Radical intermediates in the peroxidation of indoles

Article abstract of DOI:10.1039/b102915k

2-Substituted and 1,2-disubstituted indoles react with m-chloroperbenzoic acid and hydrogen peroxide in the presence of acid or calcium chloride affording 2- and 3-(3-oxoindol-2-yl)indoles; whereas 2,3-disubstituted indoles, reacting with the same oxidant

Full text of DOI:10.1039/b102915k

View Article Online / Journal Homepage / Table of Contents for this issue
Radical intermediates in the peroxidation of indoles
Paola Astolfi,a Lucedio Greci,a Corrado Rizzoli,b Paolo Sgarabottob and Giancarlo Marrosuc
a Dipartimento di Scienze dei Materiali e della Terra, Università, Via Brecce Bianche,
I-60131 Ancona, Italy
2
b Dipartimento di Chimica Generale ed Inorganica, Chimica Analitica, Chimica Fisica,
Università, Centro di Studio per la Strutturistica Diffrattometrica del CNR,
Viale delle Scienze, I-43100 Parma, Italy
c Dipartimento di Ingegneria Chimica Università di Roma “La Sapienza”,
Via Del Castro Laurenziano 7, I-00161 Roma, Italy
Received (in Cambridge, UK) 30th March 2001, Accepted 20th July 2001
First published as an Advance Article on the web 14th August 2001
2-Substituted and 1,2-disubstituted indoles react with m-chloroperbenzoic acid and hydrogen peroxide in the
presence of acid or calcium chloride affording 2- and 3-(3-oxoindol-2-yl)indoles; whereas 2,3-disubstituted indoles,
reacting with the same oxidants, lead to the formation of products typical of pentaatomic ring opening. The reaction
mechanisms are discussed in terms of electron transfer processes based on the redox potentials of the reagents, the
Marcus theory and the reaction products distribution. The reactions of 1-hydroxy-2-phenylindole, which yield
2-phenylisatogen (2-phenyl-3-oxo-3H-indole 1-oxide), bisnitrone and 3-(3-oxoindol-2-yl)indole are also explained
by an electron transfer mechanism depending on the oxidant and on the conditions of the reaction. The structures
of 2- and 3-(3-oxoindol-2-yl)indoles have been elucidated by X-ray analysis.
electron transfer and their rationalization by the Marcus
Introduction
theory,11 the increasing technology concerning detection and
The peroxidation of indoles has been extensively studied since
study of radicals12 have permitted the reinvestigation of the
the beginning of the 1950’s.1,2 Most of the work was performed
peroxidation of indoles from an electron transfer standpoint.
An outer-sphere electron transfer between an electron donor
on 2-alkylindoles3–5 whose dimers of type 1 and 2 were isolated.
(D) and an electron acceptor (A) occurs when the difference
between the oxidation potential of the donor (EDox) and the
reduction potential of the acceptor (EAred) measured versus
the same reference electrode is less than 0.4 V.13 An oxidant is
usually also a good electrophile, just as a donor is a good
nucleophile, so it is not always easy to distinguish which is the
true mechanism involved in a reaction. In fact, both the mech-
anisms could be operating competitively at the same time.
For the reasons described above, we have reinvestigated the
peroxidation of differently substituted indoles with m-chloro-
perbenzoic acid and hydrogen peroxide in order to verify the
involvment of a radical mechanism. In the present study, the
redox potentials of the reagents, the Marcus theory, the reac-
tions carried out in the presence and in the absence of oxygen
and the reaction products distribution were all considered.
The identification of these two compounds was the subject
of discussion and controversy,5 but in the end agreement
was reached.3 From the mechanistic point of view, compounds
Results
1 and 2 were explained by the intermediate formation of
The indoles studied in the present work were 2-alkylindoles
2-alkylindolone 3.3,5 The latter could truly be the intermediate
(RЈ = methyl, 5; RЈ = tert-butyl, 6; RЈ = phenyl, 7), 1,2-dialkyl-
in the formation of 1; in fact 1 could derive from the nucleo-
indoles (R = RЈ = methyl, 8; R = ethyl, RЈ = phenyl, 9) (Scheme
philic attack of an indole molecule on indolone 3.6 However,
1), 2,3-dialkylindoles (R = RЈ = methyl, 14a; R = RЈ = phenyl
the ionic mechanism proposed is highly unlikely to explain the
14b; see Scheme 2) and 1-hydroxy-2-phenylindole 16. All indoles
formation of compound 2,5 which is more likely to arise from
were reacted in the presence of oxygen with m-chloro-
the dimerization of radical 4.
perbenzoic acid and hydrogen peroxide using trichloroacetic
Indoles are generally compounds with rather low oxidation
acid or calcium chloride as catalysts.
