- Email: [email protected]
Electrochemistry of ytterbium (III) in molten alkali metal chlorides
Electrochimica Acta 54 (2008) 382–387
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Electrochemistry of ytterbium (III) in molten alkali metal chlorides V. Smolenski a , A. Novoselova a , A. Osipenko b , C. Caravaca c , G. de Córdoba c,∗ a Institute of High-Temperature Electrochemistry, Ural Division, Russian Academy of Science, Ekaterinburg, 620219, Russia b Research Institute of Atomic Reactors, Dimitrovgrad-10, Ulyanovsk Region, 433010, Russia c High Level Waste Unit, Nuclear Fission Division, CIEMAT, Madrid, 28040, Spain
a r t i c l e
i n f o
Article history: Received 23 April 2008 Received in revised form 23 June 2008 Accepted 25 July 2008 Available online 15 August 2008 Keywords: Molten alkali metal chlorides Ytterbium chloride Diffusion coefficient Voltammetry techniques Thermodynamics
a b s t r a c t This work presents the electrochemical study of Yb(III) ions in molten alkali metal chlorides in the temperature range 723–1073 K. Transient electrochemical techniques such as linear sweep, cyclic and square wave voltammetry, and potentiometry at zero current have been used to investigate the reduction mechanism, transport parameters and thermodynamic properties of the reaction YbCl2 + 1/2Cl2 = YbCl3 The results obtained show that the reduction reaction Yb(III) + e− ⇔ Yb(II) is reversible being controlled by the rate of the mass transfer. The diffusion coefficient of [YbCl6 ]3− complex ions has been determined at different temperatures in the fused eutectic LiCl–KCl, the equimolar NaCl–KCl and the CsCl media. The apparent standard potential of the soluble–soluble redox system Yb(III)/Yb(II) has been obtained by cyclic voltammetry. The influence of the nature of the solvent on the electrochemical and thermodynamic properties of ytterbium compounds is discussed. © 2008 Elsevier Ltd. All rights reserved.
1. Introduction Partitioning and Transmutation (P&T) concept is nowadays considered as one of the strategies to reduce the long-term radiotoxicity of the nuclear wastes. To achieve this, the efficient recovery and multi-recycling of actinides (An), especially TRU elements, in advanced dedicated reactors are essential. Fuels proposed to transmute the actinides into short-lived or even stable radionuclides will contain significant amounts of Pu and minor actinides (MA) (Np, Am, Cm), possibly dissolved in inert matrices (U free), and will reach high burn-ups. Pyrochemical separation techniques offer some potential advantages compared to the hydrometallurgical processes to separate actinides from fission products (FP) contained in the irradiated fuel. The high radiation stability of the salt and metallic solvents used, resulting in shorter fuel cooling times, stands out [1–5]. The aim of the separation techniques which are currently being investigated, both hydrometallurgical and pyrometallurgical ones, is to optimize the recovery efficiency of MA minimizing at the same time the FP content in the final product. Special attention is devoted
∗ Corresponding author. Tel.: +34 91 346 62 16; fax: +34 91 346 62 33. E-mail address: [email protected] (G. de Córdoba). 0013-4686/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2008.07.086
to rare earth fission products (REE) mainly due to its neutronic poison effect and the high content of these elements in the irradiated fuel. In addition, REE have similar chemical properties to those of An; hence, separation between these groups of elements is rather difficult. For this reason, a good knowledge of the basic properties of An [6–14] and REE [15–21] in the proposed separation media is very important. There is not too much information concerning the behaviour of ytterbium compounds in molten chlorides. The standard redox potentials of Yb(III)/Yb(II) versus the Pt(II)/Pt(0) reference in LiCl–KCl have been measured as well as its temperature dependence [17]. The electrochemical behaviour of Yb(III) ions on active Ni electrode in molten NaCl–KCl has been investigated at 973 K. It has been shown that the reaction occurs in two steps: the electrochemical formation of Yb(II) ions followed by the reduction of Yb(II) into metal Yb and the subsequent Ni–Yb alloys formation [18]. The electrochemistry of Yb(III) ions in molten NaCl–KCl, KCl and CsCl has been studied by transient electrochemical techniques. In these media, the apparent redox potentials of the Yb(III)/Yb(II) system as well as the diffusion coefficients of Yb(III) and Yb(II) ions were calculated [16,19]. Bermejo at al. [20] have also investigated the electrochemical behaviour of Yb(III) ions in fused LiCl–KCl. These authors have determined the kinetic and thermodynamic properties of the reaction Yb(III) + e− ↔ Yb(II). Novoselova et al. [21] have
V. Smolenski et al. / Electrochimica Acta 54 (2008) 382–387
Fig. 1. Cyclic voltammograms of fused LiCl–KCl eutectic and LiCl–KCl–YbCl3 on tungsten electrode (S = 0.31 cm2 ) at 723 K. Scan rate 0.1 V s−1 . Yb(III) concentration 2.57 × 10−2 mol kg−1 .
estimated the apparent redox potentials of the Yb(III)/Yb(II) system in molten alkali metal chlorides by using a correlation method between aqueous solutions and molten chlorides [21]. This work presents the electrochemical behaviour of Yb(III) ions in molten LiCl–KCl eutectic in the temperature range 723–973 K using transient electrochemical techniques. The experimental data obtained are compared with those in NaCl–KCl and CsCl media previously published. The influence of the nature of the solvent on the electrochemical and thermodynamic properties of ytterbium ions is discussed. 2. Experimental The experiments were carried out under inert argon atmosphere using a three electrodes electrochemical quartz sealed cell. Different transient electrochemical techniques were applied such as linear sweep (LSW), cyclic (CV) and square wave voltammetry (SWV), as well as potentiometry at zero current. The electrolyte LiCl–KCl (Aldrich, 99.9%) was dehydrated by heating under vacuum in the temperature range 293–573 K, while the solvents NaCl–KCl (Aldrich, 99.9%) and CsCl (Aldrich, 99.9%) were heated in the temperature range 293–773 K, also under vacuum. Afterwards, the salts were fused under Ar(g) atmosphere. Finally, they were purified by means of the so-called “direct crystallization method [22]. Yb(III) solutions were prepared by direct addition of anhydrous YbCl3 (Merck, 99.99%) to the electrolytic bath. The electrochemical measurements were carried out using an Autolab PGSTAT30 potentiostat (Eco-Chimie) with specific GPES electrochemical software (version 4.9).
Fig. 2. Linear sweep voltammograms of fused NaCl–KCl–YbCl3 at different sweep rates at 973 K. [Yb(III)] = 3.79 × 10−2 mol kg−1 . Working electrode: W (S = 0.27 cm2 ).
