Bis-porphyrin arrays. Part 2. The synthesis of asymmetrically substituted bis-porphyrins

Article abstract of DOI:10.1039/a907059a

A strategy for the synthesis of asymmetrically substituted tetraazaanthracene linked bis-porphyrins in which the two porphyrin rings contain differences in their peripheral substituents has been developed. The method is illustrated by the preparation of bis-porphyrins with a single meso-halophenyl and seven meso-3,5-di-tert-butylphenyl substituents. The bis-porphyrins were prepared by condensation of a porphyrin-α-dione with benzene-1,2,4,5-tetraamine to form a porphyrin diaminoquinoxaline intermediate which was subsequently condensed with a second different porphyrin-α-dione. The key issue in the synthesis was the separation of the desired asymmetrically substituted bis-porphyrin from the symmetric bis-porphyrin by-products of similar polarities. Enhanced separation of the bis-porphyrin products was achieved by chelation of a metal into one of the porphyrin rings, the metal being introduced at the porphyrin-α-dione stage. Copper was successfully used when chelated into the less polar porphyrin-α-dione but the use of zinc in the more polar porphyrin-α-dione to enhance bis-porphyrin separation was unsuccessful as the pyridinium hydrochloride produced in the reaction was found to demetallate the porphyrins.

Full text of DOI:10.1039/a907059a

View Article Online / Journal Homepage / Table of Contents for this issue
Bis-porphyrin arrays. Part 2.† The synthesis of asymmetrically
substituted bis-porphyrins
Richard Beavington and Paul L. Burn*
1
The Dyson Perrins Laboratory, South Parks Road, Oxford, UK OX1 3QY
Received (in Cambridge, UK) 1st September 1999, Accepted 24th November 1999
A strategy for the synthesis of asymmetrically substituted tetraazaanthracene linked bis-porphyrins in which the two
porphyrin rings contain differences in their peripheral substituents has been developed. The method is illustrated by
the preparation of bis-porphyrins with a single meso-halophenyl and seven meso-3,5-di-tert-butylphenyl substituents.
The bis-porphyrins were prepared by condensation of a porphyrin-α-dione with benzene-1,2,4,5-tetraamine to form a
porphyrin diaminoquinoxaline intermediate which was subsequently condensed with a second different porphyrin-α-
dione. The key issue in the synthesis was the separation of the desired asymmetrically substituted bis-porphyrin from
the symmetric bis-porphyrin by-products of similar polarities. Enhanced separation of the bis-porphyrin products
was achieved by chelation of a metal into one of the porphyrin rings, the metal being introduced at the porphyrin-α-
dione stage. Copper was successfully used when chelated into the less polar porphyrin-α-dione but the use of zinc in
the more polar porphyrin-α-dione to enhance bis-porphyrin separation was unsuccessful as the pyridinium
hydrochloride produced in the reaction was found to demetallate the porphyrins.
ested in developing synthetic routes to bis-porphyrin arrays,
Introduction
as illustrated in Fig. 1, which are linked through the meso-
New covalently linked porphyrin arrays for use in molecular
positions. Linking of the bis-porphyrins should enhance their
electronics and modelling the photosynthetic process are being
solution processability, and by having a conjugated meso-
diphenylacetylene bridge the electronic properties of the bis-
continually developed.1–8 The vast majority of porphyrin arrays
are linked through the meso-positions.2,4–8 The alternative
porphyrins may also be modulated. To prepare meso-connected
and less well studied route for connecting porphyrins is to link
bis-porphyrins it is necessary to make bis-porphyrins which
them through the β-pyrrolic positions.1,3,9–11 For the covalently-
have a meso-substituent with a linking point. To this end we
linked porphyrin arrays reported each of the porphyrins within
have reported the synthesis of porphyrin-α-diones which have a
the array uses only one type of bridging point, meso- or β-
single meso-halophenyl moiety, to allow coupling to conjugated
pyrrolic, and there have been no papers describing porphyrin
arrays in which the individual porphyrin rings have links
linkers using palladium catalysis, with the other meso-positions
carrying the solubilising 3,5-di-tert-butylphenyl groups.13 In
through both the meso- and β-pyrrolic positions. The main type
this paper we report the strategy for the preparation and purifi-
of porphyrin array connected by a conjugated bridge through
cation of asymmetrically substituted bis-porphyrins in which
the β-pyrrolic positions uses a tetraazaanthracene linker.9,10 In
one porphyrin ring contains a single meso-linking moiety as the
these tetraazaanthracene bridged bis-porphyrins the electronic
next step towards forming bis-porphyrin arrays.
properties of the two porphyrins are modulated12 so that they
may be of interest as charge transport layers in light-emitting
diodes or as materials for photovoltaics. We were therefore inter-
Results and discussion
The best strategy for the preparation of asymmetrically substi-
tuted bis-porphyrins containing tetraazaanthracene bridges is
† For Part 1, see ref. 13.
