Angewandte: Chemie

Angewandte
A Journal of the Gesellschaft Deutscher Chemiker
International Edition Chemie www.angewandte.org
Accepted Article
Title: Principle of Water Electrolysis and Recent Progress of Cobalt,

Nickel, and Iron-based Oxides for Oxygen Evolution Reaction
Authors: Mingquan Yu, Eko Budiyanto, and Harun Tüysüz
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202103824
Link to VoR: https://doi.org/10.1002/anie.202103824

Angewandte Chemie International Edition 10.1002/anie.202103824
REVIEW
Principle of Water Electrolysis and Recent Progress of Cobalt,

Nickel, and Iron-based Oxides for Oxygen Evolution Reaction
Mingquan Yu, Eko Budiyanto, and Harun Tüysüz*
Dr. M. Yu, E. Budiyanto, Priv.-Doz. Dr. H. Tüysüz

Department of Heterogeneous Catalysis
Max-Planck-Institute für Kohlenforschung
Kaiser-Wilhelm-Platz 1, 45470, Mülheim an der Ruhr, Germany
E-mail: [email protected]
Accepted Manuscript
Frontispiece Graphic
(18.5 cm in diameter)
1
This article is protected by copyright. All rights reserved.
REVIEW
Abstract: Water electrolysis that results in green hydrogen is the key
process towards a circular economy. The availabilities of the
sustainable electricity and oxygen evolution reaction (OER)
electrocatalyst are the main bottlenecks of the process for large-scale
green hydrogen production. A broad range of OER electrocatalysts
has been explored to decrease the overpotential and boost the
kinetics of this sluggish half-reaction. Co, Ni, and Fe-based catalysts
have been considered to be potential candidates to replace noble
metals due to their tunable 3d electron configuration and spin state,
versatility in the terms of crystal and electronic structures as well as
abundance in nature. This review provides some basic principles of
water electrolysis, key aspects of OER, and significant criteria for the
Accepted Manuscript
development of the catalysts. It provides also some insights on recent
advances of Co, Ni, and Fe-based oxides and a brief perspective on
green hydrogen production and the challenges of water electrolysis. Scheme 1. A typical water electrolysis cell under alkaline condition.
A thermodynamic potential of 1.23 V is required to split water into

H2 and O2. However, typically much higher applied voltage is
1. Introduction necessary to get over the energy barrier mainly related to kinetic
and mass transportation.[4] This is mainly determined by the
Water electrolysis-by using sustainable electricity—is a promising overpotentials to overcome the activation energy of the HER and
process to split water into green H2 and O2 gases. The H2 OER. Additionally, some potential is required to get over the
production demand is rapidly increasing and receiving more resistances of electrolyte, contact, membrane as well as parasitic
attention since it can be used as a feedstock and energy carrier resistance of side reactions like electrode corrosion[5] and
for diverse applications including in the chemical industry, dioxygen reduction.[6] The contact resistance is further affected by
transportation, and power segments. Moreover, green H2 the diffusion of reactants and gas products. So, the overall applied
production and consumption do not cause air pollution and waste. potential (Eap) for water electrolysis can be described as in Eq.6
Thus, it offers the solution to reduce the carbon footprint or even where ηHER and ηOER are the overpotentials while ηother represents
decarbonize some key industrial processes like Haber-Bosch the total voltage drops resulting from the system resistances.
ammonia synthesis and steel production that count for a large
Eap = 1.23 V + ηHER + ηOER + ηother (6)
amount of CO2 emission.[1] This makes the green H2 production a
key element to reach targets of the European Commission and Under the ideal conditions, a theoretical value of 3.55 kWh of
the Paris Climate Agreement towards being carbon neutral by electricity is needed to produce one liter of H2. This amount is
2050.[2] Besides, water electrolyzers can be easily integrated into increased to 4.26 kWh when the total voltage drop is taken into
photovoltaic and (on) off-shore wind farms for on sites H2 account.[7] The overpotential for both half-reactions is determined
production and usage. The price of renewable electricity is getting by accessible ions near the electrode surfaces and the activation
more comparable to the fossil fuel-based power plants, which energy to overcome the kinetics limitations of the electrochemical
makes the green H2 more affordable in comparison with the grey reactions. To increase the concentration of accessible reactant
H2 from the conventional steam reforming process.[3] ions, it is essential to speed up the ion diffusions in the electrolyte.
For instance, vigorous stirring or increasing the operating
Water splitting is an endothermic reaction and requires an energy
temperature of electrolytes could accelerate the diffusion of ions.
input of ΔG = 237.1 kJ/mol under standard conditions (25 oC,
1 atm), as shown in Eq.1. The hydrogen evolution reaction (HER) The activation energy for reaction kinetic could be significantly
and oxygen evolution reaction (OER) take place at the cathode reduced by using electrocatalysts. Among both half-reactions,
and anode, respectively, and they are preferably separated by a OER is triggered by multiple electron transfer and oxygen-oxygen
membrane as illustrated in Scheme 1. Depending on the pH of formation, which is thermodynamically more challenging than the
the electrolyte, HER and OER are proceed in different reaction two electrons transfer process in HER.[8] Therefore, high-energy
pathways. In an acidic electrolyte, water is oxidized to O2 gas by input for OER has been a key factor in limiting the overall
forming the protons at the anode. Protons are transferred to the efficiency of electrochemical water splitting. Developing OER
cathode as charge carriers and get reduced to H2 to complete the electrocatalysts with high activity and robust stability is crucial to
electrochemical circuit (Eq. 2-3). Under alkaline conditions, reach target of large-scale green H2 production for sustainable
hydroxyl anions play the role of electrochemical charge carriers, future energy and industrial landscapes.
which are formed by water reduction at the cathode and oxidized Over the past decades, Ir and Ru-based oxide electrocatalyst
at the anode to produce H2 and O2, respectively (Eq. 4-5). have been widely used in industrial proton exchange membrane
Water splitting: 2H2O(l) → 2H2(g) + O2(g) (1) water electrolyzers.[9] However, these noble metal-based
materials are not suitable for practical large-scale H2 production
In acidic electrolyte:
due to their cost as well as supply limitation.[10] Alternatively, non-
HER: 4H+ + 4e- → 2H2(g) (2)
precious metal-based catalysts have shown promising activity
OER: 2H2O(l) → 4H+ + O2(g) + 4e- (3)
towards OER, and thus attracted massive research interest,
In alkaline electrolyte: especially the first-row transition metals like Mn, Fe, Co, and Ni.
HER: 2H2O(l) + 4e- → 2H2(g) + 4OH- (4) In recent years, a wide range of active OER catalysts consist of
OER: 4OH- → O2(g) +2H2O(l) + 4e- (5) these metals have been reported, which are prepared in various
form, i.e. oxides,[11] (oxy)hydroxides,[12] phosphides,[13] borides,[14]
2
REVIEW
carbides,[15] nitrides,[16] arsenides,[17] sulfides,[18] selenides,[19] and step involves the formation of M-OOH intermediate by the
tellurides.[20] Among them, the oxides hold the greatest potential nucleophilic attack of OH- ion on M-O intermediate, where the
for industrial application despite that higher catalytic activity could third electron is feed into the circuit. Subsequently, a hydroxide
be achieved on other metal compounds, such as (oxy)hydroxides ion attacks the M-OOH intermediate and forms molecular O2, and
for the OER process.[21] The preparation of TMOs is more feasible releases the fourth electron. Although there are other proposed
in scaled-up production compared to the synthesis of other mechanisms,[21b, 24d, 26] the OER pathways are always associated
counterparts and thus, it is more appealing for practical utilizations. with the bonding interactions of the metal centers and
Unfortunately, so far, there has not been a significant intermediates, which have a key impact on the reaction kinetics.
breakthrough in this field at the level of commercialization since
Corrigan accidentally formed an active Ni-Fe oxide electrocatalyst
when studying the effect of impurity of Fe in NiO thin film.[22] For
most reported TMO-based OER catalysts, a significant
Accepted Manuscript
overpotential (~ 300 mV) is still required to reach a current density
of 10 mA cm-2 in 1 M KOH electrolyte.[23] Hence, there is still a
need for the design and development of more effective, durable,
sustainable, and economical materials for OER.
In this review, we present the basic principle of water electrolysis
by focusing on OER, and key electrochemical terms for the
evaluation of the electrocatalysts. We also discuss crucial aspects
of OER catalysts including their chemical and structural properties
as well as the defects and catalyst´s alteration under the operating
conditions by focusing on Co, Ni, and Fe-based oxides. In the end, Scheme 2. An alkaline water-splitting cell, the magnification shows a
a perspective is provided by discussing challenges and further generalized mechanism of OER in the alkaline medium over metal
electrocatalyst.
prospects in the development of water electrolysis for green H2
production and its impact on the future energy landscape. To reach fast reaction kinetics, all intermediates in each
elementary step of OER should be favored by the surface energy
and bonding interactions. The standard free energy change (ΔG)
2. Electrochemical oxygen evolution reaction of the OER is calculated to be 4.92 eV and a thermodynamically
ideal OER catalyst would require the same minimal free energy
The oxidation of water is a fundamental reaction involving in the
(ΔGs) of 1.23 eV (4.92 eV/4 = 1.23 eV) to drive the formation of
photosynthesis system that produces the molecular O2 and water-
each of oxygen intermediates, which are *OH, *O, *OOH, and O2,
derived electrons, which are used to convert CO2 to biomass, food,
respectively.[27] When either of the intermediates is strongly
and fuel. It is very challenging to have a solid insight into the
bonded to the active metal sites, a higher ΔGs is required for the
reaction process due to the alteration of the surface structure of
corresponding electron-transfer step, resulting in an overpotential
the electrocatalyst under operating conditions, availability of the
to drive the OER process. Practically, an overpotential is always
small number of catalytically active centers as well as their short
present with either high or low values depending on the type of
life-time.[6] However, with the rapid advance of high-resolution and
catalysts. This is due to the different bonding strengths between
operando techniques, great progress has been done in the last
the surface of the metal catalyst and the oxygen intermediates.
decades. [21c, 24] Below, we discuss some of those progress and
The fast diffusion of reactants and the quick release of molecular
important aspects of OER including the reaction kinetic and main
oxygen are essential to obtain high catalytic efficiency.
parameters to evaluate the performance of the electrocatalysts.
2.1. Reaction kinetics
The OER is a multiple-step reaction involving several adsorption

