i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 5 1 5 6 e1 5 1 6 3

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

Ethanol electro-oxidation: Cyclic voltammetry,

electrochemical impedance spectroscopy and galvanostatic
oscillation

Lijuan Han a, Hua Ju b, Yanhui Xu a,*

a
Institute of Chemical Power Sources, Soochow University, Moye Rd 688, Suzhou 215006, China
b
School of Urban Rail Transportation, Soochow University, 215006 Suzhou, China

article info abstract

Article history: In the present work, ethanol electro-oxidation reaction on Pt electrode has been studied in
Received 12 June 2012 details by measuring and analyzing the cyclic voltammetries (CV), the dependence of the
Received in revised form electrochemical impedance spectroscopies (EIS) on the applied potentials and the galva-
24 July 2012 nostatic potential oscillation. The CV measurement has exhibited a bistable characteristic.
Accepted 9 August 2012 The origin of all the oxidation and reduction peaks has been analyzed. The origin of the
Available online 28 August 2012 bistable characteristic has also been studied by measuring the EIS at different potentials. In
addition, the dependence of the galvanostatic oscillation on the applied potential and the
Keywords: ethanol concentration has been reported.
Galvanostatic oscillation Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
Negative real impedance reserved.
Cyclic voltammetry
Ethanol electro-oxidation

1. Introduction electroecatalytic activity of Pt-based electro-catalysts, it is of

very importance to understand the electro-oxidation reaction
In the past decades, direct alcohol fuel cells (DAFC), as one of mechanism [2,8e11,21e23], but until now the detail of the
the most promising chemical power sources, have attracted electro-oxidation reaction mechanism remain controversial,
more and more attention due to their high energy-conversion especially for long-chain alcohols. In the ethanol molecule
efficiency, proper operation temperature and simple handling there are two kinds of protons. One is bonded to the oxygen
of fuel, as well as convenience to use [1e7]. The main chal- atom. If it is firstly oxidized, then the first reaction step is
lenge faced by DAFC includes the solution of poor electro-
oxidation reaction kinetics, the poisoning tendency of the CH3 CH2 OH / ½CH3 CH2 Oad þ Hþ þ e
anode catalysts by some intermediates, such as adsorbed CO
The other is bonded to the carbon atom. If it is firstly
[8e14], as well as the large over-potential for oxygen reduction
oxidized, then the first reaction step should be
reaction.
Much attention has been paid on the Pt-based electro-
CH3 CH2 OH / ½CH3 CHOHad þ Hþ þ e
catalysts that are well known to have a good catalytic behavior
for methanol, ethanol, propanol and ethylene glycol etc. It can be not still concluded which one of the two reaction
electro-oxidation reaction [2,5,12e20]. To improve the pathways is right.

* Corresponding author. Tel.: þ86 512 67261337; fax: þ86 512 67261575.
E-mail addresses: [email protected], [email protected] (Y. Xu).
0360-3199/$ e see front matter Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ijhydene.2012.08.034
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 5 1 5 6 e1 5 1 6 3 15157

