FULL PAPER DOI: 10.1002/asia.201402338
Hierarchical Activated Mesoporous Phenolic-Resin-Based Carbons for Supercapacitors Zhao Wang, Min Zhou, Hao Chen, Jingui Jiang, and Shiyou Guan*[a] Abstract: A series of hierarchical activated mesoporous carbons (AMCs) were prepared by the activation of highly ordered, body-centered cubic mesoporous phenolic-resin-based carbon with KOH. The effect of the KOH/carbon-weight ratio on the textural properties and capacitive performance of the AMCs was investigated in detail. An AMC prepared with a KOH/carbon-weight ratio of 6:1 pos-
sessed the largest specific surface area (1118 m2 g1), with retention of the ordered mesoporous structure, and exhibited the highest specific capacitance of 260 F g1 at a current density of 0.1 A g1 in 1 m H2SO4 aqueous electroKeywords: carbon · electrochemistry · mesoporous materials · activation · supercapacitors
Introduction
At present, commercial supercapacitors are dominated by EDLCs that employ activated carbon as the electrode material, owing to its high microporosity and cost effectiveness.[14] However, the micropores (< 2 nm) in activated carbon are difficult to be fully exploited for accumulating charge, which leads to a low effective specific surface area for charge storage and, hence, to a relatively low specific capacitance.[15] Moreover, an evident decrease in capacitance is observed at high current densities, owing to the resistance of the diffusion and transportation of electrolyte ions in the inner pores.[15–17] Because the performance of a supercapacitor critically depends on the electrode materials, the development of new materials is very important. Carbon-based electrode materials, such as activated carbon, mesoporous carbon, carbon nanotubes, graphene, and carbon aerogels, have been widely studied as electrode materials for EDLCs.[18–22] Of these porous carbon materials, ordered mesoporous carbons have been considered to be promising electrode materials for supercapacitors, owing to their uniform pore size and specific channels.[20, 23] Among the ordered mesoporous carbons, highly ordered, body-centered cubic mesoporous phenolicresin-based carbon (C-FDU-16) has attracted widespread attention owing to its unique 3D networks, low-cost, and simple synthesis.[23–27] However, the relatively low specific surface area (< 600 m2 g1), complicated pore channels, and highly hydrophobic surface of C-FDU-16 result in a limited number of active sites, which is disadvantageous for charge storage. Recently, Cai et al. reported the modification of CFDU-16 through treatment with nitric acid and showed improved electrochemical capacitive performance owing to an increased number of active sites for energy storage and the promotion of wettability in aqueous electrolyte.[23] However, the introduction of oxygen-containing groups resulted in
Supercapacitors, a class of electrical-energy-storage devices with higher power density, longer cycle lifetime (> 105), and shorter charging time compared to batteries, have attracted great attention and been successfully applied in many fields.[1–5] However, the lower energy density (normally 10 Wh kg1) of supercapacitors compared with batteries (20–170 Wh kg1) limits their further application.[5] Based on the charge-storage mechanism, supercapacitors can be divided into two categories: pseudocapacitors and electrical double-layer capacitors (EDLCs).[2–8] Pseudocapacitors, in which metal oxides and conducting polymers are used as the main types of electrode materials,[4, 9–11] store electrical energy faradaically through electrosorption, reduction–oxidation reactions, and intercalation processes.[3] But their practical applications are limited because they often encounter poor electrical conductivity and cycling stability.[3, 5, 7, 12] Another class of supercapacitors, EDLCs, which usually employ porous carbons as the electrode materials, store energy by exploiting the charge separation at electrode/electrolyte interfaces. Therefore, the specific surface area, poresize distribution, surface functionality, and degree of crystallinity of porous carbon materials strongly affect the electrochemical performance of such supercapacitors.[3, 13] [a] Z. Wang,+ Dr. M. Zhou,+ H. Chen, Dr. J. Jiang, Prof. Dr. S. Guan School of Materials Science and Engineering East China University of Science and Technology Mei Long Road 130, Shanghai 200237 (P. R. China) Fax: (+ 86) 21-64251509 E-mail: [email protected] [+] These authors contributed equally to this work. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/asia.201402338.
Chem. Asian J. 2014, 9, 2789 – 2797
lyte. This material also showed excellent rate capability (163 F g1 retained at 20 A g1) and good long-term electrochemical stability. This superior capacitive performance could be attributed to a large specific surface area and an optimized micro-mesopore structure, which not only increased the effective specific surface area for charge storage but also provided a favorable pathway for efficient ion transport.
2789
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
a decrease in electric conductivity, thus leading to poor rate capability (only 49 % of the specific capacitance was retained on increasing the current density from 0.1 to 5 A g1). Moreover, the pseudo-capacitance that was derived from the Faradic redox reaction based on the oxygen-containing groups was unstable over long-term cycling. Recently, hierarchical micro-mesopore structures have been found to be highly effective for charge storage in carbon materials,[5, 7, 17, 28–30] because the hierarchical networks have good pore interconnectivity and have advantageous multimodal pores that show a synergistic effect during the electrochemical charge–discharge process.[5] Herein, we utilized a facial KOH activation process to increase the effective specific surface area and improve the pore interconnectivity of C-FDU-16. Electrochemical studies showed that the obtained hierarchical activated mesoporous carbons (AMCs) showed clearly enhanced capacitive performance as active materials for aqueous supercapacitors. An AMC that was prepared with a KOH/C-FDU-16 weight ratio of 6:1 exhibited the highest specific capacitance of 260 F g1 at a current density of 0.1 A g1 in 1 m H2SO4 aqueous electrolyte. This material also showed excellent rate capability (163 F g1 remained at a current density of 20 A g1) and good cycling stability over 10 000 cycles. The effects of the KOH/C-FDU-16 weight ratio on the textural properties and capacitive performance of the AMCs have been analyzed in detail.
Shiyou Guan et al.
FDU-16 was synthesized by using triblock copolymer F127 as a template and phenolic resin as a carbon precursor through a solvent-evaporation-induced self-assembly (EISA) method, followed by thermopolymerization and carbonization.[23–27] Then, C-FDU-16 was activated by KOH with different KOH/C-FDU-16 weight ratios (2:1, 4:1, 6:1, and 8:1) and the obtained materials were denoted as AMCX, where X was the KOH/C-FDU-16 weight ratio. Small-angle X-ray diffraction (XRD) patterns of the pristine mesoporous polymer (FDU-16), C-FDU-16, and the AMCs are shown in Figure 2. For FDU-16, one intense diffraction peak at about 0.898 and two weak diffraction peaks were clearly observed. These diffraction peaks were indexed as the 110, 200, and 211 diffractions of the body-centered cubic space group (Im3m).[25, 27, 31] After calcination at 800 8C under a nitrogen atmosphere, an intense diffraction peak was also observed in the small-angle XRD pattern of CFDU-16, thus confirming that the highly ordered cubic (Im3m) mesopore structure was thermally stable during carbonization.[25] The d-spacings were calculated by using the Bragg equation, d = l/2 sin q, and the unit-cell parameters
Results and Discussion Synthesis and Characterization Our synthesis of the AMCs is shown in Figure 1. First, C-
Figure 1. Synthesis of the AMCs.
Abstract in Chinese:
Figure 2. Small-angle XRD patterns of FDU-16, C-FDU-16, and the synthesized AMCs.
Chem. Asian J. 2014, 9, 2789 – 2797
2790
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Shiyou Guan et al.
www.chemasianj.org
p (a0) were calculated according to the formula, a0 = 2 d110, where d110 represented the d-spacing of the 110 diffraction.[25, 27] The unit-cell parameters were calculated to be 13.9 and 13.6 nm for FDU-16 and C-FDU-16, respectively. The difference was attributed to shrinkage of the framework during the calcination process.[27] After KOH activation, although the diffraction peaks of the AMCs were slightly broadened and the peak intensity decreased gradually, the diffraction peaks were maintained, even at a KOH/C-FUD16 weight ratio up to 6:1, thus suggesting excellent framework stability of C-FUD-16. However, a slight shift towards lower angle of the 110 diffraction peak was also observed, thus indicating an increase in the unit cell parameters. The unit cell parameters of AMC-2, AMC-4, and AMC-6 were 14.2, 14.3, and 14.4 nm, respectively, thus implying a framework expansion caused by etching of the mesopore wall during KOH activation.[20] During KOH activation, etching of the mesopore wall and shrinkage of the framework coexisted, whilst the former was dominant for AMC-2, AMC-4, and AMC-6. When the KOH/C-FUD-16 weight ratio was increased further to 8:1, the characteristic peaks of AMC-8 became less resolved, reflecting the degeneration of mesoporous ordering. To further investigate the textural changes after KOH activation, N2-adsorption/desorption isotherms of C-FDU-16 and AMCs were recorded (Figure 3 a). Clearly, C-FDU-16 exhibited a type-IV curve with a H2-type hysteresis loop, thus implying the presence of a 3D caged porous structure.[25, 32] The specific surface area, micropore area, mesopore area, and pore volume of C-FDU-16 were 548, 358, 190 m2 g1, and 0.31 cm3 g1, respectively (Table 1). After KOH activation, a dramatic increase in N2-sorption capacity at low relative pressure (p/p0 < 0.1) was clearly observed, thus indicating the introduction of a large number of micropores. The N2-sorption capacity at low relative pressure (p/ p0 < 0.1) increased as the KOH/C-FDU-16 weight ratio increased from 2:1 to 6:1, whereas it decreased on further increasing the KOH/C-FDU-16 weight ratio to 8:1. The micropore areas were 472, 763, 938, and 524 m2 g1 for AMC-2, AMC-4, AMC-6, and AMC-8, respectively (Table 1). In addition, the “knee” at lower relative pressures became wider as the KOH/C-FDU-16 weight ratio increased from 2:1 to 8:1, thus indicating the formation of larger pores. These results could be ascribed to collapse of the micropores, thus leading to the formation of more unordered mesopores under severe activation conditions. The capillary condensation steps at p/p0 = 0.4–0.6 became less evident after KOH activation, thus indicating degradation of the ordered mesostructure, in agreement with the XRD results (Figure 2). The mesopore areas were 176, 145, 180, and 507 m2 g1 for AMC-2, AMC-4, AMC-6, and AMC-8, respectively (Table 1). The trend of an initial decrease followed by increasing mesopore area was due to simultaneous destruction of the ordered mesopores and the formation of new unordered mesopores during KOH activation. AMC-6 possessed the largest specific surface area (1118 m2 g1), whilst the specific surface areas of AMC-2, AMC-4, and AMC-8 were
Chem. Asian J. 2014, 9, 2789 – 2797
Figure 3. a) N2-adsorption/desorption isotherms and b) pore-size distributions of C-FDU-16 and the synthesized AMCs. STP = standard temperature and pressure.
