Multiple transformations of 2-alkynyl-1,8-bis(dimethylamino)naphthalenes into Benzo[ g ]indoles. Pd/Cu-dependent switching of the electrophilic and nucleophilic sites in acetylenic bond and a puzzle of porcelain catalysis

Article abstract of DOI:10.1021/jo502363t

By means of Sonogashira reaction, a series of 2-alkynyl- and 2,7-dialkynyl derivatives of 1,8-bis(dimethylamino)naphthalene ( proton sponge ) have been obtained from the corresponding iodides. It was disclosed that changing the reaction conditions and isolation protocol or conducting the model experiments with the authentic acetylenes results in several types of palladium- and copper-assisted heterocyclizations with the participation of the C≡C bond and 1-NMe₂ group. These include: (i) a cyclization into isomeric 1H-benzo[g]indoles with [1,3] migration of the N-methyl group into the newly formed pyrrole ring; (ii) a similar cyclization with a loss of the methyl group; (iii) a tandem process of cyclization into benzo[g]indoles and their subsequent 3,3′-dimerization; and (iv) a copper-catalyzed oxidative transformation into 3-aroylbenzo[g]indoles. In most cases, the reactions occur in parallel, but under certain conditions, one of the above products becomes predominant or even the only one. Remarkably, in Pd-catalyzed cyclizations i-iii, the acetylenic bond behaves as an electrophile being attacked at the β-position by the amine nitrogen atom. In contrast, in transformation iv, the C≡C bond attacks by its C_α atom on the aminomethyl radical functionality N(Me)-CH₂· presumably arising at copper oxidation/deprotonation of the 1-NMe₂ group. Studying rearrangement i, some evidence for the porcelain catalysis was obtained.

Full text of DOI:10.1021/jo502363t

Article
pubs.acs.org/joc
Multiple Transformations of 2‑Alkynyl-1,8-
bis(dimethylamino)naphthalenes into Benzo[g]indoles. Pd/Cu-
Dependent Switching of the Electrophilic and Nucleophilic Sites in
Acetylenic Bond and a Puzzle of Porcelain Catalysis
Ekaterina A. Filatova, Alexander F. Pozharskii,* Anna V. Gulevskaya, and Valery A. Ozeryanskii
Department of Organic Chemistry, Southern Federal University, Zorge str. 7, 344090 Rostov-on-Don, Russian Federation
S
* Supporting Information
ABSTRACT: By means of Sonogashira reaction, a series of 2-
alkynyl- and 2,7-dialkynyl derivatives of 1,8-bis(dimethylamino)-
naphthalene (“proton sponge”) have been obtained from the
corresponding iodides. It was disclosed that changing the
reaction conditions and isolation protocol or conducting the
model experiments with the authentic acetylenes results in
several types of palladium- and copper-assisted heterocycliza-
tions with the participation of the CC bond and 1-NMe2
group. These include: (i) a cyclization into isomeric 1H-
benzo[g]indoles with [1,3] migration of the N-methyl group
into the newly formed pyrrole ring; (ii) a similar cyclization with
a loss of the methyl group; (iii) a tandem process of cyclization
into benzo[g]indoles and their subsequent 3,3′-dimerization;
and (iv) a copper-catalyzed oxidative transformation into 3-
aroylbenzo[g]indoles. In most cases, the reactions occur in parallel, but under certain conditions, one of the above products
becomes predominant or even the only one. Remarkably, in Pd-catalyzed cyclizations i−iii, the acetylenic bond behaves as an
electrophile being attacked at the β-position by the amine nitrogen atom. In contrast, in transformation iv, the CC bond
attacks by its Cα atom on the aminomethyl radical functionality N(Me)−CH2· presumably arising at copper oxidation/
deprotonation of the 1-NMe2 group. Studying rearrangement i, some evidence for the porcelain catalysis was obtained.
preliminary formation of a contact ion pair was registered.3g
Recently, several examples of migration of the methyl group in
INTRODUCTION
■
It is well-known that ortho-aminoarylacetylenes with primary or
such processes have been reported for 2-alkynyl-N,N-
dimethylanilines (Scheme 1b).4 The reaction proceeded
under rather drastic conditions (160 °C) with the assistance
of the gold-carbene catalyst. The mechanism of this conversion
was not discussed in the original paper.
Over the past few years, we were studying the closely related
cyclizations of 2-alkynyl- and 2,7-dialkynyl derivatives of 1,8-
bis(dimethylamino)naphthalene (“proton sponge”). Our re-
sults, mentioned briefly in the conference theses,5 have brought
quite a few new findings into this important field that seems to
originate from the specific structure and reactivity of the
“proton sponge”. In particular, a number of alternative
cyclization modes for such compounds were disclosed and
the relative importance of different catalytic and oxidative
additives was clarified. Now, we report on these findings in
more detail.
secondary amino groups are quite easily cyclized into the
corresponding indole derivatives or their numerous condensed
analogues.1 The reaction is generally accelerated under
employment of basic catalysts or transition metal complexes.
Remarkably, even arylacetylenes having tertiary ortho-amino
groups under special conditions are able to cyclize with
removing one of the N-substituents.1d,2 A typical example is the
iodine activated conversion of 2-alkynyl-N,N-dimethylanilines
into 1-methylindoles (Scheme 1a). The reaction starts with the
electrophilic activation of the CC bond, followed by
intramolecular nucleophilic attack and elimination of the N-
methyl group as methyl iodide.2
In some cases, the cyclizations are accompanied by an
intramolecular migration of the N-substituent into the newly
formed heterocyclic ring.3a Along with anilines, their oxygen
and sulfur analogues can also undergo similar cyclizations.
Virtually, in all of these instances, the migratory function was
represented by a relatively stable SN1 group: allyl,3b propargyl,3c
acyl,3d (α-alkoxyalkyl),3e (p-methoxyphenyl)methyl (MPM),3f
α-phenethyl,3g RSO2,3h R3Si.3i Very often, the exact nature of
these migrations remained unexplored, but in a few cases, the
Received: October 15, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jo502363t | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 1. Examples of Electrophile-Activated Cyclization (a) and Metal-Catalyzed Rearrangement (b) of 2-Alkynyl-N,N-
dimethylanilines into N-Methylindole Derivatives
Scheme 2. Synthesis of 2-Ethynyl- and 2,7-Diethynyl-1,8-bis(dimethylamino)naphthalenes Using Isolation Protocol A
RESULTS AND DISCUSSION
Scheme 3. Sonogashira Coupling of Iodides 1a,b with 1-
Alkynes Using Isolation Protocols B,C
■
Our work started with the synthesis of previously unknown
acetylenes 2 and 3 by the Sonogashira coupling of iodides 1a,b
and 1-alkynes using the Pd2dba3/CuI/Ph3P/K2CO3 catalytic
system (Scheme 2).6 Normally, the reactions with phenyl- and
p-tolylacetylenes were conducted by heating the reactants in
DMF solution at 60−65 °C for 8−10 h. The coupling of 1a
with trimethylsilylacetylene (TSA) was carried out with
triethylamine as a solvent and a base, simultaneously.
Usually, to isolate compounds 2 and 3, we added a saturated
aqueous solution of NaCl to the reaction mixture and then
extracted the products with diethyl ether (isolation protocol
A).6 Once, upon synthesizing compound 2a, the reaction
mixture was poured into a porcelain basin and evaporated to
dryness on a hot water bath (isolation protocol B).
Unexpectedly, after flash column chromatography of the
residue, N,N,1,3-tetramethyl-2-phenyl-1H-benzo[g]indol-9-
amine (4a) was isolated as a single product in 55% yield.
When the similar procedure was applied to the reaction of
iodide 1a with p-tolylacetylene, benzo[g]indole 4b was
obtained in 60% yield (Scheme 3). The presence of the 1,3-
dimethylindole fragment was proved by single-crystal X-ray
analysis (Figure S1, Supporting Information). The same
isolation protocol in the case of coupling diiodide 1b with
phenylacetylene (4 equiv) gave compound 6 (40%), testifying
that the second pyrrole ring was not closed, most likely due to a
non-proton sponge nature of 6. Instead, along with compound
6, diacetylene 7 was also isolated in 15−30% yield (depending
on the amount of phenylacetylene taken) and structurally
characterized (Figure S2, Supporting Information).
