J. L. MascareÇas, F. Lꢀpez, G. Ujaque et al.
Table 1. Au ([(IPr)AuCl]/AgSbF6)-catalyzed [4C+3C] cycloaddition of 1.[a]
product must arise from
a
competitive Au-catalyzed
Entry
Allenediene, 1
Cycloadducts (6, 7)
6:7
(ratio)[b]
Yield
ACHTUNGTRENNUNG
(time)[c]
G
[4C+2C] annulation process,
as the reaction doesn’t proceed
under thermal activation.[11] In
this particular case we found
that the [4C+3C] cycloadduct
7i can be more efficiently ob-
tained by using either AuCl
(74% yield, 2 h) or AuCl3
(70% yield, 2 h) (Table 1, en-
tries 10 and 11).[12]
Finally, as an additional ex-
perimental mechanistic proof,
precursor 1b was treated with
[(IPr)AuCl]/AgNTf2 (10 mol%)
in the presence of diphenyl
sulfoxide (4 equiv), conditions
previously used for oxidative
trapping of gold carbenes.[13] In
addition to observing the for-
mation of traces of 6b and 7b
and recovering unreacted 1b,
we could isolate the ketone 9b
(25% yield, Scheme 3). This
result is consistent with the
participation of carbene spe-
1
1b
6b:7b (1:2)
85 (3)
2
3
4
5
1c, R=Me
1d, R=Ph
1e, R=tBu
1e
6c:7c (1:3)
82 (1)
7d
7e
7e
50 (12)
57 (20)
85 (3)[d]
6
7
8
1 f, R1 =Me[e]
1g, R2 =Me[e]
1h, R3 =Me[e]
6 f
6g
6h
93 (2)
84 (2)
77 (2)
9
1i
1i
1i
7i:8i (3:1)
7i
7i
66 (3)
74 (2)
70 (2)
10[f]
11[g]
[a] Conditions: [(IPr)AuCl] (10%) and AgSbF6 (10%) in CH2Cl2 (0.15m) at room temperature (20–258C)
unless otherwise noted. X=C(CO2Et)2, Y=C(CO2Me)2, and Z=NTs. [b] Determined by NMR spectroscopy.
[c] Yield of isolated product (%) after the reaction time (hours). [d] Reaction at 858C in 1,2-dichloroethane.
[e] Unspecified substituents are H. [f] Carried out with AuCl (10 mol%) at room temperature. [g] Carried out
with AuCl3 (10 mol%) at room temperature.
G
ACHTUNGTRENNUNG
whereas AuCl, AuCl3, [Ph3PAuCl]/AgSbF6, chloro[(1,1’-bi-
phenyl-2-yl)di-tert-butylphosphine]gold(I)/AgSbF6 or (di-
cies of type 4 in the catalytic cycle, and confirms the poten-
tial of the method to directly obtain stereochemically rich
oxygenated bycyclic products.
chloropyridine-2-carboxylato)goldACTHNUTRGNEUNG
(III),[9] led to low conver-
sions and/or formation of several unidentified side products,
the reaction of 1c with A (10 mol%) provided the cycload-
ducts 6c and 7c in 82% combined isolated yield, in less
than 1 h at room temperature (Table 1, entry 2).
The reaction also proceeds with furan derivatives 1d and
1e, providing the expected cycloadducts with complete se-
lectivity (Table 1, entries 3–5). As expected, sterically hin-
dered substrate 1e (Table 1, entry 5, R=tBu) resulted in a
slower reaction, but full conversion could be achieved after
3 h at 858C, providing the cycloadduct 7e with complete se-
lectivity and excellent yield. The above-mentioned results
(Table 1, entries 1–5) clearly indicate that the use of com-
plex A is critical for the success in cycloadditions involving
terminally monosubstituted allenes such as 1b–e. Notewor-
thy, the well known synthetic potential of oxabridged sys-
tems similar to those present in products 7c–e, bodes well
for the potential applicability of the method in the prepara-
tion of target-relevant bicarbocyclic systems.[10]
As can be seen in Table 1, the presence of substituents at
the diene unit was well tolerated (1 f–h, Table 1, entries 6–
8), and the corresponding [4C+3C] cycloadducts were ob-
tained in good yields (77–93%), and with complete stereo-
control. Surprisingly, the reaction of cyclopentylallene deriv-
ative 1i in the presence of catalyst A, in addition to provid-
ing the expected tricycle 7i that results from a 1,2 alkyl shift
in the intermediate carbene species, led also to a minor cy-
cloadduct identified as the bicycle 8i (Table 1, entry 9). This
Scheme 3. Oxygen atom transfer from the sulfoxide to the carbene
gold(I) species.
In summary, we have discovered that the Au complex
generated in situ from [(IPr)AuCl] and AgSbF6 catalyzes
the [4C+3C] intramolecular cycloaddition of a variety of al-
lenedienes at room temperature. The new conditions repre-
sent a significant step forward in terms of the scope and ver-
satility with respect to the previous PtCl2-catalyzed process,
and opens the door for the development of asymmetric ver-
sions. According to DFT calculations, the cycloaddition
must involve the formation of a metal–allyl cation inter-
mediate and its subsequent concerted cycloaddition with a
1,3-diene. These studies point to a 1,2-hydride shift on the
generated carbene intermediate as the rate-limiting step of
the process. The mechanistic insights derived from the de-
scribed theoretical and experimental studies, in addition to
providing a detailed account of the transformation, might
help to explain and design other gold-mediated processes.
3338
ꢄ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 3336 – 3339