Optimization of pulsed electrodeposition of aluminum from AlCl 3-1-ethyl-3-methylimidazolium chloride ionic liquid

Article abstract of DOI:10.1016/j.electacta.2010.10.056

In this study, Al was electrodeposited on a platinum substrate at room temperature from an ionic liquid bath of EMIC containing AlCl₃ using potentiostatic polarization (PP), galvanostatic polarization (GP), monopolar current pulse polarization

Full text of DOI:10.1016/j.electacta.2010.10.056

Electrochimica Acta 56 (2011) 1130–1137
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Optimization of pulsed electrodeposition of aluminum from
AlCl3-1-ethyl-3-methylimidazolium chloride ionic liquid
Jinwei Tang∗, Kazuhisa Azumi
Graduate School of Engineering, Hokkaido University, N13W8, Kitaku, Sapporo 060-8628, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 August 2010
Received in revised form 18 October 2010
Accepted 18 October 2010
Available online 23 October 2010
In this study, Al was electrodeposited on a platinum substrate at room temperature from an ionic liq-
uid bath of EMIC containing AlCl3 using potentiostatic polarization (PP), galvanostatic polarization (GP),
monopolar current pulse polarization (MCP) and bipolar current pulse polarization (BCP). Transition
of current or potential during galvanostatic or pulse polarization revealed that the initial stage of the
deposition process was controlled by a nucleation process depending on the polarization condition. For
example, the average size of Al deposits decreased with increasing current density in the case of GP.
FE-SEM observation showed that dense and compact Al deposits with a smooth surface were obtained
by the current pulse method. Roughness factor evaluated from electrochemical impedance measurement
confirmed the smooth surface of these deposits. Adhesion strength of Al deposits was greatly improved
using BCP in which an anodic pulse was combined with a cathodic pulse for electrodeposition. In this
study, the optimal parameters for BCP were found to be IC = −16.0 mA cm−2, IA = 1.0 mA cm−2, rC (duty
ratio) = 0.5, and f = 2 Hz. The mechanisms of electrodeposition by these three methods are discussed.
© 2010 Elsevier Ltd. All rights reserved.
Keywords:
Ionic liquid
Aluminum
Pulsed electrodeposition
1. Introduction
ically reduced to form metallic deposits according to the following
reaction [6]:
Aluminum and its alloys have low density and high levels of cor-
rosion resistance and electric conductivity and are thus promising
materials for various applications. Because of the high corrosion
resistivity of Al, various methods for Al coating of corrodible mate-
rials have been developed. Compared with physical deposition
methods such as thermal spray coating [1], physical or chemical
vapor deposition [2], electroless plating [3] and hot dipping [4], the
electrodeposition method has many advantages such as low energy
consumption, low cost, simple operation, and better control of the
microstructure of deposited layers. However, Al coatings cannot
be electrodeposited in an aqueous solution because water decom-
poses prior to Al deposition. Room temperature ionic liquids (RTIL)
have many unique features such as low vapor pressure, good ionic
conductivity and a wide electrochemical potential window, and
they have recently been considered as media for Al electrodepo-
sition [5,6]. One of the ionic liquids used for electrodeposition of Al
is a mixture of 1-ethyl-3-methyl-imidazolium chloride (EMIC) and
aluminum chloride (AlCl3), from which various metals and alloys,
such as Al, Al–Zn [7], Al–Mo–Ti [8], and even AlInSb semiconduc-
tor [9], have been electrodeposited. In this acidic ionic liquid, it has
been suggested that the dominant ions of Al2Cl7− are electrochem-
−
−
4Al2Cl7 + 3e− ↔ Al + 7AlCl4
(1)
The pulse current technique has been used for electrodeposition
of aluminum [10–12] because it enables the properties and struc-
tures of deposits to be controlled and improved by adjusting pulse
parameters such as frequency, duty ratio, pulse intensity and polar-
ity. Yang [10] studied the effects of the electrolyte composition,
current density and current form on characteristics of Al layers
deposited from a molten salt system of AlCl3-n-butylpyridinium
chloride and found that more compact and smoother layer was
obtained with pulse current method compared with DC method.
