Effect of temperature on Co electrodeposition in the presence of boric acid

Article abstract of DOI:10.1016/j.electacta.2007.07.025

The electrodeposition of cobalt from sulphate solutions containing boric acid was investigated using EQCM technique coupled with potentiostatic measurements. The boric acid was added to electrolyte as a buffer to avoid the local pH rise caused by parallel hydrogen evolution reaction (HER). The results showed that the buffer contribution of boric acid is effective in the cobalt electrodeposition at 25 °C; however, cobalt hydroxide is formed simultaneously with cobalt deposition at 48 °C. The M/z values calculated using the Sauerbrey equation and the Faraday Law showed that in the initial stages of deposition at 48 °C, only cobalt deposits were detected, but after 2 s, an important amount of Co(OH)₂ started to be formed.

Full text of DOI:10.1016/j.electacta.2007.07.025

Electrochimica Acta 53 (2007) 644–649
Effect of temperature on Co electrodeposition
in the presence of boric acid
J.S. Santos, R. Matos, F. Trivinho-Strixino, E.C. Pereira∗
NANOFAEL, CMDMC, Departamento de Qu´ımica, Universidade Federal de Sa˜o Carlos, Cx 676, Sa˜o Carlos, SP, Brazil
Received 25 April 2007; received in revised form 11 July 2007; accepted 12 July 2007
Available online 19 July 2007
Abstract
The electrodeposition of cobalt from sulphate solutions containing boric acid was investigated using EQCM technique coupled with potentiostatic
measurements. The boric acid was added to electrolyte as a buffer to avoid the local pH rise caused by parallel hydrogen evolution reaction (HER).
The results showed that the buffer contribution of boric acid is effective in the cobalt electrodeposition at 25 ◦C; however, cobalt hydroxide is
formed simultaneously with cobalt deposition at 48 ◦C. The M/z values calculated using the Sauerbrey equation and the Faraday Law showed that
in the initial stages of deposition at 48 ◦C, only cobalt deposits were detected, but after 2 s, an important amount of Co(OH)2 started to be formed.
© 2007 Elsevier Ltd. All rights reserved.
Keywords: Cobalt; Cobalt hydroxide; Electrodeposition; Morphology; EQCM
1. Introduction
strongly affect the morphological and structural properties and
decrease the overall process efficiency [11–14]. According to
some authors [11,13], the pH near the electrode surface could
increase due to the water electrolysis leading to hydroxide ions
formation and hydrogen evolution reaction, as follows:
While the recent advances in the nanomaterial science have
opened up new possibilities for the fabrication of new devices,
they have also raised new issues. As an example, a number of
these involve the problems that arise in dealing with the elec-
trodeposition of metals with magnetic properties over substrates
with specific pattern for the storage devices industry [1–3]. Fur-
thermore, the structure and morphology control is an essential
task in the electrodeposition of metals with magnetic property.
This property, as well others such as hardness, thermal stability
and corrosion resistance of cobalt electrodeposits has moti-
vated further investigations on cobalt electrodeposition [4–6].
Depending on the preparation conditions, i.e. electrolyte com-
position, pH of the solution, temperature, current density and
the presence of additives, different morphological and structure
properties can be obtained [6–10]. As these problems are not
addressed at all in this field, the study of the cobalt electrodepo-
sition mechanism is of particular interest.
H2O + M–H + e− → M + H2(ads) + OH−
(a)
From reaction (a), one can observe that the adsorption mech-
anism is an important task to be considering in electrodeposits
carried out simultaneously with HER [8,11–14]. Hence, the
hydrogen electrodic desorption rate (Ved) must be considered for
the investigation of the overall mechanism during electrodeposi-
tion process and their parameters dependence (i.e., temperature,
the surface fraction covered with adsorbed hydrogen (θ) and the
pH in the double layer region). In this subject, some authors
[9,15] suggested that the temperature may affect the HER rais-
ing its rate. Thus, for high temperatures, the electrodeposition
process efficiency could drop down to 60% and the hydrogen
adsorption process would become less favorable modifying the
hydrogen electrodic desorption rate (Ved).
