Temperature dependence of the cathodic and anodic kinetics in a copper

electrowinning cell based on reactive electrodialysis

L. Cifuentesa, , J. Simpsonb
a Departamento de Ingeniera de Minas, Universidad de Chile, Tupper 2069, Santiago, Chile
b Departamento de Ingeniera Metalrgica, Universidad de Santiago, Av. Ecuador 3659, Santiago, Chile

Abstract
The temperature dependence of various kinetic parameters for the copper electrodeposition reaction and for the ferrous to ferric ion
oxidation has been experimentally determined. Potentiodynamic sweeps were carried out in a reactive electrodialysis cell in the 3060 C
range in order to obtain expressions for each parameter as a function of temperature. In the conditions studied, for both reactions, the
exchange current density depends on temperature as ln[i0 (A/m2 )] = a/T + b and the limiting current density, as |iL (A/m2 )| = cT d.
For the copper electrodeposition reaction, the charge transfer coefcient is constant, whereas for the ferrous to ferric ion oxidation it
depends on temperature as
= f T 2 + gT + h (ah are constants). These expressions can be used in a temperature-dependent model of a
membrane-based copper electrowinning cell.

Keywords: Temperature; Chemical reactors; Kinetics; Electrochemistry; Membrane; Modeling

1. Introduction
1.1. Objectives
This work attempts to establish the temperature dependence of various kinetic parameters (charge transfer coefcient, exchange current density, limiting current density) for
the main cathodic and anodic reactions in a reactive electrodialysis (RED) cell. This cell is made up of two compartments, the rst one containing the cathode and the catholyte,
and the second one containing the anode and the anolyte. The
compartments are separated by an anion membrane meant
to prevent cation transfer between the electrolytes. The cathodic reaction is copper electrodeposition and the anodic
reaction is ferrous to ferric ion oxidation. At the operating
cell current densities, both reactions take place under mixed
control. The design and operation of this cell have been described in detail elsewhere (Cifuentes et al., 2004a,b).
Corresponding author. Tel.: +562 678 4510; fax: +562 672 3504.

E-mail address: luicifue@ing.uchile.cl (L. Cifuentes).

The objectives of the present work are: (a) to study the

effect of temperature on the exchange current density, the
charge transfer coefcient and the limiting current density
of the Cu2+ + 2e Cu0 and Fe2+ Fe3+ + e reactions and (b) to produce mathematical relationships for the
temperature dependence of the said parameters which could
help the development of a temperature-dependent RED cell
model.
1.2. Calculation of kinetic parameters
The cell current density for a cathodic reaction under
mixed control is given by
|ic | =

i0c |iLc |

(1)

cF
i0c + |iLc | exp(
RT
c )

and, for an anodic reaction under the same type of control,

ia =

i0a iLa

a F
i0a + iLa exp( RT
a )

(2)

L. Cifuentes, J. Simpson

where i0c , i0a are cathodic and anodic exchange current densities, A/m2 , iLc, iLa are cathodic and anodic limiting current densities, A/m2 ,
c ,
a are cathodic and anodic charge
transfer coefcients, c , a are cathodic and anodic overpotentials, V, F is the Faradays constant, 96 500 C/eq, R is gas
constant, 8.314 J/mol K and T is temperature, K.
In order to determine the values of these parameters, potentiodynamic sweeps were carried out. These experiments
provided data for i = f () at various temperatures. The values of i0 , iL and
for the copper electrodeposition reaction were obtained by the least-squares method (Chapra and
Canale, 1998), i.e., by minimizing the expression

2
N


i0c |iLc |
Z=
wk |i|k
.
(3)
cF
i0c + |iLc | exp(
RT
ck )
k=j
The same parameters, for the ferrous to ferric ion oxidation,
were obtained by minimizing the expression

