Surface Reorganization on Electrochemically-Induced Zn–Ni–Co Spinel Oxides for Enhanced Oxygen Electrocatalysis

Article abstract of DOI:10.1002/anie.202000690

Herein, we highlight redox-inert Zn²⁺ in spinel-type oxide (Zn_XNi_1?XCo₂O₄) to synergistically optimize physical pore structure and increase the formation of active species on the catalyst surface. The presence of Zn²⁺ segregation has been identified experimentally and theoretically under oxygen-evolving condition, the newly formed V_Zn?O?Co allows more suitable binding interaction between the active center Co and the oxygenated species, resulting in superior ORR performance. Moreover, a liquid flow Zn–air battery is constituted employing the structurally optimized Zn_0.4Ni_0.6Co₂O₄ nanoparticles supported on N-doped carbon nanotube (ZNCO/NCNTs) as an efficient air cathode, which presents remarkable power density (109.1 mW cm^?2), high open circuit potential (1.48 V vs. Zn), excellent durability, and high-rate performance. This finding could elucidate the experimentally observed enhancement in the ORR activity of Zn_XNi_1?XCo₂O₄ oxides after the OER test.

Full text of DOI:10.1002/anie.202000690

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Surface Reorganization on Electrochemically-Induced Zn-Ni-Co
Spinel Oxides for Enhanced Oxygen Electrocatalysis
Authors: Zhao-Qing Liu, Xiao-Tong Wang, Ting ouyang, Ling Wang,
and Jia-Huan Zhong
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202000690
Angew. Chem. 10.1002/ange.202000690
Link to VoR: http://dx.doi.org/10.1002/anie.202000690
http://dx.doi.org/10.1002/ange.202000690

10.1002/anie.202000690
Angewandte Chemie International Edition
RESEARCH ARTICLE
Surface Reorganization on Electrochemically-Induced Zn-Ni-Co
Spinel Oxides for Enhanced Oxygen Electrocatalysis
Xiao-Tong Wang, Ting Ouyang, Ling Wang, Jia-Huan Zhong and Zhao-Qing Liu*
promise as an electrode for energy equipment, owing to its earth
abundance, excellent stability, and rich redox reactions.[3]
However, the performance of NiCo2O4 without further tuning is
still falling short of the level required for the actual demand,
thereby inspiring us to find more effective strategies for structure
optimization.
Abstract: Comprehensive investigation of inter-functional
coordination between OER and ORR takes on the important action
of designing highly active and durable bifunctional oxygen catalysts.
Here, we highlight redox-inert Zn2+ in spinel-type oxide
(ZnXNi1−XCo2O4) to synergistically optimize physical pore structure
and increase the formation of active species on the surface of
catalyst. The presence of Zn2+ segregation has been identified
experimentally and theoretically under oxygen-evolving condition,
the newly formed VZn−O−Co allows more suitable binding interaction
between the active center Co and the oxygenated species, resulting
in superior ORR performance. Moreover, a liquid flow Zn-air battery
is constituted employing the structurally optimized Zn0.4Ni0.6Co2O4
nanoparticles supported on N-doped carbon nanotube
(ZNCO/NCNTs) as an efficient air cathode, which presents
remarkable power density (109.1 mW cm−2), high open circuit
Although spinel materials have been widely studied due to
their excellent catalytic activity for OER and ORR, the
aggregation or segregation of cations near the surface and
interface of catalysts have been frequently captured during the
reaction, which has factored heavily into the electrode dynamics
and durability.[4] A profound understanding of the correlation
between the surface nanostructure and the electrocatalytic
performance is then accentuated, which is vital to achieve the
rational design of more efficient electrode materials with
enhanced stability. Much can be learned from previously
reported literature, as well as efforts to control cation segregation
for high electrocatalytic activity and durability. For cobalt-based
spinel, during OER operation, the irreversible transformation of
Co2+ oxide to Co3+Oh-containing CoOx(OH)y is accompanied by
the weakening of crystallinity. The change of valence state of two
or more redox active cobalt atoms, from the valence of +2 to +4
and then reduced, is considered to be the active site in OER,
which are connected by μ-OH or μ-O bridges.[5] Wang et al.
reported that Co2+ in tetrahedral sites can be easily oxidized to
form CoOx(OH)y, which then acts as the active phase, while the
Co3+ in octahedral sites is rather inactive toward the OER as it
might be blocked by OH-groups.[6] For ORR, CoOx(OH)y may not
be considered as an active phase. A general principle is
concluded that the Co3+ octahedral site of spinel can be directly
recognized as an ORR performance descriptor. Wei et al.
proposed that the eg occupancy of the active cation in the
octahedral site is closely related to the oxygen catalyst
performance.[7] Besides, partial substitution of redox-inert Zn2+
into the thermodynamically preferred tetrahedral coordination,
rendering structural support and providing cooperative effects for
the catalytically active cobalt.[8] However, during the reaction, the
dissolution/leaching of zinc remains a matter of concern. Wahl et
al. reported that Zn0.35Co0.65O can transform in situ from the
wurtzite structure to the layered γ-Co(O)OH during the OER
process, while most of its zinc leaches out. Coordinately
unsaturated cobalt octahedral on the surface is indeed the active
center for the OER.[9] Menezes et al. examined cobalt-containing
spinel oxides toward OER and showed that partly leaching of
Zn2+ from the tetrahedral sites leads to defective sites and thus
increases the activity.[10]
potential (1.48
V vs. Zn), excellent durability and high-rate
performance. This finding could elucidate the experimentally
observed enhancement in the ORR activity of ZnXNi1−XCo2O4 oxides
after the OER test. The atomic-level insight into the new catalyst
activation strategy based on Zn vacancies is adaptable for
developing dual-functional Zn-contained spinel electrocatalysts.
