Electrochimica Acta 55 (2010) 4677–4684

Contents lists available at ScienceDirect

Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta

Electrochemical performances of poly(3,4-ethylenedioxythiophene)–NiFe2 O4

nanocomposite as electrode for supercapacitor
Pintu Sen a,∗ , Amitabha De b
a
Physics Department, Variable Energy Cyclotron Centre, 1/AF Bidhannagar, Kolkata 700064, India
b
Saha Institute of Nuclear Physics, 1/AF Bidhannagar, Kolkata 700064, India

a r t i c l e i n f o a b s t r a c t

Article history: Poly 3,4-ethylenedioxythiophene (PEDOT)-based NiFe2 O4 conducting nanocomposites were synthesized
Received 13 January 2010 and their electrochemical properties were studied in order to find out their suitability as electrode
Received in revised form 23 March 2010 materials for supercapacitor. Nanocrystalline nickel ferrites (5–20 nm) have been synthesized by sol–gel
Accepted 26 March 2010
method. Reverse microemulsion polymerization in n-hexane medium for PEDOT nanotube and aque-
Available online 1 April 2010
ous miceller dispersion polymerization for bulk PEDOT formation using different surfactants have
been adopted. Structural morphology and characterization were studied using XRD, SEM, TEM and IR
Keywords:
spectroscopy. Electrochemical performances of these electrode materials were carried out using cyclic
PEDOT
NiFe2 O4
voltammetry at different scan rates (2–20 mV/s) and galvanostatic charge–discharge at different constant
Supercapacitor current densities (0.5–10 mA/cm2 ) in acetonitrile solvent containing 1 M LiClO4 electrolyte. Nanocom-
Electrical double layer capacitance posite electrode material shows high specific capacitance (251 F/g) in comparison to its constituents viz
Pseudocapacitance NiFe2 O4 (127 F/g) and PEDOT (156 F/g) where morphology of the pore structure plays a significant role
over the total surface area. Contribution of pseudocapacitance (CFS ) arising from the redox reactions
over the electrical double layer capacitance (CDL ) in the composite materials have also been investigated
through the measurement of AC impedance in the frequency range 10 kHz–10 mHz with a potential
amplitude of 5 mV. The small attenuation (∼16%) in capacitance of PEDOT–NiFe2 O4 composite over 500
continuous charging/discharging cycles suggests its excellent electrochemical stability.
© 2010 Elsevier Ltd. All rights reserved.

1. Introduction they have both electrochemical double layer capacitance and

pseudocapacitance arising mainly due to the fast and reversible
Now a days, it has been widely accepted that electrochemical oxidation and reduction processes related to the ␲-conjugated
capacitor (EC) popularly known as supercapacitor is one of the best polymer chain [2,3]. Among various conducting polymers, poly
candidates to provide high power and long cycle life, essential for 3,4-ethylenedioxythiophene (PEDOT) has recently attracted world-
new energy devices such as hybrid electric vehicles, backup sources wide research interest due to its environmental stability and
for various electrical devices and uninterrupted power supplies [1]. controllable electrical conductivity. However, PEDOT like other
Therefore, exploring new electrode materials for supercapacitor to ECPs also suffers a serious problem of typical volumetric swelling
meet the requirements of high power density and long durability and shrinkage during the insertion and ejection of counter ions
devices has become mandatory for the researchers working in the [4–8]. In order to solve this problem, nanocomposite comprising of
field of energy storage device. Depending on the charge storage PEDOT and inorganic transition metal oxide nanoparticles acting
mechanism, electrochemical capacitors are categorized as electro- as filler has been considered as the electrode materials for super-
chemical double layer capacitors (EDLC) and pseudocapacitors. The capacitor where a synergistic effect of composite formation plays a
origin of capacitance in the EDLC is the charge separation at the significant role to increase the capacitance value. Moreover, tran-
electrode–electrolyte interface, whereas pseudocapacitance arises sition metal oxides have generally been reported to be suitable
from fast, reversible faradic redox reactions taking place on or near as electrode material for pseudocapacitors because of their large
the surface of the electrode. capacitance and fast redox kinetics [9].
Electronically conducting polymers (ECPs) have advantage over Among the transition metal oxides, hydrous ruthenium dioxide
active carbon, as an electrode material for supercapacitor, as is recognized as one of the most promising candidates for elec-
trode materials in electrochemical capacitors, as it can store charges
reversibly by redox reaction [10]. But the lack of abundance and
∗ Corresponding author. Fax: +91 3323346871. the cost of the precious metal (Ru) are the major disadvantages
E-mail address: [email protected] (P. Sen). for the commercial production of RuO2 . As an alternative mate-

