“This document is the Accepted Manuscript version of a Published Work that appeared in final
form in Nature Reviews Chemistry, copyright © 2019 Springer Nature Publishing AG. after peer
review and technical editing by the publisher. To access the final edited and published work see:

https://www.nature.com/articles/s41570-019-0096-0#rightslink

The development of molecular water oxidation catalysts
Roc Matheu,a,b Pablo Garrido-Barros,a,b Marcos Gil-Sepulcre,a Mehmed Z. Ertem,c Xavier Sala,d
Carolina Gimbert-Suriñacha & Antoni Llobet a,d,*
a

Institute of Chemical Research of Catalonia (ICIQ), Barcelona Institute of Science and

Technology (BIST), Avinguda Països Catalans 16, 43007 Tarragona, Spain.
b

Departament de Química Física i Inorgànica, Universitat Rovira i Virgili, Marcel·lí Domingo s/n,

43007 Tarragona, Spain.
c

Chemistry Division, Energy & Photon Sciences Directorate, Brookhaven National Laboratory,

Upton, New York, 11973-5000, USA.
d

Departament de Química, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, 08193

Barcelona, Spain.

Email: allobet@iciq.cat

Abstract | There is an urgent need to transition from fossil fuels to solar fuels — not only to lower
CO2 emissions that cause global warming, but also to ration fossil resources. Splitting H2O with
sunlight emerges as a clean and sustainable energy conversion scheme that can afford practical
technologies in the short to midterm. A crucial component in such a device is a water oxidation
catalyst (WOC). These artificial catalysts have mainly been developed over the last two decades,
which is in contrast to Nature’s WOCs, which have featured in its photosynthetic apparatus for
more than a billion years. This time period has seen the development of increasingly active
molecular WOCs, the study of which affords an understanding of catalytic mechanisms and
decomposition pathways. This Perspective offers a historical description of the landmark molecular
WOCs, particularly ruthenium systems, that have guided research to our present degree of
understanding.

Graphical abstract

Table of contents blurb
Water oxidation catalysts are key components in water-splitting devices for sustainable schemes to
synthesize fuels by using energy, including that from sunlight. This Perspective presents historical
developments in the molecular water oxidation catalysis field, placing particular emphasis on
studies of ruthenium complexes that have taught us much about how to design optimal catalysts.

[H1] Introduction
Global warming of our planet threatens the wellbeing of modern societies as we know them.1,2,3,4
This threat has brought together environmentally educated scientists and laypersons to urge world
leaders and politicians to develop policies that ensure the long-term preservation of our planet while
maintaining our societal lifestyle.5,6 Among the most promising of these policies are those based on
a circular economy and low carbon emissions. The concept of a circular economy relies on green
chemistry and catalysis, and thus highlights the key role chemistry plays in society today as well as
the important role it will likely play in solving problems in the near future.7 The main contributors
to global warming are fossil fuels, the use of which generates 80–85% of all the anthropogenic
energy used on Earth.8,9 This warming, as well as the non-renewable nature of these fuels, see it
imperative to develop sustainable and clean energy conversion schemes.
Nature has been using photosynthesis for a billion years10,11,12 to generate carbohydrates from CO2
and H2O (eq. 1), storing solar energy in chemical bonds by driving a thermodynamically uphill
conversion. Splitting H2O with sunlight (eq. 2) can be regarded as an analogous reaction to
photosynthesis13,14,15,16,17 in that both reactions are thermodynamically uphill and driven by sunlight.
Moreover, in both cases the use of sunlight to drive the reaction requires intermediate molecules (or
materials) capable of transferring the sunlight energy to activate catalysts. Lastly, both conversions
are redox reactions in which the oxidation half-reaction is the conversion of H2O to O2 (eq. 3). The
reactions only differ in their reductions: in photosynthesis, CO2 is reduced to hydrocarbons, while in
water splitting H+ is reduced to H2 (eq. 4).

6CO2(g) + 6H2O(l) + hν → C6H12O6(aq) + 6O2(g) (1)
2H2O(l) + hν → O2(g) + 2H2(g) (2)
2H2O(l) → O2(g) + 4H+(aq) + 4e– Eo = 1.23 V vs. NHE at pH 0.0 (3)
2H+(aq) + 2e– → H2(g) Eo ≡ 0.0 V vs. NHE at pH 0.0 (4)

The slow kinetics of the H2O oxidation reaction calls for catalysis, which, if economically viable,
could see sunlight-driven H2O splitting (WS-hν) as a potential solution to today’s energy problem.
Concomitantly, the H2 evolved at the cathode could be used later to react with O2 to generate H2O
and release energy. Alternatively, the H2 can be used to hydrogenate CO2 to give CO, MeOH, CH4
or synthetic hydrocarbons using the Fischer–Tropsch process.18,19,20 These energy vectors, if
obtained in this way, are referred to as solar fuels and could form the basis for an energy democracy
within a circular economy.21

Performing WS-hν in an efficient manner requires access to an active and robust water oxidation
catalyst (WOC). This role is filled in Nature by the oxygen evolving complex in photosystem II
(OEC-PSII), which features a Mn4Ca cubane cluster (FIG. 1a).22,23,24,25 The OEC-PSII converts two
H2O molecules into O2 on the millisecond time scale (1.6 ms) by a series of sunlight-induced charge
transfer steps known as the Kok cycle.26,27,28,29,30 The Mn4Ca cluster is reassembled approximately
every 10–15 minutes due to the damage induced on the amino acid residues coordinated to the
cluster, as a result of reacting with photogenerated 1O2.31,32 Given the lifetime of the cluster and the
rate at which it evolves O2, we can estimate that its turnover number (TON) is on the order of 106.

Figure 1 | Nature uses metalloenzymes to effect redox reactions that consume or afford H2O. a
| The oxygen-evolving complex in photosystem II is a Mn4Ca cluster that mediates the Kok cycle,
in which light energy input enables the oxidation of 2H2O to O2. A proposed structure of the cluster
from the protein in its S0 state is drawn. b | The active site of cytochrome P450 enzymes activates
O2 to give H2O and the oxo complex [Fe(porphyrinato)(cysteinato)(O)], which goes on to oxidize
organic substrates. An X-ray structure of the oxo complex is presented. Part a is adapted with
permission from REF. 30, Elsevier. Part b is adapted with permission from REF. 79.

One of the key features of the OEC-PSII is its capacity to carry out a series of consecutive redox
reactions at potentials close to the thermodynamic value of the 2H2O(l) → O2(g) + 4H+(aq) + 4e− halfreaction. Indeed, one can imagine that sequential oxidations should become more difficult, but not
if the phenomenon of redox leveling is at play. This is enabled by proton-coupled electron-transfer
(PCET) processes33,34,35,36 that see the initial resting state S0 undergoing four light-driven oxidations
to give S4, the state that can release O2. Three of these steps involve the simultaneous removal of H+
and e− or a single H atom. In this way, the Mn4Ca cluster sees its Mn centres undergo sequential
oxidations, which in three cases do not change the overall charge of the cluster until the generation
of O2.37 Thus, the cluster is progressively activated until it reaches a state that is sufficiently reactive
to generate an O–O bond. The exact molecular details of how this occurs remains a topic of
debate.38,39,40,41 For contrast, Nature can not only convert H2O to O2 but can also reduce O2 back to
H2O with concomitant oxidation of an organic substrate, as occurs in cytochrome (Cyt) P450
enzymes (FIG. 1b).
Synthetic WOCs can be broadly classified into two main groups: oxides42,43,44 and molecular
complexes.45,46,47 The use of oxides as H2O oxidation catalysts can be traced back to 1903 and a

report by Cohen and Gläser about their attempted measurement of the potential of a CoIII/II couple,
which resulted instead in the deposition of CoOx on their electrode and large currents that reflect
catalytic H2O oxidation.48 In contrast, the first well-defined molecular WOC came in 1982, and this
so-called ‘blue dimer’ (1, FIG. 2) was reported at a time when coordination chemistry was already
well established.49 Since then, a large number of complexes capable of oxidizing H2O to O2 have
emerged, prominent examples of which are presented in FIG. 2. This Perspective presents a
historical account of molecular WOCs and places emphasis on systems that, at the time they were
reported, advanced our understanding of the field.

Figure 2 | Transition metal complexes of N-donor ligands are prominent H2O oxidation
catalysts. bbp−, (3,5-bis(2,2′-bipyridin-6-yl)pyrazolate; bda2−, 2,2′-bipyridine-6,6′-dicarboxylate;
bpm, 2,2′-bipyrimidine; bpp−, 3,5-bis(2-pyridyl)pyrazolate; bpy, 2,2′-bipyridine; misoq, 6methoxyisoquinoline; mox4−, N1,N1′-(1,2-phenylene)bis(N2-methyloxalamide); ppy−, 2-(2phenylido)pyridine; py, pyridine; tda2−, 2,2′:6′,2′′-terpyridine-6,6′′-dicarboxylate; tpm, tris(2pyridyl)methane; tpy, 2,2′:6′,2′′-terpyridine.

