Syntheses and CO2 reduction activities of π-expanded/extended iron porphyrin complexes

Article abstract of DOI:10.1007/s00775-017-1438-3

The construction of molecular catalysts that are active toward CO₂ reduction is of great significance for designing sustainable energy conversion systems. In this study, we aimed to develop catalysts for CO₂ reduction by introducing

Full text of DOI:10.1007/s00775-017-1438-3

J Biol Inorg Chem
DOI 10.1007/s00775-017-1438-3
ORIGINAL PAPER
Syntheses and CO2 reduction activities of π‑expanded/extended
iron porphyrin complexes
Yuki Okabe1,2 · Sze Koon Lee1,2 · Mio Kondo1,2,3,4 · Shigeyuki Masaoka1,2,4
Received: 11 November 2016 / Accepted: 4 January 2017
© SBIC 2017
Abstract The construction of molecular catalysts that
are active toward CO2 reduction is of great signifi-
cance for designing sustainable energy conversion sys-
tems. In this study, we aimed to develop catalysts for
CO2 reduction by introducing aromatic substituents to
the meso-positions of iron porphyrin complexes. Three
novel iron porphyrin complexes with π-expanded sub-
measurements under Ar in comparison with an iron com-
plex with a basic framework, 5,10,15,20-tetrakis(phenyl)
porphyrinato iron(III) chloride (Fe‑Ph). Moreover, the
catalytic activity of the complexes was studied by elec-
trochemical measurements under CO2, and it is found that
the complex with the π-expanded substituents exhibits the
highest activity among these complexes.
stituents
(5,10,15,20-tetrakis(pyren-1-yl)porphyrinato
iron(III) chloride (Fe‑Py)), π-extended substituents
(5,10,15,20-tetrakis((1,1′-biphenyl)-4-yl)porphyrinato
iron(III) chloride (Fe‑PPh)) and π-expanded and extended
substituents (5,10,15,20-tetrakis(4-(pyren-1-yl)phenyl)por-
phyrinato iron(III) chloride (Fe‑PPy)) were successfully
synthesized, and their physical properties were investigated
by UV–vis absorption spectroscopy and electrochemical
Keywords Iron complex · Porphyrin · Pyrene ·
Electrochemistry · CO2 Reduction
Introduction
Growing global energy demands and environmental prob-
lems, such as climate change and air pollution, currently
underpin the broad interest in the development of sustain-
able energy conversion systems. In natural photosynthetic
reactions, renewable but not storable solar energy is suc-
cessfully converted into storable chemical energy by reduc-
ing carbon dioxide. This reductive transformation of carbon
dioxide, which results in the production of energy-rich car-
bon compounds, is a very attractive reaction because this
technology will allow us to solve both energy and envi-
ronmental problems [1, 2]. Based on such considerations,
there have been numerous studies that aimed to develop
molecular-based artificial catalysts for the reduction of car-
bon dioxide in recent years [3–20].
Iron porphyrin complexes are fascinating candidates as
catalysts for carbon dioxide reduction [21–35]. One of the
advantageous points of the complexes is that we can eas-
ily tune their physical properties by introducing various
substituents at the meso-positions of the porphyrin frame-
work. In fact, several reports revealed that the introduction
of appropriate functional groups, such as acidic/basic and
Electronic supplementary material The online version of this
article (doi:10.1007/s00775-017-1438-3) contains supplementary
material, which is available to authorized users.
* Mio Kondo
mio@ims.ac.jp
* Shigeyuki Masaoka
masaoka@ims.ac.jp
1
Institute for Molecular Science (IMS), 5-1 Higashiyama,
Myodaiji, Okazaki, Aichi 444-8787, Japan
2
Department of Structural Molecular Science, School
of Physical Sciences, SOKENDAI (The Graduate University
for Advanced Studies), Shonan Village, Hayama-cho,
Kanagawa 240-0193, Japan
3
Japan Science and Technology Agency (JST), ACT-C, 4-1-8
Honcho, Kawaguchi, Saitama 332-0012, Japan
4
Research Center of Integrative Molecular Systems (CIMoS),
Institute for Molecular Science, Nishigo-naka 38, Myodaiji,
Okazaki, Aichi 444-8585, Japan
1 3

