Cesium carborane as an unconventional non-aqueous electrolyte salt for electrochemical capacitors

Electrochimica Acta 125 (2014) 482–487

Contents lists available at ScienceDirect

Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta

Cesium carborane as an unconventional non-aqueous electrolyte salt for electrochemical capacitors Ann Laheäär ∗ , Alar Jänes, Enn Lust Institute of Chemistry, University of Tartu, 14a Ravila Street, 50411 Tartu, Estonia

a r t i c l e

i n f o

Article history: Received 1 January 2014 Received in revised form 27 January 2014 Accepted 27 January 2014 Available online 8 February 2014 Keywords: Cesium carborane Specific adsorption Electrochemical capacitor Carbide derived carbon Non-aqueous electrolyte

a b s t r a c t A novel electrolyte salt for non-aqueous electrochemical capacitors, cesium carborane, was electrochemically characterized in two organic solvent systems at different concentrations–acetonitrile and a 1:1 mixture of ethylene carbonate and dimethyl carbonate. It was found that Cs+ ion specific adsorption takes place, being dependent on the electrolyte concentration and on the solvent used. Stronger specific adsorption effect with partial charge transfer was observed from the acetonitrile based electrolytes, where the cesium salt has low solubility. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction The enhancement in the power and energy performance of the electrochemical capacitors is mainly focused on developing new electrode materials of high capacitance, such as carbon nanostructures of different surface chemistry and texture, redox active metal oxide or electrically conductive polymer based materials, intercalation compounds, electrochemical hydrogen storage materials, etc. [1,2]. However, the chemical composition and surface morphology (i.e., wetting properties) of separators [3] as well as the properties of electrolytes, e.g., improved ion adsorption characteristics, electrochemical stability and compatibility with electrode materials, are equally important features that have to be addressed in the development of high performance electrochemical capacitors of extended life time. The conventional non-aqueous electrochemical capacitor electrolytes, e.g., alkyl ammonium salts in acetonitrile (AN) or in propylene carbonate (PC), allow applying cell potentials up to 2.7 V while maintaining long-lasting cyclability. It is important to find alternatives of improved electrochemical stability that allow applying higher cell potentials and, if possible, to increase the capacitance of the electrode/electrolyte interface by reversible faradaic reactions or specific adsorption of the electrolyte ions.

∗ Corresponding author. Tel.: +372 737 6636, fax: +372 737 5264. E-mail address: [email protected] (A. Laheäär). 0013-4686/$ – see front matter © 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.electacta.2014.01.145

Specific adsorption of halide ions on the bismuth single crystal planes has been demonstrated with the strongest effect established for the I− ions in aqueous and non-aqueous electrolytes [4,5]. The redox-active I− based aqueous electrolytes [6–8] as well as I− containing ionic liquid mixtures [9,10] and redox-active Br− ionic liquids [11], have been thoroughly studied for carbon based electrochemical capacitor applications. Weak specific adsorption effect was also observed for ClO4 − anions in aqueous electrolyte [12]. The influence of the specific adsorption of bulky tetraalkylammonium cations on the charge-compensation mechanism in carbon micropores, together with the effect of coupling confined specific adsorption of cations and charge-induced desorption of anions, has been analyzed in Ref. [13]. Ion solvation and distortions in the solvation shell (partial removal) during the adsorption process at micro-meso-porous carbon electrodes depend on the combination of solvents and ions, and thus have direct effect on the capacitance of the electrode/electrolyte interface [14]. In this study novel non-aqueous electrolytes based on a superacid salt cesium carborane (CsCB11 H12 ) are tested in order to describe the specific adsorption effects in relation to the electrolyte solvent used and to establish the electrochemical behavior of a large spherical carborane anion. For better comparison, electrolytes of different concentration were prepared in the traditional electrochemical capacitor solvent AN, and in an environmentally friendlier solvent system of ethylene carbonate (EC) and dimethyl carbonate (DMC), commonly applied in Li-ion batteries [15], but also tested for carbon based electrical double layer and hybrid electrochemical capacitors with Li- and Na-salt electrolytes [16–20].

