www.advancedsciencenews.com www.advenergymat.

de

Rational Design of Metal-Organic Framework Derived

Full paper
Hollow NiCo2O4 Arrays for Flexible Supercapacitor
and Electrocatalysis
Cao Guan,* Ximeng Liu, Weina Ren, Xin Li, Chuanwei Cheng,* and John Wang*

both supercapacitors and electrocatalysis

Metal-organic frameworks (MOFs) are promising porous precursors for the due to the fact that they can provide rich
construction of various functional materials for high-performance electrochem- reaction sites and short ion diffusion
ical energy storage and conversion. Herein, a facile two-step solution method length as well as release the strain during
the electrochemical reactions, thus main-
to rational design of a novel electrode of hollow NiCo2O4 nanowall arrays on taining the integrity of the electrodes.[11–19]
flexible carbon cloth substrate is reported. Uniform 2D cobalt-based wall- For example, a novel hollowed cubic NiS
like MOFs are first synthesized via a solution reaction, and then the 2D solid nanoframe electrode was achieved by
nanowall arrays are converted into hollow and porous NiCo2O4 nanostructures anion exchange of a Ni–Co MOF with
through an ion-exchange and etching process with an additional annealing S2− ions, which demonstrated enhanced
capacitive performance for electrochem-
treatment. The as-obtained NiCo2O4 nanostructure arrays can provide rich
ical capacitors.[20] Aijaz et al. report a Co@
reaction sites and short ion diffusion path. When evaluated as a flexible elec- Co3O4 encapsulated in N-doped carbon
trode material for supercapacitor, the as-fabricated NiCo2O4 nanowall electrode nanotube-grafted carbon polyhedral elec-
shows remarkable electrochemical performance with excellent rate capability trode derived from a cobalt MOF as a
and long cycle life. In addition, the hollow NiCo2O4 nanowall electrode exhibits bifunctional electrocatalyst with excellent
promising electrocatalytic activity for oxygen evolution reaction. This work catalytic performance under alkaline con-
ditions.[21] Despite the progress achieved
provides an example of rational design of hollow nanostructured metal oxide to date, most of the reported MOF-based/
arrays with high electrochemical performance and mechanical flexibility, derived materials are limited to micro-/
holding great potential for future flexible multifunctional electronic devices. nanoparticles. In addition, most of the pre-
mentioned MOF-derived hollow structures
are in powder forms, which include the
1. Introduction unwanted polymer binder and carbon addictive in electrochem-
ical reaction.[22] These “dead mass” can block the active sites
Metal-organic frameworks (MOFs) have been regarded as a and the resulted inflexibility of the electrodes can set a big bar-
promising class of advanced materials for gas adsorption, rier for the potential usage in flexible electronics.[23–27] As such,
sensing, supercapacitors, and catalysis, owing to their crystal- it would be of great interest to develop 2D MOFs and MOFs
line structures with high porosities and tunable functionali- derived arrayed structures on flexible current collector, which
ties.[1–6] In particular, due to their porous structures with high may offer unique physical and chemical properties with larger
surface areas, MOFs are ideal sacrificial templates to derive surface area and more accessible active sites.[28–33]
into various porous active materials, such as carbonaceous Herein, we report a novel approach to fabricate hollow and
materials, metal oxides, and metal phosphides.[7–10] Hollow porous NiCo2O4 nanowall arrays on flexible carbon cloth cur-
and porous electrode materials have shown great potential for rent collector. This strategy involves in the construction of 2D
cobalt-based MOF solid nanowall arrays on carbon cloth and
an ion-exchange and etching reaction process with Ni(NO3)2 in
ethanol followed by a thermal treatment. This unique electrode
Dr. C. Guan, X. M. Liu, X. Li, Prof. J. Wang
Department of Materials Science and Engineering design processes several advantages as follows. First, the highly
National University of Singapore hollow and porous structures allow the intimate electrolyte pen-
117574, Singapore etration and fast ion/mass transport. Second, the flexible and
E-mail: [email protected]; [email protected] conductive carbon cloth with tight connection to the NiCo2O4
W. N. Ren, Prof. C. W. Cheng arrays can be directly used as electrode without any binder
Shanghai Key Laboratory of Special
Artificial Microstructure Materials and Technology
additives, which can ensure high mechanical stability and elec-
School of Physics Science and Engineering tric conductivity. Third, the unique 2D nanostructures exhibit
Tongji University high surface area and shortened ion diffusion length, which
Shanghai 200092, P. R. China is beneficial to improve the electrochemical performance. As
E-mail: [email protected] a demonstration, a flexible supercapacitor is assembled with
DOI: 10.1002/aenm.201602391 the hollow NiCo2O4 nanoarrays as cathode, which achieves

Adv. Energy Mater. 2017, 7, 1602391 © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com (1 of 8)  1602391
 www.advenergymat.de www.advancedsciencenews.com