Fig. 1 A diphenylacetylene linked bis-porphyrin array.
DOI: 10.1039/a907059a
J. Chem. Soc., Perkin Trans. 1, 2000, 605–609
This journal is © The Royal Society of Chemistry 2000
605

View Article Online
Scheme 1 Reagents and conditions: i, pyridine, benzene-1,2,4,5-tetraamine tetrahydrochloride, Ar, ∆ then second porphyrin-α-dione, Ar, ∆.
possibility of forming symmetric bis-porphyrins with eight 3,5-
di-tert-butylphenyl groups, or with six 3,5-di-tert-butylphenyl
groups and two halophenyl substituents as well as the desired
asymmetrically substituted bis-porphyrin. Given the small
changes in outer peripheral substituents the difference in polar-
ity between each of the potential bis-porphyrins might be small
and hence separations based on polarity would be difficult.
To determine the best method for achieving the synthesis
of a mono-meso-halophenylbis-porphyrin we chose the more
easily accessed mono-meso-bromophenylporphyrin-α-dione 1,
although for palladium catalysed couplings to make the desired
bis-porphyrin arrays the iodophenyl substituent would be
better.
The first experiment was to determine whether the polarities
of the free-base (4Ј-bromophenyl)heptakis(3Љ,5Љ-di-tert-butyl-
phenyl)bis-porphyrin 7 (Scheme 1) and the “symmetric” bis-
porphyrin by-products 6 and 8 were sufficiently different to
allow easy separation. The (4-bromophenyl)porphyrin-α-dione
1 was reacted with 1.1 equivalents of benzene-1,2,4,5-tetra-
amine tetrahydrochloride in pyridine heated at reflux. We found
that the use of refluxing pyridine was advantageous over the
reactions carried out at room temperature as both the first and
second condensations occurred much more quickly. After one
hour the porphyrin diaminoquinoxaline intermediate was
added to an excess of the symmetric porphyrin-α-dione 4 and
the reaction mixture heated at reflux overnight. This procedure
gave, as expected, a mixture of the three bis-porphyrins and
as anticipated they had similar polarities. However, they could
be separated with difficulty using chromatotron and column
Fig. 2 Porphyrin diaminoquinoxaline intermediate.
to react one equivalent of a porphyrin-α-dione with one equiv-
alent of benzene-1,2,4,5-tetraamine to form a porphyrin
diaminoquinoxaline (Fig. 2) which is then condensed with a
second porphyrin-α-dione.14 This procedure works well pro-
viding the meso-substituents flanking the α-dione of the first
porphyrin are sterically demanding and that the two porphyrin
components have significantly different properties. The latter is
important because in addition to the desired asymmetrically
substituted bis-porphyrin there are often small amounts of the
undesired symmetric bis-porphyrins which arise from conden-
sation of two equivalents of the same porphyrin-α-dione with
the benzene-1,2,4,5-tetraamine. If the polarities of the two
porphyrin components used in the synthesis are significantly
different then the bis-porphyrin products can be easily separ-
ated by column chromatography over silica.14 However, in our
study we require the bis-porphyrins to have only one of the
meso-phenyls substituted with a halo group (for linking to form
the array) with the other seven containing solubilising tert-butyl
groups (for ease of manipulation). This means that there is the
606
J. Chem. Soc., Perkin Trans. 1, 2000, 605–609

View Article Online
chromatography over silica and under these conditions the
desired (4Ј-bromophenyl)heptakis(3Љ,5Љ-di-tert-butylphenyl)-
bis-porphyrin 7 was isolated in a 56% yield, with octakis(3Љ,5Љ-
di-tert-butylphenyl)bis-porphyrin 6 and bis(4Ј-bromophenyl)-
hexakis(3Љ,5Љ-di-tert-butylphenyl)bis-porphyrin 8 collected in
7% and ~9% yields respectively [8 was not isolated in a pure
form]. Even after careful separation some of the bis-porphyrin
material remained as a mixture of 6, 7, and 8 (~16% by mass). It
is important to note that in the case of bis(4Ј-bromophenyl)-
hexakis(3Љ,5Љ-di-tert-butylphenyl)bis-porphyrin 8 there are two
possible isomers, the “cis”, as illustrated in Scheme 1, and the
“trans” which has one of the bromophenyl groups in a “top”
meso-position. We believe that there is no chemical advantage
for the formation of one isomer of 8 over the other and there-
fore a mixture of the two would occur. In addition, the excess
symmetric porphyrin-α-dione 4 was recovered. Although the
desired bis-porphyrin was the major product, a more efficient
method for separating the products was clearly required.
