"This is the peer reviewed version of the following article: Quantum Chemical Study of the
Mechanism of Water Oxidation Catalyzed by a Heterotrinuclear Ru2Mn Complex, which has
been published in final form at:

https://onlinelibrary.wiley.com/doi/10.1002/cssc.201802395

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

Quantum Chemical Study of the Mechanism of Water Oxidation
Catalyzed by a Heterotrinuclear Ru2Mn Complex

Ying-Ying Lia, Carolina Gimbertb, Antoni Llobetb, Per E. M. Siegbahnc, Rong-Zhen Liaoa*

aKey

Laboratory of Material Chemistry for Energy Conversion and Storage, Ministry

of Education, Hubei Key Laboratory of Bioinorganic Chemistry and Materia Medica,
Hubei Key Laboratory of Materials Chemistry and Service Failure, School of
Chemistry and Chemical Engineering, Huazhong University of Science and
Technology, Wuhan 430074, China
bInstitute

of Chemical Research of Catalonia (ICIQ-BIST), Avinguda Països Catalans

16, 43007 Tarragona, Spain
cDepartment

of Organic Chemistry, Arrhenius Laboratory, Stockholm University,

Stockholm 10691, Sweden
*Corresponding author: rongzhen@hust.edu.cn

1

Abstract
The heterotrinuclear complex A {[RuII(H2O)(trpy)]2(μ-[MnII(H2O)2(bpp)2])}4+ was
disclosed to catalyze water oxidation both electrochemically and photochemically
with [Ru(bpy)3]3+ as the photosensitizer and Na2S2O8 as the electron acceptor in
neutral phosphate buffer. The mechanism of water oxidation catalyzed by this
unprecedented trinuclear complex was studied via density functional calculations. The
calculations showed that a series of oxidation and deprotonation events take place
from A, leading to the formation of complex 1 (formal oxidation state of
Ru1IVMnIIIRu2III), which is the starting species for the catalytic cycle. Three sequential
oxidations of 1 result in the generation of the catalytically competing species 4
(formal oxidation state of Ru1IVMnVRu2IV) that triggers the O-O bond formation. The
direct coupling of two adjacent oxo ligands bound to Ru and Mn leads to the
production of a superoxide intermediate Int1. This step was calculated to have a
barrier of 7.2 kcal mol-1 at the B3LYP*-D3 level. Subsequent O2 release from Int1
turns out to be quite facile. Other possible pathways were found to be much less
favorable, including water nucleophilic attack, the coupling of an oxo and a hydroxide,
and the direct coupling pathway at a lower oxidation state (Ru IVMnIVRuIV).
Keywords: water oxidation; mechanism; density functional calculations; ruthenium;
manganese; trinuclear.

2

1. Introduction
A promising scheme to convert solar energy into chemical energy is the water
splitting driven by sunlight. Water oxidation occurs at a potential of 1.23 V vs the
standard hydrogen electrode (SHE) and is both thermodynamically and kinetically
very unfavorable. In photosystem II, a Mn4Ca cluster in the oxygen evolution center
(OEC) is known to catalyze the oxidation of water into molecular dioxygen. 1-5 During
the past few decades, extraordinary efforts have been dedicated to the synthesis of
manganese-based molecular complexes to mimic the structure and function of OEC. 610

Other transition metals have also been used for the synthesis of water oxidation

catalyst, including Ru,11-24 Ir,25-26 Fe,27-31 Co,32-34 Ni,35-37 and Cu.38-40
Recently, the Llobet group disclosed the synthesis of a heterotrinuclear Ru2Mn
complex A0 {[RuII(trpy)]2(μ-[MnII(bpp)2])(OAc)2}2+.41 This pre-catalyst undergoes
water/acetate ligand exchange to form complex A {[RuII(H2O)(trpy)]2(μ-[MnII
(H2O)2(bpp)2])}4+ (Figure 1) that is capable of mediating water oxidation both
electrochemically and photochemically.41 The photochemical reaction was performed
using [Ru(bpy)3]2+ as the photosensitizer and Na2S2O8 as the sacrificial electron
acceptor in neutral phosphate buffer. A turnover number of 8 was obtained in about
10 minutes. For the electrochemical reaction, the differential pulse voltammetry (DPV)
study of A displayed a catalytic onset potential of 1.5 V. 41 A related CoRu2 complex
was also reported, and the TON was estimated to be 50 using the same
photochemical water oxidation condition. Since the complex water oxidation process
involves the removal of four protons and four electrons, this trinuclear complex
represents an elegant example that the multiple redox-active metal centers can
cooperate during the four electron transfer processes. In addition, all three metals
possess one or two aqua ligands. As a consequence, the high oxidation states can be
accessed more easily via proton-coupled electron transfer.
To improve the efficiency of water oxidation catalysts and design new catalysts, a
detailed mechanistic understanding of the reaction mechanism is of great importance.
In recent years, quantum chemical calculations have been successfully applied to
3

elucidate the catalytic mechanism for a large number of homogeneous water
oxidation catalysts.42 In this paper, we carried out quantum chemical calculations to
delineate the mechanism of water oxidation catalyzed by this trinuclear Ru2Mn
complex. Various possible O-O bond formation pathways have been considered and
compared. This study provides a particular opportunity for understanding how three
transition metal centers cooperate in water oxidation.

Figure 1. Optimized structures of the trinuclear RuII2MnII complex (A, sextet). Distances are given in
Å. Spin densities for selected atoms are indicated in italics. Unimportant hydrogen atoms are not
shown for clarity.

2. Computational details
The quantum chemical calculations were performed using the Gaussian 09 program
package.43 All geometries were optimized with the B3LYP-D3 functional (including
Grimme’s D3 dispersion)44,45 in combination with the SDD46 pseudopotential for Ru
and Mn as well as 6-31G (d,p) basis set for the C, H, O and N elements. The final and
the solvation energies in water solution were obtained by performing single-point
calculations on the optimized geometries using the SMD 47 continuum solvation model
at the B3LYP*-D3 (with 15% Hartree-Fock exchange, including D3 dispersion and
Gibbs free energy corrections from B3LYP-D3)48 level with a larger basis set, where
6-311+G(2df,2p) were used for all atoms except for Ru and Mn (SDD). The harmonic
frequencies were computed to characterize the nature of various stationary points as
either minima or transition states, and also to obtain the Gibbs free energy
4