A. Laheäär et al. / Electrochimica Acta 125 (2014) 482–487

2. Experimental Two- and three-electrode special hermetic aluminium test cells ( 2TC and  3TC , Hohsen Corp.) were assembled for the electrochemical characterization by cyclic voltammetry and electrochemical impedance spectroscopy methods (1252A Solartron frequency response analyzer and SI1287 potentiostat with a 5 mV modulation). Carbide derived carbon powder was synthesized from molybdenum carbide by the high-temperature chlorination method at 800 ◦ C, noted as C(Mo2 C), and was mixed with polytetrafluoroethylene (60% solution in H2 O (Aldrich)) to form a paste with 5 wt% binder content. Thin (∼105 ± 5 ␮m) electrode layers were rollpressed from the active material paste and ∼2 ␮m thick Al layer was deposited onto one side of the electrode layer by the magnetron sputtering method (AJA International) to reduce the ohmic potential drop (IR-drop). The C(Mo2 C) powder specific surface area was SBET ≈ 1675 m2 g−1 , calculated from the Brunauer-EmmettTeller theory, and micropore area from the t-plot method was Sm ≈ 1560 m2 g−1 with two main pore size distribution maxima located at 1.15 nm and 3.80 nm. After preparing the electrode layers from powder and the Al-deposition, less than 10% decrease was measured for both total specific surface area and microporous surface area (1535 m2 g−1 and 1480 m2 g−1 , respectively). More precise physical characterization (Raman and XRD analysis) of the partly graphitized amorphous C(Mo2 C) powder can be found in Ref. [21]. The electrochemically tested CsCB11 H12 (Strem Chemicals) electrolytes were 0.1 M and 0.2 M solutions (saturation concentration ∼0.22 M) in AN (99.9%, max 0.003% H2 O, Riedel-de Haën), and 0.3 M, 0.5 M and 0.8 M solutions (saturation concentration >0.8 M) in an equimolecular mixture of EC (Selectipur® , Merck) and DMC (> 99%, H2 O < 0.002%, Sigma-Aldrich). All test cells were assembled in a glove box (O2 and H2 O < 0.1 ppm, MBraun). The specific ionic conductivities () of the electrolyte solutions were measured using a self-prepared conductivity cell with platinum plate electrodes, for which the cell constant was determined from the measured values of the high-frequency impedance for two standard electrolyte solutions of known ionic conductivities: 0.1 M and 0.01 M KCl aqueous solutions at T = 23 ◦ C. The cross-sectional surface area of the working electrodes (WE) and counter electrodes (CE) in 3TCs was 0.28 cm2 and 2.0 cm2 , respectively, and 2.0 cm2 for both electrodes in 2TCs. The carbon loading of electrodes was ∼5.5 mg cm−2 . The reference electrode (RE) was Ag/AgCl in the same electrolyte solution (+0.2 V vs. SHE), prepared by depositing a layer of AgCl on a positively polarized Ag wire from 0.1 M HCl aqueous solution. Mesoporous polypropylene separator (Celgard® 2400) was applied in the systems with carbonate solvent mixture based electrolytes, and cellulose separator (TF4425, Nippon Kodoshi) with the acetonitrile based electrolytes for better chemical compatibility [16,17,21,22]. A simple notation system was adapted for the test systems, e.g., 0.2AN-2TC indicating a two-electrode test cell based on 0.2 M CsCB11 H12 solution in acetonitrile. All the calculation methods used have been explained in a previous publication [16].

483

measured for 0.2 M CsCB11 H12 + AN. The measured  value of 8.9 mS cm−1 for 0.5 M CsCB11 H12 solution in EC:DMC is rather close to the values for 1 M solutions of NaClO4 and NaPF6 in the same solvent mixture ( ≈ 9.15 mS cm−1 ), indicating that the larger Cs+ and CB11 H12 − ions are not strongly solvated in the mixture of carbonates (also in AN). The CB11 H12 − anion has the lowest nucleophilic ability even known [23], and has an estimated diameter of 1.04 nm, which is slightly smaller than the first pore size distribution maximum for the C(Mo2 C) carbon, located at 1.15 nm [24]. 3.2. Cyclic voltammetry measurements for three-electrode test cells Cyclic voltammograms (CVs) were measured in various working electrode potential regions at potential scan rates () from 0.5 to 50 mV s−1 . Nearly rectangular shape of CVs can be observed for all the systems under study in a working electrode potential (E) region from -2.0 to 1.0 V (vs. Ag/AgCl), characteristic to the nearly ideal capacitive behavior with working electrode capacitance CWE (Figs. 1 and 2). CVs measured at 1 mV s−1 for 0.2AN-3TC (Fig. 1a) show a large capacitive peak at the negative potential scan direction (cathodic scan) at E ∼ -2.2 V. Quick capacitance increase at more negative potentials (E < -2.35 V) is probably initiated by the faradaic reduction of the electrolyte and/or reduction of the water/oxygen traces in the electrolyte. Taking into account the well expressed concentration dependence of CWE , the peak at E ∼ -2.2 V could be attributed to the Cs+ cation specific adsorption on the micro-meso-porous carbon surface with partial charge transfer, as the peak capacitance is nearly two times lower in a more dilute 0.1 M CsCB11 H12 + AN electrolyte system (Fig. 1b).

