Redox electrochemistry and formal standard redox potentials of the Eu(III)/Eu(II) redox couple in an equimolar mixture of molten NaCl-KCl

Article abstract of DOI:10.1016/S0013-4686(00)00708-8

The electrochemical redox process Eu(III)+e^-?Eu(II) in an equimolar NaCl-KCl melt in the temperature range 973-1123 K on glassy carbon electrodes was studied by linear sweep voltammetry, cyclic voltammetry, chronopotentiometry and reversal chronopotentiometry. It was determined that at a sweep rate of ν≤0.1 V s^-1, the electroreduction of Eu(III) to Eu(II) is reversible, but at 0.1<ν<0.3 V s^-1, mixed diffusion and electron-transfer control is observed. Further increase of polarization rate, ν>0.5 V s^-1 results in electron-transfer control. The diffusion coefficients of Eu(III) and Eu(II) were determined by linear sweep voltammetry and chronopotentiometry methods. The values found by these methods are in a good agreement with each other. The diffusion coefficients of Eu(III) and Eu(II) in the NaCl-KCl melt are discussed in connection with the strength and stability of these complex ions. The standard rate constants for the reduction of Eu(III) to Eu(II) were calculated on the basis of cyclic voltammetry data. The sluggish kinetics of this reaction is discussed in terms of substantial rearrangement of the europium coordination sphere. The formal standard redox potentials of E_{Eu(III)/Eu(II)}* were determined from linear sweep voltammetry and cyclic voltammetry.

Full text of DOI:10.1016/S0013-4686(00)00708-8

Electrochimica Acta 46 (2001) 1101–1111
www.elsevier.nl/locate/electacta
Redox electrochemistry and formal standard redox
potentials of the Eu(III)/Eu(II) redox couple in an
equimolar mixture of molten NaClꢀKCl
a,
b
S.A. Kuznetsov *, M. Gaune-Escard
a
Institute of Chemistry Kola Science Centre RAS, 184200 Apatity, Murmansk region, Russia
b
IUSTI-CNRS UMR 6595, Technopole Chateau-Gombert, Uni6ersity of Pro6ence, 5 rue Enrico Fermi,
F-13453 Marseille Cedex 13, France
Received 14 April 2000; received in revised form 4 October 2000
Abstract
−
The electrochemical redox process Eu(III)+e UEu(II) in an equimolar NaClꢀKCl melt in the temperature range
9
73–1123 K on glassy carbon electrodes was studied by linear sweep voltammetry, cyclic voltammetry, chronopoten-
−
1
tiometry and reversal chronopotentiometry. It was determined that at a sweep rate of w50.1 V s , the
−
1
electroreduction of Eu(III) to Eu(II) is reversible, but at 0.1Bw50.3 V s , mixed diffusion and electron–transfer
−
1
control is observed. Further increase of polarization rate, w]0.5 V s
results in electron–transfer control. The
diffusion coefficients of Eu(III) and Eu(II) were determined by linear sweep voltammetry and chronopotentiometry
methods. The values found by these methods are in a good agreement with each other. The diffusion coefficients of
Eu(III) and Eu(II) in the NaClꢀKCl melt are discussed in connection with the strength and stability of these complex
ions. The standard rate constants for the reduction of Eu(III) to Eu(II) were calculated on the basis of cyclic
voltammetry data. The sluggish kinetics of this reaction is discussed in terms of substantial rearrangement of the
europium coordination sphere. The formal standard redox potentials of E*Eu(III)/Eu(II) were determined from linear
sweep voltammetry and cyclic voltammetry. © 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Europium; Molten salts; Diffusion coefficient; Kinetics of electrode reaction; Formal standard potentials
1
. Introduction
cause lanthanide elements which are present in spent
fuel from fast nuclear reactors can be converted into
molten salts by anodic dissolution [6]. The actinides are
selectively deposited at the cathode due to the differ-
ences among the redox potentials of the elements while
fission products remain in the anode and in the elec-
trolyte [7,8]. The lanthanide elements are the most
difficult fission products to separate from actinides due
to their similar chemical properties. Not withstanding
the active research, especially on rare-earth metal elec-
trowinning and electrorefining, the important funda-
Rare-earth metals of high purity are widely used now
in different fields of modern techniques. Electrowinning
and electrorefining in molten salts is a promising
method for production of high-purity rare-earth metals
[1–4]. Electrochemical synthesis is an effective method
for obtaining alloys of rare-earth metals in melts [5].
