A study of tiron in aqueous solutions for redox flow battery application

Article abstract of DOI:10.1016/j.electacta.2009.09.031

In this study, the electrochemical behavior of tiron in aqueous solutions and the influence of pH were investigated. A change of pH mainly produces the following results. In acidic solutions of pH below 4, the electrode reaction of tiron exhibits a simple process at a relatively high potential with a favorable quasi-reversibility. The tiron redox reaction exhibits fast electrode kinetics and a diffusion-controlled process. In solutions of pH above 4, the electrode reaction of tiron tends to be complicated. Thus, acidic aqueous solutions of pH below 4 are favorable for the tiron as active species of a redox flow battery (RFB). Constant-current electrolysis shows that a part of capacity is irreversible and the structure of tiron is changed for the first electrolysis, which may result from an ECE process for the tiron electro-oxidation. Thus, the tiron needs an activation process for the application of a RFB. Average coulombic and energy efficiencies of the tiron/Pb battery are 93 and 82%, respectively, showing that self-discharge is small during the short-term cycling. The preliminary exploration shows that the tiron is electrochemically promising for redox flow battery application.

Full text of DOI:10.1016/j.electacta.2009.09.031

Electrochimica Acta 55 (2010) 715–720
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
A study of tiron in aqueous solutions for redox flow battery application
Yan Xu, Yue-Hua Wen∗, Jie Cheng, Gao-Ping Cao, Yu-Sheng Yang
Research Institute of Chemical Defense, Beijing 100191, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 5 July 2009
Received in revised form 7 September 2009
Accepted 8 September 2009
Available online 15 September 2009
In this study, the electrochemical behavior of tiron in aqueous solutions and the influence of pH were
investigated. A change of pH mainly produces the following results. In acidic solutions of pH below 4, the
electrode reaction of tiron exhibits a simple process at a relatively high potential with a favorable quasi-
reversibility. The tiron redox reaction exhibits fast electrode kinetics and a diffusion-controlled process.
In solutions of pH above 4, the electrode reaction of tiron tends to be complicated. Thus, acidic aqueous
solutions of pH below 4 are favorable for the tiron as active species of a redox flow battery (RFB). Constant-
current electrolysis shows that a part of capacity is irreversible and the structure of tiron is changed for
the first electrolysis, which may result from an ECE process for the tiron electro-oxidation. Thus, the tiron
needs an activation process for the application of a RFB. Average coulombic and energy efficiencies of the
tiron/Pb battery are 93 and 82%, respectively, showing that self-discharge is small during the short-term
cycling. The preliminary exploration shows that the tiron is electrochemically promising for redox flow
battery application.
Keywords:
Redox flow battery
Tiron
Electrochemical behavior
Acid aqueous solution
Tiron/Pb battery
© 2009 Elsevier Ltd. All rights reserved.
1. Introduction
in solutions, to the cell potential and to the number of electrons
transferred during discharge. Comparatively, most of organic mate-
As a new type of low cost, high-efficiency, large-scale energy
storage system [1], the redox flow battery (RFB) allows energy to
be stored in two recirculated electrolytes that contain different sol-
uble redox couples. It can not only be used for the electricity storage
of renewable energy to ensure the continuity and stabilization of
electricity generation and supply, but also can be used for load lev-
eling/peak shaving of electricity supply networks. Over the past
years, it has received widespread great attention. Since the redox
flow battery concept was proposed by Thaller [2] in 1974, a num-
ber of RFB systems, such as chromium/iron [2], zinc/bromine [3],
polysulfide/bromine [4], all-vanadium [5], vanadium/bromine [6]
and lead-acid [7], have been fabricated and developed. The related
battery techniques obtained substantial progresses. Among these
systems, the vanadium redox flow battery is the most practical can-
didate due to the use of the same element in both the half-cells,
which avoids problems of cross-contamination of the two half-
cell electrolytes during long-term use [5]. However, all the present
RFBs are based on inorganic active materials. That is, their devel-
opment is inevitably restricted by limited mineral resources. For
example, with the rapid soaring of the vanadium mineral resources
price, the commercialization of all-vanadium redox flow batteries
is faced with pressure. In addition, the specific energy of RFBs is rel-
atively low, which is related to the concentration of the redox ions
rials can be synthesized from abundant resources mainly being
composed of such elements as C, H and O. Moreover, chemical tun-
ability of organic compounds has made them even more attractive
[8]. That is, organic materials can be modified (designed) to give
additional chemical and/or electrochemical properties of inter-
est. This characteristic favors the number of electrons transferred
during discharge to be adjusted. Therefore, it inspired us to inves-
tigate the electrochemical prospect of organic active materials for
application in RFBs, expanding the space for the development of
RFBs.
