Electrodeposition of zinc-cobalt alloy from a complexing alkaline glycinate bath

Article abstract of DOI:10.1016/j.electacta.2007.01.010

The influence of cobalt on the electrodeposition of zinc onto AISI 1018 steel was studied in weakly alkaline glycine solutions. Thermodynamic calculations were performed to construct predominance-zone diagrams to identify the stability of the zinc and cobalt glycine complexes, and experimental studies of electrochemical behavior and deposit properties were conducted. When zinc is present, cobalt deposition shifts to more negative potentials, producing ZnCo alloys. Two main reduction steps were observed for electrodeposition from the ZnCo bath: the first at low potentials was due to ZnCo electrodeposition. In the second, at more negative potentials, cobalt content in the deposit increased forming a range of intermediate phases, and the hydrogen-evolution reaction became significant. The presence of Co(II) in the bath modified the morphology of the deposits as well as reducing the faradaic metal-deposition efficiency. ZnCo-deposit morphology was modified by the applied current density as well as the metal composition of the coating. X-ray diffraction studies revealed that cobalt oxide or hydroxide is formed during ZnCo electrodeposition, indicating that an elevation of the interfacial pH plays a role in the alloy deposition process.

Full text of DOI:10.1016/j.electacta.2007.01.010

Electrochimica Acta 52 (2007) 4742–4751
Electrodeposition of zinc–cobalt alloy from a complexing
alkaline glycinate bath
J.L. Ortiz-Aparicioa, Y. Measa,∗, G. Trejoa, R. Ortegaa, T.W. Chapmana,
E. Chainetb, P. Ozilb
a Centro de Investigacio´n y Desarrollo Tecnolo´gico en Electroqu´ımica, S.C., Parque Tecnolo´gico Quere´taro,
Sanfandila, Pedro Escobedo, Quere´taro, C.P. 76703, Mexico
b Laboratoire d’Electrochimie et Physico-chimie des Mate´riaux et Interfaces (UMR 5631 CNRS-INPG-UJF), Ecole Nationale
Supe´rieure d’Electrochimie et d’Electrometallurgie de Grenoble BP75, 38402 Saint Mart´ın d’He`res, France
Received 22 August 2006; received in revised form 10 January 2007; accepted 11 January 2007
Available online 26 January 2007
Abstract
The influence of cobalt on the electrodeposition of zinc onto AISI 1018 steel was studied in weakly alkaline glycine solutions. Thermodynamic
calculations were performed to construct predominance-zone diagrams to identify the stability of the zinc and cobalt glycine complexes, and
experimental studies of electrochemical behavior and deposit properties were conducted. When zinc is present, cobalt deposition shifts to more
negative potentials, producing ZnCo alloys. Two main reduction steps were observed for electrodeposition from the ZnCo bath: the first at low
potentials was due to ZnCo electrodeposition. In the second, at more negative potentials, cobalt content in the deposit increased forming a range of
intermediate phases, and the hydrogen-evolution reaction became significant. The presence of Co(II) in the bath modified the morphology of the
deposits as well as reducing the faradaic metal-deposition efficiency. ZnCo-deposit morphology was modified by the applied current density as well
as the metal composition of the coating. X-ray diffraction studies revealed that cobalt oxide or hydroxide is formed during ZnCo electrodeposition,
indicating that an elevation of the interfacial pH plays a role in the alloy deposition process.
© 2007 Elsevier Ltd. All rights reserved.
Keywords: Zinc–cobalt alloys; Electrodeposition; Glycine; Anomalous deposition
1. Introduction
anism for Zn–Co codeposition, involving the formation of a
zinc hydroxide layer on the electrode surface, which suppresses
cobalt reduction. Higashi et al. [7] carried out a study of the elec-
trodeposition of Zn in the presence of Co; the variation of the
interfacial pH was determined, and the transition from normal to
anomalous codeposition was associated with an increment of the
interfacial pH. Stankeviciute et al. [8] applied electrochemical
quartz-crystal microbalance measurements to analyze the depo-
sition of Zn–Co alloys in terms of the mechanism of hydroxide
suppression. On the other hand, other authors have argued that
during the anomalous codeposition of Ni–Fe, Ni–Zn and Fe–Zn
alloys, only soluble intermediate species are formed [9–14].
The concept of hydroxide oscillation [15] has also been used
to explain anomalous deposition. According to this model, the
thickness of the hydroxide layer changes periodically. When
the zinc hydroxide layer is depleted, the reduction of both H+
and cobalt occurs preferentially to that of Zn. Then, as the
interfacial pH increases due to the hydrogen-evolution reaction,
Zinc coatings are used widely to protect iron and steel sub-
strates against corrosion [1]. Metals like iron, cobalt and nickel
have been incorporated in zinc plating baths to obtain coatings
with higher corrosion resistance [1,2]. During zinc codeposition
with metals of the iron group, the less noble metal, zinc, is elec-
trodeposited preferentially; this phenomenon has been described
as anomalous codeposition by Brenner [3].
According to Dahms and Croll [4], in the anomalous code-
position of Fe–Ni alloys a layer of iron hydroxide is formed,
which adsorbs on the electrode and suppresses the nickel reduc-
tion. The metal hydroxide layer is formed as a consequence
of elevated interfacial pH caused by the hydrogen-evolution
reaction. Decroly and co-workers [5,6] proposed a similar mech-
∗
Corresponding author. Tel.: +52 442 211 6070.
E-mail address: yunnymeas@cideteq.mx (Y. Meas).
0013-4686/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2007.01.010

J.L. Ortiz-Aparicio et al. / Electrochimica Acta 52 (2007) 4742–4751
4743
the hydroxide layer forms again favoring Zn reduction, and the
codeposit becomes anomalous.
ating parameters on the electrodeposit properties, such as alloy
composition, morphology, and current efficiency. The ultimate
objective is to develop an efficient and environmentally benign
zinc–cobalt plating bath.
Inhibition of nickel reduction due to zinc underpotential
deposition (upd) has been proposed by Nicol and Philip [16].
