Synthesis of hexacyanotriphenylene and hexacyanotribenzylene

Article abstract of DOI:10.1055/s-2000-8207

The synthesis of hexacyanotriphenylene (1) and hexacyanotribenzylene (2) with triangular geometry is described. The new hexanitriles are considered to be precursors for oligomeric, polymeric or spherical phthalocyanine systems.

Full text of DOI:10.1055/s-2000-8207

PAPER
1705
Synthesis of Hexacyanotriphenylene and Hexacyanotribenzylene
Heiko Krempl, Robert Mattmer, Michael Hanack*
Institut für Organische Chemie, Lehrstuhl für Organische Chemie II, Universität Tübingen, Auf der Morgenstelle 18, 72076 Tübingen,
Germany
Fax +49(7071)29524; E-mail: hanack@uni-tuebingen.de
Received 20 April 2000; revised 1 August 2000
Dedicated to Professor R. Neidlein on the occasion of his 70th birthday
Abstract: The synthesis of hexacyanotriphenylene (1) and hexacy-
anotribenzylene (2) with triangular geometry is described. The new
hexanitriles are considered to be precursors for oligomeric, poly-
meric or spherical phthalocyanine systems.
Key words: phthalocyanines, triphenylenes, tribenzylenes, spheres,
cyanation
Phthalocyanines and metalphthalocyanines have been
widely used recently in materials science because they ex-
hibit several interesting properties, e.g. semi conductivity,
Figure 1
non-linear optical behavior and others. In addition, phtha-
locyanines are thermally and chemically quite stable,
statistical reaction of 1 and 2 with phthalonitrile would
lead to molecules with three phthalocyanine units (vide
supra). The self-condensation of 1 would lead to products
with polymer character.
which is another advantage for their practical applications
in the mentioned fields.1,2 The basic reaction for preparing
metalphthalocyanines is based on the reaction of phthalo-
dinitriles or diiminoisoindolines with a metal salt as
shown in Scheme 1.
Under the typical conditions for preparing phthalocya-
nines, it is in principle also possible to get branched struc-
tures similar to dendrimers. Under high dilution, the
system could build spheres via a three dimensional ring-
closure reaction. The possibility of formation of a spheri-
cal molecule built up from eight hexacyanotriphenylenes
(1) was patented by Oeste3 in 1991. The hypothetical
spherical molecule (Figure 2, left side) consists of six ph-
thalocyanine units, which are not totally planar.
Scheme 1
Figure 2
Starting with the appropriately modified phthalonitriles
(Figure 1), it should be also possible to prepare polymeric
phthalocyanines.3
Molecule 2 which is not planar due to its methylene bridg-
es would lead by self-condensation to the so-called phth-
alocuberenes as shown in Figure 2 (right side), which
have a cubic symmetry and would be built up from eight
molecules of 2.4 Calculations with force field (MM+)
We became interested in the hexacyano substituted sys-
tems 1 and 2 because the triangular geometry of these pre-
cursors should, in principle lead to 2- and 3-dimensional
structures by reacting them according to Scheme 1. The
Synthesis 2000, No. 12, 1705–1708 ISSN 0039-7881 © Thieme Stuttgart · New York

1706
H. Krempl et al.
PAPER
methods gave promising results regarding the stability of clotriveratrylene (6),13 which after demethylation with bo-
these types of spheres.5
ron tribromide,14 yields the catechol derivative 7 (Scheme
3).
In the following, we report on the synthesis of the hexacy-
ano substituted precursors 1 and 2.
To exchange the hydroxy with cyano groups, they were
first converted to nonafluorosulfonates by reacting with
nonaflyl fluoride (nonafluorobutanesulfonyl fluoride).
The reaction was carried out in DMF with triethylamine in
analogy to the preparation of other aromatic nonafluoro-
butanesulfonates.15 The obtained hexanonaflate 8 was
characterized by 1H and 13C NMR, MS, IR, and elemental
analyses. The methylene protons inside and outside the
nine-membered ring (designated as Hi, Ho) in compounds
2, 6 8 show different chemical shifts.
