Enhanced capacitance in partially exfoliated multi-walled carbon nanotubes
Gongming Wang,Yichuan Ling,Fang Qian,Xunyu Yang,Xiao-Xia Liu,Yat Li
DOI: https://doi.org/10.1016/j.jpowsour.2011.02.019
IF: 9.2
2011-01-01
Journal of Power Sources
Abstract:We report for the first time the enhanced capacitance of multi-walled carbon nanotubes (MWCNTs) after exfoliation. Transmission electron microscopy studies confirmed that the MWCNTs were partially exfoliated with improved effective surface area. Carbon cloth electrode deposited with partially exfoliated carbon nanotubes (Ex-CNTs) yielded specific capacitances in a range of 130–165 F g −1 at charging/discharging current densities from 5 to 0.5 A g −1 , with coulombic efficiencies of ∼98%. The specific capacitance of Ex-CNTs was an order of magnitude higher than untreated MWCNTs, and comparable to graphene at all charging/discharging current densities we studied. The enhanced capacitance can be attributed to improved effective surface area and increased defect density of the exfoliated tubular structure. The results declared that Ex-CNTs are promising electrode materials for high-capacitance supercapacitors. Keywords Supercapacitor Exfoliated carbon nanotubes Capacitance Graphene Multi-walled carbon nanotubes 1 Introduction Supercapacitors have recently been extensively investigated as energy storage devices because of their large specific energy and power density [1–11] . In comparison to conventional capacitors, supercapacitors have several orders of magnitude higher in specific energy density. Supercapacitors can also store and deliver a large amount of charge in a short period of time, which allows them deliver higher power than rechargeable batteries [12–14] . The capacitance of supercapacitor is built upon the formation of electrochemical double layer at the electrode/electrolyte interface, and thus, the active interfacial surface area is a key parameter that determines the capacitance [15–17] . One-dimensional (1D) carbon nanotubes (CNTs) are emerging as new electrode materials for supercapacitors, due to their high conductivity, high surface area range and good corrosion resistance [3,14,18] . The effective surface area of a supercapacitor material is referred to the active layer that is accessible by electrolyte ions [19] , however, it is energetically unfavorable for solvated ions to diffuse into the inner tube of CNT [20] , in particularly for multi-walled carbon nanotubes (MWCNTs). In this regard, the specific capacitance of MWCNTs, especially at high current operation, can be limited by the slow diffusion of ions. Although single-walled carbon nanotubes (SWCNT) have higher specific capacitance than MWCNT due to larger surface area per unit mass [1–3] , the high cost of SWCNT limits their feasibility for device applications. Considerable efforts have been placed to improve the specific capacitance of MWCNTs. For example, the CNTs treated with sulfuric acid or nitric acid showed several times enhanced capacitance due to the increased surface defect density, which facilitated the diffusion of ions into the inner tube [21,22] . Enhanced specific capacitance has also been observed in MWCNT based pseudocapacitors, where MWCNTs were decorated with metal oxides [23–27] , and/or conducting polymer [28–32] . The enhanced capacitance in pseudocapacitors is due to faradaic type charge transfer instead of double layer electrostatic interaction. However, the charging/discharging cycling performance of pseudocapacitors can be limited by the irreversible redox reactions [33] , and the possible structural changes of polymer appear in subsequent cycles [34] . The objective of this work is to develop a novel and simple way to enhance the capacitance of MWCNTs by increasing the active surface area and defect density through exfoliation. 2 Experimental methods 2.1 Materials MWCNTs (95%) and graphite powder were purchased from Alfa Aesar. H 2 SO 4 (98%), HCl (37%), HNO 3 68–70%, H 2 O 2 (30%), hydrazine (64%) and KMnO 4 were obtained from Fisher Scientific Company. All chemicals were used as received, unless otherwise stated. 2.2 Preparation of Ex-CNTs Ex-CNTs were prepared by a reported method with modest modification [35] . 