Section for details). First, a facile solution

Full paper

method was carried out to grow cobalt-based

2D MOF solid nanowall arrays (noted as
Co-MOF) on a flexible carbon cloth substrate.
Then, through a subsequent ion exchange/
etching process using Ni(NO3)2 in ethanol,
the Co-MOF will be etched and simultane-
ously Ni–Co layered double-hydroxide (noted
as Ni–Co LDH) will deposit on the surface,
resulting into a hollow wall structure. In this
step, the hollowing strategy for the nano-
Figure 1.  Schematic illustration of the formation of hollow NiCo2O4 nanostructure from 2D structure can be described as a hydrolysis-
Co-MOF solid nanowalls. controlled ion exchange and etching process
(Ni2+ + 2H2O = Ni(OH)2 + 2H+, where the
a high gravimetric density of 31.9 W h kg−1 at 2.9 kW kg−1, a H+ will etch the MOF and promote the deposition of Ni–Co
maximum power density of 27.3 W h kg−1 at 22.9 kW kg−1, and LDH), which is efficient to create a central void space.[34,35]
excellent cycling ability. Moreover, the electrocatalytic property After a thermal treatment in air, the layered double-hydroxide
of the hollow NiCo2O4 nanowall arrays for oxygen evolution will be decomposed and become porous, thus highly hollow
reaction is evaluated, which also exhibits superior performance and porous as well as crystallized NiCo2O4 nanostructures can
in terms of low onset potential (1.49 V vs reversible hydrogen be obtained on carbon cloth substrate (noted as CC@NiCo2O4),
electrode (RHE)) and excellent stability. which can be directly used as a flexible electrode for both super-
capacitor and electrocatalysis.
The 2D Co-MOF with nanowall morphology are prepared
2. Results and Discussion on flexible carbon cloth via a modified solution method.[36]
The scanning electron microscopy (SEM) observations in
2.1. Material Fabrication and Characterizations Figure 2a,b indicate the uniform coverage of the 2D Co-MOF
on carbon cloth surface. From an enlarged view in Figure 2c,
The synthetic strategy for the hollow NiCo2O4 nanowall arrays one can see the solid nature of these Co-based nanowalls with
is schematically depicted in Figure 1 (see the Experimental smooth surfaces. The X-ray diffraction (XRD) patterns of the

Figure 2.  SEM images of a–c) 2D Co-MOF nanowall, d–f) Ni–Co LDH hollow nanostructure, and g–i) NiCo2O4 hollow nanostructure. The arrows in
(f) and (i) indicate the hollow feature of the materials.

1602391  (2 of 8) wileyonlinelibrary.com © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Energy Mater. 2017, 7, 1602391
www.advancedsciencenews.com www.advenergymat.de

Co-MOF in Figure S1a (Supporting Information) confirm that without noticeable signals of residues, and its corresponding

Full paper
the crystalline Co-MOF precursors have a layered phase EDX result (Figure S1d, Supporting Information) reveals a Co/
without any impurities,[36,37] and the corresponding energy- Ni atomic ratio of about 1.8, further confirming the formation
dispersive X-ray spectroscopy (EDX) spectrum (Figure S1b, of NiCo2O4 nanoarrays.
Supporting Information) verifies the existence of N and C ele- The structure evolution from solid Co-MOF nanowalls to
ments from the organic ligand. Figure 2d–e shows the typical hollow NiCo2O4 nanowalls is further revealed by transmis-
morphology of Co-MOF after the transformation in the pres- sion electron microscopy (TEM) characterizations. As shown
ence of Ni(NO3)2. It can be observed that the overall 2D struc- in Figure 3a, the 2D Co-MOF precursor has a solid feature
tures and the uniform coverage of the materials on the carbon with typical height of ≈1 µm. After the reactions with Ni(NO3)2
cloth are maintained after the reactions, and small nanoflakes (Figure 3b,c), a hollow structure can be clearly observed with
appear on the surfaces of the 2D walls. From an enlarged SEM sharp contrast between the grained shell and the central void
image in Figure 2f, we can even see the 2D structures become space, indicating the formation of Ni–Co LDH nanostructures.
“semitransparent” (marked with an arrow), indicating that the From a magnified observation in Figure 3c, we can see that the
hollow structures have been formed. The full transformation surface of the hollow nanowall has been uniformly entangled
of the Co-MOF is also revealed by the disappeared Co-MOF with small nanoflakes (≈50 nm in length). After annealing the
peaks in XRD patterns and the lack of N signal in EDX spectra Ni–Co LDH nanostructures (Figure 3d), the hollow structure
(Figure S1c, Supporting Information). It is worth mentioning can be well preserved. As displayed in Figure 3e, the surface of
that short-time etching process will result in insufficient etching the nanostructures is composed of many interconnected nano-
and semihollow structures, and excess dosage of Ni(NO3)2 will particles with small diameters of ≈10 nm. High-resolution TEM
lead to the overgrowth of nanoflakes and severe etching with image of the hollow nanostructure is shown in Figure 3f, from
the destruction of the 2D arrays (Figure S2, Supporting Infor- which the lattice fringes of 0.286 and 0.244 nm match well with
mation). After annealing the Ni–Co LDH nanostructure in air, the (220) and (311) planes of NiCo2O4 (JCPDS card: 20-0781).[40]
the nanoarrays have been completely maintained (Figure 2g,h). The selected area electron diffraction (SAED) patterns (inset in
As shown in Figure 2i, it can be seen that the hollow feature Figure 3f) further demonstrate the cubic crystalline structure
is clearly revealed by the top open area, and the surface of the of the NiCo2O4 nanostructures (JCPDS card: 20-0781), noticing
hollow structure has been covered with interlinked ultrathin that no apparent signals of NiO or Co3O4 can be detected from
nanoflakes, indicating that hollow and porous nanowall arrays the high-resolution TEM (HRTEM) and SAED.
have been formed on the carbon cloth. All the XRD peaks of the The X-ray photoelectron spectroscopy (XPS) measurement
hollow nanoarrays (Figure S1a, Supporting Information) can is further carried out to study the chemical composition of the
be assigned to the NiCo2O4 phase (JCPDS card: 20-0781)[38,39] hollow NiCo2O4 nanoarrays, as shown in Figure 4. Broad scan

