Synthesis of a tetraazulene porphodimethene analogue

Article abstract of DOI:10.1021/jo901959k

(Chemical Equation Presented) Substituted calix[4]azulenes were prepared by reacting 6-alkylazulenes with paraformaldehyde in the presence of florisil. Hydride abstraction of a calix[4]azulene with Ph₃CPF₆ afforded a tetraazulene ana

Full text of DOI:10.1021/jo901959k

pubs.acs.org/joc
remarkable chemical properties, including the ability to form
Synthesis of a Tetraazulene Porphodimethene
Analogue†
organometallic derivatives8 and to undergo selective oxidation
reactions.9,10 The data demonstrate that the replacement of one
pyrrole moiety with a cyclopentadiene or indene unit does not
greatly diminish the aromatic properties of the system. In
principle, 2, 3, or even 4 of the nitrogens could be replaced by
carbons to give bridged annulene structures that could give
further insights into the nature of porphyrinoid aromaticity.11
The hydrocarbon porphyrin analogue “quatyrin” (3) is the
ultimate goal of these studies, but a synthesis of this “Holy
Grail” has not yet been achieved.11 Nevertheless, a series of
dicarbaporphyrinoids have been synthesized with azulene and
indene subunits.11-15 Dicarbaporphyrins with opposite carbo-
cyclic subunits have proven to be somewhat unstable,11,12 but
recent reports on the synthesis of adjacent dicarbaporphyrinoid
systems have shown that structures like 4 and 5 are quite robust
and show significant diatropic properties.14,15
Timothy D. Lash,* Jessica A. El-Beck, and Denise A. Colby
Department of Chemistry, Illinois State University, Normal,
Illinois 61790-4160
tdlash@ilstu.edu
Received September 11, 2009
Substituted calix[4]azulenes were prepared by reacting 6-
alkylazulenes with paraformaldehyde in the presence of flori-
sil. Hydride abstraction of a calix[4]azulene with Ph3CPF6
afforded a tetraazulene analogue of the porphodimethenes.
The aromatic characteristics of porphyrins (1) are commonly
attributed to the presence of [18]annulene substructures
(emphasized in bold) within the macrocycle.1,2 However, the
system has a total of 26π electrons and other models have been
invoked to explain the aromaticity and diatropic properties of
these structurally unique natural products.3,4 To gain a further
understanding of porphyrinoid aromaticity, as well as to explore
the chemistry of related macrocyclic systems, conjugated macro-
cycles related to carbaporphyrins 2 have been synthesized.5,6
Carbaporphyrins show fully aromatic characteristics and their
proton NMR spectra commonly show the interior CH reso-
nance upfield near -7 ppm.7 These structures also show
† Part 53 in the series Conjugated Macrocycles Related to the Porphyrins.
(1) Vogel, E. J. Heterocycl. Chem. 1996, 33, 1461–1487.
(2) Lash, T. D. Synlett 2000, 279–295.
(3) Cyranski, M. K.; Krygowski, T. M.; Wisiorowski, M.; Hommes,
N. J. R.; van, E.; Schleyer, P. v. R. Angew. Chem., Int. Ed. 1998, 37, 177–180.
(4) (a) Aihara, J.-I.; Kimura, E.; Krygowski, T. M. Bull. Chem. Soc. Jpn.
2008, 81, 826–835. (b) Aihara, J.-i. J. Phys. Chem. A 2008, 112, 5305–5311.
(5) Lash, T. D. In The Porphyrin Handbook; Kadish, K. M., Smith, K. M.,
Guilard, R., Eds.; Academic Press: San Diego, CA, 2000; Vol. 2, pp 125-199.
(6) Lash, T. D. Eur. J. Org. Chem. 2007, 5461–5481.
(7) (a) Lash, T. D. Angew. Chem., Int. Ed. Engl. 1995, 34, 2533–2535. (b)
Lash, T. D.; Hayes, M. J. Angew. Chem., Int. Ed. Engl. 1997, 36, 840–842. (c)
Lash, T. D.; Hayes, M. J.; Spence, J. D.; Muckey, M. A.; Ferrence, G. M.;
Szczepura, L. F. J. Org. Chem. 2002, 67, 4860–4874.
Many partially or fully conjugated systems are known that
share the same tetrapyrrolic framework as porphyrins but
(8) (a) Muckey, M. A.; Szczepura, L. F.; Ferrence, G. M.; Lash, T. D.
Inorg. Chem. 2002, 41, 4840–4842. (b) Lash, T. D.; Rasmussen, J. M.;
Bergman, K. M.; Colby, D. A. Org. Lett. 2004, 6, 549–552. (c) Lash, T. D.;
Colby, D. A.; Szczepura, L. F. Inorg. Chem. 2004, 43, 5258–5267.