Some electro-oxidation systems are able to exhibit oscil- Fig. 1 shows the CV pattern of Pt working electrode in
latory behavior. The appearance of the oscillatory behavior is 0.5 mol L1 sulfuric acid solution. For the two couples of redox
related with the co-existence of the positive and negative peaks positioned at 0.44 V and 0.59 V, the oxidation peak
feedback. For the alcohol electro-oxidation, it is commonly potential is the same as the reduction peak potential, which
accepted that the negative feedback is just the poisoning means that they represent the under-potential electro-depo-
process. Therefore, the analysis and study of the oscillatory sition and adsorption/desorption of the hydrogen. The
behavior may provide some useful information on the oxidation peaks at 0.25 V and 0.36 V should correspond to the
poisoning reaction. It is plausibly concluded that the long- two-step formation of Pt oxides/hydroxides. At the potentials
chain alcohol has a larger poisoning ability. Therefore we of more than 0.8 V the oxygen evolution reaction becomes
are planning to compare the electrochemistry of methanol, more obvious. The reduction peak at 0.02 V is believed to
ethanol, propanol, ethylene glycol, propylene glycol and pro- relate with the reduction of Pt oxides/hydroxides. The
panetriol etc by comparing their oscillatory response. Our phenomenon that this reduction peak is sharper than the
purpose is to explore the reaction mechanism and to find the corresponding oxidation peaks may imply that the reduction
ways to improve the electro-oxidation reaction kinetics of the reaction of Pt oxides/hydroxides is a fast step. The obtained
alcohols, as results, the fuel cell performance. In the present CV pattern of Pt electrode in 0.5 mol L1 H2SO4 solution is
work, the ethanol electro-oxidation was firstly reported. same as that reported in literature [3,24,25], which has
confirmed our experimental quality. There is no impurity in
the solution and on the electrode surface.
2. Experimental Fig. 2 shows the CV pattern in 0.1 mol L1 H2SO4 þ x mol L1
CH3CH2OH (x ¼ 0.05, 0.5 and 3.0) solutions at the scanning rate
The solutions were prepared by dissolving the chemicals of 1 mV s1. The first main oxidation peak pa1 is found to be at
(CH3CH2OH p.a., H2SO4 p.a.) in ultra-pure water (18.2 MU, 0.2 V for the solution containing 3.0 mol L1 CH3CH2OH. One
Hitech Ultra-pure water system). The working electrode is a Pt shoulder peak pas can be found at the left side of the first main
metal electrode with an area of 24 mm2. The purity of Pt metal oxidation peak, at about 0.01 V. In some literature [26e29]
is 99.99%. The counter electrode is a Pt electrode with much there is no shoulder peak being reported at the low-potential
larger area, and the reference electrode is Hg/Hg2SO4. The side of the main oxidation peak. The position of the
distance between the working electrode and reference elec- shoulder peak is not influenced by the methanol concentra-
trode was set up at 20 mm. The corresponding electro- tion and the scanning rates. It is reasonable to conclude that
chemical tests were performed with PE Parc 2273 or CHI 660C this shoulder peak pas should originate from the non-faradiac
electrochemical workstations. adsorption/desorption process of some organics.
To remove some possible organic impurities existing on Pt The formation of the first main oxidation peak is believed
electrode surface, before electrochemical measurement, the to relate with the competitive adsorption between OH and
working electrode was pre-treated by sweeping its electrode oxygen-containing intermediates. Here the OH adsorbent is
potential between-in 0.65 V and 0.80 V in 0.5 mol L1 H2SO4 formed via the water oxidation decomposition. Usually, the
solution until the standard CV was obtained, as showed in oxygen-containing intermediate is assigned to CO [8e14].
Fig. 1. Fig. 3 shows the dependence of the CV pattern on the
scanning rates. The potential of the oxidation peak pa2 is about
0.45 V when the scanning rate is 1 mV s1. The decrease in
ethanol concentration does make the peak potential slightly
3. Results and conclusion more positive, as showed in Fig. 2. In addition, the increase in
the scanning rate does also lead to more positive peak
The CV pattern in sulfuric acid electrolyte is useful to analyze
potential. These results may imply that the oxidation peak pa2
the electro-oxidation mechanism of ethanol at Pt electrode.

Fig. 2 e The CV pattern in 0.1 mol LL1 H2SO4 D x mol LL1

L1
Fig. 1 e The CV pattern of Pt electrode in 0.5 mol L H2SO4 CH3CH2OH (x [ 0.05, 0.5 and 3.0) solutions, the scanning
solution, the scanning rate is 50 mV sL1. rate is 1 mV sL1.
15158 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 5 1 5 6 e1 5 1 6 3

Fig. 3 e The CV patterns of Pt electrode in 3.0 M

CH3CH2OH D 0.1 M H2SO4 solution.

