Effect of catalytic graphitization on the electrochemical behavior of wood derived carbons for use in supercapacitors

Effect of catalytic graphitization on the electrochemical behavior of 1 wood derived carbons for use in supercapacitors 2 A. Gutiérrez-Pardo, J. Ramírez-Rico*, R. Cabezas-Rodríguez, J. Martínez-Fernández 3 Departamento Física de la Materia Condensada - ICMS (Universidad de Sevilla-CSIC), Avda. 4 Reina Mercedes S/N, 41012 Seville, Spain. 5 Abstract 6 Porous graphitic carbons were successfully obtained from wood precursors through 7 pyrolysis using a transition metal as catalyst. Once the catalyst is removed, the resulting 8 material mimics the microstructure of the wood and presents high surface area, open and 9 interconnected porosity and large pore volume, high crystallinity and good electrical 10 conductivity, making these carbons interesting for electrochemical devices. Carbons 11 obtained were studied as electrodes for supercapacitors in half cell experiments, obtaining 12 high capacitance values in a basic media (up to 133 F·g-1 at current densities of 20 mA·g-1 13 and 35 F·g-1 at current densities of 1 A·g-1). Long-cycling experiments showed excellent 14 stability of the electrodes with no reduction of the initial capacitance values after 1000 15 cycles in voltammetry. 16 Keywords: supercapacitors; carbon materials; pyrolysis; graphite; electrochemistry 17 1 Introduction 18 Electric double-layer capacitors (EDLCs) have received considerable attention as energy 19 storage/conversion devices because they exhibit high power density, long cycling life, a 20 high charge/discharge capability and are environmentally friendly [1-5]. Supercapacitors 21 are similar in design and manufacture to batteries, but in contrast to them, where the cycle 22 * corresponding author: [[email protected], Tel. +34 954 550963] life is limited because of the repeated contraction and expansion of the electrode upon 23 cycling, EDLC lifetime is in principle infinite, as it operates solely on electrostatic surface 24 charge accumulation, without any noticeable performance change [6, 7]. Thus, EDLCs are 25 well suited as backup sources due to their low cost and maintenance-free operation. 26 Supercapacitors have gained considerable interest in different fields such as 27 transportation (electric vehicles), industrial (uninterruptible power sources), digital 28 communications system (cell phones), and others [8]. 29 The capacitance of a supercapacitor is largely dependent on the characteristics of the 30 electrode material. The science and technology of supercapacitors has been reviewed for a 31 number of electrode materials, including carbon, mixed metal oxides, and conducting 32 polymers [9]. More work has been done using porous carbons than with any other 33 material and most of the commercially available devices use carbon electrodes. They have 34 desirable physical and chemical properties, including variety of available forms (powders, 35 fibers, aerogels, composites, monoliths, nanotubes, etc.), ease of processing, controllable 36 porosity, stability in different solutions (from strongly acidic to basic) and ability to 37 perform in a wide range of temperatures [9, 10]. The factors that dictate the selection of 38 carbon materials for electrodes include high surface area and high electrical conductivity, 39 but also interconnected pore structure, controlled pore size matching to that of the 40 electrolyte ions, good wettability towards the electrolyte and the presence of 41 electrochemically stable surface functionalities [11, 12]. The ELDC capacitance arises from 42 the charge separation at the electrode and electrolyte interface, and porous carbonaceous 43 structures can increase the effective contact area between the electrode and electrolyte. 