During the metal electrodeposition in aqueous solution par-
allel processes can occur. An important secondary process is
the parallel hydrogen evolution reaction (HER), which may
On the other hand, to avoid the pH variation near the
electrode, it is also common to add boric acid (H3BO3) in
the electrolyte to inhibit the formation of hydroxide species
[10,16–20]. Unfortunately, the real function of this additive in
electrodeposition is yet a matter of controversy and a lot of
∗
Corresponding author. Tel.: +55 16 33518214/213; fax: +55 16 33518214.
E-mail address: decp@power.ufscar.br (E.C. Pereira).
0013-4686/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2007.07.025

J.S. Santos et al. / Electrochimica Acta 53 (2007) 644–649
645
propositions were done. Most authors proposed that the boric
acid acts mainly as a buffer agent [16–18]. This effect is thought
to be a dynamic process occurring during the HER. Conse-
quently, the dissociation kinetics of the buffer must be fast
enough to supply protons during molecules discharge occurring
at the electrode surface. Studying this be−havior, Horkans [19]
concluded that the dissociation of HSO4 as well as H3BO3
is so slow that the overall buffer contribution is ineffective
during HER in sulphate electrolyte. However, this author also
proposed that this additive may act as a homogeneous catalyst
which decreases the metal electrodeposition overpotential [19].
In addition, Yin and Lin [20] proposed that H3BO3 works as
a selective membrane inhibiting movement of the ions during
diffusion controlled process at the electrode.
All potentiodynamic and potentiostatic reactions were car-
ried out at two temperatures (25 ◦C and 48 ◦C) using 0.05 mol
L−1 CoSO4·5H2O + 0.01 mol L−1 H3BO3 + 0.113 mol L−1
Na2SO4 as supporting electrolyte (pH 5.0 at T = 25 ◦C). The
electrochemical cell was maintained at a constant temperature
using a thermostatic bath. The working electrode was cleaned
using fast sweeping rate technique under 0.1 mol L−1 H2SO4
solution between −0.25 V and 1.20 V (versus SCE) until a
satisfactory response of the Pt profile was obtained. All solu-
tions were prepared with deionized water and analytical grade
reagents. Before each experiment, the supporting electrolyte
solution was bubbled with N2 flux during 25 min.
3. EQCM data processing
Considering how these parameters (temperature, the surface
fraction covered with adsorbed hydrogen (θ) and the pH in the
double layer region) could affect the electrodeposition process,
much effort has been given to elucidate the mechanism of cobalt
electrodeposition. The main idea lies on the assumption that the
pH in the vicinity of the cathode surface increases due to hydro-
gen evolution and cobalt hydroxide can be formed. From this
point of view, numerous authors proposed different and/or com-
plementary mechanism for cobalt electrodeposition. According
to Jiang and Tseung [21], Co deposition occurs through the
reduction of a cobalt complex, such as CoOH+, and the limiting
step was the reduction of CoOH to metallic Co. In another paper,
Pradhan et al. [14] also suggested that the cobalt electrodeposi-
tion mechanism occurs via formation of CoOH+ at more acidic
pH values. The propositions of these authors [14] consider that
there is a surface fraction covered by adsorbed hydrogen and the
reaction proceeds through CoOH+ and Hads reduction. However,
rising the pH (at pH 4 and 4.5) the mechanism involves Co(OH)2
formation. On the other hand, we described elsewhere [22] that
cobalt hydroxide is formed simultaneously with Co deposition
during the early stages of reduction at pH 4.10 under borate free
electrolytes. However, atpH3.33onlythedirectcobaltreduction
is observed without any hydroxide species formation.
The EQCM technique combined with cyclic voltammetry is
a convenient tool to investigate electrochemical reactions by
measuring the current, charge and related mass changes at the
working electrode [23]. According to the Sauerbrey equation
[24] the frequency variation (ꢀf) of the quartz crystal is corre-
lated with the mass change on the electrode (ꢀm), as follows:
−2f02ꢀm
ꢀf =
= −Kꢀm
(1)
√
A μiρi
where f0 is the resonant frequency of the quartz crystal, A the
piezoelectric active area, μi the shear modulus of the quartz, K
the experimental mass coefficient and ρi is the density of quartz.
