Investigation of carbon-supported Pt nanocatalyst preparation by the polyol process for fuel cell applications

Article abstract of DOI:10.1016/j.electacta.2007.05.080

Parametric investigation of the polyol process for the preparation of carbon-supported Pt nanoparticles as catalysts for fuel cells was carried out. It was found that the concentration of glycolate anion, which is a function of pH, plays an important role in controlling Pt particle size and loading on carbon. It was observed that Pt loading decreased with increasing alkalinity of the solution. As evidenced by zeta potential measurement, this was mainly due to poor adsorption or repulsive forces between the metal colloids and the supports. In order to modify the conventional polyol process, the effect of the gas purging conditions on the characteristics of Pt/C was examined. By the optimization of the gas environment during the reaction, it was possible to obtain high loading of 39.5 wt% with a 2.8 nm size of Pt particle. From the single cell test, it was found that operating in ambient O₂ at 70 °C can deliver high performance of more than 0.6 V at 1.44 A cm^-2.

Full text of DOI:10.1016/j.electacta.2007.05.080

Electrochimica Acta 52 (2007) 7278–7285
Investigation of carbon-supported Pt nanocatalyst preparation
by the polyol process for fuel cell applications
Hyung-Suk Oh, Jong-Gil Oh, Youn-Gi Hong, Hansung Kim∗
Department of Chemical Engineering, Yonsei University, 134 Shinchon-Dong, Seodaemun-gu, 120-749 Seoul, Republic of Korea
Received 10 March 2007; received in revised form 18 May 2007; accepted 29 May 2007
Available online 17 June 2007
Abstract
Parametric investigation of the polyol process for the preparation of carbon-supported Pt nanoparticles as catalysts for fuel cells was carried out.
It was found that the concentration of glycolate anion, which is a function of pH, plays an important role in controlling Pt particle size and loading
on carbon. It was observed that Pt loading decreased with increasing alkalinity of the solution. As evidenced by zeta potential measurement, this
was mainly due to poor adsorption or repulsive forces between the metal colloids and the supports. In order to modify the conventional polyol
process, the effect of the gas purging conditions on the characteristics of Pt/C was examined. By the optimization of the gas environment during the
reaction, it was possible to obtain high loading of 39.5 wt% with a 2.8 nm size of Pt particle. From the single cell test, it was found that operating
in ambient O2 at 70 ◦C can deliver high performance of more than 0.6 V at 1.44 A cm−2
© 2007 Elsevier Ltd. All rights reserved.
.
Keywords: Polyol process; Ethylene glycol; Zeta potential; PEM fuel cells
1. Introduction
al. developed organoaluminum-stabilized colloids with a parti-
cle size smaller than 2 nm at room temperature [10]. Organic
stabilizers such as polyvinyl pirrolidone (PVP) and the sur-
factant dodecyldimethyl (3-sulfo-propyl) ammonium hydroxide
(SB12) are widely used in the preparation of metal colloids
[2,11,12]. The intrinsic problem underlying this process is that
the stabilizing organic material remains on the surface of metal
colloids. Thisshouldberemovedpriortotheapplicationofmetal
particles for electrocatalysis. Removal of the organic material is
important as it hinders the access of fuel to the catalyst sites. In
general, the removal of stabilizer involves heat treatment. Conse-
quently, due to the sintering effect, the phase separation and the
distribution of metal particles are affected, resulting in lowered
catalytic performance.
In this respect, preparation via the polyol process is preferred
due to several advantages. The polyol process is a technique in
which a polyol such as ethylene glycol is used as both solvent
and reducing agent. The polyol method has been used for the
preparation of nanometal powder [13,14] and nanowires [15,16].
A unique property of the polyol process is that it does not require
any type of polymer stabilizer. In the polyol process using ethy-
lene glycol, metal ions are reduced to form a metal colloid by
receiving the electrons from the oxidation of ethylene glycol
to glycolic acid. Glycolic acid is present in its deprotonated
There is tremendous interest in the preparation of carbon-
supported electro-catalysts for fuel cell applications [1–4]. It is
well known that the performance of catalysts can be improved
by achieving nanosized particles, uniform distribution and high
loading of catalysts over large surface area carbons [5–7]. Con-
ventional preparation techniques used for the preparation of
supported catalysts are based on the wet impregnation followed
by reduction in a hydrogen atmosphere at high temperatures or
the chemical reduction of the metal precursors using reducing
agents. However, these methods do not provide adequate control
of particle size and distribution. Many studies have shown the
difficulty of high metal loadings without a significant increase
in the particle size [8,9].
