Cr(III) oxidation with lead dioxide-based anodes

Article abstract of DOI:10.1016/j.electacta.2003.07.005

A procedure for preparing PbO₂-based electrodes with a titanium substrate is proposed. A platinum underlayer is first deposited on Ti by metal organic chemical vapor deposition (MOCVD), followed by the electrodeposition of the PbO₂ layer. The prepared Ti/Pt/PbO₂ anodes were examined by scanning electron microscopy (SEM) and X-ray diffraction (XRD) before being used for oxidation of Cr(III) in sulphuric acid. The current efficiency was determined for that electrodes and the results were compared with those obtained with Pb/PbO₂ and Ebonex/PbO ₂ electrodes with different pH conditions. The Ti/Pt/PbO₂ were found to have a very good electrochemical behaviour (current efficiency: φ=0.93 for pH 2), and may be used as dimensionally stable anodes for the oxidation of Cr(III).

Full text of DOI:10.1016/j.electacta.2003.07.005

Electrochimica Acta 48 (2003) 4301–4309
Cr(III) oxidation with lead dioxide-based anodes
D. Devilliersa,∗,1, M.T. Dinh Thia,b, E. Mahéa, Q. Le Xuanb
a
LI2C Laboratory, Université Pierre & Marie Curie, CNRS UMR 7612, 4 Place Jussieu, 75252 Paris Cedex 05, France
b
Institut de Technologie Tropicale, CNST, VN, 18, Route Hoang Quoc Viet, Hanoi, Viet Nam
Received 15 July 2003; received in revised form 29 July 2003; accepted 29 July 2003
Abstract
A procedure for preparing PbO2-based electrodes with a titanium substrate is proposed. A platinum underlayer is first deposited on Ti
by metal organic chemical vapor deposition (MOCVD), followed by the electrodeposition of the PbO2 layer. The prepared Ti/Pt/PbO2
anodes were examined by scanning electron microscopy (SEM) and X-ray diffraction (XRD) before being used for oxidation of Cr(III) in
sulphuric acid. The current efficiency was determined for that electrodes and the results were compared with those obtained with Pb/PbO2 and
Ebonex®/PbO2 electrodes with different pH conditions. The Ti/Pt/PbO2 were found to have a very good electrochemical behaviour (current
efficiency: φ = 0.93 for pH 2), and may be used as dimensionally stable anodes for the oxidation of Cr(III).
© 2003 Elsevier Ltd. All rights reserved.
Keywords: Lead dioxide electrode; Titanium-based electrode; Chromate; Electrodeposition
1. Introduction
The electrodeposition of PbO2 has been studied on a va-
riety of substrates: platinum [7–10], gold [11–13], graphite,
glassy carbon [14], tin oxide [15], titanium [12,16,17], ce-
ramic materials composed of Magnéli phase titanium ox-
ides Ti4O7 and Ti5O9 called Ebonex® [18,19], stainless steel
[17], and lead or a lead alloy [1].
Soluble Cr(VI) species are useful in many sectors of in-
dustry [1]. For example, Cr(VI) is a redox mediator in the
preparation of anthraquinone from anthracene, but it is also
used for electropolishing or pickling (metal finishing). The
waste baths often contain sulphuric acid, in the concentra-
tion range 0.005–3 mol l−1 [1]. Their disposal is problemat-
ical for environment and the discharge of Cr species in efflu-
ents must be minimised. Hence, it is preferable to regenerate
Cr(VI) from those solutions by electrolysis treatment [2].
Cr(III) oxidation has already been studied during the last
30 years; work in the field has been the object of review
articles [3,4]. The anodic material used for the oxidation
of Cr(III) into Cr(VI) species is often lead dioxide, which
can be electrodeposited on different substrates. The prepa-
ration and properties of such anodes have been reviewed
[5,6]. PbO2 may be described as a non-stoichiometric com-
pound with changing stoichiometry [6]. The main interests
of PbO2 anodes are: low price compared to noble metals,
chemical stability in corrosive media, and high value of the
overpotential for the oxygen evolution reaction [6].
Lead substrates usually have poor performances. The
electrochemical deposition of PbO2 on Ebonex® substrates
leads to good adherence coatings, but it is difficult to obtain
cheap large electrodes. Precious metals are too expensive.
