Tetraazaarenes by the ceramidonine approach

Article abstract of DOI:10.1039/c2nj21033a

The synthesis of extended heteroarenes via the acid-promoted dehydrocyclisation of arylamino-anthraquinones is examined as an approach to highly conjugated electron-acceptor materials and eventually to heterographene nanoribbons. Whilst the latter perspective is found to remain challenging, the former is exemplified by the synthesis of extended tetraazaheterocycles bearing solubilising alkyl substituents. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012.

Full text of DOI:10.1039/c2nj21033a

View Article Online / Journal Homepage / Table of Contents for this issue
New Journal of Chemistry
A journal for new directions in chemistry
www.rsc.org/njc
Volume 36 | Number 3 | March 2012 | Pages 513–844
ISSN 1144-0546
COVER ARTICLE
Bock et al.
Tetraazaarenes by the ceramidonine
approach
1144-0546(2012)36:3;1-H

Dynamic ArticVleiewLiAnrktiscle Online
NJC
Cite this: New J. Chem., 2012, 36, 570–574
www.rsc.org/njc
LETTER
Tetraazaarenes by the ceramidonine approachw
Parantap Sarkar, Ie-Rang Jeon, Fabien Durola and Harald Bock*
Received (in Montpellier, France) 12th December 2011, Accepted 6th January 2012
DOI: 10.1039/c2nj21033a
The synthesis of extended heteroarenes via the acid-promoted
dehydrocyclisation of arylamino-anthraquinones is examined as
an approach to highly conjugated electron-acceptor materials
and eventually to heterographene nanoribbons. Whilst the latter
perspective is found to remain challenging, the former is
exemplified by the synthesis of extended tetraazaheterocycles
bearing solubilising alkyl substituents.
appear promising. They are accessible from phenazine via
nitration3 and separation of the two obtained dinitrophena-
zines by selective crystallisation.4 As concerns bifunctional
anthraquinones, the 1,4- and 1,5-dihydroxy derivatives are cheap
and easily transformed into the corresponding triflates 5 and 6.5
Ceramidonine 1, the heart-shaped cyclodehydration product
of 1-phenylamino-anthraquinone 2, is efficiently obtained by
treatment of the latter in sulfuric acid at elevated temperature.1,2
We assumed thus that oligo- and polymers with multiple
1-arylaminoanthraquinone fragments are convenient precursors
of extended polycyclic heteroarenes. Such extended azahetero-
cyclic nanoribbons should exhibit pronouncedly electron-
deficient and thus electron-acceptor type electronic character
that may be of interest for organic electronics. They also are
more stable against oxidation than their homoaromatic counter-
parts, and may represent novel chelating ligands for transition
metal complexation, especially if further aromatic nitrogens
are introduced by the choice of appropriate azaarylamine
precursors. We deal here with the first stage on the path to
ceramidonine based condensed polymers with high nitrogen
content, i.e. the transformation of bifunctional azaaryl-diamines
and bifunctional anthraquinones into tetraazaarenes with two
heart-shaped pentacyclic fragments.
Replacement of the triflate leaving groups on 5 or 6 by the
amino functions of 3 or 4 followed by loss of water leads to
four hypothetical polymers; these ribbon-like polymers are
linear in the cases of poly35, poly36 and poly46, and helicoidal
in the case of poly45 (a hexameric macrocycle can also be
imagined). It may anecdotically be noted that, if the three
linear ribbons are classified in analogy to carbon nanotubes,6,7
all three different basic types of hexagon orientations (with respect
to any perpendicular cut through the ribbon) are present: zigzag
(poly35), oblique (poly36) and armchair (poly46).
To obtain soluble ceramidonine dimers with these bifunctional
bricks 3–6, we aimed at condensing them with monofunctional
counterparts bearing solubilising n-butyl chains, i.e. 4-butyl-
1-triflyloxy-anthraquinone 7 and 3-butyl-5-aminoquinoline 8.
4-Butyl-1-hydroxy-anthraquinone is available from the atypical
reaction of butylamine with 1-hydroxyanthraquinone in the
presence of CoCl2 reported by M. Matsuoka et al.,8 and we
aimed at obtaining 8 from 3-bromoquinoline via nitration,
alkylation and reduction.9,10 To our surprise, when trying to
obtain 3-butyl-5-nitroquinoline from 3-bromo-5-nitroquinoline
with excess tetrabutyltin and catalytic Pd(dppf)Cl2 [dppf =
1,10-bis(diphenylphosphino)ferrocene] in DMF under argon,
we isolated 3-butyl-5-aminoquinoline directly in 39% yield.
