Simple and Clean Photoinduced Aromatic Trifluoromethylation Reaction

Article abstract of DOI:10.1021/jacs.6b02782

We describe a simple, metal-and oxidant-free photochemical strategy for the direct trifluoromethylation of unactivated arenes and heteroarenes under either ultraviolet or visible light irradiation. We demonstrated that photoexcited aliphatic ketones, such as acetone and diacetyl, can be used as promising low-cost radical initiators to generate CF₃ radicals from sodium triflinate efficiently. The broad utility of this strategy and its benefit to medicinal chemistry are demonstrated by the direct trifluoromethylation of unprotected bidentate chelating ligand, xanthine alkaloids, nucleosides, and related antiviral drug molecules.

Full text of DOI:10.1021/jacs.6b02782

Communication
pubs.acs.org/JACS
Simple and Clean Photoinduced Aromatic Trifluoromethylation
Reaction
Lu Li,†,‡ Xiaoyue Mu,† Wenbo Liu,† Yichen Wang,‡ Zetian Mi,*,‡ and Chao-Jun Li*,†
†Department of Chemistry, McGill University, 801 Sherbrooke Street West, Montreal, Quebec H3A 0B8, Canada
‡Department of Electrical and Computer Engineering, McGill University, 3480 University Street, Montreal, Quebec H3A 0E9,
Canada
S
* Supporting Information
transition metal-free method for the direct trifluoromethylation
ABSTRACT: We describe a simple, metal- and oxidant-
free photochemical strategy for the direct trifluoromethy-
lation of unactivated arenes and heteroarenes under either
ultraviolet or visible light irradiation. We demonstrated
that photoexcited aliphatic ketones, such as acetone and
diacetyl, can be used as promising low-cost radical
initiators to generate CF3 radicals from sodium triflinate
efficiently. The broad utility of this strategy and its benefit
to medicinal chemistry are demonstrated by the direct
trifluoromethylation of unprotected bidentate chelating
ligand, xanthine alkaloids, nucleosides, and related antiviral
drug molecules.
of heterocycles by using sodium triflinate (CF3SO2Na, Langlois’
reagent) and t-butyl hydroperoxide. However, the difficulty
associated with the removal of metal catalyst residue has led to
growing concerns in pharmaceuticals and materials.27 On the
other hand, the preparation of concentrated peroxides,
commonly used oxidants for the generation of CF3 radicals, is
energy intensive and dangerous. In this respect, the development
of a transition-metal-free trifluoromethylation methodology to
access benzotrifluorides under safe and environmentally benign
conditions has become an essential yet challenging objective for
today’s medical science.
As a greener alternative to the metal-catalyzed strategy in
organic synthesis, photochemical reactions possess the advantage
of avoiding the use of expensive and toxic metal catalysts.28−30
Very recently, we reported photoinduced metal-free aromatic
Finkelstein31 and Sonogashira32 reactions at room temperature.
With particular interest in photochemistry and fluorine
chemistry, we describe herein a mild, photoinduced approach
for the direct trifluoromethylation of unactivated arenes and
heteroarenes through a photoreduction mechanism without
using any metal catalysts or oxidants. As shown in Scheme 1, this
radical trifluoromethylation protocol was carried out in an
exceptionally facile manner by simply mixing substrate and solid
he incorporation of trifluoromethyl (CF3) group into drug
Tcandidates is widely prevalent in biochemical and medicinal
science because the CF3 moiety can dramatically modify the
physical and biological properties of parent molecules such as
solubility, lipophilicity, and catabolic stability.1−4 During the past
decades, an increasing attention has been paid to the
trifluoromethylation of arenes and heteroarenes, which are
fundamental building blocks of many top-selling pharmaceuticals
and agrochemicals, including celecoxib, cinacalcet, nilotinib,
beflubutamid, norfluazon, and so on.5 However, the traditional
cross-coupling methodologies usually require stoichiometric
amounts of metal complexes, preactivated substrates, or directing
groups.6 Substrate scope also suffers from severe limitations due
to the harsh reaction conditions.7 More recently, many powerful
strategies have been developed to synthesize functionalized
benzotrifluorides with high yield and broad substrate scope.8−10
One of the major improvements has been made via copper- or
palladium-catalyzed coupling of various prefunctionalized aryl
halides, boronic acids, and (hetero)arenes with either
nucleophilic or electrophilic CF3 reagents.11−17 Pivotal progress
was made by Langois, Baran, and Macmillion who pioneered the
direct radical trifluoromethylation of arenes and heteroar-
enes.18−26 The prefunctionalized strategy usually results in a
high regioselectivity, while the direct trifluoromethylation of
unactivated arenes can avoid the multistep synthesis of
complicated prefunctionalized substrates.