and desorption of surface intermediates, e.g. oxo, peroxide, and
superoxide species. The initial step of OER in the alkaline
electrolyte is the adsorption of hydroxide radical on the active
sites of the electrocatalyst.[21b, 24b] A simplified OER pathway on
the metal sites was proposed by Song and the coworkers as
depicted in Scheme 2.[25] The active metal site is displayed as “M”,
and the sequence of step reactions is shown by the direction of
the arrows. As the first step, the active metal site adsorbs a
hydroxide ion (OH-) to form the initial M-OH intermediate along
with the release of one electron. Another hydroxide ion reacts with
M-OH intermediate with coupling proton and removing electron to
form H2O molecular and M-O intermediate. Two different
pathways are available to produce molecular O2 from M-O Figure 1. The OER volcano plot of overpotentials of diverse electrocatalysts at
intermediate. The first one is through the direct combination of two 1 mA/cm2 vs energy differences (ΔG*O - ΔG*OH). Reproduced with permission.[28]
M-O species. The second one, which is seen as a more Copyright 2017, American Association for the Advancement of Science.
representative pathway, contains two reaction steps. The first
3
REVIEW
First-principles periodic density functional theory (DFT) better charge transportation.[30] In practical alkaline water
calculations are applied on broad classes of metal oxides to revisit electrolyzer, typically 5-7 M of KOH or NaOH solution is used as
the origin of the overpotential of OER, which revealed a solid electrolyte and both cells are separated with Zirfon membrane
correlation between the binding energy of *OOH and *OH (ZrO2-coated polyphenyle sulfide mesh). The system runs at
intermediates.[27] Thus, a universal descriptor based on free ambient pressure or the cell is pressurized up to 30 bar.[31]
energy and the binding energy of *O and *OH intermediates (ΔG*O
The laboratory temperatures are typically fluctuated depending on
- ΔG*OH) was introduced to reveal the origin of the overpotential.[28]
the seasons and even days’ time and this can impact the
An OER volcano plot was built up based on this descriptor as
electrolyte surroundings and the activity of the catalyst. Thus, it is
shown in Fig.1. Overpotentials at 1 mA/cm2 from experimental
suggested to use an external thermostat to have a constant cell
results turn out to have a good match with the overpotentials from
temperature. In general, electrochemical OER follows the
the theoretical calculation when employing ΔG*O - ΔG*OH as the
Arrhenius’ law where the reaction rate is proportional to the
descriptor. As seen in the volcano plot, the catalysts with optimum
temperature. However, an irregularity could be found in some
Accepted Manuscript
binding strengths like IrO2, NiFeOx, and CoOx exhibit superior
transition metal-based oxides due to the variation of rate
OER activities with lower overpotentials. Thus, design of an OER
determining step in certain temperature range. Pang et al recently
catalyst with neither too strong nor too weak bonding to oxygen
reported that for cobalt-based catalysts at optimum temperature
intermediates is key to reach faster reaction kinetic.
of 65 °C, the rate-determining steps were shifted to the
combination of adsorption and the cleavage of the OH group.[32]
2.2. Key parameters for the evaluation of the OER activity This temperature is close to the operation temperature of the
commercial alkaline water electrolyzer (60-80 °C).[31, 33] It is also
The OER performance of the electrocatalysts is very sensitive to
recommended to use a cell made of Teflon instead of glass at
experimental conditions, however, most of the used methods are
corrosive alkaline electrolytes since the silica can be leached from
not standardized and their apparatus and experimental conditions
the glassware and affects the experimental conditions.
largely fluctuate. This makes challenging to compare the
performance of different materials. In the following section, critical Electrode type and configurations: different types of materials
aspects of the experimental conditions, configuration of electrode are used as electrodes and their background current and
kits, and other key features of OER measurements are elaborated. durability vary depending on the experimental conditions. The
Some of the key aspects are summarized in Scheme 3. most often reported working electrodes are glassy carbon (GC),
Experimental conditions: surrounding of electrodes including nickel foam (NF), carbon fiber paper (CFP), and conductive thin
pressure, temperature, pH, types, and concentration of electrolyte film coated glass. Due to the surface properties and varied
play an essential role in the OER activity of the electrocatalysts. conductivity of these materials, a dramatic difference can be
The electrolyte provides pathways for ions to stream between observed in the OER performance regardless of the catalysts. For
electrodes and maintains charge balance. For transition metal instance, Ni foam has a very high background current in an
oxides, the lab-scale OER measurements are typically conducted alkaline electrolyte as the surface of NF gets easily converted to
under alkaline conditions (pH 12-14) by using KOH solution as the nickel (oxy)-hydroxide that shows very good OER activity.[34] The
electrolyte. Even a small amount of Fe impurities of KOH conductivity as well as the activity of the catalyst could be also
electrolyte have been shown to change dramatically the OER enhanced by depositing on appropriate substrate. For instance,
performances of Co, Ni, and Fe-based electrocatalysts.[29] Co, Mn, Ni and Fe-based catalysts function more effective when
Besides the type of electrolyte, its concentration plays a they are deposited on Au substrate, while Ti substrate gives a
substantial role in the fact that the higher concentration assists synergistic effect when it is coupled with Ir-based catalyst.[35]
Scheme 3. Summary of important key parameters to evaluate OER activity including experimental condition and evaluation.
4
REVIEW
In a typical three-electrode cell configuration, the counter 2.2.1 Overpotential and Tafel slope
electrode (usually Pt) provides a means of applying input potential
to the working electrode while the reference electrode (like An overpotential is required to assist the adsorption and
Hg/Hg2Cl2, Hg/AgO, Ag/AgCl) holds a constant potential and desorption of OER intermediates, and it is described as activation
assists in measuring the potential of the working electrode. The potential. This value can be lowered by a good choice of catalyst
electrode should be chosen carefully depending on the with optimized binding energy. There are two more types of
experimental conditions, for instance, Ag/AgCl electrode is not potentials, which can be called concentration and resistance
suitable for very high pH since the diffusion of OH- would cause overpotential. The concentration overpotential is caused by the
the formation of Ag2O and block the frit. Metal-based counter diffusion limitation from the local drop of reactant concentration
electrodes tend to dissolve in strong acid solutions. Reversible near the interfaces, and it can be minimized by rotating the
hydrogen electrode and Hg/HgO reference electrode (filled with working electrode or stirring the electrolyte. In an electrochemical
KOH solution) are better options for highly alkaline condition. cell, additional resistances from the electrolyte, wire, and
Accepted Manuscript
connections are always present, and they cause voltage growth
Working electrode preparation and binding agent effect: the
according to Ohm’s law (V = I*R). Since these resistances are
obtained OER results are also firmly affected by electrode
rather stable in a cell, the resistance overpotential can be offset
preparation methods, the loading of catalyst, and the binder,
simply by applying resistance compensation to the linear sweep
which affects the electrode-catalyst-electrolyte interface.
voltammetry (LSV) curves.[39]
Commonly, electrocatalyst materials are dispersed in a binder-
contained liquid to form a homogeneous ink, which is afterward The overpotential (ƞ) at a specific current density per geometric
deposited onto the clean working electrode surface via drop- area is used as a primary indicator to evaluate the OER activity.[40]
casting, spray coating, sputtering or electrophoretic deposition. Typically, the overpotential required to achieve a geometric
Depending on the types of the working electrode, a proper current density of 10 mA/cm2 (denoted as ƞ10 ) is used, which is
deposition method should be used to fabricate a uniform thin film. originated from the potential of solar cells to reach 10% of
The binder should be carefully chosen to avoid the suppression efficiency.[41] In the case of metal electrodes, such as NF
of OER activity of investigated materials. The binder should have electrodes, the “background” current is considerably large, hence,
good adhesive properties to the surface of the working electrode the overpotential at much larger current densities (100-
and good conductivity to facilitate the mass and charge 1000 mA/cm2) is used as the indicator.[42]
transfer.[29c] Fulfilling both requirements, Nafion® 117 The second most commonly used evaluation indicator is the Tafel
(perfluorinated polymer with sulfonate groups) is the most slope that reveals the dependency of the compensated
commonly used binder. The preparation of working electrode with overpotential on the current density.[25, 43] The Tafel slope is an
binder is commonly applied in lab-scale applications. For real important kinetic indicator and could be calculated from the Tafel
application, such as in alkaline water electrolyzer, the electrodes equation (Eq.7) where ƞ is the overpotential for OER, b is the
are commonly prepared from Ni mesh/plates via vacuum plasma Tafel slope while j is the measured current. [44]
spray with Ni-Al alloy or electroplating in case of Ni-Zn. Either Al ƞ = a + b log j (7)
or Zn is then leached with KOH by forming Raney Nickel
For a multiple electron transfer reaction, the Tafel slope provides
catalyst.[33, 36] Another common method to fabricate electrodes for
valuable information on the reaction mechanism since its value is
alkaline water electrolyzer is electrodeposition or Ni electroplating
dependent on the charge transfer coefficient and the number of
on stainless steel mesh.[31, 37]
transferred electrons. The Tafel slope also provides a direct
Catalyst loading is another important aspect, the low loading of correlation between the current changes and the applied voltage.
the electrocatalyst could lead to inhomogeneity and leave an A smaller value of the Tafel slope indicates a faster increase on
uncoated electrode surface. Hence, the current normalization to the catalytic current when applying higher voltage. Similar to the
geometric surface area will not be straightforward. On the other overpotential, the interpretation of the Tafel slope suffers from the
hand, high loading amount could lead to the materials stacking capacitive current. A large error could be made due to the different
and create a rough electrode surface. It might hinder the mass capacitive current measured at changed potential scan rates.
transport as well as block some active sites, resulting in lower Also, the Tafel slope could differ depending on the overpotential
mass-specific activity. region. For instance, an unusually higher Tafel slope (340 mV/dec)
Measurement techniques: individual methods are usually was reported on a diamond electrode at an overpotential of
adapted to assess the catalytic properties accordingly. The < 1.2 V, in comparison to the value (121 mV/dec) obtained in the
comparison of measurements from varied methods could lead to high overpotential region (> 1.2 V, as shown in Fig.2a).[45] Such a
contradictory results. The condition and timing of the high Tafel slope was considered as a result of the surface function
measurements are also critical since the alteration of groups hindering the kinetics of OER.[46] At a higher overpotential
electrocatalyst (e.g. corrosion, contamination, and re-construction) region, the same electrode gave a Tafel slope of around
would lead to deactivation or activation process. Since these 120 mV/dec, which fits the theoretical value.[45]
aforementioned factors can change the OER results, variable Another important aspect that influences the Tafel slope
catalytic performances are often reported on the same type of determination is Ohmic (iR) drop that mainly originated from the
material from different research groups.[38] Appropriate evaluation electrical resistance of electrolyte solution as well as distance
parameters and indicators are required to quantitatively compare between working electrode and reference electrode in three-
the catalytic performances, which is below further elaborated. electrode system. This disturbance should be measured
experimentally and compensated before performing the OER
measurements by using Electrochemical Impedance
Spectroscopy (EIS) or current-interrupted methods. EIS method
5
REVIEW
is the most common method to determine Ohmic drop that could 2.2.2 Electrochemical Impedance Spectroscopy (EIS)
be easily detected by observing the intercept of real part of
impedance for high frequencies in Nyquist plot. Current- Charge transfer at the interface of electrocatalyst and electrolyte
interrupted method uses the basic principle of Ohm’s law, is another important descriptor for the OER. EIS can elucidate
principally the current step is applied, then the resistance is such charge transfer behavior and the nature of reaction at the
calculated when the current is applied and interrupted. Nowadays, interface. Electrochemical impedance is typically measured by
most of the potentiostats are equipped with the feature to applying an alternating current (AC) potential to an
compensate Ohmic drop. However, Ohmic drop is generally electrochemical cell and then measuring the current. EIS is a one-
compensated at 80-90% since overcompensation may cause point measurement by inducing small perturbation amplitude (E0
oscillation in the measurement.[47] In practical alkaline water within the 1 - 10 mV of magnitude) of the applied potential over
electrolyzer, the Ohmic drop effect is more severe compared to in the time E(t) in various frequencies (ω) as shown in Eq. 8.[48] In a
proton exchange membrane (PEM) water electrolyzer due to the linear system (where the response is directly proportional to input),
Accepted Manuscript
cell geometry and usage of liquid electrolyte. the measured current response, I(t), is then shifted in phase (φ)
and has a different amplitude (Eq. 9). The impedance is then
stated as resistance in AC by analogous Ohm’s law (Eq. 10).
E(t) = |E0| sin ωt (8)

I(t) = |I0| sin (ωt + φ) (9)
Z(ω) = E(t) / I(t) (10)
(a)
(b)
Figure 2. (a) LSV curves (1) before and (2) after Ohmic drop correction on the
boron-doped diamond electrode. Inset shows the corresponding Tafel plots.
Reprinted with permission.[45] Copyright 2018, Elsevier. (b) Plots of NiO loading
on the electrode against the value of ƞ10 and the Tafel slope, which were derived
from the LSV result after Ohmic drop correction. Reprinted with permission.[41]
Copyright 2019 American Chemical Society.
Figure 3. Nyquist plots were obtained from EIS measurements for Co3O4, NiO,
Although the use of these two indicators, ƞ10 and the Tafel slope, and mixed spinel oxides before (a) and after (b) electrochemical activation.
can offer a fair comparison on the catalytic performance of OER Reprinted with permission.[29a] Copyright 2017 American Chemical Society.
catalysts in a similar system,[40, 43] they do not give a holistic
The EIS data are typically interpreted by using Nyquist or Bode
picture of the intrinsic activity of an OER catalyst. As pointed out
plots. In the Nyquist plot, as shown in Fig.3, the impedance is
in a recent study,[41] the value of ƞ10 and the Tafel slope are very
plotted in a complex plane against a real plane (ZIm vs ZRe), giving
much dependent on the mass loading of the catalyst. With the
a typical semicircle relation. The Bode plot describes the
increasing NiO loading on the working electrode, a decreasing
magnitude of impedance |Z| and the phase angle φ as a function
trend was shown on the values of these two evaluation
of frequency.[49] The Nyquist plot is widely used to interpret EIS
parameters, illustrating an enhancement in the catalytic activity as
data of OER due to its direct correlation to solution resistance (Rs)
seen in Fig.2b. However, as it is well-known, the intrinsic activities
and charge transfer resistance (Rct).[50] The experimental
and properties of the materials are independent of the mass and
impedance data should be always fitted to the equivalent circuit
loading amount of the catalyst. Therefore, the primary use of ƞ10
model to avoid bias in interpretation and gain relevant information
and the Tafel slope are not indicative of the intrinsic activity rather
of the aforementioned parameters. This equivalent circuit model
reflecting the ability of an electrocatalyst on catalyzing OER from
could be built by several elements, such as resistor (R),
the engineering point of view. This could lead to false
capacitance (C), constant phase element (Q), impedance of
interpretations to evaluate the catalytic activity without knowing
diffusion or Warburg diffusion (Zw), and other elements. For OER
their dependence on the loading.
6
REVIEW
catalyzed by transition metal oxide, generally the intercept at high
frequency at the real plane (axis) is related to the electrolyte
solution resistance (Rs), while the diameter of semicircle could be
directly translated into polarization or charge transfer resistance
(Rct). Fig.3 shows Nyquist plots to evaluate the charge transfer
resistance of Co3O4, NiO, and their mixed oxides before and after
electrochemical alteration in the KOH electrolyte. After CV
measurements, nickel-containing electrocatalysts got activated
due to the Fe impurities uptake from the electrolyte. This causes
a decrement of charge-transfer resistance of nickel-based
catalysts—measured by the diameter of the semi-circle in the
Nyquist plot—and enhancement of the reaction kinetic.[29a]
Accepted Manuscript
2.2.3 Electrochemical surface area (ECSA)
Figure 4. Carbon oxidation and relevant current evaluation (iC) along with the
OER (iOER) of carbon-supported nickel boride catalyst. Adapted with
To shed light on the intrinsic activity of the electrocatalyst, it is permission.[5] Copyright 2019, Wiley-VCH Verlag GmbH & Co.
essential to estimate the amount of catalytically active sites.
ECSA can be used to normalize the measured currents for the The competing oxidation of the reactants can be triggered in the
evaluation of intrinsic activities and the determination of the presence of anions as well. For instance, Cl- oxidized chlorine
number of active centers. In principle, changing the catalyst (chlorine evolution reaction) is a major reaction in the electrolysis
loading does not impact the ECSA-normalized specific activity. of brine.[54] Applying higher voltage dramatically speeds up these
However, practically the charge transfer ability of electrode is side reactions and thus lowers the Faradaic efficiency of the OER
largely affected by catalyst loading, which can also influence catalyst. Therefore, measuring Faradaic efficiency is critical for
mass transportation. Thus, the OER activity can be the evaluation of OER activity as it provides the catalytic current
underestimated for the high loading and ECSA cannot be used as rather than the total measured current. There are several methods
a universal indicator to evaluate OER performance. to calculate the Faradaic efficiency, which are mainly based on
the quantification of the produced gases through diverse methods
Several methods are available to determine the ECSA, the widely
like a sensor, rotating ring disc electrode (RRDE)[8], or differential
reported are electrochemical double-layer capacitance (EDLC),
electrochemical mass spectrometry (DEMS)[55]. For Faradaic
redox peak, impedance, hydrogen under-potential deposition,
efficiency calculation with RRDE, the measurement is conducted
and CO stripping.[40] The determination of EDLC is well defined
with chronopotentiometry at 1 mA/cm2 geometric current density.
through the employment of cyclic voltammograms by changing
Faradaic efficiency is denoted as the relative value of measured
scan rate to determine the capacitive current associated with
ring current (ir) to the disc current as shown in Eq. 11.[23] The disc
double-layer changing.[51] This method is commonly employed to
current is then calculated by multiplying constant disc current (id)
evaluate the ECSA of the material with the distinct redox couple.
with the collection efficiency of RRDE (N).
The determination of ECSA is favored via redox peaks, especially
for 3d transition metal-based catalysts. In the cases of noble = (11)
metal-based catalysts, H2 under-potential deposition and CO
stripping are widely used.[52] Regardless choice of the method, a 2.2.5 Chronopotentiometry and chronoamperometry plots
certain deviation should be expected for the calculation of ECSA. for catalyst stability and durability
2.2.4 Faradaic efficiency A desirable OER catalyst should exhibit not only high catalytic
activity but also long-term durability and stability. The catalytic
Faradaic efficiency is a quantitative indicator to determine the stability can be assessed by performing chronopotentiometry or
selectivity of the electrocatalysts. It reflects the ratio of the chronoamperometry experiments to monitor the potential change
electrons provided by the oxygen intermediates on the active sites at a constant current or current variation at a fixed potential,
through multiple-step reactions, to the total amount of electrons respectively.[29b, 56] A widely reported stability test to monitor the
that go into the external circuit. The Faradaic current is mainly overpotential is keeping the current density at 10 mA/cm2 over 12
originated from the electrochemical oxidation of water, however, hours. The variation of ƞ10 gives hint about the stability of the
significant current can be also obtained from side reactions. For catalyst as well as its activation or deactivation.
instance, corrosion reaction occurs on most of the OER catalysts, A large current density is chosen to inspect durability for industrial
especially when it is combined with carbon-based materials.[53] application.[42a, 57] Spöri et al. reported the stability performances
The thermodynamic equilibrium potential of carbon oxidation is of most active noble metal-based electrocatalysts where the
0.207 V vs. RHE in standard conditions, which is much lower than overpotential fluctuation at 1000 mA/cm2 was used as an
the OER potential. Thus, the oxidation of carbon is unavoidable indicator.[58] As shown in Fig.5, Ru-based electrocatalysts go
in the potential region of OER. Recently, Möller et al. could detect through severe deactivation while Ir-based catalysts are more
the formation of CO2 from carbon oxidation via mass spectrometry, stable by holding their high initial catalytic activities. Besides,
verifying that the measured current is not only coming from the cyclic voltammetry (CV) measurements at a higher scan rate are
water oxidation, as illustrated in Fig.4.[5] considered as an accelerated degradation test whereby up to
thousands of CV scans are conducted to check the catalyst
durability and monitor the evolution of surface redox species.[29c]
7
REVIEW
This can be also used to monitor electrochemical deposition of Fe
impurities from KOH electrolyze, which has been verified to cause
significant activation on Co and Ni-based catalysts.[29a, 29e]
The electronic, surface structures, binding energy, interaction
between the electrocatalysts, and the intermediates govern the
overall efficiency of the catalyst. Accordingly, precise adjustment
of crystal and electronic structures of the materials can tune the
binding energy and improve the OER performance of the
electrocatalysts.[61] For instance, among a range of transition
metal-based electrocatalysts, the optimum overpotential could be
achieved with the Ni-doped β-CoOOH (Fig.6a) whereby the
surface binding energy of intermediate species (ΔGO – ΔGOH and
Accepted Manuscript
ΔGOH) is not too strong nor too weak.[62] A weak binding cannot
adsorb reactants to initiate the reaction while a strong binding
causes a block of reactive sites by either reactants or products.
The adsorption behaviors are mainly governed by the surface and
Figure 5. A summary of catalytic stability of noble-metal-based electrocatalysts electronic structures of the catalyst.[63] A correlation between
tested at 1 A/cm2 in proton exchange membrane water electrolyzer. The
intrinsic activity and eg orbital filling of perovskite crystal structures
overpotential changes (Δη=ηfinal−ηinitial) indicates activity delay, with the results
compiled from the literature.[59] Reproduced with permission.[58] Copyright 2017, was likewise reported.[64] The eg orbital filling impacts the
Wiley-VCH Verlag GmbH & Co. interactions between the binding energy of oxygen intermediates
and the catalyst surface.
3. Criteria for electrocatalyst development
The electrocatalyst should fulfill diverse criteria to be considered

