Novel synthetic routes to thienocarbazoles via palladium or copper catalyzed amination or amidation of arylhalides and intramolecular cyclization

Article abstract of DOI:10.1016/S0040-4020(02)00904-3

Palladium or copper catalyzed aminations or amidations were performed to obtain diarylamines and diarylacetamides precursors of thienocarbazoles. The fact that an ortho-bromodiarylamine did not cyclize to the corresponding thienocarbazole under conditions known for carbazoles from ortho-halodiphenylamines, conducted us to a highly efficient method of palladium-catalyzed intramolecular cyclization with N-deprotection of ortho-halodiarylacetamides to thienocarbazoles. Other method of intramolecular cyclization of diarylamines based on the reoxidation of the Pd(0) formed by Cu(OAc)₂, avoiding the use of stoichiometric amounts of Pd(OAc)₂, gave thienocarbazoles in a moderate yield, including a ring A methoxylated compound. An attempt to combine palladium and copper catalyses in a 'one pot' reaction of amination and intramolecular cyclization gave as major product a N-benzo[b]thiophene substituted carbazole and the required thienocarbazole in low yield.

Full text of DOI:10.1016/S0040-4020(02)00904-3

Tetrahedron 58 (2002) 7943–7949
Novel synthetic routes to thienocarbazoles via palladium or
copper catalyzed amination or amidation of arylhalides and
intramolecular cyclization
b
Isabel C. F. R. Ferreira,a Maria-Joa˜o R. P. Queiroza and Gilbert Kirsch
,
*
a
´
Departamento de Quımica, Universidade do Minho, 4710-057 Braga, Portugal
´ ´
Laboratoire d’Ingenierie Moleculaire et Biochimie Pharmacologique, Universite de Metz, Faculte des Sciences,
b
´
Ile du Saulcy, 57045 Metz, France
Received 10 April 2002; revised 4 July 2002; accepted 23 July 2002
Abstract—Palladium or copper catalyzed aminations or amidations were performed to obtain diarylamines and diarylacetamides precursors
of thienocarbazoles. The fact that an ortho-bromodiarylamine did not cyclize to the corresponding thienocarbazole under conditions known
for carbazoles from ortho-halodiphenylamines, conducted us to a highly efficient method of palladium-catalyzed intramolecular cyclization
with N-deprotection of ortho-halodiarylacetamides to thienocarbazoles. Other method of intramolecular cyclization of diarylamines based
on the reoxidation of the Pd(0) formed by Cu(OAc)2, avoiding the use of stoichiometric amounts of Pd(OAc)2, gave thienocarbazoles in a
moderate yield, including a ring A methoxylated compound. An attempt to combine palladium and copper catalyses in a ‘one pot’ reaction of
amination and intramolecular cyclization gave as major product a N-benzo[b]thiophene substituted carbazole and the required
thienocarbazole in low yield. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
retention among different series of pharmacological agents.5
Thus thienocarbazoles I are bioisosteric analogues of the
known natural antitumoral pyridocarbazoles, ellipticines
IIa–c and olivacine IId, by substitution of the pyridine ring
by a thiophene and are being prepared to evaluate their
biological activity either as DNA intercalating compounds,
interacting or not with Topoisomerase II, or as radical
scavengers compounds.6 The achievement of less toxic
compounds than pyridocarbazoles is also a very important
goal (Fig. 1).
Due to their interesting biological activities carbazole
alkaloids constitute an important class of natural com-
pounds.1 Their isolation from different sources (terrestrial
plants, marine sources and streptomycetes) induced the
development of novel strategies of synthesis of structurally
unprecedent carbazole derivatives.2 Heteroannellated car-
bazoles are often of potential biological interest, mostly
based on their special affinity to DNA. Therefore this type of
compounds play a crucial role as potential leads for the
discovery of antitumor active drugs.3
The sulfur atom can provide interesting properties to this
type of molecules like the establishment of additional long
distance hydrogen bonds with the DNA chains or even
confer to the molecule interesting photochemical properties
for use as markers or in phototherapy applications.7,8 Some
preliminary studies of fluorescence of our new molecules
have already begun.
One of the standard methodologies that the medicinal
chemist can use as a rational approach to lead optimisation
is the bioisosteric replacement.4,5 Bioisosteres are sub-
stituents or groups, that do not necessarily have the same
size or volume, but have a similarity in chemical or physical
properties which could produce broadly similar biological
properties but being expected significant changes in
selectivity, toxicity and metabolic stability. However there
are many examples where bioisosteric replacements have
resulted in marked increases in potency as well as efficacy.
The use of classical isosteres as benzene, thiophene and
pyridine resulted in analogues with biological activity
The methyl groups and their position on the bioactive
pyridocarbazoles showed to be important for the antitumor
activity.9 Ring A substitution had also proven to be very
important for the activity being 9-methoxy and 9-hydroxy-
ellipticines IIb and c much more active than ellipticine
IIa.10
In recent years we have been interested in finding a ring B
method for the synthesis of substituted linear and angular
thienocarbazoles, in order to evaluate their structure-activity
relationship. The angularity or the linearity of the
Keywords: C–N coupling; copper; palladium; cyclization–deprotection;
thienocarbazoles.
