Assemblies of supramolecular porphyrin dimers in pentagonal and hexagonal arrays exhibiting light-harvesting antenna function

Article abstract of DOI:10.1021/ja0583214

Porphyrin-based supramolecular macrocyclic arrays were synthesized as mimics of photosynthetic light-harvesting (LH) antennae. Pentameric and hexameric macrocyclic porphyrin arrays EP5 and EP6 were constructed by complementary coordination of m-bis(ethyny

Full text of DOI:10.1021/ja0583214

Published on Web 03/18/2006
Assemblies of Supramolecular Porphyrin Dimers in
Pentagonal and Hexagonal Arrays Exhibiting Light-Harvesting
Antenna Function
Fatin Hajjaj,† Zin Seok Yoon,‡ Min-Chul Yoon,‡ Jaehong Park,‡ Akiharu Satake,†
Dongho Kim,*,‡ and Yoshiaki Kobuke*,†
Contribution from the Graduate School of Materials Science, Nara Institute of Science and
Technology, Takayama 8916-5, Ikoma, Nara 630-0192, Japan, and Department of Chemistry
and Center for Ultrafast Optical Characteristics Control, Yonsei UniVersity,
Seoul 120-749, Korea
Received December 7, 2005; E-mail: dongho@yonsei.ac.kr; kobuke@ms.aist-nara.ac.jp
Abstract: Porphyrin-based supramolecular macrocyclic arrays were synthesized as mimics of photosynthetic
light-harvesting (LH) antennae. Pentameric and hexameric macrocyclic porphyrin arrays EP5 and EP6
were constructed by complementary coordination of m-bis(ethynylene)phenylene-linked zinc-imidazolylpor-
phyrin Zn-EP-Zn. The proton NMR spectra of noncovalently linked N-EP5 and N-EP6 indicate fast rotation
of the porphyrin moieties along the ethyne axis. These macrocycles were covalently linked and identified
as C-EP5 (6832 Da) and C-EP6 (8199 Da) by mass spectrometry. Fluorescence quantum yields of C-EP2
(10.0%), C-EP5 (10.1%), and C-EP6 (11.0%), even larger than that of the unit coordination dimer C-EP1
(9.3%), were significantly increased from those of the series without the ethynylene linkage. The order of
increasing fluorescence quantum yields was parallel to that of decreasing fluorescence lifetimes (C-EP1
(1.65 ns), C-EP2 (1.45 ns), C-EP5 (1.42 ns), and C-EP6 (1.38 ns)), indicating that the radiative decay rate
kF increased relative to the other decay rates with an increase in the number of ring components. Based
on the exciton-exciton annihilation and anisotropy depolarization times, the excitation energy hopping
(EEH) times in these macrocyclic systems were obtained as 21 ps for C-EP5 and 12.8 ps for C-EP6. EEH
times depend strongly on the orientation factor of the component transition dipoles in the macrocyclic arrays.
The hexagonal macrocyclic array with an orientation of better transition dipole coupling resulted in faster
EEH time compared to the pentagonal one.
Introduction
of the circle and in transferring the energy to neighboring
antenna molecules in the ring and further to other rings by
excited energy hopping (EEH) processes.
Photosynthetic light-harvesting (LH) antennae are efficient
molecular devices for catching dilute sunlight energy and
transferring the excitation energy to the reaction center through
an antenna network. Purple bacteria adopt macrocyclically the
LH systems LH1 and LH2, in which, respectively, 30 and 16
bacteriochlorophylls (BChls) are arranged in circular arrays.1
Circular structures are favored in forming densely packed two-
dimensional arrays in biological membranes.2 They are effective
in absorbing light from any direction due to the isotropic nature
Synthetic models of LH antennae without the protein matrix
are useful for exploring photochemical solar energy conversion
and other optoelectronic devices operating at the molecular level.
Porphyrin-based LH antennae have been investigated thoroughly
due to their close relation to the chlorophyll system with respect
to structure and photophysical properties.3 The fabrication and
characterization of macrocyclic porphyrin arrays to resemble
LH complexes in purple bacteria, however, are rather limited.
So far, the researches on a diphenylbutadiyne-linked cyclic
tetramer,4 m-bis(phenylethynyl)phenylene-linked cyclic hex-
amer,5 and diphenylethyne-linked cyclic hexamer6 have been
reported. Recently, m-phenylene- and meso-meso-linked cyclic
dodecamers, in which 12 porphyrins are closely connected, were
reported to exhibit fast EEH.7
† Nara Institute of Science and Technology.
‡ Yonsei University.
(1) (a) Hu, X.; Damjanovic´, A.; Ritz, T.; Schulten, K. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 5935-5941. (b) Green, B. R., Parson, W. W., Eds. Light-
HarVesting Antennas in Photosynthesis; Advances in Photosynthesis and
Respiration series, Vol. 13; Kluwer Academic Publishers: Dordrecht, The
Netherlands, 2003. (c) 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. (d) Koepke, J.; Hu, X.; Muenke, C.;
Schulten, K.; Michel, H. Structure 1996, 4, 581-597. (e) Roszak, A. W.;
Howard, T. D.; Southall, J.; Gardiner, A. T.; Law, C. J.; Isaacs, N. W.;
Cogdell, R. J. Science 2003, 302, 1969-1972.
Self-assembling into macrocyclic porphyrin arrays through
noncovalent bonds is challenging from a viewpoint of synthetic
accessibility to nature’s methodology. Cyclic tetramers using
hydrogen bonds8 and cyclic dodecamers formed through axial
coordination of substituted pyridines to six-coordinated cobalt
(2) (a) Scheuring, S.; Sturgis, J. N.; Prima, V.; Bernadac, A.; Le´vy, D.; Rigaud,
J.-L. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 11293-11297. (b) Bahaty-
rova, S.; Frese, R. N.; Siebert, C. A.; Olsen, J. D.; van der Werf, K. O.;
van Grondelle, R.; Niederman, R. A.; Bullough, P. A.; Otto, C.; Hunter,
C. N. Nature 2004, 430, 1058-1062.
9
4612
J. AM. CHEM. SOC. 2006, 128, 4612-4623
10.1021/ja0583214 CCC: $33.50 © 2006 American Chemical Society