(9) (a) Hayes, M. J.; Spence, J. D.; Lash, T. D. Chem. Commun. 1998,
2409–2410. (b) Lash, T. D.; Muckey, M. A.; Hayes, M. J.; Liu, D.; Spence,
J. D.; Ferrence, G. M. J. Org. Chem. 2003, 68, 8558–8570.
(10) Carbaporphyrin ketals have also been shown to be promising agents
in the treatment of leishmaniasis. See: Morganthaler, J. B.; Peters, S. J.;
~
Cedeno, D. L.; Constantino, M. H.; Edwards, K. A.; Kamowski, E. M.;
Passini, J. C.; Butkus, B. E.; Young, A. M.; Lash, T. D.; Jones, M. A. Bioorg.
Med. Chem. 2008, 16, 7033–7038.
(11) Lash, T. D.; Romanic, J. L.; Hayes, M. J.; Spence, J. D. Chem.
Commun. 1999, 819–820.
8830 J. Org. Chem. 2009, 74, 8830–8833
Published on Web 10/26/2009
DOI: 10.1021/jo901959k
r
2009 American Chemical Society

Lash et al.
JOCNote
are nonaromatic. These include porphomethenes 6, porpho-
dimethenes 7,16 phlorins,17 isophlorins,18 isoporphyrins 8,19
and the formally antiaromatic didehydroporphyrins 9.20
Although initial interests in hydroporphyrins like 6 and 7
primarily related to their intermediacy in porphyrin syn-
theses,21 gem-disubstituted derivatives have been widely
studied in recent years for molecular recognition and anion
binding studies.22 It is worth noting that these systems are
generally more stable in their protonated forms. Some years
ago we reported a synthesis of calix[4]azulene 10a, a non-
conjugated macrocycle that shares the same carbon skeleton
of quatyrin.23 Although the formation of a quatyrin-like
system from 10a is unlikely, the preparation of structures
resembling 6-8 is far more plausible. In this report, we
further explore the synthesis of calixazulenes and their
conversion into hydroporphyrin analogues.
SCHEME 1
The synthesis of porphyrins commonly relies on the ability
of pyrroles to undergo electrophilic substitution at the
R-positions and thereby generate the required carbon-
carbon bonds to form the macrocycle.21 Azulene (11a) also
readily undergoes electrophilic substitution at the equivalent
1,3-positions and this property can be used to synthesize
porphyrin analogue systems.12,24-29 This principle was used
to generate tripyrrane analogues that could be used in the
synthesis of azuliporphyrins and related systems.12,24,25 In
addition, a one-pot synthesis of tetraarylazuliporphyrins
was developed where mixtures of azulene, pyrrole, and
arylaldehydes could be condensed in the presence of
BF3 Et2O, followed by oxidation with DDQ, to directly
3
afford the azuliporphyrins.26-28 However, initial attempts to
(12) Graham, S. R.; Colby, D. A.; Lash, T. D. Angew. Chem., Int. Ed.
2002, 41, 1371–1374. The synthesis of resorcinol-derived dicarbaporphyr-
inoids has also been noted: Xu, L.; Lash, T. D. Tetrahedron Lett. 2006, 47,
8863–8866.
prepare calix[4]azulene 10a from azulene proved to be diffi-
cult due to poor yields and problems encountered in puri-
fication.23 Reaction of 11a with paraformaldehyde in the
presence of acid catalysts gave moderate yields of impure 10a
that was contaminated with the related calix[5]azulene, and
attempts to purify the product by column chromatography
on silica or alumina led to extensive degradation.23 During
the course of these investigations, stepwise routes to 10a were
considered. Azulene reacts with phosphorus oxychloride and
DMF at 80 °C to give dialdehyde 12,30 and this was reduced
with sodium borohydride in an attempt to form dicarbinol
13. However, these reductions gave diazulene dialcohol 14 as
the major product. This result was unexpected but can be
rationalized by the mechanism shown in Scheme 1. Follow-
ing initial reduction to give 13, the formation of a carboca-
tion species could be envisaged and subsequent ipso-
substitution onto a second dicarbinol and elimination of
formaldehyde would give the bridged diazulene 14.