44 The storage charge can be limited in some conventional carbons because they contain an 45 abundant proportion of micropores (pore size < 2nm) that are not easily accessible to ions 46 of the electrolyte. Carbons containing interconnected mesopores (2-50 nm) are desirable 47 for ELDCs electrodes, but meso/micro porosity must exist because if only mesopores are 48 present in the material the volumetric capacitance decreases due to a lower density of the 49 carbon [13]. 50 Most porous carbon materials have the disadvantage of needing binders or additives 51 which can block part of the porosity of the carbon. The possibility of using activated 52 carbon materials without a binding substance gives an additional benefit from the 53 construction point of view. Monolithic porous carbons with large surface area have long 54 been used as electrode in supercapacitors and demonstrated to be ideal candidates due to 55 their relatively low cost, good electronic conductivity, high surface area and availability [1, 56 14, 15]. 57 The design of novel structures by mimicking the cellular tissue anatomy of natural 58 biostructures has become a matter of increasing interest [16-18]. Recent examples in the 59 field of energy storage include the use of biomass waste such as banana peel [19] or 60 naturally abundant algae [20] as templates for the synthesis of functionalized carbon 61 structures. In the past, we have employed these porous carbon materials that replicate the 62 microstructure of the natural wood as scaffolds towards the synthesis of carbide ceramics 63 [21, 22]. Carbon obtained by pyrolysis of wood presents structural disorder that prevents 64 its graphitization with a subsequent heat treatment; for instance, Pappacena et al. showed 65 that wood derived pyrolytic carbon remains essentially amorphous after exposure to 66 temperatures up to 2400ºC in vacuum [23]. For this reason, wood derived carbon is 67 considered hard or non-graphitizing. However, graphitization can be induced by using a 68 catalyst during pyrolysis, usually a transition metal [24, 25], and this process has been 69 shown to enhance capacitance in carbon materials by both increasing conductivity as well 70 as generating an abundance of mesopores. Recent examples of the use of catalysts to 71 induce graphitization of pyrolytic carbon in the field of energy storage applications include 72 nickel on carbon microspheres [26], MnO2 catalyzed carbon microspheres [27] or polymer 73 derived (Fe, Co, Ni)-catalyzed carbon monoliths [28]. 74 Here we explore the application of graphitized wood-derived carbons in energy storage 75 applications. This study proposes the use of Fe as catalyst to treat the natural precursor 76 (wood in our case) and obtain the formation of graphite during the pyrolysis process. 77 Afterwards the catalyst is removed to obtain a monolithic, macroporous, partially 78 graphitized carbon material that combines the properties of pyrolysis derived carbons, 79 such as open and interconnected porosity and large pore volume, with the properties of 80 graphite such as high crystallinity and good thermal and electrical conductivity. This paper 81 analyzes the microstructure and the electrochemical behavior of partially graphitized 82 biocarbons processed at different temperatures, and compares them to wood derived 83 pyrolytic carbons obtained without the use of a graphitization catalyst. 