Anodic oxidation of methane at noble metal electrodes: An 'in situ' surface enhanced infrared spectroelectrochemical study

Article abstract of DOI:10.1016/S0013-4686(01)00649-1

The mechanism of electrooxidation of methane at 25 °C in 0.5 M HClO₄ on the noble metal electrodes Pt, Au, Pd, Ru and Rh has been investigated by 'in situ' infrared spectroscopy. The final product of oxidation was found to be CO₂ in

Full text of DOI:10.1016/S0013-4686(01)00649-1

Electrochimica Acta 46 (2001) 3525–3534
www.elsevier.com/locate/electacta
Anodic oxidation of methane at noble metal electrodes: an
‘in situ’ surface enhanced infrared spectroelectrochemical
study
F. Hahn *, C.A. Melendres 1,2
Catalyse en Chimie Organique, Equipe Electrocatalyse, Uni6ersite´ de Poitiers, UMR-CNRS n 6503,
40, A6enue du Recteur Pineau, 86022 Poitiers Cedex, France
Received 28 February 2001; received in revised form 6 June 2001
Abstract
The mechanism of electrooxidation of methane at 25 °C in 0.5 M HClO4 on the noble metal electrodes Pt, Au, Pd,
Ru and Rh has been investigated by ‘in situ’ infrared spectroscopy. The final product of oxidation was found to be
CO2 in all cases. Using the technique of surface enhanced infrared absorption spectroscopy and deuterated water
solutions, it has been possible to detect the presence of adsorbed intermediates such as CO and ꢀCHO (or ꢀCOOH)
for the first time. The infrared signal enhancement observed could be accounted for by considering the increase in the
surface area of the electrodeposited metals as indicated by the cyclic voltammograms. Pt and Ru appear to have the
highest, and Au the lowest, electrocatalytic activity among the metals studied. © 2001 Elsevier Science Ltd. All rights
reserved.
Keywords: Methane; Noble metals; Anodic oxidation; Spectroelectrochemical study; Surface enhanced infrared spectroscopy
1. Introduction
more widespread applications. Methane and methanol
are among the most simple and readily available or-
ganic compounds that can be used as fuel and are hence
of great interest for use in fuel cell devices. Methanol is
attractive because it can be obtained from renewable
sources, such as corn and other agricultural products.
Its electrochemical oxidation has therefore been exten-
sively studied [2,3]. Little work has been done on
methane as a fuel cell feedstock. Recent discovery of
methane hydrate deposits under the ocean floor and the
Arctic permafrost suggests that they are a potentially
enormous resource that could fuel the 21st century and
the third millennium [4]. There is thus a great economic
incentive to develop a fuel cell device that could effec-
tively utilize methane.
The need for new power sources which minimize
environmental pollution has spurred a renaissance in
the further development of fuel cells for electric vehicle
propulsion and for electricity generating power stations
[1]. Fuel cell devices have proven their utility in
aerospace applications where hydrogen has been used
as fuel. However, problems with the storage of hydro-
gen in tanks or as metal hydrides, as well as costs, have
prevented the commercial use of fuel cells for other
* Corresponding author. Tel.: +33-549-453971; fax: +33-
549-453580.
E-mail addresses: francoise.hahn@univ-poitiers.fr (F.
Hahn), camelendres@hotmail.com (C.A. Melendres).
1 ISE member.
So far, because of its high stability, the use of meth-
ane as fuel has been limited to high temperature fuel
cells, i.e. molten carbonate and solid oxide [5], after
reforming into hydrogen and carbon monoxide mix-
2 On sabbatical leave of absence from Argonne National
Laboratory, Argonne, IL 60439, USA.
