Microwave-assisted synthesis of corroles

Article abstract of DOI:10.1016/S0040-4039(02)02792-2

The recently developed Gross's method for the synthesis of corroles has been modified and successfully applied to the preparation of new free base tris-aryl- and tris-pyrimidyl-corroles using solvent-free conditions and microwave irradiation. Compared to

Full text of DOI:10.1016/S0040-4039(02)02792-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1207–1210
Microwave-assisted synthesis of corroles
James P. Collman* and Richard A. Decr e´ au
Stanford University, Department of Chemistry, Stanford, CA 94305-5080, USA
Received 18 September 2002; revised 7 December 2002; accepted 9 December 2002
Abstract—The recently developed Gross’s method for the synthesis of corroles has been modified and successfully applied to the
preparation of new free base tris-aryl- and tris-pyrimidyl-corroles using solvent-free conditions and microwave irradiation.
Compared to conventional heating, the microwave technique afforded an increase in corrole yields of ca. 30% and led to
noticeably cleaner reaction mixtures. It is demonstrated that short reaction times and high temperatures are required to afford
optimum yields. © 2003 Elsevier Science Ltd. All rights reserved.
1
19
Corroles are porphyrin analogues which lack one meso
and characterized by H, F NMR, HRMS and UV–
9
carbon bridge. Recently several synthetic methods have
vis. In Table 1, yields of 1–8 are compared either when
1
a–d,2a–c
been reported
which allow corroles to be
the reaction was carried on under the precise conditions
described by Gross using conventional heating (entry
H) or after using microwave irradiation (entry MW).
Entry c gives the resulting increase of yield.
obtained in reasonable quantities and to be used for a
3
a–g
wide range of applications as catalysts.
The solvent-
free method developed by Gross on solid support is
quite straightforward, it has been developed for elec-
tron-withdrawing aldehydes and in fact only tris-penta-
fluorophenylcorrole 4 was obtained in 8–11% yield after
Previously Gross showed that 4 would be obtained in
–11% yield after 1 min–4 h under conventional heating
8
1
a,b
heating.
We report here the application of this
1a,b
at 100°C;
whereas we show here that 13–15% of
methodology to other electron-withdrawing benzyl-
aldehydes as well as to one pyrimidyl-aldehyde. Fur-
thermore, several reports have shown that using
microwave irradiation (MW) instead of conventional
heating (H) raises yields, enhances reaction rate,
reduces thermal degradation by-products for a wide
isolated yield are obtained after only 2 min of
microwave irradiation (ca. 47% increase). As pointed
out by Gross, the reproducibility of this solvent-free
synthesis on alumina in terms of isolated yields varies
because of variation in the effectiveness of stirring.
With a series of less reactive fluorinated benzaldehydes
we showed that a similar range of yield increase is
observed. Indeed, tetrafluorobenzaldehyde led to cor-
role 3 which went from 10.5% (H) to 13% (MW) (ca.
4
range of organic reactions including cyclocondensation
5
a,b
reactions leading to tetraarylporphyrins
phthalocyanines.
and
6
a–d
In an effort to improve corrole
yields, conventional heating was replaced by microwave
irradiation. Indeed the microwave dielectric heating
effect is an efficient heating source for thermally
demanding organic reactions, such as corrole synthesis.
Herein we report the first method to synthesize corroles
under microwave irradiation, and the first application
of Gross’s method to other kinds of aldehydes.
2
8% increase). However other aldehyde precursors did
not react spontaneously with pyrrole at room tempera-
ture and require thermal or microwave activation. Thus
trifluorobenzaldehyde led to corrole 2 in 6.7% (H)
which increased to 8.9% (MW, ca. 33% increase). An
experiment led to 12% which corresponds to 80% yield
increase. Similarly p-fluorobenzaldehyde led to corrole
1 in 1.5% (H) and reached 2.0–2.5% (MW, 30%
increase). We accounted for the observation Gross pre-
viously made on two precursors, 2,6-difluorobenzalde-
New corroles 1–3 and 5–7 whose structures are depicted
in Figure 1, and known ones like tris-pentafluorophenyl
1
a,b
1d
corrole 4
and tris-pyridyl-corrole 8 have been syn-
7,8
thesized following two different heating procedures
1b
hyde and pentafluorobenzaldehyde, by examining a
range of aldehydes of different degrees of fluorination
here, that the yield of corroles 1–4 is dependent upon
the percentage of electron-withdrawing F atoms of the
aldehydes (Table 1).