Figure 3.  TEM images of a) 2D Co-MOF, b,c) Ni–Co LDH hollow nanostructure, and d,e) NiCo2O4 hollow nanostructure. f) HRTEM image of hollow
NiCo2O4 nanostructure with the SAED pattern inserted.

Adv. Energy Mater. 2017, 7, 1602391 © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com (3 of 8)  1602391
 www.advenergymat.de www.advancedsciencenews.com

(b) annealing the Co-MOF precursor in air is

Full paper

(a) CC@NiCo2O4 O 1s

Ni 2p
also tested, and detailed characterization

Co 2p
of the solid Co3O4 nanowalls can be found
Intensity (a.u.)

Intensity (a.u.)
in Figure S3 (Supporting information).

O 1s
The representative cyclic voltammetry (CV)
curves of the CC@NiCo2O4 at scan rates of
5–100 mV s−1 are provided in Figure S4a

C 1s
(Supporting information). Apparent redox
peaks can be observed from the CV curves,
1200 1000 800 600 400 200 0 536 532 528 which can be assigned for the Faradic reac-
Binding Energy (eV) Binding Energy (eV) tion between Ni2+/Ni3+, Co3+/Co4+ with
OH −.[44–46] From the CV curves of CC@
(c) Ni 2p Ni 2p3/2 (d) Co 2p Co 2p3/2 NiCo2O4, the peak positions only shift a
Ni 2p1/2 little when the scan rate increased from 5 to
s s Co 2p1/2
Intensity (a.u.)

Intensity (a.u.)
100 mV s−1, indicating that the electrode has
s very good reversibility. The galvanostatic
s charge–discharge (GCD) curves of the CC@
CoNi2O4 in the same voltage range are illus-
trated in Figure S4b (Supporting informa-
tion). All the charge–discharge curves have
890 880 870 860 850 810 800 790 780 770 almost the same voltage plateaus, matching
Binding Energy (eV) Binding Energy (eV) well with the CV results. Figure 5a com-
pares the GCD curves of the CC@CoNi2O4
Figure 4.  XPS spectra of CC@NiCo2O4. a) Broad scan, b) O 1s, c) Ni 2p, and d) Co 2p of the and CC@Co3O4 at a same current den-
sample. sity of 10 mA cm−2. The discharge time of
CC@NiCo2O4 is longer than that of CC@
of the CC@NiCo2O4 shows the element of C, Co, Ni, O without Co3O4, indicating that the capacitance is improved (the CV and
any other impurities (Figure 4a). The O1 s spectra in Figure 4b GCD curves of the CC@Co3O4 can be found in Figure S4c,d
can be well fitted with three subpeaks: the peak at 529.6 eV is in the Supporting Information). Noting that the carbon cloth
the typical metal–oxygen bonds, and the peaks at 531.3 and contributes very little (less than 0.5%) to the total capacitance
531.8 eV correspond well to the defects with low oxygen and (Figure S4e,f, Supporting Information). Figure 5b shows the
adsorbed water at or near the surface, respectively.[41,42] The Co specific capacitance as a function of the current density for the
2p spectrum in Figure 4c can be fitted into both Co2+ and Co3+ two electrodes. At a small current of 2.5 mA cm−2, the CC@
each accompanied with a shakeup satellite, and the Ni 2p spec- CoNi2O4 exhibits a high capacitance of 1055.3 F g−1 and a high
trum in Figure 4d can be also best fitted into Ni2+ and Ni3+ each capacitance of 483.3 F g−1 can be maintained even when the
accompanied with a shakeup satellite. These results indicate current density increased to 60 mA cm−2, suggesting that 45.8%
that the surface of the NiCo2O4 nanostructures contains Co , 2+ of the capacitance can be remained when the current increased
Co3+, Ni2+, and Ni3+, which is in good agreement with previous 24 times. On the other hand, the CC@Co3O4 delivers a capaci-
results on NiCo2O4.[41,43] tance of 823.4 F g−1 at 5 mA cm−2, and it drops to 44.3% when
the current density increased 16 times. The enhanced capaci-
tance and rate capability of the CC@NiCo2O4 can be attributed
2.2. Supercapacitive Property to the unique hollow and porous structures that can provide suf-
ficient reaction cites and short ion diffusion length. To confirm
Based on the above discussions, a novel hollow NiCo2O4 this, the Brunauer–Emmett–Teller (BET) surface areas of the
nanoarray directly grown on carbon cloth substrate can be two electrodes have been tested, and the CC@NiCo2O4 shows
realized from a 2D Co-MOF precursor. The resulted hollow a surface area of 11.6 m2 g−1, which is larger than that of CC@
and porous NiCo2O4 nanostructures can not only provide suf- Co3O4 (8.4 m2 g−1). In addition, the direct contact between the
ficient reaction sites, short ion diffusion length, buffer the active materials with the carbon support can also ensure elec-
volume change, but also ensure direct mechanical and electrical tron transport highway for fast reaction at high rates. To further
connection with the flexible carbon support, making it a very demonstrate the potential application of the CC@NiCo2O4 for
promising electrode material for flexible energy storage and flexible energy storage devices, a flexible asymmetric superca-
electrocatalysis applications. pacitor was assembled using the CC@NiCo2O4 as the cathode
To evaluate the electrochemical performance, the CC@ and a nitrogen-doped carbon flakes on carbon cloth (CC@NC)
NiCo2O4 is first investigated as a supercapacitor electrode using as the anode with a poly(vinyl alcohol) (PVA)–KOH solid-state
a three-electrode system. A Pt plate is used as the counter electrolyte[47] (detailed characterizations of CC@NC are shown
electrode, and an Hg/HgO is used as the reference electrode. in Figure S5 in the Supporting Information). The CV curves of
A 2 m KOH (potassium hydroxide) aqueous solution is used the CC@NiCo2O4//CC@NC full cell are shown in Figure S6a
as the electrolyte. For comparison, the solid Co3O4 nanowalls (Supporting Information), in which quasirectangular curves
on carbon cloth (noted as CC@Co3O4) obtained by directly without sharp redox peaks can be observed, showing that the

