- Email: [email protected]
Manganese Oxide for Electrochemical Capacitors
Electrochimica Acta 141 (2014) 39–44
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Core-Shell Tubular Nanostructured Electrode of Hollow Carbon Nanofiber/Manganese Oxide for Electrochemical Capacitors Seungki Hong a,1 , Sangkyu Lee b,1 , Ungyu Paik a,∗ a b
Department of Energy Engineering, Hanyang University, Seoul 133-791, Korea Division of Materials Science and Engineering, Hanyang University, Seoul 133-791, Korea
a r t i c l e
i n f o
Article history: Received 14 June 2014 Received in revised form 10 July 2014 Accepted 11 July 2014 Available online 21 July 2014 Keywords: Core-shell hollow carbon nanofiber manganese oxide capacitor
a b s t r a c t Here we propose a core-shell tubular nanostructure consisted of hollow carbon nanofiber and manganese oxide (MnO2 ) for the application of high capacitance electrochemical capacitors. Hollow nanostructured carbon nanofibers are prepared using an electrospinning technique with a dual nozzle. The hollow channel of carbon nanofibers enables the uptake of MnO2 precursor solution inside the hollow carbon nanofiber, leading to the formation of MnO2 layer on both the inner and outer surfaces of hollow carbon nanofiber. The utilization of both surfaces of hollow carbon nanofiber increases the effective reaction sites of electrode materials contacted with an electrolyte as well as maximizes the loading mass of MnO2 on the surface of hollow carbon nanofiber (94% compared to carbon contents), consequently enabling the fabrication of electrochemical capacitors with the increased specific capacitance of 237 F/g. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction Electrochemical capacitors have been attracting attention as a promising energy device that exhibits high power density and long life cycle stability, and also bridges the gap in the energy density of conventional capacitors and rechargeable lithium ion batteries [1–3]. They can be categorized into electrical double layered capacitor (EDLC) and pseudocapacitor according to the charging mechanism [2]. In the case of EDLC, charges accumulate on the surface of electrode and simultaneously ions gather at the interface of electrode-electrolyte during the charging process. This reaction occurs at the interface between electrode and electrolyte, thus the increase in the surface area of electrode is prerequisite for obtaining high performance EDLC [1–3]. For this purpose, carbonaceous materials having large surface area have been utilized [1–3]. Although many attempts have been made to optimize their surface area and pore structure for achieving high capacitance EDLC, the specific capacitance of EDLC electrode materials still remains as low as 100-400 F/g [3]. To overcome such low specific capacitance of EDLC, pseudocapacitor has been proposed. In the case of pseudocapacitor, charging and discharging are based on the fast and reversible Faradaic
∗ Corresponding author. Tel.: +82 2 2220 0502. E-mail address: [email protected] (U. Paik). 1 Both authors contributed equally to this work. http://dx.doi.org/10.1016/j.electacta.2014.07.047 0013-4686/© 2014 Elsevier Ltd. All rights reserved.
reaction of electrode materials with electrolyte ions [2,4]. Metal oxides such as MnO2 , NiO, Co3 O4 , and their hydroxide forms or compounds have been widely investigated as electrode materials for pseudocapacitor application [2,4]. Some of these materials exhibit the specific capacitance as high as 2,000 F/g [5,6]. However, they suffer from their low electronic conductivity. Thus, active materials have been mixed together with conducting additives, providing efficient electron pathways to pseudocapacitor electrodes. In addition, polymeric binders that glue all the electrode constituents onto the current collector have been included. General binders are inactive and electrically insulating materials; therefore their presence in the electrode deteriorates the electrochemical performance of electrochemical capacitors. Recently, binder-free processes that utilize a freestanding conducting mat have been suggested to avoid this issue [7–11]. Such conducting mat can be fabricated with 1D or 2D carbonaceous materials such as carbon nanotube, graphene, carbon nanofiber or their mixture. Among them, carbon nanofiber is considered as a promising conducting material for constructing efficient conducting pathways because electrons can be freely transported through the intimate contact points between carbon nanofibers as well as through the body of carbon nanofiber. Poly(acrylonitrile)(PAN) has been widely utilized as a starting material for producing carbon nanofibers [12,13]. PAN nanofibers can be prepared via a facile electrospinning, which is converted into carbon nanofibers through a successive process of structure stabilization and carbonization. The resulting carbon nanofibers can be decorated with redox-based
40
S. Hong et al. / Electrochimica Acta 141 (2014) 39–44
Fig. 1. Preparation steps for hollow carbon nanofiber. (a) Electrospinning with a dual nozzle system, (b) PMMA-PAN core-shell nanofibers, and (c) hollow carbon nanofibers. All scale bars are 500 nm.
metal oxides [14] or conducting polymers [15], which has been utilized to fabricate the high performance pseudocapacitor. Meanwhile, conventional carbon nanofiber is an imporous material. Since the electrochemical reaction in the electrochemical capacitors only occurs at the interface between electrode materials and electrolyte, the core part of the carbon nanofibers does not participate in the reaction. Finally, it leads to the decrease in the specific capacitance of electrochemical capacitors. In this study, we demonstrate a unique nanostructured electrode to achieve high performance electrochemical capacitors by constructing a hollow open channel along the axis of carbon nanofiber. The hollow structured carbon nanofibers that are proposed here minimize the unreacted part of the carbon nanofiber. In addition, ions can be transported through the inner hollow channel, increasing ion-accessible surface area. Such dual contribution of hollow carbon nanofibers expects the increase in the specific capacitance of electrochemical capacitors fabricated with hollow carbon nanofibers. An electrospinning system equipped with a dual nozzle is utilized to produce core-shell nanofibers consisted of sacrificial poly(methyl methacrylate) (PMMA) core and PAN shell (Fig. 1).
One-step annealing process generates hollow carbon nanofibers. To further increase the specific capacitance of carbon nanofibers, we decorate their surface with MnO2 that is a typical electrochemically active material exhibiting high theoretical specific capacitance of 1,370 F/g [16]. Also, the precursor solution for MnO2 formation can be freely penetrated into the hollow channel of hollow carbon nanofiber, enabling the formation of MnO2 layer on both inner and outer surfaces of hollow carbon nanofibers. This approach is expected to contribute to utilize both inner and outer surfaces as reactive sites for the electrochemical reaction as well as obtain the high mass loading of MnO2 , eventually achieving high capacitance electrochemical capacitors. 2. Experimental 2.1. Preparation of starting materials 1 g of PAN (Mw 150,000, Sigma-Aldrich) and 3 g of PMMA (Mw 120,000, Alfa Aesar) were dissolved in 9 g of N,Ndimethylformamide (DMF, CHROMASOLV Plus, HPLC, 99.9%,
Fig. 2. (a) SEM (upper) and TEM (lower) images of carbon nanofibers and (b) hollow carbon nanofibers.
