Coaxial NiO/Ni nanowire arrays for high performance pseudocapacitor applications

Article abstract of DOI:10.1016/j.electacta.2011.11.039

A new and low cost facile technique has been reported to fabricate coaxial nickel oxide/nickel (NiO/Ni) nanowire (NW) arrays as high performance electrodes for pseudocapacitors. The arrays are fabricated by direct electrochemical deposition of Ni inside nanoporous alumina template and the subsequent conversion of the outer shell into oxide by oxygen plasma annealing. This resulted in a unique core/shell, one-dimensional (1D) vertical nanostructure, which uses the inner Ni core as a conductor and robust support for a large effective surface area, and thereby provides reliable electrical connection to a thin redox-active NiO shell. The morphology, crystal structure, and shell thickness of NiO are found to be dependent on the annealing conditions (input power and plasma annealing time at a constant oxygen flow). The coaxial NiO/Ni NW arrays with the optimized redox-active NiO shell thickness displayed a high capacitance value of 0.36 F cm^-2. Furthermore, the arrays show a rapid charge-discharge kinetics and stable long-term cycling performance (practically no drop in capacitance was observed even after 2000 cycles at high charging-discharging current of 5 mA cm^-2). Excellent cycling stability and faster charge-discharge kinetics are attributed to the unique core/shell (NiO/Ni) and vertically aligned nanoarchitecture that facilitates faster electron and ion transport during the charge-discharge reactions.

Full text of DOI:10.1016/j.electacta.2011.11.039

Electrochimica Acta 60 (2012) 193–200
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Coaxial NiO/Ni nanowire arrays for high performance pseudocapacitor
applications
Maksudul Hasan, Mamun Jamal, Kafil M. Razeeb∗
Tyndall National Institute, University College Cork, Lee Maltings, Dyke Parade, Cork, Ireland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 August 2011
Received in revised form 2 November 2011
Accepted 9 November 2011
Available online 18 November 2011
A new and low cost facile technique has been reported to fabricate coaxial nickel oxide/nickel (NiO/Ni)
nanowire (NW) arrays as high performance electrodes for pseudocapacitors. The arrays are fabricated
by direct electrochemical deposition of Ni inside nanoporous alumina template and the subsequent con-
version of the outer shell into oxide by oxygen plasma annealing. This resulted in a unique core/shell,
one-dimensional (1D) vertical nanostructure, which uses the inner Ni core as a conductor and robust
support for a large effective surface area, and thereby provides reliable electrical connection to a thin
redox-active NiO shell. The morphology, crystal structure, and shell thickness of NiO are found to be
dependent on the annealing conditions (input power and plasma annealing time at a constant oxygen
flow). The coaxial NiO/Ni NW arrays with the optimized redox-active NiO shell thickness displayed a
high capacitance value of 0.36 F cm−2. Furthermore, the arrays show a rapid charge–discharge kinet-
ics and stable long-term cycling performance (practically no drop in capacitance was observed even
after 2000 cycles at high charging–discharging current of 5 mA cm−2). Excellent cycling stability and
faster charge–discharge kinetics are attributed to the unique core/shell (NiO/Ni) and vertically aligned
nanoarchitecture that facilitates faster electron and ion transport during the charge–discharge reactions.
Keywords:
Nickel nanowire
NiO
Nickel oxide
Coaxial
Supercapacitor
Plasma annealing
© 2011 Elsevier Ltd. All rights reserved.
1. Introduction
nanowires (NWs) or nanotubes (NTs), fabricated directly on to the
substrate may eliminate the use of ancillary materials, and fur-
The development of high-performance electrochemical energy
storage (EES) devices is crucial for the powering of contem-
porary and future electronic systems, including multifunctional
portable equipments, hybrid electric vehicles (HEVs), large indus-
trial equipment and the wireless sensor modules deployed in smart
environments and health systems. Designing efficient and minia-
turized EES devices that achieve high energy storage or delivery
at high charge and discharge rates and with life times matching
with the specific requirements of applications is one of the major
challenges facing today’s science and technology experts [1–3].