Li et al. [11] also investigate the effects of pulse parameters on Al
electrodeposits from AlCl3-EMIC ionic liquid containing NdCl3, and
found that the micro-deposits changed from polygonal to spheri-
cal crystals as increasing DC and the pulse current gave the brighter
and flatter deposits compared with the DC method with the same
average current density.
Despite the potential improvement of aluminum deposition
using the pulse technique and ionic liquids, there has been no
fundamental and systematical investigation of the pulse polariza-
tion process and optimization of pulse conditions. In the present
work, therefore, prior to explore the electrodeposition condition
of Al on the other metals such as Mg and steels for corrosion
protection, an Al layer was electrodeposited on a Pt substrate
from EMIC/AlCl3 ionic liquid using various polarization conditions
∗
Corresponding author. Tel.: +81 11706 6749; fax: +81 11706 7897.
E-mail address: tangjw1984@gmail.com (J. Tang).
0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.10.056

J. Tang, K. Azumi / Electrochimica Acta 56 (2011) 1130–1137
1131
including potential and current pulse technique, and the results
were compared to clarify the deposition process and to obtain
optimum deposition conditions.
100
50
500 mV/s
300
200
150
100
2. Experimental
60
35
The ionic liquid [EMIm]Cl (Aldrich, BASF quality, purity ≥ 95%)
was dried at 373 K in vacuum for 24 h and stored in a glove box filled
with nitrogen gas. Anhydrous AlCl3 powders (Junsei Chemical Co.,
purity ≥ 98.0%) were used as a source of Al without further purifi-
cation. An electrodeposition bath composed of AlCl3 and [EMIm]Cl
with a molar ratio of 2:1 was prepared by slow addition of AlCl3
to [EMIm]Cl in a beaker at room temperature with stirring. Some
fumes and heat were generated during preparation, and then a
slightly dark-brown liquid was obtained. For further purification,
an Al foil (≥99.999%) was immersed in the bath for 3 days to remove
residual water by water decomposition reaction accompanied by
hydrogen gas evolution [13]. Finally, a transparent liquid with a
slightly yellow color as an electrodeposition bath was obtained.
In the electrodeposition process, a typical three-electrode sys-
tem composed of a working electrode (WE) of Pt foil (7 × 7 mm),
a reference electrode (RE) of Pt wire or Al wire (99.99%), and a
counter electrode (CE) of an Al plate (99.999%) was used. In this
paper, electrode potential is presented as a value vs. Al-RE. Before
experiments, the WE was cleaned in water by ultrasonic cleaning
for 5 min followed by acetone rinsing and then dried with a stream
of air. Then the WE was immediately transferred into the glove box
and immersed in the ionic liquid bath for 20 min prior to polariza-
tion. The WE was then polarized between 1.0 V and 2.0 V (vs. Pt-RE)
with a scan rate of 0.1 V s−1 to remove adsorbed impurities [6].
A potentio/galvanostat (Ivium Technologies, model IviumStat)
was used for electrochemical operations such as electrodeposition,
cyclic voltammetry and electrochemical impedance spectroscopy
(EIS). Electrodeposition of Al was conducted using galvanostatic
polarization (GP), monopolar current pulse (MCP), or bipolar cur-
rent pulse (BCP) polarization. A voltammogram was measured with
potential scanning from open circuit potential (OCP, approximately
1.1 V vs. Al-RE) toward the negative direction. EIS measurement of
Al deposits was conducted by applying an ac signal of 0.05 VP–P
(considerable noise appeared below this amplitude.) in the fre-
quency range of 1–105 Hz in borate buffer solution (0.15 M H3BO3
and 0.05 M Na2B4O7, pH 8.5) [14] to evaluate surface roughness of
deposits. All experiments were conducted at room temperature.