When Eq. (1) is combined with the Faraday Law, and the num-
ber of electrons involved in the reaction is known, the apparent
molar mass can be calculated [24–26]. Hence, M/z values can be
determined by fitting the slope of the mass versus charge curves:
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢁ
ꢂ
M
z
dꢀf
dQ
F
=
(2)
K
where M is the molar mass of the deposit, F the Faraday constant
and z is the number of electrons. The M/z values may also be
used to evaluate the number of reactions occurring during an
electrochemical process.
Although significant advances have been made in the under-
standing of how additives and temperature influence the cobalt
electrodeposition, very little attention appears to have been given
totheexactroleoftheseparametersontheelectrodepositionpro-
cess. Hence, the objective of the present study is to investigate
the cobalt electrodeposition using the electrochemical quartz
crystal microbalance (EQCM) in the presence of H3BO3 at two
different temperatures (25 ◦C and 48 ◦C).
In the present paper, the M/z values were determined from
potentiodynamic and potentiostatic measurements in CoSO4
solution containing boric acid as a buffer at two temperatures
(25 ◦C and 48 ◦C).
4. Results
2. Materials and methods
A preliminary investigation to evaluate the main processes
duringthecobaltelectrodepositionwasperformedusingtriangu-
lar voltammetry. Fig. 1 shows the voltammograms for the cobalt
electrodeposition performed at two different temperatures and
the corresponding mass change observed during the process. In
the scan toward negative potentials there is no process until the
beginning of cobalt deposition at −0.76 V (25 ◦C). For poten-
tials more negative than this, an increase in the current and a
mass change on the working electrode were observed (Fig. 1C).
The cobalt reduction current peak is observed at −1.04 V (25 ◦C)
followed by a significant increase in the cathodic current which
The experiments were carried out in a glass cell. The work-
ing electrode (WE) was a 9 MHz AT-cut quartz crystal coated
with a Pt film in contact with the electrolyte (A = 0.2 cm2). The
sensitivity of the EQCM used was 858.8 Hz ␮g−1. The counter
electrode(CE)wasaPtsheetandallpotentialsarereferredtosat-
urated calomel electrode (SCE). The resonance frequency shift
was measured with a Seiko EG&G quartz crystal microbalance
(model QCA 917). The electrochemical measurements were
conducted using an EG&G PAR 263 A potentiostat/galvanostat.

646
J.S. Santos et al. / Electrochimica Acta 53 (2007) 644–649
Table 1
Apparent M/z values calculated from Fig. 1A considering the total mass (ꢀm)
and charge (ꢀQ) during cobalt electrodeposition toward negative potentials
Temperature (◦C)
25
Potential range (V)
M/z (g mol−1
)
−0.80 → −1.20
31 0.5
−0.86 → −1.01
−1.01 → −1.17
−1.17 → −1.20
38 0.5
32 0.5
28 0.5
48
and changed the electrolyte temperature. In order to evaluate
the species participating during the cobalt electrodeposition, the
apparent M/z values (Tables 1 and 2) were calculated using the
Faraday Law. Considering an ideal cobalt reduction involving
2 mol of electrons, the apparent M/z value must be 29.5 g mol−1
(MW Co = 58.9 g mol−1). Following:
Co2+(aq) + 2e− → Co(s)
(b)
The Co(OH)2 formation requires an apparent M/z value of
46.5 g mol−1 (MW Co(OH)2/2e−) following the global reaction
from (c) and (d):
2H2O + 2e− → H2(g) + 2OH−(aq)
(c)
(d)
Co2+(aq) + 2OH−(aq) → Co(OH)2(ads)
Table 1 illustrates the apparent M/z value for potentiodynamic
cobalt electrodeposition toward more negative potentials (from
−0.8 V to −1.2 V), calculated using the slope of mass × charge
curves and the Faraday Law (Fig. 2A). It was observed that
for the experiment performed at 25 ◦C, there is only one appar-
ent M/z in the whole potential window (31 0.5 g mol−1). This
result shows that the electrodeposition follows the reaction
(b), i.e. direct cobalt reduction. Different apparent M/z values
were calculated for the experiment performed at 48 ◦C, as it
can be observed in the slope variation during the potentiody-
namic deposition depicted in Fig. 2B. Just at the beginning of
cobalt reduction (from −0.86 V to −1.01 V) the M/z value was
38 0.5 g mol−1, which has an important component related to
the global reaction originated from (c) and (d), where Co(OH)2
is formed. In the subsequent process, the calculated M/z val-
ues were 32 0.5 g mol−1 (from −1.01 V to −1.17 V) and
28 0.5 g mol−1 (from −1.17 V to −1.20 V) indicating only
cobalt reduction. The Co(OH)2 could be reduced to Co at poten-
tial more negative than −0.97 V (Standard Potential [29,30]),
following the reaction below:
Fig. 1. (A) Cyclic voltammograms, (B) detailed range of cobalt deposition
overpotential and (C) mass change as a function of potential in solution con-
taining 0.05 M CoSO4 + 0.01 M H3BO3 + 0.113 M Na2SO4 at pH 5.0 for cobalt
electrodeposition at 25 ◦C and 48 ◦C. Scan rate = 20 mV/s.
