Regulating the Coordination of Co sites in Co3O4/MnO2 Compounding for Facilitated Oxygen Reduction Reaction

Article abstract of DOI:10.1002/cssc.202002110

Binary transition metal oxides as a promising oxygen reduction reaction (ORR) catalyst have received significant attention. However, their exact reaction mechanisms are often too complex to be discussed. Herein, novel Co?Mn composites with a well-defined nanostructure were developed for understanding the role of each component. The growth pattern of cobalt oxide and the effects of the coordination environment of Co sites during growth on the overall activity were investigated. Based on experimental and density functional theory studies, it was found that the decaying coordination number directly affected the expression of crystal planes of cobalt oxide, which further had a great influence upon limiting current density of Co?Mn catalysts. The cuboid-Co/Mn catalyst exhibited outstanding limiting current density and showed good stability, related to more highly active (110) planes exposed in Co₃O₄. These provided many references for the preparation of related nonprecious catalysts in various domains.

Full text of DOI:10.1002/cssc.202002110

Chemistry–Sustainability–Energy–Materials
Accepted Article
Title: Regulating the coordination of Co sites in Co3O4/MnO2
Compounding for Facilitated Oxygen Reduction Reaction
Authors: Hao Zhang, Zhiqiang Wang, Chenglong Ma, Zhenhua Zhou,
Limei Cao, and Ji Yang
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.202002110
Link to VoR: https://doi.org/10.1002/cssc.202002110
0
1/2020

ChemSusChem
10.1002/cssc.202002110
FULL PAPER
Regulating the coordination of Co sites in Co
3
O
4
/MnO
2
Compounding for Facilitated Oxygen Reduction Reaction
Hao Zhang, Zhiqiang Wang, Chenglong Ma, Zhenhua Zhou, [a] Limei Cao, [a] and Ji Yang*[a]
[
a]
[b]
[a]
[
a]
Hao Zhang, Chenglong Ma, Dr. Zhenhua Zhou, Dr. Limei Cao, and Prof. Ji Yang
State Environmental Protection Key Laboratory of Environmental Risk Assessment and Control on Chemical Processes, School of Resources and
Environmental Engineering
East China University of Science and Technology
Shanghai 200237, P. R. China.
E-mail: yangji@ecust.edu.cn
[
b]
Dr. Zhiqiang Wang
Key Laboratory for Advanced Materials, Center for Computational Chemistry and Research Institute of Industrial Catalysis, School of Chemistry and
Molecular Engineering
East China University of Science and Technology
Shanghai 200237, P. R. China
Abstract: Binary transition metal oxides as a promising oxygen
reduction reaction (ORR) catalyst had received significant attention.
However, the exact reaction mechanisms of them are often too
complex to be discussed. Herein we report novel Co-Mn composites
with a well-defined nanostructure for understanding the role of each
component. And we investigated the growth pattern of cobalt oxide
and the effects of the coordination environment of Co sites during
growth on the overall activity. Based on experimental and density
functional theory studies, we found the decaying coordination number
directly affected the expression of crystal planes of cobalt oxide, which
further had a great influence upon limiting current density of Co-Mn
catalysts. The cuboid-Co/Mn catalyst exhibits outstanding limiting
current density and shows good stability, related to more highly active
stability.[14] Therefore, to figure out the change process of physical
and chemical properties for catalyst-supports is beneficial to gain
the optimal allocation between supports and activated material.
Likewise, these characteristics are also reflected in most cobalt
oxides, particularly Co
have reported that the expedient modulations of the active sites
were implemented by tailoring the morphologies of Co . This is
because Co2+ and Co3+ sites have different application
environments. For instance, the (110) faces of Co have higher
3 4
O , with spinel structure. Some studies
3
O
4
3
O
4
3
+
activity for CO oxidation because of the numerous Co sites
existing only in this plane[ . Wang et al. demonstrated that the
15]
[16]
Co2+ located in tetrahedron was the main active sites for oxygen
evolution reaction by substituting Co2+ and Co3+ of Co
O
with
other inactive metal cations. Consequently, it is essential for
Co catalysts to discuss the influence of different morphologies
3
4
3 4
(110) planes exposed in Co O . These provide many references for
the preparation of related nonprecious catalysts in various domains.
3 4
O
on catalytic activity. For further improving the ORR performance
of the non-precious catalyst, an alternative route is constructing
bimetallic oxide composite to generate well synergistic effects on
reaction. Du et al. reported that the MnO modified by Co O
2 3 4
Introduction
[17]
presented improved OER activity but a decreasing ORR activity
.[18] Similarly,
The development of renewable energy technology, such as fuel
cells, has become increasingly significant with the shrinkage of
fossil fuel.[ Oxygen reduction reaction (ORR) as a vital half-
because of the occupation of active sites by Co
3 4
O
Kim et al. demonstrated a layered Co -MnO composite, in
3
O
4
2
1]
which this catalyst exhibited more positive onset-potential than
each component but lower limited diffusion current density than
reaction on fuel cells involves
a low-kinetic four-electron
process.[ However, Pt-based noble metal, as the state-of-the-art
ORR catalysts, restricts the commercial development of fuel cells
for its expensive costs and limited resources.[3] Transition metal
oxides exhibiting inimitable shaped-dependent electronics, good
conductivity, and stability have an excellent prospect for energy
storage and conversion technologies, such as supercapacitors
and batteries/fuel cells. Hence low-budget and abundant
transition metal oxide catalysts are desired to develop for
superseding Pt-based catalysts. Among them, manganese and
cobalt oxides always attract extensive attention because of its
controllable morphology and excellent electrochemical activity.[4]
Manganese oxides have a wide application as catalyst
scaffolds due to their diverse morphology and controllable
synthesis methodologies.[5] So far, these are many developed
2]
.[19] So, it is difficult for these composites to give full play to
Co O
3 4
the advantages of all components, due mainly to the
overcomplicated catalytic mechanism of binary catalysts.
