Synthesis of gem-difluoroalkenes by copper-catalyzed regioselective hydrodefluorination of 1-trifluoromethylalkenes

Article abstract of DOI:10.1246/cl.200163

A copper-catalyzed regioselective hydrodefluorination of 1-trifluoromethylalkenes with hydrosilanes has been developed. The copper catalysis is compatible with several functional groups, including alkyl chloride, ether, ester, nitrile, and imide moieties, to form the corresponding gem-difluoroalkenes in good yields. Additionally, asymmetric induction is also possible by using the chiral DTBM-SEGPHOS ligand, and gem-difluoroalkene with point chirality at the allylic position is obtained with high enantioselectivity.

Full text of DOI:10.1246/cl.200163

Received: March 2, 2020 | Accepted: March 17, 2020 | Web Released: March 24, 2020
CL-200163
Synthesis of gem-Difluoroalkenes by Copper-catalyzed Regioselective Hydrodefluorination
of 1-Trifluoromethylalkenes
Yuki Kojima, Tatsuaki Takata, Koji Hirano,* and Masahiro Miura*
Department of Applied Chemistry, Graduate School of Engineering, Osaka University, Suita, Osaka 565-0871, Japan
E-mail: k_hirano@chem.eng.osaka-u.ac.jp
a) previous hydroamination with p-t-Bu-dppbz
A copper-catalyzed regioselective hydrodefluorination of
1-trifluoromethylalkenes with hydrosilanes has been developed.
The copper catalysis is compatible with several functional
groups, including alkyl chloride, ether, ester, nitrile, and imide
moieties, to form the corresponding gem-difluoroalkenes in good
yields. Additionally, asymmetric induction is also possible by
using the chiral DTBM-SEGPHOS ligand, and gem-difluoro-
alkene with point chirality at the allylic position is obtained with
high enantioselectivity.
H
F
F
Cu(OAc)2
F
F
F
F
PMHS
p-t-Bu-dppbz
Ph(CH2)2
+
Ph(CH2)2
CsOAc
NBn2
76%
BzO NBn2
1a
1,4-dioxane
b) mechanistic considerations
+
LnCuX Si
H
H
F
BzO NBn2
F
F
R
LnCu
H
NBn2
hydroamination
H
F
LnCuOBz
LnCuF
F
F
F
A
Keywords: Copper
| gem-Difluoroalkenes | Fluorine
F
F
R
R
LnCu
1a
B
Because of uniquely high electron-withdrawing nature,
gem-difluoroalkenes are isosteres of the carbonyl group1 and
thus frequently observed in pharmaceuticals and agrochemicals.2
Additionally, they are good building blocks for functionalized
mono-fluoroalkenes, which can work as the biologically
important amide isostere,3 via metal-catalyzed site-selective
defluorofunctionalizations.4 The gem-difluoroalkenes are gener-
ally prepared by the Wittig-type reaction of carbonyl compounds
with difluoromethylenation reagents.5 Moreover, some research
groups have recently developed more convergent coupling-type
processes with some CF2 carbene precursors or CF3CH carbene
equivalents.6 However, despite the aforementioned certain
advances, there still remains a need for efficient chemical
synthesis of gem-difluoroalkenes owing to their ubiquity in
fields of organic chemistry and medicinal chemistry.