383
Fig. 3. Cyclic voltammograms of CsCl–YbCl3 at different sweep potential rates at 973 K. Working electrode: W (S = 0.31 cm2 ). [Yb(III)] = 3.70 × 10−2 mol kg−1 .
The inert working electrode consisted of a 1 mm metallic W wire (Goodfellow, 99.9%). It was immersed into the molten bath between 3 and 10 mm. The active surface area was determined after each experiment by measuring the immersion depth of the electrode. The counter electrode consisted of a 3 mm vitreous carbon rod (SU – 2000). The 2Cl− /Cl2 system is the most reliable reference electrode since it constitutes the universal reference in molten chloride media and it can be used for direct thermodynamic calculations. The electrode device consists of a graphite tube (for the chlorine gas introduction) dipped into a quartz tube provided with a porous membrane at its bottom and containing molten chloride solvent. Chlorine gas is bubbled through the melt during the experiment [23]. The ytterbium concentration in solution was determined by taking samples from the melt which were dissolved in nitric acid solutions and then analyzed by ICP-MS. 3. Results and discussion 3.1. Study by voltammetric techniques Cyclic voltammetry was carried out on inert tungsten for all melts tested: LiCl–KCl, NaCl–KCl and CsCl, and at several temperatures (723–1073 K). The electroactivity domain in LiCl–KCl at 723 K is shown in Fig. 1. The cathodic and anodic limits of the electrochemical window correspond to the reduction of the alkali metal ions of the solvent (Li+ /Li0 , Na+ /Na0 and Cs+ /Cs0 ) and to the oxidation of chloride ions into chlorine gas (2Cl− /Cl2 ), respectively.
Fig. 4. Square wave voltammogram of LiCl–KCl–YbCl3 (9.41 × 10−2 mol kg−1 ) at 10 Hz at 723 K. Working electrode: W (S = 0.25 cm2 ).
384
V. Smolenski et al. / Electrochimica Acta 54 (2008) 382–387
Fig. 5. Variation of the cathodic and anodic peak potential as a function of the sweep rate in fused LiCl–KCl–YbCl3 (9.41 × 10−2 mol kg−1 ) at 723 K. Working electrode: W (S = 0.25 cm2 ).
Fig. 6. Variation of the cathodic and anodic peak current as a function of the potential scan rate in fused LiCl–KCl–YbCl3 (9.41 × 10−2 mol kg−1 ) at 723 K. Working electrode: W (S = 0.25 cm2 ).
Fig. 1 also plots the cyclic voltammogram of a LiCl–KCl–YbCl3 solution on W at 723 K. It shows a single cathodic peak at a potential of −1.762 V vs. 2Cl− /Cl2 and its corresponding anodic peak at −1.566 V vs. the 2Cl− /Cl2 . A similar behaviour for the reduction of Yb(III) ions has been observed in the fused equimolar NaCl–KCl mixture and CsCl. Fig. 2 shows a linear sweep voltammogram of the system NaCl–KCl–YbCl3 registered at different potential scan rates. A single cathodic peak is observed. The analysis of the peak indicates that only one electron is exchanged being associated to the reduction reaction Yb(III) + e− ↔ Yb(II). Concerning the system CsCl–YbCl3 , the cyclic voltammograms recorded at several scan rates are plotted in Fig. 3. The shape of the CV is typical for a soluble–soluble system suggesting that in this case the system corresponds to the reaction Yb(III)/Yb(II). The square wave voltammetry technique has been applied to determine the number of electrons exchanged in the reduction of Yb(III) ions in LiCl–KCl. Fig. 4 shows the Gaussian-shaped symmetric cathodic wave obtained in the LiCl–KCl–YbCl3 at 723 K. The number of electrons exchanged is determined by measuring the width at half height of the reduction peak, W1/2 (V), recorded at different frequencies (6–80 Hz), using the following equation, valid for reversible systems [10,24]:
independent of the potential scan rate (see Fig. 5); (ii) The cathodic and anodic peak currents (Ip ) are directly proportional to the square root of the polarization rate () (see Fig. 6); (iii) A linear relationship between the cathodic peak current density and the concentration of YbCl3 ions in the melt is observed (see Fig. 7). From these results and, according to the theory of the linear sweep voltammetry technique [25], it can be concluded that the redox system Yb(III)/Yb(II) is reversible and it is controlled by the rate of the mass transfer. From the transient electrochemical techniques applied, it is found that the system Yb(II)/Yb(0) cannot be observed in molten alkali chlorides media since its potential is more negative than the reduction potential of the solvent Me(I)/Me(0), Me being: Li, Na, K and Cs, (see Fig. 1).
W1/2 = 3.52
RT nF
(1)
At low frequencies, a linear relationship between the cathodic peak current and the square root of the frequency is found. Under these conditions, the system can be considered as reversible; hence it is possible to apply Eq. (1). The number of electrons exchanged was found close to one. Similar results were obtained in NaCl–KCl and CsCl media. Potentiostatic electrolysis at potentials corresponding to the cathodic peaks for all systems studied did not show the formation of any solid phase at the tungsten surface after its polarization. No plateaus were observed in the potential–time curves. Also the working electrode did not undergo any visual change. In addition, the X-ray analysis of the surface of the working electrodes showed the absence of any solid phase. The results obtained allow concluding that the reduction of Yb(III) ions takes place in a single step with the exchange of one electron and the formation of a soluble product, according to the following reaction: Yb(III) + e− ⇔ Yb(II)
3.2. Diffusion coefficient of Yb(III) ions The diffusion coefficient of [YbCl]3− 6 ions in molten chloride media was determined using the cyclic voltammetry technique and applying the Randles–Sevˇcik equation, valid for reversible soluble–soluble systems [24]: Ip = 0.446(nF)3/2 C0 S
Dv 1/2 RT
(3)
where S is the electrode surface area (cm2 ), C0 is the solute concentration (mol cm−3 ), D is the diffusion coefficient (cm2 s−1 ), F is the Faraday constant (96,500 C mol−1 ), R is the ideal gas constant (J K−1 mol−1 ), n is the number of exchanged electrons, is the potential sweep rate (V s−1 ) and T is the absolute temperature (K). The values obtained for the different solvents studied at several
(2)
The reaction mechanism of the soluble–soluble Yb(III)/Yb(II) exchange was studied by analyzing the voltammetric curves recorded at several potential scan rates. The analysis shows that: (i) The cathodic and anodic peak potentials (Ep ) are constant and
Fig. 7. Variation of the cathodic peak current as a function of the concentration YbCl3 in fused LiCl–KCl–YbCl3 at 723 K. Working electrode: W (S = 0.25 cm2 ). Scan rate = 0.1 V s−1 .
