activated carbon hybrid with ultrahigh surface area for electrochemical capacitors

Accepted Manuscript Title: Carbon nanotubes/activated carbon hybrid with ultrahigh surface area for electrochemical capacitors Author: Xin Geng Lixiang Li Feng Li PII: DOI: Reference:

S0013-4686(15)00869-5 http://dx.doi.org/doi:10.1016/j.electacta.2015.03.220 EA 24746

To appear in:

Electrochimica Acta

Received date: Revised date: Accepted date:

20-1-2015 19-3-2015 31-3-2015

Please cite this article as: Xin Geng, Lixiang Li, Feng Li, Carbon nanotubes/activated carbon hybrid with ultrahigh surface area for electrochemical capacitors, Electrochimica Acta http://dx.doi.org/10.1016/j.electacta.2015.03.220 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Carbon nanotubes/activated carbon hybrid with ultrahigh surface area for electrochemical capacitors Xin Geng ab, Lixiang Li a, a

Feng Li

*b

Institute of material electrochemistry, School of Chemical Engineering, University of Science

and technology Liaoning, Anshan 114015, China. b

Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese

Academy of Sciences, Shenyang 110016, China.

*

Corresponding author. Tel.: 86-24-83970065; Fax: +86 24 2390 3126. E-mail: [email protected] (F.

Li).

Abstract A hybrid carbon materials with three dimensional network carbon nanotubes in activated carbon was prepared by an in situ activation process. This hybrid carbon material has ultrahigh specific surface area (2879 m2/g), pore volume (1.36 cm3/g) and developed small mesopore network, which is very favorable for fast electrolyte ion transport in the porous channel. As evidenced by N2 adsorption, cyclic voltammetry and electrochemical impedance measurements, the hybrid carbon material shows superior capacitive behaviors (exhibiting a high capacitance of 140 F/g, better capacitance retention ratio of 80% even at very high sweep rate of 200 mV/s) and provides much higher power density while still maintaining good energy density. This hybrid carbon material offers a great potential in electrochemical

capacitors, particularly for applications where high power output and good high-frequency capacitive performances are required.

Keywords: Carbon nanotube; activated carbon; electrochemical capacitor

1. Introduction In response to the increasing demands for clean energy technologies, electrochemical capacitors (ECs) are considered as the most promising energy storage and power output technology for portable electronics, electric vehicles, and renewable energy systems operated on intermittent sources such as solar and wind. Porous carbons, transition metal oxides, and conducting polymers are fundamental candidates as electrode materials of ECs. However, metal oxides and conducting polymers are disturbed by several key drawbacks. For example, the metal oxides are either too expensive (RuO2) or poorly conductive (NiO, MnO2, etc.), and the conducting polymers show short cycling life [1-11]. Porous carbon materials have been commercially applied in ECs as the availability with stable physicochemical properties, good conductivity, and low cost [1-11]. Even though, for the emerging applications, the fundamental improvements of carbon electrode materials are needed. Currently, the performance specifications of EC, e.g. power density and corresponding energy density, intensively depend on the characteristics of porous carbon. The porous textures and the conductivity of the carbon materials directly

impact on the equivalent series resistance (ESR) and the internal distribution of electrolyte resistance (IER) in the porous carbon [12-14]. These properties ultimately determine the power densities and energy densities of ECs. Therefore, two important parameters, the ESR of the electrode itself and ion transport resistance (which corresponds to the IER) of the electrolyte in pore channels, need to be controlled as low as possible [15-17]. Although activated carbon (AC) shows the advantage of higher energy density, their power densities are restrained by the microporous structure and low conductivity. As a result, a variety of mesoporous carbons have been developed. Especially, the hierarchical porous structure of porous carbon, where micropore is nesting inside mesopore, greatly improves the power performances of ECs. Carbon nanotubes (CNTs) have characteristics of high stability, low resistance, and narrow pore size distribution. It is promised that CNTs can resolve the low conductivity and adjust the pore size of the AC, if CNTs and AC were combined. Few papers have evaluated the effects of CNTs added in the AC on the ESR, IER and pore structure of AC [18]. In our previous research, we found that the characteristic of capacitance response versus frequency was improved and the electrode resistance was decrease by the addition of CNTs to activated carbon [19]. Meanwhile, the capacitance and the efficiency of energy storage for ECs were improved, when activated carbon was mixed with a mass fraction of 3-15% of MWCNTs. Based on the above outline of the porous textures and the conductivity of the carbon materials in ECs, this paper reports the one-step preparation of CNTs/AC

