Electrochemical behavior of IrxRu1-xO2 oxides as anodic electrocatalyst for electrosynthesis of dinitrogen pentoxide

Article abstract of DOI:10.1016/j.electacta.2012.04.060

The Ir_xRu_1-xO₂/Ti electrodes (x = 0-1) were prepared by thermal decomposition of non-aqueous solutions of H ₂IrCl₆ and RuCl₃. The surface morphology and microstructure of the Ir_xRu_1-xO₂ coatings were characterized by XRD and SEM. The electrochemical characterization of the surface was carried out in a 0.5 mol L^-1 H₂SO₄ solution. The Ir_xRu_1-xO₂ coating with x = 0.5 showed a minimal nanocrystalline particle size and a maximum surface charge (q*). The electrochemical behaviors of the Ir_xRu _1-xO₂ coatings for oxidation of N₂O₄ were determined by cyclic voltammetry (CV), the open circuit potential (OCP) and Tafel plot in 1.7 mol L^-1 N₂O₄/HNO ₃ solutions. The open-circuit potential (E_ocp) in N ₂O₄/HNO₃ solution was about 1.18 V and it was almost independent of the surface composition. Cyclic voltammetry showed a broad wave feature at 1.4-1.8 V in the anodic scan, which was attributed to the oxidation of NO₂ derived from N₂O₄ to NO ₂⁺. The redox couple of NO₂/NO₂ ⁺ showed a quasi-reversible behavior and the process was under diffusion control. The electrocatalytic activities of various Ir _xRu_1-xO₂ coatings for oxidation of N ₂O₄ were evaluated by potentiostatic polarization curves under the quasi-steady state conditions. The Ru_0.5Ir _0.5O₂ coating showed the lowest Tafel slope, 35.7 mV, in the low current densities, indicating that this coating had the highest electrocatalytic activity. The experiments of electrochemical synthesis of N₂O₅ from N₂O₄ in nitric acid also indicated that the Ru_0.5Ir_0.5O₂ coating had the best electrocatalytic performance.

Full text of DOI:10.1016/j.electacta.2012.04.060

Electrochimica Acta 74 (2012) 227–234
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Electrochemical behavior of IrxRu1−xO2 oxides as anodic electrocatalyst for
electrosynthesis of dinitrogen pentoxide
Qingfa Wang∗, Fangmin Wu, Nan Wang, Li Wang, Xiangwen Zhang
Key Laboratory for Green Chemical Technology, School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
The IrxRu1−xO2/Ti electrodes (x = 0–1) were prepared by thermal decomposition of non-aqueous solu-
tions of H2IrCl6 and RuCl3. The surface morphology and microstructure of the IrxRu1−xO2 coatings were
characterized by XRD and SEM. The electrochemical characterization of the surface was carried out in a
0.5 mol L−1 H2SO4 solution. The IrxRu1−xO2 coating with x = 0.5 showed a minimal nanocrystalline par-
ticle size and a maximum surface charge (q*). The electrochemical behaviors of the IrxRu1−xO2 coatings
for oxidation of N2O4 were determined by cyclic voltammetry (CV), the open circuit potential (OCP) and
Tafel plot in 1.7 mol L−1 N2O4/HNO3 solutions. The open-circuit potential (Eocp) in N2O4/HNO3 solution
was about 1.18 V and it was almost independent of the surface composition. Cyclic voltammetry showed
a broad wave feature at 1.4–1.8 V in the anodic scan, which was attributed to the oxidation of NO2 derived
Received 4 February 2012
Received in revised form 13 April 2012
Accepted 17 April 2012
Available online 24 April 2012
Keywords:
Electrochemical synthesis
Dinitrogen pentoxide
N2O4 oxidation
IrxRu1−xO2 oxides
Anodic electrocatalyst
+
from N2O4 to NO2+. The redox couple of NO2/NO2 showed a quasi-reversible behavior and the process
was under diffusion control. The electrocatalytic activities of various IrxRu1−xO2 coatings for oxidation
of N2O4 were evaluated by potentiostatic polarization curves under the quasi-steady state conditions.
The Ru0.5Ir0.5O2 coating showed the lowest Tafel slope, 35.7 mV, in the low current densities, indicating
that this coating had the highest electrocatalytic activity. The experiments of electrochemical synthesis
of N2O5 from N2O4 in nitric acid also indicated that the Ru0.5Ir0.5O2 coating had the best electrocatalytic
performance.
© 2012 Elsevier Ltd. All rights reserved.
1. Introduction
the systematic investigation of the electrode materials which are
suitable for electrosynthesis of N2O5 from N2O4 in HNO3.
Dinitrogen pentoxide (N2O5) is generally believed as a green
nitrating agent and has been widely used for the synthesis of ener-
getic materials [1,2]. The electrosynthesis of N2O5 from N2O4 in
nitric acid (HNO3) is a promising method and have attracted more
and more researchers’ attention [1–12]. Previous studies, includ-
ing the performance evaluation of a range of electrodes as well as
scale-up studies, have shown that this electrochemical process is
very viable. The basic anodic reaction for this method is to oxidize
N2O4 in HNO3 as follows:
Recently, modulation to improve the electrocatalytic activity of
dimensionally stable anodes (DSA®) was best achieved by mix-
ing two or more components [13,14]. The iridium dioxide (IrO2)
mixed with ruthenium dioxide (RuO2) has been widely used as the
coatings for the oxygen evolution reaction (OER) in acid medium
for different fuel cells [15–25] and chlorine evolution reaction
for chlor-alkali industry [26–28]. Cheng et al. [15] observed that
the IrxRu1−xO2 coating prepared by Adams’ fusion method was
more active than IrO2 and more stable than RuO2 in solid poly-
mer electrolyte (SPE) water electrolysis. The Ir0.2Ru0.8O2 coating
had a better performance. The structure of RuO2/IrO2 mixed oxide
films deposited on Ti support has been comprehensively investi-
gated using extended X-ray absorption fine-structure spectroscopy
(EXAFS) [16]. Kotz and co-workers [17] investigated the electro-
chemical and electrocatalytic properties with respect to the OER
and the oxygen reduction reaction (ORR) of a number of irid-
ium compounds (including pyrochlores, perovskites and fluorides)
in concentrated base (45 wt% KOH). Avaca and co-workers [18]
characterized the electrode surfaces modified by sol–gel derived
RuxIr1−xO2 coatings for OER in acid medium. Arico and co-workers
[19] described the preparation and evaluation of RuO2/IrO2, IrO2–Pt
and IrO2/Ta2O5 electrodes for the OER in an SPE electrolysis cell.