2
N


i0a iLa
Z=
wk |i|k
.
(4)

a F
i0a + iLa exp( RT
ak )
k=j
In Eqs. (3) and (4): wk is the weighting factor for each experimental point in the potentiodynamic sweep, N is number
of data points, j is rst data point in the mixed control range
and k is kth data point in the potentiodynamic sweep.
These equations were solved by an adapted Newton
Raphson method (Chapra and Canale, 1998), producing i0 ,
iL and
values for each studied temperature.
In the present work, 800 data points were captured for
each sweep and the resulting data were processed by Excel
software, rst, in order to obtain parameter values at each
studied temperature, and then to obtain expressions for each
parameter as a function of temperature.
1.3. Previous work
There have been numerous studies of copper electrodeposition (Cu2+ + 2e Cu0 ) kinetics (e.g., Andersen
et al.,1973; Peykova et al., 1995; Zagorovskii et al., 2001;
Vicenzo and Cavallotti, 2002; Cifuentes et al., 2004a). The
values of the kinetic parameters (exchange current density,
charge transfer coefcient and limiting current density)
obtained by these authors vary with cell geometry and temperature, composition and hydrodynamics of the electrolyte;
they also vary with the cathode materials used.
The ferrous to ferric ion oxidation (Fe2+ Fe3+ + e)
has been chosen as an alternative to water decomposition
(2H2 O O2 +4H+ +4e) in the design of new copper electrowinning cells because its use allows a drastic reduction
in the cell voltage, from about 2 V (typical of conventional
Cu electrowinning) to about 1 V (Sandoval, 1995; Cifuentes
et al., 2004a). In addition, this reaction eliminates acid mist
and the produced ferric ion can be used as an oxidizing
agent in combined leaching processes. The ferrous to ferric
ion oxidation kinetics have also been studied by a number

of authors (e.g., Cooke et al., 1989; Lee, 1993; Rao et al.,

1995; Sandoval et al., 1995; Das, 1996; Dew et al., 1985a,b;
Jiricny and Evans, 1999; Marconi et al., 1996; Ronnholm et
al. 1999a,b; Cifuentes et al., 2004a).
Various aspects of the design and operation of RED cells
have been discussed by Cifuentes et al. (2004a,b). These
cells have been shown to produce electrolytic copper at high
values of the cell current density, cathodic current efciency
and specic energy consumption. A time-dependent model
for the operation of a RED cell has been produced and validated (Cifuentes et al., 2004c). The test variables were copper production rate, ferric ion production rate, cell voltage as
a function of time, catholyte and anolyte conductivities and
specic energy consumption. The model was validated only
for operation at 50 C, therefore, further work is required to
allow for model predictions at a range of temperatures. This
work intends to advance towards this aim.

2. Experimental
Potentiodynamic sweeps were carried out at 30, 40, 50
and 60 C (303, 313, 323 and 333 K) for both copper electrodeposition and ferrous to ferric ion oxidation. In order
to approximate the conditions found in a RED copper electrowinning cell (cell geometry, electrolyte composition and
hydrodynamics), the sweeps were conducted, for the cathodic reaction, in the cathode compartment of this cell and
for the anodic reaction, in the anode compartment. In both
cases, there was simultaneous catholyte and anolyte recirculation to ad hoc tanks. The experimental arrangements are
shown in Figs. 1 and 2.
During normal operation of this cell, a current is generated
by a rectier and it is passed between anode and cathode. In
the present case, anode, cathode and rectier were absent,
but the working, reference and counter electrodes depicted
in Figs. 1 and 2 were connected to a Voltamaster 10 electrochemical interface in order to carry out the potentiodynamic
sweeps.
In the copper electrodeposition case, the working electrode was a 1 cm 1 cm 0.05 cm electrolytic copper sheet
and the counter electrode was a 2 cm 2 cm 0.01 cm platinum sheet. In the ferrous/ferric oxidation case, the working
electrode was a 1 cm 1 cm 0.05 cm platinum sheet and
the counter electrode was a 4 cm 4 cm 0.19 cm titanium
mesh. In both cases, a saturated Hg2 SO4 /Hg reference electrode with Luggin capillary was used and the sweep rate was
0.5 mV/s.
The two-compartment RED cell was made of 15 mm thick
acrylic plates held together by seven 5-mm diameter stainless steel bars bolted at both ends. Compartment dimensions
were 7 8.5 1.4 cm3 .
In both cases, the anolyte composition was 60 g/L
Fe (from ferrous sulfate) and 190 g/L sulfuric acid. The
catholyte composition was 30 g/L Cu (from cupric sulfate) and 190 g/L sulfuric acid. The source chemicals were

L. Cifuentes, J. Simpson

Fig. 1. Experimental arrangement for the study of the cathodic Cu2+ /Cu0 reaction. WE = working electrode; RE = reference electrode; CE = counter
electrode.

Fig. 2. Experimental arrangement for the study of the anodic Fe2+ /Fe3+ reaction. WE = working electrode; RE = reference electrode; CE = counter
electrode.

analytical grade CuSO4 5H2 O, FeSO4 7H2 O and 96%

(w/w) sulfuric acid.
An Ionac MA-3475 anion membrane was used to separate anolyte from catholyte; it was tted in a 4 4 cm2
window cut in the acrylic plates between the compartments.
This membrane was selected on the basis of availability
and performance in previous works. It was kept in place
by 2 mm thick rubber seals. Before each experiment, the
membrane was pre-treated. Membrane properties and pretreatment methods have been given elsewhere (Cifuentes et
al., 2004b).
The electrolytes were separately recirculated at a ow
rate of 700 cm3 / min, driven by two Cole-Parker WZIR051
peristaltic pumps. Two double-jacket, 1-L recirculation glass

tanks were used. The inner diameter of the recirculation inlet

and outlet tubes tted to the cell was 1 cm. Total volume for
anolyte and catholyte was 600 cm3 each. The temperature of
both electrolytes was kept constant during each experiment
by a Julabo thermostatic bath.

3. Results and discussion

3.1. Cathodic kinetics
Cathodic kinetics in the mixed control range is represented in Fig. 3 as potentiodynamic curves at 30, 40, 50 and
60 C. It is clear that, the higher the temperature, the higher

L. Cifuentes, J. Simpson

Fig. 3. Experimental and simulated potentiodynamic curves (ln |i| vs. E) for the cathodic Cu2+ /Cu0 reaction at various temperatures.

Table 1
Kinetic parameters for the Cu2+ /Cu0 reaction on a copper sheet cathode
at various temperatures: a rst approximation
Temperature
( C) (K)

Exchange c. d. (i0 ) Charge transfer Limiting c. d. (iL )

(A/m2 )
coefcient (
c ) (A/m2 )

30
40
50
60

4.5
19.6
158.7
1268.2

303
313
323
333

0.7440
0.7430
0.7429
0.7439

614
841
1170
1351

The remaining parameters (i0c and iLc ) were recalculated

by the procedure already described, but this time
c was considered constant. These corrected (second approximation)
parameters are given in Table 2.
Using the parameter values in Table 2 and an expression
for the reaction rate under mixed control (Eq. (1)), simulated potentiodynamic sweeps were produced and plotted in
Fig. 3. The simulated curves are in good agreement with
experimental results.
3.3. Cathodic exchange current density

the cathodic current density throughout the studied potential

range.
The experimental curves in Fig. 3 were used to obtain a
rst approximation to the values for the kinetic parameters
(i0 ,
and iL ) from the procedure described in Section 1.2
above. These values are in Table 1. The simulated results in
Fig. 3 are discussed below.
3.2. Cathodic charge transfer coefcient
Cathodic charge transfer coefcient (
c ) values from
Table 1 are plotted against temperature in Fig. 4.
c is practically constant in the studied temperature range. As this
coefcient depends on the reaction mechanism, it becomes
apparent that the mechanism for copper electrodeposition
does not change with temperature in the studied range.
From these considerations, the cathodic charge transfer
coefcient was assumed constant and it was calculated as
the mean value from the results in Table 1, so, for the studied
temperature range,

c = 0.7434 0.0006.

(5)

Various relationships between the exchange current density (i0 in ampere per square meter) for the cathodic reaction and the absolute temperature (in Kelvin) were tried. A
linear t was found between the logarithm of the exchange
current density (values in Table 2) and the reciprocal value
of the absolute temperature. The plot is shown in Fig. 5.
The numerical relationship was obtained with Excel software as
ln[i0c (A/m2 )] = 20 438

1
+ 68.33.
T (K)

(6)

The correlation coefcient was 0.999.