0013-4686/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0013-4686(01)00649-1

3526
F. Hahn, C.A. Melendres / Electrochimica Acta 46 (2001) 3525–3534
tures [6]. There have been efforts to oxidize methane
directly in low temperature fuel cells [7–9]. Recently,
Berthelot et al. [10] showed that electrooxidation of
methane could be carried out in a polymer electrolyte
membrane fuel cell using high surface area electrodes
with Pt and dispersed Pt–Ru as electrocatalysts. They
reported that different products resulted from the use
of the two different electrodes, i.e. carbon dioxide was
produced in the case of Pt and methanol resulted when
using Pt–Ru. Attempts to determine the mechanism of
electrooxidation of methane from purely electrochemi-
cal studies have also been made [11,12]. Hsieh and
Chen interpreted the results of their polarization mea-
surements in acidic sulfate electrolyte at 80 °C by pos-
tulating the formation of intermediates from adsorbed
CH3 [11]. Sustersic et al. suggested the formation of
adsorbed intermediates such as CO2 and COH from
their cyclic voltammetry (CV) studies in 1 N H2SO4
solution at 60 °C [12]. In the present work, we provide
‘in situ’ infrared spectroscopic evidence that unequivo-
cably establishes for the first time the nature of some of
the intermediates involved in the oxidation of CH4. The
technique of surface enhancement of the IR signal by
electrodeposition was found particularly useful in en-
abling the detection of weak signals from the
intermediates.
interfacial Fourier transform infrared reflectance spec-
troscopy (SNIFTIRS). In the LPS-FTIRS technique
[14], the potential of the electrode is scanned at 1 mV/s
and interferograms are taken at regular intervals, say
100 mV. The interferograms are co-added and con-
verted into reflectances, which are then ratioed to that
of a reference in order to obtain spectra (transmit-
tance). The second method (PDIRS) is very similar to
the first one except that instead of scanning the poten-
tial at a very slow rate, it is stepped from an initial
value (the reference) to a higher value and interfero-
grams are collected for preset periods of accumulation
at each potential [15]. It has the advantage that long
counting periods could be employed to collect the
spectra and thus improve on the signal to noise ratio.
In the SNIFTIRS technique [16], the potential is modu-
lated between two values while interferograms are
taken at each potential and then co-added. Surface
enhancement of the IR signal on Pt was carried out by
platinization of the electrode following the procedure of
Bjerke et al. [17]. Surface enhanced infrared absorption
spectroscopy (SEIRAS) studies on Ru and Pd were
carried out by electrodeposition of the metal on ruthe-
nium disc and glassy carbon (GC) substrates, respec-
tively. The conditions used for electrodeposition were
similar to those outlined by Tian et al. [18] for surface
enhanced Raman spectroscopy (SERS), but with some
modifications. Thus, our solutions consisted of 0.01 M
PdCl2 in 0.1 M KCl for Pd plating and 0.01 M
K2RuCl5·xH2O in 0.1 M KCl for Ru deposition. More-
over, the latter was carried out at a potential of −0.2
V versus RHE for 5 min rather than the conditions
used by Tian et al. [18]. The roughening procedure
prescribed by Weaver and coworkers [19] was also used
in an attempt to obtain SEIRAS on Au. In all cases,
following electrodeposition, the electrodes were rinsed
thoroughly with water before transfer to the IR spec-
troelectrochemical cell. All electrochemical and infrared
spectroscopic measurements were carried out at ambi-
ent temperature.
2. Experimental
The IR spectroelectrochemical cell and spectrometer
have been described previously [13]. The cell was fitted
with a flat CaF2 window (25 mm diameter×4 mm
thick) to admit the IR radiation. The Fourier transform
infrared (FTIR) spectrometer was a Bruker model IFS
66v with a liquid nitrogen-cooled MCT detector. The
working electrodes were in the form of discs about 8
mm in diameter and 0.5–2 mm thick. They were at-
tached to the end of a glass tube holder using epoxy,
then polished with successive grades of alumina powder
down to 0.3 mm, and rinsed thoroughly. The counter
electrode consisted of a coil of Pt wire, while a hydro-
gen electrode was used as reference (RHE). The elec-
trolyte was 0.5 M HClO4; it was prepared from a
suprapure material and ultrahigh purity water (Mil-
lipore-Milli Q). The solution was deaerated by bubbling
ultrapure nitrogen gas (Air Liquide, quality U). The
methane used was also obtained from Air Liquide
(Alphagaz, purity N55). It was adsorbed on the elec-
trodes by bubbling through the electrolyte solution for
1–3 h while the potential of the electrode was held at a
certain value. ‘In situ’ IR spectra were obtained by
either one of three techniques: (1) linear potential
sweep-Fourier transform infrared spectroscopy (LPS-
FTIRS); (2) potential difference infrared reflectance
spectroscopy (PDIRS); and (3) subtractively normalized
3. Results
Fig. 1 shows cyclic voltammograms of Pt before
(solid line) and after adsorption of methane (broken
line) at 0.26 V while bubbling for 2 h in 0.5 M HClO4.
