DOI: 10.1002/cctc.201600007

Full Papers

This is the peer reviewed version of the following article: ChemCatChem 2016, 8, 1792 – 1798, which has been published
in final form at http://dx.doi.org/10.1002/cctc.201600007. This article may be used for non-commercial
purposes in accordance with Wiley Terms and Conditions for Self-Archiving.

Computationally Probing the Performance of Hybrid,
Heterogeneous, and Homogeneous Iridium-Based
Catalysts for Water Oxidation
Max García-Melchor,*[a] Laia Vilella,[b, c] Nfflria López,[b] and Aleksandra Vojvodic*[d]
An attractive strategy to improve the performance of water oxidation catalysts would be to anchor a homogeneous molecular catalyst onto a heterogeneous solid surface to create
a hybrid catalyst. The idea of this combined system is to take
advantage of the individual properties of each of the two catalyst components. We use DFT calculations to determine the
stability and activity of a model hybrid water oxidation catalyst

that consists of a dimeric Ir complex attached on the IrO2(11 0)
surface through two oxygen atoms. We find that homogeneous catalysts can be anchored to oxide surfaces without a significant loss of activity. Hence, the design of hybrid systems
that benefit from both the high tunability of the activity of homogeneous catalysts and the stability of heterogeneous systems seems feasible.

Introduction
With the depletion of fossil fuels and the environmental concerns over their use, there is an increasing need to replace
these energy sources with more sustainable and renewable
ones.[1, 2] One promising solution would be to use electrical
energy from a renewable source, for example, sunlight and
store it in the form of chemical bonds such as H2 and O2
through artificial photosynthesis.[3, 4] The bottleneck of this process is the oxidation of water to molecular oxygen, also known
as the oxygen evolution reaction (OER).[5] This reaction is particularly energy demanding because of the mechanistic requirements for the removal of four electrons and protons from
two water molecules with the concomitant formation of an OÀ
O bond.[6, 7] Hence, the search for active OER catalysts that provide high current densities and operate for substantial periods
of time without degradation has been a quest for decades. To
[a] Dr. M. García-Melchor
SUNCAT Center for Interface Science and Catalysis
Department of Chemical Engineering
Stanford University
Stanford, CA 94305 (USA)
E-mail: maxg@slac.stanford.edu
[b] Dr. L. Vilella, Prof. N. López
Institute of Chemical Research of Catalonia (ICIQ)
The Barcelona Institute of Science and Technology (BIST)
Av. Països Catalans, 16, E-43007 Tarragona (Spain)
[c] Dr. L. Vilella
Departament de Química
Universitat Autònoma de Barcelona
Cerdanyola del Vall›s, E-08193 Barcelona (Spain)
[d] Dr. A. Vojvodic
SUNCAT Center for Interface Science and Catalysis
SLAC National Accelerator Laboratory
Menlo Park, CA 94025 (USA)
E-mail: alevoj@slac.stanford.edu
Supporting Information and the ORCID identification number(s) for the
author(s) of this article can be found under http://dx.doi.org/10.1002/
cctc.201600007.

ChemCatChem 2016, 8, 1792 – 1798

date, the best homogeneous and heterogeneous OER catalysts
are based on Ir and Ru,[8–11] but the scarcity and high cost of
these elements limit the scale-up of the energy conversion
technologies.[12] In addition, even the most active Ir- and Rubased catalysts reported in the literature either require a high
overpotential or are not stable enough for industrial application. To overcome these issues, several approaches have been
undertaken such as the synthesis of catalysts inspired by
nature[13–15] and based on non-precious metals,[16–19] the addition of dopants[20–23] or promoters,[22, 24, 25] or the creation of
confined reaction environments.[26, 27]
Another appealing strategy, which is far from fully exploited
yet, consists of the anchoring of active homogeneous OER catalysts on a solid support (either active or inactive) to create
a hybrid type of catalyst.[28–30] The fundamental idea behind
these combined systems is that they might benefit from the
individual properties of their homogeneous and heterogeneous constituents, as the limitations of one are usually the main
advantages of the other.[31] Homogeneous catalysts typically
exhibit a high tunability of activity but they lack the high stability and good electron transport often found in heterogeneous catalyst systems. Besides the coexistence of these benefits,
a synergistic effect that arises from a combination of the two
catalytic systems might also emerge. To date, a few successful
examples of this strategy have been demonstrated, which include the recent experimental studies on water oxidation
using an Ir-based HEDTA (HEDTA = monoprotonated ethylenediaminetetraacetic acid) catalyst supported on TiO2[32] and a dimeric Ir complex anchored on Sn-doped indium oxide nanoparticles.[33] However, the interplay between the anchored molecular catalyst and the support is still unknown.
Herein, we present the first DFT study on the OER performance (stability and activity) of a hybrid model Ir-based catalyst (Figure 1 a) and compare it to its corresponding homogeneous and heterogeneous constituents (Figure 1 b–c). Our cal-

