Iron-Catalyzed Water Oxidation: O–O Bond Formation via Intramolecular Oxo–Oxo Interaction

Article abstract of DOI:10.1002/anie.202100060

Herein, we report the importance of structure regulation on the O?O bond formation process in binuclear iron catalysts. Three complexes, [Fe₂(μ-O)(OH₂)₂(TPA)₂]⁴⁺ (1), [Fe₂(μ-O)(OH₂)₂(6-HPA)]⁴⁺ (2) and [Fe₂(μ-O)(OH₂)₂(BPMAN)]⁴⁺ (3), have been designed as electrocatalysts for water oxidation in 0.1 M NaHCO₃ solution (pH 8.4). We found that 1 and 2 are molecular catalysts and that O?O bond formation proceeds via oxo–oxo coupling rather than by the water nucleophilic attack (WNA) pathway. In contrast, complex 3 displays negligible catalytic activity. DFT calculations suggested that the anti to syn isomerization of the two high-valent Fe=O moieties in these catalysts takes place via the axial rotation of one Fe=O unit around the Fe-O-Fe center. This is followed by the O?O bond formation via an oxo–oxo coupling pathway at the Fe^IVFe^IV state or via oxo–oxyl coupling pathway at the Fe^IVFe^V state. Importantly, the rigid BPMAN ligand in complex 3 limits the anti to syn isomerization and axial rotation of the Fe=O moiety, which accounts for the negligible catalytic activity.

Full text of DOI:10.1002/anie.202100060

Angewandte
Research Articles
Chemie
How to cite:
International Edition: doi.org/10.1002/anie.202100060
German Edition: doi.org/10.1002/ange.202100060
Water Oxidation
Hot Paper
Iron-Catalyzed Water Oxidation: O–O Bond Formation via
Intramolecular Oxo–Oxo Interaction
Hong-Tao Zhang, Xiao-Jun Su, Fei Xie, Rong-Zhen Liao,* and Ming-Tian Zhang*
Dedicated to the 100th anniversary of the Department of Chemistry at Nankai University
Abstract: Herein, we report the importance of structure
inherent difficulty of water oxidation catalysis, lack of enough
regulation on the O O bond formation process in binuclear
understanding of multielectron/multiproton transfer (4eꢀ/
ꢀ
iron catalysts. Three complexes, [Fe2(m-O)(OH2)2(TPA)2]4+
(1), [Fe2(m-O)(OH2)2(6-HPA)]4+ (2) and [Fe2(m-O)(OH2)2-
(BPMAN)]4+ (3), have been designed as electrocatalysts for
water oxidation in 0.1 M NaHCO3 solution (pH 8.4). We found
4H+ for water oxidation) and O O bond formation. To
ꢀ
develop earth-abundant catalysts with performance as well as
or better than precious materials is one of the challenges for
water oxidation catalysis. Precise understanding of catalytic
mechanisms and identification of structures and oxidation
states of transient species during catalytic turnover are crucial
for rational catalyst design.
ꢀ
that 1 and 2 are molecular catalysts and that O O bond
formation proceeds via oxo–oxo coupling rather than by the
water nucleophilic attack (WNA) pathway. In contrast, com-
plex 3 displays negligible catalytic activity. DFT calculations
suggested that the anti to syn isomerization of the two high-
Iron is considered as the most attractive candidate for
water oxidation due to its cheapness, abundance and biocom-
patibility. Collins and co-workers firstly reported a monoiron
complex (Fe-TAML) for catalytic water oxidation at pH 0.7
with CeIV(CAN) as sacrificial oxidants.[6a] Fillol and Costas
described a family of neutral tetradentate ligands based iron
complexes as WOCs in which the FeV(O)(OH) was proposed
=
valent Fe O moieties in these catalysts takes place via the axial
=
rotation of one Fe O unit around the Fe-O-Fe center. This is
ꢀ
followed by the O O bond formation via an oxo–oxo coupling
pathway at the FeIVFeIV state or via oxo–oxyl coupling pathway
at the FeIVFeV state. Importantly, the rigid BPMAN ligand in
complex 3 limits the anti to syn isomerization and axial
ꢀ
as the key intermediate for O O bond formation via water
rotation of the Fe O moiety, which accounts for the negligible
nucleophilic type mechanism.[6b] Meyer and Dhar also
=
V
=
catalytic activity.
demonstrated that high valent intermediate of Fe O can
ꢀ
promote the O O bond formation by WNA pathway with
Introduction
unsatisfying efficiency.[6c,d] Recently, Lloret-Fillol and Costas
found that highly active FeV O can cause the collapse of itself
=
Water oxidation (2H2O ! O2 + 4H+ + 4eꢀ) is an
essential step in both natural and artificial photosynthesis,
because it is the ideal electron/proton source for chemical
fuels production.[1] To date, numerous transition-metal com-
plexes (such as Ru,[2] Ir,[3] Mn,[4] Co,[5] Fe,[6] Ni[7] and Cu[8])
have been used as molecular water oxidation catalysts
(WOCs) to lower the kinetics barriers of this reaction. These
previous studies have partially revealed the catalytic mech-
skeleton via hydrogen atom transfer and pointed out the
possible stability problems of mononuclear Fe-based cata-
lyst.[9] To further tune the O O bond formation step, di- or
ꢀ
polynuclear iron-based WOCs also appeared on the re-
search.[10] Najafpour and co-workers first found that dinuclear
[Fe2(m-O)(OH2)2(TPA)2]4+ was six times faster than mono-
nuclear [Fe(TPA) (OTf or Cl)2] with CeIV as oxidant.[10a]
However, Sakai and his co-workers reported an analogous
possessing TPA ligands could be dissociated into more active
single Fe-site catalyst when lowering the pH.[10b] Interestingly,
Que and his-co-workers reported a (m-oxo)diiron(IV) com-
plex supported by a pentadentate ligand could activate H2O
via proton-coupled electron transfer (PCET) to generate the
hydroxyl radical. Unfortunately, the generated hydroxyl
radical was consumed by the solvent, acetonitrile, and no O2
or H2O2 was detected.[11] Recently, Masaoka developed
a highly efficient pentanuclear iron catalyst with TOF of
1900 sꢀ1, in which the charge is effectively distributed on
ꢀ
anism of water oxidation. For example, two types of O O
bond formation mechanism, water nucleophilic attack
(WNA) and interaction of two M-O units (I2M), are proposed
for guiding the catalyst design.[1f] However, water oxidation is
still the bottleneck for artificial photosynthesis because of the
[*] H.-T. Zhang, X.-J. Su, F. Xie, Prof. Dr. M.-T. Zhang
Center of Basic Molecular Science (CBMS)
Department of Chemistry, Tsinghua University
Beijing, 100084 (China)
E-mail: mtzhang@mail.tsinghua.edu.cn
ꢀ
multi-irons to improve the catalytic performance and O O
R.-Z. Liao
bond formation was proposed via the coupling of two
terminal oxo groups.[10e,12] The stability of these Fe-based
molecular catalysts is also a challenge because the foamed
Key Laboratory for Large-Format Battery Materials and System
School of Chemistry and Chemical Engineering
Huazhong University of Science and Technology
Wuhan 430074 (China)
=
high valent Fe O could abstract the hydrogen atom from the
ligand.[13] Above all, the study of iron catalyzed water
oxidation has significant advance and the catalytic activity is
strongly dependent on the ligand architecture and iron
E-mail: rongzhen@hust.edu.cn
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.202100060.
ꢀ
nuclearity. The further studies on key intermediates and O
&&&&
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
IV
IV
=
ꢀ ꢀ
=
O bond formation mechanism are helpful to regulate its
catalytic activity and design new catalysts.
[O Fe O Fe O] intermediate, they could get close to
each other by rotating axially around the Fe-O-Fe and further
couple in a relatively flexible structure. When the molecular
According to the principle of microreversibility, well-
structure becomes rigid, the rotation of FeIV O will be
ꢀ
=
studied O O bond cleavage in oxygen activation via diiron
complexes can give us more inspiration to explore the detail
limited, making them difficult to couple face-to-face, and the
catalyst would have lower activity. To verify this proposal, we
employed [Fe2(m-O)(OH2)2(TPA)2]4+ (1), [Fe2(m-O)(OH2)2(6-
HPA)]4+ (2) and [Fe2(m-O)(OH2)2(BPMAN)]4+ (3) as electro-
catalysts to catalyze water oxidation in an alkaline condition.
Complexes 1 and 2 exhibit higher activity among the three
complexes in NaHCO3 solution (pH 8.4), while complex 3 is
inactive under the form of FeIVFeIV species. As designed, rigid
=
ꢀ
of O O bond formation in water oxidation. Soluble methane
monooxygenase (sMMO) which includes diiron active site
can efficiently activate oxygen and subsequently form active
intermediate.[14] Now, groups of Que and Kodera have
ꢀ
extensively studied O O bond cleavage by designing the
sMMO models.[15] Que and co-workers have reported that
intermediate of (m-oxo)(m-hydroperoxo)diiron(III,III) can
turn into (m-oxo)(hydroxo)(oxo)diiron(IV,IV) through cleav-
age of peroxide bond, and the latter can undergo proton
BPMAN architecture in 3 limits the axial rotation of FeIV
O
and further prevents the coupling of two FeIV O. Combined
=
IV
IV
=
ꢀ ꢀ
=
ꢀ
transfer to yield [O Fe O Fe O] intermediate in the
with experimental results and DFT calculations, O O bond
presence of base (Scheme 1a).[15d,h] The high-valent diiron
formation via face-to-face coupling between two FeIV O was
=
IV
IV
=
ꢀ ꢀ
=
intermediate [O Fe O Fe O] has also been described by
Kodera and co-workers through reacting corresponding
diiron(III) complex with H2O2 and Et3N, and it was in
thermal equilibrium with its (m-oxo)(m-1,2-peroxo)diiron(III)
confirmed and thus provides a direction for the design of
molecular catalysts and the regulation of catalytic efficiency.
isomer (Scheme 1b).[15f,16] Interestingly, O O bond is formed Results and Discussion
ꢀ
V
in this thermal equilibrium via FeIV O rather than Fe
O
=
=
which is extensively proposed in mononuclear iron modulated
water oxidation catalysis. So, this encourages us to explore
Solid-state structures and catalytic species for three diiron
complexes. The synthetic details are listed in SI. The crystal
structures of three diiron(III) complexes designed by subtle
modification of ligand are shown in Figure 1.[17] Similar to the
literature,[10a,16] the two Fe atoms are separated by 3.567 ꢀ
and 3.601 ꢀ, with an Fe-O-Fe angle of 169.78 and 179.08, of
complex 1 and complex 2, respectively. Complex 3 with
inflexible BPMAN ligand shows shorter Fe1···Fe2 distance
(3.201 ꢀ) which is closer to the Fe···Fe distance (3.221(6) ꢀ)
of [Fe2(BPMAN)(m-OH)-(m-O2CPhCy)](OTf)2,[18] and an
angle with bridge O1 (]Fe1-O1-Fe2) at 129.08. The two
ꢀ
whether the reverse O O bond formation can be achieved
through the coupling of two FeIV O moieties. Structure-
=
activity relationship can usually provide pivotal and solid
ꢀ
mechanism information, which can be used to study O O
bond formation. As designed in Scheme 1c, while two
terminal FeIV O moieties initially locate at anti position in
=
Figure 1. The X-ray ORTEP plots of [Fe2(m-O)(OH2)2 (TPA)2]4+ (1, (a)),
[Fe2(m-O)(OH2)2(6-HPA)]4+ (2, (b)) and [Fe2(m-O)(OH2)2(BPMAN)]4+
ꢀ
ꢀ
(3, (c)). H atoms (except for O H bonds), solvents, ClO4 and
ꢀ
CF3SO3 anions have been omitted for clarity. Ellipsoids are plotted at
Scheme 1. Transformation of m-oxo-m-peroxodiiron(III) to m-oxo-
dioxodiiron(IV) in O2 activation by Que (a) and Kodera (b); The
99.5% (1) or 50% (2 and 3) probability. d) Comparison of pivotal
crystal data, involving the distance between two iron atoms (Fe1···Fe2),
the angle ]Fe1-O1-Fe2 and the dihedral angle formed by four atoms
with Fe1-Fe2 as the axis (]O2-Fe1Fe2-O3).
ꢀ
proposed alternative O O bond formation in this work (c).
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
H2O molecules in the structure of 1–3 all bind in an anti
fashion with respect to the diiron vector. In particular, the
difference in ligand flexibility and coordination geometry
among 1–3 provides a research platform for understanding
the relationship between structure and catalytic activity and
exploring the details of mechanism.
different scan rates, the peak currents varied linearly with the
square root of the scan rate (Figure S9a), which indicates
a diffusion-controlled process. This is assigned as the one-
electron reduction of Fe2(III,III) to Fe2(III,II). After two
irreversible oxidative processes, the CV of 0.5 mM complex
1 exhibited a significant catalytic activity, during the scanning
at 100 mVsꢀ1 from 0.24 to 2.0 V vs. NHE (Figure 2a, Fig-
ure S10). CVs were further recorded at different scan rates
and the resulting normalized catalytic currents increased with
decreasing scan rate (Figure S9b), consistent with a water
oxidation catalysis process. The onset potential for water
oxidation appears at 1.57 V vs. NHE with an overpotential of
ꢁ 830 mV followed by two oxidative waves. When complex 2
was used as WOC under the same condition, only one
irreversible catalytic process was observed at 1.76 V vs. NHE
(Figure 2b, Figures S11 and S12b) and the catalytic current is
lower than 1. Complex 3 displays two oxidative waves at 1.36
and 1.64 V vs. NHE (Figure 2c, Figure S13), but normalized
catalytic currents of 3 nearly keep constant at different scan
rates (Figure S14), indicating that complex 3 cannot catalyze
water oxidation. Additionally, simple Fe3+ ion catalyzed water
oxidation was excluded by control experiments. 1.0 mM of
Fe(NO3)3, Fe(ClO4)3 and FeCl3 were examined under iden-
tical conditions, respectively (Figure S15). All Fe3+ ions
showed no activity towards water oxidation indicating that
the observed catalytic activity is not caused by free Fe3+ ions
or formed iron oxides which may possibly exist during
catalysis.
The oxygen evolution was further confirmed by con-
trolled-potential electrolysis (CPE) (Figure 3a) using a 2 cm2
BDD disk working electrode. For CPE at 1.57 V vs. NHE,
complex 1 exhibits the highest catalytic current density of
0.7 mAcmꢀ2, followed by complex 2 with 0.3 mAcmꢀ2 and
complex 3 shows negligible 30 mAcmꢀ2 equivalent to blank
current. The dissolved oxygen was also measured by a cali-
brated Ocean Optics FOXY probe during CPE (Figure 3b).
The concentration of O2 dissolved in the solution phase
increased from 54 mM to 301 mM for 1, from 54 mM to 180 mM
for 2 and negligible for 3. For both 1 and 2, the Faraday
efficiency for the O2 evolution is above 90% in initial 5
minutes (Table S1). The oxygen evolution rate for 1 is about 2
times faster than 2, and 3 is inactive toward catalytic water
oxidation, indicating that the catalytic activity is strongly
dependent on the choice of ligand and the regulation of the
ligand framework can achieve tuning its catalytic perfor-
mance.
We firstly determined the pKa values of the aqua ligand of
1–3 [Eq. (1)–(2)] by potentiometric titration. The pKa1 values
are 4.41, 4.93 and 4.96 for 1H O=H O, 2H O=H O and 3H O=H O
,
2
2
2
2
2
2
respectively. The pKa2 values are 5.22, 5.81 and 6.10 for
H O=HO, 2H O=HO and 3H O=HO, respectively (Figures S4–S6 and
1
2
2
2
Table S2). It can be found that as the rigidity of the ligand
increases, the acidity of the aqua ligand gradually decreases.
To a certain extent, this indicates that the chemical reactivity
of the aqua ligand is sensitive to the regulation of the ligand.
According to the pKa2 values, the HO-FeIII-O-FeIII-OH
species is the dominant existence at pH > 8 buffer solution
(Figures S4–S6). In terms of acetate,[19] phosphate[20] and
sulphate[10b,20] could play as bridge ligand to inhibit the
catalytic efficiency by occupying the coordination sites of
water molecules, diiron(III) complexes in 0.1 M NaHCO3
(pH 8.4) solution was applied for electrocatalytic water
oxidation investigation.[21]
Electrochemical behavior and oxygen evolution of diiron
complexes 1–3. Figure 2 shows the CVs of 0.5 mM complex 1–
3 in 0.1 M NaHCO3 (pH 8.4) using BDD electrode. For
complex 1, on the cathodic direction, an irreversible cathodic
process at Epc = ꢀ0.23 V vs. NHE was observed in the CV, and
an irreversible anodic process at Epa = 0.40 V vs. NHE was
observed in the reverse scan (Figure S8, with peak–peak
splitting DEp = 630 mV). When the CV was measured at
The stability of 1 and 2 during the electrocatalytic process
was further examined by electrode analysis after long time
bulk electrolysis (Figure S16). The morphology of electrode
did not change before and after electrolysis in scanning
electron microscopy (SEM) images (Figure S17), which
indicated that there was no FexOy formation during the
electrocatalytic process. Moreover, X-ray photoelectron
spectroscopy (XPS, Figure S18) analysis indicated that no
Fe species such as FexOy were deposited on the electrode
surface, and rinse tests also showed that there was no active
species attached to the electrode surface (Figure S19). CVand
UV-vis absorption of the solution after electrolysis was
consistent with the fresh solution (Figures S20–S22), indicat-
Figure 2. CVs recorded in 0.1 M NaHCO3 (pH 8.4) solution using
a BDD electrode without (black line) and with 0.5 mM 1 ((a) red line),
0.5 mM 2 ((b) blue line), and 0.5 mM 3 ((c) green line); the scan rate
was 100 mVsꢀ1
.
&&&&
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
current of the catalytic wave could be expressed by Equa-
tion (4),[22]
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ð4Þ
icat ¼ ncatFA½Cat:ꢂ kobsDcat
where F is the Faraday constant, A is the electrode surface
area, id is the peak current of the diffusion-controlled wave, icat
is the peak current of the catalytic wave, [Cat.] is the bulk
concentration of the diiron catalyst, Dcat is the diffusion
coefficient of the catalyst, nd = 1 is the number of electrons
transferred in this diffusion-controlled process, a = 0.5 is the
transfer coefficient of the catalyst and ncat = 4 is the electro-
chemical stoichiometry for water oxidation. In order to
eliminate background interference, the catalytic current is
corrected by subtracting the background current (Figure S30)
at the same potential, marked as icat’. According to the plot of
pffiffi
icat’/id versus 1/ v (Figure S31), the observed rate constant for
water oxidation catalysis at 1.89 V vs. NHE, kobs, was
determined as 0.55 sꢀ1 for 1 and 0.04 sꢀ1 for 2. This apparent
catalytic efficiency (kobs) is consistent with the oxygen
evolution rate observed in the bulk electrolysis experiment
in which 1 is much faster than 2, and 3 is almost catalytic
silence. The catalytic performance of 1 and 2 is comparable
with mostly reported Fe-based WOCs (Table S3). The kinet-
ics of water oxidation are also investigated in D2O and the
resulted kinetic isotope effect (KIE = kcat,H O/kcat,D O) of
2
2
catalysts 1 and 2 are 1.23 and 1.12, respectively (Figures S32,
S33). The observed KIE is in contrast to the single-site RuV
=
Figure 3. a) Controlled-potential electrolysis with blank solution (black
line) and 0.5 mM 1 (red line), 0.5 mM 2 (blue line), and 0.5 mM 3
(green line) with BDD disk working electrode (2 cm2) in 0.1 M
NaHCO3 (pH 8.4) solution at 1.57 V vs. NHE. b) The O2 evolution
during the CPE with blank solution (black line), 0.5 mM 1 (red line),
0.5 mM 2 (blue line), and 0.5 mM 3 (green line) was measured with
a fluorescence probe.
O cases[2n,23] in which the remarkable KIE is attributed to
atom-proton transfer (APT) with O O bond formation. Both
the KIE and the catalytic behavior differences among 1–3
ꢀ
ꢀ
indicate that WNA pathway of O O bond formation via
V
single site FeIV O or Fe O can be excluded. More crucially,
=
=
ꢀ
more details on the O O bond formation could be obtained
by understanding the special case 3. Complex 3 has similar
electrochemistry oxidation behavior in E-pH diagram as
IV
IV
=
ꢀ ꢀ
=
catalyst 1 (Figures S24, S26), indicating that O Fe O Fe
ing the species in solution was robust toward the bulk
electrocatalysis, and this was further confirmed by the
constant CV shape in multiple CVs scan (Figure S23). All
experimental results indicate that both complexes 1 and 2 are
the stable molecular electrocatalyst for water oxidation under
the conditions in this work rather than nanoparticles, despite
the difficulty of definitively excluding a colloid material.
Water oxidation kinetics and mechanistic analysis. Accord-
ing to the above discussion, 1 and 2 are able to act as
homogeneous catalysts to electrocatalyze water oxidation,
while 3 is catalytic silence, although all these three com-
pounds, 1–3, could generate corresponding high-valent
FeIVFeIV intermediates via two-electron oxidation (Figur-
es S24–S26). The currents of both the diffusion-controlled
waves and catalytic waves for water oxidation by 1 and 2 are
linearly dependent on the concentration of the catalysts,
[Cat.], (Figures S27, S28). This suggests that the irreversible
ꢀ
O species with rigid BPMAN ligand is inactive for O O bond
formation, because two terminal FeIV O moieties of O Fe
O Fe O initially locate at anti position. While the Fe
IV
=
=
ꢀ
IV
IV
ꢀ
=
=
O
moieties with flexible ligand could rotate axially around the
Fe O Fe. This results in the direct coupling of two Fe O to
form O O bond and further exhibit the ability of water
IV
ꢀ ꢀ
=
ꢀ
oxidation. This process is a reverse process of oxygen
activation reported by both Que[15h] and Kodera.[15f] In O2
activation, Kodera suggested that a syn-dioxo form of catalyst
2 was transformed into anti-dioxo form through a t-oxo-di(m-
oxo)diiron(IV) transition state.[15f,j]
The DFT calculations[24] were performed to further
explain the different catalytic activity of 2 and 3. For these
dinuclear iron complexes with a bridging oxo ligand, the more
stable antiferromagnetically coupled state[25] was considered
and presented. The protonation state of the diferric complex 2
at the working pH 8.4 was first identified. The pKa1 and pKa2
of 2 with two water molecules coordinated were calculated to
be 4.9 and 6.8, respectively. These two values are in excellent
agreement with our experimental data (4.93 and 5.81). This
suggested that the most stable form of 2 has two hydroxide
ligands in anti position (labeled as 2III,III), with a total charge
processes at ꢀ0.23 V (in 1) and ꢀ0.26 V (in 2) vs. NHE for the
III,III/III,II
Fe2
couple should obey the Randles-Svecik equation
in Equation (3) (Figures S8, S9a, S12a and S29) and the
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ð3Þ
ꢀ 2021 Wiley-VCH GmbH
id ¼ 0:496ndFA½Cat:ꢂ ndaFvDcat=RT
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
of + 2. The electronic structure of 2III,III can be described as
a featuring a broken-symmetry open-shell singlet with the two
high spin ferric ions (S = 5/2) interacting in an antiferromag-
netic fashion. The spin densities on the two iron ions are 4.09
and ꢀ4.09, respectively.
The oxidation of 2III,III was calculated to a proton-coupled
electron transfer process, with a potential of 1.20 V vs. NHE,
leading to the formation of a doublet complex 2III,IV (pKa of
6.2 for its protonated form). With a reference potential of
1.57 V vs. NHE (experimental potential for CPE), this step
was calculated to be exergonic by 8.5 kcalmolꢀ1 (Figure 4).
The electronic structure of 2III,IV can be interpreted as a high
ꢀ
Figure 5. Structures of key transition states for O O bond formation
starting from 2IV,IV and 2IV,V. Distances are given in Angstrom, while
spin densities on key atoms are indicated in red italics.
ꢀ
O O bond formation at a higher oxidation state has also
Figure 4. Gibbs energy diagram (in kcalmolꢀ1) at the B3LYP*-D3 level
for water oxidation catalyzed by 2.
been considered. Further one-electron oxidation of 2IV,IV to
produce 2IV,V (total charge of + 3) is associated with a potential
of 1.68 V vs. NHE. With an applied potential of 1.57 V vs.
NHE, this process is endergonic by 2.5 kcalmolꢀ1. The anti to
syn isomerization of 2IV,V now becomes more favorable, the
structure of the corresponding transition state 2-TS1’ is
displayed in Figure 5. The barrier for 2-TS1’ was calculated
to be only 10.2 kcalmolꢀ1 relative to 2IV,V. A plausible reason
is that the charge repulsion between the two metal centers
(IV, V vs. IV, IV) becomes larger, which makes the
dissociation of the bridging O1 from Fe2 easier. This is
evidenced by the significantly longer Fe2-O1 and Fe2-O2
distances in 2-TS1’ (2.28 ꢀ and 2.61 ꢀ) compared with those
in 2-TS1 (2.20 ꢀ and 2.08 ꢀ). In addition, this isomerization
to form 2-Int1’ becomes exergonic by 2.1 kcalmolꢀ1. Similarly,
spin FeIII ion (S = 5/2) antiferromagnetically coupled to the
high spin FeIV ion (S = 2). Further PCET oxidation of 2III,IV
(potential of 1.32 V vs. NHE) affords an open-shell singlet
complex 2IV,IV, which has two antiferromagnetically coupled
FeIV ions (S = 2 for each iron). The Mulliken atomic spin
densities on the two iron ions become 3.13 and ꢀ3.13,
respectively, while they are 0.43 and ꢀ0.43 on the two
terminal oxygen atoms, respectively. In 2IV,IV, the two oxo
groups are in anti position, and an anti to syn isomerization
has to take place via the axial rotation of one of the two FeIV
=
O moieties. The transition state for this process has been
optimized and labeled as 2-TS1 (Figure 5), which has a barrier
of 19.2 kcalmolꢀ1 relative to 2IV,IV, and this isomeric formation
of 2-Int1 is endergonic by 3.3 kcalmolꢀ1. In 2-TS1, both O1
and O2 become the bridging ligands, which results in the
formation of a hexa-coordinated Fe1 and a hepta-coordinated
ꢀ
the following O O bond formation by the coupling of the two
antiferromagnetically coupled oxo/oxyl groups via 2-TS2’ has
a very facile barrier of 9.3 kcalmolꢀ1 relative to 2-Int1’. This
leads to the production of a diferric superoxide intermediate
2-Int2’, which can also be generated by the one-electron
oxidation of 2-Int2 (potential of 0.59 V vs. NHE). The
alternative water nucleophilic attack on 2-Int1’ (2-TS3’ in
Figure S34 in SI) has also been considered. However, its
barrier was found to be 23.4 kcalmolꢀ1 relative to 2-Int1’ plus
a water molecule, and the total barrier becomes 23.8 kcal
molꢀ1 when the energy cost for the formation of 2-Int1’ is
added. This is much higher than that via the oxo-oxo coupling.