Electrochemical capacitors (ECs), also known as supercapacitors or
ultracapacitors have drawn much attention recently owing to much
higher power delivery or uptake (>10 kW kg−1) capability when
compared to batteries (<1 kW kg−1 for Li-ion). Nanostructured
pseudocapacitor electrodes are beneficial in asymmetric and/or
ther increase the active surface area and the energy capacity of the
electrodes. In recent years, there has been extensive research in
developing alternative pseudocapacitor electrode materials, such
as manganese oxide [6,7], cobalt oxide [8], nickel oxide [9–12], and
nickel hydroxide [13] either as porous and/or one-dimensional (1D)
nanostructures. NiO is one of the most promising materials owing
to its low cost, high specific capacitance and good capacity retention
[4]. Furthermore, it is environmentally friendly. However, nickel
oxide is an electrically poor conductive material (p-type semicon-
ductor). Therefore, in order to improve the intrinsic conductivity
and the maximum utilization of NiO, introduction of a conductive
metal as a coaxial supporting structure is essential.
In this work, we demonstrate the fabrication of coaxial nanowire
(NW) arrays of nickel oxide (NiO) and nickel (Ni) as high per-
formance electrodes for pseudocapacitors. The engineered coaxial
NiO/Ni NW arrays combining both high charge storage capacity
metal oxide (NiO) and conductive metal (Ni) enhanced the pseu-
docapacitance properties. Recently, oriented NiO–TiO2 NT arrays
were reported by electrochemical anodization of Ni–Ti alloy foils
followed by thermal annealing (at 600 ◦C) where Ti component in
the foil was used as the core supporting structure [10]. However,
the anodization of Ti or Ni–Ti films do not lead to the formation of
high aspect ratio and well porous nanostructures, as the reported
hybrid configurations due to their high specific capacitances (F g−1
)
and high volumetric capacitances (F cm−3) resulting from their high
packing density and large active surface areas, respectively [4,5].
In particular, one-dimensional (1D) nanostructured materials e.g.,
∗
Corresponding author. Tel.: +353 21 490 4078; fax: +353 21 427 0271.
E-mail address: kafil.mahmood@tyndall.ie (K.M. Razeeb).
0013-4686/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2011.11.039

194
M. Hasan et al. / Electrochimica Acta 60 (2012) 193–200
NT film’s thickness was only 200–500 nm (maximum aspect ratio
is 13:1 with pore diameter of 38 nm) with film’s porosity of 49%,
and the capacitance was measured to be 4 mF cm−2 [10,14]. Fur-
thermore, this alloy foil is quite expensive, and requires longer
processing time and complicated processing steps to fabricate the
final electrodes. In another recent report, NiO nanorod arrays were
fabricated by thermal hydrolysis of a mixture of urea and nickel
salts on Ni foam substrate at 100 ◦C for 12 h followed by annealing at
250–400 ◦C for 3 h to facilitate NiO crystallization to be able to show
pseudocapacitance [15]. Hence, we adopted a facile method for
the fabrication of the electrode by a simple electrodeposition of Ni
NWs using template and controlled conversion of the deposited Ni
NWs into NiO/Ni core/shell structure by low power oxygen plasma
annealing within a few minutes. Remarkably, the fabricated coax-
ial NiO/Ni NW arrays demonstrated high specific capacitance and
excellent capacity retention, which are important properties for
supercapacitor applications.
(HR-TEM, JEOL 2100 HRTEM instrument operating at an accel-
eration voltage of 200 kV) coupled with energy dispersive X-ray
spectroscopy (EDS Oxford Instruments INCA energy system). TEM
cross section samples were prepared using focused ion beam (FIB,
FEI DualBeam FIB Helios Nanolab 600i). FIB protective layers of
carbon and platinum was deposited onto a lamella by electron
beam induced deposition (EBID) and after deposition the lamella
was cut and attached to a TEM grid. The lamella was thinned
down to approximately 100–150 nm using the FIB at 30 kV. The
final polishing was done with 2 kV in order to minimize the FIB
induced damage to 2 nm per side. The targeted final thickness
was in the range of 30–60 nm. The cross sections were analyzed
by HR-TEM at 200 kV and bright field (BF) images were taken
with a medium sized objective aperture to increase the contrast.
