Electrochemical synthesis of N2O5 by oxidation of N2O4 in nitric acid with PTFE membrane

Article abstract of DOI:10.1016/j.electacta.2006.10.034

Electrochemical synthesis of dinitrogen pentoxide (N₂O₅) by oxidation of dinitrogen tetroxide (N₂O₄) in a plate-and-frame electrolyzer was investigated. As the separator, different porous polytetrafluoroethylene (PTFE) membranes were tested in this process and the effects of hydrophilicity and of hydrophobicity on the electrolysis were discussed. The transport of N₂O₄ and water from catholyte to anolyte through membrane occurred in the electrolysis, especially at the end of the electrolysis. The water transport had a much more effect on the electrolysis than that of the N₂O₄ diffusion. The hydrophobic PTFE membranes had better performance on control of water transport from catholyte to anolyte than that of the hydrophilic ones. Hydrophobicity can increase the chemical yield of N₂O₅. The membranes with a low hydrophobic surface were preferred. All the hydrophobic PTFE membranes with low resistance have the specific energy of 1.1-1.5 kWh kg^-1 N₂O₅. The current efficiency of 67.3-80.2% and chemical yield of 58.9-60.9% were achieved in production of N₂O₅. The technique of replacing the catholyte with fresh nitric acid can minimize the transport of N₂O₄ and water to a great extent, it can further improve the chemical yield and reduce the specific energy.

Full text of DOI:10.1016/j.electacta.2006.10.034

Electrochimica Acta 52 (2007) 3667–3672
Electrochemical synthesis of N2O5 by oxidation of N2O4 in
nitric acid with PTFE membrane
Qingfa Wang, Min Su, Xiangwen Zhang∗, Li Wang,
Jixiao Wang, Zhentao Mi
Key Laboratory for Green Chemical Technology of State Education Ministry, School of Chemical Engineering & Technology,
Tianjin University, Tianjin 300072, PR China
Received 2 August 2006; received in revised form 15 October 2006; accepted 19 October 2006
Available online 21 November 2006
Abstract
Electrochemical synthesis of dinitrogen pentoxide (N2O5) by oxidation of dinitrogen tetroxide (N2O4) in a plate-and-frame electrolyzer was
investigated. As the separator, different porous polytetrafluoroethylene (PTFE) membranes were tested in this process and the effects of hydrophilic-
ity and of hydrophobicity on the electrolysis were discussed. The transport of N2O4 and water from catholyte to anolyte through membrane occurred
in the electrolysis, especially at the end of the electrolysis. The water transport had a much more effect on the electrolysis than that of the N2O4
diffusion. The hydrophobic PTFE membranes had better performance on control of water transport from catholyte to anolyte than that of the
hydrophilic ones. Hydrophobicity can increase the chemical yield of N2O5. The membranes with a low hydrophobic surface were preferred. All the
hydrophobic PTFE membranes with low resistance have the specific energy of 1.1–1.5 kWh kg−1 N2O5. The current efficiency of 67.3–80.2% and
chemical yield of 58.9–60.9% were achieved in production of N2O5. The technique of replacing the catholyte with fresh nitric acid can minimize
the transport of N2O4 and water to a great extent, it can further improve the chemical yield and reduce the specific energy.
© 2006 Elsevier Ltd. All rights reserved.
Keywords: Dinitrogen pentoxide; Electrochemical synthesis; Porous PTFE membrane; Dinitrogen tetroxide; Hydrophobicity
1. Introduction
pounds, alkanes, alkenes, alcohols, sugars, oximes, amines and
amides in solution or in gas phase [4].
Nitration products have many applications in pharmaceu-
tical, agricultural, pest control, pigment and polyurethane or
polyamide products as well as in high-energy materials [1,2].
