View other article » Take our survey »
Copyright © 2011 Elsevier B.V. All rights reserved.
Elsevier logo
Chemical Physics Letters

Volume 492, Issue 4-6, June 2010, Pages 276-280

Chemical Physics Letters journal image

The electrode potentials of the Group I alkali metals in the ionic liquid N-butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide

  1. ⁎ Corresponding author. Fax: +44 (0) 1865 275 410.
  2. Department of Chemistry, Physical and Theoretical Chemistry Laboratory, Oxford University, South Parks Road, Oxford OX1 3QZ, United Kingdom

Research highlights

  • Despite the widespread use of ionic liquids, the fundamental parameters and trends of solutes in such systems are still largely unknown.
  • The Group I metals Li, Na, K, Rb and Cs have characteristic trends, such as in Ef°, and are important with regards to energy storage, synthesis, etc.
  • The first systematic investigation of Li, Na, K, Rb and Cs in a non-haloalumiante ionic liquid has now been demonstrated.
  • All metals were successful electrodeposited from the ionic liquid N-butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide, [C4mpyrr][NTf2].
  • Simulation of the voltammetry has yielded the Group I parameters and trends in terms of Ef° (vs. Fc/Fc+), α, k0, D and ΔG°(solvation) (vs. Fc/Fc+).
  • The trends in Ef° demonstrate that the behaviour of the Group I metals in [C4mpyrr][NTf2] most closely resembles that known in propylene carbonate, as opposed to water, acetonitrile, liquid ammonia, etc.

Graphical abstract

Abstract

Trends in the formal potential of the Group I alkali metals, for the ionic liquid [C4mpyrr][NTf2] and a range of common solvents.

The redox couples M/M+ of the Group I alkali metals Lithium, Sodium, Potassium, Rubidium and Caesium have been extensively investigated in a room temperature ionic liquid (IL) and compared for the first time. Cyclic voltammetric experiments in the IL N-butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide ([C4mpyrr][NTf2]) and subsequent simulation of the data has allowed the determination of the formal potential ( vs. ferrocene/ferrocenium), standard electrochemical rate constant (k0) and transfer coefficient (α) for each couple in the group. The trend in in [C4mpyrr][NTf2] is remarkably similar to the established trend in the common battery electrolyte, propylene carbonate.

1. Introduction

The Group I alkali metals, consisting of Lithium (Li), Sodium (Na), Potassium (K), Rubidium (Rb), Caesium (Cs) and Francium (Fr) are all extremely strong reducing agents, although their relative reducing ability depends upon which solvents they are in contact with. Their electrochemistry is of importance to a number of areas, most notably with respect to batteries and energy storage [1,2].

Ionic liquids (ILs) generally possess extremely low volatility and inherent conductivity, and can possess good chemical robustness, non-flammability (with respect to common organic solvents) and wide electrochemical windows, amongst other desirable properties [3,4]. The IL N-butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide ([C4mpyrr][NTf2]), possessing the above said properties, has allowed us the investigate the electrochemical redox properties of the first five Group I metals in a non-haloaluminate IL for the first time; by necessity the radioactive and extremely rare Francium had to be excluded.

We have previously reported in-depth studies on the behaviour of the Li/Li+[5,6] and Na/Na+[7] redox couple in a range of ILs at Platinum (Pt) and Nickel (Ni) microelectrodes. No evidence of alloy formation between the Group I metals and Ni was observed, while [C4mpyrr][NTf2] gave the most defined metal deposition and stripping processes, allowing accurate determination of fundamental parameters such as , k0 and α. In this study we complete the Group I series by investigating the K/K+, Rb/Rb+ and Cs/Cs+ redox couples in [C4mpyrr][NTf2] at a Ni microelectrode. By focussing attention upon the reverse sweep of the CV, when the microelectrode has been electrodeposited with the Group I metal, the system behaves as an M/M+ half-cell, yielding fundamental values that have been accurately referenced with respect to ferrocene/ferrocenium.

2. Experimental

2.1. Instruments and chemicals

The structure of ,N-butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide [C4mpyrr][NTf2], is shown in Fig. 1, which was kindly donated by Merck KGaA. The alkali metal–[NTf2]-salts were prepared by the reaction of aqueous solutions of the appropriate metal hydroxide salt with a 5% molar excess of H[NTf2], followed by repeated dissolution in water then drying under vacuum, until pH neutral.