S. Hong et al. / Electrochimica Acta 141 (2014) 39–44
Sigma-Aldrich), respectively. To prepare carbon nanofibers, the PAN-DMF precursor solution was electrospun on an aluminum (Al) foil current collector at a flow rate of 1.5 mL/h and an electric field of 1 kV/cm (voltage: 20 kV, distance: 20 cm). The electrospun PAN nanofibers mat was detached from the Al foil, stabilized at 280 ◦ C in an ambient condition, and successively annealed at 1,000 ◦ C for 5 hr in an argon atmosphere. Hollow carbon nanofibers were prepared from core-shell nanofibers. First, the core-shell nanofibers were electrospun using a dual nozzle. The PMMA-DMF solution was injected into the core capillary of dual nozzle while the PANDMF solution was injected into the shell capillary of dual nozzle. The condition of electric field for generating PMMA-PAN coreshell nanofibers was the same with that of PAN carbon nanofibers. The flow rates of core and shell solutions were 1.0 and 1.5 mL/h, respectively. The immiscibility between PMMA and PAN enables the formation of core-shell nanofibers. The mat of PMMA-PAN coreshell nanofibers was annealed at 280 ◦ C for 1 hr in an ambient condition for the stabilization of PAN phase, annealed at 450 ◦ C for 1 hr in a mixed atmosphere of hydrogen and argon for the removal of the core PMMA phase. Finally, the hollow carbon nanofibers were obtained by annealing at 1,000 ◦ C for 5 hr in an argon atmosphere. To decorate the surface of carbon nanofibers with MnO2 , the carbon nanofiber mats (area: 0.785 cm2 ) were immersed into an aqueous solution containing 0.1 M of potassium permanganate (KMnO4 , 99.3%, Daejung Chemical Co.) and 0.1 M of sodium sulfate (Na2 SO4 , anhydrous, 99.0%, Samchun Chemical Co.), and then stored in an oven heated to 80 ◦ C. The contents of MnO2 on carbon nanofibers were controlled by dipping time.
(a)
3. Results and discussion Fig. 2 shows the scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images of carbon nanofibers and hollow carbon nanofibers. Carbon nanofibers have a diameter of ∼ 500 nm and dense microstructure. The morphology of hollow carbon nanofibers is exactly the same with those of the carbon nanofibers. TEM image clearly shows the formation of hollow channel in the hollow carbon nanofibers. The diameter of hollow channel is ∼ 500 nm and the wall thickness is ∼ 50 nm. Also, the thickness of carbon nanofiber mats was controlled to be ∼ 18 m (Fig. S1). To examine the influence of inner channel inside hollow carbon nanofibers on their electrochemical performance, cyclic voltammetry (CV) was utilized (Fig. 3). The freestanding films of carbon nanofibers and hollow carbon nanofibers are mechanically robust to be utilized as working electrodes without adding polymeric binder. The measurement of CV for both carbon nanofibers was conducted in the 0.1 M Na2 SO4 aqueous electrolyte with a 3-electrode
Current density (A/g)
1.0 0.5
100 mV/s
0.0
10 mV/s
-0.5 -1.0
(b)
0.0
0.2
0.4
0.6
1.0
Current density (A/g)
0.8
Potential (V vs. Ag/AgCl)
1.0
100 mV/s
0.5 0.0
10 mV/s
-0.5
(c)
Specific capacitance (F/g)
-1.0
2.2. Characterization The morphology of carbon nanofibers was observed using a JSM 4700F field emission scanning electron microscope (FE-SEM, JEOL) and a JEM 2100F field emission transmission electron microscope (FE-TEM, JEOL). The crystal structure of MnO2 coating layer was characterized using an X-ray diffractometer (Bruker Miller) with a Cu-K␣ radiation. The masses of all samples were measured using a Sartorius S2 microbalance (resolution: 0.1 g, Sartorius). All characterizations of electrochemical performance of electrochemical capacitors were carried out with a 3-electrode configuration. An adequate amount of Na2 SO4 was dissolved in de-ionized water to prepare the aqueous Na2 SO4 electrolyte solution with a concentration of 0.1 M. The results of cyclic voltammetry and galvanostatic charge-discharge tests were obtained using an Autolab PGSTAT 302 N potentiostat/galvanostat apparatus (Metrohm AG).
41
0.0
0.2
0.4
0.6
0.8
Potential (V vs. Ag/AgCl)
1.0
30 Carbon nanofiber Hollow carbon nanofiber
20
10
0
0
20
40
60
80
100
Scan rate (mV/s) Fig. 3. Electrochemical performance of electrodes prepared with carbon nanofibers and hollow carbon nanofibers. (a) CV curves of carbon nanofiber electrode and (b) hollow carbon nanofiber electrode obtained at the scan rate of 10, 20, 30, 40, 50, and 100 mV/s. (c) Specific capacitance calculated from the results of CV curves.
configuration. The curves were obtained in the voltage range of 0 to 0.8 V with increasing scan rate from 10 to 100 mV/s. The CV curves for both electrodes show nearly rectangular shape. Although the scan rate is increased, the shape of CV curves is well maintained, indicating the good capacitive behavior of both carbon nanofiberbased electrodes. The specific capacitance was also estimated with the results of CV curves using the following equation [17] and the results are plotted in Fig. 3c. Csp, CV =
Q mE
(1)
where Q is the integrated charges, m is the mass of carbon nanofiber electrode, E is the potential window for the measurement of CV curves (in this wok, this value is 0.8 V). While carbon nanofibers have the specific capacitance of 5.5 F/g at the scan rate of 10 mV/s, much higher capacitance value (19.5 F/g) is obtained in the case of hollow carbon nanofibers at the same measurement condition. This increment in the specific capacitance is attributed to the presence of inner surface to which ions can be accessible as well as the diminished weight of hollow carbon nanofibers. With increasing scan rate, ion accessible sites are decreased, leading to the
42
S. Hong et al. / Electrochimica Acta 141 (2014) 39–44
Fig. 4. MnO2 -decorated carbon nanofibers. (a) Loading level of MnO2 as a function of dipping time. The actual loading masses of MnO2 and carbon nanofibers are presented as (loading mass of MnO2 /weight of carbon nanofibers, unit: mg). (b) SEM images of MnO2 -decorated carbon nanofibers and (c) hollow carbon nanofibers. (d) Cross-sectional image of MnO2 -decorated hollow carbon nanofibers.