Conventional nitrating agents normally involved nitric acid
mixed with other reagents such as sulphuric acid in different
proportions. During this nitration process, undesirable reactions
and serious pollution often occurred. In addition, conventional
nitration agents cannot be used selectively for nitration and for
deactivated substrates. In contrast to the conventional nitrating
agents, dinitrogenpentoxide(N2O5)hasseveraladvantagessuch
as simple isolation of the product, easy control of temperature
and absence of spent acid for disposal [3]. N2O5 is an alterna-
tive nitrating agent. N2O5 has been proven to be very useful in
the nitration of a variety of substrates including aromatic com-
The conventional methods of N2O5 preparation are based
on the dehydration of white fuming HNO3 with phosphorous
pentoxide [5], or trifluoracetic anhydride [6]. These methods
are costly unacceptable for scale-up due to their harsh opera-
tion conditions, low yield and poor product quality [7]. Pure
N2O5 can also be synthesized by oxidation of dinitrogen tetrox-
ide (N2O4) with ozone [8–10]. This method is of limited value
owing to the capability of ozone generator. Furthermore, the
low temperature operation of N2O5 trap would also increase the
cost of this process. A potentially convenient method is oxida-
tion of N2O4 electrochemically, which is a friendly route firstly
described by Anon. in 1910 [11]. Much work on this process has
been done in recent years to improve the current efficiency and
the chemical yield of N2O5 product by a controlled-potential
technique [12–20,24,25], as well as to reduce the contamina-
tion of N2O4 by a multistage electrolysis method [21–23]. In
these processes, ion exchange membrane, porous ceramic or
glass were used as the separator. When the ion exchange mem-
∗
Corresponding author. Tel.: +86 22 27892340; fax: +86 22 27402604.
E-mail address: qingfaw@yahoo.com.cn (X. Zhang).
0013-4686/$ – see front matter © 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2006.10.034

3668
Q. Wang et al. / Electrochimica Acta 52 (2007) 3667–3672
Table 1
The electrochemical characteristics of the PTFE membranes in the electrosynthesis of N2O5
The membranesa
Supplier
Surface contact
angle of water (◦)
Rcell
,
Analyte (%)
Chemical
yield (%)
Current
efficiency (%)
Special energy
(kWh/kg N2O5)
(ꢀ cm−2
)
N2O5
N2O4
GA
GD
GL
GS
BS01
BS01/GS
Gore company
Gore company
Gore company
Gore company
BPRIb
135.9 2.9
96.6 5.0
73.1 2.1
65.0 2.5
125.8 1.6
106.5 2.0
34.0
24.7
30.3
26.6
26.4
34.3
16.6
25.2
17.9
9.4
26.1
24.6
17.1
13.1
8.8
10.8
11.7
13.4
40.56
58.90
37.24
37.72
50.60
60.90
51.2
80.2
35.9
40.8
59.1
67.3
1.5
1.1
1.9
1.5
1.2
1.3
–
a
The thickness of the membranes was 0.08 mm except BS01/GS membrane (0.14 mm).
Beijing Plastics Research Institute.
b
brane is used, the diffusion of N2O4 and the electrodialysis of
+
ions such as NO+ and NO2 (derived from N2O4 and N2O5,
respectively), aswellasthesolutionosmoticandelectro-osmotic
of water occurred and thus, reducing the chemical yield of
N2O5. Porous ceramic and glass are too fragile to be scaled
up. Harrar et al. [24] first adopted the polytetrafluoroethylene
(PTFE) membrane as the separator for electrochemical oxida-
tion of N2O4 to prepare N2O5. Compared with the ion exchange
membranes, the PTFE membrane that Harrar et al used had a
similar performance with the anion membrane, which was con-
sidered as the best cell separator for this process [24,25]. But the
PTFE membrane used in literature was loaded with inorganic
oxides for hydrophilicity, which might enhance the transport
of water. What influence the physicochemical properties of the
PTFE membrane would have on the electrolysis process? To
our knowledge, the work on this has not been reported up to
now.
In present work, the effect of PTFE membranes as a sep-
arator on the electrochemical synthesis of N2O5 by oxidizing
N2O4 was investigated. Various PTFE membranes were tested.
The hydrophobility and hydrophilicity of the PTFE membrane
was considered seriously. The GL membrane used in this work
(Table 1) almost has the same performance as that reported in
reference [24].
Fig. 1. The schematic diagraph of N2O5 electrosynthesis system.
are depicted as
Anode reaction
N2O4 + 2HNO3 → 2N2O5 + 2H++2e−
(1)
Cathode reaction
2HNO3 + 2H+ + 2e− → 2N2O4
+ 2H2O
2. Experimental
(2)
(3)
Nitric acid used in the present study (commercial-grade,
manufactured by Shandong Chemical Fertilizer Company, PR
China) was 98% concentration, containing 1.7–1.9 wt% H2O
and 0.2–0.3 wt% N2O4. This acid was used directly as the
catholyte. The initial anolyte solution of 25–33% N2O4 in nitric
acid was prepared by adding N2O4 (CP grade) into the cooled
nitric acid. The membranes used as separator were listed in
Table 1. All the membranes were used as received except for
BS01/GS membrane, which was physically prepared with BS01
membrane as top layer and GS membrane as support layer.