Therefore, this paper was paid more attention to local electronic
structure, crystal field effect, and the exact reaction mechanisms
of binary transition metal oxides. For all this, the coordination-
controllable Co
prepared by a two-step hydrothermal route. The coordination
numbers of Co NPs were adjusted by varying the initial
concentration of Co-precursor. The distinguishable nanostructure
of Co NPs contributed to understanding the role of cobalt
oxides in binary transition metal oxides. Meanwhile, the growth
patterns of cobalt oxides on MnO were also discussed in detail.
3
O
4
2
NPs grown on -MnO nanowire were
3
O
4
3
O
4
2
And the changes of oxidation state on the surface and bulk of the
Co-Mn catalysts during the growth process were discussed more
comprehensively by XPS and XAS analysis, respectively, which
was further related to the changes of coordination numbers.
Combined with density functional theory (DFT) calculations, we
try to illuminate that the changed exposed crystal faces resulting
from coordination numbers of Co sites significantly impacted on
ORR performance. Hence, our findings achieved from the
methods for the formation of various morphologies of manganese
oxides, such as nanowires,[
6]
nanorods,[6-7] nanotubes,[8]
[9]
[10]
2
nanoparticles, nanosheets, etc. Specifically, the -MnO with
a 22 tunnel structure exhibits superb ORR performance in
comparison to other MnO
2
phases,[ also a suitable catalyst
11]
supports for generating synergistic effects,[12] facilitating electrons
and ion transport,[13] increasing active sites or improving
1
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202002110
FULL PAPER
conducted study have a referenced role in preparing such Results and Discussion
catalysts.
Figure 1. (a) XRD patterns of prepared catalysts. (b) Surface element ratio of prepared samples. (c) Mn 2p core-level XPS spectra of prepared samples. (d) Mn-K
edge XANES of prepared materials.
Growth patterns of cobalt oxides: The XRD pattern reveals the
The Mn-2p core-level XPS spectra were presented in Figure 1c,
crystalline structure of as-prepared catalysts, showed in Figure 1a. from which two main peaks at approximately 641.34 eV-641.64
The explicit diffraction peaks of all catalysts match well with the α-
MnO phase (PDF 44-0141). Meanwhile, with a greater content of
the cobalt precursor, the emerging Co phase (PDF 42-1467)
diffraction intensity significantly increases, and the α-MnO
substrate still maintains its crystalline structure well.[20] This
phenomenon also is proved by the extended X-ray absorption fine
structure (EXAFS) spectra of the Mn-K edge ( Fig.S1).
eV (Mn-2p3/2) and 653.29 eV-653.59 eV (Mn-2p1/2) can be
identified for all composites.[ Further, the Mn-2p spectra (Figure
1c) was fitted by using Gaussian fitting methods. The Mn-2p3/2
22]
2
3
O
4
2
spectra, as
a
representative for further analysis, was
4+
3+
[23]
deconvoluted into two Mn peaks and one Mn peak.
Moreover, the peak positions and areas of three valence peaks in
Mn-2p3/2 spectra were listed in detail in Table S2, from which we
3
+
4+
Further research discovered that the growth situation of cobalt
can obtain an ever-decreasing ratio of Mn : Mn . This result also
indicated the increasing cobalt oxide occupied on the low valence
oxide is related to the surface oxidation state of the α-MnO
substrate. The not heat-treated MnO nanorod provided a more
proper growing environment than heat-treated MnO for cobalt
2
Mn region, leading to less exposure of Mn3+ on the α-MnO
2
2
2
substrate. Meanwhile, reduced the exposure of low valence Mn
allowed an apparent shift of Mn-2p3/2 peak to higher energy.
Besides, the Mn 3s (Fig.S4) reflecting the average valence state
oxide. This can be demonstrated by the atomic ratio of Co: Mn on
catalyst surface (Figure 1b insert) and the unremarkable XRD
diffraction peak of Co-oxides of Co
The relative content of cobalt oxide, grown on not heat-treated
3
O
4
/heat-treated MnO
2
(Fig.S2). of uncovering surface of MnO
conclusion.[24]
The Mn–K edge X-ray absorption near-edge structure (XANES)
of α-MnO , showed in Figure 1d, was separated into the pre-edge
peak region and white line region. As we all know, this is not
possible that electron transition from 1s to 3d orbital in [MnO
2
, came to a consistent
MnO
MnO
2
2
, apparently increased with the cobalt precursor, while the
that had been heated in the air at 400℃ for two hours
2
presented an almost unchanged ratio of Co: Mn on the catalyst
surface. Hence a reasonable inference was proposed that cobalt
oxide tends to grow on lower coordination manganese region.
And on account of the uneven distribution of K (potassium) in the
6
]
octahedron for centrosymmetric complexities.[25] The emergence
of the pre-edge peak in the region I, indicated the existence of low
coordination manganese caused by the strong interaction
between accumulated K and oxygen in the [2×2] tunnels.[26] In
addition, the attached growth of cobalt oxide resulted in the
decline of pre-edge peak intensity, which could be achieved by
electron transfer at the interface of Co-O-Mn.[27] Meanwhile, the
increased coordination number of manganese can also be proven
[
2×2] tunnels formed by double chains of [MnO
the valence of manganese will be lower in the region of excess K.
