E. Villedieu-Percheron et al. / Tetrahedron Letters 55 (2014) 2446–2449
2447
decomposition and loss of material. Some of these products were
found to be light sensitive and all the manipulations (including
workup, purification, and evaporation) were carried out in the
dark.11 The reaction tolerates a large variety of amides; even hin-
dered nitrogen substituents gave good results. However, the mor-
pholine derivative 3f was isolated in low yield. Even an increase in
temperature to 60 °C did not lead to full conversion. Side-product
formation was observed during the reaction which led to a lower
yield of isolated product 3f. In this particular case, the product it-
self could be reactive toward ketene-iminium salt due to the nucle-
ophilic oxygen atom of the morpholine ring, leading to loss of
material.
O
OH
O
N
O
N
iii)
i) ii)
I
I
1a
2a
Scheme 1. Preparation of N,N-dimethyl-2-(2-vinylphenyl)acetamide 2. Reagents
and conditions: (i) oxalyl chloride (2 equiv), DMF (cat.), CH2Cl2; (ii) NHMe2
(4 equiv), CH2Cl2, 0 °C, 99%; (iii) vinylstannane (1.5 equiv), Pd(PPh3)4 (3 %), toluene,
reflux, 16 h, 84%.
on the yield of 3a, as expected for ketene-iminium chemistry
(entry 3).
We then focused on the influence of different substituents on
the aromatic ring to the outcome of the 6p/10p-electrocyclization
Higher amounts of sym-collidine were detrimental (entry 4).
Stirring in dichloromethane between ambient temperature and re-
flux gave the best results. When the reaction was carried out below
room temperature or above 40 °C, decomposition occurred (entries
5, 8 and 10). The use of 2,6-ditertbutyl-4-methylpyridine did not
improve the yield (Entry 7). When the reaction was carried out
in the dark, the naphthalene derivative 3a was obtained in similar
yield (entry 9). Changing the solvent to acetonitrile did not lead to
any further improvements (entry 11).
Having established suitable reaction conditions, we then
focused on the scope by varying substituents (R1, R2) on the amide
2a–h (Table 2), on the aromatic ring (R3) 2i–k (Table 3) and on the
vinyl group (R4, R5) 2l–q (Table 4).
We started by varying the amide moiety (Table 2). Linear and
cyclic alkyl chains were chosen, as well as aryl substituents which
would lead to more or less hindered and/or electrophilic ketene-
iminium intermediates. These amides 2a–h were prepared accord-
ing to the previously described two-step synthesis (Scheme 1).
Table 2 describes the first set of different naphthalene deriva-
tives obtained by using 2.4 equiv of sym-collidine, 1.1 equiv of
triflic acid, in dichloromethane (0.05 M) at 40 °C. For some of these
reactions, 1.8 equiv of Tf2O were necessary to reach full conversion.
Increasing the amount or triflic acid led to decomposition.
Moreover, product 3g was obtained by conducting the reaction at
room temperature. Once again, higher temperatures led to
(Table 3). Full conversion was realized with 1.8 equiv of Tf2O under
reflux, although 1.5 equiv were sufficient at room temperature. Ta-
ble 3 describes the results obtained for various derivatives. Yields
decreased when the reaction was carried out under reflux, proba-
bly due to the degradation of the products under the reaction con-
ditions. Working at room temperature improved the results with
completion after 10 min. For example the product 3j where
R3 = OMe was rather unstable in the reaction mixture.
We then turned our attention to the influence of substituents on
the olefin. Table 4 summarizes the effect of varying substituents R4
and R5 on the olefin part on the reaction outcome. Compound 3l
was isolated in 60% yield despite its light sensitivity. Interestingly,
the tricyclic naphthylamines 3m–o were isolated in better yields
although the steric hindrance at the extremity of the double bond
could have decreased the reactivity of such reagents. The tricyclic
structures of the products 3m–o do not allow the coplanar orien-
tation of the amine and of the adjacent aromatic ring due to steric
repulsion between the saturated ring and the N,N-diisopropyl moi-
ety. Consequently, the derivatives 3m–o were less light sensitive
than the other products we have described previously. These types
of structures 3m–o are so far difficult to access by other methodol-
ogies.12 Electron-donating groups were introduced on both posi-
tions of the double bond. Whereas 3p (R4 = OEt, R5 = H) was
obtained in very good yield, compound 3q (R4 = H, R5 = OEt) could
not be obtained at all. In the latter case, only starting material was
O
N
N+
N
Tf2O
-H+
N+
C
sym-collidine
2a
3a
4a
Scheme 2. Electrocyclization of ketene-iminium salt 4a to naphthylamine 3a.
Table 1
Optimization of the condition for electrocyclization of amide 2a to naphthylamine 3a
Entry
sym-Collidine (equiv)
Triflic anhydride (equiv)
Concentration of 2a (mol LÀ1
)
Solvent
T (°C)
Yield (%)
1
2
3
4
5
6
7
8
1.2
2.4
2.4
5
2.4
2.4
2.4a
2.4
2.4
2.4
2.4
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
0.05
0.05
0.02
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
(CH2)2Cl2
CH2Cl2
CHCl3
rt
rt
rt
rt
0
40
rt
80
40
60
40
48
58
54
44
23
58
0
39
60
Traces
35
9b
10b
11
MeCN
a
2,6-Diterbutyl-4-methylpyridine.
The reaction was carried out in the dark.
b