Pd deposition onto Au(111) electrodes from sulphuric acid solution

Article abstract of DOI:10.1016/j.electacta.2005.04.009

The initial stages of palladium deposition onto Au(111) from 0.1 M H ₂SO₄ + 0.1 mM PdSO₄ have been studied by cyclic voltammetry and in situ scanning tunnelling microscopy. While Pd is commonly deposited from chloride solutions, the effect of sulphate adsorption is considered in this work. A Pd monolayer is formed at underpotentials before a second monolayer grows at overpotentials. There is strong evidence that these two Pd layers are pseudomorphic with the Au(111) substrate. Sulphate is adsorbed on the pseudomorphic Pd layers over a wide potential range in a (3×7)R19.1° structure like in the case of massive Pd(111). The deposition process changes to three-dimensional growth with the third Pd layer, which has already bulk properties. This is indicated by the appearance of a voltammetric peak in the hydrogen adsorption region, which is characteristic for the behaviour of massive Pd(111). Differences to Pd deposition from chloride solutions are discussed.

Full text of DOI:10.1016/j.electacta.2005.04.009

Electrochimica Acta 51 (2005) 125–132
Pd deposition onto Au(111) electrodes from sulphuric acid solution
J. Tang , M. Petri , L.A. Kibler , D.M. Kolb3
1
2
∗,3
Department of Electrochemistry, University of Ulm, 89069 Ulm, Germany
Received 4 February 2005; received in revised form 8 April 2005; accepted 15 April 2005
Available online 11 May 2005
Abstract
The initial stages of palladium deposition onto Au(111) from 0.1 M H
2
4 4
SO + 0.1 mM PdSO have been studied by cyclic voltammetry and in
situ scanning tunnelling microscopy. While Pd is commonly deposited from chloride solutions, the effect of sulphate adsorption is considered
in this work. A Pd monolayer is formed at underpotentials before a second monolayer grows at overpotentials. There is strong evidence that
these two Pd layers are pseudomorphic with the Au(111) substrate. Sulphate is adsorbed on the pseudomorphic Pd layers over a wide potential
√
√
◦
range in a ( 3 × 7)R19.1 structure like in the case of massive Pd(111). The deposition process changes to three-dimensional growth with
the third Pd layer, which has already bulk properties. This is indicated by the appearance of a voltammetric peak in the hydrogen adsorption
region, which is characteristic for the behaviour of massive Pd(111). Differences to Pd deposition from chloride solutions are discussed.
©
2005 Elsevier Ltd. All rights reserved.
Keywords: Cyclic voltammetry; Scanning tunnelling microscopy; Surface structure; Pseudomorphism; Gold; Palladium
1
. Introduction
from surface X-ray diffraction [8] and scanning tunnelling
microscopy(STM) [1,9–11] withtheoretical studieslike DFT
calculations [12–14] made a substantial contribution to the
interpretation of electrochemical data. Especially the first
and the second Pd monolayer on Au(111), which have been
shown to be pseudomorphic with the substrate [8], reveal an
altered electrochemical behaviour as compared to massive
Pd(111) [3].
Thin metal overlayers have gained considerable interest in
various fields of physical chemistry, including electrocatal-
ysis. Epitaxially grown thin Pd films on Au(111) have been
used as model systems to study various electrochemical reac-
tions, like adsorption and absorption of hydrogen [1], formic
acid oxidation [2,3], oxygen reduction [4], formaldehyde oxi-
dation [5] and CO adlayer oxidation [6]. Systematic changes
in the activity of Pd on Au(111) with coverage have been
observed for all of these reactions.
