Journal of

Materials Chemistry A
View Article Online
PAPER View Journal | View Issue
Published on 21 April 2017. Downloaded by National Institute of Technology - Rourkela on 3/8/2019 5:30:07 AM.

Simple synthesis of graphene-doped flower-like

Cite this: J. Mater. Chem. A, 2017, 5,
cobalt–nickel–tungsten–boron oxides with self-
9907 oxidation for high-performance supercapacitors†
Cuili Xiang,ab Qingyong Wang,ab Yongjin Zou, *ab Pengru Huang,ab
Hailiang Chu, ab Shujun Qiu,ab Fen Xuab and Lixian Sun*ab

In this study, we devised an easy and simple approach to synthesize a composite of flower-like cobalt–
nickel–tungsten–boron oxides (Co–Ni–W–B–O) that were doped with reduced graphene oxide (rGO);
the composite was designed for supercapacitor applications. A Co–Ni–W–B alloy was first deposited on
rGO through one-pot chemical reduction in an ethanol solution at room temperature. The resulting Co–
Ni–W–B alloy self-oxidized in air on the rGO surface. The Co–Ni–W–B–O/rGO composites resembled
three-dimensional flowers with a high surface area; they also exhibited superior electrochemical
performance when compared to most previously reported electrodes based on nickel–cobalt oxides.
Furthermore, the Co–Ni–W–B–O/rGO composite prepared in an ethanol solution showed much higher
Received 8th January 2017
Accepted 21st April 2017
electrochemical performance than the composite prepared in water. The Co–Ni–W–B–O/rGO
electrode showed an ultrahigh specific capacitance of 1189.1 F g
1 at 1 A g
1 and exhibited a high energy
DOI: 10.1039/c7ta00234c
density of 49.9 W h kg
1 along with remarkable cycle stability (10 000 cycles with 80.7% capacitance
rsc.li/materials-a retention at 15 A g
1), which is promising for its application in energy storage devices.

exhibit 10–100 times higher energy density but poor cycle

1. Introduction performance compared to EDLCs.6 Conducting polymers and
In recent decades, the increasing consumption of traditional transition-metal oxides are commonly used for pseudocapaci-
fossil fuels and growing concerns regarding air pollution have tors. As a matter of fact, hybrid SCs exhibiting both pseudoca-
spurred an urgent need to develop renewable and green energy pacitance and electrical-double-layer capacitance have attracted
sources. Meanwhile, new energy-storage devices with high signicant research interest in recent years. These hybrid SCs
power density and charge–discharge efficiency are also inten- have the advantage of high supercapacitance as well as excellent
sively required. As a result, supercapacitors (SCs), a promising stability.
energy-storage device, have attracted much attention owing to Transition-metal oxides and hydroxides are regarded as ideal
their higher power capability, better reversibility, faster pseudocapacitive materials for SCs because they possess
recharge ability, and longer cycling life when compared to multiple oxidation states that enable rich redox reactions to
batteries.1–5 Depending on the energy-storage mechanism, SCs store energy. A variety of nanostructured metal oxides such as
can generally be divided into electrical-double-layer capacitors Fe2O3, MnO2, NiO, Co3O4, and their composites are commonly
(EDLCs) and pseudocapacitors. EDLCs usually show high used.7–13 For example, Li et al. fabricated Ni/Co layered double
stability but very limited specic capacitance. Carbon-based hydroxides via self-assembling; the prepared composites deliv-
conductive nanomaterials are oen used to store energy by ered an excellent capacitance of about 2228 F g
1 at a current
charge accumulation at the electrochemical electrode/ density of 1 A g
1 and exhibited excellent cycling stability.14 Zuo
electrolyte interface in EDLCs. Pseudocapacitors, which accu- et al. synthesized a composite nanolm of Co3O4/NiO by the
mulate charge via the faradaic electron-transfer reaction, calcination of a precursor comprising a-Co(OH)2–a-Ni(OH)2;
the composite has attracted considerable attention for super-
capacitor applications because of its superior electrochemical
a
Guangxi Key Laboratory of Information Materials, Guilin University of Electronic performance at relatively low cost.15 Unfortunately, the observed
Technology, Guilin 541004, P. R. China. E-mail: [email protected]; sunlx@guet. specic capacitances are much lower than the theoretical value
edu.cn in all these oxides. Owing to their poor electronic conductivity
b
Guangxi Collaborative Innovation Center of Structure and Property for New Energy and the large volume change during recurring redox reaction
Materials, Guilin 541004, P. R. China
processes, the practical use of these materials is also hindered
† Electronic supplementary information (ESI) available: SEM images and EDX
images of samples; XPS spectra, CV data, adsorption data, XRD patterns, and
by rapid capacitance fading upon extended cycling. To resolve
electronic data of samples. See DOI: 10.1039/c7ta00234c these problems, pseudocapacitive materials are prepared with

This journal is © The Royal Society of Chemistry 2017 J. Mater. Chem. A, 2017, 5, 9907–9916 | 9907
 View Article Online

Journal of Materials Chemistry A Paper

specially designed hierarchical nanostructures and doped with a homogeneous solution. Next, 20 mL of a freshly prepared
nanocarbon materials, and they simultaneously serve as NaBH4 solution (10 wt%) was slowly released in droplets to the
building blocks with nanostructures and conductive above mixture under vigorous stirring. Aer 1 h, the black
components.16 sample was collected and washed thoroughly with deionized
Cobalt-based double hydroxides or oxides are important water and then with absolute ethanol. The sample was then
transition-metal-oxide materials for SCs largely because of their dried in air for 24 h at 25  C. During this process, the Co–Ni–W–
Published on 21 April 2017. Downloaded by National Institute of Technology - Rourkela on 3/8/2019 5:30:07 AM.