Porphyrin Dimers in Pentagonal and Hexagonal Arrays
A R T I C L E S
porphyrins9 have been demonstrated. Although the latter
structurally resembles natural LH2, no LH function is expected
from nonfluorescent cobalt porphyrins.
Here, we report the efficient synthesis of noncovalently linked
N-EP5 and N-EP6 and covalently linked C-EP5 and C-EP6.
1
Their dynamic behaviors are examined by H NMR, and LH
We have recently developed a complementary coordination
of bis(zinc-imidazolylporphyrin) to construct macrocyclic por-
phyrin arrays.10 When two zinc-imidazolylporphyrins are linked
with an appropriate spacer, various macrocyclic porphyrin arrays
can be obtained in good yields. In the coordination dimer
(hereafter called as “dimer”), two porphyrins form a slipped-
cofacial structure, like the chlorophyll dimer in LH2, and EEH
in the dimer occurs very rapidly on a subpicosecond time scale.11
In the macrorings, a different EEH process across the covalently
connected spacer was observed.
Previously, we reported the first macrocyclic pentamer P5
and hexamer P6 supramolecules from bis(zinc-imidazolylpor-
phyrin) linked through a m-phenylene moiety.10a,11 In these rings,
because porphyrin and phenylene planes are orthogonal to each
other, the π-conjugation between porphyrins is prohibited. In
the present work, we have explored the supramolecular forma-
tion of the cyclic pentamer EP5 and hexamer EP6 from bis-
(zinc-imidazolylporphyrin) linked by a m-bis(ethynyl)phenylene
spacer. Introduction of the ethyne moiety between porphyrin
and phenylene allows free rotation along the ethyne axis, and
an intimate electronic communication among porphyrins would
occur as a result of the coplanar conformation. In addition, the
transition dipole of porphyrin along the axis of π-conjugation
is increased, as observed for the red-shift and increased intensity
of the Q-bands, otherwise symmetry forbidden. In natural LH
systems, vectorial energy flow is realized by arranging multiple
LH units, such as LH2 and LH1 in purple bacteria. In this
context, EP5 and EP6 can be potential candidates for accepting
high-excitation energy from P5 and P6.
antenna functions are evaluated by steady-state absorption,
emission, and time-resolved transient absorption techniques. We
compare the LH functions with those of P5 and P6 and discuss
the effect of ring size and orientation factor in the EEH process.
Results
Synthesis of N-EP5, N-EP6, C-EP5, and C-EP6. We have
examined the self-assembly of macrocyclic porphyrin arrays
from various bis(zinc-imidazolylporphyrin)s. If an appropriate
bisporphyrin unit is prepared, the desired porphyrin array can
be produced in good yields by the reorganization technique.
EP5 and EP6 require the bisporphyrin unit Zn-EP-Zn. In our
first attempt at preparing Zn-EP-Zn, the double Sonogashira
coupling12 of 1,3-diiodo-benzene with porphyrin 2 was exam-
ined. When porphyrin 2 (2 equiv) and 1,3-diiodo-benzene (1
equiv) were reacted in the presence of the catalyst Pd2(dba)3/
AsPh3, the desired H2-EP-H2 was obtained along with 4, H2-
EP, and H2-3 with the recovery of 2. Oxidative coupling product
4 generally accompanies the palladium catalyzed coupling
reaction of ethynylporphyrins. Unfortunately, H2-EP-H2 could
not be separated from 4 effectively by various chromatographical
techniques (SiO2, Al2O3, and GPC) owing to the similarity of
polarities and molecular volumes. Finally, we have established
the synthetic scheme for Zn-EP-Zn shown in Scheme 1. First,
the monocoupling product H2-3 was prepared in 78% yield from
2 with an excess 1,3-diiodo-benzene in the presence of the
catalyst Pd2(dba)3/AsPh3. After zinc insertion into H2-3, the
coupling reaction of Zn-3 with 2 gave H2-EP-Zn as the
dominant product along with H2-EP and 4. This time, H2-EP-
Zn was isolated successfully in 50% yield by gel permeation
chromatography. H2-EP was also isolated in 10% yield, and
its zinc complex Zn-EP was used as a reference compound for
steady state and transient absorption measurements.
(3) (a) Harvey, P. D. In The Porphyrin Handbook; Kadish, K. M., Smith, K.
M., Guilard, R., Eds.; Academic Press: New York, 2003; Vol. 18, pp 63-
250. (b) Kim, D.; Osuka, A. Acc. Chem. Res. 2004, 37, 735. (c) Holten,
D.; Bocian, D. F.; Lindsey, J. S. Acc. Chem. Res. 2002, 35, 57-69. (d)
Choi, M.-S.; Yamazaki, T.; Yamazaki, I.; Aida, T. Angew. Chem., Int. Ed.
2004, 43, 150-158. (e) Imahori, H. J. Phys. Chem. B 2004, 108, 6130-
6143. (f) Satake, A.; Kobuke, Y. Tetrahedron 2005, 61, 13-41. (g) Balaban,
T. S. Acc. Chem. Res. 2005, 38, 612-623. (h) Prathapan, S.; Johnson, T.
E.; Lindsey, J. S. J. Am. Chem. Soc. 1993, 115, 7519-7520. (i) Nakano,
A.; Osuka, A.; Yamazaki, I.; Yamazaki, T.; Nishimura, Y. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1096. (j) Maruo, N.; Uchiyama, M.; Kato, T.; Arai,
T.; Akisada, H.; Nishino, N. Chem. Commun. 1999, 2057. (k) Yeow, E.
K. L.; Ghiggino, K. P.; Reek, J. N. H.; Crossley, M. J.; Bosman, A. W.;
Schenning, A. P. H. J.; Meijer, E. W. J. Phys. Chem. B 2000, 104, 2596-
2606. (l) Choi, M.-S.; Aida, T.; Yamazaki, T.; Yamazaki, I. Angew. Chem.,
Int. Ed. 2001, 40, 3194. (m) Aratani, N.; Osuka, A.; Kim, Y. H.; Jeong, D.
H.; Kim, D. Angew. Chem., Int. Ed. 2000, 39, 1458. (n) Ogawa, K.; Kobuke,
Y. Angew. Chem., Int. Ed. 2000, 39, 4070. (o) Furutsu, D.; Satake, A.;
Kobuke, Y. Inorg. Chem. 2005, 44, 4460-4462. (p) Drain, C. M.; Goldberg,
I.; Sylvain, I.; Falber, A. Top. Curr. Chem. 2005, 245, 55. (q) You, C.-C.;
Dobrawa, R.; Saha-Mo¨ller, C. R.; Wu¨rthner, F. Top. Curr. Chem. 2005,
258, 39.
Zinc insertion into H2-EP-Zn afforded the bis-zinc complex
Zn-EP-Zn, which spontaneously organized into oligomers by
the complementary coordination of imidazolyl residues to zinc
centers.13 GPC analysis of the as-prepared (pristine) Zn-EP-
Zn oligomers showed a broad molecular weight distribution with
sharp spikes at longer retention times (Figure S1, red curve).
The pattern of the first chromatogram was close to those
obtained in the cases of P5 and P6,10a suggesting that the
formation of specific cyclic oligomers was accompanied by the
formation of a significant amount of linear oligomers.
To dominate the intramolecular cyclization, the Zn-EP-Zn
oligomeric mixture was reorganized as follows. A 20.0 µM
solution of the mixture was prepared in CHCl3/MeOH ) 10/1
(v/v), and the solution was allowed to stand at rt for 1 day.
During this period, conversion into N-EP5 and N-EP6 pro-
ceeded in good yield. The solvents were then evaporated at 26
°C under reduced pressure as quickly as possible. The evapora-
tion process must be controlled carefully. During the concentra-
tion process, the solvent becomes methanol-rich, and the dimer
tends to dissociate. The cyclic assemblies dissociated are
transformed into polymeric assemblies at higher concentra-
(4) Anderson, S.; Anderson, H. L.; Bashall, A.; McPartlin, M.; Sanders, J. K.
Angew. Chem., Int. Ed. Engl. 1995, 34, 1096-1099.
(5) Mongin, O.; Schuwey, A.; Vallot, M.-A.; Gossauer, A. Tetrahedron Lett.
1999, 40, 8347.
(6) Tomizaki, K.-Y.; Yu, L.; Wei, L.; Bocian, D. F.; Lindsey, J. S. J. Org.
Chem. 2003, 68, 8199-8207.
(7) (a) Hwang, I.-W.; Ko, D. M.; Ahn, T. K.; Yoon, Z. S.; Kim, D.; Peng, X.;
Aratani, N.; Osuka, A. J. Phys. Chem. B 2005, 109, 8643-8651. (b) 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.
(8) Shi, X.; Barkigia, K. M.; Fajer, J.; Drain, C. M. J. Org. Chem. 2001, 66,
6513-6522.
(9) Haycock, R. A.; Hunter, C. A.; James, D. A.; Michelsen, U.; Sutton, L. R.
Org. Lett. 2000, 2, 2435-2438.
(10) (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)
Kuramochi, Y.; Satake, A.; Kobuke, Y. J. Am. Chem. Soc. 2004, 126,
8668-8669. (d) Shoji, O.; Okada, S.; Satake, A.; Kobuke, Y. J. Am. Chem.
Soc. 2005, 127, 2201-2210. (e) Takahashi, R.; Kobuke, Y. J. Org. Chem.
2005, 70, 2745-2753.
(12) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975, 16,
4467-4470.
(13) Kobuke, Y.; Miyaji, H. J. Am. Chem. Soc.1994, 116, 4111-4112.
(11) Kim, D.; Miyasaka, H.; Kobuke, Y.; et al. Chem.sEur. J. 2005, 11, 3753.
9
J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006 4613