3. Results and discussion 3.1. Electrolyte ionic conductivities The measured electrolyte  values are presented in Table 1, together with some electrolytes studied earlier [16–19]. Despite the low concentration, the  value for 0.1 M CsCB11 H12 electrolyte in a low-viscosity AN solvent (10.6 mS cm−1 ) is comparable to the 1 M solutions of Li- and Na-salt electrolytes with symmetric PF6 − and ClO4 − anions in a more viscous EC:DMC mixture, and higher  was

Fig. 1. Cyclic voltammograms expressed as working electrode gravimetric capacitance CWE vs. potential E for 3TCs with 0.2 M (a) and 0.1 M (b) CsCB11 H12 electrolytes in AN.

484

A. Laheäär et al. / Electrochimica Acta 125 (2014) 482–487

Table 1 Specific ionic conductivities of CsCB11 H12 electrolytes in comparison with some other previously tested electrolytes [16–19]. CsCB11 H12 electrolyte /mS cm−1 1 M electrolyte in EC:DMC /mS cm−1

NaPF6 12.29

0.1 M (AN) 10.6 LiPF6 10.54

The Cs+ desorption process is very slow as there is no evident desorption peak on the anodic scan at close proximity to the adsorption peak (Fig. 1a). In addition to the adsorption processes, there should be formation of a surface film at high negative electrode potentials. It should be noted that the intensity of the capacitive peak related to the Cs+ ions specific adsorption, decreases somewhat during the first few cycles for the 0.2AN-3TC and then the system stabilizes (Fig. 1a). However, there is no stabilization in the

NaClO4 9.18

0.2 M (AN) 19.1 LiClO4 9.15

LiB(C2 O4 )2 5.85

0.5 M (EC:DMC) 8.9 LiCF3 SO3 3.01

0.1AN-3TC test system during the first 10 cycles (Fig. 1b), which could be related to the instability of the surface film caused by the low concentration of electrolyte ions in the surface film. An alternative process to the Cs+ specific adsorption could be the trapping of Cs+ cations in the narrow micropores of the C(Mo2 C) carbon electrode, discussed for YP-17 carbon in contact with the standard tetraethylammonium tetrafluoroborate solution in propylene carbonate [25]. The cations trapped at high electrode charge density cannot be desorbed in the native potential domain of the negatively charged carbon electrode. The potential of zero charge (pzc) is noted in the Figs. 1a and 1b [26]. In situ XPS or Fourier transmission infrared measurements should be made for more accurate characterization of the surface processes under discussion. In the test systems with the EC:DMC solvent mixture based CsCB11 H12 electrolytes, surprisingly the kind of peak observed in AN did not appear (Fig. 2). For 0.8EC:DMC-3TC, steep increase in capacitance was established at E < -2.0 V and a peculiar reduction peak occurred on the reversed potential scan direction at negative electrode potentials (Fig. 2a, cycles 1 and 2). The observed peak declined during the cycling of the working electrode potential between 1.0 V and -2.5 V, and the nearly rectangular shape of CVs from the 7th cycle indicates system stabilization (Fig. 2a). Thus, during the potential cycling, a stable surface film is formed on the electrode surface. In case of 0.5EC:DMC-3TC, the stabilization of the system takes even more time (Fig. 2b). Therefore, some irreversible reduction processes take place in addition to the very slow specific adsorption of Cs+ ions. There is no remarkable increase in capacitance at negative potentials in the 0.3EC:DMC3TC (Fig. 2c). We suggest that the increase in capacitance for the 0.8 M and 0.5 M + EC:DMC electrolyte systems could still be related to the adsorption of Cs+ cations. However, the surface film formation process, and charge transfer process kinetics and mechanism in this solvent mixture, differ from AN based electrolytes. The specific adsorption effect is in correlation with the solubility of CsCB11 H12 salt. Thus, weaker and non-equilibrium adsorption effect can be seen from the EC:DMC solvent system, where the Cs-salt has higher solubility compared to AN. Therefore, stronger solvent-ion interactions take place in EC:DMC, decreasing the extent of the partial charge transfer step during Cs+ ion adsorption on the carbon electrode surface. When extending the cathodic potential limit to more negative values, noticeable current density (and capacitance) increase can be observed on the reverse scan at more positive potentials in Fig. 1a, especially after cycling up to the Cs+ specific adsorption region (E < -2.0 V). The same trend was observed for all the systems under study and is explained by the oxidation processes taking place with the surface-bound Cs+ species.