84 2 Experimental 85 2.1 Materials 86 Wood-derived carbons were produced from beech (Fagus silvatica) precursors. Pieces of 87 wood were cut into blocks with dimensions ca. (15 x 15 x 75) mm3 and dried in an oven at 88 100 °C to release adsorbed water from the pores. Iron was chosen as the graphitization 89 catalyst which was prepared into a 3M FeCl3 solution in isopropanol and used to 90 impregnate the wood samples in vacuum for 2 hours to ensure complete filling of the 91 pores. Subsequently, the wood samples were heated until complete drying, evaluated by 92 periodic weight measurements, and stored in a stove until further use. 93 The pyrolysis process took place in a tube furnace with controlled and constant flow of 94 nitrogen. Pyrolysis was carried out with a ramp rate of 1ºC/min to 500ºC; 5ºC/min to 95 maximum temperature (1000 to 1600ºC), soak time of 30 minutes and 5ºC/min for 96 cooling down to room temperature. This temperature program was chosen to ensure that 97 crack-free carbon monoliths could be obtained. 98 After pyrolysis, the residual iron was removed by stirring in concentrated HNO3 (69%, 99 Panreac). The resulting biocarbon monoliths were rinsed and stirred in acetone until 100 neutral pH, and then dried. Samples without catalyst were made to evaluate the effect of 101 graphitization on the electrochemical behavior under similar conditions on 102 treated/untreated carbons. 103 2.2 Characterization 104 For a microstructural evaluation of the samples, scanning (SEM, JEOL 6460-LV) and 105 transmission electron microscopy (TEM, Philips CM-200) were used. To prepare samples 106 for TEM, carbon samples were ground, dispersed in ethanol and sonicated for 30 minutes. 107 A drop of this dispersion was deposited on a copper grid with a reticulated amorphous 108 carbon film and allowed to dry. 109 To determine the thermodynamic characteristics of the graphitization process, 110 thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) were 111 carried out using a dual beam thermobalance (Thermal Advantage SDT-600) which 112 monitored the weight loss and the heat flow in-situ during the pyrolysis, both for treated 113 and untreated wood precursors. Measurements were performed in nitrogen flow using a 114 20 °C/min heating rate up to 1400ºC. 115 Residual amounts of hydrogen and nitrogen were determined in a fast combustion furnace 116 using pure oxygen (LECO-CHNS 932), while residual Fe content was measured by ICP-OES 117 (Horiba Ultima 2). Oxygen content was determined by x-ray fluorescence spectroscopy 118 (PANalytical AXIOS). 119 Samples were studied to determine the effect of the catalytic treatment and the maximum 120 pyrolysis temperature on the degree of graphitization. Crystallinity was determined by X-121 ray diffraction in a powder diffractometer (A25 D8 Advance, Bruker), and the degree of 122 structural disorder in the carbon structures was evaluated by Raman spectroscopy 123 (LabRam Jobin Yvon provided with a confocal microscope) with an excitation wavelength 124 of 532 nm, in the spectral region between 1000 and 1800 cm-1 which yields the most 125 interesting structural information on carbonaceous materials [29, 30]. 126 The specific surface area of the studied carbon materials was determined by the BET 127 method based on nitrogen adsorption isotherms using a Micromeritics ASAP 2420 128 Analyzer. Samples previously cut into small pieces to feed into a glass bulb were outgassed 129 at 400ºC for 10 hours and placed in a Dewar vessel to analyze the adsorbed and desorbed 130 amount in the material at 77K in the relative pressure range of P/P0 = 0-1. 131 Bulk electrical resistivity was measured at room temperature using the four-terminal 132 sensing method. Carbons were previously cut into bars with dimensions of about (50 x 3 x 133 3) mm3, and current intensities from 1 to 100 mA were applied in the extremes of the bars, 134 using a constant intensity power supply (Iso-Tech IPS 3202), measuring the voltage 135 between two fixed points at 15 mm span. The electrical resistance was obtained according 136 to Ohm´s Law (V = IR), and the resistivity was calculated from the resistance using the 137 measured specimen dimensions. 138 2.3 Evaluation of the electrochemical properties. 