A R T I C L E S
Hajjaj et al.
Figure 1. Structures of macrocyclic porphyrin arrays.
tions.10a,d,e,14 Evaporation under these circumstances increases
the amount of oligomer and, thus, must be performed quickly.
The use of large volumes should be avoided for a reason. An
analytical GPC trace of the sample reorganized from Zn-EP-
Zn (Figure S1, blue curve) showed two converged peaks,
putative cyclic hexamer and pentamer N-EP6 and N-EP5,
respectively, and the higher molecular weight components
disappeared. N-EP6 and N-EP5 were separated using prepara-
tive GPC, and their purities were checked by analytical GPC
(Figure S2).
The isolated components are stable in solid form and in
noncoordinating solvents such as CHCl3 (∼10 µM) and toluene.
The dimer units in the isolated assemblies were covalently linked
by olefin metathesis reactions. Each covalently linked macroring
was purified by preparative GPC with CHCl3/MeOH (10/1) as
eluent, and its purity was checked by analytical GPC (Figure
S3). The metathesis reaction enables the detection of molecular
ion peaks of m/z 6832 and 8199, corresponding to the covalently
linked pentamer C-EP5 and hexamer C-EP6, respectively.
Covalent linkage also allows photophysical measurements even
in coordinating solvents such as pyridine.15 As reference
molecules, porphyrins C-EP1 and C-EP2 were prepared by
mixing Zn-EP and Zn-EP-Zn in pyridine followed by evapora-
tion and metathesis reaction. C-EP1 and C-EP2 were isolated
by recycling GPC (Figure S6).
Structural Analysis by 1H NMR. Proton NMR analysis of
N-EP5, N-EP6, H2-3, and Zn-3 was carried out in CDCl3 at
25 °C (Figure 3). Under these conditions, zinc-imidazolylpor-
phyrin exists solely as a dimer. Protons Im5, Im4, N-Me, â1,
and â2 of Zn-3 are upfield shifted considerably to 5.51, 2.09,
1.69, 5.42, and 8.90 ppm, respectively, compared to the
corresponding peaks of H2-3. By contrast, â3 and â4 protons of
Zn-3 are downfield shifted as a result of the deshielding effect.
(14) (a) Chi, X.; Guerin, A. J.; Haycock, R. A.; Hunter, C. A.; Sarson, L. D. J.
Chem. Soc., Chem. Commun. 1995, 2563. (b) Ercolani, G. J. Phys. Chem.
B 1998, 102, 5699.
(15) Ohashi, A.; Satake, A.; Kobuke, Y. Bull. Chem. Soc. Jpn. 2004, 77, 365-
374.
9
4614 J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006

