Synthesis of locked meso-β-substituted chlorins via 1,3-dipolar cycloaddition

Article abstract of DOI:10.1021/jo060545x

A novel approach toward "locked" chlorins with increased stability has been studied in detail. The chlorin skeleton is assembled in a convergent fashion from two fragments via a porphyrin forming reaction, followed by 1,3-dipolar cycloaddition of azomethi

Full text of DOI:10.1021/jo060545x

Synthesis of Locked meso-â-Substituted Chlorins via 1,3-Dipolar
Cycloaddition
Michał Gałe¸zowski and Daniel T. Gryko*
Institute of Organic Chemistry of the Polish Academy of Sciences, Warsaw, Poland
daniel@icho.edu.pl
ReceiVed March 13, 2006
A novel approach toward “locked” chlorins with increased stability has been studied in detail. The chlorin
skeleton is assembled in a convergent fashion from two fragments via a porphyrin forming reaction,
followed by 1,3-dipolar cycloaddition of azomethine ylides, which are formed in situ. Central to the
success of the process is the presence of two electron-withdrawing groups in vicinal positions at the
perimeter of the porphyrin. As a result, the 1,3-dipolar cycloaddition took place regioselectively, on the
bond activated by two electron-withdrawing groups. Moreover, the chlorins formed are locked and hence
more stable because of the presence of two quaternary carbon atoms. Overall, in just six steps locked
chlorins were constructed from easily available materials. The large array of functionalities tolerated in
this approach validates it for a broad use in more advanced studies. The correlation between the results
of the 1,3-dipolar cycloaddition and dipolarophile (porphyrin) LUMO energy was extensively studied.
There was a definite correlation between the reaction time and the LUMO energy level, and a partial
correlation between the reaction yield and the distribution of the LUMO. Additionally, various approaches
toward crucial building blocks, namely 3,4-disubstituted-2,5-diformylpyrroles, were investigated.
Introduction
Another promising area of chlorins applications is photody-
namic therapy (PDT); a method utilizing visible or ultraviolet
light in combination with a photosensitizing agent to induce a
phototoxic reaction, which results in cell damage or death. With
a growing importance of PDT,4 which seems to be endowed
with several favorable features also for the treatment of localized
microbial5 and fungi infections,6 there is a need for the
The green color of leaves has intrigued humans for millennia.
It was not until the twentieth century that chlorin was discovered
to be the basic structural element responsible for this color.
Being involved in one of the most elaborate nanobiological
machines, the natural photosynthetic system,1 chlorins were
intensively studied and synthesized for more than a century.2
Recently, many studies have been focused on the use of artificial
photosynthesis to develop light-energy conversion systems.3
Although porphyrins lack some crucial photophysical features
they are often used instead of chlorins in the energy and electron-
transfer studies. The photophysical development of chlorins is
hampered by the lack of a reasonable methodology to construct
suitable chlorin building blocks.
(2) (a) Montforts, F.-P.; Glasenapp-Breiling, M. Prog. Heterocycl. Chem.
1988, 10, 1-24. (b) Smith, K. M. In Chlorophylls; Sheer, H., Ed.; CRC
Press: Boca Raton, FL, 1991; pp 115-143. (c). Montforts, F.-P.; Gerlach,
B.; Ho¨per, F. Chem. ReV. 1994, 94, 327-347. (d) Pavlov, V. Y.; Ponomarev,
G. V. Khim. Geterotsikl. Soedin. 2004, 483-519. (e) Huber, V.; Katterle,
M.; Lysetska, M.; Wu¨rthner, F. Angew. Chem., Int. Ed. 2005, 44, 3147-
3151. (f) Balaban, T. S.; Tamiaki, H. R.; A. Holzwarth. Supramolecular
Dye Chemistry; Wu¨rthner, F., Ed.; Topics in Current Chemistry, Vol. 258;
Springer: Berlin, Germany, 2005; pp 1-38. (g) Kozyrev, A. N.; Chen. Y.;
Goswami, L. N.; Tabaczynski, W. A.; Pandey, R. K. J. Org. Chem. 2006,
71, 1949-1960. (h) Sasaki, S.; Mizoguchi, T.; Tamiaki, H. Tetrahedron
2005, 61, 8041-8048.
(1) (a) McDermott, G.; Prince, S. M.; Freer, A. A.; Hawthornthwaite-
Lawless, A. M.; Papiz, M. Z.; Cogdell, R. J.; Isaacs, N. W. Nature 1995,
374, 517-520. (b) Roszak, A. W.; Howard, T. D.; Southall, J.; Gardiner,
A. T.; Law, C. J.; Isaacs, N. W.; Cogdell, R. J. Science 2003, 302, 1969-
1972. (c) Ben-Shem, A.; Frolow, F.; Nelson, N. Nature 2003, 426, 630-
635.
(3) (a) Wasielewski, M. R. Chem. ReV. 1992, 92, 435-461. (b) Gust,
D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 2001, 34, 40-48.
(4) (a) Jasat, A.; Dolphin, D. Chem. ReV. 1997, 97, 2267-2340. (b) Ali,
H.; van Lier, J. E. Chem. ReV. 1999, 99, 2379-2450.
10.1021/jo060545x CCC: $33.50 © 2006 American Chemical Society
Published on Web 06/30/2006
5942
J. Org. Chem. 2006, 71, 5942-5950

Locked meso-â-Substituted Chlorins
development of new agents. One potential advantage of chlorins
in PDT stems from their photophysical properties, the red shift
of the longest wavelength Q band (which enables lower energy
light to be used) along with the increased molar extinction of
this band. The majority of second generation PDT sensitizers
are chlorins.7-9
In our preliminary communication we presented a successful
realization of our idea of locked chlorin synthesis via regiose-
lective 1,3-dipolar cycloaddition.17 Here, we present the full
account which outlines the scope and limitations of this method,
the building block synthesis, and the relationship between the
structure of porphyrins and the efficiency of chlorin formation.
The major obstacle preventing broader use of chlorins in PDT
and artificial photosynthesis is the lack of their efficient
synthesis. Routes to chlorins for materials and biological
chemistry applications typically employ one of three strate-
gies: (1) derivatization of naturally occurring chlorins,2 (2)
transformation of synthetic porphyrins,10,11 or (3) total syntheses.
The first route provides access to large quantities of starting
materials, but restricts control over the pattern of substituents.
The transformation of synthetic porphyrins is compatible with
diverse substituents, yet it often suffers from adventitious
dehydrogenation that regenerates the porphyrin.10,12 In total
syntheses, it is possible to incorporate quaternary carbon atoms
in the reduced ring, thereby precluding adventitious dehydro-
genation and “locking” the chlorin oxidation state. However,
the total syntheses of such stable chlorins are rare and lengthy.
Typically they comprise 11-17 steps with the overall yield in
the range 0.04-1.1%.13-15 Despite further remarkable progress
in chlorin chemistry over the past decade, mainly due to the
work of Lindsey and co-workers,16 the task of developing novel
strategies for the preparation of these tetrapyrrole macrocycles
remains a challenge to state-of-the-art synthesis. This fact
provided us with the motivation to develop a simpler procedure
for the synthesis of inherently stable chlorins.
Results and Discussion
Design. Several issues merit particular consideration when
contemplating the synthesis of complex chlorins. From the
standpoint of synthetic economy, it is desirable to avoid the
protecting group manipulations which have come to dominate
traditional syntheses. Stabilization of the compounds can be
achieved by the introduction of a quaternary carbon atom(s) at
the reduced bond. To reduce the total number of steps, while
maintaining the ‘lock’, we considered various strategies.
Eventually we chose an approach based on Cavailero’s discov-
ery that simple meso-substituted A4-porprhyrins could act as
dienophiles in the Diels-Alder reaction and as dipolarophiles
in the 1,3-dipolar cycloaddition.18 In these reactions, chlorins
were formed as the main products but with bacteriochlorin and
isobacteriochlorin side products. These interesting results were
expanded19 to also include corroles, sapphyrins, hexaphyrins,
and octaphyrins.20 However, Cavaleiro’s approach has some
drawbacks: (1) the reaction is not regioselective, (2) chlorins
(and bacteriochlorins) formed are intrinsically unstable and can
easily oxidize back to porphyrins, and (3) separation could often
only be accomplished by preparative TLC. We envisioned that
if two electron-withdrawing groups were placed in vicinal
positions at the perimeter of the porphyrins, two problems could
be solved at the same time. First, the reaction might take place
regioselectively, on the bond activated by the two electron-
withdrawing groups. Second, the chlorins formed would be
locked because of the presence of two quaternary carbon atoms.
Such chlorins would be very interesting candidates for various
studies provided that (1) the synthesis of respective porphyrins
was straightforward, and (2) a potentially broad variety of
functional groups could be introduced. An alternative approach,
involving only one electron-withdrawing group placed at the â
position, was rejected because of uncertain regioselectivity, the
(5) (a) Maisch, T.; Szeimies, R.-M.; Jori, G.; Abels, C. Photochem.
Photobiol. Sci. 2004, 3, 907-917. (b) Jori, G. J. EnViron. Pathol., Toxicol.
2006, 25, 505-519.
(6) Smijs, T. G. M.; van der Haas, R. N. S.; Lugtenburg, J.; Liu, Y.; de
Jong, R. L. P.; Schuitmaker, H. J. Photochem. Photobiol. 2004, 80, 197-
202.
(7) Pandey, R. K.; Zheng, G. In Porphyrin Handbook; Kadish, K. M.,
Smith, K. M., Guilard, R., Eds.; Academic Press: San Diego, CA, 2000;
Vol. 6, pp 157-230.
(8) Vicente, M. G. H. Porphyrin-based sensitizers in the detection and
treatment of cancer: Recent progress. Curr. Med. Chem., Anti-cancer Agents
2001, 1, 175-194.
(9) One chlorin is already approved by EMEA (European Medicines
Agency) (m-tetra(hydroxyphenyl)chlorin, Foscan), for use in head squamous
cell carcinomas.
(10) Montforts, F.-P.; Glasenapp-Breiling, M.; Kusch, D. In Houben-
Weyl Methods of Organic Chemistry; Schaumann, E., Ed.; New York, 1998;
Vol, E9d, pp 614-635.
(17) Gryko, D. T.; Gałe¸zowski, M. Org. Lett. 2005, 7, 1749-1752.
(18) (a) Tome´, A. C.; Lacerda, P. S. S.; Neves, M. G. P. M. S.; Cavaleiro,
J. A. S. Chem. Commun. 1997, 1199-1200. (b) Silva, A. M. G.; Tome´, A.
C.; Neves, M. G. P. M. S.; Silva, A. M. S.; Cavaleiro, J. A. S. Chem.
Commun. 1999, 1767-1768. (c) Cavaleiro, J. A. S.; Neves, M. G. P. M.
S.; Tome´, A. C.; Silva, A. M. S.; Faustino, M. A. F.; Lacerda, P. S. S.;
Silva, A. M. G. J. Heterocycl. Chem. 2000, 37, 527-534. (d) Cavaleiro, J.
A. S.; Neves, M. G. P. M. S.; Tome´, A. C. ArkiVoc 2003, 14, 107-130.
(19) (a) Tome´, A. C.; Lacerda, P. S. S.; Silva, A. M. G.; Neves, M. G.
P. M. S.; Cavaleiro, J. A. S. J. Porphyrins Phthalocyanines 2000, 4, 532-
537. (b) Silva, A. M. G.; Tome´, A. C.; Neves, M. G. P. M. S.; Cavaleiro,
J. A. S. Tetrahedron Lett. 2000, 41, 3065-3068. (c) Silva, A. M. G.; Tome´,
A. C.; Neves, M. G. P. M. S.; Cavaleiro J. A. S. Synlett 2002, 1155-1157.
(d) Silva, A. M. G.; Tome´, A. C.; Neves, M. G. P. M. S.; Silva, A. M. S.;
Cavaleiro J. A. S.; Perrone, D.; Dondoni, A. Tetrahedron Lett. 2002, 43,
603-605. (e) Flemming, J.; Dolphin, D. Tetrahedron Lett. 2002, 43, 7281-
7283.
(20) (a) Barata, J. F. B.; Silva, A. M. G.; Faustino, M. A. F.; Neves, M.
G. P. M. S.; Tome´, A. C.; Silva, A. M. S.; Cavaleiro, J. A. S. Synlett 2004,
1291-1293. (b) Hata, H.; Shinokubo, H.; Osuka, A. Angew. Chem., Int.
Ed. 2005, 44, 932-935. (c) Hata, H.; Kamimura, Y.; Shinokubo, H.; Osuka,
A. Org. Lett. 2006, 8, 1169-1172. (d) Tome´, J. P. C.; Cho D.-G.; Sessler,
J. L.; Neves, M. G. P. M. S.; Tome´, A. C.; Silva, A. M. S.; Cavaleiro J. A.
S. Tetrahedron Lett. 2006, 47, 3131-3134. (e) Ostrowski, S.; Wyreˆbek,
P.; Mikus, A. Heterocycles 2006, 68, 885-888.
(11) For selected recent examples see the following: (a) Jaquinoid, L.;
Gros, C.; Khoury, R. G.; Smith, K. M. Chem. Commun. 1996, 2581. (b)
Balaban, T. S.; Linke-Schaetzel, M.; Bhise, A. D.; Vanthuyne, N.; Roussel,
C. Eur. J. Org. Chem. 2004, 3919-3930. (c) McCarthy, J. R.; Jenkins, H.
A.; Bru¨ckner, C. Org. Lett. 2003, 5, 19-22. (d) Fouchet, J.; Jeandon, C.;
Ruppert, R.; Callot, H. J. Org. Lett. 2005, 7, 5257-5260.
(12) Burns, D. H.; Shi, D. C.; Lash, T. D. Chem. Commun. 2000, 299-
300.
(13) Battersby, A. R.; Dutton, C. J.; Fookes, C. J. R. J. Chem. Soc.,
Perkin Trans. 1 1988, 1569-1576.
(14) (a) Strachan, J.-P.; O’Shea, D. F.; Balasubramanian, T.; Lindsey,
J. S. J. Org. Chem. 2000, 65, 3160-3172. (b) Balasubramanian, T.;
Strachan, J.-P.; Boyle, P. D.; Lindsey, J. S. J. Org. Chem. 2000, 65, 7919-
7929. (c) Jacobi, P. A.; Lanz, S.; Ghosh, I.; Leung, S. H.; Lo¨wer, F.; Pippin,
D. Org. Lett. 2001, 3, 831-834.
(15) (a) Montforts F.-P. Angew. Chem., Int. Ed. Engl. 1981, 20, 778-
779. (b) Schmidt, W.; Montforts, F.-P. Synlett 1997, 903-904. (c) Montforts,
F.-P.; Kutzki, O. Angew. Chem., Int. Ed. 2000, 39, 599-601.
(16) (a) Taniguchi, M.; Ra, D.; Mo, G.; Balasubramanian, T.; Lindsey,
J. S. J. Org. Chem. 2001, 66, 7342-7354. (b) Taniguchi, M.; Kim, M. N.;
Ra, D.; Lindsey, J. S. J. Org. Chem. 2005, 70, 275-285. (c) Kim, H.-J.;
Lindsey, J. S. J. Org. Chem. 2005, 70, 5475-5486. (d) Laha, J. K.; Muthiah,
C.; Taniguchi, M.; McDowell, B. E.; Ptaszek, M.; Lindsey, J. S. J. Org.
Chem. 2006, 71, 4092-4192.