Before use, each membrane was immersed in the N2O4/HNO3
solution for one day.
Overall cell reaction
2HNO3 → N2O5 + H2O
The IrO2/Ti and RuO2-IrO2/Ti, prepared by thermal decom-
position method according to reference [26], were used as the
anode and cathode, respectively. The apparent electrode surface
area in this electrolysis was 130 cm2. Prior to the experiments,
thetwocompartmentsoftheelectrolyzerwererinsedbycirculat-
ing 98% nitric acid for 10 min. 98% nitric acid was pumped into
the reservoirs and N2O4 was added into the anolyte. The tested
solutions were cooled to 10 ◦C and pumped into the electrolyzer.
Then the electrolysis was started. The chiller temperature was
adjusted to maintain the electrolyte between 10 and 15 ◦C. And
the applied voltage (2–3 V) to the electrolyzer was adjusted to
give the required current density (600–1000 A m−2). The solu-
tion circulation rate was kept at least at 20 mL/min to minimize
The scheme of electrosynthesis of N2O5 was shown in Fig. 1.
Theelectrolysiswascarriedoutinaplate-and-frameelectrolyzer
with a volume of 240 mL. The basic cell reactions of this process

Q. Wang et al. / Electrochimica Acta 52 (2007) 3667–3672
3669
Table 2
the solution electronic resistance. The electrolysis was carried
The XPS result of different membranes
out for 7 h. During the electrolysis, samples were taken from
the anolyte compartment at different intervals. The voltage, cur-
rent and temperature were recorded. The concentration of N2O4
was determined by redox titration method according to refer-
ence [27], and the concentration of N2O5 present in the anolyte
was determined using NMR spectrometry [28].
The membranes
Chemical composition of surface (%)
C 1s
F 1s
O 1s
Si 2p
GL
GD
BS01
29.5
37.4
37.0
35.3
61.8
62.2
25.7
0.8
0.8
9.5
0
0
The film bond configuration was obtained using Fourier
transform infrared spectrum analyzer (Nicolet 560 FTIR) with
an attenuated total reflection element (ATR) in the wave
numbers range of 600–4000 cm−1. For the composite mem-
brane, the structure of top-layer was characterized. The surface
hydrophilicity of the PTFE membrane was conducted by the sur-
face contact angle of water measured in air at room temperature
using the anglemeter (Dataphysics, DCAT21, Germany). X-
ray photoelectron spectroscopy (XPS) analysis was conducted
using a Perkin–Elmer PHI 1600 ESCA (USA) with a Mg K␣
X-ray source (1253.6 eV) in fixed analyzer transmission (FAT)
mode. The background pressure of the analytical chamber was
267 × 10−8 Pa (2.0 × 10−8 Torr) and the take-off angle of pho-
toelectrons was 90◦ with respect to the sample surface.
peaks might be attributed to the stretching vibrations of car-
bonyl group and SiO2-group, respectively, further identified by
the XPS measurement (as shown in Table 2). These hydrophilic
functional groups could be incorporated to improve the PTFE
polymer surface properties by chemical modifications such as
grafting, cross-linking or adding inorganic reinforcements into
membranes [33–36]. This observation corresponded well with
the result shown in Table 1.
