Electrochimica Acta 113 (2013) 748–754

Contents lists available at ScienceDirect

Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta

The mechanism of anodic dissolution of cobalt in neutral and alkaline

electrolyte at high current density
N. Schubert a , M. Schneider b,∗ , A. Michealis a,b
a
TU Dresden, Inst. f. Werkstoffwissenschaft, Helmholtzstr. 7, 01062 Dresden, Germany
b
Fraunhofer IKTS, Winterbergstr. 28, 01277 Dresden, Germany

a r t i c l e i n f o a b s t r a c t

Article history: The anodic dissolution of cobalt is investigated in neutral and alkaline electrolytes at high current
Received 1 November 2012 densities. Beside electrochemical measurements, the amount of dissolved cobalt is investigated by
Received in revised form 11 June 2013 UV–vis-spectroscopy. The measurements are completed by optical measurements of the metal removing.
Accepted 16 June 2013
It is shown that cobalt can be anodically dissolved in alkaline solutions by using a complexing agent. In
Available online 2 July 2013
this case the current efficiency is approximately 91%. The continuous anodic dissolution of cobalt starts
in the transpassive range.
Keywords:
© 2013 Elsevier Ltd. All rights reserved.
ECM
Anodic dissolution
Cobalt
Electrochemical machining

1. Introduction quality. Therefore, it is necessary to dissolve either the hard phase

Cobalt is the most important binder material for cemented car-
bides (90–94% WC, 6–10% Co), which are commonly used for tools
such as in replaceable cutting inserts [1,2]. Cemented carbides are
characterized by enhanced hardness and wear resistance. Due to
these properties cemented carbides are commonly sintered near-
net-shaped to avoid high tool wear by extensive finishing. Tool
wear is negligible in electrochemical machining (ECM). Therefore,
ECM is a promising option to shape cemented carbides without
tool wear. ECM is known to avoid mechanical stress and thermal
impact to the work piece and provides the opportunity of shaping
even complex forms of materials with a high surface quality, inde-
pendent of the hardness of the material [3,4]. The principle of the
ECM process is shown in Fig. 1. The gap between anode (work piece)
and cathode (tool) is approximately 0.1–1 mm. The current density
amounts typically to 10–300 A/cm2 and the electrolyte flow speed
varies between 5 and 50 m/s [5,6].
Investigating the mechanism of ECM and evaluating the suit-
ability of parameters for machining metal-ceramic-compounds is
the objective of this work. Considering the contrary nature of elec-
trochemical behavior of hard phase (WC) and binder phase (Co), the
electrochemical machining of a composite material such as WCCo
is a challenge. A homogenous dissolution process of both phases is
required to retain the material properties and to get high surface
or the binder phase in the transpassive state. As is well known from
the pH-potential-diagram WC can be dissolved electrochemically
in an alkaline solution and thus forms a passive layer in neutral
and acidic electrolytes. On the other hand, Co forms passive lay-
ers in an alkaline solution and can be machined in the neutral and
acidic electrolyte [7]. Due to the fact that tungsten carbide is the
main component of cemented carbides and forms non reducible ion
conducting oxides in acidic and neutral electrolyte, alkaline elec-
trolytes for ECM would be preferable. The challenge is to enable
dissolution of Co in an alkaline electrolyte. A possible achievement
of this aim could be the usage of a complexing agent for cobalt,
which avoids the passive film formation or the degrading of exist-
ing passive layers. In the 1970s Vu et al. investigated the leaching
behavior of cobalt and cobalt oxides in ammoniacal solution and
calculated a pH-potential-diagram for the system Co–H2 O–NH3
[8,9]. Thus ammonia could be a promising chemical for ECM of
cobalt. A special feature is that ammonia acts as a base itself and
no further chemicals for adjustment of the pH value are necessary.
With this background the authors investigated the electrochemical
behavior of cobalt in an ammoniacal electrolyte and compared the
dissolution mechanism at different pH values. The work is focused
on pH 6 (NaNO3 ), pH 12 (NH3 + NaNO3 ) and pH 14 (NaOH).
2. Experimental

∗ Corresponding author. Tel.: +49 0351 2553 793; fax: +49 0351 2554 108. Electrochemical measurements were carried out with a capil-
E-mail address: Michael.Schneider@ikts.fraunhofer.de (M. Schneider). lary flow-through cell (so called -cell). Such a cell concept was

0013-4686/$ – see front matter © 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.electacta.2013.06.093
 N. Schubert et al. / Electrochimica Acta 113 (2013) 748–754
Fig. 1. Basic concept of electrochemical machining (ECM). State I: (initial state) closed electric circuit of power supply, cathode (tool), electrolyte flow and anode (work
piece),state II: lowering the cathode, specific electrochemical dissolution at the anode, state III: (final state) anode with negative shape of the cathode.
Fig. 2. Photo of the capillary flow-through-cell and enlarged scheme of the framed
section of the capillary.
introduced by A. Moehring at first [10,11]. Fig. 2 shows a photo of
the -cell used in the present work. It is mounted on a force sensor.
The capillary was positioned on the sample and limits the area of
the working electrode to approximately 0.22 mm2 . As the schemat-
ically enlarged section of the capillary in Fig. 2 shows, an inner
glass wall separates two channels inside the capillary. A gap was
prepared between the channels at the capillary tip. By using one
channel as the electrolyte inlet and one as the outlet a contin-
uous electrolyte exchange in front of the working electrode was
guaranteed. A gold wire was positioned in the outlet channel. The
electrode gap was approximately 0.6 mm. A silver/silver chloride
reference electrode was also implemented in the cell. This com-
plete 3-electrode-setup allows measurements with high current
density typical for ECM by using commercial potentiostat systems.
A computer-controlled PGU 20 V–2 A potentiostat (IPS Ing.-Büro
P.Schrems) was used. Samples of a cobalt foil with a purity of 99.95%
(Fa. MaTeck) were used. Table 1 shows typical analytical results of
this cobalt foil [12]. The samples were grinded with SiC up to P1200-
paper and polished with diamond suspension (particle size: 6 ␮m,
3 ␮m and 1 ␮m) and finally SiO2 suspension (0.04 ␮m), followed
by an etching process with a mixture of 100 mL CH3 OH (99.8%) and
6 mL HNO3 (65%).
All electrochemical experiments were carried out in the elec-
trolytes listed in Table 2.
The electrochemical behavior of cobalt was investigated by
cyclic voltammetry using a potential range of −0.8 VSHE to
Table 1
Typical analytical results of a cobalt foil (MaTeck) with a purity of 99.95%.
Element Fe Cu
Mass fraction (ppm) 100 10
749
10.2 VSHE (dE/dt = 500 mV s−1 ). Additionally, linear sweep voltam-
mograms focused on the early state of cobalt dissolution performed
in a limited potential range between −0.8 VSHE and 1.2 VSHE
(dE/dt = 100mV s−1 ).
Further on, the current efficiency of the dissolution process of
cobalt in different electrolytes was estimated. Therefore, current
controlled pulse experiments were conducted (5 pulses, ton = 1 s,
toff = 1 s, ion = 25 A cm−2 ). By using a current controlled procedure an
equal amount of charge was applied evenly in different electrolytes.
This procedure enhances the comparability of the analytical results.
The used electrolyte of each experiment was collected in a beaker.
The authors investigated the amount of dissolved cobalt ions in
this electrolyte by UV–vis-spectroscopy (StellarNet EPP2000) using
Nitroso-R-salt (C10 H5 NNa2 O8 S2 ) as complexing agent, which leads
to the formation of a red cobalt complex in aqueous solutions
[13–15]. The absorption intensity of this complex was evaluated
at 504 nm as recommended by Shipman et al. [13] and Ghasemi
et al. [15]. The standard addition method was used. A linear relation
between absorbance and concentration of the cobalt ion com-
plex was observed in the chosen concentration range of 0.3 to
2.0 mg/L Co. Additionally, the removal of cobalt was optically deter-
mined using a digital 3D-microscope (Keyence VH-Z50L/VHX-500S).
The microscope allows a quantitative geometric measuring of the
removed material volume. Furthermore, the machined samples
were investigated by SEM and EDX with an NVision 40 (Carl Zeiss
SMT GmbH).
3. Results and discussion
Electrochemical behavior in neutral pH range (NaNO3 , pH 6)
Fig. 3a shows the potentiodynamic polarization curve of cobalt
in sodium nitrate electrolyte at pH 6. The potential range was
between −0.8 VSHE and 1.2 VSHE (dE/dt = 100mV s−1 ).
Due to the Pourbaix-diagram of the system cobalt–water an
active dissolution can be expected according to Eq. (1) [7]. A cor-
rosion potential at −0.25 VSHE was observed in NaNO3 as shown in
Fig. 3a. This corresponds to the standard potential of active cobalt
dissolution.
Co → Co2+ + 2e− , E 0 = −0.277 VSHE (1)
The cyclic voltammogram (Fig. 3b) between −0.8 VSHE and
10.2 VSHE exhibits a linear increasing current density of up to
Table 2
Electrolyte characterization.
Electrolyte Ionic concentration Conductivity pH value
mol L−1 mS cm−1
S NaNO3 2.9 159 6
NaNO3 /NH3 2.9/2 158 12
100
NaOH 1 182 14
750 N. Schubert et al. / Electrochimica Acta 113 (2013) 748–754

(a) 1000 (b)

NaNO3
-2
12 NaNO3

-2
Current density j / mA cm

Current density j / A cm
100
10
10
8

1 6

0.1 4

2
0.01

0
-0.5 0.0 0.5 1.0 0 2 4 6 8 10
Potential E vs. SHE / V Potential E vs. SHE / V

Fig. 3. (a) potentiodynamic polarization curve of Co in NaNO3 , E = −0.8 to 1.2VSHE , dE/dt = 100mV s−1 . (b) Cyclic voltammogram of Co in NaNO3 , E = −0.8 to 10.2VSHE ,
dE/dt = 500mV s−1 .

10.2 A cm−2 . The intersection point of the extrapolated curve with
the potential axis is well below the standard potential of oxygen
evolution (Eq. (2)). This confirms the assumption of an active dis-
solution mechanism at high current densities.
6H2 O → 4H3 O+ + O2 + 4e− , E 0 = 1.23 VSHE − 0.0591 V pH
With the reciprocal of the slope and the cell constants
of the electrode surface A and electrode gap d we were able to
calculate the resistance between working and counter electrode
to 16.7  cm. This resistance is slightly higher than the calcu-
lated electrolyte resistance of 6.3  cm (Table 2). This fact might
be caused by the rapid increase of the concentration of reaction
products in front of the electrode. As explained in the literature,
a supersaturated product layer is formed at high current density
[16–18]. The electrolyte flow keeps the product layer very small
but nevertheless the dissolution process at the anode is controlled
by a mass transported through the layer and thus the product layer
causes an additional resistance Rpl . The sum of the electrolyte resis-
tance Rel and product layer resistance Rpl lead to a higher total
resistance R between the electrodes (Eq. (3)).
R = Rel + Rpl
Within the supersaturated layer the concentration of solvent
molecules is very low and the molecules were immediately con-
sumed by the solvation of cobalt ions. Hence, side reactions of
solvent molecules are suppressed.
3.1. Electrochemical behavior in alkaline pH range
3.1.1. Cobalt dissolution in NaOH, pH 14
Fig. 4 shows the electrochemical dissolution behavior of cobalt
in NaOH in comparison with cobalt in NaNO3. As shown in Fig. 4a
(2) Co exhibits active, passive and transpassive behavior in NaOH.
(1  cm2 ) There is a sudden increase of current density beyond a poten-
tial of −0.57 VSHE to a maximum of 3.3 mA cm−2 at −0.43 VSHE ,
which limits the active dissolution range. According to the lit-
erature the oxidation of Co at pH 14 leads to the formation of
CoO or Co(OH)2 (Eqs. (4) and (5)), respectively [7,19–26]. The
oxide (hydr.) passivates the surface of the electrode, which can be
observed in the potential range of −0.43 VSHE to 0.5 VSHE , where,
except for the small peak at 0.2 VSHE , the current decreases to
0.42 mA cm−2 .
Co + 3H2 O → CoO + 2H3 O+ + 2e− , E 0
= 0.095 VSHE − 0.0591 V pH (4)
Co + 4H2 O → Co(OH)2 + 2H3 O+ + 2e− , E 0
(3)
= 0.166 VSHE − 0.0591 V pH (5)
The peak potentials of the higher cobalt oxides are very close and
also dependent on hydration effects. As in the literature described,
it is difficult to assign the peaks to a specific reaction [20,24,27–29].

(a) (b)
1000
-2

NaNO3 12 NaNO3
-2
Current density j / mA cm

Current density j / A cm

100 NaOH NaOH

10
10
8

1 6

0.1 4

2
0.01

0
-0.5 0.0 0.5 1.0 0 2 4 6 8 10
Potential E vs. SHE / V Potential E vs. SHE / V

Fig. 4. (a) Potentiodynamic polarization curve of Co in NaOH (black) and NaNO3 (gray), E = −0.8 to 1.2VSHE , dE/dt = 100mV s−1 . (b) Cyclic voltammogram of Co in NaOH (black)
and NaNO3 (gray), E = −0.8 to 10.2VSHE , dE/dt = 500mV s−1 .
 N. Schubert et al. / Electrochimica Acta 113 (2013) 748–754 751

(a)1000 (b)
NaNO3 12 NaNO3

-2
-2

Current density j / A cm
Current density j / mA cm 100 NH3/NaNO3 NH3/NaNO3
10 NaOH
NaOH
10
8

1 6

0.1 4

2
0.01
0
-0.5 0.0 0.5 1.0 0 2 4 6 8 10
Potential E vs. SHE / V
Potential E vs. SHE / V

Figure 5. (a) Potentiodynamic polarization curve of Co in NH3 /NaNO3 (black), NaOH (gray dotted), NaNO3 (gray), E = −0.8 to 1.2VSHE , dE/dt = 100mV s−1 . (b) Cyclic voltam-
mogram of Co in NH3 /NaNO3 (black), NaOH (gray dotted), NaNO3 (gray), E = −0.8 to 10.2VSHE , dE/dt = 500mV s−1 .

Thus several reactions are possible for the peak at 0.2 VSHE , as listed 3.1.2. Cobalt dissolution in NH3 + NaNO3 , pH 12
in Eqs. (6–8) [7]. To customize electrochemical machining of cobalt in the alka-
line pH range the authors used ammonia, which is known to act as
3CoO + 3H2 O → Co3 O4 + 2H3 O+ + 2e− , E 0 complexing agent for different cobalt species according to Eq. (11):
= 0.993 VSHE − 0.0591 V pH (6)

Co2+/3+ + 6NH3 → [Co(NH3 )6 ]2+/3+ (11)

+ − 0
3Co(OH)2 → Co3 O4 + 2H3 O + 2e , E
= 0.777 VSHE − 0.0591 V pH
Co3 O4 +6H2 O → 3Co(OH)3 + H3 O+ + e− , E 0
= 1.018 VSHE − 0.0591 V pH
At potentials higher than 0.5 VSHE the dissolution shows transpas-
sive behavior. This correlates with the standard potential of oxygen
evolution, which can take place at the passivated surface, due to
the electron conductive nature of the cobalt oxides/hydroxides
(Eq. (2)). Initially, the current slightly increases due to the higher
resistance caused by the electrode surface allocation with oxy-
gen. After passing 0.69 VSHE the current strongly increases and gas
bubbling is observed. In addition to the oxygen evolution, a fur-
ther oxidation of the hydroxide to cobalt(IV)-oxide is discussed
in the literature (Eq. (9)) [7]. Thereby, CoO2 is described as highly
unstable and decomposes water by the degeneration of Co(OH)3 .
Thus an alternating process between the electrochemical reac-
tion in Eq. (9) and the chemical reaction in Eq. (10) is possible,
which would lead to an enhancement of the oxygen formation.
The insertion of Eq. (8) into Eq. (9) results in the well known
formula of oxygen evolution (Eq. (2)). As reported by Göhr [22]
the existence of Co(IV)-oxide is still unproven caused by its low
stability. Whether CoO2 is formed as an intermediate step, oxy-
gen evolution can be assumed as the main reaction at potentials
>0.69 VSHE .
To obtain higher conductivity sodium nitrate was added, which is
a common conducting salt for ECM [3,4].
(7)
The anodic potential sweep of cobalt in NH3 /NaNO3 in the
potential range of −0.8 VSHE to 1.2 VSHE is shown in Fig. 5a (dashed
line). The corrosion potential of active dissolution of cobalt is iden-
tified at −0.34 VSHE. A passive range follows between −0.19 VSHE
(8) and 0.22 VSHE . Subsequently the current density rises and a peak
occurs at 0.59 VSHE . At higher potentials the current rises again.
The slope of the graph is similar to the slope observed on Co in
NaOH (see Fig. 5b). In contrast to the behavior in NaOH, a signifi-
cant dissolution of Co is evident in NH3 /NaNO3 . This result agrees
with the potential-pH diagram of cobalt in ammoniacal solution
(Fig. 6) reported by Vu et al. [8,9].
Co(OH)2 is formed in a first step due to Eq. (5). This fact explains
the passivation of Co shown in Fig. 5a. However, the Co(OH)2 -layer
is affected by ammonia and a soluble Co–NH3 -complex is formed.
This process can be associated with the observed peak at approx-
imately 0.59 VSHE (Fig. 5a). The consequence is the breakdown of
passivity and transpassive dissolution at higher potentials. Finally,
the electrochemical machining of cobalt in alkaline solutions is
possible by using of ammonia.

Co(OH)3 → CoO2 + H3 O+ + e− , E 0 = 1.477 VSHE − 0.0591 V pH (9)

4CoO2 + 6H2 O → 4Co(OH)3 + O2 (10)

A cyclic voltammogram of cobalt in sodium hydroxide in a potential

range of −0.8 VSHE to 10.2 VSHE (Fig. 4b) exhibits a linear increase in
current density. A significant removal of cobalt from the sample was
not observed. This is no surprise according to the aforementioned
explanation. Almost the whole transferred charge is consumed by
the oxygen evolution. Therefore in these investigations the machin- Fig. 6. Potential-pH-diagram of the system Co–NH3 –H2 O, activity of soluble cobalt
ing of cobalt in NaOH is impossible. ion species 10−4 mol L−1 [8].
752 N. Schubert et al. / Electrochimica Acta 113 (2013) 748–754

Fig. 7. Current efficiency of cobalt dissolution in different electrolyte solutions.

3.2. Current efficiency of cobalt dissolution

To compare the current efficiency of the dissolution of cobalt

in the different electrolytes the authors analyzed the removed
cobalt mass by two optical methods (UV–vis-spectroscopy, 3D-
microscopy) and related the determined mass mopt to the
coulometrically determined mass mcoul .

MQ M Idt
mcoul = = (12)
zF zF
Where F is the Faraday constant, M the molar mass of Co, z the
valence (z = 2), Q the consumed charge, I the current and t the sum
of the pulse time. The mass mcoul calculated according to Eq. (12)
represents the theoretical maximum of dissolved Co. The quotient
 of mopt and mcoul (Eq. (13)) corresponds to the current efficiency
of cobalt dissolution.
mopt
= (13)
mcoul
The calculated current efficiency is shown in Fig. 7.
The current efficiency of cobalt dissolution in the neutral elec-
trolyte NaNO3 is approximately 100%. In contrast, the efficiency
is 0% in NaOH. This is obvious because of the passivity of cobalt
in sodium hydroxide. In NaOH the whole transferred charge at
the electrode was consumed by oxygen evolution at the passive
layer. Fortunately the efficiency in the mixed electrolyte ammo-
nia/sodium nitrate is approximately 91%.

3.3. SEM/EDX analysis

Fig. 8 shows the sample surface of machined cobalt in the
different electrolytes at a current density of 25 A cm−2 . All SEM
investigations were completed by EDX analysis of the surface.
Table 3 presents the relation between the OK␣1,2 peak intensity and
the CoL␣1,2 peak intensity.
The surface machined in NaNO3 (Fig. 8a) is very smooth and
seems to be covered by a thin oxide film. The EDX analysis shows
a ratio of Co:O of approximately 4:1 (Table 3). This fact confirms
the aforementioned hypothesis of a very thin cobalt oxide film on
the sample surface. Obviously, the oxide film is formed during the
Table 3
EDX analysis of the machined cobalt foil, relation of CoL␣1,2 /OK␣1,2 .
Electrolyte NaNO3 NaOH
CoL␣1,2 /OK␣1,2 4/1 n.a.
Fig. 8. (a) SEM picture of the surface of a Co foil machined in NaNO3 at 25 A cm−2 .
(b) SEM picture of the surface of a Co foil machined in NaOH at 25 A cm−2 . (c) SEM
picture of the surface of a Co foil machined in NH3 /NaNO3 at 25 A cm−2 .
machining process. However, this film does not result to a dis-
tinct passivation, because the material removal is evidenced. The
dissolution mechanism of cobalt seems comparable to the dissolu-
tion mechanism of copper, which was repeatedly discussed in the
literature [30–34].
Fig. 8b shows the SEM image of the cobalt surface after the elec-
trochemical treatment in NaOH. No dissolution can be detected.
The initial topography of the surface is still visible and only slightly
roughened. No significant oxygen peak can be detected by using
EDX analysis. Hence the signal ratio of oxygen and cobalt cannot
NH3 /NaNO3 be evaluated. Nevertheless, it is no doubt that a thin oxide film
1/1
exists which causes the passivity of cobalt under these conditions
and enhances the electron transfer reaction of oxygen evolution.
 N. Schubert et al. / Electrochimica Acta 113 (2013) 748–754 753

Fig. 9. (a) Scheme of the electrochemical process at a cobalt electrode in NaNO3 at high current density, formation of a viscous layer of reaction products. (b) Scheme of
the electrochemical process at a cobalt electrode in NaOH, left: formation of Co(OH)2 layer, right: oxidation of the divalent to finally trivalent cobalt hydroxide, (c) Scheme
of transpassive cobalt dissolution in NH3 /NaNO3 , left: formation of Co(OH)2 layer in the very beginning of the dissolution process, right: solvation of cobalt hydroxide by
ammonia at the electrolyte hydroxide interface during high rate dissolution, formation of a viscous layer containing cobalt hexamine complex.

However, the information depth of EDX is approximately 1–2 ␮m
and not suitable for surface sensitive analysis.
Fig. 8c shows the surface of a cobalt sample successfully
machined in ammoniacal sodium nitrate. An oxide layer with many
stress cracks (Fig. 8c) can be identified. The EDX analysis shows high
peaks of oxygen as well as of cobalt. The ratio of the OK␣1,2 peak to
the CoL␣1,2 peak is approximately 1:1 (Table 3).
3.4. Reaction mechanism
Regarding the experimental results discussed in section 3 the
authors want to summarize the different dissolution mechanism
of cobalt along the schemes which are presented in Fig. 9.
In sodium nitrate cobalt dissolves in the active state. The cur-
rent efficiency is close to 100%. At high current densities a viscous
and supersaturated product film is formed in front of the electrode
(Fig. 9a) [16–18]. Residues might be found as thin oxide layers by
SEM and EDX respectively. Nevertheless, the film formation does
not inhibit the dissolution of cobalt in sodium nitrate at pH 6.
In contrast to this, cobalt shows passive behavior in sodium
hydroxide at pH 14, as illustrated in Fig. 9b. The authors assume
an initial formation of the divalent cobalt oxide (hydr.), which
is further oxidized to the trivalent species at higher potentials.
Cobalt oxides are electron conducting. Therefore, electron trans-
fer reactions, e.g. the decomposition of the solvent molecules, can
take place at the oxide/electrolyte interface. The most probable
side reaction in sodium hydroxide is oxygen evolution and indeed
bubble formation was observed at the electrode during the electro-
chemical experiments (Fig. 9b right).
Fig. 9c illustrates the dissolution of cobalt in NH3 /NaNO3 in the
transpassive state. In the beginning cobalt forms Co(OH)2 and the
surface becomes passive. This passive layer is strongly affected by
ammonia, which dissolves the oxide at the oxide/electrolyte inter-
face and forms a soluble Co–NH3 -complex (Fig. 9c right). The mass
transport through the product film is limited and side reactions
are suppressed. If ammonia dissolves out cobalt ions of the cobalt
hydroxide layer, cation vacancies will be formed. The vacancies
move toward the interface metal/hydroxide. At this interface Co
atoms oxidize and combine with the cation vacancies. Hence the
hydroxide layer is present during the whole machining process.
After ECM the cobalt hydroxide layer remains on the metal surface
as shown in the SEM image (Fig. 8c).
4. Conclusions
The dissolution mechanism of cobalt in neutral and alka-
line electrolytes was investigated in a special flow cell by cyclic
voltammetry under a low and high anodic current density. The elec-
trochemical measurements were confirmed by different analytical
investigations such as UV–vis-spectroscopy and SEM in combina-
tion with EDX. With respect to the results of our measurements we
assumed three different dissolution mechanisms:
In a neutral NaNO3 solution the dissolution takes place in an
active state with approximately 100% current efficiency even at a
754 N. Schubert et al. / Electrochimica Acta 113 (2013) 748–754
higher current density. After machining in a neutral NaNO3 solution
the surface is smooth and covered by a thin oxide film.
In a NaOH solution cobalt passivates with an electron con-
ducting oxide layer and oxygen evolution take place. No metal
removal at the electrode could be observed and no cobalt could
be analyzed in the waste electrolyte. The current efficiency
is 0%.
In an alkaline electrolyte containing NH3 and NaNO3 cobalt
hydroxide is built, but due to the fact that ammonia acts as a com-
plex agent for cobalt the hydroxide can be dissolved, which leads
to a transpassive dissolution mechanism. Therefore, cobalt can be
machined in the electrolyte mixture NH3 /NaNO3 with a current
efficiency of 91%. The surface after machining is covered by a thin
oxide film.
Acknowledgements
The authors wish to thank the German Research Foundation
(DFG) for the financial funds of this project (grant numbers MI
509/16-1 and SCHN 745/11-1).
References
[1] V. Böhlke, Hartmetall-ein moderner Hochleistungswerkstoff, Materialwis-
senschaft und Werkstofftechnik vol 33 (10) (2002) 575–580.
[2] W. Schedler, Hartmetall für den Praktiker, VDI-Verl. (1988).
[3] V. Degner, Elektrochemische Metallbearbeitung, Technik, Berlin, 1984.
[4] E. Rumyantsev, A. Davydov, Electrochemical machining of metals, Mir Publish-
ers, 1989.
[5] D. Landolt, Mechanistische gesichtspunkte der elektrochemischen metallbear-
beitung, Chemie -Ingenieur -Technik 45 (1973) 188–194.
[6] W. Mauz, Technik der elektrochemischen metallbearbeitung, Chemie -