H
F
R = Ph(CH2)2
R
F
hydrodefluorination
c) serendipity of hydrodefluorination with xantphos
F
H
F
Cu(OAc)2
F
F
PMHS
xantphos
+
Ph(CH2)2
Ph(CH2)2
2a, 84% (1H NMR)
F
CsOAc
1,4-dioxane
BzO NBn2
1a
t-Bu
t-Bu
P
P
O
Ph2P
PPh2
t-Bu
t-Bu
xantphos
p-t-Bu-dppbz
Meanwhile, we recently developed the Cu/p-t-Bu-dppbz-
catalyzed regioselective net-hydroamination7,8 of 1-trifluoro-
methylalkene 1a with polymethylhydrosiloxane (PMHS) and the
hydroxylamine derivative (Scheme 1a).9 Following our mecha-
nistic considerations, 1a undergoes the insertion into in-situ
generated copper hydride species A10 to form the α-CF3 alkyl-
copper intermediate B. Subsequent electrophilic amination with
the hydroxylamine delivers the corresponding hydroaminated
product. In spite of the possibility of β-F elimination from B,4,11
the p-t-Bu-dppbz ligand selectively promoted the net-hydro-
amination reaction (Scheme 1b). During the optimization
studies on this chemistry, we serendipitously found the
significant ligand effect of xantphos: even in the presence of
hydroxylamine, the hydrodefluorinated product, namely gem-
difluoroalkene 2a, was formed exclusively (Scheme 1c). The
unique propensity with xantphos toward the β-F elimination
prompted us to further investigate this reaction. Herein, we
wish to report a Cu/xantphos-catalyzed hydrodefluorination
of 1-trifluoromethylalkenes with the hydrosilane to afford the
gem-difluoroalkenes. The copper catalysis is tolerated with
several functional groups, and the corresponding gem-difluoro-
alkenes are obtained in good yields with high regioselectivity.
Additionally, by the judicious choice of supporting ligand, the
asymmetric induction is also possible. The related boryl-,12
Scheme 1. Copper-catalyzed reductive functionalization of 1-
trifluoromethylalkene 1a. a) Hydroamination with p-t-Bu-dppbz
ligand, b) mechanistic considerations, and c) serendipity of
hydrodefluorination with xantphos ligand.
silyl,13 alkyl-,14 and aryldefluorination15 reactions of 1-trifluoro-
methylalkenes were reported by several groups, but the reductive
defluorination process still remains underdeveloped.16
Inspired by our initial finding (Scheme 1c), our optimization
studies commenced with 1-trifluoromethylalkene 1a, PMHS,
and CuCl/xantphos catalyst in 1,4-dioxane solvent, and some
bases were initially screened (Table 1). As shown in Scheme 1c,
CsOAc was a good base, giving the gem-difluoroalkene 2a
1
in 75% H NMR yield (Entry 1). Consistent with our previous
observation9 and recent mechanistic insight11b that smaller alkali
metal cations can accelerate the β-F elimination process, lithium
and sodium alkoxide bases also worked well, and the targeted 2a
was obtained in yields comparable to that with CsOAc (Entries
2-5). In particular, NaO-t-Bu and NaOTMS showed better
performance to deliver 2a in 87% and 92% isolated yields,
respectively (Entries 4 and 5). Other weaker non-alkoxide bases
such as KOPiv resulted in no formation of 2a (data not shown),
suggesting that relatively strong bases are necessary in the
Chem. Lett. 2020, 49, 637–640 | doi:10.1246/cl.200163
© 2020 The Chemical Society of Japan | 637

R2
CuCl (10 mol%)
R2
Table 1. Optimization studies for regioselective hydrodefluori-
nation of 1-trifluoromethylalkene 1a.a
F
F
F
F
xantphos (10 mol%)
+
PMHS
F Cl
R1
R1
F
base, 1,4-dioxane
rt, 4 h
F
F
CuCl (10 mol%)
ligand (10 mol%)
1
2
F
F
+
Ph(CH2)2
H
Si
Ph(CH2)2
F
F
F
F
base, 1,4-dioxane
rt, 4 h
1a
2a
BnO
CH3(CH2)7
F
F
Entry H-Si
Ligand
xantphos
xantphos
xantphos
xantphos
xantphos
dppbz
Base
Yieldb/%
2c, 61% (NaO-t-Bu) 2d, 50% (NaO-t-Bu)a
2b, 87% (NaO-t-Bu)
1
2
3
4
5
6
7
8
PMHS
PMHS
PMHS
PMHS
PMHS
PMHS
PMHS
PMHS
PMHS
PMHS
PMHS
CsOAc
LiO-t-Bu
LiOTMS
NaO-t-Bu 95 (87)
NaOTMS >99 (92)
NaOTMS 13
NaOTMS 21
NaOTMS 45
NaOTMS 57
75
72
78
F
F
PivO
F
TBSO
F
2f, 78% (NaO-t-Bu)
2e, 71% (NaO-t-Bu)
O
F
F
dppe
dppf
N
O
2h, 66% (NaOTMS)b
NC
F
F
2g, 80% (NaOTMS)
9
DPEPhos
rac-BINAP NaOTMS 59
PPh3
10
11c
12
13d
F
Me
F
F
NaOTMS 43
NaOTMS 38
NaOTMS 99
F
(EtO)2MeSiH xantphos
Ph2SiH2 xantphos
Ph
F
Ph
F
BocN
2j, 77% (NaOTMS)c
2i, 83% (NaO-t-Bu)
2k, 22% (NaOTMS)d
aConditions: CuCl (0.025 mmol), ligand (0.025 mmol), 1a
(0.25 mmol), H-Si (0.75 mmol based on H-Si), base
Me3Si
SiMe3
b
(0.50 mmol), 1,4-dioxane (1.5 mL), rt, 4 h, N2. Estimated by
F
F
1H NMR with CH2Br2 as the internal standard. Isolated yields
are in parentheses but somewhat lower than the 1H NMR yield
because of the volatility of 2a. cWith 0.050 mmol of PPh3.