EDS line scans were taken in scanning TEM (STEM) mode with
medium spot size (40–60 nm sphere diameter of analyzed vol-
ume). STEM analysis was done at 30 kV and dark field images
are taken in high angle annular dark field (HAADF) mode. In all
cases, the elements were identified by the INCA software. The X-
ray photo electron spectroscopy (XPS, Kratos AXIS-165, Mono A1
X-rays) was used to determine the chemical/oxidation state of
Ni in the NiO/Ni NWs. The electrochemical measurements of the
NiO/Ni nanowire arrays were performed using a CH Instruments
660C potentiostat. Electrodes comprised of 20 ␮m long NiO/Ni
NWs, arrayed in 1 cm × 1 cm surface areas and the electrochemi-
cal results are expressed in terms of the geometrical surface area.
Potentiostatic cyclic voltammetry (CV) and chronopotentiometry
(CP) were performed with a three-electrode glass cell setup filled
with 1 mol dm−3 KOH as an electrolyte, in which the active mate-
rial (NiO/Ni NW arrays) served as the working electrode, platinized
titanium mesh as the counter and Ag/AgCl/3 mol dm−3 KCl as the
reference electrode. CV measurements were carried out in the
potential range of 0.0–0.52 V (vs. Ag/AgCl) at different scan rates
of 5–50 mV s−1 and conditioned at the initial potential for 20 s. CP
charge–discharge measurements were carried out at a fixed current
2. Experimental
2.1. Synthesis of coaxial NiO/Ni nanowires
Alumina template (Anodisc 47, Whatman, nanopore diameter
of ∼250–300 nm, length of ∼60 ␮m and density of ∼1 × 109 cm−2
)
was purchased from GE Healthcare UK Ltd. A 400 nm layer of
Ni film was evaporated (Temescal FC-2000) onto one side of the
template which served as the working electrode. A copper wire
was connected to the Ni backside of the template by silver paste
(Radionics Ltd., Ireland) and left it over night to dry before use.
Low stress nickel sulfamate bath was prepared using nickel sulfa-
mate (0.37 mol dm−3), boric acid (0.64 mol dm−3), nickel bromide
(0.18 mol dm−3) and wetting agent ANKOR (R) F (10 ml l−1). The pH
of the solution was adjusted to 3.8 by adding 1 mol dm−3 sulfonic
acid at room temperature. The solution was stirred slowly at an
rpm of 100 and the deposition was conducted at a constant tem-
perature of 50 ◦C. A two electrodes cell was used with Ni pellets in
Ti-basket as an anode and Ni/porous alumina template as a working
electrode (cathode). A 50 mA cm−2 (as the ratio of the total template
area) constant current was applied to the working electrode and the
deposition rate was found to be 0.6 ␮m min−1. One-dimensional
vertical Ni NW arrays were released from the template by dissolv-
ing the template in 1 mol dm−3 KOH solution for 1 h, washed with
plenty of de-ionized water, dried in air. A thin film of Ni (3–4 ␮m)
was also deposited onto the back side of the seed layer while Ni
nanostructures grew inside the pores, which prevented the NWs
from collapsing after removal of the template. The fabricated Ni
NW arrays were annealed in plasma (March Plasmod GCM 200) for
different times at power input ranging from 25 to 50 W and at a con-
stant oxygen flow (30 cm3 min−1). The film of Ni NW arrays were
attached onto a silicon substrate by adhesive tape (Kapton tapes,
DuPontTM, USA) at the edge so that the back side does not expose
to the plasma and therefore, only the surface of Ni NWs was con-
verted to oxides. At the front side of the seed layer (NWs side), only
a little fraction of the seed could be exposed to the plasma owing
to very high density of NWs (∼1 × 109 cm−2 with wire diameter of
250–300 nm). The plasma treated samples were detached from the
Si substrate and cut into pieces and preserved for characterization.
density of 5 mA cm−2
.
2.3. Capacitance calculation
The capacitance (C) of the electrodes were calculated from CV
profile, which is equivalent to the sum of the integrated voltam-
metric charges divided by the same potential window, as expressed
by
Qa + Qc
C =
(1)
2A × ꢁV
where Qa is the anodic charge; Qc is the cathodic charge; A is the
surface area of the electrode; and ꢁV is the potential window.
For chronopotentiometry, the capacitance was derived from the
expression.
I × ꢁt
C =
(2)
ꢁV
where I is the current density during charging–discharging and ꢁt
is the time in each segment.
