In situ FTIR spectroscopic studies of (bi)sulfate adsorption on electrodes of Pt nanoparticles supported on different substrates

Article abstract of DOI:10.1016/j.electacta.2009.11.035

Nanostructured Pt electrodes were prepared by electrodeposition of Pt nanoparticles on different substrates (GC, Pt and Au) under cyclic voltammetric conditions and with various number (n) of potential cycling, and were denoted as nm-Pt/S(n) (S = GC, Pt and Au). Adsorption of (bi)sulfate on the nm-Pt/S(n) was studied by in situ FTIR reflection spectroscopy. It has been revealed that the nanostructured Pt electrodes exhibit anomalous IR properties for (bi)sulfate adsorption regardless of the different reflectivity of substrate, i.e. the IR absorption of (bi)sulfate species adsorbed on all the nm-Pt/S(n) electrodes is significantly enhanced and the IR band direction is completely inverted in comparison with the same species adsorbed on a bulk Pt electrode. The two IR bands around 1200 and 1110 cm^-1 attributed to adsorbed (bi)sulfate species are shifted linearly with increasing electrode potential, yielding Stark tuning rates (d over(ν, ?) / d E_S) of 152.1 and 21.1 cm^-1 V^-1 on nm-Pt/GC(20), respectively. Along with increasing n, the Stark tuning rate of the IR band around 1200 cm^-1 decreases quickly and declined to 7.6 cm^-1 V^-1 on nm-Pt/GC(80), while the Stark tuning rate of the IR band near 1100 cm^-1 is fluctuated between 23.0 and 16.2 cm^-1 V^-1. It has determined that the enhancement of IR absorption of (bi)sulfate adsorbed on nanostructured Pt electrode is varied with substrate material and n, and a maximal 16-fold enhancement of the IR band near 1200 cm^-1 has been measured on the nm-Pt/GC(30) electrode. The in situ FTIR studies illustrated that the adsorption of (bi)sulfate occurs mainly in the double layer potential region, and reaches a maximum around 0.80 V. The results demonstrated also that the competitive adsorption of CO and oxygen species can inhibit completely (bi)sulfate adsorption, which has evidenced a weak interaction of (bi)sulfate with nm-Pt/S(n) electrode surface.

Full text of DOI:10.1016/j.electacta.2009.11.035

Electrochimica Acta 55 (2010) 2065–2072
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
In situ FTIR spectroscopic studies of (bi)sulfate adsorption on electrodes of Pt
nanoparticles supported on different substrates
Dong-Mei Zeng, Yan-Xia Jiang, Zhi-You Zhou, Zhang-Fei Su, Shi-Gang Sun∗,1
State Key Laboratory of Physical Chemistry of Solid Surfaces, Department of Chemistry, College of Chemistry and Chemical Engineering,
Xiamen University, Xiamen 361005, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 21 July 2009
Received in revised form 8 November 2009
Accepted 11 November 2009
Available online 4 December 2009
Nanostructured Pt electrodes were prepared by electrodeposition of Pt nanoparticles on different sub-
strates (GC, Pt and Au) under cyclic voltammetric conditions and with various number (n) of potential
cycling, and were denoted as nm-Pt/S(n) (S = GC, Pt and Au). Adsorption of (bi)sulfate on the nm-Pt/S(n)
was studied by in situ FTIR reflection spectroscopy. It has been revealed that the nanostructured Pt elec-
trodes exhibit anomalous IR properties for (bi)sulfate adsorption regardless of the different reflectivity
of substrate, i.e. the IR absorption of (bi)sulfate species adsorbed on all the nm-Pt/S(n) electrodes is
significantly enhanced and the IR band direction is completely inverted in comparison with the same
species adsorbed on a bulk Pt electrode. The two IR bands around 1200 and 1110 cm−1 attributed to
adsorbed (bi)sulfate species are shifted linearly with increasing electrode potential, yielding Stark tun-
ing rates (dꢀ˜/dES) of 152.1 and 21.1 cm−1 V−1 on nm-Pt/GC(20), respectively. Along with increasing n,
the Stark tuning rate of the IR band around 1200 cm−1 decreases quickly and declined to 7.6 cm−1 V−1
on nm-Pt/GC(80), while the Stark tuning rate of the IR band near 1100 cm−1 is fluctuated between 23.0
and 16.2 cm−1 V−1. It has determined that the enhancement of IR absorption of (bi)sulfate adsorbed on
nanostructured Pt electrode is varied with substrate material and n, and a maximal 16-fold enhancement
of the IR band near 1200 cm−1 has been measured on the nm-Pt/GC(30) electrode. The in situ FTIR stud-
ies illustrated that the adsorption of (bi)sulfate occurs mainly in the double layer potential region, and
reaches a maximum around 0.80 V. The results demonstrated also that the competitive adsorption of CO
and oxygen species can inhibit completely (bi)sulfate adsorption, which has evidenced a weak interaction
of (bi)sulfate with nm-Pt/S(n) electrode surface.
Keywords:
(Bi)sulfate
Adsorption
Pt nanoparticles
Anomalous IR property
In situ FTIRS
© 2009 Elsevier Ltd. All rights reserved.
1. Introduction
properties are strongly dependent on the structure of nanomateri-
als.
Nanomaterials exhibit peculiar properties due to their nano-size
effects, surface effects, quantum optics effects, etc., and display
potential applications in electronics [1], optics [2], catalysis [3],
magnetic data storage [4], and so on. Investigation of the IR
peculiar properties is of importance to reveal the fundamental
aspects of nanomaterials. The optical properties of nanomaterials
were extensively studied in recent years by using vibration spec-
troscopy, such as the surface-enhanced Raman scattering (SERS)
[5], surface-enhanced IR absorption spectroscopy (SEIRAS) [6],
surface-enhanced second harmonic generation (SESHG) [7] and
surface-enhanced sum frequency generation (SESFG) [8]. The for-
mer two optical properties are considered to arise from the size and
separation of surface nanoscale islands, while the latter two optical
It has been reported that the electrocatalysts of metal nano-
materials exhibit, in general, anomalous infrared property [9–17],
which consists in the inversion of the direction of IR band of adsor-
bates (anomalous absorption), the enhancement of IR absorption
and the broadening of IR band width (inhomogeneous broadening)
in comparison with the same species adsorbed on a correspond-
ing bulk metal surface. Previous studies concerning the anomalous
IR property of nanomaterials were conducted mostly by using CO
adsorption as a probe molecule. It is known that the CO adsorp-
tion on transition metals leads to the formation of a 5␴ bond
together with a d–␲* back donation [18], which results in a strong
interaction. Besides, IR absorption of CO species appears around
2100–1800 cm−1 region depending on their coordination mode to
surface, i.e. the highest IR energy region provided by modern FTIR
instruments. Such high IR energy gives rise to almost the most
sensitive spectral response in the mid-infrared range [19,20]. It is
worthwhile to investigate the anomalous IR property of nanoma-
terials using weak adsorbates such as (bi)sulfate species as probes
∗
Corresponding author. Tel.: +86 592 2180181; fax: +86 592 2183047.
