Preparation of carbon nanotube-activated carbon ...

1 downloads 0 Views 2MB Size Report
Dec 28, 2015 - Mohammad Mahmudul Huq, Chien-Te Hsieh ⁎, Chia-Yin Ho ..... S.-H. Jeong, K.-H. Lee, H.S. Kim, C.G. Park, J.H. Han, Electrical properties of.
Diamond & Related Materials 62 (2016) 58–64

Contents lists available at ScienceDirect

Diamond & Related Materials journal homepage: www.elsevier.com/locate/diamond

Preparation of carbon nanotube-activated carbon hybrid electrodes by electrophoretic deposition for supercapacitor applications Mohammad Mahmudul Huq, Chien-Te Hsieh ⁎, Chia-Yin Ho Department of Chemical Engineering and Materials Science, Yuan Ze University, Taoyuan 32003, Taiwan

a r t i c l e

i n f o

Article history: Received 27 August 2015 Received in revised form 9 December 2015 Accepted 20 December 2015 Available online 28 December 2015 Keywords: Electrophoretic deposition Activated carbon Carbon nanotubes Supercapacitors

a b s t r a c t This study examines the possibility of preparing activated carbon (AC) and carbon nanotubes (CNT) based electrodes for supercapacitors by a facile electrophoretic deposition (EPD) method. The EPD method is able to deposit AC and CNTs on a stainless steel substrate, fabricating a three dimensional porous structure. The weight proportions of CNT to AC in the as-prepared electrodes are estimated by thermogravimetric analysis. The amount of CNTs in the electrodes can be easily controlled by tuning the solution bath composition before the EPD process. The effects of the presence of CNTs in the AC electrode on supercapacitor performance are also examined in this work. The addition of CNTs not only increases the specific capacitance but also enhances the rate capability of the AC electrodes. After a cyclic stability test for 11,000 cycles, the as-prepared AC/CNT electrode shows capacitance retention of 85%, clearly demonstrating the commercial applicability of the electrodes fabricated by EPD process. © 2015 Elsevier B.V. All rights reserved.

1. Introduction As a result of the continuing drive to the development of renewable energy sources, developing electric vehicles, hybrid electric vehicles and power saving devices is the priority now. In order to back-up the development, more effective energy storage devices like batteries and supercapacitors are expected to become more efficient [1]. With their high energy and power density and long cycle life, supercapacitors have the potential to bridge the gap between traditional batteries and dielectric capacitors [2–4]. Basically, there are two kinds of supercapacitors: (i) electrochemical double layer capacitors (EDLC) and (ii) pseudocapacitors [5]. EDLCs mainly consist of porous carbon materials and offer better power density and longer cyclic stability, but suffer from low energy density [6]. On the other hand, pseudo-capacitors are metal oxides or conducting polymer based, characterized by high energy density, but show low cyclic stability and poor power density [7]. Activated carbon (AC) has been in use as an EDLC electrode material for a long time because of its high capacitance, low cost and long cycle life [8–12]. In AC, an electrical double layer of charges are formed at the AC/electrolyte interface; there is no significant faradaic reaction. However, AC has drawbacks of poor rate capability because of its microporous structure and low conductivity [13]. To alleviate this problem, some groups have developed electrodes consisting of AC mixed with a small amount of carbon nanotubes (CNTs) [14]. Recently, CNTs have

⁎ Corresponding author. E-mail address: [email protected] (C.-T. Hsieh).

http://dx.doi.org/10.1016/j.diamond.2015.12.014 0925-9635/© 2015 Elsevier B.V. All rights reserved.

