Efficient synthesis of meso-substituted corroles in a H2O-MeOH mixture

Article abstract of DOI:10.1021/jo060007k

New and efficient conditions for the synthesis of meso-substituted corroles were developed. The first step, involving the reaction of aldehydes with pyrrole, was carried out in a water-methanol mixture in the presence of HCl. A relatively narrow distribut

Full text of DOI:10.1021/jo060007k

Efficient Synthesis of meso-Substituted Corroles in a H O-MeOH
2
Mixture
Beata Koszarna and Daniel T. Gryko*
Institute of Organic Chemistry of the Polish Academy of Sciences, Kasprzaka 44/52,
0
1-224 Warsaw, Poland
daniel@icho.edu.pl
ReceiVed January 3, 2006
New and efficient conditions for the synthesis of meso-substituted corroles were developed. The first
step, involving the reaction of aldehydes with pyrrole, was carried out in a water-methanol mixture in
the presence of HCl. A relatively narrow distribution of aldehyde-pyrrole oligocondensates was obtained
through careful control of their solubility in the reaction medium. After thorough optimization of various
reaction parameters (cosolvent, reagent, and acid concentration), high yields of bilanes were obtained.
Additionally, the bilane derived from 4-cyanobenzaldehyde was isolated, and the oxidative macrocy-
clization reaction was performed under various reaction conditions (different solvents, different
concentrations, and various oxidants). As a result, triphenylcorrole was obtained in the highest yield
(32%) reported to date. The scope and limitations studied showed that this method was particularly efficient
for moderately reactive aldehydes and those bearing electron-donating groups (yields 14-27%). Using
these conditions, corroles bearing strongly electron-donating groups were obtained for the first time. In
addition, it was found that the reaction of unhindered dipyrromethanes with aldehydes under analogous
conditions afforded trans-A B-corroles in very high yields (45-56%; 8-fold higher than previously
2
reported) without scrambling. The fact that scrambling was not observed in this reaction despite a very
high HCl concentration (0.3 M) is unprecedented. Detailed studies on the oxidation of bilane, derived
from sterically hindered dipyrromethane, allowed us to unequivocally establish that the yield of
1
macrocyclization is insensitive to the concentration. It was found the H NMR spectra of corroles in
deuterated TFA gave very sharp signals.
Introduction
pyrrole. Refinement of reaction conditions has improved corrole
yields to 15-20%. A broader investigation of the physicochem-
ical properties and practical applicability of corroles necessitates
the development of powerful new methods for their synthesis,
and although there has been dramatic progress in recent years,
there is still tremendous room for improvement.
Corroles, one carbon shorter analogues of porphyrins, recently
1
emerged as an independent area of research. Their coordination
2
3
4
5
chemistry, photophysics, synthesis, chemical transformations,
6
7
electrochemistry, and other properties have recently been
studied in great detail. Optimization of corrole synthesis has
8
been a frequent focus of research activities since the initial
The one-pot synthesis of meso-substituted A3-corroles from
aldehydes and pyrroles consists of two independent steps. The
first step is an acid-mediated electrophilic substitution to yield
reports by Gross et al.4 and Paolesse et al.,
a
4b,9
revealing one-
pot syntheses of meso-substituted corroles from aldehydes and
1
0.1021/jo060007k CCC: $33.50 © 2006 American Chemical Society
Published on Web 04/13/2006
J. Org. Chem. 2006, 71, 3707-3717
3707

Koszarna and Gryko
a mixture of various aldehyde-pyrrole oligocondensates, in-
cluding bilane (tetrapyrrane), a direct precursor of corrole. The
second step is the oxidative ring closure. Maximizing bilane
formation while minimizing the formation of dipyrromethanes
decrease in the yield of dipyrromethanes, probably due to the
formation of higher oligocondensates. This could be attributed
to the better solubility of dipyrromethanes in H2O/MeOH
mixtures, which permits further reaction. We reasoned that
careful optimization of the aldehyde/pyrrole ratio in conjunction
with the amount of MeOH might be a perfect way for narrowing
the distribution of oligocondensates and, thus, open the way to
a more efficient synthesis of corroles. Here we describe the
synthesis of A3-corroles as well as trans-A2B-corroles in an
H2O/MeOH mixture, which resulted from this study.
(DPMs), tripyrrane, and higher oligocondensates, is a difficult
task due to the similar reactivity of all these compounds. An
independent improvement of both processes (i.e., bilane forma-
tion and its macrocyclization) is the key to success, but the
optimization of the first step seems to be more crucial. Initial
attempts concentrated on the fine-tuning of the reaction condi-
8
c
tions, depending on the reactivity of various aldehydes.
Although these results were satisfactory (especially for alde-
hydes with electron-withdrawing groups), there was still room
Results and Discussion
A3-Corroles. Model Optimization Study. The reaction of
benzaldehyde (1) with pyrrole was chosen as a model system
for the optimization study because the yield of corrole 5 can be
easily compared with several existing procedures (Scheme 1).
The highest yield of corrole 5 (17%) has been reported by
for further improvements. The inspiration for this work came
from an intriguing paper by Kral and co-workers,10 describing
the synthesis of dipyrromethanes in water. The authors took
advantage of the difference in water solubility between the
substrates (aldehyde, pyrrole) and the product (dipyrromethane).
Exploiting this solubility difference, it was possible to essentially
stop the reaction at the dipyrromethane level using only a 6-fold
excess of pyrrole. Furthermore, the authors briefly mentioned
that the addition of MeOH to the reaction mixture led to a
8
1
1
8
d
Paolesse and co-workers. On the basis of the paper by Kral
1
0
et al., we chose the following initial conditions for the first
acid-catalyzed step: [ald. 1] ) 18 mM; ald. 1/pyrrole ) 3:4;
[HCl] ) 0.12 M; H2O/MeOH ) 1:1 (Table 1, entry 1).
Benzaldehyde (1) reacted with pyrrole under these conditions
to give a pink suspension of oligocondensates. These products
were extracted with CH2Cl2 and oxidized with DDQ. Though
(
1) (a) Paolesse, R. In The Porphyrin Handbook; Kadish, K. M., Smith,
K. M., Guilard, R., Eds.; Academic Press: New York, 2000; Vol. 2, pp
01-232. (b) Montforts, F.-P.; Glasenapp-Breiling, M.; Kusch, D. In
2
4
g
Geier and co-workers showed that DDQ is not the best
oxidizing agent for the synthesis of triphenylcorrole 5 because
it causes its partial decomposition, it was chosen in our study
for convenience (reaction with p-chloranil required a much
longer time). The use of DDQ would certainly decrease the
yields in comparison to p-chloranil, but the relative differences
in the yields between reactions under different conditions would
remain the same.
Under the reaction conditions initially chosen, corrole 5 was
obtained in a reasonable 14% yield. Initially we concentrated
our efforts on examining the influence of the benzaldehyde (1)/
pyrrole ratio on the yield of corrole 5 (Table 1, entries 1-3).
The best yield (21%) was obtained when the ratio was 1:2 (Table
Houben-Weyl Methods of Organic Chemistry; Schaumann, E., Ed.; Thi-
eme: Stuttgart, Germany, New York, 1998; Vol. E9d, pp 665-672. (c)
Gryko, D. T. Eur. J. Org. Chem. 2002, 1735-1742. (d) Guilard, R.; Barbe,
J.-M.; Stern, C.; Kadish, K. M. In The Porphyrin Handbook; Kadish, K.
M., Smith, K. M., Guilard R., Eds.; Elsevier Science: New York, 2003;
Vol. 18, pp 303-351. (e) Gryko, D. T.; Fox, J. P.; Goldberg, D. P. J.
Porphyrins Phthalocyanines 2004, 8, 1091-1105. (f) Nardis, S.; Monti,
D.; Paolesse, R. Mini ReV. Org. Chem. 2005, 2, 355-372.
(
2) (a) Meier-Callahan, A. E.; Di Bilio, A. J.; Simkhovich, L.; Maha-
mmed, A.; Goldberg, I.; Gray, H. B.; Gross, Z. Inorg. Chem. 2001, 40,
788-6793. (b) Gross, Z. J. Biol. Inorg. Chem. 2001, 6, 733-738. (e)
Ramdhanie, B.; Stern, C. L.; Goldberg, D. P. J. Am. Chem. Soc. 2001,
23, 9447-9448. (b) Edwards, N. Y.; Eikey, R. A.; Loring, M. I.; Abu-
6
1
Omar, M. M. Inorg. Chem. 2005, 44, 3700-3708. (b) Joseph, C. A.; Ford,
P. C. J. Am. Chem. Soc. 2005, 127, 6737-6743. (c) Collman, J. P.; Wang,
H. J. H.; Decr e´ au, R. A.; Eberspacher, T. A.; Sunderland, C. J. Chem.
Commun. 2005, 2497-2499.
1
, entry 2). Hence, for further reactions, it was kept constant at
(3) (a) Ding, T.; Alem a´ n, E. A.; Modarelli, D. A.; Ziegler, C. J. J. Phys.
Chem. A 2005, 109, 7411-7417. (b) Ventura, B.; Esposti, A. D.; Koszarna,
this value. Subsequently, solvent effects were investigated by
replacing MeOH with other water-miscible, polar solvents such
as i-PrOH, CH3CN, DMF, and THF.
We observed that bilanes were formed in negligible amounts
in all solvents studied, and only traces of corrole 5 were detected
B.; Gryko, D. T.; Flamigni, L. New J. Chem. 2005, 29, 1559-1566.
(
4) (a) Gross, Z.; Galili, N.; Saltsman, I. Angew. Chem., Int. Ed. 1999,
3
8, 1427-1429. (b) Paolesse, R.; Nardis, S.; Sagone, F.; Khoury, R. G. J.