Introduction
Oxygen reduction reaction (ORR) and oxygen evolution reaction
(OER) are the key processes for various electrochemical energy
conversion devices (e.g., fuel cells and metal-air batteries).
However, the sluggish multiple electron transfer process of ORR
and OER presents a significant obstacle to the practical
application of these energy technologies, making oxygen
electrocatalysts essential for these energy conversion
systems.[1] Spinel-type (AXBYOZ) complex oxides with multiple
cations can show a variety of chemical ordering between cations
at the A or B sites, thus exhibiting various electrocatalytic
properties.[2] In particular, NiCo2O4 has shown remarkable
[a]
X.-T. Wang, Dr. T. Ouyang, L. Wang, J.-H. Zhong, Prof. Z.-Q. Liu
School of Chemistry and Chemical Engineering/Guangzhou Key
Laboratory for Clean Energy and Materials/Key Laboratory for
Water Quality and Conservation of the Pearl River Delta, Ministry
of Education, Guangzhou University.
No. 230 Wai Huan Xi Road, 510006 (P. R. China)
E-mail: lzqgzu@gzhu.edu.cn
Supporting information for this article is given via a link at the end
of the document.
As an essential principle of function coordination for the
durability of the bifunctional catalyst, the overall regulation plays
This article is protected by copyright. All rights reserved.

10.1002/anie.202000690
Angewandte Chemie International Edition
RESEARCH ARTICLE
an important role in the process of electrochemical polarization.
By summarizing the previous properly elaborated works, we got
inspired to craft a material that combines the two identified
driving forces. First, NiCo2O4 is chosen as an optimization object,
given its excellent structural stability and intrinsic activity.
Moreover, the material provides a part of tetrahedral coordinated
Zn2+, in situ generation of defective species during
electrochemical reactions is expected to produce a special
catalytic effect. Second, the unsatisfactory electrocatalytic
performance of spinel, which is subject to its intrinsically low
electrical conductivity, coupling with carbon nanotubes (CNTs)
can greatly alleviate this effect.[11] Thus we start from tuning the
structure of ZnXNi1−XCo2O4 cubic spinel anchored on N-doped
CNTs by varying the amount of Zn substitution (X = 0, 0.2, 0.4,
0.6, 0.8, 1). When X = 0.4 (Zn0.4Ni0.6Co2O4/NCNTs), the Zn/Ni
composition was optimal to yield high activities for both OER and
ORR. Especially, we propose the presence of potential dual-
reinforcement mechanism on Zn0.4Ni0.6Co2O4/NCNTs and obtain
activity trends at experimentally relevant electrochemical
conditions. Firstly, the morphology of materials has a great
impact on their catalytic activities. A desirable structure as
electrocatalyst should have abundant active sites and allow for
full electrolyte penetration as well as rapid mass transport. Zn2+
substitution can effectively optimize the physical structure of
oxides. The introduction of a large number of mesopores affords
Zn0.4Ni0.6Co2O4/NCNTs with large specific surface area and
abundant structured edge, leading to improved ORR/OER
activities. Then, using aberration-corrected transmission
electron microscopy (Ac-TEM) and corresponding spectrum
(XPS, XRD), cation segregation can be observed on the spinel
after the reaction. Specifically, plentiful Zn cation defects are
randomly distributed on the catalyst surface. The new electronic
framework VZn–O–Co shows better catalytic performance, which
can be further confirmed by alternatively-test and density
functional theory (DFT) calculation.
Figure 1. Schematic illustration of the fabrication of ZnXNi1−XCo2O4/NCNTs
hybrid in a one-stage synthesis approach.
ZnXNi1−XCo2O4/NCNTs catalysts were characterized by field-
emission scanning electron microscopy (FESEM). No large
particles were observed on the surface of NCNTs (Figure 2a and
S2). Nanoparticle spatial distributions might have a large
significance for catalyst stability. Figure 2b shows a typical low-
magnification transmission electron microscopy (TEM) image of
Zn0.4Ni0.6Co2O4/NCNTs, in which the nanoparticles on NCNTs
are composed of small particles (6−8 nm). High-resolution
transmission electron microscopy (HRTEM) images are
displayed in Figure 2c−d. The characteristic spacing of 0.28 and
0.23 nm correspond to the (220) and (222) lattice planes of face
center cubic (FCC) spinel. Figure 2e exhibits the high-angle
annular dark-field scanning TEM (HAAD-STEM) image and
corresponding energy dispersive X-ray (EDX) images of
Zn0.4Ni0.6Co2O4/NCNTs, which reveals the distinct distribution of
Zn, Ni, Co, C, N, and O elements.