for industrial applications: i) it should be efficient and deliver high
current density at lower applied potential, ii) it should be durable
and not lose its performance, iii) it has to be economical to be able
to compete with grey H2, iv) it has to be sustainable, ecological
and non-toxic, v) it needs to be based on earth-abundant
elements instead of noble metals with the potential of depletion
within next century, vi) it has to be recyclable, vii) raw-material of
the catalyst and its production should not have long-term supply
and geopolitical risks. Among others, these key factors should be
considered for the design and development of OER catalysts.
There are two general figures of merit strategies to enhance the
performance of the OER catalyst from a materials synthesis point
of view. The first one is increasing the number of active centers
by tuning the structural properties of the catalyst (such as
morphology, geometry, dimension, shape, size, and textural
parameters of the materials) and the loading on the working
electrode. The second one is boosting the intrinsic activity of each
active center by tuning intrinsic properties of the materials like
crystal, band, and electronic structures.[28] The combination of
these strategies can lead to improved catalytic performances.
It should be kept in mind that in the case of the most reported
electrochemical system, the active catalytic centers are formed
after the alteration of starting catalyst. Simple immersing catalyst
material into alkaline electrolyte leads to the alteration of its
surface and further electrochemical measurements can cause Figure 6. (a) Theoretical overpotential plot of doped β-CoOOH as a function of
crystal and structural transformation as well as element ΔGOH and ΔGO – ΔGOH. Adapted with permission.[62] Copyright 2013, American
Chemical Society. (b) Water adsorption on cobalt (111) surface of defect-free
dissolution.[21a, 24a] A good understanding of the alteration of the
(left) and Co-defected Co3–xO4 (right), blue-colored numbers denote O-H bond
surface of material helps to reveal the degradation mechanisms length of adsorbed H2O and yellow-colored numbers denote bonds length
and elucidate the activation and deactivation processes towards between adsorbed O and H from water species over Co and O sites of Co3O4.
the development of a more stable catalyst. Therefore, it is crucial Adapted with permission.[65] Copyright 2018, American Chemical Society.
to monitor the alteration of a catalyst via in situ, operando, and
Among other parameters, defects can play an essential role in the
post characterizations.[21c, 24b, 60] For instance, a combination of
OER performances of the catalysts. Various structural defects,
the electrochemical flow cell with an oxygen sensor and an
such as zero-dimensional point defects (vacancy and interstitial),
inductively coupled plasma–optical emission spectrometry (ICP-
one-dimensional line defects (dislocation) as well as two-
OES) system allows simultaneous analysis of electrocatalyst
dimensional planar defects (grain and phase boundaries) could
(de)-activation, activity, and faradaic efficiency of the catalyst.[29d]
8
REVIEW
exist within the structure of electrocatalysts.[66] Among those types operation of commercial water electrolysis.[6, 79] Among TMOs,
of defects, zero-dimensional point defects in terms of cation and perovskites with the general formula of ABO3, where A is
anion vacancies are commonly found in transition metal oxides.[65, commonly rare-earth or alkaline-earth metal, and B is transition
67]
As seen in Fig.6b, the Co-defect sites could draw a hydrogen metal, have been widely studied as remarkable catalyst materials
atom from adsorbed water molecule onto the oxygen site (shorten due to their highly tunable metal compositions, unique electronic
the H-O bond length from 2.117 Å to 1.746 Å) and decreased the structures and robust stability in alkaline solutions.[80] One specific
water adsorption energy (−1.07 eV) compared to defect-free perovskite, i.e. Ba0.5Sr0.5Co0.8Fe0.2O3-δ (BSCF), has drawn special
cobalt oxide (−0.98 eV). This favors better water adsorption and interest since Shao-Horn et al. reported the volcano plot of OER
H-OH bond cleavage during OER. 59] The oxygen vacancy could activity against eg filling.[64] In such a plot, BSCF perovskite
also promote OER activity of the oxide electrocatalyst by lowering locates at the summit with an optimal eg filling that is strongly
the adsorption energy of water molecules. Besides, oxygen related to the binding energies of the surface oxygen. As for
vacancy could provide more adsorption sites for OH• species and drawbacks, the oxides need a very high temperature to be
promote the formation of Co-OOH• intermediate phase.[67b, 68]
Accepted Manuscript
crystallized in the perovskite structure, which requires high energy
demand.[81] Besides, it is challenging to prepare high surface area
The OER performance of the catalysts can be improved by
perovskite with the conventional synthetic protocols since
varying their structural and physicochemical properties through a
particles sinter at high temperatures.[82]
proper synthetic strategy. All in all, OER electrocatalyst should: i)
offer a large surface exposed to reactants, ii) provide reactive Other than perovskites, several oxides and spinels based on first-
sites which can effectively adsorb reactants and have a balanced row transition metals such as Mn, Fe, Co, Ni, and Cu have been
binding to intermediates, iii) own a good electrical conductivity frequently studied as OER catalysts.[11, 29a, 29c, 83] Among them,
allowing efficient electron transfer from reactant to the circuit, iv) more attention has been given to Fe, Co, and Ni-based oxides.
have a porous structure for better mass transportation. Some of Table 1 presents a summary of catalytic performances of state-
these assets will be elaborated more in detail below section. of-the-art Fe, Co, and Ni-based OER electrocatalysts. For most of
the reported oxide catalysts, still, a significant overpotential is
required to achieve a current density of 10 mA/cm2 in 1 M KOH
4. Transition metal oxide electrocatalysts electrolyte. Thus, there is a need for the design and development
of more effective electrocatalysts. For this, it is essential to
The electrocatalysts consisting of first-row transition metals show establish a solid structure-activity correlation from a holistic
good OER activity as RuO2 and IrO2 in alkaline electrolytes.[25, 64, perspective and monitor electrocatalyst under operating
69]
A range of transition metal-based compounds have been conditions to gain some insight into the catalyst alteration,
developed for OER, including; (i) oxides (TMOs),[6, 25] ii) reaction intermediates, catalytically active centers as well as the
(oxy)hydroxides,[70] and iii) MX compounds and alloys (M = reaction mechanism. Some specific examples and recent
transition metals, and X = non-metals or metalloids).[71] Some of progress of Co, Ni, and Fe oxides OER catalysts are discussed in
the non-oxide electrocatalysts like Co2B, Ni3FeN, and Co1-xNixP3 the section below.
revealed stimulating OER activity, with reported overpotential
from 200 to 300 mV at 10 mA/cm2.[72] But, the preparation of MX 4.1. Cobalt oxide-based electrocatalysts
compounds is more cost demanding and they are also not very
stable. MX compounds and alloys are considered as pre-catalysts Since the beginning of the last century, cobalt oxides have been
to form active (oxy)hydroxides sites in alkaline electrolytes.[71] in the focus as a catalyst for OER.[84] The electrochemical
Transition metal-based (oxy)hydroxides have proven so far to be behavior of cobalt oxide films was intensively studied in the
the most active OER electrocatalysts.[70a, 73] In particular, NiFe- 1980s.[85] In 2008, the Nocera group reported the stimulating
based (oxy)hydroxides are reported to show the lowest OER catalytic performance and stability of cobalt phosphate (CoPi) in
overpotential in alkaline electrolytes.[70a] For instance, Gong et al. phosphate buffer.[13] Among the cobalt compounds, cobalt oxides
reported an overpotential of ~ 230 mV to reach 10 mA/cm2 over have received the largest focus due to their activity and durability
Ni-Fe (oxy)hydroxides.[74] More complex (oxy)hydroxides have in alkaline electrolytes.
been also designed as very active OER catalysts, for instance, Cobalt oxide has two common crystal structures, namely CoO and
Fe and W co-doped cobalt oxyhydroxide were reported to have Co3O4. CoO possesses a rock salt structure and consists of two
near-optimal OH- adsorption energies.[75] A gelled FeCoW interpenetrating fcc sub-lattices of Co2+and O2− while Co3O4 has a
oxyhydroxide was reported to deliver a record-low overpotential typical spinel structure where tetrahedral and octahedral sites are
of 191 mV at 10 mA/cm2. Although these (oxy)hydroxides are very occupied by Co2+ and Co3+ cations, respectively. These sites can
active OER catalysts,[76] they are commonly not resistant to “real- be substituted with a variety of transition metal cations, which
world” electrolysis conditions. Most of them are reported to keep influence the electronic structure and catalytic performance.[86]
high activity for a short time at a small current density such as Our research group has been working on the design of well-
10 mA/cm2.[42a] On the other hand, practical water electrolysis defined mesoporous cobalt oxides via the nanocasting method[87]
requires stable electrode catalysts that can work for a longer time and used them as a toolbox to find a structure-activity correlation
with large delivered current densities (e.g., 1000 mA/cm2).[77] as well as to explore the role of the key properties towards the
TMOs have been used as electrode materials since the beginning development of more effective OER catalysts.[4, 56, 88] Our earlier
of the last century[78] since they meet the requirements in terms of studies verified the importance of dimension, geometry, symmetry,
catalytic activity and stability, structural durability, elemental morphology, particle size, and surface area of Co3O4 on its OER
abundance as well as low production cost for the practical performance.[30, 88a]
9
REVIEW
Accepted Manuscript
Figure 7. (a) Electrochemical in situ surface enhanced Raman spectra (SERS) of isotope labelled Co18O2 in 0.1 M purified 16O-KOH after conditioning in 18O-KOH,
(b) in situ SERS of Co16OOH in 0.1 M Fe-free isotope labelled purified KOD, and (c) proposed OER mechanism on cobalt oxyhydroxide. Adapted with permission.[24b]
Copyright 2020, American Chemical Society.
Although Co3O4 is resistant to alkaline solutions, its structural on OO- peak (located around 1050 cm-1) that indicates the
alteration and surface amorphization have been commonly exchange of lattice oxygen and OO– with oxygen species from
observed under electrochemical operating conditions.[89] Dau and electrolyte (Fig.7a). Further D-labelling of KOH electrolyte
Strasser observed a structurally reversible alteration of Co3O4 showed a 10 cm-1 red-shift of A1g peak in compression with
spinel nanoparticles by using in situ X-ray techniques including measurement within normal KOH. This confirms the possible
grazing induced XRD and EXAFS.[90] XRD patterns collected at exchange of OH group with OD in the electrolyte or a hydrogen
different potentials demonstrated that the catalyst film retained bonding interaction on Co-O moieties (Fig.7b). On the other hand,
spinel structure under applied potentials. An in-depth analysis no change in OO– peak position has been observed, pointing out
based on the broadening of Co3O4 reflections illustrated changes that the equivalent mechanism does not occur at superoxide (OO-)
in the mean structural coherence length of the crystallites. The moieties. The mechanism of OER by surface cobalt oxyhydroxide
initial irreversible growth of Co3O4 crystallites occurred under is foreseen to involve the combination of two lattice oxygen atoms
potential up to 1.2 V, which is likely due to Ostwald ripening and/or to produce the Co-superoxide intermediate and the release of
coalescence of the particles.[90-91] Further increasing potential to dioxygen from the Co-superoxide as the rate-determining step
1.62 V resulted in a reversible structural transformation with a (Fig.7c).[24b] The formation of Co(IV) species as intermediate on
lower degree of crystallinity. This structural transformation was the surface of cobalt oxide has been also confirmed by an in situ
attributed to the formation of CoOx(OH)y shell on the Co3O4 electron paramagnetic resonance (EPR) study whereby Co(II)
crystallites. Furthermore, the local atomic structure of Co ions was EPR signal was converted into a Co(IV) with increasing of the
revealed by Fourier-transformed EXAFS over the same potential. electrode potential.[92]
By investigating the modification of Co oxidation state as well as The Co3O4 spinel has the advantage of incorporation of other di-
coordination numbers during OER, reversibly the formation of and trivalent cations, thus their intrinsic properties can be easily
amorphous CoOx(OH)y shell on crystalline Co3O4 core could be adjusted. The integration of a second metal, such as Mg, Ni, Fe,
confirmed. The Co3O4 phase provides the stable crystalline and Cu has been shown to improve the OER activity of Co3O4.[4,
structure in a resting state while the reversible surface CoOx(OH)y 88b, 93]
To investigate the effect of Fe doping and incorporation, we
phase affords active sites for OER. utilized ordered mesoporous silica templates (SBA-15 and KIT-6)
The OER mechanism in CoOOH was recently further postulated to prepare well-defined Co-Fe and Co-Ni oxides as the model
experimentally by using operando XAS and Raman spectroscopy electrocatalyst. A small amount of iron incorporation (Co:Fe
(Fig.7a-c).[24b] An additional isotope labeling experiment has been atomic ratio of 32:1) could alter the pore structure of the replica
performed to probe lattice oxygen and active oxygen species with a more open pore structure (Fig.8a) compared to the long-
exchange within the KOH electrolyte. First, the CoOOH was range order nanowires of pristine Co3O4. Magnification on the A1g
conditioned at 1.75 V in 0.1 M of 18O-KOH electrolyte. The Raman band (Fig.8b), that corresponding to the symmetric
dominating resting state was found to be Co(IV) species Co3+-O stretch of the octahedral sites, showed a clear shift
originated from CoO2. Afterwards the electrolyte was exchanged indicating a distortion in octahedral sites and alteration of the
with 16O-KOH, a blue-shift was observed on A1g and Eg Raman electronic structure with Fe incorporation.[94]
band (peaks at around 470 and 580 cm-1, respectively) as well as
10
REVIEW
Accepted Manuscript
Figure 8. (a) Morphology of mixed cobalt iron oxide (Co/Fe 32) templated from SBA-15 silica (b) Co Kβ-detected HERFD XAS spectra of sample series, (c) A1g
Raman band, and (d) LSV curve of SBA-15 templated cobalt iron oxide series, inset: magnification at 10 mA/cm2. Adapted with permission.[83a] Copyright 2020,
American Chemical Society. (e) Initial LSV curve of mesostructured cobalt nickel sample series and (f) LSV curves after 150 CV scans. Adapted with permission.
[29a]
Copyright 2017, American Chemical Society.
This finding is further supported by X-ray absorption spectroscopy obtained, illustrating that Co3+ Oh is the more optimal geometrical
(XAS) that revealed the increasing ratio of Co in tetrahedral and configuration than Co2+ Td and Co2+ Oh. The calculated Gibbs-
octahedral sites as well as distortion in the crystalline lattice free energy diagram showed a sequence of Co3O4 < MgCo2O4 <
(Fig.8c). The alteration of the electronic structure and the Co2TiO4 < CoCr2O4, matching well with the OER results. The
increase of Co in tetrahedral sites were found to be beneficial to combination of Co3+ Oh and Co2+ Td was found to promote the
enhance the formation of μ-OOH moieties and improve the OER formation of *OOH, leading to superior OER activity due to the
activity with a diluted amount of incorporated iron[95] as seen in lowest energy barrier. Besides the surface energy, spin state and
Fig.8d. Further increase of Fe content is unfavorable for the OER eg orbital play also essential roles for the OER activity.[95, 97]
due to the formation of cobalt ferrite spinel phase with low
Apart from its crystalline phase counterparts, amorphous cobalt
conductivity and poorer charge transferability.[12] All in all, the
oxide has gained a lot of interest in recent years due to its easy
incorporation of Fe tunes the intrinsic properties of cobalt oxide
preparation as well as ability to adsorb Fe impurities from the
by altering the crystal and electronic structures and overall OER
electrolyte.[21a] It has been shown that CoOOH can uptake Fe
performance depending on the loading amount.[83a, 88b] Similar
impurities from KOH electrolyte and form a highly conductive and
concept can be used as a toolbox to investigate the effect of Ni
OER active Co1–xFex(OOH).[12] The co-existence of Fe3+ and CoOx
incorporation into cobalt oxide spinel and to optimize the
gave a synergetic effect and decreases the overpotential at 10
electrocatalyst composition.[29a] As shown in Fig.8e-f, Ni
mA/cm2 up to 69 mV. [98] The XAS measurements confirmed the
containing cobalt oxide samples show different OER activities
decrement of coordination number of Fe and the formation of
after an activation process due to the uptake of iron impurities
oxygen vacancies that boost the OER activity. Going along a
from KOH electrolytes.
similar direction, our group has developed a facile method to
In an analog research-line, to study the role of different catalytic produce highly active amorphous cobalt-based electrocatalyst
sites, Liu et al. prepared three cobalt oxides with the incorporation starting from cobalt(II) nitrate that can be converted to Co-
of Mg2+, Cr3+, and Ti4+ by forming MgCo2O4, CoCr2O4, and oxyhydroxide during OER.[29c]
Co2TiO4. The catalytically non-active Mg2+, Cr3+, and Ti4+ occupy
Although cobalt oxide has a decent OER catalyst it suffers from
Td, Oh, and Td centers, respectively.[96] Therefore, only Co3+ at
its low conductivity. Their conductivities can be improved by
the octahedral sites (Co3+ Oh), Co2+ at the tetrahedral site (Co2+
merging with more conductive materials like graphitic carbon,[99]
Td), and Co2+ at the octahedral site (Co2+ Oh) were present as
gold,[100] and silver.[101] Recently, we have shown the dual
active sites for OER. The electrochemical measurements found
functions of Ag species when they are coupled with mesopores
out that substitution of Co2+ or Co3+ by an inactive element
cobalt oxide.[101] Ag can be merged into 3D mesostructure during
decreased the OER activity of pristine Co3O4. A sequence on the
the hard templating process in the form of metallic Ag and ultra-
OER activity of Co3O4 > MgCo2O4 > Co2TiO4 > CoCr2O4 is
11
REVIEW
small Ag2O nanoparticles. While metallic Ag increased the As an alternative practical post-treatment method, we have
conductivity of the composite, silver oxide moieties provided the recently established a facile selective acid leaching method to
ability of Fe uptake from KOH electrolyte and resulted in a tune textural parameters of Co3O4.[56] Compared to the
significant activation and a two-fold increment of current density aforementioned post-treatment methods, selective acid leaching
at 1.7 V. does not require a complicated and expensive setup, and hence
The surface area, crystal, and electronic structures and it is more favorable for large-scale synthesis. The process was
consequently the OER performances of cobalt oxides could be carried out by utilizing Mo, Li, Ca, or Mg as a sacrificial agent
also tuned via post-treatment methods. Liu et al [102] reported that during the synthesis, which was leached out via the post-selective
heterophase Co3O4/CoO nanosheets via Ar-plasma etching on acid leaching process. The BET surface area of the post-treated
Co3O4 could decrease the overpotential up to 92 mV. DFT Co3O4 could increase up to three-fold, as a result, more active
calculation revealed that the phase transformation of Co3O4 spinel sites could be exposed and favors fast OER reaction kinetic. By
into CoO rocksalt phase is followed by lattice Co diffusion that doing so, the current density at 1.7 V vs RHE of the OER catalyst
could be increased from 70 mA/cm2 to 123 mA/cm2.
Accepted Manuscript
causes oxygen vacancy. The lower p-state energy of adsorbed-O
species on Co3f site of Co3O4/CoO was proved to decrease the It should be kept in mind that regardless crystal structure and
adsorption energy of reactive oxygen species on the catalyst’s crystallinity of cobalt oxide, its dynamic behavior in the water
active sites. Similarly, our group reported the effect of post- environment and electrolyte needs to be always considered. A
treatment on mesostructured Co3O4 and CoO via pulsed laser recent study by Hein et al [24a] utilizing Near-Ambient Pressure X-
fragmentation in liquid.[67a] The laser fragmentation results in the ray Photoelectron Spectroscopy (NAP-XPS) showed that simple
formation of Co3O4/CoO bi-phase as well as decrement of immersion in water could induce the reversible phase
nanoparticle size into sub-5 nm that led to the three-fold increase transformation from tetrahedrally coordinated Co2+ shell and
of BET surface area. Rietveld refinement of high-resolution XRD octahedrally coordinated Co2+ core into octahedrally coordinated
pattern and surface chemical state investigation revealed the Co2+ and partially oxidized octahedrally coordinated Co3+. Thus,
formation of Co vacancy in tetrahedral sites and the oxygen in most cases, the starting material possibly goes through an
vacancy. These vacancies were found to be beneficial to boost alteration in the electrolyte even without an applied external bias.
OER activity by decreasing charge transfer resistance.
Figure 9. (a) CV curves of NiO nanoparticles deposited on a gold-coated electrode. The CV curves were collected in 0.5 M KOH before and after electrochemical
aging labeled as “a” and “b” in the figure. Reproduced with permission.[103] Copyright 2014, Wiley-VCH Verlag GmbH & Co. (b) Illustration of Bode scheme for the
phase transformation on Ni(OH)2. (c) Aging effect on the OER activity of Ni(OH)2 thin films in Fe-free (blue) and unpurified (red) KOH electrolyte. (d) Change of iron
amount in the Ni(OH)2 thin films after aging in unpurified KOH. Reproduced with permission.[104] Copyright 2015, American Chemical Society.
12
REVIEW
and the Tafel slope was obtained as shown in Fig.9c. On the
4.2. Nickel oxide-based electrocatalysts
other hand, Ni-hydroxide catalyst got activated in commercial
KOH with Fe impurity and gradually reached a stable activity after
Ni is more attractive for real application in comparison with Co
4 days of applied potential (Fig.9c). The layered structure of Ni-
since it is more abundant, cheaper, and less toxic, and has fewer
hydroxide efficiently absorbed Fe impurities from unpurified KOH
geopolitical risks.[22, 103, 105] In terms of catalytic OER performance,
solution and up to 20 at. % Fe could be incorporated into the
Ni is as active as Co at a lower potential range based on their ƞ10
catalyst, which could be experimentally confirmed by XPS and
as listed in Table 1. However, pristine NiO has an inferior
ICP analyses (Fig.9d). This indicates that the electrochemical
conductivity and this limits its performance in high voltage range
aging-induced activation on Ni-based catalyst is not due to the
where a large number of electrons are required to be transferred
phase transformation, but rather related to the uptake of Fe
into an external circuit. To overcome the conductivity issue, nickel
impurities from the electrolyte. This kind of Fe-induced activation
oxide is widely combined with other transition metals like Co and
Fe. For instance, Ni-Fe-based oxides have been reported to be was also observed on mixed Ni-Co oxide electrocatalysts.[29a]
Accepted Manuscript
one of the most active ones for alkaline water electrolysis.[22, 106] The OER activity of Ni-based oxides might be further enhanced
The catalytic performances of Ni-Fe oxides strongly depend on by increasing the number of their active cites by coupling with
the specific metal stoichiometry and preparation method. [106c] other transition metals and/or supporting conductive support. For
NiO has the rock salt structure (as CoO) with octahedral Ni2+ and instance, the electrochemical surface area of NiO could be
O2− sites and is very sensitive to the phase transformation. During doubled via self-assembly of Cu-Ni bimetal oxide grown on Ni
foam (NF). In addition, the synergetic effect between Cu and Ni
OER in alkaline electrolytes, NiO is easily transformed to the
improved activity and stability as compared to the single NiO/NF
layered hydroxide/oxyhydroxide phase as shown in Fig.9a.[103] A
electrode.[113] Similarly, it has been shown that Mo could also
narrow anodic and broad cathodic peaks observed in CV curves
promote the OER activity when combined with Ni-based oxide
are ascribed to the NiO/NiOOH redox couple of NiO. Initially, a
catalyst.[42a] In-depth XPS study revealed that Mo5+ species in
stable activity was achieved by a few cycling, as depicted as curve
NiMoO4−x/MoO2 played an important role by delocalizing the
“a” in Fig.9a. When an external voltage is applied, the initial OER
oxygen vacancies and promote the adsorption of water molecules
activity of NiO was significantly increased (b curve in Fig.9a),
which was preliminarily associated with surface amorphization onto the low-coordinated Mo5+ centers. Similarly, W-doped α-
and phase transformation of NiO to Ni(OH)2.[107] Ni(OH)2 electrocatalyst was reported with high efficiency and an
overpotential of 237 mV at 10 mA/cm2 and low Tafel slope of
The activation of nickel-based oxide catalysts has been in the
33 mV/dec in 1M KOH electrolyte.[114] The presence of low spin-
focus of the scientific community for a long while and different
state d0 orbital of W6+ was stated to be the key factor to stabilize
reasons have been given in reported studies; (i) due to the
the O radical and providing vacant d orbital, resulting in an optimal
structural transformation of nickel oxide into a more active
OH- adsorption on the exposed W sites of the Ni(OH)2.
oxyhydroxide species during applying anodic potential,[106a, 108] (ii)
because of trace amount of Fe impurities originated from The activity of Ni oxide has been demonstrated to be further
commonly used KOH electrolyte that can form highly active Ni-Fe increased when it is supported on a substrate. For instance, Bell
oxyhydroxide.[106b, 109] The phase transformation of Ni catalyst was et al. observed that a sub-monolayer of Ni oxide deposited on Au
studied by Bode et al.[110] by monitoring electrocatalyst through showed an order of magnitude higher turnover frequency
several techniques including X-ray and neutron diffractions compared to its nickel oxyhydroxide counterparts.[108] Their in situ
(Fig.9b). Briefly, immersing nickel oxide in alkaline electrolyte Raman study revealed that charge transfer from the oxide to the
turns it into a hydrous α-Ni(OH)2 phase. Upon aging in alkaline highly electronegative Au leads to the formation of Ni/Au surface
solution, α-Ni(OH)2 phase is further transferred to anhydrous oxide, which was found to be responsible for OER activity
β-Ni(OH)2 phase. Applying positive potential causes the oxidation enhancement. Likewise, anchoring nickel oxide into polymer-
of these hydroxide phases into forming γ-NiOOH and β-NiOOH. based carbon nitride was reported to give a similar effect.[115] NiO
A long-term overcharging at high potential converts β-NiOOH to and carbon nitride linked thought Ni–N bond was found to
γ-NiOOH phase. The metal sites in high oxidation state are enhance the conductivity as well as possess a low Gibbs energy
proposed to be beneficial for OER, making whether Ni3+ in for the adsorptions of OER intermediates by resulting in a low
β-NiOOH or Ni3+ in γ–NiOOH an active site, which is a subject of overpotential of 261 mV at 10 mA/cm2. A similar synergy was also
controversy. It has been also reported that nickel was first observed for the NiCeOx–Au system where an overpotential of
oxidized into γ–NiOOH followed by a reversible electrochemical 279 mV at a current density of 10 mA/cm2 was reported.[105a] The
formation of Ni4+ peroxide (NiOO2), which was proposed to be cerium dopant was proven to alter the local electronic structure,
active sites for the high OER activity.[111] resulting in favorable binding energies of the OER intermediates.
Metallic Ni foam is commonly used as the substrate of the working Recently, Strasser et al unraveled the Mars-van-Kravelen
electrode as well as an active OER catalyst.[112] Dipping Ni foam mechanism for OER over Ni(OH)2 and NiFe LDH electrocatalysts
into alkali electrolyte forms OER active nickel oxyhydroxide through continuous substitution of oxygen species of catalyst with
species. The phase transformation of metallic Ni electrode to the species from electrolyte.[117] This finding on lattice oxygen
γ-NiOOH under bias could be confirmed by in situ Raman contribution is in agreement with the OER mechanism over
spectroscopy.[108] The potential cycling transformed γ–NiOOH to CoOOH catalyst as proposed by Hu et al.[24b] The mechanism on
β-NiOOH and this newly formed β-NiOOH was found to be three the metal species, however, seems to be different since large
times more active than γ–NiOOH. However, later on, it was portion of Ni metal centers in the NiFe LDH catalyst remain in +2
reported that the experiments were conducted in unpurified KOH oxidation state under the potential bias instead of oxidized to +4
electrolytes, which contain Fe impurities from industrial as in cobalt counterpart.
production.[104] When the same experiments were conducted in
purified Fe-free electrolyte, significantly increased values of ƞ10
13
REVIEW
Even though the pure Ni-based OER electrocatalyst is often OER activity is mainly governed by the predominant facets.[118] As
suffered from bad electrical conductivity, those problems could be shown in Fig.10a, among α-Fe2O3 with diverse facets, the one
resolved by controlling the structural transformation of starting with high-index exposed facets of (012) shows decent activity. The
material into active Ni oxyhydroxide intermediate or blending catalytic stability was not much influenced by the exposed facets,
them with other hetero elements.[116] Due to its adsorption illustrating the good durability of α-Fe2O3 in the alkaline solution
capability toward Fe species, the OER activity of Ni-based oxides (Fig.10b). As seen in Fig.10c, a lower energy barrier of each step
could be boosted by coupling with iron by forming dynamic active reaction was shown on the high-index facet (012) based on the
sites. All in all, NiFe-based oxides are still considered a promising DFT calculation, which was found to be favorable for the
candidate to be used for large-scale water electrolyzer. adsorption and desorption of oxygenated intermediates.[118]
Instead of being used as a pristine oxide electrocatalyst, Fe is
commonly combined with other transition metals, especially with
4.3. Iron oxide-based electrocatalysts Co and Ni. [29b, 119] Our recent study demonstrated the usage of
Accepted Manuscript
hard templating as a toolbox to prepare Fe2O3 and NiO, and
Fe-based catalysts are very attractive in thermal catalysis optimize the composition of Fe-Ni mixed oxides for OER.[83a] A
because of the abundance of Fe (second most earth-abundant nice synergy between Fe and Ni has been observed whereby the
metal) and some fascinating chemical properties (e.g., various highest OER could be achieved with the Ni/Fe ratio of 32:1. EIS
oxidation states, non-toxicity, and its magnetic property). Among as well as electrochemical surface area analysis revealed that at
different forms of iron oxide, hematite (α-Fe2O3) that is crystallized this stoichiometric ratio, the lowest charge transfer resistance and
in a corundum structure with Fe3+ in octahedral sites, has been in the highest intrinsic activity were reached. The Ni32Fe oxide
the focus photo-electrochemical water splitting due to its good catalyst deposited on the Ni foam reached a geometrical current
light absorption. However, its application as a pristine density of 600 mA/cm2 at 1.7 V vs RHE with long-term stability.
electrocatalyst is severely hindered due to its high resistance that This value is close to the required current density for a large-scale
results in very low OER activity as seen in Table 1. water electrolyzer. Suryanto et al [119] recently reported the role of
interface between Ni metal and γ-Fe2O3. LSV measurements
revealed that the Ni metal domain interconnected to a γ-Fe2O3
has a superior OER activity (with overpotential as low as 210 mV
at 10 mA/cm2) compared to the bare Ni as well as several
benchmark catalysts. The analogous finding was reported by
Song et al with nanoclusters of γ-FeOOH covalently linked to a
γ-NiOOH support electrocatalyst.[120] Based on the DFT
calculations, Fe species were found to act as oxygen-evolving
sites while hydrogen was bonded on the neighboring O site of the
γ-NiOOH. These examples indicate the impact of grain
boundaries on the formation of new active centers for OER.
Among mixed electrocatalysts, Ni-Fe layered double hydroxide
(LDH) catalysts have been investigated intensively due to their
high OER activities.[109b, 121] Trotochaud et al. conducted a
systematic characterization on Ni1-xFex LDH catalysts with a
varied amount of Fe.[109b] Intentional addition of Fe during the
synthesis as well as the Fe impurities presented in the KOH
electrolyte was found to increase OER activity dramatically. The
addition of Fe increases the electrical conductivity of NiOOH by a
factor of 30 and the reaction intermediates are found to bond with
oxidized Fe sites much easier compared to Ni sites in the form of
oxo ligands. Additionally, a decreased stability is rose on metal-
oxo complexes with the addition of d electrons to π* orbitals.[122]
Thus, Fe with a high oxidation state has been widely accepted as
the determinative factor on the activity enhancement. A series of
in situ techniques like XAS and Mössbauer was thus applied and
discovered the presence of Fe4+ in Ni-Fe LDH.[122-123] Applying
anodic voltage induced the formation of Fe4+ with its signature
spin-flip excited-singlet to triplet emission.[122]
This observation could be further supported by an in situ
Mössbauer spectroscopy study, at the beginning in the absence
Figure 10. a) LSV curves and (b) chronopotentiometry measurements for voltage, high spin Fe3+ species were detected with isomer shift δ
(012)-O, (012), (104), and (110) facets exposed α-Fe2O3. (c) Free energy ≈ 0.34 mm/s and quadrupole splitting Δ ≈ 0.46 mm/s.[123] Upon
diagram of OER intermediates. Adapted with permission.[118] Copyright 2018, applying a voltage of 1.76 V, a shoulder appeared in the
Wiley-VCH Verlag GmbH & Co. Mössbauer spectra, which is ascribed to the oxidation of Fe3+ to
Wu et al. investigated the relationship between the exposed Fe4+. After OER and keeping the electrode at the open circuit for
facets and catalytic activity of α-Fe2O3 and concluded that the a while, the shoulder disappeared, indicating that the oxidation of
Fe3+ to Fe4+ is a reversible process for Ni-Fe LDH. It is believed
14
REVIEW
that the formation of Fe4+ is due to the stabilization effect of the 18
O isotopic labeling, operando Raman spectroscopy, and
LDH host lattice. The Fe4+ species, especially the ones located at electro-kinetic analysis (Fig.11a-b).[124] The superficial γ-FeOOH
the edges, corners, and defects, are assumed as the catalytically was detected only in the FeOOH-NiOOH system by Raman
active sites for OER.[70a, 122-123] Furthermore, Markovic’s group spectra, which is reported to be favorable to boost OER activity of
found that Fe active species are dynamically stable as a result of FeOOH-NiOOH compared to NiFe LDH with Fe doped in
dissolution and re-deposition at the electrolyte/host interface.[21a] lattice.[124-125] 18O isotopic labeling coupled with operando Raman
The strong interaction of Fe with the host is the key to control the assisted to formulate the reaction mechanism. In the initial stage
average number of Fe active sites present at the solid/liquid of OER over FeOOH-NiOOH, dimeric Ni(II) species at the dioxo
interface for an efficient OER catalyst. bridge were oxidized into dimeric Ni(III). OER mechanism in
FeOOH-NiOOH then followed bi-functional way where OH- bond
cleavage was followed by oxygen evolution and hydrogen
adsorption to Ni(III)-O species as the rate-determining step. This
is confirmed by the previous reaction mechanism predicted based
Accepted Manuscript
on DFT calculation.[120] On the other hand, OER on NiFe LDH was
initiated by oxidation of Fe(III) into Fe(IV) convoyed by Fe=O
formation on the pre-equilibrium step. The whole reaction is
followed by the conventional four-electron transfer mechanism
where the reaction is governed by the formation of Fe(III)-OOH
intermediate from OH- attack on Fe=O unit.[25, 126]
Finally, despite the low electrical conductivity of pristine Fe oxide,
the properties of Fe-based electrocatalyst could be tailored via
doping or combination with Ni, Co, or other metals. Formation of
Ni-Fe oxide interface or Fe species supported on nickel oxide
remarkably enhance the OER activity by initiating the new active
sites and unconventional reaction pathways. Nevertheless, the
nature of Fe dissolution and corrosion as well as the impurity of
electrolyte should be taken into account in developing a robust Fe
oxide-based catalyst for water electrolysis.
5. Summary and Perspective