3.3. Cyclic voltammetry measurements for two-electrode test cells

Fig. 2. Cyclic voltammograms expressed as working electrode gravimetric capacitance CWE vs. potential E for 3TCs with 0.8 M (a), 0.5 M (b) and 0.3 M (c) CsCB11 H12 electrolytes in EC:DMC.

CVs for the 2TCs were measured at various potential scan rates up to cell potential E ≤ 3.6 V. Nearly ideal capacitive behaviour was established for 0.2AN-2TC up to E = 3.2 V (Fig. 3a). Interestingly, during the first CV scans with maximum E higher than 3.0 V, an effect similar to the so-called electrolyte starvation took place [14,27], demonstrating a decrease in the gravimetric capacitance (Cg ) at E > 3.0 V. This kind of behavior occurred for the AN

A. Laheäär et al. / Electrochimica Acta 125 (2014) 482–487

485

Fig. 3. Cyclic voltammograms expressed as gravimetric capacitance Cg vs. cell potential E for 2TCs with CsCB11 H12 electrolytes: 0.2 M in AN (a) and 0.8 M in EC:DMC (b).

electrolyte based systems and for the 0.3EC:DMC-2TC, i.e., for the systems with rather low electrolyte concentration. However, after several CVs and impedance measurements (i.e., continuous system cycling), this  starvation-like behaviour was no longer observed. As the amount of ions does not change during the consecutive charging/discharging cycles, an explanation for this kind of phenomena could be the so-called pore opening or electrowetting process, i.e., some inaccessible parts of the electrode get accessible to the electrolyte during the potential cycling. No  starvation-like behaviour, but rather typical behaviour for an ideally polarizable interface was observed for the 0.8EC:DMC-2TC (Fig. 3b), where the increase in Cg at E > 2.5 V and  = 1 mV s−1 could be explained by the slow Cs+ ion specific adsorption processes on the C(Mo2 C) electrode.

Fig. 4. Impedance spectra (a) and frequency dependencies of the imaginary part of impedance -Z” (b) and imaginary part of capacitance C” (c) for 3TC with 0.2 M CsCB11 H12 electrolyte in AN.

3.4. Impedance spectroscopy data for three-electrode test cells The Nyquist plots (i.e., Z , Z -plots) were measured in an ac frequency (f) range from 1·10−3 to 3·105 Hz for both 3TCs and 2TCs. The high-frequency series resistance (Rs ) was nearly 4 times lower for 0.2AN-3TC (Fig. 4a) than for 0.8EC:DMC-3TC (Fig. 5a), indicating the much lower resistivity of the low-viscosity AN solvent based electrolytes, despite the much lower salt concentration. The depressed semi-circles at high f are characteristic to the mixed kinetic processes (partial charge transfer and adsorption) on the open surface of electrodes [17,26,28]. Much wider semi-circles were observed for the 0.8EC:DMC-3TC, thus, giving a higher total polarization resistance for this system (Figs. 4a and 5a). The finite-length diffusion limited process can be observed in Fig. 4a for 0.2AN-3TC with the Nyquist plot slope ≤ -45◦ at f from ∼0.5 to 100 Hz [26,28–30].