It is evident that the hydrogen adsorption–desorption
peaks are suppressed partially which indicates the pres-
ence of an adsorbed species covering the electrode
surface. There is an increase in current during the
anodic scan which presumably corresponds to the oxi-
dation of the adsorbed species. Fig. 2 shows ‘in situ’ IR
spectra of the Pt disc electrode obtained by LPS-
FTIRS. The technique allows the monitoring of the
products or intermediates of an electrode reaction as a

F. Hahn, C.A. Melendres / Electrochimica Acta 46 (2001) 3525–3534
3527
function of applied potential. From Fig. 2, for example,
the appearance of a band at 2345 cm−1 is associated
with the production of CO2 due to the electrooxidation
of CH4. There is some indication of the presence of a
band due to CO at about 2020 cm−1. The positive-go-
ing band at about 1650 cm−1 is due to water which is
presumably consumed in the oxidation process. How-
ever, the anticipated increase in the intensity of the
band with potential is difficult to discern because of the
poor signal to noise ratio. There is also a broad band at
about 1750 cm−1 which may be due to some dissolved
acidic, aldehydic, or ketonic species bearing a ꢀCO
moiety. We were unable to observe a band that we
could attribute to adsorbed CH4 or other surface spe-
cies using conventional FTIR spectroscopy. We there-
fore resorted to the use of the SEIRAS technique by
electrodeposition of Pt on the Pt (will be referred to as
platinized Pt) disc electrode [17]. Along with the use of
deuterated water, D2O, to shift the position of the
water bands (especially the band at around 1640
cm−1), the resulting signal enhancement is then suffi-
cient to observe the very weak signals due to the
intermediates adsorbed on the surface. Fig. 3 shows
LPS-FTIR spectra obtained ‘in situ’ following adsorp-
tion of CH4 on the surface of a platinized Pt electrode.
(Note that the reference for the spectrum is taken at
1.46 V so that the bands are inverted, i.e. going up
rather than going down.) Bands at about 1719 and 2027
cm−1 are evident due to the presence of the carbonyl
moiety of a ꢀCHO (aldehyde) or ꢀCOOH (acid) inter-
mediate and CO, respectively. A band at 2345 cm−1
due to CO2 could also be expected but the OD stretch-
ing bands due to HDO and D2O solution species which
occurs in the same region are very intense to effectively
mask the signal due to CO2. The presence of the
adsorbed species on the electrode surface is further
confirmed in a SNIFTIRS experiment. Fig. 4 shows the
results of such a measurement. Again, bands at 1706–
1713 and about 1994–2031 cm−1 (pseudo-derivative in
shape due to modulation) are observed. FTIR spec-
troelectrochemical studies of Au, Pd, Ru, and Rh also
showed the production of CO2 from the oxidation of
CH4. To obtain better sensitivity the SEIRAS technique
was employed as far as practicable using the electrode-
position technique. Fig. 5 shows cyclic voltammograms
of a smooth Ru electrode (solid line) compared to that
of ruthenium electrodeposited (ruthenized) on a Ru disc
(broken line). The considerable increase in current is
evident and is presumably due to the increased surface
area of the electrodeposited electrode. Fig. 6 shows
PDIR spectra on the smooth Ru disc as a function of
potential following the adsorption of CH4. A band at
2345 cm−1 is observed due to the production of CO2.
The upward-going band at about 1650 cm−1 is pre-
sumably due to water; there also appears to be a
downward-going band at about 1600 cm−1. For the
ruthenized Ru electrode, an additional band is observed
at about 2047 cm−1 which is assigned to adsorbed CO
(Fig. 7). There are other bands in the spectral region
from 1550 to 1850 cm−1. Adsorption and oxidation of
CH4 on smooth polycrystalline Pd showed only the
production of CO2. The use of Pd electrodeposited on
GC to obtain IR surface enhancement, allowed the
observation of a band at 2025 cm−1 due to CO. CO2 at
2345 cm−1 and an unknown species at about 1600
cm−1 were observed on both smooth and roughened
polycrystalline Au electrodes, as illustrated in Fig. 8. In
Table 1 we summarize the main results of spectroelec-
trochemical experiments on the various metals which
we have studied. We have included data obtained on
the use of GC as a substrate in order to serve as a
reference for comparing the results on palladium elec-
trodeposited on GC. It was also of interest to see the
behavior of a non-catalytic surface for methane elec-
trooxidation, in comparison with noble metals which
are generally regarded as electrocatalytically active.
