Hexameric and pentameric slipped-cofacial dimers: Toward an artificial light-harvesting complex

Article abstract of DOI:10.1021/jo047923n

(Chemical Equation Presented) We prepared a zinc complex of bis(1-methylimidazolyl)-m-gable porphyrin. The automatically assembled coordinate species showed a complex mixture of wide molecular weight distributions accompanied with assemblies of specific a

Full text of DOI:10.1021/jo047923n

Hexameric and Pentameric Slipped-Cofacial Dimers: Toward an
Artificial Light-Harvesting Complex
Ryoichi Takahashi† and Yoshiaki Kobuke*,†,‡
Graduate School of Materials Science, Nara Institute of Science and Technology, and CREST, Japan
Science and Technology Corporation (JST), 8916-5 Takayama, Ikoma, Nara 630-0192, Japan
kobuke@ms.naist.jp
Received November 23, 2004
We prepared a zinc complex of bis(1-methylimidazolyl)-m-gable porphyrin. The automatically
assembled coordinate species showed a complex mixture of wide molecular weight distributions
accompanied with assemblies of specific assembly numbers. When this zinc complex was once
dissociated by the addition of methanol and reorganized again by elimination of the methanol under
high-dilution conditions in chloroform to facilitate intramolecular coordinate structure formation,
two convergent assemblies were obtained through analysis by gel permeation chromatography.
These assemblies gave round-shaped particles on solid substrates by probe microscopies (atomic
force and scanning tunneling microscopies). These two components were separated by GPC and
evidence from small-angle X-ray scattering measurements in solution with synchrotron radiation
was consistent with hexameric and pentameric macrorings of the gable porphyrin. The porphyrin
macrorings did not show any fluorescence quenching by assembly formation, and we anticipate
that the macrorings of gable porphyrins represent a good model for an artificial light-harvesting
complex.
Introduction
chlorophylls in LH1 and 18 or 16 ones in B850 of LH2
are arranged in macroring structures with use of bacte-
riochlorophyll-a dimers, which are characterized as being
in slipped-cofacial orientations to each other. These
macrorings are believed to be associated with the most
important function in light harvesting. Therefore, struc-
tural arrangement of porphyrins into a macroring is
extremely interesting for elucidating the mechanism of
highly efficient excitation energy transfer and developing
useful light-harvesting systems. Until now several oli-
gomeric porphyrins, particularly in circular forms, have
been synthesized through both covalent3 and supra-
molecular3b,4 approaches toward light-harvesting antenna
mimics. However, there are no systematic studies that
would allow a rational design of self-assembled macro-
rings with the desired properties toward antenna func-
tion. To the best of our knowledge, there was no example
of a porphyrin macroring composed of self-assembled
dimer units.
The first event of a photosynthetic reaction is the
absorption of light by light-harvesting antenna complexes
arranged around the reaction center complex. The struc-
tures of light-harvesting antenna systems for LH1, LH2,
and LH3 complexes of photosynthetic purple bacteria
have been determined by X-ray crystallographies1 and
electron microscopies.2 In these complexes, 32 bacterio-
† Nara Institute of Science and Technology (NAIST).
‡ CREST, Japan Science and Technology Corporation (JST).
(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, 517-521. (b) Koepke, J.; Hu, X.; Muenke, C.; Schulten, K.; Michel,
H. Structure 1996, 4, 581-597. (c) McLuskey, K.; Prince, S. M.; Cogdell,
R. J.; Isaacs, N. W. Biochemistry 2001, 40, 8783-8789.
(2) (a) Karrasch, S.; Bullough, P. A.; Ghosh, R. EMBO J. 1995, 14,
631-638. (b) Savage, H.; Cyrklaff, M.; Montoya, G.; Ku¨hlbrandt, W.;
Sinning, I. Structure 1996, 4, 243-252. (c) Walz, T.; Jamieson, S. J.;
Bowers, C. M.; Bullough, P. A.; Hunter, C. N. J. Mol. Biol. 1998, 282,
833-845.
10.1021/jo047923n CCC: $30.25 © 2005 American Chemical Society
Published on Web 03/01/2005
J. Org. Chem. 2005, 70, 2745-2753
2745

Takahashi and Kobuke
SCHEME 1. Molecular Design for Formation of
Supramolecular Macrocycle 3 by Complementary
Coordination of Bis(1-methylimidazol-2-yl) Gable
Porphyrinnatozinc(II) 2 from Its Free Base 1a
In preceding reports, we introduced an idea to organize
slipped-cofacial porphyrin dimers into macrorings to
construct an artificial light-harvesting complex of the
LHs of photosynthetic purple bacteria, in a hope to unveil
the function of this unique cyclic structure.5 Bis(1-
methylimidazolyl)porphyrin dimer, free base 1 and its
zinc complex 2, was employed as the basic structural unit
(Scheme 1).