Scheme 1. Approaches to the Generation of CF3 Radicals from
Sodium Triflinate
Specific to the direct radical trifluoromethylation strategy, the
significant breakthrough made by MacMillan et al.19 showed that
ruthenium and iridium photoredox catalysts are amenable to the
direct trifluoromethylation of arenes under mild visible-light
irradiation. In the meantime, Baran et al.20 described a powerful
Received: March 16, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.6b02782
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
sodium triflinate (CF3SO2Na, Langlois’ reagent)18 in acetone
(solvent) under UV light (>300 nm) irradiation. Moreover, we
demonstrated that visible light (>400 nm) also worked smoothly
if a small amount of diacetyl cosolvent was added.
may directly activate sodium triflinate salts. Experimental results
showed that under direct photoirradiation from a 300 W xenon
lamp without any filter, a substantial yield of 1 could also be
achieved by using either acetonitrile (53%) or dichloromethane
(66%) instead of acetone as solvent (Table S1). However, if a
280 nm long-pass filter was carefully mounted in the system to
completely block wavelengths shorter than 280 nm from the
xenon lamp, which may interact with sodium triflinate, the
reaction did not proceed at all in acetonitrile or dichloromethane;
whereas 83% yield of 1 could be obtained in acetone. These
observations clearly demonstrate that photoexcited acetone is
capable of triggering the trifluoromethylation reaction efficiently.
Therefore, acetone in this study essentially acts as both solvent to
dissolve reactants and photosensitizer to harvest resonant
photons.38,39
As one of the most important and commonly used solvents in
organic synthesis and chemical industry, acetone is also the
simplest ketone and was among the first organic chromophoric
compounds described in photoreactions.33 Numerous photo-
induced processes have been successfully initiated by using n, π*
triplet-excited ketones, such as hydrogen abstraction,34 oxetane
formation (Paterno−
̀
Buchi reaction),35 photocyclization,36 and
̈
the Norrish reactions.37 As depicted in Figure 1a, acetone has a
The optimal wavelength and effective range for the
trifluoromethylation reaction in acetone were determined by
the wavelength dependence experiment. As shown in Figure 1b,
the best yield of 95% (0.1 mmol scale) for 1 could be achieved
under 2 h photoirradiation from a 300 W xenon lamp with a 300
nm long-pass filter (λ > 300 nm). The apparent quantum
efficiency at 300 nm was calculated to be ∼22% (the calculation
details can be found in the SI). The yield of 1 still remained at a
high level around 320 nm but decreased dramatically when the
wavelength of light irradiation was longer than 340 nm. The
effective wavelength range is consistent with the absorption
spectrum of acetone, further indicating that acetone was a good
photosensitizer for the trifluoromethylation reaction. A proposed
mechanism is depicted in Figure 2.
Figure 1. (a) UV−vis absorption spectra of CF3SO2Na (orange line)
and acetone (blue line). (b) Apparent quantum efficiency (AQE) (red
line and hexagon symbol, left axis) and yield of 1 (blue columns, right
axis) as a function of wavelength of the incident light.
strong absorption band in the ultraviolet range below 330 nm. Its
electronic excitation involves the transition of a lone pair electron
from oxygen to the carbonyl carbon. Based on this “radical-like”
characteristic, we envisioned that the resulting electron-deficient
oxygen atom of the carbonyl n, π* state may abstract one
electron from CF3SO2Na to generate a CF3 radical, which is
called the photoreduction of ketone.34
Initial experiments started with 1,3,5-trimethoxybenzene and
sodium triflinate salts in acetone at 20 °C under UV irradiation (λ
= 254 nm) at an intensity of 4.0 mW cm−2 by using a
photoreactor (Figure S1). The airtight quartz tube containing
reactants and solvent was evacuated by four freeze−pump−thaw
cycles and backfilled with ultrapurified argon (>99.999%) prior
to use. Optimized reaction conditions, which include the use of
0.1 mmol of substrate and 4 equiv of triflinate in 1 mL of acetone,
gave the desired CF3-substituted products 1 in excellent yield of
88% after 15 h photoreaction. Obviously, the conversion rate and
yield of a photochemical reaction depends strongly on the
incident light intensity. Therefore, if a 300 W xenon lamp with a
stronger light intensity of 20 mW cm−2 around 250 nm was
employed (Figure S1), the reaction time could be dramatically
reduced from 15 to 2 h with a good yield of 81%. It is noted that
∼5% of bis-CF3-substituted product 2 was also detected after 2 h
irradiation with strong UV light at 20 mW cm−2. Further
extending the reaction time to 4 and 20 h increased the yield of
bis-CF3-substituted product 2 to 36% and 77%, respectively.