Complex mixed kinetic processes take place in the middle f range for the EC:DMC solvent based system (Fig. 5a). In the low-frequency part of the Nyquist plots, the deviation from the ideal capacitive behaviour, indicated by a vertical shape of the plots, is more pronounced at the extreme cathodic or anodic working electrode potentials (Figs. 4a and 5a). At the potential limits, specific adsorption of Cs+ (or CB11 H12 − ) occurs or the electrolyte electrochemical stability limits are reached, and the system characteristics (Nyquist plot shape) are determined by mixed kinetic processes. Parasitic faradaic charge transfer processes such as oxidation of the carbon surface functionalities and/or decomposition of the electrolyte start to take place at E ≥ 1.2 V, characterized by the slanted Nyquist plot shape at low frequencies (Fig. 4a) and

486

A. Laheäär et al. / Electrochimica Acta 125 (2014) 482–487

especially by the horizontal plateau of the imaginary part of impedance (-Z”) vs. f plot within the region 0.01 < f < 1 Hz at E = 1.2 V (vs. Ag/AgCl) (Fig. 4b). The slope of the low-frequency part of the -Z” vs. f plot for 0.8EC:DMC-3TC is nearly -1.0 at E = -0.6 V, indicating ideal capacitive behavior, and less than -1.0 at the more positive and negative potentials, especially in the potential region of Cs+ specific adsorption, E ≤ -2.2 V (Fig. 5b). For 0.2AN-3TC, the slope of -Z” vs. f plots is less than -1.0 at all potentials studied (Fig. 4b), and the influence of Cs+ specific adsorption is even more pronounced at low f. The characteristic time constants ( R ) were calculated at different applied E from the frequency of the maxima (fmax ) on the imaginary part of impedance (C”) plots according to  R =1/(2fmax ) [31]. The smallest  R values were calculated at around E = -0.6 V,

but higher values were established at the more positive and negative potentials (Figs. 4c and 5c). According to the very small increase in the  R values at E > -0.6 V, probably anions do not adsorb specifically in the studied potential region. However, the much longer  R at high negative potentials E < -0.6 V indicate a slow cation specific adsorption effect. The  R values were an order of magnitude higher for the binary carbonate mixture based electrolytes, which suggests that the specific adsorption peak might not be seen on the CVs with the 1 mV s−1 potential scan rate (Fig. 2). Longer  R were obtained for the more dilute 0.1 M AN based electrolyte, compared to the 0.2 M one, which is explained by the increase in the Debye screening length. However, the opposite effect was observed for the EC:DMC based electrolytes, where the  R were shorter for the electrolytes of lower concentration (not shown on the figures). This suggests possible ion pairing at higher concentrations, initiating the complicated mass transfer processes within the micro-mesoporous C(Mo2 C) structure. 3.5. Impedance spectroscopy data for two-electrode test cells The phase angle () vs. f plots for the 2TCs show that at E = 1.5 V nearly ideal capacitive behaviour ( values nearly -90◦ ) is reached at low ac frequency (Fig. 6a). The slightly lower  values at E = 3.0 V are in correlation with the conclusions made based on the analysis of the Nyquist plots for 3TCs, i.e., mixed kinetic processes occur at/inside the more strongly polarized electrodes. The ideal capacitive behaviour is reached at somewhat higher f in 0.2AN-2TC due to the lower viscosity of the AN based electrolyte. The series capacitance values per one electrode (Cs ) for the 2TCs were calculated from the impedance data [31], which scale up to a limiting value (adsorption equilibrium) when f → 0. The highest Cs values were obtained for 0.8EC:DMC-2TC, i.e., corresponding to the highest electrolyte concentration (Fig. 6b). Practically equal

Fig. 5. Impedance spectra (a) and frequency dependencies of the imaginary part of impedance -Z” (b) and imaginary part of capacitance C” (c) for 3TC with 0.8 M CsCB11 H12 electrolyte in EC:DMC.

Fig. 6. Phase angle  (a) and gravimetric series capacitance Cs (b) dependencies on ac frequency at cell potentials of 1.5 V and 3.0 V for different 2TCs noted in the figure.