should represent an electrochemical process accompanied by

charge-transfer process.
The oxidation peak pa3 can not be found when the ethanol
concentration is 0.05 mol L1, as seen in Fig. 2, while it appears
at 0.96 V in the solution containing 3.0 mol L1 CH3CH2OH. The
increase in the scanning rate makes the peak potential more
positive. The fact that the oxidation peak position is influ-
enced by the scanning rate may imply that this oxidation peak
Fig. 4 e The dependence of the peak current (Ipa2, (A); Ipa3,
represents an electrochemical process in which a charge
(C)) and potential (Ppa2, (B); Ppa3, (D)) on the scanning rate in
transfer process should be involved.
3 M CH3CH2OH D 0.1 M H2SO4 solution.
In the negative sweeping potential direction, the oxidation
peak par appears at 0.03 V when the ethanol concentration is
3.0 mol L1. And, its peak potential becomes more negative as
the ethanol concentration decreased (Fig. 2) or the scanning pa3 in 3 M CH3CH2OH þ 0.1 M H2SO4 solution. The relation
rate increased (Fig. 3). At lower scanning rates, the oxidation between the peak current and the square root of the scanning
peak par is left-to-right asymmetric. It is believed that there rate is linear for the reversible and irreversible electro-
exists an oxidation peak hidden in the low-potential side of chemical reactions. The peak potential is independent of the
the oxidation peak par and the peak potential of this hidden scanning rates for the reversible reaction, while it has a linear
oxidation peak is not influenced by the scanning rates and the dependence upon the logarithm of the square root of the
ethanol concentrations. It should have a same origin with the scanning rate if the reaction is irreversible.
oxidation peak pas. The oxidation peak par is believed to orig- It can be seen from Fig. 4 that the peak current of the
inate from the competitive adsorption between OH and CO, oxidation peak pa2 has an approximate linearization with the
and to represent the ethanol electro-oxidation at metallic Pt square root of the scanning rate. The peak potential does also
electrode surface. It has a same origin with the oxidation peak
pa1.
One small reduction peak can be found at 0.015 V from the
Fig. 2, when the ethanol concentration is 0.05 mol L1. The
reduction peak is hardly to see when the ethanol concentra-
tion increased, but becomes clearer if the scanning rate
increased. These phenomena have at least conformed that the
reduction peak represents one electrochemical reaction
process and the corresponding reduction reaction is of a fast
step. In Fig. 1, the reduction peak represents the electro-
reduction of Pt oxides/hydroxides. We think, the reduction
peak in the solution containing ethanol should also represent
the electro-reduction of Pt oxides/hydroxides. The ethanol or
some intermediates can react with the Pt oxides/hydroxides,
which leads to the disappearance of reduction peak in higher
ethanol concentration. Fig. 5 e The electrochemical impedance spectroscopies of
Fig. 4 shows the dependence of the peak potential and Pt electrode in 3 M CH3CH2OH D 0.1 M H2SO4 solution at
current on the scanning rate for the oxidation peaks pa2 and different applied potentials.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 5 1 5 6 e1 5 1 6 3 15159

Fig. 6 e The electrochemical impedance spectroscopies of

Fig. 8 e The time evolution of the electrode potential at
Pt electrode in 0.5 M CH3CH2OH D 0.1 M H2SO4 solution at
different applied currents in 3.0 M CH3CH2OH D 0.1 M
different applied potentials.
H2SO4 solution.

linearly vary with the logarithm of the square root of the

scanning rate. It could be concluded from the Fig. 10 that the lowest frequency (0.001 Hz) the real impedance becomes
electrochemical reaction that is represented by the oxidation positive again. The appearance of the negative real impedance
peak pa2 is a fully irreversible. For the oxidation peak pa3, the for non-zero finite frequencies may imply that such an elec-
linearity between the peak current and the square root of the trochemical system can exhibit unstable behavior [31]. At
scanning rate is very bad, while linearity between peak higher potential than 0.36 V, the impedance spectra becomes
potential and the logarithm of the square root of the scanning “normal” as that obtained at 0.06 V.
rate is better. At the scanning rate of 50 mV/s, the peak current Fig. 6 shows the potentiostatic EIS pattern in 0.1 mol L1
is much smaller than that expected from the linear relation, H2SO4 þ 0.5 mol L1 CH3CH2OH solution. At 0.01 V, the
which may be caused by some unexpected reasons. However, impedance data emerge in a clockwise mode as the frequency
the linearity becomes better if the CV obtained at 50 mV/s is decreases. When the frequency decreased from 0.333 Hz to
excluded. Based upon the above-mentioned analysis, it could 0.264 Hz, the imaginary impedance becomes positive and an
be concluded that the oxidation peak pa3 is an irreversible inductive arc appears. In general, it is commonly accepted
reaction peak. The two oxidation peaks pa2 and pa3 should that the inductive arc in the low-frequency domain in Nyquist
represent the electro-oxidation process of the ethanol on the plot should represent the adsorption/desorption process. The
oxidized electrode surface. fast transition from the capacitive behavior to the inductive
Fig. 5 shows the potentiostatic EIS pattern in 0.1 mol L1 behavior may imply a fast reaction kinetic for adsorption/
H2SO4 þ 3.0 mol L1 CH3CH2OH solution. For 0.06 V, the desorption process. The EIS pattern obtained at 0.05 V is
impedance data in Nyquist plot emerge in a clockwise mode similar as that obtained at 0.09 V in the solution containing
as the applied frequency decreases. Such a kind of EIS pattern 3.0 mol L1 CH3CH2OH. When the electrode potential was
is commonly encountered [30]. The impedance spectrum has increased up to more than 0.37 V, the impedance spectra
a qualitative change when the applied potential is increased becomes “normal”.
up to 0.09 V. As the frequency decreases, the real impedance The EIS pattern that is obtained at 0.15 V in the solution
becomes negative for the frequency of about 0.264 Hz. At the containing 0.05 mol L1 CH3CH2OH is somewhat different, as