N2O4 + 2HNO3 → N2O5 + 2e− + 2H+
(I)
However, the drastic conditions bring out a very high require-
ment on the electrode materials. In previous work, it was shown
that the anodes consisting of noble metals or metal oxides such as
Pt, Pt–Ir, IrO2 and RuO2 had lower overpotentials for the oxidation
of N2O4 [6,7]. IrO2 showed a superior performance among the stud-
ied metal oxides [6–9]. However, little work has been done upon
∗
Corresponding author. Tel.: +86 22 27892340; fax: +86 22 27892340.
E-mail address: qingfaw@yahoo.com.cn (Q. Wang).
0013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.electacta.2012.04.060

228
Q. Wang et al. / Electrochimica Acta 74 (2012) 227–234
Takasu et al. [20] described the ORR behavior of RuO2/Ti, IrO2/Ti
and IrM (M: Ru, Mo, W, V)Ox/Ti binary oxide electrodes in H2SO4
solution. More recently, Lyons and Floquet [21] investigated a com-
prehensive examination of the catalytic properties with respect
to the OER activity of a series of conductive metallic IrxRu1−xO2
electrodes in aqueous acid and alkaline solutions. The effect of
Sb-doped SnO2 (ATO) nanoparticles as the support on the electro-
catalytic activity of IrxRu1−xO2 was investigated by Marshall and
Haverkamp [22]. The use of mixed Ru/Ir metal oxide catalysts in
SPE cells have also been discussed in works by Tunold et al. [23] and
Song et al. [24]. A continuous solid solution with the composition of
Ir0.2Ru0.8O2 obtained by reactive sputtering showed good erosion-
tolerant performance in the acidic electrolyte [25]. In addition,
Hansen and co-workers [26] investigated the electrochemical chlo-
rine evolution on rutile (1 1 0) IrO2 and RuO2 surfaces using density
functional theory (DFT) and compared the required overpotential
for the chlorine evolution reaction and for the OER. Generally, the
Ru0.3Ir0.7O2 coating shows excellent electrocatalytic performance
for the hydrogen/oxygen evolution reaction (HER/OER). Besides,
oxides such as TiO2 [29], SnO2 [30,31], Ta2O5 [32,33], SiO2 [34],
ZrO2 [35] and MnO2 [36] have also been investigated as additives
to increase the activity, selectivity, and stability of the Ir-based
electrodes.
Fig. 1. Schematic diagram of electrolysis for N2O5 synthesis.
used as the counter electrode and a reference electrode, respec-
tively. Solutions of 0.5 mol L−1 H2SO4 at 25 ◦C and 1.7 mol L−1 N2O4
in HNO3 at 0 ◦C were used as the electrolytes. The cyclic voltam-
mograms were performed from −0.4 to 1.4 V in H2SO4 solution
and from 0.8 to 2.0 V in N2O4/HNO3 solution. The voltammetric
curves were obtained at the following scan rates: 5, 10, 15, 20,
30, 40, 50, 60, 80, 100, 200, 300, 400 and 500 mV s−1. The anodic
polarization curves were performed to evaluate the apparent elec-
trocatalytic activity toward the oxidation of N2O4. The ohmic drop
(jRs) was corrected by the electrochemical impedance spectroscopy
(EIS) measurement.
In our previous work,
a commercial anode coated with
Ir0.3Ru0.7O2 layer was used in electrosynthesis of N2O5 and showed
a better electrocatalytic performance than IrO2 layer [37]. The aim
of this work is to systematically investigate the performance of the
IrxRu1−xO2 coatings prepared by thermal decomposition on Ti sub-
strate to electrochemically synthesize N2O5 from oxidizing N2O4
in HNO3.
2. Experimental
2.1. Preparation of the electrodes
2.4. Electrosynthesis of N2O5
Two
titanium
plates
with
different
sizes
The electrosynthesis of N2O5 experiments were carried out in
a plate-and-frame electrolyzer. A scheme of electrolysis appara-
tus is shown in Fig. 1. The prepared IrxRu1−xO2/Ti and IrO2/Ti
(45 mm × 70 mm × 1 mm samples) were used as the anode and
cathode, respectively, with an active electrode area of 7.1 cm2.
Prior to the experiments, these two compartments of the elec-
trolyzer were rinsed by circulating 98 wt% HNO3 for 10 min. 98 wt%
HNO3 was pumped into the reservoirs and N2O4 was added into
the anolyte. The tested solutions were cooled to 0 ◦C and pumped
into the electrolyzer. Then the electrolysis was started. The applied
potential difference to the electrolyzer was adjusted to give the
required current density (600–1000 A m−2). The electrolysis was
carried out for 4 h and then the anolyte was sampled. More details
of the electrosynthesis of N2O5 can be found elsewhere [37].
