1. Short Communication
Electrochemical investigation of graphene/cerium oxide
nanoparticles as an electrode material for supercapacitors
Mahdi Robat Sarpoushi a,n
, Mahdi Nasibi a,b,nn
, Mohammad Ali Golozar c
,
Mohammad Reza Shishesaz a
, Mohammad Reza Borhani d
, Sajad Noroozi a
a
Technical Inspection Engineering Department, Petroleum University of Technology, Abadan 63187-14331, Iran
b
Health, Safety and Environment (HSE) Engineering Office, NIOPDC, Yazd Region, Yazd 89167-84395, Iran
c
Materials Science Engineering Department, Isfahan University of Technology, Isfahan 84156-83111, Iran
d
Department of Materials Engineering, Malek Ashtar University of Technology, ShahinShahr, Isfahan, Iran
a r t i c l e i n f o
Keywords:
Electronic materials
Nanostructures
Electrical properties
Energy storage
Graphene
Cerium oxide nanoparticles
a b s t r a c t
Mechanisms of charge storage, stability, capacitance, morphology and response current of
graphene/cerium oxide (CeO2) nanoparticles as an electrode material for electrochemical
capacitors have been investigated. Electrochemical properties of the assembled electrodes
were studied using cyclic voltammetry (CV) and electrochemical impedance spectroscopy
(EIS) techniques in 3 M NaCl, NaOH and KOH electrolytes. Scanning electron microscopy
(SEM) is used to characterize the microstructure and the nature of prepared electrodes.
SEM images confirm the layered structure (12 nm thickness) of the used graphene.
The proposed electrode shows a maximum specific capacitance as high as 11.09 F gÀ1
in
the potential range between À0.55 and 0.3 (V vs. SCE) at scan rate of 5 mV sÀ1
. The
charge/discharge cycling test shows a good reversibility and confirms that capacitance will
increase after 500 cycles by 37%.
Crown Copyright & 2014 Published by Elsevier Ltd. All rights reserved.
1. Introduction
Energy is currently a vital global issue given the likely
depletion of current resources (fossil fuels) coupled
with the demand for higher-performance energy storage
systems [1]. Such systems require the advantages of
portability and energy efficiency whilst being environ-
mental friendly [2,3]. Among different energy storage
systems electrochemical capacitors can provide high
power capabilities, excellent reversibility (90–95%) and
long cycle life (4105
) and exhibit 20–200 times larger
capacitance per unit volume or mass than conventional
capacitors [4-6].
Depending on the electrode material and charge sto-
rage mechanism, electrochemical capacitors are classified
as electrochemical double layer capacitors and pseudoca-
pacitors [7-10]. The electrochemical double layer capaci-
tors arise from the charge separation at the electrode/
electrolyte interfaces, whereas pseudocapacitors exhibit
electrochemical Faradic reactions between electrode mate-
rial and electrolyte [11]. Transition metal oxides are
considered to be the most suitable candidate materials
for electrochemical capacitors. These stem from the high
specific capacitance coupled with low resistance resulting
in a high specific power which makes them suitable for
commercial applications. Because of the direct and fast
transformation of Ce(III) and Ce(IV), CeO2 nanoparticles
may be good candidate as an electrode material for
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/mssp
Materials Science in Semiconductor Processing
http://dx.doi.org/10.1016/j.mssp.2014.04.034
1369-8001/Crown Copyright & 2014 Published by Elsevier Ltd. All rights reserved.
n
Corresponding author. Tel: +98 991 55725460; fax: +98 631 4423520.
nn
Corresponding author. Tel.: +98 911 3708480; fax: +98 631 4423520.
E-mail addresses: mehdi.sarpoushi@gmail.com (M.R. Sarpoushi),
mahdi.nasibi@gmail.com (M. Nasibi).