Control experiments established the requirement of UV light, as
no reaction proceeded in the dark even under heating.
Figure 2. Proposed mechanism.
The preparative scope of the trifluoromethylation reaction was
carried out under UV irradiation with either a 300 W xenon lamp
(λ > 300 nm) or the photoreactor (λ = 254 nm). In general, this
clean photochemical protocol allows the direct incorporation of
CF3 moiety into a broad range of arenes, heteroarenes, and
nucleosides in yields ranging from 38% to 95% (Scheme 2).
Evidently, this approach is very effective for electron-rich arenes
(1, 2, 12) since CF3 radical is electron deficient.40 In addition,
although we cannot explain the phenomenon in detail yet, it was
found that the addition of 60 μL of acetic anhydride (Figure S2)
into acetone could increase the trifluoromethylation yield (Table
S2). Therefore, unsubstituted benzene and other arenes with
either electron-withdrawing groups or weak electron-donating
groups gave good yields (3−5, 9, 13) by adding 60 μL of acetic
anhydride as additive. For unsymmetrical aromatics with more
than one reactive site, products may be formed as isomeric
mixtures (6−8, 14, 15) or a single compound (16), synergisti-
cally governed by the innate electronic effect and steric effect of
To demonstrate the effect of acetone on the photoinduced
trifluoromethylation process, a series of optical filters and
different solvents were employed to run this reaction under UV
irradiation from the 300 W xenon lamp (emission wavelengths
between 200 and 1000 nm). As shown in Figure 1a, sodium
triflinate itself has a wide absorption band below 255 nm in the
UV−vis spectrum, indicating that UV light shorter than 255 nm
B
DOI: 10.1021/jacs.6b02782
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
addition of a small amount of water (0.05 or 0.1 mL) into acetone
could improve the substrate solubility and get a better yield.
Moreover, 5-trifluoromethyl-2′-deoxyuridine (trifluridine,
F3TdR, 23), a powerful antiviral drug that has been shown to
be highly useful in many biological systems with wide clinical
utility,41 was synthesized in a moderate yield via the direct
trifluoromethylation of 2′-deoxyuridine without protection.
Moreover, we conducted a large-scale reaction in a 100 mL of
quartz flask, and the results revealed that our strategy was
effective on the gram scale (1, 4, and 9, 5 mmol). Unfortunately,
substrates with nitro and carboxyl groups did not give good
yields. Amino group is not tolerated in our system due to its
intrinsic reactivity with acetone, generating imines.
Scheme 2. Radical Trifluoromethylation of Various Arenes
and Heteroarenes
ab
,
Encouraged by the experimental outcome under UV light, we
attempted to further extend this strategy into visible region (λ >
400 nm) by employing aliphatic diacetyl instead of acetone as the
radical initiator, which contains two carbonyl groups and shows
absorption in the visible wavelength range (Scheme 3a). As
shown in Scheme 3b, the preliminary results demonstrated that a
Scheme 3. (a) UV−vis Absorption Spectrum of Diacetyl and
Reaction Setup (inset), (b) Control Experiments with or
without Diacetyl under Visible Light Irradiation, and (c)
Substrate Scope under Visible Light Irradiation
a
Reaction conditions: arenes and heteroarenes (0.1 mmol),
NaSO2CF3 (0.4 mmol), and acetone (1.0 mL) at 20 °C in argon
b
c
under UV light for 2−40 h. Yield of isolated product. 0.06 mL of
d
acetic anhydride was added into 0.94 mL of acetone. Reaction
showed incomplete conversion, and a second portion of fresh
NaSO2CF3 (0.4 mmol) was added into the system to drive the
e
reaction toward completion. 0.05 mL of water was added into 0.95
f
mL of acetone. 0.1 mL of water was added into 0.9 mL of acetone.