This points to DMF acting as a reducing agent and a hydrogen
source, in concordance with the recently reported reduction of
nitro substituents to amino groups during the palladium-catalysed
1-Arylaminoanthraquinones are most conveniently obtained
by a condensation of an aminoarene with a 1-substituted
anthraquinone such as 1-bromo- or 1-triflyloxyanthraquinone.
A ceramidonine-based approach to oligomers requires bifunc-
tional diaminoarenes and dianthraquinone bricks with substituents
on opposing sides. As conveniently stable diaminoarenes with ring
nitrogen atoms in vicinal positions to the amino groups for
possible chelation, the 1,6- and 1,9-diaminophenazines 3 and 4
´
Centre de Recherche Paul Pascal, CNRS UPR8641 and Universite de
Bordeaux, 115 av. Schweitzer, 33600 Pessac, France.
E-mail: bock@crpp-bordeaux.cnrs.fr; Fax: +33-556845600;
Tel: +33-556845673
w CCDC reference numbers 856958 and 856959. For crystallographic
data in CIF or other electronic format see DOI: 10.1039/c2nj21033a
c
570 New J. Chem., 2012, 36, 570–574
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
oxygen-free azaarenes 858 and 868 via their precursors
pre737, pre747, pre858 and pre868.
Before using dearly obtained 8 in condensation attempts
with bifunctional anthraquinones 5 or 6, we investigated the
condensation of commercial aminoarenes with mono- and
bifunctional anthraquinones 7, 6 and 5. We achieved the
coupling of singly triflate-substituted anthraquinone 7 with
8-aminoquinoline as monofunctional aminoarene to give 9
first in moderate yields (25%) with a combination of catalytic
Pd(dba)2 [dba = dibenzylideneacetone] and a dppf as a chelating
ligand in the presence of sodium tert-butoxide as strong base.14
The yields were significantly improved (57%) by addition of
lithium chloride in order to deactivate the liberated triflate in
the reaction medium which may poison the palladium catalyst,
and slow addition of the triflate-substituted anthraquinones
to the reaction in order to prevent base catalysed dissociation
to hydroxyanthraquinones. Cyclisation of quinolinylamino-
butyl-anthraquinone 9 in 70% H2SO4 (10 ml) at 130 1C for
8 min gave pyridino-ceramidonine 10 in about 50% yield.
Having thus in hand a feasible method of condensation to
obtain arylamino-substituted anthraquinones, we investigated
the feasibility of double cyclisations in the presence of solubilising
coupling of nitro-substituted 2-haloanilines with alkynes to
amino-substituted indoles in DMF.11,12
Previous reports about the isomeric outcome of the mono-
nitration of 3-bromoquinoline are ambiguous,9,13 leaving some
doubt about whether the main product is the 5- or the 8-nitro-
derivative, both being poorly distinguishable by NMR. It was
in this respect fortunate that we obtained single crystals of the
derived amine 8 suitable for crystallographic structure deter-
mination, confirming that nitration predominately happens in
position 5. 3-Butyl-5-aminoquinoline crystallises in the trigonal
%
R3 space group. Three molecules of the compound are arranged
in a triangle held together by hydrogen bonds between one of
the amino hydrogen atoms and the quinoline ring nitrogen of
adjacent molecules (dotted lines in Fig. 1). The triangular units
are arranged into a hexagonal geometry in the ab plane.
The four bifunctional and two monofunctional bricks 3–8
should yield the two quinones 737 and 747 and the two
Fig. 1 Crystal structure of 3-butyl-5-aminoquinoline.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 570–574 571

View Article Online
alkyl chains, using the condensation products 11 and 13 of
p-butylaniline with the bifunctional anthraquinones 6 and 5.
Whereas 1,5-bis(p-butylphenylamino)-anthraquinone 11
yielded the corresponding bisceramidonine 12 upon double
dehydration in hot sulfuric acid, the 1,4-isomer 13 did only
give, depending on the reaction time and temperature, either
the monodehydrated intermediate 14 or tar.15 We thus aban-
doned 5 as a building block and did not attempt the synthesis
of pre858 and 858. The three other cyclisation precursors
pre737, pre747 and pre868 were obtained by our above-
mentioned improved procedure with LiCl in 40%, 52% and
40% yield respectively. The three tetraazaarene targets 737,
747 and 868 were obtained from their precursors by dehydration
in 70% sulfuric acid at 170 1C in 53%, 41% and 64% as best
yields, respectively. The yields proved to be strongly dependent
on the reaction time, short times giving incomplete cyclisations
and long times giving substantial degradation. Optimised reaction
times were 8 minutes for 737 and 747, and 30 minutes for 868.