Materials Science in Semiconductor Processing 26 (2014) 374–378

2. electrochemical capacitors [12]. In this paper, mechanical
pressing as a fast and easy method was used to fabricate
the electrodes. Then, prepared nanocomposites were eval-
uated as a novel electrode material for electrochemical
capacitors using cyclic voltammetry (CV), electrochemical
impedance spectroscopy (EIS) and scanning electron
microscopy (SEM) techniques.
2. Experimental
2.1. Materials
High purity (99.98%) cerium oxide nanoparticles
(10–30 nm) were purchased from US-nano, USA, graphene
nanoflakes (60 nm, multi-layered) with a purity of 98.5%
from graphene supermarket, USA and polytetrafluoroethy-
lene (o2 μm) from Aldrich, USA. All other chemicals used
in this study were purchased from Merck, Germany.
The mixture containing 15 wt% CeO2, 75 wt% graphene
and 10 wt% polytetrafluoroethylene (PTFE) was well mixed
using ultrasonic in ethanol bath and in a paste form for
60 min. Paste form was chosen for better dispersion of
cerium oxide nanoparticles between graphene nanoflakes.
After drying and powdering, the prepared composite was
pressed onto a 316 L stainless steel plate (5 Â 107
Pa)
which was served as a current collector (surface area was
1.22 cm2
). A steel rod and hollow cylinder of epoxy was
used for pressing and a Teflon paper was used as a
separator at the bottom of the rod during pressing. The
typical mass load of electrode material was 45 mg. 3 M
NaCl, NaOH and KOH solution was used as electrolyte.
2.2. Characterization
The electrochemical behavior of prepared nanocompo-
sites was characterized using CV and EIS tests. Electro-
chemical measurements were performed using Autolab
PGSTAT 302N (Netherlands). CV tests were performed
within the range of À0.55 and þ0.3 V (vs. SCE), using
scan rates of 5, 10, 20, 30, 40, 50, 100, 200, 300, 400 and
500 mV sÀ1
. EIS measurements were also carried out in
frequency range of 100–0.02 Hz at open circuit potential
with an AC amplitude of 10 mV. For better understanding
the effect of the surface morphology and its nature on the
charge storage and charge delivering capability of pre-
pared electrodes scanning electron microscope (TESCAN,
USA) was used.
3. Result and discussion
Graphene with flake morphology and very close inter-
layer distances has a high specific surface area which made
it suitable as an electrode material for supercapacitors.
Fig. 1 shows the surface morphology of the prepared
graphene/cerium oxide nanocomposite. Perpendicular to
these nanoflakes it shows no porosity and is completely
flat. Fig. 1 confirms that the used material is completely
porous in 2D and will be flat in 1D. With this morphology,
it seems the charge storage depends directly on the charge
separation on the upper part (which is the most accessible
surface of the electrode) and on open pore systems (which
are ion-size-dependent and less accessible) simulta-
neously. Ion size, ion diffusion and electron transfer sites
through these pores would affect the activation of these
less accessible surfaces, especially at high scan rates.
Although high molarities (3 M NaOH, KOH and NaCl ) of
these electrolytes can be used but ion mobilities through
the nanopores may be decreased and have adverse effect
on the capacitance. Increasing the ionic radius would
decrease the number of adsorbed ions on the unit surface
area of the electrode and would decrease the stored charge
on the outer Helmholtz layer. The main difference of these
electrolytes is the effective radius of their anions and
cations. Naþ
, Kþ
, ClÀ
and OHÀ
ions have effective radiuses
of 102, 138, 181 and 153 pm, respectively. Therefore the
ratio of interlayer distance of the used graphene (3.36 Å) to
ionic radius (α) of these ions would be 3.29, 2.43, 1.86
and 2.20.