(10 mm × 10 mm × 1 mm and 45 mm × 70 mm × 1 mm, grade
1, Baoji Hongshunda Titanium Industry Co., Ltd.) were used as
the substrate and treated before use according to the procedure
described in Ref. [18]. The RuC13·3H2O (Alfa Aesar, 99.9%) and
H2IrCl6·6H2O (Sigma–Aldrich 99.9%), were used as precursors and
dissolved in a mixture of alcohol/n-butanol (molar ratio = 2:1) with
a total metal ion concentration of ca. 0.027 mol mL−3. The solution
was painted on Ti substrate. The obtained samples were dried at
100 ◦C for 10 min and calcinated at 450 ◦C for additional 10 min,
and then it was cooled to room temperature. The operational
procedure was repeated for 30 times. Finally, the electrodes were
calcinated at 450 ◦C for additional 1 h.
2.2. Physical characteristics
X-ray diffraction (XRD) data in the 2ꢀ range of 5–90◦ at 2◦ min−1
were collected on a Rigaku D-max 2500 V/PC X-ray diffractometer
(Rigaku Corporation) using Cu K␣ radiation source (40 kV, 200 mA).
The crystal sizes of coatings were determined using Scherrer Equa-
tion. The SEM analysis was performed by a Hitachi X-650 scanning
electron microscope. The elemental distribution was determined
by energy dispersive X-ray analysis (EDX, Oxford Instrument ISIS
300).
3. Results and discussion
3.1. XRD characterization
The XRD patterns of IrxRu1−xO2 coatings annealed at 450 ◦C are
shown in Fig. 2. Although the signal from the Ti substrate was much
stronger than that of the film, it was possible to identify and mea-
sure the intensity of the oxide peaks except for the sample with IrO2
coating. The rutilic oxide phases, with peaks at 2ꢀ of ca. 28◦, 34.7◦
and 54.4◦, could be identified. These peaks were the reflections of
(1 1 0), (1 0 1) and (2 1 1) faces of both Ru and Ir oxides, respec-
tively [18]. Meanwhile, the oxide peaks were rather broad, which
suggested the existence of very fine crystallites in the structure.
Calculated from (1 1 0) face with use of the Scherrer formula, the
average particle sizes was about 5–8 nm for IrxRu1−xO2 (x = 0.1–1)
2.3. Electrochemical measurements
The cyclic voltammetry (CV) and polarization measurement
were carried out using a CHI 660 electrochemical station. All mea-
surements were carried out in a standard three-electrode cell. A
1 cm2 platinum foil and the saturated calomel electrode (SCE) were

Q. Wang et al. / Electrochimica Acta 74 (2012) 227–234
229
3.2. Surface morphology
The surface morphologies of the IrxRu1−xO2 coating were exam-
ined by SEM. As shown in Fig. 3, the oxide layers presented a
cracked-mud structure, typical of metal oxide coatings obtained by
thermal decomposition [38,41–44]. These cracks were produced on
the surface during the cooling of the electrode to room tempera-
ture [15,45] and/or during the solvent evaporation stage [38]. The
IrxRu1−xO2 coatings had a similar morphology. The RuO2 and IrO2
coating showed a relatively homogeneous flat, smooth morphol-
ogy consisting of a little crack island and a lot of agglomerated
particles. It could be observed that the Ir0.5Ru0.5O2 coating had
the morphology with more cracks. These different morphologies
of the IrxRu1−xO2 coatings might be ascribed to the difference in
the thermal expansion coefficient between coating and the sub-
strate, and the difference in crystallization temperature as well as
the orientation of crystalline growth of RuO2 and IrO2.
The coating compositions of the samples were determined by
EDX analysis (see Fig. 4) and were correlated with the composi-
tion of the precursor’s solution. As shown in Fig. 5, an excellent
correlation was obtained. This indicated no evident enrichment
of Ru or Ir occurred in the surface, which was different with the
previous results by X-ray photoelectron spectroscopy (XPS) anal-
ysis [46]. Generally, the sample thickness determined by EDX is
about 1000 nm and is only 1–3 nm determined by XPS. Therefore,
the results from EDX analysis and XPS analysis are usually different
for the same sample. It can be considered that the composition by
EDX analysis is more close to the really coating composition. The
similar phenomenon of the different compositions obtained from
EDX and XPS characterization was also reported for the SnO2–RuO2
oxides [47]. In addition, the Ti element was detected in EDX anal-
ysis for some samples (see Fig. 4) while not detected for the other
samples. Thus the coating thickness was estimated to about 1 ␮m.
Ti
Ti
(101)
(211)
(110)
IrO2
Ir0.7Ru0.3O2
Ir0.5Ru0.5O2
Ir0.3Ru0.7O2
Ir0.1Ru0.9O2
RuO2
20
30
40
2θ / deg
50
60
Fig. 2. XRD patterns of Ti supported IrxRu1−xO2 coating.
and about 10.1 nm for RuO2. It was found that the particle sizes
decreased with the increase of Ir content for the Ru-rich samples,
but increased with the increase of Ir content for the Ir-rich samples.
This is because the crystallization temperature of RuO2 is lower
than that of IrO2 [13,38]. The smallest oxide particles of IrxRu1−xO2
was obtained at x = 0.5. This suggested the appropriate content of
Ir contributed to the dispersion of RuO2 particles. In addition, the
main peaks of pure IrO2 were not obvious in Fig. 2. These were
mainly due to the fact that the crystallization temperature of IrO2
is higher than the preparation temperature in this study.