The two strategies that we investigated to improve the separ-
ation both involved the chelation of metals to the inner per-
ipheries of the porphyrin-α-diones. It is known that chelation
of zinc generally increases the polarity of a porphyrin when
compared to the equivalent free-base porphyrin whilst the
chelation of copper normally decreases the polarity of the
porphyrin. Therefore, the logical approach in our study was to
chelate zinc into the more polar porphyrin-α-dione, in this case
1, or copper into the less polar symmetric porphyrin-α-dione 4.
The zinc (4Ј-bromophenyl)porphyrin-α-dione 2 was easily made
in a 99% isolated yield by reacting 1 with an excess of zinc
acetate dihydrate in a dichloromethane–methanol mixture
heated at reflux for three hours. However, one of the by-
products of the condensation reaction used to form the bis-
porphyrins is hydrogen chloride and although the reactions are
carried out in pyridine we were concerned that the pyridinium
hydrochloride could remove the relatively labile zinc. To test
this we carried out a reaction on the more readily available
symmetric zinc porphyrin-α-dione 315 to make the dizinc
octakis(3Ј,5Ј-di-tert-butylphenyl)bis-porphyrin 10. Two equiv-
alents of the zinc porphyrin-α-dione 3 were reacted with
benzene-1,2,4,5-tetraamine tetrahydrochloride in pyridine
heated at reflux for 16 hours. We avoided the usual acidic work-
up, which is normally used to help remove the pyridine, to
avoid demetallation of the zinc bis-porphyrins. Under these
conditions the free-base octakis(3Ј,5Ј-di-tert-butylphenyl)bis-
porphyrin 6 was isolated in a 27% yield, with the zinc octakis-
(3Ј,5Ј-di-tert-butylphenyl)bis-porphyrin 9 isolated in a 33%
yield. The desired dizinc octakis(3Ј,5Ј-di-tert-butylphenyl)bis-
porphyrin 10 was collected in an impure form in a yield of
less than 11%. Although the three bis-porphyrins could be
separated from each other this reaction clearly showed that zinc
was being removed under the reaction conditions and therefore
the use of zinc to improve the separation in the preparation
of the (4Ј-bromophenyl)heptakis(3Љ,5Ј-di-tert-butylphenyl)bis-
porphyrin was not viable using these conditions.
dicopper octakis(3Ј,5Ј-di-tert-butylphenyl)bis-porphyrin 12 was
isolated in a 21% yield, the copper octakis(3Ј,5Ј-di-tert-butyl-
phenyl)bis-porphyrin 11 was isolated in a 78% yield, and 13%
of the copper porphyrin-α-dione 5 was returned. Importantly
the bis-porphyrin products were separable by column chrom-
atography. Given the success of using the copper chelated
porphyrin-α-dione
5 we reacted the (4Ј-bromophenyl)-
porphyrin-α-dione 1 with a small excess of benzene-1,2,4,5-
tetraamine tetrahydrochloride in pyridine heated at reflux for
1 hour. The solution was then added to an excess of the copper
porphyrin-α-dione 5 and the reaction mixture heated at reflux
for a further 15 hours. On purification the copper (4Ј-bromo-
phenyl)heptakis(3Љ,5Љ-di-tert-butylphenyl)bis-porphyrin 13 was
isolated in an 88% yield, and the excess copper diketone 5
was recovered. Only traces of two other bis-porphyrins were
observed, which were thought to be the dicopper octakis(3Ј,5Ј-
di-tert-butylphenyl)bis-porphyrin 12 and the bis(4Ј-bromo-
phenyl)hexakis(3Љ,5Љ-di-tert-butylphenyl)bis-porphyrin 8, and
these were easily removed from the desired bis-porphyrin by
chromatography over silica.
The final aspect of this study was to remove the copper
from the copper (4Ј-bromophenyl)heptakis(3Љ,5Љ-di-tert-butyl-
phenyl)bis-porphyrin 13 to give the desired free-base 7. This
was achieved by treating 13 with concentrated sulfuric acid,
however, the yields of free-base bis-porphyrin 7 were quite poor
(up to 42%). One reason for the low yield during the demetal-
lation procedure may be that protonation of the free-base por-
phyrin in the bis-porphyrin occurs, causing the bis-porphyrin to
precipitate out of the reaction mixture before the metal can be
removed.
