![Figure 2: Gibbs free energies of assumed reaction intermediates of the oxygen evolution reaction (OER) on a (1 1 0) RuO2 surface covered with O. The Gibbs free energies are depicted for three different electrode potentials: 0 V, 1.23 V (equilibrium potential of the overall reaction) and 1.60 V vs. standard hydrogen electrode (SHE). At 1.60 V vs. SHE the OER becomes thermodynamically feasible (∆RG≤0 eV under the considered conditions). Reprinted (adapted) with permission from reference [22].[28]](/figures/figure-2-gibbs-free-energies-of-assumed-reaction-3fx56mcm.webp)
![Figure 5: Binuclear homogenous oxygen evolution catalyst (blue dimer). Reprinted with permission from reference [31]. Copyright (2014) American Chemical Society.](/figures/figure-5-binuclear-homogenous-oxygen-evolution-catalyst-blue-34ym2vjn.webp)
![Figure 3: Schematic representation of (1 1 0) surface of a rutile type oxide like RuO2. The coordinatively unsaturated (cus) as well as the bridge sites are labeled. One ion in a bridge site is exchanged by a lower valent Ni ion. An OH and an OOH group are adsorbed on a cus site, respectively. In a) the different surface species are labeled. In b) possible proton transfer is exemplified, as the bridging O related to Ni are activated as proton donor-acceptor functionality.[19] Bond angles and length of adsorbed intermediates are provided as schematic representation and do not represent the actual state during the reaction.](/figures/figure-3-schematic-representation-of-1-1-0-surface-of-a-bs4yy3ju.webp)
![Figure 4: a) Illustration of the Acid-Base and Direct-Coupling OER mechanism classified according to the O-O bond formation step. Reprinted (adapted) with permission from reference [31]. Copyright (2014) American Chemical Society. b) Detailed reaction mechanism proposed for the oxygen evolution reaction on [RuII(bpy)(tpy)H2O]2+ as homogeneous catalyst (acid-base type). All intermediates of the reaction are numbered to be easily referable within the text. Reprinted (adapted) with permission from reference [40]. Copyright (2014) American Chemical Society.](/figures/figure-4-a-illustration-of-the-acid-base-and-direct-coupling-9rvynfb0.webp)
![Figure 8: OER performance of thin-film Ir oxide model catalysts in form of overpotential and Ir dissolution as a function of the calcination temperature (constant Ir loading). Overpotentials and integral Ir dissolution during 10 min were measured at a geometric current density of 2 mA cm-2. Overpotentials were taken from reference [90] and Ir dissolution results from reference [96].](/figures/figure-8-oer-performance-of-thin-film-ir-oxide-model-2i4aergt.webp)
![Figure 7: OER performance in form of overpotential and metal dissolution rates at 5 mA cm-2 geometric current density of different metal electrodes. The data are obtained from reference [76], were they were obtained by scanning flow cell measurements in conjunction with ICPMS[78] (electrolyte: 0.1 M H2SO4, scan rate: 2 mV s-1, iR corrected).](/figures/figure-7-oer-performance-in-form-of-overpotential-and-metal-1ogjd0a8.webp)
1
Electrocatalytic Oxygen Evolution Reaction in Acidic Environments
‒ Reaction Mechanisms and Catalysts
Tobias Reier
1
, Hong Nhan Nong
1
, Detre Teschner
2
, Robert Schlögl
2
and Peter Strasser
1,3
*
1
The Electrochemical Energy, Catalysis and Materials Science Laboratory,
Department of Chemistry, Chemical Engineering Division,
Technical University Berlin,
Strasse des 17. Juni 124, 10623 Berlin, Germany
2
Department of Inorganic Chemistry, Fritz-Haber-Institute of the Max-Planck-Society,
Faradayweg 4-6, 14195 Berlin, Germany
3
Ertl Center for Electrochemistry and Catalysis,
Gwangju Institute of Science and Technology, Gwangju 500-712, South Korea
Abstract
The low efficiency of the electrocatalytic oxidation of water to O
2
(oxygen evolution reaction-
OER) is considered as one of the major roadblocks for the storage of electricity from renewable
sources in form of molecular fuels like H
2
or hydrocarbons. Especially in acidic environments,
compatible with the powerful proton exchange membrane (PEM), an earth-abundant OER