decreased specific capacitance. However, hollow carbon nanofibers still deliver the specific capacitance as high as 9.4 F/g at the scan rate of 100 mV/s. The formation of inner channel inside hollow carbon nanofibers greatly enhances the specific capacitance of carbon nanofiber electrode. However, the value of specific capacitance of hollow carbon nanofibers is still low to be applied to practical application. In order to overcome such limitation of hollow carbon nanofiber, we decorate carbon nanofibers with MnO2 that is a promising material for high capacity electrochemical capacitors due to many advantages of MnO2 such as low cost, environmentally benign character, easy formation, and high theoretical capacitance [16,18–21]. Many synthesis methodologies including hydrothermal reaction [22], electrochemical deposition [19], and self-limited redox reaction [23] have been reported to coat the surface of templates with MnO2 . In the case of carbon-based template, redox-based reaction between carbon species and permanganate ions can be considered as a facile and effective strategy for generating carbon-MnO2 core-shell structure [23,24]. However, this approach is appropriate to obtain thin layer of MnO2 because the reaction is limited only around the interface between carbon and precursor solution containing permanganate ions [25]. Instead of the self-limited redox reaction, a facile dipping process has been proposed to achieve high mass loading of MnO2 on the surface of carbon [25]. Utilizing this methodology, the freestanding films of carbon nanofibers and hollow carbon nanofibers were decorated with MnO2 by dipping into a precursor solution containing KMnO4 and Na2 SO4 . And then, the solution was heated to 80 ◦ C. By controlling the dipping time, the loading level of MnO2 on both carbon nanofibers can be easily changed (Fig. 4a). Longer dipping time, higher amount of MnO2 can be decorated over the surface of carbon nanofibers. The contents almost linearly increase in this experimental condition. At the dipping time of 60 min, hollow carbon nanofibers contain 2.4 times
higher loading level of MnO2 than carbon nanofibers. Fig. 4b and c show the surface morphology of carbon nanofibers that are decorated with MnO2 by dipping in the precursor solution for 60 min. Very thin nanowall-like MnO2 are successfully synthesized on the surface of both carbon nanofibers (Fig. S2). The cross-sectional SEM image of MnO2 -decorated hollow carbon nanofiber (Fig. 4d) reveals the formation of MnO2 on both the inner and outer surfaces of hollow carbon nanofibers, explaining the higher loading level of MnO2 in MnO2 -decorated hollow carbon nanofibers. Fig. S3 shows the microstructural evolution for MnO2 -decorated hollow carbon nanofibers with increasing dipping time. The formation of three distinct layers consisted of inner MnO2 , carbon nanofiber, and outer MnO2 layers can be clearly confirmed at the dipping time of 120 min (Fig. S3 g). The crystal structure of MnO2 -carbon nanofibers was also analyzed using an X-ray diffractometer. The X-ray diffraction (XRD) pattern of MnO2 -decorated carbon nanofibers (Fig. S4) contains three peaks of 25.6, 37.3, and 66.2o , which coincides well with that of birnessite-type MnO2 [11]. These peaks can be assigned to be ¯ and (312)/(020) ¯ (002), (111), planes of birnessite-type MnO2 (JCPDS 42-1317). In the case of carbon nanofibers, on the other hand, only a halo is observed at ∼ 23.9o due to their amorphous nature. To compare the electrochemical performance of MnO2 decorated carbon nanofibers with that of MnO2 -decorated hollow carbon nanofibers, CV and galvanostatic charge-discharge tests were employed. The content and thickness of MnO2 critically affect the electrochemical performance of MnO2 -based electrochemical capacitor [16,18,19,26]. We prepared both carbon nanofibers with different loading level of MnO2 . The electrochemical performance of both carbon nanofibers with different loading level of MnO2 was evaluated by using CV measurement in the voltage range of 0 to 0.8 V with increasing scan rate from 10 to 100 mV/s (Fig. S5). Fig. 5a and b show the CV curves of carbon nanofibers and hollow carbon nanofibers that are decorated with MnO2 by dipping
S. Hong et al. / Electrochimica Acta 141 (2014) 39–44
Current density (A/g)
100 mV/s
3
(b) Current density (A/g)
2
(a)
1 0
10 mV/s
0.0
0.2
0.4
0.6
0.8
Potential (V vs. Ag/AgCl)
0
10 mV/s
300
(d)
MnO2 on carbon nanofiber 10 min 30 min 60 min
200
100
0
20
40
60
-3
1.0
80
100
Specific capacitance (F/g)
Specific capacitance (F/g)
1
-2
-2
0
100 mV/s
2
-1
-1
(c)
43
0.0
0.2
0.4
0.6
0.8
Potential (V vs. Ag/AgCl)
1.0
300 MnO2 on hollow carbon nanofiber 10 min 30 min 60 min
200
100
0
0
20
Scan rate (mV/s)
40
60
80
100
Scan rate (mV/s)
for 10 min, respectively. Both electrodes exhibit the rectangular shaped CV curves irrespective of scan rate. We calculated the specific capacitance of the electrodes with increasing MnO2 content (Fig. 5c and d). At the same dipping time, hollow carbon nanofibers decorated with MnO2 show higher specific capacitance than MnO2 decorated carbon nanofibers. With increasing the dipping time, the higher amount of MnO2 can be coated on the surface of carbon nanofibers, reaching the specific capacitance as high as 236.9 F/g in the case of MnO2 -decorated hollow carbon nanofibers. As for hollow carbon nanofibers, electrochemically active materials, in this experiment MnO2 , can be coated on the inner and outer surfaces simultaneously, enabling the dual electrochemical contribution of the active materials that are coated on the inner and outer surfaces of hollow carbon nanofibers as well as increasing the mass loading of electrochemically active materials. In general, the higher mass loading of electrochemically active materials accompanies the increase in the thickness of coating layer consisted of the materials, retarding the kinetics related to electrons and ions in the layer of active materials and consequently leading to the deterioration of their electrochemical performance. However, the utilization of a template that has large surface area enables to increase the mass loading of electrochemically active materials without increasing the thickness of coating layer [11]. Therefore, the utilization of hollow carbon nanofibers as a template for forming electrochemically active materials achieves higher capacitance in the electrochemical capacitors. Furthermore, galvanostatic charge-discharge tests were utilized to characterize the electrochemical performance of MnO2 decorated carbon nanofibers. The tests were performed in the voltage range of 0 to 0.8 V with the current density of 0.5, 1, 2, and 5 A/g. The results are shown in Fig. S6. The cycle performance of both electrodes was also measured using the results of galvanostatic charge-discharge tests (Fig. 6). The tests were repeated 1,000 cycles at the current density of 5 A/g. The specific capacitance of electrodes was calculated using the following equation [17]. Csp, galvanostatic =
it mE
(2)
Specific capacitance (F/g)
Fig. 5. Electrochemical performance of MnO2 -decorated carbon nanofibers. (a) CV curves of MnO2 -decorated carbon nanofibers and (b) hollow carbon nanofibers. (c) Specific capacitance of MnO2 -decorated carbon nanofibers and (d) hollow carbon nanofibers.