3.2. Performance of PTFE membrane and the electrodes
The cell performances with different PTFE membranes were
investigated as shown in Fig. 3. The typical current–voltage
curve of the electrolysis appeared when the applied voltage
was gradually increased. The current density was small until
the applied voltage slightly above 1.0 V and thereafter the cur-
rent density increased more rapidly. The small current was
caused by diffusion of ions away from the electrodes and a
certain amount of external electric energy was needed to coun-
terbalance this diffusion and to keep the electrodes polarized
[37]. The cell voltage was composed of the anode overpoten-
tial, the cathode overpotential, equilibrium potential and the
Ohmic drop. During electrolysis, the theoretical equilibrium
potential was about 0.37 V and the Ohmic drop contributed to
more than half of the cell voltage [24,38]. With different mem-
branes as the separator, the cell voltage at electrolysis current
density of 600 A m−2 was not more than 3.0 V, which indi-
cates that IrO2 and RuO2-IrO2 coating had a relatively low
3. Results and discussion
3.1. Characterization of the PTFE membranes
The contact angle is an important parameter in surface sci-
ences. It is a regular measurement of the surface hydrophobicity
[29]. Table 1 summarized the values of the contact angle of dif-
ferent membranes. The contact angle for pure PTFE membrane
coincides relatively well with those reported by other authors
[30,31], and denotes the hydrophobic character of PTFE related
to the interaction of Van der Waal. The surface structure of
different PTFE membranes was further characterized by FTIR-
ATR measurement. Fig. 2 shows the FTIR-ATR spectrum of the
PTFE membranes. All the samples showed two strong absorp-
tion peaks at 1210 and 1150 cm−1 due to the C-F asymmetric
and symmetrical stretching vibrations in the PTFE, respectively
[32]. Some new peaks at 1500–1700 cm−1 for GD membrane
and at 1080 cm−1 for GL membrane were observed. These
,
Fig. 3. The voltage–current density character of the electrolyzer with different
PTFE membranes as the separator, the IrO2/Ti and RuO2-IrO2/Ti as the anode
and cathode, respectively.
Fig. 2. The FTIR-ATR spectrum of PTFE membranes.

3670
Q. Wang et al. / Electrochimica Acta 52 (2007) 3667–3672
overpotential and had good electrocatalytic activity for the elec-
trode reaction. At the same current density, the difference of
the cell voltage was attributed to the diaphragm Ohmic drop.
It can be seen that the order of diaphragm resistance was
GA > BS01/GS > GD > BS01 > GL > GS. And the hydrophilic
membranes had a relative low electronic resistance. Above
400 A m−2, the current density increased linearly with the
increase in cell voltage. The overpotentials changed very little
from 400 to 800 A m−2 (about 0.25 V for the anode and cathode
according to the Butler–Volmer equation [39,40]). Therefore,
the overall process resistance (Rcell) could be obtained from the
linear relationship of the cell voltage versus current density. The
results were listed in Table 1. The overall process resistance
was considered to the sum of diaphragm resistance, electrolyte
resistance and contact resistance [24,39–41]. For the experimen-
tal PTFE membranes, the overall resistance varied from 24.7
to 34.3 ꢀ cm−2. The solution resistance mainly contributed to
the cell resistance in this electrolysis system [24]. This indi-
cates indirectly that the resistance of the membranes used in this
electrolysis was very low and had a good electric characteristic.
Fig. 4. The rate of production of N2O5 and consumption of N2O4 with the
consumption of electricity quantity in the electrolysis. (A) GL membrane as the
separator, (B) GD membrane as the separator.
solution resistance was still very low. Thus, the increase of spe-
cial energy was not so significant.
3.3. Effect of hydrophobicity of PTFE membrane on water
transport
3.4. Effect of hydrophobicity of PTFE membrane on N2O4
transport
The experimental results of the electrolysis with the dif-
ferent PTFE membranes as the separator were summarized in
Table 1. The hydrophobic membrane had a better performance
than the hydrophilic membrane had on restraining the transport
of water. The smaller the contact angle of water on the poly-
mer material, the better its wettability. The water permeability
would increase proportionally with the membrane wettability.
In the electrolysis, water was not expected in the anolyte since
it would react with the product of N2O5 to form nitric acid.
Water can be transported through membrane by the osmotic flow
From Table 1, it can also be seen that the final N2O4 concen-
tration in anolyte was still very high. This might be attributed to
the diffusion of N2O4 from the cathode to anode. For the pure
PTFE membrane, the surface tension was 18–20 mN m−1 and
it could be increased by chemical modification [42]. The sur-
face tension of N2O4 was 26.5 mN m−1[43], which is similar to
many of the more volatile organic solvents and close to that of
the pure PTFE membrane. Therefore, N2O4 molecules can dif-
fuse relatively easily through the PTFE membrane. Figs. 4 and 5
illustrated the N2O4 diffusion effect and other features of this
electrolysis. As shown in Fig. 4, the slope of the N2O4 versus
−
and “electro-osmotic” drag by the NO3 ions. The hydrophilic
membranes such as GL and GS membranes have low contact
angles of water and high surface tension. They will be swollen
in the solution when water existed. The swollen membranes
usually loose their permselectivity and suffer from poor stabil-
ity. Therefore, the osmotic flow of water was enhanced when
the hydrophilic membranes were used. Then the chemical yield
and the current efficiency were reduced and the specific energy
increased. The hydrophobic membranes such as GA, GD, BS01
and BS01/GS membranes had high contact angles of water and
the surface of these membranes were in a low-energy state. As
a result, the transport of water was inhibited to a great extent
and the hydrophobic membrane had a better performance in
this electrolysis than that of hydrophilic membrane. However,
if the hydrophobicity of the membrane was too high, the repul-
sive force to the inorganic species except for water was also
strengthened due to the interaction of Van der Waal. Thus, the
main charge transport through the membrane by NO3− ion was
restrained under the same cell voltage and the current decreased
for GA membrane. Compared with the GD, BS01 and BS01/GS
membranesasseparator, theextentofelectrolysiswithGAmem-
brane was not completely carried out in the same reaction time.