The critical point of this molecular unit is the introduc-
tion of two 1-methylimidazolyl groups at the opposite
ends of the porphyrins connected by a m-phenylene
bridge. This molecular design should provide two impor-
tant characteristics for constructing the supramolecular
antenna model. First, the complementary coordination
between imidazolyl and central zinc ion is further
facilitated by π-π stacking interaction between porphy-
rin rings and characterized by a large association con-
stant (Ka > 1010 M-1).6,7 The resulting supramolecule
shows a large splitting of Soret band through exciton
coupling interaction of two porphyrins in a slipped-
cofacial arrangement in a close distance. Therefore, it
satisfies perfectly the functional requisites of the light-
harvesting dimer unit, in terms of the distance and the
orientation of chromophores for favorable energy transfer
without quenching of excitation energy. Second, the ring
structure will be provided by connecting two mono-
(imidazolyl)porphyrinatozinc(II) with a 1,3-phenylene
spacer, a so-called “gable porphyrin”.8 In contrast to the
meso-meso-coupled bis(1-methylimidazolyl)porphyrin-
atozinc(II), which grows linearly into a giant porphyrin
array reflecting the large association constant,7 the
spatial orientation of 120° is expected to give a closed
(3) (a) Sanders, J. K. M. In Comprehensive Supramolecular Chem-
istry; Atwood, J. L., Davies, J. E. D., MacNicol, D. D., Vo¨gtle, F., Eds.;
Pergamon Press: Oxford, 1996; Vol. 9, pp 131-164. (b) Sanders, J. K.
M. In The Porphyrin Handbook; Kadish, K. M., Smith, K. M., Guilard,
R., Eds.; Academic Press: New York, 2000; Vol. 3, pp 347-368. (c)
Anderson, S.; Anderson, H. L.; Bashall, A.; McPartlin, M.; Sanders, J.
K. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1096-1099. (d) Biemans,
H. A. M.; Rowan, A. E.; Verhoeven, A.; Vanoppen, P.; Latterini, L.;
Foekema, J.; Schenning, A. P. H. J.; Meijer, E. W.; de Schryver, F. C.;
Nolte, R. J. M. J. Am. Chem. Soc. 1998, 120, 11054-11060. (e) Li, J.;
Ambroise, A.; Yang, S. I.; Diers, J. R.; Seth, J.; Wack, C. R.; Bocian,
D. F.; Holten, D.; Lindsey, J. S. J. Am. Chem. Soc. 1999, 121, 8927-
8940. (f) Mongin, O.; Schuwey, A.; Vallot, M.-A.; Gossauer, A. Tetra-
hedron Lett. 1999, 40, 8347-8350. (g) Cho, H. S.; Rhee, H.; Song, J.
K.; Min, C.-K.; Takase, M.; Aratani, N.; Cho, S.; Osuka, A.; Joo, T.;
Kim, D. J. Am. Chem. Soc. 2003, 125, 5849-5860. (h) Peng, X.; Aratani,
N.; Takagi, A.; Matsumoto, T.; Kawai, T.; Hwang, I.-W.; Ahn, T. K.;
Kim, D.; Osuka, A. J. Am. Chem. Soc. 2004, 126, 4468-4469.
(4) (a) Satake, A.; Kobuke, Y. Tetrahedron 2005, 61, 13-41. (b) Chi,
X.; Guerin, A. J.; Haycock, R. A.; Hunter, C. A.; Sarson, L. D. Chem.
Commun. 1995, 2567-2569. (c) Drain, C. M.; Russell, K. C.; Lehn, J.-
M. Chem. Commun. 1996, 337-338. (d) Knapp, S.; Vasudevan, J.;
Emge, T. J.; Arison, B. H.; Potenza, J. A.; Schugar, H. J. Angew. Chem.,
Int. Ed. 1998, 37, 2368-2370. (e) Rubtsov, I. V.; Kobuke, Y.; H. Miyaji,
H.; Yoshihara, K. Chem. Phys. Lett. 1999, 308, 323-328. (f) Kuroda,
Y.; Sugou, K.; Sasaki, K. J. Am. Chem. Soc. 2000, 122, 7833-7834.
(g) Ikeda, C.; Nagahara, N.; Yoshioka, N.; Inoue, H. New J. Chem.
2000, 24, 897-902. (h) Tsuda, A.; Nakamura, T.; Sakamoto, S.;
Yamaguchi, K.; Osuka, A. Angew. Chem., Int. Ed. 2002, 41, 2817-
2821.
a meso-Substituents of coordinated species are abbreviated for
clarity for 3.
