A Light-Harvesting/Charge-Separation Model with Energy Gradient Made of Assemblies of meta-Pyridyl Zinc Porphyrins

Article abstract of DOI:10.1002/chem.202003327

Self-assembly of porphyrins is a fascinating topic, not only for mimicking chlorophyll assemblies in photosynthetic organisms, but also for the potential of creating molecular-level devices. Herein, zinc porphyrin derivatives bearing a meta-pyridyl group at the meso position were prepared and their assemblies studied in chloroform. Among the porphyrins studied, one with a carbamoylpyridyl moiety gave a distinct ¹H NMR spectrum in CDCl₃, which allowed the supramolecular structure in solution to be probed in detail. Ring-current-induced chemical-shift changes in the ¹H NMR spectrum, together with vapor-pressure osmometry and diffusion-ordered NMR spectroscopy, among other evidence, suggested that the porphyrin molecules form a trimer with a triangular cone structure. Incorporation of a directly linked porphyrin–ferrocene dyad with the same assembling properties in the assemblies led to a rare example of a light-harvesting/charge-separation system in which an energy gradient is incorporated and reductive quenching occurs.

Full text of DOI:10.1002/chem.202003327

Accepted Article
Accepted Article
Title: A Light-Harvesting/Charge-Separation Model with Energy
Gradient Made of Assemblies of meta-pyridyl Zinc Porphyrins
Authors: Joe Otsuki, Takumi Okumura, Kosuke Sugawa, Shin-ichiro
Kawano, Kentaro Tanaka, Takehiro Hirao, Takeharu Haino,
Yu Jin Lee, Seongsoo Kang, and Dongho Kim
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202003327
Link to VoR: https://doi.org/10.1002/chem.202003327
01/2020

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
A Light-Harvesting/Charge-Separation Model with Energy
Gradient Made of Assemblies of meta-Pyridyl Zinc Porphyrins
Joe Otsuki,*[a] Takumi Okumura,[a] Kosuke Sugawa,[a] Shin-ichiro Kawano,[b] Kentaro Tanaka,[b]
Takehiro Hirao,[c] Takeharu Haino,[c] Yu Jin Lee,[d] Seongsoo Kang,[d] and Dongho Kim[d]
[a]
Professor J. Otsuki, T. Okumura, and Dr. K. Sugawa
Department of Materials and Applied Chemistry, College of Science and Technology, Nihon University
1-8-14 Kanda Surugadai, Chiyoda-ku, Tokyo 101-8308 Japan
otsuki.joe@nihon-u.ac.jp
[b]
[c]
[d]
Dr. S. Kawano and Prof. K. Tanaka
Department of Chemistry, Graduate School of Science, Nagoya University
Furocho, Chikusa-ku, Nagoya 464-8602, Japan
Dr. T. Hirao and Professor T. Haino
Department of Chemistry, Graduate School of Advanced Science and Engineering, Hiroshima University
1-3-1 Kagamiyama, Higashi-Hiroshima, 739-8526 Japan
Y. J. Lee, S. Kang, and Prof. D. Kim
Spectroscopy Laboratory for Functional π-Electronic Systems and Department of Chemistry, Yonsei University, Seoul 03722, South Korea
Supporting information for this article is given via a link at the end of the document.
Abstract: Self-assembly of porphyrins is a fascinating topic not only
for mimicking chlorophyll assemblies in photosynthetic organisms but
also for the potential of creating molecular-level devices. Herein we
have prepared zinc porphyrin derivatives bearing a meta-pyridyl group
at the meso-position and studied their assemblies in chloroform.
Among porphyrins studied, one with a carbamoylpyridyl moiety
(ZnPPyC) gave distinct 1H NMR spectrum in CDCl3, which allowed us
to probe the supramolecular structure in solution in detail. Ring-
current induced chemical shift changes in 1H NMR, together with
vapor pressure osmometry and diffusion-ordered NMR spectroscopy,
among other evidence, suggested that the porphyrin molecules form
a trimer in a triangular cone structure. Further incorporation of a
directly-linked porphyrin-ferrocene dyads with the same assembling
properties in the assemblies lead to a rare light-harvesting/charge-
separation model, in which an energy gradient is incorporated and
reductive quenching occurs.
pyridylethynyl)porphyrin also forms a square in solution.[23]
Interestingly, the same molecules form linear polymer-like
coordination chains in the solid state as revealed by a crystal
structure.[24] Formation of supramolecular square using a para-
pyridyl unit was also applied to chlorophyll derivatives to prepare
square-shaped chlorophyll assemblies.[25-28]
Much less is known for the assembling behavior of zinc meso-
(meta-pyridyl)porphyrins. Aida et al. investigated assemblies of
porphyrins with a meta-pyridyl substituent at one of the meso
positions.[29] A crystal structure of a tetrameric assembly of one of
the zinc meta-pyridiylporphyrins was reported but the information
on the structures in solutions are not available except absorption
spectra and mass spectra showing m/z signals up to the tetramer.
A chlorophyll derivative bearing a meta-pyridyl group bridged by
a phenylene ethynylene at the 20 position was reported to form a
trimer.[30]
We have investigated the assembling behavior of some new
zinc meso-(meta-pyridyl)porphyrins in chloroform solutions. The
investigated porphyrins, which are shown in Scheme 1, include
those bearing meta-pyridyl (ZnPPy), a meta-(3-acetamidopyridyl)
(ZnPPyA), and meta-(3-carbamoylpyridyl) (ZnPPyC) substituents
at the meso-position. We found that the carbamoylpyridyl
derivative, among these porphyrin derivatives, gave well-resolved
1H NMR spectra, which allowed us to investigate the structure of
the assemblies in solution in detail. Furthermore, a light-
harvesting and charge separation model was constructed by
introducing a zinc carbamoylpyridylporphyrin bearing a ferrocene
Introduction
Porphyrins have been attracting interest of many researchers for
a long time as archetypal photo/electro-active compounds
[1]
.
Their unique photophysical properties are being explored as
sensitizers in light-harvesting and charge separation model
studies[2-5] as well as for solar cells[6-10] and medical
applications,[11] among others.[12-14] Organizing porphyrin
molecules is an important subject to take best advantage of
capabilities of porphyrins.[5, 15-17]
Axial coordination of nitrogen containing heterocycles such as
pyridine to zinc ion in the center of porphyrin macrocycle is one
the simplest interactions that can be used for assembling
porphyrins. Zinc complexes of porphyrin having pyridyl moieties
assemble themselves into supramolecular species by mutual
coordination. A particularly simple case is the self-assembly of
zinc meso-(para-pyridyl)porphyrin, in which only one meso-
position is substituted by the para-pyridyl group. In nonpolar
solutions, four of these molecules self-assemble into a square by
moiety into the assemblies. This is
supramolecular porphyrin-based
separation systems in which reductive quenching is involved.
a
rare example of
light-harvesting/charge
mutual
axial
coordination.[18-22]
Zinc
meso-(para-
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
p-C12H25OPh
signals. The structure in solution turned out to be of the motif
shown on the left in Figure 1.
O
N
N
N
N
a
M
p-C12H25OPh
N
N
O
X
N
N
N
N
ꢀꢁꢂꢁꢃꢄ ꢅ ꢄꢀ
ꢅꢃꢄꢁ
Zn
C12H25
O
p-C12H25OPh
N
H2PPy: M = 2H
ZnPPy: M = Zn
b
O
N
m-C8H17OPh
H
N
O
H
N
NHAc
N
X
O
NHAc
Zn
Zn
N
N
N
N
N
N
N
a
M
m-C8H17OPh
Zn
Zn
N
C8H17
O
X
O
N
m-C8H17OPh
H
N
O
N
H
Zn
O
H2PPyA: M = 2H
ZnPPyA: M = Zn
X
H
N
b
H
Zn
N
p-C12H25OPh
Zn
CO2Me
O
COX
N
N
N
N
N
a
M
p-C12H25OPh
Figure 1. Possible motifs of assembly.
N
O
C12H25
O
p-C12H25OPh
Scheme
1
displays the synthesis of meso-(meta-
H2PPyCO2Me: M = 2H, X = OMe
H2PPyCO2H: M = 2H, X = OH
H2PPyC: M = 2H, X = NH2
ZnPPyC: M = Zn, X = NH2
c
d
b
pyridyl)porphyrins which were investigated in the present study.
Alkoxy chains were introduced to confer enough solubility to the
porphyrins. These alkoxy chains were considered not to pose any
steric problem in either type of the assemblies shown in Figure 1.
The alkoxybendaldehyde derivatives were prepared from the
CO2Me
O
p-C12H25OPh
N
COX
N
N
N
N
corresponding
bromoalkane in DMF in the presence of K2CO3.
The A3B-type porphyrins were prepared by the classical Adler-
Longo by feeding alkoxybenzaldehyde,
protocol[35]
hydroxybenzaldehyde
precursors
and
e
O
M
Fe
N
Fe
NH
NH
p-C12H25OPh
C12H25
O
pyridinecarboxaldehyde, and pyrrole in a ratio of 3:1:4. The
desired A3B-type porphyrin was eluted as a second band in silica
gel chromatography following meso-tetra(alkoxyphenyl)porphyrin
and was isolated in 3–13% yields. For the preparation of ZnPyC,
methyl 5-formyl-3-pyridinecarboxylate was used in the porphyrin
formation reaction. The methyl ester was hydrolyzed by
KOH/THF/H2O into the free carboxylic acid,[36-37] which was then
transformed into the amide by treating with excess
carbonyldiimidazole (CDI) followed by addition of aqueous
ammonia.[38] Finally, zinc was inserted into the porphyrin by
mixing the free-base porphyrin and Zn(OAc)2 at room
temperature.[39]
H2PFcPyCO2Me: M = 2H, X = OMe
H2PFcPyCO2H: M = 2H, X = OH
H2PFcPyC: M = 2H, X = NH2
ZnPFcPyC: M = Zn, X = NH2
c
d
b
Scheme 1. Preparation of zinc meso-(meta-pyridyl)porphyrins derivatives.
