3,5,7,9-Substituted hexaazaacridines: Toward structures with nearly degenerate singlet-triplet energy separations

Article abstract of DOI:10.1021/jo8005123

(Chemical Equation Presented) Toward the goal of preparing stable, neutral open-shell systems, we synthesized a novel series of p-phenyl-substituted 3,5,7,9-hexaazaacridine and 3,5,7,9-hexaazaanthracene derivatives. The effects of substitution on the mole

Full text of DOI:10.1021/jo8005123

3,5,7,9-Substituted Hexaazaacridines: Toward Structures with
Nearly Degenerate Singlet-Triplet Energy Separations
Peter Langer,*,||,‡ Shadi Amiri,§ Anja Bodtke,¶ Nehad N. R. Saleh,|| Klaus Weisz,#
Helmar Go¨rls, and Peter R. Schreiner*,§
Institut fu¨r Chemie, UniVersita¨t Rostock, Albert-Einstein-Str. 3a, D-18059 Rostock, Germany,
Leibniz-Institut fu¨r Katalyse e. V. an der UniVersita¨t Rostock, Albert-Einstein-Str. 29a,
D-18059 Rostock, Germany, Institut fu¨r Organische Chemie, Justus-Liebig-UniVersita¨t Giessen,
Heinrich-Buff-Ring 58, D-35392 Giessen, Germany, Institut fu¨r Pharmazie, UniVersita¨t Greifswald,
Friedrich-Ludwig-Jahn-Str. 17, D-17487 Greifswald, Germany, Institut fu¨r Biochemie, UniVersita¨t
Greifswald, Felix-Hausdorff-Str. 18, D-17467 Greifswald, Germany, and Institut fu¨r Anorganische and
Analytische Chemie, UniVersita¨t Jena, August-Bebel-Str. 2, 07740 Jena, Germany
peter.langer@uni-rostock.de; prs@org.chemie.uni-giessen.de
ReceiVed March 24, 2008
Toward the goal of preparing stable, neutral open-shell systems, we synthesized a novel series of p-phenyl-
substituted 3,5,7,9-hexaazaacridine and 3,5,7,9-hexaazaanthracene derivatives. The effects of substitution
on the molecular electronic properties were probed both experimentally and computationally [B3LYP/
6-311G(d,p)//B3LYP/6-31G(d,p)]. While the experimentally prepared structures already have small (20-25
kcal/mol) singlet-triplet energy gaps, systems with even smaller (<9 kcal/mol) singlet-triplet energy
separations can be realized through systematic variation of the substituent numbers, types, and patterns.
Hexaazaanthracenes show generally smaller singlet-triplet energy gaps than hexaazaacridines. Nitrogen-
bonded σ- and π-acceptor substituents that cause positive inductive and mesomeric effects as well as
carbon-bonded σ-donor substituents make substituted hexaazaanthracenes promising candidates for purely
organic high-spin systems.
Introduction
coupling through the intermolecular SOMO interactions in the
crystalline state.3 Research on molecules with promising
electronic structures for potential applications in organic
magnetism as well as organic electronics includes open-shell
ground state systems or materials with small singlet-triplet
energy separations (∆EST), as these may be valuable magnetic
materials in their ground or optically excited states.
In recent years, particular attention has been paid to tetrathi-
afulvalenes (TTFs),4 carbenes,5 fullerenes,6 oligoacenes,7 and
related species. The synthetic availability of compounds with
Paramagnetic single molecules represent attractive building
blocks for the development of ferro- and ferrimagnetic materi-
als.1 The first purely organic ferromagnet was reported in 1991
by Kinoshita et al. who found that p-nitrophenyl nitronyl
nitroxide radicals exhibit intermolecular ferromagnetic interac-
tions in a ꢀ-phase crystal at 0.60 K.2 This structure was modified
by Pratt and Blundell et al. to increase the ferromagnetic
|| Universita¨t Rostock.
‡ Leibniz-Institut fu¨r Katalyse e. V. an der Universita¨t Rostock.
§ Justus-Liebig-Universita¨t Giessen.
(2) Tamura, M.; Nakazawa, Y.; Shiomi, D.; Nozawa, K.; Hosokoshi, Y.;
Ishikawa, M.; Takahashi, M.; Kinoshita, M. Chem. Phys. Lett. 1991, 186, 401.
(3) (a) Pratt, F. L.; Valladares, R.; Caulfield, J.; Deckers, I.; Singleton, J.;
Fisher, A. J.; Hayes, W.; Kurmoo, M.; Day, P.; Sugano, T. Synth. Met. 1993,
61, 171. (b) Blundell, S. J.; Marshall, I. M.; Lovett, B. W.; Pratt, F. L.; Hayes,
W.; Kurmoo, M.; Takagi, S.; Sugano, S. Hyperfine Interact. 2001, 133, 169.
¶ Institut fu¨r Pharmazie, Universita¨t Greifswald.
# Institut fu¨r Biochemie, Universita¨t Greifswald.
Universita¨t Jena.
(1) Aubin, S. M. J.; Wemple, M. W.; Adams, D. M.; Tsai, H.-L.; Christou,
G.; Hendrickson, D. N. J. Am. Chem. Soc. 1996, 118, 7746.
5048 J. Org. Chem. 2008, 73, 5048–5063
10.1021/jo8005123 CCC: $40.75 2008 American Chemical Society
Published on Web 05/22/2008

3,5,7,9-Substituted Hexaazaacridines
SCHEME 1. Thiazyl Radicals
such unusual magnetic and/or electronic properties is crucial
for their further development. Triplet carbenes have been
considered as candidates for organic magnetic materials,8 but
their typically high reactivities make them difficult to handle
and to isolate. A recent theoretical study by Woodcock et al.
including an empirical correction showed that the big aryl
substituents such as 1-naphthyl9 and 9-anthryl groups10 reduce
∆EST of carbenes effectively due to increased resonance
stabilization, and this reduction is more pronounced for the 9-
anthryl group due to its exceptional ability to resonance stabilize
the triplet state than singlet state.11
Heterocyclic thiazyl radicals with an unpaired electron belong
to a family of molecules with the potential for application as
organic ferromagnets.12,13 The most pursued candidates contain
five-membered rings, as shown in Scheme 1. The band gaps of
these molecules are very sensitive to their substitution pattern.14
The electronic and magnetic properties of the analogous six-
membered ring systems (Scheme 1) are less well-characterized
because of their tendency to dimerize (formation of S-S
bonds).15 Recently, Leitch et al.16 introduced the first examples
of π-stacked bisthiadiazinyl radicals (1) as a new class of neutral
heterocyclic π-radicals that are monomeric in the solid state;
these radicals derive their stabilization from resonance effects.17
Replacing sulfur with selenium in these species increases
intermolecular interactions and hence causes a marked increase
in conductivity.18 Another remarkable example is the bulk
ferromagnetism of bisthiaselenazolyl 1c with an ordering
temperature (Tc) of 12.3 K.12a
(4) (a) Clemente-Leon, M.; Coronado, E.; Galan-Mascaros, J. R.; Gomez-
Garcia, C. J.; Rovira, C.; Lauhkin, V. N. Synth. Met. 1999, 103, 2339. (b)
Perepichka, D. F.; Bryce, M. R.; Pearson, C.; Petty, M. C.; McInnes, E. J. L.;
Zhao, J. P. Angew. Chem., Int. Ed. 2003, 42, 4636. (c) Ribas, X.; Mas-Torrent,
M.; Perez-Benitez, A.; Dias, J. C.; Alves, H.; Lopes, E. B.; Henriques, R. T.;
Molins, E.; Santos, I. C.; Wurst, K.; Foury-Leylekian, P.; Almeida, M.; Veciana,
J.; Rovira, C. AdV. Funct. Mater. 2005, 15, 1023. (d) Nielson, M. B.; Lomholt,
C.; Becher, J. Chem. Soc. ReV. 2000, 29, 153. (e) Bryce, M. R. AdV. Mater.
1999, 11, 11.
(5) (a) Schreiner, P. R.; Reisenauer, H. P. ChemPhysChem 2006, 7, 880. (b)
Nemirowski, A.; Reisenauer, H. P.; Schreiner, P. R. Chem.—Eur. J. 2006, 12,
7411. (c) Schreiner, P. R. J. Am. Chem. Soc. 1998, 120, 4184. (d) Hirai, K.;
Tomioka, H. J. Am. Chem. Soc. 1999, 121, 10213. (e) Itoh, T.; Jinbo, Y.; Hirai,
K.; Tomioka, H. J. Am. Chem. Soc. 2005, 127, 1650.
(6) (a) Narymbetov, B.; Omerzu, A.; Kabanov, V. V.; Tokumoto, M.;
Kobayashi, H.; Mihailovic, D. Nature 2000, 407, 883. (b) Blinc, R.; Jeglic, P.;
Apih, T.; Seliger, J.; Arcon, D.; Omerzu, A. Phys. ReV. Lett. 2002, 88, 86402/1.
(c) Gotschy, B. Phys. ReV. B 1995, 52, 7378.
(7) (a) Langer, P.; Bodtke, A.; Saleh, N. N. R.; Go¨rls, H.; Schreiner, P. R.
Angew. Chem., Int. Ed. 2005, 44, 5255. (b) Anthony, J. E. Chem. ReV. 2006,
106, 5028. (c) Nunzi, F.; Ruiz, E.; Cano, J.; Alvarez, S. J. Phys. Chem. C 2007,
111, 618. (d) Sanchez-Carrera, R. S.; Coropceanu, V.; da Silva Filho, D. A.;
Friedlein, R.; Osikowicz, W.; Murdey, R.; Suess, C.; Salaneck, W. R.; Bredas,
J.-L. J. Phys. Chem. B 2006, 110, 18904. (e) Dos Santos, M. C Phys. ReV. B
2006, 74, 45426/1. (f) Hummer, K; Ambrosch-Draxl, C. Phys. ReV. B 2005, 72,
205205/1. (g) Bendikov, M.; Wudl, F.; Perepichka, D. F. Chem. ReV. 2004, 104,
4891. (h) Schleyer, P. v. R.; Manoharan, M.; Jiao, H.; Stahl, F. Org. Lett. 2001,
3, 3643.
To date, only a small number of isolated stable neutral single
organic molecules with at least a triplet ground state (S g 1) as
high-spin organics have been reported. Recently, the first
established triplet ground state aminyl diradical (2) with strong
ferromagnetic coupling was presented by Rajca et al.19
Linear polycyclic aromatic hydrocarbons (PAHs) composed
of fused benzene rings (oligoacenes) display small ∆EST values.
The higher members of the oligoacenes were predicted to have
rather small ∆EST values and potentially open-shell singlet
character.7d,20 These systems are computationally difficult
because the wave functions of oligoacenes larger than hexacene
are not stable.7g,h Likewise, the preparation of oligoacenes larger
(8) (a) Itoh, T.; Takada, A.; Hirai, K.; Tomioka, H. Org. Lett. 2005, 7, 811.
(b) Nicolaides, A.; Enyo, T.; Miura, D.; Tomioka, H. J. Am. Chem. Soc. 2001,
123, 2628. (c) Despagnet-Ayoub, E.; Sole, S.; Gornitzka, H.; Rozhenko, A. B.;
Schoeller, W. W.; Bourissou, D.; Bertrand, G. J. Am. Chem. Soc. 2003, 125,
124. (d) Woodcock, H. L.; Moran, D.; Schleyer, P. v. R.; Schaefer, H. F. J. Am.
Chem. Soc. 2001, 123, 4331.
(9) (a) Trindle, C. J. Org. Chem. 2003, 68, 9669. (b) Tomioka, H.; Iwamoto,
E.; Itakura, H.; Hirai, K. Nature 2001, 412, 626.
(10) (a) Yoshida, K.-I.; Iiba, E.; Nozaki, Y.; Hirai, K.; Takahashi, Y.;
Tomioka, H.; Lin, C.-T.; Gaspar, P. P. Bull. Chem. Soc. Jpn. 2004, 77, 1509.
(b) Trindle, C. J. Org. Chem. 2003, 68, 9669. (c) Tsuchiya, Y.; Matsuno, M.;
Itoh, T.; Hirai, K.; Tomioka, H. Bull. Chem. Soc. Jpn. 2005, 78, 2037. (d)
Yoshida, K.-I.; Iiba, E.; Nozaki, Y.; Hirai, K.; Takahashi, Y.; Tomioka, H.; Lin,
C.-T; Gaspar, P. P. Bull. Chem. Soc. Jpn. 2004, 77, 1509. (e) Takahashi, Y.;
Tomura, M.; Yoshida, K.-I.; Murata, S.; Tomioka, H. Angew. Chem., Int. Ed.
2000, 39, 3478.
(11) Woodcock, H. L.; Moran, D.; Brooks, B. R.; Schleyer, P. v. R.; Schaefer,
H. F. J. Am. Chem. Soc. 2007, 129, 3763.
(15) (a) Hayes, P. J.; Oakley, R. T.; Cordes, A. W.; Pennington, W. T. J. Am.
Chem. Soc. 1985, 107, 1346. (b) Zienkiewicz, J.; Kaszynski, P.; Young, V. G.,
Jr J. Org. Chem. 2004, 69, 7525.
(12) (a) Robertson, C. M.; Myles, D. J. T.; Leitch, A. A.; Reed, R. W.;
Dooley, B. M.; Frank, N. L.; Dube, P. A.; Thompson, L. K.; Oakley, R. T.
J. Am. Chem. Soc. 2007, 129, 12688. (b) Alberola, A.; Less, R. J.; Pask, C. M.;
Rawson, J. M.; Palacio, F.; Oliete, P.; Paulsen, C.; Yamaguchi, A.; Farley, R. D.;
Murphy, D. M. Angew. Chem., Int. Ed. 2003, 42, 4782.
(16) Leitch, A. A.; Oakley, R. T.; Reed, R. W.; Thompson, L. K. Inorg.
Chem. 2007, 46, 6261.
(17) (a) Beer, L.; Brusso, J. L.; Cordes, A. W.; Haddon, R. C.; Itkis, M. E.;
Kirschbaum, K.; MacGregor, D. S.; Oakley, R. T.; Pinkerton, A. A.; Reed, R. W.
J. Am. Chem. Soc. 2002, 124, 9498. (b) Beer, L.; Brusso, J. L.; Haddon, R. C.;
Itkis, M. E.; Oakley, R. T.; Reed, R. W.; Richardson, J. F.; Secco, R. A.; Yu, X.
Chem. Commun. 2005, 46, 5745. (c) Brusso, J. L.; Derakhshan, S.; Itkis, M. E.;
Kleinke, H.; Haddon, R. C.; Oakley, R. T; Reed, R. W.; Richardson, J. F.;
Robertson, C. M.; Thompson, L. K. Inorg. Chem. 2006, 45, 10958.
(18) Beer, L.; Brusso, J. L.; Haddon, R. C.; Itkis, M. E.; Kleinke, H.; Leitch,
A. A.; Oakley, R. T.; Reed, R. W.; Richardson, J. F.; Secco, R. A.; Yu, X.
J. Am. Chem. Soc. 2005, 127, 18159.
(13) (a) Awaga, K.; Tanaka, T.; Shirai, T.; Fujimori, M.; Suzuki, Y.;
Yoshikawa, H.; Fujita, W. Bull. Chem. Soc. Jpn. 2006, 79, 25. (b) Mito, M.;
Fujino, M.; Deguchi, H.; Takagi, S.; Fujita, W.; Awaga, K. Polyhedron 2005,
24, 2501.
(14) (a) Barclay, T. M.; Cordes, A. W.; George, N. A.; Haddon, R. C.; Itkis,
M. E.; Mashuta, M. S.; Oakley, R. T.; Patenaude, G. W.; Reed, R. W.;
Richardson, J. F.; Zhang, H. J. Am. Chem. Soc. 1998, 120, 352. (b) Cordes,
A. W.; Minge, J. R.; Oakley, R. T.; Reed, R. W.; Zhang, H. Can. J. Chem.
2001, 79, 1352. (c) Brusso, J. L.; Clements, O. P.; Haddon, R. C.; Itkis, M. E.;
Leitch, A. A.; Oakley, R. T.; Reed, R. W.; Richardson, J. F. J. Am. Chem. Soc.
2004, 126, 8256.
(19) Rajca, A.; Shiraishi, K.; Pink, M.; Rajca, S. J. Am. Chem. Soc. 2007,
129, 7232.
J. Org. Chem. Vol. 73, No. 13, 2008 5049