Ohtsuka and Komory found evidence for the formation of a zinc
monolayer deposit that inhibits the nucleation and growth of
nickel [17]. On this basis, Swathirajan has claimed that Ni–Zn
anomalous deposition depends on the electrode potential and
that the inhibition of Ni deposition is due to the underpotential
deposition of Zn [18].
Differences in the values of exchange current densities have
also been considered in the interpretation of anomalous codepo-
sition [19,20]. The magnitude of the exchange current density
is generally much greater for zinc compared to metals like iron,
cobalt and nickel.
2. Experimental
Electrodeposits were formed from aqueous 0.2 M ZnSO4
(Merck) with CoSO4 (Merck) at various concentrations from
0.005 to 0.1 M. All tests were performed in a complexing 2.2 M
glycine (Aldrich) medium. The solution was made up first by
zinc sulfate dissolution in a previously prepared glycine solu-
tion, cobalt sulfate being added later. The electrolyte pH was
adjusted to 11 by adding sodium hydroxide. All the solutions
were prepared with deionized water (18 Mꢀ cm). The Zn and
Zn–Co coatings were electrodeposited on an AISI 1018 steel
substrate. Stripping voltammograms were recorded using a slow
sweep rate (v = 1 mV s−1) in 2.2 M glycine and pH 11 solution
under a N2 atmosphere.
The experiments were performed with a PAR 273A potentio-
stat. Cyclic and linear voltammetry were carried out in a typical
three-electrode cell. AISI 1018 steel embedded in a PTFE rod
was used as a disk working electrode with a geometrical area of
0.032 cm2, a saturated calomel electrode was the reference elec-
trode, and the counter electrode was a graphite bar. The trials
were carried out under an ultrapure nitrogen atmosphere. Before
each experiment, the steel electrode was polished with 0.05 ␮m
alumina (Buehler) to a mirror surface appearance. The temper-
ature was maintained at 25 ◦C. Deposit specimens for micro-
scopic characterization were obtained on the steel electrodes by
applying a constant current density without stirring. The mor-
phology and composition measurements were done with a scan-
ning electron microscope (JEOL DSM-5400 LV). The X-ray
diffraction patterns were obtained with Cu K␣ radiation in an X-
ray diffractometer (Brucker D8 Advance). The coating composi-
tion profile (GDS analysis) was determined by using a Glow Dis-
charge Spectrometer (Leco 850A). The current efficiency was
calculated by integrating the charge passed under the stripping
curvesandcomparing that valuewiththetotalchargepasseddur-
ing the deposition process; these were performed in duplicate.
The experimental error of the GDS experiments was estimated
to be 1% for zinc, 4% for cobalt, and 11% for carbon; these mag-
nitudes reflect the level of an element in the deposit. The primary
goal of the elemental analysis and of the current-efficiency mea-
surements was to show generally the qualitative effects of the
plating parameters during ZnCo alloy electrodeposition.
Tsuru et al. [21] showed that the nature of the solvent has an
effect on Zn–Co codeposition. Normal codeposition occurs in
pure methanol, whereas in the presence of water they obtained
anomalous deposits. Similar results were obtained in ionic liq-
uids for Zn–Fe deposition [22]. Recently, Fratesi and co-workers
[23] studied the electrodeposition of ZnCo alloy from an acidic
chloride bath. They investigated the effects of Zn2+/Co2+ ratio
and the applied potential on the formation of ZnCo-alloy phases
and obtained a transition from the Zn-rich ␩-phase of ZnCo to a
␥-phase and later to cobalt-rich deposits by increasing the cobalt
amount in the bath or by increasing the applied potential. Clearly
the composition of the coatings obtained during codeposition
depends on several factors, such as the current density, the solu-
tion composition, pH, temperature, and agitation. Anomalous
codeposition has been observed mostly at high current densi-
ties [24], whereas at low levels of current the codeposition may
be normal. On the other hand, some authors have reported that
cobalt-rich deposits are obtained at high current densities and
zinc-rich coatings at lower values [25]. The presence of other
metal ions and especially organic additives can also change the
deposit type from anomalous to normal [26,27].
Glycine has been used as a complexing agent in the electrode-
position of Zn–Ni [28], Cu–Co [29] and Zn–Co [30] alloys. It is
an attractive medium due to its ability to form complexes with
zinc and cobalt in weakly alkaline electrolytes. The characteriza-
tion of glycine-based electrolytes in alkaline solution represents
an interesting challenge because they offer a possible alterna-
tive to the use of toxic cyanide baths or zincate baths for zinc or
zinc-alloy plating. No clear mechanistic model to describe the
electrochemical behavior can yet be deduced from the previous
studies devoted to Zn–Co alloy deposition in glycinate media
[30]. The present work takes into account the solution chemistry
as it seeks to elucidate the influence of cobalt on zinc electrode-
position in an alkaline glycine medium through electrochemical
experimentation and microscopic deposit characterization. First,
an understanding of ZnCo alloy electrodeposition with glycine
requires a preliminary thermodynamic analysis to identify the
different complexes that can be formed in solution depending on
the experimental pH and, as a consequence, can affect the rela-
tive availability of the metals to form the alloy deposit. With that
background information the study presents an electrochemical
investigation of the system and reports on the effects of oper-
3. Results and discussion
3.1. Thermodynamic analysis
Glycine is a complexing amino acid (H2NCH2COOH = Lꢀ)
that provides two functional groups able to form coordination
bonds with metal cations. Glycine also presents two acidities,
the values of the acidity constants being 2.07 and 9.97 [31].
The pH used in this study (∼11) thus falls sufficiently above the
second acidity value for the divalent anion to be the predomi-

4744
J.L. Ortiz-Aparicio et al. / Electrochimica Acta 52 (2007) 4742–4751
Table 1
Equilibrium constants for the overall formation of complexes in the zinc/glycine/OH− and cobalt/glycine/OH− systems and the hydroxide solubility products obtained
from the literature [31], βn = [MLn]/[M][L]n
Metal–glycine
Metal-OH−
log β1
log β2
log β3
log β1
log β2
log β3
log β4
pKs
Zn
Co
4.88
4.69
9.06
8.46
11.56
10.81
4.4
4.3
11.3
8.4
13.7
9.7
16.6
10.2
16.92
14.8
Lꢀ is the glycine complexing agent whose form (acidity) depends
on the pH. In our experimental range of pH and analytical con-
centrations of metals and glycine, the free glycine concentration
will be about 1.6 M, corresponding to pLꢀ = −0.204. Under these
conditions the thermodynamic calculations show that MLꢀ3 is the
predominant form for both the zinc and cobalt species. For the
experimental zinc–cobalt mixed electrolytes, the glycine con-
centration was sufficiently elevated to keep these two species
predominant.
nant form. Compared to other common complexing agents like
EDTA, glycine forms less strong complexes; this makes it more
amenable to effective effluent treatment.