V. Smolenski et al. / Electrochimica Acta 54 (2008) 382–387
385
Table 1 Diffusion coefficient of [YbCl6 ]3− ions in molten alkali metal chlorides at several temperatures (Activation energy for the ytterbium ions diffusion process) Solvent
T (K)
D × 105 (cm2 s−1 )
D × 105 (cm2 s−1 ) [19]
−EA (kJ mol−1 )
−EA (kJ mol−1 ) [19]
Do × 103 (cm2 s−1 )
Do × 103 (cm2 s−1 ) [19]
LiCl–KCl
723 848 973
1.0 ± 0.1 2.7 ± 0.1 5.4 ± 0.1
– – 5.3 ± 0.2*
38.3
–
6.2
–
NaCl–KCl
973 1023 1073
2.8 ± 0.2 3.2 ± 0.2 4.1 ± 0.2
2.4 ± 0.2 2.7 ± 0.2 3.4 ± 0.2
45.4
46.4
6.6
6.2
CsCl
973 1023 1073
0.9 ± 0.1 1.2 ± 0.1 1.7 ± 0.1
0.9 ± 0.2 1.2 ± 0.2 1.6 ± 0.2
54.4
55.6
7.4
8.5
*
Extrapolated value.
temperatures are quoted in Table 1. The experimental values have also been compared with those reported in the literature. The diffusion coefficient values have been used to calculate the activation energy for the diffusion process. The influence of temperature on the diffusion coefficient obeys the Arrhenius’s law through the following equation [26]:
recorded in the YbCl3 solutions in the different alkali metal chlorides tested. According to the theory of linear sweep voltammetry, the following expressions, including the anodic and cathodic peak potentials and the half-wave potential, can be applied in the case of soluble–soluble reversible systems [19,30]:
D = Do exp −
(4)
EPC = E1/2 − 1.11
RT F
(6)
where EA is the activation energy for the diffusion process (kJ mol−1 ), Do is the pre-exponential term (cm2 s−1 ) and is the experimental error. From this expression, the value of the activation energy for the Yb(III) ions diffusion process was calculated in the different melts tested (see Table 1). The diffusion coefficient of ytterbium (III) ions becomes smaller as the radius of the alkali metal cation increases from Li to Cs (see Table 1). Such behaviour takes place due to an increase on the strength of complex ions and a decrease on the contribution of D to the “hopping” mechanism [27]. A similar trend for the diffusion coefficient has been reported elsewhere [28]. As the temperature increases the diffusion coefficient increases in all the solvents. The variation of the logarithm of the diffusion coefficient as a function of the reverse of the solvent cation radius (r) at 973 K can be expressed as follows (see Fig. 8):
EPA = E1/2 + 1.11
RT F
(7)
E A
log D[YbCl
3−
6]
RT
±
0.17 = −6.07 − ± 0.02 r
(5)
The ionic radius values have been taken from the literature [29]. 3.3. Apparent standard potential of the redox system Yb(III)/Yb(II) The apparent standard potential of the Yb(III)/Yb(II) system has been determined from the analysis of the cyclic voltammograms
(EPC + EPA ) 2
= E1/2
(8)
where the half-wave potential is given by: E1/2 =
0 EYb(III)/Yb(II)
RT + ln F
DYb(II)
1/2
DYb(III)
RT + ln F
Yb(III) Yb(II)
(9)
It is known that for concentrations of electroactive species lower than 3–5 × 10−2 in mole fraction scale, their activity coefficient is almost constant. Under these conditions, it is more convenient ∗ ) using the apparent standard redox potential concept (EYb(III)/Yb(II) expressed as follows [23]: ∗ 0 = EYb(III)/Yb(II) + EYb(III)/Yb(II)
RT ln F
Yb(III)
Yb(II)
(10)
Considering that the diffusion coefficient values for the triand bivalent ytterbium ions are close, it can be assumed that the term (RT/F)ln(DYb(II) /DYb(III) )1/2 tends to zero. As a consequence, the apparent standard potential of the redox system Yb(III)/Yb(II) can be calculated from the equations: ∗ = EPC + 1.11 EYb(III)/Yb(II)
RT F
(11)
∗ = EPA − 1.11 EYb(III)/Yb(II)
RT F
(12)
∗ = EYb(III)/Yb(II)
(EpC + EpA ) 2
(13)
From the peak potential values measured in the cyclic voltammograms, the following empirical equation for the apparent standard potential of the Yb(III)/Yb(II) system in LiCl–KCl has been obtained (potentials are expressed vs. the 2Cl− /Cl2 reference system): ∗ = −(1.915 ± 0.005) + (3.5 ± 0.2) × 10−4 T (V) EYb(III)/Yb(II)
[723 − 973 K]
Fig. 8. Linear relationship of logarithm of the [YbCl6 ]3− diffusion coefficient versus the reverse of the radius of the solvent cation at 973 K.
(14)
Lanthanide chlorides dissolved in alkali chloride melts are usually solvated by chloride ions forming different complex ions
386
V. Smolenski et al. / Electrochimica Acta 54 (2008) 382–387
Table 2 Apparent standard redox potential of the Yb(III)/Yb(II) system in molten alkali metal chlorides at several temperatures Solvent
T (K)
∗ −EYb(III)/Yb(II) (V)
∗ −EYb(III)/Yb(II) (V) [17]
∗ −EYb(III)/Yb(II) (V) [19]
∗ −EYb(III)/Yb(II) (V) [20]
LiCl–KCl
723 848 973
1.662 1.618 1.574
1.704 1.618 1.532*
– – 1.531*
1.659 – –
NaCl–KCl
973 1023 1073
1.671 1.652 1.634
– – –
1.678 1.662 1.646
1.629 – 1.551
CsCl
973 1023 1073
1.853 1.832 1.811
– – –
1.871 1.844 1.817
1.881 – 1.792
Reference system: Cl− /Cl2 . * Extrapolated value.
such as [LnCl6 ]3− and [LnCl4 ]2− [31–33]. In the case of ytterbium, [YbCl6 ]3− complex ions are present in the melt [21]. Their relative stability increases with the increase of the solvent cation radius, and the apparent standard redox potential shifts to more negative values. The results obtained here are in agreement with those reported by Smirnov [23]. The variation of the apparent standard potential of the redox couple Yb(III)/Yb(II) as a function of the reverse of the solvent cation radius (r) has been calculated at 973 K. The relation obtained is: ∗ = −2.124 + EYb(III)/Yb(II)
0.052 ± 0.003 (V) r
(15)
3.4. Thermodynamic properties
(16)
has been calculated according to following expression: (17)
Its temperature dependence allows calculating the enthalpy and entropy of the redox reaction (16) by means of the relation [24]: G∗ = H ∗ − TS ∗
[3LiCl − 2 KCl; 723 − 973 K]
(19)
G∗ = −195.96 + 0.036T ± 2.46 kJ/mol [NaCl − KCl; 973 − 1073 K] G∗ = −218.25 + 0.041T ± 2.46 kJ/mol
(20) [CsCl; 973 − 1073 K]
The variation of the thermodynamic parameters for the redox reaction (16) as a function of the solvent cation radius shows an increase of the strength of the Yb Cl bonds in the complex ions [YbCl6 ]3− . 4. Conclusions
The apparent standard Gibbs energy (G*) of the redox reaction:
∗ G∗ = −nFEYb(III)/Yb(II)
G∗ = −184.80 + 0.033T ± 2.46 kJ/mol
(21)
The experimental values of the apparent standard potential of the Yb(III)/Yb(II) system at several temperatures as well as those reported in the literature are gathered in Table 2 and plotted in Fig. 9.