Langer et al.
SCHEME 2. TPH-Acridine in Zwitterionic Ground and
than hexacene has not yet been accomplished, due to their
unstable nature (facile oxidation and dimerization).20,21
Photoexcited Triplet States
Oligoazaacenes represent analogues of oligoacenes in which
some sp2 carbon moieties (CH) are substituted with nitrogen
atoms. Recently, Winkler and Houk investigated this class of
compounds computationally and found that oligoazaacenes
generally have ∆EST values smaller than those of their all-carbon
analogues.22 Recent progress in the chemistry of polyazaacenes
includes new strategies for their preparation.23
a triplet ground state. The ultimate goal would be the identifica-
tion of stable open-shell structures that may be candidates for
magnetic materials. Of course, we currently cannot predict the
ability of these structures with respect to the coupling of their
individual spins as this is a macroscopic property. In the
following, we present a combined experimental and computa-
tional study on the preparation and characterization of novel
TPH-acridines and TPH-anthracenes. Systematic computations
on the substituent effects help explain the electronic properties
of the prepared compounds and are used to predict ways to
minimize the ∆EST values.
3,5,7,9-Tetraphenylhexaazaanthracene (TPH-anthracene, 4a)
was found to have a singlet ground state resulting in a
24
zwitterionic structure with a ∆EST of about 19 kcal mol-1
;
4a undergoes photoinduced intramolecular electron transfer to
give a radical pair.25 The electronically related azaacene 5,7-
diphenyl-5H,12H-quinoxalino[2,3-b]phenazine (5a) also pos-
sesses a singlet ground state with a pronounced zwitterionic
structure.26 Tetraazapentacenes with long-chain alkyl substitu-
ents (5b and 5c) have molecular structures that promote a range
of possible packing configurations because of their zwitterionic
character.27
Results and Discussion
A. Synthesis. The synthesis of 3,5,7,9-tetraphenyl-
hexaazaacridines 3a-i was not straightforward, and the protocol
of all steps had to be thoroughly optimized for each individual
derivative (Scheme 3 and Table 1). The reaction of 2,6-
diaminopyridine with aromatic acid chlorides afforded the
bis(amides) 6a-f which were transformed into the bis(imidoyl)
dichlorides 7a-f. The condensation of phenyl- and tolylhydra-
zine with 7a-f gave the amidrazones 8a-i, which were
transformed into the 3,5,7,9-tetraphenylhexaazaacridines 3a-i
by oxidative cyclization (DBU, air, MeOH, or EtOH).
The (optimized) yields of acridines 3a-i are based on the
corresponding bis(amides) 6a-f and are in the range of 6-27%;
the best yields were obtained for acridines 3d, 3h, and 3i. During
the optimization, the high purity of 7a-f, the reaction time of
the final cyclization step, and the workup and purification for
We reported recently the synthesis and characterization of
3,5,7,9-tetraphenylhexaazaacridine (TPH-acridine, 3a), which
again exists as a zwitterion rather than as a diradical (Scheme
2).7a In agreement with the experimental findings, density
functional theory (DFT) computations characterize both 3a and
4a as highly zwitterionic structures with very low-lying triplet
states. These structures can be considered as rare cases of stable
16 π-electrons, weakly antiaromatic entities. The present paper
focuses on the synthesis and properties of substituted hexaazaacri-
dines and hexaazaanthracenes. These molecules combine an
unusual stability with low-lying electronic triplet states.
We systematically explore substituent effects on the way to
the development of molecules with small ∆EST values or even
SCHEME 3. Synthesis of
3,5,7,9-Tetraphenylhexaazaacridines 3a-ia
(20) (a) Anthony, J. E. Angew. Chem., Int. Ed. 2007, 46, 2. (b) Lowe, J. P.;
Kafafi, S. A.; LaFemina, J. P. J. Phys. Chem. 1986, 90, 6602. (c) Kivelson, S.;
Chapman, O. L. Phys. ReV. B 1983, 28, 7236. (d) Houk, N. K.; Lee, P. S.; Nendel,
M. J. Org. Chem. 2001, 66, 5517. (e) Kao, J.; Lilly, A. C., Jr. J. Am. Chem.
Soc. 1987, 109, 4149. (f) Ali, Md. E.; Datta, S. N. J. Phys. Chem. A 2006, 110,
13232.
(21) Payne, M. M.; Parkin, S. R.; Anthony, J. E. J. Am. Chem. Soc. 2005,
127, 8028.
(22) Winkler, M.; Houk, K. N. J. Am. Chem. Soc. 2007, 129, 1805.
(23) Sto¨ckner, F; Beckert, R.; Gleich, D.; Birckner, E; Gu¨nther, W; Go¨rls,
H.; Vaughan, G Eur. J. Org. Chem. 2007, 1237.
(24) Hutchison, K.; Srdanov, G.; Hicks, R.; Yu, H.; Wudl, F.; Strassner, T.;
Nendel, M.; Houk, K. N. J. Am. Chem. Soc. 1998, 120, 2989.
(25) (a) Hutchison, K. A.; Hasharoni, K.; Wudl, F.; Berg, A.; Shuali, Z.;
Levanon, H. J. Am. Chem. Soc. 1998, 120, 6362. (b) Braunstein, P.; Siri, O.;
Taquet, J.-p.; Rohmer, M.-M.; Benard, M.; Welter, R. J. Am. Chem. Soc. 2003,
125, 12246.
a Conditions: (i) Ar1COCl (2 equiv), NEt3 (2 equiv), CH2Cl2, 14 h, 0-20
°C; (ii) PCl5 (2.9 equiv.), toluene, 1.5 h, 20 °C, then 2 h, 110 °C, 88%; (iii)
Ar2NHNH2 (2.7 equiv), n-hexane, 14 h, 0-20 °C (in some cases, NEt3
was added, see Supporting Information); (iV) air, DBU (2.8 equiv), MeOH
(for 3h,i: EtOH), 4-14 h, 20 °C.
(26) Wudl, F.; Koutentis, P. A.; Weitz, A.; Ma, B.; Strassner, T.; Houk, K. N.;
Khan, S. I. Pure Appl. Chem. 1999, 71, 295.
(27) Choi, H.; Yang, X.; Mitchell, G. W.; Collier, C. P.; Wudl, F.; Heath,
J. R. J. Phys. Chem. B 2002, 106, 1833.
5050 J. Org. Chem. Vol. 73, No. 13, 2008

3,5,7,9-Substituted Hexaazaacridines
TABLE 1. Yields and Selected NMR Data of Acridines 3a-i
SCHEME 4. Resonance Structures of Protonated
TPH-Acridines (3a-i)H+
6
3
Ar1
Ar2
% 6a
% 3b
1H NMRd
a
a
b
c
c
d
d
e
f
a
b
c
d
e
f
g
h
i
Ph
Ph
Ph
48
48
62
54
54
62
62
70
49
23b, 30c
14
6
23
11
14
11
21
27
5.75
5.85
5.88
5.68
5.73
5.63
5.69
5.54
5.68
p-MeC6H4
Ph
Ph
p-MeC6H4
Ph
p-MeC6H4
Ph
Ph
p-MeC6H4
p-ClC6H4
p-ClC6H4
p-NO2C6H4
p-NO2C6H4
naphth-2-yl
pyrid-4-yl
and Figure 1). The solvatochromism of anthracene 4b is more
pronounced than that for acridine 3c (which bears the same
substituents). The absorptions of anthracene 4b show a signifi-
cant bathochromic shift compared to those of acridines 3a-i.
The influence of the substituents on the absorptions of acridines
3a-i is relatively small. In DMF, a bathochromic shift is
observed for acceptor-substituted acridines, whereas the absorp-
tions of 3a-c are in the range of λmax ) 547-548 nm, the
absorptions of nitro derivatives 3f and 3g are λmax ) 587 and
589 nm, respectively (which is, in fact, close to the absorption
of anthracene 4b). A similar trend is observed for CH2Cl2,
acetonitrile, and toluene but not for sulfuric, phosphoric, and
acetic acid. Actually, in sulfuric acid, a hypsochromic effect is
observed for 3f and 3g relative to 3a-c. The orange colored
DMF solution of acridine 3a (λmax ) 547 nm) is highly
fluorescent at Fλmax ) 585 nm. A shift of the fluorescence to
higher wavelengths is observed for nitro-substituted acridines
3f and 3g (Fλmax ) 610 and 600 nm, respectively) and for
anthracene 4b (Fλmax ) 615 nm).
Figures 2 and 3 show the UV-vis absorption curves of
acridines 3a and 3d-i and of anthracene 4b in DMF and
concentrated sulfuric acid, respectively. As discussed above, a
bathochromic effect is observed for DMF solutions of 3f and
4b (Figure 2). Besides this effect, the UV-vis absorption curves
are quite similar. The situation is different for solutions in
concentrated sulfuric acid (Figure 3). Besides the bathochromic
effect observed for 4b, pyridine-substituted acridine 3i displays
a significant hypsochromic shift. This might be explained by
competing protonation of the pyridine rather than the acridine
nitrogen atoms.
a Yields of isolated products. b Yields based on 6. c Yield based on
8a. d Central acridine proton or carbon (TFA-d1).
each individual acridine derivative (see Supporting Information)
proved to be important parameters. In addition, it also proved
to be very important to use the bis(imidoyl) dichlorides 7 and
bis(amidrazones) 8 immediately after their preparation. Imidoyl
chlorides 7 were not recrystallized but were carefully dried.
Bis(amidrazones) 8 could not be dried completely because they
tend to decompose upon drying and standing. The amounts of
the reagents for the cyclization step were calculated based on a
theoretical 100% yield of amidrazones 8. With regard to our
previous synthesis of 3a,7a we were able to optimize the yield
of the cyclization reaction from 10 to 30% and the overall yield
(based on bis(amide) 6a) from 8 to 23%.
For reasons of comparison, the novel zwitterionic TPH-
anthracene 4b (10% yield) was prepared from 1,3-diaminoben-
zene following the procedure as given for the synthesis of
acridines 3a-i. Anthracene 4b represents the carba-analogue of
acridine derivative 3c. Anthracene 4a (the analogue of 3a) was
prepared following the procedure reported by Wudl and co-
workers. In contrast to these authors, we have been able to
isolate pure 4a, despite much synthetic efforts, in only 7% yield.
Successive dilution of a solution of 3a in concentrated sulfuric
acid with water results in a change of color from deep green
over blue and purple to pink and orange; this process is
reversible (Figure 4 and Table 3). In fact, the addition of concd
H2SO4 to a mixture of H2SO4/water ) 1.5/2.55 resulted in a
blue to green color. A similar solvatochromism was observed
by dilution of a solution of 3a in concentrated sulfuric acid with
acetone (see Supporting Information). However, the solutions
tend to be unstable at 20 °C.
B. Absorption and Fluorescence. Acridines 3a-i are red
to deep red solids, atmosphere as well as thermally stable, and
poorly soluble in DMF, CH2Cl2, acetonitrile, and toluene.
Acridine 3a is also poorly soluble in concentrated nitric acid
(red solution), phosphoric acid (blue-purple solution), perchloric
acid (green solution), and acetic acid (orange solution). In
contrast, 3a is reasonably soluble in TFA (deep purple solution),
which can be explained by the formation of cation 3aH+. An
X-ray crystal structure analysis of 3aH+ revealed that the
protonation occurred at the central nitrogen atom and that the
cation resides in the form of resonance structure A (Scheme
4).7a
Excellent solubility and a strong bathochromic effect is
observed in concentrated sulfuric acid (λmax ) 662 nm, deep
green solution). The bathochromism can be explained by
complete protonation of the acridine, formation of aggregates
(vide infra), and the influence of the counterion. A solvato-
chromic effect, depending on the polarity of the solvent,28 is
observed for all acridines 3a-i and for anthracene 4b (Table 2
Concentration-dependent UV-vis spectra of 3a in CH2Cl2
did not show shifts of the maximum absorptions, and the
Lambert-Beer law is valid for the 292 and 546 nm peaks within
4% error (Figure 5). Hence, there is no indication for the
formation of aggregates. Cyclic voltammetry (CV) measure-
ments were performed for 3a in DMF containing Bu4NClO4 as
supporting electrolyte; however, because of the low solubility,
no peak current could be observed for this substrate.
C. Computational Analysis. To elucidate and understand
the electronic structures of the prepared molecules in detail, we
report in the following a systematic computational study on a
large number of substituted hexaazaacridines and hexaazaan-
thracene analogues.
(28) Reichardt, C. SolVents and SolVent Effects in Organic Chemistry, 3rd
ed.; Wiley-VCH: Weinheim, Germany, 2002.
J. Org. Chem. Vol. 73, No. 13, 2008 5051