Fig. 7 e The electrochemical impedance spectroscopies of Fig. 9 e The time evolution of the potential at the applied
Pt electrode in 0.05 M CH3CH2OH D 0.1 M H2SO4 solution at currents in 0.1 mol LL1 H2SO4 D 0.5 mol LL1 CH3CH2OH
different applied potentials. solution.
15160 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 5 1 5 6 e1 5 1 6 3

0.1 mol L1 H2SO4 þ 3 mol L1 CH3CH2OH solution, as showed

in Fig. 8, i.e., the oscillation is always the relax-type oscilla-
tion. But when the ethanol concentration decreased down to
0.5 M, the oscillation transited from the relax-type oscillation
for the low currents to the nail-type one for the high currents.
In addition, the parameter space in which the system is able to
exhibit the oscillatory behavior becomes thinner. When the
current is increased up to 0.6 mA, the oscillatory behavior
disappears.
Fig. 10 shows the time evolution curves of the electrode
potential under a galvanostatic operation mode in 0.1 mol L1
H2SO4 þ 0.05 mol L1 CH3CH2OH solution. In our experimental
conditions there is no oscillation being found.
According to the literature [8e11,21e23], the ethanol
Fig. 10 e The time evolution of the electrode potential at electro-oxidation mechanism can be summarized as follows,
the applied currents in 0.1 mol LL1 H2SO4 D 0.05 mol LL1
nPt þ CH3 CH2 OH ðaqÞ / Ptn  CH3 CH2 OH (1)
CH3CH2OH solution.

Ptn  CH3 CH2 OH / Ptn  CH3 CHOH þ Hþ þ e (2)