Results and Discussion
We synthesized a new ZnXNi1−XCo2O4/NCNTs composite
via a simple one-step hydrothermal process (Figure 1). A series
of Zn-Ni-Co spinel oxides with uniform size were grown on N-
doped carbon nanotubes (NCNTs) by using microwave
hydrothermal reactions at 130 ºC for 3 h. The X-Ray diffraction
(XRD) patterns of as-prepared ZnXNi1−XCo2O4samples agree
well with the standard cubic spinel oxides (Figure S1a), The XRD
peak at about 27° can be attributed to the d-spacing in the
laminated CNTs stacks.[12] Surface-enhanced Raman scattering
spectroscopy of ZnXNi1−XCo2O4/NCNTs (X = 0, 0.4, 1) indicates
that ternary spinel still has the characteristic peak of spinel oxide
(Figure S1b). This result is in accordance with XRD data.
Moreover, surface defects of CNTs are at the same level, which
is determined by the intensity ratio between the D band and G
band (ID/IG).[13] The morphology of as-synthesized
The Ni 2p X-ray photoelectron spectrum (XPS) of
Zn0.4Ni0.6Co2O4/NCNTs was fitted with spin-orbit peaks,
corresponding to the Ni2+ at about 856.5 and 874.0 eV (Figure
S3a), demonstrating the +2 oxidation state of Ni species.[14] The
Zn 2p spectrum of Zn0.4Ni0.6Co2O4/NCNTs was also fitted with
spin-orbit peaks, which were assigned to characteristic peaks of
Zn2+ at about 1021.2 and 1044.2 eV (Figure S3b). Accordingly,
the Zn species was found to be mostly in the +2 oxidation
state.[15] Co 2p XPS spectrum of Zn0.4Ni0.6Co2O4/NCNTs was
fitted with one spin-orbit peak, indicating the existence of Co3+ at
about 796.2 and 780.7 eV (Figure S3c).[16] The N 1s XPS
spectrum of Zn0.4Ni0.6Co2O4/NCNTs shows the presence of
metal-N bonds and Pyrrole-N (Figure S3e). According to
quantitative analysis, ~4 at. % N was doped into CNTs. Nitrogen
This article is protected by copyright. All rights reserved.

10.1002/anie.202000690
Angewandte Chemie International Edition
RESEARCH ARTICLE
Zn0.4Ni0.6Co2O4/NCNTs is greater than that of Pt/C & RuO2
(Figure 3b). Zn-cation enables the conventional cobalt-based
spinel to rival the Pt/C & RuO2 in the fast discharge area, which
is considered as a merit for practical application. Faradaic
efficiency is a common metric for electrocatalytic systems. Under
work voltage, the decomposition of carbon materials is an
important problem for the stability of catalyst. To further assess
this information, RRDE test was carried out in N2-saturated
alkaline solution at 1600 rpm, during the electrochemical
measurement (Edisk = Ej = 10: 1.64 V, Ering: 0.4 V), the O2 bubbles
on the disk electrode were continuously generated and thrown
to the ring electrode for ORR (Figure 3c). Faradaic efficiency of
Zn0.4Ni0.6Co2O4/NCNTs is calculated to be around 94−97%. This
result arises from its low overpotential and electrochemical
stability of the crystal. The rational design of Zn-cation
substitution is immediately verified by the enhanced ORR
performance. Linear sweep voltammetry (LSV) curves were
carried out to obtain the polarization curves and Tafel slopes of
each catalyst toward ORR on a rotating disc electrode (RDE).
During the polarization tests, the RDE was kept rotating at 1600
rpm to acquire the dynamic diffusion current in O2-saturated 0.1
M
KOH. In Figure 3d−e, Zn0.4Ni0.6Co2O4/NCNTs displays
excellent electrochemical activity with an onset potential of 0.86
V, a half-wave potential (E1/2) of 0.78 V and Tafel slope of 78.0 ±
0.5 mV dec−1 together with a high current density of 4.6 mA cm−2
at working potential of 0.4 V, all of which surpass those of the
other five samples. As compared with Pt/C&RuO2, E1/2 of
Zn0.4Ni0.6Co2O4/NCNTs shifts negatively by only 50 mV, and the
limiting current density is comparable to that of the Pt/C & RuO2
catalyst. Generation of HO2− and electron transfer number (n) in
the ORR were surveyed by RRDE test (Figure S4). In this
dynamic process, around 10 % of hydrophilic peroxide yield is
detected for Zn0.4Ni0.6Co2O4/NCNTs. After conversion, it
confirmed that Zn0.4Ni0.6Co2O4/NCNTs drives the ORR by a 4-
electron transfer pathway. The ΔE value, as a crucial factor of
energy efficiency for Zn-air batteries, is defined as the difference
between the Ej = 10 and E1/2 in the ORR. In general, the
Zn0.4Ni0.6Co2O4/NCNTs have an excellent bifunctional activity for
both OER and ORR in the alkaline electrolyte (Table S1).
Figure 2. a) SEM image, b−d) TEM image, (Inset: FFT pattern of the
corresponding plane), e) HAADF-STEM image and corresponding EDX
mappings of Zn0.4Ni0.6Co2O4/NCNTs.
content of three samples are recognized to be at the same level.
The same phenomenon was observed in O and C XPS spectra
(Figure S3d−f).