E-mail address: sgsun@xmu.edu.cn (S.-G. Sun).
1
ISE member.
0013-4686/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2009.11.035

2066
D.-M. Zeng et al. / Electrochimica Acta 55 (2010) 2065–2072
that yield IR absorption in the fingerprint region (1500–800 cm−1),
where most organic fuel molecules give rise to characteristic IR
absorption. It may be expected that such investigation may provide
essential significances in three aspects: (1) to explore the anoma-
lous IR property of nanomaterials in different conditions other than
those in previous studies employed probe molecules that yielded
a strong interaction with electrode surface [9–17]; (2) to study
the adsorptive behaviors of sulfuric acid, which is the most widely
used electrolyte in electrochemical systems, on nanostructured Pt
electrodes; and (3) to extend the anomalous IR property of nano-
materials in studies of electrocatalysis of direct organic molecule
fuel cells, because nanomaterials such as Pt or Pt alloy nanoparti-
cles supported on carbon are widely used electrocatalysts, and the
anomalous IR features of the inversion of IR band direction and the
enhancement of IR absorption are very useful in determining inter-
mediates and products involved in oxidation reaction of organic
fuel molecules.
Extensive studies have been done so far concerning (bi)sulfate
adsorption on electrode surface, mostly on noble metals, such as Pt
[21–43], and Au [44], and the investigations of (bi)sulfate are mostly
focused on bulk metal electrodes. The studies have demonstrated
thatthe (bi)sulfate adsorptioncanaffect drasticallyelectrochemical
reactions of methanol oxidation [45], H2 oxidation [45], O2 reduc-
tion [46], CO2 reduction [47], coadsorption of water [48,49], and the
underpotential deposition (UPD) of Cu [50], etc. Numerous meth-
ods have been used in past decades to investigate the adsorption of
(bi)sulfate on metal electrodes, among them infrared spectroscopy
has provided the most important information about adsorbates of
(bi)sulfate at molecule level [22–29,34–41]. As the infrared bands
of adsorbed (bi)sulfate on a bulk metal surface and those of solu-
tion (bi)sulfate species appear in the same direction and at similar
wavenumbers in IR spectra, it is difficult to discern the IR bands
arising from adsorbed (bi)sulfate, which leads to the complication
of assigning the adsorbed (bi)sulfate species and of understand-
ing the adsorptive behaviors of (bi)sulfate. Undoubtedly, further
study of (bi)sulfate on surface of nonmaterial electrodes would be
very significant in order to reveal the adsorptive properties of sul-
furic acid electrolyte at molecule level, if nanostructured electrodes
also exhibit anomalous infrared property for such weakly adsorbed
species.
trodeposition. Nanostructured Pt electrodes prepared on different
substrates and with various n are denoted hereafter as nm-Pt/S(n),
where S stands for substrate of GC, Pt or Au, respectively. The sub-
strate was sealed into a Teflon holder and polished mechanically
using sand paper (6#) and alumina powder of size 5, 1 and 0.3 ␮m,
then cleaned in an ultrasonic bath, and finally subjected to poten-
tial cycling at 0.10 V s−1 in 0.1 M H2SO4 solution to remove possible
impurity on the surface before Pt nanoparticles deposition. The
surface structure of the as-prepared electrode was characterized
by scanning electron microscope (SEM) on a LEO-1530 electron
microscopy.
All solutions were prepared with super pure H2SO4 and Mil-
lipore water (18 Mꢁ cm) supplied from a Milli-Q lab apparatus
(Nihon Millipore Ltd.). The solutions were deaerated by bubbling N2
before measurement. A saturated calomel electrode (SCE) was used
as reference electrode, and a platinized platinum foil was served as
counter electrode. In order to facilitate the comparison of results of
this paper with those reported in the literatures [21–44], the elec-
trode potentials recorded versus the SCE were first converted into
the reversible hydrogen electrode (RHE) scale, and then presented
in this paper. A 263 potentiostat/galvanostat (EG&G) was used to
control the potential of electrode. All measurements were carried
out at room temperature.
Electrochemical in situ FTIR reflection spectroscopy measure-
ments were conducted on a Nexus 870 spectrometer equipped with
a liquid-nitrogen-cooled MCT-A detector. A CaF2 disk was used as
IR window. A thin layer configuration was approached by pushing
the electrode against the IR window during FTIR measurements.
The incident IR beam was aligned at about 60◦ to the normal of
electrode surface.
In situ FTIR spectra were collected using SNIFTIRS (sub-
tractively normalized interfacial FTIR spectroscopy) procedure
[22–29,34–41]. The result spectra were reported as the relative
change in reflectivity and calculated as follows:
ꢂR
R
R(ES) − R(ER)
R(ER)
=
(1)
where R(ES) and R(ER) are single-beam spectra of reflection
collected at sample potential ES and reference potential ER,
respectively. Each single-beam spectrum was co-added with 1000
interferograms at a spectral resolution of 8 cm−1. As in the SNIFTIRS
procedure, the collection of interferograms was done alternatively
at ES and ER. In order to study clearly the IR features of (bi)sulfate
adsorbates, an appropriate ER is crucial. We have compared a series
of ER and found that a maximal intensity of IR bands of adsorbed
(bi)sulfate can be obtained when the ER is set at 0.24 V (see details
in Fig. 5). So the 0.24 V was used as reference potential throughout
all in situ FTIRS studies of (bi)sulfate adsorption on nanostructured
Pt electrodes, which coincides with the literatures [30–32] that at
such low potential almost no anions are adsorbed on Pt electrode
surface.