This synthetic approach appears to be practical for the
synthesis of relatively small extended azaarenes that may offer
potential for a variety of applications such as electron acceptor
behaviour in organic electronics or double metal chelation for
metal-to-metal spin–spin interactions. But it follows from the
moderate yields accompanied by relatively fast decomposition
under the necessary stringent reaction conditions that considerable
further innovative effort may be needed to render this approach
sufficiently efficient to lead to extended electron-deficient,
potentially metal-chelating, heterographene nanoribbons such
as poly36 and poly46.
Azahydrocarbon 868 on the other hand showed two clearly
discernable reversible reduction peaks at ꢀ0.85 V and ꢀ1.25 V
vs. ferrocene (inset in Fig. 2). These values are closer than
0.1 V to the values for C60, a prominent prototype acceptor
material in organic electronics, whose reduction peaks are
found at ꢀ0.92 V and ꢀ1.32 V vs. ferrocene under identical
conditions.17 This indicates that multiple azasubstitution in
moderately long-wavelength absorbing (i.e. moderately low
band-gap) arenes such as benzannellated perylenes leads to
pronouncedly electron-deficient azaarenes of suitable
reduction potentials for organic electronic applications.
We were able to obtain single crystals of 868, which allowed
us to assess the degree of non-planarity forced upon the arene
system by the two [4]helicenic bay regions (Fig. 3). 868 crystallises
%
in the centrosymmetric triclinic space group P1, with two
molecules per unit cell that are closely stacked by p–p inter-
actions with shortest carbon–carbon distances of 3.44 A
(dashed lines in Fig. 3). The planes of the two diazatetraphene
fragments (light and dark grey in Fig. 3), both of which are in
themselves roughly planar, are tilted with respect to each other
by an angle of 241 (when observed along the two central bonds
between the two fragments).
In summary, we have explored an approach towards extended
polycyclic azaarenes based on the acid-promoted cyclising
dehydration of arylamino-anthraquinone fragments to ceramid-
onine fragments. The potential of this approach towards
extended polycyclic azaarenes is found to be geometry-dependent
with respect to anthraquinone disubstitution: 1,5-disubstituted
anthraquinones yield the desired products, whilst 1,4-substituted
ones do not. On the other hand, the approach is geometry-
independent with respect to phenazine disubstitution, with
both 1,6- and 1,9-disubstituted phenazines being similarly
reactive. Whilst the yields observed do not allow an efficient
access to polymeric nanoribbon structures, they are adequate
for double cyclisations (of the order of 50%) leading to
twisted tetraazaarenes such as 868 whose reduction potentials
come close to those of C60, the archetypal, but weakly
absorbing and weakly soluble, acceptor material in organic
electronics.
The absorption spectra (Fig. 2) of the two tetraaza-diketones
737 and 747, both of which can be seen as doubly benzoylene-
substituted tetraaza-naphthopentaphenes, show very similar,
relatively unpronounced spectra typical of kata-annellated
arenes with relatively continuous absorption up to about
500 nm. In contrast, the tetraaza-dinaphthoperylene 868 exhibits
the typical absorption features of a rylene, i.e. a gap zone of low
absorption between a zone of intense short-wavelength absorption
(below 350 nm) and intense three-peaked long-wavelength
absorption (between 450 and 560 nm).16
To assess the electron-deficient character of the three materials,
we tried to perform cyclic voltammetry on all three, but the still
limited solubility in dichloromethane of the two diketones
impeded the obtention of meaningful curves.
Fig. 2 Absorption spectra of 737 (dashed), 747 (dotted) and 868
(continuous) in chloroform. Inset: cyclic voltammogram against ferro-
cene of 868 in 1 mM solution in dichloromethane in the presence of 0.1 M
tetrabutylammonium hexafluorophosphate (scan rate: 100 mV sꢀ1).