The electrocatalytic activities of the as-obtained
ZnXNi1−XCo2O4/NCNTs catalysts for OER were investigated in
0.1
M KOH solution using a standard three-electrode
In order to understand the basic reasons for enhanced
electrochemical performance, double-layer capacitance (Cdl)
tests were performed. Zn0.4Ni0.6Co2O4/NCNTs exhibited the
highest Cdl (8.02 mF cm−2) than the other five samples (Figure
3e and S5). As depicted in Figure 3f, N2 adsorption/desorption
measurements were conducted to examine the porous nature of
Zn0.4Ni0.6Co2O4/NCNTs, NiCo2O4/NCNTs and ZnCo2O4/NCNTs.
The Brunauer-Emmett-Teller (BET) specific surface area of
Zn0.4Ni0.6Co2O4/NCNTs was calculated to be about 123.2 m2 g−1,
which is larger than NiCo2O4/NCNTs (85.1 m2 g−1) and
ZnCo2O4/NCNTs (103.6 m2 g−1). Besides, the pore size
distribution of Zn0.4Ni0.6Co2O4/NCNTs indicates that it affords a
number of pores in size of 3−15 nm, while the other samples
have plenty of mesopores in 5–10 nm (Figure 3g). BET surface
area and pore size distribution strongly support the fact that Zn
electrochemical cell. The precious metal catalyst consists of Pt/C
and RuO2 (Pt/C & RuO2, the mass ratio of Pt/C and RuO2 are
1:1). All the catalysts respond aggressively toward OER (Figure
3a). Operating potential taken at 10 mA cm−2 (Ej = 10) was as a
standard to compare the performance differences. In particular,
Zn0.4Ni0.6Co2O4/NCNTs exhibits much lower work potential (1.64
V vs. RHE) than its counterparts at identical current densities.
The
significantly
enhanced
OER
kinetics
of
Zn0.4Ni0.6Co2O4/NCNTs is also revealed by the fact that its Tafel
slope is smaller than those of ZnCo2O4/NCNTs, NiCo2O4/NCNTs
and ternary spinel contrast samples with different proportions.
Zn0.4Ni0.6Co2O4/NCNTs (118.3 ± 3 mV dec−1) demonstrates a
higher Tafel slope than Pt/C & RuO2 (108.2 ± 2 mV dec−1) at low
current density region (0.3–5 mA cm−2). In the high potential
region,
the
current
enhancement
trend
of
This article is protected by copyright. All rights reserved.

10.1002/anie.202000690
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 3. OER performance of various ZnXNi1−XCo2O4/NCNTs hybrids: a) LSV scan and b) corresponding parameters, c) Faraday efficiency test. ORR performance
of samples: d) LSV scan and e) corresponding parameters. f) Individual ORR and OER, and total overpotentials for samples. g) Cdl of samples. h) N2-
adsorption/desorption isotherms of samples and i) corresponding pore size distributions.
substitution can effectively control the specific surface area and
pore distribution, which is in good accordance with Cdl results.
Electrochemical stability is one of the key parameters used
to determine the evaluation of practicality. Unfortunately, it is well
established that RuO2 catalyst is dynamically unstable and can
be oxidized to RuO4 or suffer from agglomeration behavior at
high operating potentials, which is subsequently dissolved in the
alkaline solution.[17] However, for transition metal oxides, cation
segregation is relatively common at work potential, which is a
miscellaneous process that is impacted by combined sections,
including the nature of the materials and environmental
conditions.[18] In this work, we aim to regulate the segregation for
catalysis, so the subsequent discussion will focus on the status
of samples after the electrocatalytic process.
The robustness of the as-prepared Zn0.4Ni0.6Co2O4/NCNTs
(ZNCO/NCNTs) was finally alternatively-tested under varied
gradation of OER work potential (1.64 V) and ORR potential (0.4
V) in O2- saturated 0.1 M KOH at 1600 rpm, the benign reaction
conditions were chosen to slow down the process and make the
changes more gradual. Combining the change of the LSV curve
and the data of inductively coupled plasma mass spectrometry
(IPC-MS), we can draw the conclusion that progressive
enhancement of ORR performance is directly related to the
changes produced by the OER process (Figure S6a–h).
This article is protected by copyright. All rights reserved.

10.1002/anie.202000690
Angewandte Chemie International Edition
RESEARCH ARTICLE
Figure 4. a–b) HAADF images of cycled ZNCO/NCNTs and corresponding lattice planes. The vacancy formation energy of c) VZn and d) VNi, e) schematic diagram
of PDOS. The geometry structures and charge density difference mappings of oxygenated species (*OOH, *O, *OH) on f) ZNCO and g) VZn-ZNCO. The sky blue
and yellow isosurfaces stand for the negative and positive charges. The red and blue balls represent O and Co atoms, respectively. The cyan balls refer to Zn
atoms, gray represents Ni atoms. h) Free energy diagrams for ZNCO and VZn-ZNCO.
To gain more insight, the functional mechanism of spinel
oxides were first uncovered in reference to the crystal
transformation and the surface electronic states from XRD and
the high-resolution XPS analysis, respectively. The XRD result of
ZNCO/NCNTs after stability test still keeps the original form,
implying that the reinforcement of electrochemical properties is
not caused by phase change (Figure S7a). The high-resolution
XPS analysis further confirms that the phenomenon of chemical
shift is not observed in the Zn 2p and Ni 2p spectra after testing
new interaction between the Co and other species. The inclusion
of new exchange mode into the system was accompanied by a
decrease in the shifted Co 2p3/2 peak maxima from 781.5 to 780.7
eV. Owing to the appearance of new Co-based reaction centers,
the electron delocalization around the Co atom can be effectively
adjusted.[19]
The structural characterization analysis was obtained by
aberration-corrected transmission electron microscopy (AC-TEM).