0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.03.077
4678 P. Sen, A. De / Electrochimica Acta 55 (2010) 4677–4684

rial, some other transition metal oxide such as NiO, MnO2 , Fe3 O4 , vigorous stirring. Resulting dark blue coloured samples remained
Co3 O4 , V2 O5 , MoO3 and TiO2 have also been studied as electrode in the dispersed phase. In order to isolate the samples in the solid
materials for pseudocapacitors [11–18]. Recently, some crystalline state, precipitation of the samples were carried out using ethanol
ferrites, particularly MnFe2 O4 and CoFe2 O4 have also been reported as the non-solvent. The precipitates were washed and dried in a
to exhibit pseudocapacitance [19]. Among other metal ferrites, vacuum for 24 h at 60 ◦ C using copious amount of alcohol–water
NiFe2 O4 has not yet been explored as a candidate for pseudoca- mixture [23].
pacitor to the best of our knowledge. Therefore it is worthwhile
to investigate the suitability of the nanocomposites comprising of 2.2.3. Synthesis of PEDOT–NiFe2 O4 composite in n-hexane
PEDOT and nanocrystalline NiFe2 O4 in the field of challenging new medium
generation miniaturized supercapacitor as electrode materials of A reverse microemulsion was first prepared by dissolving
high capacitance value. Moreover, nanocrystalline NiFe2 O4 can be 19.12 mmol sodium bis(2-ethylhexyl) sulfosuccinate (AOT) in
a promising candidate for supercapacitor as it is an inexpensive and 70 ml of n-hexane. 10 mmol FeCl3 in 1.0 ml distilled water was
innocuous material. added to it and the mixture was gently stirred for 5 min. Previ-
Conducting polymer-based supercapacitors can be classified ously distilled 3.52 mmol EDOT monomer was added to the reaction
into three types [20]. Type I is a symmetric system where same p- mixture followed by slow addition of 100 mg nickel ferrite nanopar-
dopable conducting polymers are used at both the electrodes of the ticles (NiFe2 O4 ) and kept for 3 h under gentle magnetic stirring. The
capacitor. Type II is an asymmetric system based on two different blue-black precipitate of PEDOT–NiFe2 O4 was filtered and washed
p-dopable conducting polymers to be used as electrode materials. with methanol followed by acetonitrile. Composite was dried under
Type III is a symmetric system based on a conducting polymer, vacuum for 12 h at 60 ◦ C. The loading of the NiFe2 O4 nanoparticles
which can be used both as p- and n-doped electrode materials. Due in the composite sample is ∼20%. Pure PEDOT polymer in n-hexane
to the difficulties of n-doping for PEDOT [21], we have attempted medium was synthesized applying similar procedure in the absence
to build a Type I supercapacitor using two identical electrodes with of NiFe2 O4 nanoparticles [24].
same materials.
In the present work, the electrochemical performances of
nanotube based PEDOT–NiFe2 O4 nanocomposite as electrode for 2.3. Sample characterization
supercapacitor using 1 M LiClO4 in acetonitrile solution was inves-
tigated for the first time. Reverse microemulsion polymerization Phase identification and morphological characterizations of
for PEDOT nanotube and aqueous miceller dispersion polymer- PEDOT and its composites were carried using powdered X-ray
ization for bulk PEDOT formation have been adopted. Structural diffractrometer (Philips PW 1710) with CuK˛ ( = 1.5406 Å) radi-
morphology and characterization were carried out by XRD, SEM, ation, transmission electron microscope (FEI model Tecnei G2 20S
TEM and IR spectroscopy. The electrochemical performances of with 200 kV accelerating voltage and resolution of 0.2 nm) and
the supercapacitors were investigated by cyclic voltammetry scanning electron microscope (Carl Zeiss AG-EVO 50). Infrared
(CV), galvanostatic charge/discharge cycling and electrochemical spectra for the identification of the characteristic bands were
impedance spectroscopy (EIS) studies. recorded using a Perkin Elmer Spectrum 100 FTIR spectrophotome-
ter employing transmission mode through KBr disc formation. The
2. Experimental specific surface area of the samples were measured by nitrogen
gas absorption through Brunauer–Emmett–Teller (BET) method,
2.1. Materials whereas the pore volume, average pore diameter and the pore size
distribution were determined by Barrett–Joyner–Halenda (BJH)
The 3,4-ethylenedioxythiophene (EDOT) monomer was method using Quantachrome NOVA 4000e surface analyzer.
obtained from BAYTRON Co. Poly-tetrafluoroethylene and
acetylene black were obtained from Aldrich and Alpha Aesar 2.4. Electrochemical measurements
respectively. Fe(NO3 )3 ·9H2 O, FeCl3 ·6H2 O, Ni(NO3 )2 ·6H2 O, n-
hexane and ethylene glycol, dodecyl benzene sulphonic acid Electrodes for supercapacitor were prepared using following
(DBSA) and ammonium peroxy disulphate (APS) were the product procedure: 85 wt% electroactive materials (i.e. pure PEDOT-Aq,
of Merck. Sodium bis(2-ethylhexyl) sulphosuccinate (AOT) was PEDOT-Org and its composite containing 20% NiFe2 O4 nanopar-
purchased from Fluka. ticles) were mixed with 10 wt% acetylene black (AB) and 5 wt%
polytetrafluoroethylene (PTFE) to form a thick paste. The paste was
2.2. Synthesis of materials then pressed into a thin sheet of ∼100 ␮m thickness using mortar
and pestle. Finally, the sheet was compressed on a stainless steel
2.2.1. Synthesis of nickel nanoferrite particles mesh having the surface area of 1 cm2 . The prepared electrodes
NiFe2 O4 nanoparticles were prepared by sol–gel procedure. The were dried at 60 ◦ C for 6 h under vacuum. The total weight of the
stoichiometric amount of Fe(NO3 )3 , 9H2 O and Ni(NO3 )2 , 6H2 O were active material in the electrode is usually ∼5 mg.
mixed (mole ratio of Fe/Ni 2:1) and dissolved in ethylene glycol at Electrochemical behavior of the samples through cyclic voltam-
∼40 ◦ C. The sol of metal salts was then heated at ∼60 ◦ C to obtain metry (CV) measurement was investigated with AUTOLAB-30
gel. The gel product was dried at ∼100 ◦ C and fired at ∼400 ◦ C for potentiostat/galvanostat. A platinum electrode and a saturated
24 h. Finally, the sample was cooled to room temperature [22]. Ag/AgCl electrode were used as counter and the reference elec-
trodes respectively. Cyclic voltammograms were recorded between
2.2.2. Synthesis of PEDOT in aqueous medium 0 and 1 V w.r.t. reference electrode at a different scan rate
Aqueous miceller dispersion was prepared by introducing 3.48 g (2–20 mV/s). Both galvanostatic charge–discharge cycling and
dodecyl benzene sulphonic acid (DBSA) in 100 ml of deionized electrochemical impedance studies were performed with two-
water with stirring. To this solution, distilled 1.5 ml EDOT monomer electrode system having identical electrodes made of same active
was added in 1:1 mole ratio w.r.t. DBSA and solubilized with stir- electrode materials (i.e. Type I symmetry supercapacitor). Con-
ring for 1 h. 2.28 g of ammonium peroxy disulphate (APS) in 1:1 stant current densities ranging from 0.5 to 10 mA/cm2 have been
mole ratio w.r.t. monomer was added to this mixture as an oxidant. employed for charging/discharging the cell in the voltage range
Polymerization reaction was allowed to continue for 20 h under 0–1 V. The discharge capacitance (C) is estimated from the slope
 P. Sen, A. De / Electrochimica Acta 55 (2010) 4677–4684 4679