The N2O4 in catholyte was almost completely ionized and the
Fig. 5. The rate of production of N2O5 with the consumption of N2O4 during the
electrosynthesis of N2O5. (A) GL membrane as the separator, (B) GD membrane
as the separator.

Q. Wang et al. / Electrochimica Acta 52 (2007) 3667–3672
3671
charge curve (−0.98 for GL membrane and −0.73 for GD mem-
brane) was more negative initially than the theoretical one (the
theoretical slope was −0.5 for 100% current efficiency), sug-
gesting that the loss of N2O4 from the anolyte occurred. The
hydrophlilic GL membrane has a more negative slope than the
hydrophobic GD membrane has. This might be attributed to the
electrodialysis of NO2+ ions. At low concentration, N2O4 could
be almost completely ionized. At higher concentrations and in
the presence of water, ionization was suppressed to approxi-
mately 20% [24]. In the final stage of the electrolysis, water in
the catholyte gradually increased (see the cathode reaction 2).
Thus, the concentration of molecular N2O4 in the catholyte was
much higher than its concentration in the anolyte. The diffusion
of molecular N2O4 from catholyte to anolyte makes an upward
curvature in the N2O4 curve. As shown in Fig. 5, the slope of
N2O5 versus N2O4 (1.1 for GL membrane and 1.4 for GD mem-
brane) was much lower than the theoretical one (the theoretical
slope was 2) for the overall electrolysis, which was partially
attributed to the transport of N2O4 from catholyte to anolyte.
diffusion of N2O4 can increase the apparent chemical yields the
curve of N2O5 versus the charge had a significant downward
curvature (Fig. 4), further indicating that the transport of water
through the membrane was the main mass transport. The slope of
N2O5 versus N2O4 for GD membrane was much higher than that
for GL membrane (Fig. 5). This could be attributed to the trans-
port of N2O4 and water from catholyte to anolyte. As described
above, the transport of water was the main reason to cause the
deviation from the theoretical slope.
For the overall process, the diffusion of N2O4 from catholyte
to anolyte was unfavorable because it will increase the electroly-
sis time to meet with the low concentrations of N2O4 in the N2O5
product solutions. In the present work, it was concluded that,
when the catholyte was replaced by fresh nitric acid after N2O4
concentration was above 20 wt%, the diffusion effect of N2O4
could be neglected. On the other hand, the transport of water
from catholyte to anolyte was also minimized. Thus, the tech-
nique of replacing the catholyte during the process will lower the
N2O4 concentration in the anolyte more quickly and improve the
chemical yield of N2O5. It was also observed that these trans-
ports through membrane could be partly inhibited by using a
thicker membrane. Of course, the membrane resistance would
also increase so as to increase the special energy. In the elec-
trochemical synthesis of N2O5 by oxidation of N2O4 in nitric
acid, the membranes used as separator must have not only suf-
ficient chemical stability and mechanical strength to withstand
the hostile environment, but also a low voltage drop in order
to minimize electrical power consumption at any given current
density. Therefore, it is not a better way to restrain the transports
of water and N2O4 by using a thicker membrane.
