Directly meso-meso linked porphyrin rings: Synthesis, characterization, and efficient excitation energy hopping

Article abstract of DOI:10.1021/ja045254p

Directly meso-meso linked porphyrin rings CZ4, CZ6, and CZ8 that respectively comprise four, six, and eight porphyrins have been synthesized in a stepwise manner from a 5, 10-diaryl zinc(II) porphyrin building block. Symmetric cyclic structures have been

Full text of DOI:10.1021/ja045254p

Published on Web 12/03/2004
Directly meso-meso Linked Porphyrin Rings: Synthesis,
Characterization, and Efficient Excitation Energy Hopping
Yasuyuki Nakamura,† In-Wook Hwang,‡ Naoki Aratani,† Tae Kyu Ahn,‡
Dah Mee Ko,‡ Akihiko Takagi,§ Tomoji Kawai,§ Takuya Matsumoto,§
Dongho Kim,*,‡ and Atsuhiro Osuka*,†
Department of Chemistry, Graduate School of Science, Kyoto UniVersity, and Core Research for
EVolutional Science and Technology (CREST), Japan Science and Technology Agency (JST),
Sakyo-ku, Kyoto 606-8502, Japan; Center for Ultrafast Optical Characteristics Control and
Department of Chemistry, Yonsei UniVersity, Seoul 120-749, Korea; and the Institute of
Scientific and Industrial Research (ISIR), Osaka UniVersity, Core Research for EVolutional
Science and Technology (CREST), Japan Science and Technology Agency (JST), Mihogaoka,
Ibaragi, Osaka 567-0047, Japan
Received August 6, 2004; E-mail: osuka@kuchem.kyoto-u.ac.jp; dongho@yonsei.ac.kr
Abstract: Directly meso-meso linked porphyrin rings CZ4, CZ6, and CZ8 that respectively comprise four,
six, and eight porphyrins have been synthesized in a stepwise manner from a 5,10-diaryl zinc(II) porphyrin
1
building block. Symmetric cyclic structures have been indicated by their very simple H NMR spectra that
exhibit only a single set of porphyrin and their absorption spectra that display a characteristic broad nonsplit
Soret band around 460 nm. Energy minimized structures calculated at the B3LYP/6-31G* level indicate
that a dihedral angle between neighboring porphyrins decreases in order of CZ6 > CZ8 > CZ4, which is
1
consistent with the H NMR data. Photophysical properties of these molecules have been examined by
the steady-state absorption, fluorescence, fluorescence lifetime, fluorescence anisotropy decay, and transient
absorption measurements. Both the pump-power dependence on the femtosecond transient absorption
and the transient absorption anisotropy decay profiles are directly related with the excitation energy migration
processes within the porphyrin rings, where the exciton-exciton annihilation time and the polarization
anisotropy rise time are well described in terms of the Fo¨rster-type incoherent energy hopping model.
Consequently, the excitation energy hopping rates have been estimated for CZ4 (119 ( 2 fs)-1, CZ6 (342
( 59 fs)-1, and CZ8 (236 ( 31 fs)-1, which reflect the magnitude of the electronic coupling between the
neighboring porphyrins. Overall, these porphyrin rings serve as a well-defined wheel-shaped light harvesting
antenna model in light of very efficient excitation energy hopping along the ring.
Introduction
that rivals those in the natural photosynthetic antenna has hardly
been achieved so far. It is expecetd that the regular cyclic
In the bacterial photosynthetic antenna of LH1 and LH2,
photosynthetic tetrapyrrolic pigments are deliberately positioned
in a wheel-like arrangement, hence manipulating the absorbed
light energy through intrawheel efficient excitation energy
hopping (EEH) processes.1 Stimulated by these beautiful
architectures, several cyclic and dendritic porphyrin arrays have
been explored and intramolecular EEH processes therein have
been studied.2-9 Despite these extensive efforts, efficient EEH
(3) (a) Wagner, R. W.; Seth, J.; Yang, S. I.; Kim, D.; Bocian, D. F.; Holten,
D.; Lindsey, J. S. J. Org. Chem. 1998, 63, 5042. (b) 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. (c) Ambroise, A.; Li, J.;
Yu, L.; Lindsey, J. S. Org. Lett. 2000, 2, 2563.
(4) (a) 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. (b)
Schweitzer, G.; De Belder, G.; Latterini, L.; Karini, Y.; Rowan, A. E.;
Nolte, R. J. M.; de Schryver F. C. Chem. Phys. Lett. 1999, 303, 261. (c)
Lensen, M. C.; van Dingenen, S. J. T.; Elemans, J. A. A.; Dijkstra, H. P.;
van Klinc, G. P. M.; van Koten, G.; Gerritsen, J. W.; Speller, S.; Nolte, R.
J. M.; Rowan, A. E. Chem. Commun. 2004, 762. (d) Lensen, M. C.;
Takazawa, K.; Elemans, J. A. A. W.; Jeukens, C. R. L. P. N.; Christianen,
P. C. M.; Maan, J. C.; Rowan, A. E.; Nolte, R. J. M. Chem.sEur. J. 2004,
10, 831.
† Kyoto University.
‡ Yonsei University.
§ Osaka University.
(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.
(b) Karrasch, S.; Bullough, P. A.; Ghosh, R. EMBO J. 1995, 14, 631. (c)
Koepke, J.; Hu, X.; Muenke, C.; Schulten, K.; Michel, H. Structure 1996,
4, 581. (d) McLuskey, K.; Prince, S. M.; Cogdell, R. J.; Isaacs, N. W.
Biochemistry 2001, 40, 8783. (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.
(5) (a) Sugiura, K.; Fujimoto, Y.; Sakata, Y. Chem. Commun. 2000, 1105. (b)
Kato, A.; Sugiura, K.; Miyasaka, H.; Tanaka, H.; Kawai, T.; Sugimoto,
M.; Yamashita, M. Chem. Lett. 2004, 33, 578.
(6) (a) Mongin, O.; Papamicae¨l, C.; Hoyler, N.; Gossauer, A. J. Org. Chem.
1998, 63, 5568. (b) Mongin, O.; Schuwey, A.; Vallot, M.-A.; Gossauer,
A. Tetrahedron Lett. 1999, 40, 8347. (c) Brodard, P.; Matzinger, S.;
Vauthey, E.; Mongin, O.; Papamicae¨l, C.; Gossauer, A. J. Phys. Chem. A
1999, 103, 5858. (d) Mongin, O.; Hoyler, N.; Gossauer, A. Eur. J. Org.
Chem. 2000, 1193. (e) Rucareanu, S.; Mongin, O.; Schuwey, A.; Hoyler,
N.; Gossauer, A. Amrein, W.; Hediger, H.-U. J. Org. Chem. 2001, 66,
4973.