Most electrochemical studies of the initial stages of Pd de-
position onto Au(hkl), including those mentioned above, have
been performed with chloride solutions [1,9–11,15,16], since
noble metals are commonly deposited from their stable chlo-
rine compounds. However, anion effects play an important
role for Pd deposition and for the electrochemical behaviour
of the Pd overlayers [3]. While Pd is alloying with Au(111)
under ultrahigh vacuum conditions at room temperature [17],
there has been no indication of alloy formation upon electro-
chemical Pd deposition in the presence of chloride [10]. It
In order to understand such effects, geometric and elec-
tronic modifications of the thin Pd overlayers have been con-
sidered [2–7]. The combination of detailed structure data
∗
Corresponding author at: Universitaet Ulm, Abteilung Elektrochemie,
Albert-Einstein-Allee 47, 89081 Ulm, Germany. Tel.: +49 731 50 25419;
fax: +49 731 50 25409.
has been assumed that the adsorption of [PdCl4]2 plays an
−
E-mail address: ludwig.kibler@chemie.uni-ulm.de (L.A. Kibler).
Present address: Department of Chemistry, Xiamen University, Xiamen
1
important role for the monolayer growth of Pd [9].
2
−
3
61005, China.
Inthispublication, weshowthatthepresenceof[PdCl4]
2
Present address: Infineon Technologies Dresden, 01099 Dresden,
is not mandatory for the two-dimensional growth of Pd on
Au(111), i.e., Pd monolayers can also be obtained from
Germany.
3
ISE member.
0
013-4686/$ – see front matter © 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2005.04.009

1
26
J. Tang et al. / Electrochimica Acta 51 (2005) 125–132
halide-free sulphate solutions. The initial stages of Pd depo-
sition onto Au(111) have been studied by cyclic voltammetry
(CV) and in situ STM. Ordered anion structures, nucleation
and growth of Pd and the morphology of the resulting deposit
are discussed. The findings are compared with Pd deposition
from chloride solution and the effect of anion adsorption is
reassessed.
2
. Experimental
The STM measurements were performed with a PicoSPM
(
Molecular Imaging Corporation), a Nanoscope E (Digital
Fig. 1. Current–potential curves for Au(111) in 0.1 M H2SO4 with (solid
line) and without (dashed line) 0.1 mM PdSO4. Scan rate: 10 mV/s. The
measured equilibrium potential for Pd/Pd in that solution is indicated as
Instruments) and a Topometrix Discoverer 2010. All images
were recorded in the feedback mode at constant currents
of usually IT = 2 nA. Pt/Ir and W tips were prepared by
electrochemical etching of 0.25 mm wires in a lamella of
2+
vertical line at 0.53 V.
3
.4 M NaCN and 2 M NaOH, respectively. The tips were
presence of Pd2+ ions. This is indicated by the characteristic
current spike, which originates from a phase transition within
the sulphate adlayer [19]. At potentials positive of the spike,
coated with electrophoretic paint in order to minimize the
Faraday current at the tip electrode. Pt wires were used as
counter and quasi-reference electrodes (EPt = +0.55 ± 0.05 V
versus SCE) for the STM. A saturated calomel electrode
√
√
◦
the well-known ( 3 × 7)R19.1 structure of adsorbed
sulphate can be clearly seen by STM (see below, Fig. 5a) like
2+
(
SCE) served as reference for CV experiments in a standard
in Pd -free solution [20,21]. It should be noted that the ab-
sence of chloride in solution is verified by the stability of this
ordered sulphate adlayer. Otherwise, the current spike in the
voltammogram would be decreased, because sulphate would
be replaced by the more strongly adsorbing chloride [22].
In the negative-going sweep, Pd deposition commences
below 0.7 V, where the ordered sulphate structure on Au(111)
is no longer stable (Fig. 1). The deposition current decreases
after having reached a maximum at 0.54 V. In order to obtain
certain Pd coverage, the potential scan was stopped at 0.31 V,
until the charge hadreachedthe desired value. Other reactions
like hydrogen adsorption and absorption are avoided this way.
The equilibrium potential for Pd/Pd2+ in 0.1 M H2SO4
+ 0.1 mM PdSO4 was measured with a flame-annealed Pd
wire to be 0.53 V (indicated by the vertical line in Fig. 1).
