DOI: 10.1002/chem.
201803236 Full Paper
& Electronic Devices
One-Pot Synthesis of Co3O4/Ag Nanoparticles Supported on
N-Doped Graphene as Efficient Bifunctional Oxygen Catalysts for
Flexible Rechargeable Zinc–Air Batteries
Qin Wang+,[a, c] He Miao+,[b] Shanshan Sun,[a] Yejian Xue,*[a] and Zhaoping Liu*[a]
Abstract: Flexible rechargeable zinc–air batteries are consid-
ered as one of the most promising power supplies for the
emerging flexible and wearable electronic devices. However,
the development of flexible zinc–air batteries is stagnant
due to the lack of efficient bifunctional catalysts with high
oxygen catalytic activity and flexible solid-state electrolytes
with high mechanical stability and ionic conductivity. In this
work, Co3O4/Ag@NrGO composite was synthesized by a
facile one-pot method, and the catalyst shows remarkable
oxygen reduction reaction (ORR)/oxygen evolution reaction
(OER) bifunctional catalytic activity and good long-term sta-
bility. In particular, the OER overpotential of Co3O4/Ag@NrGO
reaches 437 mV, outperforming that of the commercial IrO2
Introduction
Recently, flexible and portable electronic devices such as roll-
up displays, wearable sensors, and smart clothing are becom-
ing popular in our daily life.[1] The flexible power supply is an
important research subject in the development of flexible elec-
tronic devices. Currently, lithium-ion batteries (LIBs) are usually
used as the state-of-the-art power source for flexible electronic
devices, but their low intrinsic energy density can hardly meet
the needs for long-lasting energy supply. Therefore, the devel-
opment of new energy storage technology with high-energy
density for flexible electronic devices is of great significance.[2]
[a] Dr. Q. Wang,+ Dr. S. Sun, Dr. Y. Xue, Prof. Z. Liu
Key Laboratory of Graphene Technologies
and Applications of Zhejiang Province
Ningbo Institute of Materials Technology and Engineering (NIMTE)
Chinese Academy of Sciences, Zhejiang 315201 (P.R. China)
E-mail: xueyejian@nimte.ac.cn
liuzp@nimte.ac.cn
[b] Dr. H. Miao+
Faculty of Maritime and Transportation
Ningbo University, Ningbo 315211 (P.R. China)
[c] Dr. Q. Wang+
University of Chinese Academy of Science
19 A Yuquan Rd., Shijingshan District
Beijing 100049 (P.R. China)
[+] These authors contributed equally to this work.
Supporting information and the ORCID identification number(s) for the au-
thor(s) of this article can be found under:
https://doi.org/10.1002/chem.201803236.
Chem. Eur. J. 2018, 24, 14816 – 14823
catalyst. This can be attributed to the combined effects of
Co3O4, Ag, and N-rGO. Furthermore, PAA (polyacrylic acid)
and PVA (polyvinyl alcohol) based gel-electrolytes have been
developed as flexible solid-state electrolytes for zinc–air bat-
teries. The results show that PAA-based electrolyte is more
favorable to the flexible zinc–air battery with a high power
density due to its relatively high ionic conductivity. The max-
imum power density of flexible zinc–air batteries with
Co3O4/Ag@NrGO catalyst and PAA-based electrolyte can
reach 108 mW cm@2, which is almost the highest value
reached in recent reports. This work will provide valuable
guidance for the development of flexible rechargeable zinc–
air batteries with high power density and stability.
Zinc–air batteries (ZABs) featured by a unique half-opened
system have attracted much attention. The theoretical energy
density of ZABs (1084 Wh kg@1) is about three to five times
higher than that of the LIBs. In addition, zinc–air batteries have
the advantages of being low in cost, effectively rechargeable
in alkaline electrolyte, safe, and nontoxic. Therefore, flexible
zinc–air batteries are considered as a promising power alterna-
tive for flexible electronic devices.[3]
As a flexible battery, it is necessary to maintain electrochemi-
cal and mechanical stability under repeated external loads. The
development of flexible zinc–air batteries is still in the early
stage with many problems needing to be solved. The most im-
portant issues for developing flexible zinc–air batteries are the
flexible structure and electrolyte membranes. For the flexible
structure, two kinds of structures have been usually used in
flexible zinc–air batteries, that is, cable- and sandwich-type
structures. Recently, the latter has been widely applied in flexi-
ble zinc–air and Li–air batteries.[4] For flexible electrolyte mem-
branes, they should keep high ionic conductivity and selectivi-
ty, a wide electrochemical window, chemical compatibility with
other components, and good mechanical properties.[5] One
method to prepare electrolyte membranes is to dip porous
separators, such as glass fiber paper in liquid electrolyte, but
this method has the hidden dangers of electrolyte evaporation
and leakage.[6] Another method is to adopt a solid-state elec-
trolyte, which can avoid internal short circuiting of the batter-
ies and the leakage problem, whilst always showing poor ionic
conductivity.[7] As a new generation of solid-state electrolyte,
alkaline gel electrolyte (AGE) has a good adsorption capacity
14816 T 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 Full Paper

of liquid electrolyte, which is beneficial for the OH@ ions trans- and high-performance gel-electrolyte, which will be valuable
port and participation in reactions, making it a promising can- for the development and application of flexible rechargeable
didate for electrolytes in flexible zinc–air batteries.[8] zinc–air batteries.