*
Corresponding author. Tel.: þ351-253604378; fax: þ351-253678983;
e-mail: mjrpq@quimica.uminho.pt
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)00904-3

7944
I. C. F. R. Ferreira et al. / Tetrahedron 58 (2002) 7943–7949
Figure 1. Structure of thienocarbazoles I and pyridocarbazoles II.
molecules, the position of the methyl groups, together with
the introduction of ring A substituents, namely groups that
increase water solubility and/or activity, are important
features for biological devices.
2. Results and discussion
2.1. Synthesis of an ortho-bromodiarylamine under
Buchwald’s conditions and attempted intramolecular
cyclization
Some convergent ring B methods have already been
envisaged by us for the synthesis of several methylated
thienocarbazoles but the yields were either too low from
diarylamines11,12 or fair to moderate from nitrodiaryl
compounds.13
First it was decided to couple ortho-haloanilines 1a and 1b
with 6-bromobenzo[b]thiophene 217 or 2-bromo-iodoben-
zene with 6-aminobenzo[b]thiophene 3 under Buchwald’s
conditions15 to obtain ortho-halodiarylamines. Some
modifications such as the use of higher amounts of
Pd(OAc)2 (3 mol%) and BINAP (4 mol%) comparing to
those used in literature15 were needed in our case, may be
due to some complexation of the palladium by the sulfur
atom. Bromobenzo[b]thiophene 2 gave 20% yield of
o-bromodiarylamine 4 in the coupling with o-bromoaniline
(Scheme 1). When o-bromo-iodobenzene was reacted with
aminobenzothiophene 3, the yield was increased to 40% in
one third of the reaction time. The reactions were followed
by TLC and stopped when the formation of the product
seemed not to increase. In both cases the starting materials
were recovered. Under the same conditions no iodo
diarylamine 5 could be prepared from compounds 1b and
2, occurring decomposition of the aniline.
In this paper we report a high efficient palladium-catalyzed
cyclization with deprotection of ortho-halodiarylamides,
prepared by copper-catalyzed Goldberg coupling,11 to the
corresponding novel thieno[3,2-c]carbazole. This method
enables the synthesis of differently substituted hiherto linear
and angular thienocarbazoles from appropriated precursors.
Other method of cyclization based in the reoxidation of
Pd(0) by Cu(OAc)2,14 avoiding the use of a stoichiometric
amount of Pd(OAc)2 in the oxidative cyclization of
diarylamines gave also rise to the corresponding
thieno[3,2-c]carbazoles, including a ring A methoxylated,
in a moderate yield. The diarylamines were prepared either
by hydrolysis of diarylacetamides or by palladium catalyzed
amination of aryl halides under Buchwald’s conditions.15
An ortho-bromodiarylamine obtained using the latter
conditions, did not cyclize to the corresponding thieno-
carbazole under the Sakamoto’s cyclization conditions.16
The use of Pd2(dba)3 and DPPF, as described for the
preparation of ortho-halodiphenylamines precursors of
carbazole,16 was not effective in our case.
The amine 3 was obtained under drastic basic conditions18
from the corresponding steric hindered acetamide 6 which
was prepared by Beckmann rearrangement of the oxime of
6-acetylated compound19 (Scheme 2).
Scheme 1. Synthesis of o-bromodiarylamine 4 by palladium catalyzed
amination of arylhalides. (i) Pd(OAc)2 (3 mol%), BINAP (4 mol%),
tBuONa (1.4 equiv.), toluene 908C, under Ar.
Scheme 2. Synthesis of amide 6 and amine 3. (i) NH2OH·HCl, NaOH,
H2O/EtOH, 1 h 30 min reflux; (ii) dry ether, PCl5; (iii) removal of ether,
H2O, 1 h reflux; (iv) 10 equiv. NaOH, ethyleneglycol 1 h reflux.

I. C. F. R. Ferreira et al. / Tetrahedron 58 (2002) 7943–7949
7945
Attempts to perform the C–N coupling using the acetamide
6 under Buchwald’s conditions were unsuccessfull.
reaction was also performed using the much less expensive
1,2-dibromobenzene and 6 to give amide 8a (30% yield)
which was submitted to Sakamoto’s cyclization conditions
to afford thienocarbazole 7 in high yield.
When the ortho-bromodiarylamine 4 was submitted to
Sakamoto’s intramolecular cyclization conditions, that have
worked to obtain carbazoles from ortho-bromodiphenyl-
amine,16 thienocarbazole 7 did not form (Scheme 3).
Changing the base to NEt3 no thienocarbazole was obtained
either.
1
Amide 9 showed also hindered rotation in the H NMR
spectrum giving after deprotection, in the same conditions,
the corresponding amine 10 (Scheme 4). Amide 9 was
independently synthesized in high yield using the Goldberg
coupling reaction and a stoichiometric amount of Cu2O as
Scheme 5. Golbberg coupling to obtain amide 9.
outlined in Scheme 5.
Scheme 3. Attempted intramolecular cyclization by Sakamoto’s
conditions.