Table 1. Porous properties of C-FDU-16 and the synthesized AMCs. Sample
C-FDU16 AMC-2 AMC-4 AMC-6 AMC-8
Specific surface area ACHTUNGRE[m2 g1]
Microporous area
Mesoporous area
Pore volume
ACHTUNGRE[m2 g1]
ACHTUNGRE[m2 g1]
ACHTUNGRE[cm3 g1]
548
358
190
0.31
648 908 1118 1031
472 763 938 524
176 145 180 507
0.33 0.45 0.55 0.60
648, 908, and 1031 m2 g1, respectively. The pore volumes were 0.33, 0.45, 0.55, and 0.60 cm3 g1 for AMC-2, AMC-4, AMC-6, and AMC-8, respectively (Table 1). These results confirmed that the KOH/C-FDU-16 weight ratio was an important parameter for controlling the development of porosity in the chemical activation. More information related to pore size and poresize distribution were obtained from the adsorption branch of the isotherms by using the BJH method (Figure 3 b). A clear bimodal micro-mesopore structure and increasing pore size with increasing KOH/C-FDU16 weight ratio were observed. The hierarchical micro-meso-
2791
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
Shiyou Guan et al.
pore structure had good pore interconnectivity and could exhibit the advantage of bimodal pores with a synergistic effect during the electrochemical charging–discharging process. For example, the mesopores could provide a favorable path for efficient ion transport and the micropores could strengthen the electric double-layer capacitance.[4, 5, 33] Figure 4 shows transmission electron microscopy (TEM) images of C-FDU-16 and AMC-6. Clearly, C-FDU-16 possessed a highly ordered mesopore structure as viewed along the [110] direction and the pore size was approximately 3.6 nm (Figure 4 a, b). After KOH activation, AMC-6 also exhibited an ordered mesopore structure (Figure 4 c, d), thus
Figure 5. a) Wide-angle XRD patterns and b) Raman spectra of C-FDU16 and the synthesized AMCs. Figure 4. TEM images of a, b) C-FDU-16 and c, d) AMC-6.
Raman spectra of all of the obtained samples are shown in Figure 5 b. The peaks at about 1350 and 1596 cm1 were assigned to the characteristic D and G bands, respectively, which were typical Raman peaks for carbon materials.[3, 4, 38–40] The G (graphitic) band corresponds to the E2g mode of hexagonal graphite and is related to the vibration of the sp2-hybridized carbon atoms in a graphite layer, whilst the D (defects and disorder) band corresponds to the vibration of carbon atoms with dangling bonds in the plane with termination by disordered graphite.[41] In general, the ratio of the intensity of the D/G bands is proportional to the number of defect sites. The higher the ratio, the lower the degree of graphitization.[4] The intensity of D band increased after KOH activation, thus suggesting a less ordered crystallite structure. Moreover, the tendency of these changes became more pronounced with increasing the amount of KOH, in agreement with the wide-angle XRD data. Notably, a decrease in the degree of graphitization would result in a decrease in the conductivity of carbon. Thus, it is of great importance to control the amount of KOH in the chemical activation to obtain a balanced porous microstructure and degree of graphitization for activated carbon materials. In short, a series of hierarchical activated mesoporous carbons were prepared by the KOH activation of C-FDU-16, which was synthetized by using a facile surfactant-assembly template strategy. Owing to their unique hierarchical porous
indicating good framework stability, even under severe activation conditions. However, compared with the images of C-FDU-16, chemical activation clearly generated numerous micro-mesopores in the wall. It has been generally suggested that the activation reaction between KOH and carbon proceeds according to 6 KOH+2 C!2 K+2 K2CO3+3 H2, followed by the decomposition of K2CO3 and/or the simultaneous reaction of K/K2CO3/CO2 with carbon.[20, 34–37] These TEM analyses were consistent with the small-angle XRD and N2adsorption/desorption isotherm measurements. Chemical activation can result in a well-developed microstructure and a high specific surface area, whilst excess activation could also decrease the degree of graphitization.[4] To investigate the degree of graphitization in the as-prepared samples, wide-angle XRD and Raman spectroscopy were employed (Figure 5). As shown in Figure 5 a, all of the samples displayed a broadened diffraction peak at about 238 and a weak diffraction peak at about 438, which corresponded to the (002) and (100) diffraction peaks of a graphitictype lattice, respectively, thus indicating an amorphous structure. After KOH activation, the intensity of the peaks became lower and the full-width-at-half-maximum of the peaks broadened, thus confirming that the crystalline structures were markedly destroyed during the chemical etching process.[34] On increasing the KOH/C-FDU-16 weight ratio from 2:1 to 8:1, these changes became more clear.
Chem. Asian J. 2014, 9, 2789 – 2797
2792
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
structure, large specific surface area, and increased micropore proportion, these hierarchical activated mesoporous carbons were expected to exhibit outstanding electrochemical capacitive performance as electrode materials for supercapacitors.
which was electrochemically accessible, was needed to obtain high specific capacitance. The other AMCs exhibited quasi-rectangular voltammetry curves, thus implying that they were excellent candidates as electrode materials for EDLCs. Notably, an unapparent broadened reversible hump was also observed at about 0.3 V for these AMCs, which might be due to redox reactions related to the oxygen-containing functional groups.[4, 23, 42] In general, a small amount of oxygen-containing functional groups are inevitably introduced during the KOH activation and they can provide pseudo-capacitance through redox reactions and improve the wettability of carbon materials to the electrolyte solution, thus leading to improved electrochemical capacitive performance.[43, 44] It is generally believed that electrochemical capacitance is proportional to the area under the CV curve and a higher current response results in a larger area under the CV curve.[45] Clearly, the current response of AMC-6 was the largest, thus indicating that it had the largest electrochemical capacitance. The specific capacitance of AMC-6 was up to 200 F g1 at a scan rate of 10 mV s1, higher than those of C-FDU-16 (12 F g1), AMC-2 (77 F g1), AMC-4 (155 F g1), and AMC-8 (96 F g1). Figure 6 b shows galvanostatic charge–discharge curves of all of the samples at a current density of 0.1 A g1 within the potential window of 0.1 to 0.7 V versus Ag/AgCl. C-FDU-16 and AMC-2 both exhibited short charge–discharge time with different shapes to the typical triangular shape of the EDLCs, thus implying poor electrical app:ds:double-layer capacitive performance. The charge–discharge curves of other AMCs all showed triangular shapes with good symmetry, which indicated good electrical double-layer capacitive behavior and electrochemical reversibility. Clearly, AMC-6 exhibited the longest charge–discharge time, thus indicating that it had the largest specific capacitance. The specific capacitance of AMC-6 was up to 260 F g1 at a current density of 0.1 A g1, higher than those of C-FDU-16 (32 F g1), AMC-2 (143 F g1), AMC-4 (216 F g1), and AMC-8 (115 F g1). Notably, the specific capacitance of the nitricacid-modified C-FDU-16 reported by Cai et al. was only 219 F g1 at a current density of 0.1 A g1,[23] lower than that of AMC-6. The main reason for the highest specific capacitance of AMC-6 may be due to its high specific surface area and optimized micro-mesopore structure (an ordered mesopore with a large number of micro-mesopores in the wall connecting neighboring ordered mesopore channels; Figure 3, Figure 4, and Table 1), which can increase the effective specific surface area for charge storage and improve ion penetration and transport. The specific surface area of AMC-8 was relatively large; however, its specific capacitance was undesirable. This result may be due to the excessive KOH activation, which led to collapse of the ordered mesopore structure (Figure 2) and, hence, was disadvantageous for ion penetration and transportation. These results further demonstrate that a hierarchical pore structure with interconnected pore channels is beneficial for charge storage and that higher capacitance can be obtained with a larger effective surface area.
Electrochemical Performance The electrochemical performance of the as-obtained carbon materials were characterized by cyclic voltammetry (CV), galvanostatic charge–discharge measurements, and electrochemical impedance spectroscopy (EIS) in 1 m H2SO4 aqueous solution at ambient temperature. Figure 6 a shows the CV curves of C-FDU-16 and AMCs at a scan rate of 10 mV s1 within the potential window of 0.1 to 0.7 V versus Ag/AgCl. C-FDU-16 and AMC-2 both exhibited small shapes that were different to the typical rectangular shape of the EDLCs, thus indicating poor capacitive characteristics and low capacitance. This result was ascribed to a relatively low specific surface area, complicated pore channels, and disconnection between the neighboring channels, which led to a low effective specific surface area for the formation of an electrochemical double-layer. The basis for storage of the electrical energy for supercapacitors was the separation of charge at the electrode/electrolyte interface.[30] Therefore, a large effective specific surface area,
Figure 6. a) Cyclic voltammetry curves at a scan rate of 10 mV s1 and b) galvanostatic charge–discharge curves at a current density of 0.1 A g1 of C-FDU-16 and the synthesized AMCs.