Fig. 1. X-ray diffraction patterns of (a) pure NiFe2 O4 nanoparticles and (b) nanocom-
posite sample.

(dv/dt) of the linear portion of the discharge curve using the expres-
sion.
I
C= (1)
dv/dt
The weight of the active material of the two electrodes is same
in a Type I supercapacitor. The specific capacitance (Cs ) of the single
electrode can thus be expressed as [25]
2C
Cs = (2)
m
where m is the active mass of the single electrode. The energy
density (Ed ) of the capacitor can be expressed as [26]
1
 2

Ed = 2
Cs Vmax (3)

The coulomb efficiency ‘’ was evaluated using the following

relation, where tc and td are the time of charge and discharge
respectively.
td
= × 100% (4)
tc
Electrochemical impedance spectra (EIS) were taken at open cir-
cuit potential (OCP) over the frequency range 10 kHz–10 mHz with
a potential amplitude of 5 mV. All the electrochemical experiments
(i.e. CV, charge–discharge, EIS) were carried out in an electrolyte
containing 1 M LiClO4 in acetonitrile.

3. Result and discussion

3.1. Characterization of the samples

3.1.1. X-ray diffraction, TEM and SEM studies

X-ray diffraction patterns of NiFe2 O4 nanoparticles and its com-
posite PEDOT–NiFe2 O4 are shown in Fig. 1a and b respectively. In
Fig. 2. TEM images of (a) NiFe2 O4 nanoparticles (5–20 nm) synthesized in sol–gel
Fig. 1a, all the characteristic peaks of single-phase spinel structure
technique, (b) pure PEDOT synthesized in aqueous medium and (c) PEDOT–NiFe2 O4
of NiFe2 O4 were observed. The broad peak at 2 = 25.1◦ other than nanotube having diameter ∼20 nm synthesized by reverse microemulsion polymer-
the characteristic peaks of NiFe2 O4 in Fig. 1b corresponds to the ization in n-hexane.
thiophene ring of PEDOT in the composite system [27].
The TEM picture (Fig. 2a) of nickel ferrite nanoparticles suggests
that they are in crystalline state having clear lattice fringe. The sizes
of the particles are in the range of 5–20 nm. The structural morphol-
ogy of PEDOT synthesized in aqueous medium (Fig. 2b) shows no
signature of formation of any mesoporous/microporous structure,
4680 P. Sen, A. De / Electrochimica Acta 55 (2010) 4677–4684

Fig. 4. FTIR spectra of (a) NiFe2 O4 nanoparticles and (b) PEDOT–NiFe2 O4 nanocom-
posite sample.