139 Carbon electrodes were prepared as nominal 1 cm2 slabs with a 1 mm thickness and an 140 approximate mass of 50 mg; electrical connections were performed using Cu wire and 141 silver paste for contacts. The electrochemical measurements of the electrode materials 142 were carried out using a Solartron 1287A potentiostat/galvanostat with a 1260A 143 frequency response analyzer, in a three-electrode setup: wood derived carbon samples 144 were used as working electrodes, an Ag/AgCl electrode was used as a reference and 145 platinum as the counter electrode. The electrolyte in all cases was 1M KOH. Cyclic 146 voltammetry (CV) experiments were performed using scan rates ranging from 1 to 30 mV · 147 s-1, at potentials between -1.0 and 0.0 V. Galvanostatic charge/discharge experiments were 148 performed at current densities ranging from 20 to 2000 mA·g-1 at a potential window of -149 0.8 V. Electrochemical impedance spectroscopy measurements were performed at 10 mV 150 amplitude and 0 V open circuit potential at frequencies in the range 10-3-105 Hz. The cyclic 151 stability of the working electrode was also evaluated for 1000 cycles for graphitized 152 carbons in cyclic voltammetry, at a scan rate of 10 mV·s-1. 153 3 Results and discussion 154 3.1 Structural characterization. 155 After impregnation of wood in a FeCl3 solution for 2 hours, a mean dry weight increases of 156 7 ± 2 % was determined, while geometrical volume remained essentially unchanged. After 157 pyrolysis the precursor retained its original shape, while changes in weight and volume 158 were independent of the pyrolysis temperature. The average weight reduction in the 159 samples was 66 ± 2% and the volume reduction was 68 ± 6%. 160 Microstructural characterization of the obtained biocarbons was performed using SEM 161 and TEM. Figure 1 shows SEM images for treated carbons pyrolyzed at 1000, 1300 and 162 1600ºC before and after Fe removal, both using secondary electrons to show the 163 microstructure of the carbon scaffold as well as backscattered electrons to reveal the size 164 and morphology of the Fe particles. The microstructure of the carbon scaffold replicates 165 that of the original wood, consisting of interconnected and aligned channels of a bimodal 166 size distribution. After etching with HNO3, a soluble salt of iron that eliminates the catalyst 167 is formed and Fe particles are removed from the material. A carbon skeleton which mimics 168 the microstructure of the natural precursor with open and interconnected porosity is thus 169 obtained. Residual Fe content was evaluated to determine the extent of Fe removal during 170 etching, and results are included in Table 1. It was found that residual Fe accounted for 171 less than 0.5 % weight in the case of catalyzed samples and was almost negligible 172 (attributable to oligo-elements present in the wood precursor) in the case of untreated 173 samples. Other residual elements included hydrogen, nitrogen and oxygen (Table 1). 174 Size and morphology of Fe particles in the carbon scaffold was quantified from SEM 175 measurements, consisting in globular agglomerates in the range of ~1-2 µm. No significant 176 changes in these particles were observed when the peak pyrolysis temperature exceeded 177 the melting point of iron. 178 Figure 2(a) shows a micrograph obtained by transmission electron microscopy for a 179 treated sample pyrolyzed at 1600 °C, showing Fe particles encapsulated in hollow carbon 180 spheres. The graphitization takes place probably by deposition of turbostratic layers that 181 precipitate from carbon dissolved in the Fe particles. When Fe is removed, hollow and 182 ordered carbon nanostructures remain in the material, as can be seen in Figure 2(b) and 183 will be further discussed below. Highly ordered regions with a plate like morphology and 184 consisting of stacked graphite layers were also detected in samples