Although this product could be identified by NMR spectro-
scopy, it could not be isolated in pure form. Attempts to
purify 14 by chromatography on alumina or silica again led
to decomposition and immediate color changes (blue to
pink) were evident. Due to these difficulties, alternative
materials for carrying out chromatographic purifications
were explored. Florisil proved to give remarkable results,
although this magnesium silicate did not allow us to isolate
(13) For examplesofrelateddoubly N-confused porphyrins, see: (a) Furuta,
H.; Maeda, H.; Osuka, A. J. Am. Chem. Soc. 2000, 122, 803–807. (b) Maeda, H.;
Osuka, A.; Furuta, H. J. Am. Chem. Soc. 2003, 125, 15690–15691.
(14) Lash, T. D.; Colby, D. A.; Idate, A. S.; Davis, R. N. J. Am. Chem.
Soc. 2007, 129, 13801–13802.
(15) Zhang, Z.; Ferrence, G. M.; Lash, T. D. Org. Lett. 2009, 11, 101–104.
(16) (a) Fontecave, M.; Battioni, J. P.; Mansuy, D. J. Am. Chem. Soc.
1984, 106, 5217–5222. (b) Burns, D. H.; Li, Y.-H.; Shi, D. C.; Caldwell, T. M.
J. Org. Chem. 2002, 67, 4536–4546.
(17) (a) Woodward, R. B. Ind. Chim. Belge 1962, 27, 1293–1308. (b)
Whitlock, H. W.; Oester, M. Y. J. Am. Chem. Soc. 1973, 95, 5738–5741. (c)
Jeandon, C.; Krattinger, B.; Ruppert, R.; Callot, H. J. Inorg. Chem. 2001, 40,
3149–3153. (d) O’Brien, A. Y.; McGann, J. P.; Geier, G. R. III J. Org. Chem.
2007, 72, 4084–4092.
(18) Liu, C.; Shen, D.-M.; Chen, Q.-Y. J. Am. Chem. Soc. 2007, 129,
5814–5815.
(19) (a) Barkigia, K. M.; Renner, M. W.; Xie, H.; Smith, K. M.; Fajer, J.
J. Am. Chem. Soc. 1993, 115, 7894–7895. (b) Gentemann, S.; Leung, S. H.;
Smith, K. M.; Fajer, J.; Holten, D. J. Phys. Chem. 1995, 99, 4330–4334.
(20) Yamamoto, Y.; Yamamoto, A.; Furuta, S.; Horie, M.; Kodama, M.;
Sato, W.; Akiba, K.; Tsuzuki, S.; Uchimaru, T.; Hashizume, D.; Iwasaki, F.
J. Am. Chem. Soc. 2005, 127, 14540–14541.
(21) Lash, T. D. Chem.;Eur. J. 1996, 2, 1197–1200.
ꢀ
(22) (a) Bucher, C.; Seidel, D.; Lynch, V.; Kral, V.; Sessler, J. L. Org. Lett.
ꢀ
2000, 2, 3103–3106. (b) Bernatkova, M.; Dvorakova, H.; Andrioletti, B.;
Kral, V.; Bour, P. J. Phys. Chem. A 2005, 109, 5518–5526.
ꢀ
ꢀ
ꢀ
ꢀ
(23) Colby, D. A.; Lash, T. D. J. Org. Chem. 2002, 67, 1031–1033.
(24) Lash, T. D.; Colby, D. A.; Graham, S. R.; Chaney, S. T. J. Org.
Chem. 2004, 69, 8851–8864.
(25) Lash, T. D.; El-Beck, J. A.; Ferrence, G. M. J. Org. Chem. 2007, 72,
8402–8415.
(26) Colby, D. A.; Lash, T. D. Chem.;Eur. J. 2002, 8, 5397–5402.
(27) Lash, T. D.; Colby, D. A.; Ferrence, G. M. Eur. J. Org. Chem. 2003,
4533–4548.
(28) El-Beck, J. A.; Lash, T. D. Eur. J. Org. Chem. 2007, 3981–3190.
(29) The original synthesis of azuliporphyrins involved a “3þ1” conden-
sation of 1,3-azulene dicarbaldehyde with a tripyrrane. See: Lash, T. D.;
Chaney, S. T. Angew. Chem., Int. Ed. 1997, 36, 839–840.
(30) Hafner, K.; Bernhard, C. Liebigs Ann. Chem. 1959, 625, 108–123.
J. Org. Chem. Vol. 74, No. 22, 2009 8831

Lash et al.
JOCNote
SCHEME 2
SCHEME 3
dialcohol 14. Instead, when crude 14 was loaded onto a
florisil column and eluted with dichloromethane, pure calix-
[4]azulene was collected as the only product. It was this
observation that led us to use florisil as a catalyst for the
formation of 10a. When azulene was stirred with 4 equiv of
paraformaldehyde and florisil in dichloromethane, and the
catalyst was removed by suction filtration, calix[4]azulene
was obtained in pure form in 74% yield.23 Unfortunately,
these conditions failed to give macrocyclic products with
other aldehydes and ketones.23
The conjugated units in hydroporphyrins such as 6 and 7
contain dipyrromethene components 15 (Scheme 2) that are
stabilized in their protonated forms by charge delocalization.