1602391  (4 of 8) wileyonlinelibrary.com © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Energy Mater. 2017, 7, 1602391
www.advancedsciencenews.com www.advenergymat.de

2.3. Oxygen Evolution Reaction

Full paper
As the NiCo2O4 material is a good oxygen
evolution catalyst for water splitting,[51–53]
the self-supported NiCo2O4 hollow arrays
can provide large electroactive surface area
without the usage of binders, and the elec-
tron transfer can be also enhanced by direct
electronic contact. The flexible electrode also
provides potential opportunity to be inte-
grated with other functional flexible elec-
tronic devices. The electrocatalytic activity
of as-fabricated hollow NiCo2O4 nanoarrays
was also evaluated. The tests were conducted
in a typical three-electrode cell system with
1 m KOH (pH = 14) electrolyte. A Pt plate is
used as the counter electrode, and a saturated
calomel electrode (SCE) is used as the refer-
ence electrode. The electrochemical imped-
ance spectroscopy (EIS) of the samples are
first recorded and the Nyquist plots of CC@
NiCo2O4 and CC@Co3O4 are presented in
Figure 5. a) Comparison of GCD curves of CC@NiCo2O4 and CC@Co3O4 at 10 mA cm−2. Figure 6a. In the high-frequency region,
b) The plots of specific capacitances versus current densities for CC@NiCo2O4 and CC@Co3O4.
the intercept of X axis for CC@NiCo2O4 is
c) Ragone plots of the CC@NiCo2O4//CC@NC full cell. The values reported from other MOF-
based supercapacitors are inserted for comparison. d) Cycling test result of the full cell. Digital smaller than that for CC@Co3O4, showing
images of the full cell under flat, bending, and twisting states during the test are inserted. that the bulk resistance of the hollow elec-
trode is smaller. The radius of the semicircle
of CC@NiCo2O4 is also smaller than that
full cell has good capacitive properties. The GCD curves of of CC@Co3O4, showing that the charge transfer resistance of
the full cell are shown in Figure S6b (Supporting Informa- the hollow structure is smaller. In the low-frequency region, a
tion), where all the curves have almost linear charge–discharge more vertical line for the CC@NiCo2O4 is observed, indicating
characteristics. At a current density of 5 mA cm−2, the asym- that the NiCo2O4 nanoarrays have lower ionic diffusion resist-
metric supercapacitor delivers a capacitance of ≈89.7 F g−1, and ance than the solid Co3O4 electrode. Figure 6b shows the linear
≈76.8 F g−1 can be maintained when the cur-
rent density is increased eight times, showing
excellent kinetic behavior (Figure S6c, Sup-
porting Information). Figure 5c shows the
Ragone plot (power density (P) vs energy den-
sity (E)) of the CC@NiCo2O4//CC@NC, and
data obtained from other reported MOF-based
solid-state supercapacitors are also inserted
for comparison. From the Ragone plot, the
CC@NiCo2O4//CC@NC full cell express a
maximum energy density of 31.9 W h kg−1 at
a power density of 2.9 kW kg−1, and it main-
tains 27.3 W h kg−1 at a peak power density
of 22.9 kW kg−1, which is comparable to the
previously reported MOF-based supercapaci-
tors.[9,44,48–50] Figure 5d further shows the
excellent cycling stability and mechanical flex-
ibility of the full cell, that 86.7% of the initial
capacitance can well preserved after 20 000
cycles of charge–discharge at 5 mA cm−2,
even when the full cell was bended/twisted
during the test (inset in Figure 5d). The SEM
image (Figure S6d, Supporting Information)
of the CC@NiCo2O4 after cycling tests shows Figure 6.  a) Nyquist plots, b) polarization curves, and (c) Tafel plots of CC@NiCo O and CC@
2 4
that the structure has been well maintained, Co3O4. d) Stability and flexibility test of CC@NiCo2O4 at 1.57 V versus RHE with typical flat and
further confirming the stable cycling ability. bending status as inserted.

