The effect of electrode surface modification and cathode overpotential on deposit characteristics in aluminum electrorefining using EMIC-AlCl3 ionic liquid electrolyte

Article abstract of DOI:10.1016/j.electacta.2009.06.059

Electrorefining of aluminum alloy was investigated using AlCl₃ and 1-ethyl-3-methyl-imidazolium chloride (EMIC) (molar ratio AlCl₃:EMIC = 1.65:1) ionic liquid electrolyte on copper and aluminum cathodes at temperature of 90 ± 3 °C an

Full text of DOI:10.1016/j.electacta.2009.06.059

Electrochimica Acta 54 (2009) 6661–6667
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
The effect of electrode surface modification and cathode overpotential on deposit
characteristics in aluminum electrorefining using EMIC–AlCl3 ionic liquid
electrolyte
D. Pradhan, D. Mantha, R.G. Reddy∗
Department of Metallurgical and Materials Engineering, The University of Alabama, Tuscaloosa, AL 35487, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 April 2009
Received in revised form 14 June 2009
Accepted 20 June 2009
Available online 28 June 2009
Electrorefining of aluminum alloy was investigated using AlCl3 and 1-ethyl-3-methyl-imidazolium chlo-
ride (EMIC) (molar ratio AlCl3:EMIC = 1.65:1) ionic liquid electrolyte on copper and aluminum cathodes at
temperature of 90 3 ◦C and cell voltage of 1.5 V. The effect of electrode surface modification and cathode
overpotential on deposit characteristics of aluminum was investigated. The surface modification of elec-
trodes reduced the dendritic depositions of aluminum. It was also observed that cathodic overpotentials
obtained from experiments using modified electrodes are significantly lower than those of unmodified
electrodes. A non-dendritic deposit of aluminum was observed even after prolonged electrorefining of
25 h. Pure aluminum deposits were obtained for all experiments with the current efficiencies in the range
Keywords:
Electrorefining
Ionic liquids
1-Ethyl-3-methyl-imidazolium chloride
Dendrite
of 94–99%.
© 2009 Elsevier Ltd. All rights reserved.
1. Introduction
of aluminum deposition from ionic liquid electrolytes was not
fully investigated. In electrorefining cell, coarse grained, rough and
Aluminum electrorefining/recycling from aluminum scrap using
ionic liquid electrolyte is a technologically attractive process due to
the unique chemical and physical properties of ionic liquids such as
wide temperature range for the liquid phase, high thermal stability,
negligible vapor pressure, low melting point and wide electro-
chemical window. In current scenario, two processes are used
for aluminum electrorefining: three layer process (Hoope’s pro-
cess) and the segregation process. High temperature (800–950 ◦C)
and high energy consumption (15–18 kW h/kg of aluminum) make
three layer process less economical unless for very high-purity alu-
minum production [1]. In the segregation process, the purity of
aluminum is not high enough (99.9%), which requires additional
refining steps for high-purity aluminum production.
For this reason, extraction and refining of aluminum from low
temperature or room temperature ionic liquid electrolytes is gain-
ing interest. Many researchers have studied electrodeposition of
aluminum using different ionic liquid melts [2–8]. But bulk elec-
trodeposition and commercialization is yet to be established. In
recent years, Reddy and co-workers [9–13] have reported bulk elec-
trowinning and electrorefining of aluminum using low temperature
chloroaluminate ionic liquid electrolytes, which yield high-purity
aluminum deposits with zero pollutant emissions. Optimization
adhesive deposits are required. As dendritic deposition is not adher-
ent enough to endure handling before melting and casting into
desirable shapes for further processing, they add additional cost
for handling and processing. Dendrites sometimes short the flow
of current between anodes and cathodes which is not desirable.
Although there are number of experimental parameters such as
electrolyte concentration, surface roughness of electrodes, addition
of additives, impurities, stirring, temperature and deposition time
to prevent the dendritic deposition, current density and cathode
overpotential play the major role.
