Synthetic routes to 5,10,15-triaryl-tetrabenzocorroles

Article abstract of DOI:10.1021/jo200026u

Two different synthetic routes for the preparation of 5,10,15-triaryl- tetrabenzocorroles have been developed. The first approach is based on the condensation of a tetrahydroisoindole with aryl-aldehydes, and the second pathway involves a cross-coupling reaction between a bromo-substituted corrole and a suitable substrate to form the four β-fused aryl rings. These routes are successful to lead to the target tetrabenzocorroles, obtained in both cases as the corresponding Cu complex and with the highest yield obtained via the one-step cross-coupling methodology. Demetalation of the Cu-tetrabenzocorrole produces the corresponding free base; this macrocycle was further exploited to obtain the Sn and Co complexes in good yields.

Full text of DOI:10.1021/jo200026u

ARTICLE
pubs.acs.org/joc
Synthetic Routes to 5,10,15-Triaryl-tetrabenzocorroles
Giuseppe Pomarico,† Sara Nardis,† Roberto Paolesse,*,† Owendi C. Ongayi,‡ Brandy H. Courtney,‡
Frank R. Fronczek,‡ and Maria Grac-a H. Vicente*,‡
†Department of Chemical Science and Technology, University of Rome “Tor Vergata”, Via della Ricerca Scientifica 1, 00133 Rome, Italy
‡Department of Chemistry, Louisiana State University, Baton Rouge, Louisiana 70803, United States
S Supporting Information
b
ABSTRACT:
Two different synthetic routes for the preparation of 5,10,15-triaryl-tetrabenzocorroles have been developed. The first approach is
based on the condensation of a tetrahydroisoindole with aryl-aldehydes, and the second pathway involves a cross-coupling reaction
between a bromo-substituted corrole and a suitable substrate to form the four β-fused aryl rings. These routes are successful to lead
to the target tetrabenzocorroles, obtained in both cases as the corresponding Cu complex and with the highest yield obtained via the
one-step cross-coupling methodology. Demetalation of the Cu-tetrabenzocorrole produces the corresponding free base; this
macrocycle was further exploited to obtain the Sn and Co complexes in good yields.
’ INTRODUCTION
extension of the π-aromatic system, with consequent modifica-
tions of their physicochemical properties. Considering the
peculiar photophysical properties already reported for corrole
derivatives,10ꢀ13 tetrabenzocorroles could be of relevant interest
for both theoretical and practical reasons.
To prepare such a functionalized species, two different routes
could be followed: in the first one, the corrole ring is functionalized
with proper substituents, which can later lead to the β-fused
aromatic rings; in the second approach, which follows the most
common way for the benzoporphyrin syntheses,15ꢀ18 substituted
pyrroles are used to prepared the functionalized corrole ring.
Both routes are challenging in the case of corrole, because the
peripheral functionalization of such a macrocycle is a field still in
its infancy, while the synthetic routes to triarylcorrole from
functionalized pyrroles are not yet well-defined.
Corroles are among the first members of the wide family of
porphyrin analogues, as Johnson and Kay reported their first
preparation1 in the early 1960s, during the rush for the definition
of the vitamin B12 synthetic route.2 These macrocycles are in fact
characterized by a corrin-like molecular skeleton, with a direct
pyrroleꢀpyrrole link, while retaining an 18-electron aromatic π
system like porphyrins. After a long period of silence, corroles
regained popularity at the beginning of the new century, after the
discovery of simple synthetic procedures for the preparation of
5,10,15-triarylcorroles.3ꢀ7 The possibility to prepare corroles
from commercially available precursors has allowed the great
explosion of corrole related research and a better exploration of
the intriguing chemistry of such a macrocycle, including inves-
tigations on its coordination chemistry,8,9 photophysics,10ꢀ13
and chemical reactivity,14 which have opened the way for the
application of corrole derivatives in the above-mentioned fields.
These synthetic advances have also allowed the investigation
of more complex molecular architectures based on corroles, with
the aim to investigate structureꢀproperties relationships and to
tune them for a particular application.
In this paper we investigate two different routes for the
preparation of tetrabenzocorroles, following both the direct
synthesis from pyrrole or the functionalization of a preformed
corrole via a coupling methodology.
’ RESULTS AND DISCUSSION
Among the different molecule constructs, a particularly inter-
esting structure is represented by tetrabenzocorroles, the species
represented in Figure 1.
Our first approach to tetrabenzocorrole synthesis involved
their preparation from functionalized pyrrole rings. Following a
These derivatives could be of particular interest, because
the benzene rings fused at the β-pyrrolic positions lead to the
Received: January 11, 2011
Published: April 11, 2011
r
2011 American Chemical Society
3765
dx.doi.org/10.1021/jo200026u J. Org. Chem. 2011, 76, 3765–3773
|

The Journal of Organic Chemistry
ARTICLE
route developed for the preparation of tetrabenzoporphyrins,18
we decided to use 4,5,6,7-tetrahydroisoindole as starting material,
reacting it with a corresponding arylaldehyde, adapting the
protocol normally used for the preparation of triarylcorroles6
(Scheme 1).
The product of this reaction is triaryl-tetrabutanocorrole
(1a,b), which can be converted to the final tetrabenzocorrole
upon oxidation of the peripheral rings.
on the yield of tetrabutanocorrole formation was detected
(Table 1).
The exploitation of BF3 OEt2 as the acidic catalyst, according
3
to the Lindsey procedure,20,21 gave only a slightly improvement
of the reaction yield (Table 1). To avoid the decomposition of
free base corrole during the purification process, we prepared its
copper complex, with the aim to stabilize the corrole ring toward
oxidative decomposition.
We decided to use tetrahydroisoindole as starting pyrrole,
because it is readily synthesized19 and has been successful
employed for the synthesis of tetrabenzoporphyrins.18
The coordination of copper ion to the corrole ring was readily
achieved, and the course of the reaction could be followed by the
appearance of a red spot on TLC and by UVꢀvis analysis, which
showed the appearance of a Soret band blue-shifted to 409 nm,
and a main Q-band at 552 nm.