Langer et al.
TABLE 2. Solvatochromism (λmax in nm) and Fluorescence of Acridines 3a-i
solvent
polarity28
3a
3b
3c
3d
3e
3f
3g
3h
3i
4b
H2SO4
H3PO4
HCO2H
HOAc
MeCN
DMF
CH2Cl2
toluene
fluorescence
λem (λex) DMF
662
540
a
535
a
547
a
a
a
a
a
685
654
542
542
542
548
547
553
585
(295)
667
a
a
538
a
552
a
681
537
a
539
546
553
551
556
583
(555)
631
538
a
534
a
587
a
a
649
537
a
536
584
589
582
583
600
(565)
663
a
a
546
a
557
a
558
529
a
521
556
562
559
564
583
(564)
719
655
585
568
587
593
591
599
615
(590)
0.728
0.648
0.460
0.404
0.309
0.099
538
544
548
544
535
585
(295)
a
a
a
580
(295)
579
(304)
610
(558)
a Not determined due to low solubility.
FIGURE 3. Absorptions of acridines 3a and 3d-i and of anthracene
4b in concentrated sulfuric acid.
FIGURE 1. Absorption (λmax) of acridines 3b, 3c, and 3e and of
anthracene 4b in solvents with different polarities.
FIGURE 4. Solvatochromic effect of the dilution of a solution of 3a
in concentrated sulfuric acid (the colors of the curves represent the
colors of the solutions).
FIGURE 2. Absorptions of acridines 3b-f, 3h, and 3i and of
anthracene 4b in DMF.
1. Computational Methods. Density functional theory (DFT)
using Becke’s three-parameter hybrid functional29 (U)B3LYP
with a 6-31G(d,p) basis set was employed to optimize all
geometries and energies of singlet (spin-restricted wave func-
tions) and triplet ground states (spin-unrestricted wave func-
tions). Although open-shell singlet species are potentially also
low-lying states,30 we have shown recently that DFT methods
are incapable of placing the energy of such states correctly, as
TABLE 3. Solvatochromism of 3a in Concentrated H2SO4 /Water
H2O/H2SO4
λmax (nm)
color
0.0/1.0
1.0./1.7
1.0/1.0
1.2/1.0
1.3/1.0
1.7/1.0
662
641
632
626
618
534
deep green
green-blue
blue-green
blue
purple
pink
exemplified with carbenes.31 Therefore, we refrain from such
computations; the singlet states would also be magnetically
(29) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Becke, A. D. Phys.
ReV. A 1988, 38, 3098.
(30) (a) Constantinides, C. P.; ; Koutentis, P. A.; ; Schatz, J. J. Am. Chem.
Soc. 2004, 126, 16232. (b) Bendikov, M.; ; Duong, H. M.; ; Starkey, K.; ; Houk,
K. N.; ; Carter, E. A.; ; Wudl, F. J. Am. Chem. Soc. 2004, 126, 7416. (c) Poater,
J.; ; Bofill, J. M.; ; Alemany, P.; ; Sola, M. J. Phys. Chem. A. 2005, 109, 10629.
(d) Chen, Z.; ; Jiang, D.; ; Lu, X.; ; Bettinger, H. F.; ; Dai, S.; ; Schleyer, P. v.
R.; ; Houk, K. N. Org. Lett. 2007, 9, 5449. Note that it is rather unlikely that
B3LYP/6-31G(d) gives a correct description of the open-shell singlet state. Figure
1 in this paper emphasizes this because the depicted energies are expected to
decrease with n.
(31) Schreiner, P. R.; Reisenauer, H. P.; Allen, W. D.; Sattelmeyer, K. W.
Org. Lett. 2004, 6, 1163.
(32) (a) Reddy, A. R.; Fridman-Marueli, G.; Bendikov, M. J. Org. Chem.
2007, 72, 51. (b) Reimers, J. R.; Hall, L. E.; Crossley, M. J.; Hush, N. S. J.
Phys. Chem. A 1999, 103, 4385. (c) Poater, J.; Bofill, J. M.; Alemany, P.; Sola,
M. J. Phys. Chem. A 2005, 109, 10629. (d) Portella, G.; Poater, J.; Bofill, J. M.;
Alemany, P.; Sola, M. J. Org. Chem. 2005, 70, 2509.
5052 J. Org. Chem. Vol. 73, No. 13, 2008

3,5,7,9-Substituted Hexaazaacridines
FIGURE 5. The concentration-dependent UV-vis spectra of 3a.
FIGURE 7. ORTEP drawing of 3g with atom numbering. Comparison
of experimental selected bond lengths (Å) and angles (°) (in parentheses)
with averaged computational values [B3LYP/6-31G(d,p)].
2. Structure and Energies. The structural parameters com-
puted for the minimum conformer of 3g compare well with the
X-ray crystal structure analyses (Figure 7). A high maximum
peak and a low minimum hole in the difference were observed
(0.792 and -0.696). These values are normal and correspond
to less than one electron. The crystals were extremely thin and
of low quality, resulting in a substandard data set; however,
the structure is sufficient to show connectivity and geometry
despite the high final R value. Although the effects of Peierls
distortions for oligoacenes20a,38 were indicated by Houk et al.
and for polyacetylenes as well as cyanodienes39 by Kato et al.,
the C-C bond lengths in the fused moieties show no Peierls
distortion in the molecules under consideration here, and this
emphasizes the notion of two delocalized nonalternating ribbons
joined by two longer C-C bonds (∼1.42 Å, Figure 7). Also,
the C4-N4 and C1-N7 bond lengths in the terminal rings (Figure
7) are short about the localized CN double bond and the
π-density cannot be delocalized into the outer rings.
Natural charge analyses confirm that TPH-acridine and TPH-
anthracene derivatives possess zwitterionic structures with
negative partial charges in the bottom ribbon and partial positive
charges in the top ribbon in their ground states (with an
orientation as depicted in Scheme 2). Such a charge distribution
was indicated before for tetraphenylhexaazaanthracene40 (4a)
by Wudl, Houk, and co-workers.25
The ∆EST values for the studied TPH-anthracenes are
generally smaller than those of the related TPH-acridines (Table
4). As the TPH-acridines generally display dipole moments
larger than those of TPH-anthracenes, these results might imply
that the greater charge separation or the higher polarity, the
larger the singlet-triplet energy splitting might be. To test this
correlation, we studied the relationship between ∆EST and the
total dipole moment for a wide range of molecules with various
substituents. The large permanent dipole moments (the zwitte-
rionic character) in this family of compounds make them
promising candidates for field-based switching (Table 4).41
FIGURE 6. Generic structure of para-substituted TPH-anthracenes (X
)CH) and TPH-acridines (X ) N).
inactive. Previous results show that closed-shell singlet and
triplet computations at this level of theory agree well with
experiment for acenes32 and azaacenes;24,25,33 a generic structure
is depicted in Figure 6. The computed vibrational frequencies
show that all stationary points are minima, and zero-point
vibrational energies (ZPVEs) are evaluated (no imaginary
frequencies); the B3LYP/6-31G(d,p) wave functions of the
parent systems 3a and 4a (both C2 symmetric) are stable. The
<S2> values of all singlet states were 0.00 and about 2.0 for
the triplets.
Atomic charges were estimated at the same level of theory
by using natural bond orbital analysis method (NBO 3.1). DFT
was also utilized for NMR and nucleus-independent chemical
shielding (NICS) computations with GIAO34-B3LYP/6-311G-
(d,p) on the optimized geometries. NICS(0) and NICS(1) were
determined as the negative value of the absolute isotropic
magnetic shielding35 with the SCGT36-DFT method (SCGT-
B3LYP/6-311G(d,p)) by locating a ghost atom at the ring centers
and at 1 Å above the plane of each ring. UV-vis spectra were
evaluated with time-dependent DFT (TD-B3LYP/6-311G(d,p)).
All computations employed the Gaussian03 program suite.37
(33) (a) Strassner, Th.; Weitz, A.; Rose, J.; Wudl, F.; Houk, K. N. Chem.
Phys. Lett. 2000, 321, 459. (b) Winkler, M.; Houk, K. N. J. Am. Chem. Soc.
2007, 129, 1805.
(38) (a) Norton, J. E.; Houk, K. N. J. Am. Chem. Soc. 2005, 127, 4162. (b)
Bendikov, M.; Duong, H. M.; Starkey, K.; Houk, K. N.; Carter, E. A.; Wudl, F.
J. Am. Chem. Soc. 2004, 126, 7416.
(34) Wolinski, K.; Hinton, J. F.; Pulay, P. J. Am. Chem. Soc. 1990, 112,
8251.
(35) (a) Ruiz-Morales, Y. J. Phys. Chem. A 2004, 108, 10873. (b) Chen, Z.;
Wannere, C. S.; Corminboeuf, C.; Puchta, R.; Schleyer, P. v. R. Chem. ReV.
2005, 105, 3842.
(39) (a) Kato, T.; Yamabe, T. J. Phys. Chem. A 2004, 108, 11223. (b) Kato,
T.; Yamabe, T. J. Phys. Chem. B 2005, 109, 10620. (c) Kato, T.; Yamabe, T. J.
Phys. Chem. A 2005, 109, 4804.
(36) Keith, T. A.; Bader, R. F. W. Chem. Phys. Lett. 1993, 210, 223.
(37) Frisch, M. J. Gaussian03. The full reference can be found in the
Supporting Information.
(40) Pierron, P. Ann. Chim. Phys. 1908, 15, 269.
(41) Gibbs, H. M. Optical Bistability: Controlling Light with Light; Academic
Press: New York, 1985.
J. Org. Chem. Vol. 73, No. 13, 2008 5053