Since Pd deposition on Au(111) starts at potentials more pos-
itive than 0.53 V, lowering the scan rate should yield separate
peaks for underpotential deposition (upd) and bulk deposi-
tion. It is known already that Pd deposition on Au(111) from
chloride solutions starts in the underpotential region [10].
Fig. 2 shows a current–potential curve (dotted line) for
Au(111) in 0.1 M H2SO4 + 0.1 mM PdSO4 at 0.1 mV/s. At
this very low scan rate, two well-defined deposition peaks
are observed, one being in the upd regime. The charge den-
electrochemical cell, and in the following, all potentials are
quoted with respect to SCE.
The electrolyte was prepared from H2SO4 (Merck supra-
pur), PdSO4·2H2O (99.95% Alfa Aesar), and high-quality
Milli-Q water (18.2 Mꢀ cm and <2 ppb total organic carbon).
The Au single crystal electrodes (MaTecK J u¨ lich, Ger-
◦
many) had surfaces oriented to better than 1 and were pol-
ished down to 0.03 ␮m. Disks with a diameter of 12 mm were
used for STM, while the crystals used for CV were 4 mm in
diameter. However, for some experiments, the 4 mm crystals
were transferred to the STM cell after Pd deposition and CV
characterisation in a conventional electrochemical glass cell.
The electrodes were flame-annealed prior to each measure-
ment either in a hydrogen flame (STM) or with a Bunsen
burner (CV), in order to obtain clean and structurally well-
defined surfaces. After each measurement, the electrodes
were anodically polarised in 0.1 M H2SO4 for 20 s and then
immersed in 1 M HCl to remove residues of palladium.
3
. Results and discussion
3
.1. Electrochemical measurements
−2
sity for the latter amounts to 410 ␮C cm , corresponding
Fig. 1 shows a current–potential curve (solid line)
almost perfectly to the deposition charge of a full pseudo-
−2
for Au(111) in 0.1 M H2SO4 + 0.1 mM PdSO4 and the
well-known cyclic voltammogram (dashed line) for the same
morphic Pd monolayer on Au(111) (1 ML = 445 ␮C cm ).
Thechangeofthepzcinthecourseofdeposition(from0.24 V
for bare Au(111) [23] to −0.09 V for the complete Pd mono-
layer [7]) and further adsorption of sulphate may contribute
2+
electrode in Pd -free 0.1 M H2SO4 for comparison, both at
0 mV/s. After immersing the electrode into the deposition
1
−
2
solution at 0.7 V, where the surface reconstruction is known to
be lifted [18], the potential was scanned in positive direction
until 0.8 V. Obviously, in this potential region, the electro-
chemical behaviour of Au(111) is not influenced by the
about 10 ␮C cm to the small deviation from the theoret-
ical charge density for net deposition. In addition, the Pd
monolayer may not be complete at 0.53 V due to the very
slow deposition kinetics. As will be shown below with in situ

J. Tang et al. / Electrochimica Acta 51 (2005) 125–132
127
lateral triangular islands reflecting the symmetry of the sub-
strate, and gradually covers the whole Au surface with a uni-
form monolayer, except for the Au islands (Fig. 3c–e).
The Pd monolayer appears slightly higher than the
monoatomic high Au islands, which was also observed in
STM experiments for Pd deposition from chloride solution,
where [PdCl4]2 is adsorbed on Au(111) [10]. As will be
shown below, Pd is covered by an ordered sulphate adlayer at
this potential, whereas the Pd-free Au(111) is not. This may
contribute to the height variation seen in the STM images.
The small Au islands in Fig. 3a are not covered with Pd at
−
0.52 V (Fig. 3e). Only after stepping the potential to 0.45 V,
Pd is deposited onto these islands and the first monolayer
is going to be complete (Fig. 3f). The resulting surface
reproduces nicely the topography of the Au(111) substrate,
including the small Au islands. This behaviour is similar to
Pd deposition in the presence of chloride [10] and it clearly
demonstrates that the overpotential for Pd nucleation is
markedly higher on terraces than at steps.