Conclusion
We have successfully developed a strategy for the synthesis and
separation of bis-porphyrins which have similar polarities due
to only minor differences in substituents on the porphyrin outer
periphery. We have shown that chelation of copper into the less
polar porphyrin-α-dione gives enhanced separation of the bis-
porphyrins and that the copper is stable under the reaction
conditions. In contrast, zinc is an unsuitable metal for this work
due to its lability under the reaction conditions caused by the
pyridinium hydrochloride in the reaction mixture. Finally we
have shown that the copper could be removed from the bis-
porphyrin to give the corresponding free-base compound.
Experimental
Measurements
1H NMR spectra were recorded on a Bruker AM-500 (500
MHz) or a Varian Gemini 200 (200 MHz) spectrometer. Infra-
red spectra were recorded with either a Perkin-Elmer 781 or
Paragon 1000 infrared spectrometer. UV–visible spectra were
recorded on Perkin-Elmer UV–visible spectrometers (Lambda
2 or Lambda 14P) in spectrophotometric grade dichlorometh-
ane. Fast Atom Bombardment (FAB) mass spectra (m/z) were
recorded on a VG Autospec spectrometer. MALDI mass spec-
tra were measured on a Micromass TofSpec-2E in the reflectron
mode. Samples were run from a Dithranol (1,8,9-trihydroxy-
anthracene) matrix using solutions in a chloroform–pyridine
In contrast, previous work has shown that a copper
porphyrin-α-dione and copper bis-porphyrins are stable to the
pyridinium hydrogen chloride produced during the bis-
porphyrin preparation when the reaction is carried out at room
temperature.14 However, our synthetic protocol for forming the
bis-porphyrins uses pyridine heated at reflux and although
copper porphyrins should be stable under these conditions we
nevertheless thought it important to confirm their stability by
preparing the copper octakis(3Ј,5Ј-di-tert-butylphenyl)bis-
porphyrin from porphyrin-α-diones 4 and 5. This was achieved
by reacting the free-base porphyrin-α-dione 4 with a small
excess of benzene-1,2,4,5-tetraamine tetrahydrochloride in
pyridine heated at reflux for 0.5 hours to form the correspond-
ing porphyrin diaminoquinoxaline which was subsequently
condensed with an excess of the copper porphyrin-α-dione 5
under the same conditions for 16.5 hours. On purification, the
mixture (9:1) with a sample concentration of 3–5 pmol µlϪ1
.
Due to the combination of metal chelation, halogen substitu-
ents, molecular size, and different degrees of protonation the
molecular weights observed by mass spectroscopy generally
appeared as a complex isotopic cluster around the expected
parent ion. We have quoted the mass for the base peak for each
compound. Melting points were determined on a Gallenkamp
melting point apparatus and are uncorrected. We found that 2
did not melt but partially decomposed on heating. This was
observed by analysis with thin-layer chromatography which
J. Chem. Soc., Perkin Trans. 1, 2000, 605–609
607

View Article Online
showed the presence of more than one compound after heating.
Therefore, the melting point is quoted as decomposition below
the highest temperature measured. Microanalyses were per-
formed in the Inorganic Chemistry Laboratory, Oxford. All
solvents for recrystallization were distilled before use. Pyridine
was dried and stored, after distillation, over potassium hydrox-
ide pellets. Light petroleum refers to the fraction of boiling
point 60–80 ЊC. Thin-layer chromatography was performed on
glass micro plates coated with silica GF254, or with Merck alu-
minium plates coated with silica gel 60 F254. Column chrom-
atography was performed using ACROS Organics silica gel,
0.035–0.07 mm. Where solvent mixtures are used, the propor-
tions are given by volume.
mmol), dichloromethane (25 cm3), and sulfuric acid (98%, 1.0
cm3) was stirred for 7 min at room temperature. An ice–water
mixture (20 cm3) was added carefully, and the ice allowed to
melt. The aqueous layer was separated and extracted with
dichloromethane (10 cm3). The combined organic layers were
washed with aqueous sodium bicarbonate (5%, 20 cm3) and
water (20 cm3), dried over anhydrous sodium sulfate, and the
solvent completely removed. The residue was purified by
column and chromatotron chromatography over silica using
dichloromethane–light petroleum mixtures (1:4 to 3:7) as
eluent to give (4Ј-bromophenyl)heptakis(3Љ,5Љ-di-tert-butyl-
phenyl)bis-porphyrin 7 as a brown solid (57 mg, 42%), which
co-chromatographed with and had an identical 1H NMR spec-
trum to an authentic sample.