Org. Chem. 2001, 66, 550-556. (c) Bri n˜ as, R. P.; Br u¨ ckner, C. Synlett
2
4
001, 442-444. (d) Gryko, D. T.; Jadach, K. J. Org. Chem. 2001, 66, 4267-
275. (e) Guilard, R.; Gryko, D. T.; Canard, G.; Barbe, J.-M.; Koszarna,
(Table 1, entries 4-7). Such poor results were primarily due to
B.; Brand e` s, S.; Tasior, M. Org. Lett. 2002, 4, 4491-4494. (f) Barbe, J.-
M.; Burdet, F.; Espinoza, E.; Gros, C. P.; Guilard R. J. Porphyrins
Phthalocyanines 2003, 7, 365-374. (g) Geier, G. R., III; Chick, J. F. B.;
Callinan, J. B.; Reid, C. G.; Auguscinski, W. P. J. Org. Chem. 2004, 69,
(8) (a) Ka, J.-W.; Cho, W.-S.; Lee, C.-H. Tetrahedron Lett. 2000, 41,
8121-8125. (b) Wasbotten, I. H.; Wondimagegn, T.; Ghosh, A. J. Am.
Chem. Soc. 2002, 124, 8104-8116. (c) Gryko, D. T.; Koszarna, B. Org.
Biomol. Chem. 2003, 1, 350-357. (d) 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. (e) Collman, J. P.; Decr e´ au,
R. A. Tetrahedron Lett. 2003, 44, 1207-1210.
(9) Note earlier reports: (a) Loim, N. M.; Grishko, E. V.; Pyshnograeva,
N. I.; Vorontsov, E. V.; Sokolov, V. I. IzV. Akad. Nauk, Ser. Khim. 1994,
925-927. (b) Rose, E.; Kossanyi, A.; Quelquejeu, M.; Soleilhavoup, M.;
Duwavran, F.; Bernard, N.; Lecas, A. J. Am. Chem. Soc. 1996, 118, 1567-
1568.
4
2
6
159-4169. (h) Jeandon, C.; Ruppert, R.; Callot, H. J. Chem. Commun.
004, 1090-1091. (i) Geier, G. R., III; Grindrod, S. C. J. Org. Chem. 2004,
9, 6404-6412. (j) Luguya, R. J.; Fronczek, F. R.; Smith, K. M.; Vicente,
M. G. H. Tetrahedron Lett. 2005, 46, 5365-5368.
(5) (a) Saltsman, I.; Mahammed, A.; Goldberg, I.; Tkachenko, E.;
Botoshansky, M.; Gross, Z. J. Am. Chem. Soc. 2002, 124, 7411-7420. (b)
Paolesse, R.; Nardis, S.; Venanzi, M.; Mastroianni, M.; Russo, M.; Fronczek,
F. R.; Vicente, M. G. H. Chem.sEur. J. 2003, 9, 1192-1197.
(6) Shen, J.; Shao, J.; Ou, Z.; E, W.; Koszarna, B.; Gryko, D. T.; Kadish,
K. M. Inorg. Chem. 2006, 45, 2251-2265.
(7) (a) DiNatale, C.; Salimbeni, D.; Paolesse, R.; Macagnano, A.;
(10) Kr a´ l, V.; Va sˇ ek, P.; Dolensk y´ , B. Collect. Czech. Chem. Commun.
2004, 69, 1126-1136.
DAmico, A. Sens. Actuators, B 2000, 65, 220. (b) Barbe, J.-M.; Canard,
G.; Brand e` s, S.; J e´ r oˆ me, F.; Dubois, G.; Guilard, R. Dalton Trans. 2004,
(11) Intrigued by the big difference in yields between two procedures
reported by Paolesse and co-workers (11% for neat reaction and 21% for
synthesis in CH2Cl2, respectively),8d we repeated their experiments with
very carefully purified, freshly distilled benzaldehyde. We found that the
yields are exactly 17% regardless of the method used (each experiment
was repeated three times by different researchers), which is in contrast to
what is reported. Consequently, benzaldehyde was freshly purified and
distilled before the present study to ensure repeatability of the results.
1
208-1214. (c) Radecki, J.; Stenka, I.; Dolusic, E.; Dehaen, W.; Plavec, J.
Comb. Chem. High Throughput Screening 2004, 7, 375-381. (d) Balazs,
Y. S.; Saltsman, I.; Mahammed, A.; Tkachenko, E.; Golubkov, G.; Levine,
J.; Gross, Z. Magn. Res. Chem. 2004, 42, 624-635. (e) Kadish, K. A.;
Shao, J.; Ou, Z.; Fr e´ mond, L.; Zhan, R.; Burdet, F.; Barbe, J.-M.; Gros, C.
P.; Guilard, R. Inorg. Chem. 2005, 44, 6744-6754. (e) Mahammed, A.;
Gross, Z. J. Am. Chem. Soc. 2005, 127, 2883-2887.
3708 J. Org. Chem., Vol. 71, No. 10, 2006

Efficient Synthesis of meso-Substituted Corroles
SCHEME 1
TABLE 2. Optimization of Conditions for the Conversion of
Bilane 4 into Corrole 6a
b
entry
solvent
yield of corrole 6 (%)
1
2
3
4
5
6
MeOH
CH3CN
THF
CHCl3
CH2Cl2
toluene
49
72
52
62
56
30
c
a
All reactions were performed under the following constant conditions:
[
bilane 4] ) 1.1 mM, p-chloranil (3 equiv versus bilane 4), 65 °C, 1 h.
b
Isolated yields. c Temperature was 42 °C.
the yield of corrole 5 (entries 8 and 9). When more MeOH was
used, the yield of trans-A2B2-porphyrin increased. Because the
concentration of an acid catalyst usually has a profound effect
on the yield of porphyrinoids, we next analyzed the role of HCl
concentration on the corrole 5 yield. Surprisingly, the concentra-
tion of HCl had a negligible effect on the yield of corrole 5
(entries 10 and 11). Extending the reaction time to 3 h or even
1
1
6 h also did not improve the outcome of the process (entries
2 and 13). Large volumes of solvents (MeOH and H2O) would
be required for preparative-scale synthesis if optimal conditions
were to be used directly. To eliminate this disadvantage, we
attempted to decrease the volume of the solvents (entries 14-
1
6). It should be noted that increasing the concentration of
starting materials should theoretically lead to the earlier
precipitation of dipyrromethanes and, hence, a decrease in the
yield of bilane. We observed that while a 2-fold increase in
concentration did not influence the yield of corrole 5, a further
increase gave corrole 5 in lower yield. Finally, control experi-
ments under optimized conditions with p-chloranil instead of
DDQ (entry 17) led to a substantial increase in the yield of
4g
corrole 5 to 32% (confirming previous observations ).
Oxidation Study. Our previous studies left some questions
concerning the dependence of corrole yield on the concentration
TABLE 1. Optimization of Conditions of the Benzaldehyde (1)
a
8c
Reaction with Pyrrole, Leading to Corrole 5
of bilane. Given the importance of this factor, we decided to
unequivocally establish the concentration dependence of the
bilane macrocyclization using pure bilane starting material rather
than a complicated mixture of oligocondensates. This would
simultaneously provide the maximal yield that could be reached
H2O/
cosolvent
(v/v)
yield of
ald. 1
ald.
HCl
corrole 5b
entry (mM) 1/pyrrole (mM)
cosolvent
(%)
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
18
12
6
3/4
1/2
1/4
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.06
0.36
0.12
0.12
0.24
0.36
0.48
0.25
1/1
1/1
1/1
1/1
1/1
1/1
1/1
2/1
1/2
1/1
1/1
1/1
1/1
1/1
1/1
1/1
1/1
MeOH
MeOH
MeOH
i-PrOH
CH3CN
DMF
14
21
16
1
0
0
1
15
5
19
20
21
19
22
17
15
32
4
e,8a
for this process.
Because the bilane derived from 4-cy-
anobenzaldehyde (2) is relatively stable and can be purified on
a reasonable scale, it was chosen for these studies. Bilane 4
was obtained as a mixture of diastereo- and regioisomers using
the new optimized conditions in 31% yield (Scheme 1).
Initially, we explored the role of solvent on the oxidation of
12
12
12
12
12
12
12
12
12
12
24
36
48
25
THF
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
4
with p-chloranil (Table 2). The rationale behind this study
1
1
1
1
1
1
1
1
was not only to maximize the yield, but also to check how broad
a range of solvents could be used to perform this reaction.
Corrole 6 was obtained in all solvents examined. We found that
the macrocyclization reaction afforded 60-70% yield of corrole
c
d
6
in CHCl3, CH2Cl2, and in CH3CN. Significantly lower yields
e
were obtained in toluene, MeOH, or DMF. Although CH3CN
gave the highest yield (72%), CHCl3 was chosen as the solvent
of choice for further studies because it made purification easier.
The influence of reaction concentration on the yield of corrole
a
All reactions were performed under the following constant conditions:
first step, 1 h, room temperature; second step, CHCl3, DDQ (1 equiv versus
b
c
aldehyde), room temperature. Isolated yields. Time of the first step was
d
e
12
3
h. Time of the first step was 16 h. DDQ was replaced with p-chloranil.
6 was studied next. The results of this study conducted in
CHCl3 are presented in Figure 1 and show that there is an
the improved solubility of the initially formed products and
higher oligocondensates in these solvents. Unexpectedly, only
dipyrromethane and tripyrrane were detected in THF. Increasing
or decreasing the H2O/MeOH ratio led to a sharp decrease in
(
12) The relationship between yields of corroles and dilution of the
reaction mixture after the first step was studied several times by us, but the
results were very inconsistent. Studies on pure bilane should remove any
uncertainty concerning this issue.
