Oxygen Reduction Electrocatalysts toward Practical Fuel Cells: Progress and Perspectives

Abstract

Fuel cells are an incredibly powerful renewable energy technology, but their broad applications remain lagging because of the high cost and poor reliability of cathodic electrocatalysts for the oxygen reduction reaction (ORR). This review focuses on the recent progress of ORR electrocatalysts in fuel cells. More importantly, it highlights the fundamental problems associated with the insufficient activity translation from rotating disk electrode to membrane electrode assembly in the fuel cells. Finally, for the atomic level in-depth information on ORR catalysts in fuel cells, potential perspectives are suggested, including large-scale preparation, unified assessment criteria, advanced interpretation techniques, advanced simulation and artificial intelligence. This review aims to provide valuable insights into the fundamental science and technical engineering for efficient ORR electrocatalysts in fuel cells.

Full text

Fuel Cells
Oxygen Reduction Electrocatalysts toward Practical
Fuel Cells: Progress and Perspectives
Shahid Zaman, Lei Huang, Abdoulkader Ibro Douka, Huan Yang, Bo You, and
Bao Yu Xia*
A
ngewandte
Chemi
e
Keywords:
electrocatalysts · fuel cells ·
oxygen reduction · Pt alloy
A
ngewandte
Chemi
e
Reviews
How to cite:
International Edition: doi.org/10.1002/anie.202016977
German Edition: doi.org/10.1002/ange.202016977
&&&&
 2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021,60,2–23
Ü
Ü
These are not the final page numbers!

1. Introduction
Renewable energy technologies, i.e., fuel cells and
batteries, are gaining extensive attentions from the academia
and industrial sectors due to their independence on fossil
fuels. Among various sustainable energy sources, fuel cells,
especially polymer electrolyte membrane fuel cells, are highly
trending against the declining fossil fuels. Fuel cell is a device
that converts the chemically stored energy in fuel and oxygen
into electricity.[1,2] Contrary to the lithium-ion batteries, fuel
cells have various advantages of high efficiency, high power/
energy density, low-temperature operation, zero emissions,
fast refueling time and mileage like a gasoline car.[3,4]
However, at the current stage, fuel cells could not find
a smooth way to broader commercial market due to poor
performance, limited durability and high cost.
The oxygen reduction reaction (ORR) and fuel oxidation
reaction are the immediate reactions taking place at cathode
and anode, respectively. ORR at cathode is kinetically
sluggish due to the problematic activation/cleavage of O=O
bond, especially at low-pH environment, which relies on
enormous platinum (Pt) catalyst (approx. 8 times more
quantity than anode catalyst).[5–8] Therefore, the total
amount of catalyst consumed at cathode is almost 80 % of
the whole catalyst, where it accounts for 55% of the total cost
of the commercial-scale production of fuel cells.[9,10] Mean-
while, the catalyst degradation and electrode instability are
the substantial deterrents towards widespread applications,
particularly the Pt utilization of 0.25 gPt kW1is not econom-
ical for transportations. In view of fuel cell powered electric
vehicles (EVs) such as Toyota MIRAI, the most fascinating
features are considerably short refueling time of only three
minutes, cold start capability and an attractive cruising range
of 500 km. However, the costly Pt-based cathode catalyst
affects the total cost of the overall fuel cell stacking with Pt
30 g per stack, which is considerably higher than the long-
lasting targets of 5 g per vehicle.[11] Similarly, from the
customers perspective, these EVs are still considered
luxury vehicles because only a fraction of consumers can
afford these prices, hence the wide-
spread use of the EVs is highly ques-
tionable at this price range.
Considering these issues,[12] the
reduction of Pt contents without compromising the perfor-
mance is the top priority of current research.[13] The cathode
loading must be preferentially reduced for the sustainable
cost of fuel cells, whereas decreasing the Pt loading at the
cathode needs a much thicker catalyst layer for the desired
power density, which will result in the collapse of pore
structure and poor mass transport properties with the
consequent low performance and instability of electrode.
Hence, understanding the design mechanism and integrating
efficient catalysts are vital for the broader flourishment of
fuel cell technology.[14–16] The fundamental issues which
hinder the connection between the material innovation and
device integration must be highlighted to propose plausible
solutions for the development of fuel cells.[14,17, 18] In this
regard, the last decade has perceived extraordinary advance-
ments in designing and exploring various electrocatalysts to
improve the reaction kinetics and durability. Nonetheless,
there are still numerous glitches which delay the widespread
application of fuel cells. Understanding the problems asso-
ciated with the oxygen electrocatalysis in half-cell and full cell
through experimental and theoretical aspects are highly
desired.[19–24]
This work reviews the recent progress of efficient ORR
electrocatalysis in the fuel cells. Following the brief introduc-
tion, we review various high-performance catalysts at rotating
disk electrode (RDE) in half-cell and membrane electrode
assembly (MEA) in full cell. Moreover, focusing on the
structure–activity relationship, the fundamental problems
Fuel cells are an incredibly powerful renewable energy technology,
but their broad applications remains lagging because of the high cost
and poor reliability of cathodic electrocatalysts for the oxygen
reduction reaction (ORR). This review focuses on the recent progress
of ORR electrocatalysts in fuel cells. More importantly, it highlights the
fundamental problems associated with the insufficient activity trans-
lation from rotating disk electrode to membrane electrode assembly in
the fuel cells. Finally, for the atomic-level in-depth information on
ORR catalysts in fuel cells, potential perspectives are suggested,
including large-scale preparation, unified assessment criteria,
advanced interpretation techniques, advanced simulation and artificial
intelligence. This review aims to provide valuable insights into the
fundamental science and technical engineering for efficient ORR
electrocatalysts in fuel cells.
From the Contents
1. Introduction 3
2. Pt-based catalysts for ORR 4
3. Non-platinum catalysts for ORR 7
4. ORR catalysts in fuel cells 9
5. Challenges and perspectives 14
6. Concluding remarks 19
[*] S. Zaman, L. Huang, A. I. Douka, H. Yang, B. You, B. Y. Xia
Key Laboratory of Material Chemistry for Energy Conversion and
Storage (Ministry of Education), Hubei Key Laboratory of Material
Chemistry and Service Failure, Wuhan National Laboratory for
Optoelectronics, School of Chemistry and Chemical Engineering,
Huazhong University of Science and Technology (HUST)
1037 Luoyu Road, Wuhan 430074 (China)
E-mail: byxia@hust.edu.cn
The ORCID identification number(s) for the author(s) of this article
can be found under:
https://doi.org/10.1002/anie.202016977.
A
ngewandte
Chemi
e
Reviews
&&&&
Angew. Chem. Int. Ed. 2021,60, 2 – 23  2021 Wiley-VCH GmbH www.angewandte.org
These are not the final page numbers!
Ü
Ü

raised from active components and robust supports are
highlighted. More specifically, this review focuses on the
major issues associated with the performance translation from
RDE to MEA. Finally, we propose future perspectives and
solutions for the major problems associated with ORR
catalysts involving large-scale preparation, unified evaluation
standards, advanced interpretation techniques, cutting-edge
simulations, and artificial intelligence for in-depth multiscale
understanding of the ORR in fuel cells (Figure 1). This work
may stimulate more scientific and technological endeavors to
develop efficient ORR catalysts and promote the ultimate
goals towards practical application in fuel cells.
2. Pt-based catalysts for ORR
2.1. Principles for activity improvement
Noble metals, especially Pt catalysts, are highly active in
the fuel cells. Nonetheless, the limited activity and poor
durability of Pt itself put pressure on the cost and perfor-
mance of fuel cells. The thriving
development of nanomaterials and
nanotechnologies brings opportuni-
ties for high-activity Pt catalysts.
Among the different principles, fine
structure tuning through controlled
particle size, crystal morphology and
high-index facets exposure could opti-
mize the ORR activity of Pt catalysts
(Figure 2a).[25] Firstly, a high-surface-
to-volume-ratio could be achieved by
decreasing the particle size, which will
increase the accessible surface atoms
and improve Pt utilization. Secondly,
crystal engineering can realize the
selective exposure of ORR-favorable
crystallographic and high-energy
facets. The exposure of well-charac-
terized surfaces and modified local
atomic environment generate more
unsaturated coordinated sites, which
would enhance the ORR perfor-
mance of Pt crystals. For example,
Pt(111) is 10 times more active for
ORR than Pt(110) and Pt(100) facets.[26] Thirdly, the specific
nanostructure design could lessen the particle agglomeration
and spillover/migration during the electrochemical operation
with the consequent improved activity and stability.[27] For
instance, the thin Pt nanowires (NWs) with more exposed
surface facets and isotropic properties present considerable
ORR activity and robust stability than the traditional Pt
nanoparticles.[28–29] Particularly, the jagged Pt NWs having
2–3 nm diameter and electrochemical surface area (ECSA)
of 118 m2g1Pt exhibit an ultrahigh mass activity of
13.6 Amg1Pt.[30] The unprecedently high activity is ascribed
to a highly stressed, undercoordinated rhombus-rich surface
with abundant exposed active sites.
The above-mentioned principles are useful for tuning the
Pt properties through the physical structure modifications.
However, Pt has a higher oxygen adsorption energy which
needs to be further optimized for the improved ORR kinetics.
According to the Sabatier principle, for the efficient ORR on
Pt surface, the oxygen-binding should be neither very strong
nor very weak. Density-functional theory (DFT) calculations
suggest a theoretical oxygen adsorption (Oads) energy value of
Shahid Zaman is a doctoral candidate at
School of Chemistry and Chemical Engineer-
ing, Huazhong University of Science and
Technology (HUST) under the supervision of
Prof. Bao Yu Xia. He received his master
degree in inorganic/analytical chemistry
(2016) from Quaid i Azam University
(QAU) Islamabad. His current research
interest focuses on the fabrication of PGM-
electrocatalysts for fuel cells.
Bao Yu Xia is a full professor at the School
of Chemistry and Chemical Engineering at
Huazhong University of Science and Tech-
nology (HUST). He received his Ph.D.
degree in Materials Science and Engineering
from Shanghai Jiao Tong University (SJTU)
in 2010. He worked at Nanyang Technolog-
ical University (NTU) from 2011 to 2016.
He serves as an editorial board member in
Chinese Journal of Catalysis. His research
interests focus on nanocatalysts in sustain-
able energy and environment technologies
including fuel cells, batteries and carbon
dioxide conversion.
Figure 1. ORR catalysts evolution towards practical application of fuel cells.
A
ngewandte
Chemi
e
Reviews
&&&&
www.angewandte.org  2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021,60,2–23
Ü
Ü
These are not the final page numbers!

0.4 eV for Pt, which is larger than the optimum Oads energy of
0.2 eV.[31–33] It means oxygen binds very strongly on the Pt
surface, which makes the overall ORR sluggish. The incor-
poration of transition metal into the Pt lattice produces strain
resulting in the weakened oxygen adsorption on the strained
Pt lattice. Therefore, alloying Pt with transition metals (PtM)
is a preferable choice to optimize the surface energy through
the strain-induced surface facets. An extensive analysis of Pt
biaxial strain reveals a huge difference in electrocatalytic
performance with 90% enhancement and 40 % reduction in
activity through compressive and tensile strain, respec-
tively.[34] Interestingly, a lattice strain of only 1% brings
a shift of 0.1 eV in Pt 5d-band center, which significantly
modifies the binding energies of oxygen species on the
strained Pt lattice.[35] Moreover, alloying consequences tuning
the near-edge local Pt structure by the electronic and lattice
modification, which result in the weakening of chemisorbed
surface oxygen species. Similarly, the lattice contraction
reduces the unoccupied projected electronic states resulting
in a downshift of d-band; hence both factors equally govern
the electrocatalytic properties of PtM catalyst.[36–39] Therefore,
the modified local structure of PtM as compared to pure Pt
improves the surface
adsorption of oxygen
species for the
enhanced oxygen elec-
trocatalysis.[40–42]
Additionally, alloy-
ing strategies also pro-
vide opportunities to
develop special mor-
phologies of Pt cata-
lysts due to their struc-
tural evolution diver-
sity to form core/shell
or hollow nanotubes/
nanocages/nanoframes,
e.g., nanowires,[28–30,43]
nanoplates,[44] core/
shells,[35,45] nanocrys-
tals,[46,47] and nano-
cages[33] or nano-
frames,[48,49] etc. (Fig-
ure 2b). Selective etch-
ing or dealloying
leads to the removal of
unstable components
and tunes the local
active surface, thereby
improving the activity
by geometric advan-
tages and induced
strain, ligand and
ensemble effects. The
structural evolution of
Pt catalysts from nano-
particles to high-
dimensional architec-
tures has been a signifi-
cant addition to ORR catalysis, which unveils the structure–
activity relationship through the comprehensive lattice strain,
exposed surface atoms, high-index facets and uncoordinated
atomic sites (Figure 2c). Particularly, the hollow architecture
plays a vital role in the ORR catalysis due to the highly open
and rough structure with maximum exposed active sites,
lattice strain, more unsaturated sites and high surface-to-
volume ratio that provides three-dimensional (3D) accessi-
bility to the reactants and efficient electron transfer. A typical
example of Pt3Ni nanoframes demonstrates 36 times better
specific activity than Pt/C due to the minimum coverage by
oxygenated species; meanwhile, the Pt-rich skin alleviates the
Ni dissolution during the long-run electrochemical process.[48]
Furthermore, introducing the third metal has a significant
effect on the electronic and geometric properties of Pt alloys
since it brings additional, comprehensive surface strain in
PtM alloy surface layers.[49] Binary PtNi nanoalloy has been
the most active among the transition metal alloys. Never-
theless, a near edge and center facets engineering of Pt3Ni
catalyst by Mo doping reveal one of the best ORR activities of
6.98 Amg1(Figure 3a–c).[50] Fine-tuning the chemical and
electronic properties of the surface layer by strong Mo-Pt and
Figure 2. (a) Proposed strategies for the improvement of oxygen reduction activity of Pt catalysts through
optimized size, crystal structure, electronic strucuture and metal-support interaction. (b) Graphical illustration of
different morphologies obtained by alloying Pt with transition metal for the improved ORR performance.
(c) Graphical description of mass activity and specicific activity of recently reported Pt catalysts.
A
ngewandte
Chemi
e
Reviews
&&&&
Angew. Chem. Int. Ed. 2021,60, 2 – 23  2021 Wiley-VCH GmbH www.angewandte.org
These are not the final page numbers!
Ü
Ü