Langer et al.
TABLE 4. B3LYP/6-31G(d,p) Computed Zero-Point Vibrational Energies [kcal mol–1], Singlet-Triplet Energy Separations [kcal mol–1],
Difference of Total Dipole Moments (∆µge) [D] between Ground (µg) and Excited States (µe), HOMO-LUMO Energy Gaps [kcal mol-1] and
B3LYP/6-311G(d,p)//B3LYP/6-31G(d,p) λmax [nm] of 3a-i, 4a, and 4b
ZPVE
a
No.
X
Ar1
Ar2
singlet
triplet
∆EST
µg
µe
∆µge
∆
HL
λmax
3a
3b
3c
3d
3e
3f
3g
3h
3i
N
N
N
N
N
N
N
N
N
C
C
Ph
Ph
Ph
286.1
320.4
320.4
273.9
308.2
289.2
323.5
344.8
271.2
293.5
327.8
292.5
319.4
292.6
272.9
307.2
288.1
322.5
343.7
270.2
292.6
292.5
-20.8
-21.0
-20.7
-21.0
-21.1
-21.5
-21.6
-21.2
-21.0
-16.3
-16.2
8.19
9.28
7.53
10.68
11.83
14.69
15.91
7.92
7.07
8.21
6.37
1.12
1.07
1.16
1.14
1.09
1.17
1.11
1.25
1.14
1.45
1.49
57.7
57.7
57.4
57.8
57.9
58.4
58.3
57.7
58.1
52.8
52.7
513
513
518
514
514
552
558
523
510
561
562
p-CH3C6H4
Ph
Ph
p-CH3C6H4
Ph
p-CH3C6H4
Ph
Ph
Ph
Ph
p-CH3C6H4
p-ClC6H4
p-ClC6H4
p-NO2C6H4
p-NO2C6H4
naphth-2-yl
pyrid-4-yl
Ph
9.54
10.74
13.52
14.80
6.67
11.80
5.04
12.94
6.49
5.91
4a
4b
p-CH3C6H4
4.42
a Negative value indicates a ground state singlet.
FIGURE 8. Similarity of the B3LYP/6-31G(d,p) optimized geometries
of 3h in the singlet and triplet states.
All structures show slightly twisted topologies and generally
very small geometric differences between their singlet and triplet
states (see 3h with the most pronounced distortion, Figure 8).
The rather similar geometries between singlet and triplet
optimized structures lend credibility to the small ∆EST values.
The twisting angles ꢁ(N1C14C8N7) of the fused moiety in the
TPH-acridines are more pronounced than those in TPH-
anthracenes (see Supporting Information), which is in line with
the larger ∆EST and the greater zwitterionic character. The
distortion must be due to the steric demand of the aryl groups
in the peri-positions. The structures of the hexaazaanthracene
and hexaazaacridine moieties (without the aryl groups) are both
coplanar with ꢁ(N1C14C8N7) ) 0.0° at the same level of theory.
In phenyl-substituted structures, the aryl rings are connected
by C-C (Ar1, Figure 6) and C-N bonds (Ar2) to two different
positions of the fused moiety.
FIGURE 9. The B3LYP/6-31G(d,p) frontier molecular orbitals for
TPH-acridine (3a), TPH-anthracene (4a), and tetraphenylanthracene
(TP-anthracene).
aromatic sextet is not enough to provide a sufficient degree of
stability.42 Repeated benzannulation increases their reactivity
and produces the observed bathochromic shifts. Clar was the
first to notice that the more intensively colored a benzenoid
hydrocarbon is, the less stable it generally is. He ascribed this
relationship to a bathochromic shift of the p-band as the stability
decreases.42 A similar bathochromism was observed for sub-
stituted twistacenes;38a,43 linear polyacenes are characterized by
remarkably large shifts in their UV p-bands.
For a better understanding of the UV-vis spectra, we
computed these with TD-DFT. Low transition energies (long
As shown in Figure 8, the steric repulsion of the Ar2’s in the
peri-positions and also the nature of the C-N bonds twist these
rings (Ar2’s) noticeably more away from the plane of the fused
moiety than those joined through CC bonds (Ar1’s). The
variation of the dihedral angles ꢁ(C16C15C14N13) for the Ar1
and ꢁ(C18C17N2C3) for the Ar2 in the singlet and triplet states
provides an explanation for the different degrees of π-delocal-
ization in these overall nonplanar molecules. The X-ray structure
of 3g shows no twisted topology in the fused moiety, which
we attribute to packing effects.
As discussed above, pronounced bathochromism, depending
on the polarity of the solvent, was observed for all TPH-acridines
3a-i and for TPH-anthracenes 4a,b. Following Clar’s ideas,42
only one aromatic sextet can exist in linear oligoacenes and by
extension also in their aza analogues. With an infinite number
of rings, these systems become cyclic polyenes in which one
wavelengths) imply a small HOMO-LUMO energy gap (∆HL
)
that frequently results in high reactivity;44 computed λmax values
as well as ∆HL are presented in Table 4. The bathochromic shifts
are a result of the degree of p/π-delocalization that is also
affected by the substituents. The nature of the para-substituents
(A1 and A2 in Figure 6) and the position (Ar1 and Ar2) both
affect λmax. The λmax values of the TPH-anthracenes are generally
larger than those of TPH-acridines. For example, the λmax of
unsubstituted TPH-anthracene (4a) is 561 nm and of unsubsti-
tuted TPH-acridine (3a) it is 513 nm.
The TPH-acridine and TPH-anthracene frontier orbitals,
which display significant antibonding character, are considerably
(43) (a) Duong, H. M.; Bendikov, M.; Steiger, D.; Zhang, Q.; Sonmez, G.;
Yamada, J.; Wudl, F. Org. Lett. 2003, 5, 4433. (b) Notario, R.; Abboud, J.-L.
M J. Phys. Chem. A 1998, 102, 5290.
(42) Clar, E. The Aromatic Sextet; Wiley: London, 1972.
(44) Wiberg, K. B. J. Org. Chem. 1997, 62, 5720.
5054 J. Org. Chem. Vol. 73, No. 13, 2008

3,5,7,9-Substituted Hexaazaacridines
TABLE 5. B3LYP/6-311G(d,p)//B3LYP/6-31G(d,p) Total NICS Values Related to the Ring Center for 3a-i, 4a, and 4b (in ppm)
NICS
3a
3b
3c
3d
3e
3f
3g
3h
3i
4a
4b
0.0 (inner ring)
0.0 (outer rings)
1.0 (inner ring)
1.0 (outer rings)
+0.2
+1.6
-2.9
-1.8
+0.2
+1.6
-2.9
-1.8
+0.1
+1.4
-3.1
-1.7
+0.1
+1.3
-3.1
-1.6
+0.1
+1.5
-3.2
-1.7
+0.4
+2.0
-2.7
-1.5
+0.5
+1.9
-2.8
-1.5
-0.9
+0.5
-3.5
-1.4
+0.2
+1.7
-2.9
-1.7
-2.1
+2.9
-4.7
-0.2
-2.3
+3.1
-4.7
-0.8
different from those of tetraphenylanthracene (Figure 9). In
TPH-acridines and TPH-anthracenes, the frontier orbitals have
large amplitudes on the fused moieties that are centered on the
nitrogen atoms and no coefficients on the two terminal carbon
atoms bonded to phenyl substituents. The HOMO has large
coefficients on the electronegative nitrogen atoms; this dimin-
ishes the antibonding character in nitrogen-rich TPH-acridines
relative to their TPH-anthracene analogues. As a consequence,
the experimental and computed HOMO-LUMO gaps for TPH-
magnitudes of the NICS values are rather small. Indeed, the
values are close to those of fully saturated 1,2,4-triazinane,
which is considered nonaromatic (Figure 10). The aromatic
counterpart 1,2,4-triazine is clearly defined as such by its large
negative NICS values. A comparison with the ∆EST values
shows that there is no obvious relationship between these two
quantities. If anything, a small ∆EST in a cyclic delocalized
structure manifests itself by a small absolute NICS(0) value.
4. Substituent Effects. For a comprehensive comparison, we
studied a number of substituted hexaazaacridine and hexaazaan-
thracene derivatives to determine the role of the substituents
on the ∆EST values. To maximize the effect, we first investigated
structures with substituents bonded directly to the N or C atoms
of the fused moieties without using the phenyl groups. The
structural configurations are shown in Figure 12.
A large variety of substituents Y (Figure 12) with a broad
spectrum of mesomeric and inductive effects were employed.
Halogens showed dissociation of the N-Hal bond (weakly
bound complexes form) in their optimized triplet state structures
in configurations i and ii, and these results are not discussed.
Our data demonstrate that the electronic properties, the number,
and the position of the substituents have a large effect on the
∆EST values in these molecules (Figure 13).
acridines are larger, which is in agreement with a smaller λmax
.
Carbon-bonded electron-donating and electron-withdrawing
groups (position A1, Figure 6) cause smaller and larger ∆HL
and ∆EST, respectively (Table 4). Hence, from a conceptual
viewpoint, TPH-anthracene derivatives are more attractive
targets to achieve stable organic structures with small ∆EST
values.
3. Magnetic Properties. 13C chemical shifts and nucleus-
independent chemical shielding (NICS) DFT computations
provide additional information to understand the electronic
structures of the present systems. The GIAO-computed 13C and
1H NMR chemical shifts generally agree well with experimental
data (see Supporting Information). The average deviation of the
computed δ 1H and δ 13C shifts relative to experimental data is
about 1.4 and 8.0 ppm, due to the neglect of solvent effects in
the NMR computations.
Delocalization effects were evaluated by computing NICS(0)
as well as NICS(1) values (Table 5). As references, the NICS
values for benzene, 1,2,4-triazine, cyclobutadiene, and the
related aliphatic ring structures shown in Figure 10 were
computed at the same level of theory.
The NICS data of the synthesized structures were compared
to reference structures depicted in Figure 11. As noted before,
the NICS values at the ring centers indicate weak antiaromaticity
for 3a in all core rings, but the antiaromatic character vanishes
quickly above the planes of the rings (NICS ) 1.0 Å).
Nevertheless, ring currents play a very minor role as the absolute
FIGURE 11. B3LYP/6-311G(d,p)//B3LYP/6-31G(d,p) computed
NICS(0) and NICS(1) values for TPH-acridine (3a), TPH-anthracene
(4a), and TP-anthracene.
FIGURE 10. B3LYP/6-311G(d,p)//B3LYP/6-31G(d,p) NICS(0) and
NICS(1) values (ppm) and ∆EST (kcal/mol) for benzene, cyclobutadiene
(the NICS values are for singlet state), and related structures as reference
systems. The ∆EST is not available for cyclohexane because of
dissociation upon optimization in the triplet state.
FIGURE 12. Model systems to study the substituent effects on the
electronic structures of hexaazaacridine as well as hexaazaanthracene
derivatives.
J. Org. Chem. Vol. 73, No. 13, 2008 5055

Langer et al.
FIGURE 13. Relative changes of B3LYP/6-31G(d,p) ∆EST values (kcal mol-1) depending on the substituents for hexaazaacridine (X ) N, left)
and hexaazaanthracene (X ) CH, right) derivatives.
FIGURE 14. Comparing the trends in changes of the total dipole moments due to the two different structural configurations i (left) and iv
(right) for substituted hexaazaanthracene derivatives in ground and first excited states (top). The lines should make the trends of the ∆µge
values more visible. The correlation between ∆EST values (kcal mol-1) and ∆µge’s (D) for these systems (bottom). All data were obtained
at B3LYP/6-31G(d,p).
Configurations i and ii with N-substitution in the peri-
position(s) are more susceptible to electronic effects than iii
and iv, which are substituted at carbon (Figures 12 and 13).
This can be understood on the basis that the carbons bearing
the substituents are incapable of proper π-conjugation (see MO
renderings in Figure 9). The nitrogen atom at the peri-position
is part of the conjugated system of the fused ring moieties, and
as a result, substituents at this position show relatively large
effects (configurations i and ii). The number of the substituents
is also important: The changes of ∆EST values in i are larger
than those in ii. These effects are attenuated with the C-bonded
substituents in iii and iv.
As noted above, the ∆EST values of hexaazaanthracenes (X
) CH) are smaller than those of the analogous hexaazaacridines
(X ) N). The main reason for this finding is the lower polarity
(smaller zwitterionic character) of the hexaazaanthracenes (the
total dipole moment for parent hexazaanthracene is 6.49 and
8.19 D for hexaazaacridine in their ground states). A comparison
between the computed total dipole moments and the ∆EST values
confirms that generally as the total dipole moment of a molecule
decreases the ∆EST increases (Figures 14; however, compare
discussion above). The total dipole moments decrease more
rapidly in system i than in ii. The same correlation is found for
iii and iv. For all four studied configurations, the changes of
5056 J. Org. Chem. Vol. 73, No. 13, 2008