hybrid by in situ activation process. The effects of CNTs on pore feature and electrochemical properties of CNTs/AC hybrid were intensively investigated. This approach minimizes both electrical resistance of CNTs/AC hybrid and ionic resistance of electrolyte in the resulting carbon electrodes. CNTs/AC hybrid shows the evidently improved characteristic of specific power and specific energy. 2. Experimental 2.1 Preparation and Characterization of CNTs/AC hybrids CNTs were synthesized by catalytic decomposition of hydrocarbons using a floating catalyst method with a horizontal reaction furnace [20]. After that, the raw CNTs were purified by two-step oxidation process of a nitric acid refluxing and an air oxidation. Then the CNTs were immersed in hydrochloric acid and washed with distilled water until neutrality. Phenol-formaldehyde resin (PF-resin) was used as the precursor of AC. After the resin was dissolved in KOH (a ratio of the resin and KOH = 1:4 by weight) ethanol solution in a plastic flask, the CNTs dispersed in ethanol were added to the above solution, and the solvents were removed at 393 K for 12 h in an oven to get the as-made sample. Hereafter, it was activated under the nitrogen atmosphere at 1073 K for 1 h. Finally, the mixture was impregnated in 0.1 M hydrochloric acid, washed with distilled water to eliminate residual acid, and dried at 393 K for 12 h to get the three dimension CNTs network embedded in CNTs/AC hybrid, where the yield of CNTs was around 5 wt.% of the CNTs/AC hybrid (In terms of the electrochemical

performances and cost [19], around 5% CNTs was introduced to the hybrid carbon materials). In addition, AC without the CNTs was also prepared as the blank sample under the same experimental condition (designated as AC800). Nitrogen cryo-sorption measurements were conducted on a Micromeritics ASAP2010M instrument; before measurements, the samples were evacuated at 473 K until the manifold pressure was lower than 2 Pa. The morphology and structures were characterized by SEM (LEO, SUPRA 35, 15 kV) and TEM (JEOL JEM 2010, 200 kV). The samples were dispersed in absolute ethanol and dropped onto the Cu rod for SEM or the grid for TEM observations.

2.2 Electrochemical measurements A coin-type capacitor cell was used to examine the electrochemical performance of the AC800 and CNTs/AC hybrid electrodes, which consisted of a carbon film (8 mm in diameter and about 200 µm in thickness) and nickel foil as the current collector. This cell consisted of two facing carbon electrodes, sandwiching a cellulose filter paper as the separator. All electrochemical measurements were carried out using 6 M KOH as the electrolyte solution at room temperature. The cyclic voltammetry (CV) and electrochemical impedance spectroscope (EIS) measures were collected on Solartron 1287/1260 electrochemical systems from 5~200 mV/s. The potential range for CV was 0.0~1.0 V. The bias potential applied to the electrode during EIS was 0 V, the frequency ranges were 5 mHz to 100 kHz, and the ac signal amplitude was 10 mV.

The constant-current charge-discharge measures were performed with a BT2000 Arbin cycler at different current densities from 0.01 to 10 A/g between 0 and 1.0V.

3. Results and Discussion 3.1. The pore Structure Characterization of the CNTs/AC hybrid SEM morphologies of the AC800 and CNTs/AC hybrid are presented in Fig. 1(a) and (b). The surfaces of the AC reserve the caves etched by KOH (Fig. 1(a)). These cavities result from the removal of impregnated KOH and KOH-derived compounds, leaving the space previously occupied by the compounds. The AC particles in the CNTs/AC hybrid can be linked by the CNTs after the CNTs added in Fig. 1(b). Especially, the two ends of a few CNTs are located in different AC particles, respectively. The linking manner of CNTs and AC particle facilitates the electrons transportation between the AC particles. The TEM images distinctly show that AC particles are wound by CNTs (Fig. 1(c)), and the CNTs have been embedded in the AC particles (Fig. 1(d)). These results indicate that the three dimensional network of CNTs/AC hybrid is formed and the conductivity of the CNTs/AC hybrid is to be improved. Fig. 1(e) shows the SEM image of purified CNTs, revealing two types of the diameter distribution of CNTs. The one is between 15 and 20 nm. The other is between 100 and 200 nm. The ratio of length/diameter of CNTs is high to favor the formation of conductivity network. The CNTs are clean (no amorphous deposit) and separated, and are aggregated as small bundles. Thermogravimetric analysis ( 10C/min

