Copper-mediated radical cross-coupling reaction of 2,2-dichloro-1,1,1- trifluoroethane (HCFC-123) with phenols or thiophenols

Article abstract of DOI:10.1021/ol302962p

A copper-mediated cross-coupling reaction between HCFC-123 and phenols or thiophenols has been achieved. It is found that diethyl amine, which serves as both the activator and ligand of copper, plays a key role in this reaction. Two possible radical involved processes are proposed for the reaction mechanism.

Full text of DOI:10.1021/ol302962p

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Copper-Mediated Radical Cross-Coupling
Reaction of 2,2-Dichloro-1,1,1-
trifluoroethane (HCFC-123) with Phenols
or Thiophenols
Xiao-Jun Tang and Qing-Yun Chen*
Key Laboratory of Organofluorine Chemistry, Shanghai Institute of Organic Chemistry,
Chinese Academy of Sciences, 345 Lingling Road, 200032, Shanghai, P.R. China
chenqy@sioc.ac.cn
Received October 27, 2012
ABSTRACT
A copper-mediated cross-coupling reaction between HCFC-123 and phenols or thiophenols has been achieved. It is found that diethyl amine,
which serves as both the activator and ligand of copper, plays a key role in this reaction. Two possible radical involved processes are proposed for
the reaction mechanism.
According with Montreal Protocol, ozone-depleting
hydrochlorofluorocarbons (HCFCs) will be completely
forbidden as refrigerants, aerosol propellants, or cleaning
solvents in 2030.1 However, from a synthetic viewpoint,
some HCFCs are indispensable for both syntheses of
fluorinated building blocks and industrial applications.
For example, a widely used refrigerant, difluorochloroethane
(HCFC-22), is not only a well-known difluorocarben
reagent2 in the laboratory but also a key starting material
in the manufacture of tetrafluoroethene (TFE) in the
industry.3 Because of the electronic effect and steric hin-
drance of fluorinated groups, compared with common alkyl
chlorides, most HCFCs are reluctant to undergo SN1 or SN2
reactions. In this sense, converting ozone-depleting and
easily available HCFCs to useful organfluorine com-
pounds via direct cleavage of unactivated CꢀCl bonds to
useful fluorinated compounds will contribute to both the
progress of organofluorine chemistry and environmental
protection.4 To some extent, the chemical reactivities of
CꢀCl bonds in the HCFCs are similar to those of car-
bonꢀhalogen bonds of aryl halides, which can react with
enolates,5 thiolates,6 or diphenylphosphide ions7 via Single
Electron Transfer (SET) processes.8 In most cases, com-
pared with thiolates, as a class of poor electron donors, it is
(4) For selected reports on direct cleavage of CꢀCl bonds of HCFCs,
see: (a) Tamura, M.; Sekiya, A. J. Fluorine Chem. 1995, 71, 119. (b)
Long, Z. Y.; Chen, Q. Y. J. Org. Chem. 1999, 64, 4775. (c) Tang, X. J.;
Chen, Q. Y. Chem. Sci. 2012, 3, 1694.
(5) (a) Maria, T. B.; Mariana, H. G.; Adriana, B. P. J. Org. Chem.
1998, 63, 6394. (b) Galli, C.; Gentili, P.; Rappoport, Z. J. Org. Chem.
1994, 59, 6786. (c) Leeuwen, M. V.; McKillop, A. J. Chem. Soc., Perkin
Trans. 1 1993, 2433.
(1) (a) Banks, R. E. J. Fluorine Chem. 1994, 67, 193. (b) Franciso,
J. S.; Maricq, M. M. Acc. Chem. Res. 1996, 29, 39.
(6) (a) Bowman, W. R.; Heaney, H.; Smith, Ph. H. G. Tetrahedron
Lett. 1984, 25, 5821. (b) Liedhoom, B. Acta Chem. Scand. B 1984, 8, 877.
(7) Aguiar, A. M.; Greenberg, H. J.; Rubenstein, K. E. J. Org. Chem.
1963, 28, 2091.
(8) For reviews on Single Electron Transfer (SET) reactions, see: (a)
Ashby, E. C. Acc. Chem. Res. 1988, 21, 414. (b) Marcus, R. A. Angew.