shown in Fig. 7. As the frequency decreases, the impedance Ptn  CH3 CHOH / Ptn  CH3 COH þ Hþ þ e (3)
data emerge in a counterclockwise mode. The real impedance
becomes negative for the lowest frequency, not for the non- Ptn  CH3 COH / Ptn  CH3 CO þ Hþ þ e (4)
zero finite frequencies. Based on such a kind of impedance
spectrum, it can not be concluded whether the electro-
Pt þ Ptn  CH3 CO / Ptn  CO þ Pt  CH3 (5)
chemical system is able to exhibit a nonlinear behavior. To our
knowledge, there is no literature concerning the unstable In common, the reaction pathway expressed by these five
behavior of an electrochemical system having such a kind of reactions is more possible in the acidic medium. In the alka-
impedance spectrum. In the following paragraph, it can be line medium, the most possible reaction pathway could be
found that such an electrochemical system is not able to expressed by the following reactions.
exhibit oscillatory behavior.
Ptn  CH3 CH2 OH þ OH / Ptn  CH3 CH2 O þ H2 O þ e (6)
Fig. 8 shows the galvanostatic potential oscillation curves
in 0.1 mol L1 H2SO4 þ 3.0 mol L1 CH3CH2OH solution. For the
Ptn  CH3 CH2 O þ OH / Ptn  CH3 CHO þ H2 O þ e (7)
0.05 mA current, the oscillation starts at about 2.7 h. The
oscillation amplitude can reach 0.314 V (from 0.18 V to
Ptn  CH3 CHO þ OH / Ptn  CH3 CO þ H2 O þ e (8)
0.134 V). The average potential is about 0.0 V. The oscillation
belongs to the relax-type oscillation. The potential jumps very It should be noted that some intermediates are possible to
quickly from the high-potential state to the low-potential leave the electrode surface to the solution as the incomplete
state, and then slowly wanders back to the high-potential by-products, such as CH3COOH, CH3CHO, CH4 and CH3CH3 etc.
state. As the time goes, the oscillation becomes faster. The The adsorbed CH3 group has two possible fates. It can
situation is similar for the 0.1 mA current but the oscillation combine with the adsorbed proton to form methane, or two
amplitude becomes larger and reached 0.344 V (from 0.21 V CH3 groups combine to form ethane,
to 0.134 V). As the current increases, the induction duration
becomes shorter, from 2.7 h for 0.05 mA, to 1.8 h for 0.1 mA Pt  CH3 þ Pt  H/2Pt þ CH4 (9)
and to 0.46 h for 0.5 mA, before the oscillation behavior starts
to occur. For 0.5 mA current, the oscillation amplitude is Pt  CH3 þ Pt  CH3 / 2Pt þ CH3 CH3 (10)
0.33 V. There is no induction duration being found in the There exist two possible explanations on the oxidation peak
timeeevolution curve of the potential when the applied pa1 formation. In the first explanation, the peak current is
current is 0.9 mA. The oscillation has a relatively high contributed to the direct oxidation reaction pathway of
frequency but the oscillation amplitude is relatively small for ethanol. As the potential increases some formed intermedi-
0.9 mA. As the current increased up to 1.3 mA, the oscillation ates deactivate the electrode and then the oxidation current
becomes slow again. At 1.4 mA, there is no oscillatory declined. In this explanation, the electrode surface is covered
behavior being found. by some adsorbed intermediates if the electrode potential is
Fig. 9 shows the galvanostatic potential oscillation curves kept at a potential more than the pa1, the peak potential of the
in 0.1 mol L1 H2SO4 þ 0.5 mol L1 CH3CH2OH solution. For main oxidation peak, for instance, at 0.45 V in Fig. 2.
0.1 mA current, the oscillation amplitude is about 0.29 V. And In fact, we could directly say that the direct oxidation
the induction duration is 4 h. Our experiment has showed that reaction is of very slow rate, although there is no direct
the induction duration can reach more than 12 h if the applied evidence. Usually, the anode in direct ethanol fuel cell oper-
current decreased furthermore. The oscillation type remains ates at the potential near to pa1, which does also imply that the
unchanged when the applied current is changed for the first explanation must be incorrect. The reason is simple. The
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 5 1 5 6 e1 5 1 6 3 15161