Benzene rings carrying one or two cyano groups are easy
to prepare based on several methods,6 the quite common
route being the replacement of the bromine atoms in the
aromatic ring by the cyano groups known as the Rosen-
mund von Braun reaction.7 Dehydration of oximes and
amides to nitriles also work well to obtain mono- or dini-
triles in the aromatic series. Substitution of triflates and
nonaflates with cyano groups are known for up to two
groups.8,9 However, to our knowledge there are no reports
on the preparation of aromatic compounds carrying more The replacement of all the nonaflyl with cyano groups in
than three nitrile groups by direct substitution reaction. 8 gave some problems, because the described palladium-
We report here two procedures, which we have standard- catalyzed cyanation with zinc cyanide8,9 for replacement
ized for the synthesis of hexacyanonitriles.
of triflates failed due to the deactivation of the palladium
catalyst Pd(PPh3)4 during the reaction. The replacement
reaction of nonaflates with cyano groups was therefore
optimized using 1,2-bis(nonaflyl)benzene (10) as the
model compound (Scheme 4) using a slightly modified
method.8 The bis-nonaflate 10 was obtained from catechol
(9) by reacting with nonaflyl fluoride.
Hexacyanotriphenylene (1) was synthesized starting from
triphenylene10 (3) (Scheme 2), which was selectively bro-
minated with formation of 2,3,6,7,10,11-hexabromotri-
phenylene (4), using iron as catalyst.11 The exchange of
bromine with copper cyanide in a Rosenmund von Braun
reaction7 leads only to a partial exchange of the bromine
atoms against the cyano groups. Under the usual condi- After several attempts, the optimum condition for the con-
tions, the obtained product is a mixture of mono- to pen- version was found by sequential addition of 15 to 20 equal
tacyanobromotriphenylenes.
portions of zinc cyanide to the reaction mixture over 6
hours. With this procedure, 1,2-bis(nonaflyl)benzene (10)
was converted into phthalonitrile (11) in 83% yield. Ap-
plying this method to hexakis(nonaflyl)tribenzylene (8),
pure 2 was obtained 32% yield.
After several trials, the synthesis of hexacyanotriphe-
nylene (1) from 4 was successfully achieved by reacting
with potassium cyanide in 2-chloronaphthalene (2-ClNp)
in the presence of dibenzo-18-crown-6 and tetrakis(tri-
phenylphosphine)palladium(0) as catalyst.12 Under these Preliminary studies of the reaction of 1 with an excess of
conditions, all six bromine atoms in 4 were exchanged tetraphenylphthalonitrile and nickel acetate in hexanol
with formation of 1 (41%), which was characterized by 1H gave products with phthalocyanine groups. The UV spec-
NMR, MS, IR, and elemental analyses. The hexanitrile 1 tra showed a Q band with a maximum at 748 nm and the
being planar and symmetric is very sparingly soluble in IR-spectra contained peaks typical of phthalocyanine sub-
common organic solvents and hence we encountered the structures. Because of the high molecular weight and the
problem of carrying out the self-condensation reaction.
insolubility of the reaction product, the isolation and char-
acterization was difficult. Also the self-condensation of 1
in refluxing dimethylaminoethanol containing DBU
yields phthalocyanines as seen by the Q-band with a max-
imum at 890 nm.
To improve further the solubility of the precursor with six
cyano groups, we chose the tribenzylene system 2 con-
taining a nine-membered ring with methylene bridges.
Because the aromatic systems are separated by methylene
bridges compound 2 has a non-planar structure. This In conclusion, we have synthesized for the first time two
would lead to a lower aggregation and increase the solu- different aromatic hexanitriles, which may be capable of
bility. A way for building up such nine-membered ring- forming spherical phthalocyanines of interest. Prelimi-
systems is the trimerization of veratryl alcohol (5) to cy- nary investigations on 1 have shown that phthalocyanines
Scheme 2
Synthesis 2000, No. 12, 1705–1708 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Synthesis of Hexacyanotriphenylene and Hexacyanotribenzylene
1707
Scheme 3
Scheme 4
1H NMR (250 MHz, CD3CN): = 9.36 (s, 6 H).
are formed. Further studies on the preparation of substitut-
ed and more soluble hexanitriles corresponding to 1 and 2
and their conversion into the corresponding phthalocya-
nines are in progress.