Extensive investigations have been carried out to develop
alternate routes for preparing supported Pt catalysts by the col-
loidal method using diverse stabilizing agents. In those attempts,
a stabilizing agent was used to prevent the aggregation of metal
particles during the nucleation and growth steps. Boennemann et
∗
Corresponding author. Tel.: +82 2 2123 5753; fax: +82 2 312 6401.
E-mail address: elchem@yonsei.ac.kr (H. Kim).
0013-4686/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2007.05.080

H.-S. Oh et al. / Electrochimica Acta 52 (2007) 7278–7285
7279
form as glycolate anion in alkaline solution. It is believed that
the glycolate anion acts as a stabilizer by adsorbing the metal
colloids [17]. Furthermore, removal of these organics on the
metal surface by heat treatment below 160 ◦C has been reported,
which is low enough to avoid the deleterious effects associated
with heat treatment. However, no information is available on
metal loading as a function of solution pH and different gas
environment.
In the present work, a modified polyol process is utilized for
nanosized Pt/C formation in different pH and gas environments.
The aim of this study is to seek quantitative correlations between
solution pH and metal loading on carbon. The effect of the gas
environment during preparation steps on the variation of particle
size and metal loading is examined.
obtained using a Cu K␣ source operated at 40 keV at a scan rate
of 0.033 s−1
.
2.3. Membrane electrode assembly (MEA) fabrication and
cell operation
The cathode electrodes were prepared by ultrasonically
blending a mixture of Pt/C catalyst, Nafion solution (5 wt%
from Aldrich) and isopropyl alcohol. The catalyst solution was
then sprayed onto the surface of a single sided uncatalyzed gas
diffusion layer obtained from E-TEK. The total loading of Pt
was fixed at 0.4 mg cm−2 and the geometric active area of all
MEAs was 5 cm2. A commercial E-TEK electrode (30 wt%
Pt/C, 0.5 mg cm−2) was used as the anode electrode in order
to eliminate experimental errors that may originate from elec-
trode preparation steps. Prior to the preparation of the MEA,
a total of 0.8 mg cm−2 of Nafion solution was applied to the
E-TEK anode electrode by means of the spray technique. Both
anode and cathode electrodes with pretreated Nafion 112 mem-
brane were hot-pressed at 140 ◦C for 3 min at a pressure of
3 tonnes to form MEA. The performance of the single cell was
evaluated by measuring the current density versus cell voltage
using a commercial test system (Wonatech). The reaction gases
were supplied through a humidifier and a mass flow controller
from hydrogen and oxygen tanks. The cell was operated under
ambient pressure.
2. Experimental
2.1. Preparation of Pt/C catalyst
A measured amount of PtCl4 and NaOH were dissolved in
25 ml of ethylene glycol under vigorous stirring for 30 min.
NaOH was introduced to adjust pH of solution. Since the pH
is one of the crucial operating parameters in the polyol process
it was precisely controlled and recorded at every step. After
recording the initial pH of solution, the appropriate amount of
carbon black (Ketjen black 300J) was added to solution to pro-
duce 40 wt% of Pt/C. The resulting suspension was stirred for 1 h
at room temperature followed by heating under reflux at 160 ◦C
for 3 h. The solution was allowed to cool down to room temper-
ature and left for 12 h with continuous stirring. The pH of the
solution was measured again and accepted as the final pH. The
Pt/C particles in the solution were then filtrated and thoroughly
washed with water. This carbon-supported Pt catalyst was dried
in air for 1 h at 160 ◦C and a mortar was used to homogeneously
grind the Pt/C catalyst material to powder. During each step of
experiment, different gases (N2, air and O2) were supplied to
create different atmospheres.
3. Results and discussion
Several reaction mechanisms of the polyol process are pro-
posed in the literature. By using ethylene glycol in the presence
of PVP, metallic particles are produced from the following reac-
tions, as reported by Fievet et al. [18].