Deprotection of 8a and 8b with NaOH in a vigorous
refluxing ethylene glycol (silicone bath at 2208C), resulted
in the formation of 10 in 50% yield, together with the ortho-
bromodiarylamine 4 in 10% yield and the thienocarbazole 7
in 5% yield (Scheme 6). In another experiment increasing
the time of reflux in these conditions, the proportions of the
three products did not change. The formation of thieno-
carbazole 7 in these conditions indicates that a strong basic
medium submitted to a high temperature induces in a small
extent the cyclization reaction, possibly through a benzyne
intermediate.
2.2. Synthesis of ortho-halodiarylamides by Goldberg
coupling and intramolecular cyclization
The latter unsuccessful result led us to try our Goldberg
coupling conditions,11 reacting acetamide 6 with 2-bromo-
iodobenzene using 30 mol% of Cu2O, K2CO3, heating
without solvent at 1808C. A mixture of acetamides 8a and
8b (,40% yield) was obtained and it was impossible to
separate the compounds by chromatography, being charac-
terized by mass spectrometry. Along with amides 8a and
8b the dehalogenated amide 9 was isolated in 8% yield
(Scheme 4). The use of a stoichiometric amount of Cu2O did
not increase significantly the yield for amides 8a,b and the
yield for amide 9, as a by-product was not altered.
Scheme 6. Drastic basic conditions in ethylene glycol vigorous refluxing.
(i) NaOH (10 equiv.), ethylene glycol, 1 h vigorous reflux.
Due to the hindered rotation around the amide bond in 8, 1H
NMR spectra of these compounds reveal several sets of
signals and could not be used for structure assignment.
Thus, the amides 8 were converted to amines 4 and 5,
obtained also as a mixture (,75% yield), using drastic basic
conditions18 by refluxing gently ethylene glycol (silicone
bath at 2008C) (Scheme 4).
While Sakamoto’s cyclization conditions have only been
used with o-halodiarylamines, we decided to use them with
the mixture of o-haloacetamides 8a and 8b. Surprisingly
the thienocarbazole 7 was obtained with N-deprotection,
in quantitative yield (Scheme 7). A control experiment
performed in the same conditions without the palladium
catalyst, did not provide the thienocarbazole 7.
Iododiarylamine 5 was characterized by 1H NMR excluding
the proton signals of amine 4 already prepared by palladium
catalysis (Scheme 1). The structure of amine 5 was also
confirmed deprotecting the amide 8b independently
obtained (30% yield) from Goldberg coupling (stoich.
Cu2O) of 1,2-di-iodobenzene and amide 6. The same
2.3. Intramolecular cyclization of diarylamines using
Pd(OAc)2 and Cu(OAc)2
Cyclization of the dehalogenated acetamide 9 to thieno-
carbazole 7 did not occur when the same conditions were
Scheme 4. Synthesis and deprotection of diarylamides 8 and 9. (i) Cu2O (30 mol%), K2CO3, 1808C, 12 h; (ii) NaOH (10 equiv.), ethylene glycol 1 h reflux.

7946
I. C. F. R. Ferreira et al. / Tetrahedron 58 (2002) 7943–7949
2.4. ‘One pot’ procedure of C–N coupling and
intramolecular cyclization combining copper and
palladium catalyses
A one pot procedure attempt to obtain the thienocarbazole 7,
combining copper and palladium catalyses, reacting amine
3 with 2-bromo-iodobenzene in reflux of DMF for 10 h,
gave the N-benzo[b]thiophene substituted carbazole 13 (Mþ
341) as major product in 30% yield and the thienocarbazole
7 only in 10% yield (Scheme 10). Lowering the time of
heating the two products were obtained in the same
proportion.
Scheme 7. High efficient synthesis of thienocarbazole 7.
used, but the corresponding amine 10 gave the thienocarba-
zole 7 in 30% yield when treated with palladium acetate
(50 mol%) and copper acetate (3 equiv.) in acetic acid at
1208C, (Scheme 8). The role of Cu(OAc)2 is the reoxidation
of the Pd(0) formed after electrophilic attack of Pd(OAc)2
on the aromatic rings, avoiding the use of a stoichiometric
amount of this reagent.14
Scheme 10. One pot procedure to carbazole 13 and thienocarbazole 7.
(i) Cu2O (20 mol%), Pd(OAc)2 (10 mol%), K2CO3, DMF, 10 h reflux.
The formation of a dihalogenated intermediate 14 before
cyclization to carbazole 13, is in agreement with the
synthesis of N-methylsulfonylcarbazole from N-methyl-
sulfonyl-o,o0-dibromodiphenylamine.16
Scheme 8. Cyclization of diarylamine 10 to thienocarbazole 7. (i) Pd(OAc)2
(50 mol%), Cu(OAc)2 (3 equiv.), acetic acid 1208C, 7 h.
This also constitutes a valuable method for the synthesis of
thienocarbazoles that will be applied to diarylamines
prepared either by N-deprotection of diarylacetamides or
by palladium catalysed amination of arylhalides. As an
example, the methoxydiarylamine 11 was prepared in
60% yield, coupling the arylhalide 2 with 4-methoxy-
aniline under Buchwald’s conditions and then cyclized
to the corresponding methoxylated thieno[3,2-c]carbazole
12 in 30% yield in the same oxidative conditions
(Scheme 9).
The same one pot conditions were not successful when
applied to acetamide 6 and 2-bromo-iodobenzene, resulting
in extensive decomposition.