J. Org. Chem, Vol. 71, No. 10, 2006 3709

Koszarna and Gryko
TABLE 3. Comparison of Reaction Yields of Different
A3-Corroles
a
aromatic aldehyde
C6H5CHO, 1
corrole
yield (%)
b
5
20
21
22
23
24
25
26
6
27
28
29
30
27 (17)
25 (15)
25 (2.5)
22 (10)
b
4
4
4
4
3
3
4
4
3
4
4
2
2
2
-CH3C6H4CHO, 7
-FC6H4CHO, 8
-BrC6H4CHO, 9
-MeOC6H4CHO, 10
,4-(OCH2CO2Me)C6H3CHO, 11
,4,5-(MeO)3C6H2CHO, 12
-NO2C6H4CHO, 13
-CNC6H4CHO, 2
-CNC6H4CHO, 14
-CF3C6H4CHO, 15
-(CO2Me)C6H4CHO, 16
,6-Cl2C6H3CHO, 17
,6-(MeO)2C6H4CHO, 18
-formylthiophen, 19
b
b
b
22 (15)
17
14
b
10 (22)
21 (13)
23 (15)
16 (19)
17 (14)
8 (14)
b
b
b
b
b
FIGURE 1. Dependence of the yield of corrole 6 on the concentration
of bilane 4.
0
5
31
a
Isolated yields. b Values in parentheses indicate the highest yields
optimal concentration of tetrapyrrane (∼2.2 mM). The yield of
bilanes in the first step is around 30-50%. Hence, it can be
calculated from the above data that, in reactions carried out
without purification of tetrapyrranes, for 1 mmol of starting
aldehyde, ∼45-75 mL of CHCl3 should be used for the
oxidation step to ensure the highest yield of corroles.
Although the yields for the transformation of bilanes into
corroles were high with DDQ or p-chloranil, we also investi-
gated other oxidizing agents. One of the rationales behind this
study was to look for putative intermediates, such as biladienes,
of the oxidation process. We observed that mild oxidizing agents
such as p-benzoquinone could transform bilane 4 to corrole 6.
Though this process was very slow in CH2Cl2 at rt, we did not
detect any trace intermediates, which indicates that the final
step of this process must be very fast and the lifetime of the
intermediates is rather short.
obtained using previous methods.
robenzaldehyde, etc.), the reaction worked very well and gave
yields exceeding previous results (22-27%). In contrast, for
aldehydes bearing electron-withdrawing groups (13-16, 4-cy-
anobenzaldehyde, 4-carboxymethylbenzaldehyde, etc.), yields
were moderate and usually very similar to those reported earlier
(10-23%). In all these cases, the amount of porphyrins formed
was 1-3%. Interestingly, the reaction with C6F5CHO did not
result in the corresponding corrole, and sterically hindered
aldehyde 17 afforded corrole 30 in low yield.
Given the intrinsic polarity of aldehydes bearing electron-
donating groups, we reasoned that they could give good yields
of bilanes and, hence, corroles under H2O/MeOH/HCl condi-
tions. The use of p-chloranil (milder than DDQ) should eliminate
problems with their further oxidation. Eventually, we found that
Because the reaction could not be driven to completion with
a CHCl3 reflux, it was carried out in toluene. After 16 h of
reflux, tetrapyrrane 4 could not be detected and the yield of
corrole 6 was 45%, a higher value than with p-chloranil in the
3
,4,5-trimethoxybenzaldehyde (12) reacted with pyrrole giving
the corresponding tetrapyrrane, which could be oxidized to
corrole 25 in 14% yield. Surprisingly, this corrole was stable
enough to be purified with typical bench conditions. Notably,
it was shown that aldehyde 11, which bears hydrolytically labile
ester groups, afforded corrole 24 in 17% yields. The reaction
of 2,6-dimethoxybenzaldehyde (18) with pyrrole afforded cor-
role, but this compound was so unstable that it could not be
4
a
same solvent (Table 2). Based on the work by Paolesse et al.,
we also tried molecular oxygen as the oxidizing agent. The
reaction of bilane 4 performed in boiling acetic acid gave rise
to a significant amount of 5,10,15,20-tetrakis-(4-cyanophenyl)-
porphyrin and only a low yield of corrole 6, indicating bilane
scrambling was occurring. We also tried [bis(trifluoroacetoxy)-
iodo]benzene (PIFA), which is known to promote oxidative
dimerization of aromatic compounds. However, we found that
PIFA afforded only a 10% yield of corrole 6.
Scope of the HCl/H2O/MeOH Procedure. The optimized
conditions (first step, [ald.] ) 25 mM, [HCl] ) 0.25 M, H2O/
MeOH ) 1:1, ald./pyrrole ) 1:2; second step, p-chloranil [1
equiv versus aldehyde], CHCl3, 60 mL for 1 mmol of aldehyde)
were subsequently used for the synthesis of a broad spectrum
of A3-corroles (Table 3).
The initial experiments performed at a 2.5-fold larger scale
than the optimization study showed that some dipyrromethanes
precipitated too fast. Due to the detrimental effect of the addition
of more MeOH (Table 1, entry 9), we decided to decrease the
concentration of reagents as well as HCl and increase the
reaction time (Table 1, entry 12). The yield of corroles 7 and
4a
purified. On the other hand, 2-formylthiophene (19) subjected
to optimized reaction conditions gave corrole 31, although in a
surprisingly low yield (5%). Presumably, the yield suffers due
to the participation of the electron-rich thiophene ring in the
electrophilic substitution. As expected with pyridine-4-carbox-
aldehyde, the respective corrole formed in trace amounts because
salt formation prevents the favorable solubility controlled
aldehyde-pyrrole oligocondensates distribution. Taken together,
the results of the condensation of aldehydes with pyrrole in the
4b,c,8
H2O/MeOH/HCl system are superior to earlier studies.
Characterization of the Side Products of the Reaction. In
all cases, chromatographic purification afforded an olive-green-
brown band immediately before the corrole band. We once
4d
1
0 improved significantly, and these conditions were used for
hypothesized that it can be radical cations of corrole, but this
8
d
further studies.
view has been criticized. To gather additional information on
the structure of these species, we carried out two simple
experiments: (1) we reacted triphenylcorrole (5) with 1 equiv
of p-chloranil (quantitative formation of olive-green-brown
Nevertheless, the reaction results depended significantly on
the aldehyde used in the reaction. In general, for aldehydes
similar to benzaldehyde (7-10, 4-methylbenzaldehyde, 4-fluo-
3710 J. Org. Chem., Vol. 71, No. 10, 2006

Efficient Synthesis of meso-Substituted Corroles
band) and then added NH2NH213 (anhydrous, as THF soln.);
(entry 2). An increase in the concentration of substrates to 5
mM and 10 mM resulted in corrole 43 in 51-54% yield (entry
3 and 4), while further increases led to a substantial decrease
in yield (entry 5). Altering the stoichiometric ratio of substrates
did not change the outcome of this reaction (entries 6 and 7).
Analogous to A3-corroles synthesis, the ratio of MeOH to H2O
(2) we collected olive-green bands from several experiments
and added NH2NH2. In both cases, we regenerated the respective
corroles. There are two conclusions that can be drawn from these
experiments: (1) the olive-green species cannot be open-chain
compounds as previously proposed;8d (2) we think that hydrazine
acts as a mild reducing agent that reacts with the compound
that is easily and reversibly formed from corrole and an oxidant.
All corrole-forming reactions can be quenched with hydrazine
to increase the yield of corrole.
was studied. When the ratio of MeOH to H O was increased
2
from 1:1 to 2:1, the yield of corrole 43 significantly decreased
to 18% (entry 8). In contrast, reversing the ratio (i.e., H O/
2
MeOH ) 2:1) led to essentially the same yield (entry 9).
Furthermore, we observed that the pentapyrrotetramethenes
Notably, this represents an 8-fold improvement over the
14
4d
described by Gross et al. were ubiquitous side-products that
formed under the reaction conditions from pentapyrranes. They
were observed as a characteristic red spot on the TLC plate,
located just below the corrole in virtually every case. In one
case, the reaction of 3,4,5-trimethoxybenzaldehyde with pyrrole,
pentapyrrotetramethene 32 was isolated in 5.5% yield, and its
structure was confirmed by NMR and MS techniques (Structure
previously reported yield for the same conversion.
Scope of the HCl/H2O/MeOH. The optimized procedure
(Table 4, entry 3), which does not differ significantly from one
employed for the preparation of A3-corroles from aldehydes and
pyrrole was subsequently used to examine the scope of the
corrole formation. Dipyrromethanes 34-37 bearing electron-
withdrawing or electron-donating groups and various aldehydes
1
).
1
, 2, 14, 15, 41 were employed (Scheme 2). Results showed
that trans-A2B-corroles could be obtained in yields of 45-55%
with no detectable scrambling. The only examples that gave
lower yields were 5-(4-methoxyphenyl)dipyrromethane (36) and
5-(pentafluorophenyl)dipyrromethane (38). It is noteworthy that
the amounts of porphyrins formed under these conditions were
negligible.
To prove the scalability of this process, a large scale
preparation of corrole 45 was attempted. The reaction of
dipyrromethane 34 with aldehyde 15 performed at a 20-mmol
scale (10-fold increase) furnished 1.5 g of corrole 45 in
essentially the same yield as in the small scale. Importantly, it
was also found that 1.2 g of corrole 45 could be obtained without
any chromatography by crystallization of the crude product from
EtOH and CH2Cl2.
trans-A2B-Corroles Study with Model Unhindered Dipyr-
romethane. Reactions of dipyrromethanes with aldehydes
leading to trans-A2B-corroles have been extensively studied by
Study with Sterically Hindered Dipyrromethanes. While
expanding the scope of the trans-A2B-corroles’ synthesis to
include sterically hindered dipyrromethanes, we did observed
solubility problems. The reaction of mesityldipyrromethane (39)
with 4-cyanobenzaldehyde (2; Scheme 3) carried out under
previously optimized reaction conditions gave very low conver-
sion. The combined effect of a higher ratio of MeOH to H2O,
a higher concentration of acid, and longer reaction time (first
step, [ald.] ) 5.0 mM, [HCl] ) 0.4 M, H2O/MeOH ) 1:2, ald./
DPM ) 1:2, 2 h, rt; second step, p-chloranil [3 equiv versus
aldehydes], CHCl3) afforded corrole 50 in 30% yield (the same
4d,15c
4c,8c,15a,b,d
us
and other groups.