0.25 g MWCNTs were dispersed into 15 ml concentrated H 2 SO 4 and stirred for 1 h. Then 1.5 g KMnO 4 was added into the MWCNT solution at 50 °C. When all of the KMnO 4 were consumed, the reaction mixture was quenched by adding 40 ml deionized (DI) water and 3 ml 30% H 2 O 2 . The Ex-CNTs were extracted by centrifugation and washed with DI water. 2.3 Preparation of graphene Graphite oxide was synthesized by a modified Hummers and Offenman's method from graphite powder [36–38] . 0.5 g graphite powder, 20 ml H 2 SO 4 and 5 ml HNO 3 were mixed together in an ice bath. 3 g KMnO 4 was slowly added into the mixture, and heated the solution at 35 °C for 2 h. Then the solution was diluted by adding 140 ml DI water, followed by the slow addition of 3 ml H 2 O 2 . The aqueous graphite oxide solution was sonicated vigorously for 2 h to facilitate the separation of graphene oxide sheets. The graphene oxide precipitate was extracted, washed with DI water and then finally were reduced by hydrazine into graphene. 2.4 Preparation of electrodes A carbon cloth was dipped into Ex-CNT containing aqueous solution, and then dried in an oven at 80 °C. The loading of Ex-CNTs can be controlled by repeating the dipping–drying process. Ex-CNTs were then reduced in hydrazine vapor at 60 °C from 2 to 144 h. Finally, the carbon cloth was washed with water to remove the loosely attached Ex-CNTs, and dried at 80 °C. The weight of completely dried carbon cloth was measured before and after the deposition of Ex-CNTs. All the samples for electrochemical measurements have the loading of about 1 mg cm −2 . Graphene and MWCNT electrodes were prepared by the similar method. 2.5 Methylene blue adsorption 18 mg of methylene blue (in excess) was dissolved in 400 ml ethanol solution, and then mixed with 5 mg of the analyte (Ex-CNTs or MWCNTs). The solution mixture was agitated for 12 h. The suspended particles were removed by centrifugation. 0.5 ml of the supernatant solution was collected and further diluted by a factor of 8 for UV–vis absorption spectroscopy. The amount of adsorbed methylene blue can be calculated based on the absorbance of the supernatant, by comparing it to the standard dilution curve of the methylene blue solution. 2.6 Brumauer–Emmett–Teller (BET) measurement Nitrogen adsorption–desorption isotherms were measured with a Micromeritics Quadrasorb analyzer at 77 K. Before the measurements, the samples were degassed at 180 °C in vacuum for 6 h. The BET method was utilized to calculate the specific surface areas ( S BET ) using adsorption data in a relative pressure range from 0.04 to 0.2. By using the Barrett–Joyner–Halanda model, the pore volumes ( V t ) and pore size ( D ) distributions were derived from the adsorption branches of isotherms, and the total pore volumes were estimated from the adsorbed amount at a relative pressure P / P 0 of 0.992. The micropore volumes ( V micro ) and micropore surface areas ( S micro ) were calculated from the V − t plot method using the equation V m /cm 3 = 0.001547 × I , where I represents the y intercept in the V − t plots. The t values were calculated as a function of the relative pressure using the de Bore equation, t /Å = [13.99/(log( P 0 / P ) + 0.0340)] 1/2 . 2.7 Transmission electron microscopy (TEM) studies Nanomaterials (CNTs, Ex-CNTs and graphene) were dispersed in aqueous solution and then transferred onto Cu/lacey carbon TEM grids. TEM images were collected in a JEOL 1200 EX TEM operated with a 120 kV electron beam. 2.8 Electrochemical measurements Electrochemical studies were carried out in a three-electrode cell with nanomaterial coated carbon cloth as working electrode, Pt wire as counter electrode and an Ag/AgCl electrode as a reference. Electrolyte was a 3 M NaOH aqueous solution. Electrochemical properties were measured by an electrochemical workstation (CHI 660D). 