iodide 1a with TSA, which resulted in the formation of
acetylene 2c (24% yield) together with benzo[g]indole 5a
(55%) without the 3-methyl group in the pyrrole ring (see
comment at the final part of this section). It should be noted
that, in the last case, to provide the comparable conditions of
isolation with 4a,b and 6, the reaction mixture was poured out
into a porcelain basin and evaporated to dryness. Then, a small
Obviously, the most intriguing in conversions 1a → 4 and
1b → 6 is the migration of one N-methyl group into the pyrrole
ring of the product under mild conditions that formally
represents a rather deep rearrangement of the proton sponge
acetylenes. The only exception was the Sonogashira reaction of
B
dx.doi.org/10.1021/jo502363t | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 1. Influence of Different Additives on Cyclization of 2-Phenylethynyl Derivative 2a (DMF, 90−95 °C)
a
reaction conditions
yield (%)
entry
additive
reaction vessel time, h atmosphere 2a recov. 4a
5b
8
9a
1
2
3
4
5
Pd2dba3 (5 mol %) porcelain shards (∼1.3 g)
Pd2dba3 (5 mol %)
glass flask
3
air
air
air
air
air
55
50
50
60
50
7
9
7
7
7
6
5
porcelain basin
glass flask
1.5
3
Pd2dba3 (5 mol %) Ph3P (30 mol %) porcelain shards (∼1.3 g)
Pd2dba3 (5 mol %) CuI (20 mol %)
7
porcelain basin
glass flask
1.5
3
23
25
Pd2dba3 (5 mol %) CuI (20 mol %) Ph3P (30 mol %)
porcelain shards (∼1.3 g)
1
6
7
Pd2dba3 (5 mol %) CuI (20 mol %) Ph3P (30 mol %)
porcelain basin
1.5
3
air
air
65
35
10
20
16
Pd2dba3 (5 mol %) CuI (20 mol %) Ph3P (30 mol %) K2CO3 (1 equiv) glass flask
porcelain shards (∼1.3 g)
Pd2dba3 (5 mol %) CuI (20 mol %) Ph3P (30 mol %) K2CO3 (1 equiv) porcelain basin
20
10
8
9
1.5
1.5
1.5
2
air
30
5
19
CuI (30 mol %)
glass flask
glass flask
glass flask
glass flask
glass flask
glass flask
porcelain basin
glass flask
glass flask
glass flask
glass flask, rt
air
45
49
10
11
12
13
14
15
16
17
18
19
CuI (2 equiv)
air
CuI (2 equiv)
argon
air
75
CuCl2 (2 equiv)
1.5
3
35
PdCl2 (40 mol %)
air
10
13
13
14
30
12
60
2
PdCl2 (40 mol %)
2
argon
air
PdCl2 (40 mol %)
3
PdCl2 (40 mol %) CuI (2 equiv)
PdCl2 (40 mol %) CuI (2 equiv)
PdCl2 (40 mol %) CuCl2 (2 equiv)
PdCl2 (40 mol %) CuCl2 (2 equiv)
2
argon
air
2
6
9
6
10
10
30
3
air
2
24
air
2
a
For a typical load and further details, see the Experimental Section. In most cases, the reaction was conducted at least twice.
amount of DMF was added to the residue and evaporation was
repeated (isolation protocol C).
coupling of 1a with 1-alkyne, the porcelain shards were
added into the reaction glass vessel. The resultant mixture was
then heated for 3 h under aerobic conditions and treated in
accordance with isolation protocol A. To our delight,
compounds 4a,b were isolated in this case as the main
products in 40−50% yield.
At first sight, it looks obvious that benzo[g]indoles 4 similar
to 1,3-dimethylindoles in Scheme 1b should be formed from
acetylenes 2. However, in reality, the situation turned out to be
quite uncertain. To make sense of it, we have conducted a
number of control experiments including attempts of direct
conversion of authentic acetylenes 2 into benzo[g]indoles 4
(see below).
Encouraged by this fact, we then tried to achieve a direct
transformation of acetylenes 2 into pyrroles 4. In a series of
control experiments with reference alkyne 2a, we varied
reaction conditions and additives, including porcelain shards
(Table 1). Both a glass flask and an open porcelain basin were
used as the reaction vessels, and the experiments were
conducted either in an inert (argon) or in an air atmosphere.
We found that, in all experiments with Pd2dba3 taken either
alone or with adding Ph3P (entries 1−3), the starting
compound remained mostly unchanged; at the same time,
two new benzo[g]indole derivatives were formed: compound
5b with a missed methyl group and 3,3′-di(benzo[g]indole) 8,
each in 5−9% yield. The yield of 8, whose structure was proved
by X-ray study (Figure 1),7 was markedly increased at the joint
presence of Pd2dba3 and CuI (entries 4−6). With regard to
compound 4a, the most successful was the experiment in which
we applied a full set of additives normally used for Sonogashira
coupling of 1, but with adding porcelain shards (entry 7). In
this case, compound 4a was obtained in 20% yield together
with 16% of dimer 8. The close results were obtained when the
above reaction mixture was evaporated in an open porcelain
basin with the difference that some amount of compound 5b
was isolated (entry 8). A similar experiment with p-tolylethynyl
derivative 2b gave indole 4b in 22% yield. Apparently, along
First, we made sure that just the temperature increase at the
Sonogashira coupling of 1a with phenylacetylene from 60−70
to 95−150 °C and subsequent use of isolation protocol A does
not lead to benzo[g]indole 4a. Along with this, evaporation of
DMF from the reaction mixture in a rotary evaporator resulted
in the formation of acetylene 2a and only a trace amount of 4a.
Then, we assumed that air oxygen may be involved into the
transformation (for example, via the oxidation of Pd0 to Pd2+
species, which then catalyze the pyrrole ring closure). However,
when the reaction mixture after the completion of the
Sonogashira coupling was heated for some time under aerobic
conditions, followed by isolation protocol A, no benzo[g]indole
4a was detected. From this, one can conclude that the
evaporation of DMF directly from the reaction vessel is not
sufficient for the formation of compounds 4 and 6. Bringing the
reaction mixture into an open porcelain basin and evaporating
it to dryness on a hot water bath was the only way that gave
selectively benzo[g]indole 4a in appreciable yield. To make the
possible catalytic effect of porcelain on the formation of
benzo[g]indoles 4 more convincing, we have conducted a
simple experiment. After completion of the Sonogashira
C
dx.doi.org/10.1021/jo502363t | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
CuI under an argon atmosphere (entry 16): the yield of 8 was
even more impressive (up to 60%).
Again, the use of PdCl2/CuI and PdCl2/CuCl2 mixtures
under aerobic conditions (entries 17−19) led to the tarring of
the reaction mixture, and the reaction products 5b, 8, and 9a
were isolated in low yields. Among them, ketone 9a was
predominant. As to the formation of the latter, it should be
noted that two groups of researchers have recently reported on
the copper-catalyzed10 and palladium−copper cocatalyzed11
synthesis of 3-aroylindoles 15 from ortho-alkynylated N,N-
dimethylanilines 11 (Scheme 5). They found that, on treatment
Scheme 5. Oxidative Conversion of 2-Alkynyl-N,N-
dimethylanilines into 3-Aroylindoles10,11
Figure 1. ORTEP plot for X-ray structure of 8 indicating the shortest
distance (Å) observed between the ipso-carbon atoms of the phenyl
rings (P = 50%, 120 K). Hydrogen atoms are omitted for clarity.
with the porcelain catalysis, K2CO3 may play here the key role,
which is discussed below in more detail. With regard to the
formation of 8, it likely results from the widely spread metal
catalyzed oxidative coupling of the electron-rich aromatic and
heteroaromatic compounds.8 The question is which metal
(palladium or copper) serves here as a catalyst.
Further, we found that the course of cyclization of acetylene
2a principally changed when CuI alone was used as an additive
(entries 9 and 10). In this case, previously unknown 3-
benzoylbenzo[g]indole 9a was obtained in 45−49% yield as the
only isolable product. By analogy, treatment of acetylenes 2b
and 3a with CuI gave ketones 9b and 10 in 30% and 25% yield,
respectively (Scheme 4). The structure of 9b was proved by X-
of these acetylenes with CuBr or PdBr2/CuI in the presence of
tert-butyl hydroperoxide (TBHP) in DMSO10 or toluene11 at
80−100 °C, the 3-aroylindoles 15 were formed in moderate to
good yields. Notably, either CuBr or TBHP alone did not give
any trace of the indoles. It is believed that TBHP oxidizes the
aniline NMe2 group into methyleneiminium salt 12, which then
undergoes the copper(palladium)-promoted cyclization. It was
also shown that the carbonyl oxygen in ketones 15 originates
either from water10 or from TBHP11 in the result of
nucleophilic addition 12 → 13; the subsequent oxidative
aromatization of thus formed indoline 14 gives 15.
Scheme 4. Synthesis of 3-Aroylbenzo[g]indoles 9 and 10
Taking into account this mechanistic approach and a
pronounced ability of the proton sponge NMe2 groups to be
oxidized into methyleneiminium salts by some transition
metals12 or under conditions of the so-called tert-amino
reactions,13 we first believed that the formation of ketones 9
and 10 could be represented as shown in Scheme 6. However,
two points cast doubt on the correctness of this assumption.
The first one is an inertness of the proton sponge acetylenes
toward air oxygen and copper(I) salts taken separately. The
second point is a strongly manifested tendency13 of the proton
sponge methyleneiminium intermediates to cyclize into the
dihydroperimidinium salts of type 16, which has not been
registered by us in any case.