Sustainable energy production and usage are going to be the key
features for being climate-neutral and suppressing global
warming. Although solar and wind energy has a huge potential as
renewable energy sources, only around 5% of the world's current
electricity is produced from these sources.[127] This amount has to
be significantly increased. We need to develop also effective
processes and catalytic concepts to convert solar and wind
energy into sustainable chemicals, materials, and storable fuels.
Carbon footprint-free, green H2 can be produced from water
electrolysis by using renewable electricity. This H2 can be stored
and reconverted into energy for power and heat generation to fuel
vehicles for transportation as well as can be used as feedstock to
produce valuable chemicals as illustrated in Scheme 4.
When we think about a carbon-neutral energy roadmap, green H2
production is going to be the bottleneck to reach the vision and
target. With the current technologies, production from water
electrolysis is still very energy-intensive and expensive in
comparison to steam reforming of natural gas. The commercial
HER and OER catalysts are based on Pt, Ir, and Ru, and, the high
cost and limited resources of these noble metals force us to look
Figure 11. Proposed OER reaction mechanisms of a) FeOOH-NiOOH, Fe ion
for alternatives. Thus, there is a need for the design and
is located on the surface of γ-FeOOH cluster and rate determining step is OH−
development of a new class of OER catalysts. The OER catalysts
attack on a Fe=O and H atom transfer to a Ni(III)-O site. b) NiFe LDH, Fe is
should fulfill different requirements like having high performance,
doped in the lattice of Ni LDH with rate determining step OH− attack on Fe=O.
being durable, and economical. On other hand, we should have
Reprinted with permission.[124] Copyright 2021, Wiley-VCH Verlag GmbH & Co.
a holistic approach for catalysts design by including the
recyclability, life cycle assignment, and sustainability.
Supporting that hypothesis, Hu’s group studied the OER Sustainability, at each stage of the production, is going to
mechanism on Ni-Fe LDH and FeOOH-NiOOH catalyst by using essential in fact that we are going to face supply limitations of
15
REVIEW
Accepted Manuscript
Scheme 4. Future energy landscape including usage of sustainable electricity for water electrolysis for green hydrogen production, its storage, and utilization for
power and heat generation, as fuel for transportation as well as feedstock for industrial application.
many elements soon if we keep consuming our resources like we be a good choice to enhance the conductivity of oxide-based
are doing. Thus, it is vital to focus on the development of OER catalysts as well as reduce the cost of the electrode material.
catalysts based on more abundant first-row transition metals. Thirdly, advanced spectroscopic methods are encouraged to
Co, Ni, and Fe-based oxides are promising OER electrocatalysts. resolve the electronic structure of catalytic metal sites, as well as
Stimulating catalytic behaviors are observed on these oxides and monitor the change in the local environment of the catalyst
their activities depend on their crystal and electronic structures, surface under operando conditions, which includes morphology
and the oxidation state. Co3O4 itself is an active and stable OER alteration, phase transformation, and material dissolution.
catalyst, with Co3+ Oh proposed as the most desirable Next, the utilization of machine learning and artificial intelligence
configuration. Its spinel structure can host a variety of transition might be a handful to cut-down computational costs as well as
metals to tune the electronic structure and boost its activity. easier handling of big data sets. As a result, it might shorten the
In comparison to Co, Ni is much cheaper and less toxic. The OER timescale of finding and designing optimum electrocatalyst for
activity of NiO can be significantly enhanced by Fe. NiO is OER.[128] Lastly, it is worth mentioning that joint efforts on the
sensitive to an electrochemical aging process during electrolysis, design of catalyst and separator including diaphragm and
leading to the formation of nickel oxy/hydroxides. These species membrane, the configuration of the electrolyzer, and process
are catalytically active and can be further activated by uptaking engineering are necessary to achieve efficient hydrogen
Fe impurities from KOH electrolyte. The uptake of Fe species is production for the establishment of a sustainable energy system.
the reason for the high activity of Ni-Fe catalysts with Fe4+ species Merging material synthesis and electrode preparation into one
is likely the active sites. step will not only reduce the processing time but also decrease
Fe is very abundant and Fe-based catalysts have been employed the costs of the electrode. For instance, electrochemical
in many industrial processes including Haber-Bosch ammonia deposition and in situ catalyst formation on practical support
synthesis and Fischer-Tropsch gasoline production. However, as materials like metal mesh and conductive carbon papers could
electrocatalysts iron oxides are typically not very active towards make water electrolysis more operative and affordable. Indeed,
OER due to their poor electron transferability. Nevertheless, its those type of catalysts should have long term durability and they
combination with Co and Ni can improve the overall efficiency of should be recyclable.
the electrocatalyst significantly. Raw material mining and production of catalysts have to be
As a perspective for future work, several key aspects should be sustainable without having any substantial environmental and
studied as complementary to these guidelines. Firstly, it is worth geopolitical risks. For instance, cobalt-based materials are very
investigating the mass transfer effect on the nanoscale catalyst, promising for OER, however, Co is not only toxic but has also
especially for highly porous material in which oxygen gas can be geopolitical and supplies risks. Currently, more than half of the
trapped and cover the catalytic sites. Secondly, not only metal like worldwide Co is produced in the Republic of the Congo under
Ag but also surface-modified carbon with improved stability can hazardous and unsafe working conditions.[129] Co has limited
16
REVIEW
reserves and it is depletion is growing rapidly in fact that it is a demand for a sustainable electrolysis capacity of between 7 and
getting more attention in the field of the battery sector for energy 71 GW in 2030 and 275 GW in 2050 for only Germany's energy
storage. In near future, there will be noteworthy supply limitations sector.[131] Haber-Bosch ammonia synthesis itself would need a
for large industrial-scale application of cobalt oxide. renewable electricity supply of up to 27.5 TWh per annum to be
Within the current water electrolysis technology, the electrolyte or operated with 100% green H2.[132] Now, the question is that where
water should be ultrapure to avoid other side reactions (such as this energy would come from? This cannot be provided with the
chlorine oxidation) as well as contamination that deteriorate both current technology from sustainable sources since Germany is
electrodes and separator. This is going to be another challenge not at the state to be able to cover the production of sufficient
to tackle by taking into account that many countries in the world carbon-free electrical power. There is a need for investments for
have limited clean water resources even for drinking and daily- more photovoltaic as well as for on and offshore wind turbine
based consumption. Thus, there is also a need for the farms. Nuclear energy could be an additional option to produce
development of more effective cells, reactors, and membranes electricity for on-site water electrolysis and green H2 production,
Accepted Manuscript
that could be operated in non-distilled water. however, Germany has decided to shut down all the nuclear
power facilities by the end of 2022 after the Fukushima Daiichi
Another key important aspect, besides the catalyst’s development,
nuclear disaster.
for the future would be the availability of sustainable electricity for
large-scale water electrolysis plants (Scheme 4). Let have a look Finally, there is also a need for investment in the infrastructure for
at Germany´s electricity production as an example, the the H2 storage and distribution. Indeed, there is also a necessity
Fraunhofer Institute for Solar Energy Systems ISE reported that for substantial technologic development and investment for the
Germany could produce 55.8 percent of its electricity from integration of on-site green H2 production into some existing
renewable sources for the first half of 2020 by reaching a new industrial processes such as green steel production. In the end,
record high. In the first six months of 2020, solar and wind there is a long way to go not only for large-scale green H2
produced a total of 102.9 terawatt-hours (TWh), which is only 10.6 production but also for its storage, distribution, utilization, and
integration into the existing infrastructures.
TWh more than in the first half of 2019.[130] It is predicted to have
17
REVIEW
Table 1. Performance of state-of-the-art OER electrocatalysts, based on Co, Ni, Fe, and mixed oxides.
Loading Overpotential Tafel Slope