ring under appropriate conditions. If all of these designs
work perfectly, a dodecaporphyrin array composed of
hexakis(gable porphyrin) 3 would be constructed. This
porphyrin macroring array should have a barrel structure
with center-to-center distances of 6.1 and 11.0 Å, in close
analogy to those of the light-harvesting complexes of
photosynthetic purple bacteria. The remaining four meso-
positions are substituted by alkyl or olefinic groups to
give the assembly reasonable solubility in organic sol-
vents or further functionalization such as covalent con-
nection of the complementary coordination pair.5b,c
Results and Discussion
Preparation and Characterization of Gable Por-
phyrin. In preliminary experiments, we prepared several
dipyrromethane derivatives and the corresponding por-
phyrins with candidates of alkyl and aryl meso-substit-
uents. These surveys in the preparations are discussed
in Supporting Information. Finally, we utilized m-(n-
heptyl)dipyrromethane9 and succeeded in the synthesis
of bis(imidazolyl)-substituted gable porphyrin 1 with
consideration of the previous knowledge to overcome the
scrambling reactions of meso-substituents (see Support-
(5) (a) Takahashi, R.; Kobuke, Y. J. Am. Chem. Soc. 2003, 125,
2372-2373. (b) Ikeda, C.; Satake, A.; Kobuke, Y. Org. Lett. 2003, 5,
4935-4938. (c) Ohashi, A.; Satake, A.; Kobuke, Y. Bull. Chem. Soc.
Jpn. 2004, 77, 365-374. (d) Kuramochi, Y.; Satake, A.; Kobuke, Y. J.
Am. Chem. Soc. 2004, 126, 8668-8669.
(6) Kobuke, Y.; Miyaji, H. J. Am. Chem. Soc. 1994, 116, 4111-4112.
(7) Ogawa, K.; Kobuke, Y. Angew. Chem., Int. Ed. 2000, 39, 4070-
4073.
(8) (a) Tabushi, I.; Sasaki, T. Tetrahedron Lett. 1982, 23, 1913-
1916. (b) Tabushi, I.; Kugimiya, S.; Kinnaird, M. G.; Sasaki, T. J. Am.
Chem. Soc. 1985, 107, 4192-4199.
(9) Tomohiro, Y.; Satake, A.; Kobuke, Y. J. Org. Chem. 2001, 66,
8442-8446.
2746 J. Org. Chem., Vol. 70, No. 7, 2005

Hexameric and Pentameric Slipped-Cofacial Dimers
ing Information). Purification was achieved by combining
three kinds of chromatographies (basic alumina, silica
gel, and gel permeation columns), and analytically pure
free base gable porphyrin 1 was obtained in a 2.1% total
yield.5a
1
From 600 MHz H NMR measurement of this gable
porphyrin 1, three atropisomers were observed in a 1:2:1
statistical ratio. We examined the formation and isomer-
ization barrier of atropisomers of bis(1-methylimidazolyl)-
porphyrins. The steric hindrance between 1-methyl group
of imidazolyl ring and â-protons of pyrrole rings caused
an orientation of the imidazolyl group perpendicular to
the porphyrin plane. The atropisomers were distin-
guished from different chemical shifts most clearly of the
1-methyl of imidazolyl group, induced by a ring current
effect from the porphyrin plane. The rate of exchange
between atropisomers was evaluated from NMR spec-
trum of free base gable porphyrin 1 in the temperature
range 373-301 K. A plot of ln k versus 1/T (Figure S1)
gave the activation energy ∆Gq298K at 298 K of 71.1 kJ
mol-1. In the case of m-phenyl and m-o-fluoro-phenyl,
rotational barriers were reported as 45 ( 5 kJ mol-1 at
210 K and 93.2 ( 1.3 kJ mol-1 at 298 K, respectively.10
Further, in the case of trans-bis(1-methylimidazol-2-yl)-
porphyrin, no coalescence was observed at temperatures
lower than 373 K. Considering these points, gable por-
phyrin 1 might rotate not around the 1-methylimidazolyl
sides, but around the phenyl ring sides. Relatively large
activation energy ∆Gq of 1 compared to that of m-
phenylporphyrin may come from slower rotational move-
ment of bulky 3-porphyrinylphenyl substituent.
Zinc(II) Insertion into Gable Porphyrin. We
adopted an acetate method to insert zinc(II) ion to
imidazolylporphyrin under mild condition at room tem-
perature. Zinc insertion converted free base porphyrin 1
quantitatively to the corresponding zinc complex 2, which
was spontaneously organized by complementary coordi-
nation of imidazolylporphyrinatozinc motif to give an
assembly X. The molecular size distribution of zinc gable
porphyrin assembly X was analyzed by gel permeation
chromatography (GPC) and was observed as an oligo-
meric mixture with a broad distribution of the molecular
weight (Figure 1a). However, the elution curve was
totally different from that of meso-meso coupled porphy-
rin dimer zinc complex, which showed a sharp peak at
the exclusion limit (8 min) with long tailing, relevant to
the formation of giant linear arrays.7 The elution curve
of X showed obviously longer elution time and distinct
spikes at around 12.3 min, indicating the formation of
oligomeric mixtures of smaller molecular weights along
with an assembly of somewhat specific sizes. This result
suggests a positive sign that the terminal imidazolyl
group tends to find the zinc porphyrin counterpart at its
own chain end leading to intramolecular cyclization
rather than simply elongated zigzag chain.