Chem., Int. Ed. Engl. 1993, 32, 1111.
(2) For pioneering work, see: (a) Hine, J.; Porter, J. J. J. Am. Chem.
Soc. 1957, 79, 5493. (b) Soborowskii, B. Russ. J. Org. Chem. 1959, 29,
1142.
(3) (a) Park, J.; Benning, A.; Downing, F.; Laucius, J.; McHarness,
R. Ind. Eng. Chem. 1947, 39, 354. (b) Santos, M.; Hernandez-Vara, R.;
Herreros, J. M.; Gonzales-Diaz, P. F. J. Mol. Struct. 1986, 142, 533. (c)
Maksimov, B. N. Russ. J. Org. Chem. 1994, 30, 1935.
r
10.1021/ol302962p
XXXX American Chemical Society

difficult for phenolates to initiate SRN1 reactions directly,9
so it is still a challenge to synthesize RfꢀOAr from RfꢀX
and ArOH directly.
of 2a, we decided to increase the use of copper. The
spectrum yield of 2a was updated to 89% when 1.5 equiv
of copper were used (entry 11). Excitedly, the spectrum
yield of 2a was improved to 100% when 2 equiv of copper
were added (entry 12). Furthermore, when DCE or DCM
was used as the solvent, the product of 2a was found with
68%, 78% 19F NMR yield, respectively (entries 13 and 14).
Based the on the experimental results above, entry 12 was
selected as the best conditions.
Among the family of HCFCs, 2,2-dichloro-1,1,1-tri-
fluoroethane (HCFC-123) is an easily available and largely
used cleaning solvent and refrigerant in the industry. Early
reports showed that the CꢀCl bond in HCFC-123 can be
cleaved smoothlyin the presenceofcopper.10 Similar to the
classical copper-mediated Ullmann cross-coupling reac-
tion, a SET process was also involved in the cleavage of the
CꢀCl bond of HCFC-123. Inspired by the classical cop-
per-mediated Ullmann reaction between aryl halides and
aryl halides,11 phenols, anilines, or thiophenols,12 it can
be anticipated that copper-mediated Ullmann-type cross-
coupling reactions between HCFC-123 and phenols or
thiophenols may be practicable. Herein, we report our
results in this communication.
Table 1. Optimizations of Copper-Mediated Cross-Coupling
between HCFC-123 and 1a
Optimizations of reaction conditions using p-MeOPhOH
(1a) as the model substrate are listed in the Table 1.
Initially, DMSO, a general solvent used for classical
Ullmann reaction, was selected as the solvent. Unexpect-
edly, according to 19F NMR spectra, no change in HCFC-
123 was detected even when the amount of Et3N was
increased to 3 equiv (entries 1 and 2). To our delight, a
low new 19F NMR signal at ꢀ80 ppm (d, J = 4 Hz) was
found when 1 equiv of diethyl amine was added as the
additive (entry 3). As expected, a similar result was
obtained when DMF was used in place of DMSO (entry 4).
Upon consideration of a more simple operation, we turned
toward using HCFC-123 itself as the solvent directly.
However, the CꢀCl bond of HCFC-123 was inert when
no Et2NH was added (entries 5 and 6). Surprisingly, a 61%
spectrum yield of expected product 2a was observed when
Et2NH was used as the additive (entry 7). The yield of 2a
was decreased significantly to 33% when pyrrolidine was
added as the additive (entry 8).
Theseresultsindicatethata secondaryamine, Et2NH, or
pyrrolidine may activate copper in this reaction. Further-
more, we observed a yellow Cu(I) coordination compound
when Et2NH served as a ligand, precipitating slowly in the
reaction process. In contrast, both copper and HCFC-123
remained unchanged when i-Pr2NH, a secondary amine
with steric hindrance, was used as the additive in place of
Et2NH (entry 9). As foreseen, no product of 2a was found
in the absence of copper, which confirms that a simple SN1
or SN2 reaction could not take place between phenonates
and HCFC-123 (entry 10). In order to obtain a better yield
entry
Cu/1a/Et3N
additivea
solventb
t/°C
yieldc
1
1.25:1:1.25
1.25:1:3
ꢀ
DMSO
80
80
80
80
60
60
60
60
60
60
60
60
60
60
N.R.