The lattice parameters of IrO2 and RuO2 as well as the ionic
radius of Ir4+ (0.0625 nm) and Ru4+ (0.062 nm) are quite similar
3.3. Electrochemical behavior in 0.5 mol L−1 H2SO4 solution
˚
˚
(IrO2, a = 4.4983 and c = 3.1544 A; RuO2, a = 4.4994 and c = 3.1071 A
(JCPDS 40-1290)). According to Hume-Rothery theory [39] it is pos-
sible to form a continuous solid solution of IrO2 and RuO2, where
Ru4+ and Ir4+ share the same site on the cationic sub-lattice of a
tetragonal (rutile-like) phase [40]. As shown in Fig. 2, the peak posi-
tion of (1 1 0) face of IrxRu1−xO2 (x = 0.1–0.9) fell exactly between
RuO2 and IrO2. This indicated that a solid solution of RuO2 and IrO2
was formed.
The voltammetric behaviors of the IrxRu1−xO2/Ti electrode were
measured with a 20 mV s−1 scan rate in 0.5 mol L−1 H2SO4 solu-
tion from −0.4 V to 1.4 V at room temperature. As shown in Fig. 6,
there is one broad peak at ca. 0.6 V for pure RuO2 and one peak at
0.85 V for pure IrO2, which was assigned to the redox transition of
the couple Ru(III)/Ru(IV) and Ir(III)/Ir(IV), respectively. Meanwhile,
Fig. 3. SEM images of IrxRu1−xO2/Ti electrode.

230
Q. Wang et al. / Electrochimica Acta 74 (2012) 227–234
Fig. 4. EDX spectra of the electrodes: (a) Ir0.1Ru0.9O2, (b) Ir0.3Ru0.7O2, (c) Ir0.5Ru0.5O2, and (d) Ir0.7Ru0.3O2.
the peaks near 1.2 V were attributed to the redox transitions of
Ru(IV)/Ru(VI) and Ir(IV)/Ir(VI) surface groups, respectively [18,25].
For the Ru0.5Ir0.5O2 coating, the peaks were not clear due to the
superposition of the redox process. These results were in close
agreement with those previously reported [18,25].
0.015
0.010
0.005
0.000
-0.005
The surface charge (q*) was calculated according to the method
described in Ref. [48]. According to the literature, the total charge
q∗T is attained as the scan rate v approaches 0, and the outer charge
q∗o is attained as v approaches ∞. The former showed all the active
surfaces and the latter showed the outer and more accessible active
3
2
1
surface [47]. The difference between q∗T and qo∗ gives q∗, the charge
i
related to the inner and less accessible active sites which meant
the regions forming the walls of pores, grain boundaries, crevices,
and cracks, etc. [48–50]. The effect of composition on the q* was
depicted in Fig. 7. In the case of pure IrO2 and RuO2, q* was 30.4
and 21 mC cm−2, respectively. This was consistent with the data
obtained elsewhere by sol–gel method [18], although much higher
values were reached in other cases due to the experimental condi-
tions [51]. As shown in Fig. 7, the surface charge (q*) varied with
composition with a well defined maximum around 50 mol% IrO2.
A measure of the catalyst porosity could be obtained by comput-
-0.5
0.0
0.5
1.0
1.5
E vs SCE / V
Fig. 6. Cyclic voltammograms at 20 mV s−1 scan rate in 0.5 mol L−1 H2SO4 at room
temperature of IrxRu1−xO2 coated electrodes: (1) x = 0, (2) x = 1, and (3) x = 0.5.
0.10
0.08
ing q∗/qT∗ [38]. As for the IrxRu1−xO2 electrocatalyst, the higher
i
q*T
q*i
0.06
0.04
0.02
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0.00
1.0
0.8
0.6
0.4
0.2
0.0
q*i/ q*T
q*o/ q*T
0
20
40
60
80
100
0.0
0.5
1.0
1.5
2.0
2.5
3.0
nominal composition / mol Ir%
Ir/Ru molar ratio in precursor solution
Fig. 7. Variation with nominal composition of the voltammetric charge obtained in
Fig. 5. Correlation between the concentration of the starting solution (sol) and the
surface composition for IrxRu1−xO2 coating.
a 0.5 mol L−1 H2SO4.

Q. Wang et al. / Electrochimica Acta 74 (2012) 227–234
231
RuO2
1.8
Ir0.1Ru0.9O2
Ir0.3Ru0.7O2
Ir0.5Ru0.5O2
Ir0.7Ru0.3O2
IrO2
0.2
0.1
1.6
1.4
1.2
1.0
0.0
-0.1
0.8
1.0
1.2
1.4
1.6
1.8
2.0
0.0
0.2
0.4
0.6
x( IrO2)
0.8
1.0
E vs SCE / V
Fig. 8. Cyclic voltammograms in 1.7 mol L−1 N2O4/HNO3 as a function of IrO2 con-
tent in IrxRu1−xO2/Ti electrodes.
Fig. 9. Variation with nominal composition of the peak potential in Fig. 7.
the actual anodic reaction scheme was certainly not as simple as
that depicted by reaction (IV). The possible reactions are listed in
Table 1. Under practical conditions the process of Eqs. (V)–(VIII)
was slow due to the very low concentration of water and NO [8].
Moreover, no values of standard potential (Eo) have been reported
previously for Eqs. (V)–(VIII) in anhydrous HNO3. The reference val-
ues in organic or aqueous media were presented; however, it has
been observed that the Eo of these reactions depend markedly on
the solvents [53,54].
porosity was beneficial to the reactant diffusion. As shown in the
lower panel of Fig. 6, the ratios of q∗/qT∗ was around 0.2 and it
i
almost did not vary with the composition of the oxides except
for the samples with IrO2 up to 10 mol%. This was aroused by
the effect of distortion on q*. This assertion was consistent with
the fact that no difference in the SEM micrographs was observed.
