COMMUNICATION
1,1,2,3-tetrasubstituted cyclopropene (P) under our opti-
mized reaction conditions. Instead, tetrasubstituted furan 3a
was obtained, as confirmed by X-ray diffraction and NMR
spectroscopy (Scheme 1). A mechanistic study indicated an-
other Cu/Pd relay catalytic sequence that was different from
terion played an important role in the reaction (Table 1, en-
tries 4–6). To our delight, when three equivalents of DMSO
were added to MeCN as the additive, the efficiency of this
reaction was greatly improved.[16] The reaction was complete
within 1 h in 76% yield (Table 1, entry 7).[17] Increasing or
decreasing the amount of DMSO did not lead to better re-
sults (Table 1, entries 8 and 9). Under some conditions, a
small amount of trisubstituted furan 4a was isolated as the
by-product. Using other palladium catalysts (Table 1, en-
tries 10–12) or changing the temperature to 408C and 808C
(Table 1, entries 13 and 14) produced less-satisfactory re-
sults. Finally, the efficiency of the reaction was tested by
using 1 mol% of the catalyst, which gave the product in
48% yield, along with the isomerized furan 4a in 30%
yield, after 30 min (Table 1, entry 15).
À
the general C H metalation process. Herein, we report our
preliminary results.
Initially, we explored the viability of this DHR process in
the reaction of cyclopropene 1a with methyl acrylate (2a).
Gratifyingly, the use of Pd
ACHTUNGRTEN(NUNG OAc)2 (5 mol%) as the catalyst
and Cu(OAc)2 (2 equiv) as the oxidant at 608C in MeCN af-
ACHTUNGTRENNUNG
forded a major product in 21% yield. NMR and MS data
were in accord with the formation of the DHR product (P),
except that the resonance of the two methyl ester groups
from the cyclopropene was shifted. This puzzling question
was not finally resolved until we obtained the X-ray crystal
structure of the product. Indeed, the product was not cyclo-
propene P, but instead isomerized tetrasubstituted furan 3a
(Scheme 1).[15]
Inspired by this result, we continued to optimize the reac-
tion conditions to develop an efficient catalytic system for
the synthesis of the all-substituted furan. We began our ex-
ploration by screening different oxidants and we found that
Control experiments confirmed that both the copper and
palladium catalysts were necessary in this process. Only the
isomerized product (4a) was formed without the palladium
catalyst (Table 1, entry 16) and no product was formed with-
out the copper catalyst (Table 1, entry 17).
With the optimal conditions in hand, that is, the use of
PdACHTNUGTRENNUG(OAc)2 (5 mol%), CuHCATUNGTREN(NUGN OAc)2 (2 equiv), and DMSO
(3 equiv) at 608C in MeCN, we examined the scope of this
transformation. A series of cyclopropenes were reacted with
methyl acrylate and the results are summarized in Table 2.
Both aromatic cyclopropenes and aliphatic cyclopropenes
can all generate their corresponding products in moderate-
to-good yields. An unsubstituted cyclopropene dicarboxylate
substrate also provided 2,3,5-trisubstituted furan 3d in 58%
yield. Halogen atoms, such as Br and F, and a
phenol group were tolerated in this reaction. The
reaction of a cyclopropene that was derived from
ethyl acetoacetate was more difficult in this trans-
formation. Slow injection of the cyclopropene and
increasing the palladium catalyst loading to
10 mol% afforded the 2-alkyl-tetrasubstituted furan
(3k) in 53% yield. In addition, this reaction can be
only Cu
entries 1–6). The use of other oxidants (tert-butyl hydroper-
oxide, benzoquinone, PhI(OAc)2, air; see the Supporting In-
formation for details) resulted in no coupled product. Other
copper salts, including CuCl2, CuSO4, and Cu(ClO4)2, were
ACHTUNGTRENNUNG(OAc)2 and AgOAc gave promising results (Table 1,
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
also ineffective as oxidants, which indicated that the coun-
Table 1. Optimization of the reaction conditions.[a]
Entry
Catalyst
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
PdCl2
Pd(tfa)2
[Pd(PPh3)4]
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Oxidant
Cu(OAc)2
BQ
AgOAc
CuCl2
Additive (equiv)
Yield [%][b]
scaled up to the gram scale without showing a de-
crease in reaction efficiency and yield (see the Sup-
porting Information).
1
2
3
4
5
6
7
8
9
10
11
12
13[d]
14[e]
15[f]
16[g]
17[h]
G
E
–
–
–
–
–
21
0
34
ACHTUNGTRENNUNG
R
Next, the scope of the alkene was explored
(Table 3). A variety of electron-deficient alkenes,
including different acrylates and N,N-dimethyl
acrylamide, proceeded smoothly and efficiently to
produce the corresponding products in generally
good yields. More importantly, the reaction was suc-
cessful for disubstituted alkenes, such as methyl
crotonate (product 3p) and methyl methacrylate
(products 3q and 3q’). These results are remarka-
ble, owing to the low reactivity of disubstituted al-
kenes.[18] In this case, the reaction does not require
the addition of any ligand or much excess of the
alkene and the coupled product can be formed in
moderate yield within 2 h. With methyl methacry-
late, only trace amounts of 3q’ was formed, accord-
E
trace
trace
trace
76
65 (17)
31
60
76
42
32
R
CuSO4
R
Cu
A
–
N
Cu
Cu
Cu
Cu
Cu
Cu
Cu
Cu
Cu
Cu
N
DMSO (3)
DMSO (1)
DMSO[c]
DMSO (3)
DMSO (3)
DMSO (3)
DMSO (3)
DMSO (3)
DMSO (3)
DMSO (3)
DMSO (3)
N
R
R
R
E
T
58
A
48 (30)
0 (68)
0
–
PdN
–
[a] Reaction conditions: cyclopropene 1a (0.2 mmol), methyl acrylate (2a, 0.4 mmol),
Pd catalyst (5 mol%), oxidant (0.4 mmol), MeCN (1 mL), 608C. [b] Yield of isolated
product; the number in parentheses represents the yield of by-product 4a. [c] DMSO
was used as the solvent. [d] 408C. [e] 808C. [f] Catalyst (1 mol%), 30 min. [g] Only
1
ing to H NMR spectroscopy and TLC analysis (3q/
CuACHTUNGTRENNUNG(OAc)2. [h] Only PdACHTUNGTRENNUNG(OAc)2 (5 mol%) or [PdCl2ACHTNUGTRENN(UGN MeCN)2] (5 mol%). BQ=1,4-ben-
zoquinone, TFA=trifluoroacetate.
3q’, 10:1), and the major isomer (3q) was formed
Chem. Eur. J. 2013, 19, 3584 – 3589
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3585