Previous studies showed that
reasonable yields of corroles could be obtained for sterically
hindered dipyrromethanes4
electron-withdrawing groups.1 However, from unhindered
-phenyldipyrromethane (33) and its analogues, corroles could
c,15b
and for dipyrromethanes bearing
5c
5
4c
be prepared only in very low yields (6-7%). All attempts to
increase the yield by an increase in the acid concentration led
to significant scrambling.4 To our delight, 5-phenyldipyr-
romethane (33) reacted with 4-cyanobenzaldehyde (2) under the
optimized conditions (Scheme 2) to afford the respective corrole
c
8
c
4
1
4 in 38% yield with no detectable scrambling (Table 4, entry
).
Encouraged by this remarkable result, we conducted a short
optimization study with this model reaction (Table 4). When
oxidation with p-chloranil was performed under milder condi-
tions (room temperature, overnight), the yield increased to 44%,
with a concomitant decrease in the olive-green side product
as reported). To determine the source of this difference, we
conducted a study of the oxidation of bilane derived from a
1
6
sterically hindered dipyrromethane. Bilane 53 was obtained,
using the H2O/MeOH/HCl method in 49% yield as white
crystals, and fully characterized (Scheme 3). Because this
compound was crystalline, it could be obtained in a very pure
state, which is very rare for bilanes.
The yield of corrole 50 was essentially concentration inde-
pendent (60 ( 2%) when the oxidation of bilane 53 with
p-chloranil was performed under various concentrations (20 mM,
4.0 mM, 1.0 mM, and 0.5 mM). The macrocyclization was
equally efficient for bilanes bearing sterically hindered and
unhindered substituents. The difference in the yield of corroles
was thus influenced by the first step of the synthesis. We also
carried out the oxidation of tetrapyrrane 53 with DDQ, which
resulted in a slightly lower yield of corrole 50 (55%). Corrole
(
13) Hydrazine hydrate was used by Paolesse and co-workers to remove
the excess of p-chloranil from a,c-biladiene oxidations to â-substituted
corroles: Paolesse, R.; Froiio, A.; Nardis, S.; Mastroianni, M.; Russo, M.;
Nurco, D. J.; Smith, K. M. J. Porphyrins Phthalocyanines 2003, 7, 585-
5
92.
(
14) Gross, Z.; Galili, N.; Simkhovich, L.; Saltsman, I.; Botoshansky,
M.; Blaser, D.; Boese, R.; Goldberg, I. Org. Lett. 1999, 1, 599-602.
15) (a) Asokan, C. V.; Smeets, S.; Dehaen, W. Tetrahedron Lett. 2001,
2, 4483-4485. (b) Gryko, D. T.; Piechota, K. E. J. Porphyrins Phthalo-
(
4
cyanines 2002, 6, 81-97. (c) Gryko, D. T.; Koszarna, B. Synthesis 2004,
2
2
205-2209. (d) Andrioletti, B.; Rose, E. J. Chem. Soc., Perkin Trans 1
51 was synthesized from 5-(2,6-dichlorophenyl)dipyrromethane
002, 715-716. (e) Barbe, J.-M.; Canard, G.; Brand e` s, S.; Guilard, R. Eur.
J. Org. Chem. 2005, 4601-4611.
(40; prepared using the new Lindsey procedure for sterically
J. Org. Chem, Vol. 71, No. 10, 2006 3711

Koszarna and Gryko
SCHEME 2a
a
Values in parentheses indicate the highest yield obtained using previous methods.
hindered dipyrromethanes)17 and 4-cyanobenzaldehyde (2) in
under slightly altered reaction conditions (first step, [ald.] )
8c
2
7% yield, 2-fold higher than previously reported. Recently
5.0 mM, [HCl] ) 0.78 M, H O/MeOH ) 2:1, 43:40 ) 1:2, 2
2
18
described phlorin-dipyrrin conjugates were also detected in
small amounts in these reactions, which further confirms the
idea of solubility-controlled distribution of products (hexapy-
rranes, which are direct precursors of phlorin-dipyrrin conju-
gates, could not form in sizable amounts). Interestingly,
dipyrromethane 40 reacted with pyridine-4-carboxaldehyde (43)
h, rt; second step, DDQ [3 equiv versus aldehydes], CHCl ) to
3
give corrole 55 in 16% yield (similar to previous results).1
5b
1
It is well-known that the H NMR spectra of free-base
corroles often display significant signal broadening. This
phenomenon is particularly strong for meso-substituted corroles
possessing unhindered substituents. Some of the spectra, which
we initially obtained in CDCl3 or THF-d8, had broad signals
that spread across the entire aromatic region. Consequently, we
decided to perform spectra with the addition of a small amount
of CD3OD, a procedure that was successfully used by Lee et
16) An additional motivation came from our previous studies,4d,8c,15b
(
which left some inconsistency about the dependence of the yield of the
macrocyclization of such bilanes on their concentration.
(
17) Laha, J. K.; Dhanalekshmi, S.; Taniguchi, M.; Ambroise, A.;
Lindsey, J. S. Org. Process Res. DeV. 2003, 7, 799-812.
18) Gryko, D. T.; Koszarna, B. Eur. J. Org. Chem. 2005, 3314-3318.
8
a
4g
(
al. and Geier et al. We obtained much clearer spectra for
3712 J. Org. Chem., Vol. 71, No. 10, 2006

Efficient Synthesis of meso-Substituted Corroles
TABLE 4. Optimization of Conditions of the
studied in-depth and, as a result, a broad variety of A3-corroles
has been synthesized in the highest yields yet reported. In
addition, it was found that (1) the macrocyclization of tetrapy-
rranes is strongly solvent-dependent and can be performed with
yields up to 73% (MeCN); (2) corroles bearing strongly electron-
donating substituents (like 3,4,5-trimethoxyphenyl) are stable
enough for purification; (3) methyl ester groups are stable under
the conditions of the first acid-catalyzed step; (4) linear
pentapyrrotetramethenes are typical side products in this reac-
tion; (5) p-benzoquinone can also be used for the oxidation of
bilanes to corroles; (6) p-chloranil gave higher yields than DDQ
in all cases except aldehydes with strong electron-withdrawing
groups; and (7) quenching the corrole-forming reactions with
hydrazine hydrate reduces the amount of olive-green-brown
side product, probably via its reduction to corrole.
We discovered that the reaction of sterically unhindered
dipyrromethanes with aldehydes can be performed in an H2O/
MeOH system with a high HCl concentration without any
scrambling. Although not fully explicable, this result led to the
development of a procedure that gives trans-A2B-corroles in
yields up to 56% and in turn makes these compounds easily
accessible on a large scale. Furthermore, it was observed that
the macrocyclization of bilanes bearing sterically hindered
groups led to corroles in the same yield as bilanes bearing
unhindered groups.
4
-Cyanobenzaldehyde (2) Reaction with 5-Phenyldipyrromethane
a
(33)
ald. 2
HCl H2O/MeOH yield of corrole
b
entry (mM) ald. 2/DPM 33 (M)
(v/v)
43 (%)
1c
2
3
4
5
6
7
8
9
2.5
2.5
5.0
10
20
5.0
5.0
5.0
5.0
1/2
1/2
1/2
1/2
3/4
3/4
1/3
1/2
1/2
0.30
0.30
0.30
0.30
1.20
0.30
0.30
0.20
0.20
1/1
1/1
1/1
1/1
1/1
1/1
1/1
1/2
2/1
38
44
54
51
19
50
50
18
53
a
All reactions were performed under the following constant conditions:
first step, 1 h, room temperature; second step, CHCl3, p-chloranil (1.5 equiv
b
versus dipyrromethane 33), room temperature, 16 h. Isolated yields.
c
Oxidation was performed under reflux for 1 h.
SCHEME 3
The conditions described in this paper provide by far the best
yields in both a one-pot synthesis of A3-corroles (20-30%) from
a wide range of aromatic aldehydes and in the [2+1] synthesis
of trans-A2B-corroles from sterically unhindered dipyrromethanes.
The unexpected discovery of scrambling-free high-acid con-
centration conditions opens new avenues of research in por-
phyrinoid chemistry.
Experimental Section
All chemicals were used as received unless otherwise noted.
2 2
Reagent grade solvents (CH Cl , hexanes, cyclohexane) were
1
13
distilled prior to use. All reported H and C NMR spectra were
recorded on 500- or 400-MHz spectrometers. Chemical shifts (δ
ppm) were determined with TMS as the internal reference; J values
are given in Hz. UV-vis spectra were recorded in toluene.
Chromatography was performed on silica (Kieselgel 60, 200-400
1
9
mesh), or dry column vacuum chromatography (DCVC) was
performed on preparative thin-layer chromatography silica. Mass
spectra were obtained via electrospray MS (ESI-MS). The purity
1
of all new corroles was established on the basis of H NMR spectra
1
5c
17
and elemental analysis. Aldehyde 11 and dipyrromethanes 33,
3
4-36,20 37, and 38-39 were prepared as described in the
literature.
General Procedures for the Investigation of the Reaction
Conditions. Optimization of Conditions of the Benzaldehyde
1) Reaction with Pyrrole Leading to Corrole 5 (Table 1).
21
17
(
Reactions were performed under the following conditions: ben-
zaldehyde (1 mmol, 102 µL) and pyrrole (2 mmol, 140 µL) were
dissolved in a specific amount of H O-cosolvent mixture. Sub-
2
sequently, HClaq (36%, 0.85 mL) was added, and the reaction was
stirred at room temperature. The progress of the reaction was
monitored by periodic removal of aliquots (containing precipitated
corroles 31, 45, and 46. During our study we also found that
spectra performed in pure deuterated TFA afforded sharp signals,
1
and so we obtained H NMR for corroles 24, 25, and 48 in this
solid) from the reaction mixture and vial extraction with CH
2 2
Cl ,
solvent. Pyrrolic N-H signals were obviously not detected in
either solvent.
followed by TLC examination (SiO , CH Cl /hexanes, 3:1). Fu-
2
2
2
(
(
19) Pedersen, D. S.; Rosenbohm, C. Synthesis 2001, 2431-2434.
