Electroless plating of aluminum from a room-temperature ionic liquid electrolyte

Article abstract of DOI:10.1149/1.2817880

Because aluminum is a less-noble metal which has the standard electrode potential of -1.676 V vs normal hydrogen electrode, it is impossible to obtain the electrodeposition of aluminum from an aqueous solution. No one has reported an electroless plating method of aluminum. We succeeded in demonstrating the electroless plating of aluminum from a room-temperature ionic liquid (RTIL). It was found from measurements of inductively coupled plasma, X-ray diffraction, scanning electron microscopy (SEM), SEM-energy-dispersive X-ray analysis, and glow discharge optical emission spectroscopy that dense, smooth, and pure aluminum plating was obtained from the RTIL by the electroless plating method. Moreover, the reaction mechanism of the electroless plating of aluminum from the RTIL electrolyte was electrochemically analyzed.

Full text of DOI:10.1149/1.2817880

Journal of The Electrochemical Society, 155 ͑2͒ D155-D157 ͑2008͒
D155
0013-4651/2007/155͑2͒/D155/3/$23.00 © The Electrochemical Society
Electroless Plating of Aluminum from a Room-Temperature
Ionic Liquid Electrolyte
Nobuyuki Koura,a Hiroshi Nagase,a Atsushi Sato,a Shintaro Kumakura,a
,z
,d
c
Ken Takeuchi,b, Koichi Ui,a, Tetsuya Tsuda,a, and Chun K. Loong
*
*
*
aFaculty of Science and Technology, Tokyo University of Science, Noda, Chiba 278-8510, Japan
bFaculty of Industrial Science and Technology, Tokyo University of Science, Oshamambe,
Hokkaido 049-3514, Japan
cArgonne National Laboratory, Intense Pulsed Neutron Source, Argonne, Illinois 60439, USA
Because aluminum is a less-noble metal which has the standard electrode potential of −1.676 V vs normal hydrogen electrode, it
is impossible to obtain the electrodeposition of aluminum from an aqueous solution. No one has reported an electroless plating
method of aluminum. We succeeded in demonstrating the electroless plating of aluminum from a room-temperature ionic liquid
͑RTIL͒. It was found from measurements of inductively coupled plasma, X-ray diffraction, scanning electron microscopy ͑SEM͒,
SEM-energy-dispersive X-ray analysis, and glow discharge optical emission spectroscopy that dense, smooth, and pure aluminum
plating was obtained from the RTIL by the electroless plating method. Moreover, the reaction mechanism of the electroless plating
of aluminum from the RTIL electrolyte was electrochemically analyzed.
© 2007 The Electrochemical Society. ͓DOI: 10.1149/1.2817880͔ All rights reserved.
Manuscript submitted August 24, 2007; revised manuscript received October 18, 2007.
Available electronically December 19, 2007.
Recently, there has been much research and development of new
materials and their preparation techniques due to the rapid growth of
the electronics industry. The development of technology to produce
aluminum thin films and aluminum wiring at the lowest cost has
been highly sought after. Such technology has profound conse-
quences across many electronic-equipment industries, including
hard disks, compact disks, high-density memory devices, semicon-
ductor devices, electronic circuitry, etc.1,2 Because aluminum is a
less-noble metal, it is impossible to obtain electrodeposition in an
aqueous solution. Aluminum can be deposited from a room-
temperature ionic liquid ͑RTIL͒,3-6 but a practical technology for
depositing aluminum from even RTIL has not been established.
Moreover, no one has reported the electroless plating of aluminum.
We have developed a technique of electroless plating for aluminum
based on using the RTIL as the electrolyte. There are many uses of
aluminum thin films, including aluminum wiring as stated above,
but it is necessary to use large reactors utilizing methods such as
sputtering and glow plasma in order to deposit aluminum on an
insulating substrate such as silicon or glass. If the technique of elec-
troless plating of aluminum is established, it will be able to obtain
the thin and thick film coating on the substrates of insulating mate-
rial and of quite complicated structures without the use of electricity.
We verified that the dense aluminum layer is smoothly plated with-
out any impurity phase using inductively coupled plasma ͑ICP͒,
X-ray diffraction ͑XRD͒, scanning electron microscopy ͑SEM͒,
SEM-energy dispersive X-ray analysis ͑SEM-EDX͒, and glow dis-
charge optical emission spectroscopy ͑GD-OES͒ measurements. In
addition, we determined the reaction mechanism of the aluminum
electroless plating from an electrochemical point of view.
ished the substrate with Emery paper ͑no. 320͒ and removed grease
with methanol and acetone. Second, we sensitized it with acidic
SnCl2 solution ͑0.127 mol L−1, pH 0.74͒. Third, we activated the
surface of the substrate with acidic PdCl2 solution ͑1.691
ϫ 10−3 mol L−1, pH 0.77͒. The acidity of each solution was con-
trolled by hydrochloric acid. We added LiH, LiAlH4, or diisobutyl
aluminum hydride ͑DIBAH͒ as the reducing agent into the
66.7 mol % AICl3–33.3 mol % EMIC melt. At the same time, we
immersed the substrate in the electrolyte, and then the electroless
plating of aluminum began. The temperature of the electrolyte was
kept constant with a thermocontroller. All of the procedures of
66.7 mol % AICl3–33.3 mol % EMIC melt preparation and the
electroless plating were done in a glove box filled with dry argon
gas.