Another important problem concerning rare-earth
metals and molten salts is recycling of spent fuel be-
mental
problem
of
electrochemistry
and
*
Corresponding author. Fax: +47-78914131.
E-mail
Kuznetsov).
address:
kuznet@chemy.kolasc.net.ru
(S.A.
thermodynamics of some rare-earth metals in melts has
not been addressed in depth.
0013-4686/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0013-4686(00)00708-8

1102
S.A. Kuznetso6, M. Gaune-Escard / Electrochimica Acta 46 (2001) 1101–1111
Thus a knowledge of the electrochemistry and formal
argon. The salts were handled in the glove box and
stored in sealed glass ampoules, as explained above.
The total concentration of europium was determined by
inductively coupled plasma atomic emission spec-
troscopy (ICP-AES) and potentiometric titration with
potassium dichromate used for determination of Eu(II)
in quenched samples.
standard redox potentials of europium (III, II) in
molten salts provide the data for optimization of eu-
ropium production by electrolysis and is very useful for
understanding the recycling of spent nuclear fuels.
Several electrochemical studies in molten salts were
carried out using europium trichloride (EuCl ). The
3
electrochemistry of europium (III) trichloride was stud-
ied by Gilbert [9] in molten AlCl ꢀNaCl up to a tem-
2.2. Procedures
3
perature 250°C. It was found that Eu(III) complexes
could be reduced reversibly to stable Eu(II) species in
Equimolar quantities of sodium and potassium chlo-
rides were mixed, placed in an ampoule fabricated of
glassy carbon of the SU-2000 type, and transferred to a
hermetically sealed retort of stainless steel. The latter
acidic AlCl ꢀNaCl melts, it appears to be insoluble in
3
basic AlCl ꢀNaCl melts. Electrochemical behavior of
3
Eu(III) in room temperature molten salts was studied in
several investigations [10,11]. Information about Eu(II)
chloride is very poor in the literature [12]. In a previous
study [13], standard redox potentials of the Eu(III)/
Eu(II) redox reaction in the eutectic LiClꢀKCl melt
were determined. The formal standard potentials
−
3
was evacuated to a residual pressure of 5×10
Torr,
first at room temperature and then gradually heating to
473, 673 and 873 K. After this the retort was filled with
high purity argon and the electrolyte was melted.
Electrochemical studies were performed employing
linear sweep voltammetry (LSV), cyclic voltammetry
(CV), chronopotentiometry (CP) and reversal
chronopotentiometry (RCP) methods using a Volta-
Lab-40 potentiostat with packaged software ‘VoltaMas-
ter 4’. The potential scan rate was varied between
E*
Eu(II)/Eu
from measurements of equilibrium potentials
of europium-lead alloys in
a
eutectic mixture
NaClꢀCsCl were obtained in [14]. To our best knowl-
edge no detailed electrochemistry and formal standard
redox potentials of E*Eu(III)/Eu(II) in an equimolar
NaClꢀKCl melt have previously been reported. In the
present study on the electrochemistry of Eu(III) and
Eu(II), formal standard redox potentials E*Eu(III)/Eu(II) in
an equimolar NaClꢀKCl melt have been determined.
−
3
−1
5×10
and 5.0 V s . The experiments were carried
out in a temperature range 973–1123 K. Cyclic voltam-
metric curves and chronopotentiograms were deter-
mined for 0.8–2.0 mm diameter glassy carbon and
platinum electrodes with respect to a glassy carbon
plate as a quasi- reference electrode, and versus the
silver reference electrode Ag/NaClꢀKClꢀAgCl (9.4×
2
. Experimental
−
3
10
mole fraction). The glassy carbon ampoule served
2
.1. Chemicals
as the counter electrode. Potentials versus silver refer-
ence electrode were converted to a Cl /Cl2 reference
−
Europium dichloride was synthesized from the oxide
electrode using the relation
Eu O3 (Johnson Matthey, 99.9%). Thionyl chloride
2
−4
E/V= −1.111–1.31×10 T/K
(1)
(
Johnson Matthey, 99%) was used as a chlorinating
agent during 6 h at the temperature of 823 K under
argon flow. EuCl was obtained in the first step of this
3
3. Results and discussion
synthesis. Reduction to EuCl was performed by zinc at
2
7
73 K during 3 h under dynamic vacuum. The temper-
3
.1. Cyclic 6oltammetry
ature was then increased to 1093 K and maintained
constant for 5 h under static vacuum. Finally, the
EuCl compound was purified from zinc by distillation
The
cyclic
voltammetric
curves
in
the
2
NaClꢀKClꢀEuCl3 melt obtained at the glassy carbon
electrode are presented in Fig. 1. Wave 1 is observed in
the cathodic-anodic region that indicates an appearance
in the melt of Eu(II) apparently due to reaction
at 1193 K. Due to the highly hygroscopic properties of
lanthanide compounds, EuCl and EuCl were stored in
3
2
sealed glass ampoules under vacuum. All further han-
dling of europium chlorides and filling of experimental
cells were performed in a controlled purified argon
atmosphere glove-box containing less than 2 ppm of
water.