Porphyrin Dimers in Pentagonal and Hexagonal Arrays
A R T I C L E S
Figure 2. Structures of reference porphyrins.
Scheme 1. Synthetic Scheme of H2-EP-Zn
This behavior is consistent with the previous observations of
zinc imidazolylporphyrin dimer.15
N-EP5 and N-EP6 support the formation of the dimer, and the
simplicity of 1H NMR suggests that all 10 or 12 porphyrins are
identical under the NMR conditions. Thus, inner and outer
protons of the macrocyclic arrays could not be discriminated.
Larger downfield shifts of phenylene protons Ha and Hc can be
reasonably explained by the doubly deshielding effect of both
sides of the dimers. The chemical shifts of Ha and Hc in,
Surprisingly, both N-EP5 and N-EP6 gave well-resolved 1H
NMR spectra. The spectra exhibited only a single set of
porphyrin protons, which resembles that of the starting dimer
Zn-3 except for more downfield shifted aromatic protons.
Similar upfield shifts of Im5, Im4, N-Me, â1, and â2 protons of
9
J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006 4615

A R T I C L E S
Hajjaj et al.
Figure 3. 1H NMR spectra of (A) H2-3, (B) Zn-3, (C) N-EP5, and (D) N-EP6 in CDCl3 at 25 °C. Asterisk (*) indicates impurity signals. Assignments are
indicated in Figure 4.
Scheme 2. Synthetic Scheme of C-EP5 and C-EP6
respectively, N-EP5 and N-EP6 indicate the high sensitivity to
spatial positions of the neighboring porphyrins. The slightly
larger downfield shift of Hc (9.12 ppm) observed in N-EP5
relative to that of N-EP6 (9.07 ppm) can be ascribed to the
stronger deshielding effect due to the smaller angle of the
pentagonal structure of N-EP5. On the other hand, proton Ha
in N-EP5 appears at a slightly upfield position (8.35 ppm)
compared to 8.38 ppm for N-EP6. The shift behaviors of Hc
and Ha protons are reasonably understandable on the basis of
the pentagonal and hexagonal geometries.
In the latter case, the inner and outer protons of the macrocyclic
arrays in the barrel form were discriminated because the inner
protons were more shielded by the ring current effect of nearby
porphyrins. Topological isomers were also observed with respect
to the coordination manner, indicating a slow exchange rate, if
any, among the isomers. By contrast, the dimer units in N-EP5
and N-EP6 were rotating fast on the NMR time scale along the
ethyne axis. Given that no noticeable changes were observed
in the NMR spectra at -50 °C in CDCl3, the rotational barrier
is presumably small even at -50 °C.
1
The observed H NMR spectra in Figure 3C and D sharply
contrast with those of macrocyclic porphyrin arrays of zinc-
imidazolylporphyrins connected by sterically hindered spacers.10c
Steady-State Absorption, Fluorescence, and Fluorescence
Excitation Anisotropy. The steady-state absorption spectra of
macrocyclic porphyrin arrays were measured in pyridine and
9
4616 J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006

Porphyrin Dimers in Pentagonal and Hexagonal Arrays
A R T I C L E S
Table 1. Absorption and Fluorescence Spectral Data
absorption
fluorescence
Q(0,0) emissionb
max/nm
Stokes shift
quantum yieldb,c
lifetimed
1
sample
solventa
soret/nm
Q(0,0)/nm
λ
cm-
ΦF
τ/ns
Zn-EP
C-EP1
Zn-EP-Zn
C-EP2
C-EP5
C-EP6
Zn-P
Py
Py
Py
Py
Py
Py
Py
Py
Py
Py
Tol
Tol
Tol
Tol
433.6
433.6
445.8
640.6
651.8
646.8
656.5
659.0
659.4
612.4
621.8
621
622
649.5
655
656.5
657
650.4
656.6
653.4
660.8
663.0
663.2
235
112
156
99
92
88
0.083
0.093
0.100
0.100
0.101
0.110
453.6
450.2
458.5
460.6
461.4
1.65
431
431
431
450
443
443
431.8
1.45
1.42
1.38
C-P1
415.8
410
410
433.5
433.5
431.5
431
440.0
447
449
451.5
458.5
459.0
460.5
C-P5e
425
425
626.5
627
654.8
658.8
659.2
659.4
0.035f
0.035f
0.093
0.095
0.095
0.100
2.18
2.19
C-P6e
Zn-EPg
C-EP2
N-EP5
N-EP6
124.6
88
62.4
55
442.5
442
a Py: pyridine. Tol: toluene. b Excited at 430 nm. c ZnTPP in pyridine (ΦF )0.038) was used as reference [ref 18]. d The λex ) 420 nm and λem ) 650
nm were applied for the isotropic fluorescence measurement. The fluorescence lifetimes were obtained by averaging the fitted parameters measured in
several emission wavelengthe (Figure 10). e Reference 11. f Data for N-P5 and N-P5 in toluene [ref 10e]. g Coordination dimer is formed.
formed, only the lower energy band is red-shifted by 7.8 nm,
whereas the higher energy band remains at the same wavelength.
This behavior sharply contrasts with that shown by C-P1, in
which the degenerate Soret band (431.8 nm) splits into two
bands at 415.8 and 440.0 nm due to excitonic dipole-dipole
coupling by slipped-cofacial coordination as indicated in Figure
5b. The negligible blue-shift in the case of C-EP1 may reflect
the small oscillator strength of By.16
Absorption spectra of Zn-EP-Zn, C-EP1, C-EP2, C-EP5,
and C-EP6 recorded in pyridine and Zn-EP, N-EP5, and N-EP6
recorded in toluene are shown in Figure 6. Molar extinction
coefficients of the series C-EP1-C-EP6 are approximately
proportional to the increasing number of dimer units. The
absorption spectrum of Zn-EP-Zn (Figure 6a, dashed-double-
dotted line) shows a broad Soret band caused by through-space
excitonic coupling between two porphyrin units via a m-
phenylene spacer. The shape of the spectrum resembles those
of similar bisporphyrins reported previously.17 In the case of
C-EP2, three Soret bands were observed at 431, 450, and 458.5
nm. The split bands (450 and 458.5 nm) can be ascribed to the
dipole-dipole interaction between two dimers as shown in
Figure 7. Thus, the interaction between two dipoles of two
dimers produces the new integrated dipoles X (red-shift) and
Y (blue-shift). The lowest energy Soret band and the Q(0,0)
band are red-shifted considerably by the interaction. Further red-
shifts of these bands are observed in the case of the macrocyclic
arrays C-EP5 and C-EP6 due to the interaction with two
adjacent dimers.
Figure 4. Protons assigned in Figure 3.
toluene. The absorption peaks of Soret- and Q-bands are listed
in Table 1. In pyridine, noncovalently linked dimers were
dissociated into monomeric species, and the absorption spectra
of Zn-EP and Zn-P in pyridine represent the corresponding
monomeric species with axial pyridyl coordination. In contrast,
the covalently linked arrays C-EP1, C-EP2, C-P1, C-EP5, and
C-EP6 maintain their complementary coordinated structures.15
Because of the low solubility of C-EP6 in other common
solvents, pyridine was chosen to compare the whole series of
cyclic and noncyclic porphyrins. In toluene, the complementary
coordination structure is perfectly maintained even without
covalent linking. Since the self-association constant of the dimer
unit reaches 1011 M-1 in CHCl3, the coordination structure is
dominant even in micromolar concentration without covalent
linking.
Figure 5 compares the absorption spectra of phenylethynyl-
and phenylporphyrins, Zn-EP, C-EP1, Zn-P, and C-P1. In
monomeric Zn-P, the two transition dipoles Bx and By in the
porphyrin plane are degenerate to show one strong Soret band
at 431.8 nm. By contrast, monomeric Zn-EP shows two Soret
bands at 433.6 and 445.8 nm due to the presence of nondegen-
erate Bx and By transition dipoles. The absorption intensity of
Bx at the longer wavelength (445.8 nm), Bx, is much larger.
The relative intensity of the Q(0,0) band of Zn-EP increases
as compared to that of Zn-P. When the dimer (C-EP1) is
The steady-state fluorescence spectra of C-EP1, C-EP2,
C-EP5, and C-EP6 are shown in Figure 8, and their emission
peak maxima, Stokes shifts, and quantum yields relative to
ZnTPP (ΦF ) 0.038 in pyridine)18 are listed in Table 1. The
fluorescence quantum yield of C-EP1 (9.3%) is slightly higher
than that of Zn-EP (8.3%), and its Stokes shift is smaller (112
cm-1 for C-EP1 and 235 cm-1 for Zn-EP), indicating that the
dimer provides a more rigid structure and raises the fluorescence
(16) Kasha, M. Radiat. Res. 1963, 20, 55-70.
(17) (a) Arnold, D. P.; James, D. A. J. Org. Chem. 1997, 62, 3460-3469. (b)
Shultz, D. A.; Lee, H.; Kumar, R. K.; Gwaltney, K. P. J. Org. Chem. 1999,
64, 9124-9136.
(18) Ambroise, A.; Li, J.; Yu, L.; Lindsey, J. S. Org. Lett. 2000, 2, 2563-
2566.
9
J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006 4617

A R T I C L E S
Hajjaj et al.
Figure 5. UV-vis absorption spectra in pyridine of (a) Zn-EP (dashed line) and C-EP1 (solid line) and (b) Zn-P (dashed line) and C-P1 (solid line). Inset:
schematic dipole directions and dipole-dipole coupling.
Figure 6. UV-vis absorption spectra (a) in pyridine and (b) in toluene. In (a), Zn-EP-Zn, C-EP1, C-EP2, C-EP5, and C-EP6 (from the bottom to the top).
In (b), Zn-EP, N-EP5, and N-EP6 (from the bottom to the top).
intensity compared to Zn-EP, though the vibrational degree of
freedom increased. In general, increasing the aggregation
number tends to decrease the fluorescence quantum yield. In
this case, however, the increase in the number of dimer units
(two for C-EP2, five for C-EP5, and six for C-EP6) does not
reduce the fluorescence quantum yields either, satisfying an
important prerequisite for LH antennae. If anything, the yield
increases along with increasing the number of dimer units.
Because of the inherently high fluorescence quantum yield
of Zn-EP (8.3%), as compared with those of zinc(II) porphyrin
substituted by meso-tetraaryl and alkyl groups (normally ΦF )
3-4%), macrocyclic C-EP5 and C-EP6 exhibit the highest class
of fluorescence quantum yields as porphyrin arrays.
Q-bands for all four molecular systems.3b,7a,11,19 It should be
noted that the absolute r values are smaller in C-EP5 and C-EP6
compared to C-EP1 and C-EP2, indicating the existence of
additional depolarization channels in the cyclic assembly
systems. Assuming that the magnitudes of r values originate
mainly from the depolarization processes by the rotational
diffusion motion in solution, the cyclic molecular systems should
have larger r values than those of the reference molecules C-EP1
and C-EP2 because of their larger hydrodynamic volumes.7a,20
Thus, the relatively small r values in cyclic systems imply the
involvement of another depolarization channel, such as the
redistribution of transition dipoles by the excitation energy
hopping process.19
The steady-state fluorescence excitation anisotropy (FEA)
spectra in pyridine were recorded at 670 nm to obtain informa-
tion on the relative orientation between absorption and emission
transition dipole moments (Figure 9). The steady-state FEA
spectra showed negative anisotropy values (r) in the high energy
Soret band and positive values at the low energy Soret and
Fluorescence Lifetime and Anisotropy Decay. The time-
resolved fluorescence decays were measured at several wave-
lengths in pyridine by a time-correlated single-photon-counting
(19) Osuka, A.; Kim, D.; et al. J. Phys. Chem. B 2005, 109, 11223-11230.
(20) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.; Kluwer
Academic/Plenum Publishers: New York, 1999.
9
4618 J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006