Figure 1

Fig. 1: The structure of [C4mpyrr][NTf2].

All electrochemical measurements were carried out using a computer-controlled μ-Autolab potentiostat (Eco-Chemie, Netherlands). Voltammetric measurements were conducted by applying a conventional two-electrode arrangement with a nickel microelectrode (50 μm diameter, fabricated in house as described previously [5-7]) used as a working electrode and a silver wire (Advent Research Materials Ltd., UK, 99.99%) served as a quasi-reference electrode. A section of a disposable micropipette tip (Eppendorf, Germany) was placed on the top of the working electrode, into which 20 μL of a [C4mpyrr][NTf2] solution containing 10 mM ferrocene and 0.1 M metal salt was added. The electrodes were confined in a T-shaped cell as described previously [8,9] and purged under a vacuum for 6 h prior to and during the experiments. This step is extremely important, as necessary to remove O2 and H2O, which will react with the alkali metal as well as reduce the electrochemical window of the IL [10-12].

In addition, 10 mM ferrocene (Fc) was introduced to every sample.

2. R. Wibowo et al. / Chemical Physics Letters Volume 492, Issue 4-6, June 2010, Pages 276-280

This was used as an internal reference, and the formal potential of the Fc/Fc+ redox couple (careful calculated after taking into account the discrepancies in diffusion coefficients between the two species, as described previously [7]) was set as 0 V.

2.2. Simulation conditions

The mathematical simulation model has been described [13] and applied [5-7,14,15] previously. Briefly, the model assumes a one electron transfer mechanism for the deposition and stripping of bulk alkali metal (M) on the electrode surface (Eq. (1)). Both the deposition and stripping processes are assumed to follow Butler–Volmer kinetics (Eqs. (2) and (3)).

(1).

(2).

(3).

where is the surface process rate constant (units of cm s−1) for the M+/M couple, α is the transfer coefficient and is the formal potential. [M]∗ is the standard concentration equal to 1 × 10−3 mol cm−3. R is the ideal gas constant (8.314 J K−1 mol−1), F is the Faraday constant (96 485 C mol−1) and T is the absolute temperature in K.

Fick’s second law is assumed to be followed for the diffusion process of species M+ approaching and escaping the electrode surface. The surface coverage of M on the electrode, ΓM, is dependant upon the rate of deposition and stripping. Deposition is M+ concentration dependant; therefore the concentration gradient had to be solved using a series of partial differential equations, as described previously [13]. The maximum distance simulated is set to be from the electrode surface, where D is the diffusion coefficient of M+ and t is time. It is considered that there is no concentration gradient beyond that distance. The rate of stripping is assumed to be independent of ΓM.

After the system of equations has been solved, the current, I, was calculated using the following equation:

(4).

3. Results and discussion

3.1. Voltammetric results

Fig. 2 displays a comparison of CVs for all five Group I alkali metals, recorded for 0.1 M of the relevant metals–[NTf2]-salt dissolved in [C4mpyrr][NTf2] and recorded at 10 mV s−1 at a Ni microelectrode.

Figure 2

Fig. 2: The cyclic voltammograms recorded for a 0.1 M solution of the relevant metal–[NTf2] salt in [C4mpyrr][NTf2] at a 50 μm diameter Ni electrode. Scan rate = 10 mV s−1.

Underpotential deposition (UPD) processes [5-7] were observed for all five Group I metals in the region of −2 to −3 V, and can be observed most clearly in Fig. 2 for Li+. Relatively sharp reduction profiles were observed for all metals at potentials more negative than −3 V, consistent with an overpotential being required for the initial nucleation of the Group I metal to occur on Ni, followed by the rapid deposition of metal-on-metal. The local concentration of the metal ion is therefore depleted, and in the case of Li+, Na+ and K+ this leads to a characteristic peak shape corresponding to a mass transport controlled region [16]. All scans also possess characteristic nucleation loops (where backward and forward currents cross over), as on the reverse scan Group I metal is still being deposited on Group I metal at potentials positive to that required for the initial nucleation of the metal on Ni. In the case of Rb+ and Cs+, the reduction potentials for these metals were significantly more negative than that of Li+, Na+ and K+, such that nucleation occurred only a few millivolts prior the reduction process of [C4mpyrr][NTf2]. Therefore the scan had to be reversed shortly after nucleation, and a considerable amount of metal was instead deposited on the reverse scan (direction of the scan indicated by arrows in the figure).