CH2OH–CH2OH → CH3CHO + H2O
(1)
2nCH3OH + 2Mn+ → 2M + 2nH+ + nCH3COCOCH3
2.2. Characterizations
(2)
Cyclic voltammetry was used to determine the effective sur-
face area of Pt catalysts and to calculate the particle size. The
experiment was performed in 0.5 M H2SO4 at 25 ◦C saturated
with nitrogen using a conventional three-electrode cell. A glassy
carbon electrode with a thin film of prepared Pt/C catalyst was
used as a working electrode. Platinum wire was used as a counter
electrode and a standard Hg/HgSO4 electrode served as a ref-
erence electrode. The size and distribution of the Pt particles
prepared by the polyol process was determined by using high
resolution transmission electron microscopy (HR-TEM, JEM-
30100 model). ICP-AES (Perkin-Elmer 4300DV) analysis was
carried out to estimate the Pt loading over the carbon support. In
order to understand the nature of the surface charge and explain
the stability of the colloidal solution, the zeta potential (Malvern
Instruments ZEN 3600) was measured for carbon black and the
Pt particles individually as a function of solution pH. Powder
X-ray diffraction scans (XRD) patterns of the catalysts were
PVP acts as a capping reagent. However, there is a limitation
on applying this mechanism in cases where PVP is not used. As
shown by Bock et al. in the preparation of PtRu nanocatalysts
[17], the reaction mechanism of the colloidal particle formation
involves the interaction of –OH groups of ethylene glycol with
Pt-ion sites resulting in the oxidation of the alcohol groups to
aldehydes. These aldehydes are not very stable and undergo fur-
ther oxidation to form glycolic acid and oxalic acid, respectively.
These two carboxylic acids may again be oxidized to CO2 or
carbonate in alkaline media. The electrons donated by oxidation
reactions result in the reduction of the Pt metal ions. This was
supported by the fact that both oxalic and glycolic acids were
detected in the HPLC analysis of the synthesis solutions. The
quantitative analysis revealed that glycolic acid is the dominat-
ing product in the resulting solution. The dissociation constant
of glycolic acid is known to be 1.48 × 10−4 mol L−1 at 25 ◦C
[19]. This implies that glycolic acid is present in its deproto-

7280
H.-S. Oh et al. / Electrochimica Acta 52 (2007) 7278–7285
nificant above 0.1 M of NaOH, which corresponds to pH 6.03.
This result is in good agreement with those presented in Fig. 1.
In the absence of glycolate ions where pH is 1.45, the Pt par-
ticle size is fairly large (∼8.5 nm). Upon adding NaOH to the
solution, the average Pt particle size decreases to 3.8 nm. This
decrease in particle size is therefore attributed to the presence
of glycolate ions in the solution resulting from the change in
solution pH from 1.45 to 3.82. The stabilizing action of the gly-
colate ions helps in controlling the particle size. Upon increasing
the concentration of glycolate ions in the synthesis solution by
increasing the solution pH to 4.33 and 6.03, the Pt particle size
is remarkably reduced down to 3.1 and 2.5 nm, respectively, as
shown in Fig. 2(c and d). Upon further increase of the solution
pH to 8.50 and 9.39, a marginal effect on the particle size was
found as shown in Fig. 2(e and f). This is attributed to the fact
that the concentration of glycolate anion is constant in this pH
range. The particle size of Pt with different NaOH concentration
was also confirmed with XRD patterns as shown in Fig. 3. The
mean Pt particle sizes were calculated from Scherrer’s formula
based on Pt (1 1 1) peak and listed in Table 1. With increasing
NaOH concentration, the diffraction peaks of Pt become broad
which indicates a decrease in particle size. The particle size of
Pt was remained constant after pH of 6 which was also observed
from TEM analysis. These results strongly support the role of
glycolate anion as a stabilizing agent.
Although a lot of work has been reported on the size control
mechanism in the polyol process, few focused on the influence
of metal loading on carbon as a function of the pH and synthesis
conditions. In order to determine the Pt loading over carbon
supports prepared by the polyol process, ICP-AES analysis was
performed for the samples, as shown in Fig. 4. For the case
where NaOH is not added to the solution, Pt loading on carbon
support was measured to be 38 wt%, indicating that most of
the Pt salts were reduced and loaded on carbon without a loss.
