Electrochimica Acta 163 (2015) 196–203

Enhancement of the energy storage properties of supercapacitors usinggraphene nanosheets dispersed with macro-structured porous copperoxide

Riyaz A. Dar a, Gowhar A. Naikoo b, Pramod K. Kalambate a, Lily Giri c, Farid Khan b,Shashi P. Karna c, Ashwini K. Srivastava a,*aDepartment of Chemistry, University of Mumbai, Vidyanagari, Mumbai-400098, IndiabNanomaterials discovery laboratory, Department of Chemistry, Dr. Harisingh Gour Central University, Sagar, MP 470003, IndiacUS Army Research Laboratory, Weapons and Materials Research laboratory, RDRL-WM, Aberdeen Proving Ground, MD-21005-5069, USA

A R T I C L E I N F O

Article history:Received 13 November 2014Received in revised form 13 February 2015Accepted 14 February 2015Available online 17 February 2015

Keywords:Macroporous copper oxideGraphene nanosheetsPseudocapacitorEnergy density

A B S T R A C T

Graphene nanosheets (GN) dispersed with macroporous copper oxide (macroCuO) was investigated as anelectrode material for supercapacitors. A facile and cost-effective synthesis approach was used to preparemacro-structured porous copper oxide monoliths via modified Sol–Gel route. 1, 3, 5-trimethylbenzenewas used as an organic structural directing agent to enhance the pore size, pore volume, pore density andsurface area of the resulting CuO hybrid templated with Pluronic P-123. GN/macroCuO nanocompositewas prepared by ultrasonication of the GN and macroCuO. The macroCuO and GN/macroCuOnanocomposite were characterized using various surface analytical techniques. Electrochemicalperformance of the composite electrode was investigated using cyclic voltammetry and chronopo-tentiometry. GN/macroCuO/GCE showed pseudocapacitance behaviour due to the Faradaic type ofcapacitance involving redox process between Cu (0) and Cu (II) of porous copper oxide network.Electrochemical measurements revealed the maximum specific capacitance, energy density and powerdensity of 417 F g�1, 58 Wh kg�1 and 17.85 kW kg�1, respectively for the supercapacitor based on GN/macroCuO nanocomposite electrode at a current density of 0.9 A g�1. The fabricated supercapacitordevice exhibited excellent cycle life with 91.4% of the initial specific capacitance retained after1000 cycles. The results suggest that the hybrid composite is a promising supercapacitor electrodematerial.

ã 2015 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Electrochimica Acta

journal homepa ge: www.elsev ier .com/locate /e lectacta

1. Introduction

Energy storage, an intermediate step to the versatile, clean,and efficient use of energy, has received worldwide concern andincreasing research interest [1]. Among the various power sourcedevices, supercapacitors (SCs), also known as electrochemicalcapacitors, have attracted considerable attention over the pastdecade due to their high power densities, fast charging/discharging rate, and long cycle life compared to secondarybatteries and fuel cells, and also, higher energy densities than theconventional dielectric capacitors [1–3]. The most commonlyused materials for electrical double layer capacitors (EDLCs) arecarbonaceous materials including activated carbon [4], graphene[5] and carbon nanotubes [6], however, relatively low energy

* Corresponding author. Tel.: +91 22 26543570.E-mail address: aksrivastava@chem.mu.ac.in (A.K. Srivastava).

http://dx.doi.org/10.1016/j.electacta.2015.02.1230013-4686/ã 2015 Elsevier Ltd. All rights reserved.

density, suffering from poor rate capability and/or poor capaci-tance retention upon cycling has limited their applications [2]. Incontrast, transition metal oxides or hydroxides with variablevalence, such as NiO [7], Co3O4 [8–10], MnO2 [11], Ni(OH)2 [12],and Co(OH)2 [13] can provide higher energy density for super-capacitor. Such materials not only store energy like electrostaticcarbon materials but also exhibit electrochemical faradaicreactions between electrode materials and ions within appropri-ate potential windows [2].

Among the transition metal oxides, porous copper oxide (CuO)is noteworthy to explore as a promising candidate due to its lowcost, abundant resources, non-toxicity, chemically stable, and easypreparation in diverse shapes of nano-sized dimension. Manymethods have been developed to synthesize various CuO nano-structures [14–19], and its supercapacitance behavior has attractedextensive research interest. For instance, Patake et al. [20]synthesized the porous amorphous copper oxide thin films whichexhibited a specific capacitance of 36 F g�1 in 1 M Na2SO4

Scheme 1. (a) Cu (NO3)2 .3H2O solutions mixed with Pluronic P-123 solutionfollowed by addition of TMB as a SDA (b) Light blue colour solution formed (SOL). (c)Heat cum magnetic stirring at 55 �C followed by ageing for 2–3days resulting intocross linking of precursor with P-123 (Xerogel) (d) Calcination of the resultingxerogel at 650 �C results into porous copper oxide monoliths.

