Table 1. Optimization of the reaction conditions.[a]
During the further examina-
tion, we also found that 2a
could not be obtained upon
treatment of 1a with the other
Lewis acids (5 mol%) or
Brønsted acids (5 mol%) such
Entry
Catalyst
Additive
Solvent
T [oC]
t [h]
Yield [%][b] (2a/3a)
as LnACHTUNGTRENNUNG(OTf)3 or HOTf (Tf=
1
2
3
PPh3AuCl/AgOTf
PPh3AuCl/AgOTf
PPh3AuCl
–
–
–
DCE
DCE
DCE
50
20
50
24
24
24
99 (75:25)
trace
–
CF3SO3) under identical condi-
tions. Moreover, it was found
that this reaction gave 2a in
trace at room temperature
(Table 1, entry 2). In the ab-
sence of AgOTf, no reaction
occurred, suggesting that the
in situ generated Au+ species
is the real catalyst (Table 1,
entry 3). Using AgOTf as the
catalyst afforded 2-isopropyl-
1,4-diphenylnaphthalene as the
single product. This result was
very similar as that of
BF3·OEt2-catalyzed result, sug-
gesting that H+ might be the
real catalyst in this reaction
(Table 1, entry 4).[5a] Adding
methanol to the reaction
system did not improve the
yield and selectivity (Table 1,
entry 5).[10] To clarify the influ-
ence of trace of H+ in this re-
action, we added acetic acid
(50 mol%) or various bases
4
AgOTf
–
DCE
50
24
5
6
PPh3AuCl/AgOTf
PPh3AuCl/AgOTf
PPh3AuCl/AgOTf
PPh3AuCl/AgOTf
PPh3AuCl/AgOTf
PPh3AuCl/AgOTf
AgOTf
PPh3AuCl/AgOTf
PPh3AuCl/AgOTf
PPh3AuCl/AgOTf
PPh3AuCl/AgOTf
PPh3AuCl/AgOTf
PPh3AuCl/AgSbF6
PPh3AuCl/AgBF4
PPh3AuCl/AgOTf[d]
PPh3AuCl/AgOTf[e]
MeOH
CH3CO2H
Et3N
DCE
DCE
DCE
DCE
DCE
DCE
DCE
toluene
THF
DCM
MeOH
MeCN
DCE
DCE
DCE
DCE
50
50
50
50
50
50
50
50
50
20
50
50
50
50
50
50
10
10
10
10
10
10
10
10
10
24
24
24
10
10
10
10
94 (76:24)
90 (82:18)
13 (100/0)
90 (83:17)
90 (76:24)
97 (100:0)
–
trace of 2a
–
14 (100:0)
–
7[c]
8
TTBP
tBuOK
DBU
DBU
DBU
DBU
DBU
DBU
DBU
DBU
DBU
DBU
DBU
9
10
11
12
13
14
15
16
17
18
19
20
–
94 (100:0)
88 (100:0)
99 (77:23)
96 (84:16)
[a] All reactions were carried out using 1 (0.2 mmol), additive (50 mol%) in the presence of catalyst (5 mol%)
in various solvents (1.0 mL). [b] Isolated yield, the ratio of 2a/3a was determined by 1H NMR spectroscopic
data. [c] 80% of 1a was recovered. [d] 10 mol% of DBU was added. [e] 20 mol% of DBU was added.
(50 mol%) such as Et3N, 2,4,6-tri-tert-butylpyridine (TTBP)
and KOtBu as the additives to examine the reaction out-
come. Adding acetic acid did not cause significant alteration
(Table 1, entry 6). In the presence of Et3N, most of starting
materials were recovered and 2a was obtained in 13% yield
(Table 1, entry 7).[11] As for TTBP or KOtBu, which only can
act as a base to neutralize the trace of H+ in the reaction
system, the similar result was obtained as that of PPh3AuCl/
AgOTf, suggesting again that Au+ is the real catalyst in this
reaction (Table 1, entries 8 and 9). To our delight, using 1,8-
With these optimal conditions in hand, we next examined
a variety of arylvinylcyclopropenes 1 in this reaction and the
results of these experiments are shown in Table 2. When
electron-rich arylvinylcyclopropenes 1b and 1c were used as
the substrates, the corresponding products 2b and 2c were
produced in excellent yields in the presence of PPh3AuCl/
AgOTf (Table 2, entries 1 and 2). However, in the cases of
arylvinylcyclopropenes 1d–f having electron-withdrawing
groups on the benzene rings, traces of 2d–f were obtained
under the standard conditions. Fortunately, if using
PPh3AuCl/AgSbF6 (Table 2, entries 3–5) or PPh3AuCl/
AgBF4 (Table 2, entry 6) as the catalyst, products 3d–f could
diazabicycloACHTUNGTRENNUNG[5.4.0]undec-7-ene (DBU) as an additive afford-
ed 2a as a single product in 97% yield, indicating that weak
coordinating amine can adjust the catalytic ability of gold(I)
(Table 1, entry 10).[11] In the presence of DBU, AgOTf did
not promote the rearrangement of 1a, suggesting clearly
that H+ is the real catalyst in entry 4 of Table 1 (Table 1,
entry 11). The examination of solvent effects revealed that
DCE is the best one for the reaction (Table 1, entries 12–
16). Other silver salts such as AgSbF6 and AgBF4 could also
produce 2a in excellent yields as a sole product in the pres-
ence of DBU (Table 1, entries 17 and 18). We also examined
10 mol% and 20 mol of DBU in this reaction and found
that 2a and 3a were both formed in the ratios of 77:23 and
84:16, respectively (Table 1, entries 19 and 20).
be produced in good to excellent yields. As for spiro
ACHTUNGTRENNUNG[2.5]oct-
ene (1g) and spiro[2.4]hept-ene (1h) as well as (2-(2-(2-
AHCTUNGTRENNUNG
chlorophenyl)-3,3-dimethylcycloprop-1-enyl)ethene-1,1-di-
yl)dibenzene (1i) which has an ortho-substituted chloro
atom on the aromatic ring of R2, the reactions could also
proceed smoothly to produce indene products 2g, h and i in
good yields in the presence of PPh3AuCl/AgSbF6 (Table 2,
entries 7–9). When there was a meta group on the aromatic
ring of R2, the product mixtures of 5-chloro-2-(2,2-diphenyl-
vinyl)-1,1-dimethyl-1H-indene and 7-chloro-2-(2,2-diphenyl-
vinyl)-1,1-dimethyl-1H-indene were obtained in a 1:1 ratio
and in excellent yields (Table 2, entry 10). Either R3 or R4
was hydrogen atom or both of them were hydrogen atoms,
10220
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 10219 – 10222