Keywords: corrole synthesis; microwave irradiation.
*
Corresponding author. Tel.: (650) 725-0283; fax: (650) 725-0259;
e-mail: jpc@stanford.edu
0
040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02792-2

1
208
J. P. Collman, R. A. Decr e´ au / Tetrahedron Letters 44 (2003) 1207–1210
Using 2,6-dimethyl-pyrimidyl-carbaldehyde, prepared
1
1a,b
as reported in the literature,
corrole 7, a close
analogue of the 2,6-dichloropyrimidine-bis-(2,6-
1
2
dichlorophenyl)-corrole reported by Asokan,
obtained in 4.5% yield (H) and 6.2% yield (MW, ca.
7% increase). Unfortunately with 4-pyridyl-carbalde-
was
3
hyde, the yields of corrole 8 obtained remained very
small 2.2% (H) versus 3.3% (MW) despite a greater
increase in yield than with the other precursors (50%
increase). Consequently the modified Rothemund
method developed by Paolesse using a reflux in acetic
acid appears to be the most appropriate method with a
1c,d
9
% reported yield.
With trimesitaldehyde no trace of
trimesityl-corrole was obtained (data not shown).
Replacing thermal heating by microwave irradiation led
to a noticeable increase in yield of at least 30% in most
cases. Similarly we noticed that after MW use, the
amount of impurities whose R values are close to
f
corrole’s dropped significantly. For example, the
pentapyrrotetramethene side1 product previously
b
reported in the synthesis of 4 was isolated from the
synthesis of 6 in smaller yield after MW heating (4%)
1
3
than after conventional heating (8%). The same obser-
vation was made for the syntheses of corroles 1–4, and
5
. In terms of reaction time, Gross showed that using
pentafluorobenzaldehyde, corrole
4
was obtained
1b
within 1 min. We show here that such short reaction
times are feasible using the tetrafluorinated and pyrimy-
dyl precursors which underwent an exothermic reaction
with pyrrole at room temperature affording corroles 3
and 7. But in any case heating is required to afford the
optimum yields of corroles 3, 4 and 7. On the other
hand, less reactive aldehydes leading to corroles 1–2,
5–6 and 8 required either thermal or microwave assis-
tance to react with pyrrole. It is interesting to point out
that the reaction of 2,6-dichloro-pyrimidine carbalde-
hyde with pyrrole is very exothermic and afforded
compound 7 in reasonable yields (4.5% (H) versus 6.2%
(MW)) while an earlier work reported that no corrole
was formed with 2,6-dichlorobenzaldehyde even after
Figure 1.
Another series of aldehydes was examined comparing
H versus MW wit h two trifluoromethyl-containing benz-
aldehydes. The o-monosubstituted derivatives afforded
the corrole 5 in 9.1% yield (H) and 12% yield (MW, ca.
3
1% increase), while 3,5-bis-trifluoromethyl-benzalde-
1a
hyde afforded corrole 6 in 8.5% yield (H) and 10.2–11%
yield (MW, ca. 20–29% increase). Ghosh reported ear-
heating.
10
lier 7–10% yield with the para analog of 5. This
demonstrates that the trifluoromethyl moiety exerts a
sufficient activating inductive effect to make the alde-
hyde reactive enough towards pyrrole to afford corroles
The temperature and reaction time were checked on the
synthesis of corrole 6 (Table 2). When the temperature
was raised to 200°C at a reaction time of 2 min, the
yields increased to 12.8% (ca. 45% increase), while
higher temperature (250°C) led to 7.5% of 6 only. When
the reaction time was prolonged to 20 min a loss of
20% of the overall yield was observed at 120°C. This
5
–6 in reasonable yields. However these yields did not
increase with the number of CF groups, contrary to
3
what is observed with F atoms.