ultrahigh capacitance.10 Furthermore, cobalt-based oxides are B/rGO sample self-oxidized in air (hereaer referred to as Co–
especially interesting because of their wide availability, cost Ni–W–B–O/rGO). Depending on the amount of GO (in wt%) in
effectiveness, and well-dened electrochemical properties. The the precursor suspension, the prepared samples were named
addition of another transition metal such as Ni could further Co–Ni–W–B–O/10rGO, Co–Ni–W–B–O/20rGO, Co–Ni–W–B–O/
improve the electrochemical properties owing to the decrease in 30rGO, Co–Ni–W–B–O/40rGO, and Co–Ni–W–B–O/50rGO. For
resistance and increase in oxygen overpotential.16 From comparison, the Co–Ni–W–B/20rGO composite was synthesized
a materials science point of view, the electrochemical perfor- in water, while the Co–B, Co–Ni–B, Co–W–B, and Co–Ni–W–B
mance of a cobalt-based oxide nanostructure depends on its composites were prepared in ethanol solutions under the same
surface area, particles size, crystal structure, and morphology. conditions. Control samples of Co–Ni–B, Co–W–B, and Co–Ni–
Mesoporous nanostructures have been fabricated to develop W–B doped with GO were also prepared under the same
supercapacitor electrodes with high surface areas. However, conditions. These non-oxidized samples were dried under
commonly used preparation methods such as hydrothermal vacuum, and electrodes from these samples were prepared in
methods, solvothermal methods, physical vapor deposition, a glovebox to prevent oxidation.
and electrodeposition are time-consuming, labor-intensive, and To investigate the effect of heat treatment on the electro-
costly.17–21 Surprisingly, we discovered during our study that chemical performance of Co–Ni–W–B–O/20rGO, the samples
a Co-based ternary alloy prepared in an alcohol solution can be were calcined at 200–500  C in N2 for 2 h.
easily oxidized in air at room temperature. This technique thus
provides an opportunity for the mass production of hybrid 2.2 Characterization of materials
structures of transition-metal oxides and other metal oxides in
Scanning electron microscopy (SEM) and energy-dispersive X-
a simple and effective manner.
ray (EDX) analysis (S4800, Hitachi, Japan) were used to char-
The chemical reduction of a metal alloy from a precursor
acterize the morphology and surface composition of the
solution is widely used for the preparation of catalysts. Based on
samples at an accelerating voltage of 10 kV. Structural analysis
this technique, we devised an approach that combines chemical
was conducted by transmission electron microscopy (TEM;
reduction and self-oxidation to directly grow transition-metal
JEOL 2010, JEOL, Japan) at 300 kV. Powder X-ray diffraction
oxides on a conductive substrate. Since reduced graphene
(XRD) measurements were performed using a diffractometer
oxide (rGO) possesses high conductivity and a high surface area,
(1820, Philips, The Netherlands) with Cu Ka radiation (l ¼
which can suppress volume change during repeated redox
1.5418 Å). X-ray photoelectron spectroscopy (XPS) was carried
reaction processes and particle aggregation, the transition-
out using an XP spectrometer (ESCA 250, Scienta Omicron
metal oxides were doped with rGO. The one-pot synthesis of
GmbH, Germany) equipped with a monochromatic Al Ka
rGO-doped ower-like Co–Ni–W oxides was carried out at room
(1486.6 eV) X-ray source and a hemispherical analyzer. The
temperature. To the best of our knowledge, this kind of nano-
specic surface areas of the composites were measured by N2
structure has not been studied or reported elsewhere. The effect
adsorption at
196  C using the Brunauer–Emmett–Teller
of graphene content in the composite electrode on its pseudo-
(BET) method (Autosorb iQ2, Quantachrome Instruments,
capacitive behavior was also thoroughly studied to determine
USA). The samples were degassed at 100  C until the vacuum
the optimal content ratio and provide insight into the syner-
pressure was below 10
4 Pa. The average value was calculated
gistic effect of both rGO and the oxides.22 The Co–Ni–W–B–O/
aer obtaining three measurements for each sample.
rGO composite exhibited a large specic capacitance, high
capability, and excellent cycling properties, which are prom-
ising for practical energy-storage devices such as portable 2.3 Electrochemical measurements
electronic devices and hybrid electric vehicles. Test electrodes for electrochemical capacitors were prepared by
mixing 8 mg of each synthesized sample with 1 mg carbon black
2. Experimental section and 1 mg polyvinylidene uoride (as the binder) to form
a slurry. Next, 100 mL of ethanol was added to each mixture in an
2.1 Synthesis agate mortar to make it homogeneous. The slurries were coated
All chemicals were of analytical grade and used without further onto 1  1 cm2 nickel-foam current collectors using a blade,
purication. In a typical preparation process, 6 mmol of cobalt and the resulting electrodes were dried at 60  C under vacuum
nitrate hexahydrate (Co(NO3)2$6H2O), 3 mmol of nickel nitrate for 6 h. All the electrochemical experiments were carried out
hexahydrate (Ni(NO3)2$6H2O), 3 mmol of sodium tungstate using a conventional three-electrode system. The as-prepared
dihydrate (Na2WO4$2H2O), and 20 mg (or 10, 30, 40, 50 mg) of electrode acted as the working electrode. A platinum electrode
graphene oxide (GO) were rst added to 60 mL of absolute and a saturated calomel electrode (SCE) were used as the
ethanol, and the solution was stirred for over 24 h to form counter and reference electrodes, respectively.

9908 | J. Mater. Chem. A, 2017, 5, 9907–9916 This journal is © The Royal Society of Chemistry 2017
 View Article Online

Paper Journal of Materials Chemistry A

To construct an asymmetric supercapacitor, Co–Ni–W–B–O/ was clearly observed that the color of the powder changed from
20rGO was used as the positive electrode and rGO was used as dark black to grey aer self-oxidation in air. However, the binary
the negative electrode. The electrolyte was a 6 M KOH solution. alloy Co–B and the ternary alloys Co–Ni–B and Co–W–B were
All the experiments were performed at room temperature (25  still black aer 24 h of exposure to air at room temperature,
1  C). indicating that they were not oxidized at all or not thoroughly
Cyclic voltammetry (CV), electrochemical impedance spec- oxidized. This was probably because the synergistic effect
Published on 21 April 2017. Downloaded by National Institute of Technology - Rourkela on 3/8/2019 5:30:07 AM.

