7166
S. G. Lamont et al. / Tetrahedron Letters 56 (2015) 7165–7167
This hypothesis was further supported by observations made
with the corresponding benzothiophene 2 and benzothiophene
dioxide 3. In the case of fully aromatic benzothiophene, complete
regioselectivity was observed to exclusively give methyl com-
pound 2a. However, with benzothiophene dioxide a 2:1 mixture
of 3a and 3b was obtained (Scheme 3). Because in the case of 3
the central ring is not aromatic, it was rationalised that the relative
energy barriers of the transition states leading to 3a and 3b were
similar, therefore giving rise to a mixture of products.
These observations prompted us to investigate the synthetic
utility of this regioselective reaction with additional heterocy-
cles (Table 1). In the case of indoles with no substitution on
nitrogen, it was necessary to use tosyl protected derivatives to
avoid deprotonation by the Grignard reagent (Table 1). It should
be noted that the yields of these reactions are as yet
unoptimised.
N
N
MeMgCl
H
MgCl
1
1c
N+
N+
O
O
-O
-O
O
O
Cl
Cl
CN
CN
MeMgCl
OH
O
Cl
Cl
CN
CN
N
N
H
MgCl
1d
1a
N+
Mg
Cl
N+
O
O
-O
-O
Following our initial observations with carbazole 1, analogous
carbazoles 4 and 5 were investigated, and were noted to proceed
with similar selectivity to give methyl analogues 4a and 5a. A
similar outcome was observed when nitroindoles were used as
substrates. In the cases of 5-nitro indole 7 and 6-nitro indole 6,
the reaction proceeded with the expected selectivity, however
when 4-nitro indole 10 and 7-nitro indole 11 were employed,
an inseparable mixture of isomers was obtained. This was due
to competing alkylation occurring at the para position. Because
this competing reaction does not break the aromaticity of the
heterocyclic ring it is therefore still energetically favourable (see
Scheme 4).
Additionally, we were interested to see if this methodology
could be extended to other heterocyclic systems. In the case of
benzofuran 8 and benzothiophene 9, the reaction proceeded with
the expected selectivity to give the corresponding ortho methyl
compounds 8a and 9a.
disfavoured due to aromaticity
being broken in the pyrrole ring
Scheme 2. Proposed reaction pathway and unfavourable reaction intermediate.
S
S
1) MeMgCl
2) DDQ
N+
N+
O
2
3
O
O
48%
2a
3a
-O
-O
O
O
S
O
O
O
O
S
S
1) MeMgCl
2) DDQ
+
N+
N+
N+
O
O
-O
-O
3b
-O
34%
Other Grignard reagents could also be employed in this reac-
tion. When carbazole 1 was treated with ethyl magnesium chloride
ethyl substituted carbazole 1e was the only product isolated.
When the same carbazole was treated with isopropyl magnesium
chloride, the analogous isopropyl carbazole 1f was formed exclu-
sively. Unfortunately, when carbazole 1 was treated with phenyl
magnesium chloride the corresponding phenyl carbazole 1g was
not observed, presumably due to the lower nucleophilicity of the
phenyl Grignard reagent.
In summary, we have described a highly regioselective method-
ology for the alkylation of a variety of nitro containing aromatic
heterocycles. Given the highly versatile nature of the nitro func-
tional group we believe this methodology to be of synthetic value
and allows the expedient synthesis of a wide variety of non-com-
Scheme 3. Grignard additions to benzothiophene and benzothiophene dioxide.
Reagents and conditions: 2 or 3 (3.9 mmol), MeMgCl (5.9 mmol), THF (40 mL),
À15 °C, 1 h, then DDQ (6.69 mmol), À10 °C to RT, 16 h. 2a, 48% and 3a and 3b, 34%.
as this would require breaking the aromaticity of the carbazole
pyrrole ring. It was believed that the energy barrier to disrupting
the aromaticity was significant enough to give rise to the observed
selectivity. The reaction is completed by the addition of DDQ (4,5-
dichloro-3,6-dioxocyclohexa-1,4-diene-1,2-dicarbonitrile) which
furnishes the desired product. It is well known in the literature that
DDQ can be used to oxidise semi-saturated rings to aromatic sys-
tems.7,8 The proposed mechanism involves hydride transfer to
the quinone oxygen followed by transfer of magnesium to the phe-
nolate ion (Scheme 2).
Table 1
Substrate
Product
Yielda (%)
Substrate
Product
Yielda (%)
S
S
N
N
NO2
NO2
68
39
NO2
NO2
NO
2
9
9a
1
1a
4a
Tos
N
Tos
N
Tos
N
N
N
55
74
NO2
NO2
NO2
NO2
4
10
10a:10b(1:1 mixture)