The charge passed under the reduction process can be used to quantify the amount of alkali metal reduced at the electrode surface. Assuming the metal is deposited as layers with the same geometric area as that of the Ni microelectrode, the reduction charge calculated for the bulk reduction peak in the voltammograms in Fig. 2 indicates that approximately 63, 96, 34, 33 and 68 layers of Li, Na, K, Rb and Cs were deposited, demonstrating that all systems deposit bulk metal on the Ni electrode. All systems also display sharp oxidation peaks which correspond to the stripping off of metal from the electrode surface. Therefore, the CV response after the bulk deposition peak and prior to the bulk stripping peak (e.g. where the CV crosses over the x-axis, corresponding to zero net current being passed) corresponds to the system behaving as an M/M+ interface, as opposed to Ni/M+. The response of the system in this region is therefore a feature of the fundamental parameters of the M/M+ couple in the IL. The CVs were simulated as described below in order to quantify the parameters.

3.2. Simulation results

The details of the simulation were explained briefly in Section 2. Full details of the development of the simulation model [13], as well as its application to experimental results have been previously demonstrated [5-7,14,15]. The simulation model was applied to the K/K+, Rb/Rb+ and Cs/Cs+ systems in order to extract the fundamental electrochemical parameters of the couples. Figs. 3–5 display overlays of the experimental results as well as best fits for the simulations for K/K+, Rb/Rb+ and Cs/Cs+, respectively.

3. R. Wibowo et al. / Chemical Physics Letters Volume 492, Issue 4-6, June 2010, Pages 276-280

The simulation of the Li/Li+[5] and Na/Na+[7] systems has been reported previously.

Figure 3

Fig. 3: The fit of the experimental (–) and simulation (○) data for the cyclic voltammetry experiment using 0.1 M K[NTf2] in [C4mpyrr][NTf2] on a 50 μm diameter Ni electrode. Scan rate = 10 mV s−1.

Figure 4

Fig. 4: The best fit of experimental (–) and simulation (○) data for the CV of 0.1 M Rb[NTf2] in [C4mpyrr][NTf2] on 50 μm diameter Ni electrode. Scan rate = 10 mV s−1.

Figure 5

Fig. 5: The best fit of experimental (–) and simulation (○) data for the CV of 0.1 M Cs[NTf2] in [C4mpyrr][NTf2] on 50 μm diameter Ni electrode. Scan rate = 10 mV s−1.

The simulation can be split into three areas; the alkali metal deposition peak, the alkali metal stripping peak, and the region of zero current in between the two processes.

The simulated alkali metal ion deposition peaks do not precisely match the experimental results, as shown in Figs. 3–5. This is because the simulated results do not take into account the nucleation process, replication of which is not required for an investigation of solely M/M+ parameters. The simulation was started in the region where initial nucleation was observed to occur in the experimental results. The physical parameters were put into the simulations, and the amount of metal deposited was solved. By varying the parameters a good fit could be achieved on the reverse scan, which also resulted in the simulation generating the appropriate concentration gradient for M+ extending out from the electrode surface required for a good fit during the zero current cross-over region (occurring ca. 30 s after deposition is started).

As the shape of the reduction peak after the initial nucleation process is mass transfer controlled, it is also extremely sensitive to the diffusion coefficient, D, of the ion M+. Fitting to the experimental reduction process for K+ (as shown in Fig. 3) was used to accurately determine D. The D for Li+[5] and Na+[7] have been previously reported. However, since clear mass transfer controlled regions could not be observed in the case of Rb/Rb+ and Cs/Cs+, the diffusion coefficient could not be independently determined. The D values for Li+, Na+ and K+ were found to be linear with respect to 1/r, where r is the radius of the ion. This allowed us to extrapolate the trend and estimate D values for Rb+ and Cs+ based upon their known r values. These estimated D values were found to work well in the subsequent simulations.

For all of the metals the simulation displays stripping peaks significantly higher than that of the experimental result, as the simulation assumes 100% stripping efficiency. However, the foot of the stripping peak fitted well in all cases, indicating that during the reverse scan bulk alkali metal was present on the surface of the electrode. Therefore, in the potentials between bulk deposition and bulk stripping, the electrode is behaving as an alkali metal electrode, M, in contact with a precisely defined local concentration of its ion, M+. The current in the cross-over region area (ca. ±100 mV from the x-axis cross-over potential) is a function of both oxidation and reduction processes occurring appreciably at the M/M+ interface, as defined by Butler–Volmer kinetics (see Section 2 for more details) and is therefore sufficiently sensitive to , α and k0 for these values to be determined accurately by focussing on the simulation of this region. The values of , α and k0 were therefore varied until the best fit was obtained between simulation and experimental results for the tail end of metal deposition, the cross-over region and the foot of the oxidation peak. The good fit obtained between experimental and simulation data is clearly observed in Figs. 3–5.

Table 1 presents the electrochemical parameters which include diffusion coefficient (D), electrochemical rate constant (k0), formal potential (), and transfer coefficient (α) required for the best fits shown in Figs. 3–5. The data for the Li/Li+[5] and Na/Na+[7] redox couples are those published previously.

Table 1: The electrochemical parameters for the alkali metal redox couples in [C4myprr][NTf2], extracted by simulation of the experimental data, where D is the diffusion coefficient, k0 is the standard electrochemical rate constants, is the formal potential, α is the transfer coefficient and is the Gibbs energy of solvation in H2O or [C4myprr][NTf2].

M/M+ 108D(cm2 s−1) 105k0(cm s−1) (V vs. corrected Fc/Fc+) α (kJ mol−1)c vs. Fc/Fc+(kJ mol−1)
Li/Li+a 4.5 ± 0.1 1.20 ± 0.05 −3.250 ± 0.010 0.63 ± 0.01 −509 −336 ± 10
Na/Na+a 7.8 ± 0.4 1.15 ± 0.06 −3.066 ± 0.003 0.63 ± 0.02 −409 −278 ± 10
K/K+ 10.2 ± 0.5 0.3 ± 0.05 −3.359 ± 0.005 0.80 ± 0.05 −335 −157 ± 10
Rb/Rb+ 10.9 ± 0.5b 0.3 ± 0.04 −3.367 ± 0.003 0.75 ± 0.05 −314 −132 ± 10
Cs/Cs+ 11.5 ± 0.5b 0.2 ± 0.05 −3.472 ± 0.003 0.75 ± 0.05 −283 −86 ± 10

a The data for Li/Li+ and Na/Na+ are taken from Refs. [5,7], respectively. b D values estimated by extrapolation from a plot of Li+, Na+ and K+D values vs. 1/r, where r is the ionic radii. c values from reference [26].

From Table 1 the increase in diffusion coefficient (D) down the group can clearly be observed for Li+, Na+, and K+. It should be recalled that D was estimated for Rb+ and Cs+. Similar trends down Group I have previously been observed in aqueous [17,18] and methanol solutions [19]. The electrochemical rate constant (k0) was also found to decrease slightly down the group. The transfer coefficients are all above 0.5 implying a late transition state, presumably due to the M+ ion being largely desolvated. The formal potential () displays a similar trend to that reported for a variety of other solvents (discussed below).

4. R. Wibowo et al. / Chemical Physics Letters Volume 492, Issue 4-6, June 2010, Pages 276-280

In Table 1 is reported vs. Fc/Fc+, where the formal potential of Fc/Fc+ has been carefully corrected with respect to the discrepancies in diffusion coefficients that can arise between redox couples in ILs [20], and this correction method has been described previously [7].

The trend in formal potential of the metal redox couples has been reported for a number of solvents [21]. However, quantitative comparison is made difficult by the lack of a rigorous reference potential for ILs [15,20,22]. While extensive work aimed primarily at Lithium and energy storage applications has been reported in ILs [23-25], few quantitative values have been published that can be compared with. Wilkes et al. carried out a study aimed at the reduction potentials of the alkali metals in a neutral, buffered [C2mim]Cl:AlCl3 melt, although this was limited to the reduction potentials for amalgam formation at a Hg-film electrode, due to the electrochemical window of the melt not supporting deposition of the pure metal [26].

Fig. 6 displays a comparison of the trends in for the Group I metals relative to the of the Li/Li+ couple in the same medium. It can be observed that the trend in formal potential in H2O, DMSO and DMF [21], going from most positive to negative, is Na > K > Rb > Cs > Li. In the case of amalgam formation potentials in [C2mim]Cl:AlCl3[26], the trend goes down the group, such that Li > Na > K > Rb > Cs. In [C4mpyrr][NTf2] the trend in displays a remarkable similarity to propylene carbonate, with Na > Li > K ≈ Rb > Cs.