Reagents and conditions: a) pyrrole, propionic acid, reflux (H2PPy: 3.3%;
H2PPyA: 13%; H2PPyCO2Me: 7.9%); b) Zn(OAc)2 or Zn(OAc)2•(H2O)2, CHCl3,
MeOH, rt (ZnPPy: 74%; ZnPPyA: 100%; ZnPPyC: 88%; ZnPFcPyC: 95%); c) (i)
KOH, THF, H2O, reflux, (ii) aq HCl (H2PPyCO2H: 71%; H2PFcPyCO2H: 93%);
d) (i) CDI, THF, reflux, (ii) aq NH3 (H2PPyC: 71%; H2PFcPyC: 98%); e) (1) TFA,
NH4Cl, CH2Cl2, rt, (ii) DDQ, (ii) Et3N (H2PFcPyCO2Me: 3.0%).
In the case of ZnPFcPyC, which is an A2BC-type porphyrin, we
chose the condensation of meso-(4-dodexyloxyphen-1-yl)-2,2'-
Results and Discussion
dipyrromethane
with
5-formyl-3-pyridinecarboxylate
and
Design and Synthesis
ferrocenecarboxaldehyde using TFA as an acid catalyst and
NH4Cl as a water scavenger in CH2Cl2, followed by oxidation by
DDQ.[40-41] Our concern was the scrambling of meso-substituents
during the macrocyclization. Particularly, it was inferred that the
electron-rich meso-substituent, i.e., ferrocene, had detrimental
The motivation of using acetamido (in ZnPPyA) and carbamoyl (in
ZnPPyC) moieties at the 3-position of the pyridyl group was to see
if axial coordination and hydrogen bonding work synergistically in
the binding of these molecules. The intermolecular hydrogen
bonding with simultaneous coordination interaction would result
in assemblies in which neighboring porphyrin macrocycles make
a shallow angle as shown on the right in Figure 1. Repetition of
these interactions would result in a large assembly in the form of
a ring,[31] which is reminiscent of the light-harvesting antenna
found in purple bacteria.[32-34] We found that, as discussed below,
the structural information could be obtained for the
carbamoylpyridyl derivative owing to its well-resolved 1H NMR
effects
promoting
scrambling.
Indeed,
meso-
tris(dodecyloxyphenyl)-(methoxycarbonylpyridyl)porphyrin, i.e.,
H2PPyCO2Me, was contained in the products, which was a clear
indication that scrambling took place. This byproduct, behaved
similarly to the desired H2PFcPyCO2Me on the SiO2 TLC and was
difficult to be removed by SiO2 chromatography. We could either
purify the product by GPC at this stage or proceed to the following
reactions with the mixture. In the latter case, the desired
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
1
compound was purified by GPC at the stage of H2PFcPyC. These
porphyrins generally gave clean, nice signals corresponding to M+
or MH+ without appreciable fragmentation in the laser desorption
ionization mass spectrometry without using any matrix (for MS
spectra, see supporting information). We found, however, that
signals are much smaller for ferrocene-containing porphyrins
under similar ionization conditions.
supramolecular assembly in solution. The H NMR spectrum for
ZnPPyC is displayed in Figure 2 together with the spectrum for
the free-base counterpart, H2PPyC. Zinc insertion is indicated by
the disappearance of the imine H signal observed at –2.76 ppm
for the free-base porphyrin. The signals for the free-base
porphyrin are symmetrical with respect to the pseudo-mirror plane
passing through the porphine–pyridyl bond as expected. Also,
there is no distinction between the two ortho-protons and two
meta-protons of the phenyl groups. The number of protons
assigned to each signal is, therefore, a multiple of two except for
protons of the pyridyl group. On the other hand, this symmetry is
lost in the zinc porphyrin, which suggests that some
intermolecular association occurs leading to a structure with a
lower symmetry and hindered rotation of meso-substituents.
Partial assignment of signals for the zinc porphyrin was possible
taking advantage of substantially different J-coupling constants
depending on the fragment the protons belong to. The J-coupling
constants between porphine protons, phenyl protons, and pyridyl
protons are approximately ~5 Hz, ~8 Hz, and ~2 Hz. The
fragments to which protons belong to are indicated on the peaks
in Figure 2b, as judged by the coupling constants which are
corroborated by the COSY spectrum (Figure S3). Particularly, the
Assembly Formation
The 1H NMR spectrum for ZnPPy in CDCl3 was untractably broad
(See Figure S1a). Modification of the pyridyl group by 3-
acetamido moiety (ZnPPyA) did not improve the situation (Figure
S1b). These broadening is most likely caused by formation of
some assemblies in solution. Indeed, the diffusion-ordered NMR
spectroscopy (DOSY) measurements[42-44] for solutions in CDCl3
(21 mM, 24 °C) gave values of diffusion constants (D) of 4.96 x
10–10 m2 s–1 and 2.77 x 10–10 m2 s–1 for H2PPyA and ZnPPyA,
respectively (Figure S2). These values of diffusion coefficients
translate into hydrodynamic radiuses of 0.83 nm and 1.49 nm,
respectively, according to the Stokes-Einstein formula:
4
ꢁꢀꢂ
relatively small long-range coupling among pyridyl protons, JH-H
( )
1
ꢀ =
6ꢃꢄꢅ
was found by inserting a delay time of 50 ms in the COSY
spectrum (long-range COSY). The pyridyl protons are located at
2.11, 3.37, and 6.68 ppm. These peaks are significantly shifted
upfield compared to those of the free-base porphyrin, for which all
three pyridyl protons are observed at >8.9 ppm. These large
upfield shifts indicate that the pyridyl moiety coordinates axially to
the zinc ion in the porphyrin macrocycle of another molecule.[47] It
is likely that the 2.11 and 3.37 ppm peaks with particularly large
upfield shifts are of protons ortho to the nitrogen. It is well known
that pyridine commonly coordinates to the zinc ion in porphyrin in
a perpendicular orientation.[48-49] The chemical shifts of the two
ortho-protons are expected to be quite similar if this was the case.
However, it is noteworthy that the chemical shift values for the
ortho protons in the present case differ more than 1 ppm, which
suggests that the environment around the coordination center is
rather unsymmetrical.
in which kB, T, and h (5.28 x 10–4 kg m–1 s–1 for CDCl3[45]) are the
Boltzmann constant, temperature, and the viscosity, respectively.
The former value of 0.83 nm for the free-base porphyrin is
consistent with the size of porphyrin molecule excluding the
alkoxy chains; the H–H distance between the para-hydrogens of
the phenyl moieties trans to each other of meso-
tetraphenylporphyrin is 1.80 nm.[46] The radius for the zinc
porphyrin counterpart is 1.8 times larger than that for the free-
base porphyrin, which indicates the formation of some distinct, i.e.,
not polymeric, supramolecular assembly. It was difficult to obtain
further insight into the supramolecular structure in solution,
however, because of the broad nature of the 1H NMR spectrum.
Gratifyingly, modification of the pyridyl group by the 3-
carbamoyl moiety sharpened up the peaks in the 1H NMR
spectrum, which allowed us to shed light on the structure of the
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
C12H25
O
ꢁꢄꢃ
ꢁ
ꢀ
ꢁ
ꢁ
ꢀ
ꢁ
CONH2
ꢁ
ꢀ
NH
N
ꢂ
p-C12H25OPh
ꢁ
ꢁ
N
HN
N
ꢀ
ꢁ
ꢃꢄꢀ
ꢅꢆꢄꢀ
ꢆꢄꢀ ꢆꢄꢀ ꢇꢆꢄꢀꢈꢂ ꢆꢄꢁ
ꢀ
ꢂ
ꢀ
ꢉꢃꢄ
p-C12H25OPh
ꢀꢀ
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0
-3.0
d ꢀꢁꢂꢂ
p-C12H25OPh
ꢁꢂꢃ
CONH2
N
N
Zn
p-C12H25OPh
N
N
N
ꢆꢄꢁ
ꢆꢄꢁ
ꢆꢄꢁ
ꢆꢄꢀ
ꢆꢄ
ꢆꢄꢀ
ꢆꢄꢀ ꢇꢆꢄꢀꢈꢀꢂ
ꢃꢄꢀ
ꢀ ꢃꢄꢀ ꢆꢄꢀ
p-C12H25OPh
ꢀꢀ
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0
-3.0
d ꢀꢁꢂꢂ
Figure 2. 1H NMR spectra in CDCl3. The colors represent the fragment to which a proton belongs (black: porphine, blue: pyridyl, red: dodecyloxyphenyl). (a) H2PPyC.
(b) ZnPPyC. Each of the dots represents a proton.
To get insight into the stability and possible dynamic nature of
the assembly, 1H NMR spectra for a 5 mM sample in CDCl3 were
taken in a wide range of temperatures from –50 °C to 50 °C.
Remarkably, there was almost no change in the spectra in this
wide temperature range, except for a little broadening at the
lowest and highest temperatures as shown in Figure S4. This
observation strongly suggests that the assembly comprises a
single species in solution, in which all molecules in the assembly
are equivalent in the NMR time scale. Otherwise, equilibrium
would shift from one to another if more than one species were
involved. The equivalence of constituent molecules imply that the
supramolecular assembly has a C3 symmetry on average. This
does not necessarily mean that the C3 symmetrical structure is
the most stable and the assembly stays in that structure but rather
the assembly can fluctuate about the C3 structure.
coordination by about 15 nm, the spectrum for the 0.1 mM sample
represents the axially coordinated zinc porphyrin, while the
spectrum for the 0.01 mM sample represents the dissociated free
zinc porphyrin, and the intermediate spectra indicate the transition
of equilibrium. The association constants for axial coordination of
pyridyl ligands and zinc porphyrins are typically on the order of a
few thousands M–1 in chloroform,[50] and hence the major
equilibrium transition of coordinated species to dissociated
species is expected to occur at around 1 mM. That this occurs at
less than 0.1 mM in the present case suggests the involvement of
some synergistic effect of intermolecular interactions, such as
cyclic ring formation.[20]
Size of the assembly.