[31]. The C–S bond stretching in the thiophene ring is also seen at
983, 840, and 691 cm−1 [32]. The absence of the characteristic peak
of nickel ferrite at 608 cm−1 in the composite samples has there-
fore suggested the formation of PEDOT nanotube developed over
the surface of the nickel ferrite nanoparticles.

3.2. Electrochemical characterization

3.2.1. Cyclic voltammetry analysis

Typical cyclic voltammograms (CV) of different samples at a
Fig. 3. SEM images of (a) PEDOT-Org and (b) PEDOT–NiFe2 O4 nanocomposite sam-
ples.
scan rate of 2 mV/s between 0 and 1 V in acetonitrile containing
1 M LiClO4 electrolyte are shown in Fig. 5. Cyclic voltammograms of
PEDOT–NiFe2 O4 nanocomposite at different scan rates (2–20 mV/s)
whereas PEDOT–NiFe2 O4 synthesized in n-hexane (Fig. 2c) sup- are shown in Fig. 6. Cyclic voltammograms of different samples
ports the formation of mesoporous nanotube having diameter in are nearly semi-rectangular in shape, which indicates the good
the range ∼20 nm. SEM images of PEDOT-Org and PEDOT–NiFe2 O4 capacitative behavior of the electrodes suitable for charging and
composite shown in Fig. 3a and b respectively reveal the formation discharging at a constant rate over the voltage range of 0–1 V [1,33].
of nanotubler structures in their polymer matrix. Voltammetric charge (q* ) is one of the important parameters to
evaluate the electrochemical active surface area of the electrode.
3.1.2. FTIR analysis Total surface charge (q∗total ), related to total electrochemical active
FTIR spectra of nickel ferrite nanoparticles and its composite
with PEDOT are shown in Fig. 4a and b respectively. In Fig. 4a, the
strong peak around 608 cm−1 is attributed to the tetrahedral com-
plexes of the nickel ferrite. The characteristic peaks around 3441
and 1631 cm−1 are ascribed to the stretching mode of O–H group
and bending vibration mode of H–O–H group respectively, indi-
cating the presence of water molecule [28]. The very small peaks
around 1384 and 1051 cm−1 are due to the stretching vibrations
of the anti-symmetric NO3 − group and the stretching mode of
C–O–C group respectively. The other two small peaks occurring at
2925 and 2854 cm−1 are due to the respective anti-symmetric and
symmetric-CH2 -vibrations of the carbon chains [29]. The appear-
ance of very small peaks at 1384, 1051, 2925 and 2854 cm−1 thus
revealed the presence of nitrate and other organic substances (e.g.
ethylene glycol) in insignificant amount during the synthesis of
nano-nickel ferrite from its sol–gel route.
In the composite sample (Fig. 4b), all characteristic peaks of
PEDOT along with some other characteristic peaks at higher wave
number i.e. 3441 and 2925 cm−1 are also observed. The strong peaks
occurring at 1519 and 1350 cm−1 are due to C = C and C–C stretch-
ing of the quininoid structure of the thiophene ring respectively
Fig. 5. Typical cyclic voltammogram of (a) PEDOT-Aq, (b) PEDOT-Org, (c) nano-
[30]. The peaks at 1205, 1141 and 1091 cm−1 have originated from NiFe2 O4 and (d) PEDOT–NiFe2 O4 composite electrodes in acetonitrile containing
C–O–C bond stretching in the ethylene dioxy (alkylenedioxy) group 1 M LiClO4 electrolyte at a scan rate 2 mV/s in a potential range between 0 and 1 V.
 P. Sen, A. De / Electrochimica Acta 55 (2010) 4677–4684 4681

Fig. 6. Cyclic voltammogram of PEDOT–NiFe2 O4 composite in acetonitrile contain-

ing 1 M LiClO4 electrolyte at a scan rate (a) 2 mV/s, (b) 5 mV/s, (c) 10 mV/s and (d)
20 mV/s in a potential range between 0 and 1 V.

surface area, can be divided into two parts as q∗out and q∗in . The q∗out
correlates to the outer region of the electrodes which is directly
exposed to the electrolyte whereas q∗in correlates with the inner
part of the electrodes which are hidden in pores, grain boundary,
etc. and reflects the region of difficult accessibility for the ionic
species assisting the surface redox reaction, essential for enhancing
pseudocapacitance.
Voltammetric charges (q* ) at different potential scan rate v
(mV/s) were obtained by integration of the voltammetric curves
followed by division with the geometric surface area of the sam-
ples without correction for background capacitative current. In
fact, voltammetric charge (q* ) is a mean between the cathodic and
anodic branches of CV [34]. q* increases with decrease in poten-
tial scan rate (or with v−1/2 ). At higher scan rates, the diffusion of
ions is limited to the more accessible sites, i.e. the outer surface
of the electrode. Therefore, extrapolation of q* to scan rate v = ∝
(i.e. v−1/2 = 0) from the linear portion of the q* versus v−1/2 plot
(Fig. 7a) can provide the outer charge q∗out related to the more eas- Fig. 8. Nitrogen adsorption–desorption isotherms and corresponding pore size
distribution plots (inset) for (a) PEDOT–NiFe2 O4 nanocomposite and (b) NiFe2 O4
ily accessible sites. On the other hand, the extrapolation of q* to
nanoparticles.