These results indicated that the texture of IrxRu1−xO2 was inde-
pendent with the variation of surface compositions. Similar results
were also observed by the IrxRu1−xO2 oxides as anodic electrocat-
alyst for water electrolysis [15]. It was interesting to note that for
the IrO2 + RuO2 layers on Ti prepared by thermal decomposition
at 400 ◦C, the total charge was almost equally divided by the outer
and inner charge [51]. In the case of the nanocrystalline materials in
this work, the larger proportion of outer and more easily accessible
active sites might be the result of the numerous grain boundaries
present in the material.
NO2 → NO + (1/2)O2
(II)
(3/2)N2O4 → NO + N2O5
(III)
Therefore, according to the literature the broad wave feature
at 1.4–1.8 V in the anodic scan could be attributed to the oxida-
+
tion of NO2 derived from N2O4 to NO2 [8,53–56]. Note that a
reduction peak at ca. 1.33 V was observed and attributed to the
reduction of NO2 [8]. As shown in Fig. 8, the anodic peak for the
3.4. Electrochemical behavior in 1.7 mol L−1 N2O4/HNO3 solution
+
oxidation of NO2 was different for different IrxRu1−xO2/Ti elec-
trodes. Fig. 9 shows the anodic peak potential of oxidation of
NO2 as a function of the nominal composition. The dependence
was linear which indicated that no surface enrichment occurred.
This was consistent with the findings of Kotz and Stucki [57] in
a 0.5 mol L−1 H2SO4 solution. The anodic peak (Epa) of the oxida-
The electrode potential was an intensive quantity which did not
depend on the morphology but only on the surface proportion of
active species [52]. The open-circuit potential (Eocp) in N2O4/HNO3
solution was measured to be about 1.18 V and almost independent
of the surface composition. This indicated that the coating with
different IrxRu1−xO2 composition had a similar redox equilibrium.
The cyclic voltammograms obtained on different IrxRu1−xO2/Ti
in N2O4/HNO3 solution are shown in Fig. 8. Usually, the anodic reac-
tion scheme in electrosynthesis of N2O5 has been proposed to be
a CE mechanism (see Eq. (IV) in Table 1) [7,53] and the apparent
standard potential (Eo) for the N2O5/N2O4 or the NO2+/NO2 cou-
ples in anhydrous HNO3 was estimated at 1.42 0.02 V [7], which
was similar to the value (1.32 V) in organic solvents. In addition, the
N2O4 equilibria were affected by the water that was present and
the thermal decomposition of N2O4 (see Eqs. (II) and (III)). Thus
+
tion of NO2 to NO2 during the forward scan shifted to the lower
potential with the increase of surface IrO2 content (Epa = 1.43 V,
1.60 V and 1.48 V for IrO2, RuO2, Ir0.5Ru0.5O2, respectively), while
the reverse scan revealed an almost identical cathodic peak (Epc
=
+
1.33 V) as a result of the opposite reaction (reduction of NO2
to NO2). Meanwhile, the peak separation (ꢁE = 100 mV, 270 mV
and 150 mV for IrO2, RuO2, Ir0.5Ru0.5O2, respectively) between the
anodic and cathodic peaks was greater than the expected value
+
for a reversible redox system (59 mV), suggesting that NO2/NO2
redox couple showed a quasi-reversible behavior at IrxRu1−xO2/Ti
electrodes in N2O4/HNO3 solution and the irreversibility tended to
more and more significant with the increasing RuO2 content.
The effects of scan rate on the electrochemical properties of
Table 1
Possible anodic reactions in cyclic voltammetrogram in N2O4/HNO3 solution.
+
NO2/NO2 redox system at RuO2, IrO2 and Ir0.5Ru0.5O2 are shown
Chemical reaction
Eo vs SCE/V
Ref.
in Fig. 10. The anodic peak current (Ipa) increased with the increase
in the scan rate (v) and showed a well linear relationship with the
square root of the scan rate (v1/2) for different IrxRu1−xO2/Ti elec-
trodes. This indicated that the redox process was under diffusion
control. As N2O4 was deposited as thin layer onto the electrode
surface, the rate-determining step was probably the diffusion of
counter-ions.
+
N2O4 → 2NO2 → 2NO2 + 2e−
(IV)
(V)
(VI)
(VII)
(VIII)
1.42 0.02
1.93 0.02a
1.13a
[7]
−
+
NO3 → NO2 + (1/2)O2 + 2e−
[53]
[53]
[8]
NO → NO+ + e−
−
N2O4 + 2H2O → 2NO3 + 4H+ + 2e−
1.24b
−
NO2 + H2O → NO3 + 2H+ + e−
1.22b
[8]
a
The standard potential in sulfolane.
The standard potential in acid media.
b

232
Q. Wang et al. / Electrochimica Acta 74 (2012) 227–234
200
150
100
50
y=8.3569x-16.3233
R2=0.9943
150
bhigh
120
RuO2
90
60
blow
0
80
y=3.6725x-0.1597
R2=0.9992
60
40
20
30
IrO2
0.0
0.2
0.4
x ( IrO2)
0.6
0.8
1.0
Fig. 12. Dependence of Tafel slope on the nominal composition in the low, blow, and
high, bhigh, overpotential domains.
0
120
90
y=6.2033x-4.0568
R2=0.9995
which shows the real potential value for the oxidation of N2O4,
was given by the following equation [58–60].