Chem. Asian J. 2014, 9, 2789 – 2797
Shiyou Guan et al.
2793
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
Shiyou Guan et al.
Figure 7 a shows CV curves of AMC-6 at various scan rates from 2 to 200 mV s1. As the scan rate increased, the CV curves gradually became depressed whilst maintaining a quasi-rectangular shape with slight distortions at high scan rates. Moreover, the redox peaks became less pronounced as the scan rate increased, which was attributed to the fact that the rate of the redox reaction could not match the rapid increase in scan rate. The galvanostatic charge–discharge curves of AMC-6 at different current densities from 0.1 to 20 A g1 are shown in Figure 7 b. All of the galvanostatic charge–discharge curves showed quasi-triangular shapes with good symmetry, even at a current density of 20 A g1, thus indicating excellent capacitive performance and electrochemical reversibility, even at high current densities. Figure 7 c shows a plot of specific capacitance versus scan rate for all of the samples. On increasing the scan rate, the specific capacitance of all of the samples decreased gradually. However, AMC-6 exhibited excellent capacitance at various scan rates. At a scan rate of 2 mV s1, the specific capacitance of AMC-6 was up to 224 F g1, and a high specific capacitance of 107 F g1 was obtained at a scan rate of 200 mV s1, which resulted in a capacitance-retention ratio of 48 %, thus confirming good rate capability. This result was due to a unique 3D interconnected pore structure, which led to fast ion transport within the mesopores and a short diffusion distance from the mesopores to the micropores.[46, 47] Figure 7 d shows a plot of specific capacitance
versus current density for all of the samples. On increasing the current density from 0.1 to 20 A g1, the specific capacitance of all of the samples gradually decreased, which was attributed to the fact that the charge diffusion could not match the rapid increase in current density. The specific capacitance of AMC-6 decreased from 260 F g1 at a current density of 0.1 A g1 to 191 F g1 at a current density of 5 A g1, which corresponded to a capacitanceretention ratio of 73 %, clearly higher than that (49 %) of the nitric-acidmodified C-FDU-16 reported by Cai et al.[23] Even at a current density of 20 A g1, the specific capacitance was still up to 163 F g1, thus indicating a capacitance-retention ratio of 63 %. The excellent rate capability of AMC-6 was due to the capacitance being mainly derived from charge separation at the electrode/electrolyte interface and this surfacestorage mechanism allowed very fast energy uptake and delivery, thus leading to better power performance.[48] However, the introduction of oxygen-containing groups through treatment with nitric acid resulted in a decrease in electric conductivity, which was detrimental to the rate capability. Moreover, the rate capability of pseudo-capacitance was inferior to that of double-layer capacitance.[49] Figure 8 a shows Nyquist plots that were derived from EIS analysis of the as-obtained samples. These plots all presented similarly shaped Nyquist plots that were composed of three regions: an almost vertical line in the low-frequency region, which was related to the capacitive behavior; a sloped line in the medium-frequency region, which was related to diffusion resistance; and a semi-circular line in the highfrequency region, which was related to the interfacial chargetransfer resistance.[49] Clearly, compared with C-FDU-16, the AMCs showed a much straighter line in the low-frequency region, a shortened sloped line in the medium-frequency region, and a much smaller semi-circular line in the highfrequency region, thus indicating significantly improved electrochemical capacitive performance, low diffusion resistance, and low interfacial chargetransfer resistance, respectively. In addition, the diameter of the semi-circular line for AMC-6 was smaller than those for AMC-2, AMC-4, and AMC-8, thus indicating that AMC-6 had the lowest impedance at the electrode/electrolyte interface and the fastest ion movement Figure 7. a) Cyclic voltammetry curves at different scan rates and b) galvanostatic charge–discharge curves at 1 inside the pores.[5, 50, 51] The different current densities of AMC-6; plots of c) specific capacitance versus scan rate (2–200 mV s ) and 1 length of the almostvertical line d) specific capacitance versus current density (0.1–20 A g ) of C-FDU-16 and the synthesized AMCs.
Chem. Asian J. 2014, 9, 2789 – 2797
2794
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Shiyou Guan et al.
www.chemasianj.org
Figure 9. a) Cycling stability of AMC-6 at a current density of 2 A g1; b) Nyquist plots of AMC-6 before and after 10 000 cycles.
Figure 8. a) Nyquist plots and b) plots of specific capacitance versus frequency of C-FDU-16 and the synthesized AMCs.
in the low-frequency region corresponds to the specific capacitance of the carbon electrodes.[1] The shorter the length, the higher the specific capacitance. Clearly, AMC-6 possessed the shortest length of the almostvertical line in the low-frequency region, thus indicating that it had the highest specific capacitance, consistent with the results of the CV and galvanostatic charge–discharge analyses. Figure 8 b shows a plot of specific capacitance as a function of frequency for all of the samples. At a low frequency of about 0.04 Hz, the specific capacitance was 150 F g1 for AMC-6, higher than those of C-FDU-16 (13 F g1), AMC-2 (81 F g1), AMC-4 (122 F g1), and AMC-8 (79 F g1). Clearly, the specific capacitance gradually decreased as the frequency increased in the low-frequency region and the capacitance tended towards zero in the high-frequency region. This result was because ion penetration inside the pores became difficult at a highly alternating current. Cycling stability is a key factor in the practical application of electrode materials in supercapacitors. Excellent cycling stability is needed for supercapacitor operation. To evaluate the cycling stability of AMC-6, galvanostatic charge–discharge cycling was performed at a current density of 2 A g1 within the potential window of 0.1 to 0.7 V versus Ag/ AgCl (Figure 9 a). The specific capacitance of AMC-6 was quite stable over 10 000 cycles, thus indicating excellent cycling stability. A slight increase in specific capacitance
Chem. Asian J. 2014, 9, 2789 – 2797
(about 3 %) after 10 000 cycles was observed, which may be due to continued activation of the electroactive material during the charge–discharge process. Figure 9 b shows Nyquist plots of AMC-6 before and after 10 000 cycles. In this case, two Nyquist plots were almost overlapped, thus indicating that the impedance of AMC-6 before and after 10 000 cycles were almost equal. This result demonstrated that no deterioration of the electrode materials occurred during the galvanostatic charge–discharge cycling.
Conclusion In summary, a series of hierarchical activated mesoporous carbon materials with a high specific surface area and an improved micro-mesopore structure were synthesized by the activation of highly ordered, body-centered-cubic mesoporous phenolic-resin-based carbon with KOH. Of these activated mesoporous carbon materials, owing to a large effective specific surface area and an optimized hierarchical micro-mesopore structure, AMC-6 exhibited the best electrochemical capacitive performance, including a high specific capacitance (260 F g1 at a current density of 0.1 A g1), excellent rate capability (163 F g1 retained at a current density of 20 A g1), and good cycling stability (no capacitance loss over 10 000 cycles), thus presenting great potential for
2795
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Shiyou Guan et al.
www.chemasianj.org
Acknowledgements
practical application as a high-performance electrode material for supercapacitors.
The authors are grateful for the financial support from the National Natural Science Foundation of China (No. 21274043).