The observation that the spectra shows little changes over the
wide range of temperature also indicates that the assembly is very
stable. This in turn implies that the assembly is a closed structure.
If the assembly consisted of an open structure, such as a linear
assembly, the number of associated molecules in the assembly
would be changed upon temperature variation. The stability was
DOSY experiments were conducted for H2PPyC and ZnPPyC to
assess the size of the assembly in CDCl3 (Figure S7). The
diffusion coefficients obtained for H2PPyC and ZnPPyC are 4.3 x
10–10 m2 s–1 and 2.2 x 10–10 m2 s–1, respectively. These values
translate into hydrodynamic radiuses of 0.96 nm and 1.88 nm,
respectively. The former value is again consistent with the
monomeric porphyrin, while the latter is twice as large.
1
also shown by H NMR spectra for diluted samples. There was
practically no change in the spectra down to 0.1 mM from the
spectrum at 5 mM, as shown in Figure S5. At a concentration as
low as 0.02 mM, the spectrum showed a new pattern.
Absorption spectra corroborate that the axially coordinated
species dominate at 0.1 mM. The spectra for ZnPPyC in
chloroform in the Q-band region exhibits blue-shifts from a set of
566 nm and 610 nm to a set of 552 nm and 592 nm, respectively,
upon dilution from 0.1 mM to 0.01 mM as shown in Figure S6. As
the Q-band peaks for zinc meso-tetraphenylporphyrin appear at
550 nm and 593 nm, which are red-shifted by pyridine
The molecular weight information for the species in chloroform
solution was obtained by the vapor pressure osmometry, as
shown in Figure S8. The molecular weight of H2PPy, which is
incapable of coordination or hydrogen bonding, was obtained as
1312±48, which is close to the formula value of 1169. On the other
hand, the molecular weight for ZnPPyC was obtained as
4260±220, which is 3.3 times as large as the formula value of
1275, suggesting that the assembly is a trimer.
4
This article is protected by copyright. All rights reserved.

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
Further Confirmation of Trimer Formation — Quenching
Experiments
no new species was detected by repeated and prolonged 1H NMR
measurements in CDCl3. Fluorescence was measured by exciting
at 598 nm, one of the isosbestic points. The intensities were read
at 730 nm, a long wavelength foot of the fluorescence spectra
(Figure S10) to avoid the effect of reabsorption. The plot of
fluorescence intensities versus x is shown in Figure 3b. Also
shown in the graph are lines of (1–x)n, where n is one to eight from
top to bottom. The observed fluorescence intensities are on the
curve of (1–x)3, which supports that the assembly is a trimer.
Even though the fluorescence intensities are diminished with
increasing fraction of ZnPFcPyC, the fluorescent lifetimes were
found to be practically constant, 1.16 ns, 1.23 ns, and 1.31 ns for
samples of x = 0, 0.2, and 0.3, respectively (Figure 3c), which is
consistent with the assumption that the singlet excited state in a
trimer containing a ferrocene unit is effectively quenched and the
fluorescence observed is from the same species, i.e., (ZnPPyC)3.
The short lifetime due to quenching should be much faster than
our instrument limit (<0.1 ns). This is expected on the basis of the
ultrafast electron transfer in subpicosecond time scale reported
To confirm the trimer formation, we performed a quenching
experiment using ZnPFcPyC. The idea behind this experiment
was based on three assumptions as follows. (1) As this molecule
is similar in shape to ZnPPyC and has the same components
essential for the self-assembly as ZnPPyC, i.e., the zinc ion and
carbamolypyridyl moiety, this molecule would assemble in the
same motif as ZnPPyC. This assertion was consistent with its 1H
NMR spectrum in CDCl3, which is shown in Supporting
Information. The spectrum indicated the loss of symmetry, which
appears to be similar to the case for ZnPPyC, although,
unfortunately, the peaks were not sharp enough to allow their
assignments. When ZnPPyC and ZnPFcPyC are mixed, they
would form statistically mixed assemblies. (2) The ferrocene unit
of ZnPFcPyC would quench the singlet excited state of porphyrin,
which renders the molecule nonfluorescent via reductive electron
transfer, which was confirmed for directly-linked porphyrin-
ferrocene conjugates[40, 51] Time-resolved studies demonstrated
for similar directly-linked porphyrin–ferrocene dyads, as
52]
that fast electron transfer occurs in the subpicosecond time scale
mentioned earlier,[37,
as well as our own measurements as
52]
in similar directly-linked porphyrin–ferrocene dyads.[37,
The
described later.
ultrafast timescale of the electron transfer indicates that the
process occurs before solvation, which makes solvent polarity
irrelevant; the process occurs even in nonpolar benzene.[52]
Introduction of a ferrocene group was manifested by a new redox
wave for ZnPFcPyC at +0.07 V vs ferrocenium/ferrocene, which
was less positive than the purely porphyrin-based first oxidation
at +0.38 V for ZnPPyC (Figure S9). (3) We expected that Förster-
type energy transfer would occur among porphyrin units
associated with axial coordination of a meso-pyridyl moiety much
faster than the singlet excited state lifetime of zinc porphyrins,
which was confirmed later as discussed in the last part of Results
and Discussion.[19] Therefore, energy transfer from one porphyrin
molecule to another in the assembly should occur many times
over before decaying into the ground state. The consequence of
assumptions 1–3 is that one molecule of ZnPFcPyC in the
assembly would be enough to quench the fluorescence of all the
other porphyrin molecules in the n-mer assembly. The only n-mer
that can emit fluorescence is then (ZnPPyC)n without a ferrocene
unit among various n-mers (ZnPPyC)n–k(ZnPPyC)k (k = 0, 1, ..., n).
The fraction of (ZnPPyC)n among all the n-mers in the statistical
mixture is (1–x)n, when ZnFcPPyC and ZnPPyC are mixed in a
ratio of x to 1–x. We prepared chloroform solutions with varied x
with the combined concentration being constant at 0.1 mM:
2
(a)
x
1.0
0.8
0.6
0.4
0.2
0.0
1
0
500
550
600
650
700
750
l / nm
1
(b)
n = 8 76 5 4
3
2
1
0
0
0.2
0.4
0.6
0.8
1
x
500
(c)
x = 0
x = 0.2
x = 0.4
[
]
[
]
ZnPFcPyC
ZnPFcPyC
0.1 mM
( )
2
ꢆ =
=
[
]
[
]
ZnPFcPyC + ZnPPyC
and fluorescence was measured for each solution.
0
Figure 3 collects the results of relevant spectroscopic studies
for mixtures of ZnPPyC and ZnPFcPyC in chloroform. The
absorption spectra for the mixture of ZnPPyC and ZnPFcPyC with
the total concentration of 0.1 mM showed four isosbestic points at
544 nm, 578 nm, 598 nm, and 616 nm (Figure 3a). We noted that
there was an irregularity around 700 nm: a peak was observed for
the spectrum of ZnPPyC, which grew with time. Although the
precise nature of this peak is unclear, it is likely that the fraction
of molecules that is responsible for this absorption is very small,
because clean isosbestic points are observed at different
wavelengths and the molecules/assemblies should be stable as
0
5
10
t / ns
Figure 3. Spectroscopy for mixtures of ZnPPyC (fraction: 1 – x) and ZnPFcPyC
(fraction: x) (combined to be 0.1 mM) in chloroform. (a) Visible absorption
spectra at 25 °C. (b) Fluorescence intensity at 730 nm excited at 598 nm at
25 °C. Lines show (1 – x)n, with the values of n are indicated. (c) Fluorescence
lifetimes at room temperature (~25 °C). The black dots represent the instrument
response function.
5
This article is protected by copyright. All rights reserved.

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
Figure 4 depicts the system schematically. ZnPPyC and
ZnPFcPyC together assemble into trimers with statistical ratios.
In the assemblies containing one or two molecules of ZnPFcPyC,
the ferrocene unit not only quenches the fluorescence of the
porphyrin to which it is covalently bonded but also the
fluorescence of the other porphyrins in the assembly, either
through stepwise energy and electron transfer or through direct
intermolecular electron transfer. Thus, the porphyrin
(see Supporting Information for details). Possible supramolecular
structures were searched by changing the structural parameters
defined as in Figure 5a. Specifically, (1) the porphyrin center was
placed on the x-axis away from the origin by x, i.e., at (x, 0, 0), (2)
the rotation of each meso-substituent about the porphine–
substituent bond was allowed (qPyr, qPh1, qPh2, and qPh3), (3) the
orientation of the porphyrin plane was adjusted by three angles (q,
f, and g), and finally, (4) the second and third porphyrin molecules
were generated by rotating ±120° about the z-axis. Then the ring
current effects on every proton by all the aromatic systems
including the porphine, phenyl, and pyridyl rings were calculated
using a classical magnetic dipole model.[47, 56-57] As the one-by-
one assignment of 1H NMR signals to individual protons was not
possible but only the categorization of peaks to the protons of
porphine, pyridyl, and phenyl fragments was made, the least-
squares fitting of the calculated chemical shits to the observed
shifts was done on the fragment basis. In addition to the
parameters listed above, effects of bending of porphine–pyridyl
bond as well as the doming of porphyrin plane[58] were also
investigated, but a distortionless, planar conformation gave the
best result. Details are described in Supporting Information.
The calculated chemical shifts as a result of the least-squares
fitting agreed reasonably with the observed shifts as shown in
Figure 5b, while the trimer structure on which the calculated
chemical shifts are based is shown in Figure 5c. The overall
shape of the core part is an open triangular cone, in which three
porphyrin molecules constitute the three faces and the remaining
face is open. This core part is highlighted in Figure 5d, in which
the triangular cone is shown from the open side and from the tip
side, both along the C3 axis towards the reader.
chromophores work as
a light-harvesting antenna in the
antenna/charge separation assemblies.[53-55] We will discuss this
aspect in the last part of Results and Discussion.
CONH2
CONH2
Zn
Zn
N
Zn
N
Energy
N
N
Zn transfer
H2NOC
H2NOC
One way
N
N
Zn
Zn
One way
CONH2
CONH2
Fe
Electron transfer
(ZnPPyC)2(ZnPFcPyC)
(ZnPPyC)3
3x (1 – x)2
(1 – x)3
nonfluorescent
fluorescent
Fe
CONH2
CONH2
Zn
N
Zn
N
N
Zn
Zn
N
H2NOC
Fe
H2NOC
Fe
N
N
Zn
Zn
CONH2
CONH2
Fe
Fe
The calculated shifts from the trimer model reproduced some
key points. First, upfield shifts of pyridyl protons ortho to the
nitrogen, both large but to significantly different degrees were
reproduced. It is now possible to assign the most upfield (2.11
ppm) and the second upfield (3.37 ppm) shifts to the 3- and 1-
protons of the pyridyl moiety. The different degrees of the upfield
shifts are due to the offset, slanted axial coordination of the pyridyl
moiety to the zinc ion. Second, the lowered symmetry was also
reproduced, with individual proton signals appearing separately
one by one, not necessarily as a multiple of two protons. This is
due to the cone shape in which the pseudo-mirror plane through
the pyridyl-porphine bond is lost. Third, an exceptionally large
upfield shift (5.13 ppm) of one porphine proton was reproduced,
allowing its assignment to the 2-proton of the porphine core, which
is adjacent to the pyridyl moiety in the narrow side of the cone,
(ZnPPyC)3
x3
nonfluorescent
(ZnPPyC)(ZnPFcPyC)2
3x2 (1 – x)
nonfluorescent
Figure 4. Schematic drawing of the light harvesting/charge separation model.
Formulas indicate the fractions of species in the statistical mixture when
[ZnPPyC]:[ZnPFcPyC] = (1–x):x. Energy and electron transfer pathways are
indicated by curved arrows.
Structure of the trimer
So far, the data indicate that the assembly of ZnPPyC in
chloroform is (1) a single species, (2) a trimer, (3) a closed
structure, (4) C3 symmetrical, i.e., all molecules are equivalent on
average, (5) a structure in which the porphyrin's pseudo-mirror
symmetry is lost, and (6) a structure in which porphyrins are
mutually axially coordinated. Structures similar to the reported
which is the closest to another porphyrin macrocycle.
A
discrepancy was noted on a set of phenyl protons at 5.37 ppm
and 5.99 ppm, which were J-coupled according to the COSY
result. One proton (proton-12) appeared nearby at 5.61 ppm but
the J-coupled counterpart (proton-11), appeared at 8.86 ppm by
calculation. Proton-11 and proton-12 are protons on the phenyl
moieties clogging at the tip of the cone, which are seen in the
center region of the lower model of Figure 5d. Thus, it is likely that
their chemical shifts are quite sensitive to a slight movement
about the C3 axis passing through the tip of the triangular cone,
which may account for the discrepancy of the calculated chemical
shift value of proton-11. Finally, it is seen that carbamoyl moiety
is not involved in any intermolecular interaction in the trimer.
tetrameric X-ray crystal structure for
a zinc meta-pyridyl
porphyrin[29] can be excluded not only because ZnPPyC
molecules form a trimer but because if that was the case there
should have been a distinction of inner and outer pyridyl groups,
while only one set of pyridyl protons was actually observed.
Further insight into the structure of the assembly was obtained
by analyzing the chemical shits values in the 1H NMR spectrum,
which gives a rich information on the structure of porphyrin
assemblies because of large ring-current effects of the porphyrin
macrocyle. The parameters (number, magnitudes, positions, and
orientations) for the magnetic dipoles equivalent to zinc porphyrin
macrocycles were well established,[47, 56-57] which we used here
6
This article is protected by copyright. All rights reserved.

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
q
ꢄꢁꢆꢅꢂꢇ
ꢅꢈꢁ3
ꢉ ꢊꢋ
ꢌ20ꢍ
ꢀꢁꢂ
(a)
f
C3-axis
z
1021ꢀꢁꢂꢃ
ꢂ02ꢍꢅꢄ
qPh1
y
qPh3
g
x
1021ꢀꢁꢂꢃ
1ꢄꢅꢂꢃ
N
qPy
(x, 0, 0)
152ꢁꢇ5ꢄꢎꢏꢐꢑꢏꢒ
ꢅꢈꢁ3ꢎꢉ ꢓꢋ
152ꢁꢇ5ꢄꢎ1ꢄꢅꢂꢃꢎꢉꢔꢊꢋ
ꢌ20ꢍ 1021ꢀꢁꢂꢃ 1ꢄꢅꢂꢃ
qPh2
ꢌ20ꢍ 1021ꢀꢁꢂꢃ 1ꢄꢅꢂꢃ
(b)
12
13
Figure 6. Distortion seen in the PM6-optimized trimer structure. The alkoxy
chains were replaced by hydrogen atoms in the calculation. Hydrogen atoms
are omitted for clarity in the drawing.
11
14
6
4
8
2
1
32 31
33 34
2
1
4
5
8
3
N
3
7
1
14
5
3
6
7
32 22
33 23
31
34 24
21
21
24
22
23
Photosynthetic Antenna/Reaction Center Model
13
2
11
12
2
1
3
As the supramolecular trimer structure has been revealed, the
rate of Förster energy transfer among porphyrins in the trimer was
estimated based on the structural information according to the
formula (see Supporting Information for details including the
definitions of variables):[61]
9
7
5
3
1
d / ppm
(c)
(d)
ꢆ
(
)
( ) ( )
9 ln10 ꢇ Φꢇ
ꢋꢌ ꢌꢉꢍ ꢌ ꢎꢋ
ꢌ
∫
ꢌ
( )
3
ꢁꢁöꢂꢃꢄꢅꢂ
=
128ꢃꢈꢈꢉꢀꢊꢉꢋꢊꢌ
( )
ꢋꢌ ꢍ
ꢌ
∫
ꢌ
The calculations indicate that the energy transfer from one
ZnPPyC to another ZnPPyC occurs in 7.0 ps and that from
ZnPPyC to ZnPFcPyC occurs in 8.1 ps. These values are shorter
that the singlet excited state lifetime of 1.2 ns more than two-
orders of magnitude. Interestingly, the back energy transfer rate
from ZnPFcPyC to ZnPPyC is expected to be negligibly small
because (1) the fluorescence yield of ZnPFcPyC is negligibly
small (≲1% compared to ZnPPyC) and (2) the singlet excited
state energy of ZnPFcPyC (1.92 eV) is lower than that of ZnPPyC
(2.00 eV) as estimated from the absorption (Figure 3a) and
fluorescence spectra (Figure S10). For a trimer containing both
ZnPPyC and ZnPFcPyC, once ZnPPyC absorbs a photon,
efficient energy transfer to another porphyrin occurs. While
energy transfer between ZnPPyC molecules are reversible,
energy transfer from ZnPPyC to ZnPFcPyC is one way.
Figure 5. Trimer model which accounts for the observed 1H NMR spectrum in
CDCl3. (a) Parameters to be optimized. (b) Observed (solid lines) and calculated
(dotted lines) chemical shits. (c) NMR-fitted trimer qPy = 163°, qPh1 = –3°, qPh2
=
–1°, qPh3 = 4°, x = 4.98 Å, q = 72.5°, f = –140.7°, g = 117.3°. Dodecyloxy chains
are put in arbitrary orientations. The hydrodynamic sphere (gray) obtained from
the DOSY is superimposed. (d) Dodecyloxy groups are omitted to highlight the
core part. The upper is a view from the open side (inner surface) and the lower
is a view from the tip side (outer surface).
Subsequently, the singlet excited state of ZnPFcPyC is quickly
quenched by a process involving the ferrocene moiety. To obtain
insight into this fast process, we conducted femtosecond transient
absorption measurements. Figure 7a shows the transient
absorption spectra during the initial 200 ps duration after
excitation of ZnPFcPyC by a laser with a pulse width of ~35 fs.
The ground-state bleaching at ~580 nm and ~650 nm
corresponds to the Q-bands of ZnPFcPyC. The broad absorption
feature was observed to the low energy side of the bleach, which
We checked if an energy-minimized structure was found in the
triangular cone structure by semiempirical quantum mechanical
calculations using the PM6 method,[59] which has been applied to
axially-coordinated zinc porphyrin systems.[60] The optimized
structure is shown in Figures 6 and S11. The overall, near C3
structure is similar to the one obtained from the ring-current
simulation. Some distortions are observed to realize three
coordination bonds in the congested region. Most prominently,
the pyridyl moieties are warped with respect to the porphine
plane; the angles made by the porphine plane and the pyridyl C2–
C5 axis (C2 and C5 are the farthest and closest carbon atoms
from the porphine ring) are 18.8°–19.2°. In addition, the pyridyl
ring is rotated from the orthogonal orientation with respect to the
porphine plane by 28–31°. Further, the ligand axes (pyridyl N–C4
axis) are slanted with respect to the coordinated porphine planes,
by 10.1°–10.2°. Apparently, synergy of the triple coordination
bond formation outweighs these higher energy distortions in the
supramolecular conformations, resulting in the trimer, which are
stable in a wide concentration and temperature ranges.
52, 62]
may be assigned to the zinc-porphyrin radical anion.[37,
Transient absorption spectra of ZnPFcPyC were fitted with three
decay components and the time trances at two representative
wavelengths are shown in Figure 7b.
The fast relaxation process can represent the formation of the
charge-separated state, but the exact time constant for this
subpicosecond process could not be determined because it was
shorter than our instrumental resolution. The time constant of 2.3
ps could represent the conformation changes after the formation
of the charge-separated state. The longest time component of 26
ps indicates the charge recombination process. These
assignments are fully consistent with the literature for the above
7
This article is protected by copyright. All rights reserved.

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
52]
mentioned directly-linked porphyrin-ferrocene dyads[37,
and
unaltered over a wide range of temperatures from –50 °C to
+50 °C and a wide range of concentrations from 0.1 mM to 5 mM.
The formation of assembly was evidenced by an increased
hydrodynamic radius twice as large as that of a monomer. This
molecule could have had either one of the two motifs of assembly.
In one, hydrogen bonding and axial coordination synergistically
produce shallowly angled (30°) porphyrin association that would
lead to a large ring structure, which is reminiscent of the light-
harvesting antenna in purple bacteria. In the other, a smaller ring
would result if axial coordination dominates and hydrogen
bonding was not involved. Molecular weight determination by
vapor pressure osmometry indicated that the assembly was a
trimer. The results presented here excluded the possibility of a
large ring formation. Simulation of ring-current-induced chemical
shift changes indicated that the structure of the trimer was a
triangular cone, of which the three faces were occupied by the
porphyrin molecules and the remaining one face open.
A ferrocene-containing molecule, ZnPFcPyC, with the same
intermolecular interaction units was mixed with ZnPPyC to
generate statistical mixtures containing both the sensitizer units
and charge-separation units. The quenching experiment not only
confirmed that the assembly was a trimer but exemplified a rare
system of light-harvesting antenna/charge separation complexes
in which an energy gradient is incorporated and reductive
quenching occurred very efficiently (practically 100% efficiency).
We have previously reported a light-harvesting antenna/charge
separation system, which resulted from a statistical assembly of
none of these fast processes was present for ZnPPyC without the
ferrocene unit (Figure S12).
0.025
Time / ps
ꢀꢁꢂ
0.3
0.5
1
2
0.020
0.015
0.010
0.005
0.000
-0.005
-0.010
-0.015
5
10
20
30
50
70
90
100
120
200
400
500
600
700
800
Wavelength (nm)
0.002
0.008
0.006
0.004
0.002
0.000
λprobe : 775 nm
τ1: 2.3 ps (-38.2%, rise)
τ2: 26.3 ps (61.8%)
ꢀꢃꢂ
0.001
0.000
-0.001
-0.002
-0.003
λprobe : 585 nm
τ1: 2.3 ps (-35.6%, rise)
τ2: 26.3 ps (64.4%)
0
100
200
300
400
500
0
100
200
300
400
500
Time (ps)
Time (ps)
Figure 7. Femtosecond transient absorption for ZnPFcPyC in toluene. (a)
Transient spectra. (b) Time traces at 775 nm and 585 nm.
a
fluorescent chlorophyll derivative and a nonfluorescent
chlorophyll–fullerene conjugate.[26] As the fullerene unit
oxidatively quenches the fluorescence, the present system is a
complementary system, in the sense that the porphyrin
fluorescence is reductively quenched.
The energies of the resulting intramolecular charge-separated
state was estimated to be 2.07 eV by using the Weller equation,
on the assumption that the positive and negative charges are
localized on the ferrocene and porphyrin sites, respectively (see
Supporting Information for details including the definitions of
variables):
Directing assembly of porphyrin molecules toward a larger
antenna such as the light-harvesting antennas in purple bacteria
and the one shown on the right in Figure 1 is a fascinating
challenge. We can draw some lessons from this work to realize a
large ring-like antenna structure based on improvement of the
present design. Here we have seen that axial coordination
dominates over hydrogen bonding. Hence, one way may be to
use a stronger auxiliary interaction by making modification on the
part of the carbamoyl moiety. We have also seen that three axial
bonds are realized by a congested region with forced distortions.
Therefore, another way to direct the assembly towards the large
ring formation may be to hinder the distortion, say, by introducing
substituents on or around the pyridyl moiety to restrict its rotation.
These lines of attempts are underway in our laboratory.
ꢐ
( )
4
ꢏ = ꢐ ꢑꢏꢍꢎ − ꢏꢂꢅꢏ
−
ꢒ
4ꢃꢎꢂꢎꢐꢀ
This value (2.07 eV) is slightly larger than the singlet excited state
energy of ZnPFcPyC (1.92 eV). This may be due to
approximations involved in the estimation from combination of
different methods, i.e., electrochemistry and spectroscopy.
Particularly, it is likely that the charges are delocalized over the
ferrocene unit and porphyrin unit, lowering the energy of the
charge-separated state.[37] In the meantime, energies of various
intermolecular charge-separated states, e.g., (ZnPPyC)•–
(ZnPFcPyC)•+, were all found to be higher than the intramolecular
charge-separated state (Table S3). Thus, the trimer consisting of
ZnPPyC and ZnPFcPyC represents a complex of a light-
harvesting antenna with an energy gradient and a charge
separation reaction center.
Experimental Section
Thin layer chromatography (TLC) was performed on Merck silica
gel 60F254. Flash column chromatography was performed using
Kanto Chemical silica gel (SiO2) 60N. Recycle gel permeation
chromatography (GPC) was conducted, using CHCl3 as the
eluent, on a Japan Analytical Industry LC-9201 system equipped
with Jaigel-2H and Jaigel-1H columns in tandem, with exclusion
limits of 5000 and 1000, respectively. NMR spectra were recorded
with a JEOL JNM-ECX400 NMR spectrometer (400 MHz for 1H
and 100 MHz for 13C). Laser-desorption (LDI)-HRMS data were
recorded with a JMS-S3000 MS spectrometer. The spectra were
calibrated with a mixture of 4-methoxy-N-phenylbenzenamine (M+
Conclusion
We have investigated the self-assembly of some zinc meso-
(meta-pyridyl)porphyrins. Among them, one with a carbamoyl
group on the pyridyl moiety, ZnPPyC, was found to give a distinct
1H NMR spectrum in CDCl3 which allowed us to investigate the
1
structure in solution in detail. The H NMR spectra were nearly
8
This article is protected by copyright. All rights reserved.

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
(C13H13NO)
(C44H31N4),
199.0992),
615.2543)
meso-tetraphenylporphyrin
and Zn meso-tetrakis(p-
(MH+
was chromatographed (SiO2, CHCl3 → CHCl3/MeOH 98:2) and
further purified by GPC to afford a purple solid (61 mg, 0.058
mmol, 13%). Rf = 0.45 (SiO2, CHCl3/MeOH 9:1); 1H NMR (CDCl3):
δ = 9.18 (br, 1H), 9.08 (br, 1H), 8.94 (d, J = 4.9 Hz, 2H), 8.90 (s,
5H), 8.81 (d, J = 4.9 Hz, 2H), 7.75–7.8 (6H), 7.63 (t, J = 7.9 Hz,
3H), 7.56 (s, 1H), 7.33 (dd, J = 8.5 Hz, 1.8 Hz, 3H), 4.14 (t, J =
6.7 Hz, 6H), 2.29 (s, 3H), 1.87 (quintet, J = 6.7 Hz, 6H), 1.50
(quintet, J = 7.9 Hz, 6H), 1.2–1.4 (24H), 0.85 (t, J = 6.7 Hz, 9H),
–2.83 ppm (s, 2H); HRMS (LDI): m/z calcd for C69H80N6O4+H+:
1057.6314 [M+H]+; found: 1057.6225.
ZnPPyA. A combined solution of H2PPyA (60 mg, 0.06 mmol) in
CHCl3 (7 mL) and Zn(OAc)2 (60 mg, 0.33 mmol) in MeOH (3.5
mL) was stirred at rt for 4 h. The solvents were evaporated and
the residue was redissolved in CHCl3, which was washed with
H2O three times, dried over Na2SO4, filtered, and evaporated. The
obtained material was further dried under vacuum to afford a
purple solid (60 mg, 0.06 mmol, 100%). Rf = 0.5 (SiO2,
CHCl3/MeOH 9:1); HRMS (LDI): m/z calcd for C69H78N6O4Zn+:
1118.5371 [M]+; found: 1118.5284.
H2PPyCO2Me. A solution of p-dodecyloxybenzaldehyde[63] (10.8
g, 37.2 mmol), methyl 5-formyl-3-pyridinecarboxylate[66-67] (2.05 g,
12.4 mmol), and pyrrole (3.32 g, 49.6 mmol) in propionic acid (200
mL) was refluxed in an open flask covered with aluminum foil for
2.5 h. The solvent was evaporated and the residue was dissolved
in CHCl3, which was washed with aqueous NaHCO3 twice and
with aqueous NaCl once, dried over Na2SO4, filtered, and
evaporated. The crude material was chromatographed (SiO2,
CHCl3) twice to afford a purple solid (1.20 g, 0.983 mmol, 7.9%).
Rf = 0.5 (SiO2, CHCl3); 1H NMR (CDCl3): δ = 9.65 (d, J = 1.8 Hz,
1H), 9.60 (d, J = 2.4 Hz, 1H), 9.11 (t, J = 2.1 Hz, 1H), 8.94 (d, J =
4.9 Hz, 2H), 8.89 (s, 4H), 8.69 (d, J = 4.9 Hz, 2H), 8.11 (d, J = 8.5
Hz, 6H), 7.28 (d, J = 8.5 Hz, 6H), 4.25 (t, J = 6.7 Hz, 6H), 4.05 (s,
3H), 1.98 (quintet, J = 7.3 Hz, 6H), 1.63 (quintet, J = 7.3 Hz, 6H),
1.3–1. 5 (m, 48H), 0.90 (t, J = 6.7 Hz, 9H), –2.76 ppm (s, 2H);
HRMS (LDI): m/z calcd for C81H103N5O5+H+: 1226.8032 [M+H]+;
found: 1226.8038.
H2PPyCO2H. A combined solution of H2PPyCO2Me (1.20 g, 0.909
mmol) in THF (190 mL) and KOH (5.09 g) in H2O (3 mL) was
refluxed in a flask capped by a balloon and covered with
aluminum foil for 14 h. THF was evaporated and the remaining
aqueous layer was neutralized with 1 M HCl. The aqueous layer
was extracted with CHCl3, which was dried over Na2SO4, filtered,
and evaporated. The crude material was chromatographed (SiO2,
CHCl3/MeOH 95:5) to afford a purple solid (1.17 g) including some
impurities, which was used for the subsequent reaction. Rf = 0.2
(SiO2, CHCl3/MeOH 9:1); 1H NMR (CDCl3): d = 9.57 (br, 2H), 8.86
(br, 7H), 8.62 (br, 2H), 8.06 (br, 6H), 7.22 (br, 6H), 4.23 (br, 6H),
1.93 (br, 6H), 1.58 (br, 6H), 1.2–1.5 (48H), 0.89 (m, 9H), –2.75
ppm (2H), CO2H not identified; HRMS (LDI): m/z calcd for
C80H101N5O5+H+: 1212.7842 [M+H]+; found: 1212.7839.
dodecyloxyphenyl)porphyrin (MH+ (C92H127N4O4), 1351.9852).
Vapor pressure osmometry measurement was conducted for
samples dissolved in CHCl3 stabilized with amylene at 45 ºC using
a Gonotec Osmomat 070 osmometer. The calibration curve was
made using benzil as the standard. DOSY experiments were
carried out on
a
JEOL JNM-ECA500 or JNM-ECX400
spectrometer using a bipolar pulse pair stimulated echo pulse
sequence. Visible absorption and fluorescence spectra were
measured on Jasco V-770 and FP-8600 spectrometers,
respectively. Fluorescence lifetimes were measured on
a
Hamamatsu Quantaurus-Tau C11367. Cyclic voltammetry was
conducted on an ALS/CH Instruments electrochemical analyzer
608DN. The working, counter, and pseudo reference electrodes
were glassy carbon, Pt coil, and Ag wire, respectively. The sample
(1 mM) was dissolved in 0.1
M
tetrabutylammonium
hexafluorophosphate solution in 1,2-dichloroethane. The scan
rate was 100 mV s–1. The potentials were calibrated against
internal ferrocene added after each measurement. Transient
absorption measurements were conducted using a dual-beam
femtosecond time-resolved transient absorption spectrometer
consisted of two independently-tunable home-made noncollinear
optical parametric amplifiers pumped by
a regeneratively
amplified Ti:sapphire laser system (Spectra-Physics, Hurricane-
X) operating at 3 kHz repetition rate and an optical detection
system. For more details for transient absorption, see Supporting
Information.
H2PPy. A solution of p-dodecyloxybenzaldehyde[63] (3.61 g, 12.4
mmol), 3-pyridinecarboxaldehyde (442 mg, 4.13 mmol), and
pyrrole (1.11 g, 16.5 mmol) in propionic acid (90 mL) were
refluxed in an open flask covered with aluminum foil for 2.5 h. The
solvent was evaporated and the residue was dissolved in CHCl3,
which was washed with aqueous NaHCO3 twice and aqueous
NaCl once, dried over Na2SO4, filtered, and evaporated. The
crude material was chromatographed twice (SiO2, CHCl3) and
further purified with GPC to afford a purple solid (159 mg, 0.136
1
mmol, 3,3%). Rf = 0.4 (SiO2, CHCl3); H NMR (CDCl3): δ = 9.45
(s, 1H), 9.03 (d, J = 3.7 Hz, 1H), 8.92 (d, J = 4.3 Hz, 2H), 8.89 (s,
4H), 8.76 (d, J = 4.9 Hz, 2H), 8.51 (dt, J = 7.3 Hz, 4.3 Hz, 1H),
8.11 (d, J = 7.9 Hz, 6H), 7.74 (dd, J = 7.3, 5.5 Hz, 1H), 7.28 (d, J
= 8.5 Hz, 6H), 4.25 (t, J = 6.7 Hz, 6H), 1.98 (quintet, J = 7.0 Hz,
6H), 1.63 (quintet, J = 7.0 Hz, 6H), 1.2–1.5 (48H), 0.90 (t, J = 6.7
Hz, 9H), –2.76 (s, 2H) ppm; HRMS (LDI): m/z calcd for
C79H101N5O3+H+: 1168.7977 [M+H]+; found: 1168.7927.
ZnPPy. A combined solution of H2PPy (63 mg, 0.054 mmol) in
CHCl3 (4.7 mL) and Zn(OAc)2•(H2O)2 (59 mg, 0.27 mmol) in
MeOH (1.2 mL) was stirred at room temperature in the dark for 2
h. The solvents were evaporated and the residue was dissolved
in CHCl3, which was washed with H2O three times, dried over
Na2SO4, filtered, and evaporated. The crude product was
crystallized from CHCl3/MeCN to afford a purple solid (49 mg,
0.040 mmol, 74%). Rf = 0.55 (SiO2, CHCl3/pyridine 95:5); 1H NMR
(CDCl3): a broad spectrum, see text. HRMS (LDI): m/z calcd for
C79H99N5O3Zn+ [M]+: 1229.7034; found: 1229.7080.
H2PPyC. A solution of H2PPyCO2H (1.17 g, 0.967 mmol) and
carbonyldiimidazole (CDI, 1.57 g, 9.67 mmol) in dry THF (100 mL)
was refluxed under N2 for 1 h. After the solution was cooled to
room temperature, aqueous NH3 (28–30%, 14 mL) was added
and the solution was further stirred under N2 in the dark for 21 h.
The solution was concentrated to about a half volume to which
CHCl3 (200 mL) was added. The solution was washed with
aqueous NaHCO3 three times, dried over Na2SO4, filtered, and
evaporated. The crude material was chromatographed twice
H2PPyA. A solution of m-octyloxybenzaldehyde[64] (316 mg, 1.35
mmol), 5-acetamido-3-pyrydinecarboxaldehyde[65] (76 mg, 0.46
mmol), and pyrrole (123 mg, 1.84 mmol) in propionic acid (8 mL)
was refluxed in an open flask covered with aluminum foil for 2.5
CHCl3, which was washed with aqueous NaHCO3 three times,
dried over Na2SO4, filtered, and evaporated. The crude material
(SiO2, CHCl3
→ CHCl3/MeOH 96:4 and SiO2, CHCl3 →
CHCl3/MeOH 98:2) and further purified by GPC to afford a purple
9
This article is protected by copyright. All rights reserved.

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
solid (829 mg, 0.685 mmol, 71%). Rf = 0.6 (SiO2, CHCl3/MeOH
9:1); H NMR (CDCl3): d = 9.59 (d, J = 1.8 Hz, 1H), 9.40 (d, J =
4.18 (s, 5H), 1.2–2.0 (40H),0.90 (t, J = 7.3 Hz, 6H), –2.30 ppm (br,
2H); HRMS (LDI): m/z calc for C72H81FeN5O4+H+: 1136.5711
[M+H]+; found: 1136.5630.
1
2.4 Hz, 1H), 8.85–8.95 (7H), 8.69 (d, J = 4.9 Hz, 2H), 8.05–8.15
(6H), 7.2–7.3 (6H), 6.26 (br, 1H), 5.79 (br, 1H), 4.2–4.3 (6H), 1.9–
2.0 (6H), 1.55–1.65 (6H), 1.2–1.5 (48H), 0.90 (t, J = 6.7 Hz, 9H),
–2.76 ppm (s, 2H); HRMS (LDI): m/z calcd for C80H102N6O4+H+:
1211.8035 [M+H]+; found: 1211.7998.
H2PFcPyC. A solution of H2PFcPyCO2H (243 mg, 0.214 mmol)
and carbonyldiimidazole (CDI, 347 mg, 2.14 mmol) in dry THF (22
mL) was refluxed under N2 for 1 h. After the solution was cooled
to room temperature, aqueous NH3 (28–30%, 14 mL) was added
and the solution was further stirred under N2 in the dark for 17 h.
The solution was diluted with CHCl3 which was washed with
aqueous NaHCO3 three times, dried over Na2SO4, filtered, and
evaporated to afford a purple solid, which was essentially pure
(237 mg, 0.209 mmol, 98%). Rf = 0.6 (SiO2, CHCl3/MeOH 9:1); 1H
NMR (CDCl3): d = 9.99 (d, J = 4.3 Hz, 2H), 9.57 (d, J = 2.4 Hz,
1H), 9,43 (d, J = 2.4 Hz, 1H), 8.90 (t, J = 2.4 Hz, 1H), 8.84 (d, J =
4.9 Hz, 2H), 8.82 (d, J = 5.5 Hz, 2H), 8.63 (d, J = 4.9 Hz, 2H), 8.10
(d, J = 8.5 Hz, 4H), 7.29 (d, J = 8.5 Hz, 4H), 6.32 (br, 1H), 5.79
(br, 1H), 5.57 (m, 2H), 4.86 (m, 2H), 4.26 (t, J = 6.7 Hz, 4H), 4.19
(s, 5H), 1.99 (quintet, J = 6.7 Hz, 4H), 1.63 (quintet, J = 7.9 Hz,
4H), 1.25–1.50 (32H), 0.90 (t, J = 6.7 Hz, 6H), –2.30 ppm (s, 2H).