Upon increasing the pH of the solution, the Pt loading decreased
continuously and a rapid fall in Pt loading was observed at pH
higher than 8.5. The presence of NaOH in the solution is helpful
for controlling the particle size but the metal loading is seriously
reduced. Pt is a precious metal and the production yield is really
a catalyst cost benchmark.
In an attempt to explain this result, zeta potentials of Pt parti-
cles and carbons were measured as a function of pH separately,
as illustrated in Fig. 5. Zeta potential is the potential differ-
ence, measured in the liquid, between the shear plane and the
bulk of the liquid beyond the limits of the electrical double
layer. The magnitude of the zeta potential gives an indication
of the potential stability of the colloidal system. In case of
carbons, it is observed that the zeta potential changed to neg-
ative values with increasing pH. In aqueous media, the pH of
the samples is one of the most important factors that affect
the zeta potential. If carbons are placed in alkali media then
the carbons tend to acquire more negative charge. The addi-
tion of acid causes a build up of positive charge. Therefore a
zeta potential versus pH curve will be positive at low pH and
lower or negative at high pH. On the contrary, in the case of
Pt particles, the zeta potential shifted to the negative direction
when the pH increased up to 6, after which the zeta potential
Fig. 1. Dependence of glycolate anion concentration as a function of pH.
nated form as the glycolate anion in alkaline solutions. Glycolate
ion is considered to act as a stabilizer by forming chelate-type
complexes via its carboxyl groups [17]. Consequently, a strong
correlation has been drawn between the concentration of glyco-
late anion and particle size of metal in solution. The theoretical
calculation of the glycolate anion concentration in the solution
as a function of pH was carried out using the selected initial gly-
colic acid concentration and the dissociation constant. As shown
in Fig. 1, the major change in the glycolate anion concentration
was observed in the pH range of 2–6. A constant glycolate anion
concentration was expected at pH higher than 6 and glycolate
anion was absent from the solution at pH lower than 2. Based on
the above estimation a set of Pt/C catalyst samples was prepared
by changing solution pH.
Fig. 2(a–f) shows the HR-TEM images of the Pt/C catalyst
powder prepared by using (a) 0.00 M, (b) 0.05 M, (c) 0.075 M,
(d) 0.1 M, (e) 0.15 M and (f) 0.2 M of NaOH, respectively. The
influence of the NaOH concentration on the resulting final pH
and Pt particle size determined from HR-TEM images are sum-
marized in Table 1. Normally, the initial pH of solutions before
refluxing was dropped as a result of the reduction of the Pt salts
at higher temperatures. This drop in solution pH was consis-
tent in all the solutions where NaOH was added. Therefore only
the final pH is considered for discussion. According to Table 1,
the final pH of the solution increases from 1.45 to 9.39 with
increasing NaOH concentration in the solution. As shown in
Fig. 2, it is observed that the Pt particle size decreases with an
increase in NaOH concentration, whereas the effect is not so sig-
Table 1
Effect of NaOH concentration on the resulting Pt particles size and pH
NaOH
(mol L−1
Initial
pH
Final
pH
Particle size (nm)
in XRD
Particle size (nm)
in HR-TEM image
)
0.0
0.72
11.01
11.73
11.66
11.75
11.71
1.45
3.82
4.33
6.03
8.50
9.39
5.2
3.1
2.1
1.7
1.7
1.6
8.5
3.8
3.1
2.5
2.3
2.2
0.05
0.075
0.1
0.15
0.2

H.-S. Oh et al. / Electrochimica Acta 52 (2007) 7278–7285
7281
Fig. 2. HR-TEM images of Pt particles on Ketjen black 300J prepared at different NaOH concentrations: (a) 0.0 M, (b) 0.05 M (c) 0.075 M, (d) 0.1 M, (e) 0.15 M
and (f) 0.2 M. The bar indicates a 5 nm scale.