3. Results and discussion
NiO/Ni NW arrays were fabricated by direct electrochemical
deposition of Ni inside nanoporous alumina template and the sub-
sequent conversion of outer shell into oxide by oxygen plasma
annealing, as illustrated in Scheme 1. Fig. 1 shows the SEM images
of as deposited Ni NWs (Fig. 1(a)) and plasma annealed Ni NWs
at 50 W (watt) for 60 s (Fig. 1(b)) after removal from the template
and dispersion onto Au/Si substrate. The surface morphology of the
as deposited and plasma annealed NWs are different, where the
2.2. Characterization
The crystal structures of NiO/Ni NW were characterized by X-
ray diffraction (XRD, Philips PW3710-MPD diffractometer with Cu
K␣ radiation, ꢀ = 1.54 A). The morphology and nanostructures were
examined by scanning electron microscopy (SEM, FEI Nova 630
Nano-SEM) and high resolution transmission electron microscopy
˚

M. Hasan et al. / Electrochimica Acta 60 (2012) 193–200
195
Scheme 1. Schematic diagram shows the steps involved in the fabrication of coaxial NW arrays using the simple electrodeposition through nanoporous alumina template
and the subsequent conversion of the outer shell into oxide by oxygen plasma annealing after the template is removed.
surface roughness of the annealed samples are increased due to
oxide formation. A thin oxide layer is formed around the cir-
cumference of Ni NWs, and hence, oxygen plasma annealing is
found to be efficient in the conversion of the deposited metallic
Ni into NiO. Fig. 2 shows the effect of annealing time on the as
deposited vertically aligned Ni NWs. Fig. 2(a)–(c) show the SEM
images taken at a 45◦ perspective view for different annealing times
of 120 s, 300 s and 600 s, respectively. In the short time annealed
Ni NWs, the oxide layer thickness cannot be distinguished eas-
ily. So, a high magnification HR-TEM bright filed (BF) image has
been inserted into Fig. 2(d), where the presence of a thin oxide
layer at the outer surface of Ni is clearly visible. As the annealing
time of Ni NWs was increased while the input power was kept
constant at 50 W, the resulted oxide layer thickness around the
metallic Ni circumference increased gradually (see Fig. 2(d)–(f)).
Therefore, the annealing time (at a fixed input power of 50 W)
was varied to achieve an optimized thickness of the outer shell
oxide layer, so that it can maintain a good electrical connection
with the inner core support of the metallic Ni. When the anneal-
ing time was increased to 300 s (Fig. 2(b)) and 600 s (Fig. 2(c)), the
surface morphology of the resulted oxide layers become inhomo-
geneous compared to those annealed for shorter time (i.e., 120 s,
see Fig. 2(a)). In addition, Ni NWs annealed for 300 s or higher
are agglomerated as bundles of fused wires that consumed the
Fig. 1. SEM images after removal from the template and dispersion onto Au/Si substrate of (a) as deposited Ni NWs, and (b) NiO/Ni NWs that were prepared by oxygen
plasma annealing at 50 W for 60 s. SEM images clearly show the formation of oxide layer around the circumference of the Ni NWs.

196
M. Hasan et al. / Electrochimica Acta 60 (2012) 193–200
Fig. 2. SEM images at 45◦ perspective view of the NiO/Ni NW arrays obtained from metallic Ni NWs after oxygen plasma annealing at 50 W for 120 s (a), 300 s (b) and 600 s
(c). TEM-bright field (TEM-BF) images of the corresponding coaxial NiO/Ni NWs annealed for 120 s (d), 300 s (e) and 600 s (f) show the gradual increase in thickness of the
outer shell oxide layers and thinning of the inner core support of the metallic Ni.
Fig. 3. STEM-EDS elemental mapping shows the presence of a thin oxide layer around the Ni circumference annealed at 50 W for 600 s: dark field (DF) images of (a) coaxial
NiO/Ni NWs, (b) nickel at the inner core, and (c) oxygen at the outer shell.