potentials7–9 and oxidants such as hydrogen peroxide in acids
and peroxides are strong enough to promote an electron trans-
fer process. When the peroxidation of indoles was first investi-
Reaction with m-chloroperbenzoic acid
gated, the concept of electron transfer processes was still not
well known and many reactions involving radical intermediates
were explained by ionic mechanisms. Starting from the 1960’s,
the development of electrochemistry in organic chemistry,10
the introduction of the concepts of outer- and inner-sphere
The reactions were carried out at room temperature in
dichloromethane. The 2-substituted 5–7 and the 1,2-disubsti-
tuted 8 and 9 indoles essentially afforded 3-(3-oxoindolin-2-
yl)indoles 10 and 2,2Ј-biindolinyls 11. The only exceptions were
with indole 6 where compound 10 was not isolated, whereas the
1634
J. Chem. Soc., Perkin Trans. 2, 2001, 1634–1640
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b102915k

View Article Online
Table
1 Product distribution for the reactions of indoles with
m-chloroperbenzoic acid and hydrogen peroxide under various
conditions
Indole
Oxidant
Isolated products (yields [%])
5
5
ClC6H4CO3H
ClC6H4CO3Ha
H2O2–H3Oϩ
H2O2–CaCl2
ClC6H4CO3H
H2O2–H3Oϩ
H2O2–CaCl2
ClC6H4CO3H
H2O2–H3Oϩ
H2O2–CaCl2
ClC6H4CO3H
H2O2–H3Oϩ
H2O2–CaCl2
ClC6H4CO3H
H2O2–H3Oϩ
H2O2–CaCl2
ClC6H4CO3H
H2O2–H3Oϩ
H2O2–CaCl2
ClC6H4CO3H
H2O2–H3Oϩ
H2O2–CaCl2
ClC6H4CO3H
H2O2–H3Oϩ
H2O2–CaCl2
10a (20); 11a (61)
10a (50)
10a (49); 11a (44)
10a (50)
11b (10)
11b (traces)
11b (traces)
10c (52); 12
10c (35)
5
5
6
6
6
7
7
7
10c (15); 13 (32)
10d (30)
8
8
10d (81)
8
10d (40)
9
10e (45)
9
10e (40)
9
10e (30)
14a
14a
14a
14b
14b
14b
16
16
16
15a (40)
15a (40)
15a (50)
15b (95)
15b (45)
15b (20)
17 (95)
17 (40); 18 (37); 19 (10)
18 (97)
a In the absence of oxygen.
Scheme 1
Reaction with hydrogen peroxide
All indoles (5–9, 14a and 14b) were reacted in acetone at room
temperature in the presence of trichloroacetic acid and in
methanol in the presence of CaCl2 under reflux. In all the
experiments, the products isolated using both H2O2–Cl3-
CCOOH and H2O2–CaCl2 were the same as those obtained
from the reaction with m-chloroperbenzoic acid; the yields
are reported in Table 1. Only with 2-phenylindole 7 was the
diindolylmethane 13 isolated as the main product. Compound
13 was identified by its analytical and spectroscopic data and
by comparison with a sample obtained from an independent
synthesis (see Experimental).
Scheme 2
dimers 11 were obtained only from indoles 5 and 6. In the case
of 2-phenylindole 7, the bis(m-chlorobenzoyl) peroxide 12 was
also isolated: the products obtained are shown in Scheme 1 and
the yields are reported in Table 1.
Reaction of 1-hydroxy-2-phenylindole
1-Hydroxy-2-phenylindole 16 gave rise to phenylisatogen
(2-phenyl-3-oxo-3H-indole 1-oxide) 1715 (Scheme 3) after
The structures of compounds 10 were established by compar-
ing their spectroscopic data with those of indolinone 10a whose
structure was determined by X-ray analysis. The IR spectra of
compounds 10a and 10c–e show two typical absorptions at
ca. 1700 cmϪ1 for the C᎐O group and at ca. 1600 cmϪ1 for the
᎐
group represented by the sp3 C-2, the nitrogen and the indolic
benzene ring as described by Witkop3 and others.14 The only
2,2Ј-dimers isolated, compounds 11a and 11b, were identified
by the X-ray analysis of indolinone 11a. These compounds also
show comparable IR absorptions since both kinds of dimers
1
have the same typical groups. The H NMR spectrum of com-
pound 10e shows two identical multiplets in the region 2.9–3.6
ppm, each multiplet corresponding to one of the hydrogens of
the CH2 group bonded to the aminic nitrogen, and two over-
lapping quartets in the region 3.85–3.95 ppm due to the CH2
group of the indolic nitrogen. This particular pattern is most
probably a consequence of the presence of an asymmetric
center at the C-2. The detection of bis(m-chlorobenzoyl) per-
oxide 12 in the peroxidation of 7 was of great importance for
the mechanistic interpretation; it was identified by its analytical
and spectroscopic data. Compounds 14a and 14b undergo
opening of the pentaatomic ring affording products 15a and
15b, respectively. These two compounds were identified by their
analytical and spectroscopic data.
Scheme 3
oxidation with m-chloroperbenzoic acid in almost quantitative
yield, whilst with hydrogen peroxide in the presence of CaCl2
the bis-nitrone 1816 was isolated. In the reaction with hydrogen
peroxide in the presence of trichloroacetic acid a mixture of
J. Chem. Soc., Perkin Trans. 2, 2001, 1634–1640
1635

View Article Online
Table 2 Selected bond distances (Å) and angles (Њ) for compounds 10a
and 11a
10a
11a
O(1)–C(2)
N(1)–C(1)
N(1)–C(8)
N(2)–C(11)
N(2)–C(18)
C(1)–C(1)Јa
C(1)–C(2)
C(1)–C(12)
C(2)–C(3)
C(3)–C(8)
1.230(3)
1.477(3)
1.372(3)
1.395(3)
1.381(3)
1.222(2)
1.466(2)
1.374(2)
1.559(3)
1.548(2)
1.557(3)
1.506(3)
1.440(3)
1.396(3)
1.378(3)
1.446(3)
1.404(3)
Fig. 2 A SCHAKAL perspective view of compound 11a. A prime
denotes a transformation of 1 Ϫ x, Ϫy, Ϫz.
1.455(2)
1.402(2)
from planarity is also observed for the oxoindole systems in
11a [maximum deviation: 0.008(2) Å for C(14)], which are
strictly parallel to each other for symmetry requirements. The
orientations assumed by the indole/oxoindole systems are
consistent with the presence of weak attractive intramolecular
C–H ؒ ؒ ؒ N interactions (Table 3).