So, the more region of excess K was covered by Co
nanoparticles, the more decrease in the ratio of K: Mn (Figure 1b).
After low valence manganese on the α-MnO surface was further
6
] octahedron,[21]
3
O
4
2
oxidized by heat treatment in air, the growth sites of cobalt oxide
decreased. Correspondingly, the three catalysts (Z-Co/Mn, C-
Co/Mn, and I-Co/Mn) of the atomic ratio of Co: Mn and K: Mn had
no significant change. All detailed EDS results were presented in
Fig.S3. And Table.S1 showed that calculated the crystallinity of
2
by k -normalized Mn-K edge EXAFS for prepared samples,
showed in Fig.S1. The peaks structure that its adsorption intensity
above the jump-area, named “white line”, played an important role
in determining the coordination and oxidation state.[28] In the
region Π, we can observe that the “white line” intensity increased
with cobalt oxide content, revealing that a decreasing number of
some prominent peaks for MnO
respectively.
2
before and after heating,
2
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202002110
FULL PAPER
Mn-3d states were occupied by electrons.[29] All of these data
(Fig.S5 d and e), the variform NPs for Co
15-20 nm grown on α-MnO nanorods were presented in Figure
2a-c. Meanwhile, the SEM images (Fig.S5 a-c) also showed the
3 4
O with a size range of
indicated that the growth of Co
O
3 4
nanoparticles on the α-MnO
2
2
substrate provided more coordinate oxygen for the low-valence
manganese at the contact interface of Co-O-Mn.
tiny Co-oxides NPs grew on α-MnO
2
nanorods. The phenomenon
was also consistent with previous reports.[
30]
Although no
NPs on
significant changes in sizes and growth density of Co
nanorods were observed over the prepared Co
materials, noticeable shape changes were reflected in Co
3 4
O
3
O /MnO
4
2
3
O
4
NPs.
The changes were explicitly observed in Figure 2d-f. The most
Co NPs showed a cuboid shape for both Z- Co/Mn (Figure.2d)
and C- Co/Mn (Figure.2e). In consideration of the only plane (220)
observed in the Fourier transform graph and Co NPs mainly
3
O
4
3
O
4
growing along the [110] direction, we inferred that the (110) plane
was preferentially exposed as the main crystal plane on the
surface of Co
previous studies.
cuboid-shape had an increased d-spacing of (220) plane in
comparison to that Co NPs with developed shape, which can
3 4
O NPs. This result was in accord with some
[15, 31]
Besides, the small NPs before it grew into
3
O
4
4
+
be attributed to the presence of higher Co oxidation (Co -O). This
phenomenon will be further discussed in the following. And then,
when the content of Co-precursors reached 75 percent, the shape
3 4
of Co O NPs further became irregular (Figure.2f) due to surface
stress[ . Hence the collapse of part crystalline planes allowed
more other planes to be exposed. As presented in the insert of
Figure.2f, the three planes: (111), (110), and (311) corresponding
to the first three intensity of peaks in the XRD pattern can be
observed together. It is noteworthy that the Co3 cations exist only
32]
+
on the plane (110), which may be why Z- Co/Mn and C- Co/Mn
are more active.[15, 31] The uniform distribution of NPs on nanorods
indicated that the α-MnO nanorods were an excellent growth
2
Figure 2. HRTEM image of Z-Co/Mn (a and d), C-Co/Mn (b and e) and I-Co/Mn
(c and f).
substrate for providing a suitable growth environment for Co
NPs.
3 4
O
The detailed structural modifications of nanoparticles (NPs)
were observed from transmission electron microscopic images
TEM) shown in Figure 2. Compared with pure MnO nanorods
(
2
Figure 3. (a) LSV curves of as-prepared catalysts at 1600 rpm rotation rates in O
2
saturated 0.1 M KOH solution. (b) Tafel curves of as-prepared catalysts at
1
600rpm rotation rates in O saturated 0.1 M KOH solution. (c) The DOS of (111), (311) and (110) crystal planes. (d) Standard free-energy band diagrams obtained
2
at zero potential (U=0 V) and the potential at which all intermediates proceed spontaneously at pH 14 and T=298 K. All potentials are indicated vs. reversible
hydrogen electrode (RHE) unless otherwise noted.
3
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202002110
FULL PAPER
Electrochemical performance: For evaluating the ORR catalytic
activity of prepared Co/Mn composites, the required series of
electrochemical measurements were carried on the cathode
electrode in 0.1 M KOH solution. The cyclic voltammetry (CV)
measurement (Fig.S5) of all catalysts showed a protruding
3 4
activity. By contrast, the process of Co O NPs changes from Z-
Co/Mn to C-Co/Mn can be perceived as further growth along the
[110] direction. Hence there would be an increase in ORR activity.
These suspicions were supported by the Brunner-Emmet-Teller
(BET) surface area results, shown in Fig.S8. From the preceding,
the BET surface area of Co-Mn catalysts depended mostly on the
oxygen reduction peak in O
2
-saturated solution at a scan rate of
-saturated 0.1 M
5
0 mV/s, in contrast with current curves in N
2
Co
had barely changed after the second hydrothermal reaction. With
the growth and collapse of the (110) surface of Co , the surface
3 4 2
O NPs because the morphologies and structures of α-MnO
KOH solution. The introduction of cobalt oxides had significantly
improved ORR reactivities. The ORR polarizations showed in
Figure.3a also revealed that Co-Mn catalysts established more
positive one-set potential and higher limiting current density.
