Syntheses and reactivity of meso-unsubstituted azuliporphyrins derived from 6-tert-butyl- and 6-phenylazulene

Article abstract of DOI:10.1021/jo701523s

(Chemical Equation Presented) A series of six azuliporphyrins with substituents on the seven-membered ring were prepared by two different "3 + 1" routes from 6-tert-butyl- and 6-phenylazulene. The substituted azulenes can be converted into dialdehydes und

Full text of DOI:10.1021/jo701523s

Syntheses and Reactivity of meso-Unsubstituted Azuliporphyrins
Derived from 6-tert-Butyl- and 6-Phenylazulene†
Timothy D. Lash,* Jessica A. El-Beck, and Gregory M. Ferrence
Department of Chemistry, Illinois State UniVersity, Normal, Illinois 61790-4160
tdlash@ilstu.edu
ReceiVed July 23, 2007
A series of six azuliporphyrins with substituents on the seven-membered ring were prepared by two
different “3 + 1” routes from 6-tert-butyl- and 6-phenylazulene. The substituted azulenes can be converted
into dialdehydes under Vilsmeier-Haack conditions, and these react with tripyrranes in the presence of
TFA in CH2Cl2 to give azuliporphyrins in excellent yields. Alternatively, tripyrrane analogues can be
prepared by reacting the substituted azulenes with an acetoxymethylpyrrole in the presence of acetic
acid, and following a deprotection step, these condensed with a pyrrole dialdehyde to give the related
azuliporphyrins in 45-51% yield. Five of the azuliporphyrins were sufficiently soluble in CDCl3 to
afford high-quality proton and carbon-13 NMR data. The internal CH and NH resonances were observed
near 3 ppm, although the precise values were dependent upon substituent effects. The presence of a
tert-butyl group on the azulene moiety slightly enhanced the diatropicity of the macrocycle compared to
the phenyl-substituted azuliporphyrins. Polar solvents also increased the downfield shifts to the external
protons by stabilizing the dipolar resonance contributors that are responsible for the carbaporphyrinoid
aromatic character. A tert-butyl-substituted azuliporphyrin also gave X-ray quality crystals, and this allowed
the first structural analysis of a free base azuliporphyrin to be conducted. The macrocycle is near planar,
and the azulene unit was only tilted out of the plane by 7.4°. An analysis of the bond lengths suggests
that a 17 atom delocalization pathway significantly contributes to the aromatic properties of this system.
Protonation of azuliporphyrins affords dications with enhanced diamagnetic ring currents where the internal
CH shifts to ca. -3 ppm. Again, the chemical shifts are influenced by the substituents and the presence
of an electron-donating tert-butyl group on the azulene subunit increases the macrocyclic diatropicity.
Two of the substituted azuliporphyrins were reacted with nickel(II) acetate or palladium(II) acetate in
DMF to give the corresponding organometallic derivatives, and these stable complexes were isolated in
excellent yields. Addition of pyrrolidine to NMR solutions of 23-substituted azuliporphyrins 19
demonstrated that nucleophilic addition products were present in equilibrium with the parent porphyrinoids,
but these adducts are less favored than for azuliporphyrins lacking the 23-substituents. Although nucleophilic
attack of a peroxide anion is believed to be the first step in the conversion of azuliporphyrins to
benzocarbaporphyrins with t-BuOOH and KOH, the tert-butyl or phenyl substituents in azuliporphyrins
19a and 19b did not inhibit this chemistry. Two benzocarbaporphyrin products were isolated and
characterized in each case, and mechanisms are proposed to explain the origins of these oxidative ring
contraction products.
Introduction
dialdehydes with tripyrrolic intermediates (tripyrranes) using the
“3 + 1” variant of the MacDonald condensation.1-3 Using this
approach, carbaporphyrinoid systems such as 1-4 (Chart 1)4-7
Porphyrin analogues with carbocyclic rings in place of one
of the usual pyrrole units can be synthesized by reacting
* To whom correspondence should be addressed.
† Conjugated Macrocycles Related to the Porphyrins. 45. For part 44, see:
El-Beck, J. A.; Lash, T. D. Eur. J. Org. Chem. 2007, 3981-3990.
(1) Lash, T. D. Chem. Eur. J. 1996, 2, 1197-1200.
(2) Lash, T. D. Synlett 2000, 279-295.
10.1021/jo701523s CCC: $37.00 © 2007 American Chemical Society
Published on Web 10/05/2007
8402
J. Org. Chem. 2007, 72, 8402-8415

Syntheses and ReactiVity of meso-Unsubstituted Azuliporphyrins
CHART 1
SCHEME 1
with diverse chemical and spectroscopic properties are easily
prepared. Although benzocarbaporphyrins 1 and oxybenzipor-
phyrins 2 have strongly diatropic characteristics that are
comparable to true porphyrins,4,5 this property is reduced in
tropiporphyrins 3 due to distortions resulting from the presence
of a seven-membered ring.6 Azuliporphyrins 4 have even smaller
diatropic ring currents that result from cross-conjugation by the
azulene subunit.7,8 This factor seems to be overcome to a certain
extent by dipolar canonical forms such as 4′ that give the system
tropylium ion character while simultaneously allowing for a
carbaporphyrinoid 18π electron delocalization pathway (Scheme
1).7,8 However, the effect is limited due to the associated need
SCHEME 2
for charge separation. Addition of acid leads to diprotonation
2+
and the resulting dications 4H2
show greatly enhanced
diatropicity where the internal CH resonance shifts upfield to
ca. -3 ppm (Scheme 1).7,8 This increased aromatic character is
attributed to an increased favorability for resonance structures
2+
such as 4′H2 which now aid in charge delocalization.7,8
Azuliporphyrins readily form organometallic derivatives with
nickel(II), palladium(II), and platinum(II) salts,9,10 and these
metalation reactions closely parallel results obtained for N-
confused porphyrins.11,12 This system also undergoes an unusual
oxidative metalation reaction with copper(II) salts,13 and treat-
ment of 4a with peroxides under basic conditions leads to the
related benzocarbaporphyrins 1a-c.8,14 The latter reaction is
believed to be triggered by nucleophilic attack at position 23
on the azulene subunit. Three synthetic routes to azuliporphyrins
have been developed. In the initial studies, azulene dicarbal-
dehyde 5 was reacted with tripyrrane 6 in TFA-CH2Cl2,
followed by oxidation with DDQ or ferric chloride (Scheme 2)
to give azuliporphyrins 4.7,8 Subsequently, tripyrrane analogue
7 was prepared by reacting azulene (8) with 2 equiv of an
acetoxymethylpyrrole in refluxing acetic acid and isopropyl
alcohol (Scheme 3).8,15 Cleavage of the tert-butyl ester protective
groups with TFA, condensation with a pyrrole dialdehyde 10,
and oxidation with DDQ or FeCl3 gave good yields of
azuliporphyrin 11.8,15 This route had the advantage of allowing
the synthesis of heteroazuliporphyrins and a related dicarbap-
orphyrinoid system.8,15 In the third route, meso-tetraarylazuli-
porphyrins 12 were shown to be accessible by reacting azulene,
pyrrole, and an aromatic aldehyde in a 1:3:4 ratio in the presence
of BF3‚Et2O in chloroform, followed by oxidation with DDQ
(Scheme 4).16,17 Although meso-substituted azuliporphyrins 12
(3) Lash, T. D. In The Porphyrin Handbook; Kadish, K. M., Smith, K.
M., Guilard, R., Eds.; Academic Press: San Diego, 2000; Vol. 2, pp 125-
199.
(4) (a) Lash, T. D.; Hayes, M. J. Angew. Chem., Int. Ed. Engl. 1997, 36,
840-842. (b) 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. (c)
Liu, D.; Lash, T. D. J. Org. Chem. 2003, 68, 1755-1761.
(5) (a) Lash, T. D. Angew. Chem., Int. Ed. Engl. 1995, 34, 2533-2535.
(b) Richter, D. T.; Lash, T. D. Tetrahedron 2001, 57, 3659-3673.
(6) (a) Lash, T. D.; Chaney, S. T. Tetrahedron Lett. 1996, 37, 8825-
8828.
(7) Lash, T. D.; Chaney, S. T. Angew. Chem., Int. Ed. 1997, 36, 839-
840. For an example of an azulisapphyrin, see: Richter, D. T.; Lash, T. D.
J. Org. Chem. 2004, 69, 8842-8850.
(8) Lash, T. D.; Colby, D. A.; Graham, S. R.; Chaney, S. T. J. Org.
Chem. 2004, 69, 8851-8864.
(9) Graham, S. R.; Ferrence, G. M.; Lash, T. D. Chem. Commun. 2002,
894-895.
(10) Lash, T. D.; Colby, D. A.; Graham, S. R.; Ferrence, G. M.;
Szczepura, L. F. Inorg. Chem. 2003, 42, 7326-7338.
(11) Srinivasan, A.; Furuta, H. Acc. Chem. Res. 2005, 38, 10-20.
(12) Harvey, J. D.; Ziegler, C. J. Coord. Chem. ReV. 2003, 247, 1-19.
(13) Colby, D. A.; Ferrence, G. M.; Lash, T. D. Angew. Chem., Int. Ed.
2004, 43, 1346-1349.
(14) Lash, T. D. Chem. Commun. 1998, 1683-1684.
(15) Graham, S. R.; Colby, D. A.; Lash, T. D. Angew. Chem., Int. Ed.
2002, 41, 1371-1374.
(16) Colby, D. A.; Lash, T. D. Chem. Eur. J. 2002, 8, 5397-5402.
J. Org. Chem, Vol. 72, No. 22, 2007 8403

Lash et al.
SCHEME 3
SCHEME 5
mechanism of this unusual reaction. All of these issues were
addressed in the investigations of 23-substituted azuliporphyrins
described below.19
SCHEME 4
Results and Discussion
The key starting materials for these studies were 6-tert-
butylazulene (13a) and 6-phenylazulene (13b). Although both
of these compounds are known,20,21 an alternative procedure
was developed for the synthesis of these azulenes (Scheme 5).22
Reaction of 4-tert-butyl or 4-phenylpyridine with 1-bromobutane
in refluxing ethanol gave the corresponding pyridinium salts
14. These reacted with cyclopentadiene and sodium hydride in
DMF to afford the required azulenes 13. Treatment of 13a with
phosphorus oxychloride and DMF at 90 °C gave the related
dialdehyde 15a (Scheme 5). Vilsmeier-Haack formylation of
6-phenylazulene (13b) was carried out similarly, although a
mixture of dialdehyde 15b and monaaldehyde 15c were isolated
in this case (Scheme 5). These were easily separated by column
chromatography on silica.
The azulene dialdehydes 15a and 15b were reacted with
tripyrranes 6a or 6b to give a series of four azuliporphyrins
(16a,b and 17a,b) in 55-67% yield (Scheme 2). The reaction
was conducted using TFA-dichloromethane for the condensa-
tion and the final oxidation step involved shaking the reaction
solution with 0.1% aqueous ferric chloride solution for 5
min.23,24 Purification was easily accomplished by column
chromatography on grade 3 basic alumina and further recrys-
tallization from chloroform-hexanes. The alternative “3 + 1”
route shown in Scheme 3 was also investigated.15 As azulene
favors electrophilic substitution at the 1 and 3 positions, it is
possible to construct tripyrrane analogues by reacting 13 with
2 equiv of an acetoxymethylpyrrole 9 in the presence of a weak
are reasonably soluble in solvents such as CH2Cl2 and CHCl3,
meso-unsubstituted azuliporphyrins 4 and 11 are only sparingly
soluble in organic solvents, and this has made it difficult to
obtain high-quality NMR data for the free base forms of these
molecules.8 The internal CH and NH proton resonances provide
important data on the aromatic characteristics of carbaporphy-
rinoid systems, but these signals cannot be assigned with any
degree of confidence for azuliporphyrins 4 or 11.8 In one case,
the 21-CH resonance was tentatively identified in CDCl3
solutions near 2 ppm,8 but aggregation effects can also influence
these chemical shifts and this makes the result difficult to fully
interpret. Furthermore, it was not possible to obtain carbon-13
NMR data for the free base meso-unsubstituted azuliporphyrins.8
In order to further investigate the azuliporphyrin system, 23-
tert-butyl- and 23-phenyl-substituted azuliporphyrins were tar-
geted for synthesis. The presence of bulky substituents on the
azulene ring was expected to inhibit aggregation effects18 and
thereby overcome the problems that had previously been
encountered in obtaining high-quality NMR data. In addition,
the synthesis of these azuliporphyrins was expected to be little
affected by the presence of a substituent at this remote position
on the azulene moiety. Substitution at position 23 also allows
the macrocycle to retain a plane of symmetry, which simplifies
spectroscopic characterization. In addition, compact units such
as phenyl or tert-butyl groups sometimes aid in the isolation of
X-ray quality crystals4b,6b,9 and as no examples of free base
azuliporphyrins have been structurally characterized previously
this was an important consideration. Ring contraction reactions
to give benzocarbaporphyrins are believed to involve initial
nucleophilic attack at the 23-position14 and substituted azuli-
porphyrins could also provide insightful information on the
(19) These results were presented, in part, at the following meeting: 229th
National American Chemical Society Meeting, San Diego, California, March
2005 (El-Beck, J. A.; Lash, T. D. Book of Abstracts; American Chemical
Society: Washington, DC, 2005; CHED 546).
(20) Ito, S.; Morita, N.; Asao, T. Bull. Chem. Soc. Jpn. 1995, 68, 1409-
1436.
(21) Bergmann, E. D.; Ikan, R. J. Am. Chem. Soc. 1956, 78, 1482-
1485.
(22) These procedues were adapted from a previously reported synthesis
of 6-methylazulene: Rudolf, K.; Robinette, D.; Koenig, T. J. Org. Chem.
1987, 52, 641-647.
(17) Lash, T. D.; Colby, D. A.; Ferrence, G. M. Eur. J. Org. Chem.
2003, 4533-4548.
(18) Jiao, W.; Lash, T. D. J. Org. Chem. 2003, 68, 3896-3901.
(23) Lash, T. D.; Richter, D. T.; Shiner, C. M. J. Org. Chem. 1999, 64,
7973-7982.
(24) Richter, D. T.; Lash, T. D. Tetrahedron Lett. 1999, 40, 6735-6738.
8404 J. Org. Chem., Vol. 72, No. 22, 2007

Syntheses and ReactiVity of meso-Unsubstituted Azuliporphyrins
FIGURE 1. UV-vis spectra of azuliporphyrin 16a. Black line: free
base in 1% Et3N-chloroform. Red line: dication 16aH22+ in 1% TFA-
chloroform.
acid catalyst. Azulene had been shown to react with 9 in
refluxing AcOH-i-PrOH to give azulitripyrrane 7 in 59% yield.
However, reactions of 13a or 13b with 9 were very solvent
dependent, and poor results were obtained under the original
conditions. The reactions were attempted in a number of
different solvents, and the best yields of azulitripyrrane 18a
(55%) were obtained by reacting 13a and 9 with acetic acid in
ethyl acetate. However, these conditions did not work well for
the phenyl-substituted version. The best results in this case were
obtained when 13b was reacted with 9 in refluxing acetic acid-
ethanol, and following chromatography and recrystallization,
azulitripyrrane 18b was isolated in 49% yield. The tripyrrane
analogues 18 were treated with TFA for 10 min, diluted with
dichloromethane, and reacted with pyrrole dialdehyde 10.
Following oxidation with ferric chloride, column chromatro-
gaphy on grade 3 basic alumina, and recrystallization, azuli-
porphyrins 19a and 19b were obtained in 45-51% yield.
Therefore, good results were obtained using 6-tert-butyl- and
6-phenylazulene in both of the “3 + 1” routes to azuliporphyrins.
All of the azuliporphyrins obtained in these studies showed
similar UV-vis spectra to the previously synthesized 23-
unsubstituted azuliporphyrins 4 and 11. For instance, azulipor-
phyrin 16a in 1% Et3N-CHCl3 showed four medium absorption
bands at 358, 397, 447, and 474 nm and a broad absorption
centered on 665 nm (Figure 1). These spectra are very different
from fully aromatic porphyrinoids such as 1 and 2 which show
strong Soret bands and several smaller Q bands, and this reflects
the unusual electronic features of the azuliporphyrin system.
However, it is worth noting that the UV-vis spectra of the
individual azuliporphyrins showed significant variations in the
relative intensities and specific wavelengths for the absorption
FIGURE 2. 400 MHz proton NMR spectrum of azuliporphyrin 16a
in CDCl3.
some insights into the aromatic ring current. In a nonaromatic
benziporphyrin,25 these resonances are present at 2.4 ppm but
in fully aromatic porphyrins the methyl resonances are typically
near 3.6 ppm. In 16a, the methyl resonance appears at an
intermediary value of 3.02 ppm. Most importantly, two broad
resonances were observed between 2.8 and 2.9 ppm integrating
for a total of 2H (Figures 2 and 3). These correspond to the
internal CH and NH protons, and significantly, the CH resonance
shows up approximately 1 ppm downfield from the value
tentatively assigned to azuliporphyrins 4. The upfield shift for
4 is probable due to aggregation effects as none of the other
spectroscopic features for 4 suggest that these azuliporphyrins
are more diatropic than 16a. When the NMR solution of 16a
was shaken with D2O, a single sharper 1H peak remained at
2.82 ppm (Figure 3). Although deuterium exchange of the NH
was expected, the CH undergoes an apparent upfield shift which
we tentatively attribute to minor conformational changes due
to hydrogen bonding interactions. The proton NMR spectrum
for 23-phenylazuliporphyrin 17a in CDCl3 showed the internal
protons as two broad 1H resonances at 3.23 and 3.30 ppm, and
again these were replaced by a single slightly upfield shifted
resonance following a D2O shake. The downfield shift to the
internal proton resonances indicates that the ring current for
17a is less than for 16a. The meso-proton resonances at 8.06
and 8.96 ppm and the azulene doublets at 7.81 and 9.29 ppm
are also all slightly less deshielded for 17a than the equivalent
resonances for 16a and even the methyl group resonance at 2.99
ppm is 0.03 ppm upfield from the value seen in 16a. The
opposite trends are seen for 12,13-diphenylazuliporphyrin 16b.
The internal CH and NH resonances were observed at 2.8 and
2.49 ppm, upfield from the values seen for 16a, while the
2+
bands. In 1% TFA-CHCl3, dications 16H22+, 17H22+ or 19H2
were generated that showed that showed stronger Soret-like
bands (Figure 1) that were consistent with increased porphyrin-
type characteristics.
Five of the six new azuliporphyrins were sufficiently soluble
in chloroform to give good proton and carbon-13 NMR data
for the free base species. Triphenylazuliporphyrin 17b gave
poorly resolved proton NMR spectra in CDCl3 due to aggrega-
tion, as had been the case for 23-unsubstituted azuliporphyrins
4 and 11. The proton NMR spectrum for 16a in CDCl3 showed
the meso-protons at two 2H singlets at 8.15 and 9.05 ppm, and
the azulene protons appeared as two 2H doublets at 7.89 and
9.33 ppm (Figure 2). These shifts are consistent with a moderate
diatropic ring current. The methyl substituents can also give
(25) Lash, T .D.; Chaney, S. T.; Richter, D. T. J. Org. Chem. 1998, 63,
9076-9088.
J. Org. Chem, Vol. 72, No. 22, 2007 8405

Lash et al.
FIGURE 4. Partial 400 MHz proton NMR spectrum of azuliporphyrin
2+
dication 16aH2 in TFA-CDCl3 showing the upfield and downfield
FIGURE 3. Details of the 400 MHz proton NMR spectrum of
azuliporphyrin 16a in CDCl3. (A) Internal CH and NH resonances for
free base azuliporphyrin 16a. (B) The same region following a D2O
shake. In addition to the expected deuterium exchange for the NH, the
CH sharpens and shifts slightly upfield. (C) Details of the downfield
region.
regions.
current. This inversion in the trend seen for the free base
azuliporphyrins is due to the electron-donating ethyl groups now
stabilizing the protonated carbaporphyrinoid system. Again, the
2+
2+
chemical shifts for 19aH2 and 19bH2 were similar to the
2+
2+
peripheral protons were all further deshielded. The meso-protons
gave two 2H resonances at 8.33 and 9.39 ppm, 0.18 and 0.34
ppm further downfield than the values for 16a, and the azulene
and methyl resonances also showed minor downfield shifts. The
presence of an electron-donating tert-butyl substituent on the
azulene ring helps stabilize the tropylium character in dipolar
resonance contributors 16a′ (Scheme 1), and this helps to
increase the contributions from these aromatic canonical forms.
The decreased electron-donating ability of the phenyl substit-
uents leads to the reduced diatropicity observed for 17a. In 16a
and 17a, the presence of electron-donating ethyl groups at the
12 and 13 positions would destabilize the carbaporphyrin anion
character in the dipolar resonance contributors and the phenyl
groups in this position for 16b has a beneficial rather than
detrimental effect. The data for 19a and 19b were similar to
their isomers 16a and 17a, respectively. Azuliporphyrins 16,
17a, and 19 all gave carbon-13 NMR spectra in CDCl3 that
were consistent with proposed structures and showed the
presence of a plane of symmetry. The meso-carbons gave rise
to two resonances with chemical shift values of 93-96 ppm
and 108-110 ppm.
Addition of one drop of TFA to NMR solutions of the
azuliporphyrins gave the aromatic dications. The internal CH
resonance for 16aH22+ was observed at -3.19 ppm (Figure 4),
while the values for 17a, 16b, and 17b were -2.91, -3.43,
and -2.69 ppm, respectively. The meso-protons were also
significantly shifted downfield and 16a shows two 2H reso-
nances at 9.54 and 10.39 ppm (Figure 4). The values of these
resonances in 17a, 16b and 17b were 9.50 and 10.39, 9.66 and
10.50, and 9.47 and 10.36 ppm, respectively. These and other
data from these spectra indicate that the diamagnetic ring
currents for this series increased in the sequence 17b < 17a <
16b < 16a. The tert-butyl groups in 16a and 16b clearly still
lead to an increased diatropicity compared to the related 23-
phenyl substituted azuliporphyrins by stabilizing the tropylium
ion character. However, the presence of phenyl groups at the
12,13-positions now leads to a decrease in the diamagnetic ring
isomeric species 16aH2 and 17aH2 and showed the same
general trends. Carbon-13 NMR spectra for the protonated
azuliporphyrins again showed the presence of a plane of
symmetry. In addition, the tert-butylazuliporphyrins 16a, 16b
and 19a all showed a downfield resonance near 173 ppm due
to the 23-carbon of the substituted tropylium species.
The proton NMR spectra of 16a and 17a were also examined
in acetone-d6, DMSO-d6, and DMF-d7. Although the internal
protons did not resolve in most of these spectra, useful trends
were noted for the external protons. For 16a, the meso-protons
were observed at 8.32 and 9.30, 8.33 and 9.48, and 8.42 and
9.47, for solutions in acetone-d6, DMSO-d6 and DMF-d7,
respectively. These downfield shifts are significantly increased
compared to CDCl3, particularly for the polar aprotic solvents
DMSO and DMF, and similar downfield shifts were also noted
for the azulene protons. Small downfield shifts were also
observed for the methyl resonances and the same trends were
observed for 17a. These data imply that the diatropic character
of these azuliporphyrins is solvent dependent, and the shifts are
presumably due to stabilization of the dipolar resonance
contributors by the more polar solvents. Proton NMR spectra
were also obtained in pyridine-d5, but these results were more
difficult to interpret. For 16a, the internal CH was noted at 3.36
ppm, and this downfield shift compared to spectra run in CDCl3
implies a decreased ring current effect. However, the meso-
protons were observed as two 2H singlets at 8.56 and 9.56 ppm,
values that were 0.4-0.5 ppm further downfield than is seen
for spectra in CDCl3, and a downfield shift was also noted for
one of the azulene doublets. Similar trends were seen for 17a,
16b, 17b, 19a, and 19b. The anomalous value for the 21-CH
may be due to small changes in the local environment or a minor
conformation change due to hydrogen bonding interactions with
the pyridine solvent. The remaining data suggest that pyridine
increases the diatropic ring current in these azuliporphyrins,
again by stabilizing the dipolar resonance contributors.
23-tert-Butylazuliporphyrin 16a crystallized from dichlo-
romethane-hexanes to give crystals that were suitable for X-ray
8406 J. Org. Chem., Vol. 72, No. 22, 2007

Syntheses and ReactiVity of meso-Unsubstituted Azuliporphyrins
CHART 2
of three atoms in the central cavities.4b,26 However, the related
tetraphenylbenzocarbaporphyrin 21 exhibits an indene tilt of
only 11.9° even though there are additional steric interactions
with the meso-phenyl substituents.17 Presumably packing forces
are a significant factor affecting deformation from planarity.
Nevertheless, the degree of planarity in these examples does
not correlate with the degree of aromatic character observed in
their proton NMR spectra. In 16a, the plane defined by atoms
C(2A), C(2B), C(2C), C(3B), and C(3A) has a 10.58(5)° tilt
relative to the mean macrocyclic plane and a 5.80(7)° tilt relative
to the plane defined by C(1), C(2), C(3), C(4), and C(21), which
appears to be attributable to packing forces associated with
contact between the azulene tert-butyl group and neighboring
molecules. A Mogul geometry check of 16a indicated bonding
parameters to be normal with the exception of the C(2C)-
C(25)-C(27) bond angle and the torsion angles associated with
the t-Bu tertiary carbon atom.27 At 105.9(2)°, the bond angle is
more than 2σ more compressed than the 110.3 ( 1.9° average
of parameters in a similar environment. This seems to be the
result of packing forces between the hydrogen atoms attached
to C(26) and C(28) and the neighboring molecule related by
2 - x, 1 - y, 1 - z. By canting the t-Bu group away from the
neighboring molecule, the closest intermolecular separations
between the t-Bu group and C(21) and H(21) achieve van der
Waals separation.
FIGURE 5. ORTEP-III drawing (50% probability level, hydrogen
atoms drawn arbitrarily small) of compound 16a viewed (a) normal to
and (b) edge on with core N,N,N plane. Framework bond separations
(Å): C(1)-C(2) 1.438(3), C(2)-C(2a) 1.391(3), C(2a)-C(2b) 1.390-
(3), C(2b)-C(2c) 1.410(3), C(2c)-C(3b) 1.402(3), C(3b)-C(3a) 1.391-
(3), C(3a)-C(3) 1.387(3), C(2)-C(3) 1.465(3), C(3)-C(4) 1.440(3),
C(4)-C(21) 1.403(3), C(21)-C(1) 1.406(3), C(4)-C(5) 1.427(3),
C(5)-C(6) 1.377(3), C(6)-C(7) 1.473(3), C(7)-C(8) 1.359(3), C(8)-
C(9) 1.481(3), C(9)-C(10) 1.417(3), C(10)-C(11) 1.373(3), C(11)-
C(12) 1.455(3), C(12)-C(13) 1.370(3), C(13)-C(14) 1.456(3), C(14)-
C(15) 1.372(3), C(15)-C(16) 1.416(3), C(16)-C(17) 1.475(3), C(17)-
C(18) 1.358(3), C(18)-C(19) 1.474(3), C(19)-C(20) 1.382(3), C(20)-
C(1) 1.425(3), C(6)-N(22) 1.391(3), N(22)-C(9) 1.341(3), C(11)-
N(23) 1.382(3), N(23)-C(14) 1.388(3), C(16)-N(24) 1.342(3), N(24)-
C(19) 1.390(3).
diffraction analysis (Figure 5). This result was particuarly
significant because no structural data had previously been
obtained for a free base azuliporphyrin. Difference Fourier maps
clearly indicated the presence of electron density consistent with
hydrogen atoms attached to the C(21) and N(23) atoms. The
overall macrocycle is essentially planar as evidenced by the
minimal 2.53(6)°, 1.09(6)°, and 1.65(5)° tilts between the pyrrole
planes and the mean macrocyclic plane defined by the meso,
pyrrolic, and C(1), C(2), C(3), and C(4) carbon atoms. The
azulene as a whole displays a 7.40(3)° tilt out of the mean
macrocyclic plane. This appears to be the result of both
intermolecular packing forces and intramolecular steric crowd-
ing. The plane defined by atoms C(1), C(2), C(3), C(4), and
C(21) displays a small 4.81(6)° tilt, easily ascribed to steric
crowding in the central cavity due to the presence of two
hydrogen atoms. The fully aromatic benzocarbaporphyrin 20a
and the related 21-chlorocarbaporphyrin 20b (Chart 2) display
respective 15.5° and 29.6° tilts of their indene units due to
increasing steric crowding caused by the respective presence
In common with benzocarbaporphyrins 20 and 21, macrocycle
16a shows bonds lengths consistent with a delocalized π-bond
distribution. The aromatic pathway of the benzocarbaporphyrins
are best depicted by the 18 atom pathway shown in bold,
whereas the bond length data for 16a suggests the presence of
a markedly different pathway as shown for structure 16aX in
Chart 2. Although the C-C bond lengths in the bold pathway
fall into an approximate range of 1.37-1.45 Å, the C(6)-C(7),
C(8)-C(9), C(16)-C(17), and C(18)-C(19) bonds are all
greater than 1.47 Å, which are typical values for single-bond
(26) Lash, T. D.; Muckey, M. A.; Hayes, M. J.; Liu, D.; Spence, J. D.;
Ferrence, G. M. J. Org. Chem. 2003, 68, 8558-8570.
(27) Bruno, I. J.; Cole, J. C.; Kessler, M.; Luo, J.; Motherwell, W. D.
S.; Purkis, L. H.; Smith, B. R.; Taylor, R.; Cooper, R. I.; Harris, S. E.;
Orpen, A. G. J. Chem. Inf. Comput. Sci. 2004, 44, 2133-2144. (b) Allen,
F. H Acta Crystallogr. 2002, B58, 380-388. (c) Cambridge Structural
Database, version 5.28 (January 2007).
J. Org. Chem, Vol. 72, No. 22, 2007 8407

Lash et al.
SCHEME 6
SCHEME 7
C(sp2)-C(sp2) interactions.28 However, the C(7)-C(8) and
C(17)-C(18) bond distances are the shortest C-C separations
in the structure and are consistent with these atoms being
isolated from the macrocycle’s delocalized π-bond system.
Although the C2-C3 distance is fairly long (1.465(3) Å),
metalloazuliporphyrin structures also have a similarly long bond
length at this location.9,10 The middle pyrrole ring does show
some bond length alternation, but this is much larger in the other
two pyrrole units. It is worth noting that close examination of
the framework metrics reveals sufficient variation in bond
distances such that structural metrics alone are insufficient to
determine the extent of π-delocalization in 16a. Nevertheless,
the data suggest that azuliporphyrins have significant aromatic
character and the 17 atom delocalization pathway illustrated by
structure 16aX appears to be a significant contributor for this
system.
SCHEME 8
The reactivity of 23-substituted azuliporphyrins was also
investigated, but this work was restricted to studies on azuli-
porphyrins 19a and 19b. Azuliporphyrins have been shown to
be excellent organometallic ligands,9,10 and this property was
investigated for the 23-substituted compounds (Scheme 6). The
azuliporphyrins reacted with nickel(II) acetate in refluxing DMF
to give the nickel(II) derivatives 22 in 86-97% yield. These
derivatives were easily purified by column chromatography on
grade 3 alumina. The meso-protons for 22a appeared as two
2H singlets at 8.48 an 9.11 ppm, and these resonances appeared
at virtually the same values in 22b. These chemical shifts and
other data from these spectra indicate that the two metal
complexes have similar diatropic character. Azuliporphyrins 19a
and 19b also reacted with palladium(II) acetate in refluxing
DMF to give the palladium(II) complexes 23 in >80% yield.
The meso-protons for 23a gave two 2H singlets at 8.61 and
9.24 ppm, and again the phenyl-substituted metalloazulipor-
phyrin 23b gave similar shifts. The palladium(II) complexes
23 appear to be slightly more diatropic than the nickel
derivatives 22, probably because Pd(II) is a better fit for the
macrocyclic cavity which allows for a more planar structure.
Azuliporphyrins 4 undergo an unusual oxidative ring contrac-
tion reaction in the presence of peroxides under alkaline
conditions to afford benzocarbaporphyrins 1a-c. We have
proposed8,14 that this chemistry is triggered by a nucleophilic
attack onto the azuliporphyrin macrocycle at position 23. The
availability of 23-substituted azuliporphyrins such as 19a and
19b allowed us to further probe this interesting reaction.
Azuliporphyrins were shown to reversibly form adducts with
nucleophiles such as pyrrolidine to give species like 24 (Scheme
7). Although the pyrrolidine adduct was formed in equilibrium
with azuliporphyrin 4, addition of one drop of pyrrolidine to
an NMR solution of 4 in CDCl3 was sufficient to turn the green
solution brown and to produce a proton NMR spectrum that
was consistent with structure 24.8,14 Only one major species
was present in the NMR spectrum, and a plane of symmetry
was evident in part from the presence of only two 2H singlets
for the meso-protons at 9.67 and 9.88 ppm.8,14 The aromatic
character of the adduct is greatly increased as can be seen from
the values cited for the meso-proton resonances and the 21-CH
resonance which was shifted upfield to -6.87 ppm. The methyl
groups attached to the porphyrin macrocycle gave a 6H singlet
at 3.53 ppm, providing further evidence of a powerful diatropic
ring current in this species.8,14 In order to see whether this type
of adduct could still be formed for structures like 19a and 19b,
proton NMR spectra for these azuliporphyrins were run in
CDCl3 in the presence of increasing concentrations of pyrroli-
dine-d8. Addition of one drop of pyrrolidine to 19a showed the
presence of a new species, but the major peaks were due to the
original azuliporphyrin (Figure 6A). This showed that the tert-
butyl group shifts the equilibrium away from the formation of
pyrrolidine adducts but does not prevent this chemistry from
occurring. Addition of further amounts of d8-pyrrolidine caused
the equilibrium to further shift toward the formation of the
pyrrolidine adduct and a single resonance was observed near
-7 ppm for the 21-CH unit (Figure 6). The data are consistent
with the formation of the asymmetrical adduct 25a (Scheme 8)
where the nucleophilic addition has taken place at the position
adjacent to the tert-butyl moiety. Four 1H singlets were observed
near 10 ppm for the meso-protons and the azulene protons were
replaced by a series of four 1H doublets between 5.0 and 8.4
ppm. As the amount of pyrrolidine was increased, all of the
resonances appeared to shift upfield by >0.3 ppm, but this was
an artifact due to the residual chloroform peak being used as a
reference (see the Supporting Information). When small volumes
of pyrrolidine were added to an NMR tube containing CDCl3
with a trace amount of TMS, the chloroform resonance was
(28) A Mogul search of C(sp2, aromatic)-C(sp2, nonaromatic) returned
an average 1.48 ( 0.02 Å distance for over 500 hits.
8408 J. Org. Chem., Vol. 72, No. 22, 2007

Syntheses and ReactiVity of meso-Unsubstituted Azuliporphyrins
FIGURE 6. Partial 400 MHz proton NMR spectra of tert-butylazuliporphyrin 19a in CDCl3 with 1 drop (A), 4 drops (B), or 10 drops (C) of
pyrrolidine-d8. The peaks corresponding to pyrrolidine adduct 25a, labeled with arrows in spectrum A, increase in relative intensity with increasing
concentrations of pyrrolidine-d8. The apparent upfield shift to the resonances is an artifact due to the residual chloroform signal shifting downfield
due to interactions with the pyrrolidine. The inset upfield region shown for spectrum C shows the internal CH resonance of the aromatic adduct.
observed to shift downfield relative to the TMS standard in
direct proportion to the amount of pyrrolidine added. This shift
was attributed to a weak hydrogen bonding interaction between
the chloroform and pyrrolidine and the data demonstrate that
the use of solvent peaks as a reference can be unreliable.
Therefore, the apparent shifts seen in Figure 6 are due to a
downfield shift of the chloroform and not an upfield shift to
the resonances for the pyrrolidine adduct. Proton NMR spectra
for 23-phenylazuliporphyrin 19b in the presence of pyrrolidine-
d8 gave similar results, although the presence of a second, albeit
minor, species was evident. The major pyrrolidine adduct also
lacked a plane of symmetry and the NMR data were consistent
with the formation of species 25b.
Initially, we had speculated that the presence of a tert-butyl
group on the azulene ring would block oxidative ring contraction
reactions to benzocarbaporphyrins. However, azuliporphyrins
19a and 19b both reacted rapidly with t-BuOOH and KOH at
room temperature to give ring contraction products. The
reactions were monitored by TLC, and these results indicated
that the reactions were slightly faster for the substituted
azuliporphyrins compared to reactions previously conducted for
4 under similar conditions. Treatment of these azuliporphyrins
J. Org. Chem, Vol. 72, No. 22, 2007 8409

Lash et al.
SCHEME 9
SCHEME 10
with 2.5 equiv of t-BuOOH in the presence of potassium
hydroxide in methanol-dichloromethane gave two isolatable
benzocarbaporphyrins, although 23-tert-butylazuliporphyrin 19a
gave one major product in 45% yield while 19b gave roughly
equal amounts of the two products in a combined yield of
approximately 60%. The major carbaporphyrin product derived
from 19a showed the presence of a tert-butyl group and a 1,2,4-
trisubstituted benzene moiety. On the basis of proton NMR and
carbon-13 NMR spectroscopy and mass spectrometry, the
product was identified as 26a (Scheme 9). The less polar product
from the reaction of 19b with t-BuOOH and KOH were
similarly characterized and shown to be the analogous phenyl-
substituted benzocarbaporphyrin 26b. The minor product from
19a was only formed in trace amounts. The IR spectrum for
this compound showed the presence of a carbonyl stretch at
1699 cm-1, and HRMS gave a molecular formula of C40H45N3O.
Two plausible structures were considered for this minor product,
the formyl derivative 27a and ketone 28a. The proton NMR
spectrum showed two 1H doublets at 7.81 and 8.77 ppm (J )
8 Hz) and a series of five 1H singlets at 9.67, 9.69, 9.97, 10.03,
and 11.86 ppm. If the product were the pivaloyl derivative 28a,
long-range coupling to the isolated benzene proton would be
expected, but this is not seen in the NMR data. Also, the
presence of a downfield resonance at 11.86 ppm is not consistent
with structure 28a, but the meso-proton adjacent to the formyl
moiety in 27a would be expected to give a strongly deshielded
singlet in this region. As only trace amounts of this compound
were formed, further spectroscopic analysis was not feasible.
However, the more polar benzocarbaporphyrin derived from
azuliporphyrin 19b gave similar data and could be isolated in
27-42% yield. This product gave a carbonyl stretch in its IR
spectrum at 1688 cm-1, and the HRMS data gave a molecular
formula of C42H41N3O. The proton NMR data showed two 1H
doublets for the benzo-unit at 7.77 and 9.00 ppm, and a series
of five 1H singlets at 9.59, 9.65, 9.90, 10.62 and 11.38 ppm.
These data were again a better fit for the aldehyde structure
27b than for ketone 28b, and the downfield singlet at 11.38
ppm could only be attributed to the close proximity of a carbonyl
moiety to a meso-proton. Carbon-13 NMR spectroscopy gives
a carbonyl resonance at 196.2 ppm and DEPT confirms that
this corresponds to the CH of an aldehyde.
anion at position 22 on the azulene unit of 19 would give adduct
29 (pathway a) and subsequent Cope rearrangement would lead
to the cyclopropane peroxide species 30. Elimination of tert-
butyl alcohol and extrusion of carbon monoxide would afford
the substituted benzocarbaporphyrins 26. Alternatively, addition
at the sterically hindered 21-carbon would give adduct 31
(pathway b) and this could undergo a Cope rearrangement to
give 32. Subsequent elimination of tert-butyl alcohol would then
afford the aldehydes 27. Although the pyrrolidine experiments
showed that the 22-adducts were strongly favored, particularly
for 19a, these additions are reversible and do not appear to
control the regioselectivity for this chemistry. The formation
of 27b as a major product from the oxidative ring contraction
of phenylazuliporphyrin 19b is somewhat surprising, but if the
elimination of t-BuOH is the rate-limiting step the tert-butyl
group would sterically hinder pathway b. Certainly the steric
bulk of the tert-butyl moiety in 19a does effectively control
the regioselectivity for this reaction compared to the 23-
phenylazuliporphyrin 19b.
The structures of the carbaporphyrin products isolated from
these reactions are consistent with the mechanisms shown in
Scheme 10. Initial nucleophilic addition of a tert-butyl peroxide
8410 J. Org. Chem., Vol. 72, No. 22, 2007

Syntheses and ReactiVity of meso-Unsubstituted Azuliporphyrins
Calcd for C42H50N2O4: C, 77.98; H, 7.79; N, 4.33. Found: C, 77.80;
H, 7.68; N, 4.41.
Conclusion
23-tert-Butyl-8,12,13,17-tetraethyl-7,18-dimethylazuliporphy-
rin (16a). Tripyrranedicarboxylic acid 6a31,32 (100 mg; 0.221 mmol)
was stirred with TFA (1 mL) under nitrogen for 2 min. The solution
was diluted with dichloromethane (19 mL), 6-tert-butyl-1,3-
azulenedicarbaldehyde (53 mg; 0.221 mmol) was added, and the
resulting mixture was stirred overnight in the dark. The solution
was diluted with chloroform and shaken vigorously with 0.1%
aqueous ferric chloride solution (200 mL) for 5 min. The organic
layer was separated and the aqueous solution back-extracted with
chloroform. The combined organic solutions were washed with
water and then with 5% aqueous sodium bicarbonate, and the
organic phase was evaporated under reduced pressure. The residue
was loaded onto a grade 3 basic alumina column and eluted initially
with dichloromethane and then with chloroform. A deep green
fraction was collected and evaporated under reduced pressure and
the residue recrystallized from chloroform-hexanes to give the tert-
butylazuliporphyrin (71 mg; 0.125 mmol; 57%) as purple crystals:
mp >300 °C; UV-vis (1% Et3N-CHCl3) λmax (log ꢀ) 358 (4.81),
397 (4.71), 447 (4.74), 474 (4.81), 618 (sh, 4.23), 665 nm (4.26);
UV-vis (1% TFA-CHCl3) λmax (log ꢀ) 368 (4.92), 462 (5.06),
638 nm (4.46); 1H NMR (CDCl3) δ 1.58 (6H, t, J ) 7.6 Hz), 1.60
(9H, s), 1.64 (6H, t, J ) 7.8 Hz), 2.84 (1H, v br), 2.88 (1H, br),
3.02 (6H, s), 3.37 (4H, q, J ) 7.6 Hz), 3.50 (4H, q, J ) 7.6 Hz),
7.89 (2H, d, J ) 10.8 Hz), 8.15 (2H, s), 9.05 (2H, s), 9.33 (2H, d,
J ) 11.2 Hz); 1H NMR (TFA-CDCl3) δ -3.19 (1H, s), -2.0 (1H,
v br), -0.25 (2H, br s), 1.65 (6H, t, J ) 7.8 Hz), 1.72 (9H, s), 1.75
(6H, t, J ) 7.6 Hz), 3.47 (6H, s), 3.81-3.91 (8H, 2 overlapping
quartets), 8.79 (2H, d, J ) 11.2 Hz), 9.54 (2H, s), 9.93 (2H, d, J
) 10.8 Hz), 10.39 (2H, s); 1H NMR (pyridine-d5): δ 1.49 (9H, s),
1.57-1.64 (12H, 2 overlapping triplets), 3.11 (6H, s), 3.36 (1H,
s), 3.42-3.50 (8H, 2 overlapping quartets), 7.88 (2H, d, J ) 11.2
A series of six azuliporphyrins were prepared from 6-tert-
butyl and 6-phenylazulene using two different “3 + 1” strategies.
These porphyrinoids were all formed in excellent yields and
showed significant diatropic characteristics. The presence of a
tert-butyl group on the azulene unit increased the diatropic
properties of the azuliporphyrin macrocycle, and other substitu-
ent and solvent effects have been noted. One of the new
azuliporphyrins gave crystals suitable for X-ray crystallographic
analysis and the macrocycle was shown to be fairly planar and
the bond lengths are consistent with the presence of a favored
17 atom delocalization pathway. Addition of pyrrolidine to NMR
solutions of 23-substituted azuliporphyrins showed that the
formation of nucleophilic adducts was sterically inhibited and
the additions were primarily favored at the 22-position. However,
although oxidative ring contractions of azuliporphyrins in the
presence of t-BuOOH and KOH appear to involve an initial
nucleophilic attack onto the azulene ring, the 23-substituents
did not inhibit the formation of benzocarbaporphyrin products.
Two different benzocarbaporphyrins were identified in these
reactions, and the structures of these products provides insights
into the mechanisms for these unusual oxidative ring contrac-
tions.
Experimental Section
1,3-Bis(5-tert-butoxycarbonyl-3-ethyl-4-methyl-2-pyrrolyl-
methyl)-6-tert-butylazulene (18a). 6-tert-Butylazulene (315 mg;
1.71 mmol) and tert-butyl 5-acetoxymethyl-4-ethyl-3-methylpyrrole-
2-carboxylate (9)30 (1.07 g; 3.80 mmol) were dissolved in ethyl
acetate (25 mL) and acetic acid (5 mL). The mixture was flushed
with nitrogen and stirred under reflux overnight. The solvent was
evaporated to dryness on a rotary evaporator and the residue
chromatographed on silica eluting initially with 20% hexanes-
dichloromethane. A major deep blue band corresponding to the
tripyrrane analogue was collected, the solvent evaporated, and the
residue recrystallized from toluene to give 18a (588 mg; 0.939
mmol; 55%) as blue crystals: mp 197-199 °C, with darkening at
170 °C; 1H NMR (CDCl3) δ 1.09 (6H, t, J ) 7.6 Hz), 1.43 (9H, s),
1.47 (18H, s), 2.26 (6H, s), 2.50 (4H, q, J ) 7.5 Hz), 4.28 (4H, s),
7.25 (2H, d, partially obscured by chloroform peak), 7.41 (1H, s),
8.10 (2H, br), 8.12 (2H, d, J ) 10.4 Hz); 13C NMR (CDCl3): δ
10.7, 15.7, 17.6, 24.3, 28.8, 32.1, 38.8, 80.1, 118.8, 120.6, 123.4,
124.3, 126.1, 131.8, 132.8, 135.9, 137.8, 161.5, 162.2; HRMS (EI)
calcd for C40H54N2O4 m/z 626.4084, found 626.4083. Anal. Calcd
for C40H54N2O4: C, 76.64; H, 8.68; N, 4.47. Found: C, 76.37; H,
8.80; N, 4.50.
1,3-Bis(5-tert-butoxycarbonyl-3-ethyl-4-methyl-2-pyrrolyl-
methyl)-6-phenylazulene (18b). 6-Phenylazulene (350 mg; 1.71
mmol) and acetoxymethylpyrrole 930 (1.07 g; 3.80 mmol) in acetic
acid (5 mL) and ethanol (25 mL) were reacted under the previous
conditions. Recrystallization from toluene gave the tripyrrane
analogue (539 mg; 0.834 mmol; 49%) as blue crystals, mp 177-
180 °C, with darkening at 170 °C: 1H NMR (CDCl3) δ 1.09 (6H,
t, J ) 7.6 Hz), 1.48 (18H, s), 2.26 (6H, s), 2.50 (4H, q, J ) 7.5
Hz), 4.32 (4H, s), 7.29 (2H, d, J ) 10.4 Hz), 7.39-7.44 (1H, m),
7.45-7.50 (2H, m), 7.47 (1H, s), 7.61-7.64 (2H, m), 8.16 (2H, br
s), 8.22 (2H, d, J ) 10.4 Hz); 13C NMR (CDCl3) δ 10.7, 15.7,
17.6, 24.4, 28.8, 80.2, 118.9, 123.0, 123.6, 125.4, 126.1, 128.3,
128.7, 128.9, 131.5, 133.0, 136.0, 138.6, 145.4, 152.2, 161.6; HRMS
(EI) calcd for C42H50N2O4 m/z 646.3771, found 646.3765. Anal.
1
Hz), 8.56 (2H, s), 9.53 (2H, d, J ) 10.8 Hz), 9.56 (2H, s); H
NMR (acetone-d6) δ 1.57 (6H, t, J ) 7.6 Hz), 1.66 (9H, s), 1.68
(6H, t, J ) 8 Hz), 2.62 (1H, v br), 2.77 (1H, br), 3.05 (6H, s), 3.42
(4H, q, J ) 7.6 Hz), 3.60 (4H, q, J ) 7.6 Hz), 8.18 (2H, d, J )
10.4 Hz), 8.32 (2H, s), 9.30 (2H, s), 9.69 (2H, d, J ) 10.4 Hz); 1H
NMR (DMF-d7) δ 1.57 (6H, t, J ) 7.6 Hz), 1.65 (9H, s), 1.67
(6H, t, J ) 7.2 Hz), 3.10 (6H, s), 3.45 (4H, q, obscured by water
peak), 3.63 (4H, q, J ) 7.6 Hz), 8.23 (2H, d, J ) 10.8 Hz), 8.42
(2H, s), 9.47 (2H, s), 9.88 (2H, d, J ) 10.8 Hz); 1H NMR (DMSO-
d6) δ 1.53 (6H, t, J ) 7.6 Hz), 1.60 (9H, s), 1.62 (6H, t, J ) 7.2
Hz), 3.05 (6H, s), 3.39 (4H, q, J ) 7.6 Hz), 3.57 (4H, q, J ) 7.6
Hz), 8.21 (2H, d, J ) 10.8 Hz), 8.33 (2H, s), 9.38 (2H, s), 9.82
(2H, d, J ) 11.2 Hz); 13C NMR (CDCl3) δ 11.1, 16.6, 19.1, 19.4,
31.9, 39.0, 93.3, 108.0, 126.6, 130.1, 134.2, 136.3, 139.5, 139.8,
141.9, 142.7, 148.1, 155.2, 161.6, 164.4; 13C NMR (TFA-CDCl3)
δ 11.6, 16.1, 17.0, 19.7, 19.8, 31.7, 40.6, 95.1, 110.0, 124.1, 128.6,
139.3, 140.9, 141.6, 141.9, 142.1, 143.8, 144.5, 146.6, 153.1, 158.6,
173.3; HRMS (EI) calcd for C40H45N3 + 2H m/z 569.3770, found
569.3780. Anal. Calcd for C40H45N3‚1/5CHCl3: C, 81.60; H, 7.70;
N, 7.10. Found: C, 81.56; H, 7.81; N, 7.08.
8,12,13,17-Tetraethyl-7,18-dimethyl-23-phenylazuliporphy-
rin (17a). Using the same conditions, tripyrrane 6a (100 mg 0.221
mmol) was reacted with dialdehyde 15b (57.5 mg; 0.221 mmol).
Recrystallization from chloroform-hexanes gave the phenylazuli-
porphyrin (80.5 mg; 0.137 mmol; 62%) as dark purple crystals,
mp >300 °C; UV-vis (1% Et3N-CHCl3): λmax (logꢀ) 361 (4.77),
399 (4.73), 447 (4.77), 475 (4.89), 628 nm (4.30); UV-vis (1%
TFA-CHCl3): λmax (logꢀ) 380 (4.80), 466 (5.07), 642 nm (4.49);
1H NMR (CDCl3): δ 1.56 (6H, t, J ) 7.6 Hz), 1.62 (6H, t, J ) 7.6
Hz), 2.99 (6H, s), 3.23 (1H, v br), 3.30 (1H, s) 3.33 (4H, q, J )
7.6 Hz), 3.46 (4H, q, J ) 7.7 Hz), 7.52-7.56 (1H, m), 7.57-7.61
(2H, m), 7.76 (2H, d, J ) 8 Hz), 7.81 (2H, d, J ) 9.6 Hz), 8.06
(29) Ito, S.; Morita, N.; Asao, T. Bull. Chem. Soc. Jpn. 1999, 72, 2543-
2548.
(31) Sessler, J. L.; Johnson, M. R.; Lynch, V. J. Org. Chem. 1987, 52,
4394-4397.
(32) Lash, T. D. J. Porphyrins Phthalocyanines 1997, 1, 29-44.
(30) Clezy, P. S.; Crowley, R. J.; Hai, T. T. Aust. J. Chem. 1982, 35,
411-421.
J. Org. Chem, Vol. 72, No. 22, 2007 8411

Lash et al.
1
(2H, s), 8.96 (2H, s), 9.29 (2H, d, J ) 10 Hz); H NMR (TFA-
vis (1% Et3N-CHCl3) λmax (log ꢀ) 373 (4.85), 405 (4.79), 451
(4.82), 479 (4.93), 646 nm (4.37); UV-vis (1% TFA-CHCl3) λmax
(log ꢀ) 387 (4.85), 471 (5.09), 650 nm (4.56); 1H NMR (CDCl3) δ
1.47 (6H, t, J ) 7.8 Hz), 3.03 (6H, br s), 3.17-3.28 (4H, br q),
7.50-7.63 (9H, m), 7.73 (4H, d, J ) 7 Hz), 7.78 (2H, d, J ) 7
CDCl3) δ -2.91 (1H, s), -1.79 (1H, v br), -0.11 (2H, br s), 1.66
(6H, t, J ) 7.8 Hz), 1.75 (6H, t, J ) 7.6 Hz), 3.47 (6H, s), 3.81-
3.90 (8H, 2 overlapping quartets), 7.69-7.75 (3H, m), 7.89-7.92
(2H, m), 8.77 (2H, d, J ) 10.4 Hz), 9.50 (2H, s), 9.97 (2H, d, J )
11.2 Hz), 10.39 (2H, s); 1H NMR (pyridine-d5) δ 1.56-1.62 (12H,
2 overlapping triplets), 3.02 (6H, s), 3.38-3.46 (8H, 2 overlapping
quartets), 3.7 (1H, v br), 3.78 (1H, s), 7.5-7.6 (3H, m, obscured
by solvent), 7.85-7.87 (2H, m), 7.95 (2H, d, J ) 10.4 Hz), 8.47
(2H, s), 9.52 (2H, s), 9.70 (2H, d, J ) 10.8 Hz); 1H NMR (acetone-
d6) δ 1.56 (6H, t, J ) 7.6 Hz), 1.66 (6H, t, J ) 7.4 Hz), 3.03 (6H,
s), 3.39 (4H, q, J ) 7.6 Hz), 3.56 (4H, q, J ) 7.4 Hz), 7.59-7.63
(1H, m), 7.68 (2H, t, J ) 7.4 Hz), 7.96 (2H, d, J ) 7.2 Hz), 8.12
(2H, d, J ) 10.4 Hz), 8.24 (2H, br s), 9.26 (2H, s), 9.74 (2H, d, J
1
Hz), 7.89 (2H, br), 8.24 (2H, br), 9.08 (2H, br), 9.35 (2H, br); H
NMR (TFA-CDCl3) δ -2.69 (1H, s), 0.7 (2H, v br), 1.57 (6H, t,
J ) 7.6 Hz), 3.44 (6H, s), 3.67 (4H, q, J ) 7.6 Hz), 7.65-7.71
(9H, m), 7.80-7.83 (4H, m), 7.86-7.90 (2H, m), 8.69 (2H, d, J )
10.4 Hz), 9.47 (2H, s), 9.93 (2H, d, J ) 10.8 Hz), 10.36 (2H, s);
1H NMR (pyridine-d5) δ 1.50 (6H, t, J ) 7.6 Hz), 3.05 (6H, s),
3.25 (4H, q, J ) 7.6 Hz), 3.35 (1H, br s), 3.5 (1H, v br), 7.47-
7.61 (9H, m), 7.87-7.93 (6H, m), 8.02 (2H, d, J ) 10.8 Hz), 8.68
(2H, s), 9.68 (2H, s), 9.79 (2H, d, J ) 10.4 Hz); 13C NMR (TFA-
CDCl3) δ 11.5, 15.9, 19.6, 98.6, 109.5, 123.7, 128.7, 128.9, 129.2,
129.3, 130.0, 131.1, 132.0, 132.6,139.1, 141.0, 141.4, 141.9, 142.1,
142.5, 145.1, 146.8, 152.2, 160.3; HRMS (EI) calcd for C50H41N3
+ 2H m/z 685.3457, found 685.3457. Anal. Calcd for C50H41N3‚
1/8CHCl3: C, 86.15; H, 5.93; N, 6.01. Found: C, 85.87; H, 5.81;
N, 6.23.
1
) 10.8 Hz); H NMR (DMF-d7) δ 1.57 (6H, t, J ) 7.4 Hz), 1.66
(6H, t, J ) 7.6 Hz), 2.80 (1H, s), 3.08 (6H, s), 3.42 (4H, q, obscured
by water peak), 3.60 (4H, q, J ) 7.6 Hz), 7.61-7.65 (1H, m), 7.67-
7.72 (2H, m), 8.02-8.05 (2H, m), 8.20 (2H, d, J ) 10.8 Hz), 8.34
(2H, s), 9.45 (2H, s), 9.96 (2H, d, J ) 10.4 Hz); 1H NMR (DMSO-
d6) δ 1.52 (6H, t, J ) 7.6 Hz), 1.60 (6H, t, J ) 7.6 Hz), 3.04 (6H,
s), 3.37 (4H, q, J ) 7.6 Hz), 3.54 (4H, q, J ) 7.5 Hz), 7.59-7.63
(1H, m), 7.66-7.70 (2H, m), 7.97 (2H, d, J ) 7.2 Hz), 8.17 (2H,
d, J ) 10.8 Hz), 8.25 (2H, s), 9.36 (2H, s), 9.87 (2H, d, J ) 11.2
Hz); 13C NMR (CDCl3) δ 11.1, 16.6, 17.1, 19.1, 19.3, 93.4, 108.1,
126.9, 128.3, 129.0, 129.3, 131.9, 134.2, 136.9, 139.6, 139.9, 142.0,
143.1, 144.0, 147.4, 153.7, 155.5, 162.1; 13C NMR (TFA-CDCl3)
δ 11.6, 16.0, 16.9, 19.7, 19.8, 95.1, 110.1, 123.6, 128.7, 129.0,
130.2, 131.4, 139.4, 141.4, 141.9, 142.1, 142.3, 144.1, 144.9, 147.1,
152.4; HRMS (EI) calcd for C42H41N3 + 2H m/z 589.3457, found
589.3454. Anal. Calcd for C40H45N3‚1/4CHCl3: C, 82.16; H, 6.73;
N, 6.80. Found: C, 82.05; H, 6.70; N, 6.85.
23-tert-Butyl-7,12,13,18-tetraethyl-8,17-dimethylazuliporphy-
rin (19a). Azulitripyrrane analogue 18a (220 mg; 0.35 mmol) was
stirred in TFA (6 mL) under nitrogen for 10 min. The reaction
mixture was diluted with dichloromethane (200 mL), 3,4-dieth-
ylpyrrole-2,5-dicarbaldehyde32,33 (63 mg; 0.35 mmol) was added,
and the reaction mixture was stirred overnight in the dark. The
solution was shaken vigorously in a separatory funnel with 0.1%
aqueous ferric chloride solution (300 mL) for 5 min. The organic
layer was separated and the aqueous solution back-extracted with
chloroform. The combined organic solutions were washed with
water and then with 5% aqueous sodium bicarbonate, and the
organic phase was dried over sodium sulfate and evaporated under
reduced pressure. The residue was purified on a grade 3 basic
alumina column, eluting with chloroform, and the product was
collected as a deep green fraction that followed a brown prefraction.
The solvent was removed on a rotary evaporator and the residue
recrystallized from chloroform-hexanes to give the tert-butylazu-
liporphyrin (102.3 mg; 0.18 mmol; 51%) as dark purple crystals:
mp >300 °C; UV-vis (1% Et3N-CHCl3) λmax (log ꢀ) 357 (4.78),
400 (sh), 446 (4.64), 474 (4.75), 663 nm (4.15); UV-vis (1%
TFA-CHCl3) λmax (log ꢀ) 365 (4.81), 460 (5.00), 634 (4.41), 728
nm (3.83); UV-vis (1% pyrrolidine-CHCl3) λmax (log ꢀ) 368
(4.65), 402 (4.68), 445 (4.57), 473 (4.59), 619 nm (4.01); UV-vis
(5% pyrrolidine-CHCl3) λmax (log ꢀ) 368 (4.62), 402 (4.62), 449
(4.59), 581 nm (4.11); UV-vis (10% pyrrolidine-CHCl3) λmax (log
ꢀ) 397 (4.58), 451 (4.60), 586 nm (4.11); 1H NMR (CDCl3) δ 1.61
(9H, s), 1.58-1.65 (12H, 2 overlapping triplets), 2.8 (1H, v br),
2.86 (1H, br), 2.92 (6H, s), 3.43-3.52 (2 overlapping quartets),
7.92 (2H, d, J ) 10.8 Hz), 8.16 (2H, s), 9.04 (2H, s), 9.35 (2H, d,
23-tert-Butyl-8,17-diethyl-7,18-dimethyl-12,13-diphenylazuli-
porphyrin (16b). Tripyrrane 6b4b (100 mg; 0.182 mmol) was
reacted with 6-tert-butyl-1,3-azulenedicarbaldehyde (44 mg; 0.183
mmol) under the foregoing conditions. The crude product was
chromatographed on grade 3 basic alumina, eluting with chloroform,
and recrystallized from chloroform-hexanes to afford the porphyrin
analogue (66 mg; 0.10 mmol; 55%) as purple crystals: mp
>300 °C; UV-vis (1% Et3N-CHCl3) λmax (log ꢀ) 371 (4.79), 405
(4.74), 433 (4.77), 477 (4.79), 557 (4.06), 677 nm (4.24); UV-vis
(1% TFA-CHCl3) λmax (log ꢀ) 374 (4.81), 466 (4.97), 644 nm
(4.42); 1H NMR (CDCl3) δ 1.50 (6H, t, J ) 7.6 Hz), 1.62 (9H, s),
2.49 (1H, br s), 2.8 (1H, v br), 3.04 (6H, s), 3.25 (4H, q, J ) 7.6
Hz), 7.49-7.59 (6H, m), 7.77 (4H, d, J ) 8 Hz), 7.95 (2H, d, J )
10.4 Hz), 8.33 (2H, s), 9.17 (2H, s), 9.39 (2H, d, J ) 10.4 Hz); 1H
NMR (TFA-CDCl3) δ -3.43 (1H, s), -1.8 (1H, v br), -0.53 (2H,
br s), 1.58 (6H, t, J ) 7.8 Hz), 1.73 (9H, s), 3.49 (6H, s), 3.74
(4H, q, J ) 7.6 Hz), 7.68-7.74 (6H, m), 7.82-7.85 (4H, m), 8.85
(2H, d, J ) 10.8 Hz), 9.66 (2H, s), 9.98 (2H, d, J ) 10.4 Hz), 10.5
1
J ) 10.8 Hz); H NMR (TFA-CDCl3) δ -3.33 (1H, s), -2.30
1
(2H, s, obscured by TFA); H NMR (pyridine-d5) δ 1.50 (9H, s),
(1H, br), -0.60 (2H, s), 1.65 (9H, s), 1.61-1.70 (12H, 2
overlapping triplets), 3.31 (6H, s), 3.81 (4H, q, J ) 7.8 Hz), 3.87
(4H, q, J ) 7.8 Hz), 8.75 (2H, d, J ) 10.4 Hz), 9.51 (2H, s), 9.88
(2H, d, J ) 10.4 Hz), 10.32 (2H, s); 1H NMR (pyridine-d5) δ 1.49
(9H, s), 1.61 (6H, t, J ) 7.6 Hz), 1.67 (6H, t, J ) 7.6 Hz), 2.98
(6H, s), 3.27 (1H, v br), 3.35 (1H, s), 3.49 (4H, q, J ) 7.6 Hz),
3.59 (4H, q, J ) 7.5 Hz), 7.87 (2H, d, J ) 10.4 Hz), 8.58 (2H, s),
1.52 (6H, t, J ) 7.2 Hz), 2.90 (1H, br s), 3.14 (6H, s), 3.30 (4H,
q, J ) 7.6 Hz), 7.50 (2H, t, J ) 7.4 Hz), 7.60 (4H, t, J ) 7.4 Hz),
7.92-7.98 (6H, m), 8.79 (2H, s), 9.63 (2H, d, J ) 10.8 Hz), 9.72
(2H, s); 13C NMR (CDCl3) δ 11.2, 16.5, 19.3, 31.9, 39.1, 96.5,
108.1, 126.8, 127.8, 128.5, 130.9, 132,1, 134.5, 134.9, 135.8, 139.3,
139.5, 141.9, 142.4, 148.8, 155.4, 161.5, 164.7; 13C NMR (TFA-
CDCl3) δ 11.5, 15.8, 19.6, 31.6, 40.6, 98.4, 109.6, 123.7, 129.0,
129.4, 129.6, 132.0,139.4, 141.4, 142.2, 142.4, 142.7, 146.3, 153.4,
173.6; HRMS (EI) calcd for C48H45N3 + 2H m/z 655.3770; found:
655.3771. Anal. Calcd for C48H45N3‚0.3CHCl3: C, 82.91; H, 6.52;
N, 6.01. Found: C, 82.88; H, 6.46; N, 6.25.
1
9.57 (2H, d, J ) 10 Hz), 9.58 (2H, s); H NMR (10 drops of
pyrrolidine-d8-CDCl3; peaks for 22-pyrrolidine adduct 25a only)
δ -7.06 (1H, s), -4.5 (2H, v br), 1.13 (9H, s,), 2.18-2.23 (12H,
2 overlapping triplets), 3.35 (6H, obscured by solvent impurities),
3.68 (4H, q, J ) 7.6 Hz), 3.82 (4H, q, J ) 7.6 Hz), 5.03 (1H, d,
J ) 7.6 Hz), 6.07 (1H, d, J ) 8 Hz), 6.98 (1H, d, J ) 12 Hz), 8.15
(1H, d, J ) 12 Hz), 9.46 (1H, s), 9.47 (1H, s), 9.73 (1H, s), 9.91
(1H, s); 13C NMR (CDCl3) δ 11.0, 17.1, 17.3, 19.1, 19.5, 31.9,
39.0, 93.2, 107.9, 126.6, 130.2, 134.2, 134.8, 136.2, 139.9, 142.8,
8,17-Diethyl-7,18-dimethyl-23,12,13-triphenylazuliporphy-
rin (17b). Tripyrrane 6b4b (100 mg; 0.182 mmol) was reacted with
6-phenyl-1,3-azulenedicarbaldehyde (47 mg; 0.181 mmol) under
the foregoing conditions. The crude product was chromatographed
on grade 3 basic alumina, eluting with chloroform, and recrystallized
from chloroform-hexanes to afford the porphyrin analogue (83
mg; 0.12 mmol; 67%) as a dark green powder: mp >300 °C; UV-
(33) Tardieux, C.; Bolze, F.; Gros, C. P.; Guilard, R. Synthesis 1998,
267-268.
8412 J. Org. Chem., Vol. 72, No. 22, 2007

Syntheses and ReactiVity of meso-Unsubstituted Azuliporphyrins
146.6, 148.1, 154.1, 162.3, 164.4; 13C NMR (TFA-CDCl3) δ 11.2,
16.2, 16.9, 19.8, 20.1, 31.5, 40.8, 95.4, 109.3, 122.8, 128.7, 136.1,
139.6, 141.0, 141.5, 143.8, 145.4, 147.3, 148.8, 153.2, 173.5; HR
MS (EI) calcd for C40H45N3 + 2H m/z 569.3770, found 569.3771.
Anal. Calcd for C40H45N3‚1/20CHCl3: C, 83.83; H, 7.91; N, 7.32.
Found: C, 83.52; H, 7.93; N, 7.32.
(II) derivative 22b (9.4 mg; 0.0146 mmol; 86%) as dark purple
crystals: mp >300 °C (chloroform-methanol); UV-vis (CHCl3)
λ
max (log ꢀ) 388 (4.82), 461 (4.64), 560 nm (4.56); 1H NMR (CDCl3)
δ 1.59-1.65 (12H, 2 overlapping triplets), 3.00 (6H, s), 3.45 (4H,
q, J ) 7.6 Hz), 3.53 (4H, q, J ) 7.7 Hz), 7.50-7.59 (3H, m), 7.66
(2H, d, J ) 10.8 Hz), 7.74 (2H, d, J ) 7 Hz), 8.49 (2H, s), 8.95
(2H, d, J ) 10.8 Hz), 9.11 (2H, s); 13C NMR (CDCl3) δ 11.2,
17.1, 17.4, 19.2, 19.5, 95.7, 106.5, 127.8, 128.7, 128.9, 129.2, 129.3,
134.3, 134.6, 143.6, 144.2, 144.7, 144.9, 149.0, 159.7, 152.8,
157.3; HRMS (EI) calcd for C42H39N3Ni m/z 643.2497, found
643.2493.
[23-tert-Butyl-7,12,13,18-tetraethyl-8,17-dimethylazuliporphy-
rinato]palladium(II) (23a). Azuliporphyrin 19a (10.0 mg; 0.0176
mmol) and palladium(II) acetate (10.0 mg) were reacted under the
same conditions. Following chromatography on grade 3 alumina,
eluting with 20% hexanes-dichloromethane, and recrystallization
from chloroform-methanol, the palladium(II) complex (9.6 mg;
0.0143 mmol; 81%) was isolated as dark purple crystals: mp
>300 °C; UV-vis (CHCl3) λmax (log ꢀ) 366 (4.82), 416 (4.73),
448 (4.71), 568 nm (4.43); 1H NMR (CDCl3) δ 1.46 (9H, s), 1.61
(12H, t, J ) 7.6 Hz), 3.03 (6H, s), 3.46 (4H, q, J ) 7.6 Hz), 3.57
(4H, q, J ) 7.7 Hz), 7.75 (2H, d, J ) 10.8 Hz), 8.61 (2H, s), 9.04
(2H, d, J ) 10.8 Hz), 9.24 (2H, s); 13C NMR (CDCl3) δ 11.3,
17.3, 17.5, 19.2, 19.5, 31.6, 38.8, 96.4, 108.4, 125.4, 130.1, 133.3,
133.5, 142.0, 142.8, 143.3, 147.4, 149.7, 152.8, 161.6; HRMS (EI)
calcd for C40H43N3Pd m/z 671.2492, found 671.2487.
[7,12,13,18-tetraethyl-8,17-dimethyl-23-phenylazuliporphyri-
nato]palladium(II) (23b). Phenylazuliporphyrin 19b (10.0 mg;
0.017 mmol) reacted with palladium(II) acetate (10.0 mg) under
the same conditions to give the palladium(II) derivative (9.7 mg;
0.014 mmol; 82%) as dark purple crystals: mp >300 °C (chloro-
form-methanol); UV-vis (CHCl3) λmax (log ꢀ) 368 (4.58), 422
(4.79), 449 (4.82), 583 nm (4.51); 1H NMR (CDCl3) δ 1.68 (12H,
t, J ) 7.6 Hz), 3.08 (6H, s), 3.50 (4H, q, J ) 7.6 Hz), 3.61 (4H,
q, J ) 7.6 Hz), 7.51-7.61 (3H, m), 7.72-7.77 (4H, 2 overlapping
doublets), 8.62 (2H, s), 9.06 (2H, d, J ) 10.4 Hz), 9.24 (2H, s);
13C NMR (CDCl3; 50 °C) δ 11.3, 17.2, 17.4, 19.3, 19.6, 96.5, 108.6,
125.7, 127.9, 128.8, 129.3, 129.9, 133.6, 135.2, 142.6, 143.1, 143.7,
144.2, 148.1, 150.4, 151.1, 152.2; HRMS (EI) calcd for C42H39N3-
Pd m/z 691.2179, found 691.2183.
Oxidative Ring Contraction of 23-tert-Butyl-7,12,13,18-tet-
raethyl-8,17-dimethylazuliporphyrin. Potassium hydroxide (240
mg) in methanol (30 mL) was added to a solution of tert-
butylazuliporphyrin 19a (23 mg; 0.040 mmol) in dichloromethane
(30 mL) and the resulting stirred mixture was purged with nitrogen
for 10 min. A solution of tert-butyl hydroperoxide in decane (5 M,
20 µL) was added and the resulting solution stirred under nitrogen
in the dark at room temperature for 2 h. The solution was diluted
with chloroform, washed with water (2×), dried over sodium
sulfate, and evaporated under reduced pressure. The residues were
chromatographed on a grade 3 alumina column, eluting initially
with hexanes and then with 50% dichloromethane-hexanes. The
major orange band was collected, the solvent evaporated, and the
residue recrystallized from chloroform-hexanes to give red-brown
crystals (9.8-10.0 mg; 0.0176-0.0180 mmol; 44-45%) corre-
sponding to tert-butylbenzocarbaporphyrin 26a. A second minor
band was collected (0.8-1.9 mg; 0.0014-0.0032 mmol; 3.5-8%)
corresponding to the related aldehyde 27a.
7,12,13,18-Tetraethyl-8,17-dimethyl-23-phenylazuliporphy-
rin (19b). Using the same procedure, tripyrrane analogue 18b (228
mg; 0.353 mmol) gave the phenylazuliporphyrin (93.2 mg; 0.159
mmol; 45%) as dark green crystals: mp >300 °C (chloroform-
hexanes); UV-vis (1% Et3N-CHCl3) λmax (log ꢀ) 358 (4.85), 398
(4.76), 447 (4.78), 475 (4.89), 628 nm (4.31); UV-vis (1% TFA-
CHCl3) λmax (log ꢀ) 354 (4.77), 380 (4.76), 464 (5.14), 638 (4.53);
UV-vis (1% pyrrolidine-CHCl3) λmax (log ꢀ) 368 (4.76), 403
(4.82), 445 (4.77), 474 (4.78), 526 (4.05), 567 (4.10), 624 nm (4.19);
UV-vis (5% pyrrolidine-CHCl3) λmax (log ꢀ) 407 (4.84), 472
(4.76), 525 (4.14), 564 (4.15), 616 nm (4.10); UV-vis (10%
pyrrolidine-CHCl3) λmax (log ꢀ) 403 (4.79), 451 (4.81), 536 (4.17),
585 nm (4.31); 1H NMR (CDCl3) δ 1.58-1.65 (12H, 2 overlapping
triplets), 2.90 (6H, s), 3.15 (1H, v br), 3.30 (1H, v br) 3.42-3.49
(8H, 2 overlapping quartets), 7.51-7.62 (3H, m), 7.78-7.81 (2H,
m), 7.86 (2H, d, J ) 10.8 Hz), 8.07 (2H, s), 8.97 (2H, s), 9.33
1
(2H, d, J ) 10 Hz); H NMR (TFA-CDCl3) δ -3.06 (1H, s),
-2.17 (1H, v br), -0.54 (2H, br s), 1.69-1.77 (12H, 2 overlapping
triplets), 3.38 (6H, s), 3.89 (4H, q, J ) 7.6 Hz), 3.96 (4H, q, J )
7.6 Hz), 7.72-7.75 (3H, m), 7.89-7.92 (2H, m), 8.81 (2H, d, J )
10.4 Hz), 9.57 (2H, s), 10.00 (2H, d, J ) 10.8 Hz), 10.42 (2H, s);
1H NMR (pyridine-d5) δ 1.57-1.63 (12H, 2 overlapping triplets),
2.93 (6H, s), 3.42-3.54 (8H, 2 overlapping quartets), 3.65 (1H, v
br), 3.78 (1H, br s), 7.5-7.6 (3H, m, obscured by solvent), 7.85-
7.88 (2H, m), 7.94 (2H, d, J ) 10.4 Hz), 8.48 (2H, s), 9.55 (2H,
1
s), 9.75 (2H, d, J ) 10.8 Hz); H NMR (10 drops of pyrrolidine-
d8-CDCl3; partial data for major pyrrolidine adduct 25b) δ -6.91
(1H, s), -4.0 (2H, v br), 5.33 (1H, d, J ) 8 Hz), 6.41 (1H, d, J )
8 Hz), 7.2-7.4 (4H, m), 7.50 (2H, d, J ) 7 Hz), 8.43 (1H, d, J )
11.6 Hz), 9.55 (2H, s), 9.86 (1H, s), 10.00 (1H, s); 13C NMR
(CDCl3) δ 10.9, 17.1 (2), 19.0, 19.4, 93.3, 107.6, 126.8, 128.3,
129.1, 129.3, 132.2, 134.4, 134.6, 140.2, 143.8, 146.6, 147.4, 153.9;
13C NMR (TFA-CDCl3) δ 11.1, 16.2, 16.8, 19.6, 20.0, 95.3, 109.5,
123.0, 128.5, 128.9, 130.2, 131.5, 135.9, 139.3, 140.9, 141.2, 142.3,
144.0, 145.1, 147.7, 148.7, 152.2; HRMS (ESI) calcd for C42H41N3
+ H m/z 588.3379, found 588.3378. Anal. Calcd for C42H41N3‚
1/10CHCl3: C, 84.31; H, 6.91; N, 7.01. Found: C, 84.39; H, 7.01;
N, 7.09.
[23-tert-Butyl-7,12,13,18-tetraethyl-8,17-dimethylazuliporphy-
rinato]nickel(II) (22a). A solution of tert-butylazuliporphyrin 19a
(7.7 mg, 0.0136 mmol) and nickel(II) acetate tetrahydrate (10 mg)
in DMF (10 mL) was heated with stirring under reflux for 20 min.
The solution was cooled to room temperature, diluted with
chloroform, washed with water, and dried over sodium sulfate. The
solvent was evaporated under aspirator pressure, and residual DMF
was removed using an oil pump. The residue was purified by
column chromatography on grade 3 alumina eluting with 20%
hexanes-dichloromethane. Recrystallization from chloroform-
methanol gave the nickel(II) derivative (8.2 mg, 0.0132 mmol, 97%)
as dark purple crystals: mp >300 °C; UV-vis (CHCl3) λmax (log
ꢀ) 386 (4.70), 457 (4.52), 550 nm (4.36); 1H NMR (CDCl3) δ 1.46
(9H, s), 1.53-1.59 (12H, 2 overlapping triplets), 2.95 (6H, s), 3.40
(4H, q, J ) 7.7 Hz), 3.48 (4H, q, J ) 7.5 Hz), 7.68 (2H, d, J )
10.8 Hz), 8.48 (2H, s), 8.93 (2H, d, J ) 10.4 Hz), 9.11 (2H, s);
13C NMR (CDCl3) δ 11.2, 17.2, 17.4, 19.2, 19.6, 31.7, 38.8, 95.7,
106.4, 128.7, 129.4, 132.7, 134.1, 143.5, 144.4, 144.7, 148.6, 151.5,
152.3, 155.6, 161.3; HRMS (EI) calcd for C40H43N3Ni m/z
623.2810, found 623.2812.
22-tert-Butyl-7,12,13,18-tetraethyl-8,17-dimethylbenzo[b]car-
baporphyrin (26a): mp 289 °C with decomposition at 265 °C;
UV-vis (1% Et3N-CHCl3) λmax (logꢀ) 379 (4.78), 427 (5.20), 511
(4.37), 545 (4.28), 604 (3.87), 664 nm (3.58); 1H NMR (CDCl3) δ
-6.69 (1H, s), -3.92 (2H, br), 1.84-1.91 (12H, 2 overlapping
triplets), 3.60 (3H, s), 3.61 (3H, s), 3.97 (4H, q, J ) 7.6 Hz), 4.08-
4.18 (4H, 2 overlapping quartets), 7.81 (1H, dd, J ) 1.4, 7.8 Hz),
8.76 (1H, d, J ) 8 Hz), 8.87 (1H, d, J ) 1.6 Hz), 9.80 (1H, s),
9.81 (1H, s), 10.09 (1H, s), 10.13 (1H, s); 13C NMR (CDCl3) δ
11.2, 17.5, 18.6, 19.9, 20.1, 32.2, 35.5, 95.5, 98.3, 98.5, 109.9,
[7,12,13,18-Tetraethyl-8,17-dimethyl-23-phenylazuliporphy-
rinato]nickel(II) (22b). Using the foregoing conditions, phenyla-
zuliporphyrin 19b (10.0 mg; 0.0170 mmol) reacted with nickel(II)
acetate tetrahydrate (10 mg) in DMF (10 mL) to afford the nickel-
J. Org. Chem, Vol. 72, No. 22, 2007 8413