pyrolyzed at 1600ºC; 185 in Figure 2(c) these plate-like graphitic regions are shown with the associated convergent 186 beam pattern showing individual diffraction spots. Additional evidence of this ordering is 187 the Moiré fringes visible in the micrograph. Finally and for comparison, a TEM micrograph 188 obtained for a sample pyrolyzed without catalyst is shown in Figure 2(d), where none of 189 the previously described features are visible. In contrast with previous works using Ni as 190 a catalyst [25, 31, 32], the mechanism is not dependent of the pyrolysis temperature in the 191 range studied. In the case of wood pyrolyzed at the same temperatures without catalytic 192 treatment, carbon is essentially amorphous, as it was confirmed in the images and SAED 193 patterns in Figure 2(d). 194 Thermogravimetric analysis (Figure 3) shows an initial weight loss due to the adsorbed 195 water in the carbon up to 150 °C. Decomposition of polysaccharides which are the main 196 components of wood occurs between 200ºC and 400ºC: the hemicellulose is broken up 197 first, followed by cellulose and lignin [33, 34]. In untreated samples no more steps are 198 observed, but in treated samples a third step at about 700ºC is identified. The heat flow 199 during the pyrolysis shows an endothermic curve with a pronounced endothermic peak at 200 this temperature for the treated sample which does not appear without Fe. This peak is 201 related with the onset of the graphitization process and was confirmed by the presence of 202 a graphite (0002) peak in the XRD pattern of a sample pyrolyzed at 850ºC (not shown). As 203 will be discussed below, this peak is attributed to the formation of pseudo-eutectic liquid 204 droplets in the Fe-C system responsible for the formation of graphitic structures and 205 occurs at temperatures lower than the bulk eutectic temperatures due to size effects [35]. 206 Figure 4 shows XRD patterns of samples studied. Treated carbons present a peak at 2θ = 207 26.6°, corresponding to (0002) crystal planes of graphite [pdf 26-1079] which does not 208 appear in untreated samples which exhibit an essentially amorphous pattern. When the 209 pyrolysis temperature increases, the relative intensity of this graphite peak increases 210 while the peak width decreases. Therefore, the sample pyrolyzed at 1000 °C shows a small 211 and broad peak which indicates the onset of the graphitization, while at 1600 °C the peak 212 is much taller and narrow indicating a higher degree of crystallinity. 213 In treated samples, we analyzed the specimens before the removal of Fe, obtaining one 214 peak of great intensity at 2θ = 44.8°, and other characteristic peaks [pdf 06-0696] which 215 disappear after the acid attack of samples with concentrated HNO3, confirming the almost 216 complete removal of remaining Fe. 217 Raman spectra of samples studied are shown in Figure 5. The G band, characteristic for 218 graphitic materials, is located at 1580 cm-1 and represents the vibration of the ideal 219 graphite (E2g symmetry). Other bands indicate a disordered graphitic lattice. The D1 band 220 (1350 cm-1), the most prominent, represents the vibration of graphene layer edges (A1g 221 symmetry), and the D2 band (1620 cm-1) the vibration of surface graphene layers (E2g 222 symmetry). D3 band (1500 cm-1) and D4 band (1150 cm-1) represent the vibration of 223 amorphous regions and local disorder by the presence of ionic impurities [29, 30]. 224 The relationship between the intensity of bands G and D1 increases in treated samples 225 with the increasing of the pyrolysis temperature and the width to half height (FWHM) 226 Figure 8. Electrochemical testing results. Left: galvanostatic charge/discharge cycles at 20 376 mA/g (top) and 200 mA/g (bottom) for Fe-treated carbon samples pyrolyzed at different 377 peak temperatures. Right: EIS of Fe-treated carbon samples at different pyrolysis 378 temperatures. Equivalent series resistance has been substracted to all spectra for ease of 379 comparison. 