The thermodynamic description of the complexes formed
under our experimental conditions was obtained by construct-
ing predominance-zone diagrams and Pourbaix diagrams for
the systems of Zn–glycine and Co–glycine. The diagrams were
computed using the method described in the literature by Rojas-
´
Hernandez et al. [32–34]. Table 1 gives the values of the overall
formation constants βn for the metal–glycine complexes, those
for the metal hydroxides, and the solubility products for the
hydroxides [31]. We assumed that there are no mixed complexes
(i.e., hydroxy-glycine-metal species).
Fig. 1a and b shows the predominance-zone diagrams for the
Zn(II)ꢀ and Co(II)ꢀ species. The generalized notation M(II)ꢀ for
the dominant species (with M = Zn or Co) may represent either
the free M2+ ions in the solution, the soluble metal–glycine com-
plexes, orthemetalhydroxideineithersolubleorinsolubleform;
The Pourbaix diagrams, Fig. 1c and d, show the variation
of the equilibrium potentials for the redox couples M(II)/M(0)
for zinc and for cobalt with 2.2 M glycine at pH 11. The com-
plexing agent displaces the equilibrium potentials toward more
negative values. It is seen that cobalt deposition still should occur
at more positive potentials compared to zinc. It appears that a
small pH interval is available for zinc-hydroxide precipitation
between 12.6 and 13.2 before zincate forms. Insoluble cobalt-
hydroxide formation may occur at pH higher than 12.7. We
Fig. 1. Predominance-zone diagrams for (a) Zn(II)ꢀ; (b) Co(II)ꢀ. Pourbaix diagrams for (c) Zn(II)ꢀ; (d) Co(II)ꢀ. The total metal cation concentrations considered in
the computation of these diagrams were 0.20 M for zinc and for cobalt, 0.005 M (heavy line) and 0.1 M (thin line). Free Lꢀ was taken to be 1.6 M (pLꢀ = −0.204),
and hydroxide concentration was determined by the pH.

J.L. Ortiz-Aparicio et al. / Electrochimica Acta 52 (2007) 4742–4751
4745
therefore assume that, from a purely thermodynamic point of
view and with the parameters found in the literature, under our
experimental conditions both zinc and cobalt reduction should
involve primarily the single MLꢀ3 complexes, and the overall
reactions will be the following:
due to the formation of even a sub-monolayer of Zn on ferrous
substrates. Those authors propose that zinc upd competes with
proton reduction [36]. On this basis, we suspect that a similar
effect occurs in our system.
The second peak in curve a (Ic) is clearly associated with
the massive reduction of Zn(II)ꢀ to Zn(0). When the sweep was
reversed, an oxidation peak (Ia) was observed at −1.26 V ver-
sus SCE due to the oxidation of the deposit. As expected, peak
Ic shifted to more positive potentials when the zinc concentra-
tion was increased. When the zinc concentration was 0.4 M, the
oxidation peak became wider, and the peak current decreased.
This could be attributed to changing structure of the deposit with
increased thickness, but it is also possible that under these con-
ditions the electro-oxidation process is limited by slow chemical
reactions or diffusion.
The voltammetric behavior of the 0.1 M CoSO4 solution
(curve b) presents an electroreduction peak current (IIc) located
at −1.426 V versus SCE. A subsequent increase in current is
observed at more negative potentials due to the hydrogen evo-
lution reaction, which is not inhibited as in the case of the zinc
solutions, zinc having a higher overpotential for hydrogen evo-
lution. When the direction of the potential sweep is reversed,
a crossover potential associated with nucleation is observed at
−1.23 V versus SCE. When the potential scan is reversed in the
rising portion of peak IIc, the crossover potential varies, indicat-
ing additional processes during cobalt nucleation. An oxidation
peak for pure Co is observed at −0.756 V versus SCE (peak IIa).
According to the equilibrium potential of cobalt shown in Fig. 1d
ZnLꢀ3 + 2e → Zn(0) + 3Lꢀ
CoLꢀ3 + 2e → Co(0) + 3Lꢀ
(1)
(2)
Under our experimental conditions, the stability of insoluble
zinchydroxide, i.e. Zn(OH)2, isrestrictedtoanarrowpHinterval
above 12.6. In addition, the thermodynamic study predicts that
the dominant cobalt species becomes insoluble Co(OH)2 above
pH 12.7.
4. Cyclic voltammetry
4.1. Zinc and cobalt electrodeposition from the glycinate
electrolyte
Fig. 2 presents cyclic voltammograms for the alkaline glycine
solutions of Zn(II)ꢀ and of Co(II)ꢀ individually and for the glyci-
nate solution in the absence of metals. It is seen that Zn presents
two reduction steps; a small peak A appears around −1.26 V,
and peak Ic is located at −1.53 V versus SCE. The former could
be associated with incipient hydrogen reduction, which is seen
occurring with metal-free glycine solution in Fig. 2 (curve c)
but then is presumably inhibited by underpotential deposition
(UPD) of Zn. In the absence of metals, curve c indicates that the
reductionofhydrogenbeginsat−1.2 Vandshowsitssubsequent
oxidation at peak B. On the other hand, the charge calculated
by integration of peak A gives −364 mC cm−2, which is greater
than the theoretical amount required for the formation of a Zn
monolayer. It is well known that the HER is very sensitive to the
nature of the substrate; on Zn it proceeds very slowly, whereas
on Co, Ni and Fe it exhibits a larger exchange current density
[35]. Previous studies report the inhibition of HER in the pres-
ence of zinc ions in both alkaline [35] as well as acidic [36] baths
ꢀ
(Eeq Co(II) /Co(0) = −0.912 V versus SCE) and the gap between
the oxidation and reduction peaks, the Co(II)ꢀ reduction exhibits
greater overpotential on the AISI 1018 steel substrate than does
ꢀ
Zn (Eeq Zn(II) /Zn(0) = −1.384 V versus SCE for 0.2 M ZnSO4).