in air flow, Fig.1(f) ) shows only one well-resolved peak corresponding to sharp weight loss at 6440C, corresponding to 99.4% of the initial weight, suggesting that CNTs are purified well and the catalyst is almost removed. The nitrogen adsorption isotherm of CNTs in Fig. 1(g) shows the specific characteristic of porous structure at different pressure. The isotherms are of a type I characteristic, as indicated by the fact that the amount adsorbed increases rapidly at ultra-low pressures (P/Po = 10-6 to 0.01). The isotherm in the medium pressure zone (P/Po = 0.01-0.45) shows a surface adsorption process in which the amount of nitrogen adsorption slowly increases. This is indicative of the existence of a nonporous surface, attributed by external surface of the CNTs. The capillary condensation can be observed occurring in the medium relative pressure range (P/Po = 0.45-0.85), and results from capillarity in the mesopores. The hysteresis loop (P/Po = 0.85-0.99) of the isotherms corresponds to the larger pores of 20.0-40.0 nm. The BET specific surface area of CNTs is around 120 m2/g. Consequently, the capacitance of CNTs is very small. Namely, it can be neglected that capacitance of CNTs contribute to ECs. A schematic illustration of a three dimensional network texture of CNTs/AC hybrid is shown in Fig. 1(h). The CNTs/AC hybrid is composed of CNTs embedding inside AC and winding AC. The improvement of capacitance property results from the increase of conductivity of the CNTs/AC hybrid. Nitrogen adsorption-desorption isotherm with a relatively broad knee of the CNTs/AC hybrid, shown in Fig. 2(a), exhibits combined characteristics of type I/II

[21, 22], with a Brunauer-Emmett-Teller (BET) surface area of 2879 m2/g, a total pore volume of 1.36 cm3/g, a mesopore volume of 0.68 cm3/g, a mesopore to total pore volume ratio of 0.51, as well as an average pore diameter of 2.2 nm. These pore structure parameters of CNTs/AC hybrid are much higher than those of AC800 (1759 m2/g, 0.93 cm3/g and 0.45 cm3/g), although the isotherm type of AC800 is the same as CNTs/AC hybrid, indicative of CNTs play an important role on the 3 dimension network structure of CNTs/AC hybrid. The BET surface area and pore volume are illustrated in Table 1. Specific surface area calculated from the t-plot method.

Peak value of the pore

diameter distribution curves obtained from the DFT method. The mesopore volume is determined by extracting micropore volume, which is obtained from t-plot analysis, from total pore volume. The pore-size distributions derived from the density function theory (DFT), as shown in Fig. 2(b) and 2(c), are indicative of a quite uniform pore size distribution centered in micropore ranges of about 0.64 nm, 0.80 nm, 1.2 nm, 1.48-1.59 nm and in mesopore range of about 2.8 nm for CNTs/AC hybrid and AC800, respectively. The porous volume of CNTs/AC hybrid is distinctly higher than AC800, especially much higher at 1.2 nm and 2.8 nm. For CNTs/AC hybrid, the pore size distribution curve in mesopore range (2.0~4.5 nm) demonstrates three overlapped peaks related to the three types of mesopores. The pore size distribution gradually transforms from micropores to mesopores. Consequently, it is believed that the micropores attach to these

mesopores. Because activated agents KOH are intensively adsorbed on the CNTs as well, besides adsorbed on the precursor particle, the activation reactions were simultaneously developed inside and outside of the precursors. A part of KOH were gradually transformed into liquid potassium and gaseousness potassium with the increase of activated temperature. Different states of potassium can react with carbon atoms to form different pore size at different activated temperature. The mesopores in the CNTs/AC hybrid result from liquid potassium adsorbed on the surface of the CNTs etching the carbon atoms of the precursor. Nevertheless, The gaseousness potassium developes microsopores in the CNTs/AC hybrid. Therefore, the interconnected pores with mesopores and micropores can be formed in the CNTs/AC hybrids, which will be suitable for the diffusion of nitrogen molecules and electrolyte ions. As a result, nitrogen molecules can be introduced into nearly all of micropores, which contribute to high specific surface area and pore volume. There have been some literatures to elucidate that the pore sizes (<1nm) are available for the electrolyte ions [23, 24]. The CNTs/AC hybrids with high specific surface area and mesopore volume will possess better power performance.