troscopy (EIS) and galvanostatic charge–discharge (GCD) tests between rGO and elemental Co, Ni, and W enhanced the activity
were performed using an electrochemical workstation (IM6e, of the alloys, allowing them to easily oxidize in air. A widely used
Zahner-Elektrik, Germany). The EIS spectra were recorded at an method to enhance the activity of the Co–B alloy is to introduce
open-circuit potential with an amplitude of 10 mV over the other transition metals. The combination of several transition
frequency range from 10 kHz to 100 mHz. The average specic metals can create a synergistic effect to enhance the catalytic
capacitances were calculated from the CV curves and discharge performance of the transition-metal alloy. The dopant elements
curves according to eqn (1) and (2), respectively:23 mostly change the electronic states of the active metal and vary
ð the atomic distance and coordination number to affect the
1
Cs ¼ iðV ÞdV ; (1) catalytic performance. In the case of Co–B, by sharing electrons,
2mvDV
Co becomes a more active center than when it is used alone. An
IDt efficient route to avoid the agglomeration of Co–B particles is to
Cs ¼ ; (2)
mDV introduce an atomic diffusion barrier between them in the form
of transition metals like Ni, Mo, and W. These promoter metals,
where Cs (F g
1) is the specic capacitance, m (g) is the mass of
mainly in the form of oxides, are very efficient, and even a small
active materials in the electrode, v (V s
1) is the scan rate, DV is
atomic concentration can signicantly increase the surface area
the potential window, i (V) is the instantaneous current on the
of the Co–B catalyst powder by avoiding agglomeration.27
CV curve, I (A) is the discharge current, and Dt (s) is the
Therefore, the more active the alloy, the more easily it can react
discharge time.
with oxygen and form multiple metal oxides.
The energy density (E) and power density (P) were calculated
XRD and FTIR measurements were performed to learn more
according to eqn (3) and (4), respectively:24
about the synthesized products. Powder XRD is a reliable
1 technique for determining the crystal structure of materials. As
E¼ Cs ðDV Þ2 ; (3)
2 shown in Fig. 2a, no distinct peaks were observed in the XRD
patterns of the Co–Ni–W–B–O alloy and the Co–Ni–W–O/20rGO
E
P¼ ; (4) composite, indicating that they were basically amorphous. For
Dt
the XRD pattern of GO, the peak at 2q ¼ 10 corresponds to the
where E is in units of W h kg
1 and P is in units of W kg
1. interlayer spacing of 8.6 Å dened by the (002) peak. However,
the peak is absent in the XRD pattern of rGO owing to the
reduction of GO by NaBH4, and it is also absent in the XRD
3. Results and discussion pattern of Co–Ni–W–B–O/20rGO, thus conrming that GO was
3.1 Material synthesis completely reduced to rGO by NaBH4.28
Transition-metal oxides can be synthesized by various routes FTIR spectra were recorded and examined to conrm the
such as solvothermal, electrodeposition, sol–gel, and physical composition of Co–Ni–W–B–O/20rGO. As shown in Fig. 2b, the
vapor deposition methods. In our study, we adapted the method bands located at 3410 and 3450 cm
1 are attributed to the O–H
of direct room-temperature chemical reduction, a fast synthesis bond in H2O on the surface of the samples. Meanwhile, the
technique that yields bulk quantities of transition-metal oxides. peak at 1100 cm
1 corresponds to the dissolution of CO2 from
Transition-metal oxides were previously fabricated by Li et al. air. The bands in the low wavenumber region (400–900 cm
1)
and Yang et al., who used the solvothermal method, but not the correspond to the stretching vibrations of the Ni–O, Co–O, and
direct room-temperature mixing approach.25,26 The Co–Ni–W–B W–O bonds. The bands at around 440 and 550 cm
1 and those
alloy in our study self-oxidized on the surface of rGO in air at
room temperature. The three-dimensional (3D) nanostructure
of the samples was well preserved aer oxidation. The process
of preparation of Co–Ni–W–B–O/rGO is illustrated in Fig. 1. It

Schematic diagram of the fabrication process of the flower-like

Fig. 1 Fig. 2 (a) XRD patterns and (b) FTIR spectra of the as-prepared
Co–Ni–W–B–O/20rGO composite. samples.

This journal is © The Royal Society of Chemistry 2017 J. Mater. Chem. A, 2017, 5, 9907–9916 | 9909
 View Article Online

Journal of Materials Chemistry A Paper

around 600 and 640 cm
1 are typical of Ni–O and Co–O results also revealed that Co, Ni, W, B, O, and C were uniformly
stretching vibrations, respectively, while the band at 820 cm
1 distributed throughout the entire composite (Fig. S1†). The TEM
corresponds to the W–O bond. These results indicate that the image in Fig. 3f shows that Co–Ni–W–B–O was well anchored on
elements Co, Ni, and W were oxidized in the composite. the surface of rGO. It also proves that the products consisted of
Furthermore, because of the amorphous structure and surface interconnected ower-like 3D nanosheets. The diameter of the
defects, the absorption bands of the metal–oxygen bond are ower was about 200 nm. In the corresponding selected-area
Published on 21 April 2017. Downloaded by National Institute of Technology - Rourkela on 3/8/2019 5:30:07 AM.

shied and broadened when compared with those of the crys- electron diffraction (SAED) pattern (inset of Fig. 3f), a broad and
talline phase reported in the literature.23,29–31 diffuse halo ring can be seen, suggesting that the Co–Ni–W–B–O/
rGO sample consisted of an amorphous phase, which is in
3.2 Composition and morphology of the synthesized Co–Ni– agreement with the XRD results.
W–B–O/rGO composites XPS analysis provided more detailed information on the
The composition and morphology of the samples were further chemical composition of the as-prepared nanocomposites. The
investigated by SEM and TEM. Fig. 3 shows the typical SEM and full spectrum in Fig. 4a shows that Co (14.03 at%), Ni (8.1
TEM images of the as-prepared samples. Thick nanoakes of at%), W (0.66 at%), O (54.4 at%), C (12.71 at%), and B (10.1 at%)
the pure Co–Ni–W–B–O alloy were formed (Fig. 3a and b), were present in the Co–Ni–W–B–O/20rGO sample. To conrm
whereas the Co–Ni–W–B–O/20rGO sample displayed more that O did not originate from rGO alone, the full XPS spectrum
ower-like and dispersive features, which indicate that rGO of rGO was also recorded (Fig. S2†). The atomic ratio of C/O was
effectively decreased the agglomeration and reduced the 1.75 in the rGO, which indicates that O in Co–Ni–W–B–O/20rGO
thickness of the Co–Ni–W–B–O nanoakes (Fig. 3c and d). The mainly originated from the metal oxides. To further understand
Co–Ni–W–B–O/20rGO sample exhibited a special hierarchical the electronic states of the elements, we paid more attention to
porous structure composed of vast nanosheets; the pores were the higher-resolution spectra.
formed as a result of the release of abundant H2 during The high-resolution Co 2p XPS spectrum in Fig. 4b shows
reduction. The nanosheets gathered to form the porous struc- two main peaks with binding energies of 780.7 eV (Co 2p3/2) and
ture, which provided the products with a high specic surface 796.7 eV (Co 2p1/2), with an energy separation of 6 eV, which is
area and a large pore volume. characteristic of the Co3O4 phase.32 The satellite peaks are
EDX analysis (Fig. 3e) indicated that the Co–Ni–W–B–O/20rGO located about 6 eV above the main peaks. The complex Co 2p
nanoparticles consisted of 16.9 wt% Co, 9.3 wt% Ni, 3.9 wt% W, spectrum indicates the presence of two chemically distinct
48.7 wt% O, 18.4 wt% C, and 2.8 wt% B. While B originated from species, Co2+ and Co3+.31 Similar results were also observed in
the reducing agent, NaBH4, the high O content indicates that the
surface of the sample was thoroughly oxidized. The EDX mapping

Fig. 3SEM images of (a and b) Co–Ni–W–B–O and (c and d) Co–Ni– Fig. 4 (a) Full XPS spectrum of Co–Ni–W–B–O/20rGO. Sections of
W–B–O/20rGO. (e) EDX spectrum of Co–Ni–W–B–O/20rGO. (f) TEM the XPS spectrum corresponding to various electronic states: (b) Co
image of Co–Ni–W–B–O/20rGO. 2p; (c) Ni 2p; (d) W 4f; (e) O 1s; and (f) C 1s.

9910 | J. Mater. Chem. A, 2017, 5, 9907–9916 This journal is © The Royal Society of Chemistry 2017
 View Article Online

Paper Journal of Materials Chemistry A

the Ni 2p spectrum (Fig. 4c): two spin–orbit doublets at 873.5 3.4 Electrochemical properties of the Co–Ni–W–B–O/20rGO
and 855.7 eV corresponding to Ni 2p1/2 and Ni 2p3/2, respec- electrodes
tively; two obvious satellite peaks located at 880.1 and
Next, the electrochemical properties of the samples were
861.6 eV,33,34 which are assigned to species containing Ni2+ and
investigated to evaluate their applicability as active materials in
Ni3+ ions, respectively. The valence states of W atoms can be
supercapacitors. A three-electrode cell with an SCE reference
obtained by deconvolving the W 4f core-level spectrum (Fig. 4d)
Published on 21 April 2017. Downloaded by National Institute of Technology - Rourkela on 3/8/2019 5:30:07 AM.

electrode and a Pt counter electrode was used for the evaluation,

into two spin–orbit doublets. The main peaks at 35.4 and
and a 6 M KOH solution was used as the electrolyte. For
37.6 eV correspond to the W 4f7/2 and W 4f5/2 lines, respectively, comparison, the CV curves of Co–Ni–B, Co–W–B, Co–Ni–W–B,
of W atoms in the +6 oxidation state,35 indicating that W was in Co–Ni–W–B–O and Co–Ni–W–B–O/20rGO are shown in Fig. 6a.
the form of WO3 in the composite. The O 1s core-level region in
These CV curves were recorded in the potential window from
Fig. 4e consists of a broad peak centered at 532 eV, which is

0.2 to 0.5 V and at a scan rate of 10 mV s
1. The redox peaks
associated with bound hydroxide groups (OH–) and probably
are clearly visible in the CV curves, revealing distinct pseudo-
the formation of Co or Ni hydroxides.36 The B 1s spectrum
capacitive characteristics that are associated with the faradaic
consists of a peak at 191.6 eV, which corresponds to B2O3 in the
redox reactions related to M–O, where M represents Ni, Co,
composite (Fig. S3†).37 These results demonstrate that Co–Ni–
or W.41 It should be noted here that B2O3 is not electrochemi-
W–B was well oxidized in air and the oxides in the composite cally active in the composite. However, it can play a buffering
were mainly in the form of Co3O4, Ni3O4, WO3, and B2O3. The C role, which can help to accommodate the volume change during
1s spectrum (Fig. 4f) shows a dominant peak centered at
the charge and discharge processes.42 According to studies re-
284.7 eV due to sp2 C–C bonds, with two weak peaks at 286.8
ported in the literature,16,43–45 the faradaic reactions of the
and 288.9 eV corresponding to C–O bonds in the epoxy and
surface oxycation species of the composite can be represented
ether groups and O–C]O species in the ketone and carboxylic
by the following equations:
groups in rGO, respectively,38 indicating that Co–Ni–W–B–O was
successfully doped with rGO. Co3O4 + OH
+ H2O ¼ 3CoOOH + e
, (5)

3.3 Structure and size of pores in the Co–Ni–W–B–O/rGO

composites
Brunauer–Emmett–Teller (BET) measurements were performed
to determine the porous structure and pore-size distribution of
the samples. All of the samples were degassed at 100  C in
a nitrogen ow for 12 h prior to the measurements. Fig. 5a
shows the nitrogen sorption isotherms of the samples. The
calculated specic area of Co–Ni–W–B–O/20rGO was 167.5 m2
g
1, which is about seven times higher than that of pure Co–Ni–
W–B–O (24.7 m2 g
1), indicating that the addition of rGO
substantially increased the surface area of Co–Ni–W–B–O. Re-
ported studies have proved that a larger specic surface area
provides more efficient contact between the electrolyte and
electrode materials, leading to a fast electrochemical kinetic
process.25 Fig. 5b shows the pore-size distribution of the
samples, which also proves the presence of ample meso-
pores.39,40 An abundance of mesopores plays an important role
in the process of electrolyte adsorption and transport. These
results intuitively reveal that Co–Ni–W–B–O/rGO had a hierar-
chical porous structure that would greatly contribute to its
electrochemical properties.

Fig. 6 (a) CVs of Co–Ni–W–B–O/20rGO, Co–Ni–W–B–O, Co–Ni–

W–B, Co–Ni–B, Co–W–B and Co–B electrodes at 10 mV s
1. (b)
Galvanostatic charge–discharge curves of Co–Ni–W–B–O/20rGO,
Co–Ni–W–B–O, Co–Ni–W–B, Co–Ni–B, Co–W–B and Co–B elec-
trodes at a current density of 1 A g
1. (c) Galvanostatic charge–
discharge curves of Co–Ni–W–B–O/rGO samples containing
different amounts of rGO. (d) CVs of Co–Ni–W–B–O/20rGO samples
prepared in ethanol and water. (e) Galvanostatic charge–discharge
curves of Co–Ni–W–B–O/20rGO samples prepared in ethanol and
Fig. 5 (a) Nitrogen sorption isotherms and (b) pore-size distribution water. (f) Galvanostatic charge–discharge curves of Co–Ni–W–B–O/
curves of Co–Ni–W–B–O and Co–Ni–W–B–O/20rGO. 20rGO treated at different temperatures.

This journal is © The Royal Society of Chemistry 2017 J. Mater. Chem. A, 2017, 5, 9907–9916 | 9911
 View Article Online

Journal of Materials Chemistry A Paper

CoOOH + OH
¼ CoO2 + H2O + e
, (6) in the following order: Ni–W–B/rGO < Co–W–B/rGO < Co–Ni–B/
rGO < Co–Ni–W–B/rGO < Co–Ni–W–B–O/rGO. These results
Ni3O4 + OH
+ H2O ¼ 3NiOOH + e
, (7) further conrm the successful introduction of transition
elements, which are known to effectively improve the surface
NiOOH + H2O + e
¼ Ni(OH)2 + OH
, (8) area and decrease the resistance of the alloy.27
The specic capacitance of Co–Ni–W–B–O/rGO was found to
Published on 21 April 2017. Downloaded by National Institute of Technology - Rourkela on 3/8/2019 5:30:07 AM.