4. Discussion
4.1. Adsorption and oxidation of methane on Pt
Fig. 1. Cyclic voltammograms of Pt in 0.5 M HClO4 deaerated
solution (solid line) and after bubbling CH4 through the
solution for 2 h (broken line) at 0.26 V; cyclic voltammogram
of platinized Pt (dotted line); scan rate=100 mV/s.
Most of the studies on the electrochemical oxidation
of methane have been conducted at temperatures above
ambient because of the molecule’s relative inertness and

3528
F. Hahn, C.A. Melendres / Electrochimica Acta 46 (2001) 3525–3534
Fig. 2. ‘In situ’ LPS-FTIR spectra from a Pt disc electrode as a function of potential following adsorption of CH4 at 0.16 V for 45
min (referenced to reflectance at 0.16 V).
difficulty to activate. The effective utilization of meth-
ane in fuel cells has therefore been confined so far to
temperatures above 500 °C [20]. Without doubt, there
is great economic incentive to find an electrocatalyst
which would activate methane at room temperature.
The complete anodic oxidation reaction involved is
normally given as
tion process. Thus, Marvet and Petrii [21] have sug-
gested that the equilibrium potentials observed upon
adsorption of CH4 on the platinized Pt in H2SO4
solutions is due to the dehydrogenation of CH4 result-
ing in a shift of the electrode potential to the hydrogen
region. From a study of the charge associated with the
electrooxidation of the adsorbed species, they suggested
that the latter contains an HCO entity. Niedrach and
Tochner [9] have likewise suggested a C1 oxygenated
hydrocarbon species to be the intermediate in the an-
odic oxidation of CH4 and C2H6 from an interpretation
of the results of their potentiodynamic measurements in
phosphoric acid electrolyte. Hsieh and Chen [11] pro-
posed a mechanism involving the dissociative adsorp-
tion of CH4 forming adsorbed CH3 whose further
conversion to an intermediate constitutes the rate deter-
mining step in the overall process of oxidation to CO2.
The mechanism was found to be consistent with the
authors’ results of the polarization measurements in
acid sulfate electrolyte at 80 °C. Sustersic et al. [12]
postulated the adsorption of two types of species, COH
(or COOH) and CO, in order to explain their potentio-
CH4+2H2OUCO2+8H++8e− E°=0.146 V (1)
It is easy to see from the above equation that the
reaction cannot be a simple one, since it involves water
as a co-reactant and requires the transfer of eight
electrons. A catalytic electrode with activity for the
adsorption of CH4 and water (or OHads) is thus essen-
tial in order to effectively use methane in fuel cells.
The adsorption and subsequent anodic oxidation of
CH4 on Pt at temperatures above ambient have been
established by a number of workers using various elec-
trochemical techniques [8,11,12]. From these measure-
ments, attempts have been made to deduce the nature
of the adsorbed species on the electrode surface and
proposals made on the mechanism of the electrooxida-

F. Hahn, C.A. Melendres / Electrochimica Acta 46 (2001) 3525–3534
dynamic measurements in 1 N H2SO4 solution at
3529
60 °C. Needless to say, structural information from the
spectroscopic measurements are essential in order to
confirm any of the mechanisms that have been pro-
posed. Berthelot [22] has attempted to make such mea-
surements on the carbon-supported noble metal
catalysts and their mixtures. However, the spectra were
often not of sufficiently high signal to noise ratio to
give definitive assignments of the bands to various
species.
Our CV results (Fig. 1, broken line) are consistent
with those obtained by others, although the anodic
wave associated with the oxidation process is not as
well defined at 25 °C as those observed at higher
temperatures [8]. The first scan after bubbling CH4
through the solution shows partial suppression of hy-
drogen adsorption indicating the presence of another
species on the electrode surface. A search for the vibra-
tional bands of adsorbed CH4 or CH3, did not yield
any spectrum. From our own measurements on pure
CH4, bands could be expected at about 3016 and 1304
cm−1, corresponding to the CꢀH stretching vibrational
mode and the HꢀCꢀH bending mode, respectively. The
absence of such bands might indicate that the adsorbed
Fig. 4. ‘In situ’ IR spectra from a platinized Pt in 0.5 M
HClO4 with D2O obtained by the SNIFTIRS technique; CH4
adsorbed at 0.26 V for 2 h; modulation potentials indicated in
brackets.
CH4 quickly dissociates to some other species which in
turn reacts with water (or OH) to form a CO-contain-
ing surface intermediate. This would be consistent with
Fig. 5. Cyclic voltammograms of a polycrystalline Ru disc
electrode (solid line) and a ruthenized Ru electrode (broken
line) in 0.5 M HClO4 solution; scan rate=100 mV/s.
Fig. 3. ‘In situ’ LPS-FTIR spectra from a platinized Pt in 0.5
M HClO4 with D2O; CH4 adsorbed at 0.26 V for 1 h (refer-
enced to reflectance at 1.46 V).

3530
F. Hahn, C.A. Melendres / Electrochimica Acta 46 (2001) 3525–3534
a fast dehydrogenation step as Marvet and Petrii [21]
have suggested. However, Hsieh and Chen’s [11] pro-
posal on the adsorption of CH3 cannot be confirmed.
Our LPS-FTIRS study showed the formation of CO2
starting at about 500 mV, as indicated by the appear-
ance of a band at 2345 cm−1 and possibly also CO at
about 2020 cm−1 (Fig. 2). The use of SEIRAS with
platinized Pt [17] provided for increased sensitivity.
With the use of deuterated water as solvent, we found
bands at about 1730 and 2027 cm−1 (Fig. 3). We assign
the former to the vibration of a CO moiety in a ꢀCHO
(aldehyde) or ꢀCOOH (acid) species; the 2027 cm−1
band is due to the CO adsorbed on the Pt surface. The
width of the CO band is rather broad and may be due
to the presence of linearly bonded CO (commonly
designated as COL, with the C atop the Pt) as well as
bridge-bonded CO (or COB). These species generally
give vibrational frequencies at about 2050 and 1950
cm−1, respectively [23]. There may also be contribution
to line broadening due to the heterogeneity of the
surface sites on the polycrystalline Pt electrode. Most
interestingly, the bands are already observed at 260 mV
(Table 1) which suggest that they are likely to be
intermediates in the conversion of CH4 to CO2. The
reality of these bands was confirmed by SNIFTIRS
measurements as a function of potential and with a
modulation amplitude of 300 mV. Fig. 4 shows spectral
bands at about 1706 and 2030 cm−1 with relatively
good signal to noise ratio.
The results which we have obtained so far tend to
support theories suggested by earlier workers
[9,11,12,21], remarkably made on the basis of the elec-
trochemical measurements alone, that CH4 adsorbs dis-
sociatively on the Pt surface and that a C1 oxygenated
hydrocarbon species is formed on anodic oxidation of
CH4. There have been mechanistic models proposed
[22,24] to explain the anodic oxidation of CH4.
Bagotzky et al. [24] have suggested a generalized
chemisorption scheme wherein the oxidation of a sim-
ple organic molecule involves a series of dehydrogena-
tion steps and reaction with adsorbed OH radicals. The
latter provides the essential step to introduce the oxy-
gen into the hydrocarbon chain and the CO moiety.
Berthelot [22] has similarly found CO2 to be the princi-
pal product of CH4 electrooxidation on Pt and has
proposed a number of steps involved in the oxidation
of C−4 in CH4 to C+1 in CHO and subsequently from
C
+2 in CO to C+4 in CO2. Hydrogenated free radicals,
Fig. 6. ‘In situ’ PDIR spectra from a Ru disc electrode; CH4 adsorbed at 0.1 V for 2 h (referenced to reflectance at 0.1 V).

F. Hahn, C.A. Melendres / Electrochimica Acta 46 (2001) 3525–3534
(CO)ads+H2OUCO2+2H++2e−
3531
(5)
In any case, it is most likely that other intermediates
are involved in the multiple steps and further elucida-
tion of the mechanism of CH4 oxidation will have to
await the identification of these species. It is also appar-
ent that the discovery of the intermediates would re-
quire even better sensitivity in the technique being used.