Fig. 3 Crystal structure of nonacyclic dialkyl-tetraazaarene 868.
c
572 New J. Chem., 2012, 36, 570–574
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
747: coupling between 1,9-diaminophenazine 4 (250 mg,
1.2 mmol) and 7 (990 mg, 2.4 mmol) leads to pre747 following
the above procedure with Pd(dba)2 (150 mg, 0.26 mmol, 10 mol%),
dppf (450 mg, 0.81 mmol), sodium tbutoxide (250 mg, 2.5 mmol)
and LiCl (200 mg, 4.75 mmol) in toluene (20 ml). Yield:
460 mg (52%) of deep red powder; 1H NMR (400 MHz, CDCl3):
q = 11.16 (broad s, 2H), 8.16 (d, 4H, J = 7 Hz), 8.12 (d, 2H,
J = 8 Hz), 7.73 (t, 4H, J = 7 Hz), 7.64 (t, 4H, J = 8 Hz), 7.56
(t, 2H, J = 8 Hz), 6.66 (d, 2H, J = 7 Hz), 2.73 (t, 4H, J = 8 Hz),
1.65 (m, 4H), 1.22 (m, 4H), 0.79 (t, 6H, J = 7 Hz) ppm. pre747
(440 mg, 0.598 mmol) was heated in 70% H2SO4 (15 ml) at 130 1C
for 8 min with vigorous stirring. The color changed from dark
yellow via deep green to deep red. The mixture was worked up as
described for 737. Yield: 170 mg (41%) of brown powder;
1H NMR (400 MHz, CDCl3): q = 8.92 (d, 2H, J = 9 Hz), 8.79
(d, 2H, J = 7 Hz), 8.51 (d, 2H, J = 7 Hz), 8.45 (d, 2H, J = 7 Hz),
8.10 (d, 2H, J = 9 Hz), 8.04 (d, 2H, J = 7 Hz), 7.78 (t, 2H,
J = 7 Hz), 7.62 (t, 2H, J = 7 Hz), 4.12 (t, 4H, J = 7 Hz), 2.12
(m, 4H), 1.54 (m, 4H), 1.00 (t, 6H, J = 7 Hz) ppm. 13C NMR
(100 MHz, CDCl3): q = 182.9; 152.0; 147.7; 145.6; 145.4;
142.7; 134.9; 134.1; 133.9; 132.7; 131.1; 130.8; 130.6; 130.3;
129.6; 128.8; 128.5; 126.3; 123.4; 122.8; 32.9; 31.4; 22.5;
14.5 ppm. UV-vis. (CHCl3): lmax (rel. intensity) = 385.0 nm (0.96),
439.0 nm (0.52), 464.0 nm (0.54); shoulder at 477.0 nm (0.49). MS
(m/z (%)): 699.2 (65, [M + H]+); 700.2 (50, [M + 2H]+); 721.2
(100, [M + Na]+); 722.2 (70, [M + Na + H]+).
Experimental
Syntheses of key compounds 8, 737, 747 and 868
8: nitration of 3-bromoquinoline (Alfa Aesar) and separation
of 3-bromo-5-nitroquinoline from small quantities of 3-bromo-8-
nitroquinoline by crystallization were performed following the
procedure of Crowley et al.,9 which is identical to the procedure
of Doherty et al.,13 with the sole exception that Doherty et al.
misassigned the main product, isolated by recrystallisation
from ethyl acetate, as being the 8-nitro isomer. 3-Bromo-5-
nitroquinoline (1.1 g, 4.3 mmol), PdCl2(dppf) (630 mg, 0.86 mmol)
and LiCl (360 mg, 8.6 mmol) were stirred in DMF (20 ml) for
10 min. Tetrabutyltin (6 g, 17.2 mmol) was added dropwise to
the reaction mixture, which then was refluxed under argon for
24 h. The solvent was removed under reduced pressure.
Column chromatography through silica in pentane elutes
unreacted tetrabutyltin, and 1 : 1 DCM : EtOAc elutes the
product. Yield: 340 mg (39%) of brown oil which formed
needle shaped crystals in the freezer that were adequate
for X-ray crystallography. 1H NMR (400 MHz, CDCl3):
q = 8.72 (s, 1H), 8.08 (s, 1H), 7.65 (d, 1H, J = 8 Hz), 7.49
(t, 1H, J = 8 Hz), 6.84 (d, 1H, J = 8 Hz), 4.27 (broad s, 2H),
2.81 (t, 2H, J = 8 Hz), 1.69 (m, 2H), 1.38 (m, 2H), 0.94 (t, 3H,
J = 7 Hz) ppm; MS (m/z (%)): 202.2 (100, [M + 2H]+), 201.2
(15, [M + H]+).