The reagents (pyrrole/aldehyde ratio 5/1) were dissolved in
CH2Cl2 and stirred for 10 min under inert atmosphere before the
addition of TFA as catalyst, and the resulting mixture was stirred
for 90 min. The course of the reaction was monitored by UVꢀvis
spectroscopy, and the presence of a strong absorption band
between 400 and 450 nm in the optical spectrum upon oxidation
of an aliquot of the reaction mixture with DDQ confirmed the
formation of an aromatic macrocycle. TLC analysis of the
reaction mixture showed a complex mixture of several spots
other than that corresponding to the desired corrole. The
attempts to purify this complex mixture failed, because of the
low stability of the corrole free base, which mostly decomposed
during the chromatographic separation. For this reason only a
small quantity of pure corrole was isolated. The UVꢀvis spectrum
of this compound was consistent with the formation of an aromatic
macrocycle, as indicated by a Soret band at 431 nm and three Q
bands at 531, 570, and 648 nm. When the reaction was carried out
using the more reactive methyl-4-formylbenzoate, no improvement
The reaction mixture was purified by silica gel plug, eluted
with chloroform, followed by preparative TLC on alumina,
eluted with 55:45 CHCl3/hexane. The first eluted band was
identified as the corresponding Cu-tetrabutanoporphyrin, and
the second red brownish band corresponded to the desired
1
corrole 2a, which was obtained in 8% yield. On the H NMR
spectrum of 2a two main groups of signals were seen, one related
to the phenyl protons in the range 7.23ꢀ7.38 ppm and a
multiplet between 1.12 and 1.64 ppm for the protons on the
cyclohexane rings. A broad peak at 2.96 ppm belongs to the four
protons closer to the two directly bonded pyrrole units.
Mass analysis (MALDI) showed a molecular peak at 803.94
m/z, confirming the formation of 2a, which was further char-
acterized by X-ray crystallographic analysis of a chloroform
solvate (Figure 2).
In both of the two molecules in the asymmetric unit, the
coordination of the Cu is square planar with a slight pyramidal
distortion. The CuꢀN distances are in the range 1.894-
(3)ꢀ1.922(3) Å, with a mean value of 1.905 Å. The pyramidal
twist is such that the two N atoms opposite the five-membered
chelate ring lie on average (0.31 Å out of the CuN2 plane of the
five-membered ring. The 24-atom Cu-corrole system is reason-
ably planar, with mean deviation 0.21 Å and maximum deviation
0.503(5) Å from coplanarity. The distortion observed in 2a is
similar to the conformations observed for other copper corroles,
and this saddling effect has been recently reported by Ghosh as a
common feature of such metal derivatives.22,23 This conforma-
tion has been in fact observed regardless the presence of steric
congestion at the peripheral position of the macrocycle and has
been attributed to the noninnocent nature of copper corroles.23
Figure 1. Molecular structure of a triaryl-tetrabenzocorrole.
Scheme 1. Condensation Route to Benzocorrole Formation
3766
dx.doi.org/10.1021/jo200026u |J. Org. Chem. 2011, 76, 3765–3773

The Journal of Organic Chemistry
ARTICLE
Table 1. Yield of Benzocorrole Formation via Condensation Route
aldehyde
ratio pyrrole/aldehyde
solvent
catalyst
TFA
yield of 1 (%)
benzaldehyde
5/1
5/1
5/1
2/1
CH2Cl2
<1
<1
8a
methyl-4-formylbenzoate
benzaldehyde
CH2Cl2
TFA
CH2Cl2
BF3 OEt2
3
3-cyanobenzaldehyde
a Calculated for copper complex.
CH3OH/H2O
HCl
11
Figure 2. Molecular structure of 2a.
Figure 3. Molecular structure of 3a.
The oxidation of the peripheral rings to obtain the desired
tetrabenzocorrole was carried out with DDQ in refluxing
toluene; the progress of the reaction was monitored by UVꢀvis
spectroscopy. After 10 min, the solution color turned from red to
green; both TLC and the UVꢀvis spectrum confirmed the
disappearance of starting material and the formation of a new
compound.
The reaction was stopped, and the resulting mixture was
purified by a short alumina plug eluted with CHCl3. The UVꢀvis
spectrum of 3a showed a red-shifted and split Soret band with
λmax at 447 and 459 nm and a very intense Q-band at 647 nm,
with a shoulder at 598 nm. These features were consistent with
the formation of the fused benzene rings at the β-pyrrolic
positions of the corrole ring, since the extension of the β-aromatic
system is expected to cause a red shift of the absorption bands
and an increase of intensity of the Q bands, as observed in the
case of benzoporphyrins.
However, the resulting complex 3a showed low solubility in
different solvents, which hampered its 1H NMR characterization
due to the broadness of the signals. This result could be ascribed
to a paramagnetic character of the complex or to the aggregation
of 3a, supported also by the observed splitting of the Soret band.
The formation of 3a, with an overall yield of 6% from tetrahy-
droisoindole, was unambiguously confirmed both by MALDI-
MS, by which a molecular peak was observed at 787.73 m/z, and
by X-ray crystal structure determination, performed on a single
crystal of a chloroform solvate, obtained by the slow diffusion of
methanol in chloroform (Figure 3).
The asymmetric unit contains two corrole complex molecules,
one of which lies on a 2-fold axis, and they differ in confor-
mation. The 24-atom Cu-corrole system is quite planar for the
C2-symmetric molecule, withmeandeviation0.014 Å andmaximum
deviation 0.02(1) Å from coplanarity. The molecule not lying on
a C2 axis has a much less planar corrole, with mean deviation
0.18 Å and maximum 0.46(1) Å, similar to values seen in 2a. The
distortion is a twisted saddle. Coordination of the Cu atoms is
square planar, with CuꢀN distances in the range 1.906(11)ꢀ
1.940(13) Å, and a mean value of 1.922 Å over the two
independent molecules. It is interesting to note that in this case
the characteristic saddling distortion of copper corroles is far less
evident for both conformations of 3a, suggesting that the
extension of the aromatic system drives the macrocycle toward
a more planar conformation.
While these results showed the successful preparation of the
target benzocorrole, important drawbacks were represented by
the low solubility of the target product and the relative low yields
obtained for this synthetic route. To solve these problems, we
turned our attention to the method reported by Gryko7
(Scheme 1, 1cꢀ3c), using the more reactive 3-CN-benzalde-
hyde. This aldehyde and tetrahydroisoindole were dissolved in
methanol and an aqueous solution of HCl was added. The
3767
dx.doi.org/10.1021/jo200026u |J. Org. Chem. 2011, 76, 3765–3773

The Journal of Organic Chemistry
ARTICLE
Figure 4. UVꢀvis spectra of 1c, 2c, and 3c in CH2Cl2 (5ꢀ10 μM
concentration).
mixture was stirred at room temperature for 3 h and extracted
with chloroform, and after workup, the organic phase containing
bilane was oxidized with chloranil.
The yields obtained for 1c were increased to 11%, and the
resulting corrole also showed good solubility in several solvents,
thus allowing us to better characterize the product.
Due to the instability of the free base corrole during chroma-
tographic separation, only a small amount of free base was
purified for characterization purposes, while the remaining of
the reaction mixture was converted to the complex 2c.
The free base corrole 1c was purified by silica gel column
chromatography eluted with CH2Cl2/CH3OH/TEA (98:1:1).
The addition of TEA in the eluting mixture was necessary to
avoid the protonation of the corrole ring on silica. UVꢀvis
characterization of purified 1c (Figure 4) was consistent with
corrole formation, showing a red-shifted Soret band at 434 nm
and three Q bands at 525, 576, and 650 nm.
Figure 5. X-ray crystal structure of 4.
π-conjugation with β-fused aromatic rings and is in the same
range of the Soret band changes noticed in the case of analogous
tetrabenzoporphyrins, as reported in the literature.24
The 1H NMR spectrum of 3c showed no signals below 7 ppm,
supporting the formation of a molecule with only aromatic
protons. Mass spectrometry confirmed that aromatization took
place, as suggested by the molecular peak at 861.87 m/z.
Crystals of 3c were grown by slow diffusion of methanol into
the CDCl3 solution exploited for the 1H NMR analysis; however,
X-ray crystal structure determination gave an unexpected result,
because the molecular structure revealed that the compound was
the triaryl-tetrabenzoporphyrin 4 (Figure 5).
1
The H NMR spectrum of 1c showed several signals in the
region 7.89ꢀ8.38 ppm corresponding to the phenyl protons, and
a multiplet, between 1.76 and 2.74 ppm, which belongs to the
alkyl protons, while protons adjacent to C2 and C18 appeared as
a broad signal centered at 4.09 ppm. MALDI mass spectrum of 1c
showed a molecular peak at 818.77 m/z.
The copper complex was prepared using the same methodol-
ogy described above, and product 2c was purified by silica gel
column chromatography eluted with CH2Cl2. The main red
band corresponding to Cu-corrole 2c was collected and crystal-
lized from CH2Cl2/CH3OH. The UVꢀvis spectrum of 2c shows
a Soret band at 414 nm, blue-shifted by 20 nm, and only two Q
bands, at 549 and 600 nm, instead of three, due the higher symmetry
of 2c compared with 1c, induced by metal coordination.
This porphyrin was likely formed during the slow crystal-
lization step, in a process catalyzed by air and light, as previously
described.25 The reaction probably involved two molecules of
corrole, followed by a rearrangement of the intermediate to form
one porphyrin molecule (i.e., 4) and an open chain benzobili-
verdine-like derivative, which was not isolated.
The overall yield of 3c, based on the amount of aldehyde used,
was only 8%. Therefore we decided to explore alternative synthetic
routes. Considering previous works in the porphyrin field,26,27
we decided to explore a route where the first step is the func-
tionalization of the preformed corrole ring with appropriate
substituents, which can later allow the buildup of the β-fused
benzene rings.
1
On the other hand, the H NMR spectrum of 2c was not
deeply affected by metal insertion, and only a slight shift of
the peaks was observed. Mass spectrum analysis gave further
proof of the structure of Cu-corrole 2c, with the molecular peak
at 878.79 m/z.
The oxidation of the peripheral cyclohexane rings to obtain 3c
was carried out with DDQ in refluxing toluene. After 10 min the
solvent was removed, and the mixture was purified using neutral
alumina column chromatography, eluted with CHCl3. The
UVꢀvis spectrum of the main green fraction has a 30 nm red-
shifted Soret band at 437 nm with respect to its precursor, and a
strong Q-band at 641 nm. The red shift of the main absorption
band observed going from 2c to 3c is ascribed to the extended
A recently reported route for the preparation of tetrabenzo-
porphyrins involves a “cross-coupling” methodology, based on
the Heck procedure.27 This route allows direct formation of the
final product in a one-pot reaction, with no isolation of inter-
mediate compound needed (Scheme 2).
CuBr8TPC (5a), prepared following literature methods,6,28,29
was reacted with methyl acrylate under the Heck conditions for 3
days at 125 °C. After removal of the solvent, the residue was
3768
dx.doi.org/10.1021/jo200026u |J. Org. Chem. 2011, 76, 3765–3773

The Journal of Organic Chemistry
ARTICLE
Scheme 2. Benzocorroles via a Heck Cross-Coupling Reaction
purified by alumina (grade III) column chromatography, eluted
with chloroform.
spectrum of 6b three sets of signals were observed, with the
singlet at 1.42 ppm assigned to the tert-butyl groups (27
protons), and the other two groups of signals, down shielded
with respect to the tert-butyl singlet, appeared in the range 3ꢀ4
ppm (-CO2CH3, 24 H) and 7ꢀ8 ppm (aromatic protons, 20 H).
The MALDI mass spectrum showed the molecular peak of 6b at
1422.41 m/z. It is interesting to note that also in this case the
UVꢀvis spectrum showed a broadened and red-shifted Soret
band, around 471 nm. This red shift could be reasonably
attributed to the electronic effect of the methoxycarbonyl
substituents, as already observed in the case of the corresponding
tetrabenzoporphyrins,24 considering also that the analogous
complex 3a shows the Soret band at 447 nm.
The yield for this reaction was 16%, thus allowing the preparation
of a larger amount of the free base, which was used as starting
material for the preparation of different metal complexes. These
complexes were prepared via demetalation30 of 6b to give the
corresponding free base 7, followed by insertion of Sn (8) orCo(9).
The copper corrole was demetalated upon addition of a few
drops of concentrated H2SO4 to a chloroform solution of the
complex.
The desired product was collected and further purified by
column chromatography on alumina (grade IV) eluted with
CH2Cl2. The UVꢀvis spectrum of 6a showed a Soret band
strongly red-shifted at 454 nm, with no intense Q-bands, while
the 1H NMR spectrum showed several signals between 3 and 4
ppm belonging to the -CO2CH3 groups, and a second cluster of
multiplets in the aromatic region from 7 to 8 ppm.
It is worth noting the absence of signals around 6 ppm
characteristic of unreacted olefin, supporting the ring closure of
the peripheral substituents to give the β-fused aromatic rings.
The benzocorrole formation was confirmed by MALDI mass
spectrum, which showed the molecular peak at 1252.82 m/z.
The reaction was repeated starting with 5b in place of 5a, to
increase the solubility of the corresponding corrole 6b and to
facilitate its characterization. The reaction was carried out as
described above, and the reaction mixture was purified by neutral
alumina (grade III) column chromatography eluted with CHCl3.
As expected, the increased solubility of 6b allowed the complete
spectroscopic characterization of the complex. In the 1H NMR
3769
dx.doi.org/10.1021/jo200026u |J. Org. Chem. 2011, 76, 3765–3773

The Journal of Organic Chemistry
ARTICLE
(Figure 7), typical signatures of pyridine coordination to form
the hexacoordinated complex. Also in this case MALDI MS
confirmed the complex formation with a molecular ion peak at
1418.39 m/z, with the loss of the axial pyridine ligands.
1
It should be noted that for both complexes 8 and 9 the H
NMR characterization showed severe line broadening, with
broad multiplets for the aromatic signals and unresolved groups
of singlets for the carboxymethyl and tert-butyl substituents. This
behavior could be reasonably attributed to complex dynamic
processes, involving conformers generated by macrocycle dis-
tortions upon metal coordination. Crystallographic characteriza-
tion of these complexes would be necessary to clarify such a
point, but unfortunately all our attempts to obtain single crystals
suitable for X-ray structural analysis were unsuccessful.
’ CONCLUSIONS
Figure 6. UVꢀvis spectra of 7 (blue line) at 10 μM in CH2Cl2 and
Two different approaches for the preparation of 5,10,15-
triaryl-tetrabenzocorroles have been investigated, and the first
examples of such corrole derivatives were synthesized. The first
synthetic route involves the condensation of a benzaldehyde
with tetrahydroisoindole, leading to the formation of the
tetrabenzocorrole in moderate yields, upon oxidation of the
intermediate tetrabutanocorrole. The alternative approach
exploits an intermediate octabromocorrole, which gives the
target tetrabenzocorrole in one step after reaction with methyl
acrylate under the Heck conditions; this second route could
result in an easier approach, avoiding the preparation of the
tetrahydroisoindole reagent and giving moderate yields of the
final tetrabenzocorrole.
upon addition of one drop of TFA (red line) or TEA (green line).
Tetrabenzocorroles show chemical stability similar to that of
the corrole precursors, and we observed a corrole-porphyrin ring
expansion transformation upon standing for several weeks in
solution.
Tetrabenzocorroles show interesting UVꢀvis spectra, with
features similar to those of the related tetrabenzoporphyrins; a
detailed study of their photophysical properties is currently
under way in our laboratories and these results will be reported
in due time.
Figure 7. UVꢀvis spectra of benzocorrole metal complexes (5ꢀ20 μM
in CH2Cl2).
The formation of the tetrabenzocorrole cation was indicated
by the appearance of a peak around 730 nm in the UVꢀvis
spectrum. Variations in the UVꢀvis spectrum of the free base
upon addition of acid (TFA) or base (TEA) are shown in
Figure 6, indicating the formation of the characteristic cationic
and anionic species. These changes are similar to those observed
in the case of unsubstituted triarylcorroles31 and confirm the
peculiar NꢀH acidicity also observed for tetrabenzocorrole. This
feature has been attributed to the nonplanar structure of corrole
free bases, caused by the presence of three inner core protons,
and the related steric release experienced upon deprotonation,
which can be induced even by weak bases like TEA. This steric
effect explains the limited range of stability of free base corrole
and is similar to the instability of monoprotonated porphyrin
species.
The Sn complex was synthesized by reaction of the free base 7
with SnCl2 in refluxing DMF. The isolated green complex 8
showed a UVꢀvis spectrum characterized by a red shifted and
narrow Soret band, as seen in Figure 7. The MALDI MS con-
firmed the formation of this complex, showing a parent peak at
1667.05 m/z due to the coordination of the R-cyano-4-hydro-
xycinnamic acid, used as matrix for the MALDI measurement.
The Co complex 9 was prepared by reaction of 7 with
Co(AcO)2 in pyridine; this compound shows a further red shift
of the Soret band (499 nm) and an intense Q band at 651 nm
’ EXPERIMENTAL SECTION
2-Carboxyethyl-3,4,5,6-tetrahydroisoindole19. Mp: 80ꢀ82 °C.
1H NMR (CDCl3, 300 MHz, ppm): δ 8.77 (s br, 1H), 6.64 (d, J = 2.7 Hz,
1H), 4.29 (q, J = 8.4 Hz, 2H), 2.82 (t, J = 6.0 Hz, 2H), 2.54 (t, J = 5.5 Hz,
2H), 1.74 (m, 4H) 1.34 (t, J = 8.4 Hz 3H). MS (MALDI, m/z): 193.12
(Mþ).
19
1
3,4,5,6-Tetrahydroisoindole . H NMR (CDCl3, 300 MHz,
ppm): δ 8.27 (s br, 1H), 6.11 (d, J = 2.7 Hz, 2H), 2.39 (t, J = 2.7 Hz, 4H),
1.52 (m, 4H).
5,10,15-Triphenylcorrole6. UVꢀvis (CH2Cl2), λmax nm (log ε):
415 (5.04), 567 (4.20), 615 (4.10), 648 (4.02). 1H NMR (CDCl3, 300
MHz, ppm): δ 8.93, 8.86, 8.58, 8.53 (each d, 2H, J = 4 Hz), 8.39 (d, 4H,
J = 7 Hz), 8.16 (d, 2H, J = 5 Hz), 7.82ꢀ7.72 (m, 9H), ꢀ2.91 (s, 3H). MS
(FAB, m/z): 527 (Mþ). Anal. Calcd for C37H26N4: C, 84.39; H, 4.98; N,
10.64. Found: C, 84.24; H, 4.92; N, 10.49.
(5,10,15-Triphenylcorrolato)Cu28. UVꢀvis (CH2Cl2), λmax nm
(log ε): 410 (5.09), 538 (3.89). 1H NMR (CDCl3, 300 MHz, ppm): δ
7.89 (br s, 2H), 7.76 (d, J = 7.3 Hz, 4H), 7.65 (d, J = 7.3 Hz, 4 H),
7.59ꢀ7.54 (m, 3H), 7.51ꢀ7.44 (m, 6H), 7.35 (d, J = 3.1 Hz, 2H), 7.24
(d, J = 3.1 Hz, 2H). MS (MALDI-TOF, m/z): 586.96 (Mþ). Anal. Calcd
for C37H23N4Cu: C, 75.69; H, 3.95; N, 9.54. Found: C, 75.57; H, 3.92;
N, 9.45.
3770
dx.doi.org/10.1021/jo200026u |J. Org. Chem. 2011, 76, 3765–3773

The Journal of Organic Chemistry
ARTICLE
(2,3,7,8,12,13,17,18-Octabromo-5,10,15-triphenylcorrolato)-
Cu (5a)29. UVꢀvis (CH2Cl2), λmax nm (log ε): 454 (4.93), 653 (3.80).
1H NMR (CDCl3, 300 MHz, ppm): δ 7.67ꢀ7.63 (m, 2H), 7.53ꢀ7.39
(m, 13H). MS (MALDI-TOF, m/z): 1217.65 (Mþ). Anal. Calcd for
C37H15N4Br8Cu: C, 36.48; H, 1.24; N, 4.60. Found: C, 36.39; H, 1.28;
N, 4.52.
mixture was stirred at room temperature for 90 min, monitoring the
course via UVꢀvis spectrometry.
After the appearance of the typical corrole free base absorption bands,
an aliquot (20%) of the reaction mixture was removed and solvent was
evaporated to dryness. Residue was dissolved with CH2Cl2, passed on a
silica gel column eluted with CH2Cl2/CH3OH/TEA (98:1:1). A green
fraction was collected and crystallized from CH2Cl2/CH3OH, to give
the titled corrole (13 mg, 11% yield). UVꢀvis (CH2Cl2), λmax nm
(log ε): 434 (5.12), 525 (4.25), 576 (4.46), 650 (4.18). 1H NMR (CDCl3,
300 MHz, ppm): δ 8.38 (s, 3 H), 8.36ꢀ8.34 (m, 3 H), 8.06ꢀ8.03 (d, J =
7.17 Hz, 3H), 7.89ꢀ7.83 (m, 3 H), 4.09 (br m, 4 H), 2.74ꢀ2.70 (m, 8 H),
2.44 (br m, 4 H), 2.28ꢀ2.25 (m, 4 H), 1.94ꢀ1.92 (m, 4 H), 1.79ꢀ1.76
(m, 8 H). MS (MALDI, m/z): 818.77 (Mþ). Anal. Calcd for C56H47N7:
C, 82.22; H, 5.79; N, 11.99. Found: C, 82.13; H, 5.69; N, 11.87.
[5,10,15-Tris(3-cyanophenyl)-2:3,7:8,12:13,17:18-tetra-
butanocorrolato]Cu (2c). To the mixture containing corrole 1b was
added a solution of Cu(AcO)2 in CH3OH, and the mixture was stirred
overnight at room temperature, monitoring the course of the reaction by
UVꢀvis spectrometry. When analysis showed complex formation,
solvent was removed under vacuum; purification by column chroma-
tography (silica gel, CH2Cl2) afforded a red fraction that was crystallized
from CH2Cl2/CH3OH. (50 mg, 90% yield). UVꢀvis (CH2Cl2),
λmax nm (log ε): 414 (5.12), 549 (4.04), 600 (3.71). 1H NMR
(CDCl3, 300 MHz, ppm): δ 7.79ꢀ7.64 (m, 8 H), 7.55ꢀ7.48 (m, 4 H),
3.01 (br m, 4 H), 1.74 (br s, 6 H), 1.65 (br m, 6 H), 1.52 (br m, 8 H), 1.38
(br m, 8 H). MS (MALDI, m/z): 878.79 (Mþ). Anal. Calcd for C56H44-
N7Cu: C, 76.56; H, 5.05; N, 11.16. Found: C, 76.47; H, 4.99; N, 11.02.
[5,10,15-Tris(3-cyanophenyl)-2:3,7:8,12:13,17:18-tetra-
benzocorrolato]Cu (3c). Corrole 2c (24 mg, 0.027 mmol) was
dissolved in toluene, and DDQ (59 mg, 0.26 mmol) was added to the
mixture before heating at reflux for 10 min. The progress of the reaction
was followed by TLC (alumina) for the appearance of a green streak at
the baseline and by UVꢀvis spectrometry for the red shifting of the
absorption bands. Once the reaction was complete, the mixture was
allowed to cool, filtered through a short alumina plug eluted with
chloroform, and then was purified again on column chromatography
(alumina, CH2Cl2 then CHCl3). A green fraction was collected, crystal-
lized from CH2Cl2/CH3OH (18 mg, 77% yield). UVꢀvis (CH2Cl2),
λmax nm (log ε): 437 (4.94), 641 (4.39). 1H NMR (CDCl3, 300 MHz,
ppm): δ 7.80 (br m, 7 H), 7.71 (br m, 4 H), 7.69 (br m, 4 H), 7.66 (br m,
3 H), 7.62 (br m, 4 H), 7.52 (br m, 2 H), 7.49 (br m, 3 H), 7.47 (br m, 1
H). (MALDI, m/z): 861.87 (Mþ). Anal. Calcd for C56H28N7Cu: C,
77.99; H, 3.27; N, 11.37. Found: C, 77.88; H, 3.18; N, 11.29.
5,10,15-Tris(4-tert-butylphenyl)corrole6. UVꢀvis (CH2Cl2),
λmax nm (log ε): 418 (5.11), 573 (4.25), 619 (4.22), 651 (4.14) nm. 1H
NMR (CDCl3, 300 MHz, ppm): δ 8.94 (br s, 4H), 8.63 (br s, 4H), 8.33
(br d, 4H), 8.13 (br d, 2H), 7.84 (br d, 4H), 7.79 (br d, 4H), 1.61 (s, 27
H). MS (FAB, m/z): 695 (Mþ). Anal. Calcd for C49H50N4: 84.69; H,
7.25; N, 8.06 Found: C, 84.56; H, 7.18; N, 7.99.
(5,10,15-Tris(4-tert-butylphenyl)corrolato)Cu (5b)28. UVꢀvis
(CH2Cl2), λmax nm (log ε): 421 (5.17), 539 (3.92), 627 (3.68). 1H NMR
(CDCl3, 300 MHz, ppm): δ 7.92 (br s, 2H), 7.75 (m, 6H), 7.67 (br s, 2H)
7.50 (m, 6H), 7.43 (m, 2H), 7.35 (m, 2H), 1.56 (s, 18H), 1.44 (s, 9H). MS
(FAB):m/z755 (Mþ). Anal. Calcd for C49H47N4Cu: C, 77.90; H, 6.27; N,
7.41 Found: C, 77.82; H, 6.19; N, 7.34.
[2,3,7,8,12,13,17,18-Octabromo-5,10,15-tris(4-tert-butyl-
phenyl)corrolato]Cu. This compound was prepared following lit-
erature method.29 UVꢀvis (CH2Cl2), λmax nm (log ε): 456 (4.91), 655
(3.79). 1H NMR (CDCl3, 300 MHz, ppm): δ 7.78 (br s, 8H), 7.44 (br s,
4H), 1.44 (s, 18H), 1.42 (s, 9H). Anal. Calcd for C49H39N4Br8Cu: C,
42.44; H, 2.83; N, 4.04. Found: C, 42.35; H, 2.86; N, 3.99.
(5,10,15-Triphenyl-2:3,7:8,12:13,17:18-tetrabutanocorro-
lato)Cu (2a). Tetrahydroisoindole (250 mg, 2.1 mmol) was dissolved
in 5 mL of dichloromethane and allowed to stir for 10 min. Benzalde-
hyde (0.04 mL, 0.41 mmol) and 5 μL of BF3 OEt2 (0.03 mmol) were
3
then added to the reaction and allowed to stir for another 2 h. DDQ was
then added, and mixture was allowed to stir for 6 h at room temperature.
To the mixture containing 5,10,15-triphenyl-2:3,7:8,12:13,17:18-tetra-
butanocorrole (1a) was added a saturated solution of Cu(AcO)2 in
methanol was added, and the mixture was stirred at room temperature
for 24 h. The metalation was considered complete when a Soret band at
409 nm was evident. The reaction mixture was then purified through a
silica gel plug, eluting with chloroform, and the first red band eluted was
collected. The red fraction was dissolved in dichloromethane and
purified on an alumina gel preparative TLC eluting with 55:45 chloro-
form/hexane. The first band was the Cu(II)-tetrabutanoporphyrin, and
the second reddish brownish band was the desired 2a (9 mg, 8% yield).
UVꢀvis (CH2Cl2), λmax nm (log ε): 409 (4.39), 552 (3.56). 1H NMR
(CDCl3, 300 MHz, ppm): δ 7.38ꢀ7.23 (m, 15 H), 2.96 (s br, 4H),
1.64ꢀ1.12 (m, 28 H). MS (MALDI, m/z): 803.94 (Mþ). Anal. Calcd for
C53H47N4Cu: C, 79.22; H, 5.90; N, 6.96. Found: C, 79.03; H, 5.68;
N, 6.79.
[5,10,15-Triphenyl-2:3,7:8,12:13,17:18-tetrabenzo-200,300,700,
800,1200,1300,1700,1800-octacarboxymethylcorrolato]Cu (6a).
Corrole 5a (60 mg, 0.049 mmol), Pd(AcO)2 (11 mg, 0.049 mmol), PPh3
(34 mg, 0.128 mmol), K2CO3 (58 mg, 0.42 mmol), and methyl acrylate
(437 μL, 4.9 mmol) were dissolved in anhydrous toluene/DMF (8 þ
10 mL) in a Schlenk tube under nitrogen. The mixture was then
degassed via five freezeꢀpumpꢀthaw cycles before the vessel was
purged with nitrogen again, then sealed, and heated at 125 °C for 72 h.
Solvents were removed, and the residue purified with a neutral alumina
column eluted with CHCl3. Fractions containing corrole were combined
and purified again with a neutral alumina (grade IV) column eluted with
CH2Cl2. Brown fraction collected was crystallized from CH2Cl2/
Hexane. (8 mg, 13% yield). UVꢀvis (CH2Cl2), λmax nm (log ε): 454
(5,10,15-Triphenyl-2:3,7:8,12:13,17:18-tetrabenzocorrolato)-
Cu (3a). Compound 2a (20 mg, 0.02 mmol) was dissolved in toluene,
and DDQ (45 mg, 0.19 mmol) was added to the mixture before heating
at reflux for 10 min. The reaction progress was followed by TLC
(alumina, chloroform) for the appearance of a green streak at the
baseline. Once the reaction was complete, the mixture was allowed to
cool, filtered through a very short alumina plug, and eluted with around
100 mL of chloroform. (12 mg, 77% yield). UVꢀvis (CH2Cl2), λmax nm
(log ε): 447 (4.57), 459 (4.5), 598 (3.78), 647 (4.1). MS (MALDI,
m/z): 787.73 (Mþ). Anal. Calcd for C53H31N4Cu: C, 80.85; H, 3.97; N,
7.11. Found: C, 80.63; H, 3.58; N, 7.02.
5,10,15-Tris(3-cyanophenyl)-2:3,7:8,12:13,17:18-tetrabu-
tanocorrole (1c). Tetrahydroisoindsole (500 mg, 4.1 mmol) and
3-cyano-benzaldehyde (269 mg, 2.05 mmol) were dissolved in CH3OH
(150 mL). Subsequently, a solution of HCl (36%, 7.5 mL) in H2O
(150 mL) was added, and the reaction was stirred at room temperature
for 3 h. The mixture was extracted with CHCl3, and the organic layer
washed twice with H2O, dried over Na2SO4, filtered, and diluted to
700 mL with CHCl3. Chloranil (1.47 g, 6.0 mmol) was added, and the
1
(4.90), 604 (4.16). H NMR (CDCl3, 300 MHz, ppm): δ 7.73ꢀ7.46
(m, 23 H), 3.98ꢀ3.57 (m, 24 H). MS (MALDI, m/z): 1252.82 (Mþ).
Anal. Calcd for C69H47N4O16Cu: C, 66.21; H, 3.78; N, 4.48. Found: C,
66.08; H, 3.69; N, 4.38.
[5,10,15-Tris(4-tert-butylphenyl)-2:3,7:8,12:13,17:18-tet-
rabenzo-200,300,700,800,1200,1300,1700,1800-octacarboxymethyl-
corrolato]Cu (6b). Corrole 5b (136 mg, 0.098 mmol), Pd(AcO)2
(22 mg, 0.098 mmol), PPh3 (70 mg, 0.256 mmol), K2CO3 (118 mg,
3771
dx.doi.org/10.1021/jo200026u |J. Org. Chem. 2011, 76, 3765–3773

The Journal of Organic Chemistry
ARTICLE
0.84 mmol), and methyl acrylate (874 μL, 9.8 mmol) were dissolved in
anhydrous toluene/DMF (16 þ 20 mL) in a Schlenk tube under
nitrogen. The mixture was then degassed via five freezeꢀpumpꢀthaw
cycles before the vessel was purged with nitrogen again, then sealed and
heated at 125 °C for 72 h. Solvents were removed and residue purified
with a neutral alumina column eluted with CHCl3. Brown fraction
collected was crystallized from CH2Cl2/hexane. (22 mg, 16% yield).
UVꢀvis (CH2Cl2), λmax nm (log ε): 471 (4.93), 602 (4.18). 1H NMR
(CDCl3, 300 MHz, ppm): δ 7.69ꢀ7.49 (m, 20 H), 3.95ꢀ3.47 (m, 24
H), 1.42 (br m, 27 H). MS (MALDI, m/z): 1422.41 (Mþ). Anal. Calcd
for C81H71N4O16Cu: C, 68.51; H, 5.04; N, 3.94. Found: C, 68.08; H,
4.84; N, 3.82.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: vicente@lsu.edu; roberto.paolesse@uniroma2.it.
’ ACKNOWLEDGMENT
This research was supported by MIUR ItalyꢀPRIN project
2007C8RW53 (R.P.) and partially supported from the U.S.
National Science Foundation grant number CHE-0911629 (M.
G.H.V.).
5,10,15-Tris(4-tert-butylphenyl)-2:3,7:8,12:13,17:18-tet-
rabenzo-200,300,700,800,1200,1300,1700,1800-octacarboxymethyl-
corrole (7). Corrole 6b (80 mg, 0.056 mmol) was dissolved in 8 mL of
CHCl3, and 1 mL of concentrated H2SO4 was added. An aliquot of the
reaction mixture was diluted in methanol to monitor the progress of the
demetelation process by UVꢀvis spectroscopy. After 5 min the forma-
tion of cation was observed, and distilled water (50 mL) was added to
quench the reaction. The organic phase was extracted with chloroform,
then washed twice with water, neutralized with aqueous NaHCO3, and
dried over Na2SO4 anhydrous. Solvent was removed, and the residue
was purified with a neutral alumina (grade IV) plug eluted with CHCl3.
Brown fraction collected was crystallized from CH2Cl2/hexane. (59 mg,
77% yield). UVꢀvis (CH2Cl2), λmax nm (log ε): 476 (4.72), 606 (3.94)
699 (3.86). 1H NMR (CDCl3, 300 MHz, ppm): δ 7.77ꢀ7.49 (m, 20 H),
4.01ꢀ3.51 (m, 24 H), 1.61 (br s, 27 H). Anal. Calcd for C81H74N4O16:
C, 71.56; H, 5.49; N, 4.12. Found: C, 71.47; H, 5.54; N, 4.06.
[5,10,15-Tris(4-tert-butylphenyl)-2:3,7:8,12:13,17:18-tet-
rabenzo-200,300,700,800,1200,1300,1700,1800-octacarboxymethyl-
corrolato]SnCl (8). Complex 6b (24 mg, 0.017 mmol) was
demetalated to give the corresponding free base 7, which was dissolved
in DMF with SnCl2 (15 mg, 0.079 mmol), and the mixture was stirred to
reflux under nitrogen for 3 h. The reaction was monitored by UVꢀvis
spectrometry and TLC (alumina, CH2Cl2ꢀCH3OH 5%). When the
reaction was complete, the solvent was removed and the residue purified
with short plug ran in the same condition of TLC. The fraction
containing 8 was crystallized from CH2Cl2/hexane (13 mg, 51% yield).
UVꢀvis (CH2Cl2), λmax nm (log ε): 478 (5.35), 599 (4.35), 642 (4.45).
1H NMR (CDCl3, 300 MHz, ppm): δ 8.37ꢀ7.63 (m, 20 H), 4.22ꢀ3.69
(m, 24 H), 1.37 (br m, 27 H). MS (MALDI, m/z): 1667.05 (Mþ). Anal.
Calcd for C81H71N4O16ClSn: C, 64.40; H, 4.74; N, 3.71. Found: C,
64.22; H, 4.26; N, 3.51.
[5,10,15-Tris(4-tert-butylphenyl)-2:3,7:8,12:13,17:18-tet-
rabenzo-200,300,700,800,1200,1300,1700,1800-octacarboxymethyl-
corrolato]CoPy2 (9). Complex 6b (24 mg, 0.017 mmol) was
demetalated to give the corresponding free base 7, which was dissolved
in 15 mL of pyridine with Co(AcO)2 (10 mg, 0.056 mmol). The mixture
was stirred to reflux for 2 h, monitoring the course by UVꢀvis spectro-
metry and TLC (alumina, CHCl3). After completion of the reaction, the
solvent was removed and the residue purified by a short plug on alumina,
eluting with CHCl3. The fraction containing 9 was crystallized from
CH2Cl2/CH3OH (17 mg, 63% yield). UVꢀvis (pyridine), λmax nm (log ε):
499 (4.70), 651 (4.21). 1H NMR (CDCl3, 300 MHz, ppm): δ
7.69ꢀ7.47 (m, 20 H), 4.14ꢀ3.54 (m, 24 H), 1.63 (br m, 27 H). MS
(MALDI, m/z): 1418.39 (Mþ ꢀ 2Py). Anal. Calcd for C91H81N6O16-
Co: C, 69.46; H, 5.19; N, 5.34. Found: C, 69.37; H, 5.06; N, 5.22.
’ REFERENCES
(1) Johnson, A. W.; Kay, I. T. J. Chem. Soc. 1965, 1620.
(2) Johnson, A. W. Chem. Soc. Rev. 1980, 9, 125.
(3) Paolesse, R.; Jaquinod, L.; Nurco, D. J.; Mini, S.; Sagone, F.;
Boschi, T.; Smith, K. M. Chem. Commun. 1999, 1307–1308.
(4) Gross, Z.; Galili, N.; Saltsman, I. Angew. Chem., Int. Ed. 1999,
38, 1427–1429.
(5) Gryko, D. T. Chem. Commun. 2000, 2243–2244.
(6) Paolesse, R.; Marini, A.; Nardis, S.; Froiio, A.; Mandoj, F.; Nurco,
D. J.; Prodi, L.; Montalti, M.; Smith, K. M. J. Porphyrins Phthalocyanines
2003, 7, 25–36.
(7) Koszarna, B.; Gryko, D. J. Org. Chem. 2006, 71, 3707–3717.
(8) Meier-Callahan, A. E.; Di Bilio, A. J.; Simkhovic, L.; Mahammed,
A.; Goldberg, I.; Gray, H. B.; Gross, Z. Inorg. Chem. 2001,
40, 6788–6793.
(9) Erben, C.; Will, S.; Kadish, K. M. In The Porphyrin Handbook;
Kadish, K. M.., Smith, K. M.., Guilard, R., Eds.; Academic Press: San
Diego, 2000; Vol. 2, pp 233ꢀ300.
(10) Ding, T.; Alemꢀan, E. A.; Modarelli, D. A.; Ziegler, C. J. J. Phys.
Chem. A 2005, 109, 7411–7417.
(11) Ventura, B.; Degli Esposti, A.; Koszarna, B.; Gryko, D. T.;
Flamigni, L. New J. Chem. 2005, 29, 1559–1566.
(12) Flamigni, L.; Ventura, B.; Tasior, M.; Gryko, D. T. Inorg. Chim.
Acta 2007, 360, 803–813.
(13) Flamigni, L.; Gryko, D. T. Chem. Soc. Rev. 2009, 38, 1635–1646.
(14) Paolesse, R. Synlett 2008, 15, 2215–2230.
(15) Remy, D. Tetrahedron Lett. 1983, 24, 1451–1454.
(16) Vicente, M. G. H.; Tomꢀe, A. C.; Walter, A.; Cavaleiro, J. A. S.
Tetrahedron Lett. 1997, 38, 3639–3642.
(17) Ito, S.; Murashima, T.; Ono, N. Chem. Commun. 1998, 1661–
1662.
(18) Finikova, O. S.; Chernov, S. Y.; Cheprakov, A. V.; Filatov, M. A.;
Vinogradov, S. A.; Beletskaya, I. P. Dokl. Chem. 2003, 391, 781–783.
(19) Barton, D. H. R.; Kervagoert, J.; Zard, S. Z. Tetrahedron 1990,
46, 7587–7598.
(20) Lindsey, J. S.; Hsu, H. C.; Schreiman, I. C. Tetrahedron Lett.
1986, 27, 4969–4970.
(21) Lindsey, J. S.; Hsu, H. C.; Schreiman, I. C.; Kearney, P. C.;
Marguerettaz, A. M. J. Org. Chem. 1987, 52, 827–836.
(22) Alemayehu, A. B.; Gonzalez, E.; Hansen, L. K.; Ghosh, A. Inorg.
Chem. 2009, 48, 7794–7799.
(23) Alemayehu, A. B.; Hansen, L. K.; Ghosh, A. Inorg. Chem. 2010,
49, 7608–7610.
(24) Finikova, O. S.; Cheprakov, A. V.; Beletskaya, I. P.; Carroll, P. J.;
Vinogradov, S. A. J. Org. Chem. 2004, 69, 522–535.
(25) Gros, C. P.; Barbe, J.-M-; Espinosa, E.; Guilard, R. Angew.
Chem., Int. Ed. 2006, 45, 5642–5645.
(26) Jiao, L.; Hao, E.; Fronczek, F. R.; Vicente, M. G. H.; Smith,
K. M. Chem. Commun. 2006, 3900–3902.
’ ASSOCIATED CONTENT
(27) Deshpande, R.; Jiang, L.; Schmidt, G.; Rakovan, J.; Wang, X.;
Wheeler, K.; Wang, H. Org. Lett. 2009, 11, 4251–4253.
(28) Stefanelli, M.; Mastroianni, M.; Nardis, S.; Licoccia, S.; Fronczek,
F. R.; Smith, K. M.; Zhu, W.; Ou, Z.; Kadish, K. M.; Paolesse, P. Inorg. Chem.
2007, 46, 10791–10799.
S
Supporting Information. General experimental meth-
b
1
ods, H NMR and mass spectra of compounds, CIF files for
2a, 3a, and 4. This material is available free of charge via the
Internet at http://pubs.acs.org.
3772
dx.doi.org/10.1021/jo200026u |J. Org. Chem. 2011, 76, 3765–3773

The Journal of Organic Chemistry
ARTICLE
(29) Wasbotten, I. H.; Wondimagegn, T.; Ghosh, A. J. Am. Chem.
Soc. 2002, 124, 8104–8116.
(30) Mandoj, F.; Nardis, S.; Pomarico, G.; Paolesse, R. J. Porphyrins
Phthalocyanines 2008, 12, 19–26.
(31) Mahammed, A.; Weaver, J. J.; Gray, H. B.; Abdelas, M.; Gross,
Z. Tetrahedron Lett. 2003, 44, 2077–2079.
3773
dx.doi.org/10.1021/jo200026u |J. Org. Chem. 2011, 76, 3765–3773