380 Figure 9. (a-d) CV curves of carbons treated with Fe, at different scan rates, from 5 to 30 381 mV/s, made at peak pyrolysis temperatures of (a) 1000, (b) 1300 and (c) 1600ºC. (d) CV 382 curves of carbons treated and untreated with Fe, at a scan rate of 10 mV/s. e) Ragone-type 383 plot showing specific capacitances measured at different power densities, measured using 384 both voltammetric and amperometric tests. F) Capacitance retention of carbon monoliths 385 obtained from Fe-treated wood at 1000ºC as a function of cycle number. 386 References 387 [1] M.C. Liu, L.B. Kong, P. Zhang, Y.C. Luo, L. Kang, Electrochim Acta, 60 (2012) 443-448. 388 [2] G.P. Wang, L. Zhang, J.J. Zhang, Chem Soc Rev, 41 (2012) 797-828. 389 [3] C. Peng, X.B. Yan, R.T. Wang, J.W. Lang, Y.J. Ou, Q.J. Xue, Electrochim Acta, 87 (2013) 401-390 408. 391 [4] B.L. Hu, X.Y. Qin, A.M. Asiri, K.A. Alamry, A.O. Al-Youbi, X.P. Sun, Electrochim Acta, 100 392 (2013) 24-28. 393 [5] X.Y. Zhang, X.Y. Wang, L.L. Jiang, H. Wu, C. Wu, J.C. Su, J Power Sources, 216 (2012) 290-394 296. 395 [6] A.S. Arico, P. Bruce, B. Scrosati, J.M. Tarascon, W. Van Schalkwijk, Nat Mater, 4 (2005) 396 366-377. 397 [7] Y.G. Guo, J.S. Hu, L.J. Wan, Adv Mater, 20 (2008) 2878-2887. 398 [8] P. Kossyrev, J Power Sources, 201 (2012) 347-352. 399 [9] A. Burke, J Power Sources, 91 (2000) 37-50. 400 [10] S.L. Candelaria, Y.Y. Shao, W. Zhou, X.L. Li, J. Xiao, J.G. Zhang, Y. Wang, J. Liu, J.H. Li, G.Z. 401 Cao, Nano Energy, 1 (2012) 195-220. 402 [11] A.G. Pandolfo, A.F. Hollenkamp, J Power Sources, 157 (2006) 11-27. 403 [12] L.L. Zhang, X.S. Zhao, Chem Soc Rev, 38 (2009) 2520-2531. 404 [13] A.B. Fuertes, G. Lota, T.A. Centeno, E. Frackowiak, Electrochim Acta, 50 (2005) 2799-405 2805. 406 [14] E. Frackowiak, F. Beguin, Carbon, 39 (2001) 937-950. 407 [15] V. Ruiz, C. Blanco, R. Santamaria, J.M. Ramos-Fernandez, M. Martinez-Escandell, A. 408 Sepulveda-Escribano, F. Rodriguez-Reinoso, Carbon, 47 (2009) 195-200. 409 [16] P. Greil, T. Lifka, A. Kaindl, J Eur Ceram Soc, 18 (1998) 1961-1973. 410 [17] P. Greil, J Eur Ceram Soc, 21 (2001) 105-118. 411 [18] Z. Schnepp, W. Yang, M. Antonietti, C. Giordano, Angew Chem Int Edit, 49 (2010) 6564-412 6566. 413 [19] Y.K. Lv, L.H. Gan, M.X. Liu, W. Xiong, Z.J. Xu, D.Z. Zhu, D.S. Wright, J Power Sources, 209 414 (2012) 152-157. 415 [20] Y. Gao, W.L. Zhang, Q.Y. Yue, B.Y. Gao, Y.Y. Sun, J.J. Kong, P. Zhao, J Power Sources, 270 416 (2014) 403-410. 417 [21] J.M. Fernandez, A. Munoz, A.R.D. Lopez, F.M.V. Feria, A. Dominguez-Rodriguez, M. Singh, 418 Acta Mater, 51 (2003) 3259-3275. 419 [22] J. Martinez-Fernandez, F. Valera-Feria, M. Singh, Scripta Mater, 43 (2000) 813-818. 420 [23] K.E. Pappacena, S.P. Gentry, T.E. Wilkes, M.T. Johnson, S. Xie, A. Davis, K.T. Faber, J Eur 421 Ceram Soc, 29 (2009) 3069-3077. 422 [24] A. Oya, H. Marsh, J Mater Sci, 17 (1982) 309-322. 423 [25] M.T. Johnson, K.T. Faber, J Mater Res, 26 (2011) 18-25. 424 [26] M.X. Liu, L.H. Gan, W. Xiong, F.Q. Zhao, X.Z. Fan, D.Z. Zhu, Z.J. Xu, Z.X. Hao, L.W. Chen, 425 Energ Fuel, 27 (2013) 1168-1173. 426 [27] M.X. Liu, L.H. Gan, W. Xiong, Z.J. Xu, D.Z. Zhu, L.W. Chen, J Mater Chem A, 2 (2014) 2555-427 2562. 428 [28] M. Sevilla, A.B. Fuertes, Carbon, 56 (2013) 155-166. 429 [29] Y.B. Cai, Y. Hu, L. Song, S.Y. Xuan, Y. Zhang, Z.Y. Chen, W.C. Fan, Polym Degrad Stabil, 92 430 (2007) 490-496. 431 [30] A. Sadezky, H. Muckenhuber, H. Grothe, R. Niessner, U. Poschl, Carbon, 43 (2005) 1731-432 1742. 433 [31] A. Gutierrez-Pardo, J. Ramirez-Rico, A.R. de Arellano-Lopez, J. Martinez-Fernandez, 434 Journal of Materials Science, DOI: 10.1007/s10853-014-8477-8 (2014) In Press. 