20) Littler, B. J.; Miller, M. A.; Hung, C. H.; Wagner, R. W.; O′Shea,
Conclusions
D. F.; Boyle, P. D.; Lindsey, J. S. J. Org. Chem. 1999, 64, 1391-1396.
The acid-catalyzed formation of bilanes (tetrapyrranes) from
aldehydes and pyrrole in H2O/cosolvent mixtures has been
(
21) Clausen, C.; Gryko, D. T.; Yasseri, A. A.; Diers, J. R.; Bocian, D.
F.; Kuhr, W. G.; Lindsey, J. S. J. Org. Chem. 2000, 65, 7371-7378.
J. Org. Chem, Vol. 71, No. 10, 2006 3713

Koszarna and Gryko
+
migation of the developed TLC plates with bromine vapor
CAUTIONsfume hood) stains the dipyrromethane bright red, the
(C37
for C37
H
27
N
4
), 527.2230; found, 527.2255 [M + H ]. Anal. Calcd
4
(
H
26
N
: C, 84.38; H, 4.98; N, 10.64. Found: C, 84.17; H,
tripyrrane purple, and the bilane brown. Additionally, spots were
compared with authentic samples of dipyrromethane 33 and
diphenyltripyrrane, obtained using known procedures.1 After 1h,
there was no further progress in the reaction. The reaction was
5.00; N, 10.41. Other analytical data are consistent with literature
values.4b
7,22
5,10,15-Tris(4-cyanophenyl)corrole (6). The reaction mixture
was washed with a saturated solution of NaHCO , dried (Na SO ),
3
2
4
worked up by extraction with CHCl
twice with H O, dried (Na SO ), filtered, and diluted to 300 mL
with CHCl . Then DDQ or p-chloranil (1 equiv versus aldehyde)
was added, and the mixture was gently refluxed for 1 h. Corrole 5
and tetraphenylporphyrin were detected via a TLC comparison of
the reaction mixture with authentic samples. The yield of corrole
3
. The organic layer was washed
and filtered. The resulting solution was concentrated to about 50
mL and chromatographed (silica, CH Cl /hexanes, 1:1 then 2:1,
2
2
4
2
2
3
4:1) to give pure corrole (208 mg, 21%): ESI-HR calcd exact mass
+
(C H N ), 602.2088; found, 602.2084 [M + H ]. Anal. Calcd
4
0
24
7
for C H N : C, 79.85; H, 3.85; N, 16.30. Found: C, 79.73; H,
40
23
7
3.95; N, 16.25. Other analytical data are consistent with literature
values.8
c
5
was determined by the chromatographic separation (SiO
Cl /hexanes, 3:2) and the subsequent chromatographical purification
SiO , CH Cl /hexanes, 1:1).
Optimization of Conditions of the Conversion of Bilane 4 into
2 2
, CH -
2
5,10,15-Tris(4-methylphenyl)corrole (20). The reaction mixture
was washed with a saturated solution of NaHCO , dried (Na SO ),
(
2
2
2
3
2
4
and filtered. The mixture was concentrated to half the volume and
passed over a silica column (silica, CH Cl /hexanes, 2:3). The
subsequent chromatography (silica, CH Cl /hexanes, 2:3) afforded
pure corrole (240 mg, 25%): ESI-HR calcd exact mass (C40 ),
Corrole 6. Bilane 4 (0.164 mmol, 99.7 mg) was dissolved in a
given solvent (150 mL), and p-chloranil (0.492 mmol, 121 mg)
was added. The reaction mixture was heated at 65 °C for 1 h. After
cooling to room temperature, the resulting solution was washed
2
2
2
2
H
33
N
4
+
569.2600; found, 569.2718 [M + H ]. Anal. Calcd for C40
H
32
N
4
:
with a saturated solution of NaHCO
The yield of corrole 6 was determined by the chromatographic
separation (SiO , CH Cl /hexanes, 3:2).
3 2 4
, dried (Na SO ), and filtrated.
C, 84.48; H, 5.67; N, 9.85. Found: C, 84.27; H, 5.70; N, 9.83.
4
b
Other analytical data are consistent with literature values.
2
2
2
5,10,15-Tris(4-fluorophenyl)corrole (21). The reaction mixture
Optimization of Conditions of the 4-Cyanobenzaldehyde (2)
Reaction with 5-Phenyldipyrromethane (33; Table 4). Samples
of the 4-cyanobenzaldehyde (0.5 or 0.75 mmol) and 5-phenyldipyr-
romethane (1 or 1.5 mmol) were dissolved in MeOH (12.5-100
was evaporated to about 50 mL and passed over a silica column
(CH
rated with silica, and rechromatographed (silica, CH
1:1) to give corrole containing a few impurities. The subsequent
crystallization from CH Cl afforded almost pure corrole, which
was suspended in hot EtOH and filtered to give pure crystals (229
2
Cl
2
). All fractions containing corrole were combined, evapo-
2
Cl /hexanes,
2
mL). Subsequently, the solution of HClaq (36%, 1-5 mL) in H
2
O
2
2
(
12.5-100 mL) was added, and the reaction was stirred at room
temperature. The progress of the reaction was monitored by the
periodic removal of aliquots from the reaction mixture (containing
mg, 25%): ESI-HR calcd exact mass (C37
H
24
N
4
F
3
), 581.1948;
+
found, 581.1955 [M + H ]. Anal. Calcd for C37
H
N F : C, 76.54;
23 4 3
precipitated solid) and vial extraction with CH
2
Cl
2
, followed by
H, 3.99; N, 9.65. Found: C, 76.56; H, 3.76; N, 9.48. Other
analytical data are consistent with literature values.8
a
TLC examination (SiO , CH Cl /hexanes, 3:1). Fumigation of the
2
2
2
developed TLC plates with bromine vapor (CAUTIONsfume hood)
stains the dipyrromethane bright red, the tripyrrane purple, and the
bilane brown. The presence of remaining 5-phenyldipyrromethane
5,10,15-Tris(4-bromophenyl)corrole (22). The reaction mixture
was evaporated to dryness, suspended in hot EtOH, and filtered to
give corrole containing a few impurities. The subsequent crystal-
lization from CH Cl gave crystals of pure product (207 mg).
33 was detected by a comparison with authentic sample. The
2
2
presence of higher oligocondensates was estimated based on the
intensity of more polar spots on the TLC plate. After 1 h, p-chloranil
Because the second filtrate contained corrole, after evaporation with
silica, it was chromatographed (DCVC, silica, cyclohexane then
CH Cl /cyclohexane, 1:3) to give almost pure product, which was
(1.5 equiv versus dipyrromethane 33) was added and again TLC
2
2
examination and a comparison with authentic samples of corrole
crystallized from CH Cl to afford an additional 73 mg of corrole
2
2
4d
and 5,15-diphenyl-10,20-bis(4-cyanophenyl)porphyrin23 al-
43
22 (22%). Anal. Calcd for C H N Br × 0.5CH Cl : C, 55.90;
37
23
4
3
2
2
lowed us to identify these compounds. The yield of corrole 43 was
H, 3.00; N, 6.95. Found: C, 55.88; H, 2.89; N, 6.74. Other
analytical data are consistent with literature values.4b
determined by the chromatographic separation (SiO
hexanes, 3:2) and crystallization from hot EtOH.
2 2 2
, CH Cl /
5,10,15-Tris(4-methoxyphenyl)corrole (23). The reaction mix-
Optimization of Conditions of the Conversion of Bilane 53
into Corrole 50. Bilane 53 (0.1 mmol, 64 mg) was dissolved in a
ture was washed with a saturated solution of NaHCO
SO ), and filtered. The resulting solution was concentrated to half
the volume and passed over a silica column (silica, CH Cl /hexanes,
3:2). The fractions containing corrole were evaporated to dryness
and crystallized from CH Cl /hexanes to give pure corrole (222
mg, 22%): ESI-HR calcd exact mass (C40 ), 617.2547;
found, 617.2552 [M + H ]. Anal. Calcd for C40
0.5H O: C, 76.78; H, 5.32; N, 8.95. Found: C, 77.04; H, 5.17; N,
8.79. Other analytical data are consistent with literature values.
5,10,15-Tris[3,4-bis(methoxycarbonylmethoxy)phenyl]-
corrole (24). The reaction mixture was evaporated to dryness and
2 2 2 2
chromatographed (silica, CH Cl then CH Cl /EtOAc, 98:2, 95:5,
9:1). All fractions containing corrole were combined and evaporated
to dryness. The residue was crystallized from acetone/hexanes.
Because some impurities were still present, the crystals were
3 2
, dried (Na -
4
specific amount of CHCl
3
, an oxidant (0.30 mmol) was added, and
2
2
the reaction mixture was stirred overnight at room temperature.
The yield of corrole was determined by the chromatographic
2
2
separation (SiO
General Procedure for the Preparation of A
Aldehyde (5 mmol) and pyrrole (697 µL, 10 mmol) were dissolved
in MeOH (200 mL), and H O (200 mL) was added. Subsequently,
HClaq (36%, 4.25 mL) was added, and the reaction was stirred at
room temperature for 3 h. The mixture was extracted with CHCl
and the organic layer was washed twice with H O, dried (Na SO
filtered, and diluted to 300 mL with CHCl . p-Chloranil (1.23 g, 5
2
, CH
2
Cl
2
/hexanes, 2:3).
33 4 3
H N O
+
-Corroles.
H N
32 4
O
3
×
3
2
4
b
2
3
,
2
2
4
),
3
mmol) was added, and the mixture was refluxed for 1 h. The
purification details are described for each case as follows.