J. Org. Chem, Vol. 71, No. 16, 2006 5943

Gałe¸zowski and Gryko
SCHEME 1
SCHEME 2
electron-withdrawing and electron-donating character. At the
same time, they form convenient handles for further chemical
transformations. Since regioselectivity was a crucial issue in
this approach, we felt it was very important to have strong
electron-withdrawing subunits, which might cause the formation
of two chlorins.
(3) We would like to optimize the efficiency of the building
block synthesis and the possible introduction of other electron-
withdrawing groups on the pyrrole unit. Various esters, cyano,
and keto functionalities were chosen for this study.
Building Block Synthesis. Facile generation of the pivotal
porphyrin core, endowed with suitable functionality for further
elaboration, relies on the availability of tripyrranes and pyrrole-
2,5-dicarboxaldehydes possessing additional electron-withdraw-
ing groups at positions 3 and 4. A broad range of tripyrranes
can be synthesized directly from aldehydes and pyrrole via acid-
mediated condensation. This reaction however is not selective,
and it gives a mixture of dipyrromethanes, tripyrranes, and
tetrapyrranes etc. The yields of tripyrranes can be optimized
via changing the ratio of aldehyde to pyrrole. Using such a
procedure,23 we synthesized tripyrranes 1-7 in 15-44% yield
(Scheme 2). The required tripyrranes were separated by column
chromatography but were very difficult to purify especially on
a larger scale. The limited stability and the existence of the
tripyrranes as a mixture of diastereomers and regioisomers
contributed to the purification problems. Consequently, they
were often used in a partially purified state. The only exception
was tripyrrane 8 (Table 2) possessing trifluoromethyl groups,
which was synthesized according to the new procedure devel-
oped by Dmowski et al.,24 and was a solid that could be purified
by recrystallization.
formation of two enantiomers, and the equally difficult prepara-
tion of the required porphyrins.21
Since porphyrins with such a pattern of substituents were not
previously known, we had to devise a method for their synthesis
(Scheme 1). We selected porphyrins bearing two meso and two
â substituents at designated places (i.e., 2, 3, 10, and 15). In
principle, the required porphyrins could be synthesized via a
[3 + 1] approach22 from tripyrranes and 3,4-disubstituted-2,5-
pyrroledicarboxaldehydes. Further retrosynthetic disconnection
of the required subunits produced pyrrole, aldehydes, and
â-ketoesters.
In a preliminary account,17 we showed that this strategy works
for a few combinations of substituents located at â and meso
positions. However, a study of the scope and limitations of this
promising approach was necessary before applying it to energy/
electron-transfer studies and PDT. We designed a very broad
group of target porphyrins for the following reasons:
(1) We would like to establish the range of substituents that
provide reasonable yields for the porphyrin to chlorin transfor-
mation.
(2) We would like to determine if there are any correlations
between the results of the final chlorin-forming cycloaddition
and the energy of the LUMO orbital and its local distribution.
To determine which structural differences do not interfere with
the cycloaddition, we prepared a complementary set of por-
phyrins (bearing the following substituents: 3,4,5-trimethoxy-
phenyl, 4-methoxyphenyl, mesityl-, phenyl, 4-cyanophenyl,
3-pyridyl, pentafluorophenyl and trifluoromethyl) and evaluated
their reactivity. The substituents used span a great range of
The synthesis of the second crucial building block, the
dialdehydes, represented a significant challenge, since the
presence of four electron-withdrawing groups excluded most
of the simple routes based on electrophilic substitution. We
finally settled on a simple synthesis based on the Paal-Knorr
condensation (Scheme 3). Esters of acetoacetic acid 9-11 were
25
oxidatively dimerized using I2 or cerium ammonium nitrate
(CAN)26 to give corresponding 1,4-diketones 12-14 (yield 22-
75%) bearing additional ester functionality. The Paal-Knorr
condensation27 gave symmetrically substituted pyrroles 15-17,
respectively, in 73-100% yield. Central to the success of this
strategy was the selective oxidation of methyl groups to formyl
groups. The model compound 16 was tested with a variety of
(21) Two examples of such an idea have been published (ref 19c and
Silva, A. M. G.; Tome, A. C.; Neves, M. G. P. M. S.; Silva, A. M. S.;
Cavaleiro, J. A. S. J. Org. Chem. 2005, 70, 2306-2314). Reaction of
â-nitroTPP with diazomethane and with sarcosine/CH2O system gave one
regioisomer of locked chlorin. One has to note, however, that the preparation
of starting porphyrins via nitration cannot be regioselectively done for more
complex meso-substituted porphyrins, thereby precluding the possibility of
the construction of chlorin building blocks.
(23) Ka, J.-W.; Lee, C.-H. Tetrahedron Lett. 2000, 41, 4609-4613.
(24) Dmowski, W.; Piasecka-Maciejewska, K.; Urban´czyk-Lipkowska,
Z. Synthesis 2003, 841-844.
(25) Wu, A.; Zhao, Y.; Chen, N.; Pan, X. Synth. Comm. 1997, 27, 331-
336.
(26) Cho, L. Y.; Romero, J. R. Tetrahedron Lett. 1995, 36, 8757-8760.
(27) Knorr, L. Chem. Ber. 1885, 18, 299-311.
(22) Lin, Y.; Lash, T. D. Tetrahedron Lett. 1995, 36, 9441-9444.
5944 J. Org. Chem., Vol. 71, No. 16, 2006

Locked meso-â-Substituted Chlorins
SCHEME 3
SCHEME 4
SCHEME 5
reagents, such as PbO2/Pb(CH3COO)4,28 Pb(CH3COO)4,29 2-
iodoxybenzoic acid (IBX), and PCC. In the end, we found that
the only reagent capable of oxidizing 16 to dialdehyde 19 was
CAN,30 although extensive modifications had to be made to the
existing procedure in order to obtain a reasonable yield. Both
original procedures (MeCN/H2O-RT and HOAc/THF/H2O-
RT) resulted in the formation of numerous sideproducts and
unsatisfactory yields of dialdehyde 19 (<25%). In contrast,
performing the reaction in MeCN/H2O at reflux temperature
resulted in a much cleaner reaction, higher yield of product
(50%), and, most importantly, easy separation. Application of
the optimized method to methyl and tert-butyl esters 15 and 17
led smoothly to the dialdehydes 18 and 20 in 25% and 50%
yield, respectively (Scheme 3). These fragments, that is, 18,
19, and 20, were thus prepared in three steps with the overall
yield of 26%, 37%, and 6%, respectively.
To install the necessary pyrrole building blocks in an even
more concise fashion, we employed a Huisgen 1,3-dipolar
cycloaddition of azlactones to esters of acetylenedicarboxylic
acid.31 Using this cascade reaction, pyrrole 16 could, in principle,
be synthesized from alanine and acetic anhydride. However,
Huisgen wrote that the reaction of azlactones derived from
R-amino acids bearing aliphatic side chains could not be stopped
at the level of compound 16, and Michael addition of a second
molecule of acetylenedicarboxylate occurred to give exclusively
24.32 We felt that pyrrole 16 could have been overlooked
because of the separation method used by Huisgen and
co-workers. To our delight, the exposure of N-acetyl-D,L-alanine
(22) to acetic anhydride in the presence of 2 equiv of diethyl
acetylenedicarboxylate furnished the expected 24, accompanied
by 16 in 23% yield (Scheme 4). Puzzled with this result we
resolved to repeat the reaction with dimethyl acetylenedicar-
boxylate and performed exactly according to the original
Huisgen procedure (D,L-alanine (21), 4 equiv of alkyne, 140
°C, 30 min). We obtained ester 15 in 31% yield accompanied
by adduct 23 (Scheme 4). While this shorter route to esters 15
and 16 is not comparable in terms of efficiency with previous
procedures, it can be very useful for the synthesis of unsym-
metrically substituted pyrrole building blocks.