A three-electrode cell consisting of a working electrode ͑tung-
sten͒, a counter electrode ͑aluminum͒, and a reference electrode
͑aluminum͒ for the linear sweep voltammetry measurements was
employed. An Al wire ͑99.99%, ␾: 2 mm͒ immersed in a
66.7 mol % AICl3–33.3 mol % EMIC melt in a Pyrex tube ͑␾:
10 mm͒ separated by a G4 g1ass frit was used as the reference
electrode. All of the experiments have been done at room tempera-
ture unless otherwise noted.
Results and Discussion
The plated specimen was qualitatively identified by XRD
͑Rigaku, RAD-C, Cu K␣͒ and its composition was determined by
ICP ͑Shimadzu, ICPS-7500͒. The observed diffraction peaks of the
plated film agreed with those of aluminum in the literature,8 as
shown in Fig. 1. A calculated lattice constant derived from observed
peaks was 4.046 Å in good agreement with the reported value in the
literature.7 These diffractions indicated that the deposited film from
the electrolyte with LiH was aluminum. The data from the ICP
measurement also showed that the purity of the deposited aluminum
was more than 99%. Subsequently, we performed surface analysis
using a SEM ͑Hitachi, SEM S-2600N͒ and obtained the mapping of
the elements on the surface of the film. Figures 2a and c showed that
a crystalline material was deposited on the surface of the substrate
and it proved that electroless plating took place. The elemental map-
ping result by SEM-EDX ͑EDAX, Genesis2000͒ ͑Fig. 2b͒ also in-
dicated that the plating film consisted of aluminum only, and the
aluminum was homogeneously distributed on the surface of the film.
The depth profile of the film was analyzed with a GD-OES ͑JY-
5000RF, Horiba–Jobin Yvon͒ ͑Fig. 3͒. It was confirmed that all of
the plated film consisted of ordinary aluminum only, because only
aluminum was detected from the surface of the boundary between
Experimental
The AlCl3-1-ethyl-3-methylimidazolium chloride ͑EMIC͒ ionic
liquid, used as the electrolyte, was synthesized with the following
procedure. First, AlCl3 and EMIC were weighed with the molar
ratio 2:1 and mixed by stirring. Subsequently, we treated this ionic
liquid, which was completely melted, by a substitution method in
which aluminum wire was immersed in the liquid for more than 1
week,7 and we obtained a colorless and transparent ionic liquid. We
pretreated the substrates with the following procedure. First, we pol-
*
Electrochemical Society Active Member.
d Present address: Department of Frontier Materials and Functional Engineering,
Graduate School of Engineering, Iwate University, Morioka, Iwate 020-8551, Ja-
pan.
z E-mail: ken@rs.kagu.tus.ac.jp
Downloaded on 2014-04-11 to IP 134.208.103.160 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D156
Journal of The Electrochemical Society, 155 ͑2͒ D155-D157 ͑2008͒
Figure 1. XRD measurement for the electroless plating of aluminum from
the 66.7 mol % AlCl3–33.3 mol % EMIC electrolyte with 1.0 mol L−1 LiH.
Substrate: glass, temp: 35°C, plating time: 2 h.
Cu substrate and the deposit. No impurity phases such as C, Li, and
H, which were considered to be likely contaminates from the elec-
trolyte, were detected by GD-OES.
Figure 4. Effects of factors on the aluminum electroless plating: ͑a͒ depen-
dence of plating time and electrolyte temperature on the amount of the de-
posit and ͑b-e͒ SEM images of surface morphology of aluminum plating
from the 66.7 mol % AlCl3–33.3 mol % EMIC electrolyte with 1.0 mol L−1
LiH. Substrate: glass, plating temperature and time: ͑b͒ 35°C and 15 min, ͑c͒
35°C and 2 h, ͑d͒ 25°C and 15 min, and ͑e͒ 45°C and 15 min. ͑f͒ Depen-
dence of plating time and the amount of reducing agent on the amount of
deposit. ͑g-i͒ SEM images of surface morphology of aluminum plating from
the 66.7 mol % AlCl3–33.3 mol % EMIC electrolyte with various reducing
agents. Substrate: glass, temp: 35°C, reducing agent and plating time: ͑g͒
0.5 mol L−1 LiH and 15 min, ͑h͒ 2.0 mol L−1 LiH and 15 min, and ͑i͒
0.2 mol L−1 DIBAH and 2 h.