dWith 0.50 mmol of Ph2SiH2.
Me3Si
P
Me3Si
SiMe3
SiMe3
P
Ph
F
2k’
SiMe3
Me3Si
TMS-dppbz
Scheme 2. gem-Difluoroalkene products 2 by Cu-catalyzed
hydrodefluorination of 1-trifluoromethylalkenes 1. Conditions:
CuCl (0.025 mmol), xantphos (0.025 mmol), 1 (0.25 mmol),
PMHS (0.75 mmol based on H-Si), base (0.50 mmol), 1,4-
dioxane (1.5 mL), rt, 4 h, N2. Isolated yields are shown. The base
used (NaO-t-Bu or NaOTMS) is given in parentheses. aWith
PMHS (0.38 mmol based on H-Si). bAt 50 °C. cWith TMS-
reaction. The effect of xantphos was critical: some other
common phosphine ligands also promoted the reaction but
with much lower efficiency (Entries 6-11). We next examined
other hydrosilanes. Among them, Ph2SiH2 also gave an
almost quantitative yield (Entry 13), but from the viewpoints
of stability and cost, we identified PMHS to be the best hydride
source.
d
dppbz ligand instead of xantphos. From a 9:1 E/Z mixture of
1
1k. H NMR yield. The reduced byproduct 2k¤ was also formed
With the optimal conditions in hand, we then investigated
the scope of 1-trifluoromethylalkenes 1. The products 2 obtained
are illustrated in Scheme 2, in which the best basic additive
(NaO-t-Bu or NaOTMS) was dependent on the substrate
structure and functional group. The simple aliphatic 1b was
expectedly converted to the gem-difluoroalkene 2b in a good
yield. The copper catalysis was compatible with alkyl chloride
(2c), benzyl ether (2d), and silyl ether (2e) moieties. The
carbonyl functions such as ester (2f) and nitrile (2g) were also
tolerated under the standard conditions. Moreover, the substrates
with protected nitrogen groups underwent the hydrodefluorina-
tion smoothly to form the corresponding functionalized gem-
difluoroalkenes 2h and 2i in acceptable yields. Particularly
notable is the successful conversion of trisubstituted alkene (2j)
with the aid of TMS-dppbz ligand instead of xantphos, thus
giving a chance of asymmetric induction by an appropriate
chiral ligand (vide infra).17 On the other hand, the Ph-conjugated
1k resulted in a low yield of the corresponding gem-difluor-
oalkene 2k and competitive formation of the reduced byproduct
2k¤. This is probably because the regioselectivity in the insertion
into the copper hydride (Scheme 1b) is not controlled. The Ph-
vinyl conjugation increases the coefficient of the LUMO at the
carbon α to CF3, thus competitively forming the regioisomeric β-
1
in 3% H NMR yield.