Table 1. Yields of corroles 1–8 after conventional (H) or microwave (MW) heating at 120°C. Entry (c) corresponding yield
increase (MW versus H)
Entry
C
O
R
R
O
L
E
S
1
2
3
4
5
6
7
8
H
MW
c
1–1.5
2.5
40
6.7
8.9 (12)
33 (80)
10.5
14
28
8–11
13–15
36
9.1
12
31
5.7–9.5
10.2–11.2
23
4.5
6.2
37
2
3.3
25–50

J. P. Collman, R. A. Decr e´ au / Tetrahedron Letters 44 (2003) 1207–1210
1209
Table 2. Effects of temperature and reaction times on the
yields of corrole 6 (MW heating)
Blaser, D.; Boese, R.; Goldberg, I. Org. Lett. 1999, 1,
99–602; (c) Paolesse, R.; Jaquinod, L.; Nurco, D. J.;
Mini, S.; Sagone, F. Chem. Commun. 1999, 1307–1308;
d) Paolesse, R.; Nardis, S.; Sagone, F.; Khoury, R. G. J.
5
Temperature (°C)
(
Org. Chem. 2001, 66, 550–556.
. (a) Gryco, D. T. Eur. J. Org. Chem. 2002, 11, 1735–1743;
Time (min)
2
120
10.2
9.0
8.0
160
200
12.8
–
4.2
250
2
(
(
b) Gryko, D. T. Chem. Commun. 2000, 22, 2243–2244;
c) Gryko, D. T.; Jadach, K. J. Org. Chem. 2001, 66,
10.3
8.3
–
7.5
–
1.5
1
2
0
0
4267–4275.
3
. (a) Gross, Z.; Golubkov, G.; Simkhovich, L. Angew.
Chem., Int. Ed. 2000, 39, 4045–4047; (b) Gross, Z.;
Simkhovich, L.; Galili, N. Chem. Commun. 1999, 39,
599–600; (c) Simkhovich, L.; Gross, Z. Tetrahedron Lett.
2001, 42, 8089–8092; (d) Simkhovich, L.; Mahammed, A.;
Goldberg, I.; Gross, Z. Chem. Eur. J. 2001, 7, 1041–1055;
Table 3. Yields of corrole 6 under conventional (H) or
microwave (MW) heating at 120°C, after some modifica-
tions in the general procedure. Entry (a) as reported in
Table 1. (b) NaCl introduction. (c) 3 equiv. acetic acid
(
e) Jerome, F.; Gros, C. P.; Tardieux, C.; Barbe, J.-M.;
Entry
A
B
c
Guilard, R. New J. Chem. 1998, 1327–1329; (f) Kadish,
K. M.; Ou, Z.; Shao, J.; Gros, C. P.; Barbe, J.-M.;
Jerome, F.; Bolze, F.; Burdet, F.; Guilard, F. Inorg.
Chem. 2002, 41, 3990–4005; (g) Aviezer, D.; Cotton, S.;
David, M.; Segev, A.; Khaselev, N.; Galili, N.; Gross, Z.;
Yayon, A. Cancer Res. 2000, 60, 2973–2980.
H
MW
5.7/9.5
10.2–11.2
8.5
10.5
7.4
6.7
drop is even more spectacular at 200°C (67% loss of the
overall yield) (a control experiment using a mixture of
corrole 6 and pyrrole over alumina did not show any
decomposition of 6 when heated (H or MW) for 20 min
at 200°C). We conclude that a short reaction time and
a high temperature are required to afford this corrole in
good yield.
4
5
. (a) Lidstrom, P.; Tierney, J.; Wathey, B.; Westman, J.