Tiron (4,5-dibenzoquione-1,3-benzenedisulfonate) is one kind
of weakly acid aromatic organic compound, belonging to the
derivative of catechol [9]. Usually, catechol and its derivatives are
electrochemically reversible to a certain degree with relatively high
electrode potentials in aqueous solutions [10]. Two sulfonic groups
on the benzene ring of tiron improve the solubility of tiron. Hence,
tiron has the potential to be used as the positive active material
of a RFB. To date, however, there has been no literature describing
the use of tiron in chemical power sources as an active material.
In most cases, tiron is used in chemical [11] and electrochemical
analysis [12] as a chelator and an indicator.
In this study, the electrochemical behavior of the tiron is
explored deeply in order to assess its suitability as active species
in the positive electrolyte of a RFB. On top of it, a preliminary
understanding of its electrode reaction mechanism in acidic aque-
ous solutions is developed. At the same time, the charge–discharge
performance of a small tiron/Pb test cell is reported.
∗
Corresponding author.
E-mail address: wen yuehua@126.com (Y.-H. Wen).
0013-4686/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2009.09.031

716
Y. Xu et al. / Electrochimica Acta 55 (2010) 715–720
Fig. 1. Cyclic voltammograms for 0.05 M tiron in 1 M KCl solutions of pH (a) 0; (b) 2; (c) 4; (d) 6; (e) 7; (f) 8; (g) 10; (h) 12 at a graphite electrode, scan rate: 50 mV s−1
.
2. Experimental
the positive and negative electrode compartments and a cation-
exchange membrane (Nafion 115, Du Pont) situated between
the two compartments. A flange joint held the membrane with
rectangle-shaped rubber gaskets. The positive electrode compart-
ment housed a 10 mm thick graphite felt electrode (area: 5 cm2)
contacted against one graphite plate that acted as the current-
collector. The negative electrode compartment housed a lead
electrode used in valve-regulated lead-acid (VRLA) batteries, the
area of which was calculated according to the theoretical capacity
needed. Twenty milliliters of 0.25 M tiron in 3 M H2SO4 medium
was employed as the positive electrolyte and 3 M H2SO4 solution
of 20 mL was employed as the negative electrolyte. Two Xishan
magnetic drive pumps (China) were used to pump each half-cell
electrolyte through the corresponding half-cell cavity where the
charge–discharge reactions occurred.
2.1. Chemicals and electrode pretreatment
The tiron (hydrated 4,5-dibenzoquione-1,3-benzenedisu-
lfonate) of 98% purity was from Alfa Corp., TianJin, China and all
the other reagents were analytical grade.
Prior to test, the working electrodes were pretreated as follows:
after grinding with emery paper 1000 grade, the electrodes were
washed by ultrasonic cleaning in de-ionized water for 10 min. After
cleaning, the electrodes were cycled in 3 M H2SO4 solution between
−1.0 and 1.0 V (vs. SCE) for 20 min at a scan rate of 10 mV s−1
.
2.2. Cyclic voltammetry and electrolysis
The performance of test cells was evaluated with constant-
current charge–discharge experiments. A Land CT2000A battery
test system (Jinnuo Wuhan Corporation, China) was employed
for the charge–discharge cycling experiments. During the
charge–discharge cycles, the cell was charged at a current density
of 10 mA cm−2 up to 1.2 V and discharged down to 0.6 V cut-off at
the same current density. After full charging, open-circuit voltage
of the battery was measured.
Cyclic voltammetry (CV) tests were performed in a three-
electrode cell which comprised a graphite rod embedded in epoxy
resin as working electrode (0.24 cm2), with a big area graphite plate
and a saturated calomel electrode (SCE) as counter and reference
electrodes, respectively. The electrolytic solution used was 0.05 M
tiron in 1 M KCl aqueous medium with pH adjusted with H2SO4 or
NaOH for the studied pH range of 0–12.
To compare with the CV results obtained in stationary solutions
and to get the kinetic data, voltammograms were also recorded at a
series of rotation rates using a graphite disc electrode (0.1256 cm2).
This electrode was pretreated as above before test. Cyclic voltam-
mograms were measured by CHI1100 electrochemical station (CH
Corporation, USA). The rotation rate of the rotating disc electrodes
(RDEs) was controlled with an EG & G Model 636 RDE unit. All
experiments were conducted at room temperature of 25 ◦C.