The specific surface area of graphene nanoflakes is high
enough for the large accumulation of ions but very close
interlayer distances (3.63 Å) will limit ions diffusion to
these site. By adding CeO2 particles, flat surface of the
electrode becomes nearly porous; therefore, the specific
surface area of the electrode increases. The CeO2 quasi-
spherical particles could be homogeneously dispersed
across the surface of the graphene nanoflakes, thus
preventing the CeO2 particles from agglomeration and
will effectively accommodate the volume change of CeO2
during the charge–discharge process (Fig. 1). Graphene
nanosheets possess a high electronic conductivity, which
keeps the dispersed CeO2 quasi-spheres connected and
decreases the contact resistance of the active material in
the electrode [13]. As shown in Fig. 2, bare CeO2 showed
high solution resistance and low charge transfer resistance
in 3 M NaCl electrolyte which was improved by the
addition of conductive materials such as graphene. Fig. 3
Fig. 1. Scanning electron microscopy image obtained from CeO2/graphene.
M.R. Sarpoushi et al. / Materials Science in Semiconductor Processing 26 (2014) 374–378 375

3. (a) exhibits cyclic voltammetry curves obtained at the scan
rate of 100 mV sÀ1
in 3 M KOH, NaOH and NaCl electro-
lytes. The maximum area under the cyclic voltamogrames
was obtained in NaCl electrolyte. Also Fig. 3(b) shows
reasonable performance of the 3 M NaCl electrolyte which
was selected for further investigation. Capacitance of the
graphene/CeO2 compared to CeO2 was increased about
3.05 F gÀ1
(Fig. 3(a)) due to the conductive behavior of
graphene nanoflakes and double layer formation on the
electrode (Fig. 4). Capacitance decreases with increasing
the scan rate (due to energy losses). As the scan rate
increases, the capacitance vs. potential relation would
deviate from the classical square wave form expecting
for a pure capacitor (Fig. 5). This phenomena is due to the
resistance effects down the pores [14]. In practice, the
charging current may be maintained in a voltage range
across the interface where faradaic decomposition of the
solution begins to take place at that electrode. The double-
layer continues to be increasingly charged with rise of ΔV,
but the current i, gets divided into two components: idl
and iF, where iF is the Faradaic current, which increases
exponentially with ΔV when ΔV exceeds a value corre-
sponding to a thermodynamic reversible potential for
solution decomposition at the electrode [15]. In this case,
pseudocapacitance behavior of the electrode occurred at
scan rates below 20 mV sÀ1
. In addition, efficiency is
another important parameter affecting the capacitance at
high scan rates. As the scan rate increases, loss of energy
increases and the stored charge on the electrode surface
decreases causing the capacitance to decrease.
The capacitance of each electrode was calculated from
the CV curves using:
Cs ¼
Z
i dv=m:s:Δv ð1Þ
where Cs (F gÀ1
) is the specific capacitance,
R
i dv is the
integrated area of the CV curve, m (g) is the mass of the
active material, Δv (V) is the potential range, and s (V/s) is
the scan rate.
In practice, some frequency dependence is commonly
observed, i.e., the phase angle for the double-layer
Fig. 2. Nyquist diagram obtained from CeO2in 3 M NaCl electrolyte
at OCP.
Fig. 3. (a) CV curves at 100 mV sÀ1
and (b) Nyquist diagrams obtained
from CeO2/graphene electrodes in different electrolytes at OCP.
Fig. 4. Capacitive behavior obtained from CeO2, graphene and CeO2/
graphene.
Fig. 5. CV curves obtained from CeO2 using different scan rates in 3 M
NaCl electrolyte.
M.R. Sarpoushi et al. / Materials Science in Semiconductor Processing 26 (2014) 374–378376

4. capacitance may not have the ideal value of 901 at all
frequencies, and potential dependence DC leakage can also
occur. Deviation from ideal capacitative behavior can arise
when there is some dielectric loss associated with the
solvent orientation polarization in the double-layer dielec-
tric (at very high frequency) and/or when there are some
slow anionic chemisorption processes that lead to lower
frequency losses. In either case, there is energy dissipation
in the charging and discharging cycles at an incompletely
polarizable electrode interface. Another source of nonideal
behavior is distribution of the double-layer capacitance
over a porous electrode surface. In a practical electroche-
mical capacitor device, frequency dependence of the over-
all capacitance is generally observed and is due, in
addition, to the “porous electrode” effect, to coupling with
other equivalent series resistance (ESR) components [15].