737: 1,6-diaminophenazine (3) (200 mg, 0.96 mmol),
Pd(dba)2 (150 mg, 0.26 mmol), dppf (450 mg, 0.81 mmol),
sodium tbutoxide (200 mg, 2.0 mmol), LiCl (160 mg, 3.8 mmol)
were stirred in toluene (10 ml) at 100 1C for 10 min under
argon. Then a suspension of 7 (790 mg, 1.92 mmol) in toluene
(10 ml) was added dropwise. The deep red mixture was
refluxed overnight under argon, cooled and chromatographed
through silica in chloroform to elute first a trace amount of
unreacted 7, followed by pre737 and finally unreacted amine
3. pre737 was recrystallised from butanol. Yield: 280 mg
(40%) of deep red crystals; 1H NMR (400 MHz, CDCl3):
q = 11.04 (broad s, 2H), 8.28 (m, 2H), 8.21 (m, 4H), 7.96
(d, 2H, J = 8 Hz), 7.79 (d, 2H, J = 8 Hz), 7.73 (m, 4H), 7.62
(t, 2H, J = 8 Hz), 6.59 (d, 2H, J = 8 Hz), 2.73 (t, 4H, J = 8 Hz),
1.63 (m, 4H), 1.21 (m, 4H), 0.78 (t, 6H, J = 7 Hz) ppm.
pre737 (100 mg, 0.143 mmol) was heated in 70% H2SO4
(10 ml) at 130 1C for 8 min with vigorous stirring. The color
of the mixture changed from brown to deep red. The hot
mixture was poured onto crushed ice and the precipitate was
filtered off, washed with 5% aqueous NaOH, dried under
reduced pressure, chromatographed through silica in 1 : 1
chloroform : EtOAc and crystallised from butanol. Yield:
50 mg (53%) of brown powder; 1H NMR (400 MHz, CDCl3):
q = 9.01 (d, 2H, J = 9 Hz), 8.78 (d, 2H, J = 7 Hz), 8.62 (d, 2H,
J = 8 Hz), 8.56 (d, 2H, J = 8 Hz), 8.49 (d, 2H, J = 9 Hz), 8.00
(d, 2H, J = 7 Hz), 7.89 (t, 2H, J = 8 Hz), 7.76 (t, 2H, J = 8 Hz),
3.87 (t, 4H, J = 8 Hz), 2.05 (m, 4H), 1.69 (m, 4H), 1.17 (t, 6H,
J = 7 Hz) ppm. 13C NMR (100 MHz, CDCl3): q = 182.7,
152.0, 147.2, 145.2, 144.8, 143.6, 134.8, 134.2, 133.9, 132.7,
131.1 (ꢁ2), 130.4, 129.9, 129.5, 129.4, 128.9, 126.3, 123.6,
122.9, 33.3, 32.0, 23.2, 14.2 ppm. UV-vis. (CHCl3): lmax
(rel. intensity) = 382.0 (1.00); 433.0 (0.47); 456.0 (0.46);
480.0 nm (0.30). MS (m/z (%)): 699.2 (100, [M + H]+).
868: 3-butyl-5-aminoquinoline 8 (500 mg, 2.5 mmol),
Pd(dba)2 (260 mg, 0.45 mmol, 20 mol%), dppf (810 mg,
t
1.46 mmol), sodium butoxide (260 mg, 2.7 mmol) and LiCl
(260 mg, 6.1 mmol) were stirred in toluene (10 ml) at 100 1C
for 10 min under argon. Then a suspension of 6 (480 mg,
0.96 mmol) in toluene (10 ml) was added dropwise. The deep
violet mixture was refluxed overnight and then separated by
column chromatography through silica. Chloroform elutes
traces of unreacted 6, and 1 : 1 chloroform : EtOAc elutes first
a red trace impurity (presumably a monocyclised intermediate)
and then pre868, which is recrystallised from butanol. Yield:
232 mg (40%) of deep red powder; 1H NMR (400 MHz,
CDCl3): q = 11.64 (broad s, 2H), 8.83 (s, 2H), 8.17 (s, 2H), 8.04
(d, 2H, J = 8 Hz), 7.80 (d, 2H, J = 8 Hz), 7.70 (t, 2H, J = 8 Hz),
7.60 (d, 2H, J = 8 Hz), 7.47 (t, 2H, J = 8 Hz), 7.13 (d, 2H,
J = 8 Hz), 2.79 (t, 4H, J = 8 Hz), 1.66 (m, 4H), 1.38 (m, 4H),
0.93 (t, 6H, J = 7 Hz) ppm. pre868 (150 mg, 0.248 mmol) was
heated in 70% sulfuric acid at 170 1C for 30 min with vigorous
stirring. The hot mixture was poured onto ice and the precipitate
was filtered off and washed with 5% aqueous NaOH, dried in
air, purified by column chromatography through silica in chloro-
form and recrystallised from methanol. Yield: 90 mg (64%) of
red powder. Single crystals for X-ray crystallography were
made by slow evaporation of solvent of a chloroform solution.
