The first direct synthesis of corroies from pyrrole

Full text of DOI:10.1002/(SICI)1521-3773(19990517)38:10<1427::AID-ANIE1427>3.0.CO;2-1

COMMUNICATIONS
Devadoss, P. Bharati, J. S. Moore, J. Am. Chem. Soc. 1996, 118, 9635 ±
9644; f) P.-W. Wang, Y.-J. Liu, C. Devadoss, P. Bharathi, J. S. Moore,
Adv. Mater. 1996, 3, 237 ± 241; g) G. M. Stewart, M. A. Fox, J. Am.
Chem. Soc. 1996, 118, 4354 ± 4360.
Ar:
F
F
F
F
TpFPPH2
Ar TdFPPH2
TdCPPH2
1
2
3
Ar
Ar
[5] a) G. McDermott, S. M. Prince, A. A. Freer, A. M. Hawthornthwaite-
Lawless, M. Z. Papiz, R. J. Cogdell, N. W. Isaacs, Nature 1995, 374,
517 ± 521; b) W. Kühlbrandt, Nature 1995, 374, 497 ± 498.
[6] a) T. Förster, Z. Naturforsch. A 1949, 4, 319 ± 327; b) T. Förster,
Discuss. Faraday Soc. 1959, 27, 7 ± 17; c) L. Stryer, Ann. Rev. Biochem.
1978, 47, 819 ± 846; d) B. Wieb Van Der Meer, G. Coker III, S.-Y.
Simon Chen, Resonance Energy Transfer, Theory and Data, VCH,
Weinheim, 1994.
[7] a) J. Bourson, J. Mugnier, B. Valeur, Chem. Phys. Lett. 1982, 92, 430 ±
432; b) J. Mugnier, J. Pouget, J. Bourson, B. Valeur, J. Lumin. 1985, 33,
273 ± 300; c) B. Valeur in Fluorescent Biomolecules, Methodologies
and Applications (Eds.: D. M. Jameson, G. D. Reinhart), Plenum, New
York, 1989, pp. 269 ± 303; d) B. Valeur in Molecular Luminescence
Spectroscopy, Part 3, Vol. 77 (Ed.: S. G. Schulman), Wiley, New York,
1993, chap. 2.
F
F
N
NH
N
NH
NH
Ar
Ar
HN
N
HN
F
Cl
Ar
Ar
Cl
Scheme 1. Schematic representation of tetraarylporphyrins TpFPPH2,
TdFPPH2, and TdCPPH2 (left) and the corresponding triarylcorroles
1 ± 3 (right).
[8] For the optical properties of commercial dyes see the Kodak or
Molecular Probes catalogs.
metal complexes. The reason for that is almost certainly the
lack of obvious procedures for their preparation. Thus, in spite
of the significant progress in corrole synthesis, even the most
simple procedures reported to date require starting materials
that are not commercially available.[5] Similarly, the first
synthesis of a meso-substituted corroleÐmeso-aryl substitu-
tion is most likely a prerequisite for the utilization of
metallocorroles in catalysisÐappeared only in 1993.[6]
Â
[9] a) C. J. Hawker, J. M. J. Frechet, J. Chem. Soc. Chem. Commun. 1990,
Â
1010 ± 1013; b) C. J. Hawker, J. M. J. Frechet, J. Am. Chem. Soc. 1990,
112, 7638 ± 7647.
[10] T. L. Tyler, J. E. Hanson, Polym. Mater. Sci. Eng. 1995, 73, 356 ± 357.
[11] Synthesis, detailed characterization, and quantitative photophysical
characteristics of these molecules will be described shortly: S. L. Gilat,
Â
A. Adronov, J. M. J. Frechet, unpublished results.
[12] L. Stryer, R. P. Haugland, Proc. Natl. Acad. Sci. USA 1967, 58, 719 ±
726.