2
EuCl U2EuCl +Cl
(2)
3
2
2
This is not an unexpected result because it is known
Alkali chlorides (NaCl and KCl) were purchased
from Prolabo (99.5% min). They were dehydrated by
continuous and progressive heating just above the melt-
ing point under gaseous HCl atmosphere in quartz
ampoules. Excess HCl was removed from the melt by
from [15] that the trichloride of europium starts to
decompose into dichloride and chlorine in the solid
phase at temperatures above 300°C.
The potentiostatic electrolysis at potentials of the
cathodic peak does not lead to the formation of a solid

S.A. Kuznetso6, M. Gaune-Escard / Electrochimica Acta 46 (2001) 1101–1111
1103
(6)
phase at the electrode, and the electrode itself under-
goes no visible transformation. This means that the
product of this stage is soluble in the melt. It was found
that the peak current is directly proportional to the
square root of the polarization rate (Fig. 2a), while the
peak potential does not depend on the polarization rate
C
3/2 −1/2 −1/2 3/2
ACD1/2w1/2
I =0.4463F
R
T
n
p
C
where Ip is the peak cathodic current (A), A is the
electrode area (cm ), C is the bulk concentration of
active species (mol cm ), D is the diffusion coefficient
(cm s ), w is the potential sweep rate (V s ), and n
is the number of electrons involved in the reaction.
The coefficients are described by the following empir-
ical dependence
2
−
3
2
−1
−1
−
1
up to w=0.1 V s
(Fig. 2b). The peak current of the
electroreduction process is linearly dependent on the
EuCl3 concentration, while the peak potential of the
first stage does not depend on the concentration of
europium trichloride in the melt. According to the
theory of linear sweep voltammetry [16], up to a polar-
log DEu(III)= −2.42−2152/T90.03
(7)
Eq. (6) was employed for calculation of DEu(II) in the
NaClꢀKClꢀEuCl melt on the basis of the peak current
determined for the process
−
1
ization rate of 0.1 V s
the electrode process is
2
controlled by the rate of mass transfer. The number of
electrons for the first stage was calculated by means of
the following equations [17]
−
Eu(II)−e “Eu(III)
(8)
−
1
At a polarization rate of 0.1 V s
in the tempera-
C
p/2
C
p
E
−E =2.2RT/nF
(3)
(4)
ture range 973–1123 K, the experimental values for the
diffusion coefficients (Table 3) were satisfactorily fitted
to the linear relation
A
p
C
p
E −E =2.22RT/nF
C
C
where Ep is the potential of the cathodic peak, Ep/2 is
the potential of the half-peak, n is number of electrons,
and Ep is the potential of the anodic peak.
The calculation of the number of electrons (Table 1)
reveals that one electron is transferred at the first stage:
log DEu(II)= −2.31−1983/T90.03
3.2. Chronopotentiometry
(9)
A
A chronopotentiogram for the electroreduction of
Eu(III) to Eu(II) in the NaClꢀKClꢀEuCl3 melt at a
glassy carbon electrode is shown in Fig. 3. Straight lines
−
Eu(III)+e UEu(II)
(5)
An analysis of semi-logarithmic dependencies E ver-
sus log (I −I) and E versus log [(I −I)/I] of steady-
state voltammograms indicates that the dicharge is
described by the Heyrovsky–Ilkovich equation [18],
and the number of electrons involved in this step is
essentially unity.
The diffusion coefficients (D) for the chloride com-
plexes of Eu(III) were determined at w=0.1 V s
Table 2) using of the Randles–Shevchik equation [19]
1/2
for plots of i~
concentrations of Eu(III) (Fig. 4). Fig. 5 shows the
versus i were obtained for various
d
d
1/2
linear relationship between i~
and C. Moreover the
line goes through the origin of coordinates. The
chronopotentiometric curve did not shift in the direc-
tion of negative potentials with increasing current den-
sity. Thus, on the bases of chronopotentiometric data it
is possible to conclude that the electrochemical reduc-
−
1
(
2
Fig. 1. Cyclic voltammograms (7 cycles) at a glassy carbon electrode in NaClꢀKClꢀEuCl melt. Area: 0.18 cm . Sweep rate: 0.15 V
3
−
1
−5
−3
−
s
. Temperature: 973 K. Concentration of EuCl : 6.64×10
mol cm . Reference electrode: Cl /Cl .