3,5,7,9-Substituted Hexaazaacridines
TABLE 6. Rationalization of Inductive Substituent Effect on the ∆EST (kcal mol-1) B3LYP/6-31G(d,p) for Hexaazaanthracene Derivatives
(configuration i)a [ISE ) Inductive Substituent Effect ((I)]
Y
-Lib
-OCH3
-CH3
-NMe2
-NH2
-H
-SiH3
-CF3
∆EST
-19.7
-0.011
-I
-CCl3
-11.1
+0.012
+I
-19.5
-0.002
-I
-CBr3
-11.2
+0.013
+I
-19.0
-0.003
-I
-C(O)CH3
-10.2
+0.015
+I
-18.7
-0.001
-I
-CN
-11.3
+0.025
+I
-18.7
+0.002
+I
-C(O)H
-9.3
+0.017
+I
-18.8
(0.000
-
-NO2
-9.3
+0.019
+I
-14.9
+0.004
+I
-C(O)Cl
-7.0
+0.023
+I
-14.9
+0.011
+I
-BH2
-5.2
+0.052
+I
∆qC (NBO)c
11
ISE
Y
∆EST
∆qC (NBO)c
11
ISE
b
a Negative value indicates that the singlet state is more stable. Li displays η2-coordination upon optimization. c Relative local natural charge on
carbon atom meso-C11 in Figure 6 ISE relative to -H; the natural charge for -H is -0.315.
the dipole moments are more irregular than the changes of the
∆EST values.
The same strategy was followed for meso-C11 (X ) CH) in
structure iv with different carbon-bonded substituents. A mild
inverse effect with some irregularities is observed. The most
positive ISE is obtained by using -Li (+0.20), -NMe2
(+0.015), and -NH2 (+0.013) and the most negative ISE
observed for -NO2 (-0.018), -C(O)Cl (-0.016), and -CBr3
(-0.014) relative to -H. The corresponding table is presented
in the Supporting Information.
II. Mesomeric Substituent Effects. Mesomeric substituent
effects ((M) mainly concern resonance (delocalization) effects.
Aromaticity and hence cyclic delocalization indices typically
rely on three common (but not exclusive) features: increased
thermodynamic stability, decreased bond length alternation, and
characteristic magnetic properties.32 In the present work, we
analyze the ( M effects of the substituents and their influence
on the π-electron system using the NICS aromaticity criteria.
First of all, there is no direct correlation between the
computed NICS(1) values and the ∆EST values. For the more
useful NICSzz(1) criteria, introduced by Schleyer et al. as a better
measure for the π-delocalization,48 a correlation is more apparent
(Figure 15).
In structure i, the computed NICSzz(1) values show that the
outer rings of all structures are antiaromatic with NICSzz(1)
between +7 and +18. For comparison, the NICSzz(1) values
for benzene are -29.11 and +57.81 for cyclobutadiene (C2h,
singlet) at the same level of theory. Comparison of the values
shows a decrease of the NICSzz(1) in both the inner rings and
the outer rings with growing ∆EST values; this effect is more
apparent in the inner ring than in the outer rings. As a result, in
structure i, π-acceptors (-M substituents) increase the aromatic
character of the inner ring while π-donors (+M substituents)
destabilize the system by increasing the electron density of the
outer rings.
Owing to the MO pattern of configuration iv, no regular
relationship could be identified between ∆EST values and the
NICSzz(1) values. As a result, carbon-bonded substituents have
only very little effect on the ∆EST values. This is in agreement
with our analyses of the experimentally prepared derivatives
above.
The influence of the relative magnitudes of the dipole
moments of the ground and first excited states (∆µge ) µg -
µe) on the singlet-triplet energy gaps was also studied. For
nitrogen-substituted structures (i and ii), the amount of ∆µge
and the changes promoted by the substituents are more
noticeable than that in the carbon-substituted structures (iii and
iv). It could be speculated that the greater the dipole moment
of the excited state is relative to the ground state, the more the
singlet-triplet energy gap increases; however, some hexaazaacri-
dine structures (X ) N) show irregularities. Tables containing
µg, µe, and ∆µge for all configurations and the figures presenting
these correlations for hexaazaanthracenes and hexaazaacridines
are available in the Supporting Information.
The relative energies of the singlet and triplet states of
molecules with structures i and ii (Y bonded to a nitrogen atom)
show that generally the introduction of deactivating groups
(EWG) strongly favors the triplet state. Electron-donating groups
(EDG) such as -OCH3, -NMe2, and -NH2 destabilize
preferentially the triplet relative to the singlet state. For example,
the ∆EST values for the hexaazaanthracenes (configuration i)
with -OCH3, -NH2, -CN, and -NO2 substitutions are -19.5,
-18.7, -11.3, and -9.3 kcal mol-1, respectively. The structural
configurations iii and iv (Y bonded to a carbon atom) are
generally not sensible enough to substituent effects. For a
rationalization of the substituent-induced changes of the ∆EST
values, in the following, the results based on the configuration
i with meso-carbon atom (X ) CH) as the most sensitive system
are presented (the data for other systems are available in the
Supporting Information).
I. Inductive Substituent Effects. We compared the ∆EST
values with the relative amount of intramolecular charge transfer
caused by the various substituents by examining the local natural
electron populations (NPAs) on the carbon atom meso-C11 (X
) CH) in structure i for which these effects should be most
apparent.45,46 Similarly, Komorowski et al. as well as Geerlings
et al. described substituent effects by means of reactivity indices
such as group hardness and group electronegativity.47 The
relative intramolecular charge transfer (∆qC) due to the inductive
substituent effect relative to -H shows that as ∆qC on meso-
C,11 the singlet-triplet energy gaps decrease (Table 6). As a
result, N-bonded σ-acceptor substituent groups decrease the
∆EST.
Next we studied the effects of push-pull substitution, using
a fixed NO2 group as an ideal ∆EST-lowering reference
substituent (-I and -M effects) in the N-position of hexaazaan-
thracene (I-IV, Figure 16). The computed ∆EST values of these
structures with varying the substituents are presented in Figure
17.
(45) (a) Wiberg, K. B. Acc. Chem. Res. 1999, 32, 922. (b) Perez, P.; Toro-
Labbe, A.; Contreras, R. J. Phys. Chem. A 2000, 104, 5882.
(46) Bailey, P.; Ashton, J. J. Org. Chem. 1957, 22, 98.
Generally the ∆EST values are smaller in these four configu-
rations compared to the related systems i-iv; the average
(47) (a) Komorowski, L.; Lipinski, J.; Pyka, M. J. J. Phys. Chem. 1993, 97,
3166. (b) Proft, F. D.; Langenaeker, W.; Geerlings, P. J. Phys. Chem. 1993, 97,
1826. (c) Proft, F. D.; Amira, S.; Choho, K.; Geerlings, P. J. Phys. Chem. 1994,
98, 5227.
(48) Corminboeuf, C.; Heine, T.; Seifert, G.; Schleyer, P. v. R.; Weber, J
Phys. Chem. Chem. Phys. 2004, 6, 273.
J. Org. Chem. Vol. 73, No. 13, 2008 5057

Langer et al.
FIGURE 15. Trends in NICSzz(1) at B3LYP/6-311G(d,p)//B3LYP/6-31G(d,p) in substituted hexaazaanthracene derivatives with configurations i
(substituents with strong positive resonance effect cause smaller ∆EST values) and iv (the resonance effect shows no regular influence on ∆EST
values).
5. Extension of π-Delocalization. Finally, we investigated
the effect of extending the π-delocalization when p-phenyl
groups are placed between the central moiety and the substit-
uents (as in most of the experimentally available systems). While
this has little effect on reducing the ∆EST values when the phenyl
groups are bonded to carbon, the ∆EST values are reduced for
N-bonded phenyl substituents (Figure 21).
With p-substituted phenyl groups as spacers, the ∆EST values
are generally less sensitive to substituent effects (Table 7). This
attenuation of the substituent effects can be attributed to some
FIGURE 16. Push-pull systems with fixed σ- and π-acceptor NO2
substituent(s).
twisting of the two phenyl groups (Ar2 in Figure 6) that causes
loss of overlap. Hence, the ∆EST values decrease most signifi-
cantly when strongly π-delocalized, electron-withdrawing groups
bonded to the nitrogen positions are used; electron-donating
groups attached to carbon only cause a mild reduction of the
∆EST values (Figure 22).
Conclusions
A variety of novel tetraphenylhexaazaacridines and tetraphe-
nylhexaazaanthracenes were synthesized and fully characterized.
These structures are promising candidates for purely organic
materials with very low-lying singlet-triplet energy separations.
The pronounced bathochromism and the appearance of the
fluorescence spectra as well as a comparison with computed
natural charges identify these compounds as zwitterionic singlet
ground state molecules with rather low-lying triplet states.
A comprehensive experimental and DFT investigation of the
electronic properties of azaacridines and azaanthracenes reveals
several important clues for understanding the electronic struc-
tures of these molecules. Generally, TPH-anthracenes show
∆EST values smaller than those of TPH-acridines. The larger
∆EST of the latter is in line with their somewhat more twisted
topology, greater total dipole moment (more zwitterionic
character), greater dipole moment of the first excited state
relative to the ground state, and a larger λmax in the UV
absorption spectra.
FIGURE 17. Trends of ∆EST values in substituted hexaazaanthracene
derivatives with push-pull structural configurations I-IV.
reduction is ca. 4 kcal mol-1 for I-III/i-iii and ca. 8 kcal mol-1
for IV/iv. The noticeable reduction for structure IV can be
explained by having two N-bonded NO2 substituents. The
structural configuration IV containing nitrogen-bonded substit-
uents with both -I and -M effects and carbon-bonded
substituents with a +I effect is suggested to be the optimum
combination to affect the electronic structure for hexaazaan-
thracene to provide a small ∆EST. Structure I is also presented
as the second favorable configuration with two N-bonded σ-
and π-acceptor substituents (Figure 17).
A systematic computational analysis reveals that the electronic
properties, the number, and the position of a series of substit-
uents have a large effect on the ∆EST values (Figure 20).
Nitrogen-bonded substituents show larger effects than those
5058 J. Org. Chem. Vol. 73, No. 13, 2008

3,5,7,9-Substituted Hexaazaacridines
FIGURE 18. The correlation between ∆EST values (kcal mol-1) and ∆µge values (D) for two structural configurations I (right) and IV (left) for
substituted hexaazaanthracenes. All data were obtained at B3LYP/6-31G(d,p).
respect to optimizing π-electron delocalization and also func-
tionalizing other heteroacenes to minimize the ∆EST values or
to obtain triplet ground state structures.
Experimental Section
General Comments. All solvents were dried by standard
methods, and all reactions were carried out under an inert
atmosphere. For 1H and 13C NMR spectra, the deuterated solvents
indicated were used. Mass spectrometric data (MS) were obtained
by electron ionization (EI, 70 eV), chemical ionization (CI, H2O),
or electrospray ionization (ESI). For preparative scale chromatog-
raphy, silica gel (60-200 mesh) was used. Melting points are
uncorrected.
N,N′-Di(benzoyl)-2,6-diaminopyridine (6a). To an ice-cold
suspension of 2,6-diaminopyridine (5.50 g, 50.0 mmol) and NEt3
(10.10 g, 100.0 mmol) in dry dichloromethane (150 mL) was added
dropwise a dichloromethane solution (50 mL) of benzoyl chloride
(14.06 g, 100.0 mmol). The reaction mixture was stirred at 20 °C
for 12 h and subsequently poured into 200 mL of water. The organic
layer was separated, washed with an aqueous solution of NaHCO3
(200 mL, 5%) and with water, and dried (MgSO4). The solution
was filtered and concentrated under reduced pressure, and the
residue was recrystallized from ethanol: Yield 7.60 g (48%),
colorless prisms (ethanol), mp 177-179 °C; IR (KBr) ν˜ ) 3340
(s), 3061, 3035, 3006 (w), 1653, 1586, 1526, 1487, 1459 (s), 1396
(m), 1307 (s), 1269, 1242 (m), 1189, 1161, 1128, 1078, 903 (w),
FIGURE 19. Trends of NICSzz(1) for the inner rings of selected
substituted hexaazaanthracene derivatives with configuration I-IV at
the B3LYP/6-311G(d,p)//B3LYP/6-31G(d,p) level of theory.
bonded to carbon, owing to the general molecular orbital
structure of these systems that favors the involvement of
π-delocalization via the nitrogen atoms. The ∆EST values in
N-bonded pull-pull electronic systems are smaller than in
pull-push and push-push systems. For TPH-anthracene, this
reduction is as large as -8.9 kcal mol-1 compared to the
unsubstituted parent system with a ∆EST of -18.8 kcal mol-1
.
We are currently engaged in the design of new azaacenes with
FIGURE 20. Trend of reduction in ∆EST values through improving the structural configurations.
J. Org. Chem. Vol. 73, No. 13, 2008 5059