M. Hasan et al. / Electrochimica Acta 60 (2012) 193–200
197
where change in the electrochemical behavior has been observed
with increasing the annealing time. The CV of the coaxial NiO/Ni
NW arrays annealed for 60 s shows an anodic peak at 0.294 V and
a cathodic peak at 0.193 V (vs. Ag/AgCl). The redox couple is asso-
ciated with the pseudocapacitive behavior of the NiO component
in NiO/Ni NW arrays. It results from the surface Faradaic oxida-
tion and reduction reactions (NiO + OH− ↔ NiOOH + e−), where the
anodic peak is due to the oxidation of NiO to NiOOH and the
cathodic peak is for the reverse process [9–11]. To compare, the
CV of the as deposited Ni NW arrays is shown in ESM section
(Fig. S4). However, after increasing the annealing time to 120 s,
the peak current intensities of the corresponding redox reactions
are also increased, and the peak potentials shift to higher values
at 0.41 V and 0.28 V, respectively (Fig. 6(a)). The increase in the
charge–discharge current intensity is attributed to the formation
of thicker oxide (i.e., larger quantity) around the circumference of
the Ni NWs. The sharp increase in the anodic and cathodic peak
currents also indicates that the outer shell of NiO film has higher
electrochemical activity, which maintains a good electrical con-
tact with the inner metallic Ni core. In the case of 180 s annealed
sample, the redox potentials shift further to the higher poten-
tials (oxidation peak at 0.465 V and reduction peak at 0.31 V) and
the peaks become wider, indicating low electrochemical activity
of the redox-active material. With increasing the annealing time,
the maximum peak potential separation (ꢁEp) increases (0.1 V for
60 s annealing to 0.13 V and 0.16 V for 120 and 180 s annealing,
respectively) while the redox peaks shift to the right side of the
scale. This observation implying that the redox reaction at or near
the surface of NiO/Ni is quasi-reversible for the samples annealed
for longer time, which may be attributed to the solution-electrode
and/or core–shell interface resistance [18,19]. The capacitance is
increased to 0.4 F cm−2 of the sample annealed for 180 s compared
to 0.38 F cm−2 and 0.21 F cm−2 of the samples annealed for 120 s
and 60 s, respectively (Fig. 6(b)). The capacitance initially increases
as more metallic Ni is converted to NiO, which is indicative of pseu-
docapacitive behavior of the redox-active NiO component in the
electrodes [18]. However, as the amount of NiO content increases
further (beyond 180 s annealing time), the capacitance decreases
(Fig. 6(b)) owing to the decrease in electrical conductivity with the
inner core (Ni) which slows down the rate of charge transport as
the thickness of NiO (shell) increases [18]. The low electrochemi-
cal activity of the electrodes annealed for longer time can also be
fairly attributed to the consumption of the interspaces between
wires as discussed earlier (Fig. 2 and Fig. S1 in ESM), which increas-
ingly reduces the total electroactive surface area for charge storage
and redox reactions. Therefore, these results can be summarized
as follows: (1) With increasing annealing time, more Ni turned into
NiO and thinning the conductive core of Ni, thereby electrical resis-
tance of the individual NW is increased as evident from shifting
of the redox peaks. (2) Also, the electrical conductivity between
outer shell NiO and inner core Ni metallic support might reduce
considerably. Therefore, thick and poorly conductive NiO in the
outer shell becomes practically in electrical isolation from the con-
ductive inner core Ni and this results in a decrease in capacitance
to 0.08 F cm−2 (annealed for 300 s) and 0.01 F cm−2 (annealed for
600 s).
Fig. 4. XRD patterns of the as electrodeposited Ni NW arrays (a), and after they were
annealed at 50 W for 120 s in oxygen plasma (b). The peaks marked by asterisks
correspond to the metallic Ni phases.
interspaces between wires and eventually reduced the total elec-
troactive surface area available for electrochemical reactions (see
Fig. S1 in Electronic supplementary material (ESM)). The SEM image
in Fig. 2(c) shows some charging effect, caused by the electron
being unable to ground through the thicker poorly conductive
NiO, formed at higher annealing time (600 s). Elemental mapping
by STEM-EDS (Scanning Transmission Electron Microscopy-Energy
Dispersive X-ray Spectroscopy) reveals the presence of both Ni and
O in the annealed Ni NWs (see Fig. 3).
Fig. 4 shows the XRD patterns of the electrodeposited Ni NWs
and after they were annealed at 50 W for 120 s in oxygen plasma.
