Int. J. Electrochem. Sci.

, 8 (2013) 12333 - 12347

International Journal of
ELECTROCHEMICAL
SCIENCE
www.electrochemsci.org


Electrochemical Hydrodynamics Modeling Approach for a
Copper Electrowinning Cell

K.R. Kim
1,*
, S.Y. Choi
1
, S. Paek
1
, J.Y. Park
2
, I.S. Hwang
2
, and Y.Jung
3

1
Korea Atomic Energy Research Institute, 1045 Daedeok-daero, Yuseong-gu, Daejeon, 305-353
Korea
2
Department of Nuclear Engineering, Seoul National University, Sillim-dong, Gwanak-gu, Seoul,
151-742, Korea
3
Department of Chemical Engineering, Korea University of Technology and Education, Cheonan,
330-780, Korea
*
E-mail: [email protected]

Received: 24 June 2013 / Accepted: 18 August 2013 / Published: 25 September 2013


This study demonstrates a simulation based on a full coupling of electrochemical kinetics with the 3-
dimensional transport of ionic species in a flowing electrolyte through a simplified channel cell of
copper electrowinning. The dependences of ionic electro-transport on the velocity of a stationary
electrolyte flow were studied using a coupling approach of the electrochemical reaction model. The
present model was implemented in a commercially available computational fluid dynamics (CFD)
platform, Ansys-CFX, using its customization ability through user-defined functions. The main
parameters characterizing the effect of the turbulent flow of an electrolyte between two planar
electrodes were demonstrated by means of a CFD-based multiphysics simulation approach. Simulation
was carried out for the case of the mass transport controlled copper electrowinning characteristics in a
stream of acid sulfate electrolyte. This approach was taken into account in the concentration profile at
the electrode surface, to represent the variation of the convective diffusion limited current density as a
function of the flow characteristics and of the applied current density. It was able to predict a
conventional current-voltage relation in addition to details of the electrolyte fluid dynamics and
electrochemical variable, such as the flow field, species concentrations, potential, and current
distributions throughout the galvanostatic cell.


Keywords: Computational fluid dynamics (CFD), Electrochemical hydrodynamics, Electrochemical
system simulation, Electro-transport, Copper electrowinning.

1. INTRODUCTION
The importance of the computer simulations in the design and performance assessment of
electrochemical systems has increased dramatically during the last three decades. The electrolysis
Int. J. Electrochem. Sci., Vol. 8, 2013

12334
model provides engineers the ability to simulate the influence of the cell geometry and operating
conditions in the electrochemical metallurgy industry [1, 2, 3, 4, 5].
The electrochemical reaction based on the hydrodynamics of the flowing electrolyte in an
electrometallurgical cell is a complex, highly non-linear coupled phenomenon. The effect of the flow
field to the electrode reaction in an electrolyte domain is the result of a combination of various
hydrodynamics and mass transport [6, 7, 8]. The electrometallurgical efficiency and quality are
determined by the cell configuration, operating conditions (mixing, applied current and cell voltage),
and chemical state of the electrolytic cell. The ion current in the bulk electrolyte requires knowledge
of the electrode surface, which requires an electrochemical hydrodynamics analysis of a mixing and
mass transport.
The most effective way to accelerate the development of these technologies is to formulate
physical models of the underlying electrochemical and transport processes. In addition, an integrated
multiphysics simulation offers an efficient approach to designing, testing, and implementing these
processes [9, 10].
A copper electroplating system has recently become the reference and standard technique for
metalizing interconnects in high-end electrodeposition studies [11]. This established system enables a
wide range of electrochemical features to be achieved in a single experiment and is useful to
benchmark a proposed model. In addition, a reliable database of copper electrode reaction kinetics has
been made available and provides an important parametric analysis for the simulation capability [12,
13].
In this study, an electrochemical hydrodynamics modeling approach based on a mass-transport
controlled electrode reaction using a commercial computational fluid dynamics (CFD) platform was
proposed. We demonstrated for the application of theoretical modeling to the copper electrowinning
processes in an acid electrolyte flow channel involving both convective and diffusive transport. The
applications discussed herein involve the use of computational multiphysics modeling to aid in the
development of these systems, and to provide a theoretical framework for predicting electrochemical
behavior in a future innovative process.