HRMS (LDI): m/z cald for C72H82FeN6O3+H+: 1135.5871 [M+H+];
found: 1135.5710.
ZnPPyC. A combined solution of H2PPyC (769 mg, 0.636 mmol)
in CHCl3 (60 mL) and Zn(OAc)2•(H2O)2 (696 mg, 3.18 mmol) in
MeOH (15 mL) was stirred at room temperature in the dark for 3
h. The solvents were evaporated and the residue was dissolved
in CHCl3, which was washed with H2O three times, dried over
Na2SO4, filtered, and evaporated. The crude product was
chromatographed twice (SiO2, CHCl3/MeOH 98:2) to afford a
purple/green solid (716 mg, 0.562 mmol, 88%). Rf = 0.45 (SiO2,
CHCl3/MeOH 95:5); 1H NMR (CDCl3): d = 8.99 (d, J = 4.9 Hz, 1H),
8.96 (d, J = 4.9 Hz, 1H), 8.70 (d, J = 4.9 Hz, 1H), 8.55–8.65 (2H),
8.34 (d, J = 4.3 Hz, 1H), 8.18 (d, J = 7.9 Hz, 1H), 8.0–8.1 (2H),
7.98 (d, J = 6.7 Hz, 1H), 7.72 (d, J = 7.9 Hz, 1H), 7.51 (d, J = 4.9
Hz, 1H), 7.15–7.25 (4H), 6.93 (d, J = 6.7 Hz, 1H), 6.68 (s, 1H),
6.28 (br, 1H), 6.00 (d, J = 7.9 Hz, 1H), 5.37 (d, J = 7.3, 1H), 5.13
(d, J = 3.7 Hz, 1H), 4.15–4.25 (4H), 3.93 (br, 1H), 3.4–3.6 (m, 2H),
3.37 (s, 1H), 2.11 (br, 1H), 1.9–2.0 (m, 4H), 1.15–1.7 (56H), 0.8–
0.95 ppm (9H); HRMS (LDI): m/z calcd for C80H100N6O4Zn+:
1272.7092 [M]+; found: 1272.7175.
ZnPFcPyC. A combined solution of H2PFcPyC (237 mg, 0.209
mmol) in CHCl3 (18 mL) and Zn(OAc)•(H2O)2 (230 mg, 1.05 mmol)
in MeOH (3.4 mL) was stirred at room temperature in the dark for
2 h. The solution was diluted with CHCl3, which was washed with
H2O three times, dried over Na2SO4, filtered, and evaporated. The
H2PFcPyCO2Me. A mixture of meso-(4-dodecyloxytphen-1-yl)-
2,2'-dipyrromethane[68] (6.74 g, 16.6 mmol), methyl 5-formyl-3-
crude product was chromatographed twice (SiO2, CHCl3
→
pyridinecarboxylate[66-67]
(1.37
g,
8.30
mmol),
CHCl3/MeOH 98:2) to afford a purple solid (237 mg, 0.198 mmol,
95%). Rf = 0.6 (SiO2, CHCl3/MeOH 9:1); 1H NMR (CDCl3): peaks
are resolved but broad, see the spectrum in Supporting
Information; HRMS (LDI): m/z calcd for C72H80FeN6O3Zn+:
1196.4927 [M]+; found: 1196.4866.
ferrocenecarboxaldehyde (1.78 g, 8.30 mmol), and NH4Cl (8.88 g,
166 mmol) in CH2Cl2 (1.65 L) in a flask capped with a septum was
deoxygenated by bubbling N2 for 20 min. TFA (2.5 mL, 33 mmol)
was added through the septum to the mixture, which was stirred
at room temperature in the dark for 4 h. DDQ (5.65 g, 24.9 mmol)
was then added and the mixture was further stirred at room
temperature for 14 h. Finally, Et3N (4.62 mL, 33.2 mmol) was
added and the solution was concentrated to ~500 mL, which was
passed through a pad of SiO2. The fractions containing the
desired compound were chromatographed twice (SiO2, CHCl3 →
CHCl3/MeOH 99:1 and CHCl3). The collected fractions still
contained H2PPyCO2Me, which was removed by GPC to afford a
purple solid (285 mg, 0.248 mmol, 3.0%). Rf = 0.7 (SiO2,
Acknowledgements
We thank K. Wada and M. Kudo of Center for Creative Materials
Research, Research Institute of College of Science and
Technology, Nihon University for low-temperature NMR and
precise-weight sampling, respectively. We also thank Y. Kida for
experimental help. J.O. acknowledges financial support from
Kakenhi (18K05156), MEXT, Japan for part of the work. The work
at Yonsei University (transient absorption) was supported by the
National Research Foundation of Korea (NRF) grant funded by
the Korea government (MSIT) (No. 2020R1A5A1019141).
1
CHCl3/MeOH 99:1); H NMR (CDCl3): d = 10,00 (d, J = 4.3 Hz,
2H), 9.63 (d, J = 1,8 Hz, 1H), 9.57 (d, J = 2.4 Hz, 1H), 9.09 (t, J =
2.4 Hz, 1H), 8.84 (d, J = 4.9 Hz, 2H), 8.82 (d, J = 4.9 Hz, 2H), 8.62
(d, J = 4.9 Hz, 2H), 8.10 (d, J = 8.5 Hz, 4H), 7.29 (d, J = 8.5 Hz,
4H), 5.57 (m, 2H), 4,.86 (m, 2H), 4.26 (t, J = 7.3 Hz, 4H), 4.19 (s,
5H), 4.06 (s, 3H), 1.99 (quintet, J = 7.3 Hz, 4H), 1.64 (quintet, J =
7.3 Hz, 4H), 1.25–1.55 (32H), 0.90 (t, J = 0.89 Hz, 6H), –2.30 ppm
(s, 2H); HRMS (LDI): m/z calc for C73H83FeN5O4+H+: 1150.5867
[M+H]+; found: 1150.5932.
H2PFcPyCO2H. A combined solution of H2PFcPyCO2Me (277 mg,
0.241 mmol) in THF (33 mL) and KOH (1.41 g, 25.2 mmol) in H2O
(0.82 mL) was refluxed in a flask capped by a balloon and covered
with aluminum foil for 15 h. THF was evaporated and the
remaining aqueous layer was neutralized with 1 M HCl and
extracted with CHCl3, which was dried over Na2SO4, filtered, and
evaporated. The crude material was chromatographed (SiO2,
CHCl3 → CHCl3/MeOH 9:1) to afford a purple solid (253 mg, 0.223
mmol, 93%). Rf = 0.25 (SiO2, CHCl3/MeOH 9:1); 1H NMR (CDCl3):
d = 9.98 (br, 2H), 9.54 (br, 2H), 8.81 (br, 5H), 8.58 (br, 2H), 8.08
(br, 4H), 7.2 (br, 4H), 5.58 (br, 2H), 4.86 (br, 2H), 4.24 (br, 4H),
Keywords: axial coordination • light-harvesting antenna •
porphyrinoids • ring current • self-assembly
[1] Handbook of Porphyrin Science, Vol. 1 (Eds.: K. M. Kadish, K.
M. Smith, R. Guilard), World Scientific, 2010.
[2] N. Aratani, D. Kim, A. Osuka, Acc. Chem. Res. 2009, 42, 1922–
1934.
[3] S. Fukuzumi, K. Ohkubo, Dalton Trans. 2013, 42, 15846–15858.
[4] J. Otsuki, J. Mater. Chem. A 2018, 6, 6710–6753.
[5] J. A. Wytko, R. Ruppert, C. Jeandon, J. Weiss, Chem. Commun.
2018, 54, 1550–1558.
[6] J. Min Park, J. H. Lee, W.-D. Jang, Coord. Chem. Rev. 2020,
407, 213157.
[7] S. Kundu, A. Patra, Chem. Rev. 2017, 117, 712–757.
[8] H. Song, Q. Liu, Y. Xie, Chem. Commun. 2018, 54, 1811–1824.
[9] A. Mahmood, J.-Y. Hu, B. Xiao, A. Tang, X. Wang, E. Zhou, J.
Mater. Chem. A 2018, 6, 16769–16797.
10
This article is protected by copyright. All rights reserved.

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
[10] K. Zeng, Z. Tong, L. Ma, W.-H. Zhu, W. Wu, Y. Xie, Energy
Environ. Sci. 2020, 13, 1617–1657.
[43] C. S. J. Jr., Prog. Nucl. Magn. Reson. Spectrosc. 1999, 34,
203–256.
[11] S. Shao, V. Rajendiran, J. F. Lovell, Coord. Chem. Rev. 2019,
379, 99–120.
[44] Y. Cohen, L. Avram, L. Frish, Angew. Chem. Int. Ed. 2005, 44,
520–554.
[12] N. Takeshima, K. Sugawa, H. Tahara, S. Jin, H. Wakui, M.
Fukushima, K. Tokuda, S. Igari, K. Kanakubo, Y. Hayakawa, R.
Katoh, K. Takase, J. Otsuki, ACS Nano 2019, 13, 13244–13256.
[13] K. Sugawa, K. Uchida, N. Takeshima, S. Jin, N. Tsunenari, H.
Takeda, Y. Kida, T. Akiyama, J. Otsuki, K. Takase, S. Yamada,
Nanoscale 2016, 8, 15467–15472.
[14] K. Sugawa, N. Takeshima, K. Uchida, H. Tahara, S. Jin, N.
Tsunenari, T. Akiyama, Y. Kusaka, N. Fukuda, H. Ushijima, Y.
Tsuchido, T. Hashimoto, T. Hayashita, J. Otsuki, J. Mater. Chem. C
2015, 3, 11439–11448.