60
30
0
Ru0.5Ir0.5O2
Eeff = Eappl − jRs = a + b log j
(1)
0
4
8
12
16
20
24
where Rs is the total ohmic resistance, j is the current density, Eappl
is the electrode potential, a is a constant and b is the Tafel slope. For
all the compositions investigated, Rs-values in the 9.9–16.6 ꢂ cm−2
range were obtained. After jRs correction, Tafel curve presented two
linear segments, revealing the deviation from linearity observed in
the experimental curve (raw data) is due to ohmic resistance com-
bination with changes in the electrode kinetics (e.g., change in rate
determining step and/or apparent electronic transfer coefficient
[60–62]).
Tafel slope was very sensitive to the physic-chemical nature
of the electrode/electrolyte interface [63], being affected by fac-
tors, such as: (1) oxide composition; (2) overpotential domain;
(3) electrode preparation conditions; (4) electrolyte composition
[50,63,64].
v
1/2 / mV1/2 s-1/2
Fig. 10. Plot of N2O4 oxidation peak current on the IrxRu1−xO2/Ti electrodes vs v1/2
.
The electrocatalytic activities of the various oxide materials
for oxidation of N2O4 were evaluated by potentiostatic polariza-
tion curves under quasi-steady state conditions. Each potential
scan (backward/forward) was recorded twice without interrup-
tion in order to eliminate the hysteresis phenomenon, sometimes
observed in the electrode response. Fig. 11 shows a representative
Tafel plot before (raw data) and after ohmic drop (jRs) correction
carried out using EIS measurement. The effective potential, Eeff
,
The Tafel slope data, b, determined in the low and high over-
potential domains as function of surface composition are shown
in Fig. 12. The electrocatalytic activity was clearly different for
the IrxRu1−xO2/Ti electrodes. As shown in Fig. 3, the different
IrxRu1−xO2/Ti electrodes had a similar surface morphology. There-
fore, Tafel data revealed the influence of surface compositions on
N2O4 oxidation kinetics and these are strongly dependent on the
overpotential domains. This behavior was consistent with the lit-
erature’s findings on oxide electrodes with Ru or Ir as the active
component for OER [31,60,65]. The best electrocatalytic activity
for the Ir0.5Ru0.5O2 coating was confirmed from the lowest slope
of Tafel plots, 35.7 mV, in the low current densities. This was in
agreement with the results from H2SO4 solution. In addition, pure
IrO2 showed a much higher electrocatalytic activity for oxidation of
N2O4 than RuO2 in the low current densities. In the high overpoten-
tial domain, Tafel slope data, bhigh, were strongly influenced by the
Ir content in the coating, decreasing from 155.2 mV to 102.0 mV
with the increase of Ir content. This revealed that the electrode
kinetics were strongly dependent on the surface composition. Fur-
ther research on the mechanism and kinetic for oxidation of N2O4
can help to understand this process and will be a part of our further
work.
2.2
2.0
1.8
raw data
1.6
bhigh
1.4
blow
1.2
-3.0
-2.5
-2.0
-1.5
log ( j / A cm-2)
Fig. 11. Representative Tafel plot of the oxidation of N2O4. Electrode: Ir0.5Ru0.5O2.

Q. Wang et al. / Electrochimica Acta 74 (2012) 227–234
233
Table 2
Effects of IrxRu1−xO2/Ti electrodes on the electrosynthesis of N2O5.
Electrodes
Finalanolyteconcentration(wt%)
Current efficiency (%)
Special energy (kWh kg−1 N2O5)
ꢁN2O5/ꢁN2O4 molar ratio
N2O4
N2O5
Ir0.1Ru0.9O2/Ti
Ir0.3Ru0.7O2/Ti
Ir0.5Ru0.5O2/Ti
Ir0.7Ru0.3O2/Ti
IrO2/Ti
8.8
8.3
6.8
9.5
8.7
14
16
22
16
16
64
70
88
82
68
0.97
0.88
0.71
0.76
0.92
1.03
1.18
1.35
1.07
1.05
3.5. Electrosynthesis of dinitrogen pentoxide
Acknowledgments
The electrolysis experiments with different IrxRu1−xO2/Ti elec-
trodes were carried out. The results of the electrolysis experiments
are summarized in Table 2. With the increase of iridium content
the current efficiency increased firstly and then decreased, while
the special energy showed a contrary variation. The molar ratio of
production of N2O5 and consumption of N2O4 (ꢁN2O5/ꢁN2O4),
which indicated the coefficient of N2O4 conversion in N2O5, were
obtained for different electrodes and the results are also listed in
Table 2. The ratio of ꢁN2O5/ꢁN2O4 was much lower than the the-
oretical value of two part due to the effects of the diffusion of N2O4
across separator and a loss of N2O4 or its equivalent ionic species
[7,37]. Moreover, the ratio of ꢁN2O5/ꢁN2O4 also varied with the
composition of the IrxRu1−xO2 coatings. The surface coating of
Ir0.5Ru0.5O2 had a maximum of current efficiency and coefficient
of N2O4 conversion and a minimum of special energy. In this case,
the current efficiency is 88% and the minimal special energy is
0.70 kWh kg−1 N2O5. These results showed that the Ru0.5Ir0.5O2/Ti
electrode had the best performance in electrosynthesis of N2O5
from N2O4.
The financial supports from the National Natural Science
Foundation of China-NSAF (granted No. 10976014) is gratefully
acknowledged. The authors also thank asst. Prof. Liang Xinhua (Mis-
souri University of Science and Technology, USA) for his helpful
discussion.
References
[1] J.P. Agrawal, R.D. Hodgson, Organic Chemistry of Explosive, John Wiley Sons,
Ltd., 2007.
[2] J.P. Agrawal, High Energy Materials Propellants, Explosives and Pyrotechnics,
John Wiley Sons, 2010.