After waiting some minutes at 0.45 V, a second Pd mono-
layer starts to grow, but in a different way (Fig. 3g). Small
Pd islands are formed, the shape of which is relatively
round. These islands gradually merge and form an almost
complete second Pd monolayer (Fig. 3h), which still repro-
duces the substrate morphology like the first layer. A three-
dimensional growth is observed after a potential step to 0.4 V
Fig. 2. Current–potential curves, starting at +0.7 V, for a high-quality
Au(111) electrode (dotted line) and a Au(111) with a 3 miscut (solid line)
in 0.1 M H2SO4 + 0.1 mM PdSO4. Scan rate: 0.1 mV/s.
◦
STM, Pd deposition on top of the Au islands covering about
5
% of the surface, only occurs in the overpotential region.
The formation of a second Pd monolayer equivalent takes
place in the overpotential deposition region at 0.45 V, over-
lapping with further bulk deposition (Fig. 2). This peak has
only been observed for Au(111) crystals of high-quality, i.e.,
it could not be resolved for surfaces with high defect density
◦
(
Fig. 2, solid line, for Au(111) with 3 miscut). However, the
deposition of Pd on stepped Au(111) surfaces yielded an ad-
ditional upd peak located at 0.58 V (Fig. 2, solid line), which
for the high-quality electrode surface is already discernable
as small shoulder. This peak can be attributed to Pd deposi-
tion at surface defects like monoatomic high steps; whereas
the upd peak at 0.56 V corresponds to Pd deposition onto flat
terraces. Such a behaviour is often found for upd of metals,
e.g., for Pd upd on Au(111) from chloride solutions [10] or
Cu upd on Au(111) [10,24]. Thus it is concluded that Pd pref-
erentially nucleates at step sites and that two Pd monolayers
on Au(111) have a remarkable stability, resulting in two pro-
nounced deposition peaks as seen in Fig. 2. These findings
are confirmed by the STM measurements described in the
following.
(Fig. 3i and j).
Compared to deposition from chloride solution [10], the
shape of the Pd islands is different during growth of the
first two monolayers from sulphate solution. While in the
former case a round shape is observed for submonolayers
of Pd [10], a triangular shape is seen here (Fig. 3c and d).
The situation is opposite for the formation of the second Pd
monolayer. Hence, it is concluded that the adsorbed anions,
either [PdCl4]2 or sulphate, have an opposite influence on
the diffusion of Pd atoms on a Au and on a Pd surface. For
example, the mobility of Pd atoms seems higher for Pd on
−
Au with adsorbed sulphate, but higher for Pd on Pd with
adsorbed [PdCl4]2
−
.
Experiments have been performed, where Pd was de-
posited in a conventional electrochemical cell from 0.1 M
H2SO4 + 0.1 mM PdSO4, and the electrode was subsequently
3
3
.2. In situ STM measurements
2+
.2.1. Pd deposition
A series of 10 STM images for Pd deposition onto Au(111)
transferred to the STM cell, filled with Pd -free sulphuric
acid solution. The observed morphology of the first and sec-
ond monolayer is very much the same as that in Fig. 3.
However, there is evidence that Pd forms tremendously
huge deposits at stepped sites for higher coverage. This is
seen in Fig. 4, where an equivalent of four Pd monolayers
had been deposited at 0.4 V from 0.1 M H2SO4 + 0.1 mM
PdSO4. While the second monolayer is widely uncovered,
large Pd particles are formed at step bunching sites with
a maximum height of 4 nm corresponding to almost 20 Pd
layers. This means that the electrochemical behaviour of
the second (pseudomorphic) monolayer might still be pre-
dominant for high Pd coverages, where bulk properties are
expected.
from 0.1 M H2SO4 + 0.1 mM PdSO4 is shown in Fig. 3. The
freshly prepared, reconstructed Au(111) surface was brought
in contact with the solution at 0.6 V. The spontaneous lifting
of the thermally induced reconstruction leaves a surface with
monatomic high gold islands [18,25]. The small holes seen in
Fig. 3a, probably originating from partial surface oxidation
during cell assembly, do not significantly influence the Pd
deposition process.
