2
J.-N. Xie et al. / Tetrahedron Letters xxx (2015) xxx–xxx
only the solvent but also the ligand was developed.32 Very recently,
we also introduced the heterogeneous catalyst (Cu(I)@C) to the
N
3
3
N
carboxylation of terminal alkynes with identified mechanism.
In the framework of our continuous efforts on C–C bond construc-
tion with CO , we speculated that the utilization of alternative type
N
Cu Cl
N
4
5
CuCl2
Ph
IL-1
2
of Cu(I) complex (copper(I)-based ILs) as catalyst would lead to an
effective procedure for direct carboxylation of terminal alkynes
IL-2
2
with atmospheric CO pressure at room temperature (Scheme 1),
[Cu(Im12) ][CuCl ]
[Cu(Im12) ][CuBr ]
thanks to the good solubility and long alkyl chain-containing imi-
nazole ligand. In this aspect, several copper(I)-based ILs
2
2
2
2
3
4,35
(
Scheme 2)
were found to be practicable to promote the car-
. Among them, [Cu
2 2
Im ) ][CuBr ] (Im = 1-dodecylimidazole) proved to be the most
IL-3
IL-4
boxylation of terminal alkynes with CO
2
1
2
12
Scheme 2. Copper(I)-containing ionic liquids used in this study.
(
efficient catalyst. This process features no external ligands, facile
preparation of the catalyst with good solubility, high catalytic
activity, and stability.
Table 1
Optimization of the reaction conditions for the Cu-IL catalyzed carboxylation of
a
terminal alkynes
Results and discussion
Cat., Cs2CO3
DMF, RT, t
O
CO2 + nBuI
(balloon)
For exploratory studies, we used phenylacetylene (1a) as the
model substrate, N,N-dimethylformamide (DMF) as the solvent in
the presence of copper catalyst (10% based on Cu), n-butyl iodide
Ph
H
+
Ph
n
O Bu
1a
2a
(
1.2 equiv), and Cs
2
CO
3
(1.2 equiv) with atmospheric pressure of
Entry
Cat.
Cat. (mol %)
d
t (h)
Yield (%)
b
CO . In the control experiment, only 4% yield of 2a was obtained,
2
1
2
3
4
5
6
7
8
9
—
—
12
12
12
12
12
12
12
12
12
12
12
12
12
12
9
4
42
74
53
46
9
20
53
94
99
81
88
43
43
53
46
suggesting the catalyst was essential to the carboxylation of termi-
nal alkynes at room temperature (Table 1, entry 1). Subsequently,
several commercial copper salts (CuCl, CuBr, CuI, and copper(I)-
thiophene-2-carboxylate (CuTC) were investigated. However, most
of the tested copper compounds exhibited poor to moderate activ-
ity in yield of 42–74% without any additive (entries 2–5). In partic-
ular, the privileged N-heterocyclic carbene (NHC) copper(I)
CuCl
CuBr
CuI
CuTC
IPrCuCl
IL-1
IL-2
IL-3
IL-4
CuBr + Im
CuBr + PPh
IL-3
IL-4
IL-4
IL-4
10
10
10
10
10
10
10
10
10
10
10
5
10
11
12
13
(
IPrCuCl) only afforded 9% yield (entry 6).
c
c
12
As depicted in Scheme 2, four types of copper(I)-containing ILs
3
were easily prepared from commercially available starting materi-
als and proved to be identical with authentic material reported in
14
5
10
10
3
4,35
1
1
5
6
the literature.
The catalytic performance of various ILs was
under the otherwise identical conditions.
6
evaluated in DMF/CO
2
a
Interestingly, the ILs incorporating copper solely in form of halo-
cuprate, that is, copper(I) in the anion (IL-1 and IL-2) gave a lower
activity (entry 7 and 8); whereas, the ILs containing copper(I) in
both the anion and the cation (IL-3 and IL-4), showed excellent
activity, affording 94% and 99% yield of 2a, respectively (entries 9
and 10), presumably suggesting that the catalyst activity is
remarkably affected by centrality copper(I) in the cation, originat-
ing from the coordination with 1-dodecylimidazole. As a result,
Reactions were performed with phenylacetylene 1a (0.0511 g, 0.5 mmol),
n
Cs
CO
2
CO
3
(0.1955 g, 0.6 mmol), catalyst, BuI (0.1104 g, 0.6 mmol), DMF (2.5 mL),
(99.999%, balloon) at room temperature.
2
b
Determined by gas chromatography (GC) with biphenyl as the internal
standard.
c
Physical mixture of CuBr (0.0072 g, 0.05 mmol) and ligand (0.05 mmol).
Based on Cu content.
d
copper(I)-based IL 4 ([Cu(Im12
2 2
) ][CuBr ]) was confirmed to be the
efficiently with [Cu(Im12
2 2 2
) ][CuBr ] (10 mol%) as the catalyst, Cs -
ideal catalyst (entry 10).
CO
2 h.
With the optimized conditions in hand, the generality of this
3
as the base, and DMF as the solvent at room temperature for
Furthermore, CuBr with alternative ligands such as 1-dode-
cylimidazole or triphenylphosphine could also mediate the car-
boxylation of 1a with 81% and 88% yields (entries 11 and 12).
We then examined the influence of the catalyst loading and reac-
tion time, respectively. Catalyst loading could not be decreased
to 5% because of noticeable decline in the yield (entries 13 and
1
process was also examined (Table 2). Phenylacetylenes substituted
with electro-rich groups at meta- or para-positions (e.g. Et-, Me-,
Pentyl-, and MeO-), could be converted to the corresponding
alkynoates 2a–2e in 88–96% yields (Table 2, entries 1–5). In addi-
tion, the substrate with weak electron-withdrawing group (e.g.
bromo-) could also give 85% yield of 2f (entry 6). Good results were
also achieved for the aliphatic alkyne, for example, 1-octyne, cyclo-
propyl acetylene, under standard conditions (entries 8 and 13). In
general, the presence of a strong electron-withdrawing group
could decrease the yield dramatically.
This poor efficiency may be due to the dropped nucleophilicity
of C–Cu bond. Accordingly, when raising the reaction temperature
to 40 °C, terminal aromatic alkynes with electron-withdrawing
group (F-, Cl-) and 3-ethynylpyridine were successfully converted
into the related carboxylic esters 2i–2l in good to excellent yields
1
5
4). When the reaction time was decreased to 9 h and 6 h, only
3% and 46% yield of 2a were observed, respectively (entries 15
and 16). As
a
result, the catalytic carboxylation proceeded
O
H
nBuI
Cu-IL
RT
+
CO2
balloon)
+
n
O Bu
R1
(
R1
Scheme 1. Copper(I)-based ionic liquids-promoted carboxylation of terminal
alkynes.
(
2
78–94%) with CO under mild reaction conditions (entries 9–12).