1792

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers

Figure 1. Side view of the atomic structure for the investigated a) hybrid,
b) homogeneous, and c) heterogeneous Ir-based catalysts. Color code: Ir
(dark blue), O (red), and H (white). This color code applies to all subsequent
figures.

culations show that the homogeneous catalyst binds strongly
to the support surface and performs water oxidation without
a significant loss of activity. We also identify the active site
motifs and reaction pathways for each catalytic system, and
based on that, propose that tuning the chemical properties of
the surface can be used as a complementary approach to the
metal–ligand exchange to alter the OER performance. These
results may pave the way to design more efficient and robust
water oxidation catalysts.

Computational Details
The stability and OER activity of the considered Ir-based catalysts were investigated by using DFT calculations in conjunction with the computational standard hydrogen electrode
(SHE) model.[34] All the calculations were performed using the
Perdew–Burke–Ernzenhof[35] (PBE) exchange correlation functional as implemented in the VASP code, version 5.3.2.[36, 37]
Unless otherwise indicated, the core electrons of the Ir, O, and
H atoms have been replaced by projector-augmented wave
(PAW) pseudopotentials,[38] whereas their valence electrons
have been expanded in plane waves with a kinetic energy
cutoff of 450 eV.
The lattice parameters for the bulk rutile IrO2 structure have
been optimized using a higher cutoff energy of 600 eV and
a 9 ” 9 ” 13 Monkhorst Pack (MP) k-point grid. The obtained lattice parameters are acalc = bcalc = 4.543 Š and ccalc = 3.187 Š,
which are in very good agreement with previous theoretical
findings[39, 40] and experimental values (aexp = bexp = 4.505 Š and
cexp = 3.159 Š).[41] To model the heterogeneous IrO2 catalyst surface, we used a 2 ” 1 supercell that consisted of a slab of nine
atomic layers of the most stable (11 0) facet[42] separated by at
least 13 Š of vacuum. Geometry optimizations for this system
were performed by using a 5 ” 5 ” 1 MP k-point mesh and by
allowing the adsorbed species (O*, HOO*, OO*, H2O*, and HO*,
in which the asterisk corresponds to an active site on the
oxide) and the five topmost atomic layers to relax, and the rest
of the atoms were kept fixed at their bulk positions.
In the case of the homogeneous catalyst, we simulated the
dimeric complex [Ir2(H2O)4(OH)4(m-OH)2] using a cell of 18 ” 15 ”
ChemCatChem 2016, 8, 1792 – 1798

www.chemcatchem.org

15 Š in size and sampling the Brillouin zone at the G-point.
This charge-neutral compound avoids the complexity of dealing with charged species in periodic calculations. To model the
hybrid catalyst, we anchored this dimeric Ir complex on top of
the IrO2(11 0) surface with a 6 ” 2 periodicity through two
oxygen atoms to the coordinatively unsaturated sites (CUS)
sites of the surface. As a result of the considerable size of the
combined system, calculations for this catalyst were performed
at the G-point. The energy convergence with respect to the kpoint sampling was examined for selected adsorbed species
using a denser 2 ” 3 ” 1 k-point mesh, which obtained energy
differences within 0.02 eV (Table S1). Geometry optimizations
for the hybrid catalyst were performed by relaxing the OER intermediates, the atoms in the attached dimer, and the three
topmost surface layers.
For the three model Ir catalysts, total energies were converged to values less than 10À5 eV in the self-consistent field,
and the geometries were relaxed until the energy threshold of
10À4 eV was fulfilled. The nature of all stationary points was
confirmed by calculating the vibrational frequencies of the adsorbed species and metal atoms coordinated directly to them.
Contributions to the Gibbs energies for the adsorbates were
calculated at T = 300 K and p = 1 atm, except for water, for
which the equilibrium vapor pressure at that temperature (p =
0.035 atm) was used (see Supporting Information for details).
The Gibbs energy for the water oxidation reaction was fixed to
the experimentally known value of 4.92 eV to avoid the introduction of the DFT error into the energetics related to molecular oxygen.[34, 43, 44] Spin-polarized calculations were performed if
needed, relaxing the total magnetic moment during the selfconsistent cycles.