After lowering the potential, electrodeposition of Pd is
seen to start at the monoatomic high steps including the rims
of the relatively round Au islands (Fig. 3b). Starting from
these surface imperfections, Pd progressively grows in equi-

1
28
J. Tang et al. / Electrochimica Acta 51 (2005) 125–132
Fig. 3. Sequence of STM images, showing the initial stages of Pd deposition onto Au(111) in 0.1 M H2SO4 + 0.1 mM PdSO4, 250 nm × 250 nm. Deposition
potential and deposition time are given in the images. Tip potential Etip = 0.35 V.
3
.2.2. Sulphate adsorption on Pd overlayers on Au(111)
wide potential range, both on the Au(111) surface and on the
Pd layers [9,10]. It is known that sulphate forms an ordered
structure on Au(111) only at potentials positive of the current
spike (Fig. 1) at 0.8 V. Since this ( 3 × 7)R19.1 structure
is also formed on a Pd(111) surface [26], we searched for this
structure on the pseudomorphic Pd monolayers on Au(111)
as well. Fig. 5 shows the STM images of sulphate on the bare
Au(111) surface, the first (b and c) and the second (d) Pd
layer at 0.85, 0.45 and 0.38 V, respectively. The noticeable
different appearance of the structure in the images (a) and
For Pd deposition from chloride-containing electrolytes,
the [PdCl4]2 anion is adsorbed in ordered structures over a
−
√
√
◦
(b), for example, arises from the fact that the bias voltage was
changed because the tip potential was kept constant at 0.35 V.
The unit cells for the structures shown in Fig. 5 are rotated
◦
by 60 , since the position of the substrate was not changed.
√
√
◦
Only rather small domains of the ( 3 × 7)R19.1 structure
could be found on the Pd layers by STM.
Fig. 4. STM image (570 nm × 570 nm) of Au(111), obtained at 0.25 V in
High-resolution STM images of the Pd overlayers, which
would allow a direct determination of the Pd–Pd distances
0
.1 M H2SO4, after deposition of four monolayer equivalents of Pd at 0.4 V
from 0.1 M H2SO4 + 0.1 mM PdSO4.

J. Tang et al. / Electrochimica Acta 51 (2005) 125–132
129
√
√
◦
Fig. 5. High-resolution STM images of the ( 3 × 7)R19.1 structure of adsorbed sulphate on Au(111) (a), and on the first (b and c) and the second (d) Pd
monolayer on Au(111) in 0.1 M H2SO4 + 0.1 mM PdSO4. Tip potential Etip = 0.35 V.
in the monolayers, and hence, provide direct support for
the pseudomorphism with the Au(111) substrate, were pos-
sible only at negative potentials – where sulphate is des-
orbed–formeasurementswithPd -freeelectrolytes(Fig. 6).
In this case, atomic resolution of the second monolayer
was obtained after Pd deposition in a separate electrochem-
ical cell and transfer to the STM cell filled with 0.1 M
H2SO4. Albeit some drift, the atomic distances in the STM
image of Fig. 6 are in agreement with a pseudomorphic
overlayer.
Unfortunately, it was not possible to obtain atomic res-
olution of the Pd surface at more positive potentials due to
the high stability of the ordered sulphate structure, which in-
stead was easily imaged. For a safe assignment of any adlayer
structure by STM, an internal standard for distance and angle
calibration is mandatory. Such standards can be provided by
the substrate or any known commensurate adlayer structure.