Catalyst Electrolyte Substrate Ref
(mg/cm2) (mV) (mv/dec)
Commercial samples:
[133]
Co3O4 (< 50 nm) 1M NaOH 0.8 GCE 500@10 mA/cm2 61
[133]
NiO (< 50 nm) 1M NaOH 0.8 GCE 430@10 mA/cm2 62
Fe2O3 (< 5 μm) 1M NaOH 0.8 GCE 1230@10 mA/cm2 - [133]
[133]
NiCoO2 (< 150 nm) 1M NaOH 0.8 GCE 390@10 mA/cm2 53
[133]
NiFe2O4 (30 nm) 1M NaOH 0.8 GCE 510@10 mA/cm2 -
[134]
CoFe2O4 1 M KOH 0.15 GCE 416@10 mA/cm2 ~ 95
Co3O4, NiO, Fe2O3, and mixed oxide electrocatalysts:
Accepted Manuscript
Co3O4 NPs (5.9 nm) 1 M KOH 1 NF 328@10 mA/cm2 - [135]
Co3O4 NPs (21.1 nm) 1 M KOH 1 NF 363@10 mA/cm2 - [135]
[83b]
Spent tea leaves templated Co3O4 NPs 1M KOH 0.025 GCE 401@10 mA/cm2 53
Co3O4 NCs@CoO (< 50 nm) 0.5 M KOH GCE 430@10 mA/cm2 89 [136]
[29a]
OM Co3O4 1M KOH 0.12 GCE 389@10 mA/cm2 56
[137]
NiO NPs (16 nm)/CNT 0.5 M KOH - GCE 409@10 mA/cm2 ~ 120
[138]
NiO NPs (16 nm) 0.5 M KOH 1.25 FTO 501@10 mA/cm2 -
NiO NWs 0.5 M KOH 1.25 FTO 363@10 mA/cm2 - [138]
NiO NS arrays 0.1 M KOH - CFP 422@10 mA/cm2 - [139]
+ [140]
N ion irradiated NiO NS arrays 0.1 M KOH - CFP 410@10 mA/cm2 136
OM NiO 1M KOH 0.12 GCE 414@10 mA/cm2 57 [29a]
Thin-film NiO 0.1 M KOH - Gold ~ 480@10 mA/cm2 ~ 55 [106a]
[106a]
Thin-film Ni0.8Fe0.2Ox 0.1 M KOH - Gold ~ 300@10 mA/cm2 ~ 40
[106a]
Thin-film Ni0.6Fe0.4Ox 0.1 M KOH - Gold ~ 280@10 mA/cm2 ~ 37
[106a]
Thin-film Fe2O3 0.1 M KOH - Gold ~ 530@10 mA/cm2 ~ 55
[118]
Fe2O3-(012) NCs (90 nm) 1M NaOH 0.15 NF 317@10 mA/cm2 59
[141]
Fe2O3 NPs (100 - 200 nm) 1M NaOH - FTO [email protected] mA/cm2 -
Co-Fe2O3 NPs (100 - 200 nm) 1M NaOH - FTO [email protected] mA/cm2 - [141]
Ni-Fe2O3 NPs (100 - 200 nm) 1M NaOH - FTO [email protected] mA/cm2 - [141]
OM NixCoyO4 1M KOH 0.12 GCE 382@10 mA/cm2 54 [29a]
Activated OM NixCoyO4 1M KOH 0.12 GCE 336@10 mA/cm2 36 [29a]
Ni-Fe NPs IM KOH 2.5 NF 210@10 mA/cm2 53 [119]
NF-AC-NiOx-Fe 1M KOH 0.014 NF 245@10 mA/cm2 34 [120]
NiFe-OH 1M KOH 1 NF 219@100 mA/cm2 56 [34]
Ni–Fe–OH/Ni3S2 1M KOH - NF 268@10 mA/cm2 54 [142]
[143]
Ni(Fe)OOH–FeSx 1M KOH - NF 220@10 mA/cm2 55
[143]
Ni(Fe)OOH 1M KOH - NF 300@10 mA/cm2 59
[144]
NiFe/Co9S8 1M KOH - CC 219@10 mA/cm2 55
[145]
CoNiFeO 1M KOH 1.9 NF 230@10 mA/cm2 40.7
LDH electrocatalysts:
NiCo-LDH 1M KOH 0.07 GCE 334@10 mA/cm2 41 [146]
[146]
NiFe-LDH 1M KOH 0.07 GCE 302@10 mA/cm2 40
[147]
NiFe-LDH 1M KOH - NF 210@10 mA/cm2 34
[147]
NiFeCP-LDH 1M KOH - NF 188@10 mA/cm2 29
[148]
CoNiFe LDH 1M KOH - GCE 278@10 mA/cm2 54.2
Note: “NP”, “NS”, “OM” and “NW” are the abbreviations of nanoparticle, nanosheet, ordered-mesoporous, and nanowire, respectively. “GCE”, “NF”, “CFP”, “CC”,
and “FTO” represent different electrodes, which are glassy carbon electrode, nickel foam, carbon fiber paper, carbon cloth, and fluorine-doped tin oxide on
glass, respectively.
18
REVIEW
Acknowledgments [22] D. A. Corrigan, J. Electrochem. Soc. 1987, 134, 377-384.
[23] C. C. L. McCrory, S. Jung, J. C. Peters, T. F. Jaramillo, J.
We thank Max Planck Society for the basic support and the Am. Chem. Soc. 2013, 135, 16977-16987.
Deutsche Forschungsgemeinschaft (DFG, German Research [24] a) D. Hein, G. Wartner, A. Bergmann, M. Bernal, B. R.
Foundation) Projektnummer 388390466-TRR 247 within the Cuenya, R. Seidel, ACS Nano 2020, 14, 15450-15457; b)
A. Moysiadou, S. Lee, C.-S. Hsu, H. M. Chen, X. Hu, J.
Collaborative Research Centre/Transregio 247 “Heterogeneous
Am. Chem. Soc. 2020, 142, 11901-11914; c) D. Drevon,
Oxidation Catalysis in the Liquid Phase” for the financial support. M. Görlin, P. Chernev, L. Xi, H. Dau, K. M. Lange, Sci.
Rep. 2019, 9, 1532; d) R. D. L. Smith, C. Pasquini, S.
Loos, P. Chernev, K. Klingan, P. Kubella, M. R.
Keywords: water electrolysis• oxygen evolution reaction (OER)
Mohammadi, D. Gonzalez-Flores, H. Dau, Nat. Commun.
electrocatalysis • cobalt • nickel • iron • 2017, 8, 2022.
[25] F. Song, L. Bai, A. Moysiadou, S. Lee, C. Hu, L. Liardet,
X. Hu, J. Am. Chem. Soc. 2018, 140, 7748-7759.
References
[26] K. Jiang, M. Luo, M. Peng, Y. Yu, Y.-R. Lu, T.-S. Chan, P.
Accepted Manuscript
Liu, F. M. F. de Groot, Y. Tan, Nat. Commun. 2020, 11,
[1] P. De Luna, C. Hahn, D. Higgins, S. A. Jaffer, T. F. 2701.
Jaramillo, E. H. Sargent, Science 2019, 364, eaav3506. [27] I. C. Man, H.-Y. Su, F. Calle-Vallejo, H. A. Hansen, J. I.
[2] A Hydrogen Strategy for A Climate Neutral Europe 2020, Martínez, N. G. Inoglu, J. Kitchin, T. F. Jaramillo, J. K.
https://ec.europa.eu/energy/sites/ener/files/hydrogen_strat Nørskov, J. Rossmeisl, ChemCatChem 2011, 3, 1159-
egy.pdf. 1165.
[3] S. van Renssen, Nat. Clim. Chang. 2020, 10, 799-801. [28] Z. W. Seh, J. Kibsgaard, C. F. Dickens, I. Chorkendorff, J.
[4] T. Grewe, X. Deng, C. Weidenthaler, F. Schüth, H. K. Nørskov, T. F. Jaramillo, Science 2017, 355, eaad4998.
Tüysüz, Chem. Mater. 2013, 25, 4926-4935. [29] a) X. Deng, S. Öztürk, C. Weidenthaler, H. Tüysüz, ACS
[5] S. Möller, S. Barwe, J. Masa, D. Wintrich, S. Seisel, H. Appl. Mater. Interfaces 2017, 9, 21225-21233; b) M. Yu,
Baltruschat, W. Schuhmann, Angew. Chem. Int. Ed. 2020, G.-h. Moon, E. Bill, H. Tüysüz, ACS Appl. Energy Mater.
59, 1585-1589. 2019, 2, 1199-1209; c) G.-h. Moon, M. Yu, C. K. Chan, H.
[6] B. M. Hunter, H. B. Gray, A. M. Muller, Chem. Rev. 2016, Tüysüz, Angew. Chem. Int. Ed. 2019, 58, 3491-3495; d) I.
116, 14120-14136. Spanos, A. A. Auer, S. Neugebauer, X. Deng, H. Tüysüz,
[7] S. P. Badwal, S. S. Giddey, C. Munnings, A. I. Bhatt, A. F. R. Schlögl, ACS Catal. 2017, 7, 3768-3778; e) I. Spanos,
Hollenkamp, Front. Chem. 2014, 2, 79. M. F. Tesch, M. Yu, H. Tüysüz, J. Zhang, X. Feng, K.
[8] C. C. L. McCrory, S. Jung, I. M. Ferrer, S. M. Chatman, J. Müllen, R. Schlögl, A. K. Mechler, ACS Catal. 2019, 9,
C. Peters, T. F. Jaramillo, J. Am. Chem. Soc. 2015, 137, 8165-8170.
4347-4357. [30] H. Tüysüz, Y. J. Hwang, S. B. Khan, A. M. Asiri, P. Yang,
[9] a) X. Li, L. Zhao, J. Yu, X. Liu, X. Zhang, H. Liu, W. Zhou, Nano Res. 2013, 6, 47-54.
Nano-Micro Lett. 2020, 12, 131; b) M. K. Debe, J. [31] M. David, C. Ocampo-Martínez, R. Sánchez-Peña, J.
Electrochem. Soc. 2013, 160, F522-F534. Energy Storage 2019, 23, 392-403.
[10] S. Cherevko, S. Geiger, O. Kasian, A. Mingers, K. J. J. [32] G. Zhang, H. Wang, J. Yang, Q. Zhao, L. Yang, H. Tang,
Mayrhofer, J. Electroanal. Chem. 2016, 773, 69-78. C. Liu, H. Chen, Y. Lin, F. Pan, Inorg. Chem. 2018, 57,
[11] M. Yu, C. K. Chan, H. Tüysüz, ChemSusChem 2018, 11, 2766-2772.
605-611. [33] a) M. Schalenbach, O. Kasian, K. J. J. Mayrhofer, Int. J.
[12] M. S. Burke, M. G. Kast, L. Trotochaud, A. M. Smith, S. W. Hydrogen Energy 2018, 43, 11932-11938; b) M.
Boettcher, J. Am. Chem. Soc. 2015, 137, 3638-3648. Schalenbach, G. Tjarks, M. Carmo, W. Lueke, M. Mueller,
[13] M. W. Kanan, D. G. Nocera, Science 2008, 321, 1072- D. Stolten, J. Electrochem. Soc. 2016, 163, F3197-F3208.
1075. [34] Y. Wei, C.-H. Shin, E. B. Tetteh, B.-J. Lee, J.-S. Yu, ACS
[14] J. Masa, I. Sinev, H. Mistry, E. Ventosa, M. de la Mata, J. Appl. Energy Mater. 2020, 3, 822-830.
Arbiol, M. Muhler, B. Roldan Cuenya, W. Schuhmann, [35] a) L. J. Enman, A. E. Vise, M. Burke Stevens, S. W.
Adv. Energy Mater. 2017, 7, 1700381. Boettcher, ChemPhysChem 2019, 20, 3089-3095; b) P.
[15] J.-H. Kim, K. Kawashima, B. R. Wygant, O. Mabayoje, Y. Chakthranont, J. Kibsgaard, A. Gallo, J. Park, M. Mitani,
Liu, J. H. Wang, C. B. Mullins, ACS Appl. Energy Mater. D. Sokaras, T. Kroll, R. Sinclair, M. B. Mogensen, T. F.
2018, 1, 5145-5150. Jaramillo, ACS Catal. 2017, 7, 5399-5409; c) L. C. Seitz,
[16] C. Liu, G. Bai, X. Tong, Y. Wang, B. Lv, N. Yang, X.-Y. T. J. P. Hersbach, D. Nordlund, T. F. Jaramillo, J. Phys.
Guo, Electrochem. Commun. 2019, 98, 87-91. Chem. Lett. 2015, 6, 4178-4183; d) D. Bernsmeier, M.
[17] J. Masa, S. Piontek, P. Wilde, H. Antoni, T. Eckhard, Y.-T. Bernicke, R. Schmack, R. Sachse, B. Paul, A. Bergmann,
Chen, M. Muhler, U.-P. Apfel, W. Schuhmann, Adv. P. Strasser, E. Ortel, R. Kraehnert, ChemSusChem 2018,
Energy Mater. 2019, 9, 1900796. 11, 2367-2374; e) J. Fester, A. Makoveev, D. Grumelli, R.
[18] P. Cai, J. Huang, J. Chen, Z. Wen, Angew. Chem. Int. Ed. Gutzler, Z. Sun, J. Rodríguez-Fernández, K. Kern, J. V.
2017, 56, 4858-4861. Lauritsen, Angew. Chem. Int. Ed. 2018, 57, 11893-11897.
[19] Z. Fang, L. Peng, H. Lv, Y. Zhu, C. Yan, S. Wang, P. [36] A. N. Colli, H. H. Girault, A. Battistel, Materials 2019, 12,
Kalyani, X. Wu, G. Yu, ACS Nano 2017, 11, 9550-9557. 1336.
[20] X. Wang, X. Huang, W. Gao, Y. Tang, P. Jiang, K. Lan, R. [37] M. I. Jaramillo-Gutiérrez, S. M. Sierra-González, C. A.
Yang, B. Wang, R. Li, J. Mater. Chem. A 2018, 6, 3684- Ramírez-González, J. E. Pedraza-Rosas, J. A. Pedraza-
3691. Avella, Int. J. Hydrogen Energy 2021, 46, 7667-7675.
[21] a) D. Y. Chung, P. P. Lopes, P. Farinazzo Bergamo Dias [38] R. Frydendal, E. A. Paoli, B. P. Knudsen, B. Wickman, P.
Martins, H. He, T. Kawaguchi, P. Zapol, H. You, D. Malacrida, I. E. L. Stephens, I. Chorkendorff,
Tripkovic, D. Strmcnik, Y. Zhu, S. Seifert, S. Lee, V. R. ChemElectroChem 2014, 1, 2075-2081.
Stamenkovic, N. M. Markovic, Nat. Energy 2020, 5, 222- [39] I. S. Muratova, K. N. Mikhelson, Y. Ermolenko, A.
230; b) F. Dionigi, Z. Zeng, I. Sinev, T. Merzdorf, S. Offenhäusser, Y. Mourzina, J. Solid State Electrochem.
Deshpande, M. B. Lopez, S. Kunze, I. Zegkinoglou, H. 2016, 20, 3359-3365.
Sarodnik, D. Fan, A. Bergmann, J. Drnec, J. F. d. Araujo, [40] S. Anantharaj, S. R. Ede, K. Karthick, S. S. Sankar, K.
M. Gliech, D. Teschner, J. Zhu, W.-X. Li, J. Greeley, B. R. Sangeetha, P. E. Karthik, S. Kundu, Energy Environ. Sci.
Cuenya, P. Strasser, Nat. Commun. 2020, 11, 2522; c) A. 2018, 11, 744-771.
Bergmann, T. E. Jones, E. Martinez Moreno, D. Teschner, [41] S. Anantharaj, S. Kundu, ACS Energy Lett. 2019, 4, 1260-
P. Chernev, M. Gliech, T. Reier, H. Dau, P. Strasser, Nat. 1264.
Catal. 2018, 1, 711-719.
19
REVIEW
[42] a) Y. Liu, X. Liang, L. Gu, Y. Zhang, G.-D. Li, X. Zou, J.-S. [69] J. Luo, J.-H. Im, M. T. Mayer, M. Schreier, M. K.
Chen, Nat. Commun. 2018, 9, 2609; b) Q. Qian, Y. Li, Y. Nazeeruddin, N.-G. Park, S. D. Tilley, H. J. Fan, M.
Liu, L. Yu, G. Zhang, Adv. Mater. 2019, 31, e1901139. Grätzel, Science 2014, 345, 1593.
[43] S. Anantharaj, S. R. Ede, K. Sakthikumar, K. Karthick, S. [70] a) F. Dionigi, P. Strasser, Adv. Energy Mater. 2016, 6,
Mishra, S. Kundu, ACS Catal. 2016, 6, 8069-8097. 1600621; b) W. Zhang, K. Zhou, Small 2017, 13, 1700806.
[44] T. Shinagawa, A. T. Garcia-Esparza, K. Takanabe, Sci. [71] J. Masa, W. Schuhmann, ChemCatChem 2019, 11, 5842-
Rep. 2015, 5, 13801. 5854.
[45] A. Kapałka, G. Fóti, C. Comninellis, Electrochem. [72] a) Q. Fu, T. Wu, G. Fu, T. Gao, J. Han, T. Yao, Y. Zhang,
Commun. 2008, 10, 607-610. W. Zhong, X. Wang, B. Song, ACS Energy Lett. 2018, 3,
[46] I. Duo, A. Fujishima, C. Comninellis, Electrochem. 1744-1752; b) J. Masa, P. Weide, D. Peeters, I. Sinev, W.
Commun. 2003, 5, 695-700. Xia, Z. Sun, C. Somsen, M. Muhler, W. Schuhmann, Adv.
[47] M. Dulce Maria, R. Marcel, Journal of Physics: Energy Energy Mater. 2016, 6, 1502313.
2021. [73] D. Xu, M. B. Stevens, M. R. Cosby, S. Z. Oener, A. M.
[48] A. R. C. Bredar, A. L. Chown, A. R. Burton, B. H. Farnum, Smith, L. J. Enman, K. E. Ayers, C. B. Capuano, J. N.
ACS Appl. Energy Mater. 2020, 3, 66-98. Renner, N. Danilovic, Y. Li, H. Wang, Q. Zhang, S. W.
Accepted Manuscript
[49] M. E. Orazem, B. Tribollet, Electrochemical Impedance Boettcher, ACS Catal. 2019, 9, 7-15.
Spectroscopy, John Wiley & Sons, Inc., Hoboken, New [74] M. Gong, Y. Li, H. Wang, Y. Liang, J. Z. Wu, J. Zhou, J.
Jersey, 2008. Wang, T. Regier, F. Wei, H. Dai, J. Am. Chem. Soc. 2013,
[50] Y. Lin, Z. Tian, L. Zhang, J. Ma, Z. Jiang, B. J. Deibert, R. 135, 8452-8455.
Ge, L. Chen, Nat. Commun. 2019, 10, 162. [75] B. Zhang, X. Zheng, O. Voznyy, R. Comin, M. Bajdich, M.
[51] S. Trasatti, O. A. Petrii, Pure Appl. Chem. 1991, 63, 711- Garcia-Melchor, L. Han, J. Xu, M. Liu, L. Zheng, F. P.
734. Garcia de Arquer, C. T. Dinh, F. Fan, M. Yuan, E.
[52] S. Sun, H. Li, Z. J. Xu, Joule 2018, 2, 1024-1027. Yassitepe, N. Chen, T. Regier, P. Liu, Y. Li, P. De Luna,
[53] S. Maass, F. Finsterwalder, G. Frank, R. Hartmann, C. A. Janmohamed, H. L. Xin, H. Yang, A. Vojvodic, E. H.
Merten, J. Power Sources 2008, 176, 444-451. Sargent, Science 2016, 352, 333-337.
[54] J. G. Vos, M. T. M. Koper, J. Electroanal. Chem. 2018, [76] R. D. L. Smith, M. S. Prévot, R. D. Fagan, Z. Zhang, P. A.
819, 260-268. Sedach, M. K. J. Siu, S. Trudel, C. P. Berlinguette,
[55] M. Görlin, P. Chernev, J. Ferreira de Araújo, T. Reier, S. Science 2013, 340, 60.
Dresp, B. Paul, R. Krähnert, H. Dau, P. Strasser, J. Am. [77] X. Zou, Y. Liu, G.-D. Li, Y. Wu, D.-P. Liu, W. Li, H.-W. Li,
Chem. Soc. 2016, 138, 5603-5614. D. Wang, Y. Zhang, X. Zou, Adv. Mater. 2017, 29,
[56] M. Yu, K. S. Belthle, C. Tüysüz, H. Tüysüz, J. Mater. 1700404.
Chem. A 2019, 7, 23130-23139. [78] P. Ferchland, J. Nussbaum, Google Patents, 1908.
[57] H. Zhou, F. Yu, Q. Zhu, J. Sun, F. Qin, L. Yu, J. Bao, Y. [79] a) M. S. Burke, L. J. Enman, A. S. Batchellor, S. Zou, S.
Yu, S. Chen, Z. Ren, Energy Environ. Sci. 2018, 11, 2858- W. Boettcher, Chem. Mater. 2015, 27, 7549-7558; b) N. T.
2864. Suen, S. F. Hung, Q. Quan, N. Zhang, Y. J. Xu, H. M.
[58] C. Spöri, J. T. H. Kwan, A. Bonakdarpour, D. P. Wilkinson, Chen, Chem. Soc. Rev. 2017, 46, 337-365.
P. Strasser, Angew. Chem. Int. Ed. 2017, 56, 5994-6021. [80] D. S. Baek, G. Y. Jung, B. Seo, J. C. Kim, H. W. Lee, T. J.
[59] a) V. K. Puthiyapura, S. Pasupathi, S. Basu, X. Wu, H. Su, Shin, H. Y. Jeong, S. K. Kwak, S. H. Joo, Adv. Funct.
N. Varagunapandiyan, B. Pollet, K. Scott, Int. J. Hydrogen Mater. 2019, 29, 1901217.
Energy 2013, 38, 8605-8616; b) H. N. Su, V. Linkov, B. J. [81] a) J. Zhu, H. Li, L. Zhong, P. Xiao, X. Xu, X. Yang, Z.
Bladergroen, Int. J. Hydrogen Energy 2013, 38, 9601- Zhao, J. Li, ACS Catal. 2014, 4, 2917-2940; b) Ş. Ortatatlı,
9608; c) C.-C. Sung, C.-Y. Liu, Int. J. Hydrogen Energy J. Ternieden, C. Weidenthaler, Eur. J. Inorg. Chem. 2018,
2013, 38, 10063-10067. 2018, 5238-5245.
[60] H. N. Nong, L. J. Falling, A. Bergmann, M. Klingenhof, H. [82] M. T. Colomer, D. A. Fumo, J. R. Jurado, A. M. Segadaẽs,
P. Tran, C. Spöri, R. Mom, J. Timoshenko, G. Zichittella, J. Mater. Chem. 1999, 9, 2505-2510.
A. Knop-Gericke, S. Piccinin, J. Pérez-Ramírez, B. R. [83] a) E. Budiyanto, M. Yu, M. Chen, S. DeBeer, O. Rüdiger,
Cuenya, R. Schlögl, P. Strasser, D. Teschner, T. E. Jones, H. Tüysüz, ACS Appl. Energy Mater. 2020, 3, 8583-8594;
Nature 2020, 587, 408-413. b) X. Deng, C. K. Chan, H. Tüysüz, ACS Appl. Mater.
[61] J. Suntivich, H. A. Gasteiger, N. Yabuuchi, H. Nakanishi, Interfaces 2016, 8, 32488-32495; c) F. Song, X. Hu, J.
J. B. Goodenough, Y. Shao-Horn, Nat. Chem. 2011, 3, Am. Chem. Soc. 2014, 136, 16481-16484; d) M. Q. Yu, Y.
546-550. H. Li, S. Yang, P. F. Liu, L. F. Pan, L. Zhang, H. G. Yang,
[62] M. Bajdich, M. García-Mota, A. Vojvodic, J. K. Nørskov, A. J. Mater. Chem. A 2015, 3, 14101-14104.
T. Bell, J. Am. Chem. Soc. 2013, 135, 13521-13530. [84] H. G. Byers, J. Am. Chem. Soc. 1908, 30, 1718-1742.
[63] X. Du, J. Huang, J. Zhang, Y. Yan, C. Wu, Y. Hu, C. Yan, [85] L. D. Burke, M. E. Lyons, O. J. Murphy, J. Electroanal.
T. Lei, W. Chen, C. Fan, J. Xiong, Angew. Chem. Int. Ed. Chem. 1982, 132, 247-261.
2019, 58, 4484-4502. [86] Y. Liu, Y. Ying, L. Fei, Y. Liu, Q. Hu, G. Zhang, S. Y.
[64] J. Suntivich, K. J. May, H. A. Gasteiger, J. B. Pang, W. Lu, C. L. Mak, X. Luo, L. Zhou, M. Wei, H.
Goodenough, Y. Shao-Horn, Science 2011, 334, 1383- Huang, J. Am. Chem. Soc. 2019, 141, 8136-8145.
1385. [87] X. Deng, K. Chen, H. Tüysüz, Chem. Mater. 2017, 29, 40-
[65] R. Zhang, Y.-C. Zhang, L. Pan, G.-Q. Shen, N. Mahmood, 52.
Y.-H. Ma, Y. Shi, W. Jia, L. Wang, X. Zhang, W. Xu, J.-J. [88] a) X. Deng, W. N. Schmidt, H. Tüysüz, Chem. Mater.
Zou, ACS Catal. 2018, 8, 3803-3811. 2014, 26, 6127-6134; b) T. Grewe, X. Deng, H. Tüysüz,
[66] G. Gottstein, in Physical Foundations of Materials Science Chem. Mater. 2014, 26, 3162-3168; c) X. Deng, H.-J.
(Ed.: G. Gottstein), Springer Berlin Heidelberg, Berlin, Bongard, C. K. Chan, H. Tüysüz, ChemSusChem 2016, 9,
Heidelberg, 2004, pp. 63-106. 409-415; d) J. Behnken, M. Yu, X. Deng, H. Tüysüz, C.
[67] a) M. Yu, F. Waag, C. K. Chan, C. Weidenthaler, S. Harms, A. Dyck, G. Wittstock, ChemElectroChem 2019, 6,
Barcikowski, H. Tüysüz, ChemSusChem 2020, 13, 520- 3460-3467.
528; b) Z. Xiao, Y.-C. Huang, C.-L. Dong, C. Xie, Z. Liu, S. [89] R. Subbaraman, D. Tripkovic, K.-C. Chang, D. Strmcnik,
Du, W. Chen, D. Yan, L. Tao, Z. Shu, G. Zhang, H. Duan, A. P. Paulikas, P. Hirunsit, M. Chan, J. Greeley, V.
Y. Wang, Y. Zou, R. Chen, S. Wang, J. Am. Chem. Soc. Stamenkovic, N. M. Markovic, Nat. Mater. 2012, 11, 550-
2020, 142, 12087-12095. 557.
[68] Q. Yue, C. Liu, Y. Wan, X. Wu, X. Zhang, P. Du, J. Catal. [90] A. Bergmann, E. Martinez-Moreno, D. Teschner, P.
2018, 358, 1-7. Chernev, M. Gliech, J. F. de Araujo, T. Reier, H. Dau, P.
Strasser, Nat. Commun. 2015, 6, 8625.
20
REVIEW
[91] R. Borup, J. Meyers, B. Pivovar, Y. S. Kim, R. Mukundan, [116] B. Tian, H. Shin, S. Liu, M. Fei, Z. Mu, C. Liu, Y. Pan, Y.
N. Garland, D. Myers, M. Wilson, F. Garzon, D. Wood, P. Sun, W. A. Goddard Iii, M. Ding, Angew. Chem. Int. Ed.
Zelenay, K. More, K. Stroh, T. Zawodzinski, J. Boncella, J. 2021, in press, https://doi.org/10.1002/anie.202101906.
E. McGrath, M. Inaba, K. Miyatake, M. Hori, K. Ota, Z. [117] J. Ferreira de Araújo, F. Dionigi, T. Merzdorf, H.-S. Oh, P.
Ogumi, S. Miyata, A. Nishikata, Z. Siroma, Y. Uchimoto, K. Strasser, Angew. Chem. Int. Ed. 2021, in press,
Yasuda, K. Kimijima, N. Iwashita, Chem. Rev. 2007, 107, https://doi.org/10.1002/anie.202101698.
3904-3951. [118] H. Wu, T. Yang, Y. Du, L. Shen, G. W. Ho, Adv. Mater.
[92] Y. Kutin, N. Cox, W. Lubitz, A. Schnegg, O. Rüdiger, 2018, 30, e1804341.
Catalysts 2019, 9, 926. [119] B. H. R. Suryanto, Y. Wang, R. K. Hocking, W. Adamson,
[93] a) X. Deng, H. Tüysüz, ACS Catal. 2014, 4, 3701-3714; b) C. Zhao, Nat. Commun. 2019, 10, 5599.
J. Rosen, G. S. Hutchings, F. Jiao, J. Am. Chem. Soc. [120] F. Song, M. M. Busch, B. Lassalle-Kaiser, C.-S. Hsu, E.
2013, 135, 4516-4521; c) D. Gao, R. Liu, J. Biskupek, U. Petkucheva, M. Bensimon, H. M. Chen, C. Corminboeuf,
Kaiser, Y.-F. Song, C. Streb, Angew. Chem. Int. Ed. 2019, X. Hu, ACS Cent. Sci. 2019, 5, 558-568.
58, 4644-4648. [121] a) K. He, T. Tadesse Tsega, X. Liu, J. Zai, X.-H. Li, X. Liu,
[94] X.-T. Wang, T. Ouyang, L. Wang, J.-H. Zhong, T. Ma, Z.- W. Li, N. Ali, X. Qian, Angew. Chem. Int. Ed. 2019, 58,
Accepted Manuscript
Q. Liu, Angew. Chem. Int. Ed. 2019, 58, 13291-13296. 11903-11909; b) D. Friebel, M. W. Louie, M. Bajdich, K. E.
[95] S.-H. Hsu, S.-F. Hung, H.-Y. Wang, F.-X. Xiao, L. Zhang, Sanwald, Y. Cai, A. M. Wise, M.-J. Cheng, D. Sokaras, T.-
H. Yang, H. M. Chen, J.-M. Lee, B. Liu, Small Methods C. Weng, R. Alonso-Mori, R. C. Davis, J. R. Bargar, J. K.
2018, 2, 1800001. Nørskov, A. Nilsson, A. T. Bell, J. Am. Chem. Soc. 2015,
[96] Z. Liu, G. Wang, X. Zhu, Y. Wang, Y. Zou, S. Zang, S. 137, 1305-1313.
Wang, Angew. Chem. Int. Ed. 2020, 59, 4736-4742. [122] B. M. Hunter, J. R. Winkler, H. B. Gray, Molecules 2018,
[97] a) M. Yu, G. Li, C. Fu, E. Liu, K. Manna, E. Budiyanto, Q. 23, 903.
Yang, C. Felser, H. Tüysüz, Angew. Chem. Int. Ed. 2021, [123] J. Y. Chen, L. Dang, H. Liang, W. Bi, J. B. Gerken, S. Jin,
60, 5800-5805; b) Y. Sun, X. Ren, S. Sun, Z. Liu, S. Xi, Z. E. E. Alp, S. S. Stahl, J. Am. Chem. Soc. 2015, 137,
J. Xu, Angew. Chem. Int. Ed. 2021, 60, 14536-14544 15090-15093.
[98] L. Gong, X. Y. E. Chng, Y. Du, S. Xi, B. S. Yeo, ACS [124] L. Bai, S. Lee, X. Hu, Angew. Chem. Int. Ed. 2021, 60,
Catal. 2018, 8, 807-814. 3095-3103.
[99] A. Bähr, G.-h. Moon, H. Tüysüz, ACS Appl. Energy Mater. [125] S. Lee, L. Bai, X. Hu, Angew. Chem. Int. Ed. 2020, 59,
2019, 2, 6672-6680. 8072-8077.
[100] B. S. Yeo, A. T. Bell, J. Am. Chem. Soc. 2011, 133, 5587- [126] S. Lee, K. Banjac, M. Lingenfelder, X. Hu, Angew. Chem.
5593. Int. Ed. 2019, 58, 10295-10299.
[101] M. Yu, G.-h. Moon, R. G. Castillo, S. DeBeer, C. [127] Distribution of Electricity Generation Worldwide in 2018 by
Weidenthaler, H. Tüysüz, Angew. Chem. Int. Ed. 2020, 59, Energy Source 2018,
16544-16552. https://www.statista.com/statistics/269811/world-
[102] Z. Liu, Z. Xiao, G. Luo, R. Chen, C.-L. Dong, X. Chen, J. electricity-production-by-energy-source/.
Cen, H. Yang, Y. Wang, D. Su, Y. Li, S. Wang, Small [128] a) R. Palkovits, S. Palkovits, ACS Catal. 2019, 9, 8383-
2019, 15, 1904903. 8387; b) M. J. Craig, G. Coulter, E. Dolan, J. Soriano-
[103] K. Fominykh, J. M. Feckl, J. Sicklinger, M. Döblinger, S. López, E. Mates-Torres, W. Schmitt, M. García-Melchor,
Böcklein, J. Ziegler, L. Peter, J. Rathousky, E.-W. Scheidt, Nat. Commun. 2019, 10, 4993.
T. Bein, D. Fattakhova-Rohlfing, Adv. Funct. Mater. 2014, [129] S. van den Brink, R. Kleijn, B. Sprecher, A. Tukker,
24, 3123-3129. Resour., Conserv. Recycl. 2020, 156, 104743.
[104] S. Klaus, Y. Cai, M. W. Louie, L. Trotochaud, A. T. Bell, J. [130] German Net Electricity Generation in First Half of 2020:
Phys. Chem. C 2015, 119, 7243-7254. Renewables Reach Record Share of 55.8 Percent 2020,
[105] a) J. W. D. Ng, M. Garcia-Melchor, M. Bajdich, P. https://www.ise.fraunhofer.de/en/press-media/press-
Chakthranont, C. Kirk, A. Vojvodic, T. F. Jaramillo, Nat. releases/2020/net-energy-production-first-half-of-
Energy 2016, 1, 16053; b) M. R. Gennero de Chialvo, A. 2020.html.
C. Chialvo, Electrochim. Acta 1988, 33, 825-830. [131] M. R. C. Hebling, T. Fleiter, U. Groos, D. Härle, A. Held,
[106] a) M. W. Louie, A. T. Bell, J. Am. Chem. Soc. 2013, 135, M. Jahn, N. Müller, T. Pfeifer, P. Plötz, O. Ranzmeyer, A.
12329-12337; b) M. B. Stevens, C. D. M. Trang, L. J. Schaadt, F. Sensfuß, T. Smolinka, M. Wietschel, Eine
Enman, J. Deng, S. W. Boettcher, J. Am. Chem. Soc. Wasserstoff-Roadmap für Deutschland 2019,
2017, 139, 11361-11364; c) A. Kleiman-Shwarsctein, Y.-S. https://www.ise.fraunhofer.de/content/dam/ise/de/docume
Hu, G. D. Stucky, E. W. McFarland, Electrochem. nts/publications/studies/2019-10_Fraunhofer_Wasserstoff-
Commun. 2009, 11, 1150-1153. Roadmap_fuer_Deutschland.pdf.
[107] S. L. Medway, C. A. Lucas, A. Kowal, R. J. Nichols, D. [132] R. Geres, A. Kohn, S. Lenz, F. Ausfelder, A. M.
Johnson, J. Electroanal. Chem. 2006, 587, 172-181. Bazzanella, A. Möller, Auf dem Weg zu Einer
[108] B. S. Yeo, A. T. Bell, J. Phys. Chem. C 2012, 116, 8394- Treibhausgasneutralen Chemischen Industrie in
8400. Deutschland : Eine Studie von DECHEMA und
[109] a) A. C. Garcia, T. Touzalin, C. Nieuwland, N. Perini, M. T. FutureCamp für den VCI, DECHEMA Gesellschaft für
M. Koper, Angew. Chem. Int. Ed. 2019, 58, 12999-13003; Chemische Technik und Biotechnologie e.V., Frankfurt am
b) L. Trotochaud, S. L. Young, J. K. Ranney, S. W. Main, 2019.
Boettcher, J. Am. Chem. Soc. 2014, 136, 6744-6453. [133] S. Jung, C. C. L. McCrory, I. M. Ferrer, J. C. Peters, T. F.
[110] H. Bode, K. Dehmelt, J. Witte, Electrochim. Acta 1966, 11, Jaramillo, J. Mater. Chem. A 2016, 4, 3068-3076.
1079-1087. [134] F. Waag, B. Gökce, C. Kalapu, G. Bendt, S. Salamon, J.
[111] K. Juodkazis, J. Juodkazytė, R. Vilkauskaitė, V. Landers, U. Hagemann, M. Heidelmann, S. Schulz, H.
Jasulaitienė, J. Solid State Electrochem. 2008, 12, 1469- Wende, N. Hartmann, M. Behrens, S. Barcikowski, Sci.
1479. Rep. 2017, 7, 13161.
[112] K. Sliozberg, Y. Aniskevich, U. Kayran, J. Masa, W. [135] A. J. Esswein, M. J. McMurdo, P. N. Ross, A. T. Bell, T. D.
Schuhmann, Z. Phys. Chem. 2020, 234, 995-1019. Tilley, J. Phys. Chem. C 2009, 113, 15068-15072.
[113] C. Li, B. Zhang, Y. Li, S. Hao, X. Cao, G. Yang, J. Wu, Y. [136] C.-W. Tung, Y.-Y. Hsu, Y.-P. Shen, Y. Zheng, T.-S. Chan,
Huang, Appl. Catal., B 2019, 244, 56-62. H.-S. Sheu, Y.-C. Cheng, H. M. Chen, Nat. Commun.
[114] J. Yan, L. Kong, Y. Ji, J. White, Y. Li, J. Zhang, P. An, S. 2015, 6, 8106.
Liu, S.-T. Lee, T. Ma, Nat. Commun. 2019, 10, 2149. [137] F. Chekin, H. Tahermansouri, M. R. Besharat, J. Solid
[115] C. Liao, B. Yang, N. Zhang, M. Liu, G. Chen, X. Jiang, G. State Electrochem. 2014, 18, 747-753.
Chen, J. Yang, X. Liu, T.-S. Chan, Y.-J. Lu, R. Ma, W. [138] P. Manivasakan, P. Ramasamy, J. Kim, RSC Adv. 2015,
Zhou, Adv. Funct. Mater. 2019, 29, 1904020. 5, 33269-33274.
21
REVIEW
[139] N. Cheng, Q. Liu, J. Tian, X. Sun, Y. He, S. Zhai, A. M.
Asiri, Int. J. Hydrogen Energy 2015, 40, 9866-9871.
[140] B. Xia, T. Wang, X. Jiang, J. Li, T. Zhang, P. Xi, D. Gao,
D. Xue, J. Mater. Chem. A 2019, 7, 4729-4733.
[141] Y. Yin, X. Zhang, C. Sun, Prog. Nat. Sci.: Mater. Int. 2018,
28, 430-436.
[142] W. He, G. Ren, Y. Li, D. Jia, S. Li, J. Cheng, C. Liu, Q.
Hao, J. Zhang, H. Liu, Catal. Sci. Technol. 2020, 10, 1708-
1713.
[143] H. Yang, L. Gong, H. Wang, C. Dong, J. Wang, K. Qi, H.
Liu, X. Guo, B. Y. Xia, Nat. Commun. 2020, 11, 5075.
[144] C. Zhan, Z. Liu, Y. Zhou, M. Guo, X. Zhang, J. Tu, L. Ding,
Y. Cao, Nanoscale 2019, 11, 3378-3385.
[145] L. Han, L. Guo, C. Dong, C. Zhang, H. Gao, J. Niu, Z.
Peng, Z. Zhang, Nano Res. 2019, 12, 2281-2287.
Accepted Manuscript
[146] F. Song, X. Hu, Nat. Commun. 2014, 5, 4477.
[147] W. Li, F. Li, H. Yang, X. Wu, P. Zhang, Y. Shan, L. Sun,
Nat. Commun. 2019, 10, 5074.
[148] S. Si, H.-S. Hu, R.-J. Liu, Z.-X. Xu, C.-B. Wang, Y.-Y.
Feng, Int. J. Hydrogen Energy 2020, 45, 9368-9379.
22
REVIEW
Entry for the Table of Contents
Accepted Manuscript
Mingquan Yu received his M.Sc. degree in
materials science from East China University of
Science and Technology in 2016 and his Ph.D.
degree in 2020 in the department of
Heterogeneous Catalysis of Max-Planck-
Institute für Kohlenforschung. Afterward, he
continued as a post-doctoral researcher in the
same department. His research interests
mainly focus on the design of transition metal-
based electrocatalysts for water splitting.
Eko Budiyanto obtained his M.Sc. degree in

Chemistry from Jagiellonian University in
Kraków in 2018. He is currently working as a
Ph.D. student in the department of
Heterogeneous Catalysis of Max-Planck-
Institute für Kohlenforschung. His research
interest is the design of mesostructured cobalt
iron and nickel-based electrocatalysts for
water electrolysis.
Harun Tüysüz got his PhD degree in chemistry

in the department of Heterogeneous Catalysis
of Max-Planck-Institute für Kohlenforschung
(MPI-KOFO) in 2008. Afterward, he was
awarded a research fellowship from the DFG
to perform his postdoctoral studies at the
University of California, Berkeley. In 2012, he
was appointed as head of an independent
research group of Heterogeneous Catalysis
and Sustainable Energy at the MPI-KOFO. He
completed his habilitation in inorganic chemistry in 2016 with a focus on the
advance of tailor-made nanostructured catalysts. The present interest of his
research group is the design of functional nanoscale materials for
environmental and sustainable energy-related catalytic applications including
water splitting, selective oxidation reactions, and catalytic conversion of CO
and CO2. His research achievements in the field of nanostructured solid-state
materials and heterogeneous catalysis were awarded with several awards
including the prestigious Jochen-Block-Prize 2016, DECHEMA Prize 2019,
and Forcheurs Jean-Marie Lehn Prize 2020.
23