Tetrahedron 2001, 57, 9225–9283; (b) Caddick, S. Tetra-
hedron 1995, 38, 10403–10432; (c) Strauss, C. R.; Trainor,
R. W. Aust. J. Chem. 1995, 48, 1665–1692.
. (a) Petit, A.; Loupy, A.; Maillard, Ph.; Momenteau, M.
Synth. Commun. 1992, 22, 1137–1142; (b) Chauhan, S. M.
S.; Sahoo, B. B.; Srinivas, K. A. Synth. Commun. 2001,
3
1, 33–37.
Other modifications in the procedure were developed
for the synthesis of corrole 6 by introducing charges
which would likely make the system more prone to
activation by microwave radiation (Table 3). Neither
the addition of inorganic additives (entry b), known to
raise the corrole yield during the coupling of tetra-
pyrromethanes, nor the introduction of acid (entry c),
improved the yields. In the former case they remained
unchanged (8% (H), 10.5% (MW)) while in the later
they dramatically dropped to 6.7–7.4% when 3 equiv. of
acetic acid was added (entry c).
6. (a) Unguranesu, C. Synthesis 1999, 10, 1729–1730; (b)
Shaabani, A. J. Chem. Res. (S) 1998, 672–673; (c) Jung,
K. S.; Ro, J. Y.; Lee, J. Y.; Park, S. S. J. Materials Sci.
Lett. 2001, 20, 2203–2205; (d) Villemin, D.; Hammadi,
M.; Hachemi, M.; Bar, N. Molecules 2001, 6, 831–844.
7. General Procedures for compounds 1–8. Conventional
1
4
1a,b
procedure: (H) described by Gross
and modified as
follows: Aldehyde (1.871 mmol) was mixed with oven-
activated basic alumina (50 mg) in a one-neck-conical
flask, then fast introduction of pyrrole (3 equiv., 0.366 ul)
was carried out while the mixture was under violent
agitation. With reactive aldehydes the reaction mixture
turned brown within 30 s, and only then was the vessel
plunged into a preheated bath (120°C) and heated under
agitation for 20 min. Microwave procedure: (MW) the
reagents were mixed the same way as above in a Teflon
vessel (8.0×1.5 cm) sealed with a cap and microwave-
heated for 2–20 min at 120–200°C. The microwave
apparatus used was a Smith-Synthesizer (Single-mode
microwave heating; Power 300 W). The reaction mixture
was inserted in a conical Teflon vessel (50×15 mm; 5 mm
thick walls) screw-capped by a stopper tightly sealed with
a schrimper. The agitation was performed with special
stir-bars provided by Smith-Synthesizer Ltd. Workup:
After the dark oily mixture had cooled down to room
temperature and solidified; it was digested with portions
of dichloromethane (total volume ca. 200 ml) under
sonication. Powdered DDQ (0.5–0.6 equiv.) was then
added and the reaction mixture was stirred at room
temperature for an additional 15 min. Unless otherwise
stated, the solution was filtered through an alumina pad
(5×3 cm), then the solvent was evaporated and the
As a result, microwave irradiation used in the synthesis
of corrole constitutes an improvement in terms of yield,
and by-product formation, to the straightforward
method previously described by Gross for the corrole
synthesis.
Acknowledgements
We wish to thank Professor B. M. Trost and Personal
Chemistry for providing the PowerSmith microwave
reactor used in the synthesis of compounds 1–8. R.A.D.
thanks the French Foreign Ministry for Lavoisier
fellowship.
References
1
. (a) Gross, Z.; Galili, N.; Saltsman, I. Angew. Chem., Int.