[3] C.A.C. Sequeira, D.M.F. Santos, Journal of the Brazilian Chemical Society 20
(2009) 387.
[4] T.E. Devendorf, J.R. Stacy, in: L.F. Albright, R.V.C. Carr, R.J. Schmitt (Eds.), Nitra-
tion: Recent Laboratory and Industrial Developments, ACS Symposium Series,
vol. 623, 1996 (Chapter 8).
[5] R.W. Millar, A.W. Arber, R.M. Endsor, J. Hamid, M.E. Colclough, Journal of Ener-
getic Materials 29 (2011) 88.
[6] J.E. Harrar, R. Quong, L.P. Rigdon, R.R. McGuire, US Patent 6,200,456 (2001).
[7] J.E. Harrar, R. Quong, L.P. Rigdon, R.R. McGuire, Journal of the Electrochemical
Society 144 (1997) 2032.
[8] Yu.M. Kargin, A.A. Chichirov, A.N. Ustyugov, N.M. Khusaenov, N.R. Sagdiev, A.M.
Dimiev, Russian Journal of Electrochemistry 29 (1993) 217.
[9] A.A. Chichirov, V.A. Belonogov, Yu.M. Kargin, Russian Journal of General Chem-
istry 62 (1992) 1948.
4. Conclusions
[10] G.E.G. Bagg, D.A. Salter, A.J. Sanderson, US Patent 5,128,001 (1992).
[11] G.E.G. Bagg, US Patent 5,181,996 (1993).
[12] R.J. Marshall, D.J. Shiffrin, F.C. Walsh, G.E.G. Bagg, US Patent 5,120,408 (1992).
[13] S. Trasatti, Electrochimica Acta 45 (2000) 2377.
[14] E. Guerrini, S. Trasatti, Russian Journal of Electrochemistry 42 (2006) 1017.
[15] J.B. Cheng, H.M. Zhang, G.B. Chen, Y.N. Zhang, Electrochimica Acta 54 (2009)
6250.
[16] T. Arikawa, Y. Takasu, Y. Murakami, K. Asakura, Y. Iwasawa, Journal of Physical
Chemistry B 102 (1998) 3736.
The IrxRu1−xO2/Ti electrodes (x = 0–1) prepared by thermal
decomposition of non-aqueous solutions of H2IrCl6 and RuC13 pos-
sessed very fine crystallites structure with a range of particle size
of 5–10.1 nm. The particle size decreased and then increased with
a minimum value at x = 0.5. No enrichment of Ru or Ir in the surface
was observed.
The cyclic voltammetry characterization of the IrxRu1−xO2/Ti
electrodes in a 0.5 mol L−1 H2SO4 solution showed that the sur-
face charge (q*) varied with composition. The IrxRu1−xO2 coating
with x = 0.5 had the maximum surface charge. The coating porosity
was around 0.2 and did almost not vary with the surface compo-
sition of the oxide except for the samples with IrO2 up to 10 mol%.
The texture of IrxRu1−xO2 was independent with the surface com-
position. The cyclic voltammetry of the IrxRu1−xO2/Ti electrodes in
N2O4/HNO3 solution indicated that the oxidation of NO2 derived
[17] M.V. ten Kortenaar, J.F. Vente, D.J.W. Ijdo, S. Muller, R. Kotz, Journal of Power
Sources 56 (1995) 51.
[18] F.I. Mattos-Costa, P. de Lima-Neto, S.A.S. Machado, L.A. Avaca, Electrochimica
Acta 44 (1998) 1515.
[19] A. Di Blasi, C. D’Urso, V. Baglio, V. Antonucci, A.S. Arico, R. Ornelas, F. Matteucci,
G. Orozco, D. Beltran, Y. Meas, L.G. Arriaga, Journal of Applied Electrochemistry
39 (2009) 191.
[20] Y. Takasu, N. Yoshinaga, W. Sugimoto, Electrochemistry Communications 10
(2008) 668.
[21] M.E.G. Lyons, S. Floquet, Physical Chemistry Chemical Physics 13 (2011) 5314.
[22] A.T. Marshall, R.G. Haverkamp, Electrochimica Acta 55 (2010) 1978.
[23] E. Ratsen, G. Hagen, R. Tunold, Electrochimica Acta 48 (2003) 3945.
[24] S. Song, H. Zhang, X. Ma, Z. Shao, R.T. Baker, B. Yi, International Journal of
Hydrogen Energy 33 (2008) 495.
+
from N2O4 to NO2 took place at 1.4–1.8 V. The redox process of
+
NO2 to NO2 was under diffusion control.
[25] T.C. Wen, C.C. Hu, Journal of the Electrochemical Society 139 (1992) 2158.
[26] H.A. Hansen, I.C. Man, F. Studt, F. Abild-Pedersen, T. Bligaard, J. Rossmeisl, Phys-
ical Chemistry Chemical Physics 12 (2010) 283.
[27] M. Zhou, L. Liu, Y. Jiao, Q. Wang, Q. Tan, Desalination 277 (2011) 201.
[28] S. Trasatti, Electrochimica Acta 29 (1987) 1503.
[29] L.A. da Silva, V.A. Alves, M.A.P. da Silva, S. Trasatti, J.F.C. Boodtst, Electrochimica
Acta 42 (1997) 271.
[30] C.P. De Pauli, S. Trasatti, Journal of Electroanalytical Chemistry 538–539 (2002)
145.
[31] J. Xu, G. Liu, J. Li, X. Wang, Electrochimica Acta 59 (2012) 105.