can be associated with proton reduction. In the reverse sweep
toward positive potentials, two crossing over potentials were
observed. Furthermore, from Fig. 1A, one can observe an intense
anodic peak with a shoulder during the cobalt dissolution which
was previously reported by other authors [12,15,22,27,28]. This
shoulder could be related to the dissolution of hydrogen rich
cobalt phase [28]. The possibility of passivation should not
be disregarded as cobalt(III) species could be formed, such
as CoOOH, originated from the oxidation of Co(OH)ads or
Co(OH)2 previously adsorbed on the electrode. However, at
any rate, all deposited material was dissolved during the anodic
stripping (Fig. 1C).
Although the general profile of these two voltammograms
(Fig. 1) has the same behavior for both temperatures, some
important issues can be pointed out: (i) the overpotential of the
cobaltdepositionintheexperimentcarriedoutat48 ◦Cissmaller
than the one at 25 ◦C (inset Fig. 1B), (ii) the mass–potential
curves showed an increase in the deposited mass over the work-
ing electrode for the measurement performed at 48 ◦C, (iii) the
total deposited mass on the electrode is dissolved (Fig. 1C) and
(iv) the dissolution current peak related to the rich hydrogen
phase is more evident in the electrodeposition carried out at
48 ◦C.
Co(OH)2(ads) + 2e− → Co(s) + 2OH−(aq)
(e)
Table 2
Apparent M/z values calculated considering the total mass (ꢀm) and charge
(ꢀQ) from current–time transient experiment
Temperature (◦C)
M/z (g mol−1
)
Time: 0–2 s
Time: 2–10 s
25
48
25a
27 0.5
30 0.5
30 0.5
30 0.5
44 0.5
31 0.5
In our previous work [22], we showed that during the initial
stages of cobalt electrodeposition, small quantities of Co(OH)2
are produced. In the present work, we introduced the boric acid
a
After cooling procedure.

J.S. Santos et al. / Electrochimica Acta 53 (2007) 644–649
647
Fig. 2. Mass variation as a function of charge for (A and B) potentiodynamic deposition and (C and D) potentiostatic deposition at −095 V.
Hence, the Co(OH)2 reduction requires an apparent M/z value
faradaic reaction occurs on the electrode) and −0.95 V for 10 s.
of 23.2 g mol−1 (MCo(OH) /4e−) following reactions (c)–(e). If
The chosen potential is located near the water electrolysis poten-
tial, where the HER occurs, and remains more positive than
−0.97 V, which is the Standard Potential for Co(OH)2 reduction
[29,30]. It is important to stress that in these experimental condi-
tion the major fraction of produced Co(OH)2 will not be reduced
to Co. After the potentiostatic deposition, anodic stripping was
performed to investigate the dissolution process and to evaluate
the current efficiency.
2
this is correct, one needs to consider 4 mol of electrons during all
process. An apparent M/z value smaller than 29.5 g mol−1 shows
that a mixed contribution from reactions (b) and (e) may also
occur for the deposition performed at potentials more negative
than −0.97 V (Standard Potential).