J Biol Inorg Chem
electron donating/withdrawing substituents, could affect
the catalytic activity [29–31]. In this context, the introduc-
tion of substituents with a large π plane is expected to be
one of the ways to control the catalytic activity of the iron
porphyrin catalysts because this class of substituents not
only could influence the electronic structure but could also
provide hydrophobicity, which is preferable for accumulat-
ing CO2 molecules near the active center, to the molecular
catalysts. However, the effect of such substituents on the
catalytic activity of iron porphyrin complexes has not been
explored.
In this study, we aimed to investigate the properties of iron
porphyrin complexes with various types of π-conjugated
substituents. We employed 5,10,15,20-tetrakis(phenyl)
porphyrin (H‑Ph, Fig. 1), which is known to catalyze elec-
trochemical CO2 reduction [32–35], as a basic framework
and derivatized its structure with π-expanded substituents,
π-extended substituents, and π-expanded and extended sub-
stituents to afford 5,10,15,20-tetrakis(pyren-1-yl)porphyrin
(H‑Py),
5,10,15,20-tetrakis((1,1′-biphenyl)-4-yl)porphy-
Fig. 1 Chemical structures of the porphyrin derivatives investigated
rin (H‑PPh) and 5,10,15,20-tetrakis(4-(pyren-1-yl)phenyl)
porphyrin (H‑PPy), respectively. Described herein are the
syntheses, UV–vis absorption properties, and electrochemi-
cal properties of the free-base porphyrins and iron porphy-
rin complexes. In addition, the catalytic activities of the iron
porphyrin complexes for electrochemical CO2 reduction are
also presented.
in this study
oxidation, as an alternative synthetic route. Aryl aldehydes
(1-pyrenecarbaldehyde or 4-phenylbenzaldehyde) were
reacted with pyrrole (1.0 eq.) in dichloromethane (DCM)
at room temperature in the presence of trifluoroacetic acid
(TFA),
and
2,3-dichloro-5,6-dicyano-p-benzoquinone
(DDQ) was added to the resultant reaction mixture as an
oxidant. Under these reaction conditions, the amounts of
impurities, such as polymeric pyrrole, were lower compared
to the previous reaction conditions. Thus, the desired com-
pounds were successfully isolated after their purification by
silica gel column chromatography. The isolated yields were
6 and 7% for H‑Py and H‑PPh, respectively. Note that the
synthesis of H‑PPh by a similar method has been reported
previously [41]. Since the acid-catalyzed condensation reac-
tion using Lindsey’s method was effective, the synthesis
of the π-expanded and extended derivative (H‑PPy) was
performed using the same method. An aryl aldehyde with
a pyrenylphenyl moiety, 4-(pyren-1-yl)benzaldehyde, was
prepared by a Pd-catalyzed Suzuki–Miyaura cross-cou-
pling reaction between 1-pyrenecarbaldehyde and 4-for-
mylphenylboronic acid, and 4-(pyren-1-yl)benzaldehyde
was obtained in 77% yield. The obtained aryl aldehyde was
then condensed with pyrrole, and the porphyrinogen inter-
mediate was further oxidized by 1.2 eq. of DDQ at room
temperature for 90 min. The resulting reaction mixture was
quenched by triethylamine (TEA) and then purified by silica
gel chromatography to afford H‑PPy with a 16% yield. The
obtained free-base porphyrins, H‑Ph, H‑Py, H‑PPh, and
Results and discussion
Syntheses
Syntheses of the free‑base porphyrins
Four types of free-base porphyrins were prepared following
the synthetic routes shown in Scheme 1. The porphyrin with
a basic structure, H‑Ph, was obtained in moderate yield
(19%) using Adler’s method [36], in which benzaldehyde
and pyrrole were refluxed in propionic acid under air. The
π-expanded (H‑Py) and the π-extended (H‑PPh) deriva-
tives were also previously synthesized using Adler’s method
[37, 38]. Therefore, we initially attempted the syntheses of
these compounds under the same conditions as the synthe-
sis of H‑Ph, and the formation of the macrocyclic porphyrin
skeleton was confirmed by 1H NMR spectroscopy. However,
the isolation of the desired products (H‑Py and H‑PPh)
from the reaction mixture using column chromatography
was unsuccessful due to the generation of numerous types
of side products. Therefore, we adopted Lindsey’s method
[39, 40], in which porphyrin derivatives were obtained by
the condensation of aryl aldehydes with pyrrole and the sub-
sequent aromatization of the porphyrinogen intermediate by
1
H‑PPy, were characterized by H NMR spectroscopy and
elemental analyses.
1 3

J Biol Inorg Chem
Scheme 1 Synthetic routes for
the free-base porphyrins, H‑Ph,
H‑Py, H‑PPh, and H‑PPy
1 3

J Biol Inorg Chem
In the 1H NMR spectra of the free-base porphyrins
in CDCl3, the broad singlet signal attributed to the inter-
nal N–H moieties was observed below 0 ppm (Fig. 2)
because of the ring current associated with the aro-
matic porphyrin. The chemical shifts of the signals were
−2.80, −1.98, −2.69 and −2.51 ppm for H‑Ph, H‑Py,
H‑PPh, and H‑PPy, respectively. Note that these chemi-
cal shifts were largely different depending on the substitu-
ent at the meso-positions. The order of chemical shifts
was H‑Py > H‑PPy > H‑PPh > H‑Ph, which can be
explained by considering the ring current from the aromatic
substituents.
In addition to the singlet peaks of the internal N–H
proton, the peaks originating from the aromatic substitu-
ents and the protons at the β-position of pyrrole rings
(β-protons) were observed in the aromatic region. The
signal of the β-protons was obtained as a singlet peak
for H‑Ph (8.82 ppm), H‑PPh (8.96 ppm) and H‑PPy
(9.21 ppm), whereas the β-protons of H‑Py were observed
as multiplets at 8.75–8.86 ppm and 8.39–8.42 ppm. Gener-
ally, the symmetrical tetra-substituted porphyrin structure
affords a singlet peak for the β-protons, and the asymmet-
ric porphyrins give more complicated peaks. Therefore, the
multiplet signals observed for H‑Py indicate the presence
of four stereoisomers formed due to the asymmetric struc-
ture and the large steric hindrance of the pyrenyl moiety.
For H‑PPy, a singlet peak for the β-protons was observed
despite the asymmetric structure of the pyrenylphenyl moi-
ety. This is because the phenylene linker allows the free
rotation of the pyrenyl moiety, which results in the aver-
aged symmetric structure in solution.
Fig. 2 1H NMR spectra of H‑Ph, H‑Py, H‑PPh, and H‑PPy
(400 MHz, CDCl3)
before heating the solution. This modified synthetic strat-
egy resulted in the formation of the desired complex,
5,10,15,20-tetrakis(phenyl)porphyrinato iron(III) chloride
(Fe‑Ph), in high yield (93%). The metalation of H‑Py
was also performed using the same reaction conditions,
and the desired product, 5,10,15,20-tetrakis(pyren-1-yl)
porphyrinato iron(III) chloride (Fe‑Py), was obtained
quantitatively. However, the iron complex was not formed
at all when H‑PPh was reacted under the same reac-
tion conditions due to the low solubility of the ligand
in DMF. Therefore, the reaction solvent was changed to
NMP and the iron complex, 5,10,15,20-tetrakis((1,1′-
biphenyl)-4-yl)porphyrinato iron(III) chloride (Fe‑PPh),
was obtained with moderate yield (66%). In the case of
H‑PPy, it took almost 24 h to complete the metalation in
DMF and several decomposed compounds, which formed
by heating the porphyrin for a long time, were obtained.
Based on this observation, we changed the reaction sol-
vent to NMP which possesses a higher boiling point than
DMF, to shorten the reaction time. As a result, the iron
complex, 5,10,15,20-tetrakis(4-(pyren-1-yl)phenyl)por-
phyrinato iron(III) chloride (Fe‑PPy), was successfully
obtained with high yield (87%). In all cases, the progress
of the iron insertion reaction and the consumption of
starting material were monitored by TLC and by checking
the decrease of the number of Q bands in UV–vis absorp-
tion spectra (for the assignments of absorption bands,
see “UV–vis absorption properties of free-base and iron
porphyrins”). All iron porphyrin complexes, Fe‑Ph, Fe‑
Py, Fe‑PPh and Fe‑PPy, were characterized by UV–vis
absorption spectroscopy, electrochemical measurements
Syntheses of iron porphyrin complexes
All free-base porphyrins were converted into the iron
porphyrin complexes by a reaction with iron(II) chloride
tetrahydrate in DMF or N-methylpyrrolidone (NMP). The
synthetic scheme employed to obtain the iron complexes
is shown in Scheme 2. The reaction conditions were opti-
mized using H‑Ph as a ligand. In the procedure previ-
ously reported in the literature [42, 43], the iron salt was
gradually added to the reaction mixture in several portions
within a given time period, and it is a slightly cumbersome
procedure. To simplify the experimental operation, we
first attempted the reaction by adding the solid of the iron
salt in one portion. However, the reaction yield (trace)
became significantly lower than the previous report (97%)
[43]. This might be because the oxidation of the iron salt
to give iron(III) species is promoted upon the heating of
the metal salt in the solid state, which prevents the met-
alation reaction. To avoid such a situation, in the modi-
fied method, a solution of iron(II) chloride was prepared
beforehand and added to a solution of H‑Ph dropwise
1
and elemental analyses. The characterization by H NMR
1 3