There are very few investigations reported in the literature on
the deposition of aluminum on copper [3,9–13] and aluminum
[14,15]. Aluminum deposition on aluminum or copper substrates
can be divided into two steps: (a) a thin layer deposited on the
substrate, (b) dendrites attaching to this thin layer. Adherence
depends directly on the reactivity between the aluminum and cath-
ode material. Formation of intermetallic compounds like Al2Cu
due to interdiffusion implies strong reactivity. The presence of an
underlayer ensures good adherence of the dendrites. In the case
of aluminum cathode, there is a low reactivity between the alu-
minum and the substrate. At the beginning of growth of layer,
Budevski et al. have proposed models taking into account the bind-
ing energy of the electrodeposited metal (M), either on substrate
(S), EMS or on the pre-deposited metal EMM [16,17]. When EMS ≤ EMM
(Volmer–Weber model), M becomes deposited on M rather than on
S, leading to three-dimensional islands. This initial process favors
∗
Corresponding author. Tel.: +1 205 348 4246; fax: +1 205 348 2164.
E-mail address: rreddy@eng.ua.edu (R.G. Reddy).
0013-4686/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2009.06.059

6662
D. Pradhan et al. / Electrochimica Acta 54 (2009) 6661–6667
Fig. 1. (a) Schematic diagram of the experimental cell and (b) electrorefining cell.
a rapid formation of axial dendrites on the cathode surface. In case
of copper, a surface alloy is formed by metal interdiffusion with
the underlayers of aluminum and dendrite formation occurs when
the intermetallic layer is saturated. At this time EMM is higher than
EM-alloy which leads to the formation of 3D islands. During deposi-
tion growth, orthogonal needles result from diffusional phenomena
favoring current at the top of the islands. The heterogeneity of the
electrolyte layer at the interface with the cathode material causes
a particular fluid motion which readily influences the morphology
of the deposition. Fleury et al. [18,19] demonstrated that the deple-
tionofcationsnearthecathodegeneratesstronglocalelectricfields.
According to these authors, the electrolyte motion follows contra-
rotative vortices at the tip of each needle, producing ramification
at the edges.
investigation. The cathode area of deposition and anode work-
ing area were calculated from the exposed area of electrodes into
the electrolyte. The purpose of reference electrode was only to
measure the electrode potential of anode and cathode individu-
ally, using a multimeter (Keithley® 2000 Multimeter). The constant
voltage power source (Kepco® ABC Programmable Power Supply)
supplied the required voltage across anode and cathode. The elec-
trolyte was stirred at a constant speed (60 rpm) using a magnetic
stirrer. The temperature was controlled by placing the electrorefin-
ing cell on a hot plate. A thermometer was inserted through the
teflon cap to monitor the temperature. The electrolyte was heated
stepwise to the required temperature under a controlled atmo-
sphere (argon). The electrolyte was then allowed to stand for at least
30 min at the final temperature for stabilization before starting the
experiment.
In order to determine the effect of cathode overpotential
on deposit morphology, a set of experiments were conducted
with three-electrode system using EG&G PARC model 273A
potentiostat/galvanostat® attached with Power Suite® software.
Similar experimental setup (Fig. 1) as mentioned above was used
for all experiments. Constant cathode potential ranging from
−0.25 V to −0.65 V were supplied between modified aluminum
sheet cathode (working electrode) and aluminum wire (reference
electrode). Modified aluminum alloy anode with dimensions of
45 mm × 20 mm × 1 mm was used as counter electrode.
In this article, we have investigated the effect of electrode
surface modification and cathode overpotential on deposit mor-
phology of aluminum using AlCl3 and EMIC (molar ratio 1.65:1)
electrolyte at 90 3 ◦C.
2. Experimental
The electrorefining experiments were conducted in a 40 ml
Pyrex® glass cell fitted with teflon cap. The schematic diagram
and actual electrolytic cell are shown in Fig. 1(a) and (b), respec-
tively. The anodes were prepared by cutting the aluminum alloy
ingot material into 45 mm × 20 mm × 5 mm plate. Table 1 shows
the compositional analysis of aluminum alloy. Copper and alu-
minum (99.999% pure, Alfa Aesar®) sheets with dimensions of
45 mm × 20 mm × 1 mm were used as cathodes. Pure aluminum
wire (2 mm diameter, 99.999% pure, Alfa Aesar®) was used as refer-
ence electrode. AlCl3 and 1-ethyl-3-methyl-imidazolium chloride
(EMIC) (molar ratio 1.65:1) were used as is.
All experiments were conducted in a ventilated hood, under
argon atmosphere at 90 3 ◦C with duration of 5 h except one for
25 h. After experiments, thorough washing of electrodes were car-
ried out with acetone and double distilled water to remove the
electrolyte from the deposit before characterization.
The qualitative and quantitative analysis of the electrodeposited
aluminum were carried out using Philips® XL30 scanning elec-
tron microscope (SEM) with energy dispersive spectroscope (EDS).
The X-ray diffraction (XRD) analysis was performed using Philips®
PW3830 X-ray diffractometer which uses a monochromatic CuK␣
radiation (ꢀ = 1.5406 Å).
All electrodes were polished with a 320 grit SiC emery paper
to expose fresh metal surface before electrorefining. Some of
the anode and cathode back surfaces were masked with non-
conductive tape (HP260®) and referred as “modified” in this
Table 1
Elemental analysis of aluminum alloy anode.
Elements
Wt%
Al
Si
Fe
Mn
Mg
Cu
Ti
+Cr
Ni
Zn
74.25
25.07
0.08
0.008
0.345
0.02
0.13
0.002
0.007
0.03

D. Pradhan et al. / Electrochimica Acta 54 (2009) 6661–6667
6663
Table 2
Summary of experimental results carried out using aluminum alloy anode and aluminum wire reference electrode (applied voltage = 1.5 V, temperature = 90
speed = 60 rpm).
3
◦C and stirring
Exp. no.
Electrode (cathode)
Electrode surface
modification
Highest cathode
Overpotential
(V)
Average current
density (A/m2)
Current efficiency (%)
Energy consumption
(kW h/kg of Al)
Deposit morphology
1
2
3
Cu
Cu
Al
Al
Al
Unmodified
Modified
Unmodified
Modified
−0.52
−0.33
−0.73
−0.32
−0.24
151
163
205
212
127
94.10
96.55
97.52
94.17
99.27
5.05
4.79
4.70
5.17
4.92
Dendritic
Granular
Dendritic
Granular
Granular
4
5a
Modified
a
25 h experiment.
3. Results and discussion
passage of time. After 5 h of experiment, primary dendritic arms
grow up to 4–6 mm as shown in SEM micrograph in Fig. 3. But mod-
ified electrodes did not produce dendritic aluminum deposit for the
entire experimental time. The deposit was granular, compact and
adherent with copper surface (see Fig. 5). After 25 h of electrorefin-
ing with modified electrode no dendritic deposition was produced.
Similar results were obtained by using modified and unmodified
aluminum cathodes.
Fig. 6 shows the variation of cathode current density with
time using modified and unmodified copper and aluminum cath-
odes. Modified electrodes exhibit higher average current density
(163 A/m2 for copper and 212 A/m2 for aluminum, see Table 2)
for both copper and aluminum cathodes. The modified electrodes
have a blocked back face, resulting in increased current density.
The plots also depict that the cathode current density obtained
using modified cathode is higher than that of unmodified for the
first 15–25 min of experiment, after which the behavior is sim-
ilar. It was observed that copper cathode exhibit lower average
current density than aluminum electrode. The rough deposit of alu-
minum was found on copper cathode due to reactivity between the
aluminum and cathode material. This leads to increase in actual
surface area of copper cathode and hence average current density
decreases.
3.1. Electrode reactions
In electrorefining process, aluminum alloys were used as anode.
The summary of experimental results is tabulated in Table 2.
The electrolyte exhibits Lewis acid properties which has a molar
ratio of AlCl3:EMIC = 1.65:1.−Acidic melt contains anions such as
−
−
AlCl4−, Al3Cl10 and Al4Cl13 along with Al2Cl7 ions. Aluminum
−
deposition proceeds by the reduction of Al2Cl7 ions which is
present in excess in acidic melts. The overall mechanism can be
explained by the following electrode reactions. The AlCl4 ions
present in the electrolyte react with anode metal to dissolve alu-
minum and produce Al2Cl7− ions as shown by the anodic reaction in
Eq. (1).
−
Al(anode) + 7AlCl4− → 4Al2Cl7− + 3e−
(1)
−
Dissolved aluminum in the form of Al2Cl7 ion traverses to the
cathode either by diffusion or by convection and gets discharged.
The reaction which leads to the aluminum deposition at the cathode
is given by the following reaction in Eq. (2).
4Al2Cl7− + 3e− → Al(cathode) + 7AlCl4
(2)
−
The variation of cathode current density with time shown in
Fig. 6 has three distinct regions (A–C) as we observed in our pre-
vious investigations [12,20]. The initial steep increase in current
density (region A) for initial period of about 20 min is due to acti-
vation polarization of cathode. To confirm activation polarization,
cathode overpotential (ꢁ) is plotted with cathodic current density
(i) in region A. Fig. 7 shows the linear relationship of cathode over-
potential (ꢁ) and log(i) for aluminum cathode and is represented
The energy consumption for Al electrorefining varies from
4.79 kW h/kg to 5.05 kW h/kg of Al, which is well below the com-
mercial and the molten salt electrorefining of Al (15–18 kW h/kg
of Al). The current efficiency of electrorefining is in the range of
94–99%.
3.2. Characterization of electrodeposits
The typical XRD pattern of the electrodeposits is shown in Fig. 2.
It shows only the diffraction peaks of Al, indicating that the deposit
was of high purity.
Fig. 3(a)–(c) shows the SEM microstructure of the aluminum
dendrites obtained after 5 h of experiment using unmodified cop-
per cathode and the corresponding compositional analysis by EDS.
The EDS analysis shows that the deposit is composed of pure alu-
minum. The SEM micrograph of the electrodeposit obtained by
using modified copper cathode shows granular structure as shown
in Fig. 4(a)–(c).
3.3. Effect of electrode surface modification
Fig. 5 shows the deposit morphology of aluminum using both
modified and unmodified copper electrodes. The photographs
clearly show that the use of surface modified electrodes resulted
in non-dendritic or granular deposition of aluminum on copper
cathode for an entire duration of 5 h. After 1 h both type of elec-
trodes show bright deposit of aluminum which is clearly visible
at the edges of the cathodes. Unmodified copper cathode shows
fine dendritic deposition of aluminum on the edges and on the
surfaces. The growth of dendritic deposits was pronounced with
Fig. 2. X-ray diffraction pattern of deposited aluminum using unmodified copper
electrode.

6664
D. Pradhan et al. / Electrochimica Acta 54 (2009) 6661–6667
Fig. 3. (a) Cross-sectional SEM image of aluminum deposit, (b) and (c) SEM micrographs and EDS analysis of aluminum dendrites obtained on unmodified copper cathode.
Fig. 4. (a) Cross-sectional SEM image of aluminum deposit, (b) and (c) SEM micrographs and EDS analysis of granular aluminum deposits obtained by using modified copper
cathode.

D. Pradhan et al. / Electrochimica Acta 54 (2009) 6661–6667
6665
Fig. 5. Deposit morphology of aluminum at different time intervals using modified and unmodified copper cathodes.
by cathodic Tafel equation, Eq. (3).
electrodes. Later, current density decreases steadily (Region C). This
is because of the thin layer formation on anode surface. The layer
consists of mainly silicon and other impurities which retard the dis-
solution kinetics of anode [12]. Hence, concentration polarization
occurs in this region.
The morphology of metal deposit is mainly dependent on the
kinetic parameters and the cathodic overpotential or the current
density. A certain well-defined critical overpotential value appears
to exist below which dendrites do not grow. Experiments using
modified electrodes produced non-dendritic and granular struc-
ture and it can be explained from the cathodic overpotential data
as shown in Fig. 8. Modified cathodes showed lower overpotential
compared to unmodified electrodes. For example, the maximum
overpotential obtained with modified aluminum cathode is 0.321 V
and that with unmodified is 0.731 V (see Table 2 and Fig. 8). The
maximum current densities obtained in both the experiments are
ꢀ
ꢁ
−˛cnF
log i = log i0
+
ꢁ
(3)
2.303RT
where i is the cathode current density (A/m2), i0 the exchange cur-
rentdensity(A/m2), ꢁ iscathodeoverpotential(V), ˛c isthecathodic
charge transfer coefficient, n is the number of electron exchanged,
F is the Faraday’s constant (96,500 coulomb/equiv.), R is the univer-
sal gas constant (8.314 J/mole K) and T is the operating temperature
(K).
The charge transfer coefficient ˛c can be calculated from the
slope of each plot in Fig. 7 using Tafel equation, Eq. (3). The exchange
current density (i0) can be determined from the intercept of plot in
Fig. 7 at ꢁ = 0. The approximate value of i0 is ∼57 A/m2. The cathode
charge transfer coefficient ˛c obtained from Fig. 7 is 0.2.
Region B (Fig. 6) refers to the steady-state for a period of
25–30 min. But this region is very short (10–15 min) for modified
Fig. 6. Effect of electrode surface modification on cathode current density using
Fig. 7. Cathodicpolarizationcurveshowsthevariationofcathodeoverpotentialwith
log(i) for aluminum/copper cathodes at 1.5 V and 90
copper and aluminum cathodes at 1.5 V, 90
3
◦C and 60 rpm.
3
◦C.

6666
D. Pradhan et al. / Electrochimica Acta 54 (2009) 6661–6667
Fig. 8. Cathodic overpotential as a function of time using unmodified and modified
electrodes.
nearly the same but the measured overpotentials are lower by a
factor of two for modified aluminum cathode. A similar result was
found for modified and unmodified copper cathodes. This indicates
that overpotentials play a significant role in the morphology of the
deposits.
Fig. 9. Variation of cathode current density with time at different cathode overpo-
tentials (−0.25 V to −0.65 V) using modified aluminum cathode at 90
◦C.
3
3.4. Effect of cathode overpotential
The SEM micrograph (Fig. 10) illustrates the change in mor-
phology of aluminum deposit on cathode surface. It shows that
at lower overpotential (<−0.55), aluminum particles are granular
and isolated. After that the particles begin to cluster and dendritic
deposition was found afterwards (−0.575 V, −0.60 V and −0.625 V).
These distinctive micrographs conclude that −0.55 V is the tran-
sition overpotential where morphology changes from granular to
dendritic.
To understand this behavior, a polarization curve was plotted
with average cathode current density vs. cathode overpotential
(from Fig. 9) as shown in Fig. 11. The pictures of cathode with
aluminum deposit are numbered corresponding to those on the
curve.
The polarization curve clearly shows the transition point (point
4) on the curve corresponding to −0.55 V where the deposit
morphology changes from granular to dendritic. The critical over-
potential (ꢁcrt) at which dendritic growth starts is about −0.55 V.