the scan rate v = 0 in the plot 1/q* versus v−1/2 (Fig. 7b) gives the
total charge q∗total , which is related to both inner and outer active
sites of the electrode [35]. One can thus easily calculate the charge
related to the inner sites (i.e. less accessible sites) needed for higher
capacitance, by the relation q∗in = q∗total − q∗out .
It is interesting to note that the internal accumulation of charges
in the inner surface of the electrode PEDOT-Org synthesized in
organic medium (i.e. n-hexane) have increased to 52.30% compared
to that of PEDOT-Aq (43.83%) synthesized in aqueous medium,
though there is not much difference in total surface area of PEDOT-
Org (∼11 m2 /g) and PEDOT-Aq (∼8 m2 /g). This indicates that the
specific capacitance or total accumulation of internal charge does
not depend only on the surface area. Pore size distribution may
be one of the other deciding parameters in determining the accu-
mulation of internal charge, hence specific capacitance [36]. Typical
nitrogen adsorption–desorption isotherms and corresponding pore
size distribution plots (inset) of PEDOT–NiFe2 O4 nanocomposite
and nickel ferrite nanoparticles are shown in Fig. 8a and b respec-
tively. Specific surface area, cumulative pore volume and average
pore diameter of all the electrode materials are given in Table 1.
Fig. 7. (a) Voltammetric charge (q* ) vs inverse square root of scan rate (v−1/2 ). (b) Generally, these materials have three types of pore distribu-
Inverse voltammetric charge (1/q* ) vs square root of scan rate (v1/2 ). tions, viz micropore (<2 nm), mesopore (2–50 nm) and macropore
4682 P. Sen, A. De / Electrochimica Acta 55 (2010) 4677–4684

Table 1
Surface area and pore volume and average pore diameter of different electrode materials.

Electrode material Surface area (m2 /g) Pore volume (cm3 /g) Average pore diameter (nm)

PEDOT-Aq 8.15 0. 026 9.24

PEDOT-Org 10.68 0. 040 16.87
Nano-NiFe2 O4 110.53 0.467 4.52
PEDOT-comp 11.79 0.042 16.46

Fig. 10. Typical galvanostatic charging and discharging curves between 0 and 1 V of
Fig. 9. Galvanostatic charging and discharging between 0 and 1 V of the capacitor the capacitor made of PEDOT–NiFe2 O4 composite electrode at (a) 0.5 mA/Cm2 , (b)
made of (a) PEDOT-Aq, (b) PEDOT-Org, (c) nano-NiFe2 O4 and (d) PEDOT–NiFe2 O4 1.0 mA/Cm2 , (c) 3 mA/cm2 , (d) 5 mA/cm2 , (e) 7 mA/cm2 and (f) 10 mA/cm2 current
composite electrode at a current density of ±1 mA/cm2 . density.

(>50 nm) [37]. The mesoporous nanotube formation of PEDOT-Org capacitance values of the electrodes have increased from 143 F/g
with large average pore diameter of 16.87 nm during its synthesis for PEDOT-Aq to 156 F/g for PEDOT-Org, as the presence of meso-
in n-hexane medium enhances more accessibility to the electrolyte porous nanotube allows the counter ions to readily penetrate into
for internal surface adsorption and intercalation, while the lower the polymer matrix and access their internal surface. The spe-
average pore diameter of PEDOT-Aq (9.24 nm) might be responsible cific capacitance value has further increased in PEDOT–NiFe2 O4
for its lowering internal charge accumulation inside the elec- composite system (251 F/g) due to synergistic effect of composite
trode compared to PEDOT-Org. The internal charge accumulation formation, where the specific capacitance of pure nanocrystalline
of PEDOT–NiFe2 O4 (63.23%) nanocomposite has further increased NiFe2 O4 is only 127 F/g. Energy density (Ed ) of the composite sam-
though its surface morphology is comparable to that of PEDOT-Org ple also shows the similar trend like capacitive behavior (Table 2).
(Table 1). This is due to dispersion of nickel ferrite nanoparticles Moreover, the charge and discharge profiles of the electrodes
having size range ∼5–20 nm in the mesoporous PEDOT nanotube, (Fig. 9) are not standard straight lines, but are somewhat curved,
whereas only pure nickel ferrite nanoparticle having high specific exhibiting a pseudocapacitive characteristic [38].
surface area ∼110 m2 /g has no such advantages over PEDOT nan- In general, the specific capacitance decrease gradually with
otube due to lower pore diameter (4.52 nm) close to microporous increasing discharge current density due to increasing IR
structure. The internal surface area of nickel ferrite nanoparticles drops. Fig. 10 shows the typical charge–discharge curve of
is only 46.23%. Moreover the average value of pore diameter of PEDOT–NiFe2 O4 nanocomposite at different current densities rang-
the PEDOT nanotube estimated by BJH method is also comparable ing from 0.5 to 10 mA/cm2 with voltage between 0 and 1 V.
to that of the internal pore diameter observed from TEM studies Specific capacitance and the coulomb efficiency () as a func-
(∼20 nm). tion of current density are shown in Fig. 11. It is evident from
Fig. 11 that the reduction in specific capacitance for the composite
3.2.2. Charge–discharge studies PEDOT–NiFe2 O4 electrode is very low (∼5%) in the range of cur-
The applicability of the supercapacitor can be directly evaluated rent density 1–5 mA/cm2 compared to current density >5 mA/cm2 .
by means of the galvanostatic charge–discharge studies. Specific At first, coulomb efficiency (∼75%) is found to be low at low dis-
capacitances of different electrode materials were compared at a charge current density (i.e. at 0.5 mA/cm2 ). However, coulomb
constant current density of ±1 mA/cm2 . Charge–discharge profiles efficiency increases with increasing discharge current density on
of different samples are shown in Fig. 9. The discharge specific 1 mA/cm2 onwards (∼88%) and reaches to maximum value ∼93%