Molecular packing in both compounds is mainly determined
by N–H ؒ ؒ ؒ O and N–H ؒ ؒ ؒ N intermolecular attractive inter-
actions (Table 3). Other contacts are consistent with van der
Waals interactions.
C(11)–C(12)
C(12)–C(13)
C(13)–C(14)
C(1)–N(1)–C(8)
C(11)–N(2)–C(18)
N(1)–C(1)–C(2)
C(1)–C(2)–C(3)
C(2)–C(3)–C(8)
N(1)–C(8)–C(3)
N(2)–C(11)–C(12)
C(11)–C(12)–C(13)
C(12)–C(13)–C(18)
N(2)–C(18)–C(13)
110.9(2)
109.3(2)
101.7(2)
107.5(2)
108.0(2)
111.7(2)
108.8(2)
106.8(2)
107.4(2)
107.7(2)
110.3(1)
103.1(1)
106.9(1)
107.4(1)
112.2(2)
Discussion
a A prime denotes a transformation of 1 Ϫ x, Ϫy, Ϫz.
As stated in the introduction, no clear indication is available in
the literature on the mechanism (radical or ionic) of indole
peroxidation. Only recently, an electron transfer process involv-
ing the formation of indole radical cations was admitted for
the oxidation of indoles with copper chloride19 and thallium
acetate.20 On the basis of the experimental results here
described and from those obtained in a previous electro-
chemical study on the oxidation of indoles,21 this mechanism
could be considered the one implicated in the reaction with
peracids and activated hydrogen peroxide. In fact, there is
more than one piece of evidence to support this hypothesis. The
studied indoles, except 1-hydroxy-2-phenylindole 16, have
oxidation potentials in the range 0.57–0.75 V vs. Ag/Agϩ in
CH3CN (see Table 4), whereas m-chloroperbenzoic acid and
H2O2 have reduction potentials more positive than those of all
the studied indoles (Table 4).
On the basis of the Marcus theory, the electron transfer
process is fully justified. In fact, the difference between the
oxidation potentials of indoles 5–9 and the reduction potentials
of m-chloroperbenzoic acid and hydrogen peroxide is, in all
cases, less than 0.4 V, the upper limit for the feasibility of an
electron transfer process. Considering that in an electron trans-
fer process the entropy changes are neglegible, from the redox
potentials of the reagents involved it could be deduced that all
the reactions show a negative ∆G value and hence they are
thermally favoured. Some of the reactions have been performed
in the EPR cavity, but no signals were detected and this is in
agreement with the low stability of the indole radical cation;21
experiments based on the spin trapping technique failed.
However, the reaction product distribution supports the radical
mechanism. The isolation of bis(3-chlorobenzoyl) peroxide 12
can only be explained by admitting the coupling of two
m-chlorobenzoyloxy radicals which arise from an electron
transfer process in the first step of the reaction according to
Scheme 4.
The indole radical cation 21 can give rise to a deprotonating
equilibrium with the corresponding indolyl radical 23, which
readily reacts with atmospheric oxygen to give peroxy radical.
This, through successive reactions affords the alkoxy radical 24
(see below). Indolyl radical 23 may couple with the benzoyloxy
22 leading to the formation of 25. According to Scheme 4, only
the 3-(3-oxoindolin-2-yl)indole can be explained by either the
alkoxy radical 24 or by the benzoyloxy derivative 25, and this
latter path has been previously demonstrated.21 On the other
Fig. 1 A SCHAKAL perspective view of compound 10a.
phenylisatogen 17, bis-nitrone 18 and 3-(3-oxoindolin-2-yl)-
indole 19 was obtained. Compounds 17,15 1816 and 1917 were
identified by comparison with authentic samples. Compound
19 was also oxidized with lead dioxide and the EPR spectrum
of the new product was compared with that of the compound
obtained from the reaction between 1-hydroxy-2-phenylindole
and 2-phenyl-3H-indol-3-one.17
Molecular geometry of compounds 10a and 11a
Selected bond distances and angles are quoted in Table 2. A
perspective view of 10a and 11a is given in Figs. 1 and 2,
respectively.
Compound 11a possesses a crystallographically imposed Ci
symmetry, the inversion centre lying midway along the C(1)–
C(1)Ј bond (Ј = 1 Ϫ x, Ϫy, Ϫz). In both compounds bond dis-
tances and angles are in line with the hybridization expected for
the atoms involved and in agreement with those of analogous
compounds reported in the literature.18 In 10a the planar indole
system [maximum deviation: 0.008(2) Å for C(14)] forms a
dihedral angle of 91.5(1)Њ with the oxoindole system which
shows significant distortion from planarity [maximum devi-
ation: 0.060(3) Å for C(3)]. A small but significant deviation
1636
J. Chem. Soc., Perkin Trans. 2, 2001, 1634–1640

View Article Online
Scheme 4
Table 3 Relevant hydrogen bonds in compounds 10a and 11a
D–H/Å
H ؒ ؒ ؒ A/Å
D ؒ ؒ ؒ A/Å
D–H ؒ ؒ ؒ A/Њ
Compound 10a
N(1)–H(1) ؒ ؒ ؒ N(2)a
N(2)–H(2) ؒ ؒ ؒ O(1)b
C(14)–H(14) ؒ ؒ ؒ N(1)
0.98
0.99
1.07
2.51
1.96
2.89
3.362(3)
2.925(2)
3.347(3)
146
164
106
Compound 11a
N(1)–H(1) ؒ ؒ ؒ O(1)c
N(1)–H(1) ؒ ؒ ؒ O(1)d
C(9)–H(92) ؒ ؒ ؒ O(1)
C(9)–H(92) ؒ ؒ ؒ N(1)d
0.91
0.91
1.00
1.00
2.69
2.33
2.78
2.66
3.297(2)
2.970(3)
3.011(2)
2.967(3)
125
127
94
98
a Ϫ1 Ϫ x, Ϫy, Ϫz. b Ϫ1 Ϫ x, Ϫy, Ϫz. c 1 ϩ x, y, z. d 1 Ϫ x, Ϫy, Ϫz.