N.R.
trace
trace
N.R.
N.R.
61%
33%
N.R.
N.R.
89%
100%
68%
75%
2
ꢀ
DMSO
3
1.25:1:1.25
1.25:1:1.25
1.25:1:1.25
1.25:1:3
Et2NH
Et2NH
ꢀ
DMSO
4
DMF
5
HCFC-123
HCFC-123
HCFC-123
HCFC-123
HCFC-123
HCFC-123
HCFC-123
HCFC-123
DCMd
6
ꢀ
7
1.25:1:1.25
1.25:1:1.25
1.25:1:1.25
0:1:1.25
Et2NH
pyrrolidine
i-Pr2NH
Et2NH
Et2NH
Et2NH
Et2NH
Et2NH
8
9
10
11
12
13
14
1.5:1:1.25
2:1:1.25
2:1:1.25
2:1:1.25
DCEd
a The ratio of Cu and the additive is 1:1. b [1a] = 0.5 mol/L; reaction
time was 3 h; 1a/HCFC-123 = 1:2 when DMSO or DMF was used as the
solvent. All reactions were performed in a sealed tube. c The yield is
determined by 19FNMR using PhCF3 as internal standard. d [1a] =
0.5 mol/L; 1a/HCFC-123 = 1:2; reaction time is 5 h.
With the optimized conditions in hand, this transforma-
tion was applied to substituted phenols. The representative
instances are illustrated in Scheme 1. As seen, phenols
bearing electron-donating (alkyl, methoxy and phenyl),
electron-withdrawing (cyano, nitro, aldhyde, and ester)
substituents and halogens (chloro and bromo) all resulted
in 2 with excellent yields (84ꢀ97%). A variety of poten-
tially reactive groups such as cyano, nitro, aldhyde, and
bromo are well-tolerated in this reaction. The low yield
(58%) of 2b may be due to the poor solubility of p-AcNH-
PhOH in the HCFC-123. It is also noted that the reaction
time for the substrate of pyridin-3-ol needed to be pro-
longed to 10 h. According to the observed experimental
phenomenon, thismay be attributed tothe intereference by
the nitrogen atom in the pyridin-3-ol in the activation of
copper by Et2NH.
(9) For selected reviews on SRN1 reactions, see: (a) Bowman, W. R.
Chem. Soc. Rev. 1988, 17, 283. (b) Bunnett, J. F. Acc. Chem. Res. 1978,
11, 413.
(10) (a) Aoyama, H. PCT Int. Appl.WO9412454. (b) Xu, Y. L.;
Dolbier, W. R.; Rong, X. X. J. Org. Chem. 1997, 62, 1576.
(11) For recent reviews on using Ullmann reactions for synthesis of
biaryls, see: (a) Nelson, T. D.; Crouch, R. D. Org. React. 2004, 63, 265.
(b) Hassan, J.; Sevignon, M.; Gozzi, C.; Schulz, E.; Lemaire, M. Chem.
Rev. 2002, 102, 1359.
(12) For recent reviews on copper-mediated C(aryl)ꢀO, C(aryl)ꢀN,
and C(aryl)ꢀS bond formations, see: (a) Ley, S. V.; Thomas, A. W.
Angew. Chem., Int. Ed. 2004, 43, 1043. (b) Ley, S. V.; Thomas, A. W.
Angew. Chem., Int. Ed. 2003, 42, 5400. (c) Kunz, K.; Scholz, U.; Ganzer,
D. Synlett 2003, 2428.
Subsequently, we sought to expand this reaction to
substrates of thiophenols. Selecting 4-methylbenzenethiol
(3a) as the model substrate, optimized conditions are
B
Org. Lett., Vol. XX, No. XX, XXXX

Table 2. Optimizations of Copper-Mediated Cross-Coupling
between HCFC-123 and 3a
Scheme 1. Substrate Scope for Copper-Mediated
Cross-Coupling between HCFC-123 and Phenols
entry
t/°C
Cu/Et2NH/3a/Et3Na
yieldb
1
2
3
4
5
6
7
60
60
75
75
75
75
75
2:2:1:1.25
trace
trace
62%
78%
86%
74%
65%
2:2.5:1:1.25
2:2.5:1:1.25
2:3:1:1.25
1.7:3:1:1.25
1.5:3:1:1.25
1.7:1.7:1:1.25
a Using HCFC-123 as the solvent, [3a] = 0.5 mol/L; all reactions are
performed in the sealed tube for 24 h. b The yield is determined by 19
NMR using PhCF3 as internal standard.