Although graphite substrates are still in use in the industry,
titanium-based electrodes must be considered because they
could constitute “dimensionally stable anodes” and allow
thinner and mechanically superior cell designs. However, it
is difficult to prepare adherent PbO2 coatings on titanium.
It is the reason why it is often necessary to deposit an inter-
mediate layer between the Ti substrate and the electroactive
PbO2. In addition, it is necessary to avoid corrosion and
passivation of titanium if the electrodes are to be used as
anodes in acidic media. Pre-treatment of the substrate is
very important for avoiding the formation of a TiO2 pas-
sive layer. The intermediate phase must be non porous and
conductive.
Conducting undercoatings which inhibit titanium oxide
formation (carbides or borides of Ta or Ti) have been pro-
posed [20]. Other authors recommend the deposition of
a metal, for example a 10 ␮m thick tin layer [17]. Ueda
∗
Corresponding author. Tel.: +33-1-44-27-36-77;
fax: +33-1-44-27-38-56.
E-mail address: devill@ccr.jussieu.fr (D. Devilliers).
1
ISE member.
0013-4686/$ – see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2003.07.005

4302
D. Devilliers et al. / Electrochimica Acta 48 (2003) 4301–4309
et al. [16] have proposed a conductive undercoating con-
sisting of Ti–Ta oxides prepared by several applications of
decomposable compounds of Ti and Ta [16] followed by
heat-treatment. Its conductivity is enhanced by several appli-
cations of a solution containing a thermally decomposable
compound of platinum followed by subsequent heat treat-
ments.
Electrochemical deposition of PbO2 often occurs in acidic
medium. In that conditions, tetragonal ␤-PbO2 is preferably
formed, so that the term “PbO2” used in applied electro-
chemistry usually refers to the ␤-form [16]. However, the
coating prepared in acidic medium is not sufficiently adher-
ent to the Ti substrate, except if additives such as sodium
lauryl sulphate are added to the electrolyte [21]. The second
common variety of lead dioxide, i.e. orthorhombic ␣-PbO2,
can be deposited from an alkaline solution (for example,
sodium acetate, lead acetate and KOH). Too alkaline solu-
tions give rise to the formation of red Pb3O4 crystals on the
electrodes [22]. ␣-PbO2 has a more compact structure com-
pared to the more porous ␤-PbO2 [23], resulting in a better
contact between the particles.
In this paper, we report a simplified procedure for prepar-
ing a Ti/Pt/PbO2 electrode, in which the underlayer in
contact with Ti and PbO2 is a compact film of platinum,
deposited by metal organic chemical vapor deposition
(MOCVD). The prepared anodes were used for oxidation
of Cr(III) in sulphuric acid. The results were compared with
those obtained with Pb/PbO2 and Ebonex®/PbO2 electrodes.
These experiments were realised in a two-compartment cell
containing an anionic membrane as a separator the current
yield was determined for the three electrode substrates with
different pH conditions.
2. Experimental
2.1. Preparation of the working electrodes
All chemicals were Prolabo Normapur grade, except
when specified hereafter. Titanium foils (Goodfellow; pu-
rity: 99.6%; area: 4 cm2) were carefully polished on abra-
sive papers Buehler-Met 600, 800 and 1200, and rinsed
with methanol in an ultrasonic cleaner.
When the procedure gives rise to a mixture of the two
crystallographic varieties, the quantitative determination of
each phase is not easy but one can compare, in the X-ray
diffractograms obtained with different samples, the ratio of
the peak areas ␣/(␣+␤) and ␤/(␣+␤) [23]. Other authors
[10,24] use the procedure of Dodson [25], based on calibra-
tion spectra obtained with pure ␣ and ␤ varieties. However,
the results cannot be very accurate since electrodeposited
PbO2 is always textured and the preferred crystallographic
orientations that are observed depend on several parame-
ters: temperature, potential, current density. Nevertheless,
in Pb(NO3)2–HNO3 electroplating solutions, the amount of
␣-phase decreases with temperature and, at 65 ◦C, only the
␤-phase was detected on a Pt substrate [10], with a preferen-
tial orientation of the planes: (1 1 1). An increase of tempera-
ture is reported to favour the formation of larger crystals [10].