Obviously the replacement of the ester group by another
moiety with a stronger electron-withdrawing effect could be
beneficial for both regioselectivity and efficiency of the 1,3-
dipolar addition. In this context we attempted to synthesize
pyrrole-2,5-dialdehydes with cyano or keto groups at positions
3 and 4. One of the most promising routes started from 3,4-
dicyano derivative 26, easily accessed from 2,5-dimethylpyrrole
(25) as described by Dolphin et al. (Scheme 5).33 The latter
one in turn was subjected to our refined conditions for the CAN-
(28) Battersby, A. R.; Dutton; C. J.; Fookes, C. J. R. J. Chem. Soc.,
Perkin Trans 1 1988, 1569-1576.
(29) Montforts, F.-P.; Schwartz U. M. Liebigs Ann. Chem. 1985, 2301-
2303.
(30) (a) Moreno-Vargas, A. J.; Robina, I.; Ferna´ndez-Bolan˜os J. G.;
Fuentes J. Tetrahedron Lett. 1998, 39, 9271-9274. (b) Thyrann, T.;
Lightner, D. A. Tetrahedron Lett. 1995, 36, 4345-4348.
(31) (a) Huisgen, R.; Gotthardt, H.; Bayer, H. O. Angew. Chem., Int.
Ed. 1964, 3, 135. (b) Gotthardt, H.; Huisgen, R.; Bayer, H. O. J. Am. Chem.
Soc. 1970, 92, 4340-4344. (c) Pizzorno, M. T.; Albonico, S. M. J. Org.
Chem. 1977, 42, 909-910.
(32) Bayer, H. O.; Gotthardt, H.; Huisgen, R. Chem. Ber. 1970, 103,
2356-2367.
(33) Bru¨ckner, C.; Xie, L. Y.; Dolphin, D. Tetrahedron 1998, 54, 2021-
2030.
J. Org. Chem, Vol. 71, No. 16, 2006 5945

Gałe¸zowski and Gryko
TABLE 1. Optimization of the Conditions for the Condensation of
Aldehyde 19 with Tripyrrane 6 Leading to Porphyrin 35a
with a change in the solvent to CHCl3 which might be attributed
to the presence of EtOH in commercially available CHCl3;
however, further addition of EtOH to 2.5% led to a decrease in
the yield of porphyrin 35. Changes of solvent and type of acid
did not improve the outcome of the process (Table 1, entries 8
and 9). Since the optimal conditions (entry 6) would require
large solvent volumes for preparative-scale synthesis, we
attempted to increase the reaction concentration. We observed
that a 10-fold increase in concentration gave porphyrin 35 in
much lower yield (Table 1, entry 10). This trend was overcome
to a certain extent by an increase in TFA concentration (entry
11). Finally, simultaneous addition of two reagents into a stirred
solution of TFA in a relatively small volume of CHCl3 solved
the problem (Table 1, entry 12). One of the most important
modifications which we were forced to apply was to replace
triethylamine as the neutralizing agent with another base.
Reaction quenching with triethylamine always created intractable
and persistent emulsions. The use of triethanolamine eliminated
this problem because the trifluoroacetate salt of triethanoloamine
precipitated from solution after 20 min at room temperature. A
simple filtration then removed the byproduct.
TFA
concn
[M]
substrate
concn
[mM]
yield of
porphyrin 35b
(%)
entry
solvent
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
1
2
3
4
5
0.026
0.001
0.01
0.1
40
1
1
1
1
0
3.4
8.3
17.7
19
0.5
6
0.1
1
CHCl3
26
7
8
0.1
0.1
0.1
0.1
1.0
1.0
1
1
1
10
10
10
CHCl3 + 2.5% EtOH
18.5
5.7
4
10.8
14
MeCN
9c
10
11
12
CHCl3
CHCl3
CHCl3
CHCl3
26
a All reactions were performed under the following constant conditions:
1st step, 15 min, room temperature; 2nd step, 20 min NEt3, room
temperature; 3rd step, 20 min, DDQ (1 equiv versus substrates), room
temperature. b Isolated yields. c TFA was replaced with BF3‚Et2O. d Each
substrate was dissolved in CHCl3 (concentration, 125 mM) and then
simultaneously syringed into a stirred solution of TFA in solvent over 15
min.
Model aldehyde 19, bearing ethyl ester functionalities, was
used in reactions with all remaining tripyrranes 1-5 and 7-8
(Table 2). All porphyrin forming reactions proceeded smoothly,
and they gave porphyrins 29-33, 37, and 38 in 7-30% yield.
Either electron-donating or electron-withdrawing groups could
be used with no significant differences in the results obtained
(Table 2). An exceptionally low yield was obtained for porphyrin
38. This result was attributed to low reactivity of tripyrrane 8
imparted by strongly electron-withdrawing groups. Aldehydes
18 and 20 were coupled with tripyrrane 6 to afford porphyrins
34 and 36 in yields lower (17% and 11%) than for ethyl ester
aldehyde 19 (Table 2).
mediated oxidation. Unfortunately, we obtained monoaldehyde
27 as the only isolable product in 63% yield (Scheme 5). All
attempts to further oxidize this compound to dialdehyde using
PCC, IBX, SeO2, or Pb(OAc)4 or directly oxidize 26 to
dialdehyde using the same reagents failed. Other unsuccessful
approaches to the corresponding dialdehyde are described in
the Supporting Information.
Furthermore, keto groups, on the condition that they would
not interfere with formyl groups during the porphyrin-forming
step, were another interesting option to be considered. One of
the promising building blocks was diketone 28, easily synthe-
sized via the two-step procedure described by Treibs and Jacob34
(Scheme 5). We tried to transform dimethyl derivative 28 into
the corresponding dialdehyde using CAN, PCC, IBX, or Pb-
(OAc)4 without any success (monoaldehyde was detected using
MS). Taken together, these last results show that CAN cannot
oxidize methylpyrroles bearing a few strongly electron-
withdrawing groups.
Synthesis of Porphyrins and Chlorins. The next stage of
our synthesis involved the construction of the porphyrin ring
via a [3 + 1] route. This strategy was extensively studied by
Lash,35 but there are almost no examples for the use of
â-unsubstituted tripyrranes in this reaction. Therefore the
condensation of model reagents (tripyrrane 6 with aldehyde 19)
was initially performed using the reaction conditions that were
developed during our study on the condensation of 5-pentafluo-
rophenyldipyrromethane and aldehydes leading to corroles.36
Since porphyrin 35 was not detected in this case, we modified
the reaction conditions by changing the concentration of TFA
and reagents. Various conditions were tested, and the results
are shown in Table 1. The increase in the concentration of TFA
combined with the dilution of the reaction mixture led to
porphyrin 35 in 3.4% yield. A further increase in acid
concentration gave the product in an appreciable yield of 19%
(Table 1, entry 3-5). The next improvement (26%) was reached
The stage was set for the pivotal cycloaddition reaction, which
we approached with some trepidation because of the regiose-
lectivity issue; in principle three regioisomeric chlorins could
be formed in this reaction. In this final step, we chose 1,3-dipolar
cycloaddition rather than the Diels-Alder reaction for the
following reasons: (1) Substrates are commercially available
and cheap (1,3-dipoles are generated in situ by thermal
decarboxylation of immonium salts obtained from condensation
of N-methylglycine and paraformaldehyde, in contrast to reactive
diene precursors used to date in the Diels-Alder reaction with
porphyrins). (2) The presence of a pyrrolidine ring in the product
imparts some polarity which should facilitate separation from
possible unreacted porphyrin. Our first model, porphyrin 35,
gave excellent results with 100% conversion and 30% yield of
the isolated purified product, utilizing slightly modified Cava-
leiro’s conditions (Table 2). To achieve full conversion of
porphyrin 35 we had to increase the reaction time and amounts
of dipole precursors. We tried to improve the yield by
performing the reaction with other forms of formaldehyde and
in different solvents. The reaction in toluene with trioxane gave
32% yield, whereas reactions carried out in DMF in the presence
of paraformaldehyde, trioxane, and formaline gave chlorin 45
in 0%, 0%, and 21% yield, respectively. Results revealed that
the desired chlorin 45 formed with full regioselectivity. The
structure of chlorin 45 was confirmed by the analysis of the 1H
NMR spectrum and mass spectrometry. The formation of
bacteriochlorin and isobacteriochlorin was not observed. All
other porphyrins 29-34 and 36-38 were subjected to the same
conditions and gave chlorins 39-48 in 10-57% yield (Table
(34) Treibs, A.; Jacob, K. Liebigs Ann. Chem. 1970, 740, 196-199.