catalyst that combines high activity and high stability is still unknown. Current PEM-compatible
OER catalysts still rely mostly on Ir and/or Ru as active components, which are both very scarce
elements of the platinum group. Hence, their amount in OER catalysts has to be strictly
minimized in order to facilitate the economically competitive large-scale application of PEM
electrolyzers. Unfortunately, the OER mechanism, which is the most powerful tool for catalyst
optimization, still remains unclear. In this review, we first review the current state of our
understanding of the OER mechanism of PEM-compatible heterogeneous electrocatalysts, before
we compare and contrast that to the OER mechanism on homogenous catalysts. Thereafter, an
overview over monometallic OER catalysts is provided followed by a review of current material
optimization concepts. Moreover, missing links required to complete the mechanistic picture as
well as the most promising material optimization concepts are pointed out.
2
1. Introduction
Renewable electricity generation technologies, like wind and solar power, are promising
candidates to achieve a clean and sustainable energy infrastructure. However, wind and solar
power are both characterized by an intermittent availability.
[1]
Thus, a large scale energy storage
solution is required in order to bridge the time gap between supply and demand. Molecular fuels
like hydrogen or hydrocarbons produced from renewable electricity and water or, respectively,
CO
2
can provide such a long term chemical energy storage solution.
[1, 2, 3]
Considering hydrogen,
its reconversion to electrical energy can be efficiently performed in fuel cells.
[4]
In a transition
period, hydrogen can additionally be used as fuel for combustion engines, which underlines its
versatility. Besides that, hydrogen and hydrocarbons can be appropriately used for mobile
applications due to their comparably high gravimetric energy density.
[2]
The electrocatalytic
production of molecular fuels like hydrogen from water or hydrocarbons from CO
2
is based on
reduction reactions which, in turn, require an electron donating counter reaction. In this context,
the electrocatalytic oxidation of water to molecular oxygen, the oxygen evolution reaction
(OER), is the most promising candidate with regard to availability and sustainability.
[5]
Additionally, the OER constitutes a common counter reaction in metal electrowinning.
[6]
Hence,
the OER is not only a key step for electricity storage but is furthermore of outmost importance in
other processes. Unfortunately, the OER is a complex multistep reaction, which adds a
considerably large overpotential to the actual process and, thus, distinctly reduces the process
efficiency even if current benchmark catalysts are applied.
[5]
Additionally, the inherent high
electrode potentials during the OER are demanding with respect to the catalysts stability.
In the context of water electrolysis for renewable electricity storage, proton exchange membrane
(PEM) electrolyzers offer great advantages compared to alkaline electrolyzers such as lower
ohmic losses, higher voltage efficiency, higher gas purity, a more compact system design, higher
current density, a faster system response and a larger partial load range.
[7, 8]
The aforementioned
advantages are directly or indirectly related to the PEM, which is an acidic solid polymer
electrolyte membrane. In particular, the PEM ensures a small gas crossover and provides a high
proton conductivity.
[7]
Since the gas crossover is rather independent of the applied load, gas
crossover becomes problematic at low loads where gas production rates are low.
[7]
In this case,
the transport rate of H
2
through the membrane can be high enough to form H
2
-O
2
mixtures that
approach or exceed the explosion limit, which has to be strictly avoided for safety reasons.
[9]
Therefore, electrolyzers can only be operated above a certain load. Comparing PEM and alkaline
electrolyzers, this minimal load is commonly much smaller for PEM electrolyzers, due to the
comparably small gas crossover of PEMs.