[H1] The blue dimer and the early days
It was long ago observed that simply adding [(bpy)3RuIII]3+ (bpy = 2,2′-bipyridine) to H2O resulted
in the evolution of O2. However, neither the mechanism nor the potential high oxidation states and
active species for the O–O bond formation were characterized.50,51 Thus, shortly thereafter, the
report on the blue dimer saw it become the first well-characterized molecular complex that behaves
as a H2O oxidation catalyst in the presence of a sacrificial oxidant. The ‘blue dimer’ cis[(H2O)RuIII(bpy)2(μ-O)RuIII(bpy)2(OH2)]4+ is a complex wherein two RuIII centres are bridged by an
oxo group to give a complex with a strong absorption at λmax = 637 nm (ε = 21100 M−1cm−1 at pH
1.0).52 The pseudo-octahedral RuIII sites each feature two bpy co-ligands in a cis arrangement with
the last position being occupied by an aquo group. The redox properties of this complex have been
thoroughly studied by electrochemistry, which shows that the higher Ru oxidation states required
for catalysts are indeed accessible.52 In 1 and other Ru complexes described here, at the potentials
relevant to H2O oxidation it is usually the Ru centres and/or the oxygenic ligands that undergo
redox rather than the co-ligands.

At pH 1.0, where most of the catalytic H2O oxidation reactions have been carried out for 1, the blue
dimer undergoes two redox processes. First, the [(H2O)RuIII(bpy)2(μ-O)RuIII(bpy)2(OH2)]4+ species
undergoes loss of H+ and e− at Eo = 0.79 V (all redox potentials here are reported against the normal

hydrogen electrode, NHE) to give [(H2O)RuIII(bpy)2(μ-O)RuIV(bpy)2(OH)]4+. In order to easily keep
track of electron counts, we denote formal oxidation states in this Perspective; for Ru in oxidation
states above +III, electron density is removed both from the Ru centre and any O2− ligand bonded to
it, conferring oxyl character (O−∙) on the ligand. Second, further oxidation involves loss of 3e− and
3H+ in a single step at Eo = 1.22 V to afford [(O)RuV(bpy)2(μ-O)RuV(bpy)2(O)]4+. Taken together,
these two processes result in the removal of the 4H+ and 4e− required for the oxidation of 2H2O
molecules to O2 (eq. 3), as occurs in the OEC-PSII. It was initially thought that [(O)RuV(bpy)2(μO)RuV(bpy)2(O)]4+ would undergo intramolecular coupling of the two RuV=O units to liberate O2
and give back, on aquation, the starting complex [(H2O)RuIII(bpy)2(μ-O)RuIII(bpy)2(OH2)]4+.53 This
mechanism involves the interaction of two M–O units (I2M) but is now thought not to be operative
here. Instead, kinetics and labeling experiments showed that O–O bond formation occurs through
water nucleophilic attack (WNA) on a RuV=O unit to form a Ru–O2H intermediate prior to O2
liberation (eq. 5,6).53 Despite this conclusion, isotopic labeling experiments performed under the
exact same conditions led to contradictory results, reflecting the difficulty of carrying out such
experiments, in which headspace contamination with atmospheric O2 can easily take place.54,55,56,57
[(O)RuV(bpy)2(μ-O)RuV(bpy)2(O)]4+(aq) + H2O(l) →
[(O)RuIV(bpy)2(μ-O)RuIV(bpy)2(O2H)]3+(aq) + H+(aq)

(5)

[(O)RuIV(bpy)2(μ-O)RuIV(bpy)2(O2H)]3+(aq) + H2O(l) →
[(H2O)RuIII(bpy)2(μ-O)RuIII(bpy)2(OH2)]4+(aq) + O2 + H+(aq) (6)
In the presence of excess CeIV, a sacrificial 1e− oxidant,58 1 oxidizes H2O to O2 with 4–5 turnovers
and an initial turnover frequency (TOF; FIG. 3) of 4.2×10−3 s−1.52 It is assumed that one of the major
shortcomings for this catalyst is the coordination of anions (anation), a process that competes with
aquation and deactivates the system.59,60 The TOF of a catalyst is highly sensitive to the conditions
under which it is studied, although Figure 3 affords a reasonable qualitative picture of catalyst
evolution because all the TOF values were determined at pH 1.0 (except for that of 11, which was
determined at pH 10.0).
The TOF values can be measured chemically using a sacrificial electron acceptor reagent, as is the
case just described for 1, but can also be measured electrochemically that is by applying a potential
through an electrode. In the chemical case the potential generated is limited by the redox potential
of the sacrificial reagent whereas in the electrochemical case the applied potential can be varied
basically at will. In the electrochemical case the Foot of the Wave Analysis (FOWA) developed by

Saveant et al.,61 has become one of most convenient ways to calculate TOF. Since the TOF values
are dependent on the applied potential, FOWA generally reports the maximum achievable value for
a given catalyst, that is named TOFmax.
What becomes clear is that for more than 20 years the blue dimer 1 and related derivatives62,63 were
the only well-characterized molecular WOCs. One of the main challenges to developing new Ru
catalysts was the unavailability of rational synthetic strategies for the preparation of dinuclear oxobridged aquo complexes. A breakthrough came with the realization that the O2− bridge was not
indispensable for WOC activity. Thus, it became possible to use alternative bridging ligands,
particularly organic N-donors as in [(H2O)RuII(tpy)(μ-bpp)RuII(tpy)(OH2)]3+ (2, tpy = 2,2′:6′,2′′terpyridine, bpp− = 3,5-bis(2-pyridyl)pyrazolate), where a pyrazolate moiety in bpp− bridges the two
Ru centres64. Complex 2 is notable in that its two Ru–OH2 groups are rigidly held in an orientation
in which the O atoms are very close together and thus ready to undergo intramolecular coupling
once 2 is suitably oxidized. Indeed, oxygen labeling experiments and theoretical calculations
confirmed that this process can occur, and this system is the first synthetic example in which an O–
O bond is generated through an I2M mechanism.65,66,67 This preorganization of the active Ru–O
groups greatly favours the reaction by minimizing any activation entropy and leading to a TOF of
1.4×10−2 s−1 — 3.4 times greater than that of the blue dimer 1.

Figure 3 | The evolution of molecular H2O oxidation catalysts. Refined designs enable the
turnover frequency (TOF) for the conversion of 2H2O to O2 to even exceed that of the oxygenevolving complex in photosystem II (OEC-PSII). TOF values for complexes were determined at pH
1.0 chemically using Ce(IV) as sacrificial oxidant, except in the case of 11, when pH 10.0
electrolyte was used and the TOF was determined electrochemically.

Being able to use organic bridging ligands enabled synthetic chemists to rapidly develop new Ru
WOCs. This synthetic strategy of avoiding O2−-bridged species has since afforded a family of
diruthenium complexes in which auxiliary ligands can control the mechanistic pathway through
which O–O bond formation occurs. This is nicely illustrated in the case of [(H2O)RuII(tpm)(μbpp)RuII(tpm)(OH2)]3+ (3, tpm = tris(2-pyridyl)methane), a derivative of 2 in which the meridional
tpy ligand has been substituted with the facial tpm ligand. With this new structural arrangement, the
two aquo groups that were facing one another in 2 are now situated trans to each other in 3. As a
result, the intramolecular I2M pathway is not accessible for 3 and O–O bond formation instead
proceeds through an intermolecular I2M pathway. Moving from the tetradentate ligand bpp− to the

hexadentate ligand bbp− (3,5-bis(2,2′-bipyridin-6-yl)pyrazolate) affords yet another distinct motif.
In [(H2O)RuII(py)2(μ-bbp)RuII(py)2(OH2)]3+ (4, py = pyridine),68 for example, the equatorial bbp−
ligand constrains the bond angles at the Ru centres in the equatorial plane, thereby separating the
two Ru–OH2 moieties in 4 relative to those in 2. Thus, 4 cannot participate in an I2M pathway but
instead undergoes the O–O bond formation through a WNA path.

Although isotopic labeling experiments and measurements of the relative abundance of evolved O2
isotopologues afforded contradictory results in the case of 1 (probably due to leakage of
atmospheric O2), the labeling experiments provided very consistent data for 2–4. This is important
because these experiments provide extremely valuable mechanistic information that allows the
unambiguous identification of the O–O bond formation pathway. The high quality of the data
obtained for 2–4 is partly due to today’s commercially available mass spectrometry instrumentation,
which allow one to interface atmospheric pressure reaction vessels directly to a high vacuum
measurement chamber.

[H1] The ‘one site is enough’ revolution
Although the emergence of dinucleating organic ligands does enable the synthesis of many
complexes, preparing tailored dinuclear Ru–OH2 complexes remains a difficult task that can take
years of synthetic efforts.69 In comparison, a targeted mononuclear Ru–OH2 complex can typically
be prepared in weeks, making it very desirable to see if one could develop active mononuclear
WOCs. Indeed, when the mononuclear complex [(ntp)(mpy)2RuII(OH2)]2+ (5, mpy = 4methylpyridine, ntp = 2,6-di(1,8-naphthyridin-2-yl)-4-tert-butyl-pyridine) was reported70 to behave
as a H2O oxidation catalyst, a revolution in the field ensued because easy access to Ru WOCs was
conceivable. Indeed, many research efforts were directed at studying mononuclear Ru complexes as
WOC candidates, affording a better understanding of WOC mechanisms in general. This knowledge
contributed to the development of faster catalysts that emerged from 2005 onwards (FIG. 3).
Methodology also became more reliable, although the paper describing the catalytic activity of 5
included a mass balance violation.70 The maximum amount of O2 that can be produced under the
conditions reported assuming 100% chemical efficiency would give a TON of 228, less than half of
the reported TON of 580. This error, which was corrected in a subsequent publication, is most
likely due to miscalibration of the polarographic probe used to quantify the generated O2 gas. We
stress again that the execution of catalytic experiments in this field requires utmost care.