In addition, the significant challenge for flexible zinc–air bat-
teries is the intrinsically sluggish kinetics of the oxygen elec-
trode. Many efforts have been devoted to developing efficient
Results and Discussion
electrochemical catalysts, in order to facilitate the oxygen re- The crystalline phases of the as-prepared graphene composites
duction reaction (ORR) and oxygen evolution reaction (OER).[9] have been characterized by XRD, and the diffraction patterns
Up to now, the best catalysts for ORR are Pt and its alloys, and are shown in Figure 1. It can be seen that the diffraction peaks
the favorite catalysts for OER are Ir, Ru, and their oxides. But at 38.1, 44.3, 64.5, and 77.48 can be assigned to (111), (200),
the high price and scarcity of these precious-metal-based cata- (220), and (311) planes of Ag, according to the standard data-
lysts seriously limit their practical application.[10] Instead of the base (PDF#87-0717-Ag). Whereas the diffraction peaks at 19,
precious metal-based catalysts, transition-metal-based catalysts 31.3, 36.8, and 59.48 correspond to the (111), (220), (311), and
have attracted intensive research interests due to their low (511) planes of Co3O4, according to the standard database
price and acceptable catalytic activities. Among them, Co3O4 is (PDF#74-2120-Co3O4). Therefore, the as-synthesized composites
one of most widely studied materials, though its ORR activity can be confirmed to Co3O4@rGO, Ag@rGO, Co3O4/Ag@rGO, and
is not very satisfactory.[11] On the other hand, Ag is regarded as Co3O4/Ag@NrGO, respectively, without any impurity phase.
a promising ORR catalyst due to its relatively low price and
direct four-electron reaction on the surface.[12] Also, Ag has the
highest electric conductivity (6.8213 V 107 S m@1) in all metals,
while the value is as low as & 10@2 S cm@1 for Co3O4.[13] Unfortu-
nately, the OER catalytic activity of Ag is particularly poor. The
composite of Co3O4 and Ag may achieve favorable bifunctional
catalytic effects toward ORR and OER. Amin et al.[14] reported
that the mixture of Co3O4 nanoparticles (50 nm, 99.5 %, Aldrich)
and Ag nanoparticles (Ag 311) prepared by simple physical
mixing can be used as an efficient bifunctional oxygen catalyst
of Li–air batteries. From the rotating-disk electrode tests, the
optimized composite of Co3O4/Ag showed a much lower over-
potential ( & 320 mV) for ORR than pure Co3O4 catalyst, and
was verified to be a four-electron pathway. Meanwhile, the
OER activity of the composite catalyst is about 1.5 times higher
than the single Co3O4 catalyst. Generally speaking, the compo-
sites prepared by physical mixing usually shows poor interac-
tion between each species, and cannot realize the optimal cat-
alytic activities. Therefore, more efficient synthesis strategies
should be developed to prepare Co3O4/Ag composite.[15]
Another method to improve the conductivity and catalytic
activity of Co3O4 is compositing it with the conductive carbon
materials, especially, some heteroatom-doped carbon materials
that can also provide active sites for oxygen catalysis. Dai’s
group[11a] found that, the in-suit growth of Co3O4 nanoparticles
Figure 1. XRD patterns of the as-prepared samples.
(4–8 nm) on N-doped reduced graphene (rGO) sheets can
create synergistic effects, and the Co3O4/rGO composite shows
much better ORR/OER catalytic activities than the single Co3O4 The micromorphology of the typical sample Co3O4/
catalyst or Co3O4 + rGO catalyst prepared by the physical Ag@NrGO has been observed by SEM and TEM equipment.
mixing method. The SEM image in Figure 2 a shows the curved graphene
In this work, we synthesized Co3O4/Ag@NrGO composite by sheets clearly. The TEM image in Figure 2 b shows that the
a facile one-pot solvothermal method. The Co3O4 and Ag parti- spherical particles with diameters of 10–15 nm are evenly dis-
cles were directly nucleated and anchored on the graphene tributed on the graphene sheets. The HR-TEM images (Fig-
sheets. Simultaneously, nitrogen doping on the skeleton of ure 2 c,d) reveal that the lattice fringes with a spacing of
graphene was realized. The combined effects of Co3O4, Ag, and 0.236 nm can be assigned to the Ag(111) crystalline plane, and
NrGO can significantly enhance the ORR/OER bifunctional cata- those of 0.244 nm correspond to the (311) plane of Co3O4.
lytic activity. Then, a flexible zinc–air battery was fabricated Energy dispersion spectrum (EDS) analysis (Figure 2 e) has been
using Co3O4/Ag@NrGO oxygen catalyst and PAA/PVA alkaline conducted on the red-box area in the TEM image, which con-
gel electrolytes and showed the promising application pros- firms the existence of C, Co, Ag, and Cu elements, whereas Cu
pect. This work focused on the efficient bifunctional catalyst is the substrate used in the TEM sample preparation. The N el-

Chem. Eur. J. 2018, 24, 14816 – 14823 www.chemeurj.org 14817 T 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Figure 2. SEM image (a), TEM image (b), HR-TEM images (c,d), and EDS analysis (e) of
Co3O4/Ag@NrGO.
ement has not been detected, which may due to its relatively
low content in the small detection area.