been extensively used as an electrode material for supercapacitors because of their high conductivity, stability, low density and narrow pore size distribution [15–21]. However, CNTs show a poor specific capacitance mostly due to their low specific surface area [8,13]. Unlike previous reports, this study adopts an electrophoretic deposition (EPD) method to fabricate AC/CNT electrodes. In an EPD process, the electrode current collector is immersed in colloidal solution of the nano-materials. The charged nano-materials are deposited continuously on the current collector as a voltage difference is applied across it. The EPD method has some major advantages over other coating techniques, making it a very attractive synthesis method for preparing supercapacitor electrodes in recent years. Firstly, it is a very simple process requiring just a DC current source and one solution bath container; secondly, the EPD process takes place at ambient temperature; thirdly, the process is cheap and scalable to industrial production [22–23]. Another advantage of EPD processes is that it allows dispersing different nano-materials very uniformly on the electrode, whereas other methods like slurry coating and spin coating show poor dispersion resulting from poor dispersibility of nano-sized materials in slurry composing of solvent, nano-materials, polymer binders and conducting agents [24–25]. Lastly, having no binding material in the EPD made electrode not only reduces the cost but also ensures no unwanted blockage in the microporous materials. In this research, we report the fabrication of AC-based electrode by EPD method for the first time. The effects of adding CNTs in the AC structure in EPD process have also been discussed. This study will open the pathway for EPD to fabricate AC/CNT-based electrode for highperformance supercapacitors.

M.M. Huq et al. / Diamond & Related Materials 62 (2016) 58–64

2. Experimental section The commercial AC powder used here was purchased from the First Chemical Group (Taiwan). Multi-walled CNTs were procured from Taiwan Maxwave Co. Ltd. All other chemicals such as magnesium nitrate (Mg(NO3)2), nitric acid (HNO3), and ethanol were reagent grades and used as received. Herein AC powder and CNTs were pretreated according to the following procedure: first, 500 mg of carbon samples (i.e., AC or CNTs) were stirred in 1000 ml of 1 M HNO3 at 95 °C for 2 h. Then, the suspension was vacuum filtered with a filter paper and the obtained slurry was washed with deionized water until the pH becomes 7. Lastly, the carbon powders were dried in a vacuum oven overnight. The solution bath for the EPD contained 12 mg of AC and CNTs with varying proportions, 14 mg of Mg(NO3 )2 and 50 ml of ethanol. Herein Mg(NO3 ) 2 served as a charging agent. Prior to the EPD process, the solution bath underwent 1 h of sonication treatment. In the present work, 2 cm by 2 cm of stainless steel foils (SS) were used as current collectors for the electrodes. The SS foils were also given a sonication treatment in acetone for 1 h so as to remove any oil from the surface. EPD was carried out by applying a voltage of 100 V, using the SS current collector as the negative electrode and another SS foil of the equal size as the positive electrode. Deposition time was set to 5 min for making each electrode. After the EPD process, the electrodes were dried in an oven and then given a heat treatment at 300 °C in nitrogen environment for 1 h. For identification, electrodes prepared from solution baths containing 50% and 25% CNTs content are designated as AC-CNT-1 and AC-CNT-2, respectively. In order to calculate the specific capacitance of AC in AC/CNT electrodes, an electrode containing solely CNTs was also prepared by the same method. To find out whether EPD method can retain the capacitance of the active material, one AC electrode was made by dropping ethanol containing AC particles on a carbon paper. This electrode had a weight loading of 1.5 mg cm− 2 and had no binder material. The electrode is designated as AC-D. To estimate the individual weight loading of AC and CNTs in ACCNT-1 electrodes after EPD, two separate solution baths, one containing solely AC (12 mg of AC/50 ml of ethanol) and another containing solely CNTs (12 mg of CNTs/50 ml of ethanol) were prepared. Then, AC and CNTs were deposited by EPD method on separate SS foils with different deposition times. The weights of the AC and CNT electrodes were taken with a scale (Mettler Toledo AX205, Switzerland) of high precision (±0.1 mg). Herein 5 samples of each type were prepared with different deposition times. The weight proportions of CNTs in different electrodes were estimated with a thermos-gravimetric analyzer (TGA, Perkin Elmer TA7). The TGA samples were collected by mildly scratching the electrode. Prior to the experiment, the samples were washed in nitric acid so as to remove any Mg. The experiments were carried out under air atmosphere at a heating rate of 10 °C min−1. The crystalline structure of the as prepared AC/CNT electrode material was analyzed by X-ray diffraction (XRD) with Cu-Kα radiation, using an automated X-ray diffractometer (Shimadzu Labx XRD-6000). In order to remove interference from the SS current collector during XRD, the electrode materials were carefully scratched off the SS foil. Scanning electron microscopy (FE-SEM, JEOL 2010F) was used to examine the morphology and structure of the as prepared electrodes. The electrochemical performance of the AC/CNT electrode was studied by a three-electrode system with 1 M Li 2 SO 4 electrolyte at ambient temperature, where a Pt wire and Ag/AgCl were used as counter and reference electrodes, respectively. Cyclic voltammetry (CV) was conducted within the voltage window of − 0.2 to 0.6 V at scan rates ranging from 5 to 100 mV s− 1. The galvanostatic charge and discharge (GCD) cycle was also conducted at different current densities (i.e., 100–2000 mA g − 1 ). The specific