*Minor isomeric product.
the substrates. Besides arenes, a variety of heteroarenes including
pyridine (17), pyrrole (18), indole (19), and imidazoles (20, 21)
could all be likewise tolerated with this photochemical
procedure. To further demonstrate the broad utility of this
strategy, some biologically active molecules, such as caffeine
(10), theophylline (11), and uridine (22), were successfully
subjected to this simple protocol to access their trifluoromethy-
lated derivatives with high regioselectivity. In these cases, the
a
Reaction conditions: arenes and heteroarenes (0.1 mmol),
NaSO2CF3 (0.4 mmol) and a mixture of ethyl acetate (0.8 mL)-
diacetyl (0.2 mL) at 20 °C in Ar under visible light for 10−40 h. Yield
of isolated product. Water was used instead of ethyl acetate as solvent.
b
c
C
DOI: 10.1021/jacs.6b02782
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(7) Burton, D. J.; Yang, Z.-Y. Tetrahedron 1992, 48, 189.
good yield of 66% for the trifluoromethylation of mesitylene
could be achieved under visible light irradiation with a household
fluorescent light bulb in a mixture solution of acetone-diacetyl
(9:1). Besides acetone, a variety of other solvents, such as
acetonitrile, ethyl acetate, and water, also worked well with
diacetyl (Table S3). In contrast, no reaction occurred in pure
solvent without diacetyl under the same conditions, indicating
the crucial role of diacetyl for the visible light-induced
trifluoromethylation reaction. Finally, the reaction scope was
tested under the optimized reaction conditions, which include
the use of 4 equiv of triflinate in 1.0 mL of ethyl acetate-diacetyl
mixture (4:1) under visible light irradiation from a 300 W xenon
lamp with a 400 nm long-pass filter (Scheme 3c). It was found
that most substrates listed in Scheme 2 proceeded well in the
visible light system with comparable yields. Importantly, the
reaction was also compatible with aldehyde (25) and halogen
(26, 27) groups, which are not stable under UV irradiation.
Purine (28) and bipyridine (29), widely used nitrogen-
containing heterocycles and bidentate chelating ligands,
respectively, were also operative smoothly with high regiose-
lectivity by this method.
In summary, an efficient and practical approach to the
photoinduced trifluoromethylation of arenes and heteroarenes
was developed with easily handled sodium triflinate. The value of
this strategy has been highlighted via the trifluoromethylation of
biologically active molecules under either UV or visible light
irradiation. Significantly, this photochemical strategy employed
acetone, one of the most widely used and cheapest organic
solvents, instead of expensive metal catalyst or dangerous
peroxides to generate CF3 radical, which provides a greener route
to cost-effective large-scale synthesis of trifluoromethylated
chemicals.
(8) Furuya, T.; Kamlet, A. S.; Ritter, T. Nature 2011, 473, 470.
(9) Chen, M.; Buchwald, S. L. Angew. Chem., Int. Ed. 2013, 52, 11628.
(10) Beatty, J. W.; Douglas, J. J.; Cole, K. P.; Stephenson, C. R. J. Nat.
Commun. 2015, 6, 7919.
(11) Cho, E. J.; Senecal, T. D.; Kinzel, T.; Zhang, Y.; Watson, D. A.;
Buchwald, S. L. Science 2010, 328, 1679.
(12) Wang, X.; Truesdale, L.; Yu, J.-Q. J. Am. Chem. Soc. 2010, 132,
3648.
(13) Morimoto, H.; Tsubogo, T.; Litvinas, N. D.; Hartwig, J. F. Angew.
Chem., Int. Ed. 2011, 50, 3793.
(14) Liu, T.; Shen, Q. Org. Lett. 2011, 13, 2342.
(15) Senecal, T. D.; Parsons, A. T.; Buchwald, S. L. J. Org. Chem. 2011,
76, 1174.
(16) Chu, L.; Qing, F.-L. J. Am. Chem. Soc. 2012, 134, 1298.
(17) Besset, T.; Schneider, C.; Cahard, D. Angew. Chem., Int. Ed. 2012,
51, 5048.
(18) Langlois, B. R.; Laurent, E.; Roidot, N. Tetrahedron Lett. 1991, 32,
7525.
(19) Nagib, D. A.; MacMillan, D. W. C. Nature 2011, 480, 224.
(20) Ji, Y.; Brueckl, T.; Baxter, R. D.; Fujiwara, Y.; Seiple, I. B.; Su, S.;
Blackmond, D. G.; Baran, P. S. Proc. Natl. Acad. Sci. U. S. A. 2011, 108,
14411.