Table 2
Typical charge–discharge characteristics of different capacitor cell at constant current-density ±1 mA/cm2 .

Cell Working voltage (V) Discharge specific capacitance (F/g) Energy density Ed (Wh/kg) Coulombic efficiency (%)

PEDOT-Aq 1.00 143 19.86 80

PEDOT-Org 1.00 156 21.87 89
Nano-NiFe2 O4 1.00 127 17.63 75
PEDOT-comp 1.00 251 34.86 88
 P. Sen, A. De / Electrochimica Acta 55 (2010) 4677–4684 4683

Fig. 11. Specific capacitance and the coulomb efficiency () of PEDOT–NiFe2 O4 com-
posite electrode as a function of current density. Fig. 13. Typical Nyquist impedance plot at open circuit potential (OCP) over a
frequency range of 100 kHz–10 mHz with a potential amplitude of 5 mV for (a)
PEDOT-Aq, (b) PEDOT-Org, (c) nano-NiFe2 O4 and (d) PEDOT–NiFe2 O4 composite
at 10 mA/cm2 . Hence the cycle life of the PEDOT–NiFe2 O4 compos- electrodes.
ite electrode was examined over 500 hundred cycles at a constant
current density 1 mA/cm2 having high capacitance value with opti-
mum coulomb efficiency. Fig. 12 shows the variation of specific cept of the semi-circle with real axis (Z ) at high frequencies is
capacitance as a function of cycle number (cycled between 0 and the measure of internal resistance (Rs ) which may be due to (i)
1 V). There is a small decrease in specific capacitance (∼5%) over ionic resistance of the solution or electrolyte, (ii) intrinsic resis-
first 100 cycles and then it remains almost constant. The small tance of the active electrode materials and (iii) interfacial resistance
attenuation of the specific capacitance (∼16%) over 500 cycles thus between the electrode and current collector [40]. The origin of the
suggests the excellent electrochemical stability of PEDOT–NiFe2 O4 semi-circle at higher frequency range is due to ionic charge trans-
nanocomposite material acting as electrode for supercapacitor over fer resistance (Rct ) at the electrode–electrolyte interface [41]. The
a large number of cycle operation. diameter of the semi-circle along the real axis (Z ) gives the charge
transfer resistance Rct . All the impedance spectra were analyzed
using semi-quantitative fittings programme supplied by the instru-
3.2.3. Electrochemical impedance analysis
ment (AUTOLAB-30) on the basis of the Randles equivalent circuit
Electrochemical impedance spectroscopy is a powerful tech-
modeling shown in Fig. 14. The model circuit comprised of five
nique for investigating the capacitative behavior of electrochemical
elements, the internal resistance Rs is in series with the electri-
cells. Typical Nyquist impedance spectra of PEDOT and its com-
cal double layer capacitance (Cdl ) at the interface of electrode and
posites over a frequency range of 10 kHz–10 mHz with a potential
electrolyte. Cdl is in parallel with the charge transfer resistance (Rct )
amplitude of 5 mV are shown in Fig. 13. The impedance spectra
and warburg impedance (WZ ). The warburg impedance is associ-
show a single semi-circle in the high frequency region and nearly
ated with the semi-infinite diffusion of ions in the electrode [1]. The
vertical line in the low frequency region, which indicates that the
above set is in series with pseudocapacitance (CFS ), which is used
electrode process is controlled by electrochemical reaction at high
to account for the faradic reaction. The values of Rs , Rct , WZ and
frequencies and by mass transfer at low frequencies [39]. The inter-
CFS were estimated qualitatively from the fittings of experimental
impedance spectra and presented in Table 3. The mean error of the
modulus is ∼1%.
The internal resistance (Rs ) i.e. intrinsic resistance of the PEDOT-
Org is less compared to PEDOT-Aq. This is due to the development
of internal mesoporous nanotubular structure during its synthe-
sis in n-hexane medium. The value of internal resistance further
decreases in the composite system due to introduction of nickel fer-
rite nanoparticles having average particle size 5–20 nm (Table 3).