WO3 + e
+ 3H2O ¼ [W(OH)6]
. (9)

increase with the rGO content in the composite (Fig. 6c). When
the weight of rGO was increased from 10 to 20 mg, the specic
capacitance of the resulting Co–Ni–W–B–O/rGO composite
The area of the Co–Ni–W–B–O/20rGO electrode (deduced
increased from 848.7 to 1189.1 F g
1 at a current density of 1 A
from the shape of the CV curve) is higher than the areas of the
g
1 because the higher amount of rGO facilitated charge
other electrodes, suggesting that the Co–Ni–W–B–O/20rGO
transport and energy storage. However, further increase of the
electrode possessed a greater capacitance than the others. The
rGO content to 30 mg led to decreased capacitive performance,
values of specic capacitance derived from the CV curves at
which might be attributed to the low surface area of the
a scan rate of 10 mV s
1 are shown in Table S1.† Meanwhile,
aggregated rGO nanoparticles. The nitrogen adsorption
GCD tests were conducted to compare the capacitive ability of
isotherms of Co–Ni–W–B–O/rGO composites with different
the different Co–Ni–W–B–O/rGO electrodes at the same current
amounts of rGO are shown in Fig. S5.† The calculated specic
density within the potential range from
0.2 to 0.35 V, as
surface areas of the samples with different amounts of rGO are
illustrated in Fig. 6b. It can be seen that the proles of all the
listed in Table S2.† It can be clearly seen that the Co–Ni–W–B–O/
curves have obvious discharge voltage plateaus in the discharge
20rGO sample had the highest surface area.
process, and these plateaus match well with the reduction peaks
Fig. S6 and S7† show the effect of the Co/Ni and (Co + Ni)/W
observed in the CV curves (Fig. 6a). This phenomenon indicates
molar ratio on the specic capacitance of the obtained mate-
that the samples exhibited pseudocapacitive behavior.33,41,46 The
rials, respectively. The sample prepared with a Co/Ni molar ratio
Co–Ni–W–B–O/20rGO electrode showed a longer discharge time
of 2 and a (Co + Ni)/W molar ratio of 3 showed the highest
than the other electrodes, demonstrating that it had an
specic capacitance. Therefore, the as-prepared samples dis-
enhanced capacitance. The values of specic capacitance of the
cussed in the above-mentioned studies were based on this ratio.
Co–B, Co–W–B, Co–Ni–B, Co–Ni–W–B, Co–Ni–W–B–O, and Co–
To investigate the effect of solvents on the electrical perfor-
Ni–W–B–O/20rGO electrodes were calculated from the
mance of the Co–Ni–W–B–O/20rGO electrode, two samples were
discharge curves to be 482.6, 531.3, 656.7, 753.9, 831.6, and
prepared under identical conditions, except the metal salts were
1189.1 F g
1, respectively, at a current density of 1 A g
1, which
dissolved with water for one sample and with ethanol for the
are in good agreement with the values obtained from the CV
other sample. As shown in Fig. 6d, a pair of redox peaks with
curves. The Co–Ni–W–B electrode exhibited better electro-
a potential separation of about 150 mV was observed for the Co–
chemical performance than the Co–B, Co–W–B and Co–Ni–B
Ni–W–B–O/20rGO sample prepared in water, whereas the Co–
electrodes, indicating that the combination of several metals
Ni–W–B–O/20rGO sample prepared in ethanol showed a much
created a synergistic effect to enhance the performance of the
higher anodic peak and a broad cathodic response. The specic
transition-metal alloy.27 Furthermore, the capacitance of the
capacitance of the Co–Ni–W–B–O/20rGO sample prepared in
Co–Ni–W–B electrode was substantially increased aer self-
water was calculated from the discharge curves in Fig. 6e to be
oxidation in air. The rGO-doped Co–Ni–W–B–O (Co–Ni–W–B–
758.2 F g
1 at a current density of 1 A g
1, which is much lower
O/20rGO) electrode had the highest specic capacitance when
than that of the Co–Ni–W–B–O/20rGO sample prepared in
compared to the other electrodes at the same current density. In
ethanol (1189.1 F g
1 at a current density of 1 A g
1). In our
general, the electrochemical performance of electrode materials
previous work, it was demonstrated that a metal alloy prepared
depends on their nature, porosity, crystalline structure,
in an ethanol solution by chemical reduction possesses a much
morphology, and conductivity.13 The higher specic capaci-
higher surface area than an alloy prepared in water.47 One
tance of the Co–Ni–W–B–O/20rGO electrode was obviously the
possible reason was the low solubility of the metal salts in the
result of its ower-like 3D nanostructure, which could expose
ethanol solution, which limited the reaction rate during the
the majority of the atoms to the outside with high activity and
initial stage. The hydrogen gas was conned in the pores, and
provide a high-speed pathway for ionic transportation. There-
the nanoporous structure was formed. Therefore, the improved
fore, the surface-dependent electrochemical reaction would be
electrochemical performance of the Co–Ni–W–B–O/20rGO
signicantly enhanced, thus improving the electrochemical
sample prepared in ethanol can be attributed to the nano-
performance. At the same time, the incorporation of rGO into
porous structure of the metal alloy formed on the surface of
the metal oxides could substantially improve electrical
rGO. Furthermore, rGO had better dispersity in ethanol than in
conductivity, thus facilitating the transport of charge carriers.
water, which reduced the aggregation of the metal alloy nano-
To better understand the synergistic effect between the transi-
particles. Interestingly, the Co–Ni–W–B–O/20rGO sample
tion elements, the capacitance of Co–Ni–B/rGO, Co–W–B/rGO,
prepared in ethanol was highly active and easily oxidized in air
Ni–W–B/rGO, Co–Ni–W–B/rGO, and Co–Ni–W–B–O/rGO (ob-
at room temperature; the experimental data should provide
tained from the charge/discharge curves at a current density of 1
useful information on the fabrication of high-performance
A g
1) was also compared (Fig. S4†). The capacitance increased
supercapacitors.

9912 | J. Mater. Chem. A, 2017, 5, 9907–9916 This journal is © The Royal Society of Chemistry 2017
 View Article Online

Paper Journal of Materials Chemistry A

To further investigate the electrochemical performance of 15 A g
1. Using eqn (2), the specic capacitance was calculated
Co–Ni–W–B–O/20rGO calcined at different temperatures (200– to be 1189.1, 1098.2, 923.6, 827.3, 759.2, and 668.1 F g
1 at
500  C), we compared the XRD patterns of these samples a discharge current density of 1, 2, 3, 5, 10, and 15 A g
1,
(shown in Fig. S8†). Obviously, all of the patterns showed respectively. As expected, the specic capacitance decreased
weakening diffraction peaks as all the samples gradually when the current density increased (Fig. 7c).
formed a crystalline state with increasing temperature. Fig. 6f In general, the capacitance degrades more severely at
Published on 21 April 2017. Downloaded by National Institute of Technology - Rourkela on 3/8/2019 5:30:07 AM.