3.2. Electrochemical Performance of the CNTs/AC hybrid The influence of CNTs on the ion transport and capacitive behavior of CNTs/AC hybrid can be characterized by CV measurements based on the evaluation of capacitive performance of ECs. Generally, the desired capacitive performance

requires a rectangular shape voltammogram. Fig. 3(a) and 3(b) exhibit that CNTs/AC hybrid maintains better rectangular shapes with the increase of voltage sweep rate, indicative of the excellent capacitive behavior even in quick charge-discharge operation. However the voltammogram of AC800 becomes gradually distorted, which indicates that the quick discharge capability of AC800 is inferior to CNTs/AC hybrid. At the low sweep rate of 5mV/s, all carbon materials exhibit similar capacitive behavior. The rectangular shape of voltammogram of CNTs/AC hybrid is less distorted than AC800 at voltage sweep rate of 200mV/s, which means that CNTs/AC hybrid possesses better capacitive behavior and quick discharge capability than AC800 [25, 26]. The result originates from their different conductivity which is illuminated by the dashed range of rectangular in Fig. 3(c). That is, the larger the slope of ΔI/ΔV around 1.0V, the better the conductivity of electrode material. Consequently, the conductivity of CNTs/AC hybrid is better than that of AC800, whereas AC800 resembles the resistance element because its voltammogram shows the spindle shape. On the one hand, the better conductivity of CNTs/AC hybrid should be attributed to the role of CNTs. As shown in Fig. 1, the AC particles are embedded and wound by the CNTs, which improves the conductivity of CNTs/AC hybrid. On the other hand, the better conductivity correlates highly with the pore structure of CNTs/AC hybrid, where interconnecting micro-mesopore pore structures are favorable for the electrolyte ion transport. Therefore, the more pore surface wetted, the better the conductivity as well. To evaluate the capacitive behavior of CNTs/AC

and AC800, the ratio of retained gravimetric capacitance v.s. increased voltage sweep rate is plotted in Fig. 3(d). CNTs/AC hybrid maintains 80% of its capacitance at a high voltage sweep rate of 200 mV/s, which is much more than the rate of 60% for AC800. Ragone plot for the carbon is displayed in Fig. 4. The dependence of the power and energy density is calculated by means of constant-current charge-discharge of a ECs using a cell-voltage window of 1V and current densities between 0.01 and 10 A/g. the energy and power density can be described as: E  CV 2 / 2 and P  E /  t , where C , V , t are the gravimetric capacitance, the cell voltage and the time spent in discharge. As the power density increases from 150 W/kg to 2200 W/kg, the energy density of AC800 drops from 8.4 Wh/kg to 6.1 Wh/kg. Comparatively, CNTs/AC hybrid can reach much higher power density (about 245-3778 W/kg), while still maintain high energy density (about 10.5-13.6 Wh/kg). Consequently, CNTs/AC hybrid exhibits better electrochemical properties than AC800. Especially, with the increase of power output, the energy densities of CNTs/AC hybrid and AC800 only decrease a little, which shows that CNTs/AC hybrid and AC800 have excellent electrochemical properties of high energy density and power output, therefore very promising for application in the scenarios where high power output as well as high energy density are required [33].

3.3. Electrochemical Impedance Spectroscopy of CNTs/AC hybrid

Impedance spectroscopy measurements are conducted to evaluate the electrochemical behavior of CNTs/AC hybrid electrode. Fig. 5(a) exhibits the complex-plane impedance plots in the range of high frequency. The electrolyte resistance, Rs, is constant and varies with the electrolyte. A semi-circle loop in the high frequency region represents the resistance of AC particles itself and the contact resistance at the interface between the AC particles as well as that between the current collector and AC particles, where the sum of the resistance is represented by Rf. Because the contact resistance are deemed identical and negligible in all electrodes, a decrease of the Rf (Rf1 < Rf2 ) indicates a decrease of the resistance of the AC particles itself [26]. Comparatively, the Rf1 of CNTs/AC hybrid is smaller than the Rf2 of AC800, which confirms that the conductivity of AC can be preferably improved by the combination of AC and CNTs. The loop is undesired as it decreases the performances of ECs in terms of the resistance and power [27]. Following the semi-circle loop, the impedance behaviors show a non-vertical slope line, which deviates from the pure capacitor behavior. The knee frequencies are 125 Hz for CNTs/AC hybrid and 79 Hz for AC800, respectively. Generally, the knee frequency is considered to be the critical frequency where ECs begin to exhibit the capacitive behavior

[26].