4.2. Influence of cobalt on zinc electrodeposition
When both metal ions are present in the glycinate solution,
new stripping peaks (IIIa, IVa and Va) appear in Fig. 3 between
the pure zinc and pure cobalt oxidation peaks, thus indicating
Fig. 2. Cyclic voltammograms for the electrodeposition on steel of zinc and of
cobalt in 2.2 M glycine solution, pH 11, v = 10 mV s−1: (a) 0.20 M ZnSO4; (b)
0.10 M CoSO4; (c) no metal, 2.2 M glycine, pH 11.
Fig. 3. Influence of the cobalt concentration on the cyclic voltammograms for
the electrodeposition from zinc solution (0.2 M ZnSO4) in 2.2 M glycine, pH
11, v = 10 mV s−1
.

4746
J.L. Ortiz-Aparicio et al. / Electrochimica Acta 52 (2007) 4742–4751
the formation of several ZnCo alloys or, at least, intermediate
phases. AswithpurezincZnCoelectrodepositioninthepresence
of 0.005 M cobalt presents two reduction peaks. Upon rever-
sal of the potential-sweep direction, an electrooxidation peak
was observed at −1.17 V versus SCE (IIIa), associated with the
formation of a Zn-rich ZnCo alloy phase. When the cobalt con-
centration was increased to 0.025 M, the electroreduction peak
(Ic) was practically unaffected, but the electrooxidation of the
deposit showed a decrease of peak IIIa and the appearance of
a shoulder at −1.09 V versus SCE (IVa), a peak at −0.92 V
versus SCE (Va), and another shoulder at −0.86 V versus SCE
(VIa). For this higher cobalt concentration, the broadening of
peak IIIa suggests that the deposit is composed of a mixture
of diverse ZnCo phases. When the cobalt concentration was
increased to 0.1 M, the reduction peak Ic shifted slightly toward
more positive potentials, and the peak reduction current density
decreased slightly. Increases were also observed in the currents
produced at peak Va and shoulder VIa. Both cathodic and anodic
peak positions correspond to formation of nobler Zn–Co phases,
i.e. with richer Co content. Moreover, a second cathodic peak
(IIIc) appeared, and hysteresis was observed when the scan
was inverted, both effects attributed to the enhancement of the
hydrogen-evolution reaction due to theincrease ofcobalt content
in the deposit and to the formation of a new phase.
potential evidently influences the elemental composition of the
´
´
deposit. Gomez and Valles [38,39] observed on vitreous car-
bon in an acidic chloride ZnCo solution that oxidation peaks
associated only with the formation of high-zinc-content phases
appeared when E was only moderately negative, but when
␭
the applied potential was taken to more negative values, there
appeared several oxidation peaks associated with phases richer
in cobalt content. Similar results were obtained in the ZnCo
electrodeposition on AISI 1018 steel electrode [27]; Zn-rich
deposits were obtained when the cathodic potential was inverted
at low values, whereas at more negative values higher cobalt
content was obtained in the deposit. On the other hand, Fratesi
and co-workers [23] observed three reduction processes on Pt
in the Zn upd potential region, where cobalt-rich deposits were
obtained. The Zn content increased at more negative potentials.
Although they report similar behavior on steel, these reduction
´
´
peaks were not observed by Gomez and Valles [38,39] nor by
Trejo et al. [27], who found that cobalt reduction was strongly
inhibited. The electrochemical behavior of ZnCo electrodepo-
sition observed in our conditions, i.e. in alkaline electrolyte,
shows an increase of the cobalt deposition overpotential with
Zn presence similar to that observed in acidic conditions
[27,36,37].
Fig. 4 shows the influence of E on the CV for the Zn–Co
␭
´
´
According to Swathirajan [37] and to Gomez and Valles
[38,39], the multiple oxidation peaks for the electrodeposits
of Zn–Co alloys are associated with the formation of several
phases with different cobalt contents. In our electrolyte condi-
tions, Peaks IIIa, IVa and Va may be associated with different
intermediate ZnCo alloy phases, while the peak VIa is attributed
to pure cobalt (see Fig. 2). It is interesting to note that the pres-
ence of cobalt does not affect sharply the potential position
of cathodic peak Ic, although the current density diminished
slightly as higher cobalt concentrations were introduced.
glycinate system with 0.025 M CoSO4. When the E value was
␭
low, E␭1 = −1.45 V, peak Ia was the major anodic signal and
represented Zn oxidation. With E␭4 = −1.70 V, two reduction
processes (peak Ic and signal IIc) were observed: the first one
(Ic) was due to the electroreduction of Zn, and the second (IIc)
appeared at a more negative potential and was associated with
the hydrogen evolution reaction combined with cobalt deposi-
tion. When the scan direction was reversed, several oxidation
peaks appeared. In agreement with the observations of Gomez
´
and Valles in acidic conditions [38,39], signal Ia appeared pre-
It is also observed that the presence of zinc increases the
hydrogen-evolution overpotential relative to that of pure Co as
reported previously [35,36]. In alkaline conditions [35], this
inhibition is attributed to the formation of a Zn upd layer, which
also inhibits the cobalt deposition. The latter exhibits a consider-
able increase in its electrodeposition overpotential leading to an
anomalous Zn–Co codeposit. Such an effect may be involved
also in this alkaline electrolyte. Furthermore, the hydroxide-
suppression mechanism [4–8] suggests that a zinc hydroxide
layer may also adsorb on the electrode, further inhibiting cobalt
reduction. According to Fig. 1a and b, at pLꢀ = −0.204, zinc
hydroxide precipitation can occur when the pH reaches 12.6, and
cobalt precipitation takes place at pH above 12.7. Thus, insolu-
ble hydroxides could form on the electrode surface and impede
cobalt reduction. Given that the theoretical pH values for the
precipitation of the two metals are so close, both cobalt and zinc
precipitates may be involved in the codeposition process.
dominantly with low values of the inverting potential. When the
E value was increased to more negative potentials, several other
␭
oxidation processes were observed (peaks IIIa, IVa, Va and VIa).