J Biol Inorg Chem
Scheme 2 Synthetic routes for the iron porphyrin complexes, Fe‑Ph, Fe‑Py, Fe‑PPh and Fe‑PPy
measurement was difficult due to the paramagnetic nature
of the complexes.
spectroscopic data are summarized in Table 1. All free-base
porphyrins present a strong absorption band at approxi-
mately 422–436 nm and four weaker bands widely located
at approximately 516–654 nm. These spectral features are
caused by porphyrin macrocycle-based π–π* transitions
described by Gouterman’s four orbital model [44, 45]. The
former band in the shorter wavelength region is assigned
to the S0 → S2 transition (Soret band), and the latter bands
UV–vis absorption properties of free‑base and iron
porphyrins
UV–vis absorption spectra of the free-base porphy-
rins and the iron complexes are shown in Fig. 3, and the
1 3

J Biol Inorg Chem
Fig. 3 UV–vis absorption spectra of the free-base porphyrins in 1,2-dichlorobenzene a H‑Ph, b H‑Py, c H‑PPh, d H‑PPy, e Fe‑Ph, f Fe‑Py, g
Fe‑PPh and h Fe‑PPy. (Insets) Enlarged UV-vis absorption spectra at Q-band regions
are attributed to the S0 → S1 transition (Q bands). The Soret
and Q bands of H‑Py, H‑PPh, and H‑PPy are slightly red
shifted compared with those of H‑Ph, but located at a quite
similar position. This result indicates that the electronic
structure of the porphyrin moieties was not affected by the
introduction of the aromatic substituents and is consistent
with the results of the electrochemical measurements (vide
infra). In addition to the typical porphyrin-based absorption
1 3

J Biol Inorg Chem
Table 1 Summary of the
absorption spectra for the
free-base porphyrins and the
iron porphyrin complexes in
1,2-dichlorobenzene
Compound
λmax/nm (ε/104/M/cm)
Substituent’s bands
Soret-band
Q-bands
H‑Ph
373 (1.30)
422 (25.52)
436 (19.56)
427 (44.14)
429 (28.12)
420 (5.20)
431 (5.74)
426 (6.30)
427 (11.66)
516 (1.08), 551 (0.48) 593 (0.30), 649 (0.24)
524 (1.78), 555 (1.26), 604 (0.76), 658 (3.56)
520 (1.80), 557 (1.32), 596 (0.56), 653 (0.66)
520 (1.82), 556 (1.34), 596 (0.72), 654 (0.84)
510 (0.56), 575 (0.10), 691 (0.12)
513 (1.32)
H‑Py
328 (6.78), 340 (7.46)
311 (2.08), 377 (2.20)
349 (10.54)
H‑PPh
H‑PPy
Fe‑Ph
Fe‑Py
Fe‑PPh
Fc‑PPy
378 (2.38)
346 (7.74)
383 (2.78)
513 (0.74), 695 (0.18)
351 (11.22)
512 (1.48), 694 (0.34)
bands, several absorption bands were observed at shorter
wavelength (<384 nm) in the case of H‑Py and H‑PPy, and
assigned to the π–π* transitions of the aromatic substituents
[46–49].
All iron porphyrin complexes show a strong absorption
band at approximately 420–431 nm (Soret band) and weaker
bands located in a wider region between 510 and 695 nm
(Q bands). The decrease in the number of Q bands is due to
the degeneration of the HOMO and the HOMO–1 for each
porphyrin, thus indicating the generation of more symmetric
structures via metalation [50]. There is no significant change
in the absorption maxima of the Soret bands upon metalation,
which was characteristic of iron porphyrin complexes [43].
Electrochemical properties
The cyclic voltammograms (CVs) of the free-base porphy-
rins (H‑Ph, H‑Py, H‑PPh, and H‑PPy) and the iron por-
phyrin complexes (Fe‑Ph, Fe‑Py, Fe‑PPh, and Fe‑PPy)
are shown in Fig. 4 and the redox potentials of these com-
pounds are summarized in Table 2. The CVs were meas-
ured with 0.1 M tetrabutylammonium perchlorate (TBAP)
in 1,2-dichlorobenzene. The concentrations of the free-base
porphyrins and iron porphyrin complexes were 1.0 mM
except for H‑PPh. The CV of H‑PPh was measured using
a saturated solution due to its low solubility in the electro-
lyte solution. All free-base porphyrins displayed two
reversible waves, which are assigned to the two-step one
electron reduction of the porphyrin ring [25]. The half wave
potentials (E1/2 (1)/V vs. ferrocene/ferrocenium (Fc/Fc+))
for the first reduction were −2.01 (H‑Ph), −1.99 (H‑Py),
−1.97 (H‑PPh) and −1.99 V (H‑PPy), and those for the
second reduction (E1/2 (2)) were −2.32 (H‑Ph), −2.38 (H‑
Py), −2.27 (H‑PPh) and −2.28 V (H‑PPy), respectively.
The redox potentials of the free-base porphyrins are quite
similar to each other, indicating almost no electronic inter-
action between the porphyrin framework and the substitu-
ent. In the cases of the iron porphyrin complexes, three
reversible reduction peaks, which are possibly assigned to
be Fe(III)/Fe(II), Fe(II)/Fe(I) and Fe(I)/Fe(0) redox couples
Fig. 4 Cyclic voltammograms of the free-base porphyrins (H‑Ph,
H‑Py, and H‑PPy, 1.0 mM) and the iron porphyrin complexes (Fe‑
Ph, Fe‑Py, Fe‑PPy and Fe‑PPy, 1.0 mM) in a 0.1 M TBAP/1,2-
dichlorobenzene solution under an Ar atmosphere (WE: GC; CE: Pt
wire; RE: Ag+/Ag; scan rate: 20 mV s−1). Note that the measurement
of H‑PPh was performed using a saturated solution due to its low
solubility in the electrolyte solution. Potential sweeps were started
from the open circuit potential −0.93 V (H‑Ph), −0.62 V (H‑Py),
−1.00 V (H‑PPh), −0.79 V (H‑PPy), −0.75 V (Fe‑Ph), −0.59 V
(Fe‑Py), −0.72 V (Fe‑PPh) and −0.62 V (Fe‑PPy). Arrows in the
voltammograms indicate the direction of potential sweep
according to the previous report [33], were observed [51,
52].1 Similar to the results of the CVs of the free-base por-
1
In the iron porphyrin complexes, there are several possible resonant
forms to describe the electronic structure of the reduced species, and
all resonant forms presumably contribute to the electronic structure of
the complexes. In the case of Fe‑Ph, previous reports suggested that
the three electron reduced species should be best formulated as a Fe(0)
complex, based on the results of UV–vis and resonance Raman spec-
troscopy and the studies on electrocatalytic reactions (Ref. [33, 51,
52]). Given that redox potentials of the newly synthesized complexes
(Fe‑Py, Fe‑PPh and Fe‑PPy) are quite similar to the reported complex,
Fe‑Ph, the nature of the redox processes and the electronic structure of
the reduced species are probably similar. Therefore, we have assigned
the redox processes of the newly synthesized complexes to be Fe(III)/
Fe(II), Fe(II)/Fe(I) and Fe(I)/Fe(0) according to the previous reports.
1 3

J Biol Inorg Chem
Table 2 Redox potentials of the prepared free-base porphyrins and
the iron porphyrin complexes (E1/2/V vs. Fc/Fc+) in 1,2-dichloroben-
zene under an Ar atmosphere
we measured the CVs of the iron complexes using a more
hydrophilic solvent, DMF. The results of the measurements
and a summary of the electrochemical data are shown in
Fig. 6 and Table 3, respectively. It should be noted the CV
measurement of Fe‑PPh was unsuccessful due to the low
solubility of the complex in DMF. Fe‑Ph and Fe‑Py exhib-
ited three reversible redox waves (−0.63, −1.51, and
−2.15 V for Fe‑Ph, and −0.63, −1.47, −2.08 V for Fe‑Py)
under an Ar atmosphere, which is in good agreement with
the CVs measured in 1,2-dichlorobenzene. For Fe‑PPy, in
addition to the three reversible redox waves (−0.62, −1.50,
and −2.11 V), one new reversible redox peak appeared at
−2.43 V. This additional peak was assigned to the reduc-
tion of the 1-phenylpyrene substituents by comparing the
CVs of related compounds (Figure S1 in the Electronic
Supplementary Material). Under a CO2 atmosphere, Fe‑Ph
exhibited large irreversible cathodic currents attributed to
the catalytic reduction of CO2 to CO at a potential close to
the third redox couple (−2.1 V vs. Fc/Fc+), and the result is
consistent with the previous report [35]. Similarly, an irre-
versible current in a similar potential region was observed
in the CV of Fe‑Py under CO2. To confirm the origin of the
irreversible current, controlled potential electrolysis of Fe‑
Py and Fe‑Ph was performed. Solutions containing 1.0 mM
of the complexes were electrolyzed at −2.2 V (vs. Fc/Fc+)
for 1 h. After the electrolysis, both the gas and the liquid
phases were analyzed and CO as a major product and small
amount of dihydrogen (H2) as a minor product were
detected. The formation of formic acid was not detected.
The faradaic efficiencies for the conversion of CO2 to CO
were determined to be 87 (Fe‑Py) and 93 (Fe‑Ph) % (for
experimental details, see the ESM). These results clearly
indicate the irreversible current is attributed to the electro-
catalytic CO2 reduction by the complexes. In the case of
Fe‑PPy, an increase of the irreversible current was
observed both at the Fe(I)/Fe(0) redox couple (−2.11 V)
and the reduction of the substituent (−2.43 V). Among the
three complexes, Fe‑Py exhibited the largest catalytic cur-
rent under our experimental conditions, which suggests that
the introduction of π-expanded substituents can improve
the catalytic activity. Additionally, the controlled potential
electrolysis revealed that the amounts of CO evolved dur-
ing the electrolysis were 2978 and 1999 μL for Fe‑Py and
Fe‑Ph, respectively (Table S4). We also calculated the turn-
over frequency (TOF) and turnover number (TON) of the
reaction for each complex based on the result of CPE
experiments. TOF values were 120 (Fe‑Py) and 60 (Fe‑Ph)
Compound
E1/2(1)/V
E1/2(2)/V
E1/2(3)/V
H‑Ph
−2.01
−1.99
−1.97
−1.99
−1.14
−1.18
−1.12
−1.19
−2.32
−2.38
−2.27
−2.28
−1.88
−1.89
−1.86
−1.93
–
H‑Py
–
H‑PPh
H‑PPy
Fe‑Ph
Fe‑Py
Fe‑PPh
Fe‑PPy
–
–
−2.49
−2.52
−2.45
−2.52
phyrins, the redox potentials of the iron complexes are
located at almost identical positions, reflecting the similar
electronic structure of the ligands.
Catalytic activity
Iron porphyrin complexes are known to catalyze the CO2
reduction reaction to generate CO under electrochemical
conditions [29–35]. In the cyclic voltammograms of the
complexes under a CO2 atmosphere, an increase of the irre-
versible current coupled with the third reduction wave was
observed. This irreversible current is called the catalytic
current and indicates the occurrence of the electrocatalytic
reduction of CO2. It is also known that the intensity of the
catalytic current basically is correlated to the rate of the
catalytic reaction [30, 53, 54]. Therefore, the electrochemi-
cal measurements of the newly synthesized complexes
were also conducted under a CO2 atmosphere to investigate
the catalytic activity.
Initially, the measurements were performed using 0.1 M
TBAP/1,2-dichlorobenzene solution. The results of the
measurements are shown in Fig. 5. Under a CO2 atmos-
phere, an increase of the irreversible currents was observed
at a potential close to the third redox wave for all com-
pounds, suggesting the electrocatalytic reduction of CO2
[29–35]. However, the intensities of these catalytic currents
were very small. In the CO2 reduction catalyzed by iron
porphyrin complexes, the existence of proton source is
essential to promote the reaction [30]. It is also reported
that water contained in the solvent is known to serve as a
proton source [28, 35]. However, the concentration of water
is low in 1,2-dichlorobenzene due to its hydrophobicity,
which may result in the small catalytic current.2 Therefore,
s
−1 and TON values were 4.3 × 105 (Fe‑Py) and 2.2 × 105
(Fe‑Ph) s−1 for 1 h (for details of the calculation of TOF
and TON values, see the ESM). These results clearly indi-
cate that the reaction rate of Fe‑Py is approximately 2.0
times higher than that of Fe‑Ph. Based on these observa-
tions, we can conclude that the introduction of π-expanded
2
It should be noted that the concentration of CO2 in solution can
be an important factor to affect the reaction rate. In our experiments,
CO2-saturated solutions were employed in all the electrochemical
measurements under a CO2 atmosphere although the exact concentra-
tion of CO2 in the 1,2-diclorobenzene solutions was unknown.
1 3

J Biol Inorg Chem
Fig. 5 Cyclic voltammograms of the iron porphyrin complexes ▸
(a Fe‑Ph, b Fe‑Py, c Fe‑PPy and d Fe‑PPy, 1.0 mM) in a 0.1 M
TBAP/1,2-dichlorobenzene solution under Ar (solid lines) and CO2
(dotted lines) atmospheres. (WE: GC; CE: Pt wire; RE: Ag+/Ag; scan
rate: 20 mV s−1). The solutions were saturated with Ar or CO2 prior
to the experiments. Potential sweeps were started from the open cir-
cuit potential −0.75 V (Fe‑Ph), −0.59 V (Fe‑Py), −0.72 V (Fe‑PPh)
and −0.62 V (Fe‑PPy). Arrows in the voltammograms indicate the
direction of potential sweep
substituents can improve the catalytic activity. Given that
the redox potentials attributed to the iron porphyrin com-
plexes are quite similar in these complexes (Table 3), the
substituents rarely affected the electronic structure of the
iron centers. It should be also noted that the adsorption of
Fe‑Py on the GC electrode rarely occur under the experi-
mental conditions employed in this study (Figures S2–S4
and S6). One possible explanation for the difference in the
catalytic activity is that the hydrophobic space created by
the substituents could promote the accumulation of the sub-
strate, CO2, and the faster supply of the substrate during
catalysis resulted in a larger reaction rate.
Conclusions
This study describes the syntheses and the spectroscopic
and electrochemical properties of iron porphyrin com-
plexes with π-conjugated substituents at the meso position.
Three novel iron complexes, Fe‑Py, Fe‑PPh, and Fe‑PPy,
were synthesized by the reaction of the corresponding free-
base porphyrins (H‑Py, H‑PPh, and H‑PPy) with iron(II)
chloride tetrahydrate in DMF or NMP and character-
ized by elemental analyses and UV–vis absorption spec-
troscopy. The electrochemical analyses of the complexes
under an Ar atmosphere indicated that the introduction of
the π-conjugated substituents rarely affects the electronic
structures of the iron porphyrin complexes. However, the
catalytic activity of Fe‑Py was found to be much higher
than that of Fe‑Ph. These results indicate that the introduc-
tion of large π-conjugated substituents directly to the meso-
positions of a porphyrin moiety could provide a hydropho-
bic space suitable for the accumulation of CO2 molecules.
Experimental
Materials and methods
Pyrrole was purchased from Sigma-Aldrich Co., LLC.
1,2-dichlorobenzene, ferrocene (Fc), 4-formylphenylbo-
ronic acid, iron(II) chloride tetrahydrate, 1-methylpyrro-
lidin-2-one (NMP), 1-pyrenecarbaldehyde, triethylamine
(TEA), trifluoroacetic acid (TFA), bis(triphenylphosphine)
1 3

J Biol Inorg Chem
Fig. 6 Cyclic voltammograms
of a Fe‑Ph, b Fe‑Py and c
Fe‑PPy (1.0 mM) in a 0.1 M
TBAP/DMF solution under Ar
(solid lines) and CO2 (dotted
lines) atmospheres. (WE: GC;
CE: Pt wire; RE: Ag+/Ag; scan
rate: 20 mV s−1) The solutions
were saturated with Ar or CO2
prior to the experiments. The
concentration of CO2 in a 0.1 M
TBAP/DMF solution is approxi-
mately 0.23 M [55]. Potential
sweeps were started from the
open circuit potential −0.40 V
(Fe‑Ph), −0.28 V (Fe‑Py) and
−0.41 V (Fe‑PPy). Arrows in
the voltammograms indicate the
direction of potential sweep
redox signal for the ferrocene/ferrocenium (Fc/Fc+) cou-
ple. UV–vis absorption measurements were performed on a
Shimadzu UV-1800 spectrometer at room temperature.
Table 3 Redox potential for the prepared iron porphyrin complexes
(E1/2/V vs. Fc/Fc+) in a 0.1 M TBAP/DMF solution under Ar atmos-
phere
Compound
E1/2(1)/V
E1/2(2)/V
E1/2(3)/V
Syntheses
Fe‑Ph
Fe‑Py
Fc‑PPy
−0.63
−0.63
−0.62
−1.51
−1.47
−1.5
−2.15
−2.08
−2.11
Synthesis of 5,10,15,20‑tetrakis(phenyl)porphyrin (H‑Ph)
H‑Ph was synthesized according to the published method
1
(yield 19%) [36]. H NMR (400 MHz, CDCl3) δ 8.82 (s,
8H), 8.19–8.21 (m, 8H), 7.71–7.78 (m, 12H), −2.80 (s,
2H) ppm. Anal. Calcd for C44H30N4: C, 85.97; H, 4.92; N,
9.11. Found: C, 85.90; H, 5.02; N, 9.04%.
palladium(II) dichloride and tripotassium phosphate
were purchased from Wako Pure Chemical Industries,
Ltd. Hexane (Hex) and N,N-dimethylformamide (DMF)
were purchased from Kanto Chemical Co., Inc. Tetra(n-
butyl)ammonium perchlorate (TBAP), 1-bromopyrene,
2,3-dichloro-5,6-dicyano-1,4-benzoquinone and 4-phenylb-
enzaldehyde were purchased from Tokyo Chemical Indus-
try Co., Ltd. CDCl3 was obtained from Cambridge Isotopes,
Inc. All reagents were used without further purification.
TBAP was recrystallized from absolute ethanol and dried in
vacuo. Dichloromethane (DCM) and tetrahydrofuran (THF)
were degassed and purified under an N2 atmosphere using a
GlassContour solvent system (Nikko Hansen Co., Ltd.).
1H NMR spectra were recorded on a JEOL 400 MHz
Synthesis of 5,10,15,20‑tetrakis(pyren‑1‑yl)porphyrin
(H‑Py)
To a solution of 1-pyrenecarbaldehyde (0.93 g, 4.05 mmol)
in dry DCM (600 mL) was added pyrrole (0.28 mL,
4.05 mmol) in one portion at rt. After stirring for 5 min, to
the reaction mixture was added TFA (0.93 mL, 12.2 mmol)
at rt. The mixture was stirred for 21 h at rt and then DDQ
(1.1 g, 4.9 mmol) was added. After 90 min, the reaction was
quenched by TEA (1.7 mL, 12.2 mmol). The resulting mix-
ture was passed through a silica short column to give a black
solution. The resulting residue was purified by column chro-
matography (DCM/Hex) to afford a dark purple solution.
The desired product was recrystallized from a tiny amount
of THF and diethyl ether to give a purple solid, and the prod-
uct was washed with several portions of methanol and die-
1
instrument. All H NMR spectra were referenced against
residual proton signals of CDCl3. Elemental analyses were
measured on a MICRO CORDER JM10. Electrochemical
experiments were performed under argon and CO2 atmos-
pheres using a BAS ALS Model 650DKMP electrochemi-
cal analyzer. Cyclic voltammograms were recorded in
1,2-dichlorobenzene ([complex] = 1.0 mM; 0.1 M TBAP)
using a glassy carbon disk, a platinum wire and an Ag+/Ag
electrode (Ag/0.01 M AgNO3) that were used as the work-
ing, auxiliary, and reference electrodes, respectively. The
redox potentials of the samples were calibrated against the
1
thyl ether (73 mg, yield 6%). H NMR (400 MHz, CDCl3)
δ 8.75–8.86 (m, 4H), 8.46 (d, J = 7.9 Hz, 4H), 8.39–8.42
(m, 8H), 8.33 (d, J = 9.2 Hz, 4H), 8.24–8.29 (m, 8H), 8.08
(d, J = 6.4 Hz, 4H), 7.98–8.04 (m, 4H), 7.70-7.76 (m, 4H),
7.59–7.64 (m, 1H), 7.56 (dd, J = 9.2, 5.8 Hz, 2H), 7.47–7.51
1 3