Experimental Section [1] C. Lei, N. Amini, F. Markoulidis, P. Wilson, S. Tennison, C. Lekakou, J. Mater. Chem. A 2013, 1, 6037 – 6042. [2] Y. P. Zhai, Y. Q. Dou, D. Y. Zhao, P. F. Fulvio, R. T. Mayes, S. Dai, Adv. Mater. 2011, 23, 4828 – 4850. [3] W. J. Qian, F. X. Sun, Y. H. Xu, L. H. Qiu, C. H. Liu, S. D. Wang, F. Yan, Energy Environ. Sci. 2014, 7, 379 – 386. [4] M. Zhou, F. Pu, Z. Wang, S. Y. Guan, Carbon 2014, 68, 185 – 194. [5] Y. R. Liang, F. X. Liang, H. Zhong, Z. H. Li, R. W. Fu, D. C. Wu, J. Mater. Chem. A 2013, 1, 7000 – 7005. [6] H. Zhu, X. L. Wang, X. X. Liu, X. R. Yang, Adv. Mater. 2012, 24, 6524 – 6529. [7] G. Wang, L. Zhang, J. Zhang, Chem. Soc. Rev. 2012, 41, 797 – 828. [8] G. A. Snook, P. Kao, A. S. Best, J. Power Sources 2011, 196, 1 – 12. [9] Z. S. Wu, D. W. Wang, W. C. Ren, J. P. Zhao, G. M. Zhou, F. Li, H. M. Cheng, Adv. Funct. Mater. 2010, 20, 3595 – 3602. [10] S. Chen, J. Zhu, X. Wu, Q. Han, X. Wang, ACS Nano 2010, 4, 2822 – 2830. [11] A. Sumboja, C. Y. Foo, X. Wang, P. S. Lee, Adv. Mater. 2013, 25, 2809 – 2815. [12] X. Dong, X. Wang, L. Wang, H. Song, X. Li, L. Wang, M. B. ChanPark, C. M. Li, P. Chen, Carbon 2012, 50, 4865 – 4870. [13] S. Kondrat, C. R. Perez, V. Presser, Y. Gogotsi, A. A. Kornyshev, Energy Environ. Sci. 2012, 5, 6474 – 6479. [14] B. Xu, H. Duan, M. Chu, G. P. Cao, Y. S. Yang, J. Mater. Chem. A 2013, 1, 4565 – 4570. [15] Y. Y. Lv, F. Zhang, Y. Q. Dou, Y. P. Zhai, J. X. Wang, H. J. Liu, Y. Y. Xia, B. Tu, D. Y. Zhao, J. Mater. Chem. 2012, 22, 93 – 99. [16] A. Burke, J. Power Sources 2000, 91, 37 – 50. [17] E. Frackowiak, F. Beguin, Carbon 2001, 39, 937 – 950. [18] X. Q. Wang, J. S. Lee, C. Tsouris, D. W. DePaoli, S. Dai, J. Mater. Chem. 2010, 20, 4602 – 4608. [19] H. J. Ahn, J. H. Lee, Y. Jeong, J. H. Lee, C. S. Chi, H. J. Oh, Mater. Sci. Eng. A 2007, 449, 841 – 845. [20] L. Zou, G. Morris, D. Qi, Desalination 2008, 225, 329 – 340. [21] P. Xu, J. E. Drewes, D. Heil, G. Wang, Water Res. 2008, 42, 2605 – 2617. [22] M. Zhou, F. Pu, H. Chen, Z. Wang, H. Y. Zhang, S. Y. Guan, New J. Chem. 2013, 37, 4148 – 4155. [23] T. W. Cai, M. Zhou, D. Y. Ren, G. S. Han, S. Y. Guan, J. Power Sources 2013, 231, 197 – 202. [24] Y. Meng, D. Gu, F. Q. Zhang, Y. F. Shi, L. Cheng, D. Feng, Z. X. Wu, Z. X. Chen, Y. Wan, A. Stein, D. Y. Zhao, Chem. Mater. 2006, 18, 4447 – 4464. [25] Y. Meng, D. Gu, F. Q. Zhang, Y. F. Shi, H. H. Yang, Z. Li, C. Z. Yu, B. Tu, D. Y. Zhao, Angew. Chem. Int. Ed. 2005, 44, 7053 – 7059; Angew. Chem. 2005, 117, 7215 – 7221. [26] Y. Huang, H. Q. Cai, T. Yu, X. L. Sun, B. Tu, D. Y. Zhao, Chem. Asian J. 2007, 2, 1282 – 1289. [27] F. Q. Zhang, Y. Meng, D. Gu, Y. Yan, Z. X. Chen, B. Tu, D. Y. Zhao, Chem. Mater. 2006, 18, 5279 – 5288. [28] Y. W. Zhu, S. Murali, M. D. Stoller, K. J. Ganesh, W. W. Cai, P. J. Ferreira, A. Pirkle, R. M. Wallace, K. A. Cychosz, M. Thommes, D. Su, E. A. Stach, R. S. Ruoff, Science 2011, 332, 1537 – 1541. [29] A. G. Pandolfo, A. F. Hollenkamp, J. Power Sources 2006, 157, 11 – 27. [30] D. Y. Qu, H. Shi, J. Power Sources 1998, 74, 99 – 107. [31] D. Y. Zhao, Q. S. Huo, J. L. Feng, B. F. Chmelka, G. D. Stucky, J. Am. Chem. Soc. 1998, 120, 6024 – 6036. [32] J. R. Matos, M. Kruk, L. P. Mercuri, M. Jaroniec, L. Zhao, T. Kamiyama, O. Terasaki, T. J. Pinnavaia, Y. Liu, J. Am. Chem. Soc. 2003, 125, 821 – 829. [33] X. Y. Chen, C. Chen, Z. J. Zhang, D. H. Xie, X. Deng, J. W. Liu, J. Power Sources 2013, 230, 50 – 58.
Sample Preparation The synthetic process of C-FDU-16 was similar to that reported by Zhao and co-workers.[23–27] C-FDU-16 was activated by KOH with different KOH/C-FDU-16 weight ratios (2:1, 4:1, 6:1, and 8:1). Taking the activation with a KOH/C-FDU-16 weight ratio of 6:1 as an example, C-FDU16 (0.3 g) was impregnated with KOH solution (1.8 g KOH in 11.25 mL EtOH), followed by evaporation at 65 8C under stirring in a flow of N2. The activation process was performed in a tubular furnace at 700 8C for 2 h under a flow of N2 (heating rate: 3 8C min1). Then, the obtained samples were washed with 0.2 m HCl and deionized water until the pH value of the filtrate was approximately 7. The final AMCs were obtained after drying at 80 8C for 24 h. Sample Characterization Powder XRD patterns were collected on a D/MAX 2550VB/PC diffractometer (Rigaku, Japan) that was equipped with CuKa radiation (l = 0.154 nm). The porous properties of the samples were determined from N2-adsorption/desorption isotherms and measured at 77 K on an ASAP 2010 (Micromeritics, USA) instrument. Prior to the measurements, the samples were degassed at 180 8C for 8 h. The specific surface area was determined by using BET method and the poresize distributions were derived from the adsorption branches of the isotherms by using BJH theory. The total pore volume was calculated from the amount of N2 adsorbed at a relative pressure of 0.99. TEM images were recorded on a JEM-2100 microscope (JEOL, Japan). Raman spectra were recorded on an InVia Raman microscope (Renishaw, UK) with excitation from a 514.5 nm diode laser. Electrochemical Measurements The electrochemical performance was measured in a three-electrode system in 1 m aqueous H2SO4 solution. Platinum foil and Ag/AgCl electrode were used as the counter and reference electrodes, respectively. The working electrodes were prepared by pressing a mixture of the activated material (80 wt %; C-FDU-16 and AMCs), acetylene black (10 wt %), and polytetrafluoroethylene binder (10 wt %) onto titanium mesh under 10 MPa pressure. Each working electrode contained 5 mg of the electrode material, with a size of about 1 cm2. CV and galvanostatic charge–discharge measurements were performed within the potential window of 0.1 to 0.7 V. EIS measurements were recorded from 100 kHz to 0.005 Hz, with an alternating current amplitude of 10 mV. The specific capacitance as calculated from the CV data was determined according to Equation (1), where Cm is the specific capacitance (F g1), I is the response current (A), s is the scan rate (V s1), DU is the potential window (V), and m is the mass of the active material within the working electrode (g). Cm ¼
Z
IdU=ð2 s m DUÞ
ð1Þ
The specific capacitance as calculated from the galvanostatic charge–discharge measurements was determined according to Equation (2), where I is the galvanostatic charge–discharge current (A) and Dt is the discharging time (s). Cm ¼ I Dt=ðm DUÞ
ð2Þ
The specific capacitance as calculated from the EIS measurements was determined according to Equation (3), where f is the frequency (Hz) and Zim is the imaginary part of the impedance (Ohm). Cm ¼ 1=ð2p f m Zim Þ
Chem. Asian J. 2014, 9, 2789 – 2797
ð3Þ
2796
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
[34] Q. Wang, Q. Cao, X. Y. Wang, B. Jing, H. Kuang, L. Zhou, J. Power Sources 2013, 225, 101 – 107. [35] Y. P. Guo, S. F. Yang, K. Yu, J. Z. Zhao, Z. C. Wang, H. D. Xu, Mater. Chem. Phys. 2002, 74, 320 – 323. [36] T. Otowa, Y. Nojima, T. Miyazaki, Carbon 1997, 35, 1315 – 1319. [37] M. A. Lillo-Rdenas, J. Juan-Juan, D. Cazorla-Amoros, A. LinaresSolano, Carbon 2004, 42, 1371 – 1375. [38] A. C. Ferrari, J. C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. S. Novoselov, S. Roth, A. K. Geim, Phys. Rev. Lett. 2006, 97, 187401 – 187404. [39] M. Zhou, T. W. Cai, F. Pu, H. Chen, Z. Wang, H. Y. Zhang, S. Y. Guan, ACS Appl. Mater. Interfaces 2013, 5, 3449 – 3455. [40] M. Zhou, F. Pu, Z. Wang, T. W. Cai, H. Chen, H. Y. Zhang, S. Y. Guan, Phys. Chem. Chem. Phys. 2013, 15, 11394 – 11401. [41] X. C. Chen, K. Kierzek, K. Wenelska, K. Cendrowski, J. Gong, X. Wen, T. Tang, P. K. Chu, E. Mijowska, Chem. Asian J. 2013, 8, 2627 – 2633. [42] T. W. Cai, M. Zhou, G. S. Han, S. Y. Guan, J. Power Sources 2013, 241, 6 – 11.
Chem. Asian J. 2014, 9, 2789 – 2797
[43] [44] [45] [46] [47] [48] [49] [50] [51]
2797
Shiyou Guan et al.