5
,10,15-Triphenylcorrole (5). The reaction mixture was passed
over a silica column (CH Cl ), and all fractions containing corrole
were combined and evaporated. The subsequent chromatography
silica, CH Cl /hexanes, 1:1) and crystallization (CH Cl /hexanes)
afforded pure corrole (235 mg, 27%): ESI-HR calcd exact mass
recrystallized from CH
2
Cl
2
/hexanes to afford pure corrole (304 mg,
) 0.68 (silica, CH Cl /acetone, 4:1);
17%) as dark fine crystals: R
f
2
2
2
2
1
3
H NMR (400 MHz, CF COOD) δ 3.99 (s, 3H), 4.01 (s, 6H), 4.02
(
2
2
2
2
(s, 3H), 4.08 (s, 3H), 4.11 (s, 3H), 4.97 (s, 4H), 5.01 (s, 4H), 5.09
(s, 4H), 6.77 (s, 2H), 7.02-7.11 (m, 2H), 7.19 (s, 1H), 7.27 (d, J
)
8.6 Hz, 2H), 7.36 (d, J ) 4.3 Hz, 2H), 7.47 (d, J ) 1.8 Hz, 1H),
7
7
.49 (s, 1H), 7.53 (d, J ) 4.3 Hz, 2H), 7.58 (d, J ) 1.8 Hz, 2H),
(
(
22) Ka, J.-W.; Lee, C.-H. Tetrahedron Lett. 2000, 41, 4609-4613.
23) Krebs, F. C.; Hagemann, O.; Spanggaard, H. J. Org. Chem. 2003,
.70 (d, J ) 1.8 Hz, 1H), 7.72 (d, J ) 1.8 Hz, 1H); ESI-MS obsd
1055.3 [M + H ]; ESI-HR calcd exact mass (C55
+
6
8, 2463-2466.
H
50
N
4
O
18Na),
3714 J. Org. Chem., Vol. 71, No. 10, 2006

Efficient Synthesis of meso-Substituted Corroles
077.3012; found, 1077.3033 [M + Na ]; λabs (toluene, ꢀ × 10-3)
+
71.86; H, 4.77; N, 7.80. Found: C, 72.09; H, 4.62; N, 7.60. Other
analytical data are consistent with literature values.8
1
4
c
24 (127.8), 526 (8.3), 565 (16.5), 621 (14.3), 653 (14.8) nm. Anal.
Calcd for C55
Found: C, 62.17; H, 5.20; N, 5.32.
H
50
N
4
O
18 × CH
3
OH: C, 61.87; H, 5.01; N, 5.15.
5,10,15-Tris(2,6-dichlorophenyl)corrole (30). The reaction
mixture was evaporated with silica and chromatographed (silica,
CH
were combined and evaporated to dryness. The residue was
crystallized from CH Cl /hexanes to give pure corrole (95 mg,
.8%). Anal. Calcd for C37 Cl : C, 60.60; H, 2.75; N, 7.64.
2 2
Cl /hexanes, 1:9 then 1:4, 1:2). All fractions containing corrole
5
,10,15-Tris(3,4,5-trimethoxyphenyl)corrole (25). The reaction
mixture was concentrated under reduced pressure and passed over
a silica column (silica, CH Cl then CH Cl /acetone, 96:4). All
2
2
2
2
2
2
7
H N
20 4
6
fractions containing corrole were combined and evaporated to
dryness. The residue was crystallized from acetone/cyclohexane
to give pure corrole (164 mg). Because the filtrate contained corrole,
it was rechromatographed (DCVC, silica, CH
acetone, 99:1) to obtain product which was crystallized from
acetone/cyclohexane to give an additional 22 mg of corrole (yield
Found: C, 60.66; H, 2.84; N, 7.39. Other analytical data are
consistent with literature values.4
a
2
Cl
2
then CH
2
Cl
2
/
5,10,15-Tris(2-thienyl)corrole (31). The reaction mixture was
evaporated to about 50 mL and passed over a silica column (silica,
CH
All fractions containing corrole were combined, evaporated with
silica, and rechromatographed (DCVC, silica, CH Cl /hexanes, 1:3
then 2:3) to give almost pure product, which was crystallized from
CH Cl /hexanes to afford pure corrole (46 mg, 5.0%): R ) 0.43
(silica, CH Cl /hexane, 3:2);
2 2
Cl /hexanes, 2:3) to obtain corrole containing a few impurities.
1
)
14%). R
MHz, CF
f
) 0.71 (silica, CH
2 2
Cl /acetone, 95:5); H NMR (400
3
COOD) δ 3.97 (s, 6H), 4.08 (s, 15H), 4.23 (s, 6H), 6.83
2
2
(
s, 2H), 6.84 (s, 2H), 6.87 (s, 2H), 7.35 (s, 2H), 7.47 (d, J ) 4.3
Hz, 2H), 7.52 (d, J ) 4.3 Hz, 2H), 7.62 (d, J ) 4.3 Hz, 2H); ESI-
2
2
f
+
1
H NMR (200 MHz, THF-d + 10%
MS obsd 797.3 [M + H ]; ESI-HR calcd exact mass (C46
H
45
N
4
O
9
),
)
2
2
8
+
-3
CD OD) δ 7.46-7.64 (m, 3H), 7.82 (s, 1H), 7.92-8.95 (m, 5H),
7
4
97.3181; found, 797.3182 [M + H ]; λabs (toluene, ꢀ × 10
23 (116.0), 570 (18.0), 621 (15.2), 653 (13.4) nm. Anal. Calcd
× CH OH: C, 68.10; H, 5.84; N, 6.76. Found: C,
8.01; H, 5.82; N, 6.59.
Pentapyrrotetramethene (32). All fractions containing a red
compound were combined and evaporated to dryness. The residue
was rechromatographed (DCVC, silica, CH Cl then CH Cl /acetone
8.5:1.5, 97:3, 95.5:4.5) to give the product containing a few
impurities. The subsequent crystallization from CH Cl /cyclohexane
afforded pure crystals (71 mg, 5.5%). R ) 0.72 (silica, CH Cl
) δ 3.80 (s, 12H), 3.88
br s, 12H), 3.90 (s, 6H), 3.94 (s, 6H), 5.94 (br s, 2H), 6.14 (br s,
3
8.62 (d, J ) 4.2 Hz, 2H), 8.69 (d, J ) 4.5 Hz, 2H), 8.91 (d, J )
for C46
H N
44 4
O
9
3
3.8 Hz, 2H), 9.04 (d, J ) 4.4 Hz, 2H); ESI-MS obsd 545.1 [M +
+
6
H ]; ESI-HR calcd exact mass (C31
H
21
N
4
S
3
), 545.0923; found,
+
-3
5
(
45.0897 [M + H ]; λabs (toluene, ꢀ × 10 ) 428 (163.0), 530
11.1), 578 (21.5), 629 (16.5), 664 (15.2) nm. Anal. Calcd for
: C, 68.35; H, 3.70; N, 10.29. Found: C, 68.26; H,
.64; N, 10.07.
-(2,6-Dichlorophenyl)dipyrromethane (40). 2,6-dichloroben-
zaldehyde (8.75 g, 50 mmol) was dissolved in pyrrole (347 mL, 5
mol), and MgBr (4.6 g, 25 mmol) was added. The reaction mixture
31 20 4 3
C H N S
2
2
2
2
3
9
5
2
2
f
2
2
/
1
2
acetone, 9:1). H NMR (100 MHz, CDCl
3
was stirred at room temperature for 1.5 h (slow darkening), and
powdered NaOH (10 g, 0.25 mol) was added. The mixture was
filtered through Celite, and the pyrrole was removed under vacuum.
The residue was dissolved and passed over a short silica column
(
2
3
1
1
1
1
3
H), 6.30-6.65 (m, 8H), 6.72 (br s, 6H), 6.82 (s, 2H), 12.61 (br s,
H); 13C NMR (400 MHz, CDCl
3
) δ 56.3, 60.9, 61.0, 109.1, 110.5,
14.4, 119.9, 125.7, 128.5, 128.9, 133.6, 134.0, 134.1, 135.4, 135.6,
37.6, 137.9, 146.5, 151.5, 151.9, 152.7, 165.9; ESI-MS obsd
(silica, AcOEt/hexanes, 1:3). All fractions containing product were
+
combined, evaporated, and crystallized from hot cyclohexane to
042.4 [M + H ]; ESI-HR calcd exact mass (C60
H
60
N
5
O
12),
23
+
-3
afford pure dipyrromethane 40 (8.7 g, 60%): mp 103-104 °C (lit.
Cl : C, 61.87; H, 4.15;
042.4233; found, 1042.4245 [M + H ]; λabs (toluene, ꢀ × 10
)
mp 102-103 °C). Anal. Calcd for C15
H
12
N
2
2
53 (41.1), 395 (39.5), 459 (25.6), 503 (52.9), 825 (7.3), 904 (8.9)
N, 9.62. Found: C, 62.03; H, 4.18; N, 9.74. Other analytical data
nm.
are consistent with literature values.19
5
,10,15-Tris(4-nitrophenyl)corrole (26). The reaction mixture
Cl ). All
fractions containing corrole were combined and evaporated to
dryness. The residue was crystallized from CH Cl to give pure
corrole (107 mg, 10%): ESI-HR calcd exact mass (C37 ),
2
General Procedure for the Preparation of trans-A B-Corroles
from Unhindered Dipyrromethanes: Dipyrromethane (1 mmol)
and aldehyde (0.5 mmol) were dissolved in MeOH (50 mL).
was passed over a chromatography column (silica, CH
2
2
2
2
Subsequently, a solution of HClaq (36%, 2.5 mL) in H
was added, and the reaction was stirred at room temperature for 1
h. The mixture was extracted with CHCl , and the organic layer
was washed twice with H O, dried (Na SO ), filtered, and diluted
to 250 mL with CHCl . p-Chloranil (369 mg, 1.5 mmol) was added,
2
O (50 mL)
H
H
24
N
7
O
6
+
6
62.1783; found, 662.1809 [M + H ]. Anal. Calcd for C37
0.5H
N, 14.57. Other analytical data are consistent with literature values.