Results and Discussion
In this work, we investigate the viability of a hybrid water oxidation catalyst and the role of the surface in this type of
system by anchoring the Ir complex [Ir2(H2O)4(OH)4(m-OH)2] on
the IrO2(11 0) surface through two oxygen atoms as linkers
(Figure 1). We chose this dimeric Ir complex as the homogeneous catalyst (Ir-hom) because it resembles the molecular species suggested to be involved in the formation of the purpleblue solution that catalyzes water oxidation.[45] Moreover, such
a dimeric complex allows us to compare the two main OER
mechanisms proposed in the literature, namely, the water nucleophilic attack (WNA) and the interaction between two MÀO
units (I2M).[11, 46, 47] In addition, we selected IrO2 (Ir-het) as the
heterogeneous catalyst because it is one of the most active
pure single transition metal oxides reported to date[9, 48] and it
has been studied computationally in detail.[49–51]
Notably, the hybrid catalyst considered in this work (Ir-hyb)
is intended to serve as a model to probe the feasibility and
mechanisms of this type of system and should not be considered as a potential high-performance OER catalyst. With this
aim, we first examine the stability of Ir-hyb by calculating the
adsorption energy of the dimeric Ir complex to the IrO2(11 0)
surface as follows [Eq. (1)]:

1793

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
DE ads ¼ E H2 þE Ir-hyb ÀE Ir-hom ÀE Ir-het

ð1Þ

in which EH2 , EIr-hyb, EIr-hom, and EIr-het are the potential energies
for molecular hydrogen and the Ir-hyb, Ir-hom, and Ir-het systems, respectively. Importantly, the obtained adsorption energy
from Equation (1) is exothermic by 2.19 eV, which indicates
that the dimeric Ir complex is indeed attached strongly to the
IrO2(11 0) surface through the two oxygen linkers. Interestingly,
the optimized IrÀIr bond distance in the attached Ir dimer
almost matches that of the IrO2(11 0) surface (3.207 vs.
3.190 Š). Hence, this particular hybrid catalyst that contains
only aqua and hydroxo ligands might be conceived as a defect
on the surface, or an extension of the IrO2(11 0) surface, which
would explain the strong interaction between the homogeneous and heterogeneous constituents predicted by calculations,
or even possibly a crude model for amorphous IrOx. Further insight into this interaction can be gained by evaluating the
electronic charge density difference between the molecular
and the surface fragments of the Ir-hyb catalyst (Figure S1).
This analysis reveals that both Ir atoms from the adsorbed
complex and the surface provide electron density to the O
linkers, which makes these bonds rather strong.
We also examine the stability of Ir-hyb by calculating the
energy required to detach the dimeric Ir complex from the surface upon the attack of two solvent water molecules, DEdes, according to Equation (2):
DE des ¼ E Ir-hetþ2 OH þE Ir-hom À2* E H2 O ÀE Ir-hyb

ð2Þ

in which EH2 O and EIr-het+2 OH are the energies for a water gas
+
molecule and the IrO2(11 0) surface with two OH groups adsorbed at CUS positions, respectively. In this case, we find that
the deactivation route in Equation (2) is a highly endothermic
process (1.54 eV), which reinforces the stability of the Ir-hyb
system.
Before we evaluated the OER activity for Ir-hyb and its individual constituents, it is necessary to determine the resting
state of the catalyst under relevant reaction conditions. To this
end, for each catalytic system we examined a number of potential species that might exist in aqueous solution. More specifically, for the Ir-het catalyst we considered up to 19 different
H, O, and OH surface coverages (Figure S2) and calculated
their relative Gibbs energies as a function of pH and applied
voltage (U) to construct a surface Pourbaix diagram. According
to the Pourbaix representation (Figure 2 a), the resting state for
Ir-het under typical OER conditions corresponds to a surface
with all Ir-CUS sites covered completely by O atoms, in agreement with previous theoretical findings.[27, 49]
Unlike the heterogeneous system, the assessment of the
resting state for Ir-hom is significantly more complicated given
the vast number of possible protonation states of this complex. However, a thorough analysis of the resting state for this
system is still necessary. Here, we tackle this issue by first exploring several consecutive dehydrogenation of Ir-hom at
a thermodynamic potential of 1.23 V and pH 0 (Figure S3).
Under these conditions, calculations indicate that aqua ligands
are dehydrogenated preferentially until a fully hydroxylated diChemCatChem 2016, 8, 1792 – 1798

www.chemcatchem.org

Figure 2. Calculated Pourbaix diagrams for the model a) Ir-het, b) Ir-hom,
and c) Ir-hyb catalysts. Insets of the most relevant structures are shown. The
coverages for the Ir-het system are labeled as mHb-nOHcus-kOcus, in which
mHb is the number of H atoms at the Obridge positions, and mOHcus and kOcus
are the number of OH and O groups in CUS positions. Structures for the Irhom and Ir-hyb systems are named as nH, in which n is the total number of
H atoms contained in each system.