XPS measurements have been performed on the as-pre-
pared samples to investigate the concentration and the oxida-
tive state of the elements on the surface of materials. As
shown in Table 1, the content of Co and Ag in the Co3O4/
Ag@rGO and Co3O4/Ag@NrGO samples are much lower than
Table 1. XPS quantitative analyses of the as-prepared samples.
Species concentration [wt. %]
Catalysts Co 2p Ag 3d O 1s N 1s
Co3O4@rGO 49.28 – 27.06 – 23.66
Ag@rGO – 26.54 19.32 – 54.14
Co3O4/Ag@rGO 27.05 5.79 26.03 – 41.14
Co3O4/Ag@NrGO 26.82 4.2 25.5 3.62
those of the other two samples (Co3O4@rGO and Ag@rGO),
which is consistent with the experimental design, in order to
verify the combined effects of Co and Ag species in the Co3O4/
Ag composites. Figure 3 a shows the wide spectrum of the
Co3O4/Ag@NrGO sample. The signals of Co, Ag, O, N, and C
can be detected, which is consistent with the elements in the
sample. It also confirms that the N atoms have been success-
fully doped in the graphene lattice structure.
Figure 3 b shows the high-resolution XPS spectrum of Co 2p,
which contains two major peaks Co 2p3/2 (780.8 eV) and Co
2p1/2 (796.0 eV), and a pair of satellite peaks. The major peaks
can be further fitted into four subpeaks, which correspond to
Co3 + (779.9, 795.1 eV) and Co2 + (781.9, 797.1 eV), respectively.
The spin-orbit splitting is about 15.2 eV, providing further evi-
dence for the oxidative state of the Co species.[16] The mixed
valences of Co ions lead to bifunctional activity towards ORR
and OER, by acting as donor–acceptor chemisorption sites for
reversible adsorption–desorption of oxygen.[17] Figure 3 c is the
high-resolution XPS spectrum of Ag, which is composed of
Chem. Eur. J. 2018, 24, 14816 – 14823 www.chemeurj.org
Full Paper
two peaks at 368.2 and 374.2 eV, relating to Ag 3d5/2
and Ag 3d3/2, respectively.[18] This indicates that Ag
ions have been thoroughly reduced to metal sliver
and deposited on the graphene sheets during the
solvothermal reaction. Figure 3 d shows the high-res-
olution XPS spectrum of N 1 s, which can be divided
into three peaks, representing pyridinic N (398.3 eV),
pyrrolic N (399.6 eV), and graphitic N (401.6 eV), re-
spectively.[19] Different N species lead to different
chemical and electronic environments of the neigh-
boring carbon atoms, resulting in different electroca-
talytic activities.[20] Generally, the graphitic N and the
pyridinic N are considered as the active sites toward
ORR and OER.[21]
Figure 4 a shows the LSV polarization curves of the
as-prepared composites during ORR in the potential
range of 0.1–1.1 V vs. RHE, which are obtained in
0.1 m KOH electrolyte at the rotation rate of
C 1s
39.85
Figure 3. XPS surface analysis of the Co3O4/Ag@NrGO: a) overall spectrum;
b) Co 2p; c) Ag 3d; d) N 1s.
1600 rpm and scan speed of 5 mV s@1. The half-wave potential
(E1/2) of different samples shows the order as: Ag@rGO >
Co3O4/Ag@NrGO > Co3O4/Ag@rGO > Co3O4@rGO, which are
0.751, 0.735, 0.712, and 0.629 V, respectively. Figure 4 b shows
the LSV polarization curves of the Co3O4/Ag@NrGO sample at
different rotation rates (400, 600, 900, 1200, 1600, 2000, and
2400 rpm). According to K–L equations described in the Sup-
porting Information, the electron-transfer number can be fitted
to & 3.86. Figure 4 c shows the mass specific activities of differ-
ent samples, it be seen, Co3O4/Ag@NrGO has the highest mass
specific activity (0.055 mA mg@1) at 0.7 V, a little higher than
that of Ag@rGO (0.052 mA mg@1).
Figure 4 d shows the LSV polarization curves of OER in the
potential range of 1.1 &1.9 V vs. RHE. To generate a current
density of 10 mA cm@2, the Co3O4/Ag@NrGO sample needs an
overpotential h of 437 mV, which is the smallest among those
of prepared catalysts in this work. This value is even lower
14818 T 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 Full Paper

Figure 4. Characterization of electrochemical properties: a) the ORR LSV curves; b) the LSV curves conducted at different rotation rates and the fitting lines ac-
cording to K-L equations, c) mass activities measured at 0.7 and 0.75 V; d) the OER LSV curves; e) mass activities measured at the overpotential of 500 and
600 mV; f) the potential difference DE of the as-prepared samples. Electrolyte: O2-saturated 0.1 m KOH solution; rotation rate: 1600 rpm; scan rate: 5 mV s@1.