3 Results and discussion Ex-CNTs were prepared by a reported method with modest modification [35] . TEM analysis was carried out on MWCNTs before and after the oxidation. TEM image showed that the untreated MWCNTs have tubular structures with smooth surface ( Fig. 1 a) . Significantly, the surface of tubular MWCNTs became blurry after 3-h oxidation treatment ( Fig. 1 b). Additional TEM images of Ex-CNTs are illustrated in Fig. S1, Supporting Information . These results suggested the MWCNTs were partially exfoliated ( Fig. 1 b, inset). Moreover, the MWCNTs showed substantially enhanced water solubility after exfoliation/oxidation ( Fig. 1 a, inset), as a result of the attachment of oxygen-containing groups ( e.g., –C–OH, –C O) on the tube surface. Importantly, it allows Ex-CNTs to be easily assembled on different hydrophilic substrates, including the flexible substrates such as carbon cloth or cotton cloth. In this work, we employed carbon cloth as electrode. In comparison to conventional metal electrode, carbon cloth is flexible, lightweight, low-cost and offers large surface area for active material loading. Ex-CNTs can be well entangled with the fibers of carbon cloth and expected to have low contact resistance. Ex-CNT-coated carbon cloth electrodes were characterized by cyclic voltammetry (CV) and galvanostatic charging/discharging measurements in a three-electrode electrochemical cell with 3 M NaOH solution as electrolyte. MWCNT-coated carbon cloth electrodes prepared under similar conditions were measured for comparison. Fig. 2 a compared the CVs of MWCNT and Ex-CNT samples collected at a scan rate of 5 mV s −1 . The capacitance of the blank carbon cloth is negligible ( Fig. S2, Supporting Information ). Because the specific capacitance is proportional to the area of CV curve, the results clearly showed that Ex-CNTs have substantially higher specific capacitance than MWCNTs. In addition, a small redox pair was observed on the CV curve of Ex-CNTs at −0.26 V ( Fig. 2 a), suggesting combined capacitance of pseudocapacitance and double layer capacitance for Ex-CNTs. The observed pseudocapacitance can be due to the presence of oxygen-containing groups on Ex-CNTs as a result of incomplete chemical reduction. All CV curves of Ex-CNTs collected at various scan rates up to 100 mV s −1 ( Fig. 2 b) displayed similar slightly distorted rectangular shapes, suggesting a high power density for Ex-CNTs. These data indicate efficient charge accumulation and ion diffusion at the interface between Ex-CNT electrode and electrolyte, which is especially critical for device operation at high charging/discharging current density. Ex-CNT based electrode could potentially address capacitance degradation at high charging/discharging current density that has been a limitation for commercialized activated carbon supercapacitors [39] . The specific capacitances of Ex-CNT- and MWCNT-coated electrodes were characterized by galvanostatic charging/discharging measurements. Galvanostatic cycling data obtained from the Ex-CNT at a charging/discharging current density of 1 A g −1 showed linear discharging curves in the potential range with a nearly constant slope, indicating excellent capacitive behavior of Ex-CNTs ( Fig. 3 a) [40,41] . The time required for a single charging–discharging cycle for Ex-CNT and MWCNT electrodes were compared in Fig. 3 b. It takes substantially longer time to charge Ex-CNT electrode than that of MWCNT at the same current density of 1 A g −1 , again confirming larger capacitance of Ex-CNT ( Fig. 3 b). The specific capacitances of the electrodes were calculated using the following equation: specific capacitance = i × t V × m = i S × m where S is the slope of discharging curve (V s −1 ), i is charging/discharging current, m is weight of the active material (the weight of carbon cloth is not included because of its negligible capacitance) and therefore i / m is the charging/discharging current per unit mass (A g −1 ). The calculated specific capacitances of Ex-CNT and MWCNT at the charging/discharging current density of 1 A g −1 were 154.4 F g −1 and 11.8 F g −1 , respectively. The specific capacitance of MWCNT is comparable to the best literature reported values [18,42,43] . Significantly, Ex-CNTs showed an order of magnitude enhanced capacitance compared to MWCNTs. The improved capacitance can be attributed to the improved effective surface area per unit mass of Ex-CNTs as well as the increased defect density. To confirm the increase of effective surface area in