R.A. Dar et al. / Electrochimica Acta 163 (2015) 196–203 197

electrolyte whereas, Dubal et al. [21] reported copper oxidemultilayer nanosheets thin films with a specific capacitance of43 F g�1 in 1 M Na2SO4 electrolyte. Zhang et al. [22,23] found thatCuO with flower-like nanostructures displayed a specific capaci-tance of 133.6 F g�1 in 6 M KOH electrolyte, which is about three-fold higher than commercial CuO powder. However, the specificcapacitance of CuO is still lower than other transition metal oxidesand exhibited unstable cycling performances. Wang et al. [24] havereported the synthesis of CuO nanosheets arrays directly grownonto nickel foam with higher specific capacitance of 569 F g�1,whereas the method employed is relatively complicated. Although,nanoscale CuO particles possess large surface area and relativelyhigh specific capacitance, the microstructure is easily damagedduring electrochemical cycling, resulting in a relatively poorelectrochemical stability. Recently, composites based on carbona-ceous materials (graphene and graphene oxide) and copper oxides[25–27] have been studied for improved capacitance. Integration ofgraphene into functional architectures and composites hascurrently been an active area of research [25,28].

Given this situation, a promising strategy would be to combinesupercapacitive material, graphene with low-cost pseudo-capaci-tive metal oxides such as CuO, which offers both a cost advantageand potentially a high performance benefiting from both mecha-nisms of electric double layer capacitance and pseudo-capacitance[2,29]. To the best of our knowledge, macro-structured porouscopper oxide monoliths synthesized via modified Sol–Gel routewithout using any acidic or basic medium with a reasonablespecific capacitance and better cycling performance have seldombeen reported.

Herein, we report a facile, cost-effective and scalable synthesisapproach to prepare macro-structured porous copper oxidemonoliths (macroCuO) via modified Sol–Gel route. Trimethylbenzene (TMB) is used in the synthesis as an organic structuraldirecting agent to enhance the pore size, pore volume, pore densityand surface area of the resulting CuO hybrid. This product isdispersed with graphene nanosheets to give a nanocompositematerial (GN/macroCuO) to be used as electrodes for super-capacitors.

The incorporation of macroCuO into graphene layers isexpected to improve electrolyte–electrode accessibility andelectrode conductivity by reducing the agglomeration of GN.GN/macroCuO composite has been prepared by ultrasonication ofGN and macroCuO. Supercapacitor devices are fabricated usingGN/macroCuO nanocomposite electrode materials and perfor-mance studies are conducted.

2. Experimental

2.1. Chemicals and Materials

Copper nitrate trihydrate (Sigma Aldrich as a precursor), softtemplate Pluronic P-123, (Sigma Aldrich), structural directingagent 1, 3, 5-Trimethylbenzene, TMB (Merck). Graphene (99.5 %purity) was purchased from Sisco Research LaboratoriesPvt. Ltd.

Porous copper-oxide monoliths were prepared by dissolving2.0 g Cu (NO3)2.3H2O (50 wt%, Sigma–Aldrich) in 2.0 g of ultrapurewater (50 wt %) and 2.0 g of Pluronic P-123 (14.81 wt%, Mw= 5800,Aldrich) in 11.5 g of ultrapure water (85.19 wt %) followed byaddition of 2 g of TMB at 25 �C. The gel was heated for 1 h at 55 �C ona magnetic stirrer to form the paste which gradually became lightblue in colour. The resulting gel was aged for 2–3 days at roomtemperature (xerogel formation) and then calcined at 650 �C for 2 hat a heating rate of 1 �C/min followed by cooling at a rate of 1 �C/min to room temperature in an ELLITE furnace. The overall protocolfor the synthesis is shown in Scheme 1.

2.2. Fabrication of supercapacitor based on GN/macroCuOnanocomposite modified glassy carbon electrode (GCE)

GCE was mechanically polished with a 0.05 mm aluminaslurry and then sequentially sonicated in dilute nitric acid,anhydrous ethanol and redistilled water for 15 min. Next, thecleaned GCE was dried under nitrogen stream. GN andmacroCuO in the ratio of 1:1 by weight were dispersed in100 ml of distilled water by ultrasonication for 1 h to obtain ahomogeneous GN/macroCuO suspension. Finally, the solid wasfiltered, washed several times with distilled water and alcoholand dried at 100 �C for 12 h in a vacuum oven. 10 mg of thiscomposite was dispersed in 5 mL dimethylformamide bysonicating in an ultrasound bath for 30 min to form a stablesuspension. The suspension of 10.0 mL was cast onto the GCEsurface by a micropipette, and then thoroughly dried under aninfrared lamp. Subsequently, the electrode was rinsed severaltimes by distilled water and dried in air before use; the finalobtained electrode was denoted as GN/macroCuO/GCE. Forcomparison, the GN/GCE was prepared using the GN dispersionalone by following the same procedure.

2.3. Structure characterization

After calcination of the xerogel, XRD patterns of the samplewas obtained on a Bruker D-8 advance diffractometer in thediffraction angle range 2u = 20–120�, using monochromatic Cuka radiation (l = 1.5410). The surface pore morphology andaverage pore diameter of the obtained sample was carried outby field emission scanning electron microscopy (FESEM) usingS-4800 field emission SEM system (FEI Quanta 200) operatingat 20.0 kV equipped to perform elemental chemical analysis byenergy dispersive X-ray spectroscopy (EDX). The FFT, SAEDpatterns of macroporous CuO and HRTEM image of the resultingGN/macroporous CuO composite was investigated by JEOL JEM2100F microscope operated at 200 kV using an OriusSC1000 camera. Nitrogen adsorption desorption measurementsof the sample was performed by physisorption of N2 at 77 Kover a Micromeritics ASAP 2010. Before calcination of thesample TGA/DTA and FTIR analysis were carried out by Perkin-Elmer thermal analyzer (using alumina reference crucible at theheating rate of 10 �C/min) and Shimadzu-8400S spectrometerrespectively

198 R.A. Dar et al. / Electrochimica Acta 163 (2015) 196–203

2.4. Electrochemical measurements

All voltammetric experiments were performed with a CHIelectrochemical workstation (CH Instruments Model CHI1100Bseries). The conventional three-electrode geometry was adopted.The working electrode was a GN/macroCuO modified glassy carbonelectrode (GN/macroCuO/GCE), and the auxiliary and referenceelectrodes were platinum foil and Ag/AgCl respectively. Thecapacitive behavior of GN/macroCuO/GCE was studied by cyclicvoltammetry (CV) in 1 M KOH electrolyte at room temperature. CVmeasurements were performed on the three-electrode cell in thevoltage window between �1.6 and 1.2 V at different scan rates.Chronopotentiometry was used to study the galvanostaticcharging–discharging characteristics of the material performedwith different current densities to reflect the rate ability.Electrochemical impedance measurements were performed in1 M KOH electrolyte using Eco Chemie, Electrochemical WorkStation, model Autolab PGSTAT 30 using GPES software, version4.9.005 and Frequency Response Analyser, software version 2.0,respectively and measured in the frequency range from 106Hz to0.1 Hz. An Ag/AgCl, 3 M KCl and a platinum electrode were used asreference and counter electrodes, respectively.

Measurement of capacitance in the development of electro-chemical supercapacitors is necessary to evaluate performance,application and trouble-shooting when designing and re-design-ing active electrode materials. Determining the capacitance (C) orspecific gravimetric capacitance (Cp) from CV experimental resultsoften requires the integration of the current response in a cyclicvoltammogram and can then be calculated by the followingequation [30–32].

C ¼RIdv

RV:n

(1)

where, C is capacitance, I is the average current in the complete CVloop response (A), DV = Vf� Vi (where, Vf and Vi are the integrationpotential limits of the voltammetric curve. v is the potential scanrate (V s�1). The specific capacitance of the material has beencalculated from the CV curves according to the following equation:

Csp ¼RIdv

RV:n:m

(2)

where Csp (F g�1) is the specific capacitance, m (g) is the mass of theelectroactive materials in the electrode.

The discharge capacitance (C) of the electrodes was alsocalculated at the constant current density using chronopotenti-ometry from the slope of the discharge curve applied differentcurrent densities using the following equation [30–32].

Fig. 1. (A) The TG-DTA analysis of as-synthesized macroCuO sample. (B) Nitrogen sorpticalcination of copper nitrate/P-123/TMB xerogel.

Csp ¼ Im

� dtdv

; (3)

where Csp is the specific capacitance in (F g�1), I the dischargecurrent in ampere (A) and dt

dv is the inverse of the slope of thedischarge curve in volts per second (V s�1) and m is the mass of theelectroactive material. The mass of as synthesized electrodematerial taken was 20 mg. The maximum energy density valueshave been calculated from the following equation:

E ¼ 12CspV

2i (4)

where Csp is the specific capacitance and Vi is the initial voltage ofthe discharge curve [33].

3. Results and Discussion

3.1. Characterization of macroporous CuO and GN/macroCuOcomposite

The TG-DTA profile of the as prepared CuO-P123-TMB (sampleweight taken = 10.502 mg) sample (Fig. 1A) was carried out beforecalcination. It showed a 2.08% (0.219 mg) wt. loss upto 155 �C dueto the removal of physically absorbed water followed by 29.38%(3.087 mg) wt. loss at 155–230 �C associated with decompositionof copper (II) nitrate into CuO [34]. The concomitant release of O2

facilitated oxidative decomposition of Pluronic P123-TMB gelbetween 230–375 �C with a considerable mass loss of 38.47%(4.04 mg) [34]. Subsequently, between 375–430 �C, another wt.loss of 2.01% (0.02 mg) is measured, due to the combustion ofremaining carbon species, and formation of CuO monolith. Theresidual CuO monolith left is 28.05 wt% (3.136 mg). In DTA profile,the two exothermic peaks obtained at ca. 190.35 �C and 351.28 �Care corresponding to the decomposition of P123-TMB and thecombustion of remaining carbon species [35], respectively.

FTIR study has shown bands at 2890 cm�1, 1375 cm�1 and1120 cm�1, corresponding to stretching frequencies of C—H, C—Cand C—O of P-123 (Fig. 2A) which shift to lower side in CuO/P-123/TMB (Fig. 2B). However, a band at 3460 cm�1 due to O—Hstretching almost disappears in CuO/P-123/TMB as heteroatomsare introduced during the process of synthesis, indicates theinteraction of Cu2+cations by the Ethylene oxide (EO) groups of P-123 by a combination of electrostatic and hydrogen bondinginteractions and forms ordered monoliths [34,36]. So, the use ofnon ionic surfactants provides an environmentally benign andgreen synthetic approach for the fabrication of porous metal andmetal oxide sponges while avoiding the need for acid concen-trations i.e. green synthesis of porous materials (Scheme 1).

on isotherm for macroporous copper oxide (macroCuO) monoliths prepared by the

Fig. 2. Fourier Transform Infrared spectroscopic (FT-IR) studies of (A) Pure PluronicP-123 and (B) macroCuO xerogel carried out before calcination of copper nitrate/P-123/TMB.