59

capacitances of the electrodes were calculated from CD curves by using the following equation CS ¼

iΔt mΔV

ð1Þ

where CS is the specific capacitance, i is the current density in GCD test, Δt the discharge time, m the mass of carbon in the electrode and ΔV the potential window. In order to examine the behavior of the electrode at different alternating current frequencies, electrochemical impedance spectroscopy (EIS) was carried out with an electrochemical impedance spectroscopy analyzer (CH Instrument, Inc., CHI 608) within the frequency range of 0.01 Hz to 100,000 Hz. 3. Results and discussion 3.1. Characterization of AC-CNT electrodes Electrode material prepared by EPD of pure AC, pure CNTs and mixture of AC/CNTs (AC-CNT-1) were characterized by XRD. Fig. 1 shows the XRD patterns of the electrodes. Electrode material composing AC shows amorphous pattern, since AC has very limited crystallinity. On the other hand, AC-CNT-1 and pure CNT electrodes show the (002) diffraction peak of CNTs at 2θ = 26.2 °, corresponding to d-spacing distance of 0.34 nm. The (002) diffraction peak from AC-CNT-1 electrode is not only a little weaker, but also broader than that of pure CNTs, confirming a successful deposition of amorphous AC and crystalline CNTs on the SS foil. Fig. 2 presents the FE-SEM images of the CNT, AC-CNT-1 and ACCNT-2 and AC electrodes. Fig. 2(a) shows the CNTs deposited by EPD. The CNTs are 30–40 nm in diameter and several micrometers in length.

Fig. 1. XRD patterns of neat AC, neat CNT and AC-CNT-1 electrode materials.

60

M.M. Huq et al. / Diamond & Related Materials 62 (2016) 58–64

Fig. 2. SEM images of (a) CNT, (b) AC-CNT-1 (c) AC-CNT-2 and (d) AC electrodes.

As observed in Fig. 2(b) and (c), CNTs are dispersed homogenously on AC particles. In case of AC-CNT-2 electrode presented in Fig. 2(c), it is evident that the amount of CNTs in this electrode is somewhat lower than that in AC-CNT-1 electrode (Fig. 2(b)). Lastly, Fig. 2(d) shows the electrode composing only AC. Herein, the size of AC particles lies between 1 and 3 μm. Overall, the FE-SEM images show the well dispersed 3D nano-porous structure of the AC/CNT electrodes prepared by EPD process. Fig. 3 presents the weight loading of pure AC and pure CNTs at different deposition times. As shown in the figure, the weight loading increases linearly with the deposition time, implying that the rate of deposition of AC as well as CNTs is steady over the range of deposition times. The individual weight deposition rate of AC and CNTs were 0.20 and 0.22 mg cm− 2 min−1, respectively. Therefore, it can reasonably be inferred that during EPD from solution bath containing 1:1 AC to CNTs weight ratio, AC and CNTs are deposited in nearly equal amount. According to this study, the weight proportion of CNTs in AC-CNT-1 electrode is estimated to be around 54.6%. The proportions of CNTs in the AC/CTNs films were again estimated by TGA analysis. In order to locate the temperature window in which decomposition of CNTs and AC happens, electrodes made of pristine CNTs and pristine AC were analyzed first. Fig. 4(a) shows the weight loss and the derivative of weight loss of pristine CNTs against temperature. Pristine CNTs have an onset temperature (the temperature at which the weight loss starts) of 590 °C and are completely burned off at 730 °C, agreeing with other reports [26–28]. On the other hand, Fig. 4(b) shows that oxidation of AC starts at a lower temperature of around 471 °C and ends at 600 °C. It must be noted that, there is an initial weight loss of 10–20% in case of pristine AC, AC-CNT-1 and AC-CNT-2 electrode materials, whereas pristine CNTs does not show any such initial

weight loss. This initial weight loss of AC electrode material can be attributed to the evaporation of physically absorbed ethanol and water in AC. Since the specific surface area of CNTs is many

Fig. 3. Weight loading of neat CNTs and neat AC against deposition time during EPD.