Ex-CNTs compared to MWCNTs, both methylene blue as well as BET adsorption measurements were performed on the samples. Methylene blue adsorption has been used as an indicator of graphitic material surface areas in previous studies, in which each milligram of adsorbed methylene blue represents 2.45 m 2 of surface area [44–46] . A fixed amount of analyte (Ex-CNTs or MWCNTs) was mixed with excess methylene blue solution of known concentration. By determining methylene blue concentration after the reaction using UV–vis spectroscopy (see Section 2 ), the amount of adsorbed methylene blue and thereby the surface area of analyte can be calculated. The surface area of MWCNT and Ex-CNT were calculated to be ∼280 and 480 m 2 g −1 , respectively, suggesting the effective surface area of MWCNTs was enhanced after oxidation. However, the fact that the surface topology and chemistry of material may influence the area occupied by each methylene blue molecule and cause possible inaccuracy in determining material surface area, BET surface adsorption measurements were also performed on the samples for comparison (see Section 2 ). The BET adsorption data are summarized in Table 1 . Significantly, Ex-CNTs sample showed more than two times enhancement in specific surface area (391 m 2 g −1 ) compared to untreated MWCNTs (172 m 2 g −1 ), which is consistent with the methylene blue adsorption results. In addition, Ex-CNTs have larger pore volume that is expected to be favorable for solvated ion diffusion. Taken together, these results clearly confirmed the surface area enhancement in MWCNT after oxidation, and served as a strong evidence for successful exfoliation of MWCNT. Moreover, the increased defect density during exfoliation process could also facilitate the diffusion of solvated ions into inner tube, as reported previously [47] . The defect densities of MWCNT and Ex-CNT samples were characterized by Raman spectroscopy ( Fig. S3, Supporting Information ), which is a powerful approach to investigating the structural and electronic properties of carbon-based materials. It has been reported that D band of carbon material is defect dependent, and the intensity ratio of G band to D band is a good indication for the density of defects present in the materials (the higher the ratio, the less the defect density) [48,49] . Raman data ( Fig. S3, Supporting Information ) showed that the ratio decreases from 0.98 to 0.87 after exfoliation, confirming the increased defect density in Ex-CNT. The capacitances of Ex-CNT based electrodes were measured at different charging/discharging current densities from 0.5 to 5 A g −1 ( Fig. 3 c). The Ex-CNT exhibited a maximal specific capacitance of 165 F g −1 at charging current density of 0.5 A g −1 , and slightly reduced to 133 F g −1 at a larger charging current density of 5 A g −1 . The large specific capacitances at 5 A g −1 together with the observed nearly rectangular CV curves at a high scan rate of 100 mV s −1 ( Fig. 2 b) confirmed the excellent ion accessibility between Ex-CNT and electrolyte. To determine the reversibility of charging–discharging processes, we measured the coulombic efficiency (CE) of the electrodes, which is defined by the equation [7] : C E = t d i s c h a r g i n g t c h a r g i n g where t discharging and t charging are discharging and charging time. The CE of Ex-CNTs is above 98% in the whole current density range from 0.5 to 5 A g −1 ( Fig. 3 c), indicating the charging–discharging processes were highly reversible. Electrical conductivity of electrode material is another critical parameter that determines the capacitive performance. As-prepared Ex-CNTs have low electrical conductivity due to oxidization during the exfoliation process [35] . Therefore, the Ex-CNTs were further reduced in hydrazine vapor to increase the conductivity prior to electrochemical measurements. Their specific capacitances were studied as a function of reduction time. As shown in Fig. 4 a , the specific capacitance increases drastically with reduction time because the improved electrical conductivity as a result of decrease of oxygen-containing groups on Ex-CNT. The prolonged reduction process only slightly improved the specific