3. Conclusion
Novel synthetic routes to thienocarbazoles based on the
combination of metal assisted C–N coupling and intra-
molecular cyclizations were described. The target com-
pounds could act as DNA-binding agents which may be
used as biological or medical relevant probes or drugs.
These methods will be applied to the preparation of linear
and angular methylated thienocarbazoles substituted in ring
A by electron donating or withdrawing groups, for
evaluation of their biological activity and structure activity
relationship.
Scheme 9. Synthesis of diarylamine 11 by Buchwald coupling and
cyclization to thienocarbazole 12. (i) Pd(OAc)2 (3 mol%), BINAP
(4 mol%), 1.4 equiv. tBuONa toluene 908C, 24 h, under Ar; (ii) Pd(OAc)2
(50 mol%), Cu(OAc)2 (3 equiv.), acetic.acid 1208C, 7 h.

I. C. F. R. Ferreira et al. / Tetrahedron 58 (2002) 7943–7949
7947
4. Experimental
(3–4 ml), the arylhalide, the arylamine, t-BuONa
(1.4 equiv.), Pd(OAc)2 (3 mol%), racemic BINAP
(4 mol%) and the mixture was heated at 908C for several
hours. The reaction was followed by TLC. After cooling
water and ether were added. The organic phase was
separated, dried (MgSO4) and solvent removed to give an
oil which was submitted to chromatographic purification to
give the product and starting materials.
4.1. General
Melting points (8C) were determined in a Gallenkamp
apparatus and are uncorrected. IR spectra were recorded as
nujol mulls on a Perkin–Elmer 1600-FTIR spectropho-
tometer and wavenumbers are given in cm21. H and 13C
1
NMR spectra were recorded on a Varian Unity Plus (300
1
and 75.4 MHz respectively). H–1H spin–spin decoupling
4.2.1. 6-(2-Bromophenyl)amino-2,3,5-trimethylbenzo-
[b]thiophene (4). From arylhalide 217 (0.510 g,
2.00 mmol) and bromoaniline 1a (0.430 g, 2.50 mmol)
heating for 70 h and column chromagraphy using petroleum
ether, the arylhalide 2 was recovered in 57% yield as the less
polar product, compound 4 was obtained in 20% yield as a
white solid, mp 126–128. IR: 3377 (N–H). 1H NMR:
(CDCl3) 2.29 (3H, s, Me), 2.37 (3H, s, Me), 2.47 (3H, s,
Me), 5.96 (1H, s, N–H), 6.69 (1H, oct, J¼7.93, 7, 1.5 Hz,
H-40), 6.81 (1H, dd, J¼08.24, 1.5 Hz, H-60), 7.11 (1H, sept,
J¼8.24, 7, 1.5 Hz, H-05 ), 7.46 (1H, s, H-7), 7.52 (1H, dd,
J¼7.93, 1.5 Hz, H-3 ), 7.61 (1H, s, H-4). 13C NMR:
(CDCl3) 11.37 (CH3), 13.76 (CH3), 18.33 (CH3), 110.94
(C), 114.70, 116.45, 119.75, 122.84, 126.44, 128.19, 128.99
(C), 132.69, 133.02 (C), 136.11 (C), 136.38 (C), 138.31 (C),
142.90 (C). Anal. calcd for C17H16BrNS: C 58.97, H 4.66, N
4.04, S 9.26; found: C 58.71, H 4.77, N 3.91, S 9.31.
and DEPT u 458 were used. Chemical shifts are given in
ppm and coupling constants in Hz. The mass spectra were
obtained on a Unicam GC/MS 120 spectrometer or on a
Micromass Autospec 3F by an electronic impact (70 eV)
direct injection method. Elemental analysis was performed
on a LECO CHNS 932 elemental analyser.
The reactions were monitered by thin layer chromatography
(TLC). Column chromatography was performed on
Macherey-Nagel silica gel 230–400 mesh. Preparative
Layer Chromatography (PLC) was performed in 20£
20 cm2 plates Macherey-Nagel, Layer 2 mm SIL G-200
UV254. Petroleum ether refers to the boiling range 40–608C.
Ether refers to diethylether. When solvent gradient was
used, the increase of polarity was made gradually from
petroleum ether to mixtures of ether/ petroleum ether
increasing 10% of ether until the isolation of the product.
From 2-bromo-iodobenzene (0.185 g, 0.650 mmol) and
arylamine 3 (0.100 g, 0.500 mmol) heating for 21 h,
compound 4 was obtained in 40% yield after column
chromatography.
4.1.1. Synthesis of 6-acetamido-2,3,5-trimethylbenzo-
[b]thiophene (6). To hydroxylamine hydrochloride
(8.30 g, 119 mmol) in 13 ml of water, a solution of NaOH
(3.50 g, 87.5 mmol) in 12 ml of water was added with
external cooling. 6-Acetyl-2,3,5-trimethylbenzo[b]thio-
phene19 (12.5 g, 57.0 mmol) was then added in a sufficient
amount of ethanol to promote a homogeneous solution,
which was heated at reflux for 1 h 30 min. After cooling the
precipitate formed was filtered and dried at 608C. The
colourless solid obtained showed to be the corresponding
oxime (10.6 g, 80%) mp 191–193 (lit19 191, from
4.2.2. 6-(4-Methoxyphenyl)amino-2,3,5-trimethylbenzo-
[b]thiophene (11). From arylhalide 2 (0.240 g, 2.00 mmol)
and 4-methoxyaniline (0.500 g, 2.00 mmol), heating for
24 h, and using solvent gradient from petroleum ether to
30% ether/petroleum ether in the chromatographic purifi-
cation, compound 11 was obtained as a white solid (0.350 g,
60%), mp 127–129, which after crystallization from
ether/petroleum ether gave white crystals mp 130–132.