M.M. Huq et al. / Diamond & Related Materials 62 (2016) 58–64

61

Fig. 4. TGA analysis of electrode materials from (a) CNTs film (b) AC film (c) AC-CNT-1 film and (d) AC-CNT-2 film.

times smaller than that of AC, the amount of ethanol and water absorbed on CNTs surface is insignificant, which explains the event of no initial weight loss of CNTs. As can be seen in Fig. 4(c) and (d), there are two distinct weight loss peaks in the derivative curves, the first one belonging to the loss of AC and the second one to CNTs. Two distinct regions of weight losses corresponding to the peaks of derivative curves can also be located easily. The AC-CNT-1 electrode material shows 39% weight loss due to oxidation of AC and 41% weight loss due to CNTs. Therefore, the proportion of CNTs in the AC-CNT-1 electrode material is calculated to be 51% (almost the same as estimated by the weight loading study), excluding water content. In the same way, the

weight proportion of CNTs in AC-CNT-2 electrode is calculated to be 27.4%. 3.2. Electrochemical studies Fig. 5(a) shows the cyclic voltammetry (CV) profiles of the three AC/CNT and AC-D electrodes in 1 M Li2SO4 electrolyte at a sweep rate of 50 mV s−1. The rectangular patterns of CV profiles confirm an ideal (EDLC) behavior of the electrodes. The CV curve of the drop coated electrode (AC-D) almost overlaps the AC electrode, signifying that EPD method retains the original capacitance of AC. Fig. 5(c) represents the GCD curves of the as prepared electrodes at a current density of

62

M.M. Huq et al. / Diamond & Related Materials 62 (2016) 58–64

Fig. 5. (a) Cyclic voltammetry profiles of AC-D, CNT, AC-CNT-1, AC-CNT-2 and AC electrode normalized with the weight of AC at a scan rate of 50 mV s−1 in 1 M Li2SO4 electrolyte. (b) Specific capacitance of the as prepared electrodes normalized with AC weight at different scan rates. (c) GCD curve of AC-D, CNT, AC-CNT-1, AC-CNT-2 and AC electrode normalized with the weight of AC at 1000 mA g−1 of current density. (d) Specific capacitance of different electrodes was normalized with the weight of AC at different current densities.

1000 mA g−1. The triangular shape of GCD profiles also clearly signifies the EDLC behavior. Herein, the illustrated CV and GCD curves are normalized to the weight loading of AC. It is interesting to observe that, AC electrode shows higher potential drop than AC-CNT-1 and AC-CNT-2 electrodes. Fig. 5(b) and (d) shows the change of specific capacitance of the electrodes at different scan rates and current densities, calculated from CV and GCD profiles respectively. The specific capacitance normalized to the weight loading of AC for the electrodes AC-CNT-1 and ACCNT-2 are calculated using the following equation C AC ¼

C−C CNT f

ð2Þ

where CAC is the specific capacitance normalized to AC (F g−1), C the overall capacitance, CCNT the specific capacitance of CNTs (F g−1) and f the weight proportion of AC in the corresponding electrodes, estimated by TGA analysis (i.e., 0.49 and 0.73 for AC-CNT-1 and AC-CNT-2 respectively). AC electrode shows a 46% loss of capacitance as the current density increases from 100 to 2000 mA g−1. On the other hand, AC-CNT-2 and AC-CNT-1 electrodes show a loss of 24% and 9% respectively. The CV studies at different scan rates also show the similar trend, as shown in Fig. 5(a). Moreover, as observed from the figures, although the specific capacitance of AC electrode is higher than that of AC-CNT-