Microstructures of surfaces and cross sections of deposits were
observed using a field emission scanning electron microscope (FE-
SEM, Jeol Co., model JSM-6500).
0
start potential
-50
-100
-0.5
0.0
0.5
1.0
Potential /V vs. Al/AlCl3
Fig. 1. Cyclic voltammograms of Pt in the [EMIm]Cl/AlCl3 (0.5 mol%) plating bath
measured with scan rates of 35, 60, 100, 150, 200, 300, and 500 mV s−1
.
Dependence of cathodic and anodic current peaks on square root
of the scan rate is shown in Fig. 2. The linear relationship for the
cathodic current peak indicates that the deposition rate was deter-
mined by the diffusion process. Deviation of peaks from the linear
line at high scan rates was attributed to insufficient growth of the
diffusion layer due to a fast potential sweep.
Fig. 3 shows the coulombic efficiency of Al deposition, i.e., ratio
of anodic charge consumed by Al dissolution against the cathodic
charge during the cathodic polarization, as a function of square
root of the scan rate. The efficiency increased and then decreased
with increase in scan rate. Relatively low efficiency at a low scan
rate indicates that the contribution of a side reaction other than
the deposition reaction becomes apparent. This side reaction is
thought to be related to impurities in the organic solvent such as
residual water or oxygen dissolved from incompletely deaerated
atmosphere in the glove box [15]. Decrease in efficiency at a higher
scan rate was caused by growth of a depletion layer of Al ions on
the electrode surface, which also accelerates the side reaction due
to insufficient supply of Al ions.
3.2. Potentiostatic polarization (PP)
Chronoamperometry at the initial stage of electrodeposition
provides useful information on nucleation and growth processes.
100
ipa
3. Results and discussion
50
0
3.1. Voltammetry
Typical voltammograms of a Pt substrate at various scan rates
in [EMIm]Cl/AlCl3 are shown in Fig. 1. Al deposition in the cathodic
scan and its stripping in the anodic scan can be clearly seen in
the figure. Al deposition potential was lower than −0.2 V, indicat-
ing that relatively large nucleation overpotential was necessary.
Both the deposition and stripping current of Al were enhanced by
increasing the scan rate. Appearance of a current peak during the
cathodic scan at a relatively slow scan rate indicates the formation
of a depletion layer of Al ions around the Pt-WE. In the anodic scan,
a shoulder peak around 0 V became more apparent as the scan rate
decreasing, w−hich may be attributed to the formation of depression
layer of AlCl4 on WE necessary for Al dissolution. The following
clear peak and abruption of anodic current were completion of Al
deposits dissolution.
-50
ipc
0
5
10
15
υ1/2 / (mV s-1) 1/2
Fig. 2. Relationship between cathodic peak current (ipc) and anodic peak current
(ipa) against square root of the scan rate (ꢀ) obtained from the data shown in Fig. 1.

1132
J. Tang, K. Azumi / Electrochimica Acta 56 (2011) 1130–1137
100
95
90
85
80
75
0.0
-4.5
-3.0
-0.15
-0.20
-0.25
-0.30
-0.35
-0.40
-0.1
-0.2
-0.3
-0.4
-0.5
-6.0
-8.0
12
0
4
8
-3.0
-4.5
-6.0
-8.0
mA cm-2
5000
0
1000
2000
3000
4000
6
8
10
12
14
16
18
Time /s
υ1/2 /(mV s-1) 1/2
Fig. 5. Potential–time transition of Pt electrode during galvanostatic polarization
for Al electrodeposition in the [EMIm]Cl/AlCl3 (0.5 mol%) bath at various current
densities.
Fig. 3. Efficiency of Al electrodeposition against square root of the scan rate (ꢀ)
during the potential scan shown in Fig. 1.