120 100 80 60 40 MnO2 on carbon nanofiber
20 0
MnO2 on hollow carbon nanofiber
0
200
400
600
800
1000
Cycle Number Fig. 6. Cycle performance of MnO2 -decorated carbon nanofibers measured at 5 A/g.
where i is the discharge current, t is the discharge time, m is the mass of MnO2 -decorated carbon nanofiber, and E is the potential window. At the current density of 5 A/g, the electrode of MnO2 decorated hollow carbon nanofibers generates 97.3 F/g of specific capacitance, which is higher than that of MnO2 -decorated carbon nanofibers (70.3 F/g). The specific capacitance of both electrodes is hardly changed after 1,000 cycle, indicating the excellent retention of both MnO2 -decorated carbon nanofibers. 4. Conclusions The core-shell tubular nanostructure of hollow carbon nanofiber-manganese oxide is prepared by electrospinning technique and dipping process. The hollow channel of hollow carbon nanofibers decreases the electrochemically unreacted part of carbon nanofibers. In addition, the presence of the hollow channel enables the infiltration of precursor solution for MnO2 formation inside the hollow carbon nanofibers, forming a layer of MnO2 on both inner and outer surfaces of hollow carbon nanofibers. The dual contribution of hollow carbon nanofibers increases the specific
44
S. Hong et al. / Electrochimica Acta 141 (2014) 39–44
capacitance of electrochemical capacitors prepared with MnO2 decorated hollow carbon nanofibers. This demonstration proposes an effective nanostructure for increasing the loading level of metal oxide materials with high specific capacitance and consequently achieving high capacitance in the electrochemical capacitors. Acknowledgements This work was supported by the Global Research Laboratory (GRL) Program (K20704000003TA050000310) through the National Research Foundation of Korea (KRF) funded by the Ministry of Science, ICT (Information and Communication Technologies) and Future Planning, the International Cooperation program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government of Ministry of Trade, Industry & Energy (2011T100100369). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.electacta.2014.07.047. References [1] R. Kotz, M. Carlen, Electrochim. Acta 45 (2000) 2483. [2] P. Simon, Y. Gogotsi, Nat. Mater. 7 (2008) 845. [3] L.L. Zhang, X.S. Zhao, Chem. Soc. Rev. 38 (2009) 2520.
[4] V. Augustyn, P. Simon, B. Dunn, Energ. Environ. Sci. 7 (2014) 1597. [5] J. Du, G. Zhou, H.M. Zhang, C. Cheng, J.M. Ma, W.F. Wei, L.B. Chen, T.H. Wang, ACS Appl. Mater. Interfaces 5 (2013) 7405. [6] C.Q. Shang, S.M. Dong, S. Wang, D.D. Xiao, P.X. Han, X.G. Wang, L. Gu, G.L. Cui, ACS Nano 7 (2013) 5430. [7] N. Wang, C.X. Wu, J.X. Li, G.F. Dong, L.H. Guan, ACS Appl. Mater. Interfaces 3 (2011) 4185. [8] M. Zhi, A. Manivannan, F. Meng, N. Wu, J. Power Sources 208 (2012) 345. [9] S.D. Perera, A.D. Liyanage, N. Nijem, J.P. Ferraris, Y.J. Chabal, K.J. Balkus, J. Power Sources 230 (2013) 130. [10] S.Y. Wang, R.A.W. Dryfe, Journal of Materials Chemistry A 1 (2013) 5279. [11] J.H. Kim, S. Lee, J.W. Lee, T. Song, U. Paik, Electrochim. Acta 125 (2014) 536. [12] E. Zussman, X. Chen, W. Ding, L. Calabri, D.A. Dikin, J.P. Quintana, R.S. Ruoff, Carbon 43 (2005) 2175. [13] S.K. Nataraj, K.S. Yang, T.M. Aminabhavi, Progress in Polymer Science 37 (2012) 487. [14] G. Zhang, X.W. Lou, Sci. Rep 3 (2013). [15] M. Kotal, A.K. Thakur, A.K. Bhowmick, ACS Appl. Mater. Interfaces 5 (2013) 8374. [16] M. Toupin, T. Brousse, D. Bélanger, Chem. Mater. 16 (2004) 3184. [17] H. Wang, L. Zhang, X. Tan, C.M.B. Holt, B. Zahiri, B.C. Olsen, D. Mitlin, J. Phys. Chem. C 115 (2011) 17599. [18] L. Hu, W. Chen, X. Xie, N. Liu, Y. Yang, H. Wu, Y. Yao, M. Pasta, H.N. Alshareef, Y. Cui, ACS Nano 5 (2011) 8904. [19] G. Yu, L. Hu, M. Vosgueritchian, H. Wang, X. Xie, J.R. McDonough, X. Cui, Y. Cui, Z. Bao, Nano Lett. 11 (2011) 2905. [20] Z. Lei, F. Shi, L. Lu, ACS Appl. Mater. Interfaces 4 (2012) 1058. [21] S. Deng, D. Sun, C. Wu, H. Wang, J. Liu, Y. Sun, H. Yan, Electrochim. Acta 111 (2013) 707. [22] S. Yang, X. Song, P. Zhang, L. Gao, ACS Appl. Mater. Interfaces 5 (2013) 3317. [23] X. Jin, W. Zhou, S. Zhang, G.Z. Chen, Small 3 (2007) 1513. [24] S. Lee, J. Ha, H. Cheng, J.W. Lee, T.S. Jang, Y.-G. Jung, Y. Huang, J.A. Rogers, U. Paik, Adv. Energy Mater. 4 (2014) 1300472. [25] Y. Peng, Z. Chen, J. Wen, Q. Xiao, D. Weng, S. He, H. Geng, Y. Lu, Nano Res. 4 (2011) 216. [26] W. Li, G. Li, J. Sun, R. Zou, K. Xu, Y. Sun, Z. Chen, J. Yang, J. Hu, Nanoscale 5 (2013) 2901.
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Core-Shell Tubular Nanostructured Electrode of Hollow Carbon Nanofiber/Manganese Oxide for Electrochemical Capacitors Seungki Hong a,1 , Sangkyu Lee b,1 , Ungyu Paik a,∗ a b
Department of Energy Engineering, Hanyang University, Seoul 133-791, Korea Division of Materials Science and Engineering, Hanyang University, Seoul 133-791, Korea
a r t i c l e
i n f o
Article history: Received 14 June 2014 Received in revised form 10 July 2014 Accepted 11 July 2014 Available online 21 July 2014 Keywords: Core-shell hollow carbon nanofiber manganese oxide capacitor
a b s t r a c t Here we propose a core-shell tubular nanostructure consisted of hollow carbon nanofiber and manganese oxide (MnO2 ) for the application of high capacitance electrochemical capacitors. Hollow nanostructured carbon nanofibers are prepared using an electrospinning technique with a dual nozzle. The hollow channel of carbon nanofibers enables the uptake of MnO2 precursor solution inside the hollow carbon nanofiber, leading to the formation of MnO2 layer on both the inner and outer surfaces of hollow carbon nanofiber. The utilization of both surfaces of hollow carbon nanofiber increases the effective reaction sites of electrode materials contacted with an electrolyte as well as maximizes the loading mass of MnO2 on the surface of hollow carbon nanofiber (94% compared to carbon contents), consequently enabling the fabrication of electrochemical capacitors with the increased specific capacitance of 237 F/g. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction Electrochemical capacitors have been attracting attention as a promising energy device that exhibits high power density and long life cycle stability, and also bridges the gap in the energy density of conventional capacitors and rechargeable lithium ion batteries [1–3]. They can be categorized into electrical double layered capacitor (EDLC) and pseudocapacitor according to the charging mechanism [2]. In the case of EDLC, charges accumulate on the surface of electrode and simultaneously ions gather at the interface of electrode-electrolyte during the charging process. This reaction occurs at the interface between electrode and electrolyte, thus the increase in the surface area of electrode is prerequisite for obtaining high performance EDLC [1–3]. For this purpose, carbonaceous materials having large surface area have been utilized [1–3]. Although many attempts have been made to optimize their surface area and pore structure for achieving high capacitance EDLC, the specific capacitance of EDLC electrode materials still remains as low as 100-400 F/g [3]. To overcome such low specific capacitance of EDLC, pseudocapacitor has been proposed. In the case of pseudocapacitor, charging and discharging are based on the fast and reversible Faradaic
∗ Corresponding author. Tel.: +82 2 2220 0502. E-mail address: [email protected] (U. Paik). 1 Both authors contributed equally to this work. http://dx.doi.org/10.1016/j.electacta.2014.07.047 0013-4686/© 2014 Elsevier Ltd. All rights reserved.
reaction of electrode materials with electrolyte ions [2,4]. Metal oxides such as MnO2 , NiO, Co3 O4 , and their hydroxide forms or compounds have been widely investigated as electrode materials for pseudocapacitor application [2,4]. Some of these materials exhibit the specific capacitance as high as 2,000 F/g [5,6]. However, they suffer from their low electronic conductivity. Thus, active materials have been mixed together with conducting additives, providing efficient electron pathways to pseudocapacitor electrodes. In addition, polymeric binders that glue all the electrode constituents onto the current collector have been included. General binders are inactive and electrically insulating materials; therefore their presence in the electrode deteriorates the electrochemical performance of electrochemical capacitors. Recently, binder-free processes that utilize a freestanding conducting mat have been suggested to avoid this issue [7–11]. Such conducting mat can be fabricated with 1D or 2D carbonaceous materials such as carbon nanotube, graphene, carbon nanofiber or their mixture. Among them, carbon nanofiber is considered as a promising conducting material for constructing efficient conducting pathways because electrons can be freely transported through the intimate contact points between carbon nanofibers as well as through the body of carbon nanofiber. Poly(acrylonitrile)(PAN) has been widely utilized as a starting material for producing carbon nanofibers [12,13]. PAN nanofibers can be prepared via a facile electrospinning, which is converted into carbon nanofibers through a successive process of structure stabilization and carbonization. The resulting carbon nanofibers can be decorated with redox-based
40
S. Hong et al. / Electrochimica Acta 141 (2014) 39–44
Fig. 1. Preparation steps for hollow carbon nanofiber. (a) Electrospinning with a dual nozzle system, (b) PMMA-PAN core-shell nanofibers, and (c) hollow carbon nanofibers. All scale bars are 500 nm.
metal oxides [14] or conducting polymers [15], which has been utilized to fabricate the high performance pseudocapacitor. Meanwhile, conventional carbon nanofiber is an imporous material. Since the electrochemical reaction in the electrochemical capacitors only occurs at the interface between electrode materials and electrolyte, the core part of the carbon nanofibers does not participate in the reaction. Finally, it leads to the decrease in the specific capacitance of electrochemical capacitors. In this study, we demonstrate a unique nanostructured electrode to achieve high performance electrochemical capacitors by constructing a hollow open channel along the axis of carbon nanofiber. The hollow structured carbon nanofibers that are proposed here minimize the unreacted part of the carbon nanofiber. In addition, ions can be transported through the inner hollow channel, increasing ion-accessible surface area. Such dual contribution of hollow carbon nanofibers expects the increase in the specific capacitance of electrochemical capacitors fabricated with hollow carbon nanofibers. An electrospinning system equipped with a dual nozzle is utilized to produce core-shell nanofibers consisted of sacrificial poly(methyl methacrylate) (PMMA) core and PAN shell (Fig. 1).
One-step annealing process generates hollow carbon nanofibers. To further increase the specific capacitance of carbon nanofibers, we decorate their surface with MnO2 that is a typical electrochemically active material exhibiting high theoretical specific capacitance of 1,370 F/g [16]. Also, the precursor solution for MnO2 formation can be freely penetrated into the hollow channel of hollow carbon nanofiber, enabling the formation of MnO2 layer on both inner and outer surfaces of hollow carbon nanofibers. This approach is expected to contribute to utilize both inner and outer surfaces as reactive sites for the electrochemical reaction as well as obtain the high mass loading of MnO2 , eventually achieving high capacitance electrochemical capacitors. 2. Experimental 2.1. Preparation of starting materials 1 g of PAN (Mw 150,000, Sigma-Aldrich) and 3 g of PMMA (Mw 120,000, Alfa Aesar) were dissolved in 9 g of N,Ndimethylformamide (DMF, CHROMASOLV Plus, HPLC, 99.9%,
Fig. 2. (a) SEM (upper) and TEM (lower) images of carbon nanofibers and (b) hollow carbon nanofibers.
S. Hong et al. / Electrochimica Acta 141 (2014) 39–44
Sigma-Aldrich), respectively. To prepare carbon nanofibers, the PAN-DMF precursor solution was electrospun on an aluminum (Al) foil current collector at a flow rate of 1.5 mL/h and an electric field of 1 kV/cm (voltage: 20 kV, distance: 20 cm). The electrospun PAN nanofibers mat was detached from the Al foil, stabilized at 280 ◦ C in an ambient condition, and successively annealed at 1,000 ◦ C for 5 hr in an argon atmosphere. Hollow carbon nanofibers were prepared from core-shell nanofibers. First, the core-shell nanofibers were electrospun using a dual nozzle. The PMMA-DMF solution was injected into the core capillary of dual nozzle while the PANDMF solution was injected into the shell capillary of dual nozzle. The condition of electric field for generating PMMA-PAN coreshell nanofibers was the same with that of PAN carbon nanofibers. The flow rates of core and shell solutions were 1.0 and 1.5 mL/h, respectively. The immiscibility between PMMA and PAN enables the formation of core-shell nanofibers. The mat of PMMA-PAN coreshell nanofibers was annealed at 280 ◦ C for 1 hr in an ambient condition for the stabilization of PAN phase, annealed at 450 ◦ C for 1 hr in a mixed atmosphere of hydrogen and argon for the removal of the core PMMA phase. Finally, the hollow carbon nanofibers were obtained by annealing at 1,000 ◦ C for 5 hr in an argon atmosphere. To decorate the surface of carbon nanofibers with MnO2 , the carbon nanofiber mats (area: 0.785 cm2 ) were immersed into an aqueous solution containing 0.1 M of potassium permanganate (KMnO4 , 99.3%, Daejung Chemical Co.) and 0.1 M of sodium sulfate (Na2 SO4 , anhydrous, 99.0%, Samchun Chemical Co.), and then stored in an oven heated to 80 ◦ C. The contents of MnO2 on carbon nanofibers were controlled by dipping time.
(a)
3. Results and discussion Fig. 2 shows the scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images of carbon nanofibers and hollow carbon nanofibers. Carbon nanofibers have a diameter of ∼ 500 nm and dense microstructure. The morphology of hollow carbon nanofibers is exactly the same with those of the carbon nanofibers. TEM image clearly shows the formation of hollow channel in the hollow carbon nanofibers. The diameter of hollow channel is ∼ 500 nm and the wall thickness is ∼ 50 nm. Also, the thickness of carbon nanofiber mats was controlled to be ∼ 18 m (Fig. S1). To examine the influence of inner channel inside hollow carbon nanofibers on their electrochemical performance, cyclic voltammetry (CV) was utilized (Fig. 3). The freestanding films of carbon nanofibers and hollow carbon nanofibers are mechanically robust to be utilized as working electrodes without adding polymeric binder. The measurement of CV for both carbon nanofibers was conducted in the 0.1 M Na2 SO4 aqueous electrolyte with a 3-electrode
Current density (A/g)
1.0 0.5
100 mV/s
0.0
10 mV/s
-0.5 -1.0
(b)
0.0
0.2
0.4
0.6
1.0
Current density (A/g)
0.8
Potential (V vs. Ag/AgCl)
1.0
100 mV/s
0.5 0.0
10 mV/s
-0.5
(c)
Specific capacitance (F/g)
-1.0
2.2. Characterization The morphology of carbon nanofibers was observed using a JSM 4700F field emission scanning electron microscope (FE-SEM, JEOL) and a JEM 2100F field emission transmission electron microscope (FE-TEM, JEOL). The crystal structure of MnO2 coating layer was characterized using an X-ray diffractometer (Bruker Miller) with a Cu-K␣ radiation. The masses of all samples were measured using a Sartorius S2 microbalance (resolution: 0.1 g, Sartorius). All characterizations of electrochemical performance of electrochemical capacitors were carried out with a 3-electrode configuration. An adequate amount of Na2 SO4 was dissolved in de-ionized water to prepare the aqueous Na2 SO4 electrolyte solution with a concentration of 0.1 M. The results of cyclic voltammetry and galvanostatic charge-discharge tests were obtained using an Autolab PGSTAT 302 N potentiostat/galvanostat apparatus (Metrohm AG).
41
0.0
0.2
0.4
0.6
0.8
Potential (V vs. Ag/AgCl)
1.0
30 Carbon nanofiber Hollow carbon nanofiber
20
10
0
0
20
40
60
80
100
Scan rate (mV/s) Fig. 3. Electrochemical performance of electrodes prepared with carbon nanofibers and hollow carbon nanofibers. (a) CV curves of carbon nanofiber electrode and (b) hollow carbon nanofiber electrode obtained at the scan rate of 10, 20, 30, 40, 50, and 100 mV/s. (c) Specific capacitance calculated from the results of CV curves.
configuration. The curves were obtained in the voltage range of 0 to 0.8 V with increasing scan rate from 10 to 100 mV/s. The CV curves for both electrodes show nearly rectangular shape. Although the scan rate is increased, the shape of CV curves is well maintained, indicating the good capacitive behavior of both carbon nanofiberbased electrodes. The specific capacitance was also estimated with the results of CV curves using the following equation [17] and the results are plotted in Fig. 3c. Csp, CV =
Q mE
(1)
where Q is the integrated charges, m is the mass of carbon nanofiber electrode, E is the potential window for the measurement of CV curves (in this wok, this value is 0.8 V). While carbon nanofibers have the specific capacitance of 5.5 F/g at the scan rate of 10 mV/s, much higher capacitance value (19.5 F/g) is obtained in the case of hollow carbon nanofibers at the same measurement condition. This increment in the specific capacitance is attributed to the presence of inner surface to which ions can be accessible as well as the diminished weight of hollow carbon nanofibers. With increasing scan rate, ion accessible sites are decreased, leading to the
42
S. Hong et al. / Electrochimica Acta 141 (2014) 39–44
Fig. 4. MnO2 -decorated carbon nanofibers. (a) Loading level of MnO2 as a function of dipping time. The actual loading masses of MnO2 and carbon nanofibers are presented as (loading mass of MnO2 /weight of carbon nanofibers, unit: mg). (b) SEM images of MnO2 -decorated carbon nanofibers and (c) hollow carbon nanofibers. (d) Cross-sectional image of MnO2 -decorated hollow carbon nanofibers.
decreased specific capacitance. However, hollow carbon nanofibers still deliver the specific capacitance as high as 9.4 F/g at the scan rate of 100 mV/s. The formation of inner channel inside hollow carbon nanofibers greatly enhances the specific capacitance of carbon nanofiber electrode. However, the value of specific capacitance of hollow carbon nanofibers is still low to be applied to practical application. In order to overcome such limitation of hollow carbon nanofiber, we decorate carbon nanofibers with MnO2 that is a promising material for high capacity electrochemical capacitors due to many advantages of MnO2 such as low cost, environmentally benign character, easy formation, and high theoretical capacitance [16,18–21]. Many synthesis methodologies including hydrothermal reaction [22], electrochemical deposition [19], and self-limited redox reaction [23] have been reported to coat the surface of templates with MnO2 . In the case of carbon-based template, redox-based reaction between carbon species and permanganate ions can be considered as a facile and effective strategy for generating carbon-MnO2 core-shell structure [23,24]. However, this approach is appropriate to obtain thin layer of MnO2 because the reaction is limited only around the interface between carbon and precursor solution containing permanganate ions [25]. Instead of the self-limited redox reaction, a facile dipping process has been proposed to achieve high mass loading of MnO2 on the surface of carbon [25]. Utilizing this methodology, the freestanding films of carbon nanofibers and hollow carbon nanofibers were decorated with MnO2 by dipping into a precursor solution containing KMnO4 and Na2 SO4 . And then, the solution was heated to 80 ◦ C. By controlling the dipping time, the loading level of MnO2 on both carbon nanofibers can be easily changed (Fig. 4a). Longer dipping time, higher amount of MnO2 can be decorated over the surface of carbon nanofibers. The contents almost linearly increase in this experimental condition. At the dipping time of 60 min, hollow carbon nanofibers contain 2.4 times
higher loading level of MnO2 than carbon nanofibers. Fig. 4b and c show the surface morphology of carbon nanofibers that are decorated with MnO2 by dipping in the precursor solution for 60 min. Very thin nanowall-like MnO2 are successfully synthesized on the surface of both carbon nanofibers (Fig. S2). The cross-sectional SEM image of MnO2 -decorated hollow carbon nanofiber (Fig. 4d) reveals the formation of MnO2 on both the inner and outer surfaces of hollow carbon nanofibers, explaining the higher loading level of MnO2 in MnO2 -decorated hollow carbon nanofibers. Fig. S3 shows the microstructural evolution for MnO2 -decorated hollow carbon nanofibers with increasing dipping time. The formation of three distinct layers consisted of inner MnO2 , carbon nanofiber, and outer MnO2 layers can be clearly confirmed at the dipping time of 120 min (Fig. S3 g). The crystal structure of MnO2 -carbon nanofibers was also analyzed using an X-ray diffractometer. The X-ray diffraction (XRD) pattern of MnO2 -decorated carbon nanofibers (Fig. S4) contains three peaks of 25.6, 37.3, and 66.2o , which coincides well with that of birnessite-type MnO2 [11]. These peaks can be assigned to be ¯ and (312)/(020) ¯ (002), (111), planes of birnessite-type MnO2 (JCPDS 42-1317). In the case of carbon nanofibers, on the other hand, only a halo is observed at ∼ 23.9o due to their amorphous nature. To compare the electrochemical performance of MnO2 decorated carbon nanofibers with that of MnO2 -decorated hollow carbon nanofibers, CV and galvanostatic charge-discharge tests were employed. The content and thickness of MnO2 critically affect the electrochemical performance of MnO2 -based electrochemical capacitor [16,18,19,26]. We prepared both carbon nanofibers with different loading level of MnO2 . The electrochemical performance of both carbon nanofibers with different loading level of MnO2 was evaluated by using CV measurement in the voltage range of 0 to 0.8 V with increasing scan rate from 10 to 100 mV/s (Fig. S5). Fig. 5a and b show the CV curves of carbon nanofibers and hollow carbon nanofibers that are decorated with MnO2 by dipping
S. Hong et al. / Electrochimica Acta 141 (2014) 39–44
Current density (A/g)
100 mV/s
3
(b) Current density (A/g)
2
(a)
1 0
10 mV/s
0.0
0.2
0.4
0.6
0.8
Potential (V vs. Ag/AgCl)
0
10 mV/s
300
(d)
MnO2 on carbon nanofiber 10 min 30 min 60 min
200
100
0
20
40
60
-3
1.0
80
100
Specific capacitance (F/g)
Specific capacitance (F/g)
1
-2
-2
0
100 mV/s
2
-1
-1
(c)
43
0.0
0.2
0.4
0.6
0.8
Potential (V vs. Ag/AgCl)
1.0
300 MnO2 on hollow carbon nanofiber 10 min 30 min 60 min
200
100
0
0
20
Scan rate (mV/s)
40
60
80
100
Scan rate (mV/s)
for 10 min, respectively. Both electrodes exhibit the rectangular shaped CV curves irrespective of scan rate. We calculated the specific capacitance of the electrodes with increasing MnO2 content (Fig. 5c and d). At the same dipping time, hollow carbon nanofibers decorated with MnO2 show higher specific capacitance than MnO2 decorated carbon nanofibers. With increasing the dipping time, the higher amount of MnO2 can be coated on the surface of carbon nanofibers, reaching the specific capacitance as high as 236.9 F/g in the case of MnO2 -decorated hollow carbon nanofibers. As for hollow carbon nanofibers, electrochemically active materials, in this experiment MnO2 , can be coated on the inner and outer surfaces simultaneously, enabling the dual electrochemical contribution of the active materials that are coated on the inner and outer surfaces of hollow carbon nanofibers as well as increasing the mass loading of electrochemically active materials. In general, the higher mass loading of electrochemically active materials accompanies the increase in the thickness of coating layer consisted of the materials, retarding the kinetics related to electrons and ions in the layer of active materials and consequently leading to the deterioration of their electrochemical performance. However, the utilization of a template that has large surface area enables to increase the mass loading of electrochemically active materials without increasing the thickness of coating layer [11]. Therefore, the utilization of hollow carbon nanofibers as a template for forming electrochemically active materials achieves higher capacitance in the electrochemical capacitors. Furthermore, galvanostatic charge-discharge tests were utilized to characterize the electrochemical performance of MnO2 decorated carbon nanofibers. The tests were performed in the voltage range of 0 to 0.8 V with the current density of 0.5, 1, 2, and 5 A/g. The results are shown in Fig. S6. The cycle performance of both electrodes was also measured using the results of galvanostatic charge-discharge tests (Fig. 6). The tests were repeated 1,000 cycles at the current density of 5 A/g. The specific capacitance of electrodes was calculated using the following equation [17]. Csp, galvanostatic =
it mE
(2)
Specific capacitance (F/g)
Fig. 5. Electrochemical performance of MnO2 -decorated carbon nanofibers. (a) CV curves of MnO2 -decorated carbon nanofibers and (b) hollow carbon nanofibers. (c) Specific capacitance of MnO2 -decorated carbon nanofibers and (d) hollow carbon nanofibers.
120 100 80 60 40 MnO2 on carbon nanofiber
20 0
MnO2 on hollow carbon nanofiber
0
200
400
600
800
1000
Cycle Number Fig. 6. Cycle performance of MnO2 -decorated carbon nanofibers measured at 5 A/g.
where i is the discharge current, t is the discharge time, m is the mass of MnO2 -decorated carbon nanofiber, and E is the potential window. At the current density of 5 A/g, the electrode of MnO2 decorated hollow carbon nanofibers generates 97.3 F/g of specific capacitance, which is higher than that of MnO2 -decorated carbon nanofibers (70.3 F/g). The specific capacitance of both electrodes is hardly changed after 1,000 cycle, indicating the excellent retention of both MnO2 -decorated carbon nanofibers. 4. Conclusions The core-shell tubular nanostructure of hollow carbon nanofiber-manganese oxide is prepared by electrospinning technique and dipping process. The hollow channel of hollow carbon nanofibers decreases the electrochemically unreacted part of carbon nanofibers. In addition, the presence of the hollow channel enables the infiltration of precursor solution for MnO2 formation inside the hollow carbon nanofibers, forming a layer of MnO2 on both inner and outer surfaces of hollow carbon nanofibers. The dual contribution of hollow carbon nanofibers increases the specific
44
S. Hong et al. / Electrochimica Acta 141 (2014) 39–44
capacitance of electrochemical capacitors prepared with MnO2 decorated hollow carbon nanofibers. This demonstration proposes an effective nanostructure for increasing the loading level of metal oxide materials with high specific capacitance and consequently achieving high capacitance in the electrochemical capacitors. Acknowledgements This work was supported by the Global Research Laboratory (GRL) Program (K20704000003TA050000310) through the National Research Foundation of Korea (KRF) funded by the Ministry of Science, ICT (Information and Communication Technologies) and Future Planning, the International Cooperation program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government of Ministry of Trade, Industry & Energy (2011T100100369). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.electacta.2014.07.047. References [1] R. Kotz, M. Carlen, Electrochim. Acta 45 (2000) 2483. [2] P. Simon, Y. Gogotsi, Nat. Mater. 7 (2008) 845. [3] L.L. Zhang, X.S. Zhao, Chem. Soc. Rev. 38 (2009) 2520.
[4] V. Augustyn, P. Simon, B. Dunn, Energ. Environ. Sci. 7 (2014) 1597. [5] J. Du, G. Zhou, H.M. Zhang, C. Cheng, J.M. Ma, W.F. Wei, L.B. Chen, T.H. Wang, ACS Appl. Mater. Interfaces 5 (2013) 7405. [6] C.Q. Shang, S.M. Dong, S. Wang, D.D. Xiao, P.X. Han, X.G. Wang, L. Gu, G.L. Cui, ACS Nano 7 (2013) 5430. [7] N. Wang, C.X. Wu, J.X. Li, G.F. Dong, L.H. Guan, ACS Appl. Mater. Interfaces 3 (2011) 4185. [8] M. Zhi, A. Manivannan, F. Meng, N. Wu, J. Power Sources 208 (2012) 345. [9] S.D. Perera, A.D. Liyanage, N. Nijem, J.P. Ferraris, Y.J. Chabal, K.J. Balkus, J. Power Sources 230 (2013) 130. [10] S.Y. Wang, R.A.W. Dryfe, Journal of Materials Chemistry A 1 (2013) 5279. [11] J.H. Kim, S. Lee, J.W. Lee, T. Song, U. Paik, Electrochim. Acta 125 (2014) 536. [12] E. Zussman, X. Chen, W. Ding, L. Calabri, D.A. Dikin, J.P. Quintana, R.S. Ruoff, Carbon 43 (2005) 2175. [13] S.K. Nataraj, K.S. Yang, T.M. Aminabhavi, Progress in Polymer Science 37 (2012) 487. [14] G. Zhang, X.W. Lou, Sci. Rep 3 (2013). [15] M. Kotal, A.K. Thakur, A.K. Bhowmick, ACS Appl. Mater. Interfaces 5 (2013) 8374. [16] M. Toupin, T. Brousse, D. Bélanger, Chem. Mater. 16 (2004) 3184. [17] H. Wang, L. Zhang, X. Tan, C.M.B. Holt, B. Zahiri, B.C. Olsen, D. Mitlin, J. Phys. Chem. C 115 (2011) 17599. [18] L. Hu, W. Chen, X. Xie, N. Liu, Y. Yang, H. Wu, Y. Yao, M. Pasta, H.N. Alshareef, Y. Cui, ACS Nano 5 (2011) 8904. [19] G. Yu, L. Hu, M. Vosgueritchian, H. Wang, X. Xie, J.R. McDonough, X. Cui, Y. Cui, Z. Bao, Nano Lett. 11 (2011) 2905. [20] Z. Lei, F. Shi, L. Lu, ACS Appl. Mater. Interfaces 4 (2012) 1058. [21] S. Deng, D. Sun, C. Wu, H. Wang, J. Liu, Y. Sun, H. Yan, Electrochim. Acta 111 (2013) 707. [22] S. Yang, X. Song, P. Zhang, L. Gao, ACS Appl. Mater. Interfaces 5 (2013) 3317. [23] X. Jin, W. Zhou, S. Zhang, G.Z. Chen, Small 3 (2007) 1513. [24] S. Lee, J. Ha, H. Cheng, J.W. Lee, T.S. Jang, Y.-G. Jung, Y. Huang, J.A. Rogers, U. Paik, Adv. Energy Mater. 4 (2014) 1300472. [25] Y. Peng, Z. Chen, J. Wen, Q. Xiao, D. Weng, S. He, H. Geng, Y. Lu, Nano Res. 4 (2011) 216. [26] W. Li, G. Li, J. Sun, R. Zou, K. Xu, Y. Sun, Z. Chen, J. Yang, J. Hu, Nanoscale 5 (2013) 2901.