F
a Cu/Phenol/Et2NH/Et3N = 2:1:2:1.25; HCFC-123 is used as the
solvent, and reactions are stirred in 60 °C for 3 h in a sealed tube. All
yields of 2 are isolated yields based on the phenol. b The reaction time is
prolonged to 12 h.
Table 3. Substrate Scope for Copper-Mediated Cross-Coupling
between Thiophenols and HCFC-123
shown Table 2. To our surprise, only a trace of product of
4a was obtained when previous conditions were used even
for 24 h at 60 °C (entry 1). As observed, most of the copper
was sustained because thiolates may poison copper com-
pared with phenolates. Realizing this result, we decided to
increase the use of Et2NH. Regretfully, a similar result was
obtainedwhenthe ratio ofcopperand Et2NHwas changed
to 2:2.5 (entry 2). To our delight, a 62% spectrum yield of
4a was observed when the ratio of Cu/Et2NH/3a= 2: 2.5:1
(entry 3). When changing the ratio of Cu and Et2NH to
2:3, a side product, p-tolyl(2,2,2-trifluoroethyl) sulfane13
(19F NMR: ꢀ65.8 ppm, t, J = 10 Hz), was obtained in the
reaction although the yield of 4a was increased to 78%
(entry 4). Undoubtedly, p-tolyl(2,2,2-trifluoroethyl) sul-
fane was originated from the reduction of 4a by excess
copper. By changing the ratio of copper and Et2NH
to 1.7:3, the spectrum yield of 4a was increased to 86%
(entry 5). However, the yield of 4a was reduced to 74% if
the ratio of copper and Et2NH was 1.5:3 (entry 6). As
expected, the yield of 4a was reduced to 65% when the use
of Et2NH was minimized (entry 7). Based on optimiza-
tions, the ratio of Cu/Et2NH/3a = 1.7:3:1 (entry 5) is an
appropriate condition for this reaction.
yield
c
entry
Ar-SHa
4b
40
1
2
3
4
5
6
7
p-Me-PhSH (3a)
4a, 79%
4b, 71%
4c, 81%
4d, 85%
4e, 55%
4f, 75%
4g, 18%
40a, 13%
40b, 15%
40c, 14%
40d, 10%
40e, 41%
40f, 20%
40g, 46%
p-isopropyl-PhSH (3b)
p-MeO-PhSH (3c)
p-AcNH-PhSH (3d)
o-Br-PhSH (3e)
p-Cl-PhSH (3f)
pyridine-2-thiol (3g)
a Cu/Et2NH/3/Et3N = 1.7:3:1:1.25, [3] = 0.5 mol/L. All reactions
are performed in a sealed tube for 24 h. b Isolated yield based on 3. c 19
NMR yield based on 3, using PhCF3 as internal standard.
F
amount of copper was decreased. This is because the SET
processbetween 4 and copper can be assistedby the π* MO
in 4 compared with HCFC-123.14
Asdescribed inliterature, theproduct4 can be appliedto
trifluoroethylation of aromatic compounds.15 For the
product 2, it may be used to synthesize some useful
building blocks for the existence of the Cl atom. By
treatment with Zn powder in the DMSO at 110 °C for
3 h, the product 2 can be transformed into a monomer of
The scope of this reaction is exhibited by the results
shown in Table 3. In general, good total yields of 4 and 40
were obtained from thiophenols bearing different substi-
tuents. Similarto the substrate of phenols, this reaction can
also tolerate functional groups such as ethers, halogens,
and amides. It should be noted that the product 40 origi-
nating from 4 cannot be avoided in this reaction when the
(14) For a review on an SRN1 reaction assisted by π-acceptors
(carbonyl and phenyl), see: Rossi, R. A.; Pierini, A. B.; Penenory,
A. B. Chem. Rev. 2003, 103, 71.
(15) (a) Uneyama, K.; Momota, M. Tetrahedron Lett. 1989, 30, 2265.