capacitance, indicating most of the oxidized Ex-CNTs have been reduced in the first 24 h. By optimizing the reduction time, Ex-CNT electrode yielded a maximal specific capacitance of 160 F g −1 at the charging/discharging current density of 1 A g −1 , which is comparable to the best reported values of carbon-based supercapacitors at the same current density [1,2,11] . Galvanostatic cycling studies on Ex-CNT electrodes at 1 A g −1 showed that the specific capacitance dropped quickly from 147 F g −1 to 138 F g −1 in the first 100 charging/discharging cycles, and then became stable with minor reduction of specific capacitance (138–130 F g −1 ) in the following 1900 cycles ( Fig. 4 b). Overall, the Ex-CNT electrode showed good cycling stability, with a ∼11% reduction of specific capacitance after 2000 charging/discharging cycles. These results revealed Ex-CNTs are efficient and stable electrode material for supercapacitors. Finally, the capacitive performance of Ex-CNTs was compared to graphene, which has been used as electrodes for different energy storage devices [4,5,50] . TEM image shows that the as-prepared graphene oxide has a very thin 2D structure ( Fig. S4, Supporting Information ). These non-conducting graphene oxide materials were coated on carbon cloth and then reduced in hydrazine vapor into graphene for CV and galvanostatic cycling studies. The CV of graphene-coated electrode collected at different scan rates from 5 to 100 mV s −1 showed highly distorted rectangular shapes at high scan rates ( Fig. S5, Supporting Information ), which suggested the diffusion of electrolyte ions in graphene is less efficient compared to that in Ex-CNTs. The limited ion diffusion could be ascribed to the aggregation of graphene sheets during the reduction process and storage, as a result of the strong face-to-face attraction between graphene layers. The graphene sheets located inside the aggregate may not be accessible by the solvated ions that could significantly reduce the effective surface area [4,38,51] . BET adsorption data obtained from graphene and Ex-CNT samples ( Table 1 ) shows that the surface area per unit mass of Ex-CNTs is slightly smaller than that of graphene, as expected. On the other hand, Ex-CNTs have larger total pore volume that is more favorable for solvated ion diffusion compared to graphene, which is in agree with the CV data. Fig. 5 compared the capacitive performance of MWCNT, graphene and Ex-CNT. Based on the galvanostatic measurements, the maximal specific capacitance of graphene-coated electrode was calculated to be about 127 F g −1 at charging/discharging current density of 1 A g −1 , which is comparable to the best literature reported data for graphene based supercapacitors [4,5] . Significantly, Ex-CNTs has comparable, if not better, specific capacitance with graphene sample at all charging/discharging current densities we studied. 4 Conclusions We have successfully synthesized the partially exfoliated MWCNTs and studied the capacitive performance of these Ex-CNTs-coated electrodes. They exhibited a maximal specific capacitance of 165 F g −1 at a charging/discharging current density of 0.5 A g −1 , and a coulombic efficiency of ∼98%. Significantly, the specific capacitances of Ex-CNTs are an order of magnitude higher than untreated MWCNT, and comparable to graphene at all charging/discharging current densities we studied. The enhanced capacitance is attributed to the increased effective surface area and defect density in the partially exfoliated tubular structure. Ex-CNTs hold great promise as electrode material for supercapacitors, and can possibly be used in other applications such as lithium ion batteries. Acknowledgements Y.L. acknowledges the support of this work by UCSC new faculty start up fund and NSF CAREER award ( DMR-0847786 ). X.X.L. thanks the financial support from National Natural Science Foundation of China (project number 50973013). Appendix A Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jpowsour.2011.02.019 . Appendix A Supplementary data References [1] L. Hu J.W. Choi Y. Yang S. Jeong F.L. Mantia L. Cui Y. Cui Proc. Natl. Acad. Sci. U.S.A. 106 2009 21490 21494 [2] L. Hu M. Pasta F.L. Mantia L. Cui S. Jeong H.D. Deshazer J.W. Choi S.M. Han Y. Cui Nano Lett. 10 2010 708 714 [3] M. Kaempgen C.K. Chan J. Ma Y. Cui G. Gruner Nano Lett. 9 2009 1872 1876 [4] M.D. Stoller S.J. Park Y. Zhu J. An R.S. Ruoff Nano Lett. 8 2008 3498 3502 [5] Y. Wang Z. Shi Y. Huang Y. Ma C. Wang M. Chen Y. Chen J. Phys. Chem. C 113 2009 13103 13107 [6] C. Yu C. Masarapu J. Rong B. Wei H. Jiang Adv. Mater. 