Further calculation of critical overpotential (ꢁcrt) was carried out
using Eqs. (4)–(9). The expression for current–overpotential rela-
To study the effect of cathode overpotential on deposit mor-
phology, a series of experiments were carried out using modified
aluminum cathode in a three-electrode electrorefining cell at
90 3 ◦C. The cathode overpotential varies from −0.25 V to −0.65 V
and was applied between reference electrode and working elec-
trode (modified aluminum cathode). Modified aluminum alloy
anode was used as counter electrode. Fig. 9 shows the variation of
cathode current density with time at different cathode overpoten-
tials. The curves have similar trend up to −0.55 V of overpotential
and granular deposit of aluminum was obtained. At higher overpo-
tential a steep increase in current density are observed between
−0.575 V and −0.625 V overpotential. Dendritic deposition was
obtained within this range. At −0.65 V of overpotential, a powdery
deposition was found and which was non-adherent to the cathode
surface. Theplotsalsoindicatethatwecangetgranulardepositeven
at very high current density (400–500 A/m2) which is comparable
with high temperature industrial electrorefining processes.
Fig. 10. SEM micrographs of aluminum deposit on modified aluminum cathode at different cathode overpotentials: (a) −0.25 V, (b) −0.45 V, (c) −0.55 V, (d) −0.575 V, (e)
−0.60 V, (f) −0.625 V at 90
◦C.
3

D. Pradhan et al. / Electrochimica Acta 54 (2009) 6661–6667
6667
The ꢁcrt is calculated by substituting ˛c ≈ 0.2, i0 = 57 A/m2,
iL = 518 A/m2 and T = 363 K in Eq. (8). The calculated value of ꢁcrt
is −0.536 V which is in good agreement with the value obtained
from the polarization curve (∼ −0.55 V).
Hence, it can be concluded that the surface modification of
electrodes helps in obtaining non-dendritic deposits. The cathode
overpotential of −0.55 V or lower is recommended for electrorefin-
ing of aluminum.
4. Conclusions
The electrorefining experiments were carried out using alu-
minum alloy anode and copper/aluminum cathode materials from
AlCl3–EMIC (molar ratio 1.65:1) mixture at 1.5 V, 90 3 ◦C, 60 rpm
for 5 h. Pure aluminum deposits were obtained. The electrorefin-
ing process has high current efficiency (94.10–99.27%). The surface
modification of electrodes eliminates the dendritic deposition of
aluminum. It was observed that cathode overpotentials obtained
from experiments using modified electrodes are significantly lower
than those obtained using unmodified electrodes. Thus, cathode
overpotential is the key in preventing dendrite formation of alu-
minum deposits on cathode. The critical overpotential (ꢁcrt) at
which dendritic growth begins is about −0.55 V and is very close
to the calculated value of −0.536 V. Hence, cathode overpotentials
lowerthan−0.55 Varerecommendedforthepreventionofdendrite
formation of aluminum deposits on copper/aluminum cathodes.
Fig. 11. Polarization curve for electrodeposited aluminum on modified aluminum
cathode deposit at different cathode overpotential (−0.25 V to −0.625 V) at 90
3
◦C.
tionship in electrodeposition is given by Eq. (4) [21]
i0(fc − fa)
i =
(4)
1 + (i0fc/iL)
where iL is the limiting current density (A/m2), i0 is the exchange
current density (A/m2). Considering cathodic current density and
overpotential as positive, fa and fc are expressed as follows:
Acknowledgements
ꢀ
ꢁ
ꢀ
ꢁ
˛aFꢁ
RT
˛cFꢁ
RT
The authors gratefully acknowledge the financial support from
NSF, ACIPCO and The University of Alabama.
fa = exp
−
and fc = exp
(5)
Suffix “a” and “c” refers to anode and cathode, respectively. For
flat electrode, the condition for initiation of dendritic growth is
given by Eq. (6)
References
[1] M. Kondo, H. Maeda, M. Mizuguchi, J. Metals 42 (11) (1990) 36.
[2] R.T. Carlin, W. Crawford, M. Bersch, J. Electrochem. Soc. 139 (1992) 2720.
[3] Q. Liao, W.R. Pitner, G. Stewart, C.L. Hussey, G.R. Stafford, J. Electrochem. Soc.