Beingawarethatdiffractiontechniquesaremuchbettersuited
than STM to determine exact distances, we did evaluate the
high-resolution images of the sulphate structures on Au(111)
(like the one in Fig. 5a) and on the Pd monolayers for several
experiments. Since the structures of sulphate on Au(111) and
on the Pd monolayers were measured to be isomorphic within
a precision of 3%, we have strong support that the second Pd
monolayer on Au(111) is also pseudomorphic.
For a chloride-containing solution, it had been inferred
from a Moir e´ pattern on the fifth Pd layer on Au(111) that
up to four Pd layers can grow pseudomorphically [10]. Con-
trarily, the characteristic voltammetric peak for hydrogen ad-
sorption on bulk Pd, which emerges after deposition of two or
more Pd layers from sulphate solution, suggests that only two
monolayers grow pseudomorphically on Au(111) [3]. How-
ever, the absence of a strict layer-by-layer growth for large Pd
areas, hampers the electrochemical characterization of mul-
tilayers. Therefore, only the first and second Pd monolayer
can serve as ideal model systems. It remains a challenge to
2+

1
30
J. Tang et al. / Electrochimica Acta 51 (2005) 125–132
tion onto Au(111) after transfer to 0.1 M H2SO4. The elec-
trochemical behaviour of 1 ML Pd on Au(111) is essentially
identical to that of a Pd monolayer deposited from chloride-
containing solution [3]. This supports the findings that the
Pd monolayers formed in both cases have the same structure,
i.e., Pd forms a pseudomorphic monolayer on Au(111). We
would like to mention, however, that [PdCl4]2 may still be
adsorbed on the surface after deposition from chloride solu-
tion, even after thorough rinsing with water. When the elec-
trode is subsequently immersed into 0.1 M H2SO4, a very
small amount of Pd (ca. 0.1 ML) can be deposited and traces
of chloride will be adsorbed on the surface. Fortunately, chlo-
ride can be desorbed completely at potentials negative of
−
−
0.2 V.
The various voltammetric peaks in Fig. 7 between +0.1 V
and the onset of hydrogen evolution around −0.3 V can be at-
tributed to sulphate adsorption, hydrogen adsorption and hy-
drogen absorption. The latter reaction starting around −0.2 V
on the negative scan is only present for Pd overlayers with
a thickness larger than two monolayers, as in the case of Pd
deposition from chloride solution, and can be taken as mea-
sure of bulk Pd [1]. The peaks at more positive potentials are
related to combined hydrogen adsorption/desorption and sul-
phate desorption/adsorption, as was demonstrated by the so-
called CO displacement technique [6,27]. These adsorption
peaks are less sharp for Pd on stepped Au(111) surfaces and
muchbroaderin perchloric acidsolutions [28] andthus are re-
lated to sulphate adsorption on (111)-terraces with long range
order. It is interesting to note that these adsorption peaks are
very similar for two and three monolayer equivalents (ML) of
Pd (Fig. 7). While the surface fraction of pseudomorphic Pd
(second monolayer) is almost unchanged, there is a signifi-
cant contribution of hydrogen absorption into bulk Pd for the
3 ML case. This is in good agreement with the STM obser-
vations, where a pronounced three-dimensional growth of Pd
Fig. 6. High-resolution STM image (8.5 nm × 8.5 nm) of the second Pd
monolayer on Au(111) in 0.1 M H2SO4 at −0.09 V after deposition of an
equivalent of about three Pd monolayers.
prepare a well-ordered Pd bulk phase on Au(111) with large
terraces.
3
.3. Electrochemical behaviour and stability of Pd
overlayers on Au(111) in H2SO4
Fig. 7 shows typical voltammograms for various Pd mono-
layer equivalents (ML) deposited from a chloride-free solu-
Fig. 7. Cyclic voltammograms in the hydrogen adsorption/absorption region for various Pd monolayer equivalents (1–5 ML) on Au(111) in 0.1 M H2SO4.
Scan rate: 5 mV/s.

J. Tang et al. / Electrochimica Acta 51 (2005) 125–132
131
for Au(111) emerge in the 4th cycle, such as the sulphate
adsorption peak and the current spike around 0.2 and 0.8 V,
respectively. There was no indication of alloying under the
present experimental conditions.
4. Summary
It has been shown that Pd forms two pseudomorphic
monolayers on Au(111) upon electrochemical deposition
from 0.1 M H2SO4 + 0.1 mM PdSO4. The first monolayer is
formed in the underpotential region, whereby Pd nucleates
preferentially at step sites. Dissolution of Pd is kinetically
hindered in sulphuric acid solutions. A distinctly different
electrochemical behaviour is observed for the first and second
pseudomorphic monolayer on the one hand, and bulk deposit
formed at higher coverage on the other hand. A pronounced
three-dimensional growth of Pd is observed on top of the
second monolayer. Therefore, the electrochemical behaviour
of the second, pseudomorphic Pd monolayer is still visible
up to relatively large bulk deposits. The deposition process
is accompanied by the adsorption of sulphate which forms a
Fig. 8. Cyclic voltammograms for Au(111) with one monolayer Pd in 0.1 M
H2SO4. Scan rate: 5 mV/s. The changes from first (thick solid line) to second
(dotted line), third (dashed line) and fourth oxidation–reduction cycle (thin
solid line) reflect the dissolution of Pd.
at stepped sites is observed with deposition of more than two
layers. It seems that a huge amount of Pd has to be deposited
in order to cover the second monolayer completely.
Compared with deposition from chloride solution, signif-
icant differences in the electrochemical behaviour of Pd on
Au(111) occur for coverages higher than one monolayer [3].
Fromtheadsorptionpeaksinthehydrogenregioninsulphuric
acid solution, the surface fraction of pseudomorphic (anodic
peakat−0.005 V)andofbulkPd(anodicpeakat−0.035 mV)
can be estimated [3]. Pd deposited from chloride-free solu-
tion shows a higher tendency for three-dimensional growth
than deposited from chloride-containing solution. Therefore,
at comparable coverages the properties of the pseudomorphic
Pd layer are more pronounced for deposition from sulphate
solutions.
√
√
◦
stable ( 3 × 7)R19.1 structure on the pseudomorphic Pd
terraces in the potential region between hydrogen adsorption
and surface oxidation. Anion adsorption has a strong impact
on Pd deposition kinetics as well as on exact island shape.
Acknowledgements
This work was supported by the Fonds der Chemischen
Industrie. The help of Dr. R. Hoyer is greatly acknowledged.
The dissolution kinetics of Pd in chloride-free electrolytes
is extremely slow, since Pd forms a rather stable oxide already
around 0.7 V. However, repeated oxidation–reduction cycles
in 0.1 M H2SO4 lead to a gradual dissolution of palladium.
This behaviour is illustrated in Fig. 8. After deposition of
a Pd monolayer at 0.53 V, the electrode was rinsed with ul-
trapure water and transferred to another electrochemical cell
filled with 0.1 M H2SO4 only, in order to avoid further Pd
deposition. At potentials negative of 0.5 V, where no surface
oxidation takes place, the cyclic voltammograms are rather
stable. A potential excursion up to 0.9 V and subsequent ox-
ide reduction leads to a significant decrease of the voltam-
metric peaks, both in the hydrogen adsorption region and for
the oxide formation peak at 0.73 V in the following cycle.
The peaks in the hydrogen adsorption region can be taken
as a measure for the Pd coverage. Accordingly, Pd is slowly
dissolving in the course of oxidation–reduction cycles. An
additional peak is emerging at 0.6 V, which is related to ox-
ide formation at defects created by the cycling routine. Note
that surface oxidation on massive Pd(111) is very sensitive
to defects, the latter being preferentially oxidized [6]. After
four oxidation–reduction cycles, the Pd coverage has dropped
from 1 ML to about 0.2 ML. On the average about a third of
the Pd deposit is dissolved per cycle. Characteristic peaks
References
[1] M. Baldauf, D.M. Kolb, Electrochim. Acta 38 (1993) 2145.
[2] M. Baldauf, D.M. Kolb, J. Phys. Chem. 100 (1996) 11375.
[3] L.A. Kibler, A.M. El-Aziz, D.M. Kolb, J. Mol. Catal. A 199 (2003)
57.
[4] H. Naohara, S. Ye, K. Uosaki, Electrochim. Acta 45 (2000) 3305.
[5] H. Naohara, S. Ye, K. Uosaki, J. Electroanal. Chem. 500 (2001) 435.
[
[
6] A.M. El-Aziz, L.A. Kibler, J. Electroanal. Chem. 534 (2002) 107.
7] A.M. El-Aziz, L.A. Kibler, D.M. Kolb, Electrochem. Commun. 4
(2002) 535.
[8] M. Takahasi, Y. Hayashi, J. Mizuki, K. Tamura, T. Kondo, H. Nao-
hara, K. Uosaki, Surf. Sci. 461 (2000) 213.
[9] H. Naohara, S. Ye, K. Uosaki, J. Phys. Chem. B 102 (1998) 4366.
10] L.A. Kibler, M. Kleinert, R. Randler, D.M. Kolb, Surf. Sci. 443
[
(1999) 19.
[
[
11] H. Naohara, S. Ye, K. Uosaki, Coll. Surf. A 154 (1999) 201.
12] A. Ruban, B. Hammer, P. Stoltze, H.L. Skriver, J.K. Nørskov, J.
Mol. Catal. A 115 (1997) 421.
[13] E. Christoffersen, P. Liu, A. Ruban, H.L. Skriver, J.K. Nørskov, J.
Catal. 199 (2001) 123.
[14] A. Roudgar, A. Groß, J. Electroanal. Chem. 548 (2003) 121.
[15] L.A. Kibler, M. Kleinert, D.M. Kolb, Surf. Sci. 461 (2000) 155.
[16] L.A. Kibler, M. Kleinert, V. Lazarescu, D.M. Kolb, Surf. Sci. 498
(2002) 175.
[17] B.E. Koel, A. Sellidj, M.T. Paffett, Phys. Rev. B 46 (1992) 7846.

1
32
J. Tang et al. / Electrochimica Acta 51 (2005) 125–132
[24] M.H. H o¨ lzle, V. Zwing, D.M. Kolb, Electrochim. Acta 40 (1995)
[18] D.M. Kolb, Prog. Surf. Sci. 51 (1996) 109.
[19] D.A. Scherson, D.M. Kolb, J. Electroanal. Chem. 176 (1984)
1237.
3
53.
[25] A.S. Dakkouri, D.M. Kolb, in: A. Wieckowski (Ed.), Interfacial Elec-
trochemistry: Theory, Experiment and Applications, Marcel Dekker,
New York, 1999, p. 151.
[
20] O.M. Magnussen, J. Hageb o¨ ck, J. Hotlos, R.J. Behm, Faraday Dis-
cuss. 94 (1992) 329.
[
21] A. Cuesta, M. Kleinert, D.M. Kolb, Phys. Chem. Chem. Phys. 2
[26] L.-J. Wan, T. Suzuki, K. Sashikata, J. Okada, J. Inukai, K. Itaya, J.
Electroanal. Chem. 484 (2000) 189.
(2000) 5684.
[
[
22] M. Giesen, D.M. Kolb, Surf. Sci. 468 (2000) 149.
23] D.M. Kolb, J. Schneider, Surf. Sci. 162 (1985) 764.
[27] L.A. Kibler, Ph.D. Thesis, University of Ulm, 2000.
[28] H. Naohara, S. Ye, K. Uosaki, J. Electroanal. Chem. 500 (2001) 435.