Ed. 1999, 38, 1427–1429; (b) Gross, Z.; Galili, N.; Salts-
man, I.; Simkhovich, L.; Saltsman, I.; Botoshansky, M.;
residue submitted to chromatography on basic alumina
(10×3 cm) of low activity
8
using
a
hexane/

1
210
J. P. Collman, R. A. Decr e´ au / Tetrahedron Letters 44 (2003) 1207–1210
dichloromethane mixture (gradient elution) whose ratio
MHz, CDCl ) l (ppm): 8.90 (m, 2H), 8.37 (m, 2H), 8.26
3
was varied for each corrole. A second chromatography
afforded the desired corrole in high purity. For corroles
(m, 2H), 8.18 (m, 2H), 8.05 (m, 5H), 7.81 (m, 7H), −2.01
19
(brs, 3H). F NMR (500 MHz, CFCl ) l (ppm): −61.67
3
−
3
1
–8 traces of porphyrins were sometimes detected and
(m, F), −62.31 (m, F). UV–vis: (CH Cl ) u (m×10 ): 405
2 2
were neglected. This did not vary with the mode of
(91.6), 420 (68.6), 561 (11.5), 601 (5.9), 624 (1.07) nm.
heating (H or MW).
HR-MS (EI, m/z)=730.1765 (calcd for C H F N :
40
23
9
4
8
9
. Basic alumina of low activity was prepared by leaving
activated basic alumina for 1 week in a beaker covered
with alumina foil. Indeed we observed that activated
basic alumina was responsible for a significant loss of
corrole which remained stuck to the gel (ca. 0.5–1.2% of
the overall yield).
730.1779).
(e) 5,10,15-Tris(3,5-Bis(trifluoromethyl)phenyl) corrole
(6): Purified on silica gel 15×6 cm, gradient eluent hex-
ane/dichloromethane, 8.5:1.5 vol. Then recrystallized in
1
hexane. H NMR (400 MHz, CDCl ) l (ppm): 9.03 (d,
3
J=4.01 Hz, 2H), 8.81 (d, J=4.72 Hz, 2H), 8.76 (s, 4H),
8.64 (s, 2H), 8.53 (d, J=4.29 Hz, 2H), 8.50 (d, J=4.72
Hz, 2H), 8.30 (s, 1H), 8.25 (s, 2H), −2.30 (brs, 3H).
. Characterization of compounds:
1
19
(a) 5,10,15-Tris(4-fluorophenyl) corrole (1): H NMR
F
(
8
(
400 MHz, CDCl ) l (ppm): 8.87 (d, J=3.53 Hz, 2H),
NMR (500 MHz, CFCl ) l (ppm): −66.56 (s, 6F), −66.65
(s, 12F). UV–vis: (CH Cl ) u (m×10 ): 418 (140.8), 575
2 2
3
3
−
3
.80 (d, J=4.02 Hz, 2H), 8.58 (m, 4H), 8.29 (m, 4H), 8.10
19
d, J=3.02 Hz, 2H), 7.50 (m, 6H), −2.30 (brs, 3H).
F
(22.7), 614 (13.0), 648 (8.3) nm. HR-MS (EI, m/z)=
935.7621 (calcd for M+1 C43 : 935.6401). TLC
(alumina, hexane/dichloromethane, 3:1) R 0.74.
NMR (500 MHz, CFCl ) l (ppm): −119.40 (s, 2F),
−
H F N
21 18 4
3
−
3
119.62 (s, 1F). UV–vis: (CH Cl ) u (m×10 ): 414 (68.1),
f
2
2
5
72 (11.0), 613 (8.6), 623 (8.1), 630 (8.1), 648 (7.2) nm.
(f) 5,10,15-Tris(2,6-dichloropyrimidyl) corrole (7):
HR-MS (EI, m/z)=580.1875 (calcd for C H F N :
purified on alumina gel (17×3 cm) using dichloromethane
37
23
3
4
1
5
80.1895). TLC (alumina, hexane/dichloromethane, 6:2.5)
as eluent. H NMR (400 MHz, CDCl
) l (ppm): 9.20 (m,
3
R 0.24.