than that of the commercial IrO2 catalyst (442 mV). For the Ag/
Table 2. Comparison of ORR and OER catalytic activities of the as-pre-
rGO sample, a much bigger overpotential (730 mV) is needed, pared samples with the reported Co-based electrocatalysts.[b]
indicating a poor OER activity, and the fluctuation of the OER
LSV polarization curve at 1.5–1.6 V should be ascribed to the ORR OER
E0 [V] E1/2 Ej = @3 Ej = 10 h[a] DE Ref.
oxidation of silver.[22] Figure 4 e shows the mass specific actives [V] [V] [V] [mV] [V]
of different samples at overpotentials of 500 and 600 mV, re-
Co3O4@rGO 0.933 0.629 0.626 1.703 473 1.08 this
spectively. It can be seen that Co3O4/Ag@NrGO shows the work
highest mass specific activity among the four samples, which Ag@rGO 0.895 0.751 0.750 1.960 730 1.21 this
are 0.056 (h = 500 mV) and 0.087 mA mg@1 (h = 600 mV), respec- work
tively. Whereas, the Ag/rGO sample shows the worst mass spe- Co3O4/Ag@rGO 0.937 0.712 0.717 1.708 478 0.99 This
work
cific activity. Figure 4 f shows the potential differences (DE) of Co3O4/ 0.974 0.735 0.737 1.667 437 0.93 This
the four samples between Ej = 10 and @3 mA cm@2, and the Ag@NrGO work
smaller DE represents the better bifunctionality towards ORR Pt/C 0.939 0.801 0.800 – – – this
and OER. It can be seen, among the four samples that Co3O4/ work
IrO2 – – – 1.672 442 – this
Ag@NrGO shows the smallest DE (0.93 V), which is also lower work
than those of similar catalysts recently reported in the litera- Meso-Co3O4 0.66 1.62 0.96 [23a]
ture (shown in Table 2).[23] Co/Co3O4-NG 0.617 1.667 1.05 [23b]
In order to investigate the long-term stability of Co3O4/ CoOx/CNT 0.64 1.62 0.98 [23c]
Co3O4/oCNT 0.75 1.75 1.00 [23d]
Ag@NrGO during the ORR, a 40 000 s chronoamperometry (CA)
test has been conducted at 0.4 V. The inset in Figure 5 a shows [a] The overpotential h is obtained as follows: h = Ej = 10 mA cm@2@1.23 V
(vs. RHE), in which 1.23 V (vs. RHE) is the potential of O2/H2O equilibrium.
the current retention curves of Co3O4/Ag@NrGO and Pt/C cata-
[b] All potentials in Table 2 are referenced with respect to reversible hy-
lysts during the CA tests. It can be seen, after the CA test, that drogen electrode (RHE) scale.
the current retention of Co3O4/Ag@NrGO (96 %) is higher than
that of Pt/C (88 %). Comparing the LSV polarization curves
before and after CA tests, it can be seen that the two curves of
the Co3O4/Ag@NrGO catalyst are nearly coincident, indicating unchanged for Co3O4/Ag@NrGO, but it increases obviously for
the good stability. Whereas, an evidently negative shift IrO2 catalyst. The LSV polarization curve nearly coincides with
( & 35 mV) can be observed for the Pt/C catalyst. Figure 5 b the former one for Co3O4/Ag@NrGO, but shifts positively for
shows the long-term OER stability tests of Co3O4/Ag@NrGO IrO2 catalyst. The results demonstrate that Co3O4/Ag@NrGO has
and IrO2 catalysts. A chronopotentiometry (CP) measurement a better ORR and OER stability than those of commercial pre-
has been performed at a current density of 10 mA cm@2 that cious metal catalysts.
lasted for 10 000 s. From the potential degradation curves To evaluate the performance of the as-prepared graphene-
(insert in Figure 5 b), the potential during the CP test is almost composite catalysts in practical application, the liquid zinc–air

Chem. Eur. J. 2018, 24, 14816 – 14823 www.chemeurj.org 14819 T 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
 Full Paper

Figure 5. a) Comparison of the LSV curves of Co3O4/Ag@NrGO and 20 wt. %

Pt/C before and after chronoamperometric (CA) tests, insert: the CA respons-
es at 0.4 V. b) Comparison of the LSV curves of Co3O4/Ag@NrGO and IrO2
before and after chronopotentiometry (CP) tests; insert: the CP responses at
a current density of 10 mA cm@2. Electrolyte: O2-saturated 0.1 m KOH solu-
tion; rotation rate: 1600 rpm; scan rate: 5 mV s@1.

battery has been assembled. The schematic diagram and

actual photograph are shown in Figure 6 a and b, respectively.
