1
468
R. P. A. Cruz et al. / Tetrahedron Letters 42 (2001) 1467–1469
Table 1. Friedel–Crafts acylation of N-acetylindole–AlCl complex
3
Entry
Acylating agent
a-(CH CO) O
(R)
5 (%)
6 (%)
1
2
CH3
CH CH CO H
90
75
96
92
3
2
2
2
2
3
4
5
c-(ClCH CO) O
CH Cl
CH3
71
65
80
92
95
96
2
2
2
d-CH COCl
3
e-ClCH COCl
CH Cl
2
2
though unexpected, are of great significance since there
are no reports in the literature about the formation of
2
-acylindoles unsubstituted at the 3-position under
Friedel–Crafts conditions.
Figure 2.
It is particularly intriguing that while the succinic anhy-
dride gives only substitution at the 6 position with
succinyl chloride the reaction site is the 2 position. The
mechanism that operates in these acylations with
dichlorides is not clear. A rearrangement of a 3-acyl
compound that could have been formed initially was
discharged since no other compound was detected by
TLC during the reaction time. A possible explanation
could be an ‘ortho’ effect involving the N-acyl group
(Scheme 1), which is unlikely in the case of the
anhydrides.
Fig. 2. This intermediate would be a ‘meta’ director
group towards the benzene ring, favoring substitution
at C4 and C6 and, not at C3, and, at the same time, it
should deactivate the heterocyclic ring towards elec-
trophilic substitution.
In order to test this hypothesis, we added 1-acetylindole
to a stirred suspension of AlCl in dichloromethane
3
followed by the acylating agent. Surprisingly, these
reaction conditions worked fairly well; giving, as the
only product, the 1,6 diacylated compounds in good
yields for both acid chlorides and anhydrides as acylat-
ing reagents (Table 1). However, when the reaction was
carried out without the previous formation of the
In summary, regioselective acylations at the 6 position
with acid chlorides as well as anhydrides were accom-
plished in good yields. An inedited direct acylation at
the 2 position of N-acylindoles by using acid dichlo-
rides was discovered. This represents an advance in
indole chemistry and the synthesis of indole alkaloids
by using this methodology is in progress.
AlCl –N-acetylindole complex, 1,3-diacetylindole was
3
isolated in 90% by using acetic anhydride as acylating
agent and in 55% yield for the acetylchloride case. The
latter conditions also led to the production of 1,6-
diacetylindole in 23%. However, it seems that the
regioselectivity does not depend on the nature of the
acylating agent when the complex is used as the sub-
strate, in contrast to what was suggested by Nakatsuka,
who observed such dependence.
Typical experimental procedures for the acylations of
N-acetylindole are described below.
6
-Acylation: AlCl (2.67 g, 20 mmol) was added to a
3
stirred solution of 1-acetylindole (800 mg, 5 mmol) in
CH Cl (30 mL), resulting in a red solution. After 15
Acylations at the 2 position. 2-Acylindoles are usually
made by the Fischer indole synthesis, lithiation of the
indolic ring followed by acylation, or by electrophilic
substitution of 3-alkylindoles as reported by Jackson.
Surprisingly, during the acylations of N-acetylindole by
using acid dichlorides, the only products observed were
the 1,2-diacylindoles 7 and 8 (Scheme 1). These results,
2
2
1
minutes, a solution of the acylating agent (20 mmol) in
0 mL of CH Cl was added dropwise. The reaction
6
1
2
2
7
was stirred at room temperature until no starting mate-
rial could be detected by TLC. The mixture was poured
into cold water and extracted with ethyl acetate. The
organic extracts were dried with Na SO and the sol-
†
2
4
vent removed in vacuo. The crude product was chro-
matographed on silica, to yield a pure compound.
†
1
7
4
(
. 2D H NMR (DMSO-d ): l 11.7 (s, broad, NH); 7.0–7.4 (m, 5H);
6
−
1
1
.2 (s, CH ). IR (KBr): 3300, 1780, 1658 cm 8. 2D H NMR
DMSO-d ): l 11.4 (s, broad, NH); 7.1–7.9 (m, 5H); 3,3 (t, J=7.5,
CH ). 2.65 (t, J=7.5, CH ), 3323, 1714, 1649 cm
2
2
-Acylation: AlCl (2.67 g, 20 mmol) was added to a
stirred solution of the acylating agent (20 mmol) in
3
6
−
1
.
3
3