CF3 alkylcopper intermediate. A similar trend was observed also
in our previous work.9
As a mechanistic probe, we subsequently performed the
reaction of 1a with the Ph2SiD2 instead of PMHS under
otherwise identical conditions (eq 1). Expectedly, the deuterium
atom was selectively incorporated at the allylic position, and the
corresponding 2a-d1 was obtained in a good yield with 96% D
content. This result not only supports our initial hypothesis
(Scheme 1b) but also provides access to deuterium-labeled gem-
difluoroalkenes, which can find applications in biological label-
ing studies and drug discovery.18 Finally, we attempted to apply
chiral bisphosphine ligands for the asymmetric synthesis of 2j
(eq 2). Gratifyingly, the (R)-DTBM-SEGPHOS ligand was
found to be promising to produce the enantioenriched (S)-2j in
84% yield with >99:1 enantiomeric ratio (er).19 The product
could also be transformed to the borylalkene 320 under the
reported conditions,4c which should be a useful chiral building
block for various chiral fluoroalkenes.
In conclusion, we have developed a copper/xantphos-
catalyzed hydrodefluorination of 1-trifluoromethylalkenes with
638 | Chem. Lett. 2020, 49, 637–640 | doi:10.1246/cl.200163
© 2020 The Chemical Society of Japan

hydrosilanes to form the corresponding gem-difluoroalkenes in
good yields with high regioselectivity. The copper catalysis can
also provide an avenue to deuterium-labeled as well as the
enantioenriched gem-difluoroalkenes, which are of potent
interest in medicinal and pharmaceutical chemistry. Ongoing
work seeks to elucidate the origin of uniquely high β-F elimina-
tion performance with xantphos and develop related copper-
catalyzed reductive transformations of 1-trifluoromethylalkenes.
T. Hosoya, J. Am. Chem. Soc. 2017, 139, 12855. b) R.
Kojima, K. Kubota, H. Ito, Chem. Commun. 2017, 53,
10688. c) H. Ito, T. Seo, R. Kojima, K. Kubota, Chem. Lett.
2018, 47, 1330. d) D.-H. Tan, E. Lin, W.-W. Ji, Y.-F. Zeng,
W.-X. Fan, Q. Li, H. Gao, H. Wang, Adv. Synth. Catal. 2018,
360, 1032.
For a review: a) C. Ni, J. Hu, Synthesis 2014, 46, 842.
Selected examples: b) S. A. Fuqua, W. G. Duncan, R. M.
Silverstein, J. Org. Chem. 1965, 30, 1027. c) Y. Zhao, W.
Huang, L. Zhu, J. Hu, Org. Lett. 2010, 12, 1444. d) C. S.
Thomoson, H. Martinez, W. R. Dolbier, Jr., J. Fluorine
Chem. 2013, 150, 53.
a) V. G. Nenajdenko, G. N. Varseev, V. N. Korotchenko,
A. V. Shastin, E. S. Balenkova, J. Fluorine Chem. 2003, 124,
115. b) M. Hu, Z. He, B. Gao, L. Li, C. Ni, J. Hu, J. Am.
Chem. Soc. 2013, 135, 17032. c) M. Hu, C. Ni, L. Li, Y.
Han, J. Hu, J. Am. Chem. Soc. 2015, 137, 14496. d) J.
Zheng, J.-H. Lin, L.-Y. Yu, Y. Wei, X. Zheng, J.-C. Xiao,
Org. Lett. 2015, 17, 6150. e) Z. Zhang, Q. Zhou, W. Yu, T.
Li, G. Wu, Y. Zhang, J. Wang, Org. Lett. 2015, 17, 2474.
f) Z. Zhang, W. Yu, C. Wu, C. Wang, Y. Zhang, J. Wang,
Angew. Chem., Int. Ed. 2016, 55, 273. g) Y. Ma, B. R. P.
Reddy, X. Bi, Org. Lett. 2019, 21, 9860.
For contributions from our group, see: a) Y. Miki, K. Hirano,
T. Satoh, M. Miura, Angew. Chem., Int. Ed. 2013, 52, 10830.
b) Y. Miki, K. Hirano, T. Satoh, M. Miura, Org. Lett. 2014,
16, 1498. c) D. Nishikawa, K. Hirano, M. Miura, J. Am.