There are several factors which determine the plating rate and
surface morphology for electroless plating. Some of the factors are,
for example, the temperature of electrolyte, the concentration of the
reducing agent, and the concentration of the metal ion in the elec-
trolyte. We have investigated the effects of those factors on alumi-
num electroless plating ͑Fig. 4͒. First, we examined the dependence
of plating time on both the amount of deposit and the surface mor-
phology. The amount of deposit increased with immersion time
͑plating time͒ of the substrate at any temperature ͑Fig. 4a͒. In addi-
tion, the amount of deposit increased during the shorter period with
higher temperatures, at the beginning of the plating. It suggested that
the reaction rate increased with temperature. The maximum amount
of the deposit was about 0.40 mg cm−2 at both 25 and 35°C, but the
deposit was about 0.25 mg cm−2 at 45°C. This suggested that the
electrolyte decomposed because the temperature increased and the
reducing reaction rate of LiH became extremely high before the
maximum amount of deposit was achieved.
Figure 2. ͑a and c͒ SEM and ͑b͒ SEM-EDX measurements for the surface of
aluminum plating at the same observed position. The plating electrolyte was
the 66.7 mol % AICl3–33.3 mol % EMIC electrolyte with 1.0 mol L−1 LiH.
Substrate: glass, temp: 35°C, plating time: 2 h.
Next, we investigated the surface morphology of the aluminum
plating. The particle size at the surface became larger with longer
plating time, as shown in Fig. 4b and c. The particle size became
larger with higher temperatures according to the comparisons among
Fig. 4b, d, and e.
In the same way, we studied the dependence of the amount of
reducing agent on both the amount of deposit and the morphology of
the aluminum plating. The amount of deposited aluminum increased
with increasing the reducing agent ͑Fig. 4f͒. The amount of depos-
ited aluminum increased rapidly even at short plating times in these
cases. This showed that the added reducing agent did not condense
in the electrolyte. The reaction rate proportionally increased with the
amount of reducing agent, because LiH was well dispersed in the
electrolyte. The aluminum film deposited by electroless plating at
35°C for 15 min from the 0.5 mol L−1 LiH-added electrolyte was so
thin that we could recognize the scratched mark on the surface of the
Figure 3. GD-OES depth profile of the electroless plating of aluminum from
the 66.7 mol % AlCl3–33.3 mol % EMIC electrolyte with 1.0 mol L−1 LiH.
Substrate: Cu, temp: 35°C, plating time: 2 h. Sputtering conditions pressure
900 Pa, power 45 W, anode size 4 mm. Li, C, and H were not detected.
Downloaded on 2014-04-11 to IP 134.208.103.160 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 155 ͑2͒ D155-D157 ͑2008͒
D157
Figure 6. Polarization curves for ͑a͒ cathodic reaction of Al2Cl− from the
67 mol % AlCl3–33 mol % EMIC electrolyte and ͑b͒ anodic react7ion of LiH
from the 45.0 mol % AlCl3–55.0 mol % EMIC electrolyte with 1.0 mol L−1
LiH.
Figure 5. Raman measurements for the plating electrolyte of the 66.7 mol %
AlCl3–33.3 mol % EMIC melt with 1.0 mol L−1 LiH ͑a͒ before and ͑b͒ after
the electroless plating process.
potential theory was 0.0101 mg cm−2 min−1. The electroless-plating
rate estimated from the weight of aluminum deposit was
0.0099 mg cm−2 min−1. It was almost the same as the value calcu-
lated from the mixed-potential theory. Consequently, we determined
that the mixed-potential theory was applicable for electroless alumi-
num plating. It was concluded that the reaction mechanism was
described by the combination of electrochemical reactions indicated
by Eq. 1 and 2.
substrate. However, the aluminum film deposited from the
2.0 mol L−1 LiH-added electrolyte was much thicker ͑Fig. 4g and
h͒. Furthermore, the same tendency was observed when LiAlH4 was
used instead of LiH. This indicated that we also obtained electroless
plating of aluminum with LiAlH4.
We also found a liquid reducing agent, DIBAH. It was consid-
ered that the liquid agent of DIBAH could be much better dispersed
in the electrolyte than either of the solid agents, LiH or LiAlH4.
Certainly, a smoother plating surface was obtained in the case of
DIBAH compared with LiH, as shown in Fig. 4g-i.
Subsequently, we attempted to determine the mechanism of the
aluminum electroless plating. At first, we analyzed the change of the
ionic species involved with the aluminum plating in the electrolyte
by using a laser Raman spectrophotometer ͑Jasco, NRS-2000͒. Ac-
cording to the Raman spectrum ͑Fig. 5b͒ of the electrolyte after
electroless plating, the peak corresponding to Al2Cl−7 became small,
but the peak corresponding to AlCl−4 grew, despite that only a small
peak was observed before the electroless plating ͑Fig. 5a͒. This sug-
gested that Al2Cl−7, which is the ionic species for aluminum deposit
in the acidic AlCl3–EMIC melt,9 had the reaction shown below and
the aluminum was deposited from the electrolyte10,11
Conclusion
We succeeded in demonstrating the electroless plating of dense,
smooth, and pure aluminum from a RTIL, AlCl3–EMIC, with reduc-
ing agent of powder LiH or LiAlH4 or liquid DIBAH. XRD results
indicated that the deposited film was aluminum. The data from the
ICP measurements showed that the purity of the deposited alumi-
num was more than 99%. The elemental mapping results of SEM-
EDX indicated that the plating film consisted of aluminum only, and
the aluminum was homogeneously distributed on the surface of the
film. GD-OES results showed that the plated film consisted of alu-
minum only, because only aluminum was detected from the surface
of the boundary between Cu substrate and the deposit. No impurity
phases such as C, Li, and H, which were considered to be likely
contaminates from the electrolyte, were detected. In addition, we
analyzed the reaction mechanism of the aluminum electroless plat-
ing from an electrochemical point of view. Consequently, we deter-
mined that the mixed-potential theory was applicable for the elec-
troless aluminum plating. It was concluded that the reaction
mechanism was described by the combination of electrochemical
reactions indicated by Eq. 1 and 2.
4Al2Cl−7 + 3e− → AlCl−4 + Al
͓1͔
It was also considered that the reducing agent, LiH, had the
reaction indicated below in order to cause Reaction 1, because a H2
gas was evolved during the electroless plating procedure and an Li+
ion was detected in the electrolyte after the plating
Tokyo University of Science assisted in meeting the publication costs of
this article.
LiH → Li+ + 1/2H2 + e−
͓2͔
References
1. W. E. Frank, Microelectron. Eng., 33, 85 ͑1997͒.
2. M. Ronay, J. Electrochem. Soc., 148, 494 ͑2001͒.
Next, we attempted to analyze Reactions 1 and 2 using the
mixed-potential theory,12 which is essential to describing the elec-
troless plating reaction. We measured the cathodic polarization
curve for the acidic 66.7 mol % AICl3–33.3 mol % EMIC melt, in-
cluding Al2Cl−7 and the anodic polarization curve for the reducing
agents ͑LiH͒-added basic 45.0 mol % AICl3–55.0 mol % EMIC
3. R. T. Carlin, W. Crawford, and M. Bersch, J. Electrochem. Soc., 139, 2720 ͑1992͒.
4. S. Takahashi, K. Akimoto, and S. Isao, J. Surf. Finish. Soc. Jpn., 40, 548 ͑1989͒.
5. P. K. Lai and M. Skyllas-Kazacos, Electrochim. Acta, 32, 1443 ͑1987͒.
6. P. K. Lai and M. Skyllas-Kazacos, J. Electroanal. Chem. Interfacial Electrochem.,
248, 431 ͑1988͒.
7. S. Takahashi and N. Koura, J. Electroanal. Chem. Interfacial Electrochem., 188,
245 ͑1985͒.
8. Joint Committee for Powder Diffraction Standards, Powder Diffraction File No.
4-787, JCPDS International Center for Diffraction Data, Swarthmore, PA ͑1991͒.
9. S. Takahashi, L. A. Curtiss, D. Gosztola, N. Koura, and M.-L. Saboungi, Inorg.
Chem., 34, 2990 ͑1995͒.
melt without the Al2Cl−7 as shown in Fig. 6. The polarization curves
for cathodic reaction of Al2Cl− ͑Fig. 6a͒ and anodic reaction of LiH
͑Fig. 6b͒ were obtained. Both7the reaction rate ͑current density͒ and
the mixed potential were derived by the conventional calculating
method.11 The mixed potential was −0.130 V vs Al͑III͒/Al. This
value showed good agreement with the electrode potential,
−0.125 V vs Al͑III͒/Al, which was measured during the electroless
plating of aluminum. The current density at the mixed potential was
1.85 mA cm−2. The plating reaction rate derived from the mixed-
10. C. L. Hussey, T. B. Scheffler, J. S. Wilkes, and A. A. Fannin, Jr., J. Electrochem.
Soc., 133, 1389 ͑1986͒.
11. Y. Chryssoulakis, J.-C. Poignet, and G. Manoli, J. Appl. Electrochem., 17, 857
͑1987͒.
12. M. Paunovic, Plating, 55, 1161 ͑1968͒.
Downloaded on 2014-04-11 to IP 134.208.103.160 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).