Use of underpotential deposition for evaluation of overpotential deposition kinetics of reactive metals

Article abstract of DOI:10.1149/1.1631282

The underpotential deposition (UPD) of Zn onto Pt is shown to produce an electrode surface which is suitable for the evaluation of Zn overpotential kinetics. The UPD of Zn onto polycrystalline Pt was characterized in acidic and neutral sulfate solutions using voltammetric charge measurements and X-ray photoelectron spectroscopy. The charge density for UPD Zn deposition on platinum from neutral sulfate solution was estimated to be 260 ± 30 μC cm^-2, consistent with a monolayer. The decrease in H₂ evolution rates provided further evidence of the formation of Zn UPD layer. Evidence for the formation of a Zn-Pt alloy, resulting from polarization of the Pt electrode at a potential between the Zn UPD and overpotential deposition potentials is also presented. The alloy was oxidized at the same potential as the Pt substrate, illustrating the large interaction between Zn and Pt due to the small crystallographic misfit. The alloy also decreased H₂ evolution rates, relative to the Pt substrate. The use of such a modified substrate for measurement of kinetic parameters for metal deposition is described and the results obtained are shown to compare favorably with the literature values.

Full text of DOI:10.1149/1.1631282

Journal of The Electrochemical Society, 151 ͑1͒ E1-E6 ͑2004͒
E1
0013-4651/2003/151͑1͒/E1/6/$7.00 © The Electrochemical Society, Inc.
Use of Underpotential Deposition for Evaluation of
Overpotential Deposition Kinetics of Reactive Metals
E. Guerra,a, G. H. Kelsall,
M. Bestetti,a,d D. Dreisinger,a K. Wong,b
*
a,**,c
b,**,z
K. A. R. Mitchell,b and D. Bizzotto
b
aDepartment of Metals and Materials Engineering, and Department of Chemistry, Advanced Materials
Process Engineering Laboratory,
University of British Columbia, Vancouver, British Columbia V6T 1Z4, Canada
The underpotential deposition ͑UPD͒ of Zn onto Pt is shown to produce an electrode surface which is suitable for the evaluation
of Zn overpotential kinetics. The UPD of Zn onto polycrystalline Pt was characterized in acidic and neutral sulfate solutions using
voltammetric charge measurements and X-ray photoelectron spectroscopy. The charge density for UPD Zn deposition on platinum
from neutral sulfate solution was estimated to be 260 Ϯ 30 ␮C cmϪ2, consistent with a monolayer. The decrease in H2 evolution
rates provided further evidence of the formation of Zn UPD layer. Evidence for the formation of a Zn-Pt alloy, resulting from
polarization of the Pt electrode at a potential between the Zn UPD and overpotential deposition potentials is also presented. The
alloy was oxidized at the same potential as the Pt substrate, illustrating the large interaction between Zn and Pt due to the small
crystallographic misfit. The alloy also decreased H2 evolution rates, relative to the Pt substrate. The use of such a modified
substrate for measurement of kinetic parameters for metal deposition is described and the results obtained are shown to compare
favorably with the literature values.
© 2003 The Electrochemical Society. ͓DOI: 10.1149/1.1631282͔ All rights reserved.
Manuscript received April 26, 2003. Available electronically November 21, 2003.
The extended Butler-Volmer equation is often invoked to model
the relationship between current density and overpotential for elec-
trochemical reactions. However, the equation is based on the as-
sumption that the electrode is a smooth planar surface with a corre-
sponding uniform potential distribution. The extended Butler-
Volmer equation for electrodeposition of a metal, Me, including a
provision for electrode roughness is
foreign substrate, S, is proposed as a general technique for generat-
ing suitable electrode surfaces for the study of OPD kinetics, using
the Zn-Pt system as a model case.
Underpotential deposition.—In general, underpotential deposi-
tion describes the formation of a two-dimensional layer of metal,
Me, onto a foreign substrate, S, at a potential more positive than that
for bulk deposition of the metal. In the absence of UPD layer for-
mation, or in cases where UPD results in a significant crystallo-
graphic misfit between the depositing metal and the substrate, over-
potential deposition of the metal over the substrate is heterogeneous.