Owing to our ongoing interest in the synthesis of porphyrins
and their core-modified analogues,[7] we explored a new
approach, the solvent-free condensation of pyrrole and
aldehydes. The research goal was the development of a
simple synthetic methodology for the preparation of porphyr-
ins, driven by the anticipation that the rather unusual reaction
conditions might also lead to the formation of some porphyrin
isomers. We have focused on two aldehydes: benzaldehyde, as
the prototype of other meso-aryl-substituted porphyrins, and
perfluorobenzaldehyde, because the metal complexes of the
corresponding porphyrin (TpFPPH2, Scheme 1) are among
the most efficient oxygenation catalysts. The reaction con-
ditions were very simple indeed, consisting of mixing pyrrole
and the aldehyde in equimolar quantities on a solid support,
heating to 1008C for four hours, oxidation with 2,3-dichloro-
5,6-dicyano-1,4-benzoquinone (DDQ), and chromatographic
separation. The reaction vessel was open to air, which not only
simplifies the procedure but was also found to be required.
With benzaldehyde as reactant, tetraphenylporphyrin (TPPH2)
was obtained in 5 ± 8% yield, depending on the solid support
(see Experimental Section).
[13] M. Beggren, A. Dodabalapur, R. E. Slusher, Z. Bao, Nature 1997, 389,
466 ± 469.
The First Direct Synthesis of Corroles from
Pyrrole**
Zeev Gross,* Nitsa Galili, and Irena Saltsman
Core-modified porphyrins have been receiving increased
attention in recent years,[1] mainly because of their superiority
over porphyrins in many applications, most notably as agents
for photodynamic therapy.[2] Some of these macrocycles form
complexes with transition metals, which due to the alteration
of the ligand structure have very different properties than the
analogous metalloporphyrins. The most interesting com-
pounds in this regard are corroles, whose skeletons are
contracted by one carbon atom with respect to porphyrins
(Scheme 1).[3] In contrast to the diprotonic porphyrins,
corroles act as tetradentate trianionic ligands toward metal
ions. The most remarkable feature of corroles is the stabiliza-
tion of unusually high oxidation states, such as iron(iv),
cobalt(iv), and cobalt(v).[4] Surprisingly, however, there is no
reported application in any field for either corroles or their
When the same reaction conditions (best with basic alumina
as solid support) were applied for perfluorobenzaldehyde,
very different results were observed. Only traces of the
corresponding porphyrin (TpFPPH2) were obtained, accom-
panied by a compound (1, isolated in 11% yield) which
showed a strong porphyrin-like fluorescence and an electronic
spectrum similar to that of TpFPPH2 (Figure 1). However,
[*] Dr. Z. Gross, Dr. N. Galili, Dr. I. Saltsman
Department of Chemistry
1
while the H NMR spectrum of TpFPPH2 consists of two
Technion ± Israel Institute of Technology
Haifa 32000 (Israel)
singlets in a ratio of 8:2 at d ˆ 8.91 (sharp) and 2.93 (broad)
for the b-pyrrole and the inner nitrogen hydrogen atoms,
respectively, in the spectrum of 1 three doublets (J ꢀ 4.5 Hz)
in the ratio of 1:1:2 were obtained in the pyrrole hydrogen
region (d ˆ 9.10, 8.75, and 8.57) together with an extremely
broad resonance at d ˆ 2.25 (Figure 1, inset). Similarly, the
Fax: (‡972)4-823-3735
E-mail: chr10zg@tx.technion.ac.il
[**] We acknowledge the partial support of this research by The R. and M.
Rochlin Research Fund (Z.G.) and The Center for Absorption in
Science, Ministry of Immigrant Absorption, State of Israel (I.S.).
Angew. Chem. Int. Ed. 1999, 38, No. 10
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1433-7851/99/3810-1427 $ 17.50+.50/0
1427

COMMUNICATIONS
replaced by nucleophilic substitution. This is demonstrated by
the reaction of 1 with ortho-pyridyllithium to provide
5,10,15-tris[4-(2-pyridyl)-2,3,5,6-tetrafluorophenyl]corrole (4;
Scheme 2; see Experimental Section). The reaction of 4 with
iodomethane leads to 5, which to our knowledge is the first ever
reported ionic corrole. The solubility of 5 in water is about
1
2 mgmL , and at least in the range determinable by UV/Vis
spectroscopy there is no sign of aggregation.