IR (KBr): = 3123, 3092, 3058, 2237, 1610, 1593, 1564, 1499,
1435, 1408, 1236, 1126, 1097, 912, 800, 692 cm .
MS (EI): m/z = 378.3 (M+, 7), 353.2 (17), 328.2 (5).
1
1H and 13C NMR spectra were recorded at r.t. on a Bruker AC250
(1H: 250.1 MHz, 13C: 62.9 MHz). Chemical shifts ( ) are given in
ppm. EI/FD mass spectra were performed on Finnigan ISQ 70
(70eV)/ Varian MAT 711A. IR spectra were recorded as KBr pel-
lets using a FT-IR spectrophotometer IFS 48 (Bruker). Elemental
analyses were performed on a Carlo Erba Elemental Analyser 1106.
All mps were determined with a heating rate of 2 K/min and are un-
corrected. The course of reaction and the purity of products were
followed by TLC analysis using Macherey-Nagel Alugram SIL/
UV254 plates. Hexahydroxytribenzylene (7) was prepared according
to literature procedure.12,13
2,3,7,8,12,13-Hexakis(nonaflyl)tribenzylene (8)
A solution of hexahydroxytribenzylene (7, 2.5 g, 6.83 mmol) in
DMF (30 mL) containing Et3N (12.4 g 123 mmol) was cooled to
0 °C. Then nonafluorobutanesulfonyl fluoride (37.1 g, 123 mmol)
was added dropwise. The mixture was stirred under N2 for 2 days,
then filtered to remove the Et3N•HF. The solid was washed with
CHCl3 (20 mL) and the washings were combined with the filtrate.
The combined organic layers were washed with NaOH (5%, 2 x 20
mL) and H2O (20 mL). After drying the organic phase (Na2SO4), the
solvent was evaporated. The residue was purified by column chro-
matography on silica gel (CHCl3) to give a colorless powder (9.3 g,
66%), mp 132 133 °C.
IR (KBr): = 3084, 2883, 1593, 1501, 1431, 1356, 1294, 1238,
1204, 1145, 1126, 1034, 1011, 943, 895, 845, 806, 750, 700, 684
cm .
1H NMR (250 MHz, CDCl3): = 3.88 (d, J = 13.0 Hz, 3Ho), 4.85 (d,
J = 13.0 Hz, 3Hi), 7.45 (s, 6H).
13C NMR (250 MHz, CDCl3): = 36.3 (Ca), 108 117 (C,C4F9),
124.9 (Cc), 139.2 (Cb/d), 139.8 (Cb/d).
2,3,6,7,10,11-Hexacyanotriphenylene (1)
To a solution of 2,3,6,7,10,11-hexabromotriphenylene (4, 1 g, 1.4
mmol) in 2-chloronaphthalene (25 mL) were added KCN (0.62 g,
9.5 mmol), dibenzo-18-crown-6 (1.44 g, 4 mmol) and Pd(PPh3)4
(2.3 g, 2 mmol) under N2. The mixture was heated for 20 h at
140 °C, cooled to 60 °C, diluted with a mixture of warm heptane/
toluene (1:1), and filtered at this temperature. The product was
washed with toluene, and small amounts of Et2O and MeOH. Ex-
cess of KCN and other salts were removed by washing with H2O
(1000 mL) and dried. To remove the Ph3P and the crown-ether, the
product was extracted for 24 h with Et2O in a Soxhlet-extractor. The
solvent was removed to give a green solid (220 mg, 41%).
1
MS (FD): m/z (%) = 2058.8 (M+, 100), 1776.5 (61).
Anal. calcd for C45H12F54S6O18 (2058.8): C, 26.25; H, 0.58; S, 9.34;
F, 49.83. Found: C, 26.20; H, 0.74; S, 9.2.67; F, 50.30.
Synthesis 2000, No. 12, 1705–1708 ISSN 0039-7881 © Thieme Stuttgart · New York

1708
H. Krempl et al.
PAPER
(3) Oeste, F. D. German Patent 4114536, 1991; Chem. Abstr.