The three first oxidation peaks represent phases with increasing
4.3. Influence of the switching potential, Eλ
Fig. 4. Influence of the switching potential on the anodic behavior of the
deposits from 0.20 M ZnSO4 + 0.025 M CoSO4 in 2.2 M glycine, pH 11, v =
10 mV s−1: E␭1 = −1.50 V vs. SCE (curve 1); E␭2 = −1.55 V vs. SCE (curve 2);
The anodic voltammetric profile observed with Zn–Co alloy
deposition depends strongly on the switching potential E , the
␭
potential at which sweep direction is inverted. The switching
E
␭3 = −1.60 V vs. SCE (curve 3); E␭4 = −1.70 V vs. SCE (curve 4).

J.L. Ortiz-Aparicio et al. / Electrochimica Acta 52 (2007) 4742–4751
4747
cobalt content, while signal VIa may be associated with a pure
cobalt phase.
The mechanism of anomalous electrodeposition of the ZnCo
alloy is still not clear; the hydroxide suppression model [4–8]
predicts the formation of anomalous ZnCo alloys when hydro-
gen evolution occurs. Accordingly, as the applied potential is
made more cathodic, the hydrogen reaction is favored, and
the appearance of the oxidation peaks associated with ZnCo
alloy phases are observed. However, at low switching poten-
tials, the anodic peak Ia appears, indicating that the deposit is
more anomalous when less hydrogen evolution occurs. It is pos-
sible that a combination of Zn upd with Zn and Co hydroxide
precipitation inhibits Co reduction during the ZnCo electrode-
position. It is noted that the faint reduction peak (A) presents
itself again around −1.2 V in Fig. 4. This small current is char-
acteristic of cyclic voltammetry on ferrous substrates [36,40]
and is attributed to the underpotential deposition of Zn and its
subsequent inhibition of hydrogen reduction.
Fig. 6. Faradaic efficiency and variation of the cobalt content in constant-
current deposits on steel as functions of the relative level of cobalt in the bath.
0.2 M ZnSO4, 2.2 M glycine, pH 11. Deposits produced at j = −10 mA cm−2 for
t = 300 s, no stirring.
energy dispersion spectrometer coupled to the scanning electron
microscope and presented in Fig. 6. The results demonstrate that
all the deposits were anomalous in the sense that the Co con-
tent in the deposit was always less than the relative proportion
present in the bath. It is seen that the current efficiency at this
current density decreased with increasing cobalt concentration,
falling from 90% with pure zinc to about 67% when the cobalt
concentration was 0.025 M. This result confirms the interpreta-
tion of the CV results, namely, that higher Co levels in the bath
and in the deposit promote the hydrogen-evolution reaction.
Fig. 7a shows the alloy composition profile with depth within
the deposit formed under unstirred conditions from the 0.4 M
ZnSO4–0.05 M CoSO4–glycine solution at −10 mA cm−2. It is
seen that the alloy composition was not constant across the film.
Close to the steel substrate, the cobalt content was almost 4%.
Further out in the film, in the portion formed as electrodepo-
sition continued, the cobalt content decreased and maintained
a value of about 2.7% until dropping off further near the outer
surface. This graph shows again that under these conditions the
codeposition was anomalous as the relative cobalt concentra-
tion in the solution was 11%. No significant level of oxygen was
detected in the film except on the outer surface. In this case zinc
or cobalt oxide or hydroxide may have formed on the surface, but
none was incorporated in the deposit. It is possible that the elec-
trochemical reduction reaction is faster than the rate of crystal
growth, thus releasing the hydroxide anions to the bulk solution.
This profile for cobalt level in the deposit suggests that several
steps take place during ZnCo electrodeposition, and indeed dif-
ferent mechanisms may apply as the interface changes and the
film grows.
5. Characterization of the deposits
5.1. Composition of the coating
Thin deposits were formed on the steel disks under non-
stirring conditions from a solution containing 0.2 M ZnSO4
and various levels of Co by applying a constant current for
300 s. Fig. 5 presents the stripping voltammograms of the ZnCo
deposits obtained under galvanostatic conditions at different cur-
rent densities. The profile observed is similar to that observed
in Fig. 3, i.e. several oxidation peaks appear corresponding to
different phases of the ZnCo alloy. Curve f corresponds to the
stripping of a pure cobalt deposit.
The faradaic current efficiency for metal deposition was
determined by doing anodic stripping of the deposit and inte-
grating the anodic current in order to calculate the ratio of the
anodic to cathodic charges. The results are shown in Fig. 6 as a
function of the relative level of cobalt in the bath. The elemental
composition of the alloy deposits was also determined using an
When the deposits were obtained from a 0.2 M ZnSO4–0.1 M
CoSO4–glycine solution, a change in the composition profile
was observed. Fig. 7b shows that at −10 mA cm−2 the cobalt-
content profile across the deposit seems to be similar to that
observed in Fig. 7a, but the oxygen content in the deposit
increased. Both the oxygen and the cobalt content in the deposit
increased with an increase in the applied current density to
−30 mA cm−2 as seen in Fig. 7c, indicating occlusion of a
Fig. 5. Stripping voltammograms for ZnCo alloy deposits formed from 0.20 M
ZnSO4 + 0.025 M CoSO4 in 2.2 M glycine under constant current conditions:
(a) −1 mA cm−2; (b) −5 mA cm−2; (c) −10 mA cm−2; (d) −20 mA cm−2
;
(e) −30 mA cm−2; (f) pure Co from 0.1 M CoSO4 + 2.2 M glycine, pH 11 at
−20 mA cm−2; v = 1 mV s−1, deposition for 300 s.

4748
J.L. Ortiz-Aparicio et al. / Electrochimica Acta 52 (2007) 4742–4751
Fig. 7. CompositionprofileswithdepthbyGDSanalysisforZnCodepositsobtainedfrom(a)0.4 MZnSO4 + 0.05 MCoSO4 + 2.2 Mglycine, pH11, j = −10 mA cm−2
;
(b) 0.2 M ZnSO4 + 0.1 M CoSO4 + 2.2 M glycine, pH 11, j = −10 mA cm−2; (c) same electrolyte as in (b) with j = −30 mA cm−2. Cobalt scale is amplified by 10
times in (a) and (b). Carbon scale is amplified by 100 times.
metal-hydroxide precipitate related to the occurrence of hydro-
gen reduction. Furthermore, the cobalt content is higher and
more homogeneous than that observed in Fig. 7a and b, indicat-
ing the formation of a new phase. According to the ratio of Zn
to Co, the new alloy was a ␥-phase. This result can be correlated
with the additional anodic peaks appearing in Figs. 3 and 4. At
the same time, the carbon distribution also changed with Co(II)
level. In Fig. 7a and b, it is observed that the C profile is sim-
ilar to that of cobalt, rising near the steel interface, whereas
at the higher current density it falls off from the outer surface
of the deposit. This change may be related to the morphologi-
cal changes observed in the SEM images that are presented in
Fig. 9. In an acid electrolyte, the transition from a ␩-phase to a
␥-phase was found to occur with an increase of the cobalt level
in the bath [23].
These various results show that codeposition of the two ele-
ments does not occur according to the pure-metal deposition
kinetics. Cobalt deposition is inhibited by zinc at low current
densities or low applied potential. At the same time zinc depo-
sition is retarded by the presence of cobalt. Cobalt deposition
becomes favored at more negative applied potentials and leads
to the formation of several Zn–Co alloy phases and even a pure
The effect of current density on current efficiency was also
determined for the alloy bath with 0.025 M Co. Fig. 8 shows
the metal-deposition current efficiency as a function of current
density. Also plotted is the cathode potential. It is seen that the
current efficiency is low at low current densities but increases
rapidly to about 80%. At higher current densities, the efficiency
´
falls off as the hydrogen reaction becomes significant. Alcala
et al. [25] found that anomalous codeposition is favored at low
applied overpotentials and at high Zn(II)/Co(II) ratios. Accord-
ingly, loss of current efficiency was related to increased cobalt
concentration in the deposit at higher current densities because
of activation of the hydrogen-evolution reaction by increased
cobalt content.
Fig. 8. The current efficiency for ZnCo electrodeposition (t = 300 s) from 0.2 M
ZnSO4 + 0.025 M CoSO4 + 2.2 M glycine, pH 11, plotted as a function of the
applied current density (curve a). Curve b shows the cathode potential recorded
during the galvanostatic deposition experiments, no stirring.

J.L. Ortiz-Aparicio et al. / Electrochimica Acta 52 (2007) 4742–4751
4749
Fig. 9. Influence of the cobalt concentration and current density on the morphology of the deposits obtained from glycinic 0.2 M ZnSO4 at pH 11, j = −10 mA cm−2
,
t = 300 s. Cobalt concentrations: (a) 0.0 M; (b) 0.01 M; (c) 0.02 M; (d) 0.03 M; (e) 0.05 M; (f) 0.05 M with j = −30 mA cm−2
.
cobalt phase. Hydrogen evolution is promoted by the presence
of cobalt in the deposit, and combination of the information in
Figs. 5–8 shows that the cobalt level in the deposit affects cur-
rent efficiency. The enhancement of the hydrogen reaction with
the concomitant rise in interfacial pH probably plays a role in
moderating the rate of zinc reduction as well as in the overall
alloy-formation process.
andKarwasandHepel[43]foracidic-chlorideelectrolytes. They
obtainedfine-graineddepositsinthepresenceofhighzinc/cobalt
electrolyte ratios but dendritic deposits when the relative cobalt
concentration was increased.
With a cobalt concentration of 0.02 M, a deposit with even
smaller grain size was observed (Fig. 9c), presenting an amor-
phous structure, and this was also observed at higher cobalt
concentrations (up to 0.05 M).
Fig. 9f shows the influence of the current density on the mor-
phology of the deposits of the ZnCo alloy. It is seen that the
deposit obtained with 0.05 M Co at −30 mA cm−2 presented a
compact polyhedral morphology, in contrast to the deposits with
fine grains obtained at −10 mA cm−2. Therefore, Co plays the
role of a grain-size refiner at low current densities even at high
Co concentrations. At higher current density and high Co con-
tent in the deposit, Co imposes its own structure with a more
columnar growth. In order to verify this last assumption, we
performed crystallographic analysis of all the deposits.
5.2. Morphology of the deposits
The presence of other metals in Zn plating baths modifies
the morphology of the deposits because of their influence on
the growth of the initial crystal nuclei [25,41,42]. Fig. 9 shows
the SEM images for electrodeposits obtained from the alkaline
Zn and Zn–Co glycine solutions. Pure zinc deposits (Fig. 9a)
present characteristic hexagonal crystals, which are normally
obtained in zinc deposition [43,44]. The zinc crystals exhibited
a range of sizes, and the larger grains were of approximately
5 ␮m in diameter on which smaller crystals grew. In the pres-
ence of 0.01 M cobalt in the bath, an appreciable change was
observed in the deposit morphology with the grain size decreas-
ing to around 1 ␮m diameter, the deposits being compact and
presenting a clear gray color. The presence of cobalt apparently
modified the growth of zinc nuclei, leading to the fine-grained
deposits obtained under anomalous composition. This observa-
5.3. Crystallographic structure
The preferential crystal orientation of zinc electrodeposits
depends on the experimental conditions [44,45], such as current
density, pH and temperature. Plane (0 0 2) has been associated
with the growth of hexagonal zinc crystals parallel to the sub-
strate [44]. Fig. 10 shows the X-ray diffraction patterns obtained
´
tion is in accord with the results reported by Alcala et al. [25]

4750
J.L. Ortiz-Aparicio et al. / Electrochimica Acta 52 (2007) 4742–4751
is evidence of the formation of Zn13Co due to the presence of
small peaks at 40.3◦ and 41.2◦ as well as the peak at 2θ = 43.02◦
(PDF 29 0523). The wide peak located at 2θ = 44.56◦ (peak
v) may be associated with the presence of a pure cobalt phase
(PDF 01-1255). These changes in the X-ray patterns for higher
current density correspond to the changes in the electrochemi-
cal behavior seen in Fig. 3 and to the morphology observed in
Fig. 9f, and they reflect the modifications of the composition
profiles of the ZnCo alloy shown in Fig. 7c. All of the observa-
tions with the crystallographic structures, the morphology, and
the composition of the coatings may be associated with changes
in the electrochemical conditions under which the deposits are
formed.