On the basis of this, a mechanism of the radical cyclization
involving the joint participation of copper and oxygen looks
more preferable (Scheme 7). This is supported by the fact that
no cyclization 2a → 9a occurs in an argon atmosphere, and
acetylene 2a being largely regenerated (Table 1, entry 11).
Presumably, the oxygen is involved into two reaction stages.
First, it converts Cu+ into Cu2+ ions, which oxidize acetylene 2
into radical-cation 20. The latter then loses a proton (a
tremendous CH acidity of organic radical-cations including
those of N,N-dimethylanilines is well-documented14) and thus
formed radical 21 is then cyclized on the triple bond. Further
ray study (Figure S3, Supporting Information). We also found
that the reaction with CuI additive carried out in a glass vessel
under argon does not produce any products (entry 11). Hence,
it was assumed that oxidation of Cu+ to Cu2+ by the air oxygen
is a crucial factor for transformation 2a → 9a. Indeed, the
reaction with CuCl2 additive gave 9a as the only product in
35% yield (entry 12).
From the results of entries 4−10, we have concluded that
transformation 2a → 8 is Pd2+-catalyzed or palladium−copper
cocatalyzed. Noteworthy, the use of PdCl2 without any other
additives, especially in air atmosphere, was accompanied by a
strong tarring and resulted in a rather modest overall yield of
5b and 8 (entries 13 and 15). Most likely, this is caused by the
high oxidative potential of Pd2+.9 This should lead to the easier
formation of radical species reacting with O2 and with each
other to produce oligomers and resins. The process becomes
more controlled under an inert atmosphere as in run 14
producing 5b and 8 in 13% and 30% yield, respectively. An
argument in favor of the palladium−copper cocatalysis was
received when acetylene 2a reacted with a mixture of PdCl2 and
D
dx.doi.org/10.1021/jo502363t | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Thus, we were almost unsuccessful in modeling the selective
and high yield formation of 3-methylbenzo[g]indoles 4 directly
from acetylenes 2. Evidently, a very complicated combination
of factors operates in the Sonogashira reaction itself (see, for
example, ref 17) and at the subsequent conversion of acetylenes
2 into benzo[g]indoles. These may include influence of air
oxygen, moisture, temperature changes, and diversity of ligands
along with character of their coordination with the transition
metal species. One can also assume that the proton sponge
nature of the substrates may be even more important factor.
In light of the above, an impression arises that acetylenes 2
producing in the Sonogashira coupling are presented in the
final reaction mixture in a form, which is incapable to cyclize
into benzo[g]indoles 4. Hypothetically, palladium or copper
complexes like 24 or 25 (Figure 2; for the related complexes;
Scheme 6. Ionic Pathway for Copper-Catalyzed Conversion
of 2-Alkynyl-1,8-bis(dimethylamino)naphthalenes into 3-
Aroylbenzo[g]indoles 9
Scheme 7. Radical Pathway for Copper-Catalyzed
Conversion of 2-Alkynyl-1,8-
bis(dimethylamino)naphthalenes into 3-
Aroylbenzo[g]indoles 9
Figure 2. Possible complexes of 2-alkynyl-1,8-bis(dimethylamino)-
naphthalenes resisting cyclization into benzo[g]indoles under standard
conditions of the Sonogashira reaction (L: ligands).
see refs 18 and 19) as well as protonated acetylenes 26, in
which the unshared electron pair of the 1-NMe2 group is
blocked for heterocyclization, seem the most probable
candidates for the nonreactive forms. Of these, we prefer salt
26, guided by the following considerations. First, palladium and
copper catalysts for Sonogashira coupling are normally used in
much less than 1 equiv quantity and, therefore, can bind only a
small portion of the proton sponge acetylene formed. On the
other hand, the protons are evolved in the Sonogashira catalytic
cycle (at the formation of a copper acetylenide) in an excessive
amount.17 Because of the abnormally high basicity of the
proton sponge (pKa 12.1, H2O),15 which is even higher (by
0.2−0.8 pKa units)6 in 2-alkynyl- 2 and 2,7-dialkynyl derivatives
3, the latter should exist in the crude reaction mixture mainly in
the protonated form.20,21 Obviously, in the presence of
porcelain in combination with DMF, high temperature, and
some other components, most likely K2CO3, the unreactive
form of the proton sponge ortho-acetylenes becomes capable of
the transformation into benzo[g]indoles regardless of using the
isolation protocols A or B.
At the moment, all the transformations we observed in the
present study may be generalized by Scheme 8. The cyclization
process itself obviously requires the participation of oxygen to
convert Pd0 into Pd2+. The role of Pd2+ consists in activation of
the triple bond as shown in Scheme 8. At the beginning, the
activated form of ortho-ethynyl derivative arbitrarily shown as
27 undergoes cyclization into 28. The latter can be further
transformed into 1,3-dimethylbenzo[g]indole 4 either via [1,3]
migration of the methyl group (28 → 30 → 4; see details in
conversion of the pyrrole intermediate 22 into 9 demands
participation of the superoxide radical-anion, which is typical
for the autooxidation processes. Remarkably, unlike the
reaction shown in Scheme 5, no TBHP is demanded for
conversion 2 → 9. The absence of TBHP also narrows a range
of the possible oxygen atom donors at the formation of the
carbonyl group in 9. As seen, the copper(I)-catalyzed character
of the process is provided by the regeneration of the catalyst as
the result of one-electron reduction of the Cu(II) species with
the highly electron-donor proton sponge system.15
One of the striking things in the above findings is a switching
of cyclization mode when replacing palladium catalysts by
purely copper ones. Indeed, while at the formation of pyrroles
4−8, the CC bond behaves as an electrophile, being attacked
at the Cβ atom by the nitrogen nucleophile, in the case of 9 and
10, the reaction center is moved to the Cα atom reacting with
the N-methylene group. Although details of this phenomenon
are currently unclear, most likely it results from the different
fashion and strength of coordination of palladium and copper
ions with the acetylene ligand. Copper possesses weaker
coordination ability,16 and in the presence of more powerful
palladium species, it’s binding with π-ligands should be
essentially suppressed, which is actually observed in many
runs listed in Table 1.
E
dx.doi.org/10.1021/jo502363t | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 8. Proposed Generalized Mechanism for Pd-Catalyzed Formation of Different Benzo[g]indoles
Scheme 9. Proposed Mechanism for the N-Methyl Group [1,3] Migration Assisted by Carbonate Ion
Scheme 9) or that is less probable via preliminary loss of
methyl, e.g., as MeI, with subsequent methylation of 3-
pyrrolylpalladium compound 29 (Scheme 8a). Clearly,
intermediate 29 in the case of escaping MeI from the reaction
zone undergoes protolytic demetalation to yield benzo[g]indole
5.
Similar to 28 and 29, the participation of related intermediate
31 can explain the formation of diacetylene 7 (Scheme 8b). As
to the formation of di(benzo[g]indole) 8, we believe that it
results from transmetalation between two Pd+−aryl intermedi-
ates 29, followed by reductive elimination of Pd0 from ArPdAr
complex 32 (Scheme 8c). Pd0 is then reoxidized to Pd2+ with
air oxygen.22 Another possible way involves sequential C−H
activation of indole 5 by the Pd+−aryl intermediate 29 and
reductive elimination. Recently, on the basis of kinetic studies,
H/D exchange experiments, and kinetic isotope effects, it has
been shown that Pd-catalyzed aerobic oxidative coupling of
arenes proceeds via a bimetallic/transmetalation mechanism.23
From this, transformation 29 + 29 → 32 → 8 seems to be more
probable. The high yield of dimer 8 on heating of 2a with
PdCl2/CuI additives under an argon atmosphere (Table 1,
entry 16) may be a consequence of the tendency of Cu+ to
disproportionate to Cu2+ and Cu0. The thus generated Cu2+
then triggers reoxidation of Pd0.24
It is known that the ease of the Pd0 → Pd2+ air oxidation
depends strongly on the ligand surrounding of Pd0.22
Apparently, such a surrounding in entries 1−3 is not favorable
for the oxidation, which can explain the small yields (13−14%
in the sum) of indoles 5b and 8. In contrast, in the control
experiments (entries 4−8) where CuI and K2CO3 additives
were used, the benzo[g]indole total yield doubles. The addition
of potash gives the most notable results. Actually, only in the
presence of K2CO3 (Table 1, entries 7, 8) benzo[g]indole 4a
with the transferred methyl group appears among the reaction
products. Effect of this salt may be attributed to the bridging
nature of the carbonate ligand that promotes the Me group to
be transferred from the nitrogen heteroatom to the C(3) atom,
as depicted in Scheme 9.
Another fundamental question is a possible mechanism of
the porcelain catalysis, which also plays a crucial role in the
F
dx.doi.org/10.1021/jo502363t | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
[1,3] methyl group transfer. Porcelain is known to be a solid,
porous material made by thermal treatment of clays in the
presence of various additives.25a Like clays and zeolites,25b,c
porcelain has an aluminosilicate nature. Two former materials
are widely used in the chemical industry, especially in petrol
chemistry, as very effective and cheap catalysts for many
reactions, in particular for the hydrocarbon isomerizations.