To study the electrodeposition process in a more detailed way,
transient measurements coupled with EQCM were performed
and the results are presented in Fig. 3 and Table 2. In these exper-
iments, a step potential was applied between −0.1 V (where no
Table 2 illustrates the M/z values calculated by fitting the
slope of mass versus charge (Fig. 2C and D) for cobalt depo-
sition at −0.95 V. During the cobalt deposition carried out at
25 ◦C, the calculated M/z value was 27 0.5 g mol−1 (up to 2 s)
and 30 0.5 in the subsequent deposition (Fig. 2C). These val-
ues are very close to the theoretical value for cobalt reduction,
indicating that the major reaction occurring on the electrode is
described by reaction (b). In this case, it is not expected that the
Co(OH)2 reduction takes place since it was not produced at this
experimental condition.
On the other hand, for the electrodeposition performed
at 48 ◦C, at the first 2 s, the calculated M/z value was
30 0.5 g mol−1 (Fig. 2D). Again, thissuggeststhatintheinitial
stages of cobalt reduction the process proceeds through reac-
tion (b). Within the interval between 2 s and 10 s, the calculated
M/z value was 44 0.5 g mol−1. This value is in accordance
with the global reaction originated from (c) and (d), which are
related to Co(OH)2 formation. The same result can be better
visualized in Fig. 3, where the calculated apparent M/z transient
Fig. 3. Apparent M/z transients as a function of time.

648
J.S. Santos et al. / Electrochimica Acta 53 (2007) 644–649
deposition and anodic stripping process for 25 ◦C and 48 ◦C,
respectively. Fig. 4 depicts the theoretical mass variation for Co
and Co(OH)2 calculated considering the Faraday Law for reac-
tions (b) and (d), respectively. For the experiment performed
at 25 ◦C, the experimental mass change and the Co theoret-
ical one overlapped during the potentiostatic deposition and
the anodic stripping, which indicates that the electrodeposition
occurs only via reaction (b). Hence, the current efficiency dur-
ing this process is almost 100%. On the other hand, from the
experiment performed at 48 ◦C (Fig. 5), one can observe that
the experimental and Co theoretical curves overlapped up to 2 s,
suggesting an electrodeposition mechanism via reaction (b). For
times longer than this, the experimental curve drifts apart from
the Co theoretical one and approaches the Co(OH)2, indicat-
ing an electrodeposition process occurring via reaction (d), i.e.
Co(OH)2 formation. This behavior can also be observed during
theanodicstrippingandcanberelatedtoadecreaseinthecurrent
efficiency. Thesedataalsoconfirmedtheapparentcalculated M/z
values(Table2)foundbyelectrodeposition(−0.95 V)anddisso-
lution at 48 ◦C suggesting that the electrodeposition mechanism
is followed by a mixed contribution of different intermediary
reactions, i.e. following reactions (b)–(d). After that, from the
difference between the curves presented in Fig. 5, it was esti-
mated that, at 10 s, almost 80% of the deposited material could
be Co(OH)2.
Fig. 4. Comparison of experimental and theoretical mass change of cobalt depo-
sition and anodic stripping obtained at −0.95 V from current–time transients.
T = 25 ◦C. Scan rate: 20 mV/s.
showed a continuous variation of the M/z values as a function of
time. For the electrodeposition performed at 25 ◦C, the apparent
M/z transient remains constant during all experiment indicating
an electrodeposition mechanism where only cobalt is reduced.
In addition, for the experiment carried out at 48 ◦C, one can
observe a smooth increase of the apparent M/z transient, indicat-
ing a change in the cobalt electrodeposition mechanism starting
with cobalt deposition followed by Co(OH)2 formation occur-
ring simultaneously with cobalt deposition up to 10 s. Although
it was not expected, the sharp decrease of the M/z value at the
early stage of the deposition process (0–0.5 s) could be related to
reactions (c)–(e), which are related to the Co(OH)2 reduction.
However, the large experimental error due to the slow acqui-
sition rate of the EQCM data may also be responsible for this
sharp decrease in the beginning of the reduction process.