Table 4 Redox potentials of the reagents
in Scheme 4, since the oxidants used have a similar oxidation
power. The diindolylmethane 13, isolated in the reaction of
2-phenylindole with hydrogen peroxide in the presence of
calcium chloride, may be explained by the reaction between
the indole and the formaldehyde derived from the oxidation
of methanol, the reaction solvent. In fact, the same product
is obtained in the reaction between 2-phenylindole and
formaldehyde.
2,3-Substituted indoles 14a and 14b in the first step of the
reaction may react similarly to other indoles on the basis of
their oxidation potentials. The indole radical cation 27 formed
can couple with the benzoyloxy radical 22 leading to the inter-
mediate 28 which evolves to product 15 according to Scheme 5.
Otherwise, it may interact with oxygen yielding alkoxy radical
31, which undergoes β-scission to give radical 32. Alkoxy 33 is
subsequently formed which then leads to the final product.
Indeed, with these kinds of indoles the presence or the
absence of oxygen may only modify the path taken but not the
final reaction product and therefore it is difficult to establish
which is the true pathway. The mechanism described in Scheme
5 could also be extended to the reactions with hydrogen per-
oxide. The interaction of indolyl radical with oxygen described
in Schemes 4 and 5 has been well demonstrated recently;22 the
formation of alkoxy radicals arising from peroxy radicals
formed in the interaction of indolyl radicals with oxygen is also
a well documented reaction, previously discussed.23
Compound
E
ox vs. Ag/Agϩ in CH3CN
Reference
½
5
6
0.60
0.57
0.73
0.60
0.72
0.48
0.66
0.4
27
9
28
7
8
9
14a
14b
16
27
a
a
a
29
E
red vs. SCE in H2O
½
m-ClC6H4CO3H
H2O2
0.93
1.33
1.77
30
31
31
H2O2–H3Oϩ
a This paper.
hand, the formation of the 2,2Ј-dimer can only be justified by
the dimerization of radical 24 and not through an intermediate
such as 3 as predicted by the mechanism previously proposed.3
Significant results in favour of this hypothesis were obtained
in the reaction of 2-methylindole 5 with m-chloroperbenzoic
acid in the presence and in the absence of oxygen; in fact
in the experiment carried out in the absence of oxygen only
the compound 10a was isolated. The chemical behaviour of
2-substituted 5–7 and 1,2-disubstituted 8 and 9 indoles with
m-chloroperbenzoic acid and hydrogen peroxide is that shown
1-Hydroxy-2-phenylindole 16 shows an oxidation potential
which is ca. 300 mV less positive than those of other indoles.
Thus, in this case, electron transfer certainly occurs in the reac-
tions (compound 16 undergoes autoxidation in solution24).
J. Chem. Soc., Perkin Trans. 2, 2001, 1634–1640
1637

View Article Online
Although radical 34 (Scheme 6) is an intermediate in all the
studied oxidative conditions, the products formed are different
in every case.
favoured; in fact, only bis-nitrone 18 was isolated. In the reac-
tion performed in trichloroacetic acid under atmospheric air,
compound 19 was isolated together with phenylisatogen 17
and bis-nitrone 18. In this case, the radical intermediate
36 could be formed by interaction of indolyl radical 35
with oxygen. This hypothesis is supported by the fact that
2,3-disubstituted-1-hydroxyindoles afford 2,3-disubstituted-3-
hydroxyindole N-oxides by simple exposure of their solutions
to air.25 The formation of 3-(3-oxoindolin-2-yl)indole 19 could
be reasonably explained by nucleophilic attack of 1-hydroxy-2-
phenylindole 16 on phenylisatogen 17 under acid catalysis
(Scheme 6, path b);25 however, a mixture of these two com-
pounds in the same reaction conditions did not lead to the
formation of 19. It is noteworthy that the interaction of 35 with
39 affords the intermediate 37, which is further converted into
19 (Scheme 6, path a) by dehydroxylation, in agreement with
the common behaviour of aromatic hydroxylamines in acidic
medium in the presence of electron donors.26
In particular, with m-chloroperbenzoic acid only phenyl-
isatogen 17 was obtained, whereas with hydrogen peroxide in
the presence of calcium chloride only the bis-nitrone 18 was
isolated. It is our opinion that their different behaviour could
likely be due either to the concentration of the radical inter-
mediates formed or to their tendency to couple. In other words,
if indolyl radical 35 reacts quickly with the aroyloxy radical 22,
the intermediate formed could give phenylisatogen 17, through
a mechanism similar to the one shown in Scheme 4 (see the
transformation of 25 into 26). In the reaction with hydrogen
peroxide in the presence of calcium chloride, which was carried
out in methanol under reflux, the dimerization of 35 could be
Experimental
Melting points are uncorrected and were measured with an
Electrothermal apparatus. IR spectra were recorded in the solid
state on a Nicolet Fourier Transform Infrared 20-SX Spectro-
1
photometer equipped with a Spectra Tech. H NMR spectra
were recorded at room temperature in CDCl3 solution on a
Varian Gemini 200 spectrometer (TMS was taken as reference
peak). Mass spectra were performed on a Carlo Erba QMD
1000 mass spectrometer. EPR spectra were run on a Varian E4
instrument.