Angewandte: Chemie

Uploaded by

Copyright:

Available Formats

Angewandte: Chemie

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Angewandte: Chemie

Uploaded by

Copyright:

Available Formats

Angewandte

A Journal of the Gesellschaft Deutscher Chemiker

International Edition Chemie www.angewandte.org

Title: Principle of Water Electrolysis and Recent Progress of Cobalt,

Authors: Mingquan Yu, Eko Budiyanto, and Harun Tüysüz

To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202103824

Link to VoR: https://doi.org/10.1002/anie.202103824

Principle of Water Electrolysis and Recent Progress of Cobalt,

Dr. M. Yu, E. Budiyanto, Priv.-Doz. Dr. H. Tüysüz

A thermodynamic potential of 1.23 V is required to split water into

2.1. Reaction kinetics

The OER is a multiple-step reaction involving several adsorption

E(t) = |E0| sin ωt (8)

3. Criteria for electrocatalyst development

The electrocatalyst should fulfill diverse criteria to be considered

5. Summary and Perspective

Loading Overpotential Tafel Slope

Co3O4 NPs (21.1 nm) 1 M KOH 1 NF 363@10 mA/cm2 - [135]

NiO NS arrays 0.1 M KOH - CFP 422@10 mA/cm2 - [139]

Thin-film NiO 0.1 M KOH - Gold ~ 480@10 mA/cm2 ~ 55 [106a]

OM NixCoyO4 1M KOH 0.12 GCE 382@10 mA/cm2 54 [29a]

Activated OM NixCoyO4 1M KOH 0.12 GCE 336@10 mA/cm2 36 [29a]

Ni-Fe NPs IM KOH 2.5 NF 210@10 mA/cm2 53 [119]

NF-AC-NiOx-Fe 1M KOH 0.014 NF 245@10 mA/cm2 34 [120]

NiFe-OH 1M KOH 1 NF 219@100 mA/cm2 56 [34]

Ni–Fe–OH/Ni3S2 1M KOH - NF 268@10 mA/cm2 54 [142]

Eko Budiyanto obtained his M.Sc. degree in

Harun Tüysüz got his PhD degree in chemistry

You might also like