Chem. Soc. 2015, 137, 15620. d) D. Nishikawa, R. Sakae, Y.
Miki, K. Hirano, M. Miura, J. Org. Chem. 2016, 81, 12128.
e) T. Takata, D. Nishikawa, K. Hirano, M. Miura, Chem.®
Eur. J. 2018, 24, 10975.
For contributions from the Buchwald research group, see:
a) S. Zhu, N. Niljianskul, S. L. Buchwald, J. Am. Chem. Soc.
2013, 135, 15746. b) S. Zhu, S. L. Buchwald, J. Am. Chem.
Soc. 2014, 136, 15913. c) N. Niljianskul, S. Zhu, S. L.
Buchwald, Angew. Chem., Int. Ed. 2015, 54, 1638. d) S.-L.
Shi, S. L. Buchwald, Nat. Chem. 2015, 7, 38. e) Y. Yang,
S.-L. Shi, D. Niu, P. Liu, S. L. Buchwald, Science 2015, 349,
62. f) D. Niu, S. L. Buchwald, J. Am. Chem. Soc. 2015, 137,
9716. g) H. Wang, J. C. Yang, S. L. Buchwald, J. Am. Chem.
Soc. 2017, 139, 8428. h) Y. Zhou, O. D. Engl, J. S. Bandar,
E. D. Chant, S. L. Buchwald, Angew. Chem., Int. Ed. 2018,
57, 6672. i) S. Guo, J. C. Yang, S. L. Buchwald, J. Am.
Chem. Soc. 2018, 140, 15976.
5
6
F
F
+
Ph2SiD2
Ph(CH2)2
F
(2.0 equiv)
1a
ð1Þ
ð2Þ
CuCl (10 mol%)
D
F
xantphos (10 mol%)
Ph(CH2)2
F
NaO-t-Bu, 1,4-dioxane
rt, 4 h
2a-d1, 65%
(96% D)
PMHS (3.0 equiv)
CuCl (10 mol%)
(R)-DTBM-SEGPHOS (10 mol%)
Me
F
F
F
Ph
1j
NaOTMS, 1,4-dioxane, rt, 4 h
F
F
H Me
*
H Me
*
7
8
Ph
F
Ph
BF3K
ref. 4c
(S)-2j, 84%, >99:1 er
3, 57%
This work was supported by JSPS KAKENHI Grant Nos.
JP 15H05485 (Grant-in-Aid for Young Scientists (A)) and
JP 18K19078 (Grant-in-Aid for Challenging Research
(Exploratory)) to K.H., and JP 17H06092 (Grant-in-Aid for
Specially Promoted Research) to M.M.
Supporting Information is available on https://doi.org/
10.1246/cl.200163.
References and Notes
1
a) C. Leriche, X. He, C. T. Chang, H. Liu, J. Am. Chem. Soc.
2003, 125, 6348. b) G. Magueur, B. Crousse, M. Ourévitch,
D. Bonnet-Delpon, J.-P. Bégué, J. Fluorine Chem. 2006,
127, 637. c) N. A. Meanwell, J. Med. Chem. 2011, 54, 2529.
a) W. R. Moore, G. L. Schatzman, E. T. Jarvi, R. S. Gross,
J. R. McCarthy, J. Am. Chem. Soc. 1992, 114, 360. b) Y. Pan,
J. Qiu, R. B. Silverman, J. Med. Chem. 2003, 46, 5292.
c) P. M. Weintraub, A. K. Holland, C. A. Gates, W. R.
Moore, R. J. Resvick, P. Bey, N. P. Peet, Bioorg. Med. Chem.
2003, 11, 427. d) J.-M. Altenburger, G. Y. Lassalle, M.
Matrougui, D. Galtier, J.-C. Jetha, Z. Bocskei, C. N. Berry,
C. Lunven, J. Lorrain, J.-P. Herault, P. Schaeffer, S. E.
O’Connor, J.-M. Herbert, Bioorg. Med. Chem. 2004, 12,
1713. e) S. Messaoudi, B. Tréguier, A. Hamze, O. Provot,
J.-F. Peyrat, J. R. De Losada, J.-M. Liu, J. Bignon, J.