IR: 3394 (N–H). 1H NMR: ([D6]DMSO) 2.19 (3H, s, Me),
2.29 (3H, s, Me), 2.36 (3H, s, Me), 3.70 (3H, s, OMe),06.84
(2H, d, J¼9 Hz, H-30 and 50), 6.94 (2H, d, J¼9 Hz, H-2 and
60), 7.03 (1H, s, H-7), 7.28 (1H, s, H-4), 7.38 (1H, s, N–H).
13C NMR: ([D6]DMSO) 11.12 (CH3), 13.31 (CH3), 18.39
(CH3), 55.19 (OCH3), 108.52, 114.53, 120.18, 122.66,
125.01 (C), 126.15 (C), 129.59 (C), 134.54 (C), 135.83 (C),
137.62 (C), 140.73 (C), 153.54 (C). Anal. calcd for
C18H19NOS: C 72.70, H 6.44, N 4.71, S 10.78; found: C
73.00, H 6.27, N 4.78, S 10.70.
1
petroleum ether). H NMR: (CDCl3): 2.26 (3H, s, Me),
2.28 (3H, s, Me), 2.46 (3H, s, Me), 2.50 (3H, s, Me), 7.41
(1H, s, H-7), 7.60 (1H, s, H-4). MS: 233 (100, Mþ) 218 (78),
201 (82). To the oxime (6.00 g, 25.7 mmol) in dry ether
(150 ml) PCl5 (6.00 g, 28.6 mmol) was added and the
mixture was left stirring for 30 min. The ether was removed
and water was added. The aqueous mixture was heated at
reflux for 1 h 30 min and after cooling the colourless
precipitate formed was filtered, dried and showed to be the
amide 6 (4.00 g, 67%), mp 188–190. IR: 3237 (N–H), 1651
(CvO). 1H NMR: ([D6]DMSO) 2.08 (3H, s, Me), 2.23 (3H,
s, Me), 2.31 (3H, s, Me), 2.43 (3H, s, Me), 7.47 (1H, s, H-4),
7.85 (1H, s, H-7), 9.33 (1H, s, N–H). 13C NMR:
([D6]DMSO) 11.10 (CH3), 13.49 (CH3), 18.19 (CH3),
23.31 (NCOCH3), 117.92, 122.06, 126.28 (C), 128.32 (C),
132.88 (C), 133.06 (C), 134.65 (C), 138.07 (C), 168.35 (C).
MS: 233 (73, Mþ), 191 (100), 190 (53), 176 (41). Anal.
calcd for C13H15NOS:. C 66.92, H 6.48, N 6.00, S 13.74%;
found: C 66.79, H 6.23, N 5.94, S 13.72.
4.3. Goldberg coupling
4.3.1. 6-(2-Bromophenyl)acetamido-2,3,5-trimethylbenzo-
[b]thiophene (8a), 6-(2-iodophenyl)acetamido-2,3,5-tri-
methylbenzo[b]thiophene (8b) and 6-(phenyl)acetamido-
2,3,5-trimethylbenzo[b]thiophene (9). A mixture of the
acetamide 6 (1.00 g, 4.30 mmol), 2-bromo-iodobenzene
(1.80 g, 6.40 mmol), K2CO3 (0.600 g, 4.30 mmol), Cu2O
(0.190 g, 1.33 mmol) was heated at 1808C for 12 h. After
cooling chloroform was added and the mixture was filtered.
The filtrate was evaporated to give a brown oil which was
submitted to column chromatography using solvent gradient
4.2. General procedure for the synthesis of diarylamines
4 and 11 under Buchwald’s conditions
A dry Schlenk tube was charged under Ar with dry toluene

7948
I. C. F. R. Ferreira et al. / Tetrahedron 58 (2002) 7943–7949
1
from petroleum ether to 50% ether/petroleum ether. As the
less polar product, the dehalogenated amide 9 was obtained
as a white solid (0.100 g, 8%), mp 212–214. IR: 1675
ether, a white solid was obtained (,75%) which H NMR
spectrum showed to be a mixture of amines 4 and 5 being 5
in a slightly excess. 1H NMR: (CDCl3) 2.29 (2£3H, s, 2£Me
of 4 and 5), 2.37 (2£3H, s, 2£Me of 4 and 5), 2.47 (2£3H, s,
2£Me of 4 and 5), 5.82 (1H, s, N–H of 5), 5.96 (1H, s, N–H
of 4), 6.57 (1H, oct, J¼7.93,07, 1.5 Hz, H-40 of 5), 6.69 (1H,
oct, J¼7.93, 7, 1.5 Hz, H-4 of 4), 6.77 (1H, dd, J¼8.24,
1.5 Hz, H-60 of 5), 6.81 (1H, dd, J¼8.24, 1.5 Hz, H-60 of04),
7.11 (1H, sept partially obscured, J¼8.24, 7, 1.5 Hz, H-5 of
4), 7.14 (1H, sept partially obscured, J¼8.24, 7, 1.5 Hz, H-50
of 5), 7.46 (2£1H, s, 2£H-7 of 4 and 5), 7.52 (1H, dd,
J¼7.93, 1.5 Hz, H-30 of 4), 7.60 (1H, s, 2£H-4 of 4 and 5),
7.77 (1H, dd, J¼7.93, 1.5 Hz, H-30 of 5).