(35) Lash, T. D. In Porphyrin Handbook; Kadish, K. M., Smith, K. M.,
Guilard, R., Eds.; Academic Press: San Diego, CA, 2000; Vol. 2, pp 125-
199.
(36) Gryko, D. T.; Koszarna, B. Synthesis 2004, 2205-2209.
5946 J. Org. Chem., Vol. 71, No. 16, 2006

Locked meso-â-Substituted Chlorins
TABLE 2. Results of the Synthesis of Porphyrins 29-38 and Chlorins 39-48, as Well as LUMO Energy of Chlorins
2). Reactions were always carried out until the disappearance
of the starting porphyrin. Depending on the substituents present,
this meant reaction times varied from 2 h to 1 week. The
presence of strongly electron-donating groups at meso positions
10 and 15 greatly lowered the rates of these reactions, with full
conversion taking 1 week under these conditions. The electronic
character of the meso substituent appears to dictate the yield,
with the strongest electron-donating groups affording the highest
yields. Cavaleiro and co-workers showed that the presence of
electron-withdrawing groups at meso positions increased the
yield of the cycloaddition reactions to the porphyrin.18c Our
examples prove that, if electron-withdrawing groups are intro-
duced at â positions, the presence of electron-withdrawing
groups at meso position is no longer necessary for the successful
1,3-dipolar cycloaddition. To demonstrate the practical potential
of the approach, selected examples were examined under
preparative scale conditions, furnishing 200 mg of chlorins 45
and 47.
diffraction analysis (Figure 1). This analysis fully confirms the
expected structure of this molecule. Chlorin ring is slightly
deformed from planarity.
One of our fundamental objectives to be established before
this methodology would be implemented on a larger scale was
the relationship between the structure of the porphyrin and the
yield of the chlorin. In this context, DFT calculations of the
energy of the lowest unoccupied molecular orbital (LUMO) of
the dipolarophile (porphyrin) were performed. The geometry
of the molecule was first optimized (B3LYP 3-21G*) and then
the LUMO was calculated using B3LYP 6-311G*. It is well-
known that azomethine ylide dipoles are characterized by a high-
lying highest occupied molecular orbital (HOMO) and LUMO.
Consequently, the dominant interaction is between the dipole
HOMO and the dipolarophile LUMO. The LUMO energy of
all porphyrins used is presented in Scheme 7 (and in Figure S1
in Supporting Information). The correlation between the energy
of the dipole LUMO and the reaction time is clearly visible.
The gap between the dipolarophile LUMO and the dipole
HOMO is the smallest for porphyrin 38 bearing two CF3 groups
(reaction time 2 h). On the antipode we have porphyrin 29 with
Numerous attempts to obtain X-ray quality crystals of any
of 10 prepared chlorins failed. Finally, we were able to obtain
crystals of chlorin 45 × TFA suitable for single-crystal X-ray
J. Org. Chem, Vol. 71, No. 16, 2006 5947

Gałe¸zowski and Gryko
Visual observations revealed that chlorins bearing electron-
withdrawing or sterically hindered groups at meso positions are
green, while other chlorins were distinctly violet-green. The
UV-vis spectra of these compounds (see Supporting Informa-
tion) provided an explanation of this phenomenon. The absorp-
tion spectrum of chlorins 41, 47, and 48 shows that the ratio of
a Soret band (∼408 nm) to last Q band (∼641 nm), is ∼3.5
while for the rest of the chlorins it is ∼5 (Figure 2, Supporting
Information). The last Q band is significantly stronger than in
the case of chlorins 45 and 46.
Conclusions. The chemistry described herein demonstrates
the power of 1,3-dipolar cycloaddition with porphyrins to
construct inherently stable (locked) chlorins. Our studies have
clearly documented the ability of this strategy to assemble
chlorin skeletons with excellent regioselective control. By using
this optimized sequence, chlorins could be prepared from
commercially available reagents in just six steps in a higher
overall yield than any previous approach based on total
synthesis. The scope of this approach proved to be remarkably
broad: all substituents studied were tolerated and could be used
at meso positions around the chlorin scaffold. This should allow
for fine-tuning of the photophysical properties for energy transfer
and PDT studies. The yield of cycloaddition was found to be
surprisingly dependent upon the electron-withdrawing and
electron-donating substituents located at the meso positions. The
exploration of the chemistry of pyrroles allowed us to obtain
key building blocks in very good overall yields. Additionally,
the utility and limitations of CAN-mediated oxidation of
methylpyrroles has been demonstrated. In terms of efficiency,
the standard ethyl ester group placed on the â positions generally
offered the best yield for both substrate synthesis and porphyrin
and chlorin formation. We found that the yield of porphyrins
strongly depended on the concentration of reagents and TFA.
The optimized conditions differ from the typical ones used for
â-substituted tripyrranes in that the concentration of TFA is
much higher ([TFA] ) 1.00 M). Chlorins bearing strongly
electron-withdrawing or sterically hindered substituents were
found to possess significantly stronger last Q band than the rest
of the studied chlorins. Furthermore the extraordinary ease with
which the starting materials for these reactions can be prepared,
coupled with the striking complexity which is attainable in one
step, makes this protocol ideal for the generation of a range of
chlorins. Applications of this knowledge to the synthesis of
complex chlorins should facilitate their further evaluation with
regard to their chemistry and medicine.
FIGURE 1. X-ray structure of chlorin 45 × TFA (hydrogens were
omitted for clarity).
FIGURE 2. LUMO maps of porphyrins 29, 32, 37, and 38 calculated
at the B3LYP/6-31G* level of DFT with a G-31G* basis set using
Spartan′04 for Windows.
a LUMO that is 0.53 eV higher and needed 1 week to react
completely.
Experimental Section
6,12-Bis(3,4,5-trimethoxyphenyl)tripyrrane (1). The solution
of 3,4,5-trimethoxybenzaldehyde (10 mmol, 1.96 g) in pyrrole (50
mmol, 3.47 mL) was stirred for 5 min at room temperature under
argon atmosphere, and then TFA (1 mmol, 77 µL) was added. The
whole mixture was stirred for 30 min and then combined with
aqueous NaOH (0.1 M, 10 mL) in order to quench the reaction.
The mixture was extracted with CH2Cl2, and the organic layer was
dried (Na2SO4). The solvents were removed under vacuum.
Chromatography of the resulting dark brown oil (silica, CH2Cl2/
ethyl acetate 95:5) afforded the product (0.5 g, 18%): Rf, 0.51 (CH2-
Cl2/ethyl acetate, 9:1). HRMS (EI): calcd for C32H35N3O6 [M+],
557.2526; found, 557.2520. Anal. Calcd for C32H35N3O6 C, 68.92;
H, 6.33; N, 7.54. Found: C, 68.78; H, 6.44; N, 7.72.
We also decided to calculate the so-called “LUMO-map”, in
which the absolute value of the LUMO is mapped onto a size
surface. The LUMO-map (Figure 2) shows which regions of
the molecule are most electron deficient (blue color) and, hence,
most subject to nucleophilic attack.37 For the most interesting
case, porphyrin 38, it can be seen that there is an equal
probability of the reaction occurring in two different positions,
yet only one regioisomer is formed in this reaction. In fact, the
yield of chlorins is the lowest in this case and the highest when
the difference of the local LUMO between two double bondss
the one activated by two ester groups and the opposite onesis
the highest (porphyrins 29-31, Table 2).