Notably, the three Co-Mn catalysts with remarkably close one-set
and half-wave potential had a significant difference in limiting-
diffusion current density, showed in Table S3 in detail. By a series
of RDE measurements performed with rotating speeds from 400
to 1600 rpm (Fig.S6 b-e), good linearity for Koutecky–Levich (K-
L) plots (Fig.S6) of prepared catalysts were obtained at 0.55 V.
The Co-Mn catalysts presented an excellent selectivity for the
quasi-four-electron pathway of ORR,[33] supported by the
electron-transfer number (n) reached to 3.74, 3.91 and 3.83,
corresponding to Z-Co/Mn, C-Co/Mn and I-Co/Mn respectively.
As showed in Figure 3b, the C-Co/Mn displayed optimum ORR
activity due to the lowest Tafel slope that was an important
parameter for evaluating reaction kinetics.[34] Also, the
dramatically reduced ORR performance of I-Co/Mn was probably
attributed to the decreasing reactive sites reduced by irregular
3
O
4
area showed an increase from Z-Co/Mn to C-Co/Mn and
decreased from C-Co/Mn to I-Co/Mn. Consequently, the increase
or decrease of the high-activity (110) plane can be used as a
guide to estimate the ORR activity of Co-Mn catalysts.
To further demonstrate Co-Mn catalysts have efficient
electrocatalytic properties towards ORR, the C-Co/Mn was
compared with previous reports, showed in Table S4. Meanwhile,
in comparison to commercial Pt/C catalysts was also carried out,
showed in Fig.S9. Although the half-wave potential (E1/2) was still
lower than that of the Pt/C catalyst, the C-Co/Mn catalyst showed
a strong diffusion limiting current density, much higher than that
Pt/C. Besides, C-Co/Mn presented excellent stability and
methanol tolerance. After 32000s operation, C-Co/Mn still
retained 81% of its initial current while the Pt/C catalyst was 75%.
As for methanol tolerance, the Pt/C electrode suffered a higher
fluctuation in the presence of methanol, and there were almost no
effects for C-Co/Mn catalyst. Therefore, the C-Co/Mn catalyst
exhibited an efficient ORR activity comparable to that Pt/C, but
better stability and methanol tolerance ability.
3 4
Co O NPs surface. So considered different cobalt oxidation at
surface ascribed to changes in the shape of NPs, the activity of
Co-Mn catalysts was related to the presence of corresponding
crystal plane. Judging from observed results of XRD and TEM,
the planes of (110), (111), and (311) were further explored by first-
principles DFT calculations. The optimized structure and detailed
calculation data of each plane was presented in Fig.S7. The d-
band center is a good descriptor of occupancy of orbital states
and bond characteristics.[35] Figure 3c showed that the d-band
centers of (111), (311), and (110) planes were -3.06, -2.92, and -
Our experiment results reals that cuboid-Co
α-MnO had a positive effect on promoting ORR. Although
previous studies had discussed the effects of different-shaped
Co
NPs on ORR,[31, 38] the specific internal reasons had not
been explained. Herein, to expound that the (110) plane of the
Co NPs is the highly active plane of ORR, the electronic
3 4
O NPs grown on
2
3
O
4
3
O
4
structure of Co sites was further determined by X-ray absorption
measurements. The superficial chemical oxidation state and
structure information of Co species was showed in Figure.4a. The
Co 2p spectra presenting two spin-orbit doublets with peaks
distance of approximately 15.1 eV, corresponding to Co 2p3/2 and
2
.82 eV, respectively. The (111) crystal plane presented the
lowest d-band centers of energy, indicating that high affinity of
oxygen species, meaning the difficult desorption.[35-36] Therefore,
the oxygenated intermediates would occupy the active site and
affect the further progress of the ORR. This was also reflected in
the free energy diagrams of ORR (Figure 3d), from which the free
energy of OOH*, O* and OH* on (111) plane demonstrated lower
energy than the others. Relatively, the modest d-centers providing
appropriate adsorption and desorption capacity for oxygenated
intermediates can promote the ORR forward.[35] Besides, the
limiting thermodynamic potential where all oxygen intermediates
proceed spontaneously, a certain degree can predict the degree
of ORR activity for the spinel oxides.[37] As illustrated by the red
trendline in Figure 3d, the (110) plane had the highest limiting
thermodynamic potential (U=1.05 V). Hence, on the surface of
Co 2p1/2, were consistent with previous studies on Co
In addition, the appearance of satellite peaks indicated by S
reflected that the valence state components of Co species
3
O
4
.[19, 32, 39]
and
1
S
2
consisted of Co2 and Co sites. Change in the ratio of relative
peak area between Co2+ and Co can be clearly pointed out the
transformation of the chemical environment on the exposed
crystal plane. As seen in Figure.4a, Z- Co/Mn and C- Co/Mn had
+
3+
[40]
3+
2+
3+
the same ratio of Co /Co as 0.75, while I- Co/Mn had higher
relative content of Co2+ so that the ratio increased to 1.39. This
was due to the appearance of (111) and (311) planes on the
exposed surface of I-Co/Mn. Figure.4d showed the surface
atomic configurations of (110), (111), and (311) planes, from
which the (111) and (311) planes contain only Co2+ cations in
comparison to (110) plane. The peak areas and positions of two
valence peaks in Co-2p3/2 spectra were listed in detail in Table S5.
irregular Co
instead of (110) planes appeared on the surface of Co
due to surface stress, which eventually led to a decrease in ORR
3
O
4
NPs, the slightly active (111) or (311) planes
3
O
4
NPs
4
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202002110
FULL PAPER
2
Figure 4. (a) Co 2p core-level XPS spectra of prepared samples. (b) Co-K edge XANES of prepared materials. (c) k -normalized Co-K edge EXAFS for prepared
catalysts. (d) Spinel structure and surface atomic configurations in the (110), (111) and (311) planes of Co crystal.