(4Ј-Bromophenyl)heptakis(3Љ,5Љ-di-tert-butylphenyl)bis-
porphyrin 7
[17,18-Dioxo-5-(4Ј-bromophenyl)-10,15,20-tris(3Љ,5Љ-di-tert-
butylphenyl)chlorinato]zinc(II) 2
Method 1. A solution of (4-bromophenyl)porphyrin-α-dione
1 (73 mg, 0.07 mmol) and benzene-1,2,4,5-tetraamine tetrahy-
drochloride (23 mg, 0.08 mmol) in dry, deoxygenated, pyridine
(1.8 cm3) was heated at reflux under argon for 1 h. The solution
was then transferred under argon to porphyrin-α-dione 415 (128
mg, 0.117 mmol). The first reaction flask was rinsed with pyrid-
ine (1.4 cm3) and this was transferred to the reaction mixture
under argon. The reaction mixture was heated at reflux under
argon for 15.6 h and then allowed to cool. Dichloromethane (25
cm3) was added, and the organic solution was washed with
dilute hydrochloric acid (2 M, 2 × 15 cm3), aqueous sodium
bicarbonate (10%, 20 cm3), and water (20 cm3), dried over
anhydrous sodium sulfate, filtered, and the solvent completely
removed. The residue was purified with difficulty by column
and chromatotron chromatography over silica using dichloro-
methane–light petroleum mixtures (3:20 to 2:3) as eluent, and
four compounds were isolated. Octakis(3Ј,5Ј-di-tert-butyl-
phenyl)bis-porphyrin 6 was isolated as a brown solid (9 mg,
7%), which had an identical 1H NMR spectrum to that
reported.10 (4Ј-Bromophenyl)heptakis(3Љ,5Љ-di-tert-butylphen-
yl)bis-porphyrin 7 was isolated as a brown solid (86 mg, 56%),
mp >300 ЊC (Found: C, 80.74; H, 7.15; N, 7.26. C150H169N12Br
requires C, 81.16; H, 7.67; N, 7.57%); νmax(KBr disc)/cmϪ1 3350
(N–H); λmax(CH2Cl2)/nm (log ε) 426 (5.53), 457 (5.47), 500sh
(4.91), 604sh (4.25), 612sh (4.29), 629 (4.32), and 676 (3.84);
δH(500 MHz; CDCl3) Ϫ2.38 (1 H, s, NH), Ϫ2.37 (1 H, s, NH),
Ϫ2.31 (2 H, s, NH), 1.55–1.60 (126 H, t-butyl H), 7.81 (2 H, dd,
J2Љ,4Љ = J6Љ,4Љ = 2.0 Hz, C(4Љ)H), 7.83 (1 H, dd, J2Љ,4Љ = J6Љ,4Љ = 2.0
Hz, C(4Љ)H), 7.92 and 8.13 (4 H, AAЈBBЈ, C(2Ј)H, C(3Ј)H,
C(5Ј)H, and C(6Ј)H), 8.01 (4 H, m, C(4Љ)H), 8.10 (14 H, m,
C(2Љ)H and C(6Љ)H), 8.59 (2 H, s, bridge H), 8.69 and 8.77 (2 H,
ABq, JAB = 4.5 Hz, C(4)H and C(5)H), 8.77 (2 H, s, C(35)H and
C(36)H), and 8.89 and 8.97 (8 H, 2 × m, C(9)H, C(10)H,
C(19)H, C(20)H, C(30)H, C(31)H, C(40)H, and C(41)H); m/z
(FAB) 2221.4 (100%). Bis(4Ј-bromophenyl)hexakis(3Љ,5Љ-di-
tert-butylphenyl)bis-porphyrin 8 was isolated as a brown solid
which contained a minor amount of a second porphyrin (~7
mg, ~9%). Selected data for 8: νmax(KBr disc)/cmϪ1 3350 (N–H);
δH(500 MHz; CDCl3) Ϫ2.39 and Ϫ2.37 (4 H, 2 × s, NH), 1.55–
1.59 (108 H, t-butyl H), 7.83 (2 H, m, C(4Љ)H), 7.92 and 8.13
(8 H, AAЈBBЈ, C(2Ј)H, C(3Ј)H, C(5Ј)H, and C(6Ј)H), 8.01
(4 H, m, C(4Љ)H), 8.09 (12 H, m, C(2Љ)H and C(6Љ)H), 8.58 (2 H,
s, bridge H), 8.69 and 8.77 (4 H, ABq, JAB = 4.5 Hz, C(4)H,
C(5)H, C(35)H, C(36)H), 8.87–8.91 (6 H, br m, β-pyrrolic H),
and 8.98 (2 H, br 1/2ABq, JAB = 5.0 Hz, β-pyrrolic H); m/z
(FAB) 2188.2 (100%). Porphyrin-α-dione 4 was returned as
a green solid (32 mg), a sample of which which co-chromato-
A mixture of (4-bromophenyl)porphyrin-α-dione 1 (51 mg,