XRD patterns of NiO/Ni NWs (see Fig. 4(a)) show three well-
resolved characteristic peaks of NiO that can be readily indexed
to (1 1 1, 2ꢂ = 37.3◦), (2 0 0, 2ꢂ = 43.3◦), and (2 2 0, 2ꢂ = 63◦) [16]. The
peaks designated by asterisks are associated with the metallic Ni
phases and shows no diffraction peaks for NiO in the XRD patterns
of the as deposited sample (see Fig. 4(b)). The characteristic peak
intensities of NiO are lower than that of Ni, since Ni peaks have also
been counted from the seed layer and thin film (3–4 ␮m) deposited
onto the back side of seed layer during the electrodeposition of
Ni NWs. When the annealing power was decreased to 25 W, the
characteristic peaks for NiO were not visible even after allowing
a higher annealing time of 600 s (see Fig. S2 in ESM). Hence, the
annealing power 50 W was found to be suitable for the conver-
sion of the deposited Ni into suitable oxide phases. Thereby, NiO/Ni
NW arrays obtained at 50 W for 120 s annealing were used for the
subsequent studies because of the homogeneous surface morphol-
ogy (see Fig. 2(a)) and the one-dimensional vertical nanostructures
being well preserved (see Fig. S1 in ESM). XPS profile of the NiO/Ni
NW was analyzed by focusing on the regions where the peaks of Ni
2p were expected. The principal component peak (3.0 eV FWHM)
at a binding energy of ∼855.8 eV is in good agreement with the
reported data of Ni 2p3/2 (see Fig. S3 in ESM) [17]. Thus, the results
evidently suggest that oxygen plasma annealing transforms the as
deposited Ni to uniform coaxial oxides (NiO/Ni). The HR-TEM image
in Fig. 5(a) shows domains of different crystallites or grains of shell,
and the high magnification image (see Fig. 5(b)) shows NiO crystal-
lites in the shell. Fig. 5(c) shows another area where the interface
between the Ni (core) and NiO (shell) is visible with lattice fringes
in the shell. Fig. 5(d) and (e) shows the EDS spectra of D1 (Ni core)
and D2 (NiO shell), respectively. The composition of these regions
is revealed from the EDS data, where Ni and O ratios are 6.5:3.5
(D2) and 19:1 (D1) i.e., showing the presence of small quantity of
O in the core of the annealed Ni NWs but a large quantity of O in
the shell.
Fig. 7(a) shows CVs of 120 s annealed NiO/Ni NW arrays at scan
rates between 5 and 50 mV s−1. The pair of anodic and cathodic
peak curves is symmetric over the entire range of scan rates, indi-
cating reversibility of the redox reactions at or near the coaxial
NiO/Ni NW surfaces. The peak current intensity increases linearly
with the scan rate (see Fig. S5 in ESM), implying that the redox
process associated with NiO active component in the NiO/Ni NW
system is diffusion controlled. This linear response of the peak
current intensity with the scan rate is also an indication of the
fast electronic and ionic transport rates. However, the maximum
The effect of annealing time (at a constant input power of 50 W)
on the CVs has been shown in Fig. 6(a) (at a scan rate of 10 mV s−1),

198
M. Hasan et al. / Electrochimica Acta 60 (2012) 193–200
Fig. 5. HR-TEM images of the coaxial NiO/Ni NW (annealed for 600 s at 50 W) show (a) NiO crystallites or grains in the shell, (b) magnified view of NiO crystallites, and (c)
magnified view of the interface of Ni and NiO showing the lattice fringes in the shell. EDS spectra of D1 (d) and D2 (e) correspond to domains of Ni (core) and NiO (shell).
Fig. 6. (a) Cyclic voltammograms of the annealed Ni NW arrays measured at the potential scan rate of 10 mV s−1 in 1 mol dm−3 KOH. The annealing time ranges from 60 s to
600 s at a constant input power of 50 W and oxygen flow of 30 cm3 min−1 in the plasma chamber. (b) Capacitance of the coaxial NiO/Ni NW arrays plotted as a function of
the annealed time.

M. Hasan et al. / Electrochimica Acta 60 (2012) 193–200
199
Fig. 7. (a) Cyclic voltammograms of 120 s oxygen plasma annealed NiO/Ni NW arrays measured at scan rates from 5 to 50 mV s−1. Inset shows CV taken at a low scan rate of
5 mV s−1. (b) Capacitance as a function of the scan rate.
peak potential difference (ꢁEp) between anodic and cathodic peak
increases (0.10 V for 5 mV s−1 to 0.22 V for 50 mV s−1) with the
scan rate implying that the redox reactions are quasi-reversible.