We are trying to explore the use of differential electro-
chemical mass spectroscopy to see if other reaction
products could be identified and also to further confirm
the results obtained in the present work.
4.2. In6estigations on other noble metals
Studies on the oxidation of CH4 at Au, Pd, Rh and
Ru electrodes showed that the activity of the metals
toward the electrooxidation process varies significantly
as can be discerned from the spectroelectrochemical
data presented in Table 1. CO2 has been observed as
the main oxidation product for all. From the value of
the potential at which CO2 is first observed, it can be
surmised that Pt and Ru are the most effective electro-
catalysts. The low activity of Au, wherein the appear-
ance of CO2 is first observed by IR at 0.9 V, was not
unexpected. It is consistent with the results obtained by
others on the high anodic potential at which CO oxida-
tion occurs on Au in acid solutions [25,26]. We also
observe a band at about 1600 cm−1 in the case of Au
and Ru electrodes. This frequency is too low to be
assigned to some ꢀCO moiety or water; we have no
satisfactory explanation for its origin at this time. Inter-
estingly for comparison, GC shows the highest overpo-
tential and can be said to have no electrocatalytic
activity for the anodic oxidation of CH4. Ru appears to
have a slightly higher activity for the production of
CO2 than Pd. CO has been found as an intermediate in
the case of Pt, Ru and Pd, where increased detection
sensitivity has been possible using SEIRAS. Attempts
to detect ꢀCHO or ꢀCOOH on electrodeposited Ru and
Pd using SNIFTIRS and with D2O solutions did not
yield definitive results.
Fig. 7. ‘In situ’ PDIR spectra from a ruthenized Ru electrode;
CH4 adsorbed at 0.1 V for 2 h (referenced to reflectance at 0.1
V).
as well as C1 alcohols, and acids have been postulated
as intermediates formed. The detection of these species
spectroscopically has eluded us so far. Our observation
on the presence of CO and ꢀCHO (or ꢀCOOH) at
relatively low potentials (0.26 V) does suggest the in-
volvement of a hydroxylated species (H2O or OHads
)
early in the dissociation/oxidation reaction of CH4.
This is rather surprising. CO and ꢀCHO are the same
species that have been observed in the anodic oxidation
of methanol. The formation of CHO should normally
precede that of CO and ꢀCOOH. We are unable to
distinguish between ꢀCHO and ꢀCOOH at this time.
The simultaneous observation of ꢀCHO and CO may
be suggestive of two different oxidation pathways, e.g.
4.3. Utility of the SEIRAS technique
It is worthwhile to point out here the utility of the
SEIRAS technique because it has not been as widely
recognized as the corresponding SERS technique.
Hatta et al. [27] first applied the SEIRA technique to
the study of the electrochemical interface. Osawa et al.
[28] subsequently studied the SEIRA effect more exten-
sively. The enhancement factors obtained in SEIRA are
not as high as for SERS, but they are nevertheless
sufficient to allow for the identification of weak signals.
A comparison of the intensity of the CO bands at about
2027 cm−1 in Figs. 2 and 3 suggests an enhancement of
CH4+H2OU(CHO)ads+5H++5e−
CH4+H2OU(CO)ads+6H++6e−
(2)
(3)
Still, such mechanisms will involve the simultaneous
transfer of five or six electrons and this is deemed
highly improbable. Even the further oxidation of ad-
sorbed CO and CHO species to CO2 would require
three and two electrons, respectively, i.e.
(COH)ads+H2OUCO2+3H++3e−
(4)

3532
F. Hahn, C.A. Melendres / Electrochimica Acta 46 (2001) 3525–3534
Fig. 8. ‘In situ’ LPS-FTIR spectra from a smooth Au electrode; CH4 adsorbed at 0.0 V for 1 h (referenced to reflectance at 0.0 V).
about ten times have been obtained by electrodeposi-
tion in the present work. This maybe accounted for by
the increased surface area of the electrodeposited mate-
rial. One can see this from a comparison of the cur-
rent–potential curves for the platinized Pt (dotted line)
with that of the smooth Pt disc (full line) in Fig. 1. The
magnitude of the currents for the former is nearly
about 20 times higher than for the Pt disc electrode.