[32] A.T. Marshall, S. Sunde, M. Tsypkin, R. Tunold, International Journal of Hydrogen
Energy 32 (2007) 2320.
[33] J. Ribeiro, F.L.S. Purgato, K.B. Kokoh, J.M. Léger, A.R. De Andrade, Electrochimica
Acta 53 (2008) 7845.
[34] X.M. Wang, J.M. Hu, J.Q. Zhang, Electrochimica Acta 55 (2010) 4587.
[35] A. Benedetti, P. Riello, G. Battaglin, A. De Battisti, A. Barbieri, Journal of Electro-
analytical Chemistry 376 (1994) 195.
The electrocatalytic activities for N2O4 oxidation were deter-
mined by potentiostatic polarization curves. Tafel data revealed
that the influence of surface compositions on the oxidation kinet-
ics of N2O4 strongly depended on the overpotential domains. The
lowest slope, 35.7 mV, was obtained from the Tafel plots of the
Ir0.5Ru0.5O2 coating in the low current densities, indicating that
this coating had excellent electrocatalytic activity for N2O4 oxi-
dation. In addition, iridium oxide had a better electrocatalytic
performance than ruthenium oxide for N2O4 oxidation. The exper-
iments of electrosynthesis of N2O5 from N2O4 also indicated that
the Ru0.5Ir0.5O2 coating had the best electrocatalytic performance
with current efficiency of 88% and a special energy of 0.70 kWh kg−1
N2O5.

234
Q. Wang et al. / Electrochimica Acta 74 (2012) 227–234
[36] Z.G. Ye, H.M. Meng, D. Chen, H.Y. Yu, Z.S. Huan, X.D. Wang, D.B. Sun, Solid State
Science 10 (2008) 346.
[51] I.M. Kodintsevt, S. Trasatti, M. Rubel, A. Wieckowski, N. Kaufher, Langmuir 8
(1992) 283.
[37] Q. Wang, M. Su, X. Zhang, L. Wang, J. Wang, Z. Mi, Electrochimica Acta 52 (2007)
3667.
[52] V.A. Alves, L.A.D. Silva, J.F.C. Boodts, S. Trasatti, Electrochimica Acta 39 (1994)
1585.
[38] C.P. De Pauli, S. Trasatti, Journal of Electroanalytical Chemistry 396 (1995)
161.
[39] W. Hume-Rothery, G. Raynor, The Structure of Metals and Alloys, Institute of
Metals, London, 1954.
[40] E. Balko, C. Davidson, Journal of Inorganic and Nuclear Chemistry 42 (1980)
1778.
[41] D.T. Shieh, B.J. Hwang, Electrochimica Acta 38 (1993) 2239.
[42] L. Lipp, D. Pletcher, Electrochimica Acta 42 (1997) 1091.
[43] T.A.F. Lassali, L.O.S. Bulhoes, L.M.C. Abeid, J.F.C. Boodts, Journal of the Electro-
chemical Society 144 (1997) 3348.
[53] A. Boughriet, M. Wartel, Electrochimica Acta 36 (1991) 889.
[54] K.Y. Lee, C. Amatore, J.K. Kochi, The Journal of Physical Chemistry 95 (1991)
1285.
[55] A. Boughriet, M. Wartel, J.C. Fischer, C. Bremard, Journal of Electroanalytical
Chemistry 190 (1985) 103.
[56] A. Boughriet, M. Wartel, Journal of the Chemical Society: Chemical Communi-
cations (1989) 809.
[57] R. Kotz, S. Stucki, Journal of the Chemical Society 132 (1985) 103.
[58] H. Ma, C. Liu, J. Liao, Y. Su, X. Xue, W. Xing, Journal of Molecular Catalysis A:
Chemical 247 (2006) 7.
[44] F. Montilla, E. Morallón, A. De Battisti, J.L. Vázquez, Journal of Physical Chemistry
B 108 (2004) 5036.
[45] B. Correa-Lozano, C. Comninellis, A. De Battisti, Journal of Applied Electrochem-
istry 26 (1996) 83.
[46] C. Angelinetta, S. Trasatti, L.D. Atanasoska, R.T. Atanasoski, Journal of Electro-
analytical Chemistry 214 (1986) 535.
[47] J. Gaudet, A.C. Tavares, S. Trasatti, D. Guay, Chemistry of Materials 17 (2005)
1570.
[48] S. Ardizzone, G. Fregonara, S. Trasatti, Electrochimica Acta 35 (1990)
263.
[49] G. Lodi, G. Zucchini, A. De Battisti, E. Sivieri, S. Trasatti, Materials Chemistry 3
(1978) 179.
[59] E. Tsuji, A. Imanishi, K. Fukui, Y. Nakato, Electrochimica Acta 56 (2011) 2009.
[60] K.C. Fernandes, L.M. Da Silva, J.F.C. Boodts, L.A. De Faria, Electrochimica Acta 51
(2006) 2809.
[61] L.A. De Faria, J.F.C. Boodts, S. Trasatti, Journal of Applied Electrochemistry 26
(1996) 1195.
[62] L.M. Da Silva, K.C. Fernandes, L.A. De Faria, J.F.C. Boodts, Electrochimica Acta 49
(2004) 4893.
[63] S. Trasatti (Ed.), Electrodes of Conductive Metallic Oxides. Parts A and B, Else-
vier, Amsterdam, 1981.
[64] L.M. Da Silva, D.V. Franco, L.A. De Faria, J.F.C. Boodts, Electrochimica Acta 49
(2004) 3977.
[65] J.M. Hu, J.Q. Zhang, C.N. Cao, International Journal of Hydrogen Energy 29 (2004)
791.
[50] S. Trasatti, Electrochimica Acta 36 (1991) 225.