Mo-Ni bonding mitigates the Pt and Ni dissolution during
electrochemical operation, making the Mo-Pt3Ni a highly
active and durable catalyst. Such a specific design principle
could be worthy towards next-level PtM alloy for highly
durable oxygen electrocatalysis. Highly ordered alloy or
intermetallics tune the very local Pt atomic structure, which
brings more advanced intrinsic properties due to the strong
Pt-M interaction and increased Pt-M bonding.[51–54] Contrary
to the disordered alloys, the ordered intermetallics have
specific arrangements of Pt and transition metals with
a precise crystal lattice properties. Ordered alloy provides
biaxial strain with tension in one particular direction and
compression in the other, which endows remarkable changes
in electrochemical behavior of PtM alloys.[39] Different kind
of ordered alloys have been reported with significantly
improved structural and electrocatalytic properties than
their counter disordered alloys, e.g., binary PtCo,[55–57]
PtCo3,[58] PtFe,[45] PtIn,[59] CoPt,[60] ternary PtNiCo.[61] Con-
trolling particle size in intermetallics is highly desirable, as
most of the intermetallics are prepared by annealing at high
temperature, where avoiding particle aggregation is challeng-
ing. Usually, metal oxides or carbon coatings are utilized to
avoid particle aggregation or sintering, yet it affects the
nanoparticle movement leading to a partially formed ordered
structure. Recently, ZnO is used as the bifunctional precursor
to provide the diffusion of atomic Zn into Pt lattice to form
ordered L10PtZn, where ZnO also acts as a physical barrier to
prevent particle sintering during the heat treatment (Fig-
ure 3d–g). Strong ionic/metallic binding between Zn and Pt
with a shortened Pt lattice parameter and unique biaxial
strain having tension along (110) direction and compression
along (101) and (011) directions contributed to its high ORR
activity.[62]
2.2. Strategies for stability enhancement
The stability of Pt catalysts determines the service life and
cost of fuel cells. Improving their stability, to some degree, is
called maintaining high activity during the electrochemical
operation. Previous studies suggest that the Pt corrosion/
dissolution is triggered by the surface Pt oxidation followed
by the chemical dissolution, therefore covering the surface
oxygen-binding species could avoid the Pt oxidation with
subsequent obstruction of Pt corrosion and dissolution.[63]
However, there is also a possibility of catalyst degradation
through the electrochemical dissolution of bare Pt nano-
particles. In this regard, previous investigations on the Pt
degradation reveal that the Pt mass remains unchanged at
working potential range of 0–0.8 V vs. RHE, while it
significantly drops at the potential more than 0.8 V vs. RHE
under the highly acidic conditions.[64–65] In the view of limited
durability of Pt/C, the Pt size and crystal optimization could
endow the enhanced stability as the tiny particles are
Figure 3. (a) Representative example of PtM (Pt3Ni) alloy depicting enhanced ORR perfromance by the incarporation of trace amount of third
metal (Mo-Pt3Ni), (b) ORR polarization cuvers and (c) corresponding mass activity profile of doping different metals into Pt3Ni. Reproduced from
Ref. [50] with permission from AAAS. (d,e) Representative ordered alloy (PtZn) HR-TEM and mapping profile, (f,g) ORR polarization curves and
corresponding mass activity and specific activity of PtZn intermetallic. Reproduced from Ref. [62] with permission from WILEY-VCH. (h,i) HR-TEM
and mapping of PtPb nanoplates (j,k) LSV curves of PtPb nanoplates demonstrating the outstanding stability after 50000 potential cycles and
corresponding mass activity and specific activity. Reproduced from Ref. [68] with permission from AAAS.
A
ngewandte
Chemi
e
Reviews
&&&&
www.angewandte.org  2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021,60,2–23
Ü
Ü
These are not the final page numbers!

thermodynamically unstable and more vulnerable to aggre-
gation due to the higher surface energies. Similarly, carbon
corrosion at high potential also results in the degradation of
Pt/C system. The oxygenated species bind more strongly on
the Pt particles and detach them from the support through the
particle migration or spillover on the surface with consequent
particle redeposition and aggregation. Moreover, the surface
poisoning of Pt by radicals, sulphonic groups, etc. is also
a severe degradation source, therefore increasing size and
optimizing crystal structure could alleviate the coalescence
and sintering of Pt catalyst. However, the larger size could
decrease the ECSA and activity due to the reduced surface-
to-volume ratio, yet a minor activity loss could be compro-
mised sometime for the longer stability of the catalyst. The
subsequent principle is developing complex multidimensional
Pt nanostructures, that resist migration, dissolution and
aggregation compared to the isotropic nanoparticle.[25] Stabi-
lizing Pt by metal with higher oxidation potential which is
resistant against the corrosion and sintering, is a promising
strategy, as evidenced by the Au clusters modified Pt surface,
which exhibit remarkable stability with a negligible decay
after 30,000 potential cycles.[66] Similarly, using some protec-
tive coating could be another strategy, which inhibits the
direct interaction of more vulnerable non-noble metal with
a corrosive environment.[67] Nevertheless, the optimized uni-
form coating, especially carbon coating at high temperature, is
still challenging because it may block the efficient exchange of
the reactants to the active sites.
Despite the size and structural modifications, Pt still has
serious problems of limited stability and higher cost. There-
fore, alloying is a preferable choice to endow more robustness
against the particle dissolution and decrease the cost. A core/
shell PtPb/Pt nanoplate reports a remarkable structural or
compositional durability through a large biaxial strained
structure, which accounts for the superior stability of the
catalyst up to 50,000 potential cycles (Figure 3 h–k).[68]
Advanced characterizations coupled with DFT calculations
reveal that the higher tensile strain along particular facet and
core/shell structured intermetallics phase favors a higher
durability. Although alloying strategies could meet the
durability standards, PtM alloys still suffer from metal
corrosion, leaching or spillover of the surface particle, leading
to redeposition and aggregation with consequent sever
activity degradation. This phenomenon is explored by a Pt-
Au skin in PtAuCo catalyst, which achieves a surprising
durability of only 25% decay after 100,000 cycles.[69] The
exceptional stability is explained through a self-healing
mechanism by Au fine-tuning of surface electronic structure,
where the low surface energy of isolated Au stabilizes the Pt-
Au skin, that alleviates the Co corrosion and dissolution
under low pH environment.[70] Similarly, alloying also brings
the chance to finely tune the near-surface composition, strain,
or ligand effect and enrichment with more anti-corrosion
components to protect the particle spillover during the long-
term conditions. Our recent work on PtNi bunched nanocage
morphology reports a strained Pt-rich skin structure through
selective (electro)chemical corrosion of the Ni species,
demonstrating a higher stability of 50,000 cycles without any
noticeable degradation.[33] Contrary to solid nanowires, which
has limited exposure of active sites due to unexposed internal
atoms, merging highly open nanocages in the nanowire
architecture could be interesting to improve the activity and
durability simultaneously, as the open cage morphology
provides extensive interior and exterior surface-active sites
accessibility for more extensive Pt utilization, while the Pt-
rich skin endows a strong resistance against Ni corrosion and
dissolution during the long-term fuel cell operation.
3. Non-platinum catalysts for ORR
Except for Pt, other Pt-group metals (PGM) are also
effective for the ORR, but they are suffering from the similar
problems of high cost, limited activity and insufficient
durability.[71, 72] It is a grim reality that even the most ingenious
design approaches cannot turn back concerns over the
scarcity and cost of PGM electrocatalysts. Therefore, tran-
sition metal-based non-precious metal catalysts (NPMCs) as
a substitute for costly Pt cathode in fuel cells are the ultimate
long-term solution due to abundant resources and low cost.
Following this consideration, transition metal compounds
(oxides, sulfides, etc.) are extensively investigated as non-Pt
cathode catalysts.[73] However, these materials have good
ORR activity in alkaline media only, while the acidic
environment severely affects their performance resulting
corrosion, dissolution or oxidation at higher working poten-
tials.[74–76] Contrary to transition metal compounds, transition
metal-carbon composites are more robust, which give rea-
sonable ORR activity in acidic environment. Usually, metal
center and defective carbon are the proposed active sites in
these catalysts, hence defect engineering of carbon matrix and
special morphology construction of metal-carbon composites
could be a viable strategy for improving catalytic perfor-
mance.[77] Novel design with particular geometry and mor-
phological features modify the apparent properties of tran-
sition metal-carbon catalysts by maximizing the exposure of
the active site, while heteroatom doping can further modify
their intrinsic properties for high performance (Figure 4a).[78]
Yet, limited understandings of active sites and relevant
reaction mechanisms make them feeble for fuel cells.[79]
Therefore, in-depth knowledge of catalyst nature in active
sites and reaction mechanism is desired to improve the overall
efficiency of non-precious catalysts.
Among transition metal-carbon catalysts, metal-nitrogen-
carbon (M-N-C) electrocatalysts are the most promising
models,[80–82] where M-N and M-C moieties are collectively
responsible for the improved ORR catalysis.[83,84] An exten-
sive understanding at atomic-level insight into active sites
proposes carbon-embedded N-coordinated Fe (Fe-N4) as the
potential active sites for the ORR, where the disordered
carbons with N-functionalities bind Fe species.[85] Compara-
tive analysis reveals that the micropores are not merely
responsible for the improved performance, but the N-bearing
micropores have a crucial role in active site accessibility and
ionomer distribution to achieve outstanding ORR perfor-
mance (5 % decay after 30,000 cycles) (Figure 4b).[86] There-
fore, it might be speculated that both microporosity and
heteroatom doping present the binding spots to metal
A
ngewandte
Chemi
e
Reviews
&&&&
Angew. Chem. Int. Ed. 2021,60, 2 – 23  2021 Wiley-VCH GmbH www.angewandte.org
These are not the final page numbers!
Ü
Ü

particles and create relevant active sites for ORR. Moreover,
the most widely considered active NPMCs are the Fe, Co and
Mn composites with intrinsic activity order Fe >Co >Mn.
However, each suffers from their own disadvantages: the
dissolved Fe specie acts as a Fenton reagent which damages
the ionomer or membrane by generating poisoning radicals,
and even though Co is less Fenton active, but its sourcing is
problematic; similarly, Mn forms inactive phases (oxide,
carbide) during heat treatment, leading to a low Mn-Nx active
site density.[87] Therefore, by the proper orientation of these
materials, their weaker sides could be covered to expose the
active facets and get high performance.
Single-atom catalysts (SACs), where a single transition
metal anchored/integrated into the substrate, is relatively new
frontier of advanced catalysis due to the maximum active
sites, high atom utilization and strong metal-support inter-
action.[88,89] Particularly, the heteroatom-doped SACs are one
of the most promising catalysts that present excellent ORR
performances.[90] For instance, oxygen bridged Fe-O-Fe with
two Fe-N4sites demonstrate an improved oxygen reduction
activity in acid electrolyte (Figure 4c).[91] DFT calculations
evidence that the thermodynamic barrier towards ORR is
considerably reduced due to the modified Fe d-orbital
electronic structure by FeO bond. Similarly, a Mo SAC
with oxygen and sulfur coordination is evidenced highly
Figure 4. (a) Schematic illustration of chemically modified nanocarbons through heteroatom doping as promising catalyst for ORR including N-
doped M-N-C, N-doped M-free carbon, M-free defective carbon and N-doped carbon nanotubes. (b) Schematic synthesis of PANI-M-C catalysts
demonstrating carbon shells with nitrogen functionalities coordinated with metal species. Reproduced from Ref. [86] with permission from the
AAAS. (c) Model illustraion of FeN4sites with different orientation coordinating with spontaneously formed OH ligand. Reproduced from Ref. [91]
with permission from AAAS. (d) Schematic demonstration of pyridonic N formation by OH attachement. ORR reaction pathway on nitrogen-
doped carbon materials. Reproduced from Ref. [103] with permission from AAAS.
A
ngewandte
Chemi
e
Reviews
&&&&
www.angewandte.org  2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021,60,2–23
Ü
Ü
These are not the final page numbers!