Similar azulene-containing conjugated systems 16 can be
obtained by treating di- or triazulenylmethanes with triphe-
nylcarbenium hexafluorophosphate.31 This type of conju-
gated diazulene system is also a component of the recently
described porphyrin analogue adj-diazuliporphyrin 5.15 In
an attempt to prepare conjugated tetraazulenes, 10a was
treated with 1 equiv of Ph3CBF4 in acetonitrile. The original
calixazulene and the resulting product were both rather
insoluble, and while the proton NMR data for the product
were consistent with the formation of 17a further investiga-
tions proved to be impractical. In other studies, we have
found that the presence of tert-butyl substituents aids in
increasing the solubility of azuliporphyrin systems as well as
stabilizing tropylium-type characteristics.15,25,28,32 With this
in mind, the reaction of 6-substituted azulenes 11b-d with
paraformaldehyde was investigated. 6-tert-Butylazulene
(11b)25,31 reacted with 4 equiv of paraformaldehyde in the
presence of florisil to give the tetra-tert-butylcalixazulene
10b in 68% yield. The same conditions were used with
6-methylcalixazulene (11c),33 but in this case poorer yields
of 10c were obtained. Following recrystallization from to-
luene, 10c was isolated in 17% yield. The poorer results in
this case can be attributed to the acidity of the methyl
substituent, which can lead to side reactions.33 However, 6-
phenylazulene (11d)25 gave very poor results and complex
mixtures were formed that appeared to contain 10d, a related
calix[5]azulene, and other possibly oligomeric species. It is
not clear why 6-phenylazulene no longer allows selective
formation of the calix[4]azulene under the solid phase con-
ditions. Nevertheless, we also investigated the synthesis of
13C-labeled calix[4]azulene 18 from (13CH2O)n. However, as
4 equiv of paraformaldehyde was used in the original syntheses,
we sought to modify the conditions so that less of the labeled
precursor was required. By extending the reaction time from
90 min to 6 h, we were able to form the labeled calixazulene in
40% yield by reacting 6-tert-butylazulene with 1.2 equiv of
13C-labeled paraformaldehyde. Therefore, a fairly efficient
route to labeled calixazulenes is available.
Attempts to oxidize calixazulene 10a or 10b with DDQ
gave no identifiable products. However, hydride abstraction
with Ph3CBF4 or Ph3CPF6 gave more useful results. Both
reagents could be used to generate conjugated tetraazulenes,
but the best results were usually obtained with the hexa-
fluorophosphate. In principle, the tetra-tert-butylcalixazu-
lene could form porphomethene analogue 19, porpho-
dimethene analogue 20, isoporphyrin analogue 21, or dide-
hydroporphyrin analogue 22 (Scheme 3). Unfortunately,
many of the reactions gave mixtures of products. Reaction
of 10b with 1 equiv of the carbenium reagent gave samples
that contained 19 but this system could not be isolated in
pure form. Reactions with 5 equiv of the reagent at room
temperature in acetonitrile gave the porphodimethene ana-
logue 20 and this could be isolated by precipitation with
ether. However, when 10b was heated with the carbenium
reagent, impure isoporphyrin analogue 21 was formed but
again this product could not be purified. None of the results
were consistent with the formation of tetracation 22, but this
system is likely to be antiaromatic and therefore unstable.
Finally, all attempts to form conjugated tetraazulenes from
tetramethylcalix[4]azulene 10c were unsuccessful and com-
plex mixtures were generated in reactions with the carbenium
reagents. This is not particularly surprising as the methyl
substituent is likely to deprotonate following hydride ab-
straction.
The porphodimethene analogue 20 gave deep blue solu-
tions in acetonitrile and the UV-vis spectrum showed a
strong absorption (molar absorptivity ca. 105) at 616 nm
(Figure 1). This spectrum is bathochromically shifted com-
pared to porphodimethenes 7, which show a similarly strong
(31) Ito, S.; Morita, N.; Asao, T. Bull. Chem. Soc. Jpn. 1995, 68, 1409–
1436.
(32) Zhang, Z.; Ferrence, G. M.; Lash, T. D. Org. Lett. 2009, 11, 1249–
1252.
(33) Rudolf, K.; Robinette, D.; Koenig, T. J. Org. Chem. 1987, 52, 641–
647.