Ingenieur -Technik 45 (1973) 194–197.
[7] M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, NACE
International, 1974.
[8] C. Vu, K. Han, Leaching beheviour of cobalt in ammonia solutions, Institution
of mining and metallurgy (1977) C119–C125.
[9] C. Vu, K. Han, F. Lawson, Leaching behaviour of cobaltous and cobalto-
cobalticoxides in ammonia and in acid solution, Hydrometallurgy 6 (1980)
75–87.
[10] A. Moehring, Entwicklung einer elektrochemischen mikrodurchflusszelle zur
untersuchung des elektrochemischen senkens (ECM, electrochemical machin-
ing) dissertation thesis, Heinrich-Heine-Universität Düsseldorf Germany
(2004).
[11] M. Lohrengel, C. Rosenkranz, I. Klüppel, A. Moehring, H. Bettermann, B. Van
den Bossche, J. Deconinck, A new microcell or microreactor for material sur-
face investigations at large current densities, Electrochimica Acta 49 (2004)
2863–2870.
[12] MaTeck, Product datasheet cobalt foil 99.9%, 2011.
[13] W.H. Shipman, J.R. Lai, General photometric microdetermination of cobalt with
nitroso-R salt, Analytical Chemistry 28 (1956) 1151.
[14] G. Wünsch, Photometric determination of cobalt with nitroso-R-salt, Talanta
26 (1979) 177.
[15] J. Ghasemi, N. Shahabadi, H.R. Seraji, Spectrophotometric simultaneous deter-
mination of cobalt, copper and nickel using nitroso-R-salt in alloys by partial
least squares, Analytica Chimica Acta 510 (2004) 121.
[16] P. Jacquet, Electrolytic method for obtaining bright copper surfaces, Nature 135
(1935) 1076.
[17] P. Jacquet, On the anodic behavior of copper in aqueous solutions of orthophos-
phoric acid, Transactions of the Electrochemical Society 69 (1936) 639.
[18] P. Jacquet, The electrolytic polishing of metallic surfaces, Part II, Metal Finishing
47 (1949) 48.
[19] W. Behl, J.E. Toni, Anodic oxidation of cobalt in potassium hydroxide elec-
trolytes, Journal of Electroanalytical Chemistry 31 (1971) 63–75.
[20] L. Burke, M. Lyons, O. Murphy, Formation of hydrous oxide films on coalt under
potential cycling conditions, Journal of Electroanalytical Chemistry 132 (1982)
247–261.
[21] R. Cowling, A. Riddiford, The anodic behaviour of cobalt in alkaline solutions,
Electrochimica Acta 14 (1969) 981–989.
[22] H. Göhr, Über anodisch gebildete oxidische deckschichten auf kobalt in wäss-
riger lösung-i. zur thermodynamik des systems kobalt- wässrige lösung,
Electrochimica Acta 11 (1966) 827–834.
[23] H. Jung, A. Alfantazi, Corrosion properties of electrodeposited cobalt in sulfate
solutions containing chloride ions, Electrochimica Acta 55 (2010) 865–869.
[24] T. Ohtsuka, N. Sato, Anodic oxide film on cobalt in weakly alkaline solution,
Journal of Electroanalytical Chemistry 147 (1983) 167–179.
[25] S.E. Wakkad, A. Hickling, The anodic behaviour of metals. Part VI–Cobalt, Trans-
actions of the Faraday Society 46 (1950) 820.
[26] T. Jayaraman, V. Venkatesan, H. Udupa, Cyclic voltammetric studies of electro-
less cobalt in NaOH, Electrochimica Acta 20 (1975) 209–213.
[27] A. Bewick, C. Gutiérrez, G. Larramona, An in-situ IR spectroscopic study of the
anodic oxide film on cobalt in alkaline solutions, Journal of Electroanalytical
Chemistry 333 (1992) 165–175.
[28] G. Larramona, C. Guitiérrez, An in-situ spectroscopic study of the anodic oxide
film on a cobalt electrode at pH 4-14 by potential-modulated reflectance, Jour-
nal of Electroanalytical Chemistry 293 (1990) 237–252.
[29] S.S. Abd El Rehim, A.E. Basosi, M. Osman, The influence of halide ions on the cor-
rosion and passivity of cobalt in an alkaline medium, Journal of Electroanalytical
Chemistry 348 (1993) 99–106.
[30] M. Schneider, S. Schroth, S. Richter, S. Höhn, N. Schubert, A. Michaelis, In-
situ investigation of the interplay between microstructure and anodic copper
dissolution under near-ECM conditions–Part 2: the transpassive state, Elec-
trochimica Acta 70 (2012) 76.
[31] M. Schneider, S. Schroth, N. Schubert, A. Michaelis, In-situ investigation of the
surface-topography during anodic dissolution of copper under near-ECM con-
ditions, Materials and Corrosion 63 (2012) 96.
[32] M. Schneider, S. Schroth, S. Richter, S. Höhn, N. Schubert, A. Michaelis, In-situ
investigation of the interplay between microstructure and anodic copper dis-
solution under near-ECM conditions–Part 1: the active state, Electrochimica
Acta 56 (2011) 7628.
[33] D. Landolt, in: R.P. Frankenthal, J. Kruger (Eds.), Transpassivity in Passivity of
Metals, Electrochemical Society Inc., Princeton, NJ, 1978, p. 484.
[34] K. Kinoshita, D. Landolt, R.H. Müller, C.W. Tobias, Stoichiometry of anodic cop-
per dissolution at high current densities, Journal of the Electrochemical Society
117 (1970) 1246.