1
(CvO). H NMR spectra in CDCl3 or in [D6]DMSO at
several temperatures showed hindered rotation, being this
compound identified by the 1H NMR spectrum of the
corresponding amine 10 (see below). MS: 312 (8, Mþþ2),
311 (20, Mþþ1,), 310 (100, Mþ). Anal. calcd for
C19H19NOS: C 73.75, H 6.19, N 4.53, S 10.36; found: C
73.45, H 6.37, N 4.74, S 10.34.
Another fraction was isolated as a yellow light solid and
showed to be a mixture of 8a and 8b (0.650 g, ,40%), mp
148–150. H NMR spectra in CDCl3 or in [D6]DMSO at
1
several temperatures showed hindered rotation, being the
characterization done by the obtention of the corresponding
amines 4 and 5 (see below). MS: 436 (100, Mþ of 8b, 100),
390 (30, Mþ 81Br of 8a), 388 (30, Mþ 79Br of 8a).
The signals of compound 5 were later confirmed from the
deprotection of amide 8b, independently obtained from the
Goldberg coupling of 1,2-di-iodobenzene and amide 6 (see
Section 2.2).
4.3.2. 6-(Phenyl)acetamido-2,3,5-trimethylbenzo[b]thio-
phene (9). A mixture of the acetamide 6 (0.215 g,
0.920 mmol), 2-bromobenzene (0.174 g, 1.10 mmol),
K2CO3 (0.320 g, 2.30 mmol), Cu2O (0.130 g, 0.920 mmol)
was heated at 1808C for 8 h. After cooling chloroform was
added and the mixture was filtered. The filtrate was
evaporated to give a brown oil which was submitted to
column chromatography using solvent gradient from
petroleum ether to 10% ether/petroleum ether to give
amide 9 as a white solid (0.240 g, 84%) with identical
properties to those presented above.
4.4.3. 6-(Phenyl)amino-2,3,5-trimethylbenzo[b]thio-
phene (10). A mixture of amide 9 (0.100 g, 0.323 mmol)
and sodium hydroxide (0.130 g, 3.20 mmol) in ethylene
glycol (5 ml) was refluxed for 1 h. The oil obtained was
crystallized from ether/petroleum ether to give colourless
crystals (70.0 mg, 75%), mp 138–140. IR: 3384 (N–H). 1H
NMR: (CDCl3) 2.27 (3H, s, Me), 2.37 (3H, s, Me), 2.45 (3H,
s, Me), 5.45 (1H, s, N–H), 6.93 (2H, m, Ar-H), 7.26 (2H, m,
Ar-H), 7.41 (1H, s, H-7), 7.60 (1H, s, H-4). 13C NMR:
(CDCl3) 11.37 (CH3), 13.70 (CH3), 18.40 (CH3), 112.69,
116.90, 120.13, 122.72, 126.37 (C), 126.55 (C), 129.34,
131.79 (C), 136.54 (C), 136.79 (C), 137.87 (C), 144.62 (C).
Anal. calcd for C17H17NS: C 76.36, H 6.41, N 5.24, S 11.99;
found: C 76.55, H 6.21, N 4.91, S 11.71.
4.4. General procedure of deprotection of acetamides 6,
8a,b and 9
A solution of the acetamide and sodium hydroxide
(10 equiv.) in ethylene glycol was heated at reflux (silicone
bath at 2008C). After cooling, the mixture was poured into
iced water and after stirring, extracted with ether. The
organic phase was dried (MgSO4), filtered and the solvent
removed to give an oil which was submitted to chromato-
graphic purification or to crystallization.
4.5. Palladium-catalyzed cyclization with deprotection of
acetamides 8a,b
4.5.1. 2,3,5-Trimethyl-6H-thieno-[3,2-c]carbazole (7).
The mixture of amides 8a and 8b (0.150 g,
,0.350 mmol), Na2CO3 (57.0 mg, 0.540 mmol) and
Pd(OAc)2 (10 mol%) in refluxing DMF (5 ml) were heated
for 7 h. After cooling, ethyl acetate (20 ml) and water
(30 ml) were added. The phases were separated and the
aqueous phase was extracted with ethyl acetate (2£10 ml).