Langer et al.
FIGURE 21. The effect of π-delocalization extension on reduction of the ∆EST value (kcal mol-1).
FIGURE 22. Comparison of the π-delocalization and substituent effects on the reduction of the ∆EST values (kcal/mol–1) in carbon-bonded (a) and
nitrogen-bonded (b) systems, separately.
TABLE 7. Comparison of the Substituent Effect on ∆EST Values
off, washed with water, and recrystallized from 2-propanol: Yield
10.40 g (54%), colorless prisms (2-propanol), mp 284 °C; IR (KBr)
ν˜ ) 3424 (m), 3336 (s), 3094, 3062, 3041 (w), 1652, 1589, 1522,
1484, 1457, 1312 (s), 1243, 1093, 844, 797, 751, 634, 584, 542,
(kcal mol-1) with and without Using Phenyl Spacers
X
Ar1
Ar2
∆EST
N
Ph/H
Ph/H
Ph/H
Ph/H
Ph/H
Ph/H
p-NMe2C6H4/NMe2
p-NO2C6H4/NO2
Ph/H
-20.8/-24.1
-16.3/-18.8
-17.0/-18.6
-14.1/-9.3
-15.8/-18.0
-16.8/-18.8
-13.2/-8.9
1
486 (m); H NMR (DMSO-d6, 300 MHz) δ ) 7.59-7.62 (m, 4
CH
CH
CH
CH
CH
CH
H), 7.84-7.94 (m, 3 H), 8.01-8.04 (m, 4 H), 10.64 (s, 2 H); 13C
NMR (DMSO-d6, 150 MHz) δ ) 111.4, 128.5, 129.8, 132.8, 136.8,
140.0, 150.3, 164.8; MS (EI, 70 eV) m/z (%) ) 387 (12, [M]+,
37Cl, 35Cl), 385 (21, [M]+, 2× 35Cl), 357 (12), 141 (31), 139 (100),
112 (11), 110 (35). Anal. Calcd for C19H13Cl2N3O2 (386.24): C,
59.09; H, 3.39; N, 10.88. Found: C, 58.73; H, 3.55; N, 10.85.
N,N′-Bis(4-nitrobenzoyl)-2,6-diaminopyridine (6d). The reac-
tion was carried out following the procedure as given for the
synthesis of 6a, starting with p-nitrobenzoyl chloride (18.56 g, 100.0
mmol), 2,6-diaminopyridine (5.50 g, 50.0 mmol), and NEt3 (10.10
g, 100.0 mmol) in 200 mL of dichloromethane. The solvent was
removed under reduced pressure, and the residue was suspended
in 300 mL of water and refluxed for 40 min. The solid was filtered
off, suspended in 300 mL of an aqueous solution of NaHCO3, and
stirred for 40 min. The solid was filtered off, washed with water,
and refluxed in 2-propanol (150 mL) for 40 min: Yield 10.70 g
(62%), colorless prisms, mp 270-274 °C; IR (KBr) ν˜ ) 3402,
3360, 3111 (w), 1694, 1658, 1602, 1585 (m), 1526, 1453, 1351
p-NMe2C6H4/NMe2
p-NO2C6H4/NO2
p-NMe2C6H4/NMe2
Ph/H
p-NO2C6H4/NO2
796, 710, 690, 632 (s), 582 (w); 1H NMR (CDCl3, 300 MHz) δ )
7.47-7.60 (m, 6 H, Ar), 7.81 (t, J ) 8.1 Hz, 1 H, H-4, Pyr),
7.88-7.92 (m, 4 H, Ar), 8.11 (d, J ) 8.1 Hz, 2 H, H-3, H-4, Pyr),
8.33 (br s, 2 H, NH); 1H NMR (DMSO-d6, 300 MHz) δ )
7.50-7.61 (m, 6 H), 7.91-8.40 (m, 7 H), 10.50 (s, 2 H); 13C NMR
(DMSO-d6, 75 MHz) δ ) 111.2, 127.8, 128.3, 131.9, 133.9, 139.8,
150.4, 165.7; MS (EI, 70 eV) m/z (%) ) 318 (12, [M + H]+), 317
(72, [M]+), 290 (11), 289 (85), 288 (100), 221 (11), 105 (37). Anal.
Calcd for C19H15N3O2 (317.12): C, 71.94; H, 4.73; N, 13.25. Found:
C, 72.26; H, 5.09; N, 13.47.
N,N′-Bis(4-methylbenzoyl)-2,6-diaminopyridine (6b). The re-
action was carried out following the procedure as given for the
synthesis of 6a, starting with p-methylbenzoyl chloride (15.50 g,
100.0 mmol), 2,6-diaminopyridine (5.50 g, 50.0 mmol), and NEt3
(10.10 g, 100.0 mmol) in 200 mL of dichloromethane: Yield 10.70
g (62%), colorless prisms (ethanol), mp 215-217 °C; IR (KBr) ν˜
) 3423 (m), 3344 (m), 3028, 2919 (w), 1653 (s), 1609 (m), 1584,
1
(s), 1301, 1250, 859, 715 (m); H NMR (DMSO-d6, 300 MHz) δ
) 7.59-7.62 (m, 4 H), 7.84-7.94 (m, 3 H), 8.01-8.04 (m, 4 H),
10.64 (s, 2 H); 13C NMR (DMSO-d6, 600 MHz) δ ) 111.9, 123.5,
129.5, 139.9, 140.3, 149.3, 150.2, 164.6; MS (EI, 70 eV) m/z (%)
) 407 (34, [M]+), 379 (19), 377 (24), 149 (100), 121 (38), 120
(55), 104 (51), 92 (28), 76 (25). C19H13N5O6 (407.34).
1
1529, 1497, 1456, 1302 (s), 1242, 796, 744, 616 (m); H NMR
(DMSO-d6, 300 MHz) δ ) 2.39 (s, 6 H), 7.32-7.35 (m, 4 H),
7.87-7.95 (m, 7 H), 10.36 (s, 2 H); 13C NMR (DMSO-d6, 150
MHz) δ ) 21.0, 111.0, 127.9, 129.0, 131.2, 139.9, 142.1, 150.5,
165.5; MS (EI, 70 eV) m/z (%) ) 346 (10, [M + H]+), 345 (48,
[M]+), 317 (17), 316 (12), 119 (100), 91 (37). Anal. Calcd for
C21H19N3O2 (345.40): C, 73.03; H, 5.54; N, 12.17. Found: C, 73.19;
H, 5.25; N, 12.86.
N,N′-Bis(4-chlorobenzoyl)-2,6-diaminopyridine (6c). The reac-
tion was carried out following the procedure as given for the
synthesis of 6a, starting with 17.5 g of p-chlorobenzoyl chloride
(100 mmol), 2,6-diaminopyridine (5.50 g, 50.0 mmol), and NEt3
(10.10 g, 100.0 mmol) in 200 mL of dichloromethane. The solvent
was removed under reduced pressure, the residue was suspended
in 300 mL of water, and the suspension was stirred for 40 min.
The solid was filtered off, suspended in 300 mL of an aqueous
solution of NaHCO3, and stirred for 40 min. The solid was filtered
N,N′-Bis(naphth-2-oyl)-2,6-diaminopyridine (6e). The reaction
was carried out following the procedure as given for the synthesis
of 6a, starting with 2-naphthoylchloride (7.63 g, 40.0 mmol), 2,6-
diaminopyridine (2.18 g, 20.0 mmol), and NEt3 (4.05 g, 40.0 mmol)
in 200 mL of dichloromethane. The solvent was removed under
reduced pressure, and the residue was suspended in 300 mL of
water and refluxed for 40 min. The solid was filtered off, suspended
in 300 mL of an aqueous solution of NaHCO3, and stirred for 40
min. The solid was filtered off, washed with water, and refluxed in
2-propanol (150 mL) for 40 min: Yield 5.82 g (70%), colorless
prisms, mp 220-223 °C; IR (KBr) ν˜ ) 3340 (m), 3054 (w), 1653,
1588, 1515, 1454, 1307 (s), 1237, 801, 762 (m); 1H NMR (DMSO-
d6, 300 MHz) δ ) 7.65 (cm, 4 H), 7.96-8.10 (m, 11 H), 8.69 (m,
2 H), 10.67 (s, 2 H); 13C NMR (DMSO-d6, 600 MHz) δ ) 111.1,
124.3, 126.8, 127.6, 128.0, 128.5, 129.1, 131.2, 132.0, 134.4, 140.0,
5060 J. Org. Chem. Vol. 73, No. 13, 2008

3,5,7,9-Substituted Hexaazaacridines
150.0, 165.8; MS (EI, 70 eV) m/z (%) ) 419 (24, [M + 2H]+),
155 (58, [C10H7CO]+), 139 (14), 127 (41, [C10H7]+). C27H19N3O2
(417.47).
585 nm; MS (EI, 70 eV) m/z ) 491 (M+, 100), 77 (30); HRMS
(EI, 70 eV) calcd for C31H21N7 491.1858 (M+), found m/z 491.1858
( 2 ppm.
Synthesis of 3a without Purification of the Intermediates. A
suspension of 6a (2.10 g, 6.6 mmol) and PCl5 (4.01 g, 19.3 mmol)
in dry toluene (40 mL) was stirred under reflux for 2 h. The solvent
and POCl3 were removed under reduced pressure to give 7a. The
residue (7a) was suspended in dry n-hexane (60 mL), the mixture
was cooled to 0 °C, and phenyl hydrazine (1.7 mL, 16.0 mmol)
was added under argon atmosphere. After a period of 15 min, the
ice bath was removed and the mixture was stirred at 20 °C for
12 h. The pale yellow residue (8a) was filtered off, washed with
n-hexane (30 mL), and dissolved in methanol (150 mL). To the
well stirred suspension was dropwise added DBU (4.0 mL). After
stirring for a few hours, the solution became dark red and a solid
began to precipitate. After stirring for 5 days at 20 °C, the solid
was filtered off and washed with 2-propanol. For purification, the
solid was suspended in glacial acetic acid (10 mL) and stirred at
20 °C for 1.5 h. The red solid was isolated by filtration and washed
with water and, subsequently, 2-propanol: Yield 0.742 g (23%),
dark red solid.
N,N′-Bis(pyrid-4-oyl)-2,6-diaminopyridine (6f). The reaction
was carried out following the procedure as given for the synthesis
of 6a. iso-Nicotinoyl chloride (7.12 g, 40.0 mmol) was added at 0
°C to a suspension of 2,6-diaminopyridine (2.18 g, 20.0 mmol)
and NEt3 (8.10 g, 80.0 mmol) in dichloromethane (200 mL) over
a period of 45 min. The solvent was removed under reduced
pressure, and the residue was suspended in 300 mL of water and
refluxed for 40 min. The solid was filtered off, suspended in 300
mL of an aqueous solution of NaHCO3, and stirred for 40 min.
The solid was filtered off, washed with water, and refluxed in
2-propanol (150 mL) for 40 min. The solid was filtered off and
dried in vacuo: Yield 3.16 g (49%), colorless prisms, mp 275 °C
(decomp); IR (KBr) ν˜ ) 3433 (m), 3145, 3043 (w), 1688, 1588
(s), 1554 (m), 1521 (s), 1490 (m), 1449 (s), 1413 (m), 1308 (s),
1243 (m), 803, 751, 699, 589 (w); 1H NMR (DMSO-d6, 300 MHz)
δ ) 7.88-7.99 (m, 7 H), 8.78-8.80 (m, 4 H), 10.90 (s, 2 H); 13
C
NMR (DMSO-d6, 600 MHz) δ ) 111.9, 121.5, 140.2, 141.1, 150.1,
150.2, 164.6; MS (EI, 70 eV) m/z (%) ) 319 (39, [M]+), 291 (10),
290 (21), 241 (31), 105 (79), 79 (22), 78 (100). C17H13N5O2
(319.32).
5,7-Di(4-tolyl)-3,9-diphenylhexaazaacridine (3b). A suspension
of 6a (2.10 g, 6.6 mmol) and PCl5 (4.01 g, 19.3 mmol) in dry
toluene (20 mL) was stirred under reflux for 2 h. The solvent and
POCl3 were removed under reduced pressure to give 7a. The residue
(7a) was suspended in dry n-hexane (60 mL), tolyl hydrazine (2.74
g, 17.3 mmol) added, the mixture ooled to 0 °C,and NEt3 (1.75 g,
17.3 mmol) dropwise added under agon atmosphere. After a period
of 15 min,the ice bath was removed and the mixture was stirred at
20 °C for 4 h. The pale yellow residue (8b) was filtered off, washed
with n-hexane (30 mL),and suspended in methanol (150 mL). o
the well stirred suspension was dropwise added DBU (3.4 mL).
After stirring for a few hours,the solution became dark red and a
solid began to precipitate. After stirring for 5 days at 20 °C,the
solid was filtered off and washed with 2-propanol. For purification,
the solid was suspended in glacial acetic acid (10 mL) and stirred
at 20 °C for 1.5 h. Subsequently, the solid was filtered off,
suspended in an aqueous solution of sodium hydroxide (2%, 10
mL) and of ethanol (0.3 mL), and stirred at 20 °C for 1 h. The red
dye was isolated by filtration and washed with water and,
subsequently, with methanol: Yield 0.48 g (14%), dark red solid;
IR (KBr) ν˜ ) 3419 (w), 1502 (s), 1446, 1396, 1315, 1293, 700
N,N′-Bis(phenylimidoyl)-2,6-diaminopyridine dichloride (7a).
Bis(amide) 6a (2.10 g, 6.6 mmol) and PCl5 (4.01 g, 19.3 mmol)
were suspended in dry toluene (20 mL). The mixture was stirred
at 20 °C for 1.5 h and under reflux for 2 h. The solvent and POCl3
were removed under reduced pressure. The residue was stored under
1
argon at 4 °C to give a pale yellow solid: Yield 2.05 g (88%); H
NMR (300 MHz, CDCl3) δ ) 7.13 (d, J ) 8.1 Hz, 2 H), 7.48-7.54
(m, 4 H), 7.60-7.66 (m, 2 H), 8.25 (t, J ) 8.1 Hz, 1 H), 8.27-8.31
(m, 4 H); 13C NMR (DMSO-d6, 300 MHz) δ ) 110.4, 128.1, 128.2,
128.5, 132.7, 132.7, 147.79, 167.1. MS (EI, 70 eV) m/z (%) ) 355
(15, [M]+, 37Cl, 35Cl), 353 (35, [M]+, 2× 35Cl), 320 (34), 318 (100,
[M - Cl]+ (35Cl)), 217 (15), 215 (50), 179 (28), 89 (14), 77 (14);
HRMS (EI, 70 eV) calcd for C19H13N3Cl2 353.0486 ([M]+, 2×
35Cl), found m/z 353.0486 ( 2 ppm.
N,N′-2,6-(Pyridine)diylbis(benzamidrazone) (8a). Bis(imidoyl)
dichloride 7a (2.10 g 6.0 mmol) was suspended in dry n-hexane
(60 mL), and the mixture was cooled to 0 °C. To the mixture was
dropwise added phenyl hydrazine (1.7 mL, 16.0 mmol). After
stirring for 15 min, the ice bath was removed and the mixture was
stirred at 20 °C for 14 h. The precipitated pale yellow solid was
isolated by filtration and washed with n-hexane (30 mL): Yield
1
(m); H NMR (TFA-d1, 600 MHz) δ ) 2.38 (s, 6 H), 5.85 (s, 1
H), 7.34 (m, 8 H), 7.45-7.47 (m, 4 H), 7.55-7.57 (m, 2 H),
7.99-8.00 (m, 4 H); UV-vis (DMF) λmax (log ꢀ) ) 291 (4.61),
527 (4.33), 548 (4.35) nm; UV-vis (toluene) λmax (log ꢀ) ) 292
(4.24), 303 (4.22), 525 (3.92), 553 (3.94) nm; UV-vis (CH3CN)
λmax (log ꢀ) ) 288 (3.80), 523 (3.59), 544 (3.58) nm; UV-vis
(acetic acid) λmax (log ꢀ) ) 274 (4.59), 504 (4.32), 538 (4.43) nm;
MS (EI, 70 eV) m/z (%) ) 520 (37), 519 (100, [M]+), 103 (10),
91 (16); HRMS (EI, 70 eV) calcd for C33H25N7 519.2166 (M+),
found m/z 519.2167 ( 2 ppm.
1
2.70 g (90%), colorless prisms; H NMR (300 MHz, CDCl3) δ )
6.80 (d, J ) 7.8 Hz, 2 H), 7.50-7.38 (m, 7 H), 7.44-7.56 (m, 9
H), 7.83 (t, J ) 7.8 Hz, 1 H), 8.19-8.22 (m, 4 H); MS (EI, 70 eV)
m/z ) 497 (M+, 4), 480 (12), 91 (100). C31H27N7 (497.60).
3,5,7,9-Tetraphenylhexaazaacridine (3a). To a well stirred
solution of crude 8a (2.50 g, 5.0 mmol) in methanol (120 mL) was
dropwise added DBU (3.0 mL, 14.0 mmol). After stirring for a
few minutes, the solution became dark red and a red-brownish solid
began to precipitate. After stirring for 4 days, the solid was isolated
by filtration and washed with 2-propanol. For purification, the solid
was suspended in glacial acetic acid (10 mL) and stirred at 20 °C
for 20 min. The red solid was isolated by filtration and washed
with water and 2-propanol and dried in vacuo: Yield 0.742 g (30%),
red solid; IR (KBr) ν˜ ) 3427 (m), 3059 (w), 1594 (m), 1503 (s),
1445 (s), 1396 (s), 1317 (s), 1171 (m), 1069 (m), 1006 (m), 859
(m), 777 (m), 699 (s), 634 (w); 1H NMR (TFA-d1, 600 MHz) δ )
5.75 (s, 1 H), 7.25-7.60 (m, 16 H), 7.95-8.15 (m, 4 H); 13C NMR
(TFA-d1, 150 MHz) δ ) 91.1, 126.3, 129.2, 131.2, 132.1, 132.5,
133.9, 135.4, 141.2, 141.5, 158.3, 158.6; UV-vis (AcOH) λmax
(log ꢀ) ) 273 (4.74), 535 (4.57) nm; UV-vis (DMF) λmax (log ꢀ)
) 290 (4.76), 299 (4.74), 344 (3.81), 525 (4.50), 547 (4.51) nm;
UV-vis (H3PO4) λmax (log ꢀ) ) 540 (4.29) nm; UV-vis (H2SO4)
λmax (log ꢀ) ) 662 (4.34) nm; fluorescence (DMF) Fλmax (λEx) )
5,7-Diphenyl-3,9-di(4-tolyl)-hexaazaacridine (3c). A suspen-
sion of 6b (2.28 g, 6.6 mmol) and PCl5 (4.01 g, 19.3 mmol) in dry
toluene (20 mL) was stirred under reflux for 2 h. The solvent and
POCl3 were removed under reduced pressure to give 7b. The residue
(7b) was suspended in dry n-hexane (60 mL), the mixture cooled
to 0 °C, and phenyl hydrazine (1.7 mL, 16.0 mmol) dropwise added
under an argon atmosphere. After a period of 15 min, the ice bath
was removed and the mixture was stirred at 20 °C for 4 h. The
pale yellow residue (8c) was filtered off, washed with n-hexane
(30 mL), and suspended in methanol (200 mL). To the well stirred
suspension was dropwise added DBU (3.4 mL). After stirring for
a few hours, the solution became dark red and a solid began to
precipitate. After stirring for 5 days at 20 °C, the solid was filtered
off and washed with 2-propanol. For purification, the solid was
suspended in glacial acetic acid (10 mL) and stirred at 20 °C for
1 h. Subsequently, the solid was filtered off, suspended in an
aqueous solution of sodium hydroxide (4%, 10 mL) and of ethanol
J. Org. Chem. Vol. 73, No. 13, 2008 5061