Surface oxides of cycled ZNCO/NCNTs still maintain a dispersed
(Figure S7b–d). In order to exclude the non-electrochemical factor, particle state. The HRTEM images (Figure S8a–d) show areas of
we also tested ZNCO/NCNTs of XPS after 3 h of soaking in 0.1M
KOH. There is no offset in Zn, Ni, Co spectra compared to the
original state, indicating the catalyst has good stability in alkaline
solution. However, there is a shift in the Co 2p spectrum of cycled
ZNCO/NCNTs compared with that of the original state, proving a
lattice disorder and crystallization. Abundant vacancies sites with
randomly scattered distribution can be observed on the surface,
revealing that the defective regions (pointed out by the red circle)
mainly exist on the surface of cycled ZNCO structure (V-ZNCO).
The obvious contrast between the vacancies sites and bulk atoms
This article is protected by copyright. All rights reserved.

10.1002/anie.202000690
Angewandte Chemie International Edition
RESEARCH ARTICLE
can be attributed to the ingredients in the (220) and (222)
diffraction direction (Figure 4a and b). After electrochemical cycle
testing, cation defects and lattice disorder take place at ZNCO
with relatively higher defect concentrations. The conclusion can
be reached as the following aspects: i) cationic defects tend to
occur at edge structure, and ii) there is a higher probability of
vacancy at the Zn/Ni tetrahedral site. While for the V-ZNCO,
judging species of vacancies just based on HRTEM observation
might not be precise, because Zn2+ and Ni2+ have similar contrast.
To solve this problem, DFT calculations were carried out. The
vacancy formation energy of Zn2+ (1.24 eV) is significantly lower
than Ni2+ (5.28 eV), indicating that it is easier to form Zn vacancy
we have determined that the research object is ZNCO with a Zn
vacancy (VZn-ZNCO).
In order to investigate the intrinsic electronic structure of the
catalyst, the calculated electronic density of states (DOS)
calculations for ZNCO and VZn-ZNCO were conducted to give
more accurate determination. The shrinking energy difference
between the TM 3d and O 2p-band centers indicates an increased
metal-oxygen covalency. The Co 3d and O 2p centers are
determined for ZNCO and VZn-ZNCO by integrating the partial
electronic density of states (PDOS). Compared with ZNCO, the
Co 3d and O 2p centers get closer in VZn-ZNCO (Figure S9a–d),
suggesting that the VZn-ZNCO oxide exhibits greater Co–O
(VZn) on the basis of thermodynamics (Figure 4c and d). In general, covalency than ZNCO (Figure 4e). The enlarged Co–O covalency
Figure 5. a) Schematic of the basic configuration of a liquid flow Zn–air battery. b) Polarization and power density curves of primary Zn–air batteries. c) The high-
rate charging capability test of ZNCO/NCNTs. Discharging, 5 mA cm−2, 600 s. Ⅰ) Charging, 10 mA cm−2, 300 s. Ⅱ) Charging, 15 mA cm−2, 200 s. Ⅲ) Charging, 20
mA cm−2, 150 s. Ⅳ) charging, 25 mA cm−2, 120 s. d) The high-rate discharging capability test of ZNCO/NCNTs. Charging, 5 mA cm−2, 600 s. Ⅴ) Discharging, 10
mA cm−2, 300 s. Ⅵ) Discharging, 15 mA cm−2, 200 s. Ⅶ) Discharging, 20 mA cm−2, 150 s. Ⅷ) Discharging, 25 mA cm−2, 120 s. e) Long-time durability of samples
at a current density 25 mA cm−2
.
This article is protected by copyright. All rights reserved.

10.1002/anie.202000690
Angewandte Chemie International Edition
RESEARCH ARTICLE
can promote the electron transfer between the TM cation and
(Figure S12b). At 40 mA cm−2, the ZNCO/NCNTs achieves a high
energy density of 701.1 Wh kg−1 (about 64.6 % of the theoretical
energy density). Pt/C & RuO2 exhibits an energy density of 686.4
Wh g−1. Discharging–charging tests were carried out. A high-rate
capability of the Zn–air battery with this ZNCO/NCNTs/carbon
cloth electrode was evaluated. To stabilize the high-rate cycling,
the battery was firstly standing to allow for the catalyst to fully
contact with the electrolyte and reach equilibrium. After that, it was
tested at the charging current densities of 10, 15, 20, and 25 mA
cm−2, respectively, and maintained the discharging current
density of 5 mA cm−2 in Figure 5c. The charging voltages
gradually increase with increasing current density, while the
discharging voltages are almost maintained during each cycle.
The high-rate discharging capability was further tested through
charging at the same current density of 5 mA cm−2 but discharging
at 10, 15, 20, and 25 mA cm−2, respectively. As shown in Figure
5d, the discharging voltages decreased with increasing current
density, while even at high current densities, the voltage plateau
still kept at the same level.