The constant-current electrolysis of the tiron was conducted
at 10 mA cm−2 with a flow-type cell, in which a cation-exchange
membrane (Nafion 115, Du Pont) was used as a separator. A
graphite felt electrode (10 mm in thickness) contacted against one
graphite plate was used as the working electrode. A lead negative
electrode with an area of around 20 cm2 and a SCE electrode were
used as the counter electrode and reference electrode, respectively.
The apparent surface area for the working electrode was approxi-
mately 5 cm2. Fifty milliliters of 50 mM tiron in 3 M H2SO4 medium
was used as the feed solution while 3 M H2SO4 was contained in the
counter-electrode compartment. One Xishan magnetic drive pump
(China) was used to pump the feed solution through the working
electrode where the electrolysis reactions occurred. The constant-
current electrolysis of the tiron was recorded by an electrochemical
test station (Solartron 1280B, England) at the temperature of 298 K.
3. Results and discussion
3.1. Effect of pH
The structures of some catechol derivatives are affected greatly
by the pH of aqueous solutions, thereby exhibiting different elec-
trochemical behaviors with varying the pH [13]. As a derivative
of catechol, the electrochemical behavior of tiron may be pH-
dependent. Therefore, this paper firstly deals with the influence
of pH on the voltammetric behavior of tiron to make sure the suit-
able pH range of solutions for the use of tiron in RFBs as an active
material.
Fig. 1 shows the influence of pH on the cyclic voltammograms
of 0.05 M tiron at a graphite electrode in 1 M KCl aqueous solution
of different pH. As seen in Fig. 1A, the profile of the I vs. E curves is
almost the same in solutions with pH lower than 4. And, only one
oxidation peak (Epa = 0.74 V) and a corresponding reduction peak
(Epc = 0.58 V) are obtained. However, with the increase of pH, the
oxidation peak potential shifts toward a less positive value. Also,
the peak current is somewhat reduced. When the pH of solutions
reaches 4, an additional much smaller oxidation peak and a corre-
sponding reduction peak appear at around 0.25 and 0.20 V vs. SCE,
respectively. From Fig. 1B, it is found that when the pH changes
progressively from 4 to 12, the responding currents of the new
coupled peaks increase gradually. In contrast, the peak current at
the high potential decreases continuously. When the pH is up to
2.3. Battery fabrication and charge/discharge measurements
A
model test cell was applied for the constant-current
charge/discharge tests. This test cell consisted of three main parts,

Y. Xu et al. / Electrochimica Acta 55 (2010) 715–720
717
Table 1
Data obtained from cyclic voltammograms of Fig. 1 at various pH.
pH
0
2
4
6
7
8
10
E1A/V
0.735
0.159
1.42
–
–
–
0.689
0.220
1.54
–
–
–
0.650
0.191
1.12
0.308
0.103
1.10
0.648
0.184
1.12
0.295
0.083
1.28
0.619
0.139
1.36
0.286
0.070
1.31
0.629
0.154
1.66
0.268
0.055
1.37
0.618
0.142
1.87
0.254
0.038
1.19
|E1A − E1C|/V
|I1A/I1C
|
E2A/V
|E2A − E2C|/V
|I2A/I2C
|
12, multi-oxidation peaks appear on the voltammograms without
the corresponding reduction peaks. This indicates that the oxida-
tion products of tiron are difficult to be reduced in strongly alkaline
solutions of pH up to 12. It is manifested by the above cycle voltam-
mograms that in acidic solutions of pH lower than 4, the electrode
reaction of tiron is simple with a relatively high potential, which
is favorable for the application of tiron in RFBs. In solutions of pH
higher than 4, the electrode reaction of tiron turns to be compli-
cated. Therefore, for the use of tiron in RFBs, the pH of electrolytes
should be lower than 4. For this reason, we limited our studies to
the solutions of pH 0 below.