Lash et al.
117.5, 120.2, 123.9, 130.6, 134.2, 134.5, 134.7, 136.3, 139.3, 139.4,
141.5, 144.3, 150.0, 152.7; HRMS (EI) calcd for C39H45N3 m/z
555.3613, found 555.3617.
dark green plate thereby obtained of approximate dimensions 0.48
× 0.12 × 0.04 mm was mounted on a 50 mm MicroMesh MiTeGen
Micromount and transferred to a Bruker AXS SMART APEX CCD
diffractometer. The X-ray diffraction data were collected at
-173 °C using Mo-Ka (λ ) 0.71073 Å) radiation. Data collection
and cell refinement were performed using SMART and SAINT+,
respectively.34 The unit cell parameters were obtained from a least-
squares refinement of 6582 centered reflections. Compound 16a
was found to crystallize in the triclinic crystal system with the
following unit cell parameters: a ) 10.2118(8) Å, b ) 12.0005-
(10) Å, c ) 13.2831(11) Å, R ) 87.9668(15)°, â ) 74.7962(15)°,
γ ) 88.7667(16)°, Z ) 2. The crystal system and lack of systematic
absences indicated the space group to be P1 (no. 1) or P-1 (no.
2).35 The latter was chosen on the basis that E-statistics strongly
suggested a centrosymmetric structure and a sensible, quality
refinement was obtained. A total of 16847 reflections were
collected, of which 7749 were unique, and 4342 were observed
22-tert-Butyl-7,12,13,18-tetraethyl-21-formyl-8,17-dimethylbenzo-
[b]carbaporphyrin (27a): IR νCdO 1699 cm-1; UV-vis (1%
Et3N-CHCl3) λmax (log ꢀ) 377 (4.64), 426 (5.20), 511 (4.24), 547
(4.23), 604 (3.83); 1H NMR (CDCl3) δ -6.97 (1H, s), 1.76-1.82
(12H, m), 3.50 (3H, s), 3.52 (3H, s), 3.87 (4H, q, J ) 7.6 Hz),
3.96-4.05 (4H, m), 7.81 (1H, d, J ) 8 Hz), 8.77 (1H, d, J ) 8
Hz), 9.67 (1H, s), 9.69 (1H, s), 9.97 (1H, s), 10.03 (1H, s), 11.86
(1H, s); HRMS (EI) calcd for C40H45N3O m/z 583.3563, found
583.3568.
Oxidative Ring Contraction of 7,12,13,18-Tetraethyl-8,17-
dimethyl-23-phenylazuliporphyrin. Phenylazuliporphyrin 19b (24.0
mg; 0.041 mmol) was reacted with tert-butyl hydroperoxide (5 M
in hexane, 20 µL) in the presence of potassium hydroxide using
the procedure described above. The crude product was run through
a grade 3 alumina column, eluting with 50% dichloromethane-
hexanes, and two major orange-brown bands were collected. The
first band was evaporated down and the residue recrystallized from
chloroform-methanol to give phenylbenzocarbaporphyrin 26b
(6.2-9.1 mg; 0.011-0.016 mmol; 27-39%) as a red-brown solid,
mp 289 °C with decomposition at 283 °C. The more polar band
was also evaporated down and recrystallized from chloroform-
methanol to give the carbaporphyrin aldehyde 27b (6.9-10.6 mg;
0.011-0.017 mmol; 27-42%) as red-brown crystals, mp 291 °C
with decomposition at 287 °C.
7,12,13,18-Tetraethyl-8,17-dimethyl-22-phenylbenzo[b]car-
baporphyrin (26b): UV-vis (1% Et3N-CHCl3) λmax (log ꢀ) 383
(4.76), 430 (5.19), 512 (4.32), 547 (4.35), 604 (3.84), 665 nm (3.46);
1H NMR (CDCl3) δ -6.67 (1H, s), -4.0 (2H, v br), 1.85-1.89
(12H, 2 overlapping triplets), 3.61 (6H, s), 3.97 (4H, q, J ) 7.7
Hz), 4.09-4.16 (4H, 2 overlapping quartets), 7.49 (1H, t, J ) 7.4
Hz), 7.64 (2H, t, J ) 7.8 Hz), 7.81 (1H, dd, J ) 1.6, 7.6 Hz),
8.03-8.06 (2H, m), 8.89 (1H, d, J ) 8 Hz), 9.04 (1H, d, J ) 1.6
Hz), 9.80 (2H, s), 10.12 (1H, s), 10.16 (1H, s); 13C NMR (CDCl3)
δ 11.2, 17.5, 18.6, 20.0, 20.1, 95.6, 98.5, 98.6, 110.1, 119.6, 120.8,
125.8, 127.3, 127.8, 129.1, 130.6, 133.6, 133.7, 134.7, 136.5, 139.6
(2), 140.0, 140.7, 142.2, 142.9, 144.4, 153.0; HR MS (EI) calcd
for C41H41N3 m/z 575.3300, found 575.3293.
7,12,13,18-Tetraethyl-21-formyl-8,17-dimethyl-22-phenylben-
zo[b]carbaporphyrin (27b): IR: νCdO 1688 cm-1; UV-vis (1%
Et3N-CHCl3) λmax (log ꢀ) 358 (4.74), 376 (4.75), 429 (5.27), 488
(4.15), 514 (4.31), 552 (4.41), 605 (3.94), 667 nm (3.54); 1H NMR
(CDCl3) δ -7.01 (1H, s), -4.15 (2H, v br), 1.80-1.85 (9H, m),
1.90 (3H, t, J ) 7.4 Hz), 3.52 (3H, s), 3.55 (3H, s), 3.89 (4H, q, J
) 7.8 Hz), 4.03 (2H, q, J ) 7.7 Hz), 4.12 (2H, q, J ) 7.7 Hz),
7.53-7.57 (1H, m), 7.60-7.64 (2H, m), 7.71-7.74 (2H, m), 7.77
(1H, d, J ) 7.6 Hz), 9.00 (1H, d, J ) 7.6 Hz), 9.59 (1H, s), 9.65
(1H, s), 9.90 (1H, s), 10.62 (1H, s), 11.38 (1H, s); 1H NMR (trace
acid-CDCl3; monocation) δ -6.51 (1H, s), -3.3 (1H, v br), -1.94
(1H, br s), -1.78 (1H, br s), 1.73 (3H, t, J ) 7.8 Hz), 1.78 (3H, t,
J ) 7.6 Hz), 1.87 (6H, t, J ) 7.4 Hz), 3.54 (3H, s), 3.56 (3H, s),
4.03-4.13 (8H, m), 7.52-7.57 (1H, m), 7.61 (2H, t, J ) 7.2 Hz),
7.71 (2H, d, J ) 7.2 Hz), 7.78 (1H, d, J ) 7.6 Hz), 8.97 (1H, d,
J ) 7.6 Hz), 9.99 (1H, s), 10.01 (1H, s), 10.36 (1H, s), 10.50 (1H,
s), 11.74 (1H, s); 13C NMR (CDCl3) δ 11.2, 17.4, 17.5, 18.6, 20.0,
20.1, 20.2, 94.7, 95.6, 98.1, 105.4, 111.2, 124.1, 128.1, 128.6, 128.7,
130.0, 130.1, 130.6, 130.9, 131.8, 132.6, 134.0, 134.2, 137.5, 138.6,
140.2, 140.8, 142.0, 142.7, 144.4, 144.7, 145.5, 153.3, 153.8,
196.2; HRMS (EI) calcd for C42H41N3O m/z 603.3250, found
603.3253.
2
Fo2 > 2σ(Fo ). Limiting indicies were as follows: -13 e h e 13,
-15 e k e 15, -17 e l e 17. Data reduction were accomplished
using SAINT.34b The data were corrected for absorption using the
SADABS procedure.34c
Solution and data analysis were performed using the WinGX
software package.36 The structure of 16a was solved by direct
methods using the program SIR200437 and the refinement was
completed using the program SHELX-97.38 All non-hydrogen atoms
were refined anisotropically. The H atoms attached to atoms C21
and N23 were identified through difference Fourier synthesis and
refined with isotropic displacement parameters. All other H atoms
were included in the refinement in the riding-model approximation
(C-H Ar, CH3, CH2 ) 0.95, 0.98 Å and 0.99 Å), with isotropic
displacement parameters fixed at Ar, CH2 ) 1.2Ueq and CH3 )
1.5Ueq of the parent atom. Atoms C7A and C18A were refined as
idealized disordered methyl groups with H atoms included using
the HFIX 123 instruction in SHELXL97. Full-matrix least-squares
refinement on F2 led to convergence, (∆/σ)max ) 0.000, (∆/σ)mean
) 0.0000, with R1 ) 0.0640 and wR2 ) 0.1509 for 7749 data with
2
2
Fo > 2s(Fo ) using 0 restraints and 396 parameters. A final
difference Fourier synthesis showed features in the range of ∆Fmax
) 0.416 e-/Å3 to ∆Fmin ) -0.266 e-/Å3 which were deemed of
no chemical significance. Molecular diagrams were generated using
ORTEP-III39 and POV-Ray.40 X-ray structural data has been
deposited with the CCDC.41
Acknowledgment. This material is based upon work sup-
ported by the National Science Foundation under Grant No.
CHE-0616555 (to T.D.L.), the Camille and Henry Dreyfus
Foundation, and the Petroleum Research Fund, administered by
(34) (a) Bruker SMART CCD software package, Version 5.630; Bruker
Advanced X-ray Solutions: Madison, WI, 1997-2002. (b) Bruker SAINT+
Integration Software for Single Crystal Data frames - h,k,l, intensity, Version
6.45; Bruker Advanced X-ray Solutions: Madison, WI, 2003. (c) Bruker
SADABS-Empirical absorption correction produces; Bruker Advanced X-ray
Solutions: Madison, WI, 2003.
(35) McArdle, P. J. Appl. Crystallogr. 1996, 29, 306.
(36) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837-838.
(37) Burla, M. C.; Caliandro, R.; Camalli, M.; Carrozzini, B.; Cascarano,
G. L.; De Caro, L.; Giacovazzo, C.; Polidori, G.; Spagna, R. J. Appl.
Crystallogr. 2005, 38, 381-388.
(38) Sheldrick, G. M., SHELX-97, Programs for Crystal Structure
Analysis (Release 97-2), Institu¨t fu¨r Anorganische Chemie der Universita¨t,
Tammanstrasse 4, D-3400 Go¨ttingen, Germany, 1998.
(39) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
(40) Persistence of Vision Team, 2006, www.povray.org.
(41) Crystallographic data (excluding structure factors) for 16a has been
deposited with the Cambridge Crystallographic Data Center as supplemen-
tary publication number CCDC 650546. Copies of this data can be obtained,
free of charge, on application to CCDC, 12 Union Road, Cambridge
CB2 IEZ, UK [fax: +44 (0)12233 336033 or e-mail: deposit@ccdc.cam.
ac.uk].
Crystallographic Experimental Details of 16a. X-ray quality
crystals were obtained by layering hexane onto a dichloromethane
solution of 16a and allowing the solvents to gradually diffuse into
one another. The crystals were suspended in mineral oil at ambient
temperature and a suitable crystal was selected. A mineral oil coated
8414 J. Org. Chem., Vol. 72, No. 22, 2007

Syntheses and ReactiVity of meso-Unsubstituted Azuliporphyrins
Supporting Information Available: Synthetic procedures for
13-15, crystallographic data for 16a (CIF), and copies of the UV-
the American Chemical Society. G.M.F. also gratefully ac-
knowledges the National Science Foundation (Grant No. CHE-
0348158) for support. We thank Dr. Matthias Zeller and the
Youngstown State University Structure & Chemical Instrumen-
tation Facility for collecting the low-temperature CCD X-ray
data.
1
vis, IR, H NMR, and 13C NMR spectra for selected compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
JO701523S
J. Org. Chem, Vol. 72, No. 22, 2007 8415