Characterization and stability of doped SnO
2
anodes
F. VICENT, E. MORALLO
Â
N, C. QUIJADA, J. L. VA
Â
ZQUEZ*, A. ALDAZ
Departamento de Quõ
Â
mica Fõ
Â
sica. Universidad de Alicante, Apartado 99, 03080 Alicante, Spain
F. CASES
Departamento de Ingenierõ
Â
a Textil, EPS de Alcoy, Universidad Polite
Â
cnica de Valencia,
Paseo del Viaducto 1, 03800 Alcoy, Spain
Doped tin dioxide electrodes have been prepared by a standard spray pyrolysis technique. The
electrochemical behaviour of these electrodes has been investigated by cyclic voltammetry in sul-
phuric acid using the Fe
2+
/Fe
3+
redox couple system as test reaction. Oxygen evolution has been
used to study the stability of doped SnO
2
electrodes. The SnO
2
electrodes doped with antimony and
platinum exhibit the highest stability. XPS analysis shows that the oxidation state of Sn, Sb and Pt
are +4, +3 and +2, respectively, the probable species being SnO
2
,Sb
2
O
3
and PtO.
Keywords: SnO
2
anodes, doping, high overvoltage anodes, surface analysis, oxidation, water treatment
1. Introduct ion
There are several methods for treating industrial
wastewater containing organic and inorganic pollu-
tants: biological treatment , incineration, adsorption,
chemical and electrochemical oxidation and/or re-
duction.
Although biological processes are the easiest and
the most economic pro cesses for wastewater treat-
ment, their application is not always possible, espe-
cially for euents with high concentrations of
organic or toxic compounds. In such cases, the in-
troduction of previous chemical or electrochemical
oxidation stages has become an attractive alternative
for the treatment of wastewater euents. Electro-
chemical oxidation methods have proved to be more
ecient than chemical ones for removal of some
organic pollutants (i.e., phenolic compounds). Thus,
chemical oxidation of phenol by ozone or hydrogen
peroxide with Fe
2+
as catalyst yields 30% reduction
in TOC (total organic carbon) [1, 2]. In contrast, the
use of electrochemical oxidation methods results in a
38% and 90% TOC removal when platinum and
doped SnO
2
are employed, respectively [3±8]. Ko
È
tz
et al. [5, 6] were the pioneers in the study of physical
and electrochemical properties of doped SnO
2
an-
odes. These electrodes present a low resistivity, high
chlorine and oxygen evolution overpotentials, a high
exchange current density for the Ce
3+
® Ce
4+
re-
action and an eciency and a rate of phenol removal
much higher than for Pt and PbO
2
.
The higher TOC removal when electrochemical
oxidation methods are used have been attributed by
Comninellis and Pulgarin [7, 8] to the oxidation of
phenol and intermediate products to CO
2
.
Thus, an adequate electrode for the elimination of
organic pollutants should present a high oxygen
overpotential, a high electrical conductivity and also
good stability. The doped SnO
2
electrode satis®es
these requirements.
The aims of this work are twofold: the preparation
of stable electrodes for oxidative purposes and the
characterization of SnO
2
electrodes doped with anti-
mony or antimony and platin um. The techniques
used to characterize the electrodes were scanning
electron microscopy (SEM), energy dispersion X-ray
(EDX), X-ray photoelectron spectroscopy (XPS) and
cyclic voltammetry.
2. Experimental details
2.1. Electrode preparation
Several metal substrates, temperatures and solvents
were tested for the preparation of the doped SnO
2
electrodes [9, 10]. Due to the fact that the best results
obtained by us have been using Ti and ethanolic so-
lutions, the present study will focus on this material
and solution.
Doped Ti/SnO
2
electrodes were prepared by a
standard spray-pyrolysis method [5, 11, 12]. The
spray solution used for the preparation of the SnO
2
electrode doped with antimony was 10 g SnCl
4
.5H
2
O
and 1 g SbCl
3
in 100 ml of ethanol±HCl mixture; the
same solution with 1 g Pt (2.1% H
2
PtCl
6
) was used
for obtaining the best platinum and antimony doped
electrodes. The titanium substrate, a wire of 0.5 mm
diameter (99.6% of purity, Goodfellow Metals), was
previously etched in a 10% oxalic acid solution for
1 h, then rinsed with water and heated to 400 °C. The
ethanolic solution was then sprayed onto the titanium
with an air-atomizing spray at a ®xed distance of
* Author to whom correspondence should be addressed.