main intermediate that deactivates the Pt electrode is the

adsorbed CO. According to literature [32], the CO has
a stronger blocking effect for the entrance of other species into
the Pt electrode surface. The electrode surface is full of CO if
the electrode was kept at Ppa1 for long time and if the first
explanation is correct. In such a condition, the potentiostatic
current will decay down to zero.
In the second explanation, the electrode surface is covered
by OH-type species if its potential is kept at 0.45 V in Fig. 2. The
competitive adsorption reaction between OH and CO species
leads to the formation of the oxidation peak pa1.
Once the Pt electrode is immersed into the acidic solution Fig. 11 e Multi-pathway reaction model of ethanol electro-
containing ethanol, the non-faradiac chemical reaction will oxidation.
take place on the electrode surface. As results, some inter-
mediates formed on the surface and adsorbed on it. One of the
main intermediates is CO. As the potential increases, the
indirect electro-oxidation occurs, at same time the OH species
has confirmed that the electrochemical system could exhibit
starts to form via the decomposition of the water and remove
unstable behavior. And the EIS results have evidenced that the
the adsorbed CO via the following reactions
parameter space in which the electrochemical system can
exhibit nonlinear behavior falls into the potential range where
Ptn  CO þ Pt e OH / Ptm  COOH þ ðn þ 1  mÞPt (11) the first main oxidation peak appears. The average potential
of the oscillatory potentials is about 0.0 V. All the phenomena
Ptm  COOH/mPt þ CO2 þ Hþ þ e (12) imply that the oscillation relates with the electrode processes
represented by the shoulder and the main oxidation peak pa1
The CO adsorption on the Pt surface is dependent of the
and the pas. At the potential maximum, the electrode is
electrode potential [33]. As the potential increases and the
covered by OH, while at the potential minimum it is covered
reactions (11) and (12) proceed, at last, all the CO-type species
by oxygen-containing incomplete oxidation products, such as
existing on the electrode surface are removed and the elec-
CH3CHO, CH3CO or CO. If the direct electro-oxidation reaction
trode surface is covered fully by the OH. As mentioned in the
pathway can be omitted, then the positive feedback includes
literature [32], CO has a stronger blocking effect for the
the reactions (1)e(4), while the negative feedback is related
entrance of other species. But the OH-covered surface is open
with the reaction (5).
for the entrance of other species. It can be expected that the
In practice, the reaction represented by the oxidation peak
OH-covered electrode should have smaller impedance in
pa1 is employed. There should be two strategies to improve the
comparison with the CO-covered electrode. The impedance
electroecatalytic activity depending on the electro-oxidation
measurement has partially proved this suggestion. For the
reaction rate. When the electrode works in the low-potential
0.1 mol L1 H2SO4þ3.0 mol L1 CH3CH2OH solution, the
region (LPR), as indicated in Fig. 3, the electro-oxidation
impedance arc obtained at 0.06 V (the CO-covered surface) is
reaction rate can be improved by using some additives/
larger than that obtained at 0.36 V (the OH-covered surface),
supplementary catalysts to promote the formation of adsor-
as showed in Fig. 5. The situation in 0.1 mol L1
bed OH species, because in this potential region the removal of
H2SO4þ0.5 mol L1 CH3CH2OH solution is similar, as showed
the adsorbed CO is crucial. If it is hoped that the electrode
in Fig. 6. The impedance arc obtained at 0.01 V has the biggest
works in the high-potential side (HPS) of the main oxidation,
diameter. However, the situation in 0.1 mol L1
as indicated in Fig. 3 (in this case a larger operation current is
H2SO4 þ 0.05 mol L1 CH3CH2OH solution is different. The
expected), then the improvement in the electro-oxidation
impedance spectrum in this solution does not support our
reaction rate should be realized by using such a kind of
suggestion.
additives or supplementary catalysts that are able to delay or
As mentioned above, the shoulder peak in the positive
inhibit the formation of adsorbed OH species, because in this
sweeping potential direction (PSPD) and the hidden oxidation
case the detrimental factor to increase the operation current is
peak in the negative sweeping potential direction (NSPD) have
the poisoning effect of the adsorbed OH.
a same origin and are believed to be due to the non-faradiac
adsorption/desorption process of some organics. In prin-
ciple, the peak position should be not influenced by the
potential-scanning direction for a non-faradiac adsorption 4. Conclusion
process. From the CVs showed in Figs. 2 and 3, it can be found
that their position is being affected by the ethanol concen- In the article, the ethanol electro-oxidation reaction has been
tration and the potential-scanning direction as well as the studied in details. The cyclic voltammetry measurements
scanning rates. This phenomenon should originate from the have shown that in higher ethanol concentration there exist
disturbance of the main oxidation peaks pa1 and par in our three oxidation peaks whose position is dependent of the
experimental condition. scanning rate. At the low-potential side of the first oxidation
All the possible reaction pathways for ethanol electro- peak there is a shoulder peak. In the negative sweeping
oxidation were summarized in Fig. 11. The CV measurement potential direction one asymmetric oxidation peak can be
15162 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 5 1 5 6 e1 5 1 6 3