6,12-Dimesityltripyrrane (3). The solution of mesitaldehyde (10
mmol, 1.47 mL) in pyrrole (50 mmol, 3.47 mL) was stirred for 10
min at room temperature under argon atmosphere, and then TFA
(37) Hehre, W. J. A Guide to Molecular Mechanics and Quantum
Chemical Calculations; Wavefunction Inc.: Irvine, CA, 2003.
5948 J. Org. Chem., Vol. 71, No. 16, 2006

Locked meso-â-Substituted Chlorins
(1 mmol, 77 µL) was added. The whole mixture was stirred for 30
min and then combined with aqueous NaOH (0.1 M, 10 mL) in
order to quench the reaction. The mixture was extracted with CH2-
Cl2, and the organic layer was dried (Na2SO4). The solvent was
removed under vacuum, and the resulting dark brown oil was
distilled under vacuum. 5-Mesityldipyrromethane was distilled off
at 180-200 °C (0.5 mmHg). Chromatography (silica, ethyl acetate/
hexane, 1:9) of the remaining black solid afforded the product (0.35
g, 15%): Rf, 0.60 (ethyl acetate/hexane, 1:3). HRMS (EI): calcd
for C32H35N3 [M+], 461.2831; found, 461.2821.
6,12-Bis-(3-pyridyl)tripyrrane (5). The solution of 3-pyridin-
ecarboxaldehyde (75 mmol, 7.15 mL) in pyrrole (1.0 mol, 70 mL)
was heated in 85 °C for 24 h. After the mixture was cooled, pyrrole
was evaporated under vacuum. Chromatography of the residue
(silica, CH2Cl2/MeOH, 99:1) afforded 5-(3-pyridyl)dipyrromethane
(10.3 g, 62%) and the desired product 5 (2.28 g, 16%): Rf, 0.37
(CH2Cl2 /MeOH, 9:1). HRMS (EI): calcd for C10H10NO6 [M+],
379.1797; found, 379.1790.
(s, 1H). 13C NMR (125 MHz, CDCl3, Me4Si, δ ppm): 11.1, 13.6,
14.0, 14.2, 60.0, 61.9, 63.0, 113.2, 131.8, 134.0, 135.2, 162.2, 162.4,
165.2. HRMS (EI): calcd for C20H27NO8 [M+], 409.1737; found,
409.1727. Anal. Calcd for C20H27NO8: C, 58.67; H, 6.65; N, 3.42.
Found: C, 58.50; H, 6.63; N, 3.51. Also an additional crop of
compound 16 as white needles was obtained (278 mg, overall yield
23%): mp 96-98 °C (lit. 94-97 °C).38
2,5-Dimethyl-pyrrole-3,4-dicarboxylic Acid Di-t-butyl Ester
(17). 3,4-Diacetylsuccinic acid di-tert-butyl ester (14) (17 g, 0.054
mol) was suspended in aqueous ammonia solution (80 mL, 25%)
and stirred for 1 h. Filtration and washing three times with water
afforded pure product (16.0 g, 100%): mp. 220 °C (dec). IR νmax
(KBr)/cm-1: 3315, 2979, 1705, 1447, 1365, 1302, 1179, 1100. 1H
NMR (500 MHz, CDCl3, Me4Si, δ ppm): 1.54 (s,18H), 2.30 (s,
6H), 8.35 (s, 1H). 13C NMR (125 MHz, CDCl3, Me4Si, δ ppm):
12.4, 28.4, 80.0, 114.0, 131.3, 164.6. HRMS (EI): calcd for C16H25-
NO4 [M+], 295.1784;, found, 295.1772. Anal. Calcd for C16H25-
NO4: C, 65.06; H, 8.53; N, 4.74. Found: C, 64.92; H, 8.72; N,
4.54.
3,4-Diacetylsuccinic Acid Di-t-butyl Ester (14). After sodium
metal (0.1 mol, 2.3 g) was dissolved in warm tert-butyl alcohol,
tert-butyl acetoacetate (0.1 mol, 16.6 mL) was added, and the
mixture was stirred in a homogeneous state for 1 h. Subsequently
tert-butyl alcohol was removed in vacuo, the residue was suspended
in dry ether (150 mL), and a solution of iodine (0.05 mol, 12.7 g)
in dry THF (50 mL) was added dropwise with vigorous stirring
until the reaction mixture started to get purple. Next the reaction
mixture was filtered, and the solvent was evaporated. Crystallization
from AcOEt/hexane afforded the pure product (3.45 g, 22%): mp
87-89 °C (ethyl acetate/hexane). IR νmax/cm-1: 3013, 2986, 2944,
2,5-Diformyl-pyrrole-3,4-dicarboxylic Acid Dimethyl Ester
(18). 2,5-Dimethyl-pyrrole-3,4-dicarboxylic acid dimethyl ester (15)
(5.28 g, 25 mmol) was dissolved in mixture of acetonitrile (750
mL) and water (125 mL). CAN (110 g, 0.2 mol) was added, and
the mixture was stirred at 80 °C for 5 h. Then the reaction mixture
was cooled to room temperature, and water (200 mL) was added.
The layers were separated, and the aqueous layer was extracted
with ethyl acetate (100 mL). The combined organic layers were
washed again with water, dried (Na2SO4), and concentrated. After
crystallization (hexane/ethyl acetate), the product was isolated. This
solid was dried in vacuo (3.0 g, 50%): mp. 112-113 °C (ethyl
acetate/hexane). IR νmax (KBr)/cm-1: 3267, 1735, 1717, 1682, 1508,
1
1729, 1710, 1373, 1356, 1250, 1152, 1136. H NMR (500 MHz,
CDCl3, Me4Si, δ ppm): 1.43 (s, 18H), 2.40 (s, 6H), 4.35 (s, 2H).
13C NMR (125 MHz, CDCl3, Me4Si, δ ppm): 27.8, 59.0, 82.9,
166.2, 201.9. HRMS (ESI): calcd for C16H26O6Na [M + Na+],
337.1622; found, 337.1636. Anal. Calcd for C16H26O6: C, 61.13;
H, 8.34. Found: C, 60.80; H, 8.52.
2,5-Dimethyl-pyrrole-3,4-dicarboxylic Acid Dimethyl Ester
(15). Method A. 3,4-Diacetylsuccinic acid dimethyl ester 1226 (11.5
g, 0.05 mol) was dissolved in methanolic ammonia solution (50
mL, 2 M) and stirred for 1 h. Subsequently ethyl acetate (250 mL)
was added, and the reaction mixture was extracted with aqueous
HCl solution (5%), washed three times with water, and dried (Na2-
SO4). Next, solvent was evaporated, and the product was crystallized
from AcOEt/hexane (7.7 g, 73%): mp 115-117 °C (ethyl acetate/
hexane) (lit. 118-119 °C38).
Method B. D,L-Alanine (890 mg, 10 mmol) was suspended in
Ac2O (15 mL), and dimethyl acetylenedicarboxylate (5.0 mL, 41
mmol) was added. The reaction mixture was maintained at 140 °C
for 30 min, cooled, and evaporated under reduced pressure. The
oily residue was chromatographed (dry column vacuum chroma-
tography39 (DCVC), CH2Cl2 then CH2Cl2/MeOH, 99:1, 98:2) to
afford compound 23 as yellowish crystals (2.17 g, 62% yield; mp
118-119 °C [lit. 119-120 °C]32) and compound 15 as white
crystals (647 mg, 31% yield; mp 116-118 °C [lit. 118-119 °C38]).
Diethyl 2,5-Dimethyl-pyrrole-3,4-dicarboxylate (16). This
compound was obtained according to ref 27 and via 1,3-dipolar
cycloaddition: N-Acetyl-D,L-alanine (1.31 g, 10 mmol) was sus-
pended in Ac2O (10 mL), and diethyl acetylenedicarboxylate (3.2
mL, 20 mmol) was added. The reaction mixture was maintained at
110 °C for 2 h, cooled, and evaporated under reduced pressure.