Hence, in the fast-charging mode, both the discharging voltage
and the capacity of the battery system can be maintained, but the
charging time is greatly reduced; in the fast-discharging mode, a
steady high voltage coming from the stable ORR reaction and
liquid flow system can also be delivered. In addition, long time
charge-discharge performance at 25 mA cm−2 was further
investigated (Figure 5e). Initially, the low oxygen concentration
and unstable solid-liquid interface result in the unsustainable
discharge/charge voltage plateau. With increasing the cycle
number, the gradual transition from ZNCO/NCNTs to VZn-
ZNCO/NCNTs and the system became more stable. The round-
trip overpotentials change from 0.8 V at the 10th cycle to 0.87 V
at the 100th cycle, corresponding to a 3.4 % decrease of voltaic
efficiency (discharge end voltage divided by charge end voltage).
Compared with Pt/C & RuO2, the round-trip overpotentials change
from 0.65 V at the 10th cycle to 1.05 V at the 50th cycle,
corresponding to a 17.3 % decrease of voltaic efficiency.
Consequently, the ZNCO/NCNTs electrodes exhibit remarkable
stability for Zn–air batteries. Moreover, a mechanical fan (Vf: 1.3–
1.6 V) can be run by the assembled battery (Figure S13).
oxygen adsorbates (e.g., *O2 and *O2−) or facilitate the
2−
injection/extraction of electrons from oxygen, thereby accelerating
the oxygen catalytic rate.[20]
We employed DFT calculations to estimate the reaction
mechanism of the ORR. The ORR proceeds through four
consecutive proton/electron transfer steps via the
4H+ + 4e− + 2O2 (g) → 2H2O (l)
Intermediates OOH*, O*, OH*. Here, the * indicates adsorbed
intermediates. In O* and OOH* configuration (Figure 4f and g),
the electrons (sky blue region) can be effectively aggregated
between the Co−O bond or O−O bond, which contributes to the
activation of a transition state.[21] The Co atom of VZn-ZNCO
bound to O was without an external potential at the first step: O2
(g) + ½ H2 → OOH* decline in free energy by −1.96 eV (ZNCO:
−2.11). The following reduction of OOH* to O* (while releasing an
H2O molecule) is exothermic by −4.37 eV (ZNCO: −4.67). The
third step: O* + ½ H2 → OH* is downhill by –5.05 eV (ZNCO:
−5.34), reaching a minimum of −4.92 eV relative to gas-phase O2,
lower than the overall free energy of the ORR (Figure 4h).[22]
Adsorbed OH* on modes thus constitutes a thermodynamic sink
that would prevent the reaction from continuing, indicating that the
fourth electron transfer reaction is the rate-limiting step. The G4
of VZn-ZCNO (0.13 eV) is significantly lower than ZNCO (0.42 eV),
implying that as for ZNCO, the OH* species is bound too strongly
for the ORR to proceed. Moreover, OER is the reverse reaction of
ORR, the rate-determining step (RDS) barrier of VZn-ZNCO (G3:
2.41 eV) is lower than ZNCO (G3: 2.56 eV), which can be largely
ascribed to the improved activity of Co atom next to VZn (Figure
S10).
Considering the superior bifunctional performance of
ZNCO/NCNTs, a homemade liquid flow Zn-air battery was built to
demonstrate its feasibility in practical energy devices. The battery
consists of carbon paper-supported ZNCO/NCNTs as the air
cathode, Zn plate as the anode, and 6.0 M KOH + 0.2 ZnCl2 +
M
10% Na-citrate as the electrolyte. The circulation system of the
battery is also displayed (Figure 5a and S11). Liquid flow battery
system was adopted for the following reasons: i) The presence of
Zn2+ segregation has been identified experimentally and
theoretically under oxygen-evolving conditions. A liquid flow
battery system was proposed to reduce the negative effect of
cation segregation on the electrolyte. ii) The object of this work
mainly aims at the air cathode. Accumulation of much gas like O2,
H2 on catalyst surface during the working period of metal-batteries,
making the cell unstable in charge and discharge, which restricts
the real stability of air cathode. Such effects can be reduced in a
liquid flow battery system by flushing away the bubbles in time. iii)
Due to the half-open system, carbon dioxide will enter the alkaline
electrolyte, decreasing the ionic conductivity. A closed system
can minimize the impact of this factor. As shown in the i – t test
(Figure S12a), the open circuit potential of Zn-air battery with
ZNCO/NCNTs cathode is 1.48 V, higher than that with Pt/C &
RuO2 (1.41 V). Chronopotentiometry with the current ramp is
shown in Figure 5b. The ZNCO/NCNTs driven Zn–air battery
delivers similar recharge-discharge behaviors to the high-
performance Pt/C & RuO2 catalyst, suggesting its excellent
bifunctional catalysis activity. The peak power density of
ZNCO/NCNTs was calculated to be 109.1 mW cm−2. The specific
capacity was evaluated according to the consumption of Zn
Conclusion
In conclusion, the outstanding electrocatalytic performance
of Zn0.4Ni0.6Co2O4/NCNTs (ZNCO/NCNTs) towards oxygen
reduction/evolution reaction has been validated. Moreover, the
liquid-flow battery system was employed to investigate its
performance for the actual application, in which ZNCO/NCNTs
exhibit high rate discharge/change stability and long cycle stability.