FIGURE 1. Gel permeation chromatograms for the as-
semblies X (a) and Y (b) with a column exclusion volume of
7 × 104 Da. The curves were normalized at the peak of 12.3
min. Each eluent is chloroform and 150 µM solutions were
injected. The elution curve for X is shifted upward for clear
comparison. Arrows along the X-axis indicate molecular
weights of polystyrene standards.
arrays by adding and removing coordinating solvents,
respectively. During the process removing coordinating
solvents, new porphyrin arrangements different from the
original structures can be obtained by the choice of
appropriate conditions. We applied this reorganization
method to the zinc bis(imidazolyl) gable porphyrin as-
sembly X. To dominate the macrocyclization on recon-
structing the coordination structure, we applied the high-
dilution condition, which should favor the intramolecular
process rather than the intermolecular oligomerization.
The reorganization processes are as follows: (1) cleavage
of the coordinate bond and dilution; the assembly X was
dissolved in 5 µM of CHCl3/methanol ) 7/3 (v/v); (2)
further cleavage and dilution; methanol was added to
make a 3.5 µM solution of CHCl3/methanol ) 1/1 (v/v);
and (3) finally, the solvent was evaporated at 25 ( 1 °C.
The GPC chart of the sample after the reorganization
processes showed a dramatic change (Figure 1b).
Reorganization under High Dilution Conditions.
In the previous report, we have developed a reorganiza-
tion of the once formed structure to convert to other forms
for meso-meso directly linked bis(imidazolyl)porphyrin.7
This reorganization procedure is composed of cleavage
and reformation processes of self-assembled porphyrin
In the reorganized sample Y, the part due to larger
molecular weights was eliminated almost completely and
the peaks converged to two predominant peaks corre-
sponding to the two spike peaks originally existing at the
(10) Crossley, M. J.; Field, L. D.; Forster, A. J.; Harding, M. M.;
Sternhell, S. J. Am. Chem. Soc. 1987, 109, 341-348.
J. Org. Chem, Vol. 70, No. 7, 2005 2747

Takahashi and Kobuke
FIGURE 2. Gel permeation chromatograms (exclusion vol-
ume, 7 × 104 Da; eluent, chloroform) for assembly Y before
separation (thin line) and for the isolated two components (bold
lines) (a) 3 and (b) 4 after separation by using preparative gel
permeation chromatography.
smallest molecular weight wing with smaller contribu-
tions. These converged peaks correspond to 7-9 porphy-
rins, if one applies a calibration curve obtained from
linear meso-meso oligomers.7 However, cyclic molecules
are in general eluted later than the corresponding linear
molecules because of smaller molecular sizes as shown
not only from a theoretical consideration11 but also from
experimental results (e.g., poly(oxyethylene)s,12a poly-
carbazoles,12b and polystyrenes12c), and the observed
peaks may correspond to the closed structure of hexago-
nal and related macrorings of gable porphyrins.
These two peak components were easily separated by
using preparative GPC, employing toluene or benzene as
the eluent. Each fraction was concentrated and injected
again into the analytical column to confirm the clear
separation and stability without any scrambling each
other after the separation (Figure 2). In the following
discussion, the first and the second isolated components
are named 3 and 4, respectively.
Probe Microscopy. Atomic force microscope (AFM)
measurements of the assembly Y on flat mica substrate
demonstrated the presence of round-shaped particles
(Figure 3). Observation with further enlargement gave
round-shaped particles of a uniform height (ca. 1.5 nm),
which is characteristic of the height of the barrel-shaped
macroring. The size, after correction of the radius cur-
vature of the AFM probe (ca. 10 nm), provides the net
diameter of the particles as a few nanometers. These
particles are main components of the minimum size
corresponding to the single molecule of the barrel-shaped
macroring, although accompanied by larger circular spots
presumably of their aggregates. The formation of these
aggregates could not be avoided, because utilization of a
further diluted solution caused disaggregated species of
gable porphyrins in the deposition process. Since large
linear arrays such as observed for meso-meso coupled
porphyrin7 could not be detected, the reorganization
FIGURE 3. (a) AFM images of assembly Y (nearly 1:1 mixture
of 3 and 4) spin coated on mica substrate. (b) Enlarged results
of AFM measurement for the area framed by a square in (a).
(c) Cross-section profile of each particle appearing in (b). (d)
Height and width of each particle. The widths were derived
from half widths of the images without probe size correction.
processes converted completely the polymeric assemblies
X to a mixture of macrorings Y.
It should be noted that each component 3 and 4 is very
stable both in solid and in solution states unless coordi-
nating solvents are added. In AFM measurements,
particles were observed similarly after 2 months of
standing on the mica substrate, and observation on
highly oriented pyrolytic graphite (HOPG) substrate gave
also similar results (Figure 4). However, there was a
tendency that particles on HOPG substrate were ob-
served along the steps of the surface, which indicates a
different particle distribution under different conditions.13
STM measurements on HOPG substrate were at-
tempted for the sample Y, whose uniform particles were
confirmed by AFM measurements. Similar to the AFM
images, many particles tended to locate along the steps
of the HOPG surface in the STM measurements. Because
particles of Y were not stable on repeated scanning by
STM probe,13a we observed only a single molecular image
of low resolution (Figure 5). Although the observed
(11) Skvortsov, A. M.; Gorbunov, A. A. J. Chromatogr. 1990, 507,
487-496.