was not affected by the change in pH. The key thing to note
is that pH 6 is the point where the concentration of glycolate
anion starts to be saturated, as shown in Fig. 1. Based on this
observation, it is reasonable to assume that there is a strong cor-
relation between the concentration of glycolate anion and the
zeta potential of Pt particles and that glycolate anions may be
specifically adsorbed on the surface of a Pt particle, leading to
a negatively charged surface. It is known that the stability of
a colloidal is dependent upon the balance between the attrac-
tive van der Waals forces and the electrostatic repulsive forces
between particles as they approach each other due to the Brow-
nian motion they are undergoing [20]. Considering the values
of zeta potential on both the species, it is concluded that the
adsorption of Pt particles is higher when the charge of Pt col-
loids is opposite to that of carbon supports at low pH while the
reduction of Pt loading with increasing pH of solution is caused

7282
H.-S. Oh et al. / Electrochimica Acta 52 (2007) 7278–7285
Fig. 6. Cyclic voltammograms of the Pt/C catalysts prepared at different NaOH
concentrations in 0.5 M H2SO4 solution at 5 mV s−1
Fig. 3. XRD patterns of Pt/C prepared at different NaOH concentrations.
.
mainly by the electrostatic repulsive force between carbons and
Pt particles.
Fig. 6 shows the cyclic voltammograms of Pt/C catalysts pre-
pared at different solution pH. Since the total amount of Pt/C
loaded on a glossy carbon electrode was constant, the actual
content of Pt varied based on the Pt wt% in Pt/C samples. The
CVs were recorded in 0.5 M H2SO4 as an electrolytic solution at
a scan rate of 5 mV s−1 with N2 purging. The effective surface
area of the carbon-supported Pt nanoparticles was calculated
from the area of the hydrogen desorption peaks between 0.03
and 0.3 V(SHE) after subtracting the contribution of the dou-
ble layer charge. This area is converted into the effective active
surface area of Pt using the conversion factor of 210 ␮C cm−2
[21]. The specific active surface area using the Pt wt% obtained
from ICP analysis and the particle size, which was calculated
from the specific active surface area based on the assumption
of spherical shape of particle, are displayed in Fig. 7. As seen
by the voltammograms, the area under the peak for hydrogen
desorption increases with increasing pH as a consequence of an
increasing active surface area of Pt nanoparticles, which are in
accordance with the result of HR-TEM analysis.
Fig. 4. The variation of Pt wt% deposited on carbon black as a function of NaOH
concentration.
Fig. 5. Effect of changing solution pH on the zeta potential of Pt particles and
carbon black.
Fig. 7. Specific active surface area and the particle size of Pt calculated from
the cyclic voltammograms as a function of NaOH concentration.

H.-S. Oh et al. / Electrochimica Acta 52 (2007) 7278–7285
7283
open air environment, it is designated as condition 3. The last
one, condition 4, is the mixed one. The reflux step at high tem-
perature was carried out in N2 gas and the rest of process was
performed in an open air environment at room temperature. In
all the experiments the NaOH concentration in the solution was
fixed at 0.075 M. Fig. 8 shows the effect of various gas atmo-
spheres on the Pt loading and the final pH of the solution. Fig. 9
represents the corresponding HR-TEM images. It is interest-
ing to note that during N2 purging for condition 1, significantly
reduced Pt loading (7 wt%) was obtained. It is likely that such
small loading is associated with the difference in zeta potential
and more probably the partial reduction of Pt ions. This is sup-
ported by the fact that the final pH of the solution prepared in
condition 1 was much higher than those of the other conditions.
Since the pH of the solution decreases as the reaction proceeds,
this implies that the N2 atmosphere suppresses the oxidation of
aldehydes resulting in the lack of electron supply to reduce Pt
ions completely. For condition 2, the particle size is relatively
large with visible aggregates due to the low pH and 36 wt% of
Pt loading was observed. The maximum Pt loading of 39.5 wt%
was achieved for condition 4 with a particle size of about 2.8 nm.
In order to initiate the oxidation of ethylene glycol, a reflux step
at high temperature above 140 ◦C is required. However, it is
not necessary to carry out the entire reaction at high tempera-
Fig. 8. Effect of different gas environments on the Pt wt% and the final pH. The
concentration of NaOH was fixed at 0.075 M.