The present paper has put emphasis upon the behaviors of
(bi)sulfate adsorption on Pt nonmaterial electrodes. Nanostruc-
tured Pt electrodes were prepared by electrodeposition of Pt
nanoparticles on different substrates under cyclic voltammetric
conditions. In situ FTIR spectroscopy was employed to investigate
(bi)sulfate adsorption on the nanostructured Pt electrodes. The
results demonstrated that, for the first time, the nanostructured Pt
electrodes exhibit also anomalous IR property for weakly adsorbed
(bi)sulfate species that yield IR absorption in fingerprint region,
at which the IR energy is much lower than that in the 2000 cm−1
region. The in situ FTIR studies illustrated that the adsorption of
(bi)sulfate occurs mainly in the double layer potential region on
nanostructured Pt electrodes, and reaches a maximum around
0.80 V.
3. Results and discussion
3.1. Cyclic voltammetric (CV) studies and SEM characterization
2. Experimental
A stable cyclic voltammogram (CV) of nm-Pt/GC(n) electrode
in 0.1 M H2SO4 is shown in Fig. 1a. The potential scan range
includes three regions, i.e. hydrogen region (0.03–0.40 V), double
layer region (0.40–0.75 V) and oxygen region (0.75–1.55 V). Fig. 1a
is similar to the cyclic voltammogram of a bulk Pt electrode [31].
We observe two pairs of hydrogen adsorption/desorption current
peaks around 0.12 and 0.18 V in the hydrogen region, respectively.
In the positive-going potential scan (PGPS), oxygen adsorption cur-
rent appears in the oxygen region as a broad peak from 0.75 to
1.17 V and follows a plateau; and in the negative-going potential
Nanostructured Pt electrodes were prepared through electrode-
position of Pt nanoparticles on different substrates (glassy carbon
(GC), Pt and Au of 5 mm in diameter). The deposition was carried
out in 0.002 M K2PtCl6 + 0.1 M H2SO4 solution under cyclic voltam-
metric conditions, in which the electrode potential was cycled
between −0.25 V and 0.40 V (vs. SCE, 0.05–0.70 V (RHE)) at a scan
rate 0.05 V s−1. The size and density of deposited Pt nanoparticles
depend on the numbers (n) of potential cycles applied in elec-

D.-M. Zeng et al. / Electrochimica Acta 55 (2010) 2065–2072
2067
Fig. 1. (a) Cyclic voltammograms of nm-Pt/GC(n) electrode in 0.1 M H2SO4, potential
scan rate: 0.05 V s−1; (b) relationship between Rr and n.
scan (NGPS) a reduction current peak is observed near 0.74 V. The
amplitude of all these current peaks increases with the increase of
n. Cyclic voltammograms recorded on an nm-Pt/Pt(n) and an nm-
Pt/Au(n) display similar features. While in the case of nm-Pt/Au(n)
electrodes an additional reduction current peak is observed near
1.17 V in the NGPS, which is attributed to the oxide reduction of Au
substrate. The appearance of this current peak due to the reduction
of surface Au oxide indicates that the nanostructured Pt film is so
porous (as seen in SEM images below) that the Au substrate can be
oxidized at high electrode potentials. It is worthwhile to note that
no faraday current except the current of double layer charging is
determined in the double layer region.
As a matter of fact, a polished Pt electrode will always have a
roughness higher than 1. In the present study, a roughness of ca.
2.0 has been measured from the ratio of electro-active surface area
to geometrical surface area. The electro-active surface area is cal-
culated by QH/210 ␮C cm−2, where QH is the hydrogen adsorption
charge measured by integration of the CV curve of the polished Pt
electrode, and the value 210 ␮C cm−2 is taken for adsorption of a
monolayer of hydrogen on a smooth polycrystalline Pt electrode
[31]. To normalize the adsorption surface sites of the nm-Pt/S(n)
referring to the polished Pt, we define the relative roughness (Rr)
of the nm-Pt/S(n) electrodes by comparing the electro-active area
of nm-Pt/S(n) electrodes with that of the polished Pt electrode, i.e.
Fig. 2. SEM images of (a) nm-Pt/GC(20), (b) nm-Pt/GC(40) and (c) nm-Pt/GC(80)
electrodes, the inserts are the corresponding size histograms of Pt nanoparticles of
different structure: congeries of polyhedral particles and cube-like particles.
The relationship between Rr and n acquired with nm-Pt/GC(n)
electrodes is plotted in Fig. 1b. It illustrates clearly a linear varia-
tion of Rr with n. Such linear relationship between Rr and n is also
maintained for nm-Pt/Pt(n) and nm-Pt/Au(n) electrodes.
The SEM images in Fig. 2 illustrate that the nm-Pt/GC(n = 20,
40 and 80) electrodes are composed of Pt nanoparticles dispersed
on GC substrate. These Pt nanoparticles may be classified into
two categories, i.e. cube-like particles and congeries of polyhe-
dra as seen clearly in the insert to Fig. 2b. Along with increasing
n from 20 (Fig. 2a) to 80 (Fig. 2c), the density of Pt nanoparticles
on the substrate significantly increases, and the size of both type
nanoparticles is also enlarged. As indicated by histograms in Fig. 2a,
when n = 20, the average dimension of the congeries of polyhedra
is 277 34 nm, and the average size of the cube-like particles is
75 15 nm; when n = 40, the average dimension of the congeries of
polyhedra increases to 301 23 nm (Fig. 2b), and the average size
A(nm-Pt/S(n))
A(Pt)
QH(nm-Pt/S(n))
QH(Pt)
Rr =
=
(2)
where A(nm-Pt/S(n)) and A(Pt) are electro-active surface area
of nm-Pt/S(n) and polished Pt electrodes, respectively, QH(nm-
Pt/S(n)) and QH(Pt) are the electric charge of hydrogen adsorption
that were measured by integration of corresponding CV curves in
the hydrogen region.

2068
D.-M. Zeng et al. / Electrochimica Acta 55 (2010) 2065–2072
Fig. 3. (a) In situ SNIFTIR spectra of (bi)sulfate adsorbed on a bulk Pt (dash line, 2000 scans, ER = 0.05 V, the y-scale is multiplied by a factor 2 in order to illustrate clearly the
comparison) and nm-Pt/S(n) electrodes (1000 scans, ER = 0.24 V), 0.1 M H2SO4, ES = 0.75 V, 8 cm−1; (b) In situ SNIFTIR spectra of (bi)sulfate adsorbed on nm-Pt/GC(n) electrodes
with n varying from 20 to 80, ER = 0.24 V, ES = 0.75 V. 0.1 M H2SO4, 1000 scans, 8 cm−1
.
of the cube-like particles is augmented to 95 13 nm (Fig. 2b); as
n increases further to 80, the average dimension of the congeries
of polyhedra increases further to 381 28 nm (Fig. 2c), and the
average size of the cube-like particles reaches 118 20 nm (Fig. 2c).
nated twofold adsorbed (bi)sulfate, while Nart et al. [25] attributed
the band at 1052 cm−1 to the coordinated S–O vibration of onefold
species.