[15] Multiporphyrin Arrays, Fundamentals and Applications, (Eds.:
D. Kim), Pan Stanford, 2012.
[16] J. A. A. W. Elemans, R. v. Hameren, R. J. M. Nolte, A. E.
Rowan, Adv. Mater. 2006, 18, 1251–1266.
[17] E. Iengo, E. Zangrando, E. Alessio, Eur. J. Inorg. Chem. 2003,
2003, 2371–2384.
[18] M. M. Yatskou, R. B. M. Koehorst, H. Donker, T. J. Schaafsma,
J. Phys. Chem. A 2001, 105, 11425–11431.
[45] R. Evans, Z. Deng, A. K. Rogerson, A. S. McLachlan, J. J.
Richards, M. Nilsson, G. A. Morris, Angew. Chem. Int. Ed. 2013, 52,
3199–3202.
[46] E. B. Fleischer, C. K. Miller, L. E. Webb, J. Am. Chem. Soc.
1964, 86, 2342–2347.
[47] R. J. Abraham, G. R. Bedford, D. McNeillie, B. Wright, Org.
Magn. Reson. 1980, 14, 418–425.
[48] P. S. Sobolev, V. P. Andreev, T. A. Ekimova, Russ. J. Gen.
Chem. 2014, 84, 1399–1403.
[49] V. P. Andreev, P. S. Sobolev, N. S. Lebedeva, V. A. Tafeenko,
Russ. J. Gen. Chem. 2015, 85, 906–910.
[50] V. P. Andreev, P. S. Sobolev, D. O. Zaitsev, S. M. Timofeeva,
Russ. J. Gen. Chem. 2018, 88, 2108–2113.
[51] J. Rochford, A. D. Rooney, M. T. Pryce, Inorg. Chem. 2007, 46,
7247–7249.
[52] M. Kubo, Y. Mori, M. Otani, M. Murakami, Y. Ishibashi, M.
Yasuda, K. Hosomizu, H. Miyasaka, H. Imahori, S. Nakashima, J.
Phys. Chem. A 2007, 111, 5136–5143.
[19] M. M. Yatskou, R. B. M. Koehorst, A. van Hoek, H. Donker, T. J.
Schaafsma, B. Gobets, I. van Stokkum, R. van Grondelle, J. Phys.
Chem. A 2001, 105, 11432–11440.
[53] M. K. Chahal, H. B. Gobeze, W. A. Webre, P. A. Karr, D. T.
Payne, K. Ariga, F. D’Souza, J. P. Hill, Phys. Chem. Chem. Phys.
2020, 22, 14356–14363.
[20] G. Ercolani, M. Ioele, D. Monti, New J. Chem. 2001, 25, 783–
789.
[54] Y. Kuramochi, Y. Fujisawa, A. Satake, J. Am. Chem. Soc. 2020,
142, 705–709.
[21] M. M. Yatskou, H. Donker, E. G. Novikov, R. B. M. Koehorst, A.
van Hoek, V. V. Apanasovich, T. J. Schaafsma, J. Phys. Chem. A
2001, 105, 9498–9508.
[22] K. Funatsu, T. Imamura, A. Ichimura, Y. Sasaki, Inorg. Chem.
1998, 37, 1798–1804.
[23] R. A. Jensen, R. F. Kelley, S. Joong Lee, M. R. Wasielewski, J.
T. Hupp, D. M. Tiede, Chem. Commun. 2008, 1886–1888.
[24] A. M. Shachter, E. B. Fleischer, R. C. Haltiwanger, J. Chem.
Soc. Chem. Commun. 1988, 960–961.
[25] R. F. Kelley, R. H. Goldsmith, M. R. Wasielewski, J. Am. Chem.
Soc. 2007, 129, 6384–6385.
[55] J. Fujimoto, S. Hayashi, H. Kainuma, K. Manseki, T. Udagawa,
H. Miyaji, Chem. Asian J. 2019, 14, 2567–2572.
[56] R. J. Abraham, S. C. M. Fell, K. M. Smith, Org. Magn. Reson.
1977, 9, 367–373.
[57] R. J. Abraham, C. J. Medforth, Magn. Reson. Chem. 1988, 26,
803–812.
[58] Y. Shinozaki, Y. Mizumura, Y. Kida, K. Sugawa, K. Wada, S.
Kishiro, J. Otsuki, Chem. Lett. 2016, 45, 1123–1125.
[59] J. J. P. Stewart, J. Mol. Model. 2007, 13, 1173–1213.
[60] C. Stangel, F. Plass, A. Charisiadis, E. Giannoudis, G.
Chararalambidis, K. Karikis, G. Rotas, G. E. Zervaki, N. N.
Lathiotakis, N. Tagmatarchis, A. Kahnt, A. G. Coutsolelos, Phys.
Chem. Chem. Phys. 2018, 20, 21269–21279.
[26] Y. Shinozaki, K. Ohkubo, S. Fukuzumi, K. Sugawa, J. Otsuki,
Chem. Eur. J. 2016, 22, 1165–1176.
[27] Y. Shinozaki, T. Tsubomura, K. Sugawa, J. Otsuki, Tetrahedron
Lett. 2016, 57, 48–52.
[61] P. Klán, J. Wirz, Photochemistry of Organic Compounds, (Eds.:
Wiley, 2009.
[28] Y. Shinozaki, G. Richards, K. Ogawa, A. Yamano, K. Ohara, K.
Yamaguchi, S.-i. Kawano, K. Tanaka, Y. Araki, T. Wada, J. Otsuki,
J. Am. Chem. Soc. 2013, 135, 5262–5265.
[29] A. Tsuda, S. Sakamoto, K. Yamaguchi, T. Aida, J. Am. Chem.
Soc. 2003, 125, 15722–15723.
[62] G. L. Closs, L. E. Closs, J. Am. Chem. Soc. 1963, 85, 818–819.
[63] W. Delong, W. Lanying, W. Yongling, S. Shuang, F. Juntao, Z.
Xing, Eur. J. Med. Chem. 2017, 130, 286–307.
[64] Q. Liu, C. Li, E. Jin, Z. Lu, Y. Chen, F. Li, Z. Bo, ACS Appl.
Mater. Interf. 2014, 6, 1601-1607.
[30] V. L. Gunderson, M. R. Wasielewski, Chem. Phys. Lett. 2013,
556, 303–307.
[31] R. Takahashi, Y. Kobuke, J. Am. Chem. Soc. 2003, 125, 2372–
2373.
[32] M. Z. Papiz, S. M. Prince, T. Howard, R. J. Cogdell, N. W.
Isaacs, J. Mol. Biol. 2003, 326, 1523–1538.
[33] A. W. Roszak, T. D. Howard, J. Southall, A. T. Gardiner, C. J.
Law, N. W. Isaacs, R. J. Cogdell, Science 2003, 302, 1969–1972.
[34] S. Niwa, L.-J. Yu, K. Takeda, Y. Hirano, T. Kawakami, Z.-Y.
Wang-Otomo, K. Miki, Nature 2014, 508, 228–232.
[35] A. D. Adler, F. R. Longo, J. D. Finarelli, J. Goldmacher, J.
Assour, L. Korsakoff, J. Org. Chem. 1967, 32, 476.
[36] S. V. Bhosale, S. Hackbarth, S. J. Langford, S. V. Bhosale,
Chem. Asian J. 2012, 7, 176–182.
[65] Y. Kubo, M. Ando, H. Tanaka, S. Osaka, T. Matsumoto, H.
Nakata, D. Terada, T. Nitabaru, (Fujifilm), WO2015030057A1, 2015.
[66] H. Takezawa, M. Hayashi, Y. Iwasawa, M. Hosoi, Y. Iida, Y.
Tsuchiya, M. Horie, T. Kamei, (Banyu Pharmaceutical),
WO9005132A1, 1990.
[67] C. Kremoser, U. Abel, C. Steeneck, O. Kinzel, (Phenex
Pharmaceuticals), WO2011020615A1, 2010.
[68] R. Takahashi, Y. Kobuke, J. Org. Chem. 2005, 70, 2745–2753.
[37] D. Sirbu, C. Turta, A. C. Benniston, F. Abou-Chahine, H.
Lemmetyinen, N. V. Tkachenko, C. Wood, E. Gibson, RSC Adv.
2014, 4, 22733–22742.
[38] D.-F. Shi, R. T. Wheelhouse, D. Sun, L. H. Hurley, J. Med.
Chem. 2001, 44, 4509–4523.
[39] R. S. Loewe, A. Ambroise, K. Muthukumaran, K. Padmaja, A. B.
Lysenko, G. Mathur, Q. Li, D. F. Bocian, V. Misra, J. S. Lindsey, J.
Org. Chem. 2004, 69, 1453–1460.
[40] Y. Zhou, K. T. Ngo, B. Zhang, Y. Feng, J. Rochford,
Organometallics 2014, 33, 7078–7090.
[41] B. J. Littler, Y. Ciringh, J. S. Lindsey, J. Org. Chem. 1999, 64,
2864–2872.
[42] V. V. Krishnan, J. Magn. Reson. 1997, 124, 468–473.
11
This article is protected by copyright. All rights reserved.

10.1002/chem.202003327
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
ꢁꢂꢍꢎꢇꢅ
ꢀꢁꢂꢃꢄꢅ
ꢆꢃꢇꢁꢈꢉꢂꢃ
ꢁꢂꢍꢎꢇꢅ
ꢀꢊꢂꢋꢆꢃꢌꢁ
ꢆꢃꢇꢁꢈꢉꢂꢃ
Zinc meso-(meta-pyridyl)porphyrin forms a trimer in solution over wide temperature and concentration ranges. A supramolecular
light-harvesting/charge separation system, a rare example in which energy gradient is incorporated and reductive quenching is
involved, was constructed using this motif of self-assembly.
12
This article is protected by copyright. All rights reserved.