2. MODEL AND ELECTROWINNING CELL
2.1. Modeling approach
A commercial ANSYS

-CFX code [14] was employed to solve the electrochemical

hydrodynamics model in a given electrochemical cell geometry. The finite-volume incompressible
NavierStokes and transport equations were solved to obtain the ionic species concentrations and
electric field at each position in the cell. A scalar transport equation was used to describe the electro-
transport of the reactive ion from the bulk to the electrode surface. To numerically solve these
equations, we have successfully applied CFD techniques to the elctrodeposition model with the general
form of convection-diffusion equations.
The transport equation for the conserved quantity per unit mass (| = ionic
Int. J. Electrochem. Sci., Vol. 8, 2013

12335
concentration(C)/density()) with an external volumetric source term (S
|
, with units of conserved
quantity per unit volume per unit time) is given by:

| |
| |
|
S D u
t
+ V V = V +
c
c
) ( ) ( (1)

where, u is the velocity and D
|
is the kinematic diffusivity. The concentration (C) was used as
an additional scalar variable of volumetric type.
The choice of spatial coordinate and the boundary conditions depend on the cell geometry and
electrode configurations. The transport of the ionic reactant can be tracked by solving an additional set
of transport equations, which are fully coupled with the CFD equations during the simulations.
This scalar transport equation was solved for the concentration of the ionic species, with the
source/sink (S
|
) at the appropriate anode and cathode boundary based on Faradays law of electrolysis.
The electric boundary conditions include the specification of a current flux equivalent to an applied
current (i
app
) through the electrodes with a fixed area. The boundary conditions for the ionic species for
the source/sink at the wall of electrodes were calculated as follows:

zF
i
S
app
=
|
(2)

where z is the ionic valence and F is the faraday constant. On the anode side, a positive flux is
applied, while a negative flux of a same size is applied on the cathode.
The only cathodic electrode reaction to take into account is the deposition of copper (Cu
2+
+ 2e
-
Cu). Since only one single electrode reaction is considered, the deposition efficiency in this model is
considered to be 100 % based on the definition of Faradays law. Assuming that only the reduction of
Cu
2+
takes place at the cathode, Faradays law gives the total amount of current (I) spent to reduce the
equivalent of Cu
2+
over the effective electrode surface area (A):

A i idA I
ave
= =
}
(3)

Here, i and i
ave
are the local and average current flux for Cu deposition.
No overpotential relation is considered (primary distribution, infinitely fast reaction) at the
anode, because the electrochemical processes occurring there are of no interest for this study.
Simulations using a fixed anode potential showed that (for the same current) the cathodic current
distribution was negligibly affected. This approach allowed the average current density to be fixed in
the galvanostatic electrolysis.
The electrode reaction rate is coupled as a gradient of the concentration while the potential
gradients, including migration, are neglected. The modeling approach in this study is used to
demonstrate the electrode reaction phenomena coupled with the mass transfer in the deposited wall
region between the solid boundary and turbulent electrolyte flow.

Int. J. Electrochem. Sci., Vol. 8, 2013

12336
2.2. Electrode reaction and electric field model
The polarization equation is necessary to express the dependence of the local rate of the
electrochemical reaction on the various concentrations and on the potential drop at the interface
between the electrode and electrolyte. It is common to use the Tafel approximation [15, 16] of
electrode kinetics of the form for metal/ion systems. The local current density (i) distribution on the
electrode surface is modelled by the following equation:

|
|
.
|

\
|

|
|
.
|

\
|
=
C
bulk
s
C
C
i i
|
q
exp
0
(4)

where, q is the overpotential at the electrode,
C
is the Tafel slope, i
0
is the exchange current
density, and superscripts s and bulk are the locations at the electrode surface and bulk electrolyte,
respectively.
For electrolyte salt in the interior of an electrolytic cell, there are no free electrical charges. The
potential drop across the bulk electrolyte is governed by the Laplace equation [15, 16]:

0
2
= V | (5)

where, | represents the local electrical potential. In addition, the current density distribution is
obtained by solving the following equation (k= electrical conductivity):

| kV = i (6)

Assuming that we impose a specific voltage drop (
Cell
E ) across the electrodes, the overall
voltage balance may be written as:

c a ohm Cell
E q q | + + = (7)

where
ohm
| is the ohmic voltage drop, q
a
and q
c
are the voltage drops owing to activation
polarization (i.e., kinetic effects) and concentration polarization (owing to concentration gradients
between the electrode surface and bulk electrolyte), respectively.

2.3. Electrowinning cell geometry and boundary conditions for simulation
Copper electrowinning is the process of winning copper from an electrolyte to solid form on a
cathode by passing an electric current through the electrolyte to attract copper ions to the cathode. In
this demonstration approach, a simple channel cell was selected. Its geometry is rectangular and the
distance between the electrodes is a bit large compared to their width. Copper electrowinning takes
place in a rectangular channel geometry, with two protruded planar electrodes opposing each other,
and the current moves between the electrodes and depletes the copper ions at the cathode.
Int. J. Electrochem. Sci., Vol. 8, 2013

12337
It may then be assumed that the flow is in a steady state and is turbulent, and incompressible
and is between two planar and parallel electrodes. The geometry and generated mesh for this cell is
shown in Figure 1.

Outlet
Cathode
Anode
Inlet
Cathode
plane
20 m


Figure 1. Schematic view of volume-discretized electrowinning cell geometry.

The electrolyte volume was meshed with the tetrahedral elements within the Ansys Workbench
platform. In the simulation, the fluid domains of the electrolyte were divided into a diffusion-dominant
region near the cathode and a stationary region for the outside, which were connected together with a
domain interface. In particular, the fine meshes (~1 m) were applied to the cathode surface to obtain
the velocity distribution behaviors and concentration gradient.
Zero gradient boundary conditions for the ion concentration were imposed along the walls
except the inlet, outlet and cathode surface. No-slip velocity boundary conditions were applied on all
walls. Calculations were initialized with the ion concentration equal to the bulk concentration
throughout.


3. RESULTS AND DISCUSSION
3.1 Model parameters and hydrodynamically coupled electro-transport
The relation between electro-transport and mass transport with electrolyte flow during
electrolysis was modeled using computational fluid dynamics. In this modeling approach, the
electrolyte flow may be considered to be steady, and the fluid is supposed to have a constant chemical
composition and physical properties. It was assumed that the aqueous acid electrolyte containing
approximately 50 mol/m
3
of cupric ion (Cu
2+
) was fed into the channel cell at various inlet velocities.
Table 1 is the properties of the acid sulfate electrolyte and electrode reaction kinetics used in this
simulation [12].

Int. J. Electrochem. Sci., Vol. 8, 2013

12338
Table 1. Properties and parameters used in the simulation [12]

Electrolyte properties values
Inlet electrolyte velocity, u (m/s) 0.1-0.5
Initial and inlet concentration of cupric ions, C
bulk
(mol/m
3
) 50
Diffusion coefficient of cupric ions, D
C
(m
2
/s) 4.210
-10

Electrolyte density (kg/m
3
) 998
Electrolytic conductivity (S/m) 35
Kinematic viscosity, v (m
2
/s) 1.0910
-6

Kinetic parameters for copper deposition
Tafel slope, b
C
(V) 0.0525
Exchange current density, i
0
(A/m
2
)

-0.537

Cathode
Anode
Inlet
Outlet


Figure 2. Velocity distribution between the anode and cathode (inlet velocity =0.2 m/s).

The electrolyte solution generally flows between two planar and parallel electrodes.
Considering the order of magnitude of the electrolyte velocities, the flows are generally turbulent. It is
considered that the cathode is a sink of ion transport as the deposit parallel to the anode. Under these
conditions, the modeling for the electrochemical and mass transport phenomena occurring in the cell
can be coupled in the CFD platform.
In the model, forced convection by inlet electrolyte is considered. The velocity contours that
are flowing at an inlet velocity of 0.2 m/s are shown in Figure 2. From this, the localized velocity
values are predicted between the electrodes. The electrolyte velocity profile in a rectangular channel
cell under a pressure-driven hydrodynamic flow shows parabolic patterns with a wide plateau. The
maximum velocity was found at the region between the electrode and center of the channel, because of
Int. J. Electrochem. Sci., Vol. 8, 2013

12339
the electrodes projecting out over the channel. The velocity was almost zero at all of the channel walls.
However, the fully developed parabolic profile patterns were not found at the inlet or outlet region
within the channel.
In this study, we modeled the stationary flow-assisted mass transport that controls the electrode
reaction rates in an electrochemical system using a CFD based electrochemical simulation approach.
Hence, it is necessary to know the concentration distribution of electrolyte ions in the vicinity of the
electrode surface at various flow velocities, which preludes using the coupled transport equation for
convective diffusion. To develop a comprehensive modeling approach, we focused our attention on the
deposition of copper at the cathode in this simulation. A concentration sink flux equivalent to the
current density was applied at the cathode.
0.00000 0.00025 0.00050 0.00075 0.00100
0
10
20
30
40
50
60
0.0000
0.0002
0.0004
0.0006
0.0008
0.0010
C
o
2
+