3H), 9.06 (d, J=4.39 Hz, 2H), 8.58 (d, J=4.78 Hz, 2H),
8.43 (d, J=4.29 Hz, 2H), 8.41 (d, J=4.83 Hz, 2H), −2.40
f
1
(b) 5,10,15-Tris(3,4,6-trifluorophenyl) corrole (2):
H
−
3
NMR (400 MHz, CDCl ) l (ppm): 9.03 (d, J=3.53 Hz,
(brs, 3H). UV–vis: (CH
(28.6), 562 sh (5.5), 569 (5.8), 609 (3.5) nm. HR-MS (EI ,
m/z)=737.9514 (calcd for M+2H 10:Cl
737.9691). TLC (alumina, dichloromethane) R 0.32.
(g) Compounds 4 and 8 were characterized by H,
NMR (compound 4) HRMS and UV–vis and were as
2
Cl ) u (m×10 ): 414 (31.0), 426
2
3
+
2
4
H), 8.80 (d, J=4.02 Hz, 2H), 8.58 (m, 4H), 7.94 (m,
H), 7.77 (m, 2H), −2.30 (brs, 3H). F NMR (500 MHz,
1
9
C
H
14
N
:
6
31
CFCl ) l (ppm): −141.20 (t, 2F), −140.20 (t, 1F), −166.00
f
3
−
3
1
19
(
(
6
m, 6F). UV–vis: (CH Cl ) u (m×10 ): 414 (99.7), 573
13.9), 612 (7.4), 647 (4.2) nm. HR-MS (EI, m/z)=
88.1311 (calcd for C H F N : 688.1309). TLC (alu-
F
2
2
1a,d
previously described.
3
7
17
9
4
mina, hexane/dichloromethane, 6:2.5) R 0.41.
10. Wasbotten, I. H.; Wondimagegn, T.; Ghosh, A. J. Am.
Chem. Soc. 2002, 124, 8104–8116.
f
1
(c) 5,10,15-Tris(2,3,5,6-tetrafluorophenyl) corrole (3): H
NMR (400 MHz, CDCl ) l (ppm): 9.07 (d, J=4.02 Hz,
11. (a) Klotzer, W.; Herberz, M. Monatsch. Chem. 1965, 96,
1567–1572; (b) Bredereck, H.; Simchen, G.; Santo, A. A.
Chem. Ber. 1967, 100, 1344–1352.
3
2
H), 8.78 (d, J=4.77 Hz, 2H), 8.59 (d, J=4.78 Hz, 2H),
.56 (d, J=3.90 Hz, 2H), 7.53 (m, 3H), −2.10 (brs, 3H).
8
19
F NMR (500 MHz, CFCl ) l (ppm): −142.12 (m, 2F),
12. Asokan, C. V.; Smeets, S.; Dehaen, W. Tetrahedron Lett.
3
−
142.75 (m, 8F), −143.22 (m, 2F). UV–vis: (CH Cl ) u
2001, 42, 4483–4485.
2 2
−
3
(
m×10 ): 408 (118.6), 418 sh (104.9), 562 (19.7), 605 (9.8)
13. Pentapyrrotetramethene isolated from the synthesis of 6.
1
nm. HR-MS (EI m/z)=742.2176 (calcd for C H F N :
H NMR (400 MHz, CDCl ) l (ppm)12.52 (brs, 3H),
3
37
14 12
4
7
42.1027). TLC (alumina, hexane/dichloromethane, 6:2.5)
7.91 (m, 8H), 7.71 (brs, 4H), 6.76 (s, 2H), 6.52 (s, H), 6.49
(m, 2H), 6.39 (m, 2H), 6.06 (m, 2H), 6.00 (m, 2H). MS
R 0.30.
f
+
−
(d) 5,10,15-Tris(2-trifluoromethyl-phenyl) corrole (5):
(ESI , m/z) 1226.3, (ESI , m/z) 1224.3 (calcd for
24: 1225.1).
Three atropisomers: TLC (alumina, hexane/dichloro-
C H N F
56 27 5
methane, 6:2.5) R 0.72, 0.61, 0.55, respectively. Spectro-
14. Ka, J.-W.; Cho, W.-S.; Lee, C.-H. Tetrahedron Lett.
2000, 41, 8121–8125.
f
1
scopic data of the major fraction (R 0.61). H NMR (400
f