0.048 mmol), zinc() acetate dihydrate (31 mg, 0.14 mmol),
dichloromethane (20 cm3), and methanol (2 cm3) was heated
at reflux for 4.5 h. The solvent was completely removed, and
the residue purified by column chromatography over silica
using a dichloromethane–light petroleum mixture (2:1) as
eluent to give 2 as a green solid (52 mg, 98%), mp <300 ЊC
(decomp.) (Found m/z 1120.417 (Mϩ). C68H73N4BrO2Zn
requires 1120.421); νmax(KBr disc)/cmϪ1 1724 (C᎐O); λmax(CH2-
᎐
Cl2)/nm (log ε) 414 (5.10), 494 (4.25), 631 (3.65), 723 (3.70);
δH(500 MHz; CDCl3) 1.45–1.49 (54 H, 3 × s, t-butyl H), 7.63 (4
H, m, C(2Љ)H and C(6Љ)H), 7.71 (2 H, m, C(4Љ)H), 7.76 (1 H,
dd, J2Љ,4Љ = J6Љ,4Љ = 1.77 Hz, C(4Љ)H), 7.84 and 7.94 (4 H, AAЈBBЈ,
C(2Ј)H, C(3Ј), C(5Ј)H, and C(6Ј)H), 7.91 (2 H, d, J4Љ,2Љ
=
J4Љ,6Љ = 2 Hz, C(2Љ)H and C(6Љ)H), 8.35 and 8.53 (2 H, ABq,
JAB = 4.5 Hz, β-pyrrolic H), 8.36 and 8.62 (2 H, ABq, JAB = 4.5
Hz, β-pyrrolic H), and 8.35 and 8.53 (2 H, ABq, JAB = 5 Hz,
β-pyrrolic H), and 8.45 and 8.54 (2 H, ABq, JAB = 4.5 Hz,
β-pyrrolic H).
Octakis(3Ј,5Ј-di-tert-butylphenyl)bis-porphyrin dizinc(II) 10
A solution of zinc porphyrin-α-dione 315 (45 mg, 0.04 mmol)
and benzene-1,2,4,5-tetraamine tetrahydrochloride (6 mg, 0.02
mmol) in dry, deoxygenated, pyridine (0.5 cm3) was heated at
reflux under argon in the dark for 16 h. The solution was
allowed to cool and then dichloromethane (20 cm3) was added.
The organic solution was washed with water (3 × 20 cm3), dried
over anhydrous sodium sulfate, filtered, and the solvent com-
pletely removed. The residue was purified by column chrom-
atography over silica using a dichloromethane–light petroleum
mixture (2:1) as eluent and three bands were collected.
Octakis(3Ј,5Ј-di-tert-butylphenyl)bis-porphyrin 6 was isolated
as a brown solid (12 mg, 27%), which had an identical 1H NMR
spectrum to that reported.10 Octakis(3Ј,5Ј-di-tert-butylphenyl)-
bis-porphyrin zinc() 9 was isolated as a brown solid (15 mg,
33%), which had an identical 1H NMR spectrum to that
reported14; δH(500 MHz; CDCl3) Ϫ2.25 (2 H, s, NH), 1.60,
1.61, 1.65, and 1.66 (144 H, 4 × s, t-butyl H), 7.86 (2 H, dd,
J2Ј,4Ј = J6Ј,4Ј = 2.0 Hz, C(4Ј)H), 7.87 (2 H, dd, J2Ј,4Ј = J6Ј,4Ј = 2.0
Hz, C(4Ј)H), 8.05 (2 H, dd, J2Ј,4Ј = J6Ј,4Ј = 1.5 Hz, C(4Ј)H), 8.06
(2 H, dd, J2Ј,4Ј = J6Ј,4Ј = 2.0 Hz, C(4Ј)H), 8.12 (4 H, d,
J4Ј,6Ј = J4Ј,6Ј = 1.5 Hz, C(2Ј)H and C(6Ј)H), 8.15 (4 H, d,
J4Ј,6Ј = J4Ј,6Ј = 1.5 Hz, C(2Ј)H and C(6Ј)H), 8.16 (8 H, br s,
C(2Ј)H and C(6Ј)H), 8.73 (2 H, s, bridge H), 8.83 and 8.91 (4 H,
2 × s, C(4)H, C(5)H, C(35)H, and C(36)H), 8.89 and 9.03 (4 H,
ABq, JAB = 5 Hz, C(9)H, C(10)H, C(19)H and C(20)H), and
8.95 and 9.03 (4 H, ABq, JAB = 5 Hz, C(30)H, C(31)H, C(40)H
and C(41)H); m/z (MALDI) 2317.5 (Mϩ, 100%). Octakis(3Ј,5Ј-
di-tert-butylphenyl)bis-porphyrin dizinc() 10 was collected as
an impure brown solid (5 mg, <11%). The main signals in the
1H NMR spectrum corresponded to those reported for 10;14
δH(200 MHz; CDCl3) 1.54 and 1.59 (144 H, 2 × s, t-butyl H),
1
graphed with and had an identical H NMR spectrum to an
authentic sample. Some material (~19 mg) remained as a
mixture of 6, 7, and 8.