The organic phase was dried and filtered. Removal of the
solvent left a brown residue (0.130 g) which was submitted
to PLC (50% ether/petroleum ether) to afford the product as
a white solid (83.0 mg, quantitative yield), mp 215–217. IR:
3411 (N–H). 1H NMR: (CDCl3) 2.40 (3H, s, Me), 2.59 (3H,
s, Me), 2.69 (3H, s, Me), 7.35 (1H, td, J¼8, 1.2 Hz, H-9),
7.46 (1H, td, J¼8, 1.2 Hz, H-8), 7.48 (1H, s, H-4), 7.55 (1H,
broad d, J¼8 Hz, H-7), 8.14 (1H, s, N-H), 8.16 (1H, broad d,
J¼8 Hz, H-10). 13C NMR: (CDCl3) 11.74 (CH3), 13.75
(CH3), 17.16 (CH3), 110.66, 116.05 (C), 117.07 (C), 119.57
(C), 119.78, 121.56, 122.91 (C), 125.00, 126.98 (C), 128.54
(C), 129.72 (C), 135.02 (C), 136.50 (C), 138.97 (C). Anal.
calcd for C17H15NS: C 76.94, H 5.70, N 5.30, S 12.08;
found: C 76.69, H 5.45, N 5.16, S 12.01.
4.4.1. 6-Amino-2,3,5-trimethylbenzo[b]thiophene (3).
Acetamide
6 (0.700 g, 3.00 mmol), NaOH (1.20 g,
30.0 mmol) in ethylene glycol (5 ml) were heated for 3 h.
Chromatographic column using solvent gradient from
petroleum ether to 50% ether/petroleum ether, gave the
amine 3 as a colourless solid (0.357 g, 51%), mp 96–98. IR:
3423 (N–H). 1H NMR: (CDCl3) 2.22 (3H, s, Me), 2.30 (3H,
s, Me), 2.41 (3H, s, Me), 3.60 (1H, s, N-H), 7.04 (1H, s,
H-7), 7.26 (1H, s, H-4). 13C NMR: (CDCl3), 11.37 (CH3),
13.54 (CH3), 17.90 (CH3), 106.99, 120.70 (C), 122.35,
126.25 (C), 129.12 (C), 134.16 (C), 137.03 (C), 141.60 (C).
Anal. calcd for C11H13NS: C 69.07, H 6.85, N 7,32, S 16.76;
found: C 68.96, H 6.89, N 7.25, S 16.57.
4.4.2. 6-(2-Bromophenyl)amino-2,3,5-trimethylbenzo-
[b]thiophene (4) and 6-(2-iodophenyl)amino-2,3,5-tri-
methylbenzo[b]thiophene (5). A mixture of acetamides
8a and 8b (0.200 g, ,0.460 mmol) and sodium hydroxide
(0.184 g, 4.60 mmol) in ethylene glycol (5 ml) was refluxed
for 1 h. After column chromatography using solvent
gradient from petroleum ether to 30% ether/petroleum
4.6. General procedure for intramolecular cyclization of
non ortho-halogenated amines 10 and 11
A mixture of the diarylamine, Pd(OAc)2 (0.5 equiv.),

I. C. F. R. Ferreira et al. / Tetrahedron 58 (2002) 7943–7949
7949
Cu(OAc)2 (3 equiv.) and glacial acetic acid was heated at
Acknowledgements
1208C for 7 h. After cooling, ether (15 ml) and water
(10 ml) were added. The phases were separated and the
organic phase was washed with water, dried (MgSO4) and
filtered. Solvent removal gave an oil which was submitted to
PLC 50% ether/petroleum ether to afford the product.
Starting material was recovered.
Thanks are due to Foundation for the Science and
Technology-IBQF-Univ. Minho (Portugal) for financial
support, to the Research Incitment Programme of the
Calouste Gulbenkian Foundation (Portugal) and to Escola
´
Superior Agraria-Instituto Politecnico de Braganc¸a for
supporting in part Isabel C. F. R. Ferreira PhD.
´
4.6.1. Thienocarbazole (7). From amine 10 (0.170 g,
0.640 mmol), in glacial acetic acid (4 ml), thienocarbazole
7 was obtained as a white solid (50.0 mg, 30%), which
showed identical properties to those presented above.
References
4.6.2. 9-Methoxy-2,3,5-trimethyl-6H-thieno[3,2-c]carba-
zole (12). From amine 11 (0.140 mg, 0.470 mmol) in glacial
acetic acid (5 ml), thienocarbazole 12 was obtained as a
light yellow solid (40.0 mg, 30%), giving colourless crystals
from ether/petroleum ether crystallization mp 201–203. IR:
3378 (N–H). 1H NMR: (CDCl3) 2.40 (3H, s, Me), 2.59 (3H,
s, Me), 2.67 (3H, s, Me), 4.01 (3H, s, OMe), 7.10 (1H, dd,
J¼8.7, 2.4 Hz, H-8), 7.44 (2H, broad d, J¼8.7 Hz, H-7 and
H-4 obscured), 7.62 (1H, d, J¼2.4 Hz, H-10), 8.01 (1H, s,
N–H). 13C NMR: (CDCl3) 11.78 (CH3), 13.76 (CH3), 17.19
(CH3), 56.06 (OCH3), 104.14, 111.38, 114.30, 116.06 (C),
117.26 (C), 119.52, 123.27 (C), 127.03 (C), 128.42 (C),
129.44 (C), 133.88 (C), 134.65 (C), 137.31 (C), 154.13 (C).
MS: 295 (100, Mþ), 280 (37, Mþ215). HRMS C18H17NOS:
caldt. Mþ 295.10307; found 295.10339.