Wdzieczak-Bakala, S. Thoret, J. Dubois, J.-D. Brion, M.
Alami, J. Med. Chem. 2009, 52, 4538.
2
9
T. Takata, K. Hirano, M. Miura, Org. Lett. 2019, 21, 4284.
10 For a review on Cu-H species, see: C. Deutsch, N. Krause,
B. H. Lipshutz, Chem. Rev. 2008, 108, 2916.
11 For recent detailed mechanistic studies, see: a) N. O.
Andrella, N. Xu, B. M. Gabidullin, C. Ehm, R. T. Baker,
J. Am. Chem. Soc. 2019, 141, 11506. b) P. H. S. Paioti, J. del
Pozo, M. S. Mikus, J. Lee, M. J. Koh, F. Romiti, S. Torker,
A. H. Hoveyda, J. Am. Chem. Soc. 2019, 141, 19917.
12 a) R. Corberán, N. W. Mszar, A. H. Hoveyda, Angew. Chem.,
Int. Ed. 2011, 50, 7079. b) R. Kojima, S. Akiyama, H. Ito,
Angew. Chem., Int. Ed. 2018, 57, 7196. c) Y. Liu, Y. Zhou, Y.
Zhao, J. Qu, Org. Lett. 2017, 19, 946.
13 H. Sakaguchi, M. Ohashi, S. Ogoshi, Angew. Chem., Int. Ed.
2018, 57, 328.
14 For recent selected examples, see: a) S. B. Lang, R. J. Wiles,
C. B. Kelly, G. A. Molander, Angew. Chem., Int. Ed. 2017,
3
4
a) S. Couve-Bonnaire, D. Cahard, X. Pannecoucke, Org.
Biomol. Chem. 2007, 5, 1151. b) T. Taguchi, H. Yanai, In
Fluorine in Medicinal Chemistry and Chemical Biology; I.
Ojima, Ed.; Blackwell Publishing Inc.: Malden, 2009;
pp 257.
a) H. Sakaguchi, Y. Uetake, M. Ohashi, T. Niwa, S. Ogoshi,
Chem. Lett. 2020, 49, 637–640 | doi:10.1246/cl.200163
© 2020 The Chemical Society of Japan | 639

56, 15073. b) Y. Lan, F. Yang, C. Wang, ACS Catal. 2018, 8,
9245. c) X. Lu, X.-X. Wang, T.-J. Gong, J.-J. Pi, S.-J. He, Y.
Fu, Chem. Sci. 2019, 10, 809. d) D. Ding, Y. Lan, Z. Lin, C.
Wang, Org. Lett. 2019, 21, 2723.
trisubstituted 1j. For example, the reaction of 1a with TMS-
dppbz under otherwise identical conditions provided 2a in
only 58% NMR yield (vs. Entry 5 in Table 1).
18 a) T. G. Gant, J. Med. Chem. 2014, 57, 3595. b) J. Atzrodt,
V. Derdau, W. J. Kerr, M. Reid, Angew. Chem., Int. Ed.
2018, 57, 1758. c) T. Pirali, M. Serafini, S. Cargnin, A. A.
Genazzani, J. Med. Chem. 2019, 62, 5276.
15 a) T. Miura, Y. Ito, M. Murakami, Chem. Lett. 2008, 37,
1006. b) Y. Huang, T. Hayashi, J. Am. Chem. Soc. 2016, 138,
12340.
16 During the preparation of this manuscript, Norton reported
the related nickel-catalyzed hydrodefluorination of CF3-
substituted alkenes, but the substrate scope was limited to the
Ar-conjugated systems. C. Yao, S. Wang, J. Norton, M.
Hammond, J. Am. Chem. Soc. 2020, 142, 4793.
19 See the Supporting Information for optimization studies on
asymmetric catalysis.
20 The absolute configuration of (S)-2j and 3 was determined
by the specific rotation after the derivatization into the
known product. See the Supporting Information for details.
17 The higher performance of TMS-dppbz was unique to the
640 | Chem. Lett. 2020, 49, 637–640 | doi:10.1246/cl.200163
© 2020 The Chemical Society of Japan