The charge–discharge polarization curves of zinc–air batteries
with different cathode catalysts are shown in Figure 6 c. It can
be seen that the battery with Ag@rGO catalyst shows a rela-
tively high power density, but suffers from a much higher
charging voltage. Compared with Ag@rGO, the battery with
Co3O4@rGO catalyst shows a lower charging voltage, but
worse discharge performance. For the battery using Co3O4/
Ag@NrGO catalyst, the power density and discharging voltage
at the same current density are comparable to that with Ag/
rGO catalyst, and the charging voltage at the same current
density is as low as that with Co3O4@rGO catalyst. This demon-
strates that Co3O4/Ag@NrGO is a promising bifunctional cata-
lyst with the advantages of Ag/rGO and Co3O4@rGO catalysts.
Figure 6 d shows the charge–discharge cycling curves (at
10 mA cm@2) of batteries with Co3O4/Ag@NrGO and Pt/C cata-
lysts. It can be seen, after 100 cycles that the charge–discharge
Figure 6. Schematic illustration (a) and actual photograph (b) of a liquid re-
chargeable zinc–air battery; c) Charge-discharge polarization curves of re-
chargeable zinc–air batteries with different catalysts; d) Charge-discharge cy-
cling curves of zinc–air batteries with Co3O4/Ag@NrGO and 20 wt. % Pt/C
catalysts at 10 mA cm@2.
voltage gap is 1.08 V for the Co3O4/Ag@NrGO battery, whereas
the value is as high as 2.05 V for the Pt/C battery due to its
severe degradation during the cycling tests.
We further investigate the performance of the flexible zinc–
air batteries with Co3O4/Ag@NrGO catalyst. Figure 7 a,b are the
structure illustration and actual photograph of the sandwich-
structured flexible zinc–air battery, respectively. The charge-dis-
charge polarization curves of the flexible zinc–air batteries
with Co3O4/Ag@NrGO catalyst and PAA/PVA-based gel electro-
lytes are shown in Figure 7 c. The maximum power density of

Figure 7. Schematic illustration (a) and actual photograph (b) of a flexible rechargeable zinc–air battery. c) Charge-discharge polarization curves of flexible
zinc–air batteries with Co3O4/Ag@NrGO catalyst and PAA/PVA-based electrolytes; d) the galvanostatic discharge curves at different current plateaus; e) the
charge-discharge cycling curves of flexible zinc–air batteries with PAA and PVA-based electrolytes at 2 mA cm@2 ; f) The AC impedance spectroscopy of gel-
electrolytes.

Chem. Eur. J. 2018, 24, 14816 – 14823 www.chemeurj.org 14820 T 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

the zinc–air battery with PAA and PVA-based electrolyte are
108 and 39 mW cm@2, respectively. Furthermore, the battery
with PAA-based electrolyte is more favorable to discharge
under a big current density. When the current density is more
than 50 mA cm@2, a rapid decrease of discharge voltage and
power density can be observed for the PVA-based battery due
to the large concentration polarization of the gel electrolyte.
This demonstrates that the battery with PAA-based electrolyte
can effectively eliminate the concentration polarization and
reach the higher power density. Figure 7 d shows the discharge
curves at a current density of 2, 3, 4, 5, 6, and 8 mA cm@2, and
each discharge plateau lasts for 10 min. It can be seen that the
PVA-based battery shows significant voltage decay in the last
two discharge platforms due to the concentration polarization.
Figure 7 e shows the charge–discharge cycling curves (at
2 mA cm@2) of the PAA and PVA-based batteries. Each cycle
contains a 5 min charging process and a 5 min discharging
process. It can be seen that no obvious degradation has oc-
curred for the PAA-based battery, whereas the PVA-based bat-
tery suffers somewhat degradation. The charge–discharge volt-
age gaps of the PAA and PVA-based batteries after 100 cycles
are 0.77 and 0.82 V, respectively.
The ionic conductivities of PAA- and PVA-based gel electro-
lytes have been tested by the AC impedance spectroscopy
method (described in the Supporting Information), and the re-
sults are shown in Figure 7 f. The calculated ionic conductivity
of the PAA-based electrolyte is 0.282 S cm@1, which is
1.84 times higher than that of the PVA-based electrolyte
(0.153 S cm@1). Therefore, the good discharging performance of
the PAA-based battery can be ascribed to the high ionic con-
ductivity of PAA. Figure 8 shows a demonstration of a flexible
rechargeable zinc–air battery working at different bending
angels (take the case of the PAA-based electrolyte). It can be
seen that the open circuit voltage of the single zinc–air battery
is about 1.3 V. At conditions of 0, 90, and 1808 bending, the
single flexible battery can normally power the LED light, which
confirms that the PAA-based electrolyte has good flexibility
and mechanical stability.
Figure 8. Bending property tests of the flexible zinc–air batteries.