2-tert-Butylindole 6,9 N-ethyl-2-phenylindole 932 and
N-hydroxy-2-phenylindole 1633 were synthesised according
to the literature. Indoles 1–5, 7, 8, 14a and 14b, m-chloro-
perbenzoic acid, hydrogen peroxide, calcium chloride, tri-
chloroacetic acid were Aldrich products. All solvents were
Carlo Erba or Aldrich RP-ACS grade.
Reaction of indoles with m-chloroperbenzoic acid. General
procedure
A solution of m-chloroperbenzoic acid (1.2 mmol) in dichloro-
methane was added dropwise at room temperature to a
stirred solution of indole (1 mmol) in the same solvent. The
mixture was stirred for 1 h, then poured into 5% NH4Cl and
extracted with CH2Cl2. The organic layer was dried with
Na2SO4 and evaporated to dryness; the residue was chromato-
graphed on a silica gel column (eluant cyclohexane–ethyl
acetate 8 : 2).
Scheme 5
Scheme 6
1638
J. Chem. Soc., Perkin Trans. 2, 2001, 1634–1640

View Article Online
Reaction of 2-methylindole 5 with m-chloroperbenzoic acid in the
absence of oxygen
J = 6.9 Hz), 2.95–3.12 (m, ½ CH2), 3.35–3.55 (m, ½ CH2),
3.88 (q, 2H, J = 6.9 Hz), 3.89 (q, 2H, J = 6.9 Hz), 6.45–6.60
(m, 2H), 6.82–7.00 (m, 4H), 7.10–7.55 (m, 12H) ppm; FT-IR:
ν/cmϪ1 = 2976, 1700 and 1615; MS (EIϩ): m/z = 456 (45%), 427
(70); 351 (57), 221 (30%); 149 (77), 105 (60), 84 (100).
A sample of indole 5 in 5 mL of CH2Cl2 and m-chloro-
perbenzoic acid in 5 mL of solvent were separately introduced
into the two legs of an inverted Y flask, saturated with argon
and then mixed at room temperature. The reaction mixture was
chromatographed on a small preparative silica gel column
(cyclohexane–ethyl acetate 8 : 2), the main product was isol-
ated and its mass spectrum was compared with that of the
3-(3-oxoindolin-2-yl)indole 10a.
2-Methyl-2-(2-methyl-3-oxo-2,3-dihydro-1H-indol-2-yl)-1,2-
dihydro-3H-indol-3-one (11a). Mp = 168–172 ЊC; 1H NMR (200
MHz, CDCl3, 25 ЊC): δ = 1.14 (s, 6H), 6.06 (br s, 2H), 6.8 (t, 2H,
J = 7.5 Hz), 6.94 (d, 2H, J = 8.2 Hz), 7.49 (t, 2H, J = 7.3 Hz),
7.61 (d, 2H, J = 7.7 Hz) ppm; FT-IR: ν/cmϪ1 = 3300, 1650 and
1600; MS (EIϩ): m/z = 146 (60%), 117 (95); 77 (80).
Reaction of indoles with hydrogen peroxide and trichloroacetic
acid. General procedure
2-tert-Butyl-2-(2-tert-butyl-3-oxo-2,3-dihydro-1H-indol-2-yl)-
1,2-dihydro-3H-indol-3-one (11b). Mp = 88–90 ЊC; 1H NMR
(200 MHz, CDCl3, 25 ЊC): δ = 1.06 (s, 18H), 5.00 (br s, 2H), 6.82
(m, 4H), 7.44 (dd, 2H, J = 7.1 Hz and J = 1.3 Hz), 7.55 (dd, 2H,
J = 7.7 Hz and J = 1.3 Hz) ppm; FT-IR: ν/cmϪ1 = 3370 and
1650; MS (EIϩ): m/z = 187 (93%), 120 (89); 57 (71).
A solution of hydrogen peroxide (1 mmol) and tricloroacetic
acid (1.2 mmol) in acetone was added dropwise to a stirred
solution of indole (1 mmol) in acetone. The mixture was stirred
for 2 h, poured into 5% NH4Cl and extracted with CH2Cl2. The
organic layer was dried with Na2SO4 and evaporated to dryness;
the residue was chromatographed on a silica gel column (eluant
cyclohexane–ethyl acetate 8 : 2).
3-Chlorobenzoyl 3-chlorobenzeneperoxycarboxylate (12). 1H
NMR (200 MHz, CDCl3, 25 ЊC): δ = 7.25–7.32 (m, 3H), 7.33–
7.48 (m, 2H), 7.48–7.64 (m, 3H) ppm; FT-IR: ν/cmϪ1 = 2924,
1736; MS (EIϩ): m/z = 312 (5%), 311 (35), 310 (100), 165 (40),
77 (60).
Reaction of indoles with hydrogen peroxide and calcium chloride.
General procedure
Indole (1 mmol) was dissolved in methanol (5 mL) and calcium
chloride (10 mmol) was added. The stirred solution was heated
at ca. 70 ЊC under reflux. At this temperature hydrogen peroxide
(2 mmol) was added dropwise and the mixture was stirred,
under reflux, for 4 h. The reaction mixture was then poured into
5% NH4Cl and extracted with CH2Cl2. The organic layer was
dried with Na2SO4 and evaporated to dryness; the residue was
chromatographed on a silica gel column (eluant cyclohexane–
ethyl acetate 8 : 2).
2-Phenyl-3-[(2-phenyl-1H-indol-3-yl)methyl]-1H-indole (13).