Fig. 7(b) shows the rate capability performance of the electrode
used in Fig. 7(a). The electrode shows good rate capability over
enough even at a high charging–discharging current, implying that
the electrode has fast charge–discharge properties with an excel-
lent electrochemical reversibility. A similar time length for the
charging and discharging indicates that reversible Faradaic redox
reactions occurred at or near the electroactive surfaces. The mea-
sured capacitance is 0.36 F cm−2 (using Eq. (2)), which complies
with the value calculated from CV measurements (using Eq. (1)) at
a high scan rate of 50 mV s−1 (see Fig. 7(b)). Fig. 8(b) shows the
cycling performance for the first 2000 cycles of the NiO/Ni NW
arrays illustrated in Fig. 8(a). The coaxial NiO/Ni NW arrays as
electrode displayed long-term cycling performance and no dete-
rioration in capacitance is observed over the entire cycling period.
The slight variation in capacitance may be attributed to the heating
effect, generated in the long range cycles or other physical fac-
tors rather than chemical or structural changes associated with the
electrode materials as confirmed by observing the samples in SEM
before and after the cycles. This core/shell nanoarchitecture pro-
vides a unique combination of entirely exposed, high aspect ratio,
one-dimensional vertical structures with a highly conductive metal
in the core, and thereby leads to homogeneous electrolyte accessi-
bility even to the remote sites. This resulted in a large electroactive
surface area for charge storage and redox reactions, and high rates
of electronic and ionic conductivity. We do believe that the capac-
itive performance of the coaxial NiO/Ni NW can be enhanced
further by optimizing the fabrication parameters. Moreover, these
core/shell nanostructures can be directly grown on thin conduc-
tive and flexible substrates, which is an efficient and cost-effective
the entire range of scan rates. At the scan rate of 50 mV s−1
,
the capacitance (0.36 F cm−2) of the NiO/Ni NW arrays retained
up to 94% of that measured at a lower scan rate of 5 mV s−1
(0.38 F cm−2). NiO/Ni NW arrays consisting of thinner oxide shell
e.g., annealed for very short time of 60 s, can be cycled even at
a higher scan rate of 100 mV s−1 without any decrease in capaci-
tance (see Fig. S6 in ESM). However, 60 s annealing time showed
almost 50% lower capacitance compared to 120 s samples as evi-
dent from Fig. 6(b). High capacitance retention at the higher scan
rate indicates the rapid electronic and ionic transport rates of the
Faradaic redox reactions. This faster charge–discharge kinetics can
be attributed to the unique core/shell nanoarchitecture with a
large electroactive surface area, that offers a highly conductive and
robust core for reliable connection to the NiO shell, which also
ensures easy electrolyte access to a large volume of redox-active
materials [10,18].
To further investigate the reaction kinetics and measure the
capacitance, a set of charge–discharge experiments were carried
out using chronopotentiometry with 120 s annealed NiO/Ni NW
arrays as electrode, at a constant current density of 5 mA cm−2
.
Fig. 8(a) shows the first few cycles of the representative
charge–discharge potential profiles. The curves are symmetric
Fig. 8. (a) First few cycles of the typical charge–discharge potential profiles of 120 s oxygen plasma annealed NiO/Ni NW arrays measured with chronopotentiometry at a
charging-discharging current of 5 mA cm−2. (b) Capacitance as a function of the cycle number.

200
M. Hasan et al. / Electrochimica Acta 60 (2012) 193–200
one-step approach for binder free fabrication of supercapacitor
electrodes.
Acknowledgements
This work is financially supported by Enterprise Ireland (EI)
under the commercialization fund CFTD/2008/322. The TEM
images by Michael Schmidt are gratefully acknowledged.
4. Conclusions
In summary, we have demonstrated the fabrication of core/shell
and one-dimensional vertical NiO/Ni nanostructures for pseudoca-
pacitor electrodes based on direct electrochemical deposition of
Ni inside nanoporous alumina template and the subsequent con-
version of the outer shell into oxide by oxygen plasma annealing
after removal of the template. The morphology, crystal structure,
and shell thickness were found to be dependent on the anneal-
ing conditions (input power and annealing time). The electrode
constructed of NiO/Ni NW arrays annealed at 50 W for 120 s demon-
strated high capacitance (0.36 F cm−2) with high charge–discharge
kinetics and a high degree of stability was attained for long-term
cycling (over 2000 cycles). The high charge–discharge kinetics
and high capacitance resulted from the unique NiO/Ni (core/shell)
nanoarchitecture that offers a highly conductive and robust core
for enhanced electronic transports to the NiO shell and a large
electroactive surface area for charge storage and redox reactions.