3.5. Different transports process in the electrolysis
Fig. 4 shows that there were two distinct electrolysis periods,
during which the chemical reactions, especially in the anolyte,
and the cell transport phenomena were likely to be somewhat
different. The first period (the quantity of electricity below
0.26 F for GL membrane, and 0.23 F for GD membrane dur-
ing the electrolysis) was the beginning of the electrolysis when
water was present in the anolyte, and second was the period
after the water was consumed. The division point for GD mem-
brane was lower than that for the GL membrane, which is partly
attributed to the osmotic of water. In the first stage of the elec-
trolysis, the osmotic of water could be neglected by the lower
differential concentration between the anolyte and catholyte for
the hydrophobic membranes. But it cannot be ignored for the
hydrophilic membranes. In the second stage of electrolysis,
because of the increase of water in the catholyte, this phe-
nomenonbecamesignificantattheendoftheelectrolysis. Forthe
GD membrane, the incremental current efficiency for the forma-
tion of N2O5, after the anolyte water was consumed, was almost
100% for most of this electrolysis (Fig. 4). This indicated that the
4. Conclusions
TheelectrosynthesisofN2O5 byoxidizingN2O4 innitricacid
was carried out. As the separator, the PTFE membrane had a low
diaphragm resistance and a good electric characteristic to obtain
high current efficiency and chemical yield of N2O5 as well as to
reduce the specific energy in this process. The hydrophobicity
of PTFE membranes played an important role in controlling the
transport of water and N2O4. The transport of water was the
main mass transport in the electrolysis. The hydrophilic surface
favored the osmotic flow of water through PTFE membrane and
thus, lowered the chemical yield. The hydrophobic membrane
can significantly inhibit the transport of water and improve the
chemical yield and current efficiency. However, the membrane
with high hydrophobicity restrained the charge transport and
increased the special energy of the electrolysis. The membranes
with a low hydrophobic surface were preferred. The transports
through membrane can be minimized or eliminated by replacing
thecatholytewithfreshnitricacidandusingathickermembrane.
The technique of replacing the catholyte was a better solution
to improve the chemical yield and current efficiency.
+
possible losses of NO2 by electrodialysis and/or reaction with
water via nitrate ion drag, as well as the solvent oxidation side
reaction, can be minimized to achieve theoretical current effi-
ciency. For the GL membrane, the current efficiency decreased
quickly with the increase in water content in the catholyte. The
−
bulk flow of water gradually increased and the NO3 ion were
solvated by water when the catholyte contained large amount of
water. Their migration into the anolyte thereby reduced the yield
of N2O5 and increased the special energy. The slope of N2O5
versus the charge for the GD membrane was higher than that for
the GL membrane, further showing that the hydrophobic mem-
brane has a good performance on inhibiting the water transport
so as to improve the chemical yield.
Acknowledgements
From the finial anolyte composition and the chemical yield
of N2O5, it can be concluded that the effect of N2O4 diffusion
on the chemical yield is less than that of water. Although the
ThefinancialsupportsofthisworkbyCommissionofScience
Technology and Industry for National Defence (COSTIND)

3672
Q. Wang et al. / Electrochimica Acta 52 (2007) 3667–3672
of China (Granted No. A1420060192) are gratefully acknowl-
edged. The authors would like to thank Prof. B. Zhu (Royal
Institute of Technology (KTH), Sweden) for his helpful discus-
sion.
[20] Yu.M. Kargin, A.A. Chichirov, A.N. Ustyugov, N.M. Khusaenov, N.R.
Sagdiev, A.M. Dimiev, Russ. J. Electrochem. 29 (1993) 217.
[21] G.E.G. Bagg, D.A. Salter, A.J. Sanderson, U.S. Patent 5,128,001 (1992).
[22] G.E.G. Bagg, US Patent 5,181,996 (1993).
[23] R.J. Marshall, D.J. Shiffrin, F.C. Walsh, G.E.G. Bagg, US Patent 5,120,408
(1992).
References
[24] J.E. Harrar, R. Quong, L.P. Rigdon, R.R. Mcguire, J. Electrochem. Soc.
144 (1997) 2032.
[25] J.E. Harrar, R. Quong, L.P. Rigdon, R.R. Mcguire, US Patent 6,200,456
(2001).
[1] N.V. Latypov, U. Wellmar, P. Goede, A.J. Bellamy, Org. Process. Res. Dev.
4 (2000) 156.
[26] M. Blouin, D. Guay, J. Electrochem. Soc. 144 (1997) 573.
[27] R.J. Marshall, D.J. Shiffrin, F.C. Walsh, G.E.G. Bagg, European Patent
Appl. EP 0,295,878 (1988).
[2] M.A. Zhang, E. Eaton, R. Gilardi, Angew. Chem. Int. Ed. 39 (2000) 401.