(2) (a) Anderson, H. L.; Sanders, J. K. M. J. Chem. Soc., Chem. Commun.
1989, 1714. (b) Anderson, H. L.; Sanders, J. K. M. Angew. Chem., Int.
Ed. Engl. 1990, 29, 1400. (c) Anderson, S.; Anderson, H. L.; Sanders, J.
K. M. Acc. Chem. Res. 1993, 26, 469. (d) Dubowchik, G.; Hamilton, A.
D. J. Chem. Soc., Chem. Commun. 1987, 293.
9
236
J. AM. CHEM. SOC. 2005, 127, 236-246
10.1021/ja045254p CCC: $30.25 © 2005 American Chemical Society

Directly meso−meso Linked Porphyrin Rings
A R T I C L E S
Chart 1. Structures of CZ4, CZ6, and CZ8 (Ar ) 3,5-Di-tert-butylphenyl)
arrangement and the large electronic coupling between neigh-
boring pigments are required for efficient EEH.
whose electronic π-network consists of only porphyrins and thus
may be interesting also in view of belt-shape aromatic molecules
that consist of only aromatic segments.14 In addition, these
molecules will allow direct comparison of molecular morphol-
ogy effects, linear versus cyclic, upon the overall photophysical
properties of meso-meso linked porphyrin arrays. In this paper,
we report the syntheses and characterizations of porphyrin rings
such as CZ4, CZ6, and CZ8 (Chart 1) and their photophysical
properties.
In the past decade, we have explored various directly meso-
meso linked porphyrin arrays on the basis of silver(I)-promoted
coupling of a 5,15-diaryl-substituted zinc porphyrin.10,11 Some
of these porphyrin arrays are promising as optical molecular
wire by transmitting excitation energy over a long distance.
Actually, the efficient excitation energy transfer has been
achieved over a long distance through meso-meso linked zinc-
(II) porphyrin arrays.11 This function relies on their strongly
coupled but not fully π-conjugated electronic character.12
Recently, we reported a dodecameric porphyrin wheel, in which
meso-meso linked diporphyrin subunits are bridged by a 1,3-
phenylene spacer that is used to produce curvature of array.13
The presence of such a spacer weakens the electronic coupling
and causes two different EEH steps with rates of 0.24 ps-1 and
3.6 ps-1, which correspond to EEH within a meso-meso linked
diporphyrin and between neighboring meso-meso linked di-
porphyrins.
Experimental Section
General Information. The syntheses of CZ4, CZ6, and CZ8 as
well as the intermediates leading to them are detailed in the Supporting
Information. All solvents were spectrophotometric grade. All the
calculations were carried out using the Gaussian 03 program. All
structures were optimized with Becke’s three-parameter hybrid ex-
change functional and the Lee-Yang-Parr correlation functional
(B3LYP), employing the 6-31G* basis set for all atoms.
STM Images. Clean flat Cu(100) surfaces were obtained by Ar+
sputtering and annealing (550°C) cycles for a substrate. The porphyrin
ring molecules dissolved into CH2Cl2 were deposited by spraying ca.
0.5 µL of the solution onto the substrate in a vacuum (10-6 mbar) using
a pulse injection method,15 which is suited for deposition of large fragile
molecules with escaping decomposition often encountered in sample
deposition from the gas phase. In situ STM measurements were
performed at room temperature in an ultrahigh vacuum (<10-10 mbar)
with a home-build STM by using an electrochemical etched Pt/Ir tip.
STM image was obtained in constant height mode.
Cyclic porphyrin architectures, in which the constituent
porphyrins are all directly linked at meso-meso positions, are
a more attractive target in view of synthetic challenge, higher
molecular symmetry, and large and regular electronic interaction
between neighboring porphyrins that will lead to efficient EEH.
Such molecules can be regarded as a genuine porphyrin ring
(7) (a) Takase, M.; Ismael, R.; Murakami, R.; Ikeda, M.; Kim, D.; Shinmori,
H.; Furuta, H.; Osuka, A. Tetrahedron Lett. 2002, 43, 5157. (b) 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.
(8) (a) Choi, M.-S.; Yamazaki, T.; Yamazaki, I.; Aida, T. Angew. Chem., Int.
Ed. 2004, 43, 150. (b) Choi, M.-S.; Aida, T.; Yamazaki, T.; Yamazaki, I.
Chem.sEur. J. 2002, 8, 2668. (c) Choi, M.-S.; Aida, T.; Luo, H.; Araki,
Y.; Ito, O. Angew. Chem., Int. Ed. 2003, 42, 4060. (d) 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.
Steady-State Spectra. The samples were prepared in micromolar
concentrations in THF (spectroscopic grade). Absorption spectra were
obtained with a Shimadzu model 1601 UV spectrometer, and steady-
state fluorescence was measured by a Hitachi model F-4500 fluores-
cence spectrophotometer at room temperature.
(9) Noncovalent porphyrin assemblies were reported. (a) Drain, C. M. Nifiatis,
F.; Vasenko, A.; Batteas, J. D. Angew. Chem., Int. Ed. 1998, 37, 2344. (b)
Rubtsov, I. V.; Kobuke, Y.; Miyaji, H.; Yoshihara, K. Chem. Phys. Lett.
1999, 308, 323. (c) Takahashi, R.; Kobuke, Y. J. Am. Chem. Soc. 2003,
125, 2372. (d) Ikeda, C.; Satake, A.; Kobuke, Y. Org. Lett. 2003, 5, 4935.
(e) Kuramochi, Y.; Satake, A.; Kobuke, Y. J. Am. Chem. Soc. 2004, 126,
8668, and refs 3c and 6d.
(10) (a) Osuka, A.; Shimidzu, H. Angew. Chem., Int. Ed. Engl. 1997, 36, 135.
(b) Aratani, N.; Osuka, A.; Kim, Y. H.; Jeong, D. H.; Kim, D. Angew.
Chem., Int. Ed. 2000, 39, 1458.
(11) (a) Nakano, A.; Osuka, A.; Yamazaki, I.; Yamazaki, T.; Nishimura, Y.
Angew. Chem., Int. Ed. 1998, 37, 3023. (b) Nakano, A.; Yamazaki, T.;
Nishimura, Y.; Yamazaki, I.; Osuka, A. Chem.sEur. J. 2000, 6, 3254. (c)
Aratani, N.; Cho, H. S.; Ahn, T. K.; Cho, S.; Kim, D.; Sumi, H.; Osuka,
A. J. Am. Chem. Soc. 2003, 125, 9668.
(14) (a) Ashton, P. R.; Isaacs, N. S.; Kohnke, F. H.; Slawin, A. M. Z.; Spencer,
C. M.; Stoddart, J. F.; Williams, D. J. Angew. Chem., Int. Ed. Engl. 1988,
27, 966. (b) Diederich, F. Nature 1994, 369, 199. (c) Kawase, T.; Darabi,
H. R.; Oda, M. Angew. Chem., Int. Ed. Engl. 1996, 35, 2664. (d) Hensel,
V.; Schlu¨ter, A. D. Chem.sEur. J. 1999, 5, 421. (e) Kro¨mer, J.; Rios-
Carreras, I.; Fuhrmann, G.; Musch, C.; Wunderlin, M.; Debaerdemaeker,
T.; Mena-Osteritz, E.; Ba¨uerle, R. Angew. Chem., Int. Ed. 2000, 39, 3481.
(f) Ohkita, M.; Ando, K.; Tsuji, T. Chem. Commun. 2001, 2570. (g)
Fuhrmann, G.; Debaerdemaeker, T.; Ba¨uerle, P. Chem. Commun. 2003,
948. (h) Seidel, D.; Lynch, V.; Sessler, J. L. Angew. Chem., Int. Ed. 2002,
41, 1422. (i) Ko¨hler, T.; Seidel, D.; Lynch, V.; Arp. F. O.; Ou, Z.; Kadish,
K. M.; Sessler, J. L. J. Am. Chem. Soc. 2003, 125, 6872. (j) Neudorff, W.
D.; Lentz, D.; Anibarro, M.; Schlu¨ter, A. D. Chem.sEur. J. 2003, 9, 2745.
(k) Kawase, T.; Tanaka, K.; Fujiwara, N.; Darabi, R. H.; Oda, M. Angew.
Chem., Int. Ed. 2003, 42, 1624. (l) Matsuo, Y.; Tahara, K.; Sawamura,
M.; Nakamura, E. J. Am. Chem. Soc. 2004, 126, 8725.
(12) (a) Kim, D.; Osuka, A. J. Phys. Chem. A 2003, 107, 8791. (b) Kim, Y. H.;
Jeong, D. H.; Kim, D.; Jeoung, S. C.; Cho, H. S.; Kim, S. K.; Aratani, N.;
Osuka, A. J. Am. Chem. Soc. 2001, 123, 76.
(13) 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.
(15) (a) Tanaka, H.; Nakagawa, T.; Kawai, T. Surf. Sci. 1996, 364, L575. (b)
Takagi, A.; Yanagawa, Y.; Tsuda, A.; Aratani, N.; Matsumoto, T.; Osuka,
A.; Kawai, T. Chem. Commun. 2003, 2986.
9
J. AM. CHEM. SOC. VOL. 127, NO. 1, 2005 237

A R T I C L E S
Nakamura et al.
Time-Resolved Fluorescence. A picosecond time-correlated single
photon counting (TCSPC) system was used for time-resolved fluores-
cence and fluorescence anisotropy decay measurements. The system
consisted of a self-mode-locked and cavity-dumped femtosecond Ti:
sapphire laser pumped by a continuous wave (cw) Nd:YAG laser
(Coherent, Verdi). The full width at half-maximum of the instrument
response function obtained by a dilute solution of coffee cream was
typically 50 ps in our TCSPC system. The fluorescence decays were
measured with magic-angle emission polarization, and the number of
fluorescence photons per unit time, detected by photomultiplier tube,
was always maintained to <1% of the repetition rate of the excitation
pulses, to prevent pile-up distortions in the decay profiles. Time-resolved
fluorescence anisotropy decays were obtained by changing the detection
polarization on the fluorescence path parallel or perpendicular to the
polarization of the excitation light. The anisotropy decays then were
calculated as follows:
Figure 1. (Bottom) Absorption spectrum of Z2. (Top) CD spectra of Z2
in CH2Cl2; the fast eluting isomer Z2A (bold line) and the second eluting
isomer Z2B (thin line).
I
VV(t) - GIVH(t)
r(t) )
(1)
IVV(t) + 2GIVH(t)
to calculate the absorption difference at the desired time delay between
pump and probe pulses. For the transient absorption anisotropy decay
(r(t)) measurements, the probe white-light continuum pulse was set to
have vertical polarization using a sheet polarizer. The excitation pulse
then was changed to have parallel or perpendicular polarization by
rotating a half wave plate, with respect to the polarization of the probe
pulse. Finally, the transient absorption anisotropy decay was obtained
by the following equation,
where IVV(t) (or IVH(t)) is the fluorescence decay when the excitation
light is vertically polarized and only the vertically (or horizontally)
polarized portion of fluorescence is detected, denoting that the first
and second subscripts represent excitation and detection polarization,
respectively. The factor G is defined by IVV(t)/IVH(t), which is equal to
the ratio of the sensitivities of the detection system for vertically and
horizontally polarized light. The G factor of our detection system was
1.08. The fittings for both isotropic and anisotropic decays were
performed by a least-squares deconvolution fitting process. The vertical
and horizontal components of fluorescence emission were simulta-
neously fitted to extract the anisotropy decay functions, using the
LIFETIME program with an iterative nonlinear least-squares decon-
volution procedure that was developed at the University of Pennsyl-
vania.16
Transient Absorption and Transient Absorption Anisotropy. The
dual-beam femtosecond time-resolved transient absorption spectrometer
consisted of a self-mode-locked femtosecond Ti:sapphire oscillator
(Coherent, MIRA), a Ti:sapphire regenerative amplifier (Clark MXR
model TRA-1000) that was pumped by a Q-switched Nd:YAG laser
(Clark MXR model ORC-1000), a pulse stretcher/compressor, an optical
parametric amplifier (Clark MXR OPA), and an optical detection
system. A femtosecond Ti:sapphire oscillator pumped by a cw Nd:
YVO4 laser (Coherent, Verdi) produces a train of ∼80-fs mode-locked
pulses with an averaged power of 650 mW at 800 nm. The amplified
output beam regenerated by chirped pulse amplification (CPA) had a
pulse width of ca. 150 fs and a power of ca. 1 W at a repetition rate of
1 kHz, which was divided into two parts by a 1:1 beam splitter. One
part was color-tuned for the pump beam by an optical parametric
generation and amplification (OPG-OPA). The resulting laser pulse had
a temporal width of ∼150 fs in the vis/IR range. The pump beam was
focused to a spot diameter of ∼1 mm, and the laser fluence was
adjusted, using a variable neutral-density filter. The other part was
focused onto a flowing water cell to generate a white-light continuum,
which was again split into two parts. One part of the white-light
continuum was overlapped with the pump beam at the sample to probe
the transient, while the other part of the white-light continuum was
passed through the sample without overlapping the pump beam. The
time delay between pump and probe beams was controlled by making
the pump beam travel along a variable optical delay line. The white-
light continuum beams after the sample were sent through an appropriate
interference filter and then were detected by two photodiodes. The
outputs from the two photodiodes at the selected wavelength were
processed by a combination of a boxcar averager and a lock-in amplifier,
∆AVV(t) - ∆AHV(t)
r(t) )
(2)
∆AVV(t) + 2∆AHV(t)
where the variable represents amplitude. The fittings for anisotropic
decays were performed by a least-squares deconvolution fitting process.
The vertical and horizontal components of TA decay were simulta-
neously fitted to extract the anisotropy decay functions.
Results and Discussion
Synthesis and Characterization. 5,10-Bis(3,5-di-tert-bu-
tylphenyl)-substituted zinc(II) porphyrin (Z1)5a was employed
as a building block to construct the cyclic arrays. Coupling
reaction of Z1 with AgPF6 in CHCl3 afforded dimer Z2 (24%),
trimer Z3 (7%), and higher oligomers (Chart 2). Since a free
rotation around the meso-meso bond is strictly prohibited,17
Z2 is chiral and Z3 is diastereoisomeric, which will complicate
higher coupling products of Z2 and Z3. To avoid this burden-
some situation, Z2 was optically separated into two enatiomers
Z2A and Z2B through a chiral HPLC column (Supporting
Information). Figure 1 shows the absorption and CD spectra of
the first and second eluting Z2 isomers. The CD spectra of the
two isomers are weak but perfect mirror images of each other.
In the CD spectrum of the fast eluting isomer, the first Cotton
effect with positive sense was observed around the low energy
Soret band and the second one was observed as a bisignate split
Cotton effect with positive sense at the high energy Soret band.
The bisignate Cotton effect at the high energy Soret band is
arising from coupling of Bx and By transition dipole moments
by sensing a subtle difference in chiral environment associated
with the presence or absence of side meso-aryl substituent.
Therefore, on the basis of these Cotton effects and empirical
(16) (a) Wild, U. P.; Holzwarth, A. R.; Good, H. P. ReV. Sci. Instrum. 1977,
48, 1621. (b) Kim, Y.-R.; Hahn, J.-S.; Hong, H.; Jeong, W.; Song, N. W.;
Shin, H.-C.; Kim, D. Biochim. Biophys. Acta 1999, 1429, 486.
(17) Yoshida, N.; Osuka, A. Tetrahedron Lett. 2000, 41, 9287.
9
238 J. AM. CHEM. SOC. VOL. 127, NO. 1, 2005

Directly meso−meso Linked Porphyrin Rings
A R T I C L E S
Chart 2. Porphyrin Z1 and Its Oligomers (Ar ) 3,5-Di-tert-butylphenyl)
exciton chirality methods and the related studies,18 the first
eluting isomer has been assigned as Z2A and the second one
as Z2B.
The triporphyrin fraction Z3 was separated into two diaster-
oisomers, an entiomeric pair of Z3Aa and Z3Ba and meso-
isomer Z3b, by chromatography over a silica gel column. At
this step, it is difficult to assign the isomers. As is similar to
the case of Z4Ab, the meso-isomer Z3b having two free meso-
positions on the same side is expected to give a cyclic array.
Upon treatment with AgPF6 at 0.1 mM in CHCl3 for 2 h, the
slowly eluting triporphyrin gave coupling products, from which
hexameric porphyrin ring CZ6 was isolated in 22% yield as a
first eluting fraction. On the other hand, the fast eluting
triporphyrin did not give such a porphyrin ring product. On the
basis of these experiments, the slowly eluting triporphyrin has
been assigned as the meso-isomer Z3b.
Then, the optically pure Z2A was coupled to tetraporphyrin
Z4A (40%), hexaporphyrin Z6A (10%), and higher oligomers
(ca. 6%). The Z4A products were then separated by chroma-
tography over a silica gel column into two diastereoisomers
Z4Aa and Z4Ab. At this stage, Z4Ab is expected to be a
suitable precursor of cyclic porphyrin arrays, but it is difficult
to assign these diasterosiomeric Z4A products. Thus, we
attempted the silver(I)-promoted coupling of both tetraporphy-
rins. Upon treatment with AgPF6, a slowly eluting tetramer in
CHCl3 (3.3 mM) gave octameric porphyrin ring CZ8 in 29%
yield together with other noncyclic oligomers, while the other
tetraporphyrin isomer gave only linear noncyclic oligomers.
These results led us to assign the fast and slowly eluting
diasteroisomeric tetramers as Z4Aa and Z4Ab, respectively.
Interestingly, under dilute conditions (0.02 mM), the intramo-
lecular cyclization of Z4Ab gave tetrameric porphyrin ring CZ4
preferentially in 74% yield.
The structures of CZ4, CZ6, and CZ8 were characterized
by mass spectroscopy and NMR spectroscopy. The peaks
observed in electrospray ionization mass (ESI-MS) are m/z
1493.67 for CZ4 corresponding to [M + 2H]2+, m/z 1497.00
for CZ6 corresponding to [M + 3H]3+, and m/z 1996.22 for
CZ8 corresponding to [M + 3H]3+, which all support their
structures (Supporting Information). The symmetric cyclic
1
structures have been indicated by their very simple H NMR
(18) (a) Nakanishi, K.; Berova, N. In Circular Dichroism: Pronciples and
Applications; Nakanish, K., Berova, N., Woody, R. W., Eds.; VCH
Publishers Inc.: NewYork, 1994; p 361. (b) Mason, S. F. Molecular Optical
ActiVity and the Chiral Discriminations; Cambridge University Press:
Cambridge, 1982. (c) Harada, N.; Nakanishi, K. Acc. Chem. Res. 1972, 5,
257. But the final determination of the stereochemistry will be done through
X-ray crystallography.
spectra, which all exhibit only a single set of porphyrins (Figure
2). These simple spectra are in sharp contrast to those of
noncyclic porphyrin arrays (Supporting Information). Four
â-peripheral protons (designated as H1, H2, H3, and H4 in Figure
9
J. AM. CHEM. SOC. VOL. 127, NO. 1, 2005 239

A R T I C L E S
Nakamura et al.
Figure 2. 1H NMR spectra of (a) CZ4, (b) CZ6, and (c) CZ8 in CDCl3. The nonlabeled three peaks around 8 ppm were assigned to Ar-H.
Figure 3. Energy-minimized structures of simplified (a) CZ4, (b) CZ6, and (c) CZ8, in which meso-substituents were omitted and zinc atoms were substituted
for H2.
2) are differentiated in these porphyrin rings, and their assign-
ments have been accomplished by comprehensive ROESY
measurements and are presented in Figure 2. Of the four
â-peripheral protons, the chemical shifts of H3 and H4 that are
adjacent to the meso-meso linkage exhibit large changes
depending on the molecule; the chemical shift of H3 is 9.57
ppm in CZ4, 8.25 ppm in CZ6, and 7.84 ppm in CZ8, and that
of H4 is 6.59 ppm in CZ4, 7.70 ppm in CZ6, and 8.51 ppm in
CZ8. To explain these observations, theoretical calculations at
the B3LYP/6-31G* level were performed to gain respective
energy minimized structures (Figure 3). The calculations have
revealed that CZ6 takes a nearly orthogonal conformation with
quite planar porphyrins, where 5,15-axes of the neighboring
porphyrins are aligned collinear with a 180° angle. In the energy
minimized structure of CZ4, each porphyrin ring is significantly
distorted as suggested by a mean plane deviation of 0.13 Å,
9
240 J. AM. CHEM. SOC. VOL. 127, NO. 1, 2005

Directly meso−meso Linked Porphyrin Rings
A R T I C L E S
Figure 4. STM image of CZ8 on Cu(100). (Inset) Height profile along the bar.
Chart 3. Structures of Linear Porphyrin Oligomers Z0n
images of CZ8. A dilute solution of CZ8 in CH2Cl2 was sprayed
by using a pulse injection method onto a clean flat Cu(100)
surface obtained by cycles of annealing and Ar-ion sputtering.15
In situ STM measurements were performed at room temperature
in a ultrahigh vacuum (<10-10 mbar) in a constant height mode.
It turned out, however, that the STM detection of CZ8 was
very difficult probably due to weaker adsorption of the molecule
on the metal surface. The STM images of CZ8 taken at a sample
bias voltage of 1.5 V and a tunneling current of 52 pA reveal
nonlinear but ellipsoid spots with a major axis of ca. 5.3 nm
and a minor axis of ca. 3.6 nm (Figure 4), which are
overestimated by ca. 1 nm in both axes because of a finite
curvature of the STM tip. The size estimated from the STM
image is roughly consistent with the calculated structure. An
estimated height of the STM images is ca. 5.7 Å, which is rather
higher compared with the STM images of other linear and cyclic
meso-meso linked porphyrin arrays.
and 5,15-axes of the neighboring porphyrins are bent to 164°.
In the calculated structure of CZ8, the porphyrin rings are
discernibly distorted as suggested by a less mean plane deviation
of 0.054 Å and 5,15-axes of the neighboring porphyrins are
bent to an angle of 187°. Importantly, the dihedral angles
between neighboring porphyrins in the energy minimized
structures are 90° in CZ6, 76° (inner) and 67° (outer) in CZ4,
and 100° (inner) and 105° (outer) in CZ8. In the exact
perpendicular conformation as expected for CZ6, the spatial
positions of H3 and H4 toward the neighboring porphyrin are
the same. This situation is similar to the parent meso-meso
linked diporphyrin Z02, in which the Ha proton (designated in
Chart 3) appears at 8.12 ppm, being similar to H3 (8.25 ppm)
in CZ6. Then, a more high field shifted chemical shift of H4
can be accounted for in terms of additional shielding effect of
the neighboring porphyrin ring. The observed large changes in
the chemical shifts of H3 and H4 in CZ4 and CZ8 can be
explained by the calculated structures, since the bending at the
meso-meso linkage causes different environments for H3 and
H4. In CZ4, H4 is forced to move closer and in a more shielding
region and H3 is shifted remote and in a deshielding region
toward the neighboring porphyrin. The trend is reversed and
weakened in CZ8, reflecting modest bending toward the
opposite direction. The similar features have been well dem-
onstrated in a series of strapped meso-meso linked dipor-
phyrins.19a
Excitonic and Electronic Couplings within Cyclic Por-
phyrin Arrays. The linear meso-meso linked zinc(II) porphyrin
arrays Z0n show exciton split Soret (S0fS2) transition bands
due to dipole-dipole exciton couplings between zinc(II) por-
phyrin units (Figure 5a and Scheme 1).10 Although two almost
degenerate transition dipoles Bx and By create one Soret band
in a zinc(II) porphyrin monomer, the dipole couplings between
adjacent zinc(II) porphyrin units, such as Bz - Bz at opposite,
Bx + By at orthogonal, and Bz + Bz at parallel, generate three
nondegenerate transition dipoles and concomitantly two separate
Soret bands in the directly linked zinc(II) diporphyrin (Z02).
Low energy Soret (Bz + Bz) and Q-bands are continuously red-
shifted with an increase in the number of porphyrins (Z04 <
Z06 < Z08). In sharp contrast, the porphyrin rings CZ4, CZ6,
and CZ8 exhibit characteristic broad Soret bands that are
centered around 460 nm and are slightly red-shifted in going
from CZ4 to CZ6 and CZ8 (Figure 5b and Table 1). These
observations can be well explained by their symmetric cyclic
structures, in which all the Bx and By dipole moments are
involved in the exciton coupling (Scheme 1). The symmetric
STM Images of CZ8. Recently we have succeeded in the
STM detection of the dodecameric porphyrin ring as a discrete
ring.13 Following the same method, we attempted to take STM
(19) (a) Yoshida, N.; Ishizuka, T.; Osuka, A.; Jeong, D. H.; Cho, H. S.; Kim,
D.; Matsuzaki, Y.; Nogami, A.; Tanaka, K. Chem.sEur. J. 2003, 9, 58.
(b) Shinmori, H.; Ahn, T. K.; Cho, H. S.; Kim. D.; Yoshida, N.; Osuka, A.
Angew. Chem., Int. Ed. 2003, 42, 2754.
9
J. AM. CHEM. SOC. VOL. 127, NO. 1, 2005 241

A R T I C L E S
Nakamura et al.
Scheme 1. Exciton Coupling in Monomer, meso-meso Linked Dimer, and Cyclic Hexamer CZ6
Table 1. Absorption and Fluorescence Data of CZ4, CZ6, and
CZ8 in THF
ring structures lead to the repeating same exciton coupling,
which causes a single Soret band. The broad and structureless
Soret bands indicate that there are some heterogeneities in
dipole-dipole couplings between the porphyrin units presum-
ably because of conformational flexibilities (Figure 3). On the
other hand, the Q absorption (S0fS1) tail reveals red-shifts in
going from CZ6 to CZ8 and CZ4. An interesting feature is a
sharp shoulder absorption band at 409 nm that is observed only
for CZ4.
absorptiona (nm)
fluorescenceb (nm)
Q(0,0) Q(0,1)
Φfc
soret
(high)
soret
sample
(low)
Q(1,0)
Q(0,0)
CZ4 409
(220 000) (310 000) (76 000) (31 000)
458 575 615
(550 000) (170 000) (22 000)
464 580 628
(740 000) (240 000) (64 000)
454
594
654
674 727 0.071
631 675 0.031
646 685 0.040
CZ6
CZ8
The fluorescence spectra of the linear (Z03, Z04, Z06, and
Z08) and cyclic arrays (CZ4, CZ6, and CZ8) are compared in
Figure 6. The linear porphyrin arrays show fluorescence (S1fS0)
a Molar extinction coefficient (M-1 cm-1) is presented in parentheses.
b λex ) 455 nm. c Quantum yield was determined with reference to the value
(Φf ) 0.033) of zinc(II) tetraphenylporphyrin in benzene.
spectra with two bands at 640 and 667 nm, and the intensity of
the higher energy band increases with an increase in the number
of porphyrins (Z03 < Z04 < Z06 < Z08). On the other hand,
the fluorescence spectra of the cyclic arrays look more diverse
in shape, band position, and intensity. The fluorescence of the
cyclic arrays reveal red-shifts in the order CZ6 < CZ8 < CZ4,
in line with their Q-band tail positions, and the fluorescence
intensity increases in the order CZ6 < CZ8 < CZ4. The band
positions and structures in the absorption and fluorescence
spectra suggest unique excitonic and electronic communications
between the constituent porphyrin units.
According to our previous reports, the electronic couplings
between porphyrin units can be well related to the dihedral angle
between porphyrin units.19 The strapped zinc(II) diporphyrin,
where the strapping carbon chain controls a dihedral angle
between two zinc(II) porphyrin units, shows that the smaller
dihedral angle between porphyrin units gives rise to the stronger
electronic coupling and concomitantly red-shifts in the Q
absorption and fluorescence spectra.19b As a consequence, the
Q absorption and fluorescence spectra indicate the electronic
coupling strength increases in the order CZ6 < CZ8 < CZ4 in
accordance with the dihedral angle estimated from the above
1H NMR data and geometry optimization. On the basis of the
energy minimized structures in Figure 3, we will use values of
72° for CZ4, 90° for CZ6, and 77° for CZ8 as the effective
Figure 5. Absorption spectra of porphyrin arrays in THF; (a) 5,15-meso-
meso linked linear oligomers Z04, Z06, and Z08 (absorbances are normalized
around 413 nm) and (b) CZ4, CZ6, and CZ8.
9
242 J. AM. CHEM. SOC. VOL. 127, NO. 1, 2005

Directly meso−meso Linked Porphyrin Rings
A R T I C L E S
Figure 6. Fluorescence spectra of CZ4, CZ6, and CZ8 taken for excitation
at 455 nm in THF.
dihedral angle between neighboring porphyrins in the following
discussion. It is interesting to note that the fluorescence of CZ6
is quite similar to that of Z03, in line with its estimated 90°
dihedral angle. The small absorption shoulder at 409 nm in CZ4
may be understood in terms of H-type dipole-dipole coupling
only occurring in zinc(II) porphyrin units with a small dihedral
angle, which gives rise to a more blue-shifted Soret band (409
nm) than the high energy Soret band (417 nm)20 of Z02. Finally,
we need to look into the exciton and electronic couplings from
a viewpoint of energy transfer. Because the cyclic arrays have
strong excitonic and electronic couplings between porphyrin
units, the coherent and/or incoherent energy hopping processes
are easily expected.
Fluorescence Lifetimes and Rotational Diffusion Motions
of Cyclic Porphyrin Arrays. The time-resolved fluorescence
decay profiles of the cyclic arrays are displayed in Figure 7,
and their fitted parameters are summarized in Table 2. The
fluorescence decays do not show any emission wavelength
dependence, implying single-exponential relaxations from the
lowest singlet S1 states of porphyrins. The fitted S1 state lifetimes
slightly depend on the size of the cyclic array, decreasing in
the order CZ4 < CZ8 < CZ6, which are consistent with their
steady-state fluorescence quantum yields (Tables 1 and 2). An
interesting feature is that the S1 state lifetime decreases in the
same order of the blue-shifts in the Q absorption and fluores-
cence spectra. According to the previous work on linear zinc-
(II) porphyrin arrays, the S1 state lifetimes gradually decrease
as the number of porphyrin units increases as a result of the
enhanced conformational heterogeneities in the longer arrays.12b,21
The S1 state lifetime of 1.84 ns in CZ4 is distinctly longer than
0.92 ns in linear zinc(II) porphyrin tetramer Z04.21 In addition,
Figure 7. Time-resolved fluorescence decay profiles of CZ4, CZ6, and
CZ8 measured by using the excitation wavelength 420 nm in THF.
Table 2. Fitted Fluorescence Lifetimes and Anisotropy Decay
Parameters of CZ4, CZ6, and CZ8 in THFe
fitted fluorescence lifetimea
b (ns)
anisotropy decay parametersc
d
sample
τ
r0
Φ
d (ns)
CZ4
CZ6
CZ8
1.84 ( 0.02 (100%)
1.76 ( 0.04 (100%)
1.80 ( 0.02 (100%)
0.036 ( 0.007
0.009 ( 0.005
-0.019 ( 0.005
0.94 ( 0.31
1.50 ( 0.50
2.88 ( 0.75
a The fluorescence lifetimes of the samples were obtained by averaging
the fitted single fluorescence lifetimes at several emission wavelengths.
b Using the relation I(t) ) A exp(-t/τ), where I(t) is the time-dependent
fluorescence intensity, A, the amplitude (noted in parentheses as the
percentage), and τ, the fitted fluorescence lifetime; the ø2 values of the
fittings were maintained as ∼1.0-1.3. c The fluorescence anisotropy decays
were monitored at λem ) 670 nm. d Using the relation r(t) ) r0 exp(-t/Φ),
where r(t) is the time-dependent fluorescence anisotropy [r(t) ) (I (t) -
||
GI (t))/(I (t) + 2G I (t)), r0, the initial anisotropy value, and Φ, the fitted
||
anisotropy decay time. e The excitation wavelength, 420 nm, was applied
to all experiments.
the dependence of the S1 state lifetime on the ring size is
negligible in comparison with the linear zinc(II) porphyrin array
Z0n.12b,21 These results indicate that the conformational motions
or the nonradiative relaxation processes of CZ4, CZ6, and CZ8
are considerably restricted in the ring structures. It is to be noted
that the increased S1 state lifetime has an apparent advantage
in the excitation energy transfer process.