In 2005, the activity of mononuclear complexes was puzzling because it was assumed that a
dinuclear-diaquo complex (or larger system) was needed to be able to remove 2H+ and 2e− from
each Ru–OH2 group. This mystery was solved a few years later when the mechanistic analysis of
[(tpy)(bpm)RuII(OH2)]2+ (6, bpm = 2,2′-bipyrimidine) showed that one Ru site was enough to carry
out H2O oxidation catalysis.71 It was proposed that the initial RuII–OH2 complex undergoes a series
of e−/H+ transfers to give the active species RuV=O (eq. 7).
[(tpy)(bpm)RuII(OH2)]2+(aq) → [(tpy)(bpm)RuV(O)]3+(aq) + 2H+(aq) + 3e– (7)
[(tpy)(bpm)RuV(O)]3+(aq) + H2O(l) → [(tpy)(bpm)RuIII(O2H)]2+(aq) + H+(aq) (8)
[(tpy)(bpm)RuIII(O2H)]2+(aq) → [(tpy)(bpm)RuIV(O2)]2+(aq) + H+ + e– (9)
[(tpy)(bpm)RuIV(O2)]2+(aq) + H2O(l) → [(tpy)(bpm)RuII(OH2)]2+(aq) + O2(g) (10)
A similar reasoning is also proposed for the related complex [(tpy)(bpy)RuII(OH2)]2+, and this
mechanism is consistent with its Pourbaix diagram (FIG. 4a), which, as will become clear indicates
that it is less electron-rich than [(bda--N2O)(py)2RuII(OH2)] (9(OH2)), which features an anionic
ligand (FIG. 4b).72,73,74,75 In any case, once the RuV=O species is generated, kinetics experiments
support a WNA mechanism that gives the corresponding hydroperoxo RuIII–O2H (eq. 8).

Figure 4 | Pourbaix diagrams for Ru complexes indicate stability across a wide pH range. The
schemes can be considered phase diagrams of complexes with the variables being potential (E, in
V) and pH. a | The Pourbaix diagram for [(tpy)(bpy)RuII(H2O)]2+. b | The Pourbaix diagram for
[(bda--N2O)(py)2RuII(OH2)] (9(OH2)). Blue lines correspond to the RuV/IV couples, dashed green
lines to the standard potential Eo(H2O/O2) and vertical lines to pKas. The co-ligands and overall
charges are omitted for clarity. bda2−, 2,2′-bipyridine-6,6′-dicarboxylate; bpy, 2,2′-bipyridine; py,
pyridine; tpy, 2,2′:6′,2′′-terpyridine.
The RuIII–O2H species undergoes one more oxidative process before liberating O2 and regenerating
the initial RuII–OH2 complex (eqs 9,10) to close the catalytic cycle (FIG. 5a). Although a distinct
mechanism is possible in another system (FIG. 5b is described below), the present cycle and
variations thereof have now been widely proposed for a variety of mononuclear Ru complexes and
other mononuclear transition metal complexes proposed to be molecular WOCs.76 The mechanism
resembles the one by which Cyt-P450 activates O2 and oxidizes organic substrates (FIG. 1b),
although the two processes occur in opposite directions.77,78,79

Figure 5 | Two mechanisms of molecular H2O oxidation catalysts. a | The water nucleophilic
attack (WNA) mechanism involves O–O bond formation between an oxo and H2O substrate. This
mechanism is operative for many monoruthenium complexes such as [(tpy)(bpm)RuII(OH2)]2+ (6).71
b | Alternatively, O–O bond formation can occur through interaction of two M–O units (I2M), a
mechanism at play for [(bda)(py)2Ru] complexes similar to 9.98 bda2−, 2,2′-bipyridine-6,6′dicarboxylate; bpm, 2,2′-bipyrimidine; py, pyridine; tpy, 2,2′:6′,2′′-terpyridine.

The discoveries of the mononuclear catalysts were real breakthroughs because of their ease of
preparation and synthetic versatility. However, mononuclear Ru–OH2 complexes have the
propensity to form O2−-bridged dinuclear and even polynuclear complexes80,81,82 through processes
that may compete with H2O oxidation catalysis. Indeed, the terminal Ru=O/Ru–O∙ can bridge to
another Ru centre instead of undergoing WNA or I2M. The formation of a bridged complex can
spell deactivation of the catalyst but can also afford species that are active WOCs, as is the case for
[(tpy)(bpy)RuII(OH2)]2+ (eq. 11).83 In this case, the resulting O2−-bridged complex trans[(tpy)(bpy)RuIV(μ-O)RuIV(tpy)(OH2)(O)]4+ (7) is actually a more active and robust catalyst than the
parent complex.
2 [(tpy)(bpy)RuII(OH2)]2+(aq) + H2O(l) →
trans-[(tpy)(bpy)RuIV(μ-O)RuIV(tpy)(OH2)(O)]4+(aq) + bpy(aq) + 4 H+(aq) + 4 e− (11)

The increase in popularity of catalytic H2O oxidation has afforded a growing family of catalyst
candidates featuring Ru or other transition metals. Although we mentioned above that the
complexes described here have ligands that are resistant towards oxidation, some other ligands,
including those with benzylic/picolinic methylene groups, readily undergo oxidation. Careful
analyses of the respective Ru complexes clearly reveals ligand oxidation, in many cases all the way
to CO2.84,85,86 To illustrate just how important it is to consider the possibility of oxidation, we now
consider the thermodynamics of PhCH2Me oxidation by H2O to afford acetophenone and H287 (eq.
12), analogous to the H2O splitting reaction (eq. 13).
PhCH2Me(aq) + H2O(l) → PhC(O)Me(aq) + 2 H2(g) ΔGo = 22.3 kcalmol−1 (K = 3.8×10–17 at 298 K) (12)
2 H2O(l) → O2(g) + 2 H2(g) ΔGo = 113.5 kcalmol−1 (K = 2.9×10–84 at 298 K) (13)

The equilibrium constant of the organic oxidation is more than 67 orders of magnitude larger than
that of H2O splitting (eq. 12,13). Moreover, the kinetics of PhCH2Me oxidation are also favorable,

as is the case for benzylic/picolinic methylene groups. We thus surmise that there is no way that an
initial complex featuring benzylic/picolinic methylene groups can be a true WOC; activity observed
when using such a complex likely instead arises from a complex of oxidized ligand(s) or the
corresponding oxide cluster.88 We express an additional cautionary note with respect to dynamic
light scattering (DLS), the detection limit (1–2 nm) of which is greater than the size of low
molecular weight clusters, including those derived from Co phosphates.89,90 The existence of these
oxide clusters can thus not be ruled out using DLS, leading to a typical misinterpretation of data,
particularly for first row transition metal complexes. Examples of these complexes can be labile,
such that their catalytic activity, which often is attributed to the complex itself, arises instead from
metal oxide nanoparticles.48,88,91,92

In parallel with research into mononuclear Ru WOCs, a number of Ir complexes have also been
developed to serve as WOCs. The first of these was [(ppy)2IrIII(OH2)2]+ (ppy− = 2-(2phenylido)pyridine) — the first organometallic complex to be used as a WOC. This complex was
followed by [(C5Me5)(ppy)IrIIICl] (8),93 but later studies showed that 8 readily loses its C5Me5–
ligand to form Ir–O–Ir complexes.94 In retrospect, it is not surprising that a π-acidic ligand such as
C5Me5– is not able to stabilize Ir in the high oxidation states required for H2O oxidation. In addition,
it was later shown that the putative Ir–O2H intermediate could undergo an entropically favored
intramolecular reaction with the C–H and C–Me bonds of C5Me5– to eventually form HCO2H and
MeCO2H as decomposition products.95,96

[H1] Seven-coordinate complexes and FAME ligands
The report of the complex [(bda)(py)2RuII] (9, bda2− = 2,2′-bipyridine-6,6′-dicarboxylate)97,98
represents a turning point for the H2O oxidation catalysis field. As mentioned in the previous
section, preparing mononuclear Ru–OH2 complexes is relatively easy but still poses some
challenges because they must incorporate a substitutionally labile OH2 ligand. Thankfully, the H2O
adduct of 9 forms on dissolving 9 in H2O, with the aquo ligand bonding to RuII at the expense of
one of the hemilabile carboxylates (eq. 14).99 The system is dynamic at room temperature, and the
two carboxylates can coordinate and decoordinate very quickly in order to protect or vacate
coordination sites at Ru.100
[(bda-κ-N2O2)(py)2RuII](aq) + H2O(l) ⇌ [ (bda-κ-N2O)(py)2RuII(OH2)](aq) (14)

