Organic Letters
Letter
a
a
Table 1. Optimization of Reaction Conditions
Table 2. Optimization of Reaction Conditions
b
b
entry
MXn
yield (%)
entry
solvent
NMP
DMF
yield (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
FeCl3
CoBr2
NiBr2
<5
<5
<5
<5
<5
<5
<5
1
2
3
4
5
6
7
8
9
76
45
39
70
42
23
<5
<5
<5
<5
<5
DMA
Ni(OAc)2
NaOAc
Ce(NH4)2(NO3)6
CuBr
Cu(OAc)2
CuCN
Cu(NO3)2·2H2O
Cu(acac)2
CuSO4
DMSO
MeCN
THF
DME
1,4-dioxane
DCE
c d
e
,
76 (72) (78)
<5
<5
<5
<5
<5
<5
<5
67
<5
56
51
10
11
PhCl
toluene
a
Unless otherwise noted, the reactions were performed at 60 °C for
24 by using ethynylbenzene (1a, 0.4 mmol), (1-
ethoxycyclopropoxy)trimethylsilane (2a, 0.8 mmol), and Cu(OAc)2
(0.8 mmol) in different solvents (3 mL). Yields were determined by
NMR analysis of crude reaction mixture after workup by using 1,4-
Cu(OTf)2
CuBr2
h
b
Cu(TFA)2
Cu(OAc)2
Cu(OAc)2
Cu(OAc)2
Cu(OAc)2
Cu(OAc)2
Cu(OAc)2
f
g
dimethoxybenzene as an internal standard.
h
i
(DCE), chlorobenzene (PhCl), and toluene (Table 2, entries
7−11).
j
42
53
k
We envisioned that the stoichiometric amount of Cu(OAc)2
might be reduced to a catalytic amount by introducing
stoichiometric amounts of other co-oxidants into the reaction
system. However, when we performed the model reaction
under the optimized reaction conditions but in the presence of
20 mol % Cu(OAc)2 and 2 equiv of co-oxidant (including
K2S2O8, ceric ammonium nitrate, ortho-iodoxybenzoic acid
(IBX), m-CPBA, tBuOOBz, cumene hydroperoxide, 2,3-
dichloro-5,6-dicyanobenzoquinone (DDQ), and 2,6-di-tert-
butyl-1,4-benzoquinone), the reaction proceeded sluggishly
to provide the desired product 3a in <5% yield.
a
Unless otherwise noted, the reactions were performed at 60 °C for
24 by using ethynylbenzene (1a, 0.4 mmol), (1-
ethoxycyclopropoxy)trimethylsilane (2a, 0.8 mmol), and metallic
salt MXn (0.8 mmol) in NMP (3 mL). Yields were determined by
NMR analysis of crude reaction mixture after workup by using 1,4-
dimethoxybenzene as an internal standard. Isolated yield. Homo-
coupling product of diyne was obtained in 12% yield. Isolated yield
of a 4 mmol scale reaction. Using 1 equiv of Cu(OAc)2. Using 0.2
equiv of Cu(OAc)2. Room temperature. 12 h. The ratio of 1a/2a is
h
b
c
d
e
f
g
h
i
j
k
1/1. The ratio of 1a/2a is 2/1.
Subsequently, the substrate scope of this copper-mediated
coupling reaction involving various alkynes by using Cu(OAc)2
as a mediator and NMP as solvent was examined under the
optimized reaction conditions. As summarized in Table 3, in
most cases, alkynes 1b−1o bearing either electron-withdrawing
or electron-donating substituents on the phenyl ring effectively
underwent the organic transformation under established
reaction conditions to produce the corresponding products
3b−3o in modest to high yields (Table 3, entries 1−14). In
addition, heterocyclic aryl alkynes 1p and 1q reacted in a same
fashion, affording the expected products 3p and 3q in
moderate yields (Table 3, entries 15 and 16). Note that the
mild reaction conditions also allowed the existence of a range
of functional groups or substituents (e.g., CN, CF3, Cl, Br,
OMe, and NMe2) in the substrate, which could potentially be
retained for downstream derivatization. However, as for the
reaction using alkyl-substituted terminal alkyne (e.g., 1-
hexyne) as a substrate, the reaction did not work under the
optimized reaction conditions.
contrast, the same reaction conducted in the presence of other
copper (I or II) additive almost could not take place (Table 1,
entries 9−15). In addition, it was observed that the use of 2
equiv of Cu(OAc)2 was necessary for the efficient progress of
the reaction, as the decrease of the amount of Cu(OAc)2 led to
lowered product yield (Table 1, entries 16 and 17). Moreover,
survey of other reaction parameters by performing the present
reaction either at room temperature or for only 12 h both
resulted in reduced reaction performance (Table 1, entries 18
and 19). Furthermore, changing the ratio of 1a/2a from 1/2 to
1/1 and 2/1 both led to decreased product yields of 42% and
53%, respectively (Table 1, entries 20 and 21).
With Cu(OAc)2 being recognized as the optimal mediator
for the present reaction, we proceeded to investigate the
solvent effect on the reaction by employing various solvents as
reaction medium under the above reaction conditions, hoping
to further improve the reaction performance. As shown in
Table 2, the reaction could proceed with differing efficiency in
N-methyl-2-pyrrolidinone (NMP), dimethyl formamide
(DMF), dimethylacetamide (DMA), dimethylsulfoxide
(DMSO), acetonitrile (MeCN), and tetrahydrofuran (THF)
(Table 2, entries 1−6), with the highest product yield being
obtained by utilizing NMP as solvent (Table 2, entry 1). In
comparison, the transformation almost could not occur in
other solvents, such as DME, 1,4-dioxane, dichloroethane
Next, the substrate scope of the reaction was further
examined by using different cyclopropanes 2 as starting
materials. Our initial studies showed that trimethyl(1-
phenylcyclopropoxy)silane (2b) and 1-phenylcyclopropan-1-
ol (2c) worked with comparable efficiency under the
optimized reaction conditions, leading to the expected product
4a in 74% and 69% yields, respectively (see Scheme 2). Given
B
Org. Lett. XXXX, XXX, XXX−XXX