5. Discussion
The apparent M/z values related to hydroxide species were
not expected in our experiments as H3BO3 was used to prevent
any important pH variation near electrode surface. However, the
real function of the boric acid during electrodeposition process
is yet a matter of controversy, as initially stated in the intro-
ductory section. Despite all mechanism propositions, the data
showed that the boric acid was ineffective for the experiment
performed at 48 ◦C. Several propositions can be pointed out in
order to explain our experimental data. First of all, an impor-
tant experiment was performed to ensure that the boric acid did
not lose its effectiveness during the experiments at high tem-
peratures. For this purpose, subsequent electrodeposition was
performed immediately after the experiment at 48 ◦C, cooling
the solution until 25 ◦C. In this last measurement, at 25 ◦C, the
cobalt was deposited at −0.95 V for 10 s and the apparent M/z
value was calculated. Table 2 shows that the apparent M/z val-
ues after the cooling procedure was close to 30 g mol−1 (cobalt
reduction via reaction (b)). These data revealed that the boric
acid remains effective after the experiment performed at 48 ◦C,
in which the additive was ineffective. Hence, any proposition
about the deactivation of H3BO3 could be disregarded.
Figs. 4 and 5 illustrate the theoretical and experimental
mass change over the working electrode during potentiostatic
Jeffrey et al. [9] and Elsherief [15] proposed in their work
that the HER contribution is intense in experiments carried out
at high temperatures. If we consider that the hydrogen evolution
reaction is more effective at 48 ◦C, the formation rate of OH−
ions must be larger than in the case of the experiment carried
out at 25 ◦C. This explains the observed M/z values for hydrox-
ilate species (Tables 1 and 2). Another effect of the temperature
that is observed is a drift toward more positive overpotential for
Fig. 5. Comparison of experimental and theoretical mass change of cobalt depo-
sition and anodic stripping obtained at −0.95 V from current–time transients.
T = 48 ◦C. Scan rate: 20 mV/s.

J.S. Santos et al. / Electrochimica Acta 53 (2007) 644–649
649
the cobalt and proton reduction (Fig. 1B). However, the most
suitable explanation for the inefficiency of boric acid may arise
from the propositions described in the literature [10,17–19].
One possible mechanism involving complex ions formation
can also be evoked, as the one proposed by Hoare [17]. This
author proposed that H3BO3 can form a stable complex with
the Ni ions making easier the discharge of this metal over the
electrode. This author also suggests that the boric acid behaves
like a homogeneous catalyst decreasing the overpotential for Ni
deposition. Unfortunately, the mechanism for this proposition
cannot be determined from our experimental results and more
effort must be taken in order to consider this mechanism.
However, within the scope of our experimental result, we
proposed that there is a mixed contribution of the slow disso-
ciation kinetics of the boric acid and a mechanism involving
adsorption of neutral boric acid molecules over the electrode, as
proposed by Horkans [19]. This author proposed that in Ni depo-
sition under SO42− solution, little competition between H3BO3
and sulphates ions exists, and consequently, boric acid adsorp-
tion may decrease the active surface area. A local increase of
the temperature in the electrode surface may strongly affect the
adsorption rate of this species and change the cobalt deposition
efficiency. As a result, the proton reduction rate will increase
favoring the formation of hydroxilate species due to OH− for-
mation. On the other hand, if we consider the kinetics of the
boric acid dissociation [19], it can be proposed that this disso-
ciation is the rate-determining step during HER. In this case,
the dissociation rate is not fast enough to provide the increase
of concentration of OH− at the surface of the electrode for
the experiments performed at 48 ◦C. Although both hypotheses
corroborate our experimental results, we propose that the con-
tribution of the first proposition is more important to explain our
experimental results than the one about the dissociation kinetics
of boric acid, since it was observed that at 48 ◦C, the hydrogen
evolution is intense. Besides, in the present case, the concentra-
tion of boric acid is lower (compared to the cobalt sulphate) than
those investigations described in the literature for Ni deposition
[17,19,31].
most suitable alternative is the one concerning the effect of tem-
perature rise on the adsorption mechanism of H3BO3 over the
electrode. For high temperatures, the desorption mechanism is
greater, leading to an increase of the active surface area available
for HER and, as a result, Co(OH)2 can be formed.
Acknowledgments
The authors are grateful to FAPESP and CNPq for the finan-
cial support.
References
[1] S. Armyanov, Electrochim. Acta 45 (2000) 3323.
[2] P.L. Cavallotti, A. Vicenzo, M. Bestetti, S. Franz, Surf. Coat. Technol.
169–170 (2003) 76.
[3] M.V. Rastei, S. Colis, R. Meckenstock, O. Ersen, J.P. Bucher, Surf. Sci.