[3] M.B. Talawar, R. Sivabalan, B.G. Polke, U.R. Nair, G.M. Gore, S.N.
Asthana, J. Hazard. Mater. 124 (2005) 153.
[28] J. Zhang, L. Wang, M. Su, X.W. Zhang, Z.T. Mi, Chin. J. Energ. Mater. 14
(2006) 62.
[29] L. Palacio, J.I. Calvo, P. Pradanos, A. Hernandez, P. Va¨isa¨nen, M. Nystro¨m,
J. Membr. Sci. 152 (1999) 189.
[30] C.Y. Tu, Y.C. Wang, C.L. Li, K.R. Lee, J. Huang, J.Y. Lai, Eur. Poly. J. 41
(2005) 2343.
[31] Y. Yamada, T. Yamada, S. Tasaka, N. Inagaki, Macromolecules 29 (1996)
4331.
[4] H. Feuer, A.T. Nielsen, Nitro Compounds, Purdue University Press, West
Lafayette, Indiana, 1990, pp. 268–365.
[5] T.E. Stevens, W.D. Emmons, J. Am. Chem. Soc. 79 (1957) 6008.
[6] J.H. Robson, J. Am. Chem. Soc. 77 (1955) 107.
[7] T.E. Devenelorf, J.R. Stacy, 623 ACS Symposiums Series, 1996, p. 68.
[8] G.E.G. Bagg, A.W. Arber, US Patent 5,318,768 (1994).
[9] A.W. Arber, A.C.M. Cumming, J.K. Hammond, UKPatentAppI. 2,347,670
(2000).
[32] C.-Y. Tu, Y.-L. Liu, K.-R. Lee, J.-Y. Lai, J. Membr. Sci. 274 (2006) 47.
[33] B.K. Dutta, S.K. Sikdar, Environ. Sci. Technol. 33 (1999) 1709.
[34] A.M. Polyakov, L.E. Starannikova, Y.P. Yampolskii, J. Membr. Sci. 238
(2004) 21.
[35] J. Huang, Y.C. Wang, C.L. Li, K.R. Lee, J.Y. Lai, Sep. Sci. Technol. 36
(2001) 1677.
[36] J. Huang, Y.C. Wang, C.L. Li, K.R. Lee, S.C. Fan, T.T. Wu, J.Y. Lai, Eur.
Polym. J. 38 (2002) 179.
[37] B. Zhu, I. Albinsson, C. Andersson, K. Borsand, M. Nilsson, B.E. Mellan-
der, Electrochem. Commun. 8 (2006) 495.
[10] F. Christian, J. Wieland, European Patent Appl. EP 0,983,965 (2000).
[11] Anon., German Patent 231,546 (1910).
[12] J.E. Harrar, R.K. Pearson, J. Electrochem. Soc. 130 (1983) 108.
[13] M.J. Rodgers, P.F. Swinton, 623 ACS Symposium Series, 1996, pp. 58–
67.
[14] R.R. McGuire, C.L. Coon, J.E. Harrar, R.K. Pearson, US Patent 4,432,902
(1984).
[15] C.L. Coon, J.E. Harrar, R.K. Pearson, R.R. McGuire, US Patent 4,443,308
(1984).
[16] R.R. McGuire, C.L. Coon, J.E. Harrar, R.K. Pearson, US Patent 4,525,252
(1985).
[17] Yu.M. Kargin, Y.V. Rydvanskkii, G.A. Evtyugin, D.A. Semanov, P.V.
Yagin, L.N. Punegova, G.A. Marchenko, Russ. J. Gen. Chem. 56 (1986)
1863.
[38] P.C. Foller, UK Patent Appl. GB2,223,031 (1990).
[39] C. Montella, Electrochim. Acta 50 (2005) 3746.
[40] R.F. Mann, J.C. Amphlett, B.A. Peppley, C.P. Thurgood, J. Power Sources,
in press.
[41] K. Wandachowicz, W. Gnot, P. Dydo, Przemysl Chemiczny 81 (2002) 46.
[42] J.R. Chen, T. Wakida, J. Appl. Polym. Sci. 63 (1997) 1733.
[43] C.C. Addiion, Chem. Rev. 80 (1980) 21.
[18] A.A. Chichirov, V.A. Belonogov, Yu.M. Kargin, Russ. J. Gen. Chem. 62
(1992) 1948.
[19] A. Boughriet, M. Wartel, Electrochim. Acta 36 (1991) 889.