active and selective towards ORR, by a two-electron pathway,
which demonstrates the role of local coordination environ-
ments on the electrocatalytic activity and selectivity.[92]
Although SACs are the potential candidates for fuel cells,
there is a long way to realize their actual performance and
stability in acidic environment. Because of the low active site
density of SACs, their metal loadings need to be increased
while maintaining the resolved and isolated active sites for
large-scale applications. More in-depth investigations using
advanced characterization and simulation models could help
to realize the actual active centers and distinct the SACs from
other traditional catalysts.[93] Moreover, carbon support has
a significant role in the service level and life of the catalyst.
Apart from providing anchoring sites, carbon must be highly
anticorrosive to protect the active sites during electrochem-
ical operation. Modified carbon with strong corrosion resist-
ance and high graphitic features could be a valuable addition
to more durable catalysts providing stable anchoring sites to
M-N4active sites.[94] Metal-organic frameworks (MOFs)
derived nanocarbons are promising due to the richly porous
structure, high graphitization, abundant metal ions and
tunable composition and morphology.[95–98] Moreover, novel
design for modified nanocarbons with diverse properties of
higher graphitization, heteroatom doping, rich porosity are
desired to improve the perspective fuel cell performance.
Concerning metal corrosion and leaching in acid environ-
ment, carbon-based metal-free catalysts could be another
practicable choice for durable electrocatalysis in acid envi-
ronment.[99,100] Chemically modified N-doped nanocarbons
are promising alternatives with high electrocatalytic perfor-
mance and more anticorrosive to acid environment than
transition metals.[101,102] In this regard, a recent investigation
characterizes the pyridinic N responsible for the generation of
active sites in N-doped carbon materials.[103] Based on the
model catalyst investigations, carbon atoms next to pyridinic
N are proposed to be the active sites with Lewis basicity at
which O2molecules are adsorbed as the initial ORR step
(Figure 4d). Therefore, modified structural design through
heteroatom doping could improve the intrinsic properties of
carbon-based electrocatalysts.[104] Even though various metal
free catalysts have been developed as ORR catalyst with
higher or comparable ORR performance to the M-N-Cs
counter parts, yet there is insufficient knowledge of the active
sites responsible for the origin of ORR activity in metal-free
carbon-based catalysts. In this regard, a more extensive
analysis of a series of heteroatom-doped graphene as metal-
free catalyst reveals that the heteroatom-doped graphene
could be a viable alternate to replace the Pt cathode catalyst
in fuel cells.[105] A combined experimental and theoretical
analysis revealed by the activity descriptors, exchange current
density, onset potential, reaction pathway selectivity and
kinetic current density demonstrate that the doping of
nonmetal elements in graphene conceives unique intrinsic
properties in enhancing the ORR performance of the catalyst.
Moreover, an extensive investigation on the origin of
ORR activity in metal-free nanocarbon catalyst reveals the
topological defects along with heteroatom dopants as vital
factors to improve the ORR activity of the nanocarbon
catalysts.[106] The dopants, defects and edges effectively
modify the charge/spin distribution on the sp2-conjugated
carbon matrix, leading to the optimized intermediate chem-
isorption and facilitate the electron transfer. Electronegativ-
ity/electron affinity of the dopants, work function of the
carbon, electron localization by the doping of electron
withdrawing groups, defects and edge configurations in
carbon-based metal-free catalysts modulate the adsorption
mode of the oxygen, which optimize the binding of inter-
mediates, weaken the OO bond, and facilitate the electron
transfer to improve the ORR performance by lowering the
favorable overpotential, improved kinetics, and preferential
four-electron pathway. Modification of the graphene oxide
morphology and structural properties by the removal of
intercalated water molecules from the interlayers of graphene
oxide converts it to an impressive ORR catalyst in acidic
media.[107] Controlling the macroscopic structure and mor-
phology of graphene based electrocatalysts upon de-interca-
lation of the trapped water and nitrogen doping leads to
enhanced intrinsic properties which favor the high activity
and durability for ORR in acidic electrolytes. Based on the
recent research progress on nanocarbon-based metal free
catalysts, it is predictable that the nanocarbon-based metal
free catalysts could be the potential catalysts in future, which
will step up the commercial application of fuel cell into more
common practice.
4. ORR catalysts in fuel cells
ORR catalysts are vital in determining the overall
efficiency of fuel cells, but their electrochemical investiga-
tions are always distinct from the fundamental electrochem-
istry and technical engineering in fuel cells. The evaluation of
ORR catalysts at MEAs is thus targeted for achieving
practical application of fuel cells. Even though most of the
technical and engineering factors relying on MEAs, substan-
tial improvements are still desired to meet the comprehensive
requirements of ORR catalysts in RDE and particularly
MEAs for fuel cells.
4.1. Pt-based catalysts
Currently, Pt/C is the only predominant cathode catalyst
owing to its reasonable performance according to the require-
ments of United State department of energy (DOE). Their
evaluation standards are updated with continuous progress in
ORR catalysis for fuel cells (2025 targets, MA at 0.9 ViR-free
0.44 Amg1Pt, with <40% MA loss after 30,000 cycles;
current density >0.30 Acm2at 0.8 V and peak power density
>1.00 Wcm2with PGM utilization of >8kWg
1Pt) (Fig-
ure 5a).[10] Pt/C catalyst having tiny Pt particles supported on
carbon is not a preferable choice for large-scale applications
in fuel cells. Although the surface structural engineering of Pt
nanoparticles evidences improved Pt utilization, the high-
index facets and unsaturated atomic steps, edges and kinks
are thermodynamically unstable during the fuel cell oper-
ation, the deactivated facets and deteriorated morphologies
would result in the degradation of catalytic activity. Con-
A
ngewandte
Chemi
e
Reviews
&&&&
Angew. Chem. Int. Ed. 2021,60, 2 – 23  2021 Wiley-VCH GmbH www.angewandte.org
These are not the final page numbers!
Ü
Ü

sequently, alloying Pt with some transition metals changes the
overall chemistry of the Pt catalysts by the synergetic effect of
induced strain, ligand and ensemble effects to improve the
activity and stability. At the same time, it has the advantages
of overcoming the cost by reducing the Pt usage.[108] There-
fore, PtM alloys have consequential effects on electrocatalytic
properties for practical fuel cells (Figure 5b). Based on the
compositional segregation, Pt alloy with Pt-rich skin-protect-
ing underlying transition metal layers have been more
efficient cathode catalysts. The surface Mo modified PtNi
catalyst achieves a high MEA mass activity of 0.45 A mgPt1at
0.9 V vs. RHE.[109] Surface analysis suggests that the surface
composition segregation by Mo adsorbed atoms shields the Ni
dissolution during the electrochemical operation, preserving
the active sites for long-run ORR catalysis. Composition
segregation through random distribution and synergetic effect
arising from the different electronic arrangements of the
component metals in alloy presents highly strained, unsatu-
rated active sites and high-index facets, which facilitate the
efficient ORR (Figure 5c).[110–112] Featuring a strained surface
Pt alloy, our group reported a Pt-rich skin bunched PtNi
nanocages catalyst having current density of 1.5 Acm2at
0.6 V and power density of 0.92 W cm2with long-term
stability of 180 hours in H2-air single cell (Figure 5 d).[33]
Based on the in situ investigations and theoretical analysis,
the synergistic effect of highly strained Pt-rich surface and
modified coordination environment by the appropriate incor-
poration of Ni provide a weakened Pt-O binding strength,
which accounts for the improved ORR performance.
Similarly, the well-defined lattice structure or core/shell
geometry alleviates the core metal dissolution, thereby
improving the stability at MEA.[113,114] For instance, an
intermetallic L10-PtZn cathode catalyst presents an excep-
tionally high MEA performance of 0.52 A mgPt1at 0.9 ViR-free,
and a power density of 2.0 W cm2with only 16.6% degra-
dation after 30,000 potential cycles.[62] The exceptional
stability arises from the transition of disordered lattice to
ordered/intermetallic alloy, where more pronounced surface
strain results in the optimized Pt-O binding and the increased
vacancy formation energy are responsible for the long-term
Figure 5. (a) Graphical description of US DOE 2025 MEA targets. (b) Graphical ORR activities of model D-Pt3Co NWs and O-Pt3Co NWs
compared to Pt NWs. Reproduced from Ref. [23] with permission from American Chemical Society. (c) Pictorial illustration of J-PtNWs generated
by reactive molecular dynamics simulations. Reproduced from Ref. [30] with permission from AAAS. (d) Schematic illustration of the preparation
of bunched PtNi nanocages. Reproduced from Ref. [33] with permission from AAAS. (e) MEA performance of recently reported catalyst against the
DOE 2025 target for mass activity at BOL and EOL (30,000 ADT cycles).
A
ngewandte
Chemi
e
Reviews
&&&&
www.angewandte.org  2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021,60,2–23
Ü
Ü
These are not the final page numbers!

durability. Similarly, the intermetallics catalyst with core/shell
morphology significantly improves the activity and durability
in fuel cell, as the lattice mismatch between shell and core
structures induces geometric and electronic modification,
which enhance the ORR efficiency.[52] The improved perfor-
mance of ordered alloys originates from the more optimized
Pt-M interaction, where highly ordered arrangement of
segregated atoms and unique biaxial strain improve the
surface energies for the oxygenated species (Figure 5e).[54]
Recently, a two-dimensional coplanar PtCo alloy reported
excellent ORR activity at single cell. The coplanar Pt-carbon
nanomesh demonstrates a peak power density of 1.21 W cm2
and current density of 0.36 Acm2at voltage of 0.80 V in H2/
O2cell, where the interconnected Pt network separated by the
coplanar carbon avoids aggregation and migration of Pt
revealing higher durability under realistic fuel cell condi-
tions.[115] DFT investigations demonstrate that coplanar Pt
surface is covered by the graphitic layer, with the higher
vacancy formation energy than the clean Pt, hence the more
difficult Pt vacancy formation for the covered Pt boosts the
durability of the catalyst.
Despite Pt catalyst structural optimization, carbon sup-
port has a key role in MEA performance as it provides
diffusion channels for mass transport and charge transfer.[116]
Conventional carbon black suffers from serious corrosion and
oxidation under working voltages of fuel cells, which blocks
the diffusion channels with the consequent increase in mass
transport resistance and loss of metal-support interaction.[117]
In view of crucial role of support material in MEA
performance, carbon modification is a viable solution to
improve the overall performance. Chemical modification of
carbon supports could finely tune the microporous channels
to facilitate the reactant diffusion, regulate uniform ionomer
dispersion and water management.[118] An extensive inves-
tigation demonstrate a low-Pt loading catalyst deposited on
defect engineered chemically modified carbon support for
MEA evaluation (Figure 6a).[119] This work primarily empha-
sizes on the oxygen mass transport issues across the carbon
support by chemical modification for the uniform ionomer
distribution at the Pt/support interfaces, which significantly
reduces the mass transport resistance.
In fact, carbon support would bring more benefits beyond
our imaginations through the dispersion of active component
and creation of more reaction channels as well as holding
additional active sites by defect engineering through hetero-
atoms doping and product release (Figure 6 b).[120] To this end,
metal-organic frameworks (MOFs)-derived nanocarbons
with uniformly anchored catalyst particles, bearing more
exposed active sites (alloys, SACs, M-N-C, etc.), enhanced
ECSA and anticorrosion ability are promising replacements
for conventional carbon black.[121,122] Recently, core/shell
PtCo nanoparticles engrossed into MOFs-derived nanocar-
Figure 6. (a) Schematic illustration of ionomer distribution over the Pt/C surface (top) and a modified Pt/N-carbon (bottom). Oxygen transport
resistance across catalytic layer for poorly and well distributed ionomer over catalyst. Model graphics of carbon modification process at different
conditions and the resultant H2-air fuel cell performance. Reproduced from Ref. [119] with permission from Springer Nature. (b) Drawing
demonstrating ORR kinetic and transport in catalyst layers based on different porosity of carbon due to ionomer adsorption with the fuel cell
performance comparison of PtCo catalysts. Reproduced from Ref. [120] with permission from American Chemical Society. (c) Schematics of HOR-
selective catalysts at HOR and ORR-relevant conditions. Cross-section of MEA cathode before and after SU/SD durability testing for Pt/C and Pt/
m-HxWO3as anode with polarization curves for different catalysts. Reproduced from Ref. [128] with permission from Springer Nature.
A
ngewandte
Chemi
e
Reviews
&&&&
Angew. Chem. Int. Ed. 2021,60, 2 – 23  2021 Wiley-VCH GmbH www.angewandte.org
These are not the final page numbers!
Ü
Ü