1H NMR (200 MHz, CDCl3, 25 ЊC): δ = 4.57 (s, 2H), 6.82–6.9
(m, 2H), 7.05–7.12 (m, 2H), 7.20–7.45 (m, 10H), 7.56–7.6 (m,
4H), 8.02 (br s, 1H) ppm; FT-IR: ν/cmϪ1 = 3051 and 1604; MS
(EIϩ): m/z = 398 (100%), 321 (30); 204 (55), 193 (30).
N-(2-Acetylphenyl)acetamide (15a). Mp = 70–75 ЊC; 1H
NMR (200 MHz, CDCl3, 25 ЊC): δ = 2.02 (s, 3H), 2.67 (s, 3H),
7.12 (t, 1H, J = 7.1 Hz), 7.55 (t, 3H, J = 7.1 Hz), 7.88 (d, 1H,
J = 8.4 Hz), 8.72 (d, 1 H, J = 8.4 Hz), 11.69 (br s, 1H) ppm;
FT-IR: ν/cmϪ1 = 3225, 1674, and 1640; MS (EIϩ): m/z = 177
(55%), 134 (100); 120 (95), 92 (70).
Synthesis of bis(2-phenylindol-3-yl)methane 13
Formaldehyde vapour from thermal decomposition of para-
formaldehyde (800 mg) was bubbled through a solution of
2-phenylindole 7 (2 mmol) and toluene-p-sulfonic acid (traces)
in ethanol (50 mL) cooled at ca. 7–8 ЊC. The mixture was stirred
and after 2 h evaporated to dryness. The residue was chromato-
graphed on a silca gel column (eluant cyclohexane–ethyl acetate
8 : 2).
N-(2-Benzoylphenyl)benzamide (15b). Mp = 78–82 ЊC; 1H
NMR (200 MHz, CDCl3, 25 ЊC): δ = 7.08–7.18 (m, 2H), 7.50–
7.71 (m, 12H), 8.07–8.12 (m, 3H), 8.89 (d, 1 H, J = 7.7 Hz), 11.9
(br s, 1H) ppm; FT-IR: ν/cmϪ1 = 3330, 1677, and 1634; MS
(EIϩ): m/z = 400 (20%), 371 (95), 323 (25), 295 (60), 193 (100).
2-Methyl-2-(2-methyl-1H-indol-3-yl)-1,2-dihydro-3H-indol-3-
one (10a). Mp = 195–197 ЊC; 1H NMR (200 MHz, CDCl3,
25 ЊC): δ = 1.92 (s, 3H), 2.42 (s, 3H), 5.02 (br s, 1H), 6.86–6.92
(m, 2H), 6.98–7.12 (m, 2H), 7.22–7.28 (m, 1H), 7.38–7.40 (m,
1H), 7.48–7.6 (m, 1H), 7.68–7.75 (m, 1H), 7.88 (br s, 1H) ppm;
FT-IR: ν/cmϪ1 = 3200, 1650 and 1600; MS (EIϩ): m/z = 276
(55%), 261 (100); 247 (50), 233 (70), 157 (45).
Experimental data for the X-ray diffraction studies on crystalline
compounds 10a and 11a34
Data were collected on a Siemens AED single-crystal diffract-
ometer with graphite-monochromatized Cu-Kα radiation using
θ/2θ scan mode. Unit cell parameters were determined by
automatic centering of 24 strong reflections (25.0 < θ < 39.5Њ
and 21.5 < θ < 35.0Њ for 10a and 11a, respectively) and refined
by the least-squares method. The details of the X-ray data
collection, structure solution and refinement are given in the
supplementary material.34 Three reflections were measured
every 100 reflections collected as intensity and the orientation
controlled. For both compounds no significant intensity decay
was observed. Lorentz polarization but not absorption correc-
tion was applied. The crystal quality was tested by ψ scans
showing that crystal absorption effects could be neglected. The
structures were solved by direct methods using the SHELXS35
computer program and refined with the SHELX9336 computer
program. Refinements were done by full matrix least-squares
first isotropically and then anisotropically for all non-H atoms.
The function minimized was Σw(∆F 2)2. Anomalous scattering
corrections were included in all structure factor calculations.37b
Scattering factors for neutral atoms were taken from ref. 37(a)
for non-hydrogen atoms and from ref. 38 for H. The hydrogen
atoms were located from difference Fourier maps and intro-
2-Phenyl-2-(2-phenyl-1H-indol-3-yl)-1,2-dihydro-3H-indol-3-
one (10c). Mp = 227–229 ЊC; 1H NMR (200 MHz, CDCl3,
25 ЊC): δ = 6.68–6.75 (m, 2H), 6.89–7.03 (m, 2H), 7.14–7.24 (m,
6H), 7.25–7.35 (m, 4H), 7.39–7.56 (m, 4H), 8.08 (br s, 1H) ppm;
FT-IR: ν/cmϪ1 = 1670 and 1610; MS (EIϩ): m/z = 400 (20%),
371 (95), 323 (25), 295 (60), 193 (100).
2-(1,2-Dimethyl-1H-indol-3-yl)-1,2-dimethyl-1,2-dihydro-3H-
indol-3-one (10d). Mp = 155–157 ЊC; 1H NMR (200 MHz,
CDCl3, 25 ЊC): δ = 1.90 (s, 3H), 2.28 (s, 3H), 2.86 (s, 3H), 3.64
(s, 3H), 6.71–6.80 (m, 2H), 6.93–7.01 (m, 1H), 7.08–7.16
(m, 1H), 7.42–7.58 (m, 3H), 7.66–7.7 (m, 1H) ppm; FT-IR:
ν/cmϪ1 = 1670 and 1595; MS (EIϩ): m/z = 304 (10%), 289 (5);
269 (40); 97 (40%); 69 (100).