Typical current–time transients of Pt-WE during cathodic PP are
shown in Fig. 4 as a function of polarization potential. The large
cathodic current in the first part of the transition was due to sud-
den deposition of Al to form nuclei, and the following decrease was
caused by formation of a depletion layer of Al ions. The subsequent
increase of deposition current was caused by nuclei growth, which
enabled efficient collection of Al ions due to a steric effect. The
behavior of the deposition current in the following stage depended
on the deposition potential. At a potential higher than −0.4 V, a
diffusion layer of Al ions formed near the Pt-WE and the thick-
ness of the layer was balanced with the constant deposition rate.
On the other hand, at a potential lower than −0.5 V, the continu-
ously decreasing deposition current indicates continuous growth
of the diffusion layer due to a higher rate of Al ion consumption.
The required time to reach the transition from the nucleation stage
to the diffusion control stage, ꢁ, is indicated in the figure. The value
of ꢁ decreased with lowering of the deposition potential, indicat-
ing rapid completion of the nucleation stage at a higher deposition
rate.
showed a drop for a few seconds as shown in a time-expansion plot.
This characteristic behavior of potential transition corresponds to
the nucleation process shown in Fig. 4, and the duration of this
nucleation stage became short with increasing deposition current.
SEM micrographs of the Al surface electrodeposited using GP
at various current densities are shown in Fig. 6. Higher magni-
fication images are also shown as insets. At current density of
−3.0 mA cm−2 shown in Fig. 6(a), Al deposits were composed of
flake-like particles with poor adherence and roughly covered the
substrate surface. At −4.5 mA cm−2 shown in Fig. 6(b), the deposits
became fine but were still porous. Compact and dense deposits con-
tinuously covering the substrate were obtained at −6.0 mA cm−2
as shown in Fig. 6(c). Deposits obtained at −8.0 mA cm−2 showed
smooth and dense surface composed of finer particles as can be
seen in Fig. 6(d). It is evident that the average size of deposit parti-
cles decreased with increasing current density. This tendency can
be explained by the model in which initial deposition of Al at low
current density occurs only on limited sites due to low overpoten-
tial and each deposit grows slowly due to the low deposition rate.
On the other hand, at high current density, initial deposition of Al
occurs at various sites due to large deposition overpotential, and
further deposition occurs at other sites to form new deposits rather
than growth of pre-deposits because of the high deposition rate.
In the cross-sectional SEM image of sample d shown in Fig. 6(e),
dense and compact deposits of approximately 9 ␮m in thickness
were obtained, although some of the deposits had already detached
from the Pt substrate due to poor adhesion.
3.3. Galvanostatic polarization (GP)
Variation of electrode potential during GP at various current
densities is shown in Fig. 5. Polarization potential became less-
noble with increasing deposition current. The initial potential
-12
-0.70
3.4. Monopolar current pulse polarization (MCP)
-10
τ
An example of the current pulse waveform and corresponding
potential response during MCP is shown in Fig. 7. During the rest-
ing period, tOff, potential remained around E1. When a cathodic
current was applied during tC, the potential shifted suddenly to
a less-noble value, E2, and then moved gradually in the less-noble
direction reaching E3 at the end of tC. Increase of the electrodepo-
sition overpotential during tC corresponds to growth of a depletion
layer on the surface at each cycle. Introduction of tOff after tC pro-
vided recoveryofsurfaceconcentration ofAlions asconfirmed from
the fact that E2 restituted at each cycle. Characteristic parameters
of potential transient are summarized in Table 1. From this table, it
is clear that a small duty ratio (rC), i.e., the portion of polarization
time in one whole cycle, rC = tC/(tC + tOff), provides a small poten-
tial difference, ꢂE1 = E1 − E2 and ꢂE2 = E2 − E3, i.e., suppression of
-8
-0.50
-6
-0.40
-0.37
-4
-2
0
-0.35
-0.32
-0.30
-0.25
-0.27
V vs. Al/AlCl3
8 10
0
2
4
6
Time /s
Fig. 4. Current–time transient of Pt-WE during potentiostatic cathodic polarization
in the [EMIm]Cl/AlCl3 (0.5 mol%) bath at various potentials.