2,3,7,8,12,13-Hexacyanotribenzylene (2)
A solution of 2,3,7,8,12,13-hexakisnonaflylcyclotribenzylene (8,
6.0 g, 2.9 mmol) in deoxygenated DMF (120 mL) was heated to
85 °C under N2. Pd(PPh3)4 (in total 2.00 g, 1.73 mmol) and Zn(CN)2
(in total 1.43 g, 12.2 mmol) were added in 15 to 20 equal portions
over 6 h. The mixture was cooled to r.t., diluted with EtOAc (500
mL), and washed with dil. NaHCO3 (2 x 200 mL), H2O, dil. HCl
(200 mL), and finally with H2O (200 mL). After drying (Na2SO4)
and concentrating in vacuo, the resulting solid was treated with hot
hexane and crystallized from acetone. This provided a colorless sol-
id material, which was recrystallized from MeCN (yield: 0.40 mg,
32%, mp > 350 °C).
1992, 117, 91076.
(4) Keesmann, T.; Große-Wilde, H.; Hanack, M.; Mattmer, R.;
Krempl, H. German Patent 19635058, 1997; Chem. Abstr.
1997, 126, 305594.
(5) Karfunkel, H. R.; Dressler, T.; Hirsch, A. J. Comput.-Aided.
Mol. Des. 1992, 6, 521.
(6) Patai, S. The Chemistry of the Cyano Group: 2; Wiley-
Interscience: Chichester, 1970; p 67.
(7) Ellis, G. P.; Romney-Alexander, T. M. Chem. Rev. 1987, 87,
779.
(8) Drechsler, U.; Hanack, M. Synlett 1998, 1207.
(9) Chambers, M. R. I.; Widdowson, D. A. J. Chem. Soc., Perkin
Trans. I 1989, 1365.
Selnick, H. G.; Smith, G. R.; Tebben, A. J. Synth. Commun.
1995, 25, 3255.
(10) Präfcke, K.; Kohne, B.; Singer, D. Angew. Chem., Int. Ed.
Engl. 1990, 29, 177.
IR (KBr): = 3113, 3067, 3041, 2235, 1666, 1603, 1555, 1495,
1483, 1444, 1387, 1356, 1304, 1248, 1204, 1146, 1034, 935, 924,
1
898, 831, 783, 731, 604 cm .
1H NMR (250 MHz, DMSO-d6): = 4.00 (d, J = 13.1 Hz, 3Ho),
5.20 (d, J = 13.1 Hz, 3Hi), 8.45 (s, 6H).
13C NMR (250 MHz, DMSO-d6): = 35.5 (Ca), 114.3 (Cd), 116.4
(Ce), 136.7 (Cc), 144.9 (Cb).
MS (EI): m/z (%) = 420.2 (M+, 7), 393.1 (22), 279.3 (5), 207.1 (24).
(11) Bartle, H. D.; Heaney, H.; Jones D. W.; Lees, P. Tetrahedron
1965, 21, 3289.
(12) Takagi, K.; Sasaki, K.; Sakakibara, Y. Bull. Chem. Soc. Jpn.
1991, 64, 1118.
Anal. calcd for C27H12N6 (420.44): C, 77.13; H, 2.88; N, 19.99.
Found: C, 76.04; H, 2.85; N, 19.44.
(13) Yamato, T.; Hideshima, C.; Surya Prakash, G. K.; Olah, G. A.
J. Org. Chem. 1991, 56, 2089.
(14) Lindsey, A. S. J. Chem. Soc. 1965, 1685.
(15) Subramanian, L. R.; Garcia Martinez, A.; Herrera Fernandez,
A.; Martinez Alvarez, R. Synthesis 1984, 481.
References
(1) Hanack, M.; Lang, M. Adv. Mater. 1994, 6, 819.
(2) Hanack, M.; Subramanian, L. R. In Handbook of Organic
Conductive Molecules and Polymers, Vol. 1; Nalwa, H. S.,
Ed.; Wiley: Chichester, 1997; pp 687 726.
Article Identifier:
1437-210X,E;2000,0,12,1705,1708,ftx,en;Z03300SS.pdf
Synthesis 2000, No. 12, 1705–1708 ISSN 0039-7881 © Thieme Stuttgart · New York