23
N
7
O
6
3
×
2
O: C, 66.27; H, 3.61; N, 14.62. Found: C, 66.26; H, 3.42;
4b
2
2
4
3
5,10,15-Tris(3-cyanophenyl)corrole (27). The reaction mixture
and the mixture was stirred overnight at room temperature. The
purification details are described for each case as follows.
was evaporated to dryness and chromatographed (silica, toluene
then toluene/EtOAc, 98:2). The subsequent chromatography (silica,
toluene/EtOAc, 99:1) and crystallization from THF/Et
1
0-(4-Cyanophenyl)-5,15-diphenylcorrole (43). The reaction
mixture was concentrated to half the volume and passed over a
silica column (silica, CH Cl /hexanes, 3:2). All fractions containing
2
O afforded
),
pure corrole (226 mg, 23%): ESI-HR calcd exact mass (C40
H
24
N
7
2
2
+
6
02.2088; found, 602.2085 [M + H ]. Anal. Calcd for C40
H
23
7
N :
corrole were combined and evaporated to dryness. The resulting
solid was suspended in boiling EtOH, cooled, and filtered to give
C, 79.85; H, 3.85; N, 16.30. Found: C, 79.48; H, 3.55; N, 16.11.
Other analytical data are consistent with literature values.8
c
pure corrole (155 mg, 56%): ESI-HR calcd exact mass (C38
H
H
26
N
5
),
5
5
,10,15-Tris[(4-trifluoromethyl)phenyl]corrole (28). The reac-
tion mixture was concentrated to half the volume and passed over
a silica column (silica, CH Cl /hexanes, 1:1). The subsequent
chromatography of all fractions containing corrole (silica, CH Cl
hexanes, 2:3) and crystallization from CH Cl /hexanes afforded pure
corrole (198 mg, 16%): ESI-HR calcd exact mass (C40 ),
+
5
×
52.2188; found, 552.2204 [M + H ]. Anal. Calcd for C38
25
N
H
2
O: C, 80.12; H, 4.78; N, 12.29. Found: C, 80.45; H, 4.64;
4d
2
2
N, 12.15. Other analytical data are consistent with literature values.
0-(4-Nitrophenyl)-5,15-diphenylcorrole (44). The reaction
mixture was concentrated to half the volume and passed over a
silica column (silica, CH Cl /hexanes, 1:1). All fractions containing
2
2
/
1
2
2
24 9 4
H F N
2
2
+
7
31.1852; found, 731.1862 [M + H ]. Anal. Calcd for C40
F N
9 4
.41. Other analytical data are consistent with literature values.
23
H -
corrole were combined and evaporated to dryness. The resulting
solid was suspended in boiling EtOH, cooled, and filtered to give
: C, 65.76; H, 3.17; N, 7.67. Found: C, 65.28; H, 2.87; N,
8c
7
almost pure product. The crystals was crystallized from CH
2 2
Cl /
5
,10,15-Tris(4-carboxymethylphenyl)corrole (29). The reaction
hexanes to afford pure corrole (152 mg, 53%): ESI-HR calcd exact
+
mixture was evaporated to dryness and chromatographed (silica,
toluene/EtOAc, 95:5). A second chromatography (silica, toluene/
EtOAc, 99:1) and crystallization from CH
mass (C37
Calcd for C37
H
26
N
H
5
O
25
2
), 572.2081; found, 572.2099 [M + H ]. Anal.
O: C, 75.37; H, 4.62; N, 11.88.
N
5
O
2
× H
2
2
Cl
2
/hexanes afforded pure
× H O: C,
Found: C, 75.59; H, 4.52; N, 11.92. Other analytical data are
corrole (203 mg, 17%). Anal. Calcd for C43
H
32
N O
4 6
2
consistent with literature values.4d
J. Org. Chem, Vol. 71, No. 10, 2006 3715

Koszarna and Gryko
), 644.1937; found,
+
1
0-(3-Cyanophenyl)-5,15-bis(4-methylphenyl)corrole (45). The
reaction mixture was concentrated to half the volume and passed
over a silica column (silica, CH Cl /hexanes, 1:1). All fractions
H ]; ESI-HR calcd exact mass (C40
H
30
N
5
S
2
+
-3
644.1925 [M + H ]; λabs (toluene, ꢀ × 10 ) 426 (100.3), 589
2
2
(17.1), 618 (16.0), 653 (10.7) nm. Anal. Calcd for C40
CH OH: C, 72.89; H, 4.92; N, 10.36. Found: C, 73.04; H, 4.55;
N, 10.23.
H N S
29 5 2
×
containing corrole were combined and evaporated to dryness. The
resulting solid was suspended in boiling EtOH, cooled, and filtered
3
to give pure corrole (149 mg, 51%): R
hexane, 3:2); H NMR (200 MHz, THF-d
f
) 0.38 (silica, CH
+ 1% CD OD) δ 2.64
s, 6H), 7.63 (d, J ) 7.8 Hz, 4H), 7.92 (t, J ) 7.8 Hz, 1H), 8.08
2
Cl
2
/
10-(4-Cyanophenyl)-5,15-bis(pentafluorophenyl)corrole (49).
In this case, DDQ instead of p-chloranil was used as the oxidizing
agent. The reaction mixture was concentrated to half the volume
and passed over a silica column (silica, CH Cl /hexanes, 1:1). All
1
8
3
(
(
2
4
d, J ) 7.8 Hz, 1H), 8.22 (d, J ) 7.8 Hz, 4H), 8.37 (d, J ) 4.6 Hz,
2
2
H), 8.42-8.54 (m, 4H), 8.81 (d, J ) 4.8 Hz, 2H), 8.91 (d, J )
fractions containing corrole were combined and evaporated to
dryness. A second chromatography (silica, hexanes/acetone, 9:1 then
6:1) afforded pure corrole (73 mg, 20%). Anal. Calcd for
C H N F : C, 62.39; H, 2.07; N, 9.57. Found: C, 62.17; H, 1.87;
+
.4 Hz, 2H); ESI-MS obsd 580.3 [M + H ]; ESI-HR calcd exact
+
mass (C40
H
30
N
5
), 580.2496; found, 580.2518 [M + H ]; λabs
-
3
(
6
toluene, ꢀ × 10 ) 422 (126.4), 521 (8.1), 582 (15.4), 615 (12.6),
3
8
15
5 10
4d
51 (9.7) nm. Anal. Calcd for C40
H
29
N
5
: C, 82.88; H, 5.04; N,
N, 9.63. Other analytical data are consistent with literature values.
12.08. Found: C, 82.69; H, 4.98; N, 12.14. Large Scale Prepara-
General Procedure for the Preparation of trans-A B-Corroles
2
tion. Dipyrromethane 34 (2.36 g, 10 mmol) and 3-cyanobenzal-
dehyde 15 (650 mg, 5 mmol) were dissolved in MeOH (500 mL).
from Hindered Dipyrromethanes. Dipyrromethane (1 mmol) and
aldehyde (0.5 mmol) were dissolved in MeOH (100 mL). Subse-
Subsequently, the solution of HClaq (36%, 25 mL) in H
mL) was added, and the reaction was stirred at room temperature
for 1 h. The mixture was extracted with CHCl , and the organic
layer was washed twice with H O, dried (Na SO ), filtered, and
diluted to 1.25 L with CHCl . p-Chloranil (3.69 g, 15 mmol) was
2
O (500
quently, a solution of HClaq (36%, 5 mL) in H O (50 mL) was
2
added, and the reaction was stirred at room temperature for 2 h.
3
The mixture was extracted with CHCl , and the organic layer was
3
2
2
4
washed twice with H O, dried (Na SO ), filtered, and diluted to
2
2
4
3
250 mL with CHCl . p-Chloranil (369 mg, 1.5 mmol) was added,
3
added, and the mixture was stirred overnight at room temperature.
The whole reaction mixture was evaporated to dryness and boiled
with EtOH (70 mL). After cooling in the refrigerator, it was filtered
to afford crystals, which were dissolved in hot THF and precipitated
with cyclohexane. Filtration gave crystals of corrole 45 (794 mg).
Both filtrates were combined, evaporated to the dryness, and
and the mixture was stirred overnight at room temperature. The
purification details are described for each case as follows.
10-(4-Cyanophenyl)-5,15-dimesitylcorrole (50). The reaction
mixture was evaporated to dryness and chromatographed (silica,
CH Cl /hexanes, 2:3) to afford pure corrole (97 mg, 31%). Anal.
2
2
Calcd for C H N : C, 83.12; H, 5.87; N, 11.02. Found: C, 83.32;
44
37
5
suspended in CH
2
Cl
2
. Filtration gave crystals comprised of corrole
H, 6.05; N, 10.51. Other analytical data are consistent with literature
values.4d
and 2,3,5,6-tetrachlorohydroquinone. The latter one was removed
quantitatively by washing several times with warm EtOH to give
pure corrole 45 (446 mg). The filtrate was evaporated to dryness
with silica and chromatographed (DCVC, silica, cyclohexane then
CH Cl /cyclohexane, 2:3) to afford a fraction containing corrole
2 2
contaminated with other compounds. The resulting solid was
crystallized from THF/EtOH to give an additional crop of product
10-(4-Cyanophenyl)-5,15-bis(2,6-dichlorophenyl)corrole (51).
The reaction mixture was evaporated to dryness and chromato-
graphed (silica, toluene/hexanes, 4:1). A second chromatography
(DCVC, silica, acetone/hexanes, 1:9 then 1:6) afforded pure corrole
(93 mg, 27%): ESI-HR calcd exact mass (C H N Cl ), 688.0624;
3
8
22
5
4
+
38 21 5 4
found, 688.0624 [M + H ]. Anal. Calcd for C H N Cl : C, 66.20;
(
292 mg). Total yield, 1.532 g (53%).