displays the GCD curves of ve samples at 1 A g
1. As shown by a higher charge/discharge current, which is largely due to the
the discharge time, the specic capacitance decreased stepwise voltage drop commonly reported in many studies.23,33 The
with increasing calcination temperature. These results indicate electrochemical performance of the Co–Ni–W–B–O/20rGO
that the Co–Ni–W–B–O/20rGO sample with an amorphous electrode is compared with those of previously reported elec-
structure exhibited superior electrochemical performance when trodes in Table 1.
compared to the crystalline samples. The specic capacitance of Co–Ni–W–B–O/20rGO is consid-
Fig. 7a shows representative CV curves, obtained at different erably larger than those of most of the materials based on
scan rates, of Co–Ni–W–B–O/20rGO prepared in ethanol by transition-metal oxides. Fig. 7d shows the Ragone plot (power
chemical reduction. The oxidation peaks shied toward higher density versus energy density) derived from the charge–
potentials and the reduction peaks shied toward lower discharge curves of the Co–Ni–W–B–O/20rGO electrode. The
potentials with a large potential separation when the scan rate synergistic effect of the ower-like Co–Ni–W–B–O 3D nano-
was increased. However, the shapes of these CV curves particles and graphene sheets can be clearly seen. The Co–Ni–
remained unchanged when the scan rate was increased from 10 W–B–O electrodes delivered a high energy density of 49.9 W h
to 100 mV s
1, indicating the good capacitive behavior and the kg
1 at a power density of 256.7 W kg
1, and an energy density
high-rate capability of Co–Ni–W–B–O/20rGO. Fig. 7b presents of 28.1 W h kg
1 at a high power density of 18 100 W kg
1.
the galvanostatic charge–discharge curves in the potential range Charge–discharge cycling was also carried out to investigate
from
0.2 to 0.35 V at different current densities between 1 and the electrochemical stability of the as-fabricated Co–Ni–W–B–O/
20rGO electrode in a concentrated alkaline electrolyte. The
results are shown in Fig. 7e. The test was performed for 10 000
cycles in 6 M KOH at a high discharge current of 15 A g
1 in the
potential window between
0.2 and 0.35 V. The Co–Ni–W–B–O/
20rGO electrode demonstrated remarkable cycling stability. The
overall capacitance retention dropped slightly and retained
80.7% of the original value even aer 10 000 cycles. Further-
more, the EIS results indicate that the electrochemical resis-
tance of the electrode only slightly increased aer 10 000 cycles
(Fig. 7f). The calculated charge-transfer resistance of the Co–Ni–
W–B–O/20rGO electrode before and aer the cycling was 3.8
and 5.1 U, respectively, indicating its good stability. This
phenomenon was expected owing to the high conductivity of
graphene, and the addition of graphene into the 3D nano-
structure would certainly signicantly improve charge transfer.
The direct adhesion of Co–Ni–W–B–O on the graphene sheet
enabled fast electron transport, which in turn improved its
pseudocapacitive behavior.22 These results strongly demon-
strate that the interaction between graphene and Co–Ni–W–B–
O, and the chemical stability and mechanical exibility of gra-
phene sheets provided a suitable substrate for the nucleation of
Co–Ni–W–B–O and prevented the agglomeration of the
nanoparticles.
To investigate the potential application of the Co–Ni–W–B–
O/20rGO nanocomposite, we further constructed an asym-
metric supercapacitor by pairing the Co–Ni–W–B–O/20rGO
Fig. 7 (a) CV curves of Co–Ni–W–B–O/20rGO at various scan rates.
(b) Charge–discharge curves of the Co–Ni–W–B–O/20rGO electrode positive electrode with an rGO negative electrode (denoted as
at current densities between 1 and 15 A g
1. (c) Dependence of specific Co–Ni–W–B–O/20rGO//rGO). The mass between the positive
capacitance values on the current density. (d) Ragone plot (power and negative electrodes was determined according to the well-
density versus energy density) of the Co–Ni–W–B–O/20rGO elec- known charge balance equation (Q+ ¼ Q
). Q is equal to C 
trode, derived from the charge–discharge curves at different current
DV  m. The optimal mass ratio between the Co–Ni–W–B–O/
densities. (e) Cycle life of the Co–Ni–W–B–O/20rGO electrode at
a current density of 15 A g
1; the electrode was operated for 10 000 20rGO and rGO electrodes is estimated to be m+/m
¼ 0.38
cycles. (f) Electrochemical impedance spectra of the Co–Ni–W–B–O/ from the CV curves (Fig. S9†) according to eqn (10):56
20rGO electrode before and after 10 000 cycles.

This journal is © The Royal Society of Chemistry 2017 J. Mater. Chem. A, 2017, 5, 9907–9916 | 9913
 View Article Online

Journal of Materials Chemistry A Paper

Table 1 Comparison of the electrochemical performance of the Co–Ni–W–B–O/20rGO electrode with performances of previously reported
electrodes

Electrode materials Current density (A g
1) Specic capacitance (F g
1) Electrolyte Ref.

Ni/rGO 1 547.3 2 M KOH 19

NiO–rGO 1 1016.6 6 M KOH 20
Published on 21 April 2017. Downloaded by National Institute of Technology - Rourkela on 3/8/2019 5:30:07 AM.

NiCoMn LDH/rGO 1 912 2 M KOH 25

(rGO)–NiO 1 576 2 M KOH 26
NiCo2O4/C 5 667 2 M KOH 33
N-Doped carbon/Co3O4 0.5 978.9 6 M KOH 43
Co3O4@CoWO4/rGO 0.5 386 6 M KOH 48
Graphene sheet/NiO 2 400 1 M KOH 49
3D Co3O4–rGO aerogel 0.5 660 6 M KOH 50
WO3$H2O/rGO 1 244 1 M H2SO4 51
rGO@(Co, Mn)3O4 1 1004 6 M KOH 52
Co3O4/nanowires 1 642 3 M KOH 53
NiO–graphene 2 540 2 M KOH 54
Ni–Co oxides/C–TNAs 2 934.9 2 M KOH 55
Co–Ni–W–B–O/20rGO 1 1189.1 6 M KOH This paper

m+/m
¼ C
 DV
/(C+  DV+) (10) impedance spectrum of the asymmetric supercapacitor.

Noticeably, the calculated charge-transfer resistance of the
Specically, the loading mass of the positive and negative device was 5.8 U, indicating that the as-fabricated asymmetric
electrode active materials is 8 and 21 mg, respectively. They supercapacitor device had a relatively low internal resistance.
were galvanostatically charged and discharged in a two- Furthermore, the Co–Ni–W–B–O/20rGO//rGO asymmetric
electrode setup over a wide voltage range of 0–1.6 V supercapacitor demonstrated remarkable cycling stability. The
(Fig. S10†). Specic capacitances were calculated for all of the overall capacitance retention dropped slightly and retained
active materials in both electrodes. A high specic capacitance 79.4% of the original value even aer 10 000 cycles (Fig. S13†).
of 124.1 F g
1 was obtained at a current density of 0.5 A g
1. The Given the simplicity of its preparation and low costs, it can be
specic energy density and power density of the Co–Ni–W–B–O/ concluded that the Co–Ni–W–B–O/20rGO nanocomposite is
20rGO//rGO asymmetric supercapacitor are shown in a Ragone a rather promising material for supercapacitors.
plot (Fig. S11†). It delivered a high energy density of 44.4 W h
kg
1 at a power density of 435.4 W kg
1 and an energy density of
25.3 W h kg
1 at a high power density of 4389 W kg
1. As shown 4. Conclusions
in Fig. 8, the maximum value of energy density is higher than Co–Ni–W–B–O/rGO composites were successfully synthesized
those of most asymmetric supercapacitors.57–62 An EIS spectrum through simple one-pot chemical reduction in an ethanol solu-
was obtained to further understand the electrochemical tion followed by self-oxidation in air. The as-prepared material
behavior of the as-fabricated device. Fig. S12† shows the Nyquist exhibited an ultrahigh specic capacitance of 1189.1 F g
1 at 1 A
g
1. Electrochemical tests also conrmed its remarkable perfor-
mance, including good charge–discharge stability, excellent rate
capability, and long cycling life with 80.7% retention of the
capacitance aer 10 000 cycles. By pairing the Co–Ni–W–B–O/
20rGO positive electrode with an rGO negative electrode, it
delivered an energy density of 44.4 W h kg
1 at 435.4 W kg
1.
These results clearly demonstrate that the amorphous Co–Ni–W–
B–O/rGO composite can be applied to high-performance super-
capacitors as well as other energy-storage devices.

Acknowledgements
The authors wish to express their gratitude and appreciation to
the National Science Foundation of China (51561006, 51461011,
51461010, 51401059, 51361005, 51371060, and U1501242), the
Guangxi Natural Science Foundation (2016GXNSFFA380012,
2014GXNSFBA118240, 2015GXNSFBA139208, and
Fig. 8Ragone plot of the Co–Ni–W–B–O/20rGO//rGO device 2014GXNSFAA118318), and the Innovation Project of GUET
compared with those in the literature. Graduate Education (2016YJCX23).

9914 | J. Mater. Chem. A, 2017, 5, 9907–9916 This journal is © The Royal Society of Chemistry 2017
 View Article Online

Paper Journal of Materials Chemistry A

25 M. Li, J. P. Cheng, F. Liu and X. B. Zhang, Chem. Phys. Lett.,

Notes and references
2015, 640, 5–10.
1 J. L. Yang, C. Zeng, F. X. Wei, J. Q. Jiang, K. Y. Chen and 26 Y. Y. Yang, Z. A. Hu, Z. Y. Zhang, F. H. Zhang, Y. J. Zhang,
S. H. Lu, Mater. Des., 2015, 83, 552–556. P. J. Liang, H.-Y. Zhang and H. Y. Wu, Mater. Chem. Phys.,
2 M. Li, S. H. Xu, Y. P. Zhu, P. X. Yang, L. W. Wang and 2012, 133, 363–368.
P. K. Chu, J. Alloys Compd., 2014, 589, 364–371. 27 N. Patel and A. Miotello, Int. J. Hydrogen Energy, 2015, 40,
Published on 21 April 2017. Downloaded by National Institute of Technology - Rourkela on 3/8/2019 5:30:07 AM.

3 Z. Zhu, Y. Hu, H. Jiang and C. Li, J. Power Sources, 2014, 246, 1429–1464.
402–408. 28 D. B. Mascagni, C. M. Miyazaki, N. C. da Cruz, M. L. de
4 Y. F. Tang, T. Chen, S. X. Yu, Y. Q. Qiao, S. C. Mu, Moraes, A. Riul Jr and M. Ferreira, Mater. Sci. Eng., C,
S. H. Zhang, Y. F. Zhao, L. Hou, W. W. Huang and 2016, 68, 739–745.
F. M. Gao, J. Power Sources, 2015, 295, 314–322. 29 R. Vijayalakshmi, M. Jayachandran and C. Sanjeeviraja, Curr.
5 Q. Wang, Y. Zou, C. Xiang, H. Chu, H. Zhang, F. Xu, L. Sun Appl. Phys., 2003, 3, 171–175.
and C. Tang, Ceram. Int., 2016, 42, 12129–12135. 30 X. Sun, G. K. Wang, H. T. Sun, F. Y. Lu, M. P. Yu and J. Lian, J.
6 P. C. Banerjee, D. E. Lobo, R. Middag, W. K. Ng, Power Sources, 2013, 238, 150–156.
M. E. Shaibani and M. Majumder, ACS Appl. Mater. 31 J. M. Luo, B. Gao and X. G. Zhang, Mater. Res. Bull., 2008, 43,
Interfaces, 2015, 7, 3655–3664. 1119–1125.
7 G. Binitha, M. S. Soumya, A. A. Madhavan, P. Praveen, 32 Y. Zou, I. A. Kinloch and R. A. Dryfe, ACS Appl. Mater.
A. Balakrishnan, K. R. V. Subramanian, M. V. Reddy, Interfaces, 2015, 7, 22831–22838.
S. V. Nair, A. S. Nair and N. Sivakumar, J. Mater. Chem. A, 33 V. Veeramani, R. Madhu, S.-M. Chen and M. Sivakumar, ACS
2013, 1, 11698–11704. Sustainable Chem. Eng., 2016, 4, 5013–5020.
8 Y. Li, H. Xie, J. Wang and L. Chen, Mater. Lett., 2011, 65, 403– 34 M. Yang, H. Cheng, Y. Gu, Z. Sun, J. Hu, L. Cao, F. Lv, M. Li,
405. W. Wang, Z. Wang, S. Wu, H. Liu and Z. Lu, Nano Res., 2015,
9 J. W. Lee, T. Ahn, J. H. Kim, J. M. Ko and J. D. Kim, 8, 2744–2754.
Electrochim. Acta, 2011, 56, 4849–4857. 35 S. Cong, Y. Tian, Q. Li, Z. Zhao and F. Geng, Adv. Mater.,
10 K. Qiu, H. Yan, D. Zhang, Y. Lu, J. Cheng, W. Zhao, C. Wang, 2014, 26, 4260–4267.
Y. Zhang, X. Liu, C. Cheng and Y. Luo, Electrochim. Acta, 36 U. M. Patil, J. S. Sohn, S. B. Kulkarni, S. C. Lee, H. G. Park,
2014, 141, 248–254. K. V. Gurav, J. H. Kim and S. C. Jun, ACS Appl. Mater.
11 Y. Li, Y. Zhang, Y. Li, Z. Wang, H. Fu, X. Zhang, Y. Chen, Interfaces, 2014, 6, 2450–2458.
H. Zhang and X. Li, Electrochim. Acta, 2014, 145, 177–184. 37 U. B. Demirci and P. Miele, Phys. Chem. Chem. Phys., 2014,
12 R. Kumar, H.-J. Kim, S. Park, A. Srivastava and I. K. Oh, 16, 6872–6885.
Carbon, 2014, 79, 192–202. 38 R. Kumar, R. K. Singh, P. K. Dubey, D. P. Singh and
13 R. Kumar, R. K. Singh, A. R. Vaz and S. A. Moshkalev, RSC R. M. Yadav, ACS Appl. Mater. Interfaces, 2015, 7, 15042–
Adv., 2015, 5, 67988–67995. 15051.
14 T. Li, G. H. Li, L. H. Li, L. Liu, Y. Xu, H. Y. Ding and T. Zhang, 39 C. Yin, C. Yang, M. Jiang, C. Deng, L. Yang, J. Li and D. Qian,
ACS Appl. Mater. Interfaces, 2016, 8, 2562–2572. ACS Appl. Mater. Interfaces, 2016, 8, 2741–2752.
15 Y. Zuo, J. J. Ni, J. M. Song, H. L. Niu, C. J. Mao, S. Y. Zhang 40 X. S. Yin, C. H. Tang, L. Y. Zhang, Z. G. Yu and H. Gong, Sci.
and Y. H. Shen, Appl. Surf. Sci., 2016, 370, 528–535. Rep., 2016, 6, 21566.
16 H. N. Ma, J. He, D. B. Xiong, J. S. Wu, Q. Q. Li, V. Dravid and 41 Y. Li, L. Cao, L. Qiao, M. Zhou, Y. Yang, P. Xiao and Y. Zhang,
Y. F. Zhao, ACS Appl. Mater. Interfaces, 2016, 8, 1992–2000. J. Mater. Chem. A, 2014, 2, 6540–6548.
17 H. Chen, L. Hu, M. Chen, Y. Yan and L. Wu, Adv. Funct. 42 L. Wen, X. Qin, W. Meng, N. Cao and Z. Song, Mater. Sci.
Mater., 2014, 24, 934–942. Eng., B, 2016, 23, 63–68.
18 G. Liu, B. Wang, L. Wang, Y. Yuan and D. Wang, RSC Adv., 43 Y. Zheng, Z. Li, J. Xu, T. Wang, X. Liu, X. Duan, Y. Ma,
2016, 6, 7129–7138. Y. Zhou and C. Pei, Nano Energy, 2016, 20, 94–107.
19 L. Niu, J. Wang, W. Hong, J. Sun, Z. Fan, X. Ye, H. Wang and 44 M. Yang, F. Lv, Z. Wang, Y. Xiong, M. Li, W. Wang, L. Zhang,
S. Yang, Electrochim. Acta, 2014, 123, 560–568. S. Wu, H. Liu, Y. Gu and Z. Lu, RSC Adv., 2015, 5, 31725–
20 P. Cao, L. Wang, Y. Xu, Y. Fu and X. Ma, Electrochim. Acta, 31731.
2015, 157, 359–368. 45 N. M. Shinde, A. D. Jagadale, V. S. Kumbhar, T. R. Rana,
21 W. Yang, Z. Gao, J. Ma, J. Wang, X. Zhang and L. Liu, Dalton J. Kim and C. D. Lokhande, Korean J. Chem. Eng., 2015, 32,
Trans., 2013, 42, 15706–15715. 974–979.
22 D. T. Dam, X. Wang and J. M. Lee, ACS Appl. Mater. Interfaces, 46 G. Xiong, P. He, L. Liu, T. Chen and T. S. Fisher, J. Mater.
2014, 6, 8246–8256. Chem. A, 2015, 3, 22940–22948.
23 C. Long, M. Zheng, Y. Xiao, B. Lei, H. Dong, H. Zhang, H. Hu 47 C. Xiang, J. Cheng, Z. She, Y. Zou, H. Chu, S. Qiu, H. Zhang,
and Y. Liu, ACS Appl. Mater. Interfaces, 2015, 7, 24419–24429. L. Sun and F. Xu, RSC Adv., 2015, 5, 163–166.
24 Y. M. Wang, X. Zhang, C. Y. Guo, Y. Q. Zhao, C. L. Xu and 48 X. Xu, Y. Yang, M. Wang, P. Dong, R. Baines, J. Shen and
H. L. Li, J. Mater. Chem. A, 2013, 1, 13290–13300. M. Ye, Ceram. Int., 2016, 42, 10719–10725.
49 X. Xia, J. Tu, Y. Mai, R. Chen, X. Wang, C. Gu and X. Zhao,
Chem.–Eur. J., 2011, 17, 10898–10905.

This journal is © The Royal Society of Chemistry 2017 J. Mater. Chem. A, 2017, 5, 9907–9916 | 9915
 View Article Online

Journal of Materials Chemistry A Paper

50 L. Xie, F. Su, L. Xie, X. Li, Z. Liu, Q. Kong, X. Guo, Y. Zhang, 56 L. Gao, J. U. Surjadi, K. Cao, H. Zhang, P. Li, S. Xu, C. Jiang,
L. Wan, K. Li, C. Lv and C. Chen, Chemsuschem, 2015, 8, J. Song, D. Sun and Y. Lu, ACS Appl. Mater. Interfaces, 2017, 9,
2917–2926. 5409–5418.
51 L. Ma, X. Zhou, L. Xu, X. Xu, L. Zhang, C. Ye, J. Luo and 57 X. Wang, W. Liu, X. Lu and P. S. Lee, J. Mater. Chem., 2012,
W. Chen, Colloids Surf., A, 2015, 481, 609–615. 22, 23114–23119.
52 C. Lamiel, V. H. Nguyen, C. Roh, C. Kang and J. J. Shim, 58 J. Xiao and S. Yang, J. Mater. Chem., 2012, 22, 12253–12262.
Published on 21 April 2017. Downloaded by National Institute of Technology - Rourkela on 3/8/2019 5:30:07 AM.

Electrochim. Acta, 2016, 210, 240–250. 59 K. Xu, R. Zou, W. Li, Q. Liu, X. Liu, L. An and J. Hu, J. Mater.
53 Y. Liang, Y. Yang and Z. Hu, Int. J. Electrochem. Sci., 2016, 11, Chem. A, 2014, 2, 10090–10097.
4092–4109. 60 X. Wang, J. Liu, Y. Wang, C. Zhao and W. Zheng, Mater. Res.
54 C. Ge, Z. Hou, B. He, F. Zeng, J. Cao, Y. Liu and Y. Kuang, J. Bull., 2014, 52, 89–95.
Sol-Gel Sci. Technol., 2012, 63, 146–152. 61 J. Ji, L. Zhang, H. Ji, Y. Li, X. Zhao, X. Bai, X. Fan, F. Zhang
55 C. Yu, Y. Wang, J. Zhang, X. Shu, J. Cui, Y. Qin, H. Zheng, and R. S. Ruoff, ACS Nano, 2013, 7, 6237–6243.
J. Liu, Y. Zhang and Y. Wu, New J. Chem., 2016, 40, 6881– 62 R. Ding, L. Qi, M. Jia and H. Wang, Electrochim. Acta, 2013,
6889. 107, 494–502.

9916 | J. Mater. Chem. A, 2017, 5, 9907–9916 This journal is © The Royal Society of Chemistry 2017