bon exhibits outstanding activity owing to the densely
populated Co-Nx-C sites and Co@graphene along with
some active single atomic sites.[123] The extensive synergetic
effect from several kind of catalytic sites boosts the MEA
performance at both high-voltage and high-current domains
with MA of 1.08 A mgPt1and retaining the initial values by
64% after 30,000 potential cycles, even at an ultralow-Pt
loading of 0.033 mgPt cm2. Following these reports, it is
evident that support materials have a significant contribution
in the intrinsic properties of the catalyst, particularly at MEA
level, where mass transport of reactant to the active sites and
water management are the core issues that impede the MEA
performance. Concerning the carbon corrosion and oxidation
in electrochemical operation, the non-carbon support materi-
als or self-supported carbon-free catalysts are also promising
owing to corrosion resistance nature along with higher
conductivity.[124, 125] Recently, a self-supported Pt-CoO net-
work as carbon-free catalyst demonstrate a remarkable
specific and mass activity with a promising fuel cell perfor-
mance at high temperature, high current density, and low
humidity.[126] Such self-supported design strategies are highly
inspiring to avoid carbon corrosion and mass transport issues
in fuel cells. Similarly, more comparative studies have
unveiled the support influence on ORR catalysis in fuel
cells.[127] Another intelligent catalyst-support design is pro-
posed to suppress the ORR at anode that causes cathode
corrosion during the start-up/shutdown process of fuel cells.
Pt thin layers supported on hydrogen tungsten bronze (Pt/
HxWO3) suppresses the ORR by adapting insulator behavior
following exposure to oxygen while selectively promotes the
hydrogen oxidation (Figure 6 c). Such a unique strategy for
selective electrocatalysis imparted by a metal-insulator tran-
sition in Pt/HxWO3demonstrate a remarkably enhanced
MEA durability as compared to Pt/C catalysts.[128]
Usually, catalyst size, morphology and composition are
focused in improving the ORR kinetics. Certainly, modifica-
tions of structure and composition are crucial to tune the
intrinsic activity of catalyst to improve the ORR kinetics.
However, the local chemical environment of the catalyst is
mostly neglected, even though the triple-phase-boundary
regulations, mass transfer of reactant gases and discharge of
products are extremely crucial at MEA for the overall fuel
cell performance. In this regard, exciting work reports the
porous PtNi composite integrated with the secondary phase
material ionic liquid.[129] Primarily, this work is based on the
principle of impregnating the pores with secondary phase
material having higher oxygen solubility than the exterior
aqueous phase, where the hydrophobic ionic liquid facilitates
the water (product) removal. Hence ionic liquid with the
higher O2solubility preferentially transports oxygen to the
catalyst even without a fully saturated reaction environment.
However, the impregnated ionic liquid must meet the
requirements of higher oxygen solubility, hydrophobicity to
remove the product (water), protic to transport the protons to
active sites and thermally stable. As a matter of fact, this
would be a genuine idea to incorporate such an unsupported
catalyst into MEA with locally optimized triple-phase-boun-
dary without any mass transport problems, conductivity, water
managements, support corrosion and structural stability. Yet,
it is a grand challenge to fabricate mesoscopic nanoparticles
with suitable secondary phase material.
As discussed above, SACs are the emerging addition to
heterogenous catalysis due to high atom utilization efficiency,
however, their present performance is limited to half-cell
only.[72, 130–132] SACs have promising features of their unsatu-
rated coordinated environment, which provides strong metal
support interaction to facilitate the efficient electron trans-
fer.[133] However, insufficient ORR performance of SACs in
fuel cells still requires a more profound understanding of
design strategies and active site engineering. Therefore, the
poisoning resistant support material and strong metal support
interaction through heteroatom-doped moieties are crucial
for the high performance of SACs. Recently, Pt single atom
anchored through pyridinic N (Pt1-N/BP) achieved an
encouraging MEA performance of 680 mW cm2with a re-
markable Pt utilization of 7.5 kWgPt1, especially this catalyst
shows an extraordinary larger surface area of 1102 m2g1and
outstanding durability of 74% current retention for 200 h at
80
8
C.[134] Similarly, another SAC with single Pt atoms
embedded in four carbon through defected graphitic carbon
divacancies as ORR active sites exhibits a superhigh Pt
utilization of 11.1 kWgPt1.[135] Atomic structure analysis
reveals that the local environment and strong metal-support
interaction are directly associated with the electronic struc-
ture of active sites, which determines electron transfer,
adsorption/desorption of oxygenated intermediates and cata-
lyst durability.
4.2. Non-Pt based catalysts
Here the non-Pt catalysts refer to NPMCs, as PGM other
than Pt have similar problems of limited activity, durability
and high cost. Earth-abundant NPMCs are the promising
candidates for ORR in fuel cells, but currently they could not
uphold the desired performance at realistic working con-
ditions of fuel cells as compared to their counterpart Pt-based
catalysts. Contrary to Pt-based catalysts where the active sites
are the distinct Pt particles, the active sites in NPMCs are
closely merged with carbon matrix, which could not be
separated for a clear understanding of reaction mechanism.
Moreover, the active sites are generated in carbon matrix
during the high-temperature annealing; hence the structure–
activity understandings through advanced in situ character-
ization could provide insight into NPMCs. Besides, insuffi-
cient stability of NPMCs resulted from metal leaching, carbon
corrosion, micropores flooding and active site poisoning are
the key factors for the overall degradation (Figure 7 a,b).[136]
Significant progress has been made for the beginning-of-life
(BOL) performance, however insufficient knowledge about
active sites and degradation mechanism hinders the reason-
able end-of-life (EOL) performance of NPMC in practical
fuel cells. Keeping in mind the poor durability of bulky
transition metal nanoparticles in low pH environment, the
corresponding single-atom NPMCs reveal enormous progress
towards competent ORR catalysis in fuel cells. Among the
different NPMCs, Fe composites have the most exciting ORR
performance comparable to Pt/C due to Fe-N4moieties as
A
ngewandte
Chemi
e
Reviews
&&&&
www.angewandte.org  2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021,60,2–23
Ü
Ü
These are not the final page numbers!

active sites embedded into the nitrogen-bearing defective
carbon matrix. Respective active sites provide lower adsorp-
tion energies for the oxygenated intermediates through the
synergetic effect of N-coordinated local structure. However,
the Fe dissolution in low pH leads to Fenton reaction, forming
hydroxyl radicals resulting in cell performance degradation,
which is still a huge threat to Fe-based NPMCs. In recent
work, N-coordinated single Fe in porous carbon architecture
(H-Fe-Nx-C) catalyst demonstrate a current density of
1.55 Acm2and a power density of 655 mW cm2at a working
voltage of 0.43 V, much better than the counterpart cata-
lysts.[137] However, further stability improvements are still
required before its practical applications.
Generally, the catalytic performance of NPMCs is asso-
ciated with uniformly dispersed metal active sites, local N-
binding, crystallinity, and porosity, which determine the
utilization of active centers, accessibility of reactants, and
the removal of products under the actual fuel cell conditions.
Controlled fabrication strategies could help to optimize the
accessibility of the active site in NPMCs, making outstanding
candidates for practical applications. Usually, high-temper-
ature annealing could not guarantee homogenous size and
uniform dispersion of the metal active sites. Therefore,
reasonable low-temperature annealing is viable for the
heterogeneous dispersion (from single atom to particle
formation) with uniformly distributed single-atom sites
throughout the matrix. Following the concerns, employing
cadmium as a sacrificial metal can achieve the preparation of
Fe-based SACs at a low annealing temperature, which
delivers a current density of 44 mA cm2at 0.87 V with
a power density of 680 mWcm2in MEA (iR-free).[138] It is
worthwhile to note that the central transition atom is an
essential factor in designing NPMCs, where the intersectional
activity relies on the nature of transition metals.[72] Similarly,
despite the higher ORR activity of Fe catalysts, the Fe
catalyzed Fenton reaction is unavoidable phenomena in most
of Fe-based ORR electrocatalyst, hence designing non-Fe
electrocatalyst could be more useful towards the enhanced
durability. In a recent report, Co-N-C based catalyst demon-
strate a high activity in H2/O2fuel cell with a current density
of 0.022 A cm2at 0.9 ViR-free and peak power density of
640 mWcm2.[139] High ORR activity in acidic media has been
demonstrated by the experimental results along with DFT
studies, revealing the CoN4moieties as active sites, which
show low activity towards radical-generating Fenton reaction
than the Fe-based catalyst that results extensive stability.
Moreover, the degradation mechanism studies indicate that
chemical oxidation of the catalyst by radicals and active site
demetallation are primarily responsible for the catalytic
degradation of M-N-C catalyst. Both degradation paths
appear to be alleviated for the Co-N-C catalyst relative to
those for the similarly synthesized Fe-N-C catalyst. Recently,
a unique p-block Sn-based single-metal-site non-precious
metal catalyst has emerged as ORR cathode in fuel cells that
exhibit 40–50% enhancement in current density at cell
voltages lower than 0.7 V.[140] Introduction of p-block metals
as cathode material could break the traditional transition
metal catalyst-oxygen adsorptions scenario, where the parti-
Figure 7. (a) Pictorial presentation of major issues associated NPMCs including limited active sites, metal dissolution, carbon oxidation and poor
mass transport. (b) Proposed solutions for the improved activity by modified carbon matrix through higher degree of graphitization, heteroatom
doping, improved atomic coordination sites and large active sites. (c) Schematic of MEA catalyst layer showing efficient O2diffusion through
separated N-G-CNT sheets vs. densely packed N-G-CNT sheets. Porous catalyst layer of N-G-CNT/KB/Nafion with separated N-G-CNT sheets and
polarization curves of N-G-CNT. Reproduced from Ref. [101] with permission from AAAS.
A
ngewandte
Chemi
e
Reviews
&&&&
Angew. Chem. Int. Ed. 2021,60, 2 – 23  2021 Wiley-VCH GmbH www.angewandte.org
These are not the final page numbers!
Ü
Ü

ally filled d-shells in transition metals control the strength of
the adsorbed oxygenated species. In contrast, the closed d-
shells in p-block elements would have very clear interaction
for oxygenated species, which would realize a more profound
understanding and design of active sites and reaction
mechanisms.
With a focus on metal leaching in acid environment, the
metal-free carbon-based catalysts could be another practica-
ble choice for durable fuel cells. In this regard, modified
carbon is more corrosion resistant than transition metals, but
it has unsatisfactory performance in acidic media as compared
to alkaline electrolytes, which need a long way to meetthe US
DOE standards. Despite these issues, a metal-free carbon-
based N-doped graphene/carbon nanotube (N-G-CNT) com-
posite reported as a superior cathode catalyst for single-cell
evaluation (Figure 7 c).[101] The N-G-CNT demonstrate a peak
power density of 30 Ag1(at 0.8 V) and a limiting current of
2000 Ag1at the voltage of 0.1 V due to the enhanced multi-
channel O2pathways in the porous N-G-CNT composites,
which facilitate the effective utilization of the active sites.
More interestingly, the N-G-CNT with different loadings
exhibit much better durability ( 20 % decay over 100 hours)
in acidic fuel cells at a constant voltage of 0.5 V. This work
signifies a breakthrough to replace the costly and unstable
metallic catalyst with metal-free carbon-based ORR catalysts
by translating to commercial reality and opens avenues for
ORR catalysis towards more affordable and durable fuel
cells.
5. Challenges and perspectives
The most significant issues to ORR electrocatalysts are
limited activity, poor durability, and high cost, which hinder
the scale-up applications in fuel cells. More importantly,
unsatisfactory MEA performance is challenging for the
advanced ORR electrocatalysts due to the unoptimized
MEA parameters that include physiochemical features of
catalyst and system engineering of MEA setup. Pt durability
(oxidation/corrosion/dissolution/detachment/agglomeration),
carbon corrosion/oxidation, lack of Pt-support integration,
oxidative H2O2and hydroxyl radicals, and sulphonic groups
poisoning are the fundamental problems associated with the
poor performance of catalysts at MEA.[64, 141–144] Similarly, the
MEA system engineering suffers from water management,
flooding and diffusion channel blockage, temperature and
humidity management, proton and gas conductivity resist-
ance, ionomer issues, three-phase interface management,
start-up/shutdown mechanism, maintenance of proton
exchange membrane, air permeability and hydrophobicity of
gas diffusion layer (GDL), spraying technology of catalyst
layer, and assembling technology of gas diffusion electrode
(GDE) and MEA.[145] Therefore, it is indispensable to
extensively investigate the design principles of cathode
catalysts from the fundamental catalyst side to get MEA
system for insight into the core issues that hinder the
transformation of half-cell performance to the full cell and
subsequent practical applications.
5.1. Challenges
Certainly, the huge difference in evaluation standards of
catalyst at half cell and full cell leads to considerable
mismatch in catalytic activity. In half-cell, the RDE technique
is predominant and fast evaluation for screening intrinsic
activity and durability of catalysts. RDE is a hydrodynamic
working electrode that rotates during experiments making
a constant flux of O2and electrolyte to the delicately thin
catalyst layer on electrode, where the controlled rotation
speed provides homogeneous and reproducible reactant
convection and diffusion to the electrode. Although RDE is
a quick and easily accessible, yet it is highly sensitive to the
layer quality and experimental conditions, which provides
data at very low current density. Consequently, the evaluation
of catalyst performance in the real cell could not be fully
predicted from simple RDE screening data because the
single-cell performance is at much higher current densities.
Contrary to RDE, MEA evaluation is much closer to the
fuel cells, which is fabricated by assembling the proton
exchange membrane, catalyst layer, and GDL. In a typical
MEA, catalyst layers (membranes coated with catalyst) are
sandwiched between the GDLs with appropriate compression
for the optimized transport of reactants and water. Funda-
mental problems associated with MEA are catalyst degrada-
tion, mass transport issue, water management, catalyst layer
thickness, temperature and pressure management. Based on
these issues, a considerable research gap exists between RDE
and MEA due to different testing standards, where both have
particular requirements of experimental conditions. There-
fore, many high-performance ORR catalysts could not trans-
late their RDE activities to the MEA level. Some of the core
issues resulting measurement mismatch between RDE and
MEA are described below.
Ideal conditions at RDE in half-cell (high-speed RDE/
thin catalyst layer) facilitate the efficient mass transfer,
however, the large quantity of reactant gas at MEA results
enormous mass transfer resistance at catalyst sites. Similarly,
poor solubility of oxygen in the electrolyte yields low current
density in RDE, while the current density requirements in
MEA are higher enough to meet the mandatory power
density of the fuel cell device. Therefore, MEA is combined
with GDLs for effective mass transfer of reactant gas to the
active sites in relatively thicker layers, which make it more
complicated. Likewise, RDE has thin catalyst layer, which
facilitates the complete utilization of active sites, while the
thicker catalyst layer in MEA could not guarantee the
extensive usage of active sites. Hence, the triple-phase
interface (solid–liquid–gas), an essential requirement for
efficient mass transport and active site utilization could not
be achieved. Meantime, there are no thermal, gas and water
management issues at RDE level, as the low concentration of
O2in electrolyte is very convenient for smooth operation.
However, such parameters could not be optimized well at
MEA level due to bulk concentration of O2that needs
optimized pressure and flow rate. Also, the MEA operating
temperature is typically ranging from 60–120
8
C, which is
much higher than RDE operation environment (25
8
C).
Moreover, RDE screening could be performed with minor
A
ngewandte
Chemi
e
Reviews
&&&&
www.angewandte.org  2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021,60,2–23
Ü
Ü
These are not the final page numbers!

