Communication
using cheaper NaSO2CF3 as the CF3 source could be achieved
(Scheme 1d). Herein, we report a Cu-catalyzed difunctionaliz-
ing trifluoromethylation of activated alkynes in the synthesis of
3-trifluoromethyl spiro[4.5]trienones. This strategy involves
a connection of the CF3 group with the electron-deficient
carbon–carbon triple bond to generate the spiro[4.5]decane
skeleton via a tandem cyclization/dearomatization process.
Our investigation started by using N-methyl-N,3-diphenyl-
propiolamide 1a as the model substrate with sodium trifluoro-
methanesulfinate (NaSO2CF3 or Langlois’ reagent) in the pres-
ence of CuSO4 (10%) and tert-butyl hydroperoxide (TBHP) in
a 2:1 CH3CN/H2O solvent mixture at 608C under air atmos-
phere. Gratifyingly, the desired product 2a was obtained in
13% yield (Table 1, entry 1). When Na2CO3 was added as a base
could promote the yield slightly, of which MnO2 led to the
best yield of 53% (Table 1, entries 13–17). Finally, a yield of
58% was obtained after adjusting the amount of MnO2 to
3.0 equivalents. Other attempts to promote this transformation
proved to be ineffective (see the Supporting Information).
After the detailed investigation mentioned above, the opti-
mized conditions for this process were defined as the use of
1a (0.2 mmol) with NaSO2CF3 (2.5 equiv), TBHP (7.0 equiv, 70%
aqueous), NaOAc (2.0 equiv), and MnO2 (3.0 equiv) in CH3CN/
H2O (2:1, 1.5 mL) at 608C under air atmosphere for 10 h. The
structure of 2a was confirmed by X-ray crystal structure analy-
sis (see the Supporting Information).[14]
To explore the scope of this reaction, various substituted N-
arylpropiolamides were investigated under the optimized reac-
tion conditions (Table 2). Firstly, we investigated the effect of
the substituent at nitrogen (R1). Whereas substrate 1c (R1 =H)
only gave a trace amount of the desired product, substrate 1b
(R1 =ethyl) afforded the corresponding product 2b in 41%
yield. Secondly, the effect of substituent R2 was investigated.
Both electron-withdrawing and electron-donating groups at
the ortho and meta positions were well tolerated and afforded
the desired products 2d–h in moderate yields, which demon-
strated that the electronic effect of R2 had no significant influ-
ence on the reaction. Notably, the products 2 f and 2h, in
which R2 =Cl, could be obtained smoothly in satisfactory
yields, which could provide a potential opportunity for further
transformations through orthogonal cross-couplings. The sub-
strates with two substituents on the phenyl rings were also tol-
erated to give the corresponding products 2i–k, albeit in rela-
tively low yields, which indicated that steric effect made some
difference to the yield. By changing the phenyl group to
a naphthyl group in substrate 1l, the corresponding product
2l was obtained in 44% yield. Subsequently, the effect of R3
on the reaction was studied under the optimized conditions.
Substrates 1m–s, which bore electron-donating or electron-
withdrawing at the ortho, meta, or para position of the aryl
group, were efficiently transformed into the desired products
2m–2s in moderate yields. Notably, the reaction of substrate
1t, with two electron-withdrawing chloro groups, still proceed-
ed smoothly and afforded the corresponding product 2t in
44% yield. Interestingly, a biphenyl group (1u) or 2-thienyl
group (1v) on the terminal alkyne was compatible with this
process to afford the products 2u or 2v, albeit in low yields.
Even a methyl group on the terminal alkyne did not affect the
desired reaction, giving 2w in 36% yield. Finally, we replaced
Table 1. Optimization of the reaction conditions.[a]
Entry Catalyst ([mol%]) Base ([equiv]) Additive ([equiv])
Yield[b] [%]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
CuSO4 (10)
CuSO4 (10)
CuSO4 (10)
CuSO4 (10)
CuSO4 (10)
CuSO4 (10)
CuI (10)
CuCl2 (10)
Cu(OAc)2 (10)
Cu(OTf)2 (10)
CuCl (10)
–
CuCl (10)
CuCl (10)
CuCl (10)
CuCl (10)
CuCl (10)
–
–
–
–
–
–
–
–
–
–
–
–
–
13
30
26
24
28
38
37
42
36
31
45
trace
37
49
42
53
58(21[c])
Na2CO3 (2.0)
KOH (2.0)
K2HPO4 (2.0)
KF (2.0)
NaOAc (2.0)
NaOAc (2.0)
NaOAc (2.0)
NaOAc (2.0)
NaOAc (2.0)
NaOAc (2.0)
NaOAc (2.0)
NaOAc (2.0)
NaOAc (2.0)
K2S2O8 (2.0)
KIO3 (2.0)
NaOAc (2.0) Ce(SO4)2·4H2O (2.0)
NaOAc (2.0)
NaOAc (2.0)
MnO2 (2.0)
MnO2 (3.0)
[a] Reaction conditions: 1a (0.2 mmol,1 equiv), NaSO2CF3 (2.5 equiv),
TBHP (7.0 equiv, 70% aqueous solution) in CH3CN/H2O (2:1, 1.5 mL) at
608C under air atmosphere for 10 h; [b] yield of isolated product; [c] the
amount of recovered starting material is given in parentheses.
to this system, the yield was increased to 30%. We considered
that the addition of base might accelerate trifluoromethylation,
as the slower trifluoromethylation of electron-deficient
carbon–carbon triple bond, which is disfavored in atom-trans-
fer radical addition reactions. Thus, a series of bases were
tested, including KOH, K2HPO4, KF, and NaOAc, of which
NaOAc led to the most marked improvement in the yield
(Table 1, entries 2–6). Subsequently, different copper catalysts,
such as CuI, CuCl2, Cu(OAc)2, Cu(OTf)2, and CuCl, were investi-
gated, of which CuCl was most beneficial (Table 1, entries 7–
11). A control experiment revealed that the copper catalyst
was essential for the reaction (Table 1, entry 12). A further
study on this reaction indicated that the addition of oxidants
Scheme 2. Trifluoromethylation of substrates with para substituents R2.
&
&
Chem. Eur. J. 2014, 20, 1 – 7
2
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!