J. C. Wang, S. Kaskel, J. Mater. Chem. 2012, 22, 23710 – 23725. D. Y. Qu, J. Power Sources 2002, 109, 403 – 411. Y. Li, X. Zhao, P. Yu, Q. Zhang, Langmuir 2013, 29, 493 – 500. K. S. Xia, Q. M. Gao, J. H. Jiang, J. Hu, Carbon 2008, 46, 1718 – 1726. W. T. Huang, H. Zhang, Y. Q. Huang, W. K. Wang, S. C. Wei, Carbon 2011, 49, 838 – 843. P. Simon, Y. Gogotsi, Nat. Mater. 2008, 7, 845 – 854. B. Xu, S. S. Hou, G. P. Cao, F. Wu, Y. S. Yang, J. Mater. Chem. 2012, 22, 19088 – 19093. L. Yang, S. Cheng, Y. Ding, X. Zhu, Z. L. Wang, M. Liu, Nano Lett. 2012, 12, 321 – 325. W. F. Zhang, Z. H. Huang, G. P. Cao, F. Y. Kang, Y. S. Yang, J. Power Sources 2012, 204, 230 – 235. Received: April 7, 2014 Revised: May 15, 2014 Published online: August 5, 2014
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Hierarchical Activated Mesoporous Phenolic-Resin-Based Carbons for Supercapacitors Zhao Wang, Min Zhou, Hao Chen, Jingui Jiang, and Shiyou Guan*[a] Abstract: A series of hierarchical activated mesoporous carbons (AMCs) were prepared by the activation of highly ordered, body-centered cubic mesoporous phenolic-resin-based carbon with KOH. The effect of the KOH/carbon-weight ratio on the textural properties and capacitive performance of the AMCs was investigated in detail. An AMC prepared with a KOH/carbon-weight ratio of 6:1 pos-
sessed the largest specific surface area (1118 m2 g1), with retention of the ordered mesoporous structure, and exhibited the highest specific capacitance of 260 F g1 at a current density of 0.1 A g1 in 1 m H2SO4 aqueous electroKeywords: carbon · electrochemistry · mesoporous materials · activation · supercapacitors
Introduction
At present, commercial supercapacitors are dominated by EDLCs that employ activated carbon as the electrode material, owing to its high microporosity and cost effectiveness.[14] However, the micropores (< 2 nm) in activated carbon are difficult to be fully exploited for accumulating charge, which leads to a low effective specific surface area for charge storage and, hence, to a relatively low specific capacitance.[15] Moreover, an evident decrease in capacitance is observed at high current densities, owing to the resistance of the diffusion and transportation of electrolyte ions in the inner pores.[15–17] Because the performance of a supercapacitor critically depends on the electrode materials, the development of new materials is very important. Carbon-based electrode materials, such as activated carbon, mesoporous carbon, carbon nanotubes, graphene, and carbon aerogels, have been widely studied as electrode materials for EDLCs.[18–22] Of these porous carbon materials, ordered mesoporous carbons have been considered to be promising electrode materials for supercapacitors, owing to their uniform pore size and specific channels.[20, 23] Among the ordered mesoporous carbons, highly ordered, body-centered cubic mesoporous phenolicresin-based carbon (C-FDU-16) has attracted widespread attention owing to its unique 3D networks, low-cost, and simple synthesis.[23–27] However, the relatively low specific surface area (< 600 m2 g1), complicated pore channels, and highly hydrophobic surface of C-FDU-16 result in a limited number of active sites, which is disadvantageous for charge storage. Recently, Cai et al. reported the modification of CFDU-16 through treatment with nitric acid and showed improved electrochemical capacitive performance owing to an increased number of active sites for energy storage and the promotion of wettability in aqueous electrolyte.[23] However, the introduction of oxygen-containing groups resulted in
Supercapacitors, a class of electrical-energy-storage devices with higher power density, longer cycle lifetime (> 105), and shorter charging time compared to batteries, have attracted great attention and been successfully applied in many fields.[1–5] However, the lower energy density (normally 10 Wh kg1) of supercapacitors compared with batteries (20–170 Wh kg1) limits their further application.[5] Based on the charge-storage mechanism, supercapacitors can be divided into two categories: pseudocapacitors and electrical double-layer capacitors (EDLCs).[2–8] Pseudocapacitors, in which metal oxides and conducting polymers are used as the main types of electrode materials,[4, 9–11] store electrical energy faradaically through electrosorption, reduction–oxidation reactions, and intercalation processes.[3] But their practical applications are limited because they often encounter poor electrical conductivity and cycling stability.[3, 5, 7, 12] Another class of supercapacitors, EDLCs, which usually employ porous carbons as the electrode materials, store energy by exploiting the charge separation at electrode/electrolyte interfaces. Therefore, the specific surface area, poresize distribution, surface functionality, and degree of crystallinity of porous carbon materials strongly affect the electrochemical performance of such supercapacitors.[3, 13] [a] Z. Wang,+ Dr. M. Zhou,+ H. Chen, Dr. J. Jiang, Prof. Dr. S. Guan School of Materials Science and Engineering East China University of Science and Technology Mei Long Road 130, Shanghai 200237 (P. R. China) Fax: (+ 86) 21-64251509 E-mail: [email protected] [+] These authors contributed equally to this work. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/asia.201402338.
Chem. Asian J. 2014, 9, 2789 – 2797
lyte. This material also showed excellent rate capability (163 F g1 retained at 20 A g1) and good long-term electrochemical stability. This superior capacitive performance could be attributed to a large specific surface area and an optimized micro-mesopore structure, which not only increased the effective specific surface area for charge storage but also provided a favorable pathway for efficient ion transport.
2789
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
a decrease in electric conductivity, thus leading to poor rate capability (only 49 % of the specific capacitance was retained on increasing the current density from 0.1 to 5 A g1). Moreover, the pseudo-capacitance that was derived from the Faradic redox reaction based on the oxygen-containing groups was unstable over long-term cycling. Recently, hierarchical micro-mesopore structures have been found to be highly effective for charge storage in carbon materials,[5, 7, 17, 28–30] because the hierarchical networks have good pore interconnectivity and have advantageous multimodal pores that show a synergistic effect during the electrochemical charge–discharge process.[5] Herein, we utilized a facial KOH activation process to increase the effective specific surface area and improve the pore interconnectivity of C-FDU-16. Electrochemical studies showed that the obtained hierarchical activated mesoporous carbons (AMCs) showed clearly enhanced capacitive performance as active materials for aqueous supercapacitors. An AMC that was prepared with a KOH/C-FDU-16 weight ratio of 6:1 exhibited the highest specific capacitance of 260 F g1 at a current density of 0.1 A g1 in 1 m H2SO4 aqueous electrolyte. This material also showed excellent rate capability (163 F g1 remained at a current density of 20 A g1) and good cycling stability over 10 000 cycles. The effects of the KOH/C-FDU-16 weight ratio on the textural properties and capacitive performance of the AMCs have been analyzed in detail.
Shiyou Guan et al.
FDU-16 was synthesized by using triblock copolymer F127 as a template and phenolic resin as a carbon precursor through a solvent-evaporation-induced self-assembly (EISA) method, followed by thermopolymerization and carbonization.[23–27] Then, C-FDU-16 was activated by KOH with different KOH/C-FDU-16 weight ratios (2:1, 4:1, 6:1, and 8:1) and the obtained materials were denoted as AMCX, where X was the KOH/C-FDU-16 weight ratio. Small-angle X-ray diffraction (XRD) patterns of the pristine mesoporous polymer (FDU-16), C-FDU-16, and the AMCs are shown in Figure 2. For FDU-16, one intense diffraction peak at about 0.898 and two weak diffraction peaks were clearly observed. These diffraction peaks were indexed as the 110, 200, and 211 diffractions of the body-centered cubic space group (Im3m).[25, 27, 31] After calcination at 800 8C under a nitrogen atmosphere, an intense diffraction peak was also observed in the small-angle XRD pattern of CFDU-16, thus confirming that the highly ordered cubic (Im3m) mesopore structure was thermally stable during carbonization.[25] The d-spacings were calculated by using the Bragg equation, d = l/2 sin q, and the unit-cell parameters
Results and Discussion Synthesis and Characterization Our synthesis of the AMCs is shown in Figure 1. First, C-
Figure 1. Synthesis of the AMCs.
Abstract in Chinese:
Figure 2. Small-angle XRD patterns of FDU-16, C-FDU-16, and the synthesized AMCs.
Chem. Asian J. 2014, 9, 2789 – 2797
2790
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Shiyou Guan et al.
www.chemasianj.org
p (a0) were calculated according to the formula, a0 = 2 d110, where d110 represented the d-spacing of the 110 diffraction.[25, 27] The unit-cell parameters were calculated to be 13.9 and 13.6 nm for FDU-16 and C-FDU-16, respectively. The difference was attributed to shrinkage of the framework during the calcination process.[27] After KOH activation, although the diffraction peaks of the AMCs were slightly broadened and the peak intensity decreased gradually, the diffraction peaks were maintained, even at a KOH/C-FUD16 weight ratio up to 6:1, thus suggesting excellent framework stability of C-FUD-16. However, a slight shift towards lower angle of the 110 diffraction peak was also observed, thus indicating an increase in the unit cell parameters. The unit cell parameters of AMC-2, AMC-4, and AMC-6 were 14.2, 14.3, and 14.4 nm, respectively, thus implying a framework expansion caused by etching of the mesopore wall during KOH activation.[20] During KOH activation, etching of the mesopore wall and shrinkage of the framework coexisted, whilst the former was dominant for AMC-2, AMC-4, and AMC-6. When the KOH/C-FUD-16 weight ratio was increased further to 8:1, the characteristic peaks of AMC-8 became less resolved, reflecting the degeneration of mesoporous ordering. To further investigate the textural changes after KOH activation, N2-adsorption/desorption isotherms of C-FDU-16 and AMCs were recorded (Figure 3 a). Clearly, C-FDU-16 exhibited a type-IV curve with a H2-type hysteresis loop, thus implying the presence of a 3D caged porous structure.[25, 32] The specific surface area, micropore area, mesopore area, and pore volume of C-FDU-16 were 548, 358, 190 m2 g1, and 0.31 cm3 g1, respectively (Table 1). After KOH activation, a dramatic increase in N2-sorption capacity at low relative pressure (p/p0 < 0.1) was clearly observed, thus indicating the introduction of a large number of micropores. The N2-sorption capacity at low relative pressure (p/ p0 < 0.1) increased as the KOH/C-FDU-16 weight ratio increased from 2:1 to 6:1, whereas it decreased on further increasing the KOH/C-FDU-16 weight ratio to 8:1. The micropore areas were 472, 763, 938, and 524 m2 g1 for AMC-2, AMC-4, AMC-6, and AMC-8, respectively (Table 1). In addition, the “knee” at lower relative pressures became wider as the KOH/C-FDU-16 weight ratio increased from 2:1 to 8:1, thus indicating the formation of larger pores. These results could be ascribed to collapse of the micropores, thus leading to the formation of more unordered mesopores under severe activation conditions. The capillary condensation steps at p/p0 = 0.4–0.6 became less evident after KOH activation, thus indicating degradation of the ordered mesostructure, in agreement with the XRD results (Figure 2). The mesopore areas were 176, 145, 180, and 507 m2 g1 for AMC-2, AMC-4, AMC-6, and AMC-8, respectively (Table 1). The trend of an initial decrease followed by increasing mesopore area was due to simultaneous destruction of the ordered mesopores and the formation of new unordered mesopores during KOH activation. AMC-6 possessed the largest specific surface area (1118 m2 g1), whilst the specific surface areas of AMC-2, AMC-4, and AMC-8 were
Chem. Asian J. 2014, 9, 2789 – 2797
Figure 3. a) N2-adsorption/desorption isotherms and b) pore-size distributions of C-FDU-16 and the synthesized AMCs. STP = standard temperature and pressure.