(b) Uneyama, K.; Momota, M.; Hayashida, K.; Itoh, T. J. Org. Chem.
1990, 55, 5364. (c) Tahara, R.; Fukuhara, T.; Hara, S. J. Fluorine Chem.
2011, 132, 579.
(13) Long, Z. Y.; Chen, Q. Y. J. Fluorine Chem. 1998, 91, 95.
Org. Lett., Vol. XX, No. XX, XXXX
C

fluorinated polymers, difluoroviny aryl ether 5 with ex-
cellent yields (Scheme 2). Compared with previous method
of defluorination of 2,2,2-trifluoroethyl aryl ether with a
strong organic base,16 this method has good functional
tolerance and high yields.
of the anion-radical M results in the 2,2,2-trifluoro-1-
chloroethyl radical intermediate R. On the other hand,
phenolates or thiolates can be converted to Cu(I)-ZAr N in
the presence of Cu(I).18 Starting from the intermediate
Cu(I)-ZAr N, two possible paths may exist in the following
process. For Path 1, hemolytic cleavage of the CuꢀZ bond
in N will produce Cu(0) and the radical intermediate S. A
double radical coupling between the radical intermediate S
and R leads to the product P. For Path 2, coupling with the
radical intermediate R results in the the Cu(I) intermediate
N which can be converted to another Cu(II) intermediate
O. Reductive elimination of the Cu(II) intermediate O
affords the product P.
Scheme 2. Synthesis of Difluoroviny Aryl Ether from 2
Scheme 4. Proposed Mechanism for the Copper-Mediated
Radical Cross-Coupling Reaction
Finally, we turned toward investigating the mechanism
of this cross-coupling reaction. The Atom Transfer Radical
Addition17 product of 6 was obtained with 97% isolated
yield when styrene was used as the radical trapper.
Furthermore, when the same amount of p-MeOPhOH
and styrene were added, the product 6 and the cross-
coupling product 2a were obtained in 71% and 39% 19
F
NMR yield (Scheme 3), respectively. These results indicate
that a 2,2,2-trifluoro-1-chloroethyl radical intermediate is
in a “free” status in this cross-coupling reaction.
In summary, we have completed an interesting copper-
mediated cross-coupling reaction between HCFC-123 and
phenols or thiophenols with excellent yields. The resulted
products can be used as useful fluorinated building blocks.
In this cross-coupling reaction, Et2NH serves as not only
the activator of copper powder but also the ligand for
copper species. The radical intermediate is found to be at
a free status when using styrene as the radical trapper.
Scheme 3. Trapping the Radical Intermediate with Styrene
Differing from the existing common SN1, SN2, and SRN
1
reactions, this cross-coupling reaction displays a novel path-
way for a substitution reaction in an unactived alkyl chloride,
HCFC-123. More detailed mechanism research and further
development of this cross-coupling reaction are in progress.
Acknowledgment. We thank the Chinese Academy of
Science, the National Natural Science Foundation (21032006),
and the 973 Program of China (2012CBA01200) for financial
support.
In light of these results, our proposed mechanism is
depicted as follows (Scheme 4). The Single Electron
Transfer (SET) reaction between Cu and HCFC-123 gives
rise to the anion radical M and CuCl. Quick fragmentation
Supporting Information Available. Full experimental
data and characterization for all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(16) Nakai, T.; Tanaka, K.; Ishikawa, N. Chem. Lett. 1976, 5, 1263.
(17) For reviews on ATRA reactions, see: (a) Clark, A. J. Chem. Soc.
Rev. 2002, 31, 1. (b) Severin, K. Curr. Org. Chem. 2006, 10, 217.
(18) Posner, G. H.; Whitten, C. E. Org. Synth. 1988, 6, 248.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX