Investigation of complementary reactions of a dipyrromethane with a dipyrromethanemonocarbinol leading to a 5-isocorrole

Article abstract of DOI:10.1021/jo902452c

(Chemical Equation Presented) Two complementary dipyrromethane + dipyrromethanemonocarbinol routes to a meso-substituted 5-isocorrole were investigated. While both routes could afford the identical 5-isocorrole, self-condensation of the different dipyrromethanemonocarbinol precursors could potentially lead to a second porphyrinoid of different structure (a porphyrin or a porphodimethene). The two reaction routes were examined to compare the distribution of porphyrinoid products, probe the effect of key reaction parameters on the product distribution, and identify conditions for the efficient preparation of the 5-isocorrole so that its spectral properties and stability toward light and air could be assessed. For each route, a systematic survey of reaction parameters was performed via analytical-scale reactions monitored for the yields of the 5-isocorrole and self-condensation product by HPLC. The two reaction routes were found to afford very different product distributions in accordance with the anticipated nucleophilicity of the dipyrromethane and dipyrromethanemonocarbinol precursors. The most promising reaction condition (InCl₃, 0.32 mM) was performed on a preparative-scale providing the 5-isocorrole in an isolated yield of 31% (102 mg). Spectroscopic analysis was consistent with the 5-isocorrole structure. The stability of the 5-isocorrole in dilute solution upon exposure to light and air was assessed by UV-vis spectroscopy and HPLC and was found to be excellent.

Full text of DOI:10.1021/jo902452c

pubs.acs.org/joc
Investigation of Complementary Reactions of a Dipyrromethane with a
Dipyrromethanemonocarbinol Leading to a 5-Isocorrole
Danette L. Flint, Rachel L. Fowler, Timothy D. LeSaulnier, Andrew C. Long,
Anna Y. O’Brien, and G. Richard Geier, III*
Colgate University, Department of Chemistry, 13 Oak Drive, Hamilton, New York 13346
ggeier@colgate.edu
Received November 16, 2009
Two complementary dipyrromethane þ dipyrromethanemonocarbinol routes to a meso-substituted
5-isocorrole were investigated. While both routes could afford the identical 5-isocorrole, self-
condensation of the different dipyrromethanemonocarbinol precursors could potentially lead to a
second porphyrinoid of different structure (a porphyrin or a porphodimethene). The two reaction
routes were examined to compare the distribution of porphyrinoid products, probe the effect of key
reaction parameters on the product distribution, and identify conditions for the efficient preparation
of the 5-isocorrole so that its spectral properties and stability toward light and air could be assessed.
For each route, a systematic survey of reaction parameters was performed via analytical-scale
reactions monitored for the yields of the 5-isocorrole and self-condensation product by HPLC. The
two reaction routes were found to afford very different product distributions in accordance with the
anticipated nucleophilicity of the dipyrromethane and dipyrromethanemonocarbinol precursors.
The most promising reaction condition (InCl3, 0.32 mM) was performed on a preparative-scale
providing the 5-isocorrole in an isolated yield of 31% (102 mg). Spectroscopic analysis was consistent
with the 5-isocorrole structure. The stability of the 5-isocorrole in dilute solution upon exposure to
light and air was assessed by UV-vis spectroscopy and HPLC and was found to be excellent.
Introduction
the incorporation of an sp3-hybridized carbon atom at one of
the bridging meso positions as exemplified by phlorins and
5-isocorroles (pseudocorrole2).3 The sp3-hybridized carbon
atom disrupts macrocycle conjugation, altering the electronic
structure and rendering such porphyrinoids nonaromatic.
Additionally, the presence of the sp3-hybridized carbon
atom introduces a structural perturbation due to the tetra-
hedral geometry at that position.
Porphyrinic macrocycles with core structures altered re-
lative to porphyrin are of interest as their properties and
potential applications may complement prototypical por-
phyrins.1 One structural alteration of particular interest is
(1) (a) Sessler, J. L.; Weghorn, S. Expanded, Contracted, and Isomeric
Porphyrins; Elsevier: Oxford, 1997. (b) Jasat, A.; Dolphin, D. Chem. Rev. 1997,
97, 2267–2340. (c) Sessler, J. L.; Gebauer, A.; Vogel, E. In The Porphyrin
Handbook; Kadish, K. M., Smith, K., Guilard, R., Eds.; Academic Press:
San Diego, CA, 2000; Vol. 2, pp 1-54. (d) Sessler, J. L.; Gebauer, A.; Weghorn,
S. J. In The Porphyrin Handbook; Kadish, K. M., Smith, K., Guilard, R., Eds.;
Academic Press: San Diego, CA, 2000; Vol. 2, pp 55-124. (e) Misra, R.;
Chandrashekar, T. K. Acc. Chem. Res. 2008, 41, 265–279.
(2) The name “isocorrole” was first applied to the entirely different
(2.0.1.0)-type isomer of corrole. For that reason, 5-isocorrole has also been
called “pseudocorrole” in the literature: Orlewska, C.; Maes, W.; Toppet, S.;
Dehaen, W. Tetrahedron Lett. 2005, 46, 6067–6070.
Our initial interest in porphyrinoids bearing an sp3-
hybridized meso-position was directed toward phlorins. This
interest stemmed from the intriguing structural relationship
of phlorins to the better known porphyrin and corrole
species. Corroles have received recent attention, in part,
(3) Sessler, J. L.; Zimmerman, R. S.; Bucher, C.; Kral, V.; Andrioletti, B.
Pure Appl. Chem. 2001, 73, 1041–1057.
DOI: 10.1021/jo902452c
r
Published on Web 12/30/2009
J. Org. Chem. 2010, 75, 553–563 553
2009 American Chemical Society

Flint et al.
JOCFeatured Article
due to their ability to stabilize metal ions in high oxidation
states4 and the observation of catalysis complementary to
metalloporphyrins.5 Presumably, the distinctive coordina-
tion chemistry of corroles stems from their greater electron
richness and smaller core cavity relative to porphyrins. Like
corroles, phlorins bear three “pyrrole-like” N-H groups in
their core rendering them trianionic upon deprotonation and
metal insertion. Yet, like porphyrins, phlorins lack the core
contraction of the direct bipyrrole linkage. Interestingly,
unlike either corroles or porphyrins, phlorins bear an sp3
hybridized carbon atom at one of the meso positions. Taken
together, the phlorin structure is complementary to corroles
and porphyrins in interesting ways. Unfortunately, many
phlorins are unstable toward light and air.6 Thus, we began
our work by investigating the impact of sterically bulky7 or
electron-withdrawing8 peripheral meso substituents on
phlorin stability.
SCHEME 1. Dipyrromethane þ Dipyrromethanedicarbinol Route
to a Phlorin Bearing Electron-Withdrawing Substituents (TpFPPhl)
That Also Provided a 5-Isocorrole Byproduct in Low Yield
While attempting to prepare a phlorin bearing pentafluoro-
phenyl substituents at the three sp2-hybridized meso posi-
tions (TpFPPhl, Scheme 1),8 a second green pigment was
observed in a subset of the reaction conditions. The UV-vis
spectrum of the pigment was remarkably similar to that of
TpFPPhl, and the 1H NMR spectrum of the pigment showed
features consistent with a porphyrinoid, albeit one of lower
symmetry than TpFPPhl. Laser desorption mass spectral
(LD-MS) analysis found the molecular weight of the pigment
to be lower than TpFPPhl by approximately a pentafluoro-
phenyl substituent and a carbon atom. Based on this
preliminary spectral characterization, the second pigment
appeared to be 5-isocorrole 3 [5,5-dimethyl-10,15-bis-
(pentafluorophenyl)isocorrole]. This is not unreasonable as
the 5-isocorrole could have arisen from acidolytic cleavage of
one of the carbinol groups of the dipyrromethanedicarbinol
precursor9 followed by formation of the direct bipyrrole
linkage upon oxidation with DDQ.10
Although the 5-isocorrole was observed inconsistently and
in low quantity, we were intrigued by its structure. Like
phlorins, the structure of 5-isocorroles is complementary to
porphyrins and corroles, but in a different way. 5-Isocorroles
bear two internal N-H groups like porphyrins, while the size
of the core cavity is similar to corroles (due to the direct
bipyrrole linkage). Unlike either corroles or porphyrins,
5-isocorroles bear an sp3-hybridized carbon atom. Perhaps
the properties (e.g., coordination chemistry) of 5-isocorroles
will complement porphyrins, corroles, and phlorins in inter-
esting ways. Relative to these better studied porphyrinoids;
little is known about 5-isocorroles. Only a small number of
free base 5-isocorroles have been reported.11 All bear an
oxygen atom at the sp3-hybridized position (a hydroxyl or
methoxy group), and all were obtained upon oxidation of a
corrole (twice as a byproduct from attempts to demetallate
copper or silver corrole complexes11b,c). Additionally, three
(4) (a) Vogel, E.; Will, S.; Schulze-Tilling, A; Neumann, L.; Lex, J.; Bill,
E.; Trautwein, A. X.; Wieghardt, K. Angew. Chem., Int. Ed. Engl. 1994, 33,
731–735. (b) Erben, C.; Will, S.; Kadish, K. M. In The Porphyrin Handbook;
Kadish, K. M., Smith, K., Guilard, R., Eds.; Academic Press: San Diego, CA,
2000; Vol. 2, pp 233-300. (c) Gross, Z. J. Biol. Inorg. Chem. 2001, 6, 733–738.
(d) Ghosh, A.; Steene, E. J. Inorg. Biochem. 2002, 91, 423–436. (e) Gross, Z.;
Gray, H. B. Comments Inorg. Chem. 2006, 27, 61–72.
(5) (a) Gross, Z.; Gray, H. B. Adv. Synth. Catal. 2004, 346, 165–170.
(b) Aviv, I.; Gross, Z. Chem. Commun. 2007, 1987–1999.
(6) Jeandon, C.; Krattinger, B.; Ruppert, R.; Callot, H. J. Inorg. Chem.
2001, 40, 3149–3153.
(10) Formation of a direct bipyrrole linkage upon oxidation has been
observed in the synthesis of meso-substituted corroles from tetrapyrrane
species and in the synthesis of expanded porphyrins: (a) Gryko, D. T. Eur. J.
Org. Chem. 2002, 1735–1743. (b) Nardis, S.; Monti, D.; Paolesse, R. Mini-
Rev. Org. Chem. 2005, 2, 355–374. (c) Sessler, J. L.; Seidel, D. Angew. Chem.,
Int. Ed. 2003, 42, 5134–5175.
(7) LeSaulnier, T. D.; Graham, B. W.; Geier, G. R., III. Tetrahedron Lett.
2005, 46, 5633–5637.
(8) O’Brien, A. Y.; McGann, J. P.; Geier, G. R., III. J. Org. Chem. 2007,
72, 4084–4092.
(9) Braaten, K. C.; Gordon, D. G.; Aphibal, M. M.; Geier, G. R., III.
Tetrahedron 2008, 64, 9828–9836.
(11) (a) Nardis, S.; Pomarico, G.; Fronczek, F. R.; Vicente, M. G. H.;
Paolesse, R. Tetrahedron Lett. 2007, 48, 8643–8646. (b) Mandoj, F.; Nardis,
S.; Pomarico, G.; Paolesse, R. J. Porphyrins Phthalocyanines 2008, 12, 19–26.
(c) Stefanelli, M.; Shen, J.; Zhu, W.; Mastroianni, M.; Mandoj, F.; Nardis,
S.; Ou, Z.; Kadish, K. M.; Fronczek, F. R.; Smith, K. M.; Paolesse, R. Inorg.
Chem. 2009, 48, 6879–6887.
554 J. Org. Chem. Vol. 75, No. 3, 2010

Flint et al.
JOCFeatured Article
SCHEME 2. Complementary Dipyrromethane þ Dipyrromethanemonocarbinol Routes Leading to a 5-Isocorrole and/or a Self-
Condensation Product (a Porphyrin or a Porphodimethene)
nickel complexes of 5-isocorroles have been reported.2,12
One was obtained in low quantity upon nickel insertion to
a biphenylenyl bridged biscorrole.12 The other two were
obtained as low yield (∼3%) byproducts of nickel-templated
MacDonald [2 þ 2] condensation reactions of a dipyrromethane
with a bisformyldipyrromethane.2 Nickel-templated con-
densation of a dipyrromethane with a monoformyldipyrro-
methane improved the yield of one of the nickel 5-isocorroles
to 10%.2 To the best of our knowledge, a targeted synthesis
of a free-base 5-isocorrole has not been reported. Motivated
by the intriguing structure of 5-isocorroles and by the limited
prior work in this area, we undertook studies toward the
synthesis of 5-isocorrole 3. The results of this work are
reported herein.13
5-Isocorrole 3, with pentafluorophenyl substituents at
the two sp2-hybridized meso positions, was targeted for
three key reasons. (1) A small quantity of the 5-isocorrole
had already been isolated in the course of our previous
studies of reactions leading to TpFPPhl. This authentic
sample aided the detection and identification of the com-
pound in this study. (2) The stability toward light and air of
free base 5-isocorroles is not established. The incorporation
of electron-withdrawing groups would increase the
likelihood of obtaining a stable product.8,14 (3) The elec-
tron-withdrawing substituents were expected to influence
the reaction pathway leading to the 5-isocorrole. This
important issue is discussed in greater detail in following
sections.
Complementary two-step, one-flask reactions15 of a
dipyrromethane with a dipyrromethanemonocarbinol were
selected for the synthesis of 5-isocorrole 3 (Scheme 2). Both
routes were attractive as dipyrromethanemonocarbinols
have the free R-pyrrole position required for formation of
the direct bipyrrole linkage upon oxidation.10 Additionally,
the use of dipyrromethanecarbinol species in the synthesis of
porphyrinoids is well established.8,14,16-18 However, both
routes share a potential drawback. The dipyrromethane-
monocarbinol species can undergo self-condensation leading
to an alternative porphyrinoid byproduct. In fact, the self-
condensation of dipyrromethanemonocarbinols is an impor-
tant route for the preparation of trans-A2B2-porphyrins.16a
Interestingly, the two reaction routes shown in Scheme 2 give
rise to different self-condensation products;a porphyrin 4
(meso-tetrakispentafluorophenylporphyrin) from the self-
condensation of 2a-OH and
a porphodimethene 5
[bis(pentafluorophenyl)calix[4]phyrin] from the self-conden-
sation of 2b-OH. The success of either reaction route de-
pends on suppression of dipyrromethanemonocarbinol self-
condensation. Thus, we sought to explore the distribution of
the 5-isocorrole and self-condensation products for both
reaction routes, determine the effect of reaction conditions
on the product distribution, and identify conditions for the
preparation and purification of the 5-isocorrole. Addition-
ally, we sought to examine the spectral properties of the 5-
isocorrole and assess its stability in dilute solution toward
light and air.
To explore the complementary reaction routes, required
precursors were prepared and a systematic investigation of
(12) Jerome, F.; Barbe, J.-M.; Gros, C. O.; Guilard, R.; Fischer, J.; Weiss,
R. New J. Chem. 2001, 25, 93–101.
(13) A portion of this work was presented at the 238th ACS National
Meeting: Geier, G. R., III; Long, A. C.; Fowler, R. L.; Flint, D. L.; O’Brien,
A. Y. Abstr. Papers Am. Chem. Soc. 2009, 238, 719.
(14) Geier, G. R., III; Chick, J. F. B.; Callinan, J. B.; Reid, C. G.;
Auguscinski, W. P. J. Org. Chem. 2004, 69, 4159–4169.
(16) (a) Rao, P. D.; Littler, B. J.; Geier, G. R., III; Lindsey, J. S. Org.
Chem. 2000, 65, 1084–1092. (b) Rao, P. D.; Dhanalekshmi, S.; Littler, B. J.;
Lindsey, J. S. J. Org. Chem. 2000, 65, 7323–7344. (c) Geier, G. R., III;
Callinan, J. B.; Rao, D. P.; Lindsey, J. S. J. Porphyrins Phthalocyanines 2001,
5, 810–823.
(15) Two-step, one-flask syntheses of porphyrinoids involve oligomer
formation during an acid catalyzed condensation step, followed by an
oxidation step whereby reduced precursors are converted to the porphyr-
inoids. The oxidation step is carried out without isolation of the products
from the condensation step. Lindsey, J. S. In The Porphyrin Handbook;
Kadish, K. M., Smith, K., Guilard, R., Eds.; Academic Press: San Diego,
CA, 2000; pp 45-118.
(17) (a) Cho, W. S.; Kim, H. J.; Littler, B. J.; Miller, M. A.; Lee, C. H.;
Lindsey, J. S. J. Org. Chem. 1999, 64, 7890–7901. (b) Balasubramanian, T.;
Strachan, J. P.; Boyle, P. D.; Lindsey, J. S. J. Org. Chem. 2000, 65, 7919–
7929. (c) Guilard, R.; Gryko, D. T.; Canard, G.; Barbe, J.- M.; Koszarna, B.;
Brandes, S.; Tasior, M. Org. Lett. 2002, 4, 4491–4494. (d) Gryko, D. T.;
Tasior, M.; Koszarna, B. J. Porphyrins Phthalocyanines 2003, 7, 239–248.
(18) Geier, G. R., III; Grindrod, S. C. J. Org. Chem. 2004, 69, 6404–6412.
J. Org. Chem. Vol. 75, No. 3, 2010 555

Flint et al.
JOCFeatured Article
the effect of key reaction parameters (acid catalyst, acid
catalyst concentration, quantity of DDQ oxidant, condensation
time, and ratio of dipyrromethane to dipyrromethanemono-
carbinol) was carried out for each route. The yields of the
5-isocorrole and self-condensation product in the analytical-
scale reactions were assessed by HPLC. Preparative-scale
reactions were performed under the best conditions identified
from the analytical-scale studies, and the 5-isocorrole was
isolated and characterized. The stability of the 5-isocorrole
in dilute solution toward light and air was examined by
monitoring changes in UV-vis spectra and HPLC chroma-
tograms.
(1 equiv relative to 2a-OH) afforded the porphodimethene in
a yield of 21% (29 mg). The reaction conditions were derived
from methodology previously used to prepare trans-A2B2-
porphyrins from the self-condensation of dipyrromethane-
monocarbinols.16c NMR, mass spectrometric, and UV-vis
analyses of the porphodimethene were in agreement with
reported values for the compound prepared by alternative
dipyrromethane þ aldehyde/ketone routes leading to the
porphodimethene in yields of 6-7%.21
HPLC Method for Assessing the Production of the 5-
Isocorrole and Self-Condensation Products in Analytical-
Scale Reactions. To perform a large number of analytical-
scale reactions in parallel, methodology for the efficient
monitoring of the yield of the three porphyrinoids was
required. An HPLC method used in our recent investigation
of reaction routes leading to TpFPPhl was adapted to this
study.8 The method entails oxidation of an aliquot of the
condensation reaction mixture with DDQ, basification by
the addition of triethylamine, filtration of the oxidized
reaction mixture through a silica pad to remove insoluble
and strongly polar species, and analysis by HPLC. The
separation and quantitation of the 5-isocorrole and por-
phyrin obtained from the reaction of 1a þ 2a-OH was
achieved without event. However, under identical chromato-
graphic conditions, the 5-isocorrole and the porphodi-
methene obtained from the reaction of 1b þ 2b-OH were
found to coelute. Fortunately, we found that the two com-
pounds could be spectrally resolved so that it was not
necessary to modify the chromatographic conditions. The
use of different chromatography conditions to assess the
product distribution from the two reaction routes would
have been problematic as normal phase columns equilibrate
slowly. An expanded discussion of the spectral resolution of
the 5-isocorrole and the porphodimethene as well as a
summary of control experiments, reproducibility experi-
ments, detector response calibration, calculation of porphyr-
inoid yield, and representative chromatograms may be found
in the Supporting Information.
Survey of Acid Catalysis Conditions. To examine the effect
of acid catalyst and acid catalyst concentration on the yields
of 5-isocorrole 3 and self-condensation product 4 or 5, a
survey of acid catalysis conditions was carried out for both
reaction routes. Five acid catalysts [TFA, InCl3, Sc(OTf)3,
Yb(OTf)3, and Dy(OTf)3] at five concentrations (0.32, 1.0,
3.2, 10, and 32 mM, and in some cases 0.010 and 0.032 mM)22
were investigated. Each reaction was monitored for the yield
of the 5-isocorrole and self-condensation product at 0.25, 1,
and 4 h. TFA is a benchmark acid for two-step, one-flask
reactions leading to porphyrinoids,15 and the four mild
Lewis acids have been found to be well-suited for dipyrro-
methanecarbinol routes to porphyrin16c and other porphyr-
inoids.8,14,18
Results and Discussion
Preparation of Building Block Molecules. Dipyrro-
methanes 1a,b19 and monoacyl dipyrromethane 2a20
(precursor to 2a-OH) were prepared according to the litera-
ture. Monoacyl dipyrromethane 2b (precursor to 2b-OH)
was prepared in a yield of 41% (0.76 g) from dipyrromethane
1a in accordance with literature conditions for the preparation
of related monoacyl dipyrromethanes.16a Dipyrromethane-
monocarbinol species 2a-OH and 2b-OH were prepared im-
mediately prior to use by NaBH4 reduction of 2a or 2b and were
used without purification as reported in the literature.16a We
found that the reduction of 2b required the addition of 75 equiv
of NaBH4 rather than the more typical 25 equiv for the
reduction to come to completion in the usual reaction time of
∼30 min.
Preparation of Analytical Quantities of 5-Isocorrole 3 and
Porphodimethene 5. To develop methodology for monitoring
analytical-scale reactions leading to 5-isocorrole 3 and the
self-condensation product 4 or 5, a small amount of each
compound was required. Porphyrin 4 is commercially avail-
able. A very small amount of the 5-isocorrole had been
previously isolated in our earlier work on the preparation
of TpFPPhl. Additional 5-isocorrole was prepared from a
reaction of 1a þ 2a-OH (0.500 mmol each, 2.5 mM each) in
CH2Cl2 at room temperature for 5 h catalyzed by Dy(OTf)3
(3.2 mM) followed by oxidation with DDQ (3 equiv relative
to 1a or 2a-OH). These reaction conditions had previously
provided detectable levels of the 5-isocorrole in our earlier
studies of a different reaction route leading to TpFPPhl.
While these conditions were not expected to be optimal, they
did afford the 5-isocorrole in a yield of 6% (18 mg) after
purification by filtration through a silica pad and crystal-
lization from CH2Cl2/pentane. The porphodimethene 5 is a
known compound,21 although it had not been previously
prepared by the self-condensation of 2b-OH. As we had the
precursor monoacyl dipyrromethane 2b on hand, we elected
to attempt the preparation of the porphodimethene by self-
condensation of 2b-OH. Reaction of 2b-OH (0.40 mmol,
5.0 mM) in CH2Cl2 at room temperature for 30 min cata-
lyzed by InCl3 (3.2 mM) followed by oxidation with DDQ
Results from representative conditions affording good
yields of the 5-isocorrole or self-condensation product are
summarized in Table 1. Illustrative plots of the yield of the
5-isocorrole or self-condensation product as a function of
InCl3 concentration are provided in Figure 1 (see the Sup-
porting Information for the complete set of plots).
(19) (a) Littler, B. J.; Miller, M. A.; Hung, C.-H.; Wagner, R. W.; O’Shea,
D. F.; Boyle, P. D.; Lindsey, J. S. J. Org. Chem. 1999, 64, 1391–1396.
(b) Laha, J. K.; Dhanalekshmi, S.; Taniguchi, M.; Ambroise, A.; Lindsey,
J. S. Org. Process Res. Dev. 2003, 7, 799–812.
(20) Strachan, J. P.; O’Shea, D. F.; Balasubramanian, T.; Lindsey, J. S.
J. Org. Chem. 2000, 65, 3160–3172.
(21) (a) Bernatkova, M.; Andrioletti, B.; Kral, V.; Rose, E.; Vaissermann,
J. J. Org. Chem. 2004, 69, 8140–8143. (b) Dolensky, B.; Kroulik, J.; Sessler,
J. L.; Dvorakova, H.; Bour, P.; Bernatkova, M.; Bucher, C.; Lynch, V. J. Am.
Chem. Soc. 2004, 126, 13714–13722.
(22) The quantity of acid is reported in concentration units of molarity for
convenience of comparison of reaction conditions. However, InCl3, Sc(OTf)3,
Yb(OTf)3, and Dy(OTf)3 are poorly soluble in some of the solvents, and the
reaction mixtures were heterogeneous as reported previously.16c
556 J. Org. Chem. Vol. 75, No. 3, 2010

Flint et al.
JOCFeatured Article
TABLE 1. Comparison of Macrocycle Yields from the Reaction of 1a þ 2a-OH and 1b þ 2b-OH from Representative Conditions Examined in the Survey
of Acid Catalysis Conditionsa
1a þ 2a-OH
% yield of 5-isocorroleb
5.2
37
1b þ 2b-OH
entry
acid
TFA
InCl3
[acid], mM
% yield of porphyrinb
% yield of 5-isocorroleb
% yield of porphodimetheneb
1
2
3
4
5
6
7
0.32
0.32
1.0
0.32
0.32
10
2.1
2.4
2.7
6.1
4.7
5.9
6.1
6.8
2.1
1.7
6.1
4.7
7.5
5.3
31
50
49
51
50
53
49
InCl3
31
27
26
27
21
Sc(OTf)3
Yb(OTf)3
Yb(OTf)3
Dy(OTf)3
1.0
aThe reactions were performed in CH2Cl2 with the indicated reactants (2.5 mM each) on a 5-10 mL scale at room temperature. The reactions were
monitored at 0.25, 1, and 4 h. bThe highest yield (HPLC) at any of the three time points is reported.
FIGURE 1. Yield of the 5-isocorrole (left panels) and self-condensation product (porphyrin or porphodimethene, right panels) as a function of
InCl3 concentration for the reaction of (A) 1a þ 2a-OH and (B) 1b þ 2b-OH. The reactions were performed with the indicated reactants (2.5
mM each) in CH2Cl2 at room temperature. The reactions were monitored by HPLC.
The most striking result is the observation of very different
ratios of the 5-isocorrole to self-condensation product from
the complementary reaction routes. In the reaction of 1a þ
2a-OH, generally more 5-isocorrole than porphyrin was
obtained, whereas the reaction of 1b þ 2b-OH generally
led to more porphodimethene than 5-isocorrole. It is clear
that the choice of reaction route had a substantial impact on
product distribution, with the reaction of 1a þ 2a-OH being
the superior route for the preparation of the 5-isocorrole.
For this route, a number of conditions provided yields of the
5-isocorrole of 20-40% while limiting porphyrin formation
to yields of 2-6%. Catalysis with InCl3 was particularly
promising in terms of the yield of the 5-isocorrole and sup-
pression of porphyrin byproduct (Table 1, entries 2 and 3).
The suppression of porphyrin formation is important as
the chromatographic properties of the two compounds are
very similar. Thus, purification of the 5-isocorrole from
preparative-scale reactions is more straightforward if less
porphyrin is present.
In addition to these key findings, a number of additional
observations were made. (1) The precise ratio of products
was found to vary somewhat as a function of acid catalyst.
For example, in the reaction of 1a þ 2a-OH, catalysis with
InCl3 afforded good yields of the 5-isocorrole and very little
porphyrin (Table 1, entries 2 and 3), whereas catalysis with
Sc(OTf)3, Yb(OTf)3, and Dy(OTf)3 afforded slightly lower
yields of the 5-isocorrole but about twice the level of
porphyrin as was observed from catalysis with InCl3
(entries 4-7). In the reaction 1b þ 2b-OH, similar yields of
the porphodimethene were observed from most catalysts,
but the level of the 5-isocorrole was found to vary. (2) For
each reaction route and acid catalyst, the maximum yield of
J. Org. Chem. Vol. 75, No. 3, 2010 557

Flint et al.
JOCFeatured Article
TABLE 2. Summary of Results from Reaction Time Course Experimentsa
% yield of
5-isocorrolec
% yield of self-condensation
productd
entry
reactants
acid
InCl3
InCl3
Sc(OTf)3
Yb(OTf)3
Dy(OTf)3
TFA
[acid], mM
time, hb
1
2
3
4
5
6
7
8
9
10
1a þ 2a-OH
1a þ 2a-OH
1a þ 2a-OH
1a þ 2a-OH
1a þ 2a-OH
1b þ 2b-OH
1b þ 2b-OH
1b þ 2b-OH
1b þ 2b-OH
1b þ 2b-OH
0.32
1.0
0.32
10
2
35
32
28
26
1.4
1.9
4.7
4.5
5.1
30
47
45
47
48
0.5
0.5
1
1.0
8
25
0.32
0.32
0.32
0.32
1.0
0.25
0.25
0.5
4
7.8
2.7
6.8
6.5
7.4
InCl3
Sc(OTf)3
Yb(OTf)3
Dy(OTf)3
8
aThe reactions were performed in CH2Cl2 with the indicated reactants (2.5 mM each) on a 20 mL scale at room temperature. The reactions were
monitored from 1 min to 24 h. bThe reaction time that first provided the highest yield of the 5-isocorrole. cThe highest yield of the 5-isocorrole (HPLC) is
reported. dThe yield of the porphyrin (1a þ 2a-OH) or porphodimethene (1b þ 2b-OH) (HPLC) at the time that the highest yield of the 5-isocorrole was
first obtained is reported. The yields reported here are generally within ∼2% of the highest yield obtained at any time.
the 5-isocorrole and self-condensation product was generally
similar across a wide range of acid concentration. For
example, the reaction of 1a þ 2a-OH mediated by InCl3
afforded a similar yield of the 5-isocorrole across a 10-fold
range of acid concentration (Figure 1A). The reaction of
1b þ 2b-OH mediated by InCl3 afforded a similar level of
porphodimethene across a 30-fold range of acid concentra-
tion (Figure 1B). The maximum yield of the porphyrinoids
may be observed more quickly at higher acid concentration
(Figure 1A), nevertheless, a near maximum yield could
be obtained across a wide range of acid concentration as
long as the condensation reaction time was adjusted. (3)
Extremes of acid catalyst concentration were found to
impact the product distribution. This was most evident in
the reaction of 1a þ 2a-OH under TFA catalysis. At low
TFA concentration, the 5-isocorrole was the major product
(albeit in low yield). At high TFA concentration, the por-
phyrin was the major product (see the Supporting
Information). So while the choice of reaction route appears
to be the primary determinant of the distribution of pro-
ducts, the precise choice of acid catalyst and catalyst con-
centration also has an influence.
Investigation of Oxidation Conditions. In previous work,8,14,18
we found that some two-step, one-flask syntheses of porphyr-
inoids require exacting conditions for the oxidation step, and
occasionally the best oxidation conditions deviate from the
stoichiometric quantity of oxidant. Thus, we investigated oxida-
tion conditions for both reaction pathways. We first explored
oxidation conditions prior to performing the survey of acid
catalyst conditions. Reactions of 1a þ 2a-OH or 1b þ 2b-OH
[Dy(OTf)3, 3.2 mM] were performed, and aliquots (1.2 mL)
of the reaction mixture were transferred to quantities of DDQ
(0.5, 1.0, 2.0, 4.0, 8.0, 16, or 32 mg). The stoichiometric quantity
of DDQ relative to 5-isocorrole for a 1.2 mL aliquot of crude
reaction mixture is 2.04 mg (3 equiv per 1a or 2a-OH).23 The
oxidized reaction mixtures were assessed by HPLC and TLC. To
further refine the oxidation conditions, the experiment was
repeated over a narrower range of DDQ quantities (0.5, 1.0,
1.5, 2.0, 2.5, 3.0, 3.5, 4.0 mg). Once the reaction of 1a þ 2a-OH
(InCl3, 0.32 mM) was found to provide the highest yield of the
5-isocorrole and the lowest level of the porphyrin byproduct, a
reaction of 1a þ 2a-OH (InCl3, 0.32 mM) was performed and
aliquots (1.2 mL) were transferred to quantities of DDQ (0.5,
1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0 mg).