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
o
l
/
m
3
)
Distance from cathode (m)
Inlet velocity (m/s)
0.1
0.3
0.5
Concentration
Convection
F
l
u
x

(
m
o
l
/
m
2

s
)
Diffusion

Figure 3. Concentration and mass transport (convection and diffusion) profiles close to the cathode
with three electrolyte flow conditions (0.1, 0.2 and 0.3 m/s).

During electrolysis, cations (Cu
2+
) have to be transported from bulk electrolyte to the cathode
surface. In general, this mass transport occurs by diffusion, convection, and migration. Close to the
electrode surface, the primary mass transport process is diffusion. At an electrolyte region near the
cathode, the diffusion flux is the dominating transport mechanism, while the convection is the
dominant flux in the bulk electrolyte, as shown in Figure 3. The copper ion concentration decreases
with a decreasing distance from the cathode owing to its ion consumption with the electrode reaction.
An increasing flow velocity results in an increase in surface concentration.

Int. J. Electrochem. Sci., Vol. 8, 2013

12340
Cathode
Anode
Inlet
Outlet


Figure 4. Diffusion boundary layer thickness at the cathode (i
app
=100 A/m
2
, inlet velocity =0.2 m/s).

Cathode
Anode
Inlet
Outlet


Figure 5. Surface concentration distribution at the cathode and side channel wall (i
app
=100 A/m
2
,
inlet velocity =0.2 m/s).

The distribution of the diffusion boundary layer thickness was simulated at the cathode surface,
as depicted in Figure 4. A higher thickness value of the diffusion boundary layer gives rise to a
decreased surface concentration as shown in Figure 5.
Int. J. Electrochem. Sci., Vol. 8, 2013

12341
3.2 Mass transport-coupled electric field analysis
Cathode
Anode
Inlet
Outlet


Figure 6. Electric potential distribution between the anode and cathode (i
app
=100 A/m
2
, inlet velocity
=0.2 m/s).

Cathode
Anode
Inlet
Outlet


Figure 7. Electric field distribution between the anode and cathode (i
app
=100 A/m
2
, inlet velocity
=0.2 m/s).

Metal ion reduction at the cathode yields a decrease in the ion concentration, which reduces the
electrode reaction rate. Although the current distribution at the anode as a counter electrode will be
Int. J. Electrochem. Sci., Vol. 8, 2013

12342
non-uniform under a real situation, this is found to have a negligible effect on the rate of the cathode.
Accordingly, the simulation approach using a fixed anode potential shows that (for the same current)
the cathodic current distribution is negligibly affected [12]. This approach allows the average current
density to be fixed. Based on the conservation of charge, the numerically simulated average current on
the cathode should be equal to the total current on the anode.
If the deposition rates at the cathode side are focused on the modeling approach, the rate of the
anodic reaction can be assumed reversible under a constant applied current condition. This assumption
is the ideal case of no surface overpotential at the anode. Figure 6 shows the electric potential
distribution patterns (cross section view) coupled with the cathode overpotential through the
electrolyte region between the anode and cathode. Since herein it is the assumed that the anode is
grounded as 0 V, the potential at the cathode can be calculated from the given flux-out (current density
from Tafel kinetics) using mixed boundary conditions. Owing to the electric conductivity of the
electrolyte, there is a potential increase toward the grounded anode. This is the ohmic potential drop,
which is the loss associated with resistance to the electron transport in the electrolyte region. A
simulation of the electric field between the electrodes is also shown in Figure 7. Spatial current
densities are locally proportional to theses gradients of potential. These results show the case of a
coupled computation with the concentration gradient and surface overpotential.

3.3 Local current density and overpotential
Cathode
Anode
Inlet
Outlet


(a)


Int. J. Electrochem. Sci., Vol. 8, 2013

12343
Cathode
Anode
Inlet
Outlet

(b)

Figure 8. Local overpotential (a) and current density (b) at the cathode surface (i
app
=100 A/m
2
, inlet
velocity =0.2 m/s).