R.A. Dar et al. / Electrochimica Acta 163 (2015) 196–203 199

Fig. 3A shows well resolved XRD peaks of CuO sample takenafter calcination. It shows reflections at d spacings of 2.75, 2.52,2.32, 1.86, 1.71, 1.58, 1.50, 1.41, 1.40, 1.30, 1.26, and 1.16 Å,corresponding to (110), (0 0 2), (2 0 0), (�2 0 2), (0 2 0), (2 0 2),(�113), (0 2 2), (�3 11), (311), (0 0 4), and (31 2) lattice planes of abase-centred monoclinic unit cell (tenorite) crystal (JCPDS, JointCommittee on Power Diffraction Standard, 03-0867). Fig. 4A showsthe FESEM image of porous CuO monoliths revealing the impact ofsoft template P-123 and swelling agent TMB on the surface poremorphology. The as synthesized sample is hierarchically macro-structured spheres with a continuous interconnected scaffold anda disordered network of pores. The magnified view of the samplereveals that the surface of the architecture possesses many poreswith pore diameters ranging from 0.43 mm to 1.50 mm. TMBwas added as an organic structural directing agent toCuO-P123 xerogel, responsible for enhancing pore density andpore diameter. This may be attributed to the change in the viscosityof the composite gels and their rate of decomposition orsolubilization of nonpolar TMB inside the surfactant assembly[36–38]. From the HRTEM images, the bright spot in the FastFourier Transform (FFT) pattern (Fig. 4B) and the selected areadiffraction (SAED) pattern (Fig. 4C) gives information about the

Fig. 3. (A) PXRD pattern of macroporous copper-oxide monoliths obtained aftercalcination of copper nitrate/P-123/TMB xerogel. (B) PXRD of Graphene-macro-porous copper oxide monoliths composite.

crystal structure of macroporous CuO particles and the averageparticle size was found to be 5 nm. For further investigation ofmacroporous nature of CuO-P123-TMB materials, N2 sorptionisotherms of the CuO-P123-TMB sample was closely studied at 77 Kover a Micromeritics ASAP 2010 as shown in Fig. 1B. The assynthesized sample exhibits isotherms of type II of the IUPACclassification featuring a macroporous characteristic of the samplein agreement with the FESEM results. The BET specific surface areaof macroCuO was found to be 67 m2g�1. Such types of hierarchicalsurface morphologies with well-developed pore structures areadvantageous for energy storage applications since large porechannels permit rapid electrolyte transport, while the small poresprovide more active sites for chemical reactions [1].

The synthesized GN/macroCuO composite was also character-ized by XRD, SEM/TEM and EDAX studies. Fig. 1B shows the XRDpattern of GN/macroCuO composite.

The the XRD pattern of the as-prepared GN/macroCuO shows anadditional peak at 26.52� attributed to the (0 0 2) plane of thehexagonal graphite structure, suggesting that the graphene isincorporated into the CuO nanocomposites. The other diffractionpeaks of as-synthesized GN/macroCuO composite can be ascribedto the well-crystallized macroCuO. Fig. 2D shows the SEM image ofgraphene layers interacting with each other to form an open poresystem, through which electrolyte ions easily access the surface ofgraphene to form electric double layers. Fig. 2E shows the SEMimage of GN/macroCuO composite in which macroporous CuO isencapsulated within the graphene layers which not only suppressthe agglomeration and restacking of GN but also increase theavailable surface area of the GN, leading to high electrochemicalactivity. On the other hand, GN provides a support to macroporousCuO to induce the uniform dispersion and controlled morphologyon the surface of graphene with high chemical functionality. Thefinal porous oxide encapsulated in graphene and the graphene-supported oxide can form a perfect integrated structure with adeveloped electron conductive network and shortened iontransport paths. In addition, HRTEM image (Fig. 2F) shows therandom distribution of the CuO monoliths on graphene sheets anda good dispersion of macroporous CuO monoliths with graphenewhich is in good agreement with the FESEM results. The formationof GN/macroCuO composite was further characterized by EDAX.The EDAX spectra of macroporous CuO (Fig. 5A) obtained aftercalcination show the peaks corresponding to Cu and O elements,confirming the formation of pure copper oxide monoliths. EDAXdata indicates the porous copper oxide is nearly stoichiometric.The wt.% of copper and oxygen calculated from EDAX are 84.5 and15.4, respectively. A small intense peak at 0.27 is due to a littleamount of remaining carbon during calcination. However, nearlystoichiometric ratio of copper and oxygen in the EDAX spectrumconfirms the formation of porous copper oxide monoliths. GN/macroCuO composite obtained showed the peaks corresponding toC, Cu and O (Fig. 5B). The wt.% of Cu, O and C calculated from EDAXare 23.82, 4.33 and 71.85 respectively, revealing a loading of CuO onGN sheets, which is consistent with the SEM observation. Thisstudy confirms a good dispersion of porous copper oxide monolithswith graphene.