J. Tang, K. Azumi / Electrochimica Acta 56 (2011) 1130–1137
1133
Fig. 6. SEM images of the Al surface electrodeposited from the [EMIm]Cl/AlCl3 (0.5 mol%) bath using galvanostatic polarization at (a) −3.0 mA cm−2, (b) −4.5 mA cm−2, (c)
−6.0 mA cm−2, and (d) −8.0 mA cm−2. (e) Cross-sectional SEM image of (d).
tOff
-1.2
0
0
E1
-1
0
.3
-1.4
0
-1.5
0
E2
-1.6
0
-16
tC
-1.7
0
E3
Time /s
2438.0 2438.5 2439.0 2439.5 2440.0
Time /s
Fig. 7. Current pulse waveform for Al electrodeposition and potential response during MCP polarization.

1134
J. Tang, K. Azumi / Electrochimica Acta 56 (2011) 1130–1137
Table 1
Parameters of current pulse polarization as duty ratio (rC), frequency (f), cathodic current density (IC), anodic current density (IA), potential response (E1, E2, ꢂE1, ꢂE2),
capacitance (CF) and roughness factor (ꢃ) of Al deposit films.
Sample
rC
f/Hz
IC/mA cm−2
IA/mA cm−2
E1/V
E2/V
ꢂE1/V
ꢂE2/V
CF/mF m−2
ꢃ
MCP-a
MCP-b
MCP-c
MCP-d
MCP-e
BCP-a
BCP-b
BCP-c
Al foil
0.5
3.3
2.5
1.54
1.33
1
2
2
2
–
16
16
16
16
16
16
16
16
–
0
0
0
0
−0.27
−0.25
−0.24
−0.24
−0.26
−0.24
−0.23
−0.21
–
−0.64
−0.60
−0.56
−0.61
−0.59
−0.58
−0.56
−0.53
–
0.37
0.35
0.32
0.37
0.33
0.34
0.33
0.32
–
0.08
0.05
0.04
0.09
0.10
0.02
0.02
0.01
–
22.37
21.04
17.86
19.13
20.72
22.90
20.80
15.69
9.81
2.28
2.14
1.82
1.95
2.11
2.33
2.12
1.59
1
0.375
0.23
0.33
0.5
0.5
0.5
0
0.3
0.5
1.0
–
0.5
–
Fig. 8. SEM images of the Al surface electrodeposited from the [EMIm]Cl/AlCl3 (0.5 mol%) bath using the MCP method at IC = −16.0 mA cm−2. (a) rC = 0.5, f = 3.3 Hz; (b) rC = 0.375,
f = 2.5 Hz; (c) rC = 0.23, f = 1.54 Hz; (d) rC = 0.33, f = 1.33 Hz; (e) rC = 0.5, f = 1 Hz. (f) Cross-sectional SEM image of (c).

J. Tang, K. Azumi / Electrochimica Acta 56 (2011) 1130–1137
1135
tA
E1
-1.2
0
0
-1.3
0
-1.4
0
-1.5
0
E2
-1.6
0
E3
-16
-1.7
0
tC
Time /s
2438.0
2438.4
2438.8
Time /s
Fig. 9. Current pulse waveform for Al electrodeposition and potential response during BCP polarization.
Al ion depletion and thus efficient Al deposition to obtain better
deposits.
SEM images of the Al surface electrodeposited on Pt from the
3.5. Bipolar current pulse polarization (BCP)
An example of the current pulse waveform and corresponding
potential response in the BCP polarization is shown in Fig. 9. In this
procedure, improvement in uniform deposition is induced by par-
tial dissolution of deposits during the slightly anodic polarization
pulse for tA after interruption of cathodic deposition pulse for tC.
The potential transient was similar to that for MCP shown in Fig. 7.
The potential E1 was more noble than that for MCP and moved
gradually in a noble direction during tA. The less-noble value of E1
with about −0.2 V than that of Al-RE during anodic dissolution of
Al deposits indicated that Al deposits were more active than the Al
plate used as an RE. When polarization was switched from anodic to
cathodic, the potential difference, ꢂE2 = E2 − E3, was much smaller
than that of MCP and remained almost constant during tC. This indi-
cates that the overpotential of Al deposition reaction was lessened
in the case of the BCP method, probably due to alteration of the Al
surface after slight anodic dissolution and good supply of Al ions
[EMIm]Cl/AlCl3 bath using MCP with different values of rC and fre-
quency, f, are shown in Fig. 8. The cathodic current for Al deposition
was kept constant at −16 mA cm−2 while rC and f were changed. In
all cases, a dense and compact microstructure is clearly observed
compared with the case of GP shown in Fig. 6. Surface morphology
of MCP deposits changed with changes in rC and f. For example,
the surface became smooth with increase in tOff at fixed tC and
with decrease in tC at fixed tOff. Improvement of Al deposits with
decrease in rC can be explained by the recovery of Al ion concen-
tration on the surface during tOff and thus the suppression of side
reactions, resulting in a better deposition condition in the next tC
duration. The smoothest deposit was obtained in the case of Fig. 8(c)
in which smallest rC was applied. The cross-sectional SEM image of
the specimen in Fig. 8(f) shows a compact and dense structure of
Al deposits with a thickness of about 8 ␮m.
Fig. 10. SEM images of the Al surface electrodeposited from the [EMIm]Cl/AlCl3 (0.5 mol%) bath using the BCP method with duty ratio rC = 0.5, f = 2 Hz, cathodic current
density IC = −16 mA cm−2, and anodic current density IA = (a) 0.3, (b) 0.5, and (c) 1.0 mA cm−2. (d) Cross-sectional SEM image of (c).

1136
J. Tang, K. Azumi / Electrochimica Acta 56 (2011) 1130–1137
to the surface provided by additional Al dissolution during tA to
the supply of Al ions via diffusion layer. As a result, the deple-
tion effect of Al ions on the surface during tC was weakened and
thus better electrodeposition conditions such as lower overpoten-
tial for electrodeposition and small contribution of side reactions
were obtained.
Morphology of the Al surface electrodeposited using
the BCP method with the conditions of IC = −16 mA cm−2
,
rC = tC/(tC + tA) = 0.5, f = 2 Hz and various values of anodic cur-
rent density, IA, during anodic polarization time, tA, is shown in
Fig. 10. The deposit surface became smooth with increasing IA. This
was probably can be confirmed from the value of roughness factor,
ꢃ, shown in Table 1 obtained by EIS measurement as described
below. The value of ꢃ is similar to the value measured for deposits
obtained by the MCP method shown in Table 1. However, adhesion
strength of the BCP films was considerably improved compared
with that of the MCP films. For example, the MCP film was easily
detached from the substrate in the tape test, while the BCP film
was not. The reason for the better adhesion strength of BPC films
is that partial dissolution of the deposit surface during the anodic
pulse period makes a fresher surface for the following deposition
under the condition of presence of impurities in the ionic liquid
bath such as a small amount of oxygen or water. Fig. 10(d) shows
a cross-sectional image of aluminum on the Pt substrate. The
uniform and dense Al layer adhered properly on the substrate and
exhibited excellent deposition quality. As shown in this figure, the
thickness of the deposited layer was approximately 9 ␮m, which
is very close to the theoretical value calculated from Faraday’s
electrolysis law. This result suggested that the thickness of the
deposit could be easily controlled by adjusting the total charge
passed. Therefore, it is proposed that the BCP method is better for
obtaining Al deposits with a smooth surface and better adhesivity.
Fig. 11. Bode diagram of Al deposits obtained from the [EMIm]Cl/AlCl3 (0.5 mol%)
bath using the MCP method with the conditions of rC = 0.5, f = 3.3 Hz.
were fitted by the calculation curve using an equivalent circuit, and
the capacitance value of the deposits was compared with the value
measured for the Al plate [16]. EIS results of all samples show sim-
ilar appearance as can be seen in Fig. 9. The equivalent circuit used
for simulation of impedance response was composed of an elec-
trolyte resistance, RS, a charge transfer resistance, RF, and a constant
phase element, CPE, which is a capacitor with compensation of non-
uniformity of the electrode surface. The impedance of a CPE, ZCPE
,
is given by (Fig. 11):
1
Z =
,
A(jω)Á
where j is an imaginary unit, ω is an angular frequency of the
applied ac signal, and A and Á are variables. Since the value of Á was
0.92–0.96, being close to 1, a CPE can be considered to be a capaci-
tor. The capacitance of the air-formed oxide film on the deposited
Al, CF, is given by the following equation:
3.6. Comparison of the three electrodeposition methods
In this subsection, Al deposits formed by the GP, MCP and BCP
methods are compared. As seen in SEM images of Al deposits, sur-
face morphology depends on electrodeposition condition. EIS was
thus applied to quantitatively evaluate the surface roughness of
Al deposits. The Al surface is always covered with an air-formed
oxide film in the atmosphere. Assuming that the thicknesses of
the oxide films formed on the Al deposits and Al plate are iden-
tical, the roughness factor of Al deposits can be evaluated from
the capacitance value. In this evaluation, impedance spectra of WE
εεf ꢃA
CF
=
,
d
where ε is the permittivity of vacuum, εf is permittivity of an oxide
layer, ꢃ is the roughness factor of the deposited film and A is the
geometric surface area. The value of ꢃ was assumed to be 1 for the
Fig. 12. Schematic representation of growth of Al electrodeposited from the [EMIm]Cl/AlCl3 (0.5 mol%) bath using the GP, MCP and BCP methods.

J. Tang, K. Azumi / Electrochimica Acta 56 (2011) 1130–1137
1137
flat Al foil. The total capacitance, CT, can be expressed as follows:
(1) The deposition process was controlled by the initial nucle-
ation stage and following deposit-growth stage. The former was
affected by the polarization method and current density, and
the latter was also affected by the current density via formation
of a diffusion layer of Al ions.
1
CT
1
CF
1
CH
=
+
,
where CH is capacitance of the Helmholtz layer. In the present
study, CH (a typical value of CH for most metals is in the range
CH ≈ 10–100 ␮F cm−2) can be neglected for rough estimation of ꢃ
because it is considerably larger than CF (a typical value of CF is ca.
3.6 ␮F cm−2). The value of ꢃ was determined from the ratio of the
capacitance of deposits to that on the flat Al foil as shown in Table 1.
Decrease of ꢃ with decreasing rC confirms that a smoother surface
was obtained at smaller rC, as can be seen in the SEM images shown
in Fig. 8.
(2) Surface morphology and particle size of Al deposits can be con-
trolled by the polarization conditions. For example, particle
size becomes small with increase in current density in the GP
method. Duty ratio and frequency are also adjusted to obtain
dense deposits composed of small Al particles in the current
pulse polarization method. Generally, the current pulse polar-
ization method provides dense deposits with a smooth surface.
(3) The initial nucleation process and following growth process
of Al deposits can be optimized by adjusting the wave form
of the current pulse. For example, adhesion strength of Al
deposits was improved by adding slight anodic dissolution in
the current pulse cycle. The best Al deposits with a smooth
surface and good adhesion were obtained in the conditions of
IC = −16.0 mA cm−2, IA = 1.0 mA cm−2, rC = 0.5, and f = 2 Hz in the
BCP method.
Electrodeposition processes in the three polarization methods
are schematically presented in Fig. 12. In the GP method, Al particles
grow continuously after the initial nucleation process, which occurs
only on limited sites due to low overpotential at low current density
as shown in Fig. 6. Although the density of Al deposits is improved
at relatively high current density, adhesivity of Al deposits remains
poor. In the MCP method, a large current density during tC enables
initial nucleation of Al deposits with high density on the surface.