The fluorescence anisotropy decay profiles of the cyclic arrays
are presented in Figure 8, where the excitation of the high energy
Soret band (λex ) 420 nm) was employed. The fitted decay
parameters are tabulated in Table 2. The anisotropy decay times
do not reflect the energy migration processes within CZ4, CZ6,
and CZ8, because their time scales (∼1.0-3.0 ns) are consider-
(20) Hwang, I.-W.; Cho, H. S.; Jeong, D. H.; Kim, D.; Tsuda, A.; Nakamura,
T.; Osuka, A. J. Phys. Chem. B 2003, 107, 9977.
(21) Cho, S. H.; Song, N. W.; Kim, Y. H.; Jeoung, S. C.; Hahn, S.; Kim, D.;
Kim, S. K.; Yoshida, N.; Osuka, A. J. Phys. Chem. A 2000, 104, 3287.
9
J. AM. CHEM. SOC. VOL. 127, NO. 1, 2005 243

A R T I C L E S
Nakamura et al.
Figure 8. Time-resolved fluorescence anisotropy decay profiles of CZ4,
CZ6, and CZ8. The polarized fluorescence decays (IVV(t) and IVH(t)) are
measured at the emission wavelength 670 nm, and then the anisotropy
decays r(t) are calculated by eq 1. The decays are obtained using the
excitation wavelength 420 nm in THF.
ably slow in view of the close proximity intermolecular
geometries. Instead, the decay times (Φ) are well associated
with the molecular rotational diffusion times in solution,
illustrating slower motions in larger cyclic arrays (Table 2). In
addition, the different initial r0 values plausibly result from the
different angles between absorption and emission dipoles as well
as the efficient energy transfer processes along the cyclic arrays.
EEH Processes in Cyclic Porphyrin Arrays. To investigate
the EEH process along the cyclic array, femtosecond transient
absorption (TA) and transient absorption anisotropy (TAA)
decays are simultaneously measured (Figures 9 and 10), where
the Q-band excitation (λpump ) 583 nm) is employed to avoid
an involvement of S2-S1 relaxation in the porphyrin. As
displayed in Figures 9 and 10, the cyclic arrays reveal intensity
dependent TA decays along with TAA rises in the time scale
of hundreds of femtoseconds, indicating exciton-exciton an-
nihilation and depolarization channels contributed by EEH
processes. The TA decays are deconvoluted with two decay
components (τ1 and τ2), where the slow τ2 components are fixed
as the S1 state lifetimes as revealed by the fluorescence decay
measurements (Table 3). The TAA rises are also deconvoluted
with one rising component, of which the fitted parameters are
inserted in Figure 10.
Figure 9. Transient absorption decay profiles of CZ4, CZ6, and CZ8 that
include pump-power dependences in THF, where the pump and probe
wavelengths are 583 and 460 nm, i.e., the Q-band pump and bleaching
recovery probe. The decays are measured with magic angle polarization
between pump and probe beams.
560 fs and 680 fs in going from CZ4 to CZ6 and CZ8, i.e., as
the size of the cyclic array increases. The pump-power
dependence on the TA decay is indicative of the exciton-
exciton annihilation process, because intense excitation or high
density of photons generates two or more excitons in one cyclic
array, and then the recombination process between them creates
a fast deactivation channel.22,23 According to the previous
description on the natural light-harvesting antennae such as LH1
and LH2, the exciton-exciton annihilation process is conceived
as a Fo¨rster-type incoherent EEH process from the excited donor
to the proximal excited acceptor resulting in a doubly excited
acceptor state; the latter quickly relaxes to the singly excited
state.22,23 Therefore, the exciton-exciton annihilation process
should be a direct evidence of the Fo¨rster-type incoherent EEH
process within the cyclic array. Figure 9 and Table 3 also
indicate that the exciton-exciton annihilation process is due to
(22) Bradforth, S. E.; Jimenez, R.; van Mourik, F.; van Grondelle, R.; Fleming,
G. R. J. Phys. Chem. 1995, 99, 16179.
The TA decays depend on the pump power as well as the
size of the porphyrin ring (Figure 9 and Table 3). As the pump
power is increased, the relative contribution of the fast τ1
component is enhanced as compared with the long-lived one.
In addition, the τ1 values systematically increase from 73 to
(23) (a) Trinkunas, G.; Herek, J. L.; Pol´ıvka, T.; Sundstro¨m, V.; Pullerits, T.
Phys. ReV. Lett. 2001, 86, 4167. (b) Trinkunas, G. J. Lumin. 2003, 102,
532. (c) Bru¨ggemann, B.; May, V. J. Chem. Phys. 2004, 120, 2325. (d)
Mu¨ller, M. G.; Hucke, M.; Reus, M.; Holzwarth, A. R. J. Phys. Chem.
1996, 100, 9537. (e) Bru¨ggemann, B.; Herek, J. L.; Sundstro¨m, V.; Pullerits,
T.; May, V. J. Phys. Chem. B 2001, 105, 11391.
9
244 J. AM. CHEM. SOC. VOL. 127, NO. 1, 2005

Directly meso−meso Linked Porphyrin Rings
A R T I C L E S
the size of the traveling ring increases. In addition, it is
interesting to note that the exciton migration time of 680 fs in
CZ8 is even faster than ∼1 ps22 in LH2.
The TAA measurement is also a helpful tool to investigate
the excitation energy migration process along the cyclic array,
because the EEH process even between the same molecular units
aligned in different direction creates a depolarization channel.
Figure 10 shows the TAA rises arising from the fast energy
migration processes along the cyclic arrays. The TAA rise times
systematically increase from 30 to 100 fs and 107 fs in going
from CZ4 to CZ6 and CZ8, indicating that the anisotropy rise
is also sensitive to the traveling ring size. In a multichro-
mophoric system, neither exciton-exciton annihilation nor
anisotropy depolarization time does directly represent the EEH
time, because they do not occur between single donor and
acceptor unit. The EEH time among the constituent chro-
mophores can be theoretically evaluated by modeling the
hopping process and the simultaneous observation of these two
observables.22,23
The description of the exciton-exciton annihilation and
anisotropy depolarization has been well developed for the natural
light-harvesting systems such as LH1 and LH2 that have wheel
arrangements with constant interchromophore interaction.22,23
In these complexes, an explicit Fo¨rster-type incoherent EEH
model has been described assuming a migration-limited char-
acter of the exciton-exciton annihilation process and random
walk formalism of the anisotropy decay. In this context, the
experimentally determined depolarization and exciton-exciton
annihilation times are connected with the EEH time by the
following equations
τhopping
τhopping
τdepolarization
)
)
(3)
(4)
4(1 - cos2(2π/N)) 4(1 - cos2 R)
Figure 10. Transient absorption anisotropy decay profiles of CZ4, CZ6,
and CZ8 in THF, where the polarized transient absorption decays for parallel
and perpendicular orientations between pump and probe beams are also
included. Insets show the deconvolution fitted decay parameters with a train
of 150-fs pump pulses. The pump and probe wavelengths are 583 and 460
nm, which are the Q-band pump and bleaching recovery probe.