Some data obtained from Fig. 1 and summarized in Table 1,
where EA stands for the oxidation peak potential, |EA − EC| for
the peak potential separation and |IA/IC| for the peak current
ratio, respectively. From Table 1, the voltammetric behavior can
be explained except for that at pH 12, (1) with rising the pH,
the oxidation potential of tiron decreases progressively; (2) for
the electrochemical reaction taking place at the high potential,
the potential separation is more than 100 mV and the peak cur-
rent ratio is ranged from 1.12 to 1.87, which is deviated from
the value of unity to varying extents. At pH below 8, the extent
of deviation is less than 50%, while it is especially large (>50%)
at pH above 8. This suggests that the electrode process at pH
above 8 is far from the ideal reversibility; (3) for the redox
reaction taking place at the low potential, the potential sepa-
ration is basically less than 100 mV, which is particularly small
(<60 mV) at pH above 8. Nevertheless, the peak current ratio is
still higher than 1. It suggests that in solutions of pH below 12,
the electrode reaction of tiron displays a quasi-reversible redox
process, while at pH 12, the oxidation products of tiron can-
not be reduced and the electrode process of tiron has become
irreversible.
3.2. Effect of scan rate and cycling
Fig. 2A gives the cycle voltammograms recorded at a graphite
electrode at a range of potential scan rates for 0.05 M tiron in 1 M
KCl solutions of pH 0. A plot of the peak current density (jp) as a func-
tion of the square root of scan rates (v1/2) is shown in Fig. 2B. As
seen in Fig. 2A, nearly no change in the profile of voltammograms
is observed in the range of scan rates studied. But, with increas-
ing the scan rate, the gradual shift of the peak potentials causes
the peak potential separation to be increased to a certain degree.
It is verified further that the system in strongly acidic aqueous
solutions presents a quasi-reversible behavior and an increasing
irreversibility for faster scan rates. In parallel, for both the oxida-
tion and reduction processes of tiron, an almost linear response is
obtained (see Fig. 2B), indicating relatively fast electrode kinetics
and a diffusion-controlled reaction.
Fig. 2. (A) Cyclic voltammograms recorded at potential scan rates (a) 5; (b) 10; (c)
20; (d) 50; (e) 100 mV s−1 at a graphite electrode for 0.05 M tiron in 1 M KCl solutions
of pH 0. (B) Plots of the peak current density (jp) vs. the square root of scan rate
(v1/2) for the oxidation and reduction processes in (A). (C) Cyclic voltammograms
for 0.05 M tiron in 1 M KCl solutions of pH 0 at a graphite electrode with the 1st and
100th scan cycles. Scan rate: 50 mV s−1
.
In addition, Fig. 2C illustrates the cyclic stability of tiron at the
graphite electrode during the initial 100 cycles in the same solution
at a scan rate of 50 mV s−1. Compared with the first cycle, almost no
variation is observed among the remaining voltammograms. This
suggests that the cycling stability and the reversibility of tiron at a
graphite electrode are electrochemically favorable.
3.3. Rotating disc voltammetry
Fig. 3 reports cyclic voltammograms recorded at a rotating
graphite disc electrode with various rotation rates in 0.05 M
tiron + 1 M KCl solutions of pH 0. In a stationary solution (rotation
rate = 0 rpm), the voltammogram shows well-formed reduction and

718
Y. Xu et al. / Electrochimica Acta 55 (2010) 715–720
Fig. 3. Cyclic voltammograms recorded at a rotating graphite disc electrode in
0.05 M tiron + 1 M KCl solutions of pH 0, scan rate: 50 mV s−1, rotation speeds: 0,
100, 400 rpm for curves 1–3, respectively.
Fig. 5. Electrolytic characterization of tiron in 3 M H2SO4 at 298 K.
For an electrode reaction that is activation controlled, the exchange
current density can be determined by the Butler–Volmer formula
[14]; the average value calculated was 3.9 × 10−4 A cm−2 which is
comparable to that of vanadium ions. It can be inferred that the
major polarization of tiron may be caused by the mass transporta-
tion when the tiron is used as battery active species. Herein, the
flow of electrolytes containing active species, tiron, in a RFB would
facilitate a significant decrease in the mass transportation polar-
ization.
coupled oxidation peaks at 0.74 and 0.58 V vs. SCE, respectively.
The formal potential of tiron is thus seen to be at around 0.66 V
vs. SCE, which is somewhat lower than that of the V(V)/V(IV) and
Br2/Br−couples. But, the effects of side reactions such as O2 evolu-
tion would be smaller when employing tiron as the positive active
species for a RFB since side reactions may be easier to occur at
a more positive electrode potential. Moreover, scanning the elec-
trode in the positive direction up to 1.5 V vs. SCE, no anodic side
reaction associated with O2 evolution takes place. No cathodic peak
associated with H2 evolution is observed on the reverse scan to
−0.3 V vs. SCE. With increasing the rotation rate, the oxidation peak
current increases owing to a decrease in the mass transportation
polarization. Scanning the electrode continuously in the positive
direction is seen to give rise to an increase in the oxidation current.