However, if a Me-S UPD system can be selected such that the de-
gree of crystallographic misfit is small, then OPD of the metal over
the UPD covered substrate should be homogeneous and progress
epitaxially at low current densities. Thus, the crystallographic orien-
tation and surface area of the OPD Me can be taken to be the same
as the original surface of the substrate. Therefore, the current
density-overpotential relationship for bulk deposition of the metal
can be mapped by first coating the substrate at low overpotential and
then pulsing to a potential in the region of bulk deposition, until a
steady-state current response results. In between pulses, the potential
can be returned to a value between the UPD Me and OPD Me
potentials, thereby stripping away the bulk metal from the electrode
surface and recovering the original geometry and crystallographic
orientation of the UPD-coated electrode surface. OPD metal depo-
sition on UPD modified substrates has been widely reported and the
more general literature on electrochemical phase formation and
growth reviewed, including growth on UPD modified substrates.2,3
The identification of a particular Me-S UPD system can be accom-
plished experimentally, using standard electrochemical techniques,
or predicted from theoretical calculations.4,5
Ϫ␰␣nF␩
␰͑1Ϫ␣͒nF␩
Me
Me
ͫ
ͬ
Ϫ j0,Me
e
Ϫ e
RT
RT
jMe
ϭ
͓1͔
Ϫ␰␣nF␩
Me
j0,Me
1 Ϫ
e
RT
jlim,Me
where jMe ͑A mϪ2͒ is the current density of the metal deposition
reaction, j0,Me ͑A mϪ2͒ is the exchange current density, ␩ is the
Me
overpotential, ␣ is the transfer coefficient, n is the number of elec-
trons transferred in the rate-determining step, and ␰ is the power
term of the double-layer capacitance (␰ ϭ 1.0 for a perfectly
smooth electrode and ␰ ϭ 0.5 for a porous electrode͒. By inspec-
tion, the Tafel slope for a reaction occurring on an extremely rough
electrode, which is analogous to a porous electrode, should be
double that of an otherwise identical smooth electrode.1
When studying the kinetics of overpotential deposition, OPD, of
a metal the natural desire is to construct the electrode using the
metal of interest. However, many industrially important metals are
soft and reactive, which makes it difficult to reproducibly prepare
pristine electrodes, particularly single-crystal electrodes. In addition,
the growth of metal deposits, particularly at high current density, can
significantly alter the morphology and surface area of the electrode
on which they occur. During subsequent polarization experiments
the hysteresis in the response of the electrode resulting from the
deposition of rough or dendritic deposits requires that the electrode
be extracted from the electrolytic cell and repolished after each use.
The underpotential deposition ͑UPD͒ of a reactive metal on a noble
UPD of zinc on platinum.—The potential for Zn UPD onto Pt,
ca. Ϫ0.25 V vs. SCE, was identified and studied by Aramata
et al.6-13 The potential shift for Zn UPD on platinum, i.e., the po-
tential difference between bulk Zn deposition and UPD deposition,
ranges from 1.04 to 1.19 V.7 The relatively small variations in the
UPD potential shift in different electrolytes have been attributed to
the effects of different adsorbed species over the UPD layer.
By applying relatively low overpotential pulses for extended
* Electrochemical Society Student Member.
** Electrochemical Society Active Member.
14
´
´
times, Despic and Pavlovic found that the corresponding zinc
deposition current response was independent of time. This demon-
strated that the real surface area of the electrode was independent of
time, suggesting two-dimensional layer-by-layer growth of the zinc
deposit. This result was consistent with the calculated crystallo-
c Present address: Department of Chemical Engineering and Chemical Technology,
Imperial College, London SW7 2BY, United Kingdom.
d Present address: Department of Chemistry, Materials and Chemical Engineering,
Polytechnic of Milan, 20131 Milan, Italy.
z E-mail: bizzotto@chem.ubc.ca
Downloaded on 2014-06-01 to IP 131.211.208.19 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E2
Journal of The Electrochemical Society, 151 ͑1͒ E1-E6 ͑2004͒
graphic Pt-Zn misfit. The misfit (f) is calculated from the inter-
atomic distances ͑d͒ between Zn-Zn, 2.665 Å, and Pt-Pt, 2.775 Å
d
0,Zn Ϫ d0,Pt
2.665 Ϫ 2.775
f ϭ
ϭ
ϭ Ϫ0.04
͓2͔
d0,Pt
2.775
The relatively small negative value of the crystallographic misfit
suggests that, in the absence of specifically adsorbed anions, zinc
can fully occupy interstitial sites on the platinum surface. Charac-
terization of the UPD and alloy surface of the Pt-Zn system are
presented, including an example demonstrating the utility of such a
modified surface for kinetic measurements of reactive metal reduc-
tion.
Experimental
The underpotential deposition of zinc on polycrystalline plati-
num from deoxygenated sulfuric acid and neutral magnesium sulfate
solutions was studied by cyclic voltammetry using a jacketed three-
compartment electrochemical cell with a saturated calomel ͑SCE͒
reference electrode under a nitrogen atmosphere. The temperature
was controlled using a refrigerated circulating bath ͑Fisher Scientific
Isotemp model 900͒. The electrode employed was a model MTI34
͑99.99ϩ %͒ platinum ring-disk electrode ͑RDE, Pine Instrument
Co.͒ with a disk area of 0.283 cm2. The disk surface was polished to
a mirror finish using 1 ␮m diamond slurry. The potentiostat was an
Eco-Chemie PSTAT 30, equipped with BIPOT, FRA and SCAN-
GEN modules. The zinc solutions were prepared using aliquots of a
strong, 2 mol dmϪ3, zinc sulfate solution, which was prepared from
ACS grade zinc sulfate heptahydrate, ZnSO4•7H2O ͑Fluka͒ and Mil-
lipore Milli-Q ultrapure water ͑18 M⍀ cm͒. Impurities were reduced
from the concentrated zinc solution using a platinum flag electrode,
polarized at Ϫ1.0 V vs. SCE for approximately 6 h while the solu-
tion was stirred vigorously under a nitrogen atmosphere. The sulfu-
ric acid and magnesium sulfate ͑Fisher Scientific͒ employed were
ACS grade reagents.