quantity of catalyst, where finely designed catalyst with
specific morphology and compositions demonstrates excep-
tional catalytic performance. However, large quantity catalyst
requirements in MEA and fuel cell could not maintain the
similar quality that lead to a considerable discrepancy in
material innovation and device integration. Meanwhile,
carbon corrosion and Ostwald ripening of catalyst system at
high voltage leads to the destruction of ionomer and
membrane, while RDE screening has no such serious issues.
Also, the optimized ionomer distribution is prerequisite to
avoid the poisoning of active sites during the catalysis.
Excessive ionomer concentration leads to the blockage of
active sites. Therefore, for more common practice of MEA
operations, still many efforts should be put forward to
facilitate the technology. Similarly, MEA device fabrication
with optimized parameters is complex process which affects
overall performance of the system. Finally, as a simple setup
that is accessible easily, a competent person can perform
several electrochemical RDE measurements in a single day.
However, MEA assembly and its operation need specialized
and costly equipment along with the expertise in fuel cell
investigations.
Based on the existing challenges to the ORR catalyst
towards practical applications, recently numerous efforts have
been put forward to accomplish the parallel performance at
RDE and single cell level. As mentioned above, in fuel cell
the active sites accessibility and mass transport through the
thick catalyst and ionomer layers are significant issues that
result insufficient performance and durability. In this regards
a self-standing PtCo catalyst as electrode layer directly
employed in MEA without traditional spray fabrication
evidenced outstanding performance in fuel cells. Incorporat-
ing cathode catalyst directly into MEA is vital to avoid the
extra complicated preparation of MEA and poor mass
transport properties, which will bring optimized catalyst
layer and ionomer control.[139] Recently, a newly developed
floating electrode techniques (FET) is an emerging evalua-
tion approach to obtain high current density. A comparative
investigation on RDE, FET and MEAwith state-of-the-art Pt/
C catalyst by Martens, S. et al. demonstrate that FET is an
effective technique to get higher current density at extremely
low catalyst loadings with controlled catalyst layer thickness
as compared to MEA.[146] Therefore, newly developed catalyst
must be screened with FET to evaluate the higher current
density prior to apply in MEA or realistic fuel cells. However,
when the sample size is increased beyond 1 cm2the floating
electrode geometry also did not operate well with catalyst-
coated membrane catalyst hot pressed to conventional trans-
port layers, as product water accumulate within the porous
structures due to no force convection. Therefore, a flexible
half-cell testing platform could be used with conventional fuel
cell GDEs and catalyst-coated membrane, with variable
catalyst loading and/or different catalyst/support.[146] More
importantly, such a designed GDE half-cell could be used at
extremely low cost and the utility of the half-cell reveals that
this GDE could be operated at realistic operating current
densities (0–1750 mAcm2) in the half-cell H2/O2fuel cell
with the comparable performance. More importantly the
same setup can be used without any further modifications to
investigate the HOR as well. This kind of approach bridges
the gap between the RDE and the single cell and will
accelerate the realization of actual performance at fuel cell
level.
Similarly, an interesting approach presents comparable
evaluation standards of RDE and MEA through a GDE set
up to translate the laboratory level performance to actual fuel
cell.[147] In contrast to thin-film RDE measurements, perfor-
mance data obtained from GDE measurements can be
directly compared with MEA testing results, that closes the
gap between catalyst research in academia and real applica-
tions. Despite the difference in measurement conditions
between RDE and GDE, Tafel slope for the GDE measure-
ments is almost identical to that of the mass-transport
corrected RDE measurement, which demonstrates the suffi-
cient mass transport properties of the catalyst layer in GDE.
Moreover, the comparison of I-V curves at higher and lower
potential between MEA and GDE setup discloses higher
performance due to the improved mass transport and lower
catalyst loadings. Meanwhile, the ORR specific activity in the
GDE at lower potential, 0.65 V vs. RHE, is comparable to
that in the floating electrode, regardless of the different
temperature and catalyst loadings. To this end, it could be
concluded that the GDE is an efficient technique where the
catalyst can be evaluated under realistic fuel cell operating
conditions as fast and straightforward as in RDE tests with
a very smaller quantity of the catalyst and highly improved
mass transport properties which is one of the major issues
towards high-performance MEA. The utility of the half-cell,
reported here, is demonstrated by testing commercial GDEs
at realistic operating current densities (0–1750 mA cm2)in
the half-cell and a H2/O2fuel cell to show comparable
performance.
5.2. Perspectives
Based on above discussion on problems associated with
the activity translation, following are the proposed perspec-
tives for exploring efficient catalysts to promote the fuel cell
technology (Figure 8).
i) Catalyst innovation. Pt is the priority for ORR catalysis,
however, from a long-term perspective, Pt catalysts
could not be sustained in fuel cells due to complications
of high cost, limited resources, and poor performance.
Hence, exploring novel and upgraded ORR catalysts are
highly desired for the sustainable development of fuel
cells. NPMCs would be the ultimate option in long-run
commercial applications; however, the current status of
the NPMCs is unsatisfactory. Particularly, the SACs are
the promising catalyst with nearly 100 % atom utiliza-
tion, yet their current performance could not meet the
DOE targets. Novel and progressive design strategies
could help to understand the SACs basic synthesis
criteria and structure–activity relationships to avoid the
deactivation by leaching, detaching, corrosion and the
oxidation of the coordinated heteroatoms during the
fuel cell operation. More importantly, advanced oper-
ando characterizations could be utilized to monitor the
A
ngewandte
Chemi
e
Reviews
&&&&
Angew. Chem. Int. Ed. 2021,60, 2 – 23  2021 Wiley-VCH GmbH www.angewandte.org
These are not the final page numbers!
Ü
Ü

reaction process and analyze the structure–activity
relationship for the determination of efficient electro-
catalysts.
ii) Support upgradation. Carbon has been the ideal support
material for Pt-based catalysts due to the high surface
area and reasonable conductivity. However, it suffers
from corrosion/oxidation under low pH resulting in
detachment of Pt particles with consequent performance
degradation. Therefore, modified carbon support with
stronger anchoring sites and corrosion resistance is
highly desired to avoid Pt dissolution/detachment.
Heteroatom doped carbon support not only provides
a strong metal-support interaction but also maintains
a uniform distribution of metal particles to avoid
migration and aggregation. Although in Pt/C, the
carbon acts as support to disperse the Pt particles,
however, mesoporous carbon in NPMCs has key role in
reaction as the active sites are merged in the carbon
matrix, which provides diffusion channel for the reac-
tants and water diffusion. As an active part of the
NPMCs, carbon matrix should have appropriate poros-
ity and conductivity to hold active sites and facilitate the
efficient electron and mass transport during the reac-
tion. Chemical modification of the carbon support
through heteroatom doping provides strong anchoring
sites and higher graphitization provides corrosion resist-
ance leading to improved stability of the catalyst against
the poisoning species.[145] Moreover, support material
with selective functions is promising towards improved
fuel cell catalysis. Composite materials that are selective
towards specific reactions behaving differently to par-
ticular potentials and pH conditions could alleviate the
corrosion problems to improve the performance. More
importantly, the microenvironment modulation by new
support materials is highly recommended to create
effective triple-phase interface to overcome the limited
activity, as poor triple-phase interface leads to limited
utilization of the active sites resulting in poor catalyst
activity.
iii) Large-scale preparation. The morphology sensitive,
finely designed catalyst with superior RDE performance
rarely performs well at MEA level due to poorly
controlled characteristics of catalyst in bulk quantities.
A more extensive understanding of synthesis protocol at
large scale should be explored to maintain the similar
quality at gram or even kilogram level to facilitate the
practical applications. Similarly, novel preparation
methods should be proposed at kilogram level to meet
huge demands of ORR catalysts in fuel cells by utilizing
the advanced AI-based simulation techniques along
with DFT and simulation modeling. Scale-up prepara-
tion by solid-state methods would encourage the bulk
synthesis of catalyst as compared to traditional wet
chemical method for large scale applications. Mean-
while, cost is the major issue while considering the scale-
up synthesis; therefore, higher Pt utilization strategies
could facilitate the reduction of overall cost of Pt
catalyst. Similarly, the equipment is also important for
the large quantity production to maintain the uniform
quality of the catalyst material, as the quality control at
larger scale production is highly challenging. Therefore,
sophisticated automatic equipment governed by the self-
operated machines could be utilized through simulation
modeling and machine learning to maintain the uniform
quality and quantity of products in future.
Figure 8. Perspectives from fabrication, characterization, evaluation and commercialization for the ORR catalyst in fuel cells.
A
ngewandte
Chemi
e
Reviews
&&&&
www.angewandte.org  2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021,60,2–23
Ü
Ü
These are not the final page numbers!

iv) Self-supported catalyst layer. The complicated mem-
brane electrode assembly process is one of the major
reasons for limited performance due to uncontrolled
catalyst layers thickness, mass transport resistance and
limited accessibility to active sites. Therefore, designing
a catalyst layer in the form of a self-supported mem-
brane without traditional catalyst ink spraying would be
promising solution for convenient MEA processing and
efficient catalytic activity. Well-defined superstructure
and smart membranes with optimized flow channel
could withdraw the generated water from the cathode
and internal fuel cell self-humidification of air could
resolve the humidity and water management issues of
the MEA in fuel cells. Similarly, it is crucial to
investigate the microenvironment of the catalyst mate-
rial for understanding the microstructure, actual active
sites and subsequent reaction mechanism for the evolu-
tion of self-supported catalyst system. Therefore, micro-
environment creation via innovation of catalysts from
integrated catalysts to self-supported catalyst layers (ca.
integrating the self-supported catalyst layers with the
proton exchange membrane) would facilitate the active
site utilization with the subsequent upgradation of the
effective catalyst for scale-up synthesis in commercial
fuel cells.
v) Fabrication strategy. The state-of-the-art Pt/C severely
suffers from degradation process, which is mainly
characterized by the ECSA loss associated with Pt
crystal growth due to the dissolution and redeposition of
Pt. Similarly, carbon oxidation/corrosion at high poten-
tials, particularly greater than 0.80–0.85 V, is also
alarming, especially at start-up/shutdown stages.
Hence, more profound fabrication methods could be
designed that combine optimized composition, con-
trolled particle size and geometric strain, exposed high-
index facets of Pt and strong metal-support interaction
with graphitic carbon for efficient catalysis at MEA
level. Moreover, the fabrication of controllable synthesis
through fine-tuning catalyst morphologies, and the
subsequent preparation of catalyst layers for the realistic
application in fuel cells is highly demanding. Noticeably,
fabrication of integrated or self-supported catalysts is
promising choice, specially at MEA level that could
avoid the traditional complicated preparation of spray
ink method which results Pt detachment from support.
Similarly, designing catalyst layer membranes with
appropriate porosity and thickness could be a unique
approach to prepare highly effective catalyst which can
be directly used in the MEA. Such catalyst layer would
prevent major performance loss due to poor mass
transport across the traditionally ink sprayed and hot-
pressed catalyst layer, where porosity and thickness
could not be controlled efficiently. Additionally, such
a self-supported catalyst layer would guarantee the
durability of the catalyst under realistic fuel cell
conditions without any limitations of Pt nanoparticle
dissolution/aggregation or carbon corrosion. Also, this
approach could be easily utilized at large scale prepa-
ration. Following the US DOE requirements of high
activity and stability, the Pt-nanocarbon integrated
system would provide more robust activity, oxidation
resistance, improved conductivity, and mass transport
due to synergy between Pt and supports.
vi) Unified evaluation. Currently, a considerable gap exists
in the evaluation standards of RDE and MEA. Highly
active and durable catalysts could not maintain a similar
performance at MEA level due to complicated assembly
method and dramatic difference in testing conditions of
single cell. Single cell experiments (MEA) need highly
sophisticated setup having parameters very different
from half-cell (RDE); hence a reasonable comparison
could not be guaranteed. One of the major issues is the
mass-transport phenomena that play a crucial role in the
cells performance, which is substantially different at
half cell and MEA level. Similarly, no standard operat-
ing procedure established yet, which leads to different
testing protocols and testing parameters complicating
a valid comparison between different literature results.
Therefore, standardized half-cell experiments using gas
diffusion electrode (GDE) setup could play an impor-
tant role accelerating the understanding and optimiza-
tion of catalyst layer parameters to further improve fuel
cell performance. GDE method could provide higher
current density at lower catalyst loadings with superior
mass transport properties, which is technologically
closer to single cell configuration. Similarly, MEA
experiments are time and material intensive and do
not allow direct insights into one single electrode
(usually lack of a reference electrode). To bridge the
gap between RDE and single cell experiments and to
combine their advantages, GDE experiments allow fast
and comparable testing at standardized operating con-
ditions and provide dedicated insights into one single
electrode with realistic catalyst layer parameters at
relevant potential and current ranges. Therefore, this
method can be an optimal supplement to drastically
shorten the time from successful catalyst synthesis to an
operating electrode for fuel cell applications. Similarly,
floating electrode technique has been also used to
balance the RDE and MEA testing conditions to
overcome the existing gaps between RDE screening
and MEA testing.[148] These methods provide adaptable
and matching testing conditions to practical fuel cells,
which can be employed at broad range conditions with
different Pt loading, variable catalyst layers thickness to
get higher current density. Similarly, such techniques
with similar operating conditions would help in quick
evaluation and screening of the activity and stability for
the service and failure of catalysts in fuel cells.
vii) Advanced characterization. Understanding the catalytic
process and degradation mechanism (metal dissolution/
carbon corrosion and electrode layers collapse) at
multiscale is crucial for the in-depth investigations of
catalytic sites and dynamic process at RDE and MEA
level. Therefore, advanced in situ techniques, i.e., in situ
X-ray absorption spectroscopy, incidence angle X-ray
diffraction, Raman spectroscopy, differential electro-
chemical mass spectrometry and transmission electron
A
ngewandte
Chemi
e
Reviews
&&&&
Angew. Chem. Int. Ed. 2021,60, 2 – 23  2021 Wiley-VCH GmbH www.angewandte.org
These are not the final page numbers!
Ü
Ü

