Table 1. Coupling of 1a and 2a under different reaction conditions.[a]
Table 2. Synthesis of polysubstituted furans via a cascade coupling/cycloi-
somerization process.[a]
Entry
Vinyl
Product
Yield [%][b]
iodide
Amino acid[c]
Base
Solvent
Yield [%][b]
1
75
Entry
1
2
3
4
5
6
7
8
A
B
B
B
B
B
B
B
Cs2CO3
Cs2CO3
K2CO3
K3PO4
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
dioxane
dioxane
dioxane
dioxane
DMF
DMSO
toluene
iPrOH
70
83
55
70
75
55
10
50
2
3
4
5
6
7c: R=4-CH3OC6H4
7d: R=4-FC6H4
7e: R=(CH2)2OBn
7 f: R=(CH2)2OTBS
7q: R=CH2CH2CH3
77
81
72
74
0[c]
7
89
[a] Reaction conditions: 1a (0.3 mmol), 2a (0.25 mmol), 10 mol% CuI,
30 mol% amino acid, base (0.75 mmol), solvent (0.5 mL), under Ar at-
mosphere, 808C, 36 h. [b] Yield of isolated product. [c] A: N,N-dimethyl-
glycine hydrochloride salt; B: l-proline.
8
58
tries 3 and 4). Among the other solvents tested, DMF pro-
vided the product in 75% yield (Table 1, entry 5), while low
yields were observed in the cases of DMSO, toluene, and
iPrOH (Table 1, entries 6–8). Thus, we concluded that CuI/
l-proline as the catalyst, Cs2CO3 as the base, and dioxane as
the solvent is an optimized combination for this cascade
process.
9
73
65
10
After identification of the optimized reaction conditions,
the scope and limitations of this cascade process were exam-
ined by varying vinyl iodides and 1-alkynes. As indicated in
Table 2, substituted aryl alkynes were able to couple with
2a, providing the corresponding 2,3,4-trisubstituted furans
7b–d in good yields (entries 1–3). The electronic nature of
the aromatic ring of aryl alkynes had little influence on the
reaction yield (compare entry 1 and 3). Two O-protected
propargyl alcohols gave the desired products 7e and 7 f
(Table 2, entries 4 and 5), while a simple aliphatic alkyne
did not give any furan product (entry 6). In the latter case a
coupling product was isolated in about 70% yield, indicating
that further cycloisomerization of this (Z)-2-en-4-yn-1-ol did
not take place under the present conditions. These results
demonstrated that subtle changes in electron nature of the
aliphatic alkynes could alter the cycloisomerization process.
A similar phenomenon has been noticed in our indole syn-
thesis.[6a] More studies are required to solve this problem.
Further investigation revealed that changing the substitu-
ents at the 1- or 2-positions of furans is possible, as shown
from the results of entries 7–10 in which furans 7g–j were
assembled from different 1,2,2-trisubstituted vinyl iodides
(Table 2, entries 7–10). Next, we attempted to synthesize
2,3,5-trisubstituted furans, and were pleased to find that two
2,2-disubstituted vinyl iodides were compatible with the re-
action conditions, providing the desired furans 7k and 7l in
good yields (Table 2, entries 11 and 12). When vinyl iodides
2g and 2h were employed, the reaction became sluggish,
and increasing the reaction temperature was required to
ensure good conversion (Table 2, entries 13 and 14). In the
case of vinyl iodide 2i as a substrate, the reaction proceeded
smoothly at 808C to give 2,3,4,5-tetrasubstituted furans 7o
11
12
13
79
68
67[d]
14
15
63[d]
76
16
74
[a] Reaction conditions: vinyl iodide (0.25 mmol), 1-alkyne (0.3 mmol),
CuI (0.025 mmol), l-proline (0.075 mmol), Cs2CO3 (0.75 mmol), 1,4-diox-
ane (0.5 mL), under Ar atmosphere, 808C, 36–48 h. [b] Yield of isolated
product. [c] A coupling product was isolated in about 70% yield. [d] The
reaction was carried out at 1008C.
and 7p (Table 2, entries 15 and 16). These results illustrated
that the steric hindrance or electronic nature of vinyl iodides
might influence the reaction rate. The formation of these
Chem. Asian J. 2010, 5, 74 – 76
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
75