Adv. Energy Mater. 2017, 7, 1602391 © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com (5 of 8)  1602391
 www.advenergymat.de www.advancedsciencenews.com

sweeps of CC@NiCo2O4 and CC@Co3O4 at 5 mV s−1, from was then taken out, cleaned with deionized water, and vacuum dried
Full paper

which the onset potential of CC@NiCo2O4 is only ≈1.49 V overnight.

Material Synthesis—Preparation of CC@Ni–Co LDH: A piece of CC@
(vs RHE) and it is ≈30 mV lower than that of CC@Co3O4. In addi-
Co-MOF was immersed into an ethanol solution (100 mL) containing
tion, the NiCo2O4 nanoarrays exhibit higher current densities Ni(NO3)2·6H2O (0.3 g). After stirring the solution for 15 min, the sample
across the entire potential window than those for CC@Co3O4, was taken out, washed with ethanol, and dried at 60 °C. Decreasing the
indicating a clear higher catalytic activity of the hollow nano- reaction time to 5 min will result in insufficient etching, while prolonging
structure. Furthermore, to reach anodic current densities of the reaction time to 30 min will result in the overgrowth of nanoflakes.
10 mA cm−2, the potential required for CC@NiCo2O4 is only Material Synthesis—Preparation of CC@NiCo2O4: A piece of CC@Ni–
1.57 V, which is among the most active NiCo2O4 electrocata- Co LDH was annealed in air at 350 °C for 2 h with a ramp rate of
2 °C min−1. The mass loading of NiCo2O4 was ≈0.60 mg cm−2.
lysts[54–67] (a detailed comparison can be found in Table S1 in Material Synthesis—Preparation of CC@Co3O4: A piece of
the Supporting Information). The Tafel slope of CC@NiCo2O4 CC@Co-MOF was annealed in air at 350 °C for 2 h with a ramp rate of
is 72 mV per decade as shown in Figure 6c, which is a little 2 °C min−1. The mass loading of Co3O4 was ≈0.69 mg cm−2.
lower than that of CC@Co3O4 (75 mV per decade). The capaci- Material Synthesis—Preparation of CC@NC: A piece of CC@Co-MOF
tances of the two electrodes are further tested to illustrate the was annealed in Ar at 800 °C for 2 h with a ramp rate of 1 °C min−1, after
electrochemical active surface areas (Figure S7, Supporting which the sample was immersed in 3 m FeCl3 aqueous solution for 6 h,
washed with clean water, and dried at 60 °C. The mass loading of Co3O4
Information). The calculated electrochemical double-layer
was ≈0.8 mg cm−2.
capacitance of CC@NiCo2O4 (0.088 F cm−2) is ≈4.8 times larger Material Characterization: Samples were characterized using SEM
than that for CC@Co3O4 (0.015 F cm−2), further confirming (Zeiss, 5.0–20.0 kV), TEM (JEOL-2100F), XPS (AXIS Ultra), and XRD
the better electrocatalytic performance for the hollow NiCo2O4 (Bruker D8 diffractometer). The mass of electrode materials was also
nanoarrays. As such, the enhanced electrocatalytic property recorded by an AX/MX/UMX Balance (METTLER TOLEDO, maximum
of the CC@NiCo2O4 electrode can be due to the hollow and = 5.1 g; delta = 0.001 mg). To measure the surface area of the electrode
materials, nitrogen adsorption–desorption isotherms were measured
porous structures, which can provide rich reaction cites and
on a Micromeritics ASAP 2020 at 77 K. Before measurements, samples
short ion diffusion length. To further evaluate the stability and were outgassed at 120 °C for 6 h, and the specific surface areas were
flexibility of the CC@NiCo2O4 as an oxygen evolution reaction calculated using the BET method.
catalyst, the electrode was tested under a constant potential Electrochemical Characterization: Electrochemical measurements
(1.57 V vs RHE) for 20 h, with the first 10 h under flat status were conducted using Solartron 1470E at room temperature. (a)
and the last 10 h under bended status. As shown in Figure 6d, Supercapacitor test: For three-electrode system test, CC@NiCo2O4 (or
the current density of CC@NiCo2O4 remained flat during the CC@Co3O4 and CC@NC) was directly used as the working electrode
without any metal support or other current collector. A Pt plate and
20 h test, and the bending of the electrode did not influence an Hg/HgO were used as the counter electrode and the reference
much of the current density, indicating high electrochemical electrode, respectively. KOH (2 m) was used as the electrolyte.
stability. To further confirm the stability, the postmortem study The specific capacitance (C) of the electrodes was calculated by:
has been carried out (Figure S8, Supporting Information), from C = It/mV, where I is the discharge current, V is the discharge
which the structure of CC@NiCo2O4 is well maintained after voltage, t is the discharge time, and m is the active mass of the
the long-time test. electrode. For the full cell assembly, to optimize the charge between
the two electrodes, the mass of the two electrodes were balanced
by C− × m− × V− = C+ × m+ × V+, where C− and C+ are the specific
capacitances of the negative and positive electrodes, m− and m+ are
3. Conclusions the active mass loadings of the negative and positive electrodes, and V−
and V+ are the voltage window of the negative and positive electrodes,
In summary, we have designed and fabricated novel hollow respectively. To assemble the full cell, first a solid-state electrolyte
NiCo2O4 nanoarrays on flexible carbon cloth through a facile was fabricated by dissolving 6 g PVA into 30 mL H2O and then mixed
with 30 mL 4 m KOH solution at 85 °C, then the two electrodes were
two-step strategy with 2D Co-MOF solid nanowall arrays as pre-
immersed into the electrolyte and left under ambient conditions for
cursors and templates. Excellent electrochemical performances 2 min before they were assembled face-to-face, noticing that the polymer
both for flexible supercapacitor and electrocatalytic oxygen evo- electrolyte acted as both the electrolyte and the ion-porous separator.
lution reaction properties of the CC@NiCo2O4 electrode were After the electrolyte was solidified, a mechanically robust full cell could
demonstrated, which can be attributed to unique hollow and be packed and tested. The energy density (E) and power density (P) of
porous structures that provide rich reaction cites and short the full cell were calculated by: E = 21 CV 2, P = E/t, where C is the specific
ion diffusion length. This work opens new opportunities for capacitance of the full cell, V is the discharge voltage, t is the discharge
time, and m is the active mass of the electrodes. (b) Electrocatalytic
rational design of various complex hollow structure arrays for test: The catalytic activity was tested in a standard three-electrode
high-performance flexible electrochemical storage and conver- system in 1 m KOH solution (pH = 14) after purging with oxygen
sion devices. for 30 min, using a Pt plate and a SCE as the counter and reference
electrode, respectively. CC@NiCo2O4 (or CC@Co3O4) was directly
used as the working electrode. EIS measurements were carried out by
applying an AC voltage with 5 mV amplitude in a frequency range from
4. Experimental Section 0.1 Hz to 100 kHz at open circuit potential. The polarization curves were
Material Synthesis—Preparation of CC@Co-MOF: An aqueous solution measured at 5 mV s−1 and iR corrected, and all the measured potentials
containing 2-methylimidazole (C4H6N2, 40 mL, 0.4 m) was quickly added were referred to RHE using the following equation: E(RHE) = E(SCE)
into the aqueous solution of Co(NO3)2·6H2O (40 mL, 50 × 10−3 m), + 0.0592 × pH + 0.241. The electrochemical double-layer capacitance
after which a piece of clean CC substrate (2 × 5 × 0.036 cm3, the upper (Cdl) was determined from the CV curves measured in a potential range
side and back side protected with by polytetrafluoroethylene tape) was without redox process by: Cdl = I/ν, where I is the charging current
immersed into the mixture solution. After reaction for 4 h, the sample (mA cm−2), and ν is the scan rate (mV s−1).

1602391  (6 of 8) wileyonlinelibrary.com © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Energy Mater. 2017, 7, 1602391
www.advancedsciencenews.com www.advenergymat.de

Supporting Information [22] S. Wang, S. Dou, X. Li, L. Tao, J. Huo, Chem. Commun. 2016, 52,

Full paper
9727.
Supporting Information is available from the Wiley Online Library or [23] T. Y. Ma, J. Ran, S. Dai, M. Jaroniec, S. Z. Qiao, Angew. Chem., Int.
from the author. Ed. 2015, 54, 4646.
[24] J. Feng, S. Ye, H. Xu, Y. Tong, G. Li, Adv. Mater. 2016, 28, 4698.
[25] P. Chen, K. Xu, Z. Fang, Y. Tong, J. Wu, X. Lu, X. Peng, H. Ding,
C. Wu, Y. Xie, Angew. Chem., Int. Ed. 2015, 54, 14710.
Acknowledgements [26] Z. Liu, J. Xu, D. Chen, G. Shen, Chem. Soc. Rev. 2015, 44, 161.
This research was supported by the 973 Program (Grant No. [27] X. Lu, M. Yu, G. Wang, Y. Tong, Y. Li, Energy Environ. Sci. 2014, 7,
2013CB632701), the National Natural Science Foundation of China 2160.
(Grant No. 51202163). This project was also supported by Academic [28] T. Y. Ma, S. Dai, M. Jaroniec, S. Z. Qiao, J. Am. Chem. Soc. 2014,
Research Fund (AcRF) Tier 1 R-284-000-146-112, conducted at the 136, 13925.
National University of Singapore. [29] F. Cao, M. Zhao, Y. Yu, B. Chen, Y. Huang, J. Yang, X. Cao, Q. Lu,
X. Zhang, Z. Zhang, C. Tan, H. Zhang, J. Am. Chem. Soc. 2016, 138,
Received: October 28, 2016 6924.
Revised: December 1, 2016 [30] A. Gallego, C. Hermosa, O. Castillo, I. Berlanga, C. J. Gómez-García,
Published online: January 25, 2017 E. Mateo-Martí, J. I. Martínez, F. Flores, C. Gómez-Navarro,
J. Gómez-Herrero, S. Delgado, F. Zamora, Adv. Mater. 2013, 25,
2141.
[1] I. Stassen, M. Styles, G. Grenci, H. V. Gorp, W. Vanderlinden, [31] S. C. Junggeburth, L. Diehl, S. Werner, V. Duppel, W. Sigle,
S. D. Feyter, P. Falcaro, D. D. Vos, P. Vereecken, R. Ameloot, Nat. B. V. Lotsch, J. Am. Chem. Soc. 2013, 135, 6157.
Mater. 2015, 15, 304. [32] H. Tang, S. Cai, S. Xie, Z. Wang, Y. Tong, M. Pan, X. Lu, Adv. Sci.
[2] W.-J. Li, M. Tu, R. Cao, R. A. Fischer, J. Mater. Chem. A 2016, 4, 2016, 3, 1500265.
12356. [33] S. Wang, X. Wang, X. Li, C. Ouyang, Z. Li, S. Dou, Z. Ma, L. Tao,
[3] J. Tang, R. R. Salunkhe, J. Liu, N. L. Torad, M. Imura, S. Furukawa, J. Huo, J. Mater. Chem. A 2016, 4, 9370.
Y. Yamauchi, J. Am. Chem. Soc. 2015, 137, 1572. [34] J. Liu, Y. Li, H. Fan, Z. Zhu, J. Jiang, R. Ding, Y. Hu, X. Huang,
[4] D. Sheberla, J. C. Bachman, J. S. Elias, C.-J. Sun, Y. Shao-Horn, Chem. Mater. 2009, 22, 212.
M. Dinca, Nat. Mater. 2016, doi:10.1038/nmat4766. [35] J. Luo, X. Xia, Y. Luo, C. Guan, J. Liu, X. Qi, C. F. Ng, T. Yu,
[5] L. Wang, X. Feng, L. Ren, Q. Piao, J. Zhong, Y. Wang, H. Li, Y. Chen, H. Zhang, H. J. Fan, Adv. Energy Mater. 2013, 3, 737.
B. Wang, J. Am. Chem. Soc. 2015, 137, 4920. [36] R. Chen, J. Yao, Q. Gu, S. Smeets, C. Baerlocher, H. Gu, D. Zhu,
[6] K. M. Choi, H. M. Jeong, J. H. Park, Y. B. Zhang, J. K. Kang, W. Morris, O. M. Yaghi, H. Wang, Chem. Commun. 2013, 49, 9500.
O. M. Yaghi, ACS Nano 2014, 8, 7451. [37] G. Fang, J. Zhou, C. Liang, A. Pan, C. Zhang, Y. Tang, X. Tan, J. Liu,
[7] J. Liu, C. Wu, D. Xiao, P. Kopold, L. Gu, P. A. van Aken, J. Maier, S. Liang, Nano Energy 2016, 26, 57.
Y. Yu, Small 2016, 12, 2354. [38] Z. Gao, W. Yang, J. Wang, N. Song, X. Li, Nano Energy 2015, 13, 306.
[8] R. Bendi, V. Kumar, V. Bhavanasi, K. Parida, P. S. Lee, Adv. Energy [39] C. Yuan, J. Li, L. Hou, X. Zhang, L. Shen, X. W. Lou, Adv. Funct.
Mater. 2015, 6, 1501833. Mater. 2012, 22, 4592.
[9] R. R. Salunkhe, J. Tang, Y. Kamachi, T. Nakato, J. H. Kim, [40] L. Huang, W. Zhang, J. Xiang, H. Xu, G. Li, Y. Huang, Sci. Rep. 2016,
Y. Yamauchi, ACS Nano 2015, 9, 6288. 6, 31465.
[10] R. Chen, T. Zhao, T. Tian, S. Cao, P. R. Coxon, K. Xi, [41] B. Cui, H. Lin, Y. Liu, J. Li, P. Sun, X. Zhao, C. Liu, J. Phys. Chem. C
D. Fairen-Jimenez, R. Vasant Kumar, A. K. Cheetham, APL Mater. 2009, 113, 14083.
2014, 2, 124109. [42] C. Guan, Y. Wang, Y. Hu, J. Liu, K. H. Ho, W. Zhao, Z. Fan, Z. Shen,
[11] Y. Zhao, X. Li, J. Liu, C. Wang, Y. Zhao, G. Yue, ACS Appl. Mater. H. Zhang, J. Wang, J. Mater. Chem. A 2015, 3, 23283.
Interfaces 2016, 8, 6472. [43] R. P. Antony, A. K. Satpati, K. Bhattacharyya, B. N. Jagatap, Adv.
[12] A. Morozan, F. Jaouen, Energy Environ. Sci. 2012, 5, 9269. Mater. Interfaces 2016, 6, 1600632.
[13] J. Qi, X. Lai, J. Wang, H. Tang, H. Ren, Y. Yang, Q. Jin, L. Zhang, [44] D. Yu, B. Wu, L. Ge, L. Wu, H. Wang, T. Xu, J. Mater. Chem. A 2016,
R. Yu, G. Ma, Z. Su, H. Zhao, D. Wang, Chem. Soc. Rev. 2015, 44, 4, 10878.
6749. [45] N. Jabeen, Q. Xia, M. Yang, H. Xia, ACS Appl. Mater. Interfaces 2016,
[14] A. Mahmood, W. Guo, H. Tabassum, R. Zou, Adv. Energy Mater. 8, 6093.
2016, 6, 1600423. [46] L. Hou, R. Bao, Z. Chen, M. Rehan, L. Tong, G. Pang, C. Yuan, Elec-
[15] S. Wang, J. Wang, M. Zhu, X. Bao, B. Xiao, D. Su, H. Li, Y. Wang, trochim. Acta 2016, 214, 76.
J. Am. Chem. Soc. 2015, 137, 15753. [47] C. Guan, W. Zhao, Y. Hu, Q. Ke, X. Li, H. Zhang, J. Wang, Adv.
[16] L. Shang, H. Yu, X. Huang, T. Bian, R. Shi, Y. Zhao, G. I. Waterhouse, Energy Mater. 2016, 6, 1601034.
L. Z. Wu, C. H. Tung, T. Zhang, Adv. Mater. 2015, 28, 1668. [48] R. R. Salunkhe, C. Young, J. Tang, T. Takei, Y. Ide, N. Kobayashi,
[17] S. Mao, Z. Wen, T. Huang, Y. Hou, J. Chen, Energy Environ. Sci. Y. Yamauchi, Chem. Commun. 2016, 52, 4764.
2014, 7, 609. [49] X. Cao, B. Zheng, W. Shi, J. Yang, Z. Fan, Z. Luo, X. Rui, B. Chen,
[18] C. Sun, J. Yang, X. Rui, W. Zhang, Q. Yan, P. Chen, F. Huo, Q. Yan, H. Zhang, Adv. Mater. 2015, 27, 4695.
W. Huang, X. Dong, J. Mater. Chem. A 2015, 3, 8483. [50] P. Wen, Z. Li, P. Gong, J. Sun, J. Wang, S. Yang, RSC Adv. 2016, 6,
[19] W. Zeng, L. Wang, H. Shi, G. Zhang, K. Zhang, H. Zhang, F. Gong, 13264.
T. Wang, H. Duan, J. Mater. Chem. A 2016, 4, 8233. [51] C. Jin, F. Lu, X. Cao, Z. Yang, R. Yang, J. Mater. Chem. A 2013, 1,
[20] X. Y. Yu, L. Yu, H. B. Wu, X. W. Lou, Angew. Chem., Int. Ed. 2015, 12170.
54, 5331. [52] J. Bao, X. Zhang, B. Fan, J. Zhang, M. Zhou, W. Yang, X. Hu,
[21] A. Aijaz, J. Masa, C. Rösler, W. Xia, P. Weide, A. J. R. Botz, H. Wang, B. Pan, Y. Xie, Angew. Chem., Int. Ed. 2015, 54, 7399.
R. A. Fischer, W. Schuhmann, M. Muhler, Angew. Chem., Int. Ed. [53] J. Wang, Z. Wu, L. Han, R. Lin, H. L. Xin, D. Wang, ChemCatChem
2016, 55, 4087. 2016, 8, 736.