Chem. Eur. J. 2018, 24, 14816 – 14823 www.chemeurj.org
Full Paper
Conclusions
In this work, we developed a flexible rechargeable zinc–air bat-
tery with high power density and stability by using Co3O4/
Ag@NrGO catalyst and PAA-based electrolyte. The highly effi-
cient Co3O4/Ag@NrGO catalyst was synthesized by directly nu-
cleating, growing, and anchoring of Co3O4 and Ag nanoparti-
cles on N-doped graphene sheets. The composite shows excel-
lent ORR/OER catalytic activities by combining the advantages
of Co3O4, Ag, and N-rGO. Remarkably, the OER overpotential of
Co3O4/Ag@NrGO was as low as 437 mV, outperforming that of
the commercial IrO2 catalyst. Furthermore, the flexible PAA-
based solid-state electrolyte with high ionic conductivity and
mechanical stability was obtained by a facile organic-gel
method. Using Co3O4/Ag@NrGO catalyst and PAA-based elec-
trolyte, the flexible zinc–air battery can reach the Pmax of
108 mW cm@2, and exhibit good cycling stability. We believe
that our facile and effective strategies in preparing bifunctional
oxygen catalysts and gel-electrolytes are promising for the de-
velopment of next-generation flexible zinc–air batteries.
Experimental Section
Sample preparation
Graphene oxide (GO) was prepared from a natural graphite sheet
based on the modified Hummers method, which was reported in
the literature.[24] The finally obtained GO/H2O suspension was con-
centrated, and the concentration of GO in the suspension was
measured by a freeze-drying and weighting process, which was
4.65 mg g@1. For the synthesis of Co3O4@rGO, GO suspension
(12.9 g, GO & 60 mg) was added to anhydrous ethanol (60 mL).
After sonication for 30 min, Co(CH3COO)2·4 H2O (0.44 g) was added
and dissolved in the above GO suspension. The mixture was
placed in a water bath at 80 8C with continuous stirring for 10 h. In
this process, the Co(Ac)2 was hydrolyzed, and the Co3O4 was nucle-
ating on the graphene sheets. Then, the mixture was transferred to
a 100 mL stainless-steel Teflon-lined autoclave for the solvothermal
reaction (150 8C, 3 h). In this process, Co3O4 was further crystallizing
and growing; meanwhile, the oxide graphene was reduced. The re-
action products were washed by deionized water and anhydrous
alcohol 3 times. After centrifuging and freeze-drying, the final
Co3O4@rGO composite was obtained. The Ag@rGO composite was
synthesized under the same procedure, just by replacing the
Co(Ac)2 by 120 mg silver nitrate (AgNO3).
For the synthesis of Co3O4/Ag@rGO, GO suspension (12.9 g, GO
& 60 mg) was added to anhydrous ethanol (60 mL) to form a ho-
mogeneous dispersion. Then Co(Ac)2 (0.22 g) and AgNO3 (60 mg)
were dissolved in deionized water (12 and 6 mL, respectively). The
two solutions were mixed with the GO and ethanol dispersion. The
mixture was treated in the water bath and underwent the solvo-
thermal reaction (see the previous description). During the solvo-
thermal reaction, Ag + was reduced to Ag, and the final Co3O4/
Ag@rGO product was obtained. The target Co3O4/Ag@NrGO com-
posite was prepared by adding NH3OH (2.5 mL, 30 wt.%) solution
before the solvothermal process and the remaining other steps.
Sample characterizations
The chemical phases of the four samples were identified by X-ray
diffractometer (XRD, D8 Advance, Germany) with CuKa radiation
14821 T 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

(1.5418 a). The micromorphologies of the samples were observed
by SEM (S4800, Hitachi) and TEM (JF-20, Hitachi). The element con-
tents and valence state information on the surface of the materials
were characterized by XPS (AXIS ULTARDLD).
Electrochemical measurements
The electrochemical performance of the materials has been investi-
gated by the rotating-disk electrode equipment (RDE, Pine Instru-
ments) together with an electrochemical workstation (CHI 660E,
CH Instruments). A standard three-electrode system was used in
the test with 0.1 m KOH as the electrolyte and Hg/HgO and Pt wire
as the reference and counter electrode, respectively. The working
electrode was prepared as follows: catalysts (10 mg) and carbon
support (5 mg, Vulcan-XC72) were dispersed in anhydrous ethanol
(2 mL), and then Nafion solution (80 mL, 5 wt %, DuPont) was
added. The mixture was homogenized by a 60 min sonication.
Then, 10 mL of the dispersion was coated on the polished glass
carbon electrode. After drying thoroughly, a catalyst film was
formed with a catalyst loading of 255 mgoxide cmdisk@2. For compari-
son, the Pt/C and IrO2 electrode were prepared with the same pro-
cedure and the same loading. Linear sweep voltammetry (LSV)
methods were used to study the catalytic activities of the catalyst
samples. Chronoamperometry (CA) and chronopotentiometry (CP)
methods were used to investigate the long-term stability of the
catalyst materials.
Liquid zinc–air battery tests
The liquid zinc–air battery was assembled by a metal anode, an air
cathode, and liquid electrolyte. A polished zinc plate was used as
the metal anode, and 6 m KOH solution with 0.2 m Zn(AC)2 was
used as the electrolyte. The air-cathode was prepared as follows:
Toray carbon paper (TGP-H-060, Toray Industries Inc.) was used as
the substrate, Vulcan-XC72 and PTFE dispersion (60 wt. %) was
mixed and sprayed on one side of the carbon paper (the solid con-
tent of PTFE is about 20 wt. %). Then the carbon paper was cal-
cined in a muffle furnace under 340 8C for 30 min. The formed
layer was named as the gas-diffusion layer, with the function of hy-
drophobicity and air-permeability. The catalyst ink (prepared in the
electrochemical measurement section) was coated on the surface
of the gas-diffusion layer, and the catalyst loading was controlled
at & 1 mg cm@2. The effective area of the air-cathode (exposed to
the electrolyte) was about 0.785 cm@2 (a circle with diameter of
10 mm).