corrections. Singlet-point calculations were performed at the B3LYP-D3, 45 M06-D3,49
M06L-D3,50 TPSSh-D3,49 TPSS0-D3(25% Hartree-Fock exchange using TPSSTPSS
and keyword IOp(3/76=0750002500)),51-52 and TPSSTPSS-D353 levels to assess the
sensitivities of redox potentials and barriers to the choice of functionals.
The pKa values were calculated in order to determine the protonation state of all
species at the working pH=7, and the experimental solvation energy of -264.0 kcal
mol-1 of a proton was used.54 The Gibbs free energy correction of -6.3 kcal mol-1
derived from the translational entropy of a proton in the gas phase was added. 54 The
experimental solvation free energies of water (-6.3 kcal mol-1)54 and acetate (-77.0
kcal mol-1)55,56 in aqueous solution were used. The calculated solvation free energies
of water and acetate are -7.5 kcal mol-1 and -69.2 kcal mol-1, respectively. The
experimental values for the solvation free energies of water and free anions (chloride)
has also been used in our previous study on an iron-catalyzed water oxidation
reaction.57 Importantly, the use of either the experimental values or the calculated
values does not alter any mechanistic conclusion. For all species except water, the
concentration correction of 1.9 kcal mol-1 at room temperature in water solution was
added, which was derived from the change of the standard state from 1 atm (24.5 L
mol-1, 298.15 K) for the ideal gas to 1 M (1 mol L-1) in water solution. For water, a
value of 4.3 kcal mol-1 was used as its standard state is 55.6 M. For the calculations of
the

redox

potentials,

the

experimental

absolute

potential

of

the

[Ru(bpy)2]3+/[Ru(bpy)2]2+ couple (1.26+4.281 V) was used as a reference,58 which
corresponds to an electron affinity of 127.8 kcal mol-1. This value was used to set up
the free energy diagram for the whole catalytic cycle. This kind of methodology has
been documented in details in our recent review 42 and has been successfully applied
to the study of many different molecular water oxidation catalysts. 16, 59-65

5

3. Results and discussion
3.1 Redox potentials
Our investigation starts from the conversion of A0 (Figure S1) to A (Ru2IIMnII) by
the displacement of two acetates with four water molecules. This process was
calculated to be endergonic by 3.9 kcal mol-1. A was calculated to be a sextet, which
has a high-spin MnII (SMn=5/2) and two low-spin RuII (SRu=0). This is consistent with
the EPR and magnetic susceptibility results obtained for A0 in the acetonitrile solution,
which showed a six-line signal characteristic of a high-spin Mn II ion.41 The pKas of A
and its deprotonated form Adp were calculated to be 5.1 and 5.6, respectively. Under
the working condition of pH=7, two protons are released from the two Ru-bound
water molecules in A to generate A2dp (Figure S1). The formation of A2dp from A0 is
thus exergonic by 0.6 kcal mol-1. In A2dp, both Mn-bound water molecules form
hydrogen bonds to the two Ru-bound hydroxides.
The one electron oxidation of A2dp (Ru2IIMnII) results in the generation of a septet
complex B (Ru1IIIMnIIRu2II, Scheme 1, Figure S1). The spin densities on Ru1 and Mn
are 0.83 and 4.82, respectively. The antiferromagnetic coupled quintet state (5α on
Mn and 1β on Ru1) is only 0.2 kcal mol-1 higher in energy. During the oxidation, the
electron is removed from one of the two RuII centers, rather than from the MnII center,
and the oxidation potential was calculated to be 0.30 V. When the redox potential of
the [Ru(bpy)3]3+/[Ru(bpy)3]2+ couple (1.26 V vs. SHE) was used as a reference, the
formation of complex B is exergonic by 22.1 kcal mol-1. This is followed by the one
electron oxidation of the other RuII center with a potential of 0.37 V, leading to the
formation of complex C (Ru1IIIMnIIRu2III, Figure S1). The high spin octet of C was
calculated to be the most stable, and the spin densities on Ru1, Mn, and Ru2 are 0.83,
4.85 and 0.83, respectively. The sextet and quartet states are 3.7 and 0.4 kcal mol -1
higher, respectively. From C, a proton-coupled electron transfer (PCET) oxidation
takes place to generate complex D (Ru1IIIMnIIIRu2III, Figure S1), associated with a
potential of 0.67 V. During the oxidation, the electron is removed from the Mn II center
and the proton is removed from the MnII-coordinated water molecule. D was
6

calculated to be a high-spin septet, and the spin densities on Ru1, Mn, and Ru2 are
0.82, 3.90 and 0.82, respectively. The quintet and triplet states were calculated to be
almost isogonic, being 0.2 kcal mol-1 and 0.5 kcal mol-1, respectively, higher than the
septet. From these calculations, it can be seen that in this complex the oxidation of
MnII is more difficult than that of RuII. The following oxidation also takes place first at
the RuIII center, rather than the MnIII center. The oxidation of D to generate 1
(Ru1IVMnIIIRu2III, Figure 2) was found to be a PCET process, with the proton released
from the Ru1-bound hydroxide. The redox potential was calculated to be 0.74 V. 1 is
an octet, and the spin densities on Ru1, Mn, Ru2 and O4 (Table S1) are 0.81, 3.92,
1.09 and 0.86, respectively. Consequently, a Ru1III-oxyl radical is produced. The sextet,
quartet and doublet are 11.8, 0.4 and 0.8 kcal mol-1 higher, respectively. In the sextet,
an intermediate-spin MnIII (2α) is presented, which makes its energy much higher
than the octet state with a high-spin Mn III (4α). Isomerization of 1 via proton transfer
from the MnIII bound water molecule to the Ru1III-oxyl radical to generate 1’
(Ru1IIIMnIVRu2III, Figure S3) with one MnIV center was found to be endergonic by 5.1
kcal mol-1.

Scheme 1. Conversion of A to 1. Other ligands are omitted for clarity.

Complex 1 was found to be the initial species for the whole catalytic cycle (vide
infra). The one electron oxidation of 1 to form 2 (Ru1IVMnIIIRu2IV, Figure 2) is
associated with the release of two protons, one from the Ru2III bound hydroxide and
the

other

from

the

MnIII

bound

water

7

molecule.