(21) Ye, Y.; Lee, S. H.; Sanford, M. S. Org. Lett. 2011, 13, 5464.
(22) Ye, Y.; Kunzi, S. A.; Sanford, M. S. Org. Lett. 2012, 14, 4979.
̈
(23) Cui, L.; Matusaki, Y.; Tada, N.; Miura, T.; Uno, B.; Itoh, A. Adv.
Synth. Catal. 2013, 355, 2203.
(24) Sladojevich, F.; McNeill, E.; Borgel, J.; Zheng, S.-L.; Ritter, T.
̈
Angew. Chem., Int. Ed. 2015, 54, 3712.
(25) Natte, K.; Jagadeesh, R. V.; He, L.; Rabeah, J.; Chen, J.; Taeschler,
C.; Ellinger, S.; Zaragoza, F.; Neumann, H.; Bruckner, A.; Beller, M.
̈
Angew. Chem., Int. Ed. 2016, 55, 2782.
(26) Lefebvre, Q.; Hoffmann, N.; Rueping, M. Chem. Commun. 2016,
52, 2493.
(27) Sun, C.-L.; Shi, Z.-J. Chem. Rev. 2014, 114, 9219.
(28) Li, L.; Fan, S.; Mu, X.; Mi, Z.; Li, C.-J. J. Am. Chem. Soc. 2014, 136,
7793.
(29) Sun, J.; Peng, X.; Guo, H. Tetrahedron Lett. 2015, 56, 797.
(30) Mfuh, A. M.; Doyle, J. D.; Chhetri, B.; Arman, H. D.; Larionov, O.
V. J. Am. Chem. Soc. 2016, 138, 2985.
(31) Li, L.; Liu, W.; Zeng, H.; Mu, X.; Cosa, G.; Mi, Z.; Li, C.-J. J. Am.
Chem. Soc. 2015, 137, 8328.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.6b02782.
■
S
Experimental details and data (PDF)
(32) Liu, W.; Li, L.; Li, C.-J. Nat. Commun. 2015, 6, 6526.
(33) Coyle, J. D.; Carless, H. A. J. Chem. Soc. Rev. 1972, 1, 465.
(34) Singh, P. J. Chem. Soc. C 1971, 714.
AUTHOR INFORMATION
■
(35) Buchi, G.; Inman, C. G.; Lipinsky, E. S. J. Am. Chem. Soc. 1954, 76,
̈
Corresponding Authors
*zetian.mi@mcgill.ca
*cj.li@mcgill.ca
4327.
(36) Yang, N. C.; Yang, D.-D. H. J. Am. Chem. Soc. 1958, 80, 2913.
(37) Chatgilialoglu, C.; Crich, D.; Komatsu, M.; Ryu, I. Chem. Rev.
1999, 99, 1991.
Notes
(38) Xia, J.-B.; Zhu, C.; Chen, C. J. Am. Chem. Soc. 2013, 135, 17494.
(39) Xia, J.-B.; Zhu, C.; Chen, C. Chem. Commun. 2014, 50, 11701.
(40) Dolbier, W. R. Chem. Rev. 1996, 96, 1557.
(41) Heidelberger, C. Cancer Res. 1970, 30, 1549.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was financially supported by the Canada Research
Chair (Tier 1) foundation, the Natural Sciences and Engineering
́
Research Council of Canada, the Fonds de recherche sur la
nature et les technologies, Canada Foundation for Innovation
(CFI), and McGill University.
REFERENCES
■
(1) Muller, K.; Faeh, C.; Diederich, F. Science 2007, 317, 1881.
̈
(2) Hagmann, W. K. J. Med. Chem. 2008, 51, 4359.
(3) Meanwell, N. A. J. Med. Chem. 2011, 54, 2529.
(4) Zhou, Y.; Wang, J.; Gu, Z.; Wang, S.; Zhu, W.; Acena, J. L.;
̃
Soloshonok, V. A.; Izawa, K.; Liu, H. Chem. Rev. 2016, 116, 422.
(5) Alonso, C.; Martínez de Marigorta, E.; Rubiales, G.; Palacios, F.
Chem. Rev. 2015, 115, 1847.
(6) Tomashenko, O. A.; Grushin, V. V. Chem. Rev. 2011, 111, 4475.
D
DOI: 10.1021/jacs.6b02782
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX