Because we knew that [Cu(CF3)(I)]−
reacted with methyl 4-iodobenzoate much more slowly than
[Cu(CF3)2]−, the reaction of [Cu(CF3)2]− with methyl 4-
concentration of 1a did not obviously influence the kobs1
,
thus providing further evidence that the reaction is first order
(Figure 4B). Furthermore, the initial rates for reactions of
complex [Cu(CF3)2]− 1a with methyl 4-iodobenzoate at
different concentrations at 40 °C were also studied, and it was
found that the initial rates were in a linear correlation with the
concentrations of complex [Cu(CF3)2]− 1a, thus suggesting
the reaction is also first order on the concentration of methyl
4-iodobenzoate (Figure 4C). These results illustrated that the
reaction of complex [Cu(CF3)2]− 1a with methyl 4-
iodobenzoate is a typical second-order reaction and rate
constant for the reaction in DMF at 60 °C is (1.26 0.12) ×
10−2 L·(mol·s)−1. Likewise, kinetic study for the reaction of
[Cu(CF3)(I)]− 1c with 10.0 equiv of methyl 4-iodobenzoate in
DMF at 60 °C also exhibited a first-order kinetic behavior, an
F
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
iodobenzoate could be considered as a typical continuous
reaction. Indeed, we observed the initial accumulation of
[Cu(CF3)(I)]− and then consumption slowly after reaching
the highest concentration within 25 min when the reaction of
[Ph4P]+[Cu(CF3)2]− 1a with 10.0 equiv of methyl 4-
iodobenzoate in DMF at 60 °C was monitored by 19F NMR
spectroscopy (Figure 5). Fitting the relative concentration of
Figure 6. Comparison of reactivity between [Cu(CF3)2]− and a
combination of [Cu(CF3)2]− with CuI.
°C for 15 min. The mixture was then cooled to room
temperature and analyzed by 19F NMR spectroscopy. A new
peak with a chemical shift at −27.4 ppm which was assigned as
[Cu(CF3)(I)]− and a broad peak with a chemical shift between
−25.0 and −26.5 ppm were observed. Based on the
stoichiometry of the reaction between [Cu(CF3)2]− 1a with
CuI, we suspected that the broad peak should be [CuCF3] and
the broadening of the peak is due to a rapid equilibrium of the
ligand association/dissociation process between a ligandless
[CuCF3] and a DMF-coordinated copper species [(DMF)-
CuCF3]. To probe if this is the case, we conducted low-
temperature 19F NMR experiments with the hope that the
ligand association/dissociation process could slow down at low
temperature and we would be able to observe the proposed
copper species [(DMF)CuCF3] directly. To our delight, the
peak became narrower and sharper when the temperature was
lowered from 25 °C to 0, −25, and −35 °C, respectively
(Figure 7). At −35 °C, a singlet peak with a chemical shift at
−25.8 ppm was observed. Previously, Grushin reported that a
ligandless [CuCF3] which was generated from reaction of
CuCl with fluoroform (CF3H) in DMF in the presence of 2.0
equiv of KOtBu, followed by the addition of 1 equiv of Et3N·
3HF to quench the excess KOtBu, has a chemical shift of −26.3
ppm in 19F NMR spectroscopy.11a Accordingly, we assigned
the peak with a chemical shift at −25.8 ppm we observed at
−35 °C to be [(DMF)CuCF3].
To examine whether [(DMF)CuCF3] is the active species,
we studied and compared its reactivity with those of
[Cu(CF3)(I)]− and [Cu(CF3)2]−. Reaction of a 1:1 mixture
of [Cu(CF3)2]− and CuI in DMF occurred in 56% conversion
after 1.0 h at 100 °C to generate [(DMF)CuCF3] and
[Cu(CF3)(I)]− in 53% yield, respectively. To the mixture was
added 1.0 equiv of 2-bromopyridine, and the resulting mixture
was allowed to react at 100 °C for an additional 1.0 h. It was
found that the peak at −25.8 ppm disappeared and a new peak
at −68.1 ppm appeared, which corresponded to 2-trifluor-
omethylpyridine generated in quantitative yield. In contrast,
under these conditions, [Cu(CF3)2]− was only partially reacted
while [Cu(CF3)(I)]− remained intact (Scheme 4). These
results clearly suggest that [(DMF)CuCF3] which could be
generated from the reaction of [Cu(CF3)2]− with CuI, is much
more reactive than that of [Cu(CF3)(I)]− or [Cu(CF3)2]−,
and it is the active species in the copper-mediated
trifluoromethylation reaction.
Figure 5. Plot of the relative concentration of [Cu(CF3)(I)]− from
reaction of [Cu(CF3)2]− 1a (0.050 mM) with 10.0 equiv of methyl 4-
iodobenzoate (0.5 mM) in DMF at 60 °C.
[Cu(CF3)(I)]− vs time using kinetic model for continuous
reaction gave an equation of ([[Cu(CF3)(I)]−]/[[Cu-
(CF3)2]−]0 = 1.15 [exp(−2.00 × 10−4t) − exp(−1.56 ×
10−3t)]), which allowed the extraction of the pseudo rate
constants of kobs1′ = (1.56
0.10) × 10−3 s−1 and kobs2′ =
(2.00 0.13) × 10−4 s−1 for reactions of [Cu(CF3)2]− and
[Cu(CF3)(I)]− with methyl 4-iodobenzoate, respectively.
Accordingly, the rate constants for both reactions could be
determined to be (1.56 0.10) × 10−2 L·(mol·s)−1 and (2.00
0.13) × 10−3 L·(mol·s)−1, which are within the acceptable
ranges when compared with the kobs1 and kobs2 obtained in the
previous quantitative experiments ((1.26
0.12) × 10−2 L·
(mol·s)−1 vs (3.86 0.21) × 10−3 L·(mol·s)−1, respectively).
Quantitative Comparison of the Reactivities of
[Cu(CF3)2]− 1a with a Combination of [Cu(CF3)2]−/CuI.
́
Previous studies from the groups of Perez-Temprano and Shen
independently determined that the addition of 1 equiv of CuI
would significantly accelerate the trifluoromethylation of
[Cu(CF3)2]− with aryl iodides.19,20 To quantitatively compare
the reactivity of the combination of [Cu(CF3)2]−/CuI with
[Cu(CF3)2]−, we studied the kinetics of their reactions with
electron-poor methyl 4-iodobenzoate. As shown in Figure 6,
reaction of equimolar of [Cu(CF3)2]−/CuI with 1.0 equiv of 4-
iodobenzoate occurred to full conversion of the in situ
generated [CuCF3] after 780 s at 60 °C (t1/2 = 314 s), while
reaction of [Cu(CF3)2]− with 1.0 equiv of 4-iodobenzoate in
the absence of CuI occurred to 50% conversion after 7800 s
(t1/2 = 7800 s). From these experiments, we concluded that the
reactivity of a combination of [Cu(CF3)2]−/CuI is estimated
to be at least 24 times faster than that of [Cu(CF3)2]−.
Identification of the Active Species in a Combination
of [Cu(CF3)2]− with CuI. To identify the active species in a
combination of [Cu(CF3)2]− with CuI, we carried out the
stoichiometric reaction of [Cu(CF3)2]− 1a with CuI. An
equimolar amount of [Cu(CF3)2]− 1a with CuI in DMF with
an internal standard in a J-Young NMR tube was heated at 100
G
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 7. Low-temperature 19F NMR spectra for the solution of the stoichiometric reaction between [Cu(CF3)2]− 1a and CuI after 15 min at 100
°C.
Scheme 4. Comparison of Reactivity of the in Situ
Scheme 5. Preparation and Reaction of [CuI2]− Anion
Generated [CuCF3] with [Cu(CF3)2]− and [Cu(CF3)(I)]−
equilibrium with an equilibrium constant of 2.01 was
established after 5 min (Scheme 5, eq 2). These experiments
clearly showed that the generation of CuI from [CuI2]− under
the copper-mediated trifluoromethylation conditions is un-
likely, since [CuI2]− would prefer to react with excess
[Cu(CF3)2]− in the reaction system to give less reactive
[Cu(CF3)(I)]−.
Role of [CuI2]− in Copper-Mediated Trifluoromethy-
lation of Aryl Iodides. We have now determined that even
though [Cu(CF3)(I)]− is not the active species in the copper-
mediated trifluoromethylation reactions, [Cu(CF3)(I)]− re-
acted with aryl iodides slowly to give trifluoromethylated
arenes. From the stoichiometry of the reaction [Cu(CF3)(I)]−
with aryl iodide, an ionic cuprate [CuI2]− along with ArCF3
Proposed Mechanism for Copper-Mediated Trifluor-
omethylation of Aryl Iodides. Based on our mechanistic
studies, we proposed an overall mechanistic map for the
copper-mediated trifluoromethylation of aryl iodides. First,
reaction of CuI with a trifluoromethyl source generates two
cuprates, [Cu(CF3)2]− and [Cu(CF3)(I)]−, quickly. In the
presence of excess CuI, reaction of [Cu(CF3)2]− with CuI
occurs at elevated temperature to give the active species
ligandless [CuCF3] or DMF-ligated [(DMF)CuCF3], which
then reacts with aryl iodides to give the desired trifluor-
omethylarenes and regenerate CuI (pathway A in Figure 8).
Under these conditions, pathway A is the major product-
forming pathway. Alternatively, in the absence of CuI, both
[Cu(CF3)2]− and [Cu(CF3)(I)]− react with aryl iodides but
much more slowly than that of ligandless [CuCF3] to afford
ArCF3, and pathway B is the major product-forming pathway
(pathways B and C in Figure 8). There are three points worthy
of further discussion. (1) It is known that it is difficult to
trifluoromethylate various nitrogen-containing heteroaryl hal-
ides under the copper catalyst. On the basis of our mechanistic
́
should be generated. Previously, Perez-Temprano and co-
workers proposed that [CuI2]− might undergo decomposition
to give I− and CuI, which is the activator in the reaction.19 To
clarify if this is the case, we conducted two sets of experiments.
In the first experiment, we studied the reaction of CuI with 1
equiv of nBu4NI. It was found that the reaction occurred so fast
that [nBu4N]+[CuI2]− 1d was generated after the mixture of
n
CuI and Bu4NI was shaken in CH2Cl2 for 30 s, and complex
1d was isolated as a light yellow solid in 76% yield ater
evaporation of the solvent (Scheme 5, eq 1).22 This
experiment suggests that reaction of I− with CuI occurs very
quickly, and the reverse reaction for generation of CuI from
thermal decomposition of [nBu4N]+[CuI2]− should be
unfavorable. In the second experiment, we studied the reaction
[nBu4N]+[CuI2]− with [Cu(CF3)2]−. A solution of equimolar
[Cu(CF3)2]− and [nBu4N]+[CuI2]− in DMF reacted quickly at
room temperature to generate [Cu(CF3)(I)]− until an
H
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 8. Mechanistic map for copper-mediated trifluoromethylation of aryl iodides.
map, nitrogen-containing substrates may coordinate to
ligandless [CuCF3] to form much less reactive [(substrate)-
CuCF3] and, consequently, shut down the trifluoromethylation
reaction. Under these circumstances, it would be better to
conduct the reaction that does not require excess CuI or CuI
could be fully consumed during the step for the generation of
the cuprates. (2) We found that reaction of electron-poor aryl
iodides with [Cu(CF3)2]− occurred smoothly and quickly at
50−60 °C. Therefore, for those substrates, it is not necessary
to add any activator. (3) Iodide anion is harmful for the
reaction since it is going to react with CuI to generate [CuI2]−,
which is going to further react with [Cu(CF3)2]− to give less
reactive [Cu(CF3)(I)]−.
Comments on Frequently Applied Copper-Mediated
Trifluoromethylation of Aryl Iodides. In the Introduction,
we mentioned that the reactivities of the [CuCF3] species
generated under various conditions differ significantly. Having
elucidating the [CuCF3] species and their reactivities in the
copper-mediated trifluoromethylation reaction, we could now
explain the divergent reactivities of these systems.
Trifluoromethylation Using Chen’s Reagent
(FSO2CF2CO2Me).8 Copper-mediated trifluoromethylation of
aryl halides using FSO2CF2CO2Me as the trifluoromethyl
source is another frequently applied method for the
preparation of trifluoromethyl arenes because
FSO2CF2CO2Me is cheap and available on bulk scale. To
probe the reactive species in this system, we studied the
stoichiometric reaction of FSO2CF2CO2Me and CuI in DMF
at 80 °C. 19F NMR spectroscopic study of the reaction mixture
of the stoichiometric reaction of FSO2CF2CO2Me and CuI in
DMF after 1.0 h at 80 °C showed the formation of
[Cu(CF3)(I)]−, [Cu(CF3)2]−, and [Cu(CF3)4]− in a ratio of
5.9:1.4:1, while the formation of ligandless CuCF3 was not
observed. The high ratio presence of [Cu(CF3)(I)]− suggests
that the reactivity of the reaction system is not high and
typically high temperature and long reaction time are required
for full conversion. Nevertheless, there is one advantage for this
reaction system. The iodide anion generated in the reaction
mixture would nucleophilically attack the methyl group of the
reagent to generate MeI and diminish the concentration of
[CuI2]−, which we have determined to be deleterious to the
active species in the reaction.
HOAc-Promoted Copper-Mediated Trifluoromethyla-
tion of (Hetero)aryl Halides Q+[Cu(CF3)2]− 1a. Our
mechanistic studies showed that the ligandless [CuCF3] or
DMF-ligated [(DMF)CuCF3] is the active species that
promotes the trifluoromethylation with aryl iodides, while
the initial formed copper species for the reaction of CuX with
TMSCF3 and KF are the mixture of [Cu(CF3)(I)]− and
[Cu(CF3)2]−. To overcome such a dilemma and considering
that the solubility of CuI in DMF is not high, we envisaged to
improve the reaction by the addition of an alternative additive
other than CuI that can “abstract” one of the −CF3 groups in
[Cu(CF3)2]− to generate the ligandless [CuCF3] or DMF-
ligated [(DMF)CuCF3] effectively. It is well-known that the
C−F bond in a trifluoromethylated metal species could be
easily activated by a protonic acid or a Lewis acid because of
the strong π-back donation of dπ → σ*(C−F) in the
trifluoromethylated metal species.23 We therefore chose several
Lewis acids and Brønsted acids to examine whether they could
remove one of the trifluoromethyl groups in [Cu(CF3)2]− via
abstraction of the fluorine in the trifluoromethyl group and
subsequent decomposition of the metal-bound difluorocar-
bene.24 After extensive screening, it was found that HOAc was
able to serve as the activator to effectively activate [Cu-
(CF3)2]− to generate the ligandless CuCF3. In general,
reactions of 1.5 equiv of Q+[Cu(CF3)2]− 1a and 1.5 equiv of
HOAc with a variety of (hetero)aryl iodides, bromides, and
chlorides proceeded smoothly at room temperature to 80 °C
to give the trifluoromethyalted (hetero)arenes in good yields
CuX/TMSCF3/MF System.9a Copper-mediated trifluorome-
thylation of aryl iodides using Ruppert−Prakash reagent
(TMSCF3) as the trifluoromethyl source represents one of
the most common methods for the preparation of trifluor-
omethylated arenes. The reaction using TMSCF3 for
trifluoromethylation of aryl halides was first reported by
Fuchikami and co-workers in 1991.9a Interestingly, during their
optimization of the reaction conditions, it was found that the
amount of CuI was crucial for the yield of the reaction. While
reaction using 1.0 equiv of CuI led to the formation of ArCF3
in 85% yield, the same reaction with 1.5 equiv of CuI gave
ArCF3 in quantitative yield. Our mechanistic studies now
clearly showed that excess CuI in the reaction system facilitates
the transformation of [Cu(CF3)2]− species to more reactive
ligandless CuCF3 and, consequently, gave the trifluoromethy-
lated arenes in higher yield.
In addition, Fuchikami and co-workers also found that by
switching the copper salt from CuI to CuBr the yield of the
reaction decreased to 58%. We now know that reaction of
CuBr with TMSCF3 and KF conducted by Kolomeitsev and
co-workers generated [Cu(CF3)(Br)]− and [Cu(CF3)2]− with
a ratio of 3.6:1. Our study showed that the ratio of
[Cu(CF3)(I)]− and [Cu(CF3)2]− for the reaction of CuI
with TMSCF3 and KF in DMF was 1.44:1. Since [Cu(CF3)2]−
is at least 2.3 times more reactive than [Cu(CF3)(X)]−, the
higher ratio of the [Cu(CF3)2]− in the reaction system suggests
faster reaction and higher yields since longer reaction time may
lead to decomposition of the trifluoromethylated copper
species.
I
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(Scheme 6). More specifically, reactions of electron-poor aryl
iodides such as 4-nitrophenyl iodide and methyl 4-
Nevertheless, our studies disclosed that the reactivities of both
[Cu(CF3)2]− and [Cu(CF3)(I)]− are not high, and a much
more reactive species, ligandless CuCF3 or DMF-ligated
[(DMF)CuCF3], is generated when excess CuI is presented.
Quantitative studies showed that ligandless CuCF3 or DMF-
ligated [(DMF)CuCF3] is roughly 24 times more reactive than
[Cu(CF3)2]− and 80 times more reactive than [Cu(CF3)(I)]−.
An overall mechanistic map was then provided on the basis of
these systematic studies. In the presence of excess CuI,
reaction of [Cu(CF3)2]− with CuI occurs at elevated
temperature to give the active species ligandless [CuCF3] or
DMF-ligated [(DMF)CuCF3], which then reacts with aryl
iodides to give the desired trifluoromethylarenes and
regenerate CuI. Alternatively, in the absence of CuI, both
[Cu(CF3)2]− and [Cu(CF3)(I)]− react with aryl iodides but
much more slowly than that of ligandless [CuCF3] or DMF-
ligated [(DMF)CuCF3] to afford ArCF3. Under these
conditions, reaction with [Cu(CF3)2]− is faster than that
with [Cu(CF3)(I)]−. Based on these mechanistic studies, we
uncovered a new activator HOAc which could effectively
abstract one of the trifluoromethyl groups in [Cu(CF3)2]− to
generate more reactive ligandless [CuCF3] or DMF-ligated
[(DMF)CuCF3]. Under these conditions, a broad range of aryl
iodides, bromides, and chlorides were trifluoromethylated
under mild conditions. Furthermore, our studies also suggest
that the halide anion generated from the aryl halides is
deleterious to the reaction by conversion [Cu(CF3)2]− to less
reactive [Cu(CF3)(I)]− or by quenching the activator CuI to
generate [CuI2]−. Efforts to develop efficient copper-catalyzed
trifluoromethylation of aryl halides by effectively removing the
halide anion from the reaction mixture are currently underway
in our laboratory.
Scheme 6. HOAc-Promoted Trifluoromethylation of
a,b
(Hetero)aryl Halides by Using 1a
ASSOCIATED CONTENT
■
a
sı
Reaction conditions: aryl halide (0.50 mmol), 1a (0.75 mmol),
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.1c07408.
acetate acid (0.75 mmol) in DMF (5.0 mL) at room temperature to
b
80 °C for 16−48 h under argon. Yields were determined by 19F
NMR spectrum by using fluorobenzene as internal standard; Isolated
1
yields were shown in parentheses.
Experimental details and kinetic profiles; H, 19F, and
13C NMR spectra of 1a−c and 2a−l (PDF)
iodobenzoate occurred after 24 h at room temperature to
give the desired products 2a and 2b in 91% and 82% yield,
respectively, while reactions of electron-rich aryl iodides 4-
iodo-N,N-dimethylaniline and N-(4-iodophenyl)acetamide
required reaction at 50−80 °C for 16−24 h for full conversion
(Scheme 6, 2a−d). As a comparison, previous conditions using
CuI as the activator typically required reaction at 100 °C for
full conversion for electron-rich aryl iodides.20 Electron-poor
heteroaryl bromides also showed good reactivities (2h−j),
while aryl bromide (2g), electron-rich heteroaryl bromide (2i),
and (hetero)aryl chlorides (2k−l) were trifluoromethylated in
moderate to excellent yields with the assistance of a directing
group at the ortho-position. The accelerating effect of an ortho-
ester or nitro group was consistent with the reaction that used
CuI as the activator.20
Accession Codes
CCDC 2096993 and 2096996 contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
■
Corresponding Author
Qilong Shen − Key Laboratory of Organofluorine Chemistry,
Shanghai Institute of Organic Chemistry, University of
Chinese Academy of Sciences, Chinese Academy of Sciences,
Shanghai 200032, P.R. China; orcid.org/0000-0001-
5622-153X; Email: shenql@sioc.ac.cn
CONCLUSIONS
■
Authors
In summary, in this work, we report the identification of the
previously proposed key intermediates [Ph4P]+[Cu(CF3)-
(X)]− (X = Cl, I or CF3) in copper-mediated trifluoromethy-
lation by independent synthesis. These complexes were fully
characterized by NMR spectroscopy and microanalysis and
further confirmed by X-ray diffraction of their single crystals.
He Liu − Key Laboratory of Organofluorine Chemistry,
Shanghai Institute of Organic Chemistry, University of
Chinese Academy of Sciences, Chinese Academy of Sciences,
Shanghai 200032, P.R. China; orcid.org/0000-0002-
8755-5863
J
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
ysis in Organic Synthesis; Anilkumar, A., Saranya, S., Eds.; Wiley-VCH:
Weinheim, 2020; pp 367−422.
Jian Wu − Shanghai Institute of Organic Chemistry, University
of Chinese Academy of Sciences, Chinese Academy of Sciences,
Shanghai 200032, P.R. China
Yuxuan Jin − Key Laboratory of Organofluorine Chemistry,
Shanghai Institute of Organic Chemistry, University of
Chinese Academy of Sciences, Chinese Academy of Sciences,
Shanghai 200032, P.R. China
Xuebing Leng − Shanghai Institute of Organic Chemistry,
University of Chinese Academy of Sciences, Chinese Academy
of Sciences, Shanghai 200032, P.R. China; orcid.org/
0000-0003-3291-8695
(5) (a) McLoughlin, V. C. R.; Thrower, J. Manufacture of Organic
Compounds Containing Fluorine. US3408411A, 1968. (b) McLough-
lin, V. C. R.; Thrower, J. A Route to Fluoroalkyl-Substituted Aromatic
Compounds Involving Fluoroalkylcopper Intermediates. Tetrahedron
1969, 25, 5921−5940.
(6) Kobayashi, Y.; Kumadaki, I. Trifluoromethylation of Aromatic
Compounds. Tetrahedron Lett. 1969, 10, 4095−4096.
(7) (a) Kondratenko, N. V.; Vechirko, E. P.; Yagupolskii, L. M. A
Convenient Method for the Introduction of Trifluoromethyl Groups
into Organic Compounds. Synthesis 1980, 1980, 932−933.
(b) Kremlev, M. M.; Tyrra, W.; Mushta, A. I.; Naumann, D.;
Yagupolskii, Y. L. The Solid Complex Zn(CF3)Br·2DMF as an
Alternative Reagent for the Preparation of Both, Trifluoromethyl and
Pentafluoroethyl Copper, CuCF3 and CuC2F5. J. Fluorine Chem. 2010,
131, 212−216.
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.1c07408
Notes
The authors declare no competing financial interest.
(8) (a) Xie, Q.-Q.; Hu, J.-B. Chen’s Reagent: A Versatile Reagent for
Trifluoromethylation, Difluoromethylenation, and Difluoroalkylation
in Organic Synthesis. Chin. J. Chem. 2020, 38, 202−212. (b) Chen,
Q.-Y. Trifluoromethylation of Organic Halides with Defluorocarbene
Precursors. J. Fluorine Chem. 1995, 72, 241−246. (c) Chen, Q.-Y.;
Wu, S.-W. Methyl Fluorosulphonyldifluoroacetate; a New Trifluor-
omethylating Agent. J. Chem. Soc., Chem. Commun. 1989, 705−706.
(d) Su, D.; Duan, J.; Chen, Q. Methyl Chlorodifluoroacetate a
Convenient Trifluoromethylating Agent. Tetrahedron Lett. 1991, 32,
7689−7690. (e) Wiemers, D. M.; Burton, D. J. A Remarkably Simple
Preparation of (Trifluoromethyl)cadmium and -zinc Reagents
Directly from Difluorodihalomethanes. J. Am. Chem. Soc. 1985, 107,
5014−5015.
(9) (a) Urata, H.; Fuchikami, T. A Novel and Convenient Method
for Trifluoromethylation of Organic Halides Using CF3SiR′3/KF/
Cu(I) System. Tetrahedron Lett. 1991, 32, 91−94. (b) Weng, Z. Q.;
Lee, R.; Jia, W. G.; Yuan, Y. F.; Wang, W. F.; Feng, X.; Huang, K. W.
Cooperative Effect of Silver in Copper Catalyzed Trifluoromethyla-
tion of Aryl Iodides Using Me3SiCF3. Organometallics 2011, 30,
3229−3232.
(10) (a) Matsui, K.; Tobita, E.; Ando, M.; Kondo, K. A Convenient
Trifluoromethylation of Aromatic halides with Sodium Trifluoroace-
tate. Chem. Lett. 1981, 10, 1719−1720. (b) Carr, G. E.; Chambers, R.
D.; Holmes, T. F.; Parker, D. G. Sodium perfluoroalkane carboxylates
as sources of perfluoroalkyl groups. J. Chem. Soc., Perkin Trans. 1
1988, 921−926. (c) Johansen, M. B.; Lindhardt, A. T. Copper-
Catalyzed and Additive Free Decarboxylative Trifluoromethylation of
Aromatic and Heteroaromatic Iodides. Org. Biomol. Chem. 2020, 18,
1417−1425.
(11) (a) Zanardi, A.; Novikov, M. A.; Martin, E.; Benet-Buchholz, J.;
Grushin, V. V. Direct Cupration of Fluoroform. J. Am. Chem. Soc.
2011, 133, 20901−20903. (b) Novák, P.; Lishchynskyi, A.; Grushin,
V. V. Fluoroform-Derived CuCF3 for Low-Cost, Simple, Efficient, and
Safe Trifluoromethylation of Aryl Boronic Acids in Air. Angew. Chem.,
Int. Ed. 2012, 51, 7767−7770. (c) Lishchynskyi, A.; Novikov, M. A.;
Martin, E.; Escudero-Adán, E. C.; Novák, P.; Grushin, V. V.
Trifluoromethylation of Aryl and Heteroaryl Halides with Fluoro-
form-Derived CuCF3: Scope, Limitations, and Mechanistic Features.
J. Org. Chem. 2013, 78, 11126−11146.
ACKNOWLEDGMENTS
■
The authors gratefully acknowledge financial support from the
National Natural Science Foundation of China (21625206,
21632009, 21421002, 22061160465).
REFERENCES
■
(1) (a) Hagmann, W. K. The Many Roles for Fluorine in Medicinal
Chemistry. J. Med. Chem. 2008, 51, 4359−4369. (b) Kirk, K. L.
Fluorination in Medicinal Chemistry: Methods, Strategies, and Recent
Developments. Org. Process Res. Dev. 2008, 12, 305−321. (c) Mean-
well, N. A. Synopsis of Some Recent Tactical Application of
Bioisosteres in Drug Design. J. Med. Chem. 2011, 54, 2529−2591.
(d) Wang, J.; Sánchez-Roselló, M.; Acena, J.; del Pozo, C.;
̃
Sorochinsky, A. E.; Fustero, S.; Soloshonok, V. A.; Liu, H. Fluorine
in Pharmaceutical Industry: Fluorine-Containing Drugs Introduced to
the Market in the Last Decade (2001−2011). Chem. Rev. 2014, 114,
2432−2506. (e) Zhou, Y.; Wang, J.; Gu, Z.; Wang, S.; Zhu, W.;
Acena, J.; Soloshonok, V. A.; Izawa, K.; Liu, H. Next Generation of
̃
Fluorine-Containing Pharmaceuticals, Compounds Currently in
Phase II−III Clinical Trials of Major Pharmaceutical Companies:
New Structural Trends and Therapeutic Areas. Chem. Rev. 2016, 116,
422−518.
(2) Leo, A.; Hansch, C.; Elkins, D. Partition Coefficients and their
Uses. Chem. Rev. 1971, 71, 525−616.
(3) For selected recent reviews on trifluoromethylation methods,
see: (a) Ma, J.-A.; Cahard, D. Update 1 of: Asymmetric Fluorination,
Trifluoromethylation, and Perfluoroalkylation Reactions. Chem. Rev.
2008, 108, PR1−PR43. (b) Furuya, T.; Kamlet, A. S.; Ritter, T.
Catalysis for Fluorination and Trifluoromethylation. Nature 2011,
473, 470−477. (c) Tomashenko, O. A.; Grushin, V. V. Aromatic
Trifluoromethylation with Metal Complexes. Chem. Rev. 2011, 111,
4475−4521. (d) Alonso, C.; Marigorta, E. M.; Rubiales, G.; Palacios,
F. Carbon Trifluoromethylation Reactions of Hydrocarbon Deriva-
tives and Heteroarenes. Chem. Rev. 2015, 115, 1847−1935. (e) .
(4) For selected reviews on copper-mediated/-catalyzed trifluor-
omethylation methods, see: (a) Tomashenko, O. A.; Grushin, V. V.
Aromatic Trifluoromethylation with Metal Complexes. Chem. Rev.
2011, 111, 4475−4521. (b) Roy, S.; Gregg, B. T.; Gribble, G. W.; Le,
V.-D.; Roy, S. Trifluoromethylation of aryl and Heteroaryl Halides.
Tetrahedron 2011, 67, 2161−2195. (c) Liu, T.; Shen, Q. Progresses in
Copper-Mediated Formation of Trifluoromethylated arene. Eur. J.
Org. Chem. 2012, 2012, 6679−6687. (d) Li, G.-B.; Zhang, C.; Song,
C.; Ma, Y.-D. Progress in Copper-Catalyzed Trifluoromethylation.
(12) (a) Serizawa, H.; Aikawa, K.; Mikami, K. Direct Synthesis of a
Trifluoromethyl Copper Reagent from Trifluoromethyl Ketones:
Application to Trifluoromethylation. Chem. - Eur. J. 2013, 19, 17692−
17697. (b) Li, X.-J.; Zhao, J.-W.; Zhang, L.; Hu, M.-Y.; Wang, L.-M.;
Hu, J.-B. Copper-Mediated Trifluoromethylation Using Phenyl
Trifluoromethyl Sulfoxide. Org. Lett. 2015, 17, 298−301. (c) Ye, Y.-
D.; Sanford, M. S. Merging Visible-Light Photocatalysis and
Transition-Metal Catalysis in the Copper-Catalyzed Trifluoromethy-
lation of Boronic Acids with CF3I. J. Am. Chem. Soc. 2012, 134,
9034−9037. (e) Dai, J.-J.; Fang, C.; Xiao, B.; Yi, J.; Xu, J.; Liu, Z.-J.;
Lu, X.; Liu, L.; Fu, Y. Copper-Promoted Sandmeyer Trifluoromethy-
lation Reaction. J. Am. Chem. Soc. 2013, 135 (23), 8436−8439.
́
Beilstein J. Org. Chem. 2018, 14, 155−181. (e) Mudarra, A; Martínez
de Salinas, S.; Pérez-Temprano, M. H. Nucleophilic Trifluoromethy-
lation Reactions Involving Species: From Organometallic Insights to
Scope. Synthesis 2019, 51, 2809−2820. (f) Kananovich, D. G.
Copper-Catalyzed Trifluoromethylation Reactions. In Copper Catal-
K
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(13) (a) Dubinina, G. G.; Furutachi, H.; Vicic, D. A. Active
Trifluoromethylation Agents from Well-Defined Copper(I)-CF3
Complexes. J. Am. Chem. Soc. 2008, 130, 8600−8601. (b) Morimoto,
H.; Tsubogo, T.; Litvinas, N. D.; Hartwig, J. F. A Broadly Applicable
Copper Reagent for Trifluoromethylations and Perfluoroalkluylations
of Aryl Iodides and Bromides. Angew. Chem., Int. Ed. 2011, 50, 3793−
3798. (c) Tomashenko, O. A.; Escudero-Adán, E. C.; Belmonte, M.
M.; Grushin, V. V. Simple, Stable, and Easily Accessible Well-Defined
CuCF3 Aromatic Trifluoromethylating Agents. Angew. Chem., Int. Ed.
2011, 50, 7655−7659.
(14) (a) Oishi, M.; Kondo, H.; Amii, H. Aromatic Trifluoromethy-
lation Catalytic in Copper. Chem. Commun. 2009, 1909−1911.
(b) Knauber, T.; Arikan, F.; Roschenthaler, G. V.; Goossen, L. J.
Copper-catalyzed trifluoromethylation of aryl iodides with potassium
(trifluoromethyl)trimethoxyborate. Chem. - Eur. J. 2011, 17, 2689−
2697. (c) Le, C.; Chen, T. Q.; Liang, T.; Zhang, P.; MacMillan, D. W.
C. A Radical Approach to the Copper Oxidative Addition Problem:
Trifluoromethylation of Bromoarenes. Science 2018, 360, 1010−1014.
(d) Zhao, X.-B.; MacMillan, D. W. C. Metallaphotoredox
Perfluoroalkylation of Organobromides. J. Am. Chem. Soc. 2020,
142, 19480−19486.
(15) Wiemers, D. M.; Burton, D. J. Pregeneration, Spectroscopic
Detection, and ChemicalReactivity of (Trifluoromethyl) copper, an
Elusive and Complex Species. J. Am. Chem. Soc. 1986, 108, 832−834.
(16) Willert-Porada, M. A.; Burton, D. J.; Baenziger, N. C. Synthesis
and X-Ray Structure of Bis(trifluoromethyl)(N,N-diethyldithiocarba-
mat0)-copper; a Remarkably Stable Perfluoroalkylcopper(iii) Com-
plex. J. Chem. Soc., Chem. Commun. 1989, 1633−1634.
(17) Kutt, A.; Movchun, V.; Rodina, T.; Dansauer, T.; Rusanov, E.
̈
B.; Leito, I.; Kaljurand, I.; Koppel, J.; Pihl, V.; Ovsjannikov, G.; Toom,
L.; Mishima, M.; Medebielle, M.; Lork, E.; Röschenthaler, G.-V.;
Koppel, I. A.; Kolomeitsev, A. A. Pentakis(trifluoromethyl)phenyl, a
Sterically Crowded and Electron-withdrawing Group: Synthesis and
Acidity of Pentakis(trifluoromethyl)benzene, -toluene, -phenol, and
-aniline. J. Org. Chem. 2008, 73, 2607−2620.
(18) Dubinina, G. G.; Ogikubo, J.; Vicic, D. A. Structure of
Bis(trifluoromethyl)cuprate and Its Role in Trifluoromethylation
Reactions. Organometallics 2008, 27, 6233−6235.
́
(19) Martínez de Salinas, S.; Mudarra, A. L.; Odena, C.; Martínez
Belmonte, M.; Benet-Buchholz, J.; Maseras, F.; Pérez-Temprano, M.
H. Exploring the Role of Coinage Metalates in Trifluoromethylation:
A Combined Experimental and Theoretical Study. Chem. - Eur. J.
2019, 25, 9390−9394.
(20) Liu, H.; Shen, Q. Bistrifluoromethylated Organocuprate
[Ph4P]+[Cu(CF3)2]−: Synthesis, Characterization and Its Application
for Trifluoromethylation of Activated Heteroaryl Bromides, Chlorides
and Iodides. Org. Chem. Front. 2019, 6, 2324−2328.
(21) Cuprate [Cu(CF3)(Cl)]− was previously characterized in
solution but not isolated; see: Shreiber, S. T.; Kaplan, P. T.; Hughes,
R. P.; Vasiliu, M.; Dixon, D. A.; Cramer, R. E.; Vicic, D. A. Syntheses,
Solution Behavior, and Computational Bond Length Analyses of
Trifluoromethyl and Perfluoroethyl Cuprate Salts. J. Fluorine Chem.
2020, 234, 109518.
(22) Nilsson, M. Tetrabutylammonium Inorganocuprates(I)*
−
−
−
−
nBu4N+CuCl2 , CuBr2 , CuI2 , and Cu(CN)2 . Acta Chem. Scand. B
1982, 36, 125−126.
(23) (a) Brothers, P. J.; Roper, W. R. Transition-Metal
Dihalocarbene Complexes. Chem. Rev. 1988, 88, 1293−1326.
(b) Hall, M. B.; Fenske, R. F. Electronic Structure and Bonding in
Methyl- and Perfluoromethyl(pentacarbonyl)manganese. Inorg. Chem.
1972, 11, 768−775.
(24) Romine, A. M.; Nebra, N.; Konovalov, A. I.; Martin, E.; Benet-
Buchholz, J.; Grushin, V. V. Easy Access to the Copper(III) Anion
[Cu(CF3)4]−. Angew. Chem., Int. Ed. 2015, 54, 2745−2749.
L
https://doi.org/10.1021/jacs.1c07408
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX