"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for SelfArchiving."

FULL PAPER
Electrochemically-Driven Water Oxidation by a Highly Active
Ruthenium-Based Catalyst
Andrey Shatskiy,[a] Andrey A. Bardin,[a, b] Michael Oschmann,[a] Roc Matheu,[c] Jordi Benet-Buchholz,[c]
Lars Eriksson,[d] Markus D. Kärkäs,[e] Eric V. Johnston,[a, f] Carolina Gimbert-Suriñach,[c] Antoni
Llobet,*[c, g] Björn Åkermark*[a]
Abstract: Herein is described a highly active ruthenium-based water
oxidation catalyst [RuX(mcbp)(OHn)(py)2] (5, mcbp2− = 2,6-bis(1methyl-4-(carboxylate)benzimidazol-2-yl)pyridine; n = 2, 1, and 0 for
X = II, III, and IV, respectively), which can be generated in a mixture
of RuIII/RuIV states from either [RuII(mcbp)(py)2] (4II) or
[RuIII(Hmcbp)(py)2]2+ (4III). Complexes 4II and 4III were isolated and
characterized by single crystal X-ray analysis, NMR, UV-vis, FT-IR,
ESI-HRMS, EPR, and elemental analysis, and their redox properties
were studied in detail by electrochemical and spectroscopic methods.
Unlike for the parent catalyst [Ru(tda)(py)2] (1, tda2− = [2,2′:6′,2″terpyridine]-6,6″-dicarboxylate), for which full transformation to the
catalytically active species [RuIV(tda)(O)(py)2] (2) could not be
carried out — stoichiometric generation of the catalytically active
Ru-aqua complex 5 from 4II was achieved under mild conditions (pH
7.0) and short reaction times. The redox properties of the catalyst
[a]

[b]

[c]

[d]

[e]

[f]

[g]

Dr. A. Shatskiy, Dr. A. A. Bardin, M. Oschmann, Dr. E. V. Johnston,
Prof. B. Åkermark
Department of Organic Chemistry, Arrhenius Laboratory, Stockholm
University
Svante Arrhenius väg 16C, 10691 Stockholm (Sweden)
E-mail: bjorn.akermark@su.se
Dr. A. A. Bardin
Current address:
Institute of Problems of Chemical Physics, Russian Academy of
Sciences
Prospekt Akademika Semenova 1g, 142432 Chernogolovka,
Moscow Region (Russia)
Dr. R. Matheu, Dr. J. Benet-Buchholz, Dr. C. Gimbert-Suriñach,
Prof. A. Llobet
Institute of Chemical Research of Catalonia (ICIQ), Barcelona
Institute of Science and Technology (BIST)
Avinguda Països Catalans 16, 43007 Tarragona (Spain)
E-mail: allobet@iciq.cat
Assoc. Prof. L. Eriksson
Department of Materials and Environmental Chemistry, Arrhenius
Laboratory, Stockholm University
Svante Arrhenius väg 16C, 10691 Stockholm (Sweden)
Asst. Prof. M. D. Kärkäs
Department of Chemistry, Organic Chemistry, KTH Royal Institute of
Technology
Teknikringen 30, 10044 Stockholm (Sweden)
Dr. E. V. Johnston
Current address:
Sigrid Therapeutics AB
Sankt Göransgatan 159, 11217 Stockholm (Sweden)
Prof. A. Llobet
Departament de Química, Universitat Autònoma de Barcelona
Cerdanyola del Vallès, 08193 Barcelona (Spain)
Supporting Information and the ORCID identification numbers for
the authors of this article can be found under http://dx.doi.org/

were studied and its activity for electrocatalytic water oxidation was
evaluated, reaching TOFmax ≈ 40 000 s−1 at pH 9.0 (from the foot-ofthe-wave analysis, FOWA), which is comparable to the activity of the
state-of-the-art catalyst 2.

1. Introduction
Harvesting the energy of the Sun for the production of
renewable fuels is identified as one of the prominent solutions
for reducing the negative environmental effects caused by the
use of fossil fuels.[1] For successful implementation of this
process several challenges remain unsolved. One is the
development of an efficient and durable water oxidation catalyst
(WOC), which could promote the four-electron four-proton
oxidation of water into molecular oxygen (Eq. 1).
2H2O → O2 + 4H+ + 4e− (E° = 1.23 V vs. NHE; pH 0)

(1)

A number of heterogeneous and homogeneous (molecular)
WOCs have been developed over the past two decades, with
ruthenium-based molecular catalysts serving as the principal
model for understanding the water oxidation mechanistic
pathways.[2] As a result, a number of considerations for rational
design of high performance molecular WOCs have been
identified. These include: (1) the use of oxidatively stable
organic ligands, with the nitrogen-containing aromatic heterocycles being the most common; (2) ability of the catalyst to
undergo proton-coupled electron transfer (PCET), allowing
generation of the high oxidation states of the catalytically active
species at low potentials; (3) the use of redox non-innocent
ligands, which can increase the stability of the catalyst while
facilitating access to high formal oxidation states of the metal
center; (4) the use of ligands, which can promote formation of
seven-coordinated ruthenium-aqua species, thereby lowering
the potentials of the relevant redox couples; (5) introduction of
an internal base in the second coordination sphere of the metal
center, which can assist the O–O bond formation via water
nucleophilic attack (WNA) by simultaneous abstraction of a
proton from water (intramolecular atom-proton transfer, i-APT);
(6) for catalysts which form the O–O bond via radical coupling of
two metal-oxo units (I2M mechanism) the ligand should promote
the intermolecular reaction via hydrophobic or other interactions,
or combine the two metal-oxo moieties within one molecule for
efficient intramolecular reaction. Applying the above conside-

"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
previous work

This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self2
Archiving."
H O

FULL PAPER
N

N

N

RuII

N
O

O

O

N
N

O

O

N

2

N

N
N

RuIV O
O
O

O

N

N

N

RuII

N

N

N

N

[RuII(tda)(py)2]

[RuIV (O)(tda)(py)2]

[RuII(Mebimpy)(bpy)(H2O)]2+

1

2

3

current work

Figure 2. Schematic representation of the coordination mode of
[Ru(tda)(py)2]n+ (left) and [Ru(mcbp)(py)2]n+ (right) complexes at RuIII and RuIV
states. The geometries of the complexes are adopted from the previously
published crystal structures of complex 1 with direct extrapolation of the
geometry for the structure of [Ru(mcbp)(py)2]n+. Axial pyridine ligands are
omitted for clarity.

2

N

N
N

N

N

N

RuII

N

O

O

N

O

N

N

O

N

RuIV O
O

N

O
N

N

O

RuIII HO

N

N

N

N

O
O

N
N

O

O

N

[RuII(mcbp)(py)2]

[RuIII (Hmcbp)(py)2]2+

[RuIV (O)(mcbp)(py)2]

4II

4III

5IV

Figure 1. Structures of the previously synthesized ruthenium complexes (1,
2, and 3) and the complexes synthesized in the current work (4II, 4III and 5IV).

rations to the design of molecular WOCs, an increase in the
turnover frequency (TOF) by seven orders of magnitude and the
turnover number (TON) by three orders of magnitude has been
achieved over the past decade. However, the long-term stability
and operating potentials of the developed catalysts remains
unsatisfactory and a commercially viable molecular WOC is yet
to be designed.
Catalyst 1 ([Ru(tda)(py)2], tda2− = [2,2′:6′,2″-terpyridine]-6,6″dicarboxylate) has been recently developed by Llobet and coworkers, guided by most of the aforementioned design principles
(Figure 1).[3] Catalyst 2, the active form of catalyst 1,
demonstrated a record high activity for electrocatalytic water
oxidation (TOFmax = 50 000 s−1 at pH 10, using foot-of-the-wave
analysis, FOWA),[3a] as well as high longevity (TON = 1 000 000
under bulk electrolysis with a surface-immobilized catalyst).[3b]
Moreover, the catalyst demonstrated high efficiency for lightdriven water oxidation with the use of a Ru(bpy)3-type
photosensitizer in solution,[3c] as well as immobilized on hybrid
semiconductor surfaces.[3d,f] The remarkable performance of the
catalyst is attributed to the ability of the tda2− ligand to facilitate
formation of seven-coordinated RuIII, RuIV, and RuV species at
low overpotentials. Further, the dangling carboxylate group in
complex 2 acts as an internal base, opening low-energy i-APT
pathway during the rate-limiting O–O bond formation step.
However, a significant drawback of the catalyst was the
incomplete formation of the catalytically active Ru-aqua species
2 from the initial catalyst, rendering ⅔ of the used catalyst
inactive during catalysis, which also obscured the mechanistic
studies and decreased its performance when incorporated in
hybrid anodes.[3a,b,d]
Catalyst 3 represents a well-studied ruthenium-based WOC
developed by Meyer and co-workers (Figure 1).[4a,b] Although the

catalyst had a moderate activity for electrocatalytic water
oxidation, it served as a primary model for development of a
number of strategies for the coupling of WOCs to a
photosensitizer or the electrode surface[4c-h] and was subjected
to detailed mechanistic studies.[4i-n] The related catalyst
[RuII(Mebimpy)(pic)3]2+ (Mebimpy = 2,6-bis(1-methylbenzimidazol-2-yl)pyridine, pic = 4-picoline) has also been prepared by
Sun and co-workers; however, it demonstrated no activity for
electrocatalytic water oxidation at acidic or neutral pH.[5]
In the current work we sought to address the incomplete
activation of complex 1 by preparing a ligand related to H2tda, in
which two of the coordinating pyridine moieties are substituted
with 1-methylbenzimidazoles, similar to catalyst 3. In the
proposed catalyst 4 introduction of one more carbon between
the coordinating nitrogen and carboxy-group was expected to
cause significant structural changes for the complex at the RuIII
and RuIV oxidation states, where the original tda2− ligand
coordinates to the ruthenium center in a κ-N3O2 mode. Thus
five-membered rings in catalyst 1 would expand to highlydistorted six-membered rings in catalyst 4, assuming a κ-N3O2
coordination mode (Figure 2). This distortion was expected to
weaken the Ru–O bonds in the complex, allowing easier access
to the catalytically active aqua-species for catalyst 4 compared
to catalyst 1. Indeed, full conversion of the initial complex to the
Ru-aqua species could be carried out for complex 4, resulting in
a highly active water oxidation catalyst 5, reaching TOFmax
values comparable to those of catalyst 1. The structure of
catalyst 5 was proposed based on analogy to catalyst 2, and is
fully consistent with the obtained experimental results.
Herein, we describe the preparation and redox properties of
complex 4 in the RuII oxidation state (4II) and its one-electron
oxidized form 4III with an emphasis on formation of Ru-aqua
species at the RuIII and RuIV oxidation states of the complex.
Furthermore, stoichiometric generation of complex 5 (herein also
referred to as Ru-aqua species) during spectroelectrochemical
measurements and bulk electrolysis is described, followed by
analysis of its redox properties and catalytic activity for water
oxidation.

2. Results

"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for SelfArchiving."

FULL PAPER
2.1 Synthesis and Characterization
2−

The H2mcbp (mcbp
= 2,6-bis(1-methyl-4-(carboxylate)benzimidazol-2-yl)pyridine) ligand was synthesized via reductive
cyclization between 2,6-pyridinedicarbox-aldehyde and methyl
3-(methylamino)-2-nitrobenzoate followed by hydrolysis of the
formed ester. The RuII complex 4II ([RuII(mcbp)(py)2]) was then
obtained by a prolonged reaction between the H2mcbp ligand,
pyridine, and [Ru(DMSO)4Cl2] as ruthenium precursor in
EtOH:water in the presence of NEt3 as base (see the Supporting
Information for details). X-ray quality crystals of 4II were isolated
from the reaction mixture and the structure of the complex was
confirmed by 1D and 2D NMR, SC-XRD, ESI-HRMS, elemental
analysis, EPR, ATR-FTIR and UV-vis. Oxidation of RuII complex
4II by 1 equivalent of ceric ammonium nitrate (CAN) in the
presence of NaClO4 furnished the corresponding RuIII complex
4III ([RuIII(Hmcbp)(py)2](ClO4)2). The complex was isolated as
X-ray quality crystals by recrystallization from the solution in TFE
(2,2,2-trifluoroethanol) and aqueous 0.1 M TfOH containing
NaClO4, and the structure of the complex was confirmed by
1
H NMR, SC-XRD, elemental analysis, EPR, ATR-FTIR and UVvis. All attempts to synthesize the complex in the RuIV oxidation
state were unsuccessful and resulted in isolation of RuIII
complexes, such as 4III(OTf)2 ([RuIII(Hmcbp)(py)2](OTf)2), based
on SC-XRD analysis. The RuIII complex is presumably
generated by slow oxidation of TFE or MeOH co-solvent by the
initially formed RuIV complex.
A solution of catalyst 4II in DMSO-d6 displayed a diamagnetic
1
H NMR spectrum, as expected for a d6 RuII complex (Figure

S11, top).[3a,4b,5] Initially, NMR signals from multiple species were
observed, but a clear spectrum could be obtained upon addition
of D2SO4 (Figure S11, bottom). The integrity of the complex
composition under these conditions was confirmed by HPLCESI-MS. The 1H NMR spectrum of 4II suggests a highly
symmetric structure, indicative of a fast coordination/dissociation
of the carboxylate groups to the metal center. The combination
of 1H NMR, 13C NMR, 1H-1H COSY and HSQC experiments
allowed for full assignment of the NMR spectra (Figure S13).
The 1H NMR spectrum of complex 4III displayed broadened and
shifted peaks, characteristic of a paramagnetic d5 RuIII species
(Figure S18).[3a,6] Reduction of this RuIII complex to the RuII state
with ascorbic acid resulted in a 1H NMR spectrum identical to
that of complex 4II (Figure S19).

2.2 Single Crystal X-ray Analysis
X-ray structures of complexes 4II and 4III(OTf)2 are presented in
Figure 3. In 4II the equatorial mcbp2− ligand is coordinated to the
metal center in a κ-N3O fashion, while two pyridine ligands
occupy the axial positions. For this complex a distorted
octahedral geometry is observed with typical Ru–N and Ru–O
bond lengths (see Table S4 for the comparison of relevant bond
lengths and angles). For the isolated RuIII complex 4III symmetryrelated carboxylic groups were disordered into two positions
(Figure S52, see the Supporting Information for details), while
such a disorder was not observed in the crystal of 4III(OTf)2. The
structure of the latter complex was therefore used for further
discussion.
The equatorial tda2− ligand in the previously described
[RuIII(tda)(py)2]+ complex coordinates to ruthenium in a κ-N3O2
fashion. In contrast, the Hmcbp− ligand in 4III(OTf)2 adopts κ-N3O
coordination mode with one of the carboxy-groups being
protonated and stabilized by a water molecule via hydrogen
bonding, leading to a highly distorted octahedral geometry with a
large O–Ru–N outer equatorial angle of 125°. Such a geometry
is consistent with easier access to the catalytically-active
Ru-aqua species for complex 4 relative to complex 1, discussed
in the following sections. It is, however, unclear if such a low
symmetry structure is preserved in solution, particularly for fully
deprotonated complexes in the RuIII and RuIV oxidation states,
where either κ-N3O2 coordination mode of the mcbp2− ligand or
fast coordination/dissociation of the two carboxy-groups to the
metal center is to be expected.

2.3 Electrochemical and Spectroscopic Studies of
Complexes 4II and 4III at pH 2–7

Figure 3. ORTEP plots (ellipsoids at 50% probability) of complexes 4II and
4III(OTf)2. Hydrogen bonds are shown as black dashed lines and hydrogen
atoms of the carboxy-group and water molecule are shown as black hollow
circles. Other hydrogen atoms, solvent molecules and counter-ions are
omitted for clarity. Axial pyridine ligands are omitted in the lower structures.
Color codes: C — black, N — blue, O — red, Ru — magenta.

The electrochemical properties of the synthesized complexes
were mainly studied in phosphate buffered solutions (0.1 ionic
strength), using glassy carbon disk working electrode (d = 1 mm,
S = 7.85×10−3 cm2), platinum disk auxiliary electrode, and
mercury-mercurous
sulfate reference electrode
(MSE,
Hg/Hg2SO4, sat. K2SO4). All potentials were converted to
vs. NHE by adding 0.650 V to the potentials vs. MSE. The redox
properties of the complexes were studied by cyclic voltammetry

"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for SelfArchiving."

FULL PAPER

Figure 4. CV (black) and DPV (dashed red) of 0.1 mM solutions of complex
4II at pH 2.0, 5.0, and 7.0. Three CV cycles were recorded at 0.05 V s −1 scan
rate between 0.3 V and 1.5 V and the last cycle is pre-sented for clarity.

(CV), differential pulse voltammetry (DPV), coulometry
(controlled potential electrolysis, CPE), and spectroelectrochemistry (see the Supporting Information for details).
Furthermore, the isolated and in situ generated complexes were
analyzed under similar conditions by 1H NMR, UV-vis
spectroscopy, and EPR, recorded at 4 K with an X-band
spectrometer. Due to limited solubility of the complexes in
aqueous media the experiments were typically performed using
0.05–0.2 mM solutions.

2.3.1 Redox Properties of RuII Complex 4II
At pH 2.0, the CV of complex 4II displayed two quasi-reversible
one-electron waves at 0.767 V and 1.357 V, which are assigned
to the RuIII/II and RuIV/III couples, respectively (Figure 4). The
one-electron nature of the couples is confirmed by the peak
separation in the CV (ΔEp, 0.059 V for RuIII/II and 0.068 V for
RuIV/III), peak width at half-height in DPV (W1/2, 0.096 V for RuIII/II
and 0.094 V for RuIV/III),[7] and CPE for the RuIII/II couple (transfer
of 1.0 electrons upon CPE at 1.05 V). Spectroelectrochemical
measurements at pH 2.0 revealed the disappearance of the
typical MLCT band for RuII complexes (λmax = 495 nm for 4II)
upon the RuII→RuIII transition (Figures S42 and S43a).[8]
At higher potentials, a transition to a RuIV species was observed
with two isosbestic points at 313 nm and 399 nm (Figure S43b)
and a good charge balance with the RuII→RuIII transition (Figure
S42). During the reverse CV scan, the UV-vis spectra of both
RuIII and RuII species were replenished in their initial form,
demonstrating reversibility of the involved transformations on the
experiment timescale (Figures S43c and S43d).[9] Interestingly,
the potential of the RuIII/II couple of 4II in an aprotic solvent

(MeCN) was smaller than in aqueous solutions by ca. 0.3 V, but
gradually shifted to the value observed in aqueous solutions
upon addition of water (Figures S39 and S40), highlighting the
importance of the hydrogen bonding interactions in the complex.
Upon increasing the pH, the potentials of the RuIII/II and
RuIV/III couples displayed a pH dependence of −0.019 and
−0.030 V per pH unit over pH 3–4 and pH 2–3, respectively,
while −0.059·m V per pH unit is expected for a one-electron
processes involving transfer of m protons.[10] Such behavior
can be rationalized assuming close pKa values for complex 4
at the RuII, RuIII, and RuIV oxidation states, involving
protonation/deprotonation of one of the carboxylate groups.
Indeed, simulation and fitting of the Pourbaix diagram to the
experimental values according to Eq. S1 and S2 reveals that at
low pH complex 4 at the RuII, RuIII, and RuIV states is protonated
with the pKa = 4.0, 3.3 and 2.1, respectively (Figure S32, see the
Supporting Information for details). The change in protonation
state for the RuIII/II couple is then likely to be responsible for
splitting of initially quasi-reversible RuIII/II wave into a pair of
close-lying irreversible waves (ΔEp = 0.110 V at pH 7.0, 0.05
V s−1 scan rate) and formation of an ErCrErCr square scheme
upon increasing the pH (Figures 4 and S31).[11] Accordingly,
at high scan rates the chemical reactions following/preceding the
electron transfer can be outrun and a close-lying pair of two
quasi-reversible RuIII/II couples is observed (Figure S33).
This implies the presence of two isomers of the RuII complex at
elevated pH. As the potentials for the Pourbaix diagram of RuIII/II
transition were derived from Epc of the electrochemically
irreversible RuIII/II couple, the obtained pKa value represents
the pKaII of an isomer of complex 4, dominant at the RuIII state.
The pKaII value of an isomer of 4 dominant at RuII oxidation state
was then obtained by spectrophotometric pH titration and fitting
of the experimental data to Eq. S3, resulting in a slightly different
pKa = 3.1 (Figure S27). At pH > 7 the electrochemical behavior
of RuIII/II couple becomes more complex due to formation of the
corresponding Ru-aqua species (vide infra). Therefore, the
redox behavior of the complex at high pH was only studied for
the electrochemically generated Ru-aqua species, as described
in Section 2.4.2.
At pH > 3.5 a new wave for complex 4II appears in the DPV
at high potentials (E > 1.4 V) while an electrocatalytic wave
starts growing in the CV (Figure S31). Interestingly, at pH 7.0
the catalytic wave displayed an inverse loop and the catalytic
current was growing upon repetitive cycling, which could indicate
that the initially inactive form of complex 4 is transformed into an
active catalyst at high potentials (Figure S35, left). Moreover,
formation of new species with Epa = 0.95 V could be observed
upon repetitive cycling (Figure S35, right). This new species was
assigned to the catalytically active Ru-aqua complex, generated
from complex 4 when oxidized to the RuIV state (vide infra).

"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for SelfArchiving."

FULL PAPER

Figure 5. Left: Evolution of the UV-vis spectrum of 0.05 mM solution of complex 4III upon pH jump from pH 2.0 to 7.0 over 250 min (0 min — red line, 250 min —
blue line) and a UV-vis spectrum of 0.05 mM solution of complex 4II at pH 7.0. Right: UV-vis spectra of complexes 4II (green line), 5IV (orange line, obtained from
spectroelectrochemical measurements, see Section 2.4.1), their sum (dashed pink), and a UV-vis spectrum of complex 4III at 4 h after the pH jump (blue line). For
the latter plot the absorbance of complex 4II was normalized for 0.025 mM concentration and absorbance of complex 5IV was normalized for 0.0125 mM
concentration.

2.3.2 Redox Properties of RuIII Complex 4III
As expected, at pH 2.0 the electrochemical behavior of complex
4III was identical to that of complex 4II (Figure S36).
The oxidation state of the complex was confirmed by CPE at
0.450 V, upon which transfer of 0.82 electrons was observed.
The UV-vis spectrum of complex 4III at this pH was identical to
the UV-vis spectrum of the RuIII species generated from 4II
during spectroelectrochemical experiments at pH 2.0 (Figure
S26). 1H NMR spectrum of 4III displayed broadened signals,
typical for paramagnetic d5 RuIII species (Figure S18). Moreover,
a paramagnetic EPR spectrum was observed for 4III (Figure 9),
further confirming the oxidation state of the complex.
Surprisingly, at higher pH the RuIII complex 4III is unstable
and forms RuII species, as evident from the UV-vis spectra of the
complex recorded over 4 h upon pH jump from pH 2.0 to 7.0
(Figure 5). Upon the pH jump an MLCT band with λmax = 479 nm
appeared over time along with changing ligand-based transition
bands, resulting in a UV-vis spectrum closely resembling that of
RuII complex 4II at pH 7.0, but with twice as low concentration
(Figure 5, left). The kinetics of the transformation, as monitored
by the change in the absorbance at 479 nm, displayed a
complex multi-exponential character, precluding reliable fitting to
a theoretical model (Figure S28). The changes in redox behavior
of 4III upon the pH jump were subsequently monitored
electrochemically (Figure S37), displaying formation of the Ruaqua species (see Section 2.4), which was also observed for
complex 4II upon repetitive CV cycling reaching potentials of
electrocatalytic water oxidation (Section 2.3.1) or after CPE at
1.4 V (Section 2.4.2). The disappearance of the paramagnetic
RuIII species after the pH jump was also confirmed by EPR
(Figure S30).
The above observations suggest a disproportionation type of
reaction between the RuIII species to form a RuII and a RuIV
species, concomitant with formation of the Ru-aqua complex.
The direct disproportionation of RuIII species to RuII and RuIV is
unlikely, given the large difference between the redox potentials
of the RuIII/II and RuIV/III couples with ΔE1/2 ≈ 0.6 V. Therefore, we
propose that water slowly reacts with RuIII complex 4III to form

RuIII–OH species (Eq. 2). In the RuIII and RuIII–OH mixture the
latter is easier to oxidize, since Epa(RuIV=O/RuIII–OH) = 0.95 V
(see Section 2.4.2) is significantly lower than E1/2(RuIV/III) =
1.32 V, which results in formation of RuIV=O and RuII species
(Eq. 3). The reverse transformation RuII + RuIV=O → RuIII +
RuIII–OH is then thermodynamically possible, since Epc(RuIII/II) ≈
Epc(RuIV=O/RuIII–OH) ≈ 0.7 V, but the overall equilibrium is
shifted to the products of the forward reaction due to the
indicated difference in the Epa values. The shifted equilibrium in
the solution electron transfer reaction (Eq. 3) then provides the
driving force for the otherwise unfavorable coordination of water
to the RuIII species (Eq. 2). Further, the sum of the UV-vis
spectra of RuII complex 4II and RuIV=O complex 5IV generated
spectroelectrochemically (see Section 2.4.1) resulted in a
spectrum closely overlapping with the final UV-vis spectrum of
4III at 250 min after pH jump (Figure 3, right). In this experiment,
however, absorbance of the 5IV species had to be normalized to
twice as low concentration compared to species 4II, indicating
that stoichiometry of Eq. 2 and 3 is affected by some sidereaction, possibly slow oxidation of water by the RuIV=O species,
as was observed previously for other WOCs.[3a,12] It is important
to point out that even though the described transformations
include formation of the Ru-aqua species already at the RuIII
state, the reaction is sufficiently slow on the CV timescale,
consistent with the chemical reversibility of the RuIII/II couple of 4II
at pH 7.0 observed during spectroelectrochemical measurements. The described transformations are extremely slow at low
pH, which is reasonable, given that both of the reactions require
abstraction of a proton.
RuIII + H2O ⇄ RuIII–OH + H+

(2)

Ru + Ru –OH → Ru + Ru =O + H
III

III

II

IV

+

(3)

2.4 Redox Properties and Catalytic Activity of the Ruaqua Complex 5 Generated from Complex 4
2.4.1 Stoichiometric Generation of a Catalytically
Active Ru-aqua Complex 5
Spectroelectrochemical measurements of RuII complex 4II
between 0.52 and 1.06 V at pH 7.0 demonstrated full chemical

"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for SelfArchiving."

FULL PAPER
reversibility of the RuIII/II couple (Figure S44), similar to the
measurements made at pH 2.0 (Figures S42 and S43), with a
RuIII/II cathodic wave observed at ca. 0.70 V. When reaching
potentials of up to 1.40 V (where the RuIII→RuIV transition can
occur) the reverse CV scan also displayed a cathodic wave at a
similar potential; however, the change in the UV-vis absorption
during the experiment at this potential range was dramatically
different (Figure 6). Upon the CV scan from 1.00 to 1.40 V the
UV-vis spectra displayed formation of RuIV species, but during
the reverse scan from 1.40 to 1.00 V the RuIII species were not
replenished. Instead, the UV-vis spectum continued to change in
the same manner as during the RuIII→RuIV transition. We
attribute this behavior to a water association reaction, which
follows the RuIII→RuIV transition (Eq. 4 and 5). The RuIV→RuIII
transition of the formed Ru-aqua species is then observed at
significantly lower potential of ca. 0.7 V, consistent with better
stabilization of the highly-oxidized metal center by the negatively
charged hydroxo/oxo ligand (Eq. 6). A slight increase in the UVvis absorbance at 479 nm, attributed to formation of non-aqua
RuII species, was also observed by UV-vis on the reverse CV
scan (Figure 6e). This is presumably because on the timescale

of the spectroelectrochemical experiment not all of the formed
RuIV species could be transformed into RuIV=O.
RuIII − e− → RuIV

(4)

RuIV + H2O → RuIV=O + 2H+

(5)

RuIV=O + e− + H+ → RuIII–OH

(6)

For further catalytic studies, the Ru-aqua species 5 was
generated stoichiometrically at pH 7.0 by CPE of complex 4II at
1.40 V (Figure S38). During electrolysis the non-aqua RuII
complex 4II is transformed into RuIV and then RuIV=O species,
which can catalyze water oxidation at the applied potential.
Accordingly, the current increased over time due to
accumulation of the active species and a decrease in pH of the
solution was observed. The electrolysis was terminated when
the current reached a plateau (pH of the final solution was then
adjusted back to 7.0 prior to further experiments).

2.4.2 Evaluation of the Catalytically Active Ru-aqua
Complex 5 Generated from Complex 4

"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for SelfArchiving."

FULL PAPER

Figure 6. Spectroelectrochemical measurements of ca. 0.4 mM solution of complex 4II at pH 7.0 using OTTLE cell (see the Supporting Information for details).
a) CV of complex 4II recorded at 0.002 V s −1 scan rate between 0.5 V and 1.4 V. b) Change in UV-vis absorbance during the anodic sweep from 0.5 to 1.0 V.
c) Change in UV-vis absorbance during the anodic sweep from 1.0 to 1.4 V. d) Change in UV-vis absorbance during the cathodic sweep from 1.4 to 1.0 V.
e) Change in UV-vis absorbance during the cathodic sweep from 1.0 to 0.5 V. Vertical arrows indicate the direction of change in UV-vis absorbance during CV
scan.

Figure 7. CV of 0.2 mM solution of complex 4II recorded at 0.05 V s−1 scan
rate at pH 7.0 (before CPE) and CV of complex 5 (after CPE) generated from
4II by CPE at 1.40 V at pH 7.0.

A comparison of the CV of the solution of complex 4II before and

after electrolysis revealed full transformation of the complex into
the new Ru-aqua species 5 (Figure 7). A new irreversible couple
appeared at the potentials close to the RuIII/II couple of complex
4, with Epa = 0.952 V and Epc = 0.709 V. Based on
spectroelectrochemical measurements this couple was assigned
to the RuIV/III transition, with λmax = 307 nm for RuIII and λmax =
316 nm for RuIV species (Figure 8). Two other redox couples of
smaller magnitude with E1/2 = 0.202 and 0.518 V were assigned
to two RuIII/II couples of different isomers of the Ru-aqua species.
While the couple at 0.202 V is close to being quasi-reversible
(ΔEp = 0.071 V) the couple at 0.518 V represents an ErCrErCr
square scheme (ΔEp = 0.176 V, Epa = 0.606 V, Epc = 0.430 V).
Spectroelectrochemical characterization of these couples was
unfortunately prohibited by the narrow working potential window
of the employed spectroelectrochemical cell. Importantly, the
new species have the same potentials as those emerging upon
repetitive cycling of complex 4II to the potentials of electrocatalytic water oxidation (Section 2.3.1) and upon pH jump from
2.0 to 7.0 for complex 4III (Section 2.3.2), confirming formation of
the Ru-aqua species under these conditions.
The pH of the generated solution with Ru-aqua species was
adjusted to variable values, and CV and DPV data was used to
construct the Pourbaix diagram (Figure 12, left). The observed
RuIII/II and RuIV/III couples displayed a distinct pH-dependence
close to −0.059 V per pH unit, although perturbed by coupled
chemical reactions and close-lying pKa values of the RuII, RuIII,
and RuIV complexes. As the carboxylate groups of the ligand are
deprotonated at pH > 4, proton transfer was associated with
the aqua ligand and the observed couples were assigned to

"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for SelfArchiving."

FULL PAPER

Figure 8. Spectroelectrochemical measurements of 0.2 mM solution of complex 5 generated from complex 4II by CPE at pH 7.0 using OTTLE cell (see the
Supporting Information for details). a) CV of complex 5 recorded at 0.002 V s−1 scan rate between 0.61 V and 1.11 V. b) Change in UV-vis absorbance during the
anodic sweep from 0.61 to 1.11 V. c) Change in UV-vis absorbance during the cathodic sweep from 1.11 to 0.61 V. Vertical arrows indicate the direction of
change in UV-vis absorbance during CV scan.

RuIII–OH/RuII–OH2 and RuIV=O/RuIII–OH transitions. The RuV/IV
couple displayed a pH-independence at pH 5–10, whereas at
lower pH a slope of −0.059 V per pH unit was observed due to
formation of a RuIV=OH species with pKa = 5.1.
Complex 4II and the electrochemically generated complex 5
were also studied by 1H NMR in an aqueous pD 7.0 phosphate
buffer solution (Figures S14 and S20). In analogy to the 1H NMR
spectrum of 4II obtained in DMSO-d6 upon addition of D2SO4, the
1
H NMR spectrum of 4II in aqueous solution was fully assigned,
confirming the integrity of the complex in water. After generation
of complex 5 by CPE the 1H NMR spectrum displayed shifted
and broadened signals due to the presence of the RuIII–OH
species, as observed by EPR. However, when the obtained
species was reduced to the RuII state by excess ascorbic acid a
diamagnetic spectrum identical to that of the initial 1H NMR
spectrum of RuII complex 4II was observed. This suggests that
on the timescale of the experiment (ca. 3 min between addition
of ascorbic acid and acquisition of the NMR spectrum) the
RuII–OH2 species are not stable and the water molecule
dissociates from the metal center. Additional peaks were also
observed in 1H NMR spectra after generation of the Ru-aqua
species and were still present upon reduction of the complex to

the RuII state. However, these contaminants did not correspond
to free pyridine (Figure S21), which is an expected product from
decomposition of catalyst 5.
Furthermore, the EPR spectrum of the Ru-aqua species 5
demonstrated a low intensity paramagnetic spectrum,
suggesting the presence of RuIII–OH species (Figure 9).
Importantly, the EPR spectrum of these species was drastically
different from the spectrum of the non-aqua RuIII species at the
same pH, suggesting that the generated Ru-aqua species is a
mixture of RuIII–OH and RuIV=O complexes.

2.4.3 Catalytic Activity of Complex 5
A CV of the electrochemically generated complex 5 up to 1.45 V
demonstrated a largely increased catalytic current compared to
the initial RuII complex 4II with an onset potential at ca. 1.25 V
(Figure 10). Under repetitive cycling the redox waves of the
Ru-aqua species were unchanged (except for a slight anodic
shift due to the local change in pH during water oxidation),
confirming that the catalytically active species was generated
stiochiometrically during CPE and are stable on the experiment
timescale.

Figure 10. Three CV cycles of a 0.2 mM solution of complex 4II at pH 7.0
before and after CPE at 1.40 V, recorded at 0.05 V s −1 scan rate. In the inset
only the last CV cycle is presented for clarity.

Figure 9. EPR spectra of complex 4II (green), 4 III (blue), and 5 (orange) at pH
7.0.

"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
600
This article maypH 9 used for non-commercial purposes in accordance with Wiley Terms and Conditions for Selfbe
Archiving."
pH 8

pH
FULL PAPER7
500

i/ip

400

solution before and after CPE remained unchanged, highlighting
stability of complex 5 under the studied catalytic conditions
(Figure S51).

1 i/ip

300
200

3 Discussion

100
0
-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

E (V vs. NHE)
Figure 11. CV of 0.2 mM solutions of catalyst 5 recorded at 0.1 V s −1 at pH
7.0, 8.0 and 9.0.

The catalytic activity of complex 5 for water oxidation was
evaluated using the foot-of-the-wave analysis (FOWA),13
assuming O–O bond formation through WNA as the rate
determining step (Figure 11 and Figures S45–S48, see the
Supporting Information for details). As a result, concentrationindependent TOFmax values of 1.4±0.4×103, 9.3±0.3×103, and
4.2±1.5×104 s–1 were observed at pH 7, 8, and 9, respectively,
while for the parent catalyst 2 TOFmax values of ca. 8×103,
2.5×104, and 5×104 s–1 were found at pH 7, 8, and 10. The
activities of the two catalysts are therefore similar given the
sensitivity of the method and much greater than the activities of
other WOCs operating via a WNA mechanism, as evident from
the catalytic Tafel plot (Figure S49). Furthermore, the catalytic
activity of complex 5 was evaluated on a longer timescale by
CPE at 1.45 V with monitoring of the evolved oxygen by a gasphase Clark electrode (Figure S50). After 30 min of electrolysis
98% faradaic efficiency was observed with TON = 9.3, based on
the total amount of the catalyst in the anode compartment. An
intrinsic TON of the catalyst under the employed bulk
electrolysis conditions was calculated by estimating the amount
of catalyst that participates in the catalysis at the electrode
surface (Eq. S7) and exceeded 4 million. The CV of the catalyst

Given the similarity of the catalytic activity of catalysts 2 and 5, it
is noteworthy to compare the redox properties of the complexes
in activated (2 and 5) and non-activated forms (1 and 4).
As evident from the Pourbaix diagrams of the complexes (Figure
12), both RuIII/II and RuIV/III couples for complex 4 are ca. 0.2 V
more positive than the corresponding couples for catalyst 1. This
is in line with the expected weakening of the Ru–O bonds in
complex 4 due to a change from stable five-membered rings in
catalyst 1 to highly strained six-membered rings in catalyst 4.
This is also supported by the X-ray crystal structures of the
complexes at the RuIII state, where the tda2− ligand in catalyst 1
adopts a κ-N3O2 coordination mode and the Hmcbp− ligand in
catalyst 4III coordinates to the ruthenium center in a κ-N3O mode,
providing an open site for coordination of water to the metal
center and formation of Ru-aqua species 5. The difference in the
potentials of RuIII/II and RuIV/III couples for the complexes could
also be ascribed to the difference in the electronic properties
between the 2,2′:6′,2″-terpyridine (tpy) ligand core in catalyst 1
and Mebimpy ligand core in catalyst 4. However, for the related
complexes [Ru(tpy)(py)3]2+ and [Ru(Mebimpy)(pic)3]2+ an
opposite trend in the redox potentials for the RuIII/II couple was
observed, with E1/2(RuIII/II) of [Ru(tpy)(py)3]2+ being ca. 0.25 V
more positive than E1/2(RuIII/II) of [Ru(Mebimpy)(pic)3]2+. The
latter supports that the elevated potentials for complex 4 are
indeed due to less efficient stabilization of the metal center at
high oxidation states by the more loosely bound carboxylate
ligands.
The elevated potential for the RuIII/II couple for complex 4

Figure 12. Pourbaix diagrams for complexes 5 (left) and 2 (right, simulated and extrapolated for pH < 3 using experimental data for pH 3–11 from Ref. [3a]).
Grey lines represent redox potentials of the non-aqua ruthenium species and red lines represent redox potentials of the aqua ruthenium species. The thermodynamic potential of water oxidation is shown with the blue line.

"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for SelfArchiving."

FULL PAPER

Scheme 1. Summary of the proposed transformations of the complexes
derived from 4, including catalyst activation and the catalytic cycle of water
oxidation. The mcbp2− ligand is represented with arcs and the axial pyridine
ligands are omitted for clarity. Dashed lines indicate simultaneously formed
and broken bonds.

together with the weakened Ru–O bonds at the RuIII oxidation
state allows formation of Ru-aqua species 5 already at the RuIII
oxidation state of 4, which has not been observed for catalyst 1.
The reaction is slow on a typical timescale of electrochemical
measurements but could be observed on longer timescale by
UV-vis for RuIII complex 4III. Due to the close-lying potentials of
the RuIV/III couple of Ru-aqua species 5 and the RuIII/II couple of
catalyst 4 the energetically unfavorable reaction RuIII + H2O →
RuIII–OH + H+ is driven to completion by the following
disproportionation reaction RuIII + RuIII–OH → RuII + RuIV=O +
H+, as described in Section 2.3.2. Furthermore, a large
difference in rates for the formation of the catalytically active
complexes 2 and 5 was also observed at the RuIV state of the
catalyst precursors. Complex 4 could be fully activated within ca.
4 min during the spectroelectrochemical measurements
reaching the RuIV state, while for complex 1 a 10 h electrolysis
was required to transform only ⅓ of the initial complex into the
active species 2 at pH 7. Importantly, the slow activation of
complex 1 limited performance of hybrid anodes based on 1 due
to decrease of catalyst loading in course of its activation.
For the activated catalysts 2 and 5 a striking similarity of the
pH-dependent redox potentials was observed, as evident from
their Pourbaix diagrams (Figure 12). Herein, it is important to
mention that due to incomplete activation of catalyst 1 a reliable
assignment of the newly formed redox transitions was extremely
challenging: an anodic wave at ca. 0.9 V and a cathodic wave at
ca. 0.7 V formed during catalyst activation at pH 7 appeared in
close proximity to the still present RuIII/II and RuIV/III couples of 1

and were assigned to the RuIV/III and RuIII/II transitions,
respectively. However, we believe that given the nearly identical
pH-dependent electrochemical response from catalyst 5, these
anodic and cathodic waves represent an electrochemically
irreversible RuIV/III couple, unambiguously identified for catalyst 5
by spectroelectrochemical measurements. Presumably, the
RuIII/II couple for the activated catalyst 2 is then nearly entirely
buried under the RuIII/II couple of the non-activated complex 1.
Given these observations, we propose that the catalytic
cycle for catalyst 5 should resemble the catalytic cycle
established for catalyst 2 with the help of DFT calculations
(Scheme 1). The electrochemically generated RuIV=O complex 5
undergoes one-electron oxidation to a RuV=O species, which is
responsible for the rate-limiting O–O bond formation step via
WNA. Importantly, the latter is facilitated by the simultaneous
transfer of a proton from water to one of the carboxylate groups
by an i-APT mechanism, significantly lowering the activation
energy of the reaction. The formed RuIII–OOH2 hydroperoxide
species undergoes fast one-electron/one-proton oxidation to
RuIV–OOH, while both of the hydroperoxide species are
stabilized by the hydrogen bonding to the dangling carboxylate
ligand. Oxygen release from the RuIV–OOH and coordination of
a new water molecule to ruthenium results in formation of
RuII–OH2 species. These species then undergo two oneelectron/one-proton oxidation steps, reforming the RuIV=O
complex and completing the catalytic cycle.

4 Conclusions
A new highly active water oxidation catalyst 5 was generated
from the precursor complex 4. The high activity and stability of
the catalyst was demonstrated and the redox behavior of the
Ru-non-aqua and aqua species was studied in detail. The newly
developed catalyst represents a significant improvement to the
state-of-the-art catalyst 1, which could not be activated fully,
prohibiting detailed mechanistic studies. On the other hand,
activation of catalyst 4 proceeds fully at mild conditions, resulting
in the active catalyst 5, reaching TOFmax ≈ 40 000 s−1 at pH 9.
Moreover, the developed benzimidazole-based mcbp2− ligand
provides a versatile and synthetically accessible framework for
the structure alterations by, for example, replacement of the
N-Me groups. Modification of these groups can then be used for
tuning steric and electronic properties of the ligand as well as
anchoring of the catalyst onto electrode surfaces.

Acknowledgements
B.Å. thanks Swedish Research Council (621-2013-4872),
Stiftelsen Olle Engkvist Byggmästare, and the Knut and Alice
Wallenberg Foundation (2011.0038) for the financial support.

"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for SelfArchiving."

FULL PAPER
A.L. thanks MINECO, FEDER and AGAUR for grants CTQ201680058-R, CTQ2015-73028-EXP, SEV 2013-0319, ENE201682025-REDT, CTQ2016-81923-REDC, and 2017-SGR-1631.
A.S. thanks Stiftelsen Längmanska Kulturfonden (BA17-1321)
for a PhD exchange scholarship. A.A.B. thanks Swedish Institute
(Visby Programme, 03572/2016) for a postdoctoral travel grant
and the Russian Foundation for Basic Research (18-03-01076).
Keywords: artificial photosynthesis • water oxidation •
ruthenium • molecular catalysis
[1]

[2]

[3]

[4]

a) N. S. Lewis, Science 2016, 351, 353–360; b) J. R. McKone, D. C.
Crans, C. Martin, J. Turner, A. R. Duggal, H. B. Gray, Inorg. Chem.
2016, 55, 9131–9143; c) O. Ellabban, H. Abu-Rub, F. Blaabjerg,
Renewable and Sustainable Energy Reviews 2014, 39, 748–764; d) N.
L. Panwar, S. C. Kaushik, S. Kothari, Renewable and Sustainable
Energy Reviews 2011, 15, 1513–1524; e) N. S. Lewis, D. G. Nocera,
Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 15729–15735.
a) P. Garrido-Barros, C. Gimbert-Suriñach, R. Matheu, X. Sala, A.
Llobet, Chem. Soc. Rev. 2017, 46, 6088–6098; b) I. Funes-Ardoiz, P.
Garrido-Barros, A. Llobet, F. Maseras, ACS Catal. 2017, 7, 1712–1719;
c) L. Tong, R. P. Thummel, Chem. Sci. 2016, 7, 6591–6603; d) M. D.
Kärkäs, B. Åkermark, Dalton Trans. 2016, 45, 14421–14461; e) M.
Yamamoto, K. Tanaka, ChemPlusChem 2016, 81, 1028–1044; f) X. Wu,
F. Li, B. Zhang, L. Sun, J. Photochem. Photobiol. C: Photochem. Rev.
2015, 25, 71–89; g) J. D. Blakemore, R. H. Crabtree, G. W. Brudvig,
Chem. Rev. 2015, 115, 12974–13005; h) D. L. Ashford, M. K. Gish, A.
K. Vannucci, M. K. Brennaman, J. L. Templeton, J. M. Papanikolas, T.
J. Meyer, Chem. Rev. 2015, 115, 13006−13049; i) M. D. Kärkäs, O.
Verho, E. V. Johnston, B. Åkermark, Chem. Rev. 2014, 114, 11863–
12001.
a) R. Matheu, M. Z. Ertem, J. Benet-Buchholz, E. Coronado, V. S.
Batista, X. Sala, A. Llobet, J. Am. Chem. Soc. 2015, 137, 10786–
10795; b) J. Creus, R. Matheu, I. Peñafiel, D. Moonshiram, P. Blondeau,
J. Benet-Buchholz, J. García-Antón, X. Sala, C. Godard, A. Llobet,
Angew. Chem. Int. Ed. 2016, 55, 15382–15386; c) L. Francàs, R.
Matheu, E. Pastor, A. Reynal, S. Berardi, X. Sala, A. Llobet, J. R.
Durrant, ACS Catal. 2017, 7, 5142–5150; d) R. Matheu, I. A. MorenoHernandez, X. Sala, H. B. Gray, B. S. Brunschwig, A. Llobet, N. S.
Lewis, J. Am. Chem. Soc. 2017, 139, 11345–11348; e) R. Matheu, M. Z.
Ertem, C. Gimbert-Suriñach, J. Benet-Buchholz, X. Sala, A. Llobet,
ACS Catal. 2017, 7, 6525–6532; f) S. Grau, S. Berardi, A. Moya, R.
Matheu, V. Cristino, J. José Vilatela, C. A. Bignozzi, S. Caramori, C.
Gimbert-Suriñach, A. Llobet, Sustainable Energy Fuels, 2018, 2, 1979–
1985.
a) Z. Chen, J. J. Concepcion, J. W. Jurss, T. J. Meyer, J. Am. Chem.
Soc. 2009, 131, 15580–15581; b) J. J. Concepcion, J. W. Jurss, M. R.
Norris, Z. Chen, J. L. Templeton, T. J. Meyer, Inorg. Chem. 2010, 49,
1277–1279; c) D. M. Ryan, M. K. Coggins, J. J. Concepcion, D. L.
Ashford, Z. Fang, L. Alibabaei, D. Ma, T. J. Meyer, M. L. Waters, Inorg.
Chem. 2014, 53, 8120–8128; d) D. L. Ashford, A. M. Lapides, A. K.
Vannucci, K. Hanson, D. A. Torelli, D. P. Harrison, J. L. Templeton, T. J.
Meyer, J. Am. Chem. Soc. 2014, 136, 6578–6581; e) S. E. Bettis, K.