Zinc overpotential deposition was studied by cyclic, normal
pulse, and staircase voltammetry from pure zinc sulfate electrolytes
using the same equipment and reagents as described above. The
normal pulse and staircase voltammetry experiments employed a
base potential of Ϫ0.25 V vs. SCE and were repeated three times at
each of three electrode rotation speeds, 400, 1600, and 3600 rpm.
The applied potentials were corrected for ohmic drop after the ex-
periment by estimating the solution resistance using electrochemical
impedance spectroscopy at Ϫ0.25 V vs. SCE, after all deposited Zn
had been stripped at this potential. The real and imaginary compo-
nents of the complex impedance, Z, were measured during applica-
tion of a 10 mV rms perturbation in a sine wave format in the
frequency range of 100 to 1000 Hz. The frequency response was
modeled as a series resistor-constant phase element, CPE, circuit
according to the equation below
Figure 1. Cyclic voltammograms of polycrystalline platinum in 0.1 M
H2SO4 , pH 0.7, with ͑
͒ and without ͑
͒ 0.001 M ZnSO4 at 25°C,
scan rate 20 mV sϪ1, 1600 rpm electrode rotation speed.
the electrode was conditioned at a potential in the Zn UPD potential
region, Ϫ0.2 V vs. SCE in this case, the stripping current of the
UPD Zn layer, III, diminished, possibly due to formation of a 2D
surface alloy by a site exchange process. Tadjeddine et al.15 has
described alloy formation during Zn UPD on Au and surface alloy
formation resulting from such polarization routines has been re-
ported for several UPD systems including Pb-Au,16,17 Pb-Ag,17
14
Cd-Ag,18 Cd-Au,19,20 Tl-Ag,21 and Sn-Au.22 Despic and Pavlovic,
in fact, had previously described Zn UPD and rapid alloy formation
commencing at ca. Ϫ0.8 V vs. SCE, which was reported as the UPD
potential in 1.5 M ZnSO4 buffered at pH 3.5, from which some zinc
was released from the platinum substrate only as it was oxidized ͑IV
in Fig. 2͒. Electroplating of zinc onto platinum from sulfate solution
has long been determined to result in the formation of a 3D 1:1
Zn-Pt surface alloy.23 The formation of this alloy is so rapid as to be
visible to the naked eye after only 15 min of zinc deposition, which
is reflected in the relatively large diffusion coefficient of Zn into Pt
at 100°C; 5.0 ϫ 10Ϫ15 cm2 sϪ1 ͑ca. 107 times greater than the dif-
fusion coefficient of Zn into Cu͒.23 The 3D alloy resulting from bulk
´
´
Ϫ␰
Z ϭ Rsoln
ϩ
i␻CPE͒
͓3͔
͑
X-ray photoelectron spectroscopy ͑XPS͒ analyses were per-
formed using a Leybold MAX200 spectrometer with a Mg K␣
source ͑1253.6 eV͒ operated at 15 kV, 20 mA, and a system pressure
of 1 ϫ 10Ϫ9 mbar. Spectra were obtained with a pass energy of 48
eV. Binding energies were referenced to the Au 4f1/2 peak at 84.0
eV.
Results
Behavior in sulfuric acid supporting electrolyte.—Voltam-
mograms for the Pt with and without 10Ϫ4 mol dmϪ3 ZnSO4 in 0.1
mol dmϪ3 H2SO4 are shown in Fig. 1. The voltammograms are
similar to those reported by Aramata et al.6 The features identified
are the onset of UPD Zn formation, I at ca. 0 V vs. SCE, H2 evo-
lution, II at Ͻca. Ϫ0.25 V, and stripping of UPD Zn, III at ca. Ϫ0.2
V vs. SCE; the formation and stripping of the UPD Zn layer ap-
peared to be quite reversible. However, as illustrated in Fig. 2, when
Figure 2. Cyclic voltammograms of polycrystalline platinum in 0.1 M
H2SO4 , pH 0.7, with 0.001 M ZnSO4 at 25°C, scan rate 20 mV sϪ1, 0.9 V
vs. SCE rest potential, 1600 rpm electrode rotation speed, as a function of
conditioning time at Ϫ0.200 V vs. SCE, 0 s ͑
͒, 3600 s ͑– –͒.
͒, 900 s ͑- - -͒, 1800 s
͑
Downloaded on 2014-06-01 to IP 131.211.208.19 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 151 ͑1͒ E1-E6 ͑2004͒
E3
Figure 3. Cyclic voltammograms of polycrystalline platinum in 0.1 M
Figure 4. Current density response of polycrystalline platinum in 0.1 M
MgSO4 with 0.0001 M ZnSO4 pH 5.8, at 25°C, to a potential pulse from 0.2
to Ϫ0.4 V vs. SCE, as a function of electrode rotation speed.