The oily residue was dissolved in hot cyclohexane and left in the
refrigerator overnight with crystal seeds. White needles of 16 were
filtered (274 mg). The filtrate was evaporated and chromatographed
(DCVC, CH2Cl2 then CH2Cl2/EtOH, 99:1, 98:2) to afford compound
24 as yellowish oil (1.58 g, 39% yield). 1H NMR (500 MHz, CDCl3,
Me4Si, δ ppm): 1.11 (t, J ) 7.0 Hz, 3H), 1.26-1.38 (m, 9H),
2.20 (s, 6H), 4.10 (q, J ) 7.0 Hz, 2H), 4.21-4.39 (m, 6H), 7.39
1
1292, 1227, 1104. H NMR (500 MHz, CDCl3, Me4Si, δ ppm):
3.97 (s, 6H), 10.11 (s, 2H), 10.50 (s, 1H). 13C NMR (125 MHz,
CDCl3, Me4Si, δ ppm): 52.8, 122.8, 133.3, 162.4, 181.5. HRMS
(EI): calcd for C10H9NO6 [M+], 239.0430; found, 239.0424.
2,5-Diformyl-pyrrole-3,4-dicarboxylic Acid Diethyl Ester (19).
2,5-Dimethyl-pyrrole-3,4-dicarboxylic acid diethyl ester 1627 (8.4
g, 35 mmol) was dissolved in a mixture of acetonitrile (1050 mL)
and water (175 mL). Then CAN (172.7 g, 315 mmol) was added,
and the resulting mixture was stirred at 80 °C for 5 h. Then it was
cooled to room temperature, and water (200 mL) was added. The
layers were separated, and the aqueous layer was extracted with
ethyl acetate (100 mL). The combined organic layers were washed
again with water, dried (Na2SO4), and concentrated. Column
chromatography (silica, hexane/ethyl acetate, 3:2) afforded a pale
yellow oil which was cooled in the refrigerator and then warmed
back to room temperature. After this operation was repeated, oil
turned into crystals of the product (4.7 g, 50%): mp 41-43 °C. IR
(CH2Cl2)/cm-1: 3229 (br), 2984, 1716, 1688, 1502, 1284, 1225,
1101. 1H NMR (500 MHz, CDCl3, Me4Si, δ ppm): 1.33 (t, 6H, J
) 7.1 Hz), 4.34 (q, 4H, J ) 7.1 Hz, 4H), 10.02 (s, 2H), 10.47
(s,1H). 13C NMR (125 MHz, CDCl3, Me4Si, δ ppm): 14.39, 62.70,
123.60, 134.24, 163.44, 182.61. HRMS (EI): calcd for C12H13-
NO6 [M+], 267.0743; found, 267.0732. Anal. Calcd for C12H13-
NO6: C, 53.93; H, 4.90; N, 5.24. Found: C, 53.86; H, 5.08; N,
5.30.
2,5-Diformyl-pyrrole-3,4-dicarboxylic Acid Di-t-butyl Ester
(20). 2,5-Dimethyl-pyrrole-3,4-dicarboxylic acid di-tert-butyl ester
(17) (886 mg, 3 mmol) was dissolved in mixture of acetonitrile
(90 mL) and water (15 mL). CAN (14.8 g, 27 mmol) was added,
and the mixture was stirred at room temperature for 20 h. Then
water (200 mL) was added. The layers were separated, and the
aqueous layer was extracted with ethyl acetate (100 mL). The
combined organic layers were washed again with water, dried (Na2-
SO4), and concentrated. After crystallization (hexane/ethyl acetate),
product was isolated. This solid was dried in vacuo (240 mg,
25%): mp 143-145 °C (ethyl acetate/hexane). IR νmax (KBr)/cm-1
:
3.145, 2981, 1732, 1709, 1687, 1498, 1370, 1298, 1234, 1161, 1108.
(38) Pedersen D. S.; Rosenbohm C. Synthesis 2001, 2431-2434.
(39) Gabel, N. W. J. Org. Chem. 1962, 27, 301-303.
1H NMR (500 MHz, CDCl3, Me4Si, δ ppm): 1.61 (s, 18H), 10.10
J. Org. Chem, Vol. 71, No. 16, 2006 5949

Gałe¸zowski and Gryko
(s, 2H). 13C NMR (125 MHz, CDCl3, Me4Si, δ ppm): 28.18, 83.30,
125.0, 132.7, 160.9, 181.6. HRMS (ESI): calcd for C16H25NO4Na
[M + Na+], 346.1261; found, 346.1266. Anal. Calcd for C16H21-
NO6: C, 59.43; H, 6.55; N, 4.33. Found: C, 59.23; H, 6.42; N,
4.34.
and 582 (5000). 1H NMR (500 MHz, CDCl3, Me4Si, δ ppm): -4.47
(s, 2H), 1.77 (t, 6H, J ) 7.2 Hz), 4.91 (q, 4H, J ) 7.2 Hz), 9.25
(d, 2H, J ) 4.4 Hz), 9.42 (s, 2H), 9.52 (s, 2H), 10.27 (s, 2H).
HRMS: calcd for C28H21F6N4O4 [M + H+], 591.1462; found,
591.1484. Anal. Calcd for C28H20F6N4O4‚0.5H2O: C, 56.10; H,
3.53; N, 9.35. Found: C, 55.71; H, 3.33; N, 9.16.
General Procedure for the Synthesis of Chlorins: Porphyrin
(0.1 mmol), paraformaldehyde (10 equiv), and sarcosine (4 equiv)
were placed in dry toluene (20 mL). Then flask was flushed with
argon and refluxed with Dean-Stark apparatus. Every 5 h the next
portions of paraformaldehyde (10 equiv) and sarcosine (4 equiv)
were added until all porphyrin disappeared. Subsequently, the
reaction mixture was cooled, and the solvent was removed by rotary
evaporation. The purification details are described for each case as
follows.
3,4-Dicyano-2-formyl-5-methylpyrrole (27). 2,5-Dimethyl-3,4-
dicyanopyrrole (435 mg, 3 mmol) was dissolved in mixture of
acetonitrile (90 mL) and water (15 mL). Then CAN (4.5 equiv,
7.4 g) was added, and the mixture was stirred at 80 °C for 5 h.
Then the reaction mixture was cooled to room temperature, and
water (200 mL) was added. The layers were separated, and the
aqueous layer was extracted with ethyl acetate (100 mL). The
combined organic layers were washed again with water, dried (Na2-
SO4), and concentrated. Chromatography (toluene/ethyl acetate 4:1)
afforded the product (300 mg, 63%): mp. 158-160 °C (ethyl
acetate/hexane). IR νmax (KBr)/cm-1: 3201, 3131, 2249, 2239, 1677,
Chlorin 41. Dry column vacuum chromatography [silica, toluene/
ethyl acetate (9:1)] afforded the product as a green solid which
was recrystallized from MeOH (43 mg, 57%): Rf ) 0.5 (silica,
toluene/ethyl acetate 4:1). λmax (toluene)/nm: 413 (ꢀ/dm3 mol-1
cm-1 119000), 507 (13300), 530 (5090), 587 (4760), and 637
1
1517, 1421, 1257, 1257, 1016, 881. H NMR (200 MHz, (CD3)2-
SO, δ ppm): 2.41 (s, 3H), 9.63 (s, 1H), 13.6 (s, 1H, br). 13C NMR
(50 MHz, (CD3)2CO, Me4Si, δ ppm): 11.9, 95.8, 100.6, 112.1,
112.9, 135.4, 144.6, 178.2. HRMS: calcd for C8H5N3O [M+],
159.0433; found, 159.0429.