A. Laheäär et al. / Electrochimica Acta 125 (2014) 482–487

moderate Cs values ∼50 F g−1 were established for 0.3EC:DMC-2TC and 0.2AN-2TC. The maximum specific energy (Emax ) and specific power (Pmax ) values were calculated at E = 3.0 V [29,31,32]. The highest Emax = 42 Wh kg−1 was calculated for the 0.8EC:DMC-2TC having the highest Cs (∼18 Wh kg−1 for the other two systems characterized in Fig. 6), and the highest Pmax = 110 kW kg−1 was established for the 0.2AN-2TC having the lowest Rs . It should be noted that the calculated Emax is not based on the purely electrical double layer capacitance, but there is some contribution from the specific adsorption of electrolyte ions, expressed as the increase in Cs at f ≤ 0.01 Hz (Fig. 6b). 4. Conclusions Cesium carborane (CsCB11 H12 ) solutions of different concentrations in acetonitrile (AN) or in a mixture of ethylene carbonate and dimethyl carbonate (EC:DMC) have been studied for the electrochemical capacitor electrolyte application. Occurrence of the specific adsorption of Cs+ ions was established, whereas the extent of partial charge transfer during this process is related to the solubility of the salt, i.e., the interaction forces between the salt ions and the solvent applied. The strongest specific adsorption effect was observed in case of 0.2 M CsCB11 H12 in AN, declining with the more dilute 0.1 M electrolyte. Interesting capacitance decrease behaviour was observed when applying AN (0.1 M and 0.2 M) or EC:DMC based electrolytes of low concentration (≤0.3 M). However, after repetitive potential cycling, nearly ideal capacitive behaviour has been established due to the so-called pore opening process or electrochemical microwetting of pores. In conclusion, the superacid salt cesium carborane could be an interesting option for assembling hybrid electrochemical capacitors. Acknowledgements This work has been supported by graduate school  Functional materials and processes (European Social Fund project 1.2.0401.09-0079) and European Regional Development Fund: Estonian Materials Technology project (3.2.1101.12-0019). References [1] F. Béguin, E. Frackowiak, M. Lu (Eds.), Supercapacitors. Materials, Systems, and Applications, Weinheim, 2013. [2] B.E. Conway, Electrochemical Supercapacitors: Scientific Fundamentals and Technological Applications, Kluwer Academic/Plenum Publishers, New York, 1999. [3] S.S. Zhang, A review on the separators of liquid electrolyte Li-ion batteries, J. Power Sources 164 (2007) 351. [4] K. Lust, M. Väärtnõu, E. Lust, Adsorption of anions on bismuth single crystal plane electrodes from various solvents, J. Electroanal. Chem. 532 (2002) 303. [5] L. Siinor, K. Lust, E. Lust, Impedance study of adsorption of iodide ions at Bi(001) electrode from the aqueous solutions with constant ionic strength, J. Electroanal. Chem. 601 (2007) 39. [6] G. Lota, E. Frackowiak, Striking capacitance of carbon/iodide interface, Electrochem. Commun. 11 (2009) 87. [7] G. Lota, K. Fic, E. Frackowiak, Alkali metal iodide/carbon interface as a source of pseudocapacitance, Electrochem. Commun. 13 (2011) 38. [8] E. Frackowiak, K. Fic, M. Meller, G. Lota, Electrochemistry Serving People and Nature: High-Energy Ecocapacitors based on Redox-Active Electrolytes, Chem. Sus. Chem. 5 (2012) 1181.