[7, 9]
Based on the large load range and the fast system
response, PEM electrolyzers offer a great flexibility to respond to the intermittent electricity
generation from renewable sources. Furthermore, the high proton conductivity of PEMs ensures
low ohmic losses and the applicability of high current densities,
[7]
which are not only
advantageous in the context of renewable electricity storage. In contrast to the well-established
3
fully-developed PEMs, alkaline solid polymer electrolytes are currently under development but
commercial alkaline electrolyzers still rely on liquid electrolytes in combination with
diaphragms.
[7, 10]
Furthermore, all alkaline electrolytes have the intrinsic drawback that the
equivalent conductivity of hydroxide ions (198 S cm
2
mol
-1
)
[11]
is considerably lower than that of
hydronium ions (350 S cm
2
mol
-1
)
[11]
. Thus, acidic electrolytes (membranes) potentially provide
(at similar thickness and charge carrier concentration) much lower ohmic resistances, which is
especially relevant to minimize losses at high current densities on the application level.
Besides the aforementioned disadvantages of alkaline electrolyzers, their major advantage is the
comparably wide range of abundant materials that are applicable as anode catalysts, such as Fe,
Ni, Co, Cu and Mn based oxides as well as nitrogen doped carbon materials.
[12, 13]
However,
while OER catalysis profits from alkaline conditions, the cathode reaction, the hydrogen
evolution reaction (HER), is usually impeded in an alkaline environment.
[14]
This circumstance
lowers the possible gain on the system level. In contrast to alkaline electrolyzers, one main
shortcoming of PEM electrolyzers is the limited range of materials for the anode catalyst and
related parts such as current collectors and separator plates
[7]
, since these materials must sustain
high electrode potentials in combination with the acidic environment. The vast majority of PEM
compatible anode catalysts is based on oxides of Ru and, especially, Ir which are, unfortunately,
very scarce noble metals
[7]
with annual production capacities far below that of Pt
[15]
. Hence,
pricing and availability of Ru and Ir can be considered as a major issue for the large-scale
application of PEM electrolyzers. Thus, in order to profit from the numerous advantages of PEM
electrolyzers and facilitate their large-scale application, the noble metal amount required in the
anode catalyst need to be drastically reduced. For this purpose, an in-depth fundamental
understanding of the OER mechanism and the applied catalyst materials in acidic environment is
required. Based on this knowledge, new strategies for catalyst design and optimization can be
established to minimize the noble metal content while preserving a high activity and stability for
the OER.
In this review, first the different OER mechanisms proposed in acidic environment are reviewed.
As a next step, the present status of the experimentally determined OER mechanism of
homogenous Ru catalysts is summarized to provide the basis for the subsequent discussion of in-
situ analytical insights from heterogeneous OER catalysts. Then, to provide insights into
structure-function relations of PEM compatible OER catalysts, activity and stability trends of
monometallic OER catalysts are reviewed. Based on this knowledge, new catalyst optimization
approaches are discussed to point out their potential for future research.
2. The oxygen evolution reaction mechanism in acidic environment-
Current state of knowledge
4
2.1. Heterogeneous catalysts
To date, a number of different reaction mechanisms have been proposed for the OER on
heterogeneous electro-catalysts, based on kinetic studies
[16, 17, 18]
or theoretical density functional
theory (DFT) based calculations
[19-22]
, some of which are shown in Figure 1. However, none of
the reaction mechanisms proposed for heterogeneous catalysts has been yet fully validated based
on experimental results. In a pioneering work, Bockris made up kinetic models for a variety of
different conceivable OER mechanisms, some of which are shown in Figure 1 I-III.
[16]
Bockris
demonstrated that the Tafel slope, observable in an electrocatalytic measurement, is determined
by the actual rate determining step (rds) within a certain reaction mechanism.
[16]
This analysis
was based on the assumption that one step in each reaction mechanism is the rds and that only the
reactant of the rds can build up a considerably high surface coverage (concentration).
