Table 1. Catalytic Activity over Various Gold and Platinum Catalysts
entry
substratea
catalyst (mol %)
AuCl3 (5)
time
products (yields)b
1
2
3
4
5
6
7
8
9
5a
5a
5a
5a
5a
5a
5b
5c
5d
60 min
60 min
60 min
10 min
5 h
SM (86%)c
SM (72%)
SM (82%)
6a (92%)
6a0 (81%)
6a (83%)
SM (57%)
SM (84%)
ClAuPPh3 (5) /AgSbF6 (5)
ClAuPPh3(5) /AgNTf2 (5)
PtCl2/CO (5)
PtCl2/CO (5)
Ptl2 (5)
10 min
60 min
60 min
20 min
PtCl2/CO(10)
PtCl2/CO(10)
PtCl2/CO(10)
a [Substrate] = 0.1 M. b Product yields are reported after purification from a silica column. c Recovery yields of starting materials (SM) are given in
entries 1-3, 7, and 8.
Scheme 1. The accepted mechanisms involve the 1,3- or
stepwise 1,2-electrophilic migration of key intermediate 2
and 5q (E = H),6 whereas PtCl2/CO (10 mol %) was
employed for the remaining substrates bearing carbon-
based electrophiles. Most substrates contain an alkyl sub-
stituent (R = alkyl) to ensure the kinetic stability of
epoxide products 6. Furthermore, a large R substituent
induces a small θ angle to accelerate the cyclization via the
Thorpe-Ingold effect.7 Herein, the tricyclic ketals 6j00 and
6o00 were produced from the dimerization of epoxide
products 6a; these ketals showed proton NMR spectral
patterns distinct from those of epoxides 6. The structure of
6o00 was solved by X-ray diffraction.8 Entries 1-5 show the
applicability of this catalysis to substrates 5e-i bearing
various CR2 (R = methyl, ethyl, cyclopentyl, and
cyclohexyl) and allyl groups (R = allyl, 2-methyl,
2-phenylallyl), giving epoxide species 6e-i in 58-85%
yields. We examined this reaction with unsubstituted sub-
strate 5j (R = H, E = allyl, entry 6), which gave tricyclic
ketal 6j00 in 46% yield. This catalysis is applicable also to
substrate 5k and 5l bearing a p-methoxybenzyl (PMB)
ether that gave epoxides 6k and 6l in 82% and 83% yields,
respectively (entries 7 and 8). For substrate 5m, a brief
reaction (5 min) gave no initial epoxide in pure form, but a
longer period (3 h) delivered ketone 6m0 in 56% yield (entry 9).
The migration of a p-methoxybenzyl group is feasible
also for cyclohexyl substrate 5n that gave epoxide 6n in
83% yield (entry 10). Similar to 5j, unsubstituted substrate
5o (R = H, entry 11) gave dimerization product 6o00 of
which the structure was confirmed by X-ray diffraction.8
€
to form species 3, as Yamamoto and Furstner proposed
(see Scheme 1).1 In the literature, there appears no instance
of a violation of this mechanism in the Au- and Pt-
catalyzed 1,2-carbofunctionalizations of alkynes. Herein,
we report two distinct 1,2-carboalkoxylations ofalkynes as
manifested by the cycloisomerizations of 5-alkoxypent-1-
yn-3-ol derivatives 5 and 7 to form 2,6-dioxabicyclo-
[3.1.0]hexanes 6 or dihydro-2H-pyran-4(3H)-ones 8,
respectively.4
Table 1 shows our tests of activity of substrates 5a-d
over commonly used platinum and gold catalysts. For
alkynol 5a (R = H), the use of AuCl3, ClAuPPh3/AgSbF6,
and ClAuPPh3/AgNTf2, each at 5 mol %, led only to its
exclusive recovery (72-86%, entries 1-3). To our delight,
PtCl2/CO (5 mol %)2a,5 in CH2Cl2 gave 2,6-dioxabicyclo-
[3.1.0]hexane 6a in 92% yield within 10 min (entry 4); a
protracted period (5 h) gave ketone 6a0 through a second-
ary reaction of epoxide 6a. Similarly, PtI2 gave desired 6a
in 83% yield (entry 6). We examined the same reactions of
species 5b-d bearing a methoxymethyl (MOM), methoxy,
and siloxy group, respectively, but we either recovered
unreacted 5b and 5c (entries 8 and 9) or observed a
complete decomposition of starting 5d (entry 10). The
workability of species 5a reflects the important role of its
hydroxy group in this platinum catalysis.
We prepared substrates 5e-q to examine the generality
of this new catalysis (Table 2). AuCl3 (5 mol %) was used
to implement the cycloisomerization of diol substrates 5p
(6) We obtained products 6p and 6q in complicated mixtures of
products when diol substrates 5p and 5q were treated with PtCl2/CO
(5 min) in CH2Cl2 (25 °C, 10 min).
(4) PtCl2-catalyzed cycloisomerization of 2-propargyl anilines gave
indole products through a typical 1,2-addition pathway, with no epoxide
product 6 in this case. See: Cariou, K.; Ronan, B.; Mignani, S.;
Fensterbank, L.; Malacria, M. Angew. Chem., Int. Ed. 2007, 46, 1881.
(7) For the gem-dialkyl effect of this cyclization, see selected exam-
ples: (a) Kostal, J.; Jorgensen, W. L. J. Am. Chem. Soc. 2010, 32, 8766.
(b) Jager, J.; Graafland, T.; Schenk, H.; Kirby, A. J.; Engberts, J. B. F.
N. J. Am. Chem. Soc. 1984, 106, 139. (c) Beesley, R. M.; Ingold, C. K.;
Thorpe, J. F. J. Chem. Soc. 1915, 107, 1080.
€
(5) (a) Furstner, A.; Aissa, C. J. Am. Chem. Soc. 2006, 128, 6306. (b)
Chang, H.-K.; Datta, S.; Das, A.; Liu, R.-S. Angew. Chem., Int. Ed.
2007, 46, 4744.
(8) X-ray crystallographic data of compound 6o00 is provided in the
Supporting Information.
Org. Lett., Vol. 13, No. 7, 2011
1703