The fact that one can generate the Ru–OH2 complex in situ enormously simplifies the development
of a catalyst/precatalyst. Oxidizing the aquo complex in aqueous solution affords a RuIV species in
which the demands of the electrophilic metal not only see it coordinate the py ligands and substrate,
but also all four donors in bda2−. The seven coordinate centre adopts a pseudo pentagonal
bipyramidal geometry in which the axial positions are occupied by the py ligands (eq. 15).
[(bda-κ-N2O)(py)2RuII(OH2)](aq) → [(bda-κ-N2O2)(py)2RuIV(OH)]+(aq) + 2e– + H+(aq) (15)
The anionic carboxylate groups in [(bda-κ-N2O2)(py)2RuIV(OH)]+ bind the Ru strongly and further
electron density comes from the basic OH− group, enabling the stabilization of high oxidation states
— RuIV and RuV — just when it is most needed. Indeed, the above phenomena enable 9 to mediate
H2O oxidation catalysis at pH 1.0 at an overpotential of only 170 mV. The strong effects of electron
donation and anionic charge on the carboxylates become evident by noting that the redox potential
of the RuV/IV couple for 9(OH2) is 800 mV lower than that of [(tpy)(bpy)RuII(OH2)]2+ (FIG. 4b) , a
complex that contains only charge-neutral ligands. Indeed, a correlation between the potential of the
RuV/IV couple and the anionic nature of the ligands bonded to Ru has been established that facilitates
understanding the effect of ligands on redox potentials.101 On reaching the oxidized RuV=O/RuIV–O∙
state (eq. 16), the complex dimerizes to give the corresponding peroxo intermediate [(bda-κN2O2)(py)2RuIV(μ-η1:η1-O2)RuIV(bda-κ-N2O2)(py)2]2+ (eq. 17). This dimer, in the presence of excess
oxidant, is transformed to the corresponding superoxo that can liberate O2 (eq. 19, FIG. 5b).
[(bda-κ-N2O2)(py)2RuIV(OH)]+(aq) → [(bda-κ-N2O2)(py)2RuV(O)]+(aq) + e– + H+(aq) (16)
2 [RuV(O)(bda-κ-N2O2)(py)2]+(aq) →
[(bda-κ-N2O2)(py)2RuIV(μ-η1:η1-O2)RuIV(bda-κ-N2O2)(py)2]2+(aq) (17)
[(bda-κ-N2O2)(py)2RuIV(μ-η1:η1-O2)RuIV(bda-κ-N2O2)(py)2]2+(aq) →
[(bda-κ-N2O2)(py)2RuIV(μ-η1:η1-O2)RuIV(bda-κ-N2O2)(py)2]3+(aq) + e–

(18)

[(bda-κ-N2O2)(py)2RuIV(μ-η1:η1-O2)RuIV(bda-κ-N2O2)(py)2]3+(aq) + 2 H2O(l) →
2 [(bda-κ-N2O2)(py)2RuIV(OH)]+(aq) + 2 H+(aq) + e− + O2(g) (19)

Kinetics experiments conducted at low catalyst concentrations indicate that the dimerization process
is the rate-determining step (RDS) of the catalytic cycle. Replacing the axial py ligands in 9 with π-

extended ligands such as 6-methoxyisoquinoline (misoq) affords [(bda)(misoq)2RuII] (10), a WOC
with a TOF more than two orders of magnitude greater than that for 9 measured under exactly the
same conditions.102,103,104,105 The misoq ligands allow for stronger van der Waals and π–π
interactions between the RuV=O/RuIV–O∙ complexes, thereby lowering the activation free energy
barrier of the RDS for 10 relative to that for the py derivative 9. The facile formation of the peroxo
dimer 10(μ-η1:η1-O2)10 (FIG. 2) enables TOFs of around 1000 s−1, which are on the same order of
magnitude of those for OEC-PSII (TOF = 1/(1.6 ms) ≈ 600 s−1).

We wish to caution the reader against assuming that any Ru complex of N- and O-donor ligands,
when dissolved in H2O and/or treated with an oxidant, will spontaneously afford the requisite Ru–
OH2, Ru–OH or Ru=O group and go on to behave as a WOC. Although this does occur in the case
of 9, it is by no means general. Therefore, thorough characterization of complex intermediates
remains indispensable to understanding the chemistry of a particular catalyst candidate. For
example, by making invalid assumptions one is at a high risk of reporting inaccurate mechanisms,
as transpired in the case of [(qpy)(py)2RuII] (qpy = 2,2′:6′,2′′:6′′,2′′′-quaterpyridine), which also
features a tetradentate ligand.106 The addition of Ce(IV) to an acidic solution of [(qpy)(py)2RuII]
generates a large amount of O2. However, after characterizing the species present in solution after
catalysis, it was found that the terminal pyridyl groups of qpy underwent oxidation to afford the
corresponding oxide 2,2′:6′,2′′:6′′,2′′′-quaterpyridine-1,1′′′-dioxide (qpy-O2), which remained bound
to Ru. Thus, it appears that [(qpy-O2)(py)2RuII] is the real catalyst,107 although we still lack clear
characterization of the Ru–OH2/Ru–OH/Ru=O active species. This oxidation of the terminal pyridyl
groups does not take place in related complexes such as [(tpy)(bpy)RuII(OH2)]2+. This has been
rationalized in terms of the coordination geometry of qpy, which can only bind four equatorial sites
if large distortions from the ideal octahedral Ru geometry are conceded. Thus, strain can be relieved
if the terminal pyridyls in qpy undergo decoordination and subsequent intramolecular oxidation by
a Ru=O group.
At the time of writing, the best molecular WOC, as judged from TOFs, is [(tda)(py)2RuII] (tda2− =
2,2′:6′,2′′-terpyridine-6,6′′-dicarboxylate).108,109 The tda2− ligand contains one more pyridyl group
than bda2− but still binds in a tetradentate manner in the complex [(tda-κ-N3O)(py)2RuII] because of
the electronic and steric requirements of RuII. On 2e− oxidation, the RuIV site binds the previously
dangling carboxylate to give the heptacoordinate species [(tda-κ-N3O2)(py)2RuIV], which gives rise
to a well-resolved 1H NMR spectrum consistent with it being diamagnetic. At pH 7.0 and higher
pHs, the RuIV centre undergoes OH− substitution of one carboxylate group to generate the

catalytically active species [(tda-κ-N3O)(py)2RuIV(OH)]+ (11), which now features a dangling
carboxylate (FIG. 2, eq. 20).
[(tda-κ-N3O2)(py)2RuIV]2+(aq) + OH–(aq) → [ (tda-κ-N3O)(py)2 RuIV(OH)]+(aq) (20)
Complex 11 enjoys all the benefits associated with the bda2− ligand in 9: strongly σ-donating
anionic carboxylates, a constrained geometry in the equatorial plane and accessible heptacoordinate
species at high Ru oxidation states.110 Moreover, 11 has a key additional feature in the form of a
decoordinated carboxylate, which is situated so that it can intramolecularly accept H+ during the O–
O bond formation step (RuV=O + H2O → RuIII–O2H + H+, FIG. 6a), as shown using density
functional theory calculations.108,111 This arrangement favors the WNA over the I2M mechanism by
strongly lowering the activation free energy barrier of the WNA RDS, allowing for fast catalysis; a
TOFmax ~8,000 s−1 was measured at pH 7.0 and 50,000 s−1 at pH 10.0. The TOFmax value is
calculated from the cyclic voltammograms of 11 in the presence of H2O, with large current densities
being observed in the range 1.2–1.5 V at pH 10.0 (FIG. 6a). The appearance of reversible waves
(E½ ≈ 0.6 and 1.1 V) assigned to the complex are consistent with the molecular species remaining
intact under turnover conditions.
Figure 6 | Voltammetry of [(tda-κ-N3O)(py)2RuIV(OH)]+ (11) and [(mox)CuII]2− (12) reveals
catalytic H2O oxidation waves. a | The cyclic voltammogram of 11 at pH 10.0 has two reversible
waves as well as a large catalytic wave at high potentials. Density functional theory calculations
indicate that the complex operates by using its free carboxylate to relay H+ to a hydroperoxo ligand.
Part a adapted with permission from reference 108, American Chemical Society. b | Voltammetry
for 12 at pH 11.5 features a reversible CuIII/II wave and a catalytic wave whose onset coincides with
the redox couple for coordinated mox3−/4−. Density functional theory calculations suggest, as
pictured, the intermediacy of a hydroxo complex of the radical ligand mox3−. mox4−, N1,N1′-(1,2phenylene)bis(N2-methyloxalamide); py, pyridine; 2,2′:6′,2′′-terpyridine-6,6′′-dicarboxylate. Part b
adapted with permission from reference 141, American Chemical Society.

An additional feature of complexes such as 9 and 11 is that the key chemistry occurs solely in the
equatorial plane defined by the chelating ligand and Ru–O(H) group; the apical py ligands merely
complete the coordination geometry, which is either octahedral or pentagonal bipyramidal
depending on the oxidation state. The apical ligand is therefore an ideal site to functionalize the
complex without influencing the electronic properties of the Ru centre.112,113,114,115,116 For example,

anchoring a derivative of 11 onto conductive solid supports affords electroanodes115 or lightabsorbing materials to generate hybrid molecular photonanodes.117,118 The Ru-tda2− complexes in
this hybrid molecular material still exhibit the solution-phase reactivity of 11, thus validating the
anchoring strategy. Further, the material is the most stable molecular photoanode ever reported for
H2O oxidation. Overall, the key features associated with both the bda2− and tda2− co-ligands include
flexibility, adaptability, multidenticity and equatorial coordination. For these reasons we often refer
to these privileged systems as FAME ligands.

[H1] First-row molecular WOCs
The ideal solvent in which to conduct H2O oxidation catalysis is obviously H2O, but when present
in high concentrations H2O can compete for metal coordination sites and displace co-ligands. This
is not a problem for RuII and RuIII ions because their complexes typically exist in low-spin 4d6 and
4d5 configurations that are kinetically stable. The only reactive site for polypyridyl-type Ru–OH2
catalysts is the Ru–OH2 bond — an ideal scenario because one can rely on the co-ligands remaining
firmly attached to Ru and simply being spectator ligands. The substitutional inertness of this type of
complexes is exemplified in their Pourbaix diagrams (FIG. 4), which suggest that, aside from H+/e−
transfers, most of them are perfectly stable within the 0–14 pH range.

In sharp contrast, complexes of Mn, Fe, Co, Ni and Cu are in general substitutionally labile and
their stabilities will be highly dependent on their co-ligands and solution pH. For instance, H2O
exchange between the bulk and [M(H2O)6]3+/2+ is eight orders of magnitude faster for both FeII and
FeIII relative to their RuII and RuIII counterparts.119,120 Further complicating the endeavour of
preparing first-row transition metal WOCs is that at low pHs most ligands will undergo
decoordination and protonation, as is the case for Fe–polypyridyl complexes.121,122 Further, at high
pH values the complexes will simply decompose into hydroxides and/or oxides. As the H2O
oxidation reaction proceeds, the solution pH can be drastically reduced because 4H+ are released per
O2 molecule evolved. Consequently, when using first-row transition metal WOCs it is crucial to
carry out catalysis in buffered solution. An additional problem that arises particularly with first row
transition metals is that once the corresponding oxide is formed it can adsorb onto the electrode
surface or precipitate from solution as a colloid. These are additional driving forces for the
decomposition of the initial complex.123 Interestingly, the deposition of transition metal oxides at
electrode surfaces can afford highly active electroanodes for H2O oxidation. These can be
substantially different than those generated at the same pH from the simple metal
salts.42,124,125,126,127,128,129,130,131,132 For these reasons, following electrocatalysis with the molecular

complexes it is absolutely necessary to check the catalytic activity of the electrode in a clean
electrolyte solution. In addition, an excellent way to monitor the stability of the initial molecular
catalyst is to measure a non-catalytic Faradaic wave associated with the complex (such a wave will
appear at lower potentials than the electrocatalytic process). In this way, the integrity of the initial
complex can be easily evaluated during turnover conditions.141

Only a few first-row transition metal complexes are active H2O oxidation catalysts and abide with
the conditions above.133,134,135,136,137,138,139,140 One example is [(mox)CuII]2− (12, mox4− = N1,N1′-(1,2phenylene)bis(N2-methyloxalamide)), in which CuII is strongly bound to a tetraanionic tetraamidato
ligand.141,142 The high electron density donated from this ligand enables access to CuIII at a potential
of only 0.56 V at pH = 11.5, where the complex is fully stable as evidenced by the reversibility of
this CuIII/II wave in cyclic voltammograms (FIG. 6b). At potentials >1.20 V, a large current response
is observed that must arise from an electrocatalytic process. In particular, reversible oxidation of the
aromatic ring affords a radical cation that readily accepts a OH− ligand (FIG. 6b). Following a
WNA step and single electron transfers, one obtains a complex featuring hydrogen bonded H2O2,
with this moiety then undergoing a series of intramolecular e− transfers to release O2 and regenerate
the initial complex. As mentioned above, one can check that the catalyst remains intact by scanning
the potentials at which a reversible wave characteristic of the complex appears. Indeed, the CuIII/II
wave is preserved after the electrocatalytic process, reflective of the high robustness of the catalyst
under these conditions. An additional piece of evidence supporting this proposed mechanism is the
substantial lowering of the overpotential when electron donating groups are installed at the
phenylene ring of mox4−. For instance, the dimethoxy derivative has a ~530 mV lower overpotential
relative to that of the parent catalyst 12. The catalyst 12 is further notable in that it undergoes
reversible oxidation not only at its metal centre but also at the organic ligand.

In contrast to well-characterized WOCs like 12, there are at present a large number of first-row
transition metal complexes that are proposed to operate as molecular H2O oxidation catalysts but
whose active species are actually metal oxides. To avoid the unfortunate case of misinterpreting
data, we strongly recommend that chemists rigorously check the stability of complexes in solution
as well as the activity of a used electrode in a clean electrolyte solution. Neglecting to perform these
checks and only measuring the spectroscopic and redox properties of the initial complex can lead to
a large waste of time if the initial complexes are not responsible for the H2O oxidation
activity.143,144,145,146,147,148,149,150

[H1] Conclusions and challenges
Molecular WOCs based on Ru have enjoyed a spectacular development over the last decade —
arguably a short amount of time to improve TOF values by more than seven orders of magnitude.
Additionally, this period has seen us develop a sound description of mechanisms that has helped
and will continue to help us achieve improvements in terms of catalytic performance and the
identification and avoidance of potential deactivation pathways. In parallel, the lessons learned from
studying Ru complexes are at least partially applicable to other transition metal complexes such as
those of Ir and first-row transition metals such as Fe, Co and Cu, among others.

Those working on developing improved WOCs still face several challenges. The first of these
includes the goal faced by all working in electrocatalysis: to find catalysts that turn over on the
nanosecond time scale at potentials as close as possible to the thermodynamic values. Fast catalysts
are necessary to improve the efficiencies of H2O oxidation photoanodes and H2O-splitting devices
because it is the chemical rather than photochemical steps that are typically rate-limiting. In
addition, when a charge-separated state is photogenerated we must provide it with a way to
immediately participate in the catalytic reaction instead of undergoing degradation. The second
challenge is to develop first-row transition metal complexes that are as fast and robust as the best
Ru complexes. Although there is no reason why this should be impossible, it is obvious that one
needs to pay very careful attention to the idiosyncracies of first-row transition metal complexes and
consider their potential lability. By neglecting this, one can overlook the formation of oxides that
can end up being the actual catalytically active species. The final challenge is associated with
practical electrocatalytic applications, for which molecular species in solution are typically not
useful. However, one can obtain high current densities by anchoring highly active catalysts to solid
electrodes or semiconductors with large surface coverages. In this way, one can construct
technologically useful H2O-splitting devices that could lead the way towards replacing fossil fuels
with solar fuels.

[H1] Acknowledgments
Sustained support from MINECO, FEDER and AGAUR are gratefully acknowledged. Recent
grants include CTQ2015-64261-R, CTQ2016-80058-R, CTQ2015-73028-EXP, SEV 2013-0319,
ENE2016-82025-REDT, CTQ2016-81923-REDC, and 2017-SGR-1631. The work at Brookhaven
National Laboratory (M.Z.E.) was carried out under contract DE-SC0012704 with the U.S.
Department of Energy, Office of Science, Office of Basic Energy Sciences, and utilized
computational resources at the BNL Center for Functional Nanomaterials (CFN). The CFN is a U.S.

DOE Office of Science Facility at Brookhaven National Laboratory, operating under contract No.
DE-SC0012704.

[H1] Author contributions
All authors contributed equally to the preparation of this manuscript.

[H1] Competing interests statement
The authors declare no competing interests.

[H1] Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

[H1] How to cite this article
Matheu, R. et al. The development of molecular water oxidation catalysts. Nat. Rev. Chem. 3,
xxxxx (2019).

References.
1

Steffen, W. et al. Trajectories of the Earth System in the anthropocene. Proc. Natl Acad. Sci. USA 115,

8252–8259 (2018).
2

Balmaseda, M. A., Trenberth, K. E. & Källén W. Distinctive climate signals in reanalysis of global ocean heat

content. Geophys. Res. Let. 40, 1754–1759 (2013).
3

IPCC Climate Change 2013: The Physical Science Basis (Eds Stocker,T. F. et al.) (Cambridge University Press,

Cambridge, 2013).
4

IPCC Climate Change 2014: Mitigation of Climate Change (Eds Edenhofer, O. et al.) (Cambridge University

Press, Cambridge, 2014).
5

Kyoto Protocol to the United Nations Framework Convention On Climate Change, United Nations

Framework Convention on Climate Change (UNFCCC), 1998.
6

Adoption of the Paris Agreement. Proposal by the President, Paris Climate Change Conference – November

2015, COP 21, Paris, France, 2015.
7

Staher, W. R. The circular economy. Nature 531, 435–438 (2016).

8

“Annual Energy Outlook 2018 with projections to 2050” by the U.S. Energy Information Administration,

2018. Available at the web page, “https://www.eia.gov/outlooks/aeo/”.
9

BP Statistical Review of World Energy June 2017.

10

Croce, R. & van Amerongen, H. Natural strategies for photosynthetic light harvesting. Nat. Chem. Biol. 10,

492-501 (2014).
11

Nelson, N. & Ben-She, A. The complex architecture of oxygenic photosynthesis. Nat. Rev. Mol. Cell. Biol. 5,

971–982 (2004).
12

McEvoy, J. P. & Brudvig, G. W. Water-splitting chemistry of photosystem II. Chem. Rev. 106, 4455–4483

(2006).
13

Grätzel, M. Artificial photosynthesis: water cleavage into hydrogen and oxygen by visible light. Acc. Chem.

Res. 14, 376– 384 (1981)
14

Guan, X. et al. Making of an industry-friendly artificial photosynthesis device. ACS Energy Lett. 3, 2230–

2231 (2018).
15

Lewis, N. S. Research opportunities to advance solar energy utilization. Science 351, 353–361 (2016).

16

Nocera, D. G. Solar fuels and solar chemicals industry. Acc. Chem. Res. 50, 616–619 (2017).

17

Roger, I., Shipman, M. A. & Symes, M. D. Earth-abundant catalysts for electrochemical and

photoelectrochemical water splitting. Nat. Rev. Chem. 1, 301–313 (2017).
18

Arno de Klerk (2013). "Fischer–Tropsch Process". Kirk-Othmer Encyclopedia of Chemical Technology.

Weinheim: Wiley-VCH.
19

Yu, K. M. K., Curcic, I., Gabriel, J. & Tsang, S. C. E. Recent advances in CO2 capture and utilization.

ChemSusChem, 1, 893–899 (2008).