The

pKas

of

2dpt

[Ru1IV(OH)MnIII(OH)2Ru2IV(OH)]5+ and 2pt [Ru1IV(O)MnIV(OH)2Ru2III(OH)]4+ (Figure
S3) were calculated to be -1.7 and 6.6, respectively. The redox potential for this
oxidation was calculated to be 0.98 V. 2 is a nonet, and the spin densities of Ru1, Mn,
Ru2, O1 and O4 are 0.97, 3.88, 0.97, 0.97 and 0.97(Table S1), respectively. Therefore,
2 is best described as a high-spin MnIII (SMn=2) ferromagnetically coupled with two
low-spin RuIII-oxyl radicals (SRu1=SRu2=1/2, SO1=SO4=1/2). The septet, quintet, triplet
and broken-symmetry open-shell singlet states lie at +17.7, +12.8, +6.9 and +1.2 kcal
mol-1 higher, respectively. Isomer 2’ (Figure S3), with a proton transfer from the Mn
bound hydroxide to one of the oxyl radical in 2, is 5.8 kcal mol-1 higher than 2.
Next, in the conversion of 2 to 3 (Ru2IVMnIV, Figure 2), the Mn is oxidized from
MnIII to MnIV with a potential of 0.75 V, which is even lower than that of 1 to 2.
During the oxidation, no proton is released as the pKa of 3dp was calculated to be 14.8.
3 is an octet, and the spin density on Mn of 3 is 2.83. The sextet, quartet and doublet,
are 7.6, 12.8 and 7.1 kcal mol-1 higher, respectively. Similarly, with 2, proton transfer
from Mn bound hydroxide to one of the oxyls in 3 to generate 3’ (Figure S3) is
endergonic by 19.5 kcal mol-1.
Then, a PCET of 3 leads to the formation of a Ru1IVMnVRu2IV complex 4 (Figure 2),
and the redox potential for this oxidation step was calculated to be 1.54 V. Complex 4
is a septet and the spin densities on Ru1, Mn, Ru2, O1, O2 and O4 are 0.99, 1.05, 2.77,
0.69, -0.53 and 0.90, respectively. In 4, a MnIV is antiferromagnetically coupled with
an oxyl radical, which is similar to many other Mn-based water oxidation
catalysts,62,66-72 including the oxygen evolving complex in photosystem II.66-68 The
effect of spin coupling on the energy of 4 was calculated using the Noodleman
approach (see supporting information for details).73-75 The calculated J-value is 3.1
kcal/mol, and the inclusion of the Noodleman correction decreases the energy by
1.55 kcal/mol, which is very similar to the previous study on the dinuclear Mn water
oxidation catalyst.69 In this case, a negative spin is presented in this Mn-bound oxyl
radical while positive spin in the Ru-bound oxyl radicals. This kind of spin alignment
would favor the following O-O bond formation by the coupling of two oxyl radicals

8

with opposite spin densities. The nonet (with ferromagnetic coupling between Mn IV
and the oxyl radical), quintet and triplet states are 13.0, 6.4 and 0.5 kcal mol -1 higher,
respectively.
The DPV studies of complex A in neutral phosphate buffer gave three oxidation
peaks at approximately 0.60, 0.74, and 1.05 V.41 From the present calculations, these
three peaks can be assigned as the oxidation of C to D (0.67 V), the oxidation of D to 1
(0.74 V), and the tow electron oxidation of 1 to 3 (0.86 V), respectively. Finally, the
electrocatalytic wave is observed experimentally at 1.54 V that is associated with the
3 to 4 oxidation is calculated at 1.54 V. The calculated redox potentials are in very
good agreement with the experimentally reported and especially remarkable with
the 3 to 4 transition responsible for water oxidation catalysis. The lower oxidation
state transitions from A to B and B to C are not observed experimentally probably
due to the low quality of the electrochemical response for this complex. 41
It is worth mentioning here that the active species, derived from A, that are
responsible for the O-O bond formation, reach formal oxidation states Ru1 IVMnVRu2IV.
This is in good agreement with the OEC-PSII, where a formal Mn(V) is also needed in
the S4 state.66-68

9

Figure 2. Optimized structures of 1 (Octet), 2 (Nonet), 3 (Octet) and 4 (Septet). Distances are given
in Å. Spin densities on selected atoms are indicated in red italics. Unimportant hydrogen atoms are
not shown for clarity. For all structures, only the ground state is shown.
kcal/mol
O

+4.5

OH OH

+1.5
3-TS1

RuIII MnIII RuIII

0

0.0

2H+,e-

O

OH

OH
O OH

RuIII MnIII RuIII
OH

e0.75V

-11.8
4

H+,eO

-15
-20

Int1H

-6.5
O
2

-18.3
3
O

-11.1
TS1

1.54V RuIII MnIVRuIII

OH

-19.1
Int2

-19.2

OH O

TS2

O O

O

RuIII MnIII RuIII

-30

O

O-O

-14.8
TS3

RuIII MnIII RuIII

-22.6
Int2'

RuIII MnIVRuIII

-25

O

O

OH O

OH

RuIIIMnIVRuIII

-11.6 RuIIIMnIVRuIII
Int1W
OH

O

H
O O

O

-0.4

H 2O

0.98V

-5

-10

OH2

1

TS1H

+1.4
TS1W

-25.1

e
Int1 1.37 V

-

O

RuII MnIII RuIII

-22.7
Int3

OH

O2

O

O-O

RuIII MnIII RuIII
OH

OH

O-O

O2

-28.2
Int4 1.09 V
O
RuII MnIII RuIII

-35

OH

-40

O
eIII
III
III
-32.1 Ru Mn Ru
OH
Int5

2H2O
H+

-45

-49.0
-50

1

Figure 3. Energy diagram (in kcal/mol) for water oxidation. A reference potential of 1.26 V was
used to set up the diagram.

10

3.2 O-O bond formation
O-O bond formation in water oxidation by Ru 17-20, 76-80 and Mn 62, 69-72, 81-86 catalysts
have been widely studied by quantum chemical calculations. The two popular
pathways, namely direct coupling (DC) of two adjacent oxygen units and water
nucleophilic attack (WNA), were studied herein.
For the direct coupling pathway, we first considered the coupling of two adjacent
oxo groups. The transition state (TS1) was optimized and is shown in Figure 4. TS1
prefers to be a septet, and the two oxygen atoms that form the nascent O-O bond
have opposite spins (0.60 on O1 and -0.41 on O2). This is required for an efficient OO bond formation, and the important O1-O2 distance is 1.89 Å. TS1 was
characterized to be a true transition state with only one imaginary frequency of
136.6i cm-1. The energy barrier relative to 4 was calculated to be only 0.7 kcal mol-1
(Figure 3), and the total energy barrier became 7.2 kcal mol-1 when the energy
penalty for the formation of 4 was added. The quintet and triplet barriers were
calculated to be 7.2 and 1.1 kcal mol-1, respectively relative to 4septet. Downhill from
TS1, a peroxide intermediate was expected to be formed. However, the calculations
showed that an electron is transferred from the peroxide to Mn IV, which leads to the
formation of a superoxide bridging Mn III and Ru1III. In this intermediate Int1 (Figure
4), the O1-O2 distance is 1.32 Å, which is shorter than that of a common peroxide OO single bond (around 1.4 Å). The spin densities are listed as following, Ru1: 0.81, Mn:
3.76, Ru2: 1.03, O1: -0.26, 02: -0.41, O4: 0.92. Int1 lies at -13.3 kcal mol-1 relative to 4.
The nonet, quintet, triplet and singlet states were calculated to be 3.4, 2.0, 0.3 and 2.8
kcal mol-1 higher in energy, respectively.