meric IrV species is obtained (Figure S4). From this complex, we
have further examined all possible isomers with different hydrogen contents and formal Ir oxidation states up to +7 (Figure S5). Notably, the existence of IrO4 and [IrO4]+ species with
oxidation states of +8 and +9 have been reported,[52, 53] although they have only been characterized in rare-gas matrices.
Therefore, we are only taking into account Ir species with oxidation states up to +7. The Pourbaix diagram constructed if
we consider all the above dimeric Ir complexes (Figure 2 b)
shows that Ir-hom under OER overpotentials has a total of six
H atoms and corresponds to a symmetric IrVII/IrVII dimer with
only hydroxo and oxo ligands in the equatorial and axial
planes, respectively.
As a result of the considerable size of the Ir-hyb catalyst and
the large number of possible protonation states of the adsorbed dimeric Ir complex, for this system we assume the
same surface coverage as obtained for Ir-het (Figure 2 a); that

1794

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
is, a surface with the CUS sites covered fully by O atoms. As far
as the resting state of the anchored Ir dimer is concerned, we
have considered the same consecutive dehydrogenation as investigated for Ir-hom (Figure S3). Based on all these structures
and several further oxidized species with formal Ir oxidation
states up to +6 (Figure S6), we have constructed the Pourbaix
diagram shown in Figure 2 c. We find that there are two Ir-hyb
species that might exist at relevant OER potentials (Figure 2 c,
insets). The first one, stable at lower overpotentials, contains
seven H atoms and corresponds to an Ir species in which the
equatorial plane is fully hydroxylated except for one position
that has an aqua ligand. The second one (denoted hyb-6 H) is
stable at slightly higher overpotentials and is analogous to the
resting state found for Ir-hom. Interestingly, the potentials required to reach the same Ir oxidation states in Ir-hyb are significantly higher than those needed for Ir-hom, which suggests
that the Ir atoms of Ir-hyb are less prone to be oxidized. We
also find that, in contrast to the Ir-hom catalyst, the aqua ligands in the equatorial positions of Ir-hyb are more difficult to
dehydrogenate because of the formation of H bonds between
these ligands and the Ocus surface atoms.
We next evaluate the OER activity for the Ir-het, Ir-hom, and
Ir-hyb catalysts based on the resting states derived from the
Pourbaix diagrams (Figure 2). To compare the activity of these
three catalytic systems, for each of them we have determined
the potential-limiting step, defined as the onset potential at
which all the OER steps become thermodynamically downhill.
The difference between the limiting and the standard water
oxidation potentials is the theoretical overpotential, h
[Eq. (3)]:[43, 49, 34]
h ¼ maxðDGi =eÞÀ1:23 V

ð3Þ

For rutile oxides, water oxidation in an acidic medium has
been proposed to occur through a WNA mechanism,[49, 54, 55]
which consists of the following four electrochemical steps
(WNA-1; Figure 3 a):
1Þ H2 OðlÞ þO* ! HOO* þeÀ þHþ

ð4Þ

2Þ HOO* ! O2ðgÞ þ* þeÀ þHþ

ð5Þ

3Þ * þH2 OðlÞ ! HO* þeÀ þHþ

ð6Þ

4Þ HO* ! O* þeÀ þHþ

ð7Þ

However, one variant of this mechanism would be to replace
step 2) and 3) by the following alternative steps that do not
entail unsaturated octahedral Ir atoms at the surface (WNA-2,
Figure 3 a):
20 Þ HOO* ! OO* þeÀ þHþ

ð8Þ

30 Þ OO* þH2 OðlÞ ! O2ðgÞ þHO* þeÀ þHþ

ð9Þ

In this case, a proton-coupled-electron transfer (PCET) reaction from the HOO* intermediate yields a molecularly adsorbed
OO* species that subsequently evolves to molecular oxygen
ChemCatChem 2016, 8, 1792 – 1798

www.chemcatchem.org

Figure 3. Gibbs energy diagrams for the OER activity of the a) Ir-het, b) Irhom, and c) Ir-hyb (with six H atoms) catalysts through the conventional
and alternative WNA (green and black lines, respectively) and I2M (blue
lines) mechanisms.

and a HO* group generated by a second PCET. According to
our calculations, the thermodynamically most energy-demanding step in the WNA-1 mechanism corresponds to O2 desorption, whereas in the WNA-2 it is the generation of the HOO*
species (step 1), which involves the formation of the OÀO
bond. These reaction steps result in overpotentials of 0.61 and
0.40 V for the conventional WNA-1 and alternative WNA-2
mechanisms, respectively (Figure 3 a). Thus, this alternative
pathway is more favorable by 0.21 V, and therefore, we will
consider this mechanism to compare the OER performance of
Ir-het with the other model catalysts. Moreover, the calculated
overpotential of 0.40 V through the WNA-2 mechanism is in
agreement with the reported experimental value of 0.36 V at
a current density of 10 mA cmÀ2.[9]
For homogeneous water oxidation catalysts, both the WNA
and I2 M reaction mechanisms have been shown to operate,
which depends on the nature of the catalyst.[47] Thus, for Irhom we have explored these two mechanisms and several variants of thereof (Figure S7). In the case of the WNA mechanism,
we find that the lowest energy pathway is the same as that for
Ir-het (WNA-2; Figure 3 a), in which the O*!HOO* transition is
the potential-limiting step (Figure 3 b). Nevertheless, the calcu-