Flexible zinc–air battery tests
The flexible zinc–air battery assembled in this work used a sand-
wich-type structure, as described in Figure S1 (Supporting Informa-
tion). A carbon cloth (W0S1009, Cetech, Taiwan) was used as the
support of the air-cathode, and a 0.1 mm zinc foil was used as the
anode. The gel-electrode (the preparation procedure can be seen
in the Experimental Section of the Supporting Information) was
placed between the anode and the air cathode. The nickel foam
attached on the carbon cloth functioned as the current collector.
Two PET films with holes were used as the protective layer for the
whole battery. The preparation procedure of the air cathode was
as follows: catalyst (21 mg) and carbon powder (21 mg, Vulcan-
XC72) were dispersed in isopropanol (3 mL) solution, then Nafion
solution (150 mL, 5 wt. %) was added. After sonication for 1 h, a cer-
tain amount of the catalyst ink was coated on the carbon cloth.
The carbon cloth was dried on a heating platform at 80 8C, then
the air cathode was obtained with a catalyst loading of 1 mg cm@2.
Chem. Eur. J. 2018, 24, 14816 – 14823 www.chemeurj.org 14822
Full Paper
Acknowledgements
This work was supported by National Key Research and Devel-
opment Program of China (Grant No. 2016YFB0100100), the
Key Research Program of the Chinese Academy of Sciences
(Grant No. KGZD-EW-T08), and Ningbo Natural Science Founda-
tion (2017A610058 and 2018A610015).
Conflict of interest
The authors declare no conflict of interest.
Keywords: bifunctional catalysts · electronic devices · oxygen
evolution reaction · oxygen reduction reaction · zinc–air
batteries
[1] a) B. D. Gates, Science 2009, 323, 1566 – 1567; b) H. Lee, T. K. Choi, Y. B.
Lee, H. R. Cho, R. Ghaffari, L. Wang, H. J. Choi, T. D. Chung, N. Lu, T.
Hyeon, S. H. Choi, D.-H. Kim, Nat. Nanotechnol. 2016, 11, 566; c) S.
Bauer, Nat. Mater. 2013, 12, 871.
[2] a) M.-S. Balogun, M. Yu, C. Li, T. Zhai, Y. Liu, X. Lu, Y. Tong, J. Mater.
Chem. A 2014, 2, 10825 – 10829; b) M.-S. Balogun, M. Yu, Y. Huang, C. Li,
P. Fang, Y. Liu, X. Lu, Y. Tong, Nano Energy 2015, 11, 348 – 355; c) M.-S.
Balogun, Z. Wu, Y. Luo, W. Qiu, X. Fan, B. Long, M. Huang, P. Liu, Y. Tong,
J. Power Sources 2016, 308, 7 – 17.
[3] a) J. Fu, Z. P. Cano, M. G. Park, A. Yu, M. Fowler, Z. Chen, Adv. Mater.
2017, 29, 1604685; b) Y. G. Li, H. J. Dai, Chem. Soc. Rev. 2014, 43, 5257 –
5275.
[4] a) J. Fu, F. M. Hassan, J. Li, D. U. Lee, A. R. Ghannoum, G. Lui, M. A.
Hoque, Z. Chen, Adv. Mater. 2016, 28, 6421 – 6428; b) Q. c. Liu, J. j. Xu,
Z. w. Chang, D. Xu, Y. b. Yin, X. y. Yang, T. Liu, Y. s. Jiang, J. m. Yan, X. b.
Zhang, Particle Particle Syst. Charact. 2016, 33, 500 – 505; c) C. Y. Su, H.
Cheng, W. Li, Z. Q. Liu, N. Li, Z. F. Hou, F. Q. Bai, H. X. Zhang, T. Y. Ma,
Adv. Energy Mater. 2017, 7, 12.
[5] a) E. Quartarone, P. Mustarelli, Chem. Soc. Rev. 2011, 40, 2525 – 2540;
b) Y.-S. Hu, Nat. Energy 2016, 1, 16042.
[6] F. L. Meng, H. X. Zhong, D. Bao, J. M. Yan, X. B. Zhang, J. Am. Chem. Soc.
2016, 138, 10226 – 10231.
[7] J. Park, M. Park, G. Nam, J. S. Lee, J. Cho, Adv. Mater. 2015, 27, 1396 –
1401.
[8] a) Y. F. Xu, Y. Zhang, Z. Y. Guo, J. Ren, Y. G. Wang, H. S. Peng, Angew.
Chem. Int. Ed. 2015, 54, 15390 – 15394; Angew. Chem. 2015, 127, 15610 –
15614; b) Q. Liu, Y. B. Wang, L. M. Dai, J. N. Yao, Adv. Mater. 2016, 28,
3000 – 3006; c) C. Guan, A. Sumboja, H. J. Wu, W. N. Ren, X. M. Liu, H.
Zhang, Z. L. Liu, C. W. Cheng, S. J. Pennycook, J. Wang, Adv. Mater.
2017, 29, 9.
[9] a) Y. G. Li, M. Gong, Y. Y. Liang, J. Feng, J. E. Kim, H. L. Wang, G. S. Hong,
B. Zhang, H. J. Dai, Nat. Commun. 2013, 4, 7; b) J. Zhang, Z. Zhao, Z. Xia,
L. Dai, Nat. Nanotechnol. 2015, 10, 444 – 452.
[10] a) G. Li, X. L. Wang, J. Fu, J. D. Li, M. G. Park, Y. N. Zhang, G. Lui, Z. W.
Chen, Angew. Chem. Int. Ed. 2016, 55, 4977 – 4982; Angew. Chem. 2016,
128, 5061 – 5066; b) Y. C. Lu, Z. C. Xu, H. A. Gasteiger, S. Chen, K. Hamad-
Schifferli, Y. Shao-Horn, J. Am. Chem. Soc. 2010, 132, 12170 – 12171.
[11] a) Y. Liang, Y. Li, H. Wang, J. Zhou, J. Wang, T. Regier, H. Dai, Nat. Mater.
2011, 10, 780 – 786; b) M. H. Yu, Z. K. Wang, C. Hou, Z. L. Wang, C. L.
Liang, C. Y. Zhao, Y. X. Tong, X. H. Lu, S. H. Yang, Adv. Mater. 2017, 29, 7.
[12] B. B. Blizanac, P. N. Ross, N. M. Marković, J. Phys. Chem. B 2006, 110,
4735 – 4741.
[13] H. B. Huang, S. H. Luo, C. L. Liu, Q. Wang, Z. Y. Wang, Y. H. Zhang, A. M.
Hao, Y. G. Liu, J. Z. Li, Y. C. Zhai, Y. N. Dai, J. Alloys Compd. 2017, 726,
939 – 946.
[14] H. M. A. Amin, C. J. Bondue, S. Eswara, U. Kaiser, H. Baltruschat, Electro-
catalysis 2017, 8, 540 – 553.
[15] a) T. Y. Ma, S. Dai, M. Jaroniec, S. Z. Qiao, J. Am. Chem. Soc. 2014, 136,
13925 – 13931; b) D. U. Lee, P. Xu, Z. P. Cano, A. G. Kashkooli, M. G. Park,
Z. W. Chen, J. Mater. Chem. A 2016, 4, 7107 – 7134.
T 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

[16] M. Abirami, S. M. Hwang, J. Yang, S. T. Senthilkumar, J. Kim, W. S. Go, B.
Senthilkumar, H. K. Song, Y. Kim, ACS Appl. Mater. Interfaces 2016, 8,
32778 – 32787.
[17] M. Hamdani, R. N. Singh, P. Chartier, Int. J. Electrochem. Sci. 2010, 5,
556 – 577.
[18] L. Yuan, L. Jiang, J. Liu, Z. Xia, S. Wang, G. Sun, Electrochim. Acta 2014,
135, 168 – 174.
[19] S. Li, H. Miao, Q. Xu, Y. Xue, S. Sun, Q. Wang, Z. Liu, RSC Adv. 2016, 6,
99179 – 99183.
[20] X. J. Zheng, X. C. Cao, X. W. Li, J. H. Tian, C. Jin, R. Z. Yang, Nanoscale
2017, 9, 1059 – 1067.
[21] a) D. Guo, R. Shibuya, C. Akiba, S. Saji, T. Kondo, J. Nakamura, Science
2016, 351, 361 – 365; b) Z. Zhao, M. Li, L. Zhang, L. Dai, Z. Xia, Adv.
Mater. 2015, 27, 6834 – 6840.
[22] M. Z. Iqbal, R. J. Kriek, Electrocatalysis 2018, 9, 279 – 286.
Chem. Eur. J. 2018, 24, 14816 – 14823 www.chemeurj.org 14823
Full Paper
[23] a) W. Q. Song, Z. Ren, S. Y. Chen, Y. T. Meng, S. Biswas, P. Nandi, H. A.
Elsen, P. X. Gao, S. L. Suib, ACS Appl. Mater. Interfaces 2016, 8, 20802 –
20813; b) B. He, X. Chen, J. Lu, S. Yao, J. Wei, Q. Zhao, D. Jing, X. Huang,
T. Wang, Electroanalysis 2016, 28, 1 – 10; c) B. Seo, Y. J. Sa, J. Woo, K.
Kwon, J. Park, T. J. Shin, H. Y. Jeong, S. H. Joo, ACS Catal. 2016, 6, 4347 –
4355; d) S. Zhao, B. Rasimick, W. Mustain, H. Xu, Appl. Catal. B 2017,
203, 138 – 145.
[24] X. Zhou, Z. Liu, Chem. Commun. 2010, 46, 2611 – 2613.
Manuscript received: June 26, 2018
Revised manuscript received: July 20, 2018
Accepted manuscript online: July 31, 2018
Version of record online: September 6, 2018
T 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim