Formation mechanisms and characterization of black and white cobalt electrodeposition onto stainless steel

Article abstract of DOI:10.1149/1.1393435

Cobalt electrodeposition onto a stainless steel substrate from 1.17 M Co(II) aqueous solution containing 0.98 M H₂SO₄, 0.56 M KCl, and 0.2 M H₃BO₃ was evaluated in the absence (i) and presence (ii) of 0.1 M KNO₃. Cobalt electrodeposited from the electrolytic bath (i) was white-gray colored, whereas deposition from bath (ii) formed a black-colored surface. SEM-WDX, AFM, and XRD analysis of the steel surfaces covered with these two deposits revealed distinct characteristics for black and white cobalt films. Although both deposits were composed of metallic cobalt, the white cobalt deposit was a smooth, 2D film while the black deposit consisted of many dispersed, nano-sized clusters of 150 to 250 nm in diameter. Analysis of potentiostatic current transients (I-t curves) indicated that formation of white cobalt was carried out by multiple 3D nucleation limited by lattice incorporation of cobalt adatoms to the growth centers. Formation of black cobalt was shown to involve the simultaneous processes of 3D nucleus formation and growth, limited by mass transfer, and the reduction of nitrates in the medium onto the surfaces of these nuclei. It is shown that, beside this cobalt-nitrate interaction, NO₃^- ions in solution can block active sites for cobalt reduction and the effect of this phenomenon strongly depends on the nitrate concentration. These facts could explain the observed dispersion of the black cobalt coating.

Full text of DOI:10.1149/1.1393435

Journal of The Electrochemical Society, 147 (5) 1787-1796 (2000)
1787
S0013-4651(99)05-090-9 CCC: $7.00 © The Electrochemical Society, Inc.
Formation Mechanisms and Characterization of Black and White Cobalt
Electrodeposition onto Stainless Steel
Enrique Barrera,a Manuel Palomar Pardavé,b,* Nikola Batina,c and Ignacio Gonzálezc,**,z
aDepartamento de Ingeniería de Procesos e Hidráulica, Área de Ingeniería en Recursos Energéticos and cDepartamento de
Química área de Electroquímica, Universidad Autónoma Metropolitana-Iztapalapa, C.P. 09340, México, D.F., Mexico
bDepartamento de Materiales, Área de Ciencia de los Materiales, Universidad Autónoma Metropolitana-Azcapotzalco,
C.P. 02200, México D.F., Mexico
Cobalt electrodeposition onto a stainless steel substrate from 1.17 M Co(II) aqueous solution containing 0.98 M H2SO4, 0.56 M KCl,
and 0.2 M H3BO3 was evaluated in the absence (i) and presence (ii) of 0.1 M KNO3. Cobalt electrodeposited from the electrolytic
bath (i) was white-gray colored, whereas deposition from bath (ii) formed a black-colored surface. SEM-WDX, AFM, and XRD
analysis of the steel surfaces covered with these two deposits revealed distinct characteristics for black and white cobalt films.
Although both deposits were composed of metallic cobalt, the white cobalt deposit was a smooth, 2D film while the black deposit
consisted of many dispersed, nano-sized clusters of 150 to 250 nm in diameter. Analysis of potentiostatic current transients (I-t curves)
indicated that formation of white cobalt was carried out by multiple 3D nucleation limited by lattice incorporation of cobalt adatoms
to the growth centers. Formation of black cobalt was shown to involve the simultaneous processes of 3D nucleus formation and
growth, limited by mass transfer, and the reduction of nitrates in the medium onto the surfaces of these nuclei. It is shown that, beside
this cobalt-nitrate interaction, NOϪ3 ions in solution can block active sites for cobalt reduction and the effect of this phenomenon
strongly depends on the nitrate concentration. These facts could explain the observed dispersion of the black cobalt coating.
© 2000 The Electrochemical Society. S0013-4651(99)05-090-9. All rights reserved.
Manuscript submitted May 24, 1999; revised manuscript received December 15, 1999.
cyanate in N,N-dimethyl formamide5,6 and more recently starting
from Co(II) in ammonium chloride aqueous solutions.7,8 Details of
electrodeposition, including the influence of the coordination sphere
on the mechanism of the cobalt nucleation, were included in these
publications.7,8 Reports on black cobalt electrodeposition, however,
have yet to define the mechanisms involved. McDonald9 reported that
the black cobalt coating is formed by cobalt(II) oxide when it is elec-
trodeposited from a Watts-type electrolytic bath containing hydrogen
peroxide, although he did not describe the electrocrystallization
mechanisms for this compound. Smith et al.10 prepared black cobalt
using various procedures. One of their better results was achieved in
a two-step process where the white metallic cobalt was first deposit-
ed on a nickel substrate and then converted to black cobalt through
chemical oxidation (in an ammonium persulfate medium). Once
again, however, no details of the process were included. Hutchins et
al.11 also prepared black cobalt using different techniques, one of
which involved electroformation of white cobalt. Once prepared, the
white cobalt film was activated in nitric acid aqueous solution to form
cobalt nitrate, which, as they attested, facilitated the cobalt oxide or
hydroxide formation under a strong alkaline medium (ammonium
persulfate). We report here on the deposition mechanism of cobalt
onto steel in the presence of nitrates. We found nitrates to be an essen-
tial component for direct cobalt electrodeposition, producing black
cobalt with excellent photothermal properties.2
Photothermal conversion of solar energy to generate electricity is
currently one of the most extensively applied techniques for har-
vesting solar energy worldwide. A major goal of companies devoted
to the generation of power from solar energy photothermal conver-
sion is the development of more competitive processes to reduce
energy costs through improvement of the thermal performance of
solar collectors.1 Such improvements demand the use of solar col-
lectors formed using highly efficient surfaces with high solar ab-
sortance (␣) and low thermal emittance (⑀). The former characteris-
tic is required in order to collect the maximum possible incident
solar energy in the collector lot and the latter to reduce energy loss-
es due to heat radiation. Coatings formed from black metals (i.e.,
black cobalt, black chromium, black nickel) have been shown to
form fairly efficient collector surfaces, fulfilling the prescribed char-
acteristics.1,2 Several techniques, such as chemical conversion and/
or thermal oxidation of metallic films and electrodeposition, are cur-
rently used to achieve such spectrally selective, black-metal, solar
absorber surfaces. However, the desired characteristics of the metal-
lic coating could be better controlled by directed electrodeposition.
Therefore, we propose to use electrodeposition to prepare a co-
balt photothermal material suitable for use in solar energy collection.
The mechanisms of electrochemical deposition for such a coating
must be defined to enable preparation of materials with the required
optical properties. In order to achieve control and reproducibility for
the coating to qualify as an industrially useful material, the deposi-
tion process, as well, must be properly characterized.
Experimental
We previously produced a black cobalt deposit with adequate
optical properties for the photothermal conversion of solar energy.2
This deposit was obtained by modifying the composition of an elec-
trolytic bath for white cobalt production through the addition of a
small amount of potassium nitrate (0.1 M). We reported the compo-
sition of the electrolytic bath required to produce black cobalt using
a Hull cell. One of the main goals of our present work is to describe
the mechanism of black and white cobalt deposition onto stainless
steel and the role of nitrate in black cobalt formation.
Cobalt deposition has been carried out on substrates such as
nickel, vitreous carbon, and copper, and from different electrolytic
baths containing chloride or sulfate aqueous solutions,3,4 triethylene-
diamine cobalt(III) chloride in 30% KOH, solutions of cobalt(II) thio-
Cobalt film preparation and electrochemical characterization of
cobalt film properties.—The kinetics of the cobalt electrodeposition
process were studied from a typical electrolytic bath, as described by
McDonald9 and modified as we reported previously.2 In this study,
we used two different electrolytic baths: A, without nitrate to form
white cobalt, and B, with the oxidizing agent, NOϪ3 , to obtain the
black cobalt deposit (see composition in the Table I). An electro-
chemical study was performed using cyclic voltammetry and poten-
tial-step techniques in each of these two baths.
All solutions were prepared with ultrapure, Milli-Q grade water
and analytical grade reagents. Experiments were carried out at room
temperature in a nitrogen atmosphere. Prior to electrochemical ex-
periments, all solutions were carefully deaerated using clean, nitro-
gen (99.99% purity) gas.
** Electrochemical Society Student Member.
A conventional three-electrode cell was used for the experiments.
As a working electrode (substrate for cobalt deposition), a stainless
** Electrochemical Society Active Member.
z
*
E-mail: igm@xanum.uam.mx
Downloaded on 2014-11-12 to IP 128.59.222.12 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

1788
Journal of The Electrochemical Society, 147 (5) 1787-1796 (2000)
S0013-4651(99)05-090-9 CCC: $7.00 © The Electrochemical Society, Inc.
(Fig. 2c) which is a dispersed 2D film with some occasional clusters.
Interestingly, cobalt clusters found on both surfaces were almost equal
in size (approximately 150 to 250 nm in diam). From a macroscopic
point of view, it is also interesting to estimate the surface roughness
for the clean and cobalt covered substrate. Surface roughness, express-
ed as rms, the root-mean-square function12-14 is defined as the stan-
dard deviation for the height of all imaged features. In this analysis all
image points are taken into account. Black and white cobalt possess-
es significantly different rms values of 69 vs. 40 nm, respectively. This
Table I. Chemical composition of the aqueous electrolytic baths
used in this work for white (A) and black (B) cobalt film
formation onto stainless steel.
Concentration (mol LϪ1
)
Substance
A
B
(White cobalt)
(Black cobalt)
CoSO4
CoCl2
H3BO3
KNO3
0.98
0.28
0.20
0.00
0.98
0.28
0.20
0.10
steel (SS) square (type 304) with a 1.0 cm2 surface area was embed-
ded in an epoxy resin in such a way that only one phase was exposed
to the solution. A graphite rod was utilized as the counter electrode
and a saturated calomel electrode (SCE) as the reference electrode.
Prior to each experiment, the working electrode surface was polished
to a mirror finish with several grades of alumina powder and treated
in pure water in an ultrasonic bath for 3 min. During the experi-
ments, the working electrode potential was controlled by means of
an EG&G model 263-A potentiostat-galvanostat coupled to a per-
sonal computer using Echem software. This device provided experi-
mental control and data acquisition.
Cobalt film characterization by SEM-WDX, AFM, and XRD.—
Surface characterization of the metallic coating was obtained by
X-ray diffraction (XRD) using a Siemens D500 difractometer with
Cu K␣ radiation. We used scanning electron microscopy (SEM) to
observe the overall properties of the cobalt films prepared in an elec-
trolyte solution with and without nitrate. This study was carried out
using a SEM Carl Zeis DSM 940A microscope coupled to a WDX
detector. In this work, WDX measurements were taken in the wave-
length (␭) range of 1.14 Å < ␭ < 94.36 Å. An atomic force micro-
scope (AFM), Nanoscope III, Digital Instruments, USA) operating
in air (ex situ experiments) was used to visualize the detailed struc-
ture of the cobalt film. To determine and define the cobalt film for
quantitative morphological characteristics (surface roughness, esti-
mation of the average grain size, etc.), the multimode software pack-
age accompanying the Nanoscope III was used. Measurements were
carried out by “contact” in air using standard geometry silicon
nitride probes (Digital Instruments). All images are shown in the so
called “height mode,” where higher areas appear brighter.
Results and Discussion
Cobalt Deposit Characterization
Scanning electron microscopy.—Figure 1 shows two SEM images
of the stainless steel substrate surfaces covered with white (Fig. 1a)
and black (Fig. 1b) cobalt. Cobalt was electrodeposited onto the
stainless steel substrate by means of cathodic potential steps from the
two aqueous solutions, A and B (see Table I). The electrodeposition
potentials were based on a previous study.2 In the absence of nitrates
(Fig 1a), the deposit is compact while with the addition of nitrates,
the deposit consists of dispersed particles (Fig. 1b). Both samples still
possess traces of the polishing lines (long-range scratches due to the
polishing procedure), indicating that the cobalt layer is not thicker
than approximately 0.3 ␮m. The white cobalt film appears smooth,
more like a 2D film, and compact, which indicates association among
the deposited particles. In some regions, defects in the form of small
pits were observed.
Characterization of surface morphology of cobalt films by AFM.—
Although AFM imaging provides more surface structural details than
SEM, the technique served mainly to verify our conclusions from
SEM imaging. Figure 2a shows the morphology of the stainless steel
substrate consisting mainly of large grains, with clearly visible grain
edges. This texture is completely lost after formation of the cobalt
film. The AFM analysis also showed that the black cobalt film
(Fig. 2b) consisted of small nanoclusters, unlike the white cobalt film
Figure 1. SEM images (times 5000) of the stainless steel (SS) surfaces cov-
ered with white (a, top) and black (b, bottom) cobalt. Cobalt was electrode-
posited onto the SS substrate by means of a cathodic step (Ϫ1.1 V vs. SCE
for 15 s) from an aqueous solution containing 1.17 M Co(II), 0.98 M H2SO4,
0.56 M KCl, and 0.2 M H3BO3 in the absence (a) and presence (b) of 0.1 M
KNO3.
Downloaded on 2014-11-12 to IP 128.59.222.12 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 147 (5) 1787-1796 (2000)
1789
S0013-4651(99)05-090-9 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 3. XRD patterns of SS substrates surfaces covered with (a) white
cobalt and (b) black cobalt. The standard patterns are shown from (111) and
(002) ␥-austenite (fine line) and (002) metallic cobalt (bulk line). The lines
used for estimates of the fraction of steel surface coated with cobalt are also
marked: (ooo) ␥-austenite and (ϩϩϩ) cobalt.
is quantitative confirmation of previous morphological observations
that the black cobalt, cluster-like film, is rougher than the deposited
white cobalt. The white cobalt deposit is more uniform, with a higher
fraction of the substrate surface covered, than that of black cobalt.
X-ray diffraction.—We analyzed the chemical composition of
white and black cobalt films using (XRD) techniques. Figure 3
shows the two diffraction patterns obtained for white (Fig. 3a) and
black (Fig. 3b) cobalt film coatings.
These XRD patterns revealed that in both cases the films were
composed of metallic cobalt. When the deposit was formed in the
absence of nitrates (solution A), the film was colored, white-gray,
and in the presence of small quantities of nitrates (solution B), it was
black. Comparison between diffraction peak intensities due to cobalt
(002) and that corresponding to the substrate (steel) allowed estima-
tion of the fraction of the steel surface coated with cobalt. We found
that white cobalt occupied approximately twice the substrate surface
area as black cobalt. Two important features are thus confirmed by
these results. The formation of black cobalt films was indeed ob-
tained directly after Co(II) ion reduction from an aqueous solution
containing small amounts of NOϪ3 , and this black film was com-
posed of dispersed metallic cobalt. This last feature is an essential
characteristic for good photothermal efficiency. To our knowledge
this is the first report of direct black metallic cobalt formation, since
in previous articles11,12 a black cobalt film was reported as forming
indirectly and its composition was related to the formation of cobalt
oxide rather than metallic cobalt.
WDX analysis.—In order to confirm the chemical composition of
the cobalt deposits, WDX measurements were performed. Figure 4
shows a WDX spectrum for black cobalt deposit (an identical WDX
spectrum was obtained for white cobalt deposit). This spectrum
shows lines for cobalt and other elements corresponding to the 304
stainless steel substrate (Fe, Cr, Ni). The small lines (␭ ϭ 23.5 Å)
assigned to oxygen (see insets in Fig. 4) does not allow the estab-
lishment of the presence of oxide compounds in the deposit film.
Therefore, white and black cobalt deposits were solely formed by
metallic cobalt.
Electrochemical Study
Cyclic voltammetry.—Figure 5 shows two typical voltammo-
grams recorded during cobalt electrodeposition onto stainless steel
from solutions (A) and (B). In the cathodic zone (E < Ϫ0.8 V) of
these voltammograms, the Co(II) reduction process produced differ-
ent voltammetric characteristics depending on the chemical compo-
Figure. 2. AFM images of the SS substrate surfaces (a) bare surface, (b) cov-
ered with black cobalt deposit, and (c) with white cobalt deposit. The metal-
lic cobalt deposits were obtained under the same conditions of Fig. 1.
Downloaded on 2014-11-12 to IP 128.59.222.12 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

1790
Journal of The Electrochemical Society, 147 (5) 1787-1796 (2000)
S0013-4651(99)05-090-9 CCC: $7.00 © The Electrochemical Society, Inc.
pletely oxidized in both cases (see Fig. 5) at potential values more
positive than 0.2 V. Figure 8 shows two typical potentiostatic current
transients (I-t curves) recorded during white cobalt (Fig. 8a) and
black cobalt (Fig. 8b) formation. During the first 20 s a cathodic
potential (Ec ϭ Ϫ1.1 V) was imposed to the substrate surface fol-
lowed at t > 20 s by an anodic potential (Ea ϭ 0.6 V). The current-
time evolution associated with each applied potential differed de-
pending on the chemical composition of the electrolytic bath. This
feature indicates that different mechanisms could be associated with
cobalt deposition from the two media.
Integrating each I-t curve derived from this potential program, we
calculated the charges associated with reduction (Qc) and oxidation
(Qa) processes during the potential steps. Figure 9 plots the Qa/Qc
ratio as a function of the cathodic potential step. Once again, cobalt
deposition from the electrolytic bath without nitrates resulted in a
ratio approximately equal to 1 at all applied cathodic potentials (see
Fig. 9a). For black cobalt deposition (solution B), however, the ratio
was lower (Fig. 9b). This agrees completely with the voltammetric
study and supports the hypothesis proposed above for a cobalt-nitrate
interaction. In fact, these observations are similar to those described
Figure 4. Typical WDX spectrum for the black cobalt deposit onto 304 SS
substrate in the wavelength (␭) region of 1.14 Å < ␭ < 4.0 Å. Insets show a
different ␭ range, 5 Å < ␭ < 30 Å.
sition of the electrolytic bath. While the cobalt reduction process
from solution (A) occurred without the formation of a voltammetric
peak (Fig. 5a), for black cobalt deposition, this process developed a
well-resolved, voltammetric reduction peak (IIЈ) (Fig. 5b).
In order to show that peak IIЈ is actually due to cobalt deposition,
in Fig. 6 is shown a set of voltammograms obtained at different
cathodic switching potentials (EϪ␭). Note that in all cases an associ-
ated stripping wave is always present. Moreover, the cathodic peak
current (IIЈ) varies linearly with the v1/2 (v ϭ scan rate potential),
indicating the presence of a mass-transfer process.
Note in Fig. 5 that, during the reverse potential scan (from Ϫ1.5
to 0.6V), a crossover on the cathodic branches was observed in both
cases. This feature indicates that cobalt deposition proceeds via a
nucleation and growth phenomena whether nitrates are present in the
electrolytic bath or not.6,8 When the applied potential reached more
positive values, in both cases the presence of anodic peaks (I and IЈ)
was observed. These peaks are related to oxidized cobalt deposited
during the direct cathodic potential scan. To analyze the cathodic ef-
ficiency of the cobalt deposition, we performed a voltammetric study
with different cathodic switching potentials (EϪ␭), see Fig. 6.
Charges due to the cathodic (Qc) and anodic (Qa) processes can be
obtained from integration of the cathodic and anodic branches of the
I-E curves, respectively. Figure 7 shows plots of the Qa/Qc ratio as a
function of EϪ␭ for cobalt deposition from solutions A and B. In the
case of the nitrate free solution (white cobalt formation), the Qa/Qc
ratio reached a value of 1 as EϪ␭ became more negative. This indicates
that cobalt deposited during the cathodic sweep potential is totally oxi-
dized during the anodic scan. In contrast, the Qa/Qc ratio for black
cobalt formation (solution B) reached a maximum value of 0.5 for the
more negative value of EϪ␭ considered. The difference between the
anodic and cathodic charges (Qc ϭ 2Qa) observed in this case could
be explained in terms of some process coupled to the cathodic reac-
tion. In other words, this evidence supports the simultaneous presence
of another cathodic process (that can consume electrons or recently
deposited cobalt) besides cobalt reduction. The main characteristic of
this second process is that its charge-transfer product cannot undergo
oxidation during the anodic scan. Since the only difference between
the deposition baths is the presence of nitrate, it is likely that an inter-
action between newly deposited cobalt and nitrate ions in solution is
occurring in bath B. This interaction could include a direct redox reac-
tion and/or nitrate reduction on the surface of the cobalt nuclei. The
former reaction would consume the cobalt deposit, and the latter
would provide electrons to the external circuit. Together or separately,
these reactions explain the low cobalt recovery efficiency recorded for
black cobalt formation. The cobalt deposition processes were studied
quantitatively by employing the double potential-step technique.
Potential-step technique.—We analyzed the stability of deposit-
ed metallic cobalt performing a double potential pulse study in both
deposition baths. Several reduction potentials (Er) were considered,
with the inverse anodic pulse constant (Ea ϭ 0.6 V). The reduction
potential range (Ec < Ϫ0.9 V) was chosen based on the cathodic
zone of the corresponding voltammogram. Metallic cobalt is com-
Figure 5. Typical cyclic voltammograms obtained during cobalt deposition
onto the SS electrode from an aqueous solution containing 1.17 M Co(II),
0.98 M H2SO4, 0.56 M KCl, and 0.2 M H3BO3, (a, top) in the absence (white
cobalt) and (b, bottom) in the presence (black cobalt) of 0.1 M KNO3. The
scan rate was 20 mV sϪ1. Peaks related with cobalt deposition (IIЈ) and oxi-
dation (I, IЈ) are also shown.
Downloaded on 2014-11-12 to IP 128.59.222.12 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 147 (5) 1787-1796 (2000)
1791
S0013-4651(99)05-090-9 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 6. Typical voltammograms obtained on a SS substrate in a 1.17 M Co(II) aqueous solution containing 0.98 M H2SO4, 0.56 M KCl, 0.2 M H3BO3, and 0.1
M KNO3 for different cathodic switching potentials (EϪ␭). (a) Ϫ0.95, (b) Ϫ1.0, (c) Ϫ1.05, and (d) Ϫ1.15 V. The scan rate potential was in all cases 20 mV sϪ1
.
by Serruya and Scharifker,15 who reported an electrochemical inter-
action between nitrates in solution and recently deposited thallium on
a vitreous carbon electrode.
Cobalt electrodeposition mechanism.—To confirm the nitrate reac-
tion on recently formed cobalt, we imposed a potential program to
the working electrode, as shown in Fig. 10. A reduction potential
pulse (Er) was initially applied during time, tr. The cell circuit was
subsequently opened in order to relax the diffusion layer during a
waiting time (tw), given by (tw ϭ t Ϫ tr), followed by an anodic
Figure 7. Charge ratio associated to the cathodic (Qc) and anodic (Qa)
voltammetric branches (see Fig, 6), as a function of the cathodic switching
potential (ϪEϪ␭). The charges were evaluated from voltammograms ob-
tained from 1.17 M Co(II) aqueous solution containing 0.98 M H2SO4,
0.56 M KCl, 0.2 M H3BO3, (a) in the absence (white cobalt) and (b) in the
presence (black cobalt) of 0.1 M KNO3.
Figure 8. Typical double-potential step current transient obtained in the SS
electrode from 1.17 M Co(II) aqueous solution containing 0.98 M H2SO4,
0.56 M KCl, and 0.2 M H3BO3, (a) in the absence (white cobalt) and (b) in
the presence (black cobalt) of 0.1 M KNO3. The imposed potentials are: Ec ϭ
Ϫ1.1 V vs. SCE, (t < 20 s) and Ea ϭ 0.6 V vs. SCE (t > 20 s).
Downloaded on 2014-11-12 to IP 128.59.222.12 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

1792
Journal of The Electrochemical Society, 147 (5) 1787-1796 (2000)
S0013-4651(99)05-090-9 CCC: $7.00 © The Electrochemical Society, Inc.
Co2(aϩq) ϩ 2eϪ(steel) ϭ Co/steel
[2]
2Hϩ(aq) ϩ NOϪ3(aq) ϩ Co/steel ϭ Co2(aϩq) ϩ NOϪ2(aq) ϩ H2O(l) [3]
Considering Reaction 3, we propose simultaneous nitrate reduc-
tion during the cathodic process according to the following reaction
2Hϩ(aq) ϩ NOϪ3(aq) ϩ 2eϪ(Co) ϭ NOϪ2(aq) ϩ H2O(l)
[4]
Reactions 3 and 4 were proposed based on the solution pH and
thermodynamic diagrams of predominant species of the nitrate-
nitrite system.16
Electrochemical nucleation of cobalt on stainless steel.—The poten-
tiostatic technique has been demonstrated as a powerful tool for eluci-
dating formation mechanisms of new phases (electrocrystalliza-
tion).17-22 As nitrates play a major role in the formation of cobalt coat-
ings with photothermal properties (black cobalt), analysis of the elec-
trocrystallization mechanism of Co(II) on steel must be well defined
in order to apply such methods reliably. Figures 12 show a current
family of potentiostatic transients obtained during white (Fig. 12a) and
black (Fig. 12b) cobalt coating formation, respectively.
During white cobalt deposition, the recorded current transient (Fig.
12a) developed a current, which increased over time until reaching a
stationary value (plateau). These transients clearly possess the major
characteristic of a three-dimensional (3D) nucleation process limited
by lattice incorporation of adatoms.17 For black cobalt formation (Fig.
12b), the experimental transients differed from those observed for
white cobalt, with the presence of one (lower overpotential) and two
(higher overpotentials) current maxima clearly observed. Formation of
the black cobalt deposit thus involves a different mechanism than
white cobalt formation, and these mechanisms are now described.
Figure 9. Charge ratio associated to the cathodic (Qc) and anodic (Qa) I-t
curves, as a function of the cathodic potential (ϪEr) imposed to the SS elec-
trode during the direct pulse. The charges were evaluated from the double-
potential step current transients (see Fig. 6). The potential imposed in the
reverse pulse was always the same (Ea ϭ 0.6 V). (a) White cobalt and (b)
black cobalt deposition.
potential pulse (Ea) imposed during time, ta (ta ϭ tf Ϫ t) to oxidize
the remaining metallic cobalt on the surface.
This experiment was carried out for different waiting times (tw) in
both deposition baths. For each experiment, we calculated the charge
(Qc) associated with the direct Er potential pulse and the charge (Qa)
due to the applied oxidation potential (Ea). Figure 11 shows the wait-
ing time dependence of the Qa/Qc ratio. In the free nitrate bath, the
Qa/Qc ratio was found to be approximately equal to 1 (Fig. 11a), re-
gardless of the waiting time (tw) value. When nitrate was added to the
cobalt deposition bath, the Qa/Qc ratio was highly dependent on tw,
and as tw increased, the Qa/Qc ratio decreased considerably (Fig. 11b).
These results support the presence of a redox reaction between the
recently formed cobalt deposit and nitrates in solution.
White cobalt formation mechanism.—Experimental potentiostatic
current transients obtained during white cobalt formation (Fig. 12a)
were fit to Eq. 5, proposed by Armstrong et al.17 This equation
describes a 3D nucleation limited by incorporation of adatoms. It is
important to note that this model or a variation of it has been suc-
cessfully applied to the quantitative and qualitative description of the
deposition process of a number of experimental systems23-35
I3Di–Li(t) ϭ P1[1 Ϫ exp(ϪP2t2)]
[5]
where
We propose that the following reactions take place during the
deposition of cobalt onto a stainless steel substrate from both elec-
trolytic baths.
P1 ϭ zFKgЈ
[6]
White cobalt formation, bath A, proceeds via the faradaic Reac-
tion 1
Co2(aϩq) ϩ 2eϪ(steel) ϭ Co/steel
[1]
Black cobalt deposit, bath B, occurs according to the following
faradaic (2 and 4) and chemical Reaction 3
Figure 11. Charge ratio associated to the cathodic (Qc) and anodic (Qa) I-t
curves, as a function of the waiting time (tw). The charges were evaluated
from the current transients obtained with the potential step program shown in
Fig. 8. The program was applied to the SS electrode in 1.17 M Co(II) aque-
ous solution containing 0.98 M H2SO4, 0.56 M KCl, and 0.2 M H3BO3, (a)
in the absence (white cobalt) and (b) in the presence (black cobalt) of 0.1 M
KNO3.
Figure 10. Illustration of the potential step program imposed to the SS elec-
trode, in order to study the stability of the freshly deposited cobalt in the elec-
trolytic baths here considered. The reduction potential (Er) was initially
applied during time tr. The cell circuit was opened during a waiting time,
tw ϭ (t Ϫ tr) followed by an anodic potential pulse (Ea) applied during the
time tf-t.
Downloaded on 2014-11-12 to IP 128.59.222.12 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 147 (5) 1787-1796 (2000)
1793
S0013-4651(99)05-090-9 CCC: $7.00 © The Electrochemical Society, Inc.
␲M2N0Kg2
P ϭ
[7]
2
2
␳
zF is the metallic ion molar charge, KgЈ and Kg are the growth rate
constants, perpendicular and lateral to the surface, respectively. M is
the molar mass, ␳ the atomic density of the deposit, and N0 is the
numeric density of active sites on the substrate surface.
Figure 13 shows a comparison between an experimental current
transient obtained during white cobalt formation and a theoretical
transient generated with Eq. 5. The values of the parameters P1 and
P2, which give the best fit between these curves, were calculated by
nonlinear adjustment of Eq. 5 to the experimental data, as described
elsewhere.21 Theoretical and experimental data observed here are in
satisfactorily accord, so that this case is also applicable to the other
transients of Fig. 12a.
The kinetic parameter values (P1 and P2) obtained using this
adjustment are shown in Table II. Note that, while the P1 values
slightly depend on the applied potential, the P2 values vary ex-
ponentially with it. From Eq. 6 and P1 values, the perpendicular
growth rate constants (KЈg), for each applied potential, are easily
obtained and in this case, they are in the range of (1.71-6.3) ϫ
Figure 13. Comparison of an experimental current transient (•••) obtained
during the white cobalt formation onto SS substrate surface at an imposed
potential of Ϫ0.95 V vs. SCE (see Fig. 12a) and a theoretical current transient
(—) obtained from a nonlinear fitting of Eq. 5 to the experimental data. The
fitting parameters were P1 ϭ 0.033 A cmϪ2 and P2 ϭ 1.12 sϪ2
.
10Ϫ7 mol cmϪ2 sϪ1, which is within the order of magnitude for other
analogous systems.21,23 However, the independent evaluation of N0
and Kg is not possible, see Eq. 7. Considering the P1 and P2 depen-
dence with the applied potential, and assuming that Kg in this system
should have similar values as KgЈ (4 ϫ 10Ϫ7 mol cmϪ2 sϪ1), it is then
possible to evaluate N0 from P2 values. The N0 values estimated by
this way vary exponentially with the applied potential (see Fig. 14).
This exponential relationship between N0 and the applied potential,
has been reported by others researches,18 using independent estima-
tion of N0.
Black cobalt formation mechanism.—It is easy to recognize two
different cobalt deposition behaviors for the black cobalt electrocrys-
tallization process based on the shape of the experimental potentio-
static current transients (Fig. 12b). Those obtained at less cathodic
reduction potentials possess a single current maximum, yet for the
most cathodic applied potentials (E < Ϫ1.05 V vs. SCE), recorded
transients clearly show two current maxima. Unlike white cobalt for-
mation, black cobalt deposition involves current transients with com-
plex shapes, rarely considered in the literature. The complex shape of
these experimental transients could be related to the simultaneous
presence of cobalt and nitrate reduction and cobalt-nitrate interac-
tions (Eq. 2-4). Apparently, cobalt-nitrate interactions are more im-
portant at higher cathodic potential values, as predicted from the
shape of the current transient obtained under these conditions. Al-
though quantitative analysis of the experimental current transients in
Table II. Kinetic parameters of white cobalt deposition on
stainless steel obtained from nonlinear fit of experimental i-t
curves (Fig. 12a) to Eq. 5.
3Di-li
ϪE (V)
P1 (A cmϪ2
)
P2 (sϪ2
)
Figure 12. Typical potentiostatic current transients obtained during the white
(a, top) and black (b, bottom) cobalt eletrodeposition on stainless steel, from
an electrolytic bath containing 1.17 M Co(II), 0.98 M H2SO4, 0.56 M KCl,
and 0.2 M H3BO3 with different KNO3 concentration (a) 0 M and (b) 0.1 M.
The imposed potentials (in volts vs. SCE) are shown in the figures.
0.95
1.00
1.05
1.10
1.15
0.033
0.056
0.059
0.083
0.122
11.12
13.21
10.13
28.36
72.04
Downloaded on 2014-11-12 to IP 128.59.222.12 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

1794
Journal of The Electrochemical Society, 147 (5) 1787-1796 (2000)
S0013-4651(99)05-090-9 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 15. Comparison of the theoretical nondimensional plots (I/Im)2 vs.
t/tm for instantaneous (upper curve) and progressive (lower curve) nucleation
(see Ref. 18) with experimental data (ooo). The experimental transient was
obtained during black cobalt deposition at an imposed potential of Ϫ0.95 V
vs. SCE (see Fig. 12b).
Figure 14. Potential dependence of the numeric density of active sites (N0)
for white cobalt nucleation onto stainless steel (o) evaluated from Eq. 7, the
P2 parameter values reported in Table II and assuming a constant value of
Kg ϭ 4 ϫ 10Ϫ7 mol cmϪ2 sϪ1. The continous line represents the exponential
adjustment of the experimental data.
zFD1/2
c
this case is quite complex, we present a new proposal to describe
quantitatively the potentiostatic current transient due to a nucleation
process, accompanied with a concurrent reaction, at low cathodic
potential values. Analysis of current transients obtained at higher
cathodic potential will be discussed elsewhere.36
P ϭ
3
1/2
␲
[11]
P4 ϭ N0␲kЈD
P5 ϭ A
[12]
[13]
Black cobalt formation at low cathodic potentials.—Taking in to
account the existence of a current maximum in the experimental cur-
rent transients obtained during black cobalt formation at less cathod-
ic potentials and the general form of these transients, we decided to
perform an analysis in the framework of the theoretical formalism
developed by Scharifker et al.18,19 This model describes the 3D
nucleation process limited by a mass-transfer reaction. According to
this approach,18 it is possible to distinguish between two kinds of
nucleation mechanisms. Instantaneous and progressive nucleation
are differentiated by comparing the experimental transient, previ-
ously normalized through the coordinates of the current maximum
(Im and tm), with the theoretical plots due to instantaneous and pro-
gressive nucleation7,8,18 as shown in Fig. 15. From this figure it is
clear that instantaneous nucleation describes a large part of the
experimental current transient, however, a complete adjustment does
not exist, particularly for t > tm.
It is, therefore, necessary to propose a more complete model de-
scribing the whole current transient and taking into account the pro-
cesses described by Eq. 2-4. However, as only the reactions in Eq. 2
and 4 account for electron consumption from the external circuit, we
propose to deconvolute the total current in terms of Eq. 8. This pro-
posed mechanism involve the simultaneous presence of a multiple,
three-dimensional cobalt nucleation growth, limited by diffusion of
cobalt ions (I3D-dc, Eq. 918-20) with nitrate reduction onto cobalt
nuclei surfaces (INR, Eq. 1422). This is a particular case of a most
general model proposed for deconvolution of complex current tran-
sient due to simultaneous faradaic processes8,19-22
and
INR(t) ϭ P6␪
P6 ϭ zNFKN
[14]
with
[15]
In these equations, ␪ is the surface coverage of the formed
nuclei,19 KN corresponds to the nitrate to nitrite electrochemical re-
duction constant, (Eq. 4), zN is the number of electrons involved dur-
ing this reduction, and F is the Faraday constant. This electrochem-
ical reduction is carried out on the recently deposited cobalt nuclei.
D is the cobalt diffusion coefficient, and K is the constant deter-
mined by the experimental conditions.18
The deposition process for black cobalt (at low cathodic poten-
tial) was analyzed using Eq. 8. Kinetic parameters (P3, P4, P5, and
P6) were obtained as a result of the best fit (nonlinear regression) of
Eq. 8 to experimental data (see Table III). Figure 16 compares an
experimental current transient obtained during black cobalt forma-
tion to a theoretical transient generated from Eq. 8. Individual con-
tributions to the total current of the two processes involved are also
presented. The theoretical curves describe the experimental data ade-
quately. From the above analysis of the experimental current tran-
sient for cobalt deposition, it is important to note that the presence
of nitrate in the electrolytic bath changes the limited step for cobalt
electrocrystallization, from charge transfer (in the absence of KNO3)
to mass transfer (in presence of 0.1 M KNO3). Therefore it should
be expected that there is a higher limited current in the diffusion-lim-
ited cathodic current (black cobalt deposition) than in the case of
white cobalt deposition, for the same applied potential; since Co(II)
concentration and electrode surface area are the same in both plating
systems. However, a simple comparison of the experimental current
transients in Fig. 13 and Fig. 16 shows a contrary behavior. This fact
indicates the nitrate in the electrolytic bath could play another role
in the electrolytic process of cobalt. In order to elucidate this matter,
the influence of nitrate concentration on the potentiostatic cobalt
deposition was studied.
I ϭ I3D–dc ϩ INR
[8]
where
with
I3d–dc(t) ϭ (P3tϪ1/2)␪
[9]
1 Ϫ exp(ϪP t)
5
␪ ϭ 1 Ϫ exp ϪP t Ϫ
4
P
5
[10]
Downloaded on 2014-11-12 to IP 128.59.222.12 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 147 (5) 1787-1796 (2000)
1795
S0013-4651(99)05-090-9 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 17. Potentiostatic current transient recorded during cobalt deposition
onto SS substrate from an electrolytic bath containing 1.17 M Co(II), 0.98 M
H2SO4, 0.56 M KCl, and 0.2 M H3BO3 with different KNO3 concentration,
indicated in the figure. All transients were recorded at the same applied
potential (Ϫ0.950 V vs. SCE).
Figure 16. Comparison of an experimental current transient (ooo) obtained
during the black cobalt formation onto the SS substrate surface at an imposed
potential of Ϫ0.95 V vs. SCE (see Fig. 12b) and a theoretical current tran-
sient (—) obtained from a nonlinear fitting of Eq. 8 to the experimental data.
The individual contribution to the total current due to the three-dimensional
nucleation process limited by mass-transfer reaction (I3Di-dc) and current
associated with nitrate reduction (INR) over the fresh deposit are also shown.
The fitting parameters were P3 ϭ 37.14 mA cmϪ2 s1/2 P4 ϭ 8.81 sϪ1, P5 ϭ
both electrocystallization process, for instance, the number density
of active sites (No). This kinetic parameter was evaluated from the
analysis of the current transients.
The current transients obtained in the presence of CKNO3 > 0.1 M
have been analyzed with the same procedure described for the black
cobalt deposit formation as noted above. Figure 18 shows a compar-
ison of experimental and theoretical current transients for cobalt
0.51 sϪ1, and P6 ϭ 9.49 mA cmϪ2
.
Influence of NOϪ3 concentration.—Figure 17 shows a set of exper-
imental current transients recorded during the cobalt deposition onto
stainless steel substrate electrode from an electrolytic bath containing
different KNO3 concentration, at the same applied potential (Ϫ0.95
V). The addition of KNO3 in the electrodeposition bath drastically
changes the shapes of the current transients. For KNO3 concentra-
electrodeposition in presence of CKNO > 0.1 M. The kinetics para-
3
meters obtained for the best fit procedure are shown in Table III. The
kinetics parameters for the cobalt electrodeposition in presence of
CKNO Յ 0.05 M, were obtained following the procedure described
3
tions CKNO Յ 0.05 M, the current transients possess the major fea-
for white cobalt electrodeposition formation as noted above. The
resulting fitting parameters (P1 and P2) are also shown in Table III.
It is possible to establish that the ratio of P1 values obtained for
0.0 and 0.05 M KNO3 can be directly related with KgЈ ratio (see
Eq. 6). Thus the addition of KNO3 up to 0.05 M to the electrodepo-
sition bath provokes a diminution of 2.5 times the value of the per-
3
tures described by the 3D nucleation model limited by lattice incor-
poration of adatoms (see the section on White cobalt formation), yet
for CKNO3 Ն 0.1 M the current transients show the major features for
3D nucleation limited by mass-transfer reaction, coupled with con-
current nitrate reduction (see the section on Black cobalt formation).
Note that for the nitrate concentration range that induces the tran-
sition between lattice incorporation (0.05 M KNO3) to mass-transfer
control (0.1 M KNO3), the current associated to the mass-transfer
controlled electrodeposition process is higher than that associated to
charge-transfer control, as it was expected. It is also important to
note that for the same limited step for cobalt deposition, the increase
of the nitrate concentration provokes a diminution of the associated
current to the electrodeposition process.
pendicular growth rate constant, KЈ. On the other hand, the active
g
sites (associated to P4 constant) (see Eq. 12) diminish with the incre-
ment of KNO3 concentration, causing a diminution of the cobalt
nuclei formed during the electrodeposition; and therefore the cobalt
surface area available for NOϪ3 reduction is depleting when the
nitrate concentration increases (see values of P6 in Table III).
From the appropriate parameters and the correspondent equation
for cobalt electrododeposition mechanism here considered, the
number of active sites (No) were evaluated. Despite of the use of two
different models for this evaluation, the No diminishes exponential-
ly with the increase of nitrate concentration over the whole concen-
Therefore the increase of nitrate concentration in the cobalt elec-
trodeposition bath, besides the fact that it provokes a transition of the
controlled step, should be modified a common kinetic parameter for
Table III. Influence of the NOϪ3 concentration in the cobalt electrodeposition bath, on the kinetic parameters of cobalt deposition on stainless
steel. Kinetic parameters were obtained from nonlinear fit of experimental i-t curves in Fig. 17, to Eq. 5, for [NOϪ3 ] Յ 0.5 M and Eq. 8, for
[NOϪ3 ] > 0.5 M .
3Di-li
3D-dc
NR
[NOϪ3 ]
P1
P2
P3
P4
P5
P6
(mol LϪ1
)
(mA cmϪ2
)
(sϪ2
)
(mA cmϪ2 s1/2
)
(sϪ1
)
(sϪ1
)
(mA cmϪ2
)
0.00
0.05
0.10
0.15
0.20
33.00
12.87
1.121
0.005
37.14
29.18
25.85
8.81
4.02
0.01
0.51
0.90
0.97
9.49e-8
2.71e-8
3.67e-8
Downloaded on 2014-11-12 to IP 128.59.222.12 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

1796
Journal of The Electrochemical Society, 147 (5) 1787-1796 (2000)
S0013-4651(99)05-090-9 CCC: $7.00 © The Electrochemical Society, Inc.
Figure. 18. Comparison of some experimental current transients (oooo)
obtained at the same conditions of Fig. 17, with theoretical (—) current tran-
sients obtained from a nonlinear fitting of Eq. 8 to experimental data. The
kinetic parameters that give the best fit are reported in Table III. The KNO3
concentrations in the electrolytic bath are indicated in the figure.
Figure 19. Variation of the number density of active sites (N0) toward cobalt
deposition onto SS, as a function of the KNO3 concentration in the elec-
trodeposition bath. (o) Experimental results, (—) logarithmic adjustment.
2. E. Barrera. I. González, and T. Viveros, Sol. Energy Mater. Sol. Cells, 51, 69
(1998).
3. I. M. Croll, Advances in X-Ray Analysis, Vol. 4, p.151. Plenum Press, New York
(1980).
tration range considered (see Fig. 19). Thus, we can conclude that,
the nitrate ions can block active sites on the stainless steel substrates
toward cobalt nucleation. This phenomenon could be so important
that the limiting step for cobalt reduction could be modified and the
current associated to cobalt deposition diminished when the nitrate
concentration increases.