On the reverse scan, oxidation continues to 0.7 V vs. SCE and then
the current becomes cathodic with a lowered reduction peak. When
the rotation rate reaches 400 rpm, the oxidation wave shows a well-
defined limiting current plateau and no reduction peak appears.
This corresponds to the complete diffusion-controlled electrode
process.
Fig. 4 shows the dependence of the voltammetry on the rota-
tion rate of the graphite disc electrode in the same solution. It can
be seen that a well-formed oxidation wave is observed and the
limiting current depends strongly on the rotation rate of the disc.
Indeed, a plot of the limiting current density, IL vs. w1/2 is linear
passing through the origin, confirming that the oxidation of tiron
becomes mass transport controlled and the Levich equation was
used to obtain a diffusion coefficient for tiron in this medium; the
value calculated was 2.2 × 10−7 cm2 s−1 which is one magnitude
lower than that of inorganic ions such as vanadium or lead ions.
3.4. Electrolytic characterization of tiron
Electrolytic characterization of the tiron was investigated by
constant-current electrolysis of 60 mL of 50 mM tiron in 3 M H2SO4
medium. Fig. 5 shows the electrolytic characterization of tiron in
3 M H2SO4 at 298 K. It is found that the charge potential of tiron
on the first cycle is much higher than that on the following cycles,
and the time taken for complete oxidization of tiron on the first
cycle is more than twice that on the following cycles. This suggests
that the first electro-oxidation of tiron may be a four-electron reac-
tion while the succedent electro-oxidation of tiron turns to be a
two-electron reaction. A possible explanation [15] is that the elec-
trochemical reaction mechanism of tiron is an ECE process in which
the first charge-transfer reaction is followed by a chemical reaction,
and then the secondary charge-transfer reaction occurs. The ECE
reaction process can be expressed as follows:
Consequently, the structure of tiron is changed. Starting from
the secondary electrolysis, only the secondary oxidation/reduction
Fig. 4. Voltammograms at a rotating graphite disc electrode as a function of rotation
rate. The solution is 50 mM tiron in 3 M H2SO4. Potential scan rate: 5 mV s−1
.

Y. Xu et al. / Electrochimica Acta 55 (2010) 715–720
719
reaction of tiron takes place. This phenomenon cannot be shown in
the cyclic voltammograms. The reason may be that the secondary
reaction is much easier to occur than the first one, resulting in only
one pair of peaks observed. Of course, this needs to be confirmed
with further experiments, e.g. ring-disc electrode experiments.
coulombic and energy efficiency values except for the first cycle
were calculated as 93 and 82%, respectively, which are comparable
to that for the all-vanadium and zinc–bromine systems at a rel-
atively low current density. This indicates that the self-discharge
due to diffusion of the species of tiron through the membrane is
small.
The proposed charge–discharge reactions of the Pb/tiron redox
flow cell after the activation can be described as follows:
Positive:
3.5. Charge–discharge performance of a tiron–Pb redox flow
battery
To minimize the cross-contamination of the positive and nega-
tive electrolytes through an ion-exchange membrane, a solid lead
negative electrode was combined with the tiron redox couple to
form a redox flow battery to demonstrate the suitability of tiron as
active species in the positive electrolyte of a RFB.
Negative:
Performance of a RFB employing 0.25 M tiron in 3 M H2SO4 as
positive active species and the lead electrode as negative active
species was evaluated with constant-current charge–discharge
tests and open-circuit voltage measurements, respectively. A plot
of cell voltage vs. time obtained for the initial five charge–discharge
cycles is given in Fig. 6A. The change in coulombic and energy effi-
ciencies of the Pb/tiron battery in the first 10 cycles is given in
Fig. 6B. From Fig. 6A, it can be seen that the charge voltage of the
battery for the first cycle is much higher than that on the follow-
ing cycles. In combination with Fig. 6B, the coulombic efficiency
is just 38% for the first cycle, indicative of a part of capacity irre-
versible resulting from the tiron. It suggests that the tiron needs
to be activated for the application of a RFB. From the second
cycle, the coulombic efficiency is enhanced to be over 90% which
has indicated the reversible behavior of the tiron activated elec-
trochemically. And, the battery basically shows no deterioration
during the succedent charge–discharge cycles. This corresponds to
the electrolytic characterization of tiron. Open-circuit cell voltage
after full charging remains constant at 1.10 0.05 V. The average
dischargePbSO4 + 2e−
E0 = −0.35 V
2−
Pb + SO4
ꢀ
charge
While the cell component materials such as electrodes and mem-
branes, and cell design have yet to be optimized, the above
results demonstrate that the tiron as active species in the posi-
tive electrolyte of a RFB is technically feasible and warrants further
investigations to increase the charge–discharge rate and cycling
stability of the battery.