Figure 6

Fig. 6: The plot of alkali metal redox couple formal potential relative to the Li/Li+ couple in (△) 60% DMSO/40% H2O (▽) DMSO, (○) H2O, (×) DMF, (♢) propylene carbonate, (♦)[C4mpyrr][NTf2] (this work) and (▴) neutral, buffered [C4mim]Cl:AlCl3 melt, at a Hg-film electrode (from Ref. [26]). All values with hollow symbols taken from Ref. [21].

The trend in solvation energy can be estimated by using the obtained values. A Born–Haber cycle can be used to calculate the Gibbs energy of solvation, , by comparison with the known values for the Gibbs energy of ionisation and sublimation, as well as the single-electrode reduction potential [26]. Lacking a reliable single-electrode reference potential for [C4mpyrr][NTf2], as well as knowledge of activity coefficients in the IL environment, the relative solvation energies can nevertheless be estimated vs. Fc/Fc+ by using the obtained vs. Fc/Fc+ values for the alkali metals and assuming . Table 1 displays the calculated vs. Fc/Fc+ for the metals in [C4mpyrr][NTf2], as well as the literature values for as comparison. Fig. 7 displays a plot of vs. Fc/Fc+ against , which indicates that the smaller ions are the most strongly solvated in both systems. A line of best linear fit has been included in order to guide the eye. However, it should be noted that a perfect linear fit would not be expected unless solvation behaviour is identical for the Group I ions in both H2O and [C4mpyrr][NTf2].

Figure 7

Fig. 7: The plot of the estimated Gibbs energy of solvation of the alkali metal ions in[C4mpyrr][NTf2] against the known Gibbs energy of solvation for the ions in aqueous solution.

4. Conclusions

In summary, the deposition and dissolution of all five Group I alkali metal redox couples (M/M+) have been investigated in the ionic liquid N-butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide ([C4mpyrr][NTf2]) using a Ni microelectrode. All systems displayed chemically reversible behaviour which was successfully simulated using a mathematical model and the electrochemical rate constant (k0), formal potential () transfer coefficient (α) and diffusion coefficient (D) have been extracted by simulation of the CV.

5. R. Wibowo et al. / Chemical Physics Letters Volume 492, Issue 4-6, June 2010, Pages 276-280

This constitutes the first electrochemical investigation of an entire group in an ionic liquid. The reducing powers of the alkali metals in [C4mpyrr][NTf2] follow the trend previously observed for propylene carbonate, with Na > Li > K ≈ Rb > Cs.

Acknowledgements

R.W. thanks the Directorate of Higher Education, The Ministry of National Education, Republic of Indonesia for funding.