(12) (a) Yu, G.-E.; Sinnathamby, P.; Prise, C.; Booth, C. Chem.
Commun. 1996, 31-32. (b) Zhang, Y.; Wada, T.; Sasabe, H. Chem.
Commun. 1996, 621-622. (c) Lepoittevin, B.; Dourges, M.-A.; Masure,
M.; Hemery, P.; Baran, K.; Cramail, H. Macromolecules 2000, 33,
8218-8224.
(13) (a) Han, W.; Li, S.; Lindsay, S. M.; Gust, D.; Moore, T. A.; Moore,
A. L. Langmuir 1996, 12, 5742-5744. (b) Milic, T. N.; Chi, N.; Yablon,
D. G.; Flynn, G. W.; Batteas, J. D.; Drain, C. M. Angew. Chem., Int.
Ed. 2002, 41, 2117-2119. (c) Yin, J.; Guo, Q.; Palmer, R. E.; Bampos,
N.; Sanders, J. K. M. J. Phys. Chem. B 2003, 107, 209-216.
2748 J. Org. Chem., Vol. 70, No. 7, 2005

Hexameric and Pentameric Slipped-Cofacial Dimers
FIGURE 4. (a) AFM image of assembly Y spin coated on
HOPG substrate. (b) Enlarged image of Y. (c) Cross-section
profile of particles along the step of HOPG substrate in (b).
molecular image was composed of disconnected gable
porphyrins, this image was in good agreement with the
size of the macroring of gable porphyrins.
FIGURE 5. (a) STM image of Y spin coated on HOPG
substrate. (b) Birds-eye view of image (a). Inset shows the
schematic structure of hexameric gable porphyrin 3 drawn at
the same scale.
Synchrotron Solution Small-Angle X-ray Scatter-
ing. Using the above probe microscopies, the macroring
structure of gable porphyrins was suggested to exist on
solid surface. Unfortunately, we could not determine the
specific molecular weight by MALDI-TOF MS typically
used for covalently linked organic compounds. All GPC
fractions gave predominantly a monomer peak associated
with a small dimer peak and trace higher oligomer ones.
We finally applied small-angle X-ray scattering (SAXS)
measurements for the isolated components, 3 and 4, with
a synchrotron radiation equipment. From SAXS mea-
surements, we obtained information on (1) global size
from the q2-ln I(q) plot (Guinier plot) and (2) molecular
shape information from q-ln I(q) plot, where q and I(q)
are scattering vectors in Å-1, q ) (2π sin θ)/λ, and
scattering intensity, respectively.
The results of SAXS measurements for the first
component 3 from GPC analysis with synchrotron radia-
tion are as follows. The plot of scattering intensity vs
square scattering vector (Guinier analysis) provided a
radius of gyration (Rg) of 15.59 ( 0.34 Å for the predomi-
nant (98.0%) component (Figure 6a). This Rg value
corresponds to diameters of 42 and 40 Å according to
sphere and cylinder approximations, respectively. These
values agree well with the estimation of ca. 41 Å for the
outer diameter of the cyclic hexamer from molecular
mechanics calculation with a universal force field (UNI-
VERSAL 1.02).14 Furthermore, the particle size distribu-
tion was evaluated by the Funkuchen analysis, and the
remaining 2% component gave Rg ) 49 Å, which might
correspond to particles of higher ordered aggregates.
Similarly to 3, SAXS measurements of the second com-
ponent 4 from the GPC analysis were performed with
synchrotron radiation. The Guinier analysis gave Rg )
11.12 ( 0.66 Å, and this Rg value corresponded to
diameters of 29 Å according to the cylinder approximation
(Figure 6b). Unfortunately, these SAXS results of 4 might
be relatively less reliable compared to those of 3, because
the scattering intensity was relatively low.
To analyze further the SAXS results, we compared the
experimental values with theoretical estimates by using
molecular mechanics calculations by Cerius2 with a
universal force field14 and a program package for SAXS
analysis, CRYSOL.15 The representative models are
shown in Figure S2 (Supporting Information) with their
theoretical Rg values. Three possible cyclic hexamers with
C2, C3, and C6 symmetries were considered in the
calculation, and these isomers have similar Rg values of
(14) Rappe´, A. K.; Casewit, C. J.; Colwell, K. S.; Goddard, W. A.,
III; Skiff, W. M. J. Am. Chem. Soc. 1992, 114, 10024-10035.
(15) Svergun, D. I.; Barberato, C.; Koch, M. H. J. J. Appl. Crystal-
logr. 1995, 28, 768-773.