3.2. Electrochemical properties of the graphene-macroporous copperoxide monolith (GN/macroCuO) composite

As shown in Fig. 6A (a), cyclic voltammogram (CV) of graphenenanosheets modified GCE exhibits almost rectangular typebehaviour, indicative of good charge propagation at the electrodesurface following the electric double-layer charging mechanism.Fig. 6A (b and c) are the cyclic voltammograms of macroCuOmodified GCE and GN/macroCuO modified GCE. Fig. 6A (b) shows arectangular nature of CV with additional two anodic and cathodic

Fig. 4. (A) SEM image of macroCuO monolith, (B) Fast Fourier Transform (FFT) pattern of macroCuO monolith, (C) Selected area diffraction (SAED) pattern of macroCuOmonoliths templated by Pluronic P123, (D) SEM image of GN, (E) SEM image of GN/macroCuO composite and (F) HRTEM image of GN/macroCuO monolith composite.

Fig. 5. EDAX of (a) porous copper oxide monoliths. (b) Graphene-porous copper oxide monoliths composite.

200 R.A. Dar et al. / Electrochimica Acta 163 (2015) 196–203

peaks at macroporous CuO modified GCE indicating the pseudo-capacitive characteristics of macroCuO. The peaks are due to theFaradaic redox reactions attributed to Cu (0) and Cu(II) of porousCuO [24,25,39,40]. The cathodic peak at �0.40 V (C1) corresponds

to the reduction of CuO to Cu (I) and the peak at 0.14 (C2) resultsfrom the reduction of Cu(I) to Cu(0)). The anodic signal (A1) at�0.25 V represents the oxidation of Cu (0) to Cu (I) and peak (A2) at0.35 V corresponds to oxidation of Cu(0) to Cu(I) and Cu(0) to Cu

Fig. 6. (A) Cyclic voltammograms at (a) GN/GCE and (b) macroCuO/GCE and (c) GN/macroCuO/GCE in 1 M KOH at a scan rate of 100 mV s�1. (B) Cyclic voltammograms of GN/macroCuO/GCE at different scan rates of 10, 30, 50, 100, 150 and 250 mV s�1 (a ! f). (C) Galvanostatic charge–discharge curves of (a) GN/GCE, (b) macroCuO/GCE and (c) GN/macroCuO/GCE at a current density of 2.0 A g�1. (D) Galvanostatic charge–discharge curves of GN/macroCuO/GCE at different current densities of (a) 0.9 A g�1, (b) 1.25 A g�1,(c) 2.0 A g�1, (d) 3.0 A g�1 and (e) 5.0 A g�1 in 1 M KOH.

R.A. Dar et al. / Electrochimica Acta 163 (2015) 196–203 201

(II). This explains why the oxidation peak A2 is larger in currentwith respect to A1 [24,40].

However, a large charge propagation is observed at GN/macroCuO modified GCE because of the combination of Faradaiccurrent due to macroCuO and electric double layer chargingprocess due to the porous nature of electrode material (macroCuO)and graphene nanosheets (Fig. 6A, c). Therefore, the totalcapacitance of GN/macroCuO modified GCE is the sum ofpseudocapacitance due to Faradaic redox reactions of macroCuO

Fig. 7. (A) Specific capacitance of GN/GCE, macroCuO/GCE and GN/macroCuO/GCE as a fGCE and GN/macroCuO/GCE at a current density of 0.9 A g1�. (Insert a) is the charging

and the electrical double layer capacitance due to the porousnature of macroCuO and graphene nanosheets [41].

CVs of GN/macroCuO modified GCE under the different sweeprates are shown in Fig. 6(B). It is observed that the area under theCV curves increases with the increase in scan rate. However, anodicand cathodic peaks are found to shift towards positive and negativepotentials, respectively demonstrating the quasi-reversible natureof the redox reactions. The two anodic as well as the two cathodicpeaks at higher scan rates tend to merge to a single broad anodic

unction of the current density. (B) The cycling stability tests of GN/GCE, macroCuO/discharging profile of first 5 cycles of the GN/macroCuO/GCE.

Fig. 8. Nyquist plots for GN, macroCuO and GN/macroCuO composite basedelectrodes at an amplitude of 5 mV vs. Ag/AgCl over the frequency range from 0.1 Hzand 106Hz. Z': real impedance. Z": imaginary impedance.

202 R.A. Dar et al. / Electrochimica Acta 163 (2015) 196–203

and cathodic peak, respectively. Therefore, at higher scan rates, theelectrical double layer capacitance predominates which is due tothe fast diffusion of electrolyte ions into the pores of macroporousmaterial [42]. The specific capacitances of GN/GCE, macroCuO/GCEand GN/macroCuO modified GCE obtained from CV loops are 128,223 and 419 F g�1, respectively at a scan rate of 100 mV s�1. Thesefindings suggest that the higher specific area and suitable porosityfor easy insertion/de-insertion of ions mostly contribute toexcellent redox reactions and are strongly dependent on thenanostructures of the electrode material.