4. I. M. Croll and B. A. May, in Electrodeposition Technology, Theory and Practice,
L. T. Romanikiw and D. R. Turner, Editors, PV 87-17 , p. 295, The Electrochemi-
cal Society Proceeding Series, Pennington, NJ (1987).
5. L. Brossared, Mater Chem. Phys., 27, 235 (1991).
6. C. Q. Cui, S. P. Jiang, and A. C. C. Tseung, J. Electrochem. Soc., 138, 1001 (1991)
7. A. B. Soto, E. M. Arce, M. Palomar-Pardavé, and I. González, Electrochim. Acta,
41, 2647 (1996).
Conclusions
8. M. Palomar-Pardavé, I. González, A. B. Soto, and E. M. Arce, J. Electroanal.
Chem., 443, 125 (1998).
The addition of a relatively small quantity of nitrate to a cobalt
deposition bath (Watts type), results in the electrochemical formation
of black cobalt film, instead of white cobalt, which is formed in the
absence of nitrate. XRD studies showed that the film deposited in
both cases, with and without nitrate in solution, consists of metallic
cobalt. Analysis of the deposited cobalt surface by AFM and SEM
techniques showed that the black cobalt deposit is more dispersed
and less smooth than the white cobalt. Therefore, the appropriate
photothermal characteristics of the black cobalt coating are mainly
due to its physical surface features (i.e., roughness, coverage capaci-
ty, form, and deposited grain size) rather than its chemical nature.
Voltammetric and double-potential step studies showed that dif-
ferent electrocrystallization mechanisms are involved in the white
and black cobalt deposition processes. The difference is due to an
electrochemical and/or chemical interaction between the nitrate and
newly formed metallic cobalt. This interaction provokes a more dis-
persed and rougher surface for black cobalt as compared to white
cobalt deposition.
While white cobalt deposition mechanisms were shown to occur
via multiple 3D nucleation, limited by lattice incorporation of cobalt
adatoms to the growth centers; the black cobalt was shown to involve
the simultaneous process of 3D nuclei formation and growth, limit-
ed by mass-transfer reaction, and reduction of nitrates in the medi-
um onto the surfaces of these nuclei. It is shown that besides this
cobalt-nitrate interaction, NOϪ3 ions in solution can block active sites
for cobalt reduction and the effect of this phenomenon strongly
depends on the nitrate concentration.
9. G. E. McDonald, Sol. Energy, 17, 119 (1975).
10. G. B. Smith, A. Ignatiev, and G. Zajac, J. Appl. Phys., 51, 8 (1980).
11. M. G. Hutchins, P. J. Wright, and P. D. Grebenik, Sol. Energy Mater., 16, 113
(1987).
12. R. J. Phillips, T. D. Golden, M. G. Shumsky, and J. A. Switzer, J. Electrochem.
Soc., 141, 2391 (1994).
13. W. U. Schmidt, R. C. Alkire, and A. A. Gewirth, J. Electrochem. Soc., 143, 3122
(1996).
14. D. Aurbach and Y. Cohen, J. Electrochem. Soc., 143, 3525 (1996).
15. A. Serruya and B. R. Scharifker, in Extended abstracts of the Tenth Congreso de la
Sociedad Venezolana de Electroquímica, p. 57 (1997).
16. M. T. Ramírez, Ph.D. Thesis, Universidad Autónoma Metropolitana, México City
(1997).
17. R. D. Armstrong, M. Fleischmann, and H. R. Thirsk, Trans. Faraday Soc., 58, 2200
(1962).
18. B. Scharifker and G. Hills, Electrochim. Acta, 28, 879 (1983).
19. B. R. Scharifker, J. Mostany, M. Palomar-Pardavé, and I. González, J. Electrochem.
Soc., 146, 1005 (1999).
20. M. Palomar Pardavé, Ph.D. Thesis, Universidad Autónoma Metropolitana, México
City (1998).
21. M. Palomar-Pardavé, M. Miranda-Hernández, I. González, and N. Batina, Surf.
Sci., 399, 80 (1998).
22. M. Palomar-Pardavé, B. R. Scharifker, E. M. Arce, and I. González, In preparation.
23. Y. G. Li and A. Lasia, J. Electrochem. Soc., 144, 1979 (1997).
24. U. Schmidt, M. Donten, and J. G. Osteryoung, J. Electrochem. Soc., 144, 2013
(1997).
25. H. C. De Long and R. T. Carlin, J. Electrochem. Soc., 144, 2747 (1995).
26. G. A. Tsirlira, O. A. Petrii, and S. Yu. V. Vassiliev, J. Electroanal. Chem., 414, 41
(1996).
27. I. B. Burrows, J. A. Harrison, and J. Thomson, J. Electroanal. Chem., 58, 241
(1975).
28. C. Kuhnhardt, J. Electroanal. Chem., 369, 71 (1994).
29. R. J. Phillips, T. D. Golden, M. G. Shumsky, and J. A. Switzer, J. Electrochem.
Soc., 141, 2391 (1994).
30. F. E. Varela, M. E. Vela, J. R. Vilche, and A. J. Arvia, Electrochim. Acta, 38, 1513
(1993).
31. M. Y. Abyaneh, J. Hendrikx, W. Visicher, and F. Barendrecht, J. Electrochem. Soc.,
129, 2654 (1982).
32. M. Y. Abyaneh and M. Fleischmann, J. Electroanal. Chem., 119, 197 (1981).
33. E. N. Codaro and J. R. Vilche, Electrochim. Acta, 42, 549 (1997).
34. M. Y. Abyaneh, Electrochim. Acta, 36, 727 (1991).
35. C. A. Gervasi, F. E. Varela, J. R. Vilche, and P. E. Álvarez, Electrochim. Acta, 42,
537 (1997).
Acknowledgment
This work was carried out with the financial support of CONA-
CYT, projects 400200-5-1776PA and 32158E.
Universidad Autónoma Metropolitana-Iztapalapa and Azcapotzalco
assisted in meeting the publication costs of this article.
References
1. F. Trieb, Ph.D. Thesis, DLR Institut fur Technische Termodynamik, Germany
36. E. Barrera, M. Palomar-Pardavé, and I. González, In preparation.
(1995).
Downloaded on 2014-11-12 to IP 128.59.222.12 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).