1-Ethyl-2-(1-ethyl-2-phenyl-1H-indol-3-yl)-2-phenyl-1,2-
dihydro-3H-indol-3-one (10e). Mp = 129–131 ЊC; 1H NMR (200
MHz, CDCl3, 25 ЊC): δ = 0.55 (t, 3H, J = 7.2 Hz), 1.15 (t, 3H,
J. Chem. Soc., Perkin Trans. 2, 2001, 1634–1640
1639

View Article Online
10 H. Lund in Organic Electrochemistry, 2nd edn., eds. H. Lund and
M. M. Braizer, Marcel Dekker, New York, 1983, p. 829.
11 (a) H. Taube, Angew. Chem., Int. Ed. Engl., 1984, 23, 329; (b) R. H.
Marcus, Angew. Chem., Int. Ed. Engl., 1993, 32, 1111.
12 (a) A. R. Forrester, J. M. Hay and R. H. Thomson, Organic
Chemistry of Stable Free Radicals, Academic Press, New York and
London, 1968; (b) A. Carrington and A. D. McLachlan,
Introduction to Magnetic Resonance, Harper International Edition,
Harper & Row, New York, 1969.
duced in the refinements as fixed atom contributors (Uiso = 0.08
2
2
2
Å ). For both compounds the weighting scheme w = 1/[σ (Fo ) ϩ
(aP)2] (with P = (|Fo|2 ϩ 2|Fc|2)/3 was applied in the last stage of
refinement, with a resulting in the value of 0.0108 and 0.0687
for 10a and 11a, respectively. All calculations were carried
out on a Fujitsu Personal Computer equipped with an Intel
Pentium II processor.
13 L. Eberson, Electron Transfer Reaction in Organic Chemistry,
Springer, Berlin, 1987.
14 C Berti, M. Colonna, L. Greci and L. Marchetti, Tetrahedron, 1975,
31, 1745.
15 (a) D. B. Adams, M. Hooper, C. J. Swain, E. S. Raper and
B. Stoddart, J. Chem. Soc., Perkin Trans. 1, 1986, 1005; (b) C. C.
Bond and M. Hooper, Synthesis, 1974, 443.
16 P. Bruni, E. Giorgini, G. Tommasi and L. Greci, Tetrahedron, 1998,
54, 5305.
Crystal structure of 2-methyl-2-(2-methyl-1H-indol-3-yl)-1,2-
dihydro-3H-indol-3-one 10a. C18H16N2O, M = 276.3, triclinic,
¯
space group P1, a = 10.568(3), b = 7.100(20), c = 9.755(3) Å,
α = 93.28(2), β = 106.67(4), γ = 95.05(3)Њ, V = 695.9(4) Å3,
Z = 2, Dcalcd = 1.319 g cmϪ3, F(000) = 292, λ(Cu-Kα) = 1.54178
Å, µ(Cu-Kα) = 6.18 cmϪ1, crystal dimensions 0.07 × 0.12 ×
0.29 mm. For 1549 unique observed reflections [I > 2σ(I )] the
final R is 0.036 (wR2 = 0.068 for the 2410 unique reflections
having I > 0 used in the refinement).
17 M Colonna, L. Greci and L. Marchetti, Gazz. Chim. Ital., 1975, 105,
985.
18 R. Benassi, F. Taddei, L. Marchetti, L. Greci, G. D. Andreetti,
G. Bocelli and P. Sgarabotto, J. Chem. Soc., Perkin Trans. 2, 1980,
786; A. Alberti, L. Greci, P. Stipa, P. Sgarabotto and F. Ugozzoli,
Tetrahedron, 1987, 43, 3031; P. Carloni, L. Greci, P. Stipa,
A. Alberti, C. Rizzoli, P. Sgarabotto and F. Ugozzoli, J. Chem.
Soc., Perkin Trans. 2, 1990, 185; A. L. Maniero, M. Brustolon,
F. Ottaviani, C. Rizzoli, P. Sgarabotto, F. Ugozzoli, P. Carloni
and L. Greci, Mol. Phys, 1991, 73, 1; P. Carloni, L. Greci, M.
Iacussi, M. Rossetti, P. Stipa, C. Rizzoli and P. S. Sgarabotto,
J. Chem. Res., 1996, 348, 350.
Crystal structure of 2-methyl-2-(2-methyl-3-oxo-2,3-dihydro-
1H-indol-2-yl)-1,2-dihydro-3H-indol-3-one 11a. C18H16N2O2,
M = 293.3, monoclinic, space group P21/n, a = 6.155(3),
b = 20.219(5), c = 6.556(2) Å, β = 116.08(2)Њ, V = 723.8(5) Å3,
Z = 2, Dcalcd = 1.325 g cmϪ3, F(000) = 308, λ(Cu-Kα) = 1.54178
Å, µ(Cu-Kα) = 6.67 cmϪ1, crystal dimensions 0.34 × 0.40 ×
0.56 mm. For 1255 unique observed reflections [I > 2σ(I )] the
final R is 0.042 (wR2 = 0.137 for the 1386 unique reflections
having I > 0 used in the refinement).
19 E. Balogh-Hergovic and G. Speier, J. Chem. Soc., Perkin Trans. 1,
1986, 2305.
20 J. Tholander and J. Bergman, Tetrahedron, 1999, 55, 12595.
21 P. Carloni, L. Eberson, L. Greci, P. Stipa and G. Tosi, J. Chem. Soc.,
Perkin Trans. 2, 1991, 1779.
22 L. Greci, G. Tommasi, P. Astolfi, R. Petrucci, G. Marrosu,
A. Trazza, P. Sgarabotto and L. Righi, J. Chem. Soc., Perkin Trans.
2, 2000, 1749.
23 (a) C. Berti, L. Greci, M. Poloni, G. D. Andreetti, G. Bocelli and
P. Sgarabotto, J. Chem. Soc., Perkin Trans. 2, 1980, 339; (b) M.
Nakagawa, H. Yamaguchi and T. Hino, Tetrahedron Lett., 1970, 47,
4035.
Electrochemical measurements
A three-electrode multipolarograph AMEL 472 coupled with a
digital x/y recorder AMEL 863 was employed for the voltam-
metric measurements, carried out at a pulsed (polarographic
measurements) or static (cyclic voltammetries) glassy-carbon
electrode in anhydrous MeCN containing tetraethylammonium
perchlorate (TEAP) 0.1 mol LϪ1 as supporting electrolyte. Ag/
AgClO4 0.1mol LϪ1–MeCN/sintered glass disk/TEAP 0.1 mol
L
Ϫ1–MeCN/sintered glass disk was used as reference,39 and a
24 L. Marchetti and V. Passalacqua, Ann. Chim. (Rome), 1967, 57,
1275.
platinum wire as counter electrode. These measurements were
25 S. P. Hiremath and M. Hooper, “Isatogen and Indolones” in
Advances in Heterocyclic Chemistry, ed. A. R. Katritzky and A. J.
Boulton, Academic Press, New York, 1978, p. 123.
26 T. P. Simonova, V. D. Nefedov, M. A. Toropova and N. F. Kirillov,
Russ. Chem. Rev., 1992, 61, 584.
carried out starting from 1 × 10Ϫ3 mol LϪ1 solutions.
Acknowledgements
27 A. Alberti, P. Carloni, L. Eberson, L. Greci and P. Stipa, J. Chem.
Soc., Perkin Trans 2, 1997, 887.
28 C. Berti, L. Greci, R. Andruzzi and T. Trazza, J. Org. Chem., 1982,
47, 4895.
29 R. Andruzzi, A. Trazza, C. Berti and L. Greci, J. Chem. Res. (S),
1983, 256.
30 M. Venturini and F. Secco, J. Chem. Soc., Perkin Trans. 2, 1973, 491.
31 W. M. Latimer, The Oxidation States of the Elements and
their Potentials in Aqueous Solutions, Prentice-Hall, Englewood,
New York, 1952.
This paper is dedicated to the memory of Professor Lennart
Eberson.
L. G. thanks the Italian Ministero dell’Università e della
Ricerca Scientifica e Tecnologica (MURST) and the University
of Ancona for financial support. The authors also wish to
express their gratitude to Mrs Carla Conti for recording FT-IR
spectra.
32 C. Berti, L. Greci, L. Marchetti, R. Andruzzi and A. Trazza,
J. Chem. Res. (S), 1981, 340; C. Berti, L. Greci, L. Marchetti,
R. Andruzzi and A. Trazza, J. Chem. Res. (M), 1981, 3944.
33 E. Fisher, Chem. Ber., 1896, 29, 2062.
References
1 R. J. Sundberg, The Chemistry of Indoles, Academic Press, New
York and London, 1970, pp. 282–295.
2 W. J. Houlihan, Indoles, Wiley Interscience, New York, 1972, Vol. 1,
pp. 145–160.
3 B. Witkop and J. B. Patrick, J. Am. Chem. Soc., 1951, 73, 713.
4 A. Hassner and M. J. Haddadin, J. Org. Chem., 1963, 28, 224.
5 F. Piozzi and M. R. Langella, Gazz. Chim. Ital., 1963, 93, 1373.
6 M. Colonna, L. Greci and L. Marchetti, Gazz. Chim. Ital., 1975,
105, 985.
7 M. Colonna, L. Greci, M. Poloni, G. Marrosu, A. Trazza, F. P.
Colonna and G. Distefano, J. Chem. Soc., Perkin Trans. 2, 1986,
1229.
34 Crystallographic data (excluding structure factors) for the structures
reported in this paper have been deposited with the Cambridge
Crystallographic Data Centre. CCDC reference numbers 163769
and 163770. See http://www.rsc.org/suppdata/p2/b1/b102915k/ for
crystallographic files in .cif or other electronic format.
35 G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467.
36 G. M. Sheldrick, SHELX93, Program for crystal structure
refinement, University of Göttingen, Germany, 1994.
37 (a) International Tables for X-ray Crystallography, Kynoch Press,
Birmingham, England, 1974, Vol IV, p. 99; (b) International Tables
for X-ray Crystallography, Kynoch Press, Birmingham, England,
1974, Vol IV, p. 149.
8 R. J. Waltman, A. F. Diaz and J. Bargon, J. Phys. Chem., 1984, 88,
4343.
9 P. Carloni, L. Greci, M. Iacussi, M. Rossetti, P. Cozzini
and P. Sgarabotto, J. Chem. Res. (S), 1998, 232; P. Carloni, L. Greci,
M. Iacussi, M. Rossetti, P. Cozzini and P. Sgarabotto, J. Chem. Res.
(M), 1998, 1121.
38 R. F. Stewart, E. R. Davidson and W. T. Simpson, J. Chem. Phys.,
1965, 42, 3175.
39 R. Andruzzi, A. Trazza, L. Greci and A. Marchetti, Ann. Chim.
(Rome), 1979, 69, 583.
1640
J. Chem. Soc., Perkin Trans. 2, 2001, 1634–1640