In the following stage, recovery of Al ion concentration during tOff
weakens the growth of the depletion layer. This provides further
Al deposition on various sites of the surface in the successive plat-
ing cycle, tC, resulting in dense and smooth deposits. In the BCP
method, however, not only density and smoothness but also adhe-
sion strength of Al deposits are improved. These improvements are
probably due to the following factors: (1) good supply of Al ions
to the surface, which enables uniform deposition, (2) refreshing
of the Al deposit surfaces during slight anodic dissolution, which
removes active sites for preferential Al deposition and thus uni-
formizes deposition conditions of the surface to enable deposition
reaction on many sites, resulting in lower overpotential for deposi-
tion reaction, and (3) suppression of side reactions due to lowering
of overpotential such as adsorption of organic compounds, which
may disturb crystal growth and/or formation of bonding between
deposits and pre-deposits. These possibility need to be clarified by
further investigation in the future.
Acknowledgments
This work was supported by the Global COE Program (project no.
B01: Catalysis as the Basis for Innovation in Materials Science) from
the Ministry of Education, Culture, Sports, Science and Technology,
Japan, and also supported in part by the Light Metal Educational
Foundation.
References
[1] O. Sarikaya, S. Anik, S. Aslanlar, S.C. Okumus, E. Celik, Mater. Des. 28 (2007)
2443.
[2] J.H. Yun, B.Y. Kim, S.W. Rhee, Thin Solid Films 312 (1998) 259.
[3] N. Koura, H. Nagase, A. Sato, S. Kumakura, K. Takeuchi, K. Ui, T. Tsuda, C.K. Loong,
J. Electrochem. Soc. 155 (2008) D155.
[4] W.J. Cheng, C.J. Wang, Surf. Coat. Technol. 204 (2009) 824.
[5] J.K. Chang, S.Y. Chen, W.T. Tsai, M.J. Deng, I.W. Sun, Electrochem. Conmmun. 9
(2007) 1602.
[6] T. Jiang, M.J. Chollier Brym, G. Dubé, A. Lasia, G.M. Brisard, Surf. Coat. Technol.
201 (2006) 1.
[7] S.J. Pan, W.T. Tsaia, J.K. Chang, I.W. Sun, Electrochim. Acta 55 (2010) 2158.
[8] T. Tsuda, S. Arimoto, S. Kuwabata, C.L. Hussey, J. Electrochem. Soc. 155 (2008)
D256.
4. Conclusions
[9] T. Tsuda, C.L. Hussey, Thin Solid Films 516 (2008) 6220.
[10] C.C. Yang, Mater. Chem. Phys. 37 (1994) 355.
In this study, electrodeposition of Al from Lewis acidic 1-ethyl-
3-methyl-imidazolium chloride EMIC/AlCl3 (0.5 mol%) ionic liquid
onto a Pt substrate at room temperature was investigated using
potentiostatic polarization (PP), galvanostatic polarization (GP),
monopolar current pulse polarization (MCP) and bipolar current
pulse polarization (BCP). The following conclusions could be drawn
from the experimental results:
[11] B. Li, Y. Chen, L.G. Yan, J.D. Ma, Electrochemistry 78 (2010) 523.
[12] G. Manoli, Y. Chryssoulakis, J.C. Poignet, Plat. Surf. Finish. 78 (1991) 64.
[13] J. Vaughan, D. Dreisinger, J. Electrochem. Soc. 155 (2008) D68.
[14] H.H. Elsentriecy, K. Azumi, H. Konno, Electrochim. Acta 53 (2008) 4267.
[15] P.K. Lai, M. Skyllas-Kazacos, J. Electroanal. Chem. 248 (1988) 431.
[16] M.T. Tanvir, Y. Aoki, H. Habazaki, Thin Solid Films 517 (2009) 6711.