1795

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
lated theoretical overpotential for Ir-hom is significantly higher,
that is, 1.02 V, than that of Ir-het.
However, the lowest energy path obtained for the I2M
mechanism consists of the following steps:
1Þ O* þO* ! * OO*

ð10Þ

2Þ * OO* þH2 OðlÞ ! OO* þHO* þeÀ þHþ

ð11Þ

3Þ OO* þHO* þH2 OðlÞ ! 2 HO* þO2ðgÞ þeÀ þHþ

ð12Þ

4Þ 2 HO* ! HO* þO* þeÀ þHþ

ð13Þ

5Þ HO* ! O* þeÀ þHþ

ð14Þ

This mechanism starts with the coupling between two iridium bis-oxo moieties that lead to a m-1,2-peroxo intermediate
(Figure 3 b). From this species, the cleavage of one IrÀO bond
is promoted by the nucleophilic attack of a water molecule to
result in the formation of a superoxide and an OH group generated by PCET. This superoxide species subsequently evolves
to molecular oxygen with the coordination of a second water
molecule, which gives rise to an OH group through a second
PCET. Finally, the starting bis-oxo species is regenerated after
two consecutive PCET reactions to complete the catalytic
cycle.
The potential at which all the electrochemical steps in the
I2M mechanism become downhill in energy is 1.57 V (Figure 3 b). However, a calculation of the transition state for the
I2M mechanism reveals an activation barrier of 0.76 eV. Notably, this chemical step is not affected by the applied bias. The
surmountable barrier together with the fact that the overpotential for the I2M mechanism is much lower than that of the
WNA-2 (0.34 vs. 1.02 V) suggests that these two mechanisms
are at least competitive at a certain temperature.
For the Ir-hyb catalyst, analysis of the Pourbaix diagram suggests the existence of two different resting states in the range
of typical OER overpotentials (purple and gray regions in Figure 2 c). Therefore, we have examined the activity of these two
species by considering both the WNA and I2M mechanisms for
each of them. For the two species, we find the same lowest
energy paths and potential-limiting steps as for the Ir-hom
system (Figure S7). In particular, for the Ir-hyb species that contains seven H atoms, we find that all the electrochemical steps
in the WNA-2 and I2M mechanisms are downhill in energy at
potentials of 2.24 and 1.50 V, respectively (Figure S8). However,
the I2M mechanism is hampered by the high chemical energy
required to form the m-1,2-peroxo intermediate (0.98 eV).
However, for the Ir-hyb species with six H atoms, all the electrochemical steps in the WNA-2 and I2M mechanisms become
downhill at 2.18 and 1.76 V, respectively. In this case, we find
that the formation energy for the m-1,2-peroxo intermediate is
rather low (0.39 eV), and therefore, it may be surpassed even
at room temperature. Based on these thermodynamic data, we
predict that the I2M mechanism for this species may be more
favorable than WNA-2. In other words, the height of the chemical step and the overpotential for the WNA-2 mechanism for
this species are lower than that of the one that contains seven
ChemCatChem 2016, 8, 1792 – 1798

www.chemcatchem.org

H atoms, which suggests that Ir-hyb with six H atoms may be
more representative for the OER activity of this catalyst.
Taken together, our results show that the intrinsic OER activity of the different model Ir-based catalysts through a WNA-2
mechanism increases as follows: Ir-hom % Ir-hyb < Ir-het. With
regards to the I2M mechanism, the comparison between the
Ir-hom and Ir-hyb systems is more complicated because of the
concurrence of electrochemical and chemical reaction steps.
Nevertheless, the fact that Ir-hom requires a slightly lower
overpotential compared to Ir-hyb (0.34 vs. 0.53 V) but a higher
chemical barrier (0.59 vs. 0.39 eV) indicates that these two catalysts might exhibit a comparable OER activity. The comparison
of the overall activity between the three model Ir-based catalysts is also rather complex, as the lowest energy-demanding
pathway for both Ir-hom and Ir-hyb is the I2M mechanism,
which involves a chemical step, whereas for Ir-het it is the
WNA-2 mechanism. If we assume low energy barriers for the
chemical steps for Ir-hom and Ir-hyb, the activity trend is then
dictated by the applied overpotential: Ir-hom > Ir-het > Ir-hyb.
Although this trend is not conclusive, the high OER activity
predicted for Ir-hom would support that this species might be
part of the purple-blue solution as suggested by Crabtree
et al.[45] However, the same authors also proposed that this
species eventually forms IrO2 nanoparticles, and thus it might
not be very stable in aqueous solution.
Finally, to shed light on the role of the surface in Ir-hyb, we
performed the Bader charge analyses presented in Figure 4.