Table 1. Porous properties of C-FDU-16 and the synthesized AMCs. Sample
C-FDU16 AMC-2 AMC-4 AMC-6 AMC-8
Specific surface area ACHTUNGRE[m2 g1]
Microporous area
Mesoporous area
Pore volume
ACHTUNGRE[m2 g1]
ACHTUNGRE[m2 g1]
ACHTUNGRE[cm3 g1]
548
358
190
0.31
648 908 1118 1031
472 763 938 524
176 145 180 507
0.33 0.45 0.55 0.60
648, 908, and 1031 m2 g1, respectively. The pore volumes were 0.33, 0.45, 0.55, and 0.60 cm3 g1 for AMC-2, AMC-4, AMC-6, and AMC-8, respectively (Table 1). These results confirmed that the KOH/C-FDU-16 weight ratio was an important parameter for controlling the development of porosity in the chemical activation. More information related to pore size and poresize distribution were obtained from the adsorption branch of the isotherms by using the BJH method (Figure 3 b). A clear bimodal micro-mesopore structure and increasing pore size with increasing KOH/C-FDU16 weight ratio were observed. The hierarchical micro-meso-
2791
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
Shiyou Guan et al.
pore structure had good pore interconnectivity and could exhibit the advantage of bimodal pores with a synergistic effect during the electrochemical charging–discharging process. For example, the mesopores could provide a favorable path for efficient ion transport and the micropores could strengthen the electric double-layer capacitance.[4, 5, 33] Figure 4 shows transmission electron microscopy (TEM) images of C-FDU-16 and AMC-6. Clearly, C-FDU-16 possessed a highly ordered mesopore structure as viewed along the [110] direction and the pore size was approximately 3.6 nm (Figure 4 a, b). After KOH activation, AMC-6 also exhibited an ordered mesopore structure (Figure 4 c, d), thus
Figure 5. a) Wide-angle XRD patterns and b) Raman spectra of C-FDU16 and the synthesized AMCs. Figure 4. TEM images of a, b) C-FDU-16 and c, d) AMC-6.
Raman spectra of all of the obtained samples are shown in Figure 5 b. The peaks at about 1350 and 1596 cm1 were assigned to the characteristic D and G bands, respectively, which were typical Raman peaks for carbon materials.[3, 4, 38–40] The G (graphitic) band corresponds to the E2g mode of hexagonal graphite and is related to the vibration of the sp2-hybridized carbon atoms in a graphite layer, whilst the D (defects and disorder) band corresponds to the vibration of carbon atoms with dangling bonds in the plane with termination by disordered graphite.[41] In general, the ratio of the intensity of the D/G bands is proportional to the number of defect sites. The higher the ratio, the lower the degree of graphitization.[4] The intensity of D band increased after KOH activation, thus suggesting a less ordered crystallite structure. Moreover, the tendency of these changes became more pronounced with increasing the amount of KOH, in agreement with the wide-angle XRD data. Notably, a decrease in the degree of graphitization would result in a decrease in the conductivity of carbon. Thus, it is of great importance to control the amount of KOH in the chemical activation to obtain a balanced porous microstructure and degree of graphitization for activated carbon materials. In short, a series of hierarchical activated mesoporous carbons were prepared by the KOH activation of C-FDU-16, which was synthetized by using a facile surfactant-assembly template strategy. Owing to their unique hierarchical porous
indicating good framework stability, even under severe activation conditions. However, compared with the images of C-FDU-16, chemical activation clearly generated numerous micro-mesopores in the wall. It has been generally suggested that the activation reaction between KOH and carbon proceeds according to 6 KOH+2 C!2 K+2 K2CO3+3 H2, followed by the decomposition of K2CO3 and/or the simultaneous reaction of K/K2CO3/CO2 with carbon.[20, 34–37] These TEM analyses were consistent with the small-angle XRD and N2adsorption/desorption isotherm measurements. Chemical activation can result in a well-developed microstructure and a high specific surface area, whilst excess activation could also decrease the degree of graphitization.[4] To investigate the degree of graphitization in the as-prepared samples, wide-angle XRD and Raman spectroscopy were employed (Figure 5). As shown in Figure 5 a, all of the samples displayed a broadened diffraction peak at about 238 and a weak diffraction peak at about 438, which corresponded to the (002) and (100) diffraction peaks of a graphitictype lattice, respectively, thus indicating an amorphous structure. After KOH activation, the intensity of the peaks became lower and the full-width-at-half-maximum of the peaks broadened, thus confirming that the crystalline structures were markedly destroyed during the chemical etching process.[34] On increasing the KOH/C-FDU-16 weight ratio from 2:1 to 8:1, these changes became more clear.
Chem. Asian J. 2014, 9, 2789 – 2797
2792
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
structure, large specific surface area, and increased micropore proportion, these hierarchical activated mesoporous carbons were expected to exhibit outstanding electrochemical capacitive performance as electrode materials for supercapacitors.
which was electrochemically accessible, was needed to obtain high specific capacitance. The other AMCs exhibited quasi-rectangular voltammetry curves, thus implying that they were excellent candidates as electrode materials for EDLCs. Notably, an unapparent broadened reversible hump was also observed at about 0.3 V for these AMCs, which might be due to redox reactions related to the oxygen-containing functional groups.[4, 23, 42] In general, a small amount of oxygen-containing functional groups are inevitably introduced during the KOH activation and they can provide pseudo-capacitance through redox reactions and improve the wettability of carbon materials to the electrolyte solution, thus leading to improved electrochemical capacitive performance.[43, 44] It is generally believed that electrochemical capacitance is proportional to the area under the CV curve and a higher current response results in a larger area under the CV curve.[45] Clearly, the current response of AMC-6 was the largest, thus indicating that it had the largest electrochemical capacitance. The specific capacitance of AMC-6 was up to 200 F g1 at a scan rate of 10 mV s1, higher than those of C-FDU-16 (12 F g1), AMC-2 (77 F g1), AMC-4 (155 F g1), and AMC-8 (96 F g1). Figure 6 b shows galvanostatic charge–discharge curves of all of the samples at a current density of 0.1 A g1 within the potential window of 0.1 to 0.7 V versus Ag/AgCl. C-FDU-16 and AMC-2 both exhibited short charge–discharge time with different shapes to the typical triangular shape of the EDLCs, thus implying poor electrical app:ds:double-layer capacitive performance. The charge–discharge curves of other AMCs all showed triangular shapes with good symmetry, which indicated good electrical double-layer capacitive behavior and electrochemical reversibility. Clearly, AMC-6 exhibited the longest charge–discharge time, thus indicating that it had the largest specific capacitance. The specific capacitance of AMC-6 was up to 260 F g1 at a current density of 0.1 A g1, higher than those of C-FDU-16 (32 F g1), AMC-2 (143 F g1), AMC-4 (216 F g1), and AMC-8 (115 F g1). Notably, the specific capacitance of the nitricacid-modified C-FDU-16 reported by Cai et al. was only 219 F g1 at a current density of 0.1 A g1,[23] lower than that of AMC-6. The main reason for the highest specific capacitance of AMC-6 may be due to its high specific surface area and optimized micro-mesopore structure (an ordered mesopore with a large number of micro-mesopores in the wall connecting neighboring ordered mesopore channels; Figure 3, Figure 4, and Table 1), which can increase the effective specific surface area for charge storage and improve ion penetration and transport. The specific surface area of AMC-8 was relatively large; however, its specific capacitance was undesirable. This result may be due to the excessive KOH activation, which led to collapse of the ordered mesopore structure (Figure 2) and, hence, was disadvantageous for ion penetration and transportation. These results further demonstrate that a hierarchical pore structure with interconnected pore channels is beneficial for charge storage and that higher capacitance can be obtained with a larger effective surface area.
Electrochemical Performance The electrochemical performance of the as-obtained carbon materials were characterized by cyclic voltammetry (CV), galvanostatic charge–discharge measurements, and electrochemical impedance spectroscopy (EIS) in 1 m H2SO4 aqueous solution at ambient temperature. Figure 6 a shows the CV curves of C-FDU-16 and AMCs at a scan rate of 10 mV s1 within the potential window of 0.1 to 0.7 V versus Ag/AgCl. C-FDU-16 and AMC-2 both exhibited small shapes that were different to the typical rectangular shape of the EDLCs, thus indicating poor capacitive characteristics and low capacitance. This result was ascribed to a relatively low specific surface area, complicated pore channels, and disconnection between the neighboring channels, which led to a low effective specific surface area for the formation of an electrochemical double-layer. The basis for storage of the electrical energy for supercapacitors was the separation of charge at the electrode/electrolyte interface.[30] Therefore, a large effective specific surface area,
Figure 6. a) Cyclic voltammetry curves at a scan rate of 10 mV s1 and b) galvanostatic charge–discharge curves at a current density of 0.1 A g1 of C-FDU-16 and the synthesized AMCs.