Fig. 14. Equivalent electrical circuit comprised of five elements: electrolytic solu-
Fig. 12. Cycling life of PEDOT–NiFe2 O4 composite electrode at a current density of tion resistance (Rs ), charge transfer resistance (Rct ), double layer capacitance (CDL ),
1 mA/cm2 between 0 and 1 V. pseudocapacitance (CFS ) and warburg impedance (WZ ).
4684 P. Sen, A. De / Electrochimica Acta 55 (2010) 4677–4684

Table 3 cal double layer capacitance (CDL ) in the composite materials. The
Electrochemical parameters of different capacitors obtained from impedance anal-
small decrease in specific capacitance value (∼16%) over 500 cycles
ysis based upon the proposed equivalent Randel circuit.
compared to first cycle suggests PEDOT–NiFe2 O4 nanocomposite
Cell Rs () Rt () CDL (␮F) CFS (F) WZ ( s−0.5 ) as a promising candidate for future development of safe and cost
PEDOT-Aq 12.52 33.50 8.85 0.413 0.14 effective electrochemical supercapacitors.
PEDOT-Org 11.71 26.35 15.47 0.643 0.15
Nano-NiFe2 O4 8.25 8.49 12.89 0.239 0.15 Acknowledgements
PEDOT-comp 7.26 6.08 67.90 0.781 0.40