Figure 1. UV/Vis spectra of 1 (ÐÐ), and its protonated (- - - -); reaction
* * * *
with CF3CO2H) and deprotonated forms (
; reaction with Et3N). The
spectra were measured in CH2Cl2 at identical concentrations. Inset: The 1H
and 19F NMR spectra of 1 and TpFPPH2 in CDCl3.
signal for only one type of pentafluorophenyl group is present
in the 19F NMR of TpFPPH2, but two such groups in the ratio
of 2:1 are evident for 1. This data, together with the mass
spectrum of the compound (m/z 796) allowed the identifica-
tion of 1 as 5,10,15-tri(2,3,4,5,6-pentafluorophenyl)corrole,
the first b-pyrrole-unsubstituted corrole.
Because of the difference in reactivity between benzalde-
hyde and pentaflurobenzaldehydeÐformation of porphyrin
and corrole, respectivelyÐwe turned our attention to other
halogeno-substituted benzaldehydes, namely, the 3,5-di-
chloro, 2,6-dichloro, and 2,6-difluoro derivatives. Traces of
porphyrin and 6% of 5,10,15-tri(2,6-difluorophenyl)corrole
(2) were formed in the reaction of pyrrole and 2,6-difluor-
obenzaldehyde (equimolar amounts). However, no corrole
was formed with the two chlorinated aldehydes, and only trace
amounts of porphyrin appeared with 3,5-dichlorobenzalde-
hyde. The only isolated product (7%) from the reaction of
2,6-dichlorobenzaldehyde was the meso-aryl-a-benzyldipyr-
romethene.[8, 9] Reexamination of the reactions with penta-
fluoro- and 2,6-difluorobenzaldehyde revealed that analogous
compounds were also obtained as minor components. As
dipyrromethenes are the H2O elimination products of the
chain-growing polypyrroles which lead to both porphyrin and
corrole,[10] we anticipate that further optimization of the
synthetic procedure will allow the successful preparation of
additional triarylcorroles. Indeed, a simple change of the
pyrrole:aldehyde ratio to 2:1 allowed the isolation of 5,10,15-
tri(2,6-dichlorophenyl)corrole (3) in 1% yield.
The absence of electron-donating alkyl groups in the b-
pyrrole positions and the presence of the electron-withdraw-
ing pentafluorophenyl groups at the meso-carbon atoms in
corrole 1 lead to some unique properties. For example, the
known larger NH acidity of corroles relative to porphyrins is
brought to quite an extreme in 1. In ethanolic solutions the
dominant species is the monoanion of 1, and in organic
solvents 1 is deprotonated by weak bases. This is demon-
strated in Figure 1, which shows the changes in the electronic
spectrum of 1 upon its deprotonation by triethylamine.
Interestingly, the single imine-like nitrogen atom of 1 remains
basic; that is, it is protonated by a sufficiently strong acid
(Figure 1). These spectral changes are accompanied in both
cases by a color change from violet to green. Finally, another
peculiarity of 1 is that it may serve as the precursor of many
other corroles, since its para-fluoro substituents can easily be
Scheme 2. Synthesis of 4 and water-soluble 5 from 1.
To summarize, we report the first corrole synthesis to start
with commercially available reactants (pyrrole and aldehyde),
the first corroles without substituents at the b-pyrrole
positions, and the first water-soluble corrole. We have already
prepared the iron, cobalt, rhodium, and copper complexes of
1, whose full characterization and examination in various
applications is currently in progress. We also trust that the
facile synthesis described herein will soon make corrole
derivatives commercially available.[11]
Experimental Section
General procedure described for the synthesis of TPPH2: The solid
absorbant (florisil, silica, or alumina; 1 g) was mixed in a 50-mL flask with a
solution of benzaldehyde (1.06 g, 10 mmol) and pyrrole (0.67 g, 10 mmol)
in CH2Cl2 (2 mL), and the solvent was distilled at normal pressure. The
condenser was removed, and the solid mixture was heated to 1008C, upon
which the color changed to black in 5 ± 10 min. After heating of the mixture
for 4 h, the solid support was washed with CH2Cl2 (50 mL), DDQ (1.13 mg,
5 mmol) was added, and the product was obtained by chromatography on
basic alumina with hexane/CH2Cl2 (2.5/1) as eluent. With florisil, silica,
neutral alumina, and basic alumina as solid supports the isolated chemical
yields of TPPH2 were 5, 5, 7, and 8%, respectively. When the first step of the
reaction was carried out under an inert atmosphere, neither porphyrins or
corroles were obtained.
1: Under the same reaction conditions as for TPPH2, with the solid
absorbant (0.5 g), pentafluorobenzaldehyde (0.49 g, 2.5 mmol), and pyrrole
(0.17 g, 2.5 mmol), traces of TpFPPH2 and significant amounts of 1 were
obtained (Rf (hexane/CH2Cl2 3/1) ˆ 0.52 and 0.34, respectively). The yields
of isolated 1 (after crystallization from CH2Cl2/hexane) were 4%, traces,
11%, and 11%, with florisil, silica, neutral alumina, and basic alumina as
solid support, respectively. Identical chemical yields were obtained in a
reaction with half the amount of solid support; that is, 55 mg of 1 were
isolated from the reaction supported by 0.25 g of basic alumina. UV/Vis
(CH2Cl2): lmax (e  10 3) ˆ 408 (114.0), 560 (17.6), 602 nm (9.3); H NMR
1
(CDCl3): d ˆ 9.10 (d, J ˆ 4.4 Hz, 2H), 8.75 (d, J ˆ 4.4 Hz, 2H), 8.57 (d, J ˆ
4.4 Hz, 4H), 2.25 (brs, 3H); 19F NMR (CDCl3): d ˆ 137.55 (dd, 1J ˆ
24.2, 2J ˆ 8.1 Hz, 2F), 138.14 (dd, 1J ˆ 23.03, 2J ˆ 6.9 Hz, 4F), 152.52 (t,
1
2
J ˆ 19.6 Hz, 2F), 153.10 (t, J ˆ 20.7 Hz, 1F), 161.78 (dt, J ˆ 24.2, J ˆ
1428
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1433-7851/99/3810-1428 $ 17.50+.50/0
Angew. Chem. Int. Ed. 1999, 38, No. 10

COMMUNICATIONS
8.1 Hz, 4F),
797.0854 (calcd for C37H12N4F15: 797.0822).
162.35 (dt, 1J ˆ 23.03, 2J ˆ 6.9 Hz, 2F); HR-MS: m/z:
Inorg. Chem. 1996, 35, 7260; d) Z. Gross, L. Kaustov, Tetrahedron Lett.
1995, 36, 3735; e) Z. Gross, I. Toledano, J. Org. Chem. 1994, 59, 8312.
[8] For a review on dipyrrolic intermediates, see J. B. Paine III in The
Porphyrins, Vol. I (Ed.: D. Dolphin), Academic Press, New York,
1979, Chapter 4.
[9] For the zinc complex of the same compound, see C. L. Hill, M. M.
Williamson, J. Chem. Soc. Chem. Commun. 1985, 1228.
[10] J. B. Kim, A. D. Adler, F. R. Longo in The Porphyrins, Vol. I (Ed.: D.
Dolphin), Academic Press, New York, 1979, pp. 90 ± 96.
[11] Z. Gross (Technion), Pending Israel Patent application, IL-A 126426,
1998.
2: UV/Vis (CH2Cl2): lmax (e  10 3): 406 (118.6), 562 (20.3), 602 nm (11.9);
1H NMR (CDCl3): d ˆ 8.99 (d, J ˆ 4.3 Hz, 2H), 8.71 (d, J ˆ 4.3 Hz, 2H),
8.52 (t, J ˆ 4.3 Hz, 4H), 7.72 (m, 3H), 7.33 (m, 6H), 2.1 (brs, 3H); 19F
NMR (CDCl3): d ˆ 109.32 (t, J ˆ 6.4 Hz, 2F), 109.75 (t, J ˆ 6.4 Hz, 4F);
HR-MS: m/z: 635.1660 (calcd for C37H21N4F6: 635.1670).
3: UV/Vis (CH2Cl2): lmax (e  10 3): 408 (106.1), 422 (86.6), 560 (16.7),
1
604 nm (9.3); H NMR (CDCl3): d ˆ 8.91 (d, J ˆ 4.2 Hz, 2H), 8.49 (d, J ˆ
4.8 Hz, 2H), 8.35 (d J ˆ 4.6 Hz, 4H), 7.72 (m, 3H), 7.33 (m, 6H), 1.7 (brs,
3H); HR-MS: m/z: 729.9810 (calcd for C37H20N4Cl6: 729.9819).
4: A 1.6m nBuLi solution (0.42 mL, 0.7 mmol) was added to a stirred
solution of 2-bromopyridine (0.054 mL, 0.56 mmol) in dry THF (6 mL)
under an argon atmosphere at 788C, at such a rate that the temperature
of the reaction mixture did not exceed 708C. After the addition was
complete, the reaction mixture was stirred for 1 h at 788C, resulting an a
clear yellow solution. Next, a solution of 1 (0.03 g, 0.038 mmol) in dry THF
(6 mL) was added dropwise. The mixture was stirred for 1 h at 788C and
then hydrolyzed with saturated aqueous bicarbonate solution. The layers
were separated, the aqueous layer was washed with diethyl ether, and the
combined diethyl ether extracts were dried and evaporated to yield a solid
residue. The product was purified by column chromatography on silica gel
(EtOAc/hexane 1/1) and recrystallized from CH2Cl2/hexane to provide
13 mg (35% yield) of pure 4 as a violet solid. UV/Vis (CH2Cl2): lmax (e Â
‡
Regioselective Reduction of NAD Models
‡
with [Cp*Rh(bpy)H] : Structure ± Activity
Relationships and Mechanistic Aspects in the
Formation of the 1,4-NADH Derivatives**
H. Christine Lo, Olivier Buriez, John B. Kerr, and
Richard H. Fish*
1
10 3): 414 (111.6), 564 (18.4), 606 nm (sh); H NMR (CDCl3): d ˆ 9.12 (d,
J ˆ 3.9 Hz, 2H), 8.93 (m, 5H), 8.73 (d, J ˆ 4.88 Hz, 2H), 8.66 (d, J ˆ
3.91 Hz, 2H), 8.00 (dt, 1J ˆ 7.81, 2J ˆ 1.95 Hz, 3H), 7.84 (brd, J ˆ 7.81 Hz,
3H), 7.51 (dt, 1J ˆ 6.84, 2J ˆ 1.95 Hz, 3H), 2.02 (brs, 3H); 19F NMR
(CDCl3): d ˆ 138.19 (q, J ˆ 23.79 Hz, 2F), 138.81 (q, J ˆ 23.79 Hz, 4F),
144.11 (q, J ˆ 23.79 Hz, 4F), 144.57 (q, J ˆ 23.79 Hz, 2F); HR-MS: m/z:
973.1910 (calcd for C52H23N7F12: 973.1823).
The interest in practical methods for the regeneration of the
co-enzyme 1,4-NADH, the reduced form of nicotinamide
adenine dinucleotide (NAD ), has continued to be high in the
‡
field of biocatalysis, where enzymatic reduction reactions are
important for the synthesis of chiral organic compounds.[1a, b]
Conversion of NAD into 1,4-NADH by enzymatic, chemical,
5: A mixture of 4 (11 mg, 11 mmol) and CH3I (0.8 mL, 13 mmol) in freshly
distilled DMF (2 mL) was heated to 708C for 3 h. After evaporation of the
solvent, the product was recrystallized from MeOH/diethyl ether to
provide 15.5 mg (98% yield) of 5 as a green solid. UV/Vis (MeOH): lmax
(e  10 3): 430 (76.2), 576 (10.9), 622 nm (17.8); 1H NMR ([D6]DMSO): d ˆ
9.49 (d, J ˆ 5.98 Hz, 3H), 9.16 (brm, 8H), 9.00 (t, J ˆ 8.54 Hz, 3H), 8.75 (t,
J ˆ 7.68 Hz, 3H), 8.51 (t, J ˆ 7.68 Hz, 3H), 4.68 (s, 3H), 4.65 (s, 6H); 19F
‡
photochemical, and electrochemical methods has been stud-
ied extensively in order to increase the rate of the regener-
ation, while maintaining the necessary high regioselectivity.
The regeneration is frequently the limiting step in the
eventual use of 1,4-NADH in enzymatic synthesis, particularly
for higher volume and more energy intensive processes.[1a, b]
In the search for higher rates and a more economical
regeneration process various transition metal hydrides have
been studied as catalysts for the regioselective reduction of
NMR ([D6]DMSO): d ˆ 137.26 (brm, 4F), 138.04 (brm, 6F), 138.60
‡
(brm, 2F); electron spray MS: m/z: 339.9 ([M
3I]/3, 100%).
Received: November 2, 1998 [Z12604IE]
German version: Angew. Chem. 1999, 111, 1530 ± 1533
‡
‡
NAD and NAD models to their corresponding 1,4-NADH
derivatives.[2a±g] In the most successful example, Steckhan and
Keywords: corroles ´ macrocycles ´ porphyrinoids ´ syn-
thetic methods
‡
co-workers have described the use of [Cp*Rh(bpy)(H)]
(Cp* ˆ pentamethylcyclopentadienyl, bpy ˆ 2,2'-bipyridyl),
‡
generated in situ, for the regiospecific reduction of NAD
[1] Recent reviews: a) J. L. Sessler, S. J. Weghorn in Expanded, Con-
tracted, & Isomeric Porphyrins, Pergamon, Oxford, 1997, pp. 1 ± 503;
b) A. Jasat, D. Dolphin, Chem. Rev. 1997, 97, 2267; c) E. Vogel, J.
Heterocycl. Chem. 1996, 33, 1461.
[2] a) J. L. Sessler, S. J. Weghorn in Expanded, Contracted, & Isomeric
Porphyrins, Pergamon, Oxford, 1997, pp. 429 ± 503; b) E. D. Sternberg,
D. Dolphin, C. Bruckner, Tetrahedron 1998, 54, 4151.
to 1,4-NADH,[2b] and then demonstrated the cofactor regen-
eration process in enzymatic, chiral reduction reactions.[3, 4]
‡
While the above mentioned reduction of NAD by
‡
[Cp*Rh(bpy)H] was shown to be regiospecific for
1,4-NADH,[2b, 5a] the full mechanistic details of this important
[3] S. Licoccia, R. Paolesse, Struct. Bond. 1995, 84, 71.
[4] a) E. Van Caemelbecke, S. Will, M. Autret, V. A. Adamian, J. Lex, J. P.
Gisselbrecht, M. Gross, E. Vogel, K. M. Kadish, Inorg. Chem. 1996,
35, 184; b) S. Will, J. Lex, E. Vogel, V. A. Adamian, E. Van Caemel-
becke, K. M. Kadish, Inorg. Chem. 1996, 35, 5577.
[5] a) R. Paolesse, S. Licoccia, G. Bandoli, A. Dolmella, T. Boschi, Inorg.
Chem. 1994, 33, 1171; b) S. Neya, K. Ohyama, N. Funasaki,
Tetrahedron Lett. 1997, 38, 4113.
[6] R. Paolesse, S. Licoccia, M. Fanciullo, E. Morgante, T. Boschi, Inorg.
Chim. Acta 1993, 203, 107.
[7] a) Z. Gross, I. Saltsman, R. P. Pandian, C. M. Barzilay, Tetrahedron
Lett. 1997, 38, 2383; b) S. Ini, M. Kapon, S. Cohen, Z. Gross,
Tetrahedron: Asymmetry 1996, 7, 659; c) Z. Gross, A. Mahammed,
[*] Dr. R. H. Fish, Dr. H. C. Lo, Dr. O. Buriez, Dr. J. B. Kerr
Lawrence Berkeley National Laboratory
University of California
Berkeley, CA 94720 (USA)
Fax: (‡1)510-486-7303
E-mail: rhfish@lbl.gov
[**] Bioorganometallic Chemistry Part 11. We gratefully acknowledge
Department of Energy funding from the Advanced Energy Projects
and Technology Research Division, Office of Computational and
Technology Research (DE AC03 ± 76SF00098). Part 10: S. Ogo, S.
Nakamura, H. Chen, K. Isobe, Y. Watanabe, R. H. Fish, J. Org. Chem.
1998, 63, 7151.
Angew. Chem. Int. Ed. 1999, 38, No. 10
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1433-7851/99/3810-1429 $ 17.50+.50/0
1429