1. Chakraborty, P. Carbazole Alkaloids. The Alkaloids, Brossi,
A., Ed.; Academic: New York, 1993; Vol. 44, pp 257–364.
¨
2. Knolker, H.-J. Advances in Nitrogen Heterocycles, Moody,
C. J., Ed.; JAI: Greenwich, CT, 1995; Vol. 1, pp 173–204.
3. (a) Pierre, A.; Atassi, G.; Devissaguet, M.; Bisagni, E. Drugs
Future 1997, 22, 53–59. (b) Pierre, A.; Atassi, G.;
Devissaguet, M.; Bisagni, E. Drugs Future 1998, 23, 111.
4. Thornerm, C. W. Chem. Soc. Rev. 1979, 8, 563–580.
5. Patani, G. A.; LaVoie, E. J. Chem. Rev. 1996, 96, 3147–3176.
6. Ostrovidov, S.; Franck, P.; Joseph, D.; Martarello, L.; Kirsch,
G.; Belleville, F.; Nabet, P.; Dousset, B. J. Med. Chem. 2000,
43, 1762–1769.
7. Ihmels, H.; Faulhaber, K.; Wissel, K.; Bringmann, G.; Messer,
K.; Viola, G.; Vedaldi, D.; Eur, J. Org. Chem. 2001,
1157–1161.
4.7. Synthesis of N-[6-(2,3,5-trimethylbenzo[b]-
thiophene)]carbazole (13)
8. Dougherty, T. J.; Gomer, C. J.; Henderson, G.; Jori, D.;
Kessel, M.; Korbelik, M.; Moan, J.; Peng, Q. J. Natl Cancer
Res. 1998, 90, 889–905.
Amine 3 (0.270 g, 1.40 mmol), 2-bromo-iodobenzene
(0.410 g, 1.40 mmol), Pd(OAc)2 (10 mol%), Cu2O
(20 mol%), K2CO3 (2.5 equiv.) in dimethylformamide
(10 ml) were heated at reflux for 10 h 30 min. After cooling,
water and ether were added and the organic phase was
separated, washed with water, dried (MgSO4) and filtered.
Removal of solvent left a brown solid (0.240 g) which was
submitted to PLC, 50% ether/ petroleum ether giving as the
less polar product the N-substituted carbazole 13 as a light
yellow solid (0.140 g, 30%), mp 187–189. 1H NMR:
(CDCl3) 2.05 (3H, s, Me), 2.40 (3H, s, Me), 2.56 (3H, s,
Me), 7.06 (2H, broad d, J¼8 Hz, H-1 and 8), 7.30 (2H, td,
J¼8, 1.2 Hz, H-3 and 6), 7.40 (2H, td, J¼8, 1.2 Hz, H-2 and
7), 7.67 (1H, s, H-70), 7.76 (1H, s, H-40), 8.19 (2H, broad d,
J¼8 Hz, H-4 and 5). 13C NMR: (CDCl3) 11.50 (CH3), 13.99
(CH3), 17.82 (CH3), 109.80, 119.46, 120.28, 122.69, 122.94
(C), 123.14, 123.27, 125.86, 126.67 (C), 127.20, 132.15 (C),
133.14 (C), 135.67 (C), 136.27 (C), 141.50 (C), 141.56 (C).
MS: 341 (100, Mþ). HRMS C23H19NS: caldt. Mþ
341.123822, found 341.123963.
9. Rosseau-Richard, C.; Auclair, C.; Richard, C.; Martin, R. Free
Rad. Biol. Med. 1990, 8, 223–230.
10. Dalton, L. K.; Demerac, S.; Elmes, B. C.; Loder, J. W.; Swan,
J. M.; Teitei, T. Aust. J. Chem. 1967, 20, 2715–2727.
11. Peixoto, F. M. C.; Queiroz, M.-J. R. P.; Kirsch, G. J. Chem.
Res. (S) 1998, 172–173, J. Chem. Res. (M), 801–812.
12. Queiroz, M. -J. R. P.; Kirsch, G.; 16th International Congress
of Heterocyclic Chemistry, 1997, Bozeman; US Abstr.
POII-220.
13. Ferreira, I. C. F. R.; Queiroz, M.-J. R. P.; Kirsch, G.
J. Heterocycl. Chem. 2001, 38, 749–754.
¨
14. Knolker, H.-J.; Fhrohner, W. J. Chem. Soc., Perkin Trans. 1
1998, 173–175.
15. Wolfe, J.; Buchwald, S. L. J. Org. Chem. 2000, 65,
1144–1157.
16. Iwaki, T.; Yasuhara, A.; Sakamoto, T. J. Chem. Soc., Perkin
Trans. 1 1999, 1505–1510.
17. Cagniant, P.; Faller, P.; Cagniant, D. Bull. Soc. Chim. Fr.
1966, 3055–3065.
18. Freeman, H. S.; Butler, J. R. J. Org. Chem. 1978, 26,
4975–4978.
The thienocarbazole 7 was isolated as a white solid
(30.0 mg, 10%) with identical properties to those presented
above.
19. Cagniant, P.; Faller, P.; Cagniant, D. Bull. Soc. Chim. Fr.
1964, 1525–1534.