Figure 4. Calculated oxygen (red) and iridium (blue) Bader charges for the
resting states of the a) Ir-hyb, b) Ir-hom, and c) Ir-het catalysts.

We find that the electronic charge of the O atoms in the resting states of the different Ir-based catalysts (top axial oxo
groups) increase in the order: j qIr-het j < j qIr-hyb j % j qIr-hom j . This
indicates that the oxo moieties in the Ir-hyb and Ir-hom catalysts are less electrophilic than that in Ir-het. Consequently, the
oxo groups in Ir-hom and Ir-hyb are less reactive towards the
nucleophilic attack of a water molecule, which accounts for
the predicted activity trend through the WNA-2 mechanism. In
addition, we find that the charge of the Ir atoms of the active
site varies within the different model catalysts, which points to
an increase in oxidation state as follows: Ir-het < Ir-hyb < Irhom. This trend is tightly coupled to the electronic charge of
the O atoms involved in the reaction. We also observe that the
charge of the O linkers of Ir-hyb and the bottom oxo moieties

1796

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
in Ir-hom (Figure 4 a and b) are similar, which suggests that the
former behave more similarly to oxo ligands than to hydroxo
groups (Figure S9). This can be ascribed to the additional
charge donation provided by the IrO2(11 0) surface to the O
linkers in the Ir-hyb catalyst.

Conclusions
In this work, we employed DFT calculations to evaluate the viability of a hybrid water oxidation catalyst generated by attaching a dimeric Ir complex (Ir-hom) to the IrO2(11 0) surface (Irhet). Our results indicate that the interaction between the adsorbed catalyst and the surface is rather strong as a result of
the electron donation from both Ir atoms in the anchored
complex and the surface to the O linkers. The oxygen evolution reaction (OER) activity of the hybrid catalyst (Ir-hyb) is further assessed and compared to its constituent fragments. The
activity trend for the water nucleophilic attack (WNA) mechanism is Ir-hom % Ir-hyb < Ir-het. If we assume that the chemical
barriers in the interaction between two MÀO units (I2M) mechanism for Ir-hom and Ir-hyb systems are feasible under OER
conditions, the catalytic activity may be governed by the applied potential to result in the trend: Ir-hom > Ir-het > Ir-hyb.
This is supported by the calculated activation barrier for the
I2M mechanism with Ir-hom, which should be surmountable at
room temperature. The high activity of Ir-hom shown in our
calculations is consistent with the similar structures suggested
to be involved in the formation of the purple-blue solution.[45]
Importantly, our calculations predict no significant activity loss
if the dimeric Ir complex is anchored on the IrO2(11 0) surface
through O atoms. Thus, by attaching the dimeric Ir complex to
an IrO2(11 0) surface through O linkers, the stability of the homogeneous catalyst might be enhanced, as calculations show
that this interaction is very strong.
Taken together, the results reported herein are encouraging
as they suggest that hybrid catalysts are not only viable but
they might also benefit from the activity of homogeneous catalysts and the stability of heterogeneous ones. Our findings
are in line with recent experimental findings for a stable and
water-oxidation-active dimeric Ir complex anchored on a metal
oxide surface.[33] The here demonstrated phenomena for
a hybrid system based on iridium oxide should be generalized
to more affordable materials. Furthermore, our findings suggest that there is ample room to improve the catalytic performance of hybrid catalysts by fine-tuning the ligands of the
homogeneous constituent and the electrochemical properties
of the heterogeneous constituent.

Acknowledgements
The authors acknowledge support from the DOE Office of Basic
Energy Science to the SUNCAT Center for Interface Science and
Catalysis and SLAC National Accelerator Laboratory LDRD program. M.G.-M. acknowledges funding from the Agency for Administration of University and Research Grants of Catalonia
(AGAUR, 2013 BP-A 00464). L.V. acknowledges funding from
ChemCatChem 2016, 8, 1792 – 1798