[16]
Considering RuO
2
, the Tafel slope analysis revealed a reaction mechanism similar to the
electrochemical oxide path (see Figure 1 II) with an additional chemical rearrangement step of
the M-OH species between reaction 1 and 2.
[18, 23]
Below 1.52 V this rearrangement step was
found to be rate determining whereas at higher potentials the first step (step 1 in Figure 1 II)
becomes rate determining.
[18, 23]
However, a rds cannot unambiguously be identified based on the
Tafel slope alone. Different rds in different mechanisms can result in similar Tafel slopes.
[16]
Furthermore, the actual reaction mechanism might not have been considered in the set of
mechanism for which the Tafel slopes have been deduced. Moreover, the Tafel slope itself is a
somewhat unspecific measure which can be altered by factors besides the electrocatalytic
reaction. Scheuermann et al. have for instance shown that a semiconducting oxide layer, located
between catalyst and substrate (current collector), can increase the Tafel slope.
[24]
Thus, precise
knowledge about the electrodes material properties, especially with respect to conductivity and
possible buried interfaces, is required in order to obtain valid mechanistic insights from a Tafel
slope analysis.
Figure 1: Proposed reaction mechanisms for the oxygen evolution reaction. Reaction
5
mechanisms I-III were taken (adapted) from reference [16] whereas reaction mechanism IV was
obtained from reference [22].
Another concept based on which an OER mechanism has been proposed is the thermochemical
analysis using DFT calculations, as demonstrated by Rossmeisl et al.
[20-22]
Here, first a
mechanism is proposed (see Figure 1 IV). Then, the Gibbs free energy of every reaction (∆
R
G
x
)
within the mechanism is calculated as a function of the electrode potential. In this context, only
elementary reactions in which an electron is exchanged with the electrode are considered, since
only these steps depend directly on the electrode potential. To facilitate the overall reaction, as a
necessary condition, ∆
R
G
x
of every reaction step has to be ≤ 0 J mol
-1
(compare Figure 2).
Although the sum of ∆
R
G
x
of the individual reaction steps has to equal ∆
R
G of the overall
reaction (oxidation from water to O
2
), each reaction step can have a different ∆
R
G
x
within the
mentioned boundary condition. If one electron is transferred to the catalyst in each step, ∆
R
G
x
of
each step changes equally with electrode potential. The reaction step which displays the largest
∆
R
G
x
(at the reference potential of 0 V
SHE
) requires the highest electrode potential to be become
downhill in ∆
R
G (step 3 in Figure 2 requires 1.60 V) and, thus, is referred to as the potential
determining step (pds).
[25]
Hence, there can be reaction steps that require a higher electrode
potential than the standard potential of the overall reaction (E
0
=1.23 V) to meet the condition of
∆
R
G
x
≤ 0 J mol
-1
, see Figure 2. In contrast to the overall reaction, ∆
R
G
x
of the reaction steps (see
Figure 1 IV and Figure 2) depends on the adsorption energy of the intermediates and, thus, is a
function of the catalyst. Hence, the potential required to facilitate the overall reaction is a
function of the catalyst. Since this approach describes the reactivity trend on different catalysts
fairly good, the validity of the proposed mechanism appears reasonable, although the model is
based on thermodynamics alone and does not account for any kinetic barrier. This treatment,
however, does not mean that there are no kinetic barriers but it assumes that the kinetic barriers
scale with the thermodynamic barriers and, thus, the reactivity trend can be qualitatively
explained on a thermodynamic basis. Additionally, reaction mechanism IV includes the implicit
assumption that proton and electron transfer are coupled in every step. Indeed, Nakagawa et al.
found that the OER overpotential on a heterogeneous Ir oxide catalyst does not depend on the pH
value, which supports the assumption of a coupled proton-electron transfer.
[26, 27]
However, the
occurrence of a coupled or, rather, a sequential proton electron transfer can depend on the
interaction strength between catalyst and intermediates and, hence, might be a function of the
actual catalyst.
[27]