0-Phenyl-5,15-bis(4-cyanophenyl)corrole (46). The reaction
mixture was concentrated to half the volume and passed over a
silica column (silica, CH Cl /hexanes, 3:2). All fractions containing
H, 3.07; N, 10.16. Found: C, 65.92; H, 3.29; N, 10.05. Other
analytical data are consistent with literature values.8
c
1
10-(4-Pyridyl)-5,15-bis(2,6-dichlorophenyl)corrole (52). Py-
2
2
ridine-4-carboxaldehyde (54 µL, 0.5 mmol) was dissolved in H O
2
corrole were combined and evaporated to dryness. The resulting
solid was suspended in boiling EtOH, cooled, and filtered to give
(50 mL), and HClaq (36%, 2.5 mL) was added. To this mixure was
added 2,6-dichlorophenyldipyrromethane (292 mg, 1 mmol) in
EtOH (25 mL). The reaction was stirred overnight at room
pure corrole (136 mg, 47%): R
f
) 0.54 (silica, CH
2 2
Cl /hexane,
+ 10% CD OD) δ 7.76 (s, 3H),
1
3
8
4
:2); H NMR (200 MHz, THF-d
8
3
temperature. NaOH (1.6 g) solution in H O (50 mL) was added,
2
.19 (d, J ) 7.6 Hz, 6H), 8.53 (d, J ) 5.0 Hz, 8H), 8.83 (d, J )
followed by CH Cl (50 mL). The organic phase was isolated,
2
2
.4 Hz, 2H), 8.99 (d, J ) 3.6 Hz, 2H); ESI-MS obsd 577.2 [M +
2 4
washed twice with water, dried (Na SO ), filtered, and diluted to
+
H ]; ESI-HR calcd exact mass (C39
H
25
N
6
), 577.1921; found,
250 mL with CHCl . DDQ (260 mg, 1.15 mmol) in THF (5 mL)
3
+
-3
5
77.1899 [M + H ]; λabs (toluene, ꢀ × 10 ) 432 (114.3), 577
17.8), 629 (12.8), 655 (12.9) nm. Anal. Calcd for C39 : C,
1.23; H, 4.20; N, 14.57. Found: C, 81.24; H, 4.02; N, 14.51.
0-(Pentafluorophenyl)-5,15-bis(4-methoxyphenyl)corrole (47).
The reaction mixture was evaporated to dryness and chromato-
graphed (silica, CH Cl /hexanes, 1:1). All fractions containing
corrole were combined, evaporated, and crystallized from CH Cl
hexanes to afford pure corrole (44 mg, 13%): ESI-HR calcd exact
was added, and the mixture was stirred at room temperature for 15
(
8
H N
24 6
min. The reaction mixture was concentrated to half the volume and
passed over a silica column (silica, CH Cl then acetone/hexanes,
2
2
1
5:95) to afford pure corrole (54 mg, 16%): ESI-HR calcd exact
+
mass (C H N Cl ), 664.0624; found, 664.0608 [M + H ]. Anal.
3
6
22
5
4
2
2
36 21 5 4
Calcd for C H N Cl : C, 64.98; H, 3.18; N, 10.53. Found: C,
2
2
/
64.71; H, 3.38; N, 10.41. Other analytical data are consistent with
literature values.8
c
+
mass (C39
Calcd for C39
6
H
26
N
4
O
2
F
5
), 677.1970; found, 677.1953 [M + H ]. Anal.
: C, 69.23; H, 3.72; N, 8.28. Found: C,
9.16; H, 3.74; N, 8.22. Other analytical data are consistent with
5,10,15-Tris(4-cyanophenyl)tetrapyrrane (4). Samples of 4-cy-
anobenzaldehyde 2 (7.9 g, 60 mmol) and pyrrole (8.34 mL, 120
25 4 2 5
H N O F
mmol) were dissolved in a mixture of MeOH (1 L) and H O (1 L).
2
4
d
literature values.
Subsequently, HClaq (36%, 30 mL) was added, and the reaction
1
0-(4-Cyanophenyl)-5,15-bis(4-methylthiophenyl)corrole (48).
The reaction mixture was concentrated to half the volume and
passed over a silica column (silica, CH Cl /hexanes, 3:2). All
was stirred at room temperature for 2 h. The mixture was extracted
with CH
dried (Na
2
Cl
2
, and the organic layer was washed twice with H
2
O,
2
2
2
SO
4
), and filtered. The reaction mixture was evaporated
fractions containing corrole were combined and evaporated to
dryness. The resulting solid was suspended in boiling EtOH, cooled,
to dryness and chromatographed (silica, EtOAc/hexanes, 9:1 then
4:1, 3:2, 1:1). All fractions containing product were combined and
rechromatographed (silica, toluene/EtOAc, 4:1). Removal of the
and filtered to give pure corrole (136 mg, 42%): R
f
) 0.48 (silica,
COOD) δ 2.67 (s,
H), 6.69 (s, 2H), 7.30 (d, J ) 3.4 Hz, 2H), 7.42 (s, 2H), 7.52 (d,
1
2
CH Cl
2
/hexane, 3:1); H NMR (400 MHz, CF
3
solvent afforded the product as an amorphous solid (3.75 g,
1
6
31%).%): R
MHz, CDCl
f
) 0.53 (silica, acetone/hexane, 2:3); H NMR (400
J ) 3.4 Hz, 2H), 7.58, 7.86 (AA′BB′, J ) 8.0 Hz, 2 × 4H), 7.65,
3
) δ 5.32-5.33 (m, 1H), 5.42 (d, J ) 4.7 Hz, 2H),
5.66-5.70 (m, 2H), 5.78-5.87 (m, 2H), 6.13-6.17 (m, 2H), 6.25
7.90 (AA′BB′, J ) 8.0 Hz, 2 × 2H); ESI-MS obsd 644.2 [M +
3716 J. Org. Chem., Vol. 71, No. 10, 2006

Efficient Synthesis of meso-Substituted Corroles
(
q, J ) 2.2 Hz, 1H), 6.72-6.74 (m, 2H), 6.80-6.85 (m, 1H), 7.23-
.32 (m, 6H), 7.51-7.60 (m, 6H), 7.91 (br s, 2H), 8.02 (br s, 2H);
3
12H), 2.26 (br s, 6H), 5.33 (br s, 1H), 5.67-5.70 (m, 2H), 5.80-
5.83 (m, 4H), 5.89-5.93 (m, 2H), 6.13-6.16 (m, 2H), 6.63-6.65
(m, 2H), 6.83 (br s, 4H), 7.25-7.28 (m, 2H), 7.54-7.57 (m, 2H),
7
1
3
C NMR (100 MHz, CDCl ) δ 43.98, 44.02, 44.05, 107.72, 108.08,
1
1
1
1
2
08.14, 108.69, 110.73, 110.81, 117.66, 117.86, 117.93, 118.00,
18.61, 118.70, 118.77, 126.98, 127.68, 128.20, 128.56, 128.91,
29.11, 129.16, 130.70, 130.92, 131.21, 131.47, 131.53, 132.34,
47.40, 171.20; νmax (KBr) 723, 769, 1021, 1250, 1414, 1500, 1605,
7.74 (br s, 2H), 7.83 (br s, 2H); 13C NMR (100 MHz, CDCl
3
) δ
20.53, 20.74, 38.31, 44.07, 44.11, 106.35, 106.38, 106.41, 106.85,
106.87, 106.89, 107.88, 107.91, 107.92, 108.54, 110.59, 110.61,
116.31, 116.34, 118.82, 128.00, 129.03, 129.67, 129.68, 129.71,
130.29, 131.03, 131.06, 131.09, 131.41, 131.45, 131.49, 132.22,
134.18, 136.60, 137.38, 147.88; νmax (KBr) 765, 856, 1028, 1091,
-
1
+
228, 3368 cm ; ESI-MS obsd 630.2 [M + Na ]; ESI-HR calcd
+
exact mass (C40
This compound is a mixture of regio- and stereoisomers.
0-(4-Cyanophenyl)-5,15-dimesityltetrapyrrane (53). Samples
of 5-mesityldipyrromethane 33 (2.64 g, 10 mmol) and 4-cyanoben-
zaldehyde 2 (655 mg, 5 mmol) were dissolved in MeOH (500 mL).
29 7
H N Na), 630.2377; found, 630.2372 [M + Na ].
-
1
1453, 1606, 2236, 2916, 3361 cm ; ESI-MS obsd 664.3 [M +
+
1
Na ]; ESI-HR calcd exact mass (C44
H
43
N
43 5
5
Na), 664.3411; found,
+
664.3433 [M + Na ]; Anal. Calcd for C44H N : C, 82.34; H, 6.75;
N, 10.91. Found: C, 82.56; H, 6.85; N, 10.70. This compound is
a mixture of regio- and stereoisomers.
Subsequently, HClaq (36%, 25 mL) in H
2
O (500 mL) was added,
and the reaction mixture was stirred at room temperature for a night.
The mixture was extracted with CH
2
Cl
2
, and the organic layer was
SO ), and filtered. After
Acknowledgment. This work was funded by the Ministry
of Scientific Research and Information Technology (Project 3
T09A 15728) and the Volkswagen Foundation. We thank
Mitsubishi Gas Chemical Company for the generous gift of
mesitaldehyde.
washed twice with H O, dried (Na
2
2
4
evaporation, the reaction mixture was chromatographed (silica,
EtOAc/hexanes, 9:1 then 4:1). All fractions containing product were
collected and evaporated to dryness. A subsequent crystallization
from EtOAc/hexanes gave pure tetrapyrrane (468 mg). Because the
filtrate contained product, it was chromatographed (silica, EtOAc/
hexanes, 9:1 then 4:1) to give pure tetrapyrrane as an oil that was
crystallized from EtOAc/hexanes. Filtration gave crystals (558 mg)
and an additional portion of product as a foamlike solid (554 mg).
Supporting Information Available: 1H NMR spectra of
corroles 24, 25, 31, 45, 46, and 48 and UV-vis absorption spectrum
of pentapyrromethene 32. This material is available free of charge
via the Intenet at http://pubs.acs.org.
Total yield, 1.58 g (49%); mp 203-205 °C; R
f
) 0.69 (silica,
) δ 2.02 (br s,
1
acetone/hexane, 2:3); H NMR (400 MHz, CDCl
3
JO060007K
J. Org. Chem, Vol. 71, No. 10, 2006 3717