3
O
4
Structural transformation of cobalt oxides: To further
investigate the octahedral or tetrahedral Co-O sites in Co bulk,
resulted in the decreasing of coordination number,[44] in good
agreement to the result of Co-K edge XANES. Given the above,
3
O
4
X-ray absorption fine structure spectroscopy measurements of Co
were performed (Figure.4b and 4c). The Co-K edge XANES of
prepared catalysts showed an increased absorption intensity that
in the process of Co
average coordination number of Co sites was declining,
corresponding to the proportion of [Co-O ] tetrahedra became
higher. This phenomenon led to decreasing the overall oxidation
state of Co . Although further investigation was required, a
3 4
O growing along the [110] direction, the
4
was positively correlated with the number of Co-tetrahedra in the
41]
Co
3
O
4
bulk (Figure 4b insert I).[ Hence the cobalt sites of Co-
3 4
O
tetrahedra gradually increased during the growth process of
Co NPs. This suspicion was supported by the negative shifts
similar tendency was existed in exposed planes of Co
increasing Co2+ sites on the surface.
3 4
O , scilicet
3
O
4
of peak position in the first derivative of XANES (Figure 4b insert
Π) that was often used to evaluate the oxidation state.[27, 42]
Namely, the increasing relative content of Co-tetrahedra reduced
Conclusion
3 4
the overall oxidation state of Co O in bulk. In order to
comprehensively understand the local symmetry of Co-O
coordination, Fourier transforms of k -normalized Co-K edge x-
ray absorption fine structure (EXAFS) should be acquired.
In sum, the shape-controllable Co
3
O
4
NPs grown on -MnO
2
2
nanowire has been successfully synthesized to explore the ORR
reaction mechanism in binary transition metal oxides. We found
that cobalt oxide tended to grow on the lower coordination
manganese region. Simultaneously, when the content of Co-
As shown in Figure 4c, the three typical peaks corresponded to
the nearest scattering paths of the Co-O shell, the scattering
paths from octahedrally coordinated cobalt to its adjacent cobalt
sites, and the scattering path between two adjacent tetrahedrally
coordinated cobalt sites, respectively.[16, 41] These were annotated
in the crystal diagram presented in Figure 4d. Moreover, a
scattering path observed at R =1.2578 Å was shorter than Co-O
paths, especially presented in Z- Co/Mn but almost disappeared
in I- Co/Mn catalyst. Given that Z- Co/Mn had the highest
oxidation state of Co, this scattering path might derive from the
precursors does not exceed some threshold, the resulting Co
NPs maintained a regular shape and predominantly exposed the
110) surfaces that played an important role in promoting ORR
activities. Among these Co-Mn catalysts, the C-Co /MnO
catalyst exhibits the strongest ORR performance because of its
own most (110) planes of Co . The transition from (110) to
111) or (311) planes was the main cause of decreasing ORR
3 4
O
(
3
O
4
2
3 4
O
(
activity, which was further verified by DFT studies. Besides, the
X-ray absorption measurements also revealed that the
decreasing of average coordination number of Co sites was the
internal reason for the morphology change of Co
However, more investigations are required to explain the
improved ORR performance caused by shape-controllable Co O ,
4
mainly reflects in limiting diffusion current density. It’s also a
promising strategy for enhancing the ORR activity of binary
transition metal oxides catalysts.
4
+
4+
Co -O shell. Meanwhile, excessive presence of Co cations with
a smaller ion radius in the crystal lattice will lead to changes in the
lattice parameters, which can further explain the difference in the
d-spacing of Z- Co/Mn and C- Co/Mn catalysts, even for the same
3 4
O NPs.
(
110) plane. For obtaining the detailed coordination information of
Co sties, the IFEFFIT calculation was used to fit the Co-O shell
peaks (Fig.S10) that can be divided explicitly into [Co-O
tetrahedra and [Co-O
] octahedron.[43] The fitting-results showed
3
4
]
6
that the coordination number (N) * the amplitude attenuation
2
factor (S
0
) followed the order of Z-Co/Mn > C-Co/Mn > I-Co/Mn,
which indicated that the increasing proportion of Co2 sites
+
5
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202002110
FULL PAPER
(
XAS) data of the samples were recorded at room temperature in the
Experimental Section
All chemicals are of analytical grade and were used as received without
further purification.
transmission mode using ion chambers and the BL14W1 beam line in the
Shanghai Synchrotron Radiation Facility (SSRF), China.
Density functional theory (DFT) calculation: all first-principle
calculations were implemented in the Vienna Ab-initio Simulation Package

-MnO
2
nanowires: in a typical experiment, 82 mmol KMnO
4
and 33
mmol MnSO
4
were dissolved in 30 mL deionized water, and then the
(
VASP). The spin-polarized DFT computations were performed with
Perdew-Burke-Ernzerhof (PBE) generalized gradient approximation
GGA) for exchange-correlation interactions. A plane-wave cutoff energy
solution was transferred into a 50 mL Teflon-lined stainless steel autoclave.