Langer et al.
(0.3 mL), and stirred at 20 °C for 1 h. The red solid was isolated
by filtration, washed with water and 2-propanol, and dried at 100
°C: Yield 0.206 g (6%), red solid; IR (KBr) ν˜ ) 3428, 3054, 1594
(w), 1499 (s), 1396, 1301 (m), 1174, 1083, 1005, 863, 829, 776
(w); 1H NMR (TFA-d1, 600 MHz) δ ) 2.34 (s, 6 H), 5.88 (s, 1 H),
7.29-7.52 (m, 14 H), 7.86 (m, 4 H); 13C NMR (TFA-d1, 150 MHz)
δ ) 15.7, 91.1, 119.9, 121.0, 122.3, 122.9, 125.7, 126.2, 127.7,
134.9, 141.4, 151.5, 151.9; UV-vis (AcOH) λmax (log ꢀ) ) 283
(4.64), 542 (4.50) nm; UV-vis (CH2Cl2) λmax (log ꢀ) ) 239 (4.11),
294 (4.73), 305 (4.75), 346 (3.91), 523 (4.42), 547 (4.51) nm;
UV-vis (DMF) λmax (log ꢀ) ) 294 (4.71), 525 (4.41), 548 (4.47)
nm; UV-vis (toluene) λmax (log ꢀ) ) 296 (4.67), 308 (4.71), 352
(3.64), 524 (4.35), 553 (4.43) nm; UV-vis (formic acid) λmax (log
ꢀ) ) 283 (4.63), 542 (4.50) nm; UV-vis (H2SO4) λmax (log ꢀ) )
226 (4.32), 341 (4.37), 451 (4.11), 685 (4.25) nm; fluorescence
(DMF) Fλmax (λEx) ) 585 (310) nm; MS (EI, 70 eV) m/z (%) )
520 (22), 519 (60, [M]+), 117 (13), 77 (16); HRMS (EI, 70 eV)
calcd for C33H25N7 519.2166 (M+), found m/z 519.2164 ( 2 ppm.
5,7-Diphenyl-3,9-di(4-chlorophenyl)hexaazaacridine (3d). A
suspension of 6c (2.27 g, 6.6 mmol) and PCl5 (4.01 g) in toluene
(20 mL) was refluxed for 3 h. The workup, carried out as described
for the synthesis of 7a, afforded 7c. The reaction of 7c with phenyl
hydrazine (1.7 mL) gave 8d as described for the synthesis of 8a.
A methanol solution (250 mL) of 8d and of DBU (3.4 mL) was
stirred for 14 days at 20 °C. The solid was filtered off and washed
with 2-propanol to give the crude product (0.862 g, 24%). For
purification, the solid was suspended in glacial acetic acid (10 mL)
and stirred at 20 °C for 1.5 h. Subsequently, the solid was filtered
off, suspended in a mixture of an aqueous solution of sodium
hydroxide (2%, 10 mL) and of ethanol (0.5 mL), and stirred at 20
°C for 1 h. The red solid was isolated by filtration, washed with
water and, subsequently, with 2-propanol, and dried at 100 °C: Yield
0.833 g (23%), red solid; IR (KBr) ν˜ ) 3436 (m), 3095, 3060 (w),
1630, 1596, 1538 (m), 1497 (s), 1392, 1304, 1089, 1005, 861, 699
280, 537 nm; fluorescence (DMF) Fλmax (λEx) ) 583 (555) nm;
MS (EI, 70 eV) m/z (%) ) 591 (7, [M]+, 2× 37Cl), 590 (11), 589
(29, [M]+, 37Cl, 35Cl), 588 (17), 587 (38, [M]+, 2× 35Cl), 207 (28),
137 (51), 119 (51), 91 (50); HRMS (EI, 70 eV) calcd for
C33H23Cl2N7 m/z 587.13865 ([M]+, 2× 35Cl), found m/z 587.14010
( 2 ppm.
5,7-Diphenyl-3,9-di(4-nitrophenyl)hexaazaacridine (3f). A
suspension of 6d (2.69 g, 6.6 mmol) and PCl5 (4.01 g) in toluene
(20 mL) was refluxed for 2 h. The workup, carried out as described
for 7a, afforded 7d. The reaction of 7d with phenyl hydrazine (1.7
mL) afforded 8f as described for the synthesis of 8c. A methanol
solution (400 mL) of 8f and of DBU (2.7 mL) was stirred for 14
days at 20 °C. The precipitated solid was filtered off and washed
with 2-propanol. For purification, the solid was suspended in acetic
acid (10 mL) and stirred at 20 °C for 1.5 h. The red solid was
isolated by filtration, washed with water and 2-propanol: Yield 0.542
g (14%), red solid; IR (KBr) ν˜ ) 3431 (m), 1629 (w), 1600 (m),
1499 (s), 1456, 1393 (m), 1345 (s), 1304 (m), 1235, 1213, 1170,
1
1070 (w), 1006, 857, 706 (m); H NMR (TFA-d1, 600 MHz) δ )
5.63 (s, 1 H), 7.42-7.54 (m, 10 H), 8.29-8.33 (m, 8 H); 13C NMR
(TFA-d1, 150 MHz) δ ) 82.5, 119.0, 119.1, 123.3, 125.4, 126.8,
133.1, 134.0, 135.1, 144.7, 150.8, 151.3; UV-vis (acetic acid) λmax
(log ꢀ) ) 280 (4.44), 500 (4.27), 534 (4.42) nm; UV-vis (DMF)
λmax (log ꢀ) ) 314 (4.58), 485 (4.28), 521 (4.40), 547 (4.34), 587
(4.10) nm; UV-vis (H2SO4) λmax (log ꢀ) ) 227 (3.95), 274 (4.12),
328 (3.95), 485 (3.77), 631 (4.03) nm; UV-vis (H3PO4) λmax
)
284, 538 nm; fluorescence (DMF) Fλmax (λEx) ) 610 (558) nm;
MS (EI, 70 eV) m/z (%) 581 (6, [M]+), 91 (7); HRMS (EI, 70 eV)
calcd for C31H19N9O4 m/z 581.15545 (M+), found m/z 581.15370
( 2 ppm.
5,7-Tolyl-3,9-di(4-nitrophenyl)hexaazaacridine (3g). A suspen-
sion of 6d (2.69 g, 6.6 mmol) and of PCl5 (4.01 g) in toluene (20
mL) was refluxed for 2 h. The workup, carried out as described
for 7a, afforded 7d. The reaction of 7d with tolyl hydrazine (2.74
g, 17.3 mmol) afforded 8g as described for the synthesis of 8c. A
methanol solution (250 mL) of 8g and of DBU (2.7 mL) was stirred
for 14 days at 20 °C. The precipitated solid was filtered off, washed
with 2-propanol, and recrystallized from chloroform: Yield 0.428
g (11%), dark red solid (chloroform); IR (KBr) ν˜ ) 3426, 1602
(w), 1502 (s), 1392 (w), 1345 (s), 1302 (m), 1006 (w), 857 (w),
709 (w); 1H NMR (TFA-d1, 600 MHz) δ ) 2.39 (s, 6 H), 5.69 (s,
1 H), 7.32 (d, J ) 8.1 Hz, 4 H), .35 (d, J ) 8.1 Hz, 4 H), 8.29 (d,
J ) 9.0 Hz, 4 H), 8.32 (d, J ) 9.0 Hz, 4 H); 13C NMR (TFA-d1,
150 MHz) δ ) 15.7, 119.9, 122.4, 122.9, 125.7, 126.2, 127.7, 134.9,
41.4, 151.5, 151.9; UV-vis (DMF) λmax (log ꢀ) ) 315 (4.66), 524
(4.46); UV-vis (DCM) λmax (log ꢀ) ) 240 (4.41), 310 (4.72), 481
(4.34), 522 (4.52), 543 (4.55), 582 (4.25) nm; UV-vis (toluene)
λmax (log ꢀ) ) 310 (3.72), 485 (3.38), 519 (3.46), 544 (3.59), 583
(3.38) nm; UV-vis (CH3CN) λmax (log ꢀ) ) 239 (3.99), 310 (4.28),
480 (3.94), 520 (4.07), 584 (3.78) nm; UV-vis (acetic acid) λmax
(log ꢀ) ) 280 (4.56), 501 (4.39), 536 (4.55) nm; UV-vis (H2SO4)
λmax ) 228, 274, 345, 457, 649 nm; UV-vis (H3PO4) λmax ) 279,
335, 537 nm; fluorescence (DMF) Fλmax (λEx) ) 600 (565) nm.
C33H23N9O4 (609.604).
5,7-Diphenyl-3,9-dinaphthylhexaazaacridine (3h). A suspen-
sion of 6e (2.76 g, 6.6 mmol) and of PCl5 (4.01 g) in toluene (20
mL) was refluxed for 3 h. The workup, carried out as described
for 7a, afforded 7e. The reaction of 7e with phenyl hydrazine (1.7
mL) afforded 8h as described for the synthesis of 8c. An ethanol
solution (500 mL) of 8h and of DBU (3.4 mL) was stirred for 14
days at 20 °C. The precipiated solid was filtered off and washed
with 2-propanol to give the crude product (0.823 g, 21%). For
purification, the solid was suspended in acetic acid (10 mL) and
stirred at 20 °C for 1.5 h. The solid was filtered off and suspended
in a mixture of an aqueous solution of sodium hydroxide (2%, 10
mL) and of ethanol (0.3 mL) and stirred at 20 °C for 1 h. The red
solid was isolated by filtration, washed with water and, subse-
quently, with 2-propanol, and dried at 100 °C: Yield 0.757 g (19%),
red solid; IR (KBr) ν˜ ) 3056, 1593 (w), 1498 (s), 1398, 1350,
1
(m); H NMR (TFA-d1, 600 MHz) δ ) 5.68 (s, 1 H), 7.41-7.42
(m, 8 H), 7.52 (m, 6 H), 7.97-7.99 (m, 4 H); 13C NMR (TFA-d1,
150 MHz) δ ) 83.4, 119.9, 124.2, 125.0, 126.2, 127.5, 134.9, 135.0,
135.8, 151.5, 152.6; UV-vis (acetic acid) λmax (log ꢀ) ) 280 (4.77),
537 (4.60) nm; UV-vis (DMF) λmax (log ꢀ) ) 295 (4.81), 304
(4.80), 347 (3.90), 527 (4.53), 552 (4.50) nm; UV-vis (H2SO4)
λmax (log ꢀ) ) 225 (4.39), 287 (4.42), 342 (4.34), 454 (4.04), 667
(4.43) nm; fluorescence (DMF) Fλmax (λEx) ) 579 (304) nm; MS
(EI, 70 eV) m/z (%) ) 563 (7, [M]+, 2× 37Cl), 562 (11), 561 (34,
[M]+, 37Cl, 35Cl), 560 (17), 559 (48, [M]+, 2× 35Cl), 180 (11), 137
(49), 77 (100); HRMS (EI, 70 eV) calcd for C31H19Cl2N7 m/z
559.1074 ([M]+, 2× 35Cl), found m/z 559.1064 ( 2 ppm.
5,7-Tolyl-3,9-di(4-chlorophenyl)hexaazaacridine (3e). A sus-
pension of 6c (2.55 g, 6.6 mmol) and PCl5 (4.01 g) in toluene (20
mL) was refluxed for 2 h. The workup, carried out as described
for the synthesis of 7a, afforded 7c. The reaction of 7c with tolyl
hydrazine (12.74 g, 7.3 mmol) and NEt3 (1.75 g, 17.3 mmol) gave
8e as described for the synthesis of 8c. A methanol solution (250
mL) of 8e and of DBU (2.7 mL) was stirred for 14 days at 20 °C.
The precipitated solid was filtrated off, washed with 2-propanol,
and recrystallized from chloroform: Yield 0.438 g (11%), dark red
solid (chloroform); IR (KBr) ν˜ ) 3044, 1637, 1590 (w), 1499 (s),
1392 (m), 1299, 1088, 1005, 861, 831 (w); 1H NMR (TFA-d1, 600
MHz) δ ) 2.35 (s, 6 H), 5.73 (s, 1 H), 7.27 (d, J ) 7.5 Hz, 4 H),
7.32 (d, J ) 7.5 Hz, 4 H), 7.41 (d, J ) 8.4 Hz, 4 H), 7.96 (d, J )
8.4 Hz, 4 H); 13C NMR (TFA-d1, 150 MHz) δ ) 14.6, 118.9, 123.4,
124.1, 124.3, 125.8, 131.7, 134.1, 135.0, 138.4, 150.8, 151.5;
UV-vis (DMF) λmax (log ꢀ) ) 295 (4.82), 304 (4.81), 529 (4.54),
553 (4.53) nm; UV-vis (CH2Cl2) λmax (log ꢀ) ) 295 (4.87), 305
(4.87), 527 (4.57), 551 (4.59) nm; UV-vis (CH3CN) λmax ) 292,
525, 546 nm; UV-vis (toluene) λmax (log ꢀ) ) 297 (4.72), 308
(4.73), 528 (4.42), 556 (4.43) nm; UV-vis (ethanol) λmax (log ꢀ)
) 293 (4.35), 301 (4.34), 524 (4.03), 547 (4.09) nm; UV-vis (acetic
acid) λmax (log ꢀ) ) 281 (4.80), 507 (4.53), 539 (4.63) nm; UV-vis
(H2SO4) λmax ) 248, 323, 459, 681 nm; UV-vis (H3PO4) λmax
)
5062 J. Org. Chem. Vol. 73, No. 13, 2008