found. In his low-potential side there should be a hidden modern aspects of electrochemistry, vol. 40. Springer; 2007.
oxidation peak that is belied to corresponds to the non- p. P229e84.
faradiac adsorption/desorption process. The two oxidation [9] Herranz T, Garcia S, Huerta MVM, Peria MA, Fierro JLG,
Somodi F, et al. Electrooxidation of CO and methanol on
peaks at more positive potential should correspond to the
well-characterized carbon supported PtxSn electrodes: effect
electro-oxidation at oxidized electrode surface. of crystal structure. Int J Hydrogen Energy 2012;37:7109e18.
It is found that the potentio-static impedance measure- [10] Scott K, Shukla AK. In: White RE, Vayenas CG, editors. Direct
ment exhibits a qualitative change as the applied potential methanol fuel cells: fundamentals, problems and
changes. In general, in EIS Nyquist plot, the EIS data emerge in perspectives, modern aspects of electrochemistry, vol. 40;
the clockwise mode, but in some special potentials the EIS 2007. p. P127e227.
data emerge in the counterclockwise mode. The real imped- [11] Gyenge E. In: Zhang J, editor. Electrocatalytic oxidation of
methanol, ethanol and formic acid, PEM fuel cell
ance becomes negative for non-zero finite frequencies and
electrocatalyst and catalyst layers. Springer-Verlag London
positive for the lowest frequencies in higher ethanol concen- Ltd.; 2008. p. P165e287.
trations (3.0 mol L1 and 0.5 mol L1). In 0.1 mol L1 [12] Li HY, Zhang XH, Pang HL, Huang CT, Chen JH. PMo12-
H2SO4 þ 0.05 mol L1 CH3CH2OH solution, the real impedance functionalized grapheme nanosheet-supported PtRu
becomes negative for the lowest frequencies, not for non-zero nanocatalysts for methanol electro-oxidation. J Solid State
finite frequencies. The galvanostatic measurements have Electrochem 2010;14:2267e74.
[13] Mcnicol BD, Chapman AG, Short RT. Pt/Sn bimetallic
showed that the potential oscillation can be found for the
catalysts for the electro-oxidation of methanol: poisoning by
higher-concentration ethanol solutions. There is no oscilla-
silicate and phosphate anions. J Appl Electrochem 1976;6:
tory behavior being found in 0.1 mol L1 H2SO4 þ 0.05 mol L1 221e7.
CH3CH2OH solution. The galvanostatic oscillation amplitude [14] Tayal J, Rawat B, Basu S. Bi-metallic and tri-metallic Pt-Sn/C,
can reach more than 0.34 V, and the oscillation frequency is in Pt-Ir/C, Pt-Ir-Sn/C catalysts for electro-oxidation of ethanol
an order of magnitude of about 104e101 Hz, depending on in direct ethanol fuel cell. Int J Hydrogen Energy 2011;36:
the applied current and the ethanol concentration. 14884e97.
[15] Fujiwara N, Friedrich KA, Stimming U. Ethanol oxidation on
PtRu electrodes studied by differential electrochemical mass
spectrometry. J Electroanal Chem 1999;472:120e5.
Acknowledgment [16] Jiang L, Hsu A, Chu D, Chen R. Ethanol electro-oxidation on
Pt/C and PtSn/C catalysts in alkaline and acid solutions. Int J
This work was supported by the Foundations No. SRF for Hydrogen Energy 2010;35:365e72.
[17] Garcia G, Tsiouvaras N, Pastor E, Peria MA, Fierro JLG,
ROCS, SEM and BK2009110.
Huerta MVM. Ethanol oxidation on PtRuMo/C catalysts: in
situ FTIR spectroscopy and DEMS studies. Int J Hydrogen
Energy 2012;37:7131e40.
references
[18] Otomo J, Li X, Kobayashi T, Wen C, Nagamoto H,
Takahashi H. AC-impedance spectroscopy of anodic
reactions with adsorbed intermediates: electro-oxidation of
[1] Gonzalez AS, Arco EB, Escalante J, Sandoval OJ, Gamboa SA. 2-propanol and methanol on carbon-supported Pt catalyst. J
Methanol oxidation reaction on PtSnO2 obtained by micro- Electroanal Chem 2004;573:99e109.
wave-assisted chemical reaction. Int J Hydrogen Energy 2012; [19] Gupta SS, Datta J. An investigation into the electro-oxidation
37:1752e9. of ethanol and 2-propanol for application in direct alcohol
[2] Sieben JM, Duarte MME. Nanostructured Pt and Pt-Sn fuel cells. J Chem Sci 2005;117(4):337e44.
catalysts supported on oxidized carbon nanotubes for [20] Wang H, Jusys Z, Behm RJ. Electrochemical oxidation kinetics
ethanol and ethylene glycol electro-oxidation. Int J Hydrogen and mechanism of ethylene glycol on a carbon supported Pt
Energy 2011;36:3313e21. catalyst: a quantitative DEMS study. J Electroanal Chem 2006;
[3] Brandalise M, Silva RWRV, Tusi MM, Correa OV, Farias LA, 595:23e36.
Linardi M, et al. Electro-oxidation of ethanol using PtRuBi/C [21] Mahapatra SS, Dutta A, Datta J. Temperature dependence on
electrocatalysts prepared by borohydride reduction. Ionics methanol oxidation and product formation on Pt and Pd
2009;15:743e7. modified Pt electrodes in alkaline medium. Int J Hydrogen
[4] Rodrigues RMS, Tusi MM, Chikasawa MH, Forbicini CALGO, Energy 2011;36:14873e83.
Linardi M, Spinace EV, et al. Preparation and characterization [22] Krausa M, Vielstich W. Potential oscillations during
of PtRu/C-rare earth using an alcohol-reduction process for methanol oxidation at Pt-electrodes Part 1: experimental
ethanol electro-oxidation. Ionics 2011;17:189e93. conditions. J Electroanal Chem 1995;399:7e12.
[5] Kadirgan F, Beyhan S, Atilan T. Preparation and [23] Ianniello R, Schmidt VM, Rodriguez JL, Pastor E.
characterization of nano-sized Pt-Pd/C catalysts and Electrochemical surface reactions of intermediates formed
comparison of their electro-activity toward methanol and in the oxidative ethanol adsorption on porous Pt and PtRu. J
ethanol oxidation. Int J Hydrogen Energy 2009;34:4312e20. Electroanal Chem 1999;471:167e79.
[6] Serov A, Kwak C. Review of non-platinum anode catalysts for [24] Diaz V, Real S, Teliz E, Zinola CF, Martins ME. New
DMFC and PEMFC application. Appl Catal B: Environ 2009; experimental evidence on the formation of platinum
90(3e4):313e20. superactive sites in an electrochemical environment. Int J
[7] Coutanceau C, Brimaud S, Lamy C, Beger JM. Review of Hydrogen Energy 2009;34:3519e30.
different methods for developing nanoelectrocatalysts for [25] Herron ME, Pletcher D, Robinson J, Doyle SE, Roberts KJ,
the oxidation of organic compounds. Electrochim Acta 2008; Potter RJ, et al. In-situ potential modulation difference
53:6865e80. diffraction studies of platinum in sulphuric acid suing
[8] Garcia BL, Weidner JW. In: White RE, Vayenas CG, Gamboa- synchrotron x-ray dadiation. J Electroanal Chem 1995;384:
Aldeco ME, editors. Review of direct methanol fuel cells, 39e46.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 5 1 5 6 e1 5 1 6 3 15163