Porphyrin Dimers in Pentagonal and Hexagonal Arrays
A R T I C L E S
Figure 9. Steady-state fluorescence excitation anisotropy spectra of C-EP1,
C-EP2, C-EP5, and C-EP6 in pyridine recorded at 670 nm. The polarized
excitation spectra (IVV and IVH) were measured and calculated by r ) (IVV
- GIVH)/( IVV + 2GIVH).
Figure 7. Excitonic dipole-dipole coupling within C-EP1, C-EP2, C-EP5,
and C-EP6.
Table 2. Fluorescence Excitation Anisotropy (r) and Anisotropy
Decay Time (Φ)a
b
c
sample
rmax
rmin
Φ
d/ns
C-EP1
C-EP2
C-EP5
C-EP6
0.04
0.09
0.03
0.03
-0.06
-0.07
-0.04
-0.05
1.03
2.73
a All the experiments were performed in pyridine. b The maxima ani-
sotropy values around Soret bands. (Figure 9). c The minima anisotropy
values around Soret bands (Figure 9). d The λex ) 420 nm and λem ) 650
nm were applied for the anisotropic fluorescence measurement (Figure 11).
The anisotropy decay times were obtained by averaging the fitted parameters
measured in several emission wavelengths.
which showed a time constant of less than ∼40 ps and much
longer (>10 ns). Because of the large hydrodynamic volumes
of C-EP5 and C-EP6, the fast depolarization component (∼40
ps) could not be assigned to rotational diffusion processes.
Instead, they were assigned to other depolarization channels,
such as energy migration or transition dipole redistribution
processes. Thus, we conjecture that the EEH processes in C-EP5
and C-EP6 occur in a few tens of picoseconds.
Isotropic and Anisotropic Transient Absorption Measure-
ment. To investigate the EEH processes occurring in C-EP5
and C-EP6, we measured femtosecond transient absorption (TA)
decay with pump power dependence as well as transient
absorption anisotropy (TAA) changes. We tuned the excitation
wavelength for the Q-band (λpump ) 650 nm) because the
Q-band excitation could eliminate complicated processes arising
from the involvement of the internal conversion process from
the S2 to S1 state.
Figure 8. Fluorescence spectra of C-EP1, C-EP2, C-EP5, and C-EP6 in
pyridine.
(TCSPC) technique at the magic angle, and their fitted
parameters are listed in Table 1. The fluorescence lifetimes are
shortened slightly in the order C-EP1 > C-EP2 > C-EP5 >
C-EP6. The fluorescence quantum yields increase in the same
order. Therefore, these results indicate that the radiative decay
rate kF increases relative to the other decay rates as the number
of ring components increases. The Stokes shifts also decrease
slightly from 92 cm-1 (C-EP5) to 88 cm-1 (C-EP6), indicating
that the structural difference between ground and excited states
decreases by forming cyclic pentamers and hexamers. This trend
is more evident in N-EP5 and N-EP6 in toluene. The faster kF
and smaller Stokes shift of EP6 seem to be correlated.
The fluorescence anisotropy decays of C-EP1 and C-EP2
were fitted with the time constants of 1.07 and 2.73 ns,
respectively, representing the rotational correlation times (Figure
11). In the case of C-EP5 and C-EP6, on the other hand, the
temporal profiles were fitted with the biexponential function,
The transient absorption decays corresponding to the induced
absorption signal were measured at the magic angle (Figure 12).
The fitted parameters and the pump power conditions are
summarized in Table 3.
9
J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006 4619

A R T I C L E S
Hajjaj et al.
Figure 10. Time-resolved fluorescence decay profiles of C-EP1, C-EP2, C-EP5, and C-EP6 in pyridine, where the excitation and the emission wavelengths
were 420 and 670 nm, respectively. Instrumental response function (IRF) of our TCSPC system was ∼60 ps.
Figure 11. Time-resolved fluorescence anisotropy decay profiles of C-EP1, C-EP2, C-EP5, and C-EP6 in pyridine, where the excitation and the emission
wavelengths were the same as those of the experiment of magic angle fluorescence decay. The polarized excitation spectra (IVV and IVH) were measured, and
the time-resolved anisotropy decay r(t) was calculated by r(t) ) {IVV(t) - GIVH(t)}/{IVV(t) + 2GIVH(t)}.
C-EP1 showed no pump power dependence as the intensity
of the pump pulse was varied from high (1.8 mW) to low (0.3
mW). In contrast, when C-EP2 was excited by high pump
power, two fast decay components (0.48 and 14 ps) and a slow
decay component (1.45 ns) were observed, though no fast
component was observed under low power conditions, where
the time constant of the slow decay component was fixed
according to the TCSPC data.7 In C-EP5 and C-EP6, one slow
and two fast decay components were observed even under low
pump power conditions. The amplitudes of the two fast
components became dominant under high power conditions, i.e.,
58% and 50% for C-EP5 and C-EP6, respectively. Notably,
the amplitudes of the medium component in C-EP5 (22 ps)
and C-EP6 (19 ps) increased as the pump power increased from
0.3 to 1.8 mW. These time constants were considered to
represent the S1-S1 exciton-exciton annihilation processes,
since they become manifest only under high pump power
conditions. The fastest components (1-2 ps) observed for
C-EP5 and C-EP6 are expected to be associated with higher
order annihilation processes, which could be observed in
multichromophoric systems, especially when the samples are
excited by high pump power pulses.
At the same pump/probe wavelengths, the transient absorption
anisotropy (TAA) decay profiles were obtained under low pump
power conditions to avoid multiexciton processes observed in
the power dependence experiments (Figure 13). The fitted
parameters are listed in Table 4. In C-EP1, the anisotropy decay
was fitted with a fast (∼300 fs) and a slow (∼1 ns) decay
9
4620 J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006

Porphyrin Dimers in Pentagonal and Hexagonal Arrays
A R T I C L E S
Table 3. Transient Absorption Decay Time Constants at Magic
Anglea with Pump Power Dependence of C-EP1, C-EP2, C-EP5,
and C-EP6 in Pyridine
sample
powerb
τ1
τ2
τ3
C-EP1
1.65 ns
1.8 mW
0.6 mW
0.3 mW
100%
100%
100%
C-EP2
C-EP5
C-EP6
0.5 ps
31%
14 ps
7%
1.45 ns
1.8 mW
0.6 mW
0.3 mW
62%
100%
100%
1.3 ps
22 ps
1.42 ns
1.8 mW
0.6 mW
0.3 mW
32%
24%
43%
26%
11%
3%
42%
65%
54%
2.0 ps
19 ps
1.38 ns
1.8 mW
0.6 mW
0.3 mW
29%
32%
24%
21%
2%
5%
50%
66%
71%
a The λpump)650 nm and λprobe)520 nm were applied for the transient
absorption measurement. b The pump energy per pulse was 360, 120, and
60 nJ for 1.8, 0.6, and 0.3 mW, respectively, because the repetition rate of
5 kHz was applied.
Figure 12. Transient absorption decay profiles of C-EP1, C-EP2, C-EP5,
and C-EP6 (from top to bottom) in pyridine including power dependence,
where the pump and probe wavelengths are 650 and 520 nm.
component in a short time window (13 ps). The observed fast
decay component (∼300 fs) was similar to that of the dimer
(200 fs),11 which was assigned to both the intrachromophoric
dipole equilibrium and the interchromophoric EEH process
within the dimer. The steady-state absorption spectral shapes
of C-EP1 differ significantly from those of C-P1 with the
introduction of the ethynyl moiety. A much higher oscillator
strength was observed for the Bx component as compared to
By. The uneven transition dipoles may suppress the exchange
rate significantly. In addition, the slow component (∼1 ns) is
consistent with the anisotropy decay observed in the TCSPC
experiment and is assigned to the rotational correlation time.
In contrast to C-EP1, three decay components were observed
for C-EP2, C-EP5, and C-EP6. The slowest component in
C-EP2 was fitted to 2.7 ns, which is in accordance with the
anisotropy decay observed in TCSPC, and those in C-EP5 and
C-EP6 (>5 ns) could not be assigned exactly but are expected
to be associated with the rotational correlation times. Besides
the fastest time constant (∼300 fs), which could be explained
in the same manner as the case of C-EP1, newly observed
picosecond components (4-5 ps) most likely represent ad-
ditional depolarization channels due to the excitation energy
migration processes along the entire cyclic array.
Figure 13. Transient absorption anisotropy decay profiles of C-EP1,
C-EP2, C-EP5, and C-EP6 (from top to bottom) in pyridine, where the
pump and probe wavelengths are 650 and 520 nm.
In circularly arranged multichromophoric systems, the EEH
times can be analyzed explicitly by simultaneous consideration
9
J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006 4621