microscopy could be utilized to study mechanism
associated with catalyst degradation. Advanced in–situ
and operando techniques help to comprehend the two
different characteristics of catalysis. One aspect is the
analysis of material evolution, structural changes,
boding types of intermediates and consequent activity
degradation. More advanced techniques could be devel-
oped to analyze the dynamic reaction or degradation
process at the atomic/molecular level at catalyst surface.
Another useful aspect of the operando analysis is
monitoring the reaction pathway and interaction of
intermediates to the active sites of the catalyst, which
help to build a detailed structure–activity relationship.
Recently massive research interest has been developed
towards in situ/ex-situ-operando analysis for the micro-
scale understandings of the degradation process in
MEA.[149,150] An important investigation on insitu and
operando characterization of PEM fuel cells presents
a comprehensive analysis on catalyst layer degradation,
water management and catalyst layer-ionomer interface
role in mass transport of the reactants and product
delivery during the fuel cell operation.[151] Similarly,
another operando X-ray absorption near-edge structure
computed tomography (XANES-CT) analysis have
been employed to detect the dissolution and migration
of Pt and Co in Pt-Co cathode during the MEA
operation.[152] Investigations based on the combined
studies of ex-situ/in situ measurements along with
operando XANES-CT analysis demonstrated variable
valence states of Pt and Co after different accelerated
stress cycles, which evidence that the Pt and Co
degradation is highly dependent on the structure
(crack) of the carbon support. Likewise, in situ/ex-situ
simultaneous multi-series measurements on degradation
of catalyst layer during the fuel cell operation provide
extensive visualization and understanding of the perfor-
mance degradation of the MEA. The study based on the
integrated measurements through different techniques
on similar sample simultaneously, reveal highly reliable
results by visualization analysis of the anode-gas
exchange degradation as compared to the classical
accelerated stress cycles analysis.[153] The anode-gas
exchange treatment (start-up/shut-down simulation)
causes the transformation of initial active Pt species to
less active and inactive Pt species as evidenced from
voltage-transient response kinetics by time resolved
quick scanning X-ray absorption fine structure
(QXAFS), the corrosion of the carbon support and the
associated Pt ions elution and shedding of Pt as
visualized by nano-XAFS/STEM-EDS same-view imag-
ing. These investigation conclude that among the various
degradation processes, the serious MEA degradation
due to the anode-gas exchange cycles (start-up/shut-
down operation) of polymer electrolyte fuel cells is one
of the major issues that must be solved.[154] Based on the
short overview of the accomplishment in situ/ex-situ and
operando techniques employed for the analysis of
degradation mechanism, it is speculated that advanced
microlevel analysis will figure out the actual deactiva-
tion mechanism such as dissolution, oxidation, corro-
sion, and aggregation effectively very soon which will
provide new ways to prepare more robust and efficient
ORR electrocatalysts for practical fuel cells.
viii) Theoretical investigation. Based on the comprehensive
experimental results, theoretical analysis could be per-
formed to derive the reaction mechanism and interac-
tion with active sites during catalytic reaction. DFT
studies, in conjunction with experimental investigation,
could be utilized for designing suitable parameters for
a better understanding of structure–activity correlation
at MEA level. Moreover, the multiphysical fields
simulations are also deserved for understanding these
local reaction sites and environment influence on the
expression and translation of electrochemical activity
and cell performance. Of course, the large results and
models will contribute to the subsequent construction of
material genome family, which is helpful for high-
throughput screening and identifying the efficient elec-
trocatalysis and their corresponding reaction mecha-
nisms.
ix) Artificial intelligence. Despite complicated electro-
chemical, thermodynamic and mechanical pilot experi-
ments to optimize the fuel cell parameters, the artificial
intelligence (AI) based analysis is a new, quick and
efficient tool to optimize the working parameters of fuel
cells. Machine learning could be utilized to analyze the
available data to reproduce reliable, accurate and
optimized conditions without intricate and laborious
experimentation through algorithm operation for spe-
cific problems. Computer-aided machine learning pro-
grams provide optimized models for the fuel cell, which
could be employed on a large scale.[155] Furthermore, the
AI-based protocol could be used to understand the
reaction mechanism, especially the degradation mecha-
nism at the MEA level and provide robust solutions.[156]
Combined with abundant data from the experimental
results and theoretical analysis, the autonomous learning
would promote the exploration of new and novel
catalysts for ORR in practical fuel cells.
x) System engineering. Even though the catalysis is the
core element of MEA for fuel cell operation, system
engineering of MEA, fuel cell, and cell stack is crucial to
regulate the different parameters, e.g., water manage-
ment, flooding and diffusion channel blockage, temper-
ature and controlled humidity, proton and gas conduc-
tivity resistance, ionomer issues, three-phase interface
management for the improved performance. Modifica-
tion and optimization of mechanical properties of the
catalyst, including catalyst loading and layer thickness,
ionomer ratio, humidification and water management,
should be analyzed and updated to improve the
performance of fuel cells. Similarly, the operating
conditions such as temperature, backpressure, reactant
flow, air permeability and hydrophobicity of the gas
diffusion layer could not be overlooked during the fuel
cell operation for efficient performance. Moreover, the
fuel cell cost could be compensated by appropriate
selection of MEA components based on low cost and
A
ngewandte
Chemi
e
Reviews
&&&&
www.angewandte.org  2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021,60,2–23
Ü
Ü
These are not the final page numbers!

highly durable nature. Only developing the matched
techniques for ORR catalysts, it can maximumly release
the capability of ORR catalysts, and thus fuel cells could
be stepped up much closer to large-scale application.
6. Concluding remarks
Present review summarizes recent achievements on
efficient ORR catalysis in fuel cells, specifically highlighting
the core issues related to ORR activity measurements of
cathode catalysts at RDE and MEA. Recently Pt catalyst
performance has been significantly improved by different
design strategies, including controlled size, composition,
active facets engineering and alloying Pt with other transition
metals to optimize oxygen adsorption energy. Different Pt-
based alloy with unique morphologies demonstrated remark-
able performance; however, for more swift commercial
applications, advanced novel design and their large-scale
preparation strategies should be explored to minimize the Pt
usage and improving the durability for long service life,
particularly the significant and unexpected effects raised from
supports could not be ignored. Similarly, the NPMCs show
substantial improvement in ORR catalysis; therefore, the
replacement of Pt catalyst with NPMCs will boost the
practical applications of fuel cells. However, the NPMCs
still have unsatisfactory ORR activity in acid media; thus, in
situ characterization techniques could be utilized to observe
the reaction mechanism of the catalysts, and analyze the
structure–activity relationship of the materials. Similarly,
novel cutting-edge theoretical analysis and simulations mod-
elling can realize the structure–activity relationship by under-
standing the actual reaction mechanism. Meanwhile, artificial
intelligence and big data analysis could help to select and
design a more suitable catalyst model for practical applica-
tions and machine learning can collect and interpret the
details of specific problems from the available big data in
a very quick and precise way and propose models to resolve
the issue for the efficient catalysis. We anticipate that this
contribution would stimulate academia and industry towards
broader understandings of ORR catalysts in fuel cells.
Acknowledgements
This work is funded by the National Natural Science
Foundation of China (22075092), the Program for HUST
Academic Frontier Youth Team (2018QYTD15) and the
National 1000 Young Talents Program of China.
Conflict of interest
The authors declare no conflict of interest.
[1] M. K. Debe, Nature 2012,486, 43 – 51.
[2] Y. Shao, J. P. Dodelet, G. Wu, P. Zelenay, Adv. Mater. 2019,31,
1807615 – 1807622.
[3] I. E. L. Stephens, J. Rossmeisl, I. Chorkendorff, Science 2016,
354, 1378 – 1379.
[4] G. Shen, J. Liu,H. B. Wu, P. Xu, F. Liu, C. Tongsh, K. Jiao, J. Li,
M. Liu, M. Cai, J. P. Lemmon, G. Soloveichik, H. Li, J. Zhu, Y.
Lu, Nat. Commun. 2020,11, 1191 – 1200.
[5] R. Lin, X. Cai, H. Zeng, Z. Yu, Adv. Mater. 2018,30, 1705332–
1705338.
[6] M. Shao, Q. Chang, J. P. Dodelet, R. Chenitz, Chem. Rev. 2016,
116, 3594 – 3657.
[7] Y. Li, H. Wang, C. Priest, S. Li, P. Xu, G. Wu, Adv. Mater. 2020,
2000381 – 2000417.
[8] Y. He, S. Liu, C. Priest, Q. Shi, G. Wu, Chem. Soc. Rev. 2020,49,
3484 – 3524.
[9] S. T. Thompson, D. Papageorgopoulos, Nat. Catal. 2019,2, 558 –
561.
[10] D. D. Papageorgopoulos, U.S Department of Energy, 2019.
[11] A. Holewinski, J.-C. Idrobo, S. Linic, Nat. Chem. 2014,6, 828 –
834.
[12] M. Li, X. Bi, R. Wang, Y. Li, G. Jiang, L. Li, C. Zhong, Z. Chen,
J. Lu, Matter 2020,2, 32 – 49.
[13] T. Ioroi, Z. Siroma, S.-i. Yamazaki, K. Yasuda, Adv. Energy
Mater. 2019,9, 1801284.
[14] Y. Nie, L. Li, Z. Wei, Chem. Soc. Rev. 2015,44, 2168 – 2201.
[15] A. Morozan, B. Jousselme, S. Palacin, Energy Environ. Sci.
2011,4, 1238 – 1254.
[16] Z. W. Seh, J. Kibsgaard, C. F. Dickens, I. Chorkendorff, J. K.
Norskov, T. F. Jaramillo, Science 2017,355, eaad4998.
[17] A. Ohma, T. Mashio, K. Sato, H. Iden, Y. Ono, K. Sakai, K.
Akizuki, S. Takaichi, K. Shinohara, Electrochim. Acta 2011,56,
10832 – 10841.
[18] K. Ehelebe, D. Seeberger, M. T. Y. Paul, S. Thiele, K. J. J.
Mayrhofer, S. Cherevko, J. Electrochem. Soc. 2019,166, F1259.
[19] S. Sui, X. Wang, X. Zhou, Y. Su, S. Riffat, C.-j. Liu, J. Mater.
Chem. A 2017,5, 1808 – 1825.
[20] X. X. Wang, M. T. Swihart, G. Wu, Nat. Catal. 2019,2, 578 – 589.
[21] A. V. Virkar, Y. Zhou, J. Electrochem. Soc. 2007,154, B540.
[22] L. Castanheira, W. O. Silva, F. H. B. Lima, A. Crisci, L. Dubau,
F. Maillard, ACS Catal. 2015,5, 2184 – 2194.
[23] H. Y. Kim, J. M. Kim, Y. Ha, J. Woo, A. Byun, T. J. Shin, K. H.
Park, H. Y. Jeong, H. Kim, J. Y. Kim, S. H. Joo, ACS Catal.
2019,9, 11242 – 11254.
[24] W. Ding, L. Li, K. Xiong, Y. Wang, W. Li, Y. Nie, S. Chen, X. Qi,
Z. Wei, J. Am. Chem. Soc. 2015,137, 5414 – 5420.
[25] X. L. Tian, Y. Y. Xu, W. Zhang, T. Wu, B. Y. Xia,X. Wang, ACS
Energy Lett. 2017,2, 2035 – 2043.
[26] G. Wang, Z. Yang, Y. Du, Y. Yang, Angew. Chem. Int. Ed. 2019,
58, 15848 – 15854 ; Angew. Chem. 2019,131, 15995 – 16001.
[27] S. Xu, Y. Kim, J. Park, D. Higgins, S.-J. Shen, P. Schindler, D.
Thian, J. Provine, J. Torgersen, T. Graf, T. D. Schladt, M.
Orazov, B. H. Liu, T. F. Jaramillo, F. B. Prinz, Nat. Catal. 2018,
1, 624 – 630.
[28] K. Jiang, D. Zhao, S. Guo, X. Zhang, X. Zhu, J. Guo, G. Lu, X.
Huang, Sci. adv. 2017,3, e1601705.
[29] Y. Chen, T. Cheng, W. A. Goddard III, J. Am. Chem. Soc. 2020,
142, 8625 – 8632.
[30] M. Li, Z. Zhao, T. Cheng, A. Fortunelli, C. Y. Chen, R. Yu, Q.
Zhang, L. Gu, B. V. Merinov, Z. Lin, E. Zhu, T. Yu, Q. Jia, J.
Guo, L. Zhang, W. A. Goddard III, Y. Huang, X. Duan, Science
2016,354, 1414 – 1419.
[31] J. Greeley, I. E. L. Stephens, A. S. Bondarenko, T. P. Johansson,
H. A. Hansen, T. F. Jaramillo, J. Rossmeisl, I. Chorkendorff,
J. K. Nørskov, Nat. Chem. 2009,1, 552 – 556.
[32] S. J. Hwang, S. K. Kim, J. G. Lee, S. C. Lee, J. H. Jang, P. Kim,
T. H. Lim, Y. E. Sung, S. J. Yoo, J. Am. Chem. Soc. 2012,134,
19508 – 19511.
A
ngewandte
Chemi
e
Reviews
&&&&
Angew. Chem. Int. Ed. 2021,60, 2 – 23  2021 Wiley-VCH GmbH www.angewandte.org
These are not the final page numbers!
Ü
Ü