MgSO4 , with ͑
͒ and without ͑
͒ 0.0001 M ZnSO4 , pH 5.8, at
25°C, scan rate 20 mV sϪ1, 1600 rpm electrode rotation speed.
zinc deposition cannot be removed with hot nitric or hot hydrochlo-
ric acid, but can be removed using aqua regia ͑which would have
also dissolved the platinum substrate͒.23 The implication is that the
2D surface alloy formed in the UPD region has chemical properties
similar to the bulk 3D alloy.
In a strongly acidic sulfate electrolyte, see Fig. 2, the suppression
of the hydrogen evolution rate was barely evident. Aramata et al.11
have described a shift in the onset of Zn UPD on Pt͑111͒ towards
more negative potentials with decreasing pH in a phosphate support-
ing electrolyte, which was attributed to stronger adsorption of
phosphate species at lower pHs.11 Though this effect was reportedly
absent in sulfate media, the conclusion was based on interpretations
of cyclic voltammograms in the pH range of 0.8 to 3.7.12 At
electrode rotation speeds of 400, 1600, and 3600 rpm. During the
initial stages of the pulses, the currents were determined to vary
directly with the square root of time, which is typical of semi-
infinite linear diffusion during relaxation of the boundary layer. The
plateaus in current responses, which tended to zero at Zn-filled sites
on the Pt surface, were consistent with the mass-transport control of
Zn2ϩ coupled with rapid heterogeneous UPD.25 Zn UPD on poly-
crystalline Pt is reportedly similar to that on Pt͑110͒ ͑the facet on
which Zn UPD occurs to the greatest extent͒.11 The calculated
charge density for the Zn UPD value is somewhat smaller than the
figure of 320 ␮C cmϪ2 reported for Zn UPD formation on Pt͑110͒ in
0.1 mol dmϪ3 KH2PO4 at pH 3.7.11 However, 260 ␮C cmϪ2 is not
unreasonable when considering that the theoretical charge densities
Ͻ
pH
3.3 and in the potential range of sulfate/bisulfate adsorption
for full coverage of Zn on Pt͑110͒ is 294 ␮C cmϪ2
.
on Pt͑111͒, bisulfate is the predominantly adsorbed species while
24
In order to ensure that the calculated charge density was not
overestimated, the polished platinum electrode was characterized
using atomic force microscopy ͑AFM͒. By simply calculating the
sum of the areas of triangles between all groups of three adjacent
points in the array of z position data, the real surface area of the
electrode was estimated to be was only approximately 0.4% greater
than the geometric area.
Ͼ
sulfate is the prevalent at pH
4.7. Thus, it is reasonable to as-
sume that at pHs р 3.7, bisulfate anions may block Zn UPD on
certain regions of the Pt electrode. When the potential was swept
into the region of hydrogen evolution the electrode areas which were
not occupied by Zn would have catalyzed hydrogen evolution,
which accounts for the lack of significant inhibition of hydrogen
evolution at low pH.
A notable feature of the presence of a Zn UPD layer on Pt is its
dramatic effect on hydrogen evolution kinetics, as illustrated in Fig.
5. Even at relatively large overpotentials, e.g., 0.3 V corresponding
to ca. Ϫ1.1 V vs. SCE, at which zinc bulk deposition occurs, the
corresponding hydrogen evolution current density was close to that
observed in the absence of zinc at ca. Ϫ0.8 V. Normally, hydrogen
evolution on platinum at such a high overpotential would be exces-
sive and result in the electrode surface becoming covered in hydro-
gen gas bubbles. The presence of the UPD Zn layer on platinum,
though clearly not to the same extent as pure zinc, has some en-
hanced property of inhibiting hydrogen evolution compared to bare
platinum. It is likely that the relatively neutral pH of the magnesium
sulfate solution is related to increased coverage by the zinc UPD
layer. Mascaro-Lucia et al.26 have described a strong inhibition of
hydrogen evolution due to increased coverage of Zn UPD on poly-
crystalline Pt in acidic fluoride media relative to acidic sulfate me-
dia, for which corresponding stripping charges of the Zn UPD layer
were 350 and 210 ␮C cmϪ2, respectively. From this observation it
was concluded that adsorbed sulfate or bisulfate partially blocked
Zn UPD in acidic sulfate electrolytes and that 350 ␮C cmϪ2 was
closer to the value for complete coverage of the Pt electrode by UPD
Zn. The implication is that for bulk deposition experiments from
sulfate-based electrolytes, a Zn UPD modified platinum electrode
The integration of the charge for the Zn UPD layer, required for
the estimation of the degree of coverage of the platinum, is difficult
because, at this pH, Zn UPD also occurs in the same potential region
as hydrogen adsorption, and it is not clear to what extent Zn UPD
forms preferentially with respect to hydrogen adsorption.
Behavior in magnesium sulfate supporting electrolyte.—Experi-
ments were conducted in neutral magnesium sulfate solution to
maximize the surface coverage of UPD Zn and to separate the re-
gions of UPD Zn and hydrogen adsorption on Pt. As illustrated in
Fig. 3, the potential region of hydrogen adsorption, I, was shifted to
a more negative potential, Ͻca. Ϫ0.4 V vs. SCE, together with bulk
hydrogen evolution, II, which then occurred at Ͻca. Ϫ0.8 V vs.
SCE, and oxidation of adsorbed hydrogen, III, at between ca. Ϫ0.5
to Ϫ0.2 V vs. SCE. After the addition of 10Ϫ4 mol dmϪ3 zinc sul-
fate, the potential region of Zn UPD, IV, between ca. Ϫ0.4 and 0 V
vs. SCE, appears clearly separate from hydrogen adsorption. The
potential range for Zn UPD and the shape of the voltammogram
appear quite similar to Zn UPD on polycrystalline Pt from 0.1 mol
dmϪ3 KH2PO3 buffered at pH 5.9.6
The charge density for Zn UPD formation was estimated to be
260 Ϯ 30 ␮C cmϪ2 determined by integrating the current response
͑Fig. 4͒ to a potential pulses between 0.2 to Ϫ0.4 V vs. SCE at
Downloaded on 2014-06-01 to IP 131.211.208.19 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E4
Journal of The Electrochemical Society, 151 ͑1͒ E1-E6 ͑2004͒
Figure 7. XPS analysis of bare platinum ͑
platinum electrode conditioned at Ϫ0.4 V vs. SCE for 1800 s ͑
͒, offset for clarity, and a
Figure 5. Cyclic voltammetry on polycrystalline platinum in 0.1 M ZnSO4 ,
pH 4.9 at 25°C, scan rate 20 mV sϪ1, 1600 rpm electrode rotation rate.
͒ in 0.1 M
MgSO4 with 0.02 M ZnSO4 , pH 5.0 at 25°C.
should be prepared in a near-neutral solution. This diminishes the
reversible potential and hence the kinetics of hydrogen adsorption
and/or bisulfate adsorption, thereby biasing the competitive adsorp-
tion on Pt in favor of Zn. The upper pH limit is dictated by the
ultimate precipitation of Zn͑OH)2 .
exchange to form a 2D alloy did so within 600 s. When the same
experiment was repeated using a conditioning potential of 0.2 V vs.
SCE, there was no increase in the stripping current in the region of
platinum oxidation as shown by the dotted line in Fig. 6. This was
interpreted as further evidence that the increase in the stripping cur-
rent in the region of platinum oxidation was the result of 2D alloy
formation and not due to adsorption of a contaminant, since the
same contaminant would likely have adsorbed at 0.2 V vs. SCE
before being subsequently oxidized in the region of Pt oxidation. As
shown in Fig. 7, XPS analysis of a freshly polished platinum elec-
trode subjected to a 30 min conditioning period at Ϫ0.4 V vs. SCE
in the presence of zinc ions, confirmed the retention of zinc on the
electrode surface. The XPS analysis did not contain any peaks at-
tributable to the deposition of inorganic impurities which also veri-
fied the purity of the electrolyte.
Voltammetry in aqueous zinc sulfate electrolyte solutions.—In
order to reproduce the surface alloy formed in sulfuric acid support-
ing electrolyte, pH 0.7 ͑Fig. 2͒, the electrode was polarized for vari-
ous times at Ϫ0.4 V vs. SCE in 0.1 mol dmϪ3 ZnSO4 , pH 4.9,
which is slightly more positive than potentials at which hydrogen
evolution occurs. As illustrated in Fig. 6, when the potential was
swept towards more positive potentials, the stripping current in the
UPD Zn region decreased as waiting time increased, indicated by
the down arrow, while current in the region of platinum oxidation
increased, indicated by the up arrow. The effect was limited to ap-
proximately the first 600 s of conditioning, which was interpreted as
the time for transformation of the UPD layer to a surface alloy.
However, although the possibility is not excluded, it is not claimed
that the entire UPD layer was transformed to an alloy, only that
whatever fraction of the surface which would have undergone a site
The effect of conditioning the electrode at Ϫ0.4 V vs. SCE on Zn
OPD was examined by sweeping the potential into the region of
bulk zinc deposition. As illustrated in Fig. 8, there appeared to be a
decrease in the nucleation overpotential for Zn OPD, region II,
which was also limited to the first 600 s of conditioning. This further
Figure 6. Sweep voltammetry of polycrystalline platinum in 0.1 M ZnSO4 ,
Figure 8. Sweep voltammetry of polycrystalline platinum in 0.1 M ZnSO4 ,
pH 4.9 at 25°C, scan rate 20 mV sϪ1, 1600 rpm electrode rotation rate,
• • • •
pH 4.9 at 25°C, scan rate 20 mV sϪ1, 1600 rpm electrode rotation rate,
as a function of waiting time at Ϫ0.4 V vs. SCE, 0 s ͑- - -͒, 120 s ͑
͒,
as a function of waiting time at Ϫ0.4 V vs. SCE, 0 s ͑
͒, 120 s ͑- - -͒,
• • • •
600 s ͑– –͒, 1200 s ͑
͒ and at 0.2 V vs. SCE 600 s ͑
͒.