[26] Chen S, Schell M. Bistability and excitability in the elevated temperature and pressure: a high-temperature/
electrochemical oxidation of ethanol. Electrochimica Acta high-pressure DEMS study. J Power Sources 2009;190(1):2e13.
1999;44:4773e80. [30] Xu YH, Chen Y, Wu J, Li D, Ju H, Zheng J. The determination
[27] Katikawong P, Ratana T, Veerasai W. Temperature of the kinetic parameters of electrochemical reaction in
dependence studies on the electro-oxidation of aliphatic chemical power sources: a critical review. Int J Hydrogen
alcohols with modified platinum electrodes. J Chem Sci 2009; Energy 2010;35(12):6366e80.
121(3):329e37. [31] Koper MTM. Stability study and categorization of
[28] Simões FC, Anjos DM, Vigier F, Léger JM, Hahn F, electrochemical oscillators by impedance spectroscopy.
Coutanceau C, et al. Electroactivity of tin modified platinum J Electroanal Chem 1996;409(1e2):175e82.
electrodes for ethanol electrooxidation. J Power Sources [32] Ertl G. Oscillatory kinectis and spatio-temporal self-organization
2007;167(1):1e10. in reactions at solid surfaces. Science 1991;254:1750e5.
[29] Sun S, Halseid MC, Heinen M, Jusys Z, Behm RJ. Ethanol [33] Hamnett A. Mechanism and electrocatalysis in the direct
electrooxidation on a carbon-supported Pt catalyst at methanol fuel cell. Catal Today 1997;38:445e57.