This relationship is typical of mixed control given that,
from Eq. (1):


 ic iLc 

c F |c | 1

+ ln 
(7)
ln i0c =
R
T
(ic iLc ) 
and the latter expression exhibits the same form as Eq. (6)
provided that the second term on the right-hand side of
Eq. (7) remains fairly constant with temperature. Using experimental values for Eq. (7), as temperature increases from

L. Cifuentes, J. Simpson

Fig. 4. Cathodic charge transfer coefcient (
c ) as a function of temperature for the Cu2+ /Cu0 reaction.

Table 2
Kinetic parameters for the Cu2+ /Cu0 reaction on a copper sheet cathode
at various temperatures: a second approximation

In Ficks law, D,  and km depend on temperature. The

form of Eq. (8) suggests that

Temperature
( C) (K)

Exchange c. d. (i0 ) Charge transfer Limiting c. d. (iL )

(A/m2 )
coefcient (
c ) (A/m2 )

km (T ) =

30
40
50
60

3.1
19.6
158.6
1268.8

303
313
323
333

0.7434
0.7434
0.7434
0.7434

D(T )
= aT b,
(T )

(10)

where a and b are constants, because in this case, Ficks law

becomes

616
841
1169
1269

|iLc | = zF cb (aT b) = zF cb aT zF cb b

(11)

which exhibits the same form as Eq. (8).

30 to 60 C, the rst term on the right-hand side of the equation increases by about 60%, whereas the second term increases only by about 10%.
3.4. Cathodic limiting current density
The values of the limiting current density (iLc in ampere
per square meter) for the cathodic reaction in Table 2 were
plotted against the absolute temperature (Kelvin). The obtained relationship is linear (see Fig. 6) and it is given by
|iLc (A/m2 )| = 32.61 T (K) 9309.6.

(8)

The correlation coefcient was 0.986.

On the other hand, iL can be calculated by Ficks equation
as
iL = zF D

cb
= zF k m cb ,


(9)

where cb is the Ferrous ion concentration in the bulk solution, mol/m3 , D is diffusivity, m2 /s, km is mass transfer
coefcient, m/s, and  is thickness of the diffusion layer, m

3.5. Anodic kinetics

Anodic kinetics in the mixed control range are represented
in Fig. 7 as potentiodynamic curves at 30, 40, 50 and 60 C.
The plots show that, the higher the temperature, the higher
the anodic current density throughout the studied potential
range.
The experimental curves in Fig. 7 were used to obtain
the values for the kinetic parameters (i0 ,
and iL ) from
the procedure described in Section 1.2 above. These values
are in Table 3. The simulated results were calculated with
Eq. (2) and the parameter values in Table 3. There is good
agreement between experiments and simulations.
3.6. Anodic charge transfer coefcient
Anodic charge transfer coefcient (
a ) values from Table
3 are plotted against temperature in Fig. 8.
a varies with
temperature. Considering that this coefcient depends on
the reaction mechanism, it follows that the mechanism for
ferrous to ferric ion oxidation changes with temperature in
the studied range.

L. Cifuentes, J. Simpson

Fig. 5. Logarithm of the exchange current density (i0c ) as a function of 1/T for the Cu2+ /Cu0 reaction.

Fig. 6. Absolute value of the limiting current density (|iLc |) as a function of temperature for the Cu2+ /Cu0 reaction.

The numerical relationship is

a = 2 104 T (K)2 0.1113 T (K) + 17.385.

(12)

The correlation coefcient was 0.991. Further work is required in order to establish the reaction mechanisms involved.

and the absolute temperature (in Kelvin) were tried. As in

the cathodic case, a linear t was found between the logarithm of the exchange current density (values in Table 3) and
the reciprocal value of the absolute temperature. The plot is
shown in Fig. 9.
The numerical expression is
ln[i0a (A/m2 )] = 6403

3.7. Anodic exchange current density

Various relationships between the exchange current density (i0a in ampere per square meter) for the anodic reaction

1
+ 23.828
T (K)

(13)

with a correlation coefcient of 0.996. It is of similar form

to Eq. (6) and the considerations made in Section 3.3 also
apply in this case.