4. Conclusions
In this study, the voltammetric behavior of the tiron in aqueous
solutions, influence of pH, and electrolytic characterization were
investigated. The following conclusion can be drawn:
1. The pH of solutions has a major influence on the electrochemical
behavior of tiron. In acidic solutions of pH below 4, the electrode
reaction of tiron exhibits a simple process at a relatively high
potential. In solutions of pH above 4, the electrode reactions of
tiron tend to be complicated. The acidic aqueous solution of pH
below 4 is favorable for the tiron as positive active species of a
redox flow battery.
2. The electrode process of tiron in acidic aqueous solutions of pH
below 4 is electrochemically quasi-reversible, exhibiting rela-
tively fast electrode kinetics and a diffusion-controlled reaction.
3. The electro-reaction mechanism of tiron is an ECE process, lead-
ing to a part of capacity irreversible and a change in the structure
of tiron for the first electrolysis. As a result, the tiron needs an
activation process for the application of a RFB.
4. The tiron/Pb battery employing 0.25 M tiron in 3 M H2SO4 as
positive active species and the lead electrode as negative ones
is reported. Coulombic and energy efficiencies of 93 and 82%,
respectively, are obtained in a small laboratory cell that has
a Nafion 115 membrane and graphite felt electrodes com-
pressed against graphite current-collectors. Self-discharge was
small during the short-term cycling. Further work is currently
underway to test the long-term effect of electrolyte cross-
contamination by tiron transfer across the membrane. A range
of modified graphite felt and membrane materials are presently
being evaluated over extended cycling periods to improve the
long-term performance of the battery.
Acknowledgements
Fig. 6. (A) Charge–discharge cycling curves for the Pb/tiron redox cell employing
0.25 M tiron in 3 M H2SO4 as positive active species and the lead electrode as negative
active species. Current density: 10 mA cm−2. (B) Coulombic and energy efficiencies
This work was financed by the National Basic Research Program
(973 Program) of China (2010CB227204) and the National Natural
Science Foundation of China (No. 20573135).
of the Pb/tiron battery in the first 10 cycles. Current density: 10 mA cm−2
.

720
Y. Xu et al. / Electrochimica Acta 55 (2010) 715–720
References
[7] A. Hazza, D. Pletcher, R. Wills, Phys. Chem. Chem. Phys. 6 (2004) 1773.
[8] M. Armand, J.M. Tarascon, Nature 451 (2008) 652.
[9] Y. Zeng, H.S. Zhang, Z.H. Chen, Handbook of Modern Chemical Reagent (4) Inor-
ganic Ion, 2nd edition, Chemical Industry Publishing Company, Beijing, 1993.
[10] E. Biilmann, Org. Compd. 13 (1923) 676.
[1] Y.S. Yang, S.M. Cai, Z.G. Lin, Sci. Technol. Rev. 24 (2006) 63.
[2] L.H. Thaller, US Patent 3,996,064 (1976).
[3] C.P. de Léon, A. Fr´ıas-Ferrer, J. Gonz´alez-Garc´etc, D.A. Szˇıanto, F.C. Walsh, J.
[11] G. den Boef, W. Ozinga, G.J. van Rossum, Anal. Chim. Acta 92 (1977) 387.
[12] H. Hiroshi, O. Naoki, K. Rokuro, et al., Talanta 9 (1962) 563.
[13] C. Giacomelli, K. Ckless, D. Galato, J. Braz. Chem. Soc. 13 (2002) 332.
[14] S. Zhong, M. Skyllas-Kazacos, J. Power Sources 39 (1992) 1.
[15] L. Papouchado, G. Petrie, R.N. Adams, Electroanal. Chem. Interfer. Electrochem.
38 (1972) 389.
Power Sources 160 (2006) 716.
[4] P. Zhao, H. Zhang, H. Zhou, B. Yi, Electrochim. Acta 51 (2005) 1091.
[5] M. Skyllas-Kazacos, M. Rychcik, R.G. Robins, J. Electrochem. Soc. 133 (1986)
1057.
[6] Y.S. Yang, L. Zhang, Y.H. Wen, J. Cheng, G.P. Cao, Chinese J. Power Sources 31
(2007) 175.