Their activity as solid acid catalysts stems primarily from two
features: (1) a pronounced ability to ion-exchange and (2) the
presence within their structure of a plurality of cracks and
cavities of a size that is commensurable with conventional
molecules. There are distinct grounds to believe that both of
these factors should also work in the case of porcelain. It seems
reasonable to suggest that, in our case, the porcelain catalyzes
not so much the pyrrole ring closure as the [1,3] migration of
the methyl group producing benzo[g]indoles 4 and 6.26
Presumably, the migration process shown in Scheme 9 occurs
inside the tight nanosized cavities of the porcelain matrix27 that
considerably lowers the activation energy for the migration.
One can suggest that, in the absence of the effective Me group
shuttle, e.g., K2CO3, capturing the methyl group from
intermediate 28 by other nucleophiles (I−, Ph3P) yields the
some amount of benzo[g]indoles 5a,b without C(3) methyl
groups (entries 1−6).
construction. At the moment, it is rather difficult to elucidate
the exact reaction mechanism in each particular case, especially
for transformation i due to a complex overlapping specific
reactivity of the proton sponges15 and peculiarities of the
transition metal catalysis. Under these circumstances, the above
findings, especially the porcelain catalysis, little mentioned in
the chemical literature, demand further studies, which are now
in progress.
EXPERIMENTAL SECTION
■
1
General Methods. H and 13C NMR spectra were recorded on a
250 MHz spectrometer. Chemical shifts are referred to TMS. Infrared
(IR) spectra were recorded in nujol or KBr. Mass spectra were
measured in electron impact (EI) mode. CHN analysis was
accomplished by combustion analysis (Dumas and Pregl method).
Melting points were determined in glass capillaries on a Stuart SMP30
device and are uncorrected. Flash column chromatography was
performed on Al2O3. Laboratory porcelain ware (see photos S1 and
S2 in the Supporting Information) was purchased from Rechitskiy
Porcelain Factory (www.rfz.ru).
Synthesis of 2-Alkynyl-1,8-bis(dimethylamino)naphthalenes
2a,b (General Procedure). CuI (76 mg, 0.4 mmol), Pd2dba3 (73 mg,
0.08 mmol), Ph3P (210 mg, 0.8 mmol), and K2CO3 (345 mg, 2.5
mmol) were added to 2-iodo-1,8-bis(dimethylamino)naphthalene
(1a)28 (680 mg, 2.00 mmol) in dry DMF (11 mL) under a slow
stream of argon. After stirring for 10 min under argon at 40 °C, 1-
alkyne (5.0 mmol) was added dropwise. The stirring was continued for
8 h at 60−65 °C. The reaction mixture was then mixed with saturated
solution of NaCl (20 mL) and extracted with ether (3 × 20 mL). The
organic phase was evaporated to dryness. The residue was purified by
flash column chromatography on Al2O3 (2 × 20 cm) with CHCl3/
hexane (1:3, v/v) as eluent. The yellow fraction with Rf 0.2−0.4 was
separated, and the crude product was purified additionally by flash
column chromatography on Al2O3 (2 × 20 cm) with the same eluent.
Finally, the yellow fraction with Rf 0.2 gave 2.
The structural assignment of compounds synthesized, beside
elemental analysis and X-ray measurements, was based on
1
spectral data. In this regard, IR and H and 13C NMR spectra
were especially informative. In particular, the IR and 13C NMR
spectra indicated the absence of the CC bonds in most
cyclization products (except 6, 7, and 10). Normally, the
carbon atoms of the CC bond in starting alkynes 2 and 3
give two signals at δ 82−99 ppm, which disappear after
1
cyclization. Even more important are the H NMR spectra, in
1,8-Bis(dimethylamino)-2-(phenylethynyl)naphthalene (2a). 2a
was obtained in 75% yield as a yellow solid, mp 98−100 °C
(EtOH). EIMS (m/z) (rel intensity) 314 (M+; 90), 299 (24), 282
(72), 268 (77), 254 (29), 226 (35), 207 (37), 196 (27), 167 (26), 157
(25), 149 (29), 141 (29), 133 (25), 127 (53), 113 (38), 103 (26), 91
(70), 77 (68), 58 (72), 51 (35), 44 (100). Anal. Calcd for C22H22N2:
which the cyclization product manifests itself in the three-
proton intensity decrease in the NMe2 region at δ 2.5−3.0 ppm
and the appearance of a singlet around 4 ppm, which is typical
for the pyrrolic N−Me group.
1
CONCLUSIONS
C, 84.04; H, 7.05; N, 8.91. Found: C, 83.87; H, 7.23; N, 9.03. H
■
NMR (CDCl3, 250 MHz) δ 2.78 (s, 6H), 3.16 (s, 6H), 6.94 (dd, J =
6.0, 2.8 Hz, 1H), 7.24−7.40 (m, 7H), 7.53 (dd, J = 7.7, 1.8 Hz, 2H).
13C NMR (CDCl3, 62.9 MHz) δ 45.0, 45.2, 91.5, 94.5, 114.2, 114.7,
122.2, 123.0, 124.9, 126.8, 128.2, 128.8, 131.0, 131.3, 138.3; 152.1,
152.4.
In summary, during the Sonogashira synthesis of a number of
2-alkynyl- and 2,7-dialkynyl-1,8-bis(dimethylamino)-
naphthalenes, we have disclosed their pronounced ability to
cyclize into benzo[g]indoles. Four main channels were found
for these transformations: (i) a palladium and porcelain
cocatalyzed rearrangement with [1,3] migration of one of the
N−Me groups into the newly formed pyrrole ring, (ii)
cyclization into benzo[g]indoles with elimination of the N−
Me group, (iii) a palladium-catalyzed cyclization into benzo-
[g]indoles with their subsequent dimerization, and (iv) a
copper-assisted oxidative cyclization into 3-aroylbenzo[g]-
indoles. Of these, only transformations iii and iv can be
selectively achieved with the authentic acetylenes. The
transformation i is selectively realized only at using a specific
isolation procedure (porcelain catalysis) after completion of the
Sonogashira coupling of 2-iodo-1,8-bis(dimethylamino)-
naphthalenes with 1-alkynes. Notably, in the first three
reactions, the 1-NMe2 group serves as a nucleophile, while
the β-carbon atom of the CC bond acts as an electrophile. In
transformation iv, the 1-NMe2 group is likely oxidized by
Cu(I)/air O2 into aminomethyl radical species and the reaction
center of the triple bond is moved to the α-carbon atom. The
last conversion differs not only by its mechanism but also by
the character of building blocks for the pyrrole ring
1,8-Bis(dimethylamino)-2-(p-tolylethynyl)naphthalene (2b). 2b
was obtained in 75% yield as yellow solid, mp 89−92 °C (EtOH).
EIMS (m/z) (rel intensity) 328 (M+; 100), 313 (25), 296 (74), 282
(66), 268 (17), 206 (18), 196 (20), 164 (21), 149 (22), 141 (17), 134
(18), 127 (20). Anal. Calcd for C23H24N2: C, 84.11; H, 7.37; N, 8.53.
1
Found: C, 84.24; H, 7.26; N, 8.37. H NMR (CDCl3, 250 MHz) δ
2.36 (s, 3H), 2.79 (s, 6H), 3.16 (s, 6H), 6.95 (dd, J = 6.3, 2.4 Hz, 1H),
7.15 (d, J = 7.9 Hz, 2H), 7.25−7.46 (m, 6H). 13C NMR (CDCl3, 62.9
MHz) δ 21.9, 45.0, 45.2, 90.1, 94.7, 114.2, 115.0, 121.8, 122.3, 123.0,
123.1, 126.7, 129.6, 131.0, 131.2, 138.2, 138.3, 152.0, 152.1.
Synthesis of 1,8-Bis(dimethylamino)-2-(trimethylsilyl-
ethynyl)naphthalene (2c). CuI (0.19 g, 1.0 mmol), Pd2dba3 (0.46
g, 0.5 mmol), and Ph3P (0.79 g, 3.0 mmol) were added to 2-
iodonaphthalene 1a (3.40 g, 10.0 mmol) in dry Et3N (35 mL) under a
slow stream of argon. After stirring for 10 min under argon,
trimethylsilylacetylene (3.2 mL, 2.24 g, 23.0 mmol) was added. The
flask was hermetically sealed, and the stirring was continued for 4 h at
50 °C. The reaction mixture was then evaporated to dryness. The
residue was purified by flash column chromatography on Al2O3 (2 ×
20 cm) with n-hexane as eluent. The yellow fraction with Rf 0.2−0.4
was separated, and the crude product was purified additionally by flash
column chromatography on Al2O3 (2 × 20 cm) with Et2O/hexane
G
dx.doi.org/10.1021/jo502363t | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(1:3, v/v) as eluent. The yellow fraction with Rf 0.2 gave 2.59 g (83%)
of 2c as a dark yellow oil. 2c: IR (Nujol) 2140 (CC), 1553 (ring)
cm−1; EIMS (m/z) (rel intensity) 310 (M+, 51), 295 (29), 279 (24),
264 (22), 238 (30), 221 (17), 206 (51), 192 (30), 73 (100). Anal.
