Table 2. Substrate scope of the gold(I)-catalyzed cycloisomerization.
2a was confirmed by NMR spectroscopic data and X-ray
diffraction analysis (see the Supporting Information).[8] Car-
rying out the reaction in the presence of PtCl2 or
[RhCl(CO)ACHTUNGTRENNUNG(PPh3)2]/ACHTUNGTRENNUNG[AgSbF6] (5 mol%), which were effec-
tive catalysts in other similar reactions, afforded the desired
product 2a in lower yield or complex product mixtures, re-
spectively (Table 1, entries 2 and 3).[3b–h,j,m,4d] Control experi-
ments indicated that by using [(PPh3)AuCl] or AgOTf alone
as the catalyst did not promote the reaction (Table 1, en-
tries 4 and 5). Further examination of solvent effects re-
vealed that dichloromethane (DCM) was the solvent of
choice, and 2a was formed in lower yield in other organic
solvents, such as toluene, MeCN, 1,4-dioxane, or CHCl3
(Table 1, entries 6–10). Moreover, adding [(PPh3)AuCl]/
AgOTf (10 mol%) afforded 2a in 44% yield (Table 1,
entry 11). Changing silver salt to AgSbF6 or AgBF4 did not
improve the reaction outcomes (Table 1, entries 12 and 13).
By using [(PMe3)AuCl] or [(tBu3P)AuCl] instead of
[(PPh3)AuCl] as the gold catalyst gave 2a in 47 and 90%
yields, respectively, and [(IPr)AuCl] (IPr=1,3-bis(2,6-diiso-
propylphenyl)imidazol-2-ylidene) as well as AuCl3 were not
effective gold catalysts in this reaction (Table 1, entries 14–
Entry[a]
1
X
R
Yield [%],[b]
2
1
2
3
4
5
6
7
8
1b
1c
1d
1e
1 f
1g
1h
1i
NTs
NTs
NTs
NTs
NTs
NTs
NTs
NTs
NNs
NBs
O
4-MeOPh
4-MePh
3-MePh
2-BrPh
4-ClPh
1-naphthyl
Me
Et
Ph
Ph
Ph
2b, 87
2c, 95
2d, 90
2e, 98
2 f, 99
2g, 99
2h, 73
2i, 87
2j, 88
2k, 99
2l, 71
2m,80
9
1j
1k
1l
10
11
12
1m
O
4-ClPh
13[c]
1n
1o
1p
2n, 94
NR
14
15
3p, 71
17). In the presence of [(tBuXPhos)AuACTHNUTRGEN(UNG NCMe)]SbF6
(Figure 1, cat. 1), 2a could be
obtained in 94% yield under
otherwise identical conditions
(Table 1, entry 18). Therefore,
the optimal reaction conditions
have been identified to carry
of out the reaction in DCM at
room temperature by using
[a] All reactions were carried out by using 1 (0.2 mmol) in the presence
of cat. 1 (5 mol%) in DCM (2.0 mL) at RT for 15–30 h. [b] Yield of iso-
lated product. [c] Molecular sieves (4 ꢂ, 50 mg) was added. Ts=4-tolue-
nesulfonyl; Ns=4-nitrobenzenesulfonyl; Bs=4-bromobenzenesulfonyl;
NR=no reaction.
Figure 1. Structure
[(tBuXPhos)Au(NCMe)]SbF6.
ACHTUNGTRENNUNG
strate 1n revealed that the azabicycloACTHNUTRGNEUNG[4.1.0]heptene deriva-
[(tBuXPhos)Au
(5 mol%) as the catalyst.
(NCMe)]SbF6
tive 2n could be obtained in 94% yield, when 4 ꢂ molecular
sieves was added (Table 2, entry 13). As for 1,7-enyne 1o,
no reaction occurred under the standard conditions (Table 2,
entry 14). When the terminal of alkyne moiety is a hydrogen
atom (substrate 1p), the corresponding 1,3-diene 3p could
be obtained in 71% yield rather than the tricyclic product
(Table 2, entry 15).[9] The product structures of 2b–n were
determined by NMR spectroscopic data, MS, and HRMS.
The structure of compound 3p was confirmed by NMR
spectroscopic data and X-ray diffraction analysis (see the
Supporting Information).[10]
We next examined the substrate generality of the reaction
under the optimized conditions, and the results are shown in
Table 2. As can be seen, for nitrogen-tethered substrates
1b–d with an electron-donating group on the benzene ring,
such as MeO and Me groups, the reactions proceeded
smoothly to give the desired products 2b–d in 87–99%
yields (Table 2, entries 1–3). When the electron-withdrawing
groups, Br or Cl, were introduced on the benzene ring (sub-
strates 1e and 1 f), the corresponding products 2e and 2 f
were obtained in 98 and 99% yields, respectively (Table 2,
entries 4 and 5). The aromatic group of 1 could also be
a naphthyl group (substrate 1g), giving the cycloisomerized
compound 2g in 99% yield (Table 2, entry 6). As for sub-
strates 1h (R=Me) and 1i (R=Et) with an alkyl group at
the terminal of alkyne moiety, tricyclic compounds 2h and
2i could be formed in 73 and 87% yields, respectively
(Table 2, entries 7 and 8). In the case of other N-sulfonated
amines (X=NNs or NBs), the reaction also proceeded
smoothly to give the desired products 2j and 2k in 88–99%
yields, indicating a broad substrate scope of this reaction
(Table 2, entries 9 and 10). For oxygen-tethered alkylidene-
cyclopropanes, such as substrates 1l and 1m, the reactions
produced 2l and 2m in 71 and 80% yields, respectively
(Table 2, entries 11 and 12). Further examination of sub-
On the other hand, in the case of 1,6-enyne 1n, under the
standard conditions, the reactions produced 2n, 4n, and
5n[11] in 49, 15, and 29% yields, respectively. Then, adding
molecular sieves (4 ꢂ, 50 mg) into the reaction system,
azabicycloACTHNUTRGNEUG[N 4.1.0]heptene derivative 2n could be obtained in
94% yield as the sole product. Alcohol derivative 5n was
formed in 77% yield along with a trace amount of 2,3-dihy-
drofuran derivative 4n, when H2O (10 equiv) was added,
suggesting that products 4n and 5n are derived from adven-
titious water (Scheme 2). The structure of compound 4n was
confirmed by NMR spectroscopic data and X-ray diffraction
analysis (see the Supporting Information).[12]
Plausible mechanisms for these reactions are outlined in
Scheme 3 on the basis of above results. Coordination of AuI
complex to the alkyne forms intermediate 1, which evolves
&
2
&
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!