J Biol Inorg Chem
(m, 1H), −1.98 (s, 2H) ppm. Anal. Calcd for C84H46N4·H2O:
C, 89.34; H, 4.28; N, 4.96. Found: C, 89.61; H, 4.70; N,
4.95%.
(0.38 mg, yield 16%). 1H NMR (400 MHz, CDCl3) δ 9.21
(s, 8H), 8.66 (d, J = 9.2 Hz, 4H), 8.52 (d, J = 7.9 Hz,
8H), 8.40 (dd, J = 13.0, 7.8 Hz, 8H), 8.17–8.28 (m, 20H),
8.06–8.11 (m, 12H), −2.51 (s, 2H) ppm. Anal. Calcd for
C108H62N4·2.5H2O: C, 88.80; H, 4.62; N, 3.84. Found: C,
88.87; H, 4.57; N, 3.71%.
Synthesis of 5,10,15,20‑tetrakis((1,1′‑biphenyl)‑4‑yl)
porphyrin (H‑PPh)
To a solution of 4-phenylbenzaldehyde (1.2 g, 6.6 mmol) in
dry DCM (300 mL) was added pyrrole (0.46 mL, 6.6 mmol)
in one portion at rt. After stirring for 5 min, to the reaction
mixture was added TFA (1.5 mL, 20 mmol) at rt. The mix-
ture was stirred for 1 h at rt and then DDQ (1.8 g, 8.0 mmol)
was added. After 90 min, the reaction was quenched by TEA
(2.8 mL, 20 mmol). The resulting mixture was purified by
column chromatography (DCM/Hex) to afford a dark pur-
ple solution. The desired product was recrystallized from a
tiny amount of THF and diethyl ether to give a purple solid
Synthesis of 5,10,15,20‑tetrakis(phenyl)porphyrinato
iron(III) chloride (Fe‑Ph)
Fe-Ph was prepared by the modification of a previous
report [43]. To a solution of H‑Ph (0.50 g, 0.81 mmol) in
DMF (60 mL), a DMF (21 mL) solution of FeCl2·4H2O
(0.97 g, 4.9 mmol) was added dropwise at room tempera-
ture. The mixture was refluxed for 30 min and then cooled
to room temperature. After evaporating all the solvent, the
resulting mixture was extracted with DCM. The extract
was dried over anhydrous Na2SO4, and concentrated
under reduced pressure. The desired product was precipi-
tated from DCM and Hex to give a purple solid that was
collected by filtration (0.53 g, yield 93%). Anal. Calcd for
C44H28ClFeN4: C, 75.06; H, 4.01; N, 7.96. Found: C, 75.19;
H, 4.31; N, 7.82%.
1
(0.108 mg, yield 7%). H NMR (400 MHz, CDCl3) δ 8.96
(s, 8H), 8.31 (d, J = 8.2 Hz, 8H), 8.00 (d, J = 8.2 Hz, 8H),
7.93 (d, J = 7.0 Hz, 8H), 7.60 (t, J = 7.6 Hz, 8H), 7.47 (t,
J = 7.5 Hz, 4H), −2.69 (s, 2H) ppm. Anal. Calcd for
C68H46N4·H2O: C, 87.15; H, 5.16; N, 5.98. Found: C, 87.45;
H, 5.62; N, 5.71%.
Synthesis of 4‑(pyren‑1‑yl)benzaldehyde
Synthesis of 5,10,15,20‑tetrakis(pyren‑1‑yl)porphyrinato
iron(III) chloride (Fe‑Py)
To the mixture of 1-bromopyrene (4.5 g, 16 mmol), 4-formyl-
phenylboronic acid (2.5 g, 17 mmol), tripotassium phosphate
(10 g, 48 mmol) and bis(triphenylphosphine)palladium(II)
dichloride (0.56 g, 0.80 mmol) was added dry THF (160 mL)
and water (21 mL). The mixture was refluxed with stirring for
19 h. After evaporation of THF, the precipitate was washed
by water and methanol. The mixture was purified by column
chromatography (DCM: Hex = 1:1) to afford the desired
To a solution of H‑Py (0.63 g, 0.56 mmol) in DMF
(41 mL), a DMF (15 mL) solution of FeCl2·4H2O (0.67 g,
3.4 mmol) was added dropwise at room temperature. The
mixture was refluxed for 3 h and then cooled to room tem-
perature. After evaporating all the solvent, the resulting
mixture was extracted with DCM. The extract was dried
over anhydrous Na2SO4, and concentrated under reduced
pressure. The desired product was precipitated from DCM
and Hex to give a dark brown solid (0.68 g, yield quant.).
Anal. Calcd for C84H44ClFeN4·2.5H2O: C, 81.00; H, 3.97;
N, 4.50. Found: C, 81.33; H, 4.06; N, 4.19%.
1
product (4.3 g, yield 88%). H NMR (400 MHz, CDCl3) δ
10.15 (s, 1H), 8.17–8.24 (m, 3H), 8.01–8.13 (m, 7H), 7.96 (d,
J = 7.9 Hz, 1H), 7.80 (dt, J = 8.0, 1.6 Hz, 2H) ppm.
Synthesis of 5,10,15,20‑tetrakis(4‑(pyren‑1‑yl)phenyl)
porphyrin (H‑PPy)
Synthesis of 5,10,15,20‑tetrakis((1,1′‑biphenyl)‑4‑yl)
porphyrinato iron(III) chloride (Fe‑PPh)
To a solution of 4-(pyren-1-yl)benzaldehyde (2.03 g,
6.6 mmol) in dry DCM (300 mL) was added pyrrole
(0.46 mL, 6.6 mmol) in one portion at rt. After stir-
ring for 5 min, to the reaction mixture was added TFA
(1.5 mL, 20 mmol) at rt. The mixture was stirred for 1 h
at rt and then DDQ (1.8 g, 8.0 mmol) was added. After
90 min, the reaction was quenched by TEA (2.8 mL,
20 mmol). The resulting mixture was purified by column
chromatography (DCM) to afford a dark purple solu-
tion. The desired product was recrystallized from a tiny
amount of THF and diethyl ether to give a purple solid
To the mixture of H‑PPh (41 mg, 0.045 mmol) and
FeCl2·4H2O (89 mg, 0.45 mmol) was added NMP (9 mL).
The mixture was refluxed for 1 h and then cooled to room
temperature. After evaporating all the solvent, the resulting
mixture was extracted with DCM. The extract was dried over
anhydrous Na2SO4, and concentrated under reduced pressure.
The desired product was recrystallized from a tiny amount
of THF and diethyl ether to give a dark purple solid (30 mg,
yield 66%). Anal. Calcd for C68H44ClFeN4·1.75H2O: C,
78.54; H, 4.60; N, 5.39. Found: C, 78.53; H, 4.80; N, 5.38%.
1 3