Chem. Asian J. 2014, 9, 2789 – 2797
Shiyou Guan et al.
2793
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
Shiyou Guan et al.
Figure 7 a shows CV curves of AMC-6 at various scan rates from 2 to 200 mV s1. As the scan rate increased, the CV curves gradually became depressed whilst maintaining a quasi-rectangular shape with slight distortions at high scan rates. Moreover, the redox peaks became less pronounced as the scan rate increased, which was attributed to the fact that the rate of the redox reaction could not match the rapid increase in scan rate. The galvanostatic charge–discharge curves of AMC-6 at different current densities from 0.1 to 20 A g1 are shown in Figure 7 b. All of the galvanostatic charge–discharge curves showed quasi-triangular shapes with good symmetry, even at a current density of 20 A g1, thus indicating excellent capacitive performance and electrochemical reversibility, even at high current densities. Figure 7 c shows a plot of specific capacitance versus scan rate for all of the samples. On increasing the scan rate, the specific capacitance of all of the samples decreased gradually. However, AMC-6 exhibited excellent capacitance at various scan rates. At a scan rate of 2 mV s1, the specific capacitance of AMC-6 was up to 224 F g1, and a high specific capacitance of 107 F g1 was obtained at a scan rate of 200 mV s1, which resulted in a capacitance-retention ratio of 48 %, thus confirming good rate capability. This result was due to a unique 3D interconnected pore structure, which led to fast ion transport within the mesopores and a short diffusion distance from the mesopores to the micropores.[46, 47] Figure 7 d shows a plot of specific capacitance
versus current density for all of the samples. On increasing the current density from 0.1 to 20 A g1, the specific capacitance of all of the samples gradually decreased, which was attributed to the fact that the charge diffusion could not match the rapid increase in current density. The specific capacitance of AMC-6 decreased from 260 F g1 at a current density of 0.1 A g1 to 191 F g1 at a current density of 5 A g1, which corresponded to a capacitanceretention ratio of 73 %, clearly higher than that (49 %) of the nitric-acidmodified C-FDU-16 reported by Cai et al.[23] Even at a current density of 20 A g1, the specific capacitance was still up to 163 F g1, thus indicating a capacitance-retention ratio of 63 %. The excellent rate capability of AMC-6 was due to the capacitance being mainly derived from charge separation at the electrode/electrolyte interface and this surfacestorage mechanism allowed very fast energy uptake and delivery, thus leading to better power performance.[48] However, the introduction of oxygen-containing groups through treatment with nitric acid resulted in a decrease in electric conductivity, which was detrimental to the rate capability. Moreover, the rate capability of pseudo-capacitance was inferior to that of double-layer capacitance.[49] Figure 8 a shows Nyquist plots that were derived from EIS analysis of the as-obtained samples. These plots all presented similarly shaped Nyquist plots that were composed of three regions: an almost vertical line in the low-frequency region, which was related to the capacitive behavior; a sloped line in the medium-frequency region, which was related to diffusion resistance; and a semi-circular line in the highfrequency region, which was related to the interfacial chargetransfer resistance.[49] Clearly, compared with C-FDU-16, the AMCs showed a much straighter line in the low-frequency region, a shortened sloped line in the medium-frequency region, and a much smaller semi-circular line in the highfrequency region, thus indicating significantly improved electrochemical capacitive performance, low diffusion resistance, and low interfacial chargetransfer resistance, respectively. In addition, the diameter of the semi-circular line for AMC-6 was smaller than those for AMC-2, AMC-4, and AMC-8, thus indicating that AMC-6 had the lowest impedance at the electrode/electrolyte interface and the fastest ion movement Figure 7. a) Cyclic voltammetry curves at different scan rates and b) galvanostatic charge–discharge curves at 1 inside the pores.[5, 50, 51] The different current densities of AMC-6; plots of c) specific capacitance versus scan rate (2–200 mV s ) and 1 length of the almostvertical line d) specific capacitance versus current density (0.1–20 A g ) of C-FDU-16 and the synthesized AMCs.
Chem. Asian J. 2014, 9, 2789 – 2797
2794
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Shiyou Guan et al.
www.chemasianj.org
Figure 9. a) Cycling stability of AMC-6 at a current density of 2 A g1; b) Nyquist plots of AMC-6 before and after 10 000 cycles.
Figure 8. a) Nyquist plots and b) plots of specific capacitance versus frequency of C-FDU-16 and the synthesized AMCs.
in the low-frequency region corresponds to the specific capacitance of the carbon electrodes.[1] The shorter the length, the higher the specific capacitance. Clearly, AMC-6 possessed the shortest length of the almostvertical line in the low-frequency region, thus indicating that it had the highest specific capacitance, consistent with the results of the CV and galvanostatic charge–discharge analyses. Figure 8 b shows a plot of specific capacitance as a function of frequency for all of the samples. At a low frequency of about 0.04 Hz, the specific capacitance was 150 F g1 for AMC-6, higher than those of C-FDU-16 (13 F g1), AMC-2 (81 F g1), AMC-4 (122 F g1), and AMC-8 (79 F g1). Clearly, the specific capacitance gradually decreased as the frequency increased in the low-frequency region and the capacitance tended towards zero in the high-frequency region. This result was because ion penetration inside the pores became difficult at a highly alternating current. Cycling stability is a key factor in the practical application of electrode materials in supercapacitors. Excellent cycling stability is needed for supercapacitor operation. To evaluate the cycling stability of AMC-6, galvanostatic charge–discharge cycling was performed at a current density of 2 A g1 within the potential window of 0.1 to 0.7 V versus Ag/ AgCl (Figure 9 a). The specific capacitance of AMC-6 was quite stable over 10 000 cycles, thus indicating excellent cycling stability. A slight increase in specific capacitance
Chem. Asian J. 2014, 9, 2789 – 2797
(about 3 %) after 10 000 cycles was observed, which may be due to continued activation of the electroactive material during the charge–discharge process. Figure 9 b shows Nyquist plots of AMC-6 before and after 10 000 cycles. In this case, two Nyquist plots were almost overlapped, thus indicating that the impedance of AMC-6 before and after 10 000 cycles were almost equal. This result demonstrated that no deterioration of the electrode materials occurred during the galvanostatic charge–discharge cycling.
Conclusion In summary, a series of hierarchical activated mesoporous carbon materials with a high specific surface area and an improved micro-mesopore structure were synthesized by the activation of highly ordered, body-centered-cubic mesoporous phenolic-resin-based carbon with KOH. Of these activated mesoporous carbon materials, owing to a large effective specific surface area and an optimized hierarchical micro-mesopore structure, AMC-6 exhibited the best electrochemical capacitive performance, including a high specific capacitance (260 F g1 at a current density of 0.1 A g1), excellent rate capability (163 F g1 retained at a current density of 20 A g1), and good cycling stability (no capacitance loss over 10 000 cycles), thus presenting great potential for
2795
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Shiyou Guan et al.
www.chemasianj.org
Acknowledgements
practical application as a high-performance electrode material for supercapacitors.
The authors are grateful for the financial support from the National Natural Science Foundation of China (No. 21274043).