The autoclave was sealed and hydrothermally treated at 150℃ for 12 h.
After this, the samples were filtered and washed several times with ethanol
and deionized water. The as-prepared products were dried in air at 70 ℃
overnight for further modification.
(
of all computations was set to 400 eV. The self-consistent field was
considered to converged when energetic convergence threshold changed
within 1 × 10-4 eV/atom. For modelling, the ORR behaviors were predicted
on the (110), (111) and (311) surfaces of the spinel crystal using periodic
Co
were achieved by varying the molar ratio of as-prepared MnO
Co(NO )6H O. The mixing reagents were added to 20 mL deionized water
adjusted to PH=11 by NaOH. The suspension was then homogenized by
ultrasonicating for 30 min and transferred into 50 mL Teflon-lined stainless
steel autoclave. After heating at 150 ℃ for 12 h, samples were collected
by filtration, repeatedly washed with deionized water and ethanol, then
dried overnight at 70 °C under vacuum. According to the different
3
O
4
/MnO
2
composites: the different shapes of Co
O
3 4
growth on MnO
2
2
× 1 supercell containing six O-Co-O layers. And the consideration of
2
and
selecting the crystal plane comes from the actual results of XRD and TEM.
All crystal structure information and phase parameters are produced by
the National Institute of Standards and Technology (NIST). The changes
of each oxygen-intermediate reaction during ORR in alkaline solution as
following:
3
2
−
∗
−
−
푶 (품) + ퟐ푯 푶(풍) + ퟒ풆 ⇌ 푶푶푯 + 푶푯 + 푯 푶(풍) + ퟑ풆
precursors of Co(NO
3
)6H
2
O and MnO
2
, obtain Z-Co/Mn, C-Co/Mn, and I-
ퟐ
ퟐ
ퟐ
(
(
(
(
2)
3)
4)
5)
Co/Mn, corresponding atom ratio is 0.25:1, 0.5:1, and 0.75:1 respectively.
∗
−
−
∗
−
−
푶푶푯 + 푶푯 + 푯 푶(풍) + ퟑ풆 ⇌ 푶 + ퟐ푶푯 + 푯 푶(풍) + ퟐ풆
ퟐ
ퟐ
Electrode preparation and Electrochemical characterization: the as-
prepared catalysts were combined with Vulcan carbon at a 2:3 mass ratio
푶 + ퟐ푶푯 + 푯 푶(풍) + ퟐ풆 ⇌ 푶푯 + ퟑ푶푯 + 풆−
∗
−
−
∗
−
ퟐ
(catalyst: Vulcan carbon) to enhance conductivity. Then the mixed
푶푯 + ퟑ푶푯 + 풆− ⇌ ퟒ푶푯−
∗
−
reagents (weighted at 5.0 mg) were ultrasonically dispersed in 1.5 mL
isopropanol /deionized water (2:1 v/v) with 15uL 5% Nafion solution as
solvent. The catalyst inks were deposited on a glassy carbon disk (5 mm
in diameter) as a working electrode. The total mass loadings for each
The computed Gibbs free energy (G) was obtained by the following Eq:
-
2
electrode was 200 ug/cm . Electrochemical measurements were carried
out on a CHI 650D electrochemical workstation. A common three-
electrode cell was used for the measurements, where the counter and
reference electrodes were a platinum wire and a standard calomel
electrode (SCE) respectively. The oxygen continued to flow into the 0.1
푮 = 푬(푫푭푻) + 푬(풁푷푬) – 푻푺
(6)
in which the E(DFT) is ground-state energy, E(ZPF) is zero-point energy,
T represents temperature and S is entropy corrections.
mol/L KOH electrolyte for ensuring O
from the acquisition of background current that needs N
2
saturation during overall tests, apart
saturation. Before
2
recording electrochemical data, the electrodes were kept for 100 s at the
set potential to establish a stable current response. CV measurements
were performed at a scan rate of 50 mV/s at a potential between 0.2 V and
Acknowledgements
1
.2 V (vs. RHE), and a series of polarization curves (LSV) were performed
This research is based on the work supported by the National
Natural Science Foundation of China (51778229). We thank
beamline BL14W1 (Shanghai Synchrotron Radiation Facility) for
providing the beam time.
with rotating speeds from 400 to 1600 rpm. The final spectrum of LSV was
obtained by subtracting the background current from the actual current.
The electron transfer number (n) per oxygen molecule under different
voltage were calculated by K–L plots:
Keywords: transition metal oxides • crystal plane effect • oxygen
reduction oxidation • electrocatalysis
ퟏ
풊
ퟏ
ퟏ
ퟏ
ퟏ
=
+
=
+
풊풌 풊풅 풏푭푨풌푪 ퟎ. ퟐ풏푭푨푫ퟐ/ퟑ풗−ퟏ/ퟐ푪훚ퟏ/ퟐ
(
1)
[
1]
aH. W. Kim, M. B. Ross, N. Kornienko, L. Zhang, J. Guo,
P. Yang, B. D. McCloskey, Nat. Catal. 2018, 1, 282-290;
bG. Yu, X. Xie, L. Pan, Z. Bao, Y. Cui, Nano Energy 2013,
Where i, i
k
d
, and i are the measured current, kinetically limited current and
diffusion-limited current, respectively. n is the electron transfer number. F
is the Faraday constant. A is electrode area. k stands for the rate constant.