11

Figure 4. Optimized structures of TS1, Int1, TS1H and Int1H. All structures are at septet state.
Distances are given in Å. Spin densities on selected atoms are indicated in red italics. The imaginary
frequencies of transition states are also given. Unimportant hydrogen atoms are not shown for
clarity. For all structures, only the ground state is shown.

We then considered the coupling of a Ru2 bound oxyl radical (O4) and a Mn IV bound
hydroxide (TS1H, Figure 4). The barrier of TS1H (septet) was calculated to be +15.6
kcal mol-1 higher than TS1, suggesting that this pathway is much less favorable. TS1H
was calculated to be a septet, in which the spin densities on O3 and O4 are -0.30 and
0.37, respectively. In TS1H, the important O3-O4 distance is 1.87 Å. This leads to the
formation of a hydroperoxide bridging MnIV and Ru2III (Int1H, Figure 4), and this step
is endergonic by 11.4 kcal mol-1. Therefore, this pathway is both kinetically and
thermodynamically less favorable compared with the coupling of two oxo groups.
For the alternative WNA mechanism, three scenarios were considered, namely
attack on one of the two RuIII-oxyl moieties (TS1W and TS1WH, Figure 5) or the MnIVoxyl moiety (TS1W’, Figure 5). For TS1W and TS1W’, the proton acceptors are the
adjacent MnIV-oxyl and RuIII-oxyl, respectively. The calculated barrier for TS1W is 13.2
kcal mol-1 relative to 4septet plus a separated water molecule. This is 12.5 kcal mol-1
higher than that of TS1. At TS1W, the critical O1-OW distance is 2.00 Å. TS1W prefers
to be an open-shell broken-symmetry singlet, in which the two Ru moieties are

12

coupled with the high-spin Mn in an antiferromagnetic fashion. The triplet, quintet,
and septet barriers are 5.7, 0.4, and 0.7 kcal mol-1 higher, respectively. In addition,
TS1WH, with the MnIV bound hydroxide as the proton acceptor, is 6.8 kcal mol-1 higher
than the TS1W. This suggests that the MnIV-bound oxyl group is a better proton
acceptor than the MnIV-bound hydroxide. (Figure S4) Furthermore, TS1W’ lies at
+31.3 kcal mol-1 relative to 4septet plus a water molecule (Figure S4). These
calculations suggest that the water nucleophilic attack pathway is not a viable option.

Figure 5. Optimized structures of TS1w (Broken-symmetry Singlet), Int1w (Septet), TS1WH (Septet),
Int1WH (Septet) TS1w’ (Nonet) and Int1w’ (Septet). Distances are given in Å. Spin densities on
selected atoms are indicated in red italics. The imaginary frequencies of transition state are also
given. Unimportant hydrogen atoms are not shown for clarity. For all structures, only the ground
state is shown.

O-O bond formation taking place at a lower oxidation state Ru1IVMnIVRu2IV (3) has

13

also been taken into account. The direct coupling of a RuIII-oxyl moiety and a MnIV
bound hydroxide via 3-TS1 (Figure S3) has a barrier of 19.8 kcal mol-1 (Figure 3)
relative to 3. The barrier is thus 12.6 kcal mol-1 higher than that of TS1 as described
above. These results further confirmed that O-O bond formation takes place at the
formal oxidation state of Ru1IVMnVRu2IV.
In addition to the thorough calculated energetic analysis carried out here, the
generation of the intramolecular O-O bond formation is intuitively supported by the
radical nature of all the neighboring M-O units at high oxidation states.
3.3 O2 release
For the release of a triplet O2 molecule from Int1, a number of pathways can be
envisioned. First, the superoxide moiety dissociates from the Mn III center through a
transition state (TS2 in Figure 6). We found that the first dissociation step requires
that both the superoxide and the MnIII center have positive spin densities.
Consequently, the nonet was optimized for TS2, which is shown in Figure 6. The
energy barrier was calculated to be 5.9 kcal mol-1 relative to Int1, which is 1.3 kcal
mol-1 lower (Figure 3) than that of TS1. In TS2, the important Mn-O2 distance was
calculated to be 2.50 Å, and the spin densities on Ru1, Mn, Ru2, O1 and O2 are 0.77,
3.89, 1.03, 0.54 and 0.63 respectively. TS2 was calculated to have only one imaginary
frequency of 110.3i cm-1. In the resulting intermediate Int2 (Figure 6), the Mn-O2
distance is 2.63 Å. In this step, no electron transfer occurs and the superoxide is
coordinated to Ru1. Int2 lies at +6.0 kcal mol-1 relative to Int1. From Int2, partial
intramolecular electron transfer from the superoxide moiety to the Ru1 III center leads
to the formation of the Int3 (Figure 6), which lies at -3.6 kcal mol-1 relative to Int2.
The total spin density on the dioxygen moiety increases from 1.27 in Int2 to 1.59 in
Int3. In addition, the O1-O2 distance decreases from 1.30 Å in Int2 to 1.25 Å in Int3.
The release of a dioxygen molecule from Int3 to form Int4 (Figure 6) is exergonic by
5.5 kcal mol-1. An electronic energy scan by increasing the Ru-O distance showed a
continuous increase of energy in the gas phase geometry optimization (Figure S5-1).
However, transition state can be optimized in the water solution (TS3, Figure S5-2),
14