1 and AC-CNT-2 electrodes at a very low current densities (i.e. 85, 70 and 68 F g AC−1 for AC, AC-CNT-2 and AC-CNT-1 electrodes, respectively, at a current density of 100 mA g−1), as current density increases, the specific capacitance of AC electrode drops even lower than that of ACCNT-2 and AC-CNT-1 electrodes. One reason of poor rate capability is the poor electrical conductivity of active materials. Therefore, insertion of CNTs, a better conductor than AC, into the electrode improves the overall material conductivity greatly, leading to a high rate performance. Another reason for poor rate performance is that at higher current densities the ionic migration in the porous structure of the electrode becomes too slow that ions cannot access the whole porosity, which finally leads to a decrease in capacitance. Therefore it is probable that the loss of capacitance is curbed with the introduction of CNTs, because the pore accessibility is higher with mesoporous CNTs than with microporous AC [8]. An EIS technique was adopted in order to further evaluate the performance of EPD prepared AC/CNT and AC electrodes. Fig. 6(a) represents the Nyquist plots of the as prepared electrodes within a wide range of frequencies (0.01 to 100,000 Hz). At a higher frequency, the electrodes work like resistances and in the range of low frequency the imaginary part of impedance rises, revealing the ion migration phenomena, thus showing capacitive behavior. The inset of Fig. 6(a) presents the high frequency behavior of the electrodes. Semicircle loops in the high frequency area signifies the electrode material resistance and the

M.M. Huq et al. / Diamond & Related Materials 62 (2016) 58–64

63

resistance at the interface between the current collector and the electrode material. As the interfacial resistances in all three electrodes can be assumed very low and equal, a decrease in the semicircle diameter directs to the decrease in electrode material resistance. Therefore, the high frequency behavior of the electrodes confirms that the addition of CNT improves the electrode conductivity, which is reasonable because electronic conductivity of CNTs is higher than that of AC. However, the occurrence of incomplete semicircles here might happen due to non-uniform current distribution arising from uneven electrode thickness [29]. Future work should address this problem. The Bode plots for the electrodes in low frequency region are presented in Fig. 6(b). Generally, larger frequency of approach to the phase angle means better capacitive performance and fast charge and discharge performance [30]. Herein, at frequency lower than 10 Hz, where the impedance behavior of the electrodes are diffusion controlled, the phase angles of AC-CNT-1 and AC-CNT-2 electrodes are always lower than those of AC. Therefore, rapid diffusion of Li ion happens in the CNT carrying electrodes due to the mesoporous structure of CNTs. Fig. 6(c) shows the Randle's plots for the electrodes, which relate the angular frequency to the real impedance. It is possible to estimate the ionic diffusivity of the electrolyte ions from this plot using the following equation  D¼

RT √2A F 2 σC

2 ð3Þ

where D is the diffusion coefficient, R the gas constant, T the absolute temperature, A the electrode area, F the Faraday's constant, C the electrolyte concentration and σ the coefficient of Warburg impedance [31]. Herein, Warburg impedance (σ) is actually the slope of the Randle's plot. In our calculation we took five final points from the low frequency region of the Randle's plot, as shown in the inset of Fig. 6(c). The calculated diffusion coefficients follow an order of AC-CNT-1 (5.56 × 10− 10 cm2 s− 1) N AC-CNT-2 (3.34 × 10− 10 cm2 s− 1) N AC (6.35 × 10− 11 cm2 s− 1). It is evident from the diffusion coefficient data that diffusivity in an AC/CNT-based electrode is significantly improved as compared to pure AC based ones. This enhanced diffusivity implies that Li ions can reach the active sites more easily. This phenomenon may be explained by the fact that the pore accessibility is higher with mesoporous CNTs than with microporous AC. Therefore, this result suggests that the insertion of CNTs in AC facilitates fast charge and discharge ability, thus explaining the high capacitive retention at higher current densities. In order to examine the stability of the electrodes made by EPD method, a cyclic voltammetry test was carried out for 11,000 times at a scan rate of 100 mV s− 1. An AC-CNT-1 electrode was used in this study. Fig. 7(a) represents the CV patterns. As shown in Fig. 7(b), the capacitance retention is 85% after 11,000 cycles, which implies that the carbon particles in the electrode material are strongly adhered to each other and the electrode is stable over time. This result clearly shows the high applicability of the as prepared electrode in commercial fields. 4. Conclusions

Fig. 6. (a) Nyquist plots of AC-CNT-1, AC-CNT-2 and AC electrodes. (b) Bode plots. (c) Randle's plots.