487

[9] L. Siinor, C. Siimenson, K. Lust, E. Lust, Mixture of 1-ethyl-3-methylimidazolium tetrafluoroborate and 1-ethyl-3-methylimidazolium iodide: A new potential high capacitance electrolyte for EDLCs, Electrochem. Commun. 35 (2013) 5. [10] T. Tooming, T. Thomberg, L. Siinor, K. Tõnurist, A. Jänes, E. Lust, A Type High Capacitance Supercapacitor Based on Mixed Room Temperature Ionic Liquids Containing Specifically Adsorbed Iodide Anions, J. Electrochem. Soc. 161 (3) (2014) A222. [11] S. Yamazaki, T. Ito, M. Yamagata, M. Ishikawa, Non-aqueous electrochemical capacitor utilizing electrolytic redox reactions of bromide species in ionic liquid, Electrochim. Acta 86 (2012) 294. [12] R. Jäger, E. Härk, P. Möller, J. Nerut, K. Lust, E. Lust, The kinetics of electroreduction of hexaamminecobalt(III) cation on Bi planes in aqueous HClO4 solutions, J. Electroanal. Chem. 566 (2004) 217. [13] M.D. Levi, S. Sigalov, G. Salitra, D. Aurbach, J. Maier, The Effect of Specific Adsorption of Cations and Their Size on the Charge-Compensation Mechanism in Carbon Micropores: The Role of Anion Desorption, Chem. Phys. Chem. 12 (2011) 854. [14] G. Salitra, A. Soffer, L. Eliad, Y. Cohen, D. Aurbach, Carbon Electrodes for DoubleLayer Capacitors. I. Relations Between Ion and Pore Dimensions, J. Electrochem. Soc. 147 (2000) 2486. [15] G.-A. Nazri, G. Pistoia (Eds.), Lithium Batteries: Science and Technology, Kluwer Academic Publishers, New York, 2004. [16] A. Laheäär, A. Jänes, E. Lust, NaClO4 and NaPF6 as potential non-aqueous electrolyte salts for electrical double layer capacitor application, Electrochim. Acta 82 (2012) 309. [17] A. Laheäär, A. Jänes, E. Lust, Electrochemical properties of carbide-derived carbon electrodes in non-aqueous electrolytes based on different Li-salts, Electrochim. Acta 56 (2011) 9048. [18] A. Laheäär, A. Jänes, E. Lust, Electrochemical Behaviour of Carbide Derived Carbons in LiPF6 and LiCF3 SO3 Non-Aqueous Electrolytes, ECS Trans. 28 (8) (2010) 65. [19] A. Laheäär, A. Jänes, E. Lust, Lithium bis(oxalato)borate as an electrolyte for micromesoporous carbide-derived carbon based supercapacitors, J. Electroanal. Chem. 669 (2012) 67. ˜ [20] V. Khomenko, E. Raymundo-Pinero, F. Béguin, High-energy density graphite/AC capacitor in organic electrolyte, J. Power Sources 177 (2008) 643. [21] A. Jänes, T. Thomberg, H. Kurig, E. Lust, Nanoscale fine-tuning of porosity of carbide-derived carbon prepared from molybdenum carbide, Carbon 47 (2009) 23. [22] K. Tõnurist, T. Thomberg, A. Jänes, I. Kink, E. Lust, Specific performance of electrical double layer capacitors based on different separator materials in room temperature ionic liquid, Electrochem. Commun. 22 (2012) 77. [23] A.M. Spokoyny, C.W. Machan, D.J. Clingerman, M.S. Rosen, M.J. Wiester, R.D. Kennedy, C.L. Stern, A.A. Sarjeant, C.A. Mirkin, A coordination chemistry dichotomy for icosahedral carborane-based ligands, Nature Chem. 3 (2011) 590. [24] S. Körbe, P.J. Schreiber, J. Michl, Chemistry of the Carba-closo-dodecaborate(-) Anion, CB11 H12 − , Chem. Rev. 106 (2006) 5208. [25] D. Aurbach, M.D. Levi, G. Salitra, N. Levy, E. Pollak, J. Muthu, Cation Trapping in Highly Porous Carbon Electrodes for EDLC Cells, J. Electrochem. Soc. 155 (2008) A745. [26] E. Lust, G. Nurk, A. Jänes, M. Arulepp, P. Nigu, P. Möller, S. Kallip, V. Sammelselg, Electrochemical properties of nanoporous carbon electrodes in various nonaqueous electrolytes, J. Solid State Electrochem. 7 (2003) 91. [27] A. Jänes, H. Kurig, T. Romann, E. Lust, Novel doubly charged cation based electrolytes for non-aqueous supercapacitors, Electrochem. Commun. 12 (2010) 535. [28] J.R. Macdonald, W.B. Johnson, in: J.R. Macdonald (Ed.), Impedance Spectroscopy, Wiley, New York, 1987, Ch. 4. [29] M. Arulepp, L. Permann, J. Leis, A. Perkson, K. Rumma, A. Jänes, E. Lust, Influence of the solvent properties on the characteristics of a double layer capacitor, J. Power Sources 133 (2004) 320. [30] T. Thomberg, A. Jänes, E. Lust, Energy and power performance of vanadium carbide derived carbon electrode materials for supercapacitors, J. Electroanal. Chem. 630 (2009) 55. [31] P.L. Taberna, P. Simon, J.F. Fauvarque, Electrochemical Characteristics and Impedance Spectroscopy Studies of Carbon-Carbon Supercapacitors, J. Electrochem. Soc. 150 (2003) A292. [32] P. Simon, A. Burke, Nanostructured Carbons: Double-Layer Capacitance and More, Electrochem. Soc. Interface 17 (1) (2008) 38.