As a controlling factor in the polyol process, the gas atmo-
sphere of the reaction was also examined closely. The purpose of
the experimental investigation of different gas modes is to find
a method to increase the Pt loading on carbon without increas-
ing the particle size of Pt. When N2 or O2 is purged in entire
preparation steps, this is denoted as condition 1 and 2, respec-
tively. For the case where the synthesis is performed under the
Fig. 9. HR-TEM images of Pt catalysts synthesized in ethylene glycol solution with different gas conditions: (a) N2 purging, (b) open to air, (c) O2 purging and (d)
N2 purging followed by opening to air. Spt is the specific active surface area calculated by cyclic voltammogram. The bar indicates a 5 nm scale.

7284
H.-S. Oh et al. / Electrochimica Acta 52 (2007) 7278–7285
Table 2
Kinetic parameters from regression analysis of polarization
Electrode
E0 (V)
b (V dec−1
)
R (ꢀ cm2)
i0 (A cm−2
)
Roughness factor
0 M NaOH with air open
0.075 M NaOH with N2 followed by air open
0.928
0.986
0.066
0.058
0.20
0.15
2.1 × 10−7
7.6 × 10−7
70
265
60 mV dec−1 for Pt electrodes [25]. Note that the exchange cur-
rent density of the electrode (0.075 M NaOH) is higher than
that of the other electrode (0 M NaOH). This is attributed to
the difference in the roughness factor. Exchange current den-
sity is a measure of an electrode’s readiness to proceed with an
electrochemical reaction. For this reason, for an effective elec-
trode, it is critical to maximize the exchange current density. The
exchange current density of a fuel cell is generally expressed in
terms of the geometrical surface area of the electrode rather
than the actual surface area of the catalyst. Therefore, assuming
that the nature of the oxygen reaction on Pt does not change,
the magnitude of the exchange current density in fuel cells
depends on the roughness factor, which is the ratio between
the actual reactive surface area and the geometrical electrode
area. Since the electrode prepared at 0.075 M has a higher active
surface area, it ends up with a higher exchange current den-
sity.
ture since the oxidation of aldehydes is possible even at room
temperature. Unlike conditions 2 and 3, condition 4 allows the
reaction to take place at low temperature. It probably enables
one to achieve a small Pt particle and high loading.
The performances of the MEAs in a single fuel cell are
evaluated and shown in Fig. 10. From the results of Fig. 10,
it is found that the electrode prepared at 0.075 M of NaOH
with N2 followed by open air (condition 4) shows much bet-
ter performance in comparison to the electrode with 0 M of
NaOH with open air (condition 3). This is attributed to the
difference in particle size of Pt. As described before, the par-
ticle size is affected by the concentration of glycolate anion,
which is a function of pH. In order to evaluate the electrode
kinetic parameters from the polarization curves, the experimen-
tal data were fitted to the following semi-empirical equation
[22]
E = E0 − b log i − Ri
E0 = Er + b log i0
(3)
(4)
4. Conclusions
where E and I are the experimentally measured cell voltage
and current, Er the reversible cell voltage, i0 and b are the
exchange current and the Tafel slope for oxygen reduction,
respectively. R represents the total contributions of the lin-
ear polarization components, which include the charge transfer
resistance of the hydrogen oxidation reaction, the resistance of
the electrolyte in the cell and the linear diffusion terms caused
by diffusion in the gas phase in the diffusion layer and/or in
the thin film [23,24]. The regression results are summarized
in Table 2. Both the electrodes show similar Tafel slopes of
66 and 58 mV dec−1, which is close to the typical value of
An investigation of the parameters of Pt colloid synthe-
sis using the polyol process was carried out. Although the
adjustment of pH behaves as a key factor in controlling the nan-
odimension of the Pt particles, a severe reduction in the metal
loading is observed with increasing solution pH. According to
the zeta potential study, this is attributed to the electrostatic sta-
bilization between Pt particles and carbon supports. The zeta
potential of the carbon support decreased to negative values with
increasing solution pH while that of the Pt particles remained
constant at a negative charge after pH 6. Therefore, poor adsorp-
tion or repulsive forces between the metal colloids and the
supports occurs, resulting in reduced Pt particle loading. Pt load-
ing and particle size are also affected by the gas environment
during Pt/C synthesis in the polyol process. It is observed that
carrying out the entire process of Pt/C formation in N2 showed
very good control over Pt particle size whereas the Pt load-
ing is significantly low. When the process of Pt/C formation is
carried out in the presence of O2, the Pt loading is increased
up to 36 wt%. However, the particle size of Pt increases due
to agglomeration at low solution pH. As a modification to the
polyol process, the reduction of Pt metal ions at elevated tem-
perature with N2 purging followed by the further reduction at
room temperature with air showed the best results with almost
40 wt% loading and a small particle size of 2.8 nm.