3
2

1104
S.A. Kuznetso6, M. Gaune-Escard / Electrochimica Acta 46 (2001) 1101–1111
Fig. 2. The dependencies of peak currents (a) and of peak potentials (b) of the recharge process on the polarization rate. Area: 0.19
2
−5
−3
−
cm . Temperature: 973 K. Concentration of EuCl : 9.0×10
mol cm . Reference electrode: Cl /Cl .
3
2
1
/2
tion of Eu(III) to Eu(II) is a reversible and diffusion
controlled process [17]. The chronopotentiometric curve
for the reversible electroreduction with formation a
soluble substance is described by the following equation
It was determined that the function E versus log [(~
1/2
1/2
−t )/t ] is a straight line with a slope which is very
close to the value 2.3RT/F corresponding to a one
electron transfer.
[17]
The diffusion coefficients of Eu(III) and Eu(II) in the
1
/2
NaClꢀKClꢀEuCl
3
and NaClꢀKClꢀEuCl
2
melts were de-
termined utilizing the Sand equation [18]
E=E°Eu(III)/Eu(II)+RT/nF ln(Dred/Dox)
1/2
1/2
1/2
+
RT/nF ln(k /kred)+RT/nF ln[(~ −t )/t
ox
]
I~ =(nFCD1/2y1/2A)/2
1/2
(11)
(
10)
where: D , D
diffusion coefficients and the activity coefficients of the
oxidized and reduced forms, and the transition time ~.
and kox, kred are, respectively, the
ox
red
2
where I is the current (A), A is the area (cm ) of the
working electrode, ~ is the transition time (s), C is the

S.A. Kuznetso6, M. Gaune-Escard / Electrochimica Acta 46 (2001) 1101–1111
1105
−
3
bulk concentration of the reactant (mol cm ), and D
log DEu(III)= −2.30−2272/T90.04
log DEu(II)= −2.34−1956/T90.02
(12)
(13)
2
−1
is the diffusion coefficient of the reactant (cm s ).
The main difficulty in electrochemical methods arises
from the determination of surface area of the working
electrode due to wetting between the electrode and
molten salt. In some experiments for determination of
diffusion coefficients we used the increment of the
surface area for different immersion lengths [20]. The
diffusion coefficients determined with using of surface
area and increment of surface area were in a good
agreement altogether and difference in values did not
exceed the error of experiment. Probably, this is due to
insignificant effect of wetting of working electrode by
the melt.
The current reversal chronopotentiograms were also
used for calculation of DEu(III) and DEu(II). The diffusion
coefficients are described by the following empirical
dependencies
log DEu(III)= −2.41−2171/T90.03
log DEu(II)= −2.36−1942/T90.02
All these results are given in Tables 2 and 3.
(14)
(15)
3.3. Determination of the standard rate constants of
−
the electrode reaction Eu(III)+e UEu(II)
The diffusion coefficients of Eu(III) and Eu(II) calcu-
lated by Sand equation in NaClꢀKCl are formulated as
follows
Analyzing the dependencies presented in Fig. 2 shows
that at 0.1Bw50.3 V s , process (5) is quasi-
−
1
Table 1
a
Experimental and calculated data for determination of number electrons transferred at the first stage of electroreduction
V (V s−1)
E
C
p
(V)
E
C
p/2
(V)
E
C
−E Cp (V)
p/2
E
A
p
(V)
E −E (V)
A
C
Eq. (3)
Eq. (4)
p
p
0.025
0.040
0.050
0.075
0.100
0.200
−0.846
−0.843
−0.847
−0.845
−0.843
−0.861
−0.646
−0.645
−0.649
−0.644
−0.649
−0.632
0.200
0.198
0.198
0.201
0.194
0.229
−0.642
−0.639
−0.639
−0.630
−0.626
−0.523
0.204
0.204
0.208
0.215
0.217
0.338
0.97
0.98
0.98
0.96
1.00
0.96
0.96
0.94
0.91
0.90
a
Melt NaClꢀKClꢀEuCl3 (8.84×10−5 mol cm−3), T=1023 K.