40 cm. Then, the electrode was heated for 10 min at
400 °C. The above operation was repeated at least ®ve
times. The thickness of the doped SnO
2
®lms depends
on the number of spray-pyrolysis sequences. Finally,
further heat treatment for 1 h at 600 °C was carried
out.
The same experimental proced ure was used for the
preparation of electrodes with sheet and expanded
titanium substrates (IMI 125 type, INAGASA).
2.2. Electrode characterization
Dierent techniques were applied to characterize the
properties of the doped SnO
2
electrodes. The elec-
trochemical behaviour of doped SnO
2
electrodes was
studied by cyclic voltammetry. The voltammograms
were obtained with a standard set-up using a po-
tentiostat (HQ Instruments, model 101), a generator
(EG&-G PARC, model 175) and an X±Y recorder
(Philips PM 8133). The electrolyte was a 0.5
M
sul-
phuric acid solution prepared from Merck suprapur
and Millipore water or a 0.5
M
H
2
SO
4
+ 5 ´ 10
)2
M
FeSO
4
(Merck p.a.) solution. The solutions were de-
oxygenated by bubbling nitrogen (N-50) before each
experiment and an inert atmosphere was maintained
over the working solution during the experiment. The
counter electrode was a platinum wire and the po-
tentials were measured versus a reversible hydrogen
electrode immersed in 0.5
M
sulphuric acid and con-
nected to the cell through a Luggin capillary. The
cyclic voltammograms were usually recorded at room
temperature at a sweep rate of 50 mV s
)1
.
Surface analysis was performed by X-ray photo-
electron spectroscopy (XPS) using a ESCA LAB 210
spectrometer. The X-ray source was MgK
a
of energy
1253.6 eV with a power of 240 W. The samples were
placed on `posiloc' standard sampl e holders. A
binding energy (BE) of 284.9 eV, corresponding to
the C 1s peak was used as an internal standard. The
pressure in the analysis chamber was maintained
below 2 ´ 10
)9
mbar during the measurements.
Scanning electron microscopy (SEM) was em-
ployed to observe the surface morphology of the
electrodes using a Jeol (JSM 840). SEM also provided
information on the dierent elements present on the
surface using the energy dispersive X-ray detector
(Link QX 200 EDX).
The thickness of the coatings were determined by
the step method using a pro®lometer (Surfometer SF
220).
2.3. Stability test
The oxygen evolution reaction was used to study the
stability of doped SnO
2
electrodes in 0.5
M
H
2
SO
4
or
0.5
M
K
2
SO
4
(Merck p.a.) solutions. The potential
versus time curves of the doped SnO
2
electrodes at
constant current density (referred to a geometric area
of the electrode of 0.08 cm
2
) depends on the polari-
zation time and on the previous history of the elec-
trode. An increase of 1 V in the potential was adopted
as an indication of the loss of electrocatalytic activity
of doped SnO
2
electrodes [13]. This potential increase
can be produced by the formation of a passivated
surface layer, probably caused by the hydration of
the SnO
2
layer and/or the passivation of the layer-
substrate interface [5, 14].
3. Res ults and discussion
3.1. SnO
2
electrodes doped with antimony
Figure 1 (solid line) shows the voltammogram re-
corded in a 0.5
M
H
2
SO
4
solution for a SnO
2
electrode
doped with antimony with a ®ve spray-pyrolysis se-
quence. The voltammogram is featureless in this po-
tential range and little information is obtained about
the composition and behaviour of the electrode sur-
face. At potentials below 0.3 V the onset of a cathodic
process becomes evident. The current associated with
this process increases with decreasing potential and
eventually overlaps the hydrogen evolution current.
No peaks or waves appear before oxygen evolution,
which begins at approximately 2.2 V. Compared to
the behaviour observed for a platinum electrode in
the same electrolyte and for the same current density,
oxygen evolution is shifted positively about 500 mV.
To check the charge transfer rate through the
doped SnO
2
electrode-electrolyte interface, the volt-
ammetric behaviour of the redox system Fe
3+
/Fe
2+
was studied. From the value of the peak potential and
from the value of the dierence between the potential
of the anodic and cathodic peaks (DE
p
), a direct es-
timate of the reversibility of the system, that is, the
electron transfer rate, could be obtained. Figure 2
shows the stabilized voltammogram obtained with
this electrode for a 0.5
M
H
2
SO
4
+ 5 ´ 10
)2
M
FeSO
4
solution at 50 mV s
)1
. Well de®ned reduction and
oxidation peaks are observed at about 0.65 and
0.75 V, respectively. The potentials of the anodic ( E
a
p
)
and cathodic (E
c
p
) peaks shift with scan rate, in-
creasing the value of E
a
p
ÿ E
c
p
DE
p
. The peak sep-
aration, from 96 to 120 mV (for 50 and 200 mV s
)1
,
respectively), is always greater than that expected for
a reversible one-electron process (DE
p
59 mV).
However, the doped SnO
2
electrode behaves better
than a platinum electrode at all scan rates (DE
p
values
from 104 to 136 mV).
Fig. 1. Voltammograms for a SnO
2
electrode doped with antimony
with ®ve spray-pyrolysis sequence in 0.5
M
H
2
SO
4
. v 50 mV s
)1
.
(ÐÐ) before and (- - -) after 5 h of electrolysis at 10 mA cm
)2
in
0.5
M
H
2
SO
4
.
The SEM photomicrographs of the surface of the
SnO
2
electrode doped with antimony are shown in
Fig. 3. These ®gures show a slightly rough surface,
with the particles of SnO
2
small and uniformly dis-
tributed.
The presence and distribution of Sn and Sb on the
electrode were studied by EDX. The results indicate a
homogeneous distribution of the components (Sn,
Sb) on the electrode. Sn and Sb are detected in a
proportion of 23.5% and 6%, respectively, while
titanium is detected at 66.6%. This detection indi-
cates that the thickness of the SnO
2
®lm is, according
to Rosin skaya et al. [15], lower than 2 lm.
Pro®lometric measurements were carried out on a
titanium sheet partially covered wi th SnO
2
doped
with antimony (®ve spray-pyrolysis sequences). A
coating thickness lower than 1 lm is obtained. This
result is in agreem ent with that obtained from EDX
measurements. Thus, it can be concluded that SnO
2
®lms formed by a ®ve spray pyrolysis sequence are
very thin. The holes created during the pre-treatment
of titanium (holes from 3 to 6 lm have been previ-
ously observed using this pre-treatment [9]) are
mainly covered during spray pyrolisis.
From the electrochemical results, that is, a high
oxygen overpotential and good electrochemical be-
haviour as deduced from the electron trans fer rate for
the Fe
3+
/Fe
2+
couple, the electrode obtained with a
®ve spray-pyrolysis sequence is a good candidate for
the oxidation±elimination of organic compounds. To
verify the chemical and electrochemical stability of
this electrode with time, it was employed as anode for
oxygen evolution using a 0.5
M
H
2
SO
4
solution at a
constant current density of 10 mA cm
)2
(Fig. 4,
squares). It can be seen that after 4 h electrolysis the
dierence Et ÿ Et 0 increased 1 V and reached
4.5 V after 5 h. Figure 1, dashed line, shows the volt-
ammogram of this electrode in H
2
SO
4
after 5 h of
electrolysis. The voltammetric pro®le of this electrode
has changed showing a decrease in current for all
potentials. Also, the oxygen evolution has shifted to
more positive potentials. Figure 5, solid line, shows
the voltammetric behaviour of the Fe
3+
/Fe
2+
redox
system on this electrode. The voltammogram has
dramatically changed, showing a large decrease in the
value of the anodic and cathodic current peaks and a
shift of these peaks to more positive and negative
potentials, respectively. The separation between peak
potentials is now 600 mV instead of 96 mV (Fig. 2).
This result clearly shows that the electrode has lost its
electrocatalytic activity after working in sulphuric
acid for 5 h.
A plausible explanation for this behaviour is the
formation of a passivating layer at the substrate±®lm
interface [5], which imparts low conductivity to the
electrode [9]. This passivating layer may be formed
because the electrolyte solution is in contact with the
titanium substrate due to the small thickness of the
®lm (lower than approximately 1 lm) or its high po-
rosity.
The stability of the electrode can be improved by
increasing the thickness of the SnO
2
®lm by increa s-
ing the number of spray-pyrolysis sequences. The
Fig. 2. Voltammetric behaviour of a SnO
2
electrode doped with
antimony (ÐÐ) and with antimony and platinum (0.2%) (- - -)
with a ®ve spray-pyrolysis sequence in 5 ´ 10
)2
M
. FeSO
4
+ 0.5
M
H
2
SO
4
solution. v 50 mV s
)1
.