www.chemcatchem.org

MINECO (CTQ2014-54071-P) and MECD (FPU fellowship). N.L. acknowledges funding from funding from ICIQ and computational
resources from BSC-RES. The authors would like to thank Prof.
Jens K. Nørskov and Dr. Michal Bajdich for helpful discussions.
Keywords: density functional theory · heterogeneous
catalysis · homogeneous catalysis · iridium · reaction
mechanisms
[1] T. B. Johansson, Renewable energy: sources for fuel and electricity, Island
press, Washington, D.C., 1993.
[2] R. Pike, P. Earis, Energy Environ. Sci. 2010, 3, 173.
[3] D. Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res. 2009, 42, 1890 – 1898.
[4] K. S. Joya, Y. F. Joya, K. Ocakoglu, R. van de Krol, Angew. Chem. Int. Ed.
2013, 52, 10426 – 10437; Angew. Chem. 2013, 125, 10618 – 10630.
[5] I. Katsounaros, S. Cherevko, A. R. Zeradjanin, K. J. J. Mayrhofer, Angew.
Chem. Int. Ed. 2014, 53, 102 – 121; Angew. Chem. 2014, 126, 104 – 124.
[6] H. Dau, C. Limberg, T. Reier, M. Risch, S. Roggan, P. Strasser, ChemCatChem 2010, 2, 724 – 761.
[7] H. Inoue, T. Shimada, Y. Kou, Y. Nabetani, D. Masui, S. Takagi, H. Tachibana, ChemSusChem 2011, 4, 173 – 179.
[8] L. Duan, F. Bozoglian, S. Mandal, B. Stewart, T. Privalov, A. Llobet, L. Sun,
Nat. Chem. 2012, 4, 418 – 423.
[9] C. C. L. McCrory, S. Jung, I. M. Ferrer, S. M. Chatman, J. C. Peters, T. F. Jaramillo, J. Am. Chem. Soc. 2015, 137, 4347 – 4357.
[10] J. D. Blakemore, R. H. Crabtree, G. W. Brudvig, Chem. Rev. 2015, 115,
12974 – 13005.
[11] A. Llobet, Molecular Water Oxidation Catalysis, Wiley, West Sussex, UK,
2014.
[12] P. C. K. Vesborg, T. F. Jaramillo, RSC Adv. 2012, 2, 7933 – 7947.
[13] R. Tagore, R. H. Crabtree, G. W. Brudvig, Inorg. Chem. 2008, 47, 1815 –
1823.
[14] C. S. Mullins, V. L. Pecoraro, Coord. Chem. Rev. 2008, 252, 416 – 443.
[15] C. Zhang, C. Chen, H. Dong, J.-R. Shen, H. Dau, J. Zhao, Science 2015,
348, 690 – 693.
[16] M. W. Kanan, D. G. Nocera, Science 2008, 321, 1072 – 1075.
[17] P. Du, R. Eisenberg, Energy Environ. Sci. 2012, 5, 6012 – 6021.
[18] J. A. Haber, Y. Cai, S. Jung, C. Xiang, S. Mitrovic, J. Jin, A. T. Bell, J. M. Gregoire, Energy Environ. Sci. 2014, 7, 682 – 688.
[19] D. Friebel, M. W. Louie, M. Bajdich, K. E. Sanwald, Y. Cai, A. M. Wise, M.-J.
Cheng, D. Sokaras, T.-C. Weng, R. Alonso-Mori, R. C. Davis, J. R. Bargar,
J. K. Nørskov, A. Nilsson, A. T. Bell, J. Am. Chem. Soc. 2015, 137, 1305 –
1313.
[20] M. García-Mota, A. Vojvodic, F. Abild-Pedersen, J. K. Nørskov, J. Phys.
Chem. C 2013, 117, 460 – 465.
[21] G. Novell-Leruth, G. Carchini, N. López, J. Chem. Phys. 2013, 138,
194706.
[22] E. W. McFarland, H. Metiu, Chem. Rev. 2013, 113, 4391 – 4427.
[23] K. Fominykh, P. Chernev, I. Zaharieva, J. Sicklinger, G. Stefanic, M. Dçblinger, A. Müller, A. Pokharel, S. Bçcklein, C. Scheu, T. Bein, D. Fattakhova-Rohlfing, ACS Nano 2015, 9, 5180 – 5188.
[24] J. Suntivich, E. E. Perry, H. A. Gasteiger, Y. Shao-Horn, Electrocatalysis
2013, 4, 49 – 55.
[25] M. Wiechen, I. Zaharieva, H. Dau, P. Kurz, Chem. Sci. 2012, 3, 2330 –
2339.
[26] B. Li, F. Li, S. Bai, Z. Wang, L. Sun, Q. Yang, C. Li, Energy Environ. Sci.
2012, 5, 8229 – 8233.
[27] A. D. Doyle, J. H. Montoya, A. Vojvodic, ChemCatChem 2015, 7, 709 –
709.
[28] J. J. Concepcion, R. A. Binstead, L. Alibabaei, T. J. Meyer, Inorg. Chem.
2013, 52, 10744 – 10746.
[29] D. K. Zhong, S. Zhao, D. E. Polyansky, E. Fujita, J. Catal. 2013, 307, 140 –
147.
[30] B. M. Klepser, B. M. Bartlett, J. Am. Chem. Soc. 2014, 136, 1694 – 1697.
ˇ
[31] E. Farnetti, R. Di Monte, J. Kaspar in Inorganic and Bio-inorganic Chemistry, Vol. II, Eolss Publishers, Oxford, 2007.
[32] A. Savini, A. Bucci, M. Nocchetti, R. Vivani, H. Idriss, A. Macchioni, ACS
Catal. 2015, 5, 264 – 271.