J. Org. Chem, Vol. 70, No. 7, 2005 2749

Takahashi and Kobuke
FIGURE 7. q-ln I(q) plot for SAXS measurements in methyl
benzoate: (a) 3 (2.2 mg/mL) and (b) 4 (2.0 mg/mL) (solid
curves). The theoretical curves for the cyclic hexamer, pen-
tamer, and tetramer are indicated as broken, dash-dotted,
and dotted lines, respectively.
FIGURE 6. Gunier plot for SAXS measurements in methyl
benzoate: (a) 3 (2.2 mg/mL) and (b) 4 (2.0 mg/mL), where q is
the scattering vector in Å-1 and I(q) is the scattering intensity.
with C6, C3, and C2 symmetries gave similar scattering
profiles, and we could not distinguish among them as
same as the Rg values. On the other hand, distinct
differences were observed among the cyclic hexamer,
cyclic pentamer, and cyclic tetramer, all with waved
scattering profiles showing the first minimum peaks at
0.18, 0.23, and 0.30 Å-1, respectively. It should be
noteworthy that these waved scattering profiles were not
observed for linear oligomers, which gave monotonic
profiles. Therefore if we observe waved profiles for q-ln
I(q) plots, it might be regarded as characteristics of the
hollow cylinder structures.
16.54, 16.14, and 16.67 Å, respectively. The Rg values for
the cyclic pentamer and tetramer gave 13.52 and 10.55
Å, respectively. The obtained Rg value for 3 is between
the cyclic hexamer and cyclic pentamer, and that for 4
is between the cyclic pentamer and cyclic tetramer. It
should be noted that linear arrays have Rg values
distinctly larger than those of the corresponding cyclic
arrays, 28.34, 23.76, and 19.43 Å for the linear hexamer,
pentamer, and tetramer, respectively. Combining these
results with GPC and reorganization behaviors, 3 and 4
may best be assigned as a cyclic hexamer and a cyclic
pentamer, respectively.
Furthermore, the scattering profile in a wide-angle
region was compared with theoretical models. This region
gives information about the molecular shape from q-ln
I(q) plots.16 Theoretical q-ln I(q) plot for the model
structures are shown in Figure S3. The cyclic hexamers
As expected from Figure S3, the scattering profile in a
wide-angle region for both 3 and 4 gave waved profiles
characteristic for the hollow cylinder structure (Figure
7). The first minimum peak of 3 appeared at 0.17 Å-1
followed by a rise of the intensity. This scattering
intensity plot was expressed by a theoretical calculation15
best for the cyclic hexamer with a minimum at 0.18 Å-1
,
(16) Pickover, C. A.; Engelman, D. M. Biopolymers 1982, 21, 817-
831.
in contrast to other cyclic oligomers, the minimum being
2750 J. Org. Chem., Vol. 70, No. 7, 2005

Hexameric and Pentameric Slipped-Cofacial Dimers
0.23 and 0.30 Å-1 for the pentamer and tetramer,
respectively (Figure 7a) We conclude therefore at this
point that 3 is a hexameric macroring. In the case of 4,
scattering intensity was relatively low and the first
minimum peak was observed around 0.18-0.26 Å-1
(Figure 7b). Because this value roughly corresponds to
the cyclic pentamer, we tentatively assign that 4 is a
pentameric macroring.17
SCHEME 2. Schematic Representation of
Reorganization Process under High Dilution
Conditions and Further Separation Process into
Hexameric Macroring 3 and Pentameric
Macroring 4 of Gable Porphyrins
Mechanism of Reorganization Process. Summa-
rizing the results described above, we could explain the
components generated during the reorganization and
separation processes as shown in Scheme 2. The reorga-
nization processes under high dilution conditions are
considered to proceed as follows. The dissociated gable
porphyrin tries to find the counterpart, assisted by the
large association constant of slipped-cofacial motifs. The
high dilution condition, however, makes it difficult to find
the counterpart in an intermolecular fashion in the
solution. The small linear oligomer thus formed can find
the counterpart at the other terminal in its own assembly
to satisfy mutually the demand of complementary coor-
dination. The most stable macroring must be composed
of a hexameric unit since the m-gable motif with an
interporphyrin angle of 120° is accommodated without
angle strain. A pentameric assembly is also expected to
form under forced conditions of high dilution if the angle
strain is shared by the components. The heptameric
species may be produced less favorably considering that
the dynamic process starts from the lower unit sizes in
the reorganization in accord with the experimental
observations. It is important to note that the peak at 12.3
min constantly exists even after the first zinc insertion
reaction under concentrated conditions (Figure 1), and
this component matches the idea of the most thermody-
namically stable product. On the other hand, the peak
at 12.6 min became one of the major components after
the reorganization and must be forced to generate during
the reorganization process under high dilution conditions.
Considering that (1) there are three atropisomers in
the m-gable motif and (2) once paired, slipped-cofacial
motifs are difficult to dissociate, the observed convergent
change by the reorganization process is very interesting.