We propose the presence of a potential dual-reinforcement
mechanism in ZNCO/NCNTs and obtain activity trends at
experimentally relevant electrochemical conditions. Partly by
controlling of physical state, the introduction of Zn brings a
number of mesopores to the sample and greatly increases the
specific surface area, which exactly affords edge structure and
more reaction sites. Partly by cationic segregation, we can clearly
observe that abundant vacancies sites with randomly scattered
distribution over the surface after stability test, especially at the
This article is protected by copyright. All rights reserved.

10.1002/anie.202000690
Angewandte Chemie International Edition
RESEARCH ARTICLE
[12] J. Chen, P. Cui, G. Zhao, K. Rui, M. Lao, Y. Chen, X. Zheng, Y. Jiang, H.
Pan, S. X. Dou et al., Angew. Chem. Int. Ed. 2019, 58, 12540 –12544.
[13] Y. Li, C. Zhong, J. Liu, X. Zeng, S. Qu, X. Han, Y. Deng, W. Hu, J. Lu, Adv.
Mater. 2018, 30, 1703657.
edge, in other words, the formation of vacancies benefits from the
enriched porous structure. The combined theoretical and
experimental studies for the synthesized ZNCO/NCNTs confirm
that the Co sites activated by the VZn as newly formed catalytic
active centers account for the superior ORR and OER
bifunctionality. The new catalyst activation strategy presented in
this work is adaptable for the design and synthesis of new dual-
functional Zn-contained spinel electrocatalysts.
[14] J. Romero, H. Prima-Garcia, M. Varela, S. G. Miralles, V. Oestreicher, G.
Abellán, E. Coronado, Adv. Mater. 2019, 31, 1900189.
[15] L. Hu, J. Yuan, Y. Ren, Y. Wang, J.-Q. Yang, Y. Zhou, Y.-J. Zeng, S.-T.
Han, S. Ruan, Adv. Mater. 2018, 30, 1801232.
[16]a) X.-F. Lu, L.-F. Gu, J.-W. Wang, J.-X. Wu, P.-Q. Liao, G.-R. Li, Adv. Mater.
2017, 29, 1604437; b) D. Zhou, Z. Cai, X. Lei, W. Tian, Y. Bi, Y. Jia, N. Han,
T. Gao, Q. Zhang, Y. Kuang et al., Adv. Energy Mater. 2018, 8, 1701905.
[17] a) C. Guo, Y. Zheng, J. Ran, F. Xie, M. Jaroniec, S.-Z. Qiao, Angew. Chem.
Int. Ed. 2017, 56, 8539–8543; b) B. Hua, M. Li, W. Pang, W. Tang, S. Zhao,
Z. Jin, Y. Zeng, B. Shalchi Amirkhiz, J.-L. Luo, Chem 2018. 4. 2902-2916;
c) H. Wang, R. Liu, Y. Li, X. Lü, Q. Wang, S. Zhao, K. Yuan, Z. Cui, X. Li,
S. Xin et al., Joule 2018, 2, 337–348.
Acknowledgments
This work was financially supported by National Natural Science
Foundation of China (Grant No. 21875048), Guangdong Natural
Science Foundation (Grant No. 2017A030311016), Science and
Technology Research Project of Guangdong Province (Grant No.
2016A010103043), Major Scientific Project of Guangdong University
(Grant No. 2017KZDXM059), Yangcheng Scholars Research Project
of Guangzhou (Grant No. 201831820), Guangzhou University Qs
2017 Training Program for Young Top-Notch Personnel (BJ201704).
[18] a) B. Liu, Y. Wang, H.-Q. Peng, R. Yang, Z. Jiang, X. Zhou, C.-S. Lee, H.
Zhao, W. Zhang, Mater. 2018, e1803144; b) Y.-C. Zhang, N. Afzal, L. Pan,
X. Zhang, J.-J. Zou, Adv. Sci. 2019, 6, 1900053; c) Y. Zuo, D. Rao, S. Li,
T. Li, G. Zhu, S. Chen, L. Song, Y. Chai, H. Han, Adv. Mater. 2018, 30,
1704171.
[19] X. Hao, Y. Wang, J. Zhou, Z. Cui, Y. Wang, Z. Zou, Appl. Catal. B: Environ.
2018, 221, 302–311.
[20] a) M. Bajdich, M. García-Mota, A. Vojvodic, J. K. Nørskov, A. T. Bell, J. Am.
Chem. Soc. 2013, 135, 13521–13530; b) X.-K. Gu, E. Nikolla, ACS Catal.
2017, 7, 5912–5920; c) B.-Q. Li, Z.-J. Xia, B. Zhang, C. Tang, H.-F. Wang,
Q. Zhang, Nat. Commun. 2017, 8, 934; d) S. Sun, Y. Sun, Y. Zhou, S. Xi,
X. Ren, B. Huang, H. Liao, L. P. Wang, Y. Du, Z. J. Xu, Angew. Chem. Int.
Ed. 2019, 58, 6042–6047.