In addition, the model can provide many more details on the mass transport controlled
electrochemical characteristics occurring in an electrowinning cell. An increased diffusion boundary
layer leads to a higher overpotential in this electrode region. Accordingly, an increased inlet velocity
condition results the decreased overpotential owing to a smaller diffusion boundary layer thickness. In
Figure 8, the distribution of overpotential (a) and local current density (b) throughout the cathode
surface are provided for the case of a current density of 100 A/m
2
and an inlet velocity condition of
0.2 m/s. The effects of electrolyte concentration are taken into account in this simulation. It was found
that a higher local overpotential arises from a lower ionic concentration at the electrode surface. Under
this condition, there is not a large potential drop in the electrolyte bulk. Thus the diffusion transport
has a dominant control factor at the electrode surface. It is shown that the increased inlet velocity
condition gives a decreased surface overpotential and has less of an effect on the ohmic drop. This is
evident from the modified Tafel kinetics, which is a function of the overpotential and surface
concentration, as in Equation (4). Accordingly, to maintain the given current density, a higher
overpotential is necessary for the compensation of the ionic reactant depletion.
The diffusion boundary layer () is defined in the literature [17]. Once we know the local
diffusion boundary layer, the current density can be locally estimated according to Ficks law, as in
Equation (8):

Int. J. Electrochem. Sci., Vol. 8, 2013

12344
o
s bulk
C
x
C
C C
zFD
x
C
zFD i

~ |
.
|

\
|
c
c
=
=0
(8)

Since Equation (8) serves to define the diffusion boundary layer thickness, it can be obtained
from the local concentration gradients in the direction normal to the electrode surface. Close to the
cathode surface, the primary mass transport for an electro-transport process is diffusion control. Local
current density can be approximated according to Ficks law as expressed in Equation (8). Once a
concentration gradient in the direction normal to the cathode is obtained from the mass transport
simulation, the diffusion boundary layer thickness () is able to be estimated from the definition as the
intersection between the tangent at the concentration distribution and concentration of the bulk
electrolyte.
0.00 0.01 0.02 0.03 0.04
-200
-150
-100
-50
0
0
100
200
300
400
500
0
10
20
30
40
50
Local current density
Diffusion boundary layer thickness
Surface concentration
Cathode current density: 100 A/m
2
Inlet velocity: 0.2 m/s

L
o
c
a
l

c
u
r
r
e
n
t

d
e
n
s
i
t
y

(
A
/
m
2
)
Distance along the cathode center line (m)
Simulation
Approximation

D
i
f
f
u
s
i
o
n

b
o
u
n
d
a
r
y

l
a
y
e
r

t
h
i
c
k
n
e
s
s

(

m
)

S
u
r
f
a
c
e

c
o
n
c
e
n
t
r
a
t
i
o
n

(
m
o
l
/
m
3
)


Figure 9. Simulation and approximated local current density along the center line of cathode (i
app
=100
A/m
2
, inlet velocity =0.2 m/s).

Figure 9 shows a comparison of the local current density between the simulation and
approximation from the diffusion boundary layer thickness for an applied current density of 100 A/m
2

and inlet velocity condition of 0.2 m/s. In this figure, the local current densities were drawn on the
center line along the cathode surface. The approximated local current density indicates a fairly good
agreement with the simulation results along most of the cathode surface. However, as an
approximation was considered on the concentration gradient along the normal direction to the cathode,
a slight difference was found in the vortex region of the electrolyte facing the protruded cathode.
Figure 10 shows a variation of current density as a function of overpotential based on the
Int. J. Electrochem. Sci., Vol. 8, 2013

12345
average values of the diffusion boundary layer thickness throughout the cathode. The curves were
obtained by the relationship between the current density and overpotential with the values of the
diffusion boundary layer thickness estimated from the concentration profile for various inlet velocities.
The simulated polarization curves clearly show the charge transfer, mixed (charge transfer and mass
transport) and mass transport controlled region.
-0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0
-100
-75
-50
-25
0
Charge transfer control region
Mass transport control region
0.5 m/s
0.4 m/s
0.3 m/s
0.2 m/s
Inlet velocity


L
o
c
a
l

c
u
r
r
e
n
t

d
e
n
s
i
t
y

(
A
/
m
2
)
Overpotential (V)
0.1 m/s

Figure 10. Local current density as a function of overpotential at various inlet velocities (0.1-0.5 m/s).