The shape of the cyclic voltammogram is very similar.
The increase in current is obviously a result of the
increased surface area of the electrodeposited material.
A similar increase accompanied the electrodeposition of
Ru on a ruthenium disc electrode. The difference in the
magnitudes of the currents for the two electrodes (Fig.
Table 1
Summary of spectroelectrochemical results
Metal electrode
Adsorption
Time (h)
Potential (V) at which product first
Product
Type of
potential (V)
observed
experiment
Pt
0.16
0.26
0.26
0.1
0.1
0.3
0.3
0.1
0.0
0.1
0.75
1.0
2.0
2.0
2.0
1.0
3.0
1.0
1.0
3.0
0.56
0.26
0.26
0.6
0.5
0.8
0.6
0.7
0.9
1.3
CO2
LPS-FTIRS
LPS-FTIRS
SNIFTIRS
PDIRS
Pt/Pt (in D2O)
Pt/Pt (in D2O)
Ru
Ru/Ru
Pd
Pd/GC
Rh
Au
ꢀCHO, CO
ꢀCHO, CO
CO2
CO
CO2
CO
CO2
CO2
CO2
PDIRS
LPS-FTIRS
LPS-FTIRS
LPS-FTIRS
LPS-FTIRS
LPS-FTIRS
GC

F. Hahn, C.A. Melendres / Electrochimica Acta 46 (2001) 3525–3534
3533
5) is again mainly due to the increased surface area
associated with the higher roughness factor for the
ruthenized electrode. The same could be said for the Pd
electrodeposited on GC. Thus, it does not appear nec-
essary to use a model of plasmon resonance at the
surface of the very small metal islands or particles as
have been generally invoked by others [17,27,28]. This
is not to say that such a phenomenon does not exist or
is not operative here also. Attempts were made to
utilize the SEIRA effect on gold by roughening the
electrode surface in KCl solution following the proce-
dure used by Weaver and coworkers [19] to obtain
SERS. However, these have not been successful. Sur-
prisingly, we found no increase in the surface area of
the electrode following the roughening procedure. It
appears therefore not unreasonable to find no SEIRA
in the case of Au. It can be argued that maybe the
length scale involved in SERS is not the same as for
SEIRAS. It is rather difficult to accept since Ataka et
al. [29] have effectively utilized the SEIRA effect on the
electrodeposited Au for their investigations on potential
dependent water reorientation at the electrode/elec-
trolyte interface. Finally, we might mention here that
we have not attempted to optimize the electrodeposi-
tion conditions, as well as the morphology and size of
the electrodeposited metals. Further gain in enhance-
ment factors could possibly result from such a study.
Suffice it to say, that we found the SEIRAS technique
to be a most useful adjunct in infrared spectroscopy
and the method of electrodeposition is particularly
convenient for the preparation of appropriate electrode
materials.
also be worthwhile to study and would be of great
practical importance for the development of a commer-
cially viable fuel cell system. Finally, the role of water
continues to pose a most interesting fundamental ques-
tion that remains unanswered to date.
Acknowledgements
The financial support of the Universite´ de Poitiers
and the CNRS-UMR 6503 are gratefully acknowl-
edged. One of us (C.A.M.) thanks Professor C. Lamy
for support during a sabbatical year that made this
work possible.
References
[1] A.J. Appleby, F.R. Foulkes, Fuel Cell Handbook, Van
Nostrand Reinhold, New York, 1989.
[2] K.R. Williams, M.R. Andrew, F. Jones, in: B.S. Baker
(Ed.), Hydrocarbon Fuel Cell Technology, Academic
Press, New York, 1965.
[3] N.A. Hampton, M.J. Willars, B.D. McNicol, J. Power
Sources 1 (1981) 191.
[4] R. Monastersky, Sci. News 154 (1998) 312.
[5] K.V. Kordesh, J. Electrochem. Soc. 125 (1978) 77C.
[6] S. Srinivasan, R. Mosdale, Bull. Electrochem. 12 (1996)
170.
[7] L.W. Niedrach, J. Electrochem. Soc. 111 (1964) 1309.
[8] L.W. Niedrach, J. Electrochem. Soc. 113 (1966) 645.
[9] L.W. Niedrach, M. Tochner, J. Electrochem. Soc. 114
(1967) 17.