D is the diffusivity of oxygen, ν for the kinematic viscosity. C corresponds
to the solubility of oxygen in the electrolyte and ω is the rotation rate in
rpm.
2, 213-234; cZ. Zhang, Y. Liu, L. Ren, H. Zhang, Z.
Huang, X. Qi, X. Wei, J. Zhong, Electrochim. Acta 2016,
200, 142-151.
X. Ren, J. Zhou, X. Qi, Y. Liu, Z. Huang, Z. Li, Y. Ge, S. C.
Dhanabalan, J. S. Ponraj, S. Wang, J. Zhong, H. Zhang,
Adv. Energy Mater. 2017, 7, 1700396.
[2]
Characterization: the crystal structure was determined by X-ray
diffraction (XRD) patterns with D/max2550V X-ray diffractometer using a
Cu Kα radiation, at diffraction angle range of 10 to 80. Brunner-Emmet-
Teller (BET) surface area of all catalysts were carried out by Micromeritics
Tristar 3020 SIN 993. The SEM and TEM images were determined by
[
[
3]
4]
D. F. Abbott, S. Mukerjee, V. Petrykin, Z. Bastl, N. B.
Halck, J. Rossmeisl, P. Krtil, RSC Adv. 2015, 5, 1235-
1243.
aG. Cheng, S. Xie, B. Lan, X. Zheng, F. Ye, M. Sun, X. Lu,
L. Yu, J. Mater. Chem. A 2016, 4, 16462-16468; bY.
Liang, Y. Li, H. Wang, J. Zhou, J. Wang, T. Regier, H. Dai,
Nat. Mater. 2011, 10, 780-786.
M. Huang, F. Li, F. Dong, Y. X. Zhang, L. L. Zhang, J.
Mater. Chem. A 2015, 3, 21380-21423.
3.0/S-3400N scanning electron microscope and JEM-2100 transmission
electron microscope. The chemical composition and valence states of
elements was examined by X-ray photoelectron spectroscopy (XPS,
ESCALAB 250Xi) using a monochromatic Al Ka X-ray source. And binding
energies were standardized by exogenous C 1s. The X-ray absorption
[5]
6
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202002110
FULL PAPER
[
[
[
[
6]
7]
8]
9]
N. Xu, J. Liu, J. Qiao, H. Huang, X.-D. Zhou, J. Power
Sources 2020, 455, 227992.
Y. Hang, C. Zhang, X. Luo, Y. Xie, S. Xin, Y. Li, D. Zhang,
J. B. Goodenough, J. Power Sources 2018, 392, 15-22.
B. Xu, H. Lu, W. Cai, Y. Cao, Y. Deng, W. Yang,
Electrochim. Acta 2019, 305, 360-369.
Yu, H. Liu, M. Xiong, Appl. Catal. B-Environ. 2020, 269,
118771; cQ. Deng, J. Zhao, T. Wu, G. Chen, H. A.
Hansen, T. Vegge, J. Catal. 2019, 370, 378-384.
W. Sun, Z. Wang, Q. Wang, W. Zaman, L. Cao, X.-Q.
Gong, J. Yang, Strategies of Alloying Effect for Regulating
Pt-based H2-SCR Catalytic Activity, Vol. 54, 2018.
X. Li, N. Xu, H. Li, M. Wang, L. Zhang, J. Qiao, Green.
Energy Environ. 2017, 2, 316-328.
aM. C. Biesinger, B. P. Payne, A. P. Grosvenor, L. W. M.
Lau, A. R. Gerson, R. S. C. Smart, Appl. Surf. Sci. 2011,
257, 2717-2730; bW. Wei, W. Chen, D. G. Ivey, Chem.
Mater. 2008, 20, 1941-1947.
X. Gao, J. Liu, Y. Sun, X. Wang, Z. Geng, F. Shi, X.
Wang, W. Zhang, S. Feng, Y. Wang, K. Huang, Inorg.
Chem. Front. 2019, 6, 3295-3301.
S. Yoshida, T. Tanaka, T. Hanada, T. Hiraiwa, H. Kanai,
T. Funabiki, Catal. Lett. 1992, 12, 277-285.
[38]
E. Pargoletti, G. Cappelletti, A. Minguzzi, S. Rondinini, M.
Leoni, M. Marelli, A. Vertova, J. Power Sources 2016, 325,
[39]
[40]
116-128.
[
[
[
[
[
[
[
[
[
[
[
[
10]
11]
12]
13]
14]
15]
16]
17]
18]
19]
20]
21]
X. Yang, W. Peng, K. Fu, L. Mao, J. Jin, S. Yang, G. Li,
Electrochim. Acta 2020, 340, 135989.
Y. L. Cao, H. X. Yang, X. P. Ai, L. F. Xiao, J. Electroanal.
Chem. 2003, 557, 127-134.
P. Ganesan, M. Prabu, J. Sanetuntikul, S. Shanmugam,
ACS Catal. 2015, 5, 3625-3637.
B. Chen, Z. Jiang, L. Zhou, B. Deng, Z. Jiang, J. Huang,
M. Liu, J. Power Sources 2018, 389, 178-187.
J. Liu, J. Jiang, C. Cheng, H. Li, J. Zhang, H. Gong, H. J.
Fan, Adv. Mater. 2011, 23, 2076-2081.