which gave a barrier of only 4.3 kcal mol-1 relative to Int2. Int4 prefers to be a triplet,
the septet and quintet are 0.9 and 17.2 kcal mol-1 higher, respectively. In Int4, both
Mn and Ru1 were five-coordinated. Then, one electron oxidation of Int4 results in
Int5 (Figure 6), and the electron is removed from the Ru1II center with a potential of
1.09 V. In Int5, the spin densities on Ru1, Mn, Ru2 and O4 are 0.72, 3.89, 1.07 and
0.88, respectively. Int5 is an octet, and the sextet, quartet and doublet state are 0.7,
6.9 and 1.0 kcal mol-1 higher, respectively. Subsequently, two water molecules bind to
the Ru1III and MnIII centers, coupling with the release of one proton to regenerate 1,
which can reentry the next catalytic cycle.
An alternative pathway is the one electron oxidation of Int1, leading to the
formation of Int2’ (Figure 6). The electron is removed from the superoxide moiety,
and a triplet O2 is formed in Int2’. The redox potential for this oxidation was
calculated to be 1.37 V, which suggests the formation of Int2’ is endergonic by 2.5
kcal mol-1 from Int1. Int2’ prefers to be dectet, the octet, quartet and doublet state
are slightly higher in energies (lying at +0.6, +0.9 and +1.3 kcal mol-1, respectively),
while the sextet is 10.1 kcal mol-1 higher. In Int2’, the spin densities on Ru1, Mn, Ru2,
O1, O2 and O4 are 0.82, 3.90, 1.07, 0.80, 1.01 and 0.89, respectively. The Ru1-O1
distance is 2.13 Å. However, geometry optimization in water solution using the SMD
solvation model (labeled as Int2P’, Figure S6) showed that the triplet O2 dissociates
from the Ru1III center after the oxidation of Int1, and the Ru1-O1 distance increases
to 3.01 Å. The release of a triplet O2 molecule from Int2’ also leads to Int5.

15

Figure 6. Optimized structures of TS2 (Nonet), Int2 (Nonet), Int3 (Nonet), Int4 (Triplet), Int5
(Octet) and Int2’ (Dectet). Distances are given in Å. Spin densities on selected atoms are indicated in
red italics. Unimportant hydrogen atoms are not shown for clarity. The imaginary frequencies of
transition state are also given. For all structures, only the ground state is shown.

The above discussed two pathways may compete with each other. Other possible
pathways of O2 release through water insertion into either the Ru1 center or the Mn
center were found to be energetically less favorable, as discussed in the supporting
information (Page S12-S14).
From Figure 3, it can be seen that the calculated total barrier at the B3LYP*-D3
level is only 7.2 kcal mol-1 (from 3 to TS1) assuming that the electrochemical
oxidation steps are not rate-limiting. The calculated value is somewhat
underestimated compared with the experimental kinetic data (a TON of 12.2 in 10

16

minutes; Calculated from Figure 7 in Ref 41), which is about 20 kcal mol -1. This may
attribute to the density functional used. Singlet-point calculations at the B3LYP-D3,
M06L-D3, M06-D3, TPSS0-D3 (25% HF exchange), TPSSh-D3 and TPSSTPSS-D3
levels have also been performed to evaluate how sensitive the calculated redox
potentials and total barriers are regarding to the choice of the density functionals
used. The calculated results are displayed in Table 1. Importantly, all functionals favor
the direct coupling pathway for the O-O bond formation. The total barrier increases
to 11.0 kcal mol-1 when B3LYP-D3 was used, and further to 15.4 kcal mol-1 when
M06-D3 was used. The calculated onset potentials at the B3LYP*-D3, M06L-D3 and
TPSSh-D3 level are 1.54 V, 1.44 V and 1.51 V, respectively, which are consistent with
the experiment results (1.50 V).41 However, the B3LYP-D3, M06-D3 and TPSS0-D3
functionals gave slightly higher values. It seems that density functionals with more HF
exchange give slight higher values of redox potentials than those with less HF
exchange, as found in our previous study on mononuclear iron water oxidation
catalyst.57 It is also plausible that the oxidation of 3 to 4 is the rate-limiting step in the
photochemical oxidation since the formation of 4 is endergonic by 6.5 kcal/mol. This
large endergonicity results in a much slower electronchemical oxidation than the
normal exergonic electronchemical oxidation. A stronger oxidant than [Ru(bpy)3]3+
should favor this step, and lower the total barrier. Indeed, the electrochemical oxidation has a
onset potential of 1.50 V,41 which is higher than the potential of [Ru(bpy)3]3+/[Ru(bpy)3]2+
(1.26 V).
Table1. Comparsion of the redox potentials and barriers with different functionals.
Redox Potential(V)
Total Barrier(kcal mol-1)
2/1
3/2
4/3
DCOOa
DCOOHb
WNA
B3LYP*-D3
0.98
0.75
1.54
7.2
22.8
20.2
B3LYP-D3
1.13
0.97
1.70
11.0
27.0
25.2
M06L-D3
0.82
0.32
1.44
6.1
17.9
22.7
M06-D3
0.99
1.29
1.83
15.4
32.4
32.1
c
TPSS0-D3
1.23
1.14
1.83
13.9
32.5
35.0
TPSSh-D3
1.02
0.59
1.51
6.2
23.1
20.6
TPSSTPSS-D3
0.89
0.21
1.21
0.7
17.3
14.3
direct oxo-oxo coupling mechanism.
oxo-hydroxide coupling mechanism.
c with 25% HF exchange.
a

b direct

17

Here we used the reorganization energy of 17.2 kcal/mol and second-order rate
constant of 4.2×108 M-1 s-1 for the self-exchange electron transfer between
[Ru(bpy)3]3+ and [Ru(bpy)3]2+ as a reference,87 the barrier ( G  ) for the oxidation of
3 by [Ru(bpy)3]3+ was calculated to be 12.7 kcal/mol using the Marcus theory88 by
assuming the same reorganization energy for these two different electrochemical
reactions.
( G 0   ) 2
4
where λ is the reorganization energy and G 0 is the driving force.
G  

(1)

The apparent barrier under the experimental condition (5×10 -4 M for the
concentration of [Ru(bpy)3]3+)41 can be estimated to be 17.2 kcal/mol when the
concentration correction was taken into account. This value can be considered to be
in reasonably good agreement with the experimental barrier of about 20 kcal/mol. 41
The oxidation of 3 by [Ru(bpy)3]3+ may have larger reorganization energy than the
self-exchange reaction between [Ru(bpy)3]3+ and [Ru(bpy)3]2+. If a reorganization
energy of 23.06 kcal/mol (1 eV) was used, the apparent barrier increases slightly to
18.5 kcal/mol. It should be pointed out that more accurate estimation of the rate for
this electrochemical rate for this step requires advanced calculations of
reorganization energies (reactant, product, and solvent), driving force, the electronic
coupling, and the proton transfer interface properties, 89-91 which is beyond the scope
of the present study.

4. Conclusion
In the present work, we have studied the water oxidation mechanism catalyzed by
the heterotrinuclear complex Ru2Mn-(H2O)4 (A) using density functional calculations.
The whole catalytic cycle (Scheme 2) including relevant redox intermediates and
redox potentials was calculated.

18

OH2 OH

O
2H2O

2H+,e-

RuIIIMnIIIRuIII
OH

H+

O

O

1

RuIIIMnIIIRuIII
OH
Int5

e-

OH

O

RuIIIMnIIIRuIII
OH
2

e-

O2
III

OH O

O

O
II

RuIIIMnIVRuIII

III

Ru Mn Ru
OH
Int4

OH
3

O

O-O

III
III
RuIIIMn Ru

O2

OH

H+,e-

O

O-O

O

RuIIMnIIIRuIII

O

RuIIIMnIVRuIII

OH

e-

OH O

Int3

4
O

O-O
III

III

O O
III

Ru Mn Ru

O

RuIIIMnIIIRuIII
OH
Int1

OH
Int2

Scheme 2. Suggested catalytic cycle for water oxidation by the heterotrinuclear Ru2Mn complex
on the basis of calculations.

Water oxidation commences from the formal oxidation state Ru1 IVMnIIIRu2III (1). A
RuIII-oxyl complex 1 was generated from A through a number of deprotonation and
oxidation events. From 1, one electron oxidation of Ru2 leads to 2 with a formal
oxidation state of Ru1IVMnIIIRu2IV, and this process is coupled with the release of two
protons. Then, one electron oxidation of the MnIII center in 2 results in 3
(Ru1IVMnIVRu2IV). Next, a PCET oxidation of 3 generates 4 (Ru1IVMnIVRu2IV) with a
potential of 1.54 V. Complex 4 is capable of triggering O-O bond formation through
the direct coupling of a Ru-oxo and a Mn-oxo. The direct coupling of two oxo units
leads to the superoxide intermediate Int1, with a total barrier of 7.2 kcal mol-1 at the
B3LYP*-D3 level when assuming the electrochemical oxidation of 3 is ultrafast. Other
O-O bond formation mechanisms, including the WNA mechanism, the direct coupling
of an oxo and a hydroxide, and the direct coupling at a lower oxidation state were
calculated to be much less favorable.
The O2 release may take place directly from Int1 or after a one electron oxidation of
19

Int1. For the former pathway, the superoxide first dissociates from the Mn center in
Int1 via TS2, which generates a Ru1-bound superoxide intermediate Int2. The
intramolecular electron transfer from the superoxide to Ru1 leads to the formation of
Int3, which has a triplet O2 bound to the Ru1 center. Subsequently, O2 is released,
followed by another one electron oxidation to form Int5. Finally, two water molecules
bind to Ru1 and Mn, concomitant with a proton release to regenerate 1. The one
electron oxidation of Int1 results in the direct dissociation of the dioxygen moiety
from the Mn center and the formation of a triplet O2 bound to the Ru1 center (Int2’).
Int2’ can be converted into Int5 after the release of O2. These two pathways about O2
evolution may compete with each other as they were both found to be quite facile.
Other possible pathways of O2 release via water insertion are energetically less
favorable.

Acknowledgments
This work was supported by the National Natural Science Foundation of China
(21503083 and 21873031) and the Fundamental Research Funds for the Central
Universities (2017KFKJXX014). AL thanks support from MINECO, FEDER and
AGAUR (CTQ2016-80058-R and 2017-SGR-1631).

References
[1] Y. Umena, K. Kawakami, J. R. Shen, N. Kamiya, Nature 2011, 473, 55-60.
[2] A. Guskov, J. Kern, A. Gabdulkhakov, M. Broser, A. Zouni, W. Saenger, Nat. Struct. Mol. Biol. 2009,
16, 334-342.
[3] B. Loll, J. Kern, W. Saenger, A. Zouni, J. Biesiadka, Nature 2005, 438, 1040-1044.
[4] K. N. Ferreira, T. M. Iverson, K. Maghlaoui, J. Barber, S. Iwata, Science 2004, 303, 1831-1838.
[5] M. Suga, F. Akita, M. Sugahara, M. Kubo, Y. Nakajima, T. Nakane, K. Yamashita, Y. Umena, M.
Nakabayashi, T. Yamane, T. Nakano, M. Suzuki, T. Masuda, S. Inoue, T. Kimura, T. Nomura, S.
Yonekura, L.-J. Yu, T. Sakamoto, T. Motomura, J.-H. Chen, Y. Kato, T. Noguchi, K. Tono, Y. Joti, T.
Kameshima, T. Hatsui, E. Nango, R. Tanaka, H. Naitow, Y. Matsuura, A. Yamashita, M. Yamamoto, O.
Nureki, M. Yabashi, T. Ishikawa, S. Iwata, J.-R. Shen, Nature 2017, 543, 131-135.
[6] G. Maayan, N. Gluz, G. Christou, Nat. Catal. 2018, 1, 48-54.
[7] B. Schwarz, J. Forster, T. Jacob, C. Streb, Angew. Chem. Int. Ed. 2016, 55, 6329 –6333.
[8] C. Zhang, C. Chen, H. Dong, J.-R. Shen, H. Dau, J. Zhao, Science 2015, 348, 690-693.
[9] W. T. Lee, S. B. Munoz, 3rd, D. A. Dickie, J. M. Smith, Angew. Chem. Int. Ed. 2014, 53, 9856-9859.
[10] R. K. Hocking, R. Brimblecombe, L. Y. Chang, A. Singh, M. H. Cheah, C. Glover, W. H. Casey, L.
20

Spiccia, Nat. Chem. 2011, 3, 461-466.
[11] D. Wang, S. L. Marquard, L. Troian-Gautier, M. V. Sheridan, B. D. Sherman, Y. Wang, M. S.
Eberhart, B. H. Farnum, C. J. Dares, T. J. Meyer, J. Am. Chem. Soc. 2018, 140, 719-726.
[12] R. Matheu, A. Ghaderian, L. Francas, P. Chernev, M. Ertem, J. Benet-Buchholz, V. Batista, M.
Haumann, C. Gimbert-Surinach, X. Sala, A. Llobet, Chem. Eur. J., 2018, 24, 12838-12847.
[13] Q. Daniel, L. Duan, B. J. J. Timmer, H. Chen, X. Luo, R. Ambre, Y. Wang, B. Zhang, P. Zhang, L.
Wang, F. Li, J. Sun, M. Ahlquist, L. Sun, ACS Catal. 2018, 8, 4375-4382.
[14] F. Cai, W. Su, H. A. Younus, K. Zhou, C. Chen, S. Chaemchuen, F. Verpoort, New J. Chem. 2018,
42, 2476-2482.
[15] L. Francàs, R. Matheu, E. Pastor, A. Reynal, S. Berardi, X. Sala, A. Llobet, J. R. Durrant, ACS Catal.
2017, 7, 5142-5150.
[16] S. Zhan, D. Martensson, M. Purg, S. C. L. Kamerlin, M. S. G. Ahlquist, Angew. Chem. Int. Ed.
2017, 56, 6962-6965.
[17] D. W. Shaffer, Y. Xie, J. J. Concepcion, Chem. Soc. Rev. 2017, 46, 6170-6193.
[18] D. W. Shaffer, Y. Xie, D. J. Szalda, J. J. Concepcion, J. Am. Chem. Soc. 2017, 139, 15347-15355.
[19] M. Gil-Sepulcre, M. Bohler, M. Schilling, F. Bozoglian, C. Bachmann, D. Scherrer, T. Fox, B.
Spingler, C. Gimbert-Surinach, R. Alberto, R. Bofill, X. Sala, S. Luber, C. J. Richmond, A. Llobet,
ChemSusChem 2017, 10, 4517-4525.
[20] T. Fan, L. Duan, P. Huang, H. Chen, Q. Daniel, M. S. G. Ahlquist, L. Sun, ACS Catal. 2017, 7, 29562966.
[21] M. Schulze, V. Kunz, P. D. Frischmann, F. Wurthner, Nat. Chem. 2016, 8, 576-583.
[22] Y. Tsubonouchi, S. Lin, A. R. Parent, G. W. Brudvig, K. Sakai, Chem. Commun. 2016, 52, 80188021.
[23] L. Duan, L. Wang, F. Li, F. Li, L. Sun, Acc. Chem. Res. 2015, 48, 2084-2096.
[24] M. D. Karkas, O. Verho, E. V. Johnston, B. Akermark, Chem. Rev. 2014, 114, 11863-12001.
[25] M. Li, S. Bernhard, Catal. Today 2017, 290, 19-27.
[26] I. Corbucci, K. Ellingwood, L. Fagiolari, C. Zuccaccia, F. Elisei, P. L. Gentili, A. Macchioni, Catal.
Today 2017, 290, 10-18.
[27] M. Okamura, M. Kondo, R. Kuga, Y. Kurashige, T. Yanai, S. Hayami, V. K. Praneeth, M. Yoshida, K.
Yoneda, S. Kawata, S. Masaoka, Nature 2016, 530, 465-468.
[28] Z. Codola, L. Gomez, S. T. Kleespies, L. Que, Jr., M. Costas, J. Lloret-Fillol, Nat. Commun. 2015, 6,
1-9.
[29] K. G. Kottrup, S. D'Agostini, P. H. van Langevelde, M. A. Siegler, D. G. H. Hetterscheid, ACS Catal.
2018, 8, 1052-1061.
[30] S. Pattanayak, D. R. Chowdhury, B. Garai, K. K. Singh, A. Paul, B. B. Dhar, S. S. Gupta, Chem. Eur.
J. 2017, 23, 3414-3424.
[31] B. Das, B. L. Lee, E. A. Karlsson, T. Akermark, A. Shatskiy, S. Demeshko, R. Z. Liao, T. M. Laine,
M. Haukka, E. Zeglio, A. F. Abdel-Magied, P. E. Siegbahn, F. Meyer, M. D. Karkas, E. V. Johnston, E.
Nordlander, B. Akermark, Dalton Trans. 2016, 45, 13289-13293.
[32] H. Y. Du, S. C. Chen, X. J. Su, L. Jiao, M. T. Zhang, J. Am. Chem. Soc. 2018, 140, 1557-1565.
[33] T. Ishizuka, A. Watanabe, H. Kotani, D. Hong, K. Satonaka, T. Wada, Y. Shiota, K. Yoshizawa, K.
Ohara, K. Yamaguchi, S. Kato, S. Fukuzumi, T. Kojima, Inorg. Chem. 2016, 55, 1154-1164.
[34] L. Xu, H. Lei, Z. Zhang, Z. Yao, J. Li, Z. Yu, R. Cao, Phys. Chem. Chem. Phys. 2017, 19, 9755-9761.
[35] L. Wang, L. Duan, R. B. Ambre, Q. Daniel, H. Chen, J. Sun, B. Das, A. Thapper, J. Uhlig, P. Diner, L.
Sun, J. Catal. 2016, 335, 72-78.
[36] J. Masud, P. Kyritsis, P.-C. Ioannou, N. Levesanos, a. M. Nath, ChemSusChem 2016, 9, 31283132.
[37] M. Zhang, M. T. Zhang, C. Hou, Z. F. Ke, T. B. Lu, Angew. Chem. Int. Ed. 2014, 53, 13042-13048.
[38] X. Jiang, J. Li, B. Yang, X. Z. Wei, B. W. Dong, Y. Kao, M. Y. Huang, C. H. Tung, L. Z. Wu, Angew.
Chem. 2018, 57, 7850-7854.
[39] K. J. Fisher, K. L. Materna, B. Q. Mercado, R. H. Crabtree, G. W. Brudvig, ACS Catal. 2017, 7,
3384-3387.
21