600 (2006) 2178.
[4] E. Go´mez, E. Valle´s, J. Appl. Electrochem. 32 (2002) 693.
[5] J. Dille, J. Charlier, R. Winand, J. Mater. Sci. 32 (1997) 2637.
[6] A. Azizi, A. Sahari, M.L. Felloussia, G. Schmerber, C. Miny, A. Dinia,
Appl. Surf. Sci. 228 (2004) 320.
[7] C.Q. Cui, S.P. Jiang, C.C. Tseung, J. Electrochem. Soc. 137 (1990) 3418.
[8] S. Nakahara, S. Mahajan, J. Electrochem. Soc. 127 (1980) 283.
[9] M.I. Jeffrey, W.L. Choo, P.L. Breuer, Miner. Eng. 13 (2000) 1231.
[10] B.C. Tripathy, P. Singh, D.M. Muir, Metall. Trans. B 32 (2001) 395.
[11] D.R. Gabe, J. Appl. Electrochem. 27 (1997) 908.
[12] M. Palomar-Pardave´, B.R. Scharifker, E.M. Arce, M. Romero-Romo, Elec-
trochim. Acta 50 (2005) 4736.
[13] M. Rojas, C.L. Fan, H.J. Miao, D.L. Piron, J. Appl. Electrochem. 22 (1992)
1135.
[14] N. Pradhan, T. Subbaiah, S.C. Das, U.N. Dash, J. Appl. Electrochem. 27
(1997) 713.
[15] A.E. Elsherief, J. Appl. Electrochem. 33 (2003) 43.
[16] N. Zech, D. Landolt, Electrochim. Acta 45 (2000) 3461.
[17] J.P. Hoare, J. Electrochem. Soc. 133 (1986) 2491.
[18] B.V. Tilak, A.S. Gendron, M.A. Mosoiu, J. Appl. Electrochem. 7 (1977)
495.
[19] J. Horkans, J. Electrochem. Soc. 126 (1979) 1861.
[20] K.M. Yin, B.T. Lin, Surf. Coat. Technol. 78 (1996) 205.
[21] S.P. Jiang, A.C.C. Tseung, J. Electrochem. Soc. 137 (1990) 3387.
[22] J.T. Matsushima, F. Trivinho-Strixino, E.C. Pereira, Electrochim. Acta 51
(2006) 1960.
˜
[23] D.A. Buttry, in: H.D. ABRUNA (Ed.), The Quartz Crystal Microbalance
as an In Situ Tool in Electrochemistry, 1, VHC, New York, 1991, p. 10.
[24] H.A.K.M. Saloniemi, M. Ritala, M. Leskela, J. Electroanal. Chem. 482
(2000) 139.
6. Conclusions
[25] J.G.N. Matias, J.F. Julia˜o, D.M. Soares, A. Gorenstein, J. Electroanal.
Chem. 431 (1997) 163.
[26] A. Marlot, J. Vedel, J. Electrochem. Soc. 1 (1999) 177.
[27] A.B. Soto, E.M. Arce, M. Palomar-Pardavi, I. Gonzalez, Electrochim. Acta
41 (1996) 2647.
[28] S. Jaya, T. Prasada Rao, G. Prabhakara Rao, Electrochim. Acta 32 (1987)
1073.
[29] S.P. Jiang, Y.Z. Chen, J.K. You, T.X. Chen, A.C.C. Tseung, J. Electrochem.
Soc. 137 (1990) 3374.
The analysis of the EQCM data and the current–time transient
measurements indicated that the mechanism of cobalt elec-
trodeposition changed when the temperature was increased in
sulphate solution containing boric acid. At 25 ◦C, only direct
cobalt reduction is observed, while in electrodeposition at 48 ◦C,
Co(OH)2 can be observed from the calculated apparent M/z
values. These results suggest that Co(OH)2 can be formed simul-
taneously with Co deposits and the buffer contribution of boric
acid was ineffective at 48 ◦C. Furthermore, some propositions
were considered to explain the inefficiency of the boric acid. The
[30] D.R. Lide, CRC Handbook of Chemistry and Physics, CRC Press, Florida,
1992, pp. 8–26.
[31] J. Horkans, J. Electrochem. Soc. 128 (1981) 45.