Fig. 3. Scanning electron micrograph of a SnO
2
electrode doped
with antimony with a ®ve spray-pyrolysis sequence.
Fig. 4. Et ÿ Et 0 against electrolysis time in 0.5
M
H
2
SO
4
for
SnO
2
electrodes doped with antimony with: (n) ®ve spray-pyrolysis
sequences, (d) nine spray-pyrolysis sequences. j 10 mA cm
)2
.
initial electrochemical behaviour of a SnO
2
electrode
doped with antimony obtained after nine spray-
pyrolysis sequences is similar to that obtained after
®ve sequences (Figs 1 and 2, solid line). However, the
stability of the electrode measured as the time for
obtaining Et ÿ Et 0 > 1 V, increases from 5 to
9 h of electrolysis at 10 mA cm
)2
in 0.5
M
H
2
SO
4
(Fig. 4, triangles).
3.2. SnO
2
electrodes doped with antimony
and platinum
It has been shown that the electrocatalytic activity
and the resistance to corrosion of these types of
electrode can be increased if mixed oxides are used
[16]. One of these oxides is the a ctive component
while the other improves resistance to anodic disso-
lution. The preparation procedure is crucial in de-
termining the ®nal properties.
For this reason, another possible method for en -
hancing the stability of the electrode, in addition to
the thickening of the SnO
2
®lm, is to obtain coatings
of mixed oxides of tin and platinum on titanium
substrate. Figure 6, solid line, shows the voltammo-
gram in a 0.5
M
H
2
SO
4
solution for a SnO
2
electrode
doped with antimony and platinum prepared with a
®ve spray-pyrolysis sequence with a 10%
SnCl
4
.5H
2
O + 1% SbCl
3
+ 0.42% H
2
PtCl
6
in eth-
anol±HCl mixture. The oxygen evolution is shifted to
less positive potentials than those obtained with a
SnO
2
electrode doped only with antimony (Fig. 1,
solid line). During the negative sweep a small peak
near 0.6 V is also obtained, which could be associated
with the presence of platinum because this peak does
not appear in a SnO
2
electrode without platinum
(Fig. 1, solid line).
Figure 2 (dashed line) shows the voltammogram
for this electrod e in a 0.5
M
H
2
SO
4
+ 0.05
M
FeSO
4
solution. Good behaviour with respect to the Fe
3+
/
Fe
2+
redox couple is observed. The DE
p
values are
very similar to those obtained with a platinum elec-
trode. Figure 7 (circles) shows the stability of the
SnO
2
±Sb±Pt (0.2%) electrode (®ve spray pyrolysis) in
electrolysis at a constant current density of
40 mA cm
)2
in a 0.5
M
H
2
SO
4
solution. The electrode
potential increases by 1 V from the initial value after
60 h of electrolysis. Figures 5 and 6 (dashed lines)
show the voltammograms for this electrode after 60 h
of electrolysis in 0.5
M
H
2
SO
4
+ 0.05
M
FeSO
4
and
0.5
M
H
2
SO
4
solutions, respectively. The voltammo-
grams clearly show a decrease in electrocatalytic ac-
tivity for the Fe
3+
/Fe
2+
reaction with electrolysis
time.
If the number of pyrolysis processes is increased to
15 and the proportion of platinum in the spray so-
lution also increases from 0.2 to 1% the voltammet ric
behaviour is similar to that obtaine d with the latter
electrodes but the stability increases. Figure 7 (tri-
angles) shows that the life time of this electrode is
now of 425 h for an elect rolysis at 40 mA cm
)2
in
0.5
M
H
2
SO
4
. The lifetime of the electrode for elec-
trolysis experiments conducted in neutral solution
(0.5
M
K
2
SO
4
) at a constant current density of
40 mA cm
)2
, is appro ximately 760 h (Fig. 7, squares).
Figure 8 shows the morphological details of the
surface of the electrode obtained from a 10%
SnCl
4
.H
2
O + 1% SbCl
3
+ 2.1% H
2
PtCl
6
ethanol-
ic±HCl mixture and ®fteen spray-pyrolysis processes
on a wire titanium substrate. The surface presents an
uniform distribution of particles and is more granu-
lated and spongy than that obtained for a SnO
2
electrode doped with antimony and a ®ve spray-
pyrolysis sequence (Fig. 3).