[40] X. J. Su, M. Gao, L. Jiao, R. Z. Liao, P. E. Siegbahn, J. P. Cheng, M. T. Zhang, Angew. Chem. Int. Ed.
2015, 54, 4909-4914.
[41] L. Mognon, S. Mandal, C. E. Castillo, J. Fortage, F. Molton, G. Aromí, J. Benet-Buchhlolz, M.-N.
Collomb, A. Llobet, Chem. Sci. 2016, 7, 3304-3312.
[42] R.-Z. Liao, P. E. M. Siegbahn, ChemSusChem 2017, 10, 4236-4263.
[43] R. D. Gaussian 09, G. W. T. M. J. Frisch, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.
P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A.
Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N.
Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C.
Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, R. L. Martin, K. Morokuma, V. G.Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S.
Dapprich, A. D. Daniels, O. Farkas, J. B. H.Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian,
Inc., Wallingford CT, 2013.
[44] S. Grimme,J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010,132,154104.
[45] A. D. Becke, J. Chem. Phys. 1993, 98, 5648-5652.
[46] D. Andrae, U. HaiuBermann, M. Dolg, H. Stoll, H. PreuB, Theor. Chim. Acta. 1990, 77, 123-144.
[47] A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B 2009, 113, 6378–6396.
[48] M. Reiher, O. Salomon, B. A. Hess, Theor. Chem. Acc. 2001, 114, 7894-7899.
[49] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 215-241.
[50] Y. Zhao, D. G. Truhlar, J. Chem. Phys. 2006, 125, 194101.
[51] A. Marzouk, B. Madebene, M. E. Alikhani, J. Phys. Chem. A 2013, 117, 4462-4471.
[52] V. N. Staroverov, G. E. Scuseria, J. Tao, J. P. Perdew, J. Chem. Phys. 2003, 119, 12129-12137.
[53] J. Tao, J. P. Perdew, V. N. Staroverov, G. E. Scuseria, Phys. Rev. Lett. 2003, 91, 146401.
[54] D. M. Camaioni, C. A. Schwerdtfeger, J. Phys. Chem. A 2005, 109, 10795-10797.
[55] C. J. Cramer, D. G. Truhlar, J. Am. Chem. Soc. 1991, 113, 8305-8311.
[56] C. P. Kelly, C. J. Cramer, D. G. Truhlar, J. Chem. Theory Comput. 2005, 1, 1133-1152.
[57] Y. Y. Li, L. P. Tong, R. Z. Liao, Inorg. Chem. 2018, 57, 4590-4601.
[58] A. A. Isse, A. Gennaro, J. Phys. Chem. B 2010, 114, 7894–7899.
[59] M. Z. Ertem, L. Gagliardi, C. J. Cramer, Chem. Sci. 2012, 3, 1293-1299.
[60] S. Roeser, F. Bozoglian, C. J. Richmond, A. B. League, M. Z. Ertem, L. Francas, P. Miro, J. BenetBuchholz, C. J. Cramer, A. Llobet, Catal. Sci. Technol. 2016, 6, 5088-5101.
[61] D. W. Crandell, S. Ghosh, C. P. Berlinguette, M.-H. Baik, ChemSusChem 2015, 8, 844-852.
[62] D. W. Crandell, S. Xu, J. M. Smith, M. H. Baik, Inorg. Chem. 2017, 56, 4436-4446.
[63] I. Funes-Ardoiz, P. Garrido-Barros, A. Llobet, F. Maseras, ACS Catal. 2017, 7, 1712-1719.
[64] P. Garrido-Barros, I. Funes-Ardoiz, S. Drouet, J. Benet-Buchholz, F. Maseras, A. Llobet, J. Am.
Chem. Soc. 2015, 137, 6758-6761.
[65] R.-Z. Liao, P. E. M. Siegbahn, ACS Catal. 2014, 4, 3937-3949.
[66] P. E. M. Siegbahn, Biochim. Biophys. Acta. 2013, 1827, 1003-1019.
[67] P. E. M. Siegbahn, Acc. Chem. Res. 2009, 42, 1871-1880.
[68] P. E. M. Siegbahn, Chem. Eur. J. 2006, 12, 9217-9227.
[69] M. Lundberg, M. R. A. Blomberg, P. E. M. Siegbahn, Inorg. Chem. 2004, 43, 264–274.
[70] R.-Z. Liao, P. E. M. Siegbahn, J. Catal. 2017, 354, 169-181.
[71] Y. Y. Li, Y. Ke, P. E. M. Siegbahn, R.-Z. Liao, ChemSusChem 2017, 10, 903-911.
[72] R.-Z. Liao, M. D. Karkas, B. L. Lee, B. Akermark, P. E. R. Siegbahn, Inorg. Chem. 2015, 54, 342351.
[73] L. Noodleman, J. Chem. Phys. 1981, 74, 5737-5743.
[74] L. Noodleman, D. A. Case, Adv. Inorg. Chem. 1992, 38, 423-470.
[75] J.-M. Mouesca, J. L. Chen, L. Noodleman, D. Bashford, D. A. Case, J. Am. Chem. Soc. 1994, 116,
11898-11914.
22

[76] D. Lebedev, Y. Pineda-Galvan, Y. Tokimaru, A. Fedorov, N. Kaeffer, C. Coperet, Y. Pushkar, J. Am.
Chem. Soc. 2018, 140, 451-458.
[77] R. Matheu, M. Z. Ertem, M. Pipelier, J. Lebreton, D. Dubreuil, J. Benet-Buchholz, X. Sala, A.
Tessier, A. Llobet, ACS Catal. 2018, 8, 2039-2048.
[78] D. W. S. YanXie, Anna Lewandowska-Andralojc,David J. Szalda, and, J. J. Concepcion, Angew.
Chem. Int. Ed. 2016, 55, 8067-8071.
[79] R.-Z. Liao, M. D. Kärkäs, T. M. Laine, B. Åkermark, P. E. M. Siegbahn, Catal. Sci. Technol. 2016, 6,
5031-5041.
[80] X. Sala, S. Maji, R. Bofill, J. Garcia-Anton, L. Escriche, A. Llobet, Acc. Chem. Res. 2014, 47, 504516.
[81] E. M. Sproviero, J. Inorg. Biochem. 2017, 171, 52-66.
[82] P. Xu, T. Zhou, N. Natalia, S. Hu, X. Zheng, J. Phys. Chem. A 2016, 120, 10033-10042.
[83] I. Rivalta, K. R. Yang, G. W. Brudvig, V. S. Batista, ACS Catal. 2015, 5, 2384-2390.
[84] S. Khan, K. R. Yang, M. Z. Ertem, V. S. Batista, G. W. Brudvig, ACS Catal. 2015, 5, 7104-7113.
[85] W. M. C. Sameera, C. J.McKenzieb, J. E. McGrady, Dalton Trans. 2011, 40, 3859-3870.
[86] R.-Z. Liao, P. E. M. Siegbahn, J. Photochem. Photobiol. B 2015, 152, 162-172.
[87] X. Song, Y. Lei, V. Wallendal, M. W. Perkivic, D. K. Jackman, J. F. Endicott, D. P. Rillema, J. Phys.
Chem. 1993, 97, 3225-3236.
[88] R. A. Marcus, Annu. Rev. Phys. Chem. 1964, 15, 155-196.
[89] R. I. Cukier, J. Phys. Chem. 1996, 100, 15428-15443.
[90] S. Hammes-Schiffer, N. Iordanova, Biochim. Biophys. Acta 2004, 1655, 29-36.
[91] S. Hammes-Schiffer, Acc. Chem. Res. 2009, 42, 1881-1889.

23

Graphical abstract for

Quantum Chemical Study of the Mechanism of Water Oxidation
Catalyzed by a Heterotrinuclear Ru2Mn Complex

Ying-Ying Lia, Carolina Gimbertb, Antoni Llobetb, Per E. M. Siegbahnc, Rong-Zhen Liaoa*

24