It must be noted that atropisomers can find their own
partners to lead to the formation of three geometrical
isomers of different symmetries, C2, C3 and C6, for the
hexamer 3 without any steric strain (Figure S2). If steric
strain is allowed for the formation of macrocycles, isomers
resulting from out and in coordination modes are allowed.
The latter possibility may be favored considering the fact
that macrocyclic pentamer 4, where only one symmetry-
allowed product is expected, was obtained in a yield even
competitive with the hexamer.
Absorption Spectra. In view of the C2 symmetry of
the porphyrin chromophore, it may be that the electronic
transition moment in a series of imidazolylporphyrins
runs in the meso-meso direction. In fact, UV-vis absorp-
(17) We recently developed a method in which the complementary
coordinated pair was connected covalently to each other by ring-closing
metathesis reactions between meso-olefinic substituents.5c These mac-
rocycles with meso-olefinic substituents, before the metathesis reaction,
have shown behavior toward reorganization process and GPC analysis
exactly the same as those of 3 and 4 and gave, after the metathesis,
the exact molecular ion peaks corresponding to hexameric and pen-
tameric gable porphyrins, respectively, by MALDI-TOF mass spectrom-
etry.5b
tion spectra of proximal porphyrins caused a splitting of
the Soret band region for both slipped-cofacial and
phenylene-bridged motifs. In the case of the slipped-
cofacial motif, the splitting of the Soret bands might be
qualitatively explained by the exciton coupling theory.18
When this theory is applied to the slipped-cofacial
J. Org. Chem, Vol. 70, No. 7, 2005 2751

Takahashi and Kobuke
This qualitative analysis of the absorption spectra
indicates that electronic interactions between the closed
porphyrins lead to perturbation of the Soret bands but
not the Q-band absorptions when the pigments are linked
to form the array. In the point-dipole approximation, the
exciton coupling theory indicates that the interaction
between chromophores should be inversely proportional
to the cube of the interchromophoric distance and pro-
portional to the square of the transition moments of the
interacting chromophores. Thus, long-range exciton cou-
pling in porphyrin arrays should only be detected for the
Soret bands of higher oscillator strength and not for the
Q-bands of a lower one, as is the case. In chloroform
solution, UV-vis absorption spectra of hexamer 3 and
pentamer 4 gave large splits of the Soret bands, 2073
and 2003 cm-1, respectively. These correspond to the sum
of each contribution of the splitting energy from slipped-
cofacial and phenylene-bridged interactions,18 1310 and
775 cm-1, respectively, the values of dimer 5 and of the
monomeric bis-zinc gable porphyrin 2 (measured in the
presence of excess 1-methylimidazole to dissociate the
self-assembly).
Fluorescence Spectra. Fluorescence emission spec-
tra of the macrorings (Y, 3, and 4) are similar to that of
slipped-cofacial dimer 5, reflecting the unperturbed Q-
bands absorption. In contrast to the fact that a red-
shifted and strongly quenched fluorescence was observed
for artificial cofacial zinc dimers,19 the macroring mixture
Y showed fluorescence emission with nearly the same
quantum yield (0.035) as those of tetraphenylporphy-
rinatozinc(II) (ZnTPP) (0.03)20 as the reference. This
fluorescence property ensures that the macrorings of
gable porphyrins represent a good model for the artificial
light-harvesting complex.
Conclusion
Combined evidence was consistent with the construc-
tion of porphyrin macrorings by interlocking 1,3-phen-
ylene gable porphyrins by slipped-cofacial dimer forma-
tion without any protein matrices. This model must be a
major milestone for further investigation to elucidate the
mechanism of highly efficient light-harvesting in the
natural photosynthetic system, as well as the facile
strategy of such ring structure formation.
The principal finding of this study is the successful
application of the reorganization methods that convert
the once formed structure into desired structures, mac-
roring arrays in this case. Oligomeric mixtures were
converted almost completely to ring structures under
high-dilution conditions by interlocking 1,3-phenylene
gable porphyrins. This may be regarded as a typical
advantage of supramolecular methodologies of structure
formation. Various methods of application of such struc-
tural transformation may be designed. Not only the
structure formation from only one component but also
the addition of other components to the system must also
be interesting.
FIGURE 8. UV-vis absorption spectra of the macroring Y
in benzene at room temperature and its exciton splitting
modes.
arrangement, the degenerated Soret transitions, B| and
B , in the monomeric unit are red- and blue-shifted
depending on the head-to-tail and face-to-face orienta-
tions of the transition dipoles m| and m , respectively.
In the case of the phenylene-bridged motif, the splitting
of the Soret bands might be explained similarly. The red-
shifted Soret transition B| might arise from the interac-
tion between the transition dipoles m| on the head-to-
tail orientation, and the other unperturbed Soret transition
B in the monomeric unit might be derived from both the
head-to-head (tail-to-tail) orientations of the transition
dipoles m| and the unperturbed transition dipoles m .
UV-vis absorption spectrum of the macroring Y (hex-
amer:pentamer ≈1:1) solution gave a large splitting of
the Soret bands and relatively unaffected Q-bands com-
pared to the simple slipped-cofacial imidazolyl-zinc por-
phyrin dimer 5 (Figure 8).
The second important point is that the macroring
arrays are candidates as light-harvesting antenna mim-
(19) Leighton, P.; Cowan, J. A.; Abraham, R. J.; Sanders, J. K. M.
J. Org. Chem. 1988, 53, 733-740.
(20) Seybold, P. G.; Gouterman, M. J. Mol. Spectrosc. 1969, 31,
1-13.
(18) Kasha, M.; Rawls, H. R.; El-Bayoumi, M. A. Pure Appl. Chem.
1965, 11, 371-392.
2752 J. Org. Chem., Vol. 70, No. 7, 2005

Hexameric and Pentameric Slipped-Cofacial Dimers
chloroform or benzene was freshly prepared. A droplet of 1
µL of solution of Y was deposited on the substrate and spun
at ca. 2,000 rpm for 10 s. An additional droplet of 4 µL of
solution of Y was deposited on the substrate and spun at ca.
3,000 rpm for 10 s. The substrate was sprayed with nitrogen
gas for 1 min and dried. Typical AFM parameters were as
follows: rate of amplitude attenuation ) -0.314 to -0.063;
integral gain ) 0.2273 to 0.8000; proportional gain ) 0.0415
to 0.4980; scanning area ) 5,000 to 175 nm; scanning
frequency ) 1.00 to 1.75 Hz.
STM measurements were performed with PtIr (80/20) tips.
Measurements of Y on HOPG substrate were performed
after verification of uniform particle images by preliminary
AFM measurements. Typical STM parameters were as fol-
lows: gain ) 1010 V/A for the high current mode, 1011 V/A for
the low current mode; filter ) 4.0 kHz; tip velocity ) 0.201 to
0.155 µm/s; current set point ) 500.0 to 3.00 pA; integral
gain ) 0.4000 to 1.0000; proportional gain ) 0.3500 to 1.0000;
bias ) -2,000 to 2,000 mV; scanning area ) 100 to 10 nm;
scanning rate ) 4.07 to 7.77 Hz.
ics. In general it is difficult to assemble donor molecules
in a highly condensed state because of formation of
undesired energy dissipation processes, e.g., concentra-
tion quenching, aggregation quenching, and nondirec-
tional energy transfer, in homogeneous solution. How-
ever, the arrangement of chromophores in our macroring
systems did not show any sign of fluorescence quenching
by assembling in the ring structures, notwithstanding
close interchromophoric distances, just as natural an-
tenna chlorophyll molecules do. These findings must be
correlated with the relation between the structure and
function as seen in the natural light-harvesting systems.
Elucidation of photophysical properties, especially energy
migrating processes, will soon be reported.21
A further arrangement of ring components into a two-
dimensional matrix such as a thylakoid membrane will
be interesting, and the introduction of energy accepting
units in the central cavity will also be of utmost impor-
tance. These challenges along with construction of novel
macroring analogues are under active investigation in
our laboratory. Our applications by learning plenty from
nature must be beyond the natural systems and ap-
plicable for nanoscaled molecular devices in near future.
Synchrotron SAXS measurements were performed by using
synchrotron radiation at the solution scattering station (SAX-
ES camera) installed at BL-10C, the Photon Factory, Tsukuba,
Japan.22
Acknowledgment. We thank Prof. M. Kataoka and
his laboratory members for technical assistance and
helpful discussions with the SAXS measurements. This
work was supported by CREST (Core Research for
Evolutional Science and Technology) of the Japan
Science and Technology Corporation (JST).
Experimental Section
Characterization of Porphyrin Supramolecules. AFM
measurements were performed with Si tips (351 kHz, 42 N/m)
for DFM (dynamic force mode) and tapping mode. Mica and
highly oriented pyrolytic graphite (HOPG) were used as
substrates. Typically, a 10 µM solution of assembly Y in
Supporting Information Available: Experimental de-
tails for synthesis of dipyrromethanes and porphyrins, Eyring
plot for variable temperature NMR, Guinier plot for synchro-
tron SAXS, theoretical calculations, and other experimental
conditions. This material is available free of charge via the
Internet at http://pubs.acs.org.
(21) Hwang, I.-W.; Ko, D. M.; Ahn, T. K.; Kim, D.-H.; Ito, F.;
Ishibashi, Y.; Khan, S. R.; Nagasawa, Y.; Miyasaka, H.; Ikeda, C.;
Takahashi, R.; Ogawa, K.; Satake, A.; Kobuke, Y. Submitted for
publication.
(22) Ueki, T.; Hiragi, Y.; Kataoka, M.; Inoko, Y.; Amemiya, Y.; Izumi,
Y.; Tagawa, H.; Muroga, Y. Biophys. Chem. 1985, 23, 115-124.
JO047923N
J. Org. Chem, Vol. 70, No. 7, 2005 2753