A R T I C L E S
Hajjaj et al.
Table 4. Transient Absorption Anisotropy Decay Time Constants
of C-EP1, C-EP2, C-EP5, and C-EP6 in Pyridinea
the EEH times of larger ring systems (C-EP5 and C-EP6) are
longer than those of smaller ring systems (C-P5 and C-P6),
and considering the molecular size, the EEH times of pentagonal
structures (C-EP5 and C-P5) are longer than their corresponding
hexagonal structures (C-EP6 and C-P6). Naturally, one would
expect that the larger distance between chromophores would
lead to shorter EEH times in cyclic array systems as well
because both through-bond and through-space energy transfer
rates strongly depend on the center-to-center distances propor-
tional to R-6 and exp-R, respectively. However, the shorter EEH
times in hexagonal structures as compared to those in pentagonal
structures are still questionable. This seems to stem from the
fact that the angle 120° of the 1,3-bis(ethynyl)phenylene linker
is better suited for hexagonal structures than it is for pentagonal
ones.
sample
τ1 (%)
τ2 (%)
τ3 (%)
r∞
C-EP1
C-EP2
C-EP5
C-EP6
0.3 ps (4%)
1.03 ns (96%)
2.70 ns (55%)
5 ns < (48%)
5 ns < (53%)
0.20
0.14
0.10
0.10
0.27 ps (35%)
0.36 ps (39%)
0.33 ps (29%)
4.3 ps (10%)
5.4 ps (13%)
4.2 ps (18%)
a The λpump ) 650 nm and λprobe ) 520 nm were applied for the
measurement. The anisotropy decay profiles were obtained under a low
pump power condition (0.3 mW).
of the two time constants, τann and τdep. Although neither
exciton-exciton annihilation nor anisotropy depolarization
directly represents the EEH times between the energy hopping
units, the EEH times can be determined by synchronizing the
two time constants by modeling the EEH mechanism.
In this regard, the faster EEH times in hexagonal structures
are thought to arise from the difference in the relative orientation
between the nearest neighboring chromophores. As seen in the
Fo¨rster-type energy transfer rate (eq 3) and the dipole coupling
formula (eq 4), the orientation factor (κ) is in the numerator in
both equations:
Discussion
EEH Models in Cyclic Pentamer and Hexamer Arrays.
We have analyzed the EEH processes of C-EP5 and C-EP6
using the S1-S1 exciton-exciton annihilation and anisotropy
depolarization processes on the basis of Fleming’s method.21
Assuming migration-limited annihilation and a random-walk
formalism of anisotropy decay, where the dimer unit can act as
a coherent domain within the strong coupling limit, the EEH
time can be obtained from the analytical depolarization and
exciton-exciton annihilation times using eqs 1 and 2, where N
is the number of excitation energy hopping sites, R, the angle
between the neighboring transition dipoles (108° for C-EP5 and
120° for C-EP6), τann, the slowest exciton-exciton annihilation
time, τdep, the depolarization time, and τhopping, the inverse of
the nearest neighbor energy hopping rate:21
8.8 × 10-25 κ2 Φ
kTS
)
J
(3)
(4)
n4R6τ
5.04 f 2L|bµi||bµj|κ
ꢀopR3TS
Vij )
where n is the refractive index of the solvent, R, the center-to-
center distance between donor and acceptor, τ, the fluorescence
lifetime of an energy donor, κ, the orientation factor, Φ, the
fluorescence quantum yield, J, the spectral overlap integral in
eq 3; Vij is the transition dipole coupling energy, fL, the Lorenz
local field correction factor (fL ) (ꢀop+2)/3), ꢀop, the dielectric
constant at optical frequencies, µi,j, the transition dipole moment,
and RTS, the center-to-center distance in nm, in which the
subscript TS means through-space in eq 4.22 The orientation
factor κ is calculated by eq 5:
τhopping
τhopping
τdep
)
)
(1)
(2)
4(1 - cos2(2π/N)) 4(1 - cos2 R)
N2 - 1
τann
)
τhopping
24
where τdep and τann were obtained from, respectively, TAA and
transient absorption decay depending on the pump power. For
C-EP5 where N ) 5, we substituted τdep ) 5.4 ps in eq 1 and
obtained τhopping ) 19.5 ps, and in a different approach, the
substitution of τann ) 22 ps in eq 2 gave τhopping ) 22 ps.
Similarly, for C-EP6 where N ) 6, we substituted τdep ) 4.2
ps in eq 1 and obtained τhopping ) 12.6 ps for C-EP6 and
substituted τann ) 19 ps in eq 2 and obtained τhopping ) 13 ps.
Interestingly, the two different experimental observables, τdep
and τann, result in consistent EEH times (τhopping) within small
deviations. Thus, the excitation energy migration processes in
these macrocyclic arrays can be well demonstrated by a Fo¨rster-
type incoherent energy hopping model. By taking an average
value, we determined the EEH times between the nearest
neighboring dimer units to be 21 ( 1 ps for C-EP5 and 12.8 (
0.2 ps for C-EP6. In our previous study,11 the same method
has been applied to the EEH times for C-P5 and C-P6 to
determine the EEH times of ∼8.0 and ∼5.3 ps, respectively. A
difference in the EEH rates will be discussed in the next section.
Transition Dipole Coupling Energy and EEH Times in
Cyclic Array Systems. As described in the previous section,
κ2 ) (cos θT - 3 cos θi cos θj)2
(5)
where θT is the dihedral angle between the two transition dipole
moments, and θi and θj are the angles between these dipoles
and the segment joining the two transition dipole moments.
On the assumption of perfect planar pentagonal and hexagonal
structures, κ is calculated to be 1.65 for C-EP5 and 1.75 for
C-EP6.23 The difference in κ seems small, but when the EEH
processes occur from one hopping unit to the other one
continuously, the effect of the small deviation in κ would
increase exponentially. In summary, the larger value of κ in
hexagonal structures induces a larger transition dipole coupling
(22) Morandeira, A.; Vauthey, E.; Schuwey, A.; Gossauer, A. J. Phys. Chem.
A 2004, 108, 5741-5751.
(23) The angular parameters (θT, θi, and θj) values used for estimating κ are
72°, 36°, and 36°, respectively, for C-EP5, and 60°, 30°, and 30°,
respectively, for C-EP6.
(21) Bradforth, S. E.; Jimenez, R.; van Mourik, F.; van Grondelle, R.; Fleming,
G. R. J. Phys. Chem. 1995, 99, 16179-16191.
9
4622 J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006

Porphyrin Dimers in Pentagonal and Hexagonal Arrays
A R T I C L E S
energy as compared to that in pentagonal structures, resulting
in shorter EEH times in hexagonal structures. That is, well-
oriented hexagonal arrays by the combinative conformation of
chromophores and linkers exhibit faster EEH processes than
pentagonal arrays. Thus, the difference between the experimen-
tally deduced EEH times of C-EP5 and C-EP6 is mainly
accounted for by the orientation factor induced by the confor-
mational differences. Indeed, the slightly more well-resolved
Soret bands and red-shifted Q-bands of C-EP6 as compared
with those of C-EP5 support our arguments.
Assessment of Antenna Functions and Outlook. The EEH
time constants for C-EP5 (21 ps) and for C-EP6 (12.8 ps) are
small enough compared with the lifetimes for C-EP5 (1.42 ns)
and C-EP6 (1.38 ns). In addition to their fluorescence intensities,
which are even larger than that of a monomeric unit, macrocyclic
arrays are regarded as excellent LH antennae. Different EEH
times between C-EP5 and C-EP6 indicate that EEH times
depend strongly on the orientation factor of the constituent
elements of macrocyclic arrays. However, the present study only
investigated pentagonal and hexagonal macrocyclic arrays. The
hypothesis should be confirmed for macrocycles larger than
hexagonal arrays.
zolylporphyrin Zn-EP-Zn. The proton NMR spectra of N-EP5
and N-EP6 indicate fast rotation of the porphyrin moieties along
the ethyne axis. The covalently linked macrocyclic porphyrin
arrays C-EP5 and C-EP6 were confirmed by mass spectrometry
and analyzed by various steady-state and time-resolved photo-
physical measurements in pyridine. Electronic and excitonic
dipole-dipole couplings in EP5 and EP6 macrocycles were
examined by comparison with various reference porphyrins.
Introduction of the ethyne moiety induces a large transition
dipole along the ethyne axis (Bx) and lowers the energy level
of the Q(0,0) band. Excitonic dipole-dipole coupling directed
toward Bx was dominant, whereas that toward By was negligible.
Based on the exciton-exciton annihilation and anisotropy
depolarization times, we have determined the EEH times in
cyclic multichromophoric systems to be 21 ps for C-EP5 and
12.8 ps for C-EP6. It is notable that the EEH times obtained
by the two methods closely match each other. The hexagonal
macrocyclic arrays with a better orientation of transition dipoles
afforded a shorter EEH time compared to pentagonal arrays.
This is a new finding for synthetic macrocyclic antennae.
Experimental Section
Comparison of the present EEH times with those of smaller
macrocycles, C-P5 (∼8.0 ps) and C-P6 (∼5.3 ps), suggest that
the EEH time is dependent on the center-to-center distance
between the coherent dimers. Coplanarity and electronic com-
munication through the ethyne bond in C-EP5 and C-EP6 does
not appreciably accelerate the EEH times.
Introduction of the ethyne moiety between porphyrin and
phenylene causes the red-shifted absorption spectra of C-EP5
and C-EP6 compared to C-P5 and C-P6. In addition, the Q(0,0)
band of C-EP5 and C-EP6 increases considerably. These
features are suitable as a second antenna into which excited
energy flows from a peripheral higher energy antenna. Inspection
of the overlap between the emission spectra of C-P5 and C-P6
and absorption spectra of C-EP5 and C-EP6 suggests that their
composites provide an efficient energy cascade system like LH2
and LH1.
All of the synthetic procedures and photophysical experiments
are described in the Supporting Information.
Acknowledgment. This work was supported by Grant-in-
Aids for Scientific Research (A) (Y.K.), Young Scientists (B)
(A.S.), Scientific Research on Priority Areas (No. 15036248),
Reaction Control of Dynamic Complexes (Y.K.) from the
Ministry of Education, Culture, Sports, Science and Technology,
Japan and JSPS (Japan Society for the Promotion of Science),
and the National Creative Research Initiatives Program of the
Korea Science and Engineering Foundation of Korea (D.K.).
Supporting Information Available: Synthetic procedures,
photophysical experiments; GPC analyses of C-EP1, C-EP2,
EP5, and EP6; MALDI-TOF mass spectra of C-EP1, C-EP2,
C-EP5, and C-EP6; 2D NMR spectrum of N-EP5; temperature-
1
variable H NMR spectra of N-EP5; UV-vis spectra of all
porphyrins in toluene and pyridine, and complete refs 11 and
19. This material is available free of charge via the Internet at
http://pubs.acs.org.
Conclusions
We have demonstrated the effective synthesis of the macro-
cyclic porphyrin arrays EP5 and EP6 by complementary
coordination of m-bis(ethynylene)phenylene-linked imida-
JA0583214
9
J. AM. CHEM. SOC. VOL. 128, NO. 14, 2006 4623