From 2-Int2’, the coordination of a water molecule to Fe1 to
form 2-Int3 is endergonic by 7.4 kcalmolꢀ1. In the subsequent
step, a PCET oxidation takes place to afford 2-Int4, with
a potential of 0.08 V vs. NHE. Finally, the coordination of
ꢀ
Fe2. From 2-Int1, O O bond formation proceeds via the
coupling of the two antiferromagnetically coupled oxo/oxyl
groups (2-TS2, Figure 5), similarly to the pentanuclear iron
catalyst,[12] and the heterotrinuclear MnRu2 catalyst.[26] The
barrier for 2-TS2 was calculated to be 11.0 kcalmolꢀ1 relative
to 2-Int1. 2-TS2 was characterized to have an imaginary
ꢀ1
ꢀ
frequency of 471.3i cm , which corresponds to the O O bond
ꢀ
formation. In 2-TS2, the nascent O O distance is 1.91 ꢀ, and
the atomic spin densities on O1 and O3 are 0.48 and ꢀ0.51,
respectively. The generation of the diferric peroxide inter-
mediate 2-Int2 is exergonic by 9.9 kcalmolꢀ1 from 2-Int1.
&&&&
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
a second water molecule and the release of O2 and a proton
regenerate 2III,III for the next water oxidation cycle.
achieved via intramolecular oxo/oxo or oxo/oxyl coupling and
=
thus it required the spatial position match of the two Fe O.
ꢀ
From Figure 4, it can be seen that the O O bond
This feature provides a possibility to regulate the catalytic
reactivity. The experimental results show that the tuning of
the strength of ligand rigidity can effectively regulate the
formation prefers to take place in the oxidation state of
IV,V
Fe2
under the experimental condition. The anti to syn
ꢀ
isomerization turns out to be the rate-limiting step, with
a total barrier of 12.7 kcalmolꢀ1, which is lower than that in
the oxidation state of Fe2IV,IV (19.2 kcalmolꢀ1). For complex 3,
the calculations showed that the two PCEToxidations to form
energy barrier of the O O bond formation, for example,
[Fe2(m-O)(OH2)2(BPMAN)]4+ (3) with rigid ligand has almost
no catalytic activity. This work clarifies the bimetallic
cooperative effect on the O O bond formation and provide
ꢀ
3
IV,IV are facile under the experimental condition. However,
the inspiration for molecular design and catalytic regulation
on iron catalyzed water oxidation. Further experiments are
focused on developing new catalysts based on the mechanistic
insight disclosed in this work, particular to reduce the
overpotential and enhance the catalytic activity.
the anti to syn isomerization has a very high barrier of
32.5 kcalmolꢀ1 (Figure S35 in SI), due to the usage of a too
rigid BPMAN ligand. The one-electron oxidation of 3IV,IV also
has a very high potential of 1.92 V vs. NHE.
Based on the experimental results and DFT calculations
discussed above, we can propose the following water oxida-
tion mechanism by diiron(III) complexes (Figure 6). The
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (Grant No. 219330007, 21661132006).
Conflict of interest
The authors declare no conflict of interest.
ꢀ
Keywords: artificial photosynthesis · iron catalysts · O
O bond formation · water oxidation
Figure 6. Proposed mechanism of complexes 1 and 2 catalyzed water
oxidation in 0.1 M NaHCO3 (pH 8.4).
[1] a) J. H. Alstrum-Acevedo, M. K. Brennaman, T. J. Meyer, Inorg.
Chem. 2005, 44, 6802 – 6827; b) N. S. Lewis, D. G. Nocera, Proc.
Natl. Acad. Sci. USA 2006, 103, 15729 – 15735; c) M. H. V.
Huynh, T. J. Meyer, Chem. Rev. 2007, 107, 5004 – 5064; d) T. J.
Meyer, M. H. V. Huynh, H. H. Thorp, Angew. Chem. Int. Ed.
2007, 46, 5284 – 5304; Angew. Chem. 2007, 119, 5378 – 5399;
e) T. A. Betley, Q. Wu, T. Van Voorhis, D. G. Nocera, Inorg.
Chem. 2008, 47, 1849 – 1861; f) M. D. Kꢁrkꢁs, O. Verho, E. V.
Johnston, B. ꢀkermark, Chem. Rev. 2014, 114, 11863 – 12001.
[2] a) S. W. Gersten, G. J. Samuels, T. J. Meyer, J. Am. Chem. Soc.
1982, 104, 4029 – 4030; b) R. Zong, R. P. Thummel, J. Am. Chem.
Soc. 2005, 127, 12802 – 12803; c) J. J. Concepcion, J. W. Jurss,
M. K. Brennaman, P. G. Hoertz, A. O. Patrocinio, N. Y. Mur-
akami Iha, J. L. Templeton, T. J. Meyer, Acc. Chem. Res. 2009,
42, 1954 – 1965; d) S. Romain, L. Vigara, A. Llobet, Acc. Chem.
Res. 2009, 42, 1944 – 1953; e) Y. Xu, T. ꢀkermark, V. Gyollai, D.
Zou, L. Eriksson, L. Duan, R. Zhang, B. ꢀkermark, L. Sun,
Inorg. Chem. 2009, 48, 2717 – 2719; f) L. Duan, F. Bozoglian, S.
Mandal, B. Stewart, T. Privalov, A. Llobet, L. Sun, Nat. Chem.
2012, 4, 418 – 423; g) M. D. Kꢁrkꢁs, T. ꢀkermark, H. Chen, J.
Sun, B. ꢀkermark, Angew. Chem. Int. Ed. 2013, 52, 4189 – 4193;
Angew. Chem. 2013, 125, 4283 – 4287; h) Y. Pushkar, D. Moon-
shiram, V. Purohit, L. Yan, I. Alperovich, J. Am. Chem. Soc.
2014, 136, 11938 – 11945; i) L. Duan, L. Wang, F. Li, F. Li, L. Sun,
Acc. Chem. Res. 2015, 48, 2084 – 2096; j) L. Tong, R. P. Thummel,
Chem. Sci. 2016, 7, 6591 – 6603; k) Y. Xie, D. W. Shaffer, A.
Lewandowska-Andralojc, D. J. Szalda, J. J. Concepcion, Angew.
Chem. Int. Ed. 2016, 55, 8067 – 8071; Angew. Chem. 2016, 128,
8199 – 8203; l) H. N. Kagalwala, L. Tong, R. Zong, L. Kohler,
M. S. G. Ahlquist, T. Fan, K. J. Gagnon, R. P. Thummel, ACS
Catal. 2017, 7, 2607 – 2615; m) R. Matheu, M. Z. Ertem, C.
Gimbert-Surinach, X. Sala, A. Llobet, Chem. Rev. 2019, 119,
3453 – 3471; n) J. Shi, Y.-H. Guo, F. Xie, Q.-F. Chen, M.-T.
III
III
ꢀ
ꢀ ꢀ
ꢀ
initial diiron(III) species, HO Fe O Fe OH, was oxi-
IV
IV
=
ꢀ ꢀ
=
IV
dized to O Fe O Fe O by losing two electrons and two
IV
=
ꢀ ꢀ
=
ꢀ
protons. The formed O Fe O Fe O can trigger the O O
bond formation via intramolecular oxo/oxo coupling, or it can
IV
=
ꢀ
be further oxidized to a higher valent intermediate O Fe
V
ꢀ
=
ꢀ
O Fe O, and then form an O O bond via intramolecular
oxo/oxyl coupling. What needs to be pointed out here is that
ꢀ
the O O bond formation requires the anti- to syn- config-
=
uration conversion between two Fe O moieties. This is
consistent with the results observed through the regulation
of the ligand skeleton among 1, 2 and 3. After the O O bond
formation, the obtained Fe2O2 intermediate could be further
oxidized to release the O2 to close the catalytic cycle under
the catalytic conditions.
ꢀ
Conclusion
ꢀ
Inspired by the O O bond breaking in soluble methane
monooxygenase (sMMO) and its diiron models, we have
ꢀ
investigated the possibility of diiron catalyzed O O bond
formation. Three diiron(III) complexes were intentionally
elected as water oxidation catalysts, in which [Fe2(m-O)-
(OH2)2(TPA)2]4+ (1) and [Fe2(m-O)(OH2)2(6-HPA)]4+ (2) are
active for catalyzing water oxidation. The suggested mecha-
ꢀ
nism is displayed in Figure 6. O O bond formation was
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Zhang, Angew. Chem. Int. Ed. 2020, 59, 4000 – 4008; Angew.
Chem. 2020, 132, 4029 – 4037.
Angew. Chem. 2014, 126, 12422 – 12426; e) T. Zhang, C. Wang, S.
Liu, J.-L. Wang, W. Lin, J. Am. Chem. Soc. 2014, 136, 273 – 281;
f) P. Garrido-Barros, I. Funes-Ardoiz, S. Drouet, J. Benet-
Buchholz, F. Maseras, A. Llobet, J. Am. Chem. Soc. 2015, 137,
6758 – 6761; g) X.-J. Su, M. Gao, L. Jiao, R.-Z. Liao, P. E. M.
Siegbahn, J.-P. Cheng, M.-T. Zhang, Angew. Chem. Int. Ed. 2015,
54, 4909 – 4914; Angew. Chem. 2015, 127, 4991 – 4996; h) Q.-Q.
Hu, X.-J. Su, M.-T. Zhang, Inorg. Chem. 2018, 57, 10481 – 10484;
i) X.-J. Su, C. Zheng, Q.-Q. Hu, H.-Y. Du, R.-Z. Liao, M.-T.
Zhang, Dalton Trans. 2018, 47, 8670 – 8675.
[3] a) N. D. McDaniel, F. J. Coughlin, L. L. Tinker, S. Bernhard, J.
Am. Chem. Soc. 2008, 130, 210 – 217; b) J. F. Hull, D. Balcells,
J. D. Blakemore, C. D. Incarvito, O. Eisenstein, G. W. Brudvig,
R. H. Crabtree, J. Am. Chem. Soc. 2009, 131, 8730 – 8731; c) J. D.
Blakemore, N. D. Schley, D. Balcells, J. F. Hull, G. W. Olack,
C. D. Incarvito, O. Eisenstein, G. W. Brudvig, R. H. Crabtree, J.
Am. Chem. Soc. 2010, 132, 16017 – 16029; d) R. Lalrempuia,
N. D. McDaniel, H. Mꢂller-Bunz, S. Bernhard, M. Albrecht,
Angew. Chem. Int. Ed. 2010, 49, 9765 – 9768; Angew. Chem.
2010, 122, 9959 – 9962; e) A. Lewandowska-Andralojc, D. E.
Polyansky, C.-H. Wang, W.-H. Wang, Y. Himeda, E. Fujita, Phys.
Chem. Chem. Phys. 2014, 16, 11976 – 11987; f) J. A. Woods, R.
Lalrempuia, A. Petronilho, N. D. McDaniel, H. Mꢂller-Bunz, M.
Albrecht, S. Bernhard, Energy Environ. Sci. 2014, 7, 2316 – 2328;
g) J. M. Thomsen, D. L. Huang, R. H. Crabtree, G. W. Brudvig,
Dalton Trans. 2015, 44, 12452 – 12472.
[4] a) Y. Naruta, M. Sasayama, T. Sasaki, Angew. Chem. Int. Ed.
Engl. 1994, 33, 1839 – 1841; Angew. Chem. 1994, 106, 1964 – 1965;
b) J. Limburg, J. S. Vrettos, L. M. Liable-Sands, A. L. Rheingold,
R. H. Crabtree, G. W. Brudvig, Science 1999, 283, 1524 – 1527;
c) J. Limburg, J. S. Vrettos, H. Chen, J. C. de Paula, R. H.
Crabtree, G. W. Brudvig, J. Am. Chem. Soc. 2001, 123, 423 – 430;
d) Y. Shimazaki, T. Nagano, H. Takesue, B.-H. Ye, F. Tani, Y.
Naruta, Angew. Chem. Int. Ed. 2004, 43, 98 – 100; Angew. Chem.
2004, 116, 100 – 102; e) Y. Gao, T. ꢀkermark, J. Liu, L. Sun, B.
ꢀkermark, J. Am. Chem. Soc. 2009, 131, 8726 – 8727; f) E. A.
Karlsson, B.-L. Lee, T. ꢀkermark, E. V. Johnston, M. D. Kꢁrkꢁs,
J. Sun, ꢃ. Hansson, J.-E. Bꢁckvall, B. ꢀkermark, Angew. Chem.
Int. Ed. 2011, 50, 11715 – 11718; Angew. Chem. 2011, 123, 11919 –
11922.
[5] a) M. W. Kanan, D. G. Nocera, Science 2008, 321, 1072 – 1075;
b) D. K. Dogutan, R. McGuire, Jr., D. G. Nocera, J. Am. Chem.
Soc. 2011, 133, 9178 – 9180; c) D. J. Wasylenko, C. Ganesamoor-
thy, J. Borau-Garcia, C. P. Berlinguette, Chem. Commun. 2011,
47, 4249 – 4251; d) D. J. Wasylenko, R. D. Palmer, E. Schott, C. P.
Berlinguette, Chem. Commun. 2012, 48, 2107 – 2109; e) T.
Nakazono, A. R. Parent, K. Sakai, Chem. Commun. 2013, 49,
6325 – 6327; f) H.-Y. Wang, E. Mijangos, S. Ott, A. Thapper,
Angew. Chem. Int. Ed. 2014, 53, 14499 – 14502; Angew. Chem.
2014, 126, 14727 – 14730; g) T. Nakazono, A. R. Parent, K. Sakai,
Chem. Eur. J. 2015, 21, 6723 – 6726; h) A. M. Ullman, C. N.
Brodsky, N. Li, S.-L. Zheng, D. G. Nocera, J. Am. Chem. Soc.
2016, 138, 4229 – 4236; i) H.-Y. Du, S.-C. Chen, X.-J. Su, L. Jiao,
M.-T. Zhang, J. Am. Chem. Soc. 2018, 140, 1557 – 1565.
[6] a) W. C. Ellis, N. D. McDaniel, S. Bernhard, T. J. Collins, J. Am.
Chem. Soc. 2010, 132, 10990 – 10991; b) J. L. Fillol, Z. Codolꢄ, I.
Garcia-Bosch, L. Gꢅmez, J. J. Pla, M. Costas, Nat. Chem. 2011, 3,
807 – 813; c) M. K. Coggins, M.-T. Zhang, A. K. Vannucci, C. J.
Dares, T. J. Meyer, J. Am. Chem. Soc. 2014, 136, 5531 – 5534;
d) C. Panda, J. Debgupta, D. Dꢆaz Dꢆaz, K. K. Singh, S.
Sen Gupta, B. B. Dhar, J. Am. Chem. Soc. 2014, 136, 12273 –
12282; e) Z. Codolꢄ, L. Gꢅmez, S. T. Kleespies, L. Que, Jr., M.
Costas, J. Lloret-Fillol, Nat. Commun. 2015, 6, 5865.
[9] Z. Codolꢄ, I. Gamba, F. AcuÇa-Parꢇs, C. Casadevall, M.
Clꢇmancey, J.-M. Latour, J. M. Luis, J. Lloret-Fillol, M. Costas,
J. Am. Chem. Soc. 2019, 141, 323 – 333.
[10] a) M. M. Najafpour, A. N. Moghaddam, D. J. Sedigh, M. Hoły-
´
nska, Catal. Sci. Technol. 2014, 4, 30 – 33; b) A. R. Parent, T.
Nakazono, S. Lin, S. Utsunomiya, K. Sakai, Dalton Trans. 2014,
43, 12501 – 12513; c) L. D. Wickramasinghe, R. Zhou, R. Zong,
P. Vo, K. J. Gagnon, R. P. Thummel, J. Am. Chem. Soc. 2015, 137,
13260 – 13263; d) B. Das, B.-L. Lee, E. A. Karlsson, T. ꢀker-
mark, A. Shatskiy, S. Demeshko, R.-Z. Liao, T. M. Laine, M.
Haukka, E. Zeglio, A. F. Abdel-Magied, P. E. M. Siegbahn, F.
Meyer, M. D. Kꢁrkꢁs, E. V. Johnston, E. Nordlander, B. ꢀker-
mark, Dalton Trans. 2016, 45, 13289 – 13293; e) M. Okamura, M.
Kondo, R. Kuga, Y. Kurashige, T. Yanai, S. Hayami, V. K. K.
Praneeth, M. Yoshida, K. Yoneda, S. Kawata, S. Masaoka,
Nature 2016, 530, 465 – 468.
[11] D. Wang, L. Que, Jr., Chem. Commun. 2013, 49, 10682 – 10684.
[12] R.-Z. Liao, S. Masaoka, P. E. M. Siegbahn, ACS Catal. 2018, 8,
11671 – 11678.
[13] a) G. Chen, L. Chen, S.-M. Ng, W.-L. Man, T.-C. Lau, Angew.
Chem. Int. Ed. 2013, 52, 1789 – 1791; Angew. Chem. 2013, 125,
1833 – 1835; b) D. Hong, S. Mandal, Y. Yamada, Y.-M. Lee, W.
Nam, A. Llobet, S. Fukuzumi, Inorg. Chem. 2013, 52, 9522 –
9531; c) M. M. Najafpour, R. Safdari, F. Ebrahimi, P. Rafighi, R.
Bagheri, Dalton Trans. 2016, 45, 2618 – 2623; d) P. Pelosin, M.
Gil-Sepulcre, P. Garrido-Barros, D. Moonshiram, J. Benet-
Buchholz, C. Gimbert-SuriÇach, A. Llobet, iScience 2020, 23,
101378; e) M. Puri, L. Que, Acc. Chem. Res. 2015, 48, 2443 –
2452; f) J. Serrano-Plana, A. Aguinaco, R. Belda, E. Garcꢆa-
EspaÇa, M. G. Basallote, A. Company, M. Costas, Angew. Chem.
Int. Ed. 2016, 55, 6310 – 6314; Angew. Chem. 2016, 128, 6418 –
6422; g) S. Kal, L. Que, Angew. Chem. Int. Ed. 2019, 58, 8484 –
8488; Angew. Chem. 2019, 131, 8572 – 8576.
[14] A. J. Jasniewski, L. Que, Jr., Chem. Rev. 2018, 118, 2554 – 2592.
[15] a) H.-F. Hsu, Y. Dong, L. Shu, V. G. Young, L. Que, J. Am.
Chem. Soc. 1999, 121, 5230 – 5237; b) H. Zheng, S. J. Yoo, E.
Mꢂnck, L. Que, J. Am. Chem. Soc. 2000, 122, 3789 – 3790; c) G.
Xue, D. Wang, R. De Hont, A. T. Fiedler, X. Shan, E. Mꢂnck, L.
Que, Jr., Proc. Natl. Acad. Sci. USA 2007, 104, 20713 – 20718;
d) G. Xue, A. T. Fiedler, M. Martinho, E. Mꢂnck, L. Que, Proc.
Natl. Acad. Sci. USA 2008, 105, 20615 – 20620; e) D. Wang, E. R.
Farquhar, A. Stubna, E. Mꢂnck, L. Que, Nat. Chem. 2009, 1,
145 – 150; f) M. Kodera, Y. Kawahara, Y. Hitomi, T. Nomura, T.
Ogura, Y. Kobayashi, J. Am. Chem. Soc. 2012, 134, 13236 –
13239; g) C.-W. Chiang, S. T. Kleespies, H. D. Stout, K. K.
Meier, P.-Y. Li, E. L. Bominaar, L. Que, Jr., E. Mꢂnck, W.-Z.
Lee, J. Am. Chem. Soc. 2014, 136, 10846 – 10849; h) S. A. Stoian,
G. Xue, E. L. Bominaar, L. Que, Jr., E. Mꢂnck, J. Am. Chem.
Soc. 2014, 136, 1545 – 1558; i) M. Kodera, T. Tsuji, T. Yasunaga,
Y. Kawahara, T. Hirano, Y. Hitomi, T. Nomura, T. Ogura, Y.
Kobayashi, P. K. Sajith, Y. Shiota, K. Yoshizawa, Chem. Sci.
2014, 5, 2282 – 2292; j) M. Kodera, S. Ishiga, T. Tsuji, K. Sakurai,
Y. Hitomi, Y. Shiota, P. K. Sajith, K. Yoshizawa, K. Mieda, T.
Ogura, Chem. Eur. J. 2016, 22, 5924 – 5936.
[7] a) M. Zhang, M.-T. Zhang, C. Hou, Z.-F. Ke, T.-B. Lu, Angew.
Chem. Int. Ed. 2014, 53, 13042 – 13048; Angew. Chem. 2014, 126,
13258 – 13264; b) P. Garrido-Barros, S. Grau, S. Drouet, J. Benet-
Buchholz, C. Gimbert-SuriÇach, A. Llobet, ACS Catal. 2019, 9,
3936 – 3945; c) Y. Han, Y. Wu, W. Lai, R. Cao, Inorg. Chem.
2015, 54, 5604 – 5613.
[8] a) S. M. Barnett, K. I. Goldberg, J. M. Mayer, Nat. Chem. 2012,
4, 498 – 502; b) Z. Chen, T. J. Meyer, Angew. Chem. Int. Ed. 2013,
52, 700 – 703; Angew. Chem. 2013, 125, 728 – 731; c) M.-T. Zhang,
Z. Chen, P. Kang, T. J. Meyer, J. Am. Chem. Soc. 2013, 135,
2048 – 2051; d) M. K. Coggins, M.-T. Zhang, Z. Chen, N. Song,
T. J. Meyer, Angew. Chem. Int. Ed. 2014, 53, 12226 – 12230;
[16] M. Kodera, M. Itoh, K. Kano, T. Funabiki, M. Reglier, Angew.
Chem. Int. Ed. 2005, 44, 7104 – 7106; Angew. Chem. 2005, 117,
7266 – 7268.
&&&&
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!

Angewandte
Research Articles
Chemie
[17] Deposition Numbers 2000428 (for 1), 2000435 (for 2), and
[22] a) A. J. Bard, L. R. Faulkner, Electrochemical methods: funda-
2000443 (for 3) contain the supplementary crystallographic data
for this paper. These data are provided free of charge by the joint
Cambridge Crystallographic Data Centre and Fachinforma-
tionszentrum Karlsruhe Access Structures service www.ccdc.
cam.ac.uk/structures.
mentals and applications, Wiley, New York, 2001; b) The
calculated kcat. value of the complicated 4e water oxidation
catalytic process serves only as an estimate of the catalytic rate
relative to the simple electrochemical model of catalytic
regeneration of electroactive species in a following homoge-
neous reaction (ErCcat.).
[18] C. He, S. J. Lippard, Inorg. Chem. 2001, 40, 1414 – 1420.
[19] a) Y. Liu, R. Xiang, X. Du, Y. Ding, B. Ma, Chem. Commun.
2014, 50, 12779 – 12782; b) J. Sanders-Loehr, W. D. Wheeler,
A. K. Shiemke, B. A. Averill, T. M. Loehr, J. Am. Chem. Soc.
1989, 111, 8084 – 8093; c) S. Yan, D. D. Cox, L. L. Pearce, C.
Juarez-Garcia, L. Que, J. H. Zhang, C. J. OꢈConnor, Inorg.
Chem. 1989, 28, 2507 – 2509.
[23] a) J. J. Concepcion, J. W. Jurss, J. L. Templeton, T. J. Meyer, J.
Am. Chem. Soc. 2008, 130, 16462 – 16463; b) Z. Chen, J. J.
Concepcion, X. Hu, W. Yang, P. G. Hoertz, T. J. Meyer, Proc.
Natl. Acad. Sci. USA 2010, 107, 7225 – 7229.
[24] For computational detials, see the supporting information.
[25] L. Noodleman, J. Chem. Phys. 1981, 74, 5737 – 5743.
[26] Y.-Y. Li, C. Gimbert, A. Llobet, P. E. M. Siegbahn, R.-Z. Liao,
ChemSusChem 2019, 12, 1101 – 1110.
[20] R. C. Holz, T. E. Elgren, L. L. Pearce, J. H. Zhang, C. J.
OꢈConnor, L. Que, Inorg. Chem. 1993, 32, 5844 – 5850.
ꢀ
[21] According to UV-vis spectrums in Figure S7, the form of HO
III
FeIII O Fe OH keep consistent in both 0.1 M NaNO3 and
Manuscript received: January 3, 2021
Accepted manuscript online: March 26, 2021
ꢀ ꢀ
ꢀ
NaHCO3 solutions, indicating that HCO3 (or NO3ꢀ) does not
ꢀ
affect the structure of the catalyst.
Version of record online: && &&
,
&&&&
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
Research Articles
Water Oxidation
Three binuclear Fe complexes are tested
as molecular water oxidation catalysts.
ꢀ
H.-T. Zhang, X.-J. Su, F. Xie, R.-Z. Liao,*
O O bond formation can be achieved via
M.-T. Zhang*
&&&& — &&&&
intramolecular oxo–oxo interaction rather
than by water nucleophilic attack (WNA),
and the catalytic performance can be
effectively controlled by the framework of
the ligand.
Iron-Catalyzed Water Oxidation: O–O
Bond Formation via Intramolecular Oxo–
Oxo Interaction
&&&&
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 10
These are not the final page numbers!