Method 2. A mixture of (4Ј-bromophenyl)octakis(3Љ,5Љ-di-
tert-butylphenyl)bis-porphyrin copper() 13 (138 mg, 0.060
608
J. Chem. Soc., Perkin Trans. 1, 2000, 605–609

View Article Online
7.80 (2 H, dd, J2Ј,4Ј = J6Ј,4Ј = 1.5 Hz, C(4Ј)H), 8.00 (2 H, dd,
J2Ј,4Ј = J6Ј,4Ј = 1.5 Hz, C(4Ј)H), 8.06 (8 H, d, J4Ј,2Ј = J4Ј,6Ј = 1.5 Hz,
C(2Ј)H and C(6Ј)H), 8.09 (8 H, d, J4Ј,2Ј = J4Ј,6Ј = 1.5 Hz, C(2Ј)H
and C(6Ј)H), 8.76 (2 H, s, bridge H), 8.82 and 8.97 (8 H, ABq,
JAB = 5.0 Hz, C(9)H, C(10)H, C(19)H, C(20)H, C(30)H,
C(31)H, C(40)H, and C(41)H), and 8.85 (4 H, s, C(4)H, C(5)H,
C(35)H, and C(36)H); m/z (MALDI) 2317.5 (Mϩ, 100%).
with pyridine (1.2 cm3) and this was then transferred to the
reaction mixture. The reaction mixture was heated at reflux
under argon for 15 h and then allowed to cool. Dichlorometh-
ane (20 cm3) was added, and the organic solution was washed
with water (3 × 20 cm3), dried over anhydrous sodium sulfate,
filtered, and the solvent completely removed. The residue
was purified by column chromatography over silica using
dichloromethane–light petroleum mixtures (1:4–2:3) as eluent
and two bands were collected. (4Ј-Bromophenyl)heptakis(3Љ,5Љ-
di-tert-butylphenyl)bis-porphyrin copper() 13 was isolated as
a brown solid (169 mg, 88%), mp >300 ЊC (Found: C, 78.46; H,
7.50; N, 7.08. C150H167N12BrCu requires C, 78.97; H, 7.38; N,
7.37%); νmax(CHCl3)/cmϪ1 3350 (N–H); λmax(CH2Cl2)/nm (log ε)
425 (5.51), 454 (5.34), 505 (4.83), 641 (4.30), and 680sh (4.13);
m/z (FAB) 2281.7 (100%). Copper porphyrin-α-dione 5 was
isolated as a green solid (55 mg, 33%), a sample of which co-
chromatographed with and had an identical infrared spectrum
to an authentic sample.
Octakis(3Ј,5Ј-di-tert-butylphenyl)bis-porphyrin copper(II) 11
A solution of porphyrin-α-dione 4 (29 mg, 0.03 mmol) and
benzene-1,2,4,5-tetraamine tetrahydrochloride (9 mg, 0.03
mmol) in dry, deoxygenated, pyridine (0.75 cm3) was heated at
reflux under argon for 0.5 h. The reaction mixture was then
transferred to copper porphyrin-α-dione 515 (52 mg, 0.045
mmol) under argon. The first flask was rinsed with pyridine (0.5
cm3) which was then transferred to the reaction mixture under
argon. The reaction mixture was heated at reflux under argon
for 16.5 h and then allowed to cool. Dichloromethane (20 cm3)
was added, and the organic solution was washed with dilute
hydrochloric acid (3 M, 2 × 10 cm3), aqueous sodium bicarbon-
ate (5%, 20 cm3), and water (20 cm3), dried over anhydrous
sodium sulfate, filtered, and the solvent completely removed.