N2 - 1
24
τannihilation
)
τhopping
where N is the number of hopping sites, R, the angle between
the adjacent transition dipoles, τannihilation, the slowest exciton-
exciton annihilation time, and τhopping, the inverse of the nearest
neighbor energy hopping rate. Equation 3 is understood by
considering that the depolarization is complete when the
transition dipole migrates through 90° and by considering how
many hops are required for this rotation to be accomplished.
On the other hand, eq 4 assumes that the exciton-exciton
annihilation reflects the migration limited exciton-exciton
recombination along the cyclic array and how many hops are
required for this recombination to be accomplished.
Now, we model the EEH process within the cyclic array and
simultaneously use the two different observables, exciton-
exciton annihilation and anisotropy decay times, to estimate the
EEH time between the hopping sites. Concerning the number
of hopping sites, it is conceived as N ) 4, N ) 6, and N ) 8
for CZ4, CZ6, and CZ8, respectively. Although the exciton
coherence via two porphyrin units rather than one porphyrin
unit is rationalized in the cyclic array, in view of cyclic geometry
and exciton coherence length of 4-5 porphyrin units12 in the
Q-state of the directly linked linear array Z0n, the numbers of
excitation energy hopping sites for the cyclic arrays are N ) 4,
N ) 6, and N ) 8, as a result of the bidirectional coupling with
the nearest neighboring porphyrins in the two-dimensional space
Table 3. Transient Absorption Decay Parameters for CZ4, CZ6,
and CZ8 Depending on Pump Powera
pump
fitted decay timesb (ps)
power
(mW)
τ1
τ2
CZ4
2.0
1.0
0.5
0.073 (14%)
0.073 (9%)
0.073 (8%)
1840 (86%)
1840 (91%)
1840 (92%)
CZ6
0.56 (17%)
0.56 (13%)
0.56 (11%)
2.0
1.0
0.5
1760 (83%)
1760 (87%)
1760 (89%)
CZ8
0.66 (44%)
0.70 (35%)
0.68 (19%)
2.0
1.0
0.5
1800 (56%)
1800 (65%)
1800 (81%)
a The pump and probe wavelengths are 583 and 460 nm, respectively.
b Using the relation -∆OD(t) ) A1 exp(-t/τ1) + A2 exp(-t/τ2), where
∆OD(t) is the transient absorption intensity, A, the amplitude (noted in
parentheses as the normalized percentage, i.e., [Ai/(A1 + A2)] × 100), and
τ, the fitted decay time. The solvent used was consistently THF.
the migration limited exciton-exciton recombination process
along the cyclic array, because this process becomes slow as
9
J. AM. CHEM. SOC. VOL. 127, NO. 1, 2005 245

A R T I C L E S
Nakamura et al.
Scheme 2. Excitation Energy Hopping Rates in CZ4, CZ6, and CZ8
of the cyclic arrays (Scheme 1). To describe the random walk
of the anisotropy decay, the orientations of the molecular
transition dipole moments should be considered. As shown in
Scheme 2, the transition dipoles that are attributable to the
change in anisotropy are arranged in the two-dimensional xy-
plane, where the anisotropy decay profile reflects the EEH
between the transition dipoles. This modeling indicates that the
EEH process within the cyclic array can be well described by
eqs 3 and 4, which are applicable to a well-defined cyclic
molecular array. Introducing R ) 90° and N ) 4 to eqs 3 and
4, the relations τhopping ) 4 × τdepolarization and τhopping ) 1.6 ×
demonstrating that the increased electronic interactions acceler-
ate the EEH processes between zinc(II) porphyrin units. We
attribute this trend primarily to the decreased dihedral angle
between the neighboring porphyrins in the order CZ6 > CZ8
> CZ4. Finally, it is interesting to note that the EEH times,
i.e., 119 ( 2 fs in CZ4 and 236 ( 31 fs in CZ8, are faster than
∼270 fs23b in LH2, because of the proximity and stronger
electronic couplings in the cyclic porphyrin arrays.
In summary, the directly linked porphyrin rings CZ4, CZ6,
and CZ8 were synthesized in a stepwise manner from 5,10-
diaryl zinc(II) porphyrin Z1. These porphyrin ring molecules
have been shown to serve as a platform to enable very efficient
EEH processes along the ring circuit with rates that rival those
of natural cyclic photosynthetic antenna. Future work will focus
on incorporation of these functional units into more elaborate
model systems and exploration of larger directly linked por-
phyrin arrays.
τannihilation are obtained for CZ4. Similarly, introducing R ) 60°
and N ) 6, the relations τhopping ) 3 × τdepolarization and τhopping
) 0.685 × τannihilation are obtained for CZ6. Finally, introducing
R ) 45° and N ) 8, the relations τhopping ) 2 × τdepolarization and
τhopping ) 0.38 × τannihilation are obtained for CZ8. Consequently,
the EEH time constants are calculated to be 120, 300, and 214
fs in CZ4, CZ6, and CZ8, respectively, using the anisotropy
rise times of τr ) 30, 100, and 107 fs given in Figure 10. In a
different approach, the EEH time constants are estimated to be
117, 384, and 258 fs using the annihilation times τ1 ) 73, 560,
and 680 fs given in Table 3. It is noteworthy that the two
different experimental observables, exciton-exciton annihilation
and anisotropy depolarization, result in the similar EEH times
between zinc(II) porphyrin units, which are 119 ( 2 fs in CZ4,
342 ( 59 fs in CZ6, and 236 ( 31 fs in CZ8 (Scheme 2). This
coincidence implies that the EEH processes within the cyclic
arrays are well described by the Fo¨rster-type incoherent energy
hopping model because of their well-defined dipole orientations.
Finally, the observed EEH times (119 ( 2 fs in CZ4, 342 (
59 fs in CZ6, and 236 ( 31 fs in CZ8) can be well associated
with the red-shifts in the Q absorption and fluorescence spectra,
Acknowledgment. The work at Kyoto was partly supported
by Grant-in-Aid for Scientific Research on Priority Area (No.
16033231, Reaction Control of Dynamic Complexes) from the
Ministry of Education, Culture, Sports, Science and Technology,
Japan and 21st Century COE on Kyoto University Alliance for
Chemistry. The work at Yonsei was supported by the National
Creative Research Initiatives Program of the Ministry of Science
and Technology of Korea.
Supporting Information Available: Experimental section, 1H
NMR spectra, absorption and fluorescence spectra, optical
resolution of racemic Z2, and GPC-HPLC analysis. This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA045254P
9
246 J. AM. CHEM. SOC. VOL. 127, NO. 1, 2005