The fabrication of the core/shell NW electrodes is both facile and
effective, and it can be potentially applied for volume production
of integrated electrochemical capacitors.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.electacta.2011.11.039.
References
[1] J.M. Tarascon, M. Armand, Nature 414 (2001) 359.
[2] M. Armand, J.M. Tarascon, Nature 451 (2008) 652.
[3] J. Chmiola, C. Largeot, P.L. Taberna, P. Simon, Y. Gogotsi, Science 328 (2010) 480.
[4] Y. Zhang, H. Feng, X.B. Wu, L.Z. Wang, A.Q. Zhang, T.C. Xia, H.C. Dong, X.F. Li, L.S.
Zhang, Int. J. Hydrogen Energy 34 (2009) 4889.
[5] X. Zhao, B.M. Sanchez, P.J. Dobson, P.S. Grant, Nanoscale 3 (2011) 839.
[6] T. Brousse, M. Toupin, R. Dugas, L. Athouel, O. Crosnier, D. Belanger, J. Elec-
trochem. Soc. 153 (2006) A2171.
[7] Y. Lei, B. Daffos, P.L. Taberna, P. Simon, F. Favier, Electrochim. Acta 55 (2010)
7454.
[8] T.Y. Wei, C.H. Chen, K.H. Chang, S.Y. Lu, C.C. Hu, Chem. Mater. 21 (2009) 3228.
[9] B.R. Tao, J.A. Zhang, F.J. Miao, S.C. Hui, L.J. Wan, Electrochim. Acta 55 (2010)
5258.
[10] J.H. Kim, K. Zhu, Y.F. Yan, C.L. Perkins, A.J. Frank, Nano Lett. 10 (2010) 4099.
[11] D.C. Wang, W.B. Ni, H. Pang, Q.Y. Lu, Z.J. Huang, J.W. Zhao, Electrochim. Acta 55
(2010) 6830.
[12] M.S. Wu, Y.A. Huang, C.H. Yang, H.H. Jow, Int. J. Hydrogen Energy 32 (2007)
4153.
Electronic supplementary material
[13] D.D. Zhao, S.J. Bao, W.H. Zhou, H.L. Li, Electrochem. Commun. 9 (2007) 869.
[14] According to the expression C = I × ꢁt/ꢁV, where I is 0.4 mA cm−2, ꢁt is 5 s and
ꢁV is 0.5 V, the calculated capacitance (C) is 4 mF cm−2 instead of 2.5 F cm−2
as reported in Ref. [10]. In recent (24th March 2011) communication with Kai
Zhu (Ref. [10]), it was confirmed that there was a typographical error for the
unit (see Fig. 6(b) in Ref. [10]) and it should be mF rather than F.
[15] Z. Lu, Z. Chang, J. Liu, X. Sun, Nano Res. 4 (2011) 658.
[16] X. Zhang, W. Shi, J. Zhu, W. Zhao, J. Ma, S. Mhaisalkar, T. Maria, Y. Yang, H. Zhang,
H. Hng, Q. Yan, Nano Res. 3 (2010) 643.
[17] A.P. Grosvenor, M.C. Biesinger, R.S. Smart, N.S. McIntyre, Surf. Sci. 600 (2006)
1771.
[18] J.W. Liu, J. Essner, J. Li, Chem. Mater. 22 (2010) 5022.
[19] J. Jiang, A. Kucernak, Electrochim. Acta 47 (2002) 2381.
Supplementary material (SEM images of the annealed Ni NWs
for different annealing times at 50 W, XRD pattern of the as
deposited and annealed Ni NWs at a lower power of 25 W for dif-
ferent annealing times, XPS spectra measurements of NiO/Ni NWs,
cyclic voltammograms (CVs) of the as deposited Ni NW arrays com-
pared with the those annealed at 50 W for 120 s, the scan rate
dependence of the peak current intensity for NiO/Ni NW arrays,
CVs for thin and thick shell electrodes, and the STEM-EDS line scan
of NiO/Ni NWs) is available in the online version of this article.