In order to increase the electrocatalytic activity of PbO2
towards various anodic reactions, incorporation of some ions
into the oxide layer film during its formation was examined
in the literature [23,26].
Ueda et al. [16] have proposed a galvanostatic procedure
in which a stress-free intermediate ␣-PbO2 coating is pro-
duced by electrodeposition from an alkaline lead bath. It
plays the role of binder on the top of which ␤-PbO2 can be
electrodeposited from an acid lead bath containing a suspen-
sion of fine Ta2O5 particles. These particles are incorporated
into the ␤-PbO2 layer in order to result in a coating with
low internal stress. That procedure led to an anode material
which is reported to be superior to Pb/PbO2 or Ti/Pt with
respect to oxygen overvoltage [16]. However, the prepara-
tion mode is long, especially the formation of the undercoat-
ing; in addition, its conductivity is enhanced by a platinum
deposit but its compactness is probably not sufficient for
avoiding completely the corrosion of the Ti substrate that
may occur in acidic media.
First, a thin and compact platinum film was deposited on
one face of the foil by MOCVD. This method is based on
the oxidative decomposition of (methylcyclopentadienyl)
trimethyl platinum(IV) (Strem; purity 99%) as a precursor,
diluted in cyclohexane solvent [27]. A Trijet liquid delivery
system has been used to supply and vaporize the precursor
solution (C = 0.02 M) into the reactor. Helium carrier gas
and oxygen reactant gas were used at a pressure of 1330 Pa.
The other experimental conditions for the deposition of
platinum films were: temperature of the evaporator: 48 ◦C;
platinum deposition temperature Θ = 350 ◦C; deposition
time: τ = 4000 s. With such conditions, the thickness of
the film is about 16 nm and platinum oxide is not formed:
ESCA experiments have proved that the atomic percentage
of oxygen is less than 0.1% with those experimental condi-
tions of deposition [27]. That platinum layer was prepared
for avoiding passivation or corrosion of the Ti substrate
during all the subsequent experiments. Thus, the back
side of the foils was insulated by application of an epoxy
resin.
Second, an undercoating of ␣-PbO2 was electrodeposited
from an alkaline lead bath (called S1 hereafter), as reported
by other authors [16]: 3.5 M NaOH saturated with litharge
PbO (pure; Touzart & Matignon) at Θ = 40 ◦C. The pH of
S1 is >14 and the soluble Pb(II) species are HPbO2− anions.
The cyclic voltammogram performed at v = 50 mV s−1 is
presented in Fig. 1a. It shows that the PbO2 deposition be-
gins at potentials higher than 0.25 V/ECS. Thus, for our po-
tentiostatic experiments, a constant potential (0.35 V versus
SCE) was applied for 2 h. Those potentiostatic conditions
led to homogeneous lead dioxide deposits.
Thus, the top coating (␤-PbO2) was electrodeposited from
an acid lead bath [5], called S2, and containing 30 wt.%
Pb(NO3)2 (pH 2). That bath, kept at Θ = 65 ◦C, was

D. Devilliers et al. / Electrochimica Acta 48 (2003) 4301–4309
4303
10
5
cause it is well-known that such a layer is formed when lead
electrodes are used as anodes.
The PbO2 layers were characterized by scanning electron
microscopy (SEM, Leica Stereoscan 440). In addition, X-ray
diffraction (XRD) spectra were obtained with a Philips 1380
diffractometer. The X-ray tube was operated at 30 kV with
a molybdenum target. The wavelength of the K␣1 emission
line is: λ = 0.70930 Å.
0
-5
-10
-15
2.2. Electrolysis of Cr(III) solutions
-0.4
0
0.4
0.8
(a)
E(V/SCE)
The oxidation of Cr(III) was carried out in a two-
compartment cell, under galvanostatic conditions (j =
10 mA cm−2), at 25 ◦C during 24 h, using an EG&G
273A potentiostat/galvanostat. The anodic compartment
(V = 0.2 L) was containing the PbO2 working electrode
dipped in a 0.1 M Cr(III) solution acidified by H2SO4 at
several concentrations (0.5, 0.05 or 0.005 M). Cr(III) was
introduced as KCr(SO4)2 Merck pro analysi. Before the
macro-electrolysis started, the limiting current density for
Cr(III) oxidation was determined, taking into account the
mass transfer conditions in the anodic compartment (con-
10
8
6
4
2
0
-2
-4
-6
-8
0
0.5
1
1.5
2
2.5
(b)
E(V/SCE)
tinuous stirring at constant speed). Its value was: jL
=
Fig. 1. (a) Cyclic voltammogram performed at v = 50 mV s−1 with the
alkaline solution S1. Working electrode: Ti/Pt; S = 4 cm2, (b) cyclic
voltammogram performed at v = 50 mV s−1 with the acid solution S2.
Working electrode: Ti/Pt; S = 4 cm2.
25 mA cm−2. The charge was calculated so that less than
60% of the initial Cr(III) species could be oxidized.
The cathodic compartment was containing a graphite
counter-electrode dipped in sulphuric acid.
For the study of the overall current efficiency, the Cr(VI)
concentration was determined at the end of the 24 h experi-
ment. An anion-permeable membrane Asahi Chemical A10
was used as a separator.
continuously agitated and was containing a suspension of
Ta2O5 powder (5 g/l; Aldrich; purity: 99%) in order to pre-
pare a coating with low internal stress. The cyclic voltammo-
gram performed at v = 50 mV s−1 is presented in Fig. 1b. It
shows that the PbO2 deposition begins at potentials higher
than 1.7 V/SCE. Thus, for our potentiostatic experiments, a
constant potential (1.8 V versus SCE) was applied for 3 h.
For such temperatures of the electroplating solution (65 ◦C),
the lifetime of the electrodes is reported to be higher than
that of samples obtained at room temperature [10].
The determination of Cr(VI) concentrations can be made
by titration with Mohr’s salt, with phenylanthranilic acid
as indicator [28]. It is also possible to plot the reduction
wave of Cr(VI) and to measure the limiting diffusion cur-
rent intensity, iLred, obtained with a rotating disk electrode
(RDE). However, platinum electrodes are not recommended
because acidic Cr(VI) solutions attack the metal, and the
oxide film formed hinders the electrode reaction. On the
contrary, gold electrodes are suitable [29]. Thus, we have
chosen to determine the Cr(VI) concentration in the anodic
compartment with a Tacussel RDE fitted with a gold disk
(diameter: 2 mm) and operated at a 3000 rpm rotation speed.
All the experiments were performed after 10 min argon bub-
bling for removing dissolved oxygen. A small amount of
anolyte (1 cm3) was removed and diluted in 40 cm3 of 0.5 M
sulphuric acid. The value of the limiting current intensity
due to the reduction of Cr(VI), iL, was determined and the
concentration of Cr(VI) was deduced from the calibration
curve iL = f(C). That curve was plotted previously, using
2.5 × 10−4 – 2 × 10−3 M standard solutions of Cr(VI) in
0.5 M sulphuric acid.
The loading of the undercoating (␣-PbO2) and top coating
(␤-PbO2) was estimated to be 31 and 13 ␮m, respectively.
These results were obtained by the following technique: af-
ter the PbO2 layer was formed, the electrode was immersed
in hot concentrated nitric acid (8 M); the dissolution of
PbO2 occurred and led to soluble Pb(II). Its concentration
was determined by square wave voltammetry, using a mer-
cury electrode. Thus, the thickness of the PbO2 layers was
evaluated, knowing its density (ρ = 9.375 g cm−3).
The loading of Ta2O5 mechanically incorporated into the
PbO2 layer is not known. Other authors using the same
procedure do not give that information [16].
The lead foils were provided by Weber Métaux, Paris,
and the Ebonex® ceramics (non porous grade) by Atraverda
Ltd. For the Ebonex® substrate, no platinum film was neces-
sary; the PbO2 film was directly deposited according to the
procedure reported above. On the lead substrate, no PbO2
layer was electrodeposited from the S1 or S2 solutions, be-
The instantaneous current efficiency was also determined
in a shorter time range (12.5 h). For that experiments, the
two-compartment cell was used, fitted with an anionic
membrane Ionac MA-3475. The anolyte was composed

4304
D. Devilliers et al. / Electrochimica Acta 48 (2003) 4301–4309
of KCr(SO4)2 0.1 M + H2SO4 (0.05 M) and the catholyte
was H2SO4 0.05 M. 15 ml of anolyte were periodically
removed from the electrolysis cell and transferred into
another cell fitted with a rotating gold electrode, an SCE
reference electrode and a platinum counter electrode. After
the measurement of the concentration of Cr(VI) from the
limiting reduction current intensity, the 15 ml of solution
were reintroduced into the electrolysis cell.
tentiostatic conditions described above. The corresponding
sample, called “Sample 1” hereafter, was observed by SEM.
It seems very compact (see Fig. 3). The pyramidal form of
the ␣-PbO2 crystallites cannot be easily distinguished. Their
size is about 1 ␮m. The same morphology was observed for
a PbO2 deposit on Ebonex®.
After, another sample, called “Sample 2” was prepared
by electrodeposition of PbO2 on a Ti/Pt substate from the
acid S2 solution. The resulting deposit is compact. Two
kinds of crystallites can be distinguished on the surface:
small ␤-PbO2 crystals (size < 0.5 ␮m) and also large
pyramidal-form ␣-PbO2 crystals (size: 3 ␮m), as shown in
Fig. 4. The ␣-PbO2 crystals formed in the S2 solution are
about three times larger than those formed in the S1 solution.
On several places of the surface, Ta2O5 particles are em-
bedded into the PbO2 layer. Such a particle is visible in
Fig. 5; its size is about 7 ␮m.
Other authors claim that the ␣ variety was obtained using
the S1 solution, and the ␤ variety using the S2 solution [16].
However, they did not give any structural analysis data, and
their procedure was different: they performed galvanostatic
electrodeposition, whereas we worked under potentiostatic
conditions.
In order to confirm the presence of the two varieties of
PbO2 on the electrodes, XRD experiments were performed
with Samples 1 and 2. The corresponding X-ray spectra are
presented in Figs. 6 and 7, respectively.
3. Results and discussion
3.1. Structural analysis and morphology of the PbO2
deposits
The aim of these investigations was first to verify the
nature of the PbO2 deposits (␣ or ␤ variety) and also to
determine the size of the crystallites.
The titanium substrate covered by the platinum film
deposited by MOCVD was first studied by SEM. The mi-
crography presented in Fig. 2 shows the morphology of the
surface. The stripes resulting from the mechanical polishing
of the titanium substrate are clearly visible, but it is not
necessary to have a perfectly flat substrate for depositing
adherent PbO2 on it. The thickness of the Pt deposit was es-
timated by energy dispersion spectrometry. Its value (about
20 nm) is in agreement with the expected one (16 nm, see
Section 2).
The important background is due to the absorption of the
radiation by the sample containing lead, and emission of the
absorbed energy as fluorescent radiation in all directions.
Electrodeposition of a PbO2 coating on a Ti/Pt substrate
from an alkaline solution (S1) was performed under the po-
Fig. 2. Platinum deposit on a titanium substrate (bar scale = 2 ␮m).

D. Devilliers et al. / Electrochimica Acta 48 (2003) 4301–4309
4305
Fig. 3. PbO2 electrodeposited on Ti/Pt from the alkaline solution S1 (bar scale = 1 ␮m).
Attribution of the peaks was performed by using JCPDS
data. The results are reported in Tables 1 and 2. It can be
concluded that Sample 1 is mainly composed of ␣-PbO2,
but all the characteristic peaks of ␣-PbO2 variety are not
visible. The (2 0 0) line has the highest intensity. ␤-PbO2
may also be present (the peaks at 2θ = 16.46 and 33.28◦ can
be attributed to both varieties) but only very small amounts
may be present, otherwise other characteristics peaks of the
␤-variety should be observed. The peak at 26.79◦ may be
attributed only to the ␣ variety.
On the contrary, for Sample 2, the ␤-PbO2 variety is
mainly present, and the preferred orientations planes are
(0 1 1) and (1 0 1). The ␣-variety is also clearly identified,
especially from the peak at 2θ = 13.08◦ corresponding to
the (1 1 1) plane.
Our results differ to a certain extent from other results
published in the literature. The amount of ␣-PbO2 in Sam-
ple 2, prepared at 65 ◦C, is less important that in other PbO2
layers deposited on titanium from a Pb(NO3)2–HNO3 so-
lution by Devos [30,31], working at ambient temperature.
Fig. 4. PbO2 electrodeposited on Ti/Pt/PbO2 from the acid solution S2 (bar scale = 1 ␮m).

4306
D. Devilliers et al. / Electrochimica Acta 48 (2003) 4301–4309
Fig. 5. Ta2O5 crystals embedded into the PbO2 layer electrodeposited from the S2 solution (bar scale = 1 ␮m).
However, Velichenko et al. [10] did not observe the ␣ vari-
The oxidation of Cr(III) competes with the oxidation of
water. Thus, it was necessary to determine the current effi-
ciency, φ, of the Cr(VI) production for different experimen-
tal conditions (pH 0, 1 or 2 with the three substrates).
As described in the Section 2, the calibration curve iL =
f (CrVI) was plotted, using standard solutions of Cr(VI) in
0.5 M sulphuric acid.
The determination of the amount of Cr(VI) formed in
the electrolysis cell was performed at the end of the 24 h
experiments. A small amount of anolyte was removed
from the cell and diluted in 0.5 M sulphuric acid. In such
conditions, both standard solutions and solutions with un-
known Cr(VI) concentration had similar pH value and ionic
strength.
ety in PbO2 deposited at 65 ◦C on a platinum substrate.
3.2. Electrolysis of Cr(III) solutions
During electrolysis, the following reactions occur:
• at the anode:
2 Cr3+ + 7 H2O → Cr2O72− + 14 H+ + 6 e−
or:
2 H2O → 4 H+ + O2 + 4 e−
• at the cathode:
2 H+ + 2 e− → H2
The values of the overall current efficiency, φ, for 24 h
electrolysis at j = 10 mA cm−2 are reported in Table 3.
Fig. 6. XRD spectrum obtained with Sample 1 (PbO2 electrodeposited
from the alkaline solution S1).
Fig. 7. XRD spectrum obtained with Sample 2 (PbO2 electrodeposited
from the acid solution S2).

D. Devilliers et al. / Electrochimica Acta 48 (2003) 4301–4309
4307
Table 1
XRD data for PbO2 electrodeposited with the alkaline S1 solution (Sample 1)
Experimental
diffraction
angles 2θ (◦)
Relative
intensity
PbO2 variety
Diffraction angles 2θ (◦)
for ␣-PbO2 and (h k l)
indexation JCPDS data
Relative
intensity
Diffraction angles 2θ (◦)
for ␤-PbO2 and (h k l)
indexation JCPDS data
Relative
intensity
16.46
26.79
33.28
100
46
15
␣ or ␤
␣
(2 0 0) at 16.44◦
(3 1 1) at 26.80◦
(4 0 0) at 33.24◦
20
30
30
(020) at 16.44◦
–
(040) at 33.24◦
30
–
5
␣ or ␤
The values of diffraction angles for ␣ and ␤-PbO2 are extracted from the literature.
Table 2
XRD data for PbO2 electrodeposited with the acid S2 solution (Sample 2)
Experimental
diffraction
angles 2θ (◦)
Relative
intensity
PbO2 variety
Diffraction angles 2θ (◦)
for ␣-PbO2 and (h k l)
indexation JCPDS data
Relative
intensity
Diffraction angles 2θ (◦)
for ␤-PbO2 and (h k l)
indexation JCPDS data
Relative
intensity
10.57
11.61
13.08
14.69
16.59
22.14
26.22
24
70
46
100
84
47
␣
(1 1 0) at 10.63◦
–
12
–
100
–
20
45
17
–
–
100
–
95
30
70
16
␤
(1 0 1) at 11.60◦
–
␣
(1 1 1) at 13.05◦
–
␤
(0 1 1) at 14.55◦
(0 2 0) at 16.44◦
(1 2 1) at 22.04◦
(1 3 0) at 26.11◦
␣ or ␤
␣ or ␤
␣ or ␤
(2 0 0) at 16.44◦
(1 3 0) at 22.23◦
(2 2 2) at 26.28◦
50
The values of diffraction angles for ␣- and ␤-PbO2 are extracted from the literature.
With the three substrates, φ increases with the pH value.
That result is in agreement with those obtained by Kuhn and
Clarke [29]. Higher values of φ are expected for higher val-
The instantaneous current efficiency was determined for a
12.5 h electrolysis at j = 10 mA cm−2, performed with the
Ti/Pt/PbO2 electrode as described above (Section 2). As it
can be seen from Fig. 8, ␾ decreases slightly with time from
0.985–0.885, but its value remains very high. The total volt-
age of the cell, ꢀU, was monitored continuously during the
experiment. Its initial value (20 V) is quite high because the
ohmic drop is very important in our laboratory cell. Never-
theless, the value of ꢀU continuously increased. The value
at the end of the 12.5 h experiment was 29 V. That increase
is attributed to the important decrease of the conductivity
of the catholyte. Indeed, as shown in Fig. 9, the concentra-
tion of sulphuric acid decreases sharply in that compartment
because H+ cations are consumed at the cathode (H+ are
brought to the electrode both by migration and diffusion),
whereas SO42− anions move away from the compartment by
ues of pH because protons are formed during the formation
2−
of Cr2O7
.
The highest φ values were obtained for [H2SO4] =
0.005 M (ca. pH 2) for all the substrates, but the best results
were obtained with Ti/Pt/PbO2 (φ = 0.93). Higher pH val-
ues were not tested, because they correspond to unrealistic
industrial solutions.
It is shown from Table 3 that the Ti/Pt/PbO2 electrodes ex-
hibit the best performances. Ebonex®/PbO2 electrodes also
give interesting results but it must be kept in mind that even
the “Non porous grade” Ebonex® substrates have a resid-
ual porosity that allows the electrolyte to diffuse up to the
electric contact, leading to its corrosion [32].
At last, lead anodes on which the PbO2 is formed in situ
during the electrochemical oxidation of Cr(III) do not exhibit
good results.
After the 24 h experiments, the Ti/Pt/PbO2 electrodes
were examined again by SEM. The PbO2 layer deposited
on Ti remained still compact and adherent to the surface.
100
80
60
40
20
0
Table 3
Overall current efficiency for the oxidation of 0.1 M Cr(III) on three
kinds of PbO2 electrodes at several sulphuric acid concentrations after
24 h electrolysis at j = 10 mA cm−2
[H2SO4]/M
0.5
0.05
0.005
0
2
4
6
8
10
12
14
t (h)
Ti/Pt/PbO2
Ebonex®/PbO2
Pb/PbO2
0.60
0.55
0.21
0.80
0.57
0.44
0.93
0.74
0.56
Fig. 8. Instantaneous current efficiency for the oxidation of Cr(III) on
Ti/Pt/PbO2 electrode at j = 10 mA cm−2
.

4308
D. Devilliers et al. / Electrochimica Acta 48 (2003) 4301–4309
i
K
H
1/2 H2
+
-
id
im
H
SO24
im
2
−
SO4
id
Fig. 9. Transport phenomena and electrode reactions in the electrolysis cell fitted with an anionic membrane.
[2] A.T. Kuhn, R. Clarke, J. Appl. Chem. Biotechnol. 26 (1976) 407.
[3] K.L. Pamplin, D.C. Johnson, J. Electrochem. Soc. 143 (1996)
2119.
migration. In Fig. 9, the contributions of diffusion and mi-
gration are indexed with “d” and “m” subscripts respectively.
[4] F.I. Danilov, A.B. Velichenko, Electrochim. Acta 38 (1993) 437.
[5] J.P. Carr, N.A. Hampson, Chem. Rev. 72 (1972) 679.
[6] J.P. Pohl, H. Rickert, in: S. Trasatti (Ed.), Electrodes of Conductive
Metallic Oxides, Part A, Elsevier, Amsterdam, 1980, (Chapter 4).
[7] M. Fleischmann, M. Liler, Trans. Faraday Soc. 54 (1958) 1370.
[8] B.J. Hwang, R. Santhanam, Y.W. Chang, Electroanalysis 14 (2002)
363.
4. Conclusion
The Ti/PbO2 samples have been prepared according to
the procedure reported above. The outer layer is mainly
composed of ␤-PbO2. In terms of current efficiency, these
electrodes exhibit better electrochemical behaviour for the
oxidation of Cr(III) species in sulphuric acid than other
PbO2 electrodes described in the literature. They are not
as fragile as Ebonex®/PbO2 electrodes and they are dimen-
sionally stable. Long time experiments in a flow cell need
now to be performed in order to determine the lifetime of
these electrodes. That flow cell needs to have an optimised
geometry for lowering ohmic drops.
[9] A.B. Velichenko, D.V. Girenko, F.I. Danilov, Russian J. Electrochem.
33 (1997) 96.
[10] A.B. Velichenko, R. Amadelli, A. Benedetti, D.V. Girenko, S.V.
Kovalyov, F.I. Danilov, J. Electrochem. Soc. 149 (2002) C445.
[11] H. Chang, D.C. Johnson, J. Electrochem. Soc. 136 (1989) 17.
[12] J. Lee, H. Varda, S. Uhm, Y. Tak, Electrochem. Commun. 2 (2000)
646.
[13] A.B. Velichenko, D.V. Girenko, F.I. Danilov, J. Electroanal. Chem.
405 (1996) 127.
[14] D. Velayutham, M. Noel, Electrochim. Acta 36 (1991) 2031.
[15] C.N. Ho, B.J. Hwang, Electrochim. Acta 38 (1993) 2749.
[16] M. Ueda, A. Watanabe, T. Kameyama, Y. Matsumoto, M. Sekimoto,
T. Shimamune, J. Appl. Electrochem. 25 (1995) 817.
[17] G. Cifuentes, L. Cifuentes, R. Kammel, J. Torrealba, A. Campi, Z.
Metallkd. 89 (1998) 363.
[18] J.E. Graves, D. Pletcher, R.L. Clarke, F.C. Walsh, J. Appl. Elec-
trochim. 21 (1991) 848.
[19] J.E. Graves, D. Pletcher, R.L. Clarke, F.C. Walsh, J. Appl. Elec-
trochim. 22 (1992) 200.
[20] J. Gonzalez-Garcia, J. Iniesta, E. Exposito, V. Garcia-Garcia, V.
Montiel, A. Aldaz, Thin Solid Films 352 (1999) 49.
[21] C. Bemelmans, T. O’Keefe, E. Cole, Bull. Electrochem. 12 (1996)
591.
[22] P. Rüetschi, R.T. Angstadt, B.D. Cahan, J. Electrochem. Soc. 106
(1959) 547.
Acknowledgements
The SEM experiments and the X-ray diffraction spectra
were performed in the Laboratoire de Physique des liquides
et Electrochimie, Paris. Mr Borensztajn and Dr G. Maurin
are gratefully acknowledged. The Ionac membrane was
provided by Professor G. Pourcelly (UMR CNRS 5635,
Montpellier). The authors express also their thanks to Drs
O. Valet and P. Doppelt (ESPCI, Paris) who performed the
platinum deposits and to Dr H. Groult (CNRS, Paris) for
fruitful discussions.
[23] I. Petersson, E. Ahlberg, B. Berghult, J. Power Sources 76 (1998)
98.
[24] N. Munichandraiah, J. Appl. Electrochem. 22 (1992) 825.
[25] V.H. Dodson, J. Electrochem. Soc. 108 (1961) 401.
[26] A.B. Velichenko, D.V. Girenko, S.V. Kovalyov, A.N. Gnatenko, R.
Amadelli, F.I. Danilov, J. Electroanal. Chem. 454 (1998) 203.
References
[1] D. Pletcher, F.C. Walsh, Industrial Electrochemistry, Chapman &
Hall, New York 1990 (Chapters 5 and 7).

D. Devilliers et al. / Electrochimica Acta 48 (2003) 4301–4309
4309
[27] O. Valet, P. Doppelt, P.K. Baumann, M. Schumacher, E. Balnois, F.
Bonnet, H. Guillon, Microelectron. Eng. 64 (2002) 457.
[28] B.N. Laskorin, N.M. Smirnova, Y.S. Mishukova, A.V. Komarov,
Zhurnal Prikladnoi Khimii (J. Appl. Chem. USSR) 47 (1974) 1870.
[29] F. Baumann, I. Shain, J. Am. Chem. Soc. 78 (1956) 5550.
[30] B. Devos, Thesis, P&M Curie University, Paris, 1997, p. 178.
[31] B. Devos, H. Groult, D. Devilliers, J.-F. Sarrau, M. Chemla, French
Patent 2,788,377 (2001).
[32] J. Pouilleau, H. Groult, D. Devilliers, Journées d’Electrochimie 1995,
Strasbourg, France, abstract 5–22.