L. Cifuentes, J. Simpson

Fig. 7. Experimental and simulated potentiodynamic curves (ln |i| vs. E) for the anodic Fe2+ /Fe3+ reaction at various temperatures.
Table 3
Kinetic parameters for the Fe2+ /Fe3+ reaction on a platinum sheet anode at various temperatures
Temperature
( C)

(K)

30
40
50
60

303
313
323
333

Exchange c. d. (i0 ) (A/m2 )

Charge transfer coefcient (
c )

Limiting c. d. (iL ) (A/m2 )

15.4
28.0
49.3
98.3

0.3393
0.3534
0.3782
0.4614

644
901
1091
1317

Fig. 8. Anodic charge transfer coefcient (
a ) as a function of temperature for the Fe2+ /Fe3+ reaction.

3.8. Anodic limiting current density

The values of the limiting current density (iLa in
ampere per square meter) for the anodic reaction in
Table 3 were plotted against the absolute temperature
(Kelvin). The obtained relationship is linear (see Fig. 10)

and it is given by
iLa (A/m2 ) = 23.015 T (K) 6318.9

(14)

with a correlation coefcient of 0.998. This is similar to

Eq. (8) and considerations put forward in Section 3.4 also
apply here.

L. Cifuentes, J. Simpson

Fig. 9. Logarithm of the exchange current density (i0a ) as a function of 1/T for the Fe2+ /Fe3+ reaction.

Fig. 10. Limiting current density (iLa ) as a function of temperature for the Fe2+ /Fe3+ reaction.

The obtained relationships for all six kinetic parameters (
c , i0c , iLc ,
a , i0a and iLa ) as functions of
temperatureEqs. (5), (6), (8), (12), (13) and (14)can be
included in a temperature-dependent model for the operation of a membrane-based copper electrowinning cell.

pressions for the calculation of both the cathodic and anodic

reaction rates, gives good agreement between experiment
and simulation.
(2) The charge transfer coefcient for the Cu2+ /Cu0 reaction does not signicantly change with temperature and
its value is

c = 0.7434 0.0006.
4. Conclusions
In the 3060 C temperature range:
(1) Using the kinetic parameter values obtained from the
experimental potentiodynamic sweeps and mixed control ex-

(3) The exchange current density for the Cu2+ /Cu0 reaction depends on temperature as
ln[i0c (A/m2 )] = 20 438

1
+ 68.33.
T (K)

L. Cifuentes, J. Simpson

(4) The limiting current density for the Cu2+ /Cu0 reaction
depends on temperature as
|iLc (A/m2 )| = 32.61 T (K) 9309.6.
(5) The charge transfer coefcient for the Fe2+ /Fe3+
reaction depends on temperature as

a = 2 104 T (K)2 0.1113 T (K) + 17.385.

(6) The exchange current density for the Fe2+ /Fe3+
reaction depends on temperature as
ln[i0a (A/m2 )] = 6403

1
+ 23.828.
T (K)

(7) The limiting current density for the Fe2+ /Fe3+

reaction depends on temperature as
iLa (A/m2 ) = 23.015 T (K) 6318.9.
(8) All these results can be used in a temperaturedependent mathematical model for the operation of the
RED cell.
Acknowledgements
This work was funded by the National Committee for Science and Technology (CONICYT, Chile) via FONDECYT
project no. 101 0138. Financial support from Placer Dome
to the Chair of Environmental Studies in Mining (Universidad de Chile) and continued support from the Departments
of Mining Engineering, Universidad de Chile and Metallurgical Engineering, Universidad de Santiago, are gratefully
acknowledged.
References
Andersen, T.N., Wright, C.N., Richards, K.J., 1973. Important
electrochemical aspects of electrowinning copper from acid leach
solutions. In: Evans, D.J.I., Shoemaker, R.S. (Eds.), International
Symposium on Hydrometallurgy. AIME, New York, pp. 171202.
Chapra, S.C., Canale, R.P., 1998. Numerical Methods for Engineers with
Programming and Software Applications. McGraw-Hill, New York.
Cifuentes, L., Glasner, R., Casas, J.M., 2004a. Aspects of the development
of a copper electrowinning cell based on reactive electrodialysis.
Chemical Engineering Science 59, 10871101.

Cifuentes, L., Mondaca, C., Casas, J.M., 2004b. The effectiveness of

membrane systems for the separation of anolyte and catholyte in a
lab-scale copper electrowinning cell based on reactive electrodialysis.
Minerals Engineering 17, 803809.
Cifuentes, L., Castro, J.M., Crisstomo, G., Casas, J.M., Simpson, J.,
2004c. Modeling a copper electrowinning cell based on reactive
electrodialysis. Submitted for publication.
Cooke, A.V., Chilton, J.P., Fray, D.J., 1989. Mass-transfer kinetics of
the ferrousferric electrode process in copper sulphate electrowinning
electrolytes. Transactions of the Institution of Mining and Metallurgy.
Section C: Mineral Processing and Extractive Metallurgy 98, 123131.
Das, S.C., 1996. Effect of Fe(II) during copper electrowinning at higher
current density. International Journal of Mineral Processing 91108.
Dew, D.W., Phillips, C.V., 1985a. The effect of Fe(II) and Fe(III)
on the efciency of copper electrowinning from dilute acid Cu(II)
sulphate solutions with the Chemelec cell. Part I. Cathodic and anodic
polarisation studies. Hydrometallurgy 14, 331349.
Dew, D.W., Phillips, C.V., 1985b. The effect of Fe(II) and Fe(III) on the
efciency of copper electrowinning from dilute acid Cu(II) sulphate
solutions with the Chemelec cell. Part II. The efciency of copper
electrowinning from dilute liquors. Hydrometallurgy 14, 351367.
Jiricny, V., Evans, J.W., 1999. A study of the spouted-bed electrowinning
of copper. Proceedings of the COPPER99 International Conference,
vol. 3, Electrorening and Electrowinning, TMS, Phoenix, pp. 629.
Lee, J.C., 1993. Electrogenerative oxidation of ferrous ions with graphite
electrodes. Journal of Applied Electrochemistry 23 (7), 745.
Marconi, P.F., Meunier, V., Vatistas, N., 1996. Recovery of pickling
efuents by electrochemical oxidation of ferrous to ferric chloride.
Journal of Applied Electrochemistry 26 (7), 693.
Peykova, M., Michailova, E., Stoychev, D., 1995. Galvanostatic studies
of the nucleation and growth kinetics of copper in the presence of
surfactant. Electrochimica Acta 40 (16), 25952601.
Rao, S.R., Finch, J.A., Kuyucak, N., 1995. Ferrousferric oxidation in
acid mineral efuents: comparison methods. Minerals Engineering 8,
905911.
Ronnholm, M.R., Warna, J., Salmi, T., 1999a. Kinetics of the oxidation
of ferrous sulphate with molecular oxygen. Journal of Information in
Chemical Engineering and Science 54 (19), 4223.
Ronnholm, M.R., Warna, J., Salmi, T., 1999b. Oxidation kinetics of ferrous
sulphate over active carbon. Industrial and Engineering Chemical
Research 38 (7), 2607.
Sandoval, S.P., 1995. A substitute anode for electrowinning copper. In:
Cooper, W.C. (Ed.), Proceedings of the COPPER95 International
Conference, vol. III, Electrorening and Hydrometallurgy of Copper,
The Metallurgical Society of CIM, Toronto, pp. 423435.
Vicenzo, A., Cavallotti, P.L., 2002. Copper electrodeposition from a pH
3 sulfate electrolyte. Journal of Applied Electrochemistry 32 (7),
743753.
Zagorovskii, G.M., Prikhodko, G.P., Ogenko, V.M., 2001. Deposition of
copper from dilute solutions onto disperse graphite cathode. Russian
Journal of Applied Chemistry 74 (3), 424426.