In this study we have demonstrated an EPD method to prepare AC and CNT based EDLC electrode. TGA and SEM show that the amount of CNTs in the electrode can be controlled by varying the amount of CNTs in the solution bath. The presence of CNTs in the AC electrode has positive effect on the charge and discharge performance of the electrodes. Therefore, supercapacitor electrode with optimized capacitance and charge discharge performance can be fabricated by EPD. The long term (11,000 cycles) cyclic stability test of AC-CNT-1 electrode displays capacitive retention of 85%, confirming that simple EPD method can be used commercially to fabricate AC/CNT based supercapacitor electrodes.

64

M.M. Huq et al. / Diamond & Related Materials 62 (2016) 58–64

Fig. 7. (a) Cyclic voltammogram of AC-CNT-1 electrode at 100 mV s−1 of scan rate for 11,000 cycles in 1 M Li2SO4. (b) Specific capacitance normalized with AC weight with the number of cycles.

In the future, this method can also be used to decorate pseudocapacitive materials on AC and CNT architecture by one step EPD. Prime novelty statement A facile electrophoretic deposition process has been utilized to fabricate activated carbon and CNT based 3D nanoporous electrodes for the first time. Thermogravimetric analysis of the electrode materials has shown that the amount of CNTs in the electrode can easily be controlled by varying their proportion in the EPD solution baths. Acknowledgments The authors are very grateful for the financial support from the Ministry of Science and Technology, Taiwan under the contract MOST 103-2221-E-155-014-MY2. References [1] P. Simon, Y. Gogotsi, Materials for electrochemical capacitors, Nat. Mater. 7 (2008) 845–854. [2] G. Wang, L. Zhang, J. Zhang, A review of electrode materials for electrochemical supercapacitors, Chem. Soc. Rev. 41 (2012) 797–828. [3] A. Burke, Ultracapacitors: why, how, and where is the technology, J. Power Sources 91 (2000) 37–50. [4] Q. Lu, J.G. Chen, J.Q. Xiao, Nanostructured electrodes for high performance pseudocapacitors, Angew. Chem. Int. Ed. 52 (2013) 1882–1889. [5] R. Kötz, M. Carlen, Principles and applications of electrochemical capacitors, Electrochim. Acta 45 (2000) 2483–2498. [6] A. Ghosh, Y.H. Lee, Carbon-based electrochemical capacitors, ChemSusChem 5 (2012) 480–499. [7] M. Zhi, C. Xiang, J. Li, M. Li, N. Wu, Nanostructured carbon-metal oxide composite electrodes for supercapacitors: a review, Nanoscale 5 (2013) 72–88. [8] C. Portet, P.L. Taberna, P. Simon, E. Flahaut, Influence of carbon nanotubes addition on carbon–carbon supercapacitor performances in organic electrolyte, J. Power Sources 139 (2005) 371–378. [9] J. Li, X. Wang, Q. Huang, S. Gamboa, P.J. Sebastian, Studies on preparation and performances of carbon aerogel electrodes for the application of supercapacitor, J. Power Sources 158 (2006) 784–788. [10] B. Fang, L. Binder, A modified activated carbon aerogel for high-energy storage in electric double layer capacitors, J. Power Sources 163 (2006) 616–622. [11] V. Khomenko, E. Raymundo-Piñero, F. Béguin, High-energy density graphite/AC capacitor in organic electrolyte, J. Power Sources 177 (2008) 643–651. [12] C.-C. Hu, C.-C. Wang, F.-C. Wu, R.-L. Tseng, Characterization of pistachio shell-derived carbons activated by a combination of KOH and CO2 for electric double-layer capacitors, Electrochim. Acta 52 (2007) 2498–2505.

[13] X. Geng, L. Li, F. Li, Carbon nanotubes/activated carbon hybrid with ultrahigh surface area for electrochemical capacitors, Electrochim. Acta 168 (2015) 25–31. [14] F. Markoulidis, C. Lei, C. Lekakou, D. Duff, S. Khalil, B. Martorana, I. Cannavaro, A method to increase the energy density of supercapacitor cells by the addition of multiwall carbon nanotubes into activated carbon electrodes, Carbon 68 (2014) 58–66. [15] M.F.L. De Volder, S.H. Tawfick, R.H. Baughman, A.J. Hart, Carbon nanotubes: present and future commercial applications, Science 339 (2013) 535–539. [16] S. Iijima, Helical microtubules of graphitic carbon, Nature 354 (1991) 56–58. [17] F. Frackowiak, K. Jurewicz, K. Szostak, S. Delpeux, F. Beguin, Nanotubular materials as electrodes for supercapacitors, Fuel Process. Technol. 77-78 (2002) 213–219. [18] B.-J. Yoon, S.-H. Jeong, K.-H. Lee, H.S. Kim, C.G. Park, J.H. Han, Electrical properties of electrical double layer capacitors with integrated carbon nanotube electrodes, Chem. Phys. Lett. 388 (2004) 170–174. [19] A.K. Chatterjee, M. Sharon, R. Banerjee, M. Neumann-Spallart, CVD synthesis of carbon nanotubes using a finely dispersed cobalt catalyst and their use in double layer electrochemical capacitors, Electrochim. Acta 48 (2003) 3439–3446. [20] C. Emmenegger, P. Mauron, A. Zuttel, C. Niitzenadel, A. Schneuwly, R. Gallay, L. Schlapbach, Carbon nanotube synthesized on metallic substrates, Appl. Surf. Sci. 162–163 (2000) 452–456. [21] J.H. Chen, W.Z. Li, D.Z. Wang, S.X. Yang, J.G. Wen, Z.F. Ren, Electrochemical characterization of carbon nanotubes as electrode in electrochemical double-layer capacitors, Carbon 40 (2002) 1193–1197. [22] G. Lota, K. Fic, E. Frackowiak, Carbon nanotubes and their composites in electrochemical applications, Energy Environ. Sci. 4 (2011) 1592–1605. [23] I. Corni, M.P. Ryan, A.R. Boccaccini, Electrophoretic deposition: from traditional ceramics to nanotechnology, J. Eur. Ceram. Soc. 28 (2008) 1353–1367. [24] B.J.C. Thomas, A.R. Boccaccini, M.S.P. Shaffer, Multi-walled carbon nanotube coatings using electrophoretic deposition (EPD), J. Am. Ceram. Soc. 88 (2005) 980–982. [25] M.-S. Wu, Y.-H. Fu, Tubular graphene nanoribbons with attached manganese oxide nanoparticles for use as electrodes in high-performance supercapacitors, Carbon 60 (2013) 236–245. [26] N. Chigumbu, S. Iyuke, V. Pillay, S. Ndlovu, In vitro evaluation of the physicochemical effects of drug loaded carbon nanotubes on toxicity, J. Nanosci. Nanotechnol. 3 (2012), e135. [27] H. Fang, S. Zhang, X. Wu, W. Liu, B. Wen, Z. Du, T. Jiang, Facile fabrication of multiwalled carbon nanotube/α-MnOOH coaxial nanocable films by electrophoretic deposition for supercapacitors, J. Power Sources 235 (2013) 95–104. [28] H. Fang, S. Zhang, T. Jiang, R. Lin, Y. Lin, One-step synthesis of Ni/Ni(OH)2@ multiwalled carbon nanotube coaxial nanocable film for high performance supercapacitors, Electrochim. Acta 125 (2014) 427–434. [29] Q. Cheng, Z. Chen, The cause analysis of the incomplete semi-circle observed in high frequency region of EIS obtained from TEL-covered pure copper, Int. J. Electrochem. Sci. 8 (2013) 8282–8290. [30] C. Portet, P.L. Taberna, P. Simon, C.L. Robert, Modification of Al current collector surface by sol–gel deposit for carbon–carbon supercapacitor applications, Electrochim. Acta 49 (2004) 905–912. [31] M. Shi, Z. Chen, J. Sun, Determination of chloride diffusivity in concrete by AC impedance spectroscopy, Cem. Concr. Res. 29 (1999) 1111–1115.