3,5,7,9-Substituted Hexaazaacridines
1296, 1232, 860, 762 (m); 1H NMR (TFA-d1, 600 MHz) δ ) 5.54
(s, 1 H), 7.41-7.56 (m, 14 H), 7.81-8.10 (m, 8 H), 8.67 (m, 2 H);
13C NMR (TFA-d1, 150 MHz) δ ) 85.5, 118.3, 119.9, 123.3, 123.4,
123.5, 124.6, 124.8, 125.0, 126.2, 127.6, 128.9, 131.6, 134.9, 152.2;
UV-vis (acetic acid) λmax (log ꢀ) ) 267 (4.80), 306 (4.71), 546
(4.64) nm; UV-vis (DMF) λmax ) 276, 315, 528, 557 nm; UV-vis
(H2SO4) λmax (log ꢀ) ) 217 (4.74), 262 (4.74), 310 (4.50), 460
(4.13), 663 (4.50) nm; MS (EI, 70 eV) m/z (%) 592 (23), 591 (51,
[M]+), 281 (11), 208 (14), 207 (61), 154 (16), 153 (98), 126 (20),
77 (28); HRMS (EI, 70 eV) calcd for C39H25N7 m/z 591.2166 (M+),
found m/z 591.2163 ( 2 ppm.
105 (1), 78 (2). C22H20N2O2 (344.41). The reaction of the bis(ben-
zamide) (2.27 g, 6.6 mmol) with PCl5 (4.01 g) in toluene (20 mL)
afforded, following the procedure as given for the synthesis of 7a
(2 h reflux) the bis(imidoyl) chloride. The latter was reacted with
phenyl hydrazine (1.7 mL), following the procedure as given for
the synthesis of 8a, to give the corresponding bis(amidrazone). The
reaction of the latter with DBU (1.7 mL) in MeOH (150 mL),
following the procedure as given for the synthesis of 3a, afforded
4b. Yield of the crude product: 0.629 g (18%). For purification,
the solid was suspended in acetic acid (10 mL) and stirred at 20
°C for 1 h. The red solid was isolated by filtration and washed
with water and, subsequently, with 2-propanol: Yield 0.338 g (10%),
red solid; IR (KBr) ν˜ ) 3433 (m), 1594 (w), 1504 (s), 1397, 1308
5,7-Diphenyl-3,9-di(isonicotinoyl)hexaazaacridine (3i). A sus-
pension of 6f (2.11 g, 6.6 mmol) and of PCl5 (4.01 g) was refluxed
in toluene (20 mL) for 2 h. The workup, carried out as described
for 7a, afforded 7f. The reaction of 7f with phenyl hydrazine (1.7
mL) afforded 8i as described for the synthesis of 8c. An ethanol
solution (300 mL) of 8f and of DBU (3.7 mL) was stirred for 6
days at 20 °C. The precipitated solid was filtered off and washed
with 2-propanol: Yield 0.891 g (27%), red solid; IR (KBr) ν˜ )
3428, 3039, 1636, 1597 (w), 1501 (s), 1400, 1325, 1295, 857, 695
1
(m), 1317 (s), 1176, 983, 856, 832, 765 (w); H NMR (TFA-d1,
600 MHz) δ ) 2.32 (s, 6 H), 6.09, 6.10 (2 s, 2 H), 7.26-7.46 (m,
14 H), 7.67-7.68 (m, 4 H); 13C NMR (TFA-d1, 150 MHz) δ )
15.8, 88.6, 94.7, 118.7, 119.9, 122.4, 126.2, 126.3, 128.1, 135.2,
137.6, 143.0, 144.2, 145.3; UV-vis (AcOH) λmax (log ꢀ) ) 315
(4.84), 391 (3.77), 411 (3.76), 568 (4.42) nm; UV-vis (DMF) λmax
(log ꢀ) ) 316 (4.84), 399 (3.79), 506 (4.26), 533 (4.41), 556 (4.47),
593 (4.29) nm; UV-vis (CH2Cl2) λmax (log ꢀ) ) 278 (4.41), 316
(4.87), 328 (4.92), 399 (3.79), 533 (4.41), 554 (4.47), 591 (4.29)
nm; UV-vis (toluene) λmax (log ꢀ) ) 317 (4.83), 330 (4.91), 401
(3.80), 532 (4.37), 559 (4.41), 599 (4.29) nm; UV-vis (formic acid)
λmax (log ꢀ) ) 309 (4.81), 410 (3.96), 585 (4.37) nm; UV-vis
(H2SO4) λmax (log ꢀ) ) 259 (4.21), 325 (4.57), 426 (4.38), 719
(4.14) nm; fluorescence (DMF) Fλmax (λEx) ) 395 (350) nm; MS
(EI, 70 eV) m/z (%) 519 (38), 518 (100, [M]+), 296 (12); HRMS
(EI, 70 eV) calcd for C34H26N6 m/z 518.2210 (M+), found m/z
518.2210 ppm.
1
(m); H NMR (TFA-d1, 600 MHz) δ ) 5.68 (s, 1 H), 7.43-7.44
(m, 4 H), 7.54-7.60 (m, 6 H), 8.70 (d, J ) 6.9 Hz, 4 H), 8.90 (d,
J ) 6.9 Hz, 4 H); 13C NMR (TFA-d1, 150 MHz) δ ) 84.8, 119.0,
119.9, 125.7, 127.5, 133.5, 137.0, 137.2, 145.3, 148.5, 151.9;
UV-vis (DMF) λmax (log ꢀ) ) 291 (4.72), 481 (4.28), 530 (4.49),
562 (4.25) nm; UV-vis (CH2Cl2) λmax (log ꢀ) ) 237 (4.38), 288
(4.77), 477 (4.28), 527 (4.54), 559 (4.34) nm; UV-vis (CH3CN)
λmax (log ꢀ) ) 236 (4.34), 287 (4.74), 476 (4.30), 525 (4.49), 556
(4.22) nm; UV-vis (acetone) λmax (log ꢀ) ) 342 (3.64), 477 (4.07),
528 (4.24), 560 (4.02) nm; UV-vis (acetic acid) λmax (log ꢀ) )
269 (4.53), 521 (4.50) nm; UV-vis (toluene) λmax (log ꢀ) ) 290
(4.36), 478 (3.90), 530 (4.11), 564 (3.98) nm; UV-vis (ethanol)
λmax (log ꢀ) ) 238 (4.39), 288 (4.86), 478 (4.29), 526 (4.51), 557
(4.28) nm; UV-vis (H2SO4) λmax ) 244, 381, 558 nm; UV-vis
(H3PO4) λmax (log ꢀ) ) 259, 485, 529 nm; fluorescence (DMF)
Fλmax (λEx) ) 583 (564) nm; MS (EI, 70 eV) m/z (%) 494 (22),
493 (57, [M]+), 393 (11), 207 (47), 137 (46), 123 (74), 119 (50),
91 (37); HRMS (EI, 70 eV) calcd for C29H19N9 m/z 493.17579
(M+), found m/z 493.17789 ( 2 ppm.
Acknowledgment. This work was supported by the Deutsche
Forschungsgemeinschaft. A.B. thanks the state of Mecklenburg-
Vorpommern for a scholarship. S.A. acknowledges scholarships
from the Gottlieb Daimler- and Karl Benz-Stiftung and the
DAAD. N.N.R.S. is grateful to the Dr. Gerda-von-Mach-
Geda¨chtnisstiftung for a scholarship.
Note Added after ASAP Publication. Figure 18 was
replaced, the corrected version was published June 4, 2008.
5,7-Diphenyl-3,9-ditolylhexaazaanthracene (4b). The reaction
of 4-toluoylchloride (15.50 g, 100.0 mmol), 1,3-diaminobenzene
(5.40 g, 50.0 mmol), and NEt3 (10.10 g, 100.0 mmol) in 200 mL
of dichloromethane afforded, following the procedure as given for
the synthesis of 6a, N,N′-1,3-phenylenebis(4-methylbenzamide):
Yield 9.60 g (56%), colorless prisms (ethanol), mp 216-217 °C;
IR (KBr) ν˜ ) 3326 (m), 1646, 1607, 1541, 1490 (s), 1324, 1302
(m), 839, 777, 746, 679 (m); 1H NMR (DMSO-d6, 300 MHz) δ )
2.39 (s, 6 H), 7.27-7.35 (m, 5 H), 7.47-7.50 (m, 2 H), 7.88-7.91
(m, 4 H), 8.31-8.32 (m, 1 H), 10.20 (s, 2 H); 13C NMR (DMSO-
d6, 150 MHz) δ ) 21.0, 113.0, 116.0, 127.7, 128.5, 128.9, 132.1,
139.4, 141.5, 165.3; MS (EI, 70 eV) m/z (%) ) 344 (0.2, [M]+),
Supporting Information Available: X-ray data and crystal-
lographic statistics for 3g, experimental procedures for inter-
mediates, spectroscopic results of the new compounds, copies
of NMR spectra, the tables of computed optimized geometric
parameters and energies, 13C NMR chemical shifts, natural
charge analyses for all species. This material is available free
of charge via the Internet at http://pubs.acs.org.
JO8005123
J. Org. Chem. Vol. 73, No. 13, 2008 5063