Table 2
Experimental values of diffusion coefficients (D×10 cm s ) and activation energies for diffusion (DU, kJ mol−1) of Eu(III) in
−
5
2 −1
an equimolar NaClꢀKCl melt
Method
973 (K)
1023 (K)
1073 (K)
1100 (K)
DU (kJ mol−1)
LSV1
LSV2
CP
2.35
2.26
2.28
2.27
3.02
2.88
3.04
2.90
3.78
3.61
3.72
3.60
4.23
4.05
4.30
4.14
41.2
43.2
43.5
41.6
RCP
Table 3
Experimental values of diffusion coefficients (D×10 cm s ) and activation energy for diffusion (DU, kJ mol−1) of Eu(II) in an
−
5
2 −1
equimolar NaClꢀKCl melt
Method
LSV a
LSV
CP
973 (K)
1023 (K)
1073 (K)
1100 (K)
DU (kJ mol−1)
4.46
4.35
4.51
4.43
5.61
5.45
5.64
5.52
6.90
6.70
7.02
6.88
7.67
7.47
7.64
7.46
38.0
37.8
37.5
37.2
b
RCP
a
LSV-polarization rate 0.1 V s−1
LSV-polarization rate 1.0 V s
.
.
b
−1

1106
S.A. Kuznetso6, M. Gaune-Escard / Electrochimica Acta 46 (2001) 1101–1111
−
Fig. 3. Chronopotentiogram of the process Eu(III)+e “Eu(II) at a glassy carbon electrode. Current density: 16.2 mA cm−2.
−
5
−3
Temperature: 1023 K. Concentration of EuCl : 6.32×10
mol cm . Quasi-reference electrode: glassy carbon.
3
A
C
reversible. A mixed diffusion and electron- transfer
The dependencies E −E on the function c re-
p p
−
1
rate-controlles process (5) at w\0.1 V s
is indicated
ported in [21] for a temperature of 298 K must be
recalculated for the present working temperatures, and
we used the following equations [22]
by the deviation of experimental points from linearity
C
1/2
in the Ip versus w plot (Fig. 2a), by the dependence
C
Ep on w (Fig. 2b), and by the magnitude of the differ-
ence between Ep and E , which is larger than is
(
DE ) =(DE ) 298/T
(17)
(18)
A
C
p
p 298
p T
1/2
required for a reversible process. For example, at a
„ =„298(T/298)
T
−
1
polarization rate of 0.2 V s , it is equal to 0.338 V
The values of the „ function, obtained by means of
T
(
Table 1).
Eqs. (17) and (18) and used in conjunction with expres-
sion (16) made it possible to calculate the standard rate
constants for charge transfer. The influence of tempera-
ture on the standard rate constants of electrode reac-
tion (5) is presented in the Table 4.
The problem of determining kinetic parameters on
the basis of cyclic voltammetry was considered by
Nicholson [21]. The standard rate constant of the elec-
trode process is related to the function C as follows
h/2
ꢀ
D
ox
ꢁ
kS
Table 4
Dred
C=
(16)
Influence of the temperature on the standard rate constants for
the redox reaction Eu(III)+e UEu(II) in an equimolar
1
/2
ꢀ
nF
ꢁ
−
y1/2
1/2
OX
w1/2
D
RT
NaClꢀKCl melt
Here c is a function related to the difference between
the peak potentials E −E (mV), kS is the standard
Temperature, T 973
1023
1073
2.27×10−2
A
p
C
p
(K)
−
1
ks (cm s−1)
0.80×10−2 1.25×10−2
rate constant of the electrode process (cm s ), and
h=0.5 is the transfer coefficient.

S.A. Kuznetso6, M. Gaune-Escard / Electrochimica Acta 46 (2001) 1101–1111
1107
According to the classifications in [23,24], the values
this rate the double layer charging current was virtually
insignificant relative to the Faradaic current), utilizing
the Delahay equation for irreversible electrochemical
processes [19]
−
1
of these constants confirm that at w=0.2 V s , pro-
cess (5) proceeds quasi-reversibly mostly under diffu-
sion control.
Increasing the sweep polarization rate from 0.5 to 2.0
1
/2
1/2
Ip=0.496nFCAD (hn Fw/RT)
(22)
−
1
h
V s
results in irreversible electroreduction because
1/2
the peak current is directly proportional to w , but the
slope is smaller than at polarization rate of w50.1 V
The D coefficients obtained at this polarization rate
can be determined from the empirical relations
−
1
s
(Fig. 6a) while the peak potentials shift to the
cathodic direction with increasing scan rate (Fig. 6b).
In the coordinates E −log w (Fig. 6b) at a temperature
log DEu(III)= −2.33−2256/T90.04
log DEu(II)= −2.33−1976/T90.03
(23)
(24)
p
9
73 K the next slope was found [25]
These results are given in the Tables 2 and 3. These
tables also include activation energies for diffusion
K=2.303RT/2hn F=0.208 V
(19)
h
(
−
DU) of Eu(III) and Eu(II) which were calculated from
DU/2.303R=# log D/#(1/T) (25)
where h is the transfer coefficient and n is the number
of electrons transferred.
h
Thus the product hnh can be estimated by the
equation
As can be seen from Tables 2 and 3, the diffusion
coefficients for Eu(III) and Eu(II) obtained by linear
sweep voltammetry at polarization rates 0.1 and 1.0 V
hn =2.303RT/2KF=0.46
h
(20)
−
1
s
are in a good agreement and close to the values
This value is very close to the magnitude hn =0.48
determined from chronopotentiometric and reversal
chronopotentiometric methods. The diffusion coeffi-
cients decrease when the europium oxidation state in-
creases while the activation energies for diffusion
increases. Thus D decreases and DU increases with
increasing ionic moment of the diffusing species. These
results are in agreement with numerous data on the
h
which was found from expression [26]
E −Ep/2= −1.857RT/hnhF
p
(21)
The diffusion coefficients for the Eu(III) and Eu(II)
complexes were determined in the temperature interval
−
1
9
73–1123 K, at a polarization rate w=1.0 V s
(at
1
/2
−
=4.51×10−5 mol cm−3; 2-CEuCl3 =6.32×10−5
Fig. 4. The dependence i~ and i for the process Eu(III)+e “Eu(II). 1–C
EuCl3
mol cm ; 3-CEuCl3 =8.63×10−5 mol cm ; 4-CEuCl3 =1.21×10
−
3
−3
−4
mol cm−3.

1108
S.A. Kuznetso6, M. Gaune-Escard / Electrochimica Acta 46 (2001) 1101–1111
−
Fig. 5. The relation between i~1/2 and concentration of EuCl3 for the process Eu(III)+e “Eu(II).
influence of oxidation state of central atoms on diffu-
sion coefficients, and activation energies for diffusion of
complexes [27]. The decrease in D is related to the
increased strength of the complexes, and this in turn
leads to a decrease in the contribution of the diffusion
coefficient from the ‘hopping’ mechanism as discussed
in literature [28].
reaction (5). The electron transfer itself, which in the
−
15
condensed phase may occur on a time scale of 10
s,
is preceded by the slower step of complex rearrange-
ment [32]. The energy and the time required for the
complex rearrangement decreases as the scale of chemi-
cal bond rearrangement decreases. Related with this is
the fact that redox electrochemical reaction usually is a
reversible process (even at high polarization rates up
It is necessary to note the unusual large differences
between the diffusion coefficients Eu(III) and Eu(II).
The ratio DEu(II)/DEu(III) is about 1.8–1.9 in the temper-
ature range 973–1123 K. In a previous study [29], it
was determined that the relative partial molar enthalpy
of mixing of EuCl2 with a equimolar mixture of
NaClꢀKCl when dilute solutions are formed is, within
experimental error, close to zero. Thus it can be con-
cluded that europium dichloride does not form chloride
complexes in the melt. This conclusion is in agreement
with calorimetric investigations [30] and Raman spec-
troscopy studies [31]. At the same time the relative
partial molar enthalpy of mixing of Eu(III) with the
−
1
5–10 V s ) [33], because these reactions can in princi-
ple occur only with a change of charge and without a
change in the composition of the complexes. The slug-
gish kinetics of the reaction
EuCl 36 +e “Eu +6Cl
−
−
2+
−
(27)
is connected with substantial rearrangement of the eu-
ropium coordination sphere (probably due to the loss
of six ligands) which occurs during the electrode
reaction.
3.4. Formal standard redox potentials E*Eu(III)/Eu(II)
−
1
NaClꢀKCl melt at 1100 K is −16 kJ mol
complex formation reaction
due to the
According to the theory of linear sweep and steady
state voltammetry the following relations are valid for
the reversible electrochemical reduction (5) between the
cathodic and anodic peak potentials and half-wave
potential [26]
3
+
−
3−
Eu +6Cl UEuCl6
(26)
Another peculiarity of the redox electrochemistry of
the couple Eu(III)/Eu(II) is the sluggish kinetics of

S.A. Kuznetso6, M. Gaune-Escard / Electrochimica Acta 46 (2001) 1101–1111
1109
−
Fig. 6. The dependencies of peak currents (a) and of peak potentials (b) of the process Eu(III)+e “Eu(II) on the polarization rate
−
1
2
−5
−3
−
w]0.5 V s . Area: 0.19 cm . Temperature: 973 K. Concentration of EuCl : 9.0×10
mol cm . Reference electrode: Cl /Cl .
3
2
C
E =E1/2−1.11(RT/F)
(28)
(29)
(30)
In the concentration range of ions with mole fraction
P
−
2
A
p
less than (3–5)×10
the coefficients of activity in the
E =E1/2+1.11(RT/F)
molten salts remain constant leads up to the values of
the formal standard potentials [34]
C
A
(
E +E )/2=E
p
p
1/2
where
0
Eu(III)/Eu(II)
EE*u(III)/Eu(II)=E
+RT/F ln(g /gred)
ox
(32)
0
Eu(III)/Eu(II)
1/2
E1/2=E
+RT/F ln(Dred/Dox)
Thus the formal standard redox potentials of
E* can be calculated from the following
+
RT/F ln(g /gred)
ox
(31)
Eu(III)/Eu(II)

1110
S.A. Kuznetso6, M. Gaune-Escard / Electrochimica Acta 46 (2001) 1101–1111
equations
sluggish kinetics of this reaction is connected with
substantial rearrangement of the europium coordina-
tion sphere- the loss of six ligands occurs during the
electrode reaction. The formal standard redox poten-
tials E*Eu(III)/Eu(II) were determined from electrochemical
transient techniques.
1
/2
E*Eu(III)/Eu(II)=E1/2+RT/F ln(D /Dred)
ox
(33)
1/2
E*Eu(III)/Eu(II)=Et/4+RT/F ln(D /Dred)
ox
(34)
C
p
1/2
E*
Eu(III)/Eu(II)
=E +1.11(RT/F)+RT/F ln(D /Dred)
ox
(
35)
A
p
1/2
E*
Eu(III)/Eu(II)
=E −1.11(RT/F)+RT/F ln(D /Dred)
ox
(
36)
Acknowledgements
C
A
1/2
E*
Eu(III)/Eu(II)
=(E +E )/2+RT/F ln(D /Dred) )
p p ox
(
37)
SAK is grateful for support of this study under a
NATO grant. The authors express their thanks to
Professor W. Freyland for helpful discussions. They are
also indebted to Dr L. Rycerz for help in the prepara-
tion of EuCl and EuCl .
E*
=E +RT/hn F[0.78−ln k
Eu(III)/Eu(II)
p
h
s
1/2
+
ln (hn FwD /RT)
]
(38)
we used Eqs.
a
ox
3
2
For the calculation of E*
Eu(III)/Eu(II)
C
A
(
35)–(37) because the values Ep and Ep are more
reproducible compared to the values for E1/2 and E~/4
Eq. (38) is valid for irreversible processes, and applica-
tion of this expression includes error in determination
.
References
[
1] S. Singhand, J. Balachandra, J. Electrochem. Soc. India
22 (1973) 226.
of the standard rate constant (k ).
S
Thus using the values for the potential peaks of
process (5) and the coefficients of diffusion of Eu(III)
and Eu(II), it was found that the formal standard redox
potentials are described by the following empirical
relation:
[2] V.A. Lebedev, Selectivity of Liquid Metal Electrodes in
Molten Halides, Metallurgiya, Chelyabinsk, 1993.
[3] L. Feng, C. Guo, D. Tang, J. Alloys and Comp. 234
(
1996) 183.
[
4] Y. Yamamura, I. Wu, H. Zhu, M. Endo, N. Akao, M.
Mohamedi, Y. Sato, Molten Salt Chem. Technol. 5 (1998)
355.
E*Eu(III)/Eu(II)/V= −(0.97190.006)+(1.990.2)
[
[
[
[
[
5] G. Xie, K. Ema, Y. Ito, Z.M. Shou, J. Appl. Elec-
trochem. 23 (1993) 753.
6] T. Koyama, R. Fujita, M. Iizuka, Y. Sumida, Nucl.
Technol. 110 (1995) 357.
7] M. Iizuka, T. Koyama, N. Kondo, R. Fujita, H. Tanaka,
J. Nucl. Mater. 247 (1997) 183.
8] T. Koyama, M. Iizuka, N. Kondo, R. Fujita, H. Tanaka,
J. Nucl. Mater. 247 (1997) 227.
−
4
×
10 T/K
(39)
The above results for reaction (5) were obtained
using glassy carbon electrodes. It is also possible to use
a platinum electrode to investigate the detailed mecha-
nisms associated with the overall reaction (5). However,
the use of molybdenum and tungsten electrodes which
are often used in electrochemical investigations is not
possible since they have more negative potentials of
dissolution than the redox potentials of for reaction (5).
9] B. Gilbert, V. Demarteau, G. Duyckaerts, J. Electroanal.
Chem. 89 (1978) 123.
[10] J.P. Schoebrechts, B. Gilbert, G. Duyckaerts, J. Elec-
troanal. Chem. 145 (1983) 139.
[
[
[
[
11] W.-J. Gan, I.-W. Sun, J. Electrochem. Soc. 143 (1996)
14.
12] J.E. Macintyre (Ed.), Dictionary of Inorganic Com-
pounds, Chapman and Hall London, 1992.
9
4
. Conclusions
−
The redox process for Eu(III)+e UEu(II) in an
equimolar NaClꢀKCl melt in the temperature range
73–1123 K was studied by different electrochemical
13] K.E. Johnson, J.R. Mackenzie, J. Electrochem. Soc. 116
(
1969) 1697.
9
14] B.D. Vasin, A.V. Vasil’ev, V.A. Ivanov, S.P. Raspopin,
Rasplavy (Melts) 2 (1988) 84.
techniques. The diffusion coefficients for Eu(III) and
Eu(II) obtained by linear sweep voltammetry and
chronopotentiometry methods are in a good agreement.
The diffusion coefficients decrease and activation ener-
gies for diffusion increase when the europium oxidation
state increases, or D decrease with increasing ionic
moment of the diffusing species.
The decrease of D is related to the increased strength
of the complexes, and this in turn leads to a decrease in
the contribution to the diffusion coefficients from the
[15] D.M. Laptev, T.V. Kiseleva, N.M. Kulagin, V.F.
Goryushkin, E.S. Voronkov, Russ. J. Inorg. Chem. 31
(1986) 1965.
[16] R.S. Nicholson, I. Shain, Anal. Chem. 36 (1964) 706.
[
[
[
17] Z. Galus, Fundamentals of Electrochemical Analysis, El-
lis Horwood, London, 1994.
18] A.J. Bard, L.R. Faulkner, Electrochemical Methods,
Fundamentals and Applications, Wiley, New York, 1980.
19] P. Delahay, New Instrumental Methods in Electrochem-
istry: Theory, Instrumentation and Application to Analyt-
ical and Physical Chemistry, Interscience, New York,
1954.
‘hopping’ mechanism. The standard rate constants for
−
the reaction Eu(III)+e UEu(II) were determined. The

S.A. Kuznetso6, M. Gaune-Escard / Electrochimica Acta 46 (2001) 1101–1111
1111
[
[
[
20] M. Iizuka, J. Electrochem. Soc. 145 (1998) 84.
21] R.S. Nicholson, Anal. Chem. 37 (1965) 1351.
22] S.A. Kuznetsov, S.V. Kuznetsova, P.T. Stangrit, Soviet
Electrochem. 26 (1990) 55.
23] J.A. Plambeck, Electroanalytical Chemistry: Basic Princi-
ples and Applications, Wiley, New York, 1982.
24] O.M. Ferry, G.S. Picard, B.L. Tremillon, J. Electrochem.
Soc. 135 (1988) 1443.
[28] M.V. Smirnov, O.M. Shabanov, Trans. Inst. Elec-
trochem. Ural. Branch. Ac. Sci. USSR 8 (1966) 55.
[29] S.A. Kuznetsov, M. Gaune-Escard, International G.N.
Papatheodorou Symposium ICEHT/FORTH, Patras,
Greece, 1999, p. 100.
[30] F. Da Silva, Thesis, Unversity of Provence, Marseille,
1999.
[31] Y. Koyama, R. Takagi, Y. Iwadate, K. Fukushima, J.
Alloys Comp. 260 (1997) 75.
[32] L.I. Krishtalik, Electrode Reactions. The Mechanism of
the Elementary Act, Nauka, Moscow, 1982.
[33] S.A. Kuznetsov, Russ. J. Electrochemistry 29 (1993) 1154.
[34] M.V. Smirnov, Electrode Potentials in Molten Chlorides,
Nauka, Moscow, 1973.
[
[
[
25] Southampton Electrochemistry Group, Instrumental
Methods in Electrochemistry, Ellis Horwood, London,
1
990.
[
[
26] H. Matsuda, Y. Ayabe, Z. Elektrochemie 59 (1955) 494.
27] G.J. Janz, N.P. Bansal, J. Phys. Chem. Data 11 (1982)
5
05.
.