In comparison with those IR features appeared in the spectrum
of the bulk Pt electrode, anomalous IR features are observed clearly
in spectra of (bi)sulfate adsorbed on nm-Pt/S(n) electrodes, consist-
ing mainly in four aspects: (1) the IR bands become positive-going,
i.e. the anomalous IR absorption, indicating that these bands are
derived from the adsorbed (bi)sulfate species. This has been con-
firmed also by the linear shift of band center with ES (see details
in Fig. 5). In case of bulk Pt electrode, the IR bands of adsorbed
(bi)sulfate and those of (bi)sulfate in solution are the same direction
of negative-going in spectra, which arose difficulty in discern-
ing whether these IR bands to adsorbed or to solution (bi)sulfate
species. Nevertheless the inversion of the direction of IR bands
in spectra recorded on nanomaterials is very helpful to confirm
the adsorption status of species [15]; (2) the IR band observed at
1257 cm−1 on bulk Pt is red-shifted to 1205 cm−1 on nm-Pt/S(n);
(3) the IR band appears clearly at 1052 cm−1 in spectra of both nm-
Pt/Pt(30) and nm-Pt/Au(30) electrodes, while it becomes a shoulder
peak in the spectrum of nm-Pt/GC(40) electrode; and (4) the inten-
sity of IR bands especially that of the band around 1200 cm−1
has been significantly enhanced. We can see clearly that these
anomalous IR characters are all presented in spectra of the nm-
Pt/S(n) electrodes regardless of the different substrate materials
(GC, Pt and Au) that bear different chemical nature and differ-
ent reflectivity, although slight differences in IR features may be
noticed. This result has demonstrated undoubtedly that the anoma-
lous IR property is a peculiar character of the nanostructured Pt
film.
3.2. In situ FTIR spectroscopic studies
3.2.1. Comparison of adsorption behaviors of (bi)sulfate on a bulk
Pt and nanostructured Pt electrodes
The comparison of IR features of (bi)sulfate adsorbed on nanos-
tructured Pt electrodes and on a bulk Pt electrode in 0.1 M
H2SO4 is presented in Fig. 3. In order to facilitate the com-
parison, electrodes of Pt nanoparticles deposited on different
substrates with close Rr were used, i.e. nm-Pt/GC(40) (Rr = 3.8),
nm-Pt/Pt(30) (Rr = 4.2) and nm-Pt/Au(30) (Rr = 4.4). It is known
that the (bi)sulfate adsorption on electrode surface may occur
in three configurations, i.e. the threefold, twofold and one-
fold configurations, as illustrated by the diagram below [23].
From spectra in Fig. 3, we observe three negative-going bands
around 1257, 1109 and 1052 cm−1 in spectrum recorded of the
bulk Pt electrode, corresponding to IR absorption at ES. Although
extensive studies were presented in the literature, the assignment
of these bands is still not clear. These IR bands have been attributed
to IR absorption of sulfate [24,25,37], bisulfate [26,38,39], or both
species [41]. As these IR bands correspond to IR absorption of
adsorbed species derived from sulfuric acid, they are commonly
assigned to (bi)sulfate species as described in recent literatures
[28,29,33,40]. We adopted this assignment in the present paper.
Hoshi et al. [29] assigned the band around 1257 cm−1 to the coor-
dinated S–O stretching of threefold adsorbed (bi)sulfate species,
and the band near 1109 cm−1 to the S–O stretching of the coordi-
Fig. 3b illustrates the variation of IR features of (bi)sulfate
adsorbed on nm-Pt/GC(n) electrodes in 0.1 M H2SO4 at ES = 0.75 V.
Along with increasing n, the band around 1200 cm−1 is red-
shifted, while the band near 1110 cm−1 remains almost at similar
wavenumbers. At first, the band close to 1054 cm−1 is a shoulder
peak in contrast to the strong bands around 1200 and 1110 cm−1
,
then becomes more pronounced in spectra recorded on both nm-
Pt/GC(60) and nm-Pt/GC(80) electrodes. This band does not shift
with the increase of n.

D.-M. Zeng et al. / Electrochimica Acta 55 (2010) 2065–2072
2069
from the spectrum (the solid line) recorded with ER = 0.24 V and
ES = 0.40 V after stripping completely the pre-adsorbed CO on the
nm-Pt/GC(80) electrode surface, the CO band disappears and the
three bands around 1197, 1118 and 1049 cm−1 all become positive-
going and with enhanced intensity, confirming that the (bi)sulfate
species are readsorbed on the nm-Pt/GC(80) electrode surface. This
result has indicated also that the adsorption of CO is much stronger
than that of (bi)sulfate species on the nm-Pt/GC(80) electrode
surface. Moreover, both adsorbed CO and (bi)sulfate on nanostruc-
tured Pt electrodes give rise to anomalous IR features.
We have compared also the IR features of spectrum recorded at a
potential (ES = 0.75 V) in the double layer region with that acquired
at a potential (ES = 1.00 V) in the oxygen region, the results obtained
on a nm-Pt/GC(80) electrode are displayed in Fig. 4b. It can be
seen that the three IR bands around 1200, 1100 and 1050 cm−1 in
spectrum of ES setting at the oxygen region (1.00 V) are all negative-
going, predicating that the (bi)sulfate species are in solution, while
these bands in the spectrum recorded with ES = 0.75 V are turned
into positive-going with enhanced intensity. This result may be
interpreted as follows: in the oxygen region, oxygen can adsorb
on nm-Pt/GC(80) surface to form platinum oxide species such as
Pt(OH), PtO, Pt(OH)3, PtO2 [51], which can desorb the adsorbed
(bi)sulfate or inhibit the (bi)sulfate anions adsorption on the one
hand, and on the other hand, the anions diffused from bulk solution
into in the thin layer between electrode and IR window increase sig-
nificantly as potential rising [34–36]. The large quantity of anions
gathered in the thin layer has resulted strong IR bands of solu-
tion (bi)sulfate species, which can be also observed in the spectrum
recorded on a bulk Pt electrode at the same ES, for which no anoma-
lous IR property of the bulk Pt electrode for (bi)sulfate adsorption. In
the spectra recorded on a bulk Pt electrode, the IR bands are always
negative-going regardless of ES = 0.75 V or ES = 1.0 V. It is obvious
that, from these results, the adsorption of (bi)sulfate species is
weaker than that of oxygen species on Pt electrode surface.