[10] S. Berthelot, E. Gehain, F. Hahn, J.M. Le´ger, S. Srini-
vasan, C. Lamy, Extended Abstracts, Electrochemical
Society Meeting, Boston, 98-2, 1998, Abstract 1090.
[11] S.Y. Hsieh, K.M. Chen, J. Electrochem. Soc. 124 (1977)
1171.
5. Conclusions
[12] M.G. Sustersic, R. Cordova, W.E. Triaca, A.J. Arvia, J.
Electrochem. Soc. 127 (1980) 1242.
[13] (a) B. Beden, C. Lamy, in: R.J. Gale (Ed.), Spectroelec-
trochemistry, Theory and Practice, Plenum Press, New
York, 1988;
(b) See also, H. Druliolle, K.B. Kokoh, F. Hahn, C.
Lamy, B. Beden, J. Electroanal. Chem. 426 (1997) 103.
[14] G. Orozco, C. Gutierrez, J. Electroanal. Chem. 484 (2000)
64.
[15] D.S. Corrigan, L.-W.H. Leung, M.J. Weaver, Anal.
Chem. 59 (1987) 2252.
[16] M.A. Habib, J.O’M. Bockris, J. Electroanal. Chem. 180
(1984) 287.
[17] A. Bjerke, P.R. Griffiths, W. Theiss, Anal. Chem. 71
(1999) 1967.
[18] Z.Q. Tian, J.S. Gao, X.Q. Li, B. Ren, Q.J. Juang, W.B.
Cai, F.M. Liu, B.W. Mao, J. Raman Spectrosc. 29 (1998)
703.
[19] H.Y.H. Chan, S. Zhou, M.J. Weaver, J. Phys. Chem. B
103 (1999) 11141.
The spectroscopic identification of the products of
chemisorption and oxidation of methane at noble metal
electrodes has been carried out using infrared spec-
troscopy, with and without surface enhancement via the
electrodeposition of the metal. In all cases, the final
product observed was CO2. CO and ꢀCHO (or
ꢀCOOH) have been identified as reaction intermediates
on Pt, presumably arising from the dissociative adsorp-
tion of CH4 at the electrode surface with concomitant
reaction with water. The mechanism of oxidation ap-
pears to be similar to that of methanol in that CO and
ꢀCHO (or ꢀCOOH) are produced. The associated prob-
lem of electrode poisoning will therefore have to be
dealt with. Activation of methane at room temperature
is nevertheless possible using noble metals as electrocat-
alysts. The most active appears to be Pt and Ru. It
would be interesting, in future work, to examine the
effect of alloying (e.g. Pt–Ru and Pt–Pd) on the elec-
trocatalytic activity. The effect of temperature on the
rates and mechanisms of CH4 electrooxidation should
[20] M. Stoukides, J. Appl. Electrochem. 25 (1995) 899.
[21] R.V. Marvet, O.A. Petrii, Sov. Electrochem. 3 (1966) 127.

3534
F. Hahn, C.A. Melendres / Electrochimica Acta 46 (2001) 3525–3534
[22] S. Berthelot, PhD Thesis, Universite´ de Poitiers, 1998, p.
135.
[26] H. Nakajima, H. Kita, K. Kunimatsu, A. Aramata, J.
Electroanal. Chem. 201 (1986) 175.
[23] M.I. Lopes, I. Fonseca, P. Olivi, B. Beden, F. Hahn, J.M.
Le´ger, C. Lamy, J. Electroanal. Chem. 346 (1993) 415.
[24] V.S. Bagotzky, Y.B. Vassiliev, O.A. Khazova, J. Elec-
troanal. Chem. 81 (1977) 229.
[25] B. Beden, A. Bewick, K. Kunimatsu, C. Lamy, J. Elec-
troanal. Chem. 142 (1982) 345.
[27] A. Hatta, Y. Chiba, W. Suetaka, Appl. Surf. Sci. 25
(1986) 327.
[28] M. Osawa, K. Ataka, K. Yoshii, T. Yotsuyanagi, J.
Electron Spectrosc. Relat. Phenom. 64/65 (1993) 371.
[29] K. Ataka, T. Yotsuyanagi, M. Osawa, J. Phys. Chem. 100
(1996) 10664.