J. Feng et al. / Tetrahedron Letters xxx (2014) xxx–xxx
3
Table 2
transformations, however, the former generally gave better results
Scope of the transannulations with 1,2-disubstituted alkyl enol ethersa,b,c
than the latter, particularly for the alkyl-substituted substrates
(e.g., 4g and 4h). Notably, for the substrate bearing an electron-
withdrawing carboxylate group, the reaction also worked well by
affording the corresponding product 4j in 82% yield. Finally, the
scope of the triazole partners was explored with an array of
4-Ar-1-tosyl-1,2,3-triazoles. Not surprisingly, all of the reactions
gave satisfying results (4k–p), showing little electronic or steric
effect.
Rh2(oct)4(1%)
DCE, 1 00 oC, 1-4 h;
th en TsOH, rt, 3 h
R2
N
N
N
Ts
N
Ts
OEt
+
R2
R1
R1
1
5
7
N
Ts
N
Ts
N
Ts
N
Ts
Besides the 1,1-disubstituted silyl enol ethers, we also
attempted the substrates with 1,2-disubstituted pattern. To our
disappointment, when the TBS-protected silyl ether 5aa (prepared
as a mixture of Z/E isomers in ratio of 1:0.9, 2.5 equiv) was
employed with 1a as the reaction partner, only a moderate yield
(40%) of the expected product 7a was obtained. We envisioned that
the increasing steric hindrance of the 1,2-disubstituted silyl enol
ether might decrease its nucleophilicity toward the Rh(II)-imino-
carbene intermediate. For this end, we turned to examine the less
steric alkyl enol ethers in the reaction. To our delight, while the
methyl enol ether 5ab (Z/E = 1:1) failed to give the desired product
in satisfying yield, the ethyl-derived substrate 5ac (Z/E = 20:1) dis-
played superior reactivity by affording 3,4-disubstituted pyrrole 7a
in yield of 90% (Scheme 3). Although a full rationalization of the
distinct behaviors among the above-examined silyl and alkyl enol
ethers (5aa, 5ab, and 5ac) remains yet to be achieved,14 these
observations indicated that both the steric and electronic features
of the enol ethers had profound influence on the reaction.15
With the suitable protecting group for the 1,2-disubstituted enol
ethers identified, we then evaluated the generality of the transfor-
mations. As shown in Table 2, an array of 4-Ar-1-tosyl 1,2,3-triazoles
bearing either electron-donating or withdrawing substituents
smoothly underwent the one-pot two step transformation with
5ac to afford the corresponding 3,4-disubstituted pyrroles 7a–e in
acceptable to good yields (47–90%). Comparably, the reaction seems
to be sensitive to the variants on the enol ether partner. Indeed,
while the substrates bearing an electron-deficient substituent were
amenable to the reactions (e.g., 7f and 7g), those ones carrying an
electron-donating aromatic ring failed to give the product in
synthetically useful yields (e.g., 7h). Furthermore, the reactions
could also be extended to the alkyl substituted enol ethers, as dem-
onstrated by the reactions leading to 7i–k. However, this type of
substrates showed relatively low reactivity and only resulted in
the corresponding products with moderate yields (30–60%).
To further extend the substrate scope, we also examined vari-
ous other types of substituted enol ethers (Table 3). As shown,
MeO
Me
F
7a: 90%
7b: 65%
7d: 56%
7c: 47%
Cl
F3
C
MeO
N
Ts
Ts
N
N
Ts
N
Ts
Cl
7e: 63%
7f: 55%
7g: 60%
7h : <5%
Me
Me
Me
Me
N
N
N
Ts
Ts
7i : 54%
Ts
7j: 60%
7k: 30%
a
The reaction was run with
(0.002 mmol) in DCE (1.0 mL) at 100 °C.
Refers to the isolated yield.
In most of the cases, the alkyl enol ethers were prepared as a mixture of Z/E
isomers, and used directly without further separation.
1 (0.2 mmol), 5 (0.4 mmol), and Rh2(oct)4
b
c
the ethyl vinyl ether 8a was turned out to be suitable substrates
by furnishing the monosubstituted pyrroles 9a in an excellent yield
(Table 3, entry 1). Besides, the cyclic 2,3-dihydrofuran 8b
and 3,4-dihydro-2H-pyran 8c could respectively lead to the disub-
stituted pyrroles 9b and 9c (entries 2 and 3), although the former
showed superior reactivity than the latter. Moreover, several
trisubstituted cyclic enol ethers 8d–f were evaluated, among
which, the more strained 1-ethoxycycloheptene (8e) and
1-ethoxycyclooctene (8f) afforded the better yields than the
1-ethoxycyclohexene (8d) (entries 4–6). Finally, it was found that
by giving a low yield of 9g (entry 7), most likely attributed to the
acyclic trisubstituted enol ether 8g exhibited poor reactivity unfa-
vorable steric effect associated with this type of substrates.
Finally, we also explored the possibility of employing the silyl
ketene acetals as the [2C] component in the transformations.16
As the proof-of-concept cases, three representative silyl ketene
acetals 10a–c (R = H, Me or Ph) were chosen as substrates. Interest-
ingly, the corresponding transannulation products 13a–c did form
at the beginning,17 however, they were found to be unstable and
readily underwent the deprotection of the silyl group followed
by double bond isomerization (Scheme 4) to afford the 3-pyrro-
lin-2-ones 11a–c as final products. Given that 3-pyrrolin-2-one
Ts
N
Rh2(oct)4(1%)
OR
DCE, 100 oC, 1 h
RO
5aa: R = TBS; Z:E = 1:0.9
5ab: R = Me; Z:E = 1:1
5ac: R = Et; Z:E = 20:1
6aa: R = TBS
6ab: R = Me
6ac: R = Et
represents
a key structural element in various bioactive
molecules,18 this transformation provides a new avenue to access
+
40% for 5aa
<5% for 5ab
90% for 5ac
TsOH, rt
1 h
the related aza-heterocycles.19
N
N
N
In summary, we have successfully developed a highly efficient
method for the synthesis of substituted pyrroles via a novel
Rh(II)-catalyzed transannulation of 1-tosyl-1,2,3-triazoles with
silyl or alkyl enol ethers. Moreover, the method could also be
extended to the synthesis of 3-pyrrolin-2-ones by employing silyl
ketene acetals as the reaction partner. Several advantages of the
described methodology, including the readily available starting
material, high efficiency, broad substrate scope, and controllable
regioselectivity, render it an alternative to the current toolbox of
Ts
1a
Ts
N
7a
Scheme 3. Preliminary results for the transannulations of 1-tosyl-1,2,3-triazoles
with 1,2-disubstituted silyl or alkyl enol ethers.