J Biol Inorg Chem
20. Sahara G, Kumagai H, Maeda K, Kaeffer N, Artero V, Higuchi
M, Abe R, Ishitani O (2016) J Am Chem Soc 138:14152–14158
Synthesis of 5,10,15,20‑tetrakis(4‑(pyren‑1‑yl)phenyl)
porphyrinato iron(III) chloride (Fe‑PPy)
21. Morria AJ, Meyer GJ, Fujita
42:1983–1994
E (2009) Acc Chem Res
To the mixture of H‑PPy (0.10 g, 0.071 mmol) and
FeCl2·4H2O (0.14 g, 0.71 mmol) was added NMP (14 mL).
The mixture was refluxed for 1 h and then cooled to room
temperature. After evaporating all the solvent, the result-
ing mixture was extracted with DCM. The extract was
dried over anhydrous Na2SO4, and concentrated under
reduced pressure. The desired product was recrystal-
lized from a tiny amount of THF and diethyl ether to
give a dark purple solid (92 mg, yield 87%). Anal. Calcd
for C108H60ClFeN4·2.25H2O: C, 83.93; H, 4.21; N, 3.63.
Found: C, 83.80; H, 4.31; N, 3.86%.
22. Costentin C, Robert M, Savéant J-M (2013) Chem Soc Rev
42:2423–2436
23. Costentin C, Robert M, Savéant J-M (2015) Acc Chem Res
48:2996–3006
24. Grodkowski J, Behar D, Neta P, Hambright P (1997) J Phys
Chem A 101:248–254
25. Dhanasekaran T, Grodkowski J, Neta P, Hambright P, Fujita E
(1999) J Phys Chem A 103:7742–7748
26. Mohamed EA, Zahran ZN, Naruta Y (2015) Chem Commun
51:16900–16903
27. Mondal B, Rana A, Sen P, Dey A (2015) J Am Chem Soc
137:11214–11217
28. Fukatsu A, Kondo M, Okabe Y, Masaoka S (2015) J Photochem
Photobiol A Chem 313:143–148
29. Hammouche M, Lexa D, Momenteau M, Savéant J-M (1991) J
Am Chem Soc 113:8455–8466
30. Costentin C, Drouet S, Robert M, Savéant J-M (2012) Science
338:90–94
31. Costentin C, Passard G, Robert M, Savéant J-M (2014) J Am
Chem Soc 136:11821–11829
32. Bhugun I, Lexa D, Savéant J-M (1994) J Am Chem Soc
116:5015–5016
Acknowledgements This work was supported by a Grant-in-Aid for
Scientific Research (B) (No. 16H04125) (to S.M.), a Grant-in-Aid for
Young Scientists (A) (No. 15H05480) (to M.K.), and a Grant-in-Aid
for Challenging Exploratory Research (No. 26620160) (to S.M.) from
the Japan Society for the Promotion of Science. This work was also
supported by a Grant-in-Aid for Scientific Research on Innovative
Areas “AnApple” (No. 15H00889) and ACT-C, JST.
33. Bhugun I, Lexa D, Savéant J-M (1996) J Am Chem Soc
118:1769–1776
34. Bhugun I, Lexa D, Savéant J-M (1996)
100:19981–19985
J Phys Chem
References
35. Costentin C, Drouet S, Passard G, Robert M, Savéant J-M (2013)
J Am Chem Soc 135:9023–9031
36. Adler AD, Longo FR, Shergalis W (1964) J Am Chem Soc
86:3145–3149
37. Knör G (2001) Inorg Chem Commun 4:160–163
38. Penon O, Marsico F, Santucci D, Rodriguez L, Amabilino DB,
Pérez-García L (2012) J Porphyr Phthalocyanines 16:1293–1302
39. Lindsey JS, Hsu HC, Schreiman IC (1986) Tetrahedron Lett
27:4969–4970
40. Lindsey JS, Wagner RW (1989) J Org Chem 54:828–836
41. Ishikawa R, Katoh K, Breedlove BK, Yamashita M (2012) Inorg
Chem 51:9123–9131
1. Lewis NS, Nocera DG (2006) Proc Natl Acad Sci USA
103:15729–15735
2. Gray HB (2009) Nat Chem 1:7
3. Tanaka K, Ooyama D (2002) Coord Chem Rev 226:211–218
4. Savéant J-M (2008) Chem Rev 108:2348–2378
5. Takeda H, Ishitani O (2010) Coord Chem Rev 254:346–354
6. Schneider J, Jia H, Muckermana JT, Fujita E (2012) Chem Soc
Rev 41:2036–2051
7. Windle CD, Perutz RN (2012) Coord Chem Rev 256:2562–2570
8. Kang P, Chen Z, Brookhart M, Meyer TJ (2015) Top Catal
58:30–45
9. Beley M, Collin JP, Ruppert R, Sauvage JP (1986) J Am Chem
Soc 108:7461–7467
42. Kobayashi H, Higuchi T, Kaizu Y, Osada H, Aoki M (1975) Bull
Chem Soc Jpn 48:3137–3141
43. Sun Z-C, She Y-B, Zhou Y, Song X-F, Li K (2011) Molecules
16:2960–2970
10. Nagao H, Mizukawa T, Tanaka
33:3415–3420
K (1994) Inorg Chem
11. Chen Z, Chen C, Weinberg DR, Concepcion JJ, Harri-
son DP, Brookhart MS, Meyer TJ (2011) Chem Commun
47:12607–12609
12. Smieja JM, Benson EE, Kumar B, Grice KA, Seu CS, Miller
AJM, Mayer JM, Kubiak CP (2012) Proc Natl Acad Sci USA
109:15646–15650
13. Keith JA, Grice KA, Kubiak CP, Carter EA (2013) J Am Chem
Soc 135:15823–15829
14. Jeletic MS, Mock MT, Appel AM, Linehan JC (2013) J Am
Chem Soc 135:11533–11536
15. Kobayashi K, Tanaka K (2014) Phys Chem Chem Phys
16:2240–2250
16. Chen Z, Kang P, Zhang M-T, Meyer TJ (2014) Chem Commun
50:335–337
17. Kang P, Zhang S, Meyer TJ, Brookhart M (2014) Angew Chem
Int Ed 53:8709–8713
18. Johnson BA, Agarwala H, White TA, Mijangos E, Maji S, Ott S
(2016) Chem Eur J 22:14870–14880
44. Gouterman M (1959) J Chem Phys 30:1139–1161
45. Gouterman M (1961) J Mol Spectrosc 6:138–163
46. Stockmann A, Kurzawa J, Fritz N, Acar N, Schneider S, Daub J,
Engl R, Clark T (2002) J Phys Chem A 106:7958–7970
47. Raytchev M, Pandurski E, Buchvarov I, Modrakowski C, Fiebig
T (2003) J Phys Chem A 107:4592–4600
48. Weigel W, Rwttig W, Dekhtyar M, Modrakowski C, Beinhoff M,
Schlüter AD (2003) J Phys Chem A 107:5941–5947
49. Wu W, Wu W, Ji S, Guo H, Wang X, Zhao J (2011) Dyes Pig-
ment 89:199–211
50. Mack J, Stillman MJ (2003) The porphyrin handbook. In: Kadish
K, Smith KM, Guilard R (eds) vol 16. Elsevier Science, San
Diego, pp 43–116
51. Lexa D, Savéant J-M, Su K-B, Wang D-L (1988) J Am Chem
Soc 110:7617–7625
52. Anxolabéhère É (1994) Chottard G, Lexa D. New J Chem
289:889–899
53. Rountree ES, McCarthy BD, Eisenhart TT, Dempsey JL (2014)
Inorg Chem 53:9983–10002
19. Kothandaraman J, Goeppert A, Czaun M, Olah GA, Prakash
GKS (2016) J Am Chem Soc 138:778–781
1 3

J Biol Inorg Chem
54. Okamura M, Kondo M, Kuga R, Kurashige Y, Yanai T, Hayami
S, Praneeth VKK, Yoshida M, Yoneda K, Kawata S, Masaoka S
(2016) Nature 530:465–468
55. Gennaro A, Isse AA, Vianello E (1990) J Electroanal Chem
289:203–215
1 3