Fig. 6C depicts a comparison of galvanostatic charge–dischargecurves of GN/GCE, macroCuO/GCE and GN/macroCuO modifiedGCE at the current density of 0.9 A g�1 in 1 M KOH. It is observedthat the nonlinear voltage–time curve for macroCuO (Fig. 6C, b)proved the pseudocapacitance behaviour of the macroCuO/GCE,which originates from the redox reaction between Cu (II) and Cu(0) [43,44]. This is in accordance with the cyclic voltammetryresult, where couples of redox current peaks are observed. Further,the discharging behaviour of macroCuO shows a small IR voltagedrop and the discharge curve consists of two sections, a fastpotential drop followed by a slow potential decay and then a slowdecay of voltage. This accounts for the internal resistance ofelectrode. The obvious non-linear shape of the discharge curve(Fig. 6C, b) and the non-ideal rectangular shape of the CV (Fig. 6A,b) reveal that the macroCuO shows pseudocapacitive behavior.However, the charging/discharging profile for the GN/macroCuOmodified GCE (Fig. 6C, c) is somewhat linear compared tomacroCuO modified GCE.

Fig. 6D shows the charging and discharging characteristics ofGN/macroCuO modified GCE as a function of different currentdensities at (a) 0.9 A g�1, (b) 1.25 A g�1, (c) 2.0 A g�1, (d) 3.0 A g�1

and (e) 5.0 A g�1. The charge/discharge profile of the GN/macroCuOmodified GCE remains within the potential range of -0.5–0.5 V. Forall, GN/GCE, CuO/GCE and GN/macroCuO/GCE, the specificcapacitances decrease with increasing the current densities,because of the presence of fewer electrolyte ions in the innerspace of electrodes at high current densities.

On increasing the current density from 0.9 A g�1 to 5 A g�1, only8.4% specific capacitance was retained at 5 A g�1 for GN/GCE and53.96 % for CuO/GCE in comparison with the retained specificcapacitance of 79.18% for GN/CuO/GCE at the same current density(Fig. 7A). This shows the better performance, i.e. excellentdischarge efficiency and electrochemically dynamic propertiesfor GN/CuO/GCE [45], which are in consistent with the CV results.This is because the macroporosity system of GN/macroCuOmodified GCE can form “ion-buffering reservoir” to reducediffusion path of electrolyte ions, allowing higher rate of solutioninfiltration [38].

The maximum specific capacitances at 0.9 A g�1 for GN/GCE,CuO/GCE and GN/macroCuO/GCE are estimated to be 126 F g�1,226 F g�1 and 417 F g�1 respectively. In addition, the chargedischarge duration for GN/macroCuO/GCE is greater than that ofGN/GCE and CuO/GCE (Fig. 6C) indicating a higher specificcapacitance of the former.

Pseudocapacitors can store much more energy than electricdouble-layer supercapacitors because pseudocapacitors not onlyhave fast and reversible redox reactions at the electrode surface, butalso have the electric double-layer capacitance. The maximumenergy density for GN/GCE, macroCuO/GCE and GN/macroCuO/GCEis calculated to be 17.51 Wh kg�1, 31.52 Wh kg�1 and 58 Wh kg�1. Theabove discussion clearly supports the importance of macroporousCuO in supercapacitors in order to improve the specific capacitanceto obtain the best performance of a supercapacitor.

The long-term cycle stability of a supercapacitor is importantfor its practical applications and was evaluated in this study byrepeating the cyclic charging–discharging test between �0.5 and

0.5 V (vs. Ag/AgCl) for 1000 cycles at a current density of 0.9 A g�1.Fig. 7B shows the variation of specific capacitance with cyclenumber for the GN/GCE, CuO/GCE and GN/macroCuO/GCE super-capacitors at a constant current density of 0.9 A g�1. Although apotential drop occurred in the initial cycles, the specific capaci-tance for GN/macroCuO/GCE capacitor was found to exhibitexcellent stability over the entire cycle numbers and the specificcapacitance still remains at 380.0 F g�1 (91.4 %) for GN/macroCuOmodified GCE after 1000 cycles of testing. However, in case ofgraphene and macroCuO, it drops to 76.41% and 86.24% respec-tively. This is due to the restacking of graphene sheets. This showsthat the porous CuO based supercapacitor has a long life cycle ascompared to graphene and the performance of porous CuO in asupercapacitor could be applied to other materials and devices.

Electrochemical impedance spectroscopic (EIS) study wasperformed to compare faradaic resistance and capacitive traits ofthe electrode material. The semicircle at higher frequency corre-sponds to the charge transfer resistance (Rct) caused by the Faradaicreaction and double layer capacitance (Cdl) at the interface betweenelectrode and electrolyte solution [46]. The EIS study has been donein the frequency range of 0.1 Hz to 106 Hz at open circuit potentialwith amplitude of 5 mV. The Nyquist plots for GN/GCE, macroCuO/GCE and GN/macroCuO/GCE are shown in Fig. 8. It is seen fromNyquist plot that the diameter of semicircle goes on decreasing fromGN/GCE to GN/macroCuO/GCE suggesting low charge transfer andelectrolyte resistance of the later. The magnitude of equivalent seriesresistance (ESR) values obtained for GN, macroCuO and GN/macroCuO based electrodes are 2.89, 2.19 and 1.92 ohms, respec-tively. The Nyquist plot vertical line for GN/macroCuO/GCE at lowerfrequencies almost parallel to the imaginary axis indicates a goodcapacitive behaviour, representative of the ion diffusion in thestructure of the electrode [47,48]. The reduction in the ESR value forGN/macroCuO based electrode indicates that GN/macroCuO pro-vides shorter diffusion path between sheets and improves chargetransfer performance of GN [49].

ESR data is an important factor in determining the powerdensity of a supercapacitor.The maximum power density (Pmax) ofthe supercapacitor devices are calculated from the low frequencydata of the impedance spectra, according to Eq. (5)

P max ¼ 14Rm

V2i (5)

where R is the equivalent series resistance. The ESR has beenobtained from the x-intercept of the Nyquist plot. [35].

R.A. Dar et al. / Electrochimica Acta 163 (2015) 196–203 203

The maximum power densities of 11.16, 16.55 and 21.85 kWkg�1 are obtained for graphene, macroCuO and graphene-macro-CuO based capacitors respectively. The high power capability ofporous CuO is attributed to the fast kinetic reaction of porous flakelike structures of this material permitting counter-ions whichcould diffuse into its pores and provides a short diffusion distanceto facilitate the ion transport [50,51].

A comparison of the previous works on CuO based super-capacitors in literature is given in Table S1 [20–26,39,42,52–54](Supporting information). It can be seen that the supercapacitorbased on GN/macroCuO nanocomposite exhibits a good perfor-mance in specific capacitance, energy density and power density ascompared to the earlier reported results.

4. Conclusion

A cost-effective and scalable synthesis approach was used toprepare macro-structured porous copper oxide monoliths viamodified Sol–Gel route without using any acidic or basic medium.Dispersion of this porous metal oxide with graphene helped in theimprovement of the capacitance properties of GN. Electrochemicalsupercapacitor device was fabricated using GN and GN/macroCuOcomposite electrodes. The GN/macroCuO composite showedpseudocapacitive behaviour due to faradic type of mechanism.Faradaic current in GN/macroCuO modified GCE is accompanied bythe electric double layer charging process due to the presence ofporous electrode material and graphene sheets which is responsi-ble for the enhancement in capacitance for the energy storage. Thelatter gave remarkable results with a maximum specific capaci-tance of 417 F g�1, energy density of 58 Wh kg�1 and power densityof 17.85 kW kg�1. The fabricated supercapacitor device exhibitedexcellent cycle life with 91.4% of the initial specific capacitanceretained after 1000 cycles. These features make GN/macroCuOcomposite a quite suitable and promising electrode material forefficient supercapacitors.

Acknowledgements

The funding for this work is partly by the University GrantsCommission, New Delhi, India under its Dr. D. S. Kothari Postdoctorate fellowship scheme (RAD) and partly by the US ArmyInternational Technology Center, Tokyo, Japan through contractnumber FA2386-12-1-4086.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.electacta.2015.02.123.

References

[1] C. Liu, F. Li, L.P. Ma, H.M. Cheng, Adv. Mater. 22 (2010) E62.[2] G. Wang, L. Zhang, J. Zhang, Chem. Soc. Rev. 41 (2012) 828.[3] P. Simon, Y. Gogotsi, Nat. Mater. 7 (2008) 854.[4] K. Jurewicz, K. Babel, Energy and Fuels 24 (2010) 3435.

[5] C. Liu, Z. Yu, D. Neff, A. Zhamu, B.Z. Jang, Nano Lett. 10 (2010) 4868.[6] K.H. An, W.S. Kim, Y.S. Park, J.M. Moon, D.J. Bae, S.C. Lim, Y.S. Lee, Y.H. Lee, Adv.

Func. Mater. 11 (2011) 392.[7] C. Yuan, X. Zhang, L. Su, B. Gao, L. Shen, J. Mater. Chem. 9 (2009) 5777.[8] Y. Gao, S. Chen, D. Cao, G. Wang, J. Yin, J. Power Sources 195 (2010) 760.[9] X. Qing, S. Liu, K. Huang, K. Lv, Y. Yang, Z. Lu, D. Fang, X. Liang, Electrochim. Acta

56 (2011) 4991.[10] X. Zhang, Y. Zhao, C. Xu, Nanoscale 6 (2014) 3646.[11] J. Yan, E. Khoo, A. Sumboja, P.S. Lee, ACS Nano 4 (2010) 4255.[12] H. Wang, H.S. Casalongue, Y. Liang, H. Dai, J. Am. Chem. Soc. 132 (2010) 7477.[13] J.K. Chang, C.M. Wu, I.W. Sun, J. Mater. Chem. 20 (2010) 3735.[14] R. Sahay, P.S. Kumar, V. Aravindan, J. Sundaramurthy, W.C. Ling, S.G.

Mhaisalkar, S. Ramakrishna, S. Madhavi, J. Phys. Chem. C 116 (2010) 18087.[15] C.W. Zou, J. Wang, F. Liang, W. Xie, L.X. Shao, D.J. Fu, Curr. Appl. Phys. 12 (2012)

1349.[16] V. Dhanasekaran, T. Mahalingam, R. Chandramohan, J.K. Rhee, J.P. Chu, Thin

Solid Films 520 (2012) 6608.[17] J. Liu, J. Jin, Z. Deng, S.Z. Huang, Z.Y. Hu, L. Wang, C. Wang, L.H. Chen, Y. Li, G. Van

Tendeloo, B.L. Su, J. Colloid. Interf. Sci. 384 (2012) 1.[18] I. Singh, R.K. Bedi, Solid State Sci. 13 (2011) 2011.[19] L. Zhang, R. Liu, H. Yang, Physica E 44 (2010) 1592.[20] V.D. Patake, S.S. Joshi, C.D. Lokhande, O.S. Joo, Mater. Chem. Phys. 114 (2009) 6.[21] D.P. Dubal, D.S. Dhawale, R.R. Salunkhe, V.S. Jamdade, C.D. Lokhande, J. Alloy.

Compd. 492 (2010) 26.[22] H. Zhang, M. Zhang, Mater. Chem. Phys. 108 (2008) 184.[23] H. Zhang, J. Feng, M. Zhang, Mater. Res. Bull. 43 (2008) 3221.[24] G. Wang, J. Huang, S. Chen, Y. Gao, D. Cao, J. Power Sources 196 (2011) 5756.[25] B. Zhao, P. Liu, H. Zhuang, Z. Jiao, T. Fang, W. Xu, B. Lu, Y. Jiang, J. Mater. Chem. A

1 (2013) 367.[26] X. Dong, K. Wang, C. Zhao, X. Qian, S. Chen, Z. Li, H. Liu, S. Dou, J. Alloy. Compd.

586 (2014) 753.[27] S. Chen, J.W. Zhu, X.D. Wu, Q.F. Han, X. Wang, ACS Nano 4 (2010) 2830.[28] M. Willander, K. Hasan, O. Nur, A. Zainelabdin, S. Zaman, G. Amin, J. Mater.

Chem. 22 (2012) 2337.[29] Y. Huang, J. Liang, Y. Chen, Small 8 (2012) 1834.[30] B.E. Conway, Electrochemical Supercapacitors, Scientific Fundamentals and

Technological Applications, Kluwer Academic/Plenum, New York, 1999.[31] J. Yan, T. Wei, Z. Fan, W. Qian, M. Zhang, X. Shen, F. Wei, J. Power Sources 195

(2010) 3045.[32] J. Lin, C.G. Zhang, Z. Yan, Nano Lett. 13 (2013) 72.[33] R.B. Rakhi, H.N.J. Alshareef, J. Power Sources 196 (2011) 8858.[34] F. Khan, S. Mann, J. Phys. Chem. C 113 (2009) 19871.[35] H. Zhang, C. Tang, Y. Lv, C. Sun, F. Gao, L. Dong, Y. Chen, J. Colloids interf. Sci. 380

(2012) 16.[36] G.A. Naikoo, R.A. Dar, F. Khan, J. Mater. Chem. A 2 (2014) 11798.[37] D. Walsh, I. Arceli, T. Ikoma, S. Tanaka, S. Mann, Nat. Mater. 2 (2003) 386.[38] F. Khan, M. Eswaramoorthy, C.N.R. Rao, Solid State Sci. 9 (2007) 27.[39] K. Dae-Won, R. Kyong-Yop, P. Soo-Jin, J. Alloys Compd. 530 (2012) 6.[40] W. Zhe Teo, A. Adriano, P. Martin, Electrochem. Commun. 28 (2013) 51.[41] H.J. Choi, S.M. Jung, J.M. Seo, D.W. Chang, L. Dai, J.B. Baek, Nano Energy 1 (2012)

534.[42] D.P. Dubal, G.S. Gund, R. Holze, H.S. Jadhav, C.D. Lokhande, C. Park, Dalton

Trans. 42 (2013) 6459.[43] J. Kang, S. Wei, K. Zhu, Y. Kim, J. Phys. Chem. C 115 (2011) 4909.[44] K. Chen, S. Song, K. Li, D. Xu, CrystEngComm 15 (2013) 10367.[45] T. Lu, L. Pan, H. Li, G. Zhu, T. Lv, X. Liu, Z. Sun, T. Chen, D.H.C. Chua, J. Alloys

Compd. 509 (2011) 5488.[46] Impedance Spectroscopy – Theory, Experiment, and Applications, in: E.

Barsoukov, J.R. Macdonald (Eds.), Wiley-Interscience, 2005.[47] G. Nagasubramanian, R.G. Jungst, D.H. Doughty, J. Power Sources 83 (1999)

193.[48] B.Y. Chang, S.M. Park, Annu. Rev. Anal. Chem. 3 (2010) 207.[49] Y. Wang, Z. Shi, Y. Huang, Y. Ma, C. Wang, M. Chen, Y. Chen, J. Phys. Chem. C 113

(2009) 13103.[50] T. Brezesinski, J. Wang, S.H. Tolbert, B. Dunn, Nat. Mater. 9 (2010) 146.[51] K. Wang, Y.H. Wang, H. Zhou, J. Phys. Chem. C 113 (2009) 1093.[52] X.Y. Zhang, H.M. Ming Kuang, C.P. Liu, J.L. Tan, M. Dong, Y. Yuan, X.L. Zhao, Z.

Wen, Int. J. Electrochem. Sci. 8 (2013) 1381.[53] Z. Endut, M. Hamdi, W.J. Basirun, Thin Solid Films 528 (2013) 216.[54] Y. Liutao, J. Yuanyuan, L. Lanlan, M. Jie, G. Wang, B. Gen, X. Zhang,

CrystEngComm 15 (2013) 7657.