Fig. 4. (a) In situ SNIFTIR spectra of (bi)sulfate adsorbed on nm-Pt/GC(80) electrode
with CO adsorbed on electrode surface (dash line, 6000 scans, ER = 0.10 V) and after
stripping completely the adsorbed COad (solid line, 1000 scans, ER = 0.24 V), 0.1 M
H2SO4, ES = 0.40 V, 16 cm−1. (b) In situ SNIFTIR spectra recorded on nm-Pt/GC(80)
with ES respectively at 0.75 V (double layer potential region) and at 1.00 V (oxygen
3.2.3. Effects of electrode potential on (bi)sulfate adsorption
In order to study the behaviors of potential dependence of
(bi)sulfate adsorption, an appropriate reference potential ER in the
SNIFTIRS studies is important. In the past study, the potential of
0.05 V was chosen as the reference potential on bulk and single
crystal Pt electrodes [21–44], since radioactive tracer measure-
ments showed negligible adsorption of anions at this potential
[52]. Fig. 5 shown in situ FTIR spectra recorded with ES = 0.75 V
and ER varying from 0.05 to 0.30 V (solid line). It can be seen that
the IR bands of (bi)sulfate adsorbed on nanostructured Pt surface,
i.e. the bands near 1205 and 1108 cm−1, have reached the largest
intensity with ER at 0.24 V. This result indicates that the adsorbed
(bi)sulfate adsorbed on nanostructured Pt surface at 0.24 V presents
a minimum among all other ER studied. As the result in situ FTIR
spectra were reported as the relative change in reflectivity, the
less adsorbed (bi)sulfate at ER will make clearer the IR features
of adsorbed (bi)sulfate at ES. So 0.24 V was set up as ER in the
present in situ FTIR study of (bi)sulfate adsorption on nm-Pt/S(n)
electrodes. In comparison, 0.05 V and 0.24 V were also used as ER
for in situ FTIR study of (bi)sulfate adsorption on a bulk Pt elec-
trode (see the dashed line in Fig. 5). The intensity of IR bands of
adsorbed (bi)sulfate on bulk Pt surface reached a maximum value
with ER = 0.05 V, which is in consistence with the literature [52].
By setting up ER = 0.24 V, a series of in situ FTIR spectra were
recorded on the nm-Pt/GC(40) electrode and displayed in Fig. 6a.
These spectra illustrate the potential dependence of IR features
of adsorbed (bi)sulfate. Two obvious positive-going bands appear
respectively around 1180 and 1110 cm−1 at 0.30 V, and shifted
positively and linearly with increasing ES. The Stark tuning rate
is measured from the slope of the linear variation of band center
(ꢀ˜) with ES, and values of 57.3 and 18.7 cm−1 V−1 were evaluated
potential region), 0.1 M H2SO4, ER = 0.24 V, 1000 scans, 8 cm−1
.
3.2.2. The competitive adsorption of (bi)sulfate with CO and
oxygen species
In order to study further the competition adsorption behaviors
of (bi)sulfate, CO adsorption at low potentials and oxygen adsorp-
tion at high potentials were introduced. The comparison of in situ
FTIR spectra before and after the oxidation of a saturation layer of
CO pre-adsorbed on nm-Pt/GC(80) electrode is illustrated in Fig. 4a.
When a saturation layer CO is pre-adsorbed on nm-Pt/GC(80) elec-
trode, we observe from the spectrum (the dashed line) a strong
IR band around 2060 cm−1 that is assigned to IR absorption of CO
adsorbed on nanostructured Pt electrode surface. The distortion of
this band confirms the anomalous IR property of the nm-Pt/GC(80)
for CO adsorption as reported previously for CO adsorption on
nanostructured Pt electrodes [13]. As the adsorbed CO exists at
both ER (0.10 V) and ES (0.40 V)[13] and its bonding with Pt sur-
face is so strong that it could desorb any other adsorbates, all the
bands corresponding to (bi)sulfate (1197, 1118 and 1049 cm−1) are
negative-going, which evidences that these IR bands arise from
IR absorption of solution (bi)sulfate species that were diffused at
ES from the bulk solution into the thin layer between electrode
and IR window due to electrostatic effect, and evidences that no
adsorbed (bi)sulfate species are presented on the CO covered nm-
Pt/GC(80) electrode surface. However, when the saturation layer of
pre-adsorbed CO is removed completely by stripping it at a high
electrode potential, i.e. 1.00 V in the present experiment, the IR
features of (bi)sulfate are changed dramatically. As can be seen

2070
D.-M. Zeng et al. / Electrochimica Acta 55 (2010) 2065–2072
Table 1
Variation of Stark tuning rates (dꢀ˜/dES) with n measured on nm-Pt/GC(n).
n
dꢀ˜/dES (cm−1 V−1) (1200 cm−1
)
dꢀ˜/dES (cm−1 V−1) (1100 cm−1
)
20
30
40
60
80
152.1
86.4
57.3
37.5
7.6
21.1
19.6
18.7
23.0
16.2
but it is decreased to 7.6 cm−1 V−1 when n is 80. The decrease in
Stark tuning rate has been observed also with the increase of Rr
for CO adsorbed on nanostructured Pt electrodes [14]. As the Rr
increases almost linearly with increasing n (see Fig. 1b), the rapid
decrease in Stark tuning rate of the band around 1200 cm−1 is
certainly an effect of surface structure. In contrast, only a small fluc-
tuation of dꢀ˜/dES between 23.0 and 16.2 cm−1 V−1 was measured
for the band around 1110 cm−1. It is known that the Stark tuning
rate of the IR band of (bi)sulfate is also related to the adsorbed
geometry [23–25]. A (bi)sulfate adsorbed with threefold geome-
try (ca. 1200 cm−1) yields a dꢀ˜/dES about 100 cm−1 V−1 on Pt(1 1 1)
[23,24,26,27], while a twofold geometry (bi)sulfate gives rise to
smaller dꢀ˜/dES of 12.5 cm−1 V−1 for the coordinated S–O band (ca.
1100 cm−1) [25]. The larger dꢀ˜/dES of the band at ca. 1200 cm−1
of adsorbed (bi)sulfate has been interpreted by a back donation of
electron from metal to LUMO orbit of the (bi)sulfate [23], and the
integrated effects of the consequence of the electric field on the
effective dipole moments and the coverage [25].
The difference of the coverage does not cause substantial dis-
crepancy of the dꢀ˜/dES between the threefold and the twofold
(bi)sulfate species, since the estimated dꢀ˜/dES of the singleton
threefold (bi)sulfate is still bigger than that of the twofold one
[24,29]. When n increases, the Rr also increases, while the dꢀ˜/dES
of the band near 1200 cm−1 decreases obviously. As illustrated by
Fig. 2, the density of Pt nanoparticles on GC substrate increases
along with increasing n (particles per ␮m2: nm-Pt/GC(20): 2 (cube-
like), 4 (congeries of polyhedral); nm-Pt/GC(40): 9 (cube-like), 5
Fig. 5. In situ SNIFTIR spectra of (bi)sulfate adsorbed on (a) nm-Pt/GC(40) (solid
line, 1000 scans) and (b) a bulk Pt electrodes (dash line, 2000 scans, the y-scale is
multiplied by a factor 2 in order to illustrate clearly the comparison) with different
ER in hydrogen potential region, 0.1 M H2SO4, ES = 0.75 V, 8 cm−1
.
from Fig. 6b for the two bands, respectively. It is interesting to note
that the Stark tuning rates of these two IR bands vary in differ-
ent ways with n. The increase in n causes a significant decrease in
dꢀ˜/dES of the band around 1200 cm−1. As listed in Table 1, the Stark
tuning rate of the band around 1200 cm−1 decreases quickly with
increasing n. The dꢀ˜/dES of this band is 152.1 cm−1 V−1 for n = 20,
Fig. 6. (a) In situ SNIFTIR spectra of (bi)sulfate adsorbed on nm-Pt/GC(40), the corresponding potential dependences of (b) the center and (c) the intensity of the IR bands
near 1200 cm−1(ꢀ) and 1100 cm−1(ꢀ). 0.1 M H2SO4, 8 cm−1, 1000 scans, ER = 0.24 V, ES varying progressively from 0.30 to 0.90 V with an interval of 0.05 V.

D.-M. Zeng et al. / Electrochimica Acta 55 (2010) 2065–2072
2071
at n = 30. The largest values of ꢃIR (1200 cm−1) are respectively 9.0
and 9.1 on nm-Pt/Pt(30) and nm-Pt/Au(30), while the largest values
of ꢃIR (1100 cm−1) are only 2.0 and 1.7 on nm-Pt/Pt(30) and nm-
Pt/Au(30). It is worthwhile noting that the GC substrate yields the
largest enhancement of IR absorption among the three substrates
used, illustrating the substrate effect on the IR features, which may
be ascribed to the much lower reflectivity of the GC substrate than
that of the Pt and Au metallic substrates. However, for thicker layers
the IR enhancement depends mostly on the size, shape and distri-
bution of the metal particles, which should be largely independent
of the nature of the substrate.
Zheng and Sun [10] have investigated the adsorption of CO on
nanostructured Ru film electrodes (nm-Ru/GC), and they found that
the variation of the enhancement factor ꢃIR versus the film thick-
ness is manifested an asymmetrical volcano curve; the maximum
value appears when the film thickness is 86 nm. Heaps and Griffiths
[55] have also illustrated the IR intensity of SAMs of PNTP formed
on gold films, and they observed that the IR intensity did not always
increase with film thickness, there is also a maximum at a certain
film thickness. In the current study, the thickness of nanoparticles
film definitely increases with n, so the variation of ꢃIR with n is cer-
tainly related to the variation of thickness of film of Pt nanoparticles
on substrates.
(congeries of polyhedral); nm-Pt/GC(80): 18 (cube-like), 6 (con-
geries of polyhedral)). As a result of the mechanical renormalization
[53] and the Pauli repulsion [54], the multicoordination to the sur-
face causes a blue shift of the band center [25,38], and the more
coordination between (bi)sulfate and the electrode surface, the
more sensitive the adsorbed (bi)sulfate to the surface structure.
The trend for the coordinated group frequencies is v˜(S–O,threefold)
>
v˜(S–O,twofold) > v˜(S–O,onefold), i.e. v˜(a) > v˜(b) > v˜(c). Furthermore, the
large variation of the Stark tuning rate is also dependent on the
sensitivity of adsorptive modes to surface structures. The threefold
mode is more sensitive to surface structure than the twofold mode
does. As a consequence, the above results may confirm that the IR
bands near 1200, 1100 and 1050 cm−1 could be assigned respec-
tively to the S–O vibration of the coordinated threefold (bi)sulfate,
twofold and onefold adsorbed (bi)sulfate species.
The variations of intensity (I) of the IR bands around 1200 and
1110 cm−1 are plotted versus potential in Fig. 6c. It can be seen that
I of both bands increases almost linearly with increasing ES and
reaches a maximum at 0.80 V, then decreases with further increase
of ES beyond 0.80 V. We have tested that all nm-Pt/S(n) elec-
trodes have yielded almost the similar variation, i.e. the intensity
of IR bands of (bi)sulfate adsorption on all nm-Pt/S(n) electrodes
achieves the maximal value at 0.80 V. The result implies that the
adsorption of (bi)sulfate occurs mainly in the double layer potential
region. This phenomenon is coincident with the thermodynamic
studies of (bi)sulfate adsorption on Pt electrodes (Pt(h k l), espe-
cially for Pt(1 1 1)) [30,32,33], in which a maximal adsorption has
been determined at 0.80 V (vs. RHE) as well.
4. Conclusions
In the present paper, nanostructured Pt electrodes were
prepared through electrodeposition of Pt nanoparticles on dif-
ferent substrates by applying cyclic voltammetry. Adsorption of
(bi)sulfate on as-prepared nm-Pt/S(n) (S = GC, Pt and Au) and a
bulk Pt electrodes was studied using in situ infrared reflection spec-
troscopy. An IR band around 1257 cm−1 in the spectra recorded on
bulk Pt is assigned to IR absorption of S–O stretching of adsorbed
threefold anions, while it has been shifted negatively to around
1200 cm−1 on nanostructured Pt electrodes. Other two IR bands
around 1110 and 1050 cm−1 on bulk Pt are attributed to IR absorp-
tion of coordinated S–O vibration of the adsorbed twofold anions
and onefold anions, respectively. Similar wavenumbers of these
two bands were measured in spectra recorded on nanostructured Pt
electrodes. Anomalous IR features were observed from in situ FTIR
spectra of (bi)sulfate adsorption on all nanostructured Pt electrodes
prepared in the current study, i.e. the IR absorption of (bi)sulfate
species adsorbed on all the nm-Pt/S(n) electrodes is significantly
enhanced and the IR band direction is completely inverted, in com-
parison with the same species adsorbed on a bulk Pt electrode.
These anomalous IR features were attributed to the anomalous IR
property of Pt nanoparticles film. Although the different substrates,
which present different chemical nature and different reflectiv-
ity, yield different enhancement factors, the inversion of IR band
direction for adsorbed (bi)sulfate is always maintained in spectra
recorded on Pt nanoparticles film deposited on different substrates.
This is the first demonstration that nanostructured Pt electrodes
exhibit anomalous IR property for weakly adsorbed (bi)sulfate
species, and also a first observation of anomalous IR features in the
fingerprint region. It has found that the S-O stretching frequencies
and the band intensities increase linearly with electrode potential
in double layer region and reaches a maximum at 0.80 V, indicating
a potential dependence of (bi)sulfate adsorption on nanostructured
Pt electrodes. It has demonstrated also that the adsorbed (bi)sulfate
can be desorbed by the adsorption of oxygen species with elec-
trode potentials in the oxygen region, which is evidenced by the
change of the direction of (bi)sulfate IR bands form positive-going
(ES = 0.75 V, anomalous IR absorption of adsorbates) to negative-
going (ES = 1.0 V, IR absorption of solution species that were diffused
from bulk solution into the thin layer between electrode and IR win-
3.2.4. The enhancement of IR absorption of (bi)sulfate adsorbed
on nanostructured Pt electrodes
To study quantitatively the enhancement of IR absorption, an
enhancement factor (ꢃIR) is designated to represent the degree of
enhancement of IR absorption by (bi)sulfate adsorbed on nanos-
tructured Pt electrodes referring to that on a bulk Pt electrode,
i.e.
1 IIR(nm-Pt/S(n))
ꢃIR
=
(4)
Rr
IIR(Pt)
where IIR(nm-Pt/S(n)) and IIR(Pt) are the integral IR band intensi-
ties of (bi)sulfate adsorbed on nm-Pt/S(n) and bulk Pt electrodes,
respectively. Rr is the corresponding relative roughness of the
nm-Pt/S(n) electrode evaluated by using Eq. (2), which is used to
normalize the surface adsorption sites of nm-Pt/S(n) referring to a
polished Pt electrode. The IIR(nm-Pt/S(n)) is measured in the spec-
tra collected at 0.75 V, since this potential is close to the potential
at which the maximal intensity of IR bands of (bi)sulfate adsorbed
on nm-Pt/S(n) is measured.
Table 2 lists the ꢃIR measured on nm-Pt/GC(n) electrodes. It
can be seen that the ꢃIR depends on the number of cycles applied
to electrodeposition of Pt nanoparticles on GC substrate. It is evi-
dent that the ꢃIR of the band near 1200 cm−1 (ꢃIR(1200 cm−1))
is much larger than the ꢃIR of the band around 1100 cm−1
(ꢃIR(1100 cm−1)), and the enhancement factor of both IR bands
reaches a maximum when n is 30. The largest ꢃIR (1200 cm−1) is
16.0 and the ꢃIR (1100 cm−1) is only 3.7. On the nm-Pt/Pt(n) and
nm-Pt/Au(n) electrodes, the largest ꢃIR values were also measured
Table 2
Variation of ꢃIR with n measured on nm-Pt/GC(n).
ꢃIR (1200 cm−1
)
ꢃIR (1100 cm−1
)
20
30
40
60
80
15.3
16.0
12.3
8.4
1.4
3.7
1.9
1.1
1.9
9.7

2072
D.-M. Zeng et al. / Electrochimica Acta 55 (2010) 2065–2072
dow). The facts that both CO and oxygen adsorptions can inhibit the
adsorption of (bi)sulfate anions confirmed the weak adsorption of
(bi)sulfate anions on nanostructured Pt electrodes.
[12] H.C. Wang, S.G. Sun, J.W. Yan, H.Z. Yang, Z.Y. Zhou, J. Phys. Chem. B 109 (2005)
4309.
[13] Q.S. Chen, S.G. Sun, J.W. Yan, J.T. Li, Z.Y. Zhou, Langmuir 22 (2006) 10575.
[14] H. Gong, S.G. Sun, Y.J. Chen, S.P. Chen, J. Phys. Chem. B 108 (2004) 11575.
[15] A.E. Bjerke, P.R. Griffiths, Anal. Chem. 71 (1999) 1967.
[16] S. Park, Y.Y. Tong, A. Wieckowski, M.J. Weaver, Electrochem. Commun. 3 (2001)
509.
[17] G. Orozco, C. Gutierrez, J. Electroanal. Chem. 484 (2000) 64.
[18] G. Blyholder, J. Phys. Chem. 68 (1964) 2772.
[19] H. Buijs, in: P.R. Griffiths, J.M. Chalmers (Eds.), Handbook of Vibrational Spec-
troscopy, John Wiley & Sons Ltd, Chichester, 2002.
[20] E. Theocharous, J.R. Birch, in: P.R. Griffiths, J.M. Chalmers (Eds.), Handbook of
Vibrational Spectroscopy, John Wiley & Sons Ltd, Chichester, 2002.
[21] J. Clavilier, J. Electroanal. Chem. 107 (1980) 211.
[22] K. Kunimatsu, M.G. Samant, H. Seki, J. Electroanal. Chem. 258 (1989) 163.
[23] P.W. Faguy, N. Markovicl, R.R. Adzic, C.A. Fierro, E.B. Yeager, J. Electroanal. Chem.
289 (1990) 245.
[24] F.C. Nart, T. Iwasita, M. Weber, Electrochim. Acta 39 (1994) 961.
[25] F.C. Nart, T. Iwasita, M. Weber, Electrochim. Acta 39 (1994) 2093.
[26] P.W. Faguy, N.S. Marinkovic, R.R. Adzic, Langmuir 12 (1996) 243.
[27] Y. Shingaya, M. Ito, Chem. Phys. Lett. 256 (1996) 438.
[28] V. Climen, N. Garcia-Araez, J.M. Feliu, Electrochem. Commun. 8 (2006) 1577.
[29] N. Hoshi, A. Sakurada, S. Nakamura, S. Teruya, O. Koga, Y. Hori, J. Phys. Chem. B
106 (2002) 1985.
[30] J. Mostany, E. Herrero, J.M. Feliu, J. Lipkowski, J. Phys. Chem. B 106 (2002) 12787.
[31] M.C. Santos, D.W. Miwa, S.A.S. Machado, Electrochem. Commun. 2 (2000) 692.
[32] W. Savich, S.G. Sun, J. Lipkowski, A. Wieckowski, J. Electroanal. Chem. 388
(1995) 233.
The current paper has demonstrated clearly the behaviors
of (bi)sulfate adsorption on nanostructured Pt electrodes, and
revealed also the anomalous IR property of nanostructured Pt for
weak adsorbates. The study has extended the study of anomalous
IR property of nanostructured electrodes from strongly adsorbed
species to weakly adsorbed anions, and also from the high-
est energy region (2100–1800 cm−1) provided by a modern FTIR
instrument to fingerprint region, i.e. the relatively lower IR energy
region where most organic fuel molecules give rise to characteris-
tic IR absorption. The results revealed further that the anomalous
IR properties are essential for nanostructured electrodes, and con-
tributed to the understanding of the origin of peculiar property of
nanomaterials as well. The anomalous IR features, i.e. the enhance-
ment of IR absorption and the inversion of IR band direction, are
significant for determining adsorbates including anions at molecule
level, and for investigations of surface processes and reaction mech-
anism involved in electrocatalysis of direct organic molecule fuel
cells that employ mainly Pt nanoparticles deposited on carbon
materials as catalysts.
[33] N. Garcia-Araez, V. Climent, P. Rodriguez, J.M. Feliu, Electrochim. Acta 53 (2008)
6793.
[34] M. Futamata, L.Q. Luo, C. Nishihara, Surf. Sci. 590 (2005) 196.
[35] Y. Shingaya, K. Hirota, H. Ogasawara, M. Ito, J. Electroanal. Chem. 409 (1996)
103.
Acknowledgment
[36] K. Hirota, M.B. Song, M. Ito, Chem. Phys. Lett. 250 (1996) 335.
[37] G.J. Edens, X. Gao, M.J. Weaver, J. Electroanal. Chem. 375 (1994) 357.
[38] Y. Shingaya, M. Ito, J. Electroanal. Chem. 467 (1999) 299.
[39] A. Lachenwitzer, N. Li, J. Lipkowski, J. Electroanal. Chem. 532 (2002) 85.
[40] K. Kunimatsu, T. Senzakib, G. Samjeske, M. Tsushima, M. Osawa, Electrochim.
Acta 52 (2007) 5715.
[41] F.C. Nart, T. Iwasita, J. Electroanal. Chem. 322 (1992) 289.
[42] I. Fromondi, D. Scherson, J. Phys. Chem. C 111 (2007) 10154.
[43] A. Kolics, A. Wieckowski, J. Phys. Chem. B 105 (2001) 2588.
[44] K. Sato, S. Yoshimoto, J. Inukai, K. Itaya, Electrochem. Commun. 8 (2006) 725.
[45] H. Kita, Y. Gao, T. Nakato, H. Hattori, J. Electroanal. Chem. 373 (1994) 177.
[46] H. Kita, Y. Gao, K. Ohnishi, Chem. Lett. 23 (1994) 73.
[47] N. Hoshi, T. Suzuki, Y. Hori, J. Electroanal. Chem. 416 (1996) 61.
[48] M. Nakamura, H. Kato, N. Hoshi, J. Phys. Chem. C 112 (2008) 9458.
[49] T. Kondo, J. Morita, K. Hanaoka, S. Takakusagi, K. Tamura, M. Takahasi, J. Mizuki,
K. Uosaki, J. Phys. Chem. C 111 (2007) 13197.
This study was supported by grants from National Natural Sci-
ence foundation of China (Grant Nos. 20833005, 20873116 and
20828005).
References
[1] V. Palermo, M. Palma, P. Samori, Adv. Mater. 18 (2006) 145.
[2] V.M. Shalaev, A.K. Sarychev, Phys. Rev. B 57 (1998) 13265.
[3] T.P.St Clair, D.W. Goodman, Top. Catal. 13 (2000) 5.
[4] L. Krusin-Elbaum, T. Shibauchi, B. Argyle, L. Gignac, D. Weller, Nature 410 (2001)
444.
[5] R. Emory, W.E. Haskins, S. Nie, J. Am. Chem. Soc. 120 (1998) 8009.
[6] M. Osawa, in: S. Kawata (Ed.), Near-field Optics and Surface Plasmon Polaritons,
vol. 81, Springer-Verlag, Berlin, Germany, 2001, p. 163.
[7] T.A. Leskova, M. Leyva-Lucero, E.R. Mendez, A.A. Maradudin, I.V. Novikov, Opt.
Commun. 183 (2000) 529.
[50] K. Ataka, G. Nishina, W.B. Cai, S.G. Sun, M. Osawa, Electrochem. Commun. 2
(2000) 417.
[8] S. Baldelli, A.S. Epple, E. Anderson, J. Chem. Phys. 113 (2000) 5432.
[9] G.Q. Lu, S.G. Sun, L.R. Cai, S.P. Chen, Z.W. Tian, K.K. Shiu, Langmuir 16 (2000)
778.
[51] N.H. Li, S.G. Sun, S.P. Chen, J. Electroanal. Chem. 430 (1997) 57.
[52] G. Horanyi, J. Solt, F. Nagy, J. Electroanal. Chem. 31 (1971) 95.
[53] T.B. Grimley, Proc. Phys. Soc. 79 (1962) 1203.
[10] M.S. Zheng, S.G. Sun, J. Electroanal. Chem. 500 (2001) 223.
[11] H. Gong, S.G. Sun, J.T. Li, Y.J. Chen, S.P. Chen, Electrochim. Acta 48 (2003) 2933.
[54] P.S. Bagus, W. Muller, Chem. Phys. Lett. 115 (1985) 540.
[55] D.A. Heaps, P.R. Griffiths, Vibrational Spectrosc. 42 (2006) 45.