20

Oleksandr S. B. et al. What should we make with CO2 and how can we make it? Joule 2, 825–832 (2018).

21

www.energy-democracy.net/. Accessed on August 17th, 2018.

22

Ferreira, K. N., Iverson, T. M., Maghlaoui, K., Barber, J. & Iwata, S. Architecture of the photosynthetic

oxygen-evolving center. Science 303, 1831–1838 (2004).
23

Barber, J. Mn4Ca cluster of photosynthetic oxygen-evolving center: structure, function and evolution.

Biochemistry 55, 5901–5906 (2016).
24

Umena, Y., Kawakami, K., Shen, J.-R. & Kamiya, N. Crystal structure of oxygen-evolving photosystem II at a

resolution of 1.9 Å. Nature 473, 55–60 (2011)
25

Suga, M. et al. Native structure of photosystem II at 1.95 Å resolution viewed by femtosecond X-ray

pulses. Nature 517, 99–103 (2015)
26

Klauss, A., Haumann, M. & Dau, H. Alternating electron and proton transfer steps in photosynthetic water

oxidation. Proc. Natl Acad. Sci. USA 109, 16035–16040 (2012).
27

Krewald, V. et al. Metal oxidation states in biological water splitting. Chem. Sci. 6, 1676–1695 (2015).

28

Rappaport, F., Guergova-Kuras, M., Nixon, P. J., Diner, B. A. & Lavergne, J. Kinetics and pathways of charge

recombination in photosystem II. Biochemistry 41, 8518–8527 (2002).
29

Cox, N., Pantazis, D. A., Neese, F. & Lubitz, W. Biological water oxidation. Acc. Chem. Res. 46, 1588–1596

(2013).
30

Vogt, L., Vinyard, D. J., Khan, S. & Brudvig, G. W. Oxygen-evolving complex of photosystem II: an analysis

of second-shell residues and hydrogen-bonding networks. Curr. Opin. Chem. Biol. 25, 152–158 (2015).
31

Vass, I. Molecular mechanisms of photodamage in the photosystem II complex. Biochim. Biophys. Acta

1817, 209–217 (2012).
32

Hideg, E., Kalai, T., Hideg, K. & Vass, I. Photoinhibition of photosynthesis in vivo results in singlet oxygen

production detection via nitroxide-induced fluorescence quenching in broad bean leaves. Biochemistry 37,
11405–11411 (1998).
33

Moyer, B. A. & Meyer, T. J. Proton-coupled electron transfer between [Ru(bpy)2(py)OH2]2+ and

[Ru(bpy)2(py)O]2+. A solvent isotope effect (kH2O/kD2O) of 16.1. J. Am. Chem. Soc. 100, 3601–3603 (1978)
34

Weinberg, D. R. et al. Proton-coupled electron transfer. Chem. Rev. 112, 4016–4093 (2012).

35

Gagliardi, C. J., Vannucci, A. K., Concepcion, J. J., Chen, Z. & Meyer, T. J. The role of proton coupled

electron transfer in water oxidation. Energy Environ. Sci. 5, 7704–7717 (2012).
36

Savéant, J.-M. Electrochemical approach to proton-coupled electron transfers: recent advances. Energy

Environ. Sci. 5, 7718–7731 (2012).
37

Glöckner, C. et al. Structural changes of the oxygen-evolving complex in photosystem II during the

catalytic cycle. J. Biol. Chem. 288, 22607–22620 (2013).
38

Young, I. D. et al. Structure of photosystem II and substrate binding at room temperature. Nature 540,

453–457 (2016).

39

Suga, M. et al. Light-induced structural changes and the site of O=O bond formation in PSII caught by XFEL.

Nature 543, 131–135 (2017).
40

Cox, N. et al. Electronic structure of the oxygen-evolving complex in photosystem II prior to O–O bond

formation. Science 345, 804–808 (2014).
41

Lohmiller, T. et al. The first state in the catalytic cycle of the water-oxidizing enzyme: identification of a

water-derived μ-hydroxo bridge. J. Am. Chem. Soc. 139, 14412–14424 (2017).
42

McCrory, C. C. L. et al. Benchmarking hydrogen evolving reaction and oxygen evolving reaction

electrocatalysts for solar water splitting devices. J. Am. Chem. Soc. 137, 4347–4357 (2015).
43

Smith, R. D. L. et al. Photochemical route for accessing amorphous metal oxide materials for water

oxidation catalysis. Science, 340, 60–63 (2013).
44

Godwin, I., Rovetta, A., Lyons, M. & Coleman, J. Electrochemical water oxidation: The next five years. Curr.

Opin. Electrochem. 7, 31–35 (2018).
45

Blakemore, J. D., Crabtree, R. H. & Brudvig, G. W. Molecular catalysts for water oxidation. Chem. Rev. 115,

12974–13005 (2015).
46

Berardi, S. et al. Molecular artificial photosynthesis. Chem. Soc. Rev. 43, 7501–7519 (2014).

47

Garrido-Barros, P., Gimbert-Suriñach, C., Matheu, R., Sala, X. & Llobet, A. How to make an efficient and

robust molecular catalyst for water oxidation. Chem. Soc. Rev. 46, 6088–6098 (2017).
48

Coehn, A. & Gläser, M. Studien über die Bildung von Metalloxyden. 1. Über das anodische Verhalten von

Kobalt- und Nickel-Lösungen. Zeits. Anorg. Chem. 33, 9–24 (1903).
49

Gersten, S. W., Samuels, G. J. & Meyer, T. J. Catalytic oxidation of water by an oxo-bridged ruthenium

dimer. J. Am. Chem. Soc. 104, 4029–4030 (1982).
50

Elizarova, G. L., Matvienko, L. G., Parmon, V. N. & Zamaraev, K. I. Catalysts of the oxidation of water to

molecular oxygen based on binuclear complexes of iron(III) and other compounds of transition metals. Dokl.
Akad. Nauk SSSR, 249, 863–866 (1979).
51

Elizarova, G. L., Matvienko, L. G., Lozhkina, N. V., Parmon, V. N. & Zamaraev, K. I. Homogeneous catalysts

for dioxygen evolution from water. Water oxidation by trisbipyridylruthenium(III) in the presence of cobalt,
iron and copper complexes. React. Kinet. Catal. Lett. 16, 191–194 (1981).
52

Gilbert, J. A. et al. Structure and redox properties of the water-oxidation catalyst

[(bpy)2(OH2)RuORu(OH2)(bpy)2]4+. J. Am. Chem. Soc. 107, 3855–3864 (1985).
53

Liu, F. et al. Mechanisms of water oxidation from the blue dimer to Photosystem II. Inorg. Chem. 47, 1727–

1752 (2008).
54

Geselowitz, D. & Meyer, T. J. Water oxidation by [(bpy)2(O)RuVORuV(O)(bpy)2]4+. An oxygen-labeling study.

Inorg. Chem. 29, 3894–3896 (1990).

55

Yamada, H., Siems, W. F., Koike, T. & Hurst, J. K. Mechanisms of water oxidation catalyzed by the cis,cis-

[(bpy)2Ru(OH2)]2O4+ ion. J. Am. Chem. Soc. 126, 9786–9795 (2004).
56

Cape, J. L. & Hurst, J. K. Detection and mechanistic relevance of transient ligand radicals formed during

[Ru(bpy)2(OH2)]2O4+-catalyzed water oxidation. J. Am. Chem. Soc. 130, 827–829 (2008).
57

Chronister, C. W., Binstead, R. A., Ni, J. & Meyer, T. J. Mechanism of water oxidation catalyzed by the μ-

oxo dimer [(bpy)2(OH2)RuIIIORuIII(OH2)(bpy)2]4+. Inorg. Chem. 36, 3814–3815 (1997).
58

Parent, A. R., Crabtree, R. H. & Brudvig, G. W. Comparison of primary oxidants for water-oxidation

catalysis. Chem. Soc. Rev. 42, 2247–2252 (2013).
59

Binstead, R. A., Chronister, C. W., Ni, J., Hartshorn, C. M. & Meyer, T. J. Mechanism of water oxidation by

the μ-oxo dimer [(bpy)2(H2O)RuIIIORuIII(OH2)(bpy)2]4+. J. Am. Chem. Soc. 122, 8464–8473 (2000).
60

Moonshiram, D. et al. Structure and electronic configurations of the intermediates of water oxidation in

blue ruthenium dimer catalysis. J. Am. Chem. Soc. 134, 4625–4636 (2012).
61

Costentin, C., Drouet, S., Robert, M., Savéant, J.-M. Turnover Numbers, Turnover Frequencies, and

Overpotential in Molecular Catalysis of Electrochemical Reactions. Cyclic Voltammetry and Preparative-Scale
Electrolysis. J. Am. Chem. Soc. 134, 11235-11242 (2012).
62

Rotzinger, F. P. et al. A molecular water-oxidation catalyst derived from ruthenium diaqua bis(2,2′-

bipyridyl-5,5′-dicarboxylic acid). J. Am. Chem. Soc. 109, 6619–6626 (1987).
63

Nazeeruddin, M. K., Rotzinger, F. P., Comte, P. & Grätzel, M. Spontaneous oxidation of water to oxygen by

the mixed-valence µ-oxo ruthenium dimer L2(H2O)Ru–O–Ru(OH)L2 (L = 2,2′-bipyridyl-5,5′-dicarboxylic acid).
J. Chem. Soc., Chem. Commun. 872–874 (1988).
64

Sens, C. et al. A new Ru complex capable of catalytically oxidizing water to molecular dioxygen. J. Am.

Chem. Soc. 126, 7798–7799 (2004).
65

Romain, S., Bozoglian, F., Sala, X. & Llobet, A. Oxygen−oxygen bond formation by the Ru-Hbpp water

oxidation catalyst occurs solely via an intramolecular reaction pathway. J. Am. Chem. Soc. 131, 2768–2769
(2009).
66

Bozoglian, F. et al. The Ru-Hbpp water oxidation catalyst. J. Am. Chem. Soc. 131, 15176–15187 (2009).

67

Romain, S., Vigara, L. & Llobet, A. Oxygen–oxygen bond formation pathways promoted by Ru complexes.

Acc. Chem. Res. 42, 1944–1953 (2009).
68

Neudeck, S. et al. New powerful and oxidatively rugged dinuclear Ru water oxidation catalyst: control of

mechanistic pathways by tailored ligand design. J. Am. Chem. Soc. 136, 24–27 (2014).
69

Catalano, V. J. et al. Monometallic and dimetallic ruthenium(II)−terpyridine complexes employing the

tetradentate ligands dipyridylpyrazolyl, dipyridyloxadiazole, and their dimethyl derivatives. Inorg. Chem. 42,
321–334 (2003).

70

Zong, R. & Thummel, R. P. A new family of Ru complexes for water oxidation. J. Am. Chem. Soc. 127,

12802–12803 (2005).
71

Concepcion, J. J., Jurss, J. W., Templeton, J. L. & Meyer, T. J. One site is enough. Catalytic water oxidation

by [Ru(tpy)(bpm)(OH2)] 2+ and [Ru(tpy)(bpz)(OH2)]2+. J. Am. Chem. Soc. 130, 16462–16463 (2008).
72

Masaoka, S. & Sakai, K. Evidence showing the robustness of a highly active oxygen-evolving mononuclear

ruthenium complex with an aqua ligand. Chem. Lett. 38, 182–183 (2009).
73

Wasylenko, D.J. et al. Electronic modification of the [RuII(tpy)(bpy)(OH2)]2+ scaffold: effects on catalytic

water oxidation. J. Am. Chem. Soc. 132, 16094–16106 (2010).
74

Maji, S., López, I., Bozoglian, F., Benet-Buchholz, J. & Llobet, A. Mononuclear ruthenium–water oxidation

catalysts: discerning between electronic and hydrogen-bonding effects. Inorg. Chem. 52, 3591–3593 (2013).
75

Concepcion, J. J. et al. Catalytic water oxidation by single-site ruthenium catalysts. Inorg. Chem. 49, 1277–

1279, (2010).
76

Tong, L. & Thummel, R. P. Mononuclear ruthenium polypyridine complexes that catalyze water oxidation.

Chem. Sci. 7, 6591–6603 (2016).
77

Cytochrome P450: Structure, Mechanism, and Biochemistry (Ed. de Montellano, P. R. O.) (Kluwer

Academic/Plenum, New York, ed. 3, 2005).
78

Rittle, J. & Green M. T. Cytochrome P450 compound I: capture, characterization, and C–H bond activation

kinetics. Science 330, 933–937 (2010).
79

Groves, J. T. Cytochrome P450 enzymes: understanding the biochemical hieroglyphs. F1000Research

2015, 4(F1000 Faculty Rev):178 Last updated: 25 DEC 2016.
80

López, I., Maji, S., Benet-Buchholz, J. & Llobet, A. Oxo-bridge scenario behind single-site water-oxidation

catalysts. Inorg. Chem. 54, 658–666 (2015).
81

Earley, J. E., Smith, P. M., Fealey, T. & Silverton, J. V. Crystal and molecular structure of di-mu-oxo-

bis(pentaammineruthenium)bis(ethylenediamine)ruthenium hexachloride. Ethylenediamine analog of
"ruthenium red". Inorg. Chem. 10, 1943–1947 (1971).
82

Tsubonouchi, Y., Lin, S., Parent, A.R., Brudvig, G.W. & Sakai, K. Light-induced water oxidation catalyzed by

an oxido-bridged triruthenium complex with a Ru–O–Ru–O–Ru motif. Chem. Commun. 52, 8018–8021
(2016).
83

López, I. et al. A self-improved water-oxidation catalyst: is one site really enough? Angew. Chem. Int. Ed.

53, 205–209 (2014).
84

Radaram, B. et al. Water oxidation by mononuclear ruthenium complexes with TPA-based ligands. Inorg.

Chem. 50, 10564–10571 (2011).
85

Francas, L. et al. Synthesis, structure, and reactivity of new tetranuclear Ru-Hbpp-based water-oxidation

Catalysts. Inorg. Chem. 50, 2771–2781 (2011).

86

Sander, A.C., Schober, A., Dechert, S. & Meyer, F. A Pyrazolate-bridged bis(pentadentate) ligand and its

dinuclear ruthenium complex. Eur. J. Inorg. Chem. 2015, 4348–4353 (2015).
87

Garrido-Barros, P. et al. In Water as an oxygen source for oxidation reactions in Catalytic Oxidations in

Organic Synthesis. (Ed. Muniz, K.) (Georg Thieme Verlag KG, Stuttgart, New York, 2017).
88

Harriman, A., Pickering, I. J., Thomas, J. M. & Christensen, P. A. Metal oxides as heterogeneous catalysts

for oxygen evolution under photochemical conditions. J. Chem. Soc., Faraday Trans. 84, 2795–2806 (1988).
89

Risch, M. et al. Cobalt-oxo core of a water-oxidizing catalyst film. J. Am. Chem. Soc. 131, 6936–6937

(2009).
90

Kanan, M. W. et al. Structure and valency of a cobalt−phosphate water oxidation catalyst determined by

in situ X-ray spectroscopy. J. Am. Chem. Soc., 132, 13692–13701 (2010).
91

Elizarova, G. L., Matvienko, L. G. & Parmon, V. N. Oxidation of water by Ru(bpy)3+ in the presence of Co(III)

and Fe(III) colloidal hydroxides as catalysts. J. Mol. Catal. 43, 171–181 (1987).
92

Kim, T. V. Elizarova, G. L. & Parmon, V. N. Catalytic oxidation of water to dioxygen by Ru(bpy)3+ in the

presence of mixed iron and cobalt hydroxides. React. Kinet. Catal. Lett. 26, 57–60 (1984).
93

Hull, J. F. et al. Highly active and robust Cp* iridium complexes for catalytic water oxidation. J. Am. Chem.

Soc. 131, 8730–8731 (2009).
94

Zhao, Y. et al. Stable iridium dinuclear heterogeneous catalysts supported on metal oxide substrate for

solar water oxidation. Proc. Natl Acad. Sci. U.S.A. 115, 2902–2907 (2018).
95

Rodriguez, G.M., Gatto, G., Zuccaccia, C. & Machioni, A. Benchmarking water oxidation catalysts based on

iridium complexes: clues and doubts on the nature of active species. ChemSusChem 10, 4503–4509 (2017).
96

Savini, A. et al. Activity and degradation pathways of pentamethyl-cyclopentadienyl-iridium catalysts for

water oxidation. Green Chem., 13, 3360–3374 (2011).
97

Duan, L., Fischer, A., Xu, Y. & Sun, L. Isolated seven-coordinate Ru(IV) dimer complex with [HOHOH]−

bridging ligand as an intermediate for catalytic water oxidation. J. Am. Chem. Soc. 131, 10397–10399 (2009).
98

Duan, L. et al. A molecular ruthenium catalyst with water-oxidation activity comparable to that of

photosystem II. Nat. Chem. 4, 418–423 (2012).
99

Matheu, R. et al. The behavior of the Ru–bda water oxidation catalysts at low oxidation states. Chem. Eur.

J. 24, 12838–12847 (2018).
100

Nyhlen, J., Duan, L., Aakermark, B., Sun, L. & Privalov, T. Evolution of O2 in a seven-coordinate RuIV dimer

complex with a [HOHOH]− bridge: a computational study. Angew. Chem. Int. Ed. 49, 1773–1777 (2010).
101

Matheu, R. et. al. Hydrogen bonding rescues overpotential in seven-coordinated Ru water oxidation

catalysts. ACS Catal. 7, 6525–6532 (2017).
102

Duan, L., Araujo, C. M., Ahlquist, M. S. G. & Sun, L. Highly efficient and robust molecular ruthenium

catalysts for water oxidation. Proc. Natl Acad. Sci. U. S. A. 109, 15584–15588 (2012).
103

Richmond, C. J. et al.. Chem. Eur. J. 20, 17282–17286 (2014).

104

Wang, L., Duan, L., Wang, Y., Ahlquist, M. S. G. & Sun, L. Highly efficient and robust molecular water

oxidation catalysts based on ruthenium complexes. Chem. Commun. 50, 12947–12950 (2014).
105

Xie, Y., Shaffer, D. W. & Concepcion, J. J. O–O radical coupling: from detailed mechanistic understanding

to enhanced water oxidation catalysis. Inorg. Chem. 57, 10533–10542 (2018).
105

THIS REFERENCE SHUOULD BE REMOVED

107

Liu, Y. et al. Catalytic water oxidation by ruthenium(II) quaterpyridine (qpy) complexes: evidence for

ruthenium(III) qpy-N,N′′′-dioxide as the real catalysts. Angew. Chem. Int. Ed. 53, 14468–14471 (2014).
108

Matheu, R. et al. Intramolecular proton transfer boosts water oxidation catalyzed by a Ru complex. J. Am.