The efficiency of an electrochemical capacitor is related to
the loss factor, dc [16]:
dc ¼ tan ðδÞ ¼ tan ð901ÀφÞ ð2Þ
The power dissipated as heat in the internal resistance
is determined by the cosine of φ. The minimum loss factor
of 0.492 (φ¼71.361) is achieved for the CeO2/graphene
electrode at a frequency of 20 mHz (Fig. 6). The equivalent
circuit of the CeO2/graphene electrode is shown in Fig. 7.
The equivalent circuit contains the bulk solution resistance
element (Rs), the constant phase element (Q1) which is
parallel with Charge transfer resistance (Rct), a Warburg
diffusion element (W) attributed to the diffusion of ions,
and the low frequency capacitance (Q2), which is in
parallel with (Rl), the leakage resistance. Table 1 shows
numerical values for the equivalent circuit of CeO2/gra-
phene electrode.
In charge and discharge cycles, total charge can be
written as the sum of an inner charge from the less and an
outer charge from the more accessible reaction sites, i.e.,
qn
T ¼qn
Iþqn
O, where qn
T, qn
I and qn
O are the total charge
and charges related to inner and outer surface, respec-
tively [18]. Extrapolation of qn
to s¼0 from 1/qn
vs. s1/2
plot
(Fig. 8(a)) gives total charge qn
T which is charge related to
the entire active surface of electrode. In addition, extra-
polation of qn
to s¼1 (sÀ1/2
¼0) from qn
vs. sÀ1/2
plot
(Fig. 8(b)) gives the outer charge qn
O, which is charged on
the most accessible active surface. Prepared electrodes
show the ratio of outer to total charge (qn
O/qn
T) of 0.084
in NaCl electrolyte which confirms the low current
response on voltage reversal.
The cycle stability was evaluated by repeating the CV at
a scan rate of 100 mV sÀ1
for 500 cycles. Prepared electro-
des were found to exhibit excellent stability over the entire
cycles. It is very interesting that the anodic and cathodic
currents increased after 500 cycles as well as the cyclic
voltammetry curves remained in their rectangular-shaped
profiles (Fig. 9(a)).The charge stored on the electrode after
500 cycles increased by 37% which may be used to the
large amount of oxygen vacancies within the cerium oxide
formed during reduction conditions [17]. Some of the
cerium (IV) oxide is also reduced to cerium (III) oxide
Fig. 6. (a) Bode plot obtained from CeO2/graphene electrode in 3 M NaCl
electrolyte at OCP.
Fig. 7. Equivalent circuit of CeO2/graphene electrode in 3 M NaCl elec-
trolyte at OCP.
Table 1
Electrochemical parameters obtained from electrochemical impedance spectroscopy measurement from graphene/cerium oxide nanoparticle electrodes.
Rs (Ω) Q1 (S.sec0.912
), n Rct (Ω) W (S.sec0.5
) Q2 (S.sec0.25
), n Rl (Ω)
8.351 0.084, 0.912 732 109.8 0.123, 0.250 6.436 Â 106
Fig. 8. Extrapolation of q to ν¼0 from the qÀ1
vs. ν0.5
plot given the total
charge and (b)extrapolation of q to ν¼1 from the q vs. νÀ0.5
plot given
the outer charge for CeO2/graphene electrode.