[33] X. Tian, X. Zhao, Y. Q. Su, L. Wang, H. Wang, D. Dang, B. Chi,
H. Liu, E. J. M. Hensen, X. W. D. Lou, B. Y. Xia, Science 2019,
366, 850 – 856.
[34] V. R. Stamenkovic, B. Fowler, B. S. Mun, G. F. Wang, P. N. Ross,
C. A. Lucas, N. M. Markovic, Science 2007,315, 493 – 497.
[35] P. Strasser, S. Koh, T. Anniyev, J. Greeley, K. More, C. Yu, Z.
Liu, S. Kaya, D. Nordlund, H. J. N. c. Ogasawara, Nat. Chem.
2010,2, 454 – 460.
[36] X. Shen, S. Dai, Y. Pan, L. Yao, J. Yang, X. Pan, J. Zeng, Z.
Peng, ACS Catal. 2019,9, 11431 – 11437.
[37] M. Luo, S. Guo, Nat. Rev. Mater. 2017,2, 17059.
[38] Y. J. Wang, N. Zhao, B. Fang, H. Li, X. T. Bi, H. Wang, Chem.
Rev. 2015,115, 3433 – 3467.
[39] J. Pan, X. L. Tian, S. Zaman, Z. Dong, H. Liu, H. S. Park, B. Y.
Xia, Batteries Supercaps 2019,2, 336 – 347.
[40] S. Zhang, Y. Hao, D. Su, V. V. Doan-Nguyen, Y. Wu, J. Li, S.
Sun, C. B. Murray, J. Am. Chem. Soc. 2014,136, 15921 – 15924.
[41] L. Bu, Y. Feng, J. Yao, S. Guo, J. Guo, X. Huang, Nano Res.
2016,9, 2811 – 2821.
[42] C. Zhang, W. Sandorf, Z. Peng, ACS Catal. 2015,5, 2296 – 2300.
[43] L. Bu, S. Guo, X. Zhang, X. Shen, D. Su, G. Lu, X. Zhu, J. Yao,
J. Guo, X. Huang, Nat. Commun. 2016,7, 11850.
[44] S. Luo, W. Chen, Y. Cheng, X. Song, Q. Wu, L. Li, X. Wu, T. Wu,
M. Li, Q. Yang, K. Deng, Z. Quan, Adv. Mater. 2019,31,
1903683 – 1903689.
[45] J. Li, Z. Xi, Y. T. Pan, J. S. Spendelow, P. N. Duchesne, D. Su, Q.
Li, C. Yu, Z. Yin, B. Shen, Y. S. Kim, P. Zhang, S. Sun, J. Am.
Chem. Soc. 2018,140, 2926 – 2932.
[46] L. Cao, Z. Zhao, Z. Liu, W. Gao, S.Dai, J. Gha,W. Xue, H. Sun,
X. Duan, X. Pan, T. Mueller, Y. Huang, Matter 2019,1, 1567 –
1580.
[47] S. I. Choi, S. Xie, M. Shao, J. H. Odell, N. Lu, H. C. Peng, L.
Protsailo, S. Guerrero, J. Park, X. Xia, J. Wang, M. J. Kim, Y.
Xia, Nano Lett. 2013,13, 3420 – 3425.
[48] Y. K. Chen Chen, Z. Huo, Z. Zhu, W. Huang, H. L. Xin, J. D.
Snyder, D. Li, J. A. Herron, M. Mavrikakis, M. F. Chi, K. L.
More, Y. Li, N. M. Markovic, G. A. Somorjai, P. Yang, V. R.
Stamenkovic, Science 2014,343, 1339 – 1343.
[49] P. M. M. Escudero-Escribano, M. H. Hansen, U. G. Vej-
Hansen, A. Velzquez-Palenzuela, V. Tripkovic, J. Schiøtz, J.
Rossmeisl, I. E. L. Stephens, I. Chorkendorff, Science 2016,352,
73 – 76.
[50] X. Huang, Z. Zhao, L. Cao, Y. Chen, E. Zhu, Z. Lin, M. Li, A.
Yan, A. Zettl, Y. M. J. S. Wang, Science 2015,348, 1230 – 1234.
[51] B. N. Wanjala, B. Fang, J. Luo, Y. Chen, J. Yin, M. H. Engelhard,
R. Loukrakpam, C. J. Zhong, J. Am. Chem. Soc. 2011,133,
12714 – 12727.
[52] Z. Meng, F. Xiao, Z. Wei, X. Guo, Y. Zhu, Y. Liu, G. Li, Z.-Q.
Yu, M. Shao, W.-Y. Wong, Nano Res. 2019,12, 2954 – 2959.
[53] T. Y. Yoo, J. M. Yoo, A. K. Sinha, M. S. Bootharaju, E. Jung,
H. S. Lee, B.-H. Lee, J. Kim, W. H. Antink, Y. M. Kim, J. Lee, E.
Lee, D. W. Lee, S.-P. Cho, S. J. Yoo, Y.-E. Sung, T. Hyeon, J.
Am. Chem. Soc. 2020,142, 14190 – 14200.
[54] C. Cui, L. Gan, M. Heggen, S. Rudi, P. Strasser, Nat. Mater.
2013,12, 765 – 771.
[55] X. X. Wang, S. Hwang, Y. T. Pan, K. Chen,Y. He, S. Karakalos,
H. Zhang, J. S. Spendelow, D. Su, G. Wu, Nano Lett. 2018,18,
4163 – 4171.
[56] W. Zhao, Y. Ye, W. Jiang, J. Li, H. Tang, J. Hu, L. Du, Z. Cui, S.
Liao, J. Mater. Chem. A 2020,8, 15822 – 15828.
[57] D. Wang, H. L. Xin, R. Hovden, H. Wang, Y. Yu, D. A. Muller,
F. J. DiSalvo, H. D. Abruna, Nat. Mater. 2013,12, 81 – 87.
[58] Z. Wang, X. Yao, Y. Kang, L. Miao, D. Xia, L. Gan, Adv. Funct.
Mater. 2019,29, 19029287 – 19029295.
[59] Q. Wang, Z. L. Zhao, Z. Zhang, T. Feng, R. Zhong, H. Xu, S. T.
Pantelides, M. Gu, Adv. Sci. 2020,7, 1901279 – 1901286.
[60] J. Li, S. Sharma, X. Liu, Y.-T. Pan, J. S. Spendelow, M. Chi, Y.
Jia, P. Zhang, D. A. Cullen, Z. Xi, H. Lin, Z. Yin, B. Shen, M.
Muzzio, C. Yu, Y. S. Kim, A. A. Peterson, K. L. More, H. Zhu,
S. Sun, Joule 2019,3, 124 – 135.
[61] T. Wang, J. Liang, Z. Zhao, S. Li, G. Lu, Z. Xia, C. Wang, J. Luo,
J. Han, C. Ma, Y. Huang, Q. Li, Adv. Energy Mater. 2019,9,
1803771 – 1803778.
[62] J. Liang, Z. Zhao, N. Li, X. Wang, S. Li, X. Liu, T. Wang, G. Lu,
D. Wang, B. J. Hwang, Y. Huang, D. Su, Q. Li, Adv. Energy
Mater. 2020,10, 2000179 – 2000187.
[63] A. A. Topalov, I. Katsounaros, M. Auinger, S. Cherevko, J. C.
Meier, S. O. Klemm, K. J. Mayrhofer, Angew. Chem. Int. Ed.
2012,51, 12613 – 12615 ; Angew. Chem. 2012,124, 12782 – 12785.
[64] X. Liu, J. Chen, G. Liu, L. Zhang, H. Zhang, B. Yi, J. Power
Sources 2010,195, 4098 – 4103.
[65] S. G. Rinaldo, J. Stumper, M. Eikerling, J. Phys. Chem. C 2010,
114, 5773 – 5785.
[66] J. Zhang, K. Sasaki, E. Sutter, R. R. Adzic, Science 2007,315,
220 – 222.
[67] X. Tong, J. Zhang, G. Zhang, Q. Wei, R. Chenitz, J. P. Claverie,
S. Sun, Chem. Mater. 2017,29, 9579 – 9587.
[68] L. Bu, N. Zhang, S. Guo, X. Zhang, J. Li, J. Yao, T. Wu, G. Lu,
J. Y. Ma, D. Su, X. Huang, Science 2016,354, 1410 – 1414.
[69] X. Tan, S. Prabhudev, A. Kohandehghan, D. Karpuzov, G. A.
Botton, D. Mitlin, ACS Catal. 2015,5, 1513 – 1524.
[70] K. Sasaki, H. Naohara, Y. Choi, Y. Cai, W.-F. Chen, P. Liu, R. R.
Adzic, Nat. Commun. 2012,3, 1115 – 1123.
[71] M. Luo, Z. Zhao, Y. Zhang, Y. Sun, Y. Xing, F. Lv, Y. Yang, X.
Zhang, S. Hwang, Y. Qin, J. Y. Ma, F. Lin, D. Su, G. Lu, S. Guo,
Nature 2019,574, 81 – 85.
[72] Q. Liu, Y. Li, L. Zheng, J. Shang, X. Liu, R. Yu, J. Shui, Adv.
Energy Mater. 2020,10, 2000689 – 2000696.
[73] Y. Liang, Y. Li, H. Wang, J. Zhou, J. Wang, T. Regier, H. Dai,
Nat. Mater. 2011,10, 780 – 786.
[74] Y. Xu, P. Deng, G. Chen, J. Chen, Y. Yan, K. Qi, H. Liu, B. Y.
Xia, Adv. Funct. Mater. 2020,30, 1906081 – 1906088.
[75] A. I. Douka, Y. Xu, H. Yang, S. Zaman, Y. Yan, H. Liu, M. A.
Salam, B. Y. Xia, Adv. Mater. 2020,32, 2002170 – 2002177.
[76] H. Zhang, W. Xia, H. Shen, W. Guo, Z. Liang, K. Zhang, Y. Wu,
B. Zhu, R. Zou, Angew. Chem. Int. Ed. 2020,59, 1871 – 1877;
Angew. Chem. 2020,132, 1887 – 1893.
[77] K. Wu, L. Zhang, Y. Yuan, L. Zhong, Z. Chen, X. Chi, H. Lu, Z.
Chen, R. Zou, T. Li, C. Jiang, Y. Chen, X. Peng, J. Lu, Adv.
Mater. 2020,32, 2002292 – 2002297.
[78] K. Zeng, X. Zheng, C. Li, J. Yan, J. H. Tian, C. Jin, P. Strasser,
R. Yang, Adv. Funct. Mater. 2020,30, 2000503 – 2000525.
[79] J. Masa, W. Xia, M. Muhler, W. Schuhmann, Angew. Chem. Int.
Ed. 2015,54, 10102 – 10120 ; Angew. Chem. 2015,127, 10240 –
10259.
[80] L. Du, V. Prabhakaran, X. Xie, S. Park, Y. Wang, Y. Shao, Adv.
Mater. 2020,32, 1908232 – 1908249.
[81] P. Wang, Y. Ren, R. Wang, P. Zhang, M. Ding, C. Li, D. Zhao, Z.
Qian, Z. Zhang, L. Zhang, L. Yin, Nat. Commun. 2020,11,
1576.
[82] M. Tavakkoli, E. Flahaut, P. Peljo, J. Sainio, F. Davodi, E. V.
Lobiak, K. Mustonen, E. I. Kauppinen, ACS Catal. 2020,10,
4647 – 4658.
[83] G. Chen, P. Liu, Z. Liao, F. Sun, Y. He, H. Zhong, T. Zhang, E.
Zschech, M. Chen, G. Wu, J. Zhang, X. Feng, Adv. Mater. 2020,
32, 1907399 – 1907406.
[84] S. Wang, L. Shang, L. Li, Y. Yu, C. Chi, K. Wang, J. Zhang, R.
Shi, H. Shen, G. I. Waterhouse, S. Liu, J. Tian, T. Zhang, H. Liu,
Adv. Mater. 2016,28, 8379 – 8387.
[85] X. Wang, Y. Jia, X. Mao, D. Liu, W. He, J. Li, J. Liu, X. Yan, J.
Chen, L. Song, A. Du, X. Yao, Adv. Mater. 2020,32, 2000966 –
2000974.
A
ngewandte
Chemi
e
Reviews
&&&&
www.angewandte.org  2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021,60,2–23
Ü
Ü
These are not the final page numbers!