[41]
[42]
[43]
[44]
Y. Zhang, F. Ding, C. Deng, S. Zhen, X. Li, Y. Xue, Y.
Yan, K. Sun, Catal. Commun. 2015, 67, 78-82.
L. Zhang, Y. Zhao, M. N. Banis, K. Adair, Z. Song, L.
Yang, M. Markiewicz, J. Li, S. Wang, R. Li, S. Ye, X. Sun,
Nano Energy 2019, 60, 111-118.
X. Xie, Y. Li, Z.-Q. Liu, M. Haruta, W. Shen, Nature 2009,
458, 746-749.
H.-Y. Wang, S.-F. Hung, H.-Y. Chen, T.-S. Chan, H. M.
Chen, B. Liu, J. Am. Chem. Soc. 2016, 138, 36-39.
W. Han, C. Zang, Z. Huang, H. Zhang, L. Ren, X. Qi, J.
Zhong, Int. J. Hydrogen Energy 2014, 39, 19502-19512.
G. Du, X. Liu, Y. Zong, T. S. A. Hor, A. Yu, Z. Liu,
Nanoscale 2013, 5, 4657-4661.
G. Kim, H. Sun, A. Manthiram, Nano Energy 2016, 30,
1
30-137.
Q. Wang, B. Yu, X. Li, L. Xing, X. Xue, J. Mater. Chem. A
016, 4, 425-433.
2
aM. Lehtimäki, H. Hoffmannová, O. Boytsova, Z. Bastl, M.
Busch, N. B. Halck, J. Rossmeisl, P. Krtil, Electrochim.
Acta 2016, 191, 452-461; bF. Cheng, Y. Su, J. Liang, Z.
Tao, J. Chen, Chem. Mater. 2010, 22, 898-905.
[
[
[
22]
23]
24]
H. W. Nesbitt, D. Banerjee, Am. Mineral. 1998, 83, 305-
315.
S. Sun, Y. Xue, Q. Wang, H. Huang, H. Miao, Z. Liu,
Electrochim. Acta 2018, 263, 544-554.
aM. Zhi, A. Manivannan, F. Meng, N. Wu, J. Power
Sources 2012, 208, 345-353; bJ. Wang, Y. Yang, Z.
Huang, F. Kang, Electrochim. Acta 2011, 56, 9240-9247;
cL. Bao, J. Zang, X. Li, Nano Lett. 2011, 11, 1215-1220;
dT. N. Lambert, J. A. Vigil, S. E. White, C. J. Delker, D. J.
Davis, M. Kelly, M. T. Brumbach, M. A. Rodriguez, B. S.
Swartzentruber, J. Phys. Chem. C 2017, 121, 2789-2797.
L. S. Kau, D. J. Spira-Solomon, J. E. Penner-Hahn, K. O.
Hodgson, E. I. Solomon, J. Am. Chem. Soc. 1987, 109,
[
25]
6433-6442.
[
[
[
26]
27]
28]
D. Shevchenko, M. F. Anderlund, S. Styring, H. Dau, I.
Zaharieva, A. Thapper, PCCP 2014, 16, 11965-11975.
W. Sun, Z. Zhou, W. Q. Zaman, L.-m. Cao, J. Yang, ACS
Appl. Mater. Interfaces 2017, 9, 41855-41862.
W. Zaman, W. Sun, M. Tariq, Z. Zhou, U. Farooq, Z.
Abbas, L. Cao, J. Yang, Appl. Catal. B-Environ. 2019, 244,
2
95-302.
W. Sun, Y. Song, X. Gong, L.-m. Cao, J. Yang, Chem. Sci.
015, 6, 4993-4999.
[
[
[
[
[
[
[
[
[
29]
30]
31]
32]
33]
34]
35]
36]
37]
2
Q. Wang, D.-A. Zhang, Q. Wang, J. Sun, L.-L. Xing, X.-Y.
Xue, Electrochim. Acta 2014, 146, 411-418.
J. Xiao, Q. Kuang, S. Yang, F. Xiao, S. Wang, L. Guo, Sci
Rep 2013, 3, 2300.
L. Liu, Z. Jiang, L. Fang, H. Xu, H. Zhang, X. Gu, Y.
Wang, ACS Appl. Mater. Interfaces 2017, 9, 27736-27744.
J. Wang, R. Ma, Y. Zhou, Q. Liu, J. Mater. Chem. A 2015,
3
, 12836-12844.
X. Qiao, J. Jin, H. Fan, Y. Li, S. Liao, J. Mater. Chem. A
017, 5, 12354-12360.
2
C. Si, Y. Zhang, C. Zhang, H. Gao, W. Ma, L. Lv, Z.
Zhang, Electrochim. Acta 2017, 245, 829-838.
G. Chai, K. Qiu, M. Qiao, M. Titirici, C. Shang, Z. Guo,
Energy Environ. Sci. 2017, 10, 1186-1195.
aD. U. Lee, J. Li, M. G. Park, M. H. Seo, W. Ahn, I.
Stadelmann, L. RicardezSandoval, Z. Chen,
ChemSusChem 2017, 10, 2258-2266; bC. Li, E. Zhou, Z.
7
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202002110
FULL PAPER
Entry for the Table of Contents
The development of alternative low-cost ORR catalysts is urgently needed. This work describes the electronic structure of cobalt oxides
during its growth on MnO in detail for exploring the roles of each components in binary metal oxides on ORR. Emphasis is given to
the optimization in coordination environment of Co sites.
2
8
This article is protected by copyright. All rights reserved.