600 s ͑
͒, 1200 s ͑
͒.
Downloaded on 2014-06-01 to IP 131.211.208.19 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 151 ͑1͒ E1-E6 ͑2004͒
supports the previous assertion of 2D alloy formation. The presence
E5
of a contaminant is an unlikely explanation for this observation be-
cause such an agent would be liable to block sites for zinc deposi-
tion and adversely affect Zn OPD. The rate of hydrogen evolution in
region I of Fig. 8 appeared to be diminished upon conditioning.
Epelboin et al.27 have proposed an autocatalytic mechanism of zinc
electrodeposition from acidic solutions in which zinc adsorption, as
Znϩads , competes with hydrogen adsorption, Hads, as the first step in
zinc electrocrystallization. If this model is accepted, the diminished
hydrogen evolution would be consistent with enhanced zinc deposi-
tion, since hydrogen adsorption precedes hydrogen evolution. These
results suggest that hydrogen adsorption on the Zn-Pt alloy is simply
not as favorable as on the initially formed UPD Zn layer.
If the electrode was oxidized by sweeping the potential to 1.1 V
vs. SCE, then subsequent voltammograms appeared similar to that
of the original unconditioned electrode. If on the other hand, after
conditioning, the electrode potential was swept back to Ϫ0.25 V vs.
SCE and then returned to the potential region of zinc deposition, the
enhancement of zinc deposition kinetics remained. This potential is
significant since it corresponds to the reversible potential for hydro-
gen evolution from a highly acidic solution of pH ca. 0. Thus, by
using Ϫ0.25 V vs. SCE as the stripping potential to avoid hydrogen
evolution, it should be possible to extract current efficiencies for
bulk zinc deposition from a wide composition range of acidic zinc
sulfate electrolytes by integrating the current-time transients in re-
sponse to potential steps for zinc deposition and stripping, and cal-
culating the ratio of deposition to stripping charges. However, as
mentioned, rapid 3D alloy formation occurring during Zn OPD leads
to extreme roughening of the Pt substrate which is visible to the
naked eye after extended polarization times. Limiting the Zn OPD
polarization times to Ͻ2 min was determined to be sufficient for
mitigating 3D alloy formation and associated substrate roughening.
The real surface area of the electrode, measured from AFM images,
after 100 cycles of Zn OPD and stripping increased by only ca. 2%.
Nevertheless, it is clear that 3D alloy formation can significantly
alter the crystallography of the substrate, and this must be kept in
mind when performing experiments using single-crystal electrodes.
Figure 9. Tafel plot for zinc deposition from 0.10 mol kgϪ1 ZnSO4 , pH 4.9
at 25°C, 400 ͑᭹͒, 1600 ͑᭿͒, and 3600 rpm ͑ࡗ͒ electrode rotation speed.
and SO24Ϫ at infinite dilution͒ and electrolyte viscosities and densi-
ties were calculated using empirical expressions reported by Umetsu
et al.29
Initially, all the data were fit simultaneously to obtain the values
for zinc exchange current densities, j0,Zn , corresponding to each
zinc sulfate concentration, as well as a value of the product of the
transfer coefficient, ␣, and the number of electrons transferred in the
rate determining step, n. The fitting method was a weighted least-
squared fit30 of the differences in the natural logarithms of the ob-
served and calculated current densities, using the inverse of the un-
certainty in overpotential as the weighting parameter. Using this
method, the best-fit value for ␣n was 1.08, which was close to the
theoretical value of 1.0 ͑Tafel slope of 60 mV decϪ1 at 25°C͒. Due
to covariance between the transfer coefficient, ␣, and exchange cur-
rent density, j0,Zn , a value of 1.0 was adopted for ␣n and the cor-
responding zinc exchange current densities were recalculated to be
21 Ϯ 5 A mϪ2, 23 Ϯ 7 A mϪ2, and 28 Ϯ 5 A mϪ2, for 0.1 mol
kgϪ1 ZnSO4 , 0.5 mol kgϪ1 ZnSO4 , and 1.00 mol kgϪ1 ZnSO4 ,
respectively. These results are plotted in Fig. 10 along with previ-
Tafel measurements for aqueous zinc sulfate electrolyte solu-
tions.—The kinetics of zinc deposition from pure zinc sulfate elec-
trolytes, ca. pH 4.9 at 25°C, were examined by normal pulse and
staircase voltammetry as a function of zinc sulfate concentration, for
ZnSO4 concentration levels of 0.1 mol kgϪ1 ZnSO4 , 0.5 mol kgϪ1
ZnSO4 , and 1.00 mol kgϪ1 ZnSO4 . The current efficiencies for zinc
deposition at the disk electrode were always 100% at all applied
overpotentials, whether employing the ring-disk electrode to correct
the disk current by detecting hydrogen evolution at the ring elec-
trode, or by amperometric integration of the normal potential pulse
sequences. The results from staircase voltammetry generally
matched well with normal pulse experiments and, hence, were
treated as replicates of the same experiment. A typical Tafel plot is
shown in Fig. 9 where the solid lines correspond to a fit according to
the extended Butler-Volmer equation. The limiting current densities
for zinc deposition were calculated using the Levich equation. The
diffusion coefficients for Zn2ϩ, DZؠ n2ϩ , were estimated from the
following expression
DؠZn
DؠSO
DZnSO
2ϩ
4
DZn2ϩ
Ϸ
1 ϩ
ϭ 0.84DZnSO
͓4͔
4
ͫ
ͬ
2
2Ϫ
4
where DZnSO is the diffusion coefficient for zinc sulfate from the
4
data set of Awakura et al.,28 and DZؠ n2ϩ (0.703 ϫ 10Ϫ9 m2 sϪ1͒, and
Figure 10. Zinc exchange current density in sulphate based electrolytes at
DSؠ O2Ϫ (1.065 ϫ 10Ϫ9 m2 sϪ1͒ are the diffusion coefficients of Zn2ϩ
31
25°C, from ͑᭹͒ this study, ͑᭝͒ Zouari and Lapique,32 ͑ᮀ͒ Jovic et al., ͑छ͒
´
14
Despic and Pavlovic, ͑᭺͒ Hurlen and Brevik,33 ͑ ͒ Budov and Losev,34 ͑ϫ͒
4
´
´
*
and SO24Ϫ at infinite dilution. The transport number of Zn2ϩ was
Kahanda and Tomkiewicz,35 ͑᭿͒ Parsons,36 and ͑ϩ͒ Sierra-Alcazar and
fixed at 0.4 ͑also calculated using the diffusion coefficients of Zn2ϩ
Harrison,37 plotted against zinc concentration.
Downloaded on 2014-06-01 to IP 131.211.208.19 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E6
Journal of The Electrochemical Society, 151 ͑1͒ E1-E6 ͑2004͒
Z
⌫
␰
complex impedance, ⍀
ously reported zinc exchange current densities in other weakly
acidic sulfate electrolytes. Epelboin et al.,27 have reported that ad-
dition of even a small amount of acid to a zinc electrolyte causes a
shift in the polarization curve to higher cathodic polarizations, i.e., a
small concentration of protons in the solution lowers the zinc ex-
change current density. Thus, the sensitivity of zinc deposition ki-
netics to acidity probably contributed to the scatter of previously
reported zinc exchange current densities in Fig. 10 which were mea-
sured in electrolytes with pHs ranging from 3.0 to 6.5. Even so, the
surface concentration, mol mϪ2
power term of constant phase element
frequency, rad sϪ1
␻
References
1. R. de Levie, J. Electroanal. Chem., 281, 1 ͑1990͒.
2. E. Budevski, G. Staikov, and W. J. Lorenz, Electrochim. Acta, 45, 2559 ͑2000͒.
3. E. Budevski, G. Staikov, and W. J. Lorenz, Electrochemical Phase Formation and
Growth: An Introduction to the Initial Stages of Metal Deposition, VCH Publishers,
New York ͑1996͒.
4. W. Lehnert and W. Schmickler, J. Electroanal. Chem., 310, 27 ͑1991͒.
5. C. Sanchez and E. Leiva, J. Electroanal. Chem., 458, 183 ͑1998͒.
6. A. Aramata, Md. A. Quaiyyum, W. A. Balais, T. Atoguchi, and M. Enyo, J. Elec-
troanal. Chem., 338, 367 ͑1992͒.
7. Md. A. Quaiyyum, A. Aramata, S. Moniwa, S. Taguchi, and M. Enyo, J. Electroa-
nal. Chem., 373, 61 ͑1994͒.
8. S. Taguchi and A. Aramata, J. Electroanal. Chem., 396, 131 ͑1995͒.
9. A. Aramata, S. Terui, S. Taguchi, T. Kawaguchi, and K. Shimazu, Electrochim.
Acta, 41, 761 ͑1996͒.
10. G. Horanyi and A. Aramata, J. Electroanal. Chem., 434, 201 ͑1997͒.
11. A. Aramata, S. Taguchi, T. Fukuda, M. Nakamura, and G. Horanyi, Electrochim.
Acta, 44, 999 ͑1998͒.
12. S. Taguchi and A. Aramata, J. Electroanal. Chem., 457, 73 ͑1998͒.
13. K. Igarashi, A. Aramata, and S. Taguchi, Electrochim. Acta, 46, 1773 ͑2001͒.
14. A. R. Despic and M. G. Pavlovic, Electrochim. Acta, 27, 1539 ͑1982͒.
15. A. Tadjeddine, M. Ladouceur, A. Lahrichi, D. Guay, and G. Tourillon, in X-Ray
Methods in Corrosion and Interfacial Electrochemistry, A. J. Davenport and J. G.
Gordon, Editors, PV 92-1, p. 159, The Electrochemical Society Proceedings Series,
Pennington, NJ ͑1992͒.
exchange current densities measured in this study matched relatively
31
´
well with those recently reported by Jovic et al. ͑pH 3.5͒ and
Zouari and Lapique32 ͑pH 4.8͒ using pure zinc electrodes, confirm-
ing the viability of the present methodology.
Conclusions
The UPD of Zn onto polycrystalline Pt creates a suitable surface
on which to study overpotential deposition of zinc. This UPD layer
has been characterized in acidic and neutral sulfate solutions using
voltammetric charge and XPS measurements. The charge density for
UPD Zn deposition on platinum from a neutral sulfate solution was
estimated to be 260 Ϯ 30 ␮C cmϪ2, which is in the order of a full
monolayer. A decrease in H2 evolution rates was also described as
evidence of the presence of a Zn UPD layer. The UPD layer was
found to evolve into a Zn-Pt alloy when the electrode was polarized
at potentials more negative than the UPD potential. The alloy was
oxidized at the same potential as the oxidation of Pt, illustrating the
small crystallographic misfit and large interaction between Zn and
Pt. The alloy also decreased H2 evolution rates and enhanced zinc
OPD kinetics relative to the freshly deposited UPD layer. The use of
Zn UPD on Pt as a substrate for OPD kinetic experiments was
determined to yield values for kinetic parameters in agreement with
those recently reported using pure Zn electrodes. The application of
underpotential deposition, UPD, of the reactive metal onto a noble
foreign substrate is proposed as a general method for generating
well-defined and reproducible surfaces on which to study overpoten-
tial deposition.
´
´
´
¨
16. H. J. Pauling, G. Staikov, and K. Juttner, J. Electroanal. Chem., 376, 179 ͑1994͒.
17. U. Schmidt, S. Vinzelberg, and G. Staikov, Surf. Sci., 348, 261 ͑1996͒.
18. V. D. Jovic and B. M. Jovic, Electrochim. Acta, 47, 1777 ͑2002͒.
19. R. Vindu and S. Hara, J. Electroanal. Chem., 475, 171 ͑1999͒.
20. G. Inzelt and G. Horanyi, J. Electroanal. Chem., 491, 111 ͑2000͒.
´
´
´
¨
21. D. Carnal, P. I. Oden, U. Muller, E. Schmidt, and H. Siegenthaler, Electrochim.
Acta, 40, 1223 ͑1995͒.
22. J. W. Yan, H. Tang, Y. Y. Yang, J. W. Wu, Z. X. Xie, S. G. Sun, and B. W. Mao,
Surf. Interface Anal., 32, 49 ͑2001͒.
23. H. H. Uhlig, J. S. MacNairn, and D. A. Vaugn, Acta Metall., 3, 302 ͑1955͒.
24. A. Lachenwitzet, N. Li, and J. Lipkowski, J. Electroanal. Chem., 532, 85 ͑2002͒.
25. S. Swathirajan and S. Bruckenstein, J. Electrochem. Soc., 129, 1202 ͑1982͒.
26. H. Mascaro-Lucia, M. C. Santos, S. A. S. Machado, and L. A. Avaca, J. Braz.
Chem. Soc., 13, 529 ͑2002͒.
27. I. Epelboin, M. Ksouri, and R. Wiart, J. Electrochem. Soc., 122, 1206 ͑1975͒.
28. Y. Awakura, T. Doi, and T. H. Majima, Metall. Mater. Trans. B, 19, 5 ͑1988͒.
29. Y. Umetsu, Q. Su, and K. Tozawa, Shigen to Kankya, 104, 829 ͑1988͒.
30. D. C. Montgomery, Design and Analysis of Experiments, 4th ed., John Wiley &
Sons, Inc., New York ͑1997͒.
List of Symbols
CPE constant phase element, S1/␰
d
E
F
F
i
interatomic distance, m
´
´
´
´
´
electrode potential vs. reference electrode, V
crytallographic misfit
31. V. D. Jovic, J. N. Jovicevic, M. G. Pavlovic, and R. M. Stevanovic, J. Serb. Chem.
Soc., 55, 341 ͑1990͒.
Faraday constant, 96,485, C molϪ1
current density, A mϪ2
32. I. Zouari and F. Lapique, Electrochim. Acta, 37, 439 ͑1992͒.
33. T. Hurlen and T. R. Breivik, Acta Chem. Scand., 32, 447 ͑1961͒.
34. G. M. Budov and V. V. Losev, Russ. J. Phys. Chem., 37, 784 ͑1963͒.
35. G. L. M. K. S. Kahanda and M. Tomkiewicz, J. Electrochem. Soc., 136, 1497
͑1989͒.
i0 exchange current density, A mϪ2
m
NA
molal concentration, mol kgϪ1
Avogadro number, 6.0221367 ϫ 1023, molϪ1
Rsoln solution resistance, ⍀
charge number
36. R. Parsons, Handbook of Electrochemical Constants, Butterworths, London ͑1959͒.
37. H. B. Sierra-Alcazar and J. A. Harrison, Electrochim. Acta, 22, 627 ͑1977͒.
z
Downloaded on 2014-06-01 to IP 131.211.208.19 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).