Experimental Section [1] C. Lei, N. Amini, F. Markoulidis, P. Wilson, S. Tennison, C. Lekakou, J. Mater. Chem. A 2013, 1, 6037 – 6042. [2] Y. P. Zhai, Y. Q. Dou, D. Y. Zhao, P. F. Fulvio, R. T. Mayes, S. Dai, Adv. Mater. 2011, 23, 4828 – 4850. [3] W. J. Qian, F. X. Sun, Y. H. Xu, L. H. Qiu, C. H. Liu, S. D. Wang, F. Yan, Energy Environ. Sci. 2014, 7, 379 – 386. [4] M. Zhou, F. Pu, Z. Wang, S. Y. Guan, Carbon 2014, 68, 185 – 194. [5] Y. R. Liang, F. X. Liang, H. Zhong, Z. H. Li, R. W. Fu, D. C. Wu, J. Mater. Chem. A 2013, 1, 7000 – 7005. [6] H. Zhu, X. L. Wang, X. X. Liu, X. R. Yang, Adv. Mater. 2012, 24, 6524 – 6529. [7] G. Wang, L. Zhang, J. Zhang, Chem. Soc. Rev. 2012, 41, 797 – 828. [8] G. A. Snook, P. Kao, A. S. Best, J. Power Sources 2011, 196, 1 – 12. [9] Z. S. Wu, D. W. Wang, W. C. Ren, J. P. Zhao, G. M. Zhou, F. Li, H. M. Cheng, Adv. Funct. Mater. 2010, 20, 3595 – 3602. [10] S. Chen, J. Zhu, X. Wu, Q. Han, X. Wang, ACS Nano 2010, 4, 2822 – 2830. [11] A. Sumboja, C. Y. Foo, X. Wang, P. S. Lee, Adv. Mater. 2013, 25, 2809 – 2815. [12] X. Dong, X. Wang, L. Wang, H. Song, X. Li, L. Wang, M. B. ChanPark, C. M. Li, P. Chen, Carbon 2012, 50, 4865 – 4870. [13] S. Kondrat, C. R. Perez, V. Presser, Y. Gogotsi, A. A. Kornyshev, Energy Environ. Sci. 2012, 5, 6474 – 6479. [14] B. Xu, H. Duan, M. Chu, G. P. Cao, Y. S. Yang, J. Mater. Chem. A 2013, 1, 4565 – 4570. [15] Y. Y. Lv, F. Zhang, Y. Q. Dou, Y. P. Zhai, J. X. Wang, H. J. Liu, Y. Y. Xia, B. Tu, D. Y. Zhao, J. Mater. Chem. 2012, 22, 93 – 99. [16] A. Burke, J. Power Sources 2000, 91, 37 – 50. [17] E. Frackowiak, F. Beguin, Carbon 2001, 39, 937 – 950. [18] X. Q. Wang, J. S. Lee, C. Tsouris, D. W. DePaoli, S. Dai, J. Mater. Chem. 2010, 20, 4602 – 4608. [19] H. J. Ahn, J. H. Lee, Y. Jeong, J. H. Lee, C. S. Chi, H. J. Oh, Mater. Sci. Eng. A 2007, 449, 841 – 845. [20] L. Zou, G. Morris, D. Qi, Desalination 2008, 225, 329 – 340. [21] P. Xu, J. E. Drewes, D. Heil, G. Wang, Water Res. 2008, 42, 2605 – 2617. [22] M. Zhou, F. Pu, H. Chen, Z. Wang, H. Y. Zhang, S. Y. Guan, New J. Chem. 2013, 37, 4148 – 4155. [23] T. W. Cai, M. Zhou, D. Y. Ren, G. S. Han, S. Y. Guan, J. Power Sources 2013, 231, 197 – 202. [24] Y. Meng, D. Gu, F. Q. Zhang, Y. F. Shi, L. Cheng, D. Feng, Z. X. Wu, Z. X. Chen, Y. Wan, A. Stein, D. Y. Zhao, Chem. Mater. 2006, 18, 4447 – 4464. [25] Y. Meng, D. Gu, F. Q. Zhang, Y. F. Shi, H. H. Yang, Z. Li, C. Z. Yu, B. Tu, D. Y. Zhao, Angew. Chem. Int. Ed. 2005, 44, 7053 – 7059; Angew. Chem. 2005, 117, 7215 – 7221. [26] Y. Huang, H. Q. Cai, T. Yu, X. L. Sun, B. Tu, D. Y. Zhao, Chem. Asian J. 2007, 2, 1282 – 1289. [27] F. Q. Zhang, Y. Meng, D. Gu, Y. Yan, Z. X. Chen, B. Tu, D. Y. Zhao, Chem. Mater. 2006, 18, 5279 – 5288. [28] Y. W. Zhu, S. Murali, M. D. Stoller, K. J. Ganesh, W. W. Cai, P. J. Ferreira, A. Pirkle, R. M. Wallace, K. A. Cychosz, M. Thommes, D. Su, E. A. Stach, R. S. Ruoff, Science 2011, 332, 1537 – 1541. [29] A. G. Pandolfo, A. F. Hollenkamp, J. Power Sources 2006, 157, 11 – 27. [30] D. Y. Qu, H. Shi, J. Power Sources 1998, 74, 99 – 107. [31] D. Y. Zhao, Q. S. Huo, J. L. Feng, B. F. Chmelka, G. D. Stucky, J. Am. Chem. Soc. 1998, 120, 6024 – 6036. [32] J. R. Matos, M. Kruk, L. P. Mercuri, M. Jaroniec, L. Zhao, T. Kamiyama, O. Terasaki, T. J. Pinnavaia, Y. Liu, J. Am. Chem. Soc. 2003, 125, 821 – 829. [33] X. Y. Chen, C. Chen, Z. J. Zhang, D. H. Xie, X. Deng, J. W. Liu, J. Power Sources 2013, 230, 50 – 58.
Sample Preparation The synthetic process of C-FDU-16 was similar to that reported by Zhao and co-workers.[23–27] C-FDU-16 was activated by KOH with different KOH/C-FDU-16 weight ratios (2:1, 4:1, 6:1, and 8:1). Taking the activation with a KOH/C-FDU-16 weight ratio of 6:1 as an example, C-FDU16 (0.3 g) was impregnated with KOH solution (1.8 g KOH in 11.25 mL EtOH), followed by evaporation at 65 8C under stirring in a flow of N2. The activation process was performed in a tubular furnace at 700 8C for 2 h under a flow of N2 (heating rate: 3 8C min1). Then, the obtained samples were washed with 0.2 m HCl and deionized water until the pH value of the filtrate was approximately 7. The final AMCs were obtained after drying at 80 8C for 24 h. Sample Characterization Powder XRD patterns were collected on a D/MAX 2550VB/PC diffractometer (Rigaku, Japan) that was equipped with CuKa radiation (l = 0.154 nm). The porous properties of the samples were determined from N2-adsorption/desorption isotherms and measured at 77 K on an ASAP 2010 (Micromeritics, USA) instrument. Prior to the measurements, the samples were degassed at 180 8C for 8 h. The specific surface area was determined by using BET method and the poresize distributions were derived from the adsorption branches of the isotherms by using BJH theory. The total pore volume was calculated from the amount of N2 adsorbed at a relative pressure of 0.99. TEM images were recorded on a JEM-2100 microscope (JEOL, Japan). Raman spectra were recorded on an InVia Raman microscope (Renishaw, UK) with excitation from a 514.5 nm diode laser. Electrochemical Measurements The electrochemical performance was measured in a three-electrode system in 1 m aqueous H2SO4 solution. Platinum foil and Ag/AgCl electrode were used as the counter and reference electrodes, respectively. The working electrodes were prepared by pressing a mixture of the activated material (80 wt %; C-FDU-16 and AMCs), acetylene black (10 wt %), and polytetrafluoroethylene binder (10 wt %) onto titanium mesh under 10 MPa pressure. Each working electrode contained 5 mg of the electrode material, with a size of about 1 cm2. CV and galvanostatic charge–discharge measurements were performed within the potential window of 0.1 to 0.7 V. EIS measurements were recorded from 100 kHz to 0.005 Hz, with an alternating current amplitude of 10 mV. The specific capacitance as calculated from the CV data was determined according to Equation (1), where Cm is the specific capacitance (F g1), I is the response current (A), s is the scan rate (V s1), DU is the potential window (V), and m is the mass of the active material within the working electrode (g). Cm ¼
Z
IdU=ð2 s m DUÞ
ð1Þ
The specific capacitance as calculated from the galvanostatic charge–discharge measurements was determined according to Equation (2), where I is the galvanostatic charge–discharge current (A) and Dt is the discharging time (s). Cm ¼ I Dt=ðm DUÞ
ð2Þ
The specific capacitance as calculated from the EIS measurements was determined according to Equation (3), where f is the frequency (Hz) and Zim is the imaginary part of the impedance (Ohm). Cm ¼ 1=ð2p f m Zim Þ
Chem. Asian J. 2014, 9, 2789 – 2797
ð3Þ
2796
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
[34] Q. Wang, Q. Cao, X. Y. Wang, B. Jing, H. Kuang, L. Zhou, J. Power Sources 2013, 225, 101 – 107. [35] Y. P. Guo, S. F. Yang, K. Yu, J. Z. Zhao, Z. C. Wang, H. D. Xu, Mater. Chem. Phys. 2002, 74, 320 – 323. [36] T. Otowa, Y. Nojima, T. Miyazaki, Carbon 1997, 35, 1315 – 1319. [37] M. A. Lillo-Rdenas, J. Juan-Juan, D. Cazorla-Amoros, A. LinaresSolano, Carbon 2004, 42, 1371 – 1375. [38] A. C. Ferrari, J. C. Meyer, V. Scardaci, C. Casiraghi, M. Lazzeri, F. Mauri, S. Piscanec, D. Jiang, K. S. Novoselov, S. Roth, A. K. Geim, Phys. Rev. Lett. 2006, 97, 187401 – 187404. [39] M. Zhou, T. W. Cai, F. Pu, H. Chen, Z. Wang, H. Y. Zhang, S. Y. Guan, ACS Appl. Mater. Interfaces 2013, 5, 3449 – 3455. [40] M. Zhou, F. Pu, Z. Wang, T. W. Cai, H. Chen, H. Y. Zhang, S. Y. Guan, Phys. Chem. Chem. Phys. 2013, 15, 11394 – 11401. [41] X. C. Chen, K. Kierzek, K. Wenelska, K. Cendrowski, J. Gong, X. Wen, T. Tang, P. K. Chu, E. Mijowska, Chem. Asian J. 2013, 8, 2627 – 2633. [42] T. W. Cai, M. Zhou, G. S. Han, S. Y. Guan, J. Power Sources 2013, 241, 6 – 11.
Chem. Asian J. 2014, 9, 2789 – 2797
[43] [44] [45] [46] [47] [48] [49] [50] [51]
2797
Shiyou Guan et al.
J. C. Wang, S. Kaskel, J. Mater. Chem. 2012, 22, 23710 – 23725. D. Y. Qu, J. Power Sources 2002, 109, 403 – 411. Y. Li, X. Zhao, P. Yu, Q. Zhang, Langmuir 2013, 29, 493 – 500. K. S. Xia, Q. M. Gao, J. H. Jiang, J. Hu, Carbon 2008, 46, 1718 – 1726. W. T. Huang, H. Zhang, Y. Q. Huang, W. K. Wang, S. C. Wei, Carbon 2011, 49, 838 – 843. P. Simon, Y. Gogotsi, Nat. Mater. 2008, 7, 845 – 854. B. Xu, S. S. Hou, G. P. Cao, F. Wu, Y. S. Yang, J. Mater. Chem. 2012, 22, 19088 – 19093. L. Yang, S. Cheng, Y. Ding, X. Zhu, Z. L. Wang, M. Liu, Nano Lett. 2012, 12, 321 – 325. W. F. Zhang, Z. H. Huang, G. P. Cao, F. Y. Kang, Y. S. Yang, J. Power Sources 2012, 204, 230 – 235. Received: April 7, 2014 Revised: May 15, 2014 Published online: August 5, 2014
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