144 (1997) 936.
h
iL ≤ i0 fcrt
(6)
ı
[4] J. Robinson, R.A. Osteryoung, J. Electrochem. Soc. 127 (1980) 122.
[5] C.-C. Yang, Mater. Chem. Phys. 37 (1994) 355.
where h is the height of the growing dendrite, ı is the diffusion
boundary layer thickness and fcrt corresponds to ꢁcrt. Now taking
the cathodic expression in Eq. (5) and substituting in Eq. (6) we
obtain:
[6] P.K. Lai, M. Skyllas-Kazacos, J. Electroanal. Chem. 248 (1988) 431.
[7] P.K. Lai, M. Skyllas-Kazacos, Electrochim. Acta 32 (1987) 1443.
[8] W.R. Pitner, C.L. Hussey, G.R. Stafford, J. Electrochem. Soc. 143 (1997) 130.
[9] V. Kamavaram, R.G. Reddy, Light Metals, TMS, Warrendale, 2005, p. 501.
[10] M. Zhang, V. Kamavaram, R.G. Reddy, J. Metals 55 (11) (2003) 54.
[11] V. Kamavaram, R.G. Reddy, in: R.E. Aune, M. Kekkonen (Eds.), Metal Separation
Technologies III, Helsinki University of Technology, Espoo, Finland, 2004, p. 143.
[12] V. Kamavaram, D. Mantha, R.G. Reddy, Electrochim. Acta 50 (2005) 3286.
[13] V. Kamavaram, D. Mantha, R.G. Reddy, J. Min. Met. 39 (1–2 B) (2003) 43.
[14] A.P. Abbott, C.A. Eardly, N.R.S. Faley, G.A. Griffith, A. Pratt, J. Appl. Electrochem.
31 (2001) 1345.
[15] T.P. Moffat, J. Electrochem. Soc. 141 (1994) 3059.
[16] W.J. Lorenz, G. Staikov, Surf. Sci. 335 (1995) 32.
[17] E. Budevski, G. Staikov, W.J. Lorenz, Electrochemical Phase Formation and
Growth, VCH Publications, New York, 1996.
[18] V. Fleury, J.N. Chazalviel, M. Rosso, Phys. Rev. E 48 (2) (1993) 1279.
[19] V. Fleury, J.H. Kaufman, D.B. Hibbert, Nature 367 (3) (1994) 435.
[20] D. Pradhan, R.G. Reddy, A. Lahiri, Metall. Trans. B 40B (2009) 114.
[21] K.I. Popov, S.S. Djokic´, B.N. Grgur, Fundamental Aspects of Electrometallurgy,
Springer, 2002, p. 28.
[22] D. Pradhan, R.G. Reddy, Electrochemical Studies of Al Electrodeposition
from EMIC–AlCl3 Electrolytes at Low Temperatures, Department of Metal-
lurgical and Materials Engineering, The University of Alabama, USA, 2008
(Unpublished).
ꢂ
ꢃ
RT
iL
ı
ꢁcrt
=
ln
(7)
˛cF
i0
h
The minimum overpotential for which the dendritic growth will
be initiated (when h = ı) is given by Eq. (8)
ꢀ
ꢁ
RT
iL
ꢁcrt
=
ln
(8)
˛cF
i0
The iL was calculated from Eq. (9)
nFDC0
=
ı
iL
(9)
where n is the number of electron−s exchanged (3/4, see Eq. (2)), D is
the diffusion coefficient of Al2Cl7 ion, C0 is bulk concentration of
Al2Cl7− ion and ı is the diffusion boundary layer thickness (m). Sub-
stituting the values of D = 6.89 × 10−11 (m2/s) [22], C0 = 1.8185 × 103
(mol/m3) and ı = 17.5 × 10−6 m [23] in Eq. (9) yield the limiting cur-
rent density (iL) as ∼518 A/m2.
[23] J.K. Chang, S.Y. Chen, W.T. Tsai, M.J. Deng, I.W. Sun, J. Electrochem. Soc. 155 (3)
(2008) C116.