In the charge transfer controlled region, the electrolyte turbulence has little effect on the
voltage drop. In this simulation condition, the polarizations with a lower exchange current density
would play a significant role in the total losses causing a potential drop from the ideal voltage. It was
simulated that the overpotential contributed to most of the cell potential under this demonstration
condition.
The above simulations provide a good illustration of the capabilities of the present modeling
approach. The proposed CFD model was shown to be able to predict not only the convective diffusion
behavior of the ionic reactants but also the mass-transport limited electrochemical polarizations. It was
found that the flow characteristics of the electrolyte influenced the diffusion-limited current density
and electrowinning deposit. To evaluate these influences, a coupled approach of the multiphysics
simulation for the flow and ion transport is necessary. In addition, more detailed analyses and
benchmarking tests are desirable to fully investigate a validation study.
Int. J. Electrochem. Sci., Vol. 8, 2013

12346
4. CONCLUSIONS
This study demonstrated a modeling approach based on the full coupling of electrochemical
reactions with a mass transport of ionic species using a commercially available CFD platform. An
analysis was carried out for the case of copper electrowinning from a flowing cupric acid electrolyte
through the simplified channel cell with planar and parallel electrodes. The demonstrations presented
here have focused only on copper deposition during steady-state electrolysis. A finite volume
numerical scheme based the CFD technique was successfully adapted to simulate 3-dimensional
behaviors of the electrochemical cell. The CFD model was shown to be able to predict not only the
convective diffusion behavior of the ionic reactants but also the mass-transport limited electrochemical
polarizations. This comprehensive model provides valuable information on the transport phenomena
inside an electrolytic cell such as the ion concentration, electric potential and local current density
distribution under a given current driven condition. Future work will focus on the benchmarking of this
model to validate with a proven electroplating system that controls the fluid dynamics and mass
transport conditions.


ACKNOWLEDGMENTS
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the
Korea government (MSIP).


References

1. T. Storck and A. A. Wragg, J. Appl. Electrochem., 21 (1991) 463.
2. P. Mandin and J. Deconinck, Electrochemical engineering modeling and numerical simulation to
improve the presentation of electrochemical cells and diagnostic techniques, Abstract #2661,
Proceedings 214th ESC Meeting, Honolulu, Hawaii, October 12-17 (2008).
3. P. Mandin, J. M. Cense, G. Picard and D. Lincot, Electrochim. Acta, 52 (2006) 1296.
4. P. Mandin, J. M. Cense, C. Fabian, C. Gbado and D. Lincot, Comput. Chem. Eng., 31 (2007) 980.
5. P. Mandin, R. Wthrich and H. Roustan, Comput. Aided Chem. Eng., 17 (2009) 435.
6. M. F. Santillo, A.G. Ewing and M. L. Heien, Anal. Bioanal. Chem., 399 (2011) 183.
7. S. V. Patankar, Numerical Heat Transfer and Fluid Flow, McGraw-Hill Book Company, New
York, (1980).
8. P. Mandin, J. M. Cense, B. Georges, V. Favre, T. Pauporte, Y. Fukunaka and D. Lincot,
Electrochim. Acta, 53 (2007) 233.
9. C. W. Wang, F. Albano and M. Sastry, Computational method for design and manufacture of
electrochemical systems, United States Patent, US 7,945,344 B2 (2011).
10. B. E. Conway and R. E. White, Modern Aspects of Electrochemistry, Kluwer Academic/Plenum
Publisher, New York (2002).
11. E. Malyshev, U. Landau and S. Chivlikhin, Modeling the deposit thickness distribution in copper
electroplating of semiconductor wafer interconnects, Paper 190 b, Session TK, Proceedings of the
AIChE Annual Meeting, San-Francisco, CA, November 16-21 (2003).
12. C. T. J. Low, E. P. L. Roberts and F. C. Walsh, Electrochim. Acta, 52 (2007) 3831.
13. C. T. J. Low and F. C. Walsh, Surf. Coat. Tech., 202 (2008) 1339.
14. ANSYS CFX Release 12 Users Guide, ANSYS, Inc. (2009).
15. A. J. Bard and L. R. Faulkner, Electrochemical Methods, Fundamentals and Applications, 2nd
Int. J. Electrochem. Sci., Vol. 8, 2013

12347
edition, John Wiley & Son Inc., New York (2001).
16. D. J. Pickett, Electrochemical Reactor Design, American Elsevier Inc., New York (1979).
17. K. Filzwieser, K. Hein and G. Mori, JOM, 54 (2002) 28.


2013 by ESG (www.electrochemsci.org)