Fig. 5. Voltammetric behaviour of SnO
2
electrodes doped with a
®ve spray-pyrolysis sequence in 5 ´ 10
)2
M
FeSO
4
+ 0.5
M
H
2
SO
4
.
Key: (ÐÐ) SnO
2
doped with antimony after 5 h of electrolysis at
10 mA cm
)2
; ( - - - ) SnO
2
doped with antimony and platinum after
60 h of electrolysis at 40 mA cm
)2
. v 50 mV s
)1
.
Fig. 6. Voltammetric behaviour of a SnO
2
electrode doped with
antimony and platinum with a ®ve spray-pyrolysis sequence in
0.5
M
H
2
SO
4
. Key: (ÐÐ ) before and (- - - ) after 60 h of electrolysis
at 40 mA cm
)2
. v 50 mV s
)1
.
Fig. 7. Et ÿ Et 0 against time of electrolysis for SnO
2
elec-
trodes doped with antimony and platinum with: (d) ®ve spray-
pyrolysis sequences from a 10% SnCl
4
+ 1% SbCl
3
+ 0.42%
H
2
PtCl
6
in ethanol + HCl solution in 0.5
M
H
2
SO
4
, (m) ®fteen
spray-pyrolysis sequences from a 10% SnCl
4
+ 1%
SbCl
3
+ 2.1% H
2
PtCl
6
in ethanol + HCl solution in 0.5
M
H
2
SO
4
and (n) ®fteen spray-pyrolysis sequences from a 10% SnCl
4
+ 1%
SbCl
3
+ 2.1% H
2
PtCl
6
in ethanol + HCl solution in 0.5
M
K
2
SO
4
. j 40 mA cm
)2
.
Titanium is not detected in the EDX analysis for
this electrode and only tin, antimony and platinum
are present in a proportion of 30.1, 51.5 and 18.4%,
respectively. These data also indicate that the thick-
ness of the doped SnO
2
®lm is higher than 2 lm. In
fact, pro®lometric measurement yielded a ®lm thick-
ness of approximately 4 lm.
Figure 9 shows two photomicrographs of this
electrode after 760 h of electrolysis in 0.5
M
K
2
SO
4
at
40 mA cm
)2
. The surfa ce morphology of the electrode
has changed and is smoother. The results of the EDX
data show a change in electrode composition, titani-
um now being detected. Tin, antimony, platinum and
titanium are present in a proportion of 25.5, 34.3,
23.1 and 14.1%, respectively.
For industrial electrodes to be used in ®lter-press
cells, additional experiments were carried out with
electrodes prepared on titanium sheet and expanded
titanium. The electrolysis of a 0.5
M
H
2
SO
4
solution
at 100 mA cm
)2
was carried out using a Ti(expand-
ed)/SnO
2
electrode doped with 1% antimony and 1%
platinum. The lifetime of this electrode increased to
1800 h, a value substantially higher than that ob-
tained previously (425 h) even using a current density
2.5 times higher.
3.3. XPS characterization
XPS analysis was carried out to gain more informa-
tion about the nature of the surface of the electrode
before and after prolonged oxygen evolution.
Figure 10 shows the XPS spectra for an electrode of
Fig. 8. SEM of a new SnO
2
electrode doped with antimony and
platinum with a ®fteen spray-pyrolysis sequence.
Fig. 9. Scanning electron micrograph of a SnO
2
electrode doped
with antimony and platinum with a ®fteen spray-pyrolysis sequence
after 760 h of electrolysis in a 0.5
M
K
2
SO
4
solution at 40 mA cm
)2
.
Fig. 10. XPS spectra of a doped SnO
2
electrode, obtained with a
®fteen spray-pyrolysis sequence from 10% SnCl
4
.5H
2
O + 1%
SbCl
3
+ 2.1% H
2
PtCl
6
in ethanol + HCl mixture, before the
electrolysis process (thick line) and after 760 h of electrolysis in
0.5
M
K
2
SO
4
at 40 mA cm
)2
(thin line). (a) Sn(3d
5/2
); (b) Sb(3d
5/2
)
![Table 2. Binding energies of the XPS spectra of Sb 3d5/2, Sn 3d5/2 and Pt 4f7/2, corresponding to [17 ]](/figures/table-2-binding-energies-of-the-xps-spectra-of-sb-3d5-2-sn-3hrvi9l6.webp)