435 [32] Y. Kodama, K. Sato, K. Suzuki, Y. Saito, T. Suzuki, T.J. Konno, Carbon, 50 (2012) 3486-436 3496. 437 [33] C.E. Byrne, D.C. Nagle, Carbon, 35 (1997) 267-273. 438 [34] C.E. Byrne, D.C. Nagle, Carbon, 35 (1997) 259-266. 439 [35] S. Glatzel, Z. Schnepp, C. Giordano, Angew Chem Int Edit, 52 (2013) 2355-2358. 440 [36] M. Armandi, B. Bonelli, K. Cho, R. Ryoo, E. Garrone, Int J Hydrogen Energ, 36 (2011) 441 7937-7943. 442 [37] K.S.W. Sing, Carbon, 32 (1994) 1311-1317. 443 [38] H. Marsh, D. Crawford, D.W. Taylor, Carbon, 21 (1983) 81-87. 444 [39] O.P. Krivoruchko, V.I. Zaikovskii, Mendeleev Commun, (1998) 97-100. 445 [40] X.F. Feng, S.W. Chee, R. Sharma, K. Liu, X. Xie, Q.Q. Li, S.S. Fan, K.L. Jiang, Nano Res, 4 446 (2011) 767-779. 447 [41] M. Sevilla, A.B. Fuertes, Carbon, 44 (2006) 468-474. 448 [42] M. Sevilla, C. Sanchis, T. Valdes-Solis, E. Morallon, A.B. Fuertes, J Phys Chem C, 111 449 (2007) 9749-9756. 450 [43] M. Sevilla, C. Sanchis, T. Valdes-Solis, E. Morallon, A.B. Fuertes, Electrochim Acta, 54 451 (2009) 2234-2238. 452 [44] M. Sevilla, A.B. Fuertes, Chem Phys Lett, 490 (2010) 63-68. 453 [45] J.P. Huo, H.H. Song, X.H. Chen, S.Q. Zhao, C.M. Xu, Mater Chem Phys, 101 (2007) 221-454 227. 455 [46] M. Yudasaka, K. Tasaka, R. Kikuchi, Y. Ohki, S. Yoshimura, E. Ota, J Appl Phys, 81 (1997) 456 7623-7629. 457 [47] S.A. Steiner, T.F. Baumann, B.C. Bayer, R. Blume, M.A. Worsley, W.J. MoberlyChan, E.L. 458 Shaw, R. Schlogl, A.J. Hart, S. Hofmann, B.L. Wardle, J Am Chem Soc, 131 (2009) 12144-459 12154. 460 [48] T. Pajkossy, J Electroanal Chem, 364 (1994) 111-125. 461 [49] G. Paasch, K. Micka, P. Gersdorf, Electrochim Acta, 38 (1993) 2653-2662. 462 [50] I.D. Raistrick, Electrochim Acta, 35 (1990) 1579-1586. 463 [51] O. Bohlen, J. Kowal, D.U. Sauer, J Power Sources, 173 (2007) 626-632. 464 [52] O. Bohlen, J. Kowal, D.U. Sauer, J Power Sources, 172 (2007) 468-475. 465 [53] J. Bisquert, G. Garcia-Belmonte, F. Fabregat-Santiago, N.S. Ferriols, P. Bogdanoff, E.C. 466 Pereira, J Phys Chem B, 104 (2000) 2287-2298. 467 [54] J. Bisquert, G. Garcia-Belmonte, F. Fabregat-Santiago, A. Compte, Electrochem Commun, 468 1 (1999) 429-435. 469 [55] R. de Levie, in: P. Delahay (Ed.) Advances in Electrochemistry and Electrochemical 470 Engineering, Wiley Interscience, New York, 1967, pp. 329. 471 [56] J.A. Rodrıguez, M. Fernandez Garcia, Synthesis, properties, and applications of oxide 472 nanomaterials, Wiley-Interscience, Hoboken, N.J., 2007. 473 474 Tables 475 Table 1. Most relevant results included in this work. 476 Fe-treated carbon Untreated carbon 1000ºC 1300ºC 1600ºC 1000ºC 1300ºC 1600ºC Maximum capacitance at 1 mV/s (F·g -1 ) 95 33 23 16 3.5 3.1 Maximum capacitance at 10 mV/s (F·g -1 ) 67 21 19 3.8 2.8 1.12 Maximum capacitance at 20 mA·g -1 (F·g -1 ) 133 34 24 - - - Maximum capacitance at 1A·g -1 (F·g -1 ) 35 8 5 - - - Electrical resistivity (10-4 Ω m) 2.2 0.8 0.6 4.8 3.5 1.9 Equivalent Series Resistance (Ω) 3.0 1.9 2.0 - - - BET surface area ( m 2 ·g -1 ) 370.2 217.4 167.2 2.1 1.7 1.5 Mean BET pore width (nm) 2.6 3.6 3.6 BET pore volume (cm3·g-1) 0.24 0.22 0.15 Residual hydrogen content (wt. %) 1.25 0.50 Residual nitrogen content (wt. %) 0.48 0.36 Residual iron content (wt. %) 0.38 9·10-3 Residual oxygen content (wt. %) 4.79 3.48 477 478 1000°C1300°C1600°C Before Feremoval After Feremoval 100 300 500 700 900 1100 0 20 40 60 80 100 Weightloss(%) Temperature(°C) Untreated ImpregnatedwithFeCl3 0.0 0.2 0.4 0.6 0.8 1.0 Derivativeweightloss(%/°C) 100 300 500 700 900 1100 ‐6 ‐4 ‐2 0 HeatFlow(W∙g‐1) Temperature(°C) Untreated ImpregnatedwithFeCl3 20 30 40 50 60 70 Diffractionangle,2(°) 1000°C 1300°C 1600°C 1000°C‐Treated 1300°C‐Treated 1600°C‐Treated 1000 1200 1400 1600 1800 Wavenumber(cm‐1) Untreated1000°C Untreated1300°C Untreated1600°C 1000 1200 1400 1600 1800 D1D2 Wavenumber(cm‐1) Fe‐treated1000°C Fe‐treated1300°C Fe‐treated1600°C G