Keywords: oxygen evolution reaction • oxygen reduction
reaction • Zinc-air batteries • defect chemistry
[1] a) Z. P. Cano, D. Banham, S. Ye, A. Hintennach, J. Lu, M. Fowler, Z. Chen,
Nat Energy 2018, 3, 279–289; b) G. Fu, X. Yan, Y. Chen, L. Xu, D. Sun, J.-
M. Lee, Y. Tang, Adv. Mater. 2018, 30, 1704609; c) D. Ji, L. Fan, L. Li, S.
Peng, D. Yu, J. Song, S. Ramakrishna, S. Guo, Adv. Mater. 2019, 31,
1808267; d) Y.-J. Li, L. Cui, P.-F. Da, K.-W. Qiu, W.-J. Qin, W.-B. Hu, X.-
W. Du, K. Davey, T. Ling, S.-Z. Qiao, Adv. Mater. 2018, 30, 1804653; e) X.
Zhu, T. Jin, C. Tian, C. Lu, X. Liu, M. Zeng, X. Zhuang, S. Yang, L. He, H.
Liu et al., Adv. Mater. 2017, 29, 1704091.
[21] P. Zhou, X. Hou, Y. Chao, W. Yang, W. Zhang, Z. Mu, J. Lai, F. Lv, K. Yang,
Y. Liu et al., Chem. Sci. 2019, 10, 5898–5905.
[22] R. Koitz, J. K. Nørskov, F. Studt, Phys. Chem. Chem. Phys. 2015, 17,
12722–12727.
[2] a) C. Guan, A. Sumboja, H. Wu, W. Ren, X. Liu, H. Zhang, Z. Liu, C. Cheng,
S. J. Pennycook, J. Wang, Adv. Mater. 2017, 29, 1704117; b) X. Zhang, J.
Li, Y. Yang, S. Zhang, H. Zhu, X. Zhu, H. Xing, Y. Zhang, B. Huang, S. Guo
et al., Adv. Mater. 2018, 30, 1803551.
[3] a) X. Gao, H. Zhang, Q. Li, X. Yu, Z. Hong, X. Zhang, C. Liang, Z. Lin,
Angew. Chem. Int. Ed. 2016, 55, 6290–6294; b) F. Mo, G. Liang, Z. Huang,
H. Li, D. Wang, C. Zhi, Adv. Mater. 2019, 1902151.
[4] a) Y. Li, W. Zhang, Y. Zheng, J. Chen, B. Yu, Y. Chen, M. Liu, Chem. Soc.
Rev. 2017, 46, 6345–6378; b) D. Schweke, Y. Mordehovitz, M. Halabi, L.
Shelly, S. Hayun, Adv. Mater. 2018, 30, 1706300.
[5] a) A. Bergmann, T. E. Jones, E. Martinez Moreno, D. Teschner, P. Chernev,
M. Gliech, T. Reier, H. Dau, P. Strasser, Nat Catal 2018, 1, 711–719; b) M.
Risch, F. Ringleb, M. Kohlhoff, P. Bogdanoff, P. Chernev, I. Zaharieva, H.
Dau, Energy Environ. Sci. 2015, 8, 661–674.
[6] 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.
[7] C. Wei, Z. Feng, G. G. Scherer, J. Barber, Y. Shao-Horn, Z. J. Xu, Adv.
Mater. 2017, 29, 1606800.
[8] a) X. Liu, Z. Chang, L. Luo, T. Xu, X. Lei, J. Liu, X. Sun, Chem. Mater. 2014,
26, 1889–1895; b) X. Zou, A. Goswami, T. Asefa, J. Am. Chem. Soc. 2013,
135, 17242–17245.
[9] S. Wahl, S. M. El‐Refaei, A. G. Buzanich, P. Amsalem, K.‐S. Lee, N. Koch,
M.‐L. Doublet, N. Pinna, Adv. Energy Mater. 2019, 9, 1900328.
[10] P. W. Menezes, A. Indra, A. Bergmann, P. Chernev, C. Walter, H. Dau, P.
Strasser, M. Driess, J. Mater. Chem. A 2016, 4, 10014–10022.
[11] a) H. Cheng, M.-L. Li, C.-Y. Su, N. Li, Z.-Q. Liu, Adv. Funct. Mater. 2017,
27, 1701833; b) Z.-Q. Liu, H. Cheng, N. Li, T. Y. Ma, Y.-Z. Su, Adv. Mater.
2016, 28, 3777–3784.
This article is protected by copyright. All rights reserved.

10.1002/anie.202000690
Angewandte Chemie International Edition
RESEARCH ARTICLE
RESEARCH ARTICLE
Xiao-Tong Wang, Ting Ouyang, Ling
Wang, Jia-Huan Zhong and Zhao-Qing
Liu*
The outstanding electrocatalytic performance of
Zn0.4Ni0.6Co2O4/NCNTs towards ORR/OER is
validated, which presents remarkable rate
capability and durability in liquid-flow Zn-air battery.
Moreover, the dual-reinforcement mechanism in
the Zn-Ni-Co ternary spinel is also proposed.
Benefiting from the combined effect of surface
structure optimization and cation segregation,
Zn0.4Ni0.6Co2O4/NCNTs exhibits extreme durability
and electrochemically enhanced properties, and
these advantages make it possible to apply in
practical rechargeable zinc-air batteries.
Page No. – Page No.
Surface
Reorganization
on
Electrochemically-Induced Zn-Ni-Co
Spinel Oxides for Enhanced Oxygen
Electrocatalysis
This article is protected by copyright. All rights reserved.