The residue was purified by column and chromatotron chrom-
atography over silica using a dichloromethane–light petroleum
mixture (1:4) as eluent and three bands were collected.
Octakis(3Ј,5Ј-di-tert-butylphenyl)bis-porphyrin dicopper() 12
was isolated as a brown solid (11 mg, 21%), which had a UV–
visible spectrum corresponding to an authentic sample;14
λmax(CH2Cl2)/nm (log ε) 424 (5.52), 454 (5.39), 517 (4.84),
534 (4.83), and 670 (4.36); m/z (MALDI) 2377.3 (100%).
Octakis(3Ј,5Ј-di-tert-butylphenyl)bis-porphyrin copper() 11
was isolated as a brown solid (47 mg, 78%), which had a UV–
visible spectrum corresponding to an authentic sample;14
νmax(CHCl3)/cmϪ1 3350 (N–H); λmax(CH2Cl2)/nm (log ε) 425
(5.45), 455 (5.30), 507 (4.80), 608sh (4.19), 642 (4.26), and 687sh
(4.04); m/z (MALDI) 2316.3 (100%). Copper porphyrin-
α-dione 5 was isolated as a green solid (7 mg), a sample of
which co-chromatographed with and had an identical infrared
spectrum to an authentic sample.
References
1 T. Balasubramanian and J. S. Lindsey, Tetrahedron, 1999, 55, 6771.
2 M. O. Senge and X. Feng, Tetrahedron Lett., 1999, 40, 4165.
3 A. K. Burrell and D. L. Officer, Synlett, 1998, 1297.
4 C. C. Mak, N. Bampos and J. K. M. Sanders, Chem. Commun.,
1999, 1085.
5 J. Wytko, V. Berl, M. McLaughlin, R. R. Tykwinski, M. Schreiber,
F. Diederich, C. Boudon, J.-P. Gisselbrecht and M. Gross, Helv.
Chim. Acta, 1998, 81, 1964.
6 A. Nakano, A. Osuka, I. Yamazaki, T. Yamazaki and Y. Nishimura,
Angew. Chem., Int. Ed., 1998, 37, 3023.
7 H. A. Biemans, A. R. Rowan, A. Verhoeven, P. Vanoppen, L.
Latterini, J. Foekema, A. P. H. J. Schenning, E. W. Meijer, F. C. de
Schryver and R. J. M. Nolte, J. Am. Chem. Soc., 1998, 120, 11054.
8 P. N. Taylor, A. P. Wylie, J. Huuskonen and H. L. Anderson, Angew.
Chem., Int. Ed., 1998, 37, 986.
9 M. J. Crossley and P. L. Burn, J. Chem. Soc., Chem. Commun., 1991,
1569.
10 M. J. Crossley and P. L. Burn, J. Chem. Soc., Chem. Commun., 1987,
39.
11 E. J. Atkinson, A. M. Oliver and M. N. Paddon-Row, Tetrahedron,
1993, 34, 6147.
12 P. L. Burn, M. J. Crossley and P. A. Lay, Electrochemistry: Current
and Potential Applications, Proceedings of the Seventh Australian
Electrochemistry Conference, Sydney, NSW, Australia, 1987, 245.
13 R. Beavington and P. L. Burn, J. Chem. Soc., Perkin Trans. 1, 1999,
583.
4Ј-Bromophenyl heptakis(3Љ,5Љ-di-tert-butylphenyl)bis-porphyrin
copper(II) 13
A solution of 4Ј-bromophenyl porphyrin-α-dione 1 (89 mg,
0.084 mmol) and benzene-1,2,4,5-tetraamine tetrahydrochlor-
ide (29 mg, 0.10 mmol) in dry, deoxygenated, pyridine (2.3
cm3) was heated at reflux under argon for 1 h. The reaction
mixture was transferred under argon to copper porphyrin-α-
dione 5 (165 mg, 0.143 mmol). The original flask was rinsed
14 P. L. Burn, PhD Thesis, Sydney University, Australia, 1989.
15 R. Beavington, P. A. Rees and P. L. Burn, J. Chem. Soc., Perkin
Trans. 1, 1998, 2847.
Paper a907059a
J. Chem. Soc., Perkin Trans. 1, 2000, 605–609
609