Rashwan et al. [30] also studied ZnCo electrodeposition
in glycine solution; they obtained anomalous deposits and
observed that the crystallographic structure depended on applied
current density. They reported the appearance of Co5Zn21 and
CoZn13 phases at low and higher current densities, respectively.
The GDS analysis showed the presence of oxygen within the
deposit, and this may be attributed to an increase in the inter-
facial pH during deposition. The formation of cobalt hydroxide
at increased current density or higher Co(II) concentration is
confirmed by the appearance of the X-ray peaks associated with
cobalt hydroxide.
Fig. 10. X-ray diffraction patterns for Zn and alloy electrodeposits obtained
on AISI 1018 steel, j = −10 mA cm−2, t = 300 s. Bath cobalt concentrations: (a)
0.0 M; (b) 0.01 M; (c) 0.05 M; (d) 0.10 M; (e) 0.05 M with j = −30 mA cm−2
.
for several Zn and ZnCo alloy deposits. In the absence of cobalt,
the peaks associated with the characteristic crystallographic
planes for zinc were observed, the peak associated with plane
(0 0 2) having the highest intensity.
6. Conclusions
When the deposit was formed in the presence of 0.01 M
Co(II), a considerable decrease of the (0 0 2) peak was observed,
as well as of the peaks associated with planes (1 0 2), (1 0 3) and
(1 1 0), duetothechangeinorientationoftheZncrystallographic
planes by the incorporation of cobalt. The persistence of those
peaks, although with lesser intensity, suggests the appearance of
a zinc-rich (␩-phase) ZnCo alloy [46], since the diffraction pat-
tern of the Zn–Co solid solution ␩-phase is very similar to pure
Zn diffraction pattern [44]. When the cobalt content in solution
was increased to 0.05 M, the peak associated with plane (0 0 2)
diminished further. A peak then appeared at a value of 2θ = 42.5◦
(peak i), which is associated with the formation of cobalt oxide
(PDF 43-1004).
An increase of peak (i) was observed when the cobalt level
in the solution was increased to 0.1 M. At the same time, the
intensity of the peaks for planes (0 0 2) and (1 0 1) of the ␩-phase
decreased further. The decrease of those peaks may be attributed
to the decrease of the thickness of the coatings obtained when
the cobalt concentration in the bath increased because of the loss
of current efficiency with increasing cobalt levels.
The deposit obtained at higher current density and 0.05 M
Co(II) (Fig. 9e) presents a change in the X-ray diffraction pat-
tern. Three peaks appear at 2θ = 37.64◦ (peak ii), 68.04◦ (peak
iii), and 69.56◦ (peak iv), which are associated with the presence
of cobalt hydroxide (PDF 45-0031). These results agree with the
presence of oxygen observed in Fig. 7c, its level increasing as
the distance from the substrate increases. The peak associated
with the plane (1 0 1) shifted from 43.25◦ to 43.02◦, indicating
that the ZnCo alloy crystal lattice was deformed. At the same
time, the peak associated with plane (0 0 2) disappeared com-
pletely, showing a preferential orientation of the crystals. There
This work has explored the influence of cobalt on zinc elec-
trodeposition from alkaline glycine solutions. Thermodynamic
stability calculations identified the potential stability of the
various complexes that can form in this complexing medium.
These thermodynamic calculations showed that the experimen-
tal conditions favor the formation of Zn–glycine and Co–glycine
complexes, the dominant species being ZnLꢀ3 and CoL3ꢀ . The
effects on the electrochemical behavior of the system of the zinc
and cobalt concentrations were studied experimentally.
The cobalt-deposition overpotential increased in the pres-
ence of zinc ions, while the presence of Co(II)ꢀ ions favored
the hydrogen-evolution reaction. Based on the predominance
zone diagrams and according to the electrochemical studies per-
formed, it is suggested that the alloy deposition process can be
explained in terms of the metal hydroxide suppression mech-
anism, in which the initial formation and adsorption on the
electrode surface of an insoluble Zn(OH)2 and, later on, of insol-
uble Co(OH)2, moderate the reduction of both zinc and cobalt,
especially when the hydrogen-evolution reaction becomes more
intense. A switching-potential study showed that Zn-rich ZnCo
deposits are obtained at the early stages of electrodeposition, i.e.
at moderately negative potentials.
The presence of cobalt in the deposit was evidenced by the
appearance of several stripping peaks associated with different
phases of Zn–Co that depend on the cobalt bath concentration
andontheappliedpotentialorcurrentdensity. Thecobaltcontent
modified the growth habit of the deposits, favoring the forma-
tion of smaller grains and increasing the number of crystals
formed. When the cobalt concentration was increased, finer-
grained deposits were obtained. When current density increased,

J.L. Ortiz-Aparicio et al. / Electrochimica Acta 52 (2007) 4742–4751
4751
the deposit morphology was modified, exhibiting polyhedral
crystals.
[15] H. Yan, J. Downes, P.J. Boden, S.J. Haris, J. Electrochem. Soc. 143 (1996)
1577.
[16] M.J. Nicol, H.I. Philip, J. Electroanal. Chem. 70 (1976) 233.
[17] T. Ohtsuka, A. Komory, Electrochim. Acta 43 (1998) 3269.
[18] S. Swathirajan, J. Electroanal. Chem. 221 (1987) 211.
[19] M.F. Mathias, T.W. Chapman, J. Electrochem. Soc. 134 (1987) 1408.
[20] D. Landolt, Electrochim. Acta 39 (1994) 1075.