8832 J. Org. Chem. Vol. 74, No. 22, 2009

Lash et al.
JOCNote
structure 22 was reported. Didehydroporphyrins 9 have only
been isolated as highly crowded nonplanar structures with 12
peripheral substituents,20 and the tetraazulene tetracation is
essentially an analogue of this highly distorted system. This
interesting result further extends the structural diversity of
azulene-containing macrocycles. However, unlike the diazuli-
porphyrin system 5,15 the tetraazulene tetracations lack any
porphyrin-type aromatic characteristics.34 The intermediary
conjugated calixazulenes observed in our current studies
indicate that valuable properties can be obtained at these
oxidation levels as well, but further stabilization (e.g., by
introducing gem-dialkyl groups at the meso-bridges)
will be needed to allow these systems to achieve their full
potential.
FIGURE 1. The UV-vis spectrum of 20b in acetonitrile.
Experimental Section
Porphodimethene Analogue 20b. Triphenylcarbenium hexa-
fluorophosphate (49.5 mg) was added to a stirred suspension of
calixazulene 10b (20 mg, 0.025 mmol) in 10 mL of anhydrous
acetonitrile, and the resulting solution was stirred for 30 min.
The solution initially turned dark green and then quickly
became dark blue. The volume of the solution was reduced to
ca. 1 mL on a rotary evaporator, diluted with anhydrous ether,
and placed in a freezer for 1 h. The dark precipitate was filtered
and dried in vacuo to give the bis-hexafluorophosphate salt
(14.3 mg, 0.13 mmol, 53%) as a dark solid, mp >300 °C;
UV-vis (1% Et3N-CHCl3) λmax (log ε) 616 (5.05), 701 nm
(sh, 4.53); 1H NMR (CD3CN) δ 1.56 (36H, s), 4.72 (4H, s), 8.28
(4H, dd, J = 1.9, 10.6 Hz), 8.34 (4H, dd, J = 1.9, 10.5 Hz), 8.76
(4H, d, J = 10.6 Hz), 8.97 (4H, s), 9.13 (4H, d, J = 10.5 Hz), 9.17
(2H, s); 13C NMR (CDCl3) δ 25.3, 31.8, 70.6, 131.0, 135.0, 135.6,
137.6, 138.1, 138.5, 139.7, 141.6, 148.5, 150.2, 171.0. Anal. Calcd
FIGURE 2. The 500 MHz proton NMR spectrum of porphodi-
methene analogue 20b in CD3CN.
peak near 500 nm. The proton and carbon-13 NMR data
confirmed the identity of this system. In the proton NMR
spectrum of 20, the methylene bridge protons gave a 4H
singlet at 4.72 ppm, while the conjugated methine bridges
gave a 2H singlet at 9.17 ppm (Figure 2). The four azulene
units are equivalent but asymmetrical and give rise to 5
resonances: a 4H singlet for the internal CHs at 8.96 ppm,
two 4H doublets (J = 10.6 Hz) at 8.76 and 9.13 ppm, and two
4H doublets of doublets near 8.3 ppm. The conjugated
tetraazulene was reasonably stable in solution, but addition
of small amounts of calix[4]azulene 10b gave rise to complex
mixtures of products. Some reaction was noted immediately
after the addition, but further changes were seen over a
period of 1-2 h. These data indicate that intermolecular
hydride abstraction is taking place to give mixtures including
porphomethene analogue 19. The conjugated system was
also analyzed by electrospray ionization mass spectrometry.
This gave a cluster of peaks near m/z 391 that corresponded
to the expected doubly charged ion.
for C60H62P2F12 H2O: C, 66.05; H, 5.91. Found: C, 65.82; H,
3
6.00.
Acknowledgment. This material is based upon work sup-
ported by the National Science Foundation under Grant No.
CHE-0616555 and the Petroleum Research Fund, adminis-
tered by the American Chemical Society. Funding for a 500
MHz NMR spectrometer was also provided by the National
Science Foundation under grant no. CHE-0722385.
Following the completion of this study,34 the synthesis
of a meso-substituted tetraazulene tetracation related to
Supporting Information Available: Synthetic procedures
for 10b, 10c, and 18, selected proton and carbon-13 NMR
spectra, and MS data. This material is available free of charge
via the Internet at http://pubs.acs.org.
(34) Sprutta, N.; Mackowiak, S.; Kocik, M.; Szterenberg, L.; Lis, T.;
Latos-Grazynski, L. Angew. Chem., Int. Ed. 2009, 48, 3337–3341.
J. Org. Chem. Vol. 74, No. 22, 2009 8833