[86] G. Wu, K. L. More, C. M. Johnston, P. Zelenay, Science 2011,
332, 443 – 447.
[87] J. Li, M. Chen, D. A. Cullen, S. Hwang, M. Wang, B. Li, K. Liu,
S. Karakalos, M. Lucero, H. Zhang, C. Lei, H. Xu, G. E.
Sterbinsky, Z. Feng, D. Su, K. L. More, G. Wang, Z. Wang, G.
Wu, Nat. Catal. 2018,1, 935 – 945.
[88] W. Zang, Z. Kou, S. J. Pennycook, J. Wang, Adv. Energy Mater.
2020,10, 1903181 – 1903188.
[89] B. Lu, Q. Liu, S. Chen, ACS Catal. 2020,10, 7584 – 7618.
[90] H. Xu, D. Cheng, D. Cao, X. C. Zeng, Nat. Catal. 2018,1, 339 –
348.
[91] H. T. Chung, D. A. Cullen, D. Higgins, B. T. Sneed, E. F. Holby,
K. L. More, P. Zelenay, Science 2017,357, 479 – 484.
[92] C. Tang, Y. Jiao, B. Shi, J.-N.Liu, Z. Xie, X. Chen, Q. Zhang, S.-
Z. Qiao, Angew. Chem. Int. Ed. 2020,59, 9171 – 9176; Angew.
Chem. 2020,132, 9256 – 9261.
[93] X. Wan, X. Liu, Y. Li, R. Yu, L. Zheng, W. Yan, H. Wang, M.
Xu, J. Shui, Nat. Catal. 2019,2, 259 – 268.
[94] H. Peng, Z. Mo, S. Liao, H. Liang, L. Yang, F. Luo, H. Song, Y.
Zhong, B. Zhang, Sci. Rep. 2013,3, 1765.
[95] X. X. Wang, D. A. Cullen, Y. T. Pan, S. Hwang, M. Wang, Z.
Feng, J. Wang, M. H. Engelhard, H. Zhang, Y. He, Y. Shao, D.
Su, K. L. More, J. S. Spendelow, G. Wu, Adv. Mater. 2018,30,
1706758 – 1706768.
[96] G. Lu, S. Li, Z. Guo, O. K. Farha, B. G. Hauser, X. Qi, Y. Wang,
X. Wang, S. Han, X. Liu, J. S. DuChene, H. Zhang, Q. Zhang, X.
Chen, J. Ma, S. C. Loo, W. D. Wei, Y. Yang, J. T. Hupp, F. Huo,
Nat. Chem. 2012,4, 310 – 316.
[97] Q. L. Zhu, J. Li, Q. Xu, J. Am. Chem. Soc. 2013,135, 10210 –
10213.
[98] B. Y. Xia, Y. Yan, N. Li, H. B. Wu, X. W. Lou, X. Wang, Nat.
Energy 2016,1, 15006.
[99] X. Zhou, J. Qiao, L. Yang, J. Zhang, Adv. Energy Mater. 2014,4,
1301523 – 1301547.
[100] S. Chen, J. Bi, Y. Zhao, L. Yang, C. Zhang, Y. Ma, Q. Wu, X.
Wang, Z. Hu, Adv. Mater. 2012,24, 5593 – 5597.
[101] J. Shui, F. Du, L. Dai, Sci. Adv. 2015,1, e1400129.
[102] K. P. Gong, F. Du, Z. H. Xia, M. Durstock, L. M. Dai, Science
2009,323, 760 – 764.
[103] D. H. Guo, R. Shibuya, C. Akiba, S. Saji, T. Kondo, J.
Nakamura, Science 2016,351, 361 – 365.
[104] T. Zhou, Y. Du, S. Yin, X. Tian, H. Yang, X. Wang, B. Liu, H.
Zheng, S. Qiao, R. Xu, Energy Environ. Sci. 2016,9, 2563 –
2570.
[105] Y. Jiao, Y. Zheng, M. Jaroniec, S. Z. Qiao, J. Am. Chem. Soc.
2014,136, 4394 – 4403.
[106] C. Tang, Q. Zhang, Adv. Mater. 2017,29, 1604103 – 1604112.
[107] U. Martinez, J. H. Dumont, E. F. Holby, K. Artyushkova, G. M.
Purdy, A. Singh, N. H. Mack, P. Atanassov, D. A. Cullen, K. L.
More, M. Chhowalla, P. Zelenay, A. M. Dattelbaum, A. D.
Mohite, G. Gupta, Sci. Adv. 2016,2, e1501178.
[108] Y. Bing, H. Liu, L. Zhang, D. Ghosh, J. Zhang, Chem. Soc. Rev.
2010,39, 2184 – 2202.
[109] F. Dionigi, C. C. Weber, M. Primbs, M. Gocyla, A. M. Bonastre,
C. Spori, H. Schmies, E. Hornberger, S. Kuhl, J. Drnec, M.
Heggen, J. Sharman, R. E. Dunin-Borkowski, P. Strasser, Nano
Lett. 2019,19, 6876 – 6885.
[110] W. Fang, L. Huang, S. Zaman, Z. Wang, Y. Han, B. Y. Xia,
Chem. Res. Chin. Univ. 2020,36, 611 – 621.
[111] Z. Kong, Y. Maswadeh, J. A. Vargas, S. Shan, Z. P. Wu, H.
Kareem, A. C. Leff, D. T. Tran, F. Chang, S. Yan, S. Nam, X.
Zhao, J. M. Lee, J. Luo, S. Shastri, G. Yu, V. Petkov, C. J. Zhong,
J. Am. Chem. Soc. 2020,142, 1287 – 1299.
[112] R. Chattot, O. Le Bacq, V. Beermann, S. Kuhl, J. Herranz, S.
Henning, L. Kuhn, T. Asset, L. Guetaz, G. Renou, J. Drnec, P.
Bordet, A. Pasturel, A. Eychmuller, T. J. Schmidt, P. Strasser, L.
Dubau, F. Maillard, Nat. Mater. 2018,17, 827 – 833.
[113] V. R. Stamenkovic, B. S. Mun, M. Arenz, K. J. Mayrhofer, C. A.
Lucas, G. Wang, P. N. Ross, N. M. Markovic, Nat. Mater. 2007,6,
241 – 247.
[114] Z. Liu, G. S. Jackson, B. W. Eichhorn, Angew. Chem. Int. Ed.
2010,49, 3173 – 3176 ; Angew. Chem. 2010,122, 3241 – 3244.
[115] Y. Hu, M. Zhu, X. Luo, G. Wu, T. Chao, Y. Qu, F. Zhou, R. Sun,
X. Han, H. Li, B. Jiang, Y. Wu, X. Hong, Angew. Chem. Int. Ed.
2021, https://doi.org/10.1002/anie.202014857.
[116] L. Yang, J. Shui, L. Du, Y. Shao, J. Liu, L. Dai, Z. Hu, Adv.
Mater. 2019,31, 1804799 – 1804819.
[117] Q. Ren, H. Wang, X. F. Lu, Y. X. Tong, G. R. Li, Adv. Sci. 2018,
5, 1700515 – 1700535.
[118] N. Karousis, D. Tasis, Chem. Rev. 2010,110, 5366 – 5397.
[119] S. Ott, A. Orfanidi, H. Schmies, B. Anke, H. N. Nong, J.
Hubner, U. Gernert, M. Gliech, M. Lerch, P. Strasser, Nat.
Mater. 2020,19, 77 – 85.
[120] V. Yarlagadda, M. K. Carpenter, T. E. Moylan, R. S. Kukreja,
R. Koestner, W. Gu, L. Thompson, A. Kongkanand, ACS
Energy Lett. 2018,3, 618 – 621.
[121] X. F. Lu, B. Y. Xia, S. Q. Zang, X. W. D. Lou, Angew. Chem. Int.
Ed. 2020,59, 4634 – 4650 ; Angew. Chem. 2020,132, 4662 – 4678.
[122] P. Deng, F. Yang, Z. Wang, S. Chen, Y. Zhou, S. Zaman, B. Y.
Xia, Angew. Chem. Int. Ed. 2020,59, 10807– 10813; Angew.
Chem. 2020,132, 10899 – 10905.
[123] L. Chong, J. Wen, J. Kubal, F. G. Sen, J. Zou, J. Greeley, M.
Chan, H. Barkholtz, W. Ding, D. J. Liu, Science 2018,362,
1276 – 1281.
[124] K. M. Naik, E. Higuchi, H. Inoue, J. Power Sources 2020,455,
227972 – 227982.
[125] B. Han, Y. Guo, Y. Huang, W. Xi, J. Xu, J. Luo, H. Qi, Y. Ren, X.
Liu, B. Qiao, T. Zhang, Angew. Chem. Int. Ed. 2020,59, 11824 –
11829; Angew. Chem. 2020,132, 11922 – 11927.
[126] G. W. Sievers, A. W. Jensen, J. Quinson, A. Zana, F. Bizzotto,
M. Oezaslan, A. Dworzak, J. J. K. Kirkensgaard, T. E. L.
Smitshuysen, S. Kadkhodazadeh, M. Juelsholt, K. M. O.
Jensen, K. Anklam, H. Wan, J. Schafer, K. Cepe, M. Escu-
dero-Escribano, J. Rossmeisl, A. Quade, V. Bruser, M. Arenz,
Nat. Mater. 2021,20, 208 – 213.
[127] S. Yang, Y. J. Tak, J. Kim, A. Soon, H. Lee, ACS Catal. 2017,7,
1301 – 1307.
[128] S.-M. Jung, S.-W. Yun, J.-H. Kim, S.-H. You, J. Park, S. Lee,
S. H. Chang, S. C. Chae, S. H. Joo, Y. Jung, J. Lee, J. Son, J.
Snyder, V. Stamenkovic, N. M. Markovic, Y.-T. Kim, Nat. Catal.
2020,3, 681 – 690.
[129] J. Snyder, T. Fujita, M. W. Chen, J. Erlebacher, Nat. Mater.
2010,9, 904 – 907.
[130] J. Wang, Z. Huang, W. Liu, C. Chang, H. Tang, Z. Li, W. Chen,
C. Jia, T. Yao, S. Wei, Y. Wu, Y. Li, J. Am. Chem. Soc. 2017,139,
17281 – 17284.
[131] J.-D. Yi, R. Xu, Q. Wu, T. Zhang, K.-T. Zang, J. Luo, Y.-L.
Liang, Y.-B. Huang, R. Cao, ACS Energy Lett. 2018,3, 883 –
889.
[132] H. Zhang, G. Liu, L. Shi, J. Ye, Adv. Energy Mater. 2018,8,
1701343 – 1701366.
[133] H. Zhang, S. Hwang, M. Wang, Z. Feng, S. Karakalos, L. Luo, Z.
Qiao, X. Xie, C. Wang, D. Su, Y. Shao, G. Wu, J. Am. Chem.
Soc. 2017,139, 14143 – 14149.
[134] J. Liu, M. Jiao, L. Lu, H. M. Barkholtz, Y. Li, Y. Wang, L. Jiang,
Z. Wu, D. J. Liu, L. Zhuang, C. Ma, J. Zeng, B. Zhang, D. Su, P.
Song, W. Xing, W. Xu, Y. Wang, Z. Jiang, G. Sun, Nat. Commun.
2017,8, 15938.
[135] J. Liu, M. Jiao, B. Mei, Y. Tong, Y. Li, M. Ruan, P. Song, G. Sun,
L. Jiang, Y. Wang, Z. Jiang, L. Gu, Z. Zhou, W. Xu, Angew.
Chem. Int. Ed. 2019,58, 1163 – 1167; Angew. Chem. 2019,131,
1175 – 1179.
[136] U. Martinez, S. Komini Babu, E. F. Holby, H. T. Chung, X. Yin,
P. Zelenay, Adv. Mater. 2019,31, 1806545 – 1806564.
A
ngewandte
Chemi
e
Reviews
&&&&
Angew. Chem. Int. Ed. 2021,60, 2 – 23  2021 Wiley-VCH GmbH www.angewandte.org
These are not the final page numbers!
Ü
Ü

[137] H. Yang, X. Chen, W. T. Chen, Q. Wang, N. C. Cuello, A.
Nafady, A. M. Al-Enizi, G. I. N. Waterhouse, G. A. Goenaga,
T. A. Zawodzinski, P. E. Kruger, J. E. Clements, J. Zhang, H.
Tian, S. G. Telfer, S. Ma, ACS Nano 2019,13, 8087 – 8098.
[138] T. Al-Zoubi, Y. Zhou, X. Yin, B. Janicek, C. Sun, C. E. Schulz,
X. Zhang, A. A. Gewirth, P. Huang, P. Zelenay, H. Yang, J. Am.
Chem. Soc. 2020,142, 5477 – 5481.
[139] S. Y. Lim, S. Martin, G. Gao, Y. Dou, S. B. Simonsen, J. O.
Jensen, Q. Li, K. Norrman, S. Jing, W. Zhang, Adv. Funct.
Mater. 2020,30, 2006771 – 2006778.
[140] F. Luo, A. Roy, L. Silvioli, D. A. Cullen, A. Zitolo, M. T.
Sougrati, I. C. Oguz, T. Mineva, D. Teschner, S. Wagner, J. Wen,
F. Dionigi, U. I. Kramm, J. Rossmeisl, F. Jaouen, P. Strasser,
Nat. Mater. 2020,19, 1215 – 1223.
[141] P. J. Ferreira, G. J. la O’, Y. Shao-Horn, D. Morgan, R.
Makharia, S. Kocha, H. A. Gasteiger, J. Electrochem. Soc.
2005,152, A2256.
[142] F. Hegge, J. Sharman, R. Moroni, S. Thiele, R. Zengerle, M.
Breitwieser, S. Vierrath, J. Electrochem. Soc. 2019,166, F956.
[143] Z. Y. Liu, J. L. Zhang, P. T. Yu, J. X. Zhang, R. Makharia, K. L.
More, E. A. Stach, J. Electrochem. Soc. 2010,157, B906.
[144] E. Padgett, V. Yarlagadda, M. E. Holtz, M. Ko, B. D. A. Levin,
R. S. Kukreja, J. M. Ziegelbauer, R. N. Andrews, J. Ilavsky, A.
Kongkanand, D. A. Muller, J. Electrochem. Soc. 2019,166,
F198.
[145] Y. Ono, A. Ohma, K. Shinohara, K. Fushinobu, J. Electrochem.
Soc. 2013,160, F779.
[146] S. Martens, L. Asen, G. Ercolano, F. Dionigi, C. Zalitis, A.
Hawkins, A. Martinez Bonastre, L. Seidl, A. C. Knoll, J.
Sharman, P. Strasser, D. Jones, O. Schneider, J. Power Sources
2018,392, 274 – 284.
[147] M. Inaba, A. W. Jensen, G. W. Sievers, M. Escudero-Escribano,
A. Zana, M. Arenz, Energy Environ. Sci. 2018,11, 988 – 994.
[148] L. Huang, S. Zaman, X. Tian, Z. Wang, W. Fang, B. Y. Xia, Acc.
Chem. Res. 2021,54, 311 – 322.
[149] X. Li, X. Yang, J. Zhang, Y. Huang, B. Liu, ACS Catal. 2019,9,
2521 – 2531.
[150] R. Zhou, Y. Zheng, M. Jaroniec, S.-Z. Qiao, ACS Catal. 2016,6,
4720 – 4728.
[151] Q. Meyer, Y. Zeng, C. Zhao, Adv. Mater. 2019,31, 1901900 –
1901925.
[152] Y. Tan, H. Matsui, N. Ishiguro, T. Uruga, D.-N. Nguyen, O.
Sekizawa, T. Sakata, N. Maejima, K. Higashi, H. C. Dam, M.
Tada, J. Phys. Chem. C 2019,123, 18844 – 18853.
[153] K. Higashi, S. Takao, G. Samjeske, H. Matsui, M. Tada, T.
Uruga, Y. Iwasawa, Phys. Chem. Chem. Phys. 2020,22, 18919 –
18931.
[154] S. Takao, O. Sekizawa, K. Higashi, G. Samjeske, T. Kaneko, T.
Sakata, T. Yamamoto, T. Uruga, Y. Iwasawa, ACS Appl. Mater.
Interfaces 2020,12, 2299 – 2312.
[155] S. O. T. Ogaji, R. Singh, P. Pilidis, M. Diacakis, J. Power Sources
2006,154, 192 – 197.
[156] R. Ding, R. Wang, Y. Ding, W. Yin, Y. Liu, J. Li, J. Liu, Angew.
Chem. Int. Ed. 2020,59, 19175 – 19185; Angew. Chem. 2020,
132, 19337 – 19345.
Manuscript received: December 21, 2020
Accepted manuscript online: February 2, 2021
Version of record online: &&
&&
,
&&&&
A
ngewandte
Chemi
e
Reviews
&&&&
www.angewandte.org  2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021,60,2–23
Ü
Ü
These are not the final page numbers!

Reviews
Fuel Cells
S. Zaman, L. Huang, A. I. Douka, H. Yang,
B. You, B. Y. Xia*
&&&& —&&&&
Oxygen Reduction Electrocatalysts
toward Practical Fuel Cells: Progress and
Perspectives
Fuel cells are a powerful renewable energy
technology. This review gives a compre-
hensive overview on oxygen reduction
electrocatalysts towards practical fuel
cells.
A
ngewandte
Chemi
e
Reviews
&&&&
Angew. Chem. Int. Ed. 2021,60, 2 – 23  2021 Wiley-VCH GmbH www.angewandte.org
These are not the final page numbers!
Ü
Ü