Calcd for C19H26N2Si: C, 73.49; H, 8.44; N, 9.02. Found: C, 73.58; H,
7.08 (d, J = 7.5 Hz, 1H), 7.27−7.33 (m, 3H), 7.42 (d, J = 7.9 Hz, 2H),
7.52 (t, J = 8.6 Hz, 2H), 7.64 (d, J = 8.4 Hz, 1H); 13C NMR (CDCl3,
62.9 MHz) δ 10.2, 21.8, 39.2, 43.4, 110.8, 113.8, 118.5, 119.3, 121.7,
123.1, 124.0, 127.7, 129.6, 130.4, 130.9, 134.5, 135.3, 137.5, 141.1,
148.8.
1
8.29; N, 9.23. H NMR (CDCl3, 250 MHz) δ 0.29 (s, 9H), 2.78 (s,
Synthesis of N,N,1-Trimethyl-1H-benzo[g]indol-9-amine (5a)
and N1,N1,N8,N8-Tetramethyl-2-((trimethylsilyl)ethynyl)-
naphthalene-1,8-diamine (2c). CuI (57 mg, 0.3 mmol), Pd2dba3
(128 mg, 0.14 mmol), and Ph3P (262 mg, 1.0 mmol) were added to 2-
iodonaphthalene 1a (1.02 g, 3.0 mmol) in dry Et3N (13 mL) under a
slow stream of argon. After 15 min stirring under argon, trimethyl-
silylacetylene (1 mL, 0.7 g, 7.2 mmol) was added. The flask was
hermetically sealed, and the stirring was continued for 8 h at 50−60
°C. The reaction mixture was then evaporated to dryness. To the
residue was added DMF (5 mL). The resultant mixture was poured
into a porcelain basin and evaporated to dryness on the water bath.
The residue was purified by flash column chromatography on Al2O3
with CHCl3/hexane (1:2, v/v) as eluent. The colorless fraction with Rf
0.9 gave 5a (370 mg, 55%) as beige crystals. The yellow fraction with
Rf 0.2 gave 2c (220 mg, 24%) as dark yellow oil.
6H), 3.13 (s, 6H), 6.92−6.98 (m, 1H), 7.25−7.35 (m, 4H). 13C NMR
(CDCl3, 62.9 MHz) δ 0.5, 45.0, 45.1, 82.4, 99.1, 107.3, 114.0, 122.0,
122.1, 122.7, 126.8, 131.3, 138.4, 152.1, 153.1.
Synthesis of 2,7-Dialkynyl-1,8-bis(dimethylamino)naph-
thalenes 3a,b (General Procedure). CuI (40 mg, 0.21 mmol),
Pd2dba3 (46 mg, 0.05 mmol), Ph3P (131 mg, 0.50 mmol), and K2CO3
(280 mg, 2.02 mmol) were added to 2,7-diiodo-1,8-bis(dimethyl-
amino)naphthalene (1b)28 (466 mg, 1.00 mmol) in dry DMF (12
mL) under a slow stream of argon. After stirring for 10 min under
argon at 40 °C, 1-alkyne (4.00 mmol) was added dropwise. The
stirring was continued for 10 h at 60−65 °C. The reaction mixture was
then evaporated to dryness. The residue was purified by flash column
chromatography on Al2O3 (2 × 20 cm) with CHCl3/hexane (1:3, v/v
for 3a; 1:2, v/v for 3b) as eluent. The yellow-orange fraction with Rf
0.2 gave 3.
5a: mp 73−75 °C (hexane); IR (KBr) 2959, 2926, 2853, 2790,
1600, 1594, 1556, 1525, 1510 cm−1; EIMS (m/z) (rel intensity) 224
(M+; 68), 208 (28), 193 (35), 180 (49), 166 (29), 152 (39), 139 (50),
127 (25), 112 (63), 104 (63), 97 (57), 90 (41), 83 (48), 77 (49), 69
(25), 63 (57), 57 (29), 51 (30), 42 (100). Anal. Calcd for C15H16N2:
1,8-Bis(dimethylamino)-2,7-bis(phenylethynyl)naphthalene (3a).
3a was obtained in 64% yield as a yellow solid, mp 156−157 °C
(EtOH or n-octane); IR (Nujol) 2211 cm−1 (CC); EIMS (m/z)
(rel intensity) 414 (M+; 94), 399 (39), 382 (100), 368 (31), 307 (25),
91 (29), 58 (23). Anal. Calcd for C30H26N2: C, 86.92; H, 6.32; N, 6.76.
1
C, 80.32; H, 7.19; N, 12.49. Found: C, 80.18; H, 7.03; N, 12.64. H
1
Found: C, 87.09; H, 6.17; N, 6.84. H NMR (CDCl3, 250 MHz) δ
NMR (CDCl3, 250 MHz) δ 2.73 (s, 6H), 4.06 (s, 3H), 6.64 (d, J = 3.0
Hz, 1H), 7.12−7.15 (m, 2H), 7.34 (t, J = 7.7 Hz, 1H), 7.46 (d, J = 8.4
Hz, 1H), 7.57 (d, J = 7.8 Hz, 1H), 7.66 (d, J = 8.4 Hz, 1H); 13C NMR
(CDCl3, 62.9 MHz) δ 40.1, 43.7, 103.5, 114.1, 118.9, 121.5, 121.8,
123.5, 124.1, 127.4, 131.2, 132.7, 134.3, 148.8.
3.17 (s, 12H), 7.31−7.42 (m, 10H), 7.53−7.57 (m, 4H). 13C NMR
(CDCl3, 62.9 MHz) δ 45.0, 90.8, 94.6, 116.8, 123.5, 124.3, 126.3,
128.1, 128.5, 131.1, 131.5, 137.7, 152.8.
1,8-Bis(dimethylamino)-2,7-bis(p-tolylethynyl)naphthalene (3b).
3b was obtained in 50% yield as yellow solid, mp 155−156 °C (i-
PrOH); IR (Nujol) 2210 cm−1 (CC); EIMS (m/z) (rel intensity)
442 (M+; 100), 427 (40), 410 (89), 396 (33), 320 (22), 221 (20), 205
(21), 191 (17), 175 (26). Anal. Calcd for C32H30N2: C, 86.84; H, 6.83;
Synthesis of N,N,1,3-Tetramethyl-2-phenyl-8-(phenyl-
ethynyl)-1H-benzo[g]indol-9-amine (6) and N,N,1-Trimethyl-
2-phenyl-3,8-bis(phenylethynyl)-1H-benzo[g]indol-9-amine
(7). CuI (40 mg, 0.21 mmol), Pd2dba3 (46 mg, 0.05 mmol), Ph3P (131
mg, 0.5 mmol), and K2CO3 (280 mg, 2.02 mmol) were added to 2,7-
diiodonaphthalene 1b (466 mg, 1.0 mmol) in dry DMF (12 mL)
under a slow stream of argon. After 10 min stirring under argon at 40
°C, 1-alkyne (4.0 mmol) was added dropwise. The stirring was
continued for 10 h at 60 °C. The reaction mixture was poured into a
porcelain basin and evaporated to dryness on the water bath. The
residue was purified by flash column chromatography on Al2O3 with
Et2O/hexane (1:4, v/v) as eluent. The yellow fraction with Rf 0.5 was
separated, and the crude product was purified additionally by flash
column chromatography on Al2O3 with Et2O/hexane (1:4, v/v) as
eluent. The yellow fraction with Rf 0.5 gave 6 (165 mg, 40%) as yellow
crystals. 6: mp 159−160 °C (hexane); IR (Nujol) 2205, 1596, 1536
cm−1; EIMS (m/z) (rel intensity) 414 (M+; 100), 397 (24), 321 (13),
91 (14), 77 (27). Anal. Calcd for C30H26N2: C, 86.92; H, 6.32; N, 6.76.
1
N, 6.33. Found: C, 87.00; H, 6.95; N, 6.13. H NMR (CDCl3, 250
MHz) δ 2.38 (s, 6H), 3.17 (s, 12H), 7.17 (d, J = 7.9 Hz, 4H), 7.32 (d,
J = 8.4 Hz, 2H), 7.40 (d, J = 8.4 Hz, 2H), 7.43 (d, J = 7.9 Hz, 4H). 13C
NMR (CDCl3, 62.9 MHz) δ 22.0, 45.2, 90.4, 95.0, 117.5, 121.6, 123.8,
126.8, 129.6, 131.3, 131.7, 137.8, 138.5, 152.9.
Synthesis of N,N,1,3-Tetramethyl-2-phenyl-1H-benzo[g]-
indol-9-amine (4a). CuI (76 mg, 0.4 mmol), Pd2dba3 (73 mg,
0.08 mmol), Ph3P (210 mg, 0.8 mmol), and K2CO3 (345 mg, 2.5
mmol) were added to 2-iodonaphthalene 1a (680 mg, 2.0 mmol) in
dry DMF (11 mL) under a slow stream of argon. After 10 min stirring
under argon at 40 °C, phenylacetylene (0.55 mL, 510 mg, 5.0 mmol)
was added dropwise. The stirring was continued for 8 h at 60−65 °C.
The reaction mixture was poured into a porcelain basin and
evaporated to dryness on the water bath. The residue was purified
by flash column chromatography on Al2O3 with hexane as eluent. The
colorless fraction with Rf 0.7 gave 4a (345 mg, 55%) as off-white
crystals. 4a: mp 110−112 °C (hexane); IR (Nujol) 1600, 1549 cm−1;
EIMS (m/z) (rel intensity) 314 (M+; 100), 298 (15), 283 (23), 270
(18), 254 (13), 157 (22), 149 (16), 127 (13). Anal. Calcd for
C22H22N2: C, 84.04; H, 7.05; N, 8.91. Found: C, 84.21; H, 7.19; N,
1
Found: C, 86.78; H, 6.49; N, 6.61. H NMR (CDCl3, 250 MHz) δ
2.40 (s, 3H), 3.15 (s, 6H), 3.56 (s, 3H), 7.32−7.56 (m, 13H), 7.65 (d,
J = 8.3 Hz, 1H); 13C NMR (CDCl3, 62.9 MHz) δ 10.1, 39.9, 44.0,
91.2, 94.7, 111.0, 115.6, 120.2, 121.1, 121.6, 124.6, 124.7, 127.9, 128.3,
128.7, 128.8, 128.9, 129.4, 130.9, 131.3, 133.2, 134.5, 135.7, 141.5,
150.0.
1
8.76. H NMR (CDCl3, 250 MHz) δ 2.40 (s, 3H), 2.80 (s, 6H), 3.64
The same protocol with 6.0 mmol of 1-alkyne gave a mixture of
products 6 and 7. After evaporation of the reaction mixture, the
residue was purified by flash column chromatography on Al2O3 with
Et2O/hexane (1:4, v/v) as eluent. The yellow fraction within Rf 0.3−
0.6 was separated. PTLC on Al2O3 with CH2Cl2/hexane (1:2, v/v)
elution gave 6 (Rf 0.5, 66 mg, 16%) and 7 (Rf 0.4, 150 mg, 30%).
7: yellow crystals, mp 176−178 °C (hexane); IR (Nujol) 2202
cm−1; EIMS (m/z) (rel intensity) 500 (M+; 100), 483 (15), 423 (13),
407 (19), 250 (24), 241 (14), 234 (13), 212 (12), 203 (12), 250 (24),
241 (14), 234 (13), 212 (12), 203 (12), 77 (22). Anal. Calcd for
C37H28N2: C, 88.77; H, 5.64; N, 5.60. Found: C, 88.62; H, 5.59; N,
5.78. 1H NMR (CDCl3, 250 MHz) δ 3.17 (s, 6H), 3.71 (s, 3H), 7.28−
7.60 (m, 16H), 7.86−7.92 (m, 3H); 13C NMR (CDCl3, 62.9 MHz) δ
41.2, 44.1, 84.4, 90.8, 92.9, 95.1, 99.1, 116.2, 120.9, 121.1, 123.2, 124.5,
(s, 3H), 7.08 (dd, J = 7.5, 0.9 Hz, 1H), 7.30 (t, J = 7.7 Hz, 1H), 7.34−
7.41 (m, 1H), 7.46−7.55 (m, 6H), 7.63 (d, J = 8.4 Hz, 1H); 13C NMR
(CDCl3, 62.9 MHz) δ 10.2, 39.3, 43.4, 111.2, 113.9, 118.5, 119.3,
121.9, 123.2, 124.2, 127.7, 127,8, 128.9, 131.0, 133.4, 134.6, 135.6,
141.1, 148.9.
Synthesis of N,N,1,3-Tetramethyl-2-p-tolyl-1H-benzo[g]-
indol-9-amine (4b). The reaction was carried out similarly to the
synthesis of 4a with p-tolylacetylene (580 mg, 5.0 mmol). Compound
4b (394 mg, 60%) was obtained as beige crystals, mp 152−154 °C
(hexane); IR (Nujol) 1550 cm−1; EIMS (m/z) (rel intensity) 328
(M+; 66), 164 (100), 156 (57), 149 (41), 141 (38), 134 (43), 127
(50), 121 (20), 115 (15), 91 (16). Anal. Calcd for C23H24N2: C, 84.11;
H, 7.37; N, 8.53. Found: C, 84.00; H, 7.29; N, 8.38. 1H NMR (CDCl3,
250 MHz) δ 2.40 (s, 3H), 2.45 (s, 3H), 2.81 (s, 6H), 3.64 (s, 3H),
H
dx.doi.org/10.1021/jo502363t | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
124.7, 124.8, 127.9, 128.5, 128.7, 128.8, 128.87, 128.90, 129.1, 130.1,
130.5, 131.4, 131.7, 131.9, 135.0, 135.5, 147.1, 149.9.
The colorless fraction with Rf 0.1 gave 9b (51 mg, 30%) as beige
crystals. 9b: mp 159−160 °C (hexane); IR (Nujol) 1621 cm−1; EIMS
(m/z) (rel intensity) 342 (M+; 100), 223 (20), 208 (18), 192 (12),
119 (35), 91 (30). Anal. Calcd for C23H22N2O: C, 80.67; H, 6.48; N,
8.18. Found: C, 80.45; H, 6.33; N, 8.00. 1H NMR (CDCl3, 250 MHz)
δ 2.44 (s, 3H), 2.71 (s, 6H), 4.07 (s, 3H), 7.15 (dd, J = 7.5, 0.8 Hz,
1H), 7.29 (d, J = 8.0 Hz, 2H), 7.36 (t, J = 7.7 Hz, 1H), 7.53 (s, 1H),
7.58 (dd, J = 7.9, 0.6 Hz, 1H), 7.63 (d, J = 8.6 Hz, 1H), 7.80 (d, J = 8.0
Hz, 2H), 8.49 (d, J = 8.6 Hz, 1H); 13C NMR (CDCl3, 62.9 MHz) δ
22.0, 41.1, 43.8, 114.8, 117.1, 118.2, 121.9, 123.6, 124.6, 125.1, 126.5,
129.3, 129.5, 133.8, 134.9, 138.6, 139.4, 142.1, 148.9, 191.4.
General Procedure for the Cyclization of 2a in the Presence
of Additives (Table 1). Reaction in a Glass Flask. A stirred mixture
of phenylethynylnaphthalene 2a (157 mg, 0.5 mmol), DMF (10 mL),
and additives listed in Table 1 was heated at 90−95 °C in a glass flask
for the indicated time. The reaction mixture was then mixed with a
saturated aqueous solution of NaCl (30 mL) and extracted with ether
(4 × 20 mL). The solvent was evaporated to dryness, and the residue
was purified by flash column chromatography on Al2O3 with Et2O/
hexane (1:4, v/v) as eluent. The colorless fraction with Rf 0.7−0.9 was
first collected. It contained a mixture of compounds 4a and 8. The
next fraction with Rf 0.5 gave starting compound 2a (entries 1−7) or
compound 9a (entries 9, 10, 12, 17−19). The first isolated fraction
was additionally chromatographed on Al2O3 with hexane as eluent
collecting compound 4a (entry 7) or compound 5b (entries 1−6, 13−
15, 17−19). Compound 8 was isolated after changing hexane into
Et2O/hexane (1:4, v/v) mixture.
Reaction in a Porcelain Basin. A mixture of 2a (157 mg, 0.5
mmol), DMF (10 mL), and additives listed in Table 1 was placed into
a porcelain basin and evaporated to dryness at heating on a water bath
(90−95 °C, 1.5 h). Isolation of the reaction products was carried out
similarly to the above procedure.
Reaction in a Glass Flask, Inert Atmosphere. A stirred mixture of
2a (157 mg, 0.5 mmol), DMF (10 mL), and additives listed in Table 1
was heated at 90−95 °C for 2 h in a glass flask under argon. Isolation
of the reaction products was carried out similarly to the above
procedures.
Synthesis of (9-(Dimethylamino)-1-methyl-8-(phenyl-
ethynyl)-1H-benzo[g]indol-3-yl)(phenyl)methanone (10). The
reaction was carried out similarly to the synthesis of 9b with 2,7-
bis(phenylethynyl)naphthalene 3a (207 mg, 0.5 mmol). Compound
10 was obtained in 25% yield (54 mg) as yellowish crystals with mp
163−165 °C (heptane); IR (Nujol) 2200 cm−1; EIMS (m/z) (rel
intensity) 428 (M+; 83), 351 (17), 321 (15), 307 (18), 105 (100), 77
(63). Anal. Calcd for C30H24N2O: C, 84.08; H, 5.65; N, 6.54. Found:
1
C, 84.23; H, 5.49; N, 6.71. H NMR (CDCl3, 250 MHz) δ 3.04 (s,
6H), 4.00 (s, 3H), 7.33−7.41 (m, 3H), 7.46−7.57 (m, 7H), 7.62 (dd, J
= 8.5, 2.1 Hz, 2H), 7.87 (dd, J = 7.9, 1.6 Hz, 2H), 8.55 (d, J = 8.5 Hz,
1H); 13C NMR (CDCl3, 62.9 MHz) δ 42.2, 44.2, 90.3, 95.6, 116.7,
117.5, 121.5, 122.8, 124.3, 124.7, 125.4, 127.7, 128.6, 128.7, 128.9,
129.3, 130.5, 131.3, 131.7, 133.8, 134.9, 140.2, 141.3, 149.9, 191.6.
X-ray Diffraction Analysis. Crystals suitable for X-ray studies
were grown by slow evaporation from solutions of compounds in the
appropriate solvents or solvent mixtures: 4b (n-hexane), 7 (EtOAc), 8
(PhMe−CH2Cl2), 9b (Et2O). X-ray experiments were carried out
using a SMART APEX2 CCD [λ(Mo−Kα) = 0.71073 Å, graphite
monochromator, ω-scans] diffractometer. Collected data were
analyzed by the SAINT and SADABS programs incorporated into
the APEX2 program package.29a All structures were solved by the
direct methods and refined by the full-matrix least-squares procedure
against F2 in anisotropic approximation. All hydrogen atoms were
placed in geometrically calculated positions and were refined in
isotropic approximation in riding model. The refinement was carried
out with the SHELXTL program.29b The details of data collection and
crystal structures refinement are summarized in Table S1 (Supporting
Information). CCDC 1029140−1029143 contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from the Cambridge Crystallographic Data Centre via www.
ccdc.cam.ac.uk/data_request/cif.
N,N,1-Trimethyl-2-phenyl-1H-benzo[g]indol-9-amines (5b). 5b
was obtained as beige crystals, mp 83−85 °C; IR (Nujol) 1600,
1578, 1560 cm−1; EIMS (m/z) (rel intensity) 300 (M+, 100), 284
(20), 269 (30), 256 (25), 150 (30), 142 (23), 128 (18). Anal. Calcd
for C21H20N2: C, 83.96; H, 6.71; N, 9.33. Found: C, 84.11; H, 6.59; N,
1
9.18. H NMR (CDCl3, 250 MHz) δ 2.83 (s, 6H), 3.78 (s, 3H), 6.79
(s, 1H), 7.13 (d, J = 7.5 Hz, 1H), 7.30−7.41 (m, 2H), 7.47−7.57 (m,
4H), 7.65−7.68 (m, 3H); 13C NMR (CDCl3, 62.9 MHz) δ 39.8, 43.4,
104.3, 114.1, 118.6, 121.1, 122.6, 123.2, 124.3, 127.0, 127.9, 129.0,
129.4, 133.7, 134.5, 136.6, 145.0, 148.7.
N9,N9,N9′,N9′,1,1′-Hexamethyl-2,2′-diphenyl-1H,1′H-3,3′-
bibenzo[g]indole-9,9′-diamine (8). 8 was obtained as off-white
crystals, mp 284−286 °C (heptane); Rf 0.3 (hexane); IR (Nujol)
1599, 1552 cm−1; EIMS (m/z) (rel intensity) 598 (M+; 64), 299
(100), 291 (21), 283 (24), 277 (16), 268 (26), 260 (18), 254 (22),
215 (13), 168 (21), 118 (17). Anal. Calcd for C42H38N4: C, 84.25; H,
1
6.40; N, 9.36. Found: C, 84.09; H, 6.27; N, 9.57. H NMR (CDCl3,
ASSOCIATED CONTENT
■
250 MHz) δ 2.73 (s, 6H), 2.93 (s, 6H), 3.67 (s, 6H), 6.86 (dd, J = 8.4,
1.5 Hz, 4H), 6.95−7.07 (m, 6H), 7.11 (dd, J = 7.6, 1.0 Hz, 2H), 7.32
(t, J = 7.7 Hz, 2H), 7.50 (d, J = 8.4 Hz, 2H), 7.56 (d, J = 7.9 Hz, 2H),
7.68 (d, J = 8.4 Hz, 2H); 13C NMR (CDCl3, 62.9 MHz) δ 39.7, 42.3,
44.3, 110.5, 113.9, 118.5, 121.0, 122.1, 123.2, 124.2, 126.9, 128.0,
128.6, 130.3, 133.0, 134.4, 135.8, 142.8, 148.8.
(9-(Dimethylamino)-1-methyl-1H-benzo[g]indol-3-yl)(phenyl)-
methanone (9a). 9a was obtained as off-white crystals, mp 129−130
°C (octane); IR (Nujol) 1625 cm−1; EIMS (m/z) (rel intensity) 328
(M+; 100), 312 (15), 284 (13), 223 (36), 208 (37), 192 (24), 105
(68), 77 (67). Anal. Calcd for C22H201N2O: C, 80.46; H, 6.14; N, 8.53.
Found: C, 80.33; H, 5.96; N, 8.47. H NMR (CDCl3, 250 MHz) δ
2.71 (s, 6H), 4.08 (s, 3H), 7.15 (dd, J = 7.6, 1.1 Hz, 1H), 7.37 (t, J =
7.7 Hz, 1H), 7.45−7.60 (m, 5H), 7.64 (d, J = 8.6 Hz, 1H), 7.85−7.89
(m, 2H), 8.51 (d, J = 8.6 Hz, 1H); 13C NMR (CDCl3, 62.9 MHz) δ
41.2, 43.7, 114.9, 117.0, 118.2, 121.9, 123.6, 124.8, 125.2, 126.4, 128.7,
129.3, 131.6, 133.9, 134.9, 139.8, 141.4, 148.9, 191.7.
Synthesis of (9-(Dimethylamino)-1-methyl-1H-benzo[g]-
indol-3-yl)(p-tolyl)methanone (9b). A stirred mixture of p-
tolylethynylnaphthalene 2b (164 mg, 0.5 mmol) and CuI (29 mg,
0.15 mmol) in DMF (15 mL) was heated at 90−95 °C for 2.5 h. The
reaction mixture was then mixed with a saturated aqueous solution of
NaCl (30 mL) and extracted with ether (3 × 20 mL). The solvent was
evaporated to dryness. The residue was purified by flash column
chromatography on Al2O3 with Et2O/CH2Cl2 (1:1, v/v) as eluent.
S
* Supporting Information
1H and 13C NMR spectra for the products, photos of porcelain
basins, ORTEP plots for crystal structures, and crystal data and
structure refinement for compounds 4b, 7, 8, and 9b. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: apozharskii@sfedu.ru (A.F.P.).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Russian Foundation for Basic
Research (projects 11-03-00073 and 14-03-00010). We thank
Drs. Z. A. Starikova (deceased on September 5, 2013) and K. Y.
Suponitsky (A. N. Nesmeyanov Institute of Organoelement
Compounds, Russian Academy of Sciences, Russian Feder-
ation) for X-ray experiments.
I
dx.doi.org/10.1021/jo502363t | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Kazheva, O. N.; Chekhlov, A. N.; Dyachenko, O. A. Org. Biomol. Chem.
2011, 9, 1887−1900.
REFERENCES
■
(1) For reviews, see: (a) Patil, N. T.; Yamamoto, Y. Chem. Rev. 2008,
108, 3395−3442. (b) Godoi, B.; Schumacher, R. F.; Zeni, G. Chem.
Rev. 2011, 111, 2937−2980. (c) Zeni, G.; Larock, R. C. Chem. Rev.
2004, 104, 2285−2309. (d) Adcock, H. V.; Davies, P. W. Synthesis
2012, 44, 3401−3420.
(14) Parker, D. V.; Tilset, M. J. Am. Chem. Soc. 1991, 113, 8778−
8781.
(15) Pozharskii, A. F.; Ozeryanskii, V. A. In The Chemistry of Anilines;
Rappoport, Z., Ed.; J. Wiley & Sons: Chichester, U.K., 2007; Part 2,
Chapter 17, pp 931−1026.
(16) Worrell, B. T.; Malik, J. A.; Fokin, V. V. Science 2013, 340, 457−
460.
(17) Chinchilla, R.; Najera, C. Chem. Rev. 2007, 107, 874−922.
(18) Yamasaki, T.; Ozaki, N.; Saika, Y.; Ohta, K.; Goboh, K.;
Nakamura, F.; Hashimoto, M.; Okeya, S. Chem. Lett. 2004, 33, 928−
929.
(2) Yue, D.; Larock, R. C. Org. Lett. 2004, 6, 1037−1040.
(3) (a) Patil, N. T.; Kavthe, R. D.; Yamamoto, Y. Adv. Heterocycl.