M.R. Sarpoushi et al. / Materials Science in Semiconductor Processing 26 (2014) 374–378 377

5. under these conditions, which consequently increases the
electronic conductivity of the material. Therefore increas-
ing the conductivity of the proposed material may be an
acceptable reason of increasing the capacitance after 500
cycles. EIS tests were used to evaluate the electrode
changes (Fig. 9(b)). This behavior confirmed by EIS evalua-
tions. During the 500 charge/discharge cycles, the equiva-
lent series resistance decreased (Fig. 9(b)) and Nyquist
plots shifted a little toward slower values.
4. Conclusion
Studies confirmed the presence of flat and porous
surfaces, simultaneously, and it seems the charge storage
depends directly on the charge separation and Faradic
reactions on the surface of the electrode. Also, charge/
discharge process changed partly the CeO2 particles struc-
ture which act as a pseudo-material from CeO2 to Ce2O3.
Prepared electrode showed a low ratio of the outer charge
to total charge (qO/qT) of 0.084 which confirms the low
current response on voltage reversal. Proposed electrodes
showed a maximum capacitance of as high as 11.09 F gÀ1
at 5 mV sÀ1
in 3 M NaCl electrolyte and exhibit excellent
stability over the first 500 cycles (charge storage ability
increase 37% after 500 cycles).
References
[1] K. Zhang, L.L. Zhang, X.S. Zhao, J. Wu, Chem. Mater. 22 (2010)
1392–1401.
[2] S.M. Paek, E. Yoo, I. Honma, Nano Lett. 9 (2009) 72–75.
[3] L Dong, R.R.S. Gari, Z. Li, M.M. Craig, S. Hou, Carbon 48 (2010)
781–787.
[4] C. Chen, D. Zhao, X. Wang, Mater. Chem. Phys. 97 (2006) 156–161.
[5] Y. Zhang, Y. Gui, X. Wu, H. Feng, A. Zhang, L. Wang, et al., Int.
J. Hydrog. Energy 34 (2009) 2467–2470.
[6] L. Qu, Y.J.B. Baek, L. Dai, ACS Nano 4 (2010) 1321.
[7] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.
V. Dubonos, I.V. Grigorieva, A.A. Firsov, Science 306 (2004) 666–669.
[8] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, M.I. Katsnelson,
I.V. Grigorieva, S.V. Dubonos, S. Ardizzone, G. Fregonara, S. Trasatti,
Electrochim. Acta 35 (1990) 263–267.
[9] A.K. Geim, K.S. Novoselov, Nat. Mater. 6 (2007) 183–191.
[10] A.K. Geim, Science 324 (2009) 1530–1534.
[11] D.D. Zhao, S.J. Bao, W.J. Zhou, H.L. Li., Electrochem. Commun. 9
(2007) 869–874.
[12] G. Wang, J. Bai, Y. Wang, Z. Ren, J. Baic, Scr. Mater. 6 (2011) 339–342.
[13] F. Zhou, X. Zhao, H. Xu, C. Yuan, J. Phys. Chem. C 111 (2007)
1651–1657.
[14] S. Ardizzone, G. Fregonara, S. Trasatti, Electrochim. Acta 35 (1990)
263–267.
[15] B.E. Conway, Electrochemical Capacitors Scientific Fundamental and
Technological Applications, Kluwer Academic/Plenum, New York,
1999.
[16] R. Kötz, M. Carlen, Electrochim. Acta 45 (2000) 2483–2498.
[17] M. Itagaki, S. Suzuki, I. Shitanda, K. Watanabe, H. Nakazawa, J. Power
Sources 164 (2007) 415–424.
[18] I. Danaee, M. Jafarian, F. Forouzandeh, F. Gobal, MG. Mahjani, Int. J.
Hydrog. Energy 34 (2009) 859–869.
Fig. 9. (a) CV curves obtained at 100 mV sÀ1
and, and (b) Nyquist plots
after different cycles for CeO2/graphene in 3 M NaCl electrolyte at OCP.
M.R. Sarpoushi et al. / Materials Science in Semiconductor Processing 26 (2014) 374–378378