References

View references in table layout

View references as list

# Authors Year Source title Journal name Volume Issue Pages Cited by
1 Armand M., Endres F., MacFarlane D.R., Ohno H., Scrosati B. 2009 Ionic-liquid materials for the electrochemical challenges of the future Nature Materials 8 8 621–629 66
2 Wishart J.F. 2009 Energy applications of ionic liquids Energy and Environmental Science 2 9 956–961 12
3 Silvester D.S., Compton R.G. 2006 Electrochemistry in room temperature ionic liquids: A review and some possible applications Zeitschrift fur Physikalische Chemie 220 10-Nov 1247–1274 105
4 Liu H., Liu Y., Li J. 2010 Ionic liquids in surface electrochemistry Physical Chemistry Chemical Physics 12 8 1685–1697 3
5 Wibowo R., Ward Jones S.E., Compton R.G. 2009 Kinetic and thermodynamic parameters of the Li/Li+ couple in the room temperature ionic liquid n-butyl-N-methylpyrrolidinium bis(trifluoromethylsulfonyl) imide in the temperature range 298-318 K: a theoretical and experimental study using pt and Ni electrodes Journal of Physical Chemistry B 113 36 12293–12298 9
6 Wibowo R., Jones S.E.W., Compton R.G. 2010 Investigating the electrode kinetics of the Li/Li+ Couple in a wide range of room temperature ionic liquids at 298 K Journal of Chemical and Engineering Data 55 3 1374–1376 1
7 Wibowo R., Aldous L., Rogers E.I., Ward Jones S.E., Compton R.G. 2010 A study of the Na/Na+ redox couple in some room temperature ionic liquids Journal of Physical Chemistry C 114 8 3618–3626 2
8 Evans R.G., Klymenko O.V., Price P.D., Davies S.G., Hardacre C., Compton R.G. 2005 A comparative electrochemical study of diffusion in room temperature ionic liquid solvents versus acetonitrile ChemPhysChem 6 3 526–533 76
9 Schroder U., Wadhawan J.D., Compton R.G., Marken F., Suarez P.A.Z., Consorti C.S., De Souza R.F., Dupont J. 2000 New Journal of Chemistry 24 12 1009–1015 234
10 Buzzeo M.C., Hardacre C., Compton R.G. 2006 Extended electrochemical windows made accessible by room temperature ionic liquid/organic solvent electrolyte systems ChemPhysChem 7 1 176–180 32
11 O'Mahony A.M., Silvester D.S., Aldous L., Hardacre C., Compton R.G. 2008 Journal of Chemical and Engineering Data 53 12 2884–2891 32
12 Rogers E.I., Huang X.-J., Dickinson E.J.F., Hardacre C., Compton R.G. 2009 Investigating the mechanism and electrode kinetics of the oxygenlsuperoxide (O2IO2-) couple in various room-temperature ionic liquids at gold and platinum electrodes in the temperature range 298-318 K Journal of Physical Chemistry C 113 41 17811–17823 9
13 Ward Jones S.E., Chevallier F.G., Paddon C.A., Compton R.G. 2007 Analytical Chemistry 79 11 4110–4119 16
14 Paddon C.A., Ward Jones S.E., Bhatti F.L., Donohoe T.J., Compton R.G. 2007 Kinetics and thermodynamics of the Li/Li+ couple in tetrahydrofuran at low temperatures (195-295 K) Journal of Physical Organic Chemistry 20 9 677–684 9
15 Rogers E.I., Silvester D.S., Jones S.E.W., Aldous L., Hardacre C., Russell A.J., Davies S.G., Compton R.G. 2007 Electrochemical kinetics of Ag|Ag+ and TMPD|TMPD+ in the room-temperature ionic liquid [C4mpyrr][NTf2]; toward optimizing reference electrodes for voltammetry in RTILs Journal of Physical Chemistry C 111 37 13957–13966 19
16 Compton R.G., Banks C.E. 2007 Understanding Voltammetry World Scientific Publishing 118
17 Harned H.S., Blander M., Hildreth Jr. C.L. 1954 The diffusion coefficient of cesium chloride in dilute aqueous solution at 25° Journal of the American Chemical Society 76 16 4219–4220 1
18 Leaist D.G., Goldik J. 2001 Diffusion and ion association in concentrated solutions of aqueous lithium, sodium, and potassium sulfates Journal of Solution Chemistry 30 2 103–118 6
19 Chowdhuri S., Chandra A. 2006 J. Chem. Phys. 124 1
20 Rogers E.I., Silvester D.S., Poole D.L., Aldous L., Hardacre C., Compton R.G. 2008 Voltammetric characterization of the ferrocene|ferrocenium and cobaltocenium|cobaltocene redox couples in RTILs Journal of Physical Chemistry C 112 7 2729–2735 46
21 Marcus Y. 1985 Pure Appl. Chem. 57 1129–1132 6
22 Silvester D.S., Rogers E.I., Compton R.G. 2008 Reference electrodes for use in room temperature ionic liquids Electrodeposition from Ionic Liquids 296 3
23 Howlett P.C., MacFarlane D.R., Hollenkamp A.F. 2004 High lithium metal cycling efficiency in a room-temperature ionic liquid Electrochemical and Solid-State Letters 7 5 130
24 Katayama Y., Morita T., Yamagata M., Miura T. 2003 Electrodeposition of Metallic Lithium on a Tungsten Electrode in 1-Butyl-1-methylpyrrolidinium Bis(trifluoromethanesulfone)imide Room-temperature Molten Salt Electrochemistry 71 12 1033–1035 34
25 Endres F., Zein El Abedin S., Borissenko N. 2006 Probing lithium and alumina impurities in air- and water stable ionic liquids by cyclic voltammetry and in situ scanning tunneling microscopy Zeitschrift fur Physikalische Chemie 220 10-Nov 1377–1394 24
26 Scordilis-Kelley C., Fuller J., Carlin R.T., Wilkes J.S. 1992 Alkali metal reduction potentials measured in chloroaluminate ambient-temperature molten salts Journal of the Electrochemical Society 139 3 694–699 49
Elsevier logo
Copyright © 2011 Elsevier B.V. All rights reserved. SciVerse® is a registered trademark of Elsevier Properties S.A., used under license. ScienceDirect® is a registered trademark of Elsevier B.V.