Adv. Energy Mater. 2017, 7, 1602391 © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim wileyonlinelibrary.com (7 of 8)  1602391
 www.advenergymat.de www.advancedsciencenews.com
Full paper

[54] H. Hu, B. Guan, B. Xia, X. W. Lou, J. Am. Chem. Soc. 2015, 137, [61] J. Wang, T. Qiu, X. Chen, Y. Lu, W. Yang, J. Power Sources 2014, 268,
5590. 341.
[55] Y. Li, P. Hasin, Y. Wu, Adv. Mater. 2010, 22, 1926. [62] W. Xia, N. Li, Q. Li, K. Ye, C. Xu, Sci. Rep. 2016, 6, 23398.
[56] R. Chen, H. Wang, J. Miao, H. Yang, B. Liu, Nano Energy 2015, 11, [63] C. Xiao, Y. Li, X. Lu, C. Zhao, Adv. Funct. Mater. 2016, 26, 3515.
333. [64] S. Chen, J. Duan, W. Han, S. Z. Qiao, Chem. Commun. 2014, 50,
[57] S. Chen, S. Z. Qiao, ACS Nano 2013, 7, 10190. 207.
[58] L. Liu, J. Wang, Y. Hou, J. Chen, H.-K. Liu, J. Wang, Y. Wu, Small [65] J. Geng, H. Wu, A. M. Al-Enizi, A. A. Elzatahry, G. Zheng, Nanoscale
2016, 12, 602. 2015, 7, 14378.
[59] X. Gao, H. Zhang, Q. Li, X. Yu, Z. Hong, X. Zhang, C. Liang, Z. Lin, [66] H. Cheng, Y. Su, P. Kuang, G. Chen, Z. Liu, J. Mater. Chem. A 2015,
Angew. Chem., Int. Ed. 2016, 55, 6290. 3, 19314.
[60] X. Lv, Y. Zhu, H. Jiang, X. Yang, Y. Liu, Y. Su, J. Huang, Y. Yao, C. Li, [67] J. Yin, P. Zhou, L. An, L. Huang, C. Shao, J. Wang, H. Liu, P. Xi,
Dalton Trans. 2015, 44, 4148. Nanoscale 2016, 8, 1390.

1602391  (8 of 8) wileyonlinelibrary.com © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Energy Mater. 2017, 7, 1602391