Therefore,

CNTs/AC

hybrid

can

be

anticipated

to

be

charged/discharged at faster speed than AC800, which can be estimated by the change

of specific frequency in the phase angle vs. frequency plot (Fig. 5(b)). In general, the larger the frequency of the approach to the phase angle of 90°, the better the capacitive behavior, as well as the better the performance of quick charge-discharge [28]. In Fig. 5(b), the frequencies of CNTs/AC hybrid are always high than AC800 at the same phase angle from 100 Hz to 5 mHz. When the frequency is lower than 10 Hz, where the impedance behavior of both CNTs/AC hybrid and AC800 would be controlled by ion-diffusion, the phase angle of CNTs/AC hybrid is always smaller than that of AC800 at the same frequency. These results indicate more rapid diffusion of electrolyte ions in the abundant mesoporous channels of CNTs/AC hybrid. Furthermore, the impedance frequency behavior can be studied by the complex model of capacitance, which can be used to evaluate the capacitive behavior, the pulse power performance and the capacitance change of ECs. The capacitance can be defined as the following [19]: (1)

C ( w )  C ' ( w )  jC ' ' ( w )

leading to 2

(2)

2

(3)

C ' ( w )  Z ' ' ( w ) /( w Z ( w ) ) C ' ' ( w )  Z ' ( w ) /( w Z ( w ) )

where C ' ( w ) is the real part of the capacitance C (w ) , corresponding to the capacitance of the ECs measured during the constant current discharge, and C ' ' ( w ) is the imaginary part of the capacitance C (w ) , corresponding to losses in the form of energy dissipation.

Fig. 5(c) represents the variations of the imaginary part of the capacitance C (w ) with the frequencies. C ' ' ( w ) pass through a maximum at 0.15 Hz for CNTs/AC hybrid and 0.12 Hz for AC800, respectively, where the relaxation time constants are deduced from the peak frequency (f0). That is, the relaxation time can be represented by the expression  0  2f 0 . Accordingly, CNTs/AC hybrid and AC800 have the relaxation time of 1.06 s and 1.33 s. It also demonstrates that CNTs/AC hybrid possesses better power discharge performance. Fig. 5(d) presents the variations of the real part of the capacitance C (w ) with the frequency for CNTs/AC hybrid and AC800. They illustrate until which frequency how much energy of ECs can be withdrawn. Generally, the larger pores are easier for electrolyte ions to penetrate at high frequencies, whereas some micropores can only be penetrated at very low frequencies. At low frequency (5 mHz), the capacitance of the electrode has reached 141 F/g for CNTs/AC hybrid and 90 F/g for AC800, which means that the ionic diffusion to the active surface is more efficient in the case of CNTs/AC hybrid at higher frequencies. Namely, CNTs/AC hybrid exhibits superior power discharge capability. The electrochemical impedance behaviors of CNTs/AC hybrid can be demonstrated by the simulative equivalent circuit based on transmission line model as presented in Fig. 5(e). The fitting data almost coincide with the experimental data. As a result, the equivalent circuit is reasonably simulated. The simulative equivalent circuit of CNTs/AC hybrid electrode is primarily composed of two parts, a depressed

semi loop and a transmission line element [29, 30]. The semi loop can be represented by a parallel Ri-CPEi ( CPEi, constant phase element ) circuit, associated with a series resistance Rs ( the electrolyte resistance ) [12, 28, 31-34]. CPEi represents the interfacial capacitance at the AC particles/electrolyte interface; αi traduces the non ideal behavior of this capacitance (0 < αi < 1). Depending on the Ri and CPEi values, the semi loop delays or blocks the capacitive behavior. The transmission line can be represented by a series Rsc(ω)-C(ω) circuit, traducing the double-layer capacitive behavior. Rsc(ω) is ion transport resistance of electrolyte within the porous passage, depending on the frequency; C(ω) is the capacitance also depending on the frequency. The total series resistance (R) of the ECs is given by R = Rs + Ri + R(ω). This value strongly depends on the impedance of ion diffusion inside the porous channels of AC. Consequently, suitable pore sizes can decrease the resistance of ion diffusion, which will greatly improve the capability of power density, while maintaining high conductivity and energy density.

4. Conclusions CNTs/AC hybrid with three dimensional CNT network was successfully synthesized by an in situ activation process. The capacitive behaviors of CNTs/AC hybrid was investigated based on CV and EIS. The conductivity of AC is enhanced due to CNTs embedding inside/winding AC particles. Besides the conductivity, it is also considered that CNTs play an important role in enhancing the specific surface

area as well as pore volume of AC in during of activated process. Cyclic voltammograms show that the superior capacitive behaviors of CNTs/AC hybrid are attributed to its superior conductivity and developed small mesopores (<5 nm). Such material structures are suitable for the transport of electrolyte ions. CNTs/AC hybrid demonstrates excellent high-frequency performances due to its higher surface area and developed small mesopores and can also provide much higher power density while still maintaining good energy density. In conclusion, CNTs/AC hybrid offers a great potential in ECs, particularly for applications where high power output and good high-frequency capacitive performances are required.

Acknowledgements: This work was supported by Ministry of Science and Technology of China (2011CB932604, 2014CB932402), National Science Foundation of China (Nos. 51221264, 51372253, U1401243), and the Key Research Program of Chinese Academy of Sciences (Grant No. KGZD-EW-T06), Science and Technology Program of Liaoning Education Office (No. 2008330).

Reference [1] B.E. Conway, Electrochemical Supercapacitors: Scientific Fundamentals and Technological Applications, Kluwer Academic/Plenum Publishers, 1999. [2] E. Frackowiak, F. Beguin, Carbon materials for the electrochemical storage of energy in capacitors, Carbon; 39 (2004) 937-950. [3] A.G. Pandolfo, A.F. Hollenkamp, Carbon properties and their role in supercapacitors, Journal of Power Sources, 157 (2006) 11-27. [4] Y.R. Liang, R.W. Fu, D.C. Wu, Reactive template-induced self-assembly to ordered mesoporous polymeric and carbonaceous materials, ACS Nano, 7(2013) 1748-1754 [5] Q.H. Liang, L. Ye, Z. H. Huang, Q. Xu, Y. Bai, F. Y. Kang, Q. H. Yang，A honeycomb-like porous carbon derived from pomelo peel for use in high-performance supercapacitors, Nanoscale, 6 (2014) 13831-13837 [6] D.C. Wu, Z.H. Li, M.J. Zhong, T. Kowalewski, K. Matyjaszewski, Templated synthesis of nitrogen-enriched nanoporous carbon materials from porogenic organic precursors prepared by ATRP, Angew. Chem. Int. Ed., 53 (2014) 3957-3960 [7] Z.H. Li, D.C. Wu, Y.R. Liang, R.W. Fu, K. Matyjaszewski, Synthesis of well-defined microporous carbons by molecular-scale templating with polyhedral oligomeric silsesquioxane moieties, J. Am. Chem. Soc., 136 (2014) 4805−4808 [8] Z.H. Li, D.C. Wu, Y.R. Liang, F. Xu, R.W. Fu, Facile fabrication of novel highly microporous carbons with superior size-selective adsorption and supercapacitance

properties, Nanoscale, 5 (2013)10824–10828 [9] T.T. Xie, W. Lv, W. Wei, Z.J. Li, B.H. Li, F.Y. Kang, Q.H. Yang, A unique carbon with a high specific surface area produced by the carbonization of agar in the presence of graphene, Chem. Commun., 49 (2013) 10427 [10] Z.H. Li, D.C. Wu, X. Huang, J.H. Ma, H. Liu, Y.R. Liang, R.W. Fu, K. Matyjaszewski, Fabrication of novel polymeric and carbonaceous nanoscale networks by the union of self-assembly and hypercrosslinking, Energy Environ. Sci., 7 (2014) 3006 [11] Z.H. Li, D. C. Wu, Y.R. Liang, F. Xu, R.W. Fu, Facile fabrication of novel highly

microporous

carbons

with

superior

size-selective

adsorption

and

supercapacitance properties, Nanoscale, 5 (2103) 10824 [12] W.G. Pell, B.E. Conway, N. Marincic, Analysis of non-uniform charge/discharge and rate effects in porous carbon capacitors containing sub-optimal electrolyte concentrations, Journal of Electroanalytical Chemistry, 491 (2000) 9-21. [13] M. Itagaki, S. Suzuki, I. Shitanda, K. Watanabe, H. Nakazawa, Impedance analysis on electric double layer capacitor with transmission line model, Journal of Power Sources, 164 (2007) 415-424. [14] J.G. Lee, J. Kim, S.H. Kim, Effects of microporosity on the specific capacitance of polyacrylonitrile-based activated carbon fiber, Journal of Power Sources, 160 (2006) 1495-1500. [15] H. Nishihara, H. Itoi, T. Kogure, P.X, Hou, H. Touhara, F. Okino, T. Kyotani;

Investigation of the ion storage/transfer behavior in an electrical double-layer capacitor by using ordered microporous carbons as model materials, Chemistry - A European Journal, 15 (2009) 5355-5363. [16] E. Frackowiak, K. Metenier, V. Bertagna, F. Beguin, Supercapacitor electrodes from multiwalled carbon nanotubes, Applied Physics Letter, 77(2000) 2421-2423. [17] S. Shiraishi, H. Kurihara, K. Okabe, D. Hulicova, A. Oya, Electric double layer capacitance of highly pure single-walled carbon nanotubes (HiPco™ Buckytubes™) in propylene carbonate electrolytes, Electrochemistry Communications, 4 (2002) 593-598. [18] C. Portet, P.L. Taberna, P. Simon, E. Flahaut, Influence of carbon nanotubes addition on carbon–carbon supercapacitor performances in organic electrolyte, Journal of Power Sources, 139 (2005) 371-378. [19] X. Geng, F. Li, D.W. Wang, H.M. Cheng, The electrochemical performance of a multi-wall carbon nanotube/activated carbon mixture as the electrode of electric double layer capacitors analyzed by electrochemical impedance, New Carbon Materials, 26 (2011) 305-312. [20] Y.Y. Fan, F. Li, H.M. Cheng, Preparation, morphology, and microstructure of diameter-controllable vapor-grown carbon nanofibers, Journal of Material Research, 13 (1998) 2342-2346. [21] X. Du, C.Y. Wang, M.M. Chen, S. Zhao, J. Wang, Effects of carbonization temperature on microstructure and electrochemical performances of phenolic

resin-based carbon spheres, Journal of Physics and Chemistry of Solids, 71(2010) 214-218. [22] D.W. Wang, F. Li, H.T. Fang, M. Liu, G.Q. Lu, H.M. Cheng, Effect of pore packing defects in 2-D ordered mesoporous carbons on ionic transport, Journal of Physical Chemistry B, 110 (2006) 8570-8575. [23] E. Raymundo-Pinero, K. Kierzek, J. Machnikowski, F. Beguin, Relationship between the nanoporous texture of activated carbons and their capacitance properties in different electrolytes, Carbon, 44 (2006) 2498-2507. [24] T.A. Centeno, O. Sereda, F. Stoeckli, Capacitance in carbon pores of 0.7 to 15 nm: a regular pattern, Physical Chemistry Chemical Physics, 13 (2011) 12403-12406. [25] W. Xing, S.Z. Qiao, R.G. Ding, F. Li, G.Q. Lu, Z.F. Yan, H.M. Cheng, Superior electric double layer capacitors using ordered mesoporous carbons, Carbon, 44 (2006) 216-224. [26] D.W. Wang, F. Li, M. Liu, G.Q. Lu, H.M. Cheng, 3D aperiodic hierarchical porous graphitic carbon material for high-rate electrochemical capacitive energy storage, Angewandte Chemie International Edition, 47 (2008) 373-376. [27] K.H. An, W.S. Kim, Y.S. Park, Y.C. Hoi, S.M. Lee, D.C. Chung, D.J. Bae, S.C. Lim, Y.H. Lee, Supercapacitors using single-walled carbon nanotubes electrodes, Advanced Materials, 13 (2001) 497-500. [28] 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,

Electrochimica Acta, 49 (2004) 905-912. [29] E. Barsoukov, J.R. Macdonald, Impedance spectroscopy: theory, experiment, and applications; John Wiley & Sons 2005 [30] C. Portet, P.L. Taberna, P. Simon, E. Flahaut, Influence of carbon nanotubes addition on carbon–carbon supercapacitor performances in organic electrolyte, Journal of Power Sources, 139(2005) 371-378. [32] R. Kötz, M. Carlen, Principles and applications of electrochemical capacitors, Electrochimica Acta, 45 (2000) 2483-2498. [32] H.K. Song, H.Y. Hwang, K.H. Lee, L.H. Dao, The effect of pore size distribution on the frequency dispersion of porous electrodes, Electrochimica Acta, 45 (2000) 2241-2257. [33] Y.R. Nian, H. Teng, Influence of surface oxides on the impedance behavior of carbon-based electrochemical capacitors, Journal of Electroanalytical Chemistry, 540 (2003) 119-127. [34] W.C. Chen, T.C. Wen, H. Teng, Polyaniline-deposited porous carbon electrode for supercapacitor, Electrochimica Acta, 48 (2003) 641-649.

Fig.1. SEM imagine of a) AC800 and b) CNTs/AC hybrid, c) TEM image of CNTs/AC hybrid, d) HRTEM image of CNTs/AC hybrid, e) SEM image of purified CNTs, f) Thermogravimetric analysis of purified CNTs, g) Nitrogen adsorption isotherm of purified CNTs, h) Schematic representation of the 3 dimension network texture of CNTs/AC hybrid. Fig.2. Nitrogen adsorption-desorption isotherm (a) and Pore size distribution of AC800 and CNTs/AC hybrid: (b) micropore, (c) mesopore. Fig. 3. CV for (a) CNTs/AC hybrid, (b) AC800 in 6 M KOH at different voltage scan rates, (c) CV behavior (sweep rate, 100mV/s), (d) Capacitance retention ratio. Fig. 4. Ragone plots of CNTs/AC hybrid and AC800 electrochemical capacitors. Fig.5. Nyquist plots (a), Bode plots (b) for CNTs/AC hybrid and AC800, imaginary capacitance (c) and real capacitance versus frequency for CNTs/AC hybrid and AC800.

Table 1 Pore characteristics of CNTs/activated carbon hybrid Sample

VT

Vmicro

Vmeso

pore diameter

SBET

(cm3/g)

(cm3/g)

(cm3/g)

(nm)

(m2/g)

AC800

0.93

0.48

0.45

0.88

1759

ACM800p5

1.36

0.71

0.65

2.2

2879

60

80

(f)

TG DSC

40 60

30

40

20

20

0.6% 10 0

0 0 200 400 600 800 1000 1200 0 Temperature / C

Volume Adsorbed(cm3/g STP)

120

(g)

CNTs

100 80 60 40 20 0 0.0

0.2 0.4 0.6 0.8 Relative pressure (P/P0)

50

1.0

Figure 1

DSC

Mass Loss/%

100

-10

Volume Adsorbed(cm3/g)

1000

(a)

800 600 400

AC800 CNTs/AC hybrid

200 0 0.0

0.2

0.4

0.6

0.8

1.0

Relative pressure(P/P0) 0.14

dD/dV (cm3/g.nm)

0.12

(b)

0.10 0.08 0.06 0.04 0.02 0.00 0.4

0.14 0.12

dD/dV (cm3/g.nm)

AC800 CNTs/AC hybrid

0.8 1.2 Pore Width (nm)

(c)

1.6

AC800 CNTs/AC hybrid

0.10 0.08 0.06 0.04 0.02 0.00 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Pore Width (nm)

Figure 2

5mV/g 100mV/g

I (A/g)

0.2

0.4 0.6 Voltage (V)

0.8

4 0

AC800 CNTs/AC hybrid

5 0 -5 0.2

0.4

0.6

Voltage (V)

0.8

1.0

1.0

0.0

Ratio of Retained Capacitance (%)

0.0

10

Figure 3

8

50mV/g 200mV/g

-8

(c)

0.0

5mV/g 100mV/g

(b)

-4

15

-10

12

50mV/g 200mV/g

I (A/g)

(a)

I (A/g)

25 20 15 10 5 0 -5 -10 -15 -20

0.2

0.4 0.6 Voltage (V)

0.8

1.0

100 80

(d)

60 40

AC800 CNTs/AC hybrid

20 0

0

50 100 150 200 Voltage Scan Rate (mV/s)

Power Density/W kg-1

10000

1000

100 1

Figure 4

AC800 CNTs/AC hybrid

10 -1 Energy Density/Wh kg

100

0.6 0.4

f1=125Hz f2=79Hz

Rs

0.2 0.0 0.0

0.2

Rf

1

Rf

2

0.4

0.6

(b)

AC800 CNTs/AC hybrid

80

Phase Angle (degree)

AC800 CNTs/AC hybrid

0.8

Z''/ohm

100

(a)

1.0

60 40 20 0

0.8

-2

10

Z'/ohm

-1

0

(c)

AC800 CNTs/AC hybrid

15

C'' (F/g)

Specific Capacitance (F/g)

150

20

10 5 0

0.01

0.1

1 10 Frequency (Hz)

100

1.0

(d)

50

0

1000

-2

10

-1

10

(e)

Z''/ohm

Fitting experimental

0.2 0.4 Z'/ohm

Figure 5

0

10

1

10

Frequency(Hz)

0.2

3

10

100

0.4

0.0 0.0

2

10

AC800 CNTs/AC hybrid

0.8 0.6

1

10 10 10 Frequency (Hz)

0.6

2

10

3

10