21 2009 4793 4797 [7] L. Chen L.J. Sun F. Luan Y. Liang Y. Li X.X. Liu J. Power Sources 145 2010 3742 3747 [8] H. Wang H.S. Casalongue Y. Liang H. Dai J. Am. Chem. Soc. 132 2010 7472 7477 [9] C. Liu Z. Yu D. Neff A. Zhamu B.Z. Jang Nano Lett. 10 2010 4863 4868 [10] M. Wu M. Wang Chem. Commun. 46 2010 6968 6970 [11] K. Zhang L.L. Zhang X.S. Zhao J. Wu Chem. Mater. 22 2010 1392 1401 [12] M. Winter R.J. Brodd Chem. Rev. 104 2004 4245 4270 [13] A. Burke J. Power Sources 91 2000 37 50 [14] A.G. Pandolfo A.F. Hollenkamp J. Power Sources 157 2006 11 27 [15] J. Chmiola G. Yushin Y. Gogotsi C. Portet P. Simon P.L. Taberna Science 313 2006 1760 1763 [16] E. Frackowiak F. Beguin Carbon 39 2001 937 950 [17] J. Gamby P.L. Taberna P. Simon J.F. Fauvarques M. Chesneau J. Power Sources 101 2001 109 116 [18] S. Talapatra S. Kar S.K. Pal R. Vajtai L. Ci P. Victor M.M. Shaijumon S. Kaur O. Nalamasu P.M. Ajayan Nat. Nanotech. 1 2006 112 116 [19] A.L.M. Reddy S. Ramaprabhu J. Phys. Chem. C 111 2007 7727 7734 [20] H. Shimoda B. Gao X.P. Tang A. Kleinhammes L. Fleming Y. Wu O. Zhou Phys. Rev. Lett. 88 2002 015502 [21] C.G. Liu H.T. Fang F. Li M. Liu H.M. Chen J. Power Sources 160 2006 758 761 [22] K. Jurewicz K. Babel R. Pietrzak S. Delpeux H. Wachowska Carbon 44 2006 2368 2375 [23] R.R. Bi X.L. Wu F.F. Cao L.Y. Jiang Y.G. Guo L.J. Wang J. Phys. Chem. C 114 2010 2448 2451 [24] X. Xie L. Gao Carbon 45 2007 2365 2373 [25] A.K. Mishra S. Ramaprabhu J. Phys. Chem. C 114 2010 2583 2590 [26] X. Du C. Wang M. Chen Y. Jiao J. Wang J. Phys. Chem. C 113 2009 2643 2646 [27] A. Reddy M.M. Shaijumon S.R. Gowda P.M. Ajayan J. Phys. Chem. C 114 2010 658 663 [28] M. Wu L. Zhang D. Wang J. Gao S. Zhang Nanotechnology 18 2007 385603 [29] Y. Hou Y. Cheng T. Hobson J. Liu Nano Lett. 10 2010 2727 2733 [30] Y. Fang J. Liu D.J. Yu J.P. Wicksted K. Kalkan C.O. Topal B.N. Flanders J. Wu J. Li J. Power Sources 195 2010 674 679 [31] Q. Li J. Liu J. Zou A. Chunder Y. Chen L. Zhai J. Power Sources 196 2010 565 572 [32] A. Snook P. Kao A.S. Best J. Power Sources 196 2010 1 12 [33] E. Raymundo-Pinero V. Khomenko E. Frackowiak F. Beguin J. Electrochem. Soc. 152 2005 A229 A235 [34] L. Li H. Song Q. Zhang J. Yao X. Chen J. Power Sources 187 2009 268 274 [35] D.V. Kosynkin A.L. Higginbotham A. Sinitskii J.R. Lomeda A. Dimiev B.K. Price J.M. Tour Nature 458 2009 872 876 [36] W.S. Hummers R.E. Offeman J. Am. Chem. Soc. 80 1958 1339 [37] S. Gilje S. Han M. Wang K.L. Wang R.B. Kaner Nano Lett. 7 2007 3394 3398 [38] L.J. Cote F. Kim J. Huang J. Am. Chem. Soc. 131 2009 1043 1049 [39] D. Hulicova J. Yamashita Y. Sooneda H. Hatori M. Kodama Chem. Mater. 17 2005 1241 1247 [40] D.Y. Qu J. Power Sources 109 2002 403 411 [41] A.J. Burke J. Power Sources 91 2000 37 50 [42] A.K. Cuentas-Gallegos R. Martinez-Rosales M. Baibarac P. Gomez-Romero M.E. Rincon Electrochem. Commun. 9 2007 2088 2092 [43] G.P. Pandey S.A. Hashmi Y. Kumar J. Electrochem. Soc. 157 2010 A105 114 [44] C.S. Hannes J. Li M.J. McAlister S. Hiroaki M. Herrera-Alonso D.H. Adamson R.K. Prud’homme R. Car D.A. Saville I.A. Aksay J. Phys. Chem. B 110 2006 8535 8539 [45] M.J. McAlister J. Li D.H. Adamson H.C. Schniepp A.A. Abdala J. Liu M. Herrera-Alonso D.L. Milius R. Car R.K. Prud’homme I.A. Aksay Chem. Mater. 19 2007 4396 4404 [46] H. Kim A.A. Abdala C.W. Macosko Macromolecules 43 2010 6515 6530 [47] C. Liu H. Cao P.Y. Li H. Xu Y. Zhang Carbon 44 2006 2919 2924 [48] L. Jiao L. Zhang X. Wang G. Diankov H. Dai Nature 458 2009 877 880 [49] S. Costa B. Scheibe M. Rummeli E. Borowiak-Palen Phys. Status Solidi. 246 2009 2717 2720 [50] D. Wang F. Li J. Zhao W. Ren Z. Chen J. Tan Z. Wu I. Gentle G.Q. Lu H. Cheng ACS Nano 3 2009 1745 1752 [51] S. Stankovich D.A. Dikin R.D. Piner K.A. Kohlhass A. Kleinhammes Y. Jia Y. Wu S.T. Nguyen R.S. Ruoff Carbon 45 2007 1558 1565