The authors would like to thank Dr. D.K. Srivastava for constant
It is quite interesting in the case of charge transfer resistance, encouragement and support. The authors also would like to express
where introduction of redox system having multivalent cations their gratitude to Mr. Pulak Roy of Bio-Physics Division of SINP for
(e.g. NiFe2 O4 ) plays a significant role over the pore size distri- rendering the facility of transmission electron microscope.
bution as it is reflected in the lower value of the charge transfer
resistance Rct of PEDOT–NiFe2 O4 composite and NiFe2 O4 nanopar- References
ticles compared to PEDOT-Org and PEDOT-Aq (Table 3) samples.
Thus, introduction of nanoferrite in the composite material not only [1] B.E. Conway, Electrochemical Supercapacitors, Scientific Fundamentals and
Technological Applications, Kluwer Academic/Plenum Publishers, New York,
helps to reduce the intrinsic resistance (Rs ) through the develop- 1999.
ment of more number of mesoporous structure as nanotube but [2] S.W. Woo, K. Dokk, K. Kanamura, J. Power Sources 185 (2008) 1589.
also increase the kinetics of electron transfer through redox pro- [3] B.C. Kim, J.M. Ko, G.G. Wallace, J. Power Sources 177 (2008) 665.
[4] X. Chen, O. Inganas, Synth. Met. 74 (1995) 159.
cess leading to the enhancement of pseudocapacitance (CFS ) in [5] X. Chen, K.Z. Xing, O. Inganas, Chem. Mater. 8 (1996) 2439.
the composite materials (Table 3). Moreover, the contribution of [6] N.S. Murthy, L.W. Shacklette, R.H. Baughmann, J. Chem. Phys. 87 (1987) 2346.
the pseudocapacitance (CFS ) is quite large compared to electrical [7] M. Winokur, P. Walmsley, J. Smith, A.J. Heeger, Macromolecules 24 (1991) 3812.
[8] Z.A. Hu, Y.L. Xie, Y.X. Wang, L.P. Mo, Y.Y. Yang, Z.Y. Zhang, Mater. Chem. Phys.
double layer capacitance (CF ) as the conducting polymer PEDOT 114 (2009) 990.
itself behaves as a redox system which is further enhanced by the [9] F. Fusalba, H.A. Ho, L. Breau, D. Belanger, Chem. Mater. 12 (2000) 2581.
introduction of nickel ferrite nanoparticles in the composite system [10] K.S. Ryu, S.K. Jeong, J. Joo, K. Kim, J. Phys. Chem. B 111 (2007) 731.
[11] M.S. Wu, Y.A. Huang, C.H. Yang, J.J. Jow, Int. J. Hydrogen Energy 32 (2007) 4153.
(Table 3). The lower value of warburg impedance WZ (Table 3) is
[12] D. Choi, G.E. Blomgren, P.N. Kumata, Adv. Mater. 18 (2006) 1178.
also attributed to the diffusion process of the electrolyte (i.e. mass [13] E.B. Castro, S.G. Real, L.F. Pinheiro Dick, Int. J. Hydrogen Energy 29 (2004) 255.
transfer) in the mesoporous/macroporous structure of electrode at [14] G.X. Wang, B.L. Zhang, Z.L. Yu, M.Z. Qu, Solid State Ionics 176 (2005) 1169.
[15] N.L. Wu, Mater. Chem. Phys. 75 (2002) 6.
lower frequency.
[16] G. Spinolo, S. Ardizzone, S. Trasatti, J. Electroanal. Chem. 423 (1997) 49.
[17] T. Kudo, Y. Ikeda, T. Watanabe, M. Hibino, M. Miyayama, H. Abe, K. Kajita, Solid
4. Conclusion State Ionics 152–153 (2002) 833.
[18] P. Novak, K. Mulle, K.S.V. Santhanam, O. Hass, Chem. Rev. 97 (1997) 207.
[19] S.L. Kuo, N.L. Wu, Electrochem. Solid State Lett. 8 (2005) A495.
Sol–gel method has been adopted for the preparation of NiFe2 O4 [20] A. Rudge, J. Davey, I. Raistrick, S. Gottesfeld, J.P. Ferraris, J. Power Sources 47
nanoparticles having particle size in the range of 5–20 nm. TEM (1994) 89.
and SEM images of PEDOT–NiFe2 O4 nanocomposite synthesized [21] H.J. Ahonen, J. Lukkari, J. Kankare, Macromolecules 33 (2000) 6787.
[22] A. Poddar, R.N. Bhowmik, A. De, P. Sen, J. Magn. Magn. Mater. 321 (2009) 2015.
in n-hexane reveals the formation of mesoporous nanotube over [23] A. De, Pintu Sen, A. Poddar, A. Das, Synth. Met. 159 (2009) 1002.
the surface of nickel ferrite nanoparticles, supporting the FTIR [24] X. Zhang, J.S. Lee, G.S. Lee, D.K. Cha, M.J. Kim, D.J. Yang, S.K. Manohar, Macro-
spectra of the nanocomposite where the characteristic peak of molecules 39 (2006) 470.
[25] C. Portet, P.L. Taberna, P. Simon, C. Laberty-Robert, Electrochem. Acta 49 (2004)
nickel ferrite at 608 cm−1 is not prominent. The inner voltam- 905.
metric charge, corresponds to the inaccessible active surface sites [26] S.K. Tripathi, A. Kumar, S.A. Hasmi, Solid State Ionics 177 (2006) 2979;
of PEDOT synthesized in reverse microemulsion polymerization J.P. Zheng, J. Electrochem. Soc. 152 (2005) A1864.
[27] L. Zhang, H. Peng, P.A. Kilmartin, C. Soeller, J. Travas-Sejdic, Macromolecules 41
method (52.30%), has increased compared to that synthesized in (2008) 7671.
aqueous miceller dispersion polymerization method (43.83%). In [28] S. Yan, J. Geng, L. Yin, E. Zhou, J. Magn. Magn. Mater. 277 (2004) 84.
the former case, mesoporous nanotube formation plays a sig- [29] M. Srivastava, A.K. Ojhaa, S. Chaubeya, A. Materny, J. Alloy Compd. 481 (2009)
515.
nificant role in the accumulation of internal charge compared
[30] C. Kvarnstrom, H. Neugebauer, S. Blomquist, J.H. Ahonen, J. Kankare, A. Ivaska,
to total surface area. This phenomenon is more pronounced in N.S. Sariciftci, Synth. Met. 101 (1999) 66.
PEDOT–NiFe2 O4 nanocomposite possessing highest inner voltam- [31] C. Kvarnstrom, H. Neugebauer, A. Ivaska, N.S. Sariciftci, J. Mol. Struct. 521 (2000)
271.
metric charge ∼63.23%, exhibiting highest specific capacitance
[32] P. Damlin, C. Kvarnstrom, A. Ivaska, J. Electroanal. Chem. 570 (2004) 113.
value (251 F/g), where introduction of nickel nanoferrite having [33] H.Y. Lee, J.B. Goodeough, J. Solid State Chem. 144 (1999) 220.
lower value of specific capacitance (127 F/g) is playing a role of syn- [34] J. Bobacka, A. Lewenstam, A. Ivaska, J. Electroanal. Chem. 498 (2000) 17.
ergistic agent. Moreover, electrochemical impedance study reveals [35] S. Trasatti, Electrochem. Acta 35 (1990) 263.
[36] R.N. Reddy, R.G. Reddy, J. Power Sources 124 (2003) 330.
that the nano-nickel ferrite in the composite material not only helps [37] C. Lastoskie, K.E. Gubbins, N. Quirkeft, J. Phys. Chem. 97 (1993) 4786.
to reduce the intrinsic resistance (Rs ) through the development of [38] A. Yuan, X. Wang, Y. Wang, J. Hu, J. Electrochim. Acta 54 (2009) 1021.
more number of mesoporous structure but also increases the kinet- [39] N. Li, D. Shan, H. Xue, Eur. Polym. J. 43 (2007) 2532.
[40] D. Kalpana, J. Power Sources 190 (2009) 587.
ics of electron transfer through the redox process leading to the [41] X. Ren, P.G. Pickup, J. Electrochem. Chem. 420 (1997) 251.
enhancement of pseudocapacitance (CFS ) as compared to electri-