1797

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[33] S. W. Sheehan, J. M. Thomsen, U. Hintermair, R. H. Crabtree, G. W. Brudvig, C. A. Schmuttenmaer, Nat. Commun. 2015, 6, 6469.
[34] J. K. Nørskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, J. R. Kitchin, T. Bligaard, H. Jónsson, J. Phys. Chem. B 2004, 108, 17886 – 17892.
[35] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865 –
3868.
[36] G. Kresse, J. Furthmüller, Phys. Rev. B 1996, 54, 11169 – 11186.
[37] G. Kresse, J. Furthmüller, Comput. Mater. Sci. 1996, 6, 15 – 50.
[38] P. E. Blçchl, Phys. Rev. B 1994, 50, 17953 – 17979.
[39] H. Zhang, A. Soon, B. Delley, C. Stampfl, Phys. Rev. B 2008, 78, 045436.
[40] F. G. Sen, A. Kinaci, B. Narayanan, S. K. Gray, M. J. Davis, S. K. R. S. Sankaranarayanan, M. K. Y. Chan, J. Mater. Chem. A 2015, DOI: 10.1039/
C5A04678E.
[41] A. A. Bolzan, C. Fong, B. J. Kennedy, C. J. Howard, Acta Crystallogr. Sect.
B 1997, 53, 373 – 380.
[42] Y. Lee, J. Suntivich, K. J. May, E. E. Perry, Y. Shao-Horn, J. Phys. Chem. Lett.
2012, 3, 399 – 404.
[43] J. Rossmeisl, A. Logadottir, J. K. Nørskov, Chem. Phys. 2005, 319, 178 –
184.
[44] S. Klüpfel, P. Klüpfel, H. Jónsson, J. Chem. Phys. 2012, 137, 124102.
[45] U. Hintermair, S. M. Hashmi, M. Elimelech, R. H. Crabtree, J. Am. Chem.
Soc. 2012, 134, 9785 – 9795.
[46] S. Romain, L. Vigara, A. Llobet, Acc. Chem. Res. 2009, 42, 1944 – 1953.

ChemCatChem 2016, 8, 1792 – 1798

www.chemcatchem.org

[47] X. Sala, S. Maji, R. Bofill, J. García-Antón, L. Escriche, A. Llobet, Acc.
Chem. Res. 2014, 47, 504 – 516.
[48] C. C. L. McCrory, S. Jung, J. C. Peters, T. F. Jaramillo, J. Am. Chem. Soc.
2013, 135, 16977 – 16987.
[49] J. Rossmeisl, Z.-W. Qu, H. Zhu, G.-J. Kroes, J. K. Nørskov, J. Electroanal.
Chem. 2007, 607, 83 – 89.
[50] I. C. Man, H.-Y. Su, F. Calle-Vallejo, H. A. Hansen, J. I. Martínez, N. G.
Inoglu, J. Kitchin, T. F. Jaramillo, J. K. Nørskov, J. Rossmeisl, ChemCatChem 2011, 3, 1159 – 1165.
[51] S. Siahrostami, A. Vojvodic, J. Phys. Chem. C 2015, 119, 1032 – 1037.
[52] Y. Gong, M. Zhou, M. Kaupp, S. Riedel, Angew. Chem. Int. Ed. 2009, 48,
7879 – 7883; Angew. Chem. 2009, 121, 8019 – 8023.
[53] G. Wang, M. Zhou, J. T. Goettel, G. J. Schrobilgen, J. Su, J. Li, T. Schlçder,
S. Riedel, Nature 2014, 514, 475 – 477.
[54] M. García-Mota, A. Vojvodic, H. Metiu, I. C. Man, H.-Y. Su, J. Rossmeisl,
J. K. Nørskov, ChemCatChem 2011, 3, 1607 – 1611.
[55] Y.-H. Fang, Z.-P. Liu, J. Am. Chem. Soc. 2010, 132, 18214 – 18222.

Received: January 4, 2016
Revised: February 18, 2016
Published online on April 29, 2016

1798

 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim