Electromicrogravimetric study of underpotential deposition of Co on textured gold electrode in ammonia medium

Article abstract of DOI:10.1039/b704891b

Formation of a layer of a metal M on a foreign metal substrate, S, at potentials positive to its reversible potential, E_r, (underpotential deposition, UPD) is a phenomenon that appears only in some systems. In the case of the UPD process of Co on a gold substrate, there is still a controversy about its existence, for which reason, in this study, voltammetry and chronoamperometry were coupled with a quartz microbalance in ammonium solution to better understand its formation mechanism. The results obtained show that the Co forms only one layer on the substrate before the bulk deposition of cobalt occurs. During the UPD process, the Co atom completely transfers its two electrons to the electrode and is adsorbed as a discharged species. The monolayer charge density, determined by electromicrogravimetry, is 448 μC cm^-2 and probably corresponds to a commensurate overlayer on the gold substrate. In addition, the UPD process is controlled by adsorption as shown by voltammetric experiments. Finally, the analysis of the process of Co monolayer destruction (desorption) shows a mechanism different from that in the formation process, possibly due to H adsorption on the substrate produced during the process of adsorption of Co. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique.

Full text of DOI:10.1039/b704891b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Electromicrogravimetric study of underpotential deposition of Co on
textured gold electrode in ammonia medium
Antonio Montes-Rojas,* Luz Marıa Torres-Rodrıguez and Cesar Nieto-Delgado
´ ´
Received (in Montpellier, France) 3rd April 2007, Accepted 8th June 2007
First published as an Advance Article on the web 27th June 2007
DOI: 10.1039/b704891b
Formation of a layer of a metal M on a foreign metal substrate, S, at potentials positive to its
reversible potential, E , (underpotential deposition, UPD) is a phenomenon that appears only in
r
some systems. In the case of the UPD process of Co on a gold substrate, there is still a
controversy about its existence, for which reason, in this study, voltammetry and
chronoamperometry were coupled with a quartz microbalance in ammonium solution to better
understand its formation mechanism. The results obtained show that the Co forms only one layer
on the substrate before the bulk deposition of cobalt occurs. During the UPD process, the Co
atom completely transfers its two electrons to the electrode and is adsorbed as a discharged
ꢀ
2
species. The monolayer charge density, determined by electromicrogravimetry, is 448 mC cm
and probably corresponds to a commensurate overlayer on the gold substrate. In addition, the
UPD process is controlled by adsorption as shown by voltammetric experiments. Finally, the
analysis of the process of Co monolayer destruction (desorption) shows a mechanism different
from that in the formation process, possibly due to H adsorption on the substrate produced
during the process of adsorption of Co.
strates. These authors found that a cobalt adlayer is formed on
a polycrystalline gold electrode during application of potential
in the UPD region. In addition, formation of this cobalt
adlayer involves the simultaneous presence of both adsorption
and 2D nucleation mass transfer-controlled processes. More
1
Introduction
The underpotential deposition (UPD) is defined as the process
of deposition of one metal (M) onto a dissimilar metal
substrate (S) at a positive Nernst potential for bulk deposition,
and it is undoubtedly one of the most extensively studied
processes in surface electrochemistry. A metal UPD process
involves not only a unique possibility to study the first step of
the electrocrystallization of a metal on a foreign substrate, but
also allows variation of its coverage as a function of the
potential to be used in electrocatalysis.
21
recently, Flis-Kabulska studied Co electrodeposited on
4 2 4
Au(111) from CoSO solutions in 0.5 mM H SO . Using
STM and cyclic voltammetry he showed that the electrodepo-
sition started at underpotential and that the substrate was
covered by about one monolayer. In the second viewpoint,
22
2+
Schindler et al. have mentioned that in the Au/Co system
there is a relatively weak substrate–deposit interaction com-
pared to the strong interaction characteristic of underpotential
UPD studies have been carried out with polycrystalline
1
,2
metal electrodes, well-defined single crystal surfaces or,
3,4
5
,6
more recently, with highly ordered electrodes. Numerous
bimetallic systems were investigated, mainly because of their
unique structural, electronic, magnetic and chemisorption
properties compared with either of the bulk metal
components.
23
deposition systems. In this same viewpoint, Kleinert et al.
investigated the initial stages of Co deposition onto Au(111)
and Au(100) electrodes in neutral sulfate solutions and they
reported that the UPD process could be excluded. These
authors reported that Co formed monoatomic high islands
only at the very beginning of growth. Further deposition on
top of these islands starts rather quickly, leading to a twin
layer which soon covers the whole surface. Finally, after
completion of the twin layer, the third layer starts growing.
These results are interesting since it is known that the UPD
process of a metal onto a substrate occurs when the work
function of the latter is larger than that of the deposited metal.
In this case this rule is observed since the value of the work
In recent years, Co monolayers on Au or Ag have received
attention from several authors and this system has been
7
–17
studied by different techniques
because of the giant mag-
netoresistance properties of Co. Therefore, this metal is certain
to be placed at the center of many studies in the coming years.
An exciting issue concerning this metal is its UPD process
onto gold substrates which has been considered from two
1
8–20
viewpoints. In the first, Mendoza-Huizar et al.
reported a
series of studies relating to cobalt UPD onto different sub-
24
function of the substrate is superior to that of the cobalt
(
4.78 and 4.70 eV, respectively).
Laboratorio de Electroquı ´m ica, Centro de Investigacio´n y Estudios de
Posgrado, Facultad de Ciencias Quı´micas, Universidad Auto´noma de
San Luis Potosı´, Av. Dr Manuel Nava No. 6, Zona Universitaria,
SLP, SLP Me´xico 78210. E-mail: antonio.montes@uaslp.mx; Fax:
Despite these results, the question of whether or not the
initial stage of electrochemical deposition of Co on Au
proceeds at underpotentials remains open and in this sense,
we propose to use a special tool that could clarify this issue.
(0052) 48262371; Tel: (0052) 48262372
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1769–1776 | 1769

View Article Online
Many techniques can be used to obtain information on this
kind of system, nevertheless, the electrochemical quartz-
crystal microbalance (EQCM) has made a significant impact
in recent years as an in situ technique, capable of measuring
the mass increase or decrease at the electrode surface as a
Potentials were measured against and are quoted versus the
Ag|AgCl|3 M NaCl reference electrode and a platinum wire
was used as counter-electrode. All the experiments were
performed at room temperature.
2
5–27
function of potential
using the Sauerbrey expression:
Dm = ꢀC Df (1)
f
3
Results and discussion
f
where C , the sensitivity constant,
3
.1 Substrate characterization
Fig. 1 shows a typical current–potential (I–E) curve obtained
at a Au film electrode supported on quartz in 0.5 M H SO
4
2
2
fo
ꢀ
1
C
¼
f
1=2
2
ðr m Þ
q
q
after the cyclic voltammetric procedure. This voltammogram
possesses certain characteristics such as the very small current
observed in the zone A which corresponds to charging of the
electrical double layer and this characteristic is a criterion for
the cleanliness of the surface of our electrodes. In addition, the
current peaks that appear in the zone B are associated with the
processes of oxide formation and reduction on gold in the
depends only on the properties of the quartz resonator itself:
and r are the shear modulus and density of quartz,
respectively, and fo is the fundamental frequency of the
m
q
q
resonator.
This equation holds only under certain conditions such as
that the resonant frequency shifts by no more than 2% of the
fundamental.
29
presence of specific adsorption.
30
31
ˇ
According to Hamelin et al. and Strbac et al., the
position of anodic peaks on the potential axis and their shape
are typical of each crystalline orientation, for example
the most important anodic current peak in the region B at
These properties of EQCM permitted the UPD systems to
become popular targets of EQCM study with a wide variety of
substrates, metals and electrolytes.
In this paper we report new results obtained on this system
when the Co UPD process is examined by cyclic voltammetry
and chronoamperometry coupled with a quartz-crystal micro-
balance.
+
1442 mV is characteristic of the oxidation of terraces of the
(
111) orientation (less active sites). However, this peak does
not have such an important intensity as that of a true mono-
crystal, perhaps because our electrodes are made up of small
grains that produce a very rugged surface and consequently a
decrease in its intensity. In addition, the small prepeaks
probably correspond to the formation of an OH monolayer
2
Experimental
3
2
The test solution was prepared from Co(NO
3
)
2
(Baker) in
ꢀ
4 2 4
M (NH ) SO (Sigma) 1 M concentration, and ultrapure
on defects (more active sites) or on other plane crystallines.
Furthermore, cyclic voltammetry carried out in a 5 mM
CuSO + 0.5 M H SO solution revealed the two-stage
formation of a Cu UPD layer typical of a well-defined Au
3
1
0
water. The pH of the solutions was fixed with NaOH (Sigma)
at 9.3 to diminish interferences of hydrogen evolution and to
4
2
4
2+
have the predominant species [Co(NH ) ] . Prior to each
33,34
{111} substrate
and not only one current peak distinctive
3
3
6
5,36
experiment the solution was degassed by bubbling N through
2
of a polycrystalline substrate.
Also, the change of fre-
the solution for at least 10 min, during which the atmosphere
was saturated with this gas.
quency–potential response [Fig. 2] showed a behavior similar
to that reported by Borges et al. and Watanabe et al., who
3
used textured gold substrates.
7,38
The electromicrogravimetric measurements were performed
with a PAR model 273A coupled with a quartz microbalance
Seiko model QCA922.
Finally, our X-ray pattern of the thin gold layer showed a
big peak at 38.37 (2y), which is associated with the plane
mentioned above {111}. So, all these results show that our
electrodes have a textured surface in the direction {111}.
EQCM experiments were performed with polished AT-cut
gold-coated 9 MHz quartz crystals (Seiko) of mass sensitive
2
area (Sgeo) 0.196 cm . Before each experiment, all the electro-
des were subjected to a cyclic voltammetric treatment in 0.5 M
H SO . Next, the electrodes were transferred to the test
2
4
solution. This procedure gives rise to reproducible and well-
defined voltammetric responses, which bear witness to the
quality of the electrodes and the cleanliness of the working
solutions.
The sensitivity factor in the Sauerbrey equation (eqn (1))
was determined by chronoamperometric silver deposition and
its value was very close to the theoretical value (1.714 ꢂ 0.012
8
ꢀ1
2
ꢃ10 Hz g cm ). The electroactive area (Select) of these
working electrodes was determined using the adsorption–
2
8
desorption process of oxygen onto gold. Using Sgeo and
Select the roughness factor was calculated, defined by
r = Select/Sgeo which was 1.25 ꢂ 0.04.
Fig. 1 Current–potential (I–E) curve for Au EQCM film obtained in
ꢀ1
0.5 M H SO : scan rate, 50 mV s . The geometric surface (Sgeo) is
2
4
2
0.1965 cm .
1
770 | New J. Chem., 2007, 31, 1769–1776
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007

View Article Online
Fig. 2 Change of frequency (Df) as a function of potential of a Au
EQCM electrode in 5 mM CuSO + 0.5 M H SO solution. The scan
Fig. 4 Some experimental Df–t curves recorded in the UPD zone of
Au in 1 M (NH ) SO (pH 9.3) solution by means of a program of
4
2
4
ꢀ1
rate was 5 mV s
.
4 2
4
three potential steps. In all cases the starting potential and the final
potential (step 2) were +250 mV. The second potential step (step 1)
was (a) 0 mV, (b) ꢀ200 mV, (c) ꢀ400 mV, (d) ꢀ450 mV and
3
.2 Voltammetric characterization of Co UPD on Au
A large potential zone was scanned using the electrodes of
EQCM in (NH SO 1 M solution adjusted to pH 9.3 in the
(
e) ꢀ500 mV.
4
)
2
4
absence of metallic ion with the aim of establishing the effect
of the support electrolyte on the voltammogram and the
variation of frequency vs. potential curve, Df–E. These curves
are shown in Fig. 3.
At this point of the study, it was proper to wonder if these
3
responses had a contribution from adsorption of NH mole-
cules on the electrode. To clear up this doubt, potential pulses
were applied in this region and the Df response was recorded
(Fig. 4).
Between +350 mV and ꢀ650 mV, the curve j–E shows a
ꢀ2
small constant current of about ꢂ0.56 mA cm relative to the
Fig. 4 shows only some of the curves obtained in these
experiments. As may be observed, the common feature of the
curves is that they all have the same behavior, i.e., frequency
variation is practically constant during the pulse time and
independent of the applied potential. The Df value recorded
was ꢂ3 Hz during all experiments, which represents a mini-
mum variation of Df, equivalent to that obtained in the
potential scan toward negative potentials from Fig. 3.
These results confirm that frequency variation as a conse-
electrical double layer charge. At potentials more negative
+
than ꢀ650 mV, the reduction of H to H starts to take place
2
at the substrate. This process has been associated with an
increase in the cathodic current in the voltammogram.
In the case of the variation of frequency versus potential
curve, (—) Df–E, in the beginning of the direct potential scan
at 0 mV, the frequency variation value is only of ꢂ2 Hz up to
ꢀ
700 mV. This small change of frequency value implies that
there are no processes with a mass change produced in this
potential region. At potentials more negative than ꢀ700 mV
the Df decreases up to ꢀ17 ꢂ 2 Hz. It is important to mention
that this value is not really associated with an increase of the
mass on the substrate but possibly with a change of viscosity
quence of the process of adsorption of NH
3
species on the
electrode is not important in the UPD region.
The next step was to define the UPD region of the Co/Au
system, for which reason it was necessary to obtain the I–E
curve in the presence of metallic ion in solution. Fig. 5 shows a
typical I–E curve obtained at a Au electrode in the presence of
+
on the substrate resulting from the process of reduction of H
ꢀ
3
to H
2
which begins at about ꢀ780 mV. As the potential scan is
Co(II) 10 M.
reversed the behavior of the Df becomes almost constant at
The potential region was scanned between +100 mV and
ꢀ1100 mV. The experiment started at +100 mV up to
ꢀ1100 mV and then it was reversed to the starting potential.
During the direct potential scan, only one current peak
appeared at ꢀ480 mV associated with the Co UPD process
ꢀ
8 ꢂ 2 Hz. Only when the potential scan is about to end is the
value of Df the same as in the beginning.
Fig. 3 Voltammogram, j–E, (—) and variation of frequency versus
potential curve, Df–E, (—) of the gold electrode obtained in the
ꢀ1
supporting electrolyte (NH
4
)
2
SO
4
1 M pH 9.3. The scan rate was
Fig. 5 Voltammogram at 5 mV s for Au EQCM electrodes in
(NH SO 1 M (pH 9.3) + 1 mM Co(II).
ꢀ1
5
mV s
.
4
)
2
4
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1769–1776 | 1771

View Article Online
(
peak B). It is important to mention that this voltammetric
response is very similar to that reported by Huizar et al. in the
same Co/Au system using monocrystalline substrate
which confirms the ordered surface of our
a
1
8–20
Au(111),
electrodes. In addition, an important characteristic of this
response is that the unique current peak associated with the
UPD process of cobalt reveals that the overlayer forma-
tion–destruction processes have a different nature from, for
example, the Tl UPD system which exhibits two complemen-
tary peaks for adsorption–desorption processes in the UPD
3
9
region. As the electrode potential changes to more negative
values, the current increases rapidly due to the bulk cobalt
deposit on the Au electrode and simultaneously due to the
hydrogen evolution reaction. On the other hand, during the
inverse potential scan, two crossovers of the cathodic and
anodic branches at ꢀ962 mV and ꢀ754 mV can be seen. The
more cathodic crossover (ꢀ962 mV) is associated with the
Fig. 7 Plots of the experimental cathodic peak potential as a function
of scan rate (v) for peak B (see Fig. 5).
adsorption and confirms that the peak B corresponds to a Co
UPD process on the gold electrode.
4
0
nucleation process of a new phase and the parameters of the
nucleation-growth mechanism can be taken. At potentials more
negative than ꢀ962 mV, lower current is found during the scan
toward positive potentials than toward negative potentials,
probably due to the changes in the interface concentration of
In addition, Fig. 7 shows the behavior of the cathodic peak
potential, Epc, as a function of the sweep rate in the range of
ꢀ1
5
–100 mV s . We can see that the peak shifts in the negative
direction with an increasing sweep rate, as a result of the
42,46
irreversible behavior of the system.
2+
]
40,41
[
Co(NH
)
3 6
ions, as a result of the deposition process.
In relation to this discussion it is important to say that
23
Kleinert et al. have reported a clear pre-wave for the Co
However, in the range between ꢀ962 mV and ꢀ754 mV, the
current shows an opposite behavior in both scan directions
because the energy required for cobalt deposition on the cobalt
film, during the forward sweep, is lower than that required for
cobalt deposition on the Au electrode modified by the UPD Co.
As the electrode potential was changed toward more posi-
tive values, only one peak of anodic current at ꢀ620 mV was
observed (peak A). This peak is related to the oxidation of
cobalt deposited previously during the direct potential scan.
In order to ensure that the cathodic peak at ꢀ480 mV (peak
B) actually occurs due to the UPD process of Co onto Au, a
previous analysis of this peak as a function of scan rate was
carried out. We obtained I–E curves at different scan rates and
we prepared curves of current peak as a function of scan rate
as shown in Fig. 6.
deposition on Au(100) around ꢀ0.8 V versus SCE, which is
only slightly discernible for Au(111). They propose that this
wave may originate from an ‘‘adsorption’’ of Co species at
defects because the pre-wave is not observed in pure electrolyte
solution, and it is almost absent for a defect-free Au(111)
2+
electrode in the Co containing solution. They found that the
charge due to this process corresponds to about 1% of a ML
only.
In this sense, we think that this process of adsorption
corresponds to the Co UPD process which can be promoted
by the defects or by the grain boundary localized in our
3,47,48
electrodes.
3
.3 Electromicrogravimetric characterization of Co UPD
In this figure, a linear relationship between the cathodic
peak current density (j) and the scan rate (v) associated with
the peak B can be observed. This tendency implies that the Co
atoms are adsorbed on different types of sites present at the
3.3.1 Electromicrogravimetric response. In Fig. 8 two
curves are shown: the j–E curve (a) has been described above
and the curve (b) corresponds to the change of the frequen-
cy–potential (Df–E) curve recorded simultaneously with (a).
According to this latter curve, during the negative scan from
4
2–45
substrate, as has been pointed out by different authors.
In
this case, it shows that the UPD process is controlled by
+
100 mV to ꢀ400 mV, there is no detectable change in
frequency and the current detected in the voltammogram
corresponds to double-layer charging. However, a smooth
variation of slope of frequency change is observed at poten-
tials more negative than ꢀ400 mV but more positive than
ꢀ
907 mV. We think that this behavior of the change of
frequency in this potential zone is associated with the forma-
tion process of a UPD layer of Co on the substrate. It is
important to mention that at this potential (ꢀ907 mV) the
change of frequency is ꢀ23.8 Hz (mass increases) and we think
that this value is related to the mass of the UPD monolayer.
On continuing the negative scan at potentials more negative
than ꢀ907 mV, a rapid decrease in frequency (change of slope)
can be observed associated with the formation of a massive
deposit of Co on the UPD monolayer. This process still
Fig. 6 Plot of the experimental cathodic peak current density (j) as a
function of scan rate (v) for peak B (see Fig. 5).
1
772 | New J. Chem., 2007, 31, 1769–1776
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007

View Article Online
Fig. 8 (a) Current density–potential (j–E) curve for Au EQCM film
ꢀ ꢀ1
3
Fig. 10 Some experimental Df–t curves recorded in the UPD of Co
onto Au in 1 mM Co(II) + 1 M (NH ) SO (pH 9.3) solution by means
4 2 4
in [(NH
b) Change in frequency–potential (Df–E) curve for Au EQCM film in
4 2 4
) SO ] 1 M pH 9.3 + [Co(II)] 10 M: scan rate, 5 mV s .
(
ꢀ ꢀ1
3
of a program of three potential steps. In all cases the starting potential
and the final potential were +250 mV. The second potential step was
[
4 2 4
(NH ) SO ] 1 M pH 9.3 + [Co(II)] 10 M: scan rate, 5 mV s . The
curve (b) was obtained concomitantly with the j–E curve in (a).
(
a) ꢀ300 mV, (b) ꢀ450 mV, (c) ꢀ550 mV, (d) ꢀ650 mV, (e) ꢀ700 mV,
f) ꢀ750 mV, (g) ꢀ800 mV, (h) ꢀ910 mV.
(
continues when the scan reversal is produced at ꢀ1100 mV
since the change of frequency keeps decreasing. Following
scan reversal the frequency remains constant between ꢀ911
mV and ꢀ640 mV before increasing (mass decreases) very
rapidly in the region ca. ꢀ640 to ꢀ450 mV corresponding to
anodic dissolution of the Co deposit. Finally, the destruction
of the Co monolayer on the substrate takes place since the
frequency increases (mass decreases) slowly at potentials more
positive than ꢀ450 mV until the original value of Df is attained
in the double-layer region.
According to the behavior of the change of frequency, these
curves are classified in two categories:
I. The change of frequency (mass) is constant at any time
(
curves a–e).
II. The change of frequency decreases (mass increases)
slowly as a function of time (curves f–h).
For the first category, the value of Df is almost zero
when the step potential is more positive than the potential of
peak B (curves a and b), which implies that the mass is not
deposited on the substrate. On the contrary, when the
step potential is more negative than the potential of the peak
B (curves c and d), the value of Df (mass) is constant at any
step time and Df is only a function of the potential step.
This behavior of the Df implies that the substrate coverage is a
function of the energy of the adsorption sites, for example,
defect-like sites. It is important to state that these results are
opposite to Kleinert’s, because if the process of Co deposition
occurs via a 2D nucleation mechanism then the response of the
Df should slowly diminish as a function of time.
Additional curves Df–E were obtained at different negative
potential limits which are shown in Fig. 9.
These curves Df–E show behavior the same as that described
for the curve b in Fig. 8. Furthermore, it is important to
mention that the Df attain a plateau that depends on the
cathodic limit potential (see curves a and b) and this plateau is
only observed when the cathodic limit potential is more
negative than ꢀ900 mV (see curve c). We assume that this
plateau is associated with the formation of bulk deposition of
Co onto the substrate when the limit potential is more negative
than the UPD region. So the UPD region is limited by
potentials more positive than ꢀ900 mV. We obtained the
curves of frequency change (Df) versus time (t) at different
potential steps in the supposed UPD region which are shown
in Fig. 10.
On the other hand, for the second category, the Df decreases
mass increases) slowly at potential steps more negative than
(
ꢀ
700 mV but more positive than ꢀ800 mV, curves e–g. This
behavior, perhaps, is due to the fact that cobalt atoms occupy
other adsorption sites, such as reconstruction defects or
defects present on terraces.
Finally, the curves g and h present a behavior typical
of a growing mass of a metal on an electrode resulting from
the gradual decrease of Df (mass increase) as a function of the
step potential due to the bulk deposition occurring at poten-
tials more negative than ꢀ800 mV. In this case the process
depends on the time of the potential step. It is proper to
mention that in this part of the deposit formation process,
cobalt may either grow layer by layer, as stated by other
authors, or the Co monolayer formed in the UPD zone may
give rise to an even more compact structure. However, it is
clear that any of these processes develops from a full cobalt
monolayer deposited on the substrate, because the Df changes
at an early point of the step potential to a value of approxi-
mately ꢀ23 Hz, which is in agreement with what was deter-
mined by voltammetry.
Fig. 9 A set of experimental curves Df–E obtained simultaneously
with I–E curves in the Au/1 mM Co(II) + 1 M (NH ) SO (pH 9.3)
4 2 4
system at different cathodic limit potentials: (a) ꢀ925 mV, (b)
1
.
ꢀ
950 mV and (c) ꢀ975 mV. The scan potential was 5 mV sꢀ
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1769–1776 | 1773

View Article Online
3
.3.2 Charge of Co monolayer. The charge of the Co UPD
Table 1 Number of electrons exchanged (z) by the Co with the
substrate at different potentials
monolayer can be obtained by EQCM if the number of
electrons (z) exchanged in the process is known. This task is
accomplished using Faraday’s first law
E/mV
550
z
ꢀ
1.95
2.24
1.94
2.04
1.15
1.10
1.17
1.21
1.30
1.35
MM Dm
z
ꢀ650
¼
ꢃ F
Q
ð2Þ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
750
800
850
900
910
920
where MM is the molar mass of Co, z is the number of
electrons exchanged, MM/z is the molar mass equivalent, Q
is the charge density, Dm is the mass density and F is the
Faraday constant.
ꢀ930
4
9
ꢀ940
Some authors utilize an equation equivalent to eqn (2):
MM
z
f ðt2Þ ꢀ f ðt1Þ
¼
ꢀFCf ꢃ
ð3Þ
Additionally, the two electrons exchanged by the Co in the
process of monolayer formation would be confirmed if the
following route is considered:
jQðt2Þ ꢀ Qðt1Þj
It is important to mention that to obtain z in any case it is
necessary to determine MM/z, after which Dm and Q have to
be obtained separately. Consequently Q is determined by
chronoamperometry and Dm (or Df) is obtained using EQCM.
It is noteworthy that we determine z with the anodic response
of the step program since this portion is weakly influenced by
the double layer charge and is directly related to the UPD
process.
If the Co atom is considered to be adsorbed as an adatom,
which transfers two electrons to form a commensurate over-
layer onto the highly ordered gold electrode (111), the Df
calculated with the eqn (4), resulting from the expression (1)
and Faraday’s first law (eqn (2)), is 23.3 Hz. This value of Df is
in agreement with our experimental value of 23.8 Hz or
23.3 Hz (see Figs. 8–10) which shows that the Co atoms
exchanged totally their charge.
In Fig. 11 we show some curves DQ vs. Df (or Dm) used in
the determination of z and in Table 1, we summarise some
data on z produced using the expression (3).
ꢀ
ꢁ
Cf MM
zF
Df ¼ ꢀ
ꢃ DQ
ð4Þ
As can be seen, all values of the number of electrons
exchanged, determined (Table 1) at constant potential in the
UPD region, approach two. These results indicate that the
atoms of Co are adsorbed onto the gold substrate as dis-
On the other hand, by substituting the data z = 2 and
Df = 23.8 Hz in the expression (4), the experimental density
ꢀ2
5
0
of the Co UPD monolayer charge of 443.39 ꢂ 1.87 mC cm is
obtained. This charge corresponds effectively to a commensu-
rate Co overlayer on the gold substrate. Our result is
charged atoms. These values are in agreement with Schultze,
since this author established that if the substrate electronega-
tivity is more important than that of the adsorbate and their
difference is less than 0.5, then the adsorbate will be adsorbed
as discharged species.
2
3
supported by the results of Kleinert et al., who indicated
that the reconstruction of the substrate was lifted during
Co deposition, thereby implying that the cobalt monolayer
could copy the substrate structure. In addition, the 1 ꢃ 1
structure is supported by the fact that the same authors
mentioned it as a final step in the process of the monolayer
formation.
On the other hand, as the potential is more negative than
Bꢀ800 mV the values of z decrease from 2 to around 1.25 and
it is possible to say that its values will increase as a function of
cathodic potential. This behavior is probably due to the fact
+
that H reduction influences the density of charge (DQ) and
consequently this calculus, since this reaction is thermodyna-
3
.3.3 Adsorption isotherm. Fig. 12 shows the isotherm of
mically produced at potentials more negative than ꢀ780 mV
+
2
E(H /H )).
adsorption and desorption processes of the Co UPD process
on a gold substrate. These curves were prepared using the Df
response produced during the scan of the UPD region and the
(
a
Fig. 11 Some plots of anodic charge density (Q ) versus change of
frequency (Df) at different potential steps (see Table 1).
Fig. 12 Isotherm of the Co UPD process on a gold substrate.
1
774 | New J. Chem., 2007, 31, 1769–1776
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007

View Article Online
expression (4). In addition, the covering degree (y) was calcu-
lated using
Acknowledgements
This research was performed under the auspices of the Uni-
versidad Autonoma de San Luis Potosı through the Teaching
´
QE
Support Fund and the Ministry of Public Education through
the Program for Teachers’ Improvement (PROMEP).
y ¼
ð5Þ
Qml
References
where Qml is the density of charge of the monolayer deter-
mined by EQCM and Q is the density charge at E potential.
1 A. Aramata, in Modern Aspects of Electrochemistry, ed. J. O’M.
BocKris, R. E. White and B. E. Conway, Plenum Press, New York,
E
From +100 mV to negative potentials, the process of the
Co monolayer formation starts progressively at around
1
997, vol. 31, pp. 181–249.
2
D. M. Kolb, in Advances in Electrochemistry and Electrochemical
Engineering, ed. H. Gerischer and C. W. Tobias, Wiley-Inter-
science Publication, New York, 1978, vol. 11, pp. 125–271.
A. Hamelin, J. Electroanal. Chem., 1979, 101, 285.
ꢀ
400 mV and ends at –875 mV. These values indicate that
in a potential region of 400 mV, the monolayer is completed
and the behavior of the curve shows that Co atoms adsorb
selectively on the sites of the substrate surface (skink, step,
terraces, etc.). In other words, in the process of monolayer
formation on the substrate the first adsorption sites to be
occupied by the Co atoms are those of low energy (skink, step)
and subsequently, the adsorption sites of high energy (sites like
terraces).
3
4
S. Takahashi, K. Hasebe and A. Aramata, Electrochem. Commun.,
1
999, 1, 301.
5 H. Uchida, M. Hiei and M. Watanabe, J. Electroanal. Chem.,
998, 452, 97.
1
6
7
B. K. Niece and A. A. Gewirth, J. Phys. Chem. B, 1998, 102, 818.
A. E. Berkowitz, J. R. Mitchell, M. J. Carey, A. P. Young, S.
Zhang, F. E. Spada, F. T. Parker, A. Hutten and G. Thomas,
Phys. Rev. Lett., 1992, 68, 3745.
8
9
J. Q. Xiao, J. S. Jiang and C. L. Chien, Phys. Rev. Lett., 1992, 68,
3749.
C. L. Chien, J. Q. Xiao and J. S. Jianq, J. Appl. Phys., 1993, 73,
The analysis of the desorption process shows that this
occurs differently from the adsorption process. The desorption
process follows a linear behavior and it is produced slowly
with a different slope with regard to the opposite process. It is
important to mention that the covering degree at ꢀ400 mV is
almost 50% of the monolayer in the desorption process and
that the remaining 50% is desorbed in a potential region of
5309.
0 L. Piraux, J. M. George, J. F. Despres, C. Leroy, E. Ferain, R.
Legras, K. Ounadjela and A. Fert, Appl. Phys. Lett., 1994, 65,
1
2484.
1 J. Q. Wang, E. Price and G. Xiao, J. Appl. Phys., 1994, 75, 6903.
1
12 Y. J. Chen, W. Y. Cheung, I. H. Wilson, N. Ke, S. P. Wong, J. B.
Xu, H. Sang and G. Ni, Appl. Phys. Lett., 1998, 72, 2472.
3 Y. D. Zhang, J. I. Budnick, A. Hines, C. L. Chien and J. Q. Xiao,
Appl. Phys. Lett., 1998, 72, 2053.
14 A. Gerber, A. Milner, I. Y. Korenblit, M. Karpovsky, A. Gladkikh
and A. Sulpice, Phys. Rev. B: Condens. Matter Mater. Phys., 1998,
6
00 mV. This behavior indicates that the desorption process
1
and its opposite follow different mechanisms and that the
process of monolayer formation is, in essence, irreversible,
which is in agreement with the results found in voltammetric
experiments. This irreversible characteristic of the isotherms is
possibly due to the adsorption competence of H on the
substrate produced during the process of adsorption.
5
5 J. A. De Toro, J. P. Andres, J. A. Gonzalez, J. P. Goff, A. J.
7, 13667.
1
´
´
Barbero and J. M. Riveiro, Phys. Rev. B: Condens. Matter Mater.
Phys., 2004, 70, 224412.
1
1
6 S. Kenane and E. Chainet, Electrochem. Commun., 2002, 4, 167.
7 L. Cagnon, A. Gundel, T. Devolder, A. Morrone, C. Chappert, J.
E. Schmidt and P. Allongue, Appl. Surf. Sci., 2000, 164, 22.
1
1
8 L. H. Mendoza-Huizar, J. Robles and M. Palomar-Pardave
´
Electroanal. Chem., 2002, 521, 95.
9 L. H. Mendoza-Huizar, J. Robles and M. Palomar-Pardave
, J.
, J.
, J.
4
Conclusions
´
The Co UPD process on a gold substrate was studied by
coupling voltammetry and chronoamperometry with a quartz
microbalance in ammonium solution. The results obtained
show that the Co forms only one layer at the substrate before
the bulk deposition of cobalt occurs. During the UPD process,
the Co atom completely transfers its two electrons to the
electrode and is adsorbed as discharged species. The Co atoms
occupy the sites on the substrate in accordance with their
energy. In other words, the low-energy sites are occupied at
low covering degrees and the high-energy sites are occupied at
high covering degrees. The monolayer charge density, deter-
Electroanal. Chem., 2003, 545, 39.
20 L. H. Mendoza-Huizar, J. Robles and M. Palomar-Pardave
´
Electrochem. Soc., 2005, 152, C265.
1 I. Flis-Kabulska, J. Appl. Electrochem., 2006, 36, 131.
2
2
2 W. Schindler, D. Hofmann and J. Kirsner, J. Electrochem. Soc.,
2001, 148, C124.
23 M. Kleinert, H.-F. Waibel, G. E. Engelmann, H. Martin and D.
M. Kolb, Electrochim. Acta, 2001, 46, 3129.
4 S. Trasatti, Chim. Ind. (Milan, Italy), 1971, 53, 559.
5 D. A. Buttry and M. D. Ward, Chem. Rev., 1992, 92, 1355.
2
2
26 S. Langerock and L. Heerman, J. Electrochem. Soc., 2004, 151,
C155.
7 G. A. Snook, A. M. Bond and S. Fletcher, J. Electroanal. Chem.,
2
2
2
3
2
002, 526, 1.
ꢀ2
mined by electromicrogravimetry, is 448 mC cm , which
possibly corresponds to a 1 ꢃ 1 structure at the gold substrate.
In addition, the UPD formation process is controlled by
adsorption as shown by voltammetry. Finally, the analysis
of the process of monolayer destruction (desorption) shows
that this follows a different mechanism with regard to the
formation process possibly due to the adsorption of H onto
the substrate produced during the process of adsorption of the
metallic monolayer.
8 S. Hadzi-Jordanov, H. Argenstein-Kozlowska, M. Vukovic and B.
E. Conway, J. Electrochem. Soc., 1978, 125, 1471.
9 H. Angerstein-Kozlowska, B. E. Conway, B. Barnett and J.
Mozota, J. Electroanal. Chem., 1979, 100, 417.
0 A. Hamelin, J. Electroanal. Chem., 1996, 407, 1.
ˇ
31 S. S
trbac, R. R. Ad ꢀz ic
´
and A. Hamelin, J. Electroanal. Chem.,
1988, 249, 291.
2 M. C. Santos, L. H. Mascaro and S. A. S. Machado, Electrochim.
Acta, 1998, 43, 2263.
3 T. Hachiya, H. Honbo and K. Itaya, J. Electroanal. Chem., 1991,
315, 275.
3
3
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1769–1776 | 1775

View Article Online
3
3
3
3
3
3
4
4 E. Herrero, L. J. Buller and H. D. Abrun
˜
a, Chem. Rev., 2001, 101,
41 S. Fletcher, C. S. Halliday, D. Gates, M. Westcott, T. Lwin and G.
Nelson, J. Electroanal. Chem., 1983, 159, 267.
1897.
5 D. M. Kolb, M. Przasnyski and H. Gerischer, J. Electroanal.
Chem., 1974, 54, 25.
6 W. J. Lorenz, H. D. Hermann, N. Wuthrich and F. Hilbert, J.
Electrochem. Soc., 1974, 121, 1167.
7 G. L. Borges, K. K. Kanazawa, J. G. Gordon II, K. Ashley and J.
Richer, J. Electroanal. Chem., 1994, 364, 281.
8 M. Watanabe, H. Uchida, M. Miura and N. Ikeda, J. Electroanal.
Chem., 1995, 384, 191.
9 A. Montes-Rojas and E. Chainet, J. Mex. Chem. Soc., 2005, 49,
42 B. E. Conway and E. Gileadi, Trans. Faraday Soc., 1962, 2493, 68.
43 S. Srinivasan and E. Gileadi, Electrochim. Acta, 1966, 321, 11.
44 V. A. Vincent and S. Bruckenstein, Anal. Chem., 1973, 45, 2036.
45 E. Kirowa-Eisner, R. Gepshtein and E. Gileadi, J. Electroanal.
Chem., 2005, 583, 273.
46 T. Berzins and P. Delahay, J. Am. Chem. Soc., 1953, 75, 555.
47 A. Hamelin, J. Electroanal. Chem., 1984, 165, 167.
48 M. S. Zei, Y. Nakai, G. Lehmpfuhl and D. M. Kolb, J. Electro-
anal. Chem., 1983, 150, 201.
341.
49 G. G. Lang, M. Ujvari and G. Horanyi, J. Electroanal. Chem.,
2002, 522, 179.
50 J. W. Schultze and F. D. Koppitz, Electrochim. Acta, 1976, 21, 327.
0 A. B. Soto, E. M. Arce, M. Palomar-Pardave and I. Gonzalez,
Electrochim. Acta, 1996, 41, 2647.
1
776 | New J. Chem., 2007, 31, 1769–1776
This journal is ꢁc the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007