[5]
[6]

[7]
[8]

[9]

[10]

[11]

[12]

[13]

Hanson, L. Wang, M. K. Gish, J. J. Concepcion, Z. Fang, T. J. Meyer, J.
M. Papanikolas, J. Phys. Chem. A 2014, 118, 10301–10308; f) S. E.
Bettis, D. M. Ryan, M. K. Gish, L. Alibabaei, T. J. Meyer, M. L. Waters,
J. M. Papanikolas, J. Phys. Chem. C 2014, 118, 6029−6037; g) A. M.
Lapides, B. D. Sherman, M. K. Brennaman, C. J. Dares, K. R. Skinner,
J. L. Templeton, T. J. Meyer, Chem. Sci. 2015, 6, 6398–6406; h) M. S.
Eberhart, K.-R. Wee, S. Marquard, K. Skinner, D. Wang, A. Nayak, T. J.
Meyer, ChemSusChem 2017, 10, 2380–2384; i) Z. Chen, J. J.
Concepcion, H. Luo, J. F. Hull, A. Paul, T. J. Meyer, J. Am. Chem. Soc.
2010, 132, 17670–17673; j) J. J. Concepcion, M.-K. Tsai, J. T.
Muckerman, T. J. Meyer, J. Am. Chem. Soc. 2010, 132, 1545–1557; k)
W. Song, A. Ito, R. A. Binstead, K. Hanson, H. Luo, M. K. Brennaman,
J. J. Concepcion, T. J. Meyer, J. Am. Chem. Soc. 2013, 135,
11587−11594; l) J. J. Concepcion, R. A. Binstead, L. Alibabaei, T. J.
Meyer, Inorg. Chem. 2013, 52, 10744−10746; m) Z. Chen, J. J.
Concepcion, N. Song, T. J. Meyer, Chem. Commun. 2014, 50, 8053–
8056; n) K. L. Materna, B. J. Brennan, G. W. Brudvig, Dalton Trans.
2015, 44, 20312–20315.
L. Duan, Y. Xu, L. Tong, L. Sun, ChemSusChem 2011, 4, 238–244.
L. Francàs, R. M. González-Gil, D. Moyano, J. Benet-Buchholz, J.
García-Antón, L. Escriche, A. Llobet, X. Sala, Inorg. Chem. 2014, 53,
10394–10402.
In theory for n-electron diffusional couples ΔEp = 0.059/n V and W 1/2 =
0.094/n V (limiting value for DPV pulse amplitude ΔE → 0).
a) N. Planas, L. Vigara, C. Cady, P. Miró, P. Huang, L. Hammarström,
S. Styring, N. Leidel, H. Dau, M. Haumann, L. Gagliardi, C. J. Cramer,
A. Llobet, Inorg. Chem. 2011, 50, 11134–11142; b) K. J. Takeuchi, M.
S. Thompson, D. W. Pipes, T. J. Meyer, Inorg. Chem. 1984, 23, 1845–
1851.
It is important to mention that during spectroelectrochemical studies the
slow scan rate (2 mV s−1) and the geometry of the cell (ca. 0.2 mm
optical path through the cell containing Pt mesh working electrode)
ensures bulk electrolysis of all the electroactive material at any given
potential. However, due to the cell geometry and sensitivity of the Ag
wire pseudo-reference electrode the peak separation is not always
consistent with the CV measurements made with planar working
electrodes under diffusional control (see the Supporting Information for
details).
D. R. Weinberg, C. J. Gagliardi, J. F. Hull, C. F. Murphy, C. A. Kent, B.
C. Westlake, A. Paul, D. H. Ess, D. G. McCafferty, T. J. Meyer, Chem.
Rev. 2012, 112, 4016–4093.
a) D. H. Evans, Chem. Rev. 1990, 90, 739–751; b) A. J. Bard, L. R.
Faulkner. Electrode Reactions with Coupled Homogeneous Chemical
Reactions. In Electrochemical Methods: Fundamentals and
Applications (2nd ed.); John Wiley & Sons: New York, 2001; pp 471–
533.
a) S. Roeser, P. Farras, F. Bozoglian, M. Martínez-Belmonte, J. BenetBuchholz, A. Llobet, ChemSusChem 2011, 4, 197−207; b) P. Farras, S.
Maji, J. Benet-Buchholz, A. Llobet, Chem. – Eur. J. 2013, 19,
7162−7172.
a) C. Costentin, S. Drouet, M. Robert, J.-M. Saveant, J. Am. Chem. Soc.
2012, 134, 11235−11242; b) R. Matheu, S. Neudeck, F. Meyer, X. Sala,
A. Llobet, ChemSusChem 2016, 9, 3361–3369.

"This is the peer reviewed version of the following article:Electrochemically-Driven Water Oxidation by a Highly
Active Ruthenium-Based Catalyst, which has been published in final form at:
https://onlinelibrary.wiley.com/doi/10.1002/cssc.201900097
This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for SelfArchiving."

FULL PAPER
FULL PAPER
A. Shatskiy, A. A. Bardin, M. Oschmann,
R. Matheu, J. Benet-Buchholz, L.
Eriksson, M. D. Kärkäs, E. V. Johnston,
C. Gimbert-Suriñach, A. Llobet,* B.
Åkermark*
Page No. – Page No.
Text for Table of Contents

Electrochemically-Driven Water
Oxidation by a Highly Active
Ruthenium-Based Catalyst