YbCl2(l) + 12 Cl2(g) ⇔ YbCl3(l)
The apparent standard Gibbs energy of the redox reaction (16) in the different solvents tested can be expressed as:
(18)
The electrochemical behaviour of [YbCl6 ]3− complex ions in the molten LiCl–KCl eutectic was investigated. It was found that the reduction of Yb(III) to Yb(II) ions is a reversible process controlled by the rate of the mass transfer. The diffusion coefficient of [YbCl6 ]3− species was determined at different temperatures. The apparent standard electrode potential of the redox couple Yb(III)/Yb(II) was calculated from the analysis of the cyclic voltammograms registered at different temperatures (E* Yb(III)/Yb(II) = −1.662 V vs. 2Cl− /Cl2 at 723 K). In addition, the influence of the nature of the solvent cation (ionic radius) on the thermodynamic properties of Yb(III) ions was studied. It was found that the strength of the Yb Cl bonds increases as we shift from Li to Cs cation. Acknowledgement The authors wish to thank Dr. A. Bovet for the technical support. References
Fig. 9. Apparent standard potential of the redox couple Yb(III)/Yb(II) versus the reverse of the radius of the solvent cation at 973 K.
[1] OCDE/NEA Report, Accelerator-driven systems (ADS) and Fast Reactors (FR) in Advanced Nuclear Fuel Cycles, A Comparative Study (2002) 349. [2] M.V. Kormilitsyn, A.V. Bychkov, V.S. Ishunin, Pyroelectrochemical reprocessing of irradiated fuel of fast reactors, Global 2003, New Orleans, USA (2003) 782. [3] V.V. Smolenski, V.E. Komarov, V.K. Afonichkin, Melts 2 (2000) 59 (in Russian). [4] K. Kinoshita, T. Inoue, S.P. Fusselman, et al., J. Nucl. Sci. Technol. 36 (1999) 189. [5] M. Kurata, Y. Sakamura, T. Inoue, Proceedings of the 3rd Information Exchange Meeting on Actinide and Fission Products Partitioning and Transmutation, Cadarache, France, 1994, p. 407. [6] O. Conocar, N. Douyere, J. Lacquement, J. Alloys Compd. 389 (2005) 29. [7] O. Shirai, K. Uozumi, I. Iwai, Y. Arai, J. Appl. Electrochem. 34 (2004) 323.
V. Smolenski et al. / Electrochimica Acta 54 (2008) 382–387 [8] J. Serp, M. Allibert, A.L. Terrier, et al., J. Electrochem. Soc. 152 (2005) C167. [9] L. Martinot, Molten Salt Chemistry of Actinides, Elsevier science, North Holland, 1991, 241. [10] K. Serrano, Ph.D. thesis, Univertisé Paul Sabatier, 1998. [11] J. Serp, R.J.M. Konings, R. Malmbeck, et al., J. Electroanal. Chem. 561 (2004) 143. [12] D. Lambertin, Ph.D. thesis, Université Pierre et Marie Curie, Paris VI, 2001. [13] C. Pernel, Ph.D. thesis, Insitute National Polytechnique de Grenoble, 2002. [14] S.P. Fusselman, J.J. Roy, D.L. Grimmett, et al., J. Electrochem. Soc. 146 (1999) 2573. [15] M.R. Bermejo, J. Gómez, A.M. Martínez, et al., Electrochim. Acta 53 (2008) 5106. [16] V.V. Smolenski, A.V. Novoselova, A.L. Bovet, Russ. J. Appl. Chem. 80 (2007) 1661. [17] K.E. Johnson, J.R. Mackenzie, J. Electrochem. Soc. 116 (1969) 1697. [18] L. Guankun, T. Yexiang, H. Quichan, H. Hong, Trans. Nonferrous Met. Soc. China 8 (1998) 516. [19] S.A. Kuznetsov, M. Gaune-Escard, Proceedings of the 7th International Symposium on Molten Salts Chemistry and Technology, MS 7, 2005, p. 855. [20] M.R. Bermejo, E. Barrado, M. Herguedas et al., EUCHEM Conference on Molten Salts and Ionic Liquids, Hammamet, Tunisia, 2006, p. 127. [21] A.V. Novoselova, A.M. Potapov, V.A. Khokhlov, Proceedings of the EUCHEM Molten Salts Conference, 2004, p. 270.
387
[22] V.Yu. Shishkin, V.S. Mityaev, Inorg. Mater. 18 (1982) 1917 (in Russian). [23] M.V. Smirnov, Electrode Potentials in Molten Chlorides, Nauka, Moscow, 1973 (in Russian). [24] A.J. Bard, L.R. Faulkner, Electrochemical Methods, Fundamentals and Applications, John Wiley & Sons, New York, 1980. [25] R.S. Nicholson, J. Am. Soc. 76 (1954) 2539. [26] T. Store, Ph.D. thesis, Norwegian University of Science and Technology, Electrochemistry Department, 1999. [27] M.V. Smirnov, O.M. Shabanov, Trans. Inst. Electrochem. Ural Branch Ac. Sci. USSR 8 (1966) 55. [28] G.J. Janz, N.P. Bansal, J. Phys. Chem. Data 11 (1982) 505. [29] V.A. Lebedev, Selectivity of Liquid Metal Electrodes in Molten Halide, Metallurgia, Chelyabinsk (in Russian), 1993. [30] H. Matsuda, Y. Ayabe, Z. Electrochem. Soc. 59 (1955) 494. [31] G.N. Papatheodorou, O.J. Kleppa, J. Phys. Chem. 78 (1974) 176. [32] Yu.A. Barbanell, Coordination Chemistry of f-elements in Melts, Energoatomizdat, Moscow (in Russian), 1985. [33] H. Yamana, T. Fujii, O. Shirai, International Symposium on Ionic Liquids in Honour of Marcelle Gaune-Escard, Carry le Rouet, France, 2003, p. 123.
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Electrochemistry of ytterbium (III) in molten alkali metal chlorides V. Smolenski a , A. Novoselova a , A. Osipenko b , C. Caravaca c , G. de Córdoba c,∗ a Institute of High-Temperature Electrochemistry, Ural Division, Russian Academy of Science, Ekaterinburg, 620219, Russia b Research Institute of Atomic Reactors, Dimitrovgrad-10, Ulyanovsk Region, 433010, Russia c High Level Waste Unit, Nuclear Fission Division, CIEMAT, Madrid, 28040, Spain
a r t i c l e
i n f o
Article history: Received 23 April 2008 Received in revised form 23 June 2008 Accepted 25 July 2008 Available online 15 August 2008 Keywords: Molten alkali metal chlorides Ytterbium chloride Diffusion coefficient Voltammetry techniques Thermodynamics
a b s t r a c t This work presents the electrochemical study of Yb(III) ions in molten alkali metal chlorides in the temperature range 723–1073 K. Transient electrochemical techniques such as linear sweep, cyclic and square wave voltammetry, and potentiometry at zero current have been used to investigate the reduction mechanism, transport parameters and thermodynamic properties of the reaction YbCl2 + 1/2Cl2 = YbCl3 The results obtained show that the reduction reaction Yb(III) + e− ⇔ Yb(II) is reversible being controlled by the rate of the mass transfer. The diffusion coefficient of [YbCl6 ]3− complex ions has been determined at different temperatures in the fused eutectic LiCl–KCl, the equimolar NaCl–KCl and the CsCl media. The apparent standard potential of the soluble–soluble redox system Yb(III)/Yb(II) has been obtained by cyclic voltammetry. The influence of the nature of the solvent on the electrochemical and thermodynamic properties of ytterbium compounds is discussed. © 2008 Elsevier Ltd. All rights reserved.
1. Introduction Partitioning and Transmutation (P&T) concept is nowadays considered as one of the strategies to reduce the long-term radiotoxicity of the nuclear wastes. To achieve this, the efficient recovery and multi-recycling of actinides (An), especially TRU elements, in advanced dedicated reactors are essential. Fuels proposed to transmute the actinides into short-lived or even stable radionuclides will contain significant amounts of Pu and minor actinides (MA) (Np, Am, Cm), possibly dissolved in inert matrices (U free), and will reach high burn-ups. Pyrochemical separation techniques offer some potential advantages compared to the hydrometallurgical processes to separate actinides from fission products (FP) contained in the irradiated fuel. The high radiation stability of the salt and metallic solvents used, resulting in shorter fuel cooling times, stands out [1–5]. The aim of the separation techniques which are currently being investigated, both hydrometallurgical and pyrometallurgical ones, is to optimize the recovery efficiency of MA minimizing at the same time the FP content in the final product. Special attention is devoted
∗ Corresponding author. Tel.: +34 91 346 62 16; fax: +34 91 346 62 33. E-mail address: [email protected] (G. de Córdoba). 0013-4686/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2008.07.086
to rare earth fission products (REE) mainly due to its neutronic poison effect and the high content of these elements in the irradiated fuel. In addition, REE have similar chemical properties to those of An; hence, separation between these groups of elements is rather difficult. For this reason, a good knowledge of the basic properties of An [6–14] and REE [15–21] in the proposed separation media is very important. There is not too much information concerning the behaviour of ytterbium compounds in molten chlorides. The standard redox potentials of Yb(III)/Yb(II) versus the Pt(II)/Pt(0) reference in LiCl–KCl have been measured as well as its temperature dependence [17]. The electrochemical behaviour of Yb(III) ions on active Ni electrode in molten NaCl–KCl has been investigated at 973 K. It has been shown that the reaction occurs in two steps: the electrochemical formation of Yb(II) ions followed by the reduction of Yb(II) into metal Yb and the subsequent Ni–Yb alloys formation [18]. The electrochemistry of Yb(III) ions in molten NaCl–KCl, KCl and CsCl has been studied by transient electrochemical techniques. In these media, the apparent redox potentials of the Yb(III)/Yb(II) system as well as the diffusion coefficients of Yb(III) and Yb(II) ions were calculated [16,19]. Bermejo at al. [20] have also investigated the electrochemical behaviour of Yb(III) ions in fused LiCl–KCl. These authors have determined the kinetic and thermodynamic properties of the reaction Yb(III) + e− ↔ Yb(II). Novoselova et al. [21] have
V. Smolenski et al. / Electrochimica Acta 54 (2008) 382–387
Fig. 1. Cyclic voltammograms of fused LiCl–KCl eutectic and LiCl–KCl–YbCl3 on tungsten electrode (S = 0.31 cm2 ) at 723 K. Scan rate 0.1 V s−1 . Yb(III) concentration 2.57 × 10−2 mol kg−1 .
estimated the apparent redox potentials of the Yb(III)/Yb(II) system in molten alkali metal chlorides by using a correlation method between aqueous solutions and molten chlorides [21]. This work presents the electrochemical behaviour of Yb(III) ions in molten LiCl–KCl eutectic in the temperature range 723–973 K using transient electrochemical techniques. The experimental data obtained are compared with those in NaCl–KCl and CsCl media previously published. The influence of the nature of the solvent on the electrochemical and thermodynamic properties of ytterbium ions is discussed. 2. Experimental The experiments were carried out under inert argon atmosphere using a three electrodes electrochemical quartz sealed cell. Different transient electrochemical techniques were applied such as linear sweep (LSW), cyclic (CV) and square wave voltammetry (SWV), as well as potentiometry at zero current. The electrolyte LiCl–KCl (Aldrich, 99.9%) was dehydrated by heating under vacuum in the temperature range 293–573 K, while the solvents NaCl–KCl (Aldrich, 99.9%) and CsCl (Aldrich, 99.9%) were heated in the temperature range 293–773 K, also under vacuum. Afterwards, the salts were fused under Ar(g) atmosphere. Finally, they were purified by means of the so-called “direct crystallization method [22]. Yb(III) solutions were prepared by direct addition of anhydrous YbCl3 (Merck, 99.99%) to the electrolytic bath. The electrochemical measurements were carried out using an Autolab PGSTAT30 potentiostat (Eco-Chimie) with specific GPES electrochemical software (version 4.9).
Fig. 2. Linear sweep voltammograms of fused NaCl–KCl–YbCl3 at different sweep rates at 973 K. [Yb(III)] = 3.79 × 10−2 mol kg−1 . Working electrode: W (S = 0.27 cm2 ).
383
Fig. 3. Cyclic voltammograms of CsCl–YbCl3 at different sweep potential rates at 973 K. Working electrode: W (S = 0.31 cm2 ). [Yb(III)] = 3.70 × 10−2 mol kg−1 .
The inert working electrode consisted of a 1 mm metallic W wire (Goodfellow, 99.9%). It was immersed into the molten bath between 3 and 10 mm. The active surface area was determined after each experiment by measuring the immersion depth of the electrode. The counter electrode consisted of a 3 mm vitreous carbon rod (SU – 2000). The 2Cl− /Cl2 system is the most reliable reference electrode since it constitutes the universal reference in molten chloride media and it can be used for direct thermodynamic calculations. The electrode device consists of a graphite tube (for the chlorine gas introduction) dipped into a quartz tube provided with a porous membrane at its bottom and containing molten chloride solvent. Chlorine gas is bubbled through the melt during the experiment [23]. The ytterbium concentration in solution was determined by taking samples from the melt which were dissolved in nitric acid solutions and then analyzed by ICP-MS. 3. Results and discussion 3.1. Study by voltammetric techniques Cyclic voltammetry was carried out on inert tungsten for all melts tested: LiCl–KCl, NaCl–KCl and CsCl, and at several temperatures (723–1073 K). The electroactivity domain in LiCl–KCl at 723 K is shown in Fig. 1. The cathodic and anodic limits of the electrochemical window correspond to the reduction of the alkali metal ions of the solvent (Li+ /Li0 , Na+ /Na0 and Cs+ /Cs0 ) and to the oxidation of chloride ions into chlorine gas (2Cl− /Cl2 ), respectively.
Fig. 4. Square wave voltammogram of LiCl–KCl–YbCl3 (9.41 × 10−2 mol kg−1 ) at 10 Hz at 723 K. Working electrode: W (S = 0.25 cm2 ).
384
V. Smolenski et al. / Electrochimica Acta 54 (2008) 382–387
Fig. 5. Variation of the cathodic and anodic peak potential as a function of the sweep rate in fused LiCl–KCl–YbCl3 (9.41 × 10−2 mol kg−1 ) at 723 K. Working electrode: W (S = 0.25 cm2 ).
Fig. 6. Variation of the cathodic and anodic peak current as a function of the potential scan rate in fused LiCl–KCl–YbCl3 (9.41 × 10−2 mol kg−1 ) at 723 K. Working electrode: W (S = 0.25 cm2 ).
Fig. 1 also plots the cyclic voltammogram of a LiCl–KCl–YbCl3 solution on W at 723 K. It shows a single cathodic peak at a potential of −1.762 V vs. 2Cl− /Cl2 and its corresponding anodic peak at −1.566 V vs. the 2Cl− /Cl2 . A similar behaviour for the reduction of Yb(III) ions has been observed in the fused equimolar NaCl–KCl mixture and CsCl. Fig. 2 shows a linear sweep voltammogram of the system NaCl–KCl–YbCl3 registered at different potential scan rates. A single cathodic peak is observed. The analysis of the peak indicates that only one electron is exchanged being associated to the reduction reaction Yb(III) + e− ↔ Yb(II). Concerning the system CsCl–YbCl3 , the cyclic voltammograms recorded at several scan rates are plotted in Fig. 3. The shape of the CV is typical for a soluble–soluble system suggesting that in this case the system corresponds to the reaction Yb(III)/Yb(II). The square wave voltammetry technique has been applied to determine the number of electrons exchanged in the reduction of Yb(III) ions in LiCl–KCl. Fig. 4 shows the Gaussian-shaped symmetric cathodic wave obtained in the LiCl–KCl–YbCl3 at 723 K. The number of electrons exchanged is determined by measuring the width at half height of the reduction peak, W1/2 (V), recorded at different frequencies (6–80 Hz), using the following equation, valid for reversible systems [10,24]:
independent of the potential scan rate (see Fig. 5); (ii) The cathodic and anodic peak currents (Ip ) are directly proportional to the square root of the polarization rate () (see Fig. 6); (iii) A linear relationship between the cathodic peak current density and the concentration of YbCl3 ions in the melt is observed (see Fig. 7). From these results and, according to the theory of the linear sweep voltammetry technique [25], it can be concluded that the redox system Yb(III)/Yb(II) is reversible and it is controlled by the rate of the mass transfer. From the transient electrochemical techniques applied, it is found that the system Yb(II)/Yb(0) cannot be observed in molten alkali chlorides media since its potential is more negative than the reduction potential of the solvent Me(I)/Me(0), Me being: Li, Na, K and Cs, (see Fig. 1).
W1/2 = 3.52
RT nF
(1)
At low frequencies, a linear relationship between the cathodic peak current and the square root of the frequency is found. Under these conditions, the system can be considered as reversible; hence it is possible to apply Eq. (1). The number of electrons exchanged was found close to one. Similar results were obtained in NaCl–KCl and CsCl media. Potentiostatic electrolysis at potentials corresponding to the cathodic peaks for all systems studied did not show the formation of any solid phase at the tungsten surface after its polarization. No plateaus were observed in the potential–time curves. Also the working electrode did not undergo any visual change. In addition, the X-ray analysis of the surface of the working electrodes showed the absence of any solid phase. The results obtained allow concluding that the reduction of Yb(III) ions takes place in a single step with the exchange of one electron and the formation of a soluble product, according to the following reaction: Yb(III) + e− ⇔ Yb(II)
3.2. Diffusion coefficient of Yb(III) ions The diffusion coefficient of [YbCl]3− 6 ions in molten chloride media was determined using the cyclic voltammetry technique and applying the Randles–Sevˇcik equation, valid for reversible soluble–soluble systems [24]: Ip = 0.446(nF)3/2 C0 S
Dv 1/2 RT
(3)
where S is the electrode surface area (cm2 ), C0 is the solute concentration (mol cm−3 ), D is the diffusion coefficient (cm2 s−1 ), F is the Faraday constant (96,500 C mol−1 ), R is the ideal gas constant (J K−1 mol−1 ), n is the number of exchanged electrons, is the potential sweep rate (V s−1 ) and T is the absolute temperature (K). The values obtained for the different solvents studied at several
(2)
The reaction mechanism of the soluble–soluble Yb(III)/Yb(II) exchange was studied by analyzing the voltammetric curves recorded at several potential scan rates. The analysis shows that: (i) The cathodic and anodic peak potentials (Ep ) are constant and
Fig. 7. Variation of the cathodic peak current as a function of the concentration YbCl3 in fused LiCl–KCl–YbCl3 at 723 K. Working electrode: W (S = 0.25 cm2 ). Scan rate = 0.1 V s−1 .
V. Smolenski et al. / Electrochimica Acta 54 (2008) 382–387
385
Table 1 Diffusion coefficient of [YbCl6 ]3− ions in molten alkali metal chlorides at several temperatures (Activation energy for the ytterbium ions diffusion process) Solvent
T (K)
D × 105 (cm2 s−1 )
D × 105 (cm2 s−1 ) [19]
−EA (kJ mol−1 )
−EA (kJ mol−1 ) [19]
Do × 103 (cm2 s−1 )
Do × 103 (cm2 s−1 ) [19]
LiCl–KCl
723 848 973
1.0 ± 0.1 2.7 ± 0.1 5.4 ± 0.1
– – 5.3 ± 0.2*
38.3
–
6.2
–
NaCl–KCl
973 1023 1073
2.8 ± 0.2 3.2 ± 0.2 4.1 ± 0.2
2.4 ± 0.2 2.7 ± 0.2 3.4 ± 0.2
45.4
46.4
6.6
6.2
CsCl
973 1023 1073
0.9 ± 0.1 1.2 ± 0.1 1.7 ± 0.1
0.9 ± 0.2 1.2 ± 0.2 1.6 ± 0.2
54.4
55.6
7.4
8.5
*
Extrapolated value.
temperatures are quoted in Table 1. The experimental values have also been compared with those reported in the literature. The diffusion coefficient values have been used to calculate the activation energy for the diffusion process. The influence of temperature on the diffusion coefficient obeys the Arrhenius’s law through the following equation [26]:
recorded in the YbCl3 solutions in the different alkali metal chlorides tested. According to the theory of linear sweep voltammetry, the following expressions, including the anodic and cathodic peak potentials and the half-wave potential, can be applied in the case of soluble–soluble reversible systems [19,30]:
D = Do exp −
(4)
EPC = E1/2 − 1.11
RT F
(6)
where EA is the activation energy for the diffusion process (kJ mol−1 ), Do is the pre-exponential term (cm2 s−1 ) and is the experimental error. From this expression, the value of the activation energy for the Yb(III) ions diffusion process was calculated in the different melts tested (see Table 1). The diffusion coefficient of ytterbium (III) ions becomes smaller as the radius of the alkali metal cation increases from Li to Cs (see Table 1). Such behaviour takes place due to an increase on the strength of complex ions and a decrease on the contribution of D to the “hopping” mechanism [27]. A similar trend for the diffusion coefficient has been reported elsewhere [28]. As the temperature increases the diffusion coefficient increases in all the solvents. The variation of the logarithm of the diffusion coefficient as a function of the reverse of the solvent cation radius (r) at 973 K can be expressed as follows (see Fig. 8):
EPA = E1/2 + 1.11
RT F
(7)
E A
log D[YbCl
3−
6]
RT
±
0.17 = −6.07 − ± 0.02 r
(5)
The ionic radius values have been taken from the literature [29]. 3.3. Apparent standard potential of the redox system Yb(III)/Yb(II) The apparent standard potential of the Yb(III)/Yb(II) system has been determined from the analysis of the cyclic voltammograms
(EPC + EPA ) 2
= E1/2
(8)
where the half-wave potential is given by: E1/2 =
0 EYb(III)/Yb(II)
RT + ln F
DYb(II)
1/2
DYb(III)
RT + ln F
Yb(III) Yb(II)
(9)
It is known that for concentrations of electroactive species lower than 3–5 × 10−2 in mole fraction scale, their activity coefficient is almost constant. Under these conditions, it is more convenient ∗ ) using the apparent standard redox potential concept (EYb(III)/Yb(II) expressed as follows [23]: ∗ 0 = EYb(III)/Yb(II) + EYb(III)/Yb(II)
RT ln F
Yb(III)
Yb(II)
(10)
Considering that the diffusion coefficient values for the triand bivalent ytterbium ions are close, it can be assumed that the term (RT/F)ln(DYb(II) /DYb(III) )1/2 tends to zero. As a consequence, the apparent standard potential of the redox system Yb(III)/Yb(II) can be calculated from the equations: ∗ = EPC + 1.11 EYb(III)/Yb(II)
RT F
(11)
∗ = EPA − 1.11 EYb(III)/Yb(II)
RT F
(12)
∗ = EYb(III)/Yb(II)
(EpC + EpA ) 2
(13)
From the peak potential values measured in the cyclic voltammograms, the following empirical equation for the apparent standard potential of the Yb(III)/Yb(II) system in LiCl–KCl has been obtained (potentials are expressed vs. the 2Cl− /Cl2 reference system): ∗ = −(1.915 ± 0.005) + (3.5 ± 0.2) × 10−4 T (V) EYb(III)/Yb(II)
[723 − 973 K]
Fig. 8. Linear relationship of logarithm of the [YbCl6 ]3− diffusion coefficient versus the reverse of the radius of the solvent cation at 973 K.
(14)
Lanthanide chlorides dissolved in alkali chloride melts are usually solvated by chloride ions forming different complex ions
386
V. Smolenski et al. / Electrochimica Acta 54 (2008) 382–387
Table 2 Apparent standard redox potential of the Yb(III)/Yb(II) system in molten alkali metal chlorides at several temperatures Solvent
T (K)
∗ −EYb(III)/Yb(II) (V)
∗ −EYb(III)/Yb(II) (V) [17]
∗ −EYb(III)/Yb(II) (V) [19]
∗ −EYb(III)/Yb(II) (V) [20]
LiCl–KCl
723 848 973
1.662 1.618 1.574
1.704 1.618 1.532*
– – 1.531*
1.659 – –
NaCl–KCl
973 1023 1073
1.671 1.652 1.634
– – –
1.678 1.662 1.646
1.629 – 1.551
CsCl
973 1023 1073
1.853 1.832 1.811
– – –
1.871 1.844 1.817
1.881 – 1.792
Reference system: Cl− /Cl2 . * Extrapolated value.
such as [LnCl6 ]3− and [LnCl4 ]2− [31–33]. In the case of ytterbium, [YbCl6 ]3− complex ions are present in the melt [21]. Their relative stability increases with the increase of the solvent cation radius, and the apparent standard redox potential shifts to more negative values. The results obtained here are in agreement with those reported by Smirnov [23]. The variation of the apparent standard potential of the redox couple Yb(III)/Yb(II) as a function of the reverse of the solvent cation radius (r) has been calculated at 973 K. The relation obtained is: ∗ = −2.124 + EYb(III)/Yb(II)
0.052 ± 0.003 (V) r
(15)
3.4. Thermodynamic properties
(16)
has been calculated according to following expression: (17)
Its temperature dependence allows calculating the enthalpy and entropy of the redox reaction (16) by means of the relation [24]: G∗ = H ∗ − TS ∗
[3LiCl − 2 KCl; 723 − 973 K]
(19)
G∗ = −195.96 + 0.036T ± 2.46 kJ/mol [NaCl − KCl; 973 − 1073 K] G∗ = −218.25 + 0.041T ± 2.46 kJ/mol
(20) [CsCl; 973 − 1073 K]
The variation of the thermodynamic parameters for the redox reaction (16) as a function of the solvent cation radius shows an increase of the strength of the Yb Cl bonds in the complex ions [YbCl6 ]3− . 4. Conclusions
The apparent standard Gibbs energy (G*) of the redox reaction:
∗ G∗ = −nFEYb(III)/Yb(II)
G∗ = −184.80 + 0.033T ± 2.46 kJ/mol
(21)
The experimental values of the apparent standard potential of the Yb(III)/Yb(II) system at several temperatures as well as those reported in the literature are gathered in Table 2 and plotted in Fig. 9.
YbCl2(l) + 12 Cl2(g) ⇔ YbCl3(l)
The apparent standard Gibbs energy of the redox reaction (16) in the different solvents tested can be expressed as:
(18)
The electrochemical behaviour of [YbCl6 ]3− complex ions in the molten LiCl–KCl eutectic was investigated. It was found that the reduction of Yb(III) to Yb(II) ions is a reversible process controlled by the rate of the mass transfer. The diffusion coefficient of [YbCl6 ]3− species was determined at different temperatures. The apparent standard electrode potential of the redox couple Yb(III)/Yb(II) was calculated from the analysis of the cyclic voltammograms registered at different temperatures (E* Yb(III)/Yb(II) = −1.662 V vs. 2Cl− /Cl2 at 723 K). In addition, the influence of the nature of the solvent cation (ionic radius) on the thermodynamic properties of Yb(III) ions was studied. It was found that the strength of the Yb Cl bonds increases as we shift from Li to Cs cation. Acknowledgement The authors wish to thank Dr. A. Bovet for the technical support. References
Fig. 9. Apparent standard potential of the redox couple Yb(III)/Yb(II) versus the reverse of the radius of the solvent cation at 973 K.
[1] OCDE/NEA Report, Accelerator-driven systems (ADS) and Fast Reactors (FR) in Advanced Nuclear Fuel Cycles, A Comparative Study (2002) 349. [2] M.V. Kormilitsyn, A.V. Bychkov, V.S. Ishunin, Pyroelectrochemical reprocessing of irradiated fuel of fast reactors, Global 2003, New Orleans, USA (2003) 782. [3] V.V. Smolenski, V.E. Komarov, V.K. Afonichkin, Melts 2 (2000) 59 (in Russian). [4] K. Kinoshita, T. Inoue, S.P. Fusselman, et al., J. Nucl. Sci. Technol. 36 (1999) 189. [5] M. Kurata, Y. Sakamura, T. Inoue, Proceedings of the 3rd Information Exchange Meeting on Actinide and Fission Products Partitioning and Transmutation, Cadarache, France, 1994, p. 407. [6] O. Conocar, N. Douyere, J. Lacquement, J. Alloys Compd. 389 (2005) 29. [7] O. Shirai, K. Uozumi, I. Iwai, Y. Arai, J. Appl. Electrochem. 34 (2004) 323.
V. Smolenski et al. / Electrochimica Acta 54 (2008) 382–387 [8] J. Serp, M. Allibert, A.L. Terrier, et al., J. Electrochem. Soc. 152 (2005) C167. [9] L. Martinot, Molten Salt Chemistry of Actinides, Elsevier science, North Holland, 1991, 241. [10] K. Serrano, Ph.D. thesis, Univertisé Paul Sabatier, 1998. [11] J. Serp, R.J.M. Konings, R. Malmbeck, et al., J. Electroanal. Chem. 561 (2004) 143. [12] D. Lambertin, Ph.D. thesis, Université Pierre et Marie Curie, Paris VI, 2001. [13] C. Pernel, Ph.D. thesis, Insitute National Polytechnique de Grenoble, 2002. [14] S.P. Fusselman, J.J. Roy, D.L. Grimmett, et al., J. Electrochem. Soc. 146 (1999) 2573. [15] M.R. Bermejo, J. Gómez, A.M. Martínez, et al., Electrochim. Acta 53 (2008) 5106. [16] V.V. Smolenski, A.V. Novoselova, A.L. Bovet, Russ. J. Appl. Chem. 80 (2007) 1661. [17] K.E. Johnson, J.R. Mackenzie, J. Electrochem. Soc. 116 (1969) 1697. [18] L. Guankun, T. Yexiang, H. Quichan, H. Hong, Trans. Nonferrous Met. Soc. China 8 (1998) 516. [19] S.A. Kuznetsov, M. Gaune-Escard, Proceedings of the 7th International Symposium on Molten Salts Chemistry and Technology, MS 7, 2005, p. 855. [20] M.R. Bermejo, E. Barrado, M. Herguedas et al., EUCHEM Conference on Molten Salts and Ionic Liquids, Hammamet, Tunisia, 2006, p. 127. [21] A.V. Novoselova, A.M. Potapov, V.A. Khokhlov, Proceedings of the EUCHEM Molten Salts Conference, 2004, p. 270.
387
[22] V.Yu. Shishkin, V.S. Mityaev, Inorg. Mater. 18 (1982) 1917 (in Russian). [23] M.V. Smirnov, Electrode Potentials in Molten Chlorides, Nauka, Moscow, 1973 (in Russian). [24] A.J. Bard, L.R. Faulkner, Electrochemical Methods, Fundamentals and Applications, John Wiley & Sons, New York, 1980. [25] R.S. Nicholson, J. Am. Soc. 76 (1954) 2539. [26] T. Store, Ph.D. thesis, Norwegian University of Science and Technology, Electrochemistry Department, 1999. [27] M.V. Smirnov, O.M. Shabanov, Trans. Inst. Electrochem. Ural Branch Ac. Sci. USSR 8 (1966) 55. [28] G.J. Janz, N.P. Bansal, J. Phys. Chem. Data 11 (1982) 505. [29] V.A. Lebedev, Selectivity of Liquid Metal Electrodes in Molten Halide, Metallurgia, Chelyabinsk (in Russian), 1993. [30] H. Matsuda, Y. Ayabe, Z. Electrochem. Soc. 59 (1955) 494. [31] G.N. Papatheodorou, O.J. Kleppa, J. Phys. Chem. 78 (1974) 176. [32] Yu.A. Barbanell, Coordination Chemistry of f-elements in Melts, Energoatomizdat, Moscow (in Russian), 1985. [33] H. Yamana, T. Fujii, O. Shirai, International Symposium on Ionic Liquids in Honour of Marcelle Gaune-Escard, Carry le Rouet, France, 2003, p. 123.