Chem. Soc. 137, 10786–10795 (2015).
109

Matheu, R., Neudeck, S., Meyer, F., Sala, X. & Llobet, A. Foot of the wave analysis for mechanistic

elucidation and benchmarking applications in molecular water oxidation catalysis. ChemSusChem 9, 3361–
3369 (2016).
110

Duan, L., Wang, L., Li, F., Li, F. & Sun, L. Highly efficient bioinspired molecular Ru water oxidation catalysts

with negative charged backbone ligands. Acc. Chem. Res. 48, 2084–2096 (2015).
111

Govindarajan, N., Tiwari, A., Ensing, B. & Meijer, E. J. Impact of the ligand flexibility and solvent on the O–

O bond formation step in a highly active ruthenium water oxidation catalyst. Inorg. Chem. 57, 13063–13066
(2018).
112

Gao, Y. et al. Visible light driven water splitting in a molecular device with unprecedentedly high

photocurrent density. J. Am. Chem. Soc. 135, 4219–4222 (2013).
113

Li, F. et al. Highly efficient oxidation of water by a molecular catalyst immobilized on carbon nanotubes.

Angew. Chem. Int. Ed. 50, 12276–12279 (2011).
114

Sheridan, M.V. et al. Evaluation of chromophore and assembly design in light-driven water splitting with a

molecular water oxidation catalyst. ACS Energy Lett. 1, 231–236 (2016).
115

Creus, J. et al. A million turnover molecular anode for catalytic water oxidation. Angew. Chem. Int. Ed. 55,

15382–15386 (2016).
116

Wang, D. et al. Interfacial deposition of Ru(II) bipyridine-dicarboxylate complexes by ligand substitution

for applications in water oxidation catalysis. J. Am. Chem. Soc. 140, 719–726 (2018).
117

Matheu, R. et al. Photoelectrochemical behavior of a molecular Ru-based water-oxidation catalyst bound

to TiO2-protected Si photoanodes. J. Am. Chem. Soc. 139, 11345–11348 (2017).
118

Grau, S. et al. A hybrid molecular photoanode for efficient light induced water oxidation. Sustainable

Energy Fuels 2, 1979–1985 (2018).
119

Rapaport, I., Helm, L., Merbach, A. E., Bernhard, P. & Ludi, A. High-pressure NMR kinetics. Part 34.

Variable-temperature and variable-pressure NMR kinetic study of solvent exchange on
hexaaquaruthenium(3+) and -(2+) and hexakis(acetonitrile)ruthenium(2+). Inorg. Chem. 27, 873–879 (1988).

120

Helm, L. & Merbach, A. E. Inorganic and bioinorganic solvent exchange mechanisms. Chem. Rev. 105,

1923–1960 (2005).
121

Draksharapu, A. et al. Ligand exchange and spin state equilibria of FeII(N4Py) and related complexes in

aqueous media. Inorg. Chem. 51, 900–913 (2012).
122

Hong, D. et al. Water oxidation catalysis with nonheme iron complexes under acidic and basic conditions:

homogeneous or heterogeneous? Inorg. Chem. 52, 9522–9531 (2013).
123

Wang, D., Ghirlanda, G. & Allen, J. P. Water oxidation by a nickel–hlycine catalyst. J. Am. Chem. Soc. 136,

10198–10201 (2014).
124

Kanan, M. W. & Nocera, D. G. In situ formation of an oxygen-evolving catalyst in neutral water containing

phosphate and Co2+. Science 321, 1072–1075 (2008).
125

Singh, A., Chang, S. L. Y., Hocking, R. K., Bach, U. & Spiccia, L. Highly active nickel oxide water oxidation

catalysts deposited from molecular complexes. Energy Environ. Sci. 6, 579–586 (2013).
126

Singh, A., Chang, S. L. Y., Hocking, R. K., Bach, U. & Spiccia, L. Anodic deposition of NiOx water oxidation

catalysts from macrocyclic nickel(II) complexes. Catal. Sci. Technol. 3, 1725–1732 (2013).
127

Brimblecombe, R., Kolling, D. R. J., Bond, A. M., Dismukes, G. C., Swiegers, G. F.& Spiccia, L. Sustained

water oxidation by [Mn4O4]7+ core complexes inspired by oxygenic photosynthesis. Inorg. Chem. 48, 7269–
7279 (2009).
128

Hocking, R. K. et al. Water-oxidation catalysis by manganese in a geochemical-like cycle. Nat. Chem. 3,

461–466 (2011).
129

Kuznetsov, D. A. et al. Ni-based heterogeneous catalyst from a designed molecular precursor for the

efficient electrochemical water oxidation. Chem. Commun. 52, 9255–92558 (2016).
130

Lai, Y.-H., King, T. C., Wright, D. S. & Reisner, E. Scalable one-step assembly of an inexpensive

photoelectrode for water oxidation by deposition of a Ti- and Ni-containing molecular precursor on
nanostructured WO3. Chem. Eur. J. 19, 12943–12947(2013).
131

Lai, Y. H., Kato, M., Mersch, D. & Reisner, E. Comparison of photoelectrochemical water oxidation activity

of a synthetic photocatalyst system with photosystem II. Faraday Discuss. 176, 199–211 (2014).
132

Lai, Y.-H., Palm, D. W. & Reisner, E. Multifunctional coatings from scalable single source precursor

chemistry in tandem photoelectrochemical water splitting. Adv. Energy Mater. 5, 1501668 (2015).
133

Barnett, S. M., Goldberg, K. I. & Mayer, J. M. A soluble copper–bipyridine water-oxidation electrocatalyst.

Nat. Chem. 4, 498–502 (2012).
134

Gerlach, D. L. et al. Studies of the pathways open to copper water oxidation catalysts containing proximal

hydroxy groups during basic electrocatalysis. Inorg. Chem. 53, 12689−12698 (2014).
135

Zhang, T., Wang, C., Liu, S., Wang, J.-L. & Lin, W. A biomimetic copper water oxidation catalyst with low

overpotential. J. Am. Chem. Soc. 136, 273–281 (2014).

136

Fisher, K. J., Materna, K. L., Mercado, B. Q., Crabtree, R. H. & Brudvig, G. W. Electrocatalytic water

oxidation by a copper(II) complex of an oxidation-resistant ligand. ACS Catal. 7, 3384–3387 (2017).
137

Wasylenko, D. J., Ganesamoorthy, C., Borau-Garcia, J. & Berlinguette, C. P. Electrochemical evidence for

catalytic water oxidation mediated by a high-valent cobalt complex. Chem. Commun. 47, 4249–4251 (2011).
138

Rigsby, M. L. Cobalt analogs of Ru-based water oxidation catalysts: overcoming thermodynamic instability

and kinetic lability to achieve electrocatalytic O2 evolution. Chem. Sci. 3, 3058–3062 (2012).
139

Gimbert-Suriñach, C. et al. Structural and spectroscopic characterization of reaction intermediates

involved in a dinuclear Co–Hbpp water oxidation catalyst. J. Am. Chem. Soc. 138, 15291–15294 (2016).
140

Han, Y., Wu, Y., Lai, W. & Cao, R. Electrocatalytic water oxidation by a water-soluble nickel porphyrin

complex at neutral pH with low overpotential. Inorg. Chem. 54, 5604–5613 (2015).
141

Garrido-Barros, P. et al. Redox non-innocent ligand controls water oxidation overpotential in a new family

of mononuclear Cu-based efficient catalysts. J. Am. Chem. Soc. 137, 6758–6761 (2015).
142

Funes-Ardoiz, I., Garrido-Barros, P., Llobet, A. & Maseras, F. Single electron transfer steps in water

oxidation catalysis. Redifining the mechanistic scenario. ACS Catal. 7, 1712–1719 (2017).
143

Wang, H.-Y., Mijangos, E., Ott, S. & Thapper, A. Water oxidation catalyzed by a dinuclear cobalt–

polypyridine complex. Angew. Chem. Int. Ed. 53, 14499–14502 (2014).
144

Ishizuka, T. et al. Homogeneous photocatalytic water oxidation with a dinuclear CoIII−pyridylmethylamine

complex. Inorg. Chem. 55, 1154−1164 (2016).
145

Wang, J.-W., Sahoo, P. & Lu, T.-B. Reinvestigation of water oxidation catalyzed by a dinuclear cobalt–

polypyridine complex: identification of CoOx as real heterogeneous catalyst. ACS Catal. 6, 5062–5068 (2016)
146

Zhang, M., Zhang, M.-T., Hou, C., Ke, Z.-F. & Lu, T.-B. Homogeneous electrocatalytic water oxidation at

neutral pH by a robust macrocyclic nickel(II) complex. Angew. Chem. Int. Ed. 53, 13042–13048 (2014).
147

Wang, J.-W. Further insight into the electrocatalytic water oxidation by macrocyclic nickel(II) complexes:

the influence of steric effect on catalytic activity. Catal. Sci. Technol. 7, 5585–5593 (2017).
148

Najafpour, M. M. & Feizi, H. Water oxidation by Ni(1,4,8,11-tetraazacyclotetradecane)2+ in the presence

of carbonate: new findings and an alternative mechanism. Dalton Trans. 47, 6519–6527 (2018).
149

Pizzolato, E. et al. light driven water oxidation by a single site cobalt salophen catalyst. Chem. Commun.

49, 9941–9943 (2013).
150

Najafpour, M. M. & Feizi, H. Water oxidation catalyzed by two cobalt complexes: new challenges and

questions. Catal. Sci. Technol. 8, 1840–1848 (2018).