Acknowledgement
This work is supported by Hyundai Motors and the Ministry
of Commerce, Industry and Energy.
Fig. 10. Comparison of MEA performance between the electrodes prepared at
different NaOH concentrations and gas atmospheres.

H.-S. Oh et al. / Electrochimica Acta 52 (2007) 7278–7285
7285
References
[13] M. Watanabe, M. Uchida, S. Motoo, J. Electroanal. Chem. 229 (1987)
395.
[14] T. Masaharu, H. Masayuki, N. Yuki, Mater. Lett. 58 (2004) 2326.
[15] Y.J. Zhu, X.L. Hu, Mater. Lett. 58 (2004) 1234.
[16] Y.J. Zhu, X.L. Hu, Chem. Lett. 33 (2004) 760.
[17] C. Bock, C. Paquet, M. Couillard, G.A. Botton, B.R. MacDougall, J. Am.
Chem. Soc. 126 (2004) 8028.
[18] F. Fievet, J.P. Lagier, B. Blin, B. Beaudoin, M. Fiflarz, Solid State Ionics
32/33 (1989) 198.
[19] D.R. Lide, CRC Handbook of Chemistry and Physics, 71st ed., CRC Press,
1990.
[1] I.S. Ahmadi, Z.L. Wang, T.C. Green, A. Henglein, M.A. El Sayed, Science
272 (1996) 1924.
[2] Z. Tang, D. Geng, G. Lu, J. Colloid Interface Sci. 287 (2005) 159.
[3] H. Kim, B.N. Popov, Electrochem. Solid-State Lett. 7 (2004) A71.
[4] S. Hirano, J. Kim, S. Srinivasan, Electrochim. Acta 42 (1997) 1587.
[5] M.S. Wilsn, S. Gottesfeld, J. Appl. Electrochem. 1 (1992) 1.
[6] Z. Liu, L.M. Gan, L. Hong, W. Chen, J.Y. Lee, J. Power Sources 139 (2005)
73.
[7] X. Yan, H. Liu, K.Y. Liew, J. Mater. Chem. 11 (2001) 3387.
[8] A. Gamez, D. Richard, P. Gallezot, F. Gloaguen, R. Faure, R. Durand,
Electrochim. Acta 41 (1996) 307.
[20] R.J. Hunter, Zeta Potential in Colloid Science, Academic Press, UK,
1981.
[21] A.J. Bard, Electroanalytical Chemistry, vol. 9, Marcel Dekker, 1976.
[22] J. Kim, S.M. Lee, S. Srinivasan, J. Electrochem. Soc. 142 (1988) 2670.
[23] Y.W. Rho, S. Srinivasan, J. Electrochem. Soc. 141 (1994) 2089.
[24] V.A. Paganin, E.A. Ticianelli, E.R. Gonzalez, J. Appl. Electrochem. 26
(1996) 297.
[9] M. Watanabe, K. Sakairi, US Patent 5,728,485 (1998).
[10] H. Boennemann, R. Binkmann, P. Britz, U. Endruschat, R. Moertel, U.A.
Paulus, G.J. Feldmeyer, T.J. Schmidt, H.A. Gasteiger, R.J. Behm, J. New
Mater. Electrochem. Syst. 3 (2000) 199.
[11] N. Toshima, Y. Wang, Langmuir 10 (1994) 4574.
[12] F. Bonet, V. Delmas, S. Grugeon, R.H. Urbina, P.-Y. Silvert, K. Tekaia-
Elhsissen, Nanostruct. Mater. 11 (1999) 1277.
[25] A. Parthasarathy, S. Srinivasan, A.J. Appleby, C.R. Martin, J. Electroanal.
Chem. 339 (1992) 101.