1H NMR (400 MHz, CDCl3): q = 9.55 (s, 2H), 8.92 (s, 2H), 8.72
(d, 2H, J = 9 Hz), 8.42 (m, 4H), 8.05 (d, 2H, J = 8 Hz), 7.99 (t,
2H, J = 8 Hz), 2.96 (t, 4H, J = 8 Hz), 1.84 (m, 4H), 1.51 (m, 4H),
1.02 (t, 6H, J = 7 Hz) ppm. 13C NMR (100 MHz, CDCl3): q =
152.8, 148.6, 148.2, 147.0, 137.1, 135.5, 132.6, 131.2, 130.1, 130.1,
129.5, 128.7, 127.2, 126.8, 123.5, 121.0, 33.8, 33.4, 22.6, 14.1 ppm.
UV-vis. (CHCl3): lmax (rel. intensity) = 338.0 nm (0.92); 469.0 nm
(0.21); 499.0 nm (0.48); 535.0 nm (0.66). MS (m/z (%)): 568.2
(40, [M]+); 569.2 (100, [M + H]+); 570.2 (80, [M + 2H]+).
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 570–574 573

View Article Online
10 M. Viaud, P. Jamoneau, L. Savelon and G. Guillaumet, Heterocycles,
1995, 41, 2799.
´
11 R. Sanz, V. Guilarte and A. Perez, Tetrahedron Lett., 2009,
50, 4423.
12 J. Muzart, Tetrahedron, 2009, 65, 8313–8323.
13 S. Doherty, E. G. Robbins, I. Pal, C. R. Newman, C. Hardacre,
D. Rooney and D. A. Mooney, Tetrahedron: Asymmetry, 2003,
14, 1517.
14 J. Louie, M. S. Driver, B. C. Hamann and J. F. Hartwig, J. Org.
Chem., 1997, 62, 1268.
15 We are puzzled by this difference in outcome. The expected double
dehydration product from 13 may be formally considered as
consisting of a benzene and a 5,8-diaza-pentaphene fragment
linked by two single bonds, and 5,8-diaza-pentaphene has been
reported to be exceedingly resistant to further oxidation even with
CrO3 in sulfuric acid (see G. M. Badger and R. Pettit, J. Chem.
Soc., 1952, 1874–1877). Fragility of the double cyclisation product
from 13 under the harsh reaction conditions therefore does not
seem to be an evident explanation for our failure to isolate it.
16 E. Clar, Polycyclic Hydrocarbons, Academic Press & Springer
Verlag, 1964.
Acknowledgements
This research was financed partially by CNRS and ERASMUS
MUNDUS (external co-operation window with Asia) grants.
We are grateful to Rodolphe Clerac for his continuous support.
´
Notes and references
1 A. H. Cook and W. Waddington, J. Chem. Soc., 1945, 402–405.
2 D. W. Rangnekar and S. V. Sunthankar, Indian J. Technol., 1974,
12, 548.
3 H. Otomasu, Chem. Pharm. Bull., 1958, 6, 77.
4 P. Huszthy, E. Samu, B. Vermes, G. Mezey-Vandor, M. Nogradi,
J. S. Bradshaw and R. M. Lzatt, Tetrahedron, 1999, 55, 1491–1504.
5 M. S. Taylor and T. M. Swager, Angew. Chem., Int. Ed., 2007,
46, 8480.
6 X. Lu, F. Tian, N. Wang and Q. Zhang, Org. Lett., 2002, 4, 4313.
7 R. Gletier, B. Esser and S. C. Kornmayer, Acc. Chem. Res., 2009,
42, 1108.
8 M. Matsuoka, T. Takei, I. Nakamura, K. Yoshida and T. Kitao,
Bull. Chem. Soc. Jpn., 1981, 54, 2225.
17 P.-M. Allemand, A. Koch, F. Wudl, Y. Rubin, F. Diederich,
M. M. Alvarez, S. J. Anz and R. L. Whetten, J. Am. Chem. Soc.,
1991, 113, 1050.
9 J. D. Crowley, I. M. Steele and B. Bosnich, Chem.–Eur. J., 2006,
12, 8935.
c
574 New J. Chem., 2012, 36, 570–574
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012