1
(34000). H NMR (500 MHz, CDCl3, Me4Si, δ ppm): -2.04 (s,
General Procedure for the Synthesis of Porphyrins: To a
vigorously stirred solution of TFA (7.7 mL, 0.1 mol) in chloroform
(100 mL), solutions of tripyrrane (1 mmol) in chloroform (10 mL)
and aldehyde (1 mmol) in chloroform (10 mL) were simultaneously
added during 15 min. Then the reaction was stirred for 10 min,
and the solution of DDQ (227 mg, 1 mmol) in toluene (2 mL) was
added. After an additional 10 min of stirring, the reaction mixture
was quenched using triethanoloamine (15.2 g, 0.1 mol). The reaction
mixture was then cooled for 20 min in the refrigerator and filtered.
Black crystals of salt (triethanoloammonium trifluoroacetate) were
washed with ethyl acetate until they became pale green and solvent
was evaporated. The purification details are described for each case
as follows.
2H), 1.15 (t, 6H, J ) 7.1 Hz), 1.79 (s, 6H), 1.89 (s, 6H), 2.24
(s,3H), 2.59 (s, 6H), 3.79 (d, 2H, J ) 9.6 Hz), 4.07 (d, 2H, J ) 9.6
Hz), 4.25 (q, 4H, J ) 7.1 Hz), 7.22 (s, 2H), 7.23 (s, 2H), 8.29 (s,
2H), 8.61 (d, 2H, J ) 4.5 Hz), 8.82 (d, 2H, J ) 4.5 Hz), 9.09 (s,
2H). HRMS (ESI): calcd for C47H50N5O4 [M + H+], 748.3857;
found, 748.3848.
Chlorin 43. Dry column vacuum chromatography (silica, CH2-
Cl2/MeOH, 95:5) afforded the product as a purple solid which was
recrystallized from CH2Cl2/hexane (28 mg, 42%): Rf ) 0.63 (silica,
CH2Cl2/MeOH 95:5). λmax (toluene)/nm: 413 (ꢀ/dm3 mol-1 cm-1
116000), 507 (10300), 587 (2840), and 637 (22700). 1H NMR (500
MHz, (CD3)2SO, Me4Si, 100 °C, δ ppm): -2.3 (s, 2H), 1.16 (t,
6H, J ) 7.1 Hz), 2.19 (s, 3H), 3.77 (d, 2H, J ) 10.0 Hz), 4.08 (d,
2H, J ) 10.0 Hz), 4.30-4.22 (m, 4H), 7.84 (ddd, 2H, J1 ) 0.7 Hz,
J2 ) 5.0 Hz, J3 ) 7.7 Hz), 8.44 (s,2H), 8.52 (m, 2H), 8.79 (d, 2H,
J ) 4.7 Hz), 9.02 (dd, 2H, J1 ) 1.6 Hz, J2 ) 5.0 Hz), 9.25 (d, 2H,
J ) 4.7 Hz), 9.27 (dd, 2H, J1 ) 0.7 Hz, J2 ) 2.2 Hz), 9.38 (s, 2H).
HRMS (ESI): calcd for C39H36N7O4 [M + H+], 666.2823; found,
666.2824.
Chlorin 48. Dry column vacuum chromatography (silica, toluene/
ethyl acetate, 9:1) afforded the product as a purple-green solid which
was recrystallized from CH2Cl2/hexane (4.5 mg, 7%): Rf ) 0.38
(silica, toluene/ethyl acetate 4:1). λmax (toluene)/nm: 400 (ꢀ/dm3
mol-1 cm-1 128000), 501 (12900), 599 (5510), and 654 (40000).
1H NMR (500 MHz, CDCl3, Me4Si, δ ppm): -2.97 (s, 2H), 1.14
(t, 6H, J ) 7.1 Hz), 2.26 (s, 3H), 3.82 (d, 2H, J ) 10.1 Hz), 4.13
(d, 2H, J ) 10.1 Hz), 4.27 (q, 4H, J ) 7.1 Hz), 9.18 (dd, 2H, J1
) 1.5 Hz, J2 ) 4 Hz), 9.45 (s, 2H), 9.47 (m, 2H), 9.73 (m, 2H).
HRMS (ESI): calcd for C31H28F6N5O4 [M + H+], 648.2040; found,
648.2065.
10,15-Dimesityl-22,24-dihydro-porphine-3,4-dicarboxylic Acid
Diethyl Ester (31). Dry column vacuum chromatography (silica,
toluene/ethyl acetate, 99:1) afforded product as a purple solid which
was recrystallized from CH2Cl2/hexane (207 mg, 30%): Rf ) 0.73
(silica, toluene/ethyl acetate 9:1). λmax (toluene)/nm: 421 (ꢀ/dm3
mol-1 cm-1 210000), 515 (13000), 553 (6000), 596 (4300), and
1
651 (4100). H NMR (500 MHz, CDCl3, Me4Si, δ ppm): -2.96
(s, 2H), 1.73 (t, 6H, J ) 7.2 Hz), 1.84 (s, 12H), 2.63 (s, 6H), 4.90
(q, 4H, J ) 7.2 Hz), 7.28 (s, 4H), 8.60 (s, 2H), 8.87 (d, 2H, J )
2.7 Hz), 9.40 (d, 2H, J ) 2.7 Hz), 10.70 (s, 2H). HRMS (ESI):
calcd for C44H43N4O4 [M + H+], 691.3279; found, 691.3250. Anal.
Calcd for C44H42N4O4: C, 76.50, H, 6.13, N, 8.11; Found: C, 76.25,
H, 6.32, N, 8.01.
10,15-Bis(3-pyridyl)-22,24-dihydro-porphine-3,4-dicarboxy-
lic Acid Diethyl Ester (33). Dry column vacuum chromatography
(silica, CH2Cl2/MeOH, 98:2) afforded the product as a purple solid
which was recrystallized from CH2Cl2/hexane (103 mg, 17%): Rf
) 0.65 (silica, CH2Cl2/MeOH 95:5). λmax (toluene)/nm: 422 (ꢀ/
dm3 mol-1 cm-1 199000), 515 (12200), 553 (5270), 594 (4270),
and 649 (3820). 1H NMR (200 MHz, CDCl3, Me4Si, δ ppm): -3.23
(s, 2H), 1.75 (t, 6H, J ) 7.1 Hz), 4.93 (q, 4H, J ) 7.1 Hz), 7.77-
7.83 (m, 2H), 8.51 (s, 2H), 8.79 (s, 2H), 9.03 (d, 2H, J ) 4.7 Hz),
9.10 (dd, 2H, J ) 4.9 Hz), 9.47 (s, 2H), 9.55 (d, 2H, J ) 4.7 Hz),
10.83 (s, 2H). HRMS (EI): calcd for C36H29N6O4 [M + H+],
609.2245; found, 609.2250.
10,15-Bis(trifluoromethyl)-22,24-dihydro-porphine-3,4-dicar-
boxylic Acid Diethyl Ester (38). Dry column vacuum chroma-
tography (silica, toluene/ethyl acetate, 99:1) afforded product as
purple solid which was recrystallized from CH2Cl2/hexane (41 mg,
7%): Rf ) 0.5 (silica, toluene/ethyl acetate 9:1). λmax (toluene)/
nm: 413 (ꢀ/dm3 mol-1 cm-1 106000), 515 (6660), 551 (10700),
Acknowledgment. This work was funded by the Ministry
of Scientific Research and Information Technology (Project 3
T09A 12429) and the Volkswagen Foundation. We thank
Mitsubishi Gas Chemical Company for the generous gift of
mesitaldehyde. We thank Mr. Maciej Rogacki for help.
Supporting Information Available: Description of unsuccessful
attempts to synthesize 2,5-diformyl-3,4-dicyano-pyrrole; 1H NMR
spectra for compounds 18, 27, 29, 30, 32, 33, 35, 37-48; 13C NMR
spectra for compounds 18 and 27 as well as UV-vis spectra of
chlorins 41, 45, 47, and 48. This material is available free of charge
via the Internet at http://pubs.acs.org.
JO060545X
5950 J. Org. Chem., Vol. 71, No. 16, 2006