Chem. 2010, 101, 75−95. (b) Cacchi, S.; Fabrizi, G.; Pace, P. J. Org.
Chem. 1998, 63, 1001−1011. (c) Cacchi, S.; Fabrizi, G.; Moro, L.
Tetrahedron Lett. 1998, 39, 5101−5104. (d) Shimada, T.; Nakamura,
I.; Yamamoto, Y. J. Am. Chem. Soc. 2004, 126, 10546−10547.
(e) Nakamura, I.; Mizushima, Y.; Yamamoto, Y. J. Am. Chem. Soc.
2005, 127, 15022−15023. (f) Nakamura, I.; Sato, T.; Yamamoto, Y.
Angew. Chem., Int. Ed. 2006, 45, 4473−4475. (g) Nakamura, I.; Sato,
T.; Terada, M.; Yamamoto, Y. Org. Lett. 2008, 10, 2649−2651.
(h) Nakamura, I.; Yamagishi, U.; Song, D.; Konta, S.; Yamamoto, Y.
Angew. Chem., Int. Ed. 2007, 119, 2334−2337. (i) Nakamura, I.; Sato,
T.; Terada, M.; Yamamoto, Y. Org. Lett. 2007, 9, 4081−4083.
(4) Zeng, X.; Kinjo, R.; Donnadieu, B.; Bertrand, G. Angew. Chem.,
Int. Ed. 2010, 49, 942−945.
(19) Farrer, N. J.; McDonald, R.; McIndoe, J. S. Dalton Trans. 2006,
4570−4579.
(20) For comparison, pKa values of other basic additives using in the
2−
Sonogashira reaction are considerably lower: 10.33 for CO3 and
10.87 for Et3N: Albert, A.; Serjeant, E. P. Ionization Constants of Acids
and Bases: A Laboratory Manual; Methuen and Co., Ltd.: London,
1962.
(21) We have established that the protonated salts of 2-alkynyl- and
2,7-dialkynyl-1,8-bis(dimethylamino)naphthalenes, e.g., hydroiodides
of 2a and 3a, under some specific conditions are able to cyclize into
benzo[g]indole derivatives of type 5b or 6 (without 3-Me group).
These data will be published in a separate communication.
(22) Jin, L.-Q.; Lei, A.-W. Sci. China: Chem. 2012, 55, 2027−2035.
(23) Wang, D.; Stahl, S. S. J. Am. Chem. Soc. 2014, 136, 9914−9917.
(24) Aufiero, M.; Proutiere, F.; Schoenebeck, F. Angew. Chem., Int.
Ed. 2012, 51, 1−6.
(25) (a) Anastas, P. T.; Williamson, T. C. Green Chemistry: Frontiers
in Benign Chemical Syntheses and Processes; Oxford University Press:
Oxford, U.K., 1998. (b) Vaccari, A. Appl. Clay Sci. 1999, 14, 161.
(c) van Bekkum, H. Introduction to Zeolite Science and Practice;
Elsevier: Amsterdam, 2001.
(26) The fact that all readily migrating groups belong to the so-called
SN1 type can indicate that, in the transition state (TS), they form the
corresponding cation or some kind of ionic pair, followed by the C−C
bond formation. Convincing arguments in favor of this mechanism are
given in ref 3g. Since methyl cation is not stabilized, the methyl group
transfer most likely occurs as the direct [1,3] migration (actually as
two consecutive [1,2] migrations).3g It is well-known that the TS of
[1,3] migrations is sterically highly demanded (Mikhailov, I. E.;
Dushenko, G. A.; Minkin, V. I. Molecular Rearrangements of
Cyclopolyenes (in Russian); Nauka: Moscow, 2008; p 19). Probably,
this is a reason why [1,3] migrations of the Me group in the reactions
under consideration occur only under specific conditions and with
high activation energy (see ref 4 along with our case). This view also
elucidates why compound 2c regardless of using isolation protocols B
or C is converted into benzo[g]indole 5a. Apparently, the bulky SiMe3
group strongly inhibits both intra- and intermolecular migration of the
methyl group and the latter is eliminated from 28 reacting with other
nucleophiles.
(27) By request of one of the reviewers, we have carried out the
following experiment. After completion of Sonogashira reaction of 1
with phenylacetylene, molecular sieves 3Å and 4Å (Sigma−Aldrich)
were added into the reaction mixture instead of porcelain shards. The
subsequent use of isolation protocol A gave 4a in 8% and 16%,
respectively. The major part of the reaction products seems to be
oligomers. Clearly, such experiments involving different kinds of
zeolites and clays deserve further studies.
(5) Filatova, E. A.; Pozharskii, A. F.; Boiko, L. Z. In Proceedings of
The III International Conference on Heterocyclic Chemistry
Dedicated to Professor A. N. Kost, October 18−21, Erebus-Press:
Moscow, 2010; U-62.
(6) The detailed protocols of these preparations have been recently
published: Filatova, E. A.; Pozharskii, A. F.; Gulevskaya, A. V.;
Vistorobskii, N. V.; Ozeryanskii, V. A. Synlett 2013, 24, 2515−2518.
(7) Notably, that two phenyl substituents in the solid 8 are in cis-
rather than in trans-orientation to each other, as one could expect.
Even more surprisingly, that this is accompanied by nearly perfect face-
to-face π-stacking, which is known to be less favorable as compared
with edge-to-face interaction due to a quadrupole nature of the benzene
ring: (a) Meyer, E. A.; Kastellano, R. K.; Diederich, F. Angew. Chem.,
Int. Ed. 2003, 42, 1210−1250. (b) Anslyn, E. V.; Dougherty, D. A.
Modern Physical Organic Chemistry; University Science Books:
Sausalito, CA, 2006; p 184. (c) Suponitsky, K. Yu.; Masunov, A. E.
J. Chem. Phys. 2013, 139, 094310 Apparently, the free energy loss in
the observed conformation is compensated via two mechanisms: more
profitable crystal packing and slightly unparallel disposition of both
phenyl groups. The angle between their planes is equal to 10.8°.
Because of the latter circumstance, only the C(1)−C(1′) distance
(3.19 Å) in the molecule is less than two half-thickness of the benzene
ring (3.4 Å), whilst the distances between centroids (3.85 Å) and C(4)
and C(4′) atoms (4.89 Å) of those rings are considerably larger.
(8) (a) For a review on the intermolecular oxidative cross-coupling
of arenes, see: Ashenhurst, J. A. Chem. Soc. Rev. 2010, 39, 540−548.
(b) Stuart, D. R.; Fagnou, K. Science 2007, 316, 1172−1175.
(c) Itahara, T.; Hashimoto, M.; Yumisashi, H. Synthesis 1984, 255−
256. (d) Grzybowski, M.; Skonieczny, K.; Butenschon, H.; Gryko, D.
T. Angew. Chem., Int. Ed. 2013, 52, 9900−9930. (e) Eberson, L.;
Hartshorn, M. P.; Persson, O. J. Chem. Soc., Perkin Trans. 2 1995,
409−416.
(9) (a) Bard, A. J.; Faulkner, L. R. Electrochemical Methods:
Fundamentals and Applications, 2nd ed.; J. Wiley: New York, 2001.
(b) Bard, A. J.; Parsons, R.; Jordan, J. Standard Potentials in Aqueous
Solutions; Marcel Dekker: New York, 1985.
(10) Gogoi, A.; Guin, S.; Rout, S. K.; Patel, B. Org. Lett. 2013, 15,
1802−1805.
(28) Pozharskii, A. F.; Ryabtsova, O. V.; Ozeryanskii, V. A.;
Degtyarev, A. V.; Kazheva, O. N.; Alexandrov, G. G.; Dyachenko, O.
A. J. Org. Chem. 2003, 68, 10109−10122.
(29) (a) APEX2; Bruker AXS Inc.: Madison, WI, 2009. (b) Sheldrick,
G. M. Acta Crystallogr. 2008, A64, 112−122.
(11) Xia, X.-F.; Zhang, L.-L.; Song, X.-R.; Niu, Y.-N.; Liu, X.-Y.;
Liang, Y.-M. Chem. Commun. 2013, 49, 1410−1412.
(12) (a) Gamage, S. N.; Morris, R. H.; Rettig, S. J.; Thackray, D. C.;
Thornburn, I. S.; James, B. R. Chem. Commun. 1987, 894−895.
(b) Hughes, R. P.; Kovacik, I.; Lindner, D.; Smith, J. M.; Willemsen,
S.; Zhang, D.; Guzei, I. A.; Rheingold, A. L. Organometallics 2001, 20,
3190−3197.
(13) Pozharskii, A. F.; Povalyakhina, M. A.; Degtyarev, A. V.;
Ryabtsova, O. V.; Ozeryanskii, V. A.; Dyablo, O. V.; Tkachuk, A. V.;
J
dx.doi.org/10.1021/jo502363t | J. Org. Chem. XXXX, XXX, XXX−XXX