For the reaction of 1a þ 2a-OH under either acid catalyst,
the highest yield of the 5-isocorrole was obtained in the range
of 1.5-2.5 mg of DDQ. At both higher and lower quantities
of DDQ the yield of the 5-isocorrole decreased, and no
5-isocorrole was detected at DDQ quantities of g8 mg.
The yield of porphyrin self-condensation product was less
sensitive to the quantity of DDQ with similar yields obtained
from 1.5 to 32 mg DDQ. For the reaction of 1b þ 2b-OH, the
highest yield of the 5-isocorrole and the porphodimethene
were both in the range of 1.0-4.0 mg DDQ indicating that
this reaction route is a little less sensitive to the amount of
DDQ than the complementary route. For both reaction
routes, the stoichiometric amount of DDQ was found to
be appropriate. This amount was used for analytical and
preparative-scale reactions.
Reaction Time Course Experiments. Representative reac-
tion conditions identified from the survey of catalysis con-
ditions were performed for the reaction of 1a þ 2a-OH and
1b þ 2b-OH with monitoring by HPLC and TLC at time
points ranging from 1 min to 24 h. A summary of the reaction
conditions investigated and the highest yield of the 5-iso-
corrole and self-condensation product (porphyrin or
porphodimethene) are provided in Table 2. Representative
plots of the yield of the 5-isocorrole and self-condensation
product as a function of condensation time are provided in
Figure 2 (see the Supporting Information for the complete
set of plots).
As with the survey of catalysis conditions, the two reaction
routes provided very different product distributions. The
reaction of 1a þ 2a-OH afforded the 5-isocorrole as the
major product, and the reaction of 1b þ 2b-OH primarily led
to the porphodimethene self-condensation product. The
reaction of 1a þ 2a-OH (InCl3, 0.32 mM) again provided
the best result in terms of the yield of the 5-isocorrole and
suppression of porphyrin byproduct (Table 2, entry 1). The
reaction trajectory provided by all conditions explored
showed a turnover and decline in the yield of the 5-isocorrole
at extended reaction times (Figure 2A). Such a turnover in
yield in two-step, one-flask reactions leading to porphyri-
noids is not uncommon,24 but reactions showing such a
(23) The stoichiometric amount of DDQ for the porphyrin (self-conden-
sation of 1a-OH) from a 1.2 mL aliquot of the reaction mixture is 1.02 mg
(1.5 equiv relative to 1a-OH). The stoichiometric amount of DDQ for the
porphodimethene (self-condensation of 1b-OH) from a 1.2 mL aliquot of the
reaction mixture is 0.68 mg (1.0 equiv relative to 1b-OH).
(24) Geier, G. R., III; Lindsey, J. S. J. Chem. Soc., Perkin Trans. 2 2001,
677–686.
558 J. Org. Chem. Vol. 75, No. 3, 2010

Flint et al.
JOCFeatured Article
FIGURE 2. Yields of the 5-isocorrole and self-condensation pro-
duct (porphyrin or porphodimethene) as a function of condensation
time. The reactions were performed with the indicated reactants
(2.5 mM each) in CH2Cl2 at room temperature under catalysis by
InCl3 (0.32 mM). The reactions were monitored by HPLC. Note the
logarithmic scale for time.
turnover should be monitored so that the DDQ oxidant
is added at an appropriate time. Fortunately, the yield
trajectory provided by the best reaction condition (InCl3,
0.32 mM) afforded a similarly good yield of the 5-isocorrole
over a condensation time of 2-4 h. Thus, the reaction time
is not overly exacting for this reaction condition. This
contrasts with the majority of the other conditions surveyed
(Table 2, entries 2-4) which displayed a turnover in the
5-isocorrole yield at an earlier point in the reaction leading
to a shorter window of time for optimal addition of the
oxidant.
Other key observations are as follows. (1) The appearance
of the low level of porphyrin byproduct closely tracked the
production of the 5-isocorrole in reactions of 1a þ 2a-OH.
However, no turnover in porphyrin yield was observed. (2)
The appearance of the 5-isocorrole and porphodimethene
byproduct also closely tracked each other in reactions of
1b þ 2b-OH (Figure 2B). Again, at longer reaction times the
yield of the 5-isocorrole tended to turnover and decline, while
the yield of the porphodimethene remained steady. Similar
trajectories were observed across all conditions examined.
(3) The yield of the 5-isocorrole and self-condensation
products observed in these experiments were similar to those
obtained at equivalent time points in the survey of acid
catalysis conditions.
FIGURE 3. Correlation of the level of self-condensation of (A) 2a-
OH or (B) 2b-OH with differences in the nucleophilicity of the
available R-pyrrole positions of the two dipyrromethanes and two
dipyrromethanemonocarbinols.
to the 5-isocorrole, whereas the reaction of 1b þ 2b-OH
predominantly led to self-condensation of 2b-OH. This
finding correlates with anticipated differences in the nucleophilicity
of the available R-pyrrole positions of the two dipyrro-
methanes and two dipyrromethanemonocarbinols (Figure 3).
In both reaction routes, there is a competition between the
free R-pyrrole positions for reaction with the electrophilic
carbinol group of 2-OH. If the R-pyrrole positions of the
dipyrromethane are more nucleophilic than the R-pyrrole
position of the dipyrromethanemonocarbinol, then the
5-isocorrole product may be the major porphyrinoid pro-
duct. If the free R-pyrrole position of the dipyrromethane-
monocarbinol is instead more nucleophilic, then the product
of self-condensation may predominate. In the reaction of
1a þ 2a-OH, the pyrrole rings of 1a are appended to a
bridging carbon atom bearing two electron-releasing methyl
groups, whereas the pyrrole rings of 2a-OH are appended to
a bridging carbon atom bearing an electron-withdrawing
substituent. In the reaction of 1b þ 2b-OH, the substituents
are reversed and 2b-OH bears pyrrole rings adjacent to a
On the Two Reaction Routes. The survey of catalysis
conditions and time-course experiments both revealed con-
trasting product distributions from the complementary
reaction routes. The reaction of 1a þ 2a-OH generally led
J. Org. Chem. Vol. 75, No. 3, 2010 559

Flint et al.
JOCFeatured Article
bridging carbon atom with the two methyl substituents. The
effect of the substituents at the bridging carbon atom on the
various R-pyrrole positions is apparent in the 1H NMR
spectra of the dipyrromethanes and dipyrromethanemono-
carbinols. In CDCl3, the R-pyrrole protons of 1a and 2b-OH
are both found at ∼6.58 ppm, whereas the analogous pro-
tons of 1b and 2a-OH are both more deshielded, appearing at
∼6.71 ppm.
did increase slightly with increasing DDQ reaching a max-
imum level of ∼7% with 8.0 mg of DDQ. No porphyrin was
detected under any condition. Overall, this preliminary
survey of reactant ratio found the yield of both products to
be suppressed. Whether the same result will be obtained with
other substituent combinations remains to be determined.
Fortunately, the inability to further suppress porphyrin
formation without lowering the yield of the 5-isocorrole
was not a major problem in this study as we were able to
purify the 5-isocorrole from the low level of porphyrin
byproduct by crystallization (vide infra).
The apparently profound effect of the substituents at
the bridging carbon atom of the dipyrromethanes and di-
pyrromethanemonocarbinols is important to consider in
synthetic planning. It appears that the reaction of a dipyrro-
methane with a dipyrromethanemonocarbinol will afford
the best ratio of the 5-isocorrole to self-condensation
product when substituents are present that cause the
nucleophilicity of the precursors to differ. For example,
replacement of the electron-withdrawing pentafluorophenyl
substituent with a phenyl group at the bridging carbon
atom of 1b and 2a-OH should result in lower selectivity
for both reaction routes. We are currently investigating
this possibility as well as exploring whether other reaction
parameters can be adjusted so as to obtain a satisfactory
product ratio.
Investigation of the Ratio of the Dipyrromethane to the
Dipyrromethanemonocarbinol. The ratio of dipyrromethane
to dipyrromethanemonocarbinol could be an important
parameter in suppressing self-condensation as presumably
an excess of dipyrromethane would lower the level of self-
condensation. With this in mind, we sought to determine if
the very low level of porphyrin byproduct obtained from
reaction of 1a þ 2a-OH (InCl3, 0.32 mM) could be further
lowered by altering the ratio of the starting materials.
Complete suppression of porphyrin formation would sim-
plify purification of the 5-isocorrole as the chromatographic
properties of the 5-isocorrole and the porphyrin are very
similar. Of course, any improvement in product distribution
would need to be weighed against the use of excess dipyrro-
methane, a synthetic precursor.
Preparative-Scale Synthesis of 5-Isocorrole 3. Preparative-
scale reactions were performed to confirm findings from
analytical-scale experiments, to determine whether the
5-isocorrole could be conveniently purified from the low
level of porphyrin byproduct, and to obtain sufficient quan-
tity of the 5-isocorrole for characterization and stability
studies. Preparative-scale reactions were carried out under
the best conditions identified from the analytical-scale
experiments: reaction of 1a þ 2a-OH (0.500 mmol, 2.5 mM
each), InCl3 (0.32 mM), in CH2Cl2 at room temperature.
Reactions were monitored by HPLC, and DDQ (3 equiv
relative to 1a or 2a-OH) and triethylamine (5 equiv relative
to InCl3) were added when the yield of the 5-isocorrole
reached a maximum (generally at a reaction time of 2 h).
The crude, oxidized reaction mixture was assessed by HPLC.
The yields of the 5-isocorrole and porphyrin determined
from an aliquot prior to oxidation were 34% and 1.4%,
respectively, in good agreement with results from earlier
analytical-scale reactions (Table 2, entry 1). After oxida-
tion of the entire reaction mixture, the HPLC yields of the
5-isocorrole, and porphyrin were found to be 37% and
1.8%, respectively. The crude reaction mixture was passed
through a silica pad to remove polar impurities. Silica gel
chromatography with CH2Cl2/hexanes (1:5) afforded the
5-isocorrole (132 mg) still contaminated with a very low
level of porphyrin but largely devoid of other impurities.
The percentage by mass of the 5-isocorrole was found
1
We began by exploring nine reaction conditions ranging
from equimolar to 10-fold excess of 1a relative to 2a-OH
(2.5 mM) under conditions of InCl3 catalysis (0.32 mM) in
CHCl2 at room temperature. The reactions were monitored
at 1, 2, and 4 h. The yield of both the porphyrin and the
5-isocorrole were found to decline with increasing excess of
dipyrromethane. At a roughly 5-fold excess of 1a, neither
product was detected. It is possible that the excess dipyrro-
methane interfered with either the acid catalyst or oxidant
necessitating further refinement of reaction conditions. We
found it less likely that the oxidation conditions posed a
problem as excess 1a should not consume oxidant (the
presence of two methyl groups at the bridging carbon atom
renders the dipyrromethane inert toward oxidation). Never-
theless, we explored both possibilities. A survey of InCl3
concentration (0.32, 1.0, 3.2, 10, and 32 mM) was carried out
using a fixed ratio of 1a to 2a-OH (5:1). The reactions were
monitored at 0.25, 1, and 4 h. While porphyrin was not
detected under any condition, the yield of the 5-isocorrole
was quite low ranging from 2 to 6%. InCl3 (3.2 mM)
provided the modestly best result. A survey of DDQ quantity
(0.5, 1.0, 2.0, 3.0, 4.0, 8.0, 16, 32 mg) was carried out using
aliquots (1.2 mL) from a reaction of 1a to 2a-OH (5:1)
mediated by InCl3 (3.2 mM). The yield of the 5-isocorrole
by HPLC and H NMR analyses to be 97% and 98%,
respectively. The good agreement between HPLC and
NMR analyses is supportive of accurate HPLC detector
response calibration. The approximate yield of the 5-iso-
corrole after chromatography (corrected for the low level
of porphyrin) is in agreement with the yield estimated by
HPLC during reaction monitoring and after oxidation.
Isolation of the 5-isocorrole from the porphyrin was
achieved by crystallization from CH2Cl2/hexanes (31%,
102 mg). Absence of the porphyrin was confirmed by
1
HPLC and H NMR analyses (Supporting Information).
The isolated yield and scale compare favorably to syn-
theses of other porphyrinoids via dipyrromethanecarbinol
routes.8,14,16-18
Characterization of 5-Isocorrole 3. The UV-vis spectrum
of the 5-isocorrole is quite similar to that of TpFPPhl,
consistent with an interruption of macrocycle conjugation
due to an sp3-hybridized meso-carbon atom. LD-MS and
high-resolution FAB-MS data were satisfactory. 13C NMR
analysis (CDCl3) revealed 18 peaks in the region of 105-185 ppm
consistent with the 18 sp2-hybridized carbon atoms of the
5-isocorrole ring (Supporting Information). Peaks at 25 and
39 ppm were also observed from the equivalent methyl
substituents and the sp3-hybridized meso-carbon atom,
560 J. Org. Chem. Vol. 75, No. 3, 2010

Flint et al.
JOCFeatured Article
respectively. In addition, weak, split signals from carbon
atoms of the pentafluorophenyl substituents were observed.
1H NMR analysis (CDCl3) revealed two separate signals at
13.7 and 13.9 ppm for the two core N-H groups (Supporting
Information). The assignment of the N-H protons was
supported by the absence of correlation to carbon atoms in
the HMQC spectrum. The downfield location of the N-H
signals is consistent with the absence of overall macrocycle
aromaticity. The observation of two distinct signals for the
N-H protons is in accord with reported NMR data for
5-isocorroles bearing hydroxy11b,c or methoxy11a substitu-
ents at the sp3-hybridized carbon atom. The presence of two
distinct peaks for the N-H protons is suggestive of slow
tautomerism. The tautomeric form of the 5-isocorrole drawn
in Scheme 1 was supported by COSY and NOESY analysis
(Supporting Information), and it is the same tautomer
recently reported for a 5-isocorrole bearing a hydroxy group
at the sp3-hybridized position based on X-ray crystallo-
graphy.11c The β-pyrrole region of the NMR spectrum
revealed four sets of doublets (7.15, 7.04, 7.02, and 6.82 ppm)
consistent with protons on pyrrole rings lacking a N-H
proton, two sets of doublets of doublets (6.97 and 6.49 ppm)
consistent with protons on a pyrrole ring bearing a N-H
proton, and overlapping signals at 6.36 ppm. In addition, a
singlet was observed at 1.8 ppm with integration in accord
withtwo equivalent methylgroups. A further discussion ofthe
FIGURE 4. (A) Overlaid UV-vis spectra of a solution of the
5-isocorrole in CH2Cl2 recorded after exposure to light and air over
a period of 14 days. (B) Plot of absorbance as a function of time
of exposure to light for a solution of the 5-isocorrole in CH2Cl2
(717 nm).
1
assignment of the H NMR spectrum may be found in the
Supporting Information.
Stability of 5-Isocorrole 3. The stability toward light and
air of the 5-isocorrole in dilute solution was investigated.
Similar to our previous studies of the stability of meso-
substituted corroles,14 octaphyrins,18 and phlorins,7,8 solu-
tions of the 5-isocorrole were prepared in solvents of varying
polarity (hexanes, toluene, CH2Cl2, THF, ethyl acetate,
acetone, acetonitrile, and methanol), and changes in UV-vis
spectra were followed as a function of time of exposure to
standard room lighting (i.e., light from conventional overhead
fluorescent lighting). The solutions were monitored for a
period of two weeks. A representative plot of overlaid
UV-vis spectra (CH2Cl2) is shown in Figure 4A (see the
Supporting Information for the complete set of overlaid
spectra). A representative plot of the 5-isocorrole absor-
bance (717 nm, CH2Cl2) as a function of time is shown in
Figure 4B (see the Supporting Information for the complete
set of plots).
As shown by the illustrative data, dilute solutions of the
5-isocorrole are very stable over 14 days of exposure to light
and air. The solution in CH2Cl2 retained better than 99%
of its initial absorbance at 717 nm. The other solutions
generally displayed similar stability. To confirm the results
of UV-vis monitoring, fresh solutions were prepared in
CH2Cl2 or methanol, exposed to light, and the disappear-
ance of the 5-isocorrole was monitored by HPLC for a period
of 14 days. Consistent with the UV-vis monitoring, little
change was observed in the peak area of the 5-isocorrole for
both solutions. The 5-isocorrole appears to be quite stable
toward light and air, even more so than the closely related
phlorin, TpFPPhl.8 It will be interesting to determine if
such stability is a general feature of 5-isocorroles or whether
the electron-withdrawing pentafluorophenyl substituents
substantially enhance the stability as is the case with
TpFPPhl.
Conclusions
Complementary reactions of a dipyrromethane with a
dipyrromethanemonocarbinol were investigated to examine
the distribution of a 5-isocorrole and self-condensation
product (a porphyrin or a porphodimethene), probe the
effect of key reaction parameters on the product distribution,
and identify appropriate reaction conditions for the prepara-
tion of the 5-isocorrole so that its spectral properties and
stability could be assessed. The two reaction routes produced
very different results across a wide range of conditions with
the reaction of 1a þ 2a-OH primarily providing the 5-isocorrole
and the reaction of 1b þ 2b-OH primarily affording the self-
condensation product (a porphodimethene). Reaction condi-
tions (acid catalyst, acid concentration, reaction time, and ratio
of dipyrromethane to dipyrromethanemonocarbinol) had some
effect on the product distribution; however, the primary deter-
minant was the choice of reaction route. The substantially
different outcomes obtained from the complementary routes
show that the choice of starting materials can have a profound
impact on the outcome of the reaction. The product distributions
observed from the two reaction routes are consistent with
anticipated differences in the nucleophilicity of the dipyrro-
methanes and dipyrromethanemonocarbinol species. The best
yield of the 5-isocorrole with the lowest level of porphyrin
byproduct was obtained from the reaction of 1a þ 2a-OH
(InCl3, 0.32 mM). Preparative-scale reactions were in accord
with the results of analytical-scale reactions monitored by
HPLC. On the preparative-scale, the low level of porphyrin
byproduct could be removed by crystallization with minimal
impact on the isolated yield of the 5-isocorrole. Spectral char-
acterization was consistent with the proposed 5-isocorrole
J. Org. Chem. Vol. 75, No. 3, 2010 561

Flint et al.
JOCFeatured Article
structure. Dilute solutions of the 5-isocorrole were found to be
quite stable toward light and air. The fairly efficient preparation
of the 5-isocorrole along with the observation of interesting
spectral properties and good stability provides encouragement
for further studies of this compound and for investigations of
5-isocorroles bearing a broader array of substituents.
solvent was driven off the column with a hand-held pipet tool.
The eluant was transferred to an autosampler vial and capped.
HPLC analysis was performed with an injection volume of 1 μL,
a normal-phase silica column (Alltech, Altima, 4.6 mm ꢀ
250 mm), using an isocratic solvent mixture of 92% hexanes
and 8% acetone. The hexanes solvent was water saturated by
storing over water prior to use. The solvent flow rate was
controlled as follows: T = 0-8 min, 1 mL/min; T = 8-9
min, linear increase to 2 mL/min; T = 9-14 min, 2 mL/min;
T = 14-15 min, linear decrease to 1 mL/min. The 5-isocorrole,
porphyrin, and porphodimethene eluted at 6.9, 8.2, and 6.9 min,
respectively. Detection was performed at 335 nm (5-isocorrole),
417 nm (porphyrin), and 423 nm (5-isocorrole and porpho-
dimethene). The yields of the 5-isocorrole and porphyrin or
porphodimethene were determined from the peak area by
calibration of the detector response to each compound. The
peak area for the porphodimethene required correction to
account for the contribution of the coeluting 5-isocorrole.
Representative chromatograms and further details on the analysis
method development, control and reproducibility experiments,
HPLC calibration, and spectral resolution of the 5-isocorrole
and porphodimethene may be found in the Supporting Infor-
mation.
Survey of Acid Catalysis Conditions. Following a literature
procedure,16a immediately prior to the condensation reactions,
the monoacyl dipyrromethane [2a (192 mg, 0.375 mmol) or 2b
(138 mg, 0.375 mmol)] was reduced to the corresponding
dipyrromethanemonocarbinol (2a-OH or 2b-OH) with NaBH4
[2a (0.355 g, 9.38 mmol); 2b (1.07 g, 28.1 mmol) in THF/
methanol (30 mL, 3:1)]. The reduction reactions were monitored
by TLC [2a, alumina, CH2Cl2; 2b alumina, EtOAc/CH2Cl2
(1:5)]. After drying under vacuum for 30 min, the monocarbinol
was dissolved in CH2Cl2, dipyrromethane 1a (65.3 mg, 0.375 mmol)
or 1b (117 mg, 0.375 mmol) was added to the flask, and the volume
was brought to 150 mL by adding CH2Cl2. Reactions were per-
formed at room temperature in tightly capped 20-mL vials that were
stirred with a micro stir bar. Solid acids were weighed into all
reaction vials prior to the beginning of the reaction sequence for the
day, and each reaction was initiated by the addition of 5-10 mL of
the reactant solution via volumetric pipet. Reactions involving TFA
were initiated by the addition of TFA to reaction vials already
containing 5 mL of the reactant solution. The reactions were
monitored by HPLC at 0.25, 1, and 4 h as described above. TLC
was performed on the crude, oxidized mixture [silica, CH2Cl2/
hexanes (1:1)].
Reaction Time-Course Experiments. Reaction monitoring as
a function of time was performed as described above for the
survey of acid catalysis conditions with the exception of using a
20 mL reaction volume. The reactions were monitored by HPLC
as described above at 1 min, 4 min, 8 min, 15 min, 30 min, 1 h,
2 h, 4 h, 8 h, and 24 h. Selected conditions were also monitored at
6 h. TLC was performed on the crude, oxidized mixture [silica,
CH2Cl2/hexanes (1:1)].
5,5-Dimethyl-10,15-bis(pentafluorophenyl)isocorrole (3). The
reduction of 2a (253 mg, 0.500 mmol) with NaBH4 (473 mg,
12.5 mmol) in THF/methanol (40 mL, 3:1) afforded the corre-
sponding monocarbinol 2a-OH which was used without pur-
ification.16a The monocarbinol was dried under vacuum for
30 min and then immediately subjected to condensation with
1a (87.1 mg, 0.500 mmol) in the presence of InCl3 (14.2 mg,
0.0640 mmol) in CH2Cl2 (200 mL) at room temperature. At 1.5 h,
the reaction was monitored by HPLC as described above, and
the yield of the 5-isocorrole was found to be satisfactory. At a
total reaction time of 2 h, the reaction mixture was oxidized by
the addition of DDQ (330 mg, 1.50 mmol) at room temperature.
After 5 min, triethylamine (0.045 mL, 0.32 mmol) was added,
and the mixture was stirred at room temperature for a further 1 h.
An aliquot (1.2 mL) of the oxidized reaction mixture was
Experimental Section
1-Pentafluorobenzoyl-5,5-dimethyldipyrromethane (2b). Following
a general monoacylation procedure from the literature,16a ethylmag-
nesium bromide (12.5 mL, 12.5 mmol, 1 M in THF) was added
dropwise over 15 min to a solution of 5,5-dimethyldipyrromethane 1a
(0.871 g, 5.00 mmol) in dry THF (5 mL) under argon. The reaction
mixture was stirred for 10 min at room temperature and then cooled
to -78 °C. A suspension of S-2-pyridyl pentafluorobenzothioate16a
(1.53 g, 5.00 mmol) in THF (10 mL) was added over 1 min. The
solution was kept at -78 °C for 10 min, the cooling bath was
removed, and the reaction was allowed to stir for an additional
40 min. The reaction was quenched by the addition of saturated
aqueous NH4Cl (50 mL). The mixture was poured into CH2Cl2
(50 mL), washed with water (50 mL), and dried over Na2SO4, andthe
solvent was removed providing a brown viscous liquid. Flash column
chromatography [silica, CH2Cl2/hexanes (2:1) to CH2Cl2/hexanes
(4:1)] afforded a pale yellow viscous liquid. Crystallization (EtOH/
H2O) provided light tan crystals (760 mg, 41%): mp 111-113 °C; 1H
NMR (CDCl3) δ 9.34 (s, 1H), 1.70 (s, 6H), 6.13 (m, 1H), 6.16
(m, 1H), 6.19 (m, 1H), 6.62 (m, 1H), 6.71 (m, 1H), 8.13 (s, 1H); 13
C
NMR (CDCl3) δ 28.7, 35.9, 105.0, 108.3, 108.6, 114.0 (t, J = 21 Hz),
118.0, 122.5, 130.5, 136.2, 137.6 (d, J = 243 Hz), 142.1 (d, J = 257
Hz), 143.9 (d, J = 251 Hz), 151.1, 171.5; EI-MS obsd 368, calcd 368;
IR νmax (thin film)/cm-1 1617. Anal. Calcd for C18F5OH13N2: C,
58.70; H, 3.56; N, 7.61. Found: C, 58.74; H, 3.51; N, 7.57.
Bis(pentafluorophenyl)calix[4]phyrin (Porphodimethene, 5).
Following a literature procedure,16a the reduction of 2b (147 mg,
0.400 mmol) with NaBH4 (1.14 g, 30.0 mmol) in THF/methanol
(32 mL, 3:1) afforded the corresponding monocarbinol 2b-OH
which was used without purification. The monocarbinol was dried
under vacuum for 30 min and then immediately subjected to self-
condensation in the presence of InCl3 (56.6 mg, 0.256 mmol) in
CH2Cl2 (80 mL) for 30 min at room temperature. Oxidation of the
reaction mixture was carried out by the addition of DDQ (90.8 mg,
0.400 mmol) at room temperature. The reaction mixture was
basified by the addition of triethylamine (0.178 mL, 1.28 mmol)
and allowed to stir for 30 min. The reaction mixture was filtered
through a pad of silica gel and eluted with CH2Cl2 until the eluant
was no longer red/orange. The filtrate was concentrated to a bright
orange solid. Flash column chromatography [silica, CH2Cl2/hexanes
(1:5)] afforded a red powder (54 mg) upon evaporation of
the solvent. Crystallization (CH2Cl2/hexanes) provided red needle
crystals upon slow evaporation of the CH2Cl2 solvent (29 mg,
21%). Analytical data were in agreement with those previously
1
reported21 (see the Supporting Information for the H and 13C
NMR spectra).
HPLC Determination of the Yield of the 5-Isocorrole (3),
Porphyrin (4), and Porphodimethene (5). Analytical-scale reac-
tions of 1a þ 2a-OH and 1b þ 2b-OH were monitored for the
yield of the 5-isocorrole and porphyrin or porphodimethene by
adaptation of a literature method for the analysis of TpFPPhl in
crude reaction mixtures.8 An aliquot (1.2 mL) of a condensation
reaction mixture was transferred by adjustable pipet to a 1.8 mL
microcentrifuge tube containing DDQ (2.00 mg, 0.00881 mmol),
and the mixture was vortex mixed for 2-5 s. Triethylamine
(1 equiv relative to acid) was added. A portion of the oxidized
reaction mixture (1.0 mL) was transferred via adjustable pipet to
a Pasteur pipet filled two-thirds full with silica gel (∼1.5 g). The
sample was eluted with three 1-mL portions of CH2Cl2, and
562 J. Org. Chem. Vol. 75, No. 3, 2010

Flint et al.
JOCFeatured Article
removed for HPLC analysis. The remaining reaction mixture
was filtered through a pad of silica gel and eluted with CH2Cl2
until the eluant was no longer green. The filtrate was concentrated
and adsorbed onto silica gel (10 g), concentrated to dryness, and
purified by chromatography [silica, CH2Cl2/hexanes (1:5)] afford-
ing the 5-isocorrole contaminated with a low level of porphyrin
(132 mg, 98% 5-isocorrole by mass). Crystallization by slow
evaporation of CH2Cl2 from a CH2Cl2/hexanes solution provided
purple crystals of the 5-isocorrole (102 mg, 31%) devoid of
porphyrin: IR λabs (toluene, ε ꢀ 103) 339 (30.7), 423 (39.4), 671
(9.74), 731 (11.4); 1H NMR (CDCl3) δ 1.81 (s, 6H), 6.38 (m, 2H),
6.49 (dd, J = 2.5, 3.7 Hz, 1H), 6.83 (d, J = 4.7 Hz, 1H), 6.98 (dd,
J = 2.6, 3.7 Hz, 1H), 7.01 (d, J = 4.6 Hz, 1H). 7.03 (d, J = 4.7 Hz,
5-Isocorrole Stability Experiments (HPLC)8. In a darkened
laboratory, solutions of 5-isocorrole 3 were prepared in CH2Cl2
and MeOH. The concentration of each solution was adjusted so
that the initial peak area was between 600 and 700 area units.
The samples were analyzed in the dark by HPLC using the same
analysis parameters as described above for analysis of oxidized
reaction mixtures, and then the solutions were exposed to light
(conventional overhead fluorescent lighting). Chromatograms
were recorded at 15 min, 1 h, 2 h, 4 h, 6, h 8 h, 1 d, 1.5 d, 2 d, 3 d,
4 d, 5 d, 6 d, 7 d, 8 d, 9 d, 12 d, 13 d, and 14 d of continuous
exposure to room lights. Decomposition of the 5-isocorrole was
inferred by a decline in the peak area.
1H), 7.16 (d, J = 4.6 Hz, 1H), 13.70 (s, 1H), 13.87 (s. 1H) ; 13
C
Acknowledgment. Support for this work was provided by
the National Science Foundation under Grant No. 0517882.
High-resolution FAB mass spectra were obtained at the
Mass Spectrometry Laboratory for Biotechnology at North
Carolina State University. Partial funding for the facility was
obtained from the North Carolina Biotechnology Center
and the National Science Foundation.
NMR (CDCl3) δ24.9, 38.6, 108.3, 110.9, 111.7, 112.0, 114.0, 118.5,
119.9, 122.4, 124.2, 125.8, 126.3, 133.8, 134.4, 135.2, 141.0, 147.9,
149.1, 153.7, 161.2, 183.4 (additional weak signals were present in
the region of 135-147 ppm due to carbon atoms bearing a
fluorine); LD-MS obsd 658.1 (Mþ), calcd 658.1; HRMS (FAB)
659.1288 (M þ Hþ), calcd 659.1288 (C33H17F10N4).
5-Isocorrole Stability Experiments (UV-vis)8. In a darkened
laboratory, solutions of 5-isocorrole 3 were prepared in hexanes,
toluene, CH2Cl2, THF, ethyl acetate, acetone, acetonitrile, and
methanol. The concentration of each solution was adjusted so
that the maximum absorbance in the visible range (705-730 nm)
was between 0.3 and 0.5. UV-vis spectra were recorded in the
dark, and then the solutions were exposed to room lights
(conventional overhead fluorescent lighting). Spectra were
recorded at 15 min, 30 min, 1 h, 1.5 h, 2 h, 4 h, 6 h, 8 h, 1 d,
1.5 d, 2 d, 3 d, 4 d, 5 d, 6 d, 7 d, 8 d, 9 d, 10 d, 11 d, 12 d, 13 d, and
14 d of continuous exposure to room lights. Decomposition of
the 5-isocorrole was inferred from changes in the intensity of the
visible band (705-730 nm). Data were normalized with respect
to the maximum absorbance in the visible range prior to
exposure to light. As a control, an analogous experiment was
performed with solutions of the 5-isocorrole (hexanes, CH2Cl2,
and MeOH) maintained in the dark.
Supporting Information Available: General experimental
methods; discussion of HPLC method development, control
and reproducibility experiments, and calibration; discussion of
spectral resolution of the 5-isocorrole and porphodimethene,
discussion of the calculation of porphyrinoid yield from HPLC
peak area; representative chromatograms of reaction mixtures;
plots of the yield of the 5-isocorrole and self-condensation
product as a function of acid concentration; plots of the yield
of the 5-isocorrole and self-condensation product as a function
of condensation reaction time; data on the stability of the
5-isocorrole; NMR spectra of 2b and porphodimethene 5; data
on the purity of the 5-isocorrole; discussion of the assignment of
the 1H NMR spectrum of the 5-isocorrole; and NMR spectra of
the 5-isocorrole. This material is available free of charge via the
Internet at http://pubs.acs.org.
J. Org. Chem. Vol. 75, No. 3, 2010 563