[21] T. Tsuru, S. Kobayashi, T. Akiyama, H. Fukushima, S.K. Gogia, R. Kam-
mel, J. Appl. Electrochem. 27 (1997) 209.
[22] J.-F. Huang, I.-W. Sun, J. Electrochem. Soc. 151 (2004) C8.
[23] G. Roventi, T. Bellezze, R. Fratesi, Electrochim. Acta 51 (2006) 2691.
[24] E. Chassaing, R. Wiart, Electrochim. Acta 37 (1992) 545.
[25] M.L. Alcala´, E. Go´mez, E. Valle´s, J. Electroanal. Chem. 370 (1994)
73.
[26] H.-M. Wang, T.J. O’Keefe, J. Appl. Electrochem. 24 (1994) 900.
[27] G. Trejo, R. Ortega, Y. Meas, E. Chainet, P. Ozil, J. Appl. Electrochem. 33
(2003) 373.
[28] I. Rodriguez-Torres, G. Valentin, S. Chanel, F. Lapicque, Electrochim. Acta
46 (2000) 279.
The study of X-ray diffraction patterns showed that cobalt
affects the dominant zinc plane (0 0 2). The presence of cobalt
oxide and hydroxide in the alloy was also indicated, possibly
formed by the occlusion of Co(OH)2(s) from a surface precipi-
tate. Cobalt content was not homogeneous across the deposited
film, indicating that several processes are probably involved
during electrodeposition and film growth.
Further studies of the system, leading to quantitative
electrode-kinetics models, are required to predict the alloy
composition of coatings produced under various operating
conditions.
Acknowledgements
[29] A.E. Mohamed, S.M. Rashwan, S.M. Abdel-Wahaab, M.M. Kamel, J.
Appl. Electrochem. 33 (2003) 1085.
[30] S.M. Rashwan, A.E. Mohamed, S.M. Abdel-Wahaab, M.M. Kamel, J.
Appl. Electrochem. 33 (2003) 1035.
[31] R.M. Smith, A.E. Martell, Critical Stability Constants, Plenum, New York,
1989.
[32] A. Rojas-Herna´ndez, M.T. Ram´ırez, I. Gonza´lez, J. Iban˜ez, Anal. Chim.
Acta 259 (1992) 95.
[33] A. Rojas-Herna´ndez, M.T. Ram´ırez, I. Gonza´lez, Anal. Chim. Acta 278
(1993) 321.
[34] A. Rojas-Herna´ndez, M.T. Ram´ırez, I. Gonza´lez, Anal. Chim. Acta 278
(1993) 335.
[35] J.H.O.J. Wijenberg, J.T. Stevels, J.H.W. de Wit, Electrochim. Acta 43
(1997) 649.
[36] J. Lee, J.-W. Kim, M.-K. Lee, H.-J. Shin, H.-T. Kim, S.-M. Park, J. Elec-
trochem. Soc. 151 (2004) C25.
[37] S. Swathirajan, J. Electrochem. Soc. 133 (1986) 671.
[38] E. Go´mez, E. Valle´s, J. Electroanal. Chem. 397 (1995) 177.
[39] E. Go´mez, E. Valle´s, J. Electroanal. Chem. 421 (1997) 157.
[40] E.A. Maleeva, K.S. Pedan, I.I. Ponjomarev, Russ. J. Electrochem. 32 (1996)
1380.
[41] G. Barcelo´, J. Garc´ıa, M. Sarret, C. Muller, J. Appl. Electrochem. 24 (1994)
1249.
[42] R. Fratesi, G. Roventi, J. Appl. Electrochem. 22 (1992) 657.
[43] C. Karwas, T. Hepel, J. Electrochem. Soc. 136 (1989) 1672.
[44] J.B. Bajat, V.B. Miskovic-Stankovoc, M.D. Maksimovic, D.M. Drazic, S.
Zec, Electrochim. Acta 47 (2002) 4101.
This work was supported by the Consejo Nacional de Ciencia
´
´
y Tecnologıa (CONACYT, Mexico) via Grants 31411-U. JLO
thanks CONACyT-Mexico for the fellowship. TWC is a CONA-
CyT Cooperante in CIDETEQ under the sponsorship of the US
Peace Corps.
References
[1] H. Herb, Zinc Plating, ASM International, Great Britain, 1988.
[2] E.C. David, Met. Finish. 89 (1991) 58.
[3] A. Brenner, Electrodeposition of Alloys, vol. I, Academic Press, New York,
1963.
[4] H. Dahms, I.M. Croll, J. Electrochem. Soc. 112 (1965) 771.
[5] M. Yunus, C. Capel-Boute, C. Decroly, Electrochim. Acta 10 (1965) 885.
[6] J. Mindowicz, C. Capel-Boute, C. Decroly, Electrochim. Acta 10 (1965)
901.
[7] K. Higashi, H. Fukushima, T. Urakawa, T. Adaniya, T. Matsudo, J. Elec-
trochem. Soc. 128 (1981) 2081.
[8] A. Stankeviciute, K. Leinartas, G. Bikulcius, D. Virbalyte, A. Sudavicius,
E. Juzeliunas, J. Appl. Electrochem. 28 (1998) 89.
[9] H. Deliglianni, L.T. Romankiw, IBM J. Res. Dev. 37 (1993) 85.
[10] S. Hesami, C.W. Tobias, AIChE J. 39 (1993) 149.
[11] J. Vaes, J. Fransaer, J.-P. Celis, J. Electrochem. Soc. 149 (2002) C567.
[12] F.J. Fabri Miranda, O.E. Barcia, O.R. Mattos, R. Wiart, J. Electrochem.
Soc. 144 (1997) 3441.
[13] F.J. Fabri Miranda, O.E. Barcia, O.R. Mattos, R. Wiart, J. Electrochem.
Soc. 144 (1997) 3449.
[14] S.L. D´ıaz, O.R. Mattos, O.E. Barcia, F.J. Fabri Miranda, Electrochim. Acta
47 (2002) 4091.
[45] Y